PVC AND HDPE GEOMEMBRANES IN
MUNICIPAL WASTE LANDFILL LINERS AND COVERS:

THE FACTS

Presented to

MICHIGAN DEPARTMENT OF NATURAL RESOURCES

 

By

Dr. I.D. Peggs
I-CORP INTERNATIONAL
Ocean Ridge, FL
 

THE PVC GEOMEMBRANE INSTITUTE

  

1.0 INTRODUCTION

The selection and design of the geomembrane components of landfill lining and cover systems are not simple matters if optimum durability is required. A designer cannot blindly take any material of a predetermined regulated (implying satisfactory) minimum thickness "off the shelf" and apply it for all systems, on all slopes, on all subgrades, with all leachates, in all environments, and for all landfill operating procedures. If this practice is followed, and often it is, the liner is not "designed" and will probably fail before its intended service life. And there have been a significant number of geomembrane failures in all types of landfills and liquid impoundments.

There is not one universally acceptable geomembrane material because all materials have their Achilles Heel - it is simply a matter of recognizing the negative feature of each material and designing around it, to take advantage of the positive aspects of each material. This is why geomembrane lining systems need to be competently designed in the first place, and then be approved by informed regulatory design engineers. Geomembrane thickness is an integral parameter in the design process and should not be arbitrarily legislated at one value to cover all potential materials. The flexibility to properly design a given system is essential, if optimum performance is to be achieved, and if advances in materials are to be taken advantage of.

It is not my intention in this document, and in the associated meeting, to show that PVC is a better geomembrane than HDPE, and that all lining and cover systems can be built with PVC. However, it is my objective to show that PVC, in its variant forms, has a large number of very desirable properties and, therefore, is an excellent candidate material for consideration in all lining systems.

HDPE cannot arbitrarily be used in all lining systems. I am in the midst of a failure analysis of an HDPE lining system that failed after 18 months use despite there being a written guarantee from the manufacturer that the material was chemically compatible with the impoundment contents. I have a second failure that occurred after 5 years in a plant that produced the HDPE resin from which the geomembrane was manufactured to line its own waste facilities. In this latter case the resin manufacturers were not aware that their resin was inadequate for geomembrane applications.

There is a general feeling at the grass roots level of the regulatory and designing arenas that HDPE is the geomembrane of choice and will meet most waste containment performance requirements. It will not do that, and those who have had failures certainly would not agree with that feeling. I believe that the extreme confidence that the geomembrane users place in HDPE is as unjustified as the lack of confidence they place in other materials such as PVC. And this is somewhat strange since PVC has been so successfully used for waste containment for over 36 years. I would like to dispel the myths of PVC and HDPE so that they will both come out of the starting gate being considered by competent designers on their technical merits (and demerits) only - not with preconceived emotional feelings.

The regulatory agencies must play a large part in this procedure by positively requiring comprehensive design processes, and not by imposing restrictions that unintentionally discourage necessary design efforts. Certainly it may be necessary to define the "entry level", or to make the "initial cut", but this must not be done in such a way that the intended minimum criteria be perceived as the only criteria that need to be met. And the minimum criteria should not be written in such a way that, unintentionally, adequate candidate materials, both old and new, are eliminated.

In the first draft of Act 641 a single minimum thickness (60 mil) for any geomembrane used in lining and cover systems was proposed. Such a requirement effectively defined that only HDPE could be used as a geomembrane. The Department of Natural Resources is to be commended for recognizing the inadequacies of this draft and for making appropriate, professional allowance for the use of three specific materials and for leaving the door open for new geomembranes as they become available. Such a regulatory approach is like a breath of fresh air for designers and offers the potential for competent, durable lining system designs.

There are two groups of people and institutions that always have, and always will lose their battles: those who stick with the old, ignoring new developments, and those who blindly jump to the new, not fully recognizing the problems of the new. Those who remain flexible and take advantage of both will be the winners. The PVC geomembrane manufacturers are recovering from the former. Those who have had HDPE failures are suffering from the latter.

As you consider PVC and HDPE geomembranes remember that there are many types and grades of each of them. Saying "HDPE" and "PVC" is like saying "steel": there are many different types of steels (ferritic, austenitic, martensitic, and others) formulated to meet different service conditions. There are many different PVCs formulated with different additives to meet different environmental conditions. HDPEs also are different but they are different not so much by design but by default, due to the different resin types. The factors responsible for the vastly variable fundamental stress cracking resistance (mechanical durability) of HDPE are not yet fully understood, and, therefore, cannot be intentionally engineered. Just because a failure has occurred in a PVC or HDPE geomembrane does not automatically mean that all other PVC and HDPE products will fail under the same circumstances.

Except for a relatively small number of failures that have occurred due to the selection of inappropriate materials the majority of geomembrane failures have occurred due to inadequate design, poor installation workmanship, or inattentive CQA. There is no question that, to prevent failures, the emphasis must be placed on proper design and CQA. And this, most certainly, includes proper design after the correct material has been selected.

Michigan's Act 641 will certainly help achieve this goal in its second draft form.

2.0 PVC PERFORMANCE

I will attempt to dismiss some of the myths of PVC and put the performance
of HDPE into perspective. The following concerns are usually expressed about
PVC:

• It contains pinholes and therefore a lining cannot be made leak-free.

• It does not have adequate chemical resistance.
   
• It does not have adequate weathering resistance to UV and thermal radiation.

• It becomes brittle at low temperatures.

• Plasticizers leach out, or volatilize, causing unacceptable material degradation.

• The CQA associated with PVC is inadequate.

 In assessing these features one should keep in mind that the major factor is whether the material concerned continues to function as intended. The fact that some changes occur is irrelevant unless they affect the functional performance of the geomembrane. In other words, different materials may change at different rates as they age, but both could still provide in excess of the required performance. The fact that one changes more than another is irrelevant.

 2.1 Pinholes

Since a geomembrane is intended to be an impermeable barrier, pinholes are not desirable features. However, no geomembrane is absolutely impermeable, and it is doubtful that any sizeable geomembrane will be installed without some construction induced damage and a few sections of inadequate seaming that might leak. After all, the premise of the double lining system is that a single liner cannot be made leak-free. This is, however, no reason to tolerate the existence of pinholes in any geomembrane, but it may put the existence of the one or two pinholes that may exist into practical perspective. In all CQA Plans, project specifications, and regulatory aspects, it is necessary to remember that ideals are unachievable and that the real world (practical) situations require some compromise.

PVC sheeting has been made in thin gages (as low as 4 mil) for vapor barriers, swimming pool liners, and other critical medical applications for many years without pinholes. With continuous in-plant backlighting and QC techniques, geomembranes with pinholes should not appear on site. There are ASTM standards (e.g. D4451) and Federal Specifications (e.g. L-P-375C) that demand that PVC membranes in thickness greater than 10 mil contain absolutely no pinholes. Alberta Environment has used many millions of square feet of 20 mil PVC geomembrane irrigation canals and, several years ago, had a specification of 1 pinhole (maximum) per 100 ft2 of geomembrane. After thorough examination of a large amount of geomembrane they concluded that pinholes were statistically non-existent. Improved methods of calendering have effectively eliminated pinholes in PVC sheet and geomembranes. Three or four separate rolling actions ensure there are no holes.

Alberta Environment did find a minor problem with gel defects (cold roll) under startup conditions, but discussions with the manufacturers prevented such material being delivered to the site. Similar (gel) problems can occur in HDPE when agglomerates of decomposed HDPE find their way through the extruder into the geomembrane. They can sometimes be identified as bumps, or solid blisters on the surface of the geomembrane. They may also be confused with carbon black agglomerates when making carbon dispersion measurements, if the carbon black is not properly dispersed. Agglomerates of carbon black have been seen to initiate stress cracking.

The US Bureau of Reclamation has had similar experience with (the lack of) pinholes in PVC geomembrane. Many years ago (1960's) they experienced a few pinholes in 10 mil material, but after moving to 20 mil geomembrane, found it was unnecessary to have a pinhole specification. They have done a large amount of laboratory testing with 20 mil PVC, such as hydrostatic burst and puncture testing, and have never had a premature test failure due to pinholes. They have also used 20 mil PVC, in 5 x 10 ft panels, to provide critical water proofing in other laboratory tests, without premature failures.

It would, however, be foolish to state that PVC geomembrane contains absolutely no pinholes since one cannot be absolute. But the numbers that may escape manufacturing and fabrication QC procedures are very small, and are effectively zero when compared to other construction damage that can produce leakage.

It would be even more foolish to state that PVC contains pinholes and HDPE does not. In my last visit to an HDPE geomembrane manufacturing plant I saw holes in geomembrane from a few millimeters diameter to holes one could put one's fist through. In a recent CQA project I.CORP found a 60 ft length of 60 mil HDPE geomembrane that contained about 10 holes up to 80 mils in diameter.

Both HDPE and PVC geomembranes can, therefore, appear in the field with holes. However, with appropriate specifications, and good in-plant QC and field CQA, holes in both PVC and HDPE can be effectively eliminated, or be caught and repaired.

2.2 Chemical Resistance

For landfill applications the chemical resistance of the proposed geomembrane to the leachate is assessed by EPA Method 9090 "Compatibility Test for Waste and Membrane Liners". In this test the geomembrane is exposed to leachate at 23 and 50°C for 120 days. Changes in properties are measured every 30 days. If there is no continuing degradation trend, or if changes that reach an equilibrium condition within certain limits occur, the geomembrane is considered to be compatible with the leachate.

For practical purposes it is immaterial whether one of two types of the same material, or one of two different materials, degrades more than the other if both still, and will continue to, adequately meet the project specifications.

There is no question that for any given chemical containment requirement there is more chance that HDPE (but not all HDPE's), rather than PVC, will provide the better chemical resistance. However, for most municipal waste leachate applications it is probable that, although HDPE may be technically more resistant, PVC may provide more than adequate resistance for the required service. The historical performance of PVC geomembrane in Michigan and elsewhere has proven this. It could be an unreasonable trade off to use the fundamentally more chemically resistant material if it is more expensive, and with poor design, poor installation practice, and poor CQA could lead to other durability problems. Most installers surveyed (Appendix A) for this project feel that PVC is much easier to install correctly than HDPE. A European contact stated, appropriately and realistically: "PVC is not suitable for ALL chemical contacts, but PVC is one of the best materials for the ratio price/properties, mechanical properties, weldability, and permanent elongation."

Three laboratories that perform EPA Method 9090 testing were surveyed for this project: GeoSyntec Consultants, Precision Laboratories, and Texas Research Institute. All have performed, or are presently performing, tests on PVC in hazardous and municipal leachates. None of the PVC geomembranes has failed the test. One of the tests in hazardous leachate has been performed at 85°C and, even though there has been some loss of plasticizer, the PVC geomembrane successfully passed the test. This confirms that it is extremely important to recognize that, even though some changes in properties occur the material may still provide adequate service. Two of the laboratories commented that if owners and regulatory agencies were concerned about the chemical resistance of PVC, a move to Oil Resistant grades of PVC should more than put those concerns to rest.

Out of a total of approximately 55 EPA Method 9090 tests recently performed by all three laboratories on PVC, only two displayed problems; with solvent seams. However, not all solvent seams were problematic in EPA Method 9090 testing. None of the dielectric seams tested gave problems. It Is clear, that to obtain the optimum combination of material and performance, it is essential to experimentally assess the leachate resistance of each material and its seams. Semi-crystalline materials, such as HDPE, should, in addition, be tested under stress in the leachate to assess their stress cracking resistances.

Chemical resistance work presently underway in France, in which PVC, HDPE, and other geomembranes have been exposed to an MSW leachate for 1 year, has shown that both HDPE and PVC have suffered loss of additives from only a 100 m m surface layer. (Artieres, Sardinia 91 Landfill Conference).

The second draft of Act 641 specifies that resistance to leachate shall be assessed by performing the EPA Method 9090 test. On this basis, of the 55 tests performed by the three laboratories surveyed, PVC geomembranes were appropriate for containing 53 of the leachates. PVC is unquestionably a stable material for containing landfill leachates, as has been proven in Michigan.

I have investigated three HDPE liner failures in which the standard chemical resistance tables indicate that the HDPE is compatible with the contents of the ponds (nitric acid and black liquor). In one case the liner manufacturer had provided a written guarantee that the HDPE would be resistant to the black liquor for 10 years. The black liquor pond failed after 12 to 18 months by environmental stress cracking. The nitric acid pond failed after 9 months, also by environmental stress cracking. In the latter case there was even sufficient residual stress in the extruded fillet seam bead to initiate stress cracking in the bead, independently of the service stress on the geomembrane.

Two of the laboratories commented that municipal waste leachates are getting weaker, as the wastes placed in landfills become more controlled. The third agreed in principal, but identified two municipal leachates they have tested as containing "bad actors". In the recently promulgated Part 258 of 40 CFR "Criteria for Classification of Solid Waste Disposal Facilities and Practice" (colloquially known as Subtitle D), there are indications (pp 24 and 25) that there is little difference in the toxic constituents of leachates generated in true municipal waste landfills (built since 1980) and those operated prior to 1980 that contained industrial wastes in addition to municipal wastes. It is, therefore, possible that in Michigan the municipal landfills that accept industrial wastes may not have significantly more obnoxious leachates than those that accept municipal waste only. Such co-disposal of municipal and industrial wastes can be used to advantage since there is evidence that it can be done in such a way as to promote the degradation of the municipal waste (Sardinia '91 Landfill Conference).

There is a large amount of evidence, not only from EPA Method 9090 testing, but also from field experience, that PVC is satisfactorily containing municipal waste leachates. There are more than approximately 30 PVC lined landfills in Michigan that are performing satisfactorily. The few problems that have occurred are, as in most HDPE failures, related to inadequate design, poor installation, and/or poor CQA. Samples of PVC geomembrane removed from the sump of Lycoming County, PA, landfill after exposure to leachate for 11 years are still very flexible and show no visible signs of degradation.

In summary, there is no justification for taking the following two approaches to geomembrane chemical resistance; to dismiss PVC out of hand saying it does not have adequate leachate resistance, and; to assume that HDPE will perform adequately without appropriately testing it, even though the manufacturer may provide a written guarantee.

2.3 Weathering and Thermal Resistance

This is the most confusing area of concern with PVC, and may relate to our need to understand that there are different grades or PVC, since we, in North America, are adamant that PVC should not be left exposed to the elements. Yet, in Europe, PVC is quite regularly left exposed on critical installations, such as the upstream faces of hydroelectric dams. In one installation PVC has provided excellent service for more than 12 years at an elevation of over 6000 ft. In Sicily an exposed PVC roofing membrane has provided a watertight seal at a chemical plant for over 18 years.

The US Bureau of Reclamation has investigated 10 mil thick PVC geomembrane that has been installed on irrigation canal side slopes for up to 27 years, and while it has lost some plasticizer, there is still sufficient plasticizer remaining for the geomembrane to have adequate ductility and flexibility for continued service. Once again, the important factor is that material still performs its intended function despite the fact that it has aged. Plasticizers are added in sufficient quantities for the intended service, recognizing that some might be lost in service. Similarly, antioxidant packages are added to HDPE to prevent damaging thermal oxidation during extrusion, seaming, repairing, and service. Just as PVC can lose its plasticizer during aging, HDPE can degrade by consumption of antioxidant as it experiences temperatures as high as 80^oC or more in summer sunshine. In other words it is expected that these additives will be lost during different phases of service, but that fact alone is immaterial provided sufficient additive remains.

It has been reported that 40 mil PVC has successfully survived one year EMMAQUA (standard sunlight exposure) testing and that 60 mil HDPE has failed this test. On the other hand it is probable that there will be HDPEs that would survive this test and PVCs that would not. In fact, it has been stated by the US Bureau of Reclamation that a one year EMMAQUA test "may be too long, resulting in accelerated weathering conditions too severe for some materials". All PVCs and all HDPEs are not created equal. Therefore, it is not a matter of simply stating that PVC is better than HDPE (or vice versa) for a specific use; one should also consider that one grade of HDPE may be better than another, or one type of PVC may be better than another.

PVC geomembranes are now being successfully heat seamed in the US. They have been successfully heat seamed for many years in Europe. If the geomembrane can tolerate seaming temperatures without degrading in service at the seams, it can tolerate exposure to sunlight for extended service periods, as proven on dams, roots, and canal banks.

For landfill uses most weathering/thermal problems are eliminated by the regulatory requirement for covering the liner with various types of soil and other geosynthetic layers. Such layers protect the geomembrane from ultraviolet and severe thermal effects. Act 641 requires that geomembranes be covered by soil within 30 days.

Concern has been expressed about the potential problems of a landfill PVC cover material being exposed to weathering when overlying soils are eroded. This should be considered to be a problem with the design of the cover system rather than a potential problem with the geomembrane. It is, surely, unacceptable for any soil cover, on any landfill cover system, to erode to expose the geomembrane, whatever the geomembrane material. Proper separation, filtration, and drainage layers (probably geosynthetics) as required in EPA recommended designs will prevent soil erosion. Act 641 requires that erosion of soil in the cover system shall be prevented.

2.4 Low Temperature Brittleness

PVC canal liners with an Alberta Environment brittleness temperature specification of -20°C (ASTM D1790) have been successfully installed in Canadian winters. When deployed, a covering layer of stones is dumped onto the geomembrane from a conveyor belt. This is a severe cold impact test. National Sanitation Foundation International Standard 54 "Flexible Membrane Liners" specifies a maximum brittleness temperature of -29°C for the geomembrane. This is even more stringent than the Canadian specifications which have proven satisfactory in harsh service environments. Since most landfill CQA Plans require that geomembrane shall not be seamed (thereby meaning "installed") below 5°C, PVC should be able to withstand most installation environments. Once installed, the soil cover will protect the geomembrane from extremely low temperatures. Even if the temperature does reach the -20°C range under the soil, a PVC geomembrane would continue to provide adequate protection. If the geomembrane is only subject to static loading it will be protective to much lower temperatures than -30 ^o C, since the brittleness temperature is determined by an impact procedure (ASTM D1790). The measured brittleness temperature of HDPE appears to be considerably lower than that for PVC, but once again, is this really of practical significance? It will depend on the particular installation. (It should also be noted that PVC is tested by impacting a bent strip, while HDPE is tested by striking a single thickness cantilevered specimen. The PVC test is far more severe and will, therefore, show a higher brittleness temperature.) If PVC has other favorable properties, but for some project specific reason must be installed when it is cold, it is only necessary to take a little extra care during its installation. This is no different to the extra care that is required when installing HDPE to minimize its potential for stress cracking in service.

A majority of designers and installers surveyed (Appendix A) for this project agree that PVC has better mechanical property characteristics and is easier to install than HDPE. HDPE is a problem because of the yield point in its stress/strain curve that occurs at approximately 12% strain. This point of instability is of major concern to the designer. When allowing for biaxial stress conditions, as occur in geomembranes in the field, and low temperatures, it is necessary to design for maximum strains in the order of 2 or 3%. Since PVC does not have a yield point the designer can make use of strains to the breaking point - in excess of 300% at room temperature under uniaxial conditions. The level of comfort gained by the designer when using a material that has a steadily changing stress/strain curve is significant; the designer is able to give a facility owner a much safer installation, or one with a lower factor of safety and, therefore, a wider range of safe operating conditions. However, once again, mechanical properties are not the only factors that should be considered in liner/cover design. If the subgrade is not subject to settlement, tensile properties may be of little significance. If slopes are steep, and friction angles are high on one side of the geomembrane and low on the other, mechanical properties could be extremely important.

2.5 Plasticizer Leaching and Volatilization

This has been dealt with partially in Section 2.2. Plasticizer loss does occur, but in most cases has been shown to be of no practical consequence. Similarly the consumption of antioxidants in HDPE occurs and has been shown to be of little practical consequence. When the fundamental chemical resistance of the geomembrane to the leachate is correctly assessed, or materials with the correct additive packages are exposed to UV/thermal radiation, both PVC and HDPE will provide adequate durability in landfill liner/cover applications.

The Bureau of Reclamation has shown that 30% of the plasticizer in a 10 mil PVC geomembrane may be lost within 4 years, but that after 19 years the geomembrane still contains over 50% of its plasticizer. Such material is still very flexible. As with most processes, the rate of loss of plasticizer decreases with time.

 A number of studies have shown that even if 75% of the plasticizer is removed from PVC causing the geomembrane strength, and puncture and tear resistances, to increase and the ductility to decrease, the elongation at break may still exceed 100%. This value is still more than 10 times greater than the useful strain allowable in HDPE. In HDPE, once the yield strain has been exceeded the material will continue to elongate at a stress lower than the yield stress. For design purposes HDPE has failed at the yield point.

Plasticizer has been claimed as being a desirable food to rodents, which there by chew holes in PVC liners. Where rodents have consumed PVC in the field they have only done it to gain access to the warmth behind the liner, not because they find it good to eat. A survey of users identified only one positive case where this has occurred. On the other hand, HDPE is claimed to be inedible by rodents. However, at the Somerset Generating Station of New York State Electric and Gas there is ample evidence, in the form of 1-in. diameter holes, that vole-type rodents find HDPE extremely edible. Both these cases may be the exceptions that prove the rule, but they also identify the danger of generic statements and the resultant possibility of avoiding perfectly adequate materials based on misstatements.

2.6 Quality Control and Construction Quality Assurance.

There is a feeling that the qualities of HDPE geomembranes and geomembrane seams are higher than those of PVC since the HDPE industry has more visible QC documentation, and CQA documentation is more extensive. Comprehensive CQA is absolutely essential for HDPE, particularly in locations such as Michigan, where it can be very cold, since its service performance is critically dependent upon proper installation - it is not as forgiving a material as PVC. HDPE's window of seaming parameters is narrow, allowance for expansion/contraction and wrinkling must be provided, and the conditions that produce stress cracking must be avoided. It is not surprising that more CQA attention is paid to HDPE.

Many owners of facilities have demanded detailed CQA Plans for HDPE, but only recently have requested that similar CQA Plans be prepared for PVC installations. The manufacturers and installers of PVC geomembrane, at their own discretion, have also recently generated comprehensive QC documents and CQC Plans.

The past absence of CQA Plans for PVC has unquestionably been a function of most designers' experiences with PVC, and the fact that PVC has generally performed satisfactorily without major installation controls. The survey (Appendix A) of nationally recognized designers and geomembrane installers (those who regularly install PVC and HDPE) in the USA and Canada elicited the fact that most felt most comfortable working with PVC, then VLDPE, and finally HDPE.

A number of significant comments returned with the survey were as follows:

• "Some HDPE resins are a (expletive) to weld and some are easier, the whole gamut of seamability".
   

• "PVC is hard to generalize, there are so many types available".

• "Although the mechanical and temperature characteristics of HDPE are less than ideal, its chemical inertness and weldability make it the material of choice".

• "PVC has the best overall mechanical properties".
   

• "I feel most comfortable with all the types of geomembranes and all design aspects, except repairability".

• "PVC is low tech seaming high performance product. HDPE is high tech seaming".

• "Eliminate chemical resistance and long term stability and HDPE doesn't have much going for it".

It is clear that neither PVC nor HDPE is the better geomembrane in all applications. It is also clear that a selection must be made on a project specific basis, with due allowance being made for the less desirable features of each material. Experienced designers will take each material on its own merits and not pre-select it based on generalities.

With the promulgation of Subtitle D, and the development of Michigan's Act 641, all geomembrane installations will require comprehensive CQA programs, thereby putting HDPE, PVC, and other materials on an equal footing. However, it will still be necessary to ensure that the correct information is requested in the CQA Plan, and that each Plan is customized to account for the potentially problematic parameters for each material, whether that material is PVC or HDPE.

2.7 Seams

All geomembranes require some field seaming. It is universally recognized that field prepared seams are potentially the most problematic feature of lining systems.

The width of HDPE geomembrane rolls is steadily increasing in order to minimize the number of field seams required. PVC, on the other hand, is seamed under controlled conditions in the fabrication plant to produce larger panels, thereby reducing the number of field seams required. In a given liner area the length of field seam required in PVC may be 20% of that required in HDPE. It is, therefore, a matter of rationalizing whether, for any specific project, it is better to have fewer field seams with the potential problems that PVC might have, or more seams with the problems that HDPE might have. This is a decision only the designer can make. Such a point was concluded by the Bureau of Reclamation in their study on the chemical exposure and weathering of FML field seams, (EPA/600/S2-87/015): "Generic-type material specifications are not sufficient to ensure satisfactory performance of FML seams when used for hazardous waste containment applications".

2.8 Thickness

For many years EPA has recognized, as now has Michigan DNR, that a single minimum thickness is not an adequate criterion for geomembranes used in landfill liner and cover applications. All materials are not equal, and cannot so simply be reduced to a single common denominator. Due to its semicrystalline nature HDPE is a different breed of geomembrane, and within its ranks are many sub-species. Even if a single thickness requirement is based on HDPE there are many types of HDPE geomembrane that would not perform satisfactorily even at twice a minimum thickness of 60 mil. At this moment I am examining a stress cracking failure that occurred, after five years service, in an HDPE liner 100 mil thick that was not exposed to low temperatures. In such instances thickness has absolutely no influence on the durability of the geomembrane - other parameters must be considered.

Other parameters must be considered for every type of traditional and novel geomembrane. A single thickness value, such as 60 mil, is selected based on HDPE, but where realistic technical considerations have been given to other materials, such as PVC and Hypalon, thicknesses in the range of 30 mil are considered adequate. In Subtitle D (p32) a composite bottom liner is required to have a primary geomembrane with a minimum thickness of 30 mil. If the geomembrane is HDPE it must be at least 60 mil thick. A higher thickness for HDPE is required to make allowance for its problematic features; it is extremely difficult to seam at thicknesses less than 40 mil, and the grinding required on preparation for fillet extrusion seaming produces reduced thickness in the notch sensitive area at the edge of the seam. In fact, with a tolerance of +10% (NSF 54) on HDPE geomembrane thickness and the supposed maximum grinding depth (EPA/530/SW-91/051) of 10% of thickness, 60 mil HDPE geomembrane could be almost 40 mil thick adjacent to extruded seams, patches, and penetrations. EPA has stated (EPA/530/SW-91/054): "the design engineer should recognize that some geomembrane materials may require greater thicknesses to prevent failure or to accommodate unique seaming requirements." The norm is the 30 mil figure. HDPE is the exception that requires additional thickness. PVC and the other geomembranes should not be penalized because of HDPE's perceived deficiencies.

One disadvantage, therefore, of specifying a minimum thickness is that it may eliminate excellent candidate geomembranes. Even the best available geomembrane, that would perform adequately at lower thicknesses, may be eliminated on the basis of cost at higher thicknesses. The intent of regulations is to provide safer waste containment, not simply more expensive waste containment. And, as we have seen, a minimum thickness may not be functional, even if that thickness is tailored for the specific material selected. The second disadvantage of specifying a minimum thickness, whether or not there are one or two values to suit different materials, is the danger that some "designers" will read it as a specification, i.e. a geomembrane of the minimum thickness will perform the job, and no other factors need be considered.

The same sentiment was expressed by Bob Landreth, chief of EPA's Landfill Technology Section in a letter dated 29 September 1989 to the Corps of Engineers in which he stated: 

Thickness of materials should be a function of design which implies site specific information and considerations. Although other thicknesses, 30 and 60 mils, are allowed (page v), this approach, we believe, ties the hands of the designer and will force the use of generic designs and could lead to increases in overall project costs. We also strongly believe and as part of our recommendations to consultants that a minimum thickness of material type should be specified then let the consultants "design" the system. Our recommendations based on seamability, punctureabilty and installability is as follows:

Type             Min Thickness (mil)

PVC	20-30 (30 is very tough)
CPE	30
CSPE-R	36
Polyethylene*	60

*Polyethylene is set as a 60 mil minimum primarily from a seamability standpoint. It has not been clearly demonstrated to us that PE products less than 60 mil can be constantly seamed in the field. There is also concern that this is at the lower limit for creating conditions that encourages stress cracking. While stress cracking is still under review we are starting to see improvements in seaming techniques. It is interesting to note that the West Germans are now requiring PE thickness greater than 100 mil.

On the other hand, a minimum thickness is a preliminary regulatory means of eliminating totally inadequate designs. However, it should be made perfectly clear that this is only a guide, and that engineering calculations should be made to confirm the thickness required for any liner or cover application.

In addition to allowing for different minimum thicknesses of geomembranes Michigan Act 641 requires that a "geomembrane shall be of sufficient tensile strength to withstand anticipated stressed without failure". This provides an opportunity for the regulator to confirm that a proper design has been done and that the related geomembrane thickness is based on design calculations and not the regulations themselves.

3. CLOSURE

A number of geomembranes, including PVC geomembranes, have been used successfully, for many years, to contain municipal and hazardous wastes in Michigan and elsewhere. PVC does not suddenly become inappropriate because HDPE geomembranes have become available. HDPE has many positive attributes, but it also has a number of negative attributes that are not yet fully understood. New materials (e.g. polypropylene/EPDM alloys) that offer specific advantages and, hopefully, fewer disadvantages are appearing and will continue to appear. It is desirable to design geomembrane lining systems using the least expensive material that will best achieve the performance specifications. Regulations should be written, as in the second draft of Act 641, to accommodate all candidate materials that will adequately perform the required function, and the performance of that function should be decided by proper engineering design, not by a regulatory recipe. In no way can a regulation provide a satisfactory design.

In the September/October 1990 issue of Geotechnical Fabrics Report Bob
Landreth of EPA's Risk Reduction Engineering Laboratory stated: 

"The modifications (to chemical analysis techniques and control of wastes) should increase the number of geomembrane compositions available for use. The increased number of geomembrane compositions should now allow the designer to develop innovative designs. We (EPA) believe innovative designs will be more economical, technically viable, and be more reliable."
 

As designers and Michigan DNR recognize, optimum design cannot be achieved by regulating a single minimum thickness. At least two minimum thicknesses are required to accommodate two fundamental types of materials; amorphous thermoplastics (such as PVC) and semi-crystalline thermoplastics (such as HDPE). The minimum thicknesses, (30 mil and 60 mil respectively), specified in Subtitle D are acceptable to producers of both these classes of materials.  Fortunately, a minimum thickness of 30 mil is also acceptable to those producing elastomeric geomembranes and fully-crosslinked elastomeric alloys, two other classes of materials.

There is little doubt that HDPE has become the landfill liner geomembrane of choice, but it has achieved this position with a significant dose of effective marketing and regulatory support. At the same time that the emotive pendulum has swung in favor of HDPE it has unjustifiably swung away from PVC. At present the potential problems of HDPE are overlooked, as are the many advantages of PVC.

Michigan's Act 641 has now been written to include three values, rather than one value, of minimum thickness, and to stress the importance of thorough geomembrane design in order that the people of Michigan get optimum cost effective landfill waste containment from the complete range of geomembranes available. This is a responsible approach. A fair technical hearing, should now be given to PVC as well as HDPE. PVC has performed well in the past in Michigan and, as it continues to be improved, it will perform well in municipal solid waste landfills in the future. Designers and regulators can achieve more, and widen the window of environmental protection, by taking advantage of the many unique properties of PVC. The second draft of Act 641 has opened the door for responsible design, DNR should consider each design on its site-specific, technical merits.