132
Energy & Fuels 1988,2, 132-136
Electrophoretic Mobility Distribution in Aqueous Dispersions of Bituminous Coal and Residual Hydrocarbon Materials? R. E. Marganskit and R. L. Rowell* Department of Chemistry, University of Massachusetts, Amherst, Massachusetts 01003 Received July 29, 1987. Revised Manuscript Received October 5, 1987
A new automated electrokinetics analyzer has been used to measure the dependence of the electrophoretic mobility on pH for three bituminous coal samples and a friable unconverted vacuum bottoms materials (uvb). The point of zero mobility (pzm) varied from pH 6.0 to pH 6.4 for the coals and was pH 6.8 for the uvb. The specific nature of each sample was shown by differences in the dependence of the average mobility on pH and even more pronounced in the variation of the distribution of electrophoretic mobility on pH. The results are discussed and compared with other electrophoretic data on bituminous coals, suggesting that the pzm of unoxidized bituminous vitrain is very close to pH 6.0. Introduction The effect of mineral content on the electrokinetics of coal is very pronounced and contributes to a more hydrophilic surface that contains a substantial quantity of bound water. Mineral composition is known to vary markedly from coal to coal as does water content. Dewatering is an important aspect of coal preparati0n.l In addition, there is evidence that the native mineral residue in coal plays an important catalytic role in direct hydroliquifaction. All of the above can be related to the electrophoretic behavior of the coal particles, which in turn depends on the surface functional groups present. Even though intrinsic heterogeneity in coal produces no single and universal structure, enough similarity exists in those functionalities to predict the surface charging mechanisms. Previous explanations for surface charging on coal, based on oxide-like hydration and dissociation, fail to explain the positive charge at low pH, so that the surface functionalities involved must include more than simple oxide groups. These ideas were extended to the surface chemistry of residual hydrocarbon materials, the unconverted vacuum bottoms (uvb) obtained in petroleum refining. The uvb products are an unknown material that may have an asphaltene character, but the test is not conclusive. A significant result of this work was the finding that the electrophoretic properties of the uvb products resemble those of bituminous coal. Coal fiies and uvb producta have not been commercially successful materials for large-scale electric power generation in the past. Recent efforts in coal-oil and coal-water systems, utilizing pulverized coal, have been promising. A similar role may be forseen for the uvb products in slurry form. In this paper, we report the full electrophoretic mobility distribution for several bituminous coals and a friable uvb sample as a function of pH in aqueous media. Discussion of the main structural features that affect electrophoresis is presented. The key parameters include determination Presented a t the Symposium on the Surface Chemistry of Coals, 193rd National Meeting of the American Chemical Society, Denver, CO, April 5-10, 1987. 'Present address: E. I. du Pont de Nemours & Co., Inc., P.O. Box 430, Pass Christian, MS 39571.
0887-0624/88/2502-0132$01.50/0
Table I. Ash Content and Sieve Fractions sample (no.) Island Creek (C26) GM (C8) Bishop (C28) Chyoda uvb (C27)
% ash"
sieve mesh 200 X 325 -200 200 X 325 -200
16.0 4.8 4.7 0.9
" Determined by the University of Massachusetts Microanalysis Laboratory at 1000 "C in oxygen. Table 11. Elemental Analysisaof Chyoda uvb element carbon hydrogen nitrogen sulfur oxygen
wt%
86.5 6.0 1.5 5.3 0.7
Determined by the University of Massachusetts Microanalysis Laboratory.
of heteroatoms, carbon framework, mineral matter, and physical structure, especially pores.
Experimental Section Materials. The bituminous coal samples were kindly supplied by the Atlantic Research Corp., Alexandria, VA, and were studied along with a reference sample obtained from the General Motors Corp. (GM) that had been extensively studied in earlier T h e GM coal was a low ash (5-8%), medium-volatile to highvolatile (35%) Pittsburgh No. 8 seam. All samples have been ground in the dry state and then sieved, as designated in Table I. T h e ash content was determined by the University of Massachusetts Microanalysis Laboratory a t 1000 O C in oxygen and is also reported Table I. The coal was stored in closed containers until used. (1) %port to the APS by the study group on research planning for coal utilization and synthetic fuel production: Rev. Mod. Phvs. 1981.53. . . No. 4, part 11, SI-Sl68. (2) Marlow, B. J.: Rowell, R. L. In Proceedings, Third International Symposium on Coal-Oil Mixture Combustion, 1581;National Technical Information Service: Springfield, VA, 1981; CONF-810498 (Vol. 2), pp 640-650. (3) Rowell, R. L.; Vaaconcellos, S. R.; Saia, R. J.; Farinato, R. S. Ind. Eng. Chem. Process Des. Deu. 1981,20, 283-288. (4) Rowell, R. L.; Vasconcellos, S.R.; Saia, R. J.; Medalia, A. I.; Yarmoska, B. S.; Cohen, R. E. Ind. Eng. Chem. Process Des. Deu. 1981,20, 289-394. ( 5 ) Rowell, R. L. EPRI Final Report, Projects RP-1030 and RP-1455-5; Electric Power Research Institute: Palo Alto, CA, 1981. (6) Kosman, J. J.; Rowell, R. L. Colloids Surf. 1982, 4 , 245-254.
0 1988 American Chemical Society
Energy & Fuels, Vol. 2, No. 2, 1988 133
Electrophoretic Mobility Distribution
k.
, PY\ -4 -2
0
,
,
2
4
..
-4 -2
0
4
2
MOBILITY (um/s)/(V/cm)
Figure 1. Electrophoretic mobility distribution as a function of pH for General Motors coal sample C8.
The Chyoda sample, an unconverted vacuum bottoms residue, was obtained from Japan and was supplied by KSE, Inc., Amherst, MA. The initial condition was 0.5-cm friable black lumps. Samples were ground by hand in a mortar and pestle and then sieved dry to 100-200 mesh. An elemental analysis was determined at the University MicroanalysisLaboratory and is reported in Table II. The percentages are in quantitative agreement with the elemental analysis generally found in these material^.^ The water was distilled a second time from the in-house supply in a Corning AG-lb all-Pyrex still. Certified ACS grade NaOH and HC1 were obtained from Fisher Chemical Co. Sample Preparation. A stock slurry was made by adding 1 g of powder to 20 mL of twicedistilled water. This was then mixed continuously on a stirring plate at medium speed for 48 h at room temperature (23-25 "C). The Chyoda sample was stirred for 72 h at the same temperature. Clean polyethylene containers (NalgeneLPE) were partly filled with 50 mL of distilled water, and the pH was adjusted to a preliminary value by using NaOH or HCl respectively. Ten drops of the stock coal slurry were then added to the bottles containing the water of known pH, giving a volumetric dilution factor of approximately 1/200. The samples were then stirred on a stirring plate for 3 min at fast speed and then allowed to settle overnight at 30 "C. After equilibration, the pH of the supernatant was taken, the mixture stirred at medium speed for 2 min, and the mobility then measured. Microelectrophoresis. Electrophoretic mobility was measured with a Pen Kem 3000 automated electrokinetics analyzeras The instrument has a silica sample chamber with permanently bonded palladium electrodes. Particles are illuminated with a 2-mW He-Ne laser, and their image is projected onto the surface of a rotating radial grating. Electrophoretic movement causes a frequency shift in the light transmitted through the grating as compared to a reference detector. A fast Fourier transform analyzer computes the frequency spectrum from the individual contributions of the light scattered from many particles in order to obtain a representation of the electrophoretic mobility distribution (emd). Focusing at the stationary layer is under automatic control. Measurementswere made at both front and back stationary layers,under computer control, and the results averaged to ensure maximum accuracy. Ionic Strength. No electrolyte was added so that the results could serve as a model for the interpretation of the behavior of practical coal slurry fuels that are prepared without added electrolyte. Changing the pH does change the added electrolyte but does not change the main ideas of the results reported.
Results We have reduced our presentation to three representative figures, but in every case reproducibility was checked. The Pen Kem 3000 analyzer readily generated reams of (7) Long, R. B. Adv. Chem. Ser. 1981, No. 195, 17-28. (8) The Pen Kem 3000 Electrokinetics Analyzer Reference Manual; Pen Kem: Bedford Hills: NY, 1981.
NOBILITY ( u m / s ) / ( V / c m )
Figure 2. Electrophoretic mobility distribution as a function of
pH for Chyoda uvb sample C27.
+2 -
0 -
-2
-
-4
L
I
-4
I -2
I
I
I
0
+2
+4
APH
Figure 3. Average electrophoretic mobility as a function of pH
shift from the isoelectric point for unconverted vacuum bottoms (uvb) and several coals. Samples (and isoelectric points) are as follows: solid line, General Motors coal C8 (6.0); dashed line, Bishop coal C28 (6.4);dotted line, Island Creek coal C26 (6.2); dashed-dotted lines, Chyoda uvb C27 (6.8). output. In Figures 1and 2 we have presented selected and representative results from individual runs. In order to avoid clutter, the individual data points are not shown in Figure 3, but the curves shown are justified by the data. In Figure 1we show the distribution of the electrophoretic mobility for GM coal C8 at selected pH over the range studied. The pronounced features, multiple peaks, and variation in peak breadth were also observed in the other coal samples. In Figure 2 we show the dependence of the electrophoretic mobility distribution on pH for the Chyoda uvb C27 sample. The general features, multiple peaks, and variation in peak breadth are shown by the Chyoda sample as well. The average mobility as computed by the Pen Kem 3000 analyzer is shown as a function of pH shift from the isoelectric point (iep) for all four samples in Figure 3. The ieps for the coals were 6.0, 6.2, and 6.4, somewhat lower than 6.8, the iep of the Chyoda uvb sample. The curves reflect the specific nature of each sample within a similar overall envelope and a trend that shows extrema in mobility at low and high pH.
Discussion Multimodal Distribution. GM coal had a multimodal distribution, shown in Figure 1, which suggested the presence of a major second component such as a mineral oxide. The effect of polydispersity in size, although a complicating factor, cannot explain the wide differences
Marganski and Rowell
134 Energy & Fuels, Vol. 2, No. 2, 1988
in electrophoretic mobility of the individual peaks. For example, at a pH of 6.65,the emd peaks occurred at mobilities of -2.2, -0.9, -0.4, and +0.1 (pm/s)/(V/cm), respectively. The differences observed here are far larger than the 50% mobility variation expected between the Huckel and Smoluchowski limits at a given {potential. In addition, since there is a continuous distribution in size as verified in photomicrographs, separate peaks would not be expected to occur. It is interesting to note that the individual peaks appeared to be reproducible over a wide pH range, and individual peaks widened away from the iep in a manner similar to the other coals studied in this work. Statistical variation in sampling plus a small amount of flocculation near the iep could have had some effect on the emd. Amphoteric Behavior. The hydrogen and hydroxyl ions were found to cause amphoteric behavior in all samples measured as shown in Figures 1-3. This observation is in agreement with the conclusions of Fuerstenau et. al.? who observed amphoteric behavior in both native and demineralized coal samples. Amphoteric behavior is caused by dissociation of a single surface
Table 111. Slope of the Mobility at the Isoelectric Point
Bishop (C28) Island Creek (C26) Chyoda uvb (C27) a-alumina
6.4 6.2 6.8 9.2
35.1 27.9 9.1 26.7
"Reference 15. *Mobility slope was multiplied by the Smoluchowski factor 12.9 for 25 "C, to give the { potential slope in mV/pH unit.
(MOH~)+ MOH +H~O (1) where M represents a metal ion or, in the case where there are two distnct functional groups present, one that accepts a proton to become positively charged and one that dissociates and leaves a negative site. The zwitterionic surface is common in biological systems where the two groups responsible are usually the carboxylic and the amino groups. The change in {potential (derivable from the mobility) with pH, d{/dpH, a t the isoelectric point (iep) gives an indication of how closely H+ and OH- can be considered to be potential-determining ions. It is assumed here that the surface potential is close to the potential at the surface of shear, near the iep. Addition of acid or base will alter the chemistry of the solid, creating irreversible effects, which causes deviation from the ideal Nernstian behavior of 59.2 mV/pH unit at 25 OC.l0 Smith"-13 has shown that deviations from the ideal Nernst equation become important when the fraction of sites ionized at the point of zero charge is small. Under these conditions the Nernstian slope is reduced by a fador equal to In (e,/e), where 0 is the fraction of sites of given charge that are ionized at the iep. The iep is then characteristic of an absence of charge rather than an equal number of positively and negatively charged sites. At the opposite extreme, e+ = 8- = 0.5, the logarithmic term vanishes and Nernst's law is obeyed. This is the case for silver iodide particles in aqueous solution where an excess of ions in the solid lattice determines the surface charge. Levine and Smith12and Healy and White14have both shown that oxides were significantly different than perfectly reversible silver halide surfaces. All coals measured in this work had slopes roughly half of the Nemstian value as shown in Table 111. The simple oxide, a-alumina, reported elsewhere,15had a slope similar
to that of the coal samples. The uvb had a much lower slope, indicating that a smaller amount of charge was present at the iep. This suggests that the surface charging on coals used here behaved more like an oxide than a hydrophilic carbon. Fuerstenaug has demonstrated the importance of using low-ash coal in order to avoid measuring an oxide rather than a coal surface. It is interesting to note the significant rise in mobility at the extremes of pH for the uvb sample. This can be interpreted as ionization of heteroatoms at high and low pH values but little or no ionization at the iep. Zwitterionic surfaces such as these were shown to be expected to depart markedly from Nernstian behavior as discussed by Hunter.lG The interpretation of the dependence of the breadth of the distribution on pH has been given in an earlier paper.17 Increased ionization at larger pH shift from the iep produces a wider spectrum in the electrophoretic mobility distribution. Several authors have proposed that the principal charge bearing groups on demineralized coal were oxygen functionalities such as aromatic alcohols (phenolic) and carboxylic a c i d ~ . ~ J If * these groups were alone responsible for surface charge on coal, hydrolysis in water would cause the donation of protons to water, the weaker Brransted acid, creating only negative charge. Amphoteric behavior would not be observed. This is predicted by the theoretical work of Healy and White.14 Attempts to explain the positive surface of coal by the charging mechanismlg C-OH + H+ = C-OHZ' (2) are based on the unsupported hypothesis that the adsorbed proton bears the positive charge. The origin of this concept lies in the attempt to compare the charging mechanism on coal to that of the metal oxides such as A10. Protonation of a metallic hydroxide, MOH, produces a coordinated water molecule that gives an unbalanced and positive formal charge to the metal ion, i.e., M+:OH2. The hydrogen atoms of water do not bear the charge. This is supported by the fact that charge on a hydrated metal ion in solution residues on the ion and not on any protons of the coordinated water molecules.20 Carbon, which is unable to coordinate water molecules, cannot give rise to positive charge in this manner. We propose an alternative mechanism giving a positive charge through hydrogen bonding of an hydronium ion. C-OH + (H30)+ = C-OH*..OH,+ (3) The decrease in mobility giving maxima at both high and
(9)Fuerstenau, D. W.; Rosenbaum, J. M.; Laskowski, J. Colloids Surf. 1983,8,153-174. (10)Arp, P. A. J . Colloid Interface Sci. 1983,96, 80-89. (11)Smith, A. L.In Dispersions of Powders in Liquid+ 3d ed.;Parfitt, G. D., Ed.; Applied Science: Englewood, NJ, 1981;pp 99-148. (12)Levine, S.;Smith, A. L. Discuss. Faraday SOC. 1971,52,290-301. (13)Smith, A. L. J. Colloid Interface Sci. 1976,55,525-530. (14)Healy, T.W.; White, L. R. Adu. Colloid Interface Sci. 1978,9, 303-345. (15)Marganski, R. E.;Rowell, R. L. Polym. Mater. Sci. Erg. 1986,52, 399-403.
(16)Hunter, R. J. Zeta Potential in Colloid Science, Principles and Application; Academic: New York, 1981. (17)Marganski, R. E.; Rowell, R. L. J. Dispersion Sci. Technol. 1983, 4,415-424. (18)Tingley, G.L.;Morrey, J. A. "Coal Structure and Reactivity"; Battelle Energy Report; Battelle Pacific Northwest Laboratories: Richland, WA, 1973. (19)Kelebek. S.: Salman. T.: Smith. C. W. Can. Metall. B. 1982.21, 205-209. (20)Yoon, R. H.;Salman, T.; Donnay, G. J. Colloid Interface Sci. 1979,70,483-493.
H+
F-Mo-
Electrophoretic Mobility Distribution low pH follows as a consequence of increased screening at high ionic strength. In addition, most coals have mineral matter incorporated into their structures in various degrees. These are classified into three catagories: (1)admixed minerals of larger size; (2) incorporated minerals as a result of metamorphic processes; (3) inherent minerals uniformly distributed within the organic matrix.21y22 Demineralization procedures probably do not remove all of the mineral matter. For example, treated coal had as much as 10% of the ash content of untreated samples in Fuentenau’s work? Table I shows the ash content of untreated coals used in this work. The degree of entrapment determines how easily coal can be cleaned and is of current interest in beneficiation. The extensive pore structure of coal will contain embedded water with dissolved salts, minerals, humic acids, and organic matter. The release and subsequent adsorption of various species from these pores is likely to be a function of pH. One of the techniques used to determine the moisture retention and pore structure is to observe the behavior of metallic ion adsorption on Such selectivity will allow the transport of only certain materials into and out of those pores, analogous to a molecular sieve. Evidence for this behavior is shown in Figure 3, where the average mobility for several samples is plotted as the function of pH units away from the iep. Two features can be noted: first, the maximum negative mobilities are all similar in value, and second, the maximum positive mobilities have a larger amount of variation. We interpret the above as follows: at high pH, the carboxylic and phenolic groups on the carbon skeleton of coal are mainly responsible for the negative charge. The organic framework is basically similar for the fully oxidized surfaces in this work and will result in similar electrophoretic mobility. This interpretation is supported by results obtained by sequential elution by specific solvents chromatography, SESC.24 Findings indicated that although different coals varied greatly in overall structure, many of the same chemical groups were always present. A t low pH, positive functionalities, metallic oxides, and adsorption of metal ions leached from pores all contribute to the observed mobility. The larger amount of variability expected in these processes will be reflected in a greater difference in the mobility that is observed from sample to sample. This is in agreement with the results of Wen and Sun,% who showed that at lower pH, the l potential of oxidized coal was controlled by the presence of electrolytes such as Fez+and A13+, both of which form hydroxides at pH values below 8. Above this pH, the metal ions behaved much like an indifferent electrolyte with little or no effect on the { potential. In addition, they also suggested that variations due to oxidation, shown to be small above pH 7 as compared to larger variation below pH 7, might be explainable due to the solubility of humic acids on the coal surface. I t is important to note that the measurements will be very sensitive to the solids concentration of the slurry as discussed below. Previous workers have not specified this important parameter. In addition, the aging effects will be most pronounced in the high-concentration region. Conditioning and aging procedures are generally not re(21)Strechlow, R. A.; Harris,L. A.; Yust, C. S. Fuel 1978,57,185-186. (22)Fuller, E. L.,Jr. J. Colloid Interface Sci. 1982,89, 309-327. (23)Mahajan, 0.P.In Coal Structure; Myers, R. M., Ed.; Academic: NY,1982;pp 51-86. (24)Chem. Eng. News. 1982,60 (August 9),17-26. (25)Wen, W.W.;Sun,S. C. Trans. SOC.Min. Eng. AIME 1977,262, 174-180.
Energy & Fuels, Vol. 2, No. 2, 1988 135
ported in the literature either. In this work, cme was taken to properly condition a dilute stock solution for 3 days followed by further dilution to a very low particle concentration at a given pH in order to avoid any high-concentration anomalies. By avoiding excessive leaching of inorganic ions, we are seeing the electrophoretic mobility of the coal surface and not a layer of inorganic hydrolysis precipitate. This could account for the generally high iep observed in this work as compared to other values in the l i t e r a t ~ r e . ~Clearly, * ~ ~ more research is required on this point. The composition of the uvb sample, while not completely understood, appears to consist of condensed polynuclear ring systems that have alkyl side chains containing heteroatoms such as nitrogen, oxygen, and sulfur.’ Limited data suggests that oxygen is present as non-hydrogenbonded phenolic groups. The concentration of mineral oxide in the uvb material is very low; it was found to be 0.9% for the sample reported in this work. The observed amphoteric behavior may then be explainable in terms of zwitterionic surface groups. If the uvb materials are really primarily asphaltenes, then this is in agreement with the conclusions of Cratin, who has suggested that the amphoteric nature of the asphaltenes might be similar to that of amino acids.26 In summary, we have suggested the mechanism of hydrogen bonding as a major factor controlling the surface charge and explaining the general similarity for the variety of coals shown in Figure 3. However, the smaller specific differences characteristic of each coal, must arise from real property differences as suggested above. We have illustrated the mechanism of hydrogen bonding with C-OH but we cannot rule out other Lewis base sites such as C=O, =N-, and -0-, which would also hydrogen bond to hydronium ion. Ash content certainly is also responsible for part of the contribution to the differences exhibited by each material as shown by comparing Table I and Figure 3. In such a comparison the Chyoda uvb might be regarded as a very low-ash coal. Since our work was directed at understanding the properties of commercial coals, “deashing” was not included in the study. Contemporary Comparisons. Marlow and Rowe1lZ7 have independently measured the iep of coal sample GM C8 by electrophoresis and acoustophoresis and reported a pHiepof 5.8, which is within the experimental error of comparison of our value of 6.0 reported in Table 111. In Laskowski’s reviewz8of the origin of charge at the coal/water interface, he reports the iep for fresh fines from crushed coal ranging from 3.8 to 6.6, which is consistent with our result. If the same coal samples were not crushed but were sieved to prepare fines then Laskowski’s results showed no iep but an everywhere negative { potential ranging from -25 to -35 mV at pH 3.4 to around -60 mV at pH 11.5. He attributed negative sites on the coal surface to carboxyl groups, phenolic groups, and inorganic impurities such as silica. Fuerstenau et al.29reported on hand-picked samples of vitrain bands from a West Virginia bituminous seam supplied by the Coal Research Bureau, West Virginia University. The ash content was 3.33%, which is close to (26)Cratin, P.D.In Chemistry and Physics of Interfaces-II; Ross, S., Ed.; American Chemical Society: Washington, DC, 1971,pp 35-47. (27)Marlow, B.J.;Rowell, R. L. Prepr. P a p - A m . Chem. SOC.,Diu. Fuel Chem. 1987,32, 357-366. (28)Laskowski, J. S.Prepr. P a p - A m . Chem. SOC.,Diu. Fuel Chem. 1987,32, 367-387. (29)Fuerstenau, D.W.; Rosenbaum, J. M.; You, Y. S. Prepr. P a p Am. Chem. SOC.,Diu. Fuel Chem. 1987, 32, 378-385.
136
Energy & Fuels 1988,2, 136-141
that of our bituminous sample GM(C8), 4.8%. Their coal was ground to -400 mesh and suspended in water for 9 h while being agitated so that the handling was similar to our work except for a shorter suspension time. The iep for unoxidized vitrain shown in their Figure 7 was 5.8, in quantitative agreement with our work for sample GM(C8). Moreover, their slope (&"E/~H)iep appears to be identical with ours within experimental error. They also reported an iep of the same vitrain at pH 4.5 after 192 h of oxidation in pure oxygen at 80 "C. Casassa and Toor30 reported the iep of high Kaolin Splash Dam (17% ash) and lower Kittanning (8.7% ash) samples along with the iep of beneficiated samples, Splash Dam (1.6% ash) and lower Kittanning (2.3%). Both were bituminous coals that can be compared with our work. For the high-ash samples the iep was near pH 3, where kaolin has been reported to have a zero point of charge. The beneficiated samples had a positive mobility up to about pH 6, which they suggested "appeared to be the pattern of mobility associated with an unoxidized mainly carbonaceous surface." Casassa and Toor30cited a second paper by Wen and Sun31that included a study of unoxidized and oxidized vitrain. The oxidized vitrain showed no iep but the potential was always negative and ranged from -20 mV at pH 3 to -60 mV at pH 11.5. This is the same characteristic as reported by Laskowski for sieved samples, which might be expected to contain fines in a highly oxidized state. However, Wen and Sun also report that "unoxidized" vitrain had an iep at pH 4.5. The oxidized vitrain showed
an iep at pH 4.3 after treatment with M benzidene (4,4'-NH2C6H4C6H4NH2), which further increased to 6.2 on treatment with M benzidene. In order to reconcile their work with our results, Fuerstenau's itr rain:^ and the Splash Dam and Kittanning bituminous coals of Casassa and T o o P (the latter three of which suggest an iep at pH 5.8-61, we must conclude that Wen and Sun's "unoxidized"vitrain3' had acquired an acidic character perhaps through handling or sample history and that the treatment with benzidine removed the weakly associated acidic character. Marlow and Rowel12' have found the iep of GM sample C8 at pH 4.4 for a volume fraction of 0.4, a pronounced acid shift from the acoustophoretic value of 5.8 at a volume fraction of 0.04. We must consider that result as well in reconciliation of our work with the Laskowski results. Our discussion has been in terms of the isoelectric point (iep), which is the pH of zero net surface charge. What is measurable by electrophoresis is the point of zero mobility (pzm), which is zero net charge at the average surface of shear surrounding the moving particle. In the absence of specific adsorption, the iep and pzm are identical. The differences in the various samples compared may have been a consequence of differences in specific adsorption arising from handling and sample history. We prefer the term pzm because it is the measured quantity rather than the widely used term iep or the term pzr (point of { reversal) because the t potential is a derived quantity whose calculation depends on the assumption of a model of the charge distribution in the double layer.
(30)Casassa, E.Z.;Toor, E. W. Prepr. Pap.-Am. Chem. Soc., Diu. Fuel Chem. 1987,32,386-394. (31)Wen, W. W.; Sun, S. C. Sep. Sci. Technol. 1981,16,1491-1521.
Acknowledgment. This work was supported by a grant from the Atlantic Research Corp., Alexandria, VA. Technical assistance and instrument support was provided by Pen Kem, Inc., Bedford Hills, NY.
TPD Study on H20-Gasified and 02-Chemisorbed Coal Chars? Takashi Kyotani, Zhan-Guo Zhang, Shigetoshi Hayashi, and Akira Tomita* Chemical Research Institute of Non-Aqueous Solutions, Tohoku University, Sendai 980, Japan Received August 4, 1987. Revised Manuscript Received November 25, 1987 Temperature-programmed desorption (TPD) profiles of H20-gasifiedand 02-chemisorbedcoal chars were investigated with five coals of different ranks. Coals were devolatilized in Nz and then gasified in HzO at 1100 K. Initially, TPD of the partially gasified char was determined to 1100 K in vacuo. Later the TPD experiment was repeated after the 02-chemisorptionat 420 K on the above heat-treated sample. The sharp peaks in TPD patterns, corresponding to HzO, COP,and CO evolution, all resulted from the presence of mineral matter, since no sharp peak was observed in the TPD patterns of the demineralized coal chars. The gas evolution pattern from the demineralized char was almost independent of coal type, and all of the demineralized coal chars exhibited almost the same gasification reactivities. The relationship between the gasification reactivity and the total amount of C02 and CO evolution during TPD of gasified chars and 02-chemisorbedchars is discussed. Introduction It is generally accepted that the first step of the steam gasification reaction is the chemisorption of steam on a carbon substrate to form surface oxygen complexes and
that such complexes function as intermediates in the gasification reaction.'p2 In the reaction of carbon with oxygen, some investigators carried out oxygen chemisorption studies to estimate the amount of surface oxygen com-
Presented a t the Symposium on the Surface Chemistry of Coals, 193rd National Meeting of the American Chemical Society, Denver CO, April 5-10, 1987.
(1) Walker, P. L., Jr.; Rusinko, F., Jr.; Austin, L. G.; Adu. Catal. 1959, 11, 133. (2)Ergun, S.; Mentser, M. Chem. Phys. Carbon 1965,1, 203.
088~-0S24/88/2502-0l36$01.50/0 0 1988 American Chemical Society
