Polymeric Foams Technology and Developments in Regulation Process and Products

Citation preview

61259_C000.indd i

11/6/2008 4:39:29 PM

POLYMERIC FOAMS series editor Shau-Tarng Lee

INCLUDED TITLES

Polymeric Foams: Mechanisms and Materials Edited by Shau-Tarng Lee and N.S. Ramesh Thermoplastic Foam Processing: Principles and Development Edited by Richard Gendron Polymeric Foams: Science and Technology Shau-Tarng Lee, Chul B. Park, and N.S. Ramesh Polymeric Foams: Technology and Developments in Regulation, Process, and Products Edited by Shau-Tarng Lee and Dieter Scholz

61259_C000.indd ii

11/6/2008 4:39:30 PM

61259_C000.indd iii

11/6/2008 4:39:30 PM

CRC Press Taylor & Francis Group 6000 Broken Sound Parkway NW, Suite 300 Boca Raton, FL 33487-2742 © 2009 by Taylor & Francis Group, LLC CRC Press is an imprint of Taylor & Francis Group, an Informa business No claim to original U.S. Government works Printed in the United States of America on acid-free paper 10 9 8 7 6 5 4 3 2 1 International Standard Book Number-13: 978-1-4200-6125-3 (Hardcover) This book contains information obtained from authentic and highly regarded sources. Reasonable efforts have been made to publish reliable data and information, but the author and publisher cannot assume responsibility for the validity of all materials or the consequences of their use. The authors and publishers have attempted to trace the copyright holders of all material reproduced in this publication and apologize to copyright holders if permission to publish in this form has not been obtained. If any copyright material has not been acknowledged please write and let us know so we may rectify in any future reprint. Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced, transmitted, or utilized in any form by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying, microfilming, and recording, or in any information storage or retrieval system, without written permission from the publishers. For permission to photocopy or use material electronically from this work, please access www.copyright.com (http://www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA 01923, 978-750-8400. CCC is a not-for-profit organization that provides licenses and registration for a variety of users. For organizations that have been granted a photocopy license by the CCC, a separate system of payment has been arranged. Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. Library of Congress Cataloging-in-Publication Data Polymeric foams : technology and developments in regulation, process, and products / editors, Shau-Tarng Lee and Dieter P.K. Scholz. p. cm. -- (Polymeric foams series ; 4) Includes bibliographical references and index. ISBN-13: 978-1-4200-6125-3 (alk. paper) ISBN-10: 1-4200-6125-9 (alk. paper) 1. Plastic foams. I. Lee, S.-T. (Shau-Tarng), 1956- II. Scholz, Dieter P.K. III. Title. IV. Series. TP1183.F6P6482 2008 668.4’93--dc22

2008042301

Visit the Taylor & Francis Web site at http://www.taylorandfrancis.com and the CRC Press Web site at http://www.crcpress.com

61259_C000.indd iv

11/6/2008 4:39:30 PM

To the Lord who said, “For everyone who has will be given more, and he will have an abundance. Whoever does not have, even what he has will be taken from him.” (New Testament; Matthew Chap. 25 verse 29) This simply summarizes the bubble nucleation and growth in polymeric foaming way before we understood them.

61259_C000.indd v

11/6/2008 4:39:30 PM

61259_C000.indd vi

11/6/2008 4:39:30 PM

Contents

Preface ............................................................................................................. ix Series Statement ............................................................................................ xi Acknowledgments ..................................................................................... xiii Editors ............................................................................................................ xv Contributors ............................................................................................... xvii 1. History and Trends of Polymeric Foams: From Process/Product to Performance/Regulation ................................... 1 Shau-Tarng Lee 2. Development of Endothermic Chemical Foaming/ Nucleation Agents and Its Processes ............................................... 41 Dieter Scholz 3. Foam Extrusion Using Carbon Dioxide as a Blowing Agent ...................................................................................... 69 Walter Michaeli, Dirk Kropp, Robert Heinz, and Holger Schumacher 4. Processes and Process Analysis of Foam Injection Molding with Physical Blowing Agents ....................................... 101 Walter Michaeli, Axel Cramer, and Laura Flórez 5. Foaming Analysis of Poly(e-Caprolactone) and Poly(Lactic Acid) and Their Nanocomposites .............................. 143 Ernesto Di Maio and Salvatore Iannace

61259_C000.indd vii

11/6/2008 4:39:30 PM

viii

Contents

6. Nanostructure Development and Foam Processing in Polymer/Layered Silicate Nanocomposites .................................. 175 Masami Okamoto 7. New Material Developments from the Nitrogen Autoclave Process .............................................................................. 219 Neil Witten 8. Polystyrene Foam and Its Improvement in Vacuum Insulated Panel Insulation .............................................................. 255 Chang-Ming Wong Author Index ............................................................................................... 291 Subject Index ............................................................................................... 299

61259_C000.indd viii

11/6/2008 4:39:30 PM

Preface

Since the 1960s, polymeric foams have grown into a solid industry that affects almost every aspect of our lives. As our style of living improved, foam consumption increased as well. Solid efforts have laid a good foundation in foam science and technology, which allowed it to weather the energy crisis in the ’70s, ozone issues in the ’80s, and recycle/reuse in the ’90s. Above all, innovations in Foam science and technology made the development of new applications and improvements in industry performance possible. It is no wonder this industry continues healthy growth into the 21st century. As a result of the globalization of the manufacturing industries, emerging nations benefited greatly from enhanced skills and capabilities, and economic prosperity quickly followed. With living standards rising across the world, the global polymeric foam industries grew in quantity as well as in quality to meet the new demands for consumer goods. Every year, new concepts, innovations, and developments continue to be showcased at international foam conferences. As published in earlier volumes in this Polymeric Foam series, mechanisms, processing, science, technology, and materials have been addressed. But the pace of development and the social climate are rapidly changing. As reflected at the recent foam conferences, performance, sustainable resources, and energy security are becoming increasingly important. In 2005, we began dialogue about gathering a collection of renowned polymeric foam publishings and assimilating them into a book so that readers could get a clear picture of foam development. After frequent communications, certain commitments were secured, some unfortunately could not carry through, but there was enough to make the editing of this book possible. In general, it covers from early developments in blowing agent,

61259_C000.indd ix

11/6/2008 4:39:30 PM

x

Preface

process optimization, specialty foam development, and current trends. This book is divided into chapters written by world renowned authors, primarily from Europe, and are clearly practitioner oriented. The first chapter gives a historical perspective clearly showing a foam development trend from the ’50s into global production in the 21st century. The next chapter focuses on the blowing agent evolution and how this emerging technology was turned into an industry. Physical blowing agent in foam extrusion and injection moulding are addressed in Chapters 3 and 4, respectively. Chapter 5 illustrates interesting works in sustainable foam development. Nanocomposite foam is presented by a pioneer researcher from Japan in Chapter 6. Novel foam products and energy security foam are included in Chapters 7 and 8. This book is intended to present the development picture to benefit the industrial practitioner, researcher, academic faculty, and graduate school student and hopefully direct more enthusiasm and dedication into the already active flow for an even more prosperous future.

61259_C000.indd x

11/6/2008 4:39:30 PM

Series Statement

The bubble is a wonderful creation: a perfect spherical shape, beautiful arch in various degrees of curvature, and minimum surface area per given volume. Without the bubble, both art and science would definitely have a narrower scope. The bubble consists of a weak phase surrounded by, and sustained in, a strong phase. It is like the traditional Chinese virtue, Qian ). Although foaming may be one of the more mysterious pheXu ( nomena of the universe, it is fortunate that researchers and practitioners have been able to employ it to advantage with foamed products now commonplace in our daily lives. Foaming in polymers involves delicate scientific mechanisms, subtle processing techniques, unique morphology transformations, and structure formations. It combines material principles, engineering designs, processing methodologies, and property characterization. Polymeric foams are a 20th-century success story, which during the last quarter of the century, have evolved from laboratory-scale products to pilot-line samples and then onto commercial success. Today, it is viewed not only as a technique, but also as a well-established industry. Through challenges such as ozone depletion, recycling, and environmental regulation, in addition to upgrades, it has become a strong industry. Since polymeric foams have encountered various upgrades—materials/ technology, emissions/environmental concerns, properties/applications— it has been crucial to maintain the cohesiveness of polymeric foam by looking at it from various perspectives. This series aims to cover materials/ mechanisms, science/technologies, structures/properties, applications/ post-usage, and so on. The reader will be given an overall view, together with some fascinating insights, of polymeric foam. It has to be admitted that foaming is still mysterious in quite a few areas. It is my hope that a

61259_C000.indd xi

11/6/2008 4:39:30 PM

xii

Series Statement

healthy and cohesive understanding will not only strengthen the structure of the existing polymeric foam industry, but will also generate further developments to reveal more basic truths. Let us not forget that life and truth should go hand in hand. Shau-Tarng Lee, Series editor Sealed Air Corp., New Jersey

61259_C000.indd xii

11/6/2008 4:39:30 PM

Acknowledgments

We appreciate the depth of the editorial responsibility in creating this volume, especially as it contains work from some of the foremost and dedicated exponents of polymeric foams. While it is now time to breathe a sigh of relief regarding the completion of this volume, polymer foam work is, nevertheless, an ongoing and challenging pursuit. We give thanks for inspiration and comfort throughout the editing process, especially when on the brink of giving up or at the point of going nowhere. We would like to convey our heartfelt appreciation to those who diligently met the deadlines in an ever-increasing workload, sought approval for publication, and secured copyright permissions. In addition, the authors, who are all of diverse international origins, certainly deserve appreciation for their efforts to overcome the cultural differences involved with their contributions to this book. Without their dedication, it would have been impossible for this volume to have been produced. The principal editor would like to thank Richard Gendron of the Canadian National Research Council, Professor James Lee of The Ohio State University, Professor Tom Turng of University of Wisconsin at Madison, and Dr. Andrew Pacquet (Formerly of Dow) for offering insightful comments during the review process. Their valuable critiques have definitely helped to improve the quality of this work and in updating the references. Luis Costa and Slavek Kubicz of the Sealed Air Corporation helped prepare formatted drawings and reference searches for which their help is greatly appreciated. Our special thanks go to our spouses, Friederike Scholz and Mjau-Lin Lee, for their perseverance and patience during our absence, when on necessary trips, and when tension and discord inevitably materialized in the

61259_C000.indd xiii

11/6/2008 4:39:30 PM

xiv

Acknowledgments

family during the time-consuming and arduous editing process. We would also like to extend our thanks to Joseph Lee of Nyack Theological College and Matthew Lee, a Penn State graduate student, for their assistance with grammatical correctness. Above all, may this book be used by God to inspire more to explore the truth to benefit future generations.

61259_C000.indd xiv

11/6/2008 4:39:30 PM

Editors

Shau-Tarng Lee was born and raised in Taipei, Taiwan, Republic of China. He received his Bachelor of Engineering degree from the National TsingHua University. In 1980, he joined the Chemical Engineering department at Stevens Institute of Technology and was under Professor Joseph Biesenberger’s guidance in foam-enhanced devolatilization. In 1981 and 1982, he received summer internships from Farrel Company to investigate bubble phenomena in devolatilization. He also received a Stanley fellowship and a grant from the National Science Foundation (NSF) to support his research work at Stevens Institute of Technology. After receiving his Master’s Degree in Engineering and Doctor of Philosophy (PhD), he joined Sealed Air Corporation in 1986. Since then, he has specialized in foam extrusion research, development, and production support as a development engineer, assistant research director, and research director. Dr. Lee has over 100 publications to his credit, including 26 US patents. He was elected a Fellow of the Society of Plastics Engineers in 2001 and was inducted into Sealed Air’s Inventor Hall of Fame in 2003. In July 2000, he was the editor for Foam Extrusion: Principles and Practice, published by CRC Press (now part of the Taylor & Francis Group). He is also the series editor for the Polymeric Foam Series (CRC Press) that began with Polymeric Foams: Mechanisms and Materials (edited by S. T. Lee and N. S. Ramesh) published in 2004 and was then followed by Thermoplastic Foam Processing: Principles and Development (edited by Richard Gendron). The third volume, Polymeric Foams: Science and Technology (edited by S. T. Lee, C. B. Park, and N. S. Ramesh), was published in 2007. In addition, he is also the co-editorin-chief for the Journal of Cellular Plastics. He is married to Mjau-Lin Tsai and has three children, Joseph, Matthew, and Thomas. Currently, they reside in Oakland, New Jersey. Dr. Lee is a born again Christian and is actively involved in mission works in Asia.

61259_C000.indd xv

11/6/2008 4:39:30 PM

xvi

Editors

Dieter Scholz was born on March 1, 1938 in Germany and, in 1955 after finishing school, he started his career with an apprenticeship contract and education in chemistry. Organic chemistry and organic synthesis of chemicals (fine chemicals) for various applications (mainly intermediates for R&D, colorants, pharmaceuticals and other areas) were his main focus. After finishing the education program, he worked as a chemist for the same company until 1960 when he joined the Battelle Memorial Institute (Frankfurt) as a chemist working partly on organic synthesis again, but mainly on monomers and polymers (polyimids). In order to enhance his education in chemistry, he undertook evening courses in mechanics and process engineering and finished these engineering studies in 1965. The following six years were spent working for Diversey, and then, in the summer of 1972, he joined Boehringer Ingelheim. In the initial years, he was responsible for the application of the chemicals produced. Those activities involved dealing with German and International Food Regulators (FDA and others) in order to negotiate approvals. From this experience, a wide overview of the many processes concerning chemicals used in the various industries was gained. One of them, the Hydrocerol product range, was the first ready-to-use system for gas nucleation (distribution) in the production of polystyrene and polyethylene foams. Subsequently, it was discovered that this material had a use in chemical foaming as well and this was the start of ready-to-use endothermic blowing agents in the plastics processing industry. After more than 25 years with Boehringer Ingelheim, Dieter Scholz elected for retirement in 1998. Since then, he has participated in smaller, time-limited projects related to foam and plastics processing and has helped in the setting up of international plastics foam conferences and in keeping contact with former and new “colleagues” from the relevant industries and R&D facilities.

61259_C000.indd xvi

11/6/2008 4:39:30 PM

Contributors

Axel Cramer Institute of Plastics Processing (IKV), RWTH-Aachen University, Aachen, Germany Ernesto Di Maio Department of Materials and Production Engineering, Faculty of Engineering, University of Naples Federico II, Naples, Italy Laura Flórez Institute of Plastics Processing (IKV), RWTH-Aachen University, Aachen, Germany Robert Heinz Institute of Plastics Processing (IKV), RWTH-Aachen University, Aachen, Germany Salvatore Iannace Institute of Composite Materials and Biomaterials, National Research Council, Portici, Italy Dirk Kropp Institute of Plastics Processing (IKV), RWTH-Aachen University, Aachen, Germany Shau-Tarng Lee

Sealed Air Corporation, Saddle Brook, New Jersey

Walter Michaeli Institute of Plastics Processing (IKV), RWTH-Aachen University, Aachen, Germany Masami Okamoto Advanced Polymeric Materials Engineering, Graduate School of Engineering, Toyota Technological Institute, Tempaku, Nagoya, Japan

61259_C000.indd xvii

11/6/2008 4:39:31 PM

xviii

Dieter Scholz Germany

Contributors

(Retired from Boehringer Ingelheim) Gau-Algesheim,

Holger Schumacher Institute of Plastics Processing (IKV), RWTH-Aachen University, Aachen, Germany Neil Witten Zotefoams plc, Croydon, England, United Kingdom Chang-Ming Wong Material and Chemical Research Laboratories, Industrial Technology Research Institute, Hsinchu, Taiwan, People’s Republic of China

61259_C000.indd xviii

11/6/2008 4:39:31 PM

1 History and Trends of Polymeric Foams: From Process/Product to Performance/ Regulation Shau-Tarng Lee

CONTENTS 1.1 Introduction ...................................................................................... 1 1.2 Foundation: Science Lab/Pilot ....................................................... 7 1.3 Technology: Process/Product ...................................................... 13 1.4 Performance: Properties and Applications ................................. 20 1.4.1 Physical Properties ............................................................... 21 1.4.2 Mechanical Properties ......................................................... 22 1.4.3 Thermal Properties .............................................................. 25 1.4.4 Acoustic Properties .............................................................. 27 1.5 Regulation: Environmental and Regulatory .............................. 28 References .............................................................................................. 38

1.1 Introduction Every material that we come across has a natural origin. We humans are incapable of creating something out of nothing. No matter what we do, we can only create something out of something else, and are bound by the mass conservation law. As more scientific advancements are made,

61259_C001.indd 1

10/25/2008 12:33:06 PM

2

Polymeric Foams

we are better able to harness natural resources and use them for our benefit. Polymers, like wood and metal, are products made from natural sources that can be used for a variety of purposes. Indeed, producing and using polymers has greatly enhanced our standard of living and altered modern culture.1,2 The basic ingredients of polymers come from petroleum, plants or animals. After these ingredients are synthesized, polymers are imbued with the durability useful for many applications. This branch of science began to draw global attention in the nineteenth century and took off in the twentieth century, when polymer synthesis was established in the lab, followed by a scaling up to mass production. It became extremely popular after World War II (WWII) and within 30 years, its consumption exceeded both wood and metal as the most used material in the modern world. When nylon was introduced in the late 1930s by DuPont, it soon became a favorite in the textile and clothing industry.3 Nowadays, plastic bags and foam packaging are commonly used as everyday polymer products. Polymers inherently possess unique structural arrangements that allow them to combine chemical intra-bond and van der Waals inter-bond forces to form distinct melting and crystallizing transitions, which offers a special property spectrum for applications. The processing cost and performance per unit weight for polymers became so favorable that capital investments have continually expanded to meet demand. The profits were invested for further research and innovation to bring forth new technologies for a wide variety of applications, resulting in a positive investment cycle. However, the life cycle of polymers, as illustrated in Figure 1.1, suggests time-magnitude issues that are the basis for life-cycle assessment (LCA). It is clear that petroleum formation from buried plants and organisms is a magnitude too large to be practical for regenerating resources. As a result, resources will continue to diminish. The second largest magnitude is the decomposition on polymeric products after use. No wonder then that landfill space has been, and continues to be, short on demand. It will continue to be so for as long as the degradation of (or alternatives to) polymers remains unresolved. It is worth noting that the decomposition of polymers will only resolve space or “presence” concerns,4 and will not necessarily help promote the cycle even if decomposed to the original elements. For instance, when polyethylene is cracked into ethylene, which is lighter than air, it evaporates into the atmosphere and does not participate any further in the life cycle. When the polymer industry began to take off in the 1930s, polymeric foam science was established to follow the expansion of the polymer industry. Its success in application (i.e. in WWII) encouraged technology development to start a very positive development–application reciprocal cycle. First, its lighter nature enlarged the already wide property spectrum of polymers by offering special cushioning, insulation, and absorption

61259_C001.indd 2

10/25/2008 12:33:06 PM

3

History and Trends of Polymeric Foams

103

Time (years)

102

101

1

Fossil

Feed Plants CO2 + H2O

Cracking/ Polymerization

Landfill, Incineration, Degradation

10–1

Customer

Processing (foam, film, etc.)

10–2 Process FIGURE 1.1 Organic life cycle.

properties. Its market shares continued to improve by extending into new applications and by replacing conventional business. Before WWII, the chemistry required to generate gaseous bubbles was discovered in the polycondensation reaction.5 Polyurethane (PU) development was first focused on rivaling polyamide (e.g. nylon) in the fiber industry. However, when PU foam was developed, its properties soon enabled it to far outrun fiber research. Its uses have rapidly extended into furniture, construction, and transportation since then. In the same period, the cryogenic industry found chlorofluorocarbon (CFC) to be a superb cooling chemical, (which was also found as good in blowing polymeric foam.6) An important “pillar“ was the emergence of a technique for making thermoplastic polymers into which huge investment was poured. Polystyrene (PS) was adopted for foam-making and, instead of a batch process, a continuous extrusion process was developed to greatly enhance productivity. When resins and polymers were available, the final “pillar” was the processing technology. Significant improvements to machinery were made by Lavorazione Materie Plastiche (LMP) in the late 1930s7,8 and benefits were also derived from “pasta” processing. Post-WWII, more support to civilian manufacturing was secured which proved effective in upgrading processing technologies. The four “pillars” forming the foundation of polymeric foam technology are listed in Table 1.1.

61259_C001.indd 3

10/25/2008 12:33:06 PM

Polymeric Foams

4

TABLE 1.1 Four Pillars of Polymeric Foam Foundation Process 1. 2. 3. 4.

Foaming chemistry Blowing agent Resin manufacture Processing technology

Attributes Generate gas within polymeric matrix Soluble and stable gas for polymer Material strength to hold foaming Melting, mixing, and cooling in one stage

From this foundation, laid in the 1940s and 1950s, two further advances resulted. One was thermoset foams such as polyurethane (PU); the other was thermoplastic foams such as polystyrene (PS), polyvinylchloride (PVC), and polyolefin. Since foam appeared to be so useful, both foams were accepted from the outset, and that acceptance has been a solid driver for the foam industry ever since. During the 1950s to the 1970s, many new technologies were developed one after another and it would be fitting to refer to this period as the “technology period”. With regard to thermoset foams, two branches evolved: soft foam and rigid foam. It was amazing that the same chemistry applied to different raw materials in different morphological structures could result in completely different properties ranging from as soft as a sponge to as hard as a rock. A varied range of equipment was developed to capture the reaction for desired products. Now, after 40 years, PU foam has become the dominant force in a variety of markets: furniture, automotive, construction, packaging, transportation, and recreation. As for thermoplastic foam, extrusion of polystyrene was first used by Dow for floating dock construction during WWII. The floating and insulation characteristics were soon recognized and subsequently, thermoforming, packaging, and structural designs were developed. The technology was further strengthened by the development of polyolefin foam extrusion, PS molded bead, injection molding, and cross-linked polyethylene (X-PE) technologies. Thermoplastics would not only serve the public through mass production (e.g. cups, trays) but would also offer high quality products (e.g. X-PE foam) for niche applications. The 1980s can be termed “application period.” Foam found inroads into automotive interiors, transportation, aviation, recreation, construction, medical, military, oil drilling, and reflotation markets. More often than not, market specification was built around foam products to direct development efforts. Nonetheless, more attributes were established such as insulation, shock absorption, acoustics, durable modulus, and light emission.9 Although the depletion of the ozone layer and the advent of microcellular foaming technology became interesting topics during the 1980s, application developments maintained a steady momentum. The whole development is akin to a tree with its root, trunk, leaves, and fruits as shown in Figure 1.2.

61259_C001.indd 4

10/25/2008 12:33:06 PM

5

History and Trends of Polymeric Foams

1960–now

Applications

1950–1970

Technology

1930–1950

Science

FIGURE 1.2 Foam development tree.

Now, with the beginning of the twenty-first century, social and environmental climates are dramatically changing. Global warming, degradable foam, energy security, and sustainable sources have become increasingly more important. Recycling including reuse and reprocessing became popular in the mid-1990s. In the US, since organizing a sorting system for the general public has not proved as convenient as in Japan, for instance, due to the residential distribution and collecting system, the results have not been so evident. Another factor concerns the petroleum price, which was stable for a quarter of a century. Then, in 2004, its price began to increase with the crude oil price increasing from $20 per barrel to over $100 per barrel in 2007 and continuing to climb in 2008. In addition to gasoline prices being forced to adjust, the raw material price for polymers was also heavily influenced. Therefore, the pressures for alternative resources to polymers have also increased significantly. Due to the discovery of the depletion of the ozone layer in the late 1980s, the Montreal Protocol was drafted to phase out CFCs first then and hydrochlorofluorocarbons (HCFCs) the two halogenated gases implemented in degrading the ozone layer.10 The threat of potential major climate change due to resulted global warming. Kyoto Protocol in 1996, which was drafted to maintain a neutral inorganic carbon dioxide flux in the low atmosphere. The Earth is mostly in a state of equilibrium with quite diverse parameters, all contained within a boundary. If one parameter moves across the boundary, however, a chain reaction in the other

61259_C001.indd 5

10/25/2008 12:33:06 PM

Polymeric Foams

6

parameters may ensure resulting in a global catastrophe, which may already have been set in motion, may thus be taking place. An increase in Earth’s temperature will contribute to an accelerated melting of the polar ice and, consequently, to sea level rises that will reduce the available land area and tighten living resources. Since the emission of inorganic carbon dioxide is inevitable in any energy generation process, it has created a huge problem for industrialized countries. Technology, on the one hand, facilitates our living, but it generates carbon dioxide emission which may cause global warming to hurt our living quality. Since population and location play critical roles, a credible measurement to quantify the carbon footprint issue is necessary to the success of the Kyoto Protocol. Knowing that plants can convert inorganic carbon dioxide to organic carbon, which is a reversal of the energy process in generating inorganic carbon, and that some other processes can make a similar conversion too, it was therefore been proposed to generate “credit” for such systems to balance out the emissions of industrial systems in order to achieve a neutral carbon flux. Carbon dioxide emission occurs in most conventional energy generation processes such as the burning of gasoline in vehicles for example. Polymer resin preparation, processing, converting, and even collecting a blowing agent to convert into atmospheric carbon, will all involve carbon dioxide emission. Growing plants to make polymers, which can turn inorganic carbon to organic carbon for the purposes of obtaining “credit” seems a viable use for polymer and polymeric foam. Table 1.2 lists some common ways of generating carbon dioxide. More and more international communities are now focusing on polymers made from sustainable sources. In the early 1990s, Mitsui Toatsu Chemicals Inc. succeeded in making polylactide foam.11 The concept seems quite intriguing. At present, petroleum price rises alone may be enough of an incentive for the development of a sustainable foam. PLA development has gained momentum in innovation and technology recenty.12,13 Others include ethanol from sugar cane, which can be converted to ethylene as the basic feedstock for polyethylene. In short, the fourth generation is clearly defined by environmental and societal issues (the “global regulation” generation). TABLE 1.2 Common Ways to Generate CO2 Action 1. 2. 3. 4.

61259_C001.indd 6

Burning fossil fuels Decomposing polymer Human and animal breathing Chemical Reaction

Consequence Organic carbon to inorganic carbon Organic carbon to inorganic carbon Inorganic oxygen to inorganic carbon Inorganic carbon to inorganic carbon

10/25/2008 12:33:06 PM

7

History and Trends of Polymeric Foams

Since the first appearance of polymeric foam, four distinct generations have been outlined: science, technology, application, and regulation. Each of these generations will be further addressed in the subsequent sections.

1.2 Foundation: Science Lab/Pilot Since most polymers are not tightly aligned, there are holes and microspaces that exist between polymer backbones that allow gas molecules to dwell.14 Gas is even present in the polymers. Foaming may not occur until enough gas molecules are clustered to reach the thermodynamic instability threshold. Unstable bubble nucleation and growth will then occur. The threshold is known as the “critical bubble radius,” defined as: Rcr  2s/(Pb  P)

(1.1)

where s, Pb, and P represent surface tension, bubble pressure, and surrounding pressure, respectively. As mentioned before, during a gradual change of thermodynamic conditions, diffusion and/or vaporization may be sufficient to restore the least energy state or equilibrium. Only when enough instability or disturbance occurs does the gas phase tend to follow

Work (W )

Wmin

V0

VC

Bubble size

FIGURE 1.3 Nucleation critical size.

61259_C001.indd 7

10/25/2008 12:33:07 PM

8

Polymeric Foams

FIGURE 1.4 Open cell structure.

a rigorous foaming path to re-equilibrate. A spherical shape has the least surface area at a given volume in parallel with the least energy principle. As illustrated in Figure 1.3, after critical size is reached, growth favors energy dissipation until another equilibrium is reached, or there is enough resistance to stop the growth, which is very likely when the surrounding temperature continues to decrease.15 It is noteworthy that s is an inherent fluid property, which suggests foaming is highly preferred in the molten state, or, at least, above its glass transition temperature, Tg. When temperature is reduced to solid state to hold the gaseous bubbles, surface tension literally disappears. The residual stress out of the entanglement of the polymer is formed, which may relax with time. Under certain circumstances, the bubble can grow to exceed the material limit causing cell rupture, which can lead to open cells as shown in Figure 1.4. It can be imagined that when foaming (nucleation and growth) is completed, the polymeric foam is saturated with the blowing agent. When exposed to the atmospheric environment for most applications, the gas concentration gradient can induce counter-diffusion fluxes. When the blowing agent exists, air comes in. After enough time, the concentration gradient diminishes to render a foam product filled with primarily air. In other words, the role of the blowing agent is like a catalyst. A simple illustration is presented in Figure 1.5.

61259_C001.indd 8

10/25/2008 12:33:07 PM

9

History and Trends of Polymeric Foams

Expansion

Solid

Gas

Homogenization

Nucleation

Foamed product

FIGURE 1.5 Blowing agent as “catalyst.”

Quite a few scientific breakthroughs have arisen by accident rather than by design. Foaming is no exception. Although cellular material readily exists on Earth, we have seldom succeeded in duplicating it under laboratory conditions, especially in the first half of the twentieth century. Around the 1930s, polycondensation was established as a viable way to synthesize polymers through functional group reaction and bonding. A chemical with a difunctional group, for instance, can self-react to build up the length of the backbone.16 A simple reaction is shown below: A-dimer  B-dimer Æ (A  B)-dimer

(1.2)

Polyester and polyamide (known as nylon) are typical examples. Polyethylene terephthalate (PET), a member of polyester family, is a popular polymer for foaming in food applications. Its esterification reaction is: HO–(CH2)n–OH  (C6H5)C–COOH Æ (C6H5)C–COO–(CH2)n–OH (1.3) The condensation concept was easily extended into branched polymers by implementing multifunctional chemicals as catalysts or branching agents in order to modify the nature of the polymer’s into elastic characteristics,17,18 which is deemed necessary in extrusion foaming. Moreover, when multifunctional monomers were introduced for reaction, instead of a 2D linear structure, a 3D matrix was built up. This is known as a thermoset polymer, which is quite different from the thermoplastic polymers in heat processing. The latter possesses thermo-reversible morphology, whereas the former is no longer reprocessible once set. A simple comparison is presented in Table 1.3.

TABLE 1.3 Thermoset and Thermoplastic Foam Comparison Thermoset Gas formation Physical blending Foaming path Polymer bond Foam nature

61259_C001.indd 9

Chemical reaction Spinodal decomposition Permanent bond for structure Open and closed cell Very soft to very rigid

Thermoplastic Chemical reaction Nucleation and growth van der Waal for inter-polymer Closed cell Soft to rigid

10/25/2008 12:33:07 PM

Polymeric Foams

10

Foam

Foam

Blowing agent

PU PIU Gas generated during polymerization

PET Nylon

Polycondensation FIGURE 1.6 Polycondensation foam architecture.

Sometimes, a gaseous byproduct can be generated during the condensation reaction, which naturally becomes the blowing agent. As long as the backbone is providing enough strength to hold the gas in the polymer, polymeric foam will be formed. An exothermic reaction facilitates the activity of gas molecules for more expansion. An endothermic reaction goes the other way but reinforces the strength of the polymeric surrounding. A summary polycondensation tree is illustrated in Figure 1.6. Nonetheless, PU fits nicely into the foaming criteria. Its reaction is in the following: R–N=C=O  H2O Æ RNHCOOH  R–NH2  CO2  heat (1.4) R–N=C=O  R–CH2 –OH Æ R–NH–COO–CH2 –R

(1.5)

R–N=C=O  R–NH2 Æ R–NH–CO–NH–R

(1.6)

The number of functional groups, the chain length between functional groups, and different catalyst systems can alter the product from soft as a seat cushion to for hard as for an appliance door. The degree of reaction can be controlled to make an open-cell product, as in a sponge, or one with a solid skin surface. A variety of PU foam products is therefore possible using the condensation principle. Table 1.4 presents the foam chemistry of PU and its product attributes. TABLE 1.4 PU Chemistry and Foam Properties Foam Flexible PU Rigid PU Elastomeric PU

61259_C001.indd 10

Chemistry Toluene diisocyanate (TDI) diphenyl Methylene diisocyanate (MDI) Rubber cross-linked into TDI

Attribute Cushion Insulation/structure Repeated usage

10/25/2008 12:33:07 PM

History and Trends of Polymeric Foams

11

Dr. Otto Bayer and his research group were responsibe for the isocyanate polyaddition procedure, which led to the preparation of a variety of polyurethanes.19 The success of polyamide (nylon), developed by DuPont, evidently encouraged research into PU. During WWII, the research program in the laboratories of Farbenfabriken Bayer was responsible for various PU products such as rigid foams, coatings, and adhesives.2,20–22 After WWII, Allied scientific teams studied German developments in PU, which subsequently stimulated worldwide efforts. It then became a solid pillar in the polymeric foam industry. When CFC-11 was developed in the 1930s, it was soon found to be very useful as an auxiliary blowing agent in PU foaming. In the exothermic urethane reaction, CFC-11 easily diffuses into cells for stable expansion.23 Its low conductivity opened the door for PU foam board for application in insulation markets. As for thermoplastic foam development, this was heavily dependent upon the processing technique. Lack of injection and metering control made a friendly blowing agent necessary for overcoming laboratorytesting shortcomings. Polystyrene with pentane was developed and patented in 1935.24 Since then, the development of the refrigerant known as chlorofluorocarbon (CFC) by Sir Thomas Midgley at General Motors became another powerful driving force for thermoplastic foam development, simply because of its solubility, stability, and non-flammability.6,25 Extruded PS foam was first made available in the US by the Dow Chemical Company in 1943 and was used for floating docks during WWII. Not long ago, researchers became aware of just how critical the solubility and diffusivity of the gas/melt are, especially in foam extrusion. A high solubility can reduce the time to reach a homogeneous gas/melt solution, which dictates the size of the extruder. The solubility curves for common polymers with carbon dioxide are illustrated in Figure 1.7.26 Although extrusion is an inherently efficient processing technology, it contributes heat during the melt process of the polymer. The removal of this excess heat is necessary to stabilize the melt solution for maximum foaming. From the solubility curve, it is clear that the lower the temperature, the higher the solubility. Reducing the heat input/removal and higher solubility became logical. Foam extrusion can be avoided by simply exposing solid PS to a blowing agent or the polymer soaking in the blowing agent reservoir until it becomes saturated. Reducing the pressure at an elevated temperature, normally above Tg and below Tm (melting temperature), will then allow foaming to occur followed by solidification into a cellular morphology. Fortunately, PS possesses a very useful glass transition temperature, Tg, slightly over 100°C, which allows steam to be a readily available medium for heating the chamber for ideal foaming conditions. In the 1960s, PS beading blossomed into an interesting foaming method. From an energy conservation perspective, it avoids heat addition and the consequent removal that is inherent in the common extrusion and injection molding processes.

61259_C001.indd 11

10/25/2008 12:33:07 PM

Polymeric Foams

12

Solubility (g-gas/g-polymer)

0.10

0.08

0.06

0.04

0.02

0.00

0

5

10 Pressure (MPa)

15

FIGURE 1.7 Solubility chart for CO2. ®, Polypropylene; ○, Low-density polyethylene; Δ, highdensity polyethylene; •, ethylene-ethyl acrylate copolymer; ▲, Polystyrene; ——, Estimated by Sanchez–Lacombe equation of state. (From Areerat, S., “Solubility, Diffusion Coefficient and Viscosity in Polymer/CO2 system.” Research thesis, University, Kyoto, Japan, 2002.)

When there was an attempt to use polyolefin to make foam, due to its huge volume in the market, there was a certain amount of doubt surrounding the resin characteristics when compared with polystyrene. It had neither an amorphous structure like foaming-grade PS, nor did it have a Tg like PS for strength and rigidity. In the early 1960s, Rubens et al., working for Dow Chemical, first attempted cross-linking PE via ultra violet (UV) light by setting a window before the extruder exit.27 Foaming was achieved when UV light was not applied. This evidently enlarged the base material spectrum from amorphous to branched semi-crystallization for foaming. It led to the contined growth in PE foam extrusion. Japan Styrene Paper (JSP) Corporation experimented with PE foam extrusion by foaming PS with pentane on a single screw extruder and had a enough success to be able to invest in a continued growth in foam extrusion.28 Extrusion and molded bead development helped define resin structural morphology for foaming, in which adequate material strength is key to holding the aggressive unstable expansion. Since heat is necessary to enhance expansion, the material strength at the foaming temperature is crucial for successful foaming. The necessary attributes are listed in Table 1.5. Another technique is the X-PE foaming methodology. Instead of using extrusion system for foaming, extrusion can be used for compounding

61259_C001.indd 12

10/25/2008 12:33:07 PM

13

History and Trends of Polymeric Foams

TABLE 1.5 Polymeric Foam Scientific Attributes Attributes

Mechanisms

1. Enough gas in polymer 2a. Nucleation and growth 2b. Phase separation 3. Product stabilization

Solubility: dissolution Reaction: generate gas Thermodynamic instability Spinodal decomposition Curing: cooling, permeation

to include a chemical blowing agent into the PE sheet, which is in turn subjected to a cross-linking reaction to build up extra material strength through chemical bonding. The chemistry is as follows: R–H (heat, UV, electron beam) Æ R•  H•

(1.7)

R•  R• Æ R–R

(1.8)

where a chemical bond is formed with H2 as a byproduct. After that, it is sent to an oven to activate the chemical blowing agent to expand the crosslinked sheet. The result is a fine cell and soft-feel foamed product. The cross-linked structure confers extraordinary thermal strength for thermoforming, which definitely extends the application spectrum. Heat-induced gas generation through decomposition is another viable way to generate gas within a molten polymer. It has been uned as a chemical blowing agent (CBA). Most suitable reactions occur at a temperature much higher than the foaming temperature. A typical sodium bicarbonate and citric acid reaction is: C6H8O7  3NaHCO3 Æ (C6H5Na3O7)2H2O  3CO2  H2O

(1.9)

Since the decomposition temperature is so high, only subsequent cooling or extra material strength can bring the foaming under control. Extrusion and cross-linking have become suitable for CBA foaming. A simple development chart is illustrated in Figure 1.8 with laboratory and pilot history included.

1.3 Technology: Process/Product When the science of ploymer foaming was understood, processes were then developed to produce foamed products that were then sold to generate

61259_C001.indd 13

10/25/2008 12:33:08 PM

Polymeric Foams

14

PET foam extrusion 1990

Microcellular PP foam extrusion X-PE + CBA

1970 PS molded bead

1950

PIR

PBA PS, PE

PU

CBA

1930 FIGURE 1.8 Foam science and process development. PiR, polyisocyanurate; PBA, physical blowing agent.

profits to support further developmental efforts. In essence, equipment design and scale-up have been the major challenges. However, the design challenge was very different in nature for thermoset foam in comparison with thermoplastic foam. The challenge for the former lay in the chemical reaction, and for the latter in the physical mixing. Both require high pressure, but this high pressure comes from gas generation in the thermoset reaction, and in thermoplastic foaming it is required to suppress the gas in the polymer. The chemistry of PU foam is straightforward. Different reactant ratios and degrees of reaction will result in a wide variety of PU foam products, that can be tailored to suit various applications. In general, the necessary product characteristics dictate the formulation chemistry and processes are then developed to deliver the product. Processing and process design has become one of the key areas of the thermoset methodology. With the process design support, PU evolved into rigid, flexible, and integral skin early on.29–31 A mixing chamber is necessary to keep chemicals thoroughly mixed until the reaction occurs. Foaming can take place in the constraints of a mold or in a free expansion format such as on a conveyer belt. After foaming and curing are completed, the product can undergo further fabrication to suit various applications. A simple PU foaming process is presented in Figure 1.9.32 The foam-in-place, or foam-in-mold, was developed to fill irregular voids or molds with foam. After setting, the foam can either be ejected for usage or stay in place as a component of the assembly for absorption, filling, or insulation. This process can be employed for flexible and rigid PU foams. Reactive injection molding (RIM), as illustrated in Figure 1.10,

61259_C001.indd 14

10/25/2008 12:33:08 PM

15

History and Trends of Polymeric Foams

Chemical A

Chemical B

Mixing chamber

nozzle

Applications Mold: Reactive/injection molding Open space: Foam in place Conveyor belt: Sheet or slab FIGURE 1.9 PU foaming.

and foam to fill the void for packaging are typical examples. Rigid PU for appliance doors can also be made using this method. When additional blowing agent was found to be effective in additional foaming promoted by the exothermic reaction between the isocyanate and hydroxyl reactants in order to reduce the thermoconductivity of PU foam, the insulation board process was established to meet construction codes. This process required PU streams to be poured on to the conveyer belt for free expansion and to set before cutting into the desired dimensions for shipping. Figure 1.11 demonstrates the process. Another development was “spray in place,” as shown in Figure 1.12, in which atomization of the raw components was achieved by spraying onto the substrate to form a thin coating of PU foam. An expansion of up to 30 (e.g. about 2 lb/ft3 or 33 kg/m3) could be easily accomplished. The process is applicable to simple as well as complex designs in indoor as well as outdoor applications. Process automation, mold design, and additives for performance were the main drivers in PU technology development. It also encouraged other thermoset foam developments such as isocyanurate foam for its flame retardancy33 and phenolic foam.34 In essence, fitting for application has been the focal point, and success has driven technology, thereby enlarging the application spectrum. It is little wonder that thermoset foam quickly became the dominant branch in the foam family from the outset. With regard to thermoplastics, their development has been more diversified especially from the plasticator design perspective. Table 1.6 shows the outline. In general, processes have been designed to fit the polymeric

61259_C001.indd 15

10/25/2008 12:33:08 PM

Polymeric Foams

16

Polyol Isocyanate

Pumps

Mix head

5 Axes robot

Mold

FIGURE 1.10

Reactive injection molding process.

rheology and foaming chemistry for desired cellular products. Extrusion was first adapted to accommodate the blowing agent into the polymeric melt for foaming into useful products.35 After WWII, some military expertise in machinery was made available to civilian industry and helped introduce precision processing into polymer machinery. Krobe and JSW (Japan Steel Works) have made solid contributions to extruders, single and twin screw, and injection molding systems. When blowing agent characteristics and their impact to polymer rheology were established, the extruder became more like a physical “reactor” for polymer and blowing agent to homogenize under pressure and temperature. When pressure is released at the exit, cellular structure will be developed. Unit operations such as melting, mixing, cooling, and foaming have improved tremendously in the last two decades.36–38

61259_C001.indd 16

10/25/2008 12:33:08 PM

17

History and Trends of Polymeric Foams

B A Mixing head

PU foam slab

Conveyor belt Kraft paper

FIGURE 1.11

PU board process.

Air

Chemical A Chemical B

Wall/Roof

Pressure chamber Spray nozzle

FIGURE 1.12

61259_C001.indd 17

PU spray process.

10/25/2008 12:33:09 PM

Polymeric Foams

18

TABLE 1.6 Common Foams and Their Foaming Technologies Product Flexible PU foam Rigid PU foam PU foam with skin PS foam PE and PP foam PVC foam

Technology Reactive foaming in mold Reactive foaming in place Foam in place with auxiliary liquid Foam in mold Foaming with mold temperature control Extrusion, molded bead Extrusion, molded bead, oven foaming Mold foaming, extrusion

Resin technology has also improved. Catalyst science has brought insight into morphological design to fit foaming. Branched polymers showed favorable elastic characteristics in radial expansion control, which is critical in foam extrusion.39 Injection molding was developed for irregularly-shaped plastic products. Although batch-like in nature (mold filling, setting, and opening), quite a few molds can be filled at the same time with sophisticated mold distribution design and increased injection tonnage. As a result, productivity is no poorer than the continuous extrusion process. Foaming in the mold filling exhibited constrained foaming, expansion being constrained by the mold and surface structure by mold temperature. It became particularly appealing for mass reduction and was especially attractive for expensive materials. Again, design and process have gone hand in hand to introduce new and intriguing products, with varying structure and shape, into the markets since the 1960s. When carbon dioxide was proposed as super critical blowing agent by Trexel, the benefits of cycle-time reduction were realised and embraced to improve product quality and productivity.40 As mentioned earlier, for gas/melt systems, a higher solubility at a lower temperature can save energy simply by loading polymer pellets with blowing agent at an elevated pressure and temperature and then allowing the expansion to complete below its melting temperature. Polystyrene can thus be saturated with pentane to expand in the mold into various shapes such as: cups, clamshells, and trays. Although it was possible to make thermoforming polystyrene foam sheets in the same way, early economics seemed to favor extrusion sheeting. However, when energy costs continued to increase, the mold bead process gained acceptance in more applications. In addition to energy input and subsequent removal in the extrusion process, the heat distribution and shear history posed concerns with regard to product performance. Poor heat distribution cannot only lower the

61259_C001.indd 18

10/25/2008 12:33:10 PM

19

History and Trends of Polymeric Foams

Processing

Compounding

Shear

Temperature

Foaming

Foaming

Extrusion FIGURE 1.13

Cross-linking

Extrusion versus cross-linking.

material strength, which directly diminishes the foaming window, but also causes poor property distribution determined by the weakest point. Another interesting technology was developed by shifting the processing domain as shown in Figure 1.13. Clearly, the higher the foaming temperature, the more material strength that is required. Extrusion basically characterizes the foaming process by pressure and temperature. For volatile inorganic blowing agents, it is barely possible to load enough for adequate foaming (e.g. 20 times expansion). Improvement of the material strength is necessary to sustain the volatile foaming of the inorganic blowing agent. However, when material strength is improved with more crystallinity and/or extra chemical bonding, the corresponding heat generation during processing and sharp crystallization kinetics can narrow down the processing window to improbably possible. A shearfree foaming process, cross-linked prior to foaming, was developed. The chemical blowing agent and cross-linking agent were compounded into polyolefin and were passed through a staged oven to activate the crosslinking reaction first, then chemical foaming. Cross-linked foam was thus made. Figure 1.14 shows the general cross-linking process, which is used for polyethylene. Its fine cell structure, resiliency, and thermal strength were quite unique and it became a separate brand technology. Foam nucleation is a very complex, thermodynamically-driven, but kinetically-controlled, phenomenon. When it occurs in the crystallization process, two sets of phase transformation—liquid to gas and melt to solid—are taking place in a very short time period. The co-dependent foaming process still remains mysterious in certain ways. However, in the last decade, thermoplastic foam nucleation researchers have dedicated themselves to the correlation of process design, processing conditions, and foaming design.36,41,42 Recently, a more advanced design has made use

61259_C001.indd 19

10/25/2008 12:33:10 PM

Polymeric Foams

20

Oven 1 Oven 2 Compounding

Cooling Foam roll

FIGURE 1.14

Cross-linking process.

of a volatile inorganic blowing agent in the foam extrusion process and is known as the super critical fluid (SCF) blowing agent technique.43 Although not yet fully developed, the phase separation control in the foaming stage may enable it to attain large-scale success by upgrading polymeric foams into high-tech applications. In summary, the last half-century has seen the number and quality of polymeric foam products grow and blossom. The products and technology continue to move forward hand-in-hand, so extending the polymer property spectrum. The on-going development has itself demonstrated not just survival of the fittest, but also adaptivity, flexibility, and creativity.

1.4 Performance: Properties and Applications When urethane science was established in the 1930s, the main focus was on fibers in order to duplicate the success of nylon (polyamide). During WWII, a German company, I.G. Farbenindustrie, developed lightweight and high-strength rigid PU foams for aircraft components and insulation materials for submarines and tanks.44 Technical improvements were found to be necessary in residual components and there were performance deficiencies. After nearly 80 years of foam development, the properties and

61259_C001.indd 20

10/25/2008 12:33:10 PM

History and Trends of Polymeric Foams

21

applications are deemed as the necessary measurement for the technology development, and to define the market, which success definitely brought reward to the scientific and technological efforts. In order to cover foam performance in full, it is necessary to begin with basics. Polymers possess an interesting organic long chain structure and exhibit unique properties compared with inorganic materials. As with foamed inorganic materials, foamed organic materials also extend the polymer material spectrum. Their morphology and foam structure are the main components for their properties. The emphasis here is on properties and applications. It is known that most thermodynamic properties are weight based—for example, heat capacity, enthalpy, and latent heat—and, since gas has negligent weight, most properties remain similar. However, volume-based properties and structurally-related properties appear quite different. The four areas to be addressed are physical, mechanical, thermal, and acoustical properties. 1.4.1 Physical Properties Polymers are characterized by their unique properties. In the presence of gaseous voids, gaseous cells are separated and supported by a polymeric skeleton. Since the skeleton is made of polymer, physical properties such as glass transition, melting, crystallization, and decomposition remain basically unchanged. It is worth noting that although the thermodynamic properties remain similar, kinetics may change dramatically. For instance, before reaching melting point, foam may soften much faster. This could be a serious thermally-related concern in some applications. The volume-based property, however, shows dramatic change (e.g. density). Polymeric foam density is defined as: Wp  Wg rF  ________ Vp  Vg

(1.10)

where r, W, and V denote density, weight, and volume, and subscripts F, p, and g represent foam, polymer, and gas, respectively. Since gases possess a large volume to weight ratio under normal circumstances, the more gas, the lower the density. How much gas can be dissolved into a polymer is dependent upon the solubility, whereas how soon the dissolution is complete is dependent upon the diffusivity. When the foam density is less than 1, it will automatically float on water. As long as water soaking is kept under control, the product can be used in flotation devices such as life jackets and floating docks. When cells are dispersed within a polymer, a mass reduction per unit volume is anticipated, which is important in the application of expensive engineered polymers. As long as performance needs are met, foaming is a form of mass conservation. How to meet the performance requirement

61259_C001.indd 21

10/25/2008 12:33:11 PM

Polymeric Foams

22

FIGURE 1.15

ASTM method for cell size measurement; r  (p/4)2 y  0.617y.

with the minimum weight of material, or the highest expansion, is of economic importance. Since quite a few properties are dependent upon the weakest point of the product, the challenge has been to improve the foam property distribution by improving thermal distribution in processing. Another physical aspect that can deeply affect polymer properties is cell size and distribution.45 Quite a few applications are dependent upon the internal surface area; for example, diffusion-controlled phenomena such as devolatilization,46 and sound attenuation. Surface area can change greatly at a given expansion ratio when cell size changes. In measurement terms, when foam is sliced through, not all of the cell diameter is exposed. Some cells may be sliced at a certain height of the sphere. The American Society for Testing and Materials (ASTM) have proposed a method to calculate the cell size as illustrated in Figure 1.15. This method is good for spherical cells and cells of equal size. Either or both are violated when expansion exceeds four times according to the packing proposals,47 which is very common for polymeric foam. The average cell size calculation can become quite laborious, requiring various magnification devices, and even scanning electron microscopy and 3D imaging are used when accuracy is important. Cell size and cell size distribution contribute to foam structure and therefore foam properties.48,49

1.4.2 Mechanical Properties In almost all applications, even for absorption or filling, some mechanical strength is needed. Material science has been so advanced that structure and property models can be borrowed for foam mechanical property calculation. Figure 1.16 demonstrates the skeleton model; in other words,

61259_C001.indd 22

10/25/2008 12:33:11 PM

23

History and Trends of Polymeric Foams

Cell edge

Open cell face

FIGURE 1.16

Strut—only open cell model.

an open cell structure in which compressive or tensile modulus can be established:49 Ef Ê rf ˆ = cÁ ˜ Ep Ë rp ¯

2

(1.11)

where c, subscripts f and p denote shape factor, foam, and polymer, respectively. Basically, it follows the square of density reduction. When a closed cell is considered, the cell wall stretch under disturbance becomes quite unique in terms of dissipating the disturbance. The enclosed cell functions like an elastic balloon, in which expansion can absorb disturbance. This makes closed cell foam more effective in energy absorption with a smaller reduction in modulus than open cell foam at the same expansion. The modulus dependence of density reduction becomes: Ef rf = (1 - c ) Ep rp

61259_C001.indd 23

(1.12)

10/25/2008 12:33:11 PM

Polymeric Foams

24

1

Relative Young’s modulus, E */E1

10–1 E * r* —=— E 1 r1

10–2

10–3 c = 0.8

10–4 r* E* —= — r1 E1 10–5 10–3

FIGURE 1.17

3

10–2 10–1 Relative density, r0 /r1

1

Modulus.

Figure 1.17 presents the modulus distribution at various foam densities with closed cell content as a parameter. For repeated usage, either an elastic wall and strut open cell foam or closed cell foam with stretchable cell wall is necessary. Seat cushions and shoe soles are among the applications. In general, compression and tensile strength can be deducted from the above equations. Foam rigidity and flex modulus are determined by the type of polymer and the expansion ratio, and, sometimes, the cell structure of the foam. Low mechanical strength is needed in packaging and filling types of applications. For seating, mattresses and trays, medium mechanical strength is required. For high-end structural applications, engineered polymer foam is necessary for strength and insulation. When requirements become stringent, foam composite can serve the purpose. A related area is shock absorption. Under sharp disturbance, certain foams tend to deform to absorb a good portion of the disturbance. This has proved effective in protecting a packed material from being damaged. When protection performance was improved, it became popular for athletic protection such as in helmet liners and knee and elbow pads.

61259_C001.indd 24

10/25/2008 12:33:12 PM

25

History and Trends of Polymeric Foams

1.4.3 Thermal Properties We are aware that thermal flow resistance is a material property. In steadystate composite conduction, as shown in Figure 1.18, the heat resistance for series composite and parallel composite can be expressed as:50–53 Li R  ÂRi  Â _____ Series Ali l  Âu ili Parallel

(1.13) (1.14)

where R, L, A, l, and u denote resistance, length of the component, surface area, conductivity, and volume fraction, respectively. The subscript i represents the number of components. Note that l  1/R. Foam is seen as a combination of series and parallel composite and its conduction properties were addressed in Reference 51. Lp ˆ Ê Lg lc = u g Á + ˜ l l Ë g p¯

-1

+ u p lp

Series -Parallel

(1.15)

where subscripts c, g, and p represent conduction, gas, and polymer, respectively. It is known that after a certain time, the majority of gas may be replaced by air which, in general, has a higher conductivity. However, when cell size becomes small enough, the retention of the gas over time may make a difference to thermal properties.

FIGURE 1.18 Heat conduction schematics: (a) parallel; (b) series; (c) combination of parallel and series. [Redrawn from Leach, A. G., Journal of Physics D: Applied Physics 26 (1993): 733–739.]

61259_C001.indd 25

10/25/2008 12:33:12 PM

Polymeric Foams

26

In addition to conduction, there are convection and radiation terms for foam. It is basically a structure-based calculation and can be expressed as: lF  l c  l v  l r

(1.16)

0.350

4.5

0.315

4.0

0.280

3.5

0.245

0.210

3.0 Cellulose

m2 • k w

R-value per inch

5.0

R-value per cm

F • h • fr2 Btu

where l represents the thermal conductivity and subscripts c, v, and r denote conduction, convection, and radiation, respectively. When cell size is in the micron range, the radiation term becomes non-negligible.54 In general, the higher the conductivity, the less the insulation capability. Note that conductivity increases for the foam compared to that of the polymer itself. But converting into amount of material, it turns out that much less material for foam is required to reach a given insulation value. The insulation properties were easily adapted for the purposes of heat retention and opened up contained opportunities such as cups, cup sleeves and thermal chamber insulation. Some medical systems proved to be extremely sensitive to a thermal environment and necessitated the design of special foam packaging systems to meet their needs. Its use in building applications as insulation board colder was especially important in climates where energy loss is more critical. As energy prices continue to escalate, its role in energy security is becoming increasingly important. At present, polymeric foam is now widely used for appliance doors and panels, building insulation, and thermal chambers, as energy saving has an obvious economic value. As a consequence, less energy is consumed, gas emissions are reduced, and a contribution towards global warming is reduced. Figure 1.19 illustrates the comparison between polymeric foam and fiberglass board in insulation.55

Fiberglass Polyurethane

0.175

2.5

2.0 0.0

FIGURE 1.19

61259_C001.indd 26

0.5

1.0

1.5

2.0

2.5 3.0 3.5 Density (pcf)

4.0

4.5

5.0

5.5

0.140 6.0

Foam versus fiberglass as insulation.

10/25/2008 12:33:13 PM

History and Trends of Polymeric Foams

27

Thermoforming is an extension of the thermal applications. Food trays made of polystyrene foam sheets and X-PE foam for utility cavities are typical examples of the utilization of the unique thermal strength while in a molten state. The heat-induced volume expansion and strength reduction to a soft state yet without sagging state allow it to be formed into various shapes. This property has allowed polymers to make inroads into automotive interiors such as dashboards, water shields, and liners. 1.4.4 Acoustic Properties The sound wave is considered to be an energy wave or pressure transfer. The wave tends to dampen out after transfer through a medium, like a voice through air. It can also reflect from a surface and penetrate through material by vibration. The reflective index is a material property. However, the penetration can be controlled by the structure. An open cell structure and cell wall vibration are possible mechanisms for attenuating sound waves. Low frequency sound is difficult to attenuate due to its long wavelength. In general, a large open cell PU wedge is required to reduce low frequency sound. It is far easier to deal with high frequency sound, possibly due to its short wavelength, which can be cancelled out in each reflection. Foam, due to its cellular structure, has been shown to be effective in dealing with two kinds of sound waves: airborne and contact. Airborne refers to voice-like sounds; mid-frequency speech and high-frequency engine room noise are good examples. Contact sound refers to, for example, walking on a hard floor. Floor underlay is a growing market for foam as living space becomes more restricted in areas with a rising population. The American Society for Testing and Materials (ASTM) have proposed two standards for airborne sound absorption to the next chamber (E90/ E413), and contact impact transfer to the room downstairs (E492/E989). ISO 140-8 (1997) also covered similar issues. Typical foams for flooring assembly sound absorption are presented in Figures 1.20 and 1.21. The higher numbers indicate more absorption. Table 1.7 shows the sound reduction coefficient for foams and fiberglass.56 It has become a very important construction code in relation to warehouses, gyms, studios, residences, and auditoriums. In populated areas, with living space becoming increasingly limited, privacy and the reduction of noise and disturbance from neighboring residents has become more desirable. As a result, sound absorption foam is increasingly used. A comparative chart (with fiberglass) is presented in Figure 1.22. As the years have passed by, the evolution of polymeric foam applications has continued, from flotation to furniture, to food, recreation, and electronics. Table 1.8 provides a summary. The expansion in applications is better illustrated in Figure 1.23. The application trend continues to expand although the emphasis is shifting.

61259_C001.indd 27

10/25/2008 12:33:13 PM

Polymeric Foams

28

Sound transmission loss 80 STC 54 70

Sound transmission loss (dB)

60

50

40

30

20

10

0 100 125 160 200 250 315 400 500 630 800 1000 1250 1600 2000 2500 3150 4000

One-third octave band center frequency (Hz) Sound transmission loss FIGURE 1.20

STC contour

Transmission loss in acoustics.

1.5 Regulation: Environmental and Regulatory Up until recently, the foam development picture has been clear: from scientific foundation to established technology to performance products. As we moved into the twenty-first century, increased regulation became more significant.57,58 It is, in general, intended to protect living environmental life quality. Although some have not been firmly confirmed, or never will be, but there is only one earth, it is not wise to take chances. A fine line has to be drawn between scientific evidence and political speculation. The wide consumption of polymeric foam and its use in our daily lives has raised environmental, health, and safety concerns. Since the 1990s,

61259_C001.indd 28

10/25/2008 12:33:13 PM

29

History and Trends of Polymeric Foams

Impact insulation class 30

IIC 60

70

40

60

50

50

60

40

70

30

80

20

90

10

100

0

110

Impact insulation class

Sound pressure level (dB re: 0.0002 microbar)

80

100 125 160 200 250 315 400 500 630 800 1000 1250 1600 2000 2500 3150

One-third octave band center frequency (Hz) Impact sound pressure level FIGURE 1.21

IIC contour

Impact insulation coefficient.

TABLE 1.7 Noise Reduction Coefficient (NRC) for Several Foam Products at 1” Thick from Reference 56 with the Exception of 2” Fiberglass from Owens Corning Product Sheet Material Polystyrene foam Rigid polyurethane foam Flexible polyurethane foam Phenolic foam Fiberglass board at 2” thick

61259_C001.indd 29

Density (lb/ft3) 2.5 2.0 1.9 2.0–4.0 1.0

NRC 0.18 0.32 0.6–0.7 0.5–0.75 1.0

10/25/2008 12:33:13 PM

Polymeric Foams

30

FIGURE 1.22

Fiberglass versus foam in sound absorption.

close attention has been paid to health and safety issues. Material safety data sheets (MSDS) became mandatory in the 1990s and a renewed safety and toxic list was published on the regular basis. On the environmental front, there are two main subjects. One is environmental impact; how production and decomposition impact the ecology cycle of Earth. The other is that at the present consumption rate, can Earth's resources meet demands in the long run? In short, emission, decomposition (or degradation), and resource issues must be considered. In the 1980s, satellite pictures revealed ozone thinning and the formation of holes over both Antarctic and Arctic region. There were immediate TABLE 1.8 General Thermoplastic Foam Processes Process

Equipment

Blowing Agent

Extrusion

Screw extruder

PBA and CBA

Injection molding

PBA and CBA

Molded bead

Injection and mold system Heat/cool mold

X-linking

Oven/mold

CBA

61259_C001.indd 30

PBA

Features Continuous and high speed for simple shape Semi- and continuous for specific shape Energy saving, specific shape, batch process Fine cell, smooth and flat product for PO and PVC

10/25/2008 12:33:13 PM

History and Trends of Polymeric Foams

1990

1970 Flotation medium Structural void filler Packaging Cushioning Flotation dock Insulation

Appliance insulation Structural insulation Shock-absorbent padding Carpet backing Mattresses Upholstery Food containers Thermoforming Food trays Automotive liners

31

Wind blade Structural reinforcement Decorative material Garment insulation Electronic encapsulation Electrical insulation Gaskets and filters Acoustical insulation Floral displays Toys, novelties Leather substitute Medical tape Sporting goods Protective pads

1950 FIGURE 1.23

Application tree.

concerns that a diminished ozone layer, which acts as a filter to reduce harmful radiation from the Sun, would result in UV-related damage including skin cancers, eye impairments, and crop reductions. Scientists claimed that the massive usage and emission of the very stable halogenated hydrocarbons (HC), especially chlorinated and brominated HC, had drifted into the higher atmosphere to hinder ozone formation. In 1987, the Montreal Protocol was established and signed to phase out these agents. Extra taxes were demanded for the banned chemicals to force them out production and consumption. Alternative blowing agents were formulated and introduced into the polymeric foam industry that not only helped it through the crisis without damage but also added to the technical strength of the industry. The global cooperation in reducing CFC and HCFC usage seems to have been successful via the Montreal Protocol. As illustrated in Figure 1.24,59 as CFCs are phased out, HFC and HC have increased. Nonetheless, the concerns over global warming caused by gas emissions initiated concerted efforts to regulate global warming gases. In 1997, the Kyoto Conference proposed global regulations in order to reduce the emission of greenhouse gases and so slow down global warming. Carbon dioxide became a focal point. On the one hand, it is integral to photosynthesis and the consequent production of oxygen. On the other hand, it is one of the final gases after the decomposition of organic materials. Fossil burning emission is a common phenomenon among developing countries and gasoline emission among the developed countries. Now the global challenge is how to slow down carbon dioxide emission without slowing down industrial progress. For polymeric foam, where carbon dioxide is used as a blowing agent, the direct implication of curbing carbon dioxide emission has had little impact. Instead of generating carbon dioxide, it is only

61259_C001.indd 31

10/25/2008 12:33:13 PM

Polymeric Foams

Consumption (thousands of tonnes)

32

350 300 250 Total CFCs Total HCFCs Total HFCs Total HCs Combined total

200 150 100 50 0 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 2016

Year FIGURE 1.24 Blowing agent consumption chart. (From Ashford, P. “Impact of Regulatory Developments on the Demand for Blowing Agents in Foams.” Paper presented at the Blowing Agents and Foaming Processes Conference. Rapra, Stuttgart, 2005.)

used to replace blowing agents which may possess a higher potential to cause the environment in themselves and from the production of them. However, the use of hydrocarbons, as shown in Figure 1.24, are steadily increasing in blown polymeric foam and their emission has caused serious environmental concerns. Figure 1.25 illustrates a PE foam plank emission sketch. Even after incineration, carbon dioxide and carbon monoxide are emitted. Pentane, used in the molded bead foam process, has a similar scenario as shown in Figure 1.26.60 In brief, smog is a direct concern for hydrocarbon emission, and the corresponding emission of global warming gases while hydrocarbon is destroyed in the incinerator. The social climate is a double-edged sword. On the one hand, people like to protect forests by encouraging the use of plastic in order to replace

Emissions

NOx CO2

Butane destroyed (90%)

Thermal oxidizer

Butane collected (90%)

1 Loss

Plank process

FIGURE 1.25

61259_C001.indd 32

Resin storage

>0.1 Extrusion 6

1 >0.1

Smog VOCs Butane loss (10%)

3 >0.1

Foaming 5 Handling* 2

0.8 >0.1 Aging

0.2 1

Shipping to customer

PE foam plank emission estimate.

10/25/2008 12:33:14 PM

33

History and Trends of Polymeric Foams

High atmosphere

Ozone depleting smog

Emissions Low atmosphere

Pentane destroyed (3.5-7.5%)

NOx CO2

VOCs

NOx CO2

Pentane not collected

Boiler Thermal oxidizer fuel

Pentane collected Loss Molding process

0.7-1.5

0.5-1.2

0.3-0.9

0.0-1.0

0.0-0.8

0.0-0.7

0.3-0.9

0.1-1.1

1.0

Resin 6.0 PrePre- 3.3 4.5 Operation 2.1 Storage 1.0 Finished product storage expander storage

FIGURE 1.26 PS molded bead process emission estimate. (From Kannah, K., “Pentane— Environmental and Regulatory Considerations.” Paper presented at the Blowing Agents and Foaming Processes Conference. Rapra, Frankfurt, 2007. With permission from RAPRA.)

wood-based products (e.g. foamed PP binder to replace pulp binder). On the other hand, plants or bioplastics, also known as sustainably sourced polymer, are being promoted in order to reduce the use of petroleumbased plastics. In the past decade, more and more biotechnologies have been facilitated to produce bio-based “monomers” through fermentation and/or enzymatic treatment, which, in turn, are polymerized to make polymers.61 This reduces the dependence on petroleum, and the material can be reused, as demonstrated in the cycle period (raw material to product to raw material) shown in Figure 1.1. A good example is ethanol from the fermentation of sugar cane, which can be converted to ethane. It can be used as a blowing agent for PE and PS.62 It can also be reduced to ethylene as a feedstock for polyethylene. Polylactic acid (PLA) is another example and is made from corn as shown in Figure 1.27.63,64 Fiber, film, and foam can be derived from PLA. Other common bio-based polymers are listed in Table 1.9. Another benefit associated with bio-based polymers is the use of green plants as a starting material, which can convert inorganic carbon into organic carbon during the photosynthesis reaction. In other words, the growth of agricultural feedstock can contribute to inorganic carbon conversion to organic carbon, which is accredited to the success of bio-based polymers. For example, the production of PLA required 56 MJ/kg in 2003 (about twice as high as polyethylene, 29 MJ/kg).56,61 After improvements in PLA processing, its manufacture reached carbon neutral in 2006, which means inorganic carbon emission was equal to inorganic absorption by

61259_C001.indd 33

10/25/2008 12:33:14 PM

Polymeric Foams

34

Corn field

Fresh Feed

Effluent

Fermentation

Biomass Separation System Fermeter

Polymerization

Foam processing

Blowing Agents

Expanded bead

Thermoforming

FIGURE 1.27 PLA foaming.

TABLE 1.9 Foam Property and Application Attributes Property

Attributes

Physical Mechanical

Density reduction Compressive strength High modulus Rigidity and open cell Flexibility and strength

Energy

Soft and absorption Rigid and absorption High heat retention

Thermal

Acoustics

61259_C001.indd 34

Rigid and energy saving Thermoforming Rigid and sound blocking Soft and sound absorption Semi-rigid and absorption

Applications Save material, reduce cycle time Floating device, toys, surfboards Coring component Trays, liquid retention Seat cushions, matrices, shoe soles, seals, gaskets Packaging Car bumpers Insulation boards, trays, cups, thermocontainers Appliance panels and doors Trays, dashboards Sound barrier panels Studio panels, ear phones Floor assembly

10/25/2008 12:33:14 PM

35

History and Trends of Polymeric Foams

Moisture

Energy (heat, UV, ...)

PVOH Starch Polyester PET PLA PCL

PE PP PS

Measurement ASTM D6400 6868 7021 EN 13432 ISO 17088

Degradation

FIGURE 1.28

Polymer decomposition mechanisms and test methods.

the plants Polyethylene, however, has a manufacturing emission of 2 kg CO2 emission for every kg of polyethylene. The burden is still with energy consumption for PLA manufacturing and processing, but the emphasis has enlarged the horizon. It is no longer carbon emission, but carbon and energy together in the production of a bio-based polymer. Polymer decomposition has two mechanisms as illustrated in Figure 1.28: moisture and oxidation. The relevant test methods are also included. Some, like PLA, decompose a lot faster under landfill conditions and can be categorized as degradable polymers. The ASTM established criteria for degradable and compostable materials. It is well known that esterification is a reversible reaction. In the presence of moisture and heat, decomposition can occur. Table 1.10 shows that different polyesters have different decomposition temperatures with moisture. In essence, some polyesters seem to have intriguing characteristics: made from plants, carbon neutral processing, and are compostable or degradable. Most environmental issues can be resolved using them. TABLE 1.10 Common Polyesters in the Market Polyester PET PLA PCL Copolyester*

Tg (°C)

Tm (°C)

75 55–60

High, under 30 ppm for extrusion Medium, under 300 ppm for extrusion

60

265 130–160 60

30

110–115

Low, drying not necessary for extrusion

Moisture Sensitivity

* Aliphatic–aromatic copolyester as Ecoflex™ from BASF.

61259_C001.indd 35

10/25/2008 12:33:15 PM

Polymeric Foams

36

TABLE 1.11 Common Bioplastics for Foam Polymer

Source

Polylactic acid (PLA) Starch Polyvinylalcohol (PVOH) Poly caprolacton (PCL)

Corn Corn, potato, wheat Vinyl acetate monomer (VAM) Crude oil

Starch-based foams and polyvinylalcohol (PVOH) foams are basically water soluble. As long as moisture can be properly maintained in processing as well as in usage, they can then disintegrate in water after disposal. Table 1.11 shows common degradable foam. It is worth noting that when foam is degraded or has disappeared from sight, it hastens the reduction to a basic unit process but does not solve all the problems. It could be a major concern to recycle, reuse, and ensure durable usage of the products. When dissolved, it may generate global warming byproducts. A summary for common disposal is presented in Table 1.12. It should be pointed out that the advances in the recycling and reuse of common polymers in the 1990s should be continued. The municipal sorting and segregation program is worth further support from every individual and local government. A plausible route appears to be the use of green plants to produce the monomer, during which inorganic carbon can be converted to organic carbon. The next step is an energy efficient process to make polymer and foam. The final products can, for instance, save more energy or conserve material to become “less for more” components in the whole cycle. Foam has the potential, in terms of saving energy and improvement of performance/ weight ratio, to change its role from commodity to environmental solution. The insulation benefits of foam are presented in Table 1.13. It shows a great energy security advantage. The ideal characteristics for environmental foam are summarized in Table 1.14. Nonetheless, polymeric foam is not a simple manufacturing and consuming issue any more. Since its volume is so significant, it not only affects our living standard, but also impacts on the ecology cycle. A bigger picture is TABLE 1.12 Analysis of Common Disposal Methods Methods

61259_C001.indd 36

Advantage

Incineration Landfill

Generate energy Low cost

Water soluble Biodegradable

Easy Quick volume reduction

Disadvantage Produce global warming gas Polymer takes time to decompose, need huge space Expensive Negate recycle and reuse

10/25/2008 12:33:15 PM

37

History and Trends of Polymeric Foams

TABLE 1.13 Insulation Foam Advantages Direct Advantage

Indirect Advantage

Heat retention Reduce CO2 and NOx emission by less boiler usage Sound absorption Reduce heating fluid usage to warm houses Save energy or higher energy efficiency Save the material Less prone to mold and fungus growth Improve living quality Rigidity as part of the appliance structure

necessary to define its role. Few relevant questions are: what is the source material for polymer, how is it processed, what is the life time of the final product and its impact index toward enriching our living by performance or by replacing other products, what is the potential in reuse and recycle, how to dispose it, how it degrades. As the first step, the energy index and carbon status in making, consuming, and degrading into raw components should be established and compared with other materials, especially the materials replaced by foam with other materials on the same basis; by weight and by cost. We are all familiar with the second law of thermodynamics: the output energy is less than the input in all systems. The continued challenge, not just for polymeric foam but for every industry, will be higher efficiency and effective inorganic to organic carbon processing, as well as decomposition back to inorganic material to complete the life cycle. The intriguing question is: what follows after the environmental regulation for polymeric foam? We know foam plays an important role in our civilization, making our lives much more convenient, although a lot of problems remain to be solved. It is clear that environmental conditions tend to have more impact on human living conditions. Issues such as the ozone layer, carbon concentration, earthquakes, and climate-induced famine will not go away. How foam plays a more active role in those pending topics, yet continues to enrich human convenience, is a challenge to the global foam society. TABLE 1.14 Ideal Environmental Polymeric Foam Characteristics Items for Foam Raw material Processing Foam product Disposal

61259_C001.indd 37

Environmental Benefits From green plants to negate carbon credits Energy efficiency (one step extrusion process) Using environmental friendly blowing agent Save energy in insulation Favorable performance/weight ratio Recycle and reuse Water soluble and degradable

10/25/2008 12:33:15 PM

38

Polymeric Foams

References 1. Meikle, J. L. American Plastic: A Cultural History. Rutgers University Press, New Brunswick, NJ, 1995. 2. Frisch, K. C. “Historical developments of polyurethanes.” In 60 Years of Polyurethanes, ed. J. E. Kresta and E. W. Eldred. Technomic, Lancaster, PA, 1998. 3. Ndiaye, P. A. Nylon and Bombs: Dupont and the March of Modern America. Belin, Paris, 2007. 4. Franklin, W. E., Franklin, B. S., and Franklin, M. F. “Recycling of materials in municipal solid waste—A view to the year 2000.” Paper presented at the Society of Plastics Engineers Annual Technical Conference (ANTEC), preprint 3504-08. Society of Plastics Engineers, Boston, MA, 1995. 5. Saunders, J. H. and Frisch, K. C. Polyurethanes: Chemistry and Technology, Vols I and II. Interscience-Wiley, New York, 1962, 1964. 6. Midgley, T., Henne, A. L., and McNary, R. R. Manufacture of Aliphatic Fluoro Compounds. US Patent 1,930,129, 1933. 7. Colombo, R. Italian Patent 370,578, assigned to LMP, 1939. 8. Martelli, F. Twin-Screw Extruders. Van Nostrand Reinhold, New York, 1983. 9. Kojima, J., Takada, T., and Jinno, F. “Thin microcellular plastics sheet incorporating designed foaming patterns made by photochemical foaming technology.” Journal of Cellular Plastics 43 (2007): 103–109. 10. Hampson, R. F., Kurylo, M. J., and Sander, S. P. “Reaction rate constants for selected HCFCs and HFCs with OH and O(1D).” In Scientific Assessment of Stratospheric Ozone: 1989 Volume II. World Meteorological Organization, Global Ozone Research and Monitoring Project, Report No. 20, 1990. 11. Morita, K., Uchiki, K., and Shinoda, H. High Polymer Network. US Patent 5,223,546, assigned to Mitsui Toatsu Chem., 1993. 12. Lee, S. T., Kareko, L., and Jun, J. “Study of thermoplastic PLA foam extrusion.” Journal of Cellular Plastics 44 (2008): 293–305. 13. Reignier, J., Gendron, R., and Champagne, M. “Extrusion foaming of poly(lactic acid) blown with CO2.” Paper presented at the Society of Plastics Engineers Annual Technical Conference (ANTEC), spon. Society of Plastics Engineers, Charlotte, NC, 2006. 14. Chen, C., Cheng, M. L., Jean, Y. C., Lee, J., and Yang, J. “Effect of CO2 exposure on free volumes in polystyrene studied by positron annihilation spectroscopy.” Journal of Polymer Science B: Polymer Physics 46 (2008): 388–405. 15. Blander, M. and Katz, J. L. “Bubble nucleation in liquids.” American Institute of Chemical Engineers Journal 21 (1975): 833–848. 16. Flory, P. J. Principles of Polymesr Chemistry. Cornell University Press, Ithaca, NY, 1953. 17. Muschiatti, L. C. US Patents 5,229,432 and 5,391,582 assigned to E. I. DuPont de Nemours and Co., 1993, 1995. 18. Al Ghatta, H. A. K., Severini, T., and Astarita, L. US Patents 5,362,763 and 4,422,381, assigned to M. & G. Richerche S.p.A., 1994, 1995. 19. Bayer, O., Rinke, H., Siefken, W., Ortner, L., and Schild, H. German Patent 728,981, assigned to I. G. Farbenindustrie, 1942. 20. Bayer, O., Angew. Chem., A59 (1947): 257. 21. Benning, C. J. “Polyurethane foam.” In Plastics Foams I: Chemistry and Physics of Foam Formation. Interscience-Wiley, New York, 1969.

61259_C001.indd 38

10/25/2008 12:33:15 PM

History and Trends of Polymeric Foams

39

22. Hoff, G. P., Eicker, D. B., and Perlon, U. Polyurethanes at I. G. Farben, Boringen, Augsburg P. B. Report 1122, September 1945. 23. Margedant, J. A. Freon-Blown Rigid Foams. E. I. DuPont de Nemours & Inc., Elastomers Division Bulletin HR-31, July 1958 (CFC-11). 24. Munters, G. and Tandberg, J. G. US Patent 2,023,204, 1935. 25. Vachon, V. “Researches for alternative blowing agents.” In Thermoplastic Foam Processing, ed. R. Gendron, Taylor and Francis, Boca Raton, FL, 2004. 26. Areerat, S., “Solubility, diffusion coefficient and viscosity in polymer/CO2 system.” Research thesis, Department of Chemical Engineering, Kyoto University, Kyoto, Japan, 2002. 27. Rubens, L. C., Griffin, J. D., and Urchick, D. US Patent 3,067,147, assigned to Dow Chemical, 1962. 28. 40th Anniversary for Japan Styrene Paper, JSP, Tokyo, 2004. 29. Wirtz, H. Chapter 7.1. In Polyurethane Handbook, ed. G. Oertel, Hanser, Munich, 1985. 30. Schlack, P. “Preparation process of polyesters containing high molecular weight amide groups.” French Patent 869,243, 1940. 31. Cellular Plastics. National Academy of Science, Washington DC, 1967, p. 218 and 230. 32. Barito, R. W. and Eastman, W. O., “Plastics and elastomeric foams.” Chapter 7 of Handbook of Plastics and Elastomers, ed. C. A. Harper, McGraw-Hill, New York, 1975. 33. Ashida, K. “Polyisocyanurate foams.” In Polymeric foams, ed. D. Klempner and K. Frisch. Hanser, Munich, 1991. 34. Mao, J., Chang, J., Chen, Y., and Fang, D. “Review of phenolic foam.” Chemical Industry Engineering 15 (1998): 38–43. 35. Kennedy, R. N., “Extruded expanded polystyrene.” Sec. XII of Handbook of Foamed Plastics, ed. R. J. Bender, Lake, Libertyville, IL, 1965. 36. Collins, F. H. “Controlled density polystyrene foam extrusion.” Society of Petroleum Engineering Journal 705 (1960). 37. Chung, C. I. Extrusion of Polymers. Hanser, Munich, 2000. 38. Rauwendaal, C. Polymer Mixing. Hanser, Munich, 1998. 39. Walczak, K., Gupta, M., Koppi, K. A., Dooley, J., and Spalding, M. A. “Elongational viscosity of LDPEs and polystyrenes using entrance loss data.” Polymer Engineering Science 48 (2008): 223–232. 40. Okamoto, K. T. “General Description of the MuCell Process.” Chapter 2 in Microcellular Processing. Hanser, Munich, 2003. 41. Lee, S. T., Park, C. B., and Ramesh, N. S. Polymeric Foams; Science and Technology. Taylor and Francis, Boca Raton, FL, 2006. 42. Lee, L. J., Wingert, M. J., Guo, Z., Shen, J., Han, X., Tomasko, D., and Koelling, K. W. “Foaming using a polystyrene/poly(methyl methacrylate) blend and nanocomposites.” Paper presented at the American Institute of Chemical Engineers Annual Meeting, 139d, San Francisco, 2006. 43. Suh, N. P. “Microcellular plastics.” In Innovation in Polymer Processing, ed. J. F. Stevenson. Hanser, Munich, 1996. 44. Reed, D. Urethane Technologies, March, 1987. 45. Bureau, M. N. “Relationship Between Morphology and Mechanical Properties in Thermoplastic Foams.” Chapter 6 of Thermoplastic Foam Processing, ed. R. Gendron. Taylor and Francis, Boca Raton, FL, 2005.

61259_C001.indd 39

10/25/2008 12:33:15 PM

40

Polymeric Foams

46. Biesenberger, J. A. and Todd, D. “Section I: Fundamentals in devolatilization of polymers.” In Devolatilization of Polymers, ed. J. A. Biesenberger. Hanser, Munich, 1983. 47. Maron, S. H. and Lando, J. B. “The Solid State.” Chapter 2 of Fundamentals of Physical Chemistry. Macmillan, New York, 1974. 48. Hamza, R., Zhang, X., Macosko, C., Stevens, R., and Listemann, M. “Imaging open-cell polyurethane foam via confocal microscopy.” In Polymeric Foams, ed. K. C. Khemani. ACS Symposium Series 669, American Chemical Society, Washington DC, 1997. 49. Gibson, L. J. and Ashby, M. F. Cellular Solids. Pergamon, Oxford, London, 1988. 50. Welty, J. R., Wicks, C. E., and Wilson, R. E. “Fundamentals of heat transfer.” Fundamentals of Momentum, Heat and Mass Transfer. John Wiley & Sons, New York, 1973. 51. Leach, A. G., “The thermal conductivity of foams I: Models for heat conduction.” Journal of Physics D: Applied Physics 26 (1993): 733–739. 52. Glicksman, L. “Foams and cellular materials: Thermal and mechanical properties.” Massachusetts Institute of Technology, summer course note, Cambridge, MA, 1992. 53. Ahern, A., Verbist, G., Weaire, D., Phelan, R., and Fleurent, H. “The conductivity of foams: a generalization of the electrical and the thermal case.” Colloids And Surfaces A: Physicochemical and Engineering Aspects 263 (2005): 275–279. 54. Zhu, Zhengjin, “Modeling of foam expansion and collapse of extruded filament foam through cell-to-cell diffusion.” PhD thesis, Department of Mechanical and Industrial Engineering, University of Toronto, 2007. 55. Ned Nisson, J. D. The Fiber War: Loose Insulation for Houses. Cutter Information Corp., Arlington, MA, 1997. 56. “Plastic foams.” Material Design Engineering 236 (1966). 57. Narayan, R. “Fundamental principles and perspectives of biodegradable/ compostable plastics and bioplastics.” Paper presented at the International Symposium on Polymers and Environment: Emerging Technologies and Science. BioEnvironmental Polymer Society (BEPS), Vancouver, October 2007. 58. Ashford, P. “Impact of regulatory developments on the demand for blowing agents in foams.” Paper presented at the Blowing Agents and Foaming Processes Conference. Rapra, Stuttgart, 2005. 59. Ashford, P. “Climate change, energy security and thermal efficiency—Are foams the answer?” Foams 2006. Society of Plastics Engineers, Chicago, 2006. 60. Kannah, K., “Pentane—Environmental and regulatory considerations.” Paper presented at the Blowing Agents and Foaming Processes Conference. Rapra, Frankfurt, 2007. 61. Huneault, M. A., “What does the bio really mean in bioplastics? A personal viewpoint on biobased plastics.” Paper presented at the Polymer Processing Society Annual Meeting, Salvador, Brazil, 2007. 62. Lee, S. T., “Expandable composition and process for extruded thermoplastic foams”. US Patent 5,462,974, assigned to Sealed Air Corp., 1995. 63. Bopp, R. C. and Whelan, J. “Method for producing semicrystalline polylactic acid articles.” US Patent Pub. 2003/0038405, 2003. 64. Ma, P. “Biomimetic materials for tissue engineering”, Advanced Drug Delivery Reviews 60 (2008): 184–198.

61259_C001.indd 40

10/25/2008 12:33:15 PM

2 Development of Endothermic Chemical Foaming/Nucleation Agents and Its Processes

Dieter Scholz

CONTENTS 2.1 History ............................................................................................. 2.1.1 How Did Foam Come into Use? ........................................... 2.1.2 Nucleating and Foaming (Blowing) Agents ....................... 2.2 Physical Foaming and its Nucleation ............................................ 2.2.1 Direct Gassed Foam Extrusion............................................. 2.2.1.1 Nucleation in PBA Extrusion.................................... 2.2.2 Injection Molding Direct Gassing........................................ 2.2.2.1 General Remarks ....................................................... 2.2.2.2 Problems and Solutions ............................................ 2.3 Chemical Blowing Agents (CBAs) ................................................. 2.3.1 Injection-Molded Foams ........................................................ 2.3.1.1 Methods ....................................................................... 2.3.1.2 Role of Endothermics versus Exothermics ............. 2.3.2 Extrusion ................................................................................. 2.3.2.1 Chemically Foamed Extrudates .............................. 2.3.2.2 General Aspects of Foam Extrusion ....................... 2.3.2.3 Extruder Overview: Screw, Melt Filters, and Dies ...................................................................... 2.3.2.4 Coextrusion ................................................................

61259_C002.indd 41

42 42 43 44 45 48 50 50 50 51 51 52 53 57 57 58 59 61

11/5/2008 4:26:12 PM

Polymeric Foams

42

2.3.3 Other Foam Processes ........................................................... 2.3.3.1 Foam Blow Molding with Endothermics .............. 2.3.3.2 Rotational Molding Aspects of Endothermic Blowing Agents .......................................................... 2.4 Summary ........................................................................................... 2.5 Abbreviations.................................................................................... References .................................................................................................

62 62 62 63 64 65

2.1 History 2.1.1 How Did Foam Come into Use? Gases that have been chemically liberated under heat have been found useful in generating polymeric cellular structure through common processes such as extrusion and injection molding. In the past 40 years or so, quite a few technical hurdles were successfully overcome to make the technology known as chemical blowing agent (CBA) very popular in the global polymeric foam industry. It simply became indispensable in cross-linked PE and PP foam, PVC foam, structural and profile foam, and woodplastic manufacturing. The application list continues to expand. Foam history dates as far back as to when humans first discovered techniques to make material lighter, softer and sponge-like for specific purposes. They came across natural foam structures like bones and porous minerals, and discovered fermentation to prepare food. Foams are modified substrates and belong to dispersed systems where non-dissolving properties are used with a mix of material properties in different aggregate states (gaseous, liquid, solid). Foams are dispersed systems that require gases and solid/liquid components for it to form. Foam can consist of open, closed, and combined cell structures. Gases can be introduced directly or released from other material by heat and/or chemical/biological decomposition or reaction. The polymeric universe started more than 100 years ago with trials using gases like carbon dioxide mainly resulting from the decomposition of traditional kind of baking powders like sodium carbonates with acids (mainly fruit acids like tartaric acid or citric acid). It was easy to achieve and produce these materials. The baking powder industry had the advantage based on synthetic NaHCO3 provided by the Ernest Solvay Ammonia Process,1 which was first developed in 1861/1885. With this process large scale production was possible, making an affordable source of carbonate/ bicarbonate available. In relevant plastics literature, very early gases like air, nitrogen, carbon dioxide and decomposing chemicals like carbonates (either by

61259_C002.indd 42

11/5/2008 4:26:12 PM

Development of Endothermic Chemical Foaming/Nucleation Agents

43

itself or in combination with mono- and multi-basic acids, some of which decompose), are named in a large number of publications, patents, and patent applications. The real breakthrough into expanded or foamed polymeric substrates happened after the Second World War. Industrial production and use of foamed materials started in the 1950s in rubber and plastics processing.2–4 Various techniques came into use by utilizing gases and decomposable substances and most are still in foam production today as additives. 2.1.2 Nucleating and Foaming (Blowing) Agents Though this group is rather small compared with the other additive groups, its growth surpasses the overall growth of the rest of the field.5,6 There are two major groups: 1. physical blowing agents (PBA); mainly gases or liquids; 2. chemical blowing agents (CBA); release gas by decomposition reaction, and nucleants. In general, nucleating agents are necessary in the polymeric foaming process for even dispersion in the gaseous phase, and in turn, bubble nucleation. The common nucleating agents are porous minerals (e.g. talc), whose surface becomes the residence for blowing agents. Cell size distribution improves significantly. In some processes, such as foaming with volatile inorganic gas, adding nucleating agent is not necessary. Since most physical blowing agents are not as volatile as inorganic blowing agents, the latter can be used as nucleating agents. The inorganic can be added through an injection system for PBA or as CBA. The main benefit is there is no residue left in the foam, and reuse or recycling has minimum impact on nucleation. In conventional foam extrusion, PBA is generally favored over CBA due to expansion ratio and economics. However, CBA can generate fine cell structures by liberating inorganic gases. From an environmental and emission perspective, CBA is heavily favored. When hydrocarbon is used for low-density foam, safety devices and inventory for aging are vital, which again makes CBA attractive. Moreover, CBA can be added as a nucleating agent in PBA foaming, which, when compared with talc, leaves no residue, making reuse much easier in nucleation control. However, the main hurdle is the solubility of inorganic gas which is much less than that of the organic gas in common thermoplastic polymers, and when translated to economics, presents itself as CBA’s major challenge. There are two kinds of CBA: endothermic and exothermic, depending upon its reaction thermogram. It is well known that polymeric material strength plays a critical role in holding the dynamic foam nucleation and growth, especially in volatile inorganic foaming agent. Although the heat liberated or absorbed is not so significant when compared with

61259_C002.indd 43

11/5/2008 4:26:12 PM

Polymeric Foams

44

the processing heat, the distribution of the reaction makes it impossible to ignore in certain applications. The absorbed heat has the capacity to cool down the surrounding polymeric melt to stabilize it for better cell integrity. The quality benefit can be easily transferred to performance benefit. It is no wonder that endothermic CBA has carved itself a niche among applications.7

2.2 Physical Foaming and its Nucleation It is understood that an aid can be used to improve gas distribution in the relevant substrate. It can be an inert substance such as talc or another material or a decomposable (reactive) ingredient such as a chemical foaming agent. Endothermal systems based on fruit acids (presented in Table 2.1) such as carboxylic, polycarboxylic and polyhydroxy carboxylic acids (preferably citric-based), and carbonates (preferably bicarbonates) tend to have the best performance.8 Many acids are applicable in endothermic foaming agents as well. The criteria of use are: efficiency, pricing, decomposition products, reactivity (discoloration, physiological aspects, smell, remaining substances in the substrate), availability, process condition, and stability. Organic fruit/food

TABLE 2.1 Relevant Fruit/Food Acids Acid/Acidic Salts Tartaric acid Monopotassium tartrate Citric acid Monosodium citrate Acid citric esters Gluconic acid Gluconates Malic acid Fumaric acid Succinic acid Oxalic acid Ascorbic acid Glutaric acid Lactic acid (pure) Calcium lactate

Data (mp, decomp. etc.)/Comments mp 170°C mp 250°C/decomp. 250–300°C mp 153°C/decomp. 153–170°C mp/decomp. 180–210°C7 210–240°C (decomp.) liquid/pastes8,9 mp 125–126°C No information mp c. 100°C mp ? 276–287 (300–303)°C mp 185–190°C/anhydride: 235°C (?) mp 189°C sublimation/decomp./toxic!! mp 199°C/decomp. mp 97°C mp 18–26°C No information

mp, melting point; decomp., decomposition.

61259_C002.indd 44

11/5/2008 4:26:12 PM

Development of Endothermic Chemical Foaming/Nucleation Agents

45

acids, non-fruit inorganic acids, and carbonates have been found useful as nucleating agents in physical foaming systems. They can also be used as chemical blowing agents. Relevant food/fruit acids are given in Table 2.1. Relevant non-food/inorganic acids include: • boric acid; • acidic phosphates (e.g. monocalcium phosphate, sodium acid pyrophosphate, disodium pyrophosphate, sodium aluminum phosphate); • acidic sulfate (e.g. sodium aluminum sulfate). • Relevant carbonates used as CBA include: – NH4HCO3 – Ca(HCO3)2 – KHCO3 – NaHCO3 – NaAl(OH)2CO3 10,11 – other alkali and earth-alkali bicarbonates or carbonates.12 Exothermals like azodicarbonamide (ADC) are used in non-food applications like cable insulation together with nitrogen as blowing gas. Hansen and Martin13 reported about nucleation’s tendency to generate “hot spots” in the substrate when creating fine cell structure. ADC is used in cable applications today. It may be doubtful if the hot spots or decomposition products nucleate at all due to the fact that when energy is released, relative to the energy content of a polymer melt, it is extremely small in practical technical applications. It is worth pointing out that exothermic ADC has a better foaming efficiency (220 cm3/g), than the endothermic citric acid/sodium bicarbonate (120 cm3/g). The former liberated nitrogen-based gases and the latter carbon oxides. When the polymeric strength is strong enough, such as after cross-linking, ADC became a popular foaming agent in low-density foam. 2.2.1 Direct Gassed Foam Extrusion Gases are introduced directly into the machines (extruder or similar) via pressure in the melt, distributed and mixed before leaving the machine exit (i.e. die, nozzle, etc.) as a blend. There are a number of gases found their way for industrial use. The newly introduced gas acts like a temporary “plasticizer” and reduces the melt viscosity drastically. This allows lower processing temperatures. Major groups include gases from the atmosphere like nitrogen, carbon dioxide, and even just air. Further gases from gasoline production— the alkanes like propane, butane, pentane, and others—or blends and halogenated alkanes were used. They are the most debated materials today.

61259_C002.indd 45

11/5/2008 4:26:12 PM

Polymeric Foams

46

For improved distribution of the gas in substrates, nucleating agents came into use. These included any kind of fine minerals (talc, silica, calcium carbonate, pigments, etc.) used to chemically produce carbon dioxide, blends of carbonates (sodium, calcium, magnesium, etc.) with organic acids (citric, tartaric, numeric, etc.) or inorganic acids (boric acid, acidic phosphates like monocalcium phosphate, sodium acid pyrophosphate, disodium pyrophosphate, sodium aluminum phosphate, etc.). Nucleating agents found their way into various formulations, patents, and publications.16 Desired properties of a gas are as follows14,15: • • • • • • • • • •

Environmentally acceptable Non-flammable Non-toxic Adequate solubility Non-reactive Low vapor thermal conductivity Low diffusion rate Appropriate latent/specific heat Low molecular weight Low cost.

The idea was to get the acid and carbonates to react with each other and form tiny spots from the gas and other decomposition or reaction components to create an environment for better gas distribution and foaming, resulting in a finer cell structure and hopefully lower densities. Furthermore it was important to develop highly efficient and affordable systems. Very soon the blends from sodium bicarbonate and citric acid components found their way into various processes. The major applications in the 1950s16–19 were extrusion processes of general purpose polystyrene (PS/GPPS) called direct gassing and the extrusion of polystyrene beads, which contained mainly butane already as described.18 Originally the direct gassing of PS was done by using halogenated hydrocarbons until most of them were banned or substituted by low- or non-ozone depletion materials or plain hydrocarbons. Later on, the direct gassing of low-density polyethylene (LDPE) proved to be a similar process to PS. Today, atmospheric gases have made their way into the relevant industry.20 For a list of those relevant gases refer to Table 2.2. The major nucleants still in use today consist of special grades of talc and balanced endothermic systems based on the chemistry described previously. Due to the moisture pickup from the atmosphere, Boehringer Ingelheim has sold hydrophobic coated or a kind of encapsulated citric acid to the “foam” industry since about 1967.21 The first ready-to-use endothermic system for nucleation (and expansion) was developed by

61259_C002.indd 46

11/5/2008 4:26:12 PM

61259_C002.indd 47

N2 CO2 H2O C2H5OH C3H8 C4H10 C4H10 C5H12 C5H12 C5H10 CHF2Cl CF2ClCH3 CHF2CH3 CH2FCF3 CHF2CH2CF3 CF3CH2CF2CH3 CFCl3 CF2Cl2 CF2ClCF2Cl CFCl2CF2Cl CH3OCH2OCH3

Formula 28.01 44.01 18.02 46.07 44.10 58.12 58.12 72.15 72.15 70.13 86.47 100.49 66.05 102.03 134.05 148.08 137.37 120.91 187.37 170.92 76.09

Molecular Weight (g/ mol) 195.81 78 100 78.35 42.05 0.15 11.15 36.05 27.95 49.25 40.78 9.95 24.7 26.55 15.35 40.2 23.75 30.15 47.85 3.75 41.85

Boiling Point (°C) — 829.62 0.339 0.849 121.495 30.098 43.628 8.202 11.122 5.020 132.093 42.06 84.697 82.53 17.98 6.28 12.860 82.24 5.49 26.665 6.267

Vapour Pressure at 20°C (psi) 0.807 1.52 1 0.79 0.507 0.6 0.562 0.626 0.62 0.745 1.194 1.12 0.9 1.21 1.35 1.27 1.46 1.29 1.55 1.44 0.8669

Liquid Specific Gravity (Water at 20°C ⫽ 1.0) None None None 3.3–19 2.1–9.5 1.9–8.5 1.8–9.5 1.4–7.8 1.3–8.3 1.1–8.7 None 6.0–18.0 5.1–17.1 None None 3.8–13.3 None None None None 2.2–13.8

Flammability LEL–UEL (Vol. % in Air) — — — — — — — — — — 0.05 0.07 — — — — 1 1 1 1 —

ODP

1 — — — — — — — — 1810 2310 140 1300 950 890 4750 10890 6130 10040 —

—

GWP

ODP, ozone depletion potential with ODP for CFC-11 defined as 1.0.; GWP, global warming potential with GWP for CO2 defined as 1.0; LEL, lower explosive limit; UEL, upper explosive limit.

Nitrogen Carbon dioxide Water Ethanol Propane n-butane i-butane n-pentane i-pentane Cyclopentane HCFC-22 HCFC-142b HFC-152a HFC-134a HFC-245fa HFC-365mfc CFC-11 CFC-12 CFC-113 CFC-114 Methylal

Foaming Agent

Relevant Gases

TABLE 2.2

Development of Endothermic Chemical Foaming/Nucleation Agents 47

11/5/2008 4:26:12 PM

48

Polymeric Foams

C.H. Boehringer Sohn in Ingelheim, Germany and came onto the market end of 1972 under the name Hydrocerol® (ideas and realization: Elmar Bisle PhD. and D. Scholz). Even having other formulations like acid and bicarbonates together in one blend, Boehringer could guarantee a shelf life of one year. The first ready-to-use active nucleant and endothermic blowing agent was born. Hydrocerol was registered in July 1972. Clariant bought the technology in 1999. Later some “reinventions” came onto the market in England, Japan and in the 1980s/1990s in North America and other areas. Today this concept is a well-known and widely accepted commodity with growing markets and additional modifications.22,23

2.2.1.1 Nucleation in PBA Extrusion The presence of nucleating agent evidently makes the foaming more heterogeneous than homogeneous in nature. As a result, a higher nucleation rate occurs, and cells become much finer. Nucleation examples with active (decomposable/reactive) materials can be found in various references.6,14,24 The LDPE profile made with HCFC-142b and butane are shown in Figures 2.1 and 2.2. In the past, the fastest and easiest information regarding general behavior was for GPPS and gases like CFC-11/CFC-12 (see Table 2.2) or alkanes. Here, drastic results were obtained by the various changes; for example, with citric acid and sodium bicarbonate.17 Parameter investigation helped process improvement significantly. A vast amount of research

FIGURE 2.1 LDPE profile made with HCFC-142b, with and without CBA nucleating agent.

61259_C002.indd 48

11/5/2008 4:26:13 PM

Development of Endothermic Chemical Foaming/Nucleation Agents

49

1.4

Cell size (mm)

1.2 1.0 0.8

GPPS LDPE

0.6 0.4 0.2 0 0.125

0.250 0.300 0.500 Loading (%)

0.750

1.000

FIGURE 2.2 Cell size effects at different loadings of CBA for LDPE and PS foam extrusion with butane blowing agent.

was dedicated to the formula optimization. Obscure formulations were reported in the literature, such as talc, silica/silicates, solvents like acetone and ethanol with doubtful final results. The “equimolar” citric acid and sodium bicarbonate was proved to be effective in generating fine cell structure as illustrated in Figures 2.1 and 2.2. As soon as the equimolar ratio is disturbed, the cell size increases as suggested in Figure 2.3. Tests were carried out where only the molar ratio was varied while the processing temperature, nucleant, and other additives such as colorants, were kept constant. After the test was completed, the foam product was kept in storage for a few days before the cell size

NaHCO3 100% mm

General aspects to formulate active nucleants

Citric comp 100% mm

Lowest possible cell size under constant Area of equimolar Optimizing formulation for nucleation FIGURE 2.3 Equimolar cell optimization results, the oval area represents the finest cell size.

61259_C002.indd 49

11/5/2008 4:26:13 PM

Polymeric Foams

50

and structure were measured. It was found that the reproducibility in cell structure and its variation during storage were very consistent. Another aspect is the combined use of equimolar systems and talc. There is a synergism observed and published.24 In GPPS foam extrusion, finer cells, lower density and faster outgassing were reported. Another advantage is the reduced storage time between production (extrusion) and thermoforming, since the foam sheet seemed “stiffer.” 2.2.2 Injection Molding Direct Gassing 2.2.2.1 General Remarks Foam injection molding is often referred to as structural foam molding25–27 due to the formation of an integral skin. Injection molding machines are built using two main concepts. Today’s standard is a screw which also acts like a piston for pushing. The classical concept was using a piston, as in rubber processing, but feeding with an extruder into an accumulator and shooting with a piston-operated system into the mold. This helps introduce gases in the molten polymer through the barrel of the extruder to make foamed parts as shown in Figure 2.4.26 This process was known as UCC (Union Carbide Corporation process). 2.2.2.2 Problems and Solutions When nucleants started to be used in extrusion, the question became: Why not introduce nucleants into the injection molding of structural foams? Early results in testing this concept in the late 1970s in Europe and early 1980s in North America gave very promising results and the endothermic materials naturally found their way into this application with some advantages: 1. The gas was more uniformly distributed in the melt and less gas loss through the hopper was observed. Therefore, the gas pressure could be lowered to receive the same results. 2. The skin appeared better and somewhat smoother, and the cell structure became more uniform with less dense (weight) parts. Thus, faster cycles could be realized. 3. Because this application used aluminum-based molds that were used in lower pressure applications, it gave longer running/life times for the molds and, in general, generated more parts per shift. 4. When either the nucleant was overdosed or the wrong formulation was used, or both, corrosion of the aluminum mold would occur after some time. Today these problems have been solved and endothermic nucleants generally arouse no corrosion concerns.

61259_C002.indd 50

11/5/2008 4:26:13 PM

Development of Endothermic Chemical Foaming/Nucleation Agents

51

FIGURE 2.4 Union Carbide Corporation process.

This process allows the production of rather huge parts with high weight at reasonable costs and a good lifespan of the molds due to the fact that the lower injection pressure is aluminum based and not steel. The active endonucleants have proven useful in this process. Today, common machine manufacturers include Wilmington, Johnson Controls, and Battenfeld.

2.3 Chemical Blowing Agents (CBAs) The CBAs were and are widely distributed all over major processing technologies, they can be used for all thermoplastic resins, and play a major role in the foamed plastics industry.2,18 The endothermals found a good place in practice due to certain advantages over the classical ADC (azodicarbonamide). There are applications like cross-linked polyethylene and EVA (ethylene-vinyl acetate) which still is the major market for ADC. On the other hand there is a synergistic effect of both systems in various areas to improve properties and processes.4–6,16,28,29 Major groups of foaming/blowing agents are given in Table 2.3. 2.3.1 Injection-Molded Foams When the endothermics were introduced by Boehringer Ingelheim in the early 1970s, the standard foaming or blowing agent was azodicarbonamide (ADC) or other exothermics with similar chemistry such as sulfohydrazides, OBSH, 5-PT, and so on (see Table 2.3). They were well established due to the availability, good gas yield, ease of blending, decomposition range,

61259_C002.indd 51

11/5/2008 4:26:13 PM

Polymeric Foams

52

TABLE 2.3 Major Properties of Chemical Blowing Agents Product ADC (ADCA) ADC (ADCA) activated DNPT THT TSH OBSH TSSC 5-PT NaHCO3 NaHCO3/citric comp.

Decomp. Range (°C)

Appr. Gas Evolution (ml/g)

Main Gases

200–215 140– 215 190–200 245–285 105–110 155–165 225–235 240–250 110–150 130–230

220 130–220 190–200 180–210 115 110–125 120–140 190–210 160–190 110–180

CO, CO2, NH3 N2, CO, CO2, NH3 N2, NH3, CH2O N2, NH3 N2, H2O N2, H2O N2, CO2, NH3 N2 CO2, H2O CO2, H2O

and handling. The disadvantages of exothermics in most cases were smell, discoloration, corrosion, and non-food grades. The endothermics were based on raw materials which matched the requirements for food or food additives (according to most regulations in Europe, America, and Asia). The endothermics filled a gap for food packaging, toys, and pharmaceutical and cosmetics applications.31 There was no smell, no discoloration, and in most substrates an even finer cell structure and smoother surface could be obtained. Painting or decoration could be done “in-line,” without long outgassing or storage time. The total liberated gas is much less than that of ADC, but the distribution is excellent and smaller gas losses owing to slower decomposition can be achieved, making the item of very practical use. Today, endothermics are commodities with many “reinventions” and well established in the global market. They are used almost in every kind of injection-molding processes.26,27,32 The advantage of lower pressures can be applied in the molding process. For this reason, lower clamping force and less wear at the molds can be expected. Shut-off nozzles at machines are necessary to produce good and reliable results. 2.3.1.1 Methods 2.3.1.1.1 Standard (conventional) injection molding The reduction of sink marks and chemical foaming of polymers for weight reduction can help achieve certain properties such as a matt surface, good floating, better flow and mold filling, and faster “outgassing” than ADCs. In addition, CBA powder can be easily fed into hopper of an injectionmolding machine.

61259_C002.indd 52

11/5/2008 4:26:13 PM

Development of Endothermic Chemical Foaming/Nucleation Agents

53

2.3.1.1.2 Direct gassing method An accumulator is filled through an extruder,26 which is gassed primarily with nitrogen under the use of nucleants for improved foam structure, a smoother surface and less gas losses. A piston or plunger injects the melt into the mold via a single or multi-nozzle system (e.g. Figure 2.4). 2.3.1.1.3 Gas counter process Here a good surface area is wanted. Compared with ADC, the endothermics give better results. In a rather thin wall molding, ADC may give better flow and less sinks. Also modern methods like injection of “supercritical” fluids in the melt may achieve better results due to the higher and controlled gas pressure of the introduced gases. Compared with the standard process, here the mold is sealed and preloaded with gas (nitrogen) and the injection takes place against this pressure buffer to obtain a good surface; the pressure is released during the shot and the mold filled with the help of the gas pressure inside the melt. 2.3.1.1.4 Co-injection process This process (sandwich process, 2K process, etc.) allows the use of foaming agents in the inner, outer or inside and outside, depending on what final results and applications are required. When using the foaming agent inside, the surface from the outside layer is like compact molding. The inner gas/blowing agent-containing material avoids sink marks and helps for weight reduction and excellent mold filling. 2.3.1.1.5 Special processes In processes with expandable or “opening” molds after and/or during the shot, the endothermics help to achieve good surface, low densities, no or low warpage and good cell distribution with good controllable foaming.26,27 Today, there are combinations of various processes where the blowing agents are applied accordingly. In summary, the main benefits of using foaming agents in injection molding are weight reduction, less resin consumption, no sinks, less clamping force, and a faster cycles. The kind and level of CBA affects the degree of the benefits. 2.3.1.2 Role of Endothermics versus Exothermics10,11,33 Table 2.4 shows a comparative summary between endothermals and exothermals, and their blends in injection molding process. 2.3.1.2.1 Surface roughness With the increase in exothermic ADC in resins such as ABS, PPE/PPO, and PS, the increase of the surface roughness is quite significant14 compared with endothermics.

61259_C002.indd 53

11/5/2008 4:26:13 PM

Polymeric Foams

54

TABLE 2.4 Comparative Summary between Exothermals, Endothermals, and Their Blends in the Injection Molding Process Properties Weight/density reduction Surface roughness Fine cell structure Sink mark reduction Cycle time reduction Food grade status Color Discoloration tendency Smell Gas evolution Gas pressure Environmental aspect

Exothermals

Enexothermals*

Endothermals

Excellent Poor Coarse Good to excellent 0% Limited Yellow Yes Pungent (NH3) Nitrogen High Limited impact

Very good Reasonable Reasonable Good to excellent 10–30% Limited White to yellowish Reasonable Reasonable Nitrogen, CO2 Medium Limited impact

Good Excellent Excellent Excellent 20–40% No problem White No Little CO2 Medium/low Little/no impact

*Blends between exo- and endothermic foaming agents.

Let us take ABS injection molding as an example. Conditions were constant, values taking on parts with the same weight with the measurement of the roughness in the centre between gating and outside rim. Comparison allows producing the same part and the same settings. The roughness data are as follows (plotted in Figure 2.5): • Non-foamed: 100 • Standard endothermic: 208

Blowing agent

ADC

Enexo

Endo Ry max Ra min

No 0

0.5

1

1.5 2 2.5 3 3.5 Value (microns)

4

4.5

FIGURE 2.5 Surface roughness comparison for ABS without and with various CBAs in injection molding.

61259_C002.indd 54

11/5/2008 4:26:13 PM

Development of Endothermic Chemical Foaming/Nucleation Agents

55

90.0 80.0

Pressure/bar

70.0 60.0

Pressure Development; Autoclave-Testing (conditions see above)

1

TEST PRODUCTS: 1 Azodicarbonamide 2 Blend Endo/Exo 1:1 3 Endo Standard like Hydrocerol-Compound/CF 4 Sodium Bicarbonate 5 Blend Endo/Exo 4:1 like Exocerol 232

50.0 2

40.0

3 4

30.0

5

20.0 10.0 0.0 300

330

360

390

420 450 480 Temperature/K

510

540

570

600

FIGURE 2.6 Pressure profiles for various CBAs at 0.5 K/min heating rate with nominal content of 0.11 g/1 cm3.

• Enexo (50 : 50 Endo/Exo blend): 292 • ADC: 340 2.3.1.2.2 Gas pressure development The pressure exposure during the decomposition is particularly important in regards to the cooling time, the tendency of uncontrolled post-expansion of the molded parts, and the cell structure. This can also influence the physical properties of the parts, depending on the grade of expansion rate. It is quite obvious that CBAs that liberate nitrogen have the highest gas pressure development, which is illustrated in the pressure development graph (Figure 2.6). The differences of the remaining pressure from above test under room temperature are quite drastic, as demonstrated in Table 2.5. 2.3.1.2.3 Cooling phenomena The most significant difference in application is the cooling times in comparison with the calculation of the needed CBA. In this case the cooling time under the test conditions with a somewhat low loading of CBA will be constant. Similar results and details have been reported.9,32 The blowing agent percentage for the cooling time test was taken from gas evolution testing in Figure 2.6 and compared in Figure 2.7. It is fair to conclude that endo- or exothermal CBAs have similar cooling time results, owing to the enormous energy (heat) content in the melt. At normal loading in injection molding the pressure of the gas during the

61259_C002.indd 55

11/5/2008 4:26:14 PM

Polymeric Foams

56

TABLE 2.5 Maximum Gas Pressure at Decomposition and Remaining Pressure in Room Temperature for Various Chemical Blowing Agents Substance

Max. Pr. (Bar)**

Rem. Pr. (Bar)***

85.4 40.6 42.2 41.6 28.1 24.2

30.3 21.1 15.8 11.9 11.2 11.0

ADC 5–PT* ADC/endo 1 : 1 Stand. endo ADC/bic. 1 : 4 Bicarbonate *See Reference 4. **At 495 K. ***At 300 K.

process and the remaining pressure are responsible for weight reduction and cooling time. 2.3.1.2.4 Weight reduction and outlook The weight reduction in foam molding is drastically influenced by the wall thickness of the parts as demonstrated in these tests (Figure 2.8) just by comparing the 5 mm and 8 mm parts (same shape, different thickness). The blowing agent was a standard endothermic CBA. In injection molding, the differences in the nature of various blowing agents are rather simple to demonstrate and the conditions can be easily repeated, and to a certain extent be translated into extrusion or similar processes where the test series are not easily attainable. It should be noted

Cooling time sec. 90 80 70 60

ADC

Enexoth.

Endoth.

0.4

0.51

0.82

50 40

NON

30 20 10 0 0

Blowing agent loading (%) FIGURE 2.7 Cooling time comparison for various CBAs.

61259_C002.indd 56

11/5/2008 4:26:14 PM

Development of Endothermic Chemical Foaming/Nucleation Agents

57

% Density reduction 50 41.2

8 mm

40 35.4 5 mm

27.2

30

32.7 29.2 22.6

20

0

0.2

0.4

0.6

0.8

1.0

Blowing agent loading (%)

FIGURE 2.8 PMMA density reduction at different CBA percentages for two-part thickness: 5 mm and 8 mm.

that the major aspect in any case is not the thermal nature of the foaming agent, but the gas pressure in the melt and the final results for the suitable foaming agent. 2.3.2 Extrusion (See also Section 2.2) 2.3.2.1 Chemically Foamed Extrudates Made with suitable chemical blowing agents (Table 2.3), the decomposition must yield sufficient gases to obtain the cellular structure. The gas formation has to take place at a temperature range close to the processing temperature. The gas should be easily dispersible within the polymer melt. The received decomposition products should be compatible with the resin and not have any negative influence on properties, color, and plate out, corrosion, toxic, and environmental impact.35 Generally, there is no perfect blowing agent in the market, but certain substances and blends have significant market shares. Most import materials are either inorganic (like sodium bicarbonate, sodium borohydride, etc.) or organic (azodicarbonamide, hydrazide/sulfonylhydrazides, organic acids, semicarbazides and tetrazoles). There are exothermic and endothermic decomposition characteristics depending on which group of CBAs are preferred. Mixtures of both groups are called “enexothermics,” and showed interesting improvements in properties. In extrusion, the endothermals found inroads in new products, where fine cell structure, high throughput, lack discoloration, lack of smell, and in many cases food grade classification are needed. Due to slower decomposition reactions, the distribution in the substrates is rather easy.

61259_C002.indd 57

11/5/2008 4:26:14 PM

Polymeric Foams

58

In some traditional applications like rigid PVC foams, both “enexothermal” and plain ADC have been used. The typical examples of chemically foamed extrudates are: PVC PS PP

sheets, boards, profiles, pipes sheets, profiles, picture frames decorative ribbons, strapping material, cable wrapping, monofilaments, sheets, sheets for thermoformed food packaging and microwave heating, packaging material, carpet backing LDPE cable insulation like coaxial cables, sealants from sheet material, etc. HDPE cable insulation, shopping bags, foam netting, marine piles, etc. PET sheets for food packaging etc. PLA rather new foamed material for food packaging.

2.3.2.2 General Aspects of Foam Extrusion15,34–36 For the production of foamed extrusion products, thermoplastic polymers are extruded with chemical foaming agents (Table 2.3). The foaming agent decomposes in the melted polymer and the resulting gas dissolves and disperses in the polymer melt. All common extruders can be used for foaming if the following requirements are fulfilled: 1. The melt temperature must be high enough to guarantee a total decomposition of the foaming agent. 2. The pressure of the melt must be kept high enough to keep the gas generated by the decomposition of the foaming agent dissolved in the polymer melt until the melt exits the extrusion die. 3. The pressure is controlling the gas escape through the hopper, which is undesirable. Pressure or gas losses in the feeding section lead to uneven and irregular results. If the melt temperature is too low, the decomposition of the foaming agent will be incomplete, which is uneconomical, and the non-decomposed foaming agent particles can form agglomerates to clog the melt filter or cause undesirable pressure increase. As a result, voids, irregular cell structure, or poor surface appearance are obtained. Generally, an unusually low pressure profile leads to what is called prefoaming; even with a subsequent pressure increase, the gas cannot be “redissolved.” This results in a large and irregular cell structure, and broken, collapsed cells. The coarse foam produced this way leads to holes in flat and or rupture in blown films, while profiles and sheets get a rough and uneven surface (shark skin).

61259_C002.indd 58

11/5/2008 4:26:14 PM

Development of Endothermic Chemical Foaming/Nucleation Agents

59

2.3.2.3 Extruder Overview: Screw, Melt Filters, and Dies8,15,34 (See also Figures 2.9–2.12) After decades of foam extrusion, it became clear that the right understanding of foam extrusion process is essential. Boehringer Ingelheim published an extrusion guide for foam producers to avoid a rough start-up.34 Most single screw extruders are suitable for chemical foam extrusion. The L : D (length to diameter) ratio should be at least 24 : 1 and screws with an L : D ratio of 30 : 1 are more common. The temperature in the feeding zone should be lower than the initial decomposition temperature of the used foaming agent. The use of a grooved barrel leads to a relatively quick pressure increase in the extruder. This is very advantageous when using foaming agent batches with low melting temperatures. As the foaming agent masterbatch melts in an early stage, and reaches the decomposition temperature earlier, the resulting gas will dissolve the melt owing to the high pressures present at the beginning and can be very well distributed in a short period of time. When using smooth barrels, a sufficient melt pressure is reached more slowly. Foaming agent batches with a low melting point can melt too early at the barrel wall, and the resulting gas can partially or completely escape through the hopper. In this case, the temperature of the feeding zone should be adjusted to a lower temperature, to prevent the premature decomposition of the foaming agent. 2.3.2.3.1 Screw geometry All common screws can be used for foam extrusion, as long as there is no large pressure decrease in the single zones of the screw, which leads to undesirable prefoaming in the melt. Established screws for processing are three-zone screws (feeding, compression, metering/mixing). Good results have also been achieved with degassing screws (PS), as long as you consider the conditions mentioned at the beginning of this chapter.

Forming zone

Metering zone

Grooved feeding zone

FIGURE 2.9 Grooved barrel extruder.

61259_C002.indd 59

11/5/2008 4:26:14 PM

Polymeric Foams

60

Forming zone

Metering zone

Melting & compression zone

FIGURE 2.10

Smooth barrel extruder.

FIGURE 2.11

Pressure profile.

2. Metering zone Degassing zone FIGURE 2.12

61259_C002.indd 60

1. Metering Melting & zone compression zone

Smooth feeding zone

Feeding zone

Extruder with degassing zone.

11/5/2008 4:26:14 PM

Development of Endothermic Chemical Foaming/Nucleation Agents

61

New problems were encountered while introducing the barrier screws in CBA foam extrusion. One of them was an increased tendency to premature foaming, caused by a high pressure gradient between the barrier flight and the driving pitch. When using short barrier segments, the pressure decrease can be compensated for with higher screw speeds. 2.3.2.3.2 Melt filters The use of melt strains, such as screen changers, is generally not necessary in foam extrusion. Owing to foam structure and impurities, gels and additive agglomerates are usually not visible and do not affect the foam product. However, when the impurity becomes too high to cause foaming concerns, screen pack is necessary to maintain quality products. Moreover, screen pack can improve thermal uniformity of the melt flow. If melt filters are used, it is important not to use screens that are too fine. This can cause a high pressure drop after the screens, possibly resulting in prefoaming. 2.3.2.3.3 Slot dies (T, flat, and coathanger dies, etc.) This type of die is used for the production of thin films (with a chill-roll calendar stack), or thicker films, and sheets (with a calendar), respectively. When using slot dies with a restrictor bar (for a better dispersion of the melt), the restrictor bar should not be closed too much. If the restrictor bar is closed too much, pressure reduction can occur right after the restrictor bar, which can result in prefoaming. 2.3.2.3.4 Profile dies, tubular dies, etc. Compared with the cross-section of a slot die, the cross-sections of profile dies generally cannot be changed or adjusted. The product geometry seems to dictate the die design. Generally, the best practice in die design is to keep a short land length to maintain high melt pressure up to the die lip. This is also true for the production of foamed blown film. It must also be noted that the pressure characteristics of a given die can be affected by many factors, such as type of resin, resin viscosity, temperature, desired density reduction, output rate, and actual product cross-section. 2.3.2.4 Coextrusion In general, single layer systems can be translated to the coex design with minimum modification. Many coextruded foam products are produced with a foamed inner layer and solid, non-foamed outer layers. In this case, it is very important, to select the right material. For the outer layer, a “softer” material is recommended, while a material that is somewhat harder is recommended for the inner foamed layer. This type of structure is suggested due to the fact that a foamable melt has better flow characteristics (lower apparent viscosity) compared with a solid melt of the same resin. If the layers differ in viscosity, it can result in poor or damaged foam structure.

61259_C002.indd 61

11/5/2008 4:26:15 PM

Polymeric Foams

62

2.3.3 Other Foam Processes Quite a few popular exothermic CBA foams can be found elsewhere;37,38 for example, PVC (plasticized), different processes in carpet backing, cross-linked foams, woodplastics, and the classic rubber materials. 2.3.3.1 Foam Blow Molding with Endothermics16,28,39 The first trials for the use on an industrial scale where done at the end of 1980s and early 1990s. Some applications had already demonstrated the possibilities in packaging, transportation, automotives, and leisure. The endothermal systems worked out satisfactorily and gave surprisingly good results, due to the fine cell structure (self-nucleation) and slow gas release during the process (refer to graphs and pictures below). Foamed walls have higher resistance against UV influence, which is an advantage in containers and bottles. Figure 2.13 shows the bottle products, cell structure, and foam density at various blow pressures. 2.3.3.2 Rotational Molding Aspects of Endothermic Blowing Agents The use of blowing agents in rotational molding is not new, but only recently the market began to accept this alternative technology compared with other Blow pressure

Density (g/cm3) 0.95 0.9

Blow pressure

Gas pressure inside of the foam cells

0.85 0.8 0.75 0.7 0.8

1

1.2

1.5

1.8

Blow pressure (bar)

FIGURE 2.13

61259_C002.indd 62

Blow molding bottles: products, cell structure, and density chart.

11/5/2008 4:26:15 PM

Development of Endothermic Chemical Foaming/Nucleation Agents

Density (g/cm3) 1

63

Endotherm. B Enexoth 3:1 Stand. Endoth. ADC modif. OBSH ADC

0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 150

FIGURE 2.14

160

170

180 190 Temperature (°C)

200

220

240

LLDPE rotational molding with various CBAs.

molding techniques.11,40–42 The advantages are similar to the blow molding process: weight reduction, small gap between walls, higher stiffness or thicker walls. In reality, modified ADC, endothermics, enexothermics and other exothermic blowing agent can be used. The traditional blowing agent for this application is mainly OBSH. This material is also found in ready-to-use compounds. Unfortunately OBSH is not food approved and the most recent endothermics gave undesirable results in the testing. It is very likely a new combination of food-approved endothermics (acid and carbonate components) could do the trick. The temperature plays an important role in this process, owing to the fact that no “screw machines” with good mixing, shear and temperature control could be used. The best results are obtained by using precompounded materials. The basic results and problems are demonstrated in Figure 2.14. The test was a static test by using LLDPE with drum blended blowing agents under various temperature conditions and measuring later the obtained density as illustrated in Figure 2.14.

2.4 Summary Today, the use of blowing or foaming agents is a well-established technology especially in the well-known additive field. Many polymeric foam products contain CBA. More and more endothermic CBA has been used as blowing agent and nucleating agent. After decades of practice, most exothermics are produced in Asia. The traditional manufacturers such as Bayer gave up or transferred production to China or other countries. Also, the markets for ADC and other exothermics are very large in Asia. Up to

61259_C002.indd 63

11/5/2008 4:26:16 PM

64

Polymeric Foams

80% of the applications are in PVC, and cross-linked material (EVA, PE, rubber, etc.) primarily in Asia (India, China, Indonesia, and others). But there is still room for new developments, as the possibilities for endothermics have yet to be fully realized. Foaming agents will continue to carve out their place in the global markets and in various applications with healthy, rising interest. After CBA foaming, polymer’s property spectrum seemed to expand: lower density, better insulation properties, energy absorption, shock dissipation, and many other production advantages with almost unlimited possibilities. The success in microcellular processing with supercritical carbon dioxide, and cross-linked PE and PP into low-density foam with CBA certainly brings a strong message to the CBA foaming industry. The endothermic CBA has the unique feature of dispersed cooling to locally stabilize the polymeric melt as opposed to melt cooling devices in PBA foaming. Combining the above processes, low-density CBA foam processing could be developed when chemistry and mechanical developments are optimized. It can certainly open up a lot of application gates. A bright future can thus be anticipated.

2.5 Abbreviations 2K 5-PT ABS ADC (ADCA) CBAs cm3 Citric comp. Decomp. DNPT Enexo EVA g GCP GPPS HDPE K LDPE LLDPE Max. pr. min

61259_C002.indd 64

Two-component molding 5-Phenyl tetrazole Acrylonitril butadiene styrene Azodicarbonamide Chemical blowing agents Cubic centimeter Citric acid component Decomposition Dinitroso pentamethylene tetramine Endo-/exothermic blends Ethylene vinyl acetate Gram Gas counter-pressure method General purpose polystyrene High density polyethylene Degrees Kelvin Low-density polyethylene Linear low-density polyethylene Maximum pressure Minute

11/5/2008 4:26:16 PM

Development of Endothermic Chemical Foaming/Nucleation Agents

Min. pr. mm mp NaHCO3 NaHCO3/citric comp. OBSH PE PET PLA PP PPE/PPO PS PS/GPPS PVC R11/R12 Ra min Ry max SBH Stand. endo THT TSH TSSC USM

65

Minimum pressure Millimeter Melting point Sodium bicarbonate Sodium bicarbonate/citric acid component 4,4’-Oxybis(benzenesulfonyl hydrazide) Polyethylene Polyethylene terephthalate Polylactic acid Polypropylene Polyphenylene ether/polyphenylene oxide Polystyrene Polystyrene/general purpose polystyrene Polyvinyl chloride CFC-11/CFC-12  CFCl3/CF2Cl2 Lowest value in microns Highest value in microns Sodium borohydride Standard endothermal foaming agent (hydrocerol-like compound) 2,4,6-Trihydrazino-1,3,5-triaazine Toluene sulfonyl hydrazide p-Toluene sulfonyl semicarbazide United Shoe Machinery Company

References 1. Neumüller, O.-A. “Römpps Chemie-Lexikon.” 7. Aufl., Band (part) 5 W. Keller & Co., stuttgart, (1975): 3254. 2. Landrock, A. H. Handbook of Plastic Foams. Plastics Techn. Evaluation Centre, Dover, NJ, 1985. 3. Müller, E. et al. “Aliphatische Diazo- und Azoverbindungen in der Kunststoffchemie.” Angewandte Chemie 63 (1951): 18–20. 4. Lober, F. “Entwicklung und Bedeutung von Treibmittel bei der Herstellung von Schaumstoffen aus Kautschuk und Kunststoffen.” Angewandte Chemie 64 (1952): 65–76. 5. Kirkland, C. “Blowing agents—New alternatives on all fronts.” Plastics Technology (1986): 83–87. 6. Scholz, D. and Amecke B. “Foaming agents.” In Modern Plastics Encyclopaedia, Modern Plastics, New York, 1989, pp. 166–170. 7. Scholz, D. “Structural foams in technical resins using endothermal blowing agents.” In Proceedings of the Structural Plastics Conference, April 25–28 1993, San Francisco, pp. 95–98.

61259_C002.indd 65

11/5/2008 4:26:16 PM

66

Polymeric Foams

8. Scholz, D. “How to get the most out of endothermic blowing agents.” In RAPRA Conference Blowing Agent Systems; Formulations & Processes, 19 February 1998, Shrewsbury Shropshire, UK. 9. Kretzschmann, G. et al. “Citric acid esters.” USP 4,572,740. 10. Kosin, J. A. et al. “High temperature endothermic blowing agents compositions and applications.” US patent 5,009,809. 11. Wason, S. K. “Endothermic blowing agents compositions and applications.” US patent 5,009,810. 12. Garcia, R. A. “Endothermic blowing agents for strengthening weld lines in molded thermoplastic resins and products.” US patent 5,037,580. 13. Hansen, R. H. and Martin, W. M. I&SC Product Research and Development 3 (1964): 137–141. 14. Pontiff, T. M. “Factors affecting foam cell nucleation in direct gassed foam extrusion.” In Proceedings of the FOAMPLAS Conference, 4–5 November 1997, Mainz, pp. 251–261. 15. Scholz, D. “25 Years endothermic blowing or foaming agents—Is that enough?” In Proceedings of the Structural Plastics Conference, 27th annual design competition held by the Society of the Plastics Industry Inc.’s Structural Plastics Division, 18–21 April 1999, Boston, MA. 16. “Foamed Plastics.” In Ullmann’s Encyclopaedia of Industrial Chemistry, Gerhartz, W., Schulz, G., et al., Eds. 5th ed., Vol. A11, 1985–1996, Weinheim, Germany, 435–464. 17. Houston, J. C. et al. Koppers & Comp. (Composition Comprising Polystyrene. . .). US Patent 2,941,964, 1960. 18. Collins, F. H. “(Dow Chemical), Controlled density—Polystyrene foam.” SPE Journal 16 (1960): 705–709. 19. Hansen, R. H. “Production of fine cells in the extrusion of foams.” SPE Journal 18 (1962): 77–82. 20. Kropp, D. et al. “Foam extrusion of thermoplastic elastomers using CO2 as blowing agent.” In Annual Technical Conference Antec ’97 April 27–May 2, Toronto, Canada, pp. 3473–3478. 21. Fricke, H. “Coated citric acid/Zitronensäure Type H.” Technical Information, C.H. Boehringer Sohn, Ingelheim, April 1967. 22. Lübke, G. “Thin-walled components.” KU Kunststoffe Plast Europe 92 (2002): 12, 36–38/79–82. 23. Padareva, V. et al. “Modification of blowing agent system based on sodium bicarbonate with activated natural zeolite.” Journal of Materials Science Letters 17 (1998): 107–109. 24. Kretzschmann, G. et al. “Nucleating agents (Pore Regulators) for the Preparation of Direct Gassed Thermoplastic Foams.” US patent 5,225,107. 25. Scholz, D. “The position of endothermic nucleating and blowing agents in thermoplastic foams.” Paper presented at Ausplas Conference, 12–17 October 1987, Melbourne, Australia. 26. “Engineering structural foam.” Celanese Plastics Company Brochure, Newark, NJ, 10, 1979/1980(?). 27. “Engineering structural foam.” GEP—General Electric Plastics Europe Brochure, 15–17, 1980/1981(?).

61259_C002.indd 66

11/5/2008 4:26:16 PM

Development of Endothermic Chemical Foaming/Nucleation Agents

67

28. Spiekermann, R. “New applications for chemical foaming agents in thermoplastics.” In Proceedings of the FOAMPLAS Conference, 1997, Mainz, pp. 263–275. 29. Hurnik, H. et al. “Chemische treibmittel.” Kunststoffe 86 (1996): 997–1001. 30. Lapierre, R. et al. “Chemical blowing agent compositions.” US patent 4,769,397. 31. Jacobs, P. M. “Endothermic blowing agents and structural foam quality.” In Proceedings of 12th Annual Structural Foam Conference and Part Competition, The Society of the Plastic Industry, Inc., 7–9 May 1984, San Francisco, CA. 32. Scott, R. M. “Polymeric performance in thin wall structural foam.” Proceedings of 12th Annual Structural Foam Conference and Part Competition, 1984. 33. Scholz, D. “New aspects using endothermal foaming agents.” Paper presented at Ausplas Conference, Stream 3, Commercial & Technical Sessions, 13 October 1993, Melbourne, Australia. 34. Boehringer Ingelheim, K.G. “Chemically blown thermoplast foams: Extrusion.” Technical Information/Processing Guides. Issues: 07/1991 and 03/1997. 35. Berghaus, U. et al. “Foaming of film and sheet from polypropylene.” Paper presented at the Polypropylene ‘95 Conference, September 11–12, 1995, Zurich. 36. Hensen, F. Plastics Extrusion Technology. Hanser Publishers, Munich, New York, 1998, pp. 430–487. 37. Throne, J. L. Thermoplastic Foams. Sherwood, Hinckley, Ohio, 1996. 38. Lee, S. T., Park, C. B., and Ramesh, N. S. “Wood composite foams.” In Polymeric Foams: Science and Technology. Taylor and Francis, Boca Raton, 2007. 39. Spiekermann, R. “Development of large and small applications for foamed blow moulding.” Presentation at National Plastics Exhibition, Chicago 1997. 40. Scholz, D. “Blowing agents for rotational moulding.” Presented at the 16th Annual ARM Australia Conference, March 13–15, 1994, Melbourne. 41. “Foaming of thermoplastics—Blowing agents for rotational moulding: Plastics technology.” Hong Kong Plastics Technology Centre Ltd. 18 (1994): 30–32. 42. Büttner, H. “Foaming of thermoplastic resins by rotational moulding”, Presented at the ARM Australasia Conference, Bali, Indonesia, 1997.

61259_C002.indd 67

11/5/2008 4:26:16 PM

61259_C002.indd 68

11/5/2008 4:26:16 PM

3 Foam Extrusion Using Carbon Dioxide as a Blowing Agent Walter Michaeli, Dirk Kropp, Robert Heinz, and Holger Schumacher

CONTENTS 3.1 Introduction .................................................................................... 3.2 Effect of Carbon Dioxide on the Flow Behavior of Polymer Melts ................................................................................. 3.2.1 Measuring the Shear Viscosity of Blowing Agent Containing Melts .................................................................. 3.2.2 Calculation of the Viscosity of the Polymer/Blowing Agent Mixtures ..................................................................... 3.2.3 Flow Behavior of Carbon Dioxide Containing Polymer Melts ....................................................................... 3.3 Influence of the Flow Channel Geometry on the Foam Quality .................................................................................. 3.4 Influence of Spider Legs on the Thickness Distribution of Foamed Sheets ........................................................................... 3.4.1 Influence of Melt Temperature ........................................... 3.4.2 Influence of Die Temperature ............................................. 3.4.3 Influence of Spider Geometry ............................................. 3.5 Influence of the Pressure Profile in the Extrusion Die on the Foam Structure ................................................................... 3.6 Nomenclature ................................................................................. References ..............................................................................................

61259_C003.indd 69

70 70 71 72 75 78 81 84 86 88 90 97 97

10/25/2008 12:35:28 PM

70

Polymeric Foams

3.1 Introduction Since chlorofluorocarbons (CFCs) and some hydrochlorofluorocarbons (HCFCs) were banned due to their ozone depletion potential and their contribution to the greenhouse effect, the use of inert gases as physical blowing agents in foam extrusion has gained interest. Among these, carbon dioxide (CO2) is one of the most promising blowing agents for foam extrusion. It is environmentally benign and, due to its high availability, economical reasonable. On the other hand, it has a lower solubility, a higher diffusivity, and a stronger plasticizing effect in most polymers compared to traditional blowing agents. Therefore, the foam extrusion process needs to be adjusted to the properties of CO2. Particularly in die design, the specific requirements for foam extrusion have to be taken into account when CO2 is used as a blowing agent. For instance, the plasticizing effect of the blowing agent changes the flow behavior of the polymer melt that leads to a reduced pressure in the die. In combination with the lower solubility of CO2 in most polymer melts, the risk of premature foaming inside the die is increased. The shape of the flow channel and spider legs therefore also need special attention to prevent visible defects in the foam product. In particular, the pressure profile at the die exit has a big influence on the foam structure.

3.2 Effect of Carbon Dioxide on the Flow Behavior of Polymer Melts The rheological behavior of polymer melts is one of the most important factors to describe processes in polymer processing by a theoretical approach. In particular, the flow behavior of polymer melts is an essential key factor in the design of extrusion dies. In order to predict the pressure profile, the output rate, and the velocity distribution in the die, the viscosity of the polymer has to be known. In foam extrusion with physical blowing agents, the blowing agent is injected into the extruder barrel and dissolved within the polymer melt. The dissolved blowing agent acts like a plasticizer and decreases the viscosity of the polymer melt. Depending on the blowing agent concentration and the polymer/blowing agent interaction, the pressure profile can be affected noticeably. However, the pressure in the die has to be kept above the solubility pressure of the blowing agent in the polymer to prevent a premature foaming of the melt. Since the entire extrusion process is affected by the blowing agent, it is essential to gain knowledge of the impact of the blowing agent on the flow behavior.

61259_C003.indd 70

10/25/2008 12:35:28 PM

Foam Extrusion Using Carbon Dioxide as a Blowing Agent

71

3.2.1 Measuring the Shear Viscosity of Blowing Agent Containing Melts The standard rheometers that are commonly used for the measurement of shear viscosity of polymer melts are the capillary rheometer and the rotary rheometer. Both types of rheometer can be used in a wide shear rate range but it is not possible to introduce a defined amount of blowing agent into the melt. An alternative method is to charge polymer granules with blowing agent prior to the viscosity measurement, but the actual blowing agent concentration in the polymer melt during the measurement can hardly be determined. Since rotary rheometers are not capable of building up a pressure that is high enough to prevent foaming of the melt, even pre-charged polymer cannot be measured in this type of rheometer. By contrast, capillary rheometers can operate under high pressure, but during the melting of the polymer, blowing agent is lost through the capillary and the gap between cylinder and plunger. Furthermore, a premature foaming of the melt inside the capillary will lead to incorrect values. Consequently, standard rheometers are not capable of measuring the viscosity of polymer melts in dependence on the blowing agent concentration. Han et al. used a capillary die to investigate the flow behavior of polyethylene (PE) and polystyrene (PS) melt containing chemical blowing agents1 and charged with fluorocarbon blowing agents.2,3 The volume flow rate through the die was varied by changing the rotational speed of the extruder screw. Thus, extrusion conditions; that is, shear rate, mixing and residence time in the extruder were also changed. The shear rate in the die could only be varied in a range of approximately 100–300 s1. Gendron et al. used a commercial online rheometer for rheological measurements of mixtures of PS with fluorocarbons.4,5 The online rheometer was positioned in a side stream of the extruder and the flow rate through the rheometer slit was controlled by two gear pumps. This guarantees constant process conditions in the rheometer isolated from process fluctuations. The shear rates in the rheometer were varied between 1 and 100 s1. However, an observation of the melt stream is not possible. Kropp developed a special in-line rheometer die which enables the measurement of shear viscosity of blowing agent charged melts at process conditions.6,7 The rheometer is designed as a slit capillary rheometer and can be mounted directly on the extruder. Using different inserts, the height of the 200 mm long slit can be varied between 2, 3.5, and 5 mm. The width of the slit is 50 mm. Along the flow channel, the pressure is measured at three different points in an interval of 60 mm in order to determine the pressure gradient in the slit. The level of pressure in the in-line rheometer die can be adjusted by means of a throttle at the exit of the in-line rheometer die. Glass inserts at both sides of the rheometer allow the observation of the melt flow in the slit. Figure 3.1 shows a picture of this in-line rheometer die.

61259_C003.indd 71

10/25/2008 12:35:28 PM

72

Polymeric Foams

FIGURE 3.1 In-line rheometer die.

During the operation of the in-line rheometer die, the mass flow rate through the slit capillary can be varied by a bypass upstream of the rheometer. By opening a valve, more material flows through the bypass and the flow rate through the rheometer decreases. Using this technique, the flow rate through the rheometer and thus the shear rate can be varied in a wide range without changing the process settings of the extruder. A schematic of the in-line rheometer die and the corresponding pressure course is depicted in Figure 3.2. Assuming that no premature foaming occurs in the rheometer, a linear pressure drop rate in the slit is expected. By closing the throttle at the exit of the in-line rheometer die, the level of the pressure is increased as illustrated in the diagram (dotted lines). Besides the viscosity measurement, the in-line rheometer die can also be used to investigate the bubble formation (Figure 3.3). Releasing the pressure in the rheometer by opening the throttle at the exit of the die, the initial point of bubble formation moves into the observable region in the rheometer. Since the pressure profile in the slit is known, the critical pressure; that is, the pressure when first bubbles are initiated, can be determined. This critical pressure is an essential value in the design of foam extrusion dies. 3.2.2 Calculation of the Viscosity of the Polymer/Blowing Agent Mixtures The shear rate and the melt viscosity are derived using the concept of representative viscosity.8–10 This method assumes that certain points exist in every laminar flow where the shear rate of both Newtonian and shearthinning fluids are identical. These points are called representative points.9

61259_C003.indd 72

10/25/2008 12:35:28 PM

Foam Extrusion Using Carbon Dioxide as a Blowing Agent

73

Adjustable throttle In-line rheometer die

p1

p2

p3

Pressure transducer

Pressure p (bar)

Decreasing throttle gap p1 p2

p3

Flow length L (mm) FIGURE 3.2 Pressure course in the in-line rheometer die with small throttle gap. [From Kropp, D. Extrusion thermoplastischer Schäume mit alternativen Treibmitteln (Extrusion of thermoplastic foams using alternative blowing agents). PhD Thesis, RWTH Aachen, 1999.]

The true shear rate for shear thinning fluids is derived from the Newtonian shear rate multiplied with a factor e considering the geometry of the flow channel (Equation 3.1). . _. 6 ¥ Vm ______ g  e (3.1) B ¥ H2  In principle, the geometric factor e depends on the flow characteristics of the material. However, for most polymer melts with flow exponents m in the range 2–4, it can be considered as constant for a given flow channel shape.11 Equation 3.2 shows the geometric factor for a rectangular slit geometry of the flow channel. e   0.772

61259_C003.indd 73

(3.2)

10/25/2008 12:35:28 PM

74

Polymeric Foams

FIGURE 3.3 Pressure course and bubble formation in the in-line rheometer die with large throttle gap. [From Kropp, D. Extrusion thermoplastischer Schäume mit alternativen Treibmitteln (Extrusion of thermoplastic foams using alternative blowing agents). PhD Thesis, RWTH Aachen, 1999.]

The true viscosity of the melt/blowing agent mixture can be calculated from the pressure loss p at a certain flow channel length L with Equation 3.3: Dp ¥ B ¥ H3 h–  __________ (3.3) 12 ¥ L ¥ V˙ m The mass flow rate m˙ through the rheometer is determined by weighing the melt extruded in a certain period of time. The volume flow rate of the mixture V˙ m needed for the calculation of shear rate and viscosity is derived by the following equation: m˙ V˙ m  ___ r m

61259_C003.indd 74

(3.4)

10/25/2008 12:35:28 PM

75

Foam Extrusion Using Carbon Dioxide as a Blowing Agent

TABLE 3.1 Parameters for the Calculation of Melt Density of Unfilled Standard Polymers Expansion Coefficient a (1/K)

Reference Density r0 (g/cm)

Reference Temperature T0 (°C)

LDPE

0.69 ¥ 103

0.801

115

HDPE

0.69 ¥ 103

0.792

131

PP

0.61 ¥ 103

0.759

186

PS

0.56 ¥ 103

1.029

84

Source: Menges, G. “4. korrigierte und aktualisierte.” In Werkstoffkunde Kunststoffe. Auflage, Carl Hanser Verlag, München, 1998.

The density of the polymer/blowing agent mixture r m can be derived using a linear mixing law. Since blowing agents are generally soluble in polymer melts, the density of the mixture can be considered as the density of a polymer solution with a certain weight fraction of blowing agent w S:12 rP  r s rm  _____________________ (1  w S)  rS  w S  rP

(3.5)

The temperature-dependent density of the polymer rP(T) is given by the melt density at a reference temperature T0 and the expansion coefficient a:13 r(T0) rP(T)  _______________ 1  a (T  T0)

(3.6)

In Table 3.1,13 the expansion coefficient and the density at a reference temperature are listed for some standard thermoplastics. Since the density of polymers is also dependent on the pressure, more accurate values can be taken of p–v–T plots. Since most blowing agents are in a supercritical state at process conditions, they cannot be considered as an incompressible fluid. In this state their density is dependent on both temperature and pressure. Table 3.2 lists the density of CO2 in a range relevant for foam extrusion.14 3.2.3 Flow Behavior of Carbon Dioxide Containing Polymer Melts Using the in-line rheometer die, the viscosity of polymers charged with different concentrations of blowing agent were measured.15,16 Three polymer types that are commonly used in foam extrusion were chosen for these measurements: polystyrene (PS), low-density polyethylene (LDPE) and polypropylene (PP). The viscosity curves for the polymers without blowing agent were determined using a high-pressure capillary rheometer with a slit capillary having nearly the same height/width ratio like the in-line rheometer die. Viscosity measurements of PS using both the

61259_C003.indd 75

10/25/2008 12:35:29 PM

76

Polymeric Foams

TABLE 3.2 Density of CO2 in kg/m3 Depending on Pressure and Temperature Temperature

Pressure (Bar)

50°C

100°C

150°C

200°C

250°C

300°C

10 20 30 40 50 60 70 80 90 100 150 200 250 300 350 400

17.05 35.61 56.05 78.90 104.96 135.47 172.50 219.78 285.40 388.06 699.55 784.96 834.50 870.59 899.48 923.72

14.52 29.77 45.82 62.75 80.68 99.69 119.90 141.41 164.31 188.67 332.65 481.18 590.49 663.91 716.05 756.74

12.69 25.76 39.23 53.11 67.43 82.19 97.39 113.02 129.08 145.56 233.90 327.52 415.35 492.46 556.52 608.01

11.29 22.78 34.48 46.40 58.54 70.89 83.45 96.18 109.09 122.16 189.47 258.89 326.65 389.63 447.00 497.72

10.17 20.46 30.86 41.37 52.00 62.73 73.57 84.49 95.48 106.54 162.43 218.76 274.16 326.92 376.35 422.11

9.27 18.59 27.97 37.41 46.91 56.46 66.06 75.69 85.36 95.04 143.52 191.71 238.99 284.47 327.64 368.40

Source: N. N. VDI-Wärmeatlas. VDI-Verlag, Düsseldorf, 1984.

high-pressure capillary rheometer and the in-line rheometer die have proven that these methods lead to the same results.17 Since the viscosity measurement of uncharged melts at such low temperatures which are required in foam extrusion is impossible, the curves were determined by means of the time–temperature superposition principle developed by Williams et al.18 Figure 3.4 shows the viscosity curves of PS 158 K (BASF AG, Ludwigshafen, Germany) at 180°C with different CO2 concentrations. The decreasing viscosity with increasing concentration of CO2 shows the plasticizing effect of the blowing agent. However, this plasticizing effect seems to reach a certain limit: the higher the blowing agent concentration, the lower the degree of viscosity reduction. Furthermore, the curves of different blowing agent concentrations are approximately parallel; that is, The shear thinning behavior of PS is not affected by CO2. The viscosity of LDPE (Stamylan 2102 TN 26, DSM,* Geleen, The Netherlands) was measured at temperatures of 120°C, 130°C, and 140°C, each with 0.5%, 1%, and 1.5% of CO2. As an example, the viscosity curves at 140°C are shown in Figure 3.5. Again, the viscosity reduction due to the

* LDPE 2102 is now distributed by Saudi Basic Industries Corporation (SABIC).

61259_C003.indd 76

10/25/2008 12:35:29 PM

77

Foam Extrusion Using Carbon Dioxide as a Blowing Agent

Viscosity (Pas)

10000 0% CO2 1% CO2 2% CO2 3% CO2 1000

100 10

100 Shear rate (1/s)

1000

FIGURE 3.4 Viscosity curves of polystyrene PS 158 K at 180°C and different CO2 concentrations. [From Kropp, D. Extrusion thermoplastischer Schäume mit alternativen Treibmitteln (Extrusion of thermoplastic foams using alternative blowing agents). PhD Thesis, RWTH Aachen, 1999.]

10000

Viscosity (Pas)

0.5% CO2 1% CO2 1.5% CO2

1000

100 10

100 Shear rate (1/s)

FIGURE 3.5 Viscosity curves of LDPE stamylan 2102 at 140°C and different CO2 concentrations. [From Kropp, D. Extrusion thermoplastischer Schäume mit alternativen Treibmitteln (Extrusion of thermoplastic foams using alternative blowing agents). PhD Thesis, RWTH Aachen, 1999.]

blowing agent can be seen. Contrary to the results with PS, the plasticizing effect of CO2 on LDPE is significantly lower. Since the development of new PP grades with improved melt elasticity and high melt strength (HMS-PP), the use of PP in foam extrusion has been risen.19 The viscosity of a HMS-PP developed for foam extrusion (Pro-fax 814, Montell*) was measured at 170°C using the in-line rheometer * Profax PF814 is now distributed by Basell Polyolefi ns.

61259_C003.indd 77

10/25/2008 12:35:29 PM

78

Polymeric Foams

Viscosity (Pa s)

10000 0% CO2 2% CO2 3% CO2 4% CO2 1000

100 10

100 Shear rate (1/s)

1000

FIGURE 3.6 Viscosity curves of polypropylene Profax PF 814 at 170°C and different CO2 concentrations. [From Kropp, D. Extrusion thermoplastischer Schäume mit alternativen Treibmitteln (Extrusion of thermoplastic foams using alternative blowing agents). PhD Thesis, RWTH Aachen, 1999.]

die (Figure 3.6). Similar to the results of LDPE, only a slight decrease in viscosity can be observed. Analog to the other tests, CO2 seems to have just about no effect on the shear thinning behavior of PP.

3.3 Influence of the Flow Channel Geometry on the Foam Quality Inert gases like CO2 and N2 possess a lower solubility in most polymer melts than hydrocarbons or fluorocarbons. Therefore, a higher risk of premature foaming of the polymer/blowing agent mixture in the die exists when these gases are used as blowing agents. Particularly at points of high shear rates, premature foaming in the die may occur even at pressures above the solubility pressure of the blowing agent. Such high shear rates appear primarily at variations in the cross-section of flow channels; for example, convergent or divergent flow channels, breaker plates, or spider legs in annular dies. Furthermore, a pressure decrease at dead spots can also lead to a supersaturation of the melt and thus to premature foaming. In order to examine the effect of cross-sectional variations in flow channels on bubble formation, different inserts of dissimilar shape were positioned in the 3.5 mm slit of the in-line rheometer die.16,20 The glass windows at the side of the slit allowed the observation of the melt and thus the study of

61259_C003.indd 78

10/25/2008 12:35:30 PM

Foam Extrusion Using Carbon Dioxide as a Blowing Agent

79

the initiation of bubble formation. The inserts had a height of 2.5 mm and angles of 30°, 45°, 60°, and 90° at the downstream side, respectively. A combination of LDPE (Stamylan 2102 TN26) charged with 2% CO2 at 130°C was chosen as the polymer/blowing agent mixture for the tests. By adjusting the throttle at the die exit, the level of pressure at the downstream edge of the inserts was kept constant during the tests. The flow behavior of the CO2-charged melt at the inserts with a downstream angle of 30° and 45° is compared in Figure 3.7. At the flow channel insert with a downstream angle of 30° (upper picture), no premature foaming occurred at the downstream edge. Bubbles were first initiated in the dead space downstream of the insert. Subsequently, they grew due to blowing agent diffusing into the bubbles. In contrast, the bubbles were initiated immediately at the downstream edge of the insert with 45° even though the pressure at this point was higher than the solubility pressure of the blowing agent (Figure 3.7, lower picture). Again, the bubbles grew and caused visible defects in the foam structure of the extrudate. Besides the practical examination of the bubble formation, the melt flow at the inserts was calculated with the finite-element method (FEM).21 Figure 3.8 shows the qualitative results of the isothermal, two-dimensional flow simulation at the insert with a downstream angle of 45°. Regarding the pressure distribution in the narrowed flow channel, the isobars run parallel as expected. Due to the flow deflection, a high-pressure gradient is generated at the downstream edge of the insert. This is shown by the

α = 30° Melt flow direction

α = 45° Melt flow direction 1 mm FIGURE 3.7 Bubble formation at flow channel expansions. [From Heinz, R. Prozessoptimierung bei der Extrusion thermoplastischer Schäume mit CO2 als Treibmittel (Process Optimization for the Extrusion of Thermoplastic Foams Using CO2 as a Blowing Agent). PhD Thesis at RWTH Aachen, 2002.]

61259_C003.indd 79

10/25/2008 12:35:30 PM

80

Melt flow direction

Melt flow direction

Polymeric Foams

Pressure distribution (isobars)

Velocity distribution (isotachs)

FIGURE 3.8 Simulated melt flow at a channel expansion. [From Heinz, R. Prozessoptimierung bei der Extrusion thermoplastischer Schäume mit CO2 als Treibmittel (Process Optimization for the Extrusion of Thermoplastic Foams Using CO2 as a Blowing Agent). PhD Thesis at RWTH Aachen, 2002.]

isobars running very close to each other. Downstream of the flow channel expansion, the distance between the isobars becomes wider; that is, the pressure loss is lower due to the lower velocity of the melt. The visualization of the melt flow shows a symmetrical velocity distribution in the narrowed flow channel and downstream of the flow channel expansion. The velocity at the walls is set to zero, which is represented by the dark color. The maximum velocity is reached in the center of the narrowed flow channel. The shear rate in the melt/blowing agent mixture can be assessed by the distance between two isotachs (lines of equal velocity). The maximum shear rate occurs close to the walls of the narrowed flow channel. Assuming that the bubbles are initiated solely by the pressure drop below the solubility pressure of the blowing agent, the foaming should occur along one of the isobars shown in Figure 3.8. This seems to be valid for the experiments with the 30° insert. However, when using the insert with 45° downstream angle, the foaming starts at the downstream edge reproducibly. Consequently, the bubble nucleation in the melt flow has to be influenced by further mechanisms. Shear effects are believed to have an influence on bubble nucleation in foam extrusion. Lee stated in his modified cavity model that it is not only the degree of supersaturation and the amount of nucleators that influence bubble nucleation in foam extrusion; shear forces also have an effect.22 Due to the symmetric velocity distribution in the flow channel, bubble nucleation should also be observable at the opposite side of the downstream edge of the 45° insert if shear-induced nucleation is assumed. Furthermore, the shear rate distribution in the narrowed flow channel is

61259_C003.indd 80

10/25/2008 12:35:30 PM

81

Foam Extrusion Using Carbon Dioxide as a Blowing Agent

20

10

Pressure gradient in the narrowed slit 0

30

Pressure gradient (MPa/s)

Pressure gradient (MPa/s)

200 Pressure gradient at the downstream edge

100

0 45 60 Downstream angle (°)

30

45 60 Downstream angle (°)

FIGURE 3.9 Pressure gradients at the downstream edge of the flow channel inserts. [From Heinz, R. Prozessoptimierung bei der Extrusion thermoplastischer Schäume mit CO2 als Treibmittel (Process Optimization for the Extrusion of Thermoplastic Foams Using CO2 as a Blowing Agent). PhD Thesis at RWTH Aachen, 2002.]

the same in all experiments, but bubbles are only nucleated at the downstream edge when the downstream angle of the insert exceeds 45°. As illustrated in Figure 3.8, the local pressure gradient at the downstream edge of the insert is very high. Heinz explains the premature foaming at the downstream edge of the inserts by the high pressure gradient at this point.20 Figure 3.9 shows the pressure gradients in the narrowed flow channel and at the downstream edge of the inserts with a downstream angle of 30°, 45°, and 60°, respectively. It is obvious that the pressure gradient at the downstream edge is significantly higher than in the flow channel. Furthermore, it can be seen that the pressure gradient at the downstream edge increases with the downstream angle of the insert. To prevent premature foaming in foam extrusion dies, Heinz proposed avoiding high local pressure gradients, particularly at variations of crosssections and breaker plates. Therefore, variations in cross-section should be provided with small angles and adequate radii.20

3.4 Influence of Spider Legs on the Thickness Distribution of Foamed Sheets In extrusion of low-density foam sheets, annular dies are the most commonly used dies.23 The mandrel in these dies is supported either by one spider leg (single-spider die), two spider legs (dual-spider die), or

61259_C003.indd 81

10/25/2008 12:35:30 PM

82

Polymeric Foams

Single -spider

Dual -spider

Breaker plate

Mandrel Spider leg Housing FIGURE 3.10 Different concepts of mandrel supports. [From Heinz, R. Prozessoptimierung bei der Extrusion thermoplastischer Schäume mit CO2 als Treibmittel (Process Optimization for the Extrusion of Thermoplastic Foams Using CO2 as a Blowing Agent). PhD Thesis at RWTH Aachen, 2002.]

a breaker plate (Figure 3.10). All these supports cause spider lines (streaks) in the extrudate. In foam extrusion, these spider lines are even more pronounced than they are in the extrusion of solid sheets or tubes.24 In the region of the spider lines, foamed sheets exhibit local thin sections and poor foam structure. If CO2 is used as a blowing agent, additionally, a higher risk of premature foaming at the spider legs exists due to the low solubility of CO2.20 Based on the results of the experiments with different inserts in the flow channel of the in-line rheometer die, Heinz conducted experiments to optimize the spider leg geometry in annular dies.20 Since foam sheets are cut at the spider lines, the quality of the resulting foam sheets is almost unaffected. The objective of these experiments was not the minimization of the difference in thickness but more the reduction of the width of the spider lines in order to reduce edge trim waste. The influence of melt and die temperature on the spider lines were also examined. Experiments were conducted using an annular die with an outlet diameter of 50 mm that was mounted on a 60 mm single screw foam extruder. The mandrel of the die was supported by one spider leg. At the opposite side of this spider leg, spider dummies of various geometries were installed. Figure 3.11 shows the cross-section of the die with the spider leg (bottom) and an installed spider dummy (top). The geometries of the spider dummies are depicted in Figures 3.12 and 3.13. Spiders I and II have different angles at the downstream side of 20° and 45° towards the centerline, respectively (Figure 3.12). Upstream angle, length, and width of these two spiders are equal. Spiders III–V have the same angles at the upstream and downstream side (Figure 3.13), while featuring different sizes. Spider IV is shorter and Spider V is thicker than Spider III.

61259_C003.indd 82

10/25/2008 12:35:31 PM

83

Foam Extrusion Using Carbon Dioxide as a Blowing Agent

FIGURE 3.11 Mandrel with spider leg and spider dummy. [From Heinz, R. Prozessoptimierung bei der Extrusion thermoplastischer Schäume mit CO2 als Treibmittel (Process Optimization for the Extrusion of Thermoplastic Foams Using CO2 as a Blowing Agent). PhD Thesis at RWTH Aachen, 2002.]

The experiments were conducted with LDPE (Lupolen 1810 H, Basell, Hoofddorp, the Netherlands) and HMS-PP (Profax PF814, Basell). As a nucleating agent, a masterbatch-concentrate (Hydrocerol CF 20 E, Clariant Masterbatch, Lahnstein, Germany) based on a sodium carbonate/citric acid system was added at a concentration of 1.5%. When LDPE was used,

Spider I

12,5

40°

Melt flow direction

40°

70

Spider II

12,5

90°

Melt flow direction

40°

70

FIGURE 3.12 Spider geometries with different downstream angles. [From Heinz, R. Prozessoptimierung bei der Extrusion thermoplastischer Schäume mit CO2 als Treibmittel (Process Optimization for the Extrusion of Thermoplastic Foams Using CO2 as a Blowing Agent). PhD Thesis at RWTH Aachen, 2002.]

61259_C003.indd 83

10/25/2008 12:35:31 PM

84

Polymeric Foams

12,5

40°

Melt flow

40° 56

Spider IV

40°

direction

40°

Melt flow

12,5

70

Spider III

direction 70

20

40°

Melt flow

40°

Spider V

direction FIGURE 3.13 Spider geometries with different length and width. [From Heinz, R. Prozessoptimierung bei der Extrusion thermoplastischer Schäume mit CO2 als Treibmittel (Process Optimization for the Extrusion of Thermoplastic Foams Using CO2 as a Blowing Agent). PhD Thesis at RWTH Aachen, 2002.]

a cell stabilizer (Activex CT 325, Clariant Masterbatch) and an ionomer (Surlyn 9910, DuPont de Nemours International, Geneva, Switzerland) were added. The thickness of the foam sheets was measured at several points using a hall-type thickness gauge (Magnamike 8000, Panametrics, Hofheim, Germany) with a 3/16” ball. The data were plotted against the position defined by the angle. An angle of 0° corresponds to the position of the cutter; that is, the edge of the slit foam sheet. The whole die was rotated by 45° so that the spider leg of the die was at 315° and the spider dummy at 135°. As illustrated in Figure 3.14, both the spider leg and the spider dummy generated pronounced local thin sections. 3.4.1 Influence of Melt Temperature At first, the influence of the melt temperature on the thickness profile of the LDPE foam sheets was examined. Therefore, the melt temperature was varied between 110°C and 115°C whereas the die temperature was kept constant at 105°C. Figure 3.15 shows the thickness profile of the foam sheets at different melt temperatures in the region of the spider dummy for Spider IV exemplarily. The thickness of the sheets increases with decreasing melt temperature over the entire circumference. Simultaneously, the foam density decreases from 212 kg/m3 at 115°C to 166 kg/m3 at 111°C melt temperature. This might be caused by the lower

61259_C003.indd 84

10/25/2008 12:35:31 PM

85

Thickness (mm)

Foam Extrusion Using Carbon Dioxide as a Blowing Agent

2.4 2.2 2.0 1.8 1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2

Spider dummy LDPE PP 0

45

Spider 90 135 180 225 270 Circumference of the foam tube (°)

315

360

FIGURE 3.14 Thickness profile of the foam sheet. [From Heinz, R. Prozessoptimierung bei der Extrusion thermoplastischer Schäume mit CO2 als Treibmittel (Process Optimization for the Extrusion of Thermoplastic Foams Using CO2 as a Blowing Agent). PhD Thesis at RWTH Aachen, 2002.]

diffusivity of the blowing agent at lower temperatures. Thus, the loss of blowing agent through the surface of the foam sheets is reduced and more blowing agent is available for density reduction. The thickness of the spider line at the position of the spider dummy increases, too. In contrast, the thickness of the sheets at the position of the spider leg does not change with the melt temperature. This indicates that

2.5

Thickness (mm)

2.0

1.5

1.0

0.5 0.0 45

Spider IV LDPE Die temperature 105°C

90

Melt temperature 111°C 113°C 115°C

135 180 Circumference of the foam tube (°)

225

FIGURE 3.15 Thickness profile for different melt temperatures (LDPE). [From Heinz, R. Prozessoptimierung bei der Extrusion thermoplastischer Schäume mit CO2 als Treibmittel (Process Optimization for the Extrusion of Thermoplastic Foams Using CO2 as a Blowing Agent). PhD Thesis at RWTH Aachen, 2002.]

61259_C003.indd 85

10/25/2008 12:35:32 PM

86

Polymeric Foams

2.2

Thickness (mm)

2.0 1.8 1.6 1.4 1.2 1.0

Melt temperature 175°C 177°C 179°C

0.8 90

Spider IV PP Die temperature 170°C 135

180

Circumference of the foam tube (°) FIGURE 3.16 Thickness profile for different melt temperatures (PP). [From Heinz, R. Prozessoptimierung bei der Extrusion thermoplastischer Schäume mit CO2 als Treibmittel (Process Optimization for the Extrusion of Thermoplastic Foams Using CO2 as a Blowing Agent). PhD Thesis at RWTH Aachen, 2002.]

the spider dummy is thermally capsuled from the die. Thus, the temperature of the spider dummy adapts to the melt temperature. The thickness profile of the PP foam sheets at various temperatures between 175°C and 180°C is illustrated in Figure 3.16. The thickness of the sheets increases with decreasing melt temperature. In contrast to the LDPE sheets, the thickness of the spider line after the spider dummy does not increase with falling melt temperature. At 175°C melt temperature, a noticeable spider line is observable independent of the spider geometry. This could be due to the low viscosity of the PP melt at 175°C, reducing the intermixing of the melt streams behind the spider. The spider lines of the spider leg are not influenced by the melt temperature for PP. 3.4.2 Influence of Die Temperature In order to examine the influence of the die temperature on the thickness distribution of LDPE foam sheets, the temperatures of die housing, mandrel and die lips were varied from 105°C to 135°C whereas the melt temperature was kept constant at 115°C. Figure 3.17 shows a distinct influence of the die temperature on the spider lines. At a die temperature of 105°C, the thickness profile is only slightly affected by the spider leg, whereas a pronounced thin section emerges when the die temperature exceeds 120°C. The influence of the die temperature on the spider line of the spider dummy is rather low. The thickness distribution of the sheets at 105°C and 120°C die temperature is nearly the same. At a die temperature of 135°C, the thickness of the foam sheet is slightly lower. This is probably caused by an increase of the melt temperature due to the higher die temperature.

61259_C003.indd 86

10/25/2008 12:35:32 PM

87

Foam Extrusion Using Carbon Dioxide as a Blowing Agent

1.4

1.4 Die temperature 105°C 120°C 135°C Melt temperature 115°C

1.0 0.8

1.2 Thickness (mm)

Thickness (mm)

1.2

0.6 0.4 0.2

305

310

315

0.8 0.6 0.4 Spider IV LDPE

0.2

Spider leg LDPE

0.0 300

1.0

0.0 125

320

Circumference of the foam tube (°)

130

135

140

145

Circumference of the foam tube (°)

FIGURE 3.17 Influence of temperature settings on the thickness profile (LDPE). [From Heinz, R. Prozessoptimierung bei der Extrusion thermoplastischer Schäume mit CO2 als Treibmittel (Process Optimization for the Extrusion of Thermoplastic Foams Using CO2 as a Blowing Agent). PhD Thesis at RWTH Aachen, 2002.]

2.0

2.0

1.8

1.8

1.6

1.6

1.4 1.2 1.0 0.8 0.6

Melt temperature 175°C 177°C 179°C Die temperature 170°C

0.4 305 310 315 320 325 Circumference of foam tube (°)

Thickness (mm)

Thickness (mm)

Because of the narrow processing window, the die temperature can only be varied in a range of 165°C to 175°C in the experiments with PP. As Figure 3.18 depicts, the spider lines in the PP foam sheets are not affected significantly by the die temperature. Compared to the melt temperature of 175°C, the considered temperature range is probably too small to have a significant effect on the thickness distribution of the foam sheets.

Spider leg PP

1.4 1.2 1.0 0.8 0.6

Die temperature 165°C 170°C 175°C Melt temperature 175°C

0.4 305 310 315 320 325 Circumference of foam tube (°)

FIGURE 3.18 Influence of temperature settings on the thickness profile at the spider (PP). [From Heinz, R. Prozessoptimierung bei der Extrusion thermoplastischer Schäume mit CO2 als Treibmittel (Process Optimization for the Extrusion of Thermoplastic Foams Using CO2 as a Blowing Agent). PhD Thesis at RWTH Aachen, 2002.]

61259_C003.indd 87

10/25/2008 12:35:32 PM

88

Polymeric Foams

3.4.3 Influence of Spider Geometry In the first experiments on the spider geometry, the influence of the downstream angle of the spiders was examined. As the experiments with the inserts in the flow channel of the in-line rheometer die have shown, a too steep downstream angle leads to a premature foaming in the die. Figure 3.19 shows the thickness profile of foam sheets of LDPE and PP produced with the Spiders I and II (cp. Figure 3.12). Neither the thickness distribution of the sheets nor the form of the spider lines exhibit significant differences for both spider geometries. This is true for both materials and all considered melt temperatures. The downstream angle of the spider seems to have no effect on the quality of the produced foam sheets at the examined conditions. However, the risk of premature foaming at the relatively low melt temperatures and blowing agent contents of 0.5–1.5% is rather low. In further experiments, the length and the width of the spiders were varied. In Figure 3.20, the thickness distribution of LDPE sheets at a melt temperature of 111°C is plotted for spider geometries III–V. The wider spider geometry (Spider V) leads to a wider thin section in the foam sheet than the standard geometry (Spider III). However, the thickness of the foam sheet at the spider line is higher when the wider spider geometry is used (Figure 3.20, right side). The shorter spider geometry (Spider IV) produces foam sheets with a smaller spider line that is as thin as that of Spider III. The other melt temperatures show similar results.

2.4

LDPE

1.8

2.2

1.6

2.0

Thickness (mm)

Thickness (mm)

2.0

1.4 1.2 1.0 0.8 Geometry I Geometry II

0.6 0.4 0

45 90 135 180 225 270 315 360

Circumference of the foam tube (°)

PP

1.8 1.6 1.4 1.2 1.0 0.8 0

45 90 135 180 225 270 315 360

Circumference of the foam tube (°)

FIGURE 3.19 Influence of the downstream angle on the thickness profile. [From Heinz, R. Prozessoptimierung bei der Extrusion thermoplastischer Schäume mit CO2 als Treibmittel (Process Optimization for the Extrusion of Thermoplastic Foams Using CO2 as a Blowing Agent). PhD Thesis at RWTH Aachen, 2002.]

61259_C003.indd 88

10/25/2008 12:35:33 PM

2.6 2.4 2.2 2.0 1.8 1.6 1.4 1.2 1.0 0.8 Geometry III 0.6 Geometry IV (short) 0.4 Geometry V (wide) 0.2 45 90 135 180 225 Circumference of the foam tube (°)

89

1.6 1.4 Thickness (mm)

Thickness (mm)

Foam Extrusion Using Carbon Dioxide as a Blowing Agent

1.2 1.0 0.8 0.6 0.4

LDPE Melt temperature 111°C Die temperature 105°C 125 130 135 140 Circumference of the foam tube (°)

FIGURE 3.20 Influence of the spider geometry on the thickness profile (LDPE). [From Heinz, R. Prozessoptimierung bei der Extrusion thermoplastischer Schäume mit CO2 als Treibmittel (Process Optimization for the Extrusion of Thermoplastic Foams Using CO2 as a Blowing Agent). PhD Thesis at RWTH Aachen, 2002.]

The experiments with PP were conducted at a melt temperature of 179°C. The results are depicted in Figure 3.21. In total, the thickness distribution of the PP foam sheets is only slightly affected by the spider geometry. Again, the melt temperature shows no significant effect on the thickness distribution.

Thickness (mm)

2.2

1.8

Geometry III Geometry IV (short) Geometry V (wide)

2.0 1.8 1.6 1.4 1.2

1.6 Thickness (mm)

2.4

PP Melt temperature 179°C Die temperature 170°C

1.4 1.2

1.0

1.0 45 90 135 180 225 Circumference of the foam tube (°)

0.8 120 125 130 135 140 Circumference of the foam tube (°)

FIGURE 3.21 Influence of the spider geometry on the thickness profile (PP). [From Heinz, R. Prozessoptimierung bei der Extrusion thermoplastischer Schäume mit CO2 als Treibmittel (Process Optimization for the Extrusion of Thermoplastic Foams Using CO2 as a Blowing Agent). PhD Thesis at RWTH Aachen, 2002.]

61259_C003.indd 89

10/25/2008 12:35:33 PM

90

Polymeric Foams

3.5 Influence of the Pressure Profile in the Extrusion Die on the Foam Structure According to the classical nucleation theory (Equations 3.7 and 3.8), the heterogeneous nucleation rate Nhet is dependent on the difference between the solubility pressure pS of the blowing agent and the surrounding pressure p0.25 Since the pressure is not released abruptly in the foam extrusion process, some cells are nucleated before the melt has exited the die; that is, before the pressure of the polymer/blowing agent mixture has dropped down to ambient pressure. This means that some cells begin to grow whereas others have not been nucleated yet. Consequently, nucleation, bubble growth, and thus the resulting foam structure depend on the rate of pressure release. Nhet  f  c  e -DG

/k  T

het

16  p  g 3 DGhet  ____________2  g(q) 3  (pS  p0)

(3.7) (3.8)

Park et al. validated this assumption in microcellular foaming experiments using different nozzles with dissimilar length and diameter.26 In their tests with high impact polystyrene (HIPS) charged with 10% of CO2, they found that the cell density of the foams increases significantly with increasing pressure gradient in the nozzle. Heinz conducted experiments to study the influence of pressure gradient at the die exit on the foam structure of LDPE foam sheets.20,27,28 Therefore, an annular die with exchangeable die lips is used. Three pairs of die lips with different die gaps and lengths are constructed in order to vary the pressure gradient. The total pressure loss of the die is equal for every pair of die lips, so that all extruding conditions can be kept constant during the tests. Figure 3.22 shows the die exit with the different lip geometries schematically. The shape of the lips is derived using computer software for flow calculations in axially symmetric flow channels developed at the Institute of Plastics Processing (IKV, Aachen, Germany). The calculations are based on viscosity data of LDPE charged with 1% CO2 at a temperature of 130°C. The results of these calculations in Figure 3.23 show the pressure course over the flow length in the die for three different pairs of die lips. Upstream of the die lips, the pressure loss is rather low in order to prevent a premature foaming in the die. The pressure is then released very quickly to initiate cell nucleation. The different pressure gradients are represented by the different slopes of the plots at the end of the die (270–290 mm). Due to the simplifications and assumptions in the simulation, the pressure loss encountered in the die during preliminary foam extrusion

61259_C003.indd 90

10/25/2008 12:35:34 PM

91

Foam Extrusion Using Carbon Dioxide as a Blowing Agent

FIGURE 3.22 Die lip geometries. [From Heinz, R. Prozessoptimierung bei der Extrusion thermoplastischer Schäume mit CO2 als Treibmittel (Process Optimization for the Extrusion of Thermoplastic Foams Using CO2 as a Blowing Agent). PhD Thesis at RWTH Aachen, 2002.]

tests differs from the calculated values. Therefore the die lips are reworked by adjusting the length and the die gap in order to obtain the same total pressure loss in the die with all three pairs of lips. The geometries of the reworked die lips are listed in Table 3.3.20 The pressure gradients of the reworked die lips are calculated via the pressure loss divided by the residence time in the die lip section. As shown in Table 3.3, they differ by almost two orders of magnitude. Higherpressure gradients at the die exit would lead to corrugation of the foam

50

Pressure (bar)

40 Die lips no. 1 Die lips no. 2 Die lips no. 3

30 20 10 0

18 kg/h LDPE 1% CO2 ϑM = 130°C 160

180

200

220

240

260

280

300

Flow length (mm) FIGURE 3.23 Simulated pressure course for different die lips. [From Heinz, R. Prozessoptimierung bei der Extrusion thermoplastischer Schäume mit CO2 als Treibmittel (Process Optimization for the Extrusion of Thermoplastic Foams Using CO2 as a Blowing Agent). PhD Thesis at RWTH Aachen, 2002.]

61259_C003.indd 91

10/25/2008 12:35:34 PM

92

Polymeric Foams

TABLE 3.3 Dimensions of the Reworked Die Lips Die Lips 1 2 3

Length (mm)

Die Gap (mm)

Pressure Loss (Bar)

Residence Time (s)

Pressure Gradient (MPa/s)

1.9 3.4 5.1

0.4 0.6 1.1

40 40 40

0.039 0.170 0.847

103 24 5

Source: Heinz, R. Prozessoptimierung bei der Extrusion thermoplastischer Schäume mit CO2 als Treibmittel (Process Optimization for the Extrusion of Thermoplastic Foams Using CO2 as a Blowing Agent). PhD Thesis at RWTH Aachen, 2002.

sheets that could not be handled. The residence time of polymer in the die lips is between 39 and 847 ms. The experiments with different pressure gradients are conducted with Lupolen 1810 H (Basell, Hoofddorp, the Netherlands) with 0.8% CO2 at a melt temperature of 116°C. The mass throughput rate is 20 kg/m3. As a nucleating agent, either a talc masterbatch concentrate (Hydrocerol CT 316, Clariant Masterbatch, Lahnstein, Germany) or a chemical blowing agent masterbatch concentrate (Hydrocerol CF 20 E, Clariant Masterbatch) are added in different concentrations. The cell structure of the foams is analyzed using micrographs of the foamed sheets. The mean cell diameter and the cell density; that is, the number of cells per volume, are measured to evaluate the foam quality. Figure 3.24 shows the cell density and the cell diameter in dependence on the nucleating agent content (talc) for three different pressure gradients. For example, at a constant nucleating agent content of 0.6%, an increase of the pressure gradient from 5 to 24 MPa/s leads to an increase in cell density from 3.4  103 to 1.5  104 cells/cm3. Simultaneously, the mean cell diameter decreases from 1.15 mm to 0.67 mm. When the pressure gradient is further increased up to 103 MPa/s, the cell density rises up to 2.8  104 cells/cm3 whereas the mean cell diameter decreases to 0.59 mm. Within the considered range of nucleating agent content, this effect can be observed in all tests. In total, the experiments show three main effects20,28: • The influence of the pressure gradient on the foam structure decreases with increasing pressure gradient. Although the difference between 103 and 24 MPa/s is higher, the increase in cell density is higher when the pressure gradient rises from 5 to 24 MPa/s. • The influence of talc content on the foam structure decreases with increasing pressure gradient. This becomes obvious when the dependence of cell diameter on the nucleating agent content is compared between 5 and 103 MPa/s.

61259_C003.indd 92

10/25/2008 12:35:34 PM

93

Foam Extrusion Using Carbon Dioxide as a Blowing Agent

106

Δp/Δt = 103 MPa/s Δp/Δt = 24 MPa/s Δp/Δt = 5 MPa/s

105

104

1.2 Average cell diameter (mm)

Cell density (1/cm3)

107

1.0

0.8

0.6

0.4 103 0.6 0.8 1.0 1.2 1.4 1.6 1.8 Nucleating agent content (%)

0.6 0.8 1.0 1.2 1.4 1.6 1.8 Nucleating agent content (%)

FIGURE 3.24 Influence of the pressure gradient on cell density and cell size (talc). [From Heinz, R. Prozessoptimierung bei der Extrusion thermoplastischer Schäume mit CO2 als Treibmittel (Process Optimization for the Extrusion of Thermoplastic Foams Using CO2 as a Blowing Agent). PhD Thesis at RWTH Aachen, 2002.]

• At high talc contents, the cell density and the cell diameter seem to reach a limit asymptotically, which is not crossed even if the nucleating agent content is further increased. The micrographs of foamed sheets produced with the different die lips in Figure 3.25 illustrate the effect of pressure gradient on foam structure. At constant process conditions and a nucleating agent content of 1.8% talc, only the pressure gradient at the die exit was varied in the range from 5 to 103 MPa/s. In contrast to this, when Hydrocerol CF 20 E is used as nucleating agent, the cell densities of the foam sheets increase by one order of magnitude while the cell diameters decrease tremendously below 250 μm. However, the influence of pressure gradient on the foam structure is weaker when Hydrocerol is used instead of talc. Figure 3.26 presents the cell density and the cell diameter of the foam sheets produced with Hydrocerol CF 20 E at different pressure gradients. As observed with talc, the cell density increases with the pressure gradient at constant nucleating agent content, but the courses of 24 MPa/s and 103 MPa/s do not differ very much. Also the cell diameter decreases when the pressure gradient is raised from 5 to 24 MPa/s. A further increase in pressure gradient to 103 MPa/s does not show any effect. However, at high nucleating agent contents of 0.6% Hydrocerol, the cell diameter of the foams produced with 103 MPa/s at the die exit increases again. Heinz explains this phenomenon by collapsing cells.20 Figure 3.27 shows micrographs of foams produced with 0.3%

61259_C003.indd 93

10/25/2008 12:35:34 PM

94

Polymeric Foams

FIGURE 3.25 Influence of the pressure gradient on the foam structure (1.8% talc). [From Heinz, R. Prozessoptimierung bei der Extrusion thermoplastischer Schäume mit CO2 als Treibmittel (Process Optimization for the Extrusion of Thermoplastic Foams Using CO2 as a Blowing Agent). PhD Thesis at RWTH Aachen, 2002.]

Hydrocerol at various pressure gradients. According to the diagrams, the difference in cell size and cell density at various pressure gradients is lower than with talc. The pressure gradient at the die exit also influences the foam density. Figure 3.28 shows the density of the foams produced with talc in dependence on the nucleating agent content for three different pressure gradients. All three plots of foam density exhibit a pronounced minimum. The higher the pressure gradient, the lower the nucleating agent content at which the density becomes a minimum.

0.30 Δp/Δt = 103 MPa/s Δp/Δt = 124 MPa/s Δp/Δt = 5 MPa/s

106

105

Average cell diameter (mm)

Cell density (1/cm3)

107

0.25

0.20

0.15

0.10 0.2

0.4

0.6

Nucleating agent content (%)

0.2

0.4

0.6

Nucleating agent content (%)

FIGURE 3.26 Influence of the pressure gradient on cell density and cell size (Hydrocerol). [From Heinz, R. Prozessoptimierung bei der Extrusion thermoplastischer Schäume mit CO2 als Treibmittel (Process Optimization for the Extrusion of Thermoplastic Foams Using CO2 as a Blowing Agent). PhD Thesis at RWTH Aachen, 2002.]

61259_C003.indd 94

10/25/2008 12:35:35 PM

95

Foam Extrusion Using Carbon Dioxide as a Blowing Agent

FIGURE 3.27 Influence of the pressure gradient on the foam structure (0.3% Hydrocerol). [From Heinz, R. Prozessoptimierung bei der Extrusion thermoplastischer Schäume mit CO2 als Treibmittel (Process Optimization for the Extrusion of Thermoplastic Foams Using CO2 as a Blowing Agent). PhD Thesis at RWTH Aachen, 2002.]

Similar results can be observed when little amounts of chemical blowing agents (Hydrocerol) are used as a nucleator. Figure 3.29 shows the course of density over nucleating agent content for the three pressure gradients. Again, a minimum can be detected in the plots but it is reached at lower nucleating agent contents. Furthermore, the lowest densities are achieved in a small range of nucleating agent content; that is, the shift of minimal density in dependence on pressure gradient is less pronounced than it is with talc. The rise in density with increasing nucleating agent content is very high when the chemical blowing agent is used as a nucleating agent. This is caused by emerging cell collapse at high nucleating agent contents.

0.34 Δp/Δt = 103 MPa/s Δp/Δt = 24 MPa/s Δp/Δt = 5 MPa/s

Density (g/cm3)

0.32 0.30 0.28 0.26 0.24 0.22 0.20 0.5

1.0 1.5 Nucleating agent content (%)

2.0

FIGURE 3.28 Influence of the pressure gradient on foam density (talc). [From Heinz, R. Prozessoptimierung bei der Extrusion thermoplastischer Schäume mit CO2 als Treibmittel (Process Optimization for the Extrusion of Thermoplastic Foams Using CO2 as a Blowing Agent). PhD Thesis at RWTH Aachen, 2002.]

61259_C003.indd 95

10/25/2008 12:35:36 PM

96

Polymeric Foams

0.35 Δp/Δt = 103 MPa/s Δp/Δt = 24 MPa/s Δp/Δt = 5 MPa/s

Density (g/cm3)

0.30

0.25

0.20 0.1

0.2

0.3

0.4

0.5

0.6

0.7

Nucleating agent content (%) FIGURE 3.29 Influence of the pressure gradient on foam density (Hydrocerol). [From Heinz, R. Prozessoptimierung bei der Extrusion thermoplastischer Schäume mit CO2 als Treibmittel (Process Optimization for the Extrusion of Thermoplastic Foams Using CO2 as a Blowing Agent). PhD. Thesis at RWTH Aachen, 2002.]

For both nucleating agents, the content that produces foam with the lowest density is dependent on the pressure gradient. If a low foam density is desired, the most efficient nucleating agent content can be defined as that at which the lowest density is reached. Figure 3.30 shows this most efficient content in dependence on the pressure gradient for the considered nucleating agents.

Nucleating agent content (%)

2.0 Talc Hydrocerol 1.5

1.0

0.5

0

20

40 60 80 Pressure gradient (MPa/s)

100

120

FIGURE 3.30 Influence of the pressure gradient on the most efficient nucleating agent content. [From Heinz, R. Prozessoptimierung bei der Extrusion thermoplastischer Schäume mit CO2 als Treibmittel (Process Optimization for the Extrusion of Thermoplastic Foams Using CO2 as a Blowing Agent). PhD Thesis at RWTH Aachen, 2002.]

61259_C003.indd 96

10/25/2008 12:35:36 PM

97

Foam Extrusion Using Carbon Dioxide as a Blowing Agent

3.6 Nomenclature B C e F G H K L m˙ Nhet p0 pS T T0 V˙ m Ghet p a s – g˙ h– rm rP rS q wS

(m) — — (s1) — (m) (J/K) (m) (kg/s) [1/(m3 · s)] (Pa) (Pa) (K) (K) (m3/s) ( J) (Pa) (1/K) (N/m) (s1) (Pa · s) (kg/m3) (kg/m3) (kg/m3) (°) —

Width of flow channel Concentration of heterogeneous nucleation sites Geometric factor for rectangular slits Frequency factor Geometric factor Height of flow channel Boltzman’s constant Length Mass flow rate Heterogeneous cell nucleation rate Ambient pressure Solubility pressure Temperature Reference temperature Volume flow of melt/blowing agent mixture Change in Gibbs free energy Change in pressure Expansion coefficient Interfacial tension Representative shear rate Viscosity Density of melt/blowing agent mixture Density of polymer melt Density of solvent (blowing agent) Wetting angle Weight fraction of solvent (blowing agent)

References 1. Han, C. D. and Villamizar, C. A. “Studies on structural foam processing: I. The rheology of foam extrusion.” Polymer Engineering and Science 18 (1978): 687–698. 2. Han, C. D. and Ma, C. Y. “Rheological properties of mixtures of molten polymer and fluorocarbon blowing agent: I. Mixtures of low-density polyethylene and fluorocarbon blowing agent.” Journal of Applied Polymer Science 28 (1983): 831–850. 3. Han, C. D. and Ma, C. Y. “Rheological properties of mixtures of molten polymer and fluorocarbon blowing agent: II. Mixtures of polystyrene and fluorocarbon blowing agent.” Journal of Applied Polymer Science 28 (1983): 851–860.

61259_C003.indd 97

10/25/2008 12:35:37 PM

98

Polymeric Foams

4. Gendron, R., Daigneault, L. E., Dumoulin, M. M., Dufour, J., and Caron, L.-M. “Rheological measurements of PS/blowing agent mixtures.” In Proceedings of the 10th Annual Meeting, Polymer Processing Society (PPS), Akron, OH, 1994: 415–416. 5. Gendron, R., Daigneault, L. E., and Caron, L.-M. “Rheological behavior of polystyrene/blowing agent mixtures.” In Proceedings of the Annual Technical Conference (ANTEC), Society of Plastics Engineers (SPE), Indianapolis, IN, 1996: 1118–1122. 6. Kropp, D. and Michaeli, W. “Rheological melt behaviour during foam extrusion using carbon dioxide as blowing agent.” In Proceedings of the Polymer Processing Society European Meeting, Stuttgart, 1995. 7. Michaeli, W., Kropp, D., Pfannschmidt, O., Rogalla, A., and Seibt, S. “Kohlendioxid als Verarbeitungshilfs- und Treibmittel beim Spritzgießen und Extrudieren von Thermoplasten.” Gummi Fasern Kunststoffe (GAK) 49 (1996): 652–661. 8. Chmiel, H. and Schümmer, P. “Eine neue Methode zur Auswertung von Rohrrheometer-Daten.” Chemie–Ingenieur–Technik (CIT) 43 (1971): 1257–1259. 9. Schümmer, P. and Worthoff, R. H. “An elementary method for the evaluation of a flow curve.” Chemical Engineering Science 33 (1978): 759–763. 10. Michaeli, W. Extrusion Dies for Plastics and Rubber: Design and Engineering Computations, 2nd revised edition. Carl Hanser Verlag, München, 1992. 11. Wortberg, J. “Werkzeugauslegung für die Ein- und Mehrschichtextrusion.” PhD Thesis, RWTH Aachen, 1978. 12. Gundert, F. and Wolf, B. A. “Polymer-Solvent Interaction Parameters.” In Polymer Handbook, ed. J. Brandrup and E. H. Immergut. Wiley, New York, 1989. 13. Menges, G. “4. korrigierte und aktualisierte.” In Werkstoffkunde Kunststoffe. Auflage, Carl Hanser Verlag, München, 1998. 14. N. N. VDI-Wärmeatlas. VDI-Verlag, Düsseldorf, 1984. 15. Kropp, D. Extrusion von amorphen thermoplastischen Schäumen niedriger Dichte mit CO2 als Treibmittel. Report of research project, Aachen, 1998. 16. Kropp, D. Extrusion thermoplastischer Schäume mit alternativen Treibmitteln (Extrusion of thermoplastic foams using alternative blowing agents). PhD Thesis, RWTH Aachen, 1999. 17. Niggemann, M. Untersuchung des Fließverhaltens bei der Schaumextrusion von Polystyrol mit CO2 als Treibmittel. Unpublished Thesis at RWTH Aachen, 1995. 18. Williams, M. L., Landel, R. F., and Ferry, J. D. “The temperature dependence of relaxation mechanisms in amorphous polymers and other glass-forming liquids.” Journal of the American Chemical Society 77 (1955): 3701–3707. 19. Stadlbauer, M. “Polypropylen Schaum.” In Proceedings of Thermoplastische Schaumstoffe—Verarbeitungstechnik und Möglichkeiten der Prozessanalyse, Institut für Kunststoffverarbeitung an der RWTH Aachen, Aachen, 2004. 20. Heinz, R. Prozessoptimierung bei der Extrusion thermoplastischer Schäume mit CO2 als Treibmittel (Process Optimization for the Extrusion of Thermoplastic Foams Using CO2 as a Blowing Agent). PhD Thesis at RWTH Aachen, 2002. 21. Herrmann, T., Analyse der scherinduzierten Blasenbildung bei der thermoplastischen Schaumextrusion. Thesis at RWTH Aachen, 2000.

61259_C003.indd 98

10/25/2008 12:35:37 PM

Foam Extrusion Using Carbon Dioxide as a Blowing Agent

99

22. Lee, S. T. “Shear effects on thermoplastic foam nucleation.” Polymer Engineering and Science 33 (1993): 418–422. 23. Throne, J. L. Thermoplastic Foam Extrusion: An Introduction. Carl Hanser Verlag, München, 2004. 24. Lauterberg, W. “Dickentoleranzen bei Polyethylenschaumfolien.” Plaste und Kautschuk 25 (1978): 294–295. 25. Colton, J. S. and Suh, N. P. “The nucleation of microcellular thermoplastic foam with additives: Part I: theoretical considerations.” Polymer Engineering and Science 27 (1987): 485–492. 26. Park, C. B., Baldwin, D. F., and Suh, N. P. “Effect of the pressure drop rate on cell nucleation in continuous processing of microcellular polymers.” Polymer Engineering and Science 35 (1995): 432–440. 27. Michaeli, W. and Heinz, R. “Extrusion of thermoplastic foams using CO2 as blowing agent.” In Proceedings of the Blowing Agents and Foaming Processes Conference, Rapra Technology Ltd., Heidelberg, 2002. 28. Michaeli, W. and Heinz, R. Analyse des Einflusses der Werkzeuggeometrie und des Nukleierungsmittels auf die Schaumqualität bei der thermoplastischen Schaumextrusion mit CO2 als Treibmittel. Report of Research Project, Aachen, 2002.

61259_C003.indd 99

10/25/2008 12:35:37 PM

61259_C003.indd 100

10/25/2008 12:35:37 PM

4 Processes and Process Analysis of Foam Injection Molding with Physical Blowing Agents Walter Michaeli, Axel Cramer, and Laura Flórez

CONTENTS 4.1 Characteristics of Injection Molded Thermoplastic Foams ....... 4.1.1 Benefits and Drawbacks of the Foam Injection Molding Process ................................................. 4.1.2 Chemical and Physical Blowing Agents ......................... 4.1.3 Properties of the Polymer Matrix ..................................... 4.1.4 Mechanisms of Nucleation and Bubble Growth ............ 4.2 Process Concepts for Foam Injection Molding (FIM) .............. 4.2.1 Fundamentals of the Foam Injection Molding Process ................................................................. 4.2.2 Injection of the Blowing Agent into an Extruder ........... 4.2.3 Injection of the Blowing Agent into the Plastification Unit .......................................................... 4.2.4 Injection of the Blowing Agent into a Special Gassing Unit ........................................................ 4.2.5 Processing of Pellets Pre-Charged with Blowing Agent ..................................................................... 4.2.6 Injection of the Blowing Agent with Aid of a Special Nozzle .............................................................. 4.3 Experimental Analysis of the Processing Parameters in Foam Injection Molding ....................................

61259_C004.indd 101

102 102 104 107 108 114 114 114 115 116 118 118 120

10/25/2008 12:36:12 PM

102

Polymeric Foams

4.3.1 Experiments with the IKV Blowing Agent Injection Nozzle .................................................................. 4.3.1.1 Influence of the Process Parameters on the Foamed Part Characteristics .................... 4.3.1.2 Investigations on Cycle Time Reduction ............ 4.3.2 Experiments on Pre-Loaded Polycarbonate ................... 4.3.2.1 Morphological Characterization Methods ........... 4.3.2.2 Influence of the Process Parameters on the Foamed Part Characteristics ................................ 4.4 Optimization of the Surface Quality of Foamed Injection Molded Parts ................................................................ 4.4.1 Occurrence of Silver Streaks ............................................. 4.4.2 Possibilities of Increasing the Surface Quality Through Mold and Process Technology ......................... 4.4.3 Investigation of “Breathing” Molds and Gas Counter-Pressure ................................................................ 4.5 Summary ....................................................................................... Acknowledgments .............................................................................. References .............................................................................................

121 121 123 125 126 128 131 132 134 137 139 140 140

4.1 Characteristics of Injection Molded Thermoplastic Foams 4.1.1 Benefits and Drawbacks of the Foam Injection Molding Process Injection-molded foams are counted among the most promising developments to lower material consumption and reduce both cycle time and molding tonnage in the injection molding of plastic parts. Their convenient specific material properties, besides their processing and performance advantages, make them the ideal choice for the production of different kinds of structural parts. These types of materials, also known as structural foams, have a typical “sandwich” morphology, characterized by a foamed core surrounded by a compact skin. In comparison to a compact part, which exhibits an almost constant density profile through its cross-section, structural foams have a density close to that of the compact material in the surrounding skin, where the largest concentration of material is present, and a lower density in the part’s middle, where the closed-cell foamed core is found. This particular configuration is responsible for some significant advantages during the processing and the service life of the molded parts. The most obvious effect of foam injection molding is the density reduction leading both to raw material savings and to a decrease in the part’s weight, offering not only economical but also design benefits. By keeping the same weight, a foamed part will exhibit higher specific stiffness, specifically when

61259_C004.indd 102

10/25/2008 12:36:12 PM

Processes and Process Analysis of Foam Injection Molding

103

bending forces are applied. This corresponds to a larger geometric moment of inertia in comparison to a compact molded part of the same weight, because of the accumulation of material away from the part’s neutral axis.1 Due to the lower pressure level during injection, machines with lower clamping forces can be used in comparison to conventional injection molding.2 The holding pressure is often not needed, and using unchanged processing temperatures a reduction of the internal mold pressure of up to 50% is possible.3 If the viscosity-reducing effect of the blowing agent is used for the reduction of the processing temperature, a decrease in cooling time is also possible. Furthermore, the reduction of internal stresses, the reduction of warpage, and the decrease of sink marks enhance the dimensional accuracy of ribs, openings, or dimensional changes in the part’s cross-section. However, if a part is foamed without changing its dimensions, its mechanical properties drop as a function of the weight reduction. It has been proven that the reduction in the elastic modulus and the part’s resistance can be correlated with the density reduction: X Xc

r rc

( )

___f  __f

n

(4.1)

where X is either the tensile or flexural stress or elastic modulus,  is the density, and the subscripts f and c represent respectively the foamed and the compact polymer. The exponent of this correlation, n, is determined experimentally, and commonly takes values between 1.0 and 2.0.4 This correlation makes clear that the properties of the foam drop at a steeper rate than the density reduction. Current studies show that the drop in mechanical properties can be counteracted through an optimization of the foam morphology. Another disadvantage is the achievable surface quality of foamed parts, which is rather poor in comparison to compact parts.5 By further process developments and the combination of already existing mold concepts (gas counter-pressure, breathable molds, etc.) these drawbacks are subjected to prospective optimizations. Some of the earliest applications of structural foams were made in the fields of appliances and furniture, where foaming was achieved with chemical blowing agents. In these cases the objective was to improve the dimensional stability while reducing the part’s weight. Housings for electric appliances have also traditionally made use of the advantages of foam injection molding, to eliminate the sink marks and the warpage typical of their ribbed geometries. Larger parts, such as front-ends and door panels for automotive applications, are currently being developed, and an interesting trend arises in the combination of the foam injection molding with other injection processes, such as multi-component or back molding, to achieve an optimized properties profile in the final part. A typical application is presented in Figure 4.1.

61259_C004.indd 103

10/25/2008 12:36:12 PM

104

Polymeric Foams

FIGURE 4.1 Application produced through foam injection molding.

The achievable weight reduction depends strongly on the thickness of the part and the type of foaming agent used, but for typical applications in injection molding, with wall thicknesses between 1.0 and 3.0 mm, weight reductions between 10% and 20% are common. For thicker applications weight reductions of up to 60% are possible. Due to their broad application potential, injection molded thermoplastic foams have gained a well-established position during the last few years. The process of foam injection molding is growing in importance, triggered by the recent developments in blowing agents and processing technologies. The associated reduction in consumption of raw material is gaining relevance due to the increase in the price of resin, and in the automotive industry, the ever-tightening regulations that head for the reduction of vehicle’s weight drive a struggle for the production of reliable foamed parts. 4.1.2 Chemical and Physical Blowing Agents The production of thermoplastic foams is achieved with the aid of foaming agents, either physical or chemical, that can be dosed into the polymer in different ways. Chemical blowing agents have existed in many foaming applications for a long time. However, the straight development of different process technologies for the production of physically blown foams is responsible for the rising demand on foamed parts. Whereas the development of marketable techniques for the extrusion of physically blown foams had already begun at the end of the 1970s, research activities in the injection molding area has intensified since the middle of the 1990s.1 The chemical blowing agents are added to the polymer pellets in solid form and are activated through addition of heat, releasing a fluid, mostly nitrogen, carbon dioxide, or water.6 However, the appearance of residual products is a disadvantage, given the fact that they can represent up to 70% of the final composition of the agent.1,2 Their decomposition can lead to a degradation of the polymer matrix, to a decrease in mechanical properties, to coloration of the part and to corrosion and contamination of the mold. For these reasons only a defined amount of foaming agent ought to be incorporated into the polymer melt when using chemical blowing agents. A list of chemical blowing agents is provided in Table 4.1.

61259_C004.indd 104

10/25/2008 12:36:12 PM

105

Processes and Process Analysis of Foam Injection Molding

TABLE 4.1 Commonly Used Chemical Blowing Agents Chemical Denomination

Decomposition Abbreviation Temperature (°C)

Azodicarbonamide

ADC

Main Gas Components

200–220

250–320

155–220

150–300

OBSH

140–165

120–150

N2, CO, CO2 (NH3) N2, CO, CO2 (NH3) N2, H2O

5-PT TSS

240–250 215–235

190–210 120–140

N2 N2, CO2

TSH

110–140

120–140

N2, H2O

Bicab

120–150 200–220

120–170 90–120

CO2, H2O CO2, H2O

Modified ADC 4,4-Oxybis (benzenesulfonylhydrazide) 5-Phenyltetrazole p-Toluenesulfonylsemicarbazide p-Toluenesulfonylhydrazide Sodium carbonate Citric acid

Rate of Gas Yield (ml/g)

Foaming agents like carbon dioxide, nitrogen, hydrocarbonates like pentane, and even water,1 are dosed directly into the polymer melt and are referred to as physical foaming agents. Compared with the chemical agents these foaming agents do not evoke residual products, but their dosage is usually more complex from the technological point of view, due to higher demands on the control and the transient fluid incorporation in the injection molding process. A list of physical blowing agents can be found in Table 4.2.

TABLE 4.2 Commonly Used Physical Blowing Agents Blowing Agent

Chemical Formula

Molar Weight (g/mol)

Boiling Temperature (°C)

Flammable

ODP

GWP

Isobutane

C4H10

58.1

11.7

Yes

—

—

Cyclopentane Isopentane CFC-11 HCFC-22

C5H10 C5H12 CFCl3 CHF2Cl

70.1 72.1 137.4 86.5

49.3 29.0 23.8 40.8

Yes Yes No No

— — 1.0 0.05

0.00275 — 1.0 0.35

HCFC-142b

CF2ClCH3

100.5

9.2

Yes

0.05

0.38

HFC-134a

CH2FCF3

102.0

26.5

No

—

0.27

Nitrogen

N2

28.0

195.7

No

—

—

Carbon dioxide

CO2

44.0

56.5

No

—

0.00025

61259_C004.indd 105

10/25/2008 12:36:12 PM

106

Polymeric Foams

Thanks to the developments achieved in control and valve technology, physical foaming agents, particularly the inert gases, are increasingly important. The type and amount of foaming agent added to the thermoplastics will determine the achievable foam density as well as the required engineering system for its production. The possible density reductions achievable with each type of foaming agent are illustrated in Figure 4.2, taking as an example the process of film foam extrusion. As we can see, when higher foaming grades are to be reached, foaming should be achieved through physical blowing agents because of the great extent of decomposition products associated with the use of large amounts of chemical blowing agents. One of the factors to be considered concerning the selection of the appropriate foaming agent in a specific foam application is its solubility in the selected thermoplastic material. The cell nucleation is originated by a pressure drop as a consequence of a thermodynamic instability in the mixture polymer and blowing agent. A higher pressure drop rate implies a higher nucleation rate and the formation of a larger number of cells. If the diffusion rate of the blowing agent through the polymer is slow, a larger number of cells is formed due to the longer time needed for the blowing agent to diffuse into existing cells. Another issue is, of course, the cost of the blowing agent and last but not least its environmental and toxicological impact. Due to their compliance of this particular requirement, nitrogen and carbon dioxide are broadening their acceptance as physical foaming agents.3 On a volumetric basis, carbon dioxide can be up to 18 times more soluble than nitrogen in semi-crystalline polymers, such as LDPE, at 25°C and a pressure of 1 bar. Under the same conditions, in amorphous polymers such as PVC, the solubility of carbon dioxide is 20 times larger than that of nitrogen.5 To guarantee the reproducibility of the dosage of carbon dioxide, it is advisable to bring its condition to a supercritical state (above 1000

Density (kg/m3)

800

Chemically foamed

Physically foamed

600 400 200 0

FIGURE 4.2 agents.

61259_C004.indd 106

Achievable densities in foam extrusion with chemical and physical blowing

10/25/2008 12:36:12 PM

107

Processes and Process Analysis of Foam Injection Molding

Critical point:

p/pc

CO2 Temp.: 31.04 °C Pressure: 73.83 bar

Supercritical state

N2 Temp.: –146.9 °C Pressure: 33.98 bar

1 Solid

Fluid

Gaseous State at storage N2

CO2

Triple point

State at injection N2

CO2

J/Jc

1

FIGURE 4.3 Phase diagram of carbon dioxide and nitrogen.

critical values of temperature and pressure of 31°C and 73.83 bar respectively). Once this value is reached, neither more condensation nor phase changes will take place, independently of the pressure and the temperature induced. The states of nitrogen and carbon dioxide at storage and at injection are shown in Figure 4.3. The properties of the supercritical fluids are in between those of the liquid and the gas phase.6,7 In this area these agents exhibit a low viscosity, low surface tension, comparatively good diffusion properties and a similar fluid molar volume (Table 4.3). These properties are responsible for an excellent dissolving power as well as for fast loading response. As a consequence, higher nucleation densities are achievable, and therefore a finer cell structure of the foam. The processing of plastics with supercritical fluids is a technology that has been around for years in industry, and used, for example, in gas-assisted injection molding (GAIM). 4.1.3 Properties of the Polymer Matrix In principle, all thermoplastic materials can be foamed. However, the effort required to foam a thermoplastic material with sufficient quality TABLE 4.3 Physical Properties of Carbon Dioxide Parameter

Unit

Gas

3

3

Density

g/cm

Viscosity

g/(cm·s)

61259_C004.indd 107

10

104

Fluid

Super Critical

1

0.6

102

104

10/25/2008 12:36:12 PM

108

Polymeric Foams

specifications varies widely depending on its type. Independent of the process, a material will require less effort if it exhibits a broad processing window. This processing window is strongly subjected to the dependence of the viscosity on the melt temperature. For the formation of cells a low viscosity is advantageous, but if the viscosity is too low after cell growth, the foam collapses. Due to their low viscosity, semi-crystalline polymers should be processed near to their crystallization temperature to stabilize the foam structure before collapsing. With branched polyolefins such as LDPE, because they exhibit strain hardening in their elongational viscosity, this feature enables the stabilization over a broader range of temperature before crystallization takes place. A larger processing window will be given if the viscosity increases slowly with declining temperature. For this reason amorphous polymers that are processed at temperatures closer to their Tg have a larger processing window. Also, the plasticization induced by the physical blowing agent has an impact on the processing of amorphous polymers. The maximum viscosity at which a polymer may be foamed is determined by the maximum admissible viscosity in the process and by the processing technology. Above this viscosity it may occur, for example, that the flow resistance in the mold is too high to be overcome. On the other hand, the type and the amount of the foaming agent will establish the lower limit for the viscosity. If the viscosity of the melt is too low and the pressure in the forming cells too high, a collapse of the growing cells may result. In comparison to semi-crystalline thermoplastics, the amorphous ones are easier to process. Their processing window defined by the melt temperature is broader. As consequence of a slowly and homogeneously decreasing melt viscosity a more stable foam structure can be achieved. Whereas the limits for processing will be set through the flow behavior and therefore through the shear viscosity of the melt, the limit for cell growing will essentially be determined by the elongational viscosity.8 Figure 4.4 illustrates the dependence of the shear viscosity of semi-crystalline and amorphous thermoplastics as a function of temperature.9

4.1.4 Mechanisms of Nucleation and Bubble Growth A homogeneous foam structure can be reached through the charging of the polymer with a high amount of foaming agent, thereby achieving high levels of sorption concentration. The charging phase is characterized by mechanisms of sorption and diffusion. While the sorption capacity of a polymer describes its ability to achieve a maximum degree of foaming agent concentration, the diffusion capacity determines the velocity at which the fluid transport will occur. These mechanisms connect the formation of the nucleation sites and the subsequent stage of bubble growing. These

61259_C004.indd 108

10/25/2008 12:36:13 PM

Processes and Process Analysis of Foam Injection Molding

Amorphous

109

Semi-crystalline

Shear viscosity h

Processing limit

Limit by cell collapse

DJa

DJt

Temperature J DJa: Processing range of amorphous thermoplastics DJt: Processing range of semi-crystalline thermoplastics FIGURE 4.4 Viscosity as a function of temperature for amorphous and semi-crystalline thermoplastics.

separated mechanisms have consequences for process control. However, first of all a model developed by Pfannschmidt8 will be presented, demonstrating each mechanism separately. To understand the behavior of a polymer/foaming agent system the dependence of each phase on the temperature will be outlined first. Figure 4.5 shows the behavior of a polymer under variations of pressure and temperature. For example, if the temperature is increased in the system, the velocity of the molecular movement and the kinetic energy will be increased as well. As a consequence of this energy increase, the number and frequency of collisions between molecules will rise, resulting in an increase of the residence volume of the molecules. On the other hand, the volume of the system will decline due to its compressibility if the pressure is increased in the fluid phase. T0, p0, V0

T1 > T0, p1 = p0, V1 > V0

T2 = T1, p2 > p0, V2 > V1

1 0

2

Fluid-molecule

Residence volume

FIGURE 4.5 Impact of temperature and pressure on the blowing agent.

61259_C004.indd 109

10/25/2008 12:36:13 PM

110

Polymeric Foams

T0, p0, V0

T1 > T0, p1 = p0, V1 > V0 1

0

Polymer chain

Residence volume

FIGURE 4.6 Impact of temperature on the polymeric phase.

Figure 4.6 illustrates the behavior of a polymer under variations of temperature. In comparison to a gas, a polymer will exhibit a quasiincompressible behavior, particularly within the range of low temperatures. The separated macromolecules of the considered system are not fixed to a position, but oscillate due to the molecular movements inside a determined volume, that will be referred to as “residence volume.” When the temperature is increased the macromolecules will oscillate on a larger scale, increasing the residence volume. Once the process of absorption takes place, the molecules of the blowing agent surpass the surface of the polymer and fill the spaces in between the molecular chains of the polymer matrix. The incorporation of blowing agent molecules depends on the pressure and temperature. If the pressure of the system is increased, a higher concentration of the blowing agent will be dissolved. Figure 4.7 shows schematically the influence of pressure on the number of incorporated blowing agent molecules. A mathematical formula to describe the correlation between the concentration of the dissolved material in the system, denoted below as c, and the partial pressure of the blowing agent phase, p, is given through Henry’s equation, which proposes a linear dependence: cSρ

(4.2)

The S factor is referred as solubility, and is a substance dependent value. The blowing agent molecules located in between the polymer chains reduce the intermolecular forces. Thus the distance between the polymer chains can be increased independently of temperature, which in turn leads to the embedding of more fluid molecules. The separation of molecules will lead to an increased plastification effect, as well as to an enlargement of the volume of the polymer.

61259_C004.indd 110

10/25/2008 12:36:13 PM

Processes and Process Analysis of Foam Injection Molding

T0, p0, V0

0

111

T1 = T0, p1 > p0, V1 ≈ V0

1

FIGURE 4.7 Influence of pressure on the number of incorporated fluid molecules.

A mathematical formula that describes the solubility of polymers above the glass or melt temperature is the Flory–Huggins theory. In this theory the real system is replaced by a model where the relative amount of polymer and blowing agent is represented with a proportional number of elements. The model is able to represent the exchange between all the elements present in the system, and can thus deliver some thermodynamic parameters relevant for the sorption process, such as the free energy or the chemical potential of the system. For a deeper insight into this theory the reader is referred to References 8 and 10. The influence of temperature on the solubility of the system polymer/ blowing agent can be seen in Figure 4.8, for the case where the fluid molecules have a small diameter.

T0, p0, V0

0

T1 > T0, p1 = p0, V1 > V0 1

FIGURE 4.8 Influence of temperature on the number of incorporated fluid molecules.

61259_C004.indd 111

10/25/2008 12:36:14 PM

112

Polymeric Foams

If the blowing agent is absorbed, it will only fit in the free spaces between the polymer chains. The residence volume of each polymer chain will be increased by a temperature rise, thus leaving less space for the blowing agent molecules. Therefore, it can be expected that the solubility of the blowing agent in a polymer will be reduced. While the term “sorption” denotes the solubility of a fluid in a polymer, the term “diffusion” describes the transport of material driven by concentration gradients. The mathematical theory related to diffusion in isotropic media says that the amount of transported material is proportional to the concentration gradient (Fick’s first law):11,12 F  –D  grad c

(4.3)

The flow density F describes the number of particles that cross a determined sectional area per unit of time. The concentration of the diffusing material is denoted by c, and D is the diffusion coefficient. Attending to the equation of continuity, a differential equation for the diffusion phenomena can be stated as follows (Fick’s second law):11,12 ∂c  div(D  grad c) __ ∂t

(4.4)

The diffusion of a low molecular fluid in a polymer is based on the movement capability of the molecules (Brownian molecular movement) and is therefore dependent on the temperature. To achieve a homogeneous distribution of the concentration, the fluid molecules are induced to penetrate the polymer chain bonds. As the temperature arises, the movement freedom of the macromolecular chains is also increased as well as that of the fluid molecules. The increasing movement of the small fluid molecules facilitates a faster crossing of the entangled macromolecules. The stronger oscillation of the polymer chains leads to a further separation between the molecules, which in turn results in an increase in residence volume. When the concentration increases, so will also the velocity of diffusion. The foaming is initiated through the appearance of nucleation sites. If these nucleation sites are stable, they will lead to bubble growth. The apparition of such sites is dependent on the surface tension,13–15 the saturation pressure, and the pressure of the fluid in the polymer phase. According to the traditional nucleation theory, there are three possible mechanisms for the occurrence of this phenomenon: • Homogeneous nucleation • Heterogeneous nucleation • Combination of these two mechanisms. With homogeneous nucleation the nucleation sites develop as a consequence of density fluctuations and intermolecular movements. By contrast,

61259_C004.indd 112

10/25/2008 12:36:15 PM

Processes and Process Analysis of Foam Injection Molding

113

with heterogeneous nucleation the nucleation sites appear in the interfaces of a given phase (which may be impurities in the melt, nucleation agents, etc.). For heterogeneous nucleation a lower super-saturation grade will be required to initiate the bubble growth in comparison with homogeneous nucleation.16–19 The super-saturation of the polymer, and therefore the nucleation, can either appear through a sudden drop of pressure or through an increase in temperature. In processes of industrial application, like injection molding or extrusion, the foam expansion is initiated following a sudden pressure drop. While in injection molding this effect occurs when the gas-loaded melt reaches the mold, in foam extrusion the pressure is maintained through control systems or gear pumps, for example, to secure a certain pressure drop in the extrusion die. The velocity of the pressure drop can be controlled through an appropriate geometry. To understand the influence of the pressure drop velocity, the concurrent mechanisms of cell nucleation and cell growth have to be taken into account. During the pressure drop and the resulting instable thermodynamic state, stable nucleation sites will be created initiating the first growing cells. The dissolved blowing agent diffuses towards these cells to diminish the free energy of the system. Around the nucleated cells, regions of smaller blowing agent concentration remain, in which the appearance of nucleation sites is less probable. A further decrease in saturation pressure leads to an expansion of the existing cells and, eventually, a later cell nucleation. The appearance of a later cell nucleation depends on the low concentration regions of the blowing agents. If these regions overlap, there will be no further bubble development because the blowing agent will prefer to flow into the already formed cells. The influence of pressure drop velocity can be explained by Figure 4.9. Here, pressure versus time is presented for two systems. Due to the higher pressure gradient the nucleation velocity of system A is higher than that of system B. At the time t1 the pressure drop is larger for the system A than the pressure drop for the system B. At t2 the nucleation points grow. This leads to a diminishing of the blowing agent concentration available for further bubble growth. In the regions in which the blowing agent is not used, the velocity of the nucleation site’s formation grows exponentially for both systems, but always faster for system A, due to the high pressure difference. As the blowing agent is used for system A for a shorter time, the cells will coincide with each other sooner. Due to the fact that the cells in system B have more time to grow, the zones of low blowing agent concentration have more time to increase the size of the already formed cells in such a way that the concentration of blowing agent serves the size increase of the existing cells instead of the formation of new ones. Following this observation it can be stated that an increase in pressure drop velocity leads to the appearance of a larger number of nucleated cells. On the other hand, for slower pressure drops the diameter but not the number of firstformed cells will increase.

61259_C004.indd 113

10/25/2008 12:36:15 PM

114

Polymeric Foams

Pressure gradient

Polymer/fluid-solution

A > B

Fluid bubble

Pressure p

B

B A A t1

t2

Time t

t1

t2

FIGURE 4.9 Influence of pressure gradient on the formation of nucleation sites.

4.2 Process Concepts for Foam Injection Molding (FIM) 4.2.1 Fundamentals of the Foam Injection Molding Process In a similar way to the conventional injection molding process, in the production of injection molded thermoplastic foams a melt charged with blowing agent is processed by means of a plastification unit and afterwards injected at high velocity into a mold. Triggered by the abrupt drop of pressure during flow, the cells start to grow through expansion of the foaming agent. The final foamed part is composed of a solid outer skin surrounding a core of closed-cellular foamed material. Whereas no special modifications have to be made to the injection molding machine when chemical blowing agents are used (except the use of a shut-off nozzle and ideally a position control of the screw), physical blowing agents require a specially engineered system. Different process concepts for the incorporation of physical blowing agents compete with one another in foam injection molding. The differences in these concepts involve mainly the way the blowing agent is incorporated into the melt and, to a lesser degree, the processing itself. These concepts are presented in the following chapter. 4.2.2 Injection of the Blowing Agent into an Extruder One of the oldest processes for the injection molding of physically blown foams was developed in the 1970s in North America. Using this technology it was possible to transfer the manufacturing of foamed parts from the extrusion to injection molding technology for the first time. In this case,

61259_C004.indd 114

10/25/2008 12:36:15 PM

Processes and Process Analysis of Foam Injection Molding

115

Injection piston Injection unit

Shut-off valve Nitrogen inlet

Pressure release valve FIGURE 4.10

Extruder

Injection of blowing agent in an extruder.

the plastification of the polymer and the injection of the melt into the cavity are separated from each other. The blowing agent is injected into a continuously working extruder first of all. The polymer/blowing agent mixture is then homogenized and subsequently delivered to the space in front of the injecting piston, as shown in Figure 4.10. Afterwards, the dosed melt is injected with high speed into the cavity.4 Due to the approximately constant counter-pressure in the extrusion unit the blowing agent can be incorporated by pressure control. The leakage remains small even at high injection velocities to ensure high precision regarding the mass of the part because of the application of a piston injection unit. A decisive disadvantage of this concept is the fact that a special machine is necessary for the foam injection molding. The engineering of the process is advantageous, on the other hand, because the system design is relatively simple. In any case, the first physically blown foams could be produced by this concept. Their quality and the reproducibility were not very high, because the technology was not yet fully developed. Nevertheless, it worked satisfactorily over a long period and represented the counterpart to chemically blown foams, which were used predominantly in Europe. 4.2.3 Injection of the Blowing Agent into the Plastification Unit In the middle of the 1990s a new process was developed in the US, which incorporated the blowing agent into the melt over one or several injectors in the second half of the plastification cylinder. The screw equipped with mixing and shear elements provided the homogenization of the polymer/ blowing agent mixture. This process is relatively complex, although it is a flexibly applicable system technology. Apart from the installation of one or several injectors in the plastification cylinder and a suitable, precisely working, gas dosing station, a special screw (25–28D) with mixing and

61259_C004.indd 115

10/25/2008 12:36:15 PM

116

Polymeric Foams

Controller valve

Blowing fluid duct

Mixing- and shear-elements Shut-off nozzle FIGURE 4.11

Blowing fluid dosing station

Injection of blowing agent in the plastification unit (MuCell®).

shear elements is essential. This special screw requires the extension of the plastification unit. Another difficulty is the varying effective mixing length of the screw, as the screw moves relative to the injector during plastification. To compensate for this effect, several injectors can be piloted in a cascade, which requires an accordingly sophisticated control technique. Furthermore, the position of the gassing port determines the minimum dosing volume. Therefore, the application scope of this special plastification unit is limited compared with conventional aggregates. This technology is closely linked with the term of the MuCell® technology20 of the company Trexel (Woburn, US). The concept is shown in Figure 4.11. After introduction onto the market, this process variant drew large attention. Because of the advantages related to a reduction in the molding pressure and in the cycle time, as well as the diminution of internal stresses and the increase in dimensional accuracy of the molded part, the foam injection molding process achieved a lot of interest again. Foam injection molding was rediscovered by the rising interest in the market. 4.2.4 Injection of the Blowing Agent into a Special Gassing Unit Another new concept for the dosing of physical blowing agents into the melt was firstly presented on the K-2001 show under the trade name ErgoCell® by the company DEMAG Ergotech GmbH, Schwaig, Germany.21 For the loading of the melt an additional equipment component is inserted between the plastification cylinder and the injection nozzle of a modified injection molding machine. In the original design this special gassing unit consisted of a blowing agent inlet zone, a mixing zone, and an attached piston injecting unit, in which the homogenized polymer/blowing agent is stored under pressure until the injection phase. The supply of the blowing agent was achieved with a high-pressure piston pump.

61259_C004.indd 116

10/25/2008 12:36:15 PM

Processes and Process Analysis of Foam Injection Molding

117

Gas duct

Ergocell-unit Shut-off nozzle

FIGURE 4.12

Blowing fluid dosing station

ErgoCell® process engineering scheme.

An advanced concept uses an axial arrangement of the different components. The function of the injection piston is accomplished as in the conventional injection molding by the screw. Another function of the screw is to drive a dynamic mixer, which consists both of fixed and mobile elements (stator/rotor principle). Figure 4.12 shows the principle of this process. The layout of the gassing unit is shown in Figure 4.13. While the original concept requires a special injection molding machine owing to both an additional hydraulic system and the size of the system, the new variant only needs a conventional system design with an appropriately dimensioned gassing unit. Thus, in principle, this system is retrofittable. The advantage of the original design was the separation of the different process steps of polymer plastification, incorporation of the blowing agent, Single phase (melt/gas)

Gas injection nozzle Melt

Injection piston

Screw Spline

Mixer

ErgoCell-Module

Plastification cylinder [Demag Ergotech]

FIGURE 4.13

61259_C004.indd 117

ErgoCell® module scheme.

10/25/2008 12:36:16 PM

118

Polymeric Foams

and injection of the polymer/blowing agent system. In the new design the incorporation of the blowing agent is coupled to the dosing parameters of the plastification unit. In theory this is of no disadvantage, but it offers less freedom to the operator during the optimization of the process. 4.2.5 Processing of Pellets Pre-Charged with Blowing Agent Not only can the blowing agent be injected into the melt, the pellets can also be charged with blowing agent in advance. Whereas the charging takes place in an autoclave the pellets are processed on conventional injection molding machines. Using this concept, the concentration in the polymer can be varied by adjusting the pressure, temperature and charging pressure within the autoclave. After an adequate charging time, the system pressure is reduced to ambient pressure and the pellets are removed. The subsequent desorption process lowers the concentration of blowing fluid in the polymer. After an initial sharp decline of blowing fluid content, the concentration gradient falls with an increase in desorption time. The charged polymer pellets need a long, defined desorption time under ambient conditions before the foam injection molding process becomes reproducible. Subsequently they are processed on conventional injection molding machines. By an adequate course of the process, microcellular foams can be produced with this method. A major advantage of this design is that no modifications to the injection molding machine are necessary. Only a pressure vessel is required for charging the pellets with blowing fluid. Within this vessel the material to be molded has to be charged up to 2 days before processing. In this context the blowing fluid concentration in the polymer is not only determined by the parameters during the charging in the blowing fluid, but also by the desorption time after charging. As there is a continuous loss of blowing agent and no possibility of varying the concentration at short notice the fabrication of mass-produced articles is rather complicated and even unrealistic. 4.2.6 Injection of the Blowing Agent with Aid of a Special Nozzle A further process variant of the foam injection molding doses the blowing agent into the melt through a special injection nozzle, which is installed between the shut-off nozzle and the plastification unit. During the injection phase the melt flows through the injection nozzle and is loaded with the blowing agent there. This process, developed at the Institute of Plastics Processing (IKV) in Aachen22 and first presented at the IKV-Colloquium in 2000, is licensed to the company Sulzer Chemtech AG, Winterthur (Switzerland), which distributes it under the name Optifoam™. As the melt flows through the blowing fluid injection nozzle, it is transferred from a pipe flow over a torpedo into an annular slit flow. Here the

61259_C004.indd 118

10/25/2008 12:36:17 PM

Processes and Process Analysis of Foam Injection Molding

119

Gas duct

Blowing fluid injection nozzle Static mixer Shut-off nozzle Blowing fluid dosing station FIGURE 4.14

Fluid injection into a special nozzle.

blowing agent flows across permeable sinter metal sleeves and finally reaches the polymer melt. Thus, the agent is incorporated into the moving melt over the entire surface of the sinter metal sleeves. After the enrichment with blowing agent, the melt flows through downstream static mixing elements, before it is injected into the cavity through the shut-off nozzle.23 The principle of the blowing agent injection nozzle is shown in Figures 4.14 and 4.15. The possibility of using conventional injection molding machines without modifications of screw or plastification aggregate is of great advantage. This reduces substantially the investment costs and improves the flexibility concerning the application scope. In this way, conventional injection molding machines can be retrofitted for foam injection molding applications without substantial change. Figure 4.16 shows the system

Sinter metal sleeves Melt flow Phys. blowing fluid

FIGURE 4.15

61259_C004.indd 119

Optifoam™ process scheme.

10/25/2008 12:36:17 PM

120

Polymeric Foams

Shut-off nozzle FIGURE 4.16

Static mixer

Blowing fluid injection nozzle

Optifoam™ system mounted on an injection molding machine.

mounted on an injection molding machine as presented by the IKV on the K-2004 show in Düsseldorf, Germany. As the melt loading takes place before the head of the screw, the complete dosing volume of the injection unit can be used. However, depending on the design of the injection molding machine the amount of space in front of the plastification aggregate can be rather small.

4.3 Experimental Analysis of the Processing Parameters in Foam Injection Molding In the production of injection molded foamed parts, the material properties and the process parameters have a decisive influence on the quality of the final part. The foam injection molding process involves more parameters than conventional injection molding, which are not independent but correlate with each other. A thorough understanding of their interactions is thus required to control the resulting foam structure with their corresponding mechanical properties. The following section will present two studies conducted at the IKV, in which the influence of process parameter variations on the foam part characteristics was analyzed. In the first instance, a set of experiments performed with the IKV blowing fluid injection nozzle will be presented, with the aim of evaluating the potential of cycle time reduction and analyzing the effects of the process on the resulting foam morphology. Afterwards the results of a second set of experiments will be presented, describing the effects of the process parameter variations on the density distribution and foam morphology. For these investigations, foamed parts

61259_C004.indd 120

10/25/2008 12:36:17 PM

Processes and Process Analysis of Foam Injection Molding

121

molded with pre-loaded polycarbonate were considered. In this case a regular injection molding machine was used without modifications of the plastification unit. The prior process of charging the polymer with blowing agent was accomplished using an autoclave. 4.3.1 Experiments with the IKV Blowing Agent Injection Nozzle In the foaming process with physical blowing agents the following parameters were identified as the most relevant: • Injection velocity • Melt temperature • Blowing agent content related to the gas mass flow. Further parameters like the mold temperature, the back pressure, and the system pressure of the gas dosing station are of interest. A set of experiments was conducted with molding plates of 20% talcfilled polypropylene with a wall thickness of 4 mm. 4.3.1.1 Influence of the Process Parameters on the Foamed Part Characteristics First, the influence of variations in the injection velocity was analyzed. As can be observed in Figure 4.17, for the same temperature in the melt and the same rate of blowing agent dosage, two injection speeds were tested. A finer and more even foam structure was reached with the increase in injection velocity. The reason is that with an increase in the injection velocity a

FIGURE 4.17

61259_C004.indd 121

Tm: vinj: mCO : 2

240°C 40 mm/s 0.2 g/s

Tm: vinj: mCO : 2

240°C 80 mm/s 0.2 g/s

Δr:

31.5%

Δr:

35.3%

Influence of injection velocity on foam structure and density reduction.

10/25/2008 12:36:18 PM

122

Polymeric Foams

FIGURE 4.18

Tm: vinj: mCO : 2

210°C 80 mm/s 0.2 g/s

Tm: vinj: mCO : 2

240°C 80 mm/s 0.2 g/s

Δr:

26.5%

Δr:

35.3%

Influence of melt temperature on foam structure and density reduction.

steeper pressure gradient in the cavity—and specifically in the runner system—is achieved, increasing therefore the rate of nucleation. Subsequently the influence of variations of melt temperature was investigated. As can be seen in Figure 4.18, for the same injection velocity and rate of blowing agent dosage, a more homogeneous foam structure was achieved by increasing the melt temperature. It is remarkable that a variation of the melt temperature also influences the density reduction, and lighter foams are achieved with higher temperatures. A thickness reduction of the outer skin is another effect of the melt temperature increase, because the melt freezes later and allows foaming of the polymer closer to the cavity walls. However, as can be seen in Figure 4.19, the foam structure is not only dependent on the process parameters, but also on the location in the part.

r (g/cm3)

L/D = 5

L/D = 23

0.68 0.64 0.60 Close to the gate

FIGURE 4.19

61259_C004.indd 122

L/D = 14

Far from the gate

Comparison of foam structures along the flow path.

10/25/2008 12:36:19 PM

123

Processes and Process Analysis of Foam Injection Molding

4.3.1.2 Investigations on Cycle Time Reduction Previous investigations showed that the use of the process-specific characteristics of FIM offers the possibility of reducing cycle time, and thus achieve more economic processing. However, today only vague knowledge exists about the mechanisms giving origin to this phenomenon. Furthermore, it is unclear how far the part characteristics (for example the wall thickness), the foam structure, or the process parameters (e.g. blowing agent, pressure and/or concentration) have an effect on the potential of the cooling time reduction. Possibilities for the reduction of cycle time arise because of the lack of a packing phase. Otherwise, a lower melt viscosity grants shorter injection times. Beside the fact that lower melt temperatures can be applied and less mass is injected per shot, a smaller amount of heat has to be removed from the melt, reducing thus the cooling time. The part will also experience a more intensive cooling due to the evenly distributed cavity pressure. Another contribution to the removal of heat from the melt might come from a phase transition of the blowing agent, which is regarded of minor influence. In foam injection molding trials, the cycle time reduction potential was investigated using a design of experiments. A disk shaped part with a diameter of 160 mm and different thicknesses was selected as testing geometry (Figure 4.20). Results for three sample thicknesses of 3, 4, and 6 mm were analyzed. The used material was polypropylene with 20% of talc content; carbon dioxide was used as blowing agent. The temperature of the part was measured after demolding by means of IR thermography. The digital infrared camera allows the detection of the surface temperature distribution as well as the recording of the cooling process for a defined time period in a film sequence. As a reference, the cooling time of compact parts was measured.

Nomenclature m cp p

Tm (°C) 230 240 250

vinj (mm/s) pCO2 (bar) 100 150 200

60 75 90

2n + 1 (n = 3) ⇒ 8 + 1 sets of parameters Geometry: disc (Δ = 160 mm, d = 3, 4 and 6 mm)

∅

Gate system: bar gate

d Material: Polypropylene (talcum filled – 20 wt%) FIGURE 4.20

61259_C004.indd 123

Design of experiments to research cooling time reduction.

10/25/2008 12:36:21 PM

124

Polymeric Foams

95.0°C 80 60 40 25.0°C Compact Cooling time: tC = 28 s FIGURE 4.21

Foamed setting: ppp

Surface temperature distribution after demolding.

Figure 4.21 shows the surface temperature distribution of two 3-mmthick parts just after demolding. The left picture shows the recording of the compact part, which was manufactured with the settings of the central point (cp, melt temperature 240°C), and the picture on the right side shows a part foamed in the range of the high level of process parameters (ppp, melt temperature 250°C). The foamed part is demolded with a clearly lower surface temperature than the compact one. In this case the difference in the averaged surface temperatures amounts to 17°C. This result is particularly remarkable, because it can be seen as a substantial potential for cooling time reduction. The fact that a lower surface temperature can be achieved even at a higher melt temperature is evidence that, contrary to expectations, there is a relatively large processing window, and therefore a high potential for the optimization of the part characteristics exists. Figure 4.22 shows the mean demolding temperatures of parts produced with different settings of parameters and their corresponding density reductions. This diagram demonstrates that a lower demolding temperature can be achieved for the 3-mm-thick foamed parts, in comparison to the compact part, regardless of the variations in process parameter settings. As the demolding temperature holds a relatively constant value for dissimilar process conditions, the influence of melt temperature variations seems to be relatively small. Following the comparative investigations for the determination of the demolding temperature, the possibility of reducing of the cooling time was examined. Some of the test points for which the lowest demolding temperatures were determined in the test plan were repeated, applying a gradual reduction of the cooling time. The cooling time was shortened until a demolding without additional expansion could not be avoided. Figure 4.23 shows the results of these tests. It is interesting that for the wall thicknesses of 3 and 4 mm a cooling time reduction of over 20% is

61259_C004.indd 124

10/25/2008 12:36:22 PM

125

Processes and Process Analysis of Foam Injection Molding

35

d = 3 mm tC = 28 s

60

30

50

25

40

20

30

15

20

10

density reduction demolding temperature

10

5

Density reduction (%)

Demolding temperature (°C)

70

0

0

t ac

p

om

p

m

m

m

m

m

m

pm

m

pp

pm

m

p

pm

pm

p

p pp

cp

C

FIGURE 4.22 Demolding temperature and resulting density reduction depending on the process parameters (m: low level, p: high level, cp: central point—see Figure 4.21).

possible. Thus, an enormous potential for cost savings exists for these molding/material combinations. With 6-mm wall thickness the cooling time reduction diminishes to less than 10%. In consideration of the wall thickness, this result is still remarkable; however, it is not directly transferable to other polymers. 4.3.2 Experiments on Pre-Loaded Polycarbonate In a second series pre-dried polycarbonate pellets were loaded for 18 hours with carbon dioxide in an autoclave, set at 20 bar and ambient temperature. After a certain desorption of gas the charged material was dosed

60 50

Time (s)

40

Compact Foamed

30

30

20

20 10 10 0

0 3

FIGURE 4.23

61259_C004.indd 125

Cooling time reduction (%)

40

Cooling time reduction

4 Wall thickness (mm)

6

Comparison of cooling times and attainable cooling time reductions.

10/25/2008 12:36:22 PM

126

Polymeric Foams

TABLE 4.4 Design of Experiment (DOE) of Injection Molding Parameters Parameter Variable

Constant

Injection speed Mold temperature Melt temperature Density reduction Plate thickness

Dimension (cm3) (°C) (°C) % (mm)

Levels 20 80 290 20 3

60 100 310

100 120 330

in an injection molding machine, and processed within a window of 2.5% and 2.0% CO2 content. With this processing window and a fixed dosing volume in the injection molding machine, an overall density reduction of about 20% was attained. Plates with a thickness of 3 mm, a length of 200 mm, and a width of 100 mm were molded through one gate of a hot runner system located in the center of the plate. For the production of the plates, three process parameters were varied: mold temperature, melt temperature, and injection velocity. The plan followed is depicted in Table 4.4. For different combinations of these parameters the effects on the density distribution and cell morphology were analyzed. 4.3.2.1 Morphological Characterization Methods The density distribution was analyzed in both macro- and microscopes. In the macroscope, the density distribution along the major dimensions of the part was determined through three different methods: X-rays, grid punching, and IR thermography. The density distribution and foam morphology across the transverse section of the part were observed with the aid of a microscope. Each method will be explained in detail below. With the X-ray method an image of the cell distribution along each part was obtained, as can be seen in the left image of Figure 4.24. The “grid punching” method allowed a quantitative characterization of the density distribution: each part was divided into a grid, and discs were punched from each grid cell. As the size and weight of each disc were known, the local density could be determined. In the IR thermography method of density characterization a series of images of the part was made just after demolding. Thus the temperature distribution could be observed over a time window. In IR thermography, the local density of the part can be correlated to the local heat; where there is more concentration of material and therefore less foam, the temperatures are higher. On the contrary, in the regions where the temperature is low, there are more cells and less material to cool, and therefore the part experiences a faster heat transfer.

61259_C004.indd 126

10/25/2008 12:36:23 PM

127

Processes and Process Analysis of Foam Injection Molding

Down

Up

Material: PC Makrolon 2005

Injection gate

FIGURE 4.24 X-ray image of a foamed polycarbonate plate (left), magnification of local foam structure (right).

For the microscopic analysis small samples were cut at a location equidistant from the injection point and the end of the part. Figure 4.24 shows the exact location where the foam structure observations were made. A typical microscopic image of the foam structure achieved within this set of experiments is included. With the purpose of comparing the validity of the methods used to determine the macro distribution of the foam density, the results obtained with the grid-punching method and the IR thermography method were plotted in the same diagram, as shown in Figures 4.25 and 4.26. It can be seen that the correlation between both parameters is quite satisfactory.

Temperature distribution

Density distribution

T (°C) 120 117 115 112 109 106 104 101 98 95 93

r (g/cm3) 1.050 1.036 1.023 1.009 0.995 0.982 0.968 0.955 0.941 0.927 0.914

Conditions: Tmelt = 310°C / vinj = 60 cm3 / Tmold = 100°C FIGURE 4.25

61259_C004.indd 127

Surface temperature and density distribution.

10/25/2008 12:36:23 PM

128

Polymeric Foams

Normalized distribution (–)

1.0

0.9

m2

m1

0.8 Temperature Density 0 –100

–50

0 Sample length (mm)

50

100

FIGURE 4.26 Correlation between the surface temperature of the samples and the corresponding density distribution: plot of normalized values for temperature and density distributions.

4.3.2.2 Influence of the Process Parameters on the Foamed Part Characteristics As an initial point for the analysis, the variation in the homogeneity of the density distribution was observed. In Figure 4.27, the gradient of temperature distribution as mean values of m1 and m2 (cp. Figure 4.26), which as mentioned before correlates directly with the density distribution, is plotted against the variables of melt temperature, injection velocity, and mold temperature. As the gradient of temperature distribution increases, a less homogeneous distribution of the foam density is observed. The dependence of the density distribution on each of the injection parameters is explained below. Figure 4.27 reveals that a more homogeneous foam structure is achieved along the flow path if the temperature of the melt is increased. This can be explained by the effect of temperature on the viscosity of the melt. When the temperature is increased, the viscosity and therefore the resistance to cell formation decreases, and thus a homogeneous cell structure will be likely to appear, even at locations distant from the injection gate. The increase in melt temperature also has a positive effect on the cell structure. Once the maximum value is reached, a finer cell morphology is achieved, as can be seen in Figure 4.28. The number of nucleation sites will be increased at higher temperatures, and therefore the primary mechanism of cell forming, nucleation, will predominate over the secondary mechanism, the diffusion of blowing agent into primarily formed cells. Measurements made at the IKV also showed that at high melt temperatures the surface tension of the melt is reduced, thus promoting the nucleation of independent, newly formed cells with a finer structure cell morphology. The effect of different injection velocities on the gradient of temperature distribution is depicted in Figure 4.27. As expected, the temperature

61259_C004.indd 128

10/25/2008 12:36:24 PM

129

Temperature gradient (°C/mm)

Processes and Process Analysis of Foam Injection Molding

0.4

0.3

0.2

0.1

0.0 290 300 310 320 330 Melt temp. (°C)

20 40 60 80 100 Injection velocity (cm3/s)

80

90 100 110 120 Mold temp. (°C)

FIGURE 4.27 Influence of injection velocity, mold temperature, and melt temperature on the homogeneity of the density distribution.

gradient of the demolded part will be less steep with a lower injection velocity. This is mainly an effect of the appearance of a thicker outer skin. A less abrupt pressure drop, coming from lower injection velocities, will result in the appearance of less nucleation sites, and therefore the foamed portion of the part will be smaller. A thicker compact border layer will be able to carry on a more effective heat transfer through the mold surface, and in turn will generate a cooler and more homogeneous surface temperature distribution. Another interesting effect of the velocity variation can be seen in the orientation of the cell structure. Figure 4.29 shows the difference in foam structure resulting from variations in the injection velocity. As can be seen, at low velocities cells of relatively large dimensions and strong orientation are formed. This is a result of the simultaneous effect of cell

Tm = 290°C

Tm = 310°C

Tm = 330°C

Flow direction FIGURE 4.28 Dependence of foam structure on melt temperature. Samples taken in the middle of the flow path.

61259_C004.indd 129

10/25/2008 12:36:24 PM

130

Polymeric Foams

vi = 20 cm3/s

vi = 60 cm3/s

vi = 100 cm3/s

Flow direction FIGURE 4.29 Dependence of foam structure on injection velocity. Samples taken in the middle of the flow path.

growth and melt displacement. On the other hand, if the injection velocity is increased, a much finer and homogeneous foam structure is revealed: a steeper pressure gradient grants more nucleation sites and a later bubble growth. Here the mechanism of primary cell nucleation predominates over that of gas diffusion in the process of foam formation. The effect of the mold temperature on the foam structure is illustrated in Figure 4.30. At a low mold temperature, a foam structure with clear orientation and large cells can be attained. With an increasing temperature difference between mold and melt the heat will be transferred faster, giving place to an earlier increase in viscosity and therefore to less nucleation sites. Another consequence of this effect is the increase of the melt surface tension. The growth mechanism of blowing agent diffusion into the primarily formed cells is responsible for most of the resulting foam

Tm = 80°C

Tm = 100°C

Tm = 120°C

Flow direction FIGURE 4.30 Dependence of foam structure on mold temperature. Samples taken in the middle of the flow path.

61259_C004.indd 130

10/25/2008 12:36:26 PM

131

Mold temperature

Melt temperature

Injection velocity

Melt temperature

Skin thickness

Cell size

Cell number

Melt temperature

Mold temperature

Skin thickness

Cell size

Cell number

Mold temperature

Skin thickness

Cell size

Cell number

Processes and Process Analysis of Foam Injection Molding

Injection velocity

Injection velocity

FIGURE 4.31 Summary of the influence of the process parameter on the resulting foam structure.

structure, and thus the cells exhibit larger mean diameters. On the other side, with a high mold temperature the heat is conducted through a less steep gradient, and the material is kept viscous enough to allow the apparition and growth of a more homogeneous foam structure. A comprehensive summary of the effects of changes in the injection molding parameters on the cell size, cell number and compact skin thickness is given in Figure 4.31.

4.4 Optimization of the Surface Quality of Foamed Injection Molded Parts Besides the many advantages of foam injection molding (FIM), the achievable surface qualities are rather poor in many cases. Occurring silver streaks, melt eruptions, and cold-displaced polymer melt areas cause more uneven and non-uniform part surfaces in comparison to conventional injection molding. For this reason, foamed parts are often excluded as visually exposed parts. A comprehensive understanding of the effects arising during the filling phase establishes new possibilities for increasing

61259_C004.indd 131

10/25/2008 12:36:28 PM

132

Polymeric Foams

the surface qualities in foam injection molding. New research shows that different process variants of FIM such as “breathing” molds, gas counterpressure, and structured and coated cavity surfaces can increase the surface quality effectively. 4.4.1 Occurrence of Silver Streaks If no particular measures are taken, for example, in mold technology, foamed parts show rather poor surfaces. In many cases low surface qualities are caused by bright silver streaks, which are oriented in the direction of the flow path. Semerdjev and Popov explained the development of these streaks with gas bubbles leaving the flow front during filling, which are sheared when contacting the cavity wall.24 A schematic description of their development is shown in Figure 4.32. The example shows the surface of a physically foamed part of polypropylene (PP-T20) filled with 20 wt% talc with a wall thickness of 6 mm, which was gated in the center. The surface was photographed with a flatbed scanner and increased in contrast to point out the occurring surface defects. This surface shows clearly a multitude of silver streaks in the form of bright shining lines oriented in direction of the flow path. The development of these silver streaks can be analyzed in detail by looking at the surfaces of foamed polypropylene parts. Figure 4.33 shows two impinging light microscopy pictures of two cut-outs of a PP-surface. On the left picture, a 420 μm broad white band can be clearly seen, which can be detected as a fine white line or as a silver streak on the part surface. In the picture on the right, it can be noticed that the white band consists of many small white spots with an average diameter of 40–50 μm. This picture leads to the assumption that a macroscopic gas bubble, grown during

FIGURE 4.32

61259_C004.indd 132

Schematics of the development of silver streaks.

10/25/2008 12:36:28 PM

Processes and Process Analysis of Foam Injection Molding

FIGURE 4.33

133

Cut-out of a foamed part surface with silver streaks.

the filling phase, is split into many small microscopic gas bubbles by shear strain on the cavity wall. Thereby, these small gas bubbles seem to modify the part surfaces resulting in different reflection properties, which evoke the bright shining lines—called silver streaks—on a larger scale. A possible modification of the surface will be analyzed in further examinations. A white shining spot on the surface was investigated in detail using scanning electron microscopy, which is marked by a white dashed line (see Figure 4.34, top left). On closer inspection one can see a rather rough and uneven surface outside of the dashed range. This roughness is caused by shear strain of the gas loaded melt on the cavity wall and by the low cavity pressures in foam injection molding.25 Inside of the dashed area the surface seems rather even and sunk in comparison to the surrounding surface. A cut through the part in Figure 4.34 (top right) shows

Cut: A – B

A

B 500

Silver streak (even) Part surface (uneven)

50 mu

Deepening/dent in part surface

(Far from gate)

Distance s Pressure p FIGURE 4.34

61259_C004.indd 133

(Close to gate)

Stretched microbubble as deepening on the part surface.

10/25/2008 12:36:28 PM

134

Polymeric Foams

the geometry of the surface. This deepening gives rise to the supposition that the microscopic gas bubbles release a small amount of gas, which is enclosed between melt and cavity wall displacing the hot melt. The dimension of the deepening, which can be regarded as a dent, should be dependent on the amount of gas, on the gas pressure, and the pressure in the melt. Following this idea the displaced melt volume results from the balance of forces between gas pressure and resistance of the melt to be displaced. Despite the low cavity pressures in foam injection molding in comparison to compact injection molding,26 the pressures within the dents are sufficient to cause a deepening in the surface. Regarding the surfaces of foamed parts along the flow path, it is well known that the surface qualities decrease with increasing distance to the gate. Following the above presented idea this is caused by the decreasing melt pressure along the flow path. Corresponding to the balance of forces between gas pressure and resistance of the melt to be displaced, the dents will be smaller closer to the gate than at the end of the flow path. A second reason for inferior surface qualities over the flow path is that the gas bubbles have more time to grow within the melt depending on the diffusion rate to deliver a bigger gas volume to be enclosed within the dents. Both effects lead to inferior surface qualities along the flow path. Within the performed investigations a correlation between silver streaks and the surface deformations could be detected. However, the reason for the bright reflections has not yet been clarified. Possible explanations are on the one hand a different evenness on the surface and on the other hand a different crystallization behavior of the surfaces inside and outside of the dents. Considering a different evenness, the surface within the area of the deepenings is less rough than on the surrounding surfaces, which may be caused by less shearing and the surface tension of the melt. The smoother surface within the dents in turn has a significant effect on the reflection properties, as it leads to a more directed light reflection in comparison to a diffuse reflection of the surrounding areas (Figure 4.35). The second explanation approach is based on a different crystallization behavior inside and outside the dents. Due to the gas volume encapsulated in the deepenings, the heat transfer between mold and polymer melt is reduced resulting in higher surface temperatures in the area of the dents. Thus, the hot polymer has more time for the crystallization process, which in turn has an effect on the degree of crystallization.27 For this reason an effect on the optical density is expected increasing the degree of reflection. Both approaches will be analyzed in future examinations. 4.4.2 Possibilities of Increasing the Surface Quality Through Mold and Process Technology Owing to the examinations that the development of surface defects is strongly dependent on pressures and temperatures within the melt, the

61259_C004.indd 134

10/25/2008 12:36:29 PM

135

Processes and Process Analysis of Foam Injection Molding

Directed light reflection

Diffuse reflection

Distribution of luminosity

Distribution of luminosity

αβ Even surface

αβ Uneven surface

FIGURE 4.35 Light reflection on plain and rough surfaces.

choice of the process parameters is of high importance for the surface qualities of foamed parts. However, the impact of a sole optimization of process parameters is limited. In the past, poor surface qualities also led to many developments in the field of mold and process technology in FIM. All process variants are based on conventional foam injection molding in which the mold is only partially filled during the machine injection phase depending on the weight reduction to be achieved. This standard process leads to low pressures in the cavity resulting in poor surface quality. For this reason the gas counter-pressure process, “breathing” molds, structured cavities, heat insulating cavity coatings, and the variotherm process were investigated. The first four will be briefly described in the following paragraphs. Using the process variant “breathing” mold, which is shown in Figure 4.36, the mold cavity is completely filled with the polymer/blowing agent mixture during the machine injection phase.28 To achieve compact border

FIGURE 4.36

61259_C004.indd 135

Scheme of the process variant “breathing” mold.

10/25/2008 12:36:29 PM

136

FIGURE 4.37

Polymeric Foams

Scheme of the gas counter-pressure process.

layers, a short packing phase is applied. Subsequently the cavity is enlarged initiating the foaming phase. By the filling of the cavity under high pressures and the following packing phase, occurring surface defects can be “repaired” after the filling phase. The holding pressure in the packing phase causes the reduction or avoidance of deepenings against the encapsulated gas pressure, so that the surface qualities can be increased significantly. By use of the process variant gas counter-pressure, as shown in Figure 4.37, a foaming of the melt/blowing agent-mixture can be avoided during the injection phase by application of a gas pressure inside the cavity of about 40 bars.28 By means of counter-pressure, the blowing fluid is kept above the saturation pressure inside the melt, whereby the development and the growth of gas bubbles, which are sheared during the filling phase, are avoided. This process variant is often combined with a “breathing” mold. A suited and effective mold technology for increasing the surface quality is the modification of the cavity surface with textures. The functional principle is not based on the increase of the process pressures, but rather on a better venting of the escaping gas and concealment of possible surface defects. By the use of texturized cavities, the development of silver streaks can be avoided by reducing shear strain of the growing gas bubbles on the cavity wall, which can be reached by a mechanical anchoring of the polymer in the structured mold surface. Furthermore the structures superpose the rough and uneven surfaces of foamed parts, so that the appearance of silver streaks is reduced by a diffuse backscatter (Figure 4.38, top). A further possibility is the use of coated surfaces. A heat insulating coating on the cavity surface enables an increase of the surface quality of foamed parts with only a minor increase in cycle time.29 Those coatings cause a reduction of the cooling rate of the outer part layers, causing a better flow with less orientation, less internal stress, and optical appealing

61259_C004.indd 136

10/25/2008 12:36:30 PM

Processes and Process Analysis of Foam Injection Molding

FIGURE 4.38

137

Use of structured and coated cavity surfaces.

surfaces (Figure 4.38, bottom). The improvement of the part surface can be attributed to the fact that flow marks resulting from sheared gas bubbles can longer be deformed and corrected through a higher contact temperature on the part surface. For this reason the silver streaks, which can be seen as a multitude of dents, are minimized or completely avoided depending on the coating by the prevailing melt pressure. The encapsulated gas within the deepenings is either compacted or diffuses back into the polymer. 4.4.3 Investigation of “Breathing” Molds and Gas Counter-Pressure The process variants “breathing” mold and gas counter-pressure have a significant influence on the process pressures, which in turn have a strong impact on the surface qualities. Characteristic cavity pressure curves are presented in Figure 4.39. This figure shows that the pressures achieved with conventional foam injection molding are much lower than with compact injection molding, which is an important reason for the poor surface qualities. The process variants aim to increase the surface qualities by raising the process pressures. Using gas counter-pressure, the pressures are increased primarily during the injection phase avoiding the development of gas bubbles and melt explosions during the filling stage. The process variant “breathing” mold uses primarily the increase in pressure after the filling stage, whereby the characteristic surface defects are to be “repaired” afterwards. The effect of both process variants on surface quality were studied. Therefore surfaces of black colored parts of polypropylene filled with 20 wt% talc, being produced with the respective processes, have been photographed and graphically analyzed. Figure 4.40 shows four pictures of these part surfaces in direct comparison. Owing to the fact that the

61259_C004.indd 137

10/25/2008 12:36:30 PM

138

Polymeric Foams

pCP

pcompact

pFIM+GCP pTSG+breathing pFIM Start injection

Time

FIGURE 4.39 Characteristic cavity pressure curves of different process variants in foam injection molding.

human eye detects dark areas on white areas more easily than vice versa, all pictures are inverted with regard to color scale. Thus dark lines are bright shining silver streaks in reality. Regarding the surface pictures, multiple dark spots exist on the parts produced without gas counterpressure. These spots can be explained as melt particles, which were blasted from the flow front and were surrounded afterwards by the melt again. As one can see, these surface defects can be avoided by gas counterpressure. Beyond that the surfaces are more even and smooth in comparison to conventional injection molding. Regarding the surfaces of the parts manufactured by the process “breathing” mold, one can see an articulate brighter coloration. This means that in reality the surfaces appear darker and exhibit fewer defects in the form of silver streaks. Both process variants

FIGURE 4.40 Comparison of the surface qualities of foamed parts produced with different process variants.

61259_C004.indd 138

10/25/2008 12:36:30 PM

Processes and Process Analysis of Foam Injection Molding

139

show the capability of improving the surface qualities especially when used in combination.

4.5 Summary Foam injection molding (FIM) is gaining interest due to the advantages associated with process and part performance. In comparison with conventional injection molding, new possibilities arise from the production point of view, with a reduction in cycle time and machine tonnage required, and regarding final part performance from the reduction in warpage, material consumption, and dimensional accuracy. The specific mechanical performance, particularly under bending load, can also be improved. However, considerations associated with the complexity of the process technology, the low surface quality of the final part, and an overall reduction of mechanical properties, require a deep understanding of the process. The foaming is achieved by dosing of a blowing agent into a polymer melt. This dosing will be influenced by the temperature, the process, and the material. The mechanisms of cell nucleation and growth, which have a large influence in the final foam structure, strongly depend on the pressure drop velocity. An increase in the pressure drop gradient will lead to the formation of a finer and more even foam structure. The interest in physical blowing agents is growing parallel with the development of FIM technology, owing to the higher levels of foaming attainable as well as to the absence of residual products (as with chemical blowing agents). Their dosing into the polymer melt has required the development of special technologies, some of which are reviewed within this contribution. Each of them uses a different principle to incorporate the blowing agent into the melt. More process parameters are involved in the production of foamed injection molded parts compared with the conventional injection molding process. These parameters are correlated and affect each other. Through design of different experiments the influence of parameter variations on the molded foamed parts was analyzed. It can be concluded that with increasing injection velocity, mold temperature, and melt temperature a finer and more homogeneous foam structure can be attained, and that the thickness of the outer skin was increased by lower mold temperatures. It was also demonstrated that there exists a potential for cycle time reduction in foam injection molding, which does not show a strong dependence on the melt temperature. The poor surface quality normally associated with foam injection molding comes from the apparition of gas bubbles and their destruction on the surface. The induction of pressure in the mold cavity, either through

61259_C004.indd 139

10/25/2008 12:36:31 PM

140

Polymeric Foams

technologies such as “breathing” the mold or gas counter-pressure can counteract these negative effects. Other alternatives are the use of textures or coverings on the cavity surface, which allow the controlled migration of gas from the part surface. Despite an already broad scope of application of foamed parts, the implementation of foam injection molding is to be examined specifically for each application. Even if economic advantages seem available through a reduced cooling time, the application of this process is not always possible, and the direct substitution of a compact part by a foamed one may not always satisfy the stated requirements. In many cases, the exclusion criterion is the low surface quality or an inadmissible loss in mechanical properties. However, these problems can be avoided by an integral approach considering the requirement of the foam injection molding process and the properties of thermoplastic foams in the early design phase of the product.

Acknowledgments The investigations set out in this chapter received financial support from the Federal Ministry of Economics (BMWi) within the “Arbeitsgemeinschaft Industrieller Forschungsvereinigungen “Otto von Guericke” e. V. (AiF) (Project No. 14377 N), to whom we extend our thanks. The material was provided by Ticona GmbH, Kelsterbach, Germany, Borealis GmbH, Linz, Austria, Bayer Material Science AG, Leverkusen, Germany and Clariant Masterbatch GmbH & Co. OHG; Ahrensburg, Germany. The machine technology was provided by Sulzer Chemtech AG, Winterthur, Switzerland, and Demag Ergotech GmbH, Schwaig, Germany. We thank all of them for their support.

References 1. Lübke, G. “Jedem das Seine – Treibmittelsysteme und Nukleierungsmittel für thermoplastische Schaumstoffe.” In IKV-Seminar zur Kunststoffverarbeitung, Aachen, 26–27 June 2001. 2. Spiekermann, R. “Entwicklungstendenzen bei chemischen Treibmitteln für Thermoplastschäume.” In IKV-Seminar zur Kunststoffverarbeitung, Aachen, 23–24 October 1997. 3. Leppkes, R. “Polyurethane.” In Werkstoff mit vielen Gesichtern. Verlag Moderne Industrie AG, Landsberg, 1993.

61259_C004.indd 140

10/25/2008 12:36:31 PM

Processes and Process Analysis of Foam Injection Molding

141

4. Shutov, F. A. Integral/Structural Polymer Foams. Springer Verlag, Berlin Heidelberg, 1986. 5. Van Krevelen, D. Properties of Polymers. Elsevier, Amsterdam, 1990. 6. Li, C. C. “Critical temperature estimation for simple mixtures.” Canadian Journal Chemical Engineering 19 (1971): S.709. 7. Kreglewski, A. and Kay, W. B. “The critical constants of conformal mixtures.” Journal of Physical Chemistry 73 (1969) 10: S.3359. 8. Pfannschmidt, L. O. Herstellung resorbierbarer Implantate mit mikrozellulärer Schaumstruktur. Dissertation, Rheinisch-Westfälische Technische Hochschule Aachen, 2001. 9. Menges, G. Werkstoffkunde Kunststoffe. Carl Hanser Verlag, München, New York, 1990. 10. Flory, P. J. Principles of Polymer Chemistry. Cornell University, Ithaca, 1969. 11. Fischer, M. and Schmidt, R. “Diffusion und Permeation, Eigenschaften von Polymeren.” In Chemie und Physik Band I. Georg Thieme Verlag, Stuttgart, New York, 1985. 12. Crank, J. The Mathematics of Diffusion. Clarendon Press, Oxford, 1975. 13. Wu, S. Polymer Interface and Adhesion. Marcel Dekker, New York, 1982. 14. Reid, R. C., Prausnitz, J. M., and Poling, B. E. The Properties of Gases & Liquids, 4th edition. McGraw Hill, New York, 1988. 15. Sander, B. Mikrozelluläre Schäume – Einfluss des physikalischen Treibmittels und des Polymers auf den Schäumprozess. Institut für Kunststoffverarbeitung, RWTH Aachen, unveröffentlichte Diplomarbeit. O. Pfannschmidt, Betreuer, 1999. 16. Colton, J. S. “Making microcellular foams from crystalline foams of crystalline polymers.” Plastics Engineering (1988) 8: S.53–S.55. 17. Nicolay, A. Untersuchung zur Blasenbildung in Kunststoffen unter besonderer Berücksichtigung der Rissbildung. Dissertation, Rheinisch-Westfälische Technische Hochschule Aachen, 1976. 18. Colton, J. S. and Suh, N. P. “The nucleation of microcellular thermoplastic foam with additives. Part I: Theoretical Consideration.” Polymer Engineering Science 27 (1987) 7: S.485–S.492. 19. Han, J. H. and Han, C. D. “Bubble nucleation in polymeric liquids. Part II: Theoretical considerations.” Journal of Polymer Science Part B 28 (1990): S.743–S.761. 20. Pierick, D. E., Anderson, J. R., Cha, S. W., Stevenson, J. F., and Laing, D. E. Injection Molding of Microcellular Material. Europäische Patentanmeldung EP1264672A1, 2002. 21. Jaeger, A. “Schäumen beim Spritzgießen neu entdeck.” In: Tagungshandbuch “Präzisionsspritzguss heute.” KI Lüdenscheid, Lüdenscheid, February 2002. 22. Schröder, T. Entwicklung einer Eingasungsdüse zur Herstellung von Thermoplastschäumen im Spritzgießverfahren unter Verwendung eines physikalischen Treibmittels. Student project work (supervisor: O. Pfannschmidt), Institut für Kunststoffverarbeitung, RWTH Aachen, 2001. 23. Habibi-Naini, S. Schaumspritzgießen – Untersuchungen zur Beladung der Schmelze mit physikalischen Treibmitteln während der Einspritzphase. Advisory group meeting, IKV, Aachen, 2000. 24. Semerdjiev, S. and Popov, N. “Probleme des Gasgegendruck-Spritzgießens von thermo-plastischen Strukturschaumteilen.” Kunststoffberater 4 (1978): 198–201.

61259_C004.indd 141

10/25/2008 12:36:31 PM

142

Polymeric Foams

25. Okamoto, K.T. Microcellular Processing. Carl Hanser Verlag, München, Wien, 2002. 26. Habibi-Naini, S. Neue Verfahren für das Thermoplastschaumspritzgießen. PhD thesis, RWTH Aachen University, 2004. 27. Menges, G. Werkstoffkunde Kunststoffe. Carl Hanser Verlag, München, Wien, 2002. 28. Semerdjiev, S. Thermoplastische Strukturschaumstoffe. VEB Deutscher Verlag für Grundstoffindustrie, Leipzig, 1980. 29. Horn, B., Mohren, P., and Wübken, G., Spritzgusswerkzeuge mit wärmedämmenden Formnestbeschichtungen. Notes from the Institute of Plastics Processing (IKV) at the RWTH Aachen University, 1976.

61259_C004.indd 142

10/25/2008 12:36:31 PM

5 Foaming Analysis of Poly(e-Caprolactone) and Poly(Lactic Acid) and Their Nanocomposites Ernesto Di Maio and Salvatore Iannace

CONTENTS 5.1 Introduction .................................................................................. 5.2 Foaming of PCL and PLA with CO2 and N2: Key Issues ....... 5.2.1 Foaming of PCL .................................................................. 5.2.2 Foaming of PLA ................................................................. 5.3 Molecular Modification of PCL and PLA ................................. 5.3.1 Branching/Cross-Linking of PCL .................................... 5.3.2 Chain Extended PLA ......................................................... 5.4 Nanocomposites ........................................................................... 5.4.1 Nanocomposites from PCL and PLA: Rheology, Sorption, Mass Transport, and Crystallization .............. 5.4.2 Foaming of PCL and PLA-Based Nanocomposites .................................................................. 5.4.3 Bubble versus Crystal Nucleating Effect of Nanoparticles .................................................................. 5.5 Conclusions ................................................................................... References ............................................................................................

61259_C005.indd 143

144 145 145 149 151 151 154 156 157 162 163 168 169

10/25/2008 12:37:29 PM

144

Polymeric Foams

5.1 Introduction The large and increasing volume of foamed materials has stimulated interest in developing useful innovations related to environmentally friendly processes and materials that can be utilized in these high volume applications. Biodegradation can be considered as a possible approach to solve the disposal problem, especially when recycling is difficult or costly or when biodegradation is a functional requirement of the product.1 Biodegradable polymers are already utilized in many biomedical applications such as biodegradable sutures, wound dressing, bio-resorbable implants, and drug delivery systems, applications where the high cost of the materials is justified.2 However, their use in commodity applications, such as packaging or agriculture, is still limited either for economic reasons or for difficulties related to their processing, often due to their poor thermal stability. Biodegradable materials for industrial foaming applications must display adequate properties, and their manufacturing process must be relatively simple and inexpensive. Among the most interesting biodegradable polymers that can be potentially employed for foaming are the polyesters such as poly(e-caprolactone) (PCL) and poly(lactic acid) (PLA) and their copolymers, polyhydroxyalkanoates (PHA), polyesteramide, polyurethanes (PU), and biopolymers such as polysaccharides and proteins. Since the availability of a large volume of commercial raw material supplies is uncertain, the combination of these polymers in blends and/or composites is often the way to prepare polymeric systems with a lower cost and with properties that can be tailored for the preparation of foams. This chapter will focus on the foaming behavior of PCL and PLA named poly(e-caprolactone) and poly(lactic acid). They have received a great attention by researchers working in the biomedical field and, more recently, an increasing interest for larger scale industrial applications. More information on other biodegradable foamed materials can be found in recent reviews.3,4 The main requirements for foamability are, for any thermoplastic polymer in general, the rheological characteristics of the melt,5,6 the blowing agent solubility and diffusivity, and the presence of adequate setting mechanisms. In particular, strain-induced hardening behavior is a fundamental characteristic for the foaming process, since it allows withstanding of the stretching forces at the latter stage of the bubble growth. There are basically two possibilities to improve the elongational properties of the melt: (1) optimize the molecular weight and molecular weight distribution of the polymer and/or (2) the branching of the macromolecules.7 In principle, chemical modification (i.e. chain extension and/or branching) can be used to modify the rheological properties of biodegradable polyesters such as PCL and PLA. In particular, two different techniques are described in this chapter: (1) branching and cross-linking with peroxides

61259_C005.indd 144

10/25/2008 12:37:29 PM

Foaming Analysis of PCL and PLA

145

for PCL; and (2) chemical modifications with chain extenders for PLA. The chemically modified materials were batch foamed and the effects of the molecular modifications on the physico-chemical properties of the polymers as well as on the foaming process are discussed. Another strategy for improving the foamability of thermoplastic biodegradable polymers is the use of nanoparticles.8,9 Nanometric filler, in effect, determines extensive modification of properties such as gas absorption and diffusivity, local thermodynamic properties responsible for nucleation and growing phenomena, and rheological characteristics which affect the final morphology of the foams. The effect of type and content of nanofillers on the foaming process of PCL and PLA is reviewed here.

5.2 Foaming of PCL and PLA with CO2 and N2 : Key Issues In recent years, few research groups have focused their attention on foaming of PCL and PLA, the two biodegradable polyesters of interest in this chapter. In this paragraph, a comprehensive picture of the key aspects of the foaming process of these two polymers, with CO2 and/or N2 as blowing agent, will be given. As a general, primary comment, both PLA and PCL have proved to be difficult to foam, mainly for their poor rheological properties and small processing windows. As a significant difference between the two polymers it is important to underline that the main limitations in the use of PCL are related to low processing temperatures while, in the case of PLA, to low crystallization kinetics, both of which reduce the possible setting mechanisms of the newly formed cellular structures.

5.2.1 Foaming of PCL PCL is synthetic aliphatic biodegradable polyester with chemical formula -[-O-(CH2)5-CO-]n-, which has been foamed by different research groups in a narrow range of molecular weights (Mn from approximately 69,000 to 80,000 g/mol, from different producers). PCL has a glass transition temperature of 64.2°C and a melting point of 69.0°C.10,11 When crystallized from the melt at 10°C/min, the onset of crystallization is at approximately 30°C.12 PCL crystallization is quite rapid: when cooled isothermally from the melt at 40°C, the crystallization half-time, t1/2, is 6.5 min, with an Avrami exponent close to 2 and a final degree of crystallinity of 40%,13 while, when cooled isothermally from the melt at 1°C, the crystallization half-time, t1/2, is 140 ms.14 The described crystallization kinetics are adequate for the stabilization of the foam but, in industrial applications (e.g. extrusion foaming), it is often difficult to cool the extrudate to such low temperatures, resulting in

61259_C005.indd 145

10/25/2008 12:37:29 PM

146

Polymeric Foams

a reduced efficiency of the expansion process. In batch foaming, conversely, the better control of the processing variables allows a finer control of the morphology and the achievement of foams of lower densities. As a matter of fact, most of the experimental work on PCL foaming has been performed in batch. Regarding the sorption properties of the blowing agents of concern here, solubility and diffusivity of carbon dioxide (CO2) and nitrogen (N2) in PCL have been measured at typical processing conditions. Equilibrium sorption concentrations of CO2 and N2 in PCL at different pressures and temperatures are reported in Figure 5.1 and denote a fairly higher solubility of CO2 with respect to N2. Mutual diffusivities have also been evaluated for PCL-CO2 and PCL-N2 systems, with an expected higher mutual diffusivity for the system PCL-N2 with respect to PCL-CO2. These differences in the sorption behavior and the associated differences in the plasticization effect of the two gases on the polymer, as reported in Reference 15, are responsible for the very different foaming behavior of the two different polymer-gas systems. Figure 5.2a reports a micrograph of a PCL sample foamed at 35°C with a rather slow pressure drop rate, after being saturated with CO2 at 55 bar and 70°C. In these experimental conditions, the resulting foam showed a rather coarse morphology with mean cell size of 0.5mm and a density of 0.05g/cm3.15 Similar results were obtained by Jenkins et al.,16 Xu et al.,17 Reignier et al.,18 and by Cotugno et al.19 Figure 5.3 summarizes the morphologies achieved by the different research groups in foaming PCL with CO2. The indicated temperature would let the reader think that the processing window for PCL foaming with CO2 is indeed wide (more than 40°C).

Gas mass fraction

0.25 0.2 0.15 CO2, 70°C CO2, 80°C

0.1

CO2, 90°C 0.05 0

N2, 75°C

0

5

20 10 15 Pressure (MPa)

25

30

FIGURE 5.1 Equilibrium concentrations of CO2 and N2 in PCL at different temperatures.

61259_C005.indd 146

10/25/2008 12:37:29 PM

147

Foaming Analysis of PCL and PLA

FIGURE 5.2 SEM micrograph of PCL foam: (a) with CO2 at 35°C, 10 bar/s; (b) with CO2 at 35°C, 30 bar/s; (c) with N2 at 43.3°C; (d) with 80/20% volume mixture of N2 and CO2 at 43.3°C.

800

Mean cell diameter (μm)

Xu et al. 40°C (17) 700

Jenkins et al. 65°C (16)

600

Di Maio et al. 35°C (15) Cotugno et al. 30°C (19)

500

Cotugno et al. 27°C (19) Cotugno et al. 24°C (19)

400 300 200 100 0

5

10

15

20

25

30

35

Saturation pressure (MPa) FIGURE 5.3 Variation of the mean pore dimension of PCL foamed with CO2. See cited references for experimental details.

61259_C005.indd 147

10/25/2008 12:37:29 PM

148

Polymeric Foams

Actually, the experimental apparatuses used for the preparation of the foams by the different research groups are rather different, mostly in the control of temperature, the cooling protocols, and in the gas evacuation system. This results in very different temperature histories in the batch experiments that do not allow a direct comparison among all of the results published so far. For this reason we would say that, for each single apparatus, the useful processing window is restricted to less than 10°C. As can be observed in Figure 5.3, with an increase of the saturation pressure foams with finer morphologies compared to that reported in Figure 5.2a can be obtained. As extensively reported in the literature, finer morphologies can also be obtained by increasing the pressure drop rate. As an example, Figure 5.2b reports a micrograph of a PCL sample foamed at 35°C with a faster pressure drop rate, after being saturated with CO2 at 55 bar and 70°C. The specific design of the experimental apparatus represents, commonly, the first limitation to the increase of the pressure drop rate to further reduce the mean cell diameter of the foams. This is also the case for industrial equipment. In order to modify the foam morphology, therefore, it is possible to change the blowing agent, as reported in Reference 15. In effect, foams with cellular structures characterized by finer morphologies have been obtained with N2 (see Figure 5.2c) by using the same experimental apparatus used for the foams reported in Figures 5.2a and b. In this case, however, the lower solubility of N2 with respect to CO2 led to foams with a higher density. A better compromise between cell morphology and density has been obtained by using mixtures of CO2 and N2 as blowing agents. In this case, small amounts of CO2 supply gas to inflate the high number of bubbles generated by N2, to achieve foams with a very fine morphology and low density at the same time (see Figure 5.2d). A detailed analysis of the decoupled effect of the main processing variables, the blowing agent concentration, the pressure drop rate, and the foaming temperature has been reported by Marrazzo et al.20 on PCL foaming with N2 in a fairly restricted processing range (Psat  140–200 bar, Tfoam  44.5–48.5°C and P  260–380 bar/s) by using a batch foaming apparatus designed to independently control the three process variables and define a foaming protocol. In contrast to the batch process, the continuous extrusion process is important to achieve industrial productivity. Until now, little interest has been given to biodegradable polyesters, mostly because they are not considered to be foamable in general, due to poor properties of the melt and reduced setting mechanisms. In fact results of the batch foaming suggested that it is possible to overcome these problems by controlling and improving the whole foaming process. The extrusion foaming is characterized by the same intense collapsing, density, and morphology issues as in the case of batch foaming and the methods to be used to improve the quality of extruded foams are analogous.

61259_C005.indd 148

10/25/2008 12:37:30 PM

Foaming Analysis of PCL and PLA

149

Experimental tests, performed on lab-scale foam extrusion equipment on a PCL/CO2 system led to foam characterized by a density of 0.05 g/cm3 and mean cell diameter of 200 μm, with the following temperature profile from the hopper to the die: 90, 100, 100, 100, 70, and 40°C. Higher uniformity of cell structure but higher densities were achieved with N2 as foaming agent, by using Tdie 45°C. The minimum mean cell diameter was 50 μm with a density of 0.3 g/cm3. As in the case of the batch process, the best results in term of morphology and density were achieved using as foaming agent in the extrusion a 20–80%wt CO2–N2 mixture: foams were obtained characterized at the same time by a low density (0.15 g/cm3) and a fine morphology (average cell diameter, d  20 μm).15 5.2.2 Foaming of PLA PLA is a bio-based biodegradable aliphatic polyester with chemical formula -[-O-(CHCH3)-CO-]n-. It is widely used in biomedical applications and is now finding commercial use for disposable items. PLA is generally produced by the ring-opening polymerization of lactide, a cyclic dimer prepared by the controlled depolymerization of lactic acid. Lactic acid can be manufactured by either a chemical synthesis or a carbohydrate fermentation from renewable natural resources such as corn starch, sugar cane, and sugar beet. Lactic acid polymers consist of mainly lactyl units, of only one stereo-isoform (PLLA and PDLA) or combinations of D and L lactyl units in various ratios (PDLLA). Commercial materials have, in general, low D content (up to 10%). The L-lactic acid based polymers (PLLA) may produce polymer which is a linear homopolymer of molecular size 70kDa.21 Although crystalline PLLA possesses many desirable properties (physical and mechanical), crystallization rates are extremely slow. The fastest half-time of crystallization reported in the literature for pure PLLA is 1.9 min for a 101 kg/mol sample.22 As already mentioned, this aspect limits the use of PLLA in foaming as well as in the other thermoplastic processing. Sorption experiments on the system PLA/CO2 have been performed by the Ohshima group at 200°C and up to 14 MPa. The authors report a rather linear isotherm with an equilibrium CO2 concentration of 0.05% at 10 MPa and mutual diffusivities in the range 2–5 × 105 cm2/s.23 PLA foams have been prepared by different research groups with the temperature increase method and with the pressure quench method, in a wide range of processing conditions. The two methods differ in the way the polymer-blowing agent solution is brought out of equilibrium, to induce the formation of the new gas phase (bubbles). In fact, during foaming the polymer is first saturated with the blowing agent at high temperature and pressure and then, to induce foaming, it is brought to super-saturation conditions, in which the gas prefers to run off the solution. Since, in general, the blowing agent solubility decreases with an increase in temperature

61259_C005.indd 149

10/25/2008 12:37:30 PM

150

Polymeric Foams

and/or a decrease in pressure, two methods can be used: the temperature increase (TI) and pressure quench (PQ). Generally, the former allows the achievement of ultra-fine morphologies: nanometric to micrometric celled PLA foams have been prepared in the temperature range 50–140°C with CO2. In particular, foams with a mean cell diameter of 200 nm have been obtained by heating CO2-saturated PLA (at 5.5 MPa and 25°C) to 60°C.24 At higher foaming temperatures, morphologies characterized by bigger pores have been obtained. For example, the Okamoto group obtained foams with a mean diameter of 230 μm at 140°C on CO2-saturated PLA (at 10 MPa and approximately 150°C).8 The temperature increase method, as with other thermoplastics, led to foams of relatively high density. The pressure quench method, on the contrary, allows the achievement of lower density foams, characterized by coarser structures. Mathieu et al. foamed PLA saturated with CO2 at pressures ranging from 15 to 21 MPa and at 195°C, obtaining foams with mean cell diameters ranging from 200 to 1000 μm. The selected foaming temperature was 195°C and pressure drop rates were selected in the range 0.2–1.2 MPa/s.25 Ema et al. foamed CO2saturated PLA, obtaining foams with densities as low as 0.1 g/cm3 and mean cell diameter of 60 μm, at foaming temperature of 150°C.26 Di et al. foamed PLA at 110°C after saturating the polymer with an 80/20%vol N2/CO2 mixture, obtaining foams with a density of 0.125 g/cm3 and mean cell diameter of 230 μm.27 Figure 5.4 summarizes the described results. As already observed in describing PCL foaming, from Figure 5.4 it seems that the processing window of PLA is rather large, with the possibility of achieving a wide range of morphologies and densities. This is, again, not true: for each single apparatus and processing technique, the

Mean cell diameter (μm)

1000 800

Mathieu et al. PQ (25) Ema et al. PQ (26) Di et al. PQ (27) Fujimoto et al. TI (8) Liao et al. TI (24)

600 400 200 0 50

100 150 Foaming temperature (°C)

200

FIGURE 5.4 Variation of the mean pore dimension of PLA foams. See cited references for experimental details; PQ: pressure quench method; TI: temperature increase method.

61259_C005.indd 150

10/25/2008 12:37:30 PM

Foaming Analysis of PCL and PLA

151

useful processing window is very narrow and, with respect to PCL, PLA foaming is even more critical, for the poor rheological properties and the very slow crystallization kinetics. In the following paragraphs, two methods utilized to enhance foamability of PCL and PLA will be described. In particular it will be shown how the design of the molecular architecture targeted to foaming and the addition of nucleating agents (micro- and nano-sized) may help foam stabilization, widen the processing windows, and reduce cell coalescence.

5.3 Molecular Modification of PCL and PLA 5.3.1 Branching/Cross-Linking of PCL The rheology of melts is much affected by molecular weight, molecular weight distributions (MWD) and the presence of short (SCB) and long chain branching (LCB). In the case of polyolefin melts, literature results mostly indicate that the effect of polydispersity on viscosity, elastic character, and activation energy of flow could be very similar to that expected due to the presence of LCB. That notwithstanding, the effects of LCB seem to be stronger than those due to polydispersity for a given molecular weight. Different relaxation processes appear as a consequence of the presence of LCB: slower terminal relaxation behavior than that of linear counterparts, and a faster additional branch relaxation (gel-like behavior) at higher frequencies, clearly distinguishable from polydispersity effects. Moreover, the branch number contributes less to the rheological behavior with respect to the topology of the branched polymers. Thus, branch position and architecture along the main polymer chain are the main factors controlling the viscosity function, elastic character and activation energy of flow.28 Within the polyolefin group, the polyethylene (PE) family is a good example of how it is possible to tailor the rheological properties, including those in extensional fields, by optimizing the molecular architecture, thanks to the evolution of the industrial polymerization of PE in line with market needs. The molecular architecture of linear thermoplastic polymers may be converted to branched/cross-linked polymers by promoting free-radical reactions induced by incorporating peroxides or by solid-state irradiation techniques. Both methods yield carbon–carbon cross-links, where crosslinking by irradiation takes place in the solid state and peroxide crosslinks occur in the molten state. Irradiation cross-linking is therefore a selective process that takes place mainly in the amorphous region, whereas in peroxide cross-linking the polymer is totally amorphous and thus structurally non-selective to the reaction with the peroxide radicals.29

61259_C005.indd 151

10/25/2008 12:37:30 PM

152

Polymeric Foams

Radiation effects in saturated linear polyesters depend on the specific chemical structure of the polymer. For example, biodegradable polyesters such as poly(glycolic acid) (PGA) and PLA are known to cleave and crosslink. PGA degrades and cross-links at about the same rate30,31 whereas PLA predominantly degrades.32 It has already been reported that PCL does not cross-link well by irradiation, due to the occurrence of chain scission during treatments, while it cross-links efficiently with peroxide in the molten state.33 The viscoelastic properties of PCL can be modified through the branching of the polymeric matrix using dicumyl peroxide (DCP), which was reported to generate polymeric radicals by hydrogen abstraction reactions.29 Polymeric radicals immediately react to generate branching and cross-linking and the resulting molecular modification can be easily followed by measuring the dynamic rheological properties of the melt in isothermal experiments. As reported in Reference 34, the polymer was first blended with dicumyl peroxide at a low temperature (80°C) in order to prevent premature peroxide decomposition. The peroxide modification was then performed at different temperatures, from 110°C to 150°C. The reaction kinetics was followed by measuring the dynamical rheological properties of the melt in isothermal experiments by using a parallel plate rheometer. The evolution of the macromolecular structure during the chemical reaction was followed by analyzing the time evolution of the real and imaginary component of the complex viscosity. As reported in Reference 35 the measurement of the rheological properties allows the analysis of the branching reaction extent, even at very low branching levels. The degree of chemical modification, analyzed by measuring the increase of storage modulus during dynamic rheological experiments in isothermal conditions (see Figure 5.5), can be controlled by: (a) DCP content, (b) reaction time, and (c) temperature. As expected, the reaction rate of branching increases when the curing temperature increases. Moreover, the final degree of branching is correlated to the value of storage modulus (G) at the plateau, and increases with the amount of DCP. The degree of molecular modification and, in particular, the level of branching/cross-linking, resulted in a strong change of the levels of elastic and dissipative components and a strong variation of the dependence of these components upon frequency. Figure 5.6 reports the complex viscosity and tan d versus frequency of samples fully cured at 130°C and at different DCP concentrations. Even at very low DCP content, the complex viscosity increases rapidly, particularly at the lower frequencies, revealing a reduced Newtonian behavior. The analysis of the dependence of tan d versus frequency allows assessing the occurrence of gelation phenomena, characterized by constant values of tan d versus frequency.36 As shown in Figure 5.6 tan d decreases with frequency at concentration of DCP below 0.5% and it is constant or increases with frequency at higher concentrations of DCP. These results

61259_C005.indd 152

10/25/2008 12:37:30 PM

153

Foaming Analysis of PCL and PLA

1.2 × 105 1 × 105

G′ (Pa)

G′ (Pa)

8 × 10

3 × 104

4

6 × 104 4 × 104

104 8 × 103

2 × 104

6 × 103

0 × 100 0 × 100

5 × 103 1 × 104 time (s)

4 × 103 0 0 × 10

1.5 × 104

5 × 103 1 × 104 time (s)

1.5 × 104

FIGURE 5.5 Viscoelastic properties of PCL modified with DCP: () cured at 140°C; () cured at 130°C; () cured at 110°C. Solid lines: DCP content of 1%; dashed lines: DCP content of 0.25%.

are in agreement with the literature data29 where it has been shown that some gel content is present only for concentrations higher than 0.5% and it increases with the amount of DCP. The authors have reported that a gel content of 35% can be obtained at 1% of DCP and around 60% at 2% of DCP (180°C in a Brabender). The chemically modified materials were foamed in a batch apparatus using the following procedure. PCL samples were saturated at 75°C with N2 at saturation pressure, Psat  180 bar, for at least 6 hours. The vessel was then cooled to different foaming temperatures, Tfoam, in the range 40–55°C. The pressure drop rate was P˙  320 bar/s. A more detailed description of the experimental conditions has been provided in Reference 34. Table 5.1 reports the effect of the foaming temperature on the final foam density for both neat PCL and peroxide-modified PCL. Results show that the density of neat PCL decreases with the increase of temperature from 40°C to 45°C and then it increases at higher temperatures. 105 2

% DCP

% DCP

101

104

0.5

tan d

h* (Pa·s)

1

0.25

0.25 0.5 1 2

100

0

10–1 FIGURE 5.6 content.

61259_C005.indd 153

0

100 w (rad/s)

101

10–1

10–1

100 w (rad/s)

101

Viscoelastic properties of fully cured PCL–DCP samples; effect of DCP

10/25/2008 12:37:30 PM

154

Polymeric Foams

TABLE 5.1 PCL and Modified PCL Foam Densities Foaming Temperature (°C)

Pure PCL (g/cm3)

0.5%wt DCP (g/cm3)

40

0.0688

—

—

41

0.0538

—

—

45

0.0338

0.368

0.271

49

0.0372

0.0691

0.0791

53

0.0793

55

0.181

— 0.0508

1%wt DCP (g/cm3)

— 0.0576

This behavior is due to several mechanisms taking place during foaming. At a high temperature, cell collapse occurs due to the decrease of the melt strength of the polymeric matter, leading to foams of higher density. At lower temperatures, higher densities are related to an increase of viscosity and/or to the occurrence of partial crystallization of the polymeric matrix. Both the increase of viscosity and the occurrence of crystallization reduce the deformability of the expanding matter and thus the foaming efficiency. When using the peroxide cured PCL, the increase of viscosity due to the molecular modification led to foams with higher densities (see Table 5.1) at lower temperatures. As reported in Table 5.1, chemically modified PCL can be foamed at higher temperatures and lower densities, compared to unmodified PCL, can be obtained. Conversely, foams prepared with chemically modified PCL at temperature of 55°C showed neither coalescence nor cell wall rupture. Moreover, the morphologies of these foams are characterized by uniform cellular structure and fine cell dimensions. This is related to the enhancement of the viscoelastic properties that reduce the cell wall rupture and the collapse of the cellular structure.37 Neat PCL foams, conversely, at the same temperature have collapsed, as evidenced by the higher density and by the poor morphology. 5.3.2 Chain Extended PLA PLA presents several limitations such as low thermal, oxidative, and hydrolytic stability that leads to a decrease of molecular weight during processing. The melt viscosity and, in general, the rheological properties of the melt in both shear and elongational fields are therefore often compromised by these degradation processes. To overcome such shortcomings, considerable research efforts have been devoted to control the melt rheology by increasing the molecular weight to compensate for the molecular weight decrease caused by processing degradation. Several approaches were followed and among them we should mention the

61259_C005.indd 154

10/25/2008 12:37:30 PM

Foaming Analysis of PCL and PLA

155

free-radical branching/linking with peroxide as radical initiators,38–41 the solid-state post-polymerization technique,42 the chain extension technique.27,43–46 Moreover, among the most recent work on the use of chain extension reactions we should mention the works by Vollalobos et al. on the use of novel oligomeric and low molecular weight polymeric chain extenders based on epoxy-functional (meth)acrylic monomers and nonfunctional (meth)acrylic and/or styrenic momomers47,48 and on the use of isocyanates compounds patented by Unika Ltd.49 It has been recently reported that 1,4-butanediol (BD) and 1,4-butane diisocyanate (BDI) can be used to modify commercial PLA in two steps. In the first step, BD was selected as the first coupling agent and acid value reducer to link carboxyl groups of PLA and then, in the second step, BDI was added to let it react with hydroxyl end groups of PLA to achieve chain-extended PLA. The different ratios of two chain extenders were used to investigate their effect on the structure of modified PLA samples which were then characterized and foamed in a batch foaming process. In particular, three modified PLA samples were obtained by using different ratios of BD to BDI: modified sample 1 (PLAM1): COOH/BD  2 : 1, OH/BDI  2 : 1, that is, equimolar amount of BD and BDI relative to end groups COOH and OH of PLA, where COOH and OH contents were calculated from the acid values determined by titration; sample 2 (PLAM2): COOH/BD  2 : 1, OH/BDI  1 : 1, that is, BDI amount was excessive compared to M1; sample 3 (PLAM3): COOH/BD  1 : 1, OH/BDI  2 : 1, that is, BD was excessive compared to PLAM1. The original PLA was also processed at the same conditions for comparison purposes and is named neat PLA hereafter. Details on the materials and chemical modification procedures are reported in Reference 27. Table 5.2 reports GPC results of plain and modified PLA. It is clearly shown that molecular weight of samples PLAM1 and PLAM2 were increased compared to plain PLA because of chain extension. The excess of DBI in sample PLAM2 resulted in materials with the highest molecular weight (Mw) and larger molecular weight distribution (MwD) while the excess of BD (PLAM3) resulted in materials with lowest Mw and even larger MwD compared to plain PLA. The increase of Mw and of polydispersity in samples PLAM1 and PLAM2 resulted in materials whose rheological properties were characterized by higher viscosity and higher elasticity.27,50 While PLA exhibited typical Newtonian behavior at low frequency and a small shear thinning starting at a frequency of 10 rad s1, chemically modified materials (both PLAM1 and PLAM2) showed non-Newtonian behavior even at very low frequency (see Figure 5.7). The increase of complex viscosity at lower frequency and the lower slope in G in the same frequency range is the evidence of the presence of gel-like structures coming out from cross-linking reactions taking place during the chemical modification of PLA. It is reasonable to expect that, due to the excess of BDI, the isocyanate groups

61259_C005.indd 155

10/25/2008 12:37:30 PM

156

Polymeric Foams

TABLE 5.2 Characteristics of Pure and Modified PLA Sample

Pure PLA

3

Mn/10 (g/mol) Mw/103 (g/mol) Mw/Mn Tg (°C) Tc (°C) Tm (°C) Average cell size (μm) Cell number density (108/cm3) Foam density (g/cm3)

57 124 2.2 61.8 111.5 169.6 227 0.008 0.125

M1 84 225 2.7 63.2 129.1 154.5 37 1.9 0.067

M2 107 308 2.9 63.7 130.2 153.8 24 6.7 0.092

M3 48 156 3.3 55.7 113.8 156.3 223 0.008 0.179

can react with carboxyl groups of PLA leading to amide bond that can further react with additional BDI and cause cross-linking in PLA.27 The improved viscous and elastic properties of modified PLA samples resulted in foams of lower foam density and higher cell density, as reported in Table 5.2.

5.4 Nanocomposites Nanometric additives have been reported to dramatically change properties such as gas solubility and diffusivity, rheological characteristics, and

106

106 105 104 103

104 102

G′ (Pa)

h* (Pa·s)

105

101

103

100 102 0.01

0.1

1

10

10–1 100

Frequency (rad/s) FIGURE 5.7 Viscoelastic properties of pure PLA and modified PLA; closed symbols G¢; open symbols h*. (●) neat PLA; (■) M1; (▲) M2.

61259_C005.indd 156

10/25/2008 12:37:30 PM

Foaming Analysis of PCL and PLA

157

crystallization, which are extremely important during foaming. In the industrial practice of thermoplastic foaming, furthermore, it is of vital importance the use of nucleating agents, additives which induce the heterogeneous nucleation of bubbles, with the aim of controlling the final morphology of the foam.8,51 In this section, the effects of nanofillers in modifying the relevant properties for foaming (mass transport phenomena, rheology, and crystallization) as well as their role in bubble nucleation is reviewed for PCL and PLA-based nanocomposites. 5.4.1 Nanocomposites from PCL and PLA: Rheology, Sorption, Mass Transport, and Crystallization Several papers, reviews, and patents have been published recently in the scientific literature, following the pioneering work from the Toyota Central Research Group52 on the use of nanometric additives to improve certain properties of the polymer (mainly mechanical) at very low concentration, with respect to conventional (micrometric) fillers. These concepts were also applied in the area of biodegradable polymers and there are now numerous reviews on nanocomposites based on biodegradable polymers.53–55 Specific literature on PCL and PLA has also been presented, reporting on the preparation and properties of nanocomposites from different materials such as clays, organo-modified clays, titanate, hydroxyapatite, mica, saponite, and smectite, produced via melt mixing, solvent casting, or in-situ polymerization. Intercalated and exfoliated structures have been obtained with improved mechanical, rheological, transport, and thermal properties.56–65 In the analysis of rheological properties of these systems, the characteristic features of nanocomposites; for example, the deviation from terminal flow behavior and pseudo-solid-like behavior, have been observed by Lepoittevin et al. on PCL (of an average molecular mass of 49,000) melt mixed with a montmorillonite (MMT) modified by methyl bis(2-hydroxyethyl) ammonium cation, MMT-(OH)2 and PCL melt mixed with a MMT modified by dimethyl 2-ethylhexyl ammonium cation (MMT-Alk).66 Authors, to show the importance of organo-modification for achieving intercalation and exfoliation, also reported the rheological properties of unmodified MMT-Na, which resulted in micro-composites with unchanged rheological properties with respect to pure PCL. Di et al. melt mixed PCL (of an average molecular mass of 69,000) with MMT-(OH)2 and a MMT modified by methyl dihydrogenated tallow ammonium (MMT-2HT), reporting exfoliated structures only with the former organo-modifier.12 In order to give a picture of the extent of the rheological modification induced by the nanometric particles, complex viscosities versus frequencies are reported in Figure 5.8 for PCL-MMT-(OH)2 nanocomposites. The h*s of the pure molten polymer show only a small frequency dependence, revealing a Newtonian plateau at low frequency. The h* of the nanocomposites

61259_C005.indd 157

10/25/2008 12:37:31 PM

158

Polymeric Foams

107 Pure PCL 1% 2% 3.5% 5% 7% 10%

h* (Pa·s)

106

105

104 90°C 1000

0.1

1 w (rad/s)

10

100

FIGURE 5.8 Complex viscosity curves of PCL and PCL-MMT-(OH)2 nanocomposites.

are higher overall than that of the neat polymer within the frequency range studied and h* increases with the organoclay content. Figure 5.9 reports G and G versus frequencies for PCL-based nanocomposites. Typical patterns of exfoliated nanocomposite systems, with solid-like behavior and the presence of a yield stress can be observed.12,67–69 Similar results have been obtained on several nanocomposites based on PLA. Ray and Okamoto,70 for example, reported the rheological properties of PLA melt-mixed with MMT, organically modified with octadecyl ammonium cation (C18-MMT) and observed non-terminal flow and pseudosolid behavior at concentrations of 3–7 wt%. Similar results have been 106

Moduli (Pa)

105 104 1000 100 10 0.001

0.01

0.1 w (rad/s)

1

FIGURE 5.9 Viscoelastic properties of neat PCL and PCL/clay nanocomposite, 80°C. G¢ (open symbols), G≤ (closed symbols). (●) neat PCL; (▲) 2%wt clay; (■) 10%.

61259_C005.indd 158

10/25/2008 12:37:31 PM

159

Foaming Analysis of PCL and PLA

106 10 wt% 5 wt% 105 h* (Pa s)

2 wt% 0 wt%

104

103 10–3

10–2

10–1 100 w (rad/s)

101

102

FIGURE 5.10 Complex viscosity curves of PLA and PLA/MMT-(OH)2 nanocomposites with different weight fractions of MMT at 170°C.

obtained by Di et al.71 on PLA nanocomposites with 2–10 wt% MMT-(OH)2 and by Pluta72 on the same system at 3 wt%. Typical curves are reported in Figure 5.10. Ray and Okamoto,70 also performed elongational rheology experiments and observed the rise of a strain-induced hardening for the nanocomposites with 5 wt% C18-MMT, despite the very low rheological properties of pure PLA. As will be seen in the following, these improvements are crucial for foaming. Less numerous, but important in foaming, are the studies on mass transport and sorption thermodynamics of gases in nanocomposite systems. In particular, gas barrier properties have been shown to improve dramatically upon exfoliation of clay platelet in a number of polymeric systems.73–78 In our systems, air permeability was reduced by 50% with the addition of 5 wt% MMT-(OH)2 to PCL,79 while O2 permeability is reduced by as much as 70% when modified fluorine mica was added to PLA.80 Figure 5.11 reports the mutual diffusivity of CO2 and PCL and its nanocomposites with MMT-(OH)2, evidence of the decrease of diffusivity with the increase of clay content, as already observed. The mechanism of improvement is attributed to the increase in the tortuosity of the diffusive path for a penetrating molecule, this effect depending on the aspect ratio of the additives, their concentration, and dispersion. In fact, the observed reduction in diffusivity at 10 wt% loading was less pronounced than expected, being slightly below the diffusivity at 5 wt% loading. This behavior can be related to the poorer dispersion and the lower degree of exfoliation of clay platelets that is obtained at a concentration of 10% compared to 5%. In effect, when the concentration of clay is 10%, the peak of the organomodified clay in the X-ray diffractograms did not disappear, as in the case of

61259_C005.indd 159

10/25/2008 12:37:31 PM

160

Polymeric Foams

8 × 10–5

Mutual diffusivity (cm2/s)

7 × 10–5 T = 70°C

6 × 10–5 5 × 10–5

Pure PCL Nano-PCL 5% Nano-PCL 10%

4 × 10–5 3 × 10–5 2 × 10–5 1 × 10–5 0

0

150 200 100 mgCO2/g polymer

50

250

300

FIGURE 5.11 Mutual diffusivity as a function of CO2 concentration in pure PCL and its nanocomposites with MMT-(OH)2.81 (From Cotugno, S. “Modeling of Thermodynamic and Transport Properties of Polymer Melts and Solutions to be used in the Simulation of Foaming Processes.” PhD thesis, University of Naples Federico II, Italy, 2003.)

5%, but shifted to a lower angle suggesting that the nanoparticles were intercalated and not fully exfoliated.12 Figure 5.12 reports the equilibrium concentration of CO2 in PCL and PCL-MMT-(OH)2 nanocomposites, with respect to the polymeric matrix, evidence of a slight increase in solubility in the presence of nanometric particles, which can be justified by the introduction of defects and microscopic gas volumes close to the polymer/clay interfaces. In effect, it has 350 T = 70°C 300

mgCO2/gPCL

250 200 150

Pure PCL nc 5% nc 10%

100 50 0

0

50

100

150

200

250

300

Pressure (bar) FIGURE 5.12

61259_C005.indd 160

Sorption isotherms for pure PCL and its nanocomposites with MMT-(OH)2.

10/25/2008 12:37:31 PM

161

Foaming Analysis of PCL and PLA

20

Heat flow (mW)

15

Pure PCL 1%wt clay 2%wt clay 5%wt clay 10%wt clay

10

5

0 10

15

20

25 T (°C)

30

35

40

FIGURE 5.13 Non-isothermal, melt crystallization experiments of pure PCL and PCL/clay nanocomposites.

been observed that the concentration of the blowing agent might not be uniform throughout the polymer matrix but may be higher in some domains close to the solid particles. Thermal properties are also important in foaming, as already observed, since they take part in the setting mechanism of a newly formed cellular structure. This is particularly true for semi-crystalline polymers such as PCL and PLA. It has been extensively reported that nanocomposites added to polymeric materials favor crystallization, causing significant modification of the thermal behavior of the polymeric matrix. In Figure 5.13 we compare the crystallization process for neat PCL and PCL-based nanocomposites. The presence of the nanometric filler results in a relevant increase of the crystallization temperature during cooling. The same effect has been observed with PLA-based nanocomposites, with a faster and more intense crystallization with respect to the pure polymer.71 This phenomenon has been classically described as a heterogeneous nucleation effect of the solid particles for the polymeric crystals. Heterogeneous nucleation was observed in several polymers such as polyethylene,82 polyamide 6,83 polyamide 6,6,84 polyamide 12,12,85 polypropylene,86,87 and syndiotactic polystyrene.88 In foaming, the easier crystallization will result in an easier stabilization of the cellular structure and an improved foamability. As a counterpart anyway, a possible increase in the foam density can occur due to premature crystallization during foaming at low temperature.

61259_C005.indd 161

10/25/2008 12:37:32 PM

162

Polymeric Foams

5.4.2 Foaming of PCL and PLA-Based Nanocomposites The improved elastic and viscous properties of nanocomposites have encouraged researchers to study their foaming process. Figure 5.14 reports, as an example, the SEM images of PCL-based nanocomposites with MMT(OH)2 (neat PCL, 1 wt% and 0.4 wt%, 16a, b, and c, respectively, foamed at 40°C with an 80/20% volume mixture of N2 and CO2). It is evident that the morphologies of the foams obtained from the pure polymer and its nanocomposites are rather different, with a finer morphology of the nanocomposites, characterized by smaller and uniformly distributed cells (almost a one-order-of-magnitude reduction in the mean cell diameter). The density of the 1% nanocomposites foam is higher than the one of the neat PCL (0.07 g/cm3 instead of 0.04 g/cm3 for the neat PCL). This was due to the increased viscoelasticity and crystallization rate of the nanocomposites with respect to the neat PCL. By comparing Figure 5.14c with Figure 5.14a and b, an interesting effect can be noticed: at 0.4 wt% of clay, an open-cell, very fine structure was achieved. This behavior has been ascribed to the crystalline phase nucleation and bubble nucleating effect of the filler. In fact, the very low clay concentration (0.4 wt%) has been reported to nucleate the crystalline phase, resulting in a steep increase of the crystallization kinetic, with respect to the pure PCL and to higher clay concentration. The occurrence of crystallization increases the strength while reducing the deformability of the expanding matter, hence an opencell, relatively high density (0.12 g/cm3) structure was achieved.89 Similar results have been achieved with PLA. Figure 5.15 reports the SEM images of foamed PLA and PLA nanocomposites. We can see that, under the same foaming condition, finely dispersed cells were formed in foams of PLA and PLA nanocomposites. Table 5.3 summarizes the characteristics of PLA-based foams. We note that the average cell size of pure PLA foam is very large, while it decreases significantly with 1 wt% of clay, and then levels off at higher MMT-(OH)2 concentration. The cell densities increase steeply with the presence of organoclay [from 0.008 ¥ 108 cell/cm3 for pure PLA to 5.1 ¥ 108 cell/cm3 for PLA/5 wt% MMT-(OH)2]. The bulk foam densities were measured and found to be significantly affected by the clay content.

FIGURE 5.14 SEM micrographs of (a) neat PCL; (b) nanoclay PCL (1%); and (c) nanoclay PCL (0.4%).

61259_C005.indd 162

10/25/2008 12:37:32 PM

163

Foaming Analysis of PCL and PLA

FIGURE 5.15 SEM micrographs for foams of (a) PLA and (b) PLA nanocomposite with 2 wt% MMT-(OH)2 foamed at 110°C with an 80/20% volume mixture of N2 and CO2.

The outlined results are common to several nanocomposite systems, like in PP/clay, as reported in recent work from the Toyota Technical Institute, in which up to two orders of magnitude increase in the cell number densities have been measured.90,91 The same group also reported successful results on a polycarbonate/fluoroectorite system.51 Other examples include polystyrene,92–95 polyamides,96 polyurethanes,97 and high-density polyethylene.98 To summarize, the beneficial effect of nanometric fillers in thermoplastic foaming can be ascribed to (1) the enhancement of the strength of the polymer melt to prevent melt fracture and foam collapse during bubble growth; (2) the enhancement of the crystallization kinetics to facilitate foam setting immediately after the growth; and (3) the induction of the formation of numerous bubbles. This aspect of crystal nucleation induction and of bubble nucleation induction is further analyzed in the next paragraph. 5.4.3 Bubble versus Crystal Nucleating Effect of Nanoparticles The effect of solid particles in foaming of a semi-crystalline polymer can be considered to be of a dual nature: from one side the particles can take TABLE 5.3 Average Cell Size, Cell Density and Densities of PLA Nanocomposite Foams with Different Weight Fractions of MMT-(OH)2. The Averages were Calculated from Experimental Observations on Five Different Samples Having Densities in the Reported Range Sample PLA PLA/1 wt% MMT-(OH)2 PLA/2 wt% MMT-(OH)2 PLA/5 wt% MMT-(OH)2

61259_C005.indd 163

Average Cell Size (μm)

Average Cell Density (108/cm3)

230 42 32 25

0.008 1.098 2.38 5.18

Foam Density (g/cm3) 0.10–0.14 0.18–0.22 0.23–0.35 0.30–0.42

10/25/2008 12:37:32 PM

164

Polymeric Foams

part in the nucleation of polymeric crystals and induce, increase or fasten crystallization; from the other side the particles can take part in the nucleation of bubbles and induce an improvement of the foam morphology. The two-nucleation phenomena of crystallization and bubbling should not be, in principle, correlated. In particular, the efficiency of the different nucleating agents, which is the main interest in industry, in nucleating crystals or bubbles, can be different and selective nucleation towards crystals or bubbles can be observed. We have recently analyzed this problem and studied the effect of different micrometric and nanometric nucleants in the foaming behavior of PCL.99 Nanoparticles from different materials with different shapes and/or characteristic dimensions and a traditional micrometric nucleating agent (talc) have been melt mixed with PCL and subsequently batch foamed with nitrogen and carbon dioxide. The main characteristics of the nucleating agents are reported in Table 5.4. The isothermal crystallization of PCL nanocomposites analyzed by differential scanning calorimetry (Figure 5.16) show the different abilities in inducing the formation of the crystalline phase (crystal nucleation) by different nucleating agents. All the nanocomposites investigated showed a faster crystallization rate compared to neat PCL. In particular, nanocomposites based on carbon nanotubes showed the highest crystallization rate while those based on alumina spherical nanoparticles the smallest. The immediate consequence of these changes in the crystallization rates is that the polymer can experience premature crystallization phenomena during the cooling from the temperature at which gas is solubilized to the temperature at which the material is foamed. To better explain this relationship between foaming and crystallization, it is possible to ideally TABLE 5.4 List of the Nucleating Agents used in this Study, from Producer’s Datasheets Trade Name Talc

Finntalc M03 Alumina Aeroxide AluC 130F Hombitec RM130F 30B Cloisite 30B cn

61259_C005.indd 164

Typology

Shape

Characteristic Dimension

Aspect Ratio

Surface Area (m2/g)

Talc

Cubic

1 μm

1

13.5

Al2O3

Spherical

13 nm

1

15–100

TiO2

Spherical

15 nm

1

15–100

Exfoliated

100 (width/ thickness)

750

15 nm

20–5000 40–600 (length/O.D.)

Modified Platelet montmorillonite Aldrich multi-walled Tube 636509-2G carbon nanotubes

10/25/2008 12:37:32 PM

165

Foaming Analysis of PCL and PLA

Cooling time window cn Specific heat flow (a.u.)

130F 30B Talc

Alumina Neat PCL

0

10

20

30 Time (min)

40

50

60

FIGURE 5.16 Isothermal melt-crystallization at 45°C for the 0.4 wt% composites.

superpose a cooling time window (that is the period of time during which the polymer–gas solution is cooled prior to foaming) to the crystallization curves, as schematically shown in Figure 5.16. When crystallization is fast and the DSC peak is anticipated with respect to the cooling time window (shadowed area in Figure 5.16), crystallization has already occurred before foaming which in fact cannot take place, as in the case of carbon nanotubes. When the cooling time window precedes crystallization, foaming occurs on a molten polymer and foaming efficiency is maximum (minimum density) as is the case of pure PCL and PCL with alumina. In between these two extremes, all of the possibilities of partial superposition of the time windows, with partial densification effect, are possible. Of course, cooling rate and foaming temperatures can be increased to avoid crystallization and shift the shadowed area accordingly. Figure 5.17, as an example, reports the effect of the nucleating agents on the final densities of PCL-based foams produced with N2 at a foaming temperature of 46.5°C. As can be observed, the effect of alumina on foam densification is negligible, while the other nanoparticles affect foaming to an extent that depends on the nucleating agent type and content, in particular on its ability to induce crystal nucleation. In the case of nanocomposites based on carbon nanotubes, the selected foaming temperature and/or cooling rate were too low to allow foaming before crystallization and, as a result, final density is that of the unfoamed, bulk semi-crystalline polymer. More complex, for the more numerous inter-dependencies, is the evaluation of nucleation efficiency of bubbles by the different nucleating agents. In effect, despite the two nucleating phenomena being very different, the presence of newly formed crystals (already nucleated) may induce the bubble nucleation, adding a new possible bubble nucleation mechanism.18 Hence, in this case, bubble nucleation can be related to several

61259_C005.indd 165

10/25/2008 12:37:33 PM

166

Polymeric Foams

1.2

Foam density (g/cm3)

1.0

Neat Talc Alumina

0.8

130F 30B

0.6

cn

0.4 0.2 0

0

0.1 0.4 0.7 1.0 Nucleating agent content (wt. %)

FIGURE 5.17 Effect of concentration of the nucleating agents on the foam densities of PCL-based systems, foamed with N2 at 46.5°C.

mechanisms: (1) the homogeneous bubble nucleation in the polymer; (2) the heterogeneous nucleation on the surface of the nucleating agents; and (3) the heterogeneous nucleation induced by the forming crystals. The relative importance of the two latter phenomena (nucleation of bubbles induced by the presence of the nucleating agents and by the presence of the polymeric crystals) depends again on the superposition of the two time frames, foaming and crystallization, which are strongly dependent on the foaming temperature. Bubble nucleation is, moreover, strongly affected by the kind of nucleating agent, the number of particles, their dispersion in the polymeric matrix and by the different surface interactions between the polymer and nanoparticles which can, in turn, eventually be affected by the chemical–physical properties of the blowing agent. This effect is reported in Figure 5.18, where cell number densities per unit initial volume of the composites are reported for selected samples foamed with carbon dioxide and nitrogen. All of the nanocomposites foamed with nitrogen, except the one based on spherical TiO2 (130F), led to cellular structure characterized by higher cell densities compared with materials foamed with carbon dioxide. Nanometric Al2O3 did not result in any improvement of the cellular structure compared to neat PCL, while talc and nanoclay (30B) displayed a bubble nucleation efficiency that seemed to be independent of the type of expanding gas. A singular behavior was observed for nanocomposites containing TiO2. The highest nucleation efficiency obtained with carbon dioxide suggests that, for this type of nanoparticle, the bubble nucleation is not only based on a heterogeneous mechanism occurring at the surface of the particles but also on a more complex mechanism that may involve specific interactions

61259_C005.indd 166

10/25/2008 12:37:33 PM

167

Foaming Analysis of PCL and PLA

107

106

105

130F

30B

Talc

Alumina

104

N2 CO2

Neat

Cell density (#/cm3)

108

FIGURE 5.18 Effect of the nucleating agent (at 0.4 wt%) on the cell number densities; comparison between N2 and CO2 as the blowing agent, foamed at 47.5°C and 30°C respectively.

between gas, polymer, and particles as well as mechanisms related to premature crystallization of the polymer. All of these mechanisms are strongly affected by the foaming temperature. This is clearly showed in Figure 5.19, where cell number densities as a function of foaming temperature are reported for the different polymer/nucleating agent systems. It is evident that at lower foaming temperatures, where crystallization has already occurred, cell number density is high and, in particular, follows the same trend of the crystal nucleation efficiency (compare with Figure 5.16, where the crystallization rates for 30B talc neat PCL). By increasing the temperature, crystallization is shifted to longer times with respect to foaming and the nucleation of bubbles induced by the presence of crystals is suppressed (bubble nucleation takes place before crystallization). This results in a decrease in cell number densities. At the highest temperature investigated, the cell number density increased again for neat PCL and for all the nucleating agents except, surprisingly, nanocomposites with nanoclay (30B). These data clearly show that the bubble nucleation mechanism induced by nanoparticles is still far from completely understood since there are many unknown aspects that could contribute to the nucleation efficiency of these very small size nucleants: their size; geometry and aspect ratio; surface properties; the presence of compatibilizers (often low molecular weight compounds), which can modify the local gas solubility and diffusivity around the particles; the presence of a higher macromolecular order close the particle surface that can further modify local thermodynamic; and/or the molecular mobility of the molecular chains.

61259_C005.indd 167

10/25/2008 12:37:33 PM

168

Polymeric Foams

Cell density (#/cm3)

109 Neat Talc 30B 130F Alumina

108

107

106

105 46

47

48

49 Tfoaming

50

51

52

FIGURE 5.19 Effect of foaming temperature on the foam densities of selected composites containing 0.4 wt% of nucleating agent.

5.5 Conclusions Among the most interesting biodegradable polymers that can be potentially employed for foaming are polyesters such as PCL and PLA. However, they have several limitations due to their poor rheological properties ansd small processing window. In particular, the main limitations in the use of PCL are related to the low processing temperatures and, in the case of PLA, to the low crystallization kinetics and the poor extensional rheological properties. The branching of polymer chains as well as the optimization of molecular weight and molecular weight distribution are common methods used to improve the extensional viscosity of a polymer and to make it suitable for foam formation. In fact, peroxide modified PCL can be foamed at higher temperatures compared with unmodified PCL and the morphologies of these foams are characterized by a uniform cellular structure and fine cell dimensions. Commercial PLA can be modified by using low-molecularweight chain extenders and improved melt viscosity and elasticity can be obtained. Such improvements result in foams with much reduced cell size, increased cell density, and lowered foam density compared with unmodified PLA. Nanometric additives have shown to dramatically change properties such as gas solubility and diffusivity, rheological characteristics, and crystallization behavior, properties that are extremely important for the foam formation and lead to cellular structures characterized by different foam density, cell size, and cell density. In particular, nanometric fillers induce

61259_C005.indd 168

10/25/2008 12:37:33 PM

Foaming Analysis of PCL and PLA

169

both crystalline phase nucleation and bubble nucleation and these effects were found to depend on the type, dimension, shape, and surface functionality of the nucleating agent, on the blowing agent, and temperatures used for foaming.

References 1. Klempner, D. and Sendijarevic, V. Polymeric Foams and Foam Technology. 2nd edition. Hanser Publishers/Hanser Gardener Publications, Inc., Cincinnati, OH, 2004. 2. Boyce, S. T. “Skin repair with cultured cells and biopolymers.” In Human Biomaterials Applications, ed. D. L. Wise, The Humana Press, Inc., Totowa, NJ, 1996, pp. 347–377. 3. Cotugno, S., Di Maio, E., Mensitieri, G., Nicolais, L., and Iannace, S. “Biodegradable foams.” In Handbook of Biodegradable Polymeric Materials and their Applications, ed. B. Narasimhan and S. K. Mallapragada. American Scientific Publishers, 2006, pp. 55–106. 4. Lee, S. T., Park, C. B., and Ramesh, “Biodegradable foams.” In Polymeric Foams, Science and Technology. Taylor and Francis (CRC Press), Boca Raton, 2006, pp. 165–192. 5. Siripurapu, S., Gay, Y. J., Royer, J. L., and DeSimone, J. M. “Generation of microcellular foams of PVDF and its blends using supercritical carbon dioxide in a continuous process.” Polymer 43 (2002): 5511–5520. 6. Spitael, P. and Macosko, C. W. “Strain hardening in polypropylenes and its role in extrusion foaming.” Polymer Engineering Science 44 (2004): 2090–2100. 7. Xanthos, M., Young, M. W., Karayannidis, G. P., and Bikiaris, D. N. “Reactive modification of polyethylene terephthalate with polyepoxides.” Polymer Engineering Science 41 (2001): 643–655. 8. Fujimoto, Y., Ray, S. S., Okamoto, M., Ogami, A., Yamada, K., and Ueda, K. “Well-controlled biodegradable nanocomposite foams: from microcellular to nanocellular.” Macromolecular Rapid Communications 24 (2003): 457–461. 9. Ray, S. S. and Okamoto, M. “New polylactide/layered silicate nanocomposites, 6.” Macromolecular Material Engineering 288 (2003): 936–944. 10. Gaur, U., Lau, S.-F., Wunderlich, B. B., and Wunderlich, B. “Heat capacity and other thermodynamic properties of linear macromolecules. VIII. Polyesters and polyamides.” B. Journal of Physical Chemistry 12 (1983): 65–89. 11. Lebedev, B. and Yevstropov, A. “Thermodynamic properties of polylactones.” Makromolekulare Chemie 185 (1984): 1235–1253. 12. Di, W., Iannace, S., Di Maio, E., and Nicolais, L. “Nanocomposites by melt intercalation based on polycaprolactone and organoclay.” Journal of Polymer Science Part B: Polymer Physics 41 (2003): 670–678. 13. Di Maio, E., Iannace, S., Sorrentino, L., and Nicolais, L. “Isothermal crystallization in PCL/clay nanocomposites investigated with thermal and rheometric methods.” Polymer 45 (2004): 8893–8900. 14. Adamovsky, S. and Schick, C. “Ultra-fast isothermal calorimetry using thin film sensors.” Thermochimica Acta 415 (2004): 1–7.

61259_C005.indd 169

10/25/2008 12:37:33 PM

170

Polymeric Foams

15. Di Maio, E., Mensitieri, G., Iannace, S., Nicolais, L., Li, W., and Flumerfelt, R. W. “Structure optimization of polycaprolactone foams by using mixtures of CO2 and N2 as blowing agents.” Polymer Engineering Science 45 (2005): 432–441. 16. Jenkins, M. J., Harrison, K. L., Silva, M. M. C. G., Whitaker, M. J., Shakesheff, K. M., and Howdle S. M. “Characterisation of microcellular foams produced from semi-crystalline PCL using supercritical carbon dioxide.” European Polymer Journal 42 (2006): 3145–3151. 17. Xu, Q., Ren, X., Chang, Y., Wang, J., Yu, L., and Dean, K. “Generation of microcellular biodegradable polycaprolactone foams in supercritical carbon dioxide.” Journal of Applied Polymer Science 94 (2004): 593–597. 18. Reignier, J. Tatibouët, J., and Gendron, R. “Batch foaming of poly(ecaprolactone) using carbon dioxide: Impact of crystallization on cell nucleation as probed by ultrasonic measurements.” Polymer 47 (2006): 5012–5024. 19. Cotugno, S., Di Maio, E., Mensitieri, G., Iannace, S., Roberts, G. W., Carbonell, R. G., and Hopfenberg, H. B. “Characterization of microcellular biodegradable polymeric foams produced from supercritical carbon dioxide solutions.” Industrial and Engineering Chemical Research 44 (2005): 1795–1803. 20. Marrazzo, C., Di Maio, E., Iannace, S., and Nicolais, L. “Process-structure relationship in PCL foaming.” Journal of Cellular Plastics 44 (2008): 37–52. 21. Narayanan, N., Roychoudhury, P. K., and Srivastava, A. “L (+) lactic acid fermentation and its product polymerization.” Electronic Journal of Biotechnology 7 (2004): 167–178. 22. Iannace, S. and Nicolais, L. “Isothermal crystallization and chain mobility of poly(L-lactide).” Journal Applied Polymer Science 64 (1997): 911–919. 23. Ohshima, M., Shikuma, H., E., Kihara, S., and Taki, K. “Supercritical CO2 and polymer systems—Experiments and prediction of properties of polymer/ CO2 mixtures.” PPS Americas Regional Meeting, Quebec, 2005. 24. Liao, X., Nawaby, A. V., Whitfield, P., Day, M., Champagne, M., and Denault, J. “Layered open pore poly(L-lactic acid) nanomorphology.” Biomacromolecules 7 (2006): 2937–2941. 25. Mathieu, L. M., Montjovent, M.-O., Bourban, P.-E., Pioletti, D. P., and Månson, J.-A. E. “Bioresorbable composites prepared by supercritical fluid foaming.” Journal Biomedical Materials Research 75A (2005): 89–97. 26. Ema, Y., Ikeya, M., and Okamoto, M. “Foam processing and cellular structure of polylactide-based nanocomposites.” Polymer 47 (2006): 5350–5359. 27. Di, Y., Iannace, S., Di Maio, E., and Nicolais, L. “Reactively modified poly(lactic acid): properties and foam processing.” Macromolecular Materials Engineering 290 (2005): 1083–1090. 28. Vega, J. Aguillar, M., Peòn, J., Pastor, D., and Martìnez-Salazar, J. “Effect of long chain branching on linear-viscoelastic melt properties of polyolefins.” E-polymer 2000, no. 046. 29. Narkis, M. and Wallerstein, R. “Crosslinking of polycaprolactone with peroxides.” Polymer Communications 27 (1986): 314–317. 30. Chu, C. C. and Campbell, N. D. “Scanning electron microscopic study of the hydrolytic degradation of poly(glycolic acid) suture.” Biomedical Material Research 16 (1982): 417–430. 31. Pittman, C. U., Iqbal, M., and Chen, C. Y. “Radiation degradation of poly(yhydroxyisobutyric acid) and poly(glycolic ester).” Journal of Polymer Science, Polymer Chemistry Edition 16 (1978): 2721–2724.

61259_C005.indd 170

10/25/2008 12:37:33 PM

Foaming Analysis of PCL and PLA

171

32. Gupta, M. C. and Deshmukh, V. G. “Radiation effects on poly (lactic acid).” Polymer 24 (1983): 827–830. 33. Gandhi, K., Kriz, D., Salovey, R., Narkis, M., and Wallerstein, R. “Crosslinking of polycaprolactone in the pre-gelation region.” Polymer Engineering Science 28 (1988): 1484–1490. 34. Di Maio, E., Iannace, S., Marrazzo, C., Narkis, M., and Nicolais, L. “Effect of molecular modification on PCL foam formation and morphology of PCL.” Macromolecular Symposia 228 (2005): 219–227. 35. Wood-Adams, P. M. and Dealy, J. M. “Using rheological data to determine the branching level in metallocene polyethylenes.” Macromolecules 33 (2000): 7481–7488. 36. Winter, H. H. and Mours, M. “Rheology of polymers near liquid-solid transitions.” Advances in Polymer Science 134 (1997): 166–234. 37. Naguib, H. E., Park, C. B., Panzer, U., and Reichelt, N. “Strategies for achieving ultra low-density polypropylene foams.” Polymer Engineering Science 42 (2002): 1481–1492. 38. Carlson, D., Dubois, P., Nie, L., and Narayan, R. “Free radical branching of polylactide by reactive extrusion.” Polymer Engineering Science 38 (1998): 311–321. 39. Carlson, D. Nie, L., Narayan, R., and Dubois, P. “Maleation of polylactide (PLA) by reactive extrusion.” Journal of Applied Polymer Science 72 (1999): 477–485. 40. Sodergard, A., Niemi, M., Selin, J. F., and Nasman, J. H. “Changes in peroxide melt-modified poly(L-lactide).” Industrial and Engineering Chemical Research 34 (1995): 1203–1207. 41. Park, E., Cho, H., Kim, M., and Yoon, J. “Chain extension and mechanical properties of unsaturated aliphatic copolyesters based on poly(L-lactic acid).” Journal of Applied Polymer Science 90 (2003): 1802–1807. 42. Shinno, K., Miyamoto, M., and Rimura, Y. “Solid-state postpolymerization of L-lactide promoted by crystallization of product polymer: an effective method for reduction of remaining monomer.” Macromolecules 30 (1997): 6438–6444. 43. Kylma, J., Tuominen, J., Helminen, A., and Seppala, J. “Chain extending of lactic acid oligomers. Effect of 2,2′-bis(2-oxazoline) on 1,6-hexamethylene diisocyanate linking reaction.” Polymer 42 (2001): 3333–3343. 44. Zhong, W., Ge, J., Gu, Z., Li, W., Chen, X., Zang, Y., and Yang, Y. “Study on biodegradable polymer materials based on poly(lactic acid). I. Chain extending of low molecular weight poly(lactic acid) with methylenediphenyl diisocyanate.” Journal of Applied Polymer Science 74 (1999): 2546–2551. 45. Hiltunen, K., Seppala, J. V., and Harkonen, M. “Lactic acid based poly(esterurethanes): Use of hydroxyl terminated prepolymer in urethane synthesis.” Journal of Applied Polymer Science 63 (1997): 1091–1100. 46. Tuominen, J., Kylma, J., and Seppala, J. “Chain extending of lactic acid oligomers. 2. Increase of molecular weight with 1,6-hexamethylene diisocyanate and 2,2′-bis(2-oxazoline).” Polymer 43 (2002): 3–10. 47. Vollalobos, M. A., Awojulu, A., Greeley, T., Turco, G., and Deeter, G. “Oligomeric chain extenders for economic reprocessing and recycling of condensation plastics.” Energy 31 (2006): 3227–3234. 48. Blasius, W. G., Deeter, G. A., and Villalobos, M. A. WO Patent 2003 066704. 49. Ano, Y. T., Ueda, K., Matsuoka, F., and Matsumoto, T. US Patent 2006 276617. 50. Marrazzo, C., Di Maio, E., and Iannace, S. “Foaming of synthetic and natural biodegradable polymers.” Journal Cellular Plastics 43 (2007): 123.

61259_C005.indd 171

10/25/2008 12:37:33 PM

172

Polymeric Foams

51. Mitsunaga, M., Ito, Y., Ray, S. S., Okamoto, M., and Hironaka, K. “Intercalated polycarbonate/clay nanocomposites: nanostructure control and foam processing” Macromolecular Materials and Engineering 288 (2003): 543–546. 52. Usuki, A., Kojima, Y., Kawasumi, M., Okada, A., Fukushima, Y., Kurauchi, T., and Kamigaito, O. “Synthesis of nylon 6-clay hybrid.” Journal of Materials Research 8 (1993): 1174–1184. 53. Sinha Ray, S. and Bousmina, M. “Biodegradable polymers and their layered silicate nanocomposites: In greening the 21st century materials world.” Progress in Materials Science 50 (2005): 962–1079. 54. Okamoto, M. “Biodegradable polymer/layered silicate nanocomposites: A review.” In Handbook of Biodegradable Polymeric Materials and their Applications, ed. B. Narasimhan and S. K. Mallapragada. American Scientific Publishers, Stevenson Ranch, 2005. 55. Sinha Ray, S. and Okamoto, M. “Polymer/layered silicate nanocomposites: a review from preparation to processing.” Progress in Polymer Science 28 (2003): 1539–1641. 56. Paul, M.-A., Alexandre, M., Degée, P., Calberg, C., Jérôme, R., and Dubois, P. “Exfoliated polylactide/clay nanocomposites by in-situ coordination-insertion polymerization.” Macromolecular Rapid Communications 24 (2003): 561–566. 57. Hiroi, R., Sinha Ray, S., Okamoto, M., and Shiroi, T. “Organically modified layered titanate: a new nanofiller to improve the performance of biodegradable polylactide.” Macromolecular Rapid Communications 25 (2004): 1359–1364. 58. Sinha Ray, S., Yamada, K., Okamoto, M., Ogami, A., and Ueda, K. “New polylactide/layered silicate nanocomposites. 5. Designing of materials with desired properties.” Polymer 44 (2003): 6633–6646. 59. Hao, J., Yuan, M., and Deng X. “Biodegradable and biocompatible nanocomposites of poly(ε-caprolactone) with hydroxyapatite nanocrystals: thermal and mechanical properties.” Journal of Applied Polymer Science 86 (2002): 676–683. 60. Sinha Ray, S., Yamada, K., Ogami, A., Okamoto, M., and Ueda K. “New polylactide/layered silicate nanocomposite: nanoscale control over multiple properties.” Macromolecular Rapid Communications 23 (2002): 943–947. 61. Maiti, P., Yamada, K., Okamoto, M., Ueda, K., and Okamoto, K. “New polylactide/layered silicate nanocomposites: role of organoclays.” Chemistry of Materials 14 (2002): 4654–4661. 62. Jimenez, G., Ogata, N., Kawai, H., and Ogihara, T. “Structure and thermal/ mechanical properties of poly (ε-caprolactone)–clay blend.” Journal of Applied Polymer Sciences 64 (1997): 2211–2220. 63. Ogata, N., Jimenez, G., Kawai, H., and Ogihara, T. “Structure and thermal/ mechanical properties of poly(l-lactide)-clay blend.” Journal of Polymer Science B 35 (1997): 389–396. 64. Lepoittevin, B., Pantoustier, N., Devalckenaere, M., Alexandre, M., Kubies, D., Calberg, C., Jérôme, R., and Dubois, P. “Poly(ε-caprolactone)/clay nanocomposites by in-situ intercalative polymerization catalyzed by dibutyltin dimethoxide.” Macromolecules 35 (2002): 8385–8390. 65. Krikorian, V. and Pochan, D. “Crystallization behavior of poly(L-lactic acid) nanocomposites: nucleation and growth probed by infrared spectroscopy.” Macromolecules 38 (2005): 6520–6527.

61259_C005.indd 172

10/25/2008 12:37:33 PM

Foaming Analysis of PCL and PLA

173

66. Lepoittevin, B., Devalckenaere, M., Pantoustier, N., Alexandre, M., Kubies, D., Calberg, C., Jerome, R., and Dubois, P. “Poly(ε-caprolactone)/clay nanocomposites prepared by melt intercalation: mechanical, thermal and rheological properties.” Polymer 43 (2002): 4017–4023. 67. Krishnamoorti, R. and Giannelis, E. P. “Rheology of end-tethered polymer layered silicate nanocomposites.” Macromolecules 30 (1997): 4097–4102. 68. Hoffmann, B., Dietrich, C., Thomann, R., Friedrich, C., and Muelhaupt, R. “Morphology and rheology of polystyrene nanocomposites based upon organoclay.” Macromolecular Rapid Communications 21 (2000): 57–61. 69. Shenoy, A. V. In Rheology of Filled Polymer Systems. Kluwer Academic, New Dehli, 1999. 70. Sinha Ray, S. and Okamoto, M. “Biodegradable polylactide and its nanocomposites: opening a new dimension for plastics and composites.” Macromolecular Rapid Communications 24 (2003): 815–840. 71. Di, Y., Iannace, S., Di Maio, E., and Nicolais, L. “Poly(lactic acid)/organoclay nanocomposites: thermal, rheological properties and foam processing.” Journal of Polymer Science Part B: Polymer Physics 43 (2005): 689–698. 72. Pluta, M. “Melt compounding of polylactide/organoclay: Structure and properties of nanocomposites.” Journal of Polymer Science Part B: Polymer Physics 44 (2006): 3392–3405. 73. Kojima, Y. Usuki, A. Kawasumi, M. Okada, A., Fukushima, Y., Kurauchi, T. T., and Kamigaito, O. “Synthesis of nylon 6-clay hybrid.” Journal of Materials Research 8 (1993): 1179–1184. 74. Kojima, Y., Usuki, A., Kawasumi, M., Okada, A., Kurauchi, T. T., and Kamigaito, O. “Synthesis of nylon 6-clay hybrid by montmorillonite intercalated with ε-caprolactam.” Journal of Polymer Science Part A: Polymer Chemistry 31 (1993): 983–986. 75. Usuki, A., Kawasumi, M., Kojima, Y., Okada, A., Kurauchi, T., and Kamigaito, O. “Mechanical properties of nylon 6-clay hybrid.” Journal of Materials Research 8 (1993): 1185–1189. 76. Yano, K., Usuki, A., Okada, A., Kurauchi, T., and Kamigaito, O. “Synthesis a properties of polyimide–clay hybrid.” Journal of Polymer Science Part A: Polymer Chemistry 31 (1993): 2493–2498. 77. Messersmith, P. B. and Giannelis, E. P. “Synthesis and characterization of layered silicate-epoxy nanocomposites.” Chemistry of Materials 6 (1994): 1719–1725. 78. Xu, R. Manias, E., Snyder, A. J., and Runt, J. “New biomedical poly(urethane urea)-layered silicate nanocomposites.” Macromolecules 34 (2001): 337–339. 79. Di, Y., Iannace, S., Sanguigno, L., and Nicolais, L. “Barrier and mechanical proprerties of poly(caprolactone)/organoclay nanocomposites.” Macromolecular Symposia 228 (2005): 115–124. 80. Sinha Ray, S., Yamada, K., Okamoto, M., Ogami, A., and Ueda, K. “New polylactide/layered silicate nanocomposites. 3. High-performance biodegradable materials.” Chemistry of Materials 15 (2003): 1456–1465. 81. Cotugno, S. “Modeling of thermodynamic and transport properties of polymer melts and solutions to be used in the simulation of foaming processes.” PhD thesis, University of Naples Federico II, Italy, 2003. 82. Gopakumar, T. G., Lee, J. A., Kontopoulou, M., and Parent, J. S. “Influence of clay exfoliation on the physical properties of montmorillonite/polyethylene composites.” Polymer 43 (2002): 5483–5491.

61259_C005.indd 173

10/25/2008 12:37:34 PM

174

Polymeric Foams

83. Liu, X. and Wu, Q. “Non-isothermal crystallization behaviors of polyamide 6/clay nanocomposites.” European Polymer Journal 38 (2002): 1383–1389. 84. Liu, X., Wu, Q., and Berglund, L. A. “Polymorphism in polyamide 66/clay nanocomposites.” Polymer 43 (2002): 4967–4972. 85. Wu, Z., Chixin, Z., and Zhu, N. “The nucleating effect of montmorillonite on crystallization of nylon 1212/montmorillonite nanocomposite.” Polymer Testing 21 (2002): 479–483. 86. Li, J., Zhou, C., and Gang, W. “Study on nonisothermal crystallization of maleic anhydride grafted polypropylene/montmorillonite nanocomposite.” Polymer Testing 22 (2003): 217–223. 87. Liu, X. and Wu, Q. “PP/clay nanocomposites prepared by grafting-melt intercalation.” Polymer 42 (2001): 10013–10019. 88. Tseng, C.-R., Wu, J.-Y., Lee, H.-Y., and Chang, F.-C. “Preparation and crystallization behavior of syndiotactic polystyrene–clay nanocomposites.” Polymer 42 (2001): 10063–10070. 89. Taki, K., Yanagimoto, T., Funami, E., Okamoto, M., and Ohshima, M. “Visual observation of CO2 foaming of polypropylene-clay nanocomposites.” Polymer Engineering Science 44 (2004): 1004–1011. 90. Okamoto, M., Nam, P. H., Maiti, P., Kotaka, T., Nakayama, T., Takada, M., Ohshima, M., Usuki, A., Hasegawa, N., and Okamoto, H. “Biaxial flow-induced alignment of silicate layers in polypropylene/clay nanocomposite foam.” Nano Letters 1 (2001): 503–505. 91. Nam, P. H., Maiti, P., Okamoto, M., Kotaka, T., Nakayama, T., Takada, M. Ohshima, M., Usuki, A., Hasegawa, Ns., and Okamoto, H. “Foam processing and cellular structure of polypropylene/clay nanocomposites.” Polymer Engineering Science 42 (2002): 1907–1918. 92. Han, X., Zeng, C., Lee, L. J., Koelling, K. W., and Tomasko, D. L. “Extrusion of polystyrene nanocomposite foams with supercritical CO2.” Polymer Engineering Science 43 (2003): 1261–1275. 93. Tatibouët, J., Gendron, R., Hamel, A., and Sahnoune, A. “Effect of different nucleating agents on the degassing conditions as measured by ultrasonic sensors.” Journal of Cellular Plastics 38 (2002): 203–218. 94. Shen, J., Zeng, C., and Lee, L. J. “Synthesis of polystyrene–carbon nanofibers nanocomposite foams.” Polymer 46 (2005): 5218–5224. 95. Zeng, C., Han, X., Lee, L. J., Koelling, K. W., and Tomasko, D. L. “Polymerclay nanocomposite foams prepared using carbon dioxide.” Advanced Materials 15 (2003): 1743–1747. 96. Karbas, H., Nelson, P., Yuan, M., Gong, S., and Turng, L.-S. “Effects of nanofillers and process conditions on the microstructure and mechanical properties of microcellular injection molded polyamide nanocomposites.” Polymer Composites 24 (2003): 655–671. 97. Cao, X., Lee, L. J., Widya, T., and Macosko, C. “Polyurethane/clay nanocomposites foams: processing, structure and properties.” Polymer 46 (2005): 775–783. 98. Lee, Y. H., Park, C. B., Wang, K. H., and Lee, M. H. “HDPE-clay nanocomposite foams blown with supercritical CO2.” Journal of Cellular Plastics 41 (2005): 487–502. 99. Marrazzo, C., Di Maio, E., and Iannace, S. “Conventional and nanometric nucleating agents in PCL foaming: crystals vs. bubbles nucleation.” Polymer Engineering Science 48 (2008): 336–344.

61259_C005.indd 174

10/25/2008 12:37:34 PM

6 Nanostructure Development and Foam Processing in Polymer/Layered Silicate Nanocomposites Masami Okamoto

CONTENTS 6.1 Introduction .................................................................................... 6.2 Nanostructure Development ...................................................... 6.2.1 Melt Intercalation ................................................................. 6.2.2 Interlayer Structure of OMLFs and Intercalation .......... 6.2.2.1 Nanofillers .............................................................. 6.2.2.2 Molecular Dimensions and Interlayer Structure ............................................... 6.2.2.3 Correlation of Intercalant Structure and Interlayer Opening ........................................ 6.2.2.4 Nanocomposite Structure ..................................... 6.3 Flow-Induced Structure Development ...................................... 6.3.1 Elongational Flow and Strain-Induced Hardening ....... 6.4 Foam Processing ........................................................................... 6.4.1 Foam Processing of PP-Based Nanocomposites ............ 6.4.2 In-Situ Observation of Foaming ....................................... 6.4.3 PLA-Based Nanocomposite Foaming ............................. 6.4.4 Foaming Temperature Dependence of Cellular Structure ............................................................... 6.4.5 CO2 Pressure Dependence ................................................ 6.4.6 TEM Observation ...............................................................

61259_C006.indd 175

176 177 177 177 177 178 182 183 188 188 190 190 196 200 201 204 208

10/25/2008 12:38:06 PM

176

Polymeric Foams

6.4.7 Polycarbonate (PC)-Based Nanocomposite Foams ........ 6.4.8 Mechanical Properties of Nanocomposite Foams ............ 6.4.9 Porous Ceramic Materials via Nanocomposite ................. 6.5 Conclusions and Future Prospects .............................................. References ................................................................................................

209 211 214 215 216

6.1 Introduction A decade of research has shown that nanostructured materials have the potential to significantly impact growth at every level of the world economy in the twenty-first century. This new class of materials is now being introduced in structural applications including gas barrier films, flame retardant products, and other load-bearing applications. Of particular interest are recently developed nanocomposites consisting of a polymer and layered silicate, which often exhibit remarkably improved properties1 when compared with polymer or conventional composites (both micro- and macro-composites). In polymer/layered silicate nanocomposites, a nylon 6/layered silicate hybrid2 reported by Toyota Central Research & Development Co. Inc. was successfully prepared by in-situ polymerization of ε-caprolactam in a dispersion of montmorillonite (MMT). The silicate can be dispersed in liquid monomer or a solution of monomer. It has also been possible to melt-mix polymers with layered silicates, avoiding the use of organic solvents. This method permits the use of conventional processing techniques such as injection molding and extrusion. The extensive literature on nanocomposite research are covered in recent reviews.1,3,4 Continued progress in nanoscale controlling, as well as an improved understanding of the physico-chemical phenomena at the nanometer scale, have contributed to the rapid development of novel nanocomposites. This chapter presents current research on polymer/layered silicate nanocomposites (PLSNCs) with the primary focus on nanostructure development and foam processing operations. Development of nanocomposite foams is one of the latest evolutionary technologies of the polymeric foam following a pioneering effort by Okamoto and his colleagues.5,6 They prepared polypropylene (PP)/layered silicate, poly(l-lactide) (PLA)/layered silicate and polycarbonate (PC)/layered silicate nanocomposites foams in a batch process using supercritical CO2 as a physical foaming agent.6–8 To innovate on the material properties of nanocomposite foams, one needs to pin down the morphological correlation between the dispersed silicate particles with nanometer dimensions in the bulk and the formed closedcellular structure after foaming. This chapter is devoted to the study of evaluation of the performance potential of PLSNCs in foam application.

61259_C006.indd 176

10/25/2008 12:38:07 PM

Nanostructure Development and Foam Processing

177

6.2 Nanostructure Development 6.2.1 Melt Intercalation Since the possibility of direct melt intercalation was first demonstrated,9 melt intercalation has become a preparation of the intercalated polymer/layered silicate nanocomposites (PLSNCs). This process involves annealing, statically or under shear, a mixture of the polymer and organically modified layered fillers (OMLFs) above the softening point of the polymer. During annealing, the polymer chains diffuse from the bulk polymer melt into the nanogalleries between the layered fillers. In order to understand the thermodynamic issue associated with the nanocomposite formation, Vaia et al. have applied a mean-field statistical lattice model and found conclusions based on the mean field theory agreed with the experimental results.10,11 The entropy loss associated with confinement of a polymer melt is not prohibited to nanocomposite formation because an entropy gain associated with the layer separation balances the entropy loss of polymer intercalation, resulting in a net entropy change near to zero. Thus, from the theoretical model, the outcome of nanocomposite formation via polymer melt intercalation depends on energetic factors, which may be determined from the surface energies of the polymer and OMLF. Nevertheless, we have often faced the problem where the nanocomposite shows fine and homogeneous distribution of the nanoparticles in the polymer matrix [e.g. poly(l-lactide)] without a clear peak shift of the mean interlayer spacing of the (001) plane, as revealed by wide-angle X-ray diffraction (WAXD) analysis.12 Furthermore we sometimes encounter a decrease in interlayer spacing compared with that of pristine OMLF, despite very fine dispersion of the silicate particles. For this reason, information on the structure of the surfactant (intercalant)–polymer interface is necessary to understand the intercalation kinetics that can predict final nanocomposite morphology and overall material properties.

6.2.2 Interlayer Structure of OMLFs and Intercalation 6.2.2.1 Nanofillers In characterizing layered silicate, including layered titanate (HTO), the surface charge density is particularly important because it determines the interlayer structure of intercalants as well as cation exchange capacity (CEC). Lagaly proposed a method consisting of total elemental analysis and the dimension of the unit cell:13  e  __ surface charge: ____ nm2 ab

61259_C006.indd 177

(6.1)

10/25/2008 12:38:07 PM

178

Polymeric Foams

TABLE 6.1 Characteristic Parameters of Nanofillers Parameters

syn-FH

HTO

MMT

Chemical formula

H1.07Ti1.73O3.95·0.5H2O

Na0.66Mg2.6Si4O10(F)2

Particle size (nm) BET area (m2/g) CEC* (meq/100 g)

~100–200 ~2400 ~200 (660) 1.26

~100–200 ~800 ~120 (170) 0.971

Na0.33(Al1.67Mg0.33) Si4O10(OH)2 ~100–200 ~700 ~90 (90) 0.708

2.40 2.3

2.50 1.55

2.50 1.55

4–6

9–11

7.5–10

e (charge/nm2) Density (g/cm3) Refractive index (n20D) pH

*Methylene blue adsorption method. The values in the parenthesis are calculated from chemical formula of nanofillers. Source: From Yoshida, O. and Okamoto, M. Macromolecular Rapid Communications 27 (2006): 751–757. © 2006 Wiley-VCH. With permission.

where  is the layer charge [1.07 for HTO, 0.66 for synthetic fluorine hectrite (syn-FH), and 0.33 for montmorillonite (MMT); a and b are cell parameters of HTO (a  3.782 Å, b  2.978 Å),14 syn-FH (a  5.24 Å, b  9.08 Å),15 and MMT (a  5.18 Å, b  9.00 Å)].16 For syn-FH, however, about 30% of the interlayer Na ions are not replaced quantitatively by intercalants due to the non-activity for ion-exchange reactions.15 For HTO, only 27% of interlayer H (H3O) is active for ion-exchange reactions.12 The remaining part is the non-active site in the HTO. Thus the incomplete replacement of the interlayer ions is ascribed to the intrinsic chemical reactivity. The characteristic parameters of three nanofillers are summarized in Table 6.1.16 HTO has a high surface charge density of 1.26 e/nm2 compared with those of syn-FH (0.971 e/nm2) and MMT (0.780 e/nm2). From these results, we can estimate the average distance between exchange sites, which is calculated to be 0.888 nm for HTO, 1.014 nm for syn-FH, and 1.188 nm for MMT. This estimation assumes that the cations are evenly distributed in a cubic array over the silicate surface and that half of the cations are located on one side of the platelet and the other half reside on the other side. 6.2.2.2 Molecular Dimensions and Interlayer Structure The calculated models of the intercalant structures are presented in Figure 6.1. For octadecyl ammonium (C18H3N), obtained molecular length, thickness, and width are 2.466 nm, 0.301 nm, and 0.301 nm, respectively. Since the length of all alkyl units are more than 2 nm, these spacings (distance between exchange sites) of 0.888–1.188 nm do not allow parallel

61259_C006.indd 178

10/25/2008 12:38:07 PM

179

Nanostructure Development and Foam Processing

(a)

(b)

(c)

(d)

Intercalant

C18H3N+

C18(CH3)3N+

2C18(CH3)2N+

qC14(OH)

Length (nm)

2.466

2.601

4.766

2.090

Thickness (nm)

0.301

0.372

0.434

0.374

Width (nm)

0.301

0.372

0.318

0.881

FIGURE 6.1 Molecular dimensions of intercalans: (a) octadecyl ammonium [C18H3N]; (b) octadecyl trimethyl ammonium [C18(CH3)3N]; (c) dioctadecyl dimethyl ammonium [2C18(CH3)2N]; and (d) N-(coco alkyl)-N,N-[bis(2-hydroxyethyl)]-N-methyl ammonium [qC14(OH)] cations. [From: Yoshida, O. and Okamoto, M. Macromolecular Rapid Communications 27 (2006): 751–757. © 2006 Wiley-VCH. With permission.]

layer arrangement like flat-lying chains13 in each gallery space of the nanofillers. All of the intercalants are oriented with some inclination to the host layer in the interlayer space to form an interdigitated layer. This is suggested by the paraffin-type layer structure proposed by Lagaly, especially in the case of highly surface-charged clay minerals.13

61259_C006.indd 179

10/25/2008 12:38:07 PM

180

Polymeric Foams

Wide-angle X-ray diffraction (WAXD) patterns for three OMLF powders are presented in Figure 6.2. The mean interlayer spacing of the (001) plane (d(001)) for the HTO intercalated with qC14(OH) [HTO-qC14(OH)] obtained by WAXD measurements is 2.264 nm (diffraction angle, 2Q  3.90°). The appearances of small peaks observed at 2Q  7.78°, 11.78°, and 15.74° confirmed that these reflections were due to (002) up to (004) plane of HTO-qC14(OH). HTO-qC14(OH) showed a surprisingly well-ordered suprastructure, as demonstrated by WAXD with diffraction maxima up to the fourth order, due to the high surface charge density of the HTO layers. On the other hand, syn-FH and MMT, which have low surface charge density compared with that of HTO, show a less-ordered interlayer structure; that is, the coherent order of the silicate layers is much lower in each syn-FH and MMT intercalated with surfactants.

8000 d = 2.264 nm

HTO-qC14(OH)

(001)

6000 4000

d = 1.135 nm d = 0.751 nm (002) d = 0.563 nm (003) (004)

2000 0 Intensity (a.u.)

syn-FH-qC14(OH) d = 2.063 nm

6000

(001) 4000 d = 1.027 nm 2000

(002)

0 MMT-qC14(OH) 6000 4000 d = 1.855 nm d = 0.930 nm (001) (002)

2000 0

0

5

10

15

20

2Θ (°) FIGURE 6.2 WAXD patterns of HTO, syn-FH, and MMT intercalated with qC14(OH)+. [From: Yoshida, O. and Okamoto, M. Macromolecular Rapid Communications 27 (2006): 751–757. © 2006 Wiley-VCH. With permission.]

61259_C006.indd 180

10/25/2008 12:38:07 PM

181

Nanostructure Development and Foam Processing

From WAXD results, we can discuss the interlayer opening that is estimated after subtraction of the layer thickness value of 0.374 nm for HTO,14 0.98 nm for syn-FH15 and 0.96 nm for MMT.13 This is an important point for the following discussion of the interlayer structure. The illustration of a model of interlayer structure of the qC14(OH) in the gallery space of the HTO is shown in Figure 6.3. For nanofillers with high surface charge density, the intercalants can adopt a configuration with orientation where the alkyl chains are tilted under the effect of van der Waals forces, which decreases the chain–chain distance. For this reason, the angle α should be directly related to the packing density of the alkyl chains. The value of α decreases until close contact between the chains is attained, giving an increase in the degree of the crystallinity of the intercalants in the nanogalleries. To estimate the tilt angle α, we combined the molecular dimension, interlayer spacing, and loading amount of intercalant in the layers, which was calculated from thermogravimetry analysis (TGA). The characteristic parameters are summarized in Tables 6.2 and 6.3. Note that HTO exhibits a large layer opening accompanied with large values of α and endothermic heat flow (H) owing to the melting of the intercalants in the galleries when compared with those of syn-FH and MMT. This indicates that HTO leads to a highly interdigitated layer structure and the interlayer HTO-qC14(OH) Ti-O

N+

–

–

OH

OH N+

CH3

CH3

1.889 nm

CH3

N+

HO –

CH3

N+

HO –

α Ti-O

0.888 nm (0.794 nm2/charge) FIGURE 6.3 Illustration of a model of interlayer structure of intercalant N-(coco alkyl)N,N-[bis(2-hydroxyethyl)]-N-methyl ammonium [qC14(OH)] cation in gallery space of layer titanate (HTO). The average distance between exchange sites is 0.888 nm, calculated by surface charge density of 1.26 e/nm2. For qC14(OH), obtained molecular length, thickness, and width are 2.09 nm, 0.881 nm, and 0.374 nm, respectively (see Figure 6.1). The tilt angle  of the intercalants can be estimated by the combination of the interlayer spacing, molecular dimensions, and loading amount of intercalants when the alkyl chains adopt an all-trans conformation. [From: Yoshida, O. and Okamoto, M. Macromolecular Rapid Communications 27 (2006): 751–757. © 2006 Wiley-VCH. With permission.]

61259_C006.indd 181

10/25/2008 12:38:08 PM

182

Polymeric Foams

TABLE 6.2 Comparison of Characteristic Parameters between HTO, syn-FH, and MMT Prepared with qC14(OH) HTO-qC14(OH) Layer opening (nm) Tilt angle  (°) Organic content (wt%) Tm* (°C) Ha (J/g)

1.889 64.4 39.6 108.3 214.5

syn-FH-qC14(OH) 1.083 31.1 30.4 111.3 141.2

MMT-qC14(OH) 0.895 25.3 32.5 97.7 138.6

*The melting and heat flow of qC14(OH)Cl are 35.8°C and 69.8 J/g, respectively. Source: From Yoshida, O. and Okamoto, M. Macromolecular Rapid Communications 27 (2006): 751–757. © 2006 Wiley-VCH. With permission.

TABLE 6.3 Comparison of Characteristic Parameters MMT-Based OMLF Prepared with C18H3N⫹, C18(CH3)3N⫹, and 2C18(CH3)2N⫹ C18H3N⫹ Layer opening (nm) Tilt angle  (°) Organic content (wt%) Tm* (°C) H* (J/g)

1.350 33.2 35.5 69.9 177.7

C18(CH3)3N⫹

2C18(CH3)2N⫹

1.011 22.9 29.5 69.5 189.6

1.540 40.1 39.8 44.0 129.7

*The melting and heat flow of C18H3N, C18(CH3)3N, and 2C18(CH3)2N are 83.8°C and 95.6 J/g; 103.5°C and 161.2 J/g; and 37.0°C and 54.6 J/g, respectively. Source: From Yoshida, O. and Okamoto, M. Macromolecular Rapid Communications 27 (2006): 751–757. © 2006 Wiley-VCH. With permission.

opening becomes more uniform compared with MMT and syn-FH (possessing lower surface charge density). From this fact, we can observe well-defined diffraction peaks up to the (004) plane (see Figure 6.2). The entropic contribution of the intercalants, which leads to the entropy gain associated with the layer expansion after intercalation of the polymer chains, may not be significant because of the interdigitated layer structure. 6.2.2.3 Correlation of Intercalant Structure and Interlayer Opening For the interdigitated layer structure in MMT, alkyl chain length [i.e. C18H37, CH3 and (CH2)2OH in the amine structure] changes the interlayer opening. That is, when we compare different intercalants having the same long alkyl chain [i.e. C18H3N and C18(CH3)3N], three methyl (CH3) substituents instead of hydrogen (H) disturb the contact with silicate surfaces. The value of α decreases until close contact between the ammonium cations

61259_C006.indd 182

10/25/2008 12:38:08 PM

Nanostructure Development and Foam Processing

183

and silicate surfaces is attained, giving a decrease in the interlayer opening ( d(001)) (see Table 6.3 and Figure 6.2). In cases where the intercalant has two long alkyl chains (i.e. 2C18(CH3)2N), the packing density of the alkyl chains is reduced and sterically limited in the nanogalleries. Consequently, MMT-2C18(CH3)2N exhibits large interlayer opening accompanied by low crystallinity of the intercalant (H ~ 130 J/g) compared with MMT-C18H3N and MMT-C18(CH3)3N. Accordingly, we observe a disordered diffraction peak of (001) plane of MMT2C18(CH3)2N in the WAXD analysis (see figure 1 in Reference 17). We have to pay attention to the molecular size of the substituents instead of H attached to the nitrogen for the better understanding of the interdigitated layer structure and direct polymer melt intercalation. This feature has been observed in the results of OMLFs intercalated with various intercalants (such as octadecyl di-methyl benzyl ammonium, n-hexadecyl tri-n-butyl phosphonium, n-hexadecyl tri-phenyl phosphonium cations).18 6.2.2.4 Nanocomposite Structure Figure 6.4 shows the results of TEM bright field images of PLA-based nanocomposites, in which dark entities are the cross-section of intercalated MMT layers. The organically modified MMT content in all nanocomposites was 4 wt%. From the TEM images, it becomes clear that there are some intercalated and stacked silicate layers in the nanocomposites. Yoshida et al. estimated the form factors obtained from TEM images; that is, average value of the particle length (L), of the dispersed particles, and the correlation length () between them.19 From the WAXD patterns, the crystallite size (D) of intercalated stacked silicate layers of each nanocomposite was calculated using the Scherrer equation. The calculated value of D ( thickness of the dispersed particles) and other parameters for each nanocomposite are presented in Table 6.4. For PLA/MMT-C18(CH3)3N, L and D are in the range 200 ± 25 nm and 10.7 nm. On the other hand, PLA/MMT-C18H3N exhibits a large value of L (450 ± 200 nm) with a large level of stacking of the silicate layers (D~ 21 nm). The  value of the PLA/MMT-C18(CH3)3N (80 ± 20 nm) is lower than the value of PLA/MMT-C18H3N (260 ± 140 nm), suggesting that the intercalated layers are more homogeneously and finely dispersed in the case of PLA/MMT-C18(CH3)3N. The number of the stacked individual silicate layers ( D/d(001)  1) is 5 for PLA/MMT-C18(CH3)3N and  value of this nanocomposite is one order of magnitude lower compared with those of PLA/MMT-C18H3N and PLA/MMT-2C18(CH3)2N, suggesting that intercalated silicate layers are more homogeneously and finely dispersed. Although the (initial) interlayer opening of MMT-C18(CH3)3N at 1.011 nm is smaller than MMT-C18H3N at 1.350 nm and MMT-2C18(CH3)2N at 1.540 nm, the intercalation of the PLA in these different OMLFs gives almost the same basal spacing after preparation of the nanocomposites. Note that the existence of a sharp Bragg peak in PLA-based nanocomposites

61259_C006.indd 183

10/25/2008 12:38:08 PM

184

Polymeric Foams

(a)

500 nm (b)

500 nm (c)

300 nm FIGURE 6.4 Bright filed TEM images of PLA-based nanocomposites prepared with: (a) MMT-C18H3N+; (b) MMT-C18(CH3)3N+; and (c) MMT-2C18(CH3)2N+. The dark entities are the cross-section and/or face of intercalated-and-stacked silicate layers and the bright areas are the matrix. [From: Yoshida, O. and Okamoto, M. Macromolecular Rapid Communications 27 (2006): 751–757. © 2006 Wiley-VCH. With permission.]

61259_C006.indd 184

10/25/2008 12:38:08 PM

185

Nanostructure Development and Foam Processing

TABLE 6.4 Form Factors of Three Nanocomposites Obtained from WAXD and TEM Observations Nanocomposites d001 (nm)  opening (nm) Final layer opening (nm) D (nm) (D/d001)  1 L (nm)  (nm)

PLA/MMTC18H3N⫹ 3.03 0.72 2.07 20.9 7.9 450 ± 200 260 ± 140

PLA/MMTC18(CH3)3N⫹ 2.85 0.879 1.89 10.73 4.8 200 ± 25 80 ± 20

PLA/MMT2C18(CH3)2N⫹ 2.95 0.45 1.99 14.71 6.0 655 ± 121 300 ± 52

Source: From Yoshida, O. and Okamoto, M. Macromolecular Rapid Communications 27 (2006): 751–757. © 2006 Wiley-VCH. With permission.

after melt extrusion clearly indicates that the dispersed silicate layers still retain an ordered structure after melt extrusion. In Table 6.4 they summarized the layer expansion after preparation (  opening) of three nanocomposites, or after subtraction of the initial layer opening. For the same MMT with different intercalants [e.g. comparison between MMT-C18(CH3)3N and MMT-2C18(CH3)2N], the layer expansion of the former (0.879 nm) exhibits a large value compared with that of the latter (0.45 nm) in PLA-based nanocomposites. In other words, the smaller interlayer opening caused by the configuration with a small tilt angle [  22.9° for C18(CH3)3N] promotes a large amount of intercalation of the polymer chains. Accordingly, PLA/MMT-C18(CH3)3N exhibits finer dispersion of the nanofillers compared with PLA/MMT-2C18(CH3)2N and PLA/MMT-C18H3N as discussed previously (see Figure 6.4). A more interesting feature is the absolute value of  opening. According to the molecular modeling, the width and thickness of the PLA are 0.76 nm and 0.58 nm (see Figure 6.5). This may suggest that the polymer chains could not penetrate into galleries in the case of MMT-2C18(CH3)2N when we compare the apparent interlayer expansion (  opening). Now it is necessary to understand the meaning of the interlayer expansion in the intercalated nanocomposites. As discussed previously, we have to take the interdigitated layer structure into consideration. This structure may suggest that a different orientation angle could be adopted when the polymer chains penetrate into the galleries, giving a decrease in basal spacing after intercalation. At the same time, this structure apparently provides a balance between the polymer penetration and different orientation angle of the intercalants; that is, we have to pay attention to the polymer chain intercalation into the galleries from the result of the change of the basal spacing as revealed by WAXD. Presumably the penetration of the polymer chain is prevented or reduced by the steric limitation of the

61259_C006.indd 185

10/25/2008 12:38:08 PM

186

Polymeric Foams

MD program (MM2 in Quantum CAChe)

PLA-oligomer Width 0.393 nm

3.376

3.933

Thickness 0.338 nm FIGURE 6.5 Molecular dimensions of PLA-backbone using the molecular dynamics program (MM2 in Quantum CAChe) in consideration of van der Waals radii into consideration. Optimization of structure is based on minimization of the total energy of the molecular system.

configuration with a large value of [e.g.   40.1° for MMT-2C18(CH3)2N]. Accordingly, we sometimes observe small interlayer expansion and encounter a decrease in the interlayer spacing after melt intercalation. As seen in Table 6.4, the initial interlayer opening depends on the interlayer expansion (  opening) after melt intercalation. The smaller initial opening leads to the larger interlayer expansion, and gives almost same final interlayer opening. This feature has been observed in the results of other nanocomposites prepared by different OMLFs intercalated with different surfactants.20 From this result, the entropic contribution of the intercalants, which leads to the entropy gain associated with layer expansion after intercalation of the small molecules and/or polymer chains, may not be significant owing to the interdigitated layer structure. Presumably the penetration takes place by pressure drop within the nanogalleries, nanocapillary action, generated by the two platelets. As reported in the literature,18 the pressure drop (p) in the nanogalleries, which makes the polymer penetration more difficult, should be discussed. The estimated pressure difference (~24 MPa) is much larger than the shear stress (~0.1 MPa) during melt compounding.18 This suggests that shear stress has little effect on the delamination (exfoliation) of the layer. This reasoning is consistent with the intercalated structure reported by so many nanocomposite researchers, who can prepare only intercalated (not exfoliated) nanocomposites via the simple melt extrusion technique.1 A novel compounding process is currently in progress. Solid-state shear processing may be an innovative technique to delaminate the layered fillers.21 Compared to OMLFs, the nanocomposite structure is difficult to model using atomic scale molecular dynamics (MD) because the intercalated polymer chain conformation is complex and is rarely in an equilibrium state. However, Pricl et al.22 explored and characterized the atomic scale

61259_C006.indd 186

10/25/2008 12:38:08 PM

Nanostructure Development and Foam Processing

187

FIGURE 6.6 Three-component model used for basal spacing simulations, consisting of two layers of MMT with K cations (stick model), four molecules of trimethyl ammonium cation (a) or dimethyl stearyl ammonium cation (b) (stick and ball model), and one molecule of maleated PP (PP-MA) (ball model). [From Toth, R., Coslanicha, A., Ferronea, M., et al. Polymer 45 (2004): 8075–8083. © 2006 Elsevier Science. With permission.]

structure to predict binding energies and basal spacing of PLSNCs based on polypropylene (PP) and maleated (MA) PP (PP-MA), MMT, and different alkyl ammonium ions as intercalants (see Figure 6.6). From a global interpretation of all of these MD simulation results, they concluded that intercalants with a smaller volume are more effective for clay modification as they improve the thermodynamics of the system by increasing the binding energy. On the other hand, intercalants with longer tails are more effective for intercalation and exfoliation processes, as they lead to higher

61259_C006.indd 187

10/25/2008 12:38:09 PM

188

Polymeric Foams

basal spacing. Additional information is necessary to predict a more reasonable nanostructure of PLSNCs. Some literature related to the confined polymer chains within the silicate galleries by using coarsegrained MD simulation has been published.23–26

6.3 Flow-Induced Structure Development Rheological behavior, especially elongational and shear flow behavior in the molten state of PLSNCs, has not been well studied, although such knowledge should be indispensable in relation to their performance in processing operations. One objective of this chapter is to focus on a profound understanding of PLSNCs for their innovations in practical material production. For this purpose, it is indispensable to illuminate the nanostructure as well as rheological properties of PLSNCs to assess appropriate processing conditions for designing and controlling their hierarchical nanostructure, which must be closely related to their material performance. 6.3.1 Elongational Flow and Strain-Induced Hardening Okamoto et al.27 first conducted an elongation test of PP-based nanocom. posites (PPCN4) under molten state at constant Hencky strain rate, e 0 using an elongation flow opto-rheometry, and attempted to control the alignment of the dispersed MMT layers with nanometer dimensions of intercalated PPCNs under uniaxial elongational flow. Figure 6.7 shows double logarithmic plots of transient elongational vis. cosity E(0; t) against time t observed for a nylon 6/OMLS system (N6CN3.7: MMT  3.7 wt%) and PPCN4 (MMT  4 wt%) with different Hencky . strain rates, 0, ranging from 0.001 s1 to 1.0 s1. The solid curve represents . time development of three-fold shear viscosity, 30( ; t), at 225°C with a . . . constant shear rate  0.001 s1. In E(0; t) at any 0, N6CN3.7 melt shows a weak tendency of strain-induced hardening compared with that of PPCN4 melt. A strong behavior of strain-induced hardening for PPCN4 melt was originated from the perpendicular alignment of the silicate layers to the stretching direction as reported by Okamoto et al.28 From TEM observation,29 the N6CN3.7 forms a fine dispersion of the silicate layers of about 100 nm in Lclay, 3 nm thickness in dclay and clay of about 20–30 nm between them. The clay value is one order of magnitude lower than the value of Lclay, suggesting the formation of a spatially linked-like structure of the dispersed clay particles in the nylon 6 matrix. For N6CN3.7 melt, the silicate layers are densely dispersed into the matrix and hence difficult to align under elongational flow. Under flow fields, the silicate

61259_C006.indd 188

10/25/2008 12:38:10 PM

189

Nanostructure Development and Foam Processing

108 N6CN3.7 106

225°C

ε∗0 / s–1 1.0 0.5 0.1 0.05 0.03 0.01 0.005 0.001

104

η (Pa s)

102

3*η0

(a)

–1

(cone-plate; 0.001s )

100 PPCN4 106

150°C

104

102 (b) 100 –1 10

100

101

102

103

Time (s) . FIGURE 6.7 Time variation of elongational viscosity E(0; t) for: (a) N6CN3.7 melt at 225°C; . and (b) PPCN4 at 150°C. The solid line shows three times the shear viscosity, 3E( ; t), taken . at a low shear rate  0.001 s1 on a cone-plate rheometer. [From Okamoto, M. “Polymer/ layered silicate nanocomposites.” Rapra Review Report No. 163, Rapra Technology Ltd, London, 2003. 166 pp. © 2003 Rapra Technology Ltd. With permission.]

layers might translationally move, but not rotationally in such a way that the loss energy becomes minimum. This tendency was also observed in PPCN7.5 melt having higher content of MMT ( 7.5 wt%).30 One can observe two features for the shear viscosity curve. First, the . . extended Trouton rule, 30( ; t)  E(0; t), does not hold for both N6CN3.7 and PPCN4 melts, as opposed to the melt of ordinary homo-polymers. . . The latter, E(0; t), is more than 10 times larger than the former, 30( ; t). . Second, again unlike ordinary polymer melts, 30( ; t) of N6CN3.7 melt increases continuously with t, never showing a tendency of reaching a steady state within the time span (600 s or longer) examined here. This time-dependent thickening behavior may be called anti-thixotropy or rheopexy. . . Under slow shear flow (  0.001 s1), 30( ; t) of N6CN3.7 exhibits a much stronger rheopexy behavior with almost two orders of magnitude higher than that of PPCN4. This reflects a fact that the shear-induced

61259_C006.indd 189

10/25/2008 12:38:10 PM

190

Polymeric Foams

structural change involved a process with an extremely long relaxation time as well as for other PLSNCs having rheopexy behavior,31 especially under the weak shear field. In uniaxial elongational flow (converging low) for a PPCN4, the formation of a ‘house-of-cards’ structure is found by TEM analysis.27 The perpendicular (but not parallel) alignment of disk-like MMT clay particles with large anisotropy toward the flow direction might sound unlikely but this could be the case, especially under an elongational flow field in which the extentional flow rate is the square of the converging flow rate along the thickness direction, if the assumption of affine deformation without volume change is valid. Obviously under such conditions, the energy dissipation rate due to viscous resistance between the disk surface and the matrix polymer is minimal when the disks are aligned perpendicular to the flow direction. Some 20 years ago, van Olphen32 pointed out that the electrostatic attraction between the layers of natural clay in aqueous suspension arises from higher polar forces in the medium. The intriguing features such as yield stress thixotropy and/or rheopexy exhibited in aqueous suspensions of natural clay minerals may be taken as a reference to the present PLSNCs.

6.4 Foam Processing Flow-induced internal structural change occurs in both shear and elongational flow, but differs in each case, as noted from the above results on . . E(0; t) and 30( ; t) (see Figure 6.7). Thus, with the rheological features of the PLSNCs and the characteristics of each processing operation, tactics should be selected accordingly for a particular nanocomposite for the enhancement of its mechanical properties. . For example, the strong strain-induced hardening in E(0; t) is requisite for withstanding the stretching force during the processing, while the . rheopexy in 30( ; t) suggests that for such PLSNC a promising technology is the processing in confined space (such as injection molding) where shear force is crucial. 6.4.1 Foam Processing of PP-Based Nanocomposites PPCNs have already been shown to exhibit a tendency toward strong strain-induced hardening. On the basis of this result, the first successful nanocomposite foam, processed by using supercritical (sc)-CO2 as a physical foaming agent, appeared through a pioneering effort by Okamoto et al.5,6 A small amount of nanofillers in the polymer matrix serve as

61259_C006.indd 190

10/25/2008 12:38:10 PM

Nanostructure Development and Foam Processing

191

Pressure gauge

Autoclave Band heater Sample

Cooling water jacket

CO2 gas cylinder

FIGURE 6.8 Schematic representation of autoclave set-up. [From Nam, P. H., Okamoto, M., Maiti, P., et al. Polymer Engineering Science 42 (2002): 1907–1918. © 2002 Society of Plastic Engineers. With permission.]

nucleation sites to facilitate the bubble nucleation during foaming. Novel nanocomposite foams based on the combination of new nanofillers and sc-CO2 led to a new class of materials. The process consists of four stages: (1) saturation of CO2 in the sample at desired temperature; (2) cell nucleation when the release of CO2 pressure started (supersaturated CO2); (3) cell growth to an equilibrium size during the release of CO2; and (4) stabilization of cell via cooling of the foamed sample. The autoclave setup used in their study is shown in Figure 6.8. Figure 6.9 represents the scanning electron microscopy (SEM) images of PP-MA and various PPCNs foams conducted at various temperatures under a pressure of 10 MPa. From the SEM images it can be clearly observed that, apart from PPCN4 (MMT  4 wt%) and PPCN7.5 foams prepared at 130.6°C, all exhibit neatly closed-cell structures with cells having 12- or 14-hedron shapes. The formed cells show their faces mostly in pentagons or hexagons, which express the most energetically stable state of polygon cells. They also calculated the distribution function of cell sizes from SEM images as shown in Figure 6.10. From the distribution curve it is clearly seen that PPCN7.5 exhibited a bimodal distribution of cell size, whereas the other samples neatly follow a Gaussian distribution. Another interesting observation from Figure 6.10 is the width of the distribution peaks—the polydispersity of the cell size became narrower with the addition of clay into the matrix (PPCN2 and PPCN4). This behavior may be due to the

61259_C006.indd 191

10/25/2008 12:38:11 PM

192

Polymeric Foams

Clay content (wt %) 7.5

4

2

0

200 μm 130.6°C

134.7°C

139.2°C

143.4°C Temperature (oC)

FIGURE 6.9 SEM images for PP-MA and various PPCNs foamed at different temperature. [From Nam, P. H., Okamoto, M., Maiti, P., et al. Polymer Engineering Science 42 (2002): 1907– 1918. © 2002 Society of Plastic Engineers. With permission.]

heterogenous clay sites possibly acting for cell nucleation and their uniform dispersion in the matrix, which, if present, leads to the high homogeneity in cell size. On the other hand, the cell size of prepared foam gradually decreases with increasing clay content in the PPCNs. This behavior is due the intrinsically high viscosity of the materials with increasing clay loading, which were subjected to foam processing. In contrast, the cell density of the foams behaved in the opposite way. The characteristic parameters of pre- and post-formed samples are listed in Table 6.5. The function for determining cell density (Nc ) in cells/cm3 is defined in the following equation:6 N c = 10 4

3[1 - ( r f /r p )] 4p d 3

(6.2)

On the other hand, the mean cell wall thickness () in m was estimated by the following equation:6 _________

d  d(1/√1 - (rf/r p) - 1)

(6.3)

Figure 6.11 shows the TEM images on the structure of the mono-cell wall (a) and the junction of three cell walls (b) for PPCN4 foamed at 134.7°C. In Figure 6.11a, the dispersed clay particles in the cell wall align along

61259_C006.indd 192

10/25/2008 12:38:11 PM

193

Nanostructure Development and Foam Processing

40 PPCN7.5 134.7°C

30 20 10 0

PPCN4 134.7°C

18

Fraction (%)

12 PPCN2 134.7°C

6 0

PPCN2 134.7°C

12 8 4 0

PP-MA 134.7°C

12 8 4 0

0

20

40

60

80 100 120 140 160 180 200 Cell size (μm)

FIGURE 6.10 Typical example for cell size distribution of foamed PP-MA and PPCNs in experiment at 134.7°C. Average values of d in m and variances d2 in m2 in the Gaussian fit through the data are: 122.1 and 12.1 for PP-MA foam; 95.1 and 9.8 for PPCN2 foam; and 64.4 and 3.1 for PPCN4 foam. [From Nam, P. H., Okamoto, M., Maiti, P., et al. Polymer Engineering Science 42 (2002): 1907–1918. © 2002 Society of Plastic Engineers. With permission.]

the interface between the solid and gas phase. In other words, the clay particles arrange along the boundary of cells. The orientation angle of the dispersed clay particles (versus cell boundary), calculated statistically from TEM photographs, is about 5 ± 3.6°, indicating that plane orientation of the dispersed clay particles to the cell boundary occurred. In a previous paper for PPCN4 melt,27 the perpendicular alignment of the clay particles to stretching or elongating direction was shown, which was the main reason for causing the strain-induced hardening in the uniaxial elongational viscosity. In this foam processing, apparently, a similar structure is formed, probably by a different mechanism. Due to the biaxial flow of material during foam process, the clay particles probably either turned their face (marked

61259_C006.indd 193

10/25/2008 12:38:12 PM

194

Polymeric Foams

TABLE 6.5 Characteristic Parameters of Pre- and Post-Foamed PP and Various PPCNs Sample

Tf (°C)

PP-MA

 (g mL⫺1)

d (m)

Nc (cells mL⫺1 107)

 ( mm)

0.854

PP-MA foam

130.6

0.219

74.4

1.8

11.88

PP-MA foam

134.7

0.114

122.1

0.48

9.07

PP-MA foam

139.2

0.058

155.3

0.25

5.56

PP-MA foam

143.4

0.058

137.3

0.35

6.46

72.5

1.99

10.76

PPCN2

0.881

PPCN2 foam

130.6

0.213

PPCN2 foam

134.7

0.113

95.1

1.01

6.76

PPCN2 foam

139.2

0.058

133.3

0.39

4.62

PPCN2 foam

143.4

0.113

150.3

0.26

10.68

8.41

PPCN4

0.900

PPCN4 foam

130.6

0.423

PPCN4 foam

134.7

0.196

64.4

2.92

PPCN4 foam

139.2

0.193

93.4

0.96

11.98

PPCN4 foam

143.4

0.341

56.1

3.52

15.08

PPCN7.5

0.921

PPCN7.5 foam

130.6

0.473

PPCN7.5 foam

134.7

0.190

35.1

18.35

4.30

PPCN7.5 foam

139.2

0.131

33.9

22.00

2.70

PPCN7.5 foam

143.4

0.266

27.5

34.2

5.11

Source: From Nam, P. H., Okamoto, M., Maiti, P., et al. Polymer Engineering Science 42 (2002): 1907–1918. © 2002 Society of Plastic Engineers. With permission.

with the arrows (A) in Figure 6.11a or fixed face orientation [marked with the arrows (B) in Figure 6.11a] and aligned along the flow direction of materials; that is, along the cell boundary. The interesting point here is that such aligning behavior of the clay particles may help cells to withstand the stretching force from breaking the thin cell wall; in other words, to improve the strength of foam in mechanical properties. The clay particles seem to act as a secondary cloth layer to protect the cells from being destroyed by external forces. How do such unique alignments represent an improvement in mechanical properties? The compression modulus K of the foams are shown in Table 6.6. The K of the PPCN foams appears higher than that of PP-MA foam even though they have the same f level. This may create the improvement of mechanical properties for polymeric foams through polymeric nanocomposites.

61259_C006.indd 194

10/25/2008 12:38:12 PM

195

Nanostructure Development and Foam Processing

Cell boundary (a) (b)

(B) (A)

200 nm

500 nm

Cell boundary

FIGURE 6.11 TEM micrographs for PPCN4 foamed at 134.7°C: (a) mono-cell wall and (b) junction of three contacting cells. [From Okamoto, M., Nam, P. H., Maiti, M., et al. Nano Letters 1 (2001): 503–505. © 2001 American Chemical Society. With permission.]

In Figure 6.11b, besides the alignment of clay particles, we can observe a random dispersion of clay in the central area of the junction (marked with the arrow in Figure 6.11b). Such behavior of clay particles presumably reflects the effect of stagnation flow region of material under the growth of three contacting cells. Figure 6.12 shows the stress–strain curves and the strain recovery behavior of the PP-based nanocomposite (PPCN) foams28 in the compression mode at a constant strain rate of 5% min1. The nanocomposite foams exhibit high modulus compared to neat PP-g-MA foam. The residual strain is 17%

TABLE 6.6 Morphological Parameters and Compression Modulus of PP and PPCN Foams f (g cm⫺3)

d (m)

NC ⴛ 10⫺6 (cell cm⫺3)

PP-MA PPCN2 PPCN4

0.06 0.06 0.12

155.3 133.0 93.4

2.49 3.94 9.64

5.6 4.6 11.9

0.44 1.72 1.95

PPCN7.5

0.13

33.9

220

2.7

2.80

Foam Samples

 (m)

K⬘* (MPa)

* At 25°C. Source: From Okamoto, M., Nam, P. H., Maiti, M., et al. Nano Letters 1 (2001): 503–505. © 2001 American Chemical Society. With permission.

61259_C006.indd 195

10/25/2008 12:38:12 PM

196

Polymeric Foams

1.0 Compression set (%) PP-MA PPCN2 PPCN4 PPCN7.5

Stress (MPa)

0.8

0.6

18 17 29 41

0.4

0.2

0 0

10

20

30

40

50

Strain (%) FIGURE 6.12 Stress–strain curves and strain recovery behavior of the PP-based nanocomposites (PPCNs). [From Okamoto, M. “Polymer/layered silicate nanocomposites.” Rapra Review Report No. 163, Rapra Technology Ltd, London, 2003. 166 pp. © 2003 Rapra Technology Ltd. With permission.]

for PPCN2 (MMT  2 weight-percentage) as well as neat PP foam, providing the excellent strain recovery and the energy dissipation mechanism, probably with the “house-of-cards” structure formation in the cell wall, which enhances the mechanical properties of the nanocomposites like a spruce wood which is close to right-handed helix (see Figure 6.13).33 6.4.2 In-Situ Observation of Foaming To understand the complex mechanism of physical foaming, Taki et al. studied the dynamic behavior of bubble nucleation and growth in the batch foaming of PP-based nanocomposites.34 Employing image-processing techniques, the bubble nucleation and growth rate for different nanocomposites are analyzed from the series micrographs. Together with the solubility and diffusivity of CO2 into the PP matrix, the mechanism of nanocomposite foaming is investigated. Figure 6.14 shows a schematic diagram of the visual observation apparatus for batch physical foaming. It consists of a high-pressure cell, a gas supply line and a pump with a gas cylinder. The high-pressure cell is made of stainless steel and has two sapphire windows on the walls. The C-shape stainless steel is used for a spacer. A signal processing board (DITECT, Japan; HAS-PCI) is installed so as to record a series of micrographs onto an online computer. Figure 6.15 shows the series of micrographs of PP-MA (upper) and PPCN7.5 (lower) foaming at 150°C under a pressure of 13 MPa. The

61259_C006.indd 196

10/25/2008 12:38:14 PM

Nanostructure Development and Foam Processing

197

42 μm μ

Z-Helix

Cellulose fibrils

FIGURE 6.13 X-ray microdiffraction experiment with a 2-μm-thick section of spruce wood embedded in resin. Note the asymmetry of the patterns in the enlargement (far left) which can be used to determine the local orientation of cellulose fibrils in the cell wall (arrows). The arrows are plotted in the right image with the convention that they represent the projection of a vector parallel to the fibrils onto the plane of the cross-section. The picture clearly shows that all cells are right-handed helices. [From Fratzl, P. Current Opinion in Colloid Interface Science 8 (2003): 32–39. © 2003 Elsevier Science. With permission.]

FIGURE 6.14 A schematic diagram of the visual observation apparatus for batch physical foaming. [From Taki, K., Yanagimoto, T., Funami, E., Okamoto, M., and Ohshima, M. Polymer Engineering Science 44 (2004): 1004–1011. © 2004 Society of Plastic Engineers. With permission.]

61259_C006.indd 197

10/25/2008 12:38:15 PM

198

Polymeric Foams

1.2 mm

1.6 mm

PPMA (PP+clay0%)

0s

1s

2s

1.2 mm

1.6 mm

PPCN7.5 (PP+clay7.5%)

0s

1s

2s

FIGURE 6.15 Series of micrographs of foaming: PP-MA (upper), PPCN7.5 (lower). The black dots are bubbles and white part is the polymer matrix. The color of the bubbles in the micrograph appear black because the bubbles reflect the light entering from the oppositeside window of the high-pressure cell. [From Taki, K., Yanagimoto, T., Funami, E., Okamoto, M., and Ohshima, M. Polymer Engineering Science 44 (2004): 1004–1011. © 2004 Society of Plastic Engineers. With permission.]

dynamic behavior of bubble nucleation and growth in the very early stages of foaming can be seen in Figure 6.16. The bubble nucleation rate and the final density of bubbles were highest at PPCN7.5 foaming. Although a distinct difference in bubble nucleation rate as well as in the final bubble density could not be observed between PPCN2 and PPMA foaming, the nucleation rate and the final bubble density increased as the weight fraction of clay increased. Furthermore, the induction time became shorter as the clay content increased. The bubble growth rate is quantified by measuring temporal change in cross-sectional area of each bubble. Figure 6.17 shows the representative growth rate of the bubbles born at the designated time in PP-MA and nanocomposite foaming. Since the change in cross-sectional area of bubbles can be approximated by a linear function of time as mentioned above, the bubble growth observed by micrographs is a mass transfer-controlled process. Therefore, it can be said that the clay content changes the mass transfer rate of CO2 from the matrix polymer to the bubbles. The clay particles decrease the diffusivity of CO2 while keeping the solubility of CO2 in the matrix polymer the same. Owing to the clay-induced diffusivity depression, the

61259_C006.indd 198

10/25/2008 12:38:16 PM

199

Number density of nucleated bubble (1/mm3)

Nanostructure Development and Foam Processing

1200 1000 800 600 400

PPMA PPCN2 PPCN4 PPCN7.5

200 0 2.5

3.0 3.5 4.0 4.5 5.0 5.5 Time elapsed after the pressure release starts (s)

6.0

FIGURE 6.16 Time variation in number density of nucleated bubble of PP-MA and nanocomposite foaming. [From Taki, K., Yanagimoto, T., Funami, E., Okamoto, M., and Ohshima, M. Polymer Engineering Science 44 (2004): 1004–1011. © 2004 Society of Plastic Engineers. With permission.]

Representative average growth rate (μm2/s)

12000

10000 PPMA PPCN2

8000

PPCN4 PPCN7.5 6000

4000

2000

0 2.5

3.0 3.5 4.0 4.5 5.0 Time elapsed after the pressure release starts (s)

5.5

FIGURE 6.17 Representative average growth rates for PP-MA and nanocomposite foaming. [From Taki, K., Yanagimoto, T., Funami, E., Okamoto, M., and Ohshima, M. Polymer Engineering Science 44 (2004): 1004–1011. © 2004 Society of Plastic Engineers. With permission.]

61259_C006.indd 199

10/25/2008 12:38:17 PM

200

Polymeric Foams

increase in clay content depresses the mass transfer of CO2 from the matrix polymer to the bubbles. As a result, the bubble growth rate is decreased. 6.4.3 PLA-Based Nanocomposite Foaming Figure 6.18 shows the typical results of SEM images of the fracture surfaces of the PLA/MMT-ODA and neat PLA without clay foamed at a temperature range of 100°C to 140°C under the different isobaric saturation conditions (14, 21, and 28 MPa).35 All foams exhibit the neat closed-cell structure. We noted here that homogeneous cells were formed in the case of nanocomposite foams, while neat PLA foams show rather non-uniform cell structure having large cell size. The nanocomposite foams show smaller cell size (d) and larger cell density (Nc) compared with neat PLA foam, suggesting that the dispersed silicate particles act as nucleating sites for cell formation.5 For both foam systems, the calculated distribution function of cell size from SEM images are presented in Figure 6.19. The nanocomposite foams nicely obeyed the Gaussian distribution. In the case of PLA/ODA foamed at 150°C under high pressure of 24 MPa, we can see that the width of the distribution peaks, which indicates the dispersity for cell size, became narrow accompanied by finer dispersion of silicate particles. Obviously, with decreasing saturation pressure condition (~140°C and 14 MPa), both foams exhibit large cell size due to the low supply of CO2

PLA/MMT-ODA

PLA

2

PCO (MPa)

28

21

×7500

2 μm

×1000

20 μm

14

100

140

120

140

Tf (°C) FIGURE 6.18 Typical results of SEM images of the fracture surfaces of PLA/MMT-ODA and neat PLA foamed at temperature range of 100°C to 140°C under different isobaric saturation condition (14, 21, and 28 MPa). [Reprinted from Ema, Y., Ikeya, M., and Okamoto, M. Polymer 47 (2006): 5350–5359. © 2006, Elsevier Science. With permission.]

61259_C006.indd 200

10/25/2008 12:38:17 PM

201

Nanostructure Development and Foam Processing

50 PLA PLA/MMT-ODA

Fraction (%)

40

30

20

10

0

0

10

20

30

40

50

60

70

80

90 100

Cell size (μm) FIGURE 6.19 Typical example for cell size distribution of foamed PLA/MMT-ODA and neat PLA in experiments at 150°C under 24 MPa. Average values d in m and variances d2 in m2 in the Gaussian fit through the data are 24.2 and 19.1 for PLA/MMT-ODA foam, and 58.3 and 171.0 for PLA foam. [Reprinted from Ema, Y., Ikeya, M., and Okamoto, M. Polymer 47 (2006): 5350–5359. © 2006, Elsevier Science. With permission.]

molecules, which can subsequently form a small population of cell nuclei upon depressurization. The incorporation of nanoclay (OMLS) induces heterogeneous nucleation because of a lower activation energy barrier compared with homogeneous nucleation.36 However, the competition between homogeneous and heterogeneous nucleation is no longer discernible. 6.4.4 Foaming Temperature Dependence of Cellular Structure The dependence of the foam density (f) at the Tf under different CO2 pressures are shown in Figure 6.20. Throughout the whole CO2 pressure range, the mass density of PLA/MMT-ODA foams remains at a constant value at low foaming temperature (Tf) range and abruptly decreases beyond a certain Tf , and then attains a minimum constant value up to 150°C again. From the above results, it can be said that such behavior of mass density is due to the competition between cell nucleation and cell growth. At the low Tf range (~110°C), in which a large supply of CO2 molecules are provided, the cell nucleation is dominant, while at the high Tf , (~140°C), cell growth and the coalescence of cells are prominent due to low viscosity of the systems compared with the low Tf range (~110°C). This behavior clearly appears in the plots of the cell size ( 2d), the cell density (Nc), and the mean cell wall thickness (d) versus Tf under various pressure conditions, respectively. As seen in Figure 6.21, with increasing T.f all nanocomposite foams show an increasing tendency of 2d and/or  and attain a

61259_C006.indd 201

10/25/2008 12:38:18 PM

202

Polymeric Foams

1.2 PCO2 (MPa) 14 21 18 24 28 30

rf (g cm–3)

1.0

0.8

0.6

0.4

0.2 90

100

110

120

130

140

150

160

Tf (°C) FIGURE 6.20 Foaming temperature dependence of mass density for PLA/MMT-ODA foamed under different CO2 pressure conditions. [From Ema, Y., Ikeya, M., and Okamoto, M. Polymer 47 (2006): 5350–5359. © 2006, Elsevier Science. With permission.]

maximum. On the other hand, the temperature dependence of Nc shows opposite behavior compared with the tendency of 2d due to cell growth and coalescence. Both 2d and Nc affect the mass density of the foams. Using Tg depressions (corresponding to Tg), reconstructed plots of f versus Tf  Tg were drawn from the data of Figure 6.20. The results are shown in Figure 6.22. All of the data, including neat PLA and PLA/ MMT-SBE, neatly conform to a reduced curve with f ~ 1.0 ± 0.1 g/cm3 at Tf  Tg  140 ± 4°C (nanocellular region), whereas f values approach around 0.3 ± 0.15 g/cm3 as reduced temperature (Tf  Tg) increased well above 150°C (microcellular region). The critical temperature is thus 140 ± 4°C, above which cell growth prevails. Below the critical temperature, cell nucleation dominates and cell growth is suppressed due to the high modulus and viscosity as revealed by the temperature dependence of stage, G(), and loss, G(), moduli (G  162 MPa and viscosity component G/  2 MPa s at 140°C). Figure 6.23 shows temperature-reduced plots of 2d, Nc and  versus Tf  Tg. All data nicely conform to a reduced curve such as in Figure 6.22. Interestingly, when both Tg and Tm depressions were used37 to conduct superposition, it was recognized that the reduced curve is neatly constructed but there is no significant difference in comparison with the case of Tf  Tg. This indicates that Tg depression is important in optimizing foam processing conditions but Tm depression may be not a significant factor for processing because the Tf range is still below Tm after CO2 saturation. In Figure 6.24, the relationship between 2d and Nc, and  and 2d in this study are shown. The relationship neatly obeys Equations 6.2 and 6.3 but

61259_C006.indd 202

10/25/2008 12:38:18 PM

203

Nanostructure Development and Foam Processing

2d (μm)

(a) 102 PCO (MPa) 2 14 18 21 24 28 30

101

100

–1 (b) 10

Nc (μm)

1012 1010 108

d (mm)

106 1 (c) 10 100

10–1

10–2 90

100

110

120

130

140

150

160

Tf (°C) FIGURE 6.21 Foaming temperature dependence of: (a) cell size; (b) cell density; and (c) mean cell wall thickness under different CO2 pressure conditions. [From Ema, Y., Ikeya, M., and Okamoto, M. Polymer 47 (2006): 5350–5359. © 2006 Elsevier Science. With permission.]

the deviation occurs beyond the value of Nc ~ 1012 cell/cm3 for panel (a) and below 2d ~ 1 m for panel (b). The downward and upward deviations indicate that the heterogeneous cell distribution mechanism due to the rigid crystalline phases in the PLA matrix is caused by a high degree of crystallinity (~49 wt%) under the low foaming temperature range (~100°C). As seen in Figure 6.25, the PLACN foams exhibit a heterogeneous cell distribution. The PLA foam reduces the value of Nc accompanied by a large value . of  compared with that of PLACN foams. In the case of PLACN foams, the controlled structure of the PLACN foams is from microcellular (2d  30 m and Nc  3.0 × 107 cells/cm3) to nanocellular (2d  200 nm and Nc  2.0 × 1013 cells/cm3).

61259_C006.indd 203

10/25/2008 12:38:18 PM

204

Polymeric Foams

1.2 PLA PLA/MMT-ODA PLA/MMT-SBE

1.0

rf (g cm–3)

0.8 0.6 0.4 0.2 0 120

130

140

150

160

170

Tf + ΔT (°C) FIGURE 6.22 Plot of mass density for PLA/MMT-ODA, PLA/MMT-SBE and neat PLA versus reduced foaming temperature (Tf  Tg). The critical temperature (140 ± 4°C) is shaded. [From Ema, Y., Ikeya, M., and Okamoto, M. Polymer 47 (2006): 5350–5359. © 2006, Elsevier Science. With permission.]

6.4.5 CO2 Pressure Dependence At high pressure, both homogeneous and heterogeneous nucleation mechanisms may appear to be of comparable significance. All systems demonstrate that Nc increases systematically with increasing CO2 pressure in the low Tf region (~100–120°C). For PLA/MMT-ODA foams, the system suggests that the heterogeneous nucleation is favored in high-pressure conditions. The cell nucleation in the heterogeneous nucleation system such as PLA/MMT-ODA foams took place in the boundary between the matrix and the dispersed nanoclay particles. Accordingly, the cell size decreased without individual cell coalescence for PLA/MMT-ODA and neat PLA systems, as seen in Figure 6.25. To clearly investigate whether the addition of internal surfaces of the dispersed nanoclay may hinder CO2 diffusion by creating a more tortuous diffusive pathway,19 characterization of the interfacial tension between bubble and matrix was conducted using the modified classical nucleation theory.36 According to the theory proposed by Suh and Colton, the rate of nucle. ation of cells per unit volume (N ) can be written as

È -16pg 3 S(q ) ˘ N ~ Cf exp Í ˙ 2 ÍÎ 3(DPCO ) k B T ˙˚

(6.4)

2

61259_C006.indd 204

10/25/2008 12:38:19 PM

205

Nanostructure Development and Foam Processing

(a) 103

2d (μm)

102 101 PLA PLA/MMT–ODA PLA/MMT–SBE

100 10–1 (b) 1014

Nc (μm)

1012 1010 108 106

d (μm)

1041 (c) 10

100

10–1

10–2 120

130

140 150 Tf + ΔT (°C)

160

170

FIGURE 6.23 Temperature-reduced plots of: (a) 2d; (b) Nc ; and (c)  versus Tf  Tg for PLA/MMT-ODA, PLA/MMT-SBE and neat PLA. [From Ema, Y., Ikeya, M., and Okamoto, M. Polymer 47 (2006): 5350–5359. © 2006 Elsevier Science. With permission.]

where C is the concentration of CO2 and/or the concentration of heterogeneous nucleation sites, f is the collision frequency of CO2, is the interfacial tension between bubble and matrix, S() is the energy reduction factor for the heterogeneous nucleation (i.e. PLA/MMT-ODA), PCO2 is the magnitude of the pressure quench during depressurization, kB is the Boltzmann constant, and T is absolute temperature. The theoretical cell density is given by Ntheor =

61259_C006.indd 205

t

Ú N dt

(6.5)

0

10/25/2008 12:38:19 PM

206

Polymeric Foams

(a) 103 PLA PLA/MMT-ODA

2d (mm)

102

PLA/MMT-SBE –1/3

101

100

10–1

10–2 104

106

108

1010 Nc (cm–3)

1012

1014

(b) 102 PLA PLA/MMT-ODA

d (mm)

101

PLA/MMT-SBE

100 1

10–1

10–2 10–2

10–1

100 101 2d (mm)

102

103

FIGURE 6.24 Relationship between: (a) cell size versus cell density; and (b) cell wall thickness versus cell size for all foams. [From Ema, Y., Ikeya, M., and Okamoto, M. Polymer 47 (2006): 5350–5359. © 2006 Elsevier Science. With permission.]

where t is the foaming time that takes approximately 3 s. Assuming no effect of the coalescence of cell on the value of Nc, we estimate the interfacial tension of the systems calculated using Equations 6.4 and 6.5; that is, the slope of the plots (Nc versus 1/PCO2). The characteristic parameters of two systems are shown in Table 6.7. The interfacial tension of PLA/MMTODA and neat PLA are 6.65 mJ/m2 and 7.43 mJ/m2 at 110°C, respectively. . These estimated values are in good agreement with that of the other poly(methyl methacrylate) (PMMA)-CO2 system (~10 mJ/m2).38 The PLA/ MMT-ODA system has a low value compared to that of neat PLA. This trend reflects the relative importance of heterogeneous nucleation, which

61259_C006.indd 206

10/25/2008 12:38:19 PM

Nanostructure Development and Foam Processing

207

FIGURE 6.25 SEM images of the fracture surfaces of: (a) neat PLA; (b) PLA/MMT-ODA; and (c) PLA/MMT-SBE foamed at 100°C under 28 MPa. [From Ema, Y., Ikeya, M., and Okamoto, M. Polymer 47 (2006): 5350–5359. © 2006 Elsevier Science. With permission.]

61259_C006.indd 207

10/25/2008 12:38:19 PM

208

Polymeric Foams

TABLE 6.7 Characteristic Interfacial Parameters of Two Systems

PLA/CO2 PLA/MMT-ODA/CO2 PLA/CO2 PLA/MMT-ODA/CO2

Tf (°C)

S( )1/3 (mJ/m2)

110

7.43 6.65 7.08 5.38

120

S( )

(°)

0.717

107.3

0.439

85.3

Source: From Ema, Y., Ikeya, M., and Okamoto, M. Polymer 47 (2006): 5350–5359. © 2006 Elsevier Science. With permission.

dominates over the homogeneous one in the event that the amount of CO2 available for bubble nucleation is limited because of a lower activation energy barrier, as mentioned previously. That is, in the heterogeneous nucleation (PLA/MMT-ODA), we have to take the reduction of the critical energy into consideration because of the inclusion of nucleants, which is a function of the PLA-gas-nanoclay contact angle () and the relative curvature (W) of the nucleant surface to the critical radius of the nucleated phase.39 In the case of W  10, the energy reduction factor S() can be expressed: S()  (1/4) (2  cos )(1  cos )2

(6.6)

In the case of homogeneous nucleation S() is unity (  180°). The obtained values of the contact angle are 107.3° at 110°C and 85.3° at 120°C. The estimated reduction factor [S()  0.4–0.7] was not so small when we compared with the other nanofillers [e.g. carbon nanofibers, S()  0.006].40 However, experimentally, nanoclay particles lead to an increase in Nc. For PLA/MMT-SBE foams prepared under condition with low Tf (~100– 110°C) and high pressure (~28 MPa), the nanocomposite foams exhibit no significant difference in Nc compared with PLA/MMT-ODA foams. This reasoning is consistent with the large value of W in both systems. 6.4.6 TEM Observation To confirm the heterogeneous nucleation and the nanocellular features of foam processing, Okamoto et al. conducted TEM observation of the cell wall in the PLA/MMT-ODA foam. Figure 6.26 shows a TEM micrograph for the structure of the cell wall foamed at 100°C under 28 MPa. Interestingly, the grown cells having a diameter of ~200 nm are localized along the dispersed nanoclay particles in the cell wall. In other words, the dispersed nanoclay particles act as nucleating sites for cell formation and the cell growth occurs on the surfaces of the clays; that is, the cellular structure has an oval-faced morphology rather than spherical cellular structures for high Tf conditions (~140°C).

61259_C006.indd 208

10/25/2008 12:38:20 PM

Nanostructure Development and Foam Processing

209

nano-clay

nanocell

lamellar

nano-clay

200 nm FIGURE 6.26 TEM micrograph for the structure of PLA/MMT-ODA cell wall foamed at 100°C under 28 MPa. [From Ema, Y., Ikeya, M., and Okamoto, M. Polymer 47 (2006): 5350–5359. © 2006 Elsevier Science. With permission.]

In Figure 6.26, in addition to the nanocellular structure formation, we can observe a lammellar pattern beside the nanoclay particles. This behavior appears to arise from the formation of the -phase of the PLA crystal in the presence of nanoclay particles.41 This is a unique observation of the epitaxial crystallization of PLA grown up from clay surfaces due to the nucleation effect of the dispersed nanoclays. 6.4.7 Polycarbonate (PC)-Based Nanocomposite Foams Figure 6.27 shows the typical results of SEM images of the fracture surfaces of the PC/SMA blend (matrix) and PC-based nanocomposites foamed at 140°C under different isobaric saturation condition (10, 18, and 24 MPa).42 PC/SMA foams exhibit polygon closed-cell structures having pentagonal and hexagonal faces, which express the most energetically stable state of polygon cells. Obviously, under low saturation CO2 pressure (~10 MPa), both PC/SMA/MAE1 and PC/SMA/MAE2.5 foams exhibit larger cell

61259_C006.indd 209

10/25/2008 12:38:20 PM

210

Polymeric Foams

FIGURE 6.27 Typical SEM images of the fracture surfaces of the PC/SMA blend (matrix) and PC-based nanocomposites foamed at 140°C under different isobaric condition (10, 18, and 24 MPa). [From Ito, Y., Yamashita, M., Okamoto, M. Macromolecular Materials Engineering 291 (2006): 773–783. © 2006 Wiley-VCH. With permission.]

size compared with PC/SMA, indicating that dispersed clay particles hinder CO2 diffusion by creating a maze or a more tortuous path.8 However, high CO2 pressure (~24 MPa) provides a large supply of CO2 molecules, which can subsequently form a large population of cell nuclei upon depressurization. The incorporation of nanoclay hinders CO2 diffusion and simultaneously induces heterogeneous nucleation because of a lower activation energy barrier compared with homogeneous nucleation. In Figure 6.28, the relationship between d and Nc, and  and d are plotted. Equations 6.2 and 6.3 lead to these relations but some deviation occurs in each system. For example, PC/SMA/MAE2.5 (syn-FH-C18TM) (MMT  1 wt%) exhibits a smaller value of Nc under the same d value when compared with PC/SMA and PC/SMA/MAE1. For the relationship between  and Nc, PC/SMA/MAE2.5 (MMT  2.5 wt%) shows a large value of  compared with PC/SMA/MAE1. These deviations indicate that the heterogeneous cell distribution mechanism due to the rigid matrix phases in PC/SMA is caused by high MAE loading (MMT  2.5–5.0 wt%) as seen in Figure 6.28. As well as PLA-based nanocomposite foams, in Table 6.8, the interfacial tension and energy reduction factor of systems are summarized.42 The interfacial tension of PC/SMA/MAE1 (including energy reduction factor S()) and PC/SMA (S()  1) are 9.7 mJ/m2 and 10.9 mJ/m2 at 14 MPa, respectively. These estimated values of are of the same order of magnitude compared with the PLA system (5–7 mJ/m2). The interfacial tension slightly decreases with an increase in clay content under the same CO2 pressure conditions, indicating heterogeneous cell nucleation occurs

61259_C006.indd 210

10/25/2008 12:38:21 PM

211

Nanostructure Development and Foam Processing

(a) 1000 PC/SMA PC/SMA/MAE1 PC/SMA/MAE2.5 PC/SMA/MAE5

d (mm)

100

10 –1/3 1

0.1 105

107

1011 109 Nc cells (cm–3)

1013

1015

(b) 1000

d (mm)

100

PC/SMA PC/SMA/MAE1 PC/SMA/MAE2.5 PC/SMA/MAE5

10 1 1

0.1 0.1

1

10 d (mm)

100

1000

FIGURE 6.28 (a) Cell size versus cell density; and (b) cell wall thickness versus cell size for PC/SMA and PC based nanocomposite foams. [From Ito, Y., Yamashita, M., and Okamoto, M. Macromolecular Materials Engineering 291 (2006): 773–783. © 2006 Wiley-VCH. With permission.]

easily with increasing clay content. The estimated reduction factor [S()  0.3–0.8] is the same order compared with the foaming of PLAbased nanocomposites.35 The TEM micrograph of the structure of the cell wall foamed at 160°C is shown in Figure 6.29. The grown cells are localized along the dispersed nanoclay particles in the cell wall. 6.4.8 Mechanical Properties of Nanocomposite Foams Figure 6.30 shows the relationship of relative modulus (Kf/Kp) against relative density (f/p) of neat PLA and nanocomposite foams, taken in the parallel (a) and perpendicular (b) directions to the flow, respectively.

61259_C006.indd 211

10/25/2008 12:38:21 PM

212

Polymeric Foams

TABLE 6.8 Interfacial Tension [ S()1/3] Including Energy Reduction Factor [S()] of Systems S( )1/3 (mJ/m2)

Systems PC/SMA-CO2 PC/SMA/MAE1-CO2 PC/SMA/MAE2.5-CO2 PC/SMA-CO2 PC/SMA/MAE1-CO2 PC/SMA/MAE2.5-CO2 PC/SMA-CO2 PC/SMA/MAE1-CO2 PC/SMA/MAE2.5-CO2 PC/SMA-CO2 PC/SMA/MAE1-CO2 PC/SMA/MAE2.5-CO2

S( )

PCO2  10 MPa

10.7 8.6

1.0 0.53

PCO2  14 MPa

8.0 10.9 9.7

0.42 1.0 0.72

PCO2  18 MPa

9.9 13.6 10.2

0.77 1.0 0.42

PCO2  22 MPa

9.2 11.3 12.4

0.30 1.0 —

8.0

0.36

Source: From Ito, Y., Yamashita, M., and Okamoto, M. Macromolecular Materials Engineering 291 (2006): 773–783. © 2006 Wiley-VCH. With permission.

FIGURE 6.29 TEM micrograph of the structure of the cell wall foamed at 160°C. [From Ito, Y., Yamashita, M., and Okamoto, M. Macromolecular Materials Engineering 291 (2006): 773–783. © 2006 Wiley-VCH. With permission.]

61259_C006.indd 212

10/25/2008 12:38:22 PM

213

Nanostructure Development and Foam Processing

(a) 2.5 PLA PLA/MMT-ODA PLA/MMT-SBE

2.0

Kf /Kp

1.5

1.0

0.5

0 0

0.2

0.4

0.6

0.8

1

0.8

1

rf /rp (b) 2.5

2.0

Kf /Kp

1.5

1.0

1

0.5

0

0

0.2

0.4

0.6 rf /rp

FIGURE 6.30 The relation of relative modulus (Kf/Kp) against relative density (f/p) of neat PLA and PLA-based nanocomposite foams, taken in directions parallel (a) and perpendicular (b) to the flow.

To clarify whether the modulus enhancement of the nanocomposite foams was reasonable, we applied the following Equation 6.7 proposed previously by Kumar43 to estimate relative moduli with various foam densities: K

r

4

r

4

r

( )  (r )  (r )

___f  __f rp Kp

__f p

__f p

4

(6.7)

where Kp and Kf are the modulus of pre-foamed and post-foamed samples, respectively. The solid line in the figure represents the fit with Equation 6.7.

61259_C006.indd 213

10/25/2008 12:38:22 PM

214

Polymeric Foams

The neat PLA foams do not show any difference between the two moduli (a) and (b). On the other hand, for PLACN foams, the relative moduli exhibit a large value compared with the theoretical one. The dispersed clay particles in the cell wall align along the thickness direction of the sample. In other word, the clay particles arrange owing to the biaxial flow of material during foaming. The clay particles seem to act as a secondary cloth layer to protect the cells from being destroyed by external forces. In the directions perpendicular to the flow, the relative modulus of PLA/ MMT-ODA and PLA/MMT-SBE foams appear higher than the predicted value even at the same relative mass density in the range 0.7–0.85 (see Figure 6.30a). This upward deviation suggests that the small cell size with large cell density enhances the material property as predicted by Weaire.44 This may create the improvement of mechanical properties for polymeric foams through polymeric nanocomposites. More detailed surveys on various types of nanocomposite foaming can be also be found in the literatures.45–48 6.4.9 Porous Ceramic Materials via Nanocomposite A new route for the preparation of porous ceramic material from thermosetting epoxy/clay nanocomposite was first demonstrated by Brown et al.49 This route offers attractive potential for diversification and application of the PLFNCs. Okamoto and coworkers have reported the results on the novel porous ceramic material via burning of the PLA/MMT system (PLACN).50 The PLACN contained 3.0 wt% inorganic clay. The SEM image of the fracture surface of porous ceramic material prepared from simple burning of the PLACN in a furnace of up to 950°C is shown in Figure 6.31. After complete burning, as seen in the figure, the PLACN becomes a white mass with a porous structure. The bright lines in the SEM image correspond to the edge of the stacked silicate layers. In the porous ceramic material, the silicate layers form a house-of-cards structure, which consist of the large plates having a length of ~1000 nm and thickness of ~30–60 nm. This implies that the further stacked platelet structure is formed during burning. The material exhibits the open-cell type structure having a 100–1000 nm diameter void, a BET surface area of 31 m2 g1 and a low density of porous material of 0.187 g ml1 estimated by the buoyancy method. The BET surface area value of MMT (780 m2/g) and that of the porous ceramic material (31 m2/g), suggests about 25 MMT plates stacked together. When MMT is heated above 700°C (but below 960°C) all OH groups are first eliminated from the structure and thus MMT is decomposed into that of a non-hydrated aluminosilicate. This transformation radically disturbs the crystalline network of the MMT, and the resulting diffraction pattern is indeed often typical of an amorphous (or non-crystalline) phase. The estimated rough value of the compression modulus (K) is of the order of ~1.2 MPa, which is five orders of magnitude lower than the bulk modulus of MMT (~102 GPa).28 In the stress–strain curve, the

61259_C006.indd 214

10/25/2008 12:38:23 PM

Nanostructure Development and Foam Processing

215

FIGURE 6.31 SEM image of porous ceramic material after coating with platinum layer (~10 nm thickness). [From Sinha Ray, S., Okamoto, K., Yamada, K., and Okamoto, M. Nano Letters 2 (2002): 423–425. © 2002 American Chemical Society. With permission.]

linear deformation behavior is nicely described in the early stage of the deformation; that is, the deformation of the material closely resembles that of ordinary polymeric foams.51 This open-cell type porous ceramic material consisting of the house-of-cards structure is expected to provide strain recovery and an excellent energy dissipation mechanism after unloading in the elastic region up to 8% strain, probably each plate bend like leaf spring. This porous ceramic material is a new material possessing elastic feature and is very lightweight. This new route for the preparation of porous ceramic material via burning of nanocomposites can be expected to pave the way for a much broader range of applications of the PLSNCs. This porous ceramic material provides an excellent insulator flame retardant property for PLFNCs.28 The flame behavior must be derived from the morphological control of the shielding properties of the graphitic/clay created during polymer ablation.

6.5 Conclusions and Future Prospects Development of nanocomposite foams is one of the latest evolutionary technologies of polymeric foams. The nanocomposite foams offer attractive potential for diversification and application of conventional polymeric materials. Some of them are already commercially available and applied in

61259_C006.indd 215

10/25/2008 12:38:23 PM

216

Polymeric Foams

industrial products through Unitika Ltd., Japan, and open a new dimension for plastics and composite foams. The major impact will be at least a decade away.

References 1. Sinha R., S. and Okamoto, M. “Polymer/layered silicate nanocomposites: A review from preparation to processing.” Progress in Polymer Science 28 (2003): 1539–1641. 2. Usuki, A., Kojima, Y., Okada, A., Fukushima, Y., Kurauchi, T., and Kamigaito, O. “Swelling behavior of montmorillonite cation exchanged for ω-Amino acids by ε-Caprolactam.” Journal of Material Research 8 (1993): 1174–1178. 3. Gao, F. “Clay/polymer composites: The story.” Materials Today 7 (2004): 50–55. 4. Usuki, A., Hasegawa, N., and Kato, M. “Polymer–Clay nanocomposites.” Advances in Polymer Science 179 (2005): 135–195. 5. Okamoto, M., Nam, P. H., Maiti, M., et al. “Biaxial flow-induced alignment of silicate layers in polypropylene/clay nanocomposite foam.” Nano Letters 1 (2001): 503–505. 6. Nam, P. H., Okamoto, M., Maiti, P., et al. “Foam processing and cellular structure of polypropylene/clay nanocomposites.” Polymer Engineering Science 42 (2002): 1907–1918. 7. Fujimoto, Y., Sinha Ray, S., Okamoto, M., Ogami, A., and Ueda, K. “WellControlled biodegradable nanocomposite foams: from microcellular to nanocellular.” Macromolecular Rapid Communications 24 (2003): 457–461. 8. Mitsunaga, M., Ito, Y., Sinha Ray, S., Okamoto, M., and Hironaka, K. “Intercalated polycarbonate/clay nanocomposites: nanostructure control and foam processing.” Macromolecular Materials Engineering 288 (2003): 543–548. 9. Vaia, R. A., Ishii, H., and Giannelis, E. P. “Synthesis and properties of two-dimensional nanostructures by direct intercalation of polymer melts in layered silicates.” Chemistry of Materials 5 (1993): 1694–1696. 10. Vaia, R. A. and Giannelis, E. P. “Lattice model of polymer melt intercalation in organically-modified layered silicates.” Macromolecules 30 (1997): 7990–7999. 11. Vaia, R. A. and Giannelis, E. P. “Polymer melt intercalation in organicallymodified layered silicates: Model predications and experiment.” Macromolecules 30 (1997): 8000–8009. 12. Hiroi, R., Sinha Ray, S., Okamoto, M., and Shiroi, T. “Organically modified layered titanate: A new nanofiller to improve the performance of biodegradable polylactide.” Macromolecular Rapid Communications 25 (2004): 1359. 13. Lagaly, G. Clay Minerals 16 (1970): 1. 14. Nakano, S., Sasaki, T., Takemura, K., and Watanabe, M. “Pressure-Induced intercalation of alcohol molecules into a layered titanate.” Chemistry of Materials 10 (1998) 2044–2046. 15. Tateyama, H., Nishimura, S., Tsunematsu, K., Jinnai, K., Adachi, Y., and Kimura, M. “Synthesis of expandable fluorine mica from talc.” Clay Minerals 40 (1992): 180–185.

61259_C006.indd 216

10/25/2008 12:38:23 PM

Nanostructure Development and Foam Processing

217

16. Yoshida, O. and Okamoto, M. “Direct melt intercalation of polylactide chains into Nano-Galleries: Interlayer expansion and nanocomposite structure.” Macromolecular Rapid Communications 27 (2006): 751–757. 17. Sinha Ray, S., Yamada, K., Okamoto, M., and Ueda, K. “Biodegradable polylactide/montmorillonite nanocomposites.” Journal of Nanoscience and Nanotechnology 3 (2003): 503–510. 18. Yoshida, O. and Okamoto, M. “Direct intercalation of polymer chains into Nano-Galleries: Interdigitated layer structure and interlayer expansion.” Journal of Polymer Engineering 26 (2006): 919–940. 19. Sinha Ray, S., Yamada, K., Okamoto, M., Ogami, A., and Ueda, K. “New polylactide/layered silicate nanocomposites.3. High-Performance biodegradable materials.” Chemistry of Materials 15 (2003): 1456–1465. 20. Saito, T., Okamoto, M., Hiroi, R., Yamamoto, M., and Shiroi, T. “Intercalation and interlayer expansion in the mixtures of organically modified layered fillers and Poly(p-Phenylenesulfide).” Macromolecular Materials Engineering 291 (2006): 1367–1374. 21. Saito T., Okamoto M., Hiroi R., Yamamoto M., and Shiroi T. “Delamination of organically modified layered filler via solid-state processing” Macromolecular Rapid Communications 27 (2006): 1472–1475. 22. Toth, R., Coslanicha, A., Ferronea, M., et al. “Computer simulation of polypropylene/organoclay nanocomposites: Characterization of atomic scale structure and prediction of binding energy.” Polymer 45 (2004): 8075–8083. 23. Sinsawat, A., Anderson, K. L., Vaia, R. A., and Farmer, B. L. “Influence of polymer matrix composition and architecture on polymer nanocomposite formation: Coarse-Grained molecular dynamics simulation.” Journal of Polymer Science Part B: Polymer Physics 41 (2003): 3272–3284. 24. Kuppa, V., Menakanit, S., Krishnamoorti, R., and Manias, E. J. “Simulation insights on the structure of nanoscopically confined Poly(ethylene oxide).” Journal of Polymer Science Part B: Polymer Physics 41 (2003): 3285–3298. 25. Zeng, Q. H., Yu, A. B., Lu, G. Q., and Standish, R. K. “Molecular dynamics simulation of organic-inorganic nanocomposites: Layering behavior and interlayer structure of organoclays.” Chemistry of Materials 15 (2003): 4732–4738. 26. Sheng, N., Boyce, M. C., Parks, D. M., Rutledge, G. C., Abes, J. I., and Cohen, R. E. “Multiscale micromechanical modeling of polymer/clay nanocomposites and the effective clay particle.” Polymer 45 (2004): 487–506. 27. Okamoto, M., Nam, P. H., Maiti, P., Kotaka, T., Hasegawa, N., and Usuki, A. “A house of cards structure in polypropylene/clay nanocomposites under elongational flow.” Nano Letters 1 (2001): 295–298. 28. Okamoto, M. “Polymer/layered silicate nanocomposites.” “Polymer/ layered silicate nanocomposites.” Rapra Review Report No. 163, Rapra Technology Ltd, London, 2003. 166 pp. 29. Maiti, P. and Okamoto, M. “Crystallization control via silicate surface in nylon 6-clay nanocomposites.” Macromolecular Materials Engineering 288 (2003): 440–445. 30. Nam, P. H. The Structure and Properties of Intercalated Polypropylene/Clay Nanocomposite, MSc thesis, Toyota Technological Institute, Nagoya 2001. 31. Sinha Ray, S., Okamoto, K., and Okamoto, M. “Structure-Property relationship in biodegradable poly(butylenes succinate)/layered silicate nanocomposites.” Macromolecules 36 (2003): 2355–2367.

61259_C006.indd 217

10/25/2008 12:38:23 PM

218

Polymeric Foams

32. van Olphen, H. An Introduction to Clay Colloid Chemistry. Wiley, New York, 1977. 33. Fratzl, P. “Cellulose and collagen: From fibres to tissues.” Current Opinion in Colloid Interface Science 8 (2003): 32–39. 34. Taki, K., Yanagimoto, T., Funami, E., Okamoto, M., and Ohshima, M. “Visual observation of CO2 Foaming of Polypropylene–Clay nanocomposites.” Polymer Engineering Science 44 (2004): 1004–1011. 35. Ema, Y., Ikeya, M., and Okamoto, M. “Foam processing and cellular structure of polylactide-based nanocomposites.” Polymer 47 (2006): 5350–5359. 36. Colton, J. S. and Suh, N. P. “The nucleation of microcellular thermoplastic foam with additives. 1. Theoretical considerations.” Polymer Engineering Science 27 (1987): 485–492. 37. Takada, M. Crystallization Control and Foam Processing of Semi-crystalline Polymers via Supercritical CO2. PhD thesis, Kyoto University, 2004. 38. Goel, S. K. and Beckman, E. J. “Generation of microcellular polymeric foams using supercritical carbon-dioxide.1. Effect of pressure and temperature on nucleation.” Polymer Engineering Science 34 (1994): 1137–1147. 39. Fletcher, N. H. “Size effect in heterogeneous nucleation.” Journal of Chemical Physics 29 (1958): 572–576. 40. Shen, J., Zeng, C., and Lee, L. J. “Synthesis of Polystyrene–Carbon nanofibers nanocomposite foams.” Polymer 46 (2005): 5218–5224. 41. Nam, J. Y., Sinha, S. R., and Okamoto, M. “Crystallization behavior and morphology of biodegradable polylactide/layered silicate nanocomposite.” Macromolecules 36 (2003): 7126–7131. 42. Ito, Y., Yamashita, M., and Okamoto, M. “Foam processing and cellular structure of polycarbonate-based nanocomposites.” Macromolecular Materials Engineering 291 (2006): 773–783. 43. Kumar, V. and Weller, J. E. The 49th Annual Technical Conference (ANTEC) (1991): 1401. 44. Weaire, D. and Fu, T. L. “The mechanical-behavior of foams and emulsions.” Journal of Rheology 32 (1988): 271. 45. Lee, L. J., Zeng, C., Cao, X., Han, X., Shen, J., and Xu, G. “Polymer nanocomposite foams.” Composites Science and Technology 65 (2005): 2344–2363. 46. Cao, X., Lee, L. J., Widya, T., and Macosko, C. “Polyurethane/clay nanocomposites foams: Processing, structure and properties.” Polymer 46 (2005): 775–783. 47. Chandra, A., Gong, S., Turng, L. S., Gramann, P., and Cordes, H. “Microstructure and crystallography in microcellular injection-molded polyamide-6 nanocomposite and neat resin.” Polymer Engineering Science 45 (2005): 52–61. 48. Strauss, W. and D’Souza, N. A. “Supercritical CO2 processed polystyrene nanocomposite foams.” Journal of Cellular Plastics 40 (2004): 229–241. 49. Brown, J. M., Curliss, D. B., and Vaia, R. A. Proceedings of the Polymer Materials Science Engineering Spring Meeting, San Francisco, 2000, p. 278. 50. Sinha Ray, S., Okamoto, K., Yamada, K., and Okamoto, M. “Novel porous ceramic material via burning of polylactide/layered silicate nanocomposite.” Nano Letters 2 (2002): 423–425. 51. Gibson, L. J. and Ashby, M. F. (eds) Cellular Solids: Structures and Properties Pergamon Press, New York, 1988, p. 8.

61259_C006.indd 218

10/25/2008 12:38:23 PM

7 New Material Developments from the Nitrogen Autoclave Process Neil Witten

CONTENTS 7.1 Introduction ................................................................................... 7.2 Brief History .................................................................................. 7.3 Nitrogen Autoclave Technology ................................................ 7.3.1 Raw Materials ...................................................................... 7.3.2 Process Overview ............................................................... 7.3.3 First Stage—Sheet Extrusion ............................................. 7.3.4 Second Stage—High Pressure Autoclave ........................ 7.3.5 Third Stage—Low Pressure Autoclave ........................... 7.3.6 Quality.................................................................................... 7.3.7 Alternative Cross-Linked Polyolefin Foam Technologies ............................................................. 7.4 Block Processes ............................................................................... 7.5 Semi-Continuous (Roll) Processes ............................................. 7.6 Nitrogen Autoclave Cross-Linked Polyolefin Foam Products ...... 7.6.1 Product Range ..................................................................... 7.6.2 Cross-Linked High-Density Polyethylene Foams .......... 7.6.3 Cross-Linked Metallocene Polyethylene Foams ............ 7.6.4 Markets and Applications ................................................. 7.6.4.1 Packaging ................................................................. 7.6.4.2 Sports and Leisure .................................................. 7.6.4.3 Medical/Health Care ............................................. 7.6.4.4 Other .........................................................................

61259_C007.indd 219

220 220 221 221 223 224 225 227 228 229 229 232 233 233 236 236 237 237 238 238 238

10/25/2008 12:39:00 PM

220

Polymeric Foams

7.7 New Material Developments ...................................................... 7.7.1 ZOTEK F Polyvinylidene Fluoride (PVDF) Foams ......... 7.7.1.1 Flammability Characteristics ................................. 7.7.1.2 Environmental and Chemical Performance ........ 7.7.1.3 Basic Thermal Properties ....................................... 7.7.1.4 Other Performance Attributes ............................... 7.7.1.5 Markets and Applications ...................................... 7.7.2 ZOTEK N Polyamide (PA 6) Foams .................................. 7.7.2.1 Thermal Properties ................................................. 7.7.2.2 High Temperature Modulus and Moisture ......... 7.7.2.3 Other Performance Attributes .............................. 7.7.2.4 Markets and Applications ...................................... 7.8 Conclusions ................................................................................... References ............................................................................................

239 239 240 241 243 244 245 246 247 249 251 251 251 252

7.1 Introduction Zotefoams plc is the leading manufacturer of low-density, closed-cell, crosslinked block foams produced using a unique, proprietary high-pressure gas technology which yields significant product advantages over competitive technologies. In recent years the company has been involved in the development of a new generation of low-density foams based on materials where foaming has traditionally been an insurmountable technical challenge or limitations have existed as to the reduction in density possible. These novel foam products are based on fluoropolymers and polyamides. The characteristics of the foams, not surprisingly, reflect the general properties of these classes of polymer. This chapter will review the nitrogen autoclave technology, historically employed in the manufacture of cross-linked polyolefin foams, and consider the other major technologies used to manufacture such materials. Finally a review of the key performance attributes of the new, unique, cellular materials described above is provided.

7.2 Brief History The process technology referred to variously as the “nitrogen autoclave,” “autoclave batch,” “BXL” (a previous company name), or simply the “autoclave” process is a unique process technology within the foam industry. The process is proprietary to Zotefoams plc (headquartered in Croydon, UK) and is one that has seen much development over the past

61259_C007.indd 220

10/25/2008 12:39:00 PM

New Material Developments from the Nitrogen Autoclave Process

221

100 years or so. These developments in both process technology and materials continue apace and the process now offers an extremely wide range of foam materials for an equally wide end-user base. Before describing the technology and the more recent material developments, it is an interesting diversion to look briefly at the early process history, which is a story of individual technical adventure and entrepreneurial development dating back to the late nineteenth/early twentieth century. The origins of the process begin with the Austrian, Mr. Robert Pfleummer, and his three sons, Hans, Fritz, and Herman. The Pfleummer family often cycled and all three regularly suffered punctures. To alleviate this problem and following a particularly embarrassing failure in a cycle race, the father challenged his three sons to develop a puncture-proof tire. The father also recognized at the time that such a product could have applications in the developing automotive industry. The sons were scientifically minded and looked at a number of options before developing the process with rubber. Figure 7.1 shows a laboratory set-up of plant and equipment in London, circa 1908. Figure 7.2 shows Fritz Pfleummer with an expanded rubber cylinder, circa 1912. In the background are the gas cylinders, compressors, gauges, and so on necessary for the process. Figure 7.3 shows the rather disappointing result of a test run by one of the other sons, Hans Pfleummer, using his Ford Buick, dated around 1912. Having had the tires of his car filled with expanded rubber, the foamed elastomer is clearly seen to have burst through the outer sheath of the tire during the test. Although ultimately not successful in this particular application, the process technology was sufficiently interesting for a number of individuals to attempt to commercialize the products. It took a further 20 years or so however before a successful business was formed (with many failing in between) and many decades later before the technology was developed to large-scale production. In the early 1930s the process moved from the original base in Crystal Palace to the current location in Croydon, England where it has remained to this date. The company has passed through several different owners since these early stages. For those interested in industrial history, the development of the business from the earliest days through to 1957 was suitably documented in a book1 written by the then Technical Director, Mr. A. Cooper, which was published on the twentyfirst anniversary of the company.

7.3 Nitrogen Autoclave Technology 7.3.1 Raw Materials One of the great flexibilities of the autoclave technology is that it has proven applicable to expanding a great many synthetic materials. In the

61259_C007.indd 221

10/25/2008 12:39:00 PM

222

Polymeric Foams

FIGURE 7.1 Laboratory set-up of plant and equipment in London, ca. 1908.

simplest representation a sheet of the polymer is exposed to a gas environment (normally nitrogen) at a certain pressure and temperature for a time long enough to ensure complete saturation with gas. Ignoring for a moment the additional thermal and rheological characteristics of the material necessary for expansion to be successful, it can be seen that this fundamental process of gas dissolution can be easily applied to a wide range of polymeric materials. Although the above indicates that the process could apply to any material, there are of course some limitations. Primary restrictions on material choice2 are that first, the material must have a reasonable level of gas solubility and diffusivity at practical pressures and temperatures; and second, the material must have some degree of melt strength which may be further enhanced or controlled via cross-linking or other means.

61259_C007.indd 222

10/25/2008 12:39:00 PM

New Material Developments from the Nitrogen Autoclave Process

223

FIGURE 7.2 Fritz Pfleummer, with an expanded rubber cylinder, ca. 1912.

7.3.2 Process Overview The process has three fundamental stages, although there are some variations to this dependent on product and material type. For the most part the descriptions which follow refer to the basic production of cross-linked lowdensity polyethylene (LDPE) and ethylene-vinyl acetate (EVA) foams. The process begins with the extrusion and cross-linking of a continuous sheet of material from which a “precursor” solid sheet is cut. This

FIGURE 7.3 Result of test by Hans Pfleummer, in his Ford Buick, ca. 1912.

61259_C007.indd 223

10/25/2008 12:39:00 PM

224

Polymeric Foams

precursor sheet is then transferred to a second stage where it is placed in a large, high-pressure autoclave and exposed to nitrogen gas at high temperature and pressure. The gas is absorbed into the polymer structure under these conditions until saturation is achieved at which point a thermodynamic instability is introduced, via a rapid pressure reduction, to nucleate the cell structure. For control reasons, the sheet is not fully expanded at this stage and instead the product is cooled and removed from the autoclave in nucleated form. Finally the nucleated sheet from the previous stage is transferred to a further, lower-pressure autoclave for full expansion. In this final stage the material is subjected to a mild gas pressure and reheated until the entire sheet is at a uniform temperature. Once at thermal equilibrium, the pressure is reduced and the polymer, being soft and extensible at this higher temperature, is fully expanded. 7.3.3 First Stage—Sheet Extrusion The initial sheet extrusion is perhaps the most critical stage of the process since it is in this stage where the batch-to-batch variability of the raw materials interacts with the day-to-day process and hardware variables along with having greater operator input than in the later process stages. This process step is shown schematically in Figure 7.4. Any additives are mixed with the base polymer either via preblending or by gravimetric feed into the barrel of the extruder. In simplest form the formulation comprises a base polymer (e.g. LDPE) together with a small fraction of an organic peroxide to effect cross-linking. Obviously other additives such as pigments, flame retardants, and so on, can also be added at this point as necessary, but it is worth noting that the process is typically operated without the use of nucleating agents. Typically the melting/mixing takes place in single screw extruders, although twin screw extruders are also employed. The material is mixed and pumped through a heavy duty sheet die to produce a thick sheet. The target thickness of the sheet can range between 5 and 25 mm, with typical thicknesses being 10–12 mm. The sheet is extruded onto a moving conveyor and is then passed through a series of oven units. It is in the oven units where the cross-linking reaction occurs, which takes around 30–40 minutes at the temperatures employed. Once fully

FIGURE 7.4 Schematic of the sheet extrusion process step (stage 1).

61259_C007.indd 224

10/25/2008 12:39:00 PM

New Material Developments from the Nitrogen Autoclave Process

225

cross-linked and after subsequent cooling, the sheet is then cut to specific dimensions which are in direct proportion to the required dimensions of the final foamed sheet. For each formulation, there is then an optimum degree of cross-linking to give sufficient melt strength, while also limiting over-expansion of the sheet. The influence of the melt rheology throughout the process is paramount and is tightly controlled. It is important to note that the result of this stage is no more than a sheet of cross-linked polymer, of high purity and homogeneity. Typically the polyolefin polymers used are of high molecular weight and low melt index, thereby maximizing the final foam properties; however, the use of such materials also requires that quite low levels of cross-linking agent are employed, relative to alternative technologies. Although perhaps more directly “critical to quality,” the extrusion stage utilizes, for the most part, standard hardware and the extrusion of thick sheets is by no means a new technology. That said, there are significant difficulties which need to be overcome to extrude a very thick solid sheet of the necessary internal quality. The reader should consider that at some later point the entire volume of the sheet will be expanded by up to 65 times; any minor flaws or inhomogeneities introduced via the extrusion process will also be expanded by that same factor! 7.3.4 Second Stage—High Pressure Autoclave It is in this second stage that the process technology becomes more obviously unique. This process step is shown schematically in Figure 7.5. As noted above, the extruded sheets are completely free of blowing agent and nucleating agents. To introduce the blowing agent, the sheet is simply exposed to a gas at extremely high pressure and temperature. The sheet stock from the extrusion stage is loaded into the autoclave using specially designed carriage equipment. The entire carriage is then transferred into the autoclave where the physical blowing agent is introduced. The physical blowing agent of choice is nitrogen. As nitrogen is supercritical at temperatures above 147°C and pressures above 34 bar, then it is quite correct to describe the nitrogen autoclave technology as a supercritical fluid (SCF) technology. Once in the vessel, the gas and polymer are heated and pressurized. Pressures and temperatures up to 670 bar and 250°C respectively are employed. At 250°C the temperature is well above the melting point of a

FIGURE 7.5 Schematic of the gas absorption process step (stage 2).

61259_C007.indd 225

10/25/2008 12:39:01 PM

226

Polymeric Foams

typical LDPE polymer. While at such pressures and temperatures, the crosslinking previously undertaken enhances the polymer sheet dimensional stability thereby preventing flow. Additionally the quite low solubility of nitrogen in the polymer is overcome such that sufficient gas dissolves to allow significant expansion of the sheet in the third stage. At the conditions mentioned above, cross-linked LDPE foam of 15 kg/m3 (i.e. greater than 60 times volume expansion) can be produced commercially. It is well known that the solubility of gas in the polymer will be affected by both pressure and temperature.3,4 Varying both of these parameters in the process allows control over the amount of gas to be absorbed by the sheet of polymer. Operationally this means a wide range of final expansion ratios are possible from the same starting formulation, as gas pressure and temperature are the key controls of product density in the process. Pressure and temperature are the main drivers of gas dissolution but the third variable of importance is the sheet thickness. It is well known that the mass of gas absorbed within a given time, t, into a plane sheet is related to the square of the thickness of that sheet.5 Therefore in the nitrogen autoclave process increasing the thickness of the cross-linked sheet has a t2 effect on the process cycle time in this second stage. To emphasize the importance of this point one needs to consider first the costs associated with operating at the very high pressures and temperatures involved and second that typical process cycle times for products of, for example, 10 mm (solid) thickness are of the order of 6–8 hours. Once saturation has been assured, then the next step is to nucleate the cell structure. This is achieved via a rapid depressurization, whereby the gas pressure may be reduced to a small fraction of the saturation pressure in a matter of a few seconds. The introduction of this thermodynamic instability renders the sheet supersaturated with gas, causing the nitrogen gas to come out of solution to form cell nuclei. This process stage is again tightly controlled and the expansion limited, to allow nucleation but restrict bubble growth. This step not only nucleates the structure of the final foam but at the same time control of this depressurization, more specifically the rate of depressurization, leads to the ability to control final foam cell structures. More rapid rates of pressure reduction result in a relatively high nucleation density and therefore fine cell structures in the foam, whereas slower rates of pressure drop tend to reduce nucleation density and therefore render more coarse cellular structures in the final foam. The effect of employing thermodynamic instability in the nucleation of foam structures has been studied more recently in relation to the development of microcellular foam processes.6,7 After the rapid depressurization and once pressure has again stabilized, the sheets are then cooled and removed from the autoclave. Although technically feasible to expand fully in a single stage, as noted earlier, other considerations dictate that full expansion is undertaken in a separate step.

61259_C007.indd 226

10/25/2008 12:39:01 PM

New Material Developments from the Nitrogen Autoclave Process

227

FIGURE 7.6 SEM of the structure of the intermediate, nucleated LDPE sheet.

The density of the polymer sheet at this point following nucleation may be reduced by around 30% or more—in the case of LDPE this equates to a density of around 600–700 kg/m3. The structure of the material is typically microcellular (cell diameters  100 μm) and structurally remarkably consistent. Figure 7.6 shows an SEM of the structure of this intermediate nucleated material in LDPE, with cells of the order of 30–50 μm in diameter. The description above demonstrates some of the important operational controls on the nitrogen autoclave technology. Both density and cell size are largely determined by the process conditions and much less so by the material formulation—a single formulation therefore can give rise to a very wide range of products (sheet size, foam density, and cell dimensions). It is these points combined that have led some academics8 to study the materials to gain a better understanding on the structure/property effects in cellular polymers, without the complications introduced as a result of formulation changes. 7.3.5 Third Stage—Low Pressure Autoclave It is here that the final foam expansion is achieved. The material from the previous stage is loaded onto an open tray carriage system and transferred into a further autoclave where the gas-laden nucleated sheet is reheated. Typically the temperature used for LDPE is around 150°C and the pressure is of the order of 15–20 bar. The gas environment within the autoclave may be either air or nitrogen. Pressure is initially applied while the sheet is heating. Once the sheet has been allowed to equilibrate at the required temperature and pressure

61259_C007.indd 227

10/25/2008 12:39:01 PM

228

Polymeric Foams

FIGURE 7.7 Schematic of the final expansion process step (stage 3).

for the appropriate time, the pressure is reduced. As the temperature is some 40°C above the melting point of a typical LDPE polymer during this pressure reduction, the polymer sheet expands as the gas pressure is reduced. This final stage of the process is shown schematically in Figure 7.7. An obvious simplification of this stage could be envisaged where the nucleated sheet is expanded in a simple air-circulating oven. While this is possible, such an approach is difficult to control due to the uneven heating effects that occur in thick, low thermal conductivity sheets of gas-laden polymer. The final foam sheets, sometimes referred to as blocks or buns, are typically 2000 mm long, 1000 mm wide, and 30–50 mm in thickness; however, the ability to quickly and easily manipulate sheet sizes independently of density/structure was indicated earlier and offers great flexibility to the process. Finally, as the expansion process is unrestrained, the properties of the foam material produced are largely isotropic. This is a significant benefit to end users in certain markets where the consistency of density and properties throughout the foam sheet is a key attribute, such as in cushion packaging or thermoforming. 7.3.6 Quality One key operational advantage of the process is the ability, at each discrete stage of the foaming process, to carry out quality assurance checks prior to further processing. From the extrusion stage through crosslinking, gas absorption, and then foam expansion, each intermediate product may be quality assured prior to the next process step. Sheet weight, sheet dimensions, sheet profile (thickness), and degree of cross-linking are all monitored on a continual and routine basis. The latter is monitored to ensure that the sheet expansion and properties will be as expected, assuming that all key parameters in the later stages are fixed (most notably gas pressure and temperature). By way of example, Table 7.1 shows the effect that varying cross-link level can have on the density and performance of standard ethylene-vinyl acetate (EVA) foam.9 As would be predicted, the effect of increasing the cross-link level is to increase density by restricting expansion/bubble growth. In performance terms the effect of increasing cross-linking is to lead to less ductile behavior in tension

61259_C007.indd 228

10/25/2008 12:39:01 PM

New Material Developments from the Nitrogen Autoclave Process

229

TABLE 7.1 Effect of Varying Cross-Link Level on the Density and Basic Properties of Standard Ethylene-Vinyl Acetate (EVA) Foam Peroxide Level (100% = Standard Foam Formulation Level) Property 3

Nominal density (kg/m ) Compression stress (e  50%) (kPa) Compression set (20%/48 h/0.5 h rec) Tensile strength (kPa) Elongation at break (%)

80

100

140

47.8 102

48.6 115

52.0 129

7.7 616 304

6.7 896 197

5.1 926 171

(elongation at break falls as tensile strength increases), while the material becomes more elastic in compression. 7.3.7 Alternative Cross-Linked Polyolefin Foam Technologies The nitrogen autoclave process as described above is one of only a small number of technologies that have been developed for foaming crosslinked polyolefins. With the exception of autoclave technology, the other major technologies in use today were all developed in Japan in the 1960s. Since then these technologies have been further enhanced and successfully licensed around the globe. In practical terms the processes can be categorized firstly according to the basic product form, either block or roll, and then further by the details of the processing methods employed. Figure 7.8 shows one such categorization.

7.4 Block Processes The only major alternative to the nitrogen autoclave process for the foaming of discrete blocks (often also referred to as “sheets” or “buns”) of cross-linked polyolefin foam is the press forming technique. This technology was developed out of processes traditionally employed to foam natural and synthetic elastomers but has been adapted for the needs of polyolefin materials over the past four decades such that commercial products now cover a broad density range, from 24 kg/m3 up to around 300 kg/m3. At the higher density end of this spectrum, the materials are expanded in a single-step however to achieve the very high expansions (between 15 and 40 times volume expansion) then a dual-step process is employed, achieving greater utilization of the available gas.

61259_C007.indd 229

10/25/2008 12:39:01 PM

230

Polymeric Foams

Low density, crosslinked polyolefin foam technologies

Discrete block process technologies

Semi-continuous sheet process technologies

(sometimes referred to as ‘sheet’ ‘bun’ or ‘batch’ processes)

(sometimes referred to as ‘roll’ foam’ processe)

Physically expanded

Chemically expanded

(Physical blowing agent)

(Chemical blowing agent)

Nitrogen autoclave process

Press forming process

Physically crosslinked

Chemically crosslinked

Vertical (oven)

Horizontal (salt bath)

Sekisui process

Toray process

Furukawa process

Hitachi process

FIGURE 7.8 Categorization of alternative technologies for the production of low-density cross-linked polyolefin foams based on product form and processing methods.

The basic process technology involves the mixing and homogenization of the polymer, chemical blowing agent, normally azodicarbonamide (ADCN) of specific particle size and a peroxide cross-linking agent in a batch mixer. The output from the mixer is then loaded into a mold and press formed. During this stage the cross-linking agent is reacted with the polymer to enhance the polymer melt strength and the blowing agent decomposition occurs. In the single-step process, rapid opening of the press platens leads to foam expansion. In the dual-step process, of which a number of variations exist, the sheets from the first stage undergo a controlled and limited expansion before being transferred to a mold where they are steam heated and expanded fully. One of the drawbacks of this process is the very high pressure that is generated in the polymer which leads to high-pressure hydraulic press equipment being required. A more recent novel variation of this is described in US patent 5955015,10,11 which uses annular cylindrical molds to overcome the very high clamping pressures that would otherwise be needed in manufacturing low-density block and thereby reducing capital cost also. Relative to the physically blown products from the nitrogen autoclave process, the chemically blown block products typically suffer from a larger variation in density within a single sheet due to the difficulties in controlling the exothermic reaction of the blowing agent. At the center of the expanding chemically blown block the blowing agent typically sees a higher temperature for a longer time than the outer regions and as the gas yield is related to the time and temperature conditions during decomposition this

61259_C007.indd 230

10/25/2008 12:39:01 PM

231

New Material Developments from the Nitrogen Autoclave Process

leads to a greater variation in density through the thickness of the block than is seen in the physically blown autoclave products. This difference is measurable by cutting sections through the thickness of the product and calculating density of each section to develop a density profile for the sheet. Table 7.2 offers a comparison of the properties of nominal 33 kg/m3 foams physically blown using the nitrogen autoclave process and those of commercially available chemically blown block products. The notable differences are that the products from the nitrogen autoclave process typically exhibit higher hardness, greater stiffness in compression, and a more discernible yield point in the compression curve. These characteristics are partly attributable to the larger cell size generated in the nitrogen autoclave products but are also a result of the more regular structure that is achieved in the foams as a result of the combination of the use of physical blowing agent and the homogeneous nucleation step. The improved structure of the products also tends to lead to improved compression set properties as gas escape and diffusion through the cell

TABLE 7.2 Properties of Physically Blown Foam from the Nitrogen Autoclave Process Compared to Three Commercially Available Chemically Blown Block Products (All Nominally 30 kg/m3 Cross-Linked Polyethylene)

Property

Method

Unit

Density Cell size Shore hardness Compressive stress @ 10% strain @ 25% strain @ 40% strain @ 50% strain Compression set 22 h/50%/0.5 h recovery 22 h/50%/24 h recovery Tensile strength Elongation at break Tear strength

ISO 845 Internal ISO 868 ISO 7214

kg/m3 mm ‘00’

61259_C007.indd 231

kPa kPa kPa kPa

Nitrogen Autoclave Product 31 0.35 60

54 72 105 140

Chemically Blown Product A 32 0.19 57

43 62 95 129

Chemically Blown Product B 27 0.25 54

37 55 87 119

Chemically Blown Product C 39 0.16 57

46 66 99 135

ISO 7214 %

22.4

25.8

24.1

23.9

%

13

16.2

15.6

13.4

ISO 7214

kPa

528

322

310

403

ISO 7214

%

121

140

88

178

ISO 8067

N/m

590

728

857

518

10/25/2008 12:39:01 PM

232

Polymeric Foams

walls is more restricted. Comparing the results in Table 7.2, product C can be seen to be similar in compression set, however the density of the samples should also be noted with product C being some 25% higher in density, which naturally improves the compression set behavior. Finally, the nitrogen autoclave products tend to give higher strength values in tension while the elongation at break and tear strength values are generally lower than those of chemically blown products. In these latter properties the raw materials used, the degree of cross-linking and the cell size of the product all have a more significant bearing than they would be expected to have on the compressive properties and so direct comparisons become more challenging.

7.5 Semi-Continuous (Roll) Processes These processes are capable of producing a thin sheet on a reel which is perhaps a few hundred meters in length and a meter or more in width. The product thicknesses are from less than 1 mm to a maximum of 16 mm, which is a restriction that applies as a result of both the processing requirements as well as the practical necessity to store and ship the material in practical lengths on a reel. Roll foam processes are defined by the means of cross-linking, either chemical or physical. The chemically cross-linked processes were developed independently by the Furukawa Electric Co. and the Hitachi Co. and have been licensed successfully. Both processes employ horizontal ovens through which the material is cross-linked and expanded. In the Hitachi process these steps are slightly separated by means of a sheet preheating stage where cross-linking is initiated prior to expansion. In the Furukawa process the cross-linking and expansion processes overlap and progress simultaneously in the foaming oven. This difference means the Hitachi process is perhaps a little more controllable and flexible in material formulation choices. Other differences exist in the details of the design of the expansion ovens and the material handling equipment during the expansion step. The second group of processes, developed separately by Sekisui Chemical Co. and Toray Plastics rely on the physical cross-linking of the extruded sheet. The sheet containing polymer and blowing agent (typically azodicarbonamide) is extruded onto a reel. This reel is then physically crosslinked via electron beam irradiation at some later point. The thin section of the sheet means that there are no penetration depth issues leading to variable cross-linking and nor is a very high power beam necessary. Once cross-linked the sheet may then be expanded. In the Sekisui process expansion takes place in a vertical oven. This expansion technology has the advantage of being more energy efficient

61259_C007.indd 232

10/25/2008 12:39:01 PM

New Material Developments from the Nitrogen Autoclave Process

233

but means that the material is naturally drawn in the length direction as the sheet stretches under its own weight which leads to directionality of the properties of the finished sheet. This is not necessarily a negative point, however, as the elongated cell structure in these materials can also give a more drapeable, compliant, and ‘soft feeling’ material. Sekisui have continued to work with the process to develop ultra-thin foams down to 200 μm in thickness. In the Toray process, the sheet travels horizontally during expansion and is floated on a molten salt bath, which in tandem with infrared heating leads to expansion of the sheet. As a result of the horizontal movement during foaming, the expansion in all three dimensions is much more controlled and there is less tension in the sheet to lead to directionality. In both of the physically cross-linked roll foam processes described, the sheet surface quality is a distinguishing feature. The thin foam sheets have very high quality surfaces relative to the chemically cross-linked materials; however, they are generally of restricted thickness with the thickest products being of the order of 8 mm. Some of the advantages and disadvantages of the processes described above are found in Table 7.3. For further details of cross-linked polyolefin foam processes the interested reader is initially directed to References 12–15, which contain more comprehensive descriptions, information, and comparisons as well as further references.

7.6 Nitrogen Autoclave Cross-Linked Polyolefin Foam Products 7.6.1 Product Range The current commercial product range (under the AZOTE brand) covers effectively the entire spectrum of polyolefin materials. At one extreme are foams produced from co-polymers of ethylene such as ethylene-methyl acrylate (EMA) and ethylene-vinyl acetate (EVA) materials. At the other extreme are sheet foams produced from high-density polyethylene (HDPE) and polypropylene (PP). In the middle of this range there exist low-density polyethylene (LDPE) foams as well as metallocene polyethylene (mPE) foams and medium density polyethylene (MDPE) foams. In some cases the basic process described earlier must be modified to allow processing; for example in the case of the HDPE foams, the materials are cross-linked via an irradiation process since the extrusion temperatures for HDPE preclude the use of organic peroxides. The polyolefin products produced via the nitrogen autoclave technology range from soft, ultra-flexible materials through to semi-rigid and rigid,

61259_C007.indd 233

10/25/2008 12:39:02 PM

61259_C007.indd 234

Blowing Agent

Physical

Chemical

Nitrogen Autoclave Technology

Press Molding Technology

Block Foam Processes

Cross Linked Foam Processes

Chemical

Chemical or physical

Cross-Linking Agent

Hydraulic press

Autoclave

Expansion Hardware

 110

 60

Product Thickness (mm)

No chemical blowing agent (“clean” product), applicable to wide range of materials including engineering plastics, sheet size flexibility, density and cell size control via process, very high volume expansions possible, high quality structure, free expansion (isotropic sheet), key process stages are discrete from one another. Low capital cost, thick product (seen as a yield advantage), fine cell.

Advantages

Disadvantages

Product density variation, odor, pressures very high for low density product (press specification), mold and density variation impart property variations throughout sheet.

High capital cost, regulation of high pressure equipment, less economic for high density foam, product thickness capability is limited.

Comparison of the Major Technologies Used to Produce Low-Density Cross-Linked Polyolefin Foams

TABLE 7.3

234 Polymeric Foams

10/25/2008 12:39:02 PM

61259_C007.indd 235

Chemical

Chemical

Chemical

Furukawa Technology

Hitachi Technology

Toray Technology

Roll Foam Processes Sekisui Chemical Technology

Physical

Chemical

Chemical

Physical

Molten salt bath

Horizontal oven

Horizontal oven

Vertical oven

Relatively low capital cost, slightly thicker product possible, production cost lower at lower volume. Thin sheet, excellent surface quality, wider range of materials possible (inc. PP).

 1–7

Very thin sheet possible, energy efficiency (from vertical oven), excellent surface quality, vertical expansion enhances soft touch capabilities of the material (elongated structure is developed), wider range of materials possible (inc. PP). Relatively low capital cost, slightly thicker product possible, production cost lower at lower volume.

 5–16

 5–16

 1–8

Thickness limitations of product, salt bath safety concerns.

Surface quality can be poor. Poor cell structure. Choice of materials is more difficult. Temperature control is crucial to produce best quality. Surface quality can be poor. Poor cell structure. Choice of materials is more difficult. Temperature control is crucial to produce best quality.

Higher capital cost and production cost, thickness limitations to product, anisotropic (machine and cross-machine properties differ).

New Material Developments from the Nitrogen Autoclave Process 235

10/25/2008 12:39:02 PM

236

Polymeric Foams

energy absorbent foams. In all cases the range of density produced is between 15 kg/m3 and 120 kg/m3. Of note are the following product groups. 7.6.2 Cross-Linked High-Density Polyethylene Foams Traditional high-density polyethylene (HDPE) resins are a group of polyethylene resins which can be defined as being in the density range 950–970 kg/m3 produced using low-pressure polymerization processes. The resins typically have higher levels of crystallinity leading to improved mechanical properties. Also, the materials will typically have higher melt viscosities and extrusion processing temperatures in excess of 200°C are necessary. HDPE bead foams are well known but in cross-linked sheet form the material is less familiar due to the problems associated with the control of cross-linking and availability of suitable blowing agents to produce low-density foams. Table 7.4 shows a set of typical mechanical properties for the HDPE foams produced via the nitrogen autoclave technology. These low-density HDPE foams find applications in a number of areas where the structural rigidity and high-energy absorption properties are of benefit. 7.6.3 Cross-Linked Metallocene Polyethylene Foams In the early to mid-1990s a series of polyolefin materials were commercialized based upon metallocene catalyst technology. These methods of polymerization had been in research for many years. Foam converters, like many other polymer converters, were quick to evaluate the benefits of these materials. The products produced via the nitrogen autoclave process show a series of characteristics such as enhanced toughness, strength and durability, improved thermoformability, and reduced cell sizes. The impact of reduced cell sizes is mainly an aesthetic benefit, leading to a soft feel to the material. TABLE 7.4 Typical Mechanical Properties for the Cross-Linked Plastazote HDPE Foams Plastazote HD30

Plastazote HD60

Plastazote HD80

Plastazote HD115

Property

Method

Nominal density (kg/m3) Tensile strength (MPa) Compression stress (  25%) (kPa) Compression stress (  50%) (kPa) Flexural modulus (MPa) Tear strength (N/m)

ISO 845 ISO 7214 ISO 7214

30 1.01 158

60 1.72 326

80 2.09 523

115 2.39 791

ISO 7214

222

398

593

897

BS 4370 ISO 8067

3.6 1415

11.0 3525

15.0 4970

23.0 8310

61259_C007.indd 236

10/25/2008 12:39:02 PM

237

New Material Developments from the Nitrogen Autoclave Process

These materials have found application in thermoformed parts where high draw ratios and excellent definition are required and also in returnable packaging (dunnage) applications where enhanced toughness and durability are particularly valued. 7.6.4 Markets and Applications Cross-linked polyolefin foams from nitrogen autoclave technology find a variety of uses, covering all of the major market segments. Most notably, the highly consistent nature of the foams produced, in property, density and purity terms, means that they are the material of choice in high specification packaging. Briefly examining some of the major markets: 7.6.4.1 Packaging Typical examples of use would be for case inserts, display packaging, and highly specified cushion packaging. In a number of packaging applications, the primary reason for using the foam may be simply cost or aesthetics; however, in the case of cushion packaging the material must perform a much more serious purpose. Cushion curves (see typical example in Figure 7.9) are the basis of the traditional method of packaging design to transport delicate, high value, or fragile goods. The curves describe the ability of the foam to manage the deceleration of an article based upon the loading, the height from which the article is dropped (or in the case of design the height from which the article is likely to be dropped) and the thickness of the foam used to protect it.

Plastazote LD45 cushion curves First impact / Drop height = 1300 mm

200

Foam thickness = 25 mm

Peak deceleration (G)

180 160 140 120

Foam thickness = 50 mm

100 80

Foam thickness = 75 mm

60

Foam thickness = 100 mm

40 20 0

0

2

4

6

8 10 12 Static stress (kPa)

14

16

18

20

FIGURE 7.9 Typical cushion curves used in cushion packaging design (here showing curves for various thicknesses of a 45 kg/m3 cross-linked PE foam).

61259_C007.indd 237

10/25/2008 12:39:02 PM

238

Polymeric Foams

In closed-cell foams, the preferred structural response of the foam to such an impact is plastic deformation of the cell walls. To achieve this, semi-rigid foams tend to perform optimally. In more rigid foams, the cushioning performance is reduced. Although often the shock absorption behavior of such materials is good, in a shock impact situation the deceleration is not managed and the packaged article can suffer damage from a deceleration which is too rapid. Additional to this, more rigid foams also tend to suffer from higher compression set, which worsens performance in second or multiple drop situations. More flexible foams tend to absorb energy through pneumatic compression of the gas in the cells and often a large proportion of this absorbed energy is returned to the packaged article via rebound—in this case the deceleration/acceleration combination can also yield poor performance. Finally it is worth noting that a large volume of non-cross-linked foam, for instance EPS and LDPE, is also used in some high-specification cushion packaging applications. However, while this may be so, the more durable cross-linked LDPE foams are generally found to outperform non-crosslinked foams in the critical area of multi-impact performance. 7.6.4.2 Sports and Leisure This is a large market sector for cross-linked polyolefin foams and the entire range of these types of material are used, from soft through to rigid foams. Typical applications include body protection (for example in hockey), swimming aids and floats, camping mats, sports mats and the like. Generally the materials are valued for their low weight, durability, buoyancy, and impact absorption properties. 7.6.4.3 Medical/Health Care One of the major attributes of the products of the nitrogen autoclave technology is the ability to maintain a relatively pure foam product. As noted earlier, in the basic foam materials, following the cross-linking reaction, there is little else in the foam other than LDPE and nitrogen/air. This high purity level offers the materials a certain niche in the medical and health care sectors where the foams are used for cervical collars, orthotic support, diabetic shoe insoles, and so on. In many cases for splinting, the foam may be thermoformed directly to the body to produce a perfect fitting support. 7.6.4.4 Other In addition to the key market sectors noted above, there are very few industries that do not utilize low-density cross-linked polyolefin foams for some reason. Other examples include: • Marine industry for fenders, life jackets, and floating hoses • Building and construction industry for backer rods, expansion joints, eaves fillers, and underfloor thermal/acoustic insulation

61259_C007.indd 238

10/25/2008 12:39:02 PM

New Material Developments from the Nitrogen Autoclave Process

239

• Highly flame retardant grades are used in the aerospace industry for sealing, flotation aids, and duct insulation • Automotive industry for water barriers, gaskets/seals, vibration pads, and impact protection • Electronics and semiconductor industries, where electrically conductive and static dissipative foams are used for pin insertion packs, faraday cage shielding, tote box liners, and workstation mats.

7.7 New Material Developments From a material development perspective, efforts have diversified over recent years,16 away from developments focused upon polyolefin materials, to the study of the feasibility of foaming other polymers, typically those classed as “engineering polymers.” These material developments have again demonstrated the unique position and capability of the nitrogen autoclave technology within the industry. The commercial products that have been produced as a result are being marketed under the ZOTEK® brand, giving a range of high-performance foams based upon fluoropolymers, engineering plastics, and speciality elastomers. ZOTEK F is a patented17 range of low-density, closed-cell foams based upon the polymer polyvinylidene fluoride (PVDF). This is a remarkable material in that it offers some exceptional performance attributes along with ease of processability. Key attributes include temperature resistance, outstanding chemical and weathering resistance and exceptional flammability/smoke generation performance. ZOTEK N is a patented18 range of low-density closed-cell foams based on the engineering thermoplastic, polyamide. Initial products are based upon polyamide 6, also known as nylon 6 or PA6. The key attributes of these materials are the resistance to hydrocarbon fuels and oils allied to the high temperature performance and rigidity of the resins. In the following sections, these material types and some of the key attributes are examined in more detail.

7.7.1 ZOTEK F Polyvinylidene Fluoride Foams The current commercial grades of ZOTEK F are shown in Table 7.5 with a brief description. These materials are all based on the polymer PVDF and specifically Kynar® PVDF grades from Arkema Inc., with whom Zotefoams plc have a worldwide agreement to develop the products both technically and commercially.

61259_C007.indd 239

10/25/2008 12:39:02 PM

240

Polymeric Foams

TABLE 7.5 Current Commercial Grades of ZOTEK F PVDF Foam Product

Nominal Density

Attributes

ZOTEK F 30 ZOTEK F 38HT

30 kg/m3 (1.9 lb/ft3) 38 kg/m3 (2.4 lb/ft3)

ZOTEK F 74HT ZOTEK F 42HT LS

74 kg/m3 (4.6 lb/ft3) 40 kg/m3 (2.5 lb/ft3)

Low density, flexible, closed-cell foam Higher temperature, more rigid, closed-cell foam High density, higher temperature, more rigid Higher temperature, more rigid, closed-cell foam, low smoke

PVDF is in reality a large family of materials, similar to polyolefins, covering homopolymer grades and co-polymer grades which offer the customer a broad range of temperature stability and tailored properties. PVDF homopolymer is a rigid, crystalline polymer, melting at 171°C. Less relevant here but nonetheless interesting, the material has received some attention for the piezoelectric and pyroelectric behavior it exhibits; under certain circumstances the polymer can be made substantially more piezoelectric than crystalline quartz. By introducing comonomers such as hexafluoropropylene (HFP), tetrafluoroethylene (TFE), and chlorotrifluoroethylene (CTFE), a wide range of copolymers and terpolymers of vinylidene fluoride (VDF) have also been made possible, mostly yielding more elastomeric materials. The additional possibility of employing functional additives gives rise to an even wider range of material and performance options. The key properties of the PVDF polymer of interest to foam markets are those typical of the fluoropolymer family, namely excellent flame resistance, extremely low heat release and smoke generation, outstanding UV resistance, and broad resistance to chemical attack. Finally in addition to the above are the attributes generated from foaming the material such as buoyancy, impact absorption, and low thermal conductivity, yielding products with markedly different behavior to those of foams currently available. 7.7.1.1 Flammability Characteristics The ZOTEK F foams have recently been tested to some of the most stringent large-scale flammability specification tests that industry applies, covering markets such as aerospace, construction, semiconductor cleanroom, and at a laboratory level more generally applicable specifications such as UL94 V-0. The example of the ANSI/UL 723 (ASTM E 84-01) test standard is a good example to demonstrate foam performance. This standard is often referred to as the ‘Steiner Tunnel’ test and is used to classify the surface burning characteristics of building materials. The test results are reported as Flame Spread Index (FSI) and Smoke Developed Index (SDI) (see Table 7.6).

61259_C007.indd 240

10/25/2008 12:39:02 PM

New Material Developments from the Nitrogen Autoclave Process

241

TABLE 7.6 Results of Flame Spread Index (FSI) and Smoke Developed Index (SDI) for the ZOTEK F PVDF Foam Tested to ANSI/UL 723 (ASTM E-84) “Steiner Tunnel” Test Foam Thickness

FSI/SDI

Class

3.2 mm (1/8)

0/0

A

25.4 mm (1)

5/0

A

Values of less than 25/50, respectively, indicate that the material under test has very limited combustibility and offers the highest level of fire performance other than a non-combustible (non-organic) material. In order to achieve the lowest smoke release requirements, the LS (low smoke) foams were developed, which incorporate a small amount of a proprietary smoke suppressant. This material has been tested to the requirements of FM 4910 “Test Standard for FM Approvals Cleanroom Materials Flammability Test Protocol”19 and is the first polymeric foam to be specification tested, listed by FM Global, against this standard. In aerospace, regulatory changes to test protocols for the evaluation of materials for use as thermal/acoustic insulation in commercial aircraft have been introduced in recent years. These new regulations and the applicable test methods are described in the Code of Federal Regulations FAR §25.856.20 The ZOTEK F foams comply with the requirements of FAR 25.856(a), the so-called “radiant panel” test, and also demonstrate low heat release and smoke generation. In the aerospace industry, the combination of these FST (fire, smoke, and toxicity) properties along with low weight, flexibility, closed-cell structure (low moisture absorption) and ease of fabrication have led to a great deal of interest and development activity in this sector. Finally, although not officially certified, materials tested at laboratory level have indicated a UL94 V-0 rating at 13 mm, which is quite unusual for very low-density organic foams. 7.7.1.2 Environmental and Chemical Performance The PVDF raw material has traditionally been used as an additive in paints and coatings for exterior use because of the exceptional UV/weathering resistance of the polymer as well as the broad range of chemical resistance it possesses. These important properties have been maintained in the foam and as an example of this the ZOTEK F 30 foam has been subjected21 to several thousand hours of exposure in three separate accelerated UV/ weathering instruments. The conditions were set to simulate: • Indoor: temperature  55°C, 55% RH, wavelength 300–400 nm (glass filter) and very low power (36 W/m2).

61259_C007.indd 241

10/25/2008 12:39:03 PM

242

Polymeric Foams

• Outdoor: temperature  60°C, moisture cycling, wavelength 300–400 nm and low power (63 W/m2). • Severe/dry: temperature  60°C, dry, wavelength 300–400 nm and high power (400 W/m2). The foam was evaluated by a combination of mechanical (tensile), visual (microscopy) and color measurements (Yellowness Index). The results of the color measurements (see Figure 7.10) gave some mild bleaching of the foam surface within the first 200–300 hours of exposure followed by stabilization. The scale of the Yellowness Index should be noted with the color change barely perceptible to the naked eye. Samples were also evaluated in tensile tests and these demonstrated no significant change in tensile or elongation behavior. Finally, microscopy showed no evidence of crazing or cracking after the exposure time [ 6000 hours (250 days)], even after flexing of the samples. From a chemical resistance point of view the ZOTEK F foams were immersed in two separate “standard” fuels used in standard automotive test protocols and the results compared with those of cross-linked polyethylene (XL-PE) foam of equivalent density.6 Polyethylene foams are known to have poor resistance to diesel fuel and hydrocarbon oils/fuels so the results are perhaps not surprising for this material. The ZOTEK F 30 foam however was immersed for upward of 120 days and the equilibrium weight change after this period was of the order of 1–2% (see Figure 7.11).

10 CI 3000 Xenotest 1200 SEPAP 12-24

9 8 7 Yellowness Index

6 5 4 3 2 1 0 0 –1

500 1000 1500 2000 2500 3000 3500 4000 4500 5000 5500 6000 6500 7000 Exposure time (hours)

–2 FIGURE 7.10 Plot of Yellowness Index versus exposure time for the ZOTEK F PVDF foam exposed using three different accelerated weathering test methods.

61259_C007.indd 242

10/25/2008 12:39:03 PM

243

New Material Developments from the Nitrogen Autoclave Process

30 28

ZOTEK F30 / DIESEL FUEL 2D

30 kg/m3 XL-PE / DIESEL FUEL 2D

ZOTEK F30 / FUEL C

30 kg/m3 XL-PE / FUEL C

26 24

Weight change (%)

22 20 18 16 14 12 10 8 6 4 2 0 0

10

20

30

40

50 60 70 80 Immersion time (days)

90

100

110

120

130

FIGURE 7.11 Results for the ZOTEK F PVDF foam showing equilibrium weight change after immersion for >120 days in standard test fuels (comparison with XLPE foam of similar density).

Finally, under aerospace specification RTCA DO-160D, Cat. F, Section II, the materials were exposed to around 16 different types of fluids and fuels used in that industry. The specification is primarily an exposure test for component or functional systems that are to be used in aircraft. The tests undertaken on the foam involved exposure of the surface of the foam to the fluid (at temperatures ranging between 40°C and 150°C) for 24 hours. This third party testing confirmed that the ZOTEK F 30 and F 38HT foams comply with the requirements of this specification, again with little more than minor surface discoloration. 7.7.1.3 Basic Thermal Properties In addition to the outstanding FST properties and as a result of the highly consistent, closed-cell foam structure the thermal conductivity of the foams is low, of the order of 0.030–0.040 W/m K making them ideally suited for insulating applications. The ZOTEK F HT foams are also characterized by a relatively high melting point and therefore the foams are very dimensionally stable up to temperatures of around 150°C. Given the exceptional FST properties, high operating temperature and the low thermal conductivity, the ZOTEK F foams are not surprisingly finding increased usage for the insulation of air conditioning ducts, pipework, and other process industry installations.

61259_C007.indd 243

10/25/2008 12:39:03 PM

244

Polymeric Foams

7.7.1.4 Other Performance Attributes As demonstrated above, the ZOTEK F foams have outstanding FST, thermal, chemical and environmental resistance. Allied to this the foams have a closed-cell structure (Figure 7.12 shows a micrograph of the expanded foam structure) and are therefore buoyant, opening up potential uses as “sink-proof” flotation devices for use in harsh environments. The closedcell nature and ease of moldability also offers end-users a single material solution for insulation and moisture barrier that can be easily shaped and manipulated in situ. Finally the PVDF raw polymer is known to be of extremely high purity, giving rise to the widespread usage of the material in piping systems for high-purity water systems (typically used in the semiconductor manufacturing industry). The ZOTEK F foams, being blown with nitrogen gas, can likewise be considered to be high purity foams. As a means of demonstrating the high purity, the ZOTEK F foams have been tested in accordance with the requirements of ISO 10993: Biological Evaluation of Medical Devices. The materials were shown to pass the requirements of Pharmacopoeia Monograph USP 661 testing. In extraction tests conducted at the same time, using isopropanol and water as the extraction media, total maximum extractables were found to be of the order of 30 ppm, none of which were considered harmful in any way. The conclusion from this work is that the high-purity PVDF polymer gives rise to a high-purity PVDF foam, one that is suitable for use in medical devices for use in skin contact as well as those in contact with mucosal membranes or breached and compromised surfaces for limited or prolonged exposure (up to 30 days).

FIGURE 7.12

61259_C007.indd 244

SEM of the expanded ZOTEK F PVDF foam structure.

10/25/2008 12:39:03 PM

New Material Developments from the Nitrogen Autoclave Process

245

7.7.1.5 Markets and Applications As a result of the low density, inherent flame resistance, and low smoke properties of the product, the material has quickly been accepted by the mass transport industry. The ease of moldability, closed-cell nature (moisture barrier), and therefore the single material solution are of obvious benefit in thermal/acoustic insulation applications. A further very recent application development with the ZOTEK F grade foams is as insulation in cleanroom environments.22 This is being led by a product developed by UFP Technologies Inc. in the US. The products have been branded T-Tubes®.23 The foam is easily moldable and the product is typically formed into a clamshell configuration. The shaped foam is then simply fixed around pipework to both insulate and to protect personnel who often have to operate or maintain hot pipework and equipment within space limitations. Examples of the molded T-Tubes themselves are shown in Figure 7.13. Figure 7.14 then shows part of an installation with the T-Tubes fitted. The advantages of the T-Tubes approach are many and apart from material performance which is outstanding, they are extremely flexible, very clean and simple to clean once installed, easy to handle and easily formed/cut to shape on site. The latter is an important operational aspect which relates to speed and ease of installation without the production of dust, swarf, or other residues which could contaminate otherwise clean environments in the pharmaceutical, semiconductor, food, and other sensitive process industries. The T-Tubes themselves have been tested to both the FM 4910 test standard mentioned earlier as well as a further FM specification related to the

FIGURE 7.13

61259_C007.indd 245

Examples of the molded T-Tubes for cleanroom insulation.

10/25/2008 12:39:04 PM

246

FIGURE 7.14

Polymeric Foams

Cleanroom installation showing the T-Tubes fitted in position.

flammability of pipe insulation, namely FM 492424 and is proven compliant. It is believed that T-Tubes will be the only closed-cell foam insulation on the market supported by both approvals—truly a new industry standard for cleanroom insulation. 7.7.2 ZOTEK N Polyamide (PA 6) Foams The first ZOTEK N foam was launched commercially in late 2006. These materials remain in the market and application development phase of their lifecycle. The initial commercial grade, ZOTEK N B50, is a relatively stiff grade of foam based on the nylon 6 polymer. The solid has a density of around 1100 kg/m3 and the foam has a density of around 50 kg/m3 equating to a volume expansion of 22 times. Like PVDF, polyamides represent a very large family of resins and have also been widely exploited in blends and alloys with other polymers to modify specific properties. Nylon 6 is a relatively old polymer having been developed originally in the 1940s; however, it remains a polymer of significant importance in both fiber production and injection molding. The material is a rigid, crystalline polymer. Two important characteristics dictate the final performance of the material. The first is the level of crystallinity, which is highly dependent on the cooling regime used during processing and can vary considerably. The second is the level of moisture. Polyamides are hygroscopic in nature and equilibrium levels of moisture of 4–5% by weight are typical.

61259_C007.indd 246

10/25/2008 12:39:04 PM

247

New Material Developments from the Nitrogen Autoclave Process

The plasticization of the polymer with such levels of moisture can lead to profound property differences when compared to the “dry” product. Additives also provide a means to further tailor performance to particular application requirements. The key end-use attributes of nylon 6 are temperature performance, hydrocarbon oil/fuel resistance, toughness, and a good cost/performance balance. 7.7.2.1 Thermal Properties The temperature stability and continuous use temperature ranges of lowdensity foams are normally assessed via high temperature dimensional stability testing. The reason for this is related to the cross-linking/increasing molecular weight process step in most foaming processes. In the majority of such processes, the cross-linking or molecular weight increase must take place prior to foam expansion to ensure that there is adequate melt strength during expansion, thereby enabling a wide enough foam processing window. In the results presented in Figure 7.15, 100 ¥ 100 ¥ 25 mm samples of the foam were simply exposed to a range of temperatures in an air circulating oven for a 24-hour period and the linear dimensional change was assessed for each sample after a further period of stabilization at room temperature.

80 ZOTEK N B50 XL-PE (45 kg/m3)

75 70 65 Linear shrinkage (%)

60 55 50 45 40 35 30 25 20 15 10 5 0 0

25

50

75

100 125 150 Temperature (°C)

175

200

225

250

FIGURE 7.15 Linear shrinkage results for the ZOTEK N polyamide foam exposed to a range of temperatures (comparison with XL-PE foam of similar density).

61259_C007.indd 247

10/25/2008 12:39:04 PM

248

Polymeric Foams

In the XL-PE foam market, the widely accepted definition of the maximum operating temperature is the temperature at which 5% linear shrinkage is produced after the 24-hour exposure period. For the ZOTEK N B50 foam this point is around 210°C, some 100°C higher than most commercially available foams, including the equivalent density XL-PE foam used here for comparison. Insulation is one of the largest general applications of foam materials, making good use of the typically very low thermal conductivity (k) of such materials. The ZOTEK N B50 foam is no different in this respect and the thermal conductivity has been measured over a wide range of temperatures using the method described in ISO 8301. The results shown in Figure 7.16 show the change in thermal conductivity with mean test temperature. Of note is the reasonably flat response over the temperature range of 0 to 170°C. At ambient temperatures the k value is in the region of 0.038 W/m K, while at 170°C the material still has an extremely low k value of 0.051 W/m K. Clearly where high temperature insulation is required then this is an attractive proposition and due to the flat response over this temperature range one can envisage the displacement of other higher density or less robust insulation systems with the ZOTEK N foam.

0.060

Thermal conductivity, k (W/m K)

0.050

0.040

0.030

0.020

0.010

0.000 –20

0

20

40

60 80 100 120 Mean test temperature (°C)

140

160

180

200

FIGURE 7.16 Plot showing effect of mean test temperature on the thermal conductivity (ISO 8301) for the ZOTEK N polyamide foam.

61259_C007.indd 248

10/25/2008 12:39:05 PM

249

New Material Developments from the Nitrogen Autoclave Process

7.7.2.2 High Temperature Modulus and Moisture Thermal stability tests such as those described above are useful to a limited degree as they show only whether the material is relatively dimensionally stable but give little indication of the mechanical integrity at such temperatures. To shed more light on this aspect, dynamic mechanical thermal analysis (DMTA) testing was undertaken in flexural mode on the ZOTEK N B50 foam to evaluate the modulus over a range of temperatures in both “dry” and “conditioned” (equilibrium moisture) states. These results are shown in Figures 7.17 and 7.18. As noted previously, polyamides are hygroscopic and it is usual for these materials to be used in the conditioned state where properties may be considered relatively stable and at equilibrium. The dry and conditioned foam samples both show an expected glass transition temperature (Tg) of around 65°C, which is roughly unchanged by moisture level, although more pronounced in the dry material. In the range 10 to 10°C the effect of moisture is more clear—the dry sample yields almost no peak and very low tan d whereas in the case of the conditioned material, the peak is well defined and the peak position is shifted to the left on the axis. Of more academic interest, the dry samples show, as would be expected, a marked stiffening at ambient temperatures (relative to the conditioned samples) and it is not until a temperature of more than 75°C or so that the moduli of the dry and conditioned samples begin to converge. The lower tan d and less pronounced peak in the dry sample would suggest more

1.2×107

0.240 'Dry' ZOTEK N B50

0.220

7

1.0×10

0.200

9.0×106

0.180

8.0×106

0.160

7.0×106

0.140

6.0×106

0.120

5.0×106

0.100

4.0×106

0.080

3.0×106

0.060

2.0×106

0.040

6

0.020

1.0×10

tan d

Modulus (Pa)

1.1×107

0.000

0.0 –50

–25

0

25

50

75 100 125 Temperature (°C)

150

175

200

225

FIGURE 7.17 DMTA plots for the ZOTEK N polyamide foam in the dry state (curves show results of tests at frequencies of 1 Hz and 10 Hz).

61259_C007.indd 249

10/25/2008 12:39:05 PM

250

Polymeric Foams

1.2×107

0.240 'Conditioned' ZOTEK N B50

0.220

1.0×107

0.200

9.0×106

0.180

8.0×106

0.160

7.0×106

0.140

6.0×106

0.120

5.0×106

0.100

4.0×106

0.080

3.0×106

0.060

2.0×106

0.040

1.0×106

0.020

0.0 –50

–25

0

25

50

75 100 125 Temperature (°C)

150

175

200

tan d

Modulus (Pa)

1.1×107

0.000 225

FIGURE 7.18 DMTA plots for the ZOTEK N polyamide foam in the conditioned state (curves show results of tests at frequencies of 1 Hz and 10 Hz).

brittle behavior may be expected at or around ambient temperatures, a fact borne out by the basic mechanical results presented in Table 7.7. At around room temperature the modulus of the material using DMTA is 1.8 MPa. Applying a typical approach of 50% loss of modulus as an arbitrary limit for continuous use of the material, the plots define a useful upper temperature in the range 170–175°C. Few low-density foam materials in commercial use today are able to retain load-bearing capabilities at such temperatures. The results above have implications both for the material (i.e. use of the dry material at room temperature should be avoided or discouraged) and positive benefit in applications requiring load-bearing capability allied to light weight at high temperatures.

TABLE 7.7 Basic Mechanical Test Results for the ZOTEK N Polyamide Foam Showing Effect of Absorbed Moisture Property

Method

Dry

Compression stress @ 25% strain (kPa) Compression stress @ 50% strain (kPa) Tensile strength (MPa) Elongation at break (%) Tear strength (N/m)

ISO 7214

510

61259_C007.indd 250

ISO 7214 ISO 1798 ISO 1798 ISO 8067

605 1.54 60 1840

Conditioned 275 345 1.25 75 3000

10/25/2008 12:39:06 PM

New Material Developments from the Nitrogen Autoclave Process

251

The range of polyamide resin materials and alloys/blends available is very large. A key target for Zotefoams is to develop a more flexible polyamide foam product without significantly compromising the temperature resistance properties. Other functional properties such as recovery/ hysteresis are also being investigated. This work is progressing well and a family of polyamide foams under the ZOTEK N brand will certainly develop to cover the specific needs of various end-user markets. 7.7.2.3 Other Performance Attributes The descriptions above cover the main attributes that are derived from the polyamide polymer with regard to how they influence the behavior of the foamed product. Other characteristics, however, come solely from the use of the nitrogen autoclave process itself as a result of using a nitrogen pneumatogen. The foams tend to be low VOC (volatile organic compounds), low fogging, and low odor. These characteristics are of specific interest to the automotive sector for car interior applications. Finally, the impact performance of the ZOTEK N foams is yet to be fully evaluated but the rigid feel of the material is indicative of a material with high energy absorbing characteristics. 7.7.2.4 Markets and Applications Low-density polyamide foams are a new material for the design community to handle and to understand and for that reason much work is currently focused on application development and technical support activity. Key areas of ongoing effort are in the automotive sector where the use of the material in such applications as under-bonnet insulation and noise control is being evaluated. In the same industry the material is also being assessed in press-formed parts for spacers and seals (including wire and cable management) close to or within the engine compartment. Outside the automotive industry the ZOTEK N material has found interest in the composite-manufacturing sector as a relatively low-cost/hightemperature core material.

7.8 Conclusions The nitrogen autoclave process is a well-established process which is widely recognized within the industry as being a superior process technology. The technology has been producing the highest quality and most consistent cross-linked polyolefin foam products for four decades. The use of the technology continues to develop with the recent introduction of

61259_C007.indd 251

10/25/2008 12:39:06 PM

252

Polymeric Foams

the ZOTEK brand foams moving outside of the polyolefin family of materials for the first time commercially. These new and unique products offer the market and design community high performance through combinations of: • • • • • • • •

Outstanding flame, smoke, and toxicity performance Excellent UV resistance Broad chemical resistance Very low thermal conductivity Extremely high purity Outstanding high temperature dimensional stability Low thermal conductivity retained over wide temperature range High-temperature load-bearing capability.

References 1. Cooper, A. The Story of Expanded Rubber. Expanded Rubber Company Ltd., Croydon, U.K., 1957. 2. Eaves, D. E. “New foams from the nitrogen autoclave process.” Paper presented at the 2nd International Cellular Polymers Conference, Heriot Watt University, Edinburgh, 23–25 March 1993. 3. Lee, J. G. and Flumerfelt R. W. “Nitrogen solubilities in low-density polyethylene at high temperatures and high pressures.” Journal of Applied Polymer Science 58 (1995): 2213–2219. 4. Sato, Y. Fujiwara, K., Sumarno, T. S., and Masuoka, H. “Solubility of carbon dioxide and nitrogen in polyolefins and polystyrene under high pressures and temperatures.” Presented at the 5th Meeting on Supercritical Fluids and Natural Products Processing, Nice, France, 1998. 5. Crank, J. The Mathematics of Diffusion, 2nd edition. Clarendon Press, Oxford, 1975. 6. Park, C. B., Baldwin, D. F., and Suh, N. P. “Effect of pressure drop rate on cell nucleation in continuous processing of microcellular polymers.” Polymer Engineering and Science 35 (1995): 432–440. 7. Kim, S. G., Lee, J. W. S., Park, C. B., and Sain, M. “Strategies for enhancing cell nucleation of thermoplastic polyolefin (TPO) foam.” In Proceedings of the Annual Technical Conference (ANTEC), Society of Plastics Engineers (SPE), Cincinnatti, OH, May 2007: 2099–2104. 8. Rodriguez-Perez, M. A., Gonzalez-Pena, J. I., Witten, N., and de Saja, J. A. “The effect of cell size on the physical properties of crosslinked closed cell polyethylene foams produced by a high pressure nitrogen solution process.” Cellular Polymers 21 (2002): 165.

61259_C007.indd 252

10/25/2008 12:39:06 PM

New Material Developments from the Nitrogen Autoclave Process

253

9. Eaves, D. E. “The properties of crosslinked foams produced from metallocene polyolefins.” Presented at the Metallocene Technology Seminar, RAPRA Technology Ltd., Shawbury, U.K., 2 September 1997. 10. US 59955015 [Evolution Foam Mouldings Ltd.] 11. Moore, S. “Crosslinking technique is low-cost and efficient.” Modern Plastics (2002): 42. 12. Klempner, D. and Frisch, K. C. Handbook of Polymeric Foams and Foam Technology. Hanser, Munich, Germany, 1991. 13. Eaves, D. E. Handbook of Polymer Foams. RAPRA Technology Ltd., Shawbury, U.K., 2004. 14. Trageser, D. A. “Crosslinked polyethylene foam processes.” Radiation Physics and Chemistry 9 (1977): 261–270. 15. Puri, R. R. and Collington, K. T. “The production of cellular crosslinked polyolefins: Part 2—The injection moulding and press moulding techniques.” Cellular Polymers 7 (1988): 219–231. 16. Witten, N. “Extending the conventional boundaries of crosslinked polyolefin foam production.” Presnted at FOAMPLAS, Teaneck, NJ, 19–20 May 1998. 17. WO 2005/105907 (Zotefoams plc). 18. WO 2006/077395 (Zotefoams plc). 19. See http://www.fmglobal.com/assets/pdf/fmapprovals/4910.pdf. 20. See http://www.gpoaccess.gov/index.html. 21. Van der Weide, I., and Werth, M. High Purity Plus Outstanding Resistance— PVDF Foams. Kunstoffe, Munich, Germany, 2006, p. 54. 22. Partridge, R. “Fungal, chemical, and fire resistance of PVDF foams and polymers.” Controlled Environments (2007): 14–18. Available at: http://www. cemag.us/articles.asp?pid=743. 23. See http://www.t-tubes.com. 24. See http://www.fmglobal.com/assets/pdf/fmapprovals/4924.pdf.

61259_C007.indd 253

10/25/2008 12:39:06 PM

61259_C007.indd 254

10/25/2008 12:39:06 PM

8 Polystyrene Foam and Its Improvement in Vacuum Insulated Panel Insulation Chang-Ming Wong

CONTENTS 8.1 Introduction of Vacuum Insulation Panel ................................. 8.2 PS Characteristics ......................................................................... 8.3 PS Foaming ................................................................................... 8.4 Heat Transfer in Plastic Foams ................................................... 8.5 Thermal Conductivity of VIPs Using PS Foams as Core Materials ............................................................. 8.6 Conclusions ................................................................................... 8.7 Abbreviations ................................................................................ 8.8 Nomenclature ............................................................................... References .............................................................................................

255 259 260 272 279 285 285 286 287

8.1 Introduction of Vacuum Insulation Panel Plastics containing many cells or bubbles are called foamed plastics or plastic foams. Plastic foams have many unique characteristics such as a light weight, shock absorption, and good insulation. Generally, the cell structures of plastic foams can be classified into either a closed-cell structure or an open-cell (porous) structure. A closed-cell structure, where a

61259_C008.indd 255

10/25/2008 12:39:54 PM

256

Polymeric Foams

Multi layer-films Vacuum

(a)

Desiccant(s)/getters Core material (b)

FIGURE 8.1 Schematic representation of a VIP.

large proportion of the cells are independent of each other, is easy to produce during foaming. However, an open-cell structure, where a large proportion of the cells are connected to other cells by open passage, can be more difficult to achieve during foaming. Vacuum insulation panels (VIPs) are a new generation product for thermal insulation and have only been produced in the past ten years. With the price of crude oil increasing, energy saving is an important issue nowadays. VIPs mainly consist of an open-cell material and multilayer barrier films, the open-cell material as a core material being encapsulated by multilayer films. Figure 8.l schematically illustrates VIP structure. The outward appearance of VIP looks like a pouch, as shown in Figure 8.1a. The pouch is then evacuated to a vacuum ranging from 1.33 Pa (0.01 torr) to 133.3 Pa (1 torr) and sealed afterwards to form a VIP, as illustrated in Figure 8.1b. An addition to a core material and multilayer film, getter(s) and/or desiccant(s) are also placed in the VIP to ensure the performance of the VIP during the projected lifetime. The purpose of a getter is to absorb the diffusing gases, while the desiccant is for absorbing moisture from the core material to maintain required vacuum levels in the VIP.1 Several porous materials such as open-cell polystyrene (PS) foam,2–14 open-cell polyurethane (PU) foam,15–17 aerogel,18,19 or silica powder20–22 can be used as core materials in VIPs. Basically, these porous materials have a very high content (more than 95%) of open-cells or pores. Multilayer films mainly include a protective layer (external layer), barrier layer (core layer), and sealing layer (inside layer) and are made by lamination or coextrusion.23–25 The protective layer is responsible not only for impact resistance, scratch resistance, and chemical resistance, but also for optical properties. The barrier layer provides multilayer films with a low diffusion for gas or moisture from the atmosphere. The sealing layer is primarily responsible for sealing performance in multilayer films. Multilayer films used in VIPs can be classified into three types: 1. Those containing an aluminum (Al) foil as a barrier layer, an outermost film as a protective layer such as nylon (NY) or polyethylene terephthalate (PET) or metallized PET film and an innermost film as a sealing layer such as high-density polyethylene (HDPE) film;

61259_C008.indd 256

10/25/2008 12:39:55 PM

257

Polystyrene Foam and Its Improvement

Protect layer

Protect layer

Nylon film

Nylon film Adhesive

Adhesive PET film

PET film Adhesive

Adhesive EVOH foil

Aluminum foil

Adhesive

Adhesive HDPE film

HDPE film

Sealing layer

Sealing layer

(a)

(b) Protect layer Polyester Metallization Adhesive Metallization Polyester Adhesive HDPE film Sealing layer (c)

FIGURE 8.2 Schematic representation of multilayer films.

for example, NY (15 μm)/PET (12 μm)/Al foil (5–25 μm)/HDPE (15 μm) shown in Figure 8.2a. 2. Those containing only polymeric films, and ethylene vinyl alcohol (EVOH) as a barrier material, such as NY/PET/EVOH/HDPE described in Figure 8.2b. 3. Those containing metallized films as a barrier layer without Al foil; for instance, metallized polyester film/sealant film illustrated in Figure 8.2c. The thermal conductivity of PS foams at the condition of one atmosphere is approximately 0.03 W/m.k.26 When the internal pressure in VIP reduces to 13.3 Pa (0.1 torr), the thermal conductivity of VIP can be lowered to between 0.009 W/m k (R  16) and 0.006 W/m k (R  24). However, the plastic foams used in VIP must not only have a high content of open cells, but also have a good mechanical strength. Figure 8.3 displays the photograph of a poor VIP. The porous PS foam is unable to maintain the original dimension under vacuum packing due to a poor mechanical strength of the porous PS foam. Figure 8.4 presents the photograph of a normal VIP. The porous PS foam has a good mechanical strength to keep the original size under vacuum packaging.

61259_C008.indd 257

10/25/2008 12:39:55 PM

258

Polymeric Foams

FIGURE 8.3 Poor mechanical property of a porous PS foam as a core material for a VIP.

Because multilayer films used in VIPs are unable to maintain the performance in a high-temperature environment, VIPs are mainly introduced into refrigerating applications such as shipping containers, reefers, and refrigerators for housing, boat, vehicles, and hotels. VIPs are generally three to seven times better in thermal insulation at an equivalent thickness than conventional products such as PU foam (R ~ 5.5), PS foam (R ~ 4), and fiberglass batting (R ~ 3.3). Commercial goods using vacuum insulation use less thickness than conventional insulation under the same thermal insulation. Therefore, refrigerators using VIP can increase internal storage capacity in restricted spaces. VIPs can also reduce electrical power consumption and decrease

FIGURE 8.4 Good mechanical property of a porous PS foam as a core material for a VIP.

61259_C008.indd 258

10/25/2008 12:39:55 PM

Polystyrene Foam and Its Improvement

259

coolant requirements. PU foam is a basic material for thermal insulation in refrigerators. When 40% PU foam in the wall of a refrigerator is replaced with PS/VIPs, the energy factor (L/KWhr/month) for the refrigerator can be enhanced by about 30%.27 Many companies currently apply vacuum insulation technology for manufacturing refrigerators or shipping containers. Examples include Sharp (Japan), Sanyo (Japan), Matsushtia (Japan), AEG-Electrolux (Europe), Glacier Bay (USA), and Acutemp (USA).

8.2 PS Characteristics The structure of styrene (phenyl ethylene or vinyl benzene) is CH2 CH2C6H5. Styrene monomer can polymerize to form a homopolymer or polymerize with other monomers such as acrylonitrile, butadiene, and alpha-methyl styrene to become copolymers or terpolymers. The homopolymer, copolymers, and terpolymers are called styrene-based polymers. Generally, there are two popular types of PS available in the market, one is general-purpose PS (GPPS)28,29 and the other is high-impact PS (HIPS).28,30,31 GPPS is formed by a styrene monomer under free-radical polymerization. PS produced by free-radical polymerization has an atactic configuration that leads to an amorphous polymer with a glass transition temperature (Tg) of approximately 100°C. Moreover, styrene polymer synthesized by metallocene catalysts can yield syndiotactic PS (sPS), which is a crystalline polymer with a melting temperature around 270°C and a Tg of approximately 100°C.32,33 GPPS basically has three commercial grades: easy flow, medium flow, and high heat. • Easy-flow resins have the lowest molecular weight than the other two grades and contain 3–4% mineral oil in resin to reduce melt viscosity. In addition to the application of injection molding, the resins can be used for coextruded packaging applications. • Medium-flow resins generally contain 1–2% mineral oil and have melt flow properties intermediate between the other grades. Applications for medium-flow resins also include injectionmolded products and extruded or coextruded food packaging. • High-heat resins are the highest molecular weight resins and contain no mineral oil and low concentrations of additives such as mold-release and extrusion aids. The major applications for the resins are foam, sheet, and film extrusion. As a matter of fact, GPPS is brittle in nature and the impact strength is poor, if unoriented. In order to improve the defect, styrene is grafted with

61259_C008.indd 259

10/25/2008 12:39:55 PM

260

Polymeric Foams

TABLE 8.1 Characteristics of Polystyrene Resins GPPS Properties (Units) Melt flow index (g/10 min) Specific gravity Tg (DSC) (°C) HDT (°C) Tensile strength (kg/cm2) Tensile elongation (%) Izod impact strength (kg-cm/cm) (notched, 1/4”) Mw (103)

HIPS

PS-1 2.2 1.05 107 86 540 2.0 1.7

PS-2 5.0 1.05 97 83 480 2.0 1.6

PS-3 8.0 1.05 95 78 440 2.0 1.5

PS-4 3.0 1.05 103 83 240 45 10.8

280

264

253

225

a rubber such as polybutadiene to become HIPS. HIPS resins are much softer and have better impact strength than GPPS resins. This is due to the existence of rubber particles dispersed in a continuous matrix of polystyrene. At a constant rubber level, increasing rubber particle sizes from average diameter 0.6 to 3.5 m in HIPS can enhance Izod impact strength of HIPS from 48 to 100 J/m.34 However, if the rubber particles have diameters much larger than 10 m in HIPS, the finished product has a low gloss surface and toughness decreases. Table 8.1 shows the material characteristics of four PS resins including three GPPS and one HIPS (PS-4). The Tg is measured by a differential scanning calorimeter (DSC). Generally, a higher weight average molecular weight (Mw) for GPPS has a higher heat distortion temperature (HDT), tensile strength, and impact strength. However, impact strength is poor for GPPS and is not dramatically influenced by Mw. The material properties of HIPS do not have any relationship with Mw and the impact strength is much better than that of GPPS.

8.3 PS Foaming PS can be foamed by chemical foaming agents (CFAs) such as azodicarbonamide derivative and physical foaming agents (PFAs) such as carbon dioxide (CO2), nitrogen (N2), hydrocarbons, and hydrofluorocarbons. Batch and continuous process are general methods to manufacture PS foam. When PS is foamed by CFAs, the foam density of PS foams normally ranges from 600 kg/m3 (0.6 g/cm3) to 800 kg/m3 (0.8 g/cm3) and a closedcell structure exists in PS foams. PFAs used in foaming process can create PS foams with density less than 100 kg/m3 (0.1 g/cm3). The cell structures

61259_C008.indd 260

10/25/2008 12:39:55 PM

261

Polystyrene Foam and Its Improvement

in PS foam made by PFAs are either a closed-cell structure or an open-cell (porous) structure. There are many theoretical and experimental works associated with PS foamed by CO2 under supercritical conditions in either a batch or a continuous process.35–43 There are two types of batch process used to make microcellular foam. One is that a resin is soaked with foaming agents at room temperature or a little higher than room temperature and high pressure, and then the resin is subjected to a high-temperature environment to form a microcellular foam; the other is that a resin is saturated with foaming agents at high temperatures and high pressure, and then the resin is foamed at a high pressure drop rate to obtain a microcellular foam. The first method has a much longer saturation time than the second method. Figures 8.5 and 8.6 schematically describe the batch and continuous foaming system respectively. PS foam with a cell size of less than 10 m can be obtained by a batch process. It is difficult to obtain a PS foam with cell size smaller than 10 μm in a continuous process because the pressure drop rate for a continuous process is much smaller than that for a batch process, and the uniform foaming temperature in foaming materials for a continuous process is more difficult to control.40–43 Basically the cells in most neat microcellular PS foams produced by CO2 have a closed-cell structure. Since a foaming system using PFAs is better at making open-cell PS foam, several PFAs are generally applied to foaming systems, for example, aliphatic hydrocarbon with four to six carbon atoms2 or a mixture of gas such as CO2, ethyl chloride (EtCl), and partially halogenated aliphatic hydrocarbons including chlorodifluoromethane (HCFC-22), 1-chloro1,1-difluoroethane (HCFC-142b), 1,1,1,2-tetrafluoroethane (HFC-134a), or 1,1-difluoroethane (HFC-152a).3–11

1,2 Tank

Upper heater

3,4 Stable pressure unit 5,6 Injection unit P1 P3 Upper die Lower die 1

P2

5 3 6 7

4 2

Lower heater

FIGURE 8.5 Schematic representation of the batch foaming system.

61259_C008.indd 261

10/25/2008 12:39:55 PM

262

Polymeric Foams

Hopper Extruder Gas inlet Adaptor Die Motor & gear drive Polymer melting

Gas/polymer mixing cooling

Foaming shaping

(a)

Hopper

Extruder Gas inlet

Motor & gear drive Polymer melting

Gas/polymer mixing

Extruder

Adaptor Die

Motor & gear drive Gas/polymer cooling (b)

Foaming shaping

FIGURE 8.6 Schematic representation of the continuous foaming system: (a) one extruder; and (b) two extruders.

Blending of PS and other materials such as styrene/ethylene butylene/ styrene (SEBS),8 ethylene/octane copolymer (EO),8 and ethylene/styrene interpolymer (ESI),9–11 is also an important method of producing open-cell PS foams because SEBS, EO, and ESI can act as cell-opening agents during PS foaming. A theoretical study indicates that the development of an opencell structure includes two stages, namely bubble growth to impingement and then cell wall thinning to rupture.44 Figure 8.7 presents the foam density of four neat PS foams at various foaming temperatures and a foaming pressure of 20.7 MPa (3000 psi). The four neat PS sheets are foamed by CO2 as the foaming agent at foaming temperatures ranging from 110°C to 140°C in a batch process. The material characteristics of four heat PS resins are shown in Table 8.1. Generally, the four neat PS foams have a high foam density at lower foaming temperatures. The PS-1 foam always displays a higher foam density than PS-2, PS-3, and PS-4 foams. PS-4, HIPS resin has a lower Mw than PS-2 and PS-3 resins, but the foam density of the PS-4 resin is higher than that of the PS-2 and PS-3 resins. When the foaming temperature ranges between 110°C and 125°C, the foam density of the PS-1 resin is much higher than

61259_C008.indd 262

10/25/2008 12:39:56 PM

263

Polystyrene Foam and Its Improvement

120 PS

Density (kg/m3)

100 80 60 PS-1 PS-2 PS-3 PS-4

40 20 0 100

110

120 130 Temperature (°C)

140

150

FIGURE 8.7 Dependence of PS foam density on foaming temperatures.

that of the other three PS resins. As the foaming temperature rises above 130°C, the difference of foam density for four PS resins becomes narrow. Lower foam density for all four of the PS resins is observed in the range 130°C to 140°C. Figure 8.8 illustrates the foam density of four PS resins blended with 2 phr CaCO3 (two parts CaCO3 per hundred parts of PS resin by weight) at various foaming temperatures. A higher foam density at lower foaming temperatures is observed for the four PS/2 phr CaCO3 foams. The foam density strikingly increases at the foaming temperature of 140°C. In comparison with the foam density of the four neat PS resins in Figure 8.7, the occurrence of a low foam density for the four PS/2 phr CaCO3 compounds shifts to low foaming temperatures ranging from 125°C to 135°C. The situation indicates that CaCO3 is able to change the viscoelastic property of PS resins typically seen with inorganic fillers.45,46 The foam density of four PS resins blended with 4 phr CaCO3 and 6 phr CaCO3 are exhibited in Figures 8.9 and 8.10, respectively. The low foam 250 Density (kg/m3)

PS/2 phr CaCO3 200 150 100 50 0 100

PS-1 PS-2 PS-3 PS-4 110

120

130

140

150

Temperature (°C) FIGURE 8.8 Dependence of PS/2 phr CaCO3 foam density on foaming temperatures.

61259_C008.indd 263

10/25/2008 12:39:56 PM

264

Polymeric Foams

400

Density (kg/m3)

350

PS/4 phr CaCO3

300 250 200 150

PS-1 PS-2 PS-3 PS-4

100 50 0 100

110

120 130 Temperature (°C)

140

150

FIGURE 8.9 Dependence of PS/4 phr CaCO3 foam density on foaming temperatures.

density for the four PS/4 phr CaCO3 foams and most PS/6 phr CaCO3 foams occurs at foaming temperatures of between 120°C and 130°C. Moreover, the foaming temperatures having a low foam density for PS-3/6 phr CaCO3 foam become narrow, but are still near 120°C. This indicates more CaCO3 in PS resin cannot further decrease the foaming temperatures at which a low foam density of PS resin occurs. The foam density at the foaming temperature of 140°C for the four PS/6 phr CaCO3 foams is much higher than that for the four PS/4 phr CaCO3 and PS/2 phr CaCO3 foams. Generally, the more CaCO3 is added to the PS resin, the higher PS foam density is obtained during PS foaming. The cells of the four neat PS foams and PS/2 phr CaCO3 foams at the foaming temperature of 125°C are shown in Figures 8.11 and 8.12, respectively. Figure 8.11a–d shows PS-1, PS-2, PS-3, and PS-4 foams, respectively, and Figure 8.12a–d shows PS-1/2 phr CaCO3, PS-2/2 phr CaCO3, PS-3/2

Density (kg/m3)

700 PS/6 phr CaCO3

600 500 400 300

PS-1 PS-2 PS-3 PS-4

200 100 0 100

110

120

130

140

150

Temperature (°C) FIGURE 8.10

61259_C008.indd 264

Dependence of PS/6 phr CaCO3 foam density on foaming temperatures.

10/25/2008 12:39:56 PM

Polystyrene Foam and Its Improvement

265

FIGURE 8.11 Cells in PS foams at the foaming temperature of 125°C: (a) PS-1; (b) PS-2; (c) PS-3; and (d) PS-4.

phr CaCO3, and PS-4/2 phr CaCO3 foams, respectively. Few open cells are generally observed in either the four PS or four PS/2 phr CaCO3 foams at the foaming temperature of 125°C. Figures 8.13 and 8.14 describe the cells of four neat PS foams and PS/2 phr CaCO3 foams at the foaming temperature of 135°C, respectively. Figure 8.13a–d represents PS-1, PS-2, PS-3, and PS-4 foams, respectively. Only a few pores exist in each of the four foams. The content of open cells for PS-3 foam is slightly higher than that of the other foams and the porous structure is scarcely observed in the PS-4 foam. PS-1/2 phr CaCO3, PS-2/2 phr CaCO3, PS-3/2 phr CaCO3, and PS-4/2 phr CaCO3 foams are indicated in Figure 8.14a–d, respectively. The content of pores in the PS/2 phr CaCO3 foams is significantly high in comparison with that in the neat PS foams. In particular, PS-2/2 phr CaCO3 and PS-3/2 phr CaCO3 foams have a very high content of open cells. Therefore, CaCO3 can act as a cell-opening agent during PS foaming.47 Figures 8.15 and 8.16 show the average cell sizes of four neat PS and PS/2 phr CaCO3 foams at the foaming temperatures of 125°C and 135°C, respectively. In general, the average cell sizes of the three neat GPPS at the foaming temperature of 135°C are slightly larger than that at the foaming temperature of 125°C. However, the average cell size of neat HIPS foam at

61259_C008.indd 265

10/25/2008 12:39:57 PM

266

Polymeric Foams

FIGURE 8.12 Cells in PS/2 phr CaCO3 foams at the foaming temperature of 125°C: (a) PS-1; (b) PS-2; (c) PS-3; and (d) PS-4.

the foaming temperature of 135°C is much larger than that at the foaming temperature of 125°C. The average cell sizes of three GPPS/2 phr CaCO3 foams at the foaming temperature of 135°C are, on the contrary, smaller than that at the foaming temperature of 125°C, but the average cell sizes of HIPS/2 phr CaCO3 foam are very close at the two foaming temperatures of 125°C and 135°C. The three GPPS/2 phr CaCO3 foams have much smaller average cell sizes than the three neat GPPS foams. The foam density of four PS/2 phr LDPE, PS/5 phr LDPE, and PS/7 phr LDPE foams are presented in Figures 8.17, 8.18, and 8.19, respectively. The trend of foam density for PS blended with LDPE at the foaming temperatures from 110°C to 140°C is similar to the trend of foam density for PS blended with CaCO3 indicated in Figures 8.8–8.10. A PS/LDPE foam has a higher foam density at the same foaming temperature than a neat PS foam. A striking rise in foam density for the four PS/LDPE foams is observed at the foam temperature of 140°C, the occurrence of a low foam density for the four PS/LDPE foams shifts to low foaming temperatures ranging from 120°C to 130°C. The results indicate both LDPE and CaCO3

61259_C008.indd 266

10/25/2008 12:39:57 PM

Polystyrene Foam and Its Improvement

267

FIGURE 8.13 Cells in PS foams at the foaming temperature of 135°C: (a) PS-1; (b)PS-2; (c) PS-3; and (d) PS-4.

can cause different grades in the shrinkage of PS foams and change the viscoelastic property of PS resins. The cells of four PS blended with 2 phr LDPE and 7 phr LDPE foams at the foaming temperature of 125°C are described in Figures 8.20 and 8.21, respectively. Figure 8.20a–d represents PS-1/2 phr LDPE, PS-2/2 phr LDPE, PS-3/2 phr LDPE, and PS-4/2 phr LDPE foams, respectively. Small cell sizes for four PS/2 phr LDPE foams are observed and an open-cell structure exists in some PS/2 phr LDPE foams. Figure 8.21a–d illustrates PS-1/7 phr LDPE, PS-2/7 phr LDPE, PS-3/2 phr LDPE, and PS-4/2 phr LDPE foams, respectively. The cell sizes for the four PS/7 phr LDPE foams are close to that for the four PS/2 phr LDPE foams. However, open-cell structure slightly increases for PS/7 phr LDPE foams. Figure 8.22 shows the comparison of average cell sizes for four neat PS and PS/LDPE foams produced at the foaming temperature of 125°C. The average cell sizes of four PS blended with 2 phr, 5 phr, and 7 phr LDPE foams are around 10 m and are much smaller than that of four neat PS foams. The results indicate that PS blended with LDPE can lead to the reduction of average cell size of PS foam during PS foaming.

61259_C008.indd 267

10/25/2008 12:39:58 PM

268

Polymeric Foams

FIGURE 8.14 Cells in PS/2 phr CaCO3 foams at the foaming temperature of 135°C: (a) PS-1; (b) PS-2; (c) PS-3; and (d) PS-4.

Cell sizes (mm)

80

PS

60 40 125°C 135°C

20 0 PS-1

PS-2 Materials

PS-3

PS-4

FIGURE 8.15 Average cell sizes of four PS foams at foaming temperatures of 125°C and 135°C.

61259_C008.indd 268

10/25/2008 12:39:58 PM

269

Polystyrene Foam and Its Improvement

60 PS/2 phr CaCO3

Cell sizes (mm)

50 40 30 20

125°C 135°C

10 0

FIGURE 8.16 and 135°C.

PS-1/CaCO3

PS-2/CaCO3 PS-3/CaCO3 Materials

PS-4/CaCO3

Average cell sizes of four PS/2 phr CaCO3 at foaming temperatures of 125°C

400

Density (kg/m3)

350

PS/2 phr PE

300 250 200 150

PS-1 PS-2 PS-3 PS-4

100 50 0 100

FIGURE 8.17

110

120 130 Temperature (°C)

140

150

Dependence of PS/2 phr LDPE foam density on foaming temperatures.

600 PS/5 phr PE

Density (kg/m3)

500 400 300 PS-1 PS-2 PS-3 PS-4

200 100 0 100

FIGURE 8.18

61259_C008.indd 269

110

120 130 Temperature (°C)

140

150

Dependence of PS/5 phr LDPE foam density on foaming temperatures.

10/25/2008 12:39:58 PM

270

Polymeric Foams

500 PS/7 phr PE

Density (kg/m3)

400 300 200 100 0 100

FIGURE 8.19

PS-1 PS-2 PS-3 PS-4 110

120 130 Temperature (°C)

140

150

Dependence of PS/7 phr LDPE foam density on foaming temperatures.

FIGURE 8.20 Cells in PS/2 phr LDPE foams at the foaming temperature of 125°C: (a) PS-1; (b) PS-2; (c) PS-3; and (d) PS-4.

61259_C008.indd 270

10/25/2008 12:39:59 PM

271

Polystyrene Foam and Its Improvement

FIGURE 8.21 Cells in PS/7 phr LDPE foams at the foaming temperature of 125°C: (a) PS-1; (b) PS-2; (c) PS-3; and (d) PS-4.

70 125°C foam

Cell sizes (mm)

60

PS PS/2 phr LDPE PS/5 phr LDPE PS/7 phr LDPE

50 40 30 20 10 0

PS-1

PS-2

PS-3

PS-4

Material FIGURE 8.22 Average cell sizes of four neat PS and PS/LDPE foams at the foaming temperature of 125°C.

61259_C008.indd 271

10/25/2008 12:39:59 PM

272

Polymeric Foams

8.4 Heat Transfer in Plastic Foams Plastic foams have a lower thermal conductivity than neat plastics. Plastic foams are used in industrial or daily goods are for their heat insulation. Therefore, the insulation property is an important characteristics of plastic foams. Heat transfer in plastic foams has been widely studied. The total thermal conductivity, l, of plastic foams is the sum of the thermal conductivity contributed from solid, λs, the thermal conductivity contributed from gas, λg, the thermal conductivity contributed from radiations, λr, and the thermal conductivity contributed from convection, λc. The formula equation can be written as

λ  λs  λg  λr  λc

(8.1)

The thermal conductivity through solids depends on the cell structure. However, it is difficult to describe the exact cell structure of the plastic foam because the cell structure is diverse from cell to cell. The cell structure is not spherically shaped, but the pentagonal dodecahedron, as depicted in Figure 8.23, is generally the most common shape for cells. The cells consist of cell walls formed by thin polymeric membranes and struts formed by the impingement of several walls at the intersecting position.48 Figure 8.24 shows the cell structure of a PS foam with foam density around 50 kg/m3 . The thickness of cell walls is less than 10 m, but struts are much thicker. Based on a plastic foam with isotropic structure, References 49 and 50 propose the equation for the solid conductivity (λs) as follows: fs 2  __ λs  __ (1   )λo 3 3

FIGURE 8.23

61259_C008.indd 272

(8.2)

Schematic of dodecahedral structure for a foam cell.

10/25/2008 12:40:00 PM

273

Polystyrene Foam and Its Improvement

FIGURE 8.24

Cell structure of PS foam.

r o - r a - rf   __________ r o - rg

(8.3)

and Reference 51 simplifies Equation 8.2 to become:

f fs __ 2  __ λs  __ λ 3 3 o o

(

)( )

(8.4)

where fs is the mass fraction of the struts in the foam;  is the void fraction or porosity of the foam; λo is the solid conductivity of the bulk material; f is the foam density; o is the density of the bulk material; a is the air density; g is the gas density inside the cell. If a plastic foam is a non-isotropic structure; that is, cells with different structures in parallel and perpendicular directions to heat flux, Equation 8.2 can be expressed as:

(

)

fs (1  )λ 2  __ λs  __ o 3 3

(8.5)

where  is a non-isotropic factor.49 Typically fs and  can be deduced from scanning electron microscopy (SEM) pictures of foams and fs in plastic foams is around 0.6–0.9.51 Therefore, the thermal conductivity through solid for plastic foams relates the void fraction of the foam or foam density, the mass fraction of struts, and the solid conductivity of the bulk material. The void fraction of the foam or foam density is dominant factor for solid conductivity. When foam density increases or the void fraction of the foam decreases, the solid

61259_C008.indd 273

10/25/2008 12:40:00 PM

274

FIGURE 8.25

Polymeric Foams

Thermal conductivity of gases.

conductivity rises. The solid conductivity is not a constant and decreases with the reduction of temperature. However, the change in the solid conductivity with temperature can be considered small.52 The thermal conductivity through gas depends on the trapped gas.50,53–55 The cells of a plastic foam are initially full of a foaming agent (gas) or a mixture of foaming agents. The composition of gas in the cells then changes as time elapses due to the infusion of air from the atmosphere and the effusion of the foaming agents. The process can lead to an overall increase in the thermal conductivity of the plastic foam, since air has a greater thermal conductivity, as indicated in Figure 8.25. This phenomenon is often referred as aging.55–59 Aging is faster for plastic foams under a high thermal gradient environment than under isothermal conditions.57,58 Reference 55 suggests the mass flow rate (Jm) of one gas species when it diffuses across the cell walls of the plastic foam is Pe Jm  __________ dw(P2  P1) De Jm  __________ dw(C2  C1)

(8.6) (8.7)

where Pe and De are the permeability coefficient and diffusion coefficient; dw is the thickness of a cell wall; P2 and P1 are the high and low partial pressure imposed on the two surfaces, respectively; and P2 and P1 can be associated with the concentration, C2 and C1. Therefore, a basic parameter governing the gas conductivity (λg) in the plastic foam is the gas transport properties; that is, the diffusion of the foaming agent as well as air.26,55,56,59

61259_C008.indd 274

10/25/2008 12:40:01 PM

Polystyrene Foam and Its Improvement

275

Besides the mass flow rates of gases in the plastic foam, the thermal conductivity through gas can also be interpreted as

λg  A  g  Cv  V  (L  Lv)/(Lv  L)

(8.8)

where A is a constant; g is the gas density inside the cell; Cv is the specific heat of gas at a constant volume; V is the velocity of gas molecules; L is the mean free path of gas molecules—the average distance of a gas molecule traveling before hitting another gas molecule, and Lv is the distance between cavities (cell size in the direction of thermal conduction).17 The increase in the gas density, specific heat of gas, velocity of gas molecules, and cell size can result in the enhancement of gas conductivity in plastic foams.17 Thermal conductivity through gas also depends on temperature. It decreases with the reduction of temperature because the decrease in gas molecular motion leads to reduced transport of thermal energy.52,57,58 Reference 26 concludes that the plastic foam having a foaming agent with low diffusivity, a low volume expansion ratio, a high cell density (small cell size), and a large foam thickness can attenuate the decay of the foaming agent in the plastic foam and show better thermal insulation capability. Heat transfer through gas in the plastic foams accounts for approximate 40–50% of the overall heat transfer.49,56,57 Moreover, Reference 16 estimates heat transfer through gas around 70–80% of the total heat transfer. Eventually the thermal conduction for plastic foams can occur through bulk material itself and through gas. The volume of bulk material present in the low-density foam is of the order of 2–3% so that gases act as the main conducting medium.60 The thermal conductivity through radiation can be expressed as:

λr  16T 3/3Kr

(8.9)

where  is Stefan–Boltzmann constant (5.669  108 W/m2 k4); T is the absolute temperature of the local material and Kr is the extinction coefficient. For radiant energy traversing a material in a given solid angle with wavelengths, the decrease in intensity of radiant energy can be associated with the extinction coefficient. When radiant energy interacts with the plastic foam, part of the radiant energy is scattered by struts, part is reflected by cell walls, part is absorbed by struts as well as cell walls, and part is transmitted by struts and cell walls.26 The extinction coefficient includes the absorption coefficient and the scattering coefficient. Therefore, radiation heat transfer is directly affected by the extinction coefficient of the plastic foam. Reference 61 considers that the struts are thick enough to be opaque; the strut cross-section is constant and occupies two-thirds of the area

61259_C008.indd 275

10/25/2008 12:40:01 PM

276

Polymeric Foams

of an equilateral triangle (ABC) formed at the vertices depicted in Figure 8.23. The triangle cross-sections of struts are converted into circular cross-sections with the same area indicated in Figure 8.23 and strut diameter (fs) is larger than wall thickness, or at most equal.51,62 The assumption is used in theoretical study for the extinction coefficient. The extinction coefficient in the plastic foam is directly proportional to the foam density and inversely proportional to the cell diameter (cell size).50,61–63 References 64 and 65 use more complicated methods to describe strut junctures and the struts used in the model to study the extinction coefficient. Basically the struts are not constant dimensions in the cross-section. The solid in the plastic foam contains approximately 10–20% cell walls and 80–90% struts.66,67 Struts strongly influence the absorption and scattering properties of the plastic foam. The extinction coefficient of the plastic foam can be experimentally determined using a Fourier transform infrared spectrometer (FTIR).49,61,62,64,67 Radiative conductivity (λr) for the plastic foam can decrease by reducing cell diameter or by increasing foam density and strut diameter for a given mean cell diameter.17,50,52,61–63,66,68 The effect on the reduction of radiative conductivity occurs because struts become more opaque to radiation. Radiation for a plastic foam accounts for 2.5–34% of the overall heat transfer. This fraction of radiation may be decreased through reducing the cell size in the foaming process, lowering transmitted radiant energy and re-emitted radiant energy from the absorbed energy by additives or pigments in the formulation, as well as using a medium making boundary surfaces with low emissivity values.53 When the cell sizes in plastic foams are less than 4 mm, the effect of convective heat transfer is negligible.69 The closed-cell structure in plastic foams consists of struts and intact cell walls and cells are filled with gases. Therefore, the total thermal conductivity for closed-cell plastic foams can be expressed as

λ  λs  λg  λr

(8.10)

Figure 8.26 presents the theoretical and experimental results for the thermal conductivity of PS foam boards blown with HFC-134a. The total thermal conductivity in theoretical study is based on the sum of the thermal conductivity through gas (λg) and the thermal conductivity through radiation (λr). The theoretical results for long-term thermal insulation performance of PS foam boards are shown in Figures 8.27, 8.28, and 8.29. Figure 8.27 indicates the total thermal conductivity influenced by blowing agent type. Foams produced by both HFC-134a and HCFC-142b have good long-term thermal insulation properties, but foam made by HFC-152a does not have. Figure 8.28 is the effect of cell density on the total thermal conductivity. A high cell density in the foam can lead to a low total thermal conductivity of the foam. Figure 8.29 displays the dependence of the

61259_C008.indd 276

10/25/2008 12:40:01 PM

277

Thermal conductivity (mW/m/K)

Polystyrene Foam and Its Improvement

30

20

10

Experimental Numerical

0 0

2

4 6 Time (year)

8

10

FIGURE 8.26 Relationship between thermal conductivity and time. [From Zhu, Z., Zong, J., and Park, C.B. In Proceedings of the Society of Plastics Engineers Annual Technical Conference (ANTEC) 63 (2007): 1494–1498. With permission.]

Total thermal conductivity (mW/m/K)

40 30 20 HFC 152a HFC 134a HCFC 142b

10 0 0

10

20

30

Time (year) FIGURE 8.27 Effect of blowing agent type on the thermal conductivity. [From Zhu, Z., Zong, J., and Park, C.B. In Proceedings of the Society of Plastics Engineers Annual Technical Conference (ANTEC) 63 (2007): 1494–1498. With permission.]

Total thermal conductivity (mW/m/K)

40

30

20

Cell density = 5.75 × 105#/cc Cell density = 2.50 × 106#/cc Cell density = 1.00 × 107#/cc 0

10

20

30

Time (year) FIGURE 8.28 Effect of cell density on thermal conductivity. [From Zhu, Z., Zong, J., and Park, C.B. In Proceedings of the Society of Plastics Engineers Annual Technical Conference (ANTEC) 63 (2007): 1494–1498. With permission.]

61259_C008.indd 277

10/25/2008 12:40:01 PM

278

Polymeric Foams

Total thermal conductivity (mW/m/K)

40

30

VER = 40 VER = 35 VER = 30

20

0

10

20

30

Time (year) FIGURE 8.29 Effect of volume expansion ratio on the thermal conductivity. [From Zhu, Z., Zong, J., and Park, C.B. In Proceedings of the Society of Plastics Engineers Annual Technical Conference (ANTEC) 63 (2007): 1494–1498. With permission.]

total thermal conductivity on volume expansion. A low total thermal conductivity can be obtained by a low volume expansion ratio for the foam. The open-cell structure in plastic foams mainly comprises struts and a few cell walls. The total thermal conductivity for open-cell plastic foam is the same as Equation 8.10. However, when a porous material is used as a core material in VIPs, the thermal conductivity of the VIP decreases with the reduction of the internal pressure described in Figure 8.30. Three types of core material, open-cell PS foam, open-cell PU foam, and Nanogel (branded aerogel), are used in VIPs. The pore sizes of open-cell PS foam and open-cell

0.025

K (W/m k)

0.020

Open-cell ps Open-cell pu Nanogel

0.015

0.010

0.005

0 0.01

0.1

1

10 Pressure (Pa)

100

1000

10000

FIGURE 8.30 Thermal conductivity of VIPs at various internal pressures.

61259_C008.indd 278

10/25/2008 12:40:02 PM

Polystyrene Foam and Its Improvement

279

PU foam are microscale but Nanogel has nanoscale pores. The data of thermal conductivity for Nanogel are obtained from Reference 70. The decrease in the internal pressure for the three VIPs to a certain level can lead to an approximate constant for thermal conductivity of the three VIPs. The VIP using a Nanogel as a core material has much lower thermal conductivity than that using an open-cell PS foam and open-cell PU foam at the same internal pressure. The VIP with an open-cell PS foam as a core material shows a similar thermal conductivity to that of an open-cell PU foam at an internal pressure of less than 13.3 Pa (0.1 torr). However, the VIP having an open-cell PU foam as a core material exhibits a higher thermal conductivity than that having an open-cell PS foam at a higher internal pressure. As a result, the thermal conductivity of VIP is performed by thermal conductivity through solid and radiation and is no longer affected by thermal conductivity through gas. However, when the foam density of plastic foams increases at a constant temperature, the solid conductivity increases; radiative conductivity decreases; and gas conductivity is constant. An optimal foam density exists to have the lowest total thermal conductivity for the plastic foams with either open-cell or closed-cell structure.62 Pore sizes reaching nanoscale in the core material greatly help the thermal performance of VIP. Therefore, the requirements for an open-cell plastic foam as a core material used in VIPs include • • • •

Plastics having low thermal conductivity High content of open cells (more than 95% open cells) Small cell size (less than 100 m) Light weight and good mechanical properties to resist external pressure from the atmosphere.

8.5 Thermal Conductivity of VIPs Using PS Foams as Core Materials The thermal conductivity of VIPs using different porous PS foams as core materials is presented in Figure 8.31. PS with fillers such as CaCO3 and PS with LDPE and fillers are foamed by a gas mixture of CO2 and fluorocarbon at a constant pressure. When porous PS foams used in VIPs are made at a low foaming temperature, for instance 123°C, VIPs exhibit a high thermal conductivity. VIPs using porous PS foams produced at the foaming temperature of 127°C have the lowest thermal conductivity of the three foaming temperatures, and at the foaming temperature of 125°C VIPs exhibit an in-between thermal conductivity.

61259_C008.indd 279

10/25/2008 12:40:02 PM

280

Polymeric Foams

0.030 VIP (0.01 torr)

K (W/m k)

0.025 0.020

T: 123°C T: 125°C T: 127°C

0.015 0.010 0.005 0

0

2

4

6 LDPE (%)

8

10

12

FIGURE 8.31 Thermal conductivity of VIPs using PS/LDPE foams with fillers as core materials at various foaming temperatures and CO2/fluorocarbon as a foaming agent.

Normally, a high thermal conductivity of VIPs indicates that the porous foam has a low content of open cells and vice versa. As the content of LDPE in the porous PS foam increases, the thermal conductivity of VIPs using porous foams made at the foaming temperature of 123°C shows an initially dramatic decrease and then levels off; however, at the foaming temperatures of 125°C and 127°C do not have that trend. The results indicate that the content of open cells in porous PS foams made at the foaming temperature of 123°C is more influenced by LDPE than that made at the foaming temperatures of 125°C and 127°C. Figure 8.32 illustrates the thermal conductivity of VIPs using different porous foams as core materials manufactured under a gas mixture of CO2 and fluorocarbon, three foaming pressures, and a constant foaming temperature of 125°C. VIP foam using PS with fillers made at the foaming pressure of 6.88 MPa (1000 psi) has the highest thermal conductivity while the foaming pressure of 8.95 MPa (1300 psi) displays the lowest thermal conductivity. VIPs have an intermediate thermal conductivity when PS foam with fillers is produced at the foaming pressure of 11.02 MPa (1600 psi). As the content of LDPE in porous PS foams increases, the thermal conductivity of VIPs decreases first and then increases gradually for the three foaming pressures. However, the lowest thermal conductivity of VIPs is always observed at the foaming pressure of 8.95 MPa (1300 psi). Therefore, an optimal foaming pressure exists to manufacture porous PS foam with a high content of open cells. Because PS foam with fillers or PS/LDPE foam with fillers are produced at the foaming pressure of 11.02 MPa (1600 psi), large-cell sizes scatter in porous PS foams and porous PS foams are slightly distorted. When they are foamed at the foaming pressure of 6.88 MPa (1000 psi), the content of open cells is lower than that

61259_C008.indd 280

10/25/2008 12:40:02 PM

281

Polystyrene Foam and Its Improvement

0.025 VIP (0.01 torr)

K (W/m k)

0.020 0.015 0.010 P: 6.88 MPa P: 8.95 MPa P: 11.02 MPa

0.005 0

0

2

4

6 LDPE (%)

8

10

12

FIGURE 8.32 Thermal conductivity of VIPs using PS/LDPE foams with fillers as core materials at various foaming pressures and CO2/fluorocarbon as a foaming agent.

in porous PS foams made at the foaming pressure of 8.95 MPa (1300 psi) and average cell sizes are larger. How to balance the cell size, the content of open cells, and the mechanical strength for porous PS foams is an important issue. These factors can influence the thermal conductivity of PS foams. Figure 8.33 shows the thermal conductivity of VIPs using different porous PS foams as core materials produced at four foaming temperatures of 121°C, 123°C, 125°C, and 127°C and a constant pressure of 8.95 MPa (1300 psi). The foaming agent is a gas mixture of CO2 and nitrogen (N2). The trend of thermal conductivity for VIPs using porous PS foams made by a gas mixture of CO2 and N2 is similar to that for VIPs using porous PS foams made by a gas mixture of CO2 and fluorocarbon. However, porous PS foams made by a gas mixture of CO2 and N2 have a higher thermal conductivity and a lower content of open cells than those made by a gas mixture of CO2 and fluorocarbon at the same temperature, as indicated in Figure 8.31. Generally speaking, porous PS foams with 7% LDPE and fillers made by a gas mixture of CO2 and N2 generate a better result in thermal conductivity such as a narrow difference of thermal conductivity and low thermal conductivity for the four foaming temperatures than other PS foams with LDPE and fillers. The thermal conductivity of VIPs using different porous foams as core materials produced under a gas mixture of CO2 and N2 as a foaming agent, three foaming pressures, and a constant foaming temperature of 125°C is shown in Figure 8.34. The sequence from high to low thermal conductivity for PS foams with fillers is the foam made at the foaming pressure of 11.02 MPa (1600 psi), then at the foaming pressure of 8.95 MPa (1300 psi), then at the foaming pressure of 6.88 MPa (1000 psi). The thermal conductivity for

61259_C008.indd 281

10/25/2008 12:40:03 PM

282

Polymeric Foams

0.035 VIP (0.01 torr) 0.030

K (W/m k)

0.025

T: 121°C T: 123°C T: 125°C T: 127°C

0.020 0.015 0.010 0.005 0

0

2

4

6 LDPE (%)

8

10

12

FIGURE 8.33 Thermal conductivity of VIPs using PS/LDPE foams with fillers as core materials at various foaming temperatures and CO2/N2 as a foaming agent.

PS foams with fillers produced by the gas mixture of CO2 and N2 is higher than that by the gas mixture of CO2 and fluorocarbon indicated in Figure 8.32, and a different sequence of thermal conductivity is also observed due to different foaming agents. The thermal conductivity of VIPs for PS foams with 2% LDPE and fillers at three foaming pressures is very close and is much lower than that of VIPs for PS foams with fillers at three foaming pressures. VIPs for PS foams with 5% LDPE and fillers made at the foaming

0.025 VIP (0.01 torr)

K (W/m k)

0.020

0.015

0.010

P: 6.88 MPa P: 8.95 MPa P: 11.02 MPa

0.005

0

0

2

4

6

8

10

12

LDPE (%) FIGURE 8.34 Thermal conductivity of VIPs using PS/LDPE foams with fillers as core materials at various foaming pressures and CO2/N2 as a foaming agent.

61259_C008.indd 282

10/25/2008 12:40:03 PM

Polystyrene Foam and Its Improvement

283

pressure of 8.95 MPa (1300 psi) show the lowest thermal conductivity in the study. However, the increase of LDPE in PS can lead to the enhancement of thermal conductivity for PS foams with LDPE and fillers. Figure 8.35a–d represents the cell structures of PS foams for VIPs having different thermal conductivity of 0.027 W/m k, 0.02 W/m k, 0.01 W/m k, and 0.0065 W/m k, respectively. The content of open cells is less than 50% in the PS foam having the thermal conductivity of 0.027 W/m k; in other words, this foam can be considered a closed-cell foam. When thermal conductivity decreases, the content of open-cells increases in the PS foam; for example, the PS foam with the thermal conductivity of 0.02 W/m k and approximately 70% open cells in the PS foam, but many closed-cells are still observed in the PS foam. As the thermal conductivity of the PS foam reaches 0.01 W/m k and the content of open cells in the PS foam is approximately 90%, many open cells exist and intact cell structures, such as pentagonal and hexagonal structures, gradually disappear. The thermal conductivity of the PS foam is 0.0065 W/m k. It is difficult for the PS foam to observe an intact cell structure. The PS foam contains approximately

FIGURE 8.35 Cell structures of PS/LDPE foams with fillers as core materials for VIPs having thermal conductivity: (a) 0.027 W/m k; (b) 0.02 W/m k; (c) 0.01 W/m k; and (d) 0.0065 W/m k.

61259_C008.indd 283

10/25/2008 12:40:03 PM

284

Polymeric Foams

0.023

Getter in VIP (0) No getter in VIP (1) No getter in VIP (2)

0.021 0.019

K (W/m k)

0.017 0.015 0.013 0.011 0.009 0.007 0.005

0

50

100

150

200

250

300

350

400

450

500

Time (days) FIGURE 8.36

Relationship between thermal conductivity and time.

98% open cells. In other words, this is a PS foam with very high content of open cells. The open-cell content of rigid cellular plastics can be measured and analyzed with an air pycnometer using a standard test method.71 The relationship between thermal conductivity of three VIPs having PS foams as core materials and time is described in Figure 8.36. A getter is used within VIP(0) but not in the other two VIPs, VIP(1) and VIP(2). A low thermal conductivity at the beginning stage is observed for each VIP; the thermal conductivity of each VIP then gradually increases and then levels off. A similar thermal conductivity for VIP(0) and VIP(1) is obtained initially, but VIP(2) shows a higher thermal conductivity under the same internal pressure. Generally, a lower thermal conductivity for VIP indicates a higher content of open cells in the PS foam as described in Figure 8.35. Therefore, more closed-cells represent more gases trapped in the PS foam. When the PS foam is placed in a low-pressure environment, gases are released from the PS foam. The released gases can dramatically change the internal pressure and thermal conductivity of VIP. It takes approximately 120 days for VIP(0), 240 days for VIP(1), and 360 days for VIP(2) to achieve a stable thermal conductivity. When a getter is placed in VIP(0), the stable thermal conductivity for VIP(0) is similar to the initial thermal conductivity. However, the stable thermal conductivity for VIP(2) is much higher than the initial thermal conductivity. VIP(0) and VIP(1) apparently have a lower gas release since they have lower initial conductivities and change very little over time compared with VIP(2).The results also indicate that the PS foam with a high content of open cells and a good mechanical strength is a good core material for a VIP.

61259_C008.indd 284

10/25/2008 12:40:03 PM

Polystyrene Foam and Its Improvement

285

8.6 Conclusions Neat PS and PS with CaCO3 or PS with LDPE are foamed by CO2 as a foaming agent under supercritical conditions. The low foam density for neat PS occurs at foaming temperatures of between 130°C and 140°C. When PS blended with CaCO3 or LDPE are foamed, the occurrence of low foam density for PS/CaCO3 or PS/LDPE foams shifts to low foaming temperatures. Low foam density and the foaming temperatures which generate low foam density for PS/CaCO3 or PS/LDPE foams depend on the amounts of CaCO3 or LDPE used. Average cell sizes and cell structures such as open-cell or closed-cell for PS/CaCO3 or PS/LDPE foams are also influenced by the content of CaCO3 or LDPE. In comparison with the average cell sizes of neat PS foams, CaCO3 or LDPE in PS foams can lower the average cell sizes of PS/CaCO3 or PS/LDPE foams. VIPs achieve a lower thermal conductivity when PS/LDPE foams with fillers as core materials are made by both foaming agents, a mixture of CO2/fluorocarbon and CO2/N2 at a higher foaming temperature and a constant foaming pressure of 8.95 MPa (1300 psi). However, PS/LDPE foams with fillers are produced at various foaming pressures, 6.88 MPa (1000 psi), 8.95 MPa (1300 psi), and 11.02 MPa (1600 psi), and a constant foaming temperature of 125°C. An optimal foaming pressure exists to manufacture PS/LDPE foams with fillers having a high content of open cells. Generally, a porous PS foam with a higher content of open-cells as a core material used in the VIP can result in a lower thermal conductivity and a more stable thermal conductivity during its lifetime.

8.7 Abbreviations Al CaCO3 CFA CO2 DSC EtCl EO ESI EVOH FTIR GPPS HCFC-22 HCFC-142b

61259_C008.indd 285

Aluminum Calcium carbonate Chemical foaming agent Carbon dioxide Differential scanning calorimeter Ethyl chloride Ethylene/octane copolymer Ethylene/styrene interpolymer Ethylene vinyl alcohol Fourier transform infrared spectrometer General-purpose polystyrene Chlorodifluoro methane 1-chloro-1,1-difluoro ethane

10/25/2008 12:40:04 PM

286

Polymeric Foams

HDT HDPE HFC-134a HFC-152a HIPS LDPE Mw N2 Ny PET PFA PS PU SEBS SEM sPS VIP

Heat distortion temperature High-density polyethylene 1,1,1,2-tetrafluoroethane 1,1-difluoroethane High-impact polystyrene Low density polyethylene Weight average molecular weight Nitrogen Nylon Polyethylene terephthalate Physical foaming agent Polystyrene Polyurethane Styrene/ethylene butylenes/styrene Scanning electron microscopy Syndiotactic polystyrene Vacuum insulation panel

8.8 Nomenclature   λ λc λg λo λr λs a f g o  A C1 C2 Cv De dw fs Jm Kr

61259_C008.indd 286

Non-isotropic factor Void fraction or porosity of the foam Total thermal conductivity Thermal conductivity contributed from convection Thermal conductivity contributed from gas Solid conductivity of the bulk material Thermal conductivity contributed from radiations Thermal conductivity contributed from solid Air density Foam density Gas density inside the cell Density of the bulk material Stefan-Boltzmann constant (5.669 × 108 W/m2 k4) Constant Low partial concentration High partial concentration Specific heat of gas at a constant volume Diffusion coefficient Thickness of a cell wall Mass fraction of the struts in the foam Mass flow rate Extinction coefficient

10/25/2008 12:40:04 PM

Polystyrene Foam and Its Improvement

L Lv P1 P2 Pe T Tg V

287

Mean free path of gas molecules—the average distance of a gas molecule traveling before hitting another gas molecule Distance between cavities (cell size in the direction of thermal conduction) Low partial pressure High partial pressure Permeability coefficient Absolute temperature of the local material Glass transition temperature Velocity of gas molecules

References 1. Kullberg, R. C., Manini, P., Corberi, F., and Rizzi, E. “Getters and desiccants for lifetime and performance maintenance in vacuum insulation panels.” In Proceedings of the International SAMPE Symposium and Exhibition 46 (2001): 1239–1250. 2. Lanzani, F. and Mauri, R. European Patent 0849309A1(1998). 3. Imeokparia, D. D., Shmidt, C. D. and Suh, K. W. US Patent 5,693,687, 1997. 4. Shmidt, C. D., Imeokparia, D. D., Suh, K. W., Malone, B. A., and Franklin, K. “Extruded Open-Cell microcellular foams, and their preparation process.” WO 96/34038, 1996. 5. Shmidt, C. D., Imeokparia, D. D., Suh, K. W., Malone, B. A., and Franklin, K. US Patent 5,780,521, 1998. 6. Shmidt, C. D., Imeokparia, D. D., Suh, K. W., Malone, B. A., and Franklin, K. US Patent 5,863,960, 1999. 7. Shmidt, C. D., Suh, K. W., Imeokparia, D. D., Malone, B. A., and Franklin, K., US Patent 5,869,544, 1999. 8. Bland, D. G., Stobby, W. G., Mork, S. W., Staples, T. G., and McCann, G. D. “Absorbent, extruded thermoplastic foams.” WO 98/56430, 1998. 9. Park, C. P., Imeokparia, D. D., and Chaudhary, B. I. “Open-Cell foam and method of making.” WO 99/47592, 1999. 10. Park, C. P., Chaudhary, B. I., and Imeokparia, D. D. US Patent 6,093,752, 2000. 11. Park, C. P., Chaudhary, B. I., and Imeokparia, D. D. US Patent 6,174,471, 2001. 12. Suh, K. W., Park, C. P., Maurer, M. I., Tusim, M. H., De Genova, R., Broos, R., and Sophiea, D. P. “Lightweight cellular plastics.” Advanced Materials 12 (2000): 1779–1789. 13. Wong, C. M., Liang, W. C., Ying, C. H., and Hung, M. L. “Porous polystyrene foam produced at supercritical conditions.” Journal of Cellular Plastics 40 (2004): 421–429. 14. Wong, C. M., Tsai, S. J., Ying, C. H., and Hung, M. L. “Effect of low density polyethylene on polystyrene foam.” Journal of Cellular Plastics 42 (2006): 153–163.

61259_C008.indd 287

10/25/2008 12:40:04 PM

288

Polymeric Foams

15. Tao, W. H., Sung, W. F., and Lin, J. Y. “Development of vacuum insulation panel system.” Journal of Cellular Plastics 33 (1997): 545–555. 16. Wu, J. W., Sung, W. F., and Chu, H. S. “Thermal conductivity of polyurethane foams.” International Journal of Heat and Mass Transfer 42 (1999): 2211–2217. 17. Kodama, K., Yuge, K., Masuda, Y., and Tanimoto, Y. “Development of micro cellular open cell rigid polyurethane foams.” In Proceedings of Polyurethanes World Congress, October 10–13, (1993): 140–145. 18. Smith, D. M., Maskara, A., and Boes, U. “Aerogel-Based thermal insulation.” Journal of Non-Crystalline Solids 225 (1998): 254–259. 19. Rouanet, S., Floess, J., and Eberhardt, H. F. “Aerogels for optimized performance in vacuum insulation panels.” In Proceedings of the International Society for Advancement of Materials and Process Engineering Symposium and Exhibition 46 (2001): 1263–1270. 20. Wakili, K. G., Bundi, R., and Binder, B. “Effective thermal conductivity of vacuum insulation panels.” Building Research and Information 32 (2004): 293–299. 21. Schwab, H., Heinemann, U., Beck, A., Ebert, H. P., and Fricke, J. “Dependence of thermal conductivity on water content in vacuum insulation panels with fumed silica kernels.” Journal of Thermal Envelope and Building Science 28 (2005): 319–326. 22. Schwab, H., Heinemann, U., Beck, A., Ebert, H. P., and Fricke, J. “Prediction of service life for vacuum insulation panels with fumed silica kernel and foil cover.” Journal of Thermal Envelope and Building Science 28 (2005): 357–374. 23. Toyo Aluminum K.K. Brochure, Tokyo, Japan, 2002. 24. DuPont Teijin Films. Brochure, Hopewell, VA, 2002. 25. Jacobsen, S. “Barrier films—ways to improve functionality.” Presented at VIA 2002 Symposium, Atlanta, May 22–23, 2002. 26. Zhu, Z., Zong, J., and Park, C. B. In “Long-Term thermal insulation performance of polystyrene foam boards with decay of blowing agent.” In Proceedings of the Society of Plastics Engineers Annual Technical Conference 63 (2007): 1494–1498. 27. Chang, W. R., Chang, C. H., Chung H. D., and Lin, J. Y. “Energy efficient potential for household refrigerators/freezers with high efficiency compressor and vacuum insulation panels.” Presented at VIA 2002 Symposium, Atlanta, May 22–23, 2002. 28. Sosa, J. M. and Blackmon, K. P. Polymeric Materials Encyclopedia, Vol. 10, ed. T. C. Salamone. CRC Press, Boca Ration, 1996. p. 8032. 29. Traugott, T. D. Encyclopedia of Polymer Science and Engineering, Vol. 16, ed. J. I. Kroschwitz. John Wiley & Sons, New York, 1986. p. 63 30. Soderquist, M. E. and Dion, R. P. Encyclopedia of Polymer Science and Engineering, Vol. 16, ed. J. I. Kroschwitz. John Wiley & Sons, New York, 1986. p. 88. 31. Nanasawa, A. Polymeric Materials Encyclopedia, Vol. 9, T. C. Salamone, ed. CRC Press, Boca Raton, 1996. p. 6799. 32. Dow Plastics. “Restyling styrene.” European Plastics News 24(5) (1997): 24. 33. Dow Plastics. “Questra enters the Real World.” European Plastics News 26(11) (1999): 30. 34. Lavengood, R. E. US Patent 4,214,056, 1980. 35. Martini, J., Waldman, F., and Suh, N. P. “The production and analysis of microcellular thermoplastic foams.” In Proceedings of the Society of Plastics Engineers Annual Technical Conference 28 (1982): 674–676.

61259_C008.indd 288

10/25/2008 12:40:04 PM

Polystyrene Foam and Its Improvement

289

36. Colton J. S. and Suh, N. P. “The nucleation of microcellular thermoplastic foam with additives part II: Experimental results and discussion.” Polymer Engineering and Science 27 (1987): 493–499. 37. Goel S. K. and Beckman, E. J. “Generation of microcellular polymeric foams using supercritical carbon dioxide I: Effects of pressure and temperature on nucleation.” Polymer Engineering and Science 34 (1994): 1137–1147. 38. Arora, K. A., Lesser, A. J., and McCarthy, T. J. “Preparation and characterization of microcellular polystyrene foams processed in supercritical carbon dioxide.” Macromolecules 31 (1998): 4614–4620. 39. Stafford, C. M., Russell, T. P., and McCarthy T. J. “Expansion of polystyrene using supercritical carbon dioxide: Effects of molecular weight, polydispersity, and low molecular weight components.” Macromolecules 32 (1999): 7610–7616. 40. Park, C. B., Behravesh, A. H., and Venter, R. D. “Low-Density microcellular foam processing in extrusion using CO2.” Polymer Engineering and Science 38 (1998): 1812–1823. 41. Baldwin, D. F., Park, C. B., and Suh, N. P. “An extrusion system for the processing of microcellular polymer sheets: Shaping and cell growth control.” Polymer Engineering and Science 36 (1996): 1425–1435. 42. Han, X., Koelling, K. W., Tomasko, D. L., and Lee, L. J. “Continuous microcellular polystyrene foam extrusion with supercritical CO2.” Polymer Engineering and Science 42 (2002): 2094–2106. 43. Xu, X., Park, C. B., Xu, D., and Pop-iliev, R. “Effects of die geometry on cell nucleation of PS foams blown with CO2.” Polymer Engineering and Science 43 (2003): 1378–1390. 44. Rodeheaver, B. A. and Colton, J. S. “Open-celled microcellular thermoplastic foam.” Polymer Engineering and Science 41 (2001): 380–400. 45. Suetsugu, Y. and White, J. L. “The influence of particle size and surface coating calcium carbonate on the rheological properties of its suspension in molten polystyrene.” Journal of Applied Polymer Science 28 (1983): 1481–1501. 46. Chiu, F. C., Lai, S. M., Wong, C. M., and Chang, C. H. “Properties of calcium carbonate filled and unfilled polystyrene foams prepared using supercritical carbon dioxide.” Journal of Applied Polymer Science 102 (2006): 2276–2284. 47. Lee, P. C., Li, G., Lee, J. W. S., and Park, C. B. “Improvement of cell opening by maintaining a high temperature difference in the surface and core of a foam extrudate.” Journal of Cellular Plastics 43 (2007): 431–444. 48. Klempner, D. and Frisch, K. C. Handbook of Polymeric Foams and Foam Technology. Hanser, New York, 1991. 49. Torpey, M. R. “A study of radiative heat transfer through foam insulation.” M.Sc. thesis, Department of mechanical engineering, massachusetts institute of technology, 1987. 50. Schuetz M. A. and Glicksman, L. R. Journal of Cellular Plastics 20 (1984): 114–121. 51. Kuhn, J., Ebert, H. P., Arduini-Schuster, M. C., Büttner, D., and Fricke, J. “Thermal transport in polystyrene and polyurethane foam insulations.” International Journal of Heat Mass Transfer 35 (1992): 1795–1801. 52. Bhattacharjee, D., King, J. A., and Whitehead, K. N. “Thermal Conductivity of PU/PIR foams as a function of mean temperature.” Journal of Cellular Plastics 27 (1991): 240–251.

61259_C008.indd 289

10/25/2008 12:40:04 PM

290

Polymeric Foams

53. Williams, R. J. J. and Aldao, C. M. “Thermal conductivity of plastic foams.” Polymer Engineering and Science 23 (1983): 293–298. 54. Prociak, A., Pielichowski, J., and Sterzynski, T. “Thermal diffusivity of rigid polyurethane foams blown with different hydrocarbons.” Polymer Testing 19 (2000): 705–712. 55. Ostrogorsky, A. G., Glicksman, L. R., and Reitz, D. W. “Aging of polyureathane foams.” International Journal of Heat Mass Transfer 29 (1986): 1169–1176. 56. Albouy, A., Roux, J. D., Mouton, D., and Wu, J. “Development of HFC blowing agents. Part II: Expanded polystyrene insulating boards.” Cellular Polymers 17 (1998): 163–176. 57. Zarr, R. R. and Nguyen, T. J. “Effect of humidity and elevated temperature on the density and thermal conductivity of a rigid polyisocyanurate foam Co-Blown with CCl3 and CO2.” Thermal Insulation and Building Envelopes 17 (1994): 330–350. 58. Bomberg, M. J. “Predicting field thermal performance of a modified resol foam from laboratory data.” Thermal Insulation and Building Envelopes 17 (1993): 78–88. 59. Vo, C. V. and Paquet, A. N. “An evaluation of the thermal conductivity of extruded polystyrene foam blown with HFC-134a or HCFC-142b.” Journal of Cellular Plastics 40 (2004): 205–228. 60. Doherty, D. J., Hurd, R., and Lester, G. R. “The physical properties of rigid polyurethane foams.” Chemistry and Industry (1962): 1340–1356. 61. Glicksman L. R. and Torpey, M. R. “The influence of cell size and foam density on the thermal conductivity of foam insulation.” In Proceedings of Polyurethanes World Congress, September 29–October 2 (1987): 80–84. 62. Placido, E., Arduini-Schuster, M. C., and Kuhn, J. “Thermal properties predictive model for insulation foams.” Infrared Physics and Technology 46 (2005): 219–231. 63. Koo, M. S., Chung, K., and Youn, J. R. “Reaction injection molding of polyurethane foam for improved thermal insulation.” 41 Polymer Engineering and Science 41 (2001): 1177–1186. 64. Baillis, D., Raynaud, M., and Sacadura, J. F. “Spectral radiative properties of Open-Cell foam insulation.” Journal of Thermophysics and Heat Transfer 13 (1999): 292–298. 65. Doermann, D. and Sacadura, J. F. “Heat transfer in Open-Cell foam insulation.” Journal of Heat Transfer 18 (1996): 88–93. 66. Reitz, D. W., Schuetz, M. A., and Glicksman, L. R. “A basic study of aging of foam insulation.” Journal of Cellular Plastics 20 (1984): 104–113. 67. Glicksman, L., Schuetz, M., and Sinofsky, M. “Radiation heat transfer in foam insulation.” International Journal of Heat Mass Transfer 30 (1987): 187–197. 68. Biedermann, A., Kudoke, C., Merten, A., Minogue, E., Rotermund, U., Ebert, H. P., Heinemann, U., and Fricke, J. “Analysis of heat transfer mechanisms in polyurethane rigid foam.” Journal of Cellular Plastics 37 (2001): 467–483. 69. Skochdopole, R. E. “The thermal conductivity of foamed plastics.” Chemical Engineering Progress 57 (1961): 55–59. 70. NanoPore Incorporated. Brochure, Albuquerque, NM, 2003. 71. American Society for the Testing of Materials. “Standard test method for Open-Cell content of rigid cellular plastics by the air.” D2856, Volume 8(2) Plastics (II) (1998): 143–148.

61259_C008.indd 290

10/25/2008 12:40:04 PM

Author Index

Abes, J. I., 188 Adachi, Y., 178, 181 Adamovsky, S., 145 Aguillar, M., 151 Ahern, A., 25 Al Ghatta, H. A. K., 9 Albouy, A., 274, 275 Aldao, C. M., 274, 276 Alexandre, M., 157 Amecke, B., 43, 48, 51 Anderson, J. R., 116 Anderson, K. L., 188 Ano, Y. T., 155 Arduini-Schuester, M. C., 273, 276, 279 Areerat, S., 11, 12 Arora, K. A., 261 Ashby, M. F., 22, 215 Ashford, P., 28, 31 Ashida, K., 15 Astarita, L., 9 Awojulu, A., 155 Baillis, D., 276 Baldwin, D. F., 90, 226, 261 Barito, R. W., 14 Bayer, O., 11 Beck, A., 256 Beckman, E. J., 206, 261 Behravesh, A. H., 261 Benning, C. J., 11

61259_C009.indd 291

Berghaus, U., 57, 58 Berglund, L. A., 161 Bhattacharjee, D., 274, 275, 276 Biedermann, A., 276 Biesenberger, J. A., 22 Bikiaris, D. N., 144 Binder, B., 256 Blackmon, K. P., 259 Bland, D. G., 256, 261, 262 Blander, M., 8 Blasius, W. G., 155 Boehringer Ingelheim, K. G., 58, 59 Boes, U., 256 Bomberg, M. J., 274, 275 Bopp, R. C., 33 Bourban, P.-E., 150 Bousmina, M., 157 Boyce, M. C., 188 Boyce, S. T., 144 Broos, R., 256 Brown, J. M., 214 Büttner, D., 273, 276 Büttner, H., 63 Bundi, R., 256 Calberg, C., 157 Campbell, N. D., 152 Cao, X., 163, 214 Carbonell, R. G., 146 Carlson, D., 155

10/25/2008 12:40:30 PM

292

Caron, L.-M., 71 Cha, S. W., 116 Champagne, M., 6, 150 Chandra, A., 214 Chang, C. H., 259, 263 Chang, F.-C., 161 Chang, J., 15 Chang, W. R., 259 Chang, Y., 146 Chaudhary, B. I., 256, 261, 262 Chen, C., 7 Chen, C. Y., 152 Chen, X., 155 Chen, Y., 15 Cheng, M. L., 7 Chiu, F. C., 263 Chixin, Z., 161 Chmiel, H., 72 Cho, H., 155 Chu, C. C., 152 Chu, H. S., 256 Chung, C. I., 16 Chung, H. D., 259 Chung, K., 276 Cohen, R. E., 188 Collington, K. T., 233 Collins, F. H., 16, 19, 46, 51 Colton, J. S., 90, 113, 201, 204, 261, 262 Columbo, R., 3 Cooper, A., 221 Corberi, F., 256 Cordes, H., 214 Coslanicha, A., 186 Cotugno, S., 144, 146, 160 Crank, J., 112, 226 Curliss, D. B., 214 Daigneault, L. E., 71 Day, M., 150 De Genova, R., 256 de Saja, J. A., 227 Dealy, J. M., 152 Dean, K., 146 Deeter, G. A., 155 Degée, P., 157 Denault, J., 150 Deng, X., 157 Deshmukh, V. G., 152 DeSimone, J. M., 144

61259_C009.indd 292

Author Index

Devalckenaere, M., 157 Di Maio, E., 144, 145, 146, 148, 152, 155–161, 164 Di, W., 145, 157, 158, 160 Di, Y., 150, 155, 156, 159, 161 Dietrich, C., 158 Dion, R. P., 259 Doelling, K. W., 19 Doermann, D., 276 Doherty, D. J., 275 Dooley, J., 18 D’Souza, N. A., 214 Dubois, P., 155, 157 Dufour, J., 71 Dumoulin, M. M., 71 Eastman, W. O., 14 Eaves, D. E., 222, 228, 233 Eberhardt, H. F., 256 Ebert, H. P., 256, 273, 276 Eicker, D. B., 11 Ema, Y., 150, 200 Fang, D., 15 Farmer, B. L., 188 Ferronea, M., 186 Ferry, J. D., 76 Fischer, M., 112 Fletcher, N. H., 208 Fleurent, H., 25 Floess, J., 256 Flory, P. J., 9, 111 Flumerfelt, R. W., 146, 149, 226 Franklin, K., 256, 261 Franklin, W. E., 2 Fratzl, P., 196 Fricke, H., 46 Fricke, J., 256, 273, 276 Friedrich, C., 158 Frisch, K.C., 2, 11, 3, 233, 272 Fu, T. L., 214 Fujimoto, Y., 145, 157, 176 Fujiwara, K., 226 Fukushima, Y., 157, 159, 176 Funami, E., 162, 196 Gandhi, K., 152 Gang, W., 161 Gao, F., 176

10/25/2008 12:40:31 PM

293

Author Index

Garcia, R. A., 45 Gaur, U., 145 Gay, Y. J., 144 Ge, J., 155 Gendron, R., 6, 71, 146, 163 Gerhartz, W., 46, 51, 62 Giannelis, E. P., 158, 159, 177 Gibson, L. J., 22, 215 Glicksman, L. R., 25, 272, 274–276 Goel, S. K., 206, 261 Gong, S., 163, 214 Gonzalez-Pena, J. I., 227 Gopakumar, T. G., 161 Gramann, P., 214 Greeley, T., 155 Griffin, J. D., 12 Gu, Z., 155 Gundert, F., 75 Guo, Z., 19 Gupta, M., 18 Gupta, M. C., 152 Habibi-Naini, S., 119, 134 Hamel, A., 163 Hampson, R. F., 5 Hamza, R., 22 Han, C. D., 71, 113 Han, J. H., 113 Han, X., 19, 163, 214, 261 Hansen, R. H., 45, 46 Hao, J., 157 Harkonen, M., 155 Harrison, K. L., 146 Hasegawa, N., 163, 176, 188, 190, 193 Heinemann, U., 256, 276 Heinz, R., 78, 81, 82, 100, 101, 102 Helminen, A., 155 Henne, A. L., 3, 11 Hensen, F., 58 Herrmann, T., 79 Hiltunen, K., 155 Hiroi, R., 157, 177, 178, 186 Hironaka, K., 157, 163, 176, 210 Hoff, G. P., 11 Hoffmann, B., 158 Hopfenberg, H. B., 146 Horn, B., 136 Houston, J. C., 46, 48 Howdle, S. M., 146

61259_C009.indd 293

Huneault, M. A., 33 Hung, M. L., 256 Hurd, R., 275 Hurnik, H., 51 Iannace, S., 144–146, 148–150, 152, 155–161, 164 Ikeya, M., 150, 200 Ilto, Y., 157, 163 Imeokparia, D. D., 256, 261, 262 Iqbal, M., 152 Ishii, H., 177 Ito, Y., 176, 209, 210 Jacobs, P. M., 52 Jacobsen, S., 256 Jaeger, A., 116 Jean, Y. C., 7 Jenkins, M. J., 146 Jérôme, R., 157 Jimenez, G., 157 Jinnai, K., 178, 181 Jinno, F., 4 Jun, J., 6 Kamigaito, O., 157, 159, 176 Kannah, K., 32 Karayannidis, G. P., 144 Karbas, H., 163 Kareko, L., 6 Kato, M., 176 Katz, J.L., 8 Kawai, H., 157 Kawasumi, M., 157, 159 Kay, W. B., 107 Kennedy, R. N., 16 Kihara, S., 149 Kim, M., 155 Kim, S. G., 226 Kimura, M., 178, 181 King, J. A., 274, 275, 276 Kirkland, C., 43, 51 Klempner, D., 144, 233, 272 Kodama, K., 256, 275, 276 Koelling, K. W., 163, 261 Kojima, J., 4 Kojima, Y., 157, 159, 176 Kontopoulou, M., 161 Koo, M. S., 276

10/25/2008 12:40:31 PM

294

Koppi, K. A., 18 Kosin, J. A., 45 Kotaka, T., 163, 188, 190, 193 Kreglewski, A., 107 Kretzschmann, G., 48, 50 Krikorian, V., 157 Krishnamoorti, R., 158, 188 Kriz, D., 152 Kropp, D., 71, 74, 75, 78, 157 Kropp, D., 46 Kudoke, C., 276 Kuhn, J., 273, 276, 279 Kullberg, R. C., 256 Kumar, V., 213 Kuppa, V., 188 Kurauchi, T., 157, 159, 176 Kurylo, M. J., 5 Kylma, J., 155 Lagaly, G., 177, 179, 181 Lai, S. M., 263 Laing, D. D., 116 Landel, R. F., 76 Lando, J. B., 22 Landrock, A. H., 43, 51 Lanzani, F., 256, 261 Lau, S.-F., 145 Lauterberg, W., 82 Lavengood, R. E., 260 Leach, A. G., 25 Lebedev, B., 145 Lee, H.-Y., 161 Lee, J., 7 Lee, J. A., 161 Lee, J. G., 226 Lee, J. W. S., 226, 265 Lee, L. J., 19, 163, 208, 214, 261 Lee, M. H., 163 Lee, P. C., 265 Lee, S. T., xii, 1, 6, 19, 33, 62, 80, 144 Lee, Y. H., 163 Lepoittevin, B., 157 Leppkes, R., 103, 106 Lesser, A. J., 261 Lester, G. R., 275 Li, C. C., 107 Li, G., 265 Li, J., 161 Li, W., 146, 149, 155

61259_C009.indd 294

Author Index

Liang, W. C., 256 Liao, X., 150 Lin, J. Y., 256, 259 Listemann, M., 22 Liu, X., 161 Lober, F., 43, 51 Lu, G. Q., 188 Lübke, G., 48, 103, 104, 105 Ma, C. Y., 71 Ma, P., 33 Macosko, C. W., 22, 144, 163, 214 Maiti, M., 176, 190 Maiti, P., 157, 163, 170, 188, 190, 192, 193 Malone, B. A., 256, 261 Manias, E., 159 Manias, E. J., 188 Manini, P., 256 Månson, J.-A. E., 150 Mao, J., 15 Margedant, J. A., 11 Maron, S. H., 22 Marrazzo, C., 148, 152, 155, 164 Martelli, F., 3 Martin, W. M., 45 Martìnez-Salazar, J., 151 Martini, J., 261 Maskara, A., 256 Masuda, Y., 256, 275, 276 Masuoka, H., 226 Mathieu, L. M., 150 Matsumoto, T., 155 Matsuoka, F., 155 Maurer, M. I., 256 Mauri, R., 256, 261 McCann, G. D., 256, 261, 262 McCarthy, T. J., 261 McNary, R. R., 3, 11 Meikle, J. L., 2 Menakanit, S., 188 Menges, G., 75, 108, 134 Mensitieri, G., 144, 146, 149 Merten, A., 276 Messersmith, P. B., 159 Michaeli, W., 71, 100, 102 Midgley, T., 3, 11 Minogue, E., 276 Mitsunaga, M., 157, 163, 176, 210 Miyamoto, M., 155

10/25/2008 12:40:31 PM

295

Author Index

Mohren, P., 136 Montjovent, M.-O., 150 Moore, S., 230 Morita, K., 6 Mork, S. W., 256, 261, 262 Mours, M., 152 Mouton, D., 274, 275 Muelhaupt, R., 158 Müller, E., 43 Munters, G., Muschiatti, L. C., 9 Naguib, H. E., 154 Nakano, S., 178, 181 Nakayama, T., 163 Nam, J. Y., 209 Nam, P. H., 163, 176, 188, 189, 190, 192, 193 Nanasawa, A., 259 Narayan, R., 28, 155 Narayanan, N., 149 Narkis, M., 151, 152, 153 Nasman, J. H., 155 Nawaby, A. V., 150 Ndiaye, P.A., 2 Ned Nisson, J. D., 26 Nelson, P., 163 Neumüller, O.-A., 42 Nguyen, T. J., 274, 275 Nicolais, L., 144–146, 148–150, 155–161 Nicolay, A., 113 Nie, L., 155 Niemi, M., 155 Niggemann, M., 76 Nishimura, S., 178, 181 Ogami, A., 145, 157, 159, 176, 183, 204 Ogata, N., 157 Ogihara, T., 157 Ohshima, M., 149, 162, 163, 196 Okada, A., 157, 159, 176 Okamoto, H., 163 Okamoto, K., 190, 214 Okamoto, K. T., 18, 133, 157 Okamoto, M., 145, 150, 157–159, 162, 163, 176–178, 183, 186, 188, 190, 192, 193, 195, 196, 200, 209, 210, 214, 215

61259_C009.indd 295

Ortner, L., 11 Ostrogorsky, A. G., 274 Padareva, V., 48 Pantoustier, N., 157 Panzer, U., 154 Paquet, A. N., 274 Parent, J. S., 161 Park, C. B., 19, 62, 90, 144, 154, 163, 226, 257, 261, 265, 274, 275 Park, C. P., 256, 261, 262 Park, E., 155 Parks, D. M., 188 Partridge, R., 245 Pastor, D., 151 Paul, M.-A., 157 Peòn, J., 151 Perlon, U., 11 Pfannschmidt, L. O., 71, 108, 109, 111 Phelan, R., 25 Pielichowski, J., 274 Pierick, D. E., 116 Pioletti, D. P., 150 Pittman, C. U., 152 Placido, E., 276, 279 Pluta, M., 159 Pochan, D., 157 Poling, B. E., 112 Pop-iliev, R., 261 Popov, N., 132 Prausnitz, J. M., 112 Prociak, A., 274 Puri, R. R., 233 Ramesh, N. S., 19, 62, 144 Rauwendaal, C., 16 Ray, S. S., 145, 157, 163 Raynaud, M., 276 Reed, D., 20 Reichelt, N., 154 Reid, R. C., 112 Reignier, J., 6, 146 Reitz, D. W., 274, 276 Ren, X., 146 Rimura, Y., 155 Rinke, H., 11 Rizzi, E., 256 Roberts, G. W., 146 Rodeheaver, B. A., 262

10/25/2008 12:40:31 PM

296

Rodriguez-Perez, M. A., 227 Rogalla, A., 71 Rotermund, U., 276 Rouanet, S., 256 Roux, J. D., 274, 275 Roychoudhury, P. K., 149 Royer, J. L., 144 Rubens, L. C., 12 Runt, J., 159 Russell, T. P., 261 Rutledge, G. C., 188 Sacadura, J. F., 276 Sahnoune, A., 163 Sain, M., 226 Saito, T., 186 Salovey, R., 152 Sander, B., 112 Sander, S. P., 5 Sanguigno, L., 159 Sasaki, T., 178, 181 Sato, Y., 226 Saunders, J. H., 3 Schick, C., 145 Schild, H., 11 Schlack, P., 14 Schmidt, R., 112 Scholz, D., 43, 44, 48, 50, 51, 59, 63 Schröder, T., 118 Schümmer, P., 72 Schuetz, M. A., 272, 274, 276 Schulz, G., 46, 51, 62 Schwab, H., 256 Scott, R. M., 52, 55 Seibt, S., 71 Selin, J. F., 155 Semerdjiev, S., 132, 135, 136 Sendijarevic, V., 144 Seppala, J., 155 Severini, T., 9 Shakesheff, K. M., 146 Shen, J., 19, 163, 208, 214 Sheng, N., 188 Shenoy, A. V., 158 Shikuma, H. E., 149 Shinno, K., 155 Shinoda, H., 6 Shiroi, T., 157, 177, 178, 186

61259_C009.indd 296

Author Index

Shmidt, C. D., 256, 261 Shutov, F. A., 103, 115 Siefken, W., 11 Silva, M. M. C. G., 146 Sinha Ray, S., 157–159, 176–178, 183, 186, 204, 209, 210, 214 Sinofsky, M., 276 Sinsawat, A., 188 Siripurapu, S., 144 Skochdopole, R. E., 276 Smith, D. M., 256 Snyder, A. J., 159 Sodergard, A., 155 Soderquist, M. E., 259 Sophiea, D. P., 256 Sorrentino, L., 145 Sosa, J. M., 259 Spalding, M. A., 18 Spiekermann, R., 51, 62, 103, 104 Spitael, P., 144 Srivastava, A., 149 Stadlbauer, M., 77 Stafford, C. M., 261 Standish, R. K, 188 Staples, T. G., 256, 261, 262 Sterzynski, T., 274 Stevens, R., 22 Stevenson, J. F., 116 Stobby, W. G., 256, 261, 262 Strauss, W., 214 Suetsugu, Y., 263 Suh, K. W., 256, 261 Suh, N. P., 20, 90, 113, 201, 204, 226, 261 Sumarno, T. S., 226 Sung, W. F., 256 Takada, M., 163, 202 Takada, T., 4 Takemura, K., 178, 181 Taki, K., 149, 162, 196 Tandberg, J. G., 11 Tanimoto, Y., 256, 275, 276 Tao, W. H., 256 Tateyama, H., 178, 181 Tatibouët, J., 146, 163 Thomann, R., 158 Throne, J. L., 62, 81

10/25/2008 12:40:31 PM

297

Author Index

Todd, D., 22 Tomasko, D. L., 163, 261 Topey, M. R., 272, 273, 275, 276 Toth, R., 186 Trageser, D. A., 233 Traugott, T. D., 259 Tsai, S. J., 256 Tseng, C.-R., 161 Tsunematsu, K., 178, 181 Tuominen, J., 155 Turco, G., 155 Turng, L. S., 163, 214 Tusim, M. H., 256 Uchiki, K., 6 Ueda, K., 145, 155, 157, 159, 176, 183, 204 Urchick, D., 12 Usuki, A., 157, 159, 163, 176, 188, 190, 193 Vachon, V., 11 Vaia, R. A., 177, 188, 214 Vaillamizar, C. A., 71 Van der Weide, I., 241 Van Krevelen, D., 103, 106 van Olphen, H., 190 Vega, J., 151 Venter, R. D., 261 Verbist, G., 25 Villalobos, M. A., 155 Vo, C. V., 274 Vollalobos, M. A., 155 Wakili, K. G., 256 Walczak, K., 18 Waldman, F., 261 Wallerstein, R., 151, 152, 153 Wang, J., 146 Wang, K. H., 163 Wason, S. K., 45, 63 Watanabe, M., 178, 181 Weaire, D., 25, 214 Weller, J. E., 213 Welty, J. R., 25 Werth, M., 241 Whelan, J., 33 Whitaker, M. J., 146

61259_C009.indd 297

White, J. L., 263 Whitehead, K. N., 274, 275, 276 Whitfield, P., 150 Wicks, C. E., 25 Widya, T., 163, 214 Williams, M. L., 76 Williams, R. J. J., 274, 276 Wilson, R. E., 25, 72 Wingert, M. J., 19 Winter, H. H., 152 Wirtz, H., 14 Witten, N., 227, 239 Wolf, B. A., 75 Wong, C. M., 256, 263 Wood-Adams, P. M., 152 Wortberg, J., 73 Worthoff, R. H., 72 Wu, J., 274, 275 Wu, J. W., 256 Wu, J.-Y., 161 Wu, Q., 161 Wu, S., 112 Wu, Z., 161 Wübken, G., 136 Wunderlich, B. B., 145 Xanthos, M., 144 Xu, D., 261 Xu, G., 214 Xu, Q., 146 Xu, R., 159 Xu, X., 261 Yamada, K., 145, 157, 159, 183, 204, 214 Yamamoto, M., 186, 209 Yanagimoto, T., 162, 196 Yang, J., 7 Yang, Y., 155 Yano, K., 159 Yevstropov, A., 145 Ying, C. H., 256 Yoon, J., 155 Yoshida, O., 178, 183, 186 Youn, J. R., 276 Young, M. W., 144 Yu, A. B., 188 Yu, L., 146

10/25/2008 12:40:31 PM

298

Yuan, M., 157, 163 Yuge, K., 256, 275, 276 Zang, Y., 155 Zarr, R. R., 274, 275 Zeng, C., 163, 208, 214 Zeng, Q. H., 188

61259_C009.indd 298

Author Index

Zhang, X., 22 Zhong, W., 155 Zhou, C., 161 Zhu, N., 161 Zhu, Z., 257, 274, 275 Zhu, Zhengjin, 26 Zong, J., 257, 274, 275

10/25/2008 12:40:31 PM

Subject Index

A Absorption, 2 Accumulator, 51, 53 Acid, 44 ascorbic, 44 citric, 46 citric esters, 44 fumaric, 44 gluconic, 44 glutaric, 44 lactic, 44, 149 numeric, 46 oxalic, 46 succinic, 44 tartaric, 46 Acidic salts, 44 Acoustic, 27, 34, 238, 245 Acrylonitrile, 259 Acrylonitrile-butadiene-styrene (ABS), 53, 54 Adhesive, 257 Aging, 274 Air, 42, 45, 238, 274 Alpha-methyl styrene, 259 Alumina (Al2O3), 164–166, 168 Aluminum (AL) foil, 256 Aluminum mold, 50 American National Standard Institute (ANSI), 240

61259_C010.indd 299

Ammonium cation, 198 Amorphous, 12, 108, 109, 259 Anti-thixotropy, 189 Areospace, 239, 243 Ascorbic acid, 44 ASTM, 22 D-2872, 27, 35 D-6400, 35 D-6868, 35 D-7021, 35 E-84-01, 240 E-90, 27 E-413, 27 E-492, 27 E-989, 27 Autoclave, 219–221, 231, 234 high pressure, 225 low pressure, 227 Automotive, 4, D8462, 239, 242 Azodicarbonamide (ADC), 30, 45, 51, 52, 55, 56, 57, 58, 63, 105, 230, 232

B Barrel, 59 grooved, 59, 60 smooth, 60 Biodegradable, 36, 144, 145, 149

10/25/2008 12:40:45 PM

300

Biomedical, 145 Bioplastics, 36 Block process, 229 Blow molding, 62 Blowing Agents, 8, 19, 54, 55, 69, 70, 90, 109, 114–116, 118, 121, 109, 111, 113, 116, 123, 130, 169, 225, 277 chemical, 13, 30, 45, 49, 50, 51, 54, 55, 56, 230, 234 physical, 30, 101, 104, 105, 106, 230 Boiling point, 47 Boric acid, 45, 46 Branching, 144, 151 Breaker plate, 82 breathing mold, 137, 138 Brownian movement, 112 Bubble formation, 79 Bubble nucleation, 169, 198 Bubble pressure, 7 Bun, 229 Butadiene, 259 Butane (n-butane), 32, 47 1,4 butanediol (BD), 155 1,4 butane diisocyanate (BDI), 155, 156

C Cable insulation, 58 Calcium, 46 Calcium carbonate (CaCO3), 46, 263–266, 268, 269, 279, 285 Calcium lactate, 44 Capillary rheometer, 71, 75 Carbon Dioxide, 6, 11, 12, 31, 32, 35, 37, 42, 45–47, 54, 69, 70, 76–80, 82, 90, 95, 105–107, 145–148, 159, 164, 167, 191, 196, 198, 200, 203, 204, 206, 208–210, 212, 260, 261, 274, 279–282, 285 solubiliity, 12 Carbon Monoxide, 32 Carbon nanofibers, 208 Carbon nanotubes (cn), 164–166 Carboxylic acid, 44 Carpet backing, 58 Catalyst, 8, 10, 236 Cation exchange capacity (CEC), 177, 178

61259_C010.indd 300

Subject Index

Cavity wall, 132 Cell density, 93, 167, 168, 192, 203, 205, 211, 277 distribution, 13, 203 formation, 200, 208 morphology, 128 nucleation, 201 size, 22, 93, 131, 163, 168, 191, 203, 211, 268, 271, 279, 280 structure, 54, 62, 129, 226, 273 wall, 196, 201, 203, 211, 212, 275, 278 Cellulose, 26 Ceremic, 214 Chemical blowing agents (CBAs), (See Blowing agents) Chemical foaming agent (CFA), 260 Chemical nucleating agents, 31 Chemical potential, 111 Chlorofluorocarbons (CFCs), 3, 5, 11, 32, 70 CFC-11, 47, 48, 105 CFC-12, 47, 48, 274 CFC-113, 47 CFC-114, 47 Chlorotrifluoroethylene (CTFE), 240 Citric Acid, 13, 44, 46, 49, 83, 105 Cleanroom insulation, 245 Closed cell, 238, 243, 261, 276, 284 structure, 191, 209, 226, 279 n-(coco alkyl)N, N-[bis(2hydroxyethyl)]N-methyl ammonium, 179, 181 Co-extrusion, 61 Co-injection, 53 Color measurement, 242 Compounding, 19 Construction, 4, 238 Convection, 272 Cooling, 262 Cooling time, 123–125 Corn, 36 Corn field, 34 Cosmetics application, 52 Critical bubble radius, 7 Critical pressure, 74

10/25/2008 12:40:45 PM

301

Subject Index

Cross-linked, 42 high-density PE, 236 low desnity PE, 223, 231 Metallocene, 236 PE, 42 polyolefin, 220, 229 process, 19, 232, 238, 251 Cross-linking, 144, 151, 222, 224, 247 Crystallization, 134, 145, 157, 161, 164, 165, 167 Curing, 13 Cushioning curve, 2, 237 Cycle time, 54, 120, 123, 136, 139 Cyclopentane, 47, 105

D Degassing, 60 Degradation, 3, 35 Demolding, 124, 125 Density, 107, 273 of CO2, 76 reduction, 54, 103, 121, 122, 126 Depressurization, 226 Design of experimentation (DOE), 123, 126 Devolatilization, 22 Dicumyl peroxide (DCP), 152–154 Die, 45, 61, 70, 262 annular, 82 coathanger, 61 dual-spider, 81, 82 flat, 61 gap, 91, 92 in-line rheometer, 72, 78 lip geometry, 91 profile, 61 single-spider, 81, 82 temperature, 86, 87, 89 tubular, 61 Differential Scanning Calorimeter (DSC), 165, 260 Diffusion, 7, 46, 108, 112, 130 coefficient, 112, 274 Diffusivity, 156, 168, 196, 198 Dimer, 9 dimethyl 2-ethylhexyl ammonium cation (MMT-Alk), 157 Dinitroso pentamethylene tetramine (DNPT), 52

61259_C010.indd 301

dioctadecyl dimethyl ammonium, 2C18(CH3)2N+, 179, 182–186 diphenyl methylene diisocyanate (MDI), 10 Direct gassing , 53 Discoloration, 44, 52, 54, 57 Disodium pyrophosphate, 45, 46 Dynamic mechanical thermal analysis (DMTA), 249, 250

E Electron beam, 13 Electronics, 239 Electrostatic, 190 Elongational viscosity, 193 Emission, 32, 33 EN 13432, 35 Endothermic (Endo), 44, D14854, 55, 62, 63 Energy absorption, 23 Enexothermal (Enexo), 54, 55, 57 EPS (See polystyrene) Equimolar, 49 ErgoCell, 116, 117 Ethanol, 6, 47 Ethyl chloride (EtCl), 261 Ethylene ethyl acrylate, 12 Ethylene methyl acrylate (EMA), 233 Ethylene octane (EO) copolymer, 262 Ethylene styrene interpolymer (ESI), 262 Ethylene vinyl acetate (EVA), 51, 64, 223, 228, 229, 233 Exfoliate, 160, 164, 186 Exothermic (Exo), 54, 55, 63 Expanded rubber cylinder, 222, 223 Expansion coefficient, 75 Expansion joints, 238 Extinction coefficient, 275, 276 Extruder, 45, 59, 60, 115, 262 single screw, 59 Extrusion, 11, 18, 19, 30, 33, 50, 57, 58, 59, 70, 82, 113, 224, 225, 259 die, 90

F FAR 25.856, 241 Fermentation, 33, 34

10/25/2008 12:40:45 PM

302

Fiberglass, 27, 29, 30, 258 Fick’s law, 112 Fillers, 281, 285 Finite element method (FEM), 79 Flame retardancy, 15 Flame spread index (FSI), 240 Floating, 238 Floating devices, 244 Flory-Huggin, 111 Fluorocarbon, 71, 280, 281, 285 Flroroectorite, 163 Fluoropolymer, 239 FM 4910, 241, 245 FM 4924, 246 Foam, 1, 3, 4 Blow molding, 62 Cross-linked, 62 Density, 21, 161, 168, 263, 269, 270, 285 Extrusion, 14, 58, 61, 69, 70, 75 PET, 14 PP, 14 Injection molding (FIM), 101, 102, 104, 114, 120, 131, 139 Isocyanurate, 15 low-density, 81 netting, 58 Phenolic, 15 polyethylene (See Polyethylene) polylactide, 6 polyolefin (See Polyolefin) Polypropylene (See Polypropylene) Polystyrene (See Polystyrene) Polyurethane (See Polyurethane) phenolic, 15 process, 226 quality, 78 structure, 70, 108, 122, 127–129, 139 thermoplastic, 9, 19 thermoset, 9 Foam-in-mold, 14 Foam-in-place, 14, 15 Foaming, 8, 16, 19, 33, 262 efficiency, 165 extrusion, 9 oven, 232 path, 9 Polyurethane, 14 technologies, 18

61259_C010.indd 302

Subject Index

temperature, 201–203, 264–266, 269–271 x-linked PE, 4, 12 Food packaging, 52, 58, D235259 Forming zone, 59, 60 Forurier Transform Infrared Spectrometer (FTIR), 276 Fossil, 3 Free energy, 111 Fruit acid, 44 Fumaric acid, 44

G Gas-assisted injection molding (GAIM), 107 Gas conductivity, 274 Gas counter, 53 process, 135–138 Gas evolution, 54 Gas pressure, 54, 55 Gaussian distribution, 200 Gear pump, 71 General purpose polystyrene (GPPS), 46, 48, 49, 50 Getter, 284 Glass transition temp. Tg, 8, 11, 202, 249, 259, 260 depression, 202 Global warming, 5, 36 Global warming potential (GWP), 47 Gluconates, 44 Gluconic acid, 44 Glutaric acid, 44 Grooved feeding, 59

H Half-time, 145 Health care, 219 Heat distortion temp. (HDT), 260 Heat resistance, 25 Henry’s equation, 110 Heterogeneous nucleation, 112, 113, 161, 201, 204, 205, 210 n-hexadecyl tri-n-butyl phosphonium, 183

10/25/2008 12:40:45 PM

303

Subject Index

n-hexadecyl tri-phenyl phosphonium, 183 Hexafluoropropylene (HFP), 240 Hexagonal, 283 HMS-PP (high melt strength PP), 77 Homogeneous nucleation, 112, 201 Horizontal oven, 235 Hydrocarbons (HC), 31, 32, 247, 260 Hydrocerol, 48, 83, 92, 94, 95 Hydrochlorofluorocarbons (HCFCs), 5, 31, 32, 70 HCFC-22, 47, 105, 261, 274 HCFC-142b, 47, 48, 105, 261, 276, 277 Hydrofluorocarbon (HFC), 260 HFC-134a, 47, 105, 261, 274, 276, 277 HFC-152a, 47, 261, 276, 277 HFC-245fa, 47 HFC-365mfc, 47 Hydraulic press, 51, D348234 Hygroscopic, 246

I Ionomer, 84 Impact insulation, 29 Impact Insulation Coefficient (IIC), 29 Impact sound pressure, 29 Incineration, 3, 36 Infrared (IR), 123, 126, 127 Injection molding, 4, 30, 50, 54, 101, 102, 104, 120, 126, 131, 139 gas-assisted, 111 Injection nozzle, 117, 119, 120 Injection velocity, 121, 126, 128–131, 139 In-line rheometer die, 72, 78 Insulation, 2, 26, 34, 238, 243, 245, 248, 258 Interlayer opening, 182 Irradiation, 151 ISO 845 (desnity), 236 1408, 27 1798 (tensile/elongation), 236, 250 7214 (compression), 236, 250 8067 (tear), 236, 250 17088, 35 Isobutane, 47, 105, 274 Isocyanurate foam, 15 Isopentane, (I-pentane), 47, 105, 274

61259_C010.indd 303

Isothermal, 274 Izod, 260

L L-lactic (L-lactide), 149, 176, 177 Lactic acid, 44 Lamellar, 209 Landfill, 3, 36 Lavorazione Materie Plastiche (LMP), 3 Layered silicate, 176, 177 Layered titanate (HTO), 177, 178, 181, 182 Leisure, 219 Length to diamer, L:D, 59 Life cycle, 3, 37 Life Cyle Assessment, 2 LLDPE (See [PE]) Long chain branching (LCB), 151 Low density polyamide foam, 251 Low smoke (LS), 241

M Malic acid, 44 Mandrel support, 82 Material safety data sheets (MSDS), 30 Medical, 219, 244 Melt filter, 58, 59, 61 Melt flow index (MI), 260 Melt temperature, 84–86, 89, 122, 126, 129, 131, 139 Metering zone, 59, 60 methyl bis(2-hydroxy-ethyl) ammonium cation (MMT-(OH)2), 157, 159, 162, 163 methyl dihydrogenated tallow ammonium (MMT-2HT), 157 Methylal, 47 Microcellular, 14, 202, 226 Mineral oil, 259 Mixing zone, 116 Modulus, 195, 213, 249 Mold, 51, 53 Mold foaming, 18

10/25/2008 12:40:46 PM

304

Mold temperature, 126, 128, 130, 131, 139 Molded bead, 14, 18, 30, 32, 33 Molecular dynamics (MD), 186, 188 Monocalcium phosphate, 45 Monopotassium tartrate, 44 Monosodium citrate, 44 Montmorillonite (MMT), 157, 159, 162, 163, 176, 178, 180, 182, 183, 187–190, 196, 200, 201, 204, 206, 208, 209, 214 Morphology, 9, 148, 154, 208 MuCell, 116

N Nanoclay, 201, 209, 211, 212 Nanocomposite, 143, 156, 157, 164, 175, 183, 184, 190, 194, 200, 209, 211, 214 Nanofiller, 178, 190 Nanogel, 278 Newtonian, 72 Nitrogen, 42, 45, 47, 105, 107, 145–148, 150, 167, 219–221, 231, 260, 281, 282, 285 autoclave, 221, 225, 229, 234 Noise reduction coefficient (NRC), 29 Nucleating agent, 93–96, 166, 167, 169 Nucleation, 7, 8, 9, 19, 48, 80, 113, 128, 164, 196, 201, 226 heterogeneous, 90 Nylon (NY) A403 (see Polyamide)

O OBSH, 52 Octadecyl ammonium, C18H3N+, 178, 179, 182–185 Octadecyl di-methyl benzyl ammonium, 183 Octadecyl trimethyl ammonium, C18(CH3)3N+, 179, 182–185 Open-cell, 256, 265, 267, 278, 284, 285 structure, 8, 23 modulus, 23 Optifoam, 118–120 Order (See smell)

61259_C010.indd 304

Subject Index

Organically modified layered filler (OMLF), 175, 177, 180, 182, 183, 186 Outgassing, 52 Oxalic acid, 44 4,4-Oxybis(benzenesulfonyl hydrazide), OBSH, 51, 52, 105 Oxygen permeability, 159 Ozone, 33 Ozoe depletion potential (ODP)

P Packaging, 4, 31, 34, 62, 237, 259 PBAs (See Blowing agents [PBAs]) PE (See Polyethylene [PE]) Pentane (n-pentane), 11, 33, 47, 274 Permeability, 274 Peroxide, 144, 151, 224 Petroleum, 2, 5 Pharmaceutical, 52 Pharmacopoeia Monograph USP 661, 244 Phase separation, 13 Phenolic, 15, 29 5-Phenyl tetrazole (5-PT), 51, 52, 105 Physical blowing agents (PBA), 30, 101, 104–106 Pipe insulation, 246 Plasticizing, 77 PMMA (See Polymethyle methacrylate) PO (See Polyolefin [PO]) Polyamide, 3, 9, 11, 20, 163, 239, 256, 257 Polyamide 6, 161, 239, 246, 250, 251 Polyamide 6,6, 161 Polyamide 12,12, 161 Polycaprolactone (PCL), 36, 143, 144, 146–151, 153, 154, 157–165, 167 Polycarbonate (PC), 125, 127, 163, 176, 209–212 Polycarboxylic , 44 Polycondensation, 10 Polyester, 9, 149 Polyethylene (PE), 6, 12, 33, 46, 64, 71, 151, 161 high-density, 12, 58, 75, 163, 233, 256 linear low density (LLDPE), 63

10/25/2008 12:40:46 PM

305

Subject Index

low-density, 12, 46, 48, 49, 58, 75, 77, 79, 83, 84, 89–91, 106, 108, 223, 224, 226, 227, 233, 238, 266, 267, 269–271, 280–282, 285 medium density (MDPE), 233 metallocene (mPE), 233 Polyethylene terephthalate (PET), 9, 58, 256, 257 Polyhydroxyalkanoates (PHA), 144 Polyhydroxy carboxylic acid, 44 Polyisocyanurate foams, 6, 10 Polylactic acid (PLA), 6, 33, 35, 58, 143, 144, 149–151, 154–157, 159, 162, 163, 168, 176, 183, 185, 186, 200, 201, 203, 204, 206, 208, 209, 214 Polymethyl methacrylate (PMMA), 57, 206 Polymer/layered silicate nanocomposite (PLSNC), 176, 177, 190, 215 Polyolefin, 4 Polyphenylene ether (PPE), 53 Polyphenylene oxide (PPO), 53 Polypropylene (PP), 12, 42, 58, 75, 77, 78, 89, 121, 132, 176, 187, 233 binder, 33 HMS-, 77, 83 maleated (PP-MA), 187, 192–196, 198, 199 Polystyrene (PS), 3, 4, 11, 12, 18, 29, 33, 46, 49, 58, 71, 75, 163, 238, 256, 257, 260–264, 266–273, 279, 282–285 bead, 11, 14 molded bead, 4 open cell, 278 syndiotactic, 161 Polyurethane (PU), 3, 4, 10, 19, 20, 27, 256, 259 board process, 17 flexible, 10, 14, 18, 29 elastomeric, 10 integral skin, 14 open cell, 278, 279 rigid, 10, 14, 18, 29 spray process, 17 wedge, 27 Polyvinyl alcohol (PVOH), 35, 36

61259_C010.indd 305

Polyvinylchloride (PVC), 4, 42, 58, 106 plasticized, 62 Polyvinylidene fluoride (PVDF), 239–244 PP (See Polypropylene [PP]) PP-based nanocomposite (PPCN), 188, 189, 191–196, 198, 199 Porous ceramic, 214 Prefoaming, 58, 61 Premature foaming, 78 Pressure distribution, 80 Pressure drop, 113, 186 rate, 148 Pressure gradient, 81, 91, 92, 94, 95, 113 Pressure quench method, 149 Pressure release valve, 115 Propane, 47 PS (See Polystyrene [PS]) PU (See Polyurethane [PU]) p-v-T (pressure-volume-temp.) plot, 75

R R-Value (See Thermal resistance) Radiation, 274 Reactive foaming, 18 Reactive foaming, injection molding (RIM), 14, 16 Recreation, 4 Recycle, 36 Refrigerating application, 258 Reuse, 36 Rheology, 70, 154, 157 Rheometer (Also see in-line rheometer), 189 Rheopexy, 189, 190 Rotational molding, 62, 63 Rubber, 10, 64 Rupture, 262

S Salt bath, 230, 233, 237 Sandwich process, 53 Sapphire window, 197 Scanning electron microscopy (SEM), 162, 163, 191, 200, 207, 209, 210, 215, 227, 244, 273

10/25/2008 12:40:46 PM

306

Seals, 239 Semicarbazide, 57 Semi-continuous process, 30, 232 Semi-crystallization, 12 Shape factor, 23 Shark skin, 58 Shear-thinning, 72, 78 Shear viscosity, 188, 189 Short chain branching, 151 Shut-off nozzle, 116, 117, 119 Silica, 46 Silver streaks, 132, 133, 136 Sink mark, 54 Sinter metal, 119 Smell, 44, 54, 57 Smog, 32,33 Smoke developed index (SDI), 240 Sodium acid pyrophosphate, 45, 46 Sodium aluminum phosphate, 45, 46 Sodium aluminum sulfate, 45 Sodium bicarbonate, 13, 49, 52, 57 Sodium borohydride, 57 Sodium carbonate, 42, 83, 105 Solid state shear processing, 186 Solubility, 11, 21, 46, 110, 148, 156, 160, 168, 196 sound absorption, 34, 37 Sound transmission loss, 28 Spider, 82–84, 88 geometry, 88, 89 legs, 70, 78, 81, 83, 87 Spinoidal decomposition, 13 Sports, 219 Starch, 36 Static mixer, 119, 120 Steinen Tunnel, 240 Strain-induced hardening, 188, 190 Stress-strain curve, 196 Styrene, 259 Styrene/ethylene butylene/ styrene (SEBS), 262 Succinic acid, 44 Sulfohydrazide, 51, 57 Supercritical fluids, 107, 225 state, 106, 107 Supersaturation, 113, 191 Surface temperature, 123, 124, 127 Surface tension, 7, 112

61259_C010.indd 306

Subject Index

T Talc, 46, 92, 93, 95, 164–166, 168 Tartaric acid, 44 Tensile, 242, 260 tetrafluoroethylene (TFE), 240 Tetrazoles, 57 Thermal conduction, 275 Thermal conductivity, 26, 46, 243, 248, 252, 272, 274, 277, 281, 282, 284, 285 Thermal energy, 275 Thermal insulation, 238, 245 Thermal oxidizer, 32, 33 Thermal resistance, 26 Thermodynamic instability, 111 instability, 13 Thermoforming, 27, 34 Thermoplastic foam, 9 thermoset vs., 9 Thermoset foam, 9 Titanium dioxide (TiO2), 164, 166, 168 Toluene diisocyanate (TDI), 10 Toluene sulfonyl hydrazide (TSH), 52, 105 Toluene sulfonyl semicarbazide (TSSC), 52, 105 Transportation, 4, D45862, 245 2,4,6-Trihydrazino-1,3,5-triaazine (THT), 52 T-Tubes, 245

U UL 94 V-O, 240, 241 UL 723, 240 Ultraviolet light (UV), 12, 13, 62, 240, 252 Underfloor (or underlay), 238 Union Carbide Corp. Process (UCC), 50, 51

V Vacuum Insulation Panel (VIP), 255–258, 280–285 Vacuum packaging, 257 van der Waals +A230, 2, 9

10/25/2008 12:40:46 PM

307

Subject Index

Velocity distribution, 80 Vertical oven, 230, 232, 235 Vinyl acetate monomer (VAM), 36 Vinyl alcohol (EVOH), 257 Vinylidene fluoride (VDF), 240 Viscosity, 70, 72, 74, 76, 77, 78, 107, 108, 123 shear, 69, 71, 108 Volatile organic chemical (VOC), 251 Volume expansion ratio (VER), 278

Wide angle X-ray diffraction (WAXD), 177, 180, 181, 183, 185 Wood plastics, 62

X X-linked, 30 X-PE (or XL-PE), 4, 14, 247, 248 X-ray, 126, 127, 197 Xenotest, 242

Y Yellowness index, 242

W Water, 47, 105 Water soluble, 36, 37

61259_C010.indd 307

Z Zotefoams, 220

10/25/2008 12:40:46 PM

61259_C010.indd 308

10/25/2008 12:40:46 PM