72
Energy & Fuels 1987,1, 72-75
(61 MHz). The 2H NMR spectra were recorded in methylene chloride (Aldrich, Gold Label), and the solvent signal (5.32 ppm) due t o 2H in natural abundance was used t o calibrate the field. The infrared spectra of the coals (2-4%) in dry potassium bromide were recorded with a Nicolet 20 spectrometer. The pellets
were dried in vacuo at 65 "C for 20 h before the spectra were recorded.
Acknowledgment. We are pleased to acknowledge the support of this work by the Gas Research Institute.
Effect of Pressure on Carbon Dioxide Induced Coal Swelling P. J. Reucroft* and A. R. Sethuraman Department of Metallurgical Engineering and Materials Science, University of Kentucky, Lexington, Kentucky 40506 Received September 9, 1986. Revised Manuscript Received October 22, 1986
Dilatometric studies have been carried out on coal samples exposed to carbon dioxide at 5 , 1 0 , and 15 atm. The sample dimensions increase on exposure to carbon dioxide, the effect increasing as pressure increases. The time to reach equilibrium swelling is shorter at the higher pressure. The swelling effect increases as the coal carbon content decreases. It is estimated that the swelling effect can account for 20-50% of the surface area determined by C02 adsorption methods.
Introduction BET surface areas determined by COPadsorption are generally much higher than N2surface area values.14 This has usually been attributed to the micropore system in coal samples not being completely accessible to N2 molecules at 77 K because of an activated diffusion process and/or shrinkage of pore^.^,^ Adsorption studies on coal employing organic vapor adsorbates yielded BET surface areas ranging from 9 to 200 m2/g depending upon the adsorbate vapor used.' It was also observed that the maximum surface area was obtained when the solubility parameter (6) of the adsorbate vapor was similar in value ~m-'.~). to the solubility parameter of the coal (-10 The results suggested that swelling may play an important role in determining coal surface area values obtained by gas adsorption methods. Maximum swelling (and surface area) occurs when the solubility parameter of the adsorbate is close to that of the coal. The studies further indicated that COz adsorption results in higher surface area values compared to N2 adsorption because the solubility parameter of C 0 2 (6 = 6 . 1 caP5 ~ m - ' . ~is) closer to the solubility parameter of many coal samples than that of N2 (6 = 2.6 cal ~ m - l . ~ The ) . ~ macromolecular nature of coal and the effect of solvent vapors and liquids in determining coal swelling characteristics has been well established in recent studies.+12 (1) Mahajan, 0. P. Powder Technol. 1984, 40, 1. (2) Mahajan, 0. P. In Coal Structure; Meyers, R. A., Ed.; Academic: New York, 1982; p 51. (3) Mahajan, 0. P.; Walker, P. L. Jr. Anal. Methods Coal Coal Prod. 1978, I, 125. (4) Mahajan, 0.P. In Sample Selection, Aging and Reactivity of Coal; Klein, R., Wellek, R., Eds.; Wiley: New York, in press. (5) Anderson, R. B.; Bayer, J.; Hofer, L. J. E. Fuel 1965, 44, 443. (6) Walker, P. L., Jr.; Geller, I.; Nature (London) 1956, 178, 1001. (7) Reucroft, P. J.; Patel, K. B. Fuel 1983, 62, 279. (8) Barton, A. F. M. Handbook of Solubility Parameters and Other Cohesion Parameters; CRC Press: Boca Raton, FL, 1983.
0887-0624/87/2501-0072$01.50/0
Table I. Representative Analyses of the Three Types of Coals KCER KCER 7463 7259 KCER 7122 (sample sample (sample 1) (sample 2) 3) county Webster Hopkins Carlisle seam thickness 57.0 in. 6.0 ft 52.0 in. coal rank bituminous subbituminous lignite anal. as received (wt % ) volatile matter 40.7 34.9 33.0 fixed carbon 49.5 38.5 13.7 7.0 21.6 19.2 ash moisture 1.9 5.3 34.2 total sulfur 2.9 4.6 2.3 ultimate anal. (wt % daf)" 83.8 18.3 65.8 carbon hydrogen 5.7 5.5 7.8 nitrogen 1.0 1.0 0.1 sulfur 3.2 6.3 4.9 oxygen (by difference) 6.3 8.9 21.4
"daf = dry, ash free.
Direct confirmation of the swellling effect of CO, on coal has also been obtained by dilatometric studies.13 Volume increases of up to 1.3% were obtained on exposure to COP at 1 atm. negligible effects were obtained on exposure to N2, He, and Xe. It was estimated that swelling of this magnitude could account for up to 1 4 . 5 % of the reported C 0 2 BET surface areas. (9) Gorbaty, M. L.; Mraw, S. C.; Gethner, J. S.; Brenner, D. Fuel Process. Technol. 1986,12, 31. (10) Lucht, L. M.; Peppas, N. A. AIP Conf. Proc. 1980, No. 70, 28. (11) Larsen, J. W.; Kovac, J. In Organic Chemistry of Coal; Larsen, J. W., Ed.; ACS Symposium Series 71; American Chemical Society: Washington, DC, 1978. (12) Brenner, D. Fuel 1985, 64, 167. (13) Reucroft, P. J.; Patel, H. Fuel 1986, 65, 816.
0 1987 American Chemical Society
Energy & Fuels, Vol. 1, No. 1, 1987 73
Effect of Pressure o n Coal Swelling
Table 11. Comparison of Swelling Results of Coals after 200-h Exposure to CO, a t 5 atm dimension change, qm sample % C specimen 1 specimen 2 1 83.8 29.8 27.5 2 78.3 45.5 43.4 3 65.8 86.4 83.2
-50
b
L
Ibo
1150
& I
i50
j,
i50
O IJ
Figure 1. Swelling response of KCER 7122 (% C = 78.3) a t 5 atm of pressure of COz (specimen 2).
-1
-
Table 111. Comparison of Swelling Results of Coals after 200-h Exposure to CO, at 10 a t m dimension change, wm sample $0 c specimen 1 specimen 2 1 83.8 33.9 32.7 2 78.3 75.4 74.3 3 65.8 119.8 117.1 Table IV. Comparison of Swelling Results of Coals after 200-h Exposure to CO, at 15 atm dimension change, pm sample 9'0 C specimen 1 specimen 2 1 83.8 53.0 52.8 2 78.3 114.0 114.8 3 65.8 167.1 166.3 phases. Exposure times up t o 200 h were employed. After exposures to COz,the samples were again evacuated in order to check for reversibility of the swelling effect. Two specimens of each coal sample were tested in order t o ensure reproducibility. The dilatometric assembly employed has been described previously.ls The dilatometer was calibrated with a mechanical micrometer to read directly in terms of micrometers (lo4 m).
-IS -30
I0
50
Ibo
i50
Exposure Time (Hours)
-bo
2b
;m
is0
1
Figure 2. Swelling response of KCER 7122 (% C = 78.3) a t 10 atm of pressure of COz (specimen 2).
Recent studies on density changes produced on exposing coal with 69.1% carbon content to 1 atm of C02 have concluded that no measurable swelling effect occurs within the precision of the measurement ( ~ l % ) .A: close , ~ examination of the data does in fact show a small density change, which corresponds to a volume change of 0.9%. The results are thus consistent with the more precise dilatometric measurements. The dilatometric results at 1atm indicated that equilibrium swelling had not been reached at these COP pressures, even at exposure times in excess of 240 h. In the present study the dilatometric studies have been extended to higher C02pressures. The objective has been to evaluate the effect of gas pressure on the swelling behavior of typical coals and obtain a better estimate of the equilibrium swelling parameters.
Experimental Section Dilatometric studies were carried out on three Kentucky coals in the bituminous, sub-bituminous and lignite range employing the methods described previ0us1y.l~ T h e coal samples are described in Table I. Pencil-shaped samples of coal, l cm in length and 0.4 cm in diameter, were initially evacuated for 100 h a t 298 K and exposed t o COB a t 5, 10, and 15 a t m of pressure. The evacuation procedure removes weakly bonded or physically sorbed water. Evacuation a t higher temperatures was avoided in order t o preserve the integrity of the coal samples and avoid loss of volatile organic matter. T h e change in dimension was recorded as a function of time in both the evacuation and the gas exposure (14) Stacy, W. 0.;Jones, J. C. Fuel 1986,65, 1171.
Results and Discussion Typical response data are shown in Figures 1-3 for the KCER 7122 sample (sample 2). Initial evacuation produced shrfnkage of 30-40 pm after about 100 h (stage I). Exposure to COz then produced a length increase, which reached a steady-state value after approximately 200 h exposure at 5 and 10 atm of pressure (stage 11). The steady state was reached at much shorter times at 15 atm of pressure. Final evacuation (stage 111) approximately returned the sample length to the level attained in stage I. Some samples did not return to the initial level attained in stage I as readily as other samples, however (see Figure 2). The reasons for this behavior are not clear but may be associated with individual sample variations. Two features are noteworthy. The dimension change attained after 200-h exposure to C02 increased as the pressure increased from 5 to 15 atm. The time to reach equilibrium response became shorter as the pressure increased. Similar response curves were obtained for the KCER 7259 and 7463 samples (samples 1 and 3). The effect of pressure on the COPinduced response is shown in Figures 4 and 5 for these two samples. The solubility parameter will generally increase with increasing pressure due to a decrease in molar volume? The increased swelling effect with increased pressure may thus be due, in part, to the solubility parameter of carbon dioxide approaching a value closer to that of the coal samples. The solubility parameter of carbon dioxide governs just a minor contribution to its solubility, however. A major contribution results from its acidic and basic properties, which allow it to form hydrogen bonds (or other acid-base bonds) to coal. The swelling effect produced by C02generally increased as the carbon content decreased in agreement with the previous results at 1atm.13 The swelling response to C 0 2 of the three coals (15) Sagues, A. A. Reu. Sci. Instrum. 1979,50,48.
Energy & Fuels, Vol. 1, No. 1, 1987
Reucroft and Sethuraman
II
9-
5m
5
-I
-20
KCER 7259
-30
-40 -50
I0
40
I'OO
I'50
Z'OO
is0
do0
Exposure Time (Hours)
h
4
Figure 3. Swelling response of KCER 7122 (% C = 78.3) a t 15 atm of pressure of COz (specimen 2).
.
60
== 5m
Table V. Comparison of Initial Evacuation Shrinkage after 100 h with Coal Moisture Content sample moisture, % dimension change: fim 1 1.9 -16.9 -37.2 2 5.3 -115.5 3 34.2
15 n u
.d
" Average of six specimens.
v
30
_I 01
c
20
m 01
6
IO
6 0 40
Exposure Time (Hours)
I'm
I
20
1%
Figure 4. Effect of COz pressure on the swelling response of KCER 7259 samples (% C = 83.8).
I
-
Exposure Time (Hours)
Figure 6. Comparison of the swelling response of three coal samples on exposure t o 5 atm of COz.
At
ln
0
10 ntm
c
* 5 arm
1
C
.-
O $0
Exposure Time (Hours)
IbO
1150
20
Table VI. Estimated Contribution of Swelling to COz Surface Area ( P = 5 atm) C 0 2 surface swelling vol, % sample % C area, m2/g-' estb measd" 1 83.8 150 5.4 0.75 (13.8) 1.24 (22.9) 2 78.3 150 5.4 3 65.8 250 9.0 2.16 (24.0) a Expressed as a percentage of estimated swelling in parentheses. Estimated by assuming that all the adsorbed molecules that are contained in the "monolayer" are contributing volume to a swollen adsorbent-adsorbate system.
Table VII. Estimated Contribution of Swelling to COz Surface Area ( P = 10 atm) swelling vol, % COz surface sample %C area, m2/g-' estb measd" 1 83.8 150 5.4 0.85 (15.6) 2 78.3 150 5.4 2.23 (41.3) 3 65.8 250 9.0 3.00 (33.2) Expressed as a percentage of estimated swelling in parentheses. Estimated by assuming that all the adsorbed molecules that are contained in the "monolayer" are contributing volume to a swollen adsorbent-adsorbate system.
Table VIII. Estimated Contribution of Swelling to COz Surface Area ( P = 15 atm)
Figure 5. Effect of COz pressure on the swelling response of KCER 7463 samples (% C = 65.8).
sample
%C
investigated is shown at t w o pressures i n Figures 6 and 7. D i m e n s i o n changes produced a f t e r 200-h exposure to COz at 5,10, and 15 atm are summarized in Tables 11-IV. A g r e e m e n t b e t w e e n t w o specimens of the same coal w a s generally good. Data on the s h r i n k a g e effect produced after 100-h evacuation are s u m m a r i z e d in Table V. The s h r i n k a g e on initial evacuation increases w i t h increasing m o i s t u r e content of the coal samples, indicating that the shrinkage effect is d u e to loss of m o i s t u r e on evacuation. The d i r e c t d i l a t o m e t r i c determination of percentage l e n g t h increase c a n be u s e d to determine the percentage
1 2
3
83.8 78.3 65.8
C 0 2 surface area, mZ/g-' 150 150 250
swelling vol, %
estb 5.4 5.4 9.0
measda 1.33 (24.5) 3.11 (57.6) 4.18 (46.5)
"Expressed as a percentage of estimated swelling in parentheses. Estimated by assuming that all the adsorbed molecules that are contained in the 'monolayer" are contributing volume to swollen adsorbent-adsorbate system. volume increase if isotropic behavior is assumed. Volume increases d e t e r m i n e d in t h i s way are listed as m e a s u r e d swelling volume percent i n Tables VI-VIII. P e r c e n t a g e volume increases range from 0.75 to 2.16% at 5 a t m , f r o m 0.85 to 3.0% at 10 a t m , and from 1.33 to 4.18% at 15 atm.
Energy &Fuels, Vol. 1, No. 1, 1987 75
Effect of Pressure on Coal Swelling
-
120-
C
.u
90
B
50
I
KCER 7122
0
‘6
$0
Exposure Time (Hours)
--l I
KCER 7259
ioo
pressures are thus more uncertain. There is also some uncertainty because the surface areas reported in the literature may not exactly represent the coal samples employed in the present study. In addition, because there is an inherent pore structure in coal, pore filling undoubtedly accounts for a large fraction of the adsorption and not all adsorbate contributes to swelling. The results indicate, however, that in low carbon content coals, the swelling of coal by COz may account for a significant fraction of the measured BET surface area. The higher surface areas measured by carbon dioxide adsorption can also result from the carbon dioxide dissolving in the coal, during the swelling process, and reaching inner pores that are inaccessible to nitrogen. By this process, the carbon dioxide has solution pathways through the coal to the inner pores that nitrogen cannot reach.
I$
2ob
Figure 7. Comparison of the swelling response of three coal samples on exposure t o 10 atm of C02.
The results indicate that equilibrium swelling has been attained at the highest pressures employed (15 atm). Significant swelling is thus produced when coal samples are equilibrated with COPin the 0.1-0.2 relative pressure range. It is of interest to estimate the effect that this measured swelling (volume increase) may have on typical surface area values that have been reported for coals of these carbon contents. To facilitate this assessment it can be assumed initially that all the adsorbed molecules that are contained in the “monolayer” are contributing volume to a swollen adsorbent-adsorbate system. Surface area values reported in the literature for coals of the same carbon content were e m p l ~ y e d .The ~ estimated swelling volume increases obtained in this way are compared with the measured volume increases in Tables VI-VIII. The results in Table VI indicate that the measured swelling may account for 13.8-24% of the reported surface area values. At higher pressures the swelling effect may account for higher fractions of the reported surface area values (Tables VI1 and VIII). It should be noted, however, that adsorption experiments to determine surface area are usually carried out at low pressures with an “equilibrium” time of approximately 30 min whereas the swelling volumes were measured after more prolonged contact times. The estimated contributions of swelling to surface area a t high
Summary and Conclusions Significant swelling or volume increases ranging from 0.75 to 4.18% were observed in a range of coal samples when they were exposed to carbon dioxide at pressures up to 15 atm. Increase in pressure produced an increase in swelling response and a decrease in the time required to reach maximum response. A lower carbon content correlates with a higher degree of swelling. The order of swelling was sample 1 (% C = 83.8) < 2 ( % C = 78.3) C 3 (70C = 65.8). The shrinkage due to initial evacuation showed a dependence on the moisture content of the coals. The order of shrinkage was sample 1 < 2 < 3, correlating with moisture contents of 1.9%, 5.3%,and 34.2%,respectively. COz-inducedswelling may account for up to 50% of the reported COz surface area values in lignite and subbituminous coals. In bituminous coals, swelling may account for about 20% of the surface area values. Thus, previously reported surface area values of these coals determined by C02 adsorption may be overestimated by 20-50%, depending on the coal.
Acknowledgment. The research was supported by the Institute for Mining and Minerals Research, University of Kentucky, through the U.S.Bureau of Mines Title I11 program. Coal analysis results were provided by the Kentucky Center for Energy Research Laboratory. Registry No. C02, 124-38-9.
