Shale pore alteration: Potential implications for hydrocarbon extraction and CO₂ storage

Abstract

Shale unconventional reservoirs are currently and expected to remain substantial fossil fuel resources in the future. As CO₂ is being considered to enhance oil recovery and for storage purposes in unconventional reservoirs, it is unclear how the shale matrix and fractures will react with CO₂ and water during these efforts. Here, we examined the Utica Shale and its reactivity with CO₂ and water using scanning electron microscopy, N₂ and CO₂ sorption isotherms, mercury intrusion porosimetry, and X-ray scattering methods. During CO₂ exposure, the presence of water can inhibit CO₂ migration into the shale matrix, promote carbonate dissolution, and dramatically change the pore scale variability by opening and closing pore networks over the macro- to nano-scale range. These alterations in the shale matrix could impact flow pathways and ultimately, oil recovery factors and carbon storage potential.

Graphical Abstract

1. Introduction

With the recent shale hydrocarbon boom, shale formations in the United States are responsible for 62% of the total hydrocarbons produced [1]. In 2017, they have been shown to produce 0.382 trillion m³ (13.5 trillion cubic feet) of natural gas and 0.78 million m³ (4.9 million barrels) of crude oil [1]. Recovery of hydrocarbons ranges from 5% to 30% with high initial production rates [2]. Production rates, however, decline after the first few years by at least 20% of the initial rate [3]. As shales have very low permeability (< 1000 nD), hydraulic fracturing is necessary to gain access to naturally-existing fractures and pore space located within the matrix to retrieve hydrocarbons and improve flow [2,4]. It has also been proposed that carbon dioxide (CO₂) can be used to enhance hydrocarbon recovery [5] and once hydrocarbons are depleted from shale, CO₂ could be stored in the reservoirs by taking advantage of newly available pore space and existing well infrastructure [6,7].

It is well known that geochemical alteration reactions in porous media are intricate processes that are difficult to measure experimentally and simulate correctly. Generally, it is believed that precipitation reactions reduce the volume of total pore space while dissolution reactions increase the volume of total pore space. However, heterogenous shale systems behave in a much more complex manner [8,9]. This complexity, due to the vast range in shale pore sizes and types, includes the potential for double layer effects that can cause ion concentration changes at nano-sized pore surfaces [10]. In addition, the inter-connectivity of the shale matrix pores is variable among pore types and sizes. A recent SANS study determined that the majority of micropores are inaccessible to CO₂ and concluded that these pores would be closed for CO₂ storage purposes for Marcellus Shale [11]. They recorded no reactivity of CO₂ with the Marcellus Shale while others have observed alterations of shale with fracturing fluid [12–17], extraction cycles [18], and CO₂ [13,19–21]. Hydrocarbon extraction techniques have shown total pore volume to increase [18] while others [19] observed precipitation and dissolution reactions leading to a decrease in accessible pores with exposure to CO₂, SO₂ and O₂ with clays, feldspars, and bulk shale as caprocks.

Here, we focus on the Utica Shale play and its reactivity with CO₂ and water. Currently, 663.5 billion cubic feet of natural gas and oil are produced annually from the Utica in Eastern Ohio. The Utica Shale is a key play for hydrocarbon resources and will continue to be so far into the future [22,23]. In this work, we examine how fluids interact in the natural fractures and matrix pores of the Utica Shale, how these alterations affect flow pathways, and ultimately how these interactions impact hydrocarbon production and CO₂ storage at the reservoir scale. Using field emission - scanning electron microscopy (FE-SEM) coupled with energy dispersive X-ray spectroscopy (EDS), CO₂ and N₂ sorption isotherms, mercury intrusion porosimetry, and small-angle X-ray scattering methods, this effort aims to 1) quantify alterations in the shale matrix from reaction with supercritical CO₂ and water, 2) link these alterations with the different pore types in shale, and 3) discuss implications on whether matrix scale alterations affect transport flow pathways in the shale reservoir. This enables us to define how this reactivity plays a role in hydrocarbon extraction and future carbon dioxide storage operations [8–10,18,24–26].

2. Shale properties

Shales are formed under low energy environments from the compaction of silt and clay-size particles. This compaction process squeezes fluids out and forms natural fractures that are on the millimeter aperture scale and small pores in the matrix that range from micrometers to nanometers in size due lithification processes [30]. The schematic in Fig. 1 depicts a well-scale view of a hydraulically fractured shale with labels for the horizontal well, hydraulic fractures, natural fractures, and pores found in the matrix. The hydraulic fractures (> 10,000 nm wide) are necessary to gain access to naturally-existing fractures and pore space located within the matrix to retrieve hydrocarbons and improve flow [2]. Natural fractures (< 4000 nm wide) are vital to hydrocarbon production as they play a key role in storage and transport of hydrocarbons by connecting the matrix to the induced hydraulic fractures. It should be noted, however, that natural fractures may be naturally cemented closed with calcite or bitumen [31]. In any case, shale matrix pores comprise 99% of the shale pore/void volume and are categorized as inter-particle, intra-particle, and organic matter pores [31–38]. Inter-particle pores (< 2000 nm) are found between the mineral grains and crystals, intra-particle pores (< 2000 nm) occur within the mineral particles, and organic matter pores (< 750 nm) exist within the organic matter present [31]. Shale pores are further classified by size [31] where macropores range between 256 mm and 4 mm, mesopores range between 4 mm and 62.5 μm, micropores range between 62.5 μm and 1 μm, nanopores range between 1 μm and 1 nm, and picopores are < 1 nm in size.

Fig. 1.

Well-scale view of a horizontal well in shale with hydraulic fractures, natural fractures, and matrix pores (modified from [6,31]). Injected CO₂ accessing the fractures and matrix is represented in magenta.

The Utica Shale is an Upper Ordovician mudstone with intervals of calcareous siltstone and limestone. The Utica Shale formation lies beneath portions of Ohio, West Virginia, Pennsylvania, Kentucky, Maryland, New York, Tennessee, Virginia and Canada. However, oil and gas production have been primarily focused in Eastern Ohio. It is important to note that each shale formation is complex with high variability. The pore networks for each of these shale systems are different and depend on the depositional and diagenetic history and mineralogy composition present. In some cases, it has been observed that a dominant pore type will prevail for each shale system in terms of inter-particle, intra-particle, and organic matter pores (See Fig. 1) [31]. Arkakani et al. [39] characterized the porosity of Utica Shale and found inter-particle mineral pores to be the main control on total porosity as opposed to organic matter pores. They reported the importance of major pore types for Utica Shale in a decreasing order as 1) inter-particle inorganic matrix porosity, 2) organic matter porosity, and 3) fracture porosity. They also noted that the natural fracture network was filled with calcite, dolomite, and bitumen. This analysis is consistent with the Utica core description provided by [40] as a fine-grained shale interbedded with organic‐rich and fossiliferous calcareous components.

Utica Shale samples (Table 1) were collected from a stream bed outcrop in New York state as part of the Utica Shale Flat Creek Member (referred in the text as “outcrop”) and from a natural gas production zone from a vertical well at depth of 3,259 m (10,692 ft) drilled in Eastern Ohio as part of the Utica-Point Pleasant Member (referred in the text as “production”). Both samples contain a high percentage of inorganic carbonate, however, only the production sample contains high organic carbon content (Table 2). The high mineral content is consistent with the finding that inter-particle mineral pores influence total porosity as opposed to organic matter pores [40]. The production sample from this depth was highly fractured and was not intact as a whole core [40] and contained thin clay layers, organic rich sections, and calcareous, fossil rich sections.

Table 1.

Characterization of geologic samples.

Utica Sample	Description			Sample Location
Outcrop	Stream Bed Outcrop in New York State	Appalachian Basin	Flat Creek Member	42° 53′ 11″ N	74° 33′ 53″ W
Production	Herrick 3H well core (3259 m) Monroe County, OH	Appalachian Basin	Point Pleasant Member	39° 38’ 44” N	80° 57’ 59” W

Table 2.

Carbon Analysis.

Utica Sample	Total Carbon		Total Inorganic Carbon		Total Organic Carbon
	Carbon (%)	Std. Dev.*	Carbon (%)	Std. Dev.*	Carbon (%)	Std. Dev.*
Outcrop	9.86	0.08	9.41	0.14	0.45	0.17
Production	8.76	0.00	4.39	0.00	4.37	0.00

^*Standard deviations based on two to three independent measurements of each sample

3. Experimental methods

FE-SEM, volumetric isotherms, mercury intrusion porosimetry, and X-ray scattering techniques were used to analyze Utica Shale outcrop and production zone samples in the following three scenarios: 1) un-exposed (as received samples), 2) CO₂ exposed (pressurized using CO₂ to reservoir conditions), and 3) CO₂ and H₂O exposed (submerged in Milli-Q water and pressurized using CO₂ to reservoir conditions). Methods for preparation of reacted samples as well as detailed experimental methods are described in prior work [20,21,27] and here.

3.1. Total carbon analysis

As described in prior work, a UIC CM014 carbon dioxide coulometer integrated with a CM5130 acidification module and a CM5300 combustion furnace apparatus was used to measure the total inorganic carbon (TIC), total carbon (TC), and total organic carbon (TOC) mass content of the shale samples [20].

3.2. Autoclaves

Utica Shale samples were exposed to dry CO₂ and water in autoclave vessels. The samples were placed in 500 mL polytetrafluoroethene (PTFE) lined containers which were then placed in the autoclave. For the dry CO₂ exposure, the autoclave was heated to 40 °C and pressurized to 10.3 MPa with dry supercritical CO₂ for a duration of 14 d. For exposure to CO₂ and water, the samples were submerged in 100 mL of Milli-Q ultrapure water in the 500 mL PTFE lined containers. The autoclave was heated to 40 °C and pressurized to 10.3 MPa with dry supercritical CO₂ for a duration of 14 d. The autoclave was then de-pressurized slowly over 5 h to 6 h.

3.3. N₂ And CO₂ volumetric isotherms

A Quantachrome Autosorb 1-C surface area analyzer was used to perform volumetric sorption isotherm analysis. The samples were analyzed at three different conditions: unexposed, CO₂-exposed, and CO₂ + H₂O-exposed. Before the samples were analyzed, the samples were degassed at 110 °C for 4 h under vacuum. While the measurements were conducted, the samples were kept at a constant temperature of −196 °C and 0 °C for N₂ and CO₂, respectively. The Brunauer-Emmett-Teller (BET) surface area was calculated from the adsorbed volume of N₂ between the relative pressure (P/P0) from 0.1 to 0.3. In addition to BET, the density functional theory (DFT) method was used to determine the volume and size distribution of pores with varying sizes between 0.3 nm and 35 nm. The quenched solid density functional theory (QSDFT) model was used to fit the N₂ adsorption isotherms whereas, the non-local density functional theory (NLDFT) model was used with the CO₂ adsorption isotherms.

3.4. Mercury intrusion porosimetry

A Micromeritics Autopore 9620 mercury (Hg) intrusion porosimeter was used to determine the intrusion volume distribution and pore size distribution between 0.003 μm and 350 μm for the unexposed, CO₂-exposed, and CO₂ + H₂O-exposed samples. In addition, it was used to determine the shale mass densities (bulk and skeletal). Shale samples of mass between 0.2 g and 0.3 g were first degassed under vacuum at room temperature for 2 h before being analyzed. The pressure during measurements ranged from atmospheric pressure (≈ 100 kPa) to approximately 410 MPa.

3.5. Scanning electron microscopy

Scanning electron microscopy (SEM) was conducted on the Utica Shale samples without a conductive coating under low vacuum using a FEI Quanta 600 FEG environmental-scanning electron microscope equipped with energy dispersive X-ray spectroscopy (EDS). SEM images were collected for each sample under the following three scenarios: unexposed, CO₂-exposed, and CO₂ + H₂O-exposed samples. The post-exposure images were completed using a feature relocation method described by our prior work [14] and [28]. To provide insight into the physical morphology of these samples, secondary electron (SE) images were taken. Backscattered electron (BSE) images were taken in conjunction with EDS analysis to determine the different phases within the samples. This phase differentiation is possible because EDS uses the brightness within the BSE image to correlate phase composition to the average atomic number of the constituting minerals.

3.6. X-ray scattering methods

Ultra-small-angle, small-angle and wide-angle X-ray scattering (USAXS, SAXS and WAXS) measurements were carried out at the Advanced Photon Source (APS), Argonne Laboratory, Argonne, IL, USA, specifically the APS USAXS facility at APS sector 9-ID using an X-ray energy of 21 keV. See Appendix A for additional details.

SAXS measurements were made using a Pilatus 100 K 2D detector (Dectris AG, Baden, Switzerland) and sector-averaged to provide scattering intensity data in the range 0.5 nm⁻¹ < q < 17 nm⁻¹ with a q resolution of ≈0.06 nm⁻¹. Geometrical calibration of the SAXS detection geometry is obtained using a silver behenate small-angle diffraction standard. SAXS intensities are not necessarily absolute-intensity-calibrated but can be normalized to the USAXS data (using data in the overlap regime). By combining USAXS and SAXS data, absolute-intensity calibrated I(q) versus q were obtained over the contiguous range of 0.001 nm⁻¹ < q < 17 nm⁻¹.

WAXS measurements were made using a modified Dectris Pilatus 300KW 2D detector and sector-averaged to provide X-ray diffraction (XRD) intensity versus q over a range 13.5 nm⁻¹ < q < 62.5 nm⁻¹ with a q resolution of ≈ 0.09 nm⁻¹ (slightly coarser than for SAXS due to a smaller sample-to-detector distance). The WAXS instrument geometry (and q) was calibrated using NIST Standard Reference Material, SRM 660a (LaB6: lanthanum hexaboride) [29].

Powdered samples for the unexposed, CO₂-exposed, and CO₂ + H₂O-exposed samples (thickness of order 100 μm) were encapsulated in plastic tape, mounted on a sample paddle, and USAXS, SAXS and WAXS measurements made in sequence. The total data collection time for each sample was ≈6 min. The beam size was 0.8 mm horizontal by 0.8 mm vertical for USAXS, and 0.8 mm horizontal by 0.2 mm vertical for SAXS and WAXS.

4. Results and discussion

4.1. Chemical reactivity of CO₂ and H₂O with shale

In our recent work, we found that CO₂ alone sorbs to the clay and kerogen components and reacts with the carbonate portion of dry Utica Shale by extracting interstitial water from the interior clay layers to etch and pitch the shale matrix. When the shale was exposed to CO₂ and H₂O, CO₂ primarily dissolves in the water layer instead of sorbing to the clay and kerogen components [20,21]. H₂O in this case acts as a barrier and prevents access of CO₂ to the kerogen and clays while accelerating the dissolution of carbonate regions in the Utica Shale, which in turns expands etching and pitting of the shale.

In this work, FE-SEM illustrates the etching, pitting, and changes at the micro-pore scale from carbonate dissolution and precipitation in the outcrop Utica Shale samples before and after reaction with CO₂ and water (see Fig. 2). FE-SEM characterization primarily show the calcite matrix, quartz, and pyrite grains of the outcrop Utica sample. In Fig. 2A, the images display FE-SEM backscattered (BSE) images of unexposed Utica Shale (left), supercritical CO₂ exposed Utica Shale (middle), and supercritical CO₂ and water exposed Utica Shale (right). These images are of the same location before and after exposure to CO₂ and water. Minor etching and changes in porosity occur from CO₂ exposure (middle), while major etching, pitting, and changes in porosity are apparent after CO₂ and water exposure (right) during carbonate dissolution. These changes in porosity were quantified using the open-source software packages Ilastik and ImageJ for interactive image classification and the results are shown in Fig. 2B. These bottom images are the same FE-SEM BSE images shown in the top of the figure, however, these images were modified, segmented, enhanced, and analyzed to quantify changes in porosity. In Fig. 2B, porosity is color-coded as white, while solid space is color-coded as black. The image analysis shows that porosity increased to 7.6% following CO₂ exposure and then to 33% following CO₂ and water exposure for micro-pores between 1000 nm and 15,000 nm.

Fig. 2.

A: Field-emission scanning electron microscopy backscattered (FE-SEM BSE) images of outcrop Utica Shale unexposed (left), supercritical CO₂ exposed (middle) and supercritical CO₂ and water exposed (right). B: The same FE-SEM BSE images shown in A, however, modified with Ilastik and ImageJ where porosity is color-coded as white and solid space is color-coded as black.

For the production Utica Shale sample, high-resolution FE-SEM measurements were initiated on the unexposed sample. Unfortunately, an image set complimentary to the Utica outcrop samples (see Fig. 2) was unsuccessful as the production sample did not remain intact during reaction with CO₂ due to its thin and fractured nature. Fig. 3 illustrates the fractured nature of the production sample after reaction with CO₂. FE-SEM using EDS revealed higher levels of organic matter (see yellow bands in Fig. 3) in the production sample when compared to the outcrop sample. Unfortunately, experiments exploring the reactivity of the organic content for the production sample could not be further explored as additional production samples from at depth conditions could not be secured.

Fig. 3.

Scanning electron energy dispersive X-ray spectroscopy image of a production zone Utica Shale exposed to dry CO₂ at reservoir conditions. Blue regions are calcite rich indicating carbonate presence while yellow bands represent carbon or organic matter.

Alteration of pore sizes of the outcrop and production Utica Shale samples were also documented with CO₂ and N₂ sorption isotherm measurements and Hg intrusion porosimetry measurements. The combination of these two techniques allows us to examine the impact on the Utica Shale pore volumes and pore size distributions after exposure to CO₂ and H₂O from the picopores to mesopores range (i.e., 0.3 nm to 350 μm [350,000 nm]). Textural properties of the outcrop and production samples in terms of bulk density, apparent density, BET surface area, and pore volume were compiled (Table 3). Overall, the production samples have larger BET surface area and volumes for the pore sizes below 35 nm possibly due to their higher total organic content (Table 2). Furthermore, loss of BET surface area and volume of fine nanopores up to 2 nm was noticeable after both the outcrop and production samples were exposed to CO₂ and CO₂ + H₂O.

Table 3.

Textural properties of the Utica outcrop and production samples.

Parameter	Outcrop			Production
	Un-exposed	CO2-exposed	CO2 + H2O -exposed	Un-exposed	CO2-exposed	CO2 + H2O -exposed
Bulk density (g/cm3)a	1.1620	1.1869	1.0742	1.0786	1.0970	1.0381
Apparent density (g/cm3)a	2.7114	2.5247	2.3665	2.8299	2.8815	2.5089
BET surface area (m2/g)b	6.8	6.3	5.8	19.6	8.2	17.5
Surface area (m2/cm3)c	7.9	7.5	6.2	21.1	9.0	18.2
Pore volume (cm3/g), d < 0.7 nmd	0.00127	0.00141	0.00139	0.00463	0.00371	0.00423
d = (0.7–2) nme	0.00071	0.00030	0.00001	0.00223	0.00000	0.00181
d = (2–35) nme	0.01177	0.01239	0.01651	0.03429	0.02433	0.03172

^aMeasured from Hg intrusion porosimetry. Total intrusion volume uncertainty is ± 0.03 cc/g and median pore diameter uncertainty is ± 5.0 nm.

^bMeasured from N₂ isotherm at −196 °C and relative pressures between 0.1 and 0.3. Estimated instrumental measurement uncertainties are within ± 5%.

^cSurface area = BET surface area × bulk density.

^dMeasured from CO₂ isotherm at 0 °C.

^eMeasured from N₂ isotherm at −196 °C.

Data from N₂ isotherms for the outcrop and production samples are compiled in Tables 4 and 5. Pores between 0.7 nm and 5 nm in both the outcrop and production samples become less accessible upon exposure to CO₂ and CO₂ + H₂O, where in some cases a 100% decrease in pore volume is observed. Meanwhile, pores larger than 15 nm particularly in the outcrop sample become more accessible after reaction with CO₂ + H₂O, which is indicated by an increase in pore volume as high as 32% (For CO₂ isotherms and Hg porosimetry data see Appendix A Figs. A1 and A2 and Tables A1 and A2). Fig. 4 highlights a portion of these changes from Tables 4 and 5 in pore size distributions in the 0.7 nm to 2 nm and 5 μm to 10 μm ranges for both the outcrop and production samples. While both samples reveal a multi-modal distribution, the inset in Fig. 4 records the alteration in pore volume distributions of the outcrop and production. Between 0.7 nm and 2 nm, the outcrop pore volume changes from 5% to 0% after reaction with CO₂ and H₂O while the production pore volume changes from 6.1% to 5.7%. Between, 5 μm and 10 μm, the outcrop pore volume changes from 6.0% to 19.5% after reaction with CO₂ + H₂O while the production pore volume changes from 7.7% to 6.9%. Carbonate dissolution and precipitation are likely responsible for opening of the larger, micro-pores and are consistent with the 30% increase in porosity estimated from FE-SEM image analysis (see Fig. 2). Aggregation of clay and organic features and precipitating particles may be responsible for the decrease in the smaller, nano-pores [41].

Table 4.

Percent Pore volume distribution from N₂ isotherms for the outcrop Utica Shale. The numbers in the parentheses are percent volume changes relative to the values of the unexposed sample. See Table 3 for uncertainties.

Pore diameter range (nm)	Unexposed	CO2-exposed	CO2 + H2O-exposed
0.7–2	5.0	2.4 (−52.7)	0.0 (−100.0)
2–5	12.2	11.8 (−4.0)	8.8 (−28.5)
5–10	18.1	20.2 (+11.4)	16.3 (−9.8)
10–15	16.1	14.7 (−8.8)	16.9 (+5.2)
15–20	13.2	12.1 (−8.5)	15.2 (+14.7)
20–25	12.0	11.6 (−3.2)	15.9 (+32.3)
25–30	14.5	17.0 (+17.1)	16.5 (+13.7)
30–35	8.8	10.3 (+16.8)	10.3 (+17.0)

Table 5.

Percent Pore volume distribution from N₂ isotherms for the production Utica Shale. The numbers in the parentheses are percent volume changes relative to the values of the unexposed sample. See Table 3 for uncertainties.

Pore diameter range (nm)	Unexposed	CO2-exposed	CO2 + H2O-exposed
0.7–2	6.1	0.0 (−100.0)	5.7 (−6.3)
2–5	11.2	4.4 (−60.4)	9.7 (−13.0)
5–10	19.1	17.4 (−9.1)	20.1 (+5.2)
10–15	15.6	19.2 (+23.4)	17.8 (+14.1)
15–20	16.4	19.4 (+18.0)	15.6 (−4.9)
20–25	9.6	13.3 (+39.2)	9.1 (−4.8)
25–30	14.0	16.9 (+20.8)	13.5 (−3.1)
30–35	8.0	9.4 (+16.4)	8.4 (+4.4)

Fig. 4.

Pore size distributions analyses of the outcrop and production Utica Shale samples: (A) 0.7 nm to 10 nm (from N₂ sorption isotherms) (B) 0.003 μm to 350 μm (from Hg porosimetry). Insets show corresponding pore volume distributions of the unexposed, CO₂-exposed, and CO₂ + H₂O-exposed samples for the 0.7 nm to 2 nm and 5 μm to 10 μm pore diameter ranges. Uncertainties are within ± 5% and traces are offset form each other for clarity.

Results from ultra-small-angle X-ray scattering (USAXS), small-angle X-ray scattering (SAXS), and wide-angle X-ray scattering (WAXS) analysis were consistent with FE-SEM, N₂ and CO₂ sorption isotherm measurements, and Hg intrusion porosimetry measurements described above. The nominal porosity and surface area increased as CO₂ and CO₂ and H₂O were reacted with the production Utica samples (Fig. 5). Comparison of the scattering curves for the powdered shale samples, together with their respective attenuations of the transmitted X-ray beam, indicates that the 1 μm to 4 μm size pores in the scattering size distributions are the voids between the powder grains and account for most of the porosity of the powdered samples. Nano-size pores can be attributed to porosity within the solid powder grains.

Fig. 5.

Nominal porosity and surface area for the production Utica Shale sample as calculated from USAXS, SAXS, and WAXS measurements (Advanced Photon Source, sector 9-ID). Horizontal bars in the figures represent estimated Type B standard uncertainties (≈67% confidence) based on measurement history of similar samples [42].

5. Implications for hydrocarbon extraction and CO₂ storage

In the schematic in Fig. 6, we link the data collected in this study regarding pore changes in Utica Shale following reaction with CO₂ and H₂O with the shale fracture, pore types, and pore size classification system designed by Loucks et al. [31]. The green arrows represent pore alteration trends identified by FE-SEM. The increase in porosity between 7% and 33% detected with FE-SEM primarily corresponds to micro-scale changes associated with intraparticle and interparticle pores in the mineral matter and natural fractures as carbonate is dissolved. The blue and purple arrows represent the pore alteration trends identified by N₂ and CO₂ sorption isotherm measurements and Hg intrusion porosimetry measurements, respectively. The smaller pores in the nano-scale range of 2 nm to 15 nm are likely associated with clay and organic features and may become inaccessible due to preferential aggregation of precipitating particles in the smallest pores [41]. Orange arrows for SAXS measurements represent alteration of pores in the 1 nm to 1000 nm range and further support the micro- and nano-sized changes recorded with FE-SEM and pore size analysis techniques.

Fig. 6.

Schematic linking shale fracture and pore types and size classes with the analytical techniques applied to investigate reactivity of Utica outcrop and production samples. (A) Fracture and pores present within mudrock: hydraulic fractures, natural fractures, interparticle and intraparticle pores associated with mineral matrix, and organic-matter pores associated with organic matrix. (B) Analytical techniques used to examine pore size distributions and porosity characteristics in mudrocks. (C) Pore-size classification developed by [31]. (D) Pore-size trends with Utica outcrop and production samples following reaction with CO₂ and H₂O. Green arrows represent data from scanning electron microscopy, blue arrows represent data from CO₂ and N₂ low-pressure gas sorption isotherm measurements, purple arrows represent data from mercury intrusion porosimetry measurements, and orange arrows represent data from X-ray scattering methods.

*Utica Shale porosity is primarily controlled by inter-particle inorganic matrix porosity followed by organic matter porosity and fracture porosity

These results show that CO₂ and H₂O reactivity with shale can dramatically change the pore scale variability by opening and closing pore networks over the meso- to nano- range. Due to the heterogeneity of shales in terms of fluid content, mineralogy, TOC, thermal maturity, and pore scale variability, it is important to understand the chemical reactivity of the pore network in the shale matrix [43]. As described by [31], a predominant pore system likely plays a key role on porosity and transport for each shale play. For the Utica, inter-particle inorganic matrix porosity dominates pore distribution type followed by organic matter porosity and fracture porosity [39]. In quantifying flow and transport properties, the predominant pore system and the chemical reactivity within this network from the macro- to nano-scale should be considered. Observations specific to the Utica show that meso- and micro-pores are opening while nano-pores are closing. In general, an increase in micro-porosity could translate to an increase in inter-particle pore connectivity within the shale matrix with improved access to natural and hydraulic fractures. This in turn, could increase CO₂ storage resource potential. On the other hand, closure of nano-pores associated with organic matter porosity in the shale matrix may decrease access to networks needed to improve enhanced oil recovery potential. The vast differences in alteration reactions and changes in pore scale variability demonstrate the importance of linking how these alterations affect flow pathways. Ultimately, flow property changes will affect hydrocarbon production, CO₂ storage, and sealing units at the reservoir scale of the Utica Shale and other shale plays. This is an important step in quantifying and reducing uncertainties in the porosity, the degree to which the pores are connected, and how porosity can affect transport and flow properties of shale systems.

6. Concluding remarks

Porosity is a key parameter in subsurface systems such as shale formations because it is directly linked to fluid storage - in this case for hydrocarbon extraction and CO₂ storage. This work shows that CO₂ has the ability to alter porosity from the micro- to nano-scale. This suite of complimentary techniques show that pores are both opening at the micro-scale and closing at the nano-scale for Utica Shale depending on their pore type and size classification. The alteration of pore scale variability may affect transport and flow properties of the shale systems and may vary depending upon a shale formation’s petrological properties.