You are here
Home > BLOG > Others > Current research into the use of supercritical CO2 technology in shale gas exploitation

Current research into the use of supercritical CO2 technology in shale gas exploitation

Fig. 3. Surface comparison of shale before and after ScCO2 fluid treatment (10 MPa, 50 °C) [53].


The use of supercritical CO2 for shale gas extraction is a promising new technology. This paper explores current research into this process, looking at analysis of the mechanism of CH4 displacement in nanoporous shale, the positive and negative effects accompanying its use for sequestration as well as organic extraction, the migration of elements and the swelling process, and the macro and micro control mechanisms involved in permeability enhancement in reservoirs. Fruitful directions for future research are also considered through comparison with hydraulic fracturing.


Wang Menga,b,c, Huang Kaia,b, Xie Weidonga,b, Dai Xuguanga,b

School of Resources and Earth Science, China University of Mining & Technology, Xuzhou 221116, China. Key Laboratory of Coalbed Methane Resource & Reservoir Formation Process, Ministry of Education, China University of Mining & Technology, Xuzhou 221008, China. College of Geology and Mining Engineering, Xinjiang University, Urumqi 830046, China

Received 29 January 2018 – Accepted 12 April 2018

The research findings indicate that ScCO2 fluid replacement can be used to increase gas production and seal up greenhouse gases as an effective, clean and safe method of shale gas exploitation. It is particularly effective for promoting the desorption of CH4 in shale reservoirs that have developed fine neck-wide body pores, and the subtle structural changes effected by ScCO2 fluid in sensitive minerals in reservoirs with a high brittle mineral content also have a positive effect on permeability and storage capacity. The adsorption process has been characterized as consisting of three stages: short-term shrinkage, slow swelling, and stability; an expansion equation has been proposed for CO2/CH4 that incorporates competitive adsorption, collision desorption, and impingement re-adsorption. ScCO2 fracturing has been found to be more effective than hydraulic fracturing for dense reservoirs and more effective at linking up pore-micro-fissure-fracture systems.


In 2016, the total production of shale gas in China was 78.82 × 108 m3, up from 44.71 × 108 m3 in 2015. As is clear from this large increases, these years have been a key period of capacity growth. Shale gas exploration and development have been the subject of numerous recent studies, leading to an array of valuable new findings. Studies into reservoir heterogeneity and nanopore characteristics have provided a deeper understanding of the characteristics of shale gas occurrence, and shale gas productivity investigations based on reservoir energy have provided a theoretical basis for optimizing the mining process [1], [2], [3], [4]. Today, the industry continues to focus research on enhancing the production capacity of shale gas operations.

However, owing to an increasing need for energy savings and emission reduction, attention has also been directed towards potential coordination between capacity enhancement and environmental protection measures. Studies have shown that CO2 has a better adsorption capacity than CH4, with the intriguing implication that, for unconventional reservoirs dominated by adsorbed gas, CO2 injection would be a valuable method to both promote the displacement and drainage of CH4 and store free CO2. This offers considerable economic and environmental benefits [5], [6], [7], [8]. Based on the theory of CO2-dominant adsorption, we have analyzed the feasibility of and problems associated with using supercritical CO2 (supercritical carbon dioxide, ScCO2) in shale gas exploitation from the perspective of CO2 adsorption/desorption and changes in physical properties, summarizing the findings of recent research and considering future directions in this field.

Progress in ScCO2 technology

CO2 generally reaches its supercritical state at a depth of 800 m underground (at a critical temperature of 31.04° and a critical pressure of 7.38 MPa) [9]. Because the burial depth of shale gas reservoirs is commonly more than 1000 m, CO2 present within them is in a supercritical state. ScCO2 fluid has unique physical and chemical properties that give it an excellent adsorption capacity. Indeed, Kang et al. found that the adsorption capacity of CO2 is 5–10 times that of CH4[12]. This means that CO2 injected into a shale reservoir will displace the adsorbed CH4 and will simultaneously become sealed-in itself, providing both environmental and economic benefits [10], [11], [12].

Since the 1990s, copious research has been carried out into the environmental problems caused by excessive emission of CO2, leading to a recognition of the desirability of geological storage of CO2[13], [14], [15], [16], [17], [18]. Various studies have looked into how this could be implemented. Zatsepina et al. evaluated the practicability and safety of CO2 sequestration in depleted shale gas reservoirs [19], [20]. Holloway analyzed the security of this measure by looking at leakage of volcanic CO2 after filtering through specific formations and found that the probability of leakage is low [21]. Oldenburg et al. analyzed transport, storage and injection conditions and took a cost analysis approach to look at the economic feasibility of the process, concluding that transportation and injection constitute the main costs involved [22].

Several valuable studies have been carried out regarding the sealing mechanism. Liu proposed that there are three ways to seal CO2 in shale reservoirs: adsorption by organic matter and clay minerals, microporous capillary agglomeration, and dissolution into mineralized brines [23]. Fathi, meanwhile, highlighted the importance of the diffusion of adsorbed molecules on microporous surfaces during CO2 injection because of its counter-diffusion and competitive adsorption effects [24]. Merey applied an Ono-Kondo model to characterize the adsorption process of CO2 and determine the mode of adsorption [25].

Building on previous work, domestic scholars have confirmed the feasibility of increasing productivity using CO2 (known as CO2-EGR–enhanced gas recovery by injecting CO2), focusing primarily on the transformation of organic matter and minerals [26], [27], [28]. Wang et al. introduced this technology into the field of shale gas, proposing that shale gas technology using ScCO2 could be a valuable new approach to shale gas production [11]. They further explained the rationale for applying this technology in the shale gas industry. Li et al. experimentally verified the feasibility of the replacement of CH4 by CO2 in montmorillonite at a 2–4 km depth of burial [29].

Zhang performed CO2 injection experiments in Triassic shale in Fuxian in the Ordos Basin and found that the optimum injection amount is 0.22 times pore volume [30]. The development of enhanced shale gas recovery through ScCO2 injection is in its initial stages. Against the backdrop of economic and environmental concerns at home and abroad, however, it is probable that its effectiveness, economy, and safety will lead to increasing interest in this technology in future.

Most previous studies have been limited, focusing on the quantification of increases in shale gas production. Few have combined this with a model to simulate the entire displacement process from the perspective of reservoir energy. To address this deficiency, Sun used a Langmuir model to fit the CO2 adsorption process in shale with an isothermal adsorption curve and found that the CO2 adsorption/desorption process is similar to that of CH4 but that there are differences in the maximum adsorption capacity [31]. At pressures over 6 MPa, the Langmuir model cannot be applied to characterize CO2 adsorption in shale. Gautam and Sakurovs introduced a density index to characterize the adsorption process of ScCO2[32], [33]. And Ao et al. used a supercritical DR model to explain the adsorption behavior of CO2 at high pressure [32], [33], [34].

These models describe the CO2 adsorption process well within the given range of conditions and suggest that the swelling and deformation of reservoirs is mainly an external expression of the absorption process. Wang et al. used a grand canonical Monte Carlo method to construct a pillared pore model to simulate the adsorption process of CO2/CH4 in shale and calculated the CO2/CH4 selectivity coefficient of the displacement effect under different conditions of temperature and pressure [35]. When the pressure is below 25 MPa, the CO2/CH4 selectivity coefficient is negatively correlated with the temperature, but the influence of temperature on the selectivity coefficient is not obvious at pressures between 25 and 35 MPa. Moreover, the influence of temperature is negligible when the pressure is greater than 35 MPa. This indicates that pressure is generally the dominant factor controlling the displacement effect in shale gas reservoirs.

Macro and micro geological factors controlling ScCO2 adsorption and desorption

A shale gas reservoir is a continuous natural gas reservoir that combines generation, storage, and cover. The strata comprising the reservoir are mostly black mud shale and high-carbon shale [36]. The main factor controlling shale gas accumulation is the presence of a large number of organic nanoscale micro-pores. Apart from a small quantity of gas dissolved in kerogen and asphaltene, most is adsorbed on the surfaces of the organic matter and minerals or is present unbound in pores and fissures. The content of adsorbed gas is generally between 20% and 85% [37].

Multi-factor coupling can be observed in the mechanisms of shale gas accumulation, which includes adsorptive accumulation, piston flooding and replacement migration, and factors such as a tight reservoir and a complex structure can result in limited or no migration in space and multi-stage accumulation in time [38], [39]. Therefore, it is vital to evaluate the efficiency of ScCO2 adsorption and desorption to understand the process by which CH4 is displaced by CO2 molecules and to quantify the associated phenomena from a microcosmic perspective.

Micro-control mechanisms in the displacement of shale gas by ScCO2

Shale reservoirs feature several pore types: organic pores, clay mineral intergranular pores, intergranular pyrite pores, and some later dissolution pores [40]. Pore morphologies include open tubular pores, slit pores, wedge pores, and, less commonly, ink bottle pores [41], [42].

The diameter of the CH4 molecule (0.414 nm) is slightly larger than that of CO2 (0.330 nm). If molecules of CH4 are produced directly, the pore throat will impede their movement out of some tubular pores and slit pores and particularly from ink bottle pores. The migration rate will only increase when a specific pressure differential is achieved. However, if ScCO2 is used to displace CH4, it will react with the organic matter and some minerals in the pore walls, resulting in the opening of micro pore throats. In addition, some of the (smaller) CO2 will enter at the pore throat, replacing the CH4 adsorbed in the hole walls rapidly and with relative ease (Fig. 1), thus further increasing the production rate.

Fig. 1. Comparison of pore throat reconstruction and desorption before and after injection.

Fig. 1. Comparison of pore throat reconstruction and desorption before and after injection.

In view of the CO2 adsorption characteristics of shale reservoirs, researchers have modified the Langmuir adsorption equation by analyzing the respective partial pressure system of a two-phase gas and constructing models for CH4 desorption, wall re-adsorption, CO2 competitive adsorption, and collisional desorption under microcosmic conditions. The modified equation is:

CO2 competitive adsorption, and collisional desorption under microcosmic conditions.

where V is the total adsorption capacity of CH4 and CO2; VL is the maximum adsorption capacity of the two molecules; PL1 and PL2 are the Langmuir pressure of CH4 and CO2, respectively; P1 and P2 are the partial fluid pressure of CH4 and CO2, respectively; and σ1 and σ2 are the adjustment coefficients of two-phase fluid Langmuir pressure.

where V is the total adsorption capacity of CH4 and CO2; VL is the maximum adsorption capacity of the two molecules; PL1 and PL2 are the Langmuir pressure of CH4 and CO2, respectively; P1 and P2 are the partial fluid pressure of CH4 and CO2, respectively; and σ1 and σ2 are the adjustment coefficients of two-phase fluid Langmuir pressure.

The environment that the adsorption model needs to apply to, which features competitive adsorption, collisional desorption, and impingement and readsorption, gives rise to certain conditions. Some micro-pores have surface roughness, especially pores in organic matter and mineral particles that have been subjected to multiple diagenetic processes. This heterogeneity makes the pore surfaces relatively complex, and many adsorption sites are not occupied. In addition, due to the effect of pore space hybridity, the possibility of secondary impact desorption is relatively large.

This model is well-suited to describing the adsorption behavior of complex organic-mineral pore types. The next step in this analysis will focus on equation fitting. Using displacement test data and scanning electron microscope photos, we will compare the fitting results for reservoirs with complex intergranular pores as compared to normal reservoirs and further discuss the applicability of the equation.

Negative effect of adsorption expansion and the elastic self-regulation effect of micro-fissures

Lahann discovered the phenomenon of swelling in CO2 adsorption experiments on shales from New Albany, but there was no qualitative understanding of the expansion process [43]. With the goal of investigating negative effects in the CO2 adsorption process, Heller characterized the volume of CO2 and CH4 adsorbed on shale and pure minerals (activated carbon, kaolinite, and illite), characterized the volume expansion associated with a given adsorption volume, and compared swelling due to adsorption with that attributable to pore pressure alone [44]. Their experiments achieved a greater amount of adsorption on clay minerals than was expected on the basis of theoretical calculations.

However, this can be attributed to the drying process with which they prepared their samples, which removed water vapor that would normally be adsorbed onto the surfaces of clay particles in nature, thus increasing the capacity of the particles to adsorb CO2. This provides a warning for other calculations and measurements of adsorption capacity [44]. Lu et al. produced a shale adsorption expansion curve for the Longmaxi formation of the Sichuan Basin to quantitatively characterize the phenomenon of adsorption swelling (Fig. 2). Comparing the two curves shows that the amount of shale swelling perpendicular to the bedding plane is greater than that parallel with bedding [45].

The curves indicate that, at high confining pressure, the medium within which the sample was confined produced initial transient contraction. As adsorption of CO2 occurred and as pore-filling by CO2 increased pore pressure, the shale matrix showed gradual expansional deformation. This was the most protracted stage. At the approach to the maximum adsorption capacity of CO2, the internal force environment tended to reach equilibrium and deformation entered a stable stage.

Fig. 2. Shale expansion curve at 1.98 MPa [45].

A shale reservoir is a large elastic geological body (Young’s modulus of 4 × 103 MPa), and micro-fractures are well-developed. A micro-fracture is a bridge connecting micro-pores and fractures. It not only provides reservoir space but also contributes to late-stage desorption and seepage of shale gas. Work by Qin et al. indicates that the energy of the shale gas desorption system will change along with the elastic potential energy in a coal-bed methane (CBM) system. The expansion and deformation of the coal matrix and the tension or closure of fractures will affect the storage space for CBM [46], [47], [48].

Therefore, the effects of elastic self-regulation and effective stress tend to manifest as an increase in the CO2 adsorption capacity for reservoirs containing microfractures. The pressure in the reservoir increases with an increased injection volume and the opening of fractures increases significantly, which is beneficial to desorption of CH4. The matrix expands, and the negative effect of effective stress is the dominant factor counteracting the seepage of desorbing CH4.

Mechanisms of chemical transformation and infiltration enhancement

An increase in the space available for storage and infiltration in terms of porosity, specific surface area, permeability, reservoir pressure, and pore fracture connectivity directly affects productivity and is an important requirement for the success of this new technology. However, due to the compact nature of shale reservoirs, their productivity is also strongly controlled by subsequent fracturing effects. The key to breaking the bottleneck imposed by the tight structure of shale gas fields is a suitable fracturing technique. In recent years, some progress has been made in understanding dynamic change before and after the drainage of natural gas in coal-bearing strata.

Mechanisms of chemical transformation by ScCO2 fluid

The response of each component of organic matter to the ScCO2 fluid plays an important role in the physical and chemical reaction processes that occur after injection. The controlling effects of various physical and chemical properties must be taken into account for reservoir reconstruction. ScCO2 is a non-polar solvent and can extract small molecular weight organic matter within the reservoir, including lipids, volatile matter, and some smaller molecular weight substances. It also controls the reduction process after injection (C + CO2 → 2CO). The solubility of ScCO2 fluid is weakly affected by temperature in a low-stress environment (<16 MPa) and varies with pressure fluctuations in a high-stress environment.

The size of matrix particles also has an effect on the reaction rate, with the ideal particle size lying in the range 1–10 nm [49]. There are copious organic pores in this scale, thus, organic matter has a good potential for reaction in shales. The type of organic matter can also affect the degree of chemical transformation. Certain types have higher TOC and a more obvious pore structure, both before and after transformation, or develop clear secondary microstructures through the reaction. Different types also exhibit different degrees of transformation at different hydrocarbon and lipid storage sites [50].

Carbonaceous shale, as a clay rock, is mainly composed of clay minerals (illite, illite mixed layer, montmorillonite, etc.) as well as containing some brittle minerals (quartz, feldspar, calcite) and authigenic minerals. The transformation of sensitive minerals by CO2 is mainly controlled by geological factors such as structure, temperature, and hydrogeological conditions and is related to the mineral elements involved and the inter-particle bonding on a microcosmic scale.

Recently, many studies have investigated the mechanisms by which ScCO2 fluid transforms reservoir minerals. There has been some progresses on this issue. Clay minerals can be ordered from high to low porosity as montmorillonite, illite mixed layer, kaolinite, and illite. All of these minerals respond to CO2 injection to different degrees, thus profoundly affecting transformation in the reservoir [51]. Some metal elements are adsorbed on clay mineral surfaces in the reservoir. After CO2 injection, the H+ produced by strata under a water medium enters into competitive adsorption with the adsorbed heavy metal ions, and the metal elements (Ca, Mg, K and Al) are removed and transferred [52].

In addition, the ScCO2 fluid significantly transforms calcium feldspar and montmorillonite. In scanning electron microscope photographs of shale samples (Fig. 3), the light spots in regions marked A (mainly calcium feldspar and montmorillonite) faded after ScCO2 injection and the luminosity and number of light spots overall also decreased. The surfaces of intergranular pores in region B were rough after the reaction, and the porosity and number of micro-fissures had also increased.

Fig. 3. Surface comparison of shale before and after ScCO2 fluid treatment (10 MPa, 50 °C) [53].

The main reason for the reduction in the brightness of light spots is the effect of microstructures. There are many kinds of mineral cleavage surfaces on the surface of shale, which shows up as “bright spots.” Due to the solubility and acidity of the fluid, individual minerals are subjected to intense chemical dissolution and precipitation in the form of crystalline water after ScCO2 fluid treatment. Microstructures destroy the reflectivity of the cleavage surface, and the brightness decreases.

Benefits of fracturing through ScCO2 injection

The presence of brittle minerals in shale reservoirs leads to the potential for reservoir transformation by fracturing. Due to its low cost, the ease of access and its practicability, hydraulic fracturing is currently the favored conventional fracturing method [54], [55]. However, there are many negative aspects to this approach: during the process of hydraulic fracturing, the huge pressure difference between the inside and outside of the reservoir often produces destructive fractures, which destroys the reservoir, pollutes the formation water, and causes secondary salinization [59], [60]. Additionally, clay minerals swell and plug fractures during hydraulic fracturing, which is detrimental to shale gas desorption. Furthermore, it is not an applicable technology in areas where there is a shortage of water.

Fig. 4. Comparison the effect of water (white) and ScCO2 (black) on jet shale (8.62 MPa, 50 °C) [62].

Emanuel Martin
Emanuel Martin is a Petroleum Engineer graduate from the Faculty of Engineering and a musician educate in the Arts Faculty at National University of Cuyo. In an independent way he’s researching about shale gas & tight oil and building this website to spread the scientist knowledge of the shale industry.

Leave a Reply

three × five =