Nomenclature
Literature Cited
General
R = gasconstant T = absolute temperature, K u, I. = molar volume of species i, cm”g-mo1 x , = liquid mole fraction of species i y L = activity coefficient of species i y L = infinite dilution activity coefficient of species i
A,, = transformed Wilson parameter A,, - A,, = Wilson parameter, cal/g-mol
Specific
F = variable defining uniqueness of Wilson parameter = values that y1 may assume when Wilson parameters are not unique f l ’ = first derivative of y1 with respect to t f l ” = second derivative of y1 with respect to t gl = the Wilson activity coefficient for species 1 gz = the Wilson activity coefficient for species 2 gz’ = the first derivative of gz with respect to t gz” = the second derivative of g2 with respect to t t = transformed coordinate z = transformed coordinate fl
Aristovich, V., Polozov, A. G., Terpusov, A. K., Sabylin, I. I., 2. Prikl. Khim. (English trans/.), 42, 1531 (1969). Gardner, R. S., U.S. Naval Ordinance Test Station, China Lake, Calif., 1960. Garrett, G. R., Van Winkle, M., J. Chem. Eng. Data, 14, 304 (1969). Hankinson, R. W., Langfitt, B. D., Tassios, D., Can. J. Chem. Eng., 50, 511 (1972). Hudson, J. W., Van Winkle, M., Ind. Eng. Chem., Process Des. Dev., 9, 466 (1970). Hudson, J. W., Van Winkle. M., J. Chem. Eng. Data, 14, 310 (1969). Karr, A. E., et al., Ind. Eng. Chem., 43 (4), 961 (1951). Kudryautseva. L. S., Susarev. M. P., Zh. Prikl. Khim. (English trans/.), 36, 1231 (1963). Larson, D.,personal correspondence, 1971. Marina, J., Tassios, D., lnd. Eng. Chem., Process Des. Dev., 12, 67 (1973). Miyahara, K., Sadornote, W., Kitarnura, K.. J. Chem. Eng. Jpn., 3, 157 (1970). Nagata. I., J. Chem. Eng. Data, 7, 360, 367 (1962). Nagata, I., Hayashida, H., J. Chem. Eng. Jpn., 3, 161 (1970). Orye. R. V., Prausnitz, J. M., lnd. Eng. Chem., 57 (5), 18 (1965). Schreiber, L. B.. Eckert, C. A,, lnd. Eng. Chem., Process Des. Dev.. 10, 572 (1971). Severns, W. H.,et at., A.i.Ch.E. J., 1, 401 (1955). Silverrnan, N., Tassios. D., Paper at A.1.Ch.E. National Meeting, New Orleans, La., 1973. Vinichenko, I. G., Susarev, M. P., Zh. Prikl. Khim. (English transi.), 38, 2701 (1965). Willock. J. M., Van Winkle, M..J. Chem. Eng. Data, 15, 281 (1970). Wilson, G. M., J. Am. Chem. SOC.,86, 127 (1964).
Receiued /or reuiew October 6, 1975 Accepted September 7, 1976
Reactivity of Coal and Char. 1. In Carbon Dioxide Atmosphere S. Dutta and C. Y. Wen* Department of Chemical Engineering, West Virginia University, Morgantown, West Virginia 26506
R. J. Belt Energy Research and Development Administration, Morganto wn Energy Research Center, Morgantown, West Virginia 26506
Reactivities of a few raw coals and chars of these coals obtained from gasifiers operating under different conditions have been measured in CO2 at temperatures of 840-1 100 ‘C.The reactivities have been measured in a thermogravimetric analyzer up to complete conversions of the samples in most cases. Properties such as surface area, pore size distribution, porosity, and density have been determined for each sample. Actual pore structures of a few samples have been observed at different conversion levels by a scanning electron microscope. In order to compare the reactivities of different samples, the gasification process has been divided into two distinct stages: the first stage due to pyrolysis and the second stage due to char-C02 reaction. Reactivities due to the firs stage can be roughly related to volatile matter contents of the solids and the rate of heating. Through an Arrhenius type equation, an activation energy of about 2.5 kcal/mol is obtained for the first stage. The reactivity of a char in the second stage is found to depend more on its coal seam than on the gasification scheme in its production. Activation energy for the second stage reaction has been found to be about 59 kcal/mol. A rate equation has been proposed for the second stage that incorporates the effect of relative available pore surface area changing during reaction.
Introduction
A proper understanding of the coallchar gasification kinetics is essential for successful design of a gasifier. Effects of temperature, pressure, and gaseous environments on the rate of gasification of coal/char have been extensively studied by various investigators. In addition, the rate depends also on the nature and origin of the coal or char itself. The pore characteristics and hence the reactivity of a char have been found to vary not only with the maceral of its parent coal but also with the history of its genesis, i.e., the temperature, pressure, 20
Ind. Eng. Chem., Process Des. Dev., Vol. 16, No. 1, 1977
rate of heating, and gaseous environments, etc., prevailing during its formation. The present study is devoted to determination of reactivities of a few coal and char samples which are produced in some pilot plant experiments conducted under different gasification schemes. This investigation will help us to find the relationship, if any, between the reactivities and the physical characteristics of the samples. In the present investigation reactivities are measured in a flowing stream of pure COZ at atmospheric pressure. The rate of C-COz reaction has been studied by various investigators (Gadsby et al., 1948; Gulbransen and Andrew, 1952; Walker
Table I. C'hemical Analysis of Coal and Char Samples__ ~
---
--
_
.
_
_
_
IGT Char No. HT155
_
_
_
Hydrane Char No. 49
Synthane Char No. 122
Illinois Coal No. 6
Hydrane Char No. 150
Pittsburgh HVab coal
3.90 32.80 14.67 48.63
0.21 12.30 8.05 79.44
1.20 36.40 6.70 55.70
8.05 82.00 2.89 0.96 1.02 4.87
6.78 76.74 5.34 2.74 1.42 6.66
Proximate Analysis, wt % Moisture Volatile matter Ash Fixed carbon
1.00 2.80 22.10 74.10
0.10 4.00 26.05 69.84
0.26 33.88 -
Ultimate Analysis, w t % (Moisture-Free) Ash Carbon Hydrogen Sulfur Nitrogen Oxygen
22.34 75.10 1.25 1.37 __
0.41
26.06 70.38 1.59 0.88 0.86 0.13
et al., 1953; Petersen et al., 1955; Wicke, 1955; Rossberg and Wicke, 1956; Ergun, 1956; Walker et al., 1959; Austin and Walker, 1963; Turkdogan et al., 1968; Turkdogan and Vinters, 1969; Yoshida and Kunii, 1969; Turkdogan et al., 1970; Wen and Wu, 1976; Walker and Hippo, 1975, and Fuchs and Yavorsky, 1975). However, considerable discrepancy has been reported between the values of activation energy of this reaction, ranging from 48 to 8 6 kcal/mol. Pyrolysis of coal or char takes place prior to or concurrent with other reactions in a gasifier. The behavior of pyrolysis is not yet properly established. It is, however, known that the rate of pyrolysis and the amount and composition of volatile products from a given sample of coal or char depends on several factors (Gray et al., 1973; Anthony et al., 1975) such as (a) rate of heating, (b) final decomposition temperature attained, (c) vapor residence time, (d) the environment under which the pyrolysis takes place, (e) pressure, (f) coal particle size, and (g) coal type. In the present investigation pyrolysis of coal and char in a CO? atmosphere can be studied separately from COZ reaction. This is due to the fact that pyrolysis normally starts at about 350-400 "C and is almost complete at about 1000 "C in seconds (Essenhigh and Howard, 1966), whereas the C-CO2 reaction is hardly detectable below 800-900 "C. Therefore, at a moderate rate of heating, the two stages, the pyrolysis and the char-COz reaction, will be separable from each other, the latter starting only after the former stage is essentially complete. The char-COZ reaction, after the pyrolysis reaction is completed, takes place on the char surface and is essentially a carbon-COn reaction. Gulbransen and Andrew (1952) showed that the internal surface area of graphite increases markedly during reaction with both oxygen and COz. Walker et al. (1953) made a detailed study on the possible correlation existing between reaction rates and changes in surface area during reaction. They concluded that the reaction develops new surface by enlarging to some extent the micropores of the solid but principally by opening up pore volume not previously available to reactant gas because the microcapillaries were too small or because existing pores were unconnected. During the reaction, surface area increases up to a point when the rate of formation of new area is paralleled by the rate of destruction of the old area. Surface area decreases on further conversion. For the graphite-CO2 reaction, Petersen et al. (1955) found that the observed rates were not simple functions of the total available surface area, as determined by the low-temperature gas adsorption technique, as might be expected if the reaction was chemical reaction controlled. Turkdogan et al. (1970) made a detailed investigation on the pore characteristics of several forms of carbon. Their studies indicate that, depending on the type of carbon, about
14.67 67.43 4.44 2.08 1.46 9.92
33.88 60.00 1.16 1.23 1.12 2.35
Table 11. Conditions of Char Generation in Gasifiers Char
Temp, O
C
Pressure, psig
Feed gas Hydrogensteam Hydrogen methane Oxygen-steam Hydrogen methane
IGT No. HT155
927
1000
Hydrane No. 49
850
1000
Synthane No. 122 Hydrane No. 150
948 900
300 1100
Y4 to of the volume is isolated by micropores and hence is not available for reaction at the beginning. The surface areas of carbon investigated covered a large range from 0.1 to 1100 m2/g. In all cases, most of the internal area was attributed to the micropores, 10-50 8, in diameter.
Experimental Section A Fischer TGA apparatus (Model 120P) was used in the present investigation. Two coal samples (one from Pittsburgh seam and the other from Illinois seam) and four char samples derived from these coals under different gasification schemes were investigated. The chemical compositions of the samples and the conditions used in the gasifiers for generation of the chars are shown in Tables I and 11, respectively. To start a run, 15-30 mg of coal or char particles of -35 +60 mesh was placed in the platinum holder hanging from one arm of the balance of the TGA. After fixing the hangdown tube in position, the system was first evacuated to about 20 mmHg and then flushed with COZat a flow rate of about 3.33 cm"/s for about 2.5 h. The outlet gas was analyzed by gas chromatography to assure that it is air-free. The furnace was preheated to the desired temperature keeping the hangdown tube outside it. Then the COZgas stream was turned to the desired flow rate (2.5 cm3/s) and the furnace was quickly raised to a prefixed level to enclose the hangdown tube. The weight and the time derivative of weight loss of the sample and the sample temperature were recorded continuously throughout the experiment. The porosity, density, pore volume, and pore size distribution of the devolatilized chars and coals have been determined by mercury porosimetry using pressures up to 50 000 psi. The pore surface areas and pore size distribution of the samples have been determined by the BET nitrogen adsorption method using NUMEC surface-area-apparatus, Model AfA4. Moreover, the actual macropore matrix of a few of these samples has been visually observed, at several stages of their conversions, by a scanning electron microscope up to a magnification of 6000. Ind. Eng. Chem., Process Des. Dev., Vol. 16, No. 1, 1977
21
50
1
I
60
70
F i g u r e 1. Typical chart recordings of the weight-loss and the rate of weight.10~~ curves: 1, Hydrane char No. 150; 2, Hydrane char No. 49.
Experimental Results a n d Discussions Figure 1 is the reproduction of two typical chart recordings of the weight loss and the rate of weight loss curves by the T G A apparatus. Iri this figure the initial peaks in the rate curves are due to the very fast pyrolysis stage. After this stage, the second stage reaction begins which is comparatively slow and is mostly the C--C02 reaction. The analyses of the chars just after the pyrolysis stage show carbon contents of 95-98% on an ash-free basis for all the six samples. Figme 1also shows an intermediate region, between the two vertical lines a and b, where not only the pyrolysis but also the second stage char--C& reaction is affected by the heating rate of the sample. The sample temperature vs. time for these two cases are shown in Figure 2. In these experiments the sample temperature is assumed to be identical with that recorded by an open thermocouple placed about 3 mm from the surface oft he reacting solid sample. The rate of gasification at the first and intermediate stages will obviously depend on the sample heating rate. Once the sample attains equilibrium temperature (approximately within 4 min in the present case), the rest of the process proceeds essentially under isothermal condition. A. The First Stage-Pyrolysis. The rate of gasification and the fraction conversion due to pyrolysis are shown for four samples in Figure 2. The total conversions obtained a t this stage have been found to be nearly equal to the volatile matter contents of the chars and coals as determined by the proximate analyses. The conversion (/) in the first stage (pyrolysis) ib defined here as that conversion which is attained a t the almost-constant weight period (the region between the vertical lines a and b in Figure 2 ) immediately after the rapid weight loss at the start of the process. Such a constant-weight-period has been observed in almost all cases. E'igure 2 also shows that the pyrolysis is almost complete before a temperature of about 800 "C is reached, while the char-CO2 reaction is insignificant up to this temperature, as will be seen later. The total conversion ( f )from the pyrolysis stage increases with the increase in temperature. However, this increase (-1%) is not appreciable, at a particular heating rate, in the studied temperature range 843-1074 "C. Figure 3 shows the effect of sample heating rate on the rate of pyrolysis of hydrane char No. 150. I t is noted that the peak heights of these rate curves are roughly proportional to the average slopes of the heating rate curves. A Pyrolysis Model. Pyrolysis cannot be considered as a single-step process involving a simple reaction. I t occurs in stages or as "waves" of reactions involving many complex 22
Ind. Eng. Chem., Process Des. Dev., Vol. 16,No. 1, 1977
TIME.
I
Figure 2. Typical rate and conversioii curves for the pyrolysis of coal arid chars: - - - -, conversion vs. time curve; -, rate vs. time curve; ---- -, temperature vs. time curve; 1, Hydrane Char No. 150; 2, Hydrane Char No. 49; 3, Synthane Char No. 122; 4, Illinois Coal No. 6
TIME,
I
F i g u r e 3. Effect of sample heating rate on the pyrolysis of Hydrane Char No. 150; sample size: 18.30 mg.
steps, which in turn vary from sample to sample and with the conditions of pyrolysis. No simple model would, therefore, represent this process completely. However, on an overall basis, Wen et al. (1974) proposed an Arrhenius type equation
as follows
dx -- Ae--H:R?'(f- x )
(1)
dt
The conversion, x, due to pyrolysis has been defined here as
are the initial weight of coal/char, the where w g , w,and instantaneous weight of coal/char, and the weight of ash present in the coal/char, respectively. f is assumed constant in the studied temperature range. Equation 1 has been tested by assigning arbitrary values of A and B and seeing whether the resulting values of x and dxldt can match the experimental ones. Since temperature is changing with time in this region, the values of x have been determined as a function of time from the following equation = f [ l- e - A j A e-*'3'dt
1
(3)
The evaluation of the integral in the above equation has been done numerically from known values of temperature (5") as a function of time ( t ) .The rate, dxldt, is next calculated from eq 1.With A = 2,500 cal/mol and B = 0.33 s-l, the predicted conversion and rate curves for three chars have been found t o match the experimental curves quite closely, as is shown in Figure 4a. The predicted curves and the experimental values for different heating rates are shown in Figure 4b. Calculations have been done only up to 1.5 min in order to avoid the possible influence of the second stage process in the subsequent period. Thus if the Arrhenius-type eq 1 is assumed to approximate the pyrolysis stage, the value of the activation energy becomes 2.5 kcal/mol. I t should be noted that a simple approach as above cannot truly account for the rate of the complex pyrolysis process. However, many complex mechanistic approaches that tried to take account of the stagewise releases of different functional groups of coal/char during pyrolysis are not only complicated but also are unable to describe the reaction rates accurately. Use of a large number of parameters may explain the pyrolysis mechanism of a specific size and type of coal a t a specific condition reasonably well, but may not be useful for design purposes when complicated hydrodynamic problems and other chemical reactions are taking place simultaneously. Therefore, in this study, for convenience, we tried this simpler approach that can take account of the overall phenomenon, with all the effects being lumped into the two parameters A and B of the Arrhenius-type equation. We found that this approach, although crude, can represent the overall trend of the characteristic rate reasonably well for the samples studied. B. The Second Stage-Char-CO? Reaction. As has been mentioned earlier, the rate of pyrolysis or the first stage of char/coal gasification is very rapid and can be assumed to be nearly complete when an almost-constant-weight period is attained in the weight loss vs. time curve. The remaining fraction of the solid reacts slowly with COSwhich is termed the second stage of gasification. Although a small part of volatile matters may still remain with this fraction, this fraction consists essentially of carbon (95-98%), and ash. In this second stage the rate, dx,/dt, and conversion, x,, are based on the reactive portion of char, which is the weight of solid remaining after the first stage less the weight of ash, and is termed as the base carbon. The conversion, x c , due to the second stage char-COS reaction is thus defined as
Figure 4. Calculated (-) and experimental ( - - - -) rate and conversion curves: (a) for different char samples: I-Hydrane Char No. 150, 2-Synthane Char No. 122,3-Hydrane Char No. 49; (b) for different heating rates for hydrane Char No. 150. The numbers 1,2, and 3 correspond to the heating rate curves shown in Figure 3.
Before studying the effect of temperature the effects of sample size, sample holder, particle size, and gas (COS) flow rate on char gasification rate were examined. Three kinds of sample holders of different sizes and shapes were tested using different amounts of samples placed on them. They were (a) a shallow petri dish type holder of 0.83 cm diameter and 0.2 cm depth, holding 3.75 mg of sample, (b) a cup-shaped holder of 0.85 cm mouth diameter and 0.7 cm depth, holding 18.5 mg of sample, and (c) a perforated cylindrical holder of 0.6 cm diameter and 1.5 cm depth, holding 35.02 mg of sample. No significant difference was found between the observed rates in the three cases a t a gasification temperature of 1025 "C and a gas flow rate of 2.5 cm3/s. The cup-shaped holder was used in the rest of the experiments. Gas flow rates were varied in the range 0.70-3.5 cm3/s through the reactor tube of 19 mm diameter. No change in rate was observed a t flow rates above 1.17 cm3/s a t the gasification temperature of 1025 "C. A flow rate of 2.5 cm3/s was chosen for subsequent experiments. Particle size selected was -35 +60 mesh, which showed negligible intraparticle diffusion resistance u p to a temperature of about 980 "C. The intraparticle diffusion resistance was considered negligible under these conditions due to the following reasons. (a) Four particle sizes, -20+35, -35+60, -60+100, and -100 mesh, were used to see the particle size effect on the reactivities of four char samples a t about 1000 "C. The effect was found to be negligible. (b) The slopes of the Arrhenius plots for all the samples u p to 1000 "C gives an activation energy of 59.26 kcal/mol which falls in the range (appearing to be in the chemical-reaction-control region), reported by most of the previous investigators. Ind. Eng. Chem., Process Des. Dev., Vol. 16,No. 1, 1977
23
Fraction
003
in Char
Carbon
670
6MI
431
795
156
.r
.yo
,
.y , .y , .y
Fraction a r b o n in C h a r
.€fa
i I I
I
, /
/
,
/
0 1
1-x,
Figure 5. Gasification of Hydrane Char No. 49. Rate vs. conversion based on base-carbon: - - - - - -, calculated rates.
Fraction
10
OW
I
616
445 1
,
I
,
I
Carbon in Char 706 76 2
,
,
,
,
Figure 7. Gasification of Hydrane Char No. 150. Rate vs. conversion based on base-carbon: - - - -, calculated rates.
Fractan
000
,
-k
I
6 :
I
Carbon
.y ! , a , :
In
Char .75
,
6,&.
10670/
1 -Ic 1-
x,
Figure 6. Gasification of IGT Char No. H T 155. Rate vs. conversion based on base-carbon: - - - - - -, calculated rates.
Figures 5-8 show the rate vs. conversion curves as a function of temperature for four of the samples studied. These figures also show the base carbon content of the chars at the fraction conversions indicated. The experiments were conducted up to the complete conversions of all the samples, except at lower temperatures where rates were extremely slow. Figures 5-8 (and also for the other two samples, not shown) clearly show that every sample has its own characteristic rate-conversion curve. For Hydrane Char No. 49, the curves can be considered linear (and passing through the origin) 24
Ind. Eng. Chem., Process Des. Dev., Vol. 16, No. 1, 1977
Figure 8. Gasification of Pittsburgh HVab Coal. Rate vs. conversion based on base-carbon: - - - -, calculated rates.
without significant error. However, for the other samples this is not true. Again for Hydrane Char No. 150,the rates are almost steady up to certain conversion levels, after which they decline. For Pittsburgh coal, rate-conversion curves show maxima at lower temperatures. It is also noted that for the Pittsburgh coal and for the Hydrane Char No. 150,the nature of the curves changes also with temperature. This change is shown more clearly in Figure 9, where the rates are normalized with respect to the rates observed at 20% conversion level and plotted against con-
1;
1010
09 07
05
0.4 r\
'03
2
P' 5
02
Y
c
a 0
-
Y e
R
0.1
",,,,,,\I 0.2
C-b
09
04
l:_L_ 10-8
10
;IC
Figure 9. Normalized rates vs. conversions for Pittsburgh HVab coal.
version for the Pittsburgh coal. The rate-conversion curve shows a maximum a t lower temperatures only. The variety of rate-conversion curves is due to the fact that different coal/char samples vary greatly from one another with respect to their pore structures and the change of such pore structures with conversion and temperature. T o account for such phenomena, a term a is introduced into the rate equation, which represents the relative available pore surface area and is defined as follows:
a=
(available pore surface are per unit weight at any stage of conversion) (initial available pore surface area per unit weight)
The value of a varies with conversion and temperature. Ignoring the effect of temperature, the change of a with conversion, x , , can be fitted into an empirical equation, up to x , approaching unity, as follows: a = 1 f 100xCY~e-PXC
(0 5 v 5 1)
(5)
In this equation, v and p are the physical parameters characteristic of a given coal or char. The value of v indicates the conversion x , at which the relative available surface area reaches the maximum or minimum value. According to this equation the relative available pore surface area of the particles may increase, decrease, or may show a maximum or minimum as the reaction proceeds, according to the sign (+ or -) used in eq 5. Since a drastic change in surface area and pore size would take place a t the very end approaching complete conversion, the above equation should not be applied beyond x , > 0.9. Therefore, the rate of disappearance of char due to COz reaction, for chemical reaction control, may be expressed as dx = akCAn(l - x,) dt In eq 6, rate is expressed to be proportional to the partial pressure of COa raised to the power n,where 0 5 n 5: 1. The
10-7
C02 CONCENTRATION, 10-6 10-5 rndrlcm3 164
Figure 10. Effect of CO? concentration on the rate of gasification of carbons, a t different temperatures, as observed by previous investigators. InvesCurve tiga- Temp, no. tors "C
Type of carbon and size
Dilu- Reent marks
Coconut charcoal, Ar -30+40 mesh Ar 2 a 1000 Coconut charcoal, -30+40 mesh Ar 3 a 1300 Electrode graphite, -30+50 mesh Ar 4 a 1200 Electrode graphite, -30+50 mesh Ar g/cm,( porosity 0.34 0.22 0.36 0.53 0.18 0.43
2.3 9.2 9.2 2.4 8.8 -
1.43 1.17 1.54 1.53 1.31 -
0.765 0.809 0.767 0.655 0.864 -
424 172 281 26 18 16
Table V
Sample Pittsburgh HVab Coal Hydrane Char
Total surface area x cm2/g
%total surface area covered by pores bigger than 15 A in radius
Surface area (cm2/g)x covered by pores bigger than about 15 A in radius
ho x lo-"
cm.'/mol s
16
85
14
0.89
18
100
18
1.12
424
6
27
1.88
172
19
33
2.05
281
15
42
2.27
No. 150
IGT Char No. HT155 Hydrane Char No. 49 Syn thane Char No. 122
unchanged (except for the high swelling Pittsburgh seam coal) up to a conversion of about 80%.The highly porous matrix of the solid disintegrates into smaller fractions as the reaction proceeds further. A few typical SEM photographs of the pore structure as a function of conversion (here conversion, x , meaning that due to both pyrolysis and char-CO2 reaction on an ash-free basis) are shown in Figure 14(d-h) for one char sample. Surfaces of three char samples, before reaction, are shown in Figure 14(a-c). Table IV shows the net pore volumes, densities, and porosities of the devolatilized samples as determined by the mercury penetration method on a Micromeritics Model 905 0-50 000 psia porosimeter. The devolatilized samples for these measurements, as well as for the B E T measurements, were prepared in the following way. A small amount of sample (-35+60 mesh) supported on a basket made of 200 mesh stainless steel wire was placed inside the hangdown tube. COz was introduced a t a rate of 2.5 cm3/s.The furnace, preheated to a steady temperature of 1024 "C, was quickly raised to enclose the hangdown tube. Heating was continued for 1.5 min and the furnace was lowered quickly. The sample was cooled to room temperature. The heating rate during the devolatilization was almost identical with that represented by the temperature vs. time curve shown in Figure 2. The average pore diameters determined by this method agree well with those observed by the scanning electron microscope. I t may he possible that the pressure up to 50 000 psia, as used in this determination, would break open some closed pores. However, the pore volume and pore size distribution curves for all six samples show that more than 75% of the pore volumes have been penetrated by pressures up to 100 psia and more than 90% by pressures up to 1000 psia. Pore breakage that might happen a t much larger pressures therefore may not cause as serious an error as might be expected.
20
30
LO RADIUS,
50
60
70
i Figure 15. Cumulative surface area vs. micropore radius: 0 ,Synthane Char No. 122; A , Hydrane Char No. 49; 0 ,IGT Char No. HT155; A , Hydrane Char No. 150; X, Pittsburgh HVab coal. MICROPORE
Table IV also shows total pore surface areas of the devolatilized samples as determined by B E T nitrogen adsorption method. Computation of micropore size (
