214 SHOESMITH AND SLATER PREPARATION HYDROLYSIS XXXII.-Preparation Hydrolysis and Reduction of the Fluoro- Chloro- and Bromo-benxyl Bromides. SHOESMITH By JOHN BALDWIN and ROBERTHENRYSLATER. INprevious communications (J.,1922,121 1392 ; 1923,123,2828 ; 1924,125,1312,2278)an account was given of the manner in which the reactivity of halogen atoms in various halogenated benzenoid derivatives is influenced by oxygen and by the hydrogen of a methyl group. The investigations have been continued with a view to discovering the effect of fluorine chlorine and bromine in such compounds. From the point of view of the principle of induced alternate polarities the halogens cxcept fluorine appear to act as weakly negative " key-atoms " (as pointed out by Lapworth Hem.Munchester Phil. Soc. 1920 64,No. 3). The present investig- ation has shown that (1) fluorine is capable of inducing differences of reactivity very similar to but smaller than those met with in the cases of the methoxybenzyl bromides (J.,1922 121 1392) and the o-bromoxylenes (J.,1924 125 2278) and (2) chlorine and bromine induce still smaller differences and the chloro- and bromo-benzyl bromides provide the first examples encountered in this series in AND REDUCTION OF FLUOROBENZPL BROMIDE ETC. 215 which a change of reagent does not cause a change in the order of reactivity. The order of ease with which the bromides lose their bromine as bromidion in solution in aqueous alcohol is (1) p-fluorobenzyl bromide>benzyl bromide >o-fluoro->m-fluoro-; (2) benzyl bromide >p-chloro- >o-chloro- >m-chloro- ; (3) benzyl bromide >p-bromo- >o-bromo- >m-bromo-.The order of ease of reduction by hydriodic acid is (1) o-fluoro->nz-fluoro->benzyl bromide >p-fluoro- ; (2) o-chloro->p-chloro->m-chloro->benzyl bromide ; and (3) o-bromo- >p-bromo- >m-bromo- >benzyl bromide. The differences of ease of hydrolysis are quite marked. Those of ease of reduction are small but several repetitions of the experiments led to the same order of reactivity and as is seen from the above the order of ease of reduction of the chloro- and bromo-compounds is not analogous to that discovered in the cases of the methoxybenzyl bromides the o-bromoxylenes and the fluorobenzyl bromides.A series of experiments was undertaken to see if this non-reversal of the order of reactivity is due to any of the following causes (a) removal of halogen from the nucleus (b) liberation of iodine as follows (1) R*C,H,*CH,Br + HI +R*C,H,*CR,I + HBr. (2) 2R-C6H,-CH,I -+R*C,H,*CH,*CH,*C,H,R (compare Silberrad J. 1924 125 2196) and (c) the more rapid formation of the iodide in the case of the para-isomerides and subsequent reduction of the iodides by hydriodic acid. The bromides were reduced for a long time but the slight elimination of nuclear halogen then noted was not sufficient to reverse the discovered order of reactivity when correction was made for it. Reaction (2) is a photochemical one and does not take place in a dark thermostat such as waa used in these experiments.In order to test (c) the bromobenzyl iodides were prepared; they were reduced by hydriodic acid at the same rate as the corresponding bromides. Thus it appears as if the conversion of the bromide into iodide is due to a mass-action effect and owing to the concentration of hydriodic acid used goes almost equally rapidly in all cases. Some of the iodides lose iodine more rapidly than their isomerides and when a change in order of re-activity with change of reagent takes place alternating polar influences are very strong e.g. in the methoxybenzyl bromides. An approximate estimate of the magnitude of the three influences which affect the reactivity of the bromine atom in the bromides namely the general (represented by g) the alternating (represented by a),and the so-called steric influence (represented by B) may be arrived at from considerations similar to those put forward by Blurscheim (J.,1909 95 726).A comparison by means of reaction velocities which is the most satisfactory method is impossible 216 SHOESMITH AND SLATER PREPARATION HYDROLYSIS because only during the hydrolysis of 0-and nz-fluorobenzyl bromides and also of the isomeric nitrobenzyl bromides (Shoesmith and Hetherington J. 1924 125,1316) were the reactions slow enough to give monomolecular velocity coefficients. The reduction of the bromides moreover is not the simple bimolecular reaction investigated by West in the case of bromomalonyl compounds (J. 1924 125,710).The reciprocals of the times taken for half- completion of the reaction in the various cases may be used to obtain the necessary comparison. If KO, K, Kp and Kurepresent the reciprocals of the times taken for half-completion of the reactions for the ortho- meta- para- and unsubstituted compounds respectively the manner in which the velocities of the reactions are affected by the three influences mentioned above may be expressed by the following equations * log, KO = log, K + g + a + 8. log, Km = log10 Ku + g -a. log, Kp = log, Ku + 9 + a and hence 9 = &(log10K + log, Kp -2 log, &.A a = i!(lOgio Kp -loglo Km), and s = log, KO-log, Kp Obtained from graphs plotted from observations recorded in this and previous communications the reciprocals of the various times taken for half-hydrolysis (x)and half-reduction (y) at the temperatures stated are summarised in Table I on which the following remarks are baaed The general effect.The order in which the atoms or groups affect the reactivity of the bromine (a)towards hydrolysing agents in a general way is OMe Me F C1 Br CO,H NO, ranging from the strongly enhancing methoxy-group to the strongly retarding nitro- group and (b)towards hydrogen iodide is OMe and Me C1 Br CO,H and F. In the latter case only the methoxy- and the methyl-group have an appreciable general influence on the reduction velocity. The strong general influence of the methoxy-group is noteworthy. The alternating efSect. The order in which alternation towards hydrolysing agents is produced is similar to that given above (a); in this case however the magnitudes of the effect may be compared.The order is OMe >F >Me >Cl> Br >C0,H >NO,. The differences observed in the first five cases i.e. negative groups have positive values whilst in the last two i.e. positive groups they have negative values. For reduction the order is OMe and Me>CO,H>F Cl * Logarithms are used in order that the expressions shall finally involve a product or a quotient as the case may be. TABLEI. KO. ICn,. I<,. g. a. 9. 7'7 +-7-x -7-h Compounds. X. Y. 2. Y-z. y. Hyd. Red. Hyd. Red. Hyd. Red. Methoxybenzyl <33.3(60") 0.107(25") 0.47 12.5 >33*3* 0.00 >la54 +W 0.92 -OO nil 00 bromides.W-Bromoxylenes. 2-17 (60") >0.00 (25")t 0.77 0.15 2.78 0.00 0.68 f-a 0.28 -CO -0.11 co Fluorobenz yl 0.28 (60") 0.00(25") 0.20 0.00 0.91 0.00 -0.39 -0.33 -0.51 -bromides. Fluorobenz yl 1-0(76") 0.27 (101") 0.71 0.22 4.0 0.18 -0.22 0.00 0.37 -0.05 -0.6 0.18 bromides. Chlorobenz yl 0.55(76") 0.36 (101") 0.65 0.24 1.8 0.31 -0.60 0.27 0-22 +Om05 -0.32 0.07 bromides. Bromobenzy 1 0.84 (76") 0.36 (101") 0.71 0.23 1.43 0.29 -0.67 0.22 0.15 +On06 -0.23 0.09 bromides. Nitrohenzyl 0.29 (76") -0.33 -0.29 -1.69 -0.028 -nil -bromides. w-Bromotoluic -0.72 (76") 0.27 (1lOO) 0.56 0.65 -1.07 -tO.O6 0.055 +0*19 acids. Benzyl bromide x = 0.67 (60°) 2-17 (76"). y = 0 (25") 0.20 (lolo),0-39 (110"). * The extraordinary rapidity with which p-methoxybenzyl bromide is hydrolysed in aqueous alcoholic solution suggesh that -! t]+ figure should be much greater and hence g and a for hydrolysis approach those for reduction i.e.00 M -f w-Bromo-o-xylem is definitely but very slowly reduced at 26". A defkite value for R,(y) cannot be given but 8 (red.) H must be large on account of the stability of w-iodo-p-xylene at 26" in presence of hydrogen iodide. p 218 SHOESMITH AND SLATER PREPARATION HYDROLYSIS and Br. Here the differences in the cases of the OMe Me and F compounds have positive values and in the other cases negative values. According to the principle of induced alternate polarities the sign of the difference is determined by the positive or negative character of the substituents and the reduction of the chloro- and bromo-benzyl bromides is the only case so far met with in this series in which the sign of the difference is other than was expected.The steric effect. The values obtained for s show that this effect in some cases diminishes and in other cases increases the reactivity the former being observed in the hydrolyses the latter in the reductions. Therefore the retarding of the reactions cannot be due to steric hindrance as normally conceived and the results justify the conclusion that there is a disturbing factor governing the reactivity of ortho-compounds (see Lapworth and Shoesmith J. 1922 121,1394). The influence of atoms or groups of atoms having the same electronic shell may be compared also since the manner in which the methyl and the methoxy-group and the fluorine atom each affect the reactivity of the bromine atom of the -C,H4*CH,Br group has been investigated.The order of potency is OMe>Me>F except in the case of alternation of hydrolysis for which the order is OMe>F>Me. E XPE RIME NTAL. Preparation of the Isomeric Chlorobenxyl BromidM.-The most reliable method of preparing the ortho- and para-isomerides is indicated by the scheme The appropriate chlorotoluene mixed with 5% of its weight of phosphorus pentachloride was chlorinated at 160"until the required increase in weight had taken place. The oil thus obtained was boiled with twice its bulk of 98% formic acid (d 1-20) for 3 hour the mixture poured into an excess of cold water and the oil separated dissolved in ether washed twice with water and once with aqueous sodium hydroxide.The aldehyde was precipitated as the bisulphite compound which was carefully washed with ether dissolved in water and the pure aldehyde liberated by the addition of excess of sodium carbonate. It wits converted into the alcohol (0-, m. p. 70"; p-,m. p. 72")by treatment with 25% alcoholic potassium hydroxide and this into the bromide by means of hydrogen bromide in benzene solution. AND REDUCTION OF FLUOROBENZYL BROMIDE ETC. 219 o-ChlorobenzyE Bromide is an oil b. p. 102"/9mm. (Found Br 39-05. C,H,ClBr requires Br 38.9yo). pChlorobenzy1 bromide has m. p. 51" (compare Jackson and Field Ber. 1878 11 905) (Pound Br 38.7%). m-Chlorobenzyl Bromide.+-Nitrobenzaldehyde was converted into m-chlorobenzaldehyde (Erdmann and Schwechten Annalen 1890 260 59) b.p. 213-214" and this was reduced to the alcohol in alcoholic potassium hydroxide. The alcohol had b. p. 242" not 234" as stated by Mettler (Ber. 1905 38 1749) and from it m-chZorobenxyl bromide an oil of b. p. 109"/10 mm. (Found Br 38-8%),was obtained as before. The Isomeric Bromobenxyl Bromides.-Each of these compounds was prepared by passing a stream of air through a weighed quantity of bromine into the appropriate boiling bromotoluene. The product was distilled under diminished pressure the portion passing over between 120" and 140"/12-16 mm. being collected cooled and when solid recrystallised from alcohol. The bromides thus obtained had m.p.'s o- 31"; m- 40"; and p- 63" [Found hydrolysable Br (o-) 32.1 (m-)31.8 (p-)31.8. Calc. for C7H,Br2 hydrolysable Br 32.0~0](compare Jackson Ber. 1876 9 932). The isomeric brmobenxyl iodides were prepared by boiling aqueous acetone solutions of the corresponding bromides with rather more than the calculated quantity of potassium iodide for Q hour. The mixture was then poured into water and the iodide recrystallised from light petroleum. o-Bromobenxyl iodide crystallises in shining white needles m. p. 47" (Found I 42.75. C,H,BrI requires I 42.7%). m-Brmnobenxyl iodide crystallises in white six-sided prisms m. p. 42" (Found I 42.8%). p-Bromobenzyl iodide crystallises in white needles m. p. 73" (Hantzsch and Schultze Ber. 1896 29 2253 give 80-81") (Found I 42.9%).[With R.H. SLATER.] Preparation of the Isomeric Fluoroto1uenes.-A solution of the requisite toluidine (25 g.) in a mixture of 30 C.C. of concentrated sulphuric acid and 80 C.C. of water was cooled to -5" and diazotised with 20 g. of sodium nitrite dissolved in 50 C.C. of water the temper- ature being kept below 5". The solution of the diazotised base was added to about 400 C.C. of commercial hydrofluoric acid (5&60()'0) in a 1,500 C.C. brazed Lspun-copper flask surrounded by ice after which the flask was fitted by means of a cork with a copper reflux condenser and very carefully warmed on a water-bath for about an hour (alternatively the mixture may be left at room temperature for about 16 hours). The condenser was then reversed and the flask heated directly.The mixture of fluorotoluene hydrofluoric acid 220 SHOESMITH AND SLATER PREPARATION HYDROLYSIS and cresol that distilled was collected in a copper beaker containing 300 C.C. of 30% aqueous sodium hydroxide solution surroundediby a good freezing mixture. When all the fluorotoluene had distilled the alkaline mixture in the beaker was extracted thrice with ether the combined extracts were de-emulsified by saturated ammonium sulphate solution and dried over anhydrous sodium sulphate and the ether was evaporated. In each case the fluorotoluene distilled at 113-118" and the yield was 15-16 g. (65%)(compare Holleman and Beekman Rec. trav. chim. 1904 23 238). Preparation of the Isomeric Fluorobenxyl Bromides.-Each fluorotoluene (80 g.in four lots) was brominated by volatilising bromine (30 g.) in a slow stream of dry air and passing the vapour into the boiling fluorotoluene (20 g,); by using small quantities nuclear substitution was avoided. The brominated oil was boiled with formic acid (d 1.20; m.2 vols.) for 6 hours fluorobenzotri- bromide and fluorobenzylidene bromide being thus converted into flnorobenzoic acid and fluorobenzaldehyde respectively and some of the fluorobenzyl bromide into fluorobenzyl alcohol. The mixture was poured into a large excess of water and the oil was separated and washed in ethereal solution with 10%aqueous sodium hydroxide until free from fluorobenzoic and formic acids. After complete removal of the fluorobenzaldehyde with freshly prepared sodium bisulphite solution-the aldehydes and the benzyl bromides form constant-boiling mixtures-the ethereal solution was dried over anhydrous sodium sulphate and the ether distilled.The residual oil was saturated in benzene solution with dry hydrogen bromide to convert any fluorobepzyl alcohol into the bromide and after the removal of the benzene the fluorobenzyl bromide was fractionally distilled in a vacuum in the apparatus described by Widmer (Helw. Chim. Acta 1924 7 52). o-Fluorobenxyl bromide had b. p 84-85"/15 mm. ; m-JEuorobenxyZ bromide 77"/12 mm. ; and p-JEuoro- benxyl bromide 85"/15 mm. [Found Br 42.4 (0-); 42.0 (m->; 42.1 (p-). C,H,FBr requires Br 42-3y0]. Hydrolysis of the Brmides.-Twenty C.C. of a standard solution (105c.c.) of the bromide in absolute alcohol were placed in a standard 25 C.C.flask 5 C.C. of water added and the volume was made exactly 25 C.C. by adding absolute alcohol. The whole was thoroughly mixed immersed in the vapour of boiling carbon tetrachloride for a definite time and then poured into a large volume of water. The liberated hydrobromic acid was titrated directly with standard aqueous sodium hydroxide and methyl-red. The results are in Table 11 where w represents the total weight of benzyl bromide used x the percentage changed and t the time of hydrolysis in hours. AND REDUCTION OF FLUOROBENZYL BROMIDE ETC. 221 TABLE11. Ortho-compounds. Meta-compounds. Para-compouds. F c1 Br w = 0.5218 g.0.5600 ,,0-6702 , 0.5456 g.0.5441 , 0.6600 , - 0.5102 g.0-5418 ,,0-6693 , - X.2. X. t. F. C1. Br. F. C1. Br. F. C1. Br. 14 25 51 25 46 24 44 24 41 21 38 21 39 68 88 45 69 40 62 2 76 68 66 63 58 60 96 88 85 4 94 88 87 86 80 82 98 96 95 8 95 96 96 94 92 95 99 98 9s In Table I11 are results obtained in a similar manner in order to compare the hydrolysis of benzyl bromide m-methoxybenzyl bromide,* the fluorobenzyl bromides and the w-bromoxylenes at 00.5"(b. p. of chloroform). TABLE111. Pluorobenzyl bromides. m-Methoxy-Benzyl P-benz yl bromide. bromide. 0. m. 'U'. 0.5240 0.5252 0.5102 0-543 0.46 10 t. X. X. 2. 2. 2. ;t 9 7 26.5 14 22 1 20 11 48 29 39 2 33 23 69 49 59 4 56 41 90 72 80 8 76 63 95 88 88 16 91 81 96 -Reduction of the Bromides.-Reduction at 25" under the conditions described by Lapworth and Shoesmith and by Shoesmith and Slater (Eoc.cit.)being so slow that satisfactory results could not be obtained the bromides were reduced in a thermostat at 101".Ten C.C. of a standard solution of the bromide in 50 C.C.of glacial acetic acid were placedin the 25 C.C. standard flask 10 C.C. of freshly distilled hydriodic acid (d 1-680)added the volume was made up to 25 C.C. with glacial acetic acid and the whole thoroughly mixed and kept in the fhermostat for a definite time. The percentage reduction was then estimated as in previous cases. The concentration of the hydriodic acid and the amount of iodine in it both influence the rate of reduction and therefore a series of reductions was carried out on the same day with the same hydriodic acid.The results are in Table IV t and x having the same significance as before. * At all temperatures in aqueous alcoholic solution p-methoxybenzyl bromide is completely and the o-compound almost completely hydrolysed in 24 minutes. 222 PREPARATION EYDROLYSIS AND REDUCTION ETC. TABLEIV. Ortho -compounds. Metrt-compounds. Para-compounds. 2. X. 2. -& -&-t. F. C1. Br. F. C1. Br. F. C1. Br. 13 29-5 32 30 (31) 23 23 23 (21) 22.5 28 26 (27) 3 45 53 53 (54) 37 43 43 (41) 34 49 47 (47) 6 65 76 74 (76) 63 64 61 (60) 57 68 64 (65) BenzyI bromide x = 25 38 and 61 for t = 1* 3 and 6 respectively. The figures in brackets represent the reduction of the bromobenzyl iodides.In Table V are the results of reductions of the bromides made at a higher temperature (110") in order to determine the amount of nuclear halogen eliminated. TABLEV. Ortho-compounds. Meta-compounds. Para-compounds. X. X. X. -t. <-c1. Br. c1.T- Br c1. Br. 2 76 77 57 57 70 72 4 91 91 81 79 88 88 8 102 100 97 97 98 97 18 104 103 100 100 101 101 p-Chlorobenzyl iodide (m. p. 65") was isolated from the reaction of p-chlorobenzyl bromide and hydrogen iodide in glacial acetic acid at 25" (compare Shoesmith and Slater Zoc. cit.) but owing to the dilution at which the reductions had to be carried out end-products in the other cases could not be separated. The figures obtained however show that the reaction in the cases of the meta- and para- isomerides was quite normal in that it first involved conversion of the bromide into iodide and subsequent reduction of this to the halogenated toluene.The iodobenzyl bromides have not been examined since in a former communication we showed that hydrogen iodide reduces iodo- toluene to toluene. The authors gratefully acknowledge the many valuable sugges- tions made by Dr. W. 0. Kermack especially with regard to the formuh on pp. 216 and the receipt of a grant from the Earl of Moray Research Fund. They also thank the Trustees of the Carnegie Trust for the Universities of Scotland for a scholarship which enabled one of them (R. H. S.) to take part in this research. UNIVERSITY EDINBWEGH. [Received December 35-4 1925.1 OF