A STUDY OF THE PRODUCTS OBTAIN- ED FROM AN EASTERN BITUMINOUS COAL USING DIPHENYL ETHER AND PHENOL AS SOLVENTS GEORGE ERNEST KELLER THESIS FOR THE DEGRE E OF BACHELOR OF SCIENCE IN CHEMICAL ENGINEERING COLLEGE OF LIBERAL ARTS AND SCIENCES UNIVERSITY OF ILLINOIS / 922 K 26 UNIVERSITY OF ILLINOIS _iiaLy,_J 29 _ r 192S THIS IS TO CERTIFY THAT THE THESIS PREPARED UNDER MY SUPERVISION BY Geo r-ge- -f-r-ne^ t- -Keil e-r EN T itled _ _ _4_ ^ Jt he _ ts _ .Obtain e d _ _Fr pm _ an_ _E a 3 t er ri Bituminous Coal Using Diphenyl Ether and Phebol As Solvents IS APPROVED BY ME AS FULFILLING THIS PART OF THE REQUIREMENTS FOR THE degree OF? a _phe_lor _ _c f _ _S_c_i e_n ce_ _ _i_ n_ _C he rci cal _ _E ng i rie er in g_. Instructor in Charge Approved ACTING HEAD OF DEPARTMENT OF -CHEMISTRY '^00 Digitized by the Internet Archive in 2015 https://archive.org/details/studyofproductsoOOkell Table of Contents I. INTRODUCTION PAGE 1 . General ..1 2. Purpose of Investigation 1 3. Historical 1 4. Theoretical .6 5. Outline of Present Investigation 8 II. EXPERIMENTAL 1. Apparatus and Operation . 10 2. Purification of Extracts and Residues 10 3. Large extraction apparatus .11 4. Types of coals used . 12 III. DATA 1. Comparison of the Solvents 13 2. Ultimate Analyses of Coal, Residues, and Extracts . 15 3. General Properties of Extracts and Residues (a.) Relation of Extract and Residue to Coking- . . 16 Properties (b.) Determination of Melting Points of Extracts . 17 (c.) Fractional Carbonization of Residues. ... 18 4. Alcoholic Potash Extraction 20 IV. SUMMARY AND CONCLUSIONS 21 ■ * • A STUDY OP THE PRODUCTS OBTAINED FROM AN EASTERN HIGH VOLATILE COAL USING DIPHENYL ETHER AND PHENOL AS SOLVENTS I. Introduction. 1. General. It is beginning to be recognized that our supply of coal is not in- exhaustible, and that not only should methods be adopted for the conservation of the better grades of coals, but also means should be found for more effi- cient utilization of poorer grades. Whether our future fuel will be mainly gas, produced from the complete gasification of coal, or gas and high volatile coke from the low temperature carbonization of coal, or some other fuel, re- mains to be seen. At any rate poorer grades of coal will eventually have to be used, and it seems probable tnat all coals will be subjected to some form of carbonization, not only to save the valuable by-products, but also to improve the quality of the product to be burned. The number of coals which are carbon- izable, or which ai** used in any of the modem coking processes is small. It is not definitely known why all coals do not coke. Evidently the constituents of different coals are different, and require different treatment in the coking process. Therefore, it is necessary to make complete scientific study of the constituents together with their behavior during the carbonization process. 2. Purpose of Investigation. The purpose of this investigation has been to study the products ob- tained from an Eastern High Volatile coal using phenol and diphenyl ether as solvents. 3. History. The methods of attack in the study of coal are (1) the use of solvents 2 and reagents, (2) fractional carbonization, and (3) microscopic examination. There is no sharp distinction between the use of solvents and reagents; (many substances which were formerly considered to be solvents hav6 been shown to be reagents). The true reagents which have been used are sulfuric acid, nitric acid, caustic alkalies, bromine, ozone, and oxygen. Very little success has attended the use of these reagents either because the yields of products are so small, or the products themselves are in such an advanced stage of decomposition that little can be learned from them as to the constitution of coal, or because when the yields are high it is impossible to separate the products. Fischer and Gluud^ have recently been able to dissolve practically all of coal by sus- pending it in water and passing a stream of ozone through it. The amount that can be extracted by alkalies is increased by weathering the coal. This seems to be due to the fact that ulmic acids, which are produced by oxidation, are soluble in the alkalies. The use of solvents in the study of coal has met with more success than the use of reagents because the products are obtained with the minimum alter- ation. The problem of solvents is, however, not as easy as the name would suggest, because it is very hard to distinguish between solvents and reagents in their action on coal. The greatest difficulty lies in finding a solvent which will extract large amounts without alteration so that the main constituents can easily be studied. 2 Bone states that J. A. Smythe of Gtftt ingen was one of the first workers who systematically studied tne effects of such solvents as benzene, chloroform, ethyl alcohol, light petroleum, and acetone on a brown coal. With none of these solvents, however, could he extract more than 3%, He divided his extract into portions soluble and insoluble in ether, obtained the molting points, and on still further purification by two different methods, obtained "res ins" which 1 & 2. Bone. "Coal and Its Scientific Use." London, 1918 • ( , 3 melted at 80° C., to which he assigned the formula C13H26O. Bedson 1 followed by Anderson 2 3 4 tried pyridine on a number of coals, exr- tracting as high as 20.4$ from a Durham gas coal. The method was to treat fine- ly divided coal in a Soxhlet extraction apparatus with pyridine. The excess pyridine was distilled off from the extract under reduced pressure, and the final purification effected by passing a stream of warm air over the extract. The residue was also purified and weighed. From the results of experiments on a 3 number of coals Bedson concluded that "the extracts in each case were found to be rich in volatile constituents, and in the coking assay gave highly intumes- cent cokes, whereas the residues left after extraction were almost devoid of coking properties". In general he found that there was no connection Detween the amount extracted and tne volatile content of tne coal. Air was not ex- cluded from tne extraction apparatus in these experiments, and the final puri- fications were made in a stream of air. Wheeler^and others also used pyridine as a solvent followed by an ex- traction of tne original pyridine extract with chloroform. They claim to have made by this treatment "a complete, or nearly complete separation, between the resinous constituents, and the degradation products of celluloses of which coal is conglomerated". 85.33 1.41 10.41 16/4 C H N S 0 77.32 5.14 2.07 1.21 14.26 Vol. 31.88$ 1 Jour. Soc. Chem. Ind. 1908 Page 147 2 Jour. Soc. Chem. Ind. 1902 Page 242 3 Jour. Soc. Chem. Ind. 1908 Page 147 4 Jour. Chem. Soc. 1911 Vol. 99, page 649, ibid. 1913 Vol. 103, P. 1704 . 4 The above diagram gives a summary of their results. Bone states that these results are open to three criticisms, namely: (1) "that the sum total of nitrogen in the products is nearly double that in the original coal treated, showing that some of the solvent. nad been retained by the various fractions, despite the pains to remove it; (2) that a product containing as much as 1.71 % of nitrogen and 1.3 2 %> of sulfur, as did the portion soluble in both pyridine and chloroform, cannot be regarded as a reasonably pure "resinous body"; and (3) that inasmuch as the total oxygen in the various fractions is about 1.27 times that in the original coal, there had apparently been some absorption of oxygen during the extraction process, no precaution having been made to exclude air". Hot only is their work open to these criticisms, but also their results show a decrease of 2 , 50 % in the amount of carbon in their products over that in the coal. Something besides a solvent action has taken place in the action of the pyridine on the coal. Wahl^ working with pyridine also found that there was no relation bet- ween the amount extracted, and the volatile matter. Harger 2 found that if the extraction was carried out in a sealed glass tube under pressure he could in- crease the amqnt extracted from 22. 5 % to 40.5^. He came to the conclusion, however, that the action of pyridine was not that of an ordinary solvent, but that it was a depolymerising action. Aniline and quinoline have been used by other investigators with little success. Frazer and Hoffman 3 used phenol in studying the soluble constituents of a non-coking coal from Zeigler, Ills. Their method was to place the finely ground coal in a double walled copper vessel upon a filtering medium, and to treat with phenol for ten hours at 140° C. After filtering off the solution, the process was repeated for three or four times. They treated the extract with various reagents and solvents, separating it into a large number of sub- 1. Compt. rendus. 1912 Vol. 154, page 1094 2. Jour. Soc. Chem. Ind. 1914 Page 389 3. U. S. Bureau of Mines, Tech. Papfcr Ho. 5, 1912 * . . . . . 5 stances. From the result of their experiments they came to the following con- clusions: "In the lack of evidence to the contrary it is assumed that the coal substance soluble in phenol is present in the coal as such. "The authors believe, as a result of the investigation, that some of the substances isolated very closely approach compounds." Parr and Hadley‘S using phenol as a solvent found that the amount which could be extracted was definite, and susceptible of quantitative determination. The extract had a rather definite melting point, and a decomposition point above the melting point. This extract was found to be the vital constituent concerned in the coking of the coal. They were able to modify tne coking qual- ities of the coal by oxidizing either the residue, or extract, or both. Cherry 2 working with the same solvent also found that oxidation of the residue destroyed that coking property of the coal altnough tne coking principle was in the extract. He explained this by considering that A the fusion temperature a reaction took place between the oxidized cellulosic constituents and the res- inic bodies, altering the latter in such a manner that they were not able to bind the coal particles together. Fischer and Gluud 3 of the Kaiser Wilhelm Institute for Coal Investi- gation have recently worked with benzene at temperatures ana pressures approx- imating the critical constants of the solvent. They carried out tne extraction in a steel bomb with the coal suspended in a wire gauge basket. The temperature employed was 275° C. A coal which by extraction in a Soxhlet extraction app- aratus gave only 1$ extract, yielded 6.7$ in the bomb. On pouring the extract dissolved in some benzene into petroleum ether, a brown powder separated out which soicened at 1400 and melted at 150® c. They observed that no gas was liberated when the bomb was opened and took this as an indication that no decom- 1. Bulletin 76. Univ. of 111. Exp. Station 2. Thesis B. S. Univ. of 111. 5. 3 one. "Coal and Its Scientific Use" London 191ft . ■ 6 position had. taken place. The question as to whether a temperature this nigh does not in some way alter the coal substaiice itself, has not been settled. It would seem from these investigations that one of the best fields for solvent work lies in extraction under pressure, and at comparatively high temperatures. 4. Theory. The terms 'cellulosic constituent''' and "resinic bodies" have already been mentioned in connection with Cherry’s investigation. These terms have cone into the literature, and have been used more or less indefinitely to denote the residues, and extracts from the treatment of coals with various solvents. Lewes^ was one of the first to use these terms. He states that coal is made up of the carbon residuum from the decomposition of the humus bodies (cellulosie derivatives) together with humus bodies in varying states of decomposition, resin bodies, and hydrocarbons. His resin bodies are formed by the oxidation of such extractive matters as gums, mucilage, vegetable jelly, plant resins, etc. Since the resin bodies are more resistant to decay they accumulate in the masses of decaying vegetable matter. The hydrocarbons are formed from the resin bodies by isomeric or other changes brought about by heat and pressure. He states that the "factor which has made the oldest coal a coking ooal is the more complete degredation of the humus with the elimination of oxygen bringing the resin bodies and hydrocarbons up to a percentage which will supply enough luting material on decomposition to give a satisfactory coke." Although some of his assumptions as to tne formation and composition of coal may be correct, they need to be revised as the result of later investigations, especially those of Thiessen.~ It has been shown rather conclusively that free carbon does not exist in coal. The newer and more correct view is that the extractive matters are the decomposition products of cellulose, and are in no way to be considered as resins. 1. Lewes. "The Carbonization of Coal." 1914 2. Bulletin 117, U. S. Geological Survey. * . . ■ «■ . ' • * f 7 Burgess and Vftieelerl fractionally distilled several English coals in a small platinum retort. They summarised their results as follows: (1) "there is a well defined decomposition point between 700°C. and 800° C., which corres- ponds to a marked increase in the quantity of hydrogen evolved; (2) with bit- uminous coals the quantity of hydrogen evolved falls off at temperatures above y00° C., but with anthracitic coals it is maintained up to 1100° C; (3) the evolution of hydrocarbons of the paraffin series ceases almost entirely at temp- eratures above 700° C.; (4) ethane, propane, butane, and probably higher members of the paraffin series foiro a large percentage of the gases evolved below 450° C.; (5) the rate at which the carbon monoxide is evolved is uniform throughout a dis- tillation at any one temperature, ana is maintained rignt up to the end, while the rates in the case of other gases fall off." From these statements they con- cluded that coal consists of two parts of different degrees of stability, the less stable or resinic portion decomposing first to give chiefly paraffin hydro- carbons, while the more stable cellulosic portion decomposes yielding hydrogen. Very probably the difference between one coal and another is determined mainly by the proportion in which these two types of compounds exist, anthracite, for example containing but little of the more unstable constituents. To further verify their conclusions they separated coal into the three portions mentioned on page 3, and fractionally distilled these products. Porter ana Taylor^ carried on a series of similar experiments with Am- erican coals. In general they do not agree with Burgess and Wheeler. They concluded from their experiments that the first decomposition occurring in any type of coal is the breaking down of certain oxygen bearing substances related to cellulose, the products being chiefly water, carbon dioxide, and carbon monoxide. The "resinic 1 ' portion decomposes v/ith the formation of paraffin hydrocarbons with probably hydrogen as a direct decomposition product. Their conclusion as to the early decomposition of the cellulosic portion is more logical than that of Bur- 1. Jour. Chem. Soc. 97, 1910, pp. 1917 - 1935; 99, 1911 pp. 650 -657 — , .,,f oxter - Bartm-., MO. IT. S nyeua of Ttliar.a* V ' . . - ■ 8 gess ana Wheeler. In Doth Burgess and Y/heeler's, and Porter and Taylor's work no reference has been made to the presence of nitrogenous constituents. Bone believes that the presence of these constituents has some bearing on the coking properties of the coals. Recently there has appeared an article- 1 - in which a third class of bitumen present in coal is mentioned. It is supposed to be the degraded product of the nitrogen proteins, and is intermediary in character. There is at present, however, no experimental evidence for the existence of this third group, and its influence on the coking properties, if present, is negligible. No selective solvent has been found for this proteid constituent. The term "resinic" is more or less unfortunate in that many have taken it to mean that this material is derived from the resins of the plant substances from which coal is formed. The "resin bodies" should be distinguished from the original gum or fossil resins. Although the extractive material may contain some gum and fossil resins, by far the larger part of this material is the dec- omposition product of the cellulosic material of the plant substances. In this investigation the soluble and insoluble portions of the coal will be designated by extract and residue respectively. The view has been developed at the University of Illinois that for a coal to coke it must contain sufficient of the "resinic constituent" which does not distill without decomposition, to furnish a binder for the"cellulosic portion". There must be no reaction with oxygen at the crit- ical temperature > or large deliverance of gases to prevent the luting material from cement ing the particles together. 5. Outline of Investigation. It has been the object of this investigation to make a comparison of the two solvents, diphenyl ether and phenol. Phenol has been used frequently as a solvent for coal. Up to the present time, however, no mention has been made of the use of diphenyl ether in coal investigation. Preliminary investigation of 1. &as Journal (London) 157 No. 5060, p. 28 ' ■ ■ 9 this compound, seemed to show that it would extract a comparatively large amount from coals, and it was thought that perhaps it might he used in place of phenol, pyridine, and other solvents used heretofore. The study of the action of these solvents has involved the following points: (1) A comparison of the amounts extracted from an Eastern Bituminous coal, and a weathered lignite. (2) Development of an apparatus and method for extracting comparatively large amounts of coal. (3) Determination of the amount of decomposition during the extraction. (4) Determination and comparison of the ultimate analyses of the diff- erent residues and extracts with the original coal. (5) Determination of the influence of the various products in coking tests. (6) Determination of the melting points of the two extracts. (7) Fractional carbonization of residues with a comparison of the res- ults. (8) Determination of the amount extracted by a 10^ alcoholic potash solution. II. Experimental 10 1. Apparatus and procedure. A number of different metiiods were tried in extracting the coal. The method in which the Soxhlet Extraction apparatus was used was discarded when it was found that this apparatus would not stand the difference in temperature and consequent expansion, without breaking. The method finally employed ror small quantities was essentially that used by Cherry-*-. Five grams of coal were placed in an Erlenmeyer flask, dried at 105° C. for one hour with, a stream of dry nitro- gen passing over, 50 - 60 c. c. of the solvent added, and the mixture refluxed gently for 24 hours at the boiling points of the solvents. The flasks were heat- ed in a nichrome wound electric resistance furnace, insulated from radiation by asbestos packing. After the mixture had refluxed for 24 hours, the hot solution was allow- ed to cool slightly, and was then filtered through a Buchner funnel. The phenol residue was washed twice with alcohol, and then with etner. After washing it was quickly replaced in the flask and refluxed again for 24 hours with fresh phenol. The diphenyl ether residue v/as washed with ether alone. Two extractions were made after it was found that only a very small quantity was extracted on the third and fourth treatments. 2. Purification of extracts and residues. The solutions from the two extractions were united, and most of the ex- cess phenol distilled off. Both distillation under reduced pressure and under atmospheric pressure were used. When reduced pressure v/as used, nitrogen v/as allowed to bubble through the solution instead of air in order to prevent oxida- tion. Reduced pressure distillation was finally abandoned, because the final temperature employed in the purification was slightly above the boiling point of the phenol. Before all of the phenol v/as distilled off, the contents of the dis- tilling flask were allowed to cool slightly and were transferred to a tared Er- 1. Thesis, B. S. Univ. of 111. ■ ■ ' ‘ I 11 lenmeyer flask, and the final purification made by passing a stream of dry nitro- gen slowly over the extract at 190° - 200° C. until no odor of phenol could be detected in the nitrogen passing off* The nitrogen used was obtained in cylin- ders from the Linde Air Products Co. This was further purified by passing it through solutions of alkaline pyragallol, and concentrated sulfuric acid in order to remove any oxygen and moisture that might be present. The diphenyl ether extract was purified in the same manner except that the final temperature employed was 260° - 270° C. The residues were purified by washing first with alcohol, and then with ether in a Soxhlet extraction apparatus, or on a Buchner funnel. The last traces of the solvents were removed oy passing dry nitrogen over the residues at temp- eratures slightly above the uoiling points of the solvents. 3. Large Extraction Apparatus. It was found that in both the diphenyl ether and phenol extractions, the coal tended to cake on the bottom, causing bumping, and perhaps overheating of the coal. In order to prevent this, and to facilitate the handling of larger quanti- ties or the coal, a larger extraction apparatus was devised. It consisted of a cylindrical iron vessel 6 inches in diameter, and 7 inches deep with an iron cover ground to fit, and fastened with six 3/8 inch machine uolts. A lead gasket was used to naxe a tight seal. A one inch pipe coupling was uriven into the hole already in the center of the cover, and a pipe two feet long screwed into tne coupling. A 3/8 incn brass rod carrying two stirring padulos passed down through tnis pipe. Loose fitting brass bushings were driven into each end of the pipe. It acted, not only as a means of supporting the stirring mechanism, but also as an air condensor for the condensation of the solvents. The stirring mechanism was driven by a 1/16 H. P. motor, the speed of which was reduced by a system of pulleys to 50 - oO R. P. M. - « r 18 Table IX. Substance Melting Point Degrees C. Diphenyl Extract 210 - 245 Not sharp Phenol Extract 185 - 190 Rather sharp The melting points were taken by gradually raising the temperature of an iir bath, in which a cube of the extract was suspended from a wire, until the ex- tract melted and dropped off. Some softening took place before the temperatures listed above. The diphenyl ether residue did not have a distinct melting point, and did not become very fluid at any tine. It contains constituents which have a higher and wider range of melting points than the phenol extract. W. S. Haw- thorne working at the University of Illinois, sound the softening of this coal took place between 3 too c. and 3 70° C. (c) Fractional Carbonization of Residues. In order to determine the effects of removing different amounts of the extractive matters, fractional carbonization tests were made on the diphenyl ether ana pnenol residues. The carbonizations were carried out in a glass retort heat- ed by an electric resistance furnace. The tar and water were condensed out in a small distilling flask the Mg removed by 10$> sulfuric acid, and the gases dried by passing through a CaClg tube. The temperature was raised gradually, the total carbonization time varying from 4-5 hours. .Amounts of tar and water varied from 2-5 grams. The retort was swept out with nitrogen before and after each run. The amount of nitrogen in the retort at the beginning of each run was not measured so that the amount of nitrogen delivered by the coal could not be deter- mined. ■ Fractional Carbonization or Jellico Coax Table X. Residues 19 Gas Analysis (112 free basis) Percentage Composition Diphenyl Residues Phenol Residues Temp, inside retort 360° 410° 460° 360° 410° 4o0° Carbon dioxide 69.6 44.6 40.1 74.6 47.4 16.3 Oxygen 7.o £.2 2.9 0.7 3.2 1.2 Ethylene Series 6.3 o«2 6.7 1.7 0.1 7.3 Benzene Series 0*0 O.o 0.4 0.0 0.5 0.9 Hydrogen D# o 6.5 10.3 10 .a 10.0 22.6 Carbon Monoxide b.o 6.5 3.5 1.7 7.4 6.7 Metnane 5.1 13.2 24.7 2.o 12.7 26.1 Ethane 19.0 10.4 12.7 18.9 TOTALS 100.0 100.0 100.0 100.0 100.0 100.0 Table XI. C. C. of Gaseous Products per 50 grams. Diphenyl Residues Phenol Residues Temp, inside retort 360° 410° 4600 360° 410° * 4600 Carbon dioxide 106.0 167.0 418.0 76.5 216.5 188.0 Oxygen 11.5 12.1 30.0 8.9 14.5 13.4 Ethylene Series 9.6 23.1 70.0 1.8 27.8 84.2 Benzene Series 0.0 2.2 4.0 0.0 2.4 9.8 Hydrogen 8.6 24.2 108.0 10.7 45.9 260.5 Carbon Monoxide 8.6 24.2 38.0 1.8 33.9 76.8 Methane TV Ethane 7.7 49.6 1.59 on. 4 256.2 1.3 109.8 2.7 57.5 1.5 57.5 501.8 1.42 218.5 TOTALS 152.0 373.8 1034.0 102.4 456.0 1153.0 * Residue not from same extraction as 4 and 5. • . . . • ' 1 20 Carton dioxide forms a large percentage of the gases delivered np to 460 o C. The phenol residue, which contains more of the "cellulosic constituent", shows a marked increase in the amount of hydrogen delivered over that delivered by the diphenyl ether residues. Almost no "benzene series, and only a small per- centage of the ethylene series and carton monoxide are given off up to 460° C, Xethane and ethane begin to be delivered in comparatively large amounts at about 410° C. 4. Alcoholic potash extraction. In order to determine what effect alcoholic potash would have on the coal, 20 grams were refluxed with a 10 % solution of XOH in 95 % ethyl alcohol. The solution was filtered, the filtrate acidified, and the brown precipitate fil- tered off. The amount extracted by one treatment was Z.oZ% on the ash, moisture free basis. As much as 6.49 % could be extracted by three successive treatments, although the last two extractions took out chiefly ash. The amount of ash free substance extracted was practically 5% of the coal. The residues alone from these treatments weighed more than the original coal, indicating that it was im- possible to remove all of the XOH which had reacted in some manner with the coal. The residue had no caking properties, while the extract swelled in much the same manner as the phenol or diphenyl extracts. Ho coke was formed on heating the re- combined constituents. Saponification by means of alcoholic potash should re- move all the true resins present in the coal along with the ulmic acids. The excess of the amount extracted by phenol over this amount might be taken as an indication of the amount of cellulosic degradation products present in the coal. This would not be reliable, however, since it has not been shown that the phenol removes ulmic acids. ■ ■ 21 IV. Summary and Conclusions. 1. Diphenyl ether is not a satisfactory solvent for the types of coal used in this investigation, the residues having practically as good coking qual- ities as the original coal. It is much easier to handle than phenol. 2. Che diphenyl ether extracts contain constituents which have a higher and wider range of melting points, a higher carbon and hydrogen content, and a lower oxygen content than the corresponding phenol extracts. This seems to in- dicate that diphenyl ether extracts more hydrocarbons than phenol. 3. The temperature used in the treatment with diphenyl ether undoubt- edly causes some decomposition of the coal. 4. The coking principle is in the extract. The coking property can be altered by oxidation of either the extract or residue, the oxidation of the residue taking place morequickly and completely. 5. Part of the phenol, and perhaps some of the diphenyl ether, is re- tained by tne extract or residue or by both even after the most careful treatment for its removal, indicating that it has decomposed or has reacted in some manner with the coal substance. 6. It is possible to form a semblance of a coke structure provided enough extract is added to a lignite coal or residue. The amount needed is great- er than that contained in a good coking- coal. 7. Carbon dioxide forms a large proportion of the gases given off by the residues up to 460° C. The paraffin series begins to be liberated in quan- tity at about 400° C. A larger amount of hydrogen is given off by the phenol residues than by the dipnenyl ether residues up to 460° C , ; - a ‘