T A

1

135

v.GO

UC-NRLF

3Sb

EXCHANGE

ilAJN;^.^

UNIVERSITY OF ILLINOIS BULLETIN

Vol. IX JUNE 3, 19f2 No. 30

fEntered Feb. 14 , 1902, at Urbana, 111. , as second-class matter under Act of Congress of July 16,1394]

BULLETIN NO. 60

THE COKING OF COAL AT LOW TEMPERATURES

(WITH A PRELIMINAPvY STUDY OF THE BY-PRODUCTS)

BY S. W. PARR

AND H . L. OL1N

UNIVERS/iTV OF-' ILLINOIS ENGINEERING ^XPSPtlMEISfT, STATION

URBANA, ILLINOIS

PUBLISHED BY THE UNIVERSITY

PBICB: TWENTY-FIVE CENTS

EUBOPEAN AGENT CHAPMAN AND HALL, LTD., LONDON

TIHE Engineering Experiment Station was established by I act of the Board of Trustees, December 8, 1903. It is the purpose of the Station to carry on investigations along various lines of engineering and to study problems of importance to professional engineers and to the manu- facturing, railway, mining, constructional, and industrial inter- ests of the the State.

The control of the Engineering Experiment Station is vested in the heads of the several departments of the College of Engi- neering. These constitute the Station Staff, and with the Director, determine the character of the investigations to be undertaken. The work is carried on under the supervision of the Staff, sometimes by research fellows as graduate work, sometimes by members of the instructional staff of the College of. Engineering, but more frequently by investigators belonging to the Station corps.

The results of these investigations are published in the form of bulletins, which record mostly the experiments of the Sta- tion's own staff of investigators. There will also be issued from time to time in the form of circulars, compilations giving the results of the experiments of engineers, industrial works, technical institutions; and governmental testing departments.

The volume and number at the top of the title page of the cover are merely arbitrary numbers and refer to the general publications of the University of Illinois; above the title is given the number of the Engineering Experiment Station bulletin or circular, which should be used in referring to these publications.

For' copies of ^bulletins, circulars or other information ad- dress the Engineeri^jExn&Elm^it'SJation, Urbana, Illinois.

UNIVERSITY OF ILLINOIS ENGINEERING EXPERIMENT STATION

BULLETIN No. 60 JUNE 1912

THE COKING OF GOAL AT LOW TEMPERATURES

'WITH A PRELIMINARY STUDY OF THE BY-PRODUCTS )

BY S. W. PARR, PROFESSOR OF APPLIED CHEMISTRY, AND H. L. OLIN, RESEARCH FELLOW, DEPARTMENT OF CHEMISTRY

CONTENTS

PAGE

I. Introduction 3

II. Experimental Work 5

III. Gases 9

IV. Tar , 12

V. Coke 14

VI. The Formation of Coke 20

VII. Summary and Conclusions 26

Appendix 29

254958

THE COKING OF GOAL AT LOW TEMPERATURES

I. INTRODUCTION

1. Purpose of the Investigation.— The investigations dis- cussed in this bulletin had two general purposes in view: (1) to discover some fundamental facts pertaining to the properties and characteristics of bituminous coals; (2) to determine the feasibility of modifying the composition of raw coal in order that a different type of fuel might be produced, or possibly an alteration accomplished of the entire fuel content into forms bet- ter suited to present-day requirements.

2., Scope of Previous Investigations. — In earlier experi- ments1 (1907-1908), the information developed was mainly of the type indicated under the first division; for example, the ex- periments early indicated the important role played by small amounts of oxygen in the gases surrounding the heated masses of coal. The ease with which carbonaceous matter absorbed or united with oxygen was so striking that it seemed desirable to follow the matter into detail regarding the temperatures at which oxidation takes place, and its effect upon the material in hand. As a result, the whole matter of coal oxidation at low temperatures was opened up as one of extreme importance. One fundamental fact brought out in the study2 was the absorbent power of freshly-mined coal for oxygen, and the part oxygen played in producing certain changes in the coal and promot- ing the initial form o,f deterioration in storage. Again3, the prime element in all the phenomena was seen to be that of oxi- dation. It will thus be seen that these preliminary studies on low temperature distillation, while mainly bringing into view what might be termed the scientific or fundamental properties of the material, at the same time determined facts which have had much to do with developing the practical application of

!The present investigations are a continuation of the work carried on in 1907-8 and presented as a preliminary report under the title of "The Modification of Illinois Coal by Low Tempera - ature Distillation", Bulletin No. 24, University of Illinois, Engineering Experiment Station, by S. VV. Parr and C. K. Francis. 1908.

2 "The Occluded Gases in Coal", Bulletin No. 32, University of Illinois, Engineering Experi ment Station, by S. W, Parr and Perry Barker.

3"The Weathering of Coal", Bulletin No. 38, University of Illinois, Engineering Experi- ment Station, by S. W. Parr and W. P. Wheeler; also "The Spontaneous Combustion of Coal". Bulletin No. 46, University of Illinois, Engineering Experiment Station, by S. W. Parr and F. W. Kressmann.

4 ILLINOIS ENGINEERING EXPERIMENT STATION

the information in its relation to storage and spontaneous com- bustion.

In the second phase of the earlier study, i. e., its industrial side as related to the development of a special type of fuel, it seemed to, be established that below a certain temperature, say 700° F., the heavy hydrocarbons, those chiefly responsible for the formation of smoke, could be driven off, yielding a gas of high illuminating power, a tar with high percentage of volatile oil, and a solid which, while it could be burned without smoke, was friable and not well adapted to ordinary use as a fuel.

3. Outline of Present Investigation. — In the present studies, the friable or non-cokine- tendency of the earlier product has been found to depend directly upon the amount of oxidation that has occurred both in the preliminary exposure at ordinary temperature arid in the process of heating to moderately high temperatures.

The fact that a coke of good texture could be produced when a careful exclusion of oxygen had been effected, has given special interest to the present experiments. In addition, im- portant facts have developed in connection with the study of the various by-products. These by-products have also been more or less modified in their characteristics by the exclusion of oxygen.

Briefly outlined, the present studies have developed three lines of industrial interest.

First: The possibility of developing a smokeless fuel of good texture and admirably suited to domestic as well as to gen- eral industrial use where absence of smoke is essential. The ac- companying by-products promise to be of special value. These consist of (a) Ammonia, though smaller in quantity than the yield obtained at higher temperatures; (b) Illuminating gas of high candle-power and high heat value; and (c) Tar, which is composed almost entirely of oils, with a minimum amount of pilch and free carbon. Some of the oils produced are of pecul- iar structure and may have more than passing interest, two of the fractions, for example, being readily oxidizable. The iodine absorption numbers of the lighter fraction are found to be as high as 165.

Second : They suggest a possible method for the manufac- ture of producer gas which would be free from present difficul- ties attending the use of bituminous coal, and would convert a

PARR-OLIX — COKING OF GOAL AT LO\V TEMPERATURES 5

much higher per cent of the fuel into the gaseous form. In view of recent developments in the matter of combustion, effici- encies are possible1 where gaseous fuel is available which are almost revolutionary in character.

Third: There are opened up interesting possibilities in the production of coke, briquettes or other forms of fuel in a dense and stable form to meet certain requirements of shipping, stor- ing, foundry, and other industrial uses. Certain facts developed in these studies will be found to throw some light on the prob- lem of coking, which is at present but little understood2.

It is not intended here to enter into a discussion of these three main topics; they will be taken up again after the details of the experiments have been set forth. The results of the ex- periments may then with better understanding be made to enter into the conclusions reached.

II. EXPERIMENTAL WORK

4. Apparatus. — The apparatus employed is illustrated in Fig. 1. From the high pressure main at A, steam was admitted to BB. a %-in. pipe 11 ft. long, fitted with twro return elbows. The steam was then heated by a 26-burner combustion furnace, CC. The retort D, 18 in. by 8 in., containing the coal, was fitted with a head J held in place with set-screws and sealed with asbestos. From the retort, the distillates were conducted by a pipe to a condenser E connected in turn with a large wash bottle F. Here the oils and tars were collected while the gases passed on to the gasomeier G. A Hoskins nickel-nichromc thermocouple,, inserted through a stuffing box 5 and joined to a millivoltmeter K measured the temperature of the retort contents. A battery of burners placed directly under the retort provided a means for securing additional heat, which was retained by means of, an asbestos-lined oven which entirely surrounded both the retort and the furnace.

5. Cse of Superheated Steam. — Superheated steam was used in this series of experiments as a medium for carrying1 the

iSurf ace Combustion. Proc. Am. Gas Inst.. 1911. By Prof. W. A. Bone. In this article Prof. Bone gives data showing an efficiency in the generation of steam by use of the principle of surface combustion of 94.2 per cent. It should be noted, however, that this efficiency is based upon the net heating value of gas.

2"The question as to what really is the factor that produces the coking tendency character- istic of some coals has been a matter of some speculation among manufacturers and users of coke for two hundred years and we are no nearer to its solution now than were the investigators of two centuries ago.''— Iron Age. 1907. F. C. Keighley.

ILLINOIS ENGINEERING EXPERIMENT STATION

FIG. 1

heat into the coal mass, in order to distribute the heat evenly throughout the coal and thus obviate the necessity for revolving the container. In the earlier experiments (1907-1908 , the carboniza- tion was carried on in a cylinder heated externally and mounted on hollow trunions in order to make possible the turning of the re- ceptacle, while at the same time the hollow bearings permitted the admission of various inert gases at one end and the discharge of the distillates at the other. With that device, the frequent turning over of the coal seemed to be unfavorable to the forma- tion of coke having a homogeneous texture. Moreover, the mechanical features were not easily installed. There was positive evidence also of the activity of small quantities of oxy- gen, which entered by leakage or as an impurity in the circula- ting gas employed, thereby acting as a disturbing element. There seemed sufficient reasons, therefore, for employing a fixed retort and using superheated steam as the medium for con- veying the heat and also for securing a suitable atmosphere for the distillation. As will be seen later under the discussion of the coking of coal, the use of steam in this manner has other advantages which, while not fully appreciated at first, are directly in line with the fundamental conditions upon which depends the property of coke formation.

6. Coal Used. — Table 1 gives the data concerning the coals used, it should be noted that since these studies were made for the

PARR-OLIN — COKING OF COAL AT LOW TEMPERATURES

purpose of testing the coking powers of the different coals and not to determine their relative commercial values, many of the samples selected were cleaner than the general run-of-mine. The low ash and sulphur percentages result from the exclusion of pyrites.

TABLE 1

COMPOSITION OF COAL

Mines Counties— Illinois	Moisture	Ash	Volatile Matter	Fixed Carbon	Sulphur	B. t. u.
Vermilion	8.80	8.72	43.05	39.43	2 88	12673
Franklin	6 84	7.38	37 96	47 82	1 33	12770
Saline	3.93	5.80	37.86	52.41	1.54	13593
Macon	8 70	12.12	39 30	40 88	2 30	11 417
Perry	7 19	10 05	35 42	47 34	80	12153
Williamson	5 30	8.55	36 50	49 65	2.77	12640

7. Operation. — A quantity of coal sufficient for one run only from 2500 to 3000 grams, was crushed at one time. In the first experiments, the pieces ranged from Vi in. to buckwheat size, the dust being removed by a sieve. At first the coal was put directly into the retort, but it was found that the circulation of the steam was retarded, delaying the heating of the mass. To remedy this, a cylindrical sheet-iron container, 6 in. in diameter, perforated with small holes, was made to hold the charge. This shell (see Fig. 2) being smaller than the retort and having a sur-

O O O O O 0	O
o o o o e o	o

FIG. 2

rounding space of about 1 in., allowed a free distribution of heat. It was used throughout the remaining runs of the series.

Steam was admitted from the main and allowed to blow through the system until the air was entirely displaced. The combustion furnace was next started and then the burners under the retort. The coal was not stirred after heating had begun.

Table 2 exhibits the average working conditions. By im- proving the facilities for applying external heat to the retort, the time of the later runs was reduced to an average of about five hours.

ILLINOIS ENGINEERING EXPERIMENT STATION

TABLE 2 TEST CONDITIONS : FIRST SERIES

Run No.	3	I	5	6	7
Weight of coaJi, grams	4800	5351	2195	3498	3398
Weight of residue, grams	4030	4112	1895	2810	2895
Max temp, (degrees C )	475°	515°	450°	410°	430°
Racio of coke	84$	76.8$	86.3$	80.3$	85.2$

JNineteen runs were made in the first series, using Williamson Co. coal for the first 10 tests. In the other tests, the coal came from the following counties in the order given. Ver- milion, Williamson, Franklin, Saline, Macon, Vermilion. Vermilion, Williamson. Vermilion.

8. Distribution of Products. — Table 3 illustrates the distri- bution of products.

TABLE 3

EXPERIMENT2 NO. 1 1

Coal used3 , Electric Mine. Danville, 111.

Temperature (average) 450°

Time of distillation 5hr.

Volatile matter in original coal not including moisture 43.00

Volatile matter in coke residue 27,95

Volatile matter in coke residue referred to original coal. . . 22.01

Loss in weight of original coal, volatile matter only, not

including moisture 20.28

Total volatile matter derived as above, not including mois- ture 42.29

Total material, removed by distillation including moisture 29.10

2Selected as a typical example .

3For methods of calculating percentages of coal constituents in this and succeeding tables see Bui. No. 16, p. 209, 111. Geol. Sur.

TABLE 4

YIELD OF PRODUCTS FOR DIFFERENT PERIODS OF HEATING

Time of Heating	3hr.	6hr.
Coal	3000 grams	4000 grams
Coke . .	2327 grams	3902 grams
	77.50$	72.50$
Weight of tar , . ....	238.5 grams	316.0 grams
Per cent tar	7.93$	7.90$
Weight of total water	208.5 grams	348.4 grams
Per cent free moisture	3.38$ o oo<£	3.00$ nr.Q .,,<£
water constitution	3.55$ or6-93*	5.71$ Or8'7l#
Volume of gas at 760 mm and 0°	87 liters	134.7 liters
	.46 cu. ft.	.54 CU. ft.

From the preceding tables a fair indication is given of the ratio of distribution of the main products of decomposition. A study of these three products, gas, tar and coke, has been made, sufficient to determine their general characteristics and value.

PARR-OLIN — COKING OF GOAL AT LOW TEMPERATURES

9

III. GASES

9. Analyst of the Gas. — The methods of Hempel were used in making all gas analyses. For absorbing the illuminants, bro- mine water checked with the results from fuming sulphuric acid and was free from the disagreeable properties of the latter. The paraffin hydrocarbons were determined by the use of the ex- plosion pipette. Hydrogen was determined separately with palladium sponge, a variation from the ordinary industrial method necessary when higher paraffins are present in the gas. Calorific values were determined with the Parr gas calorimeter.

It is impossible to determine absolutely the paraffin content of a gas by any methods now in general use, when more than two of the homologues of methane are present. However, by measuring the contraction of the gases and the amount of G02 produced in burning them in the explosion pipette, the total volume of the hydrocarbons having the general formula Gn H2n+2 may be determined together with the average value for n. On the assumption that the higher homologues are all ethane, the percentages of methane and ethane may then be computed1.

Several analyses of the gases obtained early in the work were made, but on account of air leakage in the gasometer, the results obtained were misleading. Table 5 shows the average of results obtained under satisfactory conditions, from gas evolved at an average temperature of 400°.

TABLE 5

GAS FROM DANVILLE ELECTRIC MINE COAL

H2S	C02	Illuminants	CO	H2	C2H6	CH4	N2	B. t. u.
3.2	5.7	8.3	5.2	5.0	14.4	51.4	5.7	1032

The computed heat value of this gas was 1024 B. t. u. and agrees closely with that determined directly. Heat values of the different gases as given by Abady2 were used as the basis of cal- culation.3

From the agreement between the observed heat value as -shown by the gas calorimeter and the calculated value as derived

lAbady, Gas Analyst's Manual, p. 356, 1902.

2Qas Analyst's Manual, p. 521, 1902.

3According to J. H. Coste (Chemical Engineer, February, 1911) it has been found from Julius Thomson's figures that the average calorific value of the unsaturated hydrocarbons is •equivalent to that of propylene, C3H6.

10 ILLINOIS ENGINEERING EXPERIMENT STATION

from the constituents, indirect evidence is obtained as to the correctness of the assumption concerning the composition of hy- drocarbons assumed to be present in the higher forms.

10. Heat Value. — It will be noted that this gas is relatively of high heating value, 1024 B. t. u. per cu. ft. Compared with ordi- nary city gas at 600 B. t. u. per cu. ft., this gas has a heat value about 70 per cent greater, i. e., 1 cu. ft. at 1024 B. t. u. would be equal to nearly 1.7 cu. ft. at OuO B. t. u.

11. Sulphuretted Hydrogen. — The gas is practically free from naphthalene but has a considerable content of H2S. The latter feature is unexpected, since the temperature of decomposition of FeS.2 is 1000° G. and above. Doubtless, therefore, the sulphur- etted hydrogen present is in the main due to the breaking down of the organic sulphur. It seems to be entirely in the form of H2S and, therefore, easily removable by the usual methods of purification. Some of the coke residue from a coal having orig- inally 4 per cent of sulphur was examined to see if any of the iron pyrites, FeS2, had been broken down by the temperature em- ployed to ferrous sulphide, FeS. Five grams were treated with a large excess of dilute hydrochloric acid. The mass was thor- oughly washed and the percentage of sulphur remaining deter- mined. Test No. 1 gave 3.55% ; No. 2, under identical conditions,. 3.70%. A quantity of the same residue kept well moistened wyas- then exposed to air and sunlight for a period of twelve days in order to oxidize any FeS present to a sulphate. After washing, the sulphur content was 3.74%, indicating that FeS in the origi- nal sample was absent.1 It is evident, therefore, that the pyritic iron had been little affected by the temperatures of the retort.

12. Ammonia. — Any by-product process for the carboniza- tion of coal would, of course, take account of the nitrogen liberated in the form of NH3. At the temperature employed in these experi- ments, it would not be expected that any considerable part of the nitrogen organically present would be decomposed. The follow- ing values are shown in a distillation2 varying in temperature from 375°-400°G. In this work the entire distillate from a run of 3000 grams was retained and the total ammonia of the liquor de- termined. It was found to contain ammonia as NH3 sufficient to

iQn the subject of the decomposition of pyrite, Peters, in Principles of Copper Smelting, p. 268, quotes Sticht as saying ' 'At dull red heat FeSs loses 3/7 of its sulphur and becomes Fe7S8. At 1200°, it becomes for the first time FeS".

2Experimentsby Mr. E. C. Hull, Fellow in Chemistry, University of Illinois, Engineering- Experiment Station, March, 1909.

PARR-OLIN— COKING OP GOAL AT LOW TEMPERATURES 11

represent a yield of 0.8 Ib. per ton of coal, somewhat less lhan ^ of the yield from high temperature distillation. It is not certain that the value of this product would pay for its recovery.

13. Decomposition of Oxygen Compounds. — The oxygen compounds upon decomposing form water. They are, therefore, often referred to as the water of constitution. They are prop- erly considered under this division, though not forming perma- nent gases. It is a question of great interest whether any decom- posing action in connection with the temperatures employed has taken place. If such decomposition has occurred, it has by so much enriched the fuel value of the remaining coke for the rea- son that these compounds are inert and noncombustible and, when present, by so much increase in effect the ash factor so far as combustion is concerned. The fact of their decomposition is shown by the increase of water content in the distillate over and above that which would normally occur from a condensa- tion of the hygroscopic moisture alone. While this fact was not available in the case of distillation with superheated steam, the point was well established in the previous experiments1, as also by experiments conducted by Mr. E. G. Hull, not heretofore pub- lished, in which careful measurement was kept of the amount of water distillate recovered from the coal used.2 Thus, from the work of the latter we have the following :

TABLE 6

	3000 Grams Coal Distilled for3hr. Temp. 300° to 400°	•4000 Grams Coal Distilled for 6 hr. Temp. 300° to 400°
Weight of water in distillate	208.5 102.0 106.5 3.55	348.4 120 0 228.4 5.71
Weight of free moisture in original coal . .
Excess water from decomposition of oxygen compound in coal. ... Per cent of water from decomposition of oxygen compound

14. Summary of Data Concerning the Gaseous Product.— Distillation of Illinois coals at temperatures averaging 450° G. and not exceeding 500° G. produces a gas having a heating value exceeding 1000 B. t. u. per cu. ft. The yield approximates % cu. ft. per Ib. of coal which, at the heat value present, would repre- sent a yield of 1.00 cu. ft. per Ib. of a gas with a heat value of 500 B. t. u. per cu. ft. The ammonia yield is low, being approx- imately 3 Ib. of ammonium sulphate per ton of coal. Decom-

iBulletin No. 24, University of Illinois, Engineering Experiment Station, Parr and Francis.

2See also Porter and Ovitz. Bulletin No. 1, U. S. Bureau of Mines, p. 26-28.

12 ILLINOIS ENGINEERING EXPERIMENT STATION

position at this temperature extends to the oxygen compounds, which are in the main carried off and appear in the condensate instead of in the gaseous product. This feature will be referred to again under the discussion of the composition and properties of the coke residue.

IV. TAR

15. Composition. — As already noted, the amount of tar re- covered from the distillations approximates ~y% of the yield of vol- atile matter and in the sample noted where a direct weighing was made (Table 6), this material represents very nearly 8% by weight of the original coal. An exhaustive study of this ma- terial would be an elaborate topic for research in itself. We can, therefore, give only the general characteristics of the ma- terial as found by fractional distillation as follows :

TABLE 7 FRACTIONS FROM Low TEMPERATURE TAR

Amount of tar (exclusive of water carried over).	375 grams
Light oil ( 20°-100°)	39 1 '	10 5%
Fraction (,b) (100°-200°j .. ...	109 1	29 1
" (c) (200°-240°)	111 8 '	29 8
(d) (240° -275°) .	20 6 '	5 5
Coke residue	80 0 '	21 3

From the results as given in Table 7, it will be seen that 75% of the material classed as tar is in reality oils of different spe- cific gravities and thus of much greater value than the pitch proper. This latter product, moreover, is much smaller in amount than is produced with high temperature distillation. In the latter case over one-half of the tar is pitch, with a consider- able content of free carbon suspended in the material. The low temperature product is approximately one-fifth a pitch residue with some suspended carbon present, seemingly depending on the extent to which the temperature of the coal mass has been carried above 400°.

16. Properties of Oils. — The further examination of the oils distilled from the tar has developed the interesting fact that these oils are readily oxidizable. As a measure of this property the iodine absorption number was determined with results as given in Table 8. It is realized, of course, that the iodine absorption

PARR-OLIN- — COKING OF GOAL AT LOW TKMPERATGtlE&vK X 13

must include or represent other activities than simple oxida- tion especially in a complex mixture where members of the aro- matic series are present.

TABLE 8 IODINE ABSORPTION OF OIL DISTILLATE

Fraction b. 100°-200° (29% ) . . . Fraction c, 200°-240° (29.8% ).

Iodine No. 165 115-125

Further study of the oils recovered is necessary in order to determine their specific values. Their ready oxidizability opens up a very interesting and suggestive field. For example, this feature is a marked characteristic of drying oils, turpentines, etc., used in paint mixtures. The question arises as to whether these oils will have drying qualities, i. e., will they not simply evapo- rate, leaving no residue, or will they oxidize in such a manner as to produce a film-covering, which will serve as a paint vehicle. Or, in a mixture with a drying oil such as linseed or similar oil, will they promote the peculiar properties of such oils which make them of value for paint mixtures? While only a few gen- eral points in this connection have been developed, they indicate characteristics of great interest and, possibly, value. It seems fair to conclude that in some measure at least the iodine ab- sorption numbers are an indication of the avidity of the oils for oxygen. This is shown by the rapid discoloration of the oil when exposed to the air and to the fact that the lighter fraction will yield a dry film on glass at a 45° angle when exposed for 24 hr. under the usual standard requirements for such test. The second fraction has also drying properties, but the process is much slower. Or, rather, a fractionation appears to take place in which the drying oil forms a hard gelatinous film while the non-drying portion segregates into minute globules which are more or less enveloped by the films of oxidized oil. At least, it may be said of the oils which make up the element of the tar, .they are available directly as fuel or for enriching or carburet- ting water gas. For example, if the process were continued to include the manufacture of water gas from the coke residue, the oil of the tar would doubtless enter into the reaction in the same manner as the crude petroleum now used, and thus would furnish the needed enrichment without the clogging effect which results when the attempt is made to use the raw coal directly in the manufacture of water gas.

14

ILLINOIS ENGINEERING EXPERIMENT STATION

V. COKE

17. Yield of Coke. — The yield of coke, under average condi- tions, as already noted in previous tables, is approximately 75% to 80%. This factor will, of course, vary greatly with the amount of ash originally in the coal and on Ihe temperature at which the distil lalion has been carried on. These Hems of variation are shown in the following table where material of widely varying composition was used.

TABLE 9 COMPOSITION OF COKE RESIDUES

	Experiment No. 11 Vermilion Co.	Experiment No. 13 Franklin Co.	Experiment No. 14 Saline Co.
Moisture ...	34	40	.28
Ash	11.15	9.28	6.97
Volatile matter	27 61	26 60	23 50
Fixed Carbon	59.90	63.72	69.23
Sulphur	2 58	1 21	1.20
B. t. u	12892	13446	13746

TABLE 10

SHOWING THE YIELD OF COKE FROM VARIOUS COALS REFERRED TO ORIGINAL COAL — DRY BASIS

	Experiment No. 11	Experiment No. 13	Experiment No. 14
Ash . . ...	9.56	7 92	6 04
Volatile matter expelled	25.48	18.00	19 12
Residual coke.	78 10	84 72	84 86

18. Reactions Involved. — In the transformation illustrated by the change from the composition as given for the raw coal in Table 1 and the residual coke as shown by the table above, No. 9, certain facts may be deduced as follows :

First : there has been, seemingly, a decomposition of the vol- atile matter in a manner which would increase slightly the fac- tor for fixed carbon. For example, if the fixed carbon be calcu- lated as indicated in Table 11 to a percentage of ash correspond- ing to that of the raw coal, comparisons will be obtained as follows :

TABLE 11 COMPARISONS OF FIXED CARBON IN ORIGINAL COALS AND RESIDUES, DRY BASIS

	Experiment No. 11 Vermilion Co.	Experiment No. 13 Franklin Co.	Experiment No. 14 Saline Co.
Fixed carbon in original coal . . . Fixed carbon in coke residue re- ferred to original ash	43.24 46.80	51.30 54.10	54.56 59.70

PARR-OLIN — COKING OF GOAL AT LOW TEMPERATURES

15

19. Oxygep Removed. — As has already been stated, the de- compositions occurring at temperatures in the neighborhood of 400° G. include the liberation of oxygen, or, as it is frequently designated, the water of constitution. Since this ingredient of the raw coal is non-combustible1, it has the same function as so much ash. Its removal, therefore, serves to make of the result- ing material a richer or more concentrated fuel. This feature is still further promoted by the removal of the hygroscopic or free moisture which usually exceeds in amount the water of composition. This point may be illustrated by the accompany- ing table wherein the heat values per pound of the original coal are compared with the heat values per pound of the residual coke. There is also given an estimate of the amount of non- combustible material removed in the form of water in the pro- cess of decomposition.

TABLE 12

					Estimated
Samples	B. t. u. per Ib. As Received	B. t. u. After Treatment per Ib.	Gain Thermal Units	Gain per cent	Loss of Total Non-combusti- ble Free and Combined
					Moisture
Williamson Co	12695	13150	455	3.60	10.30
Saline Co	13583	13746	163	1.63	8.93
Vermilion Co	12673	12892	219	1.72	13.30

20. Properties, Porosity, Hardness, etc. — The coke material •obtained by this process varies in character somewhat with the kind of coal used, and also the amount of pressure employed during the carbonization. The Williamson Go. coal, for exam- ple, gives a coke of much finer texture and less porosity than the coal from Vermilion Go. With a view to determining the reason for this greater porosity or to finding the conditions that would modify it, the attempt was made to carry on a test with the coal sample under pressure. To this end the following apparatus was used :

21. Apparatus. — A, Fig. 3, is an iron cylinder, 8 in. by 4 in., fitted with screw caps B and B1, which received the coal. The movable piston G to which is attached a long rod D, is pressed against the charge by tightening the nuts EE. The cylinder is per- forated with small holes to allow the escape of gases. This con- trivance was fitted into the retort originally used and heat was applied as before.

ifiulletinNo. 3, Illinois Geological Survey, p. 32-33.

16

ILLINOIS ENGINEERING EXPERIMENT STATION

Exhibit 1 shows the results obtained when pressure is applied slowly during the entire heating period. The outer portions passing through the temporary state of fusion soon harden and form a wall which resists external pressure. The inner core, therefore, is extremely porous. When sufficient pres-

FIG. 3

sure is applied, the outer part fractures and, as in this case, the residue comes out broken up into small pieces. The coke shown in the figure is from coal from Perry Go. The specific gravity of the outer portions of the mass is .733 against .652 when coked without pressure.

It was evident, therefore, that in order to get a firm block, pressure must be constant. In the next run, the charge was rammed into the cylinder and the piston was screwed up tightly but not moved after heating had begun. The resulting column cohered well and showed the same increase in specific gravity as the one mentioned above.

22. Illustrations of Various Products. — An interesting fea- ture of the product is the complete fusion of the mass, where proper conditions exist, i. e., the individual particles of coal of buck- wheat or pea size have completely lost their identity, the resulting homogeneous mass showing no lines of demarcation from the original pieces of coal. The texture, however, in some cases is finer or closer than in others. These points are well illustrated in photographs of typical masses as reproduced in exhibits 2 and 3, for coals from southern Illinois. Exhibit 4 represents a somewhat coarser texture. It was made from Ver- milion Go. coal. For the composition of these samples, reference is made to Table 9, p. 14. Exhibit 2 from Saline Go. coal showed a crushing strength of 750 Ib. persq. in1; exhibits from a Franklin

iJohn Fulton, (Coke, p. 331.) gives 1200 Ib. per sq. in. as the ultimate crushing strength of standard Connellsville ooke; by-product coke is. in general, considerably stronger.

The crushing strength is important in reference to the load or burden the coke can with stand in the furnace without crushing.

Exhibit 1

Exhibit 3

Exhibit 2

Exhibit 4

Exhibit 5

Exhibit

Exhibit 8

Exhibit 7

Exhibit 9

Exhibit 10

PARR-OLIN — COKING OF GOAL AT LOW TEMPERATURES 17

Go. sample crushed at 900 Ib. On account of its coarse cellular structure, exhibit 4 showed little rigidity, and broke down at a pressure of 300 Ib.

23. Resume Relating to the Coke Product. — It is evident up- on examination of the coke product obtained, as above described, that we have here a fuel of firm texture, not readily broken down by handling and producible in the most convenient sizes for handling and for efficiency in combustion. It is, moreover, in a more concentrated form, in that for the most part, the free moisture and the water of constitution have both been removed. Thus in freshly mined coal there would be eliminated from 15 to 20 per cent of inactive material. Again, the heavy hydrocarbons have been removed. These are the constituents most directly responsible for the formation of smoke in the combustion of untreated coal. It is to be noted further that because this coke has been subjected to a temperature just approaching a red heat, it will not begin to evolve volatile matter, when thrown upon the iire, before it again comes up to or passes that temperature.

The effect of this point is twofold : first, there is obviated the cooling effect! which must be necessary in the vaporization of moisture in the raw coal which also lowers the temperature just when a high temperature is needed for burning the heavy hydro- carbons; and second, the remaining gases to be evolved consist almost wholly of ethane or marsh gas (GH4) and hydrogen, both of which are readily combustible. The hydrogen, of course, burns with a non-luminous flame and is incapable of making smoke. The marsh gas (GH4), though it has carbon in its com- position, adds but little luminosity to the flame and is almost in- capable of producing smoke in the process of combustion.

It may be well to analyze briefly the processes of combustion as they occur in an ordinary hand-fired furnace. The first result of throwing a mass of coal upon a fire is to lower the tempera- ture during the time of volatilization of the moisture in the coal. Theoretically, the temperature of the mass during this process would remain at or slightly below 100° G.

Other factors tending to lower the temperature would be the specific heat1 of the coal and the heat necessary to effect the de- composition, since it is probable that the decomposition reactions are endothermic up to approximately 300° G2.

JBulletin No. 46. University of Illinois Engineering Experiment Station, Parr and Kress- man, p. 34.

2Bulletin No. 24. University of Illinois, Engineering Experiment Station. Parr and Francis, p. 46—47.

18

ILLINOIS ENGINEERING EXPERIMENT STATION

It is to be noted that during this depression of the general temperature there is being distilled from the coal such volatile substances as are liberated at these lower temperatures. This point can best be illustrated by means of the accompanying dia- gram. In Fig. 4, the region between the lines A £nd B may be assumed to include those volatile constituents that are driven off at a temperature below 400° G. This area includes the free moisture of the coal, the combined moisture or water of con- stitution, or as some prefer, the oxygen compounds of the coal, shown on the chart as inert volatile; and, in addition, some of the pure hydrocarbons which constitute a portion of the true volatile combustible matter. It is, moreover, the nature of this latter or volatile combustible material \vith which we are just now concerned in this discussion of the processes of combus- tion. It is to be noted first that this volatile matter contains the

FIG. 4

bulk of the heavy hydrocarbons. By this is meant that they be- long to the higher series of any of the homologous compounds present which in general are characterized by a higher percent- age of carbon. For example, if the series is that of marsh gas or methane GH4, i. e., Gn H?n+2, then the next higher order of this series would be ethane or G2H6, and the next, propane or G3H8. The carbon percentages, respectively, being 75, 80,81.8, etc. Again a very considerable part of the volatile matter de- livered at this temperature belongs to the methylene series GnH2a, and the first known member of the series is ethylene or olefiant gas, C2H4, with a carbon percentage of 92.92. Moreover,

PARR-OLIN — COKING OF GOAL AT LOW TEMPERATURES 19

this last compound may be made to break down under higher heat into members of other series, as acetylene, C4H4, benzene, G6H6, and naphthalene, G10H8. Other members of the ethylene or paraffin series are found which ally the resulting complex mixture quite closely to the very complicated compounds with which we are familiar in petroleum.

The point to be noted in this phase of the discussion is the fact that these compounds discharged at this relatively low tem- perature, and having these high percentages of carbon, are the most difficult of complete combustion without the formation of smoke.

It is not necessary here to discuss the mechanics of com- bustion of hydrocarbons. As a result of the researches of H. B. Dixon1 and of Professor Bone2 the selective theory of oxygen for hydrogen or the dogma of "preferential combustion of hydro- gen" has beon obliged to give place to the theory of the inter- mediate formation of "oxygenated" or "hydroxylated" mole- cules. In any event, or whatever the theory finally developed by Professor Bone in his most important researches on combustion, the fact remains that these heavier hydrocarbons are the most difficult of all with which to effect complete combustion, and that even under favorable circumstances the tendency in their combustion is to form condensation products in which free car- bon largely predominates. The faulty reaction is thereby made visible to the eye as smoke. A good illustration of this fact is found in acetylene gas, which requires a special burner with special provision for an extra oxygen supply in order to produce a smokeless flarrie.

Smokeless combustion of raw coal is secured, therefore, by observing the principles indicated above; i. e., there must be uniform and gradual accession of fresh coal and a combustion chamber maintained at a sufficiently high temperature, and the same extending over a sufficient space to permit of ultimate mixing and contact of the oxygen with the combustible gases. Other conditions such as accelerating the reaction by introduc- ing the principle of surface combustion, as developed by Pro- fessor Bone, may at some time be added to the mechanical and physical conditions now in vogue. But while these provis- ions are readily adapted to large steam generating units, they are impossible of application to the larger members of combus-

iPhil. Trans. 1893, 159; Trans. Chem. Soc. 61. 873 (1892); 2Chem. News, 102, 309(1911).

20 ILLINOIS ENGINEERING EXPERIMENT STATION

tion processes such as are common to the small plant, house heaters, and possibly to locomotives. It is these latter cases es- pecially that demand a modified fuel which can be burned with- out the formation of smoke.

It will thus be seen that in the low temperature distillation of coal, processes have been put into operation which have taken out the heavy smoke-producing ingredients, and have also removed the moisture, both free and combined, which are chiefly responsible for the depression of temperatures under ordinary conditions. There is left, moreover, as volatile matter, practi- cally these volatile substances only; methane GH4 and hydrogen, which most easily of all the gaseous products from coal, main- tain a smokeless combustion.

VI. THE FORMATION OP COKE

The experiments as thus far conducted seem to throw some light upon the matter of coke formation. In this discussion of the theories involved, it may be helpful to formulate certain hy- pothetical conditions which have had more or less confirmation in these studies, as follows :

First: For the formation of coke there must be present cer- tain bodies which Jiave a rather definite melting point.

Second: The temperature at which decomposition takes place must be above the melting point.

Third: Where the compounds that satisfy the first and second conditions]! are unsaturated, it is possible by subjecting them to oxidation to so lower the temperature of decomposition as to alter the second condition prescribed, in which case coking will not occur.

Discussion of Conditions. — The first condition prescribed above may be well illustrated by the behavior of sucrose or cane sugar. This substance has a rather low melting point, say, 160° G. This melting point, however, is just below the temperature of decomposition. Where this point is reached, gaseous products in the form of steam, etc., are delivered, leaving behind, if the high temperature is continued, a mass of coke. On the other hand, if starch is heated in a similar manner, it does not melt but its first action is that of decomposition. When this is car- ried to completion, there remains not a strongly coherent mass but easily disintegrated particles of carbon. Pure cellulose be- haves in a still more striking manner, showing no fusion pro- perties whatever as maybe demonstrated by distilling in a closed

PARR-OLIN — COKING OF GOAL AT LOW TEMPERATURES 21

tube some cotton fiber, or other form of cellulose such as filter paper beaten to a pulp and dried. Wood, however, if not disin- tegrated, as in the form of sawdust, has enough resinous material closely associated with the fibrous structure to bind t|he carbon filaments resulting from the decomposition of the cellulose by reason of the fact that these gums, or resins, have a melting point below that of their decomposition temperatures, and thus form . a binding film of carbon throughout the mass, producing a sort of coking effect which we find in charcoal.

In the case of Illinois coals, we find the first prerequisite formulated above, as present' in a marked degree. As an illus- tration of the fact of a low melting point, reference is made to exhibit 5, which is a photograph of a mass of such ma- terial, which exuded from a sample of Vermilion Go. coal, sub- jected to the usual treatment as described on p. 7. The lump shown is a part of a mass that flowed out of the container, forming a bubble-like puddle. It would seem, therefore, that this type of coal from the north Danville field (Electric mine) has the first essential for coke formation in a marked degree.

As illustrating the conditions which exist where oxidation had been allowed to take place, an example is given in exhibit 6. This was made from a weathered sample of coal from Niantic. It had little if any tendency to fuse ; the individual particles of coal still retain their form and the mass may be eas- ily crumbled between the fingers. It should be noted that this result is not due to, any inherent quality possessed by the original coal; a Danville mine sample, for instance, weathered to a like degree, gives the same results.

Test No. 9. — Another verification of this point, though in a more marked manner, was the result of test No. 9. The coal used was the fine material which had collected from the prepar- ation of the previous tests, all of which had given excellent sam- ples of coke in their freshly prepared condition. A quantity of coal passing through a 10-mesh sieve had accumulated through a period of about six weeks and had been stored in an ordinary coal hod in the grinding room. After being heated for eleven hours under conditions identical with those of the preceding runs, it showed no signs of fusion and was entirely without cok- ing properties.

It is evident from these tests that the very great avidity of fresh coal for oxygen is evidence of the presence of those com- pounds which satisfy the first of the hypothetical conditions,.

22 ILLINOIS ENGINEERING EXPERIMENT STATION

p. 17, and the subjection of the coal to oxidation destroys the fusion property of the fresh coal and produces a condition cor- responding to that described under the third proposition, in which the coking property is lost.

Other studies on the nature of the coking process were car- ried out as follows : The apparent plasticity exhibited by the coal during certain stages of the treatment suggested the idea of compressing it into a briquette at the time when it would most easily yield to pressure and when it would presumably cohere without requiring an artificial binder. Accordingly, a cupel press with a pressure of 500 Ib. wras provided and the retort was charged with Danville mine coal. At the time of maximum evo- lution of gas, the heat was suddenly shut off and the retort quickly opened. It was found at this point that the outer and hotter portion of the mass was hard and unyielding. A soft in- ner core was discovered, however, and portions of this were put into the press. The resulting briquette is shown in exhibit 8. The escaping gases have swollen it; considerably. Determinations of the amount of volatile matter possessed by the coal when in the plastic condition showed that this constituent had been reduced very little, — from 38 % to 30 % . In short, the state of fusion seems , to exist in early stages of distillation but disappears before the process has proceeded far.

In one of the earlier tests with Danville mine coal using the apparatus described in Fig. 3, the extreme fusibility of this type was again demonstrated. As the piston was slowly forced in, pencils of bituminous matter were squeezed out, through the holes of the cylinder. Exhibit 5 includes some of these nodules. The fact that there was a selective separation of bi- tumen is proved by a comparison of the ash values, the residue as a whole having 13 %, the nodules, 8 % of ash.

The readiness wTith which the cementing material ran to waste seemed to indicate that the coal contained a superfluous amount of it — mo're than was necessary for binding itself togeth- er. The correctness of this view was shown by a series of runs in which crushed gas house coke and anthracite were heated with varying amounts of bituminous coal.

Exhibit 7 shows the hard firm product resulting from the mixture of equal parts of Majestic bituminous coal and gas house coke, both crushed to 20 mesh. Fairly good results were obtained in the next run with 75 % of the coke and only 25 % of Danville Electric coal. In like manner, powdered an-

PARR-OLIN — -COKING OF GOAL AT LOW TEMPERATURES 23

ihracite and bituminous coal in ratios varying from 1 : 1 to 3 : 1 were firmly cemented together. Pitchy material no longer exuded from the retort, being absorbed, seemingly, by the added substance.

One of the essential factors in this scheme for briquetting loose infusible material with bituminous coal is the use of the press for keeping the two substances in close contact. On ac- count of the difficulty of applying such a contrivance in indus- trial work, attempts were made to attain the desired end by us- ing temporary binders, i. e., substances which might hold the par- ticles together clbsely until the permanent coal binder could relieve them.

Mixtures of Danville mine coal and Danville mine coke resi- due No. 17 in the proportion 3 : 1 were thoroughly moistened with water and pressed (1) in the cupel machine and (2) in a testing machine up to 1000 lb. per sq. in. Neither of the bri- quettes survived the subsequent heating, being disintegrated, seemingly, by tjhe escaping steam. The same effect, though to a much less degree, was noted when coal tar was employed. The resulting briquette retained its shape, but was rather soft and fri- able. Crude molasses, of all the binding materials tride, proved to be the best for this purpose. Different percentages of the molasses, ranging from 5 to 15, were tested out at different times. Below ten per cent the strength of the briquette was much diminished. Exhibit 9 is a 3:1 mixture of Danville mine coke residue and fresh Danville mine coal, both ground to 20 mesh, first bound with 11 % of molasses and then pressed in the cupel ma- chine. The cake was next heated in the retort under the atmos- pheric conditions of all the preceding runs. This briquette 2 in. high and 2 in. in diameter, has a crushing strength of5501b. per sq. in. Exhibit 10 shows anthracite briquettes made in the same way. They have a specific gravity of 1.02 and crush at 650 lb. per sq. in.

These tests seem to show that the fusible substance of Illin- ois coals is the true binding material in the coking process; that it is present in such abundance as to produce a coke of too open and spongy a character as a result of the evolution of the large amount of gaseous products which result from its decomposi- tion. In this respect, it is paralleled by the behavior of sugar in the process of coking, which yields as a result of the large vol- ume of escaping gases a very porous mass of sugar, coke or car- bon. However, if the raw coal is mixed with a considerable

24 ILLINOIS ENGINEERING EXPERIMENT STATION

amount of material which has already gone through the coking process, or which has at least given off the larger part of its gases, and then has been reduced to a fine division like breeze, the cementing material of the fresh coal is able to disseminate throughout the mass, and the gases may also escape without blowing it into a spongy mass, with the result that a coke of good texture is formed. Exactly in a similar way, if molasses or other sucrose or glucose material be substituted for the fresh coal, we shall have again the formation of a dense coke capable of retaining its shape under conditions of firing much better than where a plastic binder is used. In both cases a strongly cohering mass is produced which meets the requirements of handling, storage, and combustion with the greatest efficiency and the least formation of smoke. A small admixture of raw coal may thus be made to serve the purpose of a binder for ma- terial otherwise wasted as coke breeze at a cost which would enable it to compete with the pitch binders now in use. It also suggests a process of fractional coking, or coking in two stages. The first result at the lower temperature furnishes a product which, when ground to a moderate degree of fineness and mixed with a small portion of fresh raw coal, would furnish the essen- tial conditions for producing a coke of dense nature with a binder so distributed as to give the material a strength quite comparable with that produced by coals of the regular coking variety. Moreover, an advantage would be evident in such ma- terial, especially for use in household appliances, in that it would be more lively in combustion and less difficult of manipu- lation in the matter of maintaining a fire than coke made by the usual methods.

One point further is to be noted in this connection. It was said at the beginning of the discussion that superheated steam was employed for the purpose of conveying heat into the material so that it would not be necessary to revolve the apparatus in order to secure an even distribution of heat. It is seen from the above detail of the essential conditions to be observed in the coking of coals, at least of this class, that an atmosphere free from oxygen is of prime importance. Indeed following the indicated require- ment, the coal should be fresh, or as recently mined as possible, and in any event retained in larger sizes than in a broken down or a fine state of division, in order that the least possible oppor- tunity be given for the absorption of oxygen. Furthermore, by first admitting steam or bringing the coal into an atmosphere of

PARR-OLIN — COKING OF GOAL AT LOW TEMPERATURES 25

superheated steam, the effect is to1 drive out such oxygen as has been occluded or absorbed by the coal and as yet not chemically combined. This is also brought about at temperatures and other conditions least conducive to a reaction between oxygen and the coal substance. Moreover, from former experiments,1 it has been shown that no reaction at these temperatures takes place between the steam itself and the coal. These principles have an important bearing on certain recent tendencies to concentrate gas production and coke manufacture in large unils and distri- bute the gaseous products at high pressure. From the above, it would seem that the nearer such units were located to the mine or pit-mouth, the belter. If it is found, as seems probable, that the coke residue is a suitable material for further continuation of the gas-making process for the manufacture of producer gas, then the above advantages and essential conditions would be magnified.

A discussion2 by Prof. Lewes, relating to English co.als, bears such a striking resemblance to the facts developed in o,ur own work on Illinois coals that the references have especial interest in this connection. Lewes develops his theories on the basis of the existence in coals of four types of degradation products which have all come from two original forms of \7egetation ; viz : cellulose or lignose, and resinous bodies such as the spores of the lycopodia. The first form of vegetation, i. e., the cellulose, has produced the coal compounds of the humic and ulmic types, while the resinous bodies have produced the other three, viz: (1) : resinous bodies with but little alteration; (2) isometric or other slight modifica- tions in form rather than of composition; and (3) decomposition products from resins produced by the action of heat and pres- sure and consisting of a long series of both saturated and unsat- urated compounds, hydrocarbons of the hydro-aromatic series, and saturated hydrocarbons, like hexane, pentane, etc.

"All these degradation- products of the original vegetation are to be found in the bituminous coals, the residual body and humus forming the basis, which is luted together by the hydrocarbons and resins, and the characteristics of the various kinds of coal are dependent upon the proportions in which the^four groups of the conglomerate are present. The resin bodies and hydrocarbons which form the cementing portion in the coal melt between 300° and 320°C, and if a coarsely pow- dered sample of the coal becomes pa'sty or semi-fluid at this temperature,

iParr and Francis, p. 45, Bulletin No. 24, Engineering: Experiment Station.

2 A recent contribution to the theories of coke formation is made in a lecture by Vivian B. Lewes. Professor of Chemistry, Royal Naval College. Greenwich. England, published in Pro- gressive Age, Dec. 15, 1911. page 1030.

26 ILLINOIS ENGINEERING EXPERIMENT STATION

it is a strong inference that the coal will coke on carbonization, a fact noted by Anderson, and which I have found very useful in practice as a rough test. About these temperatures also the resin bodies and hydro- carbons begin to decompose.

The resin bodies at low temperature yield saturated hydrocarbons, unsaturated, chiefly hexahydrides or naphthenes, together with some oxygenated compounds, while the hydrocarbons yield paraffins and liquid products, all these primary constituents undergoing further decomposi- tions at slightly higher temperatures. The liquids so produced begin to distill out as tar vapors and hydrocarbon gases, and leave behind with the residuum pitch, which at 500°G forms a mass already well coked to- gether if the residuum from the humus is not too large in quantity; the coke formed at this temperature is, however, soft, but if the heat be now raised to 1000°G, the pitch residue undergoes further decomposition, yielding gas and leaving carbon, which binds the mass into a hard coke." He discusses further the action of oxygen upon certain of the initial constituents, referring to the investigations of Bou- douard.

"Boudouard has shown that when coal is weathered humus bodies are produced and the coking power lessened or destroyed. In seven samples. of various coals the humus constituents were increased by the oxidation, which seems to show that the action of the absorbed oxygen is to attack the resin compounds, and as we know that carbon dioxide and moisture are the chief products of the earlier stages of heating of masses of coal, it seems probable that the result is a conversion of resiriic into humus bodies with evolution of these gases, and it is this change which leads to the serious deterioration in the gas and tar made coal which has been too long in store, while the fact that a cannel coal like Boghead or a shale do not weather is partly due to their dense structure and also is an indi- cation that the resin bodies of which they are chiefly composed are of a different type, a fact borne out by their resistance to certain coal sol- vents which freely attack the ordinary resin matter."

A continuation of studies along this line is being made. Ptfention has been made concerning the adaptability of the coke thus produced to use in suction gas producers for furnishing fuel to gas engines. Its freedom from tar, oils, and the heavier products of distillation, which clog and render impossible the use of raw bituminous coals of this type, would seem to offer a solution of these fundamental difficulties. Further studies along this line are also being made.

VII. SUMMARY AND CONCLUSIONS

1. Goals of the Illinois type can be coked at a temperature approximately 400° or -450° G.

2. The gaseous products consist chiefly of illuminants of high candle-power, and represent, together with the condensible

PARR-OLIN — COKING OP COAL AT LOW TEMPERATURES 27

material under (3) following, the chief elements involved in for- mation of smoke in the ordinary combustion of raw coal. The nitrogen of the coal is liberated as NH3, at these temperatures, in amounts representing approximately 20% of the total nitrogen present.

3. The cordensible distillate consists largely of oils with the minimum amount of tar and free carbon. The oils repre- sent positive values for fuel, for carburetting water gas, or for other specific uses on account of their chemical characteristics- as unsaturated compounds.

4. The coke residue has special characteristics which seem to make it o,f value as a concentrated fuel, capable of combus- tion without the formation of smoke, suitable for storing with- out the possibility of spontaneous combustion, and presumably adapted to the manufacture of gas for use in the suction gas producer.

5. Certain facts seem to have been developed concerning the principles involved in the formation of coke which may open the way to the production of a kind of coke of such texture and strength as to make it acceptable for uses that are not now possible with coke made from similar coal, but formed under ordinary conditions, such as are found in the ordinary gas- house retort practice, or that of the by-product coke-oven.

Other considerations1 are pertinent in this connection, such as losses and pollution of the atmosphere which accompany the production of smoke2.

!As illustrating the present-day appreciation of matters connected with fuel economy and activity of thought concerning remedial measures, a quotation is here made from the presi- dential address of Sir William Ramsey before the British Association for the Advancement of Science at York, Eng., July, 1911. (Science, Vol. 34, p. 302, Sept., 1911.)

"The domestic fire problem is also one which claims our instant attention. It is best grap- pled with from the point of view of smoke. Although the actual thermal loss of energy in the form of smoke is small, still the presence of smoke is a sign of waste of fuel and careless stoking. In works, mechanical stokers, which insure regularity in firing and complete combustion of fuel are more and more widely replacing hand-firing. But we are still utterly wasteful in our consumption of fuel in domestic fires. These considerations would point to the conversion at the pit-mouth of the energy, using as intermediary, turbines or preferably, gas-engines; and distributing the electrical energy to where it is wanted. The use of gas engines may, if desired, be accompanied by the production of half-distilled coal, a fuel which burns nearly without smoke, and one which is suitable for domestic fires.

2It is not necessary to multiply arguments for the prevention of smoke- However, a recent article in the Journal of the Society of Chemical Industry, December 15, 1911. by Prof . J. B. Cohen and A. G. Ruston, contains some very striking facts developed in their study of the smoke problem. A few extracts may be given as follows:

"The average per cent of soot passing up the chimney, in 12 analyses including eight of Yorkshire coals, two of Durham coals, and two of Wigan coals, amounted to 6.5 per cent on the carbon burnt. This quantity 6.5 per cent seems a very high figure, representing an annual loss of nearly two million tons on the estimated domestic consumption of 32 million tons. The average deposit of soot over the whole of Leeds will therefore correspond to at least 280 tons per square mile per annum. The tar contained in the soot adheres so tenaciously to everything that it is not easily removed by the rain. The leaves of trees and evergreens in particular get coated with this black deposit. Unfortunately, it does more than blacken the vegetation; it covers the whole leaf over with a kind of varnish, and fills up the pores or stomata, thus check- ing the natural process of transpiration and assimilation. It is in fact no uncommon thing to find in the case of leaves of conifers grown in Leeds that 80 per cent of the stomata are choked up with tar."

28 ILLINOIS ENGINEERING EXPERIMENT STATION

Further studies have in mind the carrying out of the pro- cesses as indicated with apparatus involving the continuous feature, subjecting the mass at the point of greatest fusibility to the pressure of the oncoming material and producing the coke in amount sufficient for testing its properties in the gas pro- ducer and for combustion in oiher ways which would test its properties as a smokeless fuel.

APPENDIX

PARR-OLIN— COKING OF GOAL AT LOW TEMPERATURES 31

APPENDIX I. INTRODUCTION.

HISTORICAL1 STUDIES IN THE Low TEMPERATURE DISTILLATION OP COAL.

Researches in the low temperature distillation of bituminous coals have been carried on at the University of Illinois since 19022. In a series of preliminary experiments, on heating coal to temperatures ranging from 250° to 500° for periods of less than an hour, it was found that the percentage of fixed carbon was increased by more than 25% and that there was a corre- sponding decrease in volatile matter to a point where the forma- tion of smoke was prevented altogether.

In order to eliminate as far as possible those variables which would result from oxidation, Parr and Francis in continuing this work heated Illinois coals in non-oxidizing atmospheres. Choosing nitrogen first as the most suitable medium for this pur- pose, a careful study was made of the quantity and composition of the gases and heavy residues produced at different tempera- tures below 400° G. With a view to securing an absolutely inert atmosphere, after finding that the ordinary commercial nitrogen was contaminated with oxygen, the air in the retort] was dis- placed by steam.

The coals heated in these media underwent changes which rendered them smokeless in ordinary combustion. Howrever, on account of the rotary motion given the retort in order to equalize the temperatures, the coke product came out in a loose granular state much like that of the original coal.

In the course of some of the experiments, while using oxygen as the atmospheric medium, rises in the thermometer readings were observed at unexpectedly low temperatures, seemingly inde- pendent of the amount of external heat supplied. This suggest- ed the idea of a second series of tests entirely separate from the first, to determine the temperatures at which oxidation begins. The method of procedure was to allow the retort to cool slowly

iln assembling the literature relating to the carbonization of coal, it has been planned to bring together first, all of the references to low temperature distillation including the studies of the by-products, followed by references to the theories concerning coke formation and the carbonization of coal in general.

2111. Gteol, Surv., Bull. No. 4. by S. W- Parr. p. 97, 1906.

32 ILLINOIS ENGINEERING EXPERIMENT STATION

until a drop of say 50° had been recorded. Oxygen was then ad- mitted and the resulting temperature was noted. A rise was considered proof of oxidation. This was repeated until a point was reached where no rise in temperature occurred on the read- mission of oxygen. In this way it was found that pulverized bitu- minous coals in pure oxygen began to oxidize at about 125° and that they ignited at about 160°. With diluted oxygen the tem- peratures were somewhat higher.

While making some of their tests with atmospheres of steam it was observed that near the temperature of 315° the mercury made an abrupt rise incommensurate with the amount of exter- nal heat added. After allowing the coal to coiol to 300° and then again heating up to 315° the same phenomenon was observed. No appearance of carbon dioxide accompanied this sudden rise. A tentative explanation is that it was due to the exothermic character of decompositions occurring at that stage.

In considering the subject of oxidation temperatures,1 it was found that freshly-mined coal immediately begins to exude hy- drocarbons and to absorb oxygen and that it retains its avidity for oxygen for an indefinite length of time. The exact result of this absorption was not fully determined, but it seems probable that under favorable temperature conditions it would tend to hasten combustion.

Gonstam and Schlapfer2 publishing "Studies in the Gasify- ing of the Principal Types of Goal" report that the percentage o,f oxides of carbon included in the gases given off in distilling coal varies with the oxygen content of the coal itself.

R. T. Chamberlain3, studying the causes of mine dust explo- sions, found that fresh coal absorbs a large quantity of oxygen but that even under a vacuum it gives off very little. He deter- mined further that coal bottled in air for several weeks yields some carbon dioxide but an amount equivalent to only a small part of the oxygen absorbed. This he thought, might be due to the presence of unsaturated compounds in the coal, which form addition products with oxygen.

Mahler and Gharion4 found that when dry air was passed over pulverized coal at temperatures below 100°, measurable quantities of water, carbon dioxide and carbon monoxide were

iBulletin No. 32, By Parr and Barker, Engineering Experiment Station, University of Illinois. (1910).

2jour. Gasbel. 49, 741. 774. (1906). 3Bulletin No. 383. U. S. Geol. Sur. (1909). 4Compt. rend. 150, 1521, 1604. (1910).

PARR-OLIN-^-COKING OF GOAL AT LOW TEMPERATURES 33>

given off. Between 125° and 200° the liberation of water was so greatly accelerated as to indicate the" splitting oiT of water of constitution. Above 150° the water contained considerable- quantities of acetic acid, from 20% to 40% of the total conden- sate, and showed, in addition, traces of acetones, aldehydes, and methyl alcohol. The upper limit of temperature in their studies* was 200°.

Porter and Ovitz1 made an extended study of the volatile- matter of coal' with a view to determining the influence of the gas composition factor on the efficiency in the use of coal in various industrial processes with special reference to gas pro- ducer, coke oven and gas retort operation.

Their investigations show that the composition of the vola- tile matter of a coal depends largely upon the character of the coal itself. The gases from the younger coals of the West com- pared with those from the coals of the Appalachian region have high percentages of carbon dioxide and carbon monoxide. Be- cause of the readiness with which these gases are given off even at comparatively low temperatures (300°-500°) , the writers con- clude that these western coals contain compounds having a di- rect carbon linkage such as the complex alcohols, aldehydes and acids. They show, further, that contrary to the theory of Dulong, who assumed that in combustion all the oxygen of a coal combined with hydrogen, in the case of certain low grade highly oxygenated coals nearly two-thirds of the oxygen ap- pears in the volatile products in union with carbon, and that this fact accounts for the discrepancy between the determined heat value and that calculated by Dulong's method.

Higher hydrocarbons such as ethane are produced in great- est abundance from the eastern coals and they, consequently, yield more smoke in combustion. In general, however, the gas evolved from any coal subjected to moderate heat only, is rich in the higher paraffins such as ethane and propane. In the case of Gonnellsville coal, at furnace temperatures of 500° and 600° these higher hydrocarbons constitute about 50% of the total paraffin content. At about 800° the percentage reaches a maxi- mum, when it rapidly falls on account of decomposition by heat.

They conclude that the nature of the volatile products dis- tilled from coal in the early stages of heating varies in accord- ance with the smoke producing tendencies of that coal. They

iThe Volatile Matter of Coal. Bull. 1, Bureau of Mines. 1910.

34

ILLINOIS ENGINEERING EXPERIMENT STATION

include among the smoke-producing constituents, tar, benzene, ethylene, and the higher homologues of methane.

E. Boernstein1, subjecting eight Westphalian coals to a max- imum temperature of 450°, reports that the gaseous products of distillation did not exhibit differences corresponding with those shown by the coals themselves. Compared with ordinary coal gas, they were characterized by a higher content of heavy hydro- carbons (5% to 14%) and of methane and its homologues (55% to 76%), and a lower content of hydrogen (5% to 16%). The tars had a specific gravity between .95 and .98, began to distill at about 70° to 80°, and were found to contain no aniline, thio- phene, naphthalene, or anthracene. He states that the solid paraffin content ranged from .3% to 2% (m. pt. 55° to 60°).

Inasmuch as in modern gas retort operation portions of the coal do not reach their maximum temperature for one or two hours, the subject of low temperature distillation is of real importance to the gas industry. In a paper read before the Michigan Gas Association, White, Park and Dunkley 2, report the results of their studies of the primary reactions involved in heat- ing American coals to 500°.

Gas evolution commences only above 300° and that given off in the 300° to 350° interval contains from 25% to 40% of ethane. Above the latter point the yield of ethane diminishes and very little is produced between 450° to 500°. The illuminants de- crease with increasing temperature starting with 8% at 300° and going down to zero at 500°. Methane starting with small amounts reaches its maximum in the 400° to 450° interval. They call attention to the similarity of the gases produced at low temperature to natural gas and suggest that the latter was also produced at low temperature. They give the following results of analyses :

TABLE 13

AVERAGE YIELD AND COMPOSITION OF GAS FROM COAL HEATED FOR Six TO EIGHT HOURS AT TEMPERATURES OF 300°-500°

Coal Volume in cu. ft. per Ib. of	Pittsburgh , Penna.	Bay City, Mich.	Zeigler 111.
Coal. ..	1.42	1.15	0.63
CO2	2 9	16.2	13.1
Ilium ..	2.2	4.1	1.6
CO.	6.2	5.0	5.8
H2	26.3	16.4	13.9
CH4 * . ,	47 0	37.8	38.0
C2H	13 2	11 8	19.5
N2	2.7	9.1	7.8
Calculated B t u ,	902	778	871

1 Jour. Soc. Chem. Ind. 25-213.

2 Am. Gas Light Jour. 89-621.

(1906). (1906).

PARR-OLIN— COKING OF GOAL AT LOW TEMPERATURES

35

The apparent similarity between the gases evolved from coal at low temperatures and natural gas, gives interest to the work of Gady and McFarland1 on the composition of the natural gases of Kansas. They proved the presence of paraflins heavier than methane and ethane, by condensing higher boiling hydro- carbons along with the methane in a bulb surrounded with liquid air. Some of these remained liquid up to ordinary temperatures and had an odor similar to that of light boiling petroleum distil- lates. The quantity of this residue varied in the different gases.

Professor V. B. Lewes2 in discussing the relative merits of high and, low temperatures for gas distillation, gives parallel tables showing the net cost of 1000 cu. ft. of gas produced by each of the two processes.

TABLE 14

COST OF 1000 cu. FT. OF GAS

(1) Hig	h (900°)	(2) Low (400°)
•Coal	pence 13 30	pence 26 50
•Operating expenses	6 74	5 50
	20.04	32.00

LESS VALUE OF RESIDUALS PRODUCED

•Coke	82 cwt	6 11	2 4 cwt	17 64
Tar	9 gal	1 30	4 6 gal	6 90
NH4 products		2 11		2 80
		9.52		27.34

NET COST OF GAS

10.52 4.66

B. t. u. of gas 592 750.

He points out that although the coke residues are figured at the same price, coke (2) is really more valuable since it contains 15% of volatile matter which increases its calorific value. He states also that the low temperature tar distillates contain valu- able fractions of a character different from those obtained from ordinary gas tar, one of which is especially suitable for use in motors as a fuel.

Burgess and Wheeler3 working on the problem of the pre- vention of mine dust explosions, and recognizing the relationship

iJour. of Am. Chem- Soc. 29. 1523. (1907). 2Engineering. 85-410. (1908). 3Jour. Chem. Soc. 97-1917. (1910).

36

ILLINOIS ENGINEERING EXPERIMENT STATION

that exists between the character of the volatile matter escaping- from a heated coal, and its degree of inflammability, studied the composition of the gases evolved at different temperatures.

They found that with all coals whether bituminous, semi- bituminous, or anthracite, there was a well-defined decomposi- tion point at a temperature between 700° and 800° which corre- sponds to a marked increase in the quantity of hydrogen evolved. This increase they attribute to the thermal decomposition of one or more of the higher homologues of methane yielding hydrogen and carbon. Ethane, propane, butane, and, probably, higher members of the paraflin series, form a large percentage of the gases given off at temperatures below 450° ; above 700° they no longer appear.

They believe that the smoke producing elements consist al- most entirely of the higher paraffins and differ from Porter and Ovitz in excluding ethylene and the related unsaturated gases from this class. This view is based upon experiments made- showing that elhylene decomposing at 600°, deposited very little- carbon.

A typical analysis of the gases obtained is given below.

TABLE 15

GAS FROM COAL FROM ABERTILLERY, SOUTH WALES (BITUMINOUS)

Coal (C)	Temp.	Ilium.	COS	CO	H2	CH4	C2H6
	500°	5.8	3.9	4.7	8.0	64.5	11.0
	600°	4.9	3.2	6.4	25.0	47.8	12.4
	700°	2.8	3.4	7.4	34.7	46.2	4.2
	800°	2.8	2.5	9.8	50.8	28.6	4.7
	1100°	4.2	1.4	13.0	60.7	18.8	1.8

In a second paper2 they discuss the results obtained by sub- jecting coals to a series of fractional distillations in a vacuum and determining the compositions of the gases evolved within well defined limits of temperature. They succeeded by pro- longed exhaustion at a low temperature, in removing entirely the paraffin-yielding constituents and leaving behind a compound which decomposed at a comparatively high temperature, yield- ing only hydrogen. They conclude, therefore, that coal is com- posed largely of two types of compounds, the one unstable, giv- ing no hydrogen, the other more stable yielding hydrogen only.

iJour. Soc. Chem. Ind. 5. 2. (1886). 2Jour. Chem. Soc. April, 1911, p. 649.

PARR-OLIN— COKING OF GOAL AT LOW TEMPERATURES 37

G. E. Davis1, discussing the tars formed under different con- ditions, says that at low temperatures are produced mainly such hydrocarbons as belong to the paraffin series having the general formula Gn H2n + 2> along with the olefines Gn H2n. The lower members of these series are liquid, and, furnished in the pure state, are illuminating and lubricating oils ; the higher ones are solid and form commercial paraffin. They are always accom- panied by phenols. Liquid products prevail and among the watery substances acetic acid predominates.

If, on the other hand, the coal has been decomposed at a very high temperature, the molecules are grouped quite differently. While olefines and acetylenes occur more or less the paraffins disappear almost entirely with the resultant deposition of carbon.

Some of this carbon set free is deposited in the retort in a compact graphitoidal form; some occurs in a state of extremely fine division in the tar and forms a constituent of the pitch oir coke remaining behind. At the same time the action of heat effects molecular condensations by which process compounds of a higher molecular weight are formed, such as naphthalene, anthracene and phenanthrene.

Behrens2 found that the tar obtained in the distillation of ooal in the ordinary fire-clay gas-retorts (operated at high tem- peratures) was much richer in benzene, toluene, naphthalene, etc., than the tar made inPauwel's coke ovens (operated at low tem- peratures) from the same kind of coal.

Lunge3 thinks that at low temperatures most of the nitrogen of the tar is in the form of aniline and fatty amines (ethylamine, propylamine, amylamine) ; at high temperatures in the form of pyridine bases, picoline, lutidine, viridine, etc. He admits that the statement needs verification by more detailed investigations. In general, at high temperatures the tendency to complete disso- ciation becomes far more pronounced; the products approach more and more to free carbon on the one hand and free hydrogen on the other.

Watson Smith4 states that naphthalene increases with rise of temperature. This is true also of anthracene, which is then found in the creosote oil coming over before the anthracene oil proper. Carbolic acid is also an important constituent.

iJour. Soc. Chem. Ind. 5. 2. (1886). 2Dingler's Polyt. Journal 208, 362. 3Coal Tar and Ammonia, p. 26, (1900). 4Jour. Soc. Chem. Ind. 8, 950, (1890).

38 ILLINOIS ENGINEERING EXPERIMENT STATION

II. STUDIES IN GENERAL ON THE CARBONIZATION OF COAL

F. C. Keighley1 argues that since the chemical constituents of coals from any horizon are not necessarily indicative of their coking properties, it is reasonable to assume that an important factor determining the coking quality must be one of a physical character and not altogether chemical.

It is known, he says, that the finest coking coals not only are of the bituminous class, but their structure is such that upon fracture they exhibit a fingery or prismatic form and separate vertically, while the more difficult coking coals and the ones of a bituminous character that cannot be coked at all, are of a lam- inated structure and upon fracture break into cubical form and have a tendency to separate horizontally instead of vertically. This he thinks would indicate that the coking property depends very largely upon the arrangement of the small particles of coal composing the seam. If these lie in the seam with their longer axes horizontal to the bedding of the seam they are unfavorable to the coking process. On the other hand, if they are perpendic- ular to the strike of the seam, i.e., at right angles with its bed- ding, the coking tendency is much more pronounced. He sug- gests that the superiority of Connellsville coke may be due to the structure given it in the process of formation by the peculiar geological movements of the region in which it is found.

M. A. Pishel2 suggests a simple practical test, for coking coal. Pulverize the coal to 100 mesh in an agate mortar. Pour out the dust and observe its condition. If it adheres strongly to the mortar, it will probably make ^ood coke, he says. If there is little adhesion, coking properties are absent. In his experimen- tal work he tested more than 150 different specimens. Of the four Illinois coals tried, none stuck to the mortar while most of the Eastern coals adhered. He offers no theory to account for this phenomenon.

Groves and Thorp3 classify coals with respect to their coking properties as sand coals, those devoid of coking powers; sinter coals, those possessing it to a relatively slight degree; coking coals, those which produce a good quality of coke, and anthra- cite.

Jiron Age. 80-364. Aug. 1907. Mines and Minerals Oct. 1907- 2Econ. Geol. June-July 1908. p. 265-270. 3Chem. Tech. Vol. 1, p. 122 (ed. 1889)

PARR-OLIN— COKING OF GOAL AT LOW TEMPERATURES

39

They give the following analytical table made up from the wrork of Richardson. Regnault and others:

TABLE 16

	(	Percentage	)
	c	H	0
	77	5	18
Sinter coal •	83	5	12
	87	5	g
Anthracite	95	3	2

TABLE 17

Anthracite

Blanzy sinter

Lancashire cannel sinter . .

Mons coking

Grand Croix-highly coking,

80C + 88H +0 80C + 128H +60 80C + 128H +30 80C + 24CH + 50 80C + 112H +30

It will be observed from Table 16 that the amount of hydro- gen in the first three varieties is identical, while the oxygen di- minishes as the coking property is developed. The Grand Groix coal (Table 17) has only half the amount of hydrogen contained in the coking coal from Mons. Anthracite, consisting almost entirely of carbon, may be considered a kind of natural coke.

They state in conclusion, however, that Stein of Dresden has. shown that coking and non-coking coals may have the same ul- timate composition and that simple analyses, therefore, cannot determine absolutely the coking property of the coal. They sug- gest that the real source of coking lies in a resinoid body or bedies identical in composition with the coke itself.

White and others1 mention the work of Ste. Glare Deville, consulting chemist of the Paris Gas Company, who, on the basis of results of nearly 2000 tests, divided coals into groups according to the relations of their percentages of oxygen to hydrogen. He found that all coking coals contain a percentage of oxygen ap- proximately twice that of hydrogen.

They reasoned that possibly the artificial application of heat which gives as its first products water and other compounds rich in oxygen, would lower the relatively high oxygen of the non-coking coals and possibly bring them into the coking class. They found, however, that coals which were originally non-cok- ing were not improved in this respect even though the oxygen-

lAm. Gas Light Jour. 89-621. (1906)

40 ILLINOIS ENGINEERING EXPERIMENT STATION

hydrogen ratio was brought down to 2 to 1. The coking coals tested sintered together during the heating and if the resultant mass was heated to redness it retained its shape and gave a good coke. If, however, it was powdered before being heated, it re- mained a powder.

Dr. Haberman1, in studying the spontaneous heating of coals, noted the fact that long storage tends to destroy both gasifying •and coking properties. He found that those coals that oxidized the most and gave the greatest rise in temperature absorbed the largest quantities of bromine.

Professor Fischer2 of Gottingen, working on the same prob- lem, mentions the loss of coking suffered by oxidized coals. He too suggests the bromine absorption test for determining the •chemical activity of the fuel.

Parr and Lindgren3 doing work on the weathering of coal at the University of Illinois observed that in volatile matter deter- minations, samples exposed for several months gave powdery residues instead of coke as in the case when fresh coals were used.

David White4, in his bulletin "The Effect of Oxygen in Goal," -after discussing the negative calorific value of the oxygen and the transition between various grades of coal due to progressive devolatilization brought about more or less directly by dynamic forces, takes up a study of the relative proportion of oxygen, hydrogen, and carbon, in coking coals with special reference to a theory framed to explain the coking quality.

He mentions the work of Regnault8 and Bertrand,6 who found that the high percentage of volatile matter and the high illuminating value of certain bogheads and oil shales are due to the presence of immense numbers of supposed gelatinous algae which, in these coals, seem to have exercised a selective attrac- tion for certain bituminous compounds. Likewise, the condi- tions of accumulation and deposition attending the origin of many coals were doubtless favorable for the mingling of algee and different animal remains with the debris of higher plant types.

Mr. White thinks it is more than probable that the sub- stances of these lower organisms contributed as ingredients to

iSchillingrs Jour, fur Gasbel. 49-419, (1906)

2The Gas World. April 13, 1901.

^Unpublished reports of supplementary studies to Bulletin No. 17, University of Illinois,

Engineering Experiment Station. (1911) 4Bull. U. S. G. S. 382 (1909)

5Regnault, B. Les micro organismes des combustibles fossiles. St. Etienne, 1903. . Soc.d'hist. Nat. Antun. Vol. 9, 1897, p. 193.

PARK-OLIN— COKING OF GOAL AT LOW TEMPERATURES 41

the mass of coal-forming material, and that they, therefore, ex- erted some influence on the character and quality of the final res- idues. He considers the higher percentage of bituminous matter in the older and more altered condition of the fuel, due to con- centration as the result of devolatili/ation of the coal by dyn- amochemical processes, the larger part of the concentration be- ing the result of loss of oxygen, this loss being disproportionately .great as compared with that of hydrogen. Thus, the progressive -deoxygenation of the organic matter accomplishes bituminiza- tion.

Now, he continues, the qualities of fusibility and swelling •concurrent with bituminization which appear to characterize fuels known to contain quantities of gelatinous micro-alga3, are also necessary to the coking quality in coals, and he thinks it permissible;, therefore, to inquire whether the coking property may not be due to some unascertained proportion of gelatinous algal matter entering into the original mass from which the coal was formed and imparting to it this fusibility and tendency to swell.

While the presence of micro-algal ingredients has been noted in peats and even in some brown coals, yet it is very evi- dent that their detection by microscopical means in the highly metamorphosed coking coals, is so difficult as to be practically impossible. The evidence of chemical analysis must therefore, be called into service. The coals, he says, whose large volatile combustible matter contains relatively the highest hydrogen and the lowest o,xygen, thus approaching nearest the bitumen analy- ses, are those in which the organic remains described as micro- algae are most predominant and best preserved. If then, in the high volatile coals high bituminization and gelatinous algal in- gredients go together and the presence of the latter causes the coal to fuse and swell, we may conclude that high volatile coals that show sufficiently high bituminization will coke by the ordinary process. The degree of bituminization is indicated by the rela- tive excess of hydrogen as compared with the diminished oxy- gen in dry coal and is expressed by the ratio H :0.

Data covering the tests of over 300 coals from different lo- calities furnished by the U. S. Geological Survey are given. It was found that those coals having a H:0 ratio of 59 or more, coke by the ordinary commercial process. Nearly all below 59 and above 55 so far as tested, make a coke. Those below 55 us-

42

ILLINOIS ENGINEERING EXPERIMENT STATION

ually give a poor and dark product. The best cokes obtained by the ordinary process were made from coals having a ratio of 60 or over. It was noted however that with coals with a fixed carbon value of over 79 per cent the rule breaks down.

He remarks that his hypothesis appears to harmonize with the tendency of coking coals to cohere when reduced to fine powder, discussed by M. A. Pishel.

0. Boudouard1 took up the study of coals with the specific purpose of delermining the causes of coking and selected for experiments samples of (1) English anthracite, (2) Gourrieres (y± bituminous), (3; Belgian forge coal, (4) Forge coal of un- known origin which has lain in the laboratory several years, (5) Bruay (% bituminous). (6) Goal of unknown origin, (7) Lignite.

The following table gives the results of the approximate analyses :

TABLE 18 COMPOSITION OF COALS BEFORE TREATMENT

	i	2	3	4	5	6	7
Fixed carbon	88 6	89 5	70 5	79 1	39 3	51 4	37.3
Ash Volatile matter Character of coke Hardness2	2.5 8.8 powdery 0	1.6 8.8 powdery o	4.6 21.8 hard 3	2,6 18.1 hard 3	3.1 37.6 hard 3	2.3 46.2 slightly caked 0	4.2 58.4 powdery 0

These coals were successively subjected for periods of 105 hr, each to the action of air at 15° and 100°. After the first treat- ment little change in the coal and in the appearance and char- acter of the coke was noted except that No. 6 and 7 showed traces of hurnic acid. In contrast with this, after being heated at 100°, none had retained their coking powers and all but (1) and (2) contained humic acid. A marked increase in weight due to oxygen absorption was observed, amounting in some cases to nearly 5 per cent.

He further treated 25 grains of each of the coals studied with 150 grams of concentrated nitric acid for a period of 2l/2 months. Analyses of the residues gave the following results :

JBull de Ca. Sec. Chim. 5 (series 4) 365-39-) (1909\

2The relative hardness of the coke is indicated by the figures 3, 2, 1 , 0.— 3 denoting a hard compact coke, 0. a powdery residue.

PARR-OLIN- — COKING OF COAL AT LOW TEMPERATURES

43

TABLE 19

COMPOSITION OF GOALS AFTER TREATMENT WITH NITRIC ACID

	1	2	'	i	5	6	7
Per cent change in weight . . . Fixed carbon. Ash	+ 15.6 68.1 1 8	+ 26.0 54.7 41	+ 6.4 56.5 1 5	+ 20.4 51.5 6 1	+ 17.2 49.6 1 6	— 14.0 43.2 72	- 36.8 39.4 61
Vol. matter .. Appearance of coke Humic Acid . .	30.1 powdered 0	44.8 powdered 0	41.9 powdered 15 per cent	42.2 powdered 8 per cent	48.7 traces of agglome- rate 50 per cent	56.0 traces of agglome- rate 40 per cent	59.9 powdered 27 per cent

Organic solvents such as ligroin, pyridine, benzene, carbon disulphide, carbon tetrachloride and the like, modified in no appreciable way the quality of the coke produced. Concentrated sulphuric acid destroyed the coking power; concentrated hydro- chloric acid had no effect.

In none of these coals did humic acid exist before treatment and since iis presence was always constant in the same oxidized coals which had in the process lost their coking powers, work- ing on the theory that the carbohydrates were responsible for the origin of the acid, he found that starch or sugar treated with bromine water, for instance, yielded humic acid much like that obtained from coal.

It is probable, he thinks, that the hydrocarbonaceous sub- stances giving rise to this acid do not exist in a single form but in a state of great condensation, and polymerization is a result of the decomposition of the living matter, the principal characteristics of this series of processes being the disintegration of the plant tissues and the accumulation of carbon at the expense of hydro- gen and oxygen.

In his comparative studies of natural and oxidized coals, he noted that the production of a very small quantity of humic acid (less than 1 per cent) marked the disappearance of the coking qualities of the original sample.

In this connection the theories advanced by Professor Lewes, already referred to, on page 25, are of interest, harmon- izing as they do with Boudouard's work and presenting some of the most modern lines of thought on this subject.

Dennsledt and Bun/1 hold with Boudouard that humic acids

JZeitsch. f. ang. Chem. 21, 1825. (1908).

44 ILLINOIS ENGINEERING EXPERIMENT STATION

are the ultimate oxidation products of coals and the most in- flammable coals are those that produce the largest quantities of the acid.

The exact nature and composition of the so-called humic acids, however, seem to be unknown. Boudouard1 quotes the re- sults of several experimenters who produced the substance by treating sugar with acids. The empirical formulas (no struc- tural formulas are attempted,) range from C24H1809 (Stein) to C40H<24012 (Mulder). He himself proposes G18H1409 as the compo- sition of humic products he obtained by extracting oxidized coal with potassium hydroxide.

W. G. Anderson in studying the varying coking tendencies of a number of Scotch coals, concluded that cementation is caused by the decomposition of two classes of substances;. (1) resinous materials soluble in caustic potash, which break down on rapid ignition; and (2) non-saponifiable substances, some of which were volatile at 300°, others being stable at this tempera- ture.

III. SUMMARY OF OPINIONS

A very brief review of the literature covering the decompo- sition that takes place at low temperatures in the distillation of coal, is sufficient to prove to the student that the problem in all its phases is distinctly modern. A glance at the bibliography will show (hat few, if any, references date back more than ten years and that most of the publications on the subject have ap- peared within the last two or three. Indeed, Burgess and Wheel- er2 writing in 1910, remark that "previous work has been very scanty". Furthermore, almost without exception, those inves- tigators who have already made reports announce that their first articles are more or less incomplete and that they expect to continue along the same lines of study.

While the development of the subject is evidently still in its infancy, yet results from different sources are in many cases en- tirely consistent. Of particular interest in that it bears a close relationship to the problem of smoke prevention, is the fact, men- tioned by nearly all authorities, that the heavy smoke-producing benzines and paraffins of high carbon content are given off at

iBull. de la Soc. Chim. 5 (series 4) 378. (1909). Jour. Soc. Chem. Ind.. 17-1013. Nov., 1898. 2Jour. Chem. Soc. 97— 1917 (1910)

PARR-OLIN-^-GOKING OF GOAL A:P Ea.TEElliCRfiS 45

low temperatures and are practically eliminated at 500°. At- tempts to separate and estimate the higher homologues of me- thane contained in early distillates, however, have not been en- tirely successful on account of a lack of adequate methods of gas analysis. Gady and McFarland1, using liquid air, got perhaps the best results but even their scheme leaves much to be desired. Writers reporting the paraffin content of the gases studied there- fore have been obliged to estimate the heavier members as "ethane", or, using the formula Gn H2n + 2, to give average val- ues of n.

It is generally agreed further, that as temperatures rise above 500°, methane and hydrogen are the principal gas constituents, being decomposition products of not only the coal itself but of some of the gases given off at the lower temperatures. Below 400°, hydrogen is present in very small amounts. It seems fair- ly well established, therefore, that the density and, consequently, the calorific value of a gas varies inversely with the temperature at which it is evolved and that a very moderate heating of the coal is sufficient to remove enough of the smoke producing ele- ments to make the combustion of the residue clean and econom- ical.

With a very small amount of work done in determining the character of the low temperature tar distillates, a fruitful field is left for future investigation. Paraffin oils, valuable for lubri- cating and power generating, seem to predominate, while the equally important aromatic derivatives, as anthracene, are pre- sent to a less extent than in the high temperature runs.

The investigations of Parr and Francis prove that coal, mod- ified by the application of moderate heat gains valuable prop- erties and that it retains a high calorific vfilue. In the use of cer- tain types of coal, however, such as those of the central west, the problem of putting the residues into marketable condition de- mands a solution before the process can be made an economic success.

Much has been written in attempts to explain the causes of coking, or at least to define the conditions that govern it. From the work of Parr, Chamberlain, Boudouard, and others, who have studied the reactions taking place at low temperatures, it has been proved that oxygen absorption goes on rapidly when fresh coal is exposed to the atmosphere. It has been shown

ijour. Am. Chem. Soc. 29—1523 (1907)

46 : : ill.iK@IS^N(rLNaEaiNG EXPERIMENT STATION

further that this absorption weakens or destroys altogether any coking properties that the original coal may have. In other words a high oxygen-hydrogen ratiio marks the absence of fusi- bility and cementation.

The structures of the organic compounds of the coal which furnish the cementing material for coke and which are appar- ently attacked by oxygen, have not been determined and seem to vary somewhat in different types of coals. However they yield, on oxidation, humic acids of varying composition which decom- pose into powdery residues. Because of the complex nature of these substances and the difficulty experienced in isolating and identifying them, the matter of coking is still an open problem and the explanations advanced are largely hypothetical.

PUBLICATIONS OF THE ENGINEERING EXPERIMENT STATION

* Bulletin No. 1. Tests of Reinforced Concrete Beams, by Arthur N. Talbot. 1904. None available.

*Circular No. 1. High-Speed Tool Steels, by L. P. Breckenridge. 1905 None available. * Bulletin No 2. Tests of High-Speed Tool Steels on Cast Iron, by L. P. Breckenridge and Henry B . Dirks 1905 None available.

^Circular No. 2. Drainage of Earth Roads, by Ira O. Baker 1906. None available.

* Circular No. 3- Fuel Tests with Illinois Coal. (Compiled from tests made by the Tech- nologic Branch of the U. S. G. S.. at the St. Louis. Mo.. Fuel Testing Plant, 1904-1907.) by L.P. Breckenridge and Paul Diserens. 1909. Thirty cents.

^Bulletin No. 3. The Engineering Experiment Station of the University of Illinois, by L. P. Breckenridge. 1906. None available .

*Bulletin No. 4 Tests of Reinforced Concrete Beams. Series of 1905, by Arthur N. Talbot. 1906. Forty-five cents.

"Bulletin No. 5. Resistance of Tubes to Collapse, by Albert P. Carman. 1906. Fifteen cents.

*Bulletin No. 6. Holding Power of Railroad Spikes, by Roy I. Webber. 1906. Thirty- five cents.

^Bulletin No 7f Fuel Tests with Illinois Coals, by L. P. Breckenridge. S. W. Parr and Henry B. Dirks. 1906, Thirty-five cents.

*~Bulletin No. 8 Tests of Concrete: I. Shear: II. Bond, by Arthur N. Talbot. 1906. None available.

Bulletin No. 9 An Extension of the Dewey Decimal System of Classification Ap- plied to the Engineering Industries, by L. P. Breckenridge and G. A. Goodenough. 1906. Re- print available Sept. 1, 1912. Fifty cents.

*Bulletin No. lo. Tests of Concrete and Reinforced Concrete Columns. Series of 1906. by Arthur N. Talbot. 1907. None available.

'Bulletin No. 11. The Effect of Scale on the Transmission of Heat through Locomotive Boiler Tubes, by Edward C. Schmidt and John M. Snodgrass. 1907. None available.

* Bulletin No. 12. Tests of Reinforced Concrete T-beams. Series of 1906, by Arthur N. Talbot. 1907. None available.

* Bulletin No. 13. An Extension of the Dewey Decimal System of Classification Applied to Architecture and Building, by N. Clifford Ricker. 1907. Fifty cents.

^Bulletin No. 14 Tests of Reinforced Concrete Beams. Series of 1906, by Arthur N. Talbot 1907. None available.

^Bulletin No. 15. How to Burn Illinois Coal without Smoke, by L. P. Breckenridge. 1908. Twenty- five cents.

* Bulletin No. 16. A Study of Roof Trusses, by N. Clifford Ricker. 1908. Fifteen cents. *Bulletin No 77 The Weathering of Coal, by S. W. Parr, N. D. Hamilton, and W. F.

Wheeler. 1908. T^venty cents .

Bulletin iVo. is. The Strength of Chain Links, by G. A. Goodenough and L. E. Moore. 1908. Forty cents.

* Bulletin No. 19. Comparative Tests of Carbon. Metallized Carbon and Tantalum Fila- ment Lamps, by T. H. Amrine. 1908. Twenty-five cents.

*Bulletin No. 20. Tests of Concrete and Reinforced Concrete Columns, Series of 1907, by Arthur N. Talbot. 1908 None available.

Bulletin No. 21 Tests of a Liquid Air Plant, by C. S. Hudson and C. M. Garland. 1908 Fifteen cents.

* Bulletin Vo. 22. Tests of Cast-Iron and Reinforced Concrete Culvert Pipe, by Arthur N . Talbot. 1908. Thirty- five cents.

Bulletin No- 23. Voids, Settlement and Weight of Crushed Stone, by Ira O. Baker. 1908. Fifteen cents^.

Bulletin No 24. The Modification of Illinois Coal by Low Temperature Distillation, by S. W. Parr and C. K. Francis. 1908. Thirty cents.

Bulletin No. 25. Lighting Country Homes by Private Electric Plants, by T. H. Amrine. 1908. Twenty cents.

Bulletin No. 26. High Steam-Pressures in Locomotive Service. A Review of a Report to the Carnegie Institution of Washington, by W- F. M. Goss 1908. Twenty-five cents.

Bulletin No. 27. Tests of Brick Columns and Terra Gotta Block Columns, by Arthur N. Talbot and Duff A. Abrams. 1909. Thirty cents.

Bulletin No 28. A Test of Three Large Reinforced Concrete Beams, by Arthur N. Talbot. 1909 Fifteen cents.

Bulletin No 29. Tests of Reinforced Concrete Beams: Resistance to Web Stresses, Series of 1907 and 1908. by Arthur N. Talbot. 1909. Forty- five cents

Bulletin No. 30. On the Rate of Formation of Carbon Monoxide in Gas Producers, by J. K. Clement. L. H. Adams, and C. N. Haskins. 1909. Twenty- five cents.

* Out of print; price attached.

N. B.— A limited supply of bulletins, the titles of which are not starred, is available for gratuitous distribution.

PUBLICATIONS OF THE ENGINEERING EXPERIMENT STATION

Bulletin No. 31. Fuel Tests with House-heating Boilers, by J. M. Snodgrass. 1909. Fifty -five cents

Bulletin No. 32. The Occluded Gases in Coal, by S. W. Parr and Perry Barker. 1909. Fifteen cents.

bulletin No. 33. Tests of Tungsten Lamps, by T. H. Amrine and A. Guell. 1909. Twenty cents.

Bulletin No. 34. Tests of Two Types of Tile Roof Furnaces under a Water-tube Boiler, by J M. Snodgrass. 1909. Fifteen cents.

*Bulletin No 35. A Study of Base and Bearing Plates for Columns and Beams, by N. Clifford Ricker. 1909- Twenty cents.

Bulletin No. 36 The Thermal Conductivity of Fire-Clay at High Temperatures, by J . K. Clement and W. L. Egy. 1909. Twenty cents.

* Bulletin No. 37. Unit Coal and the Composition of Coal Ash, by S. W. Parr and W. F. Wheeler. 1909. Thirty-five cents.

Bulletin No. 38. The Weathering of Coal, by S. W. Parr and W. F. Wheeler. 1909. Twenty- five cents.

Bulletin No. 39. Tests of Washed Grades of Illinois Coal, by C. S. McGovney. 1909. Seventy-five cents.

Bulletin No. 40. A Study in Heat Transmission, by J. K. Clement and C. M. Garland. 1910. Ten cents.

Bulletin No. 41. Tests of Timber Beams, by Arthur N. Talbot. 1910. Twenty cents.

Bulletin No. 42. The Effect of Key ways on the Strength of Shafts, by Herbert F. Moore.

1910. Ten cents.

^Bulletin No. 43. Freight Train Resistance, by Edward C. Schmidt. 1910. Seventy -Me cents,

Bulletin No. 44. An Investigation of Built-up Columns under Load, by Arthur N. Talbot and Herbert F. Moore. 1911 . Thirty-five cents.

Bulletin No. 45. The Strength of Oxyacetylene Welds in Steel, by Herbert L Whittemore. 1911. Thirty -five cents.

Bulletin No. 46. The Spontaneous Combustion of Coal, by S. W. Parr and F. W. Kress- mann. 1911. Forty- five cents .

Bulletin No. 47. Magnetic Properties of Heusler Alloys, by Edward B. Stephenson.

1911. Tiventy-five cents.

Bulletin No. 48. Resistance to Flow through Locomotive Water Columns, by Arthur N. Talbot and Melvin L . Enger. 1911. Forty cents.

Bulletin No. 49. Tests of Nickel-Steel Riveted Joints, by Arthur N. Talbot and Herbert F.Moore. 1911. Thirty cents.

Bulletin No. 50. Tests of a Suction Gas Producer, by C. M. Garland and A. P. Kratz.

1912. Fifty cents.

Bulletin No. 51. Street Lighting, by J. M. Bryant and H. G. Hake. 1912. Thirty-five cents.

Bulletin No. 52. An Investigation of the Strength of Rolled Zinc, by Herbert F. Moore. 1912 Fifteen cents

Bulletin No. 53. Inductance of Coils, by Morgan Brooks and H. M. Turner. 1912. Forty cents.

Bulletin No. 54. Mechanical Stresses in Transmission Lines, by A. Guell. 1912. Twenty cents.

Bulletin No. 55. Starting Currents of Transformers, with Special Reference to Trans- formers with Silicon Steel Cores, by Trygve D Yensen. 1912. Twenty cents.

Bulletin No. 56. Tests of Columns: An Investigation of the Value of Concrete as Re- inforcement for Structural Steel Columns, by Arthur N. Talbot and Arthur R. Lord. 1912. Twenty- five cents.

Bulletin No. 57. Superheated Steam in Locomotive Service. A Review of Publication No. 127 of the Carnegie Institution of Washington, by W. F. M. Goss, 1912. Forty cents.

Bulletin No. 58. A New Analysis of the Cylinder Performance of Reciprocating Engines, by J. Paul Clayton. 1912. Sixty cents

bulletin No. 59. The Effects of Cold Weather upon Train Resistance and Tonnage Rating, by Edward C. Schmidt and F. W. Marquis. 1912. Twenty cents.

Bulletin No. 60. The Coking of Coal at Low Temperatures, with a Preliminary Study of the By-Products, by S. W. Parr and H. L. Olin. 1912. Ttventy-Jlve cents.

UNIVERSITY OF IULJNOIS

THE STATE UNIVERSITY

THE UNIVERSITY INCLUDES THE

COLLEGE OF LITERATURE AND ARTS (Ancient and Moderm Languages and Literatures, Philosophical and Political Science Groups of Studies, Economics, Commerce and In- dustry) . ££

COLLEGE OF ENGINEERING Graduate and undergraduate courses in Architecture; Architectural Engineering; Civil Engineering; Electrical Engineering; Mechanical Engi- neering; Mining Engineering; Municipal and Sanitary Engineering; Railway Engineering.

COLLEGE OF SCIENCE (Astronomy, Botany, Chemistry, Ge- ology, Mathematics, Physics, Physiology, Zoology) .

COLLEGE OF AGRICULTURE (Animal Husbandry, Agron- omy, Dairy Husbandry, Horticulture, Veterinary Science, Household Science.)

COLLEGE OF LAW (Three year's course).

COLLEGE OF MEDICINE (College of Physicians and Sur- geons, Chicago) . (Four year's course) .

COLLEGE OF DENTISTRY (Chicago) . (Three year's course) .

SCHOOLS— GRADUATE SCHOOL, MUSIC (Voice, Piano, Vio- lin), LIBRARY SCIENCE, PHARMACY (Chicago), EDU- CATION, RAILWAY ENGINEERING AND ADMINISTRA- TION.

A Summer School with a session of eight weeks is open dur- ing the summer.

A Military Regiment is organized at the University for in- struction in Military Science. Closely connected with the work of the University are students' organizations fpr educational and social purposes. (Glee and Mandolin Clubs; Literary, Scientific, and Technical Societies and Clubs, Young Men's and Young Women's Christian Asso- ciations).

United States Experiment Station, State Laboratory of Natural History, BiologicalExperiment Station on Illinois River, State Water Survey, State Geological Survey.

Engineering Experiment Station. A department organized to investigate problems of importance to the engineering and manufacturing interests of the State.

The Library contains 200, 000C volumes. The University offers 628 Free Scholarships.

For catalogs and information address

C, M. McCONN, Registrar,

Urbane, Ulinoto.

Illinois State

Reformatory Print

niv.of Illinois. 2nrin,

awrfr

Bullet in , '• 4

:* 7

IS5 y. 60

254958

UNIVE/

<ARY