112 FRANCIS AND WHEELER THE OXIDATION OF XX.-The Oxidation of Banded Bituminous Coal at Low Temperatures. Studies in the Composition of Coal. By WILFRID FRANCIS and RICHARD VERNON WHEELER. OF the methods that have been used for studying the chemical composition of coal attack by reagents has not in general yielded much information. The majority of the reagents used have been oxidants such as nitric acid or Schultze’s solution and these either yield intractable gelatinous substances-the ulmins and their substitution products-or ultimately fatty acids usually oxali BA4NDED BITUMINOUS COAL AT LOW TEMPERATURES. 113 acid. The use of a milder oxidant such as atmospheric air a t lorn temperatures enables progressive changes in the character of the coal-substance to be studied and apart from the importance that a study of the Oxidation of coal by air h s s in relation to its spon-taneous combustion many of the observations made throw light on the character of some of the ingredients of coal and thus help towards an understanding of its constitution.The atmospheric Oxidation or " weathering " of coal has been found to increase the proportion of substances extractable by alkaline solutions-the ulmin compounds-or even to produce such substances where apparently none existed previously. For esaniple Dennstedt a i d Bunz (2. ungczc. Chem. 1905 25 1825), by heating a number of finely-powdered coals in open vessels during several days a t 130-150" showed that the effect of oxidation was, in some instances to prociizce as much as 30 to 36qb of illinin conipxnds.Jlahlcr (dnn. Mines 1913,4 163) pzsscd a slonr stream of air during long periods through coal heated a t low temperatures and found that traces of formic acid niethyl alcohol and acetone, and larger quantities of acetic acid were present in the condensed products of oxidation whilst the coals trheinselves became increas-ingly soluble in aqueous potassium hydroxide and their temperatures of initial decomposition decreased as oxidation proceeded. Ii the temperature at which oxidation was effected was higher than 250", the ulmin compounds (soluble in potassium hydroxide) formed at lowcr temperatures began to decompose. The present research has as its object the tracing of the changes that take place in the character and composition of the coal con-glomerate as oxidation slo~-ly proceeds ; and more particularly, the comparison of the behaviour of the macroscopically distinct ingredients into which most bituminous coals can readily be separ-ated when subjected to similar treatment.For inasmuch as these Laiided ingredients which Stopes has provisionally named vitrain, clarnin durain and fusain (Proc. Roy. Xoc. 1919 B 90 470) have been shown to possess markedly different chemical charactcristics even when taken from contiguous portions of the same lump of coal (see Tideswell and Wheeler J. 1919 115 619; Baranov and Francis Fuel 1922 1 219) an intimate study of the chemistry of '' coal " must of necessity treat of each ingredient separately ; whilst a comparison of the differences that exist between the ingredients which together form the coal as il; occurs in nature throws light on the manner of its formation.The coal chosen for this work was from the Top Hard seam at the East Kirkby Colliery LYottinghamshirc a detailed examination of which has been recorded by Baranov mid Francis (Eoc. cit.) 114 FRANCIS AND WHEELER THE OXIDATION OF The vitrain clarain and durain bands in this coal are sharply differ-entiated and it contains a fair proportion of fusain. No experi-ments were made with the clarain portion since previous work led us to believe that the results would be intermediate between those for the vitrain and durain portions. The method of experiment was briefly to draw a slow current of moist air through weighed samples (50 g.) of the three ingredients in the form of powder (through a 60's and on a 90's sieve) contained in glass tubes which could be maintained at a constant temperature during prolonged periods.After passing through the coal the air bubbled into a measured volume of cold distilled water to condense any liquid or soluble gaseous products. At stated intervals this water was examined and changed and a small fraction (5 g.) of each coal was removed from the tubes for analysis. Oxidation at 100". Liquid and Gaseous Products.-The treatment of the coals with air at 100" was carried out in four stages the first of two months' and the remainder each of 1 month's duration. The distilled water through which the air bubbled after leaving the coals had in each instance acquired a milky appearance* by the end of the first period and had increased in volume.Tests were made for alkalinity or acidity formaldehyde acetone acetaldehyde and methyl and ethyl alcohols. The chief results expressed approximately as per-centages on the coals are recorded in Table I. TABLE I. Formaldehyde Vitrain Methyl alcohol Acids Formaldehyde Acids Durain Methyl alcohol Formaldehyde Acids i { Period I (two months). 0.007 0.014 0.040 0.002 0.014 0-004 nil 9 9 9. Period I1 (one month). 0.01'7 0.006 0.006 0.010 0.002 trace 0.010 0.004 trace Period I11 (one month). 0*007 nil trace nil trace nil trace ?? 91 Period IV (one month). nil 7 9 9 9 9 9 9 9 7 ) 9 9 ? 9 9 9 None of the solutions contained acetone or acetaldehyde nor could any trace of ethyl alcohol be detected.The acids appeared to be either formic or acetic acid; sulphurous acid which might have been derived from pyrites or sulphur compounds in the coals, was absent nor was the acidity due to carbon dioxide in solution. The production of formaldehyde and methyl alcohol was not * This was found to be due to bacterial growths presumably derived from the coals ; it disappeared during subsequent periods of the oxidation BANDED BITUMINOUS COAL AT LOW TEMPERATURES. 115 continuous the liquids in the wash-bottles after the fourth oxidation period remaining apparently pure water ; we believe that they were produced from methane occluded in the coals probably by the action of ozone formed as the air bubbled through the wash-bottles leading to the coal samples.Changes in the Coal Substance during Oxidation.-The examina-tion of the samples withdrawn a t each period during the oxidation included proximate analyses of the usual type and by means of solvents and ultimate analyses. The results are surnrnarised in 1-i train Original 'Xoisturo 7; 9-58 Ash 1.14 Volatile matter (yo on ash-free dry coal) 30-74 Calcrific value (cals. ~ Per g.1 7940 'Moisture Ash Yo Durain Volatile ma,tter (7; on ash-free dry coal) Calorific value (cals. Per 6.) I Moisture "/b Ash yo ! ash-free dry coal) Calorific value (cals. I per g.) Volatile matter (yo on 3-58 4-08 28-91 7465 l.G6 17.7s 19.84 G4GG Period I.0.02 1.3i 31.01 6969 0.02 4-06 26-56 6973 0.10 17-57 23-69 597s Period 11. 0.us 1.6G 30.69 6615 0-35 3.95 25-00 6697 0.32 17-76 22.27 5s0 1 Period 111. 0.71 l - i B 31.16 6345 0-93 3-50 27-05 6530 0.92 17-80 20.98 57'70 Period IT. 0.98 1-76 31.06 6266 0.78 3-95 27-04 6515 0.90 17.96 20.78 6750 The original moisture in the coals was driven off during the first period of oxidation and thereafter each ingredient showed an increasing tendency to retain water or to absorb it rapidly from the air behaviour charactreristic of coals that are naturally of high osygen-content'. The resuhs of the '' volatile matter " determin-ations indicate differences in the character of the oxidation of the different portions of the same seam of coal the fusain and the ciurain presenting a peculiar contrast.The fusain showed an increase in the amount of volatile matt'er after the first period c,f Oxidation but the amount diminished as oxidation proceeded, assuming a nearly constant value higher than that of the original material; whilst with the durain there was a marked decrease at first and then a gradual increase to a constant value less than that of the original coal. With vitrain the values remained constant within the limits of experimental error with perhaps a tendency towards a slight increase. The vitrain was the only one of the thre 116 FRANCIS AND WHEELER THE OXIDATION OF ingredients that coked strongly before oxidation and its coking-power was destroyed after the fist period of heating.It will be noticed that the vitrain suffered the greatest depreciation in calorific value. TABLE 111. Analyses by Solvents. Oxidation at 100". Period Period Period Period Original. I. 11. 111. IV. a-Compounds 79-70,/ 85.6% 85.4% 87.6% 87.6% Vitrain /3-Compounds 14-6 10.4 11.0 9.6 9.9 y-Compounds 5.7 4.0 3.6 2.8 2.5 Clmins 0.02 11.17 11.25 11.3 11.2 a-Compounds 90-5 90.0 90.0 87-9 87.6 y-Compounds 2.7 2.95 - 2.9 3.0 { { Durain { j3-Compounds 6.8 7.05 - 9.2 9.4 Ulmins 0.01 2.6 5-58 6.6 6.52 a-Compounds 95.2 95.20 96-54 95.0 94.18 y -Compounds 1.4 1.45 - 1.6 1 *40 TJlmins nil 1.2 1.8 1.7 2.2 Fusain jI-Compounds 3.4 3.36 - 3.5 4.45 E'or convenience the nomenclature proposed by Stopes and Wheeler (" The Constitution of Coal," H.M.Stationery Office 1918), which makes no presumptions as to the nature of the fractions obtained by means of solvents has been employed " a-compounds " signifying that portion of coal insoluble in pyridine at its boiling point " p-compounds " that portion soluble in pyridine but iii-soluble in chloroform and '' y-compounds " that portion soluble both in pyridine and chloroform.* The term " ulmins " signifies those compounds in the coal as a whole that were soluble in alcoholic potassium hydroxide. The most striking charges on oxidation have occurred with the vitrain. Soluble ulmins were formed rapidly the change by oxid-ation at 100" being apparently complete at the end of the first period of heating and simultaneously about 30% of both the p- and y-com-pounds (amounting to 674 of the coal) was destroyed.The p-com-pounds obtained from the original coal formed a chocolate-brown powder whilst those from the oxidised coal appeared as brown nodules of dried jelly similar to the ulmins. It seemed probable, therefore that the p-compounds from the oxidised vitrain contained a relatively large proportion of the ulmins. The quantities of material available from this series of experiments did not permit of testing this assumption directly but i t was found that the 87.6% of a-compounds only contained about half the ulmins produced. * The a- 8- and y-compounds as thus defined are not necessarily similar in character in the oxidised and unoxidised coals BANDED BIT~~WINOUS COAL AT LOW TEMPERATURES.117 The 9.9% of &compounds recorded as being present in the oxidised coal must therefore be regarded as largely ulmified. With both the durain and the fusain the changes on oxidation were not so great as with the vitrain smaller amounts of ulrnins were produced and there was a tendency for the amounts of p-compounds (probably ulmified) to increase as oxidation proceeded. TABLE IV. (Expressed as percentages on the ash-free dry substances.) Ultimate Analyses. Oxidation at 100". Carbon Hydrogen Vitrain Oxygen Nitrogen Sulphur Carbon Hydrogen Nitrogen Durain Oxygen Sulphur Carbon Hydrogen Fusain \-Nitrogen 'Oygen Sulphur i i Original. 79.7 5.2 12.1 1.7 1.3 81.7 4.8 11.3 1.6 0.6 85.9 3.9 8-4 1.3 0-5 Period I.74.3 4.1 18.6 1.6 1.4 77.6 3.9 16.3 1-4 0.8 83.1 3.5 11.6 1.2 0-6 Period 11. 72-2 3.9 21.2 1-6 1.1 76.1 3.7 17.8 1.4 1.0 81.3 3.2 13.6 1.3 0.6 Analyses of d7ompounds. Carbon Oxygen etc. Carbon Oxygen etc. Carbon Oxygen etc. Vitrain Hydrogen 1' i Durain Hydrogen Original. 7 8.0 4.8 17-2 81.5 4.5 14.0 86.9 3.8 9.3 Period I. 73.2 4.1 22.7 75.5 3.7 20.8 83.5 3.4 13.1 Period 11. 73-2 3.8 23.0 75.9 3.7 20.4 82.2 3.2 14.6 Period 111. 71.9 3-8 21.3 1.8 1.2 '16.1 3.6 17-8 1.4 1.1 81.1 3.3 14-1 1.0 0.5 Period 111. 71.7 3.5 24.8 75.5 3.4 21.1 81.4 3.1 15.5 Period IV.72.1 3.8 21.1 1-8 1.2 76.5 3.7 17.1 1.5 1 *2 81.1 3.3 13.8 1.2 0.6 Period rv. 71.6 3.6 24.8 75.7 3.3 21-0 81.2 3.2 15-6 The vitrain durain and fusain from the original coal used showed the same gradations in analysis as were found by Tideswell and Wheeler for the Hamstead coal. During oxidation there was an increase of between 5 and 10% in the oxygen contents of each ingredient and a corresponding decrease in the carbon and hydrogen contents. With the vitrain since the proportion of a-compounds increased markedly on oxidation there was a tendency for their ultimate composition to approach that of the oxidised coal. The proportions of nitrogen and sulphur did not vary outside the limits of experimental error but the nature of the sulphur 118 FRANCIS AND WHEELER THE OXIDATION OF compounds present altered as is shown by determinations of " sulphate " sulphur as follows : TABLE V.Sulphur as Sulphates. Per cent. Original. Period I. Period 11. Period 111. Period IV. Vitrain nil 0.18 0.15 0.15 0-14 Durain 7 0.16 0.18 0.14 0.17 Fusain > 9 0.19 0.28 0.26 0.25 The sulphates were produced by the oxidation of pyrites originally present in the coal the fusain being the ingredient that contained most pyrites. A measure of the relative rates of oxidation of the three ingredients FIG. 1 . 1 2 3 Tirne-nwntb. can be obtained from the percentage increase in the amount of oxygen present in the samples withdrawn a t s u c c e s s i v e t i m e -intervals.The results are shown in Fig. 1 as graphs which closely resemble t'hose obtained in another manner by Tideswell and Wheeler for the rate of absorption of oxygen a t 100" by vitrain durain and fusain from Xamstead coal (J. 1920 117 '798, Fig. 3). Oxidation at 150". It was assumed that at the end of 5 months' treatment the action of oxygen on the coals a t 100" was complete an assumption warranted by the analytical data so far as the vitrain is concerned, but perhaps not quite correct with respect to the durain which as a whole was more resistant to attack. The remainder of each sample that had been oxidised at 100" during 5 months was now heated at 150" whilst moist air was drawn through. Heating was continued during 6 months samples of the coals being withdrawn a t intervals of 2 months and the liquids in the wash-bottles examined.Liquid and Gaseous Products .-The wash-bottle connected to the fusain sample contained only a solution of carbon dioxide but after each period of heating the solutions from the vitrain and durain samples were markedly acid (sulphurous acid) and containe BANDED BITUMIXOUS COAL AT LOW TEMPERATURES. 119 formaldehyde. The presence of acetone or methyl alcohol could not be detected. Changes in Weight.-During oxidation a t 100" the weight of each coal had a t first increased slightly and then decreased the h a 1 values being vitrain + 0.8 durain - 1.3 and fusain - 1.4% on the original weight. Oxidat<ion at 150" resulted in a progressive decrease in weight as is shown in Table V.TABLE V. Loss in weight on oxidation at 150". Per cent. Period I. Period 11. Period 111. Tota.1. Vitrain 8.8 7-3 0.2 16.3 Durain 4-6 5.5 3.0 13.1 Fusain 3.3 4.4 0.8 8.5 Changes in the Coal Substance.-The changes that took place in the character of the coals during oxidation at 150" are best appre-ciated by comparing the aiialytical data with those recorded for the final samples oxidised at 100". Each coal showed a marked increase in the amount of " volatile matter " and an increase in ash-content occasioned by the loss in weight of combustible matter. These results are recorded in Table VI. TABLE VI. Proximatte dnnlyses. Oxidation at 150". Final oxidation a t 100". Period I. Period 11. Period 111. Vitrain 31-0 40-7 40.4 40.5 Fusain 20.8 25-9 25.9 27.1 Vitrain 1-76 2.22 2-15 2.17 Durain 33-95 4.73 4-86 4-95 i Fusain 17.96 17.98 17.90 18.25 on ash-free dry Durain 27.0 31.8 32.1 32.2 substance % r Volatile matter Ash The most remarkable result of the oxidation at 150" was that the vitrain was rendered nearly completely soluble in alcoholic potass-ium hydroxide (compare Charpy and Decorps Compt.rend. 1921, 175 807) whilst the durain and fusain also contained greatly increased quantities of ulmins. The quantities together with the results of analysis by means of pyridine and chloroform are recorded in Table VII. It has been seen that oxidation a t 100" tended to dest,roy the compounds soluble in pyridine and t o a less degree those soluble both in pyridine and chloroform. Further oxidation at 150" apparently destroyed the y-compounds completely or at all events rendered them insoluble in chloroform; at the same time large 120 FRAXCIS AND WHEELER THE OXIDATION OF TABLE VII.Analyses by Solvents. Oxidation at 150". Final oxidation Period I. Period 11. Period 111. a-Compounds 87.6 75.9 57.7 55.0 9.9 23.2 42-3 45.0 y-Compounds 2.5 0.9 trace nil Ulmins 11.2 93.5 97.0 95.5 a-Compounds 87.6 87.4 80.3 77.0 Durain /3-Compounds 9.4 12.1 19.7 23-0 y -Compounds 3.0 0.5 trace nil Ulmins 6.5 36.1 43.5 46-0 94.2 94.5 82.0 80.0 a-Compounds 4-5 5-5 18.0 20.0 ni 1 y -Compounds 1.3 trace trace Ulmins 2.2 14-2 19.5 24.0 { proportions of the coals were rendered soluble in pyridine. It is doubtful however whether with these oxidised coals the degree of solubility in pyridine is of much significance since prolonged exposure of the coals to pyridine results in further extraction.TAELE VIII. Ultimate Analyses. Oxidation at 150". (Expressed as percentages on the ash-free dry substances.) Carbon Oxygen etc. Carbon Oxygen etc. Carbon Oxygen etc. Vitrain Hydrogen i { Durain Hydrogen Final oxidation at 100'. 72.1 3.8 24.1 76.5 3.7 19.8 81.1 3.3 15.6 Analyses 1 Carbon (Oxygen etc. Carbon Oxygen etc. Carbon Oxygen etc. Vitrain Hydrogen Period I. 66-4 2-7 30.9 71.9 2-5 25.6 77.8 2.4 19.8 of a-Compounds. 71.6 3.6 24.8 75.7 3.3 21.0 81.2 3.2 15.6 67.7 3.1 29.2 73-1 2.8 24.1 7 8.2 2.5 19.3 Period 11. 64-7 2.5 32.8 68.9 2-6 2 8-5 73-9 2.3 23.8 65-9 2.7 31.4 69-9 2.9 27.2 77.0 2.5 20-5 Period 111.64.9 2-3 32.8 67.0 2.3 30.7 74.2 2.2 23.6 64.9 2.3 32.8 75-2 2.3 22.5 Ultimate analyses were also made of some of the samples of ulmins produced on oxidation a t 150° as follows BANDED BITUMINOUS COAL AT LOW TEMPERATURES. 121 Carbon. Hydrogen. Oxygen etc. From vitrain. Period I. 64-8 2.8 32.4 From vitrain. Period I1[. 64.5 2.7 32-8 From vitrain. Period 111. 64.6 2.6 32.8 From durain. Period 111. 65.4 3-0 31.6 All these ulmins have similar analyses and the similarity extends to ulmins produced from the coals by oxidants other than air to which reference will be made later. With vitrain which was com-pletely ulmified by oxidation at 150" the analysis of the coal as a whole and that of the a-compounds contained in it were also similar.Thirty or more analyses of the fully oxidised vitrain have shown between 62 and 66% of carbon and between 2.6 and 3.5% of hydrogen. Some of these observed slight differences in analysis may be accounted for by the fact that the,ulmins are unstable compounds and suffer slight decomposition with elimination of water on prolonged heating at 150". It is apparent that marked changes have taken place in the coals during oxidation at 150". With vitrain almost the whole of the coal substance has been converted into alkali-soluble ulmins, wElst large proportions of both the durain and the fusain have been so converted. It is to this ulmin formation that the changes in the proximate and ultimate analyses and in the general character of the coals are due.The ultimate analyses show that the carbon content of the vitrain reached a minimum value owing to its almost com-plete conversion into alkali-soluble ulmins whilst the corresponding values for the durain and the fusain are higher because of the presence of substances more resistant to such ulmification or perhaps incapable of it. Oxidation at 200". Oxidation was continued at a higher temperature to find out whether complete ulmification of the durain and fusian would occur and to determine what changes if any would take place in the vitrain. The remaining portions of the samples heated at 150" were oxidised a t 200" during 1 month and examined.Each ingre-dient had lost about 25% in weight during this treatment with a corresponding increase in its percentage of ash. A k a l period of oxidation during a fortnight was then carried out. Analytical data are recorded in Table IX. firther ulmification of the durain had occurred but was by no means complete nor did it appear that more prolonged heating or heating at a higher temperature would be satisfactory for decom-position of the oxygenated coals (with the production of carbon dioxide and carbon monoxide) was taking place fairly rapidly. The ulmins produced from the vitrain a t 150" began to decompose (as is shown by the ultimate analyses) and were rendered partl 122 FRANCIS AND WHEELER THE OXIDATION O F Loss in weight yo Ash % Volatile matter (yo on ash-Ulmins Soluble in pyridine Carbon Hydrogen free dry substance) yo on ash-free dry sub-TABLE IX.Oxidation at 200". Vitrain. Period Period I. 11. 27.0 7.2 2-8 2.8 38.5 39.0 70.0 75.0 28.0 30.0 65.8 65.9 2-0 2.01 32.2 32.09 Durain. Period Period I. 11. 23.5 7.4 6.4 6.5 37.8 37.1 64-0 65.0 22.0 24.0 69.8 70-0 1-8 1.76 28.4 28.24 Fusain. Period Period I. 11. 19.7 7.0 21.4 21.6 29.8 31.2 20.5 22.0 11.0 13.0 72.8 71.3 1.8 1.6 25.4 27.1 insoluble in alcoholic potassium hydroxide ; indeed the solutions obtained were unlike those of the true ulmins being turbid and lack-ing the rich brown colour. So far as the vitrain is concerned the production of alkali-soluble ulmins must be regarded as having been completed by oxidation at 150".By carefully regulated oxidation it has thus been found possible to obtain from all the ingredients of the coal examined ulmins having similar analyses and properties and presumably derived from substances of the same nature that occur in different proportions in each ingredient. The vitrain consists almost entirely of these " reactive " compounds the durain contains a certain proportion of relatively " inert " substances and the fusain a greater proportion. These conclusions bear out and amplify the suggestion made by Tideswell and Wheeler (Zoc. cit. p. 633) that the difference chemic-ally between the vitrain clarain durain and fusain of a banded bituminous coal lies in the proportion of relatively " inert '' or unresponsive material with which the " reactive " portion is associated the " reactive " group of compounds being of the same chemical nature in each ingredient and containing more oxygen and less carbon than the " inert " material which may differ in character as between one ingredient and another.The present research specifies as the " reactive " group of compounds those that can be converted by oxidation into alkali-soluble ulmins. The " inert " substances are those that remain after a durain or a clarain has suffered all the ulmification possible by oxidation and i t would appear that they impart to the coal most of the specific qualities i t may possess. For examination of the residue from the durain after ulmification had been completed showed that i t consisted essentially of the plant entities (chiefly spore-exines) that originally resistant to the processes (bacterial and other) that resulted in the formation of the coal conglomerate still remained resistant to oxidation BANDED BITUMINOUS COAL AT LOW TEMPERATURES.123 This process of formation of alkali-soluble ulmins by oxidation (morc properly a process of regeneration) thus affords means which we have long sought of effecting a separation of the different plant entities from the coal mass and examining their chemical properties, according to the scheme outlined by Stopes and Wheeler (" The Constitution of Coal," H.M. Stationery Office p. 41). The process of oxidation by air at 150" is a long one and only admits of the troatment of small quantities of coal.We therefore made com-parat'ive experiments with other oxidants the action of which could-be regulated and found in hydrogen peroxide a reagent which if carefully uscd would rapidly effect the complete transformation of a vitrain into alkali-soluble ulmins and whilst rendering the bulk of a clarain or a durain soluble would leave their characteristic plant tissues substantially unaltered. The analysis of ulmins prepared by thc action of hydrogen peroxide on coal averages carbon 64.3, hydrogen 3.5 and oxygen 31.2% which is in close agreement with that of the similar compouiids produced by slow atmospheric oxidation a t 150". We have spoken of the formation of alkali-soluble ulmins from bituminous cod as a " regeneration." The solubility of the ulmin-compounds in alkaline solutions to form dark brown liquors is usually regarded as one of their chief characteristics and this property is used for their estimation.From estimations made in this way i t is found that the various groups of alkali-soluble ulmin-compounds form a large proportion of the " humus '' of decayed wood ; they occur sometimes in considerable quantities iii peats ; lignites also contain them but in normal bituminous coals they occur rarely, and then only in minute quantities whilst in anthracites their presence has never been detected. In other words the further the natural " carbonisation " or" the fuel has progressed the smaller is its content of soluble ulmins. From this i t is probable that the ulmin compounds once formed during the early processes of formation of the fuels have been subsequently so modified as to render them iunrecognisable by their solubility in alkalis.Berthelot and And& (Ann. Chirn. Phys. 1892 (vi] 25 420) observed that both ulmins formed artificially from sugar and those occurring naturally in soil began to decompose on exposure to air when moist carbon dioxide being evolved; Roger and Vulquin (Compt. rend. 1908 147 1404) made a similar observation on peat ulmins and noted also that they became insoluble in alkalis ; whilst Tidesm-ell and Wheeler have recorded in their study of dopplerite (J. 1922 121 2345) that this undoubted ulmin is less readily soluble in alkalis than that in more recent peats. Solubility in alkalis to form brown solutions should not therefore be regarded as a necessary property of th 124 THE OXIDATION OF BANDED BITUMINOUS COAL ETCL ulmins but it must be recognised that they can (through polymeris-ation and dehydration) change in character so as to lose their “ characteristic ” solubility.Neither the insoluble ulmins which we consider form the major part of newly-won bituminous coals nor the soluble ulmins regener-ated from them by oxidation are identical in character with those that form the major part of peat but they possess the same nuclear structure. The external groupings of the ulmin molecules in unoxidised coal are modified during oxidation the more easily detached groupings being eliminated to form simple oxygenated compounds whilst the residual ulmin becomes more definitely of acidic character (more so than the peat ulmins) owing to the form-ation of carboxylic groupings in place of those detached.Pearson (PueZ 1924 3 297) has reached similar conclusions to these as the result of his work on the oxidation of coals by sulphuric acid and bromine although the analyses of the ulmins so formed differ from ours (obtained either by atmospheric Oxidation or by the use of hydrogen peroxide) no doubt because chemical changes other than oxidation occurred during his experiments. Bituminous coal then consists essentially of insoluble ulmins in which morphologically organised plant tissues (that have escaped ulmification) are dispersed.” The present work shows that alkali-soluble ulmins can be regenerated from a bituminous coal by mild oxidation.If the coal is a vitrain i t consists almost solely of ulmins ; whilst the regeneration of alkali-soluble ulmins in a clarain, a durain or a fusain leaves residues characteristic of each of those ingredients which are essentially the original plant entities that were particularly resistant to or were preserved from the processes of “ decay ” that constituted the early stages of coal-formation and are apparently but little altered. The character of these residues is illustrated in Fig. 2 in which (a) and ( b ) are photomicrographs of cuticular tissues such as can be separated from a clarain by treatment with hydrogen peroxide and alcoholic potassium hydroxide ( c ) shows the exine of a megaspore, characteristic of the ddbbris from a durain obtained in the same manner and (d) a wood tracheid such as occurs in the ddbris of a fusain.Future communications will deal with the chemical * For the sake of simplicity we have here omitted discussion of the presence of natural plant-substances devoid of morphological organisation, such as “resin,” in the coal conglomerate. We have evidence that these resist oxidation at low temperatures and appear amongst the plant d&& after the process of regeneration of alkali-soluble ulmins by mean8 of hydrogen peroxide (or atmospheric oxidation). This work will be described in a forth-coming publication which deals also with the chemical properties of the regenerated Wins FIG. 2. ( a ) ancl ( b ) . Fragments of cuticular tissue characteristic plant entities in a clarain. ( a x 30 and b x 180.) ( c ) and ( d ) . The exine of a megaspore ( x 30) and a wood tracheid ( x 250), characteristic plant entities in 5t durain and a fusain respectively. [To face Trans. p . 124 TIDESWELL AND WHEELER ON FUSAIN AND ITS OXIDATION. 125 examination of similar plant-tissues separated in bulk from coal ; in particular with cuticles and spore-exines since these are so often the predominant structures to be found in clarains and durains, respectively and most bituminous coals consist mainly of clarain and durain. Work on modern cuticles and spore-exines for com-parison has already been in progress some years in the belief now justified that i t should be possible to isolate individual tissues or identifiable parts from coal so that for example any character-istic distillation products of which they are the source can be determined.DEPARTMENT OF FUEL TECHNOLOGY, SHEFFIJSD UNIVERSITY. [Received September 20th 1924. 112 FRANCIS AND WHEELER THE OXIDATION OF XX.-The Oxidation of Banded Bituminous Coal at Low Temperatures. Studies in the Composition of Coal. By WILFRID FRANCIS and RICHARD VERNON WHEELER. OF the methods that have been used for studying the chemical composition of coal attack by reagents has not in general yielded much information. The majority of the reagents used have been oxidants such as nitric acid or Schultze’s solution and these either yield intractable gelatinous substances-the ulmins and their substitution products-or ultimately fatty acids usually oxali BA4NDED BITUMINOUS COAL AT LOW TEMPERATURES.113 acid. The use of a milder oxidant such as atmospheric air a t lorn temperatures enables progressive changes in the character of the coal-substance to be studied and apart from the importance that a study of the Oxidation of coal by air h s s in relation to its spon-taneous combustion many of the observations made throw light on the character of some of the ingredients of coal and thus help towards an understanding of its constitution. The atmospheric Oxidation or " weathering " of coal has been found to increase the proportion of substances extractable by alkaline solutions-the ulmin compounds-or even to produce such substances where apparently none existed previously. For esaniple Dennstedt a i d Bunz (2. ungczc. Chem. 1905 25 1825), by heating a number of finely-powdered coals in open vessels during several days a t 130-150" showed that the effect of oxidation was, in some instances to prociizce as much as 30 to 36qb of illinin conipxnds.Jlahlcr (dnn. Mines 1913,4 163) pzsscd a slonr stream of air during long periods through coal heated a t low temperatures and found that traces of formic acid niethyl alcohol and acetone, and larger quantities of acetic acid were present in the condensed products of oxidation whilst the coals trheinselves became increas-ingly soluble in aqueous potassium hydroxide and their temperatures of initial decomposition decreased as oxidation proceeded. Ii the temperature at which oxidation was effected was higher than 250", the ulmin compounds (soluble in potassium hydroxide) formed at lowcr temperatures began to decompose.The present research has as its object the tracing of the changes that take place in the character and composition of the coal con-glomerate as oxidation slo~-ly proceeds ; and more particularly, the comparison of the behaviour of the macroscopically distinct ingredients into which most bituminous coals can readily be separ-ated when subjected to similar treatment. For inasmuch as these Laiided ingredients which Stopes has provisionally named vitrain, clarnin durain and fusain (Proc. Roy. Xoc. 1919 B 90 470) have been shown to possess markedly different chemical charactcristics even when taken from contiguous portions of the same lump of coal (see Tideswell and Wheeler J. 1919 115 619; Baranov and Francis Fuel 1922 1 219) an intimate study of the chemistry of '' coal " must of necessity treat of each ingredient separately ; whilst a comparison of the differences that exist between the ingredients which together form the coal as il; occurs in nature throws light on the manner of its formation.The coal chosen for this work was from the Top Hard seam at the East Kirkby Colliery LYottinghamshirc a detailed examination of which has been recorded by Baranov mid Francis (Eoc. cit.) 114 FRANCIS AND WHEELER THE OXIDATION OF The vitrain clarain and durain bands in this coal are sharply differ-entiated and it contains a fair proportion of fusain. No experi-ments were made with the clarain portion since previous work led us to believe that the results would be intermediate between those for the vitrain and durain portions.The method of experiment was briefly to draw a slow current of moist air through weighed samples (50 g.) of the three ingredients in the form of powder (through a 60's and on a 90's sieve) contained in glass tubes which could be maintained at a constant temperature during prolonged periods. After passing through the coal the air bubbled into a measured volume of cold distilled water to condense any liquid or soluble gaseous products. At stated intervals this water was examined and changed and a small fraction (5 g.) of each coal was removed from the tubes for analysis. Oxidation at 100". Liquid and Gaseous Products.-The treatment of the coals with air at 100" was carried out in four stages the first of two months' and the remainder each of 1 month's duration.The distilled water through which the air bubbled after leaving the coals had in each instance acquired a milky appearance* by the end of the first period and had increased in volume. Tests were made for alkalinity or acidity formaldehyde acetone acetaldehyde and methyl and ethyl alcohols. The chief results expressed approximately as per-centages on the coals are recorded in Table I. TABLE I. Formaldehyde Vitrain Methyl alcohol Acids Formaldehyde Acids Durain Methyl alcohol Formaldehyde Acids i { Period I (two months). 0.007 0.014 0.040 0.002 0.014 0-004 nil 9 9 9. Period I1 (one month). 0.01'7 0.006 0.006 0.010 0.002 trace 0.010 0.004 trace Period I11 (one month).0*007 nil trace nil trace nil trace ?? 91 Period IV (one month). nil 7 9 9 9 9 9 9 9 7 ) 9 9 ? 9 9 9 None of the solutions contained acetone or acetaldehyde nor could any trace of ethyl alcohol be detected. The acids appeared to be either formic or acetic acid; sulphurous acid which might have been derived from pyrites or sulphur compounds in the coals, was absent nor was the acidity due to carbon dioxide in solution. The production of formaldehyde and methyl alcohol was not * This was found to be due to bacterial growths presumably derived from the coals ; it disappeared during subsequent periods of the oxidation BANDED BITUMINOUS COAL AT LOW TEMPERATURES. 115 continuous the liquids in the wash-bottles after the fourth oxidation period remaining apparently pure water ; we believe that they were produced from methane occluded in the coals probably by the action of ozone formed as the air bubbled through the wash-bottles leading to the coal samples.Changes in the Coal Substance during Oxidation.-The examina-tion of the samples withdrawn a t each period during the oxidation included proximate analyses of the usual type and by means of solvents and ultimate analyses. The results are surnrnarised in 1-i train Original 'Xoisturo 7; 9-58 Ash 1.14 Volatile matter (yo on ash-free dry coal) 30-74 Calcrific value (cals. ~ Per g.1 7940 'Moisture Ash Yo Durain Volatile ma,tter (7; on ash-free dry coal) Calorific value (cals.Per 6.) I Moisture "/b Ash yo ! ash-free dry coal) Calorific value (cals. I per g.) Volatile matter (yo on 3-58 4-08 28-91 7465 l.G6 17.7s 19.84 G4GG Period I. 0.02 1.3i 31.01 6969 0.02 4-06 26-56 6973 0.10 17-57 23-69 597s Period 11. 0.us 1.6G 30.69 6615 0-35 3.95 25-00 6697 0.32 17-76 22.27 5s0 1 Period 111. 0.71 l - i B 31.16 6345 0-93 3-50 27-05 6530 0.92 17-80 20.98 57'70 Period IT. 0.98 1-76 31.06 6266 0.78 3-95 27-04 6515 0.90 17.96 20.78 6750 The original moisture in the coals was driven off during the first period of oxidation and thereafter each ingredient showed an increasing tendency to retain water or to absorb it rapidly from the air behaviour charactreristic of coals that are naturally of high osygen-content'.The resuhs of the '' volatile matter " determin-ations indicate differences in the character of the oxidation of the different portions of the same seam of coal the fusain and the ciurain presenting a peculiar contrast. The fusain showed an increase in the amount of volatile matt'er after the first period c,f Oxidation but the amount diminished as oxidation proceeded, assuming a nearly constant value higher than that of the original material; whilst with the durain there was a marked decrease at first and then a gradual increase to a constant value less than that of the original coal. With vitrain the values remained constant within the limits of experimental error with perhaps a tendency towards a slight increase.The vitrain was the only one of the thre 116 FRANCIS AND WHEELER THE OXIDATION OF ingredients that coked strongly before oxidation and its coking-power was destroyed after the fist period of heating. It will be noticed that the vitrain suffered the greatest depreciation in calorific value. TABLE 111. Analyses by Solvents. Oxidation at 100". Period Period Period Period Original. I. 11. 111. IV. a-Compounds 79-70,/ 85.6% 85.4% 87.6% 87.6% Vitrain /3-Compounds 14-6 10.4 11.0 9.6 9.9 y-Compounds 5.7 4.0 3.6 2.8 2.5 Clmins 0.02 11.17 11.25 11.3 11.2 a-Compounds 90-5 90.0 90.0 87-9 87.6 y-Compounds 2.7 2.95 - 2.9 3.0 { { Durain { j3-Compounds 6.8 7.05 - 9.2 9.4 Ulmins 0.01 2.6 5-58 6.6 6.52 a-Compounds 95.2 95.20 96-54 95.0 94.18 y -Compounds 1.4 1.45 - 1.6 1 *40 TJlmins nil 1.2 1.8 1.7 2.2 Fusain jI-Compounds 3.4 3.36 - 3.5 4.45 E'or convenience the nomenclature proposed by Stopes and Wheeler (" The Constitution of Coal," H.M.Stationery Office 1918), which makes no presumptions as to the nature of the fractions obtained by means of solvents has been employed " a-compounds " signifying that portion of coal insoluble in pyridine at its boiling point " p-compounds " that portion soluble in pyridine but iii-soluble in chloroform and '' y-compounds " that portion soluble both in pyridine and chloroform.* The term " ulmins " signifies those compounds in the coal as a whole that were soluble in alcoholic potassium hydroxide. The most striking charges on oxidation have occurred with the vitrain.Soluble ulmins were formed rapidly the change by oxid-ation at 100" being apparently complete at the end of the first period of heating and simultaneously about 30% of both the p- and y-com-pounds (amounting to 674 of the coal) was destroyed. The p-com-pounds obtained from the original coal formed a chocolate-brown powder whilst those from the oxidised coal appeared as brown nodules of dried jelly similar to the ulmins. It seemed probable, therefore that the p-compounds from the oxidised vitrain contained a relatively large proportion of the ulmins. The quantities of material available from this series of experiments did not permit of testing this assumption directly but i t was found that the 87.6% of a-compounds only contained about half the ulmins produced.* The a- 8- and y-compounds as thus defined are not necessarily similar in character in the oxidised and unoxidised coals BANDED BIT~~WINOUS COAL AT LOW TEMPERATURES. 117 The 9.9% of &compounds recorded as being present in the oxidised coal must therefore be regarded as largely ulmified. With both the durain and the fusain the changes on oxidation were not so great as with the vitrain smaller amounts of ulrnins were produced and there was a tendency for the amounts of p-compounds (probably ulmified) to increase as oxidation proceeded. TABLE IV. (Expressed as percentages on the ash-free dry substances.) Ultimate Analyses. Oxidation at 100". Carbon Hydrogen Vitrain Oxygen Nitrogen Sulphur Carbon Hydrogen Nitrogen Durain Oxygen Sulphur Carbon Hydrogen Fusain \-Nitrogen 'Oygen Sulphur i i Original.79.7 5.2 12.1 1.7 1.3 81.7 4.8 11.3 1.6 0.6 85.9 3.9 8-4 1.3 0-5 Period I. 74.3 4.1 18.6 1.6 1.4 77.6 3.9 16.3 1-4 0.8 83.1 3.5 11.6 1.2 0-6 Period 11. 72-2 3.9 21.2 1-6 1.1 76.1 3.7 17.8 1.4 1.0 81.3 3.2 13.6 1.3 0.6 Analyses of d7ompounds. Carbon Oxygen etc. Carbon Oxygen etc. Carbon Oxygen etc. Vitrain Hydrogen 1' i Durain Hydrogen Original. 7 8.0 4.8 17-2 81.5 4.5 14.0 86.9 3.8 9.3 Period I. 73.2 4.1 22.7 75.5 3.7 20.8 83.5 3.4 13.1 Period 11. 73-2 3.8 23.0 75.9 3.7 20.4 82.2 3.2 14.6 Period 111.71.9 3-8 21.3 1.8 1.2 '16.1 3.6 17-8 1.4 1.1 81.1 3.3 14-1 1.0 0.5 Period 111. 71.7 3.5 24.8 75.5 3.4 21.1 81.4 3.1 15.5 Period IV. 72.1 3.8 21.1 1-8 1.2 76.5 3.7 17.1 1.5 1 *2 81.1 3.3 13.8 1.2 0.6 Period rv. 71.6 3.6 24.8 75.7 3.3 21-0 81.2 3.2 15-6 The vitrain durain and fusain from the original coal used showed the same gradations in analysis as were found by Tideswell and Wheeler for the Hamstead coal. During oxidation there was an increase of between 5 and 10% in the oxygen contents of each ingredient and a corresponding decrease in the carbon and hydrogen contents. With the vitrain since the proportion of a-compounds increased markedly on oxidation there was a tendency for their ultimate composition to approach that of the oxidised coal.The proportions of nitrogen and sulphur did not vary outside the limits of experimental error but the nature of the sulphur 118 FRANCIS AND WHEELER THE OXIDATION OF compounds present altered as is shown by determinations of " sulphate " sulphur as follows : TABLE V. Sulphur as Sulphates. Per cent. Original. Period I. Period 11. Period 111. Period IV. Vitrain nil 0.18 0.15 0.15 0-14 Durain 7 0.16 0.18 0.14 0.17 Fusain > 9 0.19 0.28 0.26 0.25 The sulphates were produced by the oxidation of pyrites originally present in the coal the fusain being the ingredient that contained most pyrites. A measure of the relative rates of oxidation of the three ingredients FIG.1 . 1 2 3 Tirne-nwntb. can be obtained from the percentage increase in the amount of oxygen present in the samples withdrawn a t s u c c e s s i v e t i m e -intervals. The results are shown in Fig. 1 as graphs which closely resemble t'hose obtained in another manner by Tideswell and Wheeler for the rate of absorption of oxygen a t 100" by vitrain durain and fusain from Xamstead coal (J. 1920 117 '798, Fig. 3). Oxidation at 150". It was assumed that at the end of 5 months' treatment the action of oxygen on the coals a t 100" was complete an assumption warranted by the analytical data so far as the vitrain is concerned, but perhaps not quite correct with respect to the durain which as a whole was more resistant to attack. The remainder of each sample that had been oxidised at 100" during 5 months was now heated at 150" whilst moist air was drawn through.Heating was continued during 6 months samples of the coals being withdrawn a t intervals of 2 months and the liquids in the wash-bottles examined. Liquid and Gaseous Products .-The wash-bottle connected to the fusain sample contained only a solution of carbon dioxide but after each period of heating the solutions from the vitrain and durain samples were markedly acid (sulphurous acid) and containe BANDED BITUMIXOUS COAL AT LOW TEMPERATURES. 119 formaldehyde. The presence of acetone or methyl alcohol could not be detected. Changes in Weight.-During oxidation a t 100" the weight of each coal had a t first increased slightly and then decreased the h a 1 values being vitrain + 0.8 durain - 1.3 and fusain - 1.4% on the original weight.Oxidat<ion at 150" resulted in a progressive decrease in weight as is shown in Table V. TABLE V. Loss in weight on oxidation at 150". Per cent. Period I. Period 11. Period 111. Tota.1. Vitrain 8.8 7-3 0.2 16.3 Durain 4-6 5.5 3.0 13.1 Fusain 3.3 4.4 0.8 8.5 Changes in the Coal Substance.-The changes that took place in the character of the coals during oxidation at 150" are best appre-ciated by comparing the aiialytical data with those recorded for the final samples oxidised at 100". Each coal showed a marked increase in the amount of " volatile matter " and an increase in ash-content occasioned by the loss in weight of combustible matter. These results are recorded in Table VI.TABLE VI. Proximatte dnnlyses. Oxidation at 150". Final oxidation a t 100". Period I. Period 11. Period 111. Vitrain 31-0 40-7 40.4 40.5 Fusain 20.8 25-9 25.9 27.1 Vitrain 1-76 2.22 2-15 2.17 Durain 33-95 4.73 4-86 4-95 i Fusain 17.96 17.98 17.90 18.25 on ash-free dry Durain 27.0 31.8 32.1 32.2 substance % r Volatile matter Ash The most remarkable result of the oxidation at 150" was that the vitrain was rendered nearly completely soluble in alcoholic potass-ium hydroxide (compare Charpy and Decorps Compt. rend. 1921, 175 807) whilst the durain and fusain also contained greatly increased quantities of ulmins. The quantities together with the results of analysis by means of pyridine and chloroform are recorded in Table VII.It has been seen that oxidation a t 100" tended to dest,roy the compounds soluble in pyridine and t o a less degree those soluble both in pyridine and chloroform. Further oxidation at 150" apparently destroyed the y-compounds completely or at all events rendered them insoluble in chloroform; at the same time large 120 FRAXCIS AND WHEELER THE OXIDATION OF TABLE VII. Analyses by Solvents. Oxidation at 150". Final oxidation Period I. Period 11. Period 111. a-Compounds 87.6 75.9 57.7 55.0 9.9 23.2 42-3 45.0 y-Compounds 2.5 0.9 trace nil Ulmins 11.2 93.5 97.0 95.5 a-Compounds 87.6 87.4 80.3 77.0 Durain /3-Compounds 9.4 12.1 19.7 23-0 y -Compounds 3.0 0.5 trace nil Ulmins 6.5 36.1 43.5 46-0 94.2 94.5 82.0 80.0 a-Compounds 4-5 5-5 18.0 20.0 ni 1 y -Compounds 1.3 trace trace Ulmins 2.2 14-2 19.5 24.0 { proportions of the coals were rendered soluble in pyridine.It is doubtful however whether with these oxidised coals the degree of solubility in pyridine is of much significance since prolonged exposure of the coals to pyridine results in further extraction. TAELE VIII. Ultimate Analyses. Oxidation at 150". (Expressed as percentages on the ash-free dry substances.) Carbon Oxygen etc. Carbon Oxygen etc. Carbon Oxygen etc. Vitrain Hydrogen i { Durain Hydrogen Final oxidation at 100'. 72.1 3.8 24.1 76.5 3.7 19.8 81.1 3.3 15.6 Analyses 1 Carbon (Oxygen etc. Carbon Oxygen etc. Carbon Oxygen etc. Vitrain Hydrogen Period I. 66-4 2-7 30.9 71.9 2-5 25.6 77.8 2.4 19.8 of a-Compounds.71.6 3.6 24.8 75.7 3.3 21.0 81.2 3.2 15.6 67.7 3.1 29.2 73-1 2.8 24.1 7 8.2 2.5 19.3 Period 11. 64-7 2.5 32.8 68.9 2-6 2 8-5 73-9 2.3 23.8 65-9 2.7 31.4 69-9 2.9 27.2 77.0 2.5 20-5 Period 111. 64.9 2-3 32.8 67.0 2.3 30.7 74.2 2.2 23.6 64.9 2.3 32.8 75-2 2.3 22.5 Ultimate analyses were also made of some of the samples of ulmins produced on oxidation a t 150° as follows BANDED BITUMINOUS COAL AT LOW TEMPERATURES. 121 Carbon. Hydrogen. Oxygen etc. From vitrain. Period I. 64-8 2.8 32.4 From vitrain. Period I1[. 64.5 2.7 32-8 From vitrain. Period 111. 64.6 2.6 32.8 From durain. Period 111. 65.4 3-0 31.6 All these ulmins have similar analyses and the similarity extends to ulmins produced from the coals by oxidants other than air to which reference will be made later.With vitrain which was com-pletely ulmified by oxidation at 150" the analysis of the coal as a whole and that of the a-compounds contained in it were also similar. Thirty or more analyses of the fully oxidised vitrain have shown between 62 and 66% of carbon and between 2.6 and 3.5% of hydrogen. Some of these observed slight differences in analysis may be accounted for by the fact that the,ulmins are unstable compounds and suffer slight decomposition with elimination of water on prolonged heating at 150". It is apparent that marked changes have taken place in the coals during oxidation at 150". With vitrain almost the whole of the coal substance has been converted into alkali-soluble ulmins, wElst large proportions of both the durain and the fusain have been so converted.It is to this ulmin formation that the changes in the proximate and ultimate analyses and in the general character of the coals are due. The ultimate analyses show that the carbon content of the vitrain reached a minimum value owing to its almost com-plete conversion into alkali-soluble ulmins whilst the corresponding values for the durain and the fusain are higher because of the presence of substances more resistant to such ulmification or perhaps incapable of it. Oxidation at 200". Oxidation was continued at a higher temperature to find out whether complete ulmification of the durain and fusian would occur and to determine what changes if any would take place in the vitrain.The remaining portions of the samples heated at 150" were oxidised a t 200" during 1 month and examined. Each ingre-dient had lost about 25% in weight during this treatment with a corresponding increase in its percentage of ash. A k a l period of oxidation during a fortnight was then carried out. Analytical data are recorded in Table IX. firther ulmification of the durain had occurred but was by no means complete nor did it appear that more prolonged heating or heating at a higher temperature would be satisfactory for decom-position of the oxygenated coals (with the production of carbon dioxide and carbon monoxide) was taking place fairly rapidly. The ulmins produced from the vitrain a t 150" began to decompose (as is shown by the ultimate analyses) and were rendered partl 122 FRANCIS AND WHEELER THE OXIDATION O F Loss in weight yo Ash % Volatile matter (yo on ash-Ulmins Soluble in pyridine Carbon Hydrogen free dry substance) yo on ash-free dry sub-TABLE IX.Oxidation at 200". Vitrain. Period Period I. 11. 27.0 7.2 2-8 2.8 38.5 39.0 70.0 75.0 28.0 30.0 65.8 65.9 2-0 2.01 32.2 32.09 Durain. Period Period I. 11. 23.5 7.4 6.4 6.5 37.8 37.1 64-0 65.0 22.0 24.0 69.8 70-0 1-8 1.76 28.4 28.24 Fusain. Period Period I. 11. 19.7 7.0 21.4 21.6 29.8 31.2 20.5 22.0 11.0 13.0 72.8 71.3 1.8 1.6 25.4 27.1 insoluble in alcoholic potassium hydroxide ; indeed the solutions obtained were unlike those of the true ulmins being turbid and lack-ing the rich brown colour.So far as the vitrain is concerned the production of alkali-soluble ulmins must be regarded as having been completed by oxidation at 150". By carefully regulated oxidation it has thus been found possible to obtain from all the ingredients of the coal examined ulmins having similar analyses and properties and presumably derived from substances of the same nature that occur in different proportions in each ingredient. The vitrain consists almost entirely of these " reactive " compounds the durain contains a certain proportion of relatively " inert " substances and the fusain a greater proportion. These conclusions bear out and amplify the suggestion made by Tideswell and Wheeler (Zoc.cit. p. 633) that the difference chemic-ally between the vitrain clarain durain and fusain of a banded bituminous coal lies in the proportion of relatively " inert '' or unresponsive material with which the " reactive " portion is associated the " reactive " group of compounds being of the same chemical nature in each ingredient and containing more oxygen and less carbon than the " inert " material which may differ in character as between one ingredient and another. The present research specifies as the " reactive " group of compounds those that can be converted by oxidation into alkali-soluble ulmins. The " inert " substances are those that remain after a durain or a clarain has suffered all the ulmification possible by oxidation and i t would appear that they impart to the coal most of the specific qualities i t may possess.For examination of the residue from the durain after ulmification had been completed showed that i t consisted essentially of the plant entities (chiefly spore-exines) that originally resistant to the processes (bacterial and other) that resulted in the formation of the coal conglomerate still remained resistant to oxidation BANDED BITUMINOUS COAL AT LOW TEMPERATURES. 123 This process of formation of alkali-soluble ulmins by oxidation (morc properly a process of regeneration) thus affords means which we have long sought of effecting a separation of the different plant entities from the coal mass and examining their chemical properties, according to the scheme outlined by Stopes and Wheeler (" The Constitution of Coal," H.M.Stationery Office p. 41). The process of oxidation by air at 150" is a long one and only admits of the troatment of small quantities of coal. We therefore made com-parat'ive experiments with other oxidants the action of which could-be regulated and found in hydrogen peroxide a reagent which if carefully uscd would rapidly effect the complete transformation of a vitrain into alkali-soluble ulmins and whilst rendering the bulk of a clarain or a durain soluble would leave their characteristic plant tissues substantially unaltered. The analysis of ulmins prepared by thc action of hydrogen peroxide on coal averages carbon 64.3, hydrogen 3.5 and oxygen 31.2% which is in close agreement with that of the similar compouiids produced by slow atmospheric oxidation a t 150".We have spoken of the formation of alkali-soluble ulmins from bituminous cod as a " regeneration." The solubility of the ulmin-compounds in alkaline solutions to form dark brown liquors is usually regarded as one of their chief characteristics and this property is used for their estimation. From estimations made in this way i t is found that the various groups of alkali-soluble ulmin-compounds form a large proportion of the " humus '' of decayed wood ; they occur sometimes in considerable quantities iii peats ; lignites also contain them but in normal bituminous coals they occur rarely, and then only in minute quantities whilst in anthracites their presence has never been detected. In other words the further the natural " carbonisation " or" the fuel has progressed the smaller is its content of soluble ulmins.From this i t is probable that the ulmin compounds once formed during the early processes of formation of the fuels have been subsequently so modified as to render them iunrecognisable by their solubility in alkalis. Berthelot and And& (Ann. Chirn. Phys. 1892 (vi] 25 420) observed that both ulmins formed artificially from sugar and those occurring naturally in soil began to decompose on exposure to air when moist carbon dioxide being evolved; Roger and Vulquin (Compt. rend. 1908 147 1404) made a similar observation on peat ulmins and noted also that they became insoluble in alkalis ; whilst Tidesm-ell and Wheeler have recorded in their study of dopplerite (J.1922 121 2345) that this undoubted ulmin is less readily soluble in alkalis than that in more recent peats. Solubility in alkalis to form brown solutions should not therefore be regarded as a necessary property of th 124 THE OXIDATION OF BANDED BITUMINOUS COAL ETCL ulmins but it must be recognised that they can (through polymeris-ation and dehydration) change in character so as to lose their “ characteristic ” solubility. Neither the insoluble ulmins which we consider form the major part of newly-won bituminous coals nor the soluble ulmins regener-ated from them by oxidation are identical in character with those that form the major part of peat but they possess the same nuclear structure. The external groupings of the ulmin molecules in unoxidised coal are modified during oxidation the more easily detached groupings being eliminated to form simple oxygenated compounds whilst the residual ulmin becomes more definitely of acidic character (more so than the peat ulmins) owing to the form-ation of carboxylic groupings in place of those detached.Pearson (PueZ 1924 3 297) has reached similar conclusions to these as the result of his work on the oxidation of coals by sulphuric acid and bromine although the analyses of the ulmins so formed differ from ours (obtained either by atmospheric Oxidation or by the use of hydrogen peroxide) no doubt because chemical changes other than oxidation occurred during his experiments. Bituminous coal then consists essentially of insoluble ulmins in which morphologically organised plant tissues (that have escaped ulmification) are dispersed.” The present work shows that alkali-soluble ulmins can be regenerated from a bituminous coal by mild oxidation. If the coal is a vitrain i t consists almost solely of ulmins ; whilst the regeneration of alkali-soluble ulmins in a clarain, a durain or a fusain leaves residues characteristic of each of those ingredients which are essentially the original plant entities that were particularly resistant to or were preserved from the processes of “ decay ” that constituted the early stages of coal-formation and are apparently but little altered. The character of these residues is illustrated in Fig. 2 in which (a) and ( b ) are photomicrographs of cuticular tissues such as can be separated from a clarain by treatment with hydrogen peroxide and alcoholic potassium hydroxide ( c ) shows the exine of a megaspore, characteristic of the ddbbris from a durain obtained in the same manner and (d) a wood tracheid such as occurs in the ddbris of a fusain. Future communications will deal with the chemical * For the sake of simplicity we have here omitted discussion of the presence of natural plant-substances devoid of morphological organisation, such as “resin,” in the coal conglomerate. We have evidence that these resist oxidation at low temperatures and appear amongst the plant d&& after the process of regeneration of alkali-soluble ulmins by mean8 of hydrogen peroxide (or atmospheric oxidation). This work will be described in a forth-coming publication which deals also with the chemical properties of the regenerated Wins FIG. 2. ( a ) ancl ( b ) . Fragments of cuticular tissue characteristic plant entities in a clarain. ( a x 30 and b x 180. ) ( c ) and ( d ) . The exine of a megaspore ( x 30) and a wood tracheid ( x 250), characteristic plant entities in 5t durain and a fusain respectively. [To face Trans. p . 124 TIDESWELL AND WHEELER ON FUSAIN AND ITS OXIDATION. 125 examination of similar plant-tissues separated in bulk from coal ; in particular with cuticles and spore-exines since these are so often the predominant structures to be found in clarains and durains, respectively and most bituminous coals consist mainly of clarain and durain. Work on modern cuticles and spore-exines for com-parison has already been in progress some years in the belief now justified that i t should be possible to isolate individual tissues or identifiable parts from coal so that for example any character-istic distillation products of which they are the source can be determined. DEPARTMENT OF FUEL TECHNOLOGY, SHEFFIJSD UNIVERSITY. [Received September 20th 1924.