Adsorption Behavior of CO2 in Coal and Coal Char

Jul 1, 2014 - ABSTRACT: Recent interest in sequestration of carbon dioxide (CO2) in gasified coal seam (i.e., post-underground coal gasification sites...
1 downloads 0 Views 6MB Size
Download PDF
Article pubs.acs.org/EF

Adsorption Behavior of CO2 in Coal and Coal Char Shanmuganathan Ramasamy,† Pavan Pramod Sripada,† Md Moniruzzaman Khan,† Su Tian,† Japan Trivedi,‡ and Rajender Gupta*,† †

Canadian Centre for Clean Coal, Department of Chemical and Materials Engineering, ‡School of Mining and Petroleum, Department of Civil and Environmental Engineering, University of Alberta, Edmonton, Alberta T6G 2G6, Canada ABSTRACT: Recent interest in sequestration of carbon dioxide (CO2) in gasified coal seam (i.e., post-underground coal gasification sites) has created a need to understand the coal properties, specifically, the adsorption behavior of CO2 on gasified coal. In the present study, the CO2 excess adsorption isotherms were determined for four coal samples of different characteristics based on the volumetric method. Further, coal chars from a coking coal and a non-coking coal (within the studied samples) were investigated for their CO2 adsorption capacity. The coal samples of size 22−32 mm were pyrolyzed in a drop-tube furnace at 800 and 1000 °C with a heating rate of approximately 2.5 °C s−1 under an inert atmosphere. Measurements were performed up to a pressure of 65 bar for all of the studied samples. Experiments were carried out at an isothermal temperature of 45.5 °C. The influence of coal properties on adsorption was also studied and compared to the literature data. Behavior of adsorption capacities was analyzed as a function of coal properties, such as vitrinite content, coal rank, volatile matter, ash content, and surface area. Results indicated that the adsorption capacity of coal char is much higher in comparison to the virgin coal samples. It was understood from the surface area analysis that there is a significant increase in surface area when coal is pyrolyzed. In addition, for coal samples, the trend of adsorption isotherms was in good agreement with the literature data.

■

INTRODUCTION The problems of increased global warming and severe climate change have resulted in efforts of reducing CO2 emissions.1−3 Among many options to reduce CO2 levels in the atmosphere, carbon capture and storage (CCS) in geological sites (e.g., coal seams and sedimentary basins) is competitive in terms of cost effectiveness and environmentally friendly nature.1 The geological sequestration of CO2 in coal formations needs an understanding in the relationship between the coal properties and the sorption capacity. This study is concerned with the adsorption behavior of coal and coal char, with the aim of assessing the CO2 storage potential, specifically, to determine the influence of coal properties in the implementation of CCS in underground coal gasification (UCG) sites. Coal has significant importance in CO2 sequestration because of its unique pore structure and fissure system, with micropores (500 Å).4,5 Moreover, it was understood that the gas sorption capacity is mainly influenced by the micropore structure apart from the surface properties of the matrix.6 However, this study details the pore morphology behavior of coal and coal char. The amount of micro-, meso-, and macropores present is governed by the coal rank and maceral composition.7 The fissure system aids in the transport of gases within the coal seam, and the microporous structure provides a higher surface area, leading to greater adsorption capacity. The diffusivity and gas sorption characteristics make coal a better adsorbent than other reservoir materials for the storage of CO2.8 In addition, the greater affinity of coal for CO2 compared to other gases, such as methane and nitrogen, has encouraged technologies, such as enhanced coal bed methane recovery (ECBMR), for the implementation of CCS in coal seams.9 The UCG process is economically attractive for the use of unminable coal seams. This technology, first conceived in the © 2014 American Chemical Society

late 19th century, has gained a lot of attention recently with the growing demand for energy and the threat of global climate change.10 The successful commercial-scale implementation of the UCG technique can significantly increase the world coal reserve estimates.11 Several pilot-scale studies were carried out in many countries, such as Russia, the U.S.A., Canada, Australia, China, and India, for demonstration of commercial-scale feasibility.11 From the reported studies, it was understood that the post-UCG site may be a potential avenue for the implementation of the CCS system.12 Moreover, it was also understood from the literature that it is possible to store CO2 in UCG cavities below 800 m to provide a hydraulic seal.13 Further, the analysis suggests that the porosity, fractures, and cracks developed during UCG are sufficient to accommodate all CO2 produced from the process.13 It was noted that the porosity and cracks in the coal seam play an important role in not only gasification but also structural properties of the cavity after gasification. However, until today, there are very few reports that assess CO2 storage capabilities in the coal seams subsequent to gasification. Figure 1 represents the linked vertical well (LVW) UCG technique, where the coal seam is gasified in an oxidizing atmosphere (steam and oxygen mixture) to yield a mixture of H2, CO, and CO2, commonly known as syngas. During gasification, many processes, such as drying, pyrolysis, combustion of volatiles, and char combustion, take place.14 In general, post-gasification sites consist of a tear-drop shape cavity with ash and rubble left at the bottom (see Figure 1). The residual coal seam can be broadly classified as the partially gasified char layer, the pyrolyzed char layers, and the raw coal Received: January 24, 2014 Revised: June 14, 2014 Published: July 1, 2014 5241

dx.doi.org/10.1021/ef500239b | Energy Fuels 2014, 28, 5241−5251

Energy & Fuels

Article

Figure 1. Post-burn cavity for the LVW UCG process.

histories. During formation of chars, development of the pores occurs by opening originally closed pores, creating new pores, and increasing the pore size of existing and newly formed pores.13 Naturally, it is expected for coal char that the adsorption capacity will be enhanced as a result of the increase in porosity and surface area in comparison to virgin coal.15 However, the development of the porosity and surface area in coal char is dependent upon the rank, nature, and maceral composition of coal. Further, the surface area development in coal char has a great influence on the adsorption and transport characteristics of CO2. There have been several investigations in the field of coal adsorption; however, the influence of coal properties, such as rank, carbon content, maceral composition, surface area, moisture, and porosity, are not well understood for pyrolyzed coal char, particularly in the UCG process. The study presented in this paper is an attempt to address this gap in understanding. The significance of the present study is that it provides fundamental insight on the storage capacity of a coal char sample as a function of the coal rank and pyrolyzed temperature. Such fundamental knowledge is essential for the development of the CCS system, in a post-UCG site. This paper reports the results of adsorption measurements of four virgin coal samples, namely, low-volatile sub-bituminous coal A (non-coking), low-volatile bituminous coal B (coking),

layer. The partially gasified char layer consists of coal that is not completely converted under the reactive atmosphere. On the other hand, the pyrolyzed char layers comprise of a coal seam that has been exposed to high temperatures in a relatively inert atmosphere (pyrolysis). As seen in Figure 2, the pyrolyzed char

Figure 2. Schematic illustrating the different strata of the pyrolyzed coal char layer.

layers will have diverse characteristics in different regions because they have been exposed to different time−temperature

Figure 3. Schematic of the volumetric adsorption apparatus: (1−6) valves, (7) pressure-relief valve, (8) check valve, (9) high-pressure syringe pump, (10) vacuum pump, (11) reference cell, (12) sample cell, (13 and 14) temperature sensor, (15 and 16) pressure transducer, (17) water bath, and (18) data acquisition system. 5242

dx.doi.org/10.1021/ef500239b | Energy Fuels 2014, 28, 5241−5251

Energy & Fuels

Article

Figure 4. Procedure for the gas adsorption measurement. transducers, thermocouples, and valves, are placed in a water bath with high-precision temperature controller (Thermo Scientific model 253). In addition, the inline thermocouples were integrated next to each cell (i.e., reference cell and sample cell) for measuring the gas temperature, which plays a vital role in determining the gas compressibility factor. Further, the reference and sample cells were incorporated with an in-line filter to avoid fine solid particles entering the tubing and valves. The evacuation segment consists of a vacuum pump and vent. During the experiment, the excess gases were removed through a vent, which was directed to the fume hood. The data acquisition system includes both pressure (Omega PX409-2.5KGUSB) and temperature (NIDAQ 9211) monitoring devices. Figure 4 schematically shows the steps involved in the gas adsorption measurement. Experiments were carried out at an isothermal environment for single-injection pressure only. Prior to any adsorption measurement, the trace gases were evacuated from the experimental setup through a vacuum pump. Subsequently, the gas was injected into the reference cell for a desired pressure with the help of the gas injection system. Further, the gas was retained in the reference cell to stabilize and obtain an isothermal condition. Later, the gas was allowed to equilibrate with the sample cell for a prolonged period (i.e., approximately 5 h) until equilibration (pressure tolerance of 0.07 bar) was obtained. In the volumetric adsorption method, it is essential to predetermine the void volume in the sample cell for measuring the sorption capacity of a sample. The void volume is defined as the space available in the sample cell for the gas to occupy. For instance, if a porous sample is considered, the void volume is the volume occupied by the gas apart from the solid surface. In this study, the void volume was determined through a helium expansion method. To obtain the porosity of the coal sample, the bulk density measurements were also carried out using a glass bead displacement technique adopted by Ramasamy et al.18 During adsorption, two different components of the gas phase are known to exist. The gas present directly on the coal surface is referred as the adsorbed phase. On the other hand, any gas that is not adsorbed on the coal surface is a part of the bulk gas phase. Properties of the

high-volatile sub-bituminous coal C (non-coking), and medium-volatile sub-bituminous coal D (non-coking). Coals of diverse characteristics have been chosen to provide a better understanding on the influence of various coal properties, such as maceral, volatile matter, and ash contents. In addition, char samples from two of these coals (a non-coking coal A and a coking coal B) were prepared by pyrolysis at 800 and 1000 °C in a nitrogen atmosphere and were tested for CO2 adsorption capacity. The volumetric adsorption measurements were performed at a controlled temperature of 45.5 °C up to a pressure of 65 bar. Specifically, the selection was made to avoid the supercritical region of CO2 and to represent the deep coal seam temperature (27−67 °C).16 The results on effects of the pyrolysis temperature on the microstructure of coal char and the subsequent influence on adsorption are reported and interpreted.

■

EXPERIMENTAL SECTION

The adsorption capacities of the coal and the coal char samples were determined using a custom-designed volumetric adsorption setup to handle large samples weighing nearly 20 g. Such large amounts (in comparison to commercial devices) were required to characterize the heterogeneity in the coal samples. The volumetric adsorption method was chosen because of its advantages, such as simplicity in laboratoryscale implementation and cost effectiveness, in contrast to gravimetric devices, which require extremely sensitive microbalances.17 Figure 3 shows the schematic representation of the volumetric adsorption apparatus. The equipment consists of four segments, namely, the gas injection system, isothermal section, evacuation section, and data acquisition system. The gas injection system consists of a high-pressure syringe pump (Teledyne ISCO 500D) with a rated delivery pressure of 258 bar. To ensure safety, a pressure release valve was installed at the output of the syringe pump. In the isothermal segment, the components, such as the reference cell, sample cell, pressure 5243

dx.doi.org/10.1021/ef500239b | Energy Fuels 2014, 28, 5241−5251

Energy & Fuels

Article

Table 1. Properties of Coals A and B19

adsorbed phase are different from that of the bulk gas phase. However, it is difficult to distinguish the bulk gas phase and adsorbed phase macroscopically; thus, it becomes impossible to experimentally measure the properties of the adsorbed phase, such as its density and volume. This limitation prevents the representation of experimentally measured adsorption amounts directly in terms of adsorbed phase properties. In this paper, the experimentally determined amounts of gas adsorbed on coal and coal char samples are denoted by excess adsorption or Gibbs surface excess (GSE) adsorption. In the GSE model, there exists an interface separating the adsorbed gas phase and the bulk gas phase, which is called the Gibbs interface. The GSE model assumes that the adsorption occurs at the bulk gas phase density rather than the actual adsorbed phase density. Thus, the GSE adsorption is defined as the difference between the amount adsorbed at the actual adsorbed gas phase density and the amount adsorbed at the bulk gas phase density. For a volumetric adsorption system at initial conditions (i.e., before injection), the amount of gas present in the reference cell can be written as

n initial = ρ1VR

parameter

(1)

where ρ1 is the molar density of the gas at the set temperature and pressure. VR is the volume of the reference cell (inclusive of the valves and the tube fittings). After gas equilibration, the mass balance of the gas presence in the system can be written as

ρ1VR = ρ1*VR + ρ1*(Vcell − Vsolid − Va) + ρa Va

coal B

coal C

coal D

4.45 15.40 28.89 50.26

2.81 10.58 39.60 52.31

0.54 63.29 3.97 1.07 15.73

0.84 64.34 4.11 0.95 19.18

65.20 22.10 2.10 0.56

67.8 17.90 7.50 0.69

dose quantity of 0.25 cm3/step with a maximum equilibration delay of 1 h/step. The adsorption data were obtained by maintaining constant test conditions, which allow for the comparisons to be made between the various samples in the study.

■

RESULTS AND DISCUSSION Adsorption of the Virgin Coal Sample. Figure 5 shows the excess adsorption capacity in virgin coal as a function of the density for both experimental and literature samples. Within the reported samples (see Figure 5), coals A−D are obtained by experiment at 45.5 °C, while on the other hand, A3 Australian coal, S2 Switzerland coal, and S3 Switzerland coal represent the literature data22 obtained at 45 °C. The coal samples chosen from the literature data are very similar to coal samples examined in this study (see Table 1). The term density here refers to the isothermal molar density of CO2 as a function of the pressure. The figure clearly shows a significant rise in excess adsorption capacity with an increase in density for the reported samples. Even though the magnitude of adsorption is different, the increase in excess adsorption per unit rise in density is almost identical for the experimental samples (i.e., coals A−D). Within the studied range of density, except for coals C and D, it was observed that the excess sorption capacity for the reported virgin coal samples is below 1 mmol/g. The higher adsorption in coals C and D can be due to the influence of the inherent coal properties, which increase the microporous surface area. However, from the literature, it was understood that the storage potential in virgin coal is comparatively less than other commercial adsorbents.23 It is noticed that the level of adsorption is much higher in the case of literature samples compared to coals A and B. However, the adsorption of coals C and D is greater than all of the literature samples. This difference in adsorption magnitude within the experimental and literature samples may be only attributed to the properties of the coal sample. Among the reported samples, coal C has the highest adsorption capacity and coal A has the lowest adsorption capacity. It was understood from Table 2 that the percentage of ash is much higher for coal C than that of coal A. However, the moisture, volatile matter, and fixed carbon are greater for coal A. Many researchers have shown that the adsorption decreases with the increase in macroscopic properties, such as moisture, ash, and volatile matter content.24−26 However, considering the ash and fixed carbon

(2)

In eq 2, Vcell is the total volume of the sample cell (i.e., the valves and the tube fittings volumes are included). Vsolid is the skeletal volume of the coal (does not include the volume in the pores). Actual free space available in the system is given by Vcell − Vsolid − Va. Va and ρa are the volume and density of the adsorbed phase, respectively. From the definition of GSE, the mass balance in eq 2 can be represented as

nexcess = VR (ρ1 − ρ1*) − ρ1*(Vcell − Vsolid)

coal A

Proximate Analysis (wt %) moisture (ad) 5.01 1.26 ash (ad) 10.39 14.07 volatile matter (daf) 31.0 27.0 fixed carbon (ad) 58.39 61.85 Ultimate Analysis (wt %, Dry Basis) sulfur 0.56 0.61 carbon 66.8 79.4 hydrogen 3.86 4.48 nitrogen 0.78 1.46 oxygen 8.99 3.83 Petrographic Analysis vitrinite (%) 32.2 44.8 inertinite (%) 52.1 44.9 liptinite (%) 4.5 5.1 maximum virinite reflectance 0.57 1.24 (Rmax, %)

(3)

where nexcess is the GSE adsorption or excess adsorption. As stated earlier, virgin coal samples considered for the adsorption measurements include coals A, C, and D, which are of low-, high-, and medium-volatile sub-bituminous rank, respectively. On the other hand, coal B is a low-volatile bituminous coal. Further, among these coals, coals A, C, and D are non-coking coals, whereas coal B is a coking coal. The main reason for this selection of coals with varying characteristics (see Table 1) is to understand the influence of coal properties on CO2 adsorption capacity. Moreover, special emphasis is given to the coal nature because the resultant pyrolyzed char structure is significantly dependent upon the nature of coal samples. Coals A and B, representing different ranks, were pyrolyzed to understand the effect of coal properties on CO2 adsorption capacity of coal chars. The size of the coal sample ranges from 22 to 32 mm, respectively. Coal samples were pyrolyzed at 800 and 1000 °C in a nitrogen environment for 20 min under an estimated heating rate of 2.5 °C/s using a droptube furnace.19 During pyrolysis, coal develops plasticity and, subsequently, resolidifies to form char. The degree of plasticity depends upon the coal nature and has a significant impact on the coal char structure.20 It was observed that the higher rank coals develop greater plasticity compared to the low-rank coals at high temperatures,21 which can potentially block the pores in high-rank coal during the pyrolysis process. Hence, this behavior reduces the surface area available for adsorption. The surface area of the coal and coal char samples was determined by the density functional theory (DFT) model that was carried out in Micromeritics ASAP 2020 using nitrogen as the probe molecule at 77 K. Initially, samples were heated to 250 °C at a ramp rate of 5 °C/min until a vacuum level of 0.5 Pa was reached. After the vacuum set point was reached, the samples were outgassed for 4 h. Surface area and micropore analyses were carried out on ∼400 mg samples using a fixed 5244

dx.doi.org/10.1021/ef500239b | Energy Fuels 2014, 28, 5241−5251

Energy & Fuels

Article

Figure 5. Comparison of excess adsorption data for the experiment and literature in the virgin coal sample.

Further, a very weak correlation is exhibited by the three coal samples considered in the literature data.22 In the case of the reported literature data,28 a sharp rise in the adsorption level was only observed for low-volatile bituminous and the semianthracites (i.e., Rmax = 1.5−2.5%). However, a weak correlation is observed for the coals with the reflectance below 1.5%.5 The higher rank coals with more aromatic hydrocarbons, short aliphatic chains, and depleted oxygen content may be responsible for the greater sorption.30 Many researchers have observed a “U-trend” for the relationship between adsorption and reflectance for coal ranging from lignite to semianthracite.31,32 This phenomenon may arise because of the “plugging” of the pores in the medium-rank coals. These pores gradually open up as the coalification process continues, leading to higher sorption capacities in high-rank coals.33 Figure 7a compares the adsorption capacity as a function of the volatile matter on an as-received (AR) and ash-free (AF) basis in virgin coal for both experimental and literature samples. The experimental data from this study were obtained at a pressure of 24 bar and temperature of 45.5 °C. Further, the experimental data were compared to literature data22,25 obtained at a pressure of 24 bar and temperatures of 45 and 26 °C. The trends of AR and AF basis obtained from Figure 7a clearly show that the excess adsorption decreases with an increase in the volatile matter content for both experimental and literature samples. In addition, it is to be noted that there is a large difference in the adsorption magnitudes between coals A and C. The large difference may be due to few properties of coals (i.e., mineral or maceral content) that dominate the sorption in coals. However, the general decrease in adsorption with the volatile matter can be explained by the fact that the macropore volume increases with an increase in the volatile

Table 2. Properties of Literature Coal Samples in Figure 5 coal sample moisture (%) volatile matter (%) fixed carbon (%) ash (%) Rmax (%)

22

Australian A3 0.39 17.65 64.16 17.80 1.34

Switzerland S2 1 28.80 53.50 16.70 0.85

22

Switzerland S3

22

0.8 26.70 44.20 28.30 0.90

contents, the trend from the literature is in contradiction for coals A and C (see Figure 5). This observation suggests that the microscopic properties, such as the maceral content, might have a stronger influence on the CO2 adsorption.27 Hence, it is essential to understand the effect of vitrinite, inertinite, liptinite, and mineral matter composition on the adsorption capacity. Effect of Coal Properties. Figure 6 shows the relationship between excess adsorption and vitrinite reflectance for virgin coal obtained by the experiment and literature. Experimental and literature22 data were obtained at 24 bar and temperatures of 45.5 and 45 °C, respectively. On the other hand, literature data28 were obtained at a pressure of 20 bar and a temperature of 25 °C.28,29 From Figure 6, it can be seen that a weak correlation is exhibited by the experimental samples that lie in the sub-bituminous to bituminous region. Coals A and C have almost the same reflectance, but the sorption capacity of coal C is much higher. The reason behind such a large difference in adsorption magnitudes can be attributed to disparity between the properties of coals A and C. Further, coals D and B have increasingly higher reflectances than that of coal C; however, there is a steady decrease in the sorption capacity. This can be explained by the fact that the amount of vitrinite is very high in coals C and D, followed by the amount in coal B. It is not only the reflectance but also the amount of vitrinite that play important factors in determining the amount of CO2 adsorbed. 5245

dx.doi.org/10.1021/ef500239b | Energy Fuels 2014, 28, 5241−5251

Energy & Fuels

Article

Figure 6. Excess adsorption behavior of virgin coal as a function of the coal rank.

pressure of 24 bar and temperature of 45.5 °C. From Figure 8, it can be seen that, the higher the vitrinite content, the greater the adsorption for both experimental and literature data. Within the studied samples, coals C and D have much greater vitrinite content than coals A and B, leading to a larger adsorption capacity. Particularly, the greater vitrinite content in coal C is probably the cause for large adsorption magnitudes compared to the other experimental samples. Moreover, despite the similar vitrinite content in coals C and D, there is a notable difference between their adsorption magnitudes. The difference may be due to the influence of other major maceral components, such as inertinite and liptinite. However, it may be speculated that the presence of greater vitrinite overpowers the role of other coal properties in determining the adsorption capacity in virgin coals. Furthermore, it is reported in the literature that vitrinites possess greater micropore volumes in comparison to other prominent maceral components, such as inertinites and liptinites.7 For instance, from DFT surface area analysis, it was understood that the amount of micropores in coal B is much higher than that in coal A. This observation can be used to corroborate the relationship between the higher vitrinite content and higher micropores in virgin coals. In addition, it was noted that the micropores in coal account for a greater fraction of adsorption compared to the macro- and mesopores.5 Moreover, it was also observed that brighter banded coals characteristic of greater vitrinite content had higher microporous volumes in comparison to the dull isorank coals.27,37 The vitrinite trend is consistent with the work reported in the literature.30,35,37 However, few studies have reported that the influence of the vitrinite content was rankdependent.33 Adsorption studies on macerals have brought out some interesting observations. For instance, there was also a report suggesting that telocollinite, a maceral in the vitrinite group,

matter content.7,32 High-volatile coals tend to have higher liptinite content, which is mainly composed of macropores.34 As noted earlier, adsorption in coal is highly influenced by the micropore content. The difference in the excess adsorption magnitudes within the experimental samples and between the experimental and literature data is attributed mainly to the coal property rather than the minor variations in the test conditions (i.e., temperature). Specifically, for the literature data,22,25 even though there is a notable difference in the test conditions, the excess adsorption magnitudes are similar. This suggests that the influence in the coal property is much more pronounced. In addition, the very similar trends seen in Figure 7a between AR and AF basis indicate that the presence of ash does not seem to contribute to the adsorption capacity. However, from Figure 7b, it can be seen that the excess adsorption decreases slightly with an increase in the ash content for almost all samples in the experiment and literature, except for coal C. Coal C has a significantly higher sorption capacity than most coals, despite a higher ash content. This observation suggests that the maceral components may have a much more pronounced impact compared to the influence of mineral components. In general, the decreasing trend in sorption capacity versus ash content is almost identical to the studies reported in the literature.24,33 It can be speculated that the ash hinders the presence of active sites for adsorption in coal, thus implying that, the greater the ash content, the lesser the number of active sites per unit volume of coal, resulting in lesser adsorption. Further, the inorganic components, such as ash, in coal have a negative impact on adsorption,35 and CO2 adsorption in coal mainly takes place on the organic phase.36 Effect of the Maceral Content in Coal. Figure 8 shows the impact of the vitrinite content on the excess adsorption of virgin coals obtained from experimental and literature data.36 As stated earlier, the experimental data have been obtained at a 5246

dx.doi.org/10.1021/ef500239b | Energy Fuels 2014, 28, 5241−5251

Energy & Fuels

Article

Figure 7. Excess adsorption behavior of virgin coal as a function of the (a) volatile matter and (b) ash content.

may have a strong impact on gas adsorption magnitudes.35 Furthermore, the combination of vitrinite and inertinite contents gave a better correlation for adsorption in comparison to the individual maceral components.35,38 In inertinite, the lower adsorption may be attributed to the nature of porosity, cross-linking density, and surface functionality.7 On the other hand, liptinite, having a large amount of volatile matter and primary aliphatic constituents, possesses a greater amount of mesopores, leading to lesser surface area and lower adsorption.34,39 These reports help in explaining the notable difference in the adsorption magnitudes between coals C and D. From Table 1, it can be seen that coals C and D have almost similar vitrinite and inertinite contents but the liptinite content varies significantly. Because liptinite has a negative impact on adsorption, the greater liptinite content in coal D leads to lesser adsorption compared to coal C. Adsorption Behavior of the Coal Char Sample. Figure 9 shows the variation in excess adsorption capacity of coal A

and B samples at different pyrolyzed temperatures. The trend in Figure 9 shows a sharp rise in excess adsorption with an increase in CO2 pressure. For the studied range of pressures, the higher the CO2 pressure, the greater the adsorption. The impact of CO2 pressure on excess adsorption is much more pronounced for char samples in comparison to the virgin sample for both coals A and B. This behavior can be attributed to the pore structure changes that occur during pyrolysis, where the volatile matter release is accompanied by structural changes, such as the generation of new pores and coalescence of pores to form larger pores, leading to enhanced surface area.15,40 However, from Figure 9, it can be seen that there is no significant change in the adsorption capacities between chars 800 and 1000 for both coals A and B. At higher temperatures, the cross-links are broken between the aromatic rings and the resulting structural rearrangement alters the surface area of the chars. The altered surface area, either a slight increase or decrease (depending upon the coal), leads to only a marginal 5247

dx.doi.org/10.1021/ef500239b | Energy Fuels 2014, 28, 5241−5251

Energy & Fuels

Article

Figure 8. Comparison of excess adsorption data for the experiment and literature in the virgin coal sample as a function of the vitrinite content.

Figure 9. Comparison of excess adsorption data for coal and coal char samples. Figure 10. Excess adsorption behavior of virgin coal and coal char samples as a function of the surface area.

change in the sorption capacity for char 1000 species compared to char 800 species. Within the studied coal samples, the adsorption trend is much steeper for coal A char samples than the coal B counterparts. However, in the case of the virgin sample, the adsorption capacity for coal B is greater than that for coal A. In brief, this observation can be partly assigned to the nature of the virgin coals, which dictates the surface area of the produced coal char. In conclusion, it is essential to understand the influence of the surface area and the coal nature on adsorption. Figure 10 illustrates the excess adsorption as a function of the surface area for both virgin coal and coal char obtained by pyrolysis at temperatures of 800 and 1000 °C. Adsorption

measurements were performed at a pressure of 24 bar and an isothermal temperature of 45.5 °C. The trend in Figure 10 shows that there is a direct linear relationship between excess adsorption and surface area. In general, for coal samples, the greater the surface area, the higher the excess adsorption. Figure 10 clearly shows that the surface area is much higher for coal A char samples in comparison to coal B char samples. On the other hand, in the case of virgin coal samples, coal A has a lower surface area than coal B. Within coal A samples, there is a significant rise in the surface area between virgin and char 800 by 30 times. In contrast, the increase in the surface area was 5248

dx.doi.org/10.1021/ef500239b | Energy Fuels 2014, 28, 5241−5251

Energy & Fuels

Article

develop more micropores than coal B, which is responsible for the increase in the surface area for adsorption. Therefore, it can be concluded that the pore distribution and nature of the coal have substantial impacts on the adsorption behavior. Figures 12 quantifies the percentage of micro-, meso-, and macropore distribution in virgin and char 800 samples for both

only 2.5 times that for the case of the coal B sample. However, an abnormal behavior was observed for char 1000 samples unlike char 800 samples; despite the rise in the surface area, the adsorption was insignificant for coal A. Similarly, the effect of the pyrolysis temperature on the surface area was inconsequential for coal B. The above observation confirms that the characteristics of the surface area were influenced by the nature of the coal and its pyrolysis temperature, where coal B, being a coking coal, tends to develop lesser pores compared to the lower rank non-coking coal (coal A). Further, it was also understood from the literature that the surface area increases as the pyrolysis temperature increases until a critical temperature and decreases thereafter.41−43 As stated earlier, this behavior is due to the high-temperature structural rearrangement of the char matrix during the pyrolysis process.43 Although the relationship between the adsorption capacity and pyrolysis temperature can be explained in terms of the surface area, a detailed interpretation of pore properties, such as porosity and pore size distributions, will help to understand the adsorption behavior in char samples. Porosity and Pore Distribution Studies. Figure 11 shows the comparison of porosity between virgin and char samples

Figure 12. Porosity distribution in terms of the surface area for virgin coal and coal char of coals A and B.

coals A and B. The observation confirms (see Figure 12) that there is a substantial change in the pore distribution between virgin and char 800 samples, particularly for coal A. The rise in the surface area for coal A is more pronounced in micropores compared to the reduction in the surface area for meso- and macropores. However, a significant rise in the surface area was only observed for mesopores between coal B samples. Within the studied virgin samples, the micropores in coal B are reasonably higher than those in coal A. The higher micropores in coal B can be attributed to the greater vitrinite content compared to that in coal A, because vitrinites are known to be composed of more micropores. It was found from the experiment that, during pyrolysis at 800 °C, the distribution of meso- and macropores was partly transformed into micropores in coal A. As explained earlier, this behavior can be attributed to the structural changes, such as the generation of new pores and coalescence of pores that occur during pyrolysis.15,40 Conversely, this behavior was not observed for coal B, because being a coking coal, the excess tar produced may condense to form more mesopores.44 Further, the decomposition of metaplast (a viscous fluid expelled during initial stages of pyrolysis) produces carbon that may plug the pores and cause reduction in the micropore surface area.43 Therefore, it can be concluded that micropores have the greatest influence on adsorption because they offer a higher surface area compared to other pore sizes. Moreover, the pore distribution plays a vital role in understanding the impact of the coal nature on the adsorption behavior.

Figure 11. Porosity of the char species of coals A and B compared to the respective virgin coal samples.

obtained at different pyrolysis temperatures. The most significant feature is that the porosity of char samples was much higher than that of virgin samples for both coals. It was also found that the rise in porosity is around 4 times that for char 800 from the virgin sample. However, it is only 1.2 times that between chars 800 and 1000. As explained earlier, the increase in the porosity is due to the development of macropores and the growth of micropores, which has caused the pore expansion and new pore generation in char 800 samples. Further, the growth in porosity was affected because of the structural rearrangement at higher temperatures for char 1000 samples. Within the studied samples, the porosity is much more pronounced for coal A in comparison to coal B. This can be attributed to the nature of the coal samples. Coal B, being a coking coal, tends to expel more tar, which, upon condensation, may block the pores. However, it can be seen that the porosities of coals A and B vary marginally, but there is a substantial increase in the surface area. This is because coal A tends to

■

CONCLUSION In this paper, four virgin coals were tested for the CO2 adsorption using a volumetric setup. It was observed from the experimental results that the adsorption magnitude is influenced by coal rank and coal properties. Within the coal 5249

dx.doi.org/10.1021/ef500239b | Energy Fuels 2014, 28, 5241−5251

Energy & Fuels

Article

(9) Marsh, H. Adsorption methods to study microporosity in coals and carbonsA critique. Carbon 1987, 25 (1), 49−58. (10) Klimenko, A. Early ideas in underground coal gasification and their evolution. Energies 2009, 2 (2), 456−476. (11) Couch, G. R. Underground Coal Gasification; IEA CleanCoal Centre: London, U.K., 2009; Report CCC/151. (12) Self, S. J.; Reddy, B. V.; Rosen, M. A. Review of underground coal gasification technologies and carbon capture. Int. J. Energy Environ. Eng. 2012, DOI: 10.1186/2251-6832-3-16. (13) Younger, P. Hydrogeological and geomechanical aspects of underground coal gasification and its direct coupling to carbon capture and storage. Mine Water Environ. 2011, 30 (2), 127−140. (14) Bunt, J. R.; Waanders, F. B.; Nel, A.; Dreyer, L.; van Rensburg, P. W. A. An understanding of the porosity of residual coal/char/ash samples from an air-blown packed bed reactor operating on inertiniterich lump coal. J. Anal Appl. Pyrolysis 2012, 95, 241−246. (15) Mahajan, O. P.; Walker, P. L., Jr. Porosity of coals and coal products. In Analytical Methods for Coal and Coal Products; Karr, C., Jr., Ed.; Elsevier: Amsterdam, Netherlands, 1978; Chapter 4. (16) Hemert, P. V.; Floter, S. R.; Wolf, K.-H. A. A.; Bruining, J. Estimate of equation of state uncertainty for manometric sorption experiments: Case study with helium and carbon dioxide. SPE J. 2010, 15 (1), 146−151. (17) Chilev, C.; Pentchev, I.; Weinberger, B.; Lamari, B. A comparison between the different methods for the measurement of excess adsorption of pure gases on porous adsorbents at high pressure. J. Univ. Chem. Technol. Metall. 2007, 42 (1), 77−84. (18) Ramasamy, S.; Tian, S.; Khan, M. M.; Trivedi, J.; Gupta, R. Assessing CO2 storage potential in coal chars for UCG-CCS system. Proceedings of the IEA 2nd Underground Coal Gasification Network Workshop; Banff, Alberta, Canada, Aug 22−23, 2012. (19) Tian, S. Fragmentation of large coal particles at high temperature in a drop tube furnace. Master’s Thesis, University of Alberta, Edmonton, Alberta, Canada, 2011. (20) Solomon, P. R.; Serio, M. A.; Suuberg, E. M. Coal pyrolysis: Experiments, kinetic rates and mechanisms. Prog. Energy Combust. Sci. 1992, 18 (2), 133−220. (21) Suuberg, E. M.; Lee, D.; Larsen, J. W. Temperature dependence of crosslinking processes in pyrolysing coals. Fuel 1985, 64 (12), 1668−1671. (22) Pini, R. Enhanced coal bed methane recovery finalized to carbon dioxide storage. Ph.D. Thesis, ETH Zurich, Zurich, Switzerland, 2009. (23) Songolzadeh, M.; Ravanchi, M. T.; Soleimani, M. Carbon dioxide capture and storage: A general review on adsorbents. World Acad. Sci., Eng. Technol. 2012, 70, 225. (24) Gürdal, G.; Yalçın, M. N. Gas adsorption capacity of carboniferous coals in the zonguldak basin (NW turkey) and its controlling factors. Fuel 2000, 79 (15), 1913−1924. (25) Pashin, J. C.; Carroll, R. E.; Groshong, R. H. J.; Raymond, D. E.; McIntyre, M.; Payton, W. J. Geologic Screening Criteria for Sequestration of CO2 in Coal: Quantifying Potential of the Black Warrior Coalbed Methane Fairway, Alabama; National Energy Technology Laboratory, U.S. Department of Energy: Washington, D.C., 2003; Report DE-FC00NT40927. (26) Carroll, R. E.; Pashin, J. C. Relationship of sorption capacity to coal quality: CO2 sequestration potential of coalbed methane reservoirs in the black warrior basin. Proceedings of the International Coalbed Methane Symposium; Tuscaloosa, AL, May 5−7, 2003. (27) Crosdale, P. J.; Beamish, B. B.; Valix, M. Coalbed methane sorption related to coal composition. Int. J. Coal Geol. 1998, 35 (1−4), 147−158. (28) Ryan, B; Richardson, D. The potential for CO2 sequestration in British Columbia coal seams. In Carbon Dioxide Capture and Storage: A Compendium of Canada’s Participation; Legg, J. F., Campbell, F. R., Eds.; Office of Energy Research and Development Natural Resources Canada: Ottawa, Ontario, Canada, 2005. (29) Gluskoter, H.; Mastalerz, M.; Stanton, R. The potential for carbon dioxide sequestration in coal beds: New evidence from methane and carbon dioxide adsorption analysis for coals from lignite

properties, the vitrinite, volatile matter, and ash content showed a consistent trend for the studied samples with the literature data. It was particularly noted that the maceral distribution on the coal samples has a pivotal role in determining the adsorption capacities of virgin coals. From the four coal samples, coal A (a coking coal) and coal B (a non-coking coal) were chosen for the comparison of adsorption capacity to their respective pyrolyzed coal chars at 800 and 1000 °C. Results suggest that adsorption capacity of coal char samples is significantly higher than that of virgin coal samples. This increase in adsorption in coal chars is because of the enhanced surface area, which is a strong function of the coal nature and pyrolysis temperature. Further, pore size distribution studies provided a fundamental understanding about the adsorption behavior in coals A and B. From the studies, it was understood that the char obtained from non-coking coal (coal A) is more microporous than that from coking coal (coal B), thus leading to greater adsorption because of the enhanced surface area. The experiments performed in this study were below the critical pressure of CO2. However, to obtain a detail understanding of the UCG CCS system, experiments need to be performed above critical pressures for a wide range of coal samples.

■

AUTHOR INFORMATION

Corresponding Author

*Telephone: +1-780-492-6861. Fax: +1-780-492-2881. E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors thank the Canadian Centre for Clean Coal/ Carbon and Mineral Processing Technologies (C5MPT), University of Alberta, Canada, for the financial support that was received for the project. The authors also thank Dr. James Sawada and Dr. Arvind Rajendran for their valuable comments during the design of the adsorption facility. Finally, the authors also thank Todd Kinnee for his technical support.

■

REFERENCES

(1) Bachu, S. CO2 storage in geological media: Role, means, status and barriers to deployment. Prog. Energy Combust. Sci. 2008, 34 (2), 254−273. (2) Bachu, S.; Bonijoly, D.; Bradshaw, J.; Burruss, R.; Holloway, S.; Christensen, N. P.; Mathiassen, O. M. CO2 storage capacity estimation: Methodology and gaps. Int. J. Greenhouse Gas Control 2007, 1 (4), 430−443. (3) Pro Oxygen. CO2Now; Pro Oxygen: Victoria, British Columbia, Canada, 2013; www.co2now.org (accessed Nov 4, 2013). (4) Mahajan, O. P. CO2 surface area of coals: The 25-year paradox. Carbon 1991, 29 (6), 735−742. (5) Saghafi, A.; Faiz, M.; Roberts, D. CO2 storage and gas diffusivity properties of coals from sydney basin, australia. Int. J. Coal Geol. 2007, 70 (1−3), 240−254. (6) Liu, Y.; Wilcox, J. Molecular simulation of CO2 adsorption in micro- and mesoporous carbons with surface heterogeneity. Int. J. Coal Geol. 2012, 104, 83−95. (7) Clarkson, C. R.; Marc Bustin, R. Variation in micropore capacity and size distribution with composition in bituminous coal of the western Canadian sedimentary basin: Implications for coalbed methane potential. Fuel 1996, 75 (13), 1483−1498. (8) Harpalani, S.; Schraufnagel, R. A. Shrinkage of coal matrix with release of gas and its impact on permeability of coal. Fuel 1990, 69 (5), 551−556. 5250

dx.doi.org/10.1021/ef500239b | Energy Fuels 2014, 28, 5241−5251

Energy & Fuels

Article

to anthracite. Proceedings of the Geological Association of America Annual Meeting; Denver, CO, Oct 27−30, 2002. (30) Jian, X.; Guan, P.; Zhang, W. Carbon dioxide sorption and diffusion in coals: Experimental investigation and modeling. Sci. China Earth Sci. 2012, 55 (4), 633−643. (31) Day, S.; Duffy, G.; Sakurovs, R.; Weir, S. Effect of coal properties on CO2 sorption capacity under supercritical conditions. Int. J. Greenhouse Gas Control 2008, 2 (3), 342−352. (32) Prinz, D.; Littke, R. Development of the micro- and ultramicroporous structure of coals with rank as deduced from the accessibility to water. Fuel 2005, 84 (12−13), 1645−1652. (33) Laxminarayana, C.; Crosdale, P. J. Role of coal type and rank on methane sorption characteristics of Bowen basin, Australia coals. Int. J. Coal Geol. 1999, 40 (4), 309−325. (34) Hessley, K. R.; Reasoner, J. W.; Riley, J. T. Coal ScienceAn Introduction to Chemistry, Technology, and Utilization; WileyInterscience: Hoboken, NJ, 1986. (35) Mastalerz, M.; Gluskoter, H.; Rupp, J. Carbon dioxide and methane sorption in high volatile bituminous coals from Indiana, USA. Int. J. Coal Geol. 2004, 60 (1), 43−55. (36) Perera, M. S. A.; Ranjith, P. G.; Choi, S. K.; Airey, D.; Weniger, P. Estimation of gas adsorption capacity in coal: A review and an analytical study. Int. J. Coal Prep. Util. 2012, 32 (1), 25. (37) Clarkson, C. R.; Bustin, R. M. The effect of pore structure and gas pressure upon the transport properties of coal: A laboratory and modeling study. 1. Isotherms and pore volume distributions. Fuel 1999, 78 (11), 1333−1344. (38) Clarkson, C. R.; Bustin, R. M. Binary gas adsorption/desorption isotherms: Effect of moisture and coal composition upon carbon dioxide selectivity over methane. Int. J. Coal Geol. 2000, 42 (4), 241− 271. (39) Harris, L. A.; Yust, C. S. Transmission electron microscope observations of porosity in coal. Fuel 1976, 55 (3), 233−236. (40) de Koranyi, A.; Balek, V. Structural changes in coals during pyrolysis. Thermochim. Acta 1985, 93, 737−740. (41) Razouk, R. I.; Saleeb, F. Z.; Youssef, A. M. The effect of thermal treatment on the surface area of coal. Carbon 1968, 6 (3), 325−331. (42) Gutierrez, M.; Cukierman, A. L.; Lemcoff, N. O. Study of subbituminous coal chars: Effect of heat treatment temperature on their structural characteristics. J. Chem. Technol. Biotechnol. 1988, 2, 85. (43) Jiang, W. D.; Lee, I.; Yang, R. Y. K. Microstructural variations of lignite, subbituminous and bituminous coals and their high temperature chars. Fuel Process. Technol. 1988, 18 (1), 11−23. (44) Suuberg, E. M.; Lee, D.; Larsen, J. W. Temperature dependence of crosslinking processes in pyrolysing coals. Fuel 1985, 64 (12), 1668−1671.

5251

dx.doi.org/10.1021/ef500239b | Energy Fuels 2014, 28, 5241−5251