Written by: Bryan Wedin, Chief Engineer
An accurate understanding of creep resistance is essential to proper material selection when using polymers, and in the case of geocells, this science is being misapplied.
The definition of creep deformation is “the tendency of a solid material to move slowly or deform permanently under the influence of mechanical stress.”
This potential failure mode creates fear and uncertainty among designers wherever the possibility of creep factors exists. Yes, creep can occur in almost all materials including plastics, metals, and concrete. In cases such as bridge and building design, it is important to properly understand creep factors and account for creep in engineering calculations. However, in the case of designing with geocells for load support, creep factors have no relevance.
What causes creep?
In order for creep to occur, two factors must be present:
1) A constant load applied to the area and
2) A sustained deformation of the geocells.
Creep only applies when there is a sustained load on a material for an extended period. In a case of repeated on- and off-loading, this type of deformation would be governed by fatigue, not by creep, because it is not a constant applied load.
The second required factor for creep to occur is an ability to undergo sustained deformation of the material. When a polymer has a load applied, the molecules of the material start to pull apart and stretch. This leads to elongation of the material in one direction and typically results in a thinning of the material’s thickness.
Creep not a factor in load support
Now, consider a geocell load support application. The geocell material is expanded to cover the project area on-site and then an infill material is placed into the cells. At this point, there is not an applied load or deformation occurring in the material.
Next, the infill material is compacted. This compaction applies a load to the cells, but this load is removed as soon as the compaction equipment is no longer positioned over the cells.
As an individual cell is loaded, it exerts an outward force while each of the adjacent cells pushes back on it (passive resistance) and prevents any sustained deformation. Therefore, at the time of compaction, there is neither a constant load nor is there a sustained deformation. Thus far, the material is successfully installed without any creep effects.
After the geocell load support system has been installed, the two types of loads that will affect the system are driving (live) and stationary (parked) loads. When a vehicle drives over a geocell system, the load is applied vertically. As the geocell distributes the load laterally, there is a temporary load applied to the geocell material. The load is not a sustained load, and therefore, would not have a creep effect.
In the case of stationary loads, the load is continually applied to the geocell, so it meets the first criteria for creep. Due to the pressure from all adjacent cells surrounding the loaded cell(s), there is no ability for the cells to move enough to have any appreciable sustained deformation. Therefore, creep cannot affect this scenario.
ASTM D6992 creep test is not applicable
Those who make claims about creep potential in geocell load support applications have cited ASTM standard methodology as the reasoning for concern.
ASTM standards provide an accepted means for standardizing testing to be able to directly compare products. It is important to review the intention and scope of a test to ensure that it is appropriate and will give relevant results. The Stepped Isothermal Method (SIM) is used to accelerate creep testing. ASTM D6992 uses SIM to predict the expected deformation of geosynthetics over time when used for reinforcement applications. This method can be effective; however, it is not suitable for a polyethylene geocell evaluation.
ASTM D6992 5.3 Note 1 states, “Currently, SIM testing has focused mainly on woven and knitted geogrids and woven geotextiles made from polyester, aramid, polyaramid, poly-vinyl alcohol (PVA), and polypropylene yarn and narrow strips.”
Additionally, the note continues with a warning against expanded scope of the test saying, “Additional correlation studies on other materials are needed.” While this test has applicability for geogrids and geotextiles, the test is not intended for evaluating geocells and correlations for polyethylene have not yet been established.
Further, D6992 cannot be considered in isolation.
D6992 states, “Results of this method are to be used to augment results of Test Method D5262 and may not be used as the sole basis for determination of long-term creep and creep-rupture behavior of geosynthetic material.”
This reinforces the importance of reviewing each test standard to ensure that the product is within the scope of the test and that the results are relevant and complete. In the case of geocell evaluation, using ASTM D6992 is inappropriate as it has not been properly correlated to provide accurate evaluation of polyethylene and without ASTM D5262, it provides an incomplete overall evaluation of the product.
HDPE’s long history of success and repeatability
High Density Polyethylene (HDPE) has been used as the industry standard material for geocells since it was invented over 40 years ago. HDPE has been extensively researched by independent scientists across multiple industries, allowing for a comprehensive understanding of its performance capabilities. Using a virgin HDPE material allows for direct verification of resin consistency through laboratory testing to ensure that each manufacturing location and production lot have consistent material performance. This laboratory verification also allows for the comparison of the material to independent scientific results and does not rely solely on manufacturer’s claims.
Challenges with Fabricated Inelastic Blend (FIB)
A few geocell manufacturers are promoting a Fabricated Inelastic Blend (FIB) to cut manufacturing costs and increase material stiffness utilizing recycled and other unpublished polymer materials. These FIB-based materials can vary widely, even for the same product. Due to the vast number of combinations possible with FIB materials, they pose two key problems when included as a material choice: validation and consistency.
Because of the unpublished nature of the blending mixture there is no way to validate this material in comparison with published testing. Any testing of FIB materials must start from the beginning without any experience to rely on for long-term performance.
The second concern with FIB materials is controlling consistency of the blend. Because each FIB blend is so variable, there is no way for a third-party tester to fully determine consistency of the blend between different manufacturing plants or even between different production lots. This inability to determine consistency creates uncertainty because there is no way to determine if there has been improper blending or changes to material blend.
Manufacturers using FIB materials promote the advantages of increased material stiffness. This stiffness is often a function of multiple generations of recycling. It is important to review the differences between elastic and inelastic materials and how they affect geocell performance. An elastic material is able to undergo a deformation (strain) and then spring back to its original state without permanent (plastic) deformation.
Conversely, an inelastic material does not undergo immediate deformation, but rather ends in catastrophic (complete) failure. Many of engineering’s worst failures have resulted from catastrophic failures of inelastic materials that were subjected to unexpected loads. This absolute nature of inelastic failure puts projects at great risk because it does not give indication prior to collapse. With elastic materials, as material limits are reached, the material will stretch and yield prior to complete material failure. This yield zone allows for changes to loadings prior to catastrophic failure.
The mobilization of soil strain and geocell strain occurs from not only deformation of infill, but deformation of subgrade materials. For reasonable ranges of geocell stiffness, subgrade deformations will cause the strain to occur in the geocell system. For a given strain stemming from subgrade deformation, a stiffer geocell will realize larger tensile stresses, both in its wall and especially at the seam, which will result in seam rupture and system failure.
The material stiffness is not the most critical point in geocell performance. It is a combination of stiffness, tensile strength, seam strength, and passive resistance of adjacent cells. Published data shows that strain in geocell cell walls is on the order of less than 0.5 to 1.0%. At such low strains, the effect of creep should be ignored for all practical purposes. Properties such as seam strength, strip flexibility, environmental stress cracking resistance, and passive resistance from adjacent cells play a much larger role than stiffness of the material. Also, at these strain rates, HDPE (including virgin mixes, most recycled and other polymer alloy geocell) stress-strain behaviors are similar.
In load support applications, loadings are transient and quite small due to the stress-distributing behavior of the pavement material and geocell mattress effect, which further compounds the irrelevance of creep in reinforcement applications. Thus, overall system strength is not related to any performance factors that are tied to creep. Therefore, the focus should be on how much strain is mobilized in geocells.
Some manufacturers have examined theoretical 2% to 5% strain rates, both of which are far above actual field conditions, and therefore, not applicable. Five percent strains in load support applications are not applicable since compacted granular materials fail at much lower strain rates. At 2% strain, granular materials would fail within a geocell, as well as outside of the confined system, due to significant rutting and deformation. The geocell material would not be the failure point, and a stiffer geocell would not affect the failure of granular materials at 2% strain.
Geocells used in load support applications prevent high strains from occurring due to the very nature of the geometry, its confining behavior, and passive resistance of adjacent cells. Evaluating geocell material strength beyond reasonable strain values is not relevant for subgrade reinforcement as it contradicts actual measured data and represents conditions outside of practical design and application. The basis of comparison should focus on stress-strain behavior at very low strains – less than 1.0%. This information is readily available from large-scale laboratory tests, field tests, and numerical modeling.
True HDPE Performance vs FIB Results
FIB materials bring a new uncertainty to the geocell market. These materials are of unverifiable composition so connecting material to performance is nearly impossible. Ultimately, these FIB materials beg your trust in their performance touting their unnecessary creep resistance. They hide the truth that creep resistance comes at a cost – inelastic material that can fail catastrophically at the seam. FIB material prioritizes a single material property of the geocell at the expense of a uniformly designed system with measurable material consistency and applied field testing.
After 40 years, HDPE continues to be the industry standard material for geocells. Presto Geosystems proudly pioneered the use of HDPE material in its GEOWEB® Geocell products due to proven performance and reliability of that material.
Over these four decades, GEOWEB Geocells have been used in load support projects worldwide without a single failure due to creep effects. Although this consistent performance is impressive, it is not surprising given that creep forces are irrelevant in these applications.