Thermal Analysis of Oil Shales

THERMAL ANALYSIS OF OIL SHALES EDWARD WERNER COOK To obtain oil from oil shale, it is necessary to heat the shale. To state this is to carry shale to Green River, but it does focus heat and heat transfer as central to the technology of oil shale development. For our present purposes, we will only be concerned with heat requirements. Although a knowledge of heat requirements for retorting oil shale is important, surprisingly few studies have been reported.1 The most extensive were by the Bureau of Mines,2 3 which were invaluable during preliminary studies. However, much of the work was empirical, so few generalizations could be safely made. It was believed that a rational study would enable one to predict with some confidence heat requirements under a variety of circumstances. First, we shall define the thermal process that will be considered. Assume Mahogany Zone Green River shale at 77°F (298 K ) will be heated to and retorted at 891°F (750 K) under anaerobic conditions. Then we may ask several questions: 1. What is the total heat (or enthalpy) required to: ( 1) heat this shale from 77°F ( 298 K) to 891°F ( 750 K), (2) pyrolyze the kerogen, and to vaporize the products? 2. What are the important parameters influencing any variations in the total heat required? 3. What is the heat content (enthalpy) of the retorted shale at 891°F? Fortunately, the mineral distribution in Green River shale is quite consistent through the Mahogany Zone. In fact, this consistency bas been employed by Smith4 as a basis for predicting the organic richness of shale by measuring the shale density. That is, variation in shale density depends only on the relative amounts of organic and mineral matter, the density of each being invariant. Thus, one may reasonably expect to calculate and predict the thermal requirements for retorting oil shale from the components. Although these calculations cannot be definitive because assumptions must be made, the results should be confirmatory and reasonably predict variations in heat reRocky Flats Research Center, The Oil Shale Corporation, Golden, Colorado 80401 133 134 QUARTERLY OF THE COLORADO SCHOOL OF MINES quirements to be expected in a commercial operation. The distribution of minerals in Mahogany Zone shale is given in table l. The most important observation is that the minerals are predominately carbonate and silicate (Illite is a silica clay) ; so one may expect some decomposition will also occur. This will be discussed later. TABLE 1.-Mineral composition of Mahogany Zone Green River shale' Mineral Formula Dolomite Calcite Quam Illite Low-Albite Adularia Pyrite Analcime CaMg(CO3) 2 CaCO3 SiO2 NaAlSi3O8 KAlSi3O8 FeS2 NaAlSi2O6 ·H2O Percent 32 16 15 19 10 6 1 'From: Ref. 4. In table 2, the mineral enthalpies are given.* They have been calculated over the temperature range of interest: 77°F to 891°F, as the latter is within the range for optimum retorting temperatures. A search of the literature did not reveal any values for illite or analcime; 105 cal/g was assumed for these minerals as reasonable for silicates. The closeness of values is important, for it indicates that variation in relative abundance would have only a small effect on the mean value (except for pyrite, which is a minor component). We now must consider which of these minerals will decompose and to what extent under retorting conditions. Immediately, we may eliminate quartz, pyrite and the feldspars: low-albite and adularia. This leaves dolomite, calcite, illite and analcime to consider. In Green River shale, dolomite is actually ankerite, that is, there is about 15 percent Fe substitution for Mg.7 The net result is to lower the temperature of decomposi· tion somewhat.5 It is difficult to predict the temperature range or the extent of carbonate decomposition during shale retorting as these decomposition; •.More than most fields, thermodynamics is a babble of units with three different energy units in common use and an equal number of temperature scales. If the result is confusing, the author apologizes; no attempt has been made to proselytize, but the metric system and K have been retained through calculations to maintain consistency with the published literature. Final results have been converted to British units. 135 THERMAL ANALYSIS OF OIL SHALE Are dependent on the partial pressure of carbon dioxide in the retort atnosphere. Experimentally, we have found that both in the Fischer-Schrader determination6 and the TOSCO II Process, the equivalent of about 20 percent of ankerite decomposes (table 3) . TABLE 2.-Mineral enthalpies Mineral ( H750- H298) 1 (cal/g) Percent Dolomite Calcite Quartz Illite Low-albite Adularia Pyrite Analcime 119.l 112.6 107.8 (105) 2 107.5 102.0 64.47 (105)2 32 16 15 19 10 6 1 1 Interpolated from Robie and Waldbaum, U. S. Geol. Survey Bull. 1259, 1968. Assumed values. TABLE 3.-Mineral heats of decomposition H }Mineral Ankerite ( Dolomite) Illite Analcime K cal/mol) 16.41 cal/g 89.0  64 (50) Range (°C) 680-780 1 400-6952  3003 1 2 In CO2 atmosphere, significantly lower in inert atmosphere. Ref. 5. See Ref . 13. Loss of water: J. W. Smith, Private communication. Heat assumed. 3 Fortunately, calcite decomposition occurs above retorting temperatures and need not be considered. Both illite and analcime lose water within retorting temperatures; complete decomposition is assumed. The heat required for analcime decomposition was not known: so 50 cal/g was assumed. However, analcime is a minor component and a large error here would be relatively insignificant. In table 4, the mineral contributions to the heat of retorting are calculated. Note that the total heats of decomposition amount to about 17 percent of the sensible heats required for the minerals. We may conclude that 232.7 Btu/lb 136 QUARTERLY OF THE COLORADO SCHOOL OF MINES are required to heat the mineral components of Mahogany Zone shale from 77°F to 891°F. TABLE Mineral Ankerite Illite Analcime Percent 32 16 1 4.-Mineral heat of retorting H % Decomposing Total (cal/g) 20 100 100 5.8 12.2 0.5 89.0 64 50 Mineral ( H750 – H298) 18.5 110.8 129.3cal/g The organic matrix of oil shale is complex. Nevertheless, it is consistent and may be considered to be of two general fractions: bitumen and kerogen. Bitumen is the lower molecular weight, soluble fraction. Kerogen is higher molecular weight, insoluble and the predominate fraction. For our present purposes we shall ignore these differences and consider all the organic material as kerogen. Little work has been done on the heat content of kerogen. For example, Cane has calculated the specific heat of Glen Davis (Australia) kerogen from the atomic heats.10 Because of its importance in these calculations, we determined the enthalpy of kerogen directly from a sample of kerogen concentrate kindly supplied by J. W. Smith of the Bureau of Mines in Laramie, Wyoming. This kerogen concentrate had been found to have 91.49 percent organic matter. Enthalpies were determined over short temperature intervals and the values obtained were used to solve Equation (1) by the method of least squares.  HT  H 298   a  b / T  cT  dT 2 (1) where H is cal/g and T is K. This particular equation was chosen in order to obtain specific heats directly from Equation (2) which was found most satisfactory for this type of material. cp  d ( HT  H 298 ) d  c  T  bT 2 dT 2 (2) After solution of the equation the residual mineral contribution was sub· tracted out, and the kerogen value was extrapolated to the temperature range of interest (table 5) . 137 THERMAL ANALYSIS OF OIL SHALE TABLE 5 . Kerogen enthalpy and heat of decomposition (H750 –H298) Hdec = 233 cal/g = 60 cal/g 293 cal/g 527 Btu/lb required to heat kerogen from 77°F to 891°F. It is rather difficult to determine accurately the heat of decomposition of kerogen. However, from Cane's work on determining the heat requirements of Gray-King assay of torbanite11 we can calculate that 60 cal/g will be required. Interestingly, this is about what one would calculate for the heat of vaporization from Trouton's Rule; it is also similar to Gräf’s calculations for heavy 8, 9 Braunkohle oil The total heat required then, to heat and retort kerogen from 77°F to 89l°F is 527 Btu/lb (table 5) . It is probably now becoming clear that the major variation in heat requirements for retorting shale is the richness of the shale. Recalling that the • mineral requirement is 233 Btu/lb we see that the organic requirements are over twice as large. Taking advantage of the uniformity of Mahogany Zone Green River shale, Smith has developed a relationship between Fischer Assay of shale and organic content of the shale :12 (organic matter, wt% shale) = 0.580 (oil richness, gal/ton) (3) From this, we derive a formula relating gross heat of retorting lo shale richness: Hretort = 232.7 + 1.708 X (4) where Hretort is gross heat of retorting in Btu/lb and x is shale richness in gal/ton. In table 6, several values have been tabulated for different shale richness. Notice that heat requirements increase about 17 percent in going from IO gal/ton to 40 gal/ton. The actual determination of heat of retorting is difficult. On a plant scale many days of operation and extensive instrument calibrations are required. In the laboratory, evolved heavy oils tend to condense within the calorimetric apparatus to make enthalpic determinations even more difficult. Nevertheless, the agreement between the values determined by the plant and calculated here is excellent. The calculated values are within .10 Btu/lb of actual values, or about 2 - 3 percent. 138 QUARTERLY OF T H E C O L O R A D O SCHOOL O F M I N E S TABLE 6.-Heat o f retorting Gross heat of retorting: (Btu/lb) = 232.7 + 1.708 x Where x is oil richness in gal/ton Shale Richness (gal/ton) Heat Required (Btu/lb) 10 249.8 266.9 283.9 301.0 20 30 40 Thus, it should he reemphasized that, even though the results agree very well, several assumptions were made; so this equation should not be considered as ex cathedra but only as indicative of what the major heat variability will be in a commercial operation. Obviously, the mineral composition can vary considerably without affect· ing the overall heat requirements, because enthalpies of most of these minerals are similar. Important exceptions are those minerals which will decompose between 77°F and 891°F. Carbonates are the most likely minerals to be significant in this instance. Of these, nahcolite and dawsonite are most widely regarded as occurring with Green River shale and which we have 111>1 already considered. Therefore, their heats of decomposition were determined (table 7 ). TABLE Mineral 7.-Decomposition heats of other shale minerals Formula Decomposition Range, °C Hd Cal/ g Btu/lb · Nahcolite Dawsonite NaHC0 3 NaAl (OH)2CO3 90-200 325-400 177  215  387 318 It should also be remembered that any process which heats shale significantly beyond 891°F will also decompose more ankerite and, if sufficiently high temperatures are reached, calcite decomposition will begin. These additional heats would have to be considered. After pyrolysis, one has a retorted shale at 891°F. Since heat may be recovered from this spent shale it is of interest to determine its heat content. As the mineral composition of Mahogany Zone shale is reasonably consistent, as all shales contain approximately the same organic residues (5 - 6 THERMAL ANALYSIS OF OIL SHALE 139 percent) regardless of original richness,* spent shale enthalpy should be reasonably constant for Mahogany Zone shale. Spent shale from a Fischer determination was employed as a representative sample. Enthalpies were determined throughout the range of interest. The values were filled by least squares to Equation 1. The calculated smoothed values are listed in table 8. From this we calculate that 228 Btu/lb is evolved on cooling spent shale from 891°F to 77°F. TABLE T ( K) 8.- Spent shale enthalpy HT – H298 (cal/g) 0 20.28 41.68 67.22 108.7 126.9 298 400 500 600 700 750 Standard error of estimate 0.5 cal/g 228 Btu/lb released on cooling spent shale from 891 °F to 77°F. REFERENCES 1 a. McKee, R. H., and Lyder, E. E., 1921, The thermal d e c o mp o s it io n of oil shales. II. Determination of the heat of reaction i n v o l v e d i n t h e ir thermal decomposition: Ind. Eng. Chem. Jour., v. 13, p. 678. b. Zelenin, N. I., Faynherg, V. S., and Chernysheva, K. Il., 1968, Khimiya i Tekhologiya Slaitsevoy Smoly: Izd-vo 'Khimiya', Leningrad, Glava 5. 2. Shaw, R. J., 1947, Specific heat of Colorado oil shales: U.S. Bur. Mines Rept. Inv.1151. 3. Sohns, H. W and others, 1951, Heat requirements for retorting oil shale: Ind. Eng. Chem. Jour., v. 43, p. 33. 4. Smith, J. W., 1969, Theoretical relationship between density and oil yield for oil shales: U.S. Bur. Mines Rcpt. Inv. 7248. 5. Reddick, K. L., 1968, Heats of reaction for carbonate mineral decomposition in Analytical calorimetry, Porter, R . S., a n d Johnson, J. F., eds.: New York, Plenum Press, p. 297. 6. Goodfellow, L., Haberman, C. E., and Atwood, M. T., 1968, Modified F i s c h e r assay equipment, procedures and product balance determination: Am. Chem. Soc., Div. Petrol. Chem., Prepr. 13 (2), F86. *This is only approximately true, but subtle differences are of no concern here. Obviously, we are only discussing spent shale from an anaerobic process; ash from gas combustion p r o c e s s e s ha ve no organic material a n d are specifically excluded here. 140 QUARTERLY OF THE COLORADO SCHOOL OF MINES 7. Smith, J. W. and Robb, W. A., 1966, Ankerite in the Green River Formation’s Mahogany Zone: Jour. Sed. Petrology, v. 36, p, 486. B. Grife, E., 1910, The heat of vaporization of petroleum oils: Petroleum, v. 5, p. 069. 9. c.f., Valdek, R., and others, 1961, O teplote razlozheniya organicheskogo Veschestva Estonskikh goryuchikh slantsev: Eesti NSV Tead. Akad . Toim., Funs. Mat. v.10, p. 158. 10. Cane, R. F., 1943, Some thermochemical properties of the torbanite of the Glen Davis deposit: Royal Soc. New South Wales Jour. and Proc., v. 76, p. 190. 11. Cane, R. F., 1948, The chemistry of the pyrolysis of torbanite: Australian Chem. Inst., ]our. Proc., v. 62. 12. Smith, J. W., 1966, Conversion constants for Mahogany Zone oil shale: Am. Assoc. Petroleum Geologists Bull, v. 50, p. 167. 13. a. Barshad, in Smothers, W. ]., and Chiang, Y., 1966, Handbook of differential thermal analysis: New York, Chemical Publ. Co., p. 134. b. Taboadela, M. M., and Ferrandis, V. A., 1957, in The differential thermal investigation of clays. MacKenzie, R. C., ed.: London, Mineralogical Soc., chap. VI. c. Sudo, C. T., and others, 1965, Energy changes for endothermic reaction; occurring in dehydration processes of some clay minerals: Internal.Conf. Thermal Analysis, First, Aberdeen, 208 p.