BURGESS AND LOWRY NEW HALOGEN DERIVATIVES ETC. 271 XLV.-New Halogen Derivatives of Camphor. Part V I . P-Bromo~mphor-a-sulphonic Acid. Part VII. The Constitution of the Reychler Series of Camphor-sulphonic Acids. Experiments on Chbrosulph-oxides. By HENRY BURGESS and THOMAS MARTIN LOWRY. Part VI.-p-Bromocamphor-a-sulphonic Acid. THE present paper describes a new series of sulphonic derivatives of camphor in which the sulphonic group occupies the a-position. This result is of interest since sulphonation has hitherto always been found to result in an attack upon a methyl group whilst the very reactive a-carbon atom has escaped. Thus even the gentle method of sulphonation used by Reychler gives p-derivatives and not the a-sulphonic acid which he thought he had prepared; and it is only after blocking the p-position with a halogen that we have been able to induce the sulphonic radical to enter the a-position.Armstrong and Lowry (J. 1902 81 1445) had already attempted to sulphonate p-bromocamphor but had obtained only a negative result. Our experiments have shown that although most of the p-bromocamphor can be recovered unchanged about 15 to 20% of it is sulphonated a t each operation. The product was at fist thought to be a p-bromocamphor-p-sulphonic acid where p repre-sents the position of the sulphonic group in Reychler’s acid but doubts arose when it was found that the sulphonamide did not yield an anhydramide like a-bromocamphor-p-sulphonamide but gave an acetyl derivative like a-bromocamphor-a-sulphonamide. Still more striking was the fact that the sulphonyl bromide whe 272 BURGESS AND LOWRY NEW HALOGEN DERIVATIVES decomposed by heat did not give a new pp-dibromocamphor but 5t mixture of Kachler and Spitzer’s aP-dibromocamphor with our new a’p-dibromocamphor (J.1923,123,1867). This suggested that the sulphonic group occupied the a-position; but as the whole research was based upon a suspicion that the pyrogenic decom-position of a p-sulphonyl bromide might give a p-bromo-compound of different orientation we did not regard it as safe t o accept the above observation as a final proof of the structure of the acid. Confirmation was however obtained when the sulphonamide on bromination again gave rise to the same aP-dibromocamphor by a remarkable action in which the halogen displaces a sulphonamide group instead of eliminating one of the reactive a-hydrogen atoms.This result was very difficult to explain unless the sulphonic group were already in the a-position. Moreover this broinination was free from the suspicions which attached t o the formation of Np-di-bromocamphor by the pyrogenic decomposition of the sulphonyl bromide. Final proof that the sulphonic group had entered the a-position leaving the camphor nucleus intact except for the presence of the p-bromine atom was obtained when the spon-taneous oxidation of a dibromosulphonamide derived from our acid gave rise t o the hitherto unknown p-bromocarnphorquinone, and thence by a direct further oxidation to p-bromocamphoric anhydride (not the free acid although the oxidation took place in aqueous solution).The whole series of actions may be set out as follows : p-Bromocamphor-a-sulphonic acid. p-Bromocamphor-a-sulphonamide. 15-Bromocamphor-a-sulphonyl bromide. Acetyl derivative. a’b-Dibromocamphor-a-sulphonamide. .1 c8H,3Br<xE PI. 1 0 t-p-Bromocamphoric anhydride. p-Bromocamphorquinone O F CAMPHOR. PL2RTS 1'1 ANT \TI. 273 In tliesc forinulx the acetyl derivative has been assumed to be enolic as in the case of the .x-isoniericle the solubility of which (unlike that of the amide from which it is derived) is not increased by the addition of alkali (Low? and Magson J. 1906 89 1044), so that the acetyl derii-atire appears not to contain an cc-hydrogen atom. I n tlic prc-ent case the fact that the acetyl group was lost on huminat icin si:ggestetl very strongly that we were again dealing with an cnolic.iicctate arid not the -SO,*NHAc compoiind. 'I'his invebt igation C C J I ~ ~ ~ ~ I I ~ S t lie lriew of Lipp and Lausberg ( A )inale?L 1024 436 274) that P-hrornocawiphor mid the Reychlet. .~erica of Crctrii~~~"".'"(~J'hoiil'c acids hare the sattie o?-ie?ztatioiL since the blocking of the $-pu~itioii I I J ? a halcgen coiiipels the sulphonic group to enter :i I I P ~ \ ~ imsitioii in the molecule. The question whether thc [d-substituenti are attarlied to a carbon atom in the ring or to a inethyl grcbiij) i5 (liwIqscd in Part IT1 below. E X P E R I 53 I3 N T A L. S II I y lz o 1 i a t io I L of fJ - Ur o i n oca n2 phot. . - p - B r o m o c a in pho r ( 5 0 g .) was added slowly to a mixturc of acetic anhydride (80 c.c.) and con-centrated sulphuric acid (25 c.c.) when the mixture became brown and most of the fJ-broinocamphor dissolved. After standing for a week a t 30-40" itl was heated for 2 hours and poured on to crushed ice and the precipitated p-bromocamphor (41 g.) washed with water. Further heating or varying the quantities of acetic. anhydride and sulphuric acid gave a smaller yield of sulphonate and a much larger quantity of brown charred material. The solution n-as boiled till free from acetic acid neutralised with a crc~ini of c;?lcium czrlionate filtered niicl ccncentrated. The sliidge of chalk z:cd calcium sulphate (which separates only slowly in the presence of the calcium sulphonste) was washed three times with boiling u-ater to remove all the sulphonate.After recrystallising three times from methylated spirit the calciuxu sulphonate ivas obtained in colourless rhombic plates. The mother-liquors Irere decoiorised by bone charcoal. The total yield was 12 g. (about 90';;)) after allowing for the recovery of iiiost of the bromocainphor. 0.1515 gave 0.1800 of CO, 0.0660 of H,O anti 0.0273 of CaXO,. 0.1773 lost 0.0172 of H,O and gave (1.0924 of R g h C == 3241 H = 4.87 Ca = 5.30 Br = 22-1 S H,O = 9.i0. (C,,H,,e)Br*S0,),Ca,4H2(~ requires C = 32-78, H = 4-95 Cia == 5 4 7 IZr = 21.83 H,O = 9.54%. Calcium p-broinocainphor- a-szrlphotzate is very soluble in hot water but crystallises readily from hot methylated spirit,. It does not melt below 230". It has [ s ~ ] ~ ~ ~ ~ = + 50° [a]5461 - + 59", = 4 116.5" in water ( 2 g.per 100 c.c.) 274 BURGESS AND LOWRY NEW HALOGEN DERIVATIVES Potassium /3-bromocamphor-cc-sulphonate prepared by addition of potassium oxalate (calc. amt.) to a solution of the calcium salt, filtration and evaporation crystallised from hot methyl alcohol in fine colourless needles. 0.2045 gave 0.2565 of CO and 0.0730 of H,O. 0.2152 lost no water a t 150-160" in a vacuum C = 34.22 I€ = 4-00 C,,H,,OBr~SO,K requires C = 34.37 H = 4-040,/,. P-Bromocamphor- a-sulphonyl Chloride (Formula as II).-A mix-ture of powdered phosphorus pentacliloride (5.4 g.) and the dry, powdered potassium salt (7-5 9.) was shaken and warmed in a dry flask on the water-bath until liquid. After 15 minutes it was shaken vigorously with powdered ice.The light brown solid was filtered and washed with water and a little dilute potassium carbonate solution ; it crystallised from ether in large compact crystals (Fig. l) which softened a t 93" and melted a t 97". 0.2393 FIQ.' 1. FIG. 2. gave 0.0940 of AgCl C1= 9.7. C,,H,,OBr*SO,Cl requires C1= 10.76%. p-Bromocamphor-a-sulphonyl chloride is comparatively stable when pure but loses chlorine quickly if impure and especially in presence of a little hydrochloric acid. It is very soluble in benzene or ether less soluble in cold alcohol and sparingly soluble in light petroleum. The crystallographic properties are as follows (two crystals were measured) : System Orthorhombic. Axial Ratios a b c = 0.2994 1 0.3769. Habit Both crystals were flattened parallel to C(OOl) with m(110) fairly large.The crystals were not quite transparent and did not give good reflections. Pmms present C(OOl) b(010) m(110) and n(Ol1). One face only of y(021) OF CAMPHOR. PARTS VI AND VII. 275 No. of Angles measure- Mean observed. ments. Limits. observation. Calculation. bnz 010 110 8 69" 52'-GS0 59' 69" 21' I cn 001:011 8 17 12-16 30 16 40 cy 001 :021 1 30 25;-31 35$ 31 04 30" 55' tnn 1 1 O O l l 4 84 28 -84 2 s.2 11 84 12 -Optical Characters Since thc crysials could not be immersed in liquids the optical data are scanty and liable to be inaccurate. The optic axial plane is parallel to b(010) with the acute bisectrix perpendicular to C(OOl) and the double refraction positive. The refractive indices determined by the minimum deviation method are very approximately a = 1.55 p = 1.56 y -1 1.58.~-Bromocamphor-a-szilphonamide (Formula ITI).-A mixture of the potassium sulphonate (7.0 9.) and phosphorus pentachloride (4.2 g.) was covered wi5h dry chloroform (50 c.c.) and left over-night. The chloroform was washed twice with ice-cold water, dried for an hour over calcium chloride filtered and excess of dry ammonia was passed into it when much heat was developed. After 12 hours the chloroform was distilled off and the residue was washed with water and crystallised from ether (yield 65%). 0.2045 gave 0.2895 of CO and 0.0955 of H,O. 0-2268 gave 0.1382 of AgBr C = 38.62 H = 5.23 Br = 25.93. C,oH,,O,NBrS requires C = 38.71 H = 5.20 Br = 25.77%. p-Bromocamphor- a-sulphonamide is very soluble in chloroform, alcohol ether or hot benzene less soluble in cold benzene and sparingly soluble in light petroleum or water.It melts a t 100-102" with the evolution of small bubbles of gas and has + 3 9 ~ 3 " ~ [a]5461 + 46.1" + 99" in alcohol (5.62 g. per 100 c.c.). The crystals which separate from an ethereal solution are tough and fibrous when crushed they are a t first quite transparent but after about 2 hours the surface becomes opaque and then the whole of the crystal perhaps as a result of polymorphic change. When this change is complete the crystal is easily crushed the fibrous structure having been lost completely. Attempts to Prepare an Anhydramide.-Lowry's two methods (J. 1902 81 1441) were employed. (a) A mixture of the sulphonamide (0.5 g.) with 45 C.C.of con-centrated hydrochloric acid was kept for a fortnight the clear solution diluted with water (2 vols.) and neutralised with potassium carbonate. ( b ) The sulphonamide (2.5 g.) was heated with acetic anhydride (10 c.c.) on the water-bath for 4 hours,* crystals separating. After * If the mixture is boiled instead of being heated on a water-bath con-Chloroform extracted the sulphonamide unchanged. siderable charring occurs 276 BURGESS AND LOWRY NEW HALOGEN DERIVATIVES cooling these were filtered off and a further small quantity was obtained by diluting the cold mother-liquor with water (yield 80:/,). Although this product resembled a-bromocamphor-p-sul-phonanhydramide in appearance and in its sparing solubility in all solvents it was an acetyl derivative for 0.2060 gave 0.3140 of CO and 0.0955 of H,O and 0.1900 gave 0.1033 of AgRr and 0.1280 of BaSO C = 41-55 H = 5-19 Br = 22.89 S = 9-25.C,,H,,O,NBrX requires C = 40.90 H = 5.15 Rr = 22.69 S = Acetyl-p-bromocana3-7hor-v.-.~ulll,honamide (IV) prepared as de-scribed above is slightly soluble in hot acetone and in hot acetic anhydride but is sparingly soluble in the cold solvents and in all the other common organic solvents. It crystallises from acetone in small compact crystals m. 13. 217" (rapid decomp.). It has [a]5,yo - 3 7 ~ 5 " ~ [CY]~,~~ - 42" in acetone (0.613 g. per 100 c.c.). 9.10 74. The crystallographic properties are as follows : System Orthorhombic. Axial Ratios a 3 c = 1.804 1 1-206. Forms present .- a(100) m(110) 1(210) q(101) d(201) ~ ( 0 1 1 ) .Habit .- The crystals were very small (Fig. 2) the longest dimen-The faces were frequently curved and not The face a(100) was only The prism faces were always very sion being about l". well suited to accurate measurement. observed on two crystals. small with the domes predominating to give a stumpy outline. No. of Angles measure -observed. ments. Limits. mm' = 110 i i o 5 57'29'- 59'51' 11' = 210:210 5 84 22- 86 16 ' = 011 :g11 6 99 58-102 24 ;z = 101 l o 1 G 67 54- 70 23 dd' = 201 201 G 73 0- 75 19 mq = 110 101 G 71 22- 73 49 mp = 110 :011 5 46 53- 48 31 Id = 210 201 10 53 61- 56 19 Ep = 210:Oll 10 56 34- 59 25 zq = 210 101 4 65 13- 65 41 md = 110 201 2 66 41- 68 44 Average.6s" 43' 85 32 101 0 68 50 74 9 73 13 47 52 54 18 57 43 65 26 67 24 Calculated. 58" 0' 84 6 100 40 67 32 73 34 74 22 47 41 53 30 58 58 65 37 67 0 CEeavuge None observed. Specific gracity .- 1.562 determined by floating in Thoulet's solution. Decomposition of p- Bromocamphor- a-sulphonyl Bromide (Formnla II).-The method of preparation was similar to that used in making the sulphonyl chloride. As the oil obtained on evaporating off the chloroform from the washed and dried solution did not crystallise in a desiccator in 2 days it was dissolved in dry xylene and refluxed for about + hour until the evolution of sulphur dioxide whic OB CAMPHOR. PARTS VI AND W. 277 was a t first very rapid had practically ceased. The liquid charred rather badly during the heating.After the xylene had been dis-tilled off in steam a colourless oil came over and slowly solidified. This was filtered off and melted a t 46-85". When crystallised from dilute alcohol it gave three successive crops of crystals. The first fraction which melted almost constantly a t 92-94' on three successive crystallisations was a t once converted into almost pure a@-dibromocamphor on crystallking in presence of a trace of piper-idine and therefore was a mixture of .@- and a'p-dibromocamphors (compare J. 1923 123 1872). The second fraction after one further crystallisation was identified as cr'p-dibromocamphor (m. p. 133-135" not lowered by mixing; [cr]5461 - 77" instead of - 85" in acetone 1 g. per 100 c.c.). The third fraction was identi-fied after one further crystallisation as crp-dibromocamphor (m.p. 113-114" not lowered by mixing with a$-dibromocamphor ; [a]5461 125" instead of 127" in acetone 3.92 g. per 100 c.c.). The above results proved quite conclusively that the sulphonyl bromide had decomposed to a mixture of M @ - and a'@-dibromocamphors. Broinination of p-Bromocamphor-a-su1phonamide.-Bromine (2.7 g.) was added to a solution of the sulphonamide (2.0 8.) in glacial acetic (50 c.c.) and the mixture refluxed for 4 hours becoming pale yellow. From the aqueous solution neutralised with sodium carbonate ether extracted an oil (about 2 c.c.) co-ntaining some lachrymatory substance probably bromoacetic acid from which was separated about 0.3 g. of ctp-dibrcmocamphor m.p. 112-114' (alone or mixed with this substance). a'@-Dibromocamphor-cr-sulphozamide (Formula V).-Bromine (2.5 g.) was added to the acetyl derivative of p-bromocamphor-cr-sul-phonamide (0.8 g.) in glacial acetic acid (20 c.c.) and the mixture refluxed for 5 hours the acetyl derivative dissolving completely, and the bromine disappearing. After cooling the acetic acid was neutralised with ammonia tho mixture being cooled meanwhile ; the white precipitate crystallised from benzene-ligroin in beautiful, colourless prisms m. p. 145" softening at 143" (yield 0.7 g. or 80%). 0.2015 gave 0.2310 of CO and 0.0710 of water. 0.1818 gave 0.1751 of AgBr and 0.1070 of BaSO C = 31.27 H = 3.94 Br = 40.99, S -.- 8.08. C,o€I150,NBr,S reruires C = 30.86 H = 3.89 Br = 41.08 S = 8.24%).cc'@-Dibroniocamphor-~-sulphonamide is insoluble in water spar-ingly soluble in light petroleum but very soluble in other organic solvents. It has - 26" [c(]j461 - 29" in benzene (2.64 g. per 100 c.c.). When warmed with aqueous sodium hydroxide it liberates ammonia. p .Bromocamphorquinone (Formula VI).-During a cr~~stallisatio 278 BURGESS AND L O ~ Y NEW HALOGEN DERIVATIVES of dibromocarnphorsulphonamide from dilute alcohol the solution slowly became yellow and deposited bright bellow crystals as well as the colourless ones of the initial compound. After a week a large proportion of the colourless crystals had disappeared whilst the yellow compound had increased in quantitly. The mixture was filtered off and sublimed in a vacuum when bright yellow prisms, m.p. 104-107" were obtained. These were recrystallised from dilute alcohol and melted at 132" after softening above 115". 0.04766 gave 0.03655 of AgBr (micro-Carius) Br = 32.63. CloH,,O,Br requires Br = 32.61%. Attempts to prepare it by hydrolysis with sodium ethoxide and with calcium hydroxide failed; both gave the intense colour of the quinone but very little could be separated. Oxidation with hydrogen peroxide in aqueous alcoholic solution first gave the intense yellow colour of the quinone; but further heating led to its decomposition with the formation of p-bromocamphoric an-hydride m. p. about 143" (alone or mixed with a genuine specimen) ; the rotatory power (+ 7") also was in harmony with that (+ 5") of a specimen of tlhe anhydride prepared from p-bromocamphoric acid.The yellow colour also disappears slowly by oxidat'ion when the quinone is left in contact with aqueous alcohol. P-Bromocamphorquinone is very soluble in benzene ether or chloroform less soluble in alcohol and insoluble in water. It h m a characteristic pleasant odour which is unlike that of camphor-quinone. The absorption spectrum in alcoholic solution of a specimen melting at 104-107" gave an absorption band a t the same wave-length as that of camphorquinone. The results are given in Table I. TABLE I. The A6sorption o j p-nromocamp,ho~quinone in Alcohol. Conc. of solution = 0-0993 g.-mol. per lit,re. A. 4300. 4400. 4500. 4550. 4600. 4650. 4700. 4750. Length of tube = 1 dcm. log €1 ...... 0.891 1.012 1.127 1.147 1.158 1.160 1.149 1.145 log € 2 ......1.15 1.32 1.41 1.458 1.460 1.462 1.460 1.457 A. 4800. 4850. 4900. 5000. 5100. 5200. 5300. log ...... 1,120 1.010 0.790 0.261 1.947 z.735 1.626 log c2 ...... 1.425 1.365 1.230 0.525 0.085 1.442 -c1 is the extinction coefficient for B-bromocsmphorquinone. e2 is the extinction coefficient for camphorquinone (Lowry and French, J. 1924 125 1921). We are indebted to Mr. R. Jeffery of Peterhouse for the crystallo-graphic measurements which he obtained under the direction of Mr. A. Hutchinson F.R.S OF CAMPHOR. PARTS VI AND VII. 279 Part V1I.-The Coizstitution of the Reychler Series of Camphor-sulphonic Acids. Experimeiats on Chlorosulphoxides. The position occupied by the sulplionic group in Iteychler’s camphorsulphoiiic acid and the related question of the position of the halogen in p-bromocamphor have recently come up for reconsideration siiice Wcdeliind Schenlr and Stusser (Ber.1823, 56 633) have discovered a series of reactions by which Reychler’s acid can be convcrted into lretopinic acid by the destruction of one of the methyl groups of the original molecule of camphor. The evidence thus supplied that the sulphonic radical has entered a methyl group is liowcver a t variance with the equally definite evidence that the halogen of P-bromocamphor (which can be pre-pared by the thermal decomposition of the sulphonyl bromide of Reychler’s acid) has entered a methyZc.lze group in the ring. Part of this evidence was reviewed by Armstrong and Lowry (J. 1902, 81 1469) who directed attention to thc oxidation of P-bromo-camphor to p-bromocamphoric acid but were unable to oxidise it further to a tricarboxylic acid and by M.0. Forster (J. 1902, 81,265) who directed attention to the ready conversion of p-bromo-camphor into campholenic acid (compare Part V J. 1924 125, 2376) but further results have since become available which point in the same direction. Thus not only is campholenonitrile (11) readily formed by the removal of water from camphoroxime (I), but a similar action takes place in the case of epicamphoroxime, where the methyl group has been transplanted bodily to the 4-position. It is therefore remarkable that this action does not occur in p-bromocamphoroxime if as is now suggested the p-bromine which inhibits the action is in the inactive methyl group, instead of in the methylene group from which the hydrogen of the eliminated molecule of water is taken.The two lines of evidence disagree so much that it appeared impossible to reconcile them unless (i) a migration of the bromine atom had occurred in the preparation of P-bromocamphor from the sulphonyl bromide of Reychler’s acid or (ii) some change of orientation had taken place during the conversion of the sulphonyl chloride of Reychler’s acid into ketopinic acid. The first hypo-thesis is negatived by the results recorded in Part VI above; the present part deals with the second hypothesis. As a first step in the conversion of Reychler’s camphorsulphony 280 BURGESS AND L O ~ Y NEW HALOGEN DERIVATIVES chloride to ketopinic acid Wedekind and his colleagues obtained by loss of water a chlorocamphorsulphoxide which they formulated as follows : CH2-qH-CH CH2-qH-CH, CH,-v--CO -Ha0 I w e I CH2*S02C1 c1c:s:o -+ CH2-$+-- CO I VMe2 I Reychler’s camphor- 1 0-Chlorocamphor-sulphonyl chloride.sulphoxide. Since they obtained a similar compound from camphor-r-sulphonyl chloride but not from aromatic sulphonylchlorides they concluded that chlorosulphoxides can be obtained only from acids in which a methyE group has been sulphonated. An alternative view which would explain the formation of 10-ketopinic acid from a 6-derivative of camphor postulates instead that a methyl group must be contiguous to the sulphonic radical thus : CH2-vH-CH CH,-qH-CH CH,-vH-CH, CH-C-CO -+ CII-C--CO _f CH,-C--CO I \ c1 I VMe I -H,O I VMe I I (we2 I I O=T=/O H,/CH I K,.,,--.., Camphor-6-sulphonyl #(Cl ‘i.-. . 3 o=v=c, H I o=s=cc1 _.. ‘-#’ Camphor-10-chloro-sulphoxide. chloride. ..-_2. This mechanism assumes that the dehydration of the sulphonyl-chloride involves the formation of an intermediate ring-compound, and the migration of a hydrogen atom from the 10 to the 6-position in addition to the migration of chlorine from sulphur to carbon which was postulated by Wedekind Schenk and Stusser. This mechanism appeared all tlhe more plausible because in camphor-n-sulphonyl chloride there is a methyl group in a precisely similar position relative to the sulphonic radical since both com-pounds contain the group I n the case of the r-compound > CCH, however it was possible to determine whether a change of orient-ation had occurred in the dehydration of the sulplionyl chloride, since a different methyl group would then be destroyed on con-verting into a carboxylic acid (i) the a-bromocamphor obtained by the pyrogenic decomposition of camphor-r-sulphonyl bromide (pre-sumably without change of orientation compare Part V I above) ; (ii) the 7r-chlorosulphoxide obtained by dehydration of the r-sul-phony1 chloride by Wedekind’s method.If this wandering took place therefore the tricstrboxylic acid obtained by oxidising Wedekind’s isoketopinic acid should be isomeric and not identica OF CAMPHOR. PARTS VI AND VII. 281 with the acid which Kipping and Pope obtained by hydrolysing and then osidising r-hromocamphor. Since Wedekiiid and his collaborators did not mention whether they had osidised their procluct to a tricarboxylic acid we decided to test the reaction by preparing and oxidising (instead of their compound) an E-bromocaiitphor-rr-chlorosulphoxide prepared from ammoiiium a-l.)romocanipl.or-7i-sulphonate.When this was oxidised with nitric acid a tricarboxylic acid and anhydride were obtained, which were identical and not isomeric with those prepared by Kipping and Pope. There was therefore no evidence here of a change of orientation. We also attempted t o prepare chlorotoluenesulphoxide from o-tolueiiesulphonyl chloride since this compound also contains a methyl group in the appropriate position; but me were again unsuccessful as most of the sulphonyl chloride was recovered unchanged.The e\-idence therefore undoubtedly proves that the suggested iiieehanisni which appeared to offer a feasible explanation of the production of 10-lietopinic acid from a camphor-6-suiphonjc acid is incorrect. We have then no alternative but to accept the conclusion that the halogen in 3-l~roiiiocaniphor 21s well as the sulphonic group in Rcychler's acid has entered a methyl group and occupies the 10-position in the camphor molecule. In doing so however it is desirable to point oat that there is no single reactiou in which P-brorno-cni,iphor hrhcrws as if it co,#tai?is the group -CH,Br. Thus it has ncver heeii coiivertctl into a primary alcohol -CH,Br -+ -CH,*O€€, and when attenipts are made to bring about this change (see Part V), the nioiecule emerges (as a derivative of cainpholenic acid) witch a1 1 its methyl groups intact hiit with a double bond in the ring.On the other liaiicl it is aimzing that a halogen in the side chain should s1)solutely inhiLit the formation of this double bond in the ring, when the attempt is made t o convert p-bromocamphoroxime into I~-bro~ocanlpliolc-iiitrile especially in view of the fact that the side chain can be eliiniiiatecl completely without affecting the action. The conclusion appears inevitable that there must be some liiik between the G- and 10-positions which is not indicated by the conventional formulz just as there must be some con-riexion between the ketonic and the gemdimethyl group (or some mechanism which involves both groups ; see Armstrong and Lowry, J.1902 81 1409) to account for the r-sulphonation of camphor. The nature of this connexion is not yet clear but it would be shirking the facts not to recognise that there is still a problem to solve even when the admission has been made that the p-series of compounds are all 10-derivatives of camphor 282 BURGESS AND LOWRY NEW HALOGEN DERIVATIVES ETC. E x P E R I M E N T A L . u-Bromocamp~~or-.rr-chlorosulphoxide C9H,,BrO*CC1:K0 was pre pared by the method of Wedekind Schenk and Stusser (Zoc. cit.) from a- bromocamphor-7r-sulphonyl chloride using dry pyridine as the dehydrating agent; but twice as much pyridine was used because the reaction mixture went solid before the addition of the sulphonyl chloride was complete and the later portions did not then react, The brownish-red granular solid was washed and crystallised three times from acetone.0.2003 gave 0.2830 of CO and 0.0700 of H,O. 0.1930 gave 0.2044 of AgCl 3- AgBr C = 38.55 H = 3-91 Br = 25.56 Cl = 11-34. C,,H,,O,CIBrS requires C = 38-63, H = 3.88 Br = 25.65 C1 = 11.38%. a-Bromocamphor-7r-chlorosulphoxide is very soluble in benzene, hot acetone or chloroform less soluble in cold acetone and the alcohols slightly soluble in hot ligroin or ether and insoluble in water. It crystallises in thin plates which are usually pointed a t one end but are occasionally hexagonal in shape. Its colour varies from pale to deep reddish-brown according t o the state of aggre-gation. It melts a t 158-159" and has + 31" [cx!J54sl + 39" in benzene (5-5 g. per 100 c.c.).Oxidution of ci-Brornocamphor-~-chlorosulphoxide.-This compound (3.5 9.) was refluxed with 20 C.C. of nitric acid (d 1.4) for 14 hours; the mixture was then cooled and water was added. After ex-traction with benzene to remove an oily by-product the aqueous layer was evaporated until crystals began to separate and then cooled. The crystals m. p. 180-185" after filtration and washing with a little water crystallised twice from benzene-.chloroform and once from ether gave a product which softened slightly a t 193" and melted a t 194-195" (decomp.); on mixture with a specimen of trans-camphotricarboxylic acid supplied by Prof. I?. S. Kipping P.R.S. its m.p. was undepressed. The substance had solu-bilities similar to those given by Kipping and Pope (J. 1896 69, 951) for that acid and gave precipitates with barium chloride, copper acetate ferric chloride and lead acetate as recorded by these authors.It has [aID + 34" [cc]~,~ + 35" and [cx]5461 + 40" in alcohol (0.281 g . in 15 c.c.). The anhydride prepared by heating with acetic anhydride diluting with water and extracting with ether separated from the ether in beautiful plates m. p. 252-254O (Pope and Kipping's trans-camphotricarboxylic acid melts a t 195-196" [decomp.] has [a]= + 37.2" in alcohol and gives an anhydride, m. p. 253-254"). The benzene extract gave a crystalline by-product m. p. about 202" (decomp.) [a]54al + 19.4" in acetone (2.166 g. per 100 c.c.) CONVERSION or AMINO-ACIDS INTO TERTIARY AMINO-ALCOHOLS. 283 This was expected to be a 7I-chlorodinitro-a-bromocamphor ; but, unlike the compound which Wedekind obtained under analogous conditions it was soluble in alkalis as if it contained a primary or secondary instead of only tertiary nitro-groups.Moreover it appeared to contain an additional molecule of water although there was not enough material available to confirm the analysis. Since the constitution of the compound is still unknown it is hoped to investigate it more fully later. Attempt to JlaEe ToZu eri e-o -chZorosuZyhoxide.-o-Toluenesulphonyl chloride (14 c.c.) was heated with dry piperidine (11 g.) for 16 hours, the liquid becoming of an intense dark red colour as in the case of a-bromocamphor-rr-sulphonyl chloride. After washing with acidified water extracting with ether and drying over calcium chloride the oil was distilled in a vacuum 85% of the o-toluenesulphonyl chloride being recovered unchanged.The charred residue was not investigated. UNIVERSITY CHEMICAL LABORATORY, C-111 ERID GE. [Received November 18th 1924. BURGESS AND LOWRY NEW HALOGEN DERIVATIVES ETC. 271 XLV.-New Halogen Derivatives of Camphor. Part V I . P-Bromo~mphor-a-sulphonic Acid. Part VII. The Constitution of the Reychler Series of Camphor-sulphonic Acids. Experiments on Chbrosulph-oxides. By HENRY BURGESS and THOMAS MARTIN LOWRY. Part VI.-p-Bromocamphor-a-sulphonic Acid. THE present paper describes a new series of sulphonic derivatives of camphor in which the sulphonic group occupies the a-position. This result is of interest since sulphonation has hitherto always been found to result in an attack upon a methyl group whilst the very reactive a-carbon atom has escaped.Thus even the gentle method of sulphonation used by Reychler gives p-derivatives and not the a-sulphonic acid which he thought he had prepared; and it is only after blocking the p-position with a halogen that we have been able to induce the sulphonic radical to enter the a-position. Armstrong and Lowry (J. 1902 81 1445) had already attempted to sulphonate p-bromocamphor but had obtained only a negative result. Our experiments have shown that although most of the p-bromocamphor can be recovered unchanged about 15 to 20% of it is sulphonated a t each operation. The product was at fist thought to be a p-bromocamphor-p-sulphonic acid where p repre-sents the position of the sulphonic group in Reychler’s acid but doubts arose when it was found that the sulphonamide did not yield an anhydramide like a-bromocamphor-p-sulphonamide but gave an acetyl derivative like a-bromocamphor-a-sulphonamide.Still more striking was the fact that the sulphonyl bromide whe 272 BURGESS AND LOWRY NEW HALOGEN DERIVATIVES decomposed by heat did not give a new pp-dibromocamphor but 5t mixture of Kachler and Spitzer’s aP-dibromocamphor with our new a’p-dibromocamphor (J. 1923,123,1867). This suggested that the sulphonic group occupied the a-position; but as the whole research was based upon a suspicion that the pyrogenic decom-position of a p-sulphonyl bromide might give a p-bromo-compound of different orientation we did not regard it as safe t o accept the above observation as a final proof of the structure of the acid.Confirmation was however obtained when the sulphonamide on bromination again gave rise to the same aP-dibromocamphor by a remarkable action in which the halogen displaces a sulphonamide group instead of eliminating one of the reactive a-hydrogen atoms. This result was very difficult to explain unless the sulphonic group were already in the a-position. Moreover this broinination was free from the suspicions which attached t o the formation of Np-di-bromocamphor by the pyrogenic decomposition of the sulphonyl bromide. Final proof that the sulphonic group had entered the a-position leaving the camphor nucleus intact except for the presence of the p-bromine atom was obtained when the spon-taneous oxidation of a dibromosulphonamide derived from our acid gave rise t o the hitherto unknown p-bromocarnphorquinone, and thence by a direct further oxidation to p-bromocamphoric anhydride (not the free acid although the oxidation took place in aqueous solution).The whole series of actions may be set out as follows : p-Bromocamphor-a-sulphonic acid. p-Bromocamphor-a-sulphonamide. 15-Bromocamphor-a-sulphonyl bromide. Acetyl derivative. a’b-Dibromocamphor-a-sulphonamide. .1 c8H,3Br<xE PI. 1 0 t-p-Bromocamphoric anhydride. p-Bromocamphorquinone O F CAMPHOR. PL2RTS 1'1 ANT \TI. 273 In tliesc forinulx the acetyl derivative has been assumed to be enolic as in the case of the .x-isoniericle the solubility of which (unlike that of the amide from which it is derived) is not increased by the addition of alkali (Low? and Magson J.1906 89 1044), so that the acetyl derii-atire appears not to contain an cc-hydrogen atom. I n tlic prc-ent case the fact that the acetyl group was lost on huminat icin si:ggestetl very strongly that we were again dealing with an cnolic. iicctate arid not the -SO,*NHAc compoiind. 'I'his invebt igation C C J I ~ ~ ~ ~ I I ~ S t lie lriew of Lipp and Lausberg ( A )inale?L 1024 436 274) that P-hrornocawiphor mid the Reychlet. .~erica of Crctrii~~~"".'"(~J'hoiil'c acids hare the sattie o?-ie?ztatioiL since the blocking of the $-pu~itioii I I J ? a halcgen coiiipels the sulphonic group to enter :i I I P ~ \ ~ imsitioii in the molecule. The question whether thc [d-substituenti are attarlied to a carbon atom in the ring or to a inethyl grcbiij) i5 (liwIqscd in Part IT1 below.E X P E R I 53 I3 N T A L. S II I y lz o 1 i a t io I L of fJ - Ur o i n oca n2 phot. . - p - B r o m o c a in pho r ( 5 0 g . ) was added slowly to a mixturc of acetic anhydride (80 c.c.) and con-centrated sulphuric acid (25 c.c.) when the mixture became brown and most of the fJ-broinocamphor dissolved. After standing for a week a t 30-40" itl was heated for 2 hours and poured on to crushed ice and the precipitated p-bromocamphor (41 g.) washed with water. Further heating or varying the quantities of acetic. anhydride and sulphuric acid gave a smaller yield of sulphonate and a much larger quantity of brown charred material. The solution n-as boiled till free from acetic acid neutralised with a crc~ini of c;?lcium czrlionate filtered niicl ccncentrated.The sliidge of chalk z:cd calcium sulphate (which separates only slowly in the presence of the calcium sulphonste) was washed three times with boiling u-ater to remove all the sulphonate. After recrystallising three times from methylated spirit the calciuxu sulphonate ivas obtained in colourless rhombic plates. The mother-liquors Irere decoiorised by bone charcoal. The total yield was 12 g. (about 90';;)) after allowing for the recovery of iiiost of the bromocainphor. 0.1515 gave 0.1800 of CO, 0.0660 of H,O anti 0.0273 of CaXO,. 0.1773 lost 0.0172 of H,O and gave (1.0924 of R g h C == 3241 H = 4.87 Ca = 5.30 Br = 22-1 S H,O = 9.i0. (C,,H,,e)Br*S0,),Ca,4H2(~ requires C = 32-78, H = 4-95 Cia == 5 4 7 IZr = 21.83 H,O = 9.54%.Calcium p-broinocainphor- a-szrlphotzate is very soluble in hot water but crystallises readily from hot methylated spirit,. It does not melt below 230". It has [ s ~ ] ~ ~ ~ ~ = + 50° [a]5461 - + 59", = 4 116.5" in water ( 2 g. per 100 c.c.) 274 BURGESS AND LOWRY NEW HALOGEN DERIVATIVES Potassium /3-bromocamphor-cc-sulphonate prepared by addition of potassium oxalate (calc. amt.) to a solution of the calcium salt, filtration and evaporation crystallised from hot methyl alcohol in fine colourless needles. 0.2045 gave 0.2565 of CO and 0.0730 of H,O. 0.2152 lost no water a t 150-160" in a vacuum C = 34.22 I€ = 4-00 C,,H,,OBr~SO,K requires C = 34.37 H = 4-040,/,. P-Bromocamphor- a-sulphonyl Chloride (Formula as II).-A mix-ture of powdered phosphorus pentacliloride (5.4 g.) and the dry, powdered potassium salt (7-5 9.) was shaken and warmed in a dry flask on the water-bath until liquid.After 15 minutes it was shaken vigorously with powdered ice. The light brown solid was filtered and washed with water and a little dilute potassium carbonate solution ; it crystallised from ether in large compact crystals (Fig. l) which softened a t 93" and melted a t 97". 0.2393 FIQ.' 1. FIG. 2. gave 0.0940 of AgCl C1= 9.7. C,,H,,OBr*SO,Cl requires C1= 10.76%. p-Bromocamphor-a-sulphonyl chloride is comparatively stable when pure but loses chlorine quickly if impure and especially in presence of a little hydrochloric acid. It is very soluble in benzene or ether less soluble in cold alcohol and sparingly soluble in light petroleum.The crystallographic properties are as follows (two crystals were measured) : System Orthorhombic. Axial Ratios a b c = 0.2994 1 0.3769. Habit Both crystals were flattened parallel to C(OOl) with m(110) fairly large. The crystals were not quite transparent and did not give good reflections. Pmms present C(OOl) b(010) m(110) and n(Ol1). One face only of y(021) OF CAMPHOR. PARTS VI AND VII. 275 No. of Angles measure- Mean observed. ments. Limits. observation. Calculation. bnz 010 110 8 69" 52'-GS0 59' 69" 21' I cn 001:011 8 17 12-16 30 16 40 cy 001 :021 1 30 25;-31 35$ 31 04 30" 55' tnn 1 1 O O l l 4 84 28 -84 2 s.2 11 84 12 -Optical Characters Since thc crysials could not be immersed in liquids the optical data are scanty and liable to be inaccurate.The optic axial plane is parallel to b(010) with the acute bisectrix perpendicular to C(OOl) and the double refraction positive. The refractive indices determined by the minimum deviation method are very approximately a = 1.55 p = 1.56 y -1 1.58. ~-Bromocamphor-a-szilphonamide (Formula ITI).-A mixture of the potassium sulphonate (7.0 9.) and phosphorus pentachloride (4.2 g.) was covered wi5h dry chloroform (50 c.c.) and left over-night. The chloroform was washed twice with ice-cold water, dried for an hour over calcium chloride filtered and excess of dry ammonia was passed into it when much heat was developed. After 12 hours the chloroform was distilled off and the residue was washed with water and crystallised from ether (yield 65%).0.2045 gave 0.2895 of CO and 0.0955 of H,O. 0-2268 gave 0.1382 of AgBr C = 38.62 H = 5.23 Br = 25.93. C,oH,,O,NBrS requires C = 38.71 H = 5.20 Br = 25.77%. p-Bromocamphor- a-sulphonamide is very soluble in chloroform, alcohol ether or hot benzene less soluble in cold benzene and sparingly soluble in light petroleum or water. It melts a t 100-102" with the evolution of small bubbles of gas and has + 3 9 ~ 3 " ~ [a]5461 + 46.1" + 99" in alcohol (5.62 g. per 100 c.c.). The crystals which separate from an ethereal solution are tough and fibrous when crushed they are a t first quite transparent but after about 2 hours the surface becomes opaque and then the whole of the crystal perhaps as a result of polymorphic change.When this change is complete the crystal is easily crushed the fibrous structure having been lost completely. Attempts to Prepare an Anhydramide.-Lowry's two methods (J. 1902 81 1441) were employed. (a) A mixture of the sulphonamide (0.5 g.) with 45 C.C. of con-centrated hydrochloric acid was kept for a fortnight the clear solution diluted with water (2 vols.) and neutralised with potassium carbonate. ( b ) The sulphonamide (2.5 g.) was heated with acetic anhydride (10 c.c.) on the water-bath for 4 hours,* crystals separating. After * If the mixture is boiled instead of being heated on a water-bath con-Chloroform extracted the sulphonamide unchanged. siderable charring occurs 276 BURGESS AND LOWRY NEW HALOGEN DERIVATIVES cooling these were filtered off and a further small quantity was obtained by diluting the cold mother-liquor with water (yield 80:/,).Although this product resembled a-bromocamphor-p-sul-phonanhydramide in appearance and in its sparing solubility in all solvents it was an acetyl derivative for 0.2060 gave 0.3140 of CO and 0.0955 of H,O and 0.1900 gave 0.1033 of AgRr and 0.1280 of BaSO C = 41-55 H = 5-19 Br = 22.89 S = 9-25. C,,H,,O,NBrX requires C = 40.90 H = 5.15 Rr = 22.69 S = Acetyl-p-bromocana3-7hor-v.-.~ulll,honamide (IV) prepared as de-scribed above is slightly soluble in hot acetone and in hot acetic anhydride but is sparingly soluble in the cold solvents and in all the other common organic solvents. It crystallises from acetone in small compact crystals m.13. 217" (rapid decomp.). It has [a]5,yo - 3 7 ~ 5 " ~ [CY]~,~~ - 42" in acetone (0.613 g. per 100 c.c.). 9.10 74. The crystallographic properties are as follows : System Orthorhombic. Axial Ratios a 3 c = 1.804 1 1-206. Forms present .- a(100) m(110) 1(210) q(101) d(201) ~ ( 0 1 1 ) . Habit .- The crystals were very small (Fig. 2) the longest dimen-The faces were frequently curved and not The face a(100) was only The prism faces were always very sion being about l". well suited to accurate measurement. observed on two crystals. small with the domes predominating to give a stumpy outline. No. of Angles measure -observed. ments. Limits. mm' = 110 i i o 5 57'29'- 59'51' 11' = 210:210 5 84 22- 86 16 ' = 011 :g11 6 99 58-102 24 ;z = 101 l o 1 G 67 54- 70 23 dd' = 201 201 G 73 0- 75 19 mq = 110 101 G 71 22- 73 49 mp = 110 :011 5 46 53- 48 31 Id = 210 201 10 53 61- 56 19 Ep = 210:Oll 10 56 34- 59 25 zq = 210 101 4 65 13- 65 41 md = 110 201 2 66 41- 68 44 Average.6s" 43' 85 32 101 0 68 50 74 9 73 13 47 52 54 18 57 43 65 26 67 24 Calculated. 58" 0' 84 6 100 40 67 32 73 34 74 22 47 41 53 30 58 58 65 37 67 0 CEeavuge None observed. Specific gracity .- 1.562 determined by floating in Thoulet's solution. Decomposition of p- Bromocamphor- a-sulphonyl Bromide (Formnla II).-The method of preparation was similar to that used in making the sulphonyl chloride. As the oil obtained on evaporating off the chloroform from the washed and dried solution did not crystallise in a desiccator in 2 days it was dissolved in dry xylene and refluxed for about + hour until the evolution of sulphur dioxide whic OB CAMPHOR.PARTS VI AND W. 277 was a t first very rapid had practically ceased. The liquid charred rather badly during the heating. After the xylene had been dis-tilled off in steam a colourless oil came over and slowly solidified. This was filtered off and melted a t 46-85". When crystallised from dilute alcohol it gave three successive crops of crystals. The first fraction which melted almost constantly a t 92-94' on three successive crystallisations was a t once converted into almost pure a@-dibromocamphor on crystallking in presence of a trace of piper-idine and therefore was a mixture of .@- and a'p-dibromocamphors (compare J.1923 123 1872). The second fraction after one further crystallisation was identified as cr'p-dibromocamphor (m. p. 133-135" not lowered by mixing; [cr]5461 - 77" instead of - 85" in acetone 1 g. per 100 c.c.). The third fraction was identi-fied after one further crystallisation as crp-dibromocamphor (m. p. 113-114" not lowered by mixing with a$-dibromocamphor ; [a]5461 125" instead of 127" in acetone 3.92 g. per 100 c.c.). The above results proved quite conclusively that the sulphonyl bromide had decomposed to a mixture of M @ - and a'@-dibromocamphors. Broinination of p-Bromocamphor-a-su1phonamide.-Bromine (2.7 g.) was added to a solution of the sulphonamide (2.0 8.) in glacial acetic (50 c.c.) and the mixture refluxed for 4 hours becoming pale yellow.From the aqueous solution neutralised with sodium carbonate ether extracted an oil (about 2 c.c.) co-ntaining some lachrymatory substance probably bromoacetic acid from which was separated about 0.3 g. of ctp-dibrcmocamphor m. p. 112-114' (alone or mixed with this substance). a'@-Dibromocamphor-cr-sulphozamide (Formula V).-Bromine (2.5 g.) was added to the acetyl derivative of p-bromocamphor-cr-sul-phonamide (0.8 g.) in glacial acetic acid (20 c.c.) and the mixture refluxed for 5 hours the acetyl derivative dissolving completely, and the bromine disappearing. After cooling the acetic acid was neutralised with ammonia tho mixture being cooled meanwhile ; the white precipitate crystallised from benzene-ligroin in beautiful, colourless prisms m.p. 145" softening at 143" (yield 0.7 g. or 80%). 0.2015 gave 0.2310 of CO and 0.0710 of water. 0.1818 gave 0.1751 of AgBr and 0.1070 of BaSO C = 31.27 H = 3.94 Br = 40.99, S -.- 8.08. C,o€I150,NBr,S reruires C = 30.86 H = 3.89 Br = 41.08 S = 8.24%). cc'@-Dibroniocamphor-~-sulphonamide is insoluble in water spar-ingly soluble in light petroleum but very soluble in other organic solvents. It has - 26" [c(]j461 - 29" in benzene (2.64 g. per 100 c.c.). When warmed with aqueous sodium hydroxide it liberates ammonia. p .Bromocamphorquinone (Formula VI).-During a cr~~stallisatio 278 BURGESS AND L O ~ Y NEW HALOGEN DERIVATIVES of dibromocarnphorsulphonamide from dilute alcohol the solution slowly became yellow and deposited bright bellow crystals as well as the colourless ones of the initial compound.After a week a large proportion of the colourless crystals had disappeared whilst the yellow compound had increased in quantitly. The mixture was filtered off and sublimed in a vacuum when bright yellow prisms, m. p. 104-107" were obtained. These were recrystallised from dilute alcohol and melted at 132" after softening above 115". 0.04766 gave 0.03655 of AgBr (micro-Carius) Br = 32.63. CloH,,O,Br requires Br = 32.61%. Attempts to prepare it by hydrolysis with sodium ethoxide and with calcium hydroxide failed; both gave the intense colour of the quinone but very little could be separated. Oxidation with hydrogen peroxide in aqueous alcoholic solution first gave the intense yellow colour of the quinone; but further heating led to its decomposition with the formation of p-bromocamphoric an-hydride m.p. about 143" (alone or mixed with a genuine specimen) ; the rotatory power (+ 7") also was in harmony with that (+ 5") of a specimen of tlhe anhydride prepared from p-bromocamphoric acid. The yellow colour also disappears slowly by oxidat'ion when the quinone is left in contact with aqueous alcohol. P-Bromocamphorquinone is very soluble in benzene ether or chloroform less soluble in alcohol and insoluble in water. It h m a characteristic pleasant odour which is unlike that of camphor-quinone. The absorption spectrum in alcoholic solution of a specimen melting at 104-107" gave an absorption band a t the same wave-length as that of camphorquinone. The results are given in Table I.TABLE I. The A6sorption o j p-nromocamp,ho~quinone in Alcohol. Conc. of solution = 0-0993 g.-mol. per lit,re. A. 4300. 4400. 4500. 4550. 4600. 4650. 4700. 4750. Length of tube = 1 dcm. log €1 ...... 0.891 1.012 1.127 1.147 1.158 1.160 1.149 1.145 log € 2 ...... 1.15 1.32 1.41 1.458 1.460 1.462 1.460 1.457 A. 4800. 4850. 4900. 5000. 5100. 5200. 5300. log ...... 1,120 1.010 0.790 0.261 1.947 z.735 1.626 log c2 ...... 1.425 1.365 1.230 0.525 0.085 1.442 -c1 is the extinction coefficient for B-bromocsmphorquinone. e2 is the extinction coefficient for camphorquinone (Lowry and French, J. 1924 125 1921). We are indebted to Mr. R. Jeffery of Peterhouse for the crystallo-graphic measurements which he obtained under the direction of Mr. A.Hutchinson F.R.S OF CAMPHOR. PARTS VI AND VII. 279 Part V1I.-The Coizstitution of the Reychler Series of Camphor-sulphonic Acids. Experimeiats on Chlorosulphoxides. The position occupied by the sulplionic group in Iteychler’s camphorsulphoiiic acid and the related question of the position of the halogen in p-bromocamphor have recently come up for reconsideration siiice Wcdeliind Schenlr and Stusser (Ber. 1823, 56 633) have discovered a series of reactions by which Reychler’s acid can be convcrted into lretopinic acid by the destruction of one of the methyl groups of the original molecule of camphor. The evidence thus supplied that the sulphonic radical has entered a methyl group is liowcver a t variance with the equally definite evidence that the halogen of P-bromocamphor (which can be pre-pared by the thermal decomposition of the sulphonyl bromide of Reychler’s acid) has entered a methyZc.lze group in the ring.Part of this evidence was reviewed by Armstrong and Lowry (J. 1902, 81 1469) who directed attention to thc oxidation of P-bromo-camphor to p-bromocamphoric acid but were unable to oxidise it further to a tricarboxylic acid and by M. 0. Forster (J. 1902, 81,265) who directed attention to the ready conversion of p-bromo-camphor into campholenic acid (compare Part V J. 1924 125, 2376) but further results have since become available which point in the same direction. Thus not only is campholenonitrile (11) readily formed by the removal of water from camphoroxime (I), but a similar action takes place in the case of epicamphoroxime, where the methyl group has been transplanted bodily to the 4-position.It is therefore remarkable that this action does not occur in p-bromocamphoroxime if as is now suggested the p-bromine which inhibits the action is in the inactive methyl group, instead of in the methylene group from which the hydrogen of the eliminated molecule of water is taken. The two lines of evidence disagree so much that it appeared impossible to reconcile them unless (i) a migration of the bromine atom had occurred in the preparation of P-bromocamphor from the sulphonyl bromide of Reychler’s acid or (ii) some change of orientation had taken place during the conversion of the sulphonyl chloride of Reychler’s acid into ketopinic acid. The first hypo-thesis is negatived by the results recorded in Part VI above; the present part deals with the second hypothesis.As a first step in the conversion of Reychler’s camphorsulphony 280 BURGESS AND L O ~ Y NEW HALOGEN DERIVATIVES chloride to ketopinic acid Wedekind and his colleagues obtained by loss of water a chlorocamphorsulphoxide which they formulated as follows : CH2-qH-CH CH2-qH-CH, CH,-v--CO -Ha0 I w e I CH2*S02C1 c1c:s:o -+ CH2-$+-- CO I VMe2 I Reychler’s camphor- 1 0-Chlorocamphor-sulphonyl chloride. sulphoxide. Since they obtained a similar compound from camphor-r-sulphonyl chloride but not from aromatic sulphonylchlorides they concluded that chlorosulphoxides can be obtained only from acids in which a methyE group has been sulphonated. An alternative view which would explain the formation of 10-ketopinic acid from a 6-derivative of camphor postulates instead that a methyl group must be contiguous to the sulphonic radical thus : CH2-vH-CH CH,-qH-CH CH,-vH-CH, CH-C-CO -+ CII-C--CO _f CH,-C--CO I \ c1 I VMe I -H,O I VMe I I (we2 I I O=T=/O H,/CH I K,.,,--.., Camphor-6-sulphonyl #(Cl ‘i.-. . 3 o=v=c, H I o=s=cc1 _.. ‘-#’ Camphor-10-chloro-sulphoxide. chloride. ..-_2. This mechanism assumes that the dehydration of the sulphonyl-chloride involves the formation of an intermediate ring-compound, and the migration of a hydrogen atom from the 10 to the 6-position in addition to the migration of chlorine from sulphur to carbon which was postulated by Wedekind Schenk and Stusser. This mechanism appeared all tlhe more plausible because in camphor-n-sulphonyl chloride there is a methyl group in a precisely similar position relative to the sulphonic radical since both com-pounds contain the group I n the case of the r-compound > CCH, however it was possible to determine whether a change of orient-ation had occurred in the dehydration of the sulplionyl chloride, since a different methyl group would then be destroyed on con-verting into a carboxylic acid (i) the a-bromocamphor obtained by the pyrogenic decomposition of camphor-r-sulphonyl bromide (pre-sumably without change of orientation compare Part V I above) ; (ii) the 7r-chlorosulphoxide obtained by dehydration of the r-sul-phony1 chloride by Wedekind’s method.If this wandering took place therefore the tricstrboxylic acid obtained by oxidising Wedekind’s isoketopinic acid should be isomeric and not identica OF CAMPHOR.PARTS VI AND VII. 281 with the acid which Kipping and Pope obtained by hydrolysing and then osidising r-hromocamphor. Since Wedekiiid and his collaborators did not mention whether they had osidised their procluct to a tricarboxylic acid we decided to test the reaction by preparing and oxidising (instead of their compound) an E-bromocaiitphor-rr-chlorosulphoxide prepared from ammoiiium a-l.)romocanipl.or-7i-sulphonate. When this was oxidised with nitric acid a tricarboxylic acid and anhydride were obtained, which were identical and not isomeric with those prepared by Kipping and Pope. There was therefore no evidence here of a change of orientation.We also attempted t o prepare chlorotoluenesulphoxide from o-tolueiiesulphonyl chloride since this compound also contains a methyl group in the appropriate position; but me were again unsuccessful as most of the sulphonyl chloride was recovered unchanged. The e\-idence therefore undoubtedly proves that the suggested iiieehanisni which appeared to offer a feasible explanation of the production of 10-lietopinic acid from a camphor-6-suiphonjc acid is incorrect. We have then no alternative but to accept the conclusion that the halogen in 3-l~roiiiocaniphor 21s well as the sulphonic group in Rcychler's acid has entered a methyl group and occupies the 10-position in the camphor molecule. In doing so however it is desirable to point oat that there is no single reactiou in which P-brorno-cni,iphor hrhcrws as if it co,#tai?is the group -CH,Br.Thus it has ncver heeii coiivertctl into a primary alcohol -CH,Br -+ -CH,*O€€, and when attenipts are made to bring about this change (see Part V), the nioiecule emerges (as a derivative of cainpholenic acid) witch a1 1 its methyl groups intact hiit with a double bond in the ring. On the other liaiicl it is aimzing that a halogen in the side chain should s1)solutely inhiLit the formation of this double bond in the ring, when the attempt is made t o convert p-bromocamphoroxime into I~-bro~ocanlpliolc-iiitrile especially in view of the fact that the side chain can be eliiniiiatecl completely without affecting the action. The conclusion appears inevitable that there must be some liiik between the G- and 10-positions which is not indicated by the conventional formulz just as there must be some con-riexion between the ketonic and the gemdimethyl group (or some mechanism which involves both groups ; see Armstrong and Lowry, J.1902 81 1409) to account for the r-sulphonation of camphor. The nature of this connexion is not yet clear but it would be shirking the facts not to recognise that there is still a problem to solve even when the admission has been made that the p-series of compounds are all 10-derivatives of camphor 282 BURGESS AND LOWRY NEW HALOGEN DERIVATIVES ETC. E x P E R I M E N T A L . u-Bromocamp~~or-.rr-chlorosulphoxide C9H,,BrO*CC1:K0 was pre pared by the method of Wedekind Schenk and Stusser (Zoc. cit.) from a- bromocamphor-7r-sulphonyl chloride using dry pyridine as the dehydrating agent; but twice as much pyridine was used because the reaction mixture went solid before the addition of the sulphonyl chloride was complete and the later portions did not then react, The brownish-red granular solid was washed and crystallised three times from acetone.0.2003 gave 0.2830 of CO and 0.0700 of H,O. 0.1930 gave 0.2044 of AgCl 3- AgBr C = 38.55 H = 3-91 Br = 25.56 Cl = 11-34. C,,H,,O,CIBrS requires C = 38-63, H = 3.88 Br = 25.65 C1 = 11.38%. a-Bromocamphor-7r-chlorosulphoxide is very soluble in benzene, hot acetone or chloroform less soluble in cold acetone and the alcohols slightly soluble in hot ligroin or ether and insoluble in water. It crystallises in thin plates which are usually pointed a t one end but are occasionally hexagonal in shape.Its colour varies from pale to deep reddish-brown according t o the state of aggre-gation. It melts a t 158-159" and has + 31" [cx!J54sl + 39" in benzene (5-5 g. per 100 c.c.). Oxidution of ci-Brornocamphor-~-chlorosulphoxide.-This compound (3.5 9.) was refluxed with 20 C.C. of nitric acid (d 1.4) for 14 hours; the mixture was then cooled and water was added. After ex-traction with benzene to remove an oily by-product the aqueous layer was evaporated until crystals began to separate and then cooled. The crystals m. p. 180-185" after filtration and washing with a little water crystallised twice from benzene-.chloroform and once from ether gave a product which softened slightly a t 193" and melted a t 194-195" (decomp.); on mixture with a specimen of trans-camphotricarboxylic acid supplied by Prof.I?. S. Kipping P.R.S. its m.p. was undepressed. The substance had solu-bilities similar to those given by Kipping and Pope (J. 1896 69, 951) for that acid and gave precipitates with barium chloride, copper acetate ferric chloride and lead acetate as recorded by these authors. It has [aID + 34" [cc]~,~ + 35" and [cx]5461 + 40" in alcohol (0.281 g . in 15 c.c.). The anhydride prepared by heating with acetic anhydride diluting with water and extracting with ether separated from the ether in beautiful plates m. p. 252-254O (Pope and Kipping's trans-camphotricarboxylic acid melts a t 195-196" [decomp.] has [a]= + 37.2" in alcohol and gives an anhydride, m. p. 253-254"). The benzene extract gave a crystalline by-product m. p. about 202" (decomp.) [a]54al + 19.4" in acetone (2.166 g. per 100 c.c.) CONVERSION or AMINO-ACIDS INTO TERTIARY AMINO-ALCOHOLS. 283 This was expected to be a 7I-chlorodinitro-a-bromocamphor ; but, unlike the compound which Wedekind obtained under analogous conditions it was soluble in alkalis as if it contained a primary or secondary instead of only tertiary nitro-groups. Moreover it appeared to contain an additional molecule of water although there was not enough material available to confirm the analysis. Since the constitution of the compound is still unknown it is hoped to investigate it more fully later. Attempt to JlaEe ToZu eri e-o -chZorosuZyhoxide.-o-Toluenesulphonyl chloride (14 c.c.) was heated with dry piperidine (11 g.) for 16 hours, the liquid becoming of an intense dark red colour as in the case of a-bromocamphor-rr-sulphonyl chloride. After washing with acidified water extracting with ether and drying over calcium chloride the oil was distilled in a vacuum 85% of the o-toluenesulphonyl chloride being recovered unchanged. The charred residue was not investigated. UNIVERSITY CHEMICAL LABORATORY, C-111 ERID GE. [Received November 18th 1924.