COKE FROM ILLINOIS COAL Temperature Conditions in Sole-Flue Ovens GILBERT THIESSEN' Illinois State Geological Survey, Urbana, 111.
B
sure against the side walls, if serious damage to the oven is to be avoided; the coke must shrink away from the walk in the final stages of coking so that the charge may be pushed from the oven without damaging the walls; and, finally, the coke must have a coherent structure so t h a t it will move as a single block under the influence of the pusher ram, and not crush and jam against the This last requirement offers the most serious difficulty encountered when coking
Y-PRODUCT coke is conventionally made from coal in ovens that are tal1 and narrow in cross section. Such ovens may be 11 to 22 inches wide, 6 to 13.5 feet high, and 30 to 40 feet long. For satisfactory operation the coal charge must have certain definite properties. The charge must not swell excessively or develop any great pres1 Present address, Koppere Company, Tar and Chemical Division, Teohnical Dept., P. 0. Box 7372, Oakland Station, Pittsburgh, Pa.
506
R
MAY, 1937
INDUSTRIAL AND ENGINEERING CHEMISTRY
coals from the interior basin in ovens of this type. Because of their high volatile matter and moisture content, these coals have a tendency to form slender, fingery pieces during the final devolatilization and shrinkage period. These fingery pieces tend to cause the charge of coke to crumble in front of the pusher ram and t o jam in the oven, making it necessary to maintain very close control over the coking operation. The difficulties are further aggravated by the fact that these coals rapidly lose their coking properties upon storage.
Coke has been commercially produced from Illinois fine coal in an oven in which the charge lies in a wide, shallow layer and is heated from the bottom. The temperature conditions in the charge have been determined. The plastic zone travels at a rate of 0.78 inch per hour in the lower 7 inches of the charge. In this type of oven a coke can be made, the lower two-thirds of which has the properties of a high-temperature, lowvolatile coke and the upper third has more of the characteristics of a low-temperature coke. Such coke can be easily
To overcome these' difficulties and to produce a coke more suited to domestic use, a type of oven has been developed in which the coal is heated from below in a wide, flat layer. The charge is more easily removed from the oven, investment is claimed to be lower, and a product grading from low-volatile, high-temperature coke a t the bottom to high-volatile, lowtemperature coke a t the top can be produced if desired. At present there is an installation of three batteries, with a total of twenty-six ovens, of this basic type in operation in Franklin County, Ill. These ovens are known as Knowles sole-flue ovens and are regenerative (9, 14). Historically, this type of oven is derived from the recuperative beehive oven in which the products of combustion of the volatile part of the coal are brought under the oven floor in a flue system (3, 8, 9, 16). The sole-flue oven has been developed both as a recuperative (26) and as a regenerative (12)by-product recovery oven. The ovens which have found commercial use have been regenerative although some nonheat-recovery ovens of this type are in use in the petroleum industry for the production of a dense, low-volatile petroleum coke from petroleum residues (5,IS,24). The ovens in Franklin County, Ill., are the first commercial installation of this type in the United States for the coking of coal alone. Because of the unusual features involved and because of the importance to the coal industry of tho state of the possibility of producing coke commercially from Illinois coal screenings in equipment for which a comparatively low investment cost is claimed, it became of interest to study the temperature conditions throughout a coal charge during a n entire coking cycle. The importance of a rapid rate of coking to the production of good coke from poorly caking coals of high volatile content is sometimes men-
507
tioned in the coke literature (17, 19). Temperature conditions inside coal charges being coked in slot-type by-product ovens and in vertical continuous gas retorts have been studied by various investigators and are available in the literature for purposes of comparison (6, 11, 15, 23). Through the courtesy of the Radiant Fuel Corporation, owner of the Knowles ovens a t West Frankfort, it was possible t o study the temperature conditions inside a charge being coked in this type of oven.
pushed from the oven. For domestic use such coke has the advantage of ease of ignition. If desired, the coking process can be carried to completion to yield a product which has the characteristics of high-temperature coke throughout. The gas produced in these ovens is similar to normal coke-oven gas with the exception of lower hydrogen and methane contents. Analyses of coals coked, of coke and gas produced, and of horizontal sections of coke at various intervals from the oven floor are presented.
At the time these tests were made, the plant consisted of one battery of ten Knowles sole-flue ovens with the necessary auxiliary equipment for coal, coke, and by-product handling. Figure 1 shows views of the coke side of the ovens. Figure 2 shows construction details of the ovens. The ovens are charged with 5 tons of No. 5 carbon (-"le inch) from Illinois No. 6 coal screenings, making a layer 10 to 12 inches thick. The normal coking time was 8 hours. Flue temperatures are about 1360" C. (2480" F.) with the floor temperature a t the completion of a coking cycle a t about 1100" C. (2012" F.). This installation has been rather fully described in the literature (2, 14). The objectivee of these tests were to determine the temperature conditions inside a charge of Illinois coal being coked in the sole-flue type of coke oven, including a determination of the rate of travel of the plastic zone, the temperature gradients, and the final temperatures achieved in the charge a t selected points; t o measure regenerator and flue temperatures; to determine the volatile matter gradient from bottom to top in the finished coke; and to determine average gas composition.
Test Equipment Temperatures inside the oven were measured at seven points
by means of seven chromel-alumel thermocouples and two potentiometer pyrometers. These thermocouples were 12 feet long and were made of No. 22 wire, insulated with two-hole sillimanite insulators and enclosed in quarter-inch steel pipes welded shut a t the end for protection. These protecting pipes were welded to steel strips in such a way that the pipes entered the oven under the door in a horizontal row and then curved so that the ends of
the couples were located in a vertical row directly under the out-
508
SECTION A - A
FIGURE 2. CONSTRUCTIOW DETAILSOF KNOWLESSOLE-FLUE OVENS
side charging hole line, 6.5 feet from the inside of the oven door and along the center line of the oven. The pipes were so bent that the ends extended about 1 foot in a horizontal line from the vertical spacing and sup orting bar in order to minimize as much as poseiile errors caused by conduction of heat along the steel pipes. The arrangement of the thermocouples is shown in Figure 3. The ends of the couples were so laced that the lowest one was on the oven 8oor; the next couples were I, 3 , 5 , 7 , and 9 inches from the floor, respectively; and the last one was 10.5 inches from the floor. mesumablv iust
P L A N VIEW
L 1
'
'
2
1
2
'
1
charge-leveler shoes cleared it. The oven was then charged and the coal was leveled with care so that the thermocouples were not disturbed. The doors of the oven were then closed and luted, and the connections to the measuring instruments were made. The automatically recording potentiometer connected to the couples 1, 3, 5, and 7 inches from the floor was started, recording temperatures about 10 minutes after the oven was charged. Manual readinns of the temDeratures
B'IGVRE 3. 'I'HERMOCOUPLE WELL measured by the couples located 1, 3, 5 , and 7 floor, a t the top of the charge, and 9 ASSEMBLY inches from the floor were automatically reinches from the' floor were not started until almost 45 minutes after the oven corded by a four-point recording potentiometer was charged, because of the activities of all observers in connecpyrometer (Brown Instrument Company); the temperatures tion with the leveling and door-closing procedure. These temmeamred by the remaining couples were manually read every 5 peratures were then read every 5 minutes. I n the laboratory, minutes with a Leeds & Northrup portable potentiometer. Temtemperatures were read from the recorder chart a t corresponding peratures in the regenerator outlet pass and in the stack were &minute intervals and were tabulated with the manual readings measured with chromel-alumel couples and the portable potentiometer. Flue temperatures were measured with a Leeds & and plotted in various ways. Northrup optical pyrometer.
Sampling and Testing of Coal, Coke, and Gas Samples of the coal being coked at the time the tests were made were taken from the belt conveyor carrying the coal from the railroad cars to the storage hopper on the oven. Laboratory samples were quartered from the bulk sample, the remainder being used in one of the box coking tests conducted. A sample of coke representing the oven charge made during the temperature measurements was taken from the coke conveyor belt. Individual pieces representing the coke at the end of the thermocouples were carefully removed as the coke-quenchin car was being emptied. The piece of coke directly at the ends ofthe thermocouples was recovered in one large piece. The analyses were made in the laboratories of the geochemical section of the Illinois State Geological Survey using A. S. T. M. standard methods. Gas samples were taken by displacement of water in 5-gallon glass bottles. The samples were collected, over a eriod of 40 minutes, from the main gas line before and after su$r removal. Gas analyses were made both with the Orsat-type gas analysis equipment and with the Podbielniak gas-fractionating equipment.
Procedure for Temperature Measurements The thermocouple and pressure tube assembly was made ready and placed conveniently near oven 9 which was selected for the study because it was located next the end oven of the battery and because its operation was typical of the ovens. The table on which the recording and measuring equi ment was mounted was placed at the end of the battery, and thereads were so placed that connections could be made quickly. After the charge just precedin the charge under test had been pushed, the thermocouple assem%ly was lifted into the oven and placed by trial so that the
Temperature in the Charge
The temperatures measured inside the charge from a time shortly after the coal was introduced into the oven until just a few minutes before the coke was pushed out of the oven at the seven places in the charge are plotted against time in Figure 4 as smooth curves through average values. T h e actual curves for locations 1 and 2 and the latter part of the curve for location 3 show a regular wave form caused by temperature changes in the combustion flues brought about by the half-hourly reversal of regenerators and burners. Since it was closest t o the flues, t h e thermocouple resting on the oven floor definitely showed this effect. The isothermal lines for each 50" C. (90" F.) temperature interval are plotted on Figure 5, and the isochronal lines for the same data are plotted for hourly intervals on Figure 6. All of these methods of plotting the data assist in making the significance of the results apparent. The isochronal plot (Figure 6) shows clearly the relation between the method of applying the heat to the charge and the temperature conditions during the coking period. The charge received heat from both the floor and from the hot gases in the free space above the charge during the 7.5 hours, but mainly from the floor. After about 8 hours the rise in temperature of the top part of the charge due to heat from the floor completely overshadowed the effect due to heat from
MAY, 1937
INDUSTRIAL AND ENGINEERING CHEMISTRY
above, and it is evident that the charge lost heat to the space above in the last stages of the coking cycle. The heat absorbed from above apparently caused no appreciable temperature effects below 3.5 inches from the top of the charge. The charge may be divided into two parts on the basis of thermal treatment. The lower 7 inches of the charge is subjected to a high-temperature coking process and in it the plastic zone travels regularly and a t a constant rate from the bottom to the top. The coal in the plastic zone is exposed to a high heat gradient. The upper part of the charge is heated gradually and more or less uniformly throughout its mass by heat contained in gases from the plastic zone in the lower part of the charge and by heat from the hot gases in the free space above. The heat gradients are lower, the final temperature is lower, and the mass cokes throughout within a relatively short time interval after the plastic temperature of the coal is attained. The gases are probably released throughout the mass rather than a t a plastic zone of limited extent as in the lower part of the charge. This condition is also clearly brought out by the isothermal lines in Figure 5. The influence of the oven conditions on the appearance of the finished coke is clearly shown in the photograph of the piece of coke taken from the end of the thermocouples after the test (Figure 7). A oharge of coal being coked in a high-temperature coke oven consists of (a)a zone of coke extending from the hot wall or floor to (6) a plastic zone which is quite narrow and which separates the coke from ( c ) the unchanged part of the coal charge. It is common knowledge that there is a large temperature gradient across the plastic zone, on one side of which there is hot coke and on the other side coal a t a temperature a little above that of boiling water. The rate of coking is determined by the rate of travel of this plastic zone. The movement of the plastic bone can be estimated (a)from
509
I
1200,
8
7
O b 6
i
3
5
7
9
I
IO 5
DISTANCE IN CHARGE- INCHES
FIGURE5 (Above). ISOTHERMAL RELATIONS THROUGH THE CHARGE FIGURE6 (Below). ISOCHRONAL CURVESOF TEMPERATURE DISTRIBUTION THROUGH THE CHARGE w
3
2
I ' 0
1
1 5 1
200
I
400
I
1
600
1
800
1 1 1 1000 1200
T E M P E R A T U R E - DEGREES C.
FIGURE 4. TEMPERATURE-TIME RELATIONS AT
VARIOUSPOINTS IN THE CHARGE
1. 2. 3. 4.
On oven floor One inch from oven floor
Three inches from oven floor Five inches from oven floor Seven inohes from oven floor Nine inches from oven floor 7. Ten and a half inches from oven floor, on top of charge 5. 6.
measurements of gas pressures a t various points along the line of travel of the plastic zone, (b) from the movement of the point of maximum temperature gradient in the charge, or (c) from the movement of the temperature zone corresponding to the plastic range of the coal (21). A n attempt made during this test to follow the movement of the plastic zone through pressure measurements was unsuccessful because of the fact that tarry material plugged the tubes which were placed in the oven for pressure measurements. The position of maximum temperature gradient, the temperature drop over a distance located 0.25 inch each side of that position, and the temperature a t that position were obtained graphically by using values read from a large-scale plot of the isochronal lines (Figure 6). The results obtained are presented in Table I. The travel of the point of maximum gradient is shown graphically in Figure 8. Determination of the plastic temperature range of the coal by means of the Agde-Damm apparatus (7) showed that the coal being coked started to soften a t about 350" C. (662" F.) and that the decomposing plastic mass set to coke a t about 450" C. (842O F.). The 350" and 450" C . isotherms are
INDUSTRIAL AND ENGINEERI.LG CHEMlSTKY
510
Womr. 7. APPEARAXCE OF COKE TAKE": i FROM THE E N D included in Figure 8. The close correspondence between the position of the point of maximum heat gradient and the position of the 450"C . (842' F.) isotherm leaves no doubt that the maximum heat gradient occurs a t the plastic zone. Because of the Muence of heat absorbed from above, the conditions in the upper 3 or 4 inches of the charge are not the same as those in the lower part. The heat gradients are very different in these two parts.
VOL. 29, NO.5
OF THE TIiERWOCUUPLE
of the Hues a t around 1350' C. (2462' F.). The side waUS of the Hues in oven 9 at the time of the test were at temperatures averaging 1360' C. (2480' F.). The temperature of the flue gases leaving tlie last regenerator pass and entering the stack Hue ranged between 229' and 586'' C . (444' and 1087"F.) for ovens in normal operation. The temperat,ure increased regularly during a regeneration cycle. Stack temperature ranged between 304" and 338" C . ( S O 0 and 640" F.) also reflecting the regeneration cycle hiit to a lesser extent.
Analyses of Coal Coked and of Coke Produced
Time alter Chsiainn
Distanae of Mer. Gradient Point from moor
HI.
In. 2.13
2.88 3.69 4.38
5.23
6.13 8
6.38
6.511
n e s t Gradient per 0.5 In: s t Mar. Grsdlent Paint C. ' F. 290 554 210 410 170 33s 160 302 140 284 140 284 150 302 140 284
Tmm 11. ANALYSESOF COAL COKED AND DERINO
Tern s t Max. &disnt Paint C. F. 480 896 475 887 470 878 490 914 470 878 450 842 545 1013 630 1166
OB
TEST(DRYBasra)
COKEPEonucJin
Sample Sample No. Date ssrnphd
The proximate analysis of the -8/&ncl~ screenings which were being coked in the ovens a t the t.ime the temperature measurements were mado and of tlie coke produced are given in Table 11. The analysis of t,lie coke is that of the material remaining on a half-inch screen after the coke had been subjected to four drops in the standard shatter test. The -'/r inch material removed was composed largely of uncoked material and of mineral matter which came away when the coke broke. The large cokc corresponded to the coke as sold after being screened to remove the fine materials. This procedure was adopted l~ecausein taking this particular sample special care was taken t o keep the sample in as large pieces as possible, resulting in the inclusion of the unconsolidated top layer. The analysis of this fine material is included in Table 11. Table I11 presents the analyses of samples of coke taken a t other tunes from the main coke conveyor. TAW= 111. ANALYSBISOF Coiuj COKEDAND OF COKEPRODUCED AT VARIOUS TIMES IN THE SOLE-FLUE OVENSSTUDIED (MOISTURE-FREE Rnsis) snmpie
Sample No. Date sampled
Ash. %
The rate of travel of the position of maximum gradient which may also be considered to be the rate of travel of the plastic zone was found to be 0.78 inch per hour over the lower 6 inches of the charge. This is considerably greater than the rates reported as occurring in standard slot-type coke ovens. The combustion Hue temperatures were 1360O"C . (2480' F.) during the test.
Temperature in Flues,Regenerators, and Stack Temperatures of the side walls of the sole Hues in various ovens in the battery as measured with the optical pyrometer vith most ranged from 1300"to 1420"C . (2372'to 2588O F.),
volatrle msttei. % Fixed carbon. 'rotalsuir*i Svi1ate d f & ,
2
Coke c-811
:4/10/34 17.9 5.8 76.8
1.39
c.
1.73 0.04
Coke
c-1453 17.8
Coke C-1454 7/29/35 15.5
77.3 1.44
1.21
7/29/35 5.1
4.2 80.3
1.20 0.49 11.61%
Ash eoftenin. temp * F.
Coal c-1439 7/29/35 12.0 33.8 54.2
..
12,680
11,870
12.180
2,063
2.087
1.128
1.142
This coal is being coked successfully in these ovens but has given much trouble when charged into ovens of the slot type. This is due to its high fusain content which decreases its caking power.
MAY, 1937
Influence of Oven Conditions on Coke Character
INDUSTRIAL AND ENGINEERING CHEMISTRY
511
TABLEIY. ANALYSES01'CROSSSECTIONS 01 COKE SPECIXENS MADEDURINQ TEBT 1
2 GI565
Section No. 3 4 CIS86 C-1587
. 5
Weighted 6
Total
AVBIBBB Sample
The 8:30 p. M, line on FigLab. SnmPie No. C-1564 C-1588 C-1589 . .. 01568 . . . 0-10.8 ure 6 &ows the distrih,ltion of Distanoe from floor. in. 10.5-9.5 9.E-8.0 8 . M .75 fi. 75-3.75 3.75-1.76 1.751) 2.0 Thiakneae of section, in. 1.0 1.75 10.5 10.5 1.5 1.25 3.0 1S.5 16.1 15.5 17.0 15.5 15.4 15.7 temperature in the charge at Ash,' % 12.9 20.6 4.7 2.4 2.9 2.3 5.6 3.0 9.7 the time the charge was re;%%$& '% 88.6 74.9 79.6 82.1 80.1 82.2 79.0 80.9 1.23 1.18 1.31 1.34 1.41 1.20 1.31 1.36 moved from the oven. At this Efsi:,&tter,a % 23.5 11.5 5.8 2.9 3.4 2.8 0.5 3.6 78.5 m.5 94.4 07.1 98.6 97.2 03.5 98.5 time about two-thirds of the c, 490 530 590 8911 1040 1100 ... ..... had the nature highD~~ bssie. b ~ ~ i ~-h.fiee. t ~ ~ ~ . t e m p e r a t u r e coke; the top o n e - t h i r d was being coked TABLE V. ANALYSESOF Cnoss SECTIONS OF COKESPECIMEB under conditions of low-temperature gradient, TAKENFROM NEWOVENS was k i n e finished a t a low final temuerature. Seotion No. 1 2 3 4 5 6 7 8 and was Gore like low-temperature coke: Pieces Rsmple No. C-18fi5 C-1866 C-1867 C-1868 C-l8B9 C-IS70 C-IS71 C-IS72 of coke representing the entire thickness of the 2-3 1-2 0-1 3-4 DistanoefromRoor,in. 7-8 6-7 5-fi 4-5 Ash.*,% 14.2 15.3 14.4 15.6 14.9 15.4 18.3 1 6 . 2 charge and taken from a region close to the 2.3 2.6 Volatile 70 5.0 4.4 a.4 2.4 1.6 2.8 ends of the thermocouples were sectioned hori88.0 81.4 81.3 Fized carbon.* Ye 5 0 . 8 8 0 . 3 SA.:! 81.7 8 2 . , 2 9 2.0 2.7 Volatile mstter,b Yo 5 . 8 2 . 8 5 . 2 3.9 3.4 zontally at approximately 1.75-inch intervals Dry basis. b Moisturesnd ssh-free and the i n d i v i d u a l p i e c e s were analyzed. Cross sections a t a p p r o x i m a t e l y 1.25-inch intervals and a longitudinal section through an entire thickness were made for a study of ceU structure aclonger coking time upon the analysis of the coke, pieces of coke cording to the method of Rose (m).The analyses of the secfrom a charge coked in one of the new ovens were sectioned liorieontally a t 1-inch intervals and analyzed for moisture, tions are presented in Table IV. These values are plotted on Figure 9, which also shows the final temperature distribution ash, and volatile matter. (Table V) The volatile matter in the oven. content of the upper sections of this coke is much lower than those of corresponding sections from the S-hour coke. The external appearance of pieces of coke representing the entire thickness of t h e charge is shown in Figure 5. The corresponding longitudinal s e e tion b y R o s e ' s method is shown in Figure 10, and the cram sections are reproduced in F i g u r e 11. The regular gradation of coke structure from a smallcelled, much shrunken, highly devolatilized, o o 1 2 3 4 J 6 7 high-temperature DISTANCE. iwcncs coke on the bottom FIGURE 8. TRAVEL 01 PLASTIC ZoxE (Figure 11, sample 350' M 450" C. (662' TO 842' F.).4ND POSITIONOF M.~XIMUM GRADIENT 8) to a large-celled v e r y thin-walled c o k e o n the top (sample 1) is plainly visible. S i n c e the time the temperature measurements were m a d e , t w o new batteries of eight ovens each have been put into o p e r a t i o n . All o v e n s a r e now operating on a 10.5-hour e o k i n g FLOOR CHARGE c y c l e . Believing DISTANCE. INCHES that it may be of FIGURE 9. RELATION OF TEMPERAinterest toshow the TURE fi-n YOLATEEGRADIENT AT i n f l u e n c e of a THE COMPLETION OF COKING
~;g~~~~&?%~,,
512
INDUSTRIAL AND ENGINEERING CHEMISTRY
I
4 WOWRE11. CROSSSECTIONS OF COKE MADED U R I N ~THB TEST (ssmple 1 from the top oi the ohawe down
to sample 8 from
the bottom.)
VOL. 29. NO. 5
MAY, 1935
INDUSTRIAL AND ENGINEERING CHEMISTRY
Gas Analyses Five-gallon gas samples were taken a t three different times and analyzed. The samples consisted of: ( a ) a pair of samples taken July 29, 1935, from the main gas line, one just before the gas entered the sulfur removal equipment and the other just after the gas left the sulfur removal equipment; (b) a sample taken October 11, 1935, just before the gas entered the sulfur removal equipment; and ’(c) two samples taken July 22, 1936, one from the suction mains coming from the two new batteries and the other from the suction main from the old battery, both just beforc the gas entered the suction pump. A11 samples were analyzed in a modified ShepherdOrsat apparatus (%’%’),equipped for the determination of hydrogen by combustion over copper oxide (4) and for the determination of illuminants by absorption in fuming sulfuric acid. The sample taken October 11, 1935, was analyzed by fractional distillation in a Podbielniak Model A precision f r a c t i o n a t i n g unit (18). The results of these analyses together with published analyses of coke-oven gas (1, 10) are presented in Table VI. The heats of combustion are calculated from the analyses of the gas as taken from the sample bottles in the laboratory. At the time the samples were taken in the plant, an automatic recording calorimeter indicated a heat of combustion for the gas of 480 B. t. u. per cubic foot.
TABLE VI.
513
ANALYSES O F COKE-OVEN
FLUEOVENS
GASPRODUCED IN
SOLE-
Orsat Analysis 7
Sample No. G-111 Coking time, hr. 8.5 Date of sampling 7/29/35 Sampling oosition
Before HIS removal
coz. % 0 2 ,
3.7 1.3
%
2.6 14.2 44.0 17.0 15.1 1.7 0.4
Sole-Flue-Oven Gas G-112 G-113 G-128 8.5
8.5
10.5
G-129 10.5
7/29/35
10/11/35
7/22/36
7/22/36
.4fter HIS moval
Before HIS removal
4.4 1.2
3.4 0.7
Main from new batteries 5.4 0.6
Main from old battery 4.1 1.3
re-
3.3 2.2 2.8 1.9 14.9 15.7 12.2 14.4 40.0 46.5 45.7 44.3 17.3 15.2 11.7 17.7 16.1 20.0 17.6 15.1 1.3 0.2 1.6 1.2 Negative Not detd. Not detd, Not detd.
__ acetate tm l a n d
.
1
Standard Coke-Oven Gas
1
1
I
(1)
(10)
2.2 0.8
2.6 0.6
0.6
4.0 6.3 46.5 8.1 32.1
5.2 6.1 47.9 3.7 33.9
2.9 5.1 57.4 4.2 28.5
(10)
1 ..
..
.. ..
574
600
1.4
..
..
__I_
paper Calcd. gross
heat of
\
1
423
441
407
47 1
400
536
B. t.’u./ cu. ft. Fractionation Analysis G-113 Methane and lighter (Hz. CHI, CO. 01,Nz), % w.3= Ethane and ethylene, % 1.8 Pentanes and higher, % 1.9 a Too much methane and lighter constituents for satisfactory fraotionation of constituents higher than methane fraction.
Acknowledgment This work is part of the program of research on Illinois mineral resources undertaken by the Illinois State Geological Survey of which M. M. Leighton is chief. The author wishes to acknowledge the assistance and guidance of F. H. Reed, chief chemist, Geochemical Section, during the tests and in the preparation of the manuscript. The author is especially indebted to Paul Grotts for assistance in preparing for and helping with the tests and for the preparation of the charts. L. C. McCabe and C. C. Boley of the survey staff assisted during the tests. Coal, coke, and gas analyses were made, under the supervision of 0. W. Rees, by J. W. Robinson, C. S. Westerberg, and L. D. McVicker. The information in this paper could not have been secured without the full cooperation of the management and personnel of the Radiant Fuel Corporation. The author especially wishes to thank M. D . Curran, president, George Curran, plant superintendent, F. E. Dodge, chemical engineer, and C. H. Case, for their assistance and cooperation.
Literature Cited (1) Am. Gas Assoc., “Combustion, A Reference Book on Theory and Practice,” 3rd ed., p. 95 (1932). (2) Anonymous, Coal Age, 39,421-3 (Nov., 1934). (3) Bone, W. A., “Coal and Its Scientific Uses,” p. 302 (1910). (4) Burrell, G. L., Seibert, F. M . (rev. by Jones, G. W.), U. S.Bur. Mines, Bull. 197, 46-8 (1926).
(5) Campbell, 0. F., Oil Gas J.,33 (45), 68-9 (March 28, 1935). (6) Cooper, G. S., J.Iron Steel Inst., 90, Pt. 11, 17-47 (1914). (7) Fieldner, A. C., and Others, U. S. Bur. Mines, Bull. 344, 15 (1931). (8) Fulton, J., “Coke,” pp. 173-7 (1906). (9) Ibid., pp. 177-8. (10) Haslam, R. T., and Russell. R. P., “Fuels and Their Combustion,’’ p. 282 (1926). (11) Hilgenstock, R. W., J . Gasbelezccht., 45, 617-21 (1902). (12) Knowles, A. S.,U. S. Patent 1,745,996 (Feb. 4, 1930); McIntire, C. V., Ibid., 1,635,280 (July 12, 1927). (13) Knowles, A. S., Ihid., 1,717,884 (June 18, 1929). (14) McBride, R . 8., Chem. & Met. Eng., 42 (6), 300-3 (1935). (15) McBride, R. S., and Selvig, W. A., Natl. Bur. Standards, Tech. Paper 137 (1919). (16) Moss, R . S., Mines and Minerals, 21 (6), 412-14 (April, 1901). (17) Ovitz, F. K., U. S. Bur. Mines, Bull. 138, 17 (1917). (18) Podbielniak, W. J., ISD. ENG.CHEM.,-4nal. Ed., 3, 177-88 (1931). (19) Roberts, .4.,U. S. Patent 1,352,696 (Sept. 14, 1920). (20) Rose, H. J., IND.ENG.CHEM., 17 (9), 896-901 (1925). (21) Ryan, W. P., Am. Gas Assoc. Proc., 7, 861-78 (1925). (22) Shepherd, M., Bur. Standards J . Research, 6 (l), 121-67 (1931). (23) Simmersbach, Oskar, Berg- u.hiittenmann. Rundschau, 1913, 157. (24) Ziegenhain, W-.T., Oil Gas J., 30 (171, 16-17, 1 0 0 (Sept. 10, 1931). (25) Zwillinger, B., U. 9. Patents 1,428,621 (Sept. 12, 1922), 1,656,617 (Jan. 17, 1928).
RECEIVED September 30, 1936. Presented before the Division of Gas and Fuel Chemistry a t the 92nd Meeting of the American Chemical Society, Pittsburgh, Pa., September 7 to 11, 1936. Published with permission of the Chief, Illinois State Geological Surrey.
