The Interactions of Coal with CO2 and Its Effects on Coal Structure

a first-order desorption process with a single activation energy for desorption (Redhead equation).14 NP is the maximum desorption rate at peak te...
1 downloads 0 Views 112KB Size
Download PDF
2022

Energy & Fuels 2006, 20, 2022-2027

The Interactions of Coal with CO2 and Its Effects on Coal Structure Mojtaba Mirzaeian and Peter J. Hall* Department of Chemical and Process Engineering, UniVersity of Strathclyde, Glasgow, G1 1XJ, United Kingdom ReceiVed January 26, 2006. ReVised Manuscript ReceiVed May 30, 2006

The interactions of coal with CO2 at pressures of up to 30 bar concerning mechanisms of diffusion, the strength of interactions, and the irreversibility of uptake for the permanent disposal of CO2 into coal fields have been studied. Differential scanning calorimetry was used to investigate coal/CO2 interactions for North Dakota, Wyodak, Illinois No. 6, and Pittsburgh No. 8 coals. It was found that the first interactions of CO2 with coals led to strongly bound carbon dioxide on coal. Energy values attributed to the irreversible storage capacity for CO2 on coals were determined. The lowest irreversible sorption energy was found for North Dakota coal (0.44 J/g), and the highest value was for the Illinois No. 6 coal (8.93 J/g). The effect of highpressure CO2 on the macromolecular structure of coal was also studied by means of differential scanning calorimetry. It was found that the temperature of the second-order phase transition of Wyodak coal decreases with an increase in CO2 pressure significantly, indicating that high-pressure CO2 diffuses through the coal matrix, causes significant plasticization effects, and changes the macromolecular structure of the Wyodak coal. Desorption characteristics of CO2 from the Pittsburgh No. 8 coal were studied by temperature-programmed desorption mass spectrometry. It was found that CO2 desorption from the coal is an activated process and follows a first-order kinetic model. The activation energy for CO2 desorption from the Pittsburgh No. 8 coal increased with the preadsorbed CO2 pressure, indicating that CO2 binds more strongly and demands more energy to desorb from the Pittsburgh No. 8 coal at higher pressures.

Introduction The study of interactions between coals and CO2 is an essential issue in the long-term sequestration of CO2 within coal seams. Coal is a composite material consisting of a large variety of various compounds both organic and inorganic. It is a porous material with polymeric character built of aromatic and hydroaromatic units which are connected by methylene, oxygen, and sulfur cross links.1 At room-temperature, coal is a glassy solid, in the sense that it is amorphous, rigid, and brittle and the diffusion of gases and liquids in its structure is slow. When it is warmed to a temperature such that the thermal energy is greater than the intramolecular interaction energy, or it interacts with a solvent, it becomes rubbery and diffusion into its structure becomes faster.2-4 Coal behaves as a polymeric macromolecular system. Its porosity results in the entrance of fluids into its structure, and its polymeric nature accompanied with the presence of various functional groups leads to the chemical interactions of fluids with its matrix through noncovalent bonding and electron transfer. CO2 has long been used to measure coal surface areas by adsorption, and many low-pressure data are available.1,5-7 A NMR study of coals containing large amounts of dissolved * Author to whom correspondence should be addressed. Phone: 44-141548-4084. Fax: 11-141-552-2302. E-mail: [email protected]. (1) Green, T. L.; Kovac, J.; Brenner, D.; Larsen, J. W. In Coal Structure; Meyers, R. A., Ed.; Academic Press: New York, 1982; pp 199-282. (2) Larsen, J. W. Int. J. Coal Geol. 2004, 57, 63-70. (3) Nishioka, M.; Larsen, J. W. Energy Fuels 1990, 4, 100-106. (4) Giri, C. C.; Sharma, D. K. Fuel 2000, 79, 577-585. (5) Mahajan, O. P. Carbon 1991, 29 (6), 735-742. (6) Walker, P. L., Jr.; Kini, K. A. Fuel 1965, 44, 453-459. (7) Bond, R. L. Nature 1956, 178, 104-105.

molecules showed that the glass transition temperature (Tg) of coal drops with an increasing amount of dissolved molecules.8 The behavior of CO2 is similar to that of organic solvents that dissolve in and swell coals.2 It dissolves in coals and acts as a plasticizer, enabling rearrangement in the coal physical structure. Recently, an extensive review by White et al.9on the storage of CO2 in coal concerning the estimation of the CO2 storage capacity of coal; evaluation of the coal seam properties relevant to CO2 sequestration; chemical, physical, and thermodynamic events initiated when CO2 is injected into a coal bed, and other key issues related to CO2 sequestration in coal was published. This review shows that the interactions between coal and CO2 are not well-understood yet and, in particular, less is known about the high-pressure coal/CO2 interactions and possible effects of CO2 on coal structure. In the present study, we have applied differential scanning calorimetry (DSC) and temperature-programmed desorption mass spectrometry (TPD-MS) to study the high-pressure interactions between coal and CO2 at pressures up to 30 bar. Experimental Section Four coal samples obtained from Argonne Premium Samples in the lignite, sub-bituminous, and bituminous range were used in this study. An analysis of the coals is given in Table 1. DSC. The DSC measurements were done with a Mettler DSC 30 instrument in conjunction with Mettler software TA72PS for data acquisition and processing. (8) Yang, X.; Silbernagel, B. G.; Larson, J. W. Energy Fuels 1994, 8, 266-275. (9) White, C. M.; Smith, D. H.; Jones, K. L.; Goodman, A. L.; Jikich, S. A.; LaCount, R. B.; DuBose, S. B.; Ozdemir, E.; Morsi, B. I.; Schroeder, K. T. Energy Fuels 2005, 19, 659-724.

10.1021/ef060040+ CCC: $33.50 © 2006 American Chemical Society Published on Web 07/06/2006

Interactions of Coal with CO2

Energy & Fuels, Vol. 20, No. 5, 2006 2023

Table 1. Elemental Analysis of Coal Samples (% daf) coal

C

H

O

N

S

North Dakota lignite Wyodak Illinois No. 6 Pittsburgh No. 8

72.94 75.01 77.67 83.20

4.83 5.35 5.00 5.32

20.38 18.05 13.58 8.95

1.15 1.12 1.37 1.64

0.7 0.47 2.38 0.89

Standard aluminum pans were used, with two pinholes in order to minimize mass transfer limitations in the evaporation of water or the contact of gas with the sample during DSC scans. Nitrogen flowing at 10 mL/min was used as a carrier gas to keep the cell free of oxygen during measurements. Typically, 10 mg of the sample was used in an experiment. The DSC runs were performed at a heating rate of 10 °C/min. Cooling of the furnace between consecutive heating scans was carried out using a liquid nitrogen cooling accessory directly beneath the furnace. To study the adsorption of CO2 on coals, two different series of scans were conducted on each of the coal samples. (1) A sample of fresh coal was purged with N2 flowing at 10 mL/min for 10 min in a DSC chamber, then heated to 110 °C at 10 °C/min and held for 30 min and cooled at nominal rate of 100 °C/min to -60 °C. The sample was then heated from -60 to 200 °C at 10 °C/min three times in succession with a cooling rate of 10 °C/min between heating runs. In all scans, the sample always remained in the DSC chamber under nitrogen flowing at 10 mL/ min. (2) A sample of fresh coal was purged with N2 flowing at 10 mL/min for 10 min in a DSC chamber, then heated to 110 °C at 10 °C/min and held for 30 min and cooled at nominal rate of 100 °C/min to -60 °C under N2 flowing at 10 mL/min. At this point, the gas was switched to CO2, and the sample was then heated from -60 to 200 °C under CO2 flowing at 10 mL/min at 10 °C/min three times in succession with a cooling rate of 10 °C/min between heating runs. Differential scanning calorimetry was also employed to study the structural change in coal caused by high-pressure CO2. Experiments were carried out on Wyodak coal. The temperature programs were as follows: (1) A sample of Wyodak coal was purged with N2 flowing at 10 mL/min for 10 min in a DSC chamber. It was then heated from 30 to 110 °C at a heating rate of 10 °C/min, held at 110 °C for 30 min, and cooled to 30 °C in a N2 atmosphere at a flow rate of 10 mL/min. This was performed to remove water from the coal sample before determining its glass transition temperature. The dried sample was heated from 30 to 200 °C at 10 °C/min in a N2 atmosphere flowing at 10 mL/min three times in succession. This experiment was carried out to determine the phase transition in Wyodak coal in a N2 atmosphere. (2) A sample of Wyodak coal was purged with 20 bar Ar three times in succession, to flush adsorbed gases present in the pores in coal and also oxygen from the high-pressure cell. It was then heated to 110 °C, held for 30 min at 110 °C, and cooled to 30 °C at 10 °C/min under an Ar atmosphere in the high-pressure cell. The sample was loaded with CO2 to the desired pressure at room temperature. It was exposed to this CO2 environment for 24 h. After that, the CO2 pressure was rapidly released, and the sample was purged with 20 bar Ar and transferred to the DSC chamber. The sample was purged with N2 at a flow rate of 10 mL/min for 10 min in the DSC chamber, and DSC was carried out from 30 to 200 °C at 10 °C/min in a N2 atmosphere three times in succession. The purpose of this sequence was to determine the effect of highpressure CO2 atmospheres on the phase transition of coal. TPD-MS. CO2 desorption measurements from coal samples were carried out using a temperature-programmed furnace coupled to a Hiden Analytical HAL/HPR20 Quadrupole Mass Spectrometer (QMS). The mass spectrometer was set up to continuously monitor the evolution of CO2 from the sample. The desorption chamber was a quartz tube placed in a temperature-programmed furnace and connected to the gas inlet system of the QMS via a stainless steel capillary tube. Sampling of the effluent gases from the desorption chamber, operating at atmospheric pressure, was done through the

Figure 1. DSC for North Dakota coal from -60 to +200 °C in a N2 atmosphere.

Figure 2. DSC for North Dakota coal from -60 to +200 °C in a CO2 atmosphere.

capillary tube. A very precise needle valve at the entrance of the mass spectrometer was used to control the flow of gases to the MS analyzer. A Ni/Al thermocouple in the center of the furnace at a few millimeters above the sample was used for temperature measurement. The CO2 partial pressure and temperature data were collected by a computer via the HAL temperature interface and the mass spectrometer interface. Flowing helium with a flow rate of 100 mL/min was used to flush evolved gas into the mass spectrometer. A typical TPD experiment was as follows: Approximately 150 mg of the sample was placed in a sample holder and loaded with CO2 to the desired pressure at room temperature in a high-pressure cell. The sample was exposed to this high-pressure CO2 atmosphere for a certain period of time. Then, the CO2 pressure was rapidly released, and the sample was transferred to the desorption chamber and purged with high-purity helium for 10 min at 298 K, before commencing the TPD run. The gas flow rate was 100 mL/min. To perform a TPD scan, the sample was heated by linearly increasing the temperature, 20 K/min, and the evolution of CO2 from the sample was monitored by QMS.

Results and Discussion DSC. Figures 1-8 show the results of DSC runs for the adsorption of N2 and CO2 on coals. DSC thermograms for the adsorption of N2 on coal samples show that coal/N2 interactions are not strong and the adsorption of N2 on coals occurs physically and reversibly. For all coals, exothermic peaks attributed to the adsorption of N2 on the first, second, and third scans are almost the same. A small difference between the first scan and the subsequent two scans might be associated with the structural rearrangement and the relaxation process in the coal, because of heating above its glass transition temperature during the first run.10 Coal is subject to stress from overlying rocks during its formation. Because of its glassy structure, its molecular motion

2024 Energy & Fuels, Vol. 20, No. 5, 2006

Mirzaeian and Hall

Figure 3. DSC for Wyodak coal from -60 to +200 °C in a N2 atmosphere.

Figure 4. DSC for Wyodak coal from -60 to +200 °C in a CO2 atmosphere.

Figure 5. DSC for Illinois No. 6 coal from -60 to +200 °C in a N2 atmosphere.

is very slow and it cannot relax rapidly to reach equilibrium at the lowest energy state. When heated to a temperature such that the thermal energy is greater than the intramolecular interactions, coal undergoes a certain amount of irreversible structural change, which is consistent with a stress relief of the structure. In the case of CO2 adsorption on coal samples, DSC results show that coal/CO2 interactions are much stronger. The exothermic peaks for the adsorption of CO2 on the coal samples are associated with the uptake of CO2 by coals. This is an activated process, and presumably, at the temperature of

Figure 6. DSC for Illinois No. 6 coal from -60 to +200 °C in a CO2 atmosphere.

Figure 7. DSC for Pittsburgh No. 8 coal from -60 to +200 °C in a N2 atmosphere.

Figure 8. DSC for Pittsburgh No. 8 coal from -60 to +200 °C in a CO2 atmosphere.

exotherms, there is enough thermal energy to overcome the activation energy for diffusion. The enthalpy values associated with the peaks are given in Table 2. The data presented in Table 2 show that for all coals the interactions between coal and CO2 during the first scan are much stronger than the interactions during the subsequent two scans. The first interactions are not reversible and contain strong adsorption of CO2 on coal because of the strong interaction of CO2 with active sites on the coal surface. Table 2 also shows that for all coals the heats of adsorption of CO2 in the second and third scans are almost the same, and they are lower than the heat of adsorption in the first scan. These results suggest

Table 2. Characteristics of Exotherms Associated with the Interaction of CO2 with Coals coal

first scan ∆H (J/g)

second scan ∆H (J/g)

third scan ∆H (J/g)

irreversible sorption capacity ∆HIrr (J/g)

North Dakota Wyodak Illinois No. 6 Pittsburgh No. 8

23.46 27.40 20.51 10.23

23.02 20.60 11.58 6.78

23.17 18.72 11.28 6.79

0.44 6.80 8.93 3.45

Interactions of Coal with CO2

Figure 9. DSC for dried Wyodak coal from 30 to 200 °C in a N2 atmosphere.

that coal/CO2 interactions on the second and third scans are reversible and much weaker than those observed on the first scans. The exothermic peaks observed on the second and third thermograms might be attributed to the physical adsorption of CO2 on the coal because of the interaction of CO2 with saturated sites after the first interactions. It is believed that exothermic peaks on the first scans show both the irreversible interactions of CO2 with active sites (strong adsorption) and the physical adsorption of CO2 on coal. When CO2 is adsorbed on coal during the first run, it interacts with active sites on the pore surface tightly and irreversibly so that, in the next run, all active sites are saturated with CO2 and the coal surface has a new and different structure. However, exothermic peaks on the second and third scans show only reversible adsorption of CO2 on coal. The difference between enthalpy values attributed to the exothermic peaks on the first scan and those on the second scan determines the energy values associated with the irreversible storage capacity of coal samples. These energy values are also presented in Table 2. Energy values attributed to the irreversible storage capacities for coal samples vary between 0.44 and 8.93 J/g with no dependence on coal rank. The lowest irreversible sorption energy was found for the North Dakota lignite (73% C), and the highest value was encountered for the Illinois No. 6 coal (78% C). Glass transition temperatures of dried Wyodak coal and also of dried coal after exposure to CO2 at different pressures were determined by DSC at a temperature range between 30 and 200 °C. Figure 9 shows the DSC thermograms for dried Wyodak coal from 30 to 200 °C. The first scan represents an irreversible process, which may be associated with an enthalpy relaxation and structural rearrangement in the coal.11 The second and third scans are reversible and indicate a second-order process. According to the work of Mackinnon and Hall, this transition possesses the characteristics of a glass transition.11 The glass transition process is one of the most important characteristic properties of the coal. Before the transition, macromolecular motions in coal are restricted and the diffusion of gases and liquids in its structure is slow. Above the transition, coal becomes rubbery and diffusion into its structure becomes much faster. The reversible nature of the glass transition process might be due to the breakage of hydrogen bonds at the onset of the glass transition upon heating and the reforming of them upon cooling.10,11 Figures 10 and 11 show the DSC thermograms for Wyodak coal loaded with CO2 at different pressures. The first scans (10) Mackinnon, A. J.; Antxustegi, M. M.; Hall, P. J. Fuel 1994, 73 (1), 113-115.

Energy & Fuels, Vol. 20, No. 5, 2006 2025

Figure 10. DSC for dried Wyodak coal (held in a 5 bar CO2 atmosphere for 24 h) from 30 to 200 °C in a N2 atmosphere.

Figure 11. DSC for dried Wyodak coal (held in a 30 bar CO2 atmosphere for 24 h) from 30 to 200 °C in a N2 atmosphere.

indicate two endothermic effects. The first effect might be associated with the evaporation of moisture, which was adsorbed by the sample during transfer from the high-pressure cell to the DSC chamber, and also continues during the release of sorbed CO2. The second effect is an endothermic peak that might be attributed to the fast release of sorbed CO2 from the coal in the vicinity of the glass transition temperature. As the coal in the DSC chamber is heated from low to high temperatures, it will continuously release CO2. However, at the glass-to-rubber transition temperature, the desorption rate may suddenly be accelerated because the chain mobility of the coal suddenly becomes higher. This process is an irreversible process and has disappeared on the second and third scans. Figure 11 shows that the endothermic peak becomes noticeable when coal is exposed to a 30 bar CO2 atmosphere. This is because the amount of gas absorbed in coal is greater at higher pressures. Consequently, at higher pressures, the relatively larger amount of CO2 would be desorbed in the vicinity of Tg, and a significant heat effect can be detected by DSC. Figure 12 shows the change in the glass transition temperature of the coal with preadsorbed CO2 pressure. Table 3 shows the characteristics of glass transition in Wyodak coal at various pressures of CO2. The depression in Tg for Wyodak in a CO2 atmosphere might be due to the solubility of CO2 into the coal matrix and the plasticization of coal by CO2. The further solubility of gas into the coal with CO2 pressure may be partly due to the solubility parameter of CO2 approaching a value closer to that of the coal.12 (11) Mackinnon, A. J.; Hall, P. J. Energy Fuels 1995, 9, 25-32. (12) Reucroft, P. J.; Patel, H. Fuel 1986, 65, 816-820.

2026 Energy & Fuels, Vol. 20, No. 5, 2006

Mirzaeian and Hall

Figure 12. Dependence of glass transition temperature of Wyodak on the pressure of CO2.

Figure 13. TPD-MS spectrum of CO2 adsorbed on Pittsburgh No. 8 coal at 5 bar.

Table 3. Temperature of Second-Order Phase Transition for Wyodak Coal at Different Pressures of CO2 PCO2 (bar)

second scan Tg (°C)

third scan Tg (°C)

0 1 5 10 20 30

121.1 117.2 86.6 83.2 82.4 81.1

121.6 114.2 81.7 78.2 77.8 75.7

It is supposed that the acidic and basic properties of CO2 which allow it to form hydrogen bonds or other acid-base bonds play a major role in its solubility in coal. CO2 may act as a hard acid in its interactions with coals.13 It dissolves in the organic coal matrix, thus modifying the physical and possibly the chemical structure of the coal matrix by disrupting hydrogen bonds in the coal structure during its interactions with coal. TPD-MS. A TPD-MS spectrum of CO2 adsorbed on Pittsburgh No. 8 coal at 5 bar of pressure is shown in Figure 13. The spectrum has two regions of interest, the first one in the low-temperature region (in the range of 300-360 K) and the second one after the peak temperature where the desorption rates decrease with increasing temperature. To analyze the desorption data, the low-temperature part of the spectrum has been modeled using the assumption of a first-order desorption process with a single activation energy for desorption (Redhead equation).14

ln

( ) ( ) ( ) [ ( )]

Edes 1 Edes 1 N T 2 1 1 ) + exp -1 NP R T TP TP R T TP (Redhead equation)

NP is the maximum desorption rate at peak temperature, TP, and N is the desorption rate at any temperature. The embedded graph in Figure 13 shows that the first-order model provides a good fit to the experimental data in the range of 300-360 K. The value of the activation energy for desorption, Edes, is calculated from the fitting of the low-temperature part of the spectrum to the model. The value of Edes for the desorption of CO2 from Pittsburgh No. 8 coal is estimated as Edes ≈ 11 kcal/mol. This value is much smaller than the ∼30 kcal/mol value for the activation energy for the desorption of pyridine from Illinois No. 6 coal measured by Hall and Larsen.15 Pyridine is an excellent hydrogen-bond acceptor and a strong organic Lewis base;16 it interacts with coal through strong (13) Glass, A. S.; Larson, J. W. Energy Fuels 1994, 8, 629-636. (14) Redhead, P. A. Vacuum 1962, 12, 203. (15) Hall, P. J.; Larsen, J. W. Energy Fuels 1993, 7, 47-51. (16) Arnett, E. M.; Joris, E.; Murty, T. S. S. R.; Gorrie, T. M.; Schleyer, P. v. R. J. Am. Chem. Soc. 1970, 92, 2365-2377.

Figure 14. TPD-MS spectra of CO2 adsorbed on Pittsburgh No. 8 coal at various pressures.

noncovalent interactions and hydrogen bonding with coal hydroxyls.17 Therefore, the estimated value of Edes for CO2 desorption on the coal is sensible. The trail of the spectrum after the peak shows that the desorption rate decreases with increasing temperature and does not follow the first-order desorption model. The process of CO2 desorption from coal is a mass transfer process involving several elementary processes. These processes may include diffusion through macropores, diffusion through micropores, and diffusion through the coal matrix. In small micropores, diffusional resistance limits the mass transfer process. The mass exchange can also be limited because of the large energy barrier for desorption. At higher surface coverages, the diffusivity is high and the rate of desorption reaches a maximum. The reason for this increase in rate is that desorption occurs from the most occupied sites in the microporous structure leading to the release of the low-energetic physisorbed sites. As the temperature increases and the desorption process proceeds, high-energetic sites in the microporous structure, which are occupied by strong interactions with CO2, are released. However, because of the large energy barrier, the rate of desorption from these sites is very slow.18 Therefore, the deviation of the desorption spectrum from the fist-order kinetic model may be due to the contributions of activated diffusion effects, micropore diffusional resistance, barrier resistance of the high-energetic sites, and experimental error in the instrument. TPD-MS spectra of CO2 adsorbed on Pittsburgh No. 8 coal at various pressures are compared in Figure 14. It can be noted that the desorption intensities increase with pressure, indicating that the amount of CO2 sorbed in the coal is greater at higher pressures. Consequently, at higher pressures, the relatively larger amount of CO2 would be desorbed. (17) Larsen, J. W.; Baskar, A. J. Energy Fuels 1987, 1, 230-232. (18) Reid, C. R.; Thomas, K. M. Langmuir 1999, 15, 3206-3218.

Interactions of Coal with CO2

Energy & Fuels, Vol. 20, No. 5, 2006 2027

Table 4. Desorption Parameters for CO2 Adsorbed on Pittsburgh No. 8 Coal at Various Pressures for 24 h P (bar)

Edes (cal/mol)

integrated area

correlation factor

5 15 20

10 445 11 190 11 202

2.137 × 10-9 3.771 × 10-9 7.160 × 10-9

0.908 0.996 0.944

The total area under a TPD-MS spectrum is proportional to the amount of adsorbed CO2.19 The values of Edes and the integrated areas under the spectra of CO2 at various pressures are shown in Table 4. Conclusions It was found that the first interactions of CO2 with coal lead to strongly bound carbon dioxide on coal, whereas the subsequent interactions are reversible and have characteristics of weak bonds (physisorbed). The reduction in the value of the exotherm between the first and second runs suggests that some CO2 is irreversibly bound to the structure even after heating to 200 (19) Habenschaden, E.; Kuppers, J. Surf. Sci. 1984, 138, L147.

°C. The results show that Illinois No. 6 and Wyodak coals have great affinity and a high irreversible sorption capacity for CO2 among the other coals investigated in this study. The study of the effect of high-pressure CO2 on the macromolecular structure of coal shows that the glass transition temperature of coal decreases with CO2 pressure significantly, indicating that high-pressure CO2 diffuses through the coal matrix, causes significant plasticization effects, and changes the macromolecular structure of coal, as predicted by White et al.9 Desorption characteristics of coal loaded with high-pressure CO2 show that CO2 desorption from coals is an activated process and follows a first-order kinetic model. The increase in the activation energy for CO2 desorption from coal with preadsorbed CO2 pressure suggests that high-pressure interactions will demand more energy to desorb from coal probably because of the access of CO2 to more micropores. Acknowledgment. We are grateful to the British Coal Utilisation Research Association (BCURA) for support of this work. EF060040+