Crisis Management, Bob Luedeka, Barbara Weathers, J. P. Hogan, Proceedings of the Polyurethane Foam Association, May 15 & 16 1997.

The objective of this presentation is to present a how-to guide to managing business-threatening events.

Information is provided to:

1. Help establish a crisis communication policy and initiate a program
2. Help in the evaluation of an existing program Provide do-it-yourself tools for the adventurous
3. Help prevent damage

An Update on PURRC, Michael Applonia, PURRC-SPI, Proceedings of the Polyurethane Foam Association, May 15 &16, 1997.

An update on the programs of the Polyurethane Recycle and Recovery Council was presented to the Polyurethane Foam Association at their May 15, 1997 meeting. The purpose of the PURRC is to reduce polyurethane foam in landfills, and educate end users on the recyclable aspects of polyurethane foam.

Recently completed PURRC projects include the following:

1. Car seat recycling plant Grinding technologies for flexible foam
2. The use of finely ground foam in flexible foam
3. Acoustical performance of rebond in automotive
4. TVOC emissions testing for rebond

New PURRC Technical Projects are:

1. Separation of polyethylene film from packaging material
2. Rigid foam regrind into flexible foam
3. VRP joint project-foam and cloth composites for improved acoustics

Non-auto acoustical applications for rebond New Communications Projects are:

1. PURRC brochure
2. PURRC Internet Homepage
3. Increased participation in Industry Association Meetings

New, Low Fugitively Amine Catalyst for Use in Polyurethane Foam K. Kaye Robinson and Richard M. Gerken, Witco/OSI, Proceedings of the Polyurethane Foam Association, May 15 & 16, 1997.

Due to environmental concerns regarding volatile amine catalysts in the polyurethane foam industry, new product development research has centered on identifying non-fugitive (i.e. nonvolatile or reactive) products that still maintain reasonable catalytic activity. A number of reactive tertiary amines containing hydroxyl or primary/secondary amine groups have been developed as potential low fugitivity catalysts. In addition, other amine compounds have been investigated as having lower fugitivity due to their high molecular weight and low vapor pressure.

NIAX Experimental Catalyst UAX-1094 is one of these new compounds. This new catalyst achieves lower fugitivity primarily through high molecular weight and low vapor pressure, but also has the potential to reach into the foam matrix. UAX-1094 also shows reasonable catalytic activity in a number of different foam applications.

Discussed in this paper are the evaluations of UAX-1094 in several different types of polyurethane systems, with a particular focus on utility in polyester foams. The relative catalytic activity of UAX-1094 is compared to the catalytic activity of industry standard catalysts such as NIAX Catalyst A-1, NIAX Catalyst A-33, N-ethylmorpholine (NEM), N-methylmorpholine, (NMM), and N, N-dimethylbenzylamine (DMBA). Fugitivity data comparing UAX-1094 is essentially nonvolatile. It also has the catalytic activity necessary to produce good quality polyurethane foam (particularly polyester foams) at reasonable use levels.

While UAX-1094 is still an experimental product, it is in the initial stages of commercialization. It was recently placed on the Toxic Substance Control Act (TSCA) inventory, and laboratory formulation optimization studies are almost complete.

Water Based Adhesive for Flexible Polyurethane Foam Mike Magee, UPACO Adhesives, Proceedings of the Polyurethane Foam Association, May 15 & 16 1997.

Discussed in this presentation are three options that UPACO Adhesives offers to methylene chloride based products. They are: one part water based products, hot melts, and flammable solvent adhesives.

The author discusses the furniture/mattress applications, and the advantages of natural latex and synthetic neoprene water based adhesives, pressure sensitive and thermal curing hot melt adhesives, and flammable solvent adhesives. He also gives some insight into the problems, associated with these products, and the equipment needed to apply them.

Understanding Sensory Irritation Judith C. Stadler, Ph.D., Haskell Laboratory for Toxicology and Industrial Medicine, DuPont Co, Proceedings of the Polyurethane Foam Association, May 15 & 16, 1997.

This paper presents an overview of sensory irritation. The author defines sensory irritation, describes ways of predicting sensory irritation, and finally discusses the possible contribution of carpet emissions to sensory irritation. Some of the volatile chemicals commonly found in carpets are listed, and sensory irritation to some of these selected carpet chemicals are described.

Identification of BHT Yellowing on Carpet Fiber Carey Mitchell, Shaw Industries, Proceedings of the Polyurethane Foam Association, May 15 & 16 1997.

The slow migration of BHT from carpet padding has been blamed for the discoloration, which is occasionally observed in light colored carpets. This paper examines the role that BHT plays in causing yellowing on carpet fiber, and describes several ways of duplicating the problem in the laboratory. The author also looks at ways of differentiating the yellowing caused by BHT from other causes of yellowing.

Using a Burnt Gas Fumes Test Chamber equipped with a small Bunsen burner operating off natural gas inside, they were able to duplicate the yellowing observed on carpets. BHT was dissolved in methanol and the carpet fibers were immersed in the methanol solution and then dried. The loaded samples were then exposed in the Burnt Gas Fumes Chamber. The presence of BHT and BHT derived quinones on the carpet fibers were confirmed by HPLC chromatography of a methanol extract of the colored fibers. Fiber type did not have a significant effect on yellowing.

Antimicrobial Agents to Use in Flexible Polyurethane Foam, Glenn Runciman, Thomas Research Associates, Proceedings of the Polyurethane Foam Association, May 15 & 16, 1997.

Many studies on the effects of fungal growth on polyurethane materials have been published. Darby and Kaplan in their study published in 1968 showed that polyether polyurethanes were digested enzymatically by fungi. Filip in 1979 proved that polyurethane degradation by microbes takes place by a cleavage of amide and urethane groups.

Polyester polyurethanes are more readily degraded by microbial action than polyether polyurethanes because of the susceptibility of the ester group to hydrolysis which a large number of microbial enzymes (hydrolyses) catalyst.

The effects of microbial growth, such as odor formation and discoloration, are of prime importance as these effects can result in unhygienic, unaesthetic and unsaleable goods.

Several antimicrobial agents are available today for the polyurethane industry. These products, introduced during the manufacturing process, become part of the polymeric structure of the foam.

Proper selection of an antimicrobial agent, based on the requirements of the end-use product, can be of great benefit to a foam manufacturer. Well treated foam is resistant to microbial attack and is more hygienic than untreated foam. In addition, the antimicrobial properties can be marketed as a value-added feature to retailers and consumers.

Improved Prepolymers for Bonded Foam Ray Thomas and Van Delk, The Dow Chemical Co., Proceedings of the Polyurethane Foam Association, May 15 & 16, 1997.

Rebond foam manufacture is a widespread and profitable business in the United States. Most of the rebond is used for carpet pad, but some is molded for specialty applications such as transportation seating. The most popular method of making the rebond block is to press the prepolymer coated shredded foam into a mold, setting the prepolymer with steam, drying the resulting block and peeling the resulting block into appropriate lengths to be made into rolls.

Prepolymer preparation for making rebond blocks is very similar, regardless of the mechanical means used to fabricate and shape them. A diisocyanate is mixed in excess of stoiciometric requirements with a polyol to form a quasi prepolymer. Usually free isocyanate, or free NCO, of at least 10% by weight is required to make the prepolymer. Rebond producers using PMDI will often include up to 25% of a non reactive diluent in their formulation. This lowers the overall cost of the prepolymer as well as reduces viscosity. An amount of prepolymer equaling 10% to 20% of the weight of shredded foam is used to coat the foam.

This study was undertaken to further define the critical formulation parameters necessary for the manufacture of rebond binders using MDI and PMDI with focus on prepolymer viscosity and stability. The term MDI denotes lower functionality or "pure" MDI. Where PMDI refers to polymeric MDI of higher functionality.

Laboratory procedures were developed to identify relative differences in reaction rates, viscosity growth, and ultimately binder effectiveness in bonding shredded foam particles. A 12" X 12" X 1" mold was used to evaluate a prepolymers effectiveness as a binder. Moisture and temperature were held constant while functionality and acidity were varied to determine the relative differences in reaction rates for these materials.

In preparing rebond binders for carpet pad and other applications using MDI or PMDI several critical parameters can be used to lower the viscosity, moderated viscosity growth, and improve processability of the prepolymer. The final viscosity can be lowered by using lower functionality raw materials, maintaining the binder at elevated temperature, using high residual NCO, choosing the properdiluent, and protecting the binder from moisture. The viscosity can be stabilized by using catalyst or maintaining the reacting mixture at elevated temperature.

Each plant site, however, must still fit the exact parameters required to the individual equipment and product mix.

New Cleaning Mechanisms for FPF Process Equipment Tony Foglia, Cold Jet Inc., Proceedings of the Polyurethane Foam Association, May 15 & 16 1997.

Today, dry ice blasting is being used effectively in a wide array of applications. Dry ice blasting uses dry ice particles in a high velocity air flow to remove contaminants from surfaces without the added costs and inconvenience of secondary waste treatment and disposal.

One unique aspect of dry ice particles as blast media is that the particles sublimate (vaporize) upon impact with the surface. The gas which is produced expands to nearly eight hundred times the volume of the pellet in a few milliseconds in what is effectively a "micro-explosion" at the point of impact. Because of the temperature differential between the dry ice particles and the surface being treated, a phenomenon known as "fracking" or thermal shock can occur. As a material's temperature decreases, it becomes embrittled, enabling the particle impact to break-up the contaminant coating. Because the carbon dioxide is vaporized, the dry ice blasting process does not generate any secondary waste.

Dry ice blasting has been used to remove unwanted release agents from molds and dies, remove residual sugars from baking fixtures, remove or destroy biofilm build-up of listeria and salmonella and remove dried ink in the Flexographic Printing Industry. Only eight years have passed since this process was first introduced. The technology is continuing to experience rapid growth, change and advancement.

Is it Time for a Non-Halogen Fire Retardant in Flexible Foams? L. Bradford, J. Stonwell, B. Williams, Akzo Nobel Chemical, Proceedings of the Polyurethane Foam Association, May 15 & 16, 1997.

Concerns regarding a broad range of halogenated products have been raised, mostly in Europe, and primarily in regard to combustion products. While most US business is not threatened at the moment, it is important to be planning how to respond if current technology is threatened. This paper reviews some historic actions regarding regulation that have affected the polyurethane industry, and some patterns that suggest future risks that need to be addressed. For future success, plans must be made to prepare technology to meet changing needs.

Determination of the Extractability of Toluene Diisocyanate from Commercial Polyurethane Foams in Air J. M. Hugo, M. W. Spence, T. D. Lickly, The Dow Chemical Co., Proceedings of the Polyurethane Foam Association, May 15 & 16, 1997.

This study was conducted to determine if toluene diisocyanate (TDI) could be extracted from polyurethane foam into air, at the earliest possible time point at which a customer could obtain the foam from a polyurethane foam producer (3 days post production). Foam samples were manufactured, cut to the appropriate size, sealed in Tedlar Bags, and shipped to Dow Chemical for testing. Four foam samples were analyzed, each cubic foot, and indexes varied from 104 to 112. The selection of these samples was designed to reflect the range of commercially-available foams.

Analysis of the filter samples from the four extraction tests showed no detectable TDI with a detection limit of 0.3 ug/5mL acetonitrile or approximately 0.2 ppb in air based on an 8 hour sample at a flow rate of 1L/min and 56% recovery of TDI from the test chamber. (lowest recovery level observed during the chamber validation studies). Analysis of the rinses of the glass chambers after each foam study showed no removable TDI from the walls of the vessels during any of the foam experiments. The TDI detection limit was approximately 1.2ug.

As no TDI was found in air from any of the freshly produced polyurethane foams, the question now became; what would happen to TDI if it were introduced into the foam. In an attempt to answer this question, it was decided to add TDI to the foam, let the foam age for the time period that had previously determined could be the earliest possible point days (post production), and then attempt to extract the foam with air.

The results of the TDI air extraction experiments showed that the filters collected during the 3 day air extraction, after 3 day aging period, showed no detectable TDI coming from the TDI-loaded, low density, low index foam sample. The TDI detection limit was 0.3ug/sample of approximately 0.2 ppb in air based on an 8 hour sample at a flow rate of 1 L/min and 56% recovery of TDI from the test chamber (lowest recovery level observed during the chamber validation).

Further work is continuing to confirm that TDI appears to react with the foam in such a manner to be unremoveable into air. This study will use a high density, high index foam sample. The reproducibility of TDI recovery through the chamber will also be examined.

MDI Adhesive Selection and Production for Bonded Flexible Foam Trent A. Shidaker, Robert J. Lockwood, Brain Fogg, ICI Polyurethanes, Proceedings of the Polyurethane Foam Association, October 9 & 10, 1997.

Manufacturing urethane adhesive at a bonded foam production facility offers product reproductibility, cost saving, and unlimited flexibility in formulating. Bonded foam facilities should fully exploit the advantages of their current adhesive system by working closely with their polyurethane supplier in determining the optimum adhesive cooking time for their specific plant conditions.

The date shown in this paper indicates a strong correlation between adhesive conversion and physical properties. Three different isocyanates in bonded foam adhesives are investigated: Rubinate 9041 (R) (i.e., an MDI engineered for bonded Foam), PMDI, and TDI. Maximum conversion is desirable for the Rubinate 9041 adhesive to maximize tear resistance, tensile strength, elongation, compression set, and firmness. Compression set worsens with increasing conversion for PMDI adhesive due to the higher isocyanate functionality. TDI adhesives require at least 35% conversion, the TDI adhesive does not bond foam crumb due to adhesive absorption in the foam crumb, rendering the bulk of the adhesive ineffective. Two models, based on chemical engineering fundamentals, are presented to calculate adhesive temperature and conversion for specific plant conditions.

How a Furniture Manufacturer Complies with TB-133 Requirements Peter Barile, Shelby Williams, Proceedings of the Polyurethane Foam Association, October 9 & 10, 1997.

In 1984, the state of California first published a new flammability standard for public seating entitled Technical Bulletin 133. It was revised in 1988 and 1990. TB-133 differed in several ways from flammability standards then in use such as TB-117 and NFPA-701, UFAC or FAA 25.853.

First, TB-133 is not a component standard. It tests an entire chair or full size mockup. Second, TB-133 measures up to seven separate criteria simultaneously, not just one or two as in small scale tests. Failure in any one of these categories constitutes a test failure in TB-133.

Finally, TB-133 differs from most small scale tests in that the heat source is a ring of fire made up of 33 separate gas fed flames applied to the seat/back area of the test chair for 80 seconds.

The complexity of TB-133 still creates much confusion among people it affects including: interior designers, chair manufacturers, component suppliers, property owners and even some fire officials.

Shelby’s testing program resulted in two classifications and a rating system for predicting TB-133 outcomes without actual testing. Upholstery fabrics are rated from 1 (lowest risk) to 7 (quite flammable). The geometry of the chairs is also rated and extends from 1 through 5. The lowest represents a simple chair with only a small upholstered seat and back separated by about four inches or more. The rating 5 is for a fully upholstered lounge chair of sofa.

Their testing has shown that you can successfully use a high risk fabric on a low risk chair and vice versa. After extensive experience, it was determined that you can add the fabric rating to the style rating and if this sum is 7 or less, the final production chair can be expected to meet TB-133 criteria. The rating system is, however, based on the use of a flame retardant foam and a barrier.

Testing by Shelby and others has shown that a flame retardant barrier between fabric and foam is effective, and in many cases essential to passing TB-133. The Crib 5 foam with an active barrier continues to be the most effective choice in conjunction with the Shelby risk rating system.

Hazards Associated with the Storage of Flexible Polyurethane Foam in Warehouse Situation, Joe Hankins, Factory Mutual Research, Proceedings of the Polyurethane Foam Association, October 9 & 10, 1997.

The hazards associated with the bulk storage of polyurethane foam, either at foam manufacturing facilities or furniture manufacturing facilities, are discussed in this paper. The factors which have the most impact on large scale fire performance are also discussed. The Factory Mutual system is a group of insurance companies who insure industrial properties and make recommendations for protecting manufacturing facilities and minimizing risk of serious fire.

One of the fundamental properties of a material is the amount of heat it generates when burned. Wood, paper and cotton generate 6-8,000 BTU’s per pound. Plastics, polyurethane and other organic materials generate 16-20,000 BTU’s per pound. A pound of polyurethane will give off three times the heat of a pound of paper or wood.

The other property that affects fire performance is the burning rate. Burning rate is a function of the physical state of a product, i.e., whether it is expanded, as in a foam, or not. The lower the density of a product, the faster it burns. A product that melts generally produces a much more severe fire than one that chars. Flexible foam typically melts while rigid foam chars.

The heat content and the burning rate in combination produce the heat release rate of the product. This property defines the challenge that the sprinkler system is going to have to control the fire and cool the building and keep if from collapsing.

Fire retardants and flame barriers have a significant impact on the ignition characteristics of the foam, however, once the material starts to burn, the fire retardants have a much more limited effect. Fire retardants do not have any effect on the heat content of the foam, which is still about 18,000 BTU’s for polyurethane.

Fire hazards can be minimized by working with the insurance carriers to first understand the problems, and then developing the engineering standards to guard against them.

NovaFlex (TM) the New Frontier R. L. Kirschner, J. S. Pisipati, Bayer Corp., K. V. Lamb, Hennecke GmbH, Proceedings of the Polyurethane Foam Association, October 9 & 10, 1997.

The Novaflex (TM) liquid carbon dioxide blowing agent process has proven itself highly successful in the slabstock industry. NovaFlex was developed by Hennecke GmbH, a member of the global Bayer Group.

This process has demonstrated its high quality, reliability, and cost effectiveness. A wide variety of foam grades that were previously being produced with high levels of methylene chloride, can now be formulated with liquid carbon dioxide as the auxiliary blowing agent. Slabstock foam manufacturers can now produce high quality products under the newest plant safety and environmental regulations.

Bayer and Hennecke have started up more than 15 plants using the NovaFlex process in Europe and North America over the last three years. As a result, they now have the experience and the know-how needed to further advance this technology.

This paper describes some of the extensive experience and information that was developed from starting up and optimizing many different production operations. The current focus of ongoing research and development activities is also discussed.

A Modern Approach to Classical Reaction Studies of Tertiary Amine Catalysts in Flexible Polyurethane Foam E. L. Rister, Jr., F. Kohutek, Dr. R. A. Grigsby, Dr. R. Zimmerman, Huntsman Chemical, Proceedings of the Polyurethane Foam Association, October 9 & 10, 1997.

The water-isocyanate, hydroxyl-isocyanate and numerous crosslinking and side reactions that occur in the production of flexible urethane foam are extremely important. These reactions are usually investigated with emphasis upon reaction kinetics in a solvent. While these data are useful, they do not give an accurate account of the reactions taking place in flexible foam manufacture.

The reactants are not the same, nor are they present in realistic concentrations. A combination of model reaction studies, ultrasonic rate of rise, weight loss, FTIR, and mocrodielectrometry was used to study the activity of specific amines in flexible foams of various density and hardness. The activities are shown both in parts by weight and millimoles catalyst per reactive unit. Log ion viscosity was used as a measure to show the total cure per millimole catalyst. FTIR gives a real time account of the many reactions that occur from initiation to final cure.

The disappearance of hydroxyl and isocyanate groups was measured by titration with di-n-butylamine and by direct measurement with FTIR to establish Gel/Blow values of specific catalysts. Factors such as vapor pressure, amine value, spacing of active centers, molecular weight, Gel/Blow ratio, and end group analysis were considered to optimize catalysts for particular types of foam. Comparisons of hydroxyl contain tertiary amines with those containing no hydroxyl termination are compared in low density flexible foams. The authors show that due to the complexity of foam production, different amine choices are needed for specific applications.

Efficiencies of Various Formulations of Decontamination Solutions Utilizing Toluene Diisocyanate, Kathy Kiestler, ARCO Chemical Co., Proceedings of the Polyurethane Foam Association, October 9 & 10, 1997.

Traditionally, a combination of solid sorbents and liquid decontamination solutions has been utilized for toluene diisocyanate (TDI) spill remediation. The objective of this study was to focus on the liquid decontamination solution, and to specifically compare the efficiencies of various commercial and in-house formulations. To-date no such comparisons have been published. TDI reacts slowly with water to form ureas and carbon dioxide gas. Historically, users have sought to hasten the reaction by adding surfactants and alcohol to disperse the TDI, and ammonia to catalyze the reaction.

A variety of decontamination solutions were chosen for testing. Recommendations were solicited from the Analytical Subcommittee members of the International Isocyanate Institute. A commercial preparation was also tested in addition to decontamination solutions endorsed by the Polyurethane Division of the Society of the Plastics Industry, Inc.

Three of the formulations tested outperformed the others with respect to vapor suppression and reactivity. This experiment was undertaken to minimize the risk to personnel during TDI spill clean-up. Although all the solutions would have eventually reacted completely, faster decontamination is certainly more desirable. Regardless of where TDI spills occur, it is important that people are not exposed to TDI vapor. Although some solutions seemed to be quite active as evidenced by CO-2 bubbles and urea formation, their performance could not be judged by visual activity alone. The use of instrumentation to detect isocyanate concentrations in air directly above the spill is very important, given that others have suggested that isocyanate sensitivity may be the result of single high exposures.

A Shift in A-O Packages: Aromatic Amines As the Primary Heat Stabilizers for Flexible Polyurethane Foam, R. A. Calabrese, D. B. Parrish and R. A. Boccuzzi, Uniroyal Chemical Co., Inc., Proceedings of the Polyurethane Foam Association, October 9 & 10, 1997.

Recent concerns over the use of butylated hydroxytoluene (BHT) as a heat stabilizer for flexible polyurethane foam have resulted in the use of higher molecular weight, and higher cost, phenolic antioxidants as replacements. The change was prompted by concerns over the volatility of BHT, which can result in its being emitted into the environment during the foaming process. It can also slowly migrate from polyurethane foam cushions, carpet pad, or clothing padding and allegedly cause discoloration of carpets, clothing and drapery materials. These more costly alternatives are not as efficient as BHT.

This paper summarizes development work at Uniroyal Chemical Company in which the traditional ratio of high phenolic to low amine has been re-examined by way of statistically designed experiments. Laboratory flexible foams prepared from polyols containing various antioxidants are compared under thermally stressed conditions by microwave heating the freshly produced foam. Discrimination among the various antioxidant packages is determined by statistical evaluation of the color developed due to this thermal stress.

Analysis of the designed experiments has shown a number of highly effective combinations in which hindered aromatic amines comprise a larger portion of the total antioxidant package. These high amine/low phenolic antioxidant packages have the potential to give exceptional performance and markedly reduced emissions. This is done at reduced cost versus formulations, which use higher molecular weight phenolics as the primary antioxidant.

The Impact of Chain Extenders and Crosslinkers on the Polymer Morphology of Slabstock and High Resilience Foam Elastomers B. Davis, L. Latham, G. Barnes, The Dow Chemical Co., Proceedings of the Polyurethane Foam Association, October 9 & 10, 1997.

Typically chain extenders of crosslinkers are not used in polyurethane slabstock foam formulations because they tend to narrow the catalyst processing latitude to an unacceptable level for large scale production. On the other hand, diethanolamine is the crosslinker of choice for molded high resilience foams. Some of its desirable performance parameters depend on this additive.

The morphological changes induced by the incorporation of crosslinkers such as DEOA, TEOA, glycerine, or MEG as chain extenders in both slabstock and molded high resilience foam elastomer have been elucidated by the use of FTIR, DSC, DMS, SAXS. It is shown that the crosslinkers have a dramatic effect on the hard phase organization and its distribution in the soft phase. In some cases the stiffness of the polymer increases, and in others it goes through a minimum.

Lessons Learned on the Road to a Zero Risk Environment Patrick J. Coughlin, Operation Life Safety, Proceedings of the Polyurethane Foam Association, October 10, 1997.

This paper is an update on Operation Life Safety. Operation Life Safety was formed in 1983 to educate the fire service about the value of fire sprinklers. If residential fire sprinklers are installed along with smoke alarms, the probability of surviving a fire is over 90 percent better than in a home without them. Smoke alarms alone increase the probability of survival by only 47 percent. Although developed as a life safety tool, residential sprinklers also proved to significantly reduce property loss.

Patrick Coughlin presents a Fire Risk Equation, which is made up of three parts. They are:

1. The probability of a fire starting
2. The probability that the fire will grow beyond its container
3. The probability that the fire will damage property, or injure or kill

Each of the three probabilities has a primary mitigation method, which is the most effective of the three mitigation methods of education, code enforcement, and suppression. These are discussed by the author.

Determination of the Extracability of TDI from Polyurethane Foam into Air J. M. Hugo, M. W. Spence, T. D. Lickly, The Dow Chemical Company, Proceedings of the Polyurethane Foam Association, October 9 & 10, 1997.

This study was conducted to determine if toluene diisocanate (TDI) could be extracted from polyurethane foam into air, at the earliest possible time point at which a customer could obtain the foam from a polyurethane foam producer (three days post production).

The study was divided into two phases. In Phase 1, four foam samples, each produced by a different manufacturer, were prepared at high index-high density, high index-low density, low index-high density and low index-low density. The foam samples were cut to the appropriate size, sealed in Tedlar® bags, and shipped to Dow Chemical for testing. On the third day after production, each foam sample was placed into the glass test chamber and extracted using 37 degree Centigrade, 30% relative humidity (RH) air. The air was pulled through the test chamber at a flow rate of 1.0 L/min. Using a calibrated sampling pump.

In Phase 2, two of the four samples (high index-high density, and low index-low density) were "loaded" with TDI at a concentration of approximately 10 ppm. (w/w) in the foam. The samples were "aged" for three days prior to being placed in the test chamber and extracted for a three day period. The foam samples were extracted using 37 degree Centigrade, 30% relative humidity (RH) air. The air was pulled through the test chamber at a flow rate of 1.0 L/min. Using a calibrated sampling pump.

Analysis for the filter samples from the four Phase 1 foam extraction tests showed no detectable TDI with a detection limit of) .3 micro grams TDI/5ml acetonitrile (3X background) or approximately 0.12ppb in air based on an 8 hour sampling time, 1LPM showed no TDI at a concentration of 1.2 micro grams / rinse.

The Phase 2 testing indicated that:

1. Polyurethane foams could be loaded with TDI to a level of approximately 1.0 ppm (w/w) by passing air containing 50-75 ppb (v/v) TDI through them. Removal efficiency of TDI from the air stream by the foam was greater than 99.9%.
2. No detectable TDI could be extracted into air passing through polyurethane foam samples, which had been loaded with TDI to a level of approximately 1.0 ppm (w/w). The quantitation limit for TDI in the extraction air was 0.12 micro grams (0.04 and 0.02% of the theoretical load for the low and high density foams, respectively).