Fiber reinforced phenolic foam: climatic effects on mechanical properties and building applications in northern Thailand

PDF

Download

Open document

Flip pages

Summarize

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content FIBER REINFORCED PHENOLIC FOAM: CLIMATIC EFFECTS ON MECHANICAL PROPERTIES AND BUILDING APPLICATIONS IN NORTHERN THAILAND by John Paul Basbagill A Thesis Presented to the FACULTY OF THE SCHOOL OF ARCHITECTURE UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree MASTER OF BUILDING SCIENCE May 2008 Copyright 2008 John Paul Basbagill ii Dedication This thesis is dedicated to my grandfather, John Marzinski, who passed away during the early stages of this project. My grandfather was a lifelong builder and engineer and an inspiration in helping me seek my interests. He laid down the bricks on this journey, and I follow in his footsteps. ~ ever the engineer ~ iii Acknowledgments I would like to thank my thesis committee for their patience and support. My thesis chair, Professor Goetz Schierle, provided me with technical guidance and timely advice on structural topics; Professor Thomas Spiegelhalter offered help on sustainability issues; Professor Marc Schiler listened to my grievances throughout the year, helped me keep the “big picture” in perspective, and kept me on course; and Professor Steven Nutt allowed me to conduct my project in the composite materials department in the Viterbi School of Engineering at the University of Southern California. I am grateful to the department for providing funding for the project. Professor Nutt helped me in narrowing my project topic, and his in depth knowledge on materials science topics was invaluable. I am also indebted to Warren Haby, whose handyman expertise in the composite materials laboratory got me through several unforeseen challenges while conducting my experiments. Finally, I would like to thank Amit Desai and Hongbin Shen, current and former composite materials doctoral students at USC, from whose previous work I was able to build and who provided helpful tips in my project methodology. Table of Contents Dedication ii Acknowledgments iii List of Tables viii List of Figures xi Abstract xiii Chapter 1: Introduction 1 1.1 The Problem 3 1.2 Potential Solution 3 1.3 Objective 4 1.4 Final Product 5 1.5 Hypothesis 5 1.6 Scope 5 1.6.1 Base Material 7 1.6.2 Reinforcement Fibers: Synthetic Versus Natural 8 1.6.2.1 Synthetic Fibers 9 1.6.2.2 Natural Fibers 9 1.6.3 Tests 10 1.6.4 Thai Climate Simulations 10 1.6.5 Comparison Materials 12 1.6.6 Environmental Impact 12 1.6.7 Costs 13 1.7 Applications 13 Chapter 2: Background 14 2.1 The Problem 14 2.1.1 Living Conditions in northern Thailand 15 2.1.1.1 Climate 15 2.1.1.2 Thai Climate, Economy, and Building Practices 17 2.1.2 Insulation and the Environment 19 2.1.3 Insulation Materials: Overview 20 2.1.3.1 Cellulose 21 2.1.3.2 Mineral Wool 22 2.1.3.3 Fiberglass 23 2.1.3.4 Polystyrene 23 2.1.3.5 Polyisocyanurate 24 2.1.3.6 Radiant Barriers 24 2.1.4 Insulation Materials: Other Environmental Factors 24 2.2 Cladding 25 2.3 Structural Insulation 26 2.4 Phenolic Foam 27 2.5 Previous Work 28 Chapter 3: Methodology 30 3.1 Fabrication 31 3.1.1 Materials 31 3.1.2 Mixer 32 3.1.3 Fabrication Process 33 3.1.4 Problems 35 3.1.4.1 Thermal Control 35 3.1.4.2 Batch Process 36 3.1.4.3 Fiber Dispersion 36 3.1.4.4 Non-Uniform Density 36 3.1.4.5 Low Density Values 39 3.2 Cutting Samples 39 3.3 Density Measurements 40 3.4 Climate Conditioning 40 3.4.1 Climate Simulation I: Water Immersion 40 3.4.2 Climate Simulation II: Accelerated Aging 42 3.5 Testing 43 3.5.1 Test I: Compression 44 3.5.2 Test II: Shear 45 3.5.3 Test III: Cell Length (SEM) 47 3.5.4 Test IV: Flammability 48 3.5.5 Test V: Conductivity 49 3.6 Results 53 3.6.1 Mechanical and Thermal Tests 54 3.6.2 Environmental Impact Assessment 54 3.6.3 Costs 55 Chapter 4: Results 56 4.1 Density 57 4.2 Strength Tests 60 4.2.1 Compression 60 4.2.1.1 Can Phenolic Foam Be Used as a 62 Load-Bearing Material in Thailand? 4.2.1.2 Do Climate Stresses Weaken Phenolic 65 Foam? 4.2.1.3 Do Natural or Synthetic Fibers Add Greater 65 Strength to Phenolic Foam? 4.2.2 Shear 65 4.3 Climate Simulations 67 4.3.1 Water Immersion 68 4.3.2 Accelerated Aging 71 4.4 Other Tests 72 4.4.1 Cell Length (SEM) 72 4.4.2 Fire Resistance 75 4.4.3 Conductivity 76 4.5 Summary 81 Chapter 5: Environmental Impact 82 5.1 Prior Environmental Impact Studies on Insulation 83 5.2 Difficulty in Assessing the Environmental Impact of 83 Phenolic Foam 5.3 Method 84 5.4 Results: Environmental Impact Assessment of Phenolic 85 Foam 5.4.1 Heat Loss 85 5.4.1.1 Building the Model 86 5.4.1.2 Calculating Heat Loss 88 5.4.2 Moisture Resistance 90 5.4.3 Embodied Energy 92 5.4.4 Emissions 93 5.4.4.1 Phenolic Foam 94 5.4.4.2 EPS 97 5.4.4.3 Phenolic Foam Versus EPS 97 5.4.5 Toxicity 98 5.4.5.1 Phenolic Foam 98 5.4.5.2 EPS 99 5.5 Summary 99 Chapter 6: Costs 101 6.1 Method 101 6.2 Production Site 102 6.3 Construction Site 102 6.4 Suppliers 104 6.5 Raw Materials 105 6.6 Transportation 107 6.6.1 Cost to Ship Raw Materials to the Production Site 107 6.6.2 Cost to Ship End Product to the Construction Site 108 6.7 Machinery 109 6.8 Labor 110 6.8.1 Manufacturing 110 6.8.2 Installation 111 6.9 Electricity Costs Associated with Production 111 6.10 Lifetime Energy Savings 112 6.11 Summary 113 Chapter 7: Conclusion 115 7.1 Mechanical and Thermal Tests 115 7.1.1 Load-Bearing Applications 117 7.1.2 Climate Stresses 117 7.1.3 Natural or Synthetic Fibers 118 7.1.4 Best Performing Fiber 118 7.1.5 Phenolic Foam or EPS 119 7.1.6 Summary 119 7.2 Environmental Impact 120 7.2.1 Heat Loss 120 7.2.2 Moisture Resistance 120 7.2.3 Embodied Energy 121 7.2.4 Emissions 121 7.2.5 Toxicity 121 7.2.6 Summary 121 7.3 Costs 122 7.4 Recommendation 122 Chapter 8: Future Work 124 8.1 Fabrication 124 8.2 Testing 126 8.3 Comparison Materials 127 8.4 Environmental Impact Assessment 127 8.5 Cost Assessment 128 Bibliography 129 Appendix A: Glossary 135 viii List of Tables Table 1.1: Uses of indigenous woods: Ban Tun village, Northern 1 Thailand Table 2.1: Precipitation in various U.S. climate regions 15 Table 2.2: Heat loss distribution for heating a typical residence 19 Table 2.3: Heat loss distribution for cooling a typical residence 20 Table 2.4: Embodied energy of common insulation materials 21 Table 3.1: Density of fiber reinforced phenolic foam samples 38 Table 3.2: Tests conducted for each material 44 Table 4.1: Density results for all materials 57 Table 4.2: Densities of common load-bearing materials 58 Table 4.3: Percentage increase in density after 8-week water exposure for 59 all materials Table 4.4: Compression results: climatic effects on compressive strength 61 Table 4.5: Compression results: climatic effects on Young’s modulus 61 Table 4.6: Young’s modulus values for common building materials 64 Table 4.7: Shear results: climatic effects on maximum shear load 66 Table 4.8: Water absorption rates for phenolic foam and EPS expressed 69 as mass of water absorbed as a percentage of mass of material Table 4.9: Mass of water absorbed expressed as percentage of mass of 69 sample Table 4.10: Water absorption values for some common insulation 70 materials ix Table 4.11: Accelerated aging results: mass lost as a percentage of initial 71 mass Table 4.12: SEM images showing cell length across climatic stress 73 Table 4.13: SEM results: cell length across climatic stress 74 Table 4.14: Flammability test results: volume lost as a percentage of 75 initial volume Table 4.15: Flammability test results: UL94 ratings assigned to materials 76 Table 4.16: R-values recorded by Oak Ridge National Laboratory per 76 ASTM C 518 Table 4.17: Conductivity results: temperature drop across dry materials 77 Table 4.18: Conductivity results: temperature drop across wet materials 78 (one week immersion) Table 4.19: Conductivity test results: R-values 79 Table 4.20: R-values of common materials 80 Table 5.1: U-values entered into HEED to obtain energy loss data on 88 each separate material Table 5.2: Effect of water exposure on energy losses in a northern 89 Thai home Table 5.3: Water absorption rates for phenolic foam and EPS expressed 91 as mass of water absorbed as a percentage of mass of materials Table 5.4: Embodied energy of common insulation materials 93 Table 5.5: Hydrofluorocarbon (HFC) emissions in phenolic foam, in the 96 case where HFCs are used as blowing agents Table 5.6: Toxicity concentrations in phenolic foam 98 Table 6.1: Suppliers of phenolic foam raw materials 104 Table 6.2: Raw materials costs for phenolic foam 106 x Table 6.3: Cost to ship phenolic foam raw materials to production site 108 Table 6.4: Costs of providing phenolic foam and EPS insulation for a 113 typical house in northern rural Thailand Table 7.1: Summary of mechanical and thermal results 116 xi List of Figures Figure 1.1: Weather worn house in northern Thailand with degraded paper 2 and bamboo walls Figure 1.2: Local woods and grasses used as posts, walls, doors, and 9 floors in rural Thai construction Figure 2.1: Prevailing air streams in Thailand 17 Figure 2.2: Rural Thai wall construction: the author holding a panel 18 composed of cellulose-based bags woven between bamboo strips Figure 2.3: Cellulose insulation 22 Figure 2.4: Rock wool insulation 22 Figure 2.5: Fiberglass insulation 23 Figure 2.6: Adidas Village, Portland, Oregon 26 Figure 3.1: Keyence hybrid mixer used to mix phenolic foam chemicals 32 Figure 3.2: The author cutting bamboo fiber into ¼” lengths 34 Figure 3.3: Adding aramid fibers to the resin-surfactant mixture 34 Figure 3.4: Styrofoam mold covered in aluminum foil 35 Figure 3.5: Samples being prepared for water absorption testing 41 Figure 3.6: Samples in USC’s environmental testing chamber 42 Figure 3.7: Instron machine loaded with two stainless steel platens for 45 compression testing Figure 3.8: Shear samples being glued between shear plates 46 Figure 3.9: Shear sample undergoing testing 47 xii Figure 3.10: Cambridge 360 Scanning Electron Microscope 48 Figure 3.11: Sample loaded into fire testing chamber 49 Figure 3.12: Conductivity samples being glued together in 12” wooden 50 frames Figure 3.13: Cellulose fiber conductivity sample 51 Figure 3.14: Conductivity test apparatus 53 Figure 4.1: Assumed typical 40’ x 50’ house plan in northern Thailand 62 Figure 5.1: Simple one-story house in northern Thailand constructed in 87 HEED Figure 6.1: Construction site chosen as a case study to estimate 103 insulation costs xiii Abstract Construction methods and materials in the hot, humid, wet region of northern Thailand are often rudimentary. Shelters are left susceptible to harsh weather conditions, and weather-worn materials revealing gaping holes into buildings are common sights. Therefore, better insulation and more durable wall materials are needed in such underdeveloped areas prone to extreme weather. Phenolic foam warrants consideration as a building material because of several characteristic features. It is lightweight and has low thermal conductivity, allowing for applications as a building insulation material. Moreover, the material has low flammability, low smoke toxicity, and is cost-competitive with conventional foams, such as polyurethane and expanded polystyrene (EPS). This paper evaluated the potential of using phenolic foam as a building material in northern Thailand. Phenolic foam’s mechanical and insulation properties, sustainable benefits, and costs were evaluated when used as an insulation material in the hot, arid, wet region of northern Thailand. Four types of fiber reinforced phenolic foams were fabricated: bamboo, aramid, glass, and cellulose. Choosing these materials provided a comparison study of natural and synthetic materials. Water absorption and accelerated aging tests were conducted on fabricated composite foam samples to simulate the flooding and extreme heat and humidity xiv conditions of northern Thailand. Compression, shear, and conductivity tests were performed after these climate simulations. Measurements of retained mechanical properties were performed to determine if the material would be a suitable insulation and/or load-bearing material. Fire resistance testing was also performed on the samples. In addition to testing, an environmental impact assessment was performed on the composite foams. Finally, costs to insulate a home in northern Thailand with the materials were evaluated. Results were compared with expanded polystyrene, a common insulation material. The results showed that neither the fabricated fiber reinforced phenolic foam nor EPS is comparable in strength to conventional load-bearing materials and do not retain their mechanical properties after extreme climate exposure. Little distinction could be drawn between natural and synthetic fibers. Results also showed phenolic foam’s stronger fire resistance and insulative properties under dry conditions than EPS. Comparison of environmental impacts showed that, due to fiber reinforced phenolic foams’ relatively low embodied energy, the material warrants consideration as a sustainable alternative to conventional building insulation materials such as EPS. Finally, cost considerations showed that neither phenolic foam nor EPS can feasibly be used as building insulation in low-income areas of northern Thailand. Keywords: Design of fiber reinforced phenolic foam, strengthening of composite materials, environmental impact Chapter 1: Introduction Building methods and materials in the developing regions of northern Thailand and Laos are often the most basic. Construction methods are rudimentary since, as subsistence farmers, few people can afford any sort of sturdy walled protection or insulation from the elements. Residents are industrious, however, as use of indigenous timber is common in shelter construction here. Shelter walls, when they do exist, are usually composed of locally available wood materials such as bamboo and eucalyptus. For example, in Ban Tun, a highland village about 150km southwest of Chiang Mai, residents make use of 49 of 78 indigenous tree species found in the village (Schmidt-Vogt 2000). Construction and firewood are some of the most common uses highlighted in Table 1.1. Ecolocial Significance and Use of Forest Trees: Ban Tun Village, Northern Thailand 0 5 10 15 20 25 30 Construction Fences Firewood Tools Food Animal Food Medicinal Ceremonial Decoration Uses Number of Tree Species Table 1.1: Uses of indigenous woods: Ban Tun village, Northern Thailand http://www.etfrn.org/etfrn/workshop/users/chapter_13.pdf 1 Construction methods here must also consider climate, as the northern regions of Thailand and Laos receive tremendous amounts of rain in the summer. Monthly averages often exceed ten inches per month (Global Historical Climatology Network 1992). Humidity swells as well with temperatures climbing to over 35º C (Simply- Thai.com 2007). Unfortunately, shelters are often left susceptible to these harsh weather conditions. Despite assiduous use of local materials such as in Ban Tun, tattered, weather-worn materials revealing gaping holes are common sights throughout the region. Living quarters are often left exposed to harsh weather conditions, and families are left to fend against severe rainstorms (see Figure 1.1). Figure 1.1: Weather worn house in northern Thailand with degraded paper and bamboo walls Photograph produced by the author, 2008 2 3 1.1 The Problem These observations suggest better insulation and more durable wall materials are needed in underdeveloped areas prone to extreme weather. This thesis set out to produce a low-cost, environmentally-sensitive, hybrid load-bearing insulation material made of natural materials and capable of withstanding extreme heat, humidity, and rain. 1.2 Potential Solution Members of Dr. Steven Nutt’s composite materials department within the chemical engineering school at the University of Southern California have been exploring the use of phenolic foam as both an insulation and structural material. Building materials are one application with which they have been working. Yet three significant gaps in scientific data exist regarding the material. The primary question this thesis explored is how phenolic foam compares with expanded polystyrene (EPS), a common insulation material. Which material is stronger and more insulative, more environmentally sensitive, and cheaper? Chapter 4 (results), chapter 5 (environmental impact), and chapter 6 (costs), respectively, answer these three questions. Second, few studies have examined phenolic foam’s response to extreme climate conditions. How does this material perform under heavy rains or extreme heat and humidity? Chapter 4 returns an answer to this question. Third, although structural benefits of fiber reinforcement in phenolic foam have been documented, comparison studies of natural versus synthetic reinforcement fibers have not been 4 conducted. Do natural fibers offer a stronger, more insulative, cheaper, and more sustainable solution to synthetic fibers? Chapters 4, 5, and 6 address these questions. Therefore, this thesis’s primary endeavor is to compare natural and synthetic fiber reinforced phenolic foams’ response to various climate conditions, then compare the results to EPS’s performance. A secondary objective is to examine the viability of phenolic foam load-bearing applications in buildings. The thesis is valuable because it provides a recommendation on which material is better suited as an insulation material in extreme weather conditions such as in northern Thailand: phenolic foam reinforced with natural or synthetic fibers or EPS. Note: Appendix A: Glossary defines all acronyms, tests, and important concepts throughout the thesis. For example, EPS will be used throughout this thesis to refer to expanded polystyrene. 1.3 Objective The objective of this thesis was to determine whether phenolic foam is suitable as a hybrid structural insulation material in hot, humid, and rainy underdeveloped regions with better mechanical, thermal, and sustainable properties and cost considerations than currently used structural and insulation materials. As part of this determination, the benefits of using various fibers as reinforcing agents were investigated. 5 1.4 Final Product The final product of this thesis is two recommendations. A statement is produced in Chapter 7 on whether: • phenolic foam is well suited as a load bearing material in hot, humid, rainy underdeveloped regions such as in northern Thailand. • phenolic foam or EPS is better suited as an insulation material in hot, humid, rainy underdeveloped regions such as in northern Thailand 1.5 Hypothesis Phenolic foam’s mechanical and thermal properties, environmental impact, and costs demonstrate the material’s superiority as a hybrid structural insulation material. This will be tested by evaluating the mechanical and thermal properties and assessing the environmental impact and costs of fabricated unreinforced and fiber reinforced phenolic foam against common structural and insulation materials. 1.6 Scope The thesis involved six steps: Step 1: Obtain materials. Fabricate phenolic foam, both un-reinforced and with various natural and synthetic fibers: bamboo, cellulose, glass, and aramid. Purchase EPS. 6 Step 2: Condition samples in either a water or aging chamber. Control samples for fiber reinforced phenolic foam, unreinforced phenolic foam, and EPS will not be conditioned in a water or aging chamber. Step 3: Perform mechanical and thermal tests on both conditioned and unconditioned (control) samples. Step 4: Conduct an environmental impact assessment on phenolic foam. Step 5: Analyze the cost to insulate a home in northern Thailand with phenolic foam insulation. Step 6: Compare results of steps 3 through 5 with common structural materials as well as EPS. Several choices were made in order to narrow the scope of the project. These included choosing: • a base material • reinforcement fibers • mechanical and thermal tests • climate simulations • a common insulation material with which to test alongside and draw comparisons with the base material • environmental impact assessment categories • cost categories related to insulating a home in northern Thailand. Each of these choices is described in the following sections. 7 1.6.1 Base Material Phenolic foam was chosen as the core insulation material with which to experiment. Its excellent insulation and fire resistant properties were extensively taken advantage of during World War II as a material in German aircraft. The material was also widely used in the 1970s as a roof insulation material. In the late 1980s and early 1990s, however, residues of an acid catalyst in the foam, when activated by moisture in the air, caused corrosion in metal decking in contact with the foam (Greenwald 2008). As a result, phenolic foam’s use as an insulation material faded. Despite such problems, Shen has noted in his dissertation Toughening of Phenolic Foam that effective solutions have been found to reduce phenolic foam’s corrosion problems (2003, p.25). In addition, its structural applications are being explored. The main reason that phenolic foam was chosen, then, was to mechanically and thermally experiment with a material that could rival current structural and insulation materials. Fire resistance could also be measured and a sustainability profile could be created, the results of which could be compared with common building materials to reinforce the material’s superiority in building applications. Phenolic foam was also chosen to compare the benefits of natural and synthetic fibers. These fibers might not only strengthen phenolic foam for load-bearing applications but also provide additional mechanical and/or thermal benefits; consequently, a broad range of tests were conducted on the fibers. Fibers were also 8 used to compare the benefits of natural, locally available versus synthetic materials, in an effort to keep the resources of developing regions in mind. Finally, phenolic foam was chosen because little data existed on how the material responded to extreme climate conditions. Could the material withstand heavy rain and extreme heat and humidity? Coupled with environmental impact and cost data, the project would ultimately determine whether phenolic foam yielded a viable structural insulation choice for poor communities in hot, humid, rainy climates. 1.6.2 Reinforcement Fibers: Natural Versus Synthetic Various fibers were chosen with which to experiment in order to evaluate natural versus synthetic fibers. The reason fibers were chosen in the first place was to confirm whether fibers found in developing regions could strengthen phenolic foam enough for use as a load-bearing building material. Secondary reasons examined whether fibers offered any thermal benefits or improved unreinforced phenolic foam’s response to climate stresses. Natural and synthetic fibers of similar lengths were compared in order to determine the potential benefits of using indigenous, locally available materials. For example, bamboo forests and plantations cover the landscape throughout Southeast Asia, India, and China (Lehmer 1997). In addition, several types of cellulose-based wood materials are used as posts, walls, and doors in Thai buildings (see Figure 1.2). By using materials available in residents’ own backyards, this thesis would hopefully yield a cheaper, more natural, and more sustainable product. Figure 1.2: Local woods and grasses used as posts, walls, doors, and floors in rural Thai construction Photograph produced by the author, 2008 1.6.2.1 Synthetic Fibers The synthetic reinforcing fibers chosen were glass and aramid. Glass fibers, commonly used in building insulation materials, were ¼” long cut fibers supplied by Owens-Corning Inc. Aramid fibers, a plastic often used as a strengthening agent in fiber reinforced polymers, were ¼” cut Nomex fibers made by DuPont. 1.6.2.2 Natural Fibers 9 Bamboo and cellulose were chosen as the natural reinforcing fibers. These represented materials widely available in northern Thailand used in many aspects of 10 life. Woven bamboo fibers were obtained from Habu Textiles in New York and were cut into ¼” pieces. The cellulose fibers were obtained from Advanced Fiber Technology Company in Ohio. They were recycled from industrial grade newsprint, clumped together in small bunches about ¼” long, and not cut. 1.6.3 Tests Five tests were conducted in order to evaluate phenolic foam’s mechanical and thermal performance. A compression test was chosen to test the material’s compressive strength and Young’s modulus, and a shear test was chosen to test maximum shear strength. Scanning electron microscopy (SEM) was used to measure samples’ cell diameters before and after being placed in water or aging chambers, in order to measure the effect of environmental stresses on cell length. The purpose of obtaining this data was to see if longer cell diameters correlated with greater water absorption rates and shorter cell diameters correlated with slower absorption rates. SEM data would therefore hopefully serve as a predictor of phenolic foam’s water resistance ability. A conductivity test was also designed in order to measure phenolic foam’s R-value, or how easily the material resists heat flow. Finally, a flammability test was chosen in order to measure how well the material resists fire. 1.6.4 Thai Climate Simulations Two climate simulations were run in order to model the effects of harsh Thai weather conditions on phenolic foam’s mechanical and thermal properties. Water immersion, an eight week test, was chosen for five reasons. First, it modeled wet climate conditions in northern Thailand during the summer wet season. Second, this 11 simulation measured how much water phenolic foam would absorb over an extended period of time, thereby indicating how well the foam could resist water. Third, compression and shear values were compared before and after the water immersion simulation in order to see if extended water exposure would weaken the material. Fourth, cell lengths measured through SEM were compared before and after the water simulation to see if cell size correlated with water resistance. A positive correlation would suggest cell length as a good predictor of water resistance rates in phenolic foam. Finally, R-values were measured before and after water immersion, in order to see how flooding conditions may impact the thermal performance of the material. Accelerated aging was the second climate simulation, a six week test chosen for three reasons. As with the water test, compression and shear values were compared before and after the simulation in order to see if extended exposure to intense heat and humidity would weaken the material. Second, cell lengths obtained through SEM were compared before and after the simulation in order to see if aging altered cell size. This SEM data could then be a useful predictor of phenolic foam’s water resistance, assuming a positive correlation between cell size and water resistance derived from the first climate simulation. Third, the test was chosen to indicate phenolic foam’s durability by measuring the percentage of mass lost throughout the test. 12 1.6.5 Comparison Materials Data for various structural materials was referenced throughout the thesis in order to evaluate phenolic foam’s potential as a load-bearing material. Although testing all of these materials was beyond the scope of this study, the values referenced are well established and offered a reliable comparison method. Expanded polystyrene (EPS) was chosen as a commonly used insulation material in the building industry with which to compare phenolic foam. The material was bought at Home Depot and subjected to the same five tests and two climate simulations as phenolic foam. Its environmental impact and cost to insulate a home in northern Thailand were also compared against phenolic foam. The intention in choosing this material was to directly compare phenolic foam’s effectiveness against a competing building insulation material. 1.6.6 Environmental Impact The environmental impact of phenolic foam as an insulation material was compared with EPS. Since data relating to phenolic foam’s environmental impact is scarce, only categories in which data was readily available and which provided an easy comparison with EPS were chosen. These categories included heat loss, moisture resistance, embodied energy, emissions, and toxicity. 13 1.6.7 Costs The cost to insulate a home in northern Thailand using phenolic foam or EPS is not widely published or readily available. Consequently, several estimates had to be made in several cost categories, including raw materials, transportation, machinery, labor, and electricity. 1.7 Applications Phenolic foam has historically been used as a roofing insulation material. Recent research also supports fiber reinforced phenolic foam’s use as a load-bearing material (Shen & Nutt 2003). The objective of this project was to prove phenolic foam’s viability as a hybrid structural insulation material. Various applications relevant to this objective, such as load-bearing walls, rain screens, and cladding systems, are discussed as background material in Chapter 2. 14 Chapter 2: Background The initial goal of this project was to create a low-cost hybrid structural insulation material with superior mechanical, thermal, and sustainable properties for use in hot, humid, and wet underdeveloped regions like northern Thailand. The chapter begins by examining the necessity for such a material. After defining the problem, the chapter continues with an overview of current insulation materials as well as cladding and structural applications. The chapter concludes with background material on phenolic foam, its potential as a low-cost, environmentally sensitive structural insulation material for use in hot, arid, wet climates, and a brief discussion of previous studies involving phenolic foam. 2.1 The Problem The problem addressed is twofold. First, many residents in northern Thailand and northern Laos are too poor to afford insulation or quality home construction. This fact, coupled with a hot, rainy climate in the summer months, means millions must fend against the elements. The problem extends to other developing regions in Asia, Africa, and South America. Therefore, a solution is needed to better protect residents. The second problem involves improving on current insulation materials’ mechanical properties, environmental impact, and cost. Whereas current materials have high R-values, are environmentally sensitive, or are cheap, no single material 15 outperforms all others in these categories. Consequently, this thesis sought to produce a material that performed well in all three categories. 2.1.1 Living Conditions in Northern Thailand Inhabitants of northern Thailand experience a harsh rainy season. Rural areas rely on farming and generate little disposable income; therefore, residents often do not have the resources to use quality materials in homes or make basic home improvements. An affordable, environmentally sensitive insulation material is therefore needed to protect people in this region from extreme weather conditions. 2.1.1.1 Climate Thailand is located in the northern hemisphere and lies along the Southeast Asian tropical rain belt during the region’s wet season from April to September. This rain belt is home to the world’s second largest rain forest, and the climate is hot and humid for the entire year. Seasonal shifting winds, or monsoons, bring heavy rainfall in the wet months. The region’s climate statistics are especially formidable. The average yearly temperature is 80ºF, the average humidity is between 70 and 90%, and the average rainfall is between 60 to 100 inches (F 2002). Comparison with three regions in the United States shows just how extreme this rainfall is: 16 U.S. Region Climate Category Annual Precipitation (in.) Eastern U.S. Moist continental 32 (abundant precipitation) Central and southern Mediterranean (wet 17 California winter-dry summer) Rocky Mountain Range Highland 9 Table 2.1: Average precipitation in various U.S. climate regions http://www.blueplanetbiomes.org/calif_chap_climate_page.htm The region is surrounded on several sides by ocean, and shifting air streams throughout the year - particularly in Thailand, which borders ocean water on two sides - dominate climatic conditions and contribute to the high rainfall (see Figure 2.1). Figure 2.1: Prevailing air streams in Thailand http://www.cig.ensmp.fr/~iahs/hsj/250/hysj_25_02_0167.pdf 2.1.1.2 Thai Climate, Economy, and Building Practices Such extreme weather conditions define relationships between rainfall, the economy, and building practices in northern Thailand. Residents derive their income primarily from farming. Yet the non-uniform distribution of rainfall throughout the year creates severe shortages of irrigation water during many months, and severe floods wash over the region at other times (Phien et al. 1980). Consequently, water resource management directly impacts living conditions, and the efficient use of available water for farming is critical to economic and physical survival. 17 Because of such extreme weather conditions, income in rural areas goes toward such expenses as irrigation and land maintenance. Yet Thailand’s GNP per capita is only $2712 - a figure presumably less in the north, where livelihood is dominated by subsistence farming (Earth Trends 2003). Therefore, in such poorer regions, living conditions and buildings are often the most basic and utilize indigenous materials such as bamboo and other woods. Because many residents cannot afford insulation or wall enclosures, entire building faces are sometimes left open to the elements. In other cases, walls - often made of cheap, cellulose-based materials - are flimsy or severely degraded (see Figure 2.2). Given the large amount of rain in the region, walls made of such materials typically do not stay intact much longer than ten years (T Kongchumchuen 2007, pers. comm., 21 Aug.). Figure 2.2: Rural Thai wall construction: the author holding a panel composed of cellulose- based bags woven between bamboo strips Photograph produced by author, 2008 18 19 Materials used in walls also often have poor strength, insulative and water and fire resistant properties. These conditions, coupled with the fact that residents’ homes face exposure to extreme elements, suggest a low-cost wall or insulation material is essential in improving the long-term living conditions of people in this area. 2.1.2 Insulation and the Environment Creating a low-cost structural insulation material is also important for environmental reasons. As energy costs rise, energy conservation in buildings is becoming increasingly important. Heat is principally lost through ceilings, walls, windows, doors, floors, and foundations as well as through air infiltration. A large reduction in heat loss, however, can be realized by insulating attics and ceilings (Klempner 2004, p.220). Tables 2.2 and 2.3 compare heat losses for an un-insulated and insulated house during heating and air-conditioning periods. Un-insulated Insulated Peak hour 1 Day Peak hour 1 Day KW*h % kW*h % kW*h % KW*h % Ceiling 5.4 42 99.3 43 0.8 12 15.8 14 Wall 2.1 16 51.9 22 0.8 12 20.1 17 Glass conduction 2.7 22 38.1 16 2.7 40 38.1 33 and convection Infiltration 2.5 20 42.4 19 2.5 36 42.4 36 Total load 12.6 231.6 6.8 116.4 Table 2.2: Heat loss distribution for heating a typical residence (Klempner 2004, p. 220) 1 Peak hour is the hour of day that corresponds to the highest thermal load. 20 Un-insulated Insulated Peak hour 1 Day Peak hour 1 Day KW*h % kW*h % kW*h % KW*h % Ceiling 1.6 18 24.6 22 0.5 7 6.7 8 Wall 0.8 10 12.9 12 0.3 5 4.4 6 Glass conduction 0.6 7 8.6 8 0.6 9 8.6 10 and convection Infiltration 1.5 16 29.7 27 1.5 19 29.7 35 Total load 8.9 110.3 6.9 83.9 Table 2.3: Heat loss distribution for cooling a typical residence (Klempner 2004, p. 221) 1 Peak hour is the hour of day that corresponds to the highest thermal load. The tables show that insulation can save on energy costs by up to 50%, with roofs the most effective location to place insulation (Khemani 1997, p.224). However, several factors need to be considered when choosing an insulation material, given the important part insulation plays in a building’s energy conservation. Are insulation materials manufactured with CFCs? How lightweight is the material? What is the R- value of the material, and does it degrade over time? These are just three questions to consider when evaluating a material’s environmental impact. Chapter 5 gives further background information on chemicals used in the foam industry and their environmental impact. 2.1.3 Insulation Materials: Overview Decisions about insulation are among the most important in the environmental impact of buildings. Although insulation reduces building energy consumption and provides other ongoing environmental benefits throughout a building’s lifetime, insulation materials greatly differ in their environmental advantages. For example, Table 2.4 reveals a tremendous difference in embodied energy of the most commonly used insulation materials. Table 2.4: Embodied energy of common insulation materials http://www.afcee.brooks.af.mil/green/case/accsfguide.pdf 1 http://www.canadianarchitect.com/asf/perspectives_sustainibility/measures_of_sustainablity/measur es_of_sustainablity_embodied.htm The table shows that EPS has an embodied energy value over 35 times greater than cellulose. The data suggests that use of natural, locally available or recycled materials can drastically reduce embodied energy costs. This section gives an overview of the most common insulation materials and recent efforts to reduce embodied energy through the use of recycled materials. 2.1.3.1 Cellulose Cellulose insulation is usually comprised of 80% post-consumer recycled newspaper by weight, while the rest is made of fire retardant chemicals (see Figure 2.3). Efforts have been made in the last decade to produce a lower-density cellulose material by breaking down newspaper into individual fibers that are fluffier, resulting in a 21 “greener” product. The material is cleaner with a higher R-value (Building Green 1995). Figure 2.3: Cellulose insulation http://www.daviddarling.info/encyclopedia/C/AE_cellulose_insulation.html 2.1.3.2 Mineral Wool Mineral wool was at one time the most common insulation material until fiberglass gained favor in the 1960s and 1970s. The material consists of up to 75% post- industrial recycled content and is classified either as slag wool or rock wool (see Figure 2.4). Slag wool comprises about 80% of the mineral wool industry and is made from iron ore blast furnace slag, an industrial waste product. Rock wool is made from natural molten rocks such as basalt and diabase (Building Green 1995). Figure 2.4: Rock wool insulation http://science.howstuffworks.com/plasma-converter2.htm 22 2.1.3.3 Fiberglass Fiberglass is the most common insulation material and is made from molten glass spun into microfibers (see Figure 2.5). The primary manufacturers of fiberglass have been using 20% recycled glass materials in their products for decades, with the largest manufacturer, Owens Corning, using 30% recycled glass (Building Green 1995). The relatively high percentage of manufactured materials used in fiberglass when compared with mineral wool is reflected in the material’s relatively high embodied energy cost shown in Table 2.4. Figure 2.5: Fiberglass insulation http://img.alibaba.com/photo/12283858/FiberGlass_wool_Insulation.jpg 2.1.3.4 Polystyrene Polystyrene is another common insulation material made from synthetic materials, and recycled plastic resin has been used by some of the largest manufacturers such as Dow and Amoco Foam Products. Expanded polystyrene (EPS) can be made out of recycled polystyrene by crumbling then re-molding the old EPS (Building Green 23 24 1995). Expandable polystyrene beads are typically impregnated with pentane, a hydrocarbon that contributes to smog but not global warming or ozone depletion. EPS is also the only rigid foam insulation made without CFCs or HCFCs. EPS was chosen as the material with which to compare phenolic foam for several reasons. Like phenolic foam, it is lightweight, cheap, comes in rigid board form, and uses pentane as an expanding agent. It offered a good “challenge” in the insulation market, as it is well established as an insulation material with known R-values, cheap and easily accessible to the public, and widely used. 2.1.3.5 Polyisocyanurate Manufacturers of polyisocyanurate foam insulation also use recycled materials, but instead of using recycled foam, the chemical components themselves contain recycled content. 2.1.3.6 Radiant Barriers Aluminum used in radiant barriers is mostly recycled and sometimes uses recycled plastic in its foam core. 2.1.4 Insulation Materials: Other Environmental Factors R-value, durability, water resistance, fire resistance, weight, and cost are other environmental factors besides embodied energy to consider when choosing an insulation material. R-value is a measure of how well a material retards the flow of energy through the material itself. The higher the R-value, the better the insulation, and the SI units are in K*m²/W. R-value degradation, particularly after exposure to 25 wet weather, is a factor when evaluating an insulation material’s environmental performance. Durability, or how well a material resists compression, moisture, and physical degradation, is also important when considering insulation materials. A strong insulation material might find application as a structural member within a building, and a material that resists environmental deterioration and moisture may be well suited to wet regions. Likewise, fire resistance may be an important factor when considering building in dry areas prone to fires. Finally, lightweight insulation materials are generally cheaper than heavier materials and may find application as exterior cladding boards or structural members. 2.2 Cladding Cladding systems are an important weatherproofing feature of buildings. They often help keep water out of buildings and reduce mold growth and leaking. They may also reduce energy losses of a building because of good insulative properties. Other important characteristics of cladding materials are weight and fire resistance. Some of the most common cladding materials are ceramic stone, glass fiber concrete, terracotta, brick, and various woods, stones, and metals. They are typically watertight and durable. Alucobond is a relatively new and popular sandwich system made of two sheets of aluminum metal adhered to a thermoplastic core material. The material has exceptional resistance to water and extreme temperatures. Cladding systems consist of core insulative materials and outer skin layers and may be used in exterior applications. For example, the Adidas Village in Portland, Oregon uses multi-colored alucobond panels on its buildings (see Figure 2.6). The panels form a curtain wall and act as a rain screen. The attachment system has an internal gutter that routes condensation out at the joints, and the panel system also accommodates seismic movement (Alcan 2004). Figure 2.6: Adidas Village, Portland, Oregon http://www.aia.org/aiarchitect/thisweek04/tw0116/0116aia_portland.htm 2.3 Structural Insulation Structural insulated panels (SIPs) consist of foam insulation sandwiched between two layers of a structural board. The foam is usually expanded polystyrene, extruded polystrene, or polyurethane, and the composite material has structural properties similar to an I-beam or I-column. SIPs are extremely versatile and can be used in floors, roofs, walls, or foundations and replace such building components as studs 26 27 and joists, insulation, and vapor barriers (Structural Insulated Panel Association 2007). Efforts have been made to use unique materials in SIPs. For example, the addition of glass and aramid fibers to phenolic foam has been shown to increase its strength (Shen & Nutt 2003, p. 906). The resulting composite material is tough, fire-retardant, strong, insulative, and has potential for use in various structural applications. 2.4 Phenolic Foam Phenolic foam is a synthetic cellular solid made up of closed cells, or an interconnected network of plates forming the edges and faces of cells (Klempner 2004, p. 37). This hollow infrastructure has several implications. First, it allows for lightweight design options as an insulation material. Second, values of various mechanical and thermal properties - including Young’s modulus, conductivity, and strength - can be controlled by altering the density of the cellular solids. The material is a polymer foam that typically consists of phenolic resin (the base matrix), surfactants, an acid catalyst, and pentane, a blowing agent used to expand the matrix. Fibers may also be added as reinforcing agents to increase the strength to weight ratio. 28 Phenolic foam has several important properties. First, as with most cellular materials, it has very low thermal conductivity and is a good choice as an insulation material. The material also has low mass, low cost, and is exceptionally fire resistant and thermally stable. Finally, it has low smoke density, low smoke toxicity, and non- dripping during combustion. Disadvantages to the material are that it is fairly water absorbent, corrosive to metal, and brittle (Shen 2003, p. 22). Severe brittleness has prevented the material from being used extensively in load- bearing applications. Sandwich panels with a phenolic foam core have achieved only modest success, due to difficulty in bonding the foam to other materials. However, previous work done at the University of Southern California (described below) has shown that fibers embedded in the material make it stronger, more ductile and competitive with structural foams such as polyurethane or PVC (Shen & Nutt 2003, p. 904). 2.5 Previous Work Previous studies with phenolic foam have focused primarily on using fibers in the foam as reinforcing agents. Fibers have the potential to increase foam strength and reduce brittleness. In particular, short chopped glass and aramid fibers have been used to increase phenolic foam toughness (Desai et. al 2008). In another study, various approaches to toughening phenolic foam and a comprehensive evaluation of its mechanical performance were undertaken (Shen 2003). Short fiber reinforcements 29 demonstrated significant improvement in mechanical performance, and aramid fibers were shown to best enhance toughness. Additional topics explored included: peel resistance of reinforced phenolic foam, anisotropy and the effect of fiber orientation on strength, and the mechanics of phenolic foam sandwich structures. Most recent work at the University of Southern California that has built off the previous studies involves mechanical behavior of hybrid composite foam (Desai et. al 2008). Phenolic foam was reinforced with chopped glass and aramid in varied proportions, in order to increase its toughness, and various mechanical properties were assessed. The above studies all confirm that fibers can effectively be used as reinforcing agents to strengthen phenolic foam. Current work at USC is considering specific load-bearing applications of fiber reinforced phenolic foam. This thesis built off the previously described studies and attempted to produce a low density load-bearing material. The thesis also examined three topics not considered in previous studies, the methods of analyses of which are outlined in Chapter 3: • comparison of phenolic foam versus EPS: mechanical, thermal, and environmental impact properties and costs • effects of various climate simulations on mechanical properties • strength and conductivity differences between natural and synthetic fibers. 30 Chapter 3: Methodology This thesis’s goal was to determine whether phenolic foam is suitable as a hybrid structural insulation material in hot, humid, and rainy underdeveloped regions like northern Thailand. Methodology in pursuing this goal involved eight steps: • Obtain materials: fabricate phenolic foam, buy EPS, and cut samples. • Record pre-conditioning measurements (length and weight). Conditioning is defined as exposing materials to environmental conditions that simulate Thailand’s hot, arid, rainy summer climate (see Appendix A: Glossary). • Condition samples with one of two climate simulations (water immersion or accelerated aging). • Record post-conditioning measurements (length and weight). • Test samples, both unconditioned and post-conditioned, in compression, shear, cell length, flammability, or conductivity. • Assess environmental impact of materials. • Assess cost of insulating a typical home in northern Thailand with the materials. • Compare results with common structural and insulation materials. Climate tests were included in the methodology since they may alter materials’ mechanical properties (American Society for Testing and Materials 1998, p. 35; American Society for Testing and Materials 2004b, p. 532). In most tests, three 31 samples were tested in order to obtain a meaningful average. Environmental impact and cost were assessed to provide a sustainability profile of phenolic foam. Several materials and equipment items were purchased in order to execute this methodology. The project also required the use of materials from various suppliers as well as several laboratory facilities at the University of Southern California. 3.1 Fabrication The first step in the project was fabricating phenolic foam. The proprietary formula belongs to M.C. Gill Corporation, a company in El Monte, California, which develops high performance composite products and has sponsored previous projects involving phenolic foam at USC. 3.1.1 Materials The formula to make phenolic foam consists primarily of phenolic resin (HRJ- 14489, supplied by Schenectedy International Inc.). The resin was stored in a refrigerator at 4ºC to prolong its lifetime and prevent curing. The resin provided the base matrix. Two surfactants (Pel-stab, supplied by Pelron Corp., France, and Dabco, DC193, supplied by Air Products and Chemicals Inc., Pennsylvania) served to stabilize the material and refine the cell sizes. Phenolsulfonic acid (CRC-608, supplied by Capital Resin Corp., Ohio) acted as a catalyst and created heat during foam formation. Fibers added reinforcement and were supplied by Owens Corning (glass), DuPont (aramid), Advanced Fiber Technology (cellulose), and Habu Textiles (bamboo). N-pentane, a blowing agent supplied by Aldrich Chemical Co., provided gas to expand the foam. The amount of pentane largely determined foam density. 3.1.2 Mixer A mixer (Keyence hybrid mixer HM-560) involving two simultaneous rotation types combined the phenolic foam chemicals (see Figure 3.1). Figure 3.1: Keyence hybrid mixer used to mix phenolic foam chemicals Photograph produced by the author, 2008 The centrifugal force drove the resin flow in the container, which provided a shear force necessary for mixing and dispersing fibers. The advantage of using this mixing 32 33 method was that it eliminated air voids in the mixture. Because temperatures of the dispersed fiber mixtures increased after mixing, the containers were cooled for ten minutes before fibers were added. 3.1.3 Fabrication Process Five types of foams were fabricated: un-reinforced phenolic foam, glass fiber reinforced phenolic foam, bamboo fiber reinforced phenolic foam, aramid fiber reinforced phenolic foam, and cellulose fiber reinforced phenolic foam. Proportions of chemicals were kept constant in order to maintain uniform composition across foam type: acid was 5% of the resin mass (5 wt%), surfactants were 1.5 wt% (Pel- stab) and .5 wt% (Dabco), pentane was 4 wt%, and glass, aramid, and cellulose fiber were 4 wt%. Bamboo fibers were only 2 wt%, since greater proportions of bamboo created high heat and viscosity when mixing the fibers into the resin. Section 3.1.4 describes problems with the mixing process. The fabrication process began by weighing resin and surfactants in a polyethylene container and mixing for two minutes in the mixer. Fibers were cut into ¼” lengths (see Figure 3.2). Figure 3.2: The author cutting bamboo fiber into ¼” lengths Photograph produced by author, 2008 The fibers were then added to the mixture, and the materials were mixed again for two minutes and then cooled (see Figure 3.3). Figure 3.3: Adding aramid fibers to the resin-surfactant mixture Photograph produced by the author, 2008 34 Next, pentane and acid were added to the cold container and mixed together for two minutes. Hand mixing, followed by one minute of additional mixing in the mixer, was sometimes necessary to thoroughly combine the chemicals. The final mixture was transferred into a Styrofoam mold (see Figure 3.4). Figure 3.4: Styrofoam mold covered in aluminum foil Photograph produced by the author, 2008 Five containers were typically combined in the mold to create a final mixture, which was heated in the open mold in an oven at 80ºC for one hour. A similar process was followed for un-reinforced foam, although the containers did not need to be cooled. 3.1.4 Problems Several problems arose while fabricating phenolic foam, including thermal control, problems with the batch process, fiber dispersion, and the creation of uniformly dense samples. 3.1.4.1 Thermal Control Thermal control proved to be the biggest challenge while fabricating fiber reinforced phenolic foam. The addition of fibers created great friction and a highly viscous 35 36 mixture generating much heat. As a result, the top of the polyethylene container sometimes burst while in the mixture, creating a huge mess. The best solution was to cool the containers in a refrigerator at 4ºC for ten minutes after adding the fibers but before adding the acid and pentane. 3.1.4.2 Batch Process A second fabrication problem was working with a batch process. Since the containers were small (holding about 200 grams), several containers – typically four or five – had to be combined in the mold. The resulting mixture was non-uniform, due to slight variations in chemical amounts or mixing times. The solution was to keep amounts and mixing times precisely identical before combining the individual batches. Amounts were measured to the nearest hundredth of a gram. 3.1.4.3 Fiber Dispersion A third problem was fiber dispersion. Bamboo fibers generated much friction, heat and eruptions in the mixer, despite cooling. Hand mixing proved the easiest method of dispersing bamboo fibers, although this did not uniformly distribute the fibers. The other fiber types were only slightly less difficult to uniformly mix in the mixer. Non-uniform fiber dispersion, caused in part by hand versus machined mixing, could explain why certain fiber reinforced phenolic foam samples did not perform well in the strength tests. 3.1.4.4 Non-Uniform Density A fourth problem was creating uniformly dense samples. Foams with different reinforcing fibers should have reasonably close densities in order to facilitate 37 comparisons between their mechanical properties (Shen et al. 2003, p.942). For example, Paraskevopoulos (1962, 4.4) has noted that mechanical properties of foams, including compression and shear strengths, are related to density as follows: S = A(D) C where S = the mechanical property in question D = density A and C = constants varying with the type of property under consideration, but independent of D. Thus, materials should have similar densities in order to accurately compare their mechanical properties. Unfortunately, maintaining uniform densities across fiber type was a difficult task during fabrication. Density does not stay constant within the volume of a foam block but may be much greater at the surface layers, due mainly to processing conditions (Klempner 2004, p.27). Table 3.1 shows that although the densities of glass, aramid, cellulose, and un-reinforced phenolic foam samples were within 6% of 4.22 pcf, bamboo foam density was 41% less than this value at 2.51 pcf. This was likely because of the aforementioned problems with mixing and fiber dispersion. Table 3.1: Density of fiber reinforced phenolic foam samples The implications of Table 3.1 were severe for my thesis goals. Because the density of bamboo samples was significantly less than the density of other foam samples, mechanical properties of bamboo fiber reinforced phenolic foam could not be fairly compared with the other samples. Therefore, bamboo fiber reinforced phenolic foam did not undergo strength testing and was not considered for structural applications. However, it was assumed that water absorption, aging, flammability, cell length, and conductivity would be less related to density (although still related) than strength 38 39 properties, and these tests were conducted for bamboo fiber reinforced phenolic foam. Remediation of this density problem is discussed in Chapter 8: Future Work. 3.1.4.5 Low Density Values Table 3.1 reveals a final problem: samples’ density values were ultimately too low for the materials to be considered for structural applications. A study looking at structural applications of phenolic foam formulated sample densities between 12 and 15 pcf (Desai et al. 2008, p. 20). The values in Table 3.1 fall well below this range. Because strength is proportional to density, it was concluded that phenolic foam as fabricated in this thesis would not be suitable for structural applications. As a result, the project goal narrowed in scope: only insulation applications would be considered for phenolic foam. However, structural tests were still carried out to confirm that phenolic foam would not be strong enough to serve as a load-bearing material. 3.2 Cutting Samples After fabrication, phenolic foam blocks were allowed to cool for 20 minutes. The blocks were then cut into samples of various sizes, depending on the American Society for Testing and Materials (ASTM) testing standard. A metal handsaw was used to cut the foam, and a band saw and sandpaper were used to create precise dimensions. 40 3.3 Density Measurements Length, width, height, and weight measurements were recorded after samples were cut. These measurements were compared to measurements taken after climate conditioning. As mentioned, due to problems in creating uniformly dense samples across fiber type, strength tests for bamboo fiber reinforced phenolic foam were not conducted. 3.4 Climate Conditioning After measurement, samples underwent climate conditioning by being placed either in a water absorption or accelerated aging chamber. Control samples were not conditioned. The water absorption test was chosen to model heavy rain conditions in northern Thailand. Likewise, the aging test was chosen to model extreme heat and humidity conditions in the region. 3.4.1 Climate Simulation I: Water Immersion ASTM D570 (long-term immersion) was followed in order to conduct water absorption, the first climate simulation. The procedure served two purposes. First, it yielded water absorption rates, as samples were weighed at various intervals throughout the test. Second, it prepared samples for four post-conditioning tests: compression, shear, cell length (conducted with scanning electron microscopy), and conductivity. The first step in the water absorption test was to condition the cut samples in an oven at 50 ºC for 24 hours. Next, samples were placed in a flat-bottomed glass bowl, which was then filled with de-ionized water. Samples were kept at room temperature and submerged by placing aluminum foil on top of the water, aluminum cans on top of the foil, and a five pound weight on top of the cans. Samples were weighed at various intervals by removing samples from the container and wiping surface moisture with a dry cloth. Figure 3.5 shows samples being submersed in the water chamber. Figure 3.5: Samples being prepared for water absorption testing Photograph produced by the author, 2008 Deviations from ASTM D570 were as follows. The water sample sizes were dependent on the post-conditioning test: 1”x1”x1” for compression, 1”x1”x¼” for shear, 1”x1”x1” for SEM, and 4½”x4½”x1” for conductivity. Samples were weighed at 24 hours, 48 hours, one week, two weeks, three weeks, four weeks, five weeks, and six weeks after initial submersion. Conductivity samples were submersed in water for only one week due to time constraints. Samples were not re-conditioned in 41 an oven, as a large quantity of sediment was not observed after six weeks. The amount of absorbed water was calculated as the increase in weight = (wet weight – conditioned weight) / conditioned weight. Three samples were conditioned for each fiber type to obtain a meaningful average for both water absorption results as well as results for each of the three post-conditioning tests. 3.4.2 Climate Simulation II: Accelerated Aging Accelerating aging was performed per ASTM D2126 specifications. The test prepared samples for three post-conditioning tests: compression, shear, and cell measurement. Samples were first conditioned in an oven at 50 ºC for 24 hours. Next, samples were placed in an environmental testing chamber, or humidity oven, at USC’s mechanical engineering lab (see Figure 3.6). Controls were set at 40ºC and 85% humidity. Samples were conditioned for six weeks then weighed. Figure 3.6: Samples in USC’s environmental testing chamber Photograph produced by the author, 2008 42 43 Deviations from ASTM D2126 were as follows. As with the water absorption test, the sample size was dependent on the post-conditiong test: 1”x1”x1” for compression, 1”x1”x¼” for shear, and 1”x1”x1” for SEM. Samples were not conditioned in an oven (to extract moisture) before being placed in the chamber. No measurements were taken during the six week duration. Three samples were placed in the aging chamber for each fiber type to obtain a meaningful average for results of the post-conditioning tests. 3.5 Testing Following climate conditioning, samples were subjected to one of five tests designed to measure mechanical or thermal properties. Control samples that had undergone no climate conditioning were also tested in order to compare phenolic foam’s properties in both conditioned and unconditioned environments. Such a comparison showed the effects of water, heat, and humidity on phenolic foam’s properties. Table 3.2 summarizes the tests conducted. 44 Glass No Fibers Bamboo Aramid Cellulose EPS Compression No stress 3 3 3 3 3 3 Water 3 3 3 3 3 3 Aging 3 3 3 3 3 3 Shear No stress 3 3 3 3 3 3 Water 3 3 3 3 3 3 Aging 3 3 3 3 3 3 Conductivity No stress 1 1 1 1 1 1 Water 1 1 1 1 1 1 Flammability 3 3 3 3 3 3 Cell Length (measured with SEM) No stress 1 1 1 1 1 1 Water 1 1 1 1 1 1 Aging 1 1 1 1 1 1 Table 3.2: Tests conducted for each material. Numbers indicate numbers of samples tested. 3.5.1 Test I: Compression Compression testing was performed per ASTM D1621 specifications. An Instron machine in USC’s engineering department provided the necessary equipment and software (see Figure 3.7). Specimens were compressed between an upper and lower stainless steel platen, and load was applied at a crosshead speed of 0.5 mm/min. The sample was loaded in the direction of the foam rise direction since, because of the anisotropy of foam properties, load direction can yield drastically different results (Shen & Nutt 2003, p. 903). The Instron program automatically provided load- deformation curves as well as maximum load, yield strength, compressive strength, Young’s modulus, and ultimate stress for each specimen. Figure 3.7: Instron machine loaded with two stainless steel platens for compression testing Photograph produced by the author, 2008 Deviations from ASTM D1621 were as follows. Sample size was 1”x1”x1” instead of 2”x2”x1”. At least three samples were tested for each specimen, and the results were averaged. Specimens were either not conditioned or conditioned in water or aging chambers as described above. 3.5.2 Test II: Shear Shear testing was performed per ASTM C273 specifications. As with the compression test, an Instron machine provided the necessary equipment and software. Shear samples had to be glued between a bottom and top metal shear plate (see Fig. 3.8). 45 Figure 3.8: Shear samples being glued between shear plates Photograph produced by the author, 2008 Two-ton epoxy was used to adhere the sample ¼” away from each metal plate’s edge. Two-pound lead fishing weights were placed on the uppermost metal plate. After the epoxy dried for 12 hours, the sample was tested: the two plates were loaded in the machine, the upper plate was pulled in one direction, and the lower shear plate was pulled in the opposite direction (see Figure 3.9). The Instron program automatically provided outputs of maximum load and extension at maximum load. 46 Figure 3.9: Shear sample undergoing testing Photograph produced by the author, 2008 Deviations from ASTM C273 were as follows. Sample size was 1”x1”x¼” instead of 2”x12”x1”. At least three samples were tested for each specimen, and the results were averaged. Specimens were either not conditioned or conditioned in water or aging chambers as described above. 3.5.3 Test III: Cell Length (SEM) Scanning electron microscopy (SEM) was used to compare cell lengths in phenolic foam both before and after climate conditioning. This test was important because cell size can have considerable influence on the properties of phenolic foam. For example, thermal conductivity increases as cell size increases, since there are more paths available for heat transfer by radiation and convection. The Young’s modulus also increases as cell size increases (Klempner 2004, p.37). 47 A Cambridge 360 SEM machine was used at USC to conduct the test (see Figure 3.10). Samples were carefully cut using a razor blade. Small samples about 1/8”x½”x1/8” were cut from 1”x1”x1” cubes and adhered to a metal disk. The disk was placed in a gold sputtering chamber for about 15 minutes to impart electrical conductivity and then loaded into the SEM machine. The machine’s operating voltage was 10kV. Images were recorded in a high-resolution electronic format and processed with computer software. The machine provided zoom and measurement capabilities in order to measure cell length. Figure 3.10 Cambridge 360 Scanning Electron Microscope Photograph produced by the author, 2008 3.5.4 Test IV: Flammability UL 94 vertical burn test was followed in order to measure flammability. Samples were loaded into the metal fire chamber, and a Bunsen burner was lit and slid beneath the sample (see Figure 3.11). The flame was applied to samples for two 10 second intervals separated by the time it took (if any) for the combustion to cease 48 after the first application. The length of time that the sample retained flame for each application was measured. Figure 3.11: Sample loaded into fire testing chamber Photograph produced by the author, 2008 Deviations from UL94 were as follows. Sample size was 1”x1”x5” instead of ½”x½”x5”. Three samples were tested per specimen. In addition to combustion time, the volume of each sample was measured before and after the test in order to determine the percent volume of material consumed by flame. 3.5.5 Test V: Conductivity R-values were measured both before and after water immersion. The purpose was to measure R-value degradation due to water, an important factor when considering the long term performance of building materials exposed to the elements. Two separate tests were conducted. First, ASTM C518 was conducted by Oak Ridge National Laboratory in Oak Ridge, Tennessee yielding R-values. 49 Deviations from ASTM C518 were as follows. Sample size was 11”x11”x1” instead of 12”x12”x1”. A big challenge in this test was fabricating large samples, since Oak Ridge National Laboratory requires sizes of 8”x8”x1” or larger. Four smaller squares sized 5 ½”x5 ½”x1” were glued together to form the larger square. The four squares were glued together by constructing wooden frames about 12”x12” (see Figure 3.12). Warren Haby, USC’s composite materials lab technician, constructed the frames. Two-ton epoxy was applied to each square, and four squares were laid in the wooden frame. Screws were inserted in the frames every 2” and tightened so each small square pressed against each other. Wax paper and a thin piece of wood were sandwiched between each sample and screw to prevent deformation. Figure 3.12: Conductivity samples being glued together in 12” wooden frames Photograph produced by the author, 2008 After the glue dried for four hours, the screws were released and the final samples were shipped to Oak Ridge National Laboratory. Figure 3.13 shows how the overall 50 sample size of 11”x11”x1” was created by gluing four smaller panels together of size 5 ½”x5 ½”x1”. Figure 3.13: Cellulose fiber conductivity sample Photograph produced by the author, 2008 A modified conductivity test was also executed. The strategy was to compare temperature drops across the foam samples with temperature drops across a material of known R-value. By assuming that temperature and R-value had a linear relation, the R-value of the foam samples could be deduced. Samples - both conditioned after one week of water immersion and unconditioned - were separately placed inside a box consisting of three layers of materials. EPS panels comprised the outer two layers. The inner layer consisted of five EPS panels and one test panel: either unreinforced phenolic foam, fiber reinforced phenolic foam, or EPS. A 4W light bulb served as a heat source inside the box. Two iButtons (DS1921L) were positioned: inside the box and embedded in an EPS panel between the inner and middle layers. The inner iButton logged the temperature inside the box, and the middle iButton logged the temperature against the outer face of the test 51 52 panel. Finally, a radiant barrier consisting of an aluminum foil sheet was placed between the heat source and the test panel in order to minimize radiant gain. This configuration (see Figure 3.14) was used to calculate the temperature drop across the test panel. The R-value of phenolic foam under both dry and wet conditions was deduced using the following formula: R-value (test panel) = R-value (EPS) * Temperature Difference (test panel) / Temperature Difference (EPS). The R-value of EPS was specified on the product wrapping as 3.85 m²*K/W. Figure 3.14: Conductivity test apparatus Photograph produced by the author, 2008 3.6 Results Chapters 4, 5, and 6 present thesis results. Each chapter compares phenolic foam with EPS. Goals and key questions are first established. An application is chosen: the use of phenolic foam as an insulation material. Assumptions, limitations, and deviations from standard testing methods are outlined. Environmental impact categories are defined, and relevant data is referenced in order to provide quantitative results. Appropriate conclusions are drawn in Chapter 7, and a recommendation is 53 54 ultimately provided to meet the thesis goal. Future work building on these results is presented in Chapter 8. 3.6.1 Mechanical and Thermal Tests Chapter 4 presents results and analysis of the mechanical and thermal test data. Three sets of data comparisons were made: climate simulations, phenolic foam versus EPS, and fiber type. The Instron machine provided compression and shear data, the SEM machine measured cell lengths, iButtons measured temperature values, and a caliper obtained length measurements associated with the fire test. All data was arranged in Excel spreadsheets. Three samples were tested for most tests and climate simulations across each fiber type. The results were averaged in order to get a more meaningful result than just one sample. Bar graphs were constructed to aid in data analysis. 3.6.2 Environmental Impact Assessment Chapter 5 presents results of the environmental impact assessment. Various lifecycle cost assessment software programs were attempted to conduct a lifecycle cost analysis of phenolic foam and EPS, including Athena, Gemis, and GaBi. However, none of the programs included phenolic foam as a material to analyze. Instead, data was obtained by researching published information on phenolic foam’s and EPS’s embodied energy, emissions, and toxicity. In addition, the heat loss of each material was calculated by inputting the R-value of each material (obtained from the 55 conductivity test) into HEED (Home Energy Efficient Design), a energy analysis software tool. 3.6.3 Costs Chapter 6 presents results of the cost assessment to insulate a home in northern Thailand with either phenolic foam or EPS. Cost categories included raw materials, transportation, machinery, labor, and electricity. Both published data and estimates were used to arrive at a final cost estimate for each material. 56 Chapter 4: Test Results Test results compared various mechanical and thermal properties of phenolic foam and EPS. Density measurements were obtained, and then all materials underwent two environmental simulations: water and accelerated aging. Common sense dictated that only fiber materials with a density of 3.0 pcf underwent compression and shear testing. Materials below this value compressed quite easily when squeezed between index finger and thumb. Materials also underwent three other tests tied to various environmental responses: SEM cell length measurement, flammability, and conductivity. As described in Chapter 1, the objective of this thesis was to determine whether phenolic foam is suitable as a hybrid structural insulation material in hot, rainy climates like northern Thailand. Therefore, answers to the following five key questions were sought throughout analysis of the test results: • Are load-bearing applications possible? • Did phenolic foam withstand climatic stresses well? • Which material performed better in the tests, phenolic foam or EPS? • Did natural or synthetic fibers fare better? • Which material performed best overall? 4.1 Density After fabricating each type of fiber reinforced phenolic foam, length, width, height, and weight were measured. Density measurements were also taken after the two climate simulations. Table 4.1 shows density results for each fiber type (glass, bamboo, aramid, and cellulose) embedded in phenolic foam. Water and aging represent the two climate stresses. EPS represents the comparison material. Climatic Effects on Density 0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00 Material Density (pcf) No stress Water Aging No stress 3.08 4.29 2.42 4.93 4.01 0.95 Water 36.06 33.79 13.09 26.20 27.71 7.31 Aging 4.29 4.00 2.31 4.57 3.79 0.95 Glass No fiber Bamboo Aramid Cellulose EPS Table 4.1: Density results for all materials The table shows that the density of bamboo fiber reinforced phenolic foam was significantly less than the other fibers. As described in Chapter 3, the creation of uniform density samples was difficult due to mixing and fiber dispersion problems. Because mechanical properties of foam materials vary according to density, strength 57 testing was therefore not undertaken for bamboo fiber reinforced phenolic foam. The density of bamboo fiber reinforced phenolic foam also fell well below density values of common load-bearing materials (see Table 4.2). Table 4.2: Densities of common load-bearing materials http://www.engineersedge.com/manufacturing_spec/average_properties_structural_materials.htm 1 http://hypertextbook.com/facts/2004/KarenSutherland.shtml The remaining materials underwent compression and shear testing, both before and after water and aging simulations, in order to determine their viability as load- bearing materials under extreme weather conditions. It should be noted that the density of EPS was significantly less than fiber reinforced phenolic foam. For example, Table 4.1 indicates that EPS was 69% less dense than glass fiber reinforced phenolic foam. Despite this disparity, this thesis tested EPS against the same set of tests as phenolic foam, since one of the project goals was to compare mechanical properties of phenolic foam and commonly available insulation materials regardless of density differences. EPS was also bought at Home Depot, and therefore its density value could not be adjusted through the fabrication process as with phenolic foam. 58 Table 4.1 also shows the effect of water and aging climate simulations on density. The table suggests that neither phenolic foam nor EPS is a very water resistant material, with the density of glass fiber reinforced phenolic foam exceeding 36 pcf after the water absorption test. Table 4.3 shows the percentage increase in density after water exposure. Density, % Increase after Water Exposure 0.00 200.00 400.00 600.00 800.00 1000.00 1200.00 Material Density, % Increase after Water Exposure Water Density, % Increase 1070.78 687.65 440.91 431.44 591.02 669.47 Glass No fiber Bamboo Aramid Cellulose EPS Table 4.3: Percentage increase in density after 8-week water exposure for all materials Bamboo and aramid showed the least increase in density, although both materials still increased their weight by a significant percentage. Glass performed the worst, and EPS performed better than some fibers but not as well as others. Similarly, natural fibers (cellulose and bamboo) performed better than some but not all of the materials. These observations on density increase after water exposure suggest no clear advantage when comparing phenolic foam versus EPS or natural versus synthetic fibers. Further discussion of water and aging tests follows in Section 4.3. 59 60 4.2 Strength Tests Results of strength tests determined whether phenolic foam could serve as a load- bearing material. 4.2.1 Compression Fiber reinforced phenolic foams with a density value of 3.0 pcf or higher as well as EPS underwent compression and shear testing. Sample sizes were 1”x1”x1”. Compression results are shown in Tables 4.4 and 4.5. Compressive strength is defined as the maximum stress a material can sustain under crush loading. For samples that do not shatter in compression, the compressive strength is the amount of stress required to distort the material an arbitrary amount. Compressive strength is calculated by dividing the maximum load by the original cross-sectional area of the sample (Instron 2008). Climatic Effects on Compressive Strength 0.00 20.00 40.00 60.00 80.00 Material Compressive Strength (psi) No stress Water Aging No stress 21.47 44.64 47.47 49.04 12.85 Water 47.76 33.63 30.83 36.78 12.33 Aging 74.08 44.36 54.27 34.07 12.69 Glass No Fiber Aramid Cellulose EPS Table 4.4: Compression results: climatic effects on compressive strength Climatic Effects on Young's Modulus 0.00 500.00 1000.00 1500.00 2000.00 2500.00 Material Young's Modulus (psi) No stress Water Aging No stress 615.80 1935.57 2226.79 933.99 83.63 Water 1140.81 1093.53 1200.85 1043.73 153.25 Aging 937.76 1301.21 1318.61 523.56 92.82 Glass No fiber Aramid Cellulose EPS Table 4.5: Compression results: climatic effects on Young’s modulus The results of Tables 4.4 and 4.5 were used to evaluate the following three questions: • Can phenolic foam be used as a load-bearing material in Thailand? 61 • Do climate stresses weaken phenolic foam? • Do natural or synthetic fibers add greater strength to phenolic foam? 4.2.1.1 Can Phenolic Foam Be Used as a Load-Bearing Material in Thailand? A simple calculation using the results from Tables 4.4 and 4.5 shows whether phenolic foam could be used as a load-bearing material in northern Thailand. Assume a typical one-story house in the region has wood floors having weight 25 psf and dimensions 40’ x 50’. The house consists of four rooms each 20’x 25’ (see Figure 5.1). A 1’ wide wall strip is analyzed. Figure 4.1: Assumed typical 40’ x 50’ house plan in northern Thailand Diagram courtesy of G.G. Schierle, 2008 The tributary area is 1’ x 20’ = 20 ft 2 . 62 The resisting area is 8” x 12” = 96 in². The tributary length is 20’. P = uniform load x tributary area = 25 psf x 20’ x 1’ = 500 lb Wall stress f = P/A = 500 lb / 96 in² = 5.2 psi The average compressive value for all phenolic foam materials in Table 4.4 is used = 43.2 psi. A reasonable safety factor is 10, considering the data has a high standard deviation (more consistent strength results would yield a lower safety factor). 43.2 psi / 10 = 4.32 psi 4.32 < 5.2, NOT OK This result shows that phenolic foam is not strong enough to be used as a load- bearing material. Consider also Table 4.6, which gives Young’s modulus values for common load- bearing materials. Published values for industrial grade phenolic resin and polystyrene are included for comparison purposes. 63 http://www.diracdelta.co.uk/science/source/y/o/youngs%20modulus/source.html 2 Schierle, Structure and Design Table 4.6: Young’s modulus values for common building materials Industrial grade phenolic foam is significantly less stiff than almost all common building materials. Aramid fiber reinforced phenolic foam, the strongest material fabricated in Table 4.5 at 2227 psi, is over 700 times less stiff than Douglas fir-larch wood and is comparable to concrete. This is likely due to the low density of phenolic foam produced in this thesis. The table also shows that phenolic resin has a stiffness about 200 times greater than the fabricated phenolic foam. Again, the most likely reason for the difference is that the fabricated phenolic foam was not very dense. In addition, the foam contained a low percentage of fibers (4% for glass, aramid, and cellulose; 2% for bamboo) due to mixing difficulties; a higher fiber percentage would have increased the density and, correspondingly, the strength values. As noted in Chapter 3, non-uniform fiber distribution also likely contributed to low strength values. See Chapter 8: Future Work for further discussion about phenolic foam’s potential as a load-bearing material. 64 65 4.2.1.2 Do Climate Stresses Weaken Phenolic Foam? Compressive results are inconclusive when evaluating phenolic foam’s ability to withstand climate stresses. Tables 4.4 and 4.5 show that, in some cases, phenolic foam’s post-stressed strength values were higher than with no stress. In other cases, the unstressed samples were strongest. This phenomenon may have occurred because the same exact sample was not tested for each stress condition: unstressed, water, and aging. Each sample varied in its fiber content, fiber dispersion, and overall density, likely leading to the flux in strength values across climate simulation. This problem could be remedied by creating more uniform samples. Compressive strength of EPS was only slightly affected by climatic stresses, although this is hardly an advantage over phenolic foam considering unstressed EPS was much weaker than phenolic foam in the first place. 4.2.1.3 Do Natural or Synthetic Fibers Provide Greater Strength to Phenolic Foam? Compression results are also inconclusive as to whether natural or synthetic fibers withstood climate stresses better, and the distinction appeared somewhat arbitrary. For example, cellulose fiber reinforced phenolic foam had a higher compressive strength and Young’s modulus across climate stress than some fibers but had lower values than other fibers. 4.2.2 Shear Shear results are shown in Table 4.7. Climatic Effects on Maximum Load 0.00 50.00 100.00 150.00 Material Maximum Load (N) No climate stress Water Aging No climate stress 90.97 79.09 112.88 69.58 89.30 Water 84.80 74.30 119.69 91.43 45.56 Aging 97.24 71.53 84.27 86.53 62.56 Glass No fiber Aramid Cellulose EPS Table 4.7: Shear results: climatic effects on maximum shear load Shear results reinforce phenolic foam’s inability to act as a load-bearing member. For example, plywood with steel stud backing (24” O.C. spacing) has a maximum shear load of 3225 N (Dietrich Metal Framing Company n.d.). This value is over 30 times the measured shear load of phenolic foam. As with the compression test, greater shear strength could be achieved by creating samples with a higher percentage of fiber reinforcements. This issue is discussed in Chapter 8: Future Work. Shear results are inconclusive when evaluating phenolic foam’s ability to withstand climatic stresses. Water and aging would presumably reduce shear stress maximum load, yet only unreinforced phenolic foam showed a progressive drop in shear values (6% after the water test and 10% after the aging test). The reason again likely relates 66 67 to the non-uniform fiber dispersion in the samples; more uniform samples would have to be fabricated in order to create more informative results. Finally, shear results are largely inconclusive when evaluating phenolic foam versus EPS and natural versus synthetic fibers, as no one material or fiber type performed consistently well across climate stress. 4.3 Climate Simulations Strength tests outlined in 4.2 showed that phenolic foam is not strong enough to act as a load-bearing material. Non-load bearing applications were then considered. In order to determine how suitable the material is for non-load bearing insulation and cladding applications, the material’s response to water immersion and accelerated aging was measured. The water simulation determined how well the material would perform under long-term water exposure, and accelerated aging determined how well the material would perform under intense heat and humidity. Bamboo fiber reinforced phenolic foam was considered for these applications since, even though the material had a low density, load-bearing applications were not being considered. Glass, aramid, and cellulose fibers were also evaluated along with unreinforced phenolic foam and EPS. Climate simulation results answered the following questions: • How well does phenolic foam withstand extreme weather conditions? 68 • Does phenolic foam or EPS perform better under extreme weather? • Do natural or synthetic fibers perform better under extreme weather? 4.3.1 Water Immersion Two tables express water absorption values. First, Table 4.8 gives water absorption rates, where weekly amount of water absorbed was calculated according to the following equation: Water absorption rate = (wet weight – conditioned weight) / initial mass * 100. The conditioned weight represented the sample weight after being placed in an oven at 50ºC for one hour but before being placed in the water immersion chamber. Water Absorption Rates 0 100 200 300 400 500 600 700 800 900 1000 Weeks Elapsed Water Absorbed, % (mass/mass) Glass No Fibers Bamboo Aramid Cellulose EPS Glass 0 285.01 388.21 463.92 631.38 696.39 741.81 758.68 772.02 No Fibers 0 272.53 406.5 494.23 676.56 754.31 812.67 847.33 869.14 Bamboo 0 218.92 323.51 416.71 492.67 534.76 534.76 534.76 534.76 Aramid 0 267.36 347.90 401.98 585.74 636.63 678.45 703.56 709.39 Cellulose 0 453.67 672.59 750.21 792.16 817.42 817.42 817.42 817.42 EPS 0 643.55 660.41 691.54 691.54 691.54 691.54 691.54 691.54 01 234 5678 Table 4.8: Water absorption rates for phenolic foam and EPS expressed as mass of water absorbed as a percentage of mass of material Table 4.9 presents water absorption values after both 24 hours and 8 weeks. Water Absorption, % 0 200 400 600 800 1000 Material Water Absorption, % (mass/mass) 24-Hour Immersion 8-Week Immersion 24-Hour Immersion 155.42 156.21 102.96 159.35 183.91 435.56 8-Week Immersion 772.02 869.14 534.76 709.39 817.42 691.54 Glass No fibers Bamboo Aramid Cellulose EPS Table 4.9: Mass of water absorbed expressed as percentage of mass of sample 69 The results show that both phenolic foam and EPS absorbed water at a rapid rate. The EPS value, however, tapered off fairly quickly. The phenolic foam absorption rate took several weeks to slow down and converge towards a value. Also, the addition of fibers in all cases reduced water absorption after eight weeks when compared to unreinforced phenolic foam. Finally, natural fibers performed better than synthetic fibers in one case and worse in another case, as bamboo absorbed the least amount of water and cellulose absorbed the most amount of water. These results suggest that bamboo fibers used to reinforce phenolic foam may be an effective method of reducing water absorption in rain screen or other cladding applications. This is reinforced by the fact that bamboo fiber reinforced foam absorbed the least amount of water of all materials tested. Table 4.10 compares 24-hour water absorption values for some common insulation materials. Table 4.10: Water absorption values for some common insulation materials http://www.buildingscience.com/bsc/designsthatwork/buildingmaterials.htm 70 The 24-hour immersion values in Table 4.9 may be compared with the values in Table 4.10. Common insulation materials absorb far less water than phenolic foam or EPS; bamboo was the least water resistant fiber yet doubled its weight in the first 24 hours of exposure. Because northern Thailand receives significant amounts of rain in the wet season, exterior cladding applications such as rain screens using phenolic foam as a core material would require a skin layer adhered to the foam. 4.3.2 Accelerated Aging Results of the accelerated aging test are summarized in Table 4.11. Mass lost as a percentage of initial mass was calculated by subtracting the final mass of each sample after six weeks in the environmental chamber from the initial mass then dividing by the initial mass and multiplying by 100. Accelerated Aging Results 0 5 10 15 20 Material Mass Lost, % of Initial Mass Mass Lost, % 6.42 5.31 6.05 5.79 14.46 2.33 Glass No fibers Bamboo Aramid Cellulose EPS Table 4.11: Accelerated aging results: mass lost as a percentage of initial mass 71 72 The results show that phenolic foam lost a fair amount of material over the six weeks at 40ºC and 85% humidity. EPS performed somewhat better and lost half as much mass as phenolic foam. Aramid performed best among fibers, and cellulose performed the worst, suggesting natural fiber reinforced phenolic foam may not be the best choice of cladding material to withstand extreme heat and humidity. 4.4 Other Tests The three remaining tests measured cell length (measured through SEM), fire resistance, and conductivity. These tests were performed to further evaluate phenolic foam’s potential as an insulation or cladding material. 4.4.1 Cell Length (SEM) SEM was used in order to measure cell length of samples before and after climate conditioning. The test was important because it determined whether extreme weather altered materials’ cell length as well as whether a correlation could be drawn between cell length and water resistance. Table 4.12 shows SEM images of the various samples before and after climatic stresses. The numbers next to the images indicate the magnification level of the microscope. Table 4.12: SEM images showing cell length across climatic stress Table 4.13 quantifies the cell lengths. Measurements were taken using software accompanying the Cambridge 360 SEM machine. 73 SEM Results: Climatic Effects on Cell Length 0 200 400 600 Material Cell length (microns) No stress Water Aging No stress 140 330 150 170 186 140 Water 164 445 176 157 144 121 Aging 188 412 224 245 211 162 Glass No fiber Bamboo Aramid Cellulose EPS Table 4.13: SEM results: cell length across climatic stress The results indicate that unreinforced phenolic foam’s cell length was much greater than the other materials’ cell lengths. This result is important because it suggests that fiber reinforcement has the benefit of reducing cell size. Comparison of Table 4.13 with Table 4.8 (water absorption results) confirms that a correlation can be drawn between cell length and water resistance: the greater the cell size, the worse the water resistance; the smaller the cell size, the better the water resistance. Reduced cell length therefore offers an explanation of why water absorption values for fiber reinforced phenolic foam were less than unreinforced phenolic foam values. Less informative cell length conclusions are as follows. For most fibers, environmental stress had little, if any, effect on cell size. EPS cell lengths were equal to or slightly less than phenolic foam cell lengths, and natural and synthetic fibers’ cell lengths were about the same. 74 4.4.2 Fire Resistance Table 4.14 gives results of the flammability test. Table 4.14: Flammability test results: volume lost as a percentage of initial volume Results show that fire resistance is one of phenolic foam’s biggest advantages over EPS. Phenolic foam was not consumed by fire whereas EPS was consumed. This result has significant implications when considering whether to use phenolic foam or EPS as an insulation material. Table 4.15 gives another result obtained by using the classification system defined by testing standard UL 94. VO is the best rating, and V1 is not as good. 75 Table 4.15: Flammability test results: UL 94 ratings assigned to materials Results clearly show that phenolic foam rates as a more fire resistant material than EPS. 4.4.3 Conductivity Two sets of conductivity results were obtained: those recorded by Oak Ridge National Laboratory per ASTM C518, and those recorded by the author using a homemade conductivity apparatus to measure temperature drops across materials (see Chapter 3). Table 4.16 presents R-values recorded by Oak Ridge National Laboratory. R-values obtained by Oak Ridge National Laboratory 0.00 2.00 4.00 6.00 Material R-value (m2*K/W) No Stress 4.00 4.07 4.05 4.00 Glass No Fibers Aramid Cellulose Table 4.16: R-values recorded by Oak Ridge National Laboratory per ASTM C518 76 Oak Ridge was only able to record reliable values for glass fiber, aramid fiber, cellulose fiber, and unreinforced phenolic foam. The values were nearly identical, and the test offered no comparison with EPS. The conductivity test conducted by the author was more useful by providing EPS and water immersion R-values. Tables 4.17 and 4.18 show results of the conductivity test for both unconditioned materials and materials immersed in water for one week. The tables show the temperature change across each material as a function of time. Conductivity Results: Temperature Drop (No Water Immersion) 0 10 20 30 40 50 60 70 1 10192837 465564 738291 100 Time (minutes) Temperature Change (°F) Glass No Fibers Bamboo Aramid Cellulose EPS Table 4.17: Conductivity results: temperature drop across dry materials 77 Table 4.18: Conductivity results: temperature drop across wet materials (one week immersion) R-values are shown in Table 4.19, calculated using the following equation: R-value (unknown material) = Maximum temperature drop (unknown material) / Maximum temperature drop (known material) * R-value (known material). EPS served as the known material. 78 R-values 0 2 4 6 Material R-value (m2*K/W) No Stress Water No Stress 5.22 5.13 5.3 5.65 4.62 3.85 Water 2.31 2.31 1.97 2.82 1.97 3.08 Glass No Fibers Bamboo Aramid Cellulose EPS Table 4.19: Conductivity test results: R-values The most significant result was that water greatly affected R-value degradation, and EPS outperformed phenolic foam in this respect. Although EPS without any water exposure had a lower R-value than phenolic foam, the material had a higher R-value than phenolic foam after water exposure. This result showed that R-values should be carefully considered when insulating buildings. Just because a material has a lower published R-value than another material does not mean that it will not insulate better after exposure to climate stresses. Phenolic foam’s poor thermal response to climate stress therefore suggests it may not be the best insulation material in wet regions like northern Thailand. Other conclusions include aramid fiber reinforced phenolic foam performed the best among fibers, and no significant differences separated natural and synthetic fibers. 79 Also, the phenolic foam R-values recorded by Oak Ridge National Laboratory were slightly less than the values recorded by the author. This was likely for two reasons: first, Oak Ridge followed ASTM C518, whereas the author measured temperature drops using no ASTM standard. Second, each test used different samples, which likely had different densities and fiber composition. R-values of common insulation materials under no water stress are provided in Table 4.20. Table 4.20: R-values of common materials http://www.buildingscience.com/bsc/designsthatwork/buildingmaterials.htm Unstressed phenolic foam has R-values that are competitive with these materials, with only rigid polyisocyanurate insulation having a higher value. However, phenolic foam’s insulative value is not retained when exposed to climate stresses such as water. 80 81 R-value degradation is also important when considering energy losses and costs of building insulation. These issues are more thoroughly discussed in Chapter 5: Environmental Impact and Chapter 6: Costs. 4.5 Summary Results show that phenolic foam cannot be used as a load-bearing material given the low densities and strength values of the samples tested. However, phenolic foam has potential as an insulation and/or cladding material. It is lightweight, fire resistant, and insulative (under no climate stresses) and performs reasonably well after extended exposure to intense heat and humidity. Neither natural nor synthetic fibers display any consistent advantage, suggesting this distinction is somewhat arbitrary as far as mechanical tests are concerned. Bamboo fiber reinforced phenolic foam, however, has fair water resistance. Aramid fiber reinforced phenolic foam was also the strongest and most insulative material. Another useful conclusion is that cell length can be correlated with water resistance, suggesting denser materials should be better at resisting water. Chapter 8 discusses ways of increasing the density of phenolic foam. Finally, phenolic foam’s best advantages over EPS are its fire resistance and insulative value, although the material’s R-value deteriorates severely when exposed to extreme flooding conditions. 82 Chapter 5: Environmental Impact An environmental impact assessment of insulation materials is essential when comparing and choosing insulation materials. The result should benefit the environment, as a well-informed decision can reduce building energy consumption costs dramatically. This is important because buildings are a huge source of energy consumption. For example, homes account for 21% of all energy used in the United States every year, and the average annual utility bill is $1767.00 (United States Department of Energy 2008). Therefore, as an energy-saving tool, insulation plays a critical role in determining building operations costs. Insulation material selection is broad. The most popular insulation materials comprising about 90% of the insulation market are inorganic fibrous (glass wool and stone wool) and organic foamy (EPS, extruded polystyrene, and polyurethane) materials (Papadopoulos & Giama 2006, p. 2178). Environmentally friendly materials based on agricultural raw materials have found limited market share due to high cost and limited accessibility of information. Environmental impact assessments can therefore better inform the public on insulation material selection and ultimately help citizens make “greener” insulation choices. 83 5.1 Prior Environmental Impact Studies on Insulation Limited data exists on the environmental impact of insulation materials. Data on phenolic foam is especially lacking. Schmidt performed a life cycle assessment of stone wool, flax, and cellulose wool insulation (2003). Paper wool had the lowest global and regional environmental impacts, and flax insulation had the highest. Stone wool had the lowest total energy use, followed by cellulose and flax. Stone wool was also found to be the most occupationally healthy material, mainly because of the absence of carcinogenic properties. Aside from Schmidt’s results, however, no studies rigorously and comprehensively evaluate the environmental advantages and disadvantages of various insulation materials and systems. 5.2 Difficulty in Assessing the Environmental Impact of Phenolic Foam Instead of using existing studies, then, raw data was researched in order to assess the environmental impact of phenolic foam. However, perhaps because of phenolic foam’s extremely limited market share, raw data is also limited. The United States Department of the Interior notes that the environmental characteristics of foam are complex and not well understood (2008). Moreover, lifecycle cost analysis software such at Athena, Gemis, and GaBi do not list phenolic foam as an insulation material, and the material is often not found in data specification sheets on insulation materials. A thorough lifecycle cost assessment model would supply all first and future costs, including production, manufacturing, transportation, maintenance, 84 labor, installation, and repair costs. However, estimating such costs in this thesis would have required considerable guesswork and assumptions. Another reason producing a comprehensive environmental impact assessment of phenolic foam proved quite a challenge is because chemicals used in phenolic foam production have frequently changed in the last 30 years, especially as hydrocarbons have become a more appealing and environmentally sensitive substitute for hydrofluorocarbons. Section 5.4.4.1 details the evolving nature of phenolic foam’s production process. 5.3 Method The approach this thesis took on assessing the environmental impact of phenolic foam was to use two sources: • data generated in Chapter 4: Results • published data. These two sources were used to compare the environmental impact of phenolic foam with EPS. R-values calculated in Chapter 4 were found to be particularly useful in this assessment. By building on its own previously acquired data, the thesis succeeded in providing an environmental impact assessment. Several environmental impact categories, including embodied energy, CO 2 emissions, and the correlation between R-value and heat loss, helped determine which material was more environmentally sensitive. 85 Data estimates from suppliers were also used in estimating the cost of insulating a home in northern Thailand with phenolic foam, including production, transportation, labor, and purchase costs. This assessment is presented in Chapter 6 and helped determine which material was not only environmentally friendly but also economically viable. A final recommendation is presented in Chapter 7, which combines results of this chapter, the test results of Chapter 4, and the cost results of Chapter 6. This conclusion yields the thesis’s goal of providing a recommendation on which insulation material is best suited for use in hot, rainy regions such as northern Thailand. 5.4 Results: Environmental Impact Assessment of Phenolic Foam Several considerations were taken into account when determining the environmental impact of phenolic foam. Heat loss, moisture resistance, embodied energy, emissions in the form of hydrochlorofluorocarbons (HCFCs) and CO 2 , and toxicity were used to compare the environmental impact of phenolic foam with EPS. Thesis data helped analyze the first two factors, and published data were used for the final three factors. 5.4.1 Heat Loss Which material produces more heat loss in a building in northern Thailand, phenolic foam or EPS? The energy software program HEED (Home Energy Efficient Design) was used to answer this question. R-values for materials both before and after water stress were used. This was important since, as Chapter 4 showed, EPS’s R-value degrades more slowly than phenolic foam’s R-value. Therefore, pre- and post-water 86 stress R-values used in heat loss calculations would presumably yield different recommendations. 5.4.1.1 Building the Model Since only relative heat loss data was desired, only basic specifications were inputted into HEED to create a rough model of a rural house in northern Thailand: • one-story • 40’x52’ (modeled after figure 4.1; note: HEED dimensions are in multiples of 4, so 40’x50’ was not possible; 40’x52’ was used.) • 8’ tall ceilings • 4 residents • no garage • windows: o 7 on south façade, 2 on east façade, 3 on north façade, 2 on west facade o clear single pane 1/8” glass in aluminum frame (“no glass”, typical of homes in rural northern Thailand, was not an option) • insulated attic and raised floor; radiant barrier installed in attic • walls: wood siding on 2x6 wood studs at 24”, plaster board interior • roof: white (cool roof) elastomeric membrane; flat roof • floors: wood floors exposed • under first floor condition: area open to exterior • infiltration: very poorly sealed older building • ventilation: good natural ventilation (up to 5.0 air changes per hour) by opening windows as needed (no fans) • heating: no furnace • cooling: no air conditioner • overhangs: fixed all year long, never retracted Additionally, since only climate regions in the United States were options, Los Angeles (zip code: 90007) was chosen as the location. Both northern Thailand and Los Angeles have warm temperatures year-round. The HEED model of this house is shown in Figure 5.1: Figure 5.1: Simple one-story house in northern Thailand constructed in HEED 87 5.4.1.2 Calculating Heat Loss In order to calculate heat loss, wall U-values had to be entered into HEED. It was assumed that the average wall U-value was equal to the U-value of a wall constructed entirely of 1” thick phenolic foam. No interior or exterior layers were applied. Only relative heat losses were desired, so the U-value of the windows was not factored in to the calculation. It is recognized that this does not give data on the magnitude of the net impact, only relative impact between phenolic foam and EPS. Table 5.1 presents U-values of phenolic foam and EPS before and after water stress. These values were obtained by inverting the R-values obtained in the conductivity results in Chapter 4 (see Table 4.19). U-values 0.000 0.200 0.400 0.600 Material U-value (W/m2*K) No Stress 0.192 0.195 0.189 0.177 0.216 0.260 Water 0.433 0.433 0.508 0.355 0.508 0.325 Glass No Fibers Bamboo Aramid Cellulose EPS Table 5.1: U-values entered into HEED to obtain energy loss data on each separate material 88 These values were entered into HEED, and Table 5.2 presents the resulting energy losses of the building both before and after the materials had been exposed to water. Energy Losses in a Northern Thai House 0 20000 40000 60000 80000 Material BTUH No stress 44134 44270 43997 43452 45225 47226 Water 55093 55093 58503 51546 58503 50182 Glass No Fibers Bamboo Aramid Cellulose EPS Table 5.2: Effect of water exposure on energy losses in a northern Thai house Results show that EPS performed about 8.7% worse than aramid, the best- performing material, before water exposure. EPS performed 2.6% better than aramid, however, after water exposure. These results suggest that EPS would retain energy slightly better than phenolic foam after prolonged exposure to rain. Heat loss, however, is not too relevant in northern Thailand’s warm climate. The analysis could be more useful in colder regions or regions which use heaters. Still, the results show moisture effects on relative energy losses of homes insulated with phenolic foam and EPS. 89 90 The results of Table 5.2 were used in Chapter 6: Costs to estimate costs related to energy losses in buildings. 5.4.2 Moisture Resistance Moisture resistance is a more relevant indicator of the environmental impact of phenolic foam, as northern Thailand receives tremendous amounts of rain during the rainy season from April to September. The ability to resist both infiltration and absorption of water plays a critical factor in long-term thermal performance of building materials in the region. Most importantly, a stable R-value is linked to high resistance of water moisture absorption (Fabian et al. 2004). This translates to less energy loss. Table 5.3 reproduces the results of Table 4.8, which show water absorption rates of phenolic foam and EPS. Water Absorption Rates 0 100 200 300 400 500 600 700 800 900 1000 Weeks Elapsed Water Absorbed, % (mass/mass) Glass No Fibers Bamboo Aramid Cellulose EPS Glass 0 285.01 388.21 463.92 631.38 696.39 741.81 758.68 772.02 No Fibers 0 272.53 406.5 494.23 676.56 754.31 812.67 847.33 869.14 Bamboo 0 218.92 323.51 416.71 492.67 534.76 534.76 534.76 534.76 Aramid 0 267.36 347.90 401.98 585.74 636.63 678.45 703.56 709.39 Cellulose 0 453.67 672.59 750.21 792.16 817.42 817.42 817.42 817.42 EPS 0 643.55 660.41 691.54 691.54 691.54 691.54 691.54 691.54 01 234 5678 Table 5.3: Water absorption rates for phenolic foam and EPS expressed as mass of water absorbed as a percentage of mass of material As noted in Chapter 4, bamboo and aramid fiber reinforced phenolic foam and EPS resisted water the best, and unreinforced phenolic foam resisted water the worst. EPS initially absorbed more water, but its absorption rate quickly tapered off and the material ultimately absorbed less water than all other materials except bamboo fiber reinforced phenolic foam. Overall, neither phenolic foam nor EPS faired well in resisting water, since all materials showed a threefold increase in weight after water immersion. In addition, both phenolic foam and EPS R-values significantly decreased after water exposure, hurting their thermal performance. As noted in Table 5.1, phenolic foam’s R-value deteriorated more than EPS’s R-value, suggesting 91 92 buildings insulated with phenolic foam would experience slighter greater energy losses after prolonged exposure to rain. 5.4.3 Embodied Energy Embodied energy is another useful indicator of the environmental impact of phenolic foam insulation in northern Thai homes. As the United States Department of the Interior notes, insulation manufacturing processes generate pollution as a result of fossil fuel combustion. The simplest way the Department recommends comparing manufacturing impacts is to compare the manufacturing energy required, or embodied energy (2008). John Barry Associates Architects and the United States Air Force Air Combat Command produced a “Sustainable Facility Guide” to creating energy efficient buildings (2000). Besides weighing the advantages of using spray, blow-in, or rigid insulation under different applications, the guide includes a table of embodied energy values for the most common insulation materials (see Table 5.4; first appears as Table 2.4). Table 5.4: Embodied energy of common insulation materials 1 http://www.canadianarchitect.com/asf/perspectives_sustainibility/measures_of_sustainablity/measur es_of_sustainablity_embodied.htm The guide also describes the production of phenolic resins used in plastic materials and gives the embodied energy as 34,383 Btu/lb. This value is 28% less than the value for EPS. Assuming phenolic foam has an embodied energy comparable to plastic materials made out of phenolic resin (a reasonable assumption given plastics and foam are typically 60% or more phenolic resin by mass), phenolic foam’s embodied energy is considerably less, and has a considerably less environmental impact, than EPS’s embodied energy. 5.4.4 Emissions Phenolic foam and EPS are closed-cell foam insulation materials, meaning they consist of cells closed to prevent the passage of air. These cells contain air or gas- filled pockets arranged to retard heat flow, and blowing agents are required to expand these materials and create the pockets. Cellular foam insulation is typically classified as either thermoplastic foams (EPS and extruded polystyrene) or thermoset 93 94 plastic foams (phenolic, polyurethane, and polyisocyanurate) (Kalinger & Drouin 2001). As long as phenolic foam and EPS use pentane as a blowing agent, they are considered CFC (chlorofluorocarbon)-free insulation and exhibit low levels of toxic chemical emissions. However, the use of HFCs would increase the level of toxic emissions. The following investigation gives a historical look at foam emissions, provides specific quantitative data on HFC emissions, and offers a basis for comparing carbon emissions in phenolic foam and EPS. 5.4.4.1 Phenolic Foam Chlorofluorocarbons (CFCs) were used up until the 1980s as blowing agents for phenolic and other thermoset foams. Since the conductivity of CFCs is lower than air, the conductivity of the insulation was lower than air-filled materials. CFCs in phenolic foam also had low toxicity and were odorless and nonflammable. Their biggest drawback was that chlorine in CFCs degraded the Earth’s ozone layer. Hydrochlorofluorocarbons (HCFCs) appeared to be a solution, as their ozone depleting potential (ODP) was only 5-10% of CFCs (Kalinger & Drouin 2001). However, HCFCs or CFCs were still emitted into the atmosphere during foam manufacturing or on-site application, while in use, or when discarded. For example, as the foams aged, blowing agents migrated out of the foam according to a leakage rate defined mainly by foam cell structure and solubility (United States Environmental Protection Agency 2001). The Copenhagen Amendment of 1992 95 restricted their use and production, and they are planned to be phased out of foam production by 2030. Hydrocarbons and hydrofluorocarbons (HFCs) have been used as substitute blowing agents. For example, n-pentane is a hydrocarbon that was used in the fabrication of phenolic foam in this thesis (see Chapter 3). This material has high availability, low cost, and zero ODP. In addition, its low solubility means it has a low diffusion rate out of foam (Grimminger & Muha 1995). This translates into minimum carbon emissions. Unfortunately, n-pentane is not always used as a blowing agent in commercially manufactured phenolic foam, as HFCs and other ozone depleting substitutes are commonly used. The United States Environmental Protection Agency has created an emissions profile for HFC blowing agents used in phenolic foam (see Table 5.5) (Godwin 2003). The values were calculated using the following equation: Ej = O (efi x Qcj-i+1) for i = 1 -> k Where E = Emissions. Total emissions of a specific chemical in year j for closed-cell foam blowing, by weight. Ef = Emission Factor. Percent of foam’s original charge emitted in each year (1 -> k). This factor includes a rate for manufacturing emissions (occurs during the first year of foam life), annual emissions (every year throughout the foam lifetime), and disposal emissions (occurs during the final year of foam life). Qc = Quantity of Chemical. Total amount of a specific chemical used in closed-cell foams in year j. k = Lifetime. Average lifetime of foam product. HFC Emissions in Phenolic Foam 25% 36% 39% Loss at Manufacturing (%) Leakage (%) Loss at Disposal (%) Table 5.5: Hydrofluorocarbon (HFC) emissions in phenolic foam, in the case where HFCs are used as blowing agents. The leakage rate is 1.125% per year, and the leakage lifetime is 32 years. 96 97 The table shows that emissions are greatest during the final year of foam life. Emissions during the product lifetime are slightly less than this value, and emissions during manufacturing are the least. 5.4.4.2 EPS The Environmental Protection Agency describes EPS as a non-emissive plastic. It is made with neither CFCs nor HCFCs and is expanded with the hydrocarbon n- pentane, similar to the fabrication of phenolic foam in this thesis. The United States Department of the Interior notes that pentane contributes to smog but not ozone depletion or global warming (2008). The pentane leaks out of the EPS and is replaced with air. The Department further notes that in California, many plants used to manufacture EPS now recover up to 95% of pentane used in production. 5.4.4.3 Phenolic Foam Versus EPS The emissions profiles of phenolic foam and EPS production, therefore, largely depend on the chemicals used. Assuming phenolic foam production uses HFCs as blowing agents, EPS would have the better emissions profile. Assuming both use the hydrocarbon n-pentane, the environmental impact of the two materials based on emissions would be similar. One final distinction the United States Environmental Protection Agency draws between the two materials is the carbon content of phenolic and EPS resins. Phenolic resin (phenol) has a somewhat lower carbon content at 77%, whereas polystyrene’s carbon content is 92% (2007). Uncertainty in these values, particularly in phenolic foam’s carbon content, is high. Some of the carbon may be released into the atmosphere during the life of the products, but the Environmental Protection Agency notes that this amount is likely to be small (2007). Therefore, this analysis suggests that carbon emissions during the life of both phenolic foam and EPS are difficult to compare and are assumed to be reasonably similar. 5.4.5 Toxicity Both phenolic foam and EPS exhibit low toxicity. Analysis below of two separate studies on toxic properties shows that no significant toxic differences separate the two materials. 5.4.5.1 Phenolic Foam Composite Building Structures, a manufacturer of homes, prepared a fire, smoke, and toxicity (FST) study of phenolic foam (2007). The material was burned, and toxins were measured against the target values for the International Maritime Organization’s fire test procedures code (IMO FTP code). Table 5.6 presents the results. Table 5.6: Toxicity concentrations in phenolic foam 98 99 Phenolic foam gas concentrations fall below the IMO FTO Code values for every toxin. The test also confirmed flammability results shown in Table 4.14, which demonstrated that phenolic foam has a low flame spread. 5.4.5.2 EPS Airlite Plastics Company manufacturers various materials for home construction projects. The company published EPS toxicity data undertaken by Underwriters Laboratories (Fox Blocks 2008). Some of the report’s key results were: • no HCFCs or CFCs emitted during manufacturing • no toxins or formaldehyde produced • low flame spread • low smoke density The report suggested that EPS emits relatively low quantities of toxins. Consequently, phenolic foam and EPS are comparable in terms of toxicity assessments. The only clear advantage that phenolic foam has over EPS as far as burn characteristics is its much greater fire resistance. 5.5 Summary A comprehensive environmental impact comparison of phenolic foam and EPS is difficult due to limited published data. However, the environmental impact of the materials can be compared in a few categories. The method of comparison used data 100 compiled in this thesis as well as in published reports. Results can be broken down into five categories: heat loss, moisture resistance, embodied energy, emissions, and toxicity. Phenolic foam has a higher R-value and conserves more energy, except after prolonged exposure to water when EPS has the advantage in these two categories. EPS initially absorbs more water than phenolic foam, but its absorption rate quickly tapers off and becomes less in the long term than phenolic foam’s absorption rate. Phenolic foam has 28% less embodied energy than EPS, the strongest indicator that phenolic foam is a better insulation material for the environment. Phenolic foam also burns much less easily. Emissions and toxicity are assumed comparable between the two materials, largely because of the difficulty in generating precise emissions estimates. In addition to environmental impact and mechanical and thermal properties, however, the various costs associated with insulating a home in northern Thailand with both phenolic foam and EPS must be considered. This data, when included with environmental impact considerations and mechanical and thermal test results, is helpful when making a final recommendation on which material is best suited as an insulation material in northern Thailand. The following chapter assesses cost. 101 Chapter 6: Costs The goal of this chapter is to compare the cost of insulating a typical house in northern Thailand with phenolic foam and EPS. Chapter 7 builds on this chapter and the prior two chapters (test results and environmental impact) to provide a final recommendation regarding which material is better suited as an insulation material. Here, raw material, transportation, labor, machinery, and electric power costs as well as energy savings are presented. 6.1 Method A scenario was created to evaluate phenolic foam versus EPS insulation costs in northern Thailand. The scenario was as follows. First, raw materials were shipped to a production site, where the final product was manufactured and then shipped to a chosen construction site in northern Thailand. Here, the product was installed in a house. Since EPS was assumed to be available at any of three large home improvement stores near the site, the only costs considered for EPS were end product retail and installation costs. It was assumed phenolic foam insulation would not be available in end product form near the site, and its costs were broken down into raw materials, production, transportation, and labor. 102 6.2 Production Site Bangkok was chosen as the production site, since many of phenolic foam’s raw materials are supplied here (see Section 6.4). The city is a major transportation and manufacturing hub for Southeast Asia, and shipping the end product from here to the construction site would be convenient and cheap. 6.3 Construction Site The chosen construction site was in Udon Thani in Thailand’s developing tropical rain belt region near the Laos border (see Figure 6.1). A railroad line runs from the city directly to Bangkok, providing a convenient method of shipping materials. Costs were estimated for phenolic foam and EPS insulation installed in the walls and roof of a one-story home in the rural outskirts of the city. Figure 6.1: Construction site chosen as a case study to estimate insulation costs http://www.lib.utexas.edu/maps/middle_east_and_asia/thailand_pol88.jpg The chosen building was a simple one-story 40’x50’ home. The walls were 8’ high, and 1” thick insulation covered 80% of four exterior walls and 100% of the roof area. Note this is identical to the building used to calculate compression loads in Section 4.2.1.1. The total insulation coverage was: • 80% x (2 walls x 40’x8’x1”x1’/12”) = 43 ft 3 (north and south exterior walls) • 80% x (2 walls x 50’x8’x1”x1’/12”) = 53 ft 3 (east and west exterior walls) • 100% x 40’x50’x1”x1’/12” = 167 ft 3 (roof) Total insulation coverage = 263 ft 3 . 103 104 6.4 Suppliers Various suppliers were researched which provide raw materials for phenolic foam manufacturing. The strategy was to find suppliers nearest to the construction site in order to reduce shipping costs. Table 6.1 lists raw materials, the various suppliers, and their proximity to the construction site. Raw material Supplier nearest construction site Location Distance to site (miles) Phenolic resin Thai GCI Resitop Co.,Ltd Rayong, Thailand 341 Surfactants Pel-stab Pelron Corp. France 5,785 Dabco Air Products and Chemicals, Inc. Pennsylvania 8,482 n-Pentane NGL Chemicals Bangkok 294 Phenolsulphonic Nanjing Yinqi Biological Engineering Nanjing, China 1,420 acid Co., Ltd. Fibers Aramid Dupont Tokyo 2,589 Glass Fabulous Fiber Bangkok 294 Cellulose Eikonik Bangkok 294 Bamboo Kongkiat Textile Co., Ltd. Bangkok 294 Table 6.1: Suppliers of phenolic foam raw materials The table shows that many of the materials would come from Asia, and several of the fibers would come from the production site in Bangkok. Dabco would be shipped from North America, since no Asian vendors were discovered. Fortunately, Dabco was only 1% of the final product mass, and so limited quantities would not inflate transportation expenses. It should be noted that if phenolic insulation becomes more common, Asian producers would likely emerge. This assumes the startup case. 105 6.5 Raw Materials Raw material costs are presented in Table 6.2 and were calculated for the one-story house as follows. Total insulation coverage = 263 ft 3 (calculated above in Section 6.3). Aramid was chosen as the reinforcing fiber, as the overall best-performing material in strength, thermal, and climate tests described in Chapter 4. Costs were based on online published data from suppliers different from those listed in Table 6.1. As most of the suppliers in Table 6.1 did not publish or make known raw material costs, it was assumed their raw materials costs would be similar to costs published by other suppliers. Bulk cost values, based on a 40,000 lb order (roughly the size of a freight container), were considered when available and are noted in the discussion. As a final note on the table values, the only available published data on bamboo fibers came from a company in New York. The Bangkok-based company in Table 6.1 would likely supply bamboo fibers at a cost considerably cheaper than this value. 106 Material Cost ($/lb) % of end product Cost/batch ($) Cost/house ($) Phenolic resin 1.25 1 87 2.07 272.75 Surfactants Pel-stab 2.25 2 1 0.04 5.64 Dabco 107.40 3 1 2.04 269.36 n-Pentane 7.95 3 4 0.60 79.75 Phenolsulphonic acid 29.63 3 4 2.25 297.25 Fibers Aramid 11.50 4 3 0.66 86.53 Glass 1.36 5 3 0.08 10.23 Cellulose 0.20 6 3 0.01 1.50 Bamboo 58.24 7 3 3.88 511.63 Total ($) 7.66 1011.28 Table 6.2: Raw materials costs for phenolic foam 1 (J Fisher 2008, pers. comm., 8 Feb.) 2 (Sabatini 1996) 3 (Sigma-Aldrich 2007) 4 (Maleniak 2001) 5 (Cripps n.d.) 6 (Marcin & Klungness 1992) 7 (Habu Textiles 2007, pers. comm., 8 Sept.) The second column represents the per pound cost of each material. Cost for phenolic resin was reduced from $1.60/lb to $1.25/lb based on a bulk order (J Fisher 2008, pers. comm., 8 Feb). The third column gives each raw material’s percentage of the final foam product. Costs were then calculated based on production of one 1.9 lb batch of foam. This value represented a typical foam block size produced in USC’s engineering lab, and figuring costs using this relatively small size was the most manageable. For example, phenolic resin made up 87% of one batch of foam, or .87 x 1.9 lb = 1.65 lb. At a cost of $1.25/lb, it cost $2.07 in phenolic resin to make one 107 batch of foam. Following this cost procedure for the remaining raw materials and adding the results, one batch of aramid fiber reinforced phenolic foam cost $7.66 in raw materials. The final column represents the cost to insulate the house walls and roof, based on 263 ft 3 of coverage. Since the density of EPS purchased from Home Depot was .95 pcf (see Table 4.1), this value became the target density of phenolic foam as well. Therefore, a 1.9 lb batch of foam at .95 pcf would yield 1.9 lb/.95 pcf = 2 ft 3 of foam. Since the total coverage was 263 ft 3 , it would take 263 ft 3 / (2ft 3 / batch) = 132 batches of foam to insulate the house. Since each batch cost $7.66 worth of raw materials, the total raw material cost to insulate the house was $7.66 x 132 batches = $1011.28. 6.6 Transportation Transportation costs for the insulation materials were broken down as follows: • cost to ship raw materials to the production site • cost to ship end product to the construction site. 6.6.1 Cost to Ship Raw Materials to the Production Site The bulk cost to ship raw materials to the production site was $0.10/lb of material, an estimate provided by Schenectedy International, Inc., the supplier of phenolic resin for this thesis (J Fisher 2008, pers. comm., 8 Feb.). Note that this cost was not 108 estimated for EPS, since it was assumed EPS was available as an end product near the construction site. Table 6.3 presents the results. Material Weight/house (lb) Cost/house ($) Phenolic resin 218.20 21.82 Surfactants Pel-stab 2.51 0.25 Dabco 2.51 0.25 n-Pentane 10.03 1.00 Phenolsulphonic acid 10.03 1.00 Fibers Aramid 7.52 0.75 Glass 7.52 0.75 Cellulose 7.52 0.75 Bamboo 7.52 0.75 Total ($) 250.80 25.08 Table 6.3: Cost to ship phenolic foam raw materials to production site The second column represents the per house weight of each raw material. The equation used in this column was: percent of each raw material within the final product x mass of one batch of foam (1.9 lb) x total number of batches per house (132). The second column represents the per house cost to ship raw material (in bulk) to the production site, based on a shipping cost of $0.10/lb. For example, the cost to ship enough phenolic resin to produce insulation for one house was 218.20 lb x $0.10/lb = $21.82. The total cost to ship enough phenolic foam raw materials to the production site for one house, based on a bulk order of 40,000 lb of material, was $25.08. 6.6.2 Cost to Ship End Product to the Construction Site The cost to ship the final insulation product to the construction site varied according to the distance from the production or supplier site to the construction site. Since 109 EPS’s supplier site (Udon Thani) was much closer to the construction site than phenolic foam’s production site (Bangkok, about 300 miles further) was to the construction site, common sense dictated that the cost to transport EPS would be cheaper than the cost to transport phenolic foam. Therefore, $0.10/lb and $0.05/lb were used as shipping costs for phenolic foam and EPS, respectively. The equation used to calculate end product shipping costs was: density of material x volume of material needed x per pound shipping cost. • Results for phenolic foam were: .95 pcf x 263 ft 3 x $0.10/lb = $25.00. • Results for EPS were: .95 pcf x 263 ft 3 x $0.05/lb = $12.50. Therefore, the cost to ship EPS insulation to the construction site was 50% cheaper than the cost to ship phenolic foam insulation. 6.7 Machinery Machinery costs only involved phenolic foam, as it was assumed EPS was already available in end product form. Costs were broken down as follows, with the assumption that production machinery consisted primarily of mixing machines, ovens, and cutting machines. In order to produce enough insulation material for one house as quickly and cost efficiently as possible, the number of machines had to be optimized. It was assumed each machine had enough working capacity to handle 25 lbs of material (or, at .95 pcf, about 26 ft 3 ). Table 6.3 shows that about 250 lb of insulation were needed to insulate one house. Therefore, 250 lb / (25 lb/machine) = 110 10 machines needed to produce enough phenolic foam insulation for one house. Estimated machinery costs were as follows: • mixer = $100,000 • oven = $100,000 • cutting machine = $50,000 It was assumed each machine could be used to help produce enough insulation product for 5,000 homes before significant repair or replacement costs were necessary. Therefore, the machinery cost for producing insulation for one house = ($100,000 + $100,000 + $50,000) / 5000 = $50. 6.8 Labor Labor costs were broken down into cost to manufacture the end product and cost to install the end product. Wages were assumed to be $1.25/hour, based on the author’s experience with working on construction projects in the region. 6.8.1 Manufacturing The cost to manufacture the end product involved only phenolic foam, since it was assumed EPS would already be available in end product form. It was assumed it would take two hours to produce enough insulation for one home, based on one hour of mixing time and one hour of baking and cutting time. Ten workers were assumed to be needed to operate the ten mixers, ovens, and cutting machines for these two hours. Therefore, the labor cost was $1.25/hour x 2 hours x 10 workers = $25.00. 111 6.8.2 Installation Installation costs for phenolic foam and EPS were assumed to be similar. Twelve workers were assumed to be the optimal number to install both materials. This was based on two workers installing insulation on each of the four walls and four workers installing insulation on the roof. It was assumed this job would take four hours. Therefore, the installation cost for both materials was $1.25/hour x 4 hours x 12 workers = $60.00. 6.9 Electric Power Costs Associated with Production Cost estimates for the energy needed to produce the insulation only involved phenolic foam, since it was assumed EPS was available in end product form near the construction site. The electricity rate for a small business at voltage greater than 22kV is 2.4649 Baht / kWh or about $0.07 / kWh (Board of Investment, Thailand 2005). It was assumed that about 1000 kWh were needed to produce enough phenolic foam insulation for one house. Therefore, the total electricity cost to produce enough phenolic foam insulation for one house is $0.07 x 1000 kWh = $70. Note: the monthly service charge is 228.18 Baht or about $6.52. Relative to the electricity rate, however, this charge was negligible and not included in energy cost calculations. 112 6.10 Lifetime Energy Savings Energy savings over product lifetime must also be considered. Table 5.2 showed how prolonged water exposure affects energy losses of phenolic foam and EPS. This data was based on the result shown in Chapter 4 that EPS’s R-value - initially lower than phenolic foam’s R-value - degrades less rapidly when exposed to water. Consequently, aramid reinforced phenolic foam insulation saves 3,774 BTUH of energy over EPS without any water exposure, but EPS saves 1,364 BTUH of energy over aramid foam after water exposure. The Board of Investment, Thailand shows that electricity costs 2.4649 (Thai baht/kWh) ~ $.000018/BTUH (2005). Assume the lifetime of EPS and phenolic foam insulation is 20 years. Therefore, the energy savings of phenolic foam over EPS without water exposure = 3,774 BTUH x $.000018/BTUH x 20 years = $11,901.69. The energy savings of EPS over phenolic foam after long term water exposure = $4,301.51. The problem with this assessment is that energy losses are most relevant in a cold climate, and northern Thailand is hot and arid. Therefore, it is assumed that neither phenolic foam nor EPS provides any cost savings when considering energy losses in this region, and these savings are not included in the overall savings calculation. However, the analysis does show the relative advantages of each material in cold climates: phenolic foam is favored in the short term without any water exposure, and EPS is favored in the long term after R-value degradation due to prolonged water exposure. 113 6.11 Summary Table 6.4 summarizes the above cost categories and energy savings. Costs ($) Phenolic Foam EPS Raw materials 1011.28 n/a Transportation (to production site) 25.00 n/a Transportation (to construction site) 25.00 12.50 Labor (manufacturing) 25.00 n/a Labor (installation) 60.00 60.00 Machinery (manufacturing) 50.00 n/a Energy (manufacturing) 70.00 n/a Product selling price 1417.54 1083.56 Total cost for one house 1502.54 1156.06 Lifetime energy savings 1 11901.69 4301.51 Overall savings 23.06% Table 6.4: Costs of providing phenolic foam and EPS insulation for a typical house in northern rural Thailand 1 Assumes cold climate. Not used in overall savings calculation. As noted throughout the chapter, several categories were not applicable to EPS since it was assumed EPS was available in end product form near the construction site. Its product selling price was based on the cost of an 8’ x 4’ x 1” thick sheet at Home Depot = $11.00. The product cost for a 40’x50’x8’ house would be $1083.56. Transportation and installation costs were added to this value to reach a final cost of $1156.06. Costs for phenolic foam insulation, on the other hand, also included raw materials and manufacturing. The product selling cost was based on the sum of the raw materials, the cost to transport raw materials to the production site, labor costs to 114 manufacture the material, and the machinery and electricity costs to produce the material. A 20% profit mark-up was included in the product selling price. Transportation costs to the construction site and installation costs were added to the product selling price to reach a final cost of $1502.54. The results show that installing EPS represented a 23% savings. However, cost and energy savings are only two factors to consider when choosing an insulation material. Strength, durability, water resistance, and other mechanical and thermal properties evaluated in previous chapters, as well as the environmental impact of each material must also be considered before making a final recommendation. The following chapter presents conclusions and offers a recommendation based on the results of the previous three chapters. 115 Chapter 7: Conclusion This thesis evaluated phenolic foam’s potential as a hybrid structural insulation material in homes in northern Thailand. Strength tests were conducted in order to assess the material’s load-bearing potential. Climatic effects on mechanical properties were considered in order to determinate whether the foam could endure the harsh weather of Thailand’s hot, humid, rainy season. The benefits of natural versus synthetic fibers were also considered. In addition, the environmental impact and costs of the material were considered to broaden the scope of the evaluation. EPS was used as a comparison insulation material. The ultimate goal of the thesis was to provide a recommendation on which material would be best suited as an insulation material in northern Thai homes. This chapter provides this recommendation by drawing conclusions from three primary areas of analysis: mechanical and thermal tests, environmental impact, and costs. Each area is evaluated separately according to which material exhibited stronger performance. The relative weights of each of these three areas are then considered before producing a final recommendation. 7.1 Mechanical and Thermal Tests Results of the mechanical and thermal tests answered the following questions: • Can phenolic foam be used as a load-bearing material? • Can phenolic foam withstand climate stresses well? • Do natural or synthetic fibers perform better? • Which fiber performs best? • Does phenolic foam or EPS perform better? Table 7.1 summarizes the mechanical and thermal results presented in Chapter 4. Table 7.1: Summary of mechanical and thermal results Before considering each of the above questions, several notes accompanying the table must be addressed. First, conclusions cannot be drawn for climatic effects on compression and shear values. The reason is that in some cases phenolic foam was not weakened by one condition, such as water absorption, but was weakened by the other condition, such as accelerated aging. Second, shear testing showed that neither phenolic foam nor EPS exhibited a clear advantage over the other material. Finally, the separation of fibers into natural versus synthetic types proved to be somewhat arbitrary, since neither type performed consistently well in mechanical and thermal tests. In other words, finding one “good” natural fiber was not offset by finding a 116 117 “bad” one. In practical terms, the good fiber would be used, the bad one would be discarded, and the attention to natural versus synthetic would be ignored. 7.1.1 Load-bearing Applications The second column of Table 7.1 suggests that load-bearing applications of phenolic foam are not practical. Strength values were low primarily because of low density values of samples; non-uniform fiber dispersion could also explain why some samples were weaker than others. Chapter 8 discusses alternate conditions under which load-bearing applications may be possible. It should also be noted that previous work has shown that phenolic foam is strong enough to act as a load- bearing member (Desai et al. 2008). Samples were fabricated at significantly higher densities than in this thesis, however, demonstrating the relation between strength and density. Fabricating higher density samples for load-bearing applications was not the exclusive end goal of this thesis. Because of the difficulty in fabricating high density materials, insulation applications became the focus. 7.1.2 Climate Stresses The third column of Table 7.1 suggests that in most cases phenolic foam did not perform well under climatic stresses or no conclusion could be drawn. One trend noted was that fibers reduce cell length, and this length remained fairly constant across climatic stress. As described in Chapter 1, the purpose of measuring cell length was to derive a correlation between cell length and water resistance: materials with little water resistance would presumably have larger cell lengths, and materials with great water resistance would presumably have smaller cell lengths. Examination 118 of Tables 4.8 (water resistance) and 4.13 (cell length) confirms a correlation between cell length and water resistance. Therefore, reduced cell length explains why water absorption values for fiber reinforced phenolic foam are less than unreinforced phenolic foam values. In conclusion, phenolic foam’s poor response to climate stresses suggests the material should not be used for exterior applications. Exceptions would be if a skin, perhaps made of plastic, were applied to the foam so that weather would not degrade the material or its R-value. 7.1.3 Natural or Synthetic Fibers Neither natural nor synthetic fibers had any conclusive advantage over the other, and the distinction proved largely arbitrary. Bamboo did demonstrate strong water resistance. However, bamboo proved difficult to fabricate in the lab above a density of 3 pcf; therefore, its strength properties could not be fairly compared with the other fibers (all at densities near 4 pcf). The creation of a stronger, denser, more water resistant bamboo fiber reinforced phenolic foam material has strong potential and is addressed in Chapter 8: Future Work. 7.1.4 Best Performing Fiber Aramid performed the best of all fibers. Its strength, water resistance, aging properties, fire resistance, and conductivity values suggest the strongest potential as an insulation material. In addition, if fabricated at a high enough density (12 pcf to 119 15 pcf), aramid fiber reinforced phenolic foam may be used effectively as a hybrid structural insulation material. 7.1.5 Phenolic Foam or EPS Both phenolic foam and EPS have mechanical advantages. EPS’s greatest advantage is that it is lightweight and would be the better choice if a building application called for a lightweight material. EPS is a better exterior insulation choice due to slower R- value degradation due to moisture ingress. EPS also showed slightly better long-term durability. Phenolic foam’s advantages are that it is stronger and has greater potential as a load-bearing material. Phenolic foam is also much more fire resistant and has better conductivity when not exposed to water. Therefore, phenolic foam should be considered when a strong, fire resistant, or highly insulative material (under low moisture exposure applications) is needed. 7.1.6 Summary In conclusion, phenolic foam cannot be used as a load-bearing material unless fabricated at a density considerably higher than those values tested in this thesis. The material does not withstand climatic stresses particularly well, and the distinction between natural and synthetic fibers proved arbitrary when considering mechanical properties. EPS is more lightweight and has slower R-value degradation than phenolic foam, whereas phenolic foam is a stronger, more fire resistant, and more insulative material when not exposed to moisture. 120 In Thailand, residents often cook in enclosed, interior kitchen spaces. Phenolic foam’s fire resistance would suit well for this application. However, summer Thai rains are intense, suggesting EPS would be a better insulation choice unless a skin layer was applied to phenolic foam. Environmental impact and costs of phenolic foam and EPS must also be examined before producing a final recommendation. 7.2 Environmental Impact Several factors were considered when evaluating the environmental impact of phenolic foam: heat loss, moisture resistance and R-value degradation, embodied energy, emissions, and toxicity. 7.2.1 Heat Loss Results from Chapter 5 showed that phenolic foam resists heat loss fairly well under dry conditions, and EPS resists heat loss better under wet conditions. These results were produced by correlating R-values calculated in Chapter 4 with heat losses calculated in HEED in Chapter 5. However, these results are most relevant in cold climates. Since Thailand’s summer climate is hot and humid and heaters are not popular in the rural north, other environmental impact categories proved more informative in recommending a material. 7.2.2 Moisture Resistance Neither phenolic foam nor EPS resisted moisture well. However, EPS demonstrated slower R-value degradation due to moisture exposure. Since R-value degradation is an important factor when considering a building’s energy losses, EPS would have a 121 better environmental impact in wet environments where energy retention within a building - such as in cold regions - was important. 7.2.3 Embodied Energy The “Sustainable Facility Guide” produced by the United States Air Force Air Combat Command and John Barry Associates Architects shows that phenolic foam has a clear advantage over EPS when embodied energy is considered (2000). Phenolic foam’s embodied energy is 28% less than EPS and comparable to or better than many other common insulation materials. 7.2.4 Emissions Predicting the emissions of foam products is a difficult task. The United States Environmental Protection Agency has shown that phenolic foam has a lower carbon content than EPS (2007). However, the organization estimates that both products are low-emissive and their emissions rates are ultimately quite comparable. 7.2.5 Toxicity No distinction based on toxicity levels appears to separate phenolic foam or EPS. 7.2.6 Summary In summary, both phenolic foam and EPS appear to have advantages when environmental impact factors are considered. Phenolic foam has a lower embodied energy, and EPS’s R-value degrades more slowly when exposed to water – an advantage in cold climates. When building homes in Thailand, choosing phenolic foam as the material with lower embodied energy should yield less environmental strain. 122 7.3 Costs EPS is about 23% cheaper than phenolic foam to use as an insulation material in northern Thailand. Phenolic foam’s raw materials and transportation costs are relatively expensive, but these expenses would likely decrease relative to EPS if phenolic foam gained a greater share in the market. If phenolic foam’s raw materials were produced in Thailand, then the cost to insulate a house here using either phenolic foam or EPS insulation would likely be similar. The more practical issue is whether insulation can be afforded by Thai residents in the first place. Unfortunately, the cost of both EPS and phenolic foam is prohibitively out of the budget of most northern Thai citizens. Significant raw material cost decreases (or GNP increases) would have to be realized before such insulation products could realistically be installed in northern Thai homes. 7.4 Recommendation The relative importance of mechanical properties, environmental impact, and cost must be considered before providing a recommendation on whether phenolic foam or EPS should be used as an insulation material in northern Thailand. The most practical category is cost. Given that Thailand’s GNP per capita is $2712, these materials are currently too expensive to be considered. Cost considerations aside, phenolic foam is recommended based on its environmental impact. When mechanical properties are considered, phenolic foam is more beneficial when fire 123 resistance and strength are important. Phenolic foam is also more insulative, but only under dry conditions. EPS’s biggest advantages are that it is more lightweight and durable than phenolic foam. Between these two materials, then, phenolic foam should be used for applications calling for a fire resistant, strong, insulative material under dry conditions, whereas EPS should be used for applications calling for a lightweight, durable, cheap, insulative material under wet conditions. 124 Chapter 8: Future Work Several areas of research can be further explored with phenolic foam. The categories can be broken down as follows: • fabrication • testing • comparison materials • environmental impact • costs 8.1 Fabrication Future experiments can explore more reliable methods of fabricating phenolic foam at a constant density. As noted in Chapter 3, creating foam samples at the same density - which allows fair comparison of mechanical properties - represented the biggest challenge of this thesis. Problems included mixing materials, dispersing fibers, regulating thermal control while mixing, and keeping the weight percentage of fibers constant. It is suggested that future work fabricate phenolic foam using a larger mixer. This would likely alleviate many of these problems and produce samples at approximately equal densities. From there, strength and climate tests would hopefully yield a more definitive separation of materials’ mechanical and thermal properties when comparing phenolic foam versus EPS or other insulation 125 materials. This thesis saw little separation of these properties in part because of fabrication difficulties. Another area to explore is fabricating phenolic foam with bamboo fibers. For example, the strength properties of bamboo fiber reinforced phenolic foam were not fully explored, since the material proved difficult to fabricate at a density above 3 pcf. Structural and other building applications with bamboo fiber reinforced phenolic foam would be good possibilities to explore. The use of other fibers may also be investigated. For example, palm fibers and flax may offer strength or water resistant advantages. Another possibility is exploring load-bearing applications of phenolic foam. Given that strength is related to density, the material should be fabricated at a high enough density (12 to 15 pcf or higher). In addition, a greater proportion of fibers should be used in the fabrication process. This thesis used 4 weight percent of fibers, and a value two or three times that would significantly increase strength. From there, a thesis project could explore using the material as a structural member within a building. 126 8.2 Testing Phenolic foam could also undergo different sorts of tests than examined in this thesis. Tests could be modeled based on a certain climate or case study. For example, northern Thailand was chosen as a case study in this thesis, and tests were chosen to measure the effects of the region’s extreme heat, humidity, and rain on phenolic foam’s mechanical properties. Instead, a colder climate could be chosen, and tests could be chosen to model the effects of cold weather on the material’s properties. R- value degradation due to moisture exposure would have more relevance here. Another possibility would be examining how well phenolic foam withstood freeze/thaw conditions. Testing could also look at hygrothermal effects, or the change in properties due to moisture absorption and temperature change. Little work has been explored on this topic, especially with phenolic foam. Oak Ridge National Laboratory performed a study two years ago on the hygrothermal performance of various cladding materials and concluded that EPS was superior over brick, concrete block, and other materials in terms of moisture resistance and thermal performance (Karagiozis 2006). Phenolic foam could be integrated into a similar study. A thesis could also look at various weather schemes based on geographical location and determine under what temperature conditions phenolic foam best resists moisture. 127 8.3 Comparison Materials Other materials could be chosen with which to compare phenolic foam. This thesis chose EPS as a common insulation material to compare its mechanical and thermal properties as well as environmental impact and costs. As described in Chapter 2, however, numerous other insulation materials exist. Structural materials could also be chosen with which to compare phenolic foam’s load-bearing properties. 8.4 Environmental Impact Assessment Limited data exists on the environmental impact of phenolic foam. For example, none of the lifecycle cost assessment software programs examined in this thesis listed phenolic foam in their database of materials. Instead, this thesis had to extract data from several sources in order to create a rough environmental impact profile. Future work could focus on creating a website or software program on the environmental impact of phenolic foam. The tool could take a more rigorous approach to evaluating the environmental impact of phenolic foam and include more comprehensive data than included here. Such a tool would fill a missing niche in the sustainable assessment of building materials. The tool could also broaden its scope to include other insulation materials, thereby creating a central database. This could be especially valuable for homeowners when considering the environmental impact of various insulation materials. 128 8.5 Cost Assessment Significant gaps exist in data showing the cost to insulate a home with phenolic foam. This is largely due to its limited use and availability, especially in the United States. As with its environmental impact, the various costs of phenolic foam could be assembled in a central database or website. Other insulation materials could be compared on the site. This could be valuable for homeowners when considering the cost savings of various insulation materials. Further work could also explore the implementation of phenolic foam in developing regions like Thailand, a question not only of building science but also economics as well as cultural understanding. How could developing regions be able to afford this material? How can costs such as raw materials and production be reduced? Various cost-benefit scenarios could be analyzed. For example, the economic and physical health benefits of installing weather resistant insulation in homes in developing regions could be weighed against economic, health and lifestyle costs associated with not having insulation. This is especially relevant when considering that several developing regions in Asia and worldwide are prone to tsunamis and flooding, the effects of which adversely affect the lives of millions of people living in sub- standard housing conditions. 129 Bibliography Alcan, Inc. 2004, Accent on innovation at Adidas village, accessed 22 January 2008, <http://www.alucobondusa.com/news_headlines_detail.asp?id=25>. Alibaba.com 2008, Alibaba manufacturer directory, accessed 4 February 2008, <http://www.alibaba.com/trade>. American Society for Testing and Materials 1998, Standard test method for water absorption of plastics, D 570, ASTM International, Pennsylvania. American Society for Testing and Materials 2000, Standard test method for shear properties of sandwich core materials, C 273, ASTM International, Pennsylvania. American Society for Testing and Materials 2004a, Standard test method for compressive properties of rigid cellular plastics, D 1621, ASTM International, Pennsylvania. American Society for Testing and Materials 2004b, Standard test method for response of rigid cellular plastics to thermal and humid aging, D 2126, ASTM International, Pennsylvania. American Society for Testing and Materials 2004c, Standard test method for steady- state thermal transmission properties by means of the heat flow meter apparatus, C 518, ASTM International, Pennsylvania. Barrie, J 2000, Sustainable Facility Guide, The Air Force Center for Engineering and the Environment, 31 March 2000, accessed 29 January 2008, <http://www.afcee.brooks.af.mil/green/case/accsfguide.pdf >. Barry, R 2001, The Construction of Buildings, Blackwell Science, Oxford. Board of Investment, Thailand 2005, Typical costs of starting a business, Runckel & Associates, accessed 5 February 2008, <http://www.business-in-asia.com/investment_costs2.html>. Building Green 1995, ‘Insulation materials: environmental comparisons’, Environmental Building News, Vol. 4, No.1, January 1995, accessed 25 January 2008, <http://www.buildinggreen.com/auth/article.cfm?fileName=040101a.xml>. 130 Composite Building Structures, Ltd. 2007, Advanced construction composites: specifications, features & benefits, accessed 7 February 2008, <http://www.cbs-homes.com/download_files/Specs_Features_Benefits.pdf>. Convertunits.com 2008, Travel distance calculator between cities and airports, accessed 5 February 2008, <http://www.convertunits.com/distance/>. Cripps, David n.d., Guide to composites, Netcomposites.com, accessed 5 February 2008, <http://www.netcomposites.com/education.asp?sequence=33>. Desai, A, Auad, M, Shen, H & Nutt, S 2008, ‘Mechanical behaviors of hybrid composite phenolic foam’, Journal of Cellular Plastics, Vol. 44, No. 1, pp.15-36. Dietrich Metal Framing Company n.d., Danback Flyer, accessed 3 February 2008, <http://www.dietrichindustries.com/library/pdf/DMF_Danback_Flyr_FA.pdf >. Earth Trends 2003, Economic indicators – Thailand, accessed 29 January 2008, <http://earthtrends.wri.org/pdf_library/country_profiles/eco_cou_764.pdf>. Elizabeth, L & Adams, C 2000, Alternative construction, John Wiley & Sons, New York. F, L 2002, Southeast Asian rainforest climate, Blue Planet Biomes, accessed 29 January 2008, <http://www.blueplanetbiomes.org/se_asian_rnfrstclimate.htm>. Fabian, B, Herrenbruck, S & Hoffee, A 2004, The environmental and societal value of extruded polystyrene foam insulation, Earth Tech Forum 2004, 15 April 2004, accessed 8 February 2008, <http://www.xpsa.com/enviro/PDFs/Fabian_Hoffee_Herrenbruck_Earthtech_ 2004.pdf>. Food and Agriculture Organization of the United Nations 2008, Major non-wood forest products of Thailand, accessed 28 January 2008, <http://www.fao.org/docrep/X5336E/x5336e0p.htm>. Fox Blocks 2008, Technical performance data - Fox Blocks ICF wall systems, Airlite Plastics Co., accessed 4 February 2008, <http://www.foxblocks.com/technical_data/Fox_Blocks_Technical_Data.pdf >. Gatz, C 1967, Curtain Wall Construction, Frederick A. Praeger, New York. 131 The Global Historical Climatology Network, Version 1 1992, Udon Thani, Thailand, WorldClimate, accessed 7 October 2007, <http://www.worldclimate.com/cgi-bin/data.pl?ref=N17E102+2100+48354 W>. Godwin, D 2003, Modeling emissions of high global warming potential gases, United States Environmental Protection Agency, accessed 5 February 2008, <http://www.epa.gov/ttn/chief/conference/ei12/green/godwin.pdf>. Greenwald, R 2008, ‘The aftermath of phenolic foam’, Professional Roofing, January 2005, accessed 28 January 2008, <http://www.professionalroofing.net/article.aspx?A_ID=572>. Grimminger, J & Muha, K 1995, ‘Silicone surfactants for pentane blown rigid foam’, Journal of Cellular Plastics, Vol. 31, No. 1, 1995, accessed 8 February 2008, <http://cel.sagepub.com/cgi/content/abstract/31/1/48>. Grzebieta, R, Al-Mahaidi, R & Wilson, J 1997, Mechanics of Structures and Materials, A.A. Balkema, Netherlands. Heggem, Arve 1996, International code for application of fire test procedures, Metal Safe Sign International, accessed 5 February 2008, <http://www.mss-int.com/msc61.html>. Instron, Inc. 2008, Glossary of materials testing terms, accessed 19 February 2008, <http://www.instron.us/wa/resourcecenter/glossaryterm.aspx?ID=25>. Kalinger, P & Drouin, M 2001, Closed cell foam insulation: resolving the issue of thermal performance, Johns Mansville Company, accessed 8 October 2007, <http://www.jm.com/roofing_systems/literature/Closedcell.pdf>. Karagiozis, A 2006, The hygrothermal performance of exterior wall systems: key points of the Oak Ridge National Laboratory NET facilities research project, Oak Ridge National Laboratory, accessed 19 February 2008, <http://www.eima.com/pdfs/EIMA_Executive_Summary_new.pdf>. Khemani, K 1997, Polymeric Foams, American Chemical Society, Washington, D.C. Klempner, D & Sendijarevic, V 2004, Polymeric Foams and Foam Technology, Hanser Publishers, Munich. 132 Lehmer, A 1997, ‘When it comes to tree-free paper, don't be bamboozled’, Earth Island Journal, Vol. 12, No. 4, Fall 1997, accessed 28 January 2008, <http://www.earthisland.org/eijournal/fall97/wr_fall97bamboozled.html>. Maleniak, Rich 2001, Aramid fibers, accessed 29 January 2008, <http://www.chem.uwec.edu/Chem405_S01/malenirf/project.html>. Marcin, T & Klungness, J 1992, ‘The economic potential for improved fiber recovery from recycled papermills’, Forest Products Journal, Vol. 42, No. 9, September 1992, accessed 4 February 2008, <http://www.fpl.fs.fed.us/documnts/pdf1992/marci92a.pdf>. Papadopoulos, A & Giama, E 2007, ‘Environmental performance evaluation of thermal insulation materials and its impact on the building’, Building and Environment, Vol. 42, No. 5, May 2007, pp.2178-2187. Paraskevopoulos, S, Borkin, H, Darvas, R & Larson, C 1963, A preliminary investigation of the potential use of foam plastics for housing in underdeveloped areas, Architectural Research Laboratory, The University of Michigan, Ann Arbor. Phien, H, Arbhabhirama, A & Sunchindah, A 1980, ‘Rainfall distribution in northeastern Thailand’, Hydrological Sciences Bulletin, Vol. 25, No. 2, June 1980, accessed 25 January 2008, <http://www.cig.ensmp.fr/~iahs/hsj/250/hysj_25_02_0167.pdf.>. Reed, D & Simiu, E 1983, Wind Loading and Strength of Cladding Glass, National Bureau of Standards Building Science Series 154, Washington, D.C. Rockwool International A/S n.d., Glossary of marine terms, accessed 8 February 2008, <http://www1.roxul.com/graphics/RX-NA/Canada/products/marine/ GlossaryofMarineTerms.pdf>. Sabatini, D, Knox, R & Harwell, J 1996, Surfactant-enhanced DNAPL remediation: surfactant selection, hydraulic efficiency, and economic factors, United States Environmental Protection Agency, accessed 4 February 2008, <http://www.epa.gov/ada/download/briefs/enhanced.pdf>. Schierle, G 2006, Structures in Architecture, University of Southern California Custom Publishing, Los Angeles. 133 Schmidt, A, Clausen, A, Jensen, A & Kamstrup, O 2003, ‘Comparative life cycle assessment of three insulation materials: stone wool, flax and paper wool’, The International Journal of Life Cycle Assessment, Vol. 8, September 2003, accessed 28 January 2008, <http://www.scientificjournals.com/sj/lca_documents/volumes>. Schmidt-Vogt, D 2000, Indigenous knowledge and the use of fallow forests in northern Thailand, European Tropical Forest Research Network, accessed 28 January 2008, <http://www.etfrn.org/etfrn/workshop/users/chapter_13.pdf>. Shen, H 2003, ‘Toughening of phenolic foam’, PhD thesis, University of Southern California. Shen, H, Lavoie, A & Nutt, S 2003, ‘Enhanced peel resistance of fiber reinforced phenolic foams’, Composites: Part A, Vol. 34, pp. 941-948. Shen, H & Nutt, S 2003, ‘Mechanical characterization of short fiber reinforced phenolic foam’, Composites: Part A, Vol. 34, pp. 899-906. Shutov, F 1986, Integral/structural polymer foams, Springer-Verlag, Berlin. Sigma-Aldrich, Co. 2007, 4-Hydroxybenzenesulfonic acid solution, accessed 4 February 2008, <https://www.sigmaaldrich.com/catalog/search/ProductDetail/FLUKA/77670 >. Sigma-Aldrich, Co. 2007, Dabco ™ 33-LV, accessed 4 February 2008, <https://www.sigmaaldrich.com/catalog/search/ProductDetail/ALDRICH/29 0734>. Simply-Thai.com 2007, Seasons in Thailand, accessed 18 January 2008, <http://www.simply-thai.com/Thailand_page_Seasons.htm>. Snyder, C, Clatty, J & Harasin, S 2002, Low-Density Polyurethane Composites, Composites 2002 Convention and Trade Show, September 2002, accessed 15 January 2008, <http://www.rimmolding.com/resources/pdf/lowdensity.pdf>. Structural Insulated Panel Association 2007, What are SIPs?, accessed 22 January 2008, <http://www.sips.org/content/about/index.cfm?pageId=7>. Umair, S 2006, ‘Environmental impacts of fiber composite materials’, Masters Thesis, Stockholm University, accessed 27 January 2008 from Lightweight Construction Applications at Sea Reports Database. 134 Underwriters Laboratories 2008, The standard for flammability of plastic materials for parts in devices and appliances, UL 94, Underwriters Laboratories, Illinois. United States Department of Energy 2008, Builders Challenge: Home Buyers, 19 February 2008, accessed 19 February 2008, <http://www1.eere.energy.gov/buildings/challenge/homebuyers.html>. United States Department of the Interior 2008, Environmental Considerations of Building Insulation, accessed 29 January 2008, <http://www.doi.gov/greening/buildings/insulation.pdf>. United States Environmental Protection Agency 2001, Cost and emission reduction analysis of HFC emissions from foams in the United States, accessed 4 February 2008, <http://www.epa.gov/highgwp/pdfs/chap9_foams.pdf>. United States Environmental Protection Agency 2007, Inventory of U.S. greenhouse gas emissions and sinks: 1990-2005, Annex 2: methodology and data for estimating CO 2 emissions from fossil fuel combustion, accessed 8 February 2008, <http://epa.gov/climatechange/emissions/downloads07/07Annex2.pdf>. Wendle, B 1976, Engineering Guide to Structural Foam, Technomic Publishing, Connecticut. 135 Appendix A: Glossary 1. climate conditioning – exposing materials to one of two environmental conditions simulating northern Thailand’s hot and humid rainy season • water immersion • accelerated aging 2. conditioning – see climate conditioning 3. EPS – expanded polystyrene, or the primary insulation material whose mechanical and thermal properties were compared with phenolic foam 4. SEM – scanning electron microscopy, or the method used to measure the cell length of phenolic foam and EPS samples 5. SIPs – structural insulated panels, or boards that serve dual structural and thermal purposes 6. fiber reinforcement – the use of two types of fibers to improve the mechanical and, possibly, thermal characteristics of phenolic foam • natural o bamboo 136 o cellulose • synthetic o glass o aramid 7. stress – one of two environmental conditions imposed on samples • water immersion • accelerated aging 8. tests – methods used to evaluate phenolic foam’s properties • environmental o water immersion o accelerated aging • mechanical o strength compression shear o cell length measurement using SEM • thermal – conductivity • other – flammability 9. underdeveloped – describing a poor, rural region, such as in areas of northern 137 Thailand, where subsistence farming and rudimentary construction methods and materials are common

Abstract (if available)

Abstract Construction methods and materials in the hot, humid, wet region of northern Thailand are often rudimentary. Shelters are left susceptible to harsh weather conditions, and weather-worn materials revealing gaping holes into buildings are common sights. Therefore, better insulation and more durable wall materials are needed in such underdeveloped areas prone to extreme weather.

Linked assets

University of Southern California Dissertations and Theses

Asset Metadata

Creator Basbagill, John Paul (author)

Core Title Fiber reinforced phenolic foam: climatic effects on mechanical properties and building applications in northern Thailand

School School of Architecture

Degree Master of Building Science

Degree Program Building Science

Publication Date 04/25/2008

Defense Date 03/26/2008

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag design of fiber reinforced phenolic foam,environmental impact,OAI-PMH Harvest,strengthening of composite materials

Place Name Thailand (countries)

Language English

Advisor Schierle, Goetz G. (committee chair), Schiler, Marc (committee member), Spiegelhalter, Thomas (committee member)

Creator Email john_basbagill@yahoo.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m1197

Unique identifier UC1287735

Identifier etd-Basbagill-20080425 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-61297 (legacy record id),usctheses-m1197 (legacy record id)

Legacy Identifier etd-Basbagill-20080425.pdf

Dmrecord 61297

Document Type Thesis

Rights Basbagill, John Paul

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu