Inter-laboratory comparison II: CO2 isotherms measured on moisture-equilibrated Argonne premium coals at 55 °C and up to 15 MPa

Available online at www.sciencedirect.com

International Journal of Coal Geology 72 (2007) 153 – 164 www.elsevier.com/locate/ijcoalgeo

Inter-laboratory comparison II: CO2 isotherms measured on moisture-equilibrated Argonne premium coals at 55 °C and up to 15 MPa A.L. Goodman a,⁎, A. Busch c , R.M. Bustin f , L. Chikatamarla f , S. Day b , G.J. Duffy b , J.E. Fitzgerald d , K.A.M. Gasem d , Y. Gensterblum c , C. Hartman e , C. Jing d , B.M. Krooss c , S. Mohammed d , T. Pratt e , R.L. Robinson Jr. d , V. Romanov a , R. Sakurovs b , K. Schroeder a , C.M. White a a

c

U.S. Department of Energy, National Energy Technology Laboratory, Pittsburgh, PA, United States b CSIRO Energy Technology, Newcastle, NSW, Australia Institute of Geology and Geochemistry of Petroleum and Coal, RWTH Aachen University, Aachen, Germany d School of Chemical Engineering, Oklahoma State University, Stillwater, OK, United States e TICORA Geosciences, Inc.19000 West Hwy. 72, Suite 100, Arvada, CO 80007, United States f University of British Columbia, Earth and Ocean Sciences, Vancouver, Canada BC V6T 1Z4 Received 5 September 2006; received in revised form 22 January 2007; accepted 26 January 2007 Available online 1 February 2007

Abstract Sorption isotherms, which describe the coal's gas storage capacity, are important for estimating the carbon sequestration potential of coal seams. This study investigated the inter-laboratory reproducibility of carbon dioxide isotherm measurements on moisture-equilibrated Argonne premium coal samples (Pocahontas No. 3, Illinois No. 6, and Beulah Zap). Six independent laboratories provided isotherm data on the three moisture-equilibrated coal samples at 55 °C and pressures up to 15 MPa. Agreement among the laboratories was good up to 8 MPa. At the higher pressures, the data among the laboratories diverged significantly for two of the laboratories and coincided reasonably well for four of the laboratories. © 2007 Elsevier B.V. All rights reserved. Keywords: CO2 sorption; CO2 storage capacity; CO2 isotherm; Argonne premium coal; Carbon sequestration

1. Introduction Carbon dioxide storage in coal seams has recently received increased attention as a potential option to

⁎ Corresponding author. Tel.: +1 412 386 4962; fax: +1 412 386 5920. E-mail address: [email protected] (A.L. Goodman). 0166-5162/$ - see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.coal.2007.01.005

reduce greenhouse gas emissions to the atmosphere. Accurate measurements of CH4 and CO2 sorption isotherms are vital for the optimum development of techniques to either sequester CO2 or to combine CO2 storage with an enhancement of CH4 recovery (Clarkson, 1999; Gasem et al., 2001). For a given coal seam, sorption isotherm measurements provide information about the storage capacity, the overall economics of the process, and the types of operating

154

A.L. Goodman et al. / International Journal of Coal Geology 72 (2007) 153–164

conditions that can be used (White et al., 2005). The storage capacity of a coal seam is traditionally estimated from manometric, volumetric, or gravimetric isotherm measurements (Ruppel et al., 1972; Lu et al., 1995; Humayun and Tomasko, 2000). Since there is no standard procedure or technique for measuring the isotherm, results obtained from different laboratories need to be compared to determine their accuracy. In order to better understand the variation in estimates of the storage capacity of coal seams, the U.S. Department of Energy-National Energy Technology Laboratory (DOE-NETL) initiated an inter-laboratory study where the sorption isotherms measured by laboratories throughout the world were compared. In that first study, the CO2 sorption isotherm measurements on dry coals at 22 °C were compared (Goodman et al., 2004). The isotherm data varied widely among the laboratories, with the greatest variation being observed for the lower rank coal samples. The results of that study suggested that the drying procedure for removing moisture from the coal samples played a large role and that the isotherms reflected different degrees of residual moisture content. The development of a strict procedure for drying coals in order to obtain isotherm measurements on dry coals was recommended. The DOE-NETL then initiated a second inter-laboratory isotherm comparison of coals where CO2 sorption isotherms were collected on moisture-equilibrated coals at temperatures and pressures relevant to CO2 sequestration. Results of this second comparison among laboratories for moisture-equilibrated coals are reported here. Each laboratory used the same coal samples and followed the same general procedure; however, each laboratory used their own apparatus and isotherm measurement technique. Currently, there is no standard CO2-coal isotherm method as assumed by ASTM procedure E-691 (ASTM E 691– 99, 2000). Thus, these results should be considered a discovery process which was undertaken to determine if stricter control of variables is needed to obtain good inter-laboratory precision for CO2-coal isotherm measurements. The data are examined for possible sources of error, but only as a guide for future investigations, not to provide unequivocal answers to specific questions. Our intent is to address the issue of whether isotherm measurements reported by laboratories worldwide are comparable. This work provides guidance for estimating the reproducibility that might be expected when comparing published sorption isotherms on moisture-equilibrated coals from different laboratories.

2. Experimental Approximately a dozen laboratories that are measuring sorption data worldwide were invited to participate in this study. Because of their funding or time constraints, the following six independent laboratories volunteered to participate at their own expense, excluding the cost of the coal samples: CSIRO, Australia; RWTH Aachen University, Germany; Oklahoma State University, USA; TICORA Geosciences, USA; University of British Columbia, Canada; and DOE-NETL, USA. Table 1 lists the participants and their affiliations. Located throughout the world, the participants included academic, government, and industrial laboratories. The data each laboratory contributed are reported anonymously, being referred to as laboratory 1, 2, 3, 4, 5, or 6 throughout the paper, an order that is different from the listing in Table 1. For this study, DOE-NETL chose a set of three coal samples from the Argonne Premium Coal Sample Program: Pocahontas No. 3 (low volatile bituminous), Illinois No. 6 (high volatile bituminous), and Beulah Zap (lignite) (Vorres, 1990). The Argonne premium coal samples provide the research community with the highest quality samples for basic research. Samples for a specific coal are as chemically and physically identical as possible. The coals are well-characterized and are stable over long periods of time because they were prepared and stored under inert gas. This ensures that all participants received identical, homogeneous coal samples and any variations in the measured results were unlikely to be due to sample variability. The

Table 1 Inter-laboratory comparison II participants Participants G. Duffy, R. Sakurovs, S. Day

Affiliation

CSIRO Energy Technology, Newcastle, NSW, Australia B.M. Krooss, A. Busch, Institute of Geology and Y. Gensterblum Geochemistry of Petroleum and Coal, RWTH Aachen University, Germany K.A.M. Gasem, R. L. Robinson, School of Chemical Engineering, Jr., J.E. Fitzgerald, C. Jing, Oklahoma State University, Stillwater, OK S. Mohammed T. Pratt, C. Hartman TICORA Geosciences, Inc.19000 West Hwy. 72, Suite 100, Arvada, Colorado 80007 M. Bustin, L. Chikatamarla University of British Columbia, Earth and Ocean Sciences, Vancouver, BC V6T 1Z4 V. Romanov, K. Schroeder, U.S. DOE-NETL, USA C. White

A.L. Goodman et al. / International Journal of Coal Geology 72 (2007) 153–164

155

Table 2 Proximate and ultimate analyses of the Argonne premium coal samples Coal sample

Ultimate analysis (wt.%, daf) a

Proximate analysis (wt.%)

Seam

State

Rank

Moisture

Ash b

VM b

C

H

O

S

N

Pocahontas No. 3 Illinois No. 6 Beulah Zap

VA IL ND

Low vol. bit. High vol. bit. Lignite

0.65 7.97 32.24

4.74 14.25 6.59

18.48 36.86 30.45

91.05 77.67 72.94

4.44 5.00 4.83

2.47 13.51 20.34

0.50 2.38 0.70

1.33 1.37 1.15

a b

daf = dry ash free. As-received basis.

samples were supplied as powders with particles less than or equal to 500 μm (− 100 mesh) (Vorres, 1993). Proximate and ultimate analyses for the Argonne premium coal samples are shown in Table 2. Laboratories 1–5 prepared moisture-equilibrated coal samples according to a modified version of the ASTM procedure D 1412-99 “Standard Test Method for Equilibrium Moisture of Coal at 96 to 97 Percent Relative Humidity” (ASTM D 1412-99, 2000). Laboratory 6 followed the modified ASTM procedure D 1412-99 for Illinois No. 6 sample and used the asreceived Beulah Zap and Pocahontas No. 3 samples without modification from the Argonne coal bank. The minor modifications to the D 1412-99 procedure are discussed below. All laboratories handled coal samples

in an inert atmosphere instead of air to prevent surface oxidation. The coal samples were equilibrated at 55 °C instead of 30 °C. The Pocahontas No. 3 and Illinois No. 6 samples were equilibrated for 48 h and the Beulah Zap sample was equilibrated for 72 h. A subsample was taken for moisture determination before the CO2 isotherm measurement. The CO2-coal isotherm was collected at 55 °C for the three Argonne coal samples at pressures up to 15 MPa. Results are reported on a daf (dry ash free) basis. All laboratories measured CO2 sorption isotherms on moisture-equilibrated coals using manometric (five laboratories), volumetric (one laboratory), or gravimetric (one laboratory) equipment. The manometric and volumetric systems were described previously in the

Fig. 1. Gravimetric apparatus used by one laboratory to measure CO2-coal sorption isotherms.

156

A.L. Goodman et al. / International Journal of Coal Geology 72 (2007) 153–164

Table 3 Experimental parameters for each laboratory

Maximum CO2 pressure (MPa) Temperature (°C) Coal mass (g) Sample cell volume (cm3) Reference cell volume (cm3) Average void volume (cm3) Equilibration time CO2 purity a

Laboratory 1 a

Laboratory 2 a

Laboratory 3 a

Laboratory 4 a

Laboratory 5 a

Laboratory 6 a

16 ± 0.01 55 ± 0.1 110–150 310 ± 0.1 148.8 ± 0.1 100–200 ± 0.1 N2 h 99.99%

15 ± 0.005 55 ± 0.1 2.1–2.7 7.7 ± 0.08 11.5 ± 0.08 5.9 ± 0.1 30 min 99.999%

15 ± 0.005 55 ± 0.1 111.741 196 ± 2 116.4 ± 0.1 113.5 6h

15 ± 0.001 55 ± 0.1 79.17 163.905 170.238 96.65 4–12 h 99.99%

15 ± 0.008 55 ± 0.1 3.73 9.23 ± 0.006 1.67 ± 0.01 6.44 60 min 99.995%

15 ± 0.007 55 ± 0.1 60.8–74.6 110 ± 0.3 250 ± 0.01 75 6–12 h 99.99%

Isotherms were measured using gravimetric (1 laboratory), manometric (5 laboratories), and volumetric (1 laboratory) equipment.

first inter-laboratory study (Goodman et al., 2004). The gravimetric system is new to this study and is shown schematically in Fig. 1. In this system, the mass change in the sample during gas sorption is monitored. Sorption isotherms for three samples can be measured simultaneously. Measurements of the mass gain of an empty chamber at an identical pressure provide a direct measure of the density of the non-adsorbed gas without recourse to equations of state calculations. Because each laboratory used different equipment to measure the isotherms, the following laboratory parameters varied: sample weight, sample cell volume, reference cell volume, void volume, equilibration time, and data collection method (Table 3). All laboratories reported temperature stability within a tenth of a degree. The amount of coal used by the laboratories ranged from 2

Fig. 2. Pocahontas No. 3 coal — y1-axis: percent moisture-equilibrated coal moisture at 55 °C measured by laboratories 1–5 (hashed bars) and percent as-received coal moisture at 30 °C obtained from the Argonne premium coal bank (solid line). Laboratory 6 used the Pocahontas No. 3 coal directly from the Argonne coal bank. y2-axis: absolute sorption (mmol/g, daf) estimated from the Langmuir equation at 8 MPa (solid circles).

to 111 g. Sample and reference volumes of the manometric, volumetric, and gravimetric equipment ranged from 6 to 310 cm 3. The amount of time for the CO2 to equilibrate with the coal ranged from 30 min to 12 h. The laboratories then measured and provided CO2coal isotherm data in terms of Gibbs excess sorption as a function of increasing pressure to DOE-NETL as described in the first study (Goodman et al., 2004). To ensure that the data from all of the laboratories were on the same basis, all the Gibbs excess sorption isotherms were calculated using compression factors from the Span and Wagner (1996) equation of state, except for the gravimetric system where the density was measured directly. The data were corrected for moisture and ash by using the moisture values measured by each laboratory after following ASTM procedure D1412-99 (Figs. 2–4) and the ash values found on the Argonne premium coal

Fig. 3. Illinois No. 6 coal — y1-axis: percent moisture-equilibrated coal moisture at 55 °C measured by laboratories 1–6 (hashed bars) and percent as-received coal moisture at 30 °C obtained from the Argonne premium coal bank (solid line). y2-axis: absolute sorption (mmol/g, daf) estimated from the Langmuir equation at 8 MPa (solid circles).

A.L. Goodman et al. / International Journal of Coal Geology 72 (2007) 153–164

157

decreases as the adsorbed phase volume increases. By accounting for the changing void volume, the absolute sorption is calculated by the following relationship ! q a nABS ¼ nGIBSS ð1Þ qa −qg

Fig. 4. Beulah Zap coal — y1-axis: percent moisture-equilibrated coal moisture at 55 °C measured by laboratories 1–5 (hashed bars) and percent as-received coal moisture at 30 °C obtained from the Argonne premium coal bank (solid line). Laboratory 6 used the Beulah Zap coal directly from the Argonne coal bank. y2-axis: absolute sorption (mmol/g, daf) estimated from the Langmuir equation at 8 MPa (solid circles).

web site (Table 2) (Vorres, 1993). No explicit correction was made for carbon dioxide solubility in water. The assessment of the Gibbs excess sorption assumes that the void volume, which is the sum of the volumes from the gas and adsorbed phases, remains constant. As higher pressures are reached, the gas phase volume

where nABS is the absolute sorption, nGIBBS is the Gibbs sorption, ρg is the density of the gas phase, and ρa is the density of the adsorbed phase (Sudibandriyo et al., 2003). The value of ρg is calculated from an equation of state, except for the gravimetric system where it is measured directly. In this study, we used the equation of state by Span and Wagner (1996) for the manometric and volumetric systems. The value of ρa is difficult to determine experimentally and is usually assumed to be constant over the entire pressure range. In this study, we used the adsorbed phase density approximation suggested by Arri and Yee (1992) of 1.18 g/cm3 . Different authors use various adsorbed phase density values when calculating sorption isotherms (Arri and Yee, 1992; Adams et al., 2000; Sudibandriyo et al., 2003). Fitzgerald et al. (2005) showed that the calculated absolute sorption can vary by as much as 15% near 14 MPa depending on the choice of the adsorbed phase density. In addition, the calculations presented here for both Gibbs and absolute sorption are based on the assumption of constant

Table 4 Gibbs excess CO2 sorption (daf) on Pocahontas No. 3 coal at 55 °C Laboratory 1

Laboratory 2

Laboratory 3

Laboratory 4

Laboratory 5

Laboratory 6

Pressure (MPa)

Gibbs excess sorption (mmol/g)

Pressure (MPa)

Gibbs excess sorption (mmol/g)

Pressure (MPa)

Gibbs excess sorption (mmol/g)

Pressure (MPa)

Gibbs excess sorption (mmol/g)

Pressure (MPa)

Gibbs excess sorption (mmol/g)

Pressure (MPa)

Gibbs excess sorption (mmol/g)

0.26 0.53 1.16 2.18 4.14 6.41 8.65 10.34 11.89 12.87 14.13 15.11

0.239 0.423 0.633 0.810 0.952 1.008 0.971 0.825 0.685 0.644 0.455 0.437

1.6410 3.3440 5.3641 7.5084 9.9422 12.087 14.872

0.76872 0.89872 1.0006 0.93082 4.8443 4.7274 3.1595

0.08 0.18 0.38 0.76 1.55 3.28 5.14 7.00 8.33 9.73 11.08 12.49 13.91

0.097 0.195 0.346 0.535 0.759 1.015 1.1567 1.227 1.234 1.207 1.012 0.868 0.750

0.14 0.22 0.34 0.72 1.91 3.77 5.69 7.17 8.21 10.15 10.94 11.80 12.41 12.54

0.033 0.128 0.248 0.677 1.405 1.681 1.798 1.377 1.347 2.300 5.131 9.068 24.452 37.112

0.20 0.46 0.76 1.08 2.01 3.07 4.05 5.07 6.01 7.06 8.03 9.06 10.02 11.00 12.05 13.10 13.85

0.229 0.393 0.517 0.607 0.785 0.903 0.970 1.017 1.053 1.056 1.048 1.034 1.005 0.955 0.886 0.838 0.818

0.40 0.77 1.49 2.84 4.25 5.63 6.99 8.33 9.69 10.34 12.16 13.11

0.282 0.440 0.607 0.767 0.859 0.906 0.921 0.914 0.875 0.847 0.738 0.682

158

A.L. Goodman et al. / International Journal of Coal Geology 72 (2007) 153–164

Table 5 Gibbs excess CO2 sorption (daf) on Illinois No. 6 coal at 55 °C Laboratory 1

Laboratory 2

Laboratory 3

Laboratory 4

Laboratory 5

Laboratory 6

Pressure (MPa)

Gibbs excess sorption (mmol/g)

Pressure (MPa)

Gibbs excess sorption (mmol/g)

Pressure (MPa)

Gibbs excess sorption (mmol/g)

Pressure (MPa)

Gibbs excess sorption (mmol/g)

Pressure (MPa)

Gibbs excess sorption (mmol/g)

Pressure (MPa)

Gibbs excess sorption (mmol/g)

0.10 0.21 0.56 1.22 3.30 5.16 7.39 9.30 10.91 12.93 14.07 15.59

0.037 0.093 0.196 0.339 0.609 0.753 0.849 0.889 0.963 0.953 0.936 0.909

0.40 1.57 3.20 5.21 7.02 8.86 10.12 11.85 13.72 15.78

0.631 0.714 1.081 1.365 1.615 2.952 5.251 4.506 3.022 2.030

0.13 0.25 0.46 0.83 1.66 3.36 5.26 7.10 8.43 9.79 11.08 12.45 13.93

0.050 0.094 0.154 0.245 0.394 0.606 0.761 0.846 0.872 0.863 0.800 0.733 0.662

0.16 0.31 0.57 1.14 2.42 4.06 5.68 7.82 9.31 10.70 11.55 12.48

0.019 0.067 0.153 0.307 0.535 0.692 0.843 0.453 2.356 21.062 37.512 32.723

0.98 1.96 2.80 3.55 4.22 4.81 5.35 5.83 6.25 7.31 8.11 9.16 10.08 11.14 12.12 13.04 14.01

0.276 0.436 0.543 0.621 0.678 0.725 0.759 0.783 0.805 0.845 0.867 0.897 0.921 0.911 0.843 0.824 0.761

0.42 0.80 1.56 2.23 2.87 4.27 5.62 7.02 8.34 9.69 11.04 12.41 13.88

0.154 0.247 0.385 0.480 0.560 0.701 0.782 0.856 0.890 0.907 0.889 0.831 0.762

volume available to the fluid in place (gas plus adsorbed phase) and do not account for coal shrinking and swelling (Reucroft and Patel, 1986; Reucroft and Sethuraman, 1987; Walker et al., 1988; Shimizu et al., 1998; Takanohashi et al., 2000; St. George and

Barakat, 2001; Ozdemir et al., 2003a,b; Ozdemir et al., 2004; White et al., 2005) or structural changes (Hsieh and Duda, 1987; Larsen et al., 1997; Goodman et al., 2005; Kelemen et al., 2005) that may alter the coal volume during the isotherm measurement.

Table 6 Gibbs excess CO2 sorption (daf) on Beulah Zap coal at 55 °C Laboratory 1

Laboratory 2

Laboratory 3

Laboratory 4

Laboratory 5

Laboratory 6

Pressure (MPa)

Gibbs excess sorption (mmol/g)

Pressure (MPa)

Gibbs excess sorption (mmol/g)

Pressure (MPa)

Gibbs excess sorption (mmol/g)

Pressure (MPa)

Gibbs excess sorption (mmol/g)

Pressure (MPa)

Gibbs excess sorption (mmol/g)

Pressure (MPa)

Gibbs excess sorption (mmol/g)

0.26 0.54 1.16 2.18 4.14 6.41 8.65 10.34 11.89 12.87 14.13 15.11

0.053 0.125 0.286 0.455 0.699 0.926 1.137 1.273 1.409 1.456 1.473 1.476

1.66 3.25 5.25 7.35 9.87 12.00 14.85

0.394 0.603 0.765 1.067 4.850 5.172 2.260

0.10 0.19 0.37 0.78 1.58 3.27 5.24 7.18 8.48 9.74 11.13 12.49 13.84

0.020 0.041 0.080 0.157 0.281 0.503 0.686 0.793 0.831 0.833 0.749 0.679 0.653

0.13 0.17 0.30 0.60 1.21 2.50 4.01 5.77 7.75 8.96 10.11 11.78 12.25 12.83 13.64

0.005 0.035 0.096 0.209 0.387 0.633 0.904 1.031 1.235 0.690 2.092 − 0.071 19.355 21.948 34.057

0.40 0.77 1.13 1.47 2.09 2.63 3.11 4.06 5.03 6.06 7.01 8.09 9.12 10.04 11.07 12.00 13.11 14.08

0.105 0.186 0.258 0.310 0.406 0.484 0.545 0.646 0.741 0.823 0.872 0.915 0.965 1.005 1.055 1.036 0.986 0.942

1.02 1.50 2.82 4.22 5.91 7.15 8.35 9.71 11.06 12.04 13.57

0.228 0.310 0.489 0.634 0.760 0.801 0.824 0.825 0.834 0.783 0.640

A.L. Goodman et al. / International Journal of Coal Geology 72 (2007) 153–164

159

the Argonne premium coal bank. The moisture contents of these as-received coals at 30 °C are compiled in Table 2 and are shown in Figs. 2–4 on the y1-axis as a solid horizontal line. The laboratories used these asreceived samples to prepare moisture-equilibrated coal samples according to the modified ASTM D1412-99 procedure described in the Experimental section. The only discrepancy to this procedure is for laboratory 6, where they used the Pocahontas No. 3 and Beulah Zap coals samples directly from the Argonne coal bank. The percent moisture values of the moisture-equilibrated coals before CO2 exposure are shown in Figs. 2–4 on the y1-axis as hashed bars. The moisture-equilibrated coals from laboratory 4 were consistently higher in moisture than the as-received moisture content of the Argonne premium coals. Laboratories 1, 5, and 6 obtained moisture contents that were similar to the asreceived moisture content of the Argonne premium coals. The samples from laboratories 2 and 3 were sometimes dryer or wetter than the as-received Argonne premium coals. Next, the laboratories measured CO2 excess sorption isotherms on the moisture-equilibrated Argonne coals at

Fig. 5. Excess CO2 sorption isotherms at 55 °C on moistureequilibrated Pocahontas No. 3 coal. a — the y-axis extends to 6 mmol/g and b — the y-axis extends to 2 mmol/g. (Laboratory 6 used the Pocahontas No. 3 coal directly from the Argonne coal bank).

The absolute isotherm data were then linearized and fit to the simple, yet widely accepted Langmuir model: h¼

V bP ¼ Vm 1 þ bP

ð2Þ

where b is the Langmuir constant, P is pressure, and θ is the fractional sorption and is equal to V/Vm where V is the volume sorbed and Vm is the volume sorbed at the monolayer point (Adamson, 1990). Models other than Langmuir are commonly used to fit CO2-coal isotherm data and can yield better fits (Sapienza et al., 1986; Arri and Yee, 1992; DeGance et al., 1993; Chaback et al., 1996; Clarkson et al., 1997; Fitzgerald et al., 2003; Ozdemir et al., 2003a,b; Sudibandriyo et al., 2003). However, the simplest and most widely used model was chosen as the starting point to fit the inter-laboratory results reported in this study. 3. Results The six laboratories were provided with Pocahontas No. 3, Illinois No. 6, and Beulah Zap coal samples from

Fig. 6. Excess CO2 sorption isotherms at 55 °C on moistureequilibrated Illinois No. 6 coal. a — the y-axis extends to 6 mmol/g and b — the y-axis extends to 2 mmol/g.

160

A.L. Goodman et al. / International Journal of Coal Geology 72 (2007) 153–164

Fig. 7. Excess CO2 sorption isotherms at 55 °C on moistureequilibrated Beulah Zap coal. a — the y-axis extends to 6 mmol/g and b — the y-axis extends to 2 mmol/g. (Laboratory 6 used the Beulah Zap coal directly from the Argonne coal bank).

55 °C and pressures up to 15 MPa by using volumetric, manometric, or gravimetric equipment. The sorption isotherms in terms of calculated CO2 Gibbs excess

sorption (mmol/g) as a function of pressure (MPa) are tabulated for the six laboratories in Tables 4–6 and charted in Figs. 5–7. Because of the wide variation in data, the excess sorption isotherms in Figs. 5–7 are displayed with two different y-axis ranges — plot a, where the y-axis extends to 6 mmol/g and plot b, where the y-axis extends to 2 mmol/g. The excess sorption isotherm for laboratory 4 extends up to 37 mmol/g as given in Tables 4–6 and is not included in Figs. 5–7. In general for the data in Figs. 5–7a, the excess sorption isotherms for laboratories 1, 3, 5, and 6 show similar behavior in that the excess sorption increases to 8 MPa, levels off, and then decreases after 10 MPa. The data for laboratories 2 and 4 also show an increase in excess sorption up to 8 MPa. After 8 MPa, the excess sorption for laboratory 2 rises sharply and then begins to decrease after 12 MPa while the excess sorption for laboratory 4 continues to rise sharply up to 15 MPa (Tables 4–6). When the y-axis for excess sorption scale is reduced to 2 mmol/g (Figs. 5–7b), the moisture-equilibrated excess isotherm curves for the six laboratories agreed well for the three coal samples up to 8 MPa with the exception of laboratories 3 and 4 for Pocahontas No. 3 coal (Fig. 5b) and laboratory 2 for Illinois No. 6 coal (Fig. 6b). Absolute sorption was then calculated according to Eq. (1), where the CO2 adsorbed density value used was 1.18 g/cm3 and the CO2 compressibility was calculated from the Span and Wagner formulation (Span and Wagner, 1996). The absolute isotherm data were then fit to the linear Langmuir equation to obtain the Langmuir fitting parameters Vm and b. The fitting parameters Vm, b, pressure range, and R value from the linear plots are

Table 7 Best fit parameters obtained from the linear Langmuir equation Parameter

Laboratory 1

Laboratory 2

Laboratory 3

Laboratory 4

Laboratory 5

Laboratory 6

Pocahontas No. 3 Vm (mmol/g) b (MPa− 1) Range (MPa) R value

1.37 0.775 0–10 0.99951

1.26 1.011 0–8 0.99804

1.77 0.590 0–8 0.99576

2.70 0.522 1–6 0.99623

1.45 0.719 0–8 0.99758

1.37 0.558 0–10 0.99767

Illinois No. 6 Vm (mmol/g) b (MPa− 1) Range (MPa) R value

1.62 0.235 0–10 0.98693

3.38 0.165 1–7 0.98325

1.59 0.228 0–8 0.99228

1.85 0.175 1–6 0.99272

1.67 0.190 0–8 0.99502

1.76 0.184 1–8 0.99079

Beulah Zap Vm (mmol/g) b (MPa− 1) Range (MPa) R value

2.54 0.107 1–6 0.98281

1.69 0.188 1–6 0.99941

2.50 0.085 0–11 0.99389

2.84 0.128 1–8 0.98337

2.26 0.114 0–8 0.98909

2.05 0.121 0–10 0.9977

A.L. Goodman et al. / International Journal of Coal Geology 72 (2007) 153–164

161

equation of state, all laboratories should obtain essentially identical excess sorption isotherms. This should result in identical absolute sorption isotherms when using the same sorbed phase density values. Whether and to what extent the isotherm data submitted by the participants are accurate is not the scope of this paper. For the moisture equilibration measurements, the asreceived coal samples from the Argonne premium coal bank were moisture-equilibrated according to the modified standard procedure described in the Experimental section. The moisture-equilibrated values reported by the six laboratories for the same coals varied significantly among the laboratories and from the as-received moisture of the Argonne premium samples (Figs. 2–4). The moisture-equilibrated coals from laboratory 4 were consistently higher in moisture than the as-received moisture content of the Argonne premium coals. Laboratories 1, 5, and 6 obtained moisture contents that were similar to the as-received moisture content of the Argonne premium coals. The samples from laboratories 2 and 3 were sometimes dryer or wetter than the asreceived Argonne premium coals. The fact that the

Fig. 8. Absolute CO2 sorption isotherms at 55 °C on moistureequilibrated Pocahontas No. 3 coal. a — the y-axis extends to 10 mmol/g and b — the y-axis extends to 2.5 mmol/g. The solid lines represent the best fit to the Langmuir model. (Laboratory 6 used the Pocahontas No. 3 coal directly from the Argonne coal bank).

compiled in Table 7. The absolute sorption isotherms are shown in Figs. 8–10 with two y-axis ranges — plot a, where the absolute sorption extends to 10 mmol/g and plot b, where the absolute sorption extends to 2.5 mmol/g. The symbols represent the absolute data and the solid lines represent the isotherm predicted by the Langmuir equation. The Langmuir model fit most of the data up to 8 MPa. At higher pressures, the data sets deviate from the Langmuir model. 4. Discussion Until recently, there have been very few high-pressure studies of CO2-coal sorption isotherms under in-seams conditions (Krooss et al., 2001, 2002; Fitzgerald et al., 2003, 2005; Siemons and Busch, 2007). However, this void is being filled because of interest in ECBM recovery and CO2 sequestration (White et al., 2005). The main point of the study was to assess to what extent CO2 sorption isotherms are comparable. For identical temperatures and moisture contents and using the same

Fig. 9. Absolute CO2 sorption isotherms at 55 °C on moistureequilibrated Illinois No. 6 coal. a — the y-axis extends to 10 mmol/g and b — the y-axis extends to 2.5 mmol/g. The solid lines represent the best fit to the Langmuir model.

162

A.L. Goodman et al. / International Journal of Coal Geology 72 (2007) 153–164

Fig. 10. Absolute CO2 sorption isotherms at 55 °C on moistureequilibrated Beulah Zap coal. a — the y-axis extends to 10 mmol/g and b — the y-axis extends to 2.5 mmol/g. The solid lines represent the best fit to the Langmuir model. (Laboratory 6 used the Beulah Zap coal directly from the Argonne coal bank).

laboratories followed essentially the same procedure that resulted in such scattered data may indicate that the seemingly small modifications of the procedure had an unexpectedly large effect on the inter-laboratory reproducibility for coal moisture-equilibration content. For the isotherm measurements, excess sorption was calculated at pressures up to 15 MPa. In general, the excess sorption isotherms for the three Argonne premium coal samples are in fair agreement up to 15 MPa with the exceptions of laboratories 2 and 4 (Figs. 5–7a). The data for laboratory 4 decrease after 8 MPa and then increase sharply up to 30 mmol/g. The data for laboratory 2 increase sharply at 8 MPa and then begin to decrease by 12 MPa. Apparent factors such as equipment (manometric, volumetric, and gravimetric), sample size, equilibration time, apparatus dimensions, and coal moisture content do not explain the significant divergence in the data above 8 MPa for laboratories 2 and 4 (Table 3). Both laboratories 2 and 4 based their measurements on a manometric design. The amount of coal ranged from approximately 2 g for laboratory 2 to

79 g for laboratory 4. CO2 equilibration was reached in 30 min for laboratory 2 and up to 12 h for laboratory 4. Sample cell, reference cell, and void volumes were significantly smaller for laboratory 2, approximately 10 cm3, when compared to laboratory 4, approximately 170 cm3. Finally, coal moisture contents for laboratory 2 were predictably much dryer than those reported by laboratory 4. Thus, no obvious tend could explain why the isotherm data for laboratories 2 and 4 diverged from laboratories 1, 3, 5, and 6 at pressures above 8 MPa. For the isotherm measurements below 8 MPa, the overall agreement is good with minor exceptions from laboratories 3 and 4 for the Pocahontas No. 3 sample and laboratory 2 for the Illinois No. 6 sample. Again, factors such as equipment (manometric, volumetric, and gravimetric), sample size, equilibration time, and apparatus dimensions do not explain these minor differences (Table 3). However, coal moisture may account for the minor differences since laboratories 2, 3, and 4 reported moisture-equilibrated values that were either higher or lower than the as-received coal moisture reported by the Argonne premium coal program. For the Pocahontas No. 3 coal data set, laboratories 3 and 4 show greater amount of CO2 sorption when compared with the other four laboratories. The percent moisture values for laboratories 3 and 4 for the Pocahontas No. 3 coal are much greater than the other laboratories: 14% for laboratory 3 and 19% for laboratory 4 and 2%, 1%, 0.8%, 0.7% moistures for laboratories 1, 2, 5, and 6, respectively. For the Illinois No. 6 coal, the isotherm for laboratory 2 shows a higher sorption capacity when compared with the other five laboratories. Laboratory 2 reports a much lower moisture value than the other laboratories: 4% moisture for laboratory 2 and 8%, 13%, 20%, 9%, and 9% for laboratories 1, 3, 4, 5, and 6, respectively.

Fig. 11. Calculated CO2 solubility in water (mmol/g) at 55 °C from Duan and Sun (2003).

A.L. Goodman et al. / International Journal of Coal Geology 72 (2007) 153–164

Because the starting coal samples were not at the same moisture content after the standard wetting procedure, differences in the resultant excess isotherm should be expected. The excess sorption CO2-coal isotherms were corrected for moisture by using the moisture values measured by the laboratories. However, this may not be adequate since CO2 solubility in water is almost as high as CO2 sorption on the coal samples as shown by Duan and Sun (2003) in Fig. 11. A recent conference proceeding by Hartman and Pratt (2005) discusses CO2 isotherm measurements conducted on coal samples that contain moisture levels above the as-received moisture content of the coal sample. Their study indicated that excess surface moisture on the coal samples increases the CO2 storage capacity of the coal. Their data suggest that CO2 dissolves in the excess surface moisture in the coal matrix causing the resultant isotherm to over-predict the CO2 storage capacity of the coal. Conversely, if isotherms are measured on coal samples that contain moisture levels below the as-received moisture content reported by the Argonne premium coal program, then a greater CO2 storage capacity would be expected for drier coal samples (Unsworth et al., 1989; Clarkson and Bustin, 2000; Krooss et al., 2002; Ozdemir et al., 2003a, b; Goodman et al., 2004). For the absolute sorption calculations, the data from the six laboratories follow the Langmuir model up to 8 MPa. After 8 MPa, all data sets diverged from the Langmuir equation. The Langmuir model was used to compare the absolute sorption calculated for each laboratory at 8 MPa with the moisture content of the samples. Figs. 2–4 show the absolute sorption predicted by the Langmuir model at 8 MPa (y2-axis, solid circles) and the percent equilibrium moisture content of the coals (y1-axis, hashed bars). The majority of the laboratories predict similar storage capacities at 8 MPa except when the coal moisture is either above or below the as-received moisture threshold. 5. Conclusions This study provides the first inter-laboratory comparison of carbon dioxide isotherm measurements for moisture-equilibrated coal. The overall agreement between the laboratories was good up to 8 MPa with the exception of those instances where moisture content of the coals was either higher or lower than the as-received moisture threshold. At CO2 pressures above 8 MPa, the reported sorption isotherms diverged significantly. Further studies need to be conducted in order to address deviations and experimental problems associated with measuring high-pressure CO2 sorption isotherms.

163

Acknowledgements A. Busch and B.M. Krooss gratefully acknowledge support in sample preparation and helpful comments on the manuscript by D. Prinz. References Adams, E.E., Crounse, B., Harrison, T., Socolofsky, S.A., 2000. Analytical and experimental studies of droplet plumes with application to CO2 ocean sequestration. Abstracts of Papers of the American Chemical Society 220: 108-FUEL Part 1 AUG 20. Adamson, A.W., 1990. Physical Chemistry of Surfaces, 5th edition. John Wiley and Sons Inc., New York. Arri, L.E., Yee, D., Morgan, W.D., Jeansonne, M.W., 1992. Modeling coalbed methane production with binary gas sorption. SPE24363, Society of Petroleum Engineers Rocky Mountain Regional Meeting. Casper, Wyoming. ASTM D 1412-99, 2000. Standard test method for equilibrium moisture of coal at 96 to 97 percent relative humidity and 30 °C. Gaseous Fuels; Coal and Coke, Section 5 Petroleum Products, Lubricant, and Fossil Fuels. West Conshohocken, PA, vol. 05.06. ASTM E 691–99, 2000. Standard practice for conducting an interlaboratory study to determine the precision of a test method. Annual Book of ASTM Standards, West Conshohocken, PA. Chaback, J.J., Morgan, W.D., Yee, D., 1996. Sorption of nitrogen, methane, carbon dioxide and their mixtures on bituminous coals at in-situ conditions. Fluid Phase Equilibria 117, 289–296. Clarkson, C.R., 1999. Binary gas adsorption/desorption isotherms: effect of moisture and coal composition upon component selectivity. Proceedings: International Coalbed Methane Symposium. University of Alabama, Tuscaloosa, AL, pp. 91–115 (May 3–7). Clarkson, C.R., Bustin, R.M., 2000. Binary gas adsorption/desorption isotherms: effect of moisture and coal composition upon carbon dioxide selectivity over methane. International Journal of Coal Geology 42, 241–271. Clarkson, C.R., Bustin, R.M., Levy, J.H., 1997. Application of the mono/multilayer and adsorption potential theories to coal methane adsorption isotherms at elevated temperature and pressure. Carbon 35, 1689–1705. DeGance, A.E., Morgan, W.D., Yee, D., 1993. High pressure adsorption of methane, nitrogen and carbon dioxide on coal substrates. Fluid Phase Equilibria 82, 215–224. Duan, Z., Sun, R., 2003. An improved model calculating CO2 solubility in pure water and aqueous NaCl solutions from 273 to 533 K and from 1 to 1000 bar. Chemical Geology 193, 257–271. Fitzgerald, J.E., Sudibandriyo, M., Pan, Z., Robinson, R.L., Gasem, K.A.M., 2003. Modeling the adsorption of pure gases on coals with the SLD model. Carbon 41, 2203–2216. Fitzgerald, J.E., Pan, Z., Sudibandriyo, M., Robinson, R.L., Gasem, K.A.M., 2005. Adsorption of methane, nitrogen, carbon dioxide, and their mixtures on wet Tiffany coal. Fuel 84, 2351–2363. Gasem, K.A.M., Fitzgerald, J.E., Pan, Z., Sudibandriyo, M., Robinson, R.L., 2001. Modeling of gas adsorption on coals. Proceedings: Eighteenth Annual International Pittsburgh Coal Conference, Newcastle, NSW, Australia, December 3–7. Goodman, A.L., Busch, A., Duffy, G.J., Fitzgerald, J.E., Gasem, K.A.M., Gensterblum, Y., Krooss, B.M., Levy, J., Ozdemir, E., Pan, Z., Robinson, R.L., Schroeder, K.T., Sudibandriyo, M., White, C.M., 2004. An interlaboratory comparison of CO2

164

A.L. Goodman et al. / International Journal of Coal Geology 72 (2007) 153–164

isotherms measured on Argonne premium coal samples. Energy and Fuels 18, 1175–1182. Goodman, A.L., Favors, R.N., Hill, M.M., Larsen, J.W., 2005. Structure changes in Pittsburgh No. 8 coal caused by sorption of CO2 gas. Energy and Fuels 19, 1759–1760. Hartman, R.C., Pratt, T.J., 2005. A preliminary study of the effect of moisture on the carbon dioxide storage capacity in coal. Proceedings: International Coalbed Methane Symposium. University of Alabama, Tuscaloosa. Hsieh, S.T., Duda, J.L., 1987. Probing coal structure with organic vapor sorption. Fuel 66, 170–178. Humayun, R., Tomasko, D.L., 2000. High-resolution adsorption isotherms of supercritical carbon dioxide on activated carbon. American Institute of Chemical Engineers (AICHE) 46, 2065–2075. Kelemen, S.R., Kwiatek, L.M., Siskin, M., Lee, A.G., 2005. Structural response of coal to drying and pentane sorption. Energy and Fuels 20, 205–213. Krooss, B.M., Gensterblum, Y., Siemons, N., van Bergen, F., Pagnier, H.J.M., David, P., 2001. High-pressure methane and carbon dioxide adsorption on dry and moisture-equilibrated Carboniferous coals. Proceedings: International Coalbed Methane Symposium. University of Alabama, Tuscaloosa, Alabama. Krooss, B.M., van Bergen, F., Gensterblum, Y., Siemons, N., Pagnier, H.J.M., David, P., 2002. High-pressure methane and carbon dioxide adsorption on dry and moisture-equilibrated Pennsylvanian coals. International Journal of Coal Geology 51, 69–92. Larsen, J.W., Flowers, R.A., Hall, P.J., Carlson, G., 1997. Structural rearrangement of strained coals. Energy and Fuels 11, 998–1002. Lu, X.C., Li, F.C., Watson, A.T., 1995. Adsorption measurements in Devonian shales. Fuel 74, 599–603. Ozdemir, E., Morsi, B.I., Schroeder, K.T., 2003a. Importance of volume effects to adsorption isotherms of carbon dioxide on coals. Langmuir 19, 9764–9773. Ozdemir, E., Schroeder, K.T., Morsi, B.I., White, C.W., 2003b. Mechanism of carbon dioxide (CO2) adsorption on moist coals. Abstracts of the Papers of the American Chemical Society, 225: U851–U852 057-FUEL Part 1 MAR. Ozdemir, E., Morsi, B.I., Schroeder, K.T., 2004. CO2 adsorption capacity of Argonne premium coals. Fuel 83, 1085–1094. Reucroft, P.J., Patel, H., 1986. Gas-induced swelling in coal. Fuel 65, 816–820.

Reucroft, P.J., Sethuraman, A.R., 1987. Effect of pressure on carbon dioxide induced coal swelling. Energy and Fuels 1, 72–75. Ruppel, T.C., Grein, C.T., Beinstock, D., 1972. Fuel 51, 297–303. Sapienza, R., Butcher, T., Slegeir, W., Healy, F., Vorres, K.S., 1986. Carbon Dioxide/Water for Coal Beneficiation Mineral Matter and Ash in Coal. American Chemical, Washington, DC, pp. 500–512. Shimizu, K., Takanohashi, T., Iino, M., 1998. Sorption behaviors of various organic vapors to Argonne premium coal samples. Energy and Fuels 12, 891–896. Siemons, N., Busch, A., 2007. Measurement and interpretation of supercritical CO2 sorption on various coals. International Journal of Coal Geology 69, 229–242. Span, R., Wagner, W., 1996. A new equation of state for carbon dioxide covering the fluid region from the triple-point temperature to 1100 K at pressures up to 800 MPa. Journal of Physical Chemistry 25, 1509–1596. St. George, J.D., Barakat, M.A., 2001. The change in effective stress associated with shrinkage from gas desorption in coal. International Journal of Coal Geology 45, 105–113. Sudibandriyo, M., Pan, Z., Fitzgerald, J.E., Robinson, R.L., Gasem, K.A.M., 2003. Adsorption of methane, nitrogen, and carbon dioxide, and their binary mixtures on dry activated carbon at 318.2 K and pressures up to 13.6 MPa. Langmuir 19, 5323–5331. Takanohashi, T., Terao, Y., Yoshida, T., Iino, M., 2000. Adsorption and diffusion of alcohol vapors by Argonne premium coals. Energy and Fuels 14, 915–919. Unsworth, J.F., Fowler, C.S., Jones, L.F., 1989. Moisture in coal. Fuel 68, 18–26. Vorres, K.S., 1990. The Argonne premium coal sample program. Energy and Fuels 4, 420–426. Vorres, K.S., 1993. Users Handbook For The Argonne Premium Coal Sample Program. http://www.anl.gov/PCS/pcshome.html. Walker, P.L., Verma, S.K., Rivera-Utrilla, J., Khan, R., 1988. A direct measurement of expansion in coals and macerals induced by carbon dioxide and methanol. Fuel 67, 719–726. 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., 2005. Sequestration of carbon dioxide in coal with enhanced coalbed methane recovery—a review. Energy and Fuels 19, 659–724.