SEM Characterization of Shale Gas Reservoirs Using Combined Secondary and Backscatter Electron Methods<subtitle>An Example from the Haynesville Shale, Texas and Louisiana</subtitle>

Abstract

Abstract Secondary electron (SE) images of rough three-dimensional shale surfaces and backscattered electron (BSE) images of smoothed argon (Ar)-ion-milled surfaces provide valuable information to the study and understanding of texture, composition, and pore systems in unconventional shale reservoirs. This chapter describes a method to acquire both traditional SE and Ar-ion-milled BSE images from the same sample by acquiring images from adjacent sample splits. This method allows side-by-side comparisons of traditional SE and Ar-ion-milled BSE images, providing much more information than can be obtained from either method alone. This dual method of imaging representative sister samples, capturing the same features at the same scale, gives powerful information that could not be effectively addressed before. The traditional scanning electron microscopic (SEM) method images secondary electrons that are ejected from the k-orbitals of the specimen atoms by inelastic scattering interactions, with beam electrons giving a large depth of field on high-relief surfaces. This methodology cannot fully assess the exact location of the dominant porosity type or the nature and distribution of the organic matter on the high-relief surface. These considerations are essential to petrographic evaluation of reservoir and completion quality in unconventional shale reservoirs. SEM samples prepared using Ar-ion milling with BSE detection allow us to effectively evaluate the nature and distribution of the organics and the porosity by looking at electron density contrast or Z-contrast. Lower-density material, such as organic matter, appears darker, and higher-density material, such as quartz, appears brighter in SEM images. Void space, such as porosity, appears black. However, this method cannot accurately determine texture or differentiate some mineral cement because many minerals are of similar densities; hence, they have similar Z-contrasts. Using both techniques in tandem for representative sister samples on similar sample features and textures makes up for the deficiencies of each method alone. Mudstone samples from the Haynesville Shale are characterized with this technique and illustrate the complex texture, dominant porosity types, and the nature and distribution of organic matter, including bitumen, the solid residue resulting from the thermal conversion of kerogen to hydrocarbons.