THE USE OF SELENIUM OXYCHLORIDE AS A SOLVENT ON COAL BY WILLIAM ROBERT KING, JR. B. S., Monmouth College, 1920 THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE IN CHEMISTRY IN THE GRADUATE SCHOOL OF THE UNIVERSITY OF ILLINOIS 1922 URBANA, ILLINOIS /3J-a.23 /.4P I 92S K58 UNIVERSITY OF ILLINOIS THE GRADUATE SCHOOL ^^u_s± 1922 1 HEREBY RECOMMEND THAT THE THESIS PREPARED UNDER MY SUPERVISION BY„ . Willi am Rob ert K ingj Jr. ENTITLED ^ e _U S6- af Sel enium asy^c iilo F ide As a So lvent on CoaLt BE ACCEPTED AS FULFILLING THIS PART OF THE REQUIREMENTS FOR Recommendation concurred in* Committee on Final Examination* •Required for doctor’s degree but not for master’s 50940.1 Acknowl edgeraen t llils Investigation was undertaken at tiie suggestion of Dr. Thomas E. Layng, and was carried out under his direction. I wish to thank him for the interest he has shown in the investigation, and to express my appreciation of the very helpful advice and criticism he has given at all times. Digitized by the Internet Archive in 2015 https://archive.org/detaiis/useofseleniumoxyOOking Table of contents Page I* Introduction. 1. General Nature of U)al Solvait Probl^ 1 2. Historical and. Theoretical 3 3. The Purpose of tlie Present Investigation It II. Es})erimen tal. Part I. 1. The Nature of Selenium oxychloride 13 2. The Type of Coal Used 16 3. The Use of a Neutral Solvent mth Selenium oxy dilorlde. 1*. Selection of the Neutral Solvoit 17 2 \ iipparatus and. Proceedure 19 3’, Examination of Products and -Analysis 23 4’. Oaihoni zation of Residues 27 III. Results, Part I. 1. The Nature of Selenium oxychloride 2. The Action of Xylene-Selenium oxydilorlde mixture on Co al . . 29 3. The Carbonization of Residues 32 IV. Conclusions, Part I 33 V. Experimental, Part II. I. The Effect of the Tar and Volatile Content of Coal upon the Resulting Action of Selenium o:^ chloride 34 2. Effi ci Qit Methods of Handling Selenium oxychloride ••• 36 3. Extraction of Coal with Selenium oxy chloriLde. 1*. Apparatus and Pro ceedure 3? 2*. Examination of Products and Analysis 39 4. The Effect of Selenium oxydrloride Upon the Primary Vol atil e Pr’odu cts of the Carbonization of Coal .... 1*. Outline of Investigation 43 2 * • App aratu s 45 3\ Temperature Control and Measurement 46 Proceedure 46 5*. 0e termination of Products and Analysis 48 VI. Results, Part II. 1. The Effect of the Tar and Volatile Contoit of Coal Upon the Resulting Action of Selenium oxychloride • 49 2. The Effect of Selenium o:^ chloride Upon Coal 50 3. The Pri.mary Volatile Pm ducts of Coal 53 4. The Effect of Selenium- oxychloride Upon the Primary Vol atil ePmductsofCoal 78 VEI • Con clu Sion s. Part II . 81 VEII. Summary 83 IX. Bibliograiihy 85 1 THE USE OF SELENIUM OXyaiLORIDE AS A SOLVENT ON COAL. I* Introcti ction* 1. Gaieral Nature of Coal Solvent Problem: Uils investigation was tak^ up as part of a group of prob- lans being studied in tliis laboratory involving numerous so-called solvents and reagents such as Benzene and Xylene under pressure, mphenyl. Biphenyl ether, and Selenium oxychloride* Ihe powerful corrosive and solvent action of selenium oxychloride has been known for several years, in fact it has beoi called by some ” the uni versal sol ven t” • Ihe ^estion of coal solvents aix)se during the middle of the last caitury and has attained much prominoice in the past decade until tod^ the use of solvents is recognized as one of the four chief methods of attacking coal* It has long been realized that tlie results obtained from ordinary metJiods of analysis, such as proximate and ultimate, are of rather narrow application especially from a chemists point of view* The information derived from such analysis is rather for the engineer* It is such information vhich can be applied to engineering problems connected with the handling and use of coal such as purcJiase, storage, weath ering, coking and ordinary combustion* These analyses give percentage results of the individual elements obtained by the destructive analysis of the coal, but they do not convey the slightest information relative to the actual organic constituents of coal as th^ occur in nature. It is not witfi percentages of hydrogai, carbon and oxygen or their compounds as produced by coking conditions and destructive analy- ,t .IT*# \ - My ‘I (Jr jL>*i ri^n i ■ “(T -r> :) ti ‘.i,vr; • M '^n *..• ■/■' .uV, -V‘»' ’ ' ■■'’‘•■' ’• n i$r.*f ;»/>.} :-Y fffTjpti v;< o i' n aJ: '** Bii'- ■ ; /■ Of-'".; '‘^>-'i !*.^W .ff* ' V ■'■' i ' i ^ f f)V Ik*. UT’ 5* -'*■'. rr *'I*‘r''»> r" ;■ t ‘ -f S' *■ t < #> ,r> - ■, >■ * Pr «k ' : ■ ■ 1 : .^OTYfur.* ( ■■' I ■' r-' ^ V.: .( y r '; '« •' V I ^ -irv — ./o*'a ;■'’,•■! •-■.• « '■•-j t . at ' 1 , . ( I ;.■.;< f riH ■'« -.jO ‘1- 1* ■ ..hJ’C^jR m ilf” . '• I .<> « , M •: ;■'*(¥< r N“*. U(:r:i .«!.•;• *••■■ -. •‘>••7 * !; "■ i u'"'’ !: < ^ ’ .'• I r ■ ' ' ’ ■■•r ,{ t'„ " •;.. ' '.7 ■Vt'.: ■■/!>(•' 'f*"' .I".! . .' 'V ,'j /; ■'', ,ryi -i.' ,'■ *’ y .’•r'- I ’ ■ , . ’irc.Mf ^m'lt 6 ;.:i ' ■. ^t' , f -y ■ ■• > O/yr 'i ■) ’ ■ r , . " ii V ;;j •' 'tl. ,♦ ■ ' ^ > , • ' , '-.I ^.;: .>V , . ' ' , ■. -' ; • I, y'l . ', yyM.s 4id!S , ^ . »T tf.*-'* i» V-’ i'l’ , . » 2 , sis that the chesnist is fundamentally concerned, but with the unal- tered compounds in coal as th^ occur in nature. It is with these constituents that the ultimate solution of such problems as di still* ti on, combustion, carbonization, o 2d. da tion, weatliering and storage lies. Here lies the field for microscopical and solvoit woiic, particular- ly for solvents. If the coal conglomerate or any considerable part of it were to show preferential solubility in any given liquid sol- vent t?ie chemist’s probl^ would be greatly simplified. ihe question of vhat kind of a solvent is most desired by chem- ists is hard to answer. A true solven t, as distinct from a reagoit, sjiould be chemically inert toward both the substance extracted and the residue. A solvent is wanted that will remove some part of coal, it matters not at, Mi ether it be cellule sic, resent c, protein or waxes. Selective solvents are desired. lUrthemore the coal consti- tuents must be removed unchanged. Many so-called solvents largely used in recent years, such as phenol and pyridine, are open to the suspicion that results obtained by their use have been rather more than that due to a solvent action alone. Ihe admitted difficulty then with most solvents is that th^ are reagoits also. There is yet a broader and more practical field for solvents. Thus far woik: with solvents on coal has not been of very great val- ue relating to carbonization processes audit is to this end, that of practical application, tliat all coal problems Miould point. Solvents have been used more to leam of the composition of coal tfian to aid in the coking of it. It is very hard to link solvent wo lie with carbonization. From the standpoint of carbonization it is the coking and non -coking const! tuQits of coal Miich should be sep- arated. Phenol and pyridine have been largely used in this attempt. I •5 I 3 , but it is doubtful if th^ have proved the right solvents for this purpose* To some extoit tJiey have proved successful as pointed out by Jones and \iheeler (1) (1914). Generally speaking however there has not yet been found a solvent Tihicii will seperate out the coking from the non -coking constituent. 2. Historical and theoretical! A resume of the leading woric ^idtli coal solvoits and reagents follows. ue Marsilly (18G2) (3) made vhat is probably the first compre- hensive stutfy of solvent action on coal. He treated various types of coal witii al cohol, ether, caibon disulphide, benzene and chlorofom at their boiling points. He noted that all except alcohol had a dis- tinct solvent action on ” f at” bituminous coals, dhloroform having the most. Anthracite coals were not affected by these solvents. The treatment of good coking coals took away some of tlie coking property Guignet (1879) (4) used dry phenol at 110 ’ and found that it was abetter solvent for finely ground bituminous coal tlian was dilorofonn. The phoiolic extract was a deep brown color and deposit- ed brown flakes on cooling. Four percent of the coal was extracted. Don do rff, according to Mu - 1.29. iteinscii (1885) (6 ) took the view that coal was composed of two substances iVhich could be distinguished by action toward certain solutions. With alkaline solutions he isolated several SLib stances vhich were not attacked by mineral adds. » i; .» o ■ M>' fXim tlTJOU '7 'I'i r • i fl' ' , f r B'l . .^1^9 n .•!. r jrajjaf 'ijip , v'**')'.’ i 'T ! rt -' t'iJ J J iA ' L -. .' i-iKj . r. W f (k > li.' . liT ^ VUt 'ihrr ’ >' ;t/) i - li? " ' ^ ■■. I ^ ^ ^ ‘ijr J n ■ , ■ ■ -1 °I . •-.’) (■■■ ■ xt ' :’. - ; i | 0 > ■ t: ::M^} rS f .'‘- y:ri ' I I . 1 r r '- V I ';■: i\ *'iU ‘I I U '{J t:.’’ *.i f 7 -V Tt* 4 , Siepmann (1S91) (7) made an important study of Westphalian co^ by extraction with (hlorofonn in a Soviet extractor. He obtained a deei3 brov/ti solution with a green fluo resell ce. 1111 e extract equaled 1.25^ of the coal. It was a heavy, darls: brown solid with an odor like petroleum. He dissolved parts of the extract in ether, al cohol and chloroform. He also tried ether, al cohol and chloroform on fresh samples of the same coal. Analy si s of th< 3 extracts vfas as folio' C« 1 H. 0. N. S. 0 rigin al coal 80. 31 5. , 5 12 !.94 1. 25 Composite chloroform extract 8 3. 46 7. ,93 4. 27 2.7 1.63 Crude ether extract ....... 84. 82 H). , 51 4. G7 - - Hedissolved ether extract . 78. 74 9. ,64 11. 60 - - Al cohol extract ........... 7 2. 52 10, ,08 17. 40 - - Uiloroform extract 78. 82 8, , 56 0. 97 - 2.6 5 i?inal coal residie 74. 00 4, ,77 20 .09 1. 14 Watson Smith (1891) (S ) tested the solubility of several csBin^ coals and a bituminous coal from Miiki, Japan, in benzene. Not more than one percoit extraction was obtained from any of the cannels, but in the Mikke coal he got a tea percent extraction. He called the soluble extract ” so lubl e bitumen” or petroleum. Ihe extracted sub- staiice contained pyridine bases and phenol as well as hydro carlDons of the benzene series. Mderson (1897 ) (9) treated some Scottish coals with gasoline and with carbon disulphide in the cold. He obtained results similar to Siepmann, but in contradiction to ue Marsilly he noted no differ- ence in the coking properties of coal before and after extraction. i?nderson and Roberts ) 1898 (10 ) treated an oxidized or weatiier- ed Scotch coal with dilute nitric adLd and potassium hydroxide. A '> i r- ) ) i:> ' ■■ ■ u ^ •V' ■ ' X\, U’ : :> I". { i;rf 5. number of substances of acidic nature were extracted. These were tal^en to be derivatives of humic acid. In regard to the coking pro- perly of a coal attribute it to a certain substance in the cosQ. Which will volatilize or decompose easily. Ni trogen, sul fVir, coking constituents of organic complexes, and an all amounts of resoiic materials are pointed out as being the main substances. Baker (1901) (11) extracted finely divided coal ^vith pyridine in a Soviet extractor for fifty hours at 110 '-190 ’O. Two bituminous coals and an antj^racite were so treated. Ihe anthracite was not attacked to any d^ree. One bituminous gave 20.4% extraction, and the other 11.5%. Ihe bituminous coals lost all or most of their coking properties after treatment. Bedson (1908 ) (12) published results of further e3q)erimen ts with pyridine on gas coals. He showed that different constituoits of coal contained various amounts of extractive material. Mderson and K aider son (1902) (13) showed tJiat the amount of volatile matter presait in residues from pyridine extraction con- tained actually higher percents than the original coal. Ihey stated that pyridine was the best solvent they had yet used. The percentage of 0,H, and N in the extract was about the same as that in the ori- ginal coals. The coking properties of a poor coal could be entirely removed, but was only partially ronoved from a strong coking coal. Lewes (1912) (14) infers from this past woik ttiat from some coals pyridine dissolves everything but fixed carbon and the a^i, and that the pyridine must attacii itself to some constituent of the coal to fonn an additive compound insoluble in excess of pyridine. Lewes says ’’The resin constituents condition the coking of coal during destructive distillation, and th^ are of at least two kinds. , »■ I r' ' , •* i 01 ' -f ’) *1 o O' - '* -1 d .> oi «; f t o-> > •■).•*, J.*7 0;fr ♦ * . , '('. •■ .f •’. .♦ !;.*• 4 .*'^ ■ . *. ■ t, «►» •■.•.• < ' . >' '' ••fJ. ■■• If- .. J?e'WO *?■ % I- I, ‘ ^ 1 ^ C'd”'V rrrf I f . •’•<'•.■■ ■•' v:.i ' v ’ • )'■ 4 ’,vr:'i \V ' ■■ .. 1 . V' ■■ »JrW m^Ait 'f'* • : -M' '»;r/ {/,' T ,' : rr-Ao ; ! /iH. i •i.:Vs • L 7 f ' '{ ' . 1 ^* V ‘•J .‘I’ r'w'ilr! ■ ' . ' ■ I *' ■ ■ ■■ r^y f i .4 f ». W, I > :■• : ,7« : ; V I'^'J • • '» r l.!i J ■ ' / rr-. Vl'j ^ 1 ' XJ;-'' ' 4 \ ft • -*►•■■> ■ >. ' • , 1 ' 17 : : • t t ■• fjjf* ' 'Jt l' L ^ t ' ■5 tif » i’r’T’t Top t!.!' I';' ; iJIrfr. i ■ tf cxi “V w i * I. * * * ' ■•' j '••> ■ ri 6 Ihe first easily oxidi zabl e, solubl e in pyridine and. saponifialDl e by alkalies, and "i^hich on weathering is oxidized, into a humus body with evolution of water and. carbon dioxLd.e, and. is responsible for tlie heating of coal in storage. The second, class is non-oxidi zabl e,no t saponified, by alkalies, and forms ^vitii pyridine a compound insoluble in excess of the reagent, and this class may be the hydrocarbons fix)m the decomposed resins, as tlie residue in ishich th^ are presait yield rich liquid hydro carbons, a tar and pitch, but not much gas.” Thus pyridine is not suitable for a solvoit according to Lewes. Stopes and \Vheeler (1918 ) (15) contradict tfiis conclusion of Lewes by saying that the amount of volatile matter yielded depends upon the temperature, duration, manner of heating, state of division and texture of the coal sample, all of vhich extraction of coal witli pyridine greatly affects. Ih^ go on to state tliat retention of pyridine by coal has not yet been proven. Frazer and Hof fin an (1912) (15) worked on the constituents of coal soluble in phenol. Of all solvents investigated th^ reported ph0tiol,pyi:d.dine and aniline as removing the largest quantities of solubl e materi al . PhenoJ was selected for special sturdy. For lacJc of contrary evidence th^ assume that coal substance soluble in phenol is present as such in the coal itself. Th^ extracted in turn this substance with numerous solvaits and reagents and believe that some of the substances isolated very closely approach pure compounds. For these th^ present molecular weights and 0,H, and 0 analysis in several cases. A further study issing pyridine was to be undertakoi. Burgess and \ftieeler (1911) (17 ) made a very thorough investi- gation of the solvent action of pyridine on coal. Their results led to the establishment of the very definite theory that coal is com- , M . Jl ' r,- ’flir.V’; ;; It r . • •■* V'' ■*'■••<» <7 ' ''Iti'j 7 *^ ' ' ■•'' U ' ' ■■ ' r.., V !■ V r M li. •■ , if ■ , 5‘v'*' i.» r ^ * '/■'• Vi''./"' ' ('-■•.a:'’': ::■ .' !• .‘iV , M "J ■ / f V I'i . '■ ■ ‘ V'7 I ' • • ;• X'" u‘ '■ I'i . ’ fO ‘.O.- , ■ ’ yfi il ..< If V- '!'■ ' : ■' «"’•'*• • > /■«' ’• ' '> • i ;t‘ ftw> r U S^- ’ . \ ^ vj^ V I I I « r.f' I < *‘ \:i ‘ i':. '■■• n .-5 .-,. (I,, f, 0 .; M'y ' . M ‘f. ) ■? ,'■ ... i jL* .* ' ( t,i'S ' ' . I , K ,:t J i... ■ ■ ' •y' ■'■• •"/ ^r^‘- - .*n, ^ :t’' mv; . . rik^ - r;i' '.7 ■ i - ' VI y ly „■ ‘ : vJtvr .■■• •■ • ;;•> >-. -Jr. -: > •: ' -r- ' . hi- I I ];•:>.•«'• '. rJ- 'i r:r?r.f'.;i.A v • ^ ^ ^ . r . - V I ha: ' 'i 4 i.<-_ 'i- I I h IV y; t|f t'u’i , ! ' ) • j O r ! .. ;; • •• ■ <;, ''■ '. ’.:1 Cii " " ir> ■ b ■■■' ' r ■) < iM Iir'i.rvV T' . '/.:,i' 'r ,.i- . rr. , p; Oil , n^io , o" ■' . . ,V' ■ <' ,f ' ■■> ’■ I • N ^ f t f 1 1 ' . ' > T' . , ^ ' >•- ,:U- ■J ’’'#y :. -il": ■ « ?■■( ctvlv-^ lir'.' ' u.','’; i •: ' r 0^;y.. '■< - ..f . - . Jrp*> ' ■ri n.fX' i - omu -*.’ , » ■■>'>1:/?'' Iv'j TW'iJn'Ki ' ‘, f-" OD ' ■ I . J.i • ; i'i';*! t-vfc.'fjv " n.rttiitt : •• ■ 4 •:» *■ liiloT. h 1^:, r I M '* M -•5 4 *'?. ' ' .1.; • I 'XiKr'ifol' :i > .'■ Mh 'll'.' i X' f. 4’ ) ( j, L ' ' ) "*> I ‘»f ’ • 1 I 'T f ' ’’'T.'l r , Ll • * * ' ■ T »i f i|.i':)r '.■ ■■ M'*'*.* ;l i5t ‘I " .: . '■',1 •'■-'I ’"',>■ f.o ’■ I I n. . t J i 7 posed of two partsj first hydrogen yielding or cellulosic constitu- ent, and second the paraffin yielding material* Ihey illustrated their results by analyzing the gases given off at different temper- atures in the destructive distillation of coal and try to j^ow that the paraffin yielding or resenic constituents of coal break down first. This theory has later beei proved erroneous by numerous in- vestigators, among \ihom are- Porter and Taylor (1916) (18) \iho, woiiring for the bureau of mines, carri ed out similar work: tho unfortunately on weathered coals. Th^ arrrive at very different conclusions from Burgess and %eeler. 'Rieir resul ts, \flil ch are very accurate except for a few minor details prove that the cellulosic material of coal is the first to decompose upon e^osure to heat. Pictet and Ramseyer (1911) (19) made a study of the solubility of gas coals in benzene. They succeeded in isolating h exahy dro flu - oraie ^ich led to the conclusion that coals contain among other constituents polymerized hydroaromatic hydro carbons. From tire similarity of distillation products to fractions from petroleum dis- tillation the authors conclude that coal and petroleum have a sim- il ar origin. Wahl (1912) (30) introduced an imp ro vesa en t in experimental me- ttiods by mixing the coal to be extracted witii some salt soluble in water. Clark and Wheeler (1912) (21) extracted a bituminous coal with pyridine and tlioi the extracted material with chloroform. They found that between 30-40^ of the pyridine soluble material was soluble in (Siloroform or benzoie. The analysis of their different fractions and of tire original coal on the ash free basis is as follows- > '"VV. ii*'' l;:,r •■?'/' .H \ ' ■' V '■ .■' I < - ■>jy ■t ff f.' ■ ''i ■A ) II ' - • »,•• ^ ‘ ‘ .r: ilr.’Z.L'j •' '* '■ : (>: ■■ I ; ft'iifi '. « A-.i r i n ' v>» -I r-t'jr. \ ' ■ ■ ::i • v« (iJO ,r.i 1 .': .•.’ ' •<>' ■ • •'■ ■•■•-«.< 'r ..■' . ; ' ■»' 1 ) 0(11 9 , . - • i ■ i<; ^ : I ‘ t « ) • •• ' t '•■•' JJ .TH''*,v ' J';'!< ■ ' I, ■ ■■ " ’ r »'i ’ ■ • '.' 'u '• , ■f •••*./'; : , • ■' I • .w V' N..&-* :»• • . *.'•» '.■• ;...’ ' •<;?•*:•.” V /t'Ij I., 'j i>n ^ ‘ n ' ■; 4.!. , ' 'To , '‘C. * >»r', ^ ' o J t; r t-'. , f%4> '■y,-\r > ' i"r :< ■)■ u ■ •*(' ■ fi^'n lil-fr;. \ ■.'■') X i ‘''O .i'.’ r tv } i: 7:'i r' i « •y'/ ■ ?v. !•' fiu:> , $ r4 ■ T* = "■ • 'U tf ■ • . .■••«■*'.•■;.•*< ! '! - •< 4 * *'.» vI'a' ‘ •■• , . I •; ' .> ;•:» < ' f) ■■ •r') ; ' ► f ' *r . k ‘ a > I ' ‘ i V *' {'5 fill'"’ fn V ? r,^ ■ ■ 'V'- .,1 ' 1 i'o'i vj H ii i; [ •’ S ' (r 4 ; I '5 . ■' ! >q t:n^ i r‘i f wi*“ rf"; * n r* 1 1 ' ) ; ■ :< . L - r, 'f o . ' ' 1 '■■" ) ;•':■/.) '■ "i ' I-) I*) (I :'>} C.' (’.'i /■' • ririft-O.is' « r V/ I.i' ‘ f mwf i. \ ■ . ( ,! I f ' vJ ■ 1 .' ' (,.4‘.y 1 ' ,i:‘ ' ' •;*.,) nf> X*j V 'i pt ',|4 '/! ■ ' r,^') : • ■ ! ‘l‘J- 1 ^ * . ^ . /.<■ ■ - ,1 rjr ■ «'■ ' 1 Vheeler (1918 ) (23) worked witii pyrddine as a sol vest and gave tlie following conclusions* lhat an accurate quantitative determination of the extracted material can be made only if the coal is ground fine and if both coal and pyrddinelis dry* Most of Ihe ex- tractive material was removed in twelve hours in a Soshlet extractor Small portions continue to be ronoved during several we^cs* One dif- ficulty met here as with oitiiier investigations, is that the coal, extract and residue will absorb oaygen in spite of all precautions, more slowly at low temperatures and faster at high temperatures. This osygesi forms in loose combination with certain parts of the coal material* Pictet, Ram s^er and Kaiser (1916) (24) extracted five and one- half tons of Saare coal with boiling benzene for four days* From ttie extract they isolated several pure hydrocarbons both saturated and un saturated* Cherry (1919) (25) carried out a series of experiments to det- * ri . tvl rr:» ' *> V-‘"t.i ^ . ' ,Mf f'V >; .fff N '! . t • ‘ ' ’ I U -rli'''. ' •: ii.?: . '•> fl‘ <> } j < ■ ' iv.-: V -v ■■,' ,!) !>■'/ ■• •: ' 'r- •! J;'!;:;? '• ' T ^ ■'I M C'.' . *) ; ' ' ',{ .i. • .; ' ’ • • ' i’ " it, \J ii' SU'. ?**»•;• j '■ ■ r,. -j; tyb - i-r;' '•-■ • U 'I V'>45'.jSs\ r . ’» ' •>:■'•, r,' I \ ■I. '^.*rr , « • »fl \ ' ■ ijftY) • f ' f ” 1 " '• '"t .* ’ :n/ '• , .,'• r.-: ; )r , .. i ,]j •• ■ ' f '-rf-V .-' - ! ,’:i; . . •(•!.; . if.A ' f/JV! *)® i; 2X- r •i be ' ■ ■,; r :^y> ■ ■ ’ j ’’ r', ''*j { f * . ‘V •'■> • ■r'l »'« V!' r,5<-. :.’ T •■■>•]/ ' •; • . i i'. -« :, £ '•> f r V *I« . 1 . f <> it . * i\ 3£r" - r X.; (IT' <5 ',t;> ."'r-ty ' . t- : ' •* ' ?r ','OM :cf.! r- ^TilNor^'J i^‘ ';'rt " ■■ '■ : --.rf; •• -to ? f, ' ' -' f " ff'V' rv'i Lfrt » i' ,’ •{? r.Li ,. T J ■ ,■■•;■■ r^' — . • . Tt* • -■' •■• ’•■• * •: '• ipi i r‘ i'1: .1' •' j_l py**T'J •' '' ^ , . . »r'C A <>v,r, fr- > ; J.* - • !.v f» t- J 1 r:?oc» ^ ,:■• / " f Tu r;t> X ■’ » j,‘. Ti '‘V < .r, • ' ' iv- ,v - ■ f* : '• ' '••: ■■ *’ '> • ■" c '. ,» ,m' ■» i r.;» ■ •' . <)0 0 • ' ye i not V: Ti 10 . the true solvents from the reagents in all cases. The following are the most widely used: h Qizene, gasoline, ethyl ether,henzene- toluene mixture,naph tlialene, toluene, alcohol, anthracene, diphenyl, cresol,low boiling tar distillates, phenol -toluene mixture, and acetone all of wliicai are considered more truly sol vent sj and carbon disulfide, chloroform, sodium hydroxide,ni tri c add, ferric chloride, s el enium o^ chi ord.de, pyridine, phenol, quinoline, aniline, sul ftir dioxide, diphenyl ether, turp^ tine, ozone and potassium di chromate ivhich are probably not sucJi true solvents. Fats, waxes, humi c adds and in general so-call ed ” resinic” mat- erials have been separated from coal by solvent action. No solvent has yet been found for the ”cellulosic” and protein parts of coal. These terms are here used in the general sense current in tiie liter- ature today as originated by Prof.V.Lewes (2) of the University of Manchester (1911), to' apply to the two main parts of coal as separ- ated by solvents. The ” cellulosic?’ is the residue not extracted and is supposed to be a d^redation product of cellulose. The "resinitf* is the extracted material lahich contains the coal resins. Since it is believed by some that gum or fossil resins are not a true constituent of coal it might be said that su(h solvents as phenol and pyrddine, iihidi extract this part,attack only a non -coal part and hence the true solvent for coal has not yet been found. %ain, if nitrogei and sulfUr, that is the proteins, are present in both the cellule sic and resinic parts of coal either tiie solvents used have not been the right ones or else the protein was decompos- ed and scattered since the resinic extracts show no protein content. Ibus it is seen that much of the work with solvoits upon coal has led to the very generally accepted throry that the organic con- ■ v^. y {.: :v ' "^5 9 ' •■* . :'.r.o‘' •, . • •*:•.■• -'■ *".il tfif “*• o* ■ \) {'^n ?■'•■ > >.’■■■ ffrti ?{••.•> f, fjjj ’ It ; Y r>' - rr^v liH* r f ■ K^'' , ' ;i'' r , I'j'» .ti;r ^ ■ ' ' . t .' Jw f,r •'? ” 111 \ ■ ’ >*'-i ~'C^' ;?>iv . i; '.{.’I*:, ft. ' »* ' f- r >V ! • .;y , .rr(>; j « V‘'‘‘ '^'v ’■ ,•:- 'Tfjs;,^ 'ivV'!' . ^-' * *iV -r L-? fi-' ur> i.i ‘ ' 't— !r j . . f •' ;vtf Ul<. . ;.v.\- \ Cr*l4'^« • •• '••^ J»Tji>a-r. /' I' -• •■'' .’ ..« .-i .Tf i- " ' I (j f'f: rf V ’-' ?.»ii i o r.'r'V", ; ♦ "i - (*':'? •■ :f-’ .:••• " ■•;vj pr.). -j < J' ■ •:•. ''-ip r’ -• •> * ' ..' .' * ' •V'* rt» 1 ' ! ! ■■' '■- ‘ •■ ^'■'. ' • ;■ i' i * r ■ r f-,. '1 . ’ ■ , 1 . 1 . j j \ » 1 VP-!^?.-. :- : .■> Cf.rnvv ■t- sV.'K;; fti i*' ^ u (tcl , , ■• r f A ■ -i r- '*- ! '■ f'^'O i't» ■ ,: .CrNf : ' rj,' 9lfi^ 1' ■„■■ ■ 1 t ** ** {j'** ffr'i: /'*♦» " » ■ , ■j^ j n 'f ) L'v<. ;, ff. ■ : .t k 0.* t 7 rM.'" fjy j.t : -‘-i.'l : ^;: ' i n ' ' '. n T .; .fp .,V5(- ri(>>'i , •__,) , t/> 'if:!., :...J 1 '• r ■:• i'.' , ■ '■ ’ « " . "■ '■ : ! . ' 'i d' ^^'r t"’ '.(A •■ .-a <*., V ;, i.T ' '- ■ . /»»,*]'<■■■' 1 h irb V r i ■ '.i . ' ■ . Olii ' 1 '• .. t ' , . .f4 '.'.-V Jl- r.i' ir • r»P ^ I - :i ■-■" ■■"■"' V rn'v •r i'- ^ i* *th iny; ;/►£', Cl j- ' : -'pr ii. 'kI f‘.i' j ' .i.\ \ 'iA'. .‘ fj''>.i ' ■■>; , ,1 ■■" *1 ' ./■ > '. -l* !>% .■'T • * 11 . stituents of coal fall into two classes, the cellulosic and the resinic. Ihe resinic type is tliat claimed to he present in the ex- tractive matter, \ihil e the cellulosic is found in the residue. This general hypothesis has heen substantiated hy numerous investigators using various solvoits. A very plaisible interpretation of this theory may be stated in a slightly differoit way, ?hich may give a new aspect to the subject. According to various wort done on the constitution of coal it may be said that there are two kinds of resins in coal, the original gum or fossil resins afdthe plants, and a d^redation resin product of cellulose or the cellulosic res- ins. Ihe extractive matter from coal generally termed ” resini c” may thfti be thought of as containing both cellulosic resins and plant resins principally the former, i^ile the cellulosic "residue” con- sists of tme cellulose plus cellulosic degredation products. Thiessen (28) in his microscopical study of coal, bears out this view in the discovery of more triily plant resins in the "attrdtal" material, and the cellulosic resin or ” an tli ration” material ^ihich he claims has passed through a hydrogell stage. The cellulosic res- ins may be tJiought of as forming the large part of tJie bonding mat- erial for coke, vSiile ttie plant resins yield the paraffins and tar. Solvaits used so far have in general extracted the resins asa class without making any selective distinction between the types. Whether the action of seloiium o^ chloride would throw any light on this question was doubtful, but at least some action dividing the coal into an extract and a residue were expected* 3. The Purpose of the Present Investigation: The purpose of this investigation was primarily to study the action of Seleiium oxychloride on coal, in parti cul ar,i ts action as ,*iSS ' M N fin-'o tX ■jJ. ^ ;/'•!■' '.,'7 ■ v tf (, ! r,j t : !'•* .V , ’ fV fh ’ i K ■ / iT: I .’r Ufi' f'H ' iJff 'V' ill ‘‘ Oi>!> 1 ..* :jn': ‘ ''v'' ■' u . -'.Irr- fe) ■r-je.' '' I'iT, .*1 t(!» I0>’ '4T> 'Oi '•Lr^' * 0 i:! yf'l ' *"■ ^ -n I'll, **: ii“ *'ri, ' r n ' ' .{ rvf* ji ; r/ /•i' « r ' : ru.i, '■' t» 0 u .'■ ' • ; Vt'i i . ... , Pi «> . , - - ’.y: . : r»V> ,. ■ ■; " h' ‘ j 'V ■■•■'.' i-x: ‘Y.- 'j'/.O ^ . j c> ; v r; ,»v;^ ■. : r , ^ ‘ ^ ' ■ ? ■•' '*'1 . < '' ’ ’ ?#'> . < Oc rp' M ’Vi^ „u ■■' ! ' !i- V ' O' t ■' '"■iv r 0 . ■’•') . ■ 1 > ; • t “}‘o f- ■ ,i:W0 4 V^rC; i„ O' :■ /‘{fO', , ‘S.‘ ■.' .feiii '■ ;> f . ( ii, ■ •:«> ? ”'ri i'jl.J.r: =’ H'ti' i .) : .f :o- I'-, ■, .•«;■• *'-’^7?' *J' j- -.' 0 ' •/o CJ^' * . ' ■ . TvOiV ■ !r..' ! , ‘ " 0 ''d*' I'ior, *'m : 1 / «<*c rfl . ■ '' u,’,0 •■ ' r ' ' ■ f • ■ '■ . .) 1 :•■■ r !■>;' ‘V ■' ' ■ ' P 1" fi'i ;n(jfi.h V ' 1 1' ' f-'-' '•1' .f s r,, I •‘T'jr* *1,, r; ^ n tft ' ' .'I,;..; ' •■ L- A I'M ‘ ^ ,-! 'To •),;■' • <,’ : • ^ v// ' '' (t' ' ■ iM' r.t ■ f t : • : • f' T’T ' ■ t Di t;.’ -,n' vrt )> r>o ;,j X; • Mi. 1 .' '“a '.; J ■ '■ 0 :'»*tJl:;j|j etil '.i ■' J '.'■(* if .'• i '. i : ■■ ’/ ..; .;.♦ ■ I yj -n t« 5 > • )l rT? .» 0 ,: ' ' ir" I K' r’’ I : •*, ; ! i M ’. •'.-. ;)f ’T W, '1 12 , a solvent. Hie prol)lem is a new one in so far as a study of the literature shows. Not a great deal is laiown about tiie action of selenium oxychloride as it is one of tlie newer compounds, and as tiie reagent is not plentiful on the maricet. Literature on selenium oxy- (Siloride is limited to a few brief articles and to one brief moi- tion of its action on coal. Hie wort referred to is by Prof. V.Lenh er (27) \\4io is perhaps the best authority on sel^ium oxaichloride in ^erica. He reports the folloiving results from treatment of various types of coal with the reagent: With natural coals the bitumen is extracted leaving a carbon- aceous residue. Mthracite coal containing little vol atil e matter 1 ^ 0 ws little action. Semi -anthracites lose considerable amounts of extractive matter. Illinois,Ohio, cannel, and bituminous coals from Pennsylvania and Virginia lose a large amount of extractive matter. Hie insoluble residue was found to contain selenium and (hlorine. With powdered coals seloiium osy chloride reacts evolving heat. Hiorxiughly ignited coke loses nothing. With tJiese points in mind and remembering tlie extremely re- active nature of the solvent and the difficulty ivith ishich it may be handled the present work was begun. It was decided to confine tlie investigation to one representative coal. With no preddeit as a basis upon ^ich to b^in worh the following general points were laid down as a basis uiion vhich to conduct the investigation: 1- To ascertain the nature of selenium o:^d[iloride for the purpose of classifying it as a solvent or a reagent. 2- To determine the most efficient method of handling it in con- nection dlh coal solvent work. 3- To observe the effect of sel^iium oxychlordde upon coal both 13 . physically and chemically. 4- To observe ttie effect of the solvent upon any substances knowi to be in coal, or formed from coal by such processes as des- tructive distillation and carbonization. 5- To note any oxidation effect. 6- To attoapt to separate unchanged one or more fundamental parts of coal. 7- To distinguish if possible betiveen the various types of resinic material in coal. II* E^erimaital, Part I. 1* The Nature of Selenium o^ chloride: Only very pure selenium oxychloride can be used in solvoit \vork The solvait hydrolyzes very easily and must be protected from mois- ture and even from long exposure to the air of the laboratory. There are three metJiods used in manufacturing the solvent,but without spec||* i al app aratu s i t i s very di f fi cul t to tu m ou t a pu re p ro du ct an d protect it from hydrolysis while making* Hydrolysis throws out the red, anorphou s form of selenium and converts the oxychloirLde into the acid. Because of the e^DOise connected with the purchase of tire re- ag^t careful use of rather anall quantities was necessary.lt was found however, that accurate results may be obtained from use of an all portions of the solvent provided a reasonable proportion of coal is taken* Surplus of selenium o3gr chloride present during ex- traction does not materially change the results of its action on a given quantity of coal. The highly reactive nature of selenium o^cJiloride is estab- lished by a study of its chemical and physical properties, and tiiere •< c ■ *v-^^ fi r# ■’»* , i ' '■ fOijM -vi ;>{J ' ' r; J *^v;tM" ' ^ >ft^ I «'V ! '■-' '■‘V’ t • ’ "’ I '■ '' • ■ lii'jpr '. :i . ... 1 ' ^ fiJ ;•' ■ ' , -r'M I '. ■**■ f ;■ . r<‘ • 1. ••■; ./■/: ^ > , * »v v^' n*’ •. ' -f 1 ■'. •/ M i; ‘ ^ ‘ ‘ ■ 't • . t .. w' J 1 ’ ‘ '.M t/ytii-*'*' ‘ , • :.i ' f !•»' k , /? ' ■(' t > ’i/rr i.' ■ !. f I' ! n: • ( 'I‘> ■ • /■..«, Cfll^V- , I ;Vv^' h '.' ■•; :.' ijifi ■' i '} .) ^ ^ •» • ^^.>■^ r );c I '; • ;'.> "I v^i, ‘ :*i! t AlV' . . . ; , , " f . I I '-*,1 \i "• .. •,»••/«. P'v.j to * I) ): ''■ •>, ■V.; 'yu :■ . ■ * I - V, •/ . , I’vivi ‘t .-iiA i iiH i i ' losion \ri.th such substances as ihite pho sphorou s,po tassium, turpentine, and sometimes with un saturated hydrocarbons. The halogens are dissolved, carbon in form of graphite, charcoal, activated dharco^ and the diamond are not attaiSced in the cold.Practically all metals are attached,more rapidly at higher temperatures, to give the chlo- ride of the metal and red-brovm mono chloride of selenium. Carbona- tes of Na,K, Sr, Zr, react with liberation of U)2*0ther carbonates are attacked slowly .N atural asphal ts, resins, and bi turn en dissolve ^vith ease in the cold #ien they are of un saturated diaracter. Gums, resins paints, lacquer, celluloid and glues di ssolve. In solubl e phenol! c con- dQisation products such as bak elite dissolve readily. Vegetabl e and fish oils form a rubber like mass. Protein materials dissolve more readily ^en heated. With carbohydrates, cellulo se is not attacked. Starch and sugars are decomposed vhen heated. All forms of rubber react chemically in the cold. The chemical (haracter of the rubber is (hanged. The more .U': 'C ? r 1 ij I't 7 >1/ : I;: • • •■ o V. !• : o -‘t ‘ i 'j . ; . ' ' ’ . ff- >. ’i -I ra '• ;; -■ tf'. ; Vt’: iV ;jf r''; ■:.vK: f '* r •r'l if* :• -3 f ^ ■’f . ' 1 i Xi;' r.-;» i ' ■'!E» ; 3 tr \ u 'r. / ' ■ • 1 ' , , y . . ■ V "S' !t! ) r ['■ n ’Jf 16 2* The lype of Coal Used: For this investigation a high grade,low sulfur Illinois Bitum- inous coal was used. The coal is a fairly good coking coal from from Franklin county going under the trade name of Makitan,and is very representative of the better class of Illinois coals. A large supply of fre^ coal was obtained and stored in size large enough to prevent rapid weathering. A portion for analysis was ground to 60 mesh after air drying, and the follo^idng complete proximate and ultimate analysis made upon the Air dry basis. Percentages were computed to other bases as indicated. Analysis for: As Rec*d. Air dry. Dry. Combustibl Air dry loss 3.0 5 Moi sture 6.28 3. 33 - - Volatile matter 33. 21 34.36 35, 42 39.08 Ash 8.74 9.02 9. 33 - Fixed Carbon 51.77 53. 39 55. 25 60.92 To tal Carbon 69.93 7 2. 17 74.62 8 2.40 SulfUi' 0.998 1.02 1.0 54 Nitrogen 1.45 1. 50 1. 55 Osygen 8.011 8.24 8 . 55 Hydrogen 4.601 4.7 2 4.90 B.T.U. 12, 451 12,8 43 13, 280 Unit coal 14,810 The analytical wo rk was carried out along the lines a used by the i/Lvision of Applied Chemistry in coal,gas,oil and other fuel analysis. Ihe Total oarbon vias deterniined in a Parr Total Carbon appai^atus. The SulfUr by the Parr standard sodium peroxide method. The Nitrogen by the Kj el dahl -Gunning method. The Hydrogen 1 V »V ■' .* C'iS» K ; . ’ ra ;'|A' ' I'i. /yr.cji H 1 ’; . --'l I r" I:."' ■ ^ -r •_. ■ \ ' '' r^-: . r '■■'■til i 'yj V(,i '‘i V’'ti 17 % and Oigrgen were calculated by use of Dulong’s fomula excluding moisture according to the Uni^cersity of Illinois method* This method has i^own exactness better even than the ordinary combustion methods. The calorific values were determined in the Parr Oxygen and the Parr Standard calorimeters, several checfe runs being made in each The unit coal value was calculated according to the Parr formula. These methods were used throughout the investigation \i4ieneTeer anal- ysis was required. Any deviation from the standard methods will be pointed out Tihoi used. 3. Th e U se of a Neutral Solvent vdth Selenium Osychlord.de: 1*. Selection of the Neutral solvent: In some lines of work with selenium oxychloride, such as tiie separation of metals, a diluting agent like ailphurdc add is used to advantage to lessen tire powerfhl action of the selenium oxychloride. This method of handling the reagent was first applied to the present investigation on coal. An attempt was therefore ma.de to find a sat- isfactory neutral solvent to use as a dilutant for selenium oxychlo- rd.de. Such a substance must be a liquid and must have no solvent action itself upon the coal, nor react in any way with the selenium o ^ chi o rd de. I f su ch a sub s tan ce coul d b e foun d i t v/as in ten ded to use it with the selenium oxychloride in various concentrations thus obtaining selective solvent action upon the coal. At the begin- ning of Ihe investigation it was thought that pure selenium oxychlo- rdde itself alone would have too great an effect upon the coal, pro- bably dissolving most of the coal and preventing any separation of extract and residue. After studying Ihe action of selenium oxychloride with vardous Substances it was found tliat the aromatic hydrocarbons benzene. ■ - • •>i' ■ -- -... I,-. ,-'■*• ■ >' ■.—• >»-«»A«i»iil if I \ •jt'' ■ 'f . ^ _ • ■■ - ..' :'■■ ; i i -i I ; '■ ''•■■ ;.(> .'i< ■ • ", ' ' 5 ?• ’ ■'!' i.! ^r« ■ r>'- ,’■«•' 5 ? ,‘ r .'0 '•t'.u ' ’! I* ■■ V Lrrv '•; %y rr' V'V.' "•■•' ‘ ‘ ■ _* ' . -I “1 •.vftfi •S'’ ;') ■.!' , ■ -i i'w' ' V t ■' I j / ! f' ■ ' .;^i !'■'■ r -':.;*- >\MPi y y tit L-ia y» 'I'}! :'. ! Y: •*.f Jt'rrf. t,,' ■' •• ■■ f .-yri^t •* ij(y -n ! . ' ■ 1 t'T f> I tp (}t ru? <>:■ ; -"■ *r > ; 1. <^1 '■1 :' i /'> f;a'fl|; ,-Jt> V’ ’ " *' > " j ->' itb ■(::■' It.ryti u-A; --yu ■ ■■•• vv»'.J ,;r ' '■ ' ; , • • • > :se 'T ' 't^y ' ■ •} .-'it • ■' ,n ‘ .jl-. ' ■ ■■ !• • lfel. 4 -^ tV • . : ;, 1 ‘A »V 38 rK? • • ; ' ;•;'/ ' , ' ■ ^ ' ’ • . Sy.i'ff; ; Q-i ■!)ft*^ '•./* H -.7 ' '■ f< ." :r?l‘ £>*1 ,. I uCK'.,' '■»'■•' ;*.:. ■••- ' .>• . r. ' '. ■■ •■ ' ■■ ■ ' ■■'y , :S ' ■• ■'*■:•■ 'i/,4' Jry „' 7 * r:s:i v 1 -^w - 1 • i: .■a f _j-- ■ ii ‘-T . • y J'fir jr. i «.! K*r;{ n#.'v.-< -s f .»i „;j; i?t »''■ '.! .“ii.' V}* ■ !' - 1 “ ••> 4 . .':'• . .■ :*..• ' ■ •••;» ■ >.f%V'T (f> ' /»' • .J ' !* : o.a {.r '■■T ■V? '•; 1,3 * V i ^ if’"') •■ ' ■ ‘ • ' 'Z''' . 81' t r 1 'i M j. .•* < '..v ■'■ '.:/ . 3 v; 7 ! !,» . -i:, :: ' 1 ' 7 r > < i- f f. ■' ■» 5 ' J ii h \''.'‘, '7 ..^ »■■ ir-'. ; , ,;.J 18. toluene and sylene answered the requirements most fUlly. Hiey are miscihle in all proportions ^vith the reagoit with which they are not supposed to react* Two extractions of fresh coal were made with each of these hydrocarbons to detennine which had the least solvent action upon the coal* Five grams of co al , after h eing dried at 105’ for an hour, were extracted ^vith 70 cc* of Ihe hydrocarbon in a sox- hlet extractor for seven hours at the boiling temperature of the hydrocarbon* The temperature of extraction varied of course with the hydrocarbon used, and the time allowed was sufficient to allow for complete extraction since maximum reailts were desired* Upon com- pletion of the extraction period the coal residue in the cone was wa^ed for two hours with 9 5^ ethyl alcohol to ronove the hydro- carbon, said then with ethyl ether to wa^ the residue free of alcohol and leave the mass in a condition ready for quick and complete dry- ing* The hydrocarbon extract solution together with the alcohol -eth- er washings was poured into a an all distilling flask and all but fifteoi cubic centimeters of the liquid mixture distilled off* The remaining liquid containing the extract was transfered to a sjnall weighed beaker and evaporated to dryness on an electric hot plate* The dry extract material from all three hydro caibons was d.ark brown and rather gummy* After drying the extract was cooled and weighed. The gain in weight of each weighed beaker indicated the amount of extraction from five grams of coal by the hydrocarbon solvent used* Percentage extraction was calculated fix>m the weight of the extract The residues were removed from the cones and dii_ed, together vriih the cones,in an electric drying oven in a stream of dry air* %en dry and in equlihrium \d.th the atmosphere of the laboratory the resi(Jres were cooled and weighed* Percentage extraction was cal culated from .1 . K 13 ' ' ' V'.' ^ ■ ■' ■ .... — .■? I! ■ \') frV: a: r --.•H >Af^ ■ ■ I v Ifl r.»-'- ■; f n '% ■ ,'» , 1 - ' '• I.W'J I. f r J'lfc ' " ' .0;-,V . r^;.V * II. '(ftd 1 'hU’J’aV)^ ., \ 3 >r <• •- \:v:r\o ' ’ . 'T’‘ f,' i C , .' ■ t,*t ' ' " !}■ il'i '(JIV ■• .tv-v 'f ;• ,y;i ,;>v.,i ' v •> ms ■' .' i l^’- .> /■*>., H.r: ;*t ■ 5 '■ ''C >'• ' «;!■>;'. j „ '; ■•» '] .J r‘ ’’Hp i^Vf , ' ■ ','l' ' I ■! V X’’*' " 'v .■ ,, , ' \ ‘ I' •' Cv'’'" :' ' ■'"'I.,' ■ •!• !/■ > .i/r v - ■'• ;! 4 ’xy the v/eights of the residue. In all cases the percentage extraction as calculated from the extract was much higher than that calculated from the dry residue. There is a possibility that the extracts were not separated thoroughly from, the solvent. Extraction as calculated from, the residue was taken as more nearly correct, Reail ts obtained here closely check with investigation carried out elsewhere along the same lines. Table I shows the average results from, this test. Table I. Solvent io Extraction ^ Extraction (calc. from, extract) (calc, from residue) Benzene 5.41 2,6 2 Toluene 3.30 1.77 Xylene 1,22 0,25 From these results it was concluded that :^lene was the best neutral solvent to use \9lth selenium oxychloride. The ^lene used was a mixture of the o,m, and p xyloies. The clear distillates of the xyl ene, al cohol and ether were evaporated and no trace of coal extract found in than. 2*. iVpparatus and. Proceedure: A series of ten extraction runs were made using a toi percent solution of selenium oxychloride and xylene, and varying the time of extraction and the temperature. The apparatus used is shoivn in Figure I, Ten grams of air dried coal were dried at 10 5' C. for an hour in an electirLc drying oven. The coal to be extracted w^as then transferred rapidly from the drying oven to the extraction flask(h). lOOcc.of the solvent solution (OOcc.of xylene to 10 cc. sel enium oxy- chloride) were added to the flask. The extraction flask was a 250 cc. wide mouth pyrex flagk with flat bottom, three inches in height in M * ( <> V f 'I ’v,i' cj'.vr "**t . I. . r f»‘i ;( ’tij J : ff.) T;' ■•. ■ ->i V ;, V ! ■ I ;•(. j’: tKi * ♦» if' ini 1 ' 1 :»- ‘ . ; A i * .■ A.. ' i'*- - > 'i'. *i», f ti'J,) ;r,t: ‘.f h ,,, • . { ^ ' ’ ' ■ ' -N w ■ 0 . ' t ‘.•, 'V ' .r (IB r *}ii ?; »Wifi (\;/ ' I r. (»).-■ ) iJ- L r i r ' r.' ■ *1 . .V . . r r J- fo s? Tviv. < . ' ’ r' i ^ 0 ' ' ' f » / ij* r • ■ ■• . ■"' -y^A) t> ■ n .) »: r, *'* '■ r :;*1 v ’ . •.< • <•> ;• ::■(- ; , • ‘ ' I ! ' vi ; ■ 1 ( Ji ‘j .; L-: F. T'M’jn' ! '•/ ft ' f‘j p'f/, 0 i*< uXor. -i .. HI ;.|r-F> U.tiu • ) ( i 7 /> I '7 I '■. .W*; •;'£ ‘*.<> 7 ;^ . >’'? '-.V ' •:' ', •■ \ •> •' ‘ } til ^ ■ ' •' f I '- l‘i o/f ' : " , I . ; . --'f ' V • r ^,. - :■ •■' ')i r'f-r 'rt;ny '; o ‘ -S' h>,yr • ■ 7 '•■ . i .• 20 the body and measuring three and one-fourth inches in diameter at the widest point. Colic: stoppers were used and tJie flask connected to an air reflux cx)ndQiser (i ) consisting of a glass tube three feet long and one-half inch in diameter. All connections were sealed i?ith a mixture of litharge and glycerine. Dry nitrogen free from oxygen, obtained by passing the gas through the drying train (l,m, ) was pass ed into the extraction flask from the nitrogen reservoir (n ) at con- stant pressure for a time to displace all air. During extraction a slow stream of nitrogen was passed thro the apparatus escaping thro the air cxm denser and out the sulfuric acid trap (j ). Contraction in the flask was relieved by automatic taking in of air thro the alkaline pyrogallol trap (k ) which removed the ojygen. The extrac- tion flask wras held upright in an electric resistance fUmace i^own in detail in Figure I, and was ^laken occasionally (firing the extrac- tion, Tonperatu re measurement was takm in this series of runs wilh an ordinary mercury thennome ter, and heat control was obtained by use of resistance coils shorn in a later figure. The first eight runs wrere made at rooro temperature and required no use of fh mace. Runs 9 and 10 were made at 100 'C. the flask being cooled dovai to room temperature in an atmosphere of nitrogen after completing the run. The time of extraction of the eight runs made at room temperaitre (20 ’o) was varied from five hours to 137 hours. Ihe two mns at 100 ’ were of five hours duration. Very little if any heat was givm off by action of the solvent mixture during the extractions at room ton- perature. During extraction the mixture in the flask was reddish- brown in color and showed green fluorescence. After extraction the contents of the flask were poured out into a large paper filter and the flask rinsed out with xylene. The res- idue on filter was well washed with 3ylene and the extract solution ^m4 Ifj '.ii' ff! :l .■' r^.!, ‘J .'; i'*', i. ■ < * » *k: ■ i .;• i.r ->«'i i*. •TV ^rf ' ■'■*'***'^*' • ft > .,, - i-'fc' ,"'■ ai'« r ^ : ■ r ■■■‘" ' • ' .. ' > ■' ’ ■ ' V ^ i<. V/(y r"'’*' ' ■ :s .' '> > !({ I ■ 'V ■' ■ ■! ■ r' . J '( ,i • *' ''■ 1 V ' '• ; \ ^ I. ■ , r|. 4f ‘id 'i ri). • if ' hn ... r^ is;- ■ I ‘ V< 0 fts ' F ' |\ » ':■ ! ■ •• - 0- ,;)’T,n ; '-1,1 .C ' i' 'Jfl ff'U' 'If . • Ti.:. ' I 't?t r ■' '.>V, or rl ■ : J ■ ■'.' V f' ' \ ff » r j V ■ 'in ^ \9 H ' ' r, J*;),': " I,' . - i i"r *•' ii ^ i .’ .‘ r< A, •..<.' .••■ -'-f '■■ : ' tHBv ' Y \v- o ■' .'O'c ! .> a '.'T ■,,, 'T>'» ,r. ^ _V' / •' ''il't , , ■ ' itttt', • ■" <■' 'T'f; ■' J !'• > •"■V 'i !i -. Vit f *♦ O' *«' r. . - ii pA-;» 'fT j i'' ;;iT J M t'i ; ('U'< ilfk'”' i ^ I ' I • . f.. .#?•■ . , «jv»t T'fw' ’■ f; !.f .V L:-k,..:,fui' j^r<- ;•'. .' i ’ ■' V .;■. : }T' , .'^ ' ■ ■• ♦ n ■ ; ; ' .'-• ■ „' ^ ■ ■ • ' Wii; J • ' ’ Tf.!; ' i r4 ^ H If, }'■ 'H • -i' r> ' K'‘. 'Vf .ir*'ti fvi ' » r‘*r 1*0 »V. ' f : I >>'XfiV fll. f‘rr.% '. ■ V ij -lo-qfi-J ■ 1 *■>1 j>0'' ■' ' ' '■ '1 u i.'/iji ; ■ \'y '< .“T • \ A ^ 5^ l‘i ,-|. *( ..r r ”r •( * •;.V .:WJ \ > ', . V H .1 . fir »ipF'''?"’T‘,.";i^ . ;r-’!P3r?5 ■■■:■ \CT' - ■ ' ^ 5 f 3 .^a ' • ■ . ^ ' ^■'. "■■ wm/ i-V . ■# L‘ft ,' j - R “■ <* ^ n ' : ■ V' ' f iRfi f It^ii. .r*> JC ,! .. » W . /*j ■ . ^ ^ , .’ • rf , •. Cl :i '• ' ' ' t .. AU J ' I * " ■>"'''■ r' ^'''' '■ V' ® , >;V::;.‘ir ' . -.■ ■. 1 :: I'/r Ve*fe..|Si *. v ^4 im:;, J|.|:i^ * ^ u .'■S ■ A,y.. '!>ttX^ia^j," f>9Xlr»8V ^^ 4 r, -^,.|^^.-v. (■(i'J'f.ft 9>>t«. !>!>■/, Li vlft.'ij'l' .)•:>■.>, r,' , .: y . Vn' ' . ' T ■‘■■-'^ '*ri._' I i) ■•^ '* ■*^!6'*^.\V > ■*' •’ ’■' / !’■' w ■/ . , •• 5.' > ■ '■' ' • ' - ‘ ’ 1 ;'^ ' ■ :•/>'';■' ■ . •.' :’ ^ ' - •"■ ’t(J«.*¥^’'rpff:'** »5?^ O'? '• « " 'Y !" ' ' •■' ■ V.' ' 0 t /iij ' •.;* , , ’' j'l'Mtr'jfi^ '•'< '^*y ■ ;. ' '^'si, . ■; ; : :;X’‘^M ■ : ^ ' t Sy '. $■■', V’ - ? fui *if Miip# nX^^l .V". -^T " ,*’:V ij *' '' ,. ” ' ’Aij^'ti •.,.■•'■ v' jfr^t ::V f ?' T-y ^ Vi> tf'i' • \n J' ’: 1-',,V. -i V. V.- If. 22 , together Avith the 2^1 ene washings caught in^^a lai^e Erlenmeyer flai^. The colarr of this extract solution after filtering was red~ brown, much darker -Qian the original solvent mixture, and had a ratti- er oily app earan ce. Ih e color of this solution deepened according to the iQigth of time and the tonperature of the extraction. Ihe res- idue ivas next washed with small portions of ether to remove the lene. This washing was continued until the wai^i liquid was fairly colorless. These ether washings were kept separate from the extract solution. The residue at this stage was black with the general ap- pearance of wet powdered coal. It was now transfered vhile still wet with ether to a 200 cc. Erl enmey er flask. Nearly all of the residue can be transfered to the flask, but ev^ with the greatest care some is lost on the filter and during tlie repeated washing ..pro cess. Ihis flask containing the residue was placed in the electric ftimace and connected to the nitrogen drying train. Outlet to ttie flask was secured by means of a small glass tube instead of the reflux air cond^ser. The opoi end of this tube passed into a sulf\irLc acid trap to protect the drying flask from moisture. Dry nitrogen was passed through this drying flask over the residue, whi ch was dried for from 5 to 12 hours at a temperature of from 200‘-280’u. Ether ftimes were bubbled off thro the sulfUi^c add trap together with the escaping nitrogen. The flask v/as next cooled dom in this atmosphere of nitrogen and op^.ed, and the residue now quite dry and powdery was transfered immediately to a weighed sample tube and sealed. The coal residue thus cleaned and dried, was black and ^iny,much like the original coal in app earfaice. They had only a trace of odor of ether about them. m ■V'fr.r )*■ »■) [.V' '(<* r'f,''"' J Y< ..; ; •'1 ♦ V ' >' ' r/‘‘/ i' -I •.■ at > ,Mj V' ■ Y.Ot : ; ) (f. V «' Jvne and extracted in a soihlet apparatus for three and one-half hours with 70 cc, of ether. The ether was only slightly ,U:.. 'ir, u ,* cr ' .;i-^.v <. V '/•¥t Tt "I'i ■<: ‘ ) < '* '■»„r ^ i . " ♦V,fi . r- t : niU • 'f '«f- r ' ’ -T- .' f. r I 1 ■ V ■ . :;^ -■ Sf\> i ( 24 , colored at llie endof the extraction. After drying in an atmosphere of nitrogen for six hours at 200 'u. the residue was found to have lost 0.2589 grm. or 2.46^. Ihe same residue was then returned to the soviet apparatus and extracted with different portions of carhon di sulphide for several hours. Ihe carbon disulphide turned light brown very quicfely. Red and brown crystals of selenium were extracted from the residue by the carbon diailphide and were throi^ out of solution by the addition of ether and cooling. 'the ten residues from the jyl oie-sel enium o^gr chloride extract- ions were analyzed for moisture, volatil e matter, ash, fixed carbon, total carbon and sulfhr. Ihe results are shown in Table III. Per- centages are tabul ated upon both the air dry and the ash free-mois- ture free bases,and are compared by a column of difference with the corresponding results on the original coal. The extract solutions from the ^1 ene-sel enium 0 3?y chlorid.e extractions were examined as follows. The extract solution was dis- tilled at reduced pressure from a tlaussen pressure distilling flask. A clear yellow distillate came over betv;een 20* and 70’C. Ihis distillate rapidly turned light amber in color and showed a yellow- green fluo resell ce. It araelled stix)ngly of selenium oxychloride and vhite fumes were thrown off during most of the di still ati on. Exam- ination of small portions of this distillate hy the use of vrater, al cohol , eth er,H CL, ammonium hy dro xide,po tassium hyd.ro xide together ivith various distillations and fractionations showed the distillate to be composed of sylene and selenium oxychloride together with a aiall amount of selenlous acid. The color was imparted to the dis- tillate hy traces of red selenium and a gummy compound of xylene in union with decomposition products of selenium oxychloride. ' ,T .’■■■(, -'-7. |K • f ' ■ i ’ " A : 7 I t .>■' ■• ' ■ thi'r'ti*' . ^ • ••;' 7 I ', f'Ttr 'j , - , ■,. 'r ' , \ ■ " '/ O’ • ’■ ''i*H (;^ ■ ’ > V'7 c • 'i .;*iA’. •••■',;; / • •' 0 too / ■ x:l lo , t^-r*r - ■% ’ . . ; f3i._ rr' r * 1 ,■■ r.; !' r'lM m : « 4 ' . ^ I f J a I tja ^ijw ■-• '> •: t 7- Ji;,/ •' -tl \.l rr,y, • p., 1 ' . J 1' 1 ' -- . . . . vt .»i , ... ' . ' ' 'Ji i: ’ • ■■ , ^'14- f , '1 '.’si'H 'i/i .’ 'i .•>■? ..ivt'.ic « '*:ov <7 r r- J* I 1 V * l or. ■ ■•(■.■' : ,#-r Jr*! ^ y ! .. - '•'•o-i:. If /. • , * l ' < :■. ' I n; .,' '■ t < jrf," _t .j 'i'ff 4'0‘ir.i.'^ ■ ft. ' ' 1 I ■ 0''i • J;4 ’■ > V 1 . 1 . i * T- O' X t" :ii n„ h . •..’ ! '■ ‘ o j f 'mqm&Ti oo ■ 'o! t f ^ ..wtt , y^\ j , , r ; , • . ,,K ^( .cr * •- ■: .1’^ 0 ■join-: I , 7 *i'£i} ftcMi.Jliii ' « ixr. tc: '■ -C^’ 0" ' -I": i/,!' ■ .' ‘'.{r V <, ' ? y.‘ r '■' ^ . n„ r 1 > • f > ... . i r'~ k -.•V'on” ■' ’.• ■• o»iJc:,'5 (’ C i'.axi'T ■ ! f ra'jt?> i ■f 'L.;n '•^• • ’'jjVJiJ' 'X< .... ... 1 t ttX’ ' . * \Q fi i *r 'I j* •' ... ^ . L ■ ,1 1 ♦, [\ ’A . ■ '/ >■ {'■ ??«.,- 1/7 0T3W. . ^ f' . '■ ,’S ' ' / »iS ■ • • ■ { F-.tO jVj , .V' . ‘ '. <> . . Mfn'Coj •'■ 'r ■» ' iU^n.l i’> t. i-Jri) ■ ' .; .; l:.i ■: f. .' ' . - . t:-'J . ., ■ X ■' , ' or.'rv. Fa ■ J-'i '■ '■ \ J 'r',» ''' . V? L r(0 i-VJ. I fjjy ■ lil i ' ' ■•■'■''■ ' / I-‘ ’f\ : '•* • > r ' V • . it jr;’/-' 'bt?!’. it •O’ r '' .’O.;mo:' O.' I . * ' ■ _ 1 ■ ■ ir- • *. V ':■} . I* •; •.; •<., ii.cn '* i!" «' *•'>« ' " I " 'i o •'t r fr.-*!! * r ‘ i ‘ ■ • » ' I ' ' ■ ' 7' ' "’o 'viv^^nM^ t ri\ ' v ' < »» S' *• !!. V 26 There remained In the Claussen flask a very dark, red-hroA^n mass appeared to he a mixture. Decan tation,hydroly si s and filter- ing finally separated the mass into tv;o parts. The largest portion of the mass was composed of a hro^vn,gummy substance consisting of the extracted material of the coal in combination with :^lene and decomposition products of selenium o chloride. Mu di red amorphous selenium was scattered throughout this gummy substance. The other part of the residual mass consisted of gray selenium formed by tlie heating of the precipitated red sel oiium thro^m out during the dis- tillation process, ^y extract material fix>m the coal is com.bined in this brown -red, gummy symp, but a separation of the extract from the addition compound holding it was found to be impossible. The material is very thick and viscous with the same yellow-green flu- orescence that colored the sylene di still ate. Upon standing the mixture, for such it must be called, tends to hard^. %en heated the mass gives off white fumes and the odor of rotten radishes charact- eristic of selenium compounds rhen decomposed by heat. Heating does not seem to evaporate the substance to any marked degree. IXiring examination of a large portion of extract solution from one of the toi mns the solution was shaken with a large quantity of water and, quite by accidoi t, allo\ired to stand about thirty minutes before separating in a separatory fhnnel. A heavy cloud of ^ite crystals settled out in the funnel between the water layer and the oily extract-:^! ene layer on top. The water and the crop of \hite crystals were drawn off and the extract solution repeatedly wa^ed with water until no more of the Thite substance cjouldbe obtained. The crystals proved to be very pure. They were subjected to physi- cal and chQnical tests including qualitative organic analysis and \ ' ‘r... '1-^ J(L.’^!tMUi» ;. V:- '■' ■ i' . ''"■/'h * o m " - c . ■ " ■s.b< ■ ‘f.‘ ft. M -; '* I '. ' r ^ 'j.' I» ■ ^iS 3 . u >> 3 pO ' f 1* ' 'H ' -Xi 'iH §ftJ: , rv ? 0 *> Ti' !« 1 1I..1, V; J.'IG ■ # ; --O-.' • :'t J ',s ‘s.r; 4 r;-(’Ci«'f.i •; ' .•:*. ;•' •»i- t ‘v; Ijn wr : ( u r:'9 t <=; Ji-i ■ . - -‘fi . "V .■x'l'r ,; . iaA”.: s .?o,' -nri MdiSi ''r '^iL ■ * ' - ^ • ; . ■■:?i . ^ Jktr: •I.-: ,* . } :■ I i" .' »''i!ii ' y •■' ^ 1 . ' -S ' * i5fS4t.iJ;' j'-' ;• I- '•I \i ' ■ ». • rci'i. ; f ,r ;•. "i /ir' i .. - .• ' i ; s Y'.' ' O i£l ;g';>; ,i£: Y '■ ' • f o ,;(> . h j t: •; irv r*- -u. *i:. '- ?'!£>■ 7-' /> r>-- •■' V,- ‘ ^ ■ ■ ' t’. X j i ' ’ ' '-f {>« ■#; ‘i* ■i. tt :ii (‘iJ* ' .; ;iiO .; :t-i:5tr' ' • •. ■ ■• .'•*/ ;.r*r ►r; ft} ‘ . • J. ■:^': ■: m-U.m '1o i^-iT •- m.j? ■ -.v.rl :‘>j- ^..v• 'Hr '-’f • ?;•.■ ■* ' ' f 'f* . ■'?? ".!' vri'tsx'.'' Y rp jf./ I - ; *v,». AV* • . '■ . ’i"' S'. b ■ ' •.■’ I wtn l_ x-0 'Y ; ' ';■• ■' :'f- • ^V! i.\- • o.r w « ‘ o v’’ • :> '>'?'• ' ’ ! > £ r.- ., ... . . . f f-ij: - ,'K' - .;»*j.' v|.,.;iA , .<: ■ ■ ' ; w loiv '. 'a 2 . '' ■ •■■if{ .-*•./> > X f] f *’:i < . rJrr 1 .o,:v \<.,ry f' ht :a\ f- *A'i Rv:-' . rJkj ;/^r!ni5 ^ ^■AXt .‘yi: ,* J ii\y' ► i- r' .'•■ M.f ‘>0 " lA' -fli ' -'i •■ : ':•; : i i ■ • >'■ ’Yia ^ : ■ - i '■/ i ' i '*.'■? t ■ '•» }=; f<:-.; ; 1 , ' r . ^--ci .• f :'i 1 -'L .« :'; ••.'f ^ - ^ -. ' ' n {'.■■■. n ■. ' , • . » -r* • * ■ ‘ ' ■ a ' ' . ■f • ■.[■■ ■■ .\ • '' ■ ■ ' •■'» i'‘< ■ , ' :'t ■.'.• •■■('■'I* ■ ‘ ? • * » ^ r. t- nys-ra "Jn. 1 ; 'if f’-i' ,,- V: ' ;l‘- ■ ' .' ; 0 • ''X :ivrM rri;,- . ./;}■■ r 'ft . ? !J :>. I'l ■ i: ri’i t'j tin.. ’' % ■ » ( »■" "A ■ ' '<••,. • o' •• » , ■ • • ‘)S. A'll, i./o>, \ ■.' A; ' J'ivJ'V*' ' v .'jf ■ f *‘ .'""/Vyiii, U fjfj •' ^ . -WTf'J ‘f ..:’! ■ ■' ' ..'I-' r >»•«•. -4 28 point* The data and results of the run are listed in Tatoie IV* There was no sign of tar in the tar trap or in the retort* Red sel- aiium was scattered throughout all the apparatus even in the gas holder* The gas generated or yielded hy the residue had a very acrid odor causing extrone irritation to the nose and eyes, \\hoi passed over lead acetate paper it turned the paper hlad^* Selenium in the residue combined with hydrogen of the coal residue to form hydrogen selenide* This gas decomposes in part to give red selenium found throughout the apparatus* To free the gas from the poisonous hydro- gen selenide so that it might better be analyzed,it was passed thro three tubes of lead acetate* The gas free of hydrogen sel enide, whi (ii would have caused an error in the percentage of carbon dioxide, was then analyzed in a U*of I*Modified Orsat apparatus as \?ill be ex- plained later* III* Results, Part I* 1 * The Nature of Selenium oiy (iiloride! It is evident ttiat selenium 0 2y chloride must be r^arded as a very highly active reagent* It is not a good solvoit in that it is impossible to secure a satisfactory separation of extract, residu e and the reagent* Recovery of the reagent by hydrolysis, with the liberation of any extracted material, is not possible in coal work as it is in the separation of aromatic hydrocarbons and selenium oxychloride, because of chemical reaction between the reagent and coal* As prepared selenium, oxychloride is very expensive, too much so to allow of much work \vith it commercially on coal* Selenium is however very plentiful, and ^ould there be a don and for commercial use of the reagent, no doubt a cheaper and easier process of manu- facture could be perfected* • j g-iWSw'wini V ll* ■ -^.' tl^i' ^ i i-* , :>--■■ -^.'■-./Jl.t— ■..V » : ,' >'v;-‘-'V.'.'(?i ■ ‘ :.v' r •r T-' 1 , • ' ■iV.Lt: ''>4ii;-/J! w ' » J ' ■-ia%V ' »■' " - ' ' fit 'T'' } n: ;l ■ ■,. tr; c " <■ i: ■ f!,-; ' S'''- ■'•■ , ■- ■' »■'' lj> ■ o ft 1 1 ^ **' 1 ' i ■ ’ '•' \ ■ \ ‘ '1 -•- 't-.V-'i-' ij f . • ? , )K ; ' ;■. X ! ' '•■ ^ i: ' f ^■1, , ' ■ • ■ ' ''' ‘ ' [ ' '' ' ,SMf7;V: , • f f- , - v;e '• • . (tO”; ' ? if^ir' . .',i \',' r »r t. ' ' V ' ■ .'■■■ • " * r.\ ■ ' *■/> * /> ‘ , j 1 ^ ’ ■ ■■ f I,, ••( ■>, I .[>: x' ft ■i : f 29. Table IV. Weight resictie coked 30.0 gm. from extractions 2,4,8, and 10. residue 22.79 = 7 5.96% tar 0.00 lype of residue: No sign of coke, powdered like original. Combined data on lun: Time Tanperature Gas Volume Remarlis 0 ^ 00 " RooM 0 15 70 100 35 190 250 Moisture on retort. 45 265 290 55 300 400 1 00 330 575 1 15 390 800 2 00 500 1800 3 05 555 2900 3 15 575 3040 4 00 590 3400 4 30 59 5 3600 4 55 60 5 37 50 5 45 615 39 50 6 20 6 25 4100 Gas data: Tonperature (m aximum ) 6 25*C. Time in hours 6.33 Total gas 4100 cc. including H 280 Gas analysis: I II Average Nitrogen free COg 8 . 0 ^ 7.9% 7.9% 10.7 02 5.8 6.1 5.9 8.1 C2P4 0.0 0.0 0.0 9.0 0.0 0.0 0.0 0.0 «2 27. 2 24.9 26.0 35.7 CO 7.0 7.0 7.0 9.5 CH4 24.8 - 24.8 34. 5 C2P6 1.2 1.2 1.5 ”2 26.0 27.2 0.0 2 # The Action of '^lene- Selenium oxychloride Mixture on Coal: From th e results ^ovai in Table II. it is not possible to draw any conclusions as to comparative amounts of extraction due to the 1 eng th 0 f th e extraction period or to the temperature of extraction. There is undoubtedly more extraction of iiie coal at the higher temp- eratures and during the runs of greatest duration, but the chemical action of tiie reagent upon the coal and the sylene varies to sucJi an extent th at no definite quantitative results are possible. The ^ I. ■ ’ a i ■* * (■ t - ) I Ft' ^ i| fc 1- 'V,^ i. I . r 1 f i 30. deepening in color of the extract solutions from the runs of great- est length would seem to point to more extraction or to greater chemical action# Extraction of residues from the mixed solvent extractions ^oiirs the presence in the residues as prepared of selenium and also quite a little extractihle matter \ihich m_ay belong to the coal or may have been added to the residue through the formation of some jg^lene compound. Analysis of resit^es gives only comparative results at best. The volatile determinations left no cobe buttons ^ihereas the origi- nal coal did. This is an indication of the loss of coking property through extraction. Tiie volatile matter nhile burning off gave evidence of chlorine and selenium in the residue. Vol atil e matter air dry i^iows a loss in six of the ten runs. The net result for the ten runs shows an average loss of 1. 335^, On t^e pure coal basis the volatile matter shows an average loss of 2.22^. for the ten runs. The a^i of the original coal was light gray. All ash from these runs except run 1 and 2 ^owed presence of selenium by their brown - red color. Seven of the ten runs iShow a loss in ash on tlie air dry basis for an average of 0.50^ Moistures on runs 2-10 inclusive show an average loss of 2.81^ This loss in moisture is explained by the treatmoit of ttie residues ^fliich excluded absorption of moisture from fee air. The moisture figures are however not important and may be ignored. The loss in asii is so anall as to be within e^qperimen tal error. Provided fee selenium o:^chloride did not attack some inorganic constituent of the coal it is safe to assume that fee ash remains fee same after treatment as in fee original coal# Fixed carbon air dry shows a gain in six of the ten runs // •> *.t rj .T' ■ ‘V ■ ' ’> ■’!•' AO J' '■ ■■*) . ” f-.l, ,rt r> *l "i ii t:-, 0.-: '■•■;•. '*.'■ ;,' r tTit,*'.' ’■ M >< ■< ■ ' ' : .*■' Oil I ’ ,\l I , i'l 'V. ;■ ifi I 4^"' ‘tii, ’ ': "f-T- :r-» '! > }..»i‘ijf.'.>'t ■ Jf; 85.|'It,i\;*-:. V' '• .!’i:i:-.'*'f!. * ;; ' ..t\» fi--' w ' i.iio't) T 'T-t ny' ' ’.O'' :jHv ' f’*' r.ov -4>'( " . ,1 V It V,; • ' . • , , , ‘ • ;ii 'V. oi> .v: r. ' . ' 'TO Hit'' ■]. ; ' : .; • •■ ,■ •',■ * ‘0. J tJ'.O ']t •■ /v« 1 TOrT.f) ■ •■ *' i »,u , - ■ f; • i.v • MT w. » .•,41 I f j ' ■ ' ‘ '• ■ ' , 1 m ‘ ' '1 ■ft- • io '. • , ' f. ' , ' T ). •.■>•: ’*•4 'i It i' iW I ' .’ i.'.t K J u J l>^.' •>r.''^/'.S :.i '■ 4 * , * • ■ .?■ u r' •H', • ' ■'.'yrk' 1 > 4 ■'> . t r» ' . f i' !i f 'll.’' , ■ V: O '■ o'Tt .'fV .1 r ■ ' ' (• J T 1 \ Ur ■ #* i,U ilK/j . iJ f .i, / ir:fi h‘ U’- '! i. '■' ' •, -T !. 1' «l C ‘ -It v-' ‘ '1^'* 1 if. ■ ~'r O' t ' '7 ■••->:.'.■}• V : ■».’» i’Tl' f> ' • LlOH^AT'Ti ' , "IfjP .’ r„ .'•-a r }' V* ' 4 X^ ' ■'r ■ \U TO f-- ft ^ .• r ■r. 1 oiT f f I ' 'io rt h:/*f • . ' ■’: i/T :■ ••■;*r.?T.ii to llPIvil'' ''■ ’ ! 'SiF'^ ■.•"i '>,:M *'‘' -j it ::■ j fv' 'fr'** ‘‘o f'C-t’'' »' i 'I'i i',‘ '' ''O 'v^iUUrt Ijotf t,fj is4 'Vi' ■’’ ' ''/} ■■’(>’■■!■ h O' »- I A’ '■) . . . . .' 4 * ;; ! (.'• ■.'( r:. . - JO 1*1: f*r r * V , . : ' r.''^ ■ ' f ifci ?' ♦jx V . . , ,i , ,i) if.' ' ' . f • • . :'!: tfi 3 lr«i( r * .. > 1" 1 * t t >: ..J re.. , « . • > ». > t , 1 A 1 •1 . .'.'1 . 1 t.'xxAl ' f" r ‘ Toi -• ••; * « ■ ... *» '£T ?.!•; t'! 31 , corresponding of course to tlie volatile loss. For the ten runs the average gain equals 4.48^# For a check it may he seoi that the aver- age gain in fixed cathon approximately equals tlie sum of tlie average loss in moi sture, vol atil e matter and aiSi. The Fixed caihon on pure coal basis shows an average net gain of l»31°/o» The total caihon air dry siiows a loss in all tai runs for an average of 9*6 2fo, ihis verifies tlie lossin volatile matter aibove. It seems that tlie anall amount of selenium oxychloride present in the solvent mixture j| ten percQit ) was able to ronove carbon in tlie form of un saturated compounds, and it is safe to assume that there was solvent action to the extent of removing some un saturated hydrocarbons and resinic material. On the pure coal basis tlie total carbon shows a loss in all ten runs for an average of 13.0 2^ It is interesting to note tJiat the amount of loss in total carbon on this basis increases with the increased length of the extraction time and with the rise of temperature. Sulfhr air dry shows ten gains for an average of 2»37% ex- cepting #1 vhich is abnormally high. Sulfhr was determined here dirett from the total carbon residues and hence any selenium pres- ent will be determined with the sulfhr. This is an added proof of the presence of selenium in the residues. From the standpoint of extracted material the use of ijg/’lene to dilute selenium oxychloride is a failure since it is impossible to isolate any extract and because there is more chemical reaction be- tween the xylene and the reagent tlian solvent action upon the cjoal. Most of the xylene can be recovered pure, but the selenium oxycJilo- ride decomposes th the treatment. The red amorphous form of sel- enium is preset everyT^iere. It is very probable that some chemical ■ Wh .■• ,"'.V :: ' A'f ^'' 'V r:;J ’! ; r ;•■ ,vi 4 •ip!:. !..: r< :• r’.-'i.-i rtv- > i I ' I i I ► it ! ,ii! 1 fi’,^ .) ,1 S' dh ••rt's '» C l» 1 “ ,f,v v>y ■ . r'- •'^4. 1 • - ; ... M ^ 7frr « . •5 : V Sg 'cJtH %i fj. y/g • *■ ’.f ^ .-'V fnfi 0 i; ■ ♦ H ■' ; ! i • ■ ' ' • • •'^ 1 • V ' »■*''•. • ■ ’ • - 1 , ' , ' ; . 1 1 . * .‘ *, ■ M' ■•■ i • ' ' • , 'J M .1: ■ :,ii. , ' . ■ ‘ V j ' : - it /■''• ■ ■yi 'r» '.'J ■•; ..•••'; tiJ. :■!' i::i f'Kffit’.*- / f. '"' ' '■’ 'J- ','0 I'l ' , y ' , r/.t '• f :7f/* r . • . ■ J 7. ^ • • ■ 't ;i 1 £.■ ‘ - i «'» ..|-;v>?< ' ' ' • s ■ ' * » ' - I it J * ’ 'l it M-f'- •■ , .f, v : .ff>i riv, , J; v' rtr ftfHO *' i ■ : ■:! iff7^„ri : ',(Hk ■ ' ' ^ ' T. -i ■ ' ■ *• • .. •' ••■ , ‘ <_/ g ‘ ' *. ,'f t • f , ' ^ ' ”',;-:'IT)' « ‘ ' ’ ■ A ‘ ; . ■* j \ y hr' k ’ t , ♦ . » ‘ ‘ 'ar^ ' ( ;j ! {‘'V.l ' ‘ “•■) y i '■ ' ■ ’ j •• '• ;' ■' ■* '■ ) J 1 . ■ • ' ‘ j -,• * v: -'X 4- ,•'? •-»:> ': .r ’ ' ■at iir.t ■f: 1 v ■» '.> »i * ',: , -^.T If'-} ' 'J . • 7 ^ ’ ■4 / - • r<‘ •>. V'T' ‘ ' i ■/' '/■•f'' ~ .,,?1 *<'/ fr't'jil c ' rtti rif .■ ‘ f ■. V ;■. v: !< j . 'I i// 1 1 f*S 5 -A ^ r'V'JVi) ' ’ • (Y ff \! >'J ' ^ ' ' “TT-T'*"^ 'T. ‘r X •' "> A (r'‘ I r-tr; 1 32 , action has taken place between part of the jgrlene and. the selenium oxychloride. The daricjh eavy syivip substance is composed of red sel- enium together witli any jgrlene compound formed and any tar or res- inic material extracted from tlie coal. From the nature of the mat- erial it seems very probable that some resinic material has been dissolved out of tlie coal, but this can not be isolated and identi- fied as such. The hard, gray substance formed upon heating tlie ex- tract is the gray crystalline form of selenium A^ich is formed from heating the red selenium. The red amorphous selenium is thro^m out from selenium oxychloride every time hydrolysis or chonical reaction taltes place. -Any subsequent heating of tliis form over 150 *C will re- sult in the formation of tlie gray seloiium ^^icil is not soluble in carbon disulphide as is tlie red variety. Selenium o:?y chloride pmb^ly forms some new compounds either with the re^nic material of the coal, the ^yl ene, or bo th.Un saturat- ed hydrocarbons of tJie coal might be attacked,but there is notway of telling \ihat the resulting compound would be. .Any chonical reac- tion with aromatic hydrocarbons will result from action upon xylene since there are no aromatic hydrocaibons in coal as su(3i. 3. The Carbonization of Residues: Very interesting results were obtained from ttie data on the carbonization run made on the residue from the jyl ene-sel ^ium oxy- chloride solvent, and from the gas analysis of this same run. There are no un saturated hydro carbons or boizene in the gas as analyzed. Benzene in cxoal is formed from un saturated hydrocarbons. The sel- enium oxychloride here evidently ronoved tlie unsaturated hydrocarb- ons from the cjoal,thus removing both un saturated aliphatic and sat- urated aromatics from the gas. The carbon dioxide in Table IV. may "3 V ' I'r . : l( :'ki'":’«it •fA> • r, < J ■ ■ :■ • » It:; -f .!> ' <•» ■ a: ♦ .' ; 't/ ;■'* t • s . 1 '. • • . >■ • ;.r7. »■; n • ' V '•■■■ ;Tc*'. 7f ■ : SI > '' " . '■- ',:-J .-\-i C'5 r» i./ ^<9r»f ■ '■■ ’■,• itruii ■■■ r ' ? ' > ■ ’’X XT '‘iOP '■•. 4 J ,{>'}:'< i-\ r„.- 7.: - i -tr< A- ,;' bt* ^ • ■ f ft : -.r>-. fO i. . u ; ' ' ■ ■ ‘ 4t-' ; ■' ^*>rr .>•■ j-ir/ (•/: ■.«' v* !. ’. ; c ,(>i ■ •; • ' ■, ■ '■'ip; 'f : . >'; ^ "t (!.'■ .: <■ 1' ' ■ 5' ’fOTVv' ,71 ”';■ - f ■ " •■ r I' s-j,',;;; fi:'- •> !■• 1,*, ' W ^es.CtU' ■■ '- ’' ^7j ' '>»!’ .c^ . TfV r?'-; \ C" • ' .i-'-r;' V--X AT' , '■■:i •ff rv* ,1'T/) , 7^ .•7: Ikti n it ' fA tf A V4 ■ r-f ,<■ , ;■ ' aaI; '■ nfl *' !) r fV.t ’♦n ^ I '7, r •■■ 4»0'> ■fx.yr'} ' ;< j- ■JtiffJh%(' ‘iff; : i.. f 77; -<7 ‘ ' ■ ■ 'H •*{! . -j i ^ • ■ * • ix? : >’ ' 1 / , f,{7 k > , r :0 33 "be slightly high due to the fact that any hydrogen selenide not removed hy tlie lead acetate train will he ahsorhedhy the KOH along \vi til tlie cathon dioxide, ihe hydrogen also may he low due to the loss of hydrogen ifliicii united with selenium to form the selenide. Ihe most important results of the coking run is the absence of tar among the products, and the absence of the hydrocarbons from the gas, as pointed out above. IV. Conclusions, Part I. No quantitative extraction of coal is possible ivith selenium o:!^ chloride as a solvent because of diemical action and deposition of selenium vhich is difficult to remove. Ihe presence of :^lene increases the amount of chemical action and thus tlie amount of selenium left in tire residue and extract. \Vhen xylene is used to dilute the reagent there is more chemical action at higher tempera- tures and in extractions covering the most time. Most of the xylene can be recovered,bu t its presence leads to greater decjompo si tlon of the reagent. The most careful and rapid treahnent is necessary to give a minimum amount of tlie red selenium Miich upon heating turns to the gray metallic variety. Xylene or any other neutralizing agent \d.ll not work xf±th sel- enium o^ycailoride in coal researdi due to chemical action and the (Complication of products. ^Vhil e not successflil as a final method, the use of :^lene will give good preliminary results and indications of \hat may be expected from the use of pure selenium oiy carlo ride on cxral. This shows that tlie most nearly perfect neutral solvents are not fitted for use as a dilutant. If any application of selenium o:^ chloride is to be made it must be in tire concentrated fonii ihich will eliminate confusing side reactions. I ? i . • ^ ' ' ■r-^ a ti. ;>iv., ’ ■ '>, f n j* . • j> • - ■ ■'. fi c k 1 " ‘K; ; ¥ ■‘ 1 ' 1 II f if' :t; Itr >r r; ■•.•.’J; f < \r ; . ■ 'H f 1 ; i « ■ - a -f; rrni a >;• ■ < • 0 ( f ■ i t- itwt/o % j t . -'-■/'‘v 3 I. Lr. •...vf' .. f, r ' t!. .ijcp :v '•• ' r’’’ ■ .' ' r t ..f.va :,} f-'. ' • ■'«;: : II fy* 'M-X ' ■■■ .'■^ ' )■>,; i - r. ^ ' • y 4 '0 *j / '.- . . 'O'* ■ t- ! 1 - ' o <1 L i i £i . f O;;:" ft'.. " o f r-*r-; ■*' *f . ■ " I • :<:f f„- (Ji i ; : * ■ ■ i 1.!' «vf ? i/\: iVJ i<=t i ‘.'’'n"': , i'll! 'in t! !!■ rr ;' ,■_/ !>■ ■ '^t ‘4 _ ,1 , i “> V ^ i * w* i'; .tfe* ■ ■■ b -:.•■• ^' : n'.j i. '■' '•':?:•?• I ^ ry' ; ' • • '"• » -• fP- -te i 3 ^/'.' vl )( t., '.• { ■■^> t^>r - ' i' . 0 . fo t <»VMj f : - r- - ■ : r ' ’ - " '■ on cris i r ’:■> i) >\ -'Kl ;v'; ■*/!•» . • ' ^ f ' ; “• - y . • c ' r. iJli*; f* fW ^Syi -• . '• •■» r? . rn/Jti.- n* f .,'t • 'i.’a v*w ,^r ' M -rj [ ^ , ■« C»; ■ -y : "/-J 1 pron -•> , .r n c L'«'; J > M , ‘ • . «, ‘M q-‘ a »lFa 0 ii|f ^ C,Vi /» • n •: ' r .; ‘ rt :;*v- *. M n ' 'i .‘‘ - ? *t iig[ '•' ■ I - j ‘ ! r j.i ’ w ' ' . ■■ Ml uK ff*"*'* ‘ ||l■l^ll^l 34 , SelQiium oxychloride reacts diemically \vitli some part of the coal conglomerate. Chlorination with substitution of red selenium results. It may be possible to free the residue from selenium so as to obtain a pure residue for analysis. The extracted materi al , \Nhi ch is probably resinic in nature, enters into chemical combination ^vith the reagQit and no separation is possible. New compounds probably of the additive type are formed. Selenium oxychloride unites with ::ylene to fom aach compounds as xylene selenic acid. Selenium and chlorine are botii present in resiAies from tliis method of extraction. Maly sis of the residues ^lows very little action upon the ash of coal, but a loss of carbon contoit both volatile and fixed. Selenium oxydiloride destroys the coking property of coal and extracts tJie tar or tar forming constituents of the coal as shoAvn by the direct carboni zation run. The extracted material is probably composed of un saturated hj’^dro carbon s and resinic material. Hydrogen selenide must be removed from tiie coke gas before analysis. More efficient treatment of residues may give a residue free from selenium and thus eliminate the H2Se from the resulting gas. V. Experimental, Part II. 1. The Effect of tlie Tar and Volatile Content of Coal Upon tJie Resulting Action of Selenium oxychloride: It would seem from the above results that the higher the per- centage of volatile matter and the more tar or tar forming material in a coal the greater ^vill be the amount of extraction with selen- ium o:^ chloride. If this is true coke ,witii no tar and very little ^XK.'Xi " :"*-■ l*y r:y\cjl ^ trt - Aa^''®?J -,..Ufl r(f»t) pff€K> ■< ;l.. M : (b V i . '.•' vr I >‘0 ' -5 'ii ■/ 1 (*44 >,< . t.riv'- 1 ►Ili.V 'ft: »>•■,’ . ' 'f ..■ i,V.; S'. ' ■»' “■•'• I f t y ^ :.-♦ fl p X - ^ ' ' ‘"i ) fX-' /*t‘ : .' Iff til f i ’ ("■^y (I’C ' -: '• ■ w I -f i„ ■■... r V. :f!j- , •••-{* ■ V'ri' '"’Ifr ■•' r.i; • n V <•• I ► ■ :. f. '■«■ , f'' ' ,Vi,t>’ t * A ' Ti I .* ' ^ > - ^ 5’ 'rss),/ »Ti' '-jnfi -,■> tv;*.-? ■; ■ 4 • , . ' ; ' ■ . “ \'4 : ^ • ■ .'* •* •'■i < ' , f ,■ 1--T, ■) ?j fV' iii /-<■ s' ::',(■ ,' '.i ■■>■■' • • i;U?.*i-'» I c. 0 i t. T:**-*?*./ T :v .- ;•.,••■{ -s *• ;j .t’ O irV .♦ • •• . -f ' ') .KJ 5>l '2 O . T il ■<■'■' ,■'!;»■*■■; ■ b'*;* fit. ■':■'■)' Fi' , !' 7 ■ t , ■ t'^'. y t\ -^rm r '. iK Oj'lJ? . ' X*J i • . '»;» • ■’■-!. I r: 'irrt 'f. : r:?: '. '»', a 4. >< ■ f ’'■ V V .'j ha:» f. S ' .•n *• '*,?(> ■1 , 'ffi «7r .• 'f ' » irr '1.' .: , - ...... ^'■3 , rT»>V 3 *': egjSf.c! X\'(lltij||^‘ <« <■ \X£ %> .)vie- • t i’ n X (L\ '? • f !,> «>■ a .)' f. X* 'f]f !, '-} , !'. ■ ’■ .■■•■.' •'{.tl '. ■ < ■■ , " 'v'.' t> )i'»J"}'xr3 '( M \j0 ■4- ', f , j t\':> y ::t <.■ ’■■ 5 .'., 'i-M tar;-: t ■■•'■"' f!J i' *■ ■•'’ • " ‘■"'’f' * ;.' r- . .,0 ■ ■••I 0- *. f •>,•'.} 'I;-'.' ' ' :*.•».■' ■'S ■ *' Ml '. ■■> il ' t P..' ’♦v’l'.tj) Qj!.' CrM^pi .-‘Jf ’■j- i *T ; -fe ■ ’ ■* i ' 35 volatile con teai t, shoul d ^ow no extraction, and a partially coked coal residue should show only a sin all amount of extraction. To test these conclusions the following extractions were made using pure concentrated selenium oxychloride. One to three gram samples of cxxke ground to 30 mesh were ^ak€n togetlier 'with a portion of tlie reagent in a small Erlenmeyer flask for from fifteen to thirty minutes. The flask was wanned a few times The (Jon tents of the flask filtered easily and there was no sign of any reslnic or gummy material mixed witli the selenium oxychloride ^Dhich filtered through. To the filtrate, rather "black in cx)lor, was added sufficient water to hydrolyze the selenium oxychloride. Red selenium separated out hut upon refLl tering this was the only Tsuhstance left upon the filter. The resirlue remained hi ack and in the same cx>ndition as before extraction. Washing with water removed all tlie reagent hut resulted in throwing out much selenium ^ich remained in the residue. A fresh sample of air dried cxxal was kept in a small stoppered flask witli selenium oxychloride for one week. At tiie end of tiiis extraction period it was found impossible 'to filter off the reagent or any extract solution. The \hole mass was brown -hi ack in color, mth the consistency of a hea-vy paste. Water was added and the mixture stirred. %en filtered the filtrate was clear water only. This show- ed that the extract had undergone aich chemical reaction as to pro- duce a gummy substance insoluble in water. Alcohol dissolved the substance enough to allow filtering. The filtrate was chocolate bro^ in color. The residue after being treated with alcohol was blacJc with a slight brownii^ tinge. It showed traces of selenium. A coal sample was secured ^vhicii had been ground to GO me^i,air dried,and partially coked up to SOO’o so that practically all the 36 . tar had been distilled out.Tliis sample xvas treated \vith the reagent for various lengths of time. The result upon filtering was mudi the same as with coke. The mixture filtered rapidly and tlie filtrate showed no signs of resinic extract obtained from fresh coal. A saniple of fre^i coal was mixed witli tlie reagent for a half Iru hour and tlie mixture then filtered, filtration was nearly impossible due to tlie thidmess and gummy nature of the solution. 2. Effi ci ent Methods of Handling Selenium o:^chloride: One of tlie most convenient and usefUl pieces of apparatus made use of in solvent worti,the soviet extractor, was first tried out. Aii» dry coal from ^ich all traces of moisture were furtlier removed by heating to 10 5*Ci. was used in the extraction cone, and a suffici- ent amount of selenium o^g?^ chloride was taken to allow proper oper- ation of tlie apparatus, uork stoppers were used since they showed less attack from the reagent than rubber. It became evident almost at once upon starting tlie extraction that this method would be im- possible to use. The reagent attacked the corks at once, dissolving thcsn to a jelly in a few minutes at tlie tonperature developed. The ifliole of the selenium oxychloride became a dark brown color showing marked decomposition of both reagent and cortc.Some of tlie reagent after refluxing ran down into the extraction cone and that too was dissolved into a pulp in a few minutes. Another method of refluxing was tried. M ordinary glass tube was attached to a small fl ai^c, con taining the coal and reagent, to serve as a cxm denser. Here the results were also negative. The rea- gent attacked the stoppers and was itself decomposed upon heating. !■ ariiied Iiy droly si 3 was noted in tlie tube ^here fhe air came in con- tact ivith the reagent. ^ yfr f *-■ to «i *• 'lyj' ' $1 I '4' • # . r < 4 -Sa if ' ' I' ) . I i V it; /? » r rv ■'< '.;• > i'.i tf-y ' 1: . J I’J.'OftJ, r •'• '. ..1,* ■'xj> „; ,'vft ' ■' •. ' ‘ '■ '> r. i»j '■i": i.‘ i.r *> ) / i- -j '>r(i :;. ‘ < V , l:. ■■'•I. A Icl iq V 5 'I oj» If' f.:.i •••‘.'tJ'li®.. ' tft flR '» tin ff J 't' .*'■ .11 f 'V" -o - >l:.’ V' 3'^ in * 1 1 *j f< • j»*. „i -in ,.1 I- : ' ••‘^ V-'. g‘r r.Tj ff iT ♦ -P f ; ; ;>,? /' •' ‘ 1.* i:"iOi: '• f ? « ^ g.jl : ir' a i) ti • iCr-*” -Tr -; ;■ H' ■ ,‘i' ,, ■♦ ;r, a • rff ■ ' r • ■' 1'',' f-' ’ .i ■■' V ■^■ . r ;, , r> ftoi .' ?» LPO f •• / 4 . i Jlfj . - 1 ' '>.‘i rTf* • '»nnc itf r1 O' V r I ; o . I ■ ■' ' ^ ■’' . '.'■ 'i <■? / T' ' ’ > ' ’.'■ ' . '■"(„•> ,0 -i'- ■* 'f'f f. ; C.l Vj fi oi' ' ! •' r.'.e f ■ tj f> ioiH' i ' • a 5 1 T^.- •; Njjl !*»,•( jf ■f y r: /. ’At.-' ?-y'i '■ y/ * _ -V V, • / ^ '■ “*“■ ; *■ r • ': - *" < L ' ' '-• if ■ ^' * . • • L *^1 ■< » • ' ' •m these attenipts it was evident that no heating of the reagent to anything near its boiling point was advisable or possible and also that no refluxing system of extraction could be used. 3. Extraction of Coal with Selenium oxychloride: ^paratus and Proceedure: Bearing in mind the results of the previous tests the fol leaving final method was adopted as the most efficient means of treating coal v.atl"i selenium o^y ctiloride. Twenty -five gram samples of coal were weighed out and dried in an electiric drying oven for an. hour at 10 5’C to remove all moi sture. When dry each sample of c^oal was re- moved rapidly and transferee! to a small Erlenmeyer fl ask (300 cc. )£aid suffideiit selenium oxychloride was poured in to just give an exces above a thin paste. The flasks were stoppered at once with cork stop- pers. The coal mass swelled a great deal and mucii heat v/as evolved, the flasks becomihg too hot to hold. The flasks were shaken and the (x>n tents stirred from time to time. Keat was ai)plied by slight warm- ing on azi electric hot pi ate. The total time of extraction varied from thirty to forty minutes. Preliminary trials showed that any attempt to filter at this stage was impossible, so at this point, a an all qiiantity of benzene was ad.ded to the flask and the mixture well shalcen and stirred. The benz^.e vlll not hydrolyze the reagait or extract any of the coal in the presence of selenium oxychloride. The baizene being miscible with the reagent forms a perfect filter- ing medium and tJie mixture whoi poured into a suction filter dll filter with ease. A slight suction was maintained throughout tlie filtering. The contoits of the flask were washed out into the fil- ter Ylth aaall portions of b enzene. The filtering became slower,but vlth repeated washing of tJie resicTue with benzene the process was ;» I ' . ' ' ■ 't % ( •' 4.>: . „■ , ' ' ^ n^'-' , ^ . .:«««. * . I S' ' ril 1 . '2j: . , .1 '’'■ t V ’ . '■ ■ .'it M ' ; ',' ^ Y>'’ ! i ^ ,’;L "'y Livr/j(rssi|lJ; * T :o'r 1 -.. > h ■a J‘f ' 4 i ij. *t '^ - ‘ ' ' B 'rt-i ni. ‘ •r(. = 1'-' ’if' »« '/nr/, > ;■ 'r f ■■> ■' fy\i - ' ' ■■.’' ■ ■ ‘ - - *;:.K' ' ' '-I;:'" i . ' 'J V - > '• ; ■• /r: .. . f’ • ■ ».";♦?■•" '-i' r- f' >5 ‘ ; > : i o 'U' I'i’.t i.M"" •• •;».*;: p.‘ .■ ■ ll ' ' ' ■ ■ i ’r ' ' ^ ' ' -V. ■“• «».t Ipi * <. ■ ?» ■»“ t . f " 07 O'- K >v ( , » t » 1 • i •Hv ,“.fr. I-:'**’- -Hv'- i ,‘ " ,- • *J'I f'y''- •I'f'i.'W ’•■' t> rvJf I'HI 0‘J "■' ’U- ♦ -7,^' ' ll rU'if' U . -• A ' !{ .>'•«»!;?•* •■' />: fS ' « <>• ^ . ' : V" : ■).*.. .;•’ {3! '.»•'' ail-? li 1' ; ■-• ■ . ■ Ici;- ; , i ; ' ;.■ v- ■ ' rrtirf'* 'i ! no ?-rri: f > : f :0 -^py: . jqjBO"'' I -M , . 1 ■/ J I irr f ,'■ ! ■; '1 F. _/ '• 'to, «-.• .i't ✓ d *.• , • i£- T \ og 1 j 'I 1 . i !^'a ■ I Mt ( ■ v< i>tw , - 'V/-‘ •• ■ T^STfua * 38 completec!. \d.thout mucti loss of time or hydrolysis of selenium oxy- <3ilonL(^e* Ihe extract solution washlack and very viscous. It con- tain ed th e sel eniuin o xy di 1 o r± de-b en aan e mi xtu re tog e th er \vi tli th e extracted substance of the coal. Ped selenium rapidly separates out in this extract solution. The solution was stoppered and set aside for examination. The resiclue on filter was washed vdth ether until all traces oi benzene had been ronoved and the ether washings became cl ear. These ether washings were reserved in a separate flask for examination. The residue was now scraped out onto a large filter paper, powdered with a spatula and allowed to dry partially. Hie colro at this stage was black with a bro\m tinge attributed to presence of selQiium. The residue gave off a strong oclor of chonicals. tVhen nearly dry the residue was transfered to a beaker and repeatedly w’ashed by decantation with b enzene until the washings filtered cleai. This was to insure complete ronoval of any selenium oxychloride and extract remaining in the residue. WasJiing in the same manner ivith ether gave a fairly clean residue iiiiicJi was dried in a large flask in an atmosphere of nitrogOi. %en dry the residue was put through a 60 mesh screen and tlien fhrther dried in drying oven for two hour? at from 50 ’-170*0 dry nitr*ogen being passed tJiro the oven. A residue was thus secured practically free from ch emi cal s, and bl ack and pow- dery with the same appearance as the original cjoal.The residue was stoppered in a sample bottle at once to prevent absorption of moistui-e or oxidation. After being washed ivith wther the residue dries very rapidly and thus danger of much oxidation is lessened. No quantitative re- sults may be expected shoiving percentages of residue and extract .I-,:: '•v:!7:-'T'v f - V -:v 3 ; ; : - Tc --v/ ■ I;-f V*^ -■ ■ f ’■'• ■ ■' ' ' • f ■Jii !..•■ V.-; V.: ■- ;::; 3 ur !i> )H’' *1 . ' a *r r {'; Vi, ;" ,P> • y| ' ■ #, , f-'-.-f ', nii' *1 < * ■ ' ' , ■. •■■ ' ■ ^ ■ i V '} ^ f*--vr.->r,'Tit fp'* • Or?.*: ,v ■>iVs ;,;i ■■ ' M f «*• '* L') i i, I ■ I • »■•» •’•'isri'-*.; •■•;'. . ^'V “y’yi J >/1 ^ =;- ' ■i ry> o j ;-j ■.■- •> : . ■ fr ■ . • 1 , • ■ „•••■ ■• '.V i 1. 1 rj} •.":r ' < ■ . • '■' .' ■*> ;■' ft 7'.y- ',» .;• ' ;, ; *;'••■ ■ ■ .f .r, r\J : - , ’ t ‘ I ,' .1; ) r ,. . > I;/ , {:.;, ' ‘ " .■• .. ('■» .;i|t !• * r*.r’ •• <«!•; '( '\ .'■‘•V a. ' {••■ ' k l\ t. tf- tr'l id '‘V/"?) r> ■■ ■ f.ft* 'i K ' ' i ‘ f'l r:-^ 5v Oilitr V jjiO'rtK') ■•XK, r > rV ’•'0 JflJK" fitiH * ^ : ■'> • no ; : ■ ’ ;< 'll ii’>- r :r M '& ■ hsfi’c ; ' 1 ■ » .'’I ' m '1 at * 1 \t n rr,. ’* m- i A f' ■ oi,. ... >• •)> sti^' ‘‘V'* . ■*• * ,^-r : ■ T'-M !.;/3 . : ir i • /n> f>Cn* 'iT* ’ . . :ti: { I n'.j.bro '• . < *ti '• ^ b *i O c UJ-* . .' «i£ij c» i. — 39 o\?ing to chemical reaction, depo si tion of seleaiuni and lossof small amounts of residue through repeated wadiiing and fdl teri.ng.B ecause of this loss no check runs are possible. The nature of the extract sol- ution is such as to allov7 of no separation of Ihe extracted c^al sub stance, and this also rmders qiianti tative woric impossible. No extraction and filteration process ^\lthout the use of benzene to aid in filtering ivill succeed as the filter clogs and the reagoit rapidly hydrolyzes from e^osure to the air. Rapid manipul ation, exclusion of moisture, and keeping the reagoit well mixed with benz^e \^iile in contact v/ith the residue and exposed to the air are the essential safeguards for effective use of this extrQ!Tely simple method of extraction. 2’, Examination of Products and Malysis: Very little attention was given to examination of the extract solution. Preliminary tests Crowed that any separation of the ex- tract material from the coal was impossible because of the chemical- ly changed- nature of the material. The extra ct,p rob ably resinic in n atu re, en ters into combination with the reagent and is thus ciianged. Selenium is present in fairly large amounts in tiiis extract mass. Removal of this selenium may be effected only be solvents or rea-en gents vhich at the same time ftirther decompose Ihe extract. The ether washings of the residue were mixed tbgiether and distilled. Most of the ether was recovered pure,leaving a small amount of heavy, black extract material, b^ind. This was added to th( extract solution. The extract solution proper contained the extract, selenium oxychloride, and benzene. ELI trati on, hydroly si s, fractional distillation and other methods failed to isolate the extracted sub- stance. Most of the benzene was recovered 3Pa±rly pure. The selenium I) - ■ !. r ! : . ff.V.V'' • - *'' ' ': ■ ■» .<> 'iV ''■ ' U-’'- ' 7 , ••>>ta-’.*'it5C‘*‘ ■ ‘f'> ''■'ti, |i' * ’.■*•7 ■ iT't; > "-^j” ir^SW J ., k) > I : !ji-\ .'^^».H^ .’:* :■> ' * /• - ; “f r> o*.- if .''if*" ;.•* •■■■ !'»■ ■• •■ ./, '. ■•■■ ' ! : ' •j'’? .t • ' ■! ■■ ; ■ '*. Vv >. ; ■ .• . ’''-I'y " • :^, v.fc r. ' '., ; ■ '.f-i- » "• - >■ . v;'J -4' •. . 5 vr* ^ , ,^m:v i. ■;. IT’’': i i-oj m »- i'')*' !.• / '. ;vt ■V' ’ ' -v: : i ‘ ! . rtf ' tf / •■; 'iv ;■; . •i ) >. : II 7 - 1 s " ■ .•; f’ ['i.i n’;t. ^C^'* I'r

''■' ', 1 Xkmi :,*i , ,' .; ,'t . . .,’ '>.>’r.,V/‘: :?•,■' '" ' ,'i , » ■ J , i.i' , ^ .V ( ■ ■ '.'T 7 0 a 7 f-rv'- /,>? 'V ; ■tMiiaio ;| .'0 .') ' 4 i *J«r« if) :■ ’• ■> ,/ ' ' '■ i •VT^hr-'- 7iCt ' « '» ' . . I 1 r ' f, . l») , If- I.* »' .1 J »..l ■ 1 ‘ .■: ■ ; •- ' ■ ^ y* '!•'.* 4 ■' ■ . ’ V' r< r« |, Hi . \ j ' i • ♦ » ti ' ^ T* * r 40 . o clilo ri de present is decomposed to give primarily selenious acid and red selenium. The selenium can he ronoved in part hy repeated hydrolysis and filtering. Heating the solution or mixture converted a portion of the red sel oiium to the gray varlety,insoluhl e in any- thing hut concentrated sulfliric acid, \^ich further complicated the investigation.lt is certain that there is some form of extracted, coal substance present. The thi cfe, gummy , nature of the extract mass, left after the henzene and selenium oxychloride have heen removed, shows this fact. The color of the mass is hrov«i-hl ach,hu t the pres- ence of selenium gives the mass its color and so prevents the d.et- ermination of the color of the extract. The odor of the extract is that of rotten radishes vhich diaracteri zes selenium compounds. It tends to decompose r?ith evolution of heavy, hro^m fumes Then heated in the air. Distillation only dries the mixture forming a hard crust like dry clay. The appearance of this mixture is idoitical ^vith that obtained from the extraction of coal with the xylene dil- ution sol ven t. Th ere is one difference. In the former^ extraction there was present a 3yl ene-sel enium oxychloride addition substance, but in the later case the reagent does not seon to react vd th the benzene in the least. The residue obtained however was in nearly pure form and from analysis of it and comparison 6f the analysis with that of the or- iginal coal it v;as thought possible to determine something of the nature of the exti'act as well as of the residual material. To free the residue as prepared from selenium it was desired, to extract the residue with large cju an titles of carbon disulphide in which the red*, variety of selenium is soluble. It did not seem probable that this extraction wj[. th carbon disulphide would reailt in the extraction of V * -SV . hr ) !. 1 ..' r V,‘ 5 ' ' ' -Th : *(>v ' - . ' ?'!**'*l r: ■■■/ ' ■■■ a ■ ■■■ M • r? II ; I ... 4i’ jfl ' , >■ ( .i: 1 U' *.i V ■j. Ai <*;. 'X 0 K 'f'r' -^■ . 1 " , • I f .1 i ^ ^ ♦>»’■■ 1 1 ' ij ■'■{■' 1 o rrb i; J Ifi ■ V 'v’ M r) J I '•*• .i ■ •>'r-' . ^ t **'-'*^ ■■ *. ri ; / it. :Tr \ r ^ jr ' ' u . «v fi , T. ;- .' ■■ rXu-i f j t < :;i ; -t r,.;j«;. . , -J - : ,f.» rrj.|; . ■ — ', i '■■ • -, , • ,^r\. :i 1 ■ ^ ' ■ ' ■ * r '■»v' * r • •” ^ * , ' / ■ • ' if lyy •’• V* ’ '.' 't- ir< ' ■ ' V M ^.'t J>o' ,^;'t , ■ ■ . •! • I ; t-.' ’• ' o' 'i ‘"'tS ,f T . 1 . ■" ■ ' • 1 . • , « > % ‘. J . '■ " f?'> ' : r'iA pJt ■ i.) . 'j , ..-{■■ri.J'' ■) i <» o n* ; !■ r ! || •; ^ ,u' .r: ] tV • 1 i !t^ ■• r; i I” Kf*r.>u ' ' 1 - ' / . T >V; •0'< ' \ 'T.^ i?-' t. _ 'LiL 'ti .^1 •' r nr' •r • p ' ■ r ft: pj: ■M.'( f .f, ^ ••■ .i'\ I.:. , ;V‘' - ‘‘ 'O'* iri; ; r. -r ; 'H cr/; , Xt^'i •• : ' . .0 ; ■ ' ’ 1£ ' ■ i * V' H' a uT 1. ;< A • A#r(r n»/‘ i .r.,' 1 1 fr#“'*^ h< ■•f - <■ ■■■ x'l - 3 ts- J «■ ff 41 any of the coal syb stance from the resicJue along vdth the selenium,, hut the following test was made to varj. fy the assumption. A small amount of residue from a preliminary extraction with selenium oxydiloride ivas analyzed fo r moi stu re, aSh, vol atil e m.atter, fixed carbon, total carbon and sulfbr content. No coke buttons v.^ere left from tlie volatile determination, and the presence of selenium and chlorine in the volatile matter was shown by the brilliant sparks and the copper ^vire test for halogens. Five grams of the same sample of residue were then extracted in a so3hlet apparatus for twelve hours with carbon disulphide to remove the selenium present. At the end of twelve hours the residue was washed free of carbon disLilphid.e \vith ether and dried at 10 5 ’u. Addition of ether to the CS2 and cooling Ihrew the seleaiium out of solution. Uiis was filter- ed off and weighed. From the five grams of residue 0.1198 grm.of red selenium, or 2. 39^, were obtained. This test pointed out the practi- ccbility of extracting all the residue with carbon disulphide to obtain a purer residue for exanination. The an all selenium free Sample of residue v^ras analyzed for moi sture, asai, vol atil e matter and fixed, carbon. The results of these analyses tog ether’ with that of the original coal and the differences are shown in Table V, This table shows, on the dry and pure coal bases, that the only effect of the US2 on the residue is a lowering of the volatile matter by 8.670 through extraction of selenium which appears in the analysis as volatile matter largely. The fixed carbon is correspondingly in- creased by 3,6fc^ No other changes in tfie analysis of importance, together with the fact that evaporation of the ether and carbon diailphide used failed to show any extract material, pixrved that extraction of the main residue ^vith carbon disulphide does not re- ; 'J. % \ - :fr Ti • . 1 '. . ' ’,f ' ■ it ■•'- ofH? •'*’ '1 ■ ^ ji'j ■ . ' . U' •<" ■ <* . • 'W fcT- r. . * ' • V .' ■ \C.: ‘ t ' :.i A' '• '1 '• i , v V- 7 rjrt ■ 1 1 v'* 'll .H r, 1 ‘ ‘ . » * • ' ", - ;>» ». ;. r'4\ ’ ■■' ■' , ; , • i ..^ ,A. ^ ‘ d.: jr' t ■ i’' . >, U ' V . :• >:/' h ' , V* ..<•♦'■:• To ',■’1 <»): I’V . ■* '• ••:/»»•? -kjf 1. .i V 1 * * rt ■ f . :>'..M ;f,|rn i ‘•r irtx , ’ " •* «i '1 ,• f*.. A , , ■ '■ ’’v • , . , V . *-■- • '» . .?• ■■• .Jhio ■■ » ' • :,.-.r- tC. ^■ ' * '.'• 0 * • i . 1 u;v ■ '--T 1 ' ? ■ "' ' ■ . ■ - *.'■■ ‘j ' •,)■•' «•' ■;•■■• i;; ft?; : ' ' V i. < , ■ . . r-'..'-' r 1 ■' ■ -T: ! '■-;■■ i’i , ■ 1 *.■ . >■ 7..; ^ 4 wa> i ' .' ’■ J f 'f Ir.,- ‘I ; ' f ' V . : •- 'I H " ■ ■ . <''• I V . ^ ■'}• { 1' 1 r'^i c^f '• ■/ "'• ,• '! ■•'•'/ ' 'll ' ■ . 't.Uv. ^r< '• fJii' ' ■* *% % * / :.‘'I •*- “ T suit in any loss of coal material. AccorcTingly liie main residue v/as extracted in a lar^e soshlet apparatus xvi tt! two portions of carbon disulphide for tliree hours vjhich rejnoved the largest part of the selQiium present. After washing with ether for an hour the residue was dried as before and sealed up for further examination. Table V. Air diy: Moisture Volatile A^ Fixed C* Total C. I^lflir Coal 3. 33 • CO 9.02 53. 39 7 2. 17 1.02 Residu e 1.67 50.78 9.25 38. 30 51.04 2.78 Resi(iie“Se 4.42 46. 21 9 . 06 40. 21 - - Difference 1-2 -1.66 -(-16. 52 + 0.23 -15.09 -21. 13 1.76 Differoice 1-3 1.09 + 11.95 + 0.04 ' 13. 18 mm Dry basis: Coal - 35. 42 9. 33 55. 25 74.6 2 1.05 Residue - 51. 50 9. 39 39.02 51.86 2.8 2 Residu e-Se - 48.06 9.42 42. 52 - - Difference 1-2 - + 16. 17 + 0.06 - 16 . 23 - 22.76 4 1.77 Difference 1-3 - + 12.64 4-0.09 -12.73 - - Pure coal basis: Coal - 39.08 - 60.9 2 80.40 - Residue - 57.00 - 43.00 57. 30 3. 12 Residu e-Se - 53.40 - 46.60 - - Difference 1-2 - t- 17.9 2 - -17.9 2 -23. ID - Difference 1-3 mm -f- 14. 32 * - 14. 32 mm 43 Ihe main residue was analyzed on iiie Air dry l^asis. The com- plete analysis is s^ioAvn in Table VI, Table VI. Analysis for; Air Dry; Dry; Corabu stibl Moisture 2. 17 - - Volatile matter 6 3.90 6 5. 31 70.90 Ash 7.76 7.94 mm Bixed carbon 26. 17 26.7 5 29. 10 Total Carbon 54.89 56.09 60.93 Sulfhr 0.81 0.84 0.91 Ni trogen 1. 50 1. 54 1.67 Ojygen 27.73 28.34 30.79 Hy drog en 5. 14 5. 25 5.70 B.T.U. 9,06 2. 2 9, 261. 36 Uni t co al 10, 136.4 4* The Effect of Selenium o:^ciilo ride Upon the Primary Volatile Products of the (Jarboni zation of Coal. 1*. Outline of Investigation; WiQi coal is heated in the absence of air, such as ckiring a cok- ing run , ch emi cal reactions occur of a compl ex nature and new sub- stances are formed in place of the oiTiginal coal. These new aib st- ances are; the solid residue or coke; liquid products in the form of water and tar; and gas. Hie yield and character of tliese products will dep Old upon the type of coal used, the toiiperature of the car- bon! zation, th e time and the pressure. Ihe first or primary volatile products coming off are ciiangecl by secondary decomposition if allow- ed to ronain exposed to a sufficiently high temperature, so in the 44 . average high t€»iperature caiiDoni zation mn it is douhtfUl \ihetfier any of the final products are liberated from the coal as su(h. Much information as to th e constitution and coking properties of cx>al be derived from a study of the primary proclucts of carboni zation. For this end it is necessary that the products be r*fiipiiMf ivJiw #rv:-,-.. U > . t' ’ ‘ ’''' - • ' ..i-j'. ■^T. ' '• ‘ : I '- ' ' ' V, .■ ■ ■•'■ . '■;m, ■ . ;^;,^X)V.i-AXW«r '>H£r«^ .CJ ;p» W' ;• , );■ •:»■ A •••:•.■ ■ ,v 5 /’v ’-■'f .' /Pw ' ■' "j-v'cv' ■■ r ■ A<}^ ■' > ' , " ‘ '^• 1 ‘ ' '' ''^: . ’■•iir'-*^''''^^'^ , '■*%iir'^ »,; > V .' V'it:^ •• ..Jrjss-i:' m irli^ • ta<»id ^fl:#iS-;.,Sffe. !•/ iliiLi»'v:: /i ' M Hit • <) >oV /T 'Ot: rf 'li/ 1 ' . ■ . ' .' . •;> ’.. Jff r?5 0!> tc^ '>' '^A>t 'W. c -ru.ari^V ••■..*. ’I*-.’ ,.: . *'>, i^;ra,^T^v".os,.WV’7;' a ^ j •. !' jc?t. '.ri .•■4' ,-..f 1 !»| iir<)? 4^a/ 'Mi lilitf .% Huf! rh.T/ fj ;„',jfi4 ti^ii‘< '. I* •'/ ■ . . ‘ .. ■.' ■' ^ ' ’' ' y ■ '■ «r9 4V ■'’51 ,£#» ‘ i . ‘ • / . 1.“A (. If . i tp JVv J, >» ./^ h •>’> . st^Oi '*4yy? 1 ^ . ,, , : & ■ •. ■ ,i.- .. V . v^>\ iM. .»^.ft'l.- 4tjh>< 0b‘rinicaiciiQ\.J^^^^ ^ I’,, . ■> '^ !t ■ ■■„• V- ■ ^ ... ■'> ■-;,: 1^-" f'/J' ir.^ctfc'^ P \yi ■ v'E^' yfiy'^v ‘J). .‘i't Cjf>9j^-o . .,■ 1 r ' ,; ..•;. ' ,j(iS ; • .. •• ; ' ^ .' :r>'.’ .’ ' - , , - %.{' *< f ._ , * l-j'-.'lfo''* ■•-»“ ys} ^:) 'tij 5 '/I'HCV:' l&'-sl;'. ", 'll/.' .%■ '•'. '. ‘ B T .T «* ■>*] If'ys r.-ir> . ■Kt »■> '• «f r ' « r .fcVa ixi^ililif'? r’ ,/; ‘ : . .. .v;.^ i f^'^^ L 'yyzM.'uyZ' ■ ■.. ■" ■ '' !i •’ ’i 't‘A''*' •“ ' ‘ ■ ' \ '', -‘ ’ ^y!' -f' '1* -yJ'. -..'1 :a (>rh ■ i.oitiirxv’ -'ii' jASiiLrn\L '‘i ' . . ' ^...jhiliiibif- V*# '/J .'jy»i L'>iyya. " iy; ‘ -^ 45 , the tar begins to distill and that given off after the tar is all distilled over was noted especially. For this reason the temperatur of the different carbonization runs were governed by the tonperatur es at vihich the tar first showed and that at Thich it ceased to distil. Through the courtesy of Wade S.Hawthome the temperatures of theb^inning and maximum softening points of the fresh coal were obtained. Ihe Franklin county coal used Ihroughout this inves- tigation begins to soften at 37 4^0. and readies its maximum soft- ened state at 406 From this data it is safe to assume that the tar b^ins to distill from the coal around 37 5 *0. 2*. j^paratus: The complete carbonization apparatus is ^owi in Figure II. The electric resistance fhmace (A) was the same as described in Figure I. Exact temperature r^ulation was secured by means of the resistance (B ). The coking retort (u) was made of pyrex glass tubing 50 mm. in diameter and 30 cm. long. A lOmm. tube or side arm (O) was sealed on about 10 cm. from the top of the retort. The retort was held in the furnace in a vertical position with Ihe upper 10 to 15 centimaters exposed. Asbestos paper was wrapped around the exposed portion to aid in distilling over the tar. The top of the retort was closed with a #8 one hole rubber stopper carrying a pyrex til ermo coup 1 e tube (E). The stopper was protected from heat by three aluminum discs (F) placed on the thermocouple tube. The retort act- ed as a distilling flask with the side arm long enough to act as a condenser. The end of the side arm extended well into a 50 cc. dis- tilling flask (G ) in which tlie water and tar w^ere collected. The evolved gas passed out through the side arm of the small distilling flask into a aaall bulb (H ) filled with glass wool, used to remove ^i rv. - .''*. * . ■»■>.■• —V .;iAr^jL..^t^ i .• J.tw^ fx '■'i '• ‘ fv ( • ft! rr 'l . I t <-'f •<•. fl '■f itt ,.a f : .>(;/ ;.*'• «|i; vv. ... > jt/r, , r ( •. \ -,;; ■; '-1 iU ;)'t " ’ u M ;.’ iw \ n* '{ rtc< / 4 j^ * 1 . - -V €i_ - v-^rc ,* “ 1 r>; ,s -1 ,^r) I.. 't;’« ■S' .; •'* r . j-n . i?'.' f.»'iir rc; ('V ■' • ( , , ir'i 'I /■)!> (• (■ f* ;t. < V? i’"’ M. r : 1 . ' 7 « r/.‘f r ;■■’.»?:;> es/i» tfifUd mrt-^ rt' . r -. *.f ' . # i V • f r^ 46 the tar fog. Ihe U tubes (I and J ) held dilute sulphuric acid for tlie removal of ammonia from Ihe gas. During runs on the residue from the selenium o^g?^ chloride extraction tJjese tubes also removed traces of selQiium from the gas. One of these U tubes was filled with lead acetate, vti desired, to remove hydrogoi sulphide and hydrogen selo» ide. From this purification train the gas passed directly into a 12 liter aspirator bottle (L ) \ihich served as a gas holder. This bottle was graduated into 25 cc. divisions and the gas volume was measured here as well. 3'. Temperature Oontrol and Measurement: The fUmace and resistance block have been previously mentioned Some runs were made ivithout resistance thus allowing the coking pro- cess to proceed as rapidly as possible. Others were conducted more slowly. The time and temperatures used are given in Ihe tables of data. In general the tonperature was not carried past 8 25*0 so as to guard against any secondary decomposition of the pro du cts. Tempera- ture readings of the flimace and retort were talcen approximately every fifteen minutes. For tonperature readings two Ch rom el -Alum el th enno coupl es (E ) made of number 16 wire were used. These were connected as shorn in Figure II. to a Weston Direct Current Milli- voltmeter (M ). They were standardized against the freezing points of Bureau of Standards Aluminum and Tin. The couples were exact dupli- cates and the same tonperature curve was used for both. The tonpera- ture could be read from the curve to within two d^rees and the couples were accurate to the same extent. 4*. Proceedure: Thirty grams of fresh coal, or residue under observation, air E (» iT'mm ■ ' :■ r) '• ' r* 1 . 'fT.r - n ‘ 5rr,!.'^ b J . ^ u ;• ,- ■■ ;■> ' ■ . ' '■.■:> , i.i' 'T »! '»niS '*'■ ■ " *..••■ *v ;• : a^':. • .V 1-7 . . ■ ■ 1 ■■?■(■/ P-t- A| V- ■■ A«a0.«^*2 -.5 ( n-'- ' ; , . -n ' ;■<) f.' >'- ‘ < j '.' ' -M S>'> V. '■' i.i ,’".‘'i'S| - '-.oau - T . 1 ) ((;.;v ; ,■ ., . ■■ "'VWTK- ' ,*l • *1 . -■ ■ . ^•^^ Iwr - o-tj ;.' : ^ ST. ft .'^, I M 4 . . t,. •;, f < v ' -I -;i ; ; : ■' ••' ■ { t \ ' ■ •■‘•r »W ^ !‘Ji td /' * ^ ■ • L !.::'' V f _* : ' ,_ . '-.' v , ''£/>•' ^ , u.T i-i •*.■/!, /> iJ .•; ■. • . ' ll ' ; I'l ‘ r . rs *,^ tr ) • I . u.' ' vi 'fjrr f ?' •-) r.i riflUf!* :*;0 tlrfi./ i '9 0^ .1 ’■ j - r / leJRtj .'‘/r'/rv, t.»-i r- 'rl a,.; q,^ ■„ •: 5 f-'WL ii ^ f Pu , rT . ir.rr nrr .!' n , > / r . •; I ' ■ '<■' *>' r - ' ’‘'hf ,1 ir; -j , ;.f* ;. IT •. >* t 0 ' / ) j'iiH a jt ^ V «4 ; * t ^ t 5 I 'lc 5 ; 1 'BiC ^ • *■’ ■ r « V /"« 4 / tf " •' • .^, . .1 '♦ TSvl ' •t 'i >ftii lov ■■ « - r ; yr - r/r 1 \y ^< il ; • • M ■ ' - t JI 5 ,; r •! - , ;. ■ .’) '• icv'i • •) ’> :•! re TijJ " i..'!/sj ! 'T’:- " noo , r* ; i*f iJjfff *') I Ur iii • J 0 3 » : »' i ' j ’ uiv « < fcg »> wg | gwi|gwg • ■ ■’ f '■ «** .A .i . J » t - <- / ■ L A - Electric Resistance Etimace B - Resistance Board C - Pyrex Glass Retort D “ Side-ann Oondaiser E - Pyrex Glass Ih emo coupl e Hibe F - Aluminum Uiscs G - Tar Trap H - Glass Wool I - U Thbe ( 10 ^H 2 S 04 ) J - U TUbe ( Pb acetate) K - Thenno couples L - A^irator Bo ttl e l?l# M - Mllli voltmeter Eigure II* 48 dried and ground to 60 mesh were poured into the retort. The ther- mocouple tube and stopper were inserted so that the lower end of the thermocouple tube extended down th ree-fou rth s of the way into the coal ciiai^e. All connections were made as shown, and the apparatus exausted of air. The aspirator bottles were used to exaust the air. During the carbonization mn a suction was maintained upon the re- tort by holding the leveling aspirator bottle ftill length below the gas holder. No fhrther attonpt to control the pressure in the re- tort was made. The suction was sufficient to take off all volatile products as rapidly as formed and thus no secondary heating of pro- ducts resulted-. 5'. Determination of Products and AnalyMs: As has been stated in the outline the most important products from these carbonization tests were the residues, the tar aid ihe gas The amount of residue was determined by weighing. After weighing the coke,or semi-coke as the case might be, was sealed in a small sample bottle. Ihe water and tar collected together in the tar trap were ranoved and centra fUged. Very few attonpts were made to determine the weight of the water. The water was separated from the tar by use of a small pipette said the tar weighed. No attonpt was made to det- ermine the ammonia. Ihe gas evolved vfas collected, and measured in the gas holder. Portions representative of the whole were removed from the holder and analyzed in a Modified Orsat Apparatus (29), designed and built in this l^oratory. The constituents were deter- mined as follows: OO 2 hy absoiTition in KOH, O 2 in alkaline pyrogallo ^314 in bromine water, and C^Hg in filming sulfliric add. H 2 and CX) were burned in a copper oxide flimace at 300 *C., the H 2 determined by contraction and tiie U) by absorption of the resulting 00 2 •' ? ^ ,'r r ■ ij' \ ' ■■ .•! .'■ r. i ' \i rritUr, i .' i ,' * ■'■ > ' {/ ' ’ .» V’ . • r V • iiik. 'O'f.'' Q'. . .r,Vr,~r/j - > ; .5 \ «',y. at f')0W 0‘C£*y .I* ail ■' /.■'’'j''r.tra; :£) nt )k? !•' ; •M'1 - iw i. M’ur^e 'w v >t -,r* i li:u4f i:!> Jr ' I .. ■ A li' V .-. f'l’S IX !} a I.. y «it *i/\!;o9*r i ;'"f. f y i . »>.' \.--* ril'^'. '-/■ '' r<; At,' oJ ; . ■ t T ■ ■ . ^-iwi *{r;v’ i • 't -n '>f :: '(>•> '■'’>■!■* ' IXi* uo oX ■'tI ii-'.* »#«.» • ■«•’' '«» ■■> • ; ; ' m'UX , ; ? : '■».% f« «.»' ■•.'•J i‘U £*'.V.i: .' mo .t; ;-'‘- i : 'T :•;.. A'.-w -t ja*;:' ^ r.-iff ■; r • '> 'A " : ■ i J " ;;i.yy: r. h.i: 1; :• .-t-.r./ , 3 vf .i ’■; t ' • .•- zctfi* ;/ a'Xft-ft *jfi? )• r^j' f ‘‘xt *X t ■:.'‘/i; ' - • . 13 w. . ♦ n- < 'V') i)f» f; T|,r 'r/;A- -V'' ''A ''la'' * ',iS ..V ' . '■ i?. tin:* r'-;‘ 1': ■ ’f ti?; =.r t^cr; •! >.*'> ' < .c I; v foV’ Cii ■ » '>i|r -s^t*^.;V}r 1. ;';<»■ f\ *1 n ' n.4M , h'' V (1 ,'1'i •■•r ■■■ ' .' ' i • TrO -or C-'.- '•••'•■■ iJ ■ :| "orf M ‘-3 it sn ir / .) ■ vp. tit.'-ii'J ■ r’i, TjfE4 ‘fAA ' • / fTC ■■'• t ij.t'ir'rrt Ui* 'roni . -d^ fc*; c.^St I ‘ ■*■ Id 1 , '/r ■ r : r:'. v , '-r; ' ‘ r'T l • P- I: i ■t-V ■' ■ 1 • >y. '- I S!4 . , •■^l A.. -■ 'J *’ i tSia 'U '..I, 49 in KOH# CH4 and C^Iq were burned in an atmosphere of oxygen over mercury in a glass bulb and tlie two gases calailatedby Eamsiiaw’s (30) method* N2 was taken as the difference between the total of the above moitioned constituaits and the original volume ^ihich was always taken as exactly 100 cc. thus giving the values obtained in percQitage by volume* No attonpt was made to analyze tjie tar* In tests during ^■hich hyd.rogen sulphide was not removed from the gas with lead acetate it was determined as (JO2 in the gas analysis. The percentages of H2S are so anall however as not to materially effect the C0 2« Ye* Results, Part II* 1* The Effect of the Tar and Volatile Cont^t of Coal Upon the - Resulting Action of Selenium oxychloride: The loss of resinic extractable matter, that is tar and those hydrocarbons forming the volatile constituent of coal, results in lessening very greatly the action of selenium o^ chloride upon coal. TJie reagent attaches coal in proportion to the amount of volatile matter present in the coal. Coke nhich has lost its volatile matter and tar is not effected. Partially coked coal retaining a portion ol its tar is attacked giving a gummy extract difficult to filter, nhile with freiSi coal the selenium oigr chloride attacks the resinic material and un saturated hyd.ro carbons* Tar is not thought of as ex- isting in coal as tar. The reagent removes those hydro carbon s, mo re especially the un saturated hydrocarbons, and those resins ^hich coked or distilled yield the tar* The resinous extract taken out by the reagoit is not a pure resinic material, but a new compound form- ed by chemical action of the reagent upon tlie resinic material. This results in the gummy sub stance, vhi ch being mixed, with selenium from '< . s S ' i I . 4 ' ■lf> :: fit ''1 a .,q u>/v »<'•■ '■ ' >"< -t U . .:. •A » ] j li'^*'** y i ‘f ’■ Jtb f* ' .. V rr.;Tr.>- >. .> ■; iM .)$iu 'TA'l^y •K' ■ :»:. T-i';.,' fit ;> d/ tit}f ■' V' •" '.. li ( b I j W T Ci j •-; ijf ? I- ? i. d ) *>h\ y h' t v„. ■ .,; t* T'\4oX}i^ h fi'f !" , 1 , :■.'■• I# . ri;, ; ?y f r** : » ’ \i!’V ' ; • » / ■ \ > r O'l/i t. > p*-- if ■' I t?' 4 ?r/O.Db’p*; . •■;....i‘j t0«i5 *H» /’•U l*-ai »rr.i « - i •; ' • , ' ‘ M I l'-rp'y(\ r.-t'.-., •• ., st w:), ;;.r r.’ ft 4- " .;:';r:;5 at> iBWl .' -I.vi v'sstjr. V ' - rii’flOBHo r O T‘ ':’rr:.7 'fo J ■'■I’.r). .; £>' r f.r ,'i Ko't' -^5 J /:{,; • r iV.' O ‘ '■' » tcv. 48 ' • ■•'•*' li y\ r»t '» .1 n Jiv ‘ “ ' ■ : " ' ' i ^ ?* r.'k -5, ,i' ! ^ ^ : .' ;. i>iiy . JVi >'/«; •??.<. fJr ■ J f *jirJ {» ^ ->'r,V: ;.. < f» ';■ 'i V ■ .' ^ •' -f ''^ •v C f'' •.••!' ' I ' ■* 7 ■' ■• ^ ' ■ ! ')• ' ' vfj? >'»' ’ ■» i ■• r T ■•iO rl5i>'i^' r'-Jr; •, . ■ . '^ ' ' ' V r! '>3’J n'l «d'A’:’ /u; & ■ ••; J'v-'' ;i r f':.- ■ V • . I ! .■ ■) C" -rtL*'- .•;j fr-^ >;u ;)fif 1 K. k f r U*t% i ’iV J i •»f1 •-’T . • j » ' / - ,nj . i. j .f i Jil V- ’< 1 !^fp«7e»r!=sj! K/3 . ft 50 the decomposed reageji t,mal£es filtering impossible. Water will not effect this extract, but alcohol will aid in the filtering of it. The loss of the coking property of the coal at the same time as the loss of the tar material leads to the conclusion that the bonding material of the coal gives the tar or else,if the two are separate and distinct, th^ both are attacked by the reagent. A sol- ven;t to destroy tlie coking properties of a coal does not necessar- ily remove the tar also. Many common solvents being used in research at the presQit time Avill destroy the coking property of coal, but vihen the coal residue is coked it continues to yield tar, though pro- b^ly not in such large quantities. Seloiium oxychloidde is unique in this respect that it is able to ranove,in the small amounts of coal used, all traces of tar from the coal \hen distilled. If the extracted resins or tar material was not changed chon- ically by the reagent it should be possible to separate it by hy- drolysis of the reagent. Since adding vjater does not effect a sep- aration but only the d.ecompo si tion of some remaining selenium oxy- chloride, it is safe to assume that chonical action fonning some additive compound has taken place. Ihe residuesshould be cellulose- ic material if the data we have is correct which assumes that sel- ^ium o:^chloride \idll not attack cellulose. Just what the nature of the residue is ^vill have to be determined by analysis and by the study of the gases formed by carbonization mns on the residue. The Effect of Selenium oxychloride Upon Coal; Selaiium oxychloride extracts all the resinic cinstituent of coal leaving nothing which may be further extracted with the ordinal organic solvents. The action is not a solvent one but chemical in nature. Powdered coal is attacked ^vith the evolution of heat. .t -A ^ i"'-" rwa '1 - M I ai M S t ^4 ^ i| '■ . ;■ 'r fi i.}'. '.* ‘I?’ ? f !!»<■: ro^^, x'.l'i.’ '•> '»»•« !■• .‘I'l ^ r I •or* 1 > i|,r ,'lt^ 'V.«;>f ■ * » ' . M. i f-r <• ' 7 f> «£ '- • ‘l- or> ■? <;■ '’i i l»* ;>/' :j {‘O '’< y 'S I rri" : ,■' '.i-'iAo ‘t'i V..*'- : tJiSf •<> iv. i (. » 'fO ■; -M . ; • ’ ff' VO ;.v:, frtyiJ i^iX> f ir ■•.' \ • i r r ‘ c ^J /'j; . v: ■Ch. .^ V ,0 .,t ‘ fl i ^ 7 ir. rrtr ''•" ' ft f 11 rrxi '’'r;iJ-:"- ‘U ' I ,'t ^ ^ f ■-■f f.» • I'c i. .t;V'. itt 3 u* ^ •' •''' - - ■«• C lit* tf *tv , ■ ■ l.'i^x 6iii I 'f ' j ■' '■ ■••'» »...' .' f , /■'<■’» 5 r^ 'Cda'. !| „ *;»a - n ••■‘O-n-j r I'Ji?; j.-.? .•* i ■J! ; .t,p« vc'/t/' .f 't?i> *■£ ♦ •’ r?. '4 .t r r;f"‘: cJ. )!.i:'’,:»r' ' ;•« ''■ f sf ■.■•.■ •^ li ■'v^ . : ■ ', ^.cf 'ti. ty^ ’ » ' ' f f “■■KCt •''KlXi'-JOil'*'*' ”,.'*■ 'iv> -'r' ' ( '•'"•' f.*;; * •'• • •' b -/I . > ^.»cT .«!*' '.K f-:, ■'.'^X'Mf ir' .'f r:...-. i r'?> ; .-Jfr «.>f» «i • '-'V ': . •..?* '■ > fj(i, y .V • ; , <'N.‘ ;j ( Ao.n^" > oit r " ■' ' ‘fiif .y^ rii » o ' 5 .;!^ .■• • 'f ’*,' ' ' ' ■ , / '' ■« ijpl . 'JJr- Mr ■» " ' ^ ftbf ' *'V 'i a ’t :■ . •[ ^ i^' .j ■•. ■'/'ij /1 ck*'’'m: ( V' tr;i <..*1 at>< *5 0 V ,: a r V. '• ’ r;.‘ ' : ' y ' ,r*c.' Tj[> ttf ; J>; r^'■^ *.» y -J f Jjj, '.;, > 'Vli.t'.. ; ' /i !-t titoNKgen B.T.U. 4.90 13, 280 5. 25 9, 261 ^ 0.35 - 4,019 I.' f 1 b.’ ' .‘ •'■>;.:• •(/; ■ ':i,' ;i?' , ,. ‘ '■ i ' • ■ 0 ' >'J : ' ft. a.-: > |$| }r , : rf; i »> - y>) ' J.f'Ol 1 IVO JflJ , '’ ' ’■ .VX ' <» >• ■ • ': t rfi' * - ■*• J V :• i r ' (■• j .) V , , 4 ■ ■ ' - . -J *' ;:; ./)0" *! '•: ■ ' i ' '• ' 1 • .'■ i ■'©•f 4. ' K ■ ■•; in ’ ' ■>.' V *’• .’ ' ::^i.' V ..M '•■iJ.:. idro -o i 1 ' ' ' ,‘I>0 ; i:; 1 >■ .' f -0 • •« t , ’ •'• i'» ^ • V •,:■ :■ fir 1 o . '.’ • f * ■■■ ♦ * ■ ” i - ' 'T ^ ..1? .■»*, X. 1' ; T i..''.’ r.'i^v *r 0 i>l fi . ij - < •>' v-W M 0 i ■ n Xif*^'’. '['I J’vy f iL I. ■’ ■j/; ('/r •)i#, r Cfi « ;.‘ • ■ i t) s 'L':.Id'* 0I»' ' ■* •> "( ';.*' y .' I V. !;/ •.'.t-'-v ' . ’ j ijT' -ft noift'' ‘ ^ - IT 1 I !:f'tJ^ ■ ill ; r ‘ ■ '.ai ^ • >Ai ij ■ -I. Miijii r d' h ■•',■ 4 ij • ^ ( . li ft^n-?- '- iS’ii' '■ i 1/ ) fS/nA UJjLilO r. ^ . «• >5 . / >. ; I r..iv*4«» K, : < ^ U; -•' *.^‘ t'/ ,.. •■sr pr(/, ■ 'Vto.'l I* ’ ! ',' ', 4 . t L ; :s c ', ‘t.' • •'f.is ■ Jv/. 53 sul)stantial gain in any single constituent of itie coal. T^iis leads to the conclusion that tiie coal ho dies extracted were very low in oxygen and that there was a large amount of oxidation during tire extraction process, lliis oxidation may have been mechanical oxida- tion or more probably may be the result of chemical action of the reagent. Oxidation was to be expected ivith sucii a reagent as selen- ium o^ chloride. 3. The Primary Volatile Products of Coal: Four direct carbonization mns were made first on the fre^ coal. These runs were numbered 1,2,3 and 5. Data collected on tliese runs and analysis of the gas is shown in Tables V1II,IX, X, and XI. In these four coking mns no resistance was used and the tempera- ture was allowed to rise until the point set for the conclusion of the mn was readied. In general the mn was terminated iihen the tar had all beoi distilled over. Three fractional carbonization mns were then made, mns number 6,7, and 8. In these mns resistance was used and tiie coal <3iatge heated up slowly in order to get the maximum yield of products up to any certain temperature. In Rm 6 and 7 two cuts were made in the collection of gas. The first fraction or cut contained tlie gas evolved up to tlie point at vSiich the tar begins to distil. The second fraction consisted of the gas given off vhile tlie tar was distilling, aid tiie last fraction was that collected after all the tar had been removed from tlie coal. In Run 8 one more fractional cut was made to differentiate more fUlly between the gases evolved after the tar had been distilled and as tiie tetnperature readied the point of secondary decomiio si tion. Data on tliese mns aid the cor- responding gas aialysis is sho^vn in Tables XII, XIII, and }CIV. ! - f'H, , ■> )'/.•:' '.. I,r. tyi' iV '..0 •) . i ■ ',;7 ua • t^ itl ,,i!^M!' U <»• 0.<*i.; ; ' -f fcr« ' 09^ ':• ■H ■ ' 7(/';’’ ' • J»o' I.. » • r.' \i ' (t^ ' 7 0 ’ rW. ! » ;A I X«1 T,) K' a,, • '*! ’*• ■JVT ,; .'u , i .0 .M'i ' . •^■‘ OJ* '. f'V' .(.' r 4 * \ I * I. • ' ’f 'f'o’V ,' '•-, !- 'f rf»<; *m 7 ^. ' ’ 0 f M f/T;i <>i>t >c; ■ I..- \.l- t* , i.F -^11,1 . " ,;■ ^> I ■ '^• O' rv tOt7fcja4..tJC ' *' I ■ I - j *• f I «;0li t •;>o . '-Ca •. 4 ’^ > . ' ‘iliSk'ftd >if.J ; • ;/! :oo ’O*-’ . ty> ^ ijo *t 0 ' . tf.» OJ «iti ■ .' '■/■ • ) .^ ' 0 i> ^• 1 1 ^ * ' • 4 ’*>)(" ■■a'.: w •>‘^0 ■ ; i ? ♦ <" f” ' ■ • ' iJ5?j . < 58 plotted against the temperature. Eacii an all circle on the graphs indicates the percentage composition of that constituent in tlie gas fraction for tiiat temperature. The rise or fall of any certain con- stituent of the gas during any given tonperature cut or fraction may be determined by reading the curve betv/een tlie points represent- ing the two cuts in question. A glance at tJie charts will show that of the four direct carbonization rans Number 2 curves vary consid- erable from Numbers 1,3, and 5 vhich all lie practically together. Number 2 may therefore be omitted from consideration w!ien drawing conclusions. One of the three fractional carbonization curves,#3, also shows a divergmce in nearly all the charts from the other two which run practically together. Numlsrers 7 and 8 may therefore be talcQi as standard curves. A study of the Tables and Figures submitted in connection with the seven direct and fractional cxrking runs made on fre^ c»al shows the following facts: Coke- The resiclies varied from a semi-coke to well coked mass. Very different results were obtained from Taylor and Porter’s (32) woric in vhich they state that ’’slow carbonization tarids to increase the yield of coke”. Table XV, giving the percentage yields of coke and tiie time of the run for the direct carbonization runs 1,2,3 and 5 shows that tire yield of coke for these runs was inversely propor- tional to tire l^gth of the run. The best coke came from the runs of shortest length. T^le XV. Run Yield of coke Time in hours 2 69.3 6.8 1 7 2. 46 4. 3 3 7 3.5 3.0 5 76.6 2.0 The average yield of coke for tlie seven runs was 7 2.1%. V • , T* I A ■'*,-11 '*»> .- U« - ii ■ '■'■ ;" •’ t> r'l :"l .>r- J f ."nc> ‘i o, M’-. •, ^o-^t-actT V? iiJltf.. • ' :i'i ; ■* .{.*) I • } r- 0 ,• n • /, ,■' ' o i.-o » v -V J- -.•; 'V---. -'. ';o,?a>, 3 .'•• -jvj ?!£i3 . i-h* I ' i*. I .Mil ' ii ' f t/ Hj-:-' t r» OriH ;t*f 1 -- ' ^ '1^. >jj/l ,fl> j.ii’ "I , . VU'*,'" ; a'.rt X^r^.W'»« StiA'iBH'jmy ; >'y 'f-V ,'i! r*-..' f.' '} It ^ - '■» ■ •: I •• ■' , h r- 1 t V. ■ ,: ■•' .fO-UU.r: >o -. 1 :'. ;.- '■ , . rur. ifDi'Ui;*' J ' ' O ■ '■ ■ ■ IT r::l • -i . . i r,9i «r©i5i XLi£ Jr - ■ ■ i ‘tOi ■.■ ' <3 !■, J >;i urj »' .7. i n « ?a« '■ / '*B.r., oj^‘* ' . ^i; II wiis4-Wk 50 Tar- Ihe average yield of tar for the sevoi runs was 8,2fo, In general the higher the tem]3erature and tiie longer the period of carbonization t?ie larger was tlie jl'ield of tar. The first sign of tar coming over varied from 33 5 ’ -400 * U. , wi tli an average of 38 5*C. This average t^perature of the first appearance of tar is just ten d^rees higher than the softening point of the coal 375* C. The tar v/as all off from 50 5 *-6 50*0. Uie average temperature of the end point of tar distillation being 556 *0. Water- Water condensed out into the side arm of the retort at from 100 *-230 *C. Heavier condensation of water varied from 28 5* -36 E ^vith an average of 331*0. Time and Temperature- Uiring ttie four direct carbonization runs tlie temperature rose rapidly for one and one-fourth hours. At this time the temperature averaged around 390*0. From this point on the temperature rose much more gradually witli the leigtti of time. The point of falling off of temperature rise with time coincides with the start of the tar distillation. Temperature and Volume of Gas- Figure V. ^ows that the ex- ception of Hun 6, the volume of gas over at any certain temperature was approximately the same for all tlie mns. The fractional carbon- ization curves ^ow a larger amount of gas at any tonperature than the other four curves, naturally because the time was longer and resistance bMng in tlie coal was exposed to any given teirp eratu re for a longer time allowing more decomposition. However the differ- ence is not great. The curves show ttiat up to about 375* the evolu- tion of gas is gradual. At this point tiiere is a rapid evolution of gas ¥hi ch continues to about 450 * after Miidi the curves siiow a steady evolution of gas with no sudden changes. The distillation of ‘i ' f I i' I - *.■> I i 1 ' 0 ’» j •I* 4 I r! I :;/.l ll f •1 V. » ;i , ^ ^ y k \ i \ I (! It ■'■t ,1 I ■ 'i”‘ ) GO water of decomposition and tar is coincident with the rapid diange in the rate of gas given off. Table XVI, shows the relation of t^p- erature and time to the volume of gas. Table XVI, Rm Time in hours Maximum Temp. Volume of gas. 5 2.0 530 2450 cc 3 3.0 565 30 50 1 CO • 615 4500 2 6.8 660 oroo Oart)on Dioxide- Maly sis of gas fractions shows that CO 2 is present below 350*0, It increases after tlie tar starts distilling and up to about 525* ^vhere the high point is reached. From this point on and in that portion of gas collected after the tar is all off the percentage of carbon dioxide rapidly decreases. Less than 2fo remains in gas at 6 25*0. The data tallies for mns 1,2,3 and 5 show that tlie total amount of carbon dioxide yielded increased in the order of mns 5-3-1 and thQi fell off rapidly in mn 2, Table XVr, shows that 5-3-1 is the order of mns in \^ich time, terapera- ture and gas volume increased. Run 2 ^idi took tlie longest time and reached the highest maximum temperature Mows the snallest amoun of total CO 2 » may conclude that LO 2 increases ^vitti tQnperature up to about 615* as shorn in tlie graph, and then rapidjy falls off as tlie tanperature increases and secondary decomposition begins. O^grgQa- Maly sis of fractional cuts of gas shows high oxygen content below 350*. As the tar distills tlie percentage of oxygen rapidly falls off. After the tar is off the o:^gen remains practi- cally the same until 600* is readied after itiich it falls. The totals Mow til at oiygen content deoreased in tlie order of mns 5-3-1-2 as ) f- f 1 . C • 61 , the taiiperatiire, time and gas volume increased. Oxygen is present in highest quantities in tlie f^rst gas given off* by carbonization of coal. Mud! of this is held mechani cally aiid is due to weatiiering. Unsaturated Hydro carbon s- are very low in gas given off by coal before the tar starts to distill. From 37 5’ tlie percentage rapidly increases until a temperature of apprxjximately 525’ is reach ed. From tlien on until 6 25’ the percentage f all s,practi cally no un satu rated hydrx) carbons cxrming off above 6 25 ’C. Baizene or Saturated Aromatics- lire curves for benzene are nearly identical with tliose for tire unsatui^ated paraffins. There is quite a little variation in the benzene curves due to the fact that sudi anall amounts are present. Benzene is formed from the unsat- urated paraffins upon the distillation of the c»al and therefore its content in gas depends upon the amount of etJiylene present. Carbon Monoxide- Only a small amount of 00 is found in gas below 37 5 ’C.Incompl ete oxidation of cellulose upon decomposition accounts for its presence at this stage. From ttie point of distilla- tion of tar up until tlie UO 3 begins to der;' '^: ' ■ cvi 1 .%’ ' ' i _ <* I * - ' ^ * vj 1 1 f .. ?• .n • .' * •r ' f ! A'a* ' >:■ •lyy •< • / iuft r. • ■ c- I , A- , 'V II,' n ' lO.':- Ih' Jf- =T '•-.. r iC '. o T ( / I i i » ' • 1 : •" .v ".v-.l : 20 ii. •c;jn ' 1 I 1 I . r / // / «-/ DeGne6 Cev^iGroAe. I m i Z)ec!|re€6 . f - : * . i i h- I 1 1 i t V*-l . ■ ■ ■ i- t f i I t; i ■ ) - - i * •% V- "”x . . m •ttf t 7 3 To summarize tiiis data it may be stated tiiat: 1- More til an two -thirds of tlie oi^anic substance of coal is decom- posed at temperatures below 500 'C. The ronainder in mostly decompos- ed betweai 500 ’-600 2- Coke may be obtained from fresii Illinois coal of the type used in this investigation. Fast coking of the charge tends to increase the yield of coke. Heating slowly over a long length of time tends to destroy the coking property. 3- Low temperature carbonization yields a higher percentage of tar than the high tonperature process. An average yield of 8.2^ is obtained by the method here used. The tar b^ins to distil at 385’ and is practically all off by 560 ’C. 4- Ordinary moisture of coal is driven off at 10 5 ’O. Water of dec- omposition of coal is produced from coal in largest amounts above 250’. This water of decomposition is mostly off by 500’, tire larger distillation coming amund 330’. ihe largest part of tire water of coal thoi, consisting of moisture and water of decomposition, is ou t by 500 ’• 5- Rise of temperature in coking retort proceeds more rapidly during the first two hours of carbonization than later on. 6- Coal brealcs down at all tonperatures more or less. At tempera- ture of normal atmospheric temperature tfie process is one of oxida- tion or weatliering. The temperature at Miich reactions occur to any extent in experim^tal time must be considered above atmospherd.c tonperature. Bituminous coals of ttie type used begin to decompose and distil between 150 ’-200 ’C. 7- The first decomposition o ecu ring as the temperature is raised is the breaking do\wi of ttie cellulosic substance of the coal. The first .... . . ! t 'I . in,-'' 1. '; ■' ' r '■ *' • » ' < Y. i ^ ■ ;V. ' ; 7 ‘ 0 J :'' ' ;> *. '^..y.}• . .*• 4 ' ,:v . -K A >( //, /’''.■ _ , •> ' • ; ■'■'■' ,'y, jj ,i v.fi . . ..; ■■ (;: 'JlP f-' “■ ; ' ’ • ) ■ i • • *> ' . 4 i t t , ' ' ' ■> '■ \ ■ ' ;v( - ,i ' ' ‘ f'' -'1 ■ '■ - '.; " 1 ' ••'.■ •'. ,v ^ . 'i .' i . ' •■■,!:. ^ - .p * ' -.r ■- -y ■{ ■ 1 « ■ . A, "i 'V i ' .r ' ' 4 -i t .., / '. • ■ •' '. o- n'7.nUif'! .. ■ : * : ■ /■ - ■ ’■ ■ ' .v' ♦ ■ :' > ; ' J-':- t ■ ■* ' ' . '• i ’: ' ■ ' , ** t ■ ■ .t ■ ‘ ' :S *1 .V , • *•' ' ; ' !, ■..* , ' /( • •f’-J- . >• ' 1 ' ' , ',y ■ : ;vj(* • ' ■ . '> r {VI - . ' > ' v' ’ « <*'>- ’■ , «■ » ‘ i! - • ■1^ ■ • • N • 4 • ^ . i ■ %. ' *'., • s ■ ’Xt *' • 5;: T ' -1 . ■ * 1 " * ■'■ •’/•i' ' ) .a .1 ••*11 U\'Ji ■ ■ I m Jr I; •vj^^cT ' (/ .• ‘M \ ^ ^ ■ •. W'' J» . ’■m t:v » • ■> It- i ;• r ;• ; ; i I \ and water of decomposition no^v predominate. The coal is still producing gases \iiich kill tJie coking property by preventing tlie bonding togetlier of the coal material. 11- Between 525’-G00* the paraffins mn out and OO 2 b^ins to dec- ompose increasing the yield of 00 and o:^gOi. The yield of un sat- urated hydrocarbons has readied a maximum and the hydrogen is in- creasing. 12- By 500* two tliirds of the organic substance of the coal is de- composed. The last water of decomposition is coming over and tjie distillation of tar is falling off. 13- Between 625’-650* formation of benzene is stopped. Tiie tar is practically all off and mudi more hydrogen is in evidence in the gas . 14- These tests were not carried to a higher temperature. Above 650' there is a final distillation of volatile matter from tlie heavier hydro carrions and resini c materi al of the coal. Secondary decomposi- tion sets in around 700 *C. \m((ra>b*^‘ '^1 r^ " • ;*'» f=-.. ,- ; » . l!^ > rf .., . , . .* A .114 ■’." •'■ !-V. •*',. I } ,!i. ,^i^, i, h tiii ' V u: r' :. ;t / • ’ ' *'u^ ' |,f V' ' 1 •.' ■. ••“ .*'j* > - S' ■ ’ ' '• ■•' *' '■¥ '7, ' * ;. V »4 ' • ./^.4 r' t " ' ' ' ■' •'■• '*.i - a; " 1, '^' • » : r* ^ »T 1 {■ i' . 7 ' 1 « '*• '■'■ ■• MT •^uivviifor ' ''5ll f^ytX ' y - 1 ■ ' ■■' " ' ■' • ‘ • v;'M»X ;Vi ' Uvjp^ \ . 1? ■> ^ f , • • - i/ ', 1 i-' : I ■if .•,' u. - M 'lifi'y V,!'!;* i’»'; .■i-f'Mti A* ,j ■ ir-: 1 './j <•• X • fi' J iJi3‘' •■'»J-' .,^ II I : ^'W X' n# *> -J,./ uj..' . , '^ , UljA ' ;, ,•/ ^ ; n' ',“0 14) ',M X h U X jiilfc’iM iLi '' ^ *■ 'i«: 1. y Ml *■- -1 I .’ ». * ,#',|v' T'/* t-ji . « ■ • T5,, ,’ o‘ ' i a , ^ :.A'r J' ' ' ■ 1 '' •' . '( ,> , I ' J 'tv '4‘ ^ - ■" I .(v;, ' •'•■■,,■ I I »■ 1 A:.. •I — * ‘ - •? ' * 7 5, Table XVII. DATA ON fractional CARBONIZATION lUN #9 ON RESIDUE. Weight charge 30.00 gm. Residue from selenium oxjr chloride ext. residue 20.44 ” =68.1^ No tar and only a few cc* of water came off. Dilute sulfuric add and lead acetate bo ttl es in train. Type of residue: Fine powder in appearance like original chaise. Combined data on mn: Time Temperature Gas Volume 0 * 00 " Room 0 1 00 210 700 1 20 265 1300 1 45 290 1700 2 00 315 1900 2 15 325 2100 2 35 340 2300 2 45 345 2400 3 05 360 2550 3 30 363 27 50 3 45 375 First cut “ 2800 0 00 375 0 1 00 450 10 50 2 25 600 Second cut - 3100 0 00 600 0 10 60 5 300 40 60 5 6 50 55 620 Last cut - 900 Data: F’raction s or cuts Ron ark: s Much moisture vapor. Trace of Se shows I II III To tal Temperature 375* 600 * 020 * 620 * Time in hours 3.75 2. 5 1.0 7.3 Total gas 2800 3100 900 6800 Gas analysis: io I0 I0 f Ni tro gen free basis - UO 2 13. 5 20 . 2 4.8 15. 4 35.7 34.0 7.3 29.8 15. 5 8.5 6.7 11 . 1 40.9 14. 3 9.9 21.4 ^#4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 h| 2.7 11. 3 28.0 10.0 7.1 19.0 41. 3 19.3 00 6.1 7.7 7.8 7.2 16.3 12.9 11. 5 13.9 CH 4 Trace 9.3 19.4 6.8 0.0 15.6 28.6 13. 1 ^2^6 Trace 2. 5 1.0 1.3 0.0 4. 2 1.4 2. 5 «2 6 2 . 2 40. 5 32. 3 48.2 c.o 0.0 0.0 0.0 . ^ 1 .I*' . '•■t -I (? ir JT '; f<(i n ii/. illrJ 76 , TjflDie XVlII. DATA ON FRACHONAL CAFBONI ZATECN FUN # 10 ON RESIDUE. Weight of charge 15.00 gnn. resictie 10.34 =69^ No tar* lype of resi(3fcie: Powder like original chaise. Combined data on run: 'W.me Temp erature Gas Volume Remarks 0 *00" Room 0 15 120 100 30 160 17 5 45 19 5 220 1 00 225 300 Much more First cut - 0 00 225 0 45 285 400 1 15 370 900 Second cut - 0 00 370 0 30 470 500 1 00 48 5 550 1 15 520 600 Third cut - 0 00 520 0 30 570 350 35 580 425 Last ait - Gas Data: Fractions or Ol ts I II III IV To tal Temperature 225’ 370' 520 ’ 560 ’ 580 ' Time in hours 1.0 1.25 1. 25 . 55 4.0 To tal g as 300 900 600 425 2225 Gas analysis: ic $ 1o 1o iX) p 1.0 8.0 14.0 8.4 8.8 02 20.0 16.8 10. 5 9.6 14. 1 C^4 0.0 0.0 0.0 0.0 0.0 CeHe 0.0 0.0 0.0 0.0 0.0 H2 0.5 1.7 9.5 10.0 5. 2 uO 0.5 2. 4 17.0 11.0 7.9 tfl4 0.0 0.0 0.0 6.8 1.3 0^6 0.0 0.0 2.0 5.8 1.6 N2 78.0 71.1 47.0 47.8 61. 1 I *::3 ( ji'T ('■u» 4 r • «U •J ? • 'I .i 77 Taible XrX, ijATA ON FRACnONiiL CARBONIZATION RJN # 11 ON RESIOJE. Weight diarge 15.00 gm. residiT e 10 . 42 " =69.5% No tar. lype of residue: Powder. Combined data on run : Time Tonp eratu re Gas Volume 0 ^ 00 ” Room 0 30 IS 5 130 1 00 280 600 2 00 485 1450 First cAt - 0 00 48 5 0 45 535 350 1 15 565 500 Second cut - Gas Data: Fractions I II To tal Temperature 48 5’ 565’ 565’ Time in hours 2.0 1. 25 3. 25 Total gas 1450 500 19 50 Gas analysis: % % % U )2 17.0 8.3 14.9 0 2 13.6 8.4 12. 3 H 2 6.5 19.4 9.9 U) 10.0 30.0 16.7 tH 4 0.0 10 . 5 2.7 0.0 1.8 0.4 N 2 52.9 15.6 43.1 Nitrogen free basis OO 2 36 • 1 9.9 36. 2 O 2 28.8 9.9 21.6 ^2 13.8 23. 1 17.4 xJO 21. 3 42.6 29.4 CH 4 0.0 12.4 4.7 0.0 2 . 1 .7 N 2 0.0 0.0 0.0 i- V ■ 1,1 V ( ■»i‘>»'lipi. -.. o” ■ , ; , , ' — < < ,'r. f 78 Table XX. DATA ON FRACTIONAL CARBONIZATION lUN # 12 ON RESIDUE. Weight of diarge 25.30 gnn. residue 16.25 " = 6 5% No tar. l^pe of residue: Powder. Combined data on run? Time Temperature Gas Volume 0 ^00” Room 0 40 280 700 Urst cut - 0 00 280 0 15 3?0 600 Second ait - 0 00 3?0 0 50 500 900 Third cut - 0 00 500 0 15 510 250 7 20 680 2450 Gas Data: Fraction s I II III IV To tal Temperature 280 ’ 370 * 500 ’ 680 ’ 680 t Time in hours .75 . 25 .90 8. 33 9. 2 Total gas 700 600 900 2450 46 50 Gas analjfcsis: % % i % Nitrogoa free U)2 3.0 8.2 15.0 6.0 12.6 18.4 21.4 7.7 ^2 15. 3 10. 3 5.9 2. 2 6 4. 3 23. 2 8.3 2.8 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 H2 2.0 10.0 9.3 40.0 8.4 22.4 13.3 51.4 CO 3. 5 16.0 35. 2 6.0 14.7 36.0 50.4 7.7 Cfi. 0.0 0.0 2.4 21. 3 0.0 0.0' 3. 3 27,3 0.0 0.0 2.4 2.4 0.0 0.0 3. 3 3.1 N2 76. 2 55. 5 29.8 22. 1 0.0 0.0 0.0 0.0 4. The Effect of Selenium oxychloride Upon the Primary Volatile Products of Coal: Four fractional carbonization mns were made upon the residue from the selenium oxychloride extraction of coal in exactly tlie same manner as upon tlie fresh coal. These runs were numbered 9,10, 11, and 12, Data and gas analysis for these runs are given in Tables XVII , XVIII , XTX, and XX. Results for these runs were plotted in purpl ink upon the same figures III to XIV inclusive i^hidh show airves e V ■‘.: nr A' iv*' i ■ 'in 1 > 79 for the cai^oni zation results on fresh coal. Corap arL sons may thus be made at a glance. Gas data for the carbonization run on the residue from extrac- tion of coal ^vith ^1 Qie-sel enium o:^chloi'lde mixture were also plotted on figures III, IV, and V in red ink. Comparative results of these runs with those on fresh coal siiow the following; Coke - Ihere was no coke residue fomied from carbonization of the residues from the seloiium o^ chloride extraction. Hie residues left in the retort were all powdery and had the same appearance as the original charge. The reag^t entirely destroyed the coking pro- perty of the coal, ei til er through extraction of the bonding material or through oxidation of the cnal mass. The average percentage of resi(iie left in these runs was 67.90 ihicii is 4.30^ less than the average cx)ke residue from^ the runs on fresh cMial. Tar - There v;as no sign of tar distilled from the resiciies. Selaiium oxychloride removes the tar forming exmstituent of the coal en td rely . Water - Not as much water distilled into the tar trap as dur- ing carbonization of tlie fre^i cxial.The temperature of heaviest distillation of w'ater v/as from 210’- 225 ’C. naiidi is about 60 ’C. lower than tlie average maximuffi distillation point for water of dec- omposition of the c»al« lime and Temperature - Figure III shows thie curve for Run 9 .i ' with a ivide variation from Rins 10,11 and 12. Run 9 was the only one of the four made on 30 grams of residue. Hie others being mad.e on less residue naturally reached a higher temperature in the same length of tirade. Qirves for mns 6,7,8, and 9 all for runs in \hich 30 grams of charge was used, siiov/s that thettemperature increased 80. with the time approximately tli e same en coking frecj:; coal and solvent residue. Time and Gas Volume - Figure IV, shores that the residue whoi coked under the same conditions gives off more gas in any certain 1 eng Hi of time than does the fresh coal. The residue, ^^i cii is cellu- le sic, is practically all gas foming material, and also there is mudi oJ^gen present in tlie residue, 'iitiich accounts for the rapid evolution of gas and the large volumes. The curve for Run 10 shov/s that 15 grams of residue will give more than twice as much gas in a certain 1 eng Hi of time than an eqiial amount of the fre^, coal. Temperature and Volume of Gas - Figure V. ^ows Hi at the runs oh residue numbered 10,11 and 12 yielded more gas up to 500’ than dL< the fresh coal. Carbon dioxide - in the gas from Hie residues slioivs a large increase over Hie amount givoi off by the coal. This may iHiow Hi at the residue was fairly highly oxidized, by a,ction of Hie reagent, and_ that part of the oiQrgen was chQnically held, or that Hie residue being mainly cellulosic in nature gives lar^e amounts of U)^ upon thermal decomposition. 0:j^gen - The residues gave off mudli more oi^gen than the freiHi coal. This would point to the fact that the residues absorbed oxy- gen during the extraction process in spite of precautions. Un saturated hyd^ro carbons and Benzoie - arc absent from the gas. The reagent entirely removes these constitu^its fi^oin Hie coal. Carbon Monoxide - is present in mudi larger amounts in Hie gas from the residues. Decomposition of cellulo si c material and OO 2 would account for this increase. J 81 Paraffin hydrocarbons - Methane and etliane occur in some\That smaller amounts in gas from tlie residues. The maximum point of tlieir cli still aticn is readied at from 50 *-100' higher temperature than from distillation of coal. It seons that these gases are two of the last to come over from carbonization of the residues up to 6 25*0. The residue ’ohidi is mainly cellulosic thus yields these saturated hydrocarbons. Hydrogen- shows larger percentages at lower temperatures than from the frei^ coal, and also increases more gradually so that at 600* slightly less hydrogen is evolved from the residue than fi*om th e CO al . Vll. Conclusions, Part II. Selenium o5ydilorid.e attadcs coal in pi'nportion to the amount of resinic volatile or tar forming content of the coal. Partially coked coal loses little, and coke nothing upon extraction. The reagent removes tliose resins and un saturated hydro carbon s rhidi form the tar upon distillation of coal. The extract or tar m^aterlal is not separated pure, but enters into a didnical union with the reagent forming new compounds. Selenium oxy^diloride reacts more completely rvith powdei’ed coal with the evolution of heat. The action is not a solvdit one, but a diemical I’eaction. All resinic and tar forming material of tlie coal is removed, and changed by the reagent. The cellulosic portion of the coal is left unharmed. Selenium oxychloride removes a larger portion of tlie resinic constituent of the coal than do ordinary organic solvents. No quantitative extraction is possible. A portion of the rea- gent is decomposed resulting in the deposition of red seloiium -*j ;> ./>, -' k’ 7( r» r:.*» , (J f^' ■ 'M .» •■ !.■ 'tj M • » . -/M » m *■■•,? 1 X;-". O' ! ' ». rtof^ ' ;' ^ ^1^ !*j ,' a' j.'rSi'f ■ ‘ lO i#i ^ o uo.«s*-u~Li i. ■ --1 ii . : 1 4ji fi,'. ' /, , : £ iTt l/'i ' ' ' ; ” ( ff' ' ij- . ' .i' ? , ;• o - ‘ . .V> A* 7 . w’nit jp 1 : j ' -j :» . ' . \ ’ •:.''' h !' f ^r.’l ;;■ - '■?: /• /il* vI 5 ; - M ■£;‘'i'-’o f/i f: o ,: .'j ^ - ' i’^'ir.yo ’ 'te-. >r r : . I- ;D £i(>i r;! ,{£i(vJiT' dafjACv" ^ ' t'*; ’ 1 .; ' ■ ''■>; . Kf r i.if» f ^ ton iit -' 't' ' a' ' •■'■ i . •- > . '* .fr i » ! J! i . ; si V ? , -r s' .'*! : “»vft locn n, . ''sn.’-'Toq '> .£' • I :! ‘ ; "J yi ^ ■ i ’ ' ,/ (; iHlrriat; ) . ■' ■ < I ; ' ' ..0 - t 1 -'o??'?’ n; ‘ D 1. i.li' ■j ^ rni h‘'r 1 '.. I S - ^ > « 'C? O ' n ^y1^r ;/n , yTJ i>i :. ^ *■ i ■ .' ‘ : J.i oyi ' t- 1 t h‘< . s ■ 'li 8 2 . throughout tl] e extract solution said residue. The residue may he purified from most of the selenium, hut isolation of the extract is impossihle. Extraction is carried out to tlie hest advantage vd.thout the use of external heat. Ordinary solvent extraction methods are not applicable to tills wort. The use of some solvent like benzene is necessary to aid in the separation of the residue. Selenium and chlorine are present in the residue. The ash of coal does not seem to he changed to any marked extent by the reagent. Nitrogen and Sulfhr do not ^ow an appreci«- able diange. It should not he concluded from this that the nitrogen and sulfLir are not affected, since the reagent in talcing out the resini c material should remove some of these t\7o constituents. Paraffin hydrocarbons of the coal ai-e prohfUaly not attacked. The residue gives off much moi'^e oxygen than the coal. This may he due to oxidation of the residue and absorption of much o:?^gen wtiich is not chemically held. The presence of 0 3^gen together with the loss of resini c mat- erial of the coal entirely removes the coking property of the coal. Carbonization of coal residues after extraction vlth selenium o^^diloride shows no sign of tar as one of the products of carbon - i zation. Seldiium. oxydilorlde divides coal into a cellulosic residue ¥hi chyyiel ds gas chiefly, and a resini c portion \hich is extracted and vhich contains the tar foming bodies. The resini c portion also (X)n tributes to the yield of gas altho in not quite such large amount^ as does the cellulosic residue. The cellulosic portion begins to decompose upon heating before the resini c portion has reached, a •v; U . '-i ■;«t '' 1 clii t .|at; '■ • ' • ‘ ?r. . 1 -• r'"* ‘' i»i'* w -.ii , ' ';L V . ' ' ■" /i laUi ' r f' f>- ■ . J '■ oi r (■ I', •>.■: vi.;‘ '4 m* V ?■•.’* ..rto J».a -'‘i I j:?l ^ :^; V'' /’! ^ I ; if-" ' •■-r il ■ V;,' J r,‘iv i'. ■■ ■'IV.'*,' ’In li' •. -• Jii: . 41 ' iR'. ‘ i '• Ilf , 1 r.' ■ *■■• 'ir-> •■■■; ^ ff- ,v; iif . , n -J oTIT ' : '■ 1- ■■ '• 'f if^ r .y-.l". ■ i,:.' '. - r ,t i \ f '•t ‘ ■ : -. :.i> J . b j.n r. 1^4 boiJ ,0 h;\ic :.: • I ' ■■ 1 0 a "10 (.'■'i' V- ,* rl r - ' r. ■-. 'T ’ •. , V , ‘.j'Tom :i ‘f't 0 r 0 ” r-,». i>.j '» '• '*■ i. iff ■r'- f'; , » *1 • t '•» l/i (i 1:1 V li. . » *( ti : '.f ( • -r.-i I •<• ,;: .1 f . ■ ■ .« • .• ,i.v } : 'i- >■- •; ICS . I JLV-:i)li!L BSBanBn "" J '■ .' i'{'> f'l'f'' A, ■■ f.fiiv: . .0 Jit,- , L':;;'- .■■ !' ZV.-" 8 3 temperature high enough to decompose, that is,before the tar and gas yielded from tlie resins begins to distil* Hie formation of tar and decomposition of the resinic material in coal results in a rapid increase in the yield of gas, and if at tliis time, the cellu- losic portion does not give off such gases as 00,002 and O2 in such quantities as to prevent bonding, the coal mass \vill b^in to coke* Ihe residual cellulosic portion of coal upon carbonization yields, up to the t^perature of secondary decomposition, as much or even more gas tiian did the original coal at the same temperature* Hiis gas vhen analyzed shows presjence of those gases rewriting from decomposition of cellulose plus the paraffin hydro caihons and large amounts of o 3 ^gen*No gas of an un saturated nature or those formed from them are present* ^T!!!* Summary* Selenium oxychloride attacks finely powdered coal \vith the evolu- tion of heat* Hie amodnt of reaction is proportional to the volatile content of the coal* U)ke loses no tiling by attade* Selenium, o^diloride reacts diemically with coal and thus is not a true solvent* It may be used however in order to obtain a residue* No quantitative extraction is possible due to decomposition of the reagent and its dionical reaction witli the tar* Neutral solvents, such as xyloie, are not applicable for use vdth sel enium o ly chlo ri de in co al extraction wo lic * Selenium oiy chlo ride divides coal into two main portions, the cellulosic residue, and the resinic substances ^hich are extracted* Hie extracted, material consists of the tar forming 4 V 8 4. and part of Uie gas yielding constituent of tJie coal ;\tiidi is resinic o r "bi tumini c in nature. Hie residue consists of tbe cellu- Icsic portion of ISie coal together vdth the ash or mineral matter. This cellulosic residue is primarily gas fo rming , yi el ding such gases as carbon monoxide, carbon dioxide, hydrogen,o3g7^gen, me thane and water of decomposition. Ihe extracted constituent reacts ivith the reagent to form new complex compounds. Hie residue when submitted to low temperature carbonization does not coke. No tar is obtained fmm carbonization of the residual material. Selenium oxychloride removes the tar entirely from the cnal, by extraction of the tar forming constituents. Selenium and chlorine are present in the residue to a small extoi t. Hydrogen of cioal is given off in largest quantities betireen 350* and 5?5’C and not at either the beginning or end of tlie low temperature period. Methane is given off by distillation of the cellulosic con- stituent of coal. 8 5, Bibliography* 1* Jones, D.T* and "Rh eel er, R* V, (19 14 ) ‘’Ihe Composition of Coal”. Tran s. Ch an* So c* Vol.lC5. London 1014. 2, Lewes, V.B* (1911) “The Caiboni zation of Coal". London, 1912. 3. de Marsllly,de C. (186 2). " De Inaction des dissolvants sur la houill^* Mn* Chim.Phy s. Ser* 3, yo1.66. pp. 167 -17 1. Pari s, 186 2. 4# Guignet, E, (1879 ) "Sur la constitution de lahouille". Compt. Rendu Acad.Sci. Vol.88, pp. 590-50 2. Paris, 1879. 5. Mu ck, F. (188 1 )• "Grundzuge und Ziele der Steinkohl enciiemie?^ . B-onn, 188 1. 6. Rein sch,P.F. (188 5) "EiLnige neuere Beobach tungen uber die chemis- (3ie ^ sammensetamg der Steinkohl^. Dingl ers Poly tedtin. Jour. Vol. 256, pp. 224-226. Stu ttgart, 188 5. 7. Siepmann,P. (1891). "B ei trage und kenntni ss der ’harzartlgoi ’ (llislichen ) Bestandtheile der Steinkohlen” . Zeits.Berg.hu t. und Salinoiwesen. Vol. 39,pp.26 -31. Berlin, 1891. 8 • Shii th , V/. ( 189 1 ). ” A con tribu ti on to ou r kno v;l ei^ e of th e solubl e and resinoid constituents of bituminous co^”. J ou r. So c. Ch on . Ind. Vol.l0,pp. 97 5-080. London, 1891. 9. Mderson,W. C. (1897 )." A contribution to the chonistiy of coal". Trans.Glasgow Phil. Soc. Vol . 29,pp.7 2-96 . G1 asgow, 1897 . 10. derson, W. C. and Rob erts,J ." Some dhemical properties of Scotch Coals".J.S. C.I. (1S98 ) 17, 1013. 11. B alter, T. (190 1 )•" lb e solVQit action of pyridine on certain coali^l J.S.C.I.^ 789; 1901. 12. Bedson,P.P. (1908 ). "Notes on .-the proximate constituents of coal". J.S.C.I. Vol.2?,pp. 147-149, London, 1908. 13. Anderson, W. C. and Henderson, J. A. R. (190 2)." Ihe coal s of Bengal and J apan" . J . S, u.I • 21^ 237-242; London, 190 2, 14. Lewes, V.B, (1912), "The Carbonization of coal" .Pp. xiv, 315, London, 19 12.Progressive Age, 29,10 30. 1911. 15. Stopes,M.C. an d eel er, R, V, (10 18 )" Monograph on the constitution of coal" . Oep t. of Sci.and Ind. Research. Lon don, 19 IS, PP* •'5 16. Frazer, J . C, W, and Hoffhian, E.J, (19 12 )" The constituents of coal soluble in Phenol", Tedi. Paper 5. uept.of In terlor,U. S. Bureau of Mines, Waging ton, 19 12. 17. Burgess, M.J. and. IVheel er, R. V," The volatile constituents of coarf J.S.C.I, 606; 1911. 8G 18#Porter,H. C* and Taylor,G*B." The primary volatile proclucts of the carbonization of coal” . Tech -Paper 140. Bureau of Mines, 19 16 . 19. Pictet, A. and Rams^er,L. (19 11 ).”Ueb er einen Bestandteil der Steinkohl ^ .B er. deu t. dh ^.G escll. Vol. 44, pp. 2486 -2497. Ber* 191 20. Wahl, A. (1912).” Recher(hes sur la houille” . oomp t. Ren(^ Acad.Sci,. Vol. 154,p. 1094. Paris, 1912. 21. Cl ark, A.H . and % eel er, R.V, (1913).” The volatile consti tu an ts of coal”. Part III. Trans. Chem- Soc. Vol. 1X33, pp. 17C4-1713. London- 22. San?e as 1, 23. Same as 16. 24. Pi ctet. A, , Rams^er,L. and Kaiser, 0. (1916 ).” Sur qu el qu es hydro - carbon contoius dans la houill e” . Oomp t. Rencti Acad. Sd- Vol. 163, pp. 358 -361, Paris, 1916. 25. Cherry, 0. A, Ihesis U.of I.” The effect of 0 3«/^gQi on the coking of coal” ,19 20. 3S,Parr,S,W, and Hadley, H.F. (1915) The analysis of coal dth phenol as solvent^ Illinois Uni v.Ihg-Expt. Sta.Eull.No. 76, Urbana, 1915. 2T.Lenher, V, (19 21 )j,A. C. S. 4^ 29; 1921. 28, Thiessen, R- (19 20 ) "Structure in Pal eo zoi c Bi tuminou s Coals”. Bulletin 117. U.S. Bureau of Mines, Washing ton, 19 20. 29 , Parr, S. W, (19 22 ) ” Fli el s, Gas, Water and Lubri can ts” . page 95. 30, Earn shaw, E. W,” Way er G as” .Jour. Franlilin In st. vol . 146 ,pp. 16 1-176 « 31, Bo sinan, V, Ih esi s U,of I- 1£)22, "Investigation of South African coals with special reference to their high nitrogen content”, 32, Taylor, G-B, and Porter,H. u, (19 16 ) Tech. Paper 140. Bureau of Mines Washington, 19 16. "Tlie primary volatile products of tJie car- bonization of coal”.