Low Temperature Air Channel Testing of PVC

LOW TEMPERATURE AIR CHANNEL TESTING OF THERMALLY BONDED
PVC GEOMEMBRANE SEAMS

Timothy D. Stark¹
Hangseok Choi²
and Richard W. Thomas³

A technical paper submitted for review and possible publication in the Geosynthetics International Journal

November 29, 2003

---

¹ Professor of Civil and Environmental Engrg., University of Illinois, Newmark Civil Engrg. Laboratory, 205 N. Mathews Ave., Urbana, IL, 61801-2352, (217) 333-7394; E-mail: tstark@uiuc.edu

² Post-doctoral Research Associate of Civil and Environmental Engrg., University of Illinois, (217) 333-1773; E-mail: hchoi2@uiuc.edu

³ Technical Director, TRI/Environmental, Inc., 9063 Bee Caves Road, Austin, TX 78733-6201, USA, Telephone: 512-263-2101, Telefax: 512-263-2558; E-mail: Rthomas@tri-env.com.

---

ABSTRACT: The objective of this paper is development of a procedure for air channel testing of dual track thermal seams at low sheet temperatures and recommendations for reducing destructive testing of field PVC geomembrane seams. This objective is accomplished by developing relationships between seam peel strength and seam burst pressure for sheet temperatures ranging from 0.6 °C to 25.6 °C during field air channel testing. This paper refines the original correlation presented by Thomas et al. (2003) using data for low sheet temperatures and develops a polynomial equation that can be used to convert the sheet temperature during field air channel testing to the air channel pressure required to satisfy the specified seam peel strength instead of graphically finding the air channel pressure from an Arrhenius analysis. Thus, the proposed relationship and equation allow the seam peel strength to be determined from the field air channel testing without conducting destructive tests.

KEYWORDS: PVC Geomembrane, Air Channel Testing, Seams, Quality Assurance, Quality Control, Thermal Welding, Peel Strength, Burst Pressure, Low Temperature

AUTHORS: T. D. Stark, Professor of Civil and Environmental Engineering, 2217 Newmark Civil Engineering Laboratory, University of Illinois, 205 N. Mathews Ave., Urbana, IL 61801, USA, Telephone: 1/217-333-7394, Telefax: 1/217-333-9464; E-mail: tstark@uiuc.edu. H. Choi, Post-doctoral Research Associate of Civil and Environmental Engineering, B156 Newmark Civil Engineering Laboratory, University of Illinois, 205 N. Mathews Ave., Urbana, IL 61801, USA, Telephone: 1/217-333-1773, E-mail: hchoi2@uiuc.edu. R. W. Thomas, Technical Director, TRI/Environmental, Inc., 9063 Bee Caves Road, Austin, TX 78733-6201, USA, Telephone: 512-263-2101, Telefax: 512-263-2558, E-mail: Rthomas@tri-env.com.

PUBLICATION: Geosynthetics International is published by the Industrial Fabrics Association International, 1801 County Road B West, Roseville, Minnesota 55113-4061, USA, Telephone: 1/612-222-2508, Telefax: 1/612-631-9334. Geosynthetics International is registered under ISSN 1072-6349. DATES: Original manuscript received November 29, 2003

 

1 INTRODUCTION

Thermal welding has proven to be a cost-effective method of field seaming PVC geomembrane liners because PVC possesses excellent thermal welding characteristics such as a wide thermal seaming range and surface preparation/grinding is not required. The thermal welding technique allows PVC geomembranes to be constructed in cold weather when chemical seams are not applicable and utilizes prevalent QA/QC techniques. Thomas et al. (2003) show that fully automated thermal welding systems can weld PVC geomembranes as thin as 0.5 mm at temperatures as low as –8 °C. These systems allow the operator to adjust welder speed, nip-roller pressure, and welding temperature to create the best quality seam. During installation, welder speed is set according to geomembrane thickness. The welder should also be adjusted to account for large variations in ambient temperature. Depending upon the manufacturer of the welder, PVC welding temperatures vary from 315 to 480 °C. The use of thermal welding also allows common QA/QC techniques to be used for PVC geomembranes, such as air channel testing. It is important to emphasize that the amount of field seaming with PVC geomembranes is much less, as much as 80% less, than HDPE geomembranes because most of the required geomembrane seaming is performed in the factory and large panels are shipped to the site.

Thomas et al. (2003) present relationships between seam peel strength and seam burst pressure at three different sheet temperatures (i.e., 22.8, 35.0, and 46.7 °C) during field air channel testing. These relationships are used to construct a correlation between the field air channel pressure required to satisfy the specified seam peel strength of 2.6 N/mm and a range of sheet temperatures during air channel testing. The correlation is extended using an Arrhenius analysis of the test results. This correlation can be used to convert the sheet temperature during field air channel testing into the air channel pressure required to satisfy the specified seam peel strength. More importantly, the flexible nature of PVC allows the inflated air channel to be visible and thus the integrity of the seam can be investigated along the entire seam length. In addition, the air channel test is challenging the peel strength along the entire length of the seam instead of a limited seam length that is used in conventional destructive tests.

In this study, thirty-seven (37) sets of seam peel strength and seam burst pressure data are used in addition to the data presented by Thomas et al. (2003). The new data corresponds to low sheet temperatures, 0.6 °C to 25.6 °C, during field air channel testing. The data presented by Thomas et al. (2003) corresponds to sheet temperatures ranging from 22.8 °C to 46.7 °C. The main objective of this paper is development of a relationship between seam peel strength and burst pressure at low sheet temperatures. This data is also used to refine the correlation developed by Thomas et al. (2003), which relates the field air channel pressure required to satisfy the specified seam peel strength of 2.6 N/mm to sheet temperatures ranging from 0 °C to 60 °C.
 

2 THERMAL SEAM EVALUATION

To make field thermal seams, it is necessary to melt the polymer at the sheet surface using a heat source. The heat can be transferred to the sheets to be welded from hot air or a hot wedge. A hot air welder uses an air blower that blows heated air from an electrical element between the two sheets to be bonded by melting an interface strip. A hot wedge welder generates the heat energy necessary to melt the sheets at the interface by electrical elements placed directly between two sheets. Rollers are used to drive the heating machine and to apply pressure on the heated strip of the sheets.

At present, two types of PVC thermal seams are used in practice: dual track and single track seams. Both types of seams can be created with a hot air or a hot wedge and allow destructive and nondestructive testing to be carried out as soon as the seam has cooled. This rapid assessment of quality allows immediate changes to be made in the seaming process to ensure optimal productivity. This paper focuses on non-destructive air channel testing of dual track seams.

The seams used in this study were created at two different locations and using two different welders. This first location is TRI/Environmental in Austin, Texas on an asphalt subgrade. The other location is Environmental Protection, Inc. (EPI) in Mancelona, Michigan. The two welders are hot air and hot wedge. The hot air machine is a Leister Twinnie Model CH6056. The hot wedge machine is a Mini-Wedge made by Plastic Welding Technologies (formerly Columbine, Inc.). The sheet temperatures range from 10 to 38 °C during thermal welding. The 0.75 and 1.00 mm thick PVC geomembranes used in the thermal seam testing were provided by Canadian General-Tower, Ltd. of Cambridge, Ontario, Canada. Both welders have a typical, pre-set nip pressure and this was maintained throughout the seaming operation. Table 1 shows the different welding conditions used.

After eliminating test results corresponding to a film tearing bond (FTB) failure mode, twenty-five (25) data sets were selected from 9.2 m long thermal seams using a hot air welder and a hot wedge welding machine, which were created at TRI/Environmental. The exclusion of the FTB failure mode is recommended by Thomas et al. (2003) so the peel strength and burst pressure of the seams correspond to a similar failure mode and thus are comparable.

Twelve (12) data sets from the nine thermal seams constructed by EPI at low temperature conditions are also used. A hot air welder was used to thermally weld the 0.75 mm thick PVC geomembranes. The sheet temperatures range from –3.9 °C to 2.8 °C during thermal welding.

A total of 37 sets of peel strength and burst pressure data, 25 sets from the TRI/Environmental seams and 12 sets from the EPI seams, are used to develop the correlation between sheet temperature, burst pressure, and peel strength. The 37 data sets are divided into four sub-groups according to the sheet temperature at the time of the air channel test as shown in Table 1. The 37 data sets are summarized in descending order of sheet temperature during field air channel testing. Seven data sets are included in Group 1, which has sheet temperatures ranging from 24.4 °C to 25.6 °C with an average value of 25.1 °C. Seven data sets are included in Group 2, which has sheet temperatures ranging from 12.8 °C to 18.3 °C with an average value of 14.8 °C. Eleven data sets are included in Group 3, which has sheet temperatures ranging from 7.8 °C to 11.7 °C with an average value of 9.7 °C. Finally, twelve data sets are included in Group 4, which has sheet temperatures ranging from 0.6 °C to 7.2 °C with an average value of 5.3 °C.

The seams were evaluated by the standard peel test at 50 mm/min at 22.8°C (ASTM D 6392, 1999) and by an air channel test developed during this project. The air channel test is performed by sealing off one end of a seam length and pressurizing the other end with compressed air. The full procedure of the air channel test is described in Thomas et al. (2003).

Thomas et al. (2003) show that the air channel test fails the seam from the inside towards the outside of the seam whereas the peel test fails the seam from the outside towards the inside of the seam. This difference is not deemed significant because PVC seam requirements are specified in terms of peel strength and the burst pressure during air channel testing is simply being correlated to this specified parameter. The specified value for the peel strength of both 0.75 and 1.00 mm thick PVC seams according to the material specification available through the PVC Geomembrane Institute (2003) is 2.6 N/mm.
 

3 RELATIONSHIP BETWEEN SEAM PEEL STRENGTH AND BURST PRESSURE

3.1 Verification of previous relationships

Thomas et al. (2003) present relationships between seam peel strength and seem burst pressure during air channel testing for three sheet temperatures. These relationships use the hypothesis that a correlation exists between peel strength and burst pressure because both tests involve peeling apart the seam, albeit in different directions. Figure 1 shows the Thomas et al. (2003) relationships between peel strength and burst pressure for the 72 seams welded at TRI/Environmental, which exhibit a peel failure mode (i.e., non-FTB failure mode), at three sheet temperatures (i.e., 22.8, 35.0, and 46.7°C). To be useful, this relationship should be linear and should include seams that fail in identical ways, i.e., peel versus FTB. Therefore, the non-linear data points, i.e., FTB failure mode, were omitted to develop a relationship between peel strength and burst pressure (Thomas et al. 2003).

The relationship between peel strength and burst pressure can be expressed in terms of a ratio of peel strength (N/mm) to burst pressure (kPa), and the ratio is obtained from the slope of each trend line in Figure 1, using a linear regression analysis. Figure 1 shows that with an increase in sheet temperature, the ratio of peel strength to burst pressure, or slope of the trend line, increases. In other words, for a given peel strength, a lower burst pressure is expected as the sheet temperature increases.

To confirm the accuracy of the relationship presented by Thomas et al. (2003) between peel strength and burst pressure at a sheet temperature of 22.8 °C (see Figure 1(a)), this ratio between peel strength and burst pressure of 0.0108 is plotted along with the Group 1 data sets in Figure 2. Group 1 data sets have sheet temperatures ranging from 24.4 °C to 25.6 °C with an average value of 25.1 °C whereas the data from Figure 1(a) corresponds to a sheet temperature of 22.8 °C.

In general, the relationship between peel strength and burst pressure at a sheet temperature of 22.8 °C (i.e., the ratio of peel strength to burst pressure of 0.0108) is in agreement with the trend of the Group 1 data. The trend line with a slope of 0.0108 lies slightly below most data sets which have sheet temperature greater than 22.8 °C. One data point does not satisfy this trend, which is at 25 °C which plots below the line. The data shows that with a sheet temperature greater than 22.8 °C, the ratio of peel strength to burst pressure must be greater than 0.0108. This trend of higher sheet temperature resulting in a greater the ratio of peel strength to burst pressure is also observed in the relationships presented by Thomas et al. (2003) between the ratio of peel strength to burst pressure and the sheet temperature as shown in Figure 1. Thus, the accuracy of the relationship presented by Thomas et al. (2003) is reinforced.
 

3.2 Development of new relationships for low temperature

In this paper, new relationships between peel strength and burst pressure for air channel testing at temperature ranging from 0.6 °C to 18.3 °C are developed to complement the prior relationships for sheet temperatures of 22.8 °C, 35.0 °C, and 46.7 °C presented by Thomas et al. (2003). Thirty (30) data sets, which are designated Groups 2, 3, and 4 in Table 1, are used to develop the relationships for low sheet temperatures. Average sheet temperatures for Group 2, 3, and 4 data sets are 14.8 °C, 9.7 °C, and 5.3 °C, respectively, as shown in Table 1. As recommended by Thomas et al. (2003), only the peel failure mode is considered in this analysis.

The results of linear regression analyses for each data set are plotted in Figure 3. The solid line in Figure 3 represents the relationship between peel strength and burst pressure from the low temperature data and can be expressed in terms of a ratio of peel strength (N/mm) to burst pressure (kPa). The ratio is obtained from the slope of each trend line in Figures 3. The slopes are calculated as 0.0091, 0.0072, and 0.0063 for average sheet temperatures of 14.8 °C, 9.7 °C, and 5.3 °C, respectively. The trend lines show that a ratio of peel strength to burst pressure decreases with a decrease in sheet temperature during air channel testing. In other words, for a given peel strength, a greater burst pressure is expected as the sheet temperature decreases and the PVC geomembrane becomes stiffer. The ratios of peel strength to burst pressure from Figure 3 are summarized in Table 2 together with the ratios for sheet temperatures of 22.8 °C, 35.0 °C, and 46.7 °C presented by Thomas et al. (2003).

The dashed line in Figure 3 represents the expected relationship between peel strength and burst pressure from the Arrhenius analysis performed by Thomas et al. (2003). The expected slopes are 0.0083, 0.0070, and 0.0060 for sheet temperatures of 14.8 °C, 9.7 °C, and 5.3 °C, respectively. Comparing the measured and estimated slopes at each sheet temperature, the expected slope from the Arrhenius analysis underestimates the slope obtained by a linear regression analysis. The degree of difference between the two slopes is expressed as follows:

(1)

Using Equation (1), the degree of difference is calculated to be 8.8%, 2.8%, and 4.8% for sheet temperatures of 14.8 °C, 9.7 °C, and 5.3 °C, respectively. Thus, the results of the Arrhenius analysis performed by Thomas et al. (2003) do not correctly represent the measured relationship between sheet temperature, burst pressure, and peel strength and thus a new relationship is presented herein.
 

4 RELATIONSHIP BETWEEN SHEET TEMPERATURE AND REQUIRED AIR CHANNEL PRESSURE

It is proposed that the air channel test can be used as a field quality assurance/quality control test instead of destructive testing of PVC geomembrane seams. Therefore, it is necessary to develop a relationship between sheet temperature, burst pressure, and peel strength. This relationship allows field personnel to determine the air channel pressure that is required for a particular sheet temperature to ensure that the specified seam peel strength, e.g., 2.6 N/mm, is satisfied.

Table 2 shows that the ratio of peel strength to burst pressure is a function of a sheet temperature during air channel testing. Thomas et al. (2003) use the three ratios for the three sheet temperatures (i.e., 22.8 °C, 35.0 °C, and 46.7 °C) and the specified peel strength of 2.6 N/mm to calculate the minimum air channel pressure required to achieve the specified peel strength at sheet temperatures ranging from 22.8 °C to 46.7 °C. Three data points (solid circles) in Figure 4 denoted as a measured value were obtained by dividing the specified peel strength of 2.6 N/mm by the ratios at the three sheet temperatures (i.e., 22.8 °C, 35.0 °C, and 46.7 °C). These three data points are from Thomas et al. (2003).

Thomas et al. (2003) utilize Arrhenius modeling (Koerner et al. 1992, Shelton and Bright 1993) to augment these three data points and extend the applicable temperature range beyond the range of 22.8 °C to 46.7 °C used in the testing. Because it is assumed that most temperature dependent properties vary exponentially, the Arrhenius model was used to extend the measured relationship between peel strength and burst pressure to other sheet temperatures. Arrhenius modeling is typically used to determine the temperature dependence of chemical reactions, including deleterious reactions such as hydrolysis or oxidation and has been frequently used to estimate the service lifetime of geosynthetic products (Koerner et al. 1992, Shelton and Bright 1993, Risseeuw and Schmidt 1990, Salman and DiMillio 1998, Thomas 2002). The results of the Arrhenius analysis performed by Thomas et al. (2003) were used to extend the relationship between sheet temperature, burst pressure, and peel strength to sheet temperatures ranging from 0 to 22.8 °C and from 46.7 to 60.0 °C.

Considering the three measured ratios for low sheet temperatures in Table 2, three data points (open circles) are added to Figure 4, which represent the air channel pressure required to satisfy the specified peel strength of 2.6 N/mm for sheet temperatures of 14.8 °C, 9.7 °C, and 5.3 °C. Instead of performing an Arrhenius analysis, all of the six slopes in Table 2 for sheet temperatures of 5.3, 9.7, 14.8, 22.8, 35.0, and 46.7 °C are used to construct the new relationship in Figure 4 between the air channel pressure required to satisfy the specified seam peel strength of 2.6 N/mm and the sheet temperature during air channel testing. The six data points correspond to the following polynomial equation:

(2)

It is useful to express Figure 4 and Equation (2) in English units because there is still a tendency to use English units in field welding operation. Figure 5 shows the air channel pressures (lb/in²) required to satisfy the specified peel strength of 15.0 lb/in, corresponding to 2.6 N/mm in SI units, for sheet temperatures of 41.5, 49.5, 58.6, 73.0, 95.0, and 116.1 °F. The six data points correspond to the following polynomial equation in English units:

(3)

These equations can be used to convert a sheet temperature to the air channel pressure required to satisfy the specified seam peel strength instead of graphically estimating the required air channel pressure or performing an Arrhenius analysis.

Welding personnel can simply measure a sheet temperature during air channel testing, apply the required air channel pressure calculated from Equation (2) or (3) to the air channel for 30 seconds, and if the air channel maintains this pressure without peeling, it can be assumed that the seam peel strength is greater than or equal to the specified value of 2.6 N/mm (15.0 lb/in). It is proposed that this procedure can be used instead of destructive seam testing, which has the disadvantages of cutting holes in the geomembrane, patching the resulting geomembrane, and not testing 100% of the seam. The technique proposed herein evaluates 100% of the seam length and the flexible nature of a PVC geomembrane allows the inflated seam to be visually inspected over the entire length for defects. In addition, the proposed air channel test can be performed onsite regardless of the sheet temperature.
 

5 EVALUATION OF RECOMMENDED RELATIONSHIP FOR AIR CHANNEL TEST

This section evaluates the accuracy of the proposed relationship in Figure 4 between required air channel pressure to satisfy the specified seam peel strength (i.e., 2.6 N/mm) and sheet temperature during air channel testing. Thomas et al. (2003) performed the verification by predicting the burst pressure for the 72 seams created and tested at the three sheet temperatures during air channel testing (i.e., 22.8 °C, 35.0 °C, and 46.7 °C) and comparing the predicted values to the measured values. This verification utilized a pass/fail criterion to simulate typical QA/QC procedures.

In this paper, the same verification procedure is adopted for the additional three average sheet temperatures (i.e., 14.8 °C, 9.7 °C, and 5.3 °C). Table 3 summarizes the verification procedure and the number of seams that would have failed the requirement of peel strength (i.e., 2.6 N/mm) and air channel pressure. The air channel pressure required for a peel strength of 2.6 N/mm is calculated from Figure 4 and Equation (2) at each sheet temperature. For example, two welded seams out of eleven in Group 3 with an average sheet temperature of 9.7 °C fail to satisfy the requirement of peel strength of 2.6 N/mm in standard seam testing. Requirement of air channel pressure corresponding to the specified peel strength is calculated from Equation (2) to be 361.1 kPa for a sheet temperature of 9.7 °C. This required air channel pressure is compared to the actually measured burst pressures in Group 3. Four welded seams out of eleven in Group 3 fail the air channel pressure requirement (see Table 3). Thus, more seams fail the air channel pressure requirement for each of the low sheet temperatures (i.e., 14.8 °C, 9.7 °C, and 5.3 °C) than a destructive seam peel test. Therefore, the result of the air channel test is conservative because it will classify more seams as failed than the conventional peel test. It is anticipated that the extra failures were identified because the burst test challenges the entire seam and not only a limited portion of the seam.
 

6 CONCLUSIONS

The purpose of this study is to develop three relationships between seam peel strength and burst pressure for sheet temperatures of 14.8 °C, 9.7 °C and 5.3 °C during field air channel testing. With these relationships, the relationship presented by Thomas et al. (2003) between the required air channel pressure to satisfy the specified peel strength (i.e., 2.6 N/mm) and the sheet temperature during air channel testing is refined and extended to a sheet temperature of 5.3 °C. The following conclusions are based on the data and interpretation presented in this paper.

1. The ratios of peel strength to burst pressure are measured to be 0.0091, 0.0072, and 0.0063 for average sheet temperatures of 14.8 °C, 9.7 °C, and 5.3 °C, respectively. The expected ratios of peel strength to burst pressure from the Arrhenius analysis presented by Thomas et al. (2003) are 0.0083, 0.0070, and 0.0060 for these sheet temperatures. Comparing these ratios at each sheet temperature indicates that the Arrhenius analysis does not predict the measured relationship between peel strength, burst pressure, and sheet temperature at the low sheet temperatures, 5.3 °C to 14.8 °C.

2. The Arrhenius analysis presented by Thomas et al. (2003) slightly overestimates the air channel pressure required to satisfy the specified seam peel strength of 2.6 N/mm at low sheet temperatures. The error is measured to be 8.8%, 2.8%, and 4.8% for sheet temperatures of 14.8 °C, 9.7 °C, and 5.3 °C, respectively. The data presented herein is used to develop a polynomial equation to refine the relationship presented by Thomas et al. (2003) for a range of sheet temperature of 5.3 °C to 46.7 °C. The equation can be used to convert a sheet temperature during field air channel testing to the air channel pressure required to satisfy the specified seam peel strength of 2.6 N/mm instead of graphically finding the required air channel pressure or performing an Arrhenius analysis.

3. Verification of the proposed equation was performed by comparing the predicted value to the measured value along with a pass/fail criterion to simulate typical QA/QC procedures. Equal or more seams fail the requirement of air channel pressure compared to the requirement of peel strength. Therefore, the air channel test is conservative, and it will classify more seams as failed than the conventional peel test. The proposed relationship in this paper will allow field personnel to perform seam QA/QC operations without conducting destructive tests. This relationship, see Figure 4 and/or Equation (2) (and Figure 5 and/or Equation (3) in English units), allows the seam peel strength to be measured indirectly by applying air pressure to the air channel in a dual track weld, which reduces if not eliminates the need for destructive testing. This coupled with the visibility of an inflated air channel provides assurance of the integrity of the seam.
 

ACKNOWLEDGMENTS

Colorado Lining of Denver, Colorado and Environmental Protection, Inc. of Mancelona, Michigan fabricated the thermal seams used in this study. The support of these organizations and the PVC Geomembrane Institute is gratefully acknowledged.
 

REFERENCES

ASTM D 6392, 1999, “Standard Test Method for Determining the Geomembrane Seams Produced using Thermo-Fusion Methods”, American Society for Testing and Materials, West Conshohocken, Pennsylvania, USA.

Geosynthetic Research Institute (GRI), 1994, “Standard Test Method GM-6: Testing of Geomembrane Materials”, Drexel University, Philadelphia, PA, July.

Koerner, R.M., Lord, A., and Hsuan, Y.H., 1992, “Arrhenius Modeling to Predict Geosynthetic Degradation”, Geotextiles and Geomembranes, No. 11, pp. 151-183.

Koerner, R.M., 1998. Designing with Geosynthetics, 4th Edition, Prentice Hall, Upper Saddle River, New Jersey, 761 p.

PVC Geomembrane Institute (PGI), 2003, “PVC Geomembrane Material Specification 1103”, University of Illinois, Urbana, IL.

Risseeuw, P. and Schmidt, H.M., 1990, “Hydrolysis of HT Polyester Yarn in Water at Moderate Temperatures”, Proceedings of the 4^th International Conference on Geotextiles, Geomembranes, and Related Products, The Hague, Netherlands, pp. 691-696.

Rollin, A.L., and Fayoux, D., 1991, “Geomembrane Seaming Technique”, Geomembranes: Identification and Performing Testing, Rollin, A.L. and Rigo, J-M, eds., RILEM Report 4, pp. 59-79.

Salman, A., Elias, V. and DiMillio, A., 1998, “The Effect of Oxygen Pressure, Temperature and Manufacturing Processes on Laboratory Degradation of Polypropylene Geosynthetics”, Proceedings of the Sixth International Conference on Geosynthetics, Atlanta, pp. 683-690.

Shelton, W.S. and Bright, D.G., 1993, “Using the Arrhenius Equation and Rate Expressions to Predict the Long-Term Behavior of Geosynthetic Polymers”, Proceedings of the Geosynthetics ’93 Conference, Vancouver, pp. 789-802.

Thomas, R.W., 2002, “Thermal Oxidation of a Polypropylene Geotextile used in a Geosynthetic Clay Liner”, Proceedings of the International Symposium IS Nuremberg 2002, Nuremberg, Germany, pp. 87-96.

Thomas, R.W. and Stark, T.D., 2003, “Reduction of Destructive Tests for PVC Seams”, Geotechnical Fabric Report, IFAI, March, pp. 26-29.

Thomas, R.W., Stark, T.D., and Choi, H., 2003, “Air Channel Testing of Thermally Bonded PVC Geomembrane Seams”, Geosynthetics International, Vol. 10, No. 2, pp. (in press).