Journal of Applied Sciences

Year: 2014 | Volume: 14 | Issue: 17 | Page No.: 1919-1927
DOI: 10.3923/jas.2014.1919.1927

Factors Affecting Sorption Induced Strain of Coal Specimens During Carbon Dioxide Injection: A Review Study

Mustafa Abunowara, Usama Eldemerdash and Mariyamni Awang

Abstract: Carbon dioxide (CO₂) capture, utilization and storage (CCUS) are considered as a potential approach to mitigate carbon dioxide emissions into geologic formations. Deep unmineable coal seams have been identified as a possible option because it has large CO₂ adsorption capacity, long time CO₂ trapping and extra enhanced coal-bed methane recovery (ECBMR) and CO₂ sequestration. The current practice for recovering coal bed methane is to depressurize the bed, usually by pumping water out of the reservoir, the desorption of gas from coal surface, diffusion of gas to the fracture systems and flow of the gas through the fractures to the wellbores. Hence, an alternative approach is to inject CO₂ into the coal bed seams to increase the mobililbity of methane recovery. As a result induced adsorption strain (swelling) is one of the main difficulties which face CO₂ sequestration in coal seams. This phenomenon occurs, particularly when the injected carbon dioxide adsorbs on surface of the coal pores and interacts with coal in chemi-physical adsorption isotherm under extreme conditions, which causes the coal to swell. This swelling in confined conditions leads to a closure of coal matrix pores and cleat system, which hinders further CO₂ injection. However, swelling will decrease permeability and adsorption capacity of coal seams and increases CO₂ injectability potential complications. The degree of swelling would be affected by many parameters such as coal rank, water content and petrographic content, mechanical properties (e.g., stress levels and confinements), operating conditions (e.g., gas injection pressure and temperature), free gas and fluid type. Thus these parameters have significant affect on CO₂ continuous injection process in coal seams in long term. This study is a reviewing for the main parameters which have influence on coal swelling during carbon dioxide injection in coal specimens.

Keywords: Coal seams, supercritical CO2, swelling, CO2 sequestration and gas sorption

INTRODUCTION

Coal is known to swell when gases adsorb chemically and physically onto its surface (Day et al., 2010; Karacan, 2007; Kelemen and Kwiatek, 2009; Majewska et al., 2009) and when CO₂ interacts with wet coal. Swelling of the coal during adsorption of CO₂ is one of the obstacles of CO₂ sequestration in coal seams as it causes coal seam permeability to be significantly reduced (Perera et al., 2011). Coal sorption induced strain (swelling) depends on a number of factors including coal rank, coal seams pressure and temperature, gas nature and type and duration of gas injection and stress and confining pressure. In unconfined coal, swelling in CO₂ is typically less than about 5% by volume but this is nevertheless important when considering enhanced coalbed methane (ECBM) production or CO₂ sequestration because it can affect the gas transport properties of coal seams. Theoretical (Pan and Connell, 2011) and experimental (Harpalani and Mitra, 2010; Wang et al., 2010) studies predict that swelling decreases seam permeability and recent field trials confirm that gas flows are reduced when CO₂ is injected into deep coal seams (Durucan et al., 2009; Van Bergen et al., 2009; Fujioka et al., 2010). Because of the potential of swelling to seriously affect large scale ECBM and sequestration projects. This is one the difficulties which face CO₂ injection into Yubari field pilot test (Kiyama et al., 2011). However, the sorption of CO₂ in coal causes deformation of the coal matrix and affects the dynamic permeability and then decrease transport of gases in coal fracture, determining both the rate and capacity of the coalbed seam. In addition to that CO₂ adsorption and CH₄ desorption and can significantly affect the volumetric change of micro and macropores in coal leading to coal swelling or shrinkage and evolution of permeability and then controlling the transport and flow of gases through coal matrix (Wang et al., 2010).

Although there are some differences in the results of varied studies, they are mostly consistent and may be due to different techniques and procedures which used by many researchers. The general conclusions can be briefly summarized as follows: Volumetric swelling increases as a non-linear function of injection pressure, approaching a maximum value, usually above 15 MPa and depending on the rank coal, maximum volumetric swelling induced by CO₂ in unconfined samples is generally within the range of about 1-5%, with highest swelling occurring in lower rank coals. However, recently it has been reported that much higher swelling is possible in compacted aggregates of coal (Van Bergen et al., 2011). In addition, other gases apart from CO₂ also swell coal to varying degrees gases such as ethane and xenon can swell some coals as CO₂ (Day et al., 2010) whereas methane induces about half as much swelling as CO₂ (Day et al., 2010; Van Bergen et al., 2011) and some gases can swell the coal more than CO₂, particularly H₂S, which could be a component of some flue gases destined for deep coal seam injection (Cui et al., 2007). In addition, the majority of coal seams are saturated with water that has been shown to influence swell of coal in different levels, in some cases by nearly as much as CO₂ (Fry et al., 2009). Often swelling is greater in the direction perpendicular to the bedding plane compared to the parallel direction (Day et al., 2008; Levine, 1996; Van Bergen et al., 2009) although some studies report that significant anisotropy was observed (Day et al., 2008). The mechanisms of coal matrix swelling base on (Qu et al., 2012) modeling results are at the initial stage of CO₂ injection under variable temperatures, matrix swelling due to gas sorption, thermal expansion and the change in adsorption capacity is localized within the vicinity of the matrix fracture. As the injection continues, the swelling zone is widening further into the matrix and the swelling becomes macro-swelling. When the swelling is localized, coal permeability is controlled by the internal fracture boundary condition and behaves volumetrically; when the swelling becomes macro-swelling, coal permeability is controlled by the external boundary condition (Qu et al., 2012). Moreover, regarding to what have mentioned above about sorption induced strain of coal. The following section will explain in more details the parameters which have impact on swelling phenomenon during carbon dioxide injections in coal seams.

FACTORS INDUCED COAL SWELLING

CO₂ injection into coal seams under in situ conditions urges chemi-physical adsorption and chemical reactions under confined conditions lead to swelling of coal seam. The degree influence of swelling relies on coal rank and water content and temperature and stress levels and confinements. Swelling leads to volumetric change of intact coal matrix and then leads to fracture cleats blockage which alter coal permeability and decrease CO₂ injection rate. Thus there are many factors which could control swelling strain such as coal rank, water content, petrographic contents, operating conditions (e.g., injection pressure and temperature), mechanical properties (stress and confining pressure). In addition, these parameters play a viable role on altering gas permeability of coal seams and then reducing CO₂ injection rate and coal adsorption capacity.

Coal rank and moisture content: Day et al. (2011) observed that swelling was greater in carbon dioxide (CO₂) than methane (CH₄). Experiments conducted on four Australian coals specimens at high injection pressure (up to 15 MPa) and temperatures (up to 55°C). It is observed that lower rank coals swell more than higher rank material and the presence of moisture significantly reduce the amount of additional swelling by the gas compared to dry coals and the degree to which the swelling of the coals was affected by moisture depended on the rank of the coal. In the lowest rank coal, maximum swelling was about 5% in CO₂ under dry conditions compared to about 2% in the highest rank sample (Day et al., 2008, 2011). In the dry samples swelling was greater in CO₂ than CH₄ by a factor about 1.7, regardless of the rank of the coal. Meanwhile, in moist coal the ratio was about 2.5, reflecting the proportionally greater effect of water on swelling in methane. It is noticed that moisture significantly reduced the degree of gas-induced swelling in both CO₂ and CH₄. However, the mount to which the swelling was reduced was a function of the rank and nature of the coal. In lower rank coals, the effect of moisture on the amount of swelling is high and swelling was depressed by about 55% whereas the highest rank coal was barely affected (Day et al., 2011). Although moist coals specimens swell less than dry samples and this is not rely on coal rank which the same result was reported by N. Siemons and A. Busch as shown in Fig. 1. No specific trend can be observed for the coals containing water while for the dry coals the increase in coal volume decreases at low rank and increases again at higher rank with a minimum vitrinite reflectance at 1.1 to 1.3%. In particular, the low rank coals show a large increase in volume (~11-13%). Coals containing water show no rank dependency. The volume increase for all samples varies between 4 and 8% (Siemons and Busch, 2007).

Thus, if this pre-swelling is induced, the total swelling of coals is high than that induced by the gas in dry coal (Day et al., 2011). Swelling was found to be entirely elastic, even after the coal had been subject to multiple exposures to CO₂ phases (Day et al., 2008).

Below a few atmospheres pressure, swelling is low and generally unaffected by the amount of gas adsorbed, but at increased pressure, swelling becomes roughly linearly proportional to the amount of CO₂ adsorbed. Above about 8 MPa, this relationship was no longer linear, adsorption continued to increase but swelling did out as shown in Fig. 2. Volumetric swelling strain at 15 MPa ranged from about 1.9-5.5% in CO₂ and 1.0-2.5% in CH₄ depending on the rank of coal and the proportion of CO₂ in the gas mixture. Experimental results depict that there is no further enhanced swelling in mixed gases above that would be observed in the pure CO₂ at the same total pressure (Day et al., 2012).

Kiyama et al. (2011) conducted two laboratory experimental tests to simulate Yubari field pilot test on coal specimens and to understand the change of coal physical properties (permeability) during continuous injection of liquid and supercritical CO₂ and N₂ gas by measuring strain, elastic wave velocity and permeability evolution under stress-constrained conditions.

In test I, liquid CO₂ was injected into a water-saturated coal specimen and then heated and injected as supercritical CO₂ and a volumetric swelling strain 0.25 to 0.5% was observed after injecting liquid CO₂. In test II, supercritical CO₂ was injected into a coal specimen saturated with N₂ and then N₂ and CO₂ were repeatedly injected and the swelling strain was about 0.5-0.8% after injecting supercritical CO₂. Following further injection of N₂ in test II, slow strain recovery was observed in the coal and this test was to simulate the case of N₂ injection and CO₂ re-injection at Yubari. The water-saturated coal specimen swelled by 2500-5000μ during liquid CO₂ injection as depicted in Fig. 3. The N₂ saturated dry specimen from the same block swelled by 5000-8000μ during supercritical CO₂ injection as depicted in Fig. 4. The difference between results was a consequence of different injected gas type, different media injection and adsorption capacity of the wet and dry coals specimens. In addition, subsequent N₂ flooding tests following CO₂ injection showed a little strain reduction, suggesting that N₂ displaces the adsorbed CO₂ in the coal matrix. Hence, the permeability of the coal specimen was also recovered after N₂ injection, although it declined rapidly after CO₂ injection. These results suggest that when liquid CO₂ was injected into the water-saturated coal specimen, it did not completely displace the water in the coal mixture and indicate that coal swelling is likely to be the main cause for the permeability change in the Yubari field tests and swelling strains were difficult to measure (Kiyama et al., 2011).

Fujioka et al. (2010) conducted a micro-pilot test with a single well and multi-well CO₂ injection tests, involving an injection and production wells in the period between May 2004 and October 2007. However, there were a variety of tests conducted in the injection well, includes an initial water injection fall-off test and a series of CO₂ injection and fall-off tests (Fujioka et al., 2010). Although gas production rate was obviously enhanced by CO₂ injection, water production rate was not clearly affected by CO₂ injection. Several injection tests that injectivity of CO₂ into the virgin coal seam saturated with water was eventually increased as the water saturation near the injector was decreased by the injected CO₂. As result, it was estimated that low injectivity of CO₂ was caused by the reduction in permeability induced by coal swelling. N₂ gas flooding test was performed in 2006 to evaluate the effectiveness of N₂ injection on improving well injectivity and the N₂ flooding test showed that daily CO₂ injection rate was boosted, but only temporarily. Moreover, the permeability did not return to the initial value after CO₂ and N₂ were repeatedly injected. It was also indicated that the coal matrix swelling might create a high stress zone near to the injection well. In addition to that the last estimation on the fracture opening pressure, boost of injection pressure up to 19MPa was tired at the final stage of the pilot test, which exceeded predetermined 15.6 MPa of limit injection pressure and it was possible to inject CO₂ at an injection rate of over 8t/day, but the last value of CO₂ injection rate at 15.6 MPa was below 4 t day^-1 (Fujioka et al., 2010).

Van Bergen et al. (2009) observed different swelling behaviors of coal with different substances which carbon dioxide leads to higher strain than methane while exposure to organ leads to very little swelling. The experiments on moisturized specimens seem to confirm the role of moisture as a competitor to gas molecules for adsorption sites. A re-injection of carbon dioxide, after intermediate gas release, results in higher strains which indicate that drying effect of the carbon dioxide on coal specimens Van Bergen et al. (2009).

Balan and Gumrah (2009) revealed that swelling increased CO₂ breakthrough time and decreased displacement ratio and CO₂ storage for all ranks of coal. In addition, low-rank coals affected more negatively than high-rank coals by swelling and dry coal specimens are more influenced by swelling than saturated wet coals and saturated wet coals are more suitable for eliminating the negative effects of CO₂ injection and these results agree with the results that found by (Day et al., 2008, 2011). Thus it is possible to reduce swelling effect of CO₂ on cleat permeability by mixing it with N₂ before injection but this could happen temporary (Balan and Gumrah, 2009). Swelling phenomenon is sensitive to coal rank and moisture and water content.

Pressure and temperature: Perera et al. (2012) carried out experiments on naturally fractured bituminous coal specimens under five gas injection pressures (8, 9, 10, 11 and 13 MPa) under two different confinements (20 and 24 MPa) and five different temperatures (25, 30, 40, 50 and 70°C). The results showed increase in the permeability of naturally fractured black coal with increasing temperature (over 40°C) for supercritical CO₂ injection at higher injection pressures (more than 10 MPa) for any confinement and the permeability increment increases with increasing injection pressure. However, temperature has no much effect on permeability for low injection pressures (less than 9 MPa). In contrast, at low temperatures (less than around 40°C) CO₂ permeability decreases with increasing injecting pressures and this due to supercritical CO₂ adsorption-induced swell as shown in Fig. 5. Meanwhile, at higher temperatures (more than 50°C), permeability increases with increasing injecting pressures.

The influence of temperature on N₂ permeability was negligible compared to the CO₂ permeability, which is basically due to the fact that N₂ is a non-reactive gas (inert) which does not make any adsorption or swelling effect in coal matrix (Perera et al., 2012). In addition to that, temperature did not directly affect the maximum amount of swelling, however, the swelling tended to occur at lower pressures with decreasing temperature. Moreover, expressing the swelling as a function of gas density rather than gas pressure showed that swelling was independent of temperature (Day et al., 2008).

The effect of swelling strain on coal permeability was investigated using two types of Australian coal specimens and tests were carried out under low different gas injection pressures (2, 2.5, 3 and 3.4 MPa) and under confining pressures (6, 8, 10 and 11 MPa). In addition to that CO₂ and N₂ gases were used as injection media. However, gas injection was carried out with two stages of N₂ injection, prior to and after CO₂ injection. As result, for all specimens, the second N₂ injection showed a clear permeability reduction compared to the first N₂ injection, after exposure to CO₂ (Jasinge et al., 2012). The coal swelling percentage increased on exposure to carbon dioxide compared to exposure to N₂ and this effect increased as gas injection pressure increased and exposure of the coal specimens to CO₂ has contributed to a detrimental injection pressure increased. Increasing pore pressure increases swelling and decreases permeability as shown in Fig. 6 (Jasinge et al., 2012).

Injection pressure and pore pressure have highly influence on increasing coal swelling at low temperatures and effective stress decreases swelling strain at constant confining pressure as shown in Fig. 6 and 7 (Wang et al., 2011).

Swelling strain increases vastly at low CO₂ concentrations but increases steadily at high CO₂ concentrations and the same behaviour with CO₂ partial pressure in gases mixture that swelling strain increases vastly at low CO₂ partial pressure but increases steadily at high CO₂ partial pressure as demonstrated in Fig. 8 and Fig. 10. Nevertheless, Swelling decreases by time as shown in Fig. 9.

CO₂ phase and time: Perera et al. (2012) have conducted experiments on naturally fractured black coal specimens at 2-20MPa injection pressures under 10 to 24 MPa confining pressures for sub/super critical CO₂ and N₂ gas at 33°C.

The experimental results depicted that the permeability of naturally fractured black coal is significantly reduced due to matrix swelling which starts as quickly as within 1 h of CO₂ injection. A further reduction is then observed and the maximum swelling rate occurs within the first 3-4 h of CO₂ adsorption. The amount of coal matrix swelling due to CO₂ adsorption clearly depends on the phase condition of the CO₂ as depicted in Fig. 8. And super-critical CO₂ adsorption-induced swelling is about two times high than that induced by sub-critical CO₂ adsorption (Perera et al., 2011;Day et al., 2012). Interestingly, although a fractured coal specimen which has already fully swelled under sub-critical CO₂ adsorption can swell significantly more under super-critical CO₂ adsorption. However, after that conditions no further swelling effect occurs under any CO₂ pressure or phase condition as depicted in Fig. 9. Moreover, the swelling process continues longer under super-critical CO₂ adsorption. It is concluded that super-critical CO₂ adsorption can induce more matrix swelling than sub-critical CO₂ adsorption under the same adsorption pressure. If the effect of adsorption time on swelling is considered, the swelling rate decreases drastically with time as shown in Fig. 9. Furthermore, the maximum swelling rate can be observed within the first 3-4 h of CO₂ adsorption and this swelling process ends the third day of CO₂ adsorption process (Perera et al., 2011).

Gas composition: During enhanced coalbed methane recovery or CO₂ sequestration, the composition of the gas within the seam will change with time while the total gas pressure remains approximately constant. Day et al. (2012) reported that swelling results show a pressure dependence on the ratio of CO₂/methane swelling as shown in Fig. 10 and 11.

Despite the linear dependence of swelling on the CO₂ concentration for different coals as depicted in Fig. 13, slight differences in the slope of the relationship at different pressures leads to different CO₂/CH₄ swelling ratios across the pressure range and there is no enhanced swelling in fixed mixed gases (100% CO₂, 80% CO₂/20% CH₄, 50% CO₂/50% CH₄, 20% CO₂/80% CH₄ and 100% CH₄) would be observed in the pure critical CO₂ at the same total pressure (Day et al., 2012).

In gases mixture of fixed composition, the swelling induced was between that pure critical CO₂ and methane with swelling increasing in proportion to the concentration of CO₂ in the mixture. It is also apparent that in mixed gases, the primary driver for swelling is the partial pressure of the component gases (Day et al., 2012) as demonstrated in Fig. 12. Exposure of coal specimens to other gases such as amines, H₂S or NO had no effect on the sorption characteristics of the coal. In contrast, SO₂ markedly reduced the CO₂ sorption capacity of the coal by 25% and SO₂ modified the minerals matter extensively and reacted with clays and carbonated and producing a range of sulfate minerals and amorphous materials (Sakurovs, 2012). In contrary, Bustin (2004) conducted experiments for adsorption isotherm and volumetric measurements on British Colombia coals with various gases. It is concluded that the overall the sorption capacity and volumetric strain increase with increasing pressure steps until they reach their saturation pressures. Comparison of strain at about 0.6 MPa pressure step show that the volumetric strain (swelling) with H₂S is very high and about 20 times higher than CO₂ and about 70 times higher than CH₄. The order of volumetric swelling at 0.6 MPa in decreasing order is H₂S (up to 10.0%)> CO₂ (up to 0.66%)> CH₄ (up to 0.29%)> N₂ (up to 0.026%) (Bustin, 2004).

CONCLUSION

Swelling phenomenon occurs during carbon dioxide (CO₂) injection into coal seams at in situ conditions and triggers complex chemo-physical adsorption and chemical reactions within coal seams. Thus coal rank, moisture/water content, injection gas pressure, confining pressure, effective stress and temperature and CO₂ injection duration time have a significant influence on coal sorption induced strain and coal adsorption capacity. However, the relationship between swelling and CO₂ injection pressure are not constant particularly with coal specimens saturated with water meanwhile swelling in dry coal specimens increases with increasing CO₂ injection pressure. Nevertheless adsorption capacity increases by increasing CO₂ injection pressure. However, temperature has no much effect on permeability for low injection pressures. The moistured coal specimens swell less than dry coal specimens and the degree to which the swelling of the coals was affected by moisture depended on the coal rank. Super-critical CO₂ can induce more matrix swelling than sub-critical CO₂ adsorption under the same adsorption pressure and if the effect of adsorption time on swelling is considered. Swelling phenomenon is sensitive to effective stress. Hence swelling strain decreases with increasing effective stress. The volumetric strain (swelling) with SO₂ and H₂S were markedly very high compared to volumetric strain caused by CO₂ and CH₄. Similarly swelling caused by CO₂ is 2-5 times higher than the CH₄ Shrinkage and CH₄ shrinkage is 10 times higher than the swelling caused by the N₂ adsorption. Due to SO₂, CO₂ and H₂S have strong adsorption and swelling effects, sequestration of SO₂ or H₂S or CO₂ into coal specimens significantly reduced cleat permeability and affected strongly the injection efficiency. Particularly, SO₂ and H₂S injection will likely causes the coal to become impermeable. The volumetric strain is one of the main difficulties exists during CO₂ injection field trial into coal seams and it is difficult to measure.

ACKNOWLEDGEMENT

The authors gratefully acknowledge and thank the financial support provided to this research program through the Ministry of Science, Technology and Innovation (MOSTI) under e-science grant and Universiti Teknologi PETRONAS (UTP).

REFERENCES