1214 READ AND HOOK THE PREPARATION AND CXX XI1 .-The Preparation and Characterisation Xth ylenebromohydrin . of By JOHN READ and REXFORD GEORGE HOOK. ACCORDING to the1 somewhatl scanty relferences available in the literature it appears that eithylenebrolmohydrin h a hitherto been prepared only in an impure colndition. This remarkable fad in the case of so1 simple a substance is to be attributed not so much to instabiliby as to the nature of ths preparative metholds adopted Thus in the older meithods ethylene glycol was utilised as the startr ing point the second reagent being either ethylene dibromide (Lourenp Ann. Chzm. Phys. 1863 [iii] 67 275) hydrolgen bromide (Henry Jahresber. 1872 304) or phmpholrus tribromide (Demole Ber. 1876 9 48) ; whelreas more rejeelntly application has beein made of the interaction of elthylelneioadohydrin and brolmine (Henry ibld.1889 1321) and the direct additioln of hypobromous acid to ethylene (Mokievski J . Rziss. Phys. Chem. SOC. 1898 30, 900). Of the principles involved in these methods the last is undoubtedly to be preferred as least liable1 to1 give rise to impurities in the resultant ethylenebromohydrin and the advanoe which has lately been made in the experimental application of this principle (Read and Williams T. 1917 111 240; this vol. p. 359) has now enabled us t o undertake the preiparation of pure ethylenelbromo-hydrin in quantity. At the same time further study has bem directed to the course of the reaction between ethylene and dilute bromine water the investigation being facilitated by the oibservs-tion that as anticipated the amount of brolmohydrin formeld dunng definit\e intervals may be wtimated wit,h sufficient accuracy by a titrimetric determinatJon of the accompanying hydrogen bromide, produced in accordance with the folllolwing scheme! : HOH t Br HOBr + HBr -+ J.C,H4Br2 C,HE,Br*O CHARACTERISATION OF ETHYLENEBROMOHYDRIN. 12 15 The results indiaate a remarkable preponderanoe of bromohydrin in the initial stages of the reaction; thus when the concentration of brolmohydrin has reached 0*2N the molecular ratio of ethylene-brolmohydrin to ethylene dibromide is aboatl l O l and in more dilute solutions the value is still higher. 'The characteristics of the pure ethylenebroinolhydrin now obtained differ markedly from the data recorded by previous observelrs the density at 20° f o r instance being 1.7629 as compared with the value 1.7195 a t 18.6O given by Henry (loc.c i t . ) . I n view of the specifia interest of this substance as a simple derivative of ethyl alcolhol and of its general potentialities as a synthetic agent it seemed important to establish its main physical and chemical characteristics and the present paper records some of the results of inquiries directed tolwards these ends. E x P E R I M E N T A L. Yreparati,on of Pure Et hylenebromo hydrin. Bromine vapour and ethylelne were passed into 1500 C.C. of dis-tilled water under the1 conditions described previously (T. 1920, 117 359) the operatlion being continued for several days until 450 grams of bromine had reaoted.The\ rate of absorption of the bromine which was slow at first increased appreciably as the reac-tion progressed. Ultimately the liquid was neutralised in the cold by the1 addition od sodid sodium carbonate and the aqueous layer was separated frolm ethylene dibromide. The latter of which nearly half the total yield had volatilised in the course of the experiment, was shaken once with water in order to extract dissolved broma-hydrin and dried over calcium chloride; the yield amoanbd to 80 grams. The amount of hydrogen bromide produced in the reac-tion estimated by titration with standard silver nitlrate solution, corresponded with 35 per cent. of the bromine addeld and thus indicated it 70 per oent. yield of ethylenebromohydrin. I n order t o alloiw of the eventual recovery of the bromine present a.s bromide, the aqueous solution was saturated a t tlhe ordinary temperature m ith anhydrous soldium sulphate rather than with sodium chloride or calcium chloride.Extraction of the bromohydrin was acxom-plished vith three successive quantitiels of 250 C.C. of ether and after drying over anhydrous sodium sulphate the ether was expelled on the water-bath. Repelated distillation of the residual crude ethylenebromohydrin (180 grams) under atmospheric pressure demonstrated the impraa ticability of effeoting an adequate purification by suoh a method 1216 READ AND HOOK THE PREPARATION AND The results were however of intle8rest aa throwing some light upon the varying boiling points ra.nging from 1 4 7 O tlo 155O recorded for this substance by earlielr investiga,tors.On distilla.tioln the tem-pe'rature rose gra'dually to 145O and thel bulk of t,hel liquid distlilled bet8weeln 145O and 1 50° although distilla.tioln was incolmplete ewii atl 155O. The dist.illat'es welre pa,le yellow and t'he residue dark brown a.nd an appreciable1 a.mountl od hydrogeln bromide was pro-duaed during t,he distilla,tion. A f t8er setvera,l redistillatioas the purest mat'erial o1bt~aine.d in this. way distdlled beltwe,en 146O and 150° (n 1.4925; DY 1.7655 vacuum standard). On dilutiorn with wa,t,er to 60 per cent. strengt,h a slight opalescelnce appeared whilst a!i 50 per cent. a distinct second phase d l e d e . d and this persisted on further dilutdoln. ,The secolnd phase a t 35 per cent. had nEo 1.496 a,nd Wafs sholwn to contlain ethylene dibromide.The purest et,hylenelbrolmohydrin obtainable by t'his method thus contlains bolth et,hyle,nel dibromidei aad hydrogen brolmide as well as tlra,ces oif w ahr. By coin ducti ug t .h e d i s ti1 1 a-t ion ol f the crude e t 'hy 1 e,n eb r om olh y d ri n undelr diminished prelssure however these1 impurities are readily edimina,ted. After t'hel original reimoval of t'hhei et'her a lit& pure, sodium c:a#rbonat,e is a.dded if necessary to1 colrrect any acidity and on distilling under diminished pressure1 a sharp delimitatlon occurs bet4ween the portion of lolw boiling point and t'hs main fraction. I n three different prepa.rat,ions the la~t~ter boileld smoothly and con-st,aatdy at! 53'5O/14 mm. 54-'5O/14 mm.a'nd 48*5O/13 mm. respeo-tively and no1 cololur was delveloped during the distillation. The) pure substance is neutlral t801 litmus and gives no opalescence wheln diluted wit'h wa,tler olr silver nitra.te1 solut'ioln. As a rule one distil-lation under diminished preissure yields a product conf o'rming to these crit,eria but someitimee a se.cond distillation is necessa,ry. The average prima.ry yielld ot pure! ethylenebrolmohydrin obt,a.inablel from 450 gra,ms of bromine in t'he ma'nne'r 0ut~1ine.d above is aboatl 120 grams and furthe'r qua,nt,itiee may be elxt,ra.cted from t'he distillatto oif low boiling point and t'he original aqueous liquid. It may be mentioned t'hat the extlradJon with &her from the aqueous solu-tioln is rendered mom oolmplet,e by previously satarafing it wit.h sodium sulphate a,tl aboat.30° instea.d of ah the ordinary temperat-ture. Under such conditiolns a wcoad pha.sei consisting mainly of et,hylenebromohydrin separates frolm the solution concerned tba.t, is from an a,pproximat'ely 1.2N-solutdoin of ethyleinembromolhydrin cont'aining a,lso soldium bromide. This olbserva,tdon is in conf ormity with Golmberg's expelrieilzce with aqueous sohtio!ns of eShylenechloro. hydrin ( J . Amer. Chem. Soc. 1.919 41 1426). The volume of ethylene required in pra,ctice for this preparatioln amounts ac,coIrd CHARACTERISATION OF ETHYLENEBROMOHYDRIN. 121 7 ing to our apelrience; to be:tlween twice and thrice the oa.lculated quantity. Conditions affecting the Course of the Reactaon between Etltylene and Dilute Bromine Water.The efTect upon the course od t,he reaction utilised in the fore-going preparation of eert'ain factors which may be subjected to va;ria.tion ha.s been discussed in fo1rme.r colmmunicat,iosns (Zoc. cit.) ; supplemelnt8ary work of a qua,ntit,a.tdve nature has now been carried out in three! direations in order tot elstablish the influence upoln this process olf (1) concentmt'ion of the reaction-products (2) tempem-tme and (3) light(. (1) For the1 purpolse ot ascertaining accurately the amounts of ethylenebrolniohydrin pro'duced a t vasious st,ages of the rea.ct,ion, t'he delt,erminatJon from time t,o tJmel of hydroge,n bromide in an aliquolt partl of t'he reaction mixture1 proved to be satisfactolry. I n a control expe'rimeat bromine vapour was passed into 500 C.C.of u 4 e r undes the usual conditions the curreat of ethylenel being replaced by one of air; after an interval of seven hours during which al pale yehw tfintl had beein mainta.ined in the liquid the volume of AT/ 10-silver nitaat,e sollutJon required by 5 C.U. was less thaa one drop showing tha8t no1 me,a,surable decomposition of hypo-brolmous acid into hydr0ge.n brolmiclel and o,xygen had t'aken place under tlhe conditions adopted . 'In t'he succeeding prepara,t'ioas of eit'hylenebroimohydrin t'he cus-t'oma,ry proceldure wit9 observed elxcept t>hat t'he bromine was introc dumd in st'ages the a,mount used being det.elrmined in ela,ch ca,se by directl weighing. Atl the. e,nd of ea.ch st.age t.he liquid was stirred f o r ten minut'es aftle1r itl had become collo8urless and 5 C.C.oQ t\he liquid wa.s t8heln withdrawn and tltrated with st,a.nda,rd silver nitlrat4e sodutioln. Some1 of the1 resulta of t.wol inclelpendent setlies od elxperi-melnte are tabulafed beloiw. The cadculations tlake into account the chaages in vollume which occur throaghout the process but it was consi.dereld unnecessa,ry t,o apply a co1rre:ctioa f o,r the slight loss of brolmine by dif€usioin. The original volumeis of wa.t8er used were 530 C.C. and 550 U.C. re,spectively and in t,he secolnd selries a highe,r fina.1 coacent8ra.t'ion was a.ttained t,ha,n in the first) series. The two seats olf results a.re instruatIve as illustrating the order of the numeri-cal di'screpancies which may ocmr in determinations of this kind.It' should be added tha.t the t8emperature of t,he solutioln which was icemmled varied froom a.bout 7 O t o go and thab the expelriments were conduoted in diffused daylight 1218 Stage. 1 2 5 8 12 1 2 4 7 10 15 These READ AND HOOK THE PREPARATION AND Percent age Total of total bromine bromine Concentration of Total present as reacting ethylenebromohydrin. present +HBr C,H,Br*OH Grams per (grams). (grams). +HBr. litre. Normality. bromine C2H,Br'OH t o form -4.26 12.95 52-56 100.66 160.44 4-00 12.10 51-36 97-67 167.90 230.41 4.05 95.1 11-85 91.6 44.02 83.8 76.73 76-2 11 1.00 69-2 Series IZ. 5.9 0.048 17.5 0.140 65-0 0.520 112.4 0.899 161.8 2-294 3.77 94.3 5.3 11.09 91.7 15.8 43.26 84.2 61.7 75.06 76.9 106.6 11 1.29 70.5 156.9 148.3 1 64.4 207-6 0.043 0.126 0.494 0.853 1.255 1-661 da.h show that the relative velocity of reaction between bromine and ethylene and hypolbromous acid and ethylene respec-t i d y is displaaed to1 the advantage of the former relaction as the products accumulate a relsult which is probably connected with the great difference in solubility of ethylelne dibromide and ethylene bromolhydrin and the separation of the dibromide as a distinct phase.Nokwithstanding the change which has just been noted bhel molelcular ratio of bromohydrin to1 dibromide (which in the first stage was about 20 1) remained greatelr than unity througholut the range of concantrations invwtigated. By plotting conoentratiolns of ethylenebromohydrin in terms sf normality factor (2) against the correlsponding percentages (y) of bromine reacting ta form ethylenebromohydrin and hydrogen bromide a regular mrve is obtained (Fig.1) the major portion of which is alosely dsfineld by two linear equatlions namely, (i) y = 94-37 - 20.482 and (ii) y = 89-99 - 15.472. These equations embrace respectively the portions of the curve corresponding with (i) cr:=0*125 to 0.853 and (ii) 2=0'853 to 1.661. The approximate yield of ethylenelbrolmohydrin obt,ainable frolm a given amount of bromine over this range od colncentrations is thus capable of ready calculation. Attention was also directed to the rate of absolrption of the bromine in these eixperimente; this was found on the whole to increase with the concentration of the solution although irregulari-ties were1 noticed.Twenty-one successive polrtions of 8 grams o CHARACTERISATION OF ETHYLENEBROMOHYDRIN. 121 9 bromine took the following respective times expressed in hours for complete absorption when introduced with elthylene under the usual coditdons int'o a volume of water originally melamring 530 0.0. 1.9 1.6 1.6 1.1 1.1 1.1 1.2 1.0 0.9 1.1 1.1 1.1 1.3 0.8, 0.8 0.6 0.5 0.5 0'5 0-5 0.5. The gradual acceleration in .the ratel of absorption may perhaps be ahtributed to an enhanced solvent action of the brolmolhydrin solutions on the ethylene that is to a progressive increase of the molecular concentration of this reagent. I n othelr series of experiments t'he corresponding times were noticeably differelnt and i t seems possible that the acceleration or retardation was due to the presence of small amounts of foreign su bst ances.(2) I n order to1 telst the influence of temperature on the1 react3ion FIG. 1. Concentration of ethylenebrornohydrin (N). it wa.s conducted af 35O after a preliminary control experiment ha.d beleln ca,rried out atl the same temperatsure according tcm the prinoiple indicated above. A constant volume of 250 C.C. was maintalned throughout the expelriment water b&ng added to repla.cel the1 loss by evapomtion. The volume of N / 1O-silver nitrab solution required by 5 C.C. of tlhe liquid after the control experi-ment ha,d beeln continued f o r four ho,urs was 0.08 C.C. Thus at 3 5 O t h r e appeass t801 be a very slolw decomposition of hypobromous aaid intIo hydrogen bromide and oxygen and a corresponding corrdoln was introduced.Itl was also foiund nelceasa(ry when working at t'his tempera,ture to apply a correction for the loss of bromine by diffusion which amounteld to 1.20 grams in the control experiment. On passing bromine vapofur a,nd elthylener into 250 C.C. of wate 1220 READ AND HOOK THE PREPARATION AND under the conditions of the coatlroll experiment the! amount of bromine ut,ilised in the prepa,ra.tion od a 0*437N-solutdon of ethylenebrolmohydrin waa 28.2 gra,rns aft,er correcting for the loss by diffusion. Of this a,moluntt 62 pelr cent. rea.cted to form ethylenebrolmohydrin and hydrogen brolmide as compareid with 85-4 per cent,. reacting in this way for the same concentmtion in the ioe-cooled so~lutim.The time taken was 14.5 holurs the correl-sponding time at the lowes blrnpera,ture being 10.0 hours. It is themfore apparent thah by raising the telmpejratare t8he rea.ction takes place more slowly and tha8t at tlhel mrne time t,he molecular ratio of ethylenebromohydrin to ethylene dibro,mide is materially lessened. The effects just noltiaed were aocentua.ted at still higher t,empera-tures but olwing t,o the considerable loss od bromine by diffusion at tempelrafares much above 35O quantitative examinafion was rein-delrefd difficult in such cases. From further experiments to be described below the second effect is somewhat moire1 pronounced tha,n indicated by the above figures which take1 no account of the slow hydrolysis of ethylenebromohydrin occurring in warm aqueous solutbolns.The slower rate of re'acltion bet!weleJn et,hylene and brolmine: wa;telr in warm sollutions is probably to be attributed t'ol the diminish-ing sollubility of elthylene i n wa,telr as the tempeaat'ure rises. (3) I n the additioa olf hypochlolrolus slid hypolbromous acids t'oi unsa,tnrated 'subst'ances it ha.s usually belein assumeld t.ha,t t,he rea,o-tion is famured by coaduding t.he olpesation not only a.t a low ternpera,tnre but1 also in diffused light or prefe'rably in the dark. It was therefore of interest to investiga.te the influence of light upon the additive rea,ctioln beltween hypobrolmous acid and ethylene. I n the short series of experimsnts summarised below the icecoolled water (530 c.c.) was exposed to direct sunlight during the passags of t'he bromine vapour a.nd ethylene and the prolgress of the reae t i o a was fotlloweld in the ma.nner a h a d y desmibe'd.The result of a control experimeint indica.ted t,hat8 no a#pprecia,ble decolmpwitioa of hypobrolmous a.cid t,ook place undelr tlhe colnditioins laid down. Tohl bromine present Stage. (grams). 1 4-12 2 12-38 3 20.35 4 36.59 Total bromine present as C,H,Br-OH + HBr (grams). 3.81 10.97 17.80 30.75 Percentage of total Concen- bromine reacting to form C,H,Br'OH + HBr. bromohydrin In diffused ( N ) . In sunlight. light. 0.045 92.5 93.5 0- 130 88.6 91.7 0-2 12 87.5 90.0 0.368 84.0 86.8 tration of ethylene - 7-From t,he last two columns od t8he table it is seen tha.t the values obtlaineld in this eelriels of experimelntg for the molecular rafioi o OHAEAaTBRISATlON OR’ ETaYLlCNEBROlKOEYDE.1221 ethylenebromohydriin to ethylene dibromide are slightly lower than the values corresponding tvit,h the original conditions; the differ-en= howev~ are not marked. The time taken for the absorption wa,s 2.75 hours whkh is appreciably less than m y oorresponding period observed for an experiment mnduated in diffused light. On the whole therefore the rather reanarkable mndusion is reaahed that sunlight exerts an advantageous rathex than a detrimental influence upon the additive reaction betwean hypobromous acid md ethylene. Phyikal Properties of Pure Ethylenebromohpdrh. Ethylenebromohydrin is a collourlm mobile liquid miscible with watai- in all proportions.Demole’s statement ( 1 ~ . cit.) that it is only sparingly soluble in water indiaateg contamination with ethylene dibromide. The solubility in water dear- conaiderably in the presence of wrtain salts such as f o r example sodium chloride calcium ohloride and sodium sulphate. The a q u m solu-tions possess a sweet burning taste. The vappur of the pure sub-stance or of its aqueous solutions down to about 5 per cent; strength produoea a painful imtation of the eyes and nostrils shortly after inhalation. Ethylenebromohydrin diwlves readily in most of the common organic solvenh but forms a distinct phase when added to light petroleum. It distils unchanged under diminished pressure at the bmperatures noted above; when the pure mbstance is heatad under latmmpherio prwsure however it darkem 89 the temperature approach= the boiling point and decomposition ocmrs with the production of hydrogen bromide.Distillation commenoe8 at about 150° but the tempexature fluatusrtes and the boiling point is indefinite. The refractive index of several specimens was determined with the AbM refraotoaneter at 20° for the D line; the resulte were prautically identical the mean value being 1.4915. When the liquid was exposed for a few seconds on the p h this value sank in a typical instance to 1.479 ; this observation illustrates the pronounced hygrosoolpioity of ethylene bromohydrin. The observed mo1eoula.r r e f r d v e power at 200 wae 34.84 which stands in olme agreement with the d c u l a t d value 34.76.Determinations of specific gravity (reduced to a vaauum) were as follows : DS 1.7902 Di6 1.7696 Dfo 1.7629 I)? 1,7660 Q” 1.7494. The mean d c i e n t of dilatation between Oo and 30° oalaulated from the dak obtained in the above determinations is 0’00078. VOL. UXVTI. z 1222 READ AJXD HOOK THE PREPARATION AND The Belationship between the Refractive ZT&X and t h e Concentra-tion of Aqueous SolzLtions of Ethylenebrmohydrin. It has been shown by Irvine that measurements of refractive index provide a rapid and convenient method for the debrmination of the approximate strengths of aqueous solutions of ethylenechloro-hydrin; the method has also been adopted with useful reeults by Gomberg (Zoc. cit. p. 1418). For this reason and also in view of certain interesting thermal changes and related phenomena whioh ocou~ during the dilution of ethylenebromohydrin with water it was considered advisable to undertake a refraotometrio examination of a series of aque~us solutions of this substance.Soluticms of different concentrations were made by direct weighing of the con-stituenb and in each case the refraative index was determined with the Abb6 instrument at 2O0 for the D line. Some of the observa-tions are tabulated herewith: Percentage by weight of ethylene-bromohydrin. 100~000 96.340 90.032 80.007 72.201 60.686 53.620 nE* 1.4916 1.4801 1-4671 1.4464 1.4299 1-4096 1.3980 Percentage by weight of ethylene-bromohydrin. 49.179 40.372 30.999 22.441 18.810 10*010 3-361 0 (weter) n:.1.3916 1.3790 1.3667 1.3661 1.3517 1 *3422 1.3368 1.3330 When the whole of the data are representad graphically the resulting points lie on a regular uurve which d m not approximate to a straight line (Fig. 2). The character of the curve indicates that the refrautive index of aqueous solutions of ethylenebromohydrin ia always less than the value calculated from the admixture farmula, that is on mixing ethylenebromohydrin with w a k there is a am-traction in volume. It is also to be observed that the value of the refraotive index does not pass through a maximum as in the c888 of ethyl alcohol (Doroschevski and Dvorschantschik J . Russ. PhtyS. Chem. SOC. 1908 40 908). It is evident that the above curve furnishes a ready means of determining with close amuraq the concentrations of aqueous solutions of ethylenebromohydrin.The Distitlation of .d pueous Solutions of Ethylenebromohydh. From the gaieral characteristi= of ethylenebromohydrin it seemed probable thab its dilute aqueous solutions such as are obtained in the method of preparation outlined above might b CHARACTERISATION OF ETHYLENEBROMOHYDRIN. 1223 concentrated by fractional distillation. In order to test this point, solutioiIls of known strengt'h were submitted t s this operation the various fra,ctions being etxamined refraotome~trically. I n the case of a 5 per cant. solution a first fractioln containing 3 per cent. of ths t(otla1 bromohydrin had a concentration of 12 per cent. The concentrations of succeeding f ractions were lower ; when 20 per cent#.od t'he total bromohydrin had beleln collected the coaaentlra4tion of the next few drops of distillate was 9.5 per centl. and after the collleiotion of 50 per cent. the corresponding value was 7.0 per cent. FIG. 2. Percentage by weight of ethyzenebromohydrin. The boiling paint was originally 100.0°/766 mm. and it rose eventually to 1O0*lo. The boiling point1 of water at the same premure is 100*22*. With al 30 per cent. solution a first fraotion containing 11 per wnt. of the tots1 bromohydrin had a concentration of 34.0 per cent. and when tiwwocthirds of the total bromolhydrin had passed over the concentratlioln olf the next few drops of distillate was identiaal with that of the original solution that is 30 per cent. Meanwhile, the boiling point had risen from 99.0° to 99.4O; the barometric height was 768.5 mm.oorreeponding with a boiling point of 100.31O for water. 2 2 1224 READ AND HOOK THE PREPARATION AND Them observations suggested that ethylenebromohydrin forms a mixture of constant boiling pointl with water atq a concentration in the vicinity of 34 peir oeat. A similar distillatlion olf a 34 per cent. solutioln confirmed this conclusion the actual concentration oaf the constanbboliling mixture being 35.0 per cent. as is evidelnt from the fo'llolwing t'able : Fraction. 1 2 3 4 6 6 7 8 residue Boiling point. 99.1 99.1 99.1 99.1 99.2 99.2 99.2 99.5 -Weight of fraction (grams). 0.90 2.06 2-96 3.22 3.38 3.44 4-24 4.35 6.11 ,).&'LO 1-3718 1.3720 1.3717 1.3717 1.3716 1.3714 1.3708 1.3700 1.3700 I).Concentration Percentage per cent. of of total bromohydrin ,bromohydrin in fraction. in fraction. 35.1 3.0 35.2 6.9 35.0 9-43 35-0 10.7 34.9 11-2 34.8 11.3 34.3 13-7 33.7 13.9 33.7 19.6 The colmtlantl boiling point1 olf a 35.0 per cent. aqueous sollution of ethylenebromolhydrin is therefore 99*l0/ 762.4 mm. the drrwpond-ing boiling point ojf water baing 100*06°. The residue frolm the above distillation possessed a faint yellow colour and was distinctly acid ; titration with standad silver nitrate solutioa showed that it, mntlained 0.1491 gram of hydrogen bromide evidedy produced by partial hydrolysis oC tlhe bromohydrin during the distillatioa.The distillatw contained a mere trace of the acid and tlhus t!ha percant-age amolunt of the total bromolhydrin hydrollysed in the course of the operation was 2.2. Very interesting results were olbtlained by distilling more con-centrahed sollutiolns of the bromohydrin and a typical series of data is aooordingly summarised bellow f o r the case of a 74.9 per centl. wlutlion. The barolmetric height was 755 mm. corresponding with a boiling point od 99-82O folr water. Concentra-tion per Percentage Fraction. Boiling point. (grams). n:". hydrin. in fraction. 1 99.5O 1-0s 1.3798 41.0 3.0 2 100*0-101~0" 3.03 1.3838 43.9 6 1 3 101~0-102*8 3.70 1.3550 47.0 8-0 4 102*8-109*1 3.84 1.3982 53.9 9-6 5 109.1-137.0 3.81 1.4332 74.0 12.9 6 137.0-147.4 4.51 2.4840 97.0 20.0 7 247.4-149.0 5.26 1.4912 99.9 24.0 residue - 3.83 1.4905 99.7 17.5 Weight of cent.of of total fraction bromo bromohydrin When concentrakd aqueous solutdons of ethylmebrmohydrin a m distilleld the watles thus passes over in thel initia,l stages of the opem tioln and evsntlually almost pure bromohydrin distils. In th CHARACTERISATION OF ETHYLENEBROMORPDRIN. 1226 experiment desmibed it was ascertained by titration that 0.9 per cent. of the total bromohydrin present wa8s hydrolysed during the distillatioln. Of the hydrogen bromide formed it is interesting ta notl@ that 5 per cent. remained in the distlilling flask and 5 per mnt. was present in fraction 5 ; the bulk was found in fractiolns 6 (60 pm-cent.) and 7 (30 per cent.). As fractions 6 and 7 thus contained 1.7 and 0.7 per cent.of hydrogen bromide resps~t~ively the purity was distinctly lowejr than indicaked by the redractive index. From these investigatiolns it is seen that dilute aqueous solutions of ethylenebrolmohydrin might bo concentlrateld by frahonal distilla-tion but that in dealing with large quantitim of such solutiolns the loss by hydrolysis would be considerable. This objectionable feature could be partly eliminated by conduoting the distillation under diminished pressure. The Hydrolysis of Ethylenebromohydrin. The slight decomposition nohiaed during tlhe distillation of aqueous solutions of ethylelnelbrolmolhydrin seemed to indicate a greater susceptibility of this substance towards hydrolysing agents than is exhibited by etlhylenechlorolhydrin and accordingly further observations were made in ordelr to sekitle this point.In the first place when aqueolus solutions of ethylenebrmohydrin of wnmntrations ranging from 95 per cent. t o 3 per cent. were prepared and examined a t regular intelrvals no decomposition m l d be e&ablished until several weleks had elapsed. As a typical example a 50 per cent. solution after remaining for a month a t the ordinary temperature had developed a perceptible content of hydroc gen bromide corresponding with the hydrolysis of about 0.3 per cent. of the bromohydrin originally present; in the same period the refra,ctivei indelx had deolineid by 0.0007. With a 3 per c a t . solution the amount hydrolyseld in the same time approached 1 per cent. of the bromolhydrin and in general this very slow hydrolysis was most apparent in dilute solutions.At higher temperatlures the velocity of hydrolysis in aqueous solutioln was much accelerated ; on boiling a 6.25 per cent+. solutlion for thirty minutes under reflux, 19 per cent. of the bromohydrin was hydrollyseld and in ninety minutes 44 per cent. was hydrolyseld. Silver nitrate produced no immediate reactJon with cold aquemu solutions of the bromohydrin but in dilute sohtions an opdwnoe developed after fifteen or twenty minutes and quantitative estima-tiom made after longer intervals showed that the hydrolysis had undergone appreciable acceleration. Seaing that ethylenebromohydrin is prepared in acid solution it was of interest to ascertain the effeet of acid on the rate of hydro 1226 READ AND HOOK ETHYLBNEBROMOHYDRIN.lysis. A 1*8N-solutJon of hydrochloric acid containing 9.0 per cent. of ethyleinebromohydrin wits examined twenty-four hours after being prepared butl no pera&ptiblel hydrolysis hamd occurred in the mld. By boiling this solutioln folr thirty minut,es 51 pelr cent. of the1 brolmohydrin was hydrolpsed t>he prelsence of acid having there-fore accelerated the reaction. It was t,ol be1 a,nticipat'eld thah alkadi would exert ,a muoh more pronounced a8cceleratiolii tha,n acid a.nd in fa.& a 4.5 per mnt,. solut,ion od etlhylenelbromohydrin containing aBofut two equivale:nt proportlions of potassium hydroxide was hydrolyseld to the1 extent of 47 per centl. whe,n a,lTolwed to rema'in at the olrdinary tampera.ture for thirty minutes.By boiling a similar solution for thirty minutes 97 per cent. af the bramohydrin was hydralysed a.nd ninelty minutels' boIling broaght aboutl complelta hydrotlysis. Lastly an analysis olf a pure specimen of ethylenebrolmohydrin by alkalline hydrollysis may be quote'd 0.2935 gram after boliling for ninety minut,es with a.n excem of sodium hydroxide solution, required 23.5 C.C. of A T / 10-AgNO,. Br = 64.0. Cadc. Br = 64.0 per cent. Sunzmary . 1. Pure elthylenebromohydrin has been prepamred in .quantity by the action of codd dilute bromine water oln elthylene this method having been found preferable to1 a.ny other which has been described f o r the purpose. 2. I n t.he abwe reaction t'he mollecru1a.r ra.t,io of ethylenebrolmoL hydrin to ethylene dibromide detmeases colntlinuously with imreasing concentmtbn of ethyleaelbrolmohydrin a'nd hydrogen bromide in the solution aad a.lsoi with rise of telmpera.turei but ita is nolt influenced markedly by sunlight. The reaction is haatened by cooling or by exposure to1 sunlight. 3. The main physical chasacteristics of ethyledxomohydrin and also the relationship beltween the conc;entrat,ioln and re;fra,otive indelx of its aqueous solutions haw been est'ablished. 4. Ethylelnebrolmohydrin and wat4er form a mixtare olf mutant. boiling point (99.1°/762.4 mm.) a.t a conaelntra.tion of 35.0 per cent!. 5. Et,hylelnelbromohydrin is hydrollyseld mom reladily tha.n ethyle\nechlorohydrin. I n colld a.queolus solutions tqhe hydrolysis is percelptiblet alt'hmgh extremely slolw ; it' is greatly accelerat,ed by he,a,t also by the prersenm of acid a#nd notably olf alkali. DEPARTMENT OF ORGANIO CHEMISTRY, UNIVERSITY OF SYDNEY. Cognate inve&gations ,are in progress. [Received ~4ugust 30th 1920.