Library Contents

Author:

Jim Kaufmann

Chris Pastore Ph.D.

Jason Lyons Ph.D.

Sean Kroszner

Search:

Home > Library > Composite Panels from Reclaimed Textile Waste

Composite Panels from Reclaimed Textile Waste

Christopher Pastore and Alberto Morales

Novacomp

Abstract:
Textile waste is widely available from many sources. Post-consumer PET accounts for 5 billion kg per year. There is an estimated 100 million kg annually of pre-consumer PET waste fiber. Pre- and post-consumer carpet waste is reported at a level of 3 billion kg per year. In total, there is an estimated 230 billion kg of waste fiber generated annually. Currently only 33% of this waste is used for products. The transformation of textile waste into useful products will extend the life of the polymeric material from which the fibers are made. In this paper, the conversion of waste textile materials into solid panels is presented. The applications of the panels are for secondary load bearing members. Additional benefits are realized in terms of the potential for improved indoor air quality. As a potential replacment material, the panels are compared with pressed wood products. Pressed wood products are commonly used as low cost building materials in the housing industry. They are inexpensive because over two thirds of the material is the waste from sawmills. A comparison of physical and mechanical properties is made between currently available pressed wood building materials and composites made with virgin and reclaimed waste fibers. The comparison is made in terms of volatile organic compound release, moisture absorption, tensile strength, 3-point bend, and screw holding ability.

Background:
The wealth of polymeric waste material produced by the textile industry presents an exciting opportunity for low cost structural materials. As the costs of landfill increase, there is an increasing driver to find applications for these products. Additionally, EPA guidelines suggest that landfill and incineration are the two least desirable methods of processing waste. If a method can be developed to leverage on the energy investment possesed by polymeric textile fibers, a cost effective production technology can be developed. An attractive application market is the construction industry to address the replacement of high formaldehyde emitting pressed wood products. In order to address the replacement of these materials, it is necessary to develop a highly cost effective material system. Most pressed wood products are currently made of re-cycled wood chips from sawmills. A viable replacement for these secondary load bearing materials must possess adequate mechanical and physical properties, reduce undesirable emissions, and have manufacturing costs which compare favorably to the replaced material. Meeting mechanical, physical, and chemical properties can easily be realized through reinforced thermoplastic composite materials. However, the cost issue needs to be addressed. To reduce the cost of production of composite materials, it is useful to take advantage of the high volumes of thermoplastic waste. In this study, the work has focused on waste material produced in the form of short fibers. The materials under consideration are thermoplastic and glass fibers from industrial waste. Composites can offer certain other benefits due to the thermoplastic binder, reduced biological degradation and volatile organic compound release. The presented materials utilize waste material as the major content in order to reduce cost. The panels must possess mechanical properties similar to the pressed wood panels that they intend to replace.

Composite Panel Production
The production of panels was carried out through the thermoforming of blended glass/polyethylene terephthalate (PET) fibers. In particular low molecular weight PET fibers (called "low melt") were used. The composite consists of a thermoplastic matrix of polyester with glass fiber reinforcement. Panels are formed starting from fibers and/or particles. Composite materials are blended to the desired volume fractions according to design constraints. Blending is accomplished using currently available nonwoven processes. These mats can be carded after blending to increase density and provide a favorable orientation, and the resulting carded webs can be cross-lapped to provide desirable stacking sequences if needed. The appropriate number of mats are stacked together and placed in the mold to match the final thickness and density of the composite. Melting the resultant commingled glass/thermoplastic fiber web consolidates the material. The fibers used in this study typically ranged from 25 to 75 mm. The glass fibers were nominally 50 mm in length, with diameters of 10 µ m. The polyester fibers were 65 mm in length with diameters of approximately 20 µm. The waste fibers were collected from various carpet and composite producers and the glass and PET blended using traditional textile operations. The resulting material is a "web" of fibers with high bulk. This material has sufficient integrity to be manipulated, but remains highly conformable. This is a useful property for the formation of rather complex shaped parts. The composite retains shapeability after consolidation. Shaping is possible through thermoforming of the panels. The web material was placed in a press mold for densification, as illustrated schematically in Figure 1. For this study the mold was a simple matched die rectangular plate. Due to the high bulk factor associated with the preform material, the female mold part has a depth of 60 mm to accomodate the bulky fibrous web. The mold is designed to allow different thickness panels by adjusting the depth of the mold. For this study the the final densified parts had thicknesses ranging from 6-8 mm. The fibrous web (preform) is placed in the mold and the mold is placed on a hot press. The tool is heated to 265° C and the die is slowly closed to allow moisture vapor to escape. It should be noted that the melting point of the PET used in this study is approximately 200° C. For other thermoplastic polymers, the die would be heated 10-15\% above the absolute melting temperature in ° K in order to assure proper polymer flow. After the die has closed, the system is rapidly cooled to below the glass transition temperature of the thermoplastic. The composite plate is thus realized. For this study, several blend ratios of glass and PET were formed into composite plates. Because of the variations in blending operations, the actual realized density varied between plates and within plates. The panels fabricated had densities ranging from 1.37 g/cc (pure PET) to 1.73 g/cc (about 25\% V_f). This shows a relatively dense material compared to pressed wood, which has densities ranging from 0.66 - 0.81 g/cc. It was observed during fabrication that the higher blend ratio materials did not achieve the intended densities due to the difficulty in blending and compressing the material. Furthermore, the through thickness orientation of the glass fibers resists compaction, making it more difficult to realize fully dense materials. Using the current technique, optimal materials can be achieved with fiber volume fractions around 20\%, or densities around 1.65 g/cc. Higher fiber volume fractions tended to develop trapped gas pockets which are visible on the surface of the composite. Figure 2 shows some enhanced surface porosity associated with a high density specimen.

Characterization
The characterization of the composite panels was done for the purpose of comparison with the candidate pressed wood panels. The response of the composites must satisfy the current requirements associated with the current pressed wood panels.

Volatile Organic Compound Emissions
It is commonly known that many adhesive and binder materials emit volatile organic compounds (VOCs) that are hazardous to the indoor air environment initially and over time [1] . The majority of pressed wood panels are bonded using urea-formaldehyde based resins [2] , and the introduction of moisture promotes formaldehyde based resin breakdown causing the release of gaseous formaldehyde. The use of thermoplastic composite materials can alleviate these problems by eliminating the application of resins or the need for additional binders. Many thermoplastic polymers can be considered inert under normal environmental conditions. Only under extreme temperatures do these polymers begin to release organic vapors as a product of molecular chain breakdown. To evaluate the VOC release associated with these composite materials, small specimens were subject to vapor release analysis using chemical indicating Drager tubes [3] . The PET/glass materials have been tested for alcohols and aldehydes and there was no measurable concentrations of either. These tests also had cross sensitivities to styrene, ether, ketones, esters, and vinyl acetate. The test was conducted using Drager tubes over a three day period. The materials showed no measurable emissions of any of these compounds during that time.

Moisture Absorption
Moisture absorption in pressed wood causes formaldehyde release, increased weight, and significant strength loss. As such, moisture absorption is considered a principal concern. For the basis of comparison, the PET/glass panels were subject to the same evaluation procedure [4] . A sample of each panel material was submerged fully in tap water. The samples were weighed initially and then after 2, 7, and 24 hours of immersion. Table 1 describes the water retention of both wood and composite panels. This test verifies the hydrophobic nature of the thermomplastic composite material. At high volume fractions there is water absorption in the composite panels due to porosity of the surface. The porosity values were back calculated using ideal densities based on the web densities and final composite thicknesses. Residual stresses in OSB and PB resulted in significant surface deterioration of the materials when subject to moisture. Surface particles were displaced from the panels early in the moisture test. All the wood panels assumed a rough surface with the surface roughness frequency and amplitude related to the particle size. Also noticeable was a yellowing of the water color during and after the test on the wood panels, probably due to resin breakdown. The composite materials showed no noticeable changes during this test.

Mechanical Properties
Both pressed wood and composite materials were tested for tensile strength, tensile stiffness, modulus of rupture (MOR), 3-point bending stiffness, and screw holding ability. All tests were conducted in best accordance with ASTM D 3039 [5].

Tensile testing
Coupons nominally 150 mm by 25 mm were cut from the densified panels and subject to tensile loading. The gage section during testing was 50 mm and the specimens were held in the test frame using end-tabs. Load-displacement curves were measured for the load frame and the data reduced accordingly. Figure \ref{str.dens} shows the distribution of strength versus specimen density. As expected, the strength increases with increasing density values. The high variation in strength values is attributable to the variations in material densities as well as surface roughnesss. For comparison, the pressed wood materials have tensile strengths ranging from 5.3 - 16.5 MPa.

Bending test
Coupons nominally 125 mm by 25 mm were tested in a 3 point bend configuration in accordance with ASTM D 1037 . Figure \ref{mor} illustrates the variation in modulus of rupture (MOR) strength values versus density. Figure \ref{bs} shows the distribution of bending stiffness values as a function of material density. In both cases, the trends are as expected, showing increasing properties at higher density levels. Pressed wood materials have MOR strengths ranging from 2- 10 MPa and bending stiffnesses from 300 - 500 MPa.

Screw hold
The screw holding test is conducted to get an idea of the possible mechanical bonding that is possible. A screw is inserted into the board perpendicular to the surface and the force required to remove the screw from the supporting material is measured, measuring the shear strength of the material supporting the threads of the screw. The results from the screw holding test are summarized in Figure 5. Again the results are as expected - increasing holding ability with increased density.

Production
Fiber waste is widely available from many sources. Post-consumer PET accounts for 5 billion kg per year. This is reused, made into rags, and formed into webbing. There is an estimated 100 million kg annually of pre-consumer PET waste fiber. Fiberglass can be acquired from waste fiberglass composites and edge trimmings from screening. Probably there is much more waste that is undisclosed by the industry. It is possible to match and surpass the material properties of pressed wood panels and even plywood. The real task is to do this at a competitive price. The market for pressed wood products is 400 million square meters in the U.S. accounted for $1.25 billion U.S. in sales in 1993 [6] . The cost to purchase pressed wood materials ranges from $3.00 - $7.00 per square meter approximately in 1992 [7] . The cost of fiber waste material is estimated at \0.10 per kg. Considering the mass of a composite panel to provide equivalent bending strength (MOR) to a particle board, the material will have an areal density of approximately 8 kg/m^2, or a raw material cost of $0.80/m^2. Missing from this estimation is the cost of thermal processing of the fiber waste into a final, shippable form. It is clear, however, that if the composite can be processed for less than $1.00 per m^2 of panel this is potentially a cost-effective substitution in addition to the improved indoor air quality.

Discussion
The material is under consideration for the reduction of indoor air pollution by replacement of existing felons. It demonstrates relatively no VOC emissions, and virtually no moisture regain (moisture in panel material can harbor microorganisms). However, further study is needed to verify the usefulness of this material. The cost to produce these materials still needs to be addressed. The material must be made at a very low cost in order to provide a competitive advantage for large scale manufacturing. This should be obtainable through the use of waste fibers and low cost fillers. Clearly the cost of waste fiber is good, but the cost of conversion from fiber to bulk polymer for the PET fibers has not been addressed. To provide a substitute for finished wood products, a surface of paper or wood veneer can be applied to the composite without the use of adhesives. A paper surface can be painted or printed to produce a simulated wood appearance for furniture products. Other surface finishes or fillers by metals can be included to enhance the electrical, magnetic, and aesthetic properties. Texture can be applied through pressure rolling or calendaring. Flame retardancy and flammability issues need to be addressed if this material is to be considered for structural applications. The constituent materials, PET and glass, possess excellent flame retardancy, and it is expected that the final product will follow suit.

Conclusion
A composite material was fabricated using short staple PET and glass fibers. Using traditional textile processes and a hot press, relatively good materials were fabricated. The principal motivation behind this study is the development of interior secondary panels that provide the opportunity to mitigate indoor air pollution. A series of characterizations were carried out on these materials with results compared particle board, oriented strand board, and medium density fiber board. With the exception of density, all comparisons favored the composite. Strength and stiffness values showed several times higher than the pressed woods. In comparison of strength per density, the materials still perform very favorably. Notably, the composites showed no VOC emissions typical of pressed wood products, and had very little moisture absorption. Specimens were tested for strength properties after 24 hours immersion in water and showed no change. Conversely, pressed wood products typically lose more than one third of their strength after such immersions.

Acknowledgement
This research was funded in part by the Environmental Protection Agency. The authors would like to thank Mr. Patrick Duke for his experimental activities contributing to the results.

Designation	Porosity (%)	2 hours	7 hours	24 hours
OSB	40	9.1	18.2	27.3
PB	27	6.7	11.7	31.7
MDF	29	4.3	8.6	14.3
PET-0	0	0	0	0
PET-10	0.6	0	0	0
PET-15	1.3	0	0	0
PET-20	2.4	0	0	0
PET-25	5.2	0	0	0
PET-30	5.1	1.9	1.9	1.9

Figure 2. Enhanced Micrograph Showing Surface Porosity of Densified Specimen

Figure 3. Modulus of Rupture of Composite Panels with Various Densities

Figure 4. Tensile Strength of Composite Panels with Various Densities

Figure 5. Bending Stiffness of Composite Panels with Various Densities

Figure 6. Screw Hold Strength of Wood and Composite Panels

REFERENCES
[1]. Andrea Sass-Kortsak, D.~Holness, Charles Pilger, and James Nethercott. Wood dust and formaldehyde exposures in the cabinet-making industry. Am. Ind. Hyg. Assoc. J. , 47(12):747--753, December 1986.
[2]. Victor Elia and Ronald Messmer. Evaluation of methods for estimating formaldehyde released from resin containing paper and wood product dusts. Am. Ind. Hyg. Assoc. J. , 53(10):632--638, October 1992.
[3]. ASTM ASTM D-5116 Test for Organic Emissions of Indoor Materials/Products with Small Scale Environmental Chamber. American Society for Testing and Measurement, Philadelphia, PA, vol. 11-03 occupational health and safety edition.
[4]. ASTM ASTM D-1037 Evaluating Properties of Wood-based Fiber and Particle Panel Materials . American Society for Testing and Measurement, Philadelphia, PA.
[5]. Standard test for tensile properties of fiber-resin composites. Technical Report D3039, ASTM, Philadelphia, Pa., 1976.
[6]. National Particle Board Association. U.S. Annual Capacity Survey. National Particle Board Association, Gaithersburg, MD, 1993.
[7]. National Particle Board Association. U.S. Annual Shipments and Production . National Particle Board Association, Gaithersburg, MD, 1992.