Applications of SEM and ESEM in Microstructural Investigation of Shale-Weathered Expansive Soils along Swelling-Shrinkage Cycles

Highlights

• We compared micrographs of two expansive soils attained from different microscopy methods.

• Similarities were found between the images taken from the three microscopy approaches.

• Structural modifications throughout two wetting-drying cycles were shown.

Abstract

Understanding the microstructure of expansive soils is of vital importance in physical interpretation and mechanical modeling of expansive soil behavior at the macro scale. To approach this objective, two natural expansive soils were compacted and imaged to disclose their microstructure using both conventional scanning electron microscopy (SEM) and environmental scanning electron microscopy (ESEM). The micrographs of the samples at their initial compacted state showed comparable results when using SEM and ESEM techniques. The structural changes in response to two wetting-drying cycles were monitored through the ESEM experiments in order to simulate the volumetric evolution at the macro scale. Despite the significant macroscopic volume changes, it was difficult to detect structural changes at the micro scale. The ESEM test results only qualitatively described structural shrinkage upon the first drying cycle but showed little structural modification thereafter, due to the limited duration of vapor exchange between the chamber atmosphere and the sample under the microscope. In addition, the combined effects of plasticity and original microstructure determined how the microstructure evolved as a function of sample consolidation. Based on the analysis of results, limitations of both SEM and ESEM approaches in microstructural investigations of expansive soils are discussed and summarized.

Introduction

Since the interparticle physicochemical forces that determine the overall mechanical behavior of fine-grained soils take place at the microstructural level, it is of significant interest to learn about the initial microstructure and the subsequent microstructural variations in response to hygroscopic evolution. Scanning electron microscopy (SEM), which has been widely used in geology, geochemistry, material sciences and biology fields, can provide a direct characterization of this microstructure (particle/aggregate scale, < 100 μm) to help explain the mechanical responses of clayey soils at a macro scale (Romero and Simms, 2008). In the SEM technique, soil or rock samples must be completely dried, fractured and coated with gold or platinum before observation under the electron beam. No single sample preparation technique is considered most appropriate (Al-Rawas and McGown, 1999), since air drying, freeze drying and critical point drying all have associated sample disturbance concerns (Tovey, 1970, Barden and Sides, 1971, Collins, 1978, Smart and Tovey, 1982). These sample disturbance concerns include (i) air drying induces volumetric shrinkage in wet clayey soils; (ii) freeze drying introduces overall swelling owing to unavoidable partial recrystallization of water (to ice); (iii) critical drying can cause particle breakup. Recent technological developments introduced the application of environmental scanning electron microscopy (ESEM), which is an advanced SEM equipped with a (i) separate dedicated vacuum pump that controls the vapor pressure of the chamber; (ii) gaseous secondary electron detector (GSED) that prevents charging on the sample; and (iii) thermoelectric Peltier cooling stage that allows for the control of sample temperature. The technique, therefore, is capable of maintaining desired environmental conditions across a range of relative humidity (RH) during the detection of sample structure. No sample desiccation and coating are required; as a result, the micrographs are more representative of how the structure exists in nature, since the sample disturbance is minimized. The main drawback of ESEM is associated with the image quality, in terms of contrast and brightness, which strongly depends on the chamber humidity (Maison et al., 2010). The observed images could be dimmed by interference of water, to various degrees, dependent on vapor pressure; the higher the vapor pressure, the lower the image resolution. Furthermore, the structural information of assemblages bounded by the water film cannot be accurately determined because the electron beam does not penetrate through water. While observing the original structure of soil samples at a specific state is useful, the compromised resolution of the ESEM makes it essential to pair the information with conventional SEM that is able to produce micrographs of high resolution and provide complete structural information.

Applications of SEM and ESEM have been reported to analyze the effects of structure of clayey soils/rocks on their engineering behavior, or vice versa. For example, by using SEM, Katti and Shanmugasundaram (2001) found that an increase in volumetric swelling, or reduction in swelling pressure, results in a decrease in clay aggregate size. Cui et al. (2002) investigated the microstructural modification of swelling clay with decreasing suction and discovered that the macropores were being closed by aggregate deformation while the micropores remained unaffected. Avsar et al. (2009) studied the relationship between swelling anisotropy and structure of Ankara clay and disclosed higher swelling pressure perpendicular to the sheeting direction of the face-to-face sheeting sample. Langroudi and Yasrobi (2009) elucidated the impact of drainage conditions on the swelling behavior of unsaturated expansive clay. Applications of SEM were also reported in investigations of the structural alterations related to lime/cement or sludge/fly ash stabilizations that lead to an increase of sample stiffness (Choquette et al., 1987, Lin et al., 2007, Lin et al., 2013).

The applications of ESEM have been successful, for instance, in displaying microstructure of clayey soil samples at a specified environment of vapor pressure and temperature (Komine and Ogata, 1999, Komine and Ogata, 2004, Zhang et al., 2004); visualizing structural swelling at macro and mesostructural scale (> 100 μm) under constant temperature (Baker et al., 1995, Villar and Lloret, 2001, Viola et al., 2005, Koliji et al., 2010, Maison et al., 2010), and studying the microstructural (< 100 μm) modifications along hydration and dehydration paths by varying chamber temperature (with accompanying modification of vapor pressure) (Montes-H et al., 2003a, Montes-H et al., 2003b, Monte-H et al., 2004). The major conclusion of these studies was that the inter-aggregate macropores are filled by expanded clay particles or aggregates upon hydration, as was summarized in Romero and Simms (2008). The highlighted work was volumetric quantification of swelling and shrinkage of a single MX80 bentonite aggregate (Montes-H et al., 2003a, Montes-H et al., 2003b) using the ESEM technique coupled with a digital image analysis program. Additionally, microstructure information, based on ESEM or SEM, aided in proposing conceptual (Zhang et al., 2004) and constitutive (Man and Graham, 2010) models for highly plastic clay. Another technique that is able to provide quantitative information of the porosity structure (e.g., intra/inter-aggregate size/space) is mercury intrusion porosimetry (MIP), which can be used together with SEM or ESEM in microstructure investigation of clayey soils or rocks (e.g., Romero et al., 1999, Cui et al., 2002, Romero and Simms, 2008, Li and Zhang, 2009, Koliji et al., 2010).

This research effort used both SEM and ESEM techniques to study the microstructure of two natural soils; Carnisaw and Eagle Ford. It was important to first compare ESEM results with those from two SEM preparation methods (air and freeze dried) in order to assess what differences, if any, occurred because of sample preparation and subsequent disturbance. Once the sample preparation differences were noted, an investigation of the microstructural evolution as a function of suction on an ESEM platform followed. In the work of Monte-H et al. (2004), two changing environmental factors, temperature and vapor pressure, were accounted for when studying the structural modifications of an argillaceous rock. The role of one factor relative to the other on the microstructure, however, was left unsolved. The research described in this paper simplified the thermodynamics by setting the test chamber to a constant temperature of 5 °C. In addition, a moisture content-volume-suction relationship was developed based on the results from other laboratory efforts (as discussed later). These types of relationships will help us evaluate how capable the ESEM technology is in capturing, or predicting, structural modifications at the macro scale.

The microstructural changes due to sample consolidation are also of substantial interest in evaluating the influence of structure modification and densification on shear strength. With this in mind, the microstructures of Carnisaw and Eagle Ford specimens (described in the next section) were observed after consolidation to effective stresses of 100 kPa and 400 kPa. The effect of consolidation/confining pressure was assessed only on saturated samples, where interparticle physicochemical forces were fully developed. The same type of effect on samples with specified suction would be interesting to investigate; however, suction control can be challenging during sample preparation prior to ESEM observations.