Skip to Main Content
1
Previously: Aramco Research Center—Boston, 400 Technology Square, Cambridge, MA 02139

ABSTRACT

This chapter demonstrates a nondestructive, multispectral approach to evaluating chemical and spatial heterogeneities within mudstone fabrics. A combination of laser scanning confocal microscopy (LSCM) and scanning electron microscopy (SEM) was used to document nanometer- to millimeter-scale microtextures in mudstones. Additionally, micro-Fourier transform infrared (micro-FTIR) spectroscopy was used to identify both clay minerals and compositional structures, such as aromatic and aliphatic components in kerogen. A set of organic-rich mudstones with thermal maturities ranging from immature to oil prone were analyzed and used as examples to document the multispectral, multiscale approach. This work demonstrates the different spectral approaches and their applicability to the analysis of organic-rich mudstones. Single-channel fluorescence images collected with various excitation/emission wavelengths were used to access microtextural details in mudstones, whereas multichannel composite fluorescence images were used to evaluate relative thermal maturity among samples. In addition, SEM backscatter and energy dispersive X-ray microscopy were used to calibrate fluorescence signals to mineralogy and provide submicron information on grain boundaries and microfabrics. Micro-FTIR chemical maps represent the spatial distribution of chemical information related to properties of interest such as the presence and character of hydrocarbons and clay minerals. The infrared (IR) spectra associated with organic matter were also analyzed for quantitative indicators of thermal maturity. Opportunities for image processing and analysis that have the capability to integrate these multiscale, multispectral approaches are discussed for a more robust understanding of mudstone microfabrics, heterogeneity, and their impact on mudstone reservoir quality.

INTRODUCTION

Evaluation of mudstone reservoir quality relies heavily on understanding the kerogen type and maturity as well as the arrangement of the host sediments. Quantitative analysis of both the inorganic and the organic components is fundamental to understanding the bulk rock properties, such as total organic carbon content and brittleness that affect reservoir exploitation. Standard methods to evaluate and quantify kerogen type and maturity determine chemical composition from destructive bulk laboratory analyses that rely on homogenized crushed samples (Peters, 1986; Peters and Cassa, 1994). Although these methods are well known and reliable, it is difficult to correlate the results with the original mudstone microfabrics. A new approach is needed that can reliably assess geochemical and mineralogical composition without destroying the spatial information recorded in the rock texture. This chapter describes three methods for spectral characterization of mineralogy and geochemistry based on nondestructive measurements: laser scanning confocal microscopy (LSCM), scanning electron microscopy (SEM), and micro-Fourier transform infrared (micro-FTIR) imaging.

Each of the spectroscopic techniques used in this study measures specific molecular responses from which physical and chemical properties, such as molecular bonding and elemental arrangement, can be inferred (Figure 1). First, LSCM is a fluorescence imaging method that relies on the light response as a result of electronic transitions from UV to visible light excitation. Laser scanning microscopes use point illumination and a pinhole to significantly improve the signal-to-noise ratio and support quantitative measurements at relatively high spatial resolution (~250 nm). This resolution makes them superior to traditional UV light microscopes. Furthermore, advances in acquisition and image processing techniques allow automated acquisition of large-area, high-resolution scans for quantitative analysis, as well as the spatial coverage for interpretation of rock fabrics. Second, SEM is a well-documented and mature imaging technique that relies on surface scanning of a sample with a focused electron beam. The electrons interact with atoms in the sample producing backscattered and secondary electrons as well as characteristic X-rays (Reed, 1996). Backscattered electron (BSE) images provide high-resolution (down to nanometer scale) textural information, whereas X-rays provide spatially relevant elemental information. Finally, FTIR spectroscopy measures infrared (IR) absorption spectra related to the characteristic vibrational stretching and bending modes of molecular bonds that are characteristic of specific chemical functional groups (Chen et al., 2015). Typically, FTIR spectra are collected for areas the size of a few millimeters, which is not ideal for heterogeneous materials with nano- and microscale variations. To improve resolution and signal-to-noise ratio, this work used a micro-FTIR imaging system with a focal plane array to simultaneously acquire multiple spectra across a sample area in a single measurement providing multispectral information for advanced analysis. The multispectral data are used to identify minerals, for example complex clay and carbonates, and to distinguish specific compositional structures within kerogen, such as aliphatic and aromatic carbon domains. The data also provide a means to document the relationship between organic content and the surrounding microtextures and mineralogy, which is not possible using traditional, bulk FTIR analysis. Documenting the relationship between the organic content and the surrounding microtextures, and then linking this relationship to the larger scale rock fabric is a necessary first step in evaluating reservoir quality in mudstones. Thus, a combined multispectral, SEM, LSCM, and micro-FTIR imaging approach supports quantitative, nondestructive evaluation of the chemical and spatial heterogeneity within mudstone fabrics.

Figure 1.

Electromagnetic spectrum highlighting the spectral wavelengths being used in this study. Each of the imaging techniques presented here relies on specific molecular responses from which physical and chemical properties can be inferred.

Figure 1.

Electromagnetic spectrum highlighting the spectral wavelengths being used in this study. Each of the imaging techniques presented here relies on specific molecular responses from which physical and chemical properties can be inferred.

MATERIALS

A set of organic-rich mudstones with thermal maturities ranging from immature to oil prone were analyzed and used as examples to document the spectral approaches described. Samples used come from a Silurian-aged regionally extensive organic-rich mudstone in Saudi Arabia (Hayton et al., 2017). Two-centimeter (~0.8 in.) diameter minicores were extracted from mudstone samples and small (<1 cm [~0.4 in.] thick) billets were cut and ion-milled to reduce surface roughness. Organic richness and maturity for each of the minicores was determined by pyrolysis using a Rock-Eval 6©. Powdered clay standards used in calibration of micro-FTIR data were acquired from Ward’s Natural Science and sieved using a no. 100 mesh (~150 microns).

LASER SCANNING CONFOCAL MICROSCOPY

Fluorescence microscopy is commonly used to identify fluorescent minerals, characterize petroleum fluid inclusions (McDougall, 1952; Robbins, 1983; Gaft et al., 2015), or identify hydrocarbons or organic matter (Teichmuller and Wolf, 1977; Robert, 1981; Powell et al., 1982; Burruss, 1991). Wide-field fluorescence techniques, however, have several limitations, such as low resolution and polychromatic excitation that make it nonideal for imaging mudstones with nanoscale textural variations as well as significant chemical heterogeneity. LSCM relies on pixel-by-pixel laser-induced fluorescence detected by a highly sensitive photomultiplier tube (PMT) to generate high-resolution images. Many confocal microscopes use a series of lasers with various wavelengths to excite specific components of interest in heterogeneous or composite materials. The fluorescence can either be collected in series or parallel depending on the spectral overlap for the fluorophores present. Traditionally, this tool is used in the biological community to image cells, study tissues, examine bio-dynamics, and for other high-resolution imaging applications (Masters, 1998; Diaspro, 2001; Matsumoto, 2003; Kadam et al., 2017). In biological applications, either fluorescent dyes are used to label specific entities or auto-fluorescence and wavelength specificity are collected as the signal. More recently, confocal microscopy has been used in the petroleum industry to identify rock features such as porosity or pore-size distribution by imbibing carbonate and sandstone pores with fluorescently tagged epoxy and creating thin sections to provide both 2-D and 3-D information (Bereskin et al., 1993; Fredrich, 1999; Lindquist and Venkatarangan, 1999; Menéndez et al., 2001; Petford et al., 2001; Al Ibrahim et al., 2012). In other cases, the intrinsic fluorescence has been linked to differences in burial history, thermal maturity, and other geochemical factors for coals, shales, and localized macerals (Li et al., 1996; Stasiuk, 1999; Xiao et al., 2002; Hackley and Kus, 2015; Kus, 2015; Hackley and Cardott, 2016).

Single-Channel Confocal Imaging

Confocal images were generated to document textural relationships between organic matter and surrounding minerals and to make qualitative comparisons of maturity among samples with different thermal histories. Individual confocal fluorescence images were generated for textural analysis using an LSM700 confocal microscope (Zeiss Inc., Germany). This equipment is capable of sequentially collecting up to four-channel fluorescence images, where each channel was set to use a different excitation laser and fluorescence collection filter set (Table 1). Given that the excitation and emission spectra for each component of the sample was unknown, the image channels were configured using predefined Alexa® brand fluorescent dyes that are common in most dye spectral libraries. After testing the range of Alexa® dyes available in Zen Black©, it was determined that the settings for the Alexa® 405 (Blue), Alexa® 488 (Green), Alexa® 568 (Orange/Red), and Alexa® 633 (Red) dyes provided optimal fluorescence signals for imaging organic and inorganic components in mudstone samples. Individual confocal images were collected in 16-bit grayscale and a false-color map was applied, corresponding to the dye settings used. Images were generated using a 20× objective over an area of 1600 × 1600 pixels, with a pixel size of 200 nm. The LSM700 is equipped with a motorized, software-controlled microscope stage that supports large area scans. Scans up to several centimeters across were achieved by stitching individual (1600 × 1600 pixel) images together with a 10% edge overlap. Automatic stitching was done during image acquisition by the Zen Black© software. To reduce noise, the pinhole for these images was set between 1 and 1.5 Airy units (AU). The PMT gain and offset for individual images was optimized to generate the best quality image per channel. Therefore, the intensity range varies from image to image and must be normalized for comparison. Image normalization was done in Matlab® to convert the grayscale images with a dynamic intensity range (min, max) to a uniform intensity range of 0–255. For each image the normalized intensity for each pixel was calculated by y = ((x-min) * 255)/(max-min), where x is the grayscale intensity of the original pixel, y is the grayscale intensity after normalization, and min-max represents the dynamic range of the original image grayscale intensity.

Table 1.

Excitation and emission wavelengths for each channel by channel color.

MicroscopeChannel ColorExcitation Wavelength (nm)Emission Wavelength Range (nm)Emission Max (nm)
LSM 700/780Blue405400–513421
LSM 700/780Green488490–633519
LSM 700Orange/Red568550–700603
LSM 700/780Red633638–747647
MicroscopeChannel ColorExcitation Wavelength (nm)Emission Wavelength Range (nm)Emission Max (nm)
LSM 700/780Blue405400–513421
LSM 700/780Green488490–633519
LSM 700Orange/Red568550–700603
LSM 700/780Red633638–747647

Single-channel confocal images of an organic-rich mudstone were acquired to assess the differences in microtextures that could be detected by the different spectral bands recorded in each image (Figure 2). Geochemical analysis by Rock-Eval 6© pyrolysis for this sample yielded a total organic carbon of 5.33 wt. %, equivalent vitrinite reflectance (%Ro) of 1.02, and a hydrogen index of 1.85 mg HC/g TOC. Figures 2A–D show each of the false-colored fluorescence channels with their full intensity range. In a postprocessing step, the fluorescence intensities were normalized and a single-color map was applied to accentuate subtle differences in intensity that highlighted microtextural differences (Figures 2A′–D′). The normalized images show a much greater contrast in intensity for the surrounding matrix, in some cases highlighting what look like specific mineral grains and laminated bedding.

Figure 2.

Single-channel confocal images of an organic-rich mudstone. Images were collected in 16-bit grayscale and a false-color map corresponding to the settings for the Alexa® 405 (Blue), Alexa® 488 (Green), Alexa® 568 (Orange/Red), and Alexa® 633 (Red) dyes. (A–D) show each of the false-colored fluorescence channels with their full intensity range. In a postprocessing step, the fluorescence intensities were normalized and a single-color map applied (A′–D′) to accentuate subtle differences in intensity to highlight microtextural differences. The normalized images show a much greater contrast in intensity for the surrounding matrix.

Figure 2.

Single-channel confocal images of an organic-rich mudstone. Images were collected in 16-bit grayscale and a false-color map corresponding to the settings for the Alexa® 405 (Blue), Alexa® 488 (Green), Alexa® 568 (Orange/Red), and Alexa® 633 (Red) dyes. (A–D) show each of the false-colored fluorescence channels with their full intensity range. In a postprocessing step, the fluorescence intensities were normalized and a single-color map applied (A′–D′) to accentuate subtle differences in intensity to highlight microtextural differences. The normalized images show a much greater contrast in intensity for the surrounding matrix.

Correlation to SEM EDS Elemental Maps

Fluorescence images alone are insufficient to determine mineralogical and sedimentological character of a mudstone. When combined with SEM EDS (energy dispersive X-ray) maps, the fluorescence intensities can be correlated to specific elemental signatures. Furthermore, SEM BSE maps provide additional details about grain morphology and spatial distribution in the mudstone fabric, and when combined with elemental information, they provide a means to infer mineral relationships. SEM data were collected on a Zeiss Sigma HDVP (Zeiss, Inc., Germany) using a 15.0 kV accelerating voltage at a 10.0 mm (0.4 in.) working distance. Total magnification was 457× over an area of 240 × 170 microns with a pixel size of 244 nm for backscattered images. The EDS data were collected using a single Bruker XFlash® 6/30 SDD detector for polished samples to avoid shadowing and reduce the need for a second EDS detector. Elemental maps were acquired and processed using Bruker’s Espirit© 2.0 software. Although the spatial resolution of the EDS signal is limited by the detector and accelerating voltage, the elemental data were artificially mapped to 244 nm/pixel to match the BSE maps. The spatial resolution of the EDS signal is actually larger than a micron at the accelerating voltage that was chosen. The signal processing software provided by Bruker allowed for the data to be down-sampled so that the elemental maps matched the BSE image resolution.

Overlaying the fluorescence image with the BSE and EDS maps then allows the interpretation of specific minerals in several areas. As an example, Figure 3 shows a colocated SEM EDS overlay image, a BSE image, a normalized confocal fluorescence as well as all the images overlain. Most obvious are the framboidal pyrite grains, marked by high-intensity fluorescence and an abundance of iron. The black areas on the BSE and EDS maps with moderate fluorescence are interpreted as organic matter. Other mineral grains, which are high in silica and aluminum, can be identified on both the SEM and fluorescence maps and are interpreted as quartz and clay minerals, respectively. Further quantitative image analysis is needed to tie specific spectral responses from confocal and SEM to individual minerals.

Figure 3.

Examples of SEM and single-channel confocal images. (A) Backscattered electron image showing detailed grain boundaries and rock fabric. (B) Composite EDS images where colors are representative of specific elements. (C) Single-channel confocal fluorescence image using the settings for the Alexa® 633 (Red) dye, with a normalized color map applied. (D) Overlay of all the images (BSE, EDS, and confocal).

Figure 3.

Examples of SEM and single-channel confocal images. (A) Backscattered electron image showing detailed grain boundaries and rock fabric. (B) Composite EDS images where colors are representative of specific elements. (C) Single-channel confocal fluorescence image using the settings for the Alexa® 633 (Red) dye, with a normalized color map applied. (D) Overlay of all the images (BSE, EDS, and confocal).

Multi-Channel Composite Imaging

To access thermal maturity among different samples, multichannel, composite confocal images were generated for qualitative comparison using a Zeiss LSM780 (Zeiss, Inc., Germany). This microscope differs from the LSM700 in that (i) the detector (GaAsP (LSM780) vs. metal alkali (LSM700)) is significantly more sensitive and provides a more consistent response across the visible light spectrum, (ii) it provides faster data collection, and (iii) a 34-channel spectral detector is available (not used here). As in the single-channel imaging experiments, the fluorescence spectra for the individual rock components were unknown so the detector settings were configured for Alexa® dyes that were selected to provide the best materials contrast in these samples. In this case, the microscope was used to collect three-channel fluorescence images, using the settings for the Alexa® 405 (Blue), Alexa® 488 (Green), and Alexa® 633 (Red) dyes. A 20× objective was over a 1024 × 1024 pixel area making the pixel size for these images 415 nm. For each multichannel composite image, three channels were defined using the aforementioned Alexa® dye presets and a 16-bit grayscale image was collected for each channel in series. After collection, each image was false colored and put in overlay. For all images, the gain and offset for each channel were set based on the fluorescence from the least mature sample and used for all subsequent images. In addition, the pinhole was set for 1 AU to produce the best confocality and reduce the effect of out-of-plane light in the collected images. Finally, the images shown here have been scaled to show the full intensity range from 0 to 65535 for each channel. The importance of the gain, offset, bit depth, and displayed intensity range is discussed further below.

Composite confocal images for a series of mudstone samples of varying maturity were collected as described above with excitation and emission specifications (see Table 1) selected based on preset ranges for Alexa® series dyes and the instrument configuration. Representative false-colored overlay images of the red, green, and blue channels are shown in Figure 4 along with sample properties, equivalent vitrinite reflectance (%Ro), total organic carbon content (TOC in wt. %), and hydrogen index (HI in mg/g TOC). From visual inspection, there are clear differences in the overall intensity of the fluorescence from one image to the next with the overall fluorescence signal for the most mature sample (%Ro 2.25) being the lowest. Additionally, in overlay, the color mixing of the red, green, and blue channels is different for each sample. Specifically, for the least mature sample (%Ro 0.63), the intensity and signal coverage of the red and blue channels are similar, making the image appear almost violet in color. However, for the intermediate maturity sample (%Ro 1.02), the red channel is significantly more intense, appearing brighter and making that overlay appear mostly red. It should be noted that changes in fluorescence can be attributed to both the organic and the inorganic components of the rock samples. Although we cannot control completely for inorganic fluorescence, we chose to use samples that were of similar lithological and geochemical composition, thus reducing the likelihood that the changes in the fluorescence were specific to changes in mineralogy alone.

Figure 4.

Representative confocal fluorescence images showing the three-channel overlay for three samples of different maturity. The equivalent vitrinite reflectance (%Ro) from left to right is 0.63, 1.02, and 2.25, respectively.

Figure 4.

Representative confocal fluorescence images showing the three-channel overlay for three samples of different maturity. The equivalent vitrinite reflectance (%Ro) from left to right is 0.63, 1.02, and 2.25, respectively.

From the qualitative observations described here, such images can clearly be leveraged to provide rich quantitative information on thermal maturity as outlined by Hackley and Kus (2015). For image analysis purposes, there are a number of important factors that should be carefully addressed if the images are used for quantitative comparisons. One key pitfall to quantitative image analysis is related to the photomultiplier (detector) gain and offset that are used to multiply and shift image intensity scale such that the full grayscale range of the detector is exploited. Traditionally, the detector gain and offset would be adjusted for each sample to provide the best image quality and contrast, but because changing the detector settings also changes the image intensity and grayscale, these parameters should remain unchanged from image to image. The overall image intensity should be kept consistent for each sample to ensure that detector settings (gain and offset) are not skewing the image data analysis. Another key factor is related to the bit depth (dynamic range) selected for image acquisition as this is directly related to the ability to resolve small differences in intensity from the number of photons collected per pixel. Typically, 8 bits are used, providing 28 (0–255) intensity levels, but a higher bit depth, 16 bits (0–65535 intensity range), is preferred when the differences in the produced fluorescence are small from one sample to the next. Finally, postprocessing to improve the displayed image quality should be avoided for quantitative comparisons and image processing. In most cases, the instrument software provides a series of common display options to improve the image appearance on screen but caution is needed to ensure that raw or unprocessed images are exported and used for analysis rather than the optimized display images. The raw data images maintain the full image intensity range provided by the bit depth selected for each channel used in the overlay image. Adjusting the intensity range for each channel independently to optimize the display image can skew the color mixing for each image and affect data comparisons.

Micro-FTIR Spectroscopy

FTIR spectroscopy has been used in the geosciences for decades to study the molecular structure of organic and inorganic components (Chen et al., 2015). Applications of FTIR spectroscopy include melt and fluid inclusion studies, mineralogical, microfossil, and palynological identification as well as chemical characterization of coals and organic-rich shales (Iglesias et al., 1995; Di Matteo et al., 2004; Lis et al., 2005; Rossman, 2006; D’Angelo and Zodrow, 2011). A detailed introduction to the technique and its applications to organic-rich rocks can be found in Rouxhet et al. (1980), Tissot and Welte (1984), and Ganz and Kalkreuth (1991). These studies document specific absorbance peaks related to the chemical functional groups found in kerogen, bitumen, asphaltenes, and coals. Absorbance of molecular vibrations under IR radiation is proportional to the abundance of a particular functional group (e.g., C–H, O–H, C=O) (Chen et al., 2015). A challenge in FTIR spectroscopy as applied to organic-rich rocks is the correct assignment of absorbance peaks to the corresponding functional group and organic matter. Furthermore, quantification based on intensity and position of these peaks requires calibration and is often done on bulk homogenized samples. Advances in IR microscopy and the advent of mercury cadmium telluride (HgCdTe) detectors enabled new instruments with increased spatial resolution and better sensitivity (Washburn et al., 2015). Micro-FTIR imaging has recently become a popular method for characterizing compositionally complex materials such as heterogeneous organic-rich rocks (Chen et al., 2014; Gasaway et al., 2017). These two studies report chemical mapping of mudstones using reflectance techniques with a spatial resolution of 20 μm. There are three commonly used techniques for FTIR analysis: (1) transmission, (2) attenuated total reflection (ATR), and (3) diffuse reflection. Here, ATR-FTIR is used where an internal reflection element (ATR crystal) creates an evanescent wave to provide chemical information from only the functional groups nearest to the surface of the sample. The reduced light penetration provided by the ATR mode improves the FTIR signal-to-noise ratio (Chen et al., 2015). Several recent studies report the use of ATR methods for mineral analysis (Müller et al., 2014) and assessment of organic materials and their maturity (Li et al., 2007; Bonoldi et al., 2016).

Micro-FTIR Chemical Mapping

Micro-FTIR images were collected using a Bruker Hyperion 3000 (Bruker Inc, Billerica MA, USA) microscope with a 100-micron-diameter germanium ATR tip, an IR objective with 20× magnification, and a 32 μm × 32 μm fixed aperture. The microscope was equipped with a liquid nitrogen-cooled, 64 μm × 64 μm, mercury cadmium telluride focal plane array connected to a Vertex 70 FTIR spectrometer. The focal plane array simultaneously collects 4096 spectra (64 × 64) over the fixed area, in this case 32 × 32 microns. The spatial resolution over this area is diffraction limited to approximately 2 μm with no pixel binning. Pixel binning and averaging over several scans can greatly improve signal-to-noise ratios. Nitrogen gas is run at low levels into the Vertex spectrometer and into the microscope to minimize spurious peaks resulting from water vapor and CO2.

ATR measurements require careful sample preparation and care was taken to ion-mill the samples with a low energy flux to ensure that sample alteration was minimized prior to probing with the ATR. A range of pressure settings for the ATR tip was investigated, as too much pressure can cause surface damage to the sample and too low a pressure yields poor spectral data as a result of poor sample contact. A medium pressure was chosen as it yielded the best signal-to-noise ratio for the samples analyzed. A background IR signal was acquired and subtracted by slightly lifting the ATR tip from the sample to ensure that signal changes were minimized.

Micro-FTIR chemical maps were produced to document the area around a single maceral in an organic-rich mudstone. The maps were generated by stitching together several spectral maps with 4 μm/pixel resolution to provide larger areal coverage. In these examples, data was averaged over 128 scans using a 2 × 2 pixel binning. Each pixel has a corresponding average FTIR spectra, where the color of each pixel is related to the integration of the peak intensity over a defined wavenumber range. Therefore, the pixel colors correspond to the relative concentration of a specific chemical functional group with warm colors representing a higher concentration. Two example chemical maps are shown in Figure 5: one for clays (A) and one for hydrocarbons (B). Representative spectra are also shown for an individual pixel highlighting the wavenumber ranges, 3500–3800 cm−1 and 2800–3000 cm−1, respectively, over which the maps were generated. The maps clearly identify the spatial distribution and boundaries for both the clays and hydrocarbons. The lack of signal in the clay spectral maps also highlights areas of hydrocarbon abundance. The use of chemical maps such as these provides a robust analysis technique for understanding the spatial distribution and relationships among different chemical signatures.

Figure 5.

Two example micro-FTIR chemical maps for clays (A) and hydrocarbons (B). These maps were generated by stitching together several 32 μm spectral maps (inset boxes). Each inset box (27 total for each image) represents an individual point of contact with the ATR Germanium tip. FTIR data at each ATR point was collected and averaged over 128 scans using a 2 × 2 pixel binning. The resulting 32 × 32 micron spectral maps (inset boxes) that correspond to individual ATR points were then stitched together with no overlap to form the larger image. Pixel color represents the integration of the peak intensity over a defined wavenumber range. Therefore, the pixel colors correspond to the relative concentration of a specific chemical functional group where warm colors represent higher concentration. Representative spectra (C) are also shown for an individual pixel, highlighting the wavenumber ranges over which the maps were generated, 3500–3800 cm−1 and 2800–3000 cm−1, respectively.

Figure 5.

Two example micro-FTIR chemical maps for clays (A) and hydrocarbons (B). These maps were generated by stitching together several 32 μm spectral maps (inset boxes). Each inset box (27 total for each image) represents an individual point of contact with the ATR Germanium tip. FTIR data at each ATR point was collected and averaged over 128 scans using a 2 × 2 pixel binning. The resulting 32 × 32 micron spectral maps (inset boxes) that correspond to individual ATR points were then stitched together with no overlap to form the larger image. Pixel color represents the integration of the peak intensity over a defined wavenumber range. Therefore, the pixel colors correspond to the relative concentration of a specific chemical functional group where warm colors represent higher concentration. Representative spectra (C) are also shown for an individual pixel, highlighting the wavenumber ranges over which the maps were generated, 3500–3800 cm−1 and 2800–3000 cm−1, respectively.

In addition to mapping aggregate clay mineral signatures, micro-FTIR can be used to identify individual clay minerals once calibrated to standard samples. To identify the clay components present in Figure 5C, three common clay minerals were chosen for bulk FTIR analysis. These clays included illite, kaolinite, and montmorillonite. Spectra from bulk clay powder standards were acquired and analyzed using an Agilent Cary 630 FTIR spectrometer with a diamond ATR tip and a 1000 μm probe spot size. The powders were sieved to 100 mesh as this grain size maximized the IR peak signal for each of the clay standards. For each clay type in Figure 5C, we show two absorbance spectra to demonstrate the consistency of data.

Kaolinite has two pronounced narrow peaks at 3688 cm−1 and 3621 cm−1 with a relatively high absorbance. Montmorillonite has a broad peak at 3619 cm−1 and two very broad peaks at 3400 cm−1 and 3240 cm−1, with moderate absorbance values. Lastly, illite has a small peak at 3693 cm−1, a broad peak at 3619 cm−1, and a very broad one at 3386 cm−1. Illite has the lowest absorbance values among the three clay minerals. This signal is a combination of absorbance and the difference in refractive index between clays. This should be taken into account when assessing the presence of illite in a heterogeneous sample. To separate the contributions from absorbance and refractive index, careful ellipsometric measurements are required, which are beyond the scope of this work.

The unique spectral combinations illustrated here for kaolinite, montmorillonite, and to a lesser extent illite could be used, given a more extensive analysis, to map individual clays and provide a more precise approach to mapping clay minerals and other textural components in mudstones.

Spectral Maturity Indicators

Quantitative spectral responses related to hydrocarbon maturity can also be observed in addition to qualitative spatial observations from micro-FTIR chemical maps. The hydrocarbon structure of organic matter changes during the burial and maturation process and this change is reflected in the FTIR spectral response. There are distinct spectral peaks related to hydrocarbons in organic matter (Rouxhet et al., 1980; Ganz and Kalkreuth, 1991): the CH2 and CH3 aliphatic groups at around 2860 and 2930 cm−1 (C-H stretches), the carboxyl and carbonyl groups at 1710 cm−1 (C=O stretches) and the aromatic C=C bands at 1630 cm−1. With increasing maturation the aliphatic peaks decrease as hydrogen is lost.

The vibration bands around 3000 cm−1 (Figure 6) are associated with both aromatic >3000 cm−1 and aliphatic <3000 cm−1 hydrocarbon C-H stretching. For samples from the same general area and same organic matter type, differences in these bands can be used to assess the relative maturity in organic-rich samples. For example, the ratio of CH3/CH2 (defined as the intensity ratio of peaks at 2955 cm−1/2920 cm−1) is commonly used to assess the chain length and degree of branching of aliphatic side groups (Marshall et al., 2005). The area under the C-H peaks is also proportional to hydrogen content and may be used as a proxy for hydrogen index (Washburn and Birdwell, 2013). As organic matter matures, hydrogen is lost as hydrocarbons are converted from aliphatic chains to aromatic rings. An example of this can be seen in Figure 6, where the spectra from three organic-rich mudstones of varying maturity are shown. The CH3/CH2 ratios presented are calculated from the average absorbance for each of the three samples at 2955 cm−1 for CH3 and 2920 cm−1 for CH2 after a baseline correction. These ratios were then plotted against hydrogen index as determined from corresponding Rock-Eval pyrolysis data. Note that as the hydrogen index decreases, the aliphatic peak around 2920 cm−1 decreases and there is a slight increase in the broad aromatic peak around 3060 cm−1. The ratio of CH3/CH2 also decreases with increasing maturity, as shown by the peaks at 2955 cm−1 and 2920 cm−1, respectively, which are closer together and more pronounced for the more mature sample.

Figure 6.

Six representative spectra from three organic-rich mudstones of varying maturity. Each spectra (6 total, 2 per sample) is an average of 128 FTIR scans taken over a single ATR contact point. The two spectra from each sample represent adjacent pixels from one ATR contact point/spectral map. Note that as hydrogen index decreases, the aliphatic peak around 2920 cm−1 decreases and there is a slight increase in the broad aromatic peak around 3060 cm−1. The ratio of CH3/CH2 also decreases with increasing maturity, as shown by the CH3 and CH2 peaks at 2955 cm−1 and 2920 cm−1, respectively.

Figure 6.

Six representative spectra from three organic-rich mudstones of varying maturity. Each spectra (6 total, 2 per sample) is an average of 128 FTIR scans taken over a single ATR contact point. The two spectra from each sample represent adjacent pixels from one ATR contact point/spectral map. Note that as hydrogen index decreases, the aliphatic peak around 2920 cm−1 decreases and there is a slight increase in the broad aromatic peak around 3060 cm−1. The ratio of CH3/CH2 also decreases with increasing maturity, as shown by the CH3 and CH2 peaks at 2955 cm−1 and 2920 cm−1, respectively.

MULTISCALE–MULTISPECTRAL IMAGE INTEGRATION

Arrangement of organic matter and mineral components and their interactions during hydrocarbon generation are first-order controls on productivity in mudstone reservoirs. A combination of LSCM, SEM EDS elemental maps, and micro-FTIR imaging can provide qualitative and semi-quantitative insights into textural variations and chemical heterogeneity at a variety of scales. Each of these approaches provides a unique understanding of both the organic and the inorganic components needed to assess bulk rock properties such as reservoir quality and maturity. When combined, the sub-micron textural and chemical analyses can be correlated to millimeter scale confocal fluorescence images (Figure 7). Advanced image analysis and machine learning approaches used in multi-scale image processing applications, such as thin section or satellite imagery/remote sensing data classification (Marmo et al., 2005; Yu et al., 2012), can be applied to these data to quantitatively link microscale spectral responses to large area confocal images. Integration of multispectral data supports the documentation of genetic relationships between organic content, the surrounding mineralogy, and larger scale rock fabric. Traditional geochemical analyses alone cannot provide the same textural context.

Figure 7.

(A) Large-area single-channel confocal fluorescence image of an organic-rich mudstone, using the settings for the Alexa® 405 dye. Bright yellows and greens have been correlated to fluorescent organic macerals based on fluorescent signature and morphology. (B) Represents a correlative area where micro-FTIR data was acquired to document the presence of hydrocarbons. (C) Denotes an area where submicron SEM data was acquired to provide higher resolution images of microtextures and verify the morphology of organic vs. inorganic fluorescence signatures.

Figure 7.

(A) Large-area single-channel confocal fluorescence image of an organic-rich mudstone, using the settings for the Alexa® 405 dye. Bright yellows and greens have been correlated to fluorescent organic macerals based on fluorescent signature and morphology. (B) Represents a correlative area where micro-FTIR data was acquired to document the presence of hydrocarbons. (C) Denotes an area where submicron SEM data was acquired to provide higher resolution images of microtextures and verify the morphology of organic vs. inorganic fluorescence signatures.

CONCLUSIONS

This work focused on demonstrating a non-destructive, multispectral approach for the evaluation of chemical and spatial heterogeneities within mudstone fabrics. Specifically, the utility of three methods for spectral characterization of mineralogy and geochemistry were described and the following conclusions were reached.

  • (1) LSCM is a nondestructive imaging technique that can be used to characterize both organic and inorganic components in mudstones. High-resolution images (nanometers per pixel) as well as large-area, stitched images (millimeter to centimeter scale) can be generated to document textural relationships between organic matter and surrounding minerals. Single-channel confocal images can be collected with various excitation/emission wavelengths and can be normalized and used to evaluate microtextural differences in mudstones.

  • (2) SEM BSE and EDS maps provide detailed nanometer-scale information that can be calibrated to confocal fluorescence images for robust grain boundary and mineral identification.

  • (3) Multichannel composite confocal images can be used for qualitative and semi-quantitative comparison of organic maturity among mudstone samples with different thermal histories. Quantitative image analysis can also provide additional insights, but care must be taken to account for image acquisition and display parameters.

  • (4) Micro-FTIR chemical mapping is a nondestructive analytical technique that can be used to understand the spatial distribution of chemical information at the micron scale. The relative concentrations of specific chemical functional groups, such as hydrocarbons, clays, and complex minerals, can be spatially correlated and compared.

  • (5) Micro-FTIR spectra can also be used to provide quantitative indicators of thermal maturity. Relative peak intensities as well as CH3/CH2 ratios can be used as a proxy for hydrogen index and thus organic maturity.

  • (6) Integration of nanometer-scale textural information from confocal and SEM with micron-scale chemical information from FTIR delivers a powerful tool for understanding organic and inorganic components in mudstones. Moreover, image processing allows for these submicron-scale observations to be correlated to larger scale confocal fluorescence signatures that can be imaged across several millimeters.

REFERENCES CITED

Al Ibrahim
,
M. A.
,
N. F.
Hurley
,
W.
Zhao
, and
D.
Acero-Allard
,
2012
,
An automated petrographic image analysis system: Capillary pressure curves using confocal microscopy
:
SPE Annual Technical Conference and Exhibition
,
San Antonio, Texas, USA
, 8–10 October, Article 159180.
Bereskin
,
R. G. C.
,
G. D.
Lord
, and
L. B.
Owen
,
1993
,
Quantitative porosity measurements of known reservoirs using confocal microscopy
:
AAPG Bulletin
 , v.
77
, p.
690
.
Bonoldi
,
L.
,
L.
Di Paolo
, and
C.
Flego
,
2016
,
Vibrational spectroscopy assessment of kerogen maturity in organic-rich source rocks
:
Vibrational Spectroscopy
 , v.
87
, p.
14
19
.
Burruss
,
R. C.
,
1991
,
Practical aspects of fluorescence microscopy of petroleum fluid inclusions
, in
C. E.
Barker
and
O. C.
Kopp
, eds.,
Luminescence microscopy: Quantitative and qualitative aspects
 : SEPM Short Course 25, p.
1
7
.
Chen
,
Y.
,
M.
Mastalerz
, and
A.
Schimmelmann
,
2014
,
Heterogeneity of shale documented by micro‐FTIR and image analysis
:
Journal of Microscopy
 , v.
256
, no.
3
, p.
177
189
.
Chen
,
Y.
,
C.
Zou
,
M.
Mastalerz
,
S.
Hu
,
C.
Gasaway
, and
X.
Tao
,
2015
,
Applications of micro-Fourier Transform Infrared Spectroscopy (FTIR) in the geological sciences—A review
:
International Journal of Molecular Sciences
 , v.
16
, p.
30223
30250
.
D’Angelo
,
J. A.
, and
E. L.
Zodrow
,
2011
,
Chemometric study of functional groups in different layers of Trigonocarpus grandis ovules (Pennsylvanian seed fern, Canada)
,
Organic Geochemistry
 , v.
42
, p.
1039
1054
.
Di Matteo
,
V.
,
M. R.
Carroll
,
H.
Behrens
,
F.
Vetere
, and
R. A.
Brooker
,
2004
,
Water solubility in trachytic melts
:
Chemical Geology
 , v.
213
, p.
187
196
.
Diaspro
,
A.
, ed.,
2001
,
Confocal and two-photon microscopy: Foundations, applications and advances
 ,
New York
,
Wiley-VCH
,
576
p.
Fredrich
,
J.
,
1999
,
3D imaging of porous media using laser scanning confocal microscopy with application to microscale transport processes: Physics and Chemistry of the Earth, Part A
:
Solid Earth and Geodesy
 , v.
24
, no.
7
, p.
551
561
, DOI: 10.1016/S1464-1895(99)00079-4.
Gaft
,
M.
,
R.
Reisfeld
, and
G.
Panczer
,
2015
,
Modern luminescence spectroscopy of minerals and materials
 :
Switzerland
,
Springer Mineralogy, Springer International Publishing
, DOI: 10.1007/978-3-319-24765-6_1.
Ganz
,
H. H.
, and
W.
Kalkreuth
,
1991
,
IR classification of kerogen type, thermal maturation, hydrocarbon potential and lithological characteristics
:
Journal of Southeast Asian Earth Science
 , v.
5
, p.
19
28
.
Gasaway
,
C.
,
M.
Mastalerz
,
F.
Krause
,
C.
Clarkson
, and
C.
Debuhr
,
2017
,
Applicability of micro‐FTIR in detecting shale heterogeneity
:
Journal of Microscopy
 , v.
265
, no.
1
, p.
60
72
.
Hackley
,
P. C.
, and
J.
Kus
,
2015
,
Thermal maturity of Tasmanites microfossils from confocal laser scanning fluorescence microscopy
:
Fuel
 , v.
143
, p.
343
350
.
Hackley
,
P. C.
, and
B. J.
Cardott
,
2016
,
Application of organic petrography in North American shale petroleum systems: A review
:
International Journal of Coal Geology
 , v.
163
, p.
8
51
, DOI: 10.1016/j.coal.2016.06.010.
Hayton
,
S.
,
A.
Rees
, and
M.
Vecoli
,
2017
,
A punctuated Late Ordovician and Early Silurian deglaciation and transgression: Evidence from a subsurface study area in northern Saudi Arabia
:
AAPG Bulletin
 , v.
101
, p.
863
886
, DOI: 10.1306/08251616058.
Iglesias
,
M. J.
,
A.
Jiménez
,
F.
Laggoun-Défarge
, and
I.
Suárez-Ruiz
,
1995
,
FTIR study of pure vitrains and associated coals
:
Energy Fuels
 , v.
9
, p.
458
466
.
Kadam
,
P.
,
R.
McAllister
,
J. S.
Urbach
,
K.
Sandberg
, and
S. C.
Mueller
,
2017
,
Live cell imaging and 3D analysis of angiotensin receptor type 1a trafficking in transfected human embryonic kidney cells using confocal microscopy
:
Journal of Visualized Experiments
 , v.
121
, p.
e55177
, DOI: 10.3791/55177.
Kus
,
J.
,
2015
,
Application of confocal laser-scanning microscopy (CLSM) to autofluorescent organic and mineral matter in peat, coals and siliciclastic sedimentary rocks—A qualitative approach
:
International Journal of Coal Geology
 , v.
137
, p.
1
18
.
Li
,
H.
,
W. K.
Wan
, and
J.
Yuan
,
1996
,
Characterization of bitumen emulsions by confocal scanning laser microscopy
:
Petroleum Society of Canada Annual Technical Meeting
,
Calgary, Alberta
, 10–12 June 1996,
12
p., DOI: 10.2118/96-29.
Li
,
Z.
,
P. M.
Fredericks
,
L.
Rintoul
, and
C. R.
Ward
,
2007
,
Application of attenuated total reflectance micro-Fourier transform infrared (ATR-FTIR) spectroscopy to the study of coal macerals: Examples from the Bowen Basin, Australia
:
International Journal of Coal Geology
 , v.
70
, no.
1
, p.
87
94
.
Lindquist
,
W.
, and
A.
Venkatarangan
,
1999
,
Investigating 3D geometry of porous media from high resolution images: Physics and Chemistry of the Earth, Part A
:
Solid Earth and Geodesy
 , v.
24
, no.
7
, p.
593
599
.
Lis
,
G. P.
,
M.
Mastalerz
,
A.
Schimmelmann
,
M. D.
Lewan
, and
B. A.
Stankiewicz
,
2005
,
FTIR absorption indices for thermal maturity in comparison with vitrinite reflectance R0 in type-II kerogens from Devonian black shales
:
Organic Geochemistry
 , v.
36
, p.
1533
1552
.
Marmo
,
R.
,
S.
Amodio
,
R.
Tagliaferri
,
V.
Ferreri
, and
G.
Longo
,
2005
,
Textural identification of carbonate rocks by image processing and neural networks: Methodology proposal and examples
:
Computers in Geoscience
 , v.
31
, p.
649
659
.
Marshall
,
C. P.
,
E. J.
Javaux
,
A.H.
Knoll
, and
M. R.
Walter
,
2005
,
Combined micro-Fourier transform infrared (FTIR) spectroscopy and micro-Raman spectroscopy of Proterozoic acritarchs: A new approach to palaeobiology
:
Precambrian Research
 , v.
138
, no.
3
, p.
208
224
.
Masters
,
B. R.
,
1998
,
Three-dimensional confocal microscopy of the human optic nerve in vivo
:
Optics Express
 , v.
3
, no.
10
, p.
356
359
.
Matsumoto
,
B.
,
2003
,
Cell biological applications of confocal microscopy
 :
London, UK
,
Academic Press
,
507
p.
McDougall
,
D.
,
1952
,
Some factors influencing fluorescence in minerals
:
American Mineralogist
 , v.
37
, p.
427
437
.
Menéndez
,
B.
,
C.
David
, and
A. M.
Nistal
,
2001
,
Confocal scanning laser microscopy applied to the study of pore and crack networks in rocks
:
Computers & Geosciences
 , v.
27
, no.
9
, p.
1101
1109
.
Müller
,
C. M.
,
B.
Pejcic
,
L.
Esteban
,
C. D.
Piane
,
M.
Raven
, and
B.
Mizaikoff
,
2014
,
Infrared attenuated total reflectance spectroscopy: An innovative strategy for analyzing mineral components in energy relevant systems
:
Scientific Reports
 , v.
4
, no.
6764
, DOI: 10.1038/srep06764.
Peters
,
K. E.
,
1986
,
Guidelines for evaluating petroleum source rock using programmed pyrolysis
:
AAPG Bulletin
 , v.
70
, p.
318
329
.
Peters
,
K. E.
, and
M. R.
Cassa
,
1994
,
Applied source rock geochemistry
, in
L. B.
Magoon
and
W. G.
Dow
, eds.,
The petroleum system—From source to trap
 : AAPG Memoir
60
, p.
93
120
.
Petford
,
N.
,
G.
Davidson
, and
J. A.
Miller
,
2001
,
Investigation of the petrophysical properties of a porous sandstone using confocal scanning laser microscopy
:
Petroleum Geoscience
 , v.
7
, no.
2
, p.
99
105
.
Powell
,
T.
,
S.
Creaney
, and
L.
Snowdon
,
1982
,
Limitations of use of organic petrographic techniques for identification of petroleum source rocks
:
AAPG Bulletin
 , v.
66
, no.
4
, p.
430
435
.
Reed
,
S. J. B.
,
1996
,
Electron microprobe analysis and scanning electron microscopy in geology
 :
Cambridge University Press
,
201
p.
Robbins
,
M.
,
1983
,
Activators in fluorescent minerals
, in
The collector’s book of fluorescent minerals
 :
Boston, MA
,
Springer
, p.
183
191
, DOI: 10.1007/978-1-4757-4792-8_9.
Robert
,
P.
,
1981
,
Classification of organic matter by means of fluorescence; application to hydrocarbon source rocks
:
International Journal of Coal Geology
 , v.
1
, no.
2
, p.
101
137
.
Rossman
,
G. R.
,
2006
,
Analytical methods for measuring water in nominally anhydrous minerals
:
Reviews in Mineralogy and Geochemistry
 , v.
62
, no.
1
, p.
1
28
, DOI: 10.2138/rmg.2006.62.1.
Rouxhet
,
P. G.
,
P. L.
Robin
, and
G.
Nicaise
,
1980
,
Characterization of Kerogens and of their evolution by infrared spectroscopy
, in
Durand
B.
, ed.,
Kerogen: Insoluble organic matter from sedimentary rocks
 :
Paris
,
Editions Technip
, p.
163
190
.
Stasiuk
,
L. D.
,
1999
,
Confocal laser scanning fluorescence microscopy of Botroyococuss alginate from boghead oil shale, Boltysk, Ukraine
:
Selective preservation of various micro-algal components: Organic Geochemistry
 , v.
30
, p.
1021
1026
, DOI: 10.1016/S0146-6380(99)00046-7.
Teichmuller
,
M.
, and
M.
Wolf
,
1977
,
Application of fluorescence microscopy in coal petrology and oil exploration
:
Journal of Microscopy
 , v.
109
, no.
1
, p.
49
73
.
Tissot
,
D.
, and
H.
Welte
,
1984
,
Petroleum formation and occurrence: A new approach to oil and gas exploration
 :
Berlin
,
Springer-Verlag
,
143
p.
Washburn
,
K. E.
, and
J. E.
Bidwell
,
2013
,
Multivariate analysis of ATR-FTIR spectra for assessment of oil shale organic geochemical properties
:
Organic Geochemistry
 , v.
63
, p.
1
7
.
Washburn
,
K. E.
,
J. E.
Birdwell
,
M.
Foster
, and
F.
Gutierrez
,
2015
,
Detailed description of oil shale organic and mineralogical heterogeneity via Fourier transform infrared microscopy
:
Energy & Fuels
 , v.
29
, no.
7
, p.
4264
4271
.
Xiao
,
X.
,
R. W.
Wilkins
,
D.
Liu
,
Z.
Liu
, and
J.
Shen
, J,
2002
,
Laser-induced fluorescence microscopy—Application to possible high rank and carbonate source rocks
:
International Journal of Coal Geology
 , v.
51
, no.
2
, p.
129
141
.
Yu
,
L.
,
A.
Porwal
,
E.
Holden
, and
M.
Dentith
,
2012
,
Towards automatic lithological classification from remote sensing data using support vector machines
:
Computers & Geosciences
 , v.
45
, p.
229
239
.

Figures & Tables

Figure 1.

Electromagnetic spectrum highlighting the spectral wavelengths being used in this study. Each of the imaging techniques presented here relies on specific molecular responses from which physical and chemical properties can be inferred.

Figure 1.

Electromagnetic spectrum highlighting the spectral wavelengths being used in this study. Each of the imaging techniques presented here relies on specific molecular responses from which physical and chemical properties can be inferred.

Figure 2.

Single-channel confocal images of an organic-rich mudstone. Images were collected in 16-bit grayscale and a false-color map corresponding to the settings for the Alexa® 405 (Blue), Alexa® 488 (Green), Alexa® 568 (Orange/Red), and Alexa® 633 (Red) dyes. (A–D) show each of the false-colored fluorescence channels with their full intensity range. In a postprocessing step, the fluorescence intensities were normalized and a single-color map applied (A′–D′) to accentuate subtle differences in intensity to highlight microtextural differences. The normalized images show a much greater contrast in intensity for the surrounding matrix.

Figure 2.

Single-channel confocal images of an organic-rich mudstone. Images were collected in 16-bit grayscale and a false-color map corresponding to the settings for the Alexa® 405 (Blue), Alexa® 488 (Green), Alexa® 568 (Orange/Red), and Alexa® 633 (Red) dyes. (A–D) show each of the false-colored fluorescence channels with their full intensity range. In a postprocessing step, the fluorescence intensities were normalized and a single-color map applied (A′–D′) to accentuate subtle differences in intensity to highlight microtextural differences. The normalized images show a much greater contrast in intensity for the surrounding matrix.

Figure 3.

Examples of SEM and single-channel confocal images. (A) Backscattered electron image showing detailed grain boundaries and rock fabric. (B) Composite EDS images where colors are representative of specific elements. (C) Single-channel confocal fluorescence image using the settings for the Alexa® 633 (Red) dye, with a normalized color map applied. (D) Overlay of all the images (BSE, EDS, and confocal).

Figure 3.

Examples of SEM and single-channel confocal images. (A) Backscattered electron image showing detailed grain boundaries and rock fabric. (B) Composite EDS images where colors are representative of specific elements. (C) Single-channel confocal fluorescence image using the settings for the Alexa® 633 (Red) dye, with a normalized color map applied. (D) Overlay of all the images (BSE, EDS, and confocal).

Figure 4.

Representative confocal fluorescence images showing the three-channel overlay for three samples of different maturity. The equivalent vitrinite reflectance (%Ro) from left to right is 0.63, 1.02, and 2.25, respectively.

Figure 4.

Representative confocal fluorescence images showing the three-channel overlay for three samples of different maturity. The equivalent vitrinite reflectance (%Ro) from left to right is 0.63, 1.02, and 2.25, respectively.

Figure 5.

Two example micro-FTIR chemical maps for clays (A) and hydrocarbons (B). These maps were generated by stitching together several 32 μm spectral maps (inset boxes). Each inset box (27 total for each image) represents an individual point of contact with the ATR Germanium tip. FTIR data at each ATR point was collected and averaged over 128 scans using a 2 × 2 pixel binning. The resulting 32 × 32 micron spectral maps (inset boxes) that correspond to individual ATR points were then stitched together with no overlap to form the larger image. Pixel color represents the integration of the peak intensity over a defined wavenumber range. Therefore, the pixel colors correspond to the relative concentration of a specific chemical functional group where warm colors represent higher concentration. Representative spectra (C) are also shown for an individual pixel, highlighting the wavenumber ranges over which the maps were generated, 3500–3800 cm−1 and 2800–3000 cm−1, respectively.

Figure 5.

Two example micro-FTIR chemical maps for clays (A) and hydrocarbons (B). These maps were generated by stitching together several 32 μm spectral maps (inset boxes). Each inset box (27 total for each image) represents an individual point of contact with the ATR Germanium tip. FTIR data at each ATR point was collected and averaged over 128 scans using a 2 × 2 pixel binning. The resulting 32 × 32 micron spectral maps (inset boxes) that correspond to individual ATR points were then stitched together with no overlap to form the larger image. Pixel color represents the integration of the peak intensity over a defined wavenumber range. Therefore, the pixel colors correspond to the relative concentration of a specific chemical functional group where warm colors represent higher concentration. Representative spectra (C) are also shown for an individual pixel, highlighting the wavenumber ranges over which the maps were generated, 3500–3800 cm−1 and 2800–3000 cm−1, respectively.

Figure 6.

Six representative spectra from three organic-rich mudstones of varying maturity. Each spectra (6 total, 2 per sample) is an average of 128 FTIR scans taken over a single ATR contact point. The two spectra from each sample represent adjacent pixels from one ATR contact point/spectral map. Note that as hydrogen index decreases, the aliphatic peak around 2920 cm−1 decreases and there is a slight increase in the broad aromatic peak around 3060 cm−1. The ratio of CH3/CH2 also decreases with increasing maturity, as shown by the CH3 and CH2 peaks at 2955 cm−1 and 2920 cm−1, respectively.

Figure 6.

Six representative spectra from three organic-rich mudstones of varying maturity. Each spectra (6 total, 2 per sample) is an average of 128 FTIR scans taken over a single ATR contact point. The two spectra from each sample represent adjacent pixels from one ATR contact point/spectral map. Note that as hydrogen index decreases, the aliphatic peak around 2920 cm−1 decreases and there is a slight increase in the broad aromatic peak around 3060 cm−1. The ratio of CH3/CH2 also decreases with increasing maturity, as shown by the CH3 and CH2 peaks at 2955 cm−1 and 2920 cm−1, respectively.

Figure 7.

(A) Large-area single-channel confocal fluorescence image of an organic-rich mudstone, using the settings for the Alexa® 405 dye. Bright yellows and greens have been correlated to fluorescent organic macerals based on fluorescent signature and morphology. (B) Represents a correlative area where micro-FTIR data was acquired to document the presence of hydrocarbons. (C) Denotes an area where submicron SEM data was acquired to provide higher resolution images of microtextures and verify the morphology of organic vs. inorganic fluorescence signatures.

Figure 7.

(A) Large-area single-channel confocal fluorescence image of an organic-rich mudstone, using the settings for the Alexa® 405 dye. Bright yellows and greens have been correlated to fluorescent organic macerals based on fluorescent signature and morphology. (B) Represents a correlative area where micro-FTIR data was acquired to document the presence of hydrocarbons. (C) Denotes an area where submicron SEM data was acquired to provide higher resolution images of microtextures and verify the morphology of organic vs. inorganic fluorescence signatures.

Table 1.

Excitation and emission wavelengths for each channel by channel color.

MicroscopeChannel ColorExcitation Wavelength (nm)Emission Wavelength Range (nm)Emission Max (nm)
LSM 700/780Blue405400–513421
LSM 700/780Green488490–633519
LSM 700Orange/Red568550–700603
LSM 700/780Red633638–747647
MicroscopeChannel ColorExcitation Wavelength (nm)Emission Wavelength Range (nm)Emission Max (nm)
LSM 700/780Blue405400–513421
LSM 700/780Green488490–633519
LSM 700Orange/Red568550–700603
LSM 700/780Red633638–747647

Contents

GeoRef

References

REFERENCES CITED

Al Ibrahim
,
M. A.
,
N. F.
Hurley
,
W.
Zhao
, and
D.
Acero-Allard
,
2012
,
An automated petrographic image analysis system: Capillary pressure curves using confocal microscopy
:
SPE Annual Technical Conference and Exhibition
,
San Antonio, Texas, USA
, 8–10 October, Article 159180.
Bereskin
,
R. G. C.
,
G. D.
Lord
, and
L. B.
Owen
,
1993
,
Quantitative porosity measurements of known reservoirs using confocal microscopy
:
AAPG Bulletin
 , v.
77
, p.
690
.
Bonoldi
,
L.
,
L.
Di Paolo
, and
C.
Flego
,
2016
,
Vibrational spectroscopy assessment of kerogen maturity in organic-rich source rocks
:
Vibrational Spectroscopy
 , v.
87
, p.
14
19
.
Burruss
,
R. C.
,
1991
,
Practical aspects of fluorescence microscopy of petroleum fluid inclusions
, in
C. E.
Barker
and
O. C.
Kopp
, eds.,
Luminescence microscopy: Quantitative and qualitative aspects
 : SEPM Short Course 25, p.
1
7
.
Chen
,
Y.
,
M.
Mastalerz
, and
A.
Schimmelmann
,
2014
,
Heterogeneity of shale documented by micro‐FTIR and image analysis
:
Journal of Microscopy
 , v.
256
, no.
3
, p.
177
189
.
Chen
,
Y.
,
C.
Zou
,
M.
Mastalerz
,
S.
Hu
,
C.
Gasaway
, and
X.
Tao
,
2015
,
Applications of micro-Fourier Transform Infrared Spectroscopy (FTIR) in the geological sciences—A review
:
International Journal of Molecular Sciences
 , v.
16
, p.
30223
30250
.
D’Angelo
,
J. A.
, and
E. L.
Zodrow
,
2011
,
Chemometric study of functional groups in different layers of Trigonocarpus grandis ovules (Pennsylvanian seed fern, Canada)
,
Organic Geochemistry
 , v.
42
, p.
1039
1054
.
Di Matteo
,
V.
,
M. R.
Carroll
,
H.
Behrens
,
F.
Vetere
, and
R. A.
Brooker
,
2004
,
Water solubility in trachytic melts
:
Chemical Geology
 , v.
213
, p.
187
196
.
Diaspro
,
A.
, ed.,
2001
,
Confocal and two-photon microscopy: Foundations, applications and advances
 ,
New York
,
Wiley-VCH
,
576
p.
Fredrich
,
J.
,
1999
,
3D imaging of porous media using laser scanning confocal microscopy with application to microscale transport processes: Physics and Chemistry of the Earth, Part A
:
Solid Earth and Geodesy
 , v.
24
, no.
7
, p.
551
561
, DOI: 10.1016/S1464-1895(99)00079-4.
Gaft
,
M.
,
R.
Reisfeld
, and
G.
Panczer
,
2015
,
Modern luminescence spectroscopy of minerals and materials
 :
Switzerland
,
Springer Mineralogy, Springer International Publishing
, DOI: 10.1007/978-3-319-24765-6_1.
Ganz
,
H. H.
, and
W.
Kalkreuth
,
1991
,
IR classification of kerogen type, thermal maturation, hydrocarbon potential and lithological characteristics
:
Journal of Southeast Asian Earth Science
 , v.
5
, p.
19
28
.
Gasaway
,
C.
,
M.
Mastalerz
,
F.
Krause
,
C.
Clarkson
, and
C.
Debuhr
,
2017
,
Applicability of micro‐FTIR in detecting shale heterogeneity
:
Journal of Microscopy
 , v.
265
, no.
1
, p.
60
72
.
Hackley
,
P. C.
, and
J.
Kus
,
2015
,
Thermal maturity of Tasmanites microfossils from confocal laser scanning fluorescence microscopy
:
Fuel
 , v.
143
, p.
343
350
.
Hackley
,
P. C.
, and
B. J.
Cardott
,
2016
,
Application of organic petrography in North American shale petroleum systems: A review
:
International Journal of Coal Geology
 , v.
163
, p.
8
51
, DOI: 10.1016/j.coal.2016.06.010.
Hayton
,
S.
,
A.
Rees
, and
M.
Vecoli
,
2017
,
A punctuated Late Ordovician and Early Silurian deglaciation and transgression: Evidence from a subsurface study area in northern Saudi Arabia
:
AAPG Bulletin
 , v.
101
, p.
863
886
, DOI: 10.1306/08251616058.
Iglesias
,
M. J.
,
A.
Jiménez
,
F.
Laggoun-Défarge
, and
I.
Suárez-Ruiz
,
1995
,
FTIR study of pure vitrains and associated coals
:
Energy Fuels
 , v.
9
, p.
458
466
.
Kadam
,
P.
,
R.
McAllister
,
J. S.
Urbach
,
K.
Sandberg
, and
S. C.
Mueller
,
2017
,
Live cell imaging and 3D analysis of angiotensin receptor type 1a trafficking in transfected human embryonic kidney cells using confocal microscopy
:
Journal of Visualized Experiments
 , v.
121
, p.
e55177
, DOI: 10.3791/55177.
Kus
,
J.
,
2015
,
Application of confocal laser-scanning microscopy (CLSM) to autofluorescent organic and mineral matter in peat, coals and siliciclastic sedimentary rocks—A qualitative approach
:
International Journal of Coal Geology
 , v.
137
, p.
1
18
.
Li
,
H.
,
W. K.
Wan
, and
J.
Yuan
,
1996
,
Characterization of bitumen emulsions by confocal scanning laser microscopy
:
Petroleum Society of Canada Annual Technical Meeting
,
Calgary, Alberta
, 10–12 June 1996,
12
p., DOI: 10.2118/96-29.
Li
,
Z.
,
P. M.
Fredericks
,
L.
Rintoul
, and
C. R.
Ward
,
2007
,
Application of attenuated total reflectance micro-Fourier transform infrared (ATR-FTIR) spectroscopy to the study of coal macerals: Examples from the Bowen Basin, Australia
:
International Journal of Coal Geology
 , v.
70
, no.
1
, p.
87
94
.
Lindquist
,
W.
, and
A.
Venkatarangan
,
1999
,
Investigating 3D geometry of porous media from high resolution images: Physics and Chemistry of the Earth, Part A
:
Solid Earth and Geodesy
 , v.
24
, no.
7
, p.
593
599
.
Lis
,
G. P.
,
M.
Mastalerz
,
A.
Schimmelmann
,
M. D.
Lewan
, and
B. A.
Stankiewicz
,
2005
,
FTIR absorption indices for thermal maturity in comparison with vitrinite reflectance R0 in type-II kerogens from Devonian black shales
:
Organic Geochemistry
 , v.
36
, p.
1533
1552
.
Marmo
,
R.
,
S.
Amodio
,
R.
Tagliaferri
,
V.
Ferreri
, and
G.
Longo
,
2005
,
Textural identification of carbonate rocks by image processing and neural networks: Methodology proposal and examples
:
Computers in Geoscience
 , v.
31
, p.
649
659
.
Marshall
,
C. P.
,
E. J.
Javaux
,
A.H.
Knoll
, and
M. R.
Walter
,
2005
,
Combined micro-Fourier transform infrared (FTIR) spectroscopy and micro-Raman spectroscopy of Proterozoic acritarchs: A new approach to palaeobiology
:
Precambrian Research
 , v.
138
, no.
3
, p.
208
224
.
Masters
,
B. R.
,
1998
,
Three-dimensional confocal microscopy of the human optic nerve in vivo
:
Optics Express
 , v.
3
, no.
10
, p.
356
359
.
Matsumoto
,
B.
,
2003
,
Cell biological applications of confocal microscopy
 :
London, UK
,
Academic Press
,
507
p.
McDougall
,
D.
,
1952
,
Some factors influencing fluorescence in minerals
:
American Mineralogist
 , v.
37
, p.
427
437
.
Menéndez
,
B.
,
C.
David
, and
A. M.
Nistal
,
2001
,
Confocal scanning laser microscopy applied to the study of pore and crack networks in rocks
:
Computers & Geosciences
 , v.
27
, no.
9
, p.
1101
1109
.
Müller
,
C. M.
,
B.
Pejcic
,
L.
Esteban
,
C. D.
Piane
,
M.
Raven
, and
B.
Mizaikoff
,
2014
,
Infrared attenuated total reflectance spectroscopy: An innovative strategy for analyzing mineral components in energy relevant systems
:
Scientific Reports
 , v.
4
, no.
6764
, DOI: 10.1038/srep06764.
Peters
,
K. E.
,
1986
,
Guidelines for evaluating petroleum source rock using programmed pyrolysis
:
AAPG Bulletin
 , v.
70
, p.
318
329
.
Peters
,
K. E.
, and
M. R.
Cassa
,
1994
,
Applied source rock geochemistry
, in
L. B.
Magoon
and
W. G.
Dow
, eds.,
The petroleum system—From source to trap
 : AAPG Memoir
60
, p.
93
120
.
Petford
,
N.
,
G.
Davidson
, and
J. A.
Miller
,
2001
,
Investigation of the petrophysical properties of a porous sandstone using confocal scanning laser microscopy
:
Petroleum Geoscience
 , v.
7
, no.
2
, p.
99
105
.
Powell
,
T.
,
S.
Creaney
, and
L.
Snowdon
,
1982
,
Limitations of use of organic petrographic techniques for identification of petroleum source rocks
:
AAPG Bulletin
 , v.
66
, no.
4
, p.
430
435
.
Reed
,
S. J. B.
,
1996
,
Electron microprobe analysis and scanning electron microscopy in geology
 :
Cambridge University Press
,
201
p.
Robbins
,
M.
,
1983
,
Activators in fluorescent minerals
, in
The collector’s book of fluorescent minerals
 :
Boston, MA
,
Springer
, p.
183
191
, DOI: 10.1007/978-1-4757-4792-8_9.
Robert
,
P.
,
1981
,
Classification of organic matter by means of fluorescence; application to hydrocarbon source rocks
:
International Journal of Coal Geology
 , v.
1
, no.
2
, p.
101
137
.
Rossman
,
G. R.
,
2006
,
Analytical methods for measuring water in nominally anhydrous minerals
:
Reviews in Mineralogy and Geochemistry
 , v.
62
, no.
1
, p.
1
28
, DOI: 10.2138/rmg.2006.62.1.
Rouxhet
,
P. G.
,
P. L.
Robin
, and
G.
Nicaise
,
1980
,
Characterization of Kerogens and of their evolution by infrared spectroscopy
, in
Durand
B.
, ed.,
Kerogen: Insoluble organic matter from sedimentary rocks
 :
Paris
,
Editions Technip
, p.
163
190
.
Stasiuk
,
L. D.
,
1999
,
Confocal laser scanning fluorescence microscopy of Botroyococuss alginate from boghead oil shale, Boltysk, Ukraine
:
Selective preservation of various micro-algal components: Organic Geochemistry
 , v.
30
, p.
1021
1026
, DOI: 10.1016/S0146-6380(99)00046-7.
Teichmuller
,
M.
, and
M.
Wolf
,
1977
,
Application of fluorescence microscopy in coal petrology and oil exploration
:
Journal of Microscopy
 , v.
109
, no.
1
, p.
49
73
.
Tissot
,
D.
, and
H.
Welte
,
1984
,
Petroleum formation and occurrence: A new approach to oil and gas exploration
 :
Berlin
,
Springer-Verlag
,
143
p.
Washburn
,
K. E.
, and
J. E.
Bidwell
,
2013
,
Multivariate analysis of ATR-FTIR spectra for assessment of oil shale organic geochemical properties
:
Organic Geochemistry
 , v.
63
, p.
1
7
.
Washburn
,
K. E.
,
J. E.
Birdwell
,
M.
Foster
, and
F.
Gutierrez
,
2015
,
Detailed description of oil shale organic and mineralogical heterogeneity via Fourier transform infrared microscopy
:
Energy & Fuels
 , v.
29
, no.
7
, p.
4264
4271
.
Xiao
,
X.
,
R. W.
Wilkins
,
D.
Liu
,
Z.
Liu
, and
J.
Shen
, J,
2002
,
Laser-induced fluorescence microscopy—Application to possible high rank and carbonate source rocks
:
International Journal of Coal Geology
 , v.
51
, no.
2
, p.
129
141
.
Yu
,
L.
,
A.
Porwal
,
E.
Holden
, and
M.
Dentith
,
2012
,
Towards automatic lithological classification from remote sensing data using support vector machines
:
Computers & Geosciences
 , v.
45
, p.
229
239
.

Related

Citing Books via

Close Modal
This Feature Is Available To Subscribers Only

Sign In or Create an Account

Close Modal
Close Modal