s 
 
 14. GS: 
 CIR Afe<? 
 c. 3l 
 
 GW( Sujuplu 
 
 STATE OF ILLINOIS 
 
 WILLIAM G. STRATTON, Governor 
 
 DEPARTMENT OF REGISTRATION AND EDUCATION 
 
 VERA M. BINKS, Director 
 
 PLASTIC PROPERTIES 
 OF COAL 
 
 E. D. Pierron 
 O. W. Rees 
 G. L Clark 
 
 ILLINOIS GEOLOGICAL 
 
 SURVEY LIBRARY 
 
 MAY 21 1959 
 
 DIVISION OF THE 
 
 ILLINOIS STATE GEOLOGICAL SURVEY 
 
 JOHN C. FRYE, Chief URBANA 
 
 CIRCULAR 269 
 
 1959 
 
.LUNO.SSTATEGEOL0G.CAL SURVEY 
 
 3 3051 00003 8442 
 
PLASTIC PROPERTIES OF COAL 
 
 E. D. Pierron, O. W. Rees, and G. L. Clark 
 
 ABSTRACT 
 
 Two coals of similar chemical analyses and plastic tem- 
 perature characteristics but of different maximum fluidities (Giese- 
 ler) were investigated to determine what components of coal are 
 responsible for its plastic properties. An Illinois high-volatile 
 B bituminous coal and an Eastern high-volatile A bituminous coal 
 were investigated. A portion of the Illinois coal that had been 
 oxidized in air until its plastic properties were destroyed also 
 was studied. 
 
 The three samples were subjected to an extraction proce- 
 dure. The coals were extracted with pyridine, the pyridine ex- 
 tract was then extracted with chloroform, and finally the chloro- 
 form extract was extracted with n-hexane. The coals and their 
 extracts and residues were analyzed and the atomic H/C and O/C 
 ratios were calculated. They were investigated further by x-ray 
 scattering and infrared spectroscopy. Their plastic characteristics 
 were determined by the Gieseler plastometer and the hot- stage 
 microscope, and the relative thermal decomposition ranges were 
 ascertained by differential thermal analysis. Molecular weight 
 determinations were made on all materials except the original 
 coals and the residues from the pyridine extractions. 
 
 Pyridine extraction of the coals rendered the residues non- 
 plastic. The yield of extract for each solvent appears to be pro- 
 portional to the fluidity. Chemical, x-ray, and infrared data 
 showed a relative increase in aliphaticity with each progressive 
 extraction. In each series, as the progressive scheme of extrac- 
 tion is carried out, each extract possesses a lower melting point, 
 a lower molecular weight, and a similar thermal decomposition 
 temperature range. In each series, the pyridine, chloroform, and 
 n-hexane extracts are similar to the corresponding extracts of the 
 other coals studied. 
 
 INTRODUCTION 
 
 Knowledge of the plastic properties of coal is necessary to a fundamental 
 understanding of coal behavior and has practical application to the carbonization 
 of coal. 
 
 Within a limited range of rank, coals soften on heating and resolidify as 
 the temperature increases. Concurrent with this phenomenon, thermal decomposi- 
 tion occurs and gas evolves. The addition of a gaseous phase to the solid and liq- 
 uid phases complicates measurement of coal properties, and the methods of measur- 
 ing plasticity are consequently empirical. Numerous laboratory tests have been de- 
 signed. In the United States, especially in the laboratories of the Illinois State 
 Geological Survey, the Gieseler plastometer yields results that permit useful eval- 
 uation of coals for certain applications (Reed et al., 1952). The Gieseler 
 
2 ILLINOIS STATE GEOLOGICAL SURVEY 
 
 plasticity data allow the comparison and grouping of coals and coal blends on a 
 qualitative or semiquantitative basis. 
 
 This study was made to determine, by means of solvent extraction, the 
 fundamental components of coal that are responsible for its plastic characteristics, 
 as determined by the Gieseler plastometer. Knowledge of the physical properties 
 and chemical nature of such components may lead to more dependable methods of 
 classifying and comparing coals. 
 
 Acknowledgments 
 
 This report is based on a thesis by E. D. Pierron submitted in partial ful- 
 fillment of the requirements for the degree of Doctor of Philosophy in Chemistry 
 at the University of Illinois, 1958. 
 
 The authors wish to express their appreciation to Dr. John C. Frye, Chief 
 of the Illinois State Geological Survey, to Dr. W. F. Bradley for his help with 
 x-ray analysis, and to K. E. Clegg for the differential thermal analysis data. 
 
 HISTORY 
 
 Several hypotheses have been advanced to explain the plastic behavior of 
 coal. 
 
 Jones and Wheeler (1915, 1916), Fischer and Gluud (1916), Bone et al. 
 (1924), and several others, postulated that the plastic characteristics of coal were 
 induced by the "bitumen" fraction. This fraction, sometimes referred to as the 
 "coking principle", is found in certain coal extracts. 
 
 Audibert (1926) originally suggested a homogeneous melting of all the coal 
 within its plastic range. However, this phenomenon might occur only in the case 
 of high concentration of such a maceral component as pure vitrinite. In general, 
 as exinite, inertinite, and mineral matter are present to a certain degree in all 
 coals, this hypothesis may be misleading. 
 
 Blayden et al. (1944) and Hirsch (1954) offered a turbostratic lamellar 
 model based on x-ray studies, which assumes that coal may be represented as 
 flat poly-condensed aromatic "lamellae" whose average dimensions and degree 
 of ordering increase with rank. This theory explains the softening of coal by as- 
 suming that the small lamellae become mobile at higher temperatures and act as a 
 lubricant for the bigger lamellae. 
 
 Bangham et al. (1949), inferred the existence of units called "micelles" 
 with diameters of about 200 A. The arrangement of the micelles becomes more 
 compact as the rank of the coal increases. Bangham explains plasticity phenomena 
 by assuming that these colloidal particles are covered by surface films that become 
 fluid when heated. Such a concept may be considered as being halfway between 
 the "bitumen" and the "lamellae" theories. 
 
 More recently Fitzgerald (1956) and Chermin and van Krevelen (1957) pro- 
 posed a mathematical model for the carbonization process. The first reaction con- 
 sidered is a depolymerization in which an unstable intermediate product, "meta- 
 plast, " is formed. The metaplast is responsible for the plastic behavior of coal. 
 The second reaction involves both a cracking process in which nonaromatic groups 
 are split off, and a recondensation of the metaplast units to form semi-coke. The 
 third reaction is a secondary degasification in which the semi-coke units polymer- 
 ize as methane evolves and, at higher temperatures, as coke is formed hydrogen 
 evolves . 
 
PLASTIC PROPERTIES OF COAL 3 
 
 It is not surprising to find so many concepts of the plastic behavior of 
 coal considering its complex, heterogeneous nature. In spite of this complexity, 
 plasticity has been determined empirically for a number of years, especially for 
 coals used in making metallurgical coke. 
 
 Of the numerous types of equipment for measuring plastic properties of 
 coal, at least three instruments have been tried in the United States - the Agde- 
 Damm, Davis, and Gieseler plastometers. The Agde-Damm apparatus gives re- 
 producible results, but the data are not readily interpreted in terms of practical 
 application. The Davis plastometer has not given useful results with coals of low 
 plasticity. Rees and Pierron (1955) evaluated the Gieseler plastometer. They 
 found that it was useful in obtaining qualitative or semiquantitative data on 
 plastic properties but not satisfactory for quantitative interpretation. In the case 
 of highly swelling coals, addition of diluents to improve results was found unsat- 
 isfactory because their effect is not linear and is not the same for different coals. 
 The addition of increasing amounts of diluents progressively decreases, by an 
 unpredictable amount, the fluidity of the coals to which they are added. 
 
 EXPERIMENTAL PROCEDURES 
 Solvent Extraction 
 
 Preliminary Investigations 
 
 Considerable work has been done on coal by means of solvent extraction. 
 Approximately 160 different solvents and solvent mixtures have been used. In 
 addition, extractions have been made in different types of apparatus, at different 
 temperatures, pressures, and lengths of time. 
 
 Several theoretical objections have been raised as to whether the coal 
 extracts represent original entities of the coal itself, degradation products of the 
 coal, or perhaps in part the products of chemical interactions between coal and 
 solvent. Another objection to the use of solvent extraction has been the difficul- 
 ty of obtaining solvent-free extracts and residues. Traces of solvent induce 
 errors in structure studies based on analytical data. 
 
 As this study deals with the extractable constituents that render the coal 
 plastic and not with the actual molecular structure as such found in coal, the 
 above objections are not considered pertinent. 
 
 Preliminary studies were undertaken to select solvents and to establish 
 the type and technique of extraction to be used. The factors governing the choice 
 of solvents were ability of the solvent to extract components responsible for the 
 plastic characteristics of the coal, the amount of extract the solvent yields, and 
 the ease with which the solvent can be separated from the extract. Pyridine, 
 ethyl enediamine, and dimethylformamide were chosen. 
 
 In previous studies on solvent extraction of coal, pyridine and ethylenedi- 
 amine have been used more extensively than other solvents. Jones and Wheeler 
 (1915), Fischer and Gluud (1916), and several later workers indicated that pyri- 
 dine removes substances that are responsible for a coal's coking ability. 
 
 More recently the work of Dryden (1949, 1950, 1951) demonstrated the 
 possibility of using ethylenediamine as a solvent and its advantages over a fairly 
 large number of other solvents. 
 
 The use of dimethylformamide as a solvent has been reported by the British 
 Coal Utilisation Research Association Laboratories in 1954. 
 
4 ILLINOIS STATE GEOLOGICAL SURVEY 
 
 MacKensie (1909), Bone and Sarjant (1920) and Malanowicz (1932) have 
 shown the importance of using dry pyridine in order to obtain maximum yield. 
 Berkowitz (1948) and Dryden (1948, 1949) reported an increase in yield of extract 
 when anhydrous ethylenediamine was used. 
 
 To obtain a maximum yield of extract, the solvents were dried as described 
 below . 
 
 The pyridine, reagent grade, obtained from Fisher Scientific Company, was 
 first dried over NaOH pellets for 48 hours with occasional shaking, then filtered 
 into a liter distillation flask and boiled with metallic sodium for three hours. 
 Fresh sodium was added, and the pyridine distilled into a storage bottle contain- 
 ing freshly ignited BaO. The first 150 ml of distillate was discarded. 
 
 The ethylenediamine, 95 to 100 percent pure, obtained from Eastman 
 Organic Chemical Company was dried by the same procedure used for pyridine. 
 
 The dimethylformamide, technical grade, obtained from du Pont de 
 Nemours and Company, Inc., was dried with KOH following a procedure described 
 by Sears et al. (1955). 
 
 The coal used in the course of the preliminary investigation was prepared 
 as follows. A fresh sample of approximately 20 kilograms of Illinois high-volatile 
 B bituminous No. 5 Coal from Saline County, size 2 by 3 inches, was collected at 
 the mine and sealed in an airtight container. In the laboratory about four to five 
 hours later approximately 4 kilograms of the sample were reduced to less than 
 one- fourth inch by means of a jaw crusher, and air-dried overnight at 34 °C. The 
 air-dried sample was then stage-ground, producing a minimum amount of fines, to 
 pass a 40-mesh sieve. A sieve analysis indicated that only 5 percent of the 
 minus 40-mesh sample was less than 60-mesh size. The minus 40-mesh sample 
 was stored under an 02~free nitrogen atmosphere in a bottle.* Approximately 100 
 grams of the sample were introduced in a steel roll mill, under an 02~free nitro- 
 gen atmosphere, and ground to pass a 200-mesh sieve. 
 
 Three soxhlet extraction units, extraction tube 40 mm I. D. and flask 
 capacity 250 ml, were set up. Exactly 15 grams of minus 200-mesh coal were 
 weighed into the three 33 by 80 mm double-thickness extraction thimbles. Cotton 
 plugs were packed lightly on the surface of the coals, and the thimbles placed in 
 the soxhlet on top of 1/4-inch porcelain Berl saddles in such a way that the tops 
 of the thimbles were higher than the maximum height of the solvent in the extrac- 
 tion tube. Three boiling flasks, one containing 150 ml of pyridine, one 150 ml of 
 ethylenediamine, and the third 150 ml dimethlyformamide, were attached to the 
 apparatus, and the extractions carried out for 48 hours. 
 
 Thermocouples, inserted through the condensers with the hot junctions at 
 the base of the thimbles, indicated the temperature for pyridine was 60 °C, for 
 ethylenediamine 61 °C, and dimethylformamide 87 °C. 
 
 Upon completion of the extraction, it was noted that samples subjected 
 to pyridine and ethylenediamine swelled to about twice their original volume. In 
 contrast, the sample submitted to dimethylformamide did not swell; instead, the 
 volume of the sample seemed to decrease to about 85 or 90 percent of its original 
 value. 
 
 * It was found that commercially available nitrogen contained a small amount 
 of oxygen and that coal stored under such nitrogen will oxidize. To minimize the 
 effect of oxidation, all samples of coals, extracts, and residues prior to fur- 
 ther use were first submitted to vacuum, 1 or 2 mm of Hg, and then stored under an 
 C>2-free nitrogen atmosphere. 
 
PLASTIC PROPERTIES OF COAL 5 
 
 Washing the residues with water to remove the last traces of solvents and 
 drying the residue both proved unsatisfactory. Washing the residues with dilute 
 hydrochloric acid produced only slight improvement for extracts prepared with 
 pyridine and ethyl enediamine. Examination by microscope showed the residue 
 particles were somewhat coated with a brown material similar to the extract. It 
 may be postulated that each particle of residue was surrounded by a layer of sol- 
 vent that still contained traces of extract in solution; when water was added dur- 
 ing washing, the extract was precipitated on to the surfaces of the particles. 
 
 Because of the difficulties encountered during thewashingand our inability to 
 remove the last trace of solvent from the minus 200-mesh residue, it was decided 
 to carry out extraction on minus 40- to plus 60-mesh coal in the hope that by 
 decreasing the area of contact between sample and solvent, adsorption of the 
 solvent would be considerably reduced. Another factor in favor of the choice of 
 the larger particle size was that Gieseler plasticity data were based on a minus 
 40-mesh sample and a better comparison between coal and residue could be made 
 where both determinations were made on substances of comparable size. 
 
 Unfortunately, as the particle size increases the solvent action decreases. 
 To compensate for this effect the temperatures of extraction with pyridine (60 °C), 
 ethylenediamine (61 °C), and dimethyl forma mid e (87 °C) were increased to the 
 boiling points of the solvents, 114°, 115°, and 153°C, respectively. 
 
 The digestion apparatus is shown on plate 1A. To keep the coal from bump- 
 ing and splashing on the side of the liter digestion flask, a stirrer rotating at the 
 rate of 150 rpm was provided. The distance of the teflon stirrer from the bottom 
 of the flask was gauged to allow, insofar as possible, only the solvent to be agi- 
 tated, so that the possibility of reducing the particle size of the sample during 
 digestion would be minimized. 
 
 The coal studied had an air-dry moisture of 4 percent. To eliminate any 
 further drying that might influence plastic characteristics by oxidation, the diges- 
 tion was made on an air-dried sample. An air-cooled condenser was used so that 
 an anhydrous extraction could be made by removing the water at the top of the con- 
 denser. 
 
 The separation of solvent and extract from residue was made by means of 
 the filtration apparatus shown on plate IB. It consists of a 160 ml capacity funnel 
 with a fritted glass plate of fine porosity, Pyrex brand. The nominal maximum pore 
 size is 5 microns. The apparatus permits filtration under vacuum in a dry nitrogen 
 atmosphere. 
 
 The difficulties encountered in separating the residue particles from their 
 coating of solvent and extract were overcome by first washing the minus 40- to plus 
 60-mesh material with 50 ml of fresh solvent, then with 50 ml of anhydrous ether. 
 Microscopic examination of the particles after such treatment indicated the absence 
 of the brown coating material previously described. 
 
 As it is imperative to obtain solvent-free extracts, several techniques of 
 separating solvent from extract were investigated. The most common type of sep- 
 aration, precipitation of the extract by addition of a large volume of water, proved 
 unsatisfactory because 1) a gelatinous precipitate was formed that prevented 
 quantitative filtration and absolute drying of the precipitate, and 2) with a dilution 
 ratio of water to solvent-plus-extract originally set at 10 to 1, the filtrate had a 
 yellowish hue, indicating a nonquantitative precipitation of extract. When the 
 ratio was increased to 100 to 1, the filtrate remained clear, but a gelatinous pre- 
 cipitate appeared in it after 48 hours. 
 
6 ILLINOIS STATE GEOLOGICAL SURVEY 
 
 To increase the rate of precipitation and improve the precipitate, flocculat- 
 ing agents including sodium chloride, ferric chloride, thorium nitrate, methanol, 
 and ethanol were tried. Concentrations of the flocculating agents from .01 to 1 per- 
 cent showed little increase in the quality of the precipitate. Change of pH by addi- 
 tion of hydrochloric acid improved the precipitate only when the dilution ratio was 
 more than 90 to 1 . In this case, when the precipitate was washed with distilled 
 water to remove the acid, peptization occurred. 
 
 Kirkby et al. (1954) found that in certain coals a large proportion of the 
 extract produced at room temperature could be precipitated by centrifuging. In our 
 study, centrifuging for four hours at 1200 rpm allowed approximately 50 percent of 
 the estimated extract to be recovered. The remaining extract in solution proved to 
 be nonquantitatively precipitated by addition of water. 
 
 A rotating vacuum evaporator (pi. 1C) was built in our laboratories for sep- 
 aration of heat-sensitive compounds from solvents and it gave very satisfactory 
 results. The evaporating flask was about one-third immersed in water maintained 
 at a constant temperature (±.1°C) in a range from 26° to 85°C. The condensing 
 flask was immersed in water at 0° to 2°C. Through a 3-way stopcock the pressure 
 of the system could be reduced to 1 mm of Hg or a nitrogen atmosphere established. 
 
 To permit a rapid distillation of the solvents without danger of distilling the 
 extracts, control of the temperatures and pressures to which the flask was sub- 
 jected was investigated. The indexes of refraction of the solvents were used as 
 checks. 
 
 To distill 250 ml of solvent in approximately 20 minutes, the following 
 settings were necessary: 
 
 Distillation flask 
 
 Solvent Temperature °C Pressure mm/Hg 
 
 Pyridine 65 80 
 
 Ethylenediamine 66 80 
 
 Dimethylformamide 82 80 
 
 Table 1 indicates some physical properties of the solvents used during the 
 course of the preliminary study. 
 
 To compare the three solvents, extractions on representative samples of 
 the Illinois high-volatile B bituminous coal, minus 40- to plus 60-mesh size, were 
 made by the procedure described in detail below. 
 
 Ten-gram portions of coal were digested with 150 ml of each of the anhy- 
 drous solvents for exactly six hours, after which time the residues were separated 
 from solvents-plus-extracts by filtration. The extracts were in turn separated from 
 the solvents. The yields of extracts were calculated on a moisture- and ash-free 
 basis. 
 
 To obtain an indication of the relative degree to which the caking properties of 
 coal are affected by solvent extraction, volatile-matter determinations (single run) 
 
 EXPLANATION OF PLATE 1 
 A - Digestion apparatus used in experiment. 
 
 B - Filtration apparatus used to separate the residues from the solvents plus extracts. 
 C - Rotating vacuum evaporator used to separate the extracts from their solvents. 
 
PLASTIC PROPERTIES OF COAL 
 
8 
 
 ILLINOIS STATE GEOLOGICAL SURVEY 
 
 were made on the coal, on each of the three dry extracts, on each of the three 
 solvent-wet residues, and on each of the three dry residues. The relative degrees 
 of agglomeration were established by the resistance to breaking exhibited by the 
 volatile-matter buttons under manual pressure. 
 
 Table 1. - Physical Properties of Solvents 
 
 Physical 
 
 
 
 Solvents 
 
 
 properties 
 
 Pyridine 
 
 
 Ethylenediamine 
 
 Dimethylformamide 
 
 Boiling point, 
 
 114°C 
 
 
 115°C 
 
 153°C 
 
 760 mm 
 
 
 
 
 
 Viscosity, cen- 
 
 0.931 
 
 
 1.725 
 
 0.802 
 
 tipoise at 25 "C 
 
 
 
 
 
 Dipole moment, (j. 
 
 2.2 
 
 
 1.95 
 
 3.82 
 
 debye units 
 
 
 
 
 
 Dielectric con- 
 
 12.5 (20 
 
 »C) 
 
 12.9(25°C) 
 
 26.6(25°C) 
 
 stant, € 
 
 
 
 
 
 n 2 A 
 
 0.39 
 
 
 0.29 
 
 0.55 
 
 Surface tension 
 
 38.0(20° 
 
 C) 
 
 32 (?)(20°C) 
 
 35.2(25°C) 
 
 dynes/cm 
 
 
 
 
 
 Table 2. - Yields of Extraction, Agglomeration Characteristics of Extracts 
 and Residues, and Physical Properties of Solvents 
 
 
 Agglomeration characteristics 
 
 Physi 
 
 cal properties 
 
 of solvent 
 
 Yield 
 of 
 
 Residues 
 
 Extracts 
 
 2 / 
 
 Surface 
 tension: e 
 (dynes/cm) 
 
 Boiling 
 point 
 
 Solvent extract* 
 
 Solvent-wet 
 
 Dry 
 
 Dry 
 
 760 mm 
 
 Dimethyl- 19 . 4 
 
 Low 
 
 None 
 
 Strong 
 
 0.55 
 
 35.2 
 
 153°C 
 
 forma mid e 
 
 
 
 
 
 
 
 Pyridine 17.9 
 
 Medium 
 
 None 
 
 Medium 
 
 0.39 
 
 38.0 
 
 114°C 
 
 Ethylene- 16.8 
 
 Strong 
 
 Low 
 
 Medium 
 
 0.29 
 
 32.0 
 
 115°C 
 
 diamine 
 
 
 
 
 
 
 
 * Moisture- and ash-free basis. 
 
 Table 2 shows the respective yields of extracts, the agglomerating charac- 
 teristics of wet and dry residues and of dry extracts from each solvent. Some 
 physical properties of the solvents are repeated in the table for correlation. For 
 the three solvents investigated, the agglomerating characteristics of the dry resi- 
 dues were lower than those for solvent-wet residues. 
 
 Agde and Hubertus (1937) attempted to correlate the amount of solvent re- 
 tained by coal as a function of ^/e , where yi is the dipole moment ande the di- 
 electric constant of the solvent . Ostwald (1939) correlated swelling and solution 
 action of solvent on high polymers as a function of \i^/e and of surface tension. 
 
PLASTIC PROPERTIES OF COAL 9 
 
 The work of Dryden (1951), in which he classified solvents by means of four 
 simple experimental tests, indicated that in some cases it was possible to corre- 
 late the physical properties of the solvents with their solvent action on coal; in 
 other cases such correlation was not possible. 
 
 When pyridine is compared to ethylenediamine of similar boiling point 
 (table 2), the yield of extract increases as ^/e ,and surface tension increases. 
 Compared to dimethylformamide, pyridine gave a lower yield with a lower \i^-/e 
 and boiling point but with a higher surface tension. It appears that the relative 
 extraction power of pyridine-class solvents, as defined by Dryden (1950), may be 
 more closely related to the boiling points of the solvents, and to a lesser degree 
 to the dipole moments, than to any other physical properties. 
 
 Ultimate analysis with atomic H/C and O/C ratios, infrared data, and x-ray 
 scattering data, indicated that the dimethylformamide, pyridine, and ethylenedia- 
 mine extracts were similar. Gieseler plasticity study shows that in each case ex- 
 tractions render the coal nonfluid. 
 
 Extraction Procedure 
 
 On the basis of the factors governing the choice of solvent and the consid- 
 erable amount of work published by Jones and Wheeler (1915, 1916) in which they 
 claimed that the coking properties of coal depend on the presence of a pyridine- 
 soluble constituent, it was decided to choose pyridine as a solvent and to study 
 the pryidine extracts and residues. 
 
 The pyridine extracts were subsequently extracted with reagent grade chloro- 
 form, conforming to A. C. S. specifications, obtained from Merck and Company, Inc. 
 The chloroform extracts in turn were extracted with freshly redistilled pure grade 
 normal-hexane obtained from Phillips Petroleum Company. 
 
 To obtain a better basis for comparing yield of extraction, physical proper- 
 ties, and chemical nature of the extracts and residues, two coals of similar chem- 
 ical composition and plastic temperature characteristics but of different maximum 
 fluidities were chosen. 
 
 The two samples were an Illinois high-volatile B bituminous No. 5 Coal 
 from Saline County, and an Eastern high-volatile A bituminous Hernshaw Coal from 
 West Virginia. The Illinois coal, size 2 by 3 inches, was taken fresh from the 
 mine. The Eastern coal, size minus one- fourth inch, was obtained from a fresh 
 shipment at a plant in St. Louis, Missouri. Both coals were prepared in the same 
 manner described previously. 
 
 In addition, the minus 40-mesh portions were screened to obtain minus 
 40- to plus 60-mesh stock samples. Such samples were not representative of the 
 coals but were prepared specimens that permitted a better basis for comparison. 
 Half of the fresh Illinois coal minus 40- to plus 60-mesh stock sample was spread 
 in a pan to a uniform depth not exceeding one-fourth inch and left in the air at 
 36°C until oxidation completely destroyed the plastic characteristics. A period of 
 45 days sufficed for this. 
 
 Each of the three samples, the two from fresh coals plusthe oxidized sample, 
 was extracted in exactly the same manner, according to the scheme shown in the 
 flow diagram on page 10. 
 
 In order to obtain a sufficient amount of the n-hexane extract, it was nec- 
 essary to treat 100 grams of coal. The pyridine extraction was carried out on ten 
 10-gram portions of coal, two each day. The following procedure was used. The 
 
10 ILLINOIS STATE GEOLOGICAL SURVEY 
 
 Coal 
 
 pyridine at 114°C 
 
 pyridine residue pyridine extract 
 
 chloroform at 61°C 
 
 chloroform residue chloroform extract 
 
 n-hexane at 67°C 
 
 n-hexane residue n-hexane extract 
 
 digestion apparatus (pi. 1) was flushed with nitrogen for 15 minutes. Exactly 10 
 grams, weighed to the nearest 0.2 mg, of minus 40- to plus 60-mesh stock sample 
 were transferred to the digestion flask, 150 ml of anhydrous pyridine added, and 
 the stirring and heating started. After 30 minutes of boiling, the water that had 
 condensed at about half the height of the air condenser was driven out by gently 
 heating the condenser tube. Boiling was then maintained for exactly six hours. 
 
 Meanwhile, a clean evaporation flask was filled with nitrogen, weighed, 
 and connected to the filtration apparatus (pi. 1). Fifteen minutes afterthe diges- 
 tion was stopped the apparatus was cold enough to be disassembled, and the con- 
 tents of the flask were transferred into the funnel of the filtration apparatus. The 
 digestion flask was washed with 50 ml of fresh anhydrous pyridine in small amounts. 
 Care was taken to transfer the residue quantitatively into the filter. 
 
 The filtration was aided by applying vacuum to the evaporating flask. A 
 nitrogen atmosphere and the vacuum were maintained on the residue for an addi- 
 tional 15 minutes, after which the evaporator flask was transferred to the rotating 
 vacuum evaporator (pi. 1). The pyridine was removed at 85 °C under 80 mm of Hg 
 pressure. 
 
 Meanwhile the residue in the suction filter was washed with 50 ml of 
 anhydrous ether. The residue was sucked dry under a nitrogen atmosphere for an 
 additional 15 minutes, then transferred into a stainless steel tray fitted in an 
 Abderhalden type of vacuum drying apparatus. It was dried for two hours at 116°C 
 at 5 mm pressure, after which the drying chamber was flushed out by introducing 
 nitrogen and evacuating six times. The residue was left overnight at room temper- 
 ature under vacuum and weighed in a nitrogen atmosphere. 
 
 EXPLANATION OF PLATE 2 
 
 A - Gieseler plastometer used to determine the plastic characteristics of the coals 
 investigated. 
 
 B - Leitz Panphot microscope used to determine melting and solidification tempera- 
 tures. 
 
PLASTIC PROPERTIES OF COAL 
 
 11 
 
12 ILLINOIS STATE GEOLOGICAL SURVEY 
 
 When the evaporator flask appeared dry, a nitrogen atmosphere was re- 
 established, and the flask was disconnected and attached to the filtration appara- 
 tus for receiving the extract from a duplicate run. After the solvent had been re- 
 moved from the extract of the second daily run and the evaporator flask appeared 
 dry, the flask was flushed by introducing nitrogen and evacuating to about 5 to 7 
 mm pressure. This was done six times, and then the flask was disconnected from 
 the apparatus, flushed with nitrogen at atmospheric pressure while it cooled to 
 room temperature, and weighed. The extract was then chipped out of the flask and 
 transferred into a previously weighed weighing bottle filled with nitrogen. A neg- 
 ligible amount of extract remained in the flask. 
 
 The extract was further dried for two hours at 100° C and 5 mm pressure, 
 cooled to room temperature, and weighed in a nitrogen atmosphere. 
 
 The yield of the extract was calculated by means of the following formula: 
 
 100 - B 
 
 where: 
 
 Yield in %- i™ *100= A(100-B) 
 
 S S 
 
 A = weight of extract in the evaporator flask 
 
 B = a correction factor comprising the percentage loss in weight 
 during drying of the portion of extract removed from the 
 flask 
 
 S = coal sample weight 
 
 For the fresh Illinois coal.determinations of the yield of extract and plas- 
 ticity were made daily over the five-day period required to complete the extraction 
 of 100 grams. For the oxidized Illinois coal and the Eastern coal, final drying of 
 the extracts and calculation of the yields were made only at the end of the five- 
 day pyridine extraction period. 
 
 The chloroform extraction was carried out in the same manner as the pyri- 
 dine extraction, except that 10 grams of pyridine extract were used with 150 ml of 
 chloroform. After six hours of digestion, the contents of the flask were transferred 
 quantitatively into the filter apparatus, which was fitted now to a 250 ml Erlenmeyer 
 flask instead of to the evaporator flask. Digestion flask and residue were washed 
 with an additional 50 ml of chloroform. The chloroform residue was dried for three 
 hours at room temperature under vacuum. The extract was separated from the sol- 
 vent by evaporating the chloroform on a steam bath at atmospheric pressure, and 
 was then dried for three hours at room temperature under vacuum. 
 
 A weighed amount of the chloroform extract (1 to 2 grams) was digested 
 with 150 ml of n-hexane in a 250 ml round bottom digestion flask for six hours. 
 The extract and residue were obtained and dried following the procedure described 
 for the chloroform extraction. 
 
 Chemical Analysis 
 
 Because the amounts of extracts obtained were so small, micro techniques 
 were used for the ultimate analyses of extracts and residues. All oxygen values 
 were determined on an ash-free basis by subtracting from 100 the sum of carbon, 
 hydrogen, nitrogen, and sulfur. 
 
 To prevent oxidation, all samples were kept under an oxygen-free, dry 
 nitrogen atmosphere in an air-tight sample container. The Dumas method for nitro- 
 
PLASTIC PROPERTIES OF COAL 13 
 
 gen determination yielded high results. It was found that drying the sample, prior 
 to analysis, for one hour at room temperature under 2 to 5 mm of Hg gave nitrogen 
 results lower by . 2 to .6 percent, and the lower values checked with the expected 
 values based on a nitrogen balance. This may be explained by the fact that the 
 Dumas method will determine the nitrogen gas adsorbed on the coal, whereas the 
 Kjeldahl method will not. 
 
 In general, in microchemical analyses, the acceptable tolerance for values 
 of each major element is 0.3 percent, assuming the determination is made correctly, 
 For substances containing only C, H, N, and S, the sum of the elements should 
 be 100 ±0.5 percent. This applies in most cases, because it is probable that the 
 errors, in the various determinations are more or less compensating. It may be 
 expected that the oxygen values obtained by difference could be in error by ±0.5 
 percent. 
 
 Table 3 compares macro and micro ultimate analysis data for coal, pyridine 
 extract, and pyridine residue. For the most part, values obtained by the two pro- 
 cedures appear to be in reasonable agreement, thus assuring the validity of the 
 micro values for this study. 
 
 Table 3. - Ultimate Analysis by Macro* and Microt Techniques 
 (Moisture- and ash-free basis) 
 
 Item Coal Pyridine extract Pyridine residue 
 
 determined 
 
 
 Macro 
 
 Micro 
 
 Macro 
 
 Micro 
 
 Macro 
 
 Micro 
 
 Atomic 
 
 H/C 
 
 ** 
 
 .784 
 
 .743 
 
 .851 
 
 .853 
 
 .786 
 
 .787 
 
 Atomic O/C 
 
 ** 
 
 .078 
 
 .083 
 
 .050 
 
 .049 
 
 .077 
 
 .077 
 
 H 
 
 
 
 5.44 
 
 5.16 
 
 6.10 
 
 6.12 
 
 5.40 
 
 5.46 
 
 C 
 
 
 
 82.80 
 
 82.76 
 
 85.50 
 
 85.57 
 
 82.60 
 
 82.62 
 
 N 
 
 
 
 1.94 
 
 1.80 
 
 2.00 
 
 2.01 
 
 2.04 
 
 2.03 
 
 O 
 
 
 
 8.68 
 
 9.18 
 
 5.70 
 
 5.66 
 
 8.46 
 
 8.43 
 
 S 
 
 
 
 1.14 
 
 1.10 
 
 .70 
 
 .64 
 
 1.50 
 
 1 .46 
 
 * Made in accordance with ASTM methods for coal and coke D271-48. 
 
 | The improvements of these methods and a description of the combustion tube 
 
 packing for the carbon-hydrogen and the nitrogen determination were published 
 
 by Clark and Rees (1954). 
 
 **Calculated. 
 
 Plasticity Measurements 
 
 Plasticity measurements of coal, extracts, and residues were made by a 
 Gieseler plastometer (Soth and Russell, 1943). The instrument, as modified by 
 Rees and Pierron (1954) and shown on plate 2, includes a fixed position for the 
 thrust bearings in the head of the plastometer and provision for removal of the 
 decomposition gases by suction. 
 
 In contrast to coal, the residues from the extraction have no fluidity. After 
 a determination, during which a temperature of 500 °C had been attained, the barrel 
 was removed from the heating bath, cooled to room temperature, and disassembled. 
 The residues could then be poured out of the crucible, indicating that no agglomer- 
 ation had occurred. 
 
14 
 
 ILLINOIS STATE GEOLOGICAL SURVEY 
 
 To check, the possibility that return of extracts to residues might restore 
 agglomerating or fluid properties, a blend of pyridine extract and residue and a 
 second blend of dimethylformamide extract and residue were made in such a way 
 that the original percentage composition of the coal was reestablished. 
 
 Table 4 gives Gieseler plasticity data of the original coal, pyridine and 
 dimethylformamide extracts, and blends 1 and 2. The indicated maximum fluid- 
 ities of blends 1 and 2 were much higher than the maximum fluidity of the origi- 
 nal coal. However, these values will require further investigation. 
 
 Table 4 
 
 . - Gieseler 
 
 Plasticity Data for Coal and Extracts 
 
 
 
 Characteristic temperature in 
 
 °C 
 
 
 
 
 Softening 
 
 Fusion 
 
 
 
 Maximum 
 
 
 (.5 dial 
 
 (5.0 dial 
 
 Maximum 
 
 Solidifi- 
 
 
 fluidity 
 
 Sample 
 
 div /min ) 
 
 div /min ) 
 
 fluidity 
 
 cation 
 
 (dial div /min) 
 
 Coal 
 
 392 
 
 414 
 
 433 
 
 469 
 
 
 61 
 
 Pyridine extract 
 
 350 
 
 390 
 
 420 
 
 450 
 
 
 15, 000 
 
 Dimethylformamide 
 
 347 
 
 388 
 
 421 
 
 453 
 
 
 15, 000 
 
 extract 
 
 
 
 
 
 
 
 Blend 1 
 
 355 
 
 396 
 
 427 
 
 460 
 
 
 12, 000 
 
 Blend 2 
 
 356 
 
 397 
 
 425 
 
 461 
 
 
 12, 500 
 
 X-ray Investigation 
 
 Mahadevan (1929) was one of the first to study coal structure by the dif- 
 fraction method. On the basis of similarity between data obtained from coal and 
 data from graphite, he concluded that coal contains compounds having a hexagonal 
 ring structure similar to that of graphite. 
 
 Blayden et al. (1944) made detailed investigations of the x-ray patterns of 
 coals, and compared the coal carbonization products with those of cellulose, lig- 
 nin, and glycine. They also postulated a graphite-like structure for coal based 
 on the fact that data indicated the presence of interplanar spacing d(002) of the 
 same order, 3. 5 A, as that between successive layers in a graphite crystal. In 
 addition, they suggested other lamellar structures that could also stack with a 
 distance of 4.5 A between layers. 
 
 Nelson (1954), by using improved techniques, was able to calculate the 
 statistical distribution of atoms around any arbitrary center in coal. Hirsch (1954) 
 derived the average number of aromatic rings from the average diameter of the la- 
 mellae. Whitaker (1955), in his review of the ultimate structure of coal, stated 
 that although the x-ray methods allow deductions somewhat indirectly related to 
 the structure, they allow no direct interpretations. This was probably because of 
 the very diffuse nature of the halos produced by essentially noncrystalline solids, 
 in comparison with sharply resolved peaks in diffraction patterns of crystalline 
 solids. Nevertheless, if x-ray study shows the same trend or indication as do 
 other techniques, its use may be justified. We therefore decided to obtain x-ray 
 scattering data for each sample investigated by means of a General Electric Unit 
 XRD-3, copper anode tube, Ni filter, 50KV-15 ma, equipped with a 1° slit and a 
 .2° screen. 
 
PLASTIC PROPERTIES OF COAL 15 
 
 A brief preliminary study of the effect of particle size on scattering data 
 from 5° to 45° scattering angle was made. Three particle sizes (minus 60- minus 
 100- and minus 200-mesh) for the coal, for the pyridine and dimethylformamide 
 extracts, and for the residues were prepared. Each sample was mounted by press- 
 ing the powder in a 1 by j-inch sample holder. To permit a quantitative comparison 
 between sizes, care was taken to pack the samples as uniformly as possible. It 
 was found that when a minus 100-mesh sample was used, a better resolution of the 
 scattering curve was obtained. In general, there was an indication that the extracts 
 possessed a higher degree of oriented arrangement than the original coal. 
 
 The period of 3.5 A (25.4° scattering angle) is related to stacking of aro- 
 matic compounds. In coal this distance represents the space between two con- 
 densed aromatic clusters (one on top of the other, for example). Similarly, the 
 period of 4.5 A (19.7° scattering angle) represents the space between two aliphatic 
 chains parallel to each other. It is possible that this 4.5 A spacing may be related 
 to the special lamellar structures reported by Blayden et al. (1944). 
 
 The scattering curve of a high-volatile-matter bituminous coal shows a 
 maximum in the region from 17.5° to 27° scattering angle. The measurement of the 
 relative intensities at 25.4° and 19.7° may be assumed to indicate the relative 
 amounts of aromatic and aliphatic materials present in the substance analyzed. 
 Following accepted procedure, the base lines from which relative intensities were 
 measured were determined by a straight line intercepting the scattering curves at 
 10° and 35°. Table 5 gives the ratios 'calculated on the basis of these assump- 
 tions. 
 
 Intensity at 19.7° Aliphatic 
 
 Intensity at 25 .4° Aromatic 
 
 Table 5. - X-Ray Scattering Ratios 
 
 Pyridine Dimethylformamide 
 
 Aliphatic 
 Aromatic 
 
 Coal Extract Residue Extract Residue 
 
 1.24 2.06 1.73 2.32 1.73 
 
 Because the preliminary study indicated a relative increase in the alipha- 
 ticity of the extracts, scattering data were obtained for each coal and for the 
 extracts and residues of the three series investigated. 
 
 The samples were prepared to minus 100-mesh, packed uniformly into the 
 holder, and subjected to analysis. The n-hexane extract, which is a viscous 
 liquid at room temperature, was heated to about 60 °C, and smeared uniformly to a 
 thickness of 2 mm on a glass slide. The slide was placed on dry ice for a few 
 minutes to make the liquid so viscous that no difficulties were encountered during 
 the determination. It was found that subjecting the n-hexane extract to x-ray 
 study at different temperatures had no detectable effect on the resulting scattering 
 curves. 
 
 Infrared Investigation 
 
 Use of infrared methods to study coal originated in England with the work 
 of Sutherland et al. (1944). Since that time most laboratories studying coal con- 
 stitution or related subjects have published much data. Cannon and Sutherland 
 
16 ILLINOIS STATE GEOLOGICAL SURVEY 
 
 (1945) assigned some of the aosorption bands to specific chemical bonds. Cannon 
 (1953) later made further assignments such as oxygen-containing groups, CH2 and 
 CH3 groups, and both single-ring and condensed aromatic structures. Vucht et al . 
 (1955) summarized their conclusions, together with the findings of other investiga- 
 tors, in the form of a table that assigns wave lengths of the absorption bands to 
 specific molecular vibrations. 
 
 Several techniques have been developed for the infrared examination of 
 coal. Cannon and Sutherland (1945), and later Orchin et al. (1951), used success- 
 fully the "thin section technique." Gordon et al. (1952) and Cannon (1953) adopted 
 the Nujol mull method in which coal is ground with Nujol. This method has two 
 serious disadvantages. First, to keep the effect of light scattering as small as 
 possible, the coal particle size must be reduced to a minimum; and second, the 
 resultant spectrum curve contains absorption bands characteristic of the Nujol, 
 thus limiting the interpretation to certain regions of the coal spectra. 
 
 The latest technique was developed by Schiedt (1953) and used most ad- 
 vantageously by Vucht et al. (1955) and Friedel (1956). For this method the coal 
 is ground with dry potassium bromide and subsequently compressed under high 
 pressure and vacuum to form translucent pellets. 
 
 Preliminary studies using the Nujol technique indicated that, as the length 
 of time for which the coal is ground with Nujol increases (that is, particle size 
 decreases), the spectrum background decreases, especially in the frequency 
 region 4000 to 2000 cm -1 . 
 
 Early in the preliminary work it became apparent that the relative alipha- 
 ticity might be investigated, and therefore the potassium bromide pellet technique 
 was used. It was found that grinding a minus 60-mesh sample with KBr for 45 min- 
 utes was sufficient to obtain satisfactory resolution of the bands. By studying the 
 ratio between extinction coefficients of the bands of aromatic and aliphatic CH 
 groups, Brown (1955) estimated within limits the relative proportions of the two 
 forms of hydrogen in coals. He selected the 3.30(jl or 3030 cm"^ band for the aro- 
 matic CH stretching vibration, and the 3.43(j. or 2910 cm~l band for the aliphatic 
 CH stretching vibration. 
 
 In our study the sample considered had a very small band in the 3030 cm~l 
 region, and in some instances this band was absent, probably because of the 
 relatively small amount of aromatic CH stretching, or the interference of the hydro- 
 gen bonding. We decided, therefore, to choose another band as representative of 
 aromatic structure. The band at 1600 cm"\ common to all coals, was chosen. 
 This band was assigned to single and condensed aromatic ring structures by Brown 
 (1955). It may, however, be influenced by conjugated carbonyl groups that probably 
 are hydrogen-bonded, and by phenoxy structures. The width rather than the height 
 of the band, especially toward the higher wave number region, probably would be 
 most affected by such interferences. We decided to measure the height of the 
 bands at 2910 and 1600 cm - l as an indication of the relative amounts of aliphatic 
 and aromatic materials present in the samples analyzed. 
 
 A Perkin-Elmer, Model 21, double beam, recording infrared spectrophoto- 
 meter was used. The operating conditions of the instrument were: prism, NaCl; 
 resolution, 927; response, 1; gain, 5.5; speed, 7; and suppression, 2. The 
 samples were prepared for analysis in the following manner: 0.0025 grams, weighed 
 to the nearest 0.1 mg, of the minus 100-mesh sample, were transferred into a mortar 
 containing exactly 1 gram of dried potassium bromide. The mixture was ground by 
 hand for ten minutes, and approximately 200 mg of the mixture was transferred to 
 the pelleting machine. A pressure of 14,000 lb. per square inch for five minutes, 
 
PLASTIC PROPERTIES OF COAL 17 
 
 and a vacuum of 2 to 3 mm of Hg were necessary to obtain circular pellets 13 mm 
 in diameter and .5 mm thick. For the n-hexane extracts, a 0.5 percent solution of 
 the sample in carbon tetrachloride was used. This concentration of extract gave a 
 1600 cm - -'- band of approximately the same height as the one obtained when hexane 
 residue in potassium bromide was used. 
 
 Adduct Investigation 
 
 To determine whether the increase in aliphaticity was due to selective ex- 
 traction of long-chain aliphatic materials, such as waxes, or short chains attached 
 to condensed structures, the urea and thiourea complex technique was used. 
 
 Bengen (1940) in Germany discovered the urea adduct formation. He found 
 that urea forms adducts with a variety of straight-chain compounds such as aliphatic 
 hydrocarbons, acids, alcohols, aldehydes, ketones, and esters. The adducts so 
 formed could be decomposed into their components by heating, or by dissolving the 
 urea. Zimmerschied et al . (1950) confirmed and extended the work of Bengen. Smith 
 (1952) made a complete structure determination on single crystals of urea adduct. 
 Angla (1949) investigated the thiourea complex formation. Truter (1953) gave a brief 
 survey of the literature on the subject, in which eight references are given. 
 
 The main difference between the urea and thiourea adducts is that the chan- 
 nel in the thiourea complex is larger than the one in the urea. Thiourea forms com- 
 plexes with branched-chain or some cyclic types of compounds. The straight-chain 
 molecules have too small a cross section to form complexes spontaneously (that is, 
 they slip out of the channels too easily), but do so when anchored by a branched- 
 chain or cyclic group. 
 
 Nearly saturated solutions of urea and thiourea in methyl alcohol were added 
 slowly in small amounts to a solution of pyridine extract in pyridine and a solution of 
 chloroform extract in chloroform. Precipitation occurred immediately. After addition 
 of approximately 50 grams of urea or thiourea to each gram of extract, the precipi- 
 tates were allowed to settle out for one hour, and were then filtered and dissolved 
 in water. The water solution was inspected for traces of oily material and then 
 filtered. It was found that thiourea did not form an adduct, but with urea approxi- 
 mately .5 percent of the pyridine extract formed an adduct. Data from infrared 
 studies indicated that the material so obtained was aliphatic. Owing to the very 
 low yield of extraction of this material, the adduct approach was considered not 
 worth further study. 
 
 Microscopic Examination 
 
 The quantities of extracts and residues produced during the experiment were 
 so small that plasticity data could not be obtained by the Gieseler plastometer. 
 Hot- stage microscopic examinations were used to determine melting and relative 
 plastic range temperatures. 
 
 The Leitz Panphot microscope, equipped with a Leitz high-temperature heat- 
 ing stage (pi. 2), was used. The temperature was measured by a calibrated platinum- 
 platinum 10 percent rhodium thermocouple, with the hot junction in contact with the 
 quartz window on which the sample was placed. The rate of heating was maintained 
 at 3°C per minute. It was found that oxygen- free dry nitrogen must flow through the 
 hot stage to prevent oxidation of the material. In general, the melting points of the 
 substances studied were 10° to 20 °C higher in an air atmosphere than in an inert 
 atmosphere. The choice of a 30X hot-stage objective and a 4X eyepiece proved 
 
18 ILLINOIS STATE GEOLOGICAL SURVEY 
 
 satisfactory. Visual inspection was made through the viewer at the top of the micro- 
 scope; photographs were taken at specific intervals. 
 
 Differential Thermal Analysis 
 
 Differential thermal analysis was used only to estimate the relative temper- 
 ature ranges in which coal, extracts, and residues underwent thermal decomposition. 
 Clegg (1955) made a careful study of the shifting of peak and shape of curve when 
 amounts of inert diluents in the sample, particle size, rate of heating, packing, and 
 other variables were changed. 
 
 In the present study it was found that, as the extracts possess such high 
 percentages of volatile matter, dilution with alundum was necessary. Mixtures of 
 25 percent sample with 75 percent diluent, loosely packed, proved best for compar- 
 ative purposes. The samples were covered to minimize oxidation, and duplicate 
 determinations were made to ascertain that results could be duplicated when a uni- 
 form technique was used. 
 
 Molecular Weight Determination 
 
 The boiling point elevation (ebullioscopic) method was used to determine 
 molecular weight. The apparatus developed and improved by Rieche was used but 
 was further improved by providing the boiling chamber with electric instead of gas 
 heating. 
 
 The solvents used were pyridine, chloroform, and n-hexane for their respec- 
 tive extracts. The ebullioscopic factor for a known volume (4.5 ml) of each solvent 
 was determined by measuring the increase of boiling point when a known amount of 
 azobenzene (mol.wt 182.22) was added. The following formula was used to calcu- 
 late the factor: 
 
 K , = 182.22 x AT 
 solv. 
 
 where 
 
 AT is the increase of the boiling point of the 4.5 ml of solvent in degrees C 
 S is the weight of azobenzene 
 
 The samples were pressed into small pellets of 25 to 30 mg each, weighed 
 accurately on a microbalance, and introduced into the boiling chamber after a con- 
 stant boiling temperature of the solvent had been established. The increase inboil- 
 ing point, AT, was then noted. Each determination was made in triplicate, and the 
 average reported. The solvents used were as follows: pyridine for pyridine extracts 
 and chloroform residues; chloroform for chloroform extracts and n-hexane residues; 
 and n-hexane for hexane extracts. 
 
 RESULTS 
 
 Table 6 gives the chemical analyses made on a moisture- and ash- free basis, 
 the free swelling indexes, and the Gieseler plasticity dataforthe samples investi- 
 gated—Illinois high-volatile B bituminous, oxidized Illinois high-volatile B bitu- 
 minous, and Eastern high-volatile A bituminous coals. 
 
Illinois State Geological Survey 
 
 Circular 269 — Plate 3 
 
 
 
 >7* 
 
 PJiHHI 
 
 
 • 
 
 3 j 
 
 | 
 
 . » 
 
 -f 
 
 tsW^ 
 
 5r 
 
 /c^ 
 
 25°" * 
 
 7 
 
 * 
 
 
 PYRIDINE 
 
 CHLOROFORM 
 
 Hot-stage Photomicrographs of Extracts 
 
 Illinois high-volatile B bituminous coal series: temperatures, in degrees C 
 
 are those of extracts when the photographs were taken. (100X) 
 
Illinois State Geological Survey 
 
 Circular 269 — Plate 4 
 
 * "rfV- i 
 
 *? .;. 
 
 
 ^1 
 
 a. 
 
 
 % 
 
 #•* 
 
 550° ilMfeSj „%*Ti ■•• 5oo 0> . ' Jg\- ;^K ►h5* v 
 
 PYRIDINE 
 
 CHLOROFORM 
 
 N-HEXANE 
 
 Hot-stage Photomicrographs of Residues 
 
 Illinois high-volatile B bituminous coal series; temperatures, in degrees C, 
 
 are those of residues wnen the photographs were taken. (100X) 
 
PLASTIC PROPERTIES OF COAL 
 
 19 
 
 Table 6. - Chemical Analyses, Free Swelling Indexes, and 
 Gieseler Data for Coals Investigated 
 (Moisture- and ash-free coal basis) 
 
 
 Illinois 
 
 Oxid. Illinois 
 
 Eastern 
 
 
 HVBB 
 
 HVBB 
 
 HVAB 
 
 Proximate* 
 
 
 
 
 Volatile matter 
 
 40.3 
 
 40.5 
 
 41.3 
 
 Fixed carbon 
 
 59.7 
 
 59.5 
 
 58.7 
 
 Ultimate* 
 
 
 
 
 Hydrogen 
 
 5.51 
 
 5.45 
 
 5.74 
 
 Carbon 
 
 83.41 
 
 82.62 
 
 86.45 
 
 Nitrogen 
 
 1.86 
 
 1.76 
 
 1.69 
 
 Oxygen 
 
 7.99 
 
 8.93 
 
 5.22 
 
 Sulfur 
 
 1.23 
 
 1.24 
 
 .90 
 
 Calorific value: Btu/lb 
 
 14658 
 
 14579 
 
 15510 
 
 Free swelling index* 
 
 6 
 
 1 
 
 5i 
 
 Gieseler plasticity 
 
 
 
 
 Softening temperature °C 
 
 384 
 
 
 364 
 
 (0 . 5 dial div /min ) 
 
 
 >. 
 
 
 Fusion temperature °C 
 
 406 
 
 
 395 
 
 (5 .0 dial div /min ) 
 
 
 3 
 
 
 Max. fluidity temperature °C 
 
 431 
 
 £ 
 
 430 
 
 Solidification temperature °C 
 
 462 
 
 O 
 
 5 
 
 485 
 
 Max. fluidity 
 
 80 
 
 
 37, 500 
 
 (dial div /min ) 
 
 
 
 
 * Made by ASTM methods. 
 
 The daily yield of pyridine extract for the Illinois coal, on a moisture- and 
 ash-free coal basis, and the average of duplicate maximum Gieseler fluidity values 
 are reported in table 7. 
 
 Table 7. - Yields of Extract and Maximum Fluidity of 
 Nonoxidized Illinois HVBB Coal 
 
 Daily extraction 
 
 Max. fluidity 
 (dial div/min) 
 
 Pyridine extract in percent* 
 
 I 
 
 II 
 
 III 
 
 IV 
 
 V 
 
 Average 
 
 126 
 90 
 68 
 65 
 51 
 
 80 
 
 22.0 
 21.3 
 
 20, 
 
 19 
 
 19, 
 
 20.7 
 
 * Moisture- and ash-free coal basis, 
 
 Only the averages of maximum fluidity and yield of extract are used for 
 comparison in the discussion. 
 
20 
 
 ILLINOIS STATE GEOLOGICAL SURVEY 
 
 Table 8. - Yields of Extract, Free Swelling Index 
 and Maximum Fluidity of the Three Coals 
 
 
 Illinois 
 
 HVBB 
 
 
 Eastern HVAB 
 
 
 Nonoxidized 
 
 Oxidized 
 
 
 Gieseler 
 
 
 
 
 
 max. fluidity 
 
 
 
 
 
 (dial div /min ) 
 
 80 
 
 
 
 
 37,500 
 
 Free swelling index 
 
 6 
 
 
 1 
 
 5* 
 
 Yield in percent 
 
 
 
 
 
 (l)* Pyridine extract 
 
 20.7 
 
 
 13.6 
 
 23.8 
 
 Chloroform extract 
 
 5.8 
 
 
 5.1 
 
 9.0 
 
 n-hexane extract 
 
 2.4 
 
 
 .7 
 
 3.4 
 
 (2)* Chloroform extract 
 
 .9 
 
 
 .8 
 
 1.3 
 
 n-hexane extract 
 
 none 
 
 
 none 
 
 none 
 
 (1) Following the previously described extraction scheme. 
 
 (2) Submitting coal directly to chloroform and n-hexane extractions, 
 
 * Moisture- and ash-free basis 
 
 Table 8 gives for each of the coals considered the Gieseler maximum fluid- 
 ity, the free swelling index, and the yields of extract in percent obtained by the 
 pyridine scheme of extraction previously described and by chloroform and n-hexane 
 extractions applied directly to the coal. 
 
 Tables 9, 10, and 11 summarize the ultimate analysis data obtained by micro 
 technique on a moisture- and ash-free basis for the three coals considered, and 
 for the respective extracts and residues. 
 
 Table 9. - Illinois HVBB Nonoxidized Series 
 Ultimate Analysis Data Obtained by Micro Technique 
 (Moisture- and ash-free basis) 
 
 
 Coal 
 
 Pyridi 
 
 ne 
 
 Chloroform 
 
 n-hi 
 
 sxane 
 
 
 Extract 
 
 Residue 
 
 Extract 
 
 Residue 
 
 Extract 
 
 Residue 
 
 H 
 
 5.85 
 
 5.99 
 
 5.53 
 
 7.12 
 
 5.37 
 
 7.85 
 
 7.92 
 
 C 
 
 83.18 
 
 84.02 
 
 81.91 
 
 84.76 
 
 79.45 
 
 87.27 
 
 86.32 
 
 N 
 
 2.02 
 
 2.10 
 
 2.13 
 
 2.25 
 
 2.25 
 
 .56 
 
 2.42 
 
 O 
 
 7.25 
 
 7.17 
 
 9.04 
 
 4.80 
 
 11.70 
 
 3.32 
 
 2.67 
 
 s 
 
 1.70 
 
 .72 
 
 1.39 
 
 1.07 
 
 1.23 
 
 1.00 
 
 .67 
 
 Atomic 
 
 
 
 
 
 
 
 
 H/C 
 
 .837 
 
 .849 
 
 .805 
 
 1.000 
 
 .805 
 
 1.072 
 
 1.093 
 
 Atomic 
 
 
 
 
 
 
 
 
 O/C 
 
 .065 
 
 .064 
 
 .083 
 
 .042 
 
 .110 
 
 .029 
 
 .024 
 
PLASTIC PROPERTIES OF COAL 
 
 21 
 
 Table 10. - Illinois HVBB Oxidized Series 
 Ultimate Analysis Data Obtained by Micro Technique 
 (Moisture- and ash-free basis) 
 
 
 Coal 
 
 Pyrid 
 
 ine 
 
 Chloroform 
 
 n-hexane 
 
 
 Extract 
 
 Residue 
 
 Extract 
 
 Residue 
 
 Extract 
 
 Residue 
 
 H 
 
 5.95 
 
 5.95 
 
 5.42 
 
 7.11 
 
 5.32 
 
 8.30 
 
 7.28 
 
 C 
 
 82.49 
 
 83.91 
 
 81.66 
 
 84.85 
 
 80.72 
 
 87.08 
 
 84.82 
 
 N 
 
 1.56 
 
 1.61 
 
 1.45 
 
 1.63 
 
 1.65 
 
 1.54 
 
 3.69 
 
 O 
 
 8.50 
 
 7.70 
 
 10.30 
 
 5.61 
 
 11.74 
 
 2.29 
 
 3.59 
 
 S 
 
 1.50 
 
 .83 
 
 1.17 
 
 .80 
 
 .57 
 
 .79 
 
 .62 
 
 Atomic 
 
 
 
 
 
 
 
 
 H/C 
 
 .859 
 
 .844 
 
 .791 
 
 .999 
 
 .786 
 
 1.135 
 
 1.023 
 
 Atomic 
 
 
 
 
 
 
 
 
 O/C 
 
 .077 
 
 .069 
 
 .094 
 
 .050 
 
 .100 
 
 .019 
 
 .031 
 
 Table 11. - Eastern HVAB Series 
 Ultimate Analysis Data Obtained by Micro Technique 
 (Moisture- and ash-free basis) 
 
 
 Coal 
 
 PyridJ 
 
 .ne 
 
 Chloroform 
 
 n-he 
 
 xane 
 
 
 Extract 
 
 Residue 
 
 Extract 
 
 Residue 
 
 Extract 
 
 Residue 
 
 H 
 
 5.93 
 
 6.42 
 
 5.43 
 
 6.70 
 
 5.95 
 
 8.58 
 
 6.88 
 
 C 
 
 86.66 
 
 86.75 
 
 85.67 
 
 87.64 
 
 82.76 
 
 87.99 
 
 88.45 
 
 N 
 
 1.34 
 
 1.96 
 
 1.49 
 
 1.32 
 
 1.86 
 
 1.32 
 
 2.00 
 
 O 
 
 5.13 
 
 4.14 
 
 6.33 
 
 3.29 
 
 8.60 
 
 1.59 
 
 1.90 
 
 S 
 
 .90 
 
 .73 
 
 1.08 
 
 1.05 
 
 .83 
 
 .52 
 
 .77 
 
 Atomic 
 
 
 
 
 
 
 
 
 H/C 
 
 .814 
 
 .882 
 
 .756 
 
 .911 
 
 .856 
 
 1.161 
 
 .928 
 
 Atomic 
 
 
 
 
 
 
 
 
 O/C 
 
 .044 
 
 .036 
 
 .056 
 
 .029 
 
 .078 
 
 .014 
 
 .016 
 
 Figure 1 shows x-ray scattering curves of the three series of samples 
 studied. The relative intensities at 3.5 and 4.5A periods are indicated by arrows. 
 
 Figures 2, 3, and 4 show the infrared spectra of the three series investi- 
 gated. The measure of the relative intensities at 2910 and 1600 cm~l is indicated 
 by the height of the bands. 
 
 Microscopic examination revealed that identical behavior occurred in each 
 series of coals. Table 12, therefore, summarizes the visual observations for the 
 three series in regard to melting temperatures, relative degree of fluidity, color, 
 and presence or absence of residues above temperatures of 600 °C. 
 
 Plates 3 and 4 show photomicrographs at four different temperatures for the 
 three extracts and residues of the nonoxidized Illinois coal. This series is rep- 
 resentative of all coals studied because changes occur at similar temperatures. 
 
 Differential thermal analysis data indicate that the three series of coal in- 
 vestigated are similar to each other. No interpretable differences are evident. 
 Because of this, only the Eastern coal series is shown in figure 5. 
 
22 
 
 ILLINOIS STATE GEOLOGICAL SURVEY 
 
 to 
 c 
 
 3 
 
 o 
 
 -H 
 
 6 
 
 CO 
 
 
 c 
 
 u. 
 
 ■H 
 
 <D 
 
 6 
 
 Q. 
 
 (0 
 
 X 
 
 O 
 
 W 
 
 
 
 CD 
 
 co 
 
 a 
 
 a , 
 
 o 
 
 Di 
 
 o 
 
 C 
 
 tO 
 
 o 
 
 ro 
 
 CD 
 
 -M 
 
 x: 
 
 2 
 
 <+-! 
 
 
 o 
 
 CD 
 
 
 -i-j 
 
 CD 
 
 ■4-> 
 
 rO 
 
 1 
 
 
 +-■ 
 
 CD 
 
 o 
 
 >-. 
 
 ffi 
 
 <D 
 
 
 x; 
 
 1 
 
 a 
 
 
 to 
 
 • 
 
 o 
 
 C") 
 r-l 
 
 e 
 
 
 J-' 
 
 (D 
 
 fO 
 
 i — i 
 
 t 
 
 ro 
 
 (1) 
 
 H 
 
 c 
 
 o 
 
 o 
 
 o 
 
 o 
 
 o 
 
 CO 
 
 o 
 
 o 
 
 o 
 
 o 
 
 CD 
 
 S-H 
 
 3 
 +-» 
 fO 
 
 u, 
 CD 
 
 o, 
 
 e 
 
 CD 
 H 
 
 a> 
 
 0) 
 
 D 
 
 3 
 
 •n 
 
 73 
 
 •m 
 
 -•H 
 
 to 
 
 to 
 
 CD 
 
 0) 
 
 a; 
 
 a: 
 
 to 
 
 
 
 CO 
 
 
 Cn 
 
 ■ — i 
 
 
 C 
 
 a 
 
 i-H 
 
 a) 
 ra 
 
 +j 
 
 < — i 
 CD 
 
 £ 
 
 (0 
 
 2 
 
 (0 
 
 
 
 CU 
 
 
 
 c 
 
 „ o 
 ca 
 
 V 
 
 fO 
 
 < — ( 
 
 CO 
 
 to 
 ra 
 CD 
 Ih 
 
 u 
 
 CD 
 
 Q 
 
 0) 
 
 21 
 
 
 
 4-» 
 
 CD 
 
 s 
 
 4-» 
 
 
 
 
 +j 
 
 CD 
 
 
 
 u 
 
 3 
 
 Ih 
 
 u 
 
 3 
 
 CD 
 
 o 
 
 3 
 
 
 CD 
 C 
 
 73 
 
 -rH 
 
 a, 
 
 CO 
 
 73 
 
 to 
 
 0) 
 
 o 
 
 4-1 
 
 o 
 
 1-. 
 
 CO 
 
 73 
 
 -rH 
 
 CD 
 
 C 
 
 (0 
 
 73 
 
 CO 
 CD 
 
 "r0 
 O 
 
 w 
 
 Di 
 
 o 
 
 W 
 
 « 
 
 CD 
 
 x: 
 i 
 
 w 
 
 Oi 
 
 O 
 
 
 
 O 
 
 
 
 c 
 
 
 
PLASTIC PROPERTIES OF COAL 
 
 23 
 
 CD 
 < 
 > 
 
 UJ 
 
 < 
 
 UJ 
 
 CO 
 CD 
 
 > 
 I 
 
 (/) 
 
 O 
 
 o 
 
 UJ 
 M 
 Q 
 X 
 
 O 
 
 CD 
 CD 
 
 > 
 I 
 
 o 
 
 S3UISN31NI 
 
24 
 
 ILLINOIS STATE GEOLOGICAL SURVEY 
 
 100 r 
 
 3000 
 
 1000 800 
 
 2000 1800 1600 1400 1200 
 
 FREQUENCY-CM- 1 
 Fig. 2. - Infrared spectra of the Illinois high volatile B bituminous coal series. 
 
PLASTIC PROPERTIES OF COAL 
 
 25 
 
 1200 
 
 1000 
 
 3000 2000 1800 1600 1400 
 
 FREQUENCY-CM-i 
 Fig. 3. - Infrared spectra of the oxidized Illinois high volatile B bituminous 
 
 coal series. 
 
 800 
 
26 
 
 ILLINOIS STATE GEOLOGICAL SURVEY 
 
 100 
 
 — 
 
 
 80 
 
 - 
 
 
 60 
 
 -^~ 
 
 
 40 
 
 
 
 20 
 
 
 
 80 
 
 — 
 
 
 60 
 
 
 y"-^' 
 
 40 
 
 
 
 3000 
 
 2000 
 
 1800 
 
 1600 
 
 1400 
 
 1200 
 
 1000 
 
 800 
 
 FREQUENCY-CM- 1 
 Fig. 4. - Infrared spectra of the Eastern high volatile A bituminous coal series. 
 
PLASTIC PROPERTIES OF COAL 
 
 27 
 
 COAL E.H.V.A. 
 
 PYR. EXT 
 
 PYR. RES. 
 
 CH Cl 3 EXT 
 
 CH Cl 3 RES. 
 HEX. EXT 
 
 HEX. RES. 
 
 0°C 100° 200° 300° 400° 500° 600° 700° 800° 
 TEMPERATURE 
 
 Fig. 5. - Differential thermal analysis of the Eastern high volatile A bituminous 
 
 coal series. 
 
28 
 
 ILLINOIS STATE GEOLOGICAL SURVEY 
 
 Table 13. - Ebullioscopic Molecular Weights 
 
 
 Pyridine 
 Extract 
 
 Chloroform 
 
 n-hi 
 
 sxane 
 
 Coal 
 
 Extract 
 
 Residue 
 
 Extract 
 
 Residue 
 
 Illinois HVBB 
 
 
 
 
 
 
 nonoxidized 
 
 1500 
 
 475 
 
 600 
 
 362 
 
 520 
 
 Illinois HVBB 
 
 
 
 
 
 
 oxidized 
 
 1600 
 
 490 
 
 700 
 
 380 
 
 550 
 
 Eastern HVAB 
 
 1740 
 
 565 
 
 895 
 
 447 
 
 710 
 
 Table 13 lists the average of triplicate ebullioscopic molecular weight 
 determinations of pyridine extracts, chloroform extracts and residues, and n-hexane 
 extracts and residues. Tables 14, 15, and 16 summarize for each series the atomic 
 H/C and O/C ratios, x-ray and infrared aliphaticity, characteristic temperatures, 
 and molecular weights. 
 
 Table 14. - Illinois HVBB Nonoxidized Series 
 H/C and O/C Ratios, X-ray and Infrared Aliphaticity, 
 Characteristic Temperatures, and Molecular Weights 
 
 
 Coal 
 
 Pyridine 
 
 Chloroform 
 
 n-hexane 
 
 
 Extract 
 
 Residue 
 
 Extract 
 
 Residue 
 
 Extract 
 
 Residue 
 
 H/C 
 
 .837 
 
 .849 
 
 .805 
 
 1.000 
 
 .805 
 
 1.072 
 
 1.093 
 
 O/C 
 
 .065 
 
 .064 
 
 .083 
 
 .042 
 
 .110 
 
 .029 
 
 .024 
 
 X-ray Al/Ar 
 
 .75 
 
 .84 
 
 .84 
 
 1.99 
 
 .86 
 
 2.22 
 
 1.61 
 
 IR Al/Ar 
 
 .31 
 
 .45 
 
 .33 
 
 1.00 
 
 .48 
 
 2.96 
 
 .96 
 
 Softening 
 
 
 
 
 
 
 
 
 T° C 
 
 384* 
 
 300 
 
 None*| 
 
 90t 
 
 350t 
 
 20t 
 
 170| 
 
 Decomp. 
 
 
 
 350-400 
 
 
 
 
 350-400 
 
 T° C (DTA) 
 
 400-500 
 
 375-525 
 
 425-500 
 
 400-525 
 
 450-500 375-550 
 
 450-550 
 
 Mol.wt. 
 
 
 1500 
 
 
 475 
 
 600 
 
 362 
 
 520 
 
 * Obtained by Gieseler plastometer. 
 t Obtained by microscopic examination, 
 
PLASTIC PROPERTIES OF COAL 
 
 29 
 
 Table 15. - Illinois HVBB Oxidized Series 
 H/C and O/C Ratios, X-ray and Infrared Aliphaticity, 
 Characteristic Temperatures, and Molecular Weights 
 
 
 Coal 
 
 Pyridi 
 
 ne 
 
 Chloroform 
 
 n-h 
 
 exane 
 
 
 Extract 
 
 Residue 
 
 Extract 
 
 Residue 
 
 Extract 
 
 Residue 
 
 H/C 
 
 .859 
 
 .844 
 
 .791 
 
 .999 
 
 .786 
 
 1.135 
 
 1.023 
 
 O/C 
 
 .077 
 
 .069 
 
 .094 
 
 .050 
 
 .100 
 
 .019 
 
 .031 
 
 X-ray Al/Ar 
 
 .80 
 
 .89 
 
 .89 
 
 1.19 
 
 .71 
 
 1.27 
 
 1.13 
 
 IRAl/Ar 
 
 .50 
 
 .64 
 
 .27 
 
 1.17 
 
 .50 
 
 2.16 
 
 .96 
 
 Softening 
 
 
 
 
 
 
 
 
 T°C 
 
 None* 
 
 300 
 
 None* - } - 
 
 90| 
 
 350t 
 
 20| 
 
 170T 
 
 Decomp. 
 
 400-500 
 
 375-525 
 
 350-400 
 
 400-525 
 
 450-500 
 
 375-550 
 
 350-400 
 
 T°C (DTA) 
 
 
 
 425-500 
 
 
 
 
 450-550 
 
 Mol. wt 
 
 
 1600 
 
 
 490 
 
 700 
 
 380 
 
 550 
 
 * Obtained by Gieseler plastometer. 
 
 ■j- Obtained by microscopic examination, 
 
 Table 16. - Eastern HVAB Series 
 H/C and O/C Ratios, X-ray and Infrared Aliphaticity, 
 Characteristic Temperatures, and Molecular Weights 
 
 
 Pyridine 
 
 Chloroform 
 
 n- 
 
 h exane 
 
 Coal 
 
 Extract 
 
 Residue 
 
 Extract 
 
 Residue 
 
 Extract Residue 
 
 H/C . 814 
 
 .882 
 
 .756 
 
 .911 
 
 .85 6 
 
 1.161 
 
 .928 
 
 O/C .044 
 
 .036 
 
 .056 
 
 .029 
 
 .078 
 
 .014 
 
 .016 
 
 X-ray Al/Ar .74 
 
 .91 
 
 .78 
 
 1.19 
 
 .72 
 
 1.41 
 
 1.08 
 
 IRAl/Ar .58 
 
 .86 
 
 .44 
 
 1.79 
 
 .62 
 
 2.53 
 
 1.26 
 
 Softening 
 
 
 
 
 
 
 
 T°C 364* 
 
 300 
 
 Nonet* 
 
 90t* 
 
 350t 
 
 20t 
 
 170t 
 
 Decomp. 400-500 
 
 375-525 
 
 350-400 
 
 400-525 
 
 450-500 
 
 375-550 
 
 350-400 
 
 T°C (DTA) 
 
 
 425-500 
 
 
 
 
 450-550 
 
 Mol. wt 
 
 1740 
 
 
 5 65 
 
 895 
 
 447 
 
 710 
 
 * Obtained by Gieseler plastometer. 
 
 f Obtained by microscopic examination. 
 
30 
 
 ILLINOIS STATE GEOLOGICAL SURVEY 
 
 DISCUSSION 
 
 130 
 
 100 
 
 x At I day intervals 
 
 • After 45 days oxidation 
 
 The following discussion is limited to the three specially prepared samples 
 of coal investigated by the described techniques and procedures. No effort is made 
 to project similar conclusions to other coals of similar rank lest such conclusions 
 be misleading. 
 
 Whitaker (1955) stated that dimethylformamide has a greater dispersive 
 power for coal than has ethylenediamine. Table 2 confirms this statement, and in 
 addition shows that the residue, after the dimethylformamide extraction, possesses 
 no agglomeration characteristics. In contrast, the residue after ethylenediamine 
 extraction still retains such characteristics. Similar extractions made on a series 
 of duplicate samples demonstrate that the values of yields reported are within ±. 1 
 percent. Therefore, the yields of extracts reported in table 2 show a definite in- 
 crease from ethylenediamine to dimethylformamide. 
 
 Table 3 demonstrates the duplicability of results obtained by the macro and 
 micro analytical techniques. Such comparison was necessary to determine the de- 
 gree of accuracy of the latter technique compared to the ultimate analysis of coal 
 
 made by ASTM methods D 271- 
 48. In all cases, values re- 
 ported by the two methods check 
 reasonably well, the greatest 
 difference appearing in the hy- 
 drogen values of coal. The re- 
 sults obtained by the different 
 techniques check within the ex- 
 pected tolerance for micro deter- 
 minations, 0.3 percent. 
 
 Figure 6, plotted on a 
 linear scale, demonstrates the 
 relationship between yields of 
 pyridine extract for the Illinois 
 high-volatile B bituminous coal 
 compared with the Gieseler 
 maximum fluidity. The curve 
 was derived by combining the 
 yield of pyridine extract from 
 the oxidized Illinois coal with 
 data presented in cable 7. This 
 curve confirms the findings of 
 Rees, Pierron, and Bursack 
 (1955) in which it was shown 
 that as coal oxidized the Giese- 
 ler maximum fluidities decreased. 
 In this study, as oxidization 
 increased, fluidity decreased, 
 and the yield of pyridine ex- 
 tracts also decreased. 
 
 50 
 
 12 
 
 13 
 
 14 
 
 15 16 17 18 19 
 
 % PYRIDINE EXTRACT 
 
 20 
 
 21 
 
 22 
 
 Fig. 6. - Yields of pyridine extract versus maximum 
 
 Gieseler fluidity for the Illinois high volatile B 
 
 bituminous coal series. 
 
PLASTIC PROPERTIES OF COAL 31 
 
 It is noteworthy that the oxidized Illinois coal, with no fluidity detectable 
 by the Gieseler method, still gives 13.6 percent of pyridine extract. The amount 
 of fusible material necessary to achieve a given amount of plasticizing of the mix- 
 ture may depend on the volume of the porous structure. Therefore, if this amount 
 is decreased below a certain level by oxidation, plasticity characteristics may 
 disappear, even though fusible material still remains in the structure. Dryden 
 (1957), in his study of coal fluidity versus rank by means of internal surface, sug- 
 gested a similar explanation for low-rank coals. 
 
 Figure 6 also indicates that the effects of oxidation progressively decrease 
 as the elapsed time increases. 
 
 Figure 7 shows the yields of pyridine, chloroform, and n-hexane extracts 
 as a function of the Gieseler maximum fluidities for the three series of coals in- 
 vestigated. The yields of extracts are plotted on a linear scale and the fluidities 
 on a logarithmic scale. Here again,as the fluidity increases the yield of extracts 
 increases. In considering these curves, especially for the Eastern high-volatile A 
 bituminous coal, two sources of error must be considered in determining maximum 
 fluidity: 
 
 1) At such high rates of turning of the instrument pointer, an error of 0.1 
 second will increase or decrease the value obtained by approximately 4200 dial 
 divisions per minute. 
 
 2) This coal has such high fluidity and swelling properties that it expands 
 while in the plastic stage, greatly reducing the amount remaining in the cup in con- 
 tact with the stirrer. This behavior might account in part for the very high fluidities 
 observed and might be considered the most probable source of error. 
 
 Because of these two factors, the fluidity of such coal probably may be con- 
 sidered only as greater than 15,000 dial divisions per minute. 
 
 Table 8 indicates that, for the series studied, direct extraction of coals by 
 chloroform gives much lower yields than those obtained by the described extraction 
 scheme. It also shows no yield when the direct n-hexane extraction is carried out. 
 
 Ultimate analyses presented in tables 9, 10, and 11 indicate a definite in- 
 crease in atomic H/C and a decrease in atomic O/C ratio from coal to pyridine ex- 
 tract and its subsequent extracts. In all cases, the chloroform extractions of the 
 pyridine extract produced greater chemical differences between the resulting chloro- 
 form extracts and residues than between the original pyridine extracts and residues. 
 Possibly the interaction of pyridine with coal causes solution and dispersion of 
 certain colloidal-size particles of the coal structure, whereas in comparison the 
 chloroform extraction may limit itself to a higher degree of solubility. Also, in all 
 cases, pyridine residues possess a lower H/C ratio and higher O/C ratio than their 
 parent coals. In each series investigated, the H/C ratios of the n-hexane extract 
 and residue are higher than the corresponding chloroform extract from which they 
 were obtained. This seems to indicate that solvent (n-hexane) still remains in 
 either extract or residue or in both. 
 
 Following the scheme of extraction, the nitrogen values appear to remain 
 constant in each extract and residue, except in the n-hexane residues, in which 
 they are the highest. The sulfur remains more or less constant for the Eastern coal, 
 but in the case of the Illinois coal it is selectively more concentrated in the pyri- 
 dine residue than in the pyridine extract. 
 
 The aliphatic/aromatic ratios obtained from x-ray scattering data (fig. 1, 
 tables 14, 15, and 16) increase progressively in the extraction series from coal to 
 n-hexane extract. These values for the pyridine and chloroform residues are simi- 
 
32 
 
 ILLINOIS STATE GEOLOGICAL SURVEY 
 
 37,500 
 
 x EASTERN HVAB 
 
 13 14 15 16 17 y pYR. EXT. 20 21 22 23 24 
 
 5 6 7 8 9 
 
 %CHC1 3 EXT. 
 
 
 
 12 3 4 
 
 % n-HEX. EXT. 
 
 Fig. 7. - Yields of extracts versus maximum Gieseler fluidity of the three 
 series of coal investigated. 
 
PLASTIC PROPERTIES OF COAL 33 
 
 lar to those of the parent coals, but in this respect the n-hexane residues resemble 
 the chloroform extracts from which they were derived. 
 
 The progressive increase in aliphaticity of each subsequent extract also is 
 indicated by the slight shift of the band maximum toward the 4.5 A period. In gen- 
 eral, the extracts show an increasingly higher degree of orientation, as measured 
 by the relative height of the band maximum. 
 
 The peaks at 37.4° and 43.6° scattering angle, as shown, for example, in 
 the pyridine extracts and more especially in n-hexane residues of the oxidized 
 Illinois coal and the Eastern coal, are not from the coal substance but from the 
 aluminum sample holder. 
 
 The aliphatic/aromatic ratios obtained from infrared data (figs. 2, 3, 4; 
 tables 14, 15, 16) increase progressively in the extraction series from coal to n- 
 hexane extract. The effect of the choice of the base lines in the calculation of the 
 ratios is demonstrated for the n-hexane extract in figure 4, infrared spectra, East- 
 ern coal. In this case, the difference in the height of the 1600 cm band caused 
 by using a different base line produces a difference of 0.13 in the calculated ratios. 
 Such error is much smaller than the differences found between ratios of the three 
 subsequent extracts in this series, and so is assumed to be negligible. 
 
 The spectra of the residues are similar to those of the parent coals. In the 
 case of the Illinois coal and Eastern coal the shoulder on the 1600 cm~l band, lo- 
 cated at approximately 1700 cm~l and related to a carbonyl bond, becomes more 
 pronounced in the spectra of the n-hexane extracts. It appears also that this car- 
 bonyl band is greater in the oxidized Illinois coal than in the nonoxidized Illinois 
 coal. This confirms the finding of Vucht et al. (1955) that oxidation develops a 
 relatively stronger C ■ O band. 
 
 The strong OH bands at 3400 cm"l in all spectra made by the potassium 
 bromide pellet method may be ascribed largely to water absorbed by the salt. 
 
 It may be added that in spite of differences in the chemical composition and 
 in the physical properties, as described later, infrared spectroscopy and x-ray scat- 
 tering data of extracts and residues are similar to those of coal in general. There- 
 fore, the data presented in tables 14, 15, and 16 in regard to aliphatic/aromatic 
 ratios are interpreted only on a qualitative basis. They represent a trend and not an 
 actual degree of aliphaticity. Nevertheless, both the x-ray and infrared ratios show 
 the same trend indicated by the atomic H/C ratios derived from chemical analyses. 
 
 It is thought that the pyridine extracts may have contained dispersed coal 
 particles. This is suggested by the information obtained by microscopic examina- 
 tion (table 14, plates 3,4), and by the fact that residues remained after heating 
 these extracts to 600 °C. 
 
 Noteworthy is the relatively long plastic range of the chloroform extract 
 (90 to 500°C) and of the n-hexane extract (room temperature to 400 °C) in compari- 
 son with the pyridine extract or with the coal itself. The thermal decomposition 
 temperatures indicated by the base of the major endothermic peaks shown in figure 5 
 remain constant. 
 
 The ebullioscopic molecular weights of the extracts (table 14) decrease from 
 pyridine extracts (1500-1740) to chloroform extracts (475-565) to n-hexane extracts 
 (362-447). The molecular weight of the pyridine extract may be too high if it is as- 
 sumed that this material contains insoluble dispersed colloidal particles from the 
 parent coal. In general, the molecular weights are highest for the Eastern coal and 
 lowest for the nonoxidized Illinois coal. 
 
34 ILLINOIS STATE GEOLOGICAL SURVEY 
 
 CONCLUSIONS 
 
 The following conclusions have been drawn from the studies of the three 
 series of coal by the described procedures and techniques: 
 
 1) The plastic behavior of a fluid coal, as measured by the Gieseler plasto- 
 
 meter, is destroyed by pyridine extraction. 
 
 2) The yield of extract for each solvent appears to be proportional to the fluid- 
 
 ity as measured by the Gieseler plastometer. 
 
 3) In each series, as the progressive scheme of extraction was carried out, 
 
 each extract showed a higher relative degree of aliphaticity, a lower 
 melting point, a lower molecular weight, and a relatively constant thermal 
 decomposition temperature range. 
 
 4) For each coal the pyridine, chloroform, and n-hexane extracts are similar to 
 
 the corresponding extracts of the other coals studied. 
 
 5) The oxidized Illinois coal, in spite of showing no Gieseler fluidity, still 
 
 yielded 13.6 percent of pyridine extract. The significance of this phenom- 
 enon is not yet known, but it may be tentatively interpreted as the possible 
 inability of the Gieseler plastometer to detect very low fluidity. 
 
 REFERENCES 
 
 Agde, G., and Hubertus, R., 1937, The colloidal structure of coal: Fuel, v. 16, 
 p. 288-303. 
 
 American Society for Testing Materials, 1954, ASTM standards on coal and coke 
 (with related information): Appendix III, Sept. 1954. 
 
 Angla, B., 1949, Les complexes de la thio-urea avec les composes organiques: 
 Ann. Chim. Paris, v. 12-4, p. 639-698. 
 
 Audibert, E., 1926, The transient fusion of coal: Fuel, v. 5, p. 229. 
 
 Bangham, D. H., Franklin, R. E., Hirst, W., and Maggs, F. A. P., 1949, A 
 structure model for coal substance: Fuel, v. 28, p. 231-237. 
 
 Bengen, M. F., 1940, German patent application O. Z. 12438, March 18. 
 
 Berkowitz, N., 1948, Solubility of bituminous coal in alcoholic potash: Nature, 
 
 v.' 162, p. 960. 
 Blayden H. E., Gibson, J., and Riley, H. J., 1944, An x-ray study of structure 
 
 of coals, cokes and tars: Proc. Conference on the Ultra-fine Structure of 
 
 Coals and Cokes, British Coal Utilisation Research Association, London, 
 
 p. 176-231. 
 
 Bone, W. A., and Sarjant, R. J., 1920, Researches on the chemistry of coal, Part I. 
 The action of pyridine upon the coal substance: Proc. Roy. Soc. (London), 
 v. A 96, p. 119-136. 
 
 Bone, W. A., Pearson, A. R., and Quarendon, R., 1924, Researches on the chem- 
 istry of coal. Part III. The extraction of coals by benzene under pressure: 
 Proc. Roy. Soc. (London), v. A 105, p. 608-625. 
 
 Brown, J. K., 1955, The infrared spectra of coals: J. Chem. Soc, p. 744-752. 
 
 Cannon, C. G., 1953, Infra-red spectra of coals and coal products: Nature, v. 171, 
 p. 308. 
 
PLASTIC PROPERTIES OF COAL 35 
 
 Cannon, C. G., and Sutherland, G. B. B. M., 1945, The infra-red absorption 
 
 spectra of coals and coal extracts: Trans. Faraday Soc, v. 41, p. 279-288. 
 
 Chermin, H. A. G., and Krevelen, D. W. van, 1957, Chemical structure and prop- 
 erties of coal, XVII. A mathematical model of coal pyrolysis: Fuel,v. 36, 
 p. 85-104 
 
 Clark, H. S., and Rees, O. W., 1954, Microanalysis of organic compounds con- 
 taining fluorine: Illinois Geol . Survey Rept. Inv. 169. 
 
 Clegg, K. E., 1955, Some experimental factors that modify differential thermograms 
 of bituminous coal: Illinois Geol. Survey Rept. Inv. 190. 
 
 Dryden, I. G. C, 1948, Extraction of bright coals by anhydrous ethylenediamine: 
 Nature, v. 162, p. 959. 
 
 Dryden, I. G. C, 1949, Specific solvents for lower rank bituminous coals: Nature, 
 v. 163, p. 141. 
 
 Dryden, I. G. C, 1950, Behaviour of bituminous coals towards solvents, I: Fuel, 
 v. 29, p. 197-207. 
 
 Dryden, I. G. C, 1951, Action of solvents on coals at lower temperature, I. A 
 qualitative survey of the effects of liquids upon bright coals of low rank: 
 Fuel, v. 30, p. 39-44. 
 
 Dryden, I. G. C, 1957, Chemistry of coal and its relation to coal carbonization: 
 J. Inst. Fuel, v. 30, p. 193-214. 
 
 Fischer, F., and Gluud, W., 1916, Die Ergiebigkeit der Kohlenextraktion mit 
 Benzol: Ber. deut. Chem., Ges., v. 49, p. 1460-1468. 
 
 Fitzgerald, D., 1956, The kinetics of coal carbonization in the plastic state: Far- 
 aday Soc. Trans., v. 52, p. 362. 
 
 Friedel, R. A., 1956, Infrared analysis of bituminous coals and other carbonaceous 
 materials: Anal. Chem., v. 28, p. 22-30. 
 
 Gordon, R. R., Adams, W. N., and Jenkins, G. I., 1952, Infra-red spectra of 
 coals: Nature, v. 170, p. 317. 
 
 Hirsch, P. B., 1954, X-ray scattering from coals: Proc. Roy. Soc. (London), 
 v. A226, p. 143-169. 
 
 Jones, D. T., and Wheeler, R. V., 1915, The composition of coal, III: J. Chem. 
 Soc, v. 107, p. 1318-1324. 
 
 Jones, D. T., and Wheeler, R. V., 1916, The constitution of coal: J. Chem. Soc, 
 v. 109, p. 707-714. 
 
 Kirkby, W. A., Lakey, J. R. A., and Sarjant, R. J., 1954, A study of coal extracts 
 by infra-red spectroscopy: Fuel, v. 33, p. 480-495. 
 
 Mackensie, K. G., 1909, Commercial pyridine and its value as a solvent: Ind. 
 Eng. Chem., v. 1, p. 360-362. 
 
 Mahadevan, C, 1929, X-ray study of vitrain and durain and of their constituents: 
 Fuel, v. 8, p. 462-469. 
 
 Malanowicz, L., 1932, Contribution a l'etude de Taction des bases pyridiques sur 
 les charbons: Chim. & Ind. (Paris), v. 28, p. 1277. 
 
36 ILLINOIS STATE GEOLOGICAL SURVEY 
 
 Nelson, J.B., 1954, X-ray studies of the ultra-fine structure of coal: Fuel, v. 33, 
 p. 153-175, 381-393. 
 
 Orchin, M., Golumbic, C., Anderson, J.E., and Storch, H.H., 1951, Studies of 
 the extraction and coking of coal and their significance in relation to its 
 structure: U.S. Bureau of Mines Bull. 505. 
 
 Ostwald, W. , 19 39, Swelling and solubility of high molecular materials in organic 
 solvents: J. Oil & Colour Chemists' Assoc, v. 22, p. 31-46. 
 
 Reed, F.H., Jackman, H.W., Rees, O.W., and Henline, P.W., 1952, Some ob- 
 servations on the blending of coals for metallurgical coke: Blast Furnace 
 and Steel Plant, v. 40, no. 3, p. 305-311, 344. Reprinted as Illinois Gaol. 
 Circ. 178. 
 
 Rees, O.W., and Pierron, E.D., 1954, Plastic and swelling properties of Illinois 
 coals: Illinois Geol. Survey Circ. 190. 
 
 Rees, O.W., and Pierron, E.D., 1955, Effect of diluents on the plastic properties 
 of coal as measured by the Gieseler plastometer: Illinois Geol. Survey 
 Circ. 19 7. 
 
 Rees, O.W. , Pierron, E.D., and Bursack, K.F., 1955, Effect of air oxidation on 
 the plastic properties of coal as measured by the Gieseler plastometer: 
 Trans. Illinois State Acad. Sci., v. 47, p. 97-98. 
 
 Schiedt, U., 1953, German technique for pressed potassium bromide infrared sam- 
 pling: Appl. Spectroscopy, v. 7, p. 75-84. 
 
 Sears, P. G., Wilhoit, E.D., and Dawson, L.R., 19 55,' Conductance of some 
 
 quaternary ammonium bromides and iodides in dimethylformamide at 25°: 
 J. Phys. Chem., v. 59, p. 373-375. 
 
 Smith, A. E., 1952, The crystal structure of the urea-hydrocarbon complexes: Acta 
 Cryst., v. 5, p. 224-235. 
 
 Soth, G. C. , and Russell, C. C, 1943, The Gieseler method for measurement of 
 the plastic characteristics of coal: Am. Soc. Testing Materials Proc, v. 
 43, p. 1176. 
 
 Sutherland, G. B. B. M., Fellgett, P. B. , and Willis, H. A., 1944, The infra-red 
 spectrum of coal: Proc. Conference on Ultra-fine Structure of Coals and 
 Coke, British Coal Utilisation Research Association, London, p. 330-333. 
 
 Truter, E. V., 1953, Sorting molecules by size and shape: Research (London), 
 v. 6, p. 320-326. 
 
 Vucht, H. A. van, Rietveld, B. J., and Krevelen, D. W. van, 1955, Chemical 
 
 structure and properties of coals, VIII. Infra-red absorption spectra: Fuel, 
 v. 34, p. 50-59. 
 
 Whitaker, A., 1955, The ultimate structure of coal: J. Inst. Fuel, v. 28, p. 218- 
 223. 
 
 Zimmerschied, W. J., Dinerstein, R. A., Weitkamp, A. W. , and Marschner, R. F. 
 1950, Crystalline adducts of urea with linear aliphatic compounds: Ind. 
 Eng. Chem., v. 42, p. 1300-1306. 
 
 Illinois State Geological Survey Circular 269 
 36 p. , 4 pis. , 7 figs., 16 tables, 1959 
 
nncDni 
 
 
 
 CIRCULAR 269 
 
 ILLINOIS STATE GEOLOGICAL SURVEY 
 
 URBANA ■O™