I. The Reduction of Tartaric Acid. II. The Heats of Combustion of Anhydrous Oxalic Acid, Malonic Acid, dl-Malic Acid, Glutaric Acid, Adipic Acid, Azelaic Acid and Sebacic Acid Citation Baumgarten, Werner (1941) I. The Reduction of Tartaric Acid. II. The Heats of Combustion of Anhydrous Oxalic Acid, Malonic Acid, dl-Malic Acid, Glutaric Acid, Adipic Acid, Azelaic Acid and Sebacic Acid. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/h6tg-a563. https://resolver.caltech.edu/CaltechTHESIS:02112025-190035731 Abstract No abstract. Item Type: Thesis (Dissertation (Ph.D.)) Subject Keywords: (Chemistry) Degree Grantor: California Institute of Technology Division: Chemistry and Chemical Engineering Major Option: Chemistry Thesis Availability: Public (worldwide access) Research Advisor(s): Lucas, Howard J. Thesis Committee: Unknown, Unknown Defense Date: 1 January 1941 Record Number: CaltechTHESIS:02112025-190035731 Persistent URL: https://resolver.caltech.edu/CaltechTHESIS:02112025-190035731 DOI: 10.7907/h6tg-a563 Default Usage Policy: No commercial reproduction, distribution, display or performance rights in this work are provided. ID Code: 16994 Collection: CaltechTHESIS Deposited By: Benjamin Perez Deposited On: 13 Feb 2025 20:19 Last Modified: 13 Feb 2025 20:22 Thesis Files PDF - Final Version See Usage Policy. 13MB Repository Staff Only: item control page

I. II. The Reduction of Tartaric Acid. The Heats of Combustion of Anhydrous Oxalic Acid, Malonic Acid, dl-Malic Acid, Glutaric Acid, Adipic Acid, Azelaic Acid and Sebacic Acid. T'nesis by Werner :Baumgarten. In Partial Fulfilment of tne Requirements for the Degree of Doctor of Philosophy, California Institute of Technology, Pasadena, Ca.lifo r:nia, 1 Summary I. In order to correlate the configuration of threo-2,3-buta.nediol with the sugar series we have attempted to prepare an optically active 2,3-butanediol by the reduction of tartaric acid • .!.-threo-0-Carbomethoxy-~.~-diacetoxybutyryl chloride can be obtained by the action of thionyl chloride on monomethyl diacetyltartrate. This acid chloride is easily reduced to methyl diacetyl-l-threuronate with hydrogen in the presence of palladinized barium sulfate. Methyl dia cetyl-1:-threuronate can be reduced in basic solution to l-threonic acid by sodium amalgam. Reduction of 1-threonic s,cid in acid. solution by sodium amalgam yields l-threitol, the configuration of which is established by this synthesis. Unsuccessful attempts were made to reduce ethyl tartrate, isobutyl tartrate, ethyl isopropylidenetartrate, and isobutyl iso:propylidenetartrate. Also, the reduction of methyl diacetyl-l-threuronate ~~s attempted by other means, but without success. II. The isot11ermal heats of combustion at 25° of anhydrous oxalic acid , malonic acid, dl-malic acid, glutaric acid, adipic acid, azelaic acid, and sebacic acid have been determined. These data are compared '7 with the existing values. Using the experimental data and the heats of formation of wa ter and carbon dioxide, the heats of formation of the compounds have been calculated. The free energies of formation hs.ve been calculated making estimates of the entropies. The Reduction of Tartaric Acid. 2 Within recent years there has been a marked revival of interest in the tetroses and related conrpounds. diacetyl-i-xylal 1 Aside from the ozonolysis of the tetroses have been prepared mainly from the pentoses by the well known degradation methods of Ruff 2a, Wohl 2b,c,3 and Weermann. 4 The desirability of using the tartaric acids as sources for tetroses and related compounds was recognized by Emil Fischer 5--who was unsuccessful in attempts to reduce tartaric acid. Following the realization that the lactones, rather than the acids themselves, are the compounds 6 which undergo reduction he planned to reduce esters of tartaric acid . Since no subsequent statement has appeared, it is presumed that results were negative. ** This report is made now, since the synthesis of .&_-tijreitol has been accomplished. The extension of the Rosenmund7 method to sugar chemistry by Cook and Major 8 and by Glattfeld and coworkers9, who have reduced the acid. chlorides of acetylated hyd.roxy aci cL s to acetylated hydroxy aldehydes with hydrogen in the presence of palladinized barium sulfate, suggested a similar method of attack here. When tartaric acid., I, is heated with acetic anhyd.ria.e1° or with benzoyl chloride 11 , it is converted into diacetyltartaric anhydride, II, or dibenzoyltartaric anhydride. The d.ibenzoyl anhyc1.ride is converted by * Dextrorotatory tartaric acid, t-(+)-threo-dihydroxysuccinic acid, is referred to simply as tartaric-acid. ** The method of Bouveault and Blanc, Co~pt. rend. 1:2§, 1676 (1903); ill, 60 (1903) has been applied to the reduction of isobutyl tartrate, ethyl isopropylidenetartra.te ana.isobutyl isopropylidenetartrate without success. Glattfeld and Mochel, J.Am.Chem.Soc. 60, 1oot' (193g) have shown that esters of ~.~-dihydroxyisobutyric acid cannot be reduced this way. 3 water into dibenzoyltartaric acid monohydra.te. The anhydride was obtained when attempts were made to convert the acid to the acid chloride, as noted by others. 12 Diacetyltartaric acid was not obtained in satisfactory yields from the diacetyl anhydride. * An acid chloride can be prepared from a half ester, obtained from one of the anhydrides through a reaction with an alcohol. Al though dibenzoyltartaric anhydride reacts rea dily with primary alcohols, the resulting products are liquids. without decomposition . Moreover, they cannot be distilled However, diacetyltartaric anhydride, II, reacts vigorously with methanol to give the easily purifiable solid monomethyl diacetyltartrate, III. 13 This with thionyl chloride is converted into the cry stalline aci d chloride of monomethyl diacetyltartra.te, IV. The acid chloride can be reduced by the Rosenmund method in xylene at a teCTperature of 130° to 135°. When the reaction mixture stands at room temperature, methyl diacetyl-!-threuronat e , V, separates in fiftyfive to sixty percent yield. The threuronate has a mild aldehyde odor, reduces ~~rm Bene dict ' s solution and gives the characteristic uronic acid test with naphthor e sorcinoi. 14 The color of the resulting dye in ether is reddish violet 0 and it has a strong absorption in the range 5200° to 6000° A, simila r to that of the dye from glucuronic acid, which absorbs in the region of the D line. The threuronate reacts with phenylhydrazine, .:e_-nitrophenyl- hydrazine and 2,4-dinitrophenylhydrazine, yielding solid derivatives with the last two. * Austin and Park, J. Chem. Soc. l~2J.• 1926 (1925) state that diacetyltarta ric acid results from the action of water in acetone . No yield is given. 4 fOOH HrOH HOCH 6ooH H :-:-1 0 AcO :_J HO TI!OAc AcO H OOMe II I \OOH H2 0H t H OAc OOMe OOH V VI III IV Ht:7 H2 0H HOf:_J AcOJH VII IX COOMe HO H CHO VIII 5- In contact with water, hydrolysis of V takes place easily, for essentially three equiva lents of sodium hydroxide must be added before the solution becomes alkaline. When an a queous so l ution of Vis con- centratecl, darkening gradually takes place. Finally a dark gum is obtained, from which no crystalline product could be isolated. This is the first tetruron1c a.cid of which the structure is definitely established. A few previously described compou...."'l.ds may be teturonic acid~, but some doubt exists as to the structurei. Thus from the oxidation of ~-erythrito1 15 Neube rg isolated an acid, C4 H6 0 5 , in the form of the barium salt. Schmidt et al isolated the calcium salt of 3-methylthreuronic acid by hydrolyzing the ozonolysis product of the unsaturated methylation product resulting from the action of diazomethane on saccharic acid, and on saccharolactonic acid. 16 Reduction of the uronic ester, V, was attempted with aluminum isopropoxide ~nd with hydrogen in the presence of a Baney nickel catalyst. * In neither case was any reduction product isolated. Ho wever, V can be reduced in aqueous solution by sodium amalgam, keeping the solution slightly basic. powe r is lo st. The reduction is stopped when the reducing The reduction product 'llro. s a syrup which could not be obtained crystalline. The procedure of G~tzi and Reichstein17 , who obtained .f-thre onolactone, VII, in a crystalline form by means of high vacuum distillation at O. 3 mm., gave a syru:p which would not crystallize. However the aci a. , VI, was characterized as the brucine salt. This salt has been described previously. 17,lS * E. Baer and H. O. L. Fischer, J. Biol. Chem. 128, 463 (1939) , have re duced isopropylidene glyceric aldehyde to i~ropylid.ene glycerol at room tempe rature and. at high pressure. The l a tter was not available here. 6 Reduction of .f- threonic acid with sodium amalgam was carried out under acidic conditions. Under these conditions the compou.YJ.d undergoinp; re duction presumaoly is the lactone, VII, which is in equilibrium with the acid. No attempt was made to isolate the first reduction product .f-threose, VIII, the presence of which was indicated by reducing action of the solution towards Benedict's solution. The reaction was continued until the solution had no reducing action. The final reduction product, .f-threitol, was isolated as dibenzylidene-!-threitol, and from this tthreitol was obtained. The synthesis of !-threitol from tartaric acid is a confirmation of the configumtion of the threitols, a s_ c1etermined by others. Thus Maouenne 19 obtained d-threitol from d-:xylose through d-threose, ancl - ~ Hockett 2 - obtained. di benzylid.ene-_.9:-threi tol from i-:xylonic acj_c. and 2;-:xylo se, through d-th reo se. The formation of t-threi tol by the reduct ion , • h 1• t s con f.igura t ion ' •20• o f ery th nuose d oes no t es t a bl is Experimental lsobutyl Tartra.te.--A mixture of 250 g. (1-2/3 moles) of tartaric acid and 450 g. (5.3 moles) of isobutyl alcohol conta.ining 3% hydrogen chloride was heated so as to obtain a slow distillation of the azeotropic mixture of the alcohol and water through a. fra ct ionating column until anh:tclrous alcohol came ove r. After excess alcohol had been distilled out, the product, which solid.ifies on cooling , distillecl at 155-160° ( 5 mm.). Yield 350 g. (80%). Reu.uction of Tartaric Esters.--The isobutyl ester in absolute a lcohol was addec, slowly to a mixture of sodium and absolute~alcohol. After the sodium had a.isappeare cl, the alcohol was distilled off and the 7 resi c:ue was taken up in water. The solution wa. s neutralized with sulfuric aci d_, boiled with charcoal anct the colorless solution was evaporated to dryness. From this residu e no organic compound could_ be extracted with alcohol. The reduction was repeated_, except that t h e alkali was neutralized oy pota.ssiu.m bi sulfite in one case, e,no_ by carbon dioxic.e in another. The results were negative. Alkyl IsopropylidenetartrE.te.s.---When ethyl tartrate kas heated. with acetone and anhydrous copper sulfate fc,r thi rt y-six hours, as in the 1 method of Tsuzuki 2 , no reaction product was obtained. However, reaction took place when a small amount (a fe w drops per liter) of sulfuric acid was add.ed. The yielrl was 50%. It was founcl preferable to carry out the reaction at room temperature on the shaking machine. Thus t h e isobutyl d.erivative was obtained in 60% yield by shaking for a.bout twelve hours, 131 g. (0.5 mole) of isobutyl tartrate, 800 ml. (10 moles) of acetone and 195 g. (1.?. mole) of anhydrous copper sulfate. 125° ( 2mm.). The product distilled at The yield from ethyl tartra.te was similar. It wa s t h ought that blocking of the hyd.roxyl groups in the i sopropyliclene compounds would facilit a te reduction of the ester groups. However, no reduction product could be isolated when the ethyl or isobutyl isopropylidenetartrate ~~s reduced as above. Dibenzoyltartaric A.nhy1fricle.--This was prepared in practically quantitative yield by the procedure of Butler and Cretcher melting point, 11 . The 193°, was higher than values reported previously (173°). The monohydrate of the free acid, r e sulting when water is added to the anhydride, was reconverted into the anhydride by thionyl chloride or phosphorus pentachloric.e, thus confirming the observations of Zetzsche and. Hubacher. 12 Thionyl chloride in the :presence of a small amount of zinc chloricle acts similarly, although it converts phthalic, succinic ancl fumaric acids into the respective acid chlorides. 22 A crystalline half ester could not be prepared. by dissolving the anhyc.ride in methyl alcohol, followed by removal of excess methanol at reduced pressure. The residual oil would not crystallize. This substantiates the results of Zetzsche and. Hubacher, who however, heated the reactants. The syrupy product from the action cf the anhyQride on benzyl alcohol was converted into the anhydride with thionyl chloride. Diacetyltartaric Anhyc'i.rid~~.--This was obtained in 95% yield when 220 g. (1.5 moles) of tartaric acia. was heated at 60Q for 20 hrs. with 530 g. (5 moles) of acetic anhyd.ride containing 3% of hyd.rogen chloride. The solid after removal of acetic acid. and. excess acetic anhydride, was crystallized from benzene, m.p. 134°. The neutralization equivalent was determined by titration in aqueous solution at room temperature, the saponification equivalent by the method of Redemann and Lucas. 23 Anal. Calcd. for 0 H o : 8 8 7 neutral. eliuivl., 110. Neutralization equiv., 108. Saponification equiv., 54. Found: Fou.nd: Sapon. eq~iv. 53.5. Monomethyl Diacetyltartrate, III.--¼~en 4o ml. ( 1 mole) of methanol is added to 108 g. (0.5 mole) of diacetyltartaric anhydride, the solid dissolves with heating. After removal of excess methanol a.t reduced 9 pressure, 93 g. (75% yield) of monomethyl diacetyltartrate is obtained after crystallization from isopropyl ether; m.p. 124. 7° (corr).• 2 [ ~] 5 D, -18.4° (~, - 0. 75°; ];, 2; .£, 2.32; MeOH) Anal. Cale ~. for c H o: C, 9 12 8 H, 4.88. eq_uiv., neutralization eq_uiv., 62. Found: 43.6; H, 4.84. 248. Found: Found: 244. C, 43.4; Saponifice.tion 63. The Acid. Chloride of Monomethyl Diacetyltartrate, IV (t-threo-it:.= Ca.rbomethoxy- a,,p-dfacetoxybutyryl Chloride) .--Under a reflux conclenser a mixture of 45 g. of III and 45 ml. of redistilled (but yellow) thionyl chloride (b.p. 75°-,78°) was heated at 6oi, for about two hours, by which time all of the solid. had dissolved. The excess thionyl chloride was removed at reduced. pressure, and the residue was recrystallized twice from isopropyl ether, which had been carefully purified from water and peroxide, and stored over sodium. It separa ted as fine needles, yield The solid is not very stable, for it slowly liq_uefies even in a sealed tube, or in a desiccator. The compound was analyzecl for chlorine by first hydrolyzing, and then cl.eterm.ining chlori de ion by the Volhardt method. Anal. 0alcd.• for c is_ 0fl: 9 1 C, 40.15; H, 4.18; Cl, 13.1. C, 40.5; H, 4.12; Cl, 13.3. Saponification eq_uiv., 52.4. Found.: Fount 57 to 59 ■ Other Ha,lf Esters and. Acid Chlorides.--Diacetyltartaric anhyd.rid.e with ethyl, isopropyl or isobutyl alcohol gives a liquid which could not be crystallized. The ethyl and isobutyl esters were converted. to acid. chloricles with thionyl chloricle as above. made as above. The sli€;htly low values are d.ue to the tendency of the compounds to decompose slowly. * Analysis for chlorine was 1 Wrobel 3 reports m.p., 102°. 10 !-threo-__(-Carbethoxy-_£,,~-diacetoxybutyryl chloride, b.p., 137.5° at 6 mm. Anal. Calcd. for c 1½_ o 01: 10 3 7 Cl, 12.65. Fou.nd: 11.8. !-threo-_(-Carbisobuto:xy-£,,~-diacetoxybutyryl chloride, b.p. 147~ at 6.5 mm. Anal. Calc d. for c IS_ o c1: 12 7 7 Cl, 11.5. Found: 10.9. Methyl Diacetyl-t-threuronate, V.--Hydrogen was passed vigorously into 60 ml. of xylene in which 32 g. of IV was dissolved and. 5 g. of palladini zed barium sulfate was suspended.• The temperature was 130° to 135° and the time, three to four hours. The reaction was stopped when the evolution of hyd.!li"ogen chloride slackened., by which time the reduction we,s 75 to 80% complete. The rate of evolution was followed by passing the exit gas through aqueous sodium hydro xide. 1;.ras removed by centrifuging. in yields of 55 to 60%. The catalyst On standing the threuronat e crystallized An a ddi tional crop resulted when the xylene was removed at reduced pressure and the product was distilled. at 3 mm. (b.p., 120 to 130°). The total crude yield was 65 to 70%. The yield was lo wer when the acid chloride or the :xylene was not pure, or when the reduction took longer, as was the case if the catalyst was not we ll suspended, if it became less active, or if the reduction was carried on towards completion. The purity of the proc.uct in some cases was so reduced. that no solid separated. When the reduction was satisfactory, the solution was colorless, or nearly so. Whenever the solution was dark colored the yield was low. Distillation at lo w pressures was necessary whenever the prod.uct failed to crystallize, or we,s dark colored.. * Technicc1,l xylene, previously refluxed. with me tallic sodium for some time and distilled, was satisfactory. The catalyst Wc, s prepared as described in Gattermann, Wieland, "Laboratory Method.s of Organic Chemistry, 11 22nd eclition, The Macmi llan Co., N. Y., 1,2.E, :p. 369. 11 The ester crystallizes in nodules from toluene or from isopropyl ether, m.p. 84° . The former is preferable, for the loss is much less. Rotations were observed in water, in methanol and in methanol containing one percent of hydrogen chloride. The initia.l rotations (a.bout five minutes was req_uired for dissolving) were: _!, 2; ~• 3.259; H2 0); [~ ] 25 r~J 25 - ]25 D = -34.4 • (~, -2.23°; [~ D = - 55.2 (~, -3.68°; _!, 2; ~• 3.314; Me0H); D = - 35.4 (~, - 1.12° ; ]:, l; ~• 3.171; Me0H, HCl). C, 46.5; H, 5.18. Found: C, l.~6.6; H, 4.96. It gives a positive uronic test with naphthoresorcinol, 14 ancl. the resulting reda.ish violet dye in ether absorbs strongly in the range, e 5200 to 6000 A. The threuronate has a mild aldehyde-like ocLor, reduces Benedict I s solution when warmed, and reacts with phenylhydrazine, ~-nitrophenylhydraz ine ancl wi th 2,4-dinitrophenylhydrazine. two, the 15-erivatives were solids. In the case of the las:t These were prepared by carrying out the reactions in hot glacial acetic acic1. and adding water to throw out the products. These were recrystallized fro m hot alcohol. melting points were: The respective p-nitrophenylhydrazone, 143° (corr); 2,4-dinitrophenylhydrazone, 148° (corr). Anal. Ca.led.• for c H o N : 15 17 8 3 C, 49.05; H, 4.63; N, 11.43. Found: C, 43.6; H, 3.88; N, 13.5. Found: C, 49.11; H, 4.39; N, 11.06. Anal. Calcd.• for C15H16 010N4 : C, 43.~6; H, 4.5; N, 1304. The rotatio n in aq_ueous solution remained constant for about t welve hours, then descreased slowly thereafter. After twenty days 12 _D [~ j was -4° . Color developed after two or three days. After t wenty days the solution i,a s dark brown. In methanol tne rotation decreased slowly from the start and after five days reached a constant value of -38°. :Sy this time the solution was colo re d. slightly, but not so much as the aaueous solution. In methanol containing hydrogen chloride the rotation dropped rapidly, rea ching zero in about one day. in four more days then +22°. Then it became positive and The solution was quite dark by this time. Some attempts at red.uction of the methyl dia.cetylthreuronate were unsuccessful. It was hea ted three to four hours with an excess of aluminum isopropoxide. The original materia l ~~s recovered by distillation. The failure here to effect reduction agrees with the experience of others 24 with hydroxyaldehydes. Hycl.rogenation was tried in me thanol and dioxane a.s solvents. lil'ei ther low pressure hydrogenation with palladium oxid.e, nor high pressure hydrogenation with Baney nickel was successful. The recovery of the t et ruronate in the first case by distillation was essentially q_uantitativeo In tne second case the rea.ction product was a dark colorecl solution, from whi ch a small amount of the uronate coul d be recovered. pressure available was about 80 atmospheres. to 120°. The maximum The temperature was 100° Probably the hydrogenation would be :possible at h igher pressures and room temperature. polymerization. High temperatures should be avoided.. because of 13 .e-Threonic Acid_, VI, was obtained by red.ucing V with sodium amalgam in aqueous solution. When 23 g. (O.l mole) of the crystalline methyl uronate was suspended in 150 ml. of ,..,ater and cooled, tne addition of ~p:proximately 0.3 mole of sodium hydroxia.e was required to make the solution alkaline. This shows that removal of the meth;irl a,nd the t wo acetyl groups takes place easily, and. that the compounc1. being reduced was sod.ium .e-threuronate. Red.uction was accomplished by the addition of 3% sodium amalgam in 50 g. batches, keeping tne liquid slightly alkaline to ~henolphtnalein by neutralizing the excess of base with sulfuric acid, and cooling to keep the temperature in the range, 0 to 25°. After about t welve hours, when about twice tne theoretical amount of soa.ium :iad been used, . the solution no longer reduced Fehling' s solution. The solution wax filtered., evaporated. to dryness und.er - reduced nressure, enough 6 :N sulfuric acid was added to liberate the • organic acid and the solution was poured into five times its volume of hot absolute alcohol. The mixture was allowed to stand some hour~, the voluminous precipitate of sodium sulfate was removed by filtration, and the filtrate was concentrated by reduced pressure evaporation to a small volume. This was taken up in a second volume of alcohol and then after filtering, concentration was effected as before. The residue was a pale colored syrup, presumably !-threono lactone , VIII. This -was converted to the brucine salt by dissolving in warm water , add.ing brucine until alkaline, cooling, removing excess brucine with chloroform, and evaporating to dryness. After crystallizing from alcohol, the solid recovered weighed 33 g., which is 62% of the theory, calculated 14 from V. The solid melted at 203-201.~o, dee. J [ £'. 25 D = -23° (2:, -0.95°; l• l; ~' 4.13; H20). ~ Anal. Calcd. for c H 11.7. Found: o N : C, 61.25; H, 6.60; N, 5.5·2; CH 3 0, 27 34 9 2 C, 61.32; H, 6.6; N, 5.22; CH 3 0, 11.8. ~nen the reduction of V was attempted under acidic con di tions, the reaction was much slower. Even when a larger amount of amalgam was used, the reduction was not complete. This shows the importance of keeping the medium slightly alkaline, rather than acidic. Dibenzylidene-t-threitol.--The syrup from the reduction of V was dissolved in water, cooled to 0°, and with good agitation sodium amalgam was added at intervals. The mixture was kept acid to Congo Red by me8.!ls of frequent additions of dilute sulfuric acid. After a short time a test portion reduced Benedict's solution, indicating the presence of t-threose. No attempt was made to isolate this. Instead, the reduction was continued until the test was negative, req_uiring a bout 200% exces.s of sodium. After fil tering, the filtrate was concentrated to a small volume (about 50 ml.) but not to dryness, and poured into a large volume (about 300 ml.) of absolute alcohol. After standing for some time, the solid was removed by filtration, the filtrate was evporated under reduced pressure and the residue was taken up in alcohol. This was saturated with hydrogen chloride and a 10% excess of benzaldehyde added.* After a time, d.ibenzylidene-!-threi tol crystallized in :fine needles. It was allowed. to stand in the cold over night, collected by centrifuging and. washed * Since benzaldehyde is an excellent solvent for t he product, only a slight excess was taken. Preliminary e:iq,eriments on test portions were made to determine the proper amount. 15 with alcohol. Yield. 15% from V. W'.a.en crystallized from hot benzene, the melting point was 221-223° (corr) • l; ~• 0.78, CHCl3). Anal. Calcd.. for c H18o4 : C, 72.5; H, 6.05. 18 H, Found: C, 72.5; 6.27 . .e-Threitol, IX.--When o.45 g. (0.0015 mole) of dibenzylidene-.&.- threitol was heated with 20 ml. of 20% aq_ueous acetic acid, solution took place slowly and \>!as complete after one hour. The solvent vras removed at reduced pressure, the residue was dissolved in a small amount of water, the solution was filtered and the solvent was removed as before. The residue was dissolved in a small amount of absolute alcohol, the insoluble matter centrifuged down and the solution was cooled in a 11 dry ice 11 bath. Crystallization was induced by scratching. thread-like needles se1)arated. Colorless, These were centrifuged down ancl re- crystallized from a small volume of absolute alcphol as above, to give 0.137 g. (0.0011 mole) of !-threitol, m.~. 88°. An additional amount of less pure material was recovered b~r evaporating the mother liq_uors. The yield of !-threitol is 73% for this step, 6.6% on the acid. chloride, IV, and 4.6% on tartaric acid, I. [.9:1 25 D ~ -4.2 (.9:, -0.32; l, l; ~• 7.63 ; H2 0). ** Anal. Calcd. for c4H 10 o1~: C, 39.34 ; H, S.19. Found: 0, 39.28; H, 8. 25. * Previous values for dibenzyl idene-d-threitol: Hockett 2 , m.p. 220222°; a. D, - 77 .9° (-73.2°). Haa'.iienne 1 9, m.p. 231° . 20 D Bertrand found. for .e-threitol, a. -4.4, m. ~) . 88°; Maq_uenne found for i-threitol,-~ D, + 4.3, m.p. 88°. 16 Bibliography 1. W. Freudenberg, Eer. 2. £2, 168 (1932). a. Hockett, J.Am.Chem.Soc. 2,, 2260, 2265 (1935); b. Hockett, Deulofeu, Sedoff and Mendive, ibid. 60, 278 (1938); c. Hockett and Maynard, ibid. 61, 2111 (1939). -- ~ ~ 3. Iwadare, Fukunaga and Kubota , Bull. Chem. Soc. Jauan ];_g, 116 (1937); Chem. Abs. 2,t, 5767 (1937). 4. Weermann, Rec. trav. chim. 2.,, 16 (1917). 5. E. Fischer, Ber. - (1889). --22,£2, 2204 930 (1890). 6. E. Fischer, ibid. 7. Rosenmund, ibid . .2,t, 585 (1918). 8. Cook and Major, J. Am. Chem. Soc.~, 2410 (1936). 9. Gl~ttfeld and Mochel, ibid. _§Q, 1011 (1938); Glattfeld and Straitiff, ibid. 60, 1~(19)8); Glattfeld and Kribben, ibid. 61, 1720 (1939). ~ - -10. Chattaway and Parkes, J, Chem. Soc. 1:,~, 663 (1923). 11. Butler and Cretcher, J. Am. Chem. Soc. ,22, 2605 (1933). 12. Zetzsche and Hubacher, Helv, chim. Acta 1, 291 (1926). 13. Wrobel, Roczniki Chem. ]1, 588 (1933); Chem.Abs. ~, 5410 (1934). 14. B. Tollens, Ber. 2Q, 1788 (1908); C, Neuberg, Bioch. Zeit. ~. 56 (1911). 15. Neuberg, Bioch,Zeit. 24, 167 (1910); Neuberg and. Hirschberg, ibio_. KI., 327 (1910). ~ 16. Schmidt, Zeiser and Dippold, Ber. 12,, 2402 (1937); Schmidt and Guenthert, ibid. ]l, 493 (193gf." 17. Gatzi a nd Reichstein, Helv. chim. Acta _gQ, 1298 (1937); ~, 186 (1938). 18. Nef, Hedenburg and Glattfeld, J, Am. Chem. Soc.~. 1638 (1917). I~~uenne, Bull soc. chim. f3] , £2, 587 (1900); Maquenne and Bertrand, [3] , .s,2, 740 (1901f. 20. Bertrand, Compt. rend. lJ.Q, 1472 (1900); Bertrand, Bull soc. ch.dim. ibid. ~] , £2, 681 (1900). 17 Eull. Soc. Chim. Japan 1,2, 255 (1935); 11:, 362, 587 (1936). 21. Tsuzuki, 22. Kyrid_es, J. Am. Chem. 23. Redemann and Lucas, Ind. Eng. Chem., Anal. Ed.~. 24. Meerwein and Schmidt, Ann. 444, 221 (1925). Soc.~. 206 (1937). -~ 521 (1937). Attel!l]_)ted Reduction of Erythritol to 2,3-Butanediol 13 The red.uction of 1,2,3,4-butanetetrol to 2,3-butanediol was at tempted but has not yet been accor~)lished. Since 1-threitol which 1 is prepared by y~e reduction of tartaric acid , is too valuable a material , erythritol was used as a model to study the reaction. We :proposed to convert erythritol into 1,4-ditosyl 2,3-butanediol. It has been pointed out by Oldham and Rutherford 2 that p-toluene- sulfonylchloride, called for shortness tosylchloride, is a reagent which attacks preferably primary alcoholic groups. The tosylesters of primary alcohols react with sodium iodide in acetone at 80° - 100° replacing the tosylgroup by iodine. desoxysugars, 3a This reaction has been used. in the preparation of since the iodine can easily be reduced off by hydrogen and Baney nickel. Unfortunately the reaction of erythritol with tosylchloride is no t clear cut. When the reaction is allowed to proceed without cooling, the temperature of the mixture rose and no uniform product can be isolated. If, however, the reaction is carried out at 0°1 we were able to obtain a single crystalline compound, which melted sharply at 165°. The analysis of t his compound indicated that more than two molecules of tosylchlorid.e have reacted with erythrit ol. This was confirmed by a molecular weight determination. When the tosylester of erythritol is shaken with acetone and anhydrous coppersulfate, there is no reaction indicating the absence of free hydroxyl groups. 19 The tosylester was heated with sodium iodide in acetone at 100° in an autoclave for several hours. The product contained large amounts of free iodine, and no crystalline product was isolated from the gummy reaction mixture. Supposedly the iodine has replaced more than the two terminal tosylgroups,and in the course of the reaction iodine was split off a.nd 1l."1saturated compounds were formed. Experimental Tosylester of Erythritol. - -4.88 g. erythritol (1 mole) are suspended in 20 ml. anhydrous pyridine (6 moles) and cooled to 0°. To this mixture is slowly added 16 g. tosylchloricLe (2 moles) so that the temperature never rose above 59 • The reaction mixture is kept at 0 9 for 24 hours and then poured into ice water. The precipitate was filtered, was..h.ed with aqueous hydrochloric acid and crystallized from ethyl acetate. ~ne compound melted sharply at 165° and crystallized in fine needles. The yield was 7 g. Anal. Cale. for ditosyl-erythritol c H 18 2 i 8 s2 : C, 50. 25; H, 5.12; s, 14.88. S, 17.3. Found: C, 51.94; H, 4.68; S, 18.60. The molecular weight detennined by a modified Rast method approximately as 700. erythritol is 738. 4 was found The molecular weight of the tetratosyl- The result of the analysis and the molecular weight determination seem to leave little doubt that the compound in question is actually the tetratosylester of erythritol. Attempted Acetonization of Tosyl Erythritol. A glass-stoppered bottle containing 2.15 g. tosylerythritol, 50 ml. acetone, 20 g. anhydrous coppersulfate and a few drops of concentrated sulfuric acid are shaken for 24 hours. The coppersulfate is filtered off, washed with acetone and then shaken with potassium carbonate for half an hour. After filtration the acetone is distilled under reduced pressure in the presence of a small amount of potassium carbonate. acetate. The residue is boiled with ethyl Nearly 2 g. of the starting material were recovered. 21 :Bibliography 1. Lucas and :Baumgart en , J • .Am. Chem. Soc., to be published June, 2. Oldham and Rutherford, ibid.$, 366 (1932). 3. a. b. 4. Levene and Raymond, J. Biol. Chem. 102, 317 (1933); Levene and Compton, ibid. 111, 325 T'1935). -- -- M. J. Schlatter, unpublished work. 1941. The Hee,ts of Combustion of Anhydrous Oxalic Acid, llialonic Acid, dl-1$.alic Acid, G-lutaric Acid,, Adipic Acid, Azelaic Acid and Sebacic Acid. 22 Introduction. During the second half of the last century Berthelot and Thomsen accumulated a vast amount of data on the heat of combustion of organic compounds. Their aim was to obtain the chemical affinities of organic compounds purely from heat data. With the general acceptance of the second law of thermodynamics their data lost its usefulness, since it was shown that AF and not AH is tne driving force of a chemical reaction. However, with the formulation of the third law of thermodynamics the values of the heats of combustion and the heats of formation derived therefrom attained new interest. The third law allows the calculation of absolute entropies from heat capacity measurements or spectroscopic data, thus the free energy can be evaluated by the combination of AH and AS in the second law eq_u.ation AF = AH - Tb.S In the la.st twenty or thirty years there has been a renewed interest in thermal data due to its applicability to structura.l q_uestions and reaction mechanisms. This has made more accurate values of the heats of formation desirable. While the data of the older investigators are remarkably good for their time, newer chemical and physical methods allow us today to attain greater accuracy. In the l a st few years the investigation of the heat of combustion of organic compounds has been undertaken in this laboratory. This report presents the heat of combustion of seven organic dicarboxylic acids. 23 Method and Apparatus. The determination of the heat of combustion of liquid and solid material is carried out most conveniently in the bomb calorimeter. Two methods of bomb calorimetry are in use at the present time. It ha,s been shown that the ordinary and the adiabatic method. give results of essentially the same accuracy. The ordinary method has been employed by Huffman and. coworkers la-g in this laboratory. In this method the rise in temperature of the calorimeter is observed while surrounded by a constant temperature jacket,and a correction is applied for the heat lost to the surroundings. 2 system is essentially that of Dickinson • The calorimetric It consists of three parts, the constant temperature jacket , the calorimeter proper, and. the bomb. The calorimeter is completely surrounded by a double-walled metallic vessel, the jacket, containing a large mass of water. water jacket (Fig. 1 This pg. 25 ) is well stirred and the temperature kept constant within.:!: 0.002° by means of a thermoregulator. The calorimeter proper (Fig. 2 pg. 25 ) is stirred by a propeller driven at 380 r.p.m. by a constant speed motor. There are four possible causes for heat interchange between the calorimeter and the jacket: radiation, convection, conduction, and evaporation or condensation. About one- fifth of the heat is transferre d by radiation, about four- fifths by convection and conduction, and only very little by evaporation. In calculating the correction for the hea t transfer Newton I s law of cooling is assumed to hold over the small range of temperature . This assumption is only approximately true. 24 The heat transfer by convection and conduction depends largely upon the shape and size of the surfaces involved. For small tenrperature differences, however, Newton I s law is a close enough approximation. The heat loss due to evaporation does not follow a simple law. However, with proper precautions evaporation losses can be reduced to insignificant values; for this purpose the stirrer, thermometer, and electrical leads were brought into the calorimeter through watertight seals. Furthermore the main portion of the calorimeter is provided with a cover which fits the calorimeter vessel closely and has only one opening for the thermometer. The heat transfer due to radiat i on is proportional to the difference to some high power of the absolute temperatures of the two surfa.ces. To eliminate transfer of this kind the calorimeter is nickel plated to reduce radiation to a minimum. A Parr bomb (Fig. 3 pg. 25) with a capacity of 380 ml. is used. It was modified in such a way that it is made gas tight by means of a gold washer. The standard illi um electrodes were replaced by platinum electrodes equipped with threaded binding posts for making the electrical connections to the fuse wire. The bomb is placed in the calorimeter and is completely surrounded by water, the same amou.~t of water being used in the calorimeter each time within 0.1 g. The bomb is mounted on sma.11 ivory pegs; this permits the calorimeter liq_uid to pass below the bomb and so facilitates the circulation. 25 Fig. 2 Fig. 1 Fig. 3 26 The temperature is measured by means of a calibrated platinum resistance thermometer, in conjunction with a Mueller type thermometer bridge, and a high sensitivity galvanometer all supplied by Leeds and Northrup Co., the sensitivity of the system being such that 0.00001 ohm (0.0001°) was equivalent to about 0.5 mm. on the galvanometer scale. The galvanometer t,ias used as a null point instrument. The times at which a predetermined series of resistances were reached were recorded on a drum type chronograph with a precision of 0.2 seconds. The charge is fired by heating a platinum wire, electrically, which ignites a paper fuse which in turn ignites the charge. required for firing was found to be 1.4 .:!: 0.2 calories. The energyla Since the filter paper is hygroscopic it is kept in a desiccator at constant humidity approximately that of the air in the room. The weight of the fuse is adjusted to 4 .:!: 0.3 mg. but the weight is actually determined to the nearest 0.01 mg. The heat of combustion of the filter paper 1 has been determined as 3,893.2 cal/g. a. In the experiments with anhydrous oxalic acid and d.1-malic acid an auxiliary substance was found necessary for complete combustion. For this purpose Cenco vacuum pump oil was used. had been determined as 10,825 ~ 2 cal / g. Its heat of combustion la The combustion samples are pressed into pellets to simplify handling and transferring. They are then dried. taken that the samples are as dry as possible. Great care has to be For this purpose the drying process is varied in length of time and intensity. First the samples are dried to constant weight at room temperature in a desiccator 27 over phosphorus pentoxide. treatments. They are then subjected to different Some are dried at elevated temperatures in air, others in a vacuum desiccator over phosphorus pentoxide. Finally some of the samples are fused)provided no decomposition occurs during the process. When more strenuous treatment did not change the heat of combustion there was reasonable assurance the drying of the samples 1 was complete. Before the combustion the samples are weighed on a Kuhlmann micro balance to 0.01 mg. A small amount of water is placed in the bomb, the platinum crucible with pellet inserted, the bomb evacuated to a few millimeters and filled with oxygen at a pressure of 30 atmospheres. The water is placed in the bomb to saturate the space with wa,ter vapor,, thus ma..~ing the correction for the vaporization of the water formed during the combustion negligible. It is furthermore necessar.r to evacuate the bomb, since the nitrogen of the air is partially burned to nitric acid, and it is advisable to keep tkis correction as small as possible. The method of determining this correction will be discussed later. The constant range method has been adopted in this laboratory. All the combustions are started at the same initial temperature of 24~ and the mass of the samples are adjusted to give a rise in temperature of 2°. An auxiliary heater in the calorimeter is used to facilitate th~ attainment of the initial temperature. 28 The Calibrations. Either of two methods may be used for the calibration of the calorimeter. The electrical energy equivalent of a calorimetric system, consisting of calorimeter, bomb, oxygen, electrical heater, thermometer, and a weighed amount of water, is determined by supplying a measured quantity of energy electrically and observing the resultant rise in temperature. The conditions in these experiments should be as nearly identical as possible to those in the combustion experiments. The results of such observations give directly the energy equivalent of the calorimeter and its contents, in joules per degree. In this laboratory the energy equivalent of the calorimetric system, consi:ting of calorimeter, bomb, electric thermometer, and a ,-,eighed amount of water, but without the bomb contents, is determined by use of the second method in which a stand.8,rd sample of benzoic acid (sample 39 e) supplied by the Bureau of Standards is burned. Its isothermal heat of combustion under standard conditions at 25° is given as -~UB/m = 26,419 international joules per gram true rnass 3. The sample is burned at 25° in oxygen under an initial pressure of 30 atmospheres absolute, in a bomb of constant volume, the mass of sample and the mass of water placed in the bomb each being 3 g. per liter of bomb volume. Since our conditions differ from these standard ones 1 we have calculated 26,418 international joules are evolved by the combustion of one gram of benzoic acid under our bomb conditions. \fu.ile this work was in progress a new calorimeter was installed. Furthermore we decreased the amount of water added to the bomb from 29 1 ml. to 0.1 ml. Under these bomb conditions we calculated with the formula given by Jessup and Green 3 for the heat of combustion of benzoic acid at 25° the value 26,416 international jo'l,lles per gram true mass. The energy equivalent of the calorimeter was determined at frequent intervals during this work. We have found 3,153.2 ! 0. 3 cal. for the old and 3,222.72 _! 0.25 cal. for the new calorimeter. A small amount of nitric acid is always formed during the combustion. The amount was determined by titration with standard sodium hydroxide using brom cresol green as the indicator. For the heat of formation of nitric acid at 25° and constant volume the va.lue 13,9 60 cal. Mole -1 is used. 4 Our results are expressed in the conventional calorie which is derived from the absolute joule by the arbitrary definition 4.185 absolute joule= l calorie. Since l international joule= 1.0004 absolute joule was accepted as the true relation between the internatioroljoule and the absolute joule it follows that 4.1833 international joule= 4.185 absolute joule= 1 calorie. All the thermal data are given for the isothermal process at 25° and true mass. The calorimetric 1etennination is carried out between the temperatures of 24° and 26°. To correct the experimental heat of reaction to that of the isothermal reaction at 25° we applied equations given by Washburn5 involving the heat capacities of the bomb contents consisting of the platinum crucible, oxygen, and water added_ to the bomb and the reactants ana_ :products in the initial and final state. 30 The molecular weights are based on the 1940 table of atomic weig.~ts. Preparation and Purification of Combustion Samples. Anhydrous Oxalic Acid. The anhydrous acid is prepared from Baker's Acid Oxalic Special for Standardizing by dehydration in the vacuum of the oil pump over concentrated sulfuric acid at approximately 65°0. Sample A is the com.~ercial sample treated as described, sample] was recrystallized once from water prior to dehydration. The dehydrated oxalic acid contained small amounts of formic acid; tp.is impurity, however, could be removed by maintaining the sample in a vacuum desiccator over phosphorus pentoxide. It should be noted that no decomposition of oxalic acid to formic acid occurred at temperatures as high as 45° C. Malonic Acid. Samples A and B were prepared from Phanstiehl c.p. malonic acid by 4 and 3 crystallizations from redistilled ~~ter in the usual manner. Samples E and F were Eastman's malonic acid 3 and 4 times crystallized from water, respectively. dl-Ma.lic Acid. Eastman's practical d.1-malic acid was dissolved in water at 80° and after filtration the clear solution was allowed to stand at room temperature until crystallization was complete. The acid separates in very hard. CIJ Stals. The mother liq_uor was decanted off and this whole process repeated.. Samples A and] were crystallized 7 4 and 3 times respectively. Glutaric Acid. The samples A, ], and H were prepared from tri- 6 methylene cyanide by the method of Organic Syntheses . crystallized 5 times, sample B 6 times from water. Sample A was Sample H was a composite from the samples A and B and was crystallized 5 times from 31 wate r . Samples E and K were Eastman I s c.p. glutaric acid crystallized t wice from water. The differences in the last two samples were that they originated from two different batches of material. Ad.ipic Acid. adipic acid. The combustion samples were prepared from Eastman I s They were purified. by successive crystallizations from water in the usual manner. Sample A was twice, sample :B three times and sample C six times crystallized from water. Azela.ic Acid. Two combustion samples were prepared from Eastman I s azelaic acid by three and five crystallizations from redistilled water respectively. Sebacic Acid. Two combustion samples A and :B were prepared from Eastman's sebacic acid by three and five crystallizations from redistilled water respectively. Results. The results of the combustion experiments are given in tables 1 to 7. The observed temperature rise of the calorimeter was corrected 2 for heat transfer and heat of stirring by Dickinson's method . The heat produced by the combustion of the sample was calculated by multiplying the corrected temperature rise by the energy equivalent of the system. From the total heat evolved were subtracted the energy of firing, the heat evolved by the combustion of the fuse, the energy_ of formation of the nitric acid formed during the combustion, and the energy of the oil when an auxiliary substance was used. Since the mean temperature of the combustion is 25°, the experimental data have been corrected to ] ] ] ]3 A A 1 2 3 4 6 7 Sample 60 hrs. Desiccation P 2 05 vacuum .7 .7 6,567.29 6,464.33 2.28124 2.22752 45.5 hrs. 45. 5 hrs. .6 6,506.32 2.09934 .5 1.0 1.0 HN0 3 Gal.from 100 hrs. 5,767.77 6,524.13 b,512.60 Total heat evolved cal. 1.66347 2.54717 2.14191 'l'rue mass 100 hrs. 42 hrs. 450 Hours in vacuum oven Anhydrous Oxalic Acid Table 1 rn. 25 4,655.29 4.08 4,97s.31 3.44 5, 044.02 3.84 .! .53 66Lf 26 Mean 0 + .18 664.44 17.41 - .64 + .6~ - .93 + . 59 + .17 663.62 664.90 663.33 664.85 Deviation cal. 664.43 19.35 16.14 16.54 4, 829 .68 4.00 5,109.51 4.4 17.17 5,081. so 2.72 Cal.from Cal.from Cal.from -AU~/m oil carbon paper EIT ca /g \.,-J f\J 576 B B 7 9 11 168 3. 23220 6,430.54 .2 + .13 .! .35 1,986.83 1,936.70 Mean + .18 + .35 - .44 17.45 1,986.88 1,987.08 1,986.26 - .55 + .46 - .17 *Combustions done with new calorimeter and with 0.1 ml. water in bomb. Corrected by+ 0.0135% to be comparable with other data in set. Where ash observed was greater than 0.07 mg. the observed weights were corrected by subtracting the ash found. 96 F* 14 16.34 3.12217 120 96 E* 13 .2 17.01 .oo 6,194.48 3.11318 52 528 B* 6,212.33 17.35 .1 6,299.59 1,986.15 20. 26 .1 3.16778 1,987.16 18.11 .1 1,986.53 1s.54 + .60 - .23 ,.1, 986.47 1,9s7.30 - .36 Deviation cal. 1,986.34 cal/g -!:N B/m 19.23 17.99 17.32 Cal.from paper EIT 528 6,394.78 .3 .3 .2 .3 Cal.from HN0 3 6,374.41 4o 3.21394 6,300.79 6,321.72 6,326.71 6,511.04 cal. Total heat evolved 3. 20421 528 B 4 3.16725 87 912 A 10 3.17622 40 816 A 8 3.18074 600 A 3 true mass 5.27418 Hrs. in vacuum 50-55° 816 A 2 Sample Desiccation P 2 0 5 vacuum Ma.lonic Acid Table 2 v-J \_,,-J A A A A B B B 11 16 18 20 10 12 17 Sample 2. 59720 1296 2.57488 6,505.59 b,398.86 6,543.47 270.41 353.33 .4 443.83 399. 77 515.38 457.35 572.10 oil Cal. from .2 .3 .5 6,464.17 2.56084 2.58727 360 .6 6,456.44 2.50705 .4 HN0 3 lial.from .4 6,553.10 Total heat evolved cal. 6,285.52 2.45963 2.52467 True mass 168 504 70 70 34 432 Desiccation hours in vacuum oven PzOs vacuum PzOs 55° en-Ma.lie Acid 'l'able 3 .19 + .29 + .21 2,365.07 Mean - .16 - . 28 + 2,36'.).36 2,364.92 2,364.79 2,365.26 + • 27 16.34 17.21 18.13 18.57 2,365.34 2,365. 02 18. 21 18.17 - .25 2,364.82 18.17 - .05 Deviation cal. -/1U /m cafJg Cal.from pape r + EIT + v-1 .4 456 E H H H K 15 22 25 26 27* 28* 55 18.03 .3 6,332.10 1.62459 144 E 7 192 192 360 336 312 1. 6224o 44 6,336.61 6,537.62 6,417.58 1.62569 1. 67648 1.64558 26 6,304.35 1.61739 6,313.56 6,322.20 6,312.45 26 1. 61971 1.61980 44 + .44 3,892.24 Mean Corrected by + .94 + .96 3,893.20 3,893.18 17 .69 1s.9s - .26 - .01 + .01 - .02 + .38 -1.63 - .72 1.4 .4 3, 891.98 3, 892.23 3,892.25 3,892.22 3,892.62 3,890.61 3,891.52 + .32 - .13 - . 13 + .24 Devia.tion cal. 18.27 17.87 19.23 rn.9s 17.75 3, 892.56 3,892.11 3,892.11 3, 89 2.48 cafJg -liU. /m .4 .4 .3 .4 19.02 *These combustions were carried out with the new calorimeter and .1 ml. water in bomb. + 0.0117% to be comparable with other dati:,. K 17.75 .3 6,339.98 1.62607 1344 B 21 1032 .35 6,166.46 17.os 18.19 B 1.58149 .8 6,267. 43 .8 19 288 1.60804 6,314. 09 B 41~ 1.61976 1 888 360 A 17.39 14 44 .2 Cal.from paper + EIT A b,268.56 HN03 Cal.from 4 1.60825 Total heat evolved cal. A 336 oven 85° True ma s s 3 Sample Desiccation hours in P2 0 5 vacuum vacuum oven P 2 0s :,50 Glutaric Acid Table 4 vJ Vl 648 C C C C C A C 5 12 13 14 15 10 8 960 6!., fused at 165° fused at 165° 65.5 94 94 94 1.45703 1.40893 1.38149 1.39882 1.37730 1.41311 2.0 6,677.41 + .88 4,575.00 Mean O'I \J.J - .45 + .70 +1.22 - .17 - .92 - .81 4.514.55 4.575.70 4.576.22 4.574.83 4.574.08 4. 57L~.19 +2.13 -1.07 - .75 - .66 + . 77 Deviation cal. 18.65 15.42 18. 77 2.0 2.5 17.09 17.93 .8 .8 16.58 .or 4.577.13 4.573.93 18. )_~5 17.97 4.574.25 4.574.34 • 4. 575. 77 -LWj/m ca /g 21.13 17.45 18.05 Cal.from :paper+:!llIT 2.2 1.0 2.0 2.0 1.2 ~HN0 3 Cal.from 6,456.53 6,334.71 6,409.09 6,310.50 6,472.77 6,430.52 6,420.56 1.40128 150 192 B 11 1. 40231 6,276.47 1.36883 68 1580 B 7 25 6,457.91 1.40932 25 65.5 288 A 9 75° 6,442.99 PzOs 1.40566 55° b8 P20s 65.5 1676 A 6 Sample Total heat evolved cal. True mass Oven 105° Desiccation hours in P 2 05 vacuu~ vacuum oven vacuum oven Adipic Acid Table 5 A 12 A 10 A B 9 11 B 7 408 1.12911 24 hrs. 6,815.03 .4 .6 6,889.90 1.13945 6,587.3$ 1.08848 20 min. 5.5 hrs. 1.0 6,520.93 .6 .6 .4 1.07793 6,368.0l 6,431.11 35 min. fused in oven 115 for 1.05276 1.06321 46 264 24 B 6 1.07541 46 264 .4 B 3 6,505.10 17.97 .6 6,472.52 1.06998 B 2 312 16.58 .5 6,499.66 1.07448 139 264 A 8 /m 6,041.75 mean 9,10 6,026.09 6,037.59 6,042.25 6,otn.28 b,03e.7s 6,038.74 6,038.53 6,039.57 6,039.06 6,039.48 6,039.44 6,037.50 6,037.97 ca~/g -b,U 19.24 18.45 17.41 15.98 mean 17.73 18.21 17.41 16.34 .4 6,464.08 1.06882 44 336 A 5 17.25 16.70 .4 1.5 6,437.56 6,905.4o 1.06462 1.14212 Cal.from Cal.from pa:per+EIT ill'J03 44 A 4 216 Total heat evolved cal. 336 A 1 Sample Azelaic Acid Desiccation hours in True P 2 05 vacuum vacuum oven oven 65° oven 98° mass P2 06 56° Table 6 + .60 + .60 - . . o4 - .25 + .79 + .28 + .70 + .66 -1.28 - .81 Deviation cal. vJ -.J A A A B B :B B 9 10 12 7 8 11 13 Bample 6,392.39 .99878 32 624 282 6, 352.94 6,277.28 .98018 91.5 .99240 1.0 6,291.43 .98267 18.5 1.0 .5 .8 .2 6,300.93 86l~ .98496 .3 2.0 HN03 Cal.from 79 6,380.47 Total heat evolved cal . 6,389.20 .99664 True mass .99849 32 70° 1176 280 114 PzOs Des i ccation hours in P 2 05 vacuum vacuum oven oven 106° Sebacic Acid Table 7 + .60 + 1.02 6,390.03 Mean - .71 6,389.32 6,390.63 +1.84 .32 6,391.87 + -1.82 -1.05 + .83 Deviation cal. 17.49 17.97 18. 77 6,390.35 6,388 .21 16.10 18. 29 6,388.98 6,390.86 ca_Jg -t.U~/m 17.17 16.62 Cal.from pa-per+EIT 1'"\- v-1 Oil. 39 the isothermal process at this temperature by applying the equations of Washburn5 as described above in the case of benzoic acid. W'.aen the heat capacity of the compounds were not known estimates have been made. The energy evolved in the isothermal bomb process per mole is represented by the sumbol -~UE. In contrast to the actual bomb process Washburn5 proposed an analogous process. It consists in the reaction of unit quanti}y of the substance with an equivalent amount of pure oxygen gas, both under a pressure of 1 atmosphere and at the temperature tH, to produce pure carbon dioxide gas and pure liquid water, both under a pressure of 1 atmosphere and the same temperature tH' the reaction taking place without the :production of any external work. The energy change in this hypothetical process is a definite thermodynamic quantity and is equal to -6UR' the decrease in energy for the reaction at tH: Ca-7) RO C (s) or (t), 1 atm. + b - 2c a+ -....,4-- The quantity -~UR is calculated by means of the correction factor given by Washburn. From this quantity - ~~. the energy evolved in the isobaric reaction at 1 atmosphere, is calculated Dy the addition of the proper work term. Finally the heat of formation - 6Hf is calculated from -6~ Dy using 94,030 calories for the heat of formation of carbon dioxide and 68,317 calories for the heat of formation of liquid water7. Rossini 8 has pointed out the advisability of a uniform procedure 4o for the determination of the precision error assigned to the different values. This uniform comparison will be concerned, however, only with the precision, the reproducibility, of the various sets of data and not with their actual accuracy. The accuracy will depend largely upon the magnitude of the unkno~n or unaccounted for systematic errors. Serious errors may arise from impure combustion samples and from incomplete combustion experiments. Jessup and Green 3 have assigned an accuracy uncertainty of 0.023% to the determination of the heat of combustion of benzoic acid.• From the result of the calibration experiments the calculated calibration error is 0.005%. ~ From the data on the combustions the reaction errors have been determined. The only additional error which will affect the absolute accuracy of the final values is the presen<re of impurities. Any error due to this cause should be very small a.nd therefore the precision error 1 has been assumed to be the same as the accuracy error. In calculating the error in -~UR a 10% uncertainty is arbitrarily allowed for in the Washburn correction. In table 8 the derived values -~UE' -~UR,-~~. and -~Hf are compiled_. The temperature coefficients of the heat of combustion a1::.UB/a!1:. are calculated by making use of Washburn 1 s5 ec1uation and the known or estimated heat capacities. 134,088 132,114 C4HsOs C5H 0 4 8 CsH1o04 C9H1s04 C10Hl804 Malic acid_ Gluta.ric acid Adi:pic acid Azelaic acid Sebacic acid 202,244 188,218 146,14 1.5* 1.03 1.5* 1.429 (lLf.5°) 1.601 ( 20°) 1. 6305 (150) 104,062 C3H404 Ma.lonic a.cid 2.0 - 668,591 + 181 514,219 _! 129 317,128 .: 78 206,740 .: 5.6 59,807 .: 49.9 :a -6U 668,115 .: 187 513,730 .: 139 316,665 .: 92 206,316 :t 69 59 , ao8 .:!: 5 • 9 -6UR 1,292,345 1,291,902 ,: 370 ! 362 1,137 , 166 1,136 , 597 + 284 ,: 295 Density 90,036 C2 H2 04 Oxalic acid M. W. Formula Substance *Estimated (?5 Summary of Derived Data at 25.0° Table - f 1,293,382 .: 370 1,137,781 ! 295 237,534 + 204 668,411 .: 187 261,771 ! 395 255,019 ! 320 - 229,668 .: 154 - 264,998 + 106 - 213,004 + 78 197,757 + 60 -6H 513,730 .:!: 139 jl6,073 + 92 205,720 .: 69 58,620 ± 54.9 -6HR 4.4 tl04 t92 .,.55 ,-41 t- 36 +18 +- :a dt d 6U -I=' I-' 42 Discussion of Results. The acids of the oxalic acid series have been burned by Verkade 9 and. coworkers. They have obtained their data u.nd_er conditions which are different from those of this investigation and in addition have used a different calorie. For the calibration of his calorimeter Verkade uses as the isothermal heat of combustion of benzoic acid at 19.5° 6324 ca1150 per gram benzoic acid (not corrected to true mass). Their value is corrected to true mass and by applica.tion of the temperature coefficient of benzoic acid to the isothermal heat of benzoic aci d at 25° per gra.."!l true mass. It follows by comparison of the value for benzoic acid used by Verkade and the one obtained by Jesst'lp and Green, that Verkade 1 s calorie has to be multiplied by the factor 0.99959 to make it compa,rable to the conventional calorie used in this laboratory. In table~ the values so corrected are compared with our results. 4 Becker and Roth have determined the heat of combustion of the dihydrate of o:r..alic acid anc1- have found. -6~ = 52.23 Kca.1/mol. They have also measured the heats of solution of both the anhydrous and the dihydrate of oxalic acid. From these data-~~ of anhydrous oxalic acid can be calculated. This calculation gives-~~= 58.8 Kcal/mol. We found from our combustion data the value 58.6 Kcal/mol. comparison is very favorable. The Verkade found for anhydrous oxalic acid-~~= 59.7 Kcal/mol. Why his value differs so largely ce.n not be stated with certainty. 10 Paessler has also determined the heat of combustion of anhydrous oxalic aci d and found =~HR= 58.4 Kcal/mol. Table 9 Comp~risons of Verkade 1 s and Our Data at 25.0° and for Standard Condit io ns values of-~ at 25°c. % Deviation , Cal. Substance Verkade This research Oxalic acid 59,67J cal. 58,620 cal. + 1.8 + 1,055 Malonic a.cid 205,935 205,720 + .10 + 215 Glutaric acid 514,072 513,730 + .067 + 342 Adipic acid 668,000 668,411 - .062 Azelaic acid 1,140,930 1,137,781 + . 28 + 3,149 Sebacic acid 1,296,242 1,293,382 + .22 + 2,860 411 No d.etails concerning the bomb conditions a.re given which makes a comp;,rison impossible. We have not been able to find a value for the heat of combustion of dl-malic acid in the literature. 11 In Landolt-Boernstein Ta.bellen a value for 1-malic acid is given. It is interesting to note that the higher homologues in the dicarboxylic acid series contain a small percentage of water which is rather difficult to remove. Both adipic and sebacic acid lose the la.st trace of wa.ter only when dried at 105°, as indicated by the higher heat of combustion of samples dried at the elevated temperature. Azelaic acid due to its lower melting point (105°) was dried at 96°. However fused samples gave a higher heat of combustion. Since only two combustions with fused samples were carried out, the heat of combustion of the sanrples dried below the melting point were used in the calculation of the precision error. When azelaic acid, however, is fused over a long period of time, the heat of combustion decreases. It is seen from the experimental data in table 6 that the longer the time of fusio n 1 the greater is the decrease in the heat of combustion. It is assumed that this drop in the heat of combustion is due to the formation of an unsaturated substance. With glutaric acid a similar effect is observed. Since glutaric acid is known to occur in two crystal structures it might be possible that the lowering of the heat of combustion is due to isomorphism. In Table 10 the free energies of formation of the seven organic acids are compiled. Oxalic and succinic acid are the only members of the oxalic acid series with known entropies. The entropies of the other compounds have been estimated by the method described by Parks and Huffman entropy units. 12 . The uncertainty is probably as high as three The values of tne free energies are probably uncertain to at least ♦ 1,000 calories. 46 Table 10 Thermal Data at 298.1° K. Substance 8 298. 1 E.U. tJI f298-. l Tb.S b.F -· 298 . l Oxalic acid 28.7 -197,757 -30,725 -165,925 Malonic acid 35.4 -213,004 -38,582 -174 , 422 . . aci.dlf Succ1.n1.c 42.0 - 224,829 -46,269 -178,560 Glutaric acid 49.0 -229, 668 -53,950 - 175,718 Adi:pic acid 57 -237, 534 -61,283 -176,251 Azelaic acid 79 -255,019 -83,878 -171,141 Sebacic acid 86 -261, 771 -91,508 -170,213 dl-Malic acicl 40.5 -264,998 -54,088 -210, 910 Bibliography 1. ~,) H. M. Huffman and E. L. Ellis, J.Am.Chem.Soc • .5.I, 41 (1935) b) R. D. Stiehler and H. M. Huffman, i bid • .5.I, 1734 (1935~ c) H. M. Huffman, E. L. Ellis and S. Fox, ibicl. ~. 1728 (1936) d) H. M. Huffman, s. Fox and E. L. Ellis, ibid • .22, 2144 (1937) e) H. M. Huffman, ibid. 60, 1171 (1938) f) H. M. Huffman ands. Fox, ibid. 60, 14oO (1938) g) H. M. Huffman, ibid. 62, 1009 (194o) 2. Di ckinson, Bull. Bureau of Standards 11, 243 (1915) 3. Jessup and Green, Bur . Standards J. Res ea rch 1], 469 (193~) 4. Beckers and Roth, Z. Elekt r ochem. 4o, 836 (1931~) 5. Washburn, Bur. Standards J. Research 10, 525 (1933) 6. Organic Syntheses Coll ect ive Volume I, pg. 283 7. Rossini, Bur. Standards J. Research 22, 407 (1939) 8. Rossini, Chem. Rev. 18, 233 (1935) 9. Verkade, Hartmann and Coops, Rec. trav. chim. ~. 373 (1926) 10. Passler, Lundo l t-BBrnstein Tabell en, Zweiter Erganzungeband, pg . 1643 11. Lundol t-Bornstein Tabell en, Erster Ergauzungsband, pg. 873 12. Parks and Huffman, 11 The Free Energies of Some Organic Compounds," The Chemica,l Ca talog Co., lTew York, 1932 Acknowledgment I wish to thank Professors H.J. Lucas and H. M. Huffman for their help and encouragement. Propositions Submitted b y Werner Baumgarten. Ph.D. Oral Examination, May 23 , 1941; 1:00 P. M., Room 27 Gates. Committee: 1. Lucas (Chairman), Buchman, Dickinson, Huffman, Koepfli, Niemann, Pauling, Zechmeister. An explanation for the o~o tical stability o:r some biphenyls with only two ortho substituents has been given by Calvin. We may expect however that similar optically active biphenyls exist where resonance of the type postulated by Calvin is not possible. Calvin, J. Org. Chem.~. 256 (1939). 2. The u.nreactivity of certain p-toluenesulfonic esters can be attributed to steric hindrance. Levene and Tipson, J. Biol. Chem. 120, 607 (1937) Mueller and Reichstein, Rel. chim Acta. 20, 1529 (1937). 3. Contrary to the statement of Raymond, Bergmann 1 s glucal-hydrobromide-diaceta te is likely to be l - bromo-2- desoxy- diacetylglucose. Raymond, Organic Chemistry, John Wiley and Sons; pg . 1500. Fischer, Bergmann and Schotte, Ber. 53, 518 (1920). 4. The results of the ozonization experiments of Shriner and Moffet as we ll as of the nitration of flavylium salts can be explained by assuming a resonating structure for t hese salts. Shriner and Moffet , J. Am . Chem. Soc. 62, 2711 (1940). LeFevre and LeFevre, J. Chem. ~oc. 1988 (1932). 5. Freudenberg and Brauns state: 11 Der Vorschlag die links drehende Glycerlnsau.re u.nd Milchsaure der d-Reihe zuzurechnen ist umso berechtigter als die Salze, Ather and Ester dieser beiden Sauren nach rechts drehen. 11 They propose t wo independent stereochemical families for the hydroxy acids and the sugars themselves. This inconsistency should be eliminated. Freudenberg and Brauns, Ber. 2.2• 1339 (1922) Freudenberg , Stereochemie F. Deuticke, Leipzig und Wien, 1933: :9g. 668. Muller and Reich stein, Rel. chim. Acta 21, 251 (1938). 6. A course showing the historical development of chemical concepts would be especially beneficial if given to graduate students. /. Only 2- and W -deso:x:ysugars are easily synthesized. A method is proposed by which other desoxysugars can be obtained. 8. The phenomena of alternation in homologous series may be more adequately expressed by a comparison of the ~F than of the ~H for the alternating increments per CH 2 group. To show that the alternation is a property of the crystal lattice it is proposed to compare the free energies of formation in a dilute aqueous solution. 9. The formation of O -bromoacetoacetyl bromide by the action of bromine on diketene is ·best explained by a ssuming vinylaceto-~lactone as the structure of diketene. Boese, Ind. Eng. Chem. 32, 20 (194o) Hurd and Abernethy, J. Am. Chem. Sillc. 62, 1147 (1940). 10 . The reaction between an optically active acid chloride and diazomethane can be d.irected to yield a homologous acid which is still optically active . This may be compared with tne Hofmann or Curtius degradation clarifying if retention or inversion occurs during tne degradation. Arndt and Eistert, Ber. 68, 200 (1935) Preobrashenski, Poljakowa and Preobrashenski, Ber. 68, 8::,0 ~1935). 11. Clarke and l3tegeman determined the molal heat of combustion of a,-lactose monohydrate to be larger by 9 Kcal. than that of 13-lactose. Their resul ts are contrary to the knoim stability relationships. Clarke and Stegeman, J. Am. Chem. Soc. 61, 1726 (1939).