Pat Harris

Forensic evaluation of three failed cement-treated base pavements

Three cement-treated base (CTB) pavements constructed around 1990 near Houston, Texas, showed severe pavement deterioration after 3 to 4 years. The unusual distress patterns included wheelpath alligator cracking and severe pumping. A forensic investigation was initiated to determine the cause and to recommend modifications to materials specification, design, and construction practices to avoid future problems. The primary cause was determined to be chemical deterioration that resulted in destruction of the cement matrix. In all cases water was trapped within the CTB layer. In two bases moisture flowed rapidly by capillary action through the CTB; these bases showed the most rapid deterioration. Water flow in one case was attributed to smectite clay contamination of the fine aggregate and in another to a highly absorptive sandstone coarse aggregate. Extensive ettringite was found in one material. In one case deterioration was caused by the growth of two phases exhibiting different morphologies: hydrated calcium aluminum silicate with a radiating needle morphology and ettringite, which showed a bladed morphology in these samples. Practical recommendations to prevent recurrence of the problem were (a) moisture absorption as acceptance testing for coarse aggregate and (b) elimination of designs placing different stabilized materials on top of one another. The laboratory suction test that was developed is recommended as a screening test for molded CTB specimens.

MEASURING SULFATE IN SUBGRADE SOIL: DIFFICULTIES AND TRIUMPHS

The accurate analysis of sulfate sulfur in subgrade soil is essential for road construction that involves calcium-based stabilizers (lime or cement). The objective was to determine if other tests give more reproducible results in a more timely fashion than Texas’s current sulfate test method (Tex-620-J), a gravimetric method. Literature review and interviews with commercial soil testing laboratories revealed three techniques to compare with Tex-620-J: ion chromatography, conductivity, and colorimetry/spectrophotometry. Soils were manufactured with known sulfate (gypsum and anhydrite) concentrations and sent to laboratories that performed gravimetric analysis and ion chromatography. Conductivity and colorimetric testing were performed in-house. Testing showed that Tex-620-J is not very precise, which creates the need for an unrealistic number of samples to obtain an accurate estimate of the sulfate concentration. To define the 95% confidence interval for true sulfate content to within ±10% of the true known value for a concentration of 5,000 ppm, Tex-620-J requires 43 tests; ion chromatography requires 14 tests. The colorimeter achieved the desired results in only one test, based on sulfate standard solutions. Results of this study revealed the difficulty with obtaining accurate sulfate measurements in the laboratory and indicated a few inexpensive pieces of equipment that can be used in both the field and laboratory settings that may yield better results.

Stabilization of High-Sulfate Soils by Extended Mellowing

The use of lime to stabilize expansive soils has been the preferred technique for many years. However, heaving and premature pavement failures in lime-treated expansive subgrades containing sulfates led to the search for alternative stabilization techniques. Of the several techniques developed, precompaction mellowing has the potential to be effective in stabilizing sulfate-bearing soils. Yet this method needs experimental evaluation. In the current study, an attempt was made to assess the stabilization effectiveness of precompaction mellowing on high-sulfate soils. For this task, six natural expansive soils from Texas, with sulfate contents varying from 200 to 44,000 ppm, were collected. Soils with low-sulfate contents were spiked with additional sulfates to make them high-sulfate soils. Basic classification and chemical tests were performed to establish the clay mineralogy of the soils. Three mellowing periods (0, 3, and 7 days) were studied. The test soils were treated with lime and allowed to mellow for the specified periods. Following the mellowing, the samples were subjected to three-dimensional tests for volumetric swell, shrinkage, and unconfined compressive strength (UCS). To study the consumptions of alumina and silica during sulfate–soil–lime reactions, reactive alumina and silica measurements were also attempted. The authors observed that shrinkage was of no concern in treated soils because the shrinkage invariably reduced with lime treatment. In four of the six soils, precompaction mellowing reduced sulfate-induced swell to a level below the natural expansive swelling. The UCS strengths of treated soils decreased slightly with mellowing. Reasons for the anomaly in UCS strengths and ineffectiveness of precompaction mellowing in two soils were explained.

Part 3: Cementitious, Chemical, and Mechanical Stabilization: Recommendations for Stabilization of High-Sulfate Soils in Texas

In an effort to construct roads more quickly, high-plasticity index soils stabilized with lime are now routinely compacted the day after mixing. With this practice has come an increasing number of heaves due to soluble sulfates reacting with the lime to form ettringite. Soils with sulfate concentrations below 7,000 to 8,000 parts per million (ppm) can generally be treated with lime. This research was performed to identify stabilizers that can be used with sulfate concentrations above 10,000 ppm. The effectiveness of the stabilizers was determined by the measurement of three-dimensional (3-D) swell reduction and unconfined compressive strength. The researchers evaluated 12 stabilizers, including enzymes, polymers, acids, emulsions, fly ash, and ground granulated blastfurnace slag (GGBFS). Three stabilizers significantly reduced volumetric swell. A polymer and an acid reduced swell by about 8%. GGBFS plus lime reduced swell by 10%. GGBFS plus lime was the only stabilizer that reduced swell, increased streng...

Effects of Organics on Stabilized Expansive Subgrade Soils

Organic soils are the combination of decayed plant matter and weathered rock material. These soils are known for their inferior engineering behavior. In order to get the precise control over the engineering behavior of organic rich soils, they are often stabilized either with lime or cement. However, transportation agencies in the United States reported that these stabilized soils bearing organic content less than 1 percent have never achieved the desired improvement or the improvement disappeared over a period of time. Therefore, a laboratory investigation was initiated to understand the mechanisms when calcium-based additives (lime and/or cement) are amended with organic rich expansive soils. Two soils having high and low organic content were considered for detailed investigation. Firstly, optimum dosage of the stabilizer was determined for the selected soils. The strength of control (untreated) and stabilized soils were determined at different curing periods. In this paper, the influence of organic matter on the engineering behavior of the lime and cement stabilized organic rich soils are discussed in detail.

Stabilizer Diffusion in Swelling Soils

There has been much speculation about the ability of lime and cement to diffuse into highly plastic soil and chemically alter the clay minerals thereby reducing the shrink/swell capacity of the soil. This research looks at hydrated lime and type I Portland cement in a manufactured soil to obtain direct evidence of the ability of both lime and cement to diffuse into and chemically alter a clay soil. Natural soils are very complex and stabilizer/soil reaction products are difficult to identify so we chose to manufacture a soil. We manufactured five soil samples composed of pure quartz sand (70 wt. %) and pure smectite clay (30 wt. %). Electron probe microanalyses of thin-sections from the manufactured soil samples show high concentrations of calcium throughout the lime stabilized clay aggregates, but very little calcium penetrated the cement stabilized clay. X-ray diffraction confirmed the presence of smectite (clay) in all samples and suggests more calcium substitution in lime treated samples. Calcium from the lime diffuses into the clay to create a chemically altered clay that is less susceptible to shrinking and swelling. The cement forms a protective coating around the clay aggregates and very little calcium actually diffuses into the clay. Thin-section analysis of samples from an expansive, natural soil stabilized with lime and cement shows identical textures to the manufactured soil samples. This research provides direct evidence of lime and cement stabilization mechanisms in clay-rich soils which will allow engineers to make informed decisions about stabilizer options in soils that exhibit shrinking and swelling.

Killing the Ettringite Reaction in Sulfate-Bearing Soils

The formation of ettringite in sulfate-bearing clay soils often results in rapid and significant expansion when calcium-based stabilizers are added to the soil. The researchers tested the hypothesis that ettringite crystal growth could be inhibited by adding a specific chemical that would prevent nucleation of ettringite crystallites, much like retarding the setting of portland cement by using organic additives. The researchers created ettringite in the laboratory by using clay mineral standards mixed with hydrated lime, gypsum, and water, with an ultrasonic homogenizer to catalyze the reaction. A natural soil from Texas (State Highway 289) that had caused sulfate-induced heave problems was also used. Diatomaceous earth, volcanic glass (amorphous silica), and calcium phosphate monobasic monohydrate were added in different proportions to see whether the formation of ettringite could be stopped. The diatomaceous earth and volcanic glass did not inhibit the growth of ettringite in any of the clay mineral standards or natural soil tested. However, the calcium phosphate monobasic monohydrate generated tricalcium aluminate monosulfate hydrate instead of ettringite in the kaolinite and in the natural soil, prevented the formation of both ettringite and tricalcium aluminate monosulfate hydrate. Therefore, phosphates may be a viable additive for prevention of heave in sulfate-bearing soils.

Avoiding Sulfate Heave in Subgrades with a Little Help from Precision Agriculture

In recent years, the Texas Department of Transportation has observed the heaving of gypsum- or sulfate-bearing subgrade soil when it has been chemically stabilized with calcium-based additives. The occurrence of sulfate or gypsum deposits in these soils is unpredictable and often localized in small areas. Current testing procedures, in which samples are collected at specified intervals for sulfate measurements, often miss these localized concentrations. A method that provides rapid and continuous mapping of sulfate content in subgrades to a depth of at least 1 m is needed. Researchers in precision agriculture have used several devices that rapidly and continuously measure the bulk electrical conductivity of the soil at high spatial resolutions (<10 ft) and depths (≥2 ft). For the research described in this paper, the Veris 3150 was used to measure spatial variability in bulk electrical conductivity across three sulfate-bearing subgrades in Texas. The data were used to construct high-resolution maps of the subgrade conductivity and were compared with soil properties to identify potentially high-risk areas for sulfate heaving. The bulk electrical conductivity, the sulfate content, and the soil plasticity varied in tandem throughout the subgrades. This research identified areas with an electrical conductivity of >100 mS/m as having the greatest potential for sulfate problems of the subgrades tested. It also identified a technique that could be used to rapidly and effectively screen large projects for potential sulfate problems and to map the distribution of high- and low-clay subgrade soils.