An in situ FTIR step-scan photoacoustic investigation of kerogen and minerals in oil shale

Spectrochimica Acta Part A 89 (2012) 105–113

Contents lists available at SciVerse ScienceDirect

Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy journal homepage: www.elsevier.com/locate/saa

An in situ FTIR step-scan photoacoustic investigation of kerogen and minerals in oil shale Kristin N. Alstadt, Dinesh R. Katti, Kalpana S. Katti ∗ Department of Civil Engineering, North Dakota State University, Fargo, ND 58105, United States

a r t i c l e

i n f o

Article history: Received 19 July 2011 Received in revised form 7 October 2011 Accepted 14 October 2011 Keywords: Colorado Piceance Basin Oil shale Kerogen Photoacoustic step-scan FTIR

a b s t r a c t Step-scan photoacoustic infrared spectroscopy experiments were performed on Green River oil shale samples obtained from the Piceance Basin located in Colorado, USA. We have investigated the molecular nature of light and dark colored areas of the oil shale core using FTIR photoacoustic step-scan spectroscopy. This technique provided us with the means to analyze the oil shale in its original in situ form with the kerogen–mineral interactions intact. All vibrational bands characteristic of kerogen were found in the dark and light colored oil shale samples confirming that kerogen is present throughout the depth of the core. Depth profiling experiments indicated that there are changes between layers in the oil shale molecular structure at a length scale of micron. Comparisons of spectra from the light and dark colored oil shale core samples suggest that the light colored regions have high kerogen content, with spectra similar to that from isolated kerogen, whereas, the dark colored areas contain more mineral components which include clay minerals, dolomite, calcite, and pyrite. The mineral components of the oil shale are important in understanding how the kerogen is “trapped” in the oil shale. Comparing in situ kerogen spectra with spectra from isolated kerogen indicate significant band shifts suggesting important nonbonded molecular interactions between the kerogen and minerals. © 2011 Elsevier B.V. All rights reserved.

1. Introduction Oil shale is a sedimentary rock which contains minerals, kerogen, and bitumen yielding a significant amount of oil through pyrolysis [1,2]. On dry weight basis, oil shales can contain 10–60% organic matter, 20–70% carbonate minerals and 15–60% sandy–clay minerals [1]. One of the important components of oil shale, kerogen, is one of the most abundant forms of carbonaceous materials on earth and could be considered to be the most valuable and important component of oil shale [3]. Oil shales are found in 27 countries around the world. However, only a few countries including Brazil, Jordan, Morocco, Australia, China, Israel, and Estonia are mining the oil shale and producing shale oil today. Currently, the United States has the largest known deposits of oil shale in the world with the richest location being the Green River Formation covering portions of Colorado, Utah, and Wyoming. It is estimated that 1.2–1.8 trillion barrels of shale oil, is present in this formation with a conservative estimate of 800 billion barrels of recoverable oil, three times greater than the proven oil reserves of Saudi Arabia. In 2007 the U.S. oil demand was 20 million barrels per day. If the 800 billion barrels of oil met a quarter of that demand (5 million barrels per day), the Green River

∗ Corresponding author. Tel.: +1 701 231 9504; fax: +1 701 231 6185. E-mail address: [email protected] (K.S. Katti). 1386-1425/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.saa.2011.10.078

Formation would supply the U.S. with shale oil for 400 years [4,5]. Kerogen from Green River oil shale is considered to be of Type I indicating immature type of kerogen that is able to produce a significant amount of oil. Studies performed by Duncan et al. confirm the geology of the Northwestern corner of the state of Colorado, the location of the Piceance Basin. This region is made up of marlstone which consists of about 60% clay minerals, 30% quartz, 5% feldspar, 4% carbonates, 1% organic matter, and 1% iron oxides [6,7]. In addition, there are other important economically minerals in the Green River Formation such as nahcolite, trona, and dawsonite [8]. Nahcolite (NaHCO3 ) is used as baking soda. Trona is currently being mined as sodium carbonate and the mineral dawsonite occurs in oil shale of the Green River Formation in extensive beds that constitute a good source of aluminum. The overall oil shale wt% composition of the Green River Formation from Piceance Basin is shown in Table 1 [9]. The mineral matter in oil shale plays an important role during the formation of kerogen. Improving cost effectiveness of shale oil extraction would require fundamental studies on in situ kerogen and its interactions with the mineral matrix. Extensive studies on oil shale and kerogen have been carried out by the Siskin group [10–18] to elucidate the molecular structure of kerogen, as well as interactions of kerogen with minerals. The major objective for the current work was to perform an in situ oil shale spectroscopic study that would identify structural changes to kerogen due to kerogen–mineral interactions. The second objective for this study was to identify differences

106

K.N. Alstadt et al. / Spectrochimica Acta Part A 89 (2012) 105–113

Table 1 The composition of Green River oil shale by weight [9].

Mineral matter 86.2%

Pyrite (FeS2 ) Analcite (NaAlSi2 O6 ·H2 O) Quartz (SiO2 ) Montmorillonite [Na0.2 Ca0.1 Al2 Si4 O10 (OH)2 (H2 O)10 ] K-feldspar (KAlSi3 O8 ) Dolomite and calcite (CaMg(CO3 )2 ) 43.1%

Organic matter 13.80%

Bitumen 2.76% Kerogen 11.04%

between light and dark colored regions located in the oil shale cores. Dark and light colored varves are commonly observed in core samples of oil shale obtained from the Green River Formation. In previous studies on infrared spectroscopy on oil shale from literature, kerogen was isolated from the oil shale through chemical and physical isolation methods since kerogen is “trapped” in, or adsorbed on, the inorganic mineral that surrounds it [19]. In these techniques, there is often a possibility that minor degradation of the organic material occurs and also extremely difficult to isolate all the minerals from the kerogen. Spectra of isolated kerogen are available for demineralized and non-demineralized coal [20], Bulgarian oil shale, Russian oil shale [8], and also Green River oil shale [21]. Often the kerogen spectra are obtained by crushing kerogen and forming pellets with KBr. Photoacoustic step-scan spectroscopy is a non-contact method that is insensitive to surface morphology, ideal for materials that are too opaque for direct transmission in the mid-infrared spectral region, and does not expose samples to air or moisture [22]. An advantage of this technique is depth profiling enabled through varying the modulation frequency [22]. The depth of the penetrated layer depends on the frequency and the thermal diffusivity of the sample. The photoacoustic phenomena is used in conjunction with FTIR instrumentation to analyze molecular interactions in materials [22–24]. Transmission and other experiments using Fourier transform infrared (FTIR) spectroscopy have been performed by various researchers [20,25–32] on oil shale. However, photoacoustic studies on oil shale have not been attempted. This technique has been previously used to evaluate spectra from organics in proximity of minerals in nacre samples from molluscan seashells [33] and teeth [34,35]. In this work, we present our results from photoacoustic experiments conducted on undisturbed oil shale samples and an insight into interactions.

2.2. Materials The samples used to perform FTIR experiments were obtained from the Core Research Center located in Colorado. The samples are from the township 4 South, and range of 95 West near Rifle, Colorado, on the border of Garfield and Rio Blanco counties. The core is seven feet long and is obtained from a burial depth of 565–572 ft. below the earth’s surface. The core contained dark and light colored regions at various locations. The oil shale cores display brown and tan (dark and light) varves each with a layer thickness of about 0.1 mm. Samples used for our experiments are from various regions of the seven foot core. The oil shale samples used to perform FTIR experiments were cut from the 2 in. diameter core with a tile cutter into semicircular disks of about 1 in. thickness. These pieces were then cut with a high precision diamond blade isomet saw into 5 mm × 5 mm × 3 mm pieces for the parallel to the bedding plane samples and 6 mm × 4 mm × 3 mm pieces for perpendicular to the bedding plane samples. The oil shale samples are kept in an air tight container at room temperature and were handled with gloves so as to not contaminate them. A schematic diagram of the parallel and perpendicular orientations as well as an image of the core is shown in Fig. 1 . 2.3. Spectroscopic technique The photoacoustic step-scan spectroscopy is non-destructive to the sample that allows spectra to be taken of the kerogen in its original environment or, in this case, the kerogen surrounded by the mineral matrix. Depth of the photoacoustic signal can be varied by varying the modulation frequency in a step-scan experiment. The sampling depth (L) is determined by [22]: L=

 D 1/2

2. Materials and methodology 2.1. Experiment The FTIR photoacoustic spectroscopic experiments were conducted using a Nicolet 850 FT-IR spectrometer equipped with MTEC Model 300 photoacoustic detector. Data is acquired in the range of 4000–400 cm−1 at a spectral resolution of 4 cm−1 for 32 scans. In-phase (0◦ ) and quadrature (90◦ ) signals were collected at modulation frequencies of 800, 500, 200, and 50 Hz. Background spectra using a carbon black filled elastomer were collected at each of these frequencies. The photoacoustic chamber was purged with ultrahigh purity He gas. For both dark and light colored areas in the oil shale core, samples were taken from various depth locations and were evaluated parallel and perpendicular to the bedding plane. All spectra were normalized using the C H stretching band at 2900 cm−1 . This band was chosen because of its sharpness, consistent location, intensity, and presence in all the spectra taken of light and dark-colored samples.

O Ca Mg C S, N, O H C

0.86% 4.30% 8.60% 12.90% 16.40% 22.20% 9.50% 5.80% 5.60% 1.28% 1.42% 11.10%

(1)

×f

where D is the thermal diffusivity, and f is the modulation frequency. The thermal diffusivity (D) of the sample is given by: D=

K ×C

(2)

where K is the thermal conductivity,  is density and C is specific heat of the sample material [24,36]. The sampling depth (L) depends Table 2 Sampling depths of Colorado Oil Shale for varying frequencies used in photoacoustic stepscan FTIR. Frequency (Hz)

Depth (␮m)

50 100 200 300 500 800

48 33.85 24 19.54 15.14 11.97

K.N. Alstadt et al. / Spectrochimica Acta Part A 89 (2012) 105–113

107

Table 3 Thermal properties of oil shale, kerogen, and shale rock [37–41].

Density (), g/cm3 Heat capacity (Cp ), J/g/K Thermal conductivity (K), J/s/cm/K Thermal diffusivity (˛), cm2 /s Thermal capacity (Cth ), J/cm3 /K

Oil shale

Kerogen

Shale rock

2.06–2.47 [37,38] ∼0.95 0.391411–0.707748

1.05 1.5–1.82 *

2.72 0.8 0.284151

0.0026–0.0098

*

*

2.167

1.8

2.58

*Indicates information that was not available.

Fig. 1. Image (a) and schematic (b) of a 2-in. oil shale core from a depth of 570 ft showing sampling directions parallel and perpendicular to bedding plane.

on the thermal properties of Green River oil shale. The frequency and sampling depths for our study are given in Table 2. The thermal properties of Green River oil shale, shale, and isolated kerogen are as reported in literature [37–41] and are shown in Table 3. 3. Results 3.1. Mineral bands in Green River oil shale A photoacoustic FTIR spectrum showing the characteristic mineral bands is shown in Fig. 2. The Green River oil shale minerals include pyrite, analcite, quartz, montmorillonite, Kfeldspar, dolomite, calcite, nahcolite, trona, and dawsonite. The

characteristic bands observed in the spectra and their assignments based on literature [9,42,43] are listed in Table 4. In addition, the bands for oil shale minerals include the ∼3660 cm−1 O H stretch band of structural hydroxyl. Silica vibrations and vibrations due to Al OH, Al FeOH are also observed from the minerals [44–49]. Spectra from light and dark colored samples prepared from different depths of the oil shale core exhibit kerogen bands similar to that of isolated Green River kerogen bands found in literature [21]. Fig. 3 shows an FTIR photoacoustic spectrum from dark colored oil shale sample taken from the sample parallel to the bedding plane showing the characteristic kerogen bands as listed in Table 4. The light and dark colored oil shale bands, their intensities, and the kerogen and mineral band assignments are shown in Table 4 [9,20,21,25,26,42–59]. Bands at 1400–1350 cm−1 are due to CH deformation from kerogen as well as naphthalene and some mineral bands due to dolomite and calcite [60]. In the region 1230–1150 cm−1 there are weaker C O bands as well as bands of aliphatic ether (OCH3 ) rocking vibration. Bands in the region of 1000 cm−1 are attributed to CCO stretching. There are two sharp, medium intensity bands that are found at 870 cm−1 due to aromatic structures with isolated aromatic hydrogen and a 750 cm−1 band due to four adjacent aromatic hydrogen atoms [25]. Bands at around 756 cm−1 , 725 cm−1 , and 670 cm−1 are aromatic CH outof-plane deformation vibrations and skeletal vibration of straight chain alkanes. All of these bands can be found in the light and dark colored regions. The bands that we have listed are all characteristic of kerogen, confirming that the kerogen molecules are present throughout the oil shale core, and is more dominantly present in the lighter colored oil shale samples. In addition, we also observed bands at 2362 cm−1 and 2333 cm−1 characteristic bands of CO2 [61,62]. Careful experimentation was done to ensure that this band was not from the air in the chamber by long-term purging of chamber with He gas, and comparison with the background spectra. It has been also shown in literature that H2 O and CO2 are major products in the effluent before and during the stage of hydrocarbon generation [63]. Other authors also comment that decarboxylation results in formation of CO, H2 O, and CO2 [20] and kerogen gives off CO2 and H2 S to produce a rubbery polymer which gives off H2 S to form a tarry intermediate, turning into bitumen leading to the formation of oil, coke, and gas [37]. 3.2. Discussion Spectra from the dark and light colored locations throughout the core depth show characteristic bands of kerogen. Kerogen bands in the dark colored samples are overlapped with mineral bands from quartz, clay, pyrite, and calcite. Fig. 4 shows the spectra from light and dark colored oil shale sample, taken from the same location in the oil shale core, scanned parallel to the bedding plane at a frequency of 50 Hz. A relatively higher intensity is observed in the 3600 cm−1 region for light oil shale sample as seen in Fig. 4. As indicated in Table 1, the

108

K.N. Alstadt et al. / Spectrochimica Acta Part A 89 (2012) 105–113

Table 4 Light and dark colored oil shale bands, their intensities, and the kerogen and mineral band assignments (note: w = weak, m = medium, s = strong) [9,20,21,25,26,42–59]. Wavenumber (cm−1 ) Dark

Intensity

Kerogen band assignments

3800–3600

m

3450

m-w

3360–3022

3294

m

3062 and 3022 2956 2926 2896 2851 2624 2514 2362 and 2333 1818 and 1808 1747

3050

m

C O stretch overtone; free NH asymmetric stretch; primary amides OH stretch intermolecular bonding; NH2 asymmetric and symmetric stretch Aromatic C H stretch

2945 2922 2890 2813 2626 2526 2361 and 2334

m m m m w m w

Aliphatic asymmetric CH3 stretch Aliphatic asymmetric CH2 stretch Aliphatic symmetric CH3 stretch Aliphatic symmetric CH2 stretch N H and S H stretching OH stretch due to HCO3 C O stretch from CO2

1795

m m

1699 1671 1644 1606

m

1637

m m/s m

1550 1538

m-s

1505

m-s

1454

1454

m-s

1436

m-s

1395

m-s

1368

m-s

1350 1325 1302 1270 1245

m-s m-s m-s s s

1230 1184 1154 1115

1210 1180 1150

s s s s

Unsaturated ester (C O) doublet so two esters close together Carbonyl C O stretch of aliphatic carboxylic acids (ketone) Aromatic C C, quinone C O Carbonyl C O stretch highly conjugated Naphthalene; aromatic C C stretch; phenanthene C O Phenanthene; C C stretch Primary amines of NH3 deformation; phenolic hydroxyl (OH) Naphthalene; aromatic C C stretch; phenanthene; C C in-plane thiophene CH3 asymmetric deformation overlapped by CH2 scissor vibration CCO stretch; CH2 and C O scissoring; OH bending; aliphatic CH3 CH2 symmetric deformation; NH4 naphthalene; CH3 aliphatic deformation CH2 symmetric deformation; NH4 naphthalene; CH3 aliphatic deformation CH3 aliphatic deformation CCO stretch CH2 wagging vibration Phenolic OH Asymmetric C O C stretch; methyl esters of long chain aliphatic acids

1092–1063

s

1034

s

CCO stretch; long paraffin chains

1000

s

CCO stretch; C C stretch

s

Aliphatic C C stretch; COOH out-of-plane deformation Symmetric COC stretch; C O broad-out-of plane bending CCO stretch; out-of-plane deformation vibration of aromatic CH groups with 3–4 rings Out-of-plane deformation vibration of aromatic CH groups with two or more adjacent hydrogen atoms CH out-of-plane deformation from three adjacent hydrogen atoms on a ring Out-of-plane deformation vibration of 3–4 ring aromatic CH groups with two or more adjacent hydrogen atoms Skeletal vibration of straight chains with more than four CH2 groups

974

m-s

919 868

878

m-s

837–800

m-s

785

m

756

m

725

m-w

O H stretch of structural hydroxyl from MMT, Al O H stretch, unassociated OH, lattice water H OH hydrogen bonded water from MMT H OH hydrogen bonded water due to MMT and dawsonite

OH stretch from nahcolite OH stretch due to bicarbonate

C O stretch characteristic of geologic aragonite

Aliphatic ether (OCH3 ) rocking vibration C O stretch; aliphatic ether Aromatic C H deformation; COC asymmetric stretch; single chain alkane CCO symmetric stretch; single chain alkane

973–942

Mineral band assignments

Light

C O stretch due to nahcolite OH deformation from MMT; Analcite; quartz C O stretch due to dolomite and nahcolite CO3 2− from dawsonite

C O and C O stretch from nahcolite CO3 2− from dolomite and calcite Hydrated FeOH from pyrite; C O and C O stretch from nahcolite

OHO in-plane bend from nahcolite

Si O orientation dependent band from MMT Quartz, pyrite Si O inplane stretch from MMT, quartz; pyrite Si O inplane stretch from MMT; analcite; feldspar Si O out of phlane stretch from MMT; quartz; pyrite; CO3 2− from dawsonite and calcite; feldspar; C O and C O stretch from nahcolite Si O in-plane stretch from MMT; analcite; C O and C O stretch from nahcolite Si O in-plane stretch from MMT; pyrite; OHO stretch out-of-plane from nahcolite; feldspar

Al O/Al OH stretch from MMT Al FeOH deformation from MMT; CO3 2− from dawsonite dolomite, calcite; quartz CO3 2− dawsonite; CO3 out-of-plane from nahcolite; quartz Si O stretch of quartz and silica from MMT, quartz; pyrite; analcite; feldspar Analcite; fledspar

Si O deformation perpendicular to optical axis from MMT, bending CO3 2− from dolomite, calcite and dawsonite; feldspar

K.N. Alstadt et al. / Spectrochimica Acta Part A 89 (2012) 105–113

109

Table 4 (Continued) Wavenumber (cm−1 ) Dark

Intensity

Kerogen band assignments

Mineral band assignments

Light

694

m-w

670 637 and 615

m-w w

583 548 509 484 462

w w w w w

Aromatic CH

Alkanes with 3 or more branches Out-of-plane naphthalene ring vibration C C straight chain alkanes

primary mineral components of green river oil shale are Dolomite and Calcite (43.1%). In light of this, the primary mineral bands must occur in the 2500 cm−1 region (characteristic of HCO3 − groups in mineral lattice [43] or surface), as is observed in the ‘dark oil shale’ samples. In addition, the presence of structural OH from minerals such as montmorillonite (at ∼3600 cm−1 ) must be in combination with significantly higher Si O bands. The higher intensity of OH structural band in ‘light oil shale’ if attributed to larger mineral content must be concurrent with higher intensity of the ∼1100 cm−1 Si O vibrations (a fact that is not observed in Fig. 4). Although much more detailed studies are needed to describe the origin and contributions to the 3600 cm−1 region, additional contributions can arise from trapped unassociated OH and also trapped lattice water. Hence, the possibility of these bands to be from minerals is the only one not entertained because it is not substantiated by higher mineral bands in 2500 cm−1 as well as a higher Si O region. In prior work in literature on Si O vibrations from montmorillonite [44,45,48,64,65] the structural OH vibration from minerals is sharp and short as compared to broader and much stronger Si O vibrations. Further, a comparison of the light and dark colored oil shale spectra indicates differences occurring in the 2000–800 cm−1 region. The dark colored samples parallel and perpendicular to the

Si O deformation parallel to optical axis; pyrite; quartz; calcite; C O, OCO, CO, OH stretch due to nahcolite Pyrite; dolomite; feldspar Coupled Al O, Si O out-of-plane vibration from MMT; pyrite; analcite Al O Si deformation from MMT; feldspar Feldspar Quartz;pyrite Si O Quartz, Si O Fe vibration from MMT; pyrite; analcite; feldspar

bedding plane have dominant mineral bands overlapping the kerogen bands due to bonds from montmorillonite (MMT), analcite, dolomite, nahcolite, dawsonite, and quartz. The light colored oil shale spectra are very similar to that obtained from isolated kerogen, displaying the three distinct kerogen bands around 1600 cm−1 , 1455 cm−1 , and 1200 cm−1 . Observations from comparison of spectra from light and dark-colored regions of the oil shale core suggest that the light-colored regions have a relatively lower mineral content, and the spectra obtained spectra similar to that from isolated kerogen, while the dark-colored samples appear to have more mineral components which overlap some of the distinct kerogen bands. An advantage of the photoacoustic step-scan technique is that spectra can be obtained at different sampling depths of the same sample by changing the modulation frequency. Depth sampling was performed on dark and light colored oil shale samples parallel and perpendicular to the bedding plane in order to observe any changes occurring on a micron scale with depth in the samples. Representative spectra from depth sampling from a light and dark colored sample are shown in Fig. 5. The samples were scanned at frequencies of 800, 500, 200, and 50 Hz resulting in sampling depths of 12, 15, 24, and 48 ␮m. Results of sampling at different depths show that at the lower frequency, or at deeper sampling depth, spectra often have a stronger intensity in the

Fig. 2. Spectrum of a dark colored oil shale scanned parallel to the bedding plane showing the bands characteristic of Green River oil shale minerals.

110

K.N. Alstadt et al. / Spectrochimica Acta Part A 89 (2012) 105–113

Fig. 3. Spectrum of a dark colored oil shale parallel to the bedding plane showing the bands characteristic of Green River kerogen.

region of 4000–1800 cm−1 , but exhibit a lower band intensity in the 1800–400 cm−1 region. As sampling depth decreases, the intensity tends to decrease in the 4000–1800 cm−1 region and increases in the 1800–400 cm−1 region. Band intensity can be attributed to concentration of bonding group in the sample, the phase of the sample, neighboring atoms/groups, and intra/inter-molecular bonds [51]. Similar results are observed in light and dark regions as well as parallel and perpendicular orientations to the bedding plane. These experiments indicate that some oil shale layers tend to have slightly more methyl-methylene bonds, hydroxyl groups, bicarbonate, and carbonyl (C O) when a stronger intensity is observed in the region 4000–1800 cm−1 . Other oil shale layers have more C O, C C, CH3 and CH2 deformation and scissor vibration, minerals,

Si O, C O, CCO, and aromatic CH out-of-plane bonds when higher intensity bands are in the region of 1800–400 cm−1 . Varying intensities at different frequencies suggest that there are compositional differences occurring in the oil shale indicating layering at the micron scale. Oil shale is composed of kerogen, bitumen, and minerals [66,67] that may not be evenly distributed throughout the sample resulting in slightly different sampling depths than that have been calculated. When comparing spectra that are obtained parallel and perpendicular to the bedding plane, features independent of the sample orientation are observed. An example of orientation similarities is shown in Fig. 6 of a light and dark colored oil shale sample scanned at a frequency of 50 Hz parallel and perpendicular

Fig. 4. Light and dark colored oil shale spectra of samples from the same core location scanned parallel to the bedding plane at a frequency of 50 Hz.

K.N. Alstadt et al. / Spectrochimica Acta Part A 89 (2012) 105–113

111

Fig. 5. Spectra of sampling depth parallel to the bedding plane on (a) dark colored oil shale and (b) light colored oil shale.

Fig. 6. Oil shale scanned at 50 Hz parallel and perpendicular to the bedding plane of (a) light colored oil shale and (b) dark colored oil shale.

Fig. 7. Spectra of a light and dark colored oil shale scanned parallel to the bedding plane showing band shifts.

112

K.N. Alstadt et al. / Spectrochimica Acta Part A 89 (2012) 105–113

to the bedding plane. The parallel and perpendicular bands in any given sample have very similar band positions and intensities, indicating uniformity throughout the sample. There were also no orientational differences observed in the parallel and perpendicular spectra obtained at the identical frequencies (50–800 Hz). To clarify, there are spectral changes occurring at different frequencies or sampling depths, but no noticeable differences between parallel and perpendicular spectra of the same frequency. These similarities are observed for both light and dark samples. In order to identify any structural changes in the in situ kerogen as compared to extracted kerogen, we compared the kerogen bands from the FTIR photoacoustic experiments on the oil shale samples to those reported in literature of isolated Green River kerogen [21]. Changes or band shifts were noticed in band positions between isolated and in situ kerogen and is shown in Fig. 7. In addition, photoacoustic FTIR spectra from light and dark colored samples indicate presence of aromatic structures with band at 3065 cm−1 and also band at 1600 cm−1 in the dark-colored samples which is not often seen in isolated kerogen. The C O carbonyl stretch at 1700 cm−1 in Green River isolated kerogen sample [21] was found at 1637 cm−1 in our Green River oil shale. This indicates a significant band shift of about 50 cm−1 . There is another ∼50 cm−1 band shift in the dark colored oil shale sample located at 1500 cm−1 which is attributed to CH3 asym def and CH2 scissor vibration as well as other overlapped bands. This band is found in literature at 1455 cm−1 . In situ kerogen also shows CH out-of-plane deformation shift from three adjacent hydrogen atoms on a ring from bands in the 1200–1000 cm−1 region which is higher than the C C stretching bands of isolated kerogen from literature around 1000 cm−1 . Band position shifts would signify that there are interactions on the molecular scale occurring between the kerogen and mineral components. The band shifts that we have mentioned are related to bonds containing oxygen and the mineral bands signature of quartz and pyrite which is known to have a strong interaction with the kerogen [3,21,68].

4. Summary We have investigated the molecular nature of light and dark colored regions of oil shale samples. The photoacoustic step-scan method provided us with means to obtain spectroscopic data from in situ kerogen with the kerogen–mineral interactions intact. Samples were examined parallel and perpendicular to the bedding plane as well as at sample depths ranging from a 48 ␮m to 12 ␮m from the surface of the sample. Important bands representing kerogen found in the light and dark samples include the OH stretching bands, characteristic aliphatic asymmetrical and symmetrical CH3 and CH2 stretching bands, aromatic CH stretch, ester, ketone, and quinine components of C O and C C stretching bonds, CH3 asymmetrical deformation overlapped by CH2 scissor vibration, aliphatic ether (OCH3 ) rocking vibration, aromatic CH out-of-plane deformation vibration and skeletal vibration of straight chain alkanes. All kerogen identifying bands were found in the dark and light regions confirming that kerogen is present throughout the core. Comparing light and dark regions of the oil shale core indicates that the light colored region has a low mineral content and is similar to that of isolated kerogen that can found in literature while spectra from dark colored regions have more mineral bands overlapping some of the distinct kerogen bands. Sampling at different depths indicates that there are compositional changes in oil shale in layers occurring on the micron level. Oil shale samples were also examined parallel and perpendicular to the bedding plane. Parallel and perpendicular oil shale bands show similar spectra in position and intensity indicating uniformity throughout the sample with no significant orientational attributes. Comparing spectra from isolated

kerogen and kerogen in its natural state show that there are significant band shifts that indicate an interaction on the molecular scale occurring between the kerogen and mineral components. All results are consistent at varying frequencies ranging from 50 to 800 Hz with almost identical bands and similar intensities when comparing light and dark colored sample spectra taken parallel and perpendicular to the bedding plane. 5. Conclusions 1) Step-scan photoacoustic FTIR experiments comparing light and dark colored regions of the oil shale core suggest that the light region of shale samples exhibit spectra similar to that of Green River isolated kerogen that was found in literature with distinct characteristic bands at 3450 cm−1 , 2900 cm−1 , 1700 cm−1 , 1450 cm−1 , and 1050 cm−1 . The dark colored sample spectra have more mineral bands (sometimes overlapping kerogen bands) at 2500 cm−1 , 1800 cm−1 , and 1200–1000 cm−1 due to clay, quartz, dolomite, and calcite minerals all found in Colorado marlstone. More kerogen content was observed for light regions as compared to dark regions. 2) Varying intensities at different penetration depths (ranging from 12 to 48 ␮m) suggest that there are compositional differences occurring in the oil shale on a micron scale indicating layers in the oil shale core. 3) Parallel and perpendicular spectra at the same frequency indicate isotropy throughout the core samples with limited orientational effects. 4) The band shifts observed in the spectra between isolated kerogen and kerogen from oil shale suggest that the mineral components in the oil shale have significant molecular interaction with the kerogen molecules. Acknowledgements The authors wish to acknowledge a grant from the Department of Energy NNSA (National Nuclear Security Administration) under Grant #DE-FG52-08NA28921). Authors would also like to thank the Core Research Center in Colorado for providing Green River oil shale samples. The curator at the CRC is John Rhoades. References [1] M. Razvigorova, T. Budinova, B. Tsyntsarski, B. Petrova, E. Ekinci, H. Atakul, The composition of acids in bitumen and in products from saponification of kerogen: investigation of their role as connecting kerogen and mineral matrix, International Journal of Coal Geology 76 (2008) 243–249. [2] B. Hascakir, T. Babadagli, S. Akin, Experimental and numerical simulation of oil recovery from oil shales by electrical heating, Energy & Fuels 22 (2008) 3976–3985. [3] W.E. Robinson, Isolation procedures for kerogens and associated soluble organic materials, in: G. Eglinton, M.T.J. Murphy (Eds.), Organic Geochemistry: Methods and Results, Springer-Verlag, Laramie, Wyoming, 1969, pp. 181–193. [4] J.M. Huntsman, E. Fletcher, Development of America’s strategic unconventional fuels resources, by Task Force on Strategic Unconventional Fuels, Resource Specific and Cross-cut Plans 2 (2007). [5] A.W. Decora, R.D. Kerr, Processing use, and characterization of shale oil products, Environmental Health Perspectives 30 (1979) 217–223. [6] D.C. Duncan, C. Belser, Geology and oil shale resources of the eastern side of the Piceance Creek Basin, Rio Blanco and Garfield counties, Colorado, U.S. Geol. Surv. Map OM 119, OIl Gas Inv. Ser., 1950, pp. 17. [7] J.R. Donnell, W.B. Cashion, J.H.J. Brown, Geology of the Cathedral Bluffs oil-shale area, Rio Blanco and Garfield Counties, Colorado, U.S. Geol. Surv. Map OM 134, 1953. [8] M.E. Brownfield, T.J. Mercier, R.C. Johnson, J.G. Self, Nahcolite Resources in the Green River Formation, Piceance Basin, Colorado, US Geological Survey Oil Shale Assessment Team, Oil Shale and Nahcolite Resources of the Piceance Basin, Colorado, U.S. Department of the Interior U.S. Geological Survey digital data series DDS-69-Y, Reston, Virginia, 2010. [9] H.W. van der Marel, H. Beutelspacher, Atlas of Infrared Spectroscopy of Clay Minerals and Their Admixtures, Elsevier Scientific, 1976. [10] B. Silbernagel, L. Gebhard, M. Siskin, G. Brons, Electron-spin-resonance studies of kerogen conversion in shale pyrolysis, Energy & Fuels 1 (1987) 501–506.

K.N. Alstadt et al. / Spectrochimica Acta Part A 89 (2012) 105–113 [11] M. Siskin, G. Brons, J.F. Payack, Disruption of kerogen mineral interactions in oil shales, Energy & Fuels 1 (1987) 248–252. [12] G. Brons, M. Siskin, Particle-size reduction studies on Green River oil-shale, Energy & Fuels 2 (1988) 628–633. [13] S. Cramer, M. Siskin, L. Brown, G. George, Characterization of arsenic in oil-shale and oil-shale derivatives by X-ray absorption-spectroscopy, Energy & Fuels 2 (1988) 175–180. [14] G.D. Cody, J.W. Larsen, M. Siskin, Investigation of linear dichroism in coals using polarized photoacoustic FTIR, Energy & Fuels 3 (1989) 544–551. [15] M. Siskin, G. Brons, J.F. Payack, Disruption of kerogen–mineral interactions in Rundle Ramsay Crossing oil-shale, Energy & Fuels 3 (1989) 108–109. [16] A. Katritzky, A. Lapucha, M. Siskin, Aqueous high-temperature chemistry of carbocycles and heterocycles. 18. 6-Membered heterocycles with one nitrogen atom – pyridine, quinoline, acridine, and phenanthridine systems, Energy & Fuels 6 (1992) 439–450. [17] J.W. Larsen, S. Li, Solvent swelling studies of Green River kerogen, Energy & Fuels 8 (1994) 932–936. [18] S.R. Kelemen, M. Afeworki, M.L. Gorbaty, M. Sansone, P.J. Kwiatek, C.C. Walters, H. Freund, M. Siskin, A.E. Bence, D.J. Curry, M. Solum, R.J. Pugmire, M. Vandenbroucke, M. Leblond, F. Behar, Direct characterization of kerogen by Xray and solid-state C-13 nuclear magnetic resonance methods, Energy & Fuels 21 (2007) 1548–1561. [19] K.E. Goklen, T.J. Stoecker, R.F. Baddour, A method for the isolation of kerogen from Green River oil shale, Industrial & Engineering Chemistry Product Research and Development 23 (1984) 308–311. [20] H.I. Petersen, P. Rosenberg, H.P. Nytoft, Oxygen groups in coals and alginiterich kerogen revisited, International Journal of Coal Geology 74 (2008) 93–113. [21] W.E. Robinson, Kerogen of the Green River Formation, in: G. Eglinton, M.T.J. Murphy (Eds.), Organic Geochemistry: Methods and Results, Springer-Verlag, Laramie, Wyoming, 1969, pp. 619–635. [22] A. Rosencwaig, Photoacoustics and Photoacoustic Spectroscopy, John Wiley & Sons Inc., 1981. [23] F.M. Mirabella, Modern Techniques in Applied Molecular Spectroscopy, John Wiley & Sons, Ltd., 1998. [24] J.M. Chalmers, P.R. Griffiths, Handbook of Vibrational Spectroscopy, John Wiley & Sons Ltd., 2002. [25] P.I. Premovic, G.S. Nikolic, M.P. Premovic, I.R. Tonsa, Fourier transform infrared and electron spin resonance examinations of kerogen from the Gunflint stromatolitic cherts (Middle Precambrian, Ontario Canada) and related materials, Journal of the Serbian Chemical Society 65 (2000) 229–244. [26] T. Mongenot, S. Derenne, C. Largeau, N.P. Tribovillard, E. Lallier-Verges, D. Dessort, J. Connan, Spectroscopic, kinetic and pyrolytic studies of kerogen from the dark parallel laminae facies of the sulphur-rich Orbagnoux deposit (Upper Kimmeridgian Jura), Organic Geochemistry 30 (1999) 39–56. [27] N.E. Altun, Incidental release of bitumen during oil shale grinding and impacts on oil shale beneficiation, Oil Shale 26 (2009) 382–398. [28] S.M. El-Sabagh, S. Faramawy, A.M. Rashad, FTIR spectroscopic studies of kerogen and its pyrolysates from selected Egyptian oil shales, Energy Sources Part A: Recovery Utilization and Environmental Effects 31 (2009) 585–593. [29] T. Kaljuvee, E. Edro, R. Kuusik, Formation of volatile organic compounds at thermooxidation of solid fossil fuels, Oil Shale 24 (2007) 117–133. [30] T. Kaljuvee, J. Pelt, M. Radin, TG-FTIR study of gaseous compounds evolved at thermooxidation of oil shale, Journal of Thermal Analysis and Calorimetry 78 (2004) 399–414. [31] S.M. El-Sabagh, J.S. Basta, F.S. Ahmed, M.A. Barakat, Characterization of selected Egyptian oil shales from the Red Sea area: composition of inorganic matrix and potential of organic matters, Energy Sources 22 (2000) 901–911. [32] A.G. Borrego, C.G. Blanco, J.G. Prado, C. Diaz, M.D. Guillen, H-1 NMR and FTIR spectroscopic studies of bitumen and shale oil from selected Spanish oil shales, Energy & Fuels 10 (1996) 77–84. [33] D. Verma, K. Katti, D. Katti, Photoacoustic FTIR spectroscopic study of undisturbed nacre from red abalone, Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 64 (2006) 1051–1057. [34] M. Di Renzo, T.H. Ellis, E. Sacher, I. Stangel, A photoacoustic FTIRS study of the chemical modifications of human dentin surfaces. I. Demineralization, Biomaterials 22 (2001) 787–792. [35] M. Di Renzo, T.H. Ellis, E. Sacher, I. Stangel, A photoacoustic FTIRS study of the chemical modifications of human dentin surfaces. II. Deproteination, Biomaterials 22 (2001) 793–797. [36] J. Mcclelland, R. Kniseley, Photoacoustic spectroscopy with condensed samples, Applied Optics 15 (1976) 2658–2663. [37] T.F. Yen, G.V. Chilingarian, Oil Shale, Elsevier Scientific Publishing Company, New York, 1976. [38] M. Prats, S.M. O’Brien, The thermal conductivity and diffusivity of Green River oil shales, Journal of Petroleum Technology 27 (1975) 97–106. [39] K. Rajeshwar, R. Nottenburg, J. Dubow, Thermophysical properties of oil shales, Journal of Materials Science 14 (1979) 2025–2052. [40] D.W. Waples, J.S. Waples, A review and evaluation of specific heat capacities of rocks, minerals, and subsurface fluids. Part 2. Fluids and porous rocks, Natural Resources Research 13 (2004) 123–130.

113

[41] S. Lee, J.G. Speight, L.S. K., Handbook of Alternative Fuel Technologies, CRC Press Taylor and Francis Group, 2007. [42] V.C. Farmer (Ed.), Infrared Spectra of Minerals (Mineralogical Society monograph), Mineralogical Society of Great Britain & Ireland, 1977, pp. 539. [43] J. Balmain, B. Hannoyer, E. Lopez, Fourier transform infrared spectroscopy (FTIR) and X-ray diffraction analyses of mineral and organic matrix during heating of mother of pearl (nacre) from the shell of the mollusc Pinctada maxima, Journal of Biomedical Materials Research 48 (1999) 749–754. [44] K.S. Katti, D. Sikdar, D.R. Katti, P. Ghosh, D. Verma, Molecular interactions in intercalated organically modified clay and clay–polycaprolactam nanocomposites: experiments and modeling, Polymer 47 (2006) 403–414. [45] D. Sikdar, K.S. Katti, D.R. Katti, Molecular interactions alter clay and polymer structure in polymer clay nanocomposites, Journal of Nanoscience and Nanotechnology 8 (2008) 1638–1657. [46] S.R. Schmidt, D.R. Katti, P. Ghosh, K.S. Katti, Evolution of mechanical response of sodium montmorillonite interlayer with increasing hydration by molecular dynamics, Langmuir 21 (2005) 8069–8076. [47] D. Sikdar, D.R. Katti, K.S. Katti, A molecular model for epsilon-caprolactambased intercalated polymer clay nanocomposite: integrating modeling and experiments, Langmuir 22 (2006) 7738–7747. [48] P.M. Amarasinghe, K.S. Katti, D.R. Katti, Nature of organic fluid–montmorillonite interactions: an FTIR spectroscopic study, Journal of Colloid and Interface Science 337 (2009) 97–105. [49] P.M. Amarasinghe, K.S. Katti, D.R. Katti, Molecular hydraulic properties of montmorillonite: a polarized Fourier transform infrared spectroscopic study, Applied Spectroscopy 62 (2008) 1303–1313. [50] J.J. Cummins, W.E. Robinson, Thermal Degradation of Green River Kerogen at 150–350C: Rate of Product Formation, US Bureau of Mines, Report of Investigations, Laramie Energy Research Center, US Department of Interior, Laramie, WY, 1929. [51] G. Socrates, Infrared and Raman Characteristic Group Frequencies, 3rd ed., John Wiley & Sons, Ltd., 2001. [52] A.P. Radlinski, M. Mastalerz, A.L. Hinde, A. Hainbuchner, H. Rauch, M. Baron, J.S. Lin, L. Fan, P. Thiyagarajan, Application of SAXS and SANS in evaluation of porosity, pore size distribution and surface area of coal, International Journal of Coal Geology 59 (2004) 245–271. [53] V.P. Beskoski, J. Mili, B. Mandic, M. Takic, M.M. Vrvic, Removal of organically bound sulfur from oil shale by iron(III)-ion generated-regenerated from pyrite by the action of Acidithiobacillus ferrooxidans – research on a model system, Hydrometallurgy 94 (2008) 8–13. [54] M. Vandenbroucke, C. Largeau, Kerogen origin, evolution and structure, Organic Geochemistry 38 (2007) 719–833. [55] J. Jehlicka, A. Stastna, R. Prikryl, Raman spectral characterization of dispersed carbonaceous matter in decorative crystalline limestones, Spectrochimica Acta Part A 73 (2009) 404–409. [56] B. Durand, Diagenetic modification of kerogens, Philosophical Transactions of the Royal Society of London Series A: Mathematical Physical and Engineering Sciences 315 (1985) 77–88. [57] C. Snape, Composition, Geochemistry and Conversion of Oil Shales, in: NATO Science Series C: Mathematical and Physical Sciences, Kluwer Academic, 1995, pp. 505. [58] W.E. Robinson, Compositional variations of organic material from Green River oil shale: Wyoming no. 1 core, U.S. Bureau of Mines Report of Investigations 7492, United States Department of the Interior, Washington DC, 1973, pp. 32. [59] J.L. Holmes, N.E. Blair, E.L. Leithold, The fate of kerogen in a river-dominated active margin, in: Proc. Eleventh Annual V.M. Goldschmidt Conference 3207, Hot Springs, VA, 2001. [60] P. Kumar, R.K. Garg, O. Parkash, R.S. Ram, Z.H. Zaidi, Photoacoustic spectroscopic studies of polycyclic aromatic hydrocarbons: naphthalene molecule, Spectrochimica Acta Part A 53 (1997) 151–155. [61] I.M. Porto, R.A. Saiani, K.L.A. Chan, S.G. Kazarian, R.F. Gerlach, L. Bachmann, Organic and inorganic content of fluorotic rat incisors measured by FTIR spectroscopy, Spectrochimica Acta Part A 77 (2010) 59–63. [62] C.F.J. Windisch, P.K. Thallapally, B.P. McGrail, Competitive adsorption study of CO2 and SO2 on CoII3 [CoIII(CN)6 ]2 using DRIFTS, Spectrochimica Acta Part A 77 (2010) 287–291. [63] M. Vandenbroucke, Kerogen: from types to models of chemical structure, Oil & Gas Science and Technology-Revue De L Institut Francais Du Petrole 58 (2003) 243–269. [64] K.S. Katti, A.H. Ambre, N. Peterka, D.R. Katti, Use of unnatural amino acids for design of novel organomodified clays as components of nanocomposite biomaterials, Philosophical Transactions of the Royal Society A: Mathematical Physical and Engineering Sciences 368 (2010) 1963–1980. [65] K.S. Katti, D.R. Katti, Relationship of swelling and swelling pressure on silica–water interactions in montmorillonite, Langmuir 22 (2006) 532–537. [66] K.S. Katti, M.W. Urban, Conductivity model and photoacoustic FT-IR surface depth profiling of heterogeneous polymers, Polymer 44 (2003) 3319–3325. [67] A. Rosencwaig, A. Gersho, Theory of photoacoustic effect with solids, Journal of Applied Physics 47 (1976) 64–69. [68] A.L. Burlingame, B.R. Simoneit, Isoprenoid fatty acids isolated from the Kerogen Matrix of the Green River Formation (Eocene), Science 160 (1968) 531–533.