Comparative assessment of expansive soil stabilization by commercially available polymers

Abstract

The objective of this study was to determine the effectiveness of commercially available polymer soil stabilizers in expansive soil swell reduction relative to quicklime and Class C fly ash. A survey of state departments of transportation within the Mountain-Plains region of the USA (Colorado, Montana, North Dakota, South Dakota, Utah, Wyoming) was conducted to define the state-of-practice in expansive soil mitigation, from which lime and fly ash were identified as the most used soil stabilizers. Four commercially available polymers were tested for comparison with lime and fly ash. Untreated and treated soils were classified, and tested for swelling pressure, swelling potential, unconfined compressive strength, and hydraulic conductivity (k). Relative to untreated soil, polymer treatment was less effective at reducing the swelling potential (70% reduction) and increasing unconfined compressive strength (46 kPa increase) of a highly expansive soil relative to lime (100% reduction, 1260 kPa increase) and fly ash (97% swelling potential reduction, 380 kPa increase in UCS). However, lime and fly ash treatments resulted in a 52,000- and 1,100-times increases in k, respectively, while polymer resulted in only a 2-times increase in k relative to the untreated soil (k = 2.9 × 10^-11 m/s). Polymer was also shown to be effective at reducing k to below 1.9 × 10^-12 m/s when used as a spray-on coating. The results of this study illustrate that polymers reduce swell in expansive soils by mechanisms that are different than lime and fly ash.

Introduction

The shrink-swell behavior of expansive soils reduces infrastructure longevity in many regions of the world [36]. Inyang et al. [28] estimate that shrink-swell behavior of expansive soils with moisture variation cause more than 50% of soil related damage to infrastructure globally. Roadways are particularly susceptible to the effects of expansive soils due the combination of low ground pressures and large surface areas. Economical solutions to mitigate damage from expansive soils are necessary to enhance transportation system longevity. The objective of this study was to compare the effectiveness of commercially available polymer-based soil stabilizers to standard treatments used in practice to assess the relative effectiveness of commercially available polymers to mitigate shrink-swell behavior of expansive soils.

Techniques used to mitigate shrink-swell of expansive soils include physical and chemical treatments. Common physical treatments include removal and replacement, over-excavation and re-compaction, pre-wetting, and encapsulation by geomembrane barriers. Removal and replacement reduce shrink-swell behavior by removing the expansive soil within the zone of shrink-swell (active zone) to an appropriate depth and replacing with non-swelling soils [37]). Overexcavation and re-compaction mitigate shrink-swell by remolding and compacting an expansive soil at standard Proctor optimum moisture content, or more commonly, at a moisture content greater than optimum (wet-side compaction) [24], [40]. Pre-wetting reduces the tendency of an expansive soil to undergo substantial additional swell after construction by compacting at a pre-swollen (wetted) condition during construction [5]. Physical isolation of expansive soils by geomembrane barriers is also used to mitigate swelling by minimizing moisture migration, thereby, eliminating the cause of shrink-swell [33], [43]. Other physical treatments have included surcharge loading, asphalt treatment, explosive treatment for expansive shales, and electro-osmosis [27], [36], [19]; however, these techniques are less common in practice today [37].

Chemical treatments include application of stabilizing additives such as lime, fly ash, cement, polymers, salts, organic compounds, and enzymes. Chemical treatments can be further subdivided into traditional and nontraditional stabilizers. Traditional stabilizers include fly ash, lime, and cement, all with pozzolanic properties that reduce shrink-swell behavior by chemical reactions between the clay minerals and the calcium oxide molecules at the surface level that result in bonding the soil particles together as well as reducing the affinity of the soil to water [23], [39], [40], [28], [20], [32]. Traditional stabilizers have limited effectiveness in highly active soils, (i.e., soils with a plasticity index (PI) greater than 50 [40]), expansive soils containing carbonates that lead to carbonation and swelling, or expansive soils containing sulfate salts (e.g., gypsum) or sulfur that can lead to ettringite formation and swelling [31]. Traditional stabilizers typically require a curing time after placement to allow cementitious bonds to be effective. Finally, there remain environmental concerns with stabilizer production and placement both in terms of impacts on greenhouse gas production and salinization of water resources [40].

Nontraditional stabilizers provide an alternative to traditional stabilizers, and include polymers, salts, and enzymes. Salts and enzymes reduce swelling by improving the ionic composition of the soil (e.g., replacing monovalent sodium for divalent calcium) and reducing the concentration gradient between absorbed and free pore water, which reduces swelling [41], [45]. Conversely, natural and synthesized polymers reduce shrink-swell by creating a nano-composite structure that results in bonded soil particles [46], [13]. Nontraditional stabilizers are attractive alternatives to traditional stabilizers as they have potential to function in soils containing carbonates, sulphate salts or Sulphur, require zero or minimal curing time, cost less, and reduce environmental impacts [29]. An in-depth review of expansive soil mitigation practice using traditional and nontraditional stabilizers is provided in Taher [44] and Behnood [15].

This study provides a comparative assessment in terms of swelling reduction of commercially available (i.e., off the shelf) polymer soil stabilizers relative to traditional stabilizers, i.e., quicklime and Class C fly ash. Polymer-based soil stabilizers have been shown to increase unconfined compressive strength and reduce swell potential, swell pressure, and hydraulic conductivity of expansive soils [28], [19], [34], [48], [12], [13], [35]. However, direct comparisions with traditional stabilizers are limited, making effectiveness-driven decisions by practicioners difficult.

A high swell classification expansive soil (classified based on FHWA-RD-77–94; [42] was used for all tests. One-dimensional swelling, unconfined compressive strength, and hydraulic conductivity were used to comparatively assess aspects of swell mitigation by different treatments. The effectiveness of polymer treatment relative to traditional stabilizers was evaluated in the context of findings distilled from the tests that reflect mechanisms that govern swelling. The results from this study provide an example of a comparative assessment between technologies, and can be used to guide how treatments are compared.

Prior to testing, a survey of expansive soil treatment methods used by state Departments of Transportation (DOTs) in the Mountain-Plains region of the USA (Colorado, Montana, North Dakota, South Dakota, Utah, Wyoming) was conducted to identify the state-of-practice in expansive soil mitigation for roadway applications. DOTs in the Mountain Plains region were selected based on the prevalence of moderate and high swelling potential soils across much of the region [38]. Six state DOTs were solicited for the survey and five states responded. Each state DOT was asked to respond to the following seven statements and questions; state-by-state survey results are provided in Appendix A.

• List methods your DOT uses to identify expansive soils. How does your DOT decide on the severity of expansive soils based on your identification methods?

• List any remedial measures that your DOT has ever taken to eliminate or mitigate swelling problems.

• Identify and describe the mitigation techniques that have worked best.

• Identify and describe techniques that have not worked, or that your DOT no longer uses.

• Provide the names or links of the document guidelines that your DOT uses in dealing with expansive soils.

• Does your DOT use polymer as a stabilization technique? If yes, please explain why? What is the manufacturer/company that provides the polymer for your DOT?

• Please provide any additional comments/suggestions.

Based on the state-of-practice survey, the two primary physical techniques used to mitigate expansive soils are removal and replacement (5 of 5 DOTs surveyed) and remolding and compaction (3 of 5 DOTs surveyed). Lime (5 of 5 DOTs surveyed) and fly ash (2 of 5 DOTs surveyed) are the most commonly used chemical treatments but were less commonly used than physical techniques. Based on these findings, lime and fly ash were selected for comparative testing with commercially available polymers. At the time of the survey, 2016, commercial polymers were not used for mitigation of expansive soils by any of the DOTs surveyed.

Methods

Atterberg limits, standard Proctor compaction, one-dimensional swelling, unconfined compressive strength, and saturated hydraulic conductivity tests were used to evaluate untreated (baseline) and treated expansive soil. All soil was air-dried and passed through a number 40 sieve prior to specimen preparation. Lime and fly ash treated specimens were prepared according to the ASTM standard procedures for each test and were manually mixed at specific mixing ratios (described in Materials) in an air-dry condition until no heterogeneity and particle segregation was visually observed in the mixture. For the Atterberg limit tests, water was added to determine the liquid limit and plasticity index, and for the one-dimensional swell, unconfined compressive strength, and hydraulic conductivity tests water was added and soils were compacted to reach (standard Proctor) optimum moisture content and maximum dry density.

Polymer solutions were also tested as bulk treatments and all sieving, compaction, and curing procedures applied to lime and fly ash treated soils were repeated but moisture addition and mixing was performed differently because, unlike lime and fly ash, commercial polymers were supplied dissolved in water. Therefore, polymer solutions were mixed (diluted) with a designated amount of water needed to achieve the test target moisture content (i.e. optimum moisture content). The diluted polymer solution was then added to the sieved air-dry expansive soil by spraying while simultaneously manually mixing with a spatula until a visually homogeneous paste was formed.

• Atterberg Limits

Atterberg limits were performed in accordance with ASTM D4318-00 [6] and were run immediately after mixing with no curing period. Atterberg limits were used for preliminary swell prediction based on FHWA-RD-77-94 [42] to select polymer stabilizers that had a maximum impact on predicted specimen swelling.

• Standard Proctor Compaction Test

Standard Proctor compaction tests were performed in accordance with ASTM D698-12 [8] to prepare specimens for swelling, unconfined compressive strength, and hydraulic conductivity testing at maximum dry density and optimum water content for each soil. Compaction at maximum dry density and optimum water content is similar to, but slightly denser than, typical compaction criteria for over-excavation and re-compaction in practice (e.g., [21]).

• One-Dimensional Swell Tests

To assess the effects of treatments on swelling, one-dimensional swelling tests were performed in accordance with ASTM D4546-14 [9] following Method A (for reconstituted specimens). After preparation at optimum water content and maximum dry density, soil specimens within odometer rings were wrapped in plastic and cured for seven days at 40 °C; curing at 40 °C was performed to achieve accelerated curing as described in ASTM D5102-09 [7]. Excess pore water pressure was not measured during testing; however, excess pore water pressure was inferred to have dissipated by the end of the swell test based on continuation of tests until no continued deformation (swell/collapse) was measurable over a four-day period.

• Unconfined Compressive Strength Tests

Unconfined compressive strength was measured in accordance with ASTM D5102-09 [7] to assess the effect of treatments on soil strength. The strain-controlled method for brittle specimens was used with an axial strain rate of 2.0% per minute. Test specimens were prepared in plastic cylindrical split-mold with diameter equal to a standard compaction mold (101.6 mm) and height equal to double a standard compaction mold (232.9 mm) was used in accordance with ASTM D5102-09, Procedure A [7]. Soils were compacted in six layers with 25 blows per layer to meet the standard compaction effort. The cylindrical split-mold was lined with plastic wrap to facilitate removing specimens and wrapping them along the long axis for subsequent hydration and cure. After compaction and removal from the mold, soil specimens were wrapped in plastic and cured for seven days at 40 °C. Then, specimens were placed on porous stones in a pan of tap water and soaked for 24 h during which the water level in the pan was maintained at the top of the porous stone. Three specimens were performed for each treatment and the mean of each is reported herein.

• Hydraulic Conductivity Tests

Saturated hydraulic conductivity tests were performed in flexible-wall permeameters in accordance with ASTM D5084-16a [11], Method C (falling headwater, rising tailwater elevation) to assess the effect of treatment on moisture movement. The specimen preparation and compaction method described in Tong and Shackelford [47] was used in this study, and included a specimen height of 29.3 mm and diameter of 101.6 mm. Compaction was accomplished in a single layer with 19 blows of a standard Proctor compaction hammer to achieve the standard Proctor compaction energy. Hydraulic conductivity tests were performed such that the applied effective stress was equal to the soil swelling pressure, as per ASTM D5084-16a, to minimize any volume change during testing. Tap water from Fort Collins, Colorado, USA, was used as the permeant solution.