1412 CAIN AND NICOLL: THE RATE OF CXL1.-The Rate of Decomposition of Diazo-compounds. By JOHN CANNELL CAIN and FRANK NICOLL. IN view of the extended use, on the large scale, of aqueous solutions of diazo-salts, it is surprising how little has been published on the subject of their stability. In the manufacture of azo-colours, the diazo-salt is usually ‘; coupled ” very shortly after its preparation, but in the production of the so- called “ice colours ” on the cotton fibre, the solution often has to stand for a considerable length of time. It is therefore of much importance to know the proper conditions under which the solution will remain undecomposed, of which the moat important is the temperature. The chief point of interest, however, lies in the study of the rate of Purt I.Diazo-compounds of the Benzene Series.DECOMPOSITION OF DIAZO-COMPOUNDS. PART I. 1413 decomposition of diazo-salts, and the investigation of this for a number of compounds has been the subject of our work during the past year, The decomposition of an aqueous solution of a diazo-salt is, in most cases, a simple one, and belongs to the class of unimolecular processes for which me have the well-known expression 1 A - log = C (a constant). Measurements of the rate of decomposition of these compounds were first made by Hausser and Muller (Bull. Soc. Chim., 1892, {iii], 7, 721 ; 1893,9,353 ; also Compt. rend., 1892, 114, 549, 669,760,1438), 1 A t A - x who determined the value of - log - for the diazo-salts from Aniline (sulphate and hydrochloride). 0-, m-, and p-Toluidine (sulphate), 0-, m-, and p-Aminobenzoic acid (sulphate), m- and p-Sulphanilic acid, a t a temperature of from 40' to 64'.The results obtained showed that only the diazo-salts from p-sulph- anilic acid and p-toluidinesulphonic acid gave a constant value for p-Toluidine-2-sulphonic acid, 1 A - log -, all the others giving gradually diminishing values. t A - x Hausser and Muller explain this unexpected result, in the case of diazobenzene only, by adducing experiments in evidence of a specific retarding action " of the phenol formed during the decomposition. Hantzsch (Ber., 1900, 33, 2517) measured the rate of decomposition Aniline, p-Anisidine, p-Toluidine, $-Cumidine, p-Bromoaniline, and showed %hat at 25' all these substances gave a constant value for of the diazo-salts from the hydrochlorides of 1 A t 4-32 - log -.Hantzsch suggests that the non-agreement of his results with those of Hausser and Muller may be due to the existence of secondary re- actions set up at the high temperature of the experiments (40' to 64'). We have therefore measured the rate of decomposition of diazo- benzene chloride at temperatures frqm 20' to 60' in order to deter- mine, if possible, the temperature at which such secondary reactions might set in, this point being shown by the diminishing values for C. We shall show that, under- the conditions of our experiments, the1414 CAIN AND NICOLL: THE RATE OF figures for diazobenzene chloride give a constant value for the above expression at all temperatures between 20° and 60°.Similar results were obtained with the diazotoluene chlorides. The full list of amines and diamines, from which we have prepared and examined the corresponding diazo-salts, is as Eollows : Aniline, Benzidine. 0-, m-, and p-Toluidine. 0-, m-, and p-Nitraniline. Dianisidine. oo-Dichloro benzidine, psulphanilic acid. Tolidine. p-Aminoacetanilide. Most of the experiments of Hausser and Muller, and all those of Hantzsch, were made with solutions obtained by dissolving the pure, dry diazo-salt in water. The former chemists prepared the diazo-solu- tions from 0- and m-toluidine sulphate by adding a solution of sodium nitrite to an acid solution of the amine ; they state, specially, that addition of sodium sulphate to a solution of dry diazo-p-toluene sulph- ate in pure water has no effect on the rate of decomposition, and conclude that it is immaterial whether the solution is prepared one way or the other.Hantzsch (Zoc. cit.) states (and we have also found) that the presence of free mineral acid has no effect on the reaction. All our solutions have been prepared directly, as on the large scale; and as our results for diazobenzene chloride agree fairly well with those of Hantzsch, we feel justified in criticising those of Hausser and Muller from the point of view of our own. The method we used for determining the rate of decomposition was, in principle, the same as that used by the chemists to whom we have referred, namely, measurement of the volume of nitrogen evolved from time to time. The arrangement used for doing this was nearly the same as that of Hantzsch, who connected a flask containing the diazo- solution by means of a capillary tube with a nitrometer ; we have used a water-jacketed Hempel’s burette, and an ordinary purette also water-jacketed.In the determinations above 30°, we interposed a very small, spiral condenser, so as to avoid any error due to the in- complete cooling of the hot nitrogen. The question as to the measurement of time is the next considera- tion. It being, of course, impossible to bring the flask containing the diazo-solution to the required temperature instantaneously, it is necessary to fix on some point from which to reckon the time of reaction, so as to make the error as small as possible. Hausser and Muller, indeed, attempted to determine the theoretical starting point from the volume of gas evolved in successive periods of time after the right temperature was reached. By extrapolation, theyDECOMPOSITION OF DIAZO-COMPOUNDS.PART I, 1415 found the length of time which would have been required for the evolution of the gas produced during the heating of the solution. But there are certain objections to this method, for, in the first place, the “curve of the rate” (taking volumes as ordinates and times as abscissa) seems to have been produced backwards by freehand draw- ing, as by using Laplace’s interpolation formula we obtain a slightly different result, and, secondly, the first few numbers obtained in these experiments (as Hantzsch proves) are much too low, owing to some of the nitrogen (at first a large proportion) remaining dissolved in the solution, From a large number of preliminary observations, we believe that the method of Kantzsch, which we have adopted for reckoning the starting point, gives us a point which can only differ from the theo- retical point by a very few seconds.The flask then, having been nearly filled with the diazo-solution and connected by a two-holed caoutchouc stopper carrying a thermometer and a capillary tube with the measuring burette, was quickly heated to the temperature of the experiment by a Bunsen flame. The flask usually contained pieces of broken glass for purposes of adjustment and was well shaken during this operation to avoid overheating. The time was reckoned from this point, and the flask plunged into a water-bath which was kept a t the right temperature by a gas regulator. The gas evolved was now measured from time to time, and in many experiments the flask was heated to a higher temperature unti1,on cooling to the temperature of the bath, no increase in volume of the nitrogen was observed.This total volume of nitrogen determined experimen- tally always agreed well with the volume calculated from the amount of diazo-salt taken. The correction for the expansion of the liquid and the small volume of air in the flask was always- determined experimentally and sub- tracted from the readings of the burette. In the following tables, A is the total volume of nitrogen at the temperature and pressure of the experiment calculated from the weight of diazo-salt contained in the solution ; x is the corrected volume of nitrogen at the same temperature and pressure evolved during the time t (in minutes), C is the constant calculated according to the equation 1 A c=-log - t A - X .1. Diaxobenxene Chloride. 9.3 Grams of aniline (1/10 mol.) dissolved in 30 C.C. of hydrochloric acid of sp. gr. 1.16 (3/10 mol.) cooled with ice and 6.9 grams of sodium nitrite added (100 C.C. of a solution containing 69 grams NaNO, per1416 CAIN AND NICOLL: THE RATE OF litre). therefore contained 10 grams of diazobenzene chloride in a litre. made in the above proportions was used for each experiment : The solution thus obtained was made up to 1400 c.c., and A fresh solution Thirty-five C.C. were taken in each experiment. Temperature 20". 1. A=60 C.C. (13", 750 mm.).t. 116 3 31 159 180 192 355 393 422 43 9 458 481 1282 1329 1354 1424 X. 9.7 C.C. 11.1 ) ) 13% ), 15.3 ,, 16'2 ,, 26'3 ), 29'8 ,) 30-5 ), 31.5 ), 32-6 ,) 51.4 )) 33'7 )) 52.8 ), 53-9 ,) 54.5 ) ) C. 0 '000 6 6 0'00068 0'00070 0 -0 00 70 0 -000 7 1 0'000 7 0 0.00076 0 -00073 0.00073 0-00074 0*00074 0'00066 0'00069 0*00073 0.00073 Range 16 to 91%. Mean 0.00071 2. A=61 C.C. (14") 741 mm.). t. 45 1 474 486 506 532 545 578 596 607 1382 X. 31'4 C.C. 32.6 ), 33'2 ), 34'4 ) ) 36.4 ,) 37'4 ,, 39.2 ,, 40-1 ), 40.5 ,) 55.3 )) C. 0'00069 0*00070 0 *00070 0-00071 0 '00 074 0*00075 0 '000 7 7 0.00078 0.00078 0.00074 Range 52 to 92%. Mean 0'000736 Mean of the two experiments, 0'00072, Taking the first experiment, it will be seen that no value for C is recorded until 116 minutes have elapsed.The numbers obtained before this time are much lower and gradually increase until they become constant. This apparent departure from the law is due to the fact that the solution must become saturated with nitrogen before the correct volume is measured in the burette. That this is the correct explanation was very clearly proved by Hantzscb, and we have therefore left out these irregular figures from our tables. That there are considerable sources of error both a t the beginning and towards the end of the reaction is obvious, and we shall therefore usually note the range of the reaction through which our observations extended (thus, for example, " range = 20 to 60 per cent." means that our numbers extend from the point at which 20 per cent. of the diazo-salt has been decomposed until 60 per cent.hae been decomposed). The rangeDECOMPOSITION OF DIAZO-COMPOUNDS. PART I. 1417 E:f:ze oft. of our observations is usually considerably wider than that of Hausser and Muller or Hantzsch. The above numbers agree fairly well with those obtained by Hantzsch at a temperature of 2 5 O , whose results for C varied in four experiments from Om0O064 to 0.00072, the range covered by three experiments being 12 to 62 per cent. (The fourth experiment was made according to an admittedly more inexact titration method.) There is not much difference in the constants at these temperatures, but, as will be seen, the constants increase enormously at higher tem- peratures. C. Extreme values of 2. Lowest. Highest. -____ Series. 0.00278 0.00278 I.A=60*45 C.C. (16", 756 mm.) 11. A=59*8 C.C. (lS", 761 mm.) 0-00305 0'00305 I. A=59*7 C.C. (IS", 765 mm.) It. A=59'7 C.C. (13'5", 7565 mm. 45 to 382 53 to 314 16 12 12 8 15.4 C.C. to 56.1 C.C. (range 25 to 92 %) 17.2 C.C. to 53.0 C.C. (range 30 to 90 %) 25 to 75 81.5 to T8-5 Temperature 30". * 2 3 5 C.C. to 46'5 C.C. (range 40 to 78 %) 27.8 C.C. to 47.2 C.C. (range 46 t o 80 %) 0 *00864 0 -00902 Mean.. Mean. 0.002947 0 -002956 0.00295 0*00876 0-00879 0.00877 * The figures obtained in this and the following experiments are of the same character as those in the experiments a t 20". We have therefore arranged our results in an abbreviated form, giving, however, one set of figures in full for each substance. We made our next experiments at a temperature of 50°, the same as that at which Hausser and Muller worked.We have already mentioned that our experiments were not made under the same conditions as those of the above chemists, but apart from this there is much that is open to criticism in the figures which Hausser and Muller give, A serious error occurs in the numbers for diazobenzene chloride (Bull. Xoc. Chim., 1892, [iii]] 7, 721) which we quote here. We have added a new column showing the values for C, which we obtain on calculating out Hausser and Muller's results. VOL. LXXXI, 5 c1418 CAIN AND NICOLL: THE RATE OF Experiment 1.-Solution contained 6-25 grams of diazobenzene chloride per litre. A = 660 C.C. t. 16-5 26 -5 36.5 46.5 56.5 x. I C. (Hausser and Muller) Recalculated, C. and N. ~~~ 0.0303 0'0285 0,0262 0.0232 0'0205 Range 31 to 66 per cent, 0.0099 0'0099 0 -0087 0-0074 0'0063 Zxperiment 2. -Solution contained 9.47 grams of diaxobenzene chloride per litre (the solution used in our experiments contained 10 grams per litre).A = 453 C.C. 12.6 17.6 22'6 27'6 32'6 42'6 0. 206 C.Ch 255 $ 9 285 9 , 305 9 9 317.5 ,, 332.5 ,, C. Hausser and Muller) 0'0360 0'0348 0'0332 0.0317 0'0299 0'0267 Range 45 to 73 per cent. C. Recalculated, C. and N. 0.0209 0.0204 0'0190 0.0176 0*0160 0.0135 There is thus a very great discrepancy which we are quite unable to account for. We were led to check these results after drawing a curve showing the course of the decomposition. It was obvious that the curve could not represent the reaction. We may add that we have recalculated all the numbers given in the various experiments by Hausser and Muller and find them to be correct.Turning now to the figures given in the same paper for diazo- benzene sulphate, we find that four experiments were made with three different solutions, containing respectively 3'38, 6.7, and 10.8 grams of diazo-salt per litre. The corresponding values for C did not differ very much from each other ; hence Hsusser and Muller conclude that the decomposition is not influenced by the concentration, The numbers for C, however, in each experiment are not constant, they gradually diminish as the decomposition goes on j thus inDECOMPO!UTION OF DIAZO-COMPOONDS, PART I. 14 1 9 Exp. I they diminish from 0.0298 to 0*0210. Range 49 to 80 per cent. 9 , 2 9 , ,, 0.0289 ,, 0.0177.), 44 ), 82 ,? 9 ) 3 9 , ), 0.0325 ,, 0.0191. ,, 45 ,, 84 ,, 31 4 ) J ,, 0.0334 ,, 0.0206. ,, 50 ,, 86 ,) From these numbers, Hausser and Muller conclude that the phenol formed in the reaction has a specific retarding action, and to prove this made a fifth experiment, to which phenol was added. From the result of this, they claim to have shown that their ex- planation was a true one, but on comparing the figures obtained, me do not think the conclusion is justified. The solution contained 22.5 grams of diazo-salt per litre, with 10.5 grams of phenol per litre, and the values for C diminished from 0.0259 to 0.0216 (range 48 to 89 per cent . ) . We have compared the values of C in the above experiments at points representing equal amounts of decomposition. Thus, when 56 per cent.had decomposed, the value of C in two experiments mas 0-0284 and 0,0318, and in the phenol experiment 0.0259. When 66 per cent. had decomposed, we find C=O*O267, 0.0297, and in the phenol experiment 0.0255, Here the value for C in the phenol experiment is less than that in the two ordinary experiments, although these differ from each other more than one of them differs from the phenol experiment. But a t a, later stage we find : A t 75 per cent. decomposition the value of C was 0-0230 and 0,0221, and in the phenol experiment 0.0248. At 80 per cent. decomposition the value of C was 0*0210 and 0,0222, and in the phenol experiment 0*0242. Here the value for C in the phenol experimeut is greater than that in the other experiments. Experiments were made (Hausser and Muller, Zoc.cit.) adding, in- stead of phenol, phenylsulphuric acid, sugar, and oxalic acid, and from equally unconvincing data the authors conclude that substances con- taining the benzene nucleus retard the reaction, whilst those belonging t o the fatty series have no action. We think, however, that the results of Hausser and Muller do not lead to this conclusion, and in order to test this point further we have made experiments also at 5.0' t o see if, under the conditions of work- ing, the addition of phenol has any effect on the reaction. We think our numbers prove conclusively that there is no trace of either a retarding or accelerating action produced by this substance.1420 CAIN AND NICOLL: THE RATE OF Temperature 5 09 The solution contained 10 grams of diazobenzene chloride per litre as before.In Experiment 3 (p. 1421), the solution contained 4-57 grams of phenol per litre, this being the same proportion of phenol to diazo-salt as that in the experiments of Hausser and Muller : -~ t, -_____ 5.5 6 6 -5 7 7.5 8 9 10 12 13 15% 17 19 21 23 26 30 1. A=58*3 (7') 753 mm.). IL'. 17.7 C.C. 19.3 ), 20.85 ,) 22.55 ,, 24 Y , 25'4 ,, 27-8 ,, 29 9 3 , 33% ), 35.4 ), 39.1 ,) 41.0 ,, 43.1 ,, 45.0 ), 46.5 ,) 48-4 ) ) 50.25 ,) Range 30 to 84 %. C. 0.0286 0'0291 0.0295 0.0303 0.0307 0'0311 0.0313 0'0312 0.0311 0'0312 0.0311 0*0310 0'0307 0'0306 0'0301 0.0296 0.0287 Mean 0.0303 t. 9 9.5 10 11 12 13 14 16 18 20 22 24 26 2. A=58*3 ( 7 O , 753 mm.). X. 26.0 C.C. 27.3 ), 28.5 ) ) 30% ,, 32.6 ,) 36.0 ,, 38.85 ,, 41.3 ,, 43.25 ,, 45.0 ,) 46'45 )) 34.4 ,, 47.7 ) ) C.0*0285 0.0289 0.0291 0'0294 0.0296 0'0298 0.0298 0.0296 0-0297 0.0294 0 '0292 0.0288 0.0285 Range 44 to 80 %. Mean 0.0293 Mean of experiments 1 and 2, 0'0298. These figures show very clearly that the addition of phenol to the diazo-solution has absolutely no influence on the course of the re- action. A fourth experiment was made a t this temperature with a solution of different concentration, namely, containing 5 grams of diazo-salt per litre :DECOMPOSITION OF DIAZO-COMPOUNDS, PART I. 1421 i-I t. 8 9 10 12 13 16 17 18 20 22 24 27 30 16.4 C.C. to 45-4 C.C. (range 28 t o 77 %) 17.4 C.C. to 44'8 C.C. (range 30 to 76 %. X. 24.3 C.C. 27.1 ), 29.5 ,) 33'6 ), 40'6 ,, 41.9 ), 43.0 ), 45'3 ,, 47'1 ,, 48.5 ), 50-3 ,) 51.8 ,, 35'4 ,, 0.102 0'1181 0.102 0.111 Mean ...Range 40 t o 86 %. I. A=58*7 C.C. (7", 748 mm.) 11. A=58*7 C.C. (8" 750mm.) 0-028i 0.0289 04293 0.0296 0'0297 0.0305 0,0305 0 -0303 0-0304 0.0302 0'0298 0'0291 0'0286 9 10 11 13 15 17 21 24 28 15.1 C.C. 17.2 ,, 18'9 ), 20'3 ,, 22.6 ,) 23'8 ,) 25'1 ,, Mean 0.0296 Range 50 to 83 %. C. 0.02'15 0.0283 0.0286 0'0287 0.0287 0'0283 0'0278 Mean 0.0283 The very slightly less value of C in experiment 4 is sufficiently accounted for by the greater proportion of nitrogen which can be retained in the more dilute solution : thus, as Hausser and Muller and Hantzsch have shown, the concentration has no influence on the course of the reaction : Temperature 60". Extreme valueg of t. 1; to 69 1.5 to 6 r C. Extreme values 0.11 03 0-1078 ~ 0.1090 An examination of the values for C at the various tempera- tures shows that they increase enormously as the temperature is raised.This is seen very clearly in the curve where the following values for C are ordinates and the temperatures are abscissze :1422 CAIN AND NICOLL: THE RATE OF 166 147 188 199 222 Temperature. 20° 30 40 50 60 26.4 0'00178 0'00184 0*00189 29.4 32.6 33 9 0'00190 36 2 0'00189 CURVE No. 1. 26'3 29'3 32'4 33.8 36.3 C. 0-00072 0.00295 0*00877 0-0298 0.109 0.001 77 0-00183 0-00187 0'00189 0-00191 2. Diaxo-o-toluene Chloride. 10.7 Grams of o-toluidine (1/10 mol.) dissolved in 30 C.C. of hydro- chloric acid of sp. gr. 1.16 (3/10 rnol.), cooled with ice and 6.9 grams of sodium nitrite added (100 U.C. of the normal solution).The solu- tion of diazo-o-toluene chloride was made up to 1400 C.C. and therefore contained 11 grams of diazo-salt per litre, Thirty-five C.C. mere taken in each experiment, Temperature 20'. ~ ~~ 1. /I 2. Az58.3 (8*5", 758 mm.). 11 A358.3 (8V, 758 mm.). I I i I t. 147 166 188 199 222 2. 1 c.DECOMPOSITION OF DIAZO-COMPOUNDS. PART I. 1423 t. 236 299 388 430 474 X. 37.; 42.7 47 '5 49.2 50 '4 Range 45 to 85 X. 24 to 74 27 to 60 Temperature 20" (continued). 19-2 C.C. to 41.9 C.C. 0.0069 (range 31 to 70 %) 21'1 C.C. to 37.7 C.C. 0-0068t (range 35 to 62 %) C. 13 to 30 29.2 C.C. to 46.8 C.C. (range 50 to SO %) 10 to 24 23.8 C.C. to 42.5 C.C. (range 40 to 70 %) 0~00190 0.001 91 0*00188 0.00187 0 *00183 0'0229 0,0225 236 286 295 384 433 455 1.5 to 6.5 1'5 to 7 37.7 41 *9 42'6 47-3 49.2 49'8 16.6 C.C.to 44.7 C.C. (range 28 to 75 %) 16-7 C.C. to 46'3 C.C. (range 28 to 79 %) Mean 0.00187 11 Range 45 to 85 %. 0 *oo 19 1 0*00193 0'00193 0 '00 189 0.00186 0'00184 Jlean 0'00187 Seiies. I. A =60'6 C.C. (8*5", 730 mm.) 11. A=60*6 C.C. (8'5", 730 mm.) J. A=58'8 C.C. (8", 750 mm.) 11. A=58'8 C.C. (8", 750 mm.) 1. A =58*8 C.C. (SO, 750 mm.) 11. A=58'8 C.C. (8", 750 min.) Mean of the two experiments, 0.00187, Lowest of t. -4 I- 0.0954 0.0961 C. Highesl 0.0073 0-00713 Mean I . 0'0247 0,0244 Mean.. 0'103 0.113 Mean.. Mean. Om0O708 0*00704 0.00706 0.0240 0.0236 0.0238 0.0997 0.1057 0.1027 -1424 CAIN AND NICOLL: THE RATE OF The collected values for C are therefore : Temperature. 20° 30 40 50 C. OgO018'7 0*00706 0.0238 0.1 027 graphically represented by the following curve : CURVE No.2. Our results again differ from those of Hausser and Muller, who measured the rate of decomposition of diazo-0-toluene sulphate (Bull. Xoc. Chim., 1893, [iii], 9,353). The diazo-salt was prepared in exactly the same way as was ours, except that we used,hydrochloric instead of rrulphuric acid. The concentration of our solution was very near that employed in Hausser and Muller's first experiment, their second solution being tsvice as strong. Their experiments were made at 40° and the results were : 1, value for C gradually diminished from 0.026 to 0.020 ; 2, value for C gradually diminished from 0.0251 to 0.020. As is seen above, we obtain a constant value for C = 0.0238. Hausser and Muller do not offer any explanation of their results, 3. Bhxo-m-toluene Cht?oride.The diazo-salt was prepared from m-toluidine exactly as in the case of o-toluidine.DECOMPOSITION OF DIAZO-COMPOUNDS, PART I. 1425 Temperature 20". I. A=60*8 C.C. (12'5", 740 mm.) 11. A=60'9 C.C. (12'5", 739 mm.) I. A=60*2 C.C. (12", 746 mm.) 11. A=6O'O C.C. (12", 747 mm.) ~ ~~ ~ ~~ 1. ~ A=60'9 (12*5", 739 mm.). 11 11 8 8 t. 135 143 160 176 192 206 226 282 313 324 347 22.6 C.C. to 44.7 C.C. (range 37 to 74 %) (range 40 to 72 %) 23'4 C.C. tQ 43.6 C.C. 2. 0'0245 0.0253 28 '3 29 -7 32.5 34-7 37'1 38'7 40.8 45'2 47.2 48.0 49 *1 Range 47 to 81 %. C. 0*00201 0.00203 0-00207 0 000208 0'00213 0.00213 0.00213 0*00209 0'00207 0'00208 0*00205 Mean 0-00208 2. A=60'9 (12.5", 739 mm.). t. 97 109 122 135 143 160 176 206 226 243 282 324 350 X.22'4 24-8 27 '0 29 '0 30-2 33 '1 35 3 38'6 40-5 42'1 45 '1 47 '9 49 -7 Range 37 to 82 %. Extreme values of t. 40 to 122 38 to 113 84 to 24 84 to 22 C. 0*00205 0 -00208 0~00208 0'00208 0'00208 0 -0 0213 0'00214 0'00212 0~00210 0'00210 0'00208 0'00207 0~00210 llean 0'00209 Mean of the two experiments, 0*002085. Extreme values of x. Temperature 30". 28.6 C.C. to 52.2 C.C. (range 47 to 86 %) 26'9 C.C. t o 50.7 C.C. (range 44 to 84 2) Tempsrature 40". Lowest. 0.0069 0.00665 C. Highest 0*00712 0 '00720 Mean. 0.0256 0.0268 Mean.. - Mean. 0 '00702 0.0069 0-00696 0.0252 0.0262 0'0257 -1426 CAIN AND NICOLL: THE RATE OF d - 0 2 Series. !4 2 - .;: We have for the values of C, Temperature. zoo 30 40 C. E::f:r Extreme values of 2. Lowest. Highest.Mean. oft. -- ' I 1- C. 0*002085 0.00696 0.0257 A=58*9 C.C. (8", 749 ma.) graphically represented by the following curve : 9 300 to 4310 7.8 C.C. to 54.6 C.C. (range 13 to 92 %) CURVE No. 3. Hausser and Muller's results (Zoc. cit.), for diazo-rn-toluene sulphate prepared in solution (that is, as above) diminished from 0-0305 to 0.0219 at 40°. The concentration in this experiment was the same as ours. Two other experiments with solutions of different concentration gave different., although still diminishing, values for C. 4. Diazo-p-toluene Chloride. The diazo-salt was prepared from p-toluidine exactly as in the case of o-toluidine. Tempratwe 30'.DECOMPOSITION OF DIAZO-COMPOUNDS. PART I. 1427 Temperature 40'. 66 to 208 24-5 C.C. to 48.3 C.C. (range 41 t o 81 %) 80 to 144 28'2 C.C.to 41.2 C.C. (range 47 to 70 %) 80 to 181 28.2 C.C. to 45.7 C.C. (range 47 to 78 %) 0-0035 0.0035 0.0035 20 to 46 20 to 62 24-5 C.C. to 42'4 C.C. 1 (range 41 to 71 %) 24.7 C.C. to 48.8 C.C. (range 41 to 81 %) 27'3 C.C. to 46.8 C.C. (range 46 to 78 %) 20'2 C.C. to 49'3 C.C. (range 34 to 82 %) 0.0445 0.0450 2. A=59*1 (8*5", 748 mm.). 1. 8-59.1 (85", 748 mm.). -- 0. t. C. C. t. 2. 180 200 230 260 301 372 057 491 580 634 19.7 21.6 24.1 265 29.6 33.9 38.8 40 *7 43 *3 45*0 0.000978 0'000988 0'000989 0 -000994 0*00100 0-000995 0-001011 0'001032 0*000988 0'000981 210 230 260 301 368 454 491 580 630 22'3 23 -9 26'4 29 *4 34.0 39 -0 40'7 43.5 45e4 0'000980 0 *00097 9 0*000989 0'000993 0*001011 0 '001032 Om0O1032 0.000997 0'001007 Range 33 to 75 %.Mean 0'000995 Range 37 to 76 2. Hean 0*001002 Mean of the two experiments, 0 *000999. - d u o 0 p d $ 4 2 % - 9 6 5 8 9 8 11 - L I C. Series. Highest 0.00368 0*00360 0'00358 Mean.. 0*0119 0-0128 Mean.., 0.0463 0'0479 Mean ... Mean. 0*00361 0 -003 56 0 -00356 0.00358 0*0118 0'0122 I I- I. A=59*1 C.C. (a", 747 mni.) 11. A=59*1 C.C. (So, 747 mm.) 111. A=59*1 C.C. (SO, 747 mm.) I. A=59*5 C.C. ( S O , 742 mm.) 11. A=59*6 C.C. (So, 740 mm.) 0.0115 0*0116 0'0120 0.0455 0,0467 0'0461 - I. A=595 C.C. (8.5", 743 mm.) 11. A=59*5 C.C. (8'5", 743 mm.) 6 to 15 4 to 171428 CAIN AND NICOLL: THE RATE OF Collecting all the results for diazo-p-toluene chloride, we have : Temperature. 30° 40 50 60 70 or graphically the following curve : CURVE NO. 4. C. 0*000209 0*0009 9 9 0*00358 0-0120 0.046 1 Hantzsch obtained the value 0.000081 at 25O, and Hausser and Muller numbers diminishing from 0-024 to 0*018 for the sulphate at 64O.The above experiments on diazobenzene and the diazotoluenes ex- tend over a wide range of temperature, and show that at all these temperatures the rate of decomposition is in accordance with the law, thus confirming Hantzsch’s results at a low temperature and being opposed to those of Hausser and Muller at higher temperatures. Of the substances still to be described, only one (diazo-p-sulphanilic acid) was examined by IIausser and Muller, who found a constant value for C. We have therefore only measured the rates of decomposition of the following diazo-salts at one or two temperatures.DECOMPOSITION OF DIAZO-COMPOUNDS.PART I. 1429 ~ j t. 5. Diazo-p-sulpianilic Acid. 17.3 Grams of sulphanilic acid were dissolved in water and caustic sod:i. Forty C.C. of hydrochloric acid of sp. gr. 1.16 (4/10 mol.) were added and then 6.9 grams of sodium nitrite in solution. The solution of diazo-salt was made up to 1400 C.C. and then contained 15.7 grams of diazo-salt per litre. Thirty-five C.C. mere taken in each experiment. ~ l1 15 20 32 35 43 51 76 90 Temperature 60". 25 30 35 45 51 59 66 76 90 t. 1 2. 1 18'1 21-3 23'9 28 -3 31.6 33 '4 35.7 38'9 42 -9 Series. C. o(1 3; Extreme d g! values Pi ?G! o f t , +? 0*00640 0'00652 0'00648 0 *00635 0 -00658 0 .ooe19 0 -0061 6 0 -0 0 6 21 0.00633 16'4 C.C. to 44.7 C . C . (range 28 to 75 %) 20'2 C . C . to 42.9 C . C . (range 34 to 72 %) 0.0681 0.0711 0.0696 0.0712 0.0744 0.0723 Mean ...0.070S I. A=58.7 C.C. (go, 752 mm.) 11. A=58-7 C.C. (go, 752 mm.) c. 2* I 9 2 to 9 8 2.5 t o 8 8-6 11.8 15.5 22-3 23 *7 27.1 29 *9 38'2 42'5 0'00625 0'00649 0-006E5 0.00648 0'00641 0-00625 0'00606 0.00601 0.00621 Range 30 to 72 %. Mean 0'00636 11 Range 15 to 72 %. Mean 0.00631 Mean of the two experiments, 0.00633. Temperature 80". G. Extreme values of x. We have drawn a curve through the two points given by the above determination, and it is interesting to observe that from this curve the value for C a t 64O, the temperature a t which Hausser and Muller1430 CAIN AND NfCOLL: TEE RATE Ol made their experiments on this substance, corresponds almost exactly with the value they found, namely, 0.0106, as the mean of four experiments.CURVE No. 5. We are here, for the only time, in agreement with Hausser and Muller’s results, and it is important to note that their experiments were made (I) with the solid diazo-compound, (2) in three different concentrations. 6. Diazo-o-nitro6enzene Chiwide. 3*45 Grams of o-nitraniline (1/40 mol.) were dissolved in 11 C.C. of hydrochloric acid of sp. gr. 1-16 (4.5140 mol.) and diazotised by the addition of a solution of 1.73 grams of sodium nitrite. The diazo- solution was made up to 700 C.C. Seventy C.C. of this solution were taken in each experiment. This solution differs from the corresponding ones from aniline and the toluidines in being more dilute; instead of a diazo-solution con- taining 1/10 mol. in 1400 c.c., we have 1/20 mol. in the same volume, and, as is well known, it is necessary to use a fairly large excess of acid in order to diazotise the nitranilines successfully, so that we have here 44 mols.of acid to 1 mol. of base, We do not think that these changes are sufficiently great t o prevent us from comparing the con- stants for the diazo-salts from the nitranilines with those from the other amines examined. This diazo-salt is exceedingly stable : at 70°, only about 50 per cent. was decomposed after 174 hours, so that we have determined the rate of decomposition at looo,DECOMPOSITION Ol? DIAZO-COMPOUNDS. PART I. 1431 Temperature 100". 2. A=59'1 (8'5', 748 mm.). 1. A=59'1 (8*5", 748 mm.). C. t. 2. C. t. 5'5 10 17 25 35 45 55 65 79 115 135 z. 7 17 20 31 35 45 50 70 80 90 100 5 *1 11.6 13 '2 19.5 21 *5 26.1 28'3 35.5 38-6 41 '2 43-5 0 *00560 0.00553 0.00549 0.00561 0.00561 0*00.562 0.00566 0.00569 0.00515 0.00576 0.00578 4.0 7.1 11 '2 15.6 20.8 25.3 29.4 32-6 37 '4 45'7 48'9 0.00553 0-00556 0*00537 0.00532 0.00538 0.00539 0 -00543 0 '00536 0-00551 0 -005 60 0*00565 Range 9 to 73 %.Mean 0.00564 Range 7 to 82 %. Mean 0.00546 Mean of the two experiments, 0.00555. 7. Diaxo-rn-mitrobenxene Chloride. The diazo-salt was prepared from m-nitraniline exactly as in the case of o-nitraniline. Temperature 809 1, A= 59.9 (13", 753 mm.). 2. As59*9 (13', 763 mm.). t. C, C: t. (E. &. 74 81 99 148 164 181 212 230 254 277 9s 60 70 85 100 120 157 168 206 219 0*00303 0'00307 0.00306 0*00308 0 *003 12 0.00319 0*00328 0*00313 0'00319 24 '1 26 '1 29.1 ao ~6 39 -5 41.7 44.6 47.7 49 -5 50% 52 *6 0'00302 0.00307 0.00310 0.00314 0 *00316 0.00315 0.00326 0.00326 0'00330 0-00313 0*00330 20'5 234 26 9 30-4 34.6 41 '0 42'7 46.3 47.9 Range 40 to 88 %.Mean 0'00317 Range 34 t o 80 %. Mean 0'00312 Mean of the two experiments, 0*003145.1432 m 53 d 2 4 z 3 14 14 CAIN AND NICOLL: THE RATE OF Extreme values o f t . 4 to 25 2 to23 Temperature 1009 C. Highest. Extreme values of x. Series. Mean. 0'0322 0-0329 0.0325 Lowest I. A = 58'2 C.C. (So, 758 mm.) 11. A=58'2 C.C. (8", 758 mm.) 14.8 C.C. t o 49.6 c.c (range 25 to 83 %) 8.3 C.C. to 48'2 C.C. (range 14 to 81 %) 0.0315 0.0324 0.0332 0.0334 Mean ... 8. Diaxo-p-nitrobenzene Chloride. This diazo-salt was prepared from p-nitraniline exactly as in the case of o-nitraniline. Temnperatz~re 80".1. A=58'3 ( 8 V , 758 mm.). 2. A=58-3 (8*5', 758 mm.). -- t. C. C. t . X. 0. 22 25 28 32 36 41 46 53 66 76 a2 96 0'00784 0 -00748 0'00748 0*00756 0 -00750 0.00751 0*00730 0.00734 0.00714 0*00728 0.00737 0.00742 25 28 32 36 41 46 53 58 66 71 89 96 19.9 22 '0 24'6 26-9 29 '6 31'5 34.7 36'6 39.0 40'4 45 -1 46.5 0.00726 0-00735 0'00744 0.00747 0.00751 0.00731 0 -0074 1 0 '00740 0.00727 0.00722 0.00725 0-00723 18.1 20'4 22'3 24'9 27.0 29.6 31.4 34.5 38 -6 42 -0 43'8 47.0 Range 31 to 80 %. Mean 0.00739 Range 34 t o 80 %. Mean 0'00734 Mean of the two experiments, 0-00736. The foregoing figures show that, in each case, the reaction is a unimolecular one, and before describing our experiments on the remaining substances, namely, aminoacetanilide and the diamines of the benzene series, we shall now discuss the results up to this point, as the decomposition of the diazo-salts of the latter substances is not quite so simple.I n most of the above series of experiments, the valueDECOMPOSITION OF DIAZO-COMPOUNDS. PART I. 1433 of C seems to rise to a maximum at about the point of 50 per cent. decomposition. This is no doubt due t o the slight increase of tem- perature in the flask (sometimes 0.5' to 1") produced by the heat evolved in the reaction. Before proceeding t o compare these constants more closely, we must here .remark that the solutions of the diazo-salts from aniline, the toluidines, and sulphanilic acid mere all equivalent, that is, contained the same number of molecules. The amount of free hydrochloric acid was the same, but there was an extra mol.of sodium chloride present in the diazo-solution of sulphanilic acid, as we started from the sodium salt. I n the case of the diazo-salts from the nitranilines, although their diazo-solutions were all exactly equivalent among themselves, yet it wag necessary to use a larger amount of hydrochloric acid, and it was found more convenient to make up the solutions to half the concentra- tion of the previous ones. We think that we are quite justified in comparing the constants of the diazonitrobenzene salts with those of the other substances, as Hantzsch has shown very clearly (Zoc. cit.) that difference in concen- tration has no effect on the rate of the reaction in the substances which he examined, and this is, of course, applicable to all unimolecular reactions.Hantzsch has also shown that the presence of hydrochloric acid has no influence on the rate of decomposition. Comparing first of all the values for C for the three diazotoluene chlorides and diazobenzene chloride, we find that the introduction of the methyl group in the ortho- or meta-position in the latter makes the compound less stable, but in the para-position renders i t much more stable. Of the diazo-o- and diazo-m-toluene chlorides, the m-compound is the less stable. This conclusion is the same as that at whkh Hausser and Muller arrived (Bull. Xoc. Chirn., 1893, [iii], 9, 353). The introduction of the acid groups SO,H and NO, has the effect of enormously increasing the stability. Diazosulphanilic acid is much more stable than diazo-p-toluene chloride, and all three diazonitro- benzene chlorides are much more stable than diazosulphanilic acid.I n the case of the diazoriitrobenzene salts, the most stable is the ortho-, then comes the meta-, and lastly, the para-cornpound. The influence of the position of the nitro-group on the stability of the diazo-compound is thus quite different from that of the methyl group. We have collected the values for C in the following table and curve (p. 1434) : VOL, LXXXL Z D1434 - Temp- erature. 20" 30 40 50 60 70 80 90 100 Aniline. 100 90 80 70 60 50 40 < 30 20 10 0 Toluidine. Sulph- anilic Ortho. 1 Meta. I Para. 1 acid* CAIN AND NICOLL: THE RATE OF Ortho. Meta. -1:- - - 1 - l - - - I - Diazo-salts from Para. - - - - 0'000999 0'00358 0.012 0 *00072 0 '00295 0.00877 0'0298 0*109 - - - - 0'00633 0*00187 0 -00 706 0'0238 0.1027 - - - 0.00208 1 0 '00696" 0.0257 - - - - * This figure appears to be slightly too low.CURVE No. 6. 0.02 0.04 0.06 0'08 0'10 0'12 Before we can compare the stability of, for iniltance, diazo-p-nitro- benzene chloride with the diazochlorides from aniline and the tolu- idinee, it is necessary t o know whether the ratio of the constants at different temperatures is the same.DECOMPOSITION OF DIAZO-COMPOUNDS. PART I. 1435 S ~ l p h - By putting the value of C for diaaobenaene chloride at each tempera- ture equal to unity, and calculating the corresponding values of C for the three diazotoluene chlorides, we obtain the following : Nitraniline. Temperature. Ortho. Meta. 20" 30 40 50 60 Para.Diazobenzene I Diazo-o-toluene Diazo-m-toluene Diazo-p-toluene chloride. chloride. I I chloride. , chloride. 1 2.6 1 1 2-4 1 2.7 1 1 , 3.4 I I - I 2.9 2.36 * 2.9 - - 0 *07 0.11 0.12 0.11 * See note on previous page. We shall, therefore, not be far from the truth if we compare diazo- pnitrobenzene chloride (at 80') with diazosulphanilic acid (at SOo), diazosulphanilic acid (at 60') with diazobenzene chloride or diazo-p- toluene chloride (at 60°), and hence obtain a comparison between the stabilities of these diazo-sat ts. Carrying this out, and putting the value of C equal to unity for each substance successively, we obtain the following numbers, which show the relative rate of decomposition of each diazo-salt for any given temperature. (For comparing diazobenzene chloride and the diazo- toluene chlorides, the constants at 40' have been used.) Diazo-salts from .Aniline. - 1 8.7 16 '6 160 378 2195 Toluidine. 1 2 7 23'8 4 5 ' 1 435 1028 5962 1-08 2.9 25-7 48'7 470 1110 6436 - 1 1.9 18-3 43 -1 250 - 1 9-6 22-8 132 - 1 5.8 I 1 2 -36 13.7 The very great stability of diazo-o-nitrobenzene chloride is thus seen at a glance; i t is more than 6000 times as stable as diazo-m-toluene chloride, the rate of decomposition of the latter being 6436 times greater than that of the former. We may here remark that Hantzsch 5 ~ 21436 CAIN AND NICOLL: THE RATE 0% 0*00301 1 ' 35 56 69 95 19'4 0.00184 94 2o I 1;:; I 0.00231 392 0-00064 334 1238 i ~ 0.00034 1119 found diazo-p-toluene chloride to be about 8 times more stable than diazobenzene chloride ; this agrees with our result (8.7).By using this table and the table of constantg, it is easy to find what percentage of any of the above diazo-salts would be decomposed in a given time. For instance, if a' solution of diazo-p-nitrobenzene chloride is prepared and kept at 20' for 1 week, the amount of decom- position may be calculated as follows. The constant a t 20' would be the constant for diazobenzene chloride (0*00072) divided by 160 = 0.0000045 ; t = 1 week = 10080 minutes, so we have : 10.5 1 0 00245 17'0 0.00215 19.3 0*00;84 25.5 1 0.00074 37 -1 0-00039 1 A . c = - log - t A - x ' log ____ loo whence 1 0*0000045 = - 10080 100-x' x = 9.9; thus, 9.9 per cent. of thediazo-solution a t 20' would be decomposed in one week. 9-Diaxo-p-cLcetclminobenzene ChZoride.Fifteen grams of p-amineacetanilide ( l / l O mol.) were dissolved in 30 C.C. of hydrochloric acid of sp. gr. 1.16 (3/10 mol.), the solution cooled, and 6.9 grams of sodium nitrite added. The solution thus obtained was made up to 1400 C.C. and contained 19-75 grams of diazo-salt per litre. We have measured the rate of decomposition of the diazoacetanilide at 80' and looo, and a t each temperature find that the value of C diminihes very rapidly. Temperature 80".DECOMPOSITION OF DIAZO-COMPOUNDS. PART I. 1437 Temperature 100". A=58'6 (SO, 751 mm.). - __ t. 2. ~ ~ 3 8.9 10 18.1 47 27-7 173 39.4 282 45-3 472 51 '6 ____.-____ C. 0 0238 0.016 0.0059 0'0028 0'0023 0 0019 The reaction is thus not a unimolecular one, being obviously com- plicated by the simultaneous decomposition of the diazo-salt and elimination of the acetyl group by the free mineral acid.By using acetic acid instead of hydrochloric acid, the hydrolysis does not take place, and the reaction proceeds in accordance with the law. The p-aminoacetanilide was diazotised as before, but 3/10 mol. of acetic acid was used in place of the mineral acid. Temperature SO". t. 3 4 6 10 12 14 17 20 23 28 33 36 40 48 60 84 X. 4'5 6-3 9'5 12% 14'2 16'3 18.5 21'3 24.3 27'3 31'6 35'2 36.8 38'8 42'2 46'3 C. 0'0110 0.0117 0'0121 0'0117 0'0114 0*0111 0*0111 0'0108 0'0109 0'0111 0.0112 0~0111 0'0110 00108 Oo0104 0'0101 Range 7 to 75 %. Mean 0*0111. t. 6 10 12 14 1 7 20 24 28 36 62 a ~ x. 8'6 11 *1 13'2 15'1 17-1 20'3 23'5 27 *5 31 -1 37 '4 48'5 0'0109 O'G108 0'0105 0'0102 0'0101 0'0102 0-0104 0*0107 0'0109 0-0113 0'0108 Range 1 4 t o 78 %.Mean 0-0106. Mean of the two experiments, 0'01085.1438 CAIN AND NICOLL: THE RATE OF 100 140 189 232 420 The diazoacetate is thus about 6.5 times more stable than diazo- sulphanilic acid. I n examining the tetrazo-salts prepared from the diamines, benz- idine, tolidine, dianisidine, and oo-dichlorobenzidine, we have ob- served singular differences in their behaviour. It is obvious that there might be a difference in the stability of the diazo-groups of any one substance, If this were the case, the re- action could not proceed unimolecularly, as an intermediate substance mould be formed containing a single diazo-group, which would then proceed to evolve nitrogen. The effect of this on the value of C (calculated as if the formula for a unimolecular reaction applied in this case) mould be either t o diminish or increase it.I n two cases, namely, the tetrazo-salts from benzidine and tolidine, the value of C gradually diminishes; in the case of the tetrazo-salt from dianisidine, the value increases, but in the case of the tetrazo- salt; from oo-dichlorobenzidine, the value is constant, showing that the two diazo-groups in this substance are decomposed a t the same rate. 22.1 0-00203 100 21 -9 0 *o 020 1 33.0 0.00187 189 33.2 0 -001 89 36'5 0.00179 232 37'0 0'00184 28.0 0~00200 140 1 27.8 0.00198 45.5 0.00151 420 1 45'4 0*00161 I 10. Tetraxodipr7lenyZ Chloride. 9.2 Grams of benzidine (1/20 mol.) were dissolved in 30 C.C. of hydrochloric acid of sp.gr. 1*16 (3/10 mol.), the solution cooled, di- azotised with 6.9 grams of sodium nitrite, and made up t o 1400 C.C. Thirty-five C.C. were taken for the experiments : Temperature 60'.DECOMPOSITION OF DIAZO-COMPOCNDS. PART I. 1439 Temperature 60". A=59*0 (lo", 754 mm.). 33 53 71 95 248 23.3 31 '2 35.7 40.2 50.5 0.0066 0.0062 0.0057 0 0052 0 '0034 The reaction is thus not a unimolecular one, as the vaIue of C is not constant. We are of opinion that the intermediate compound, diazoy -h ydrox ydi pheny 1 chloride, OH* C6H,*C, H, *B ,GI, is continually being formed and decomposed. 11. Tetraxoditolyl Chloride. The solution of this salt mas prepared as the preceding one, substi- tuting 10.6 grams of o-tolidine for the benzidine : Temperature 50".t. 36 57 79 104 144 309 I 23'6 30.9 36 '6 41 -2 45.1 52.3 0*0062 0-0057 0.0054 OmO051 0.0044 0.0031 36 57 79 104 343 385 X. 22 *5 30.6 36-5 40.5 43'8 51 '2 C. 0.0058 0 0056 0.0054 0 %04 9 0-0042 0.0023 Here, again, the reaction is not unimolecular, probably due to the formation and decomposition of the intermediate diazop-hydroxydi- tolyl chloride, OH*C,H,Me*C,H,~~e*N~*cl. 12. Tetruxo-oo-dichlorodiphenyl Chloride. The solution of this salt mas prepared as No. 10, substituting 12.6 grams of oo-dichlorobenzidine for the benzidine :1440 THE RATE OF DECOUPOYI'I'ION OF DLAZO-COMPOUNDS. Tenzperature 10 0" A=59-9 (9*5", 741 mm.). C. t. X. 9 20 46 109 161 185 202 3 *9 7.5 15.2 31'1 38.9 43'0 45.3 Range 6.5 to 75 %. Mean 0'00294. 0'0033 0.0029 0.0028 0.0029 0.0028 0'0080 0-0030 'l'he decomposition in this case follows the lam.* 1 3. Telraxo-oo-dirnetlhoxydiphenyl Chloride. The solution of this salt was prepared as No. 10, substituting 12.2 grams of dianisidine for the benzidine : Temperature 100". 1. A=59*2 (lo", 752 mm.). t. i x. 69 27 -7 80 38'4 87 1 35.2 119 45 -5 149 1 53'0 C. 0.0035 0.0037 0.0040 0'0043 0.0045 0'0053 OYjO65 2. A = 59.1 (lo", 753 mm.). t. 61 73 92 104 116 129 156 2. 22.6 28.0 35-6 39.8 43 '4 47 0 51.9 C. 0.0034 0.0038 0'0043 0.0047 0'0050 0.0053 0 -0058 This reaction also is not unimolecular. As the tetrazo-salt from dichlorobenzidine is the only salt which decomposes in accordance with the law, we can compare its constant only with those of the foregoing table. It is seen that its constant at 100' (0.00294) is less than that of o-nitroaniline (0-00555), so that it is 1.9 times more stable than the latter. * The product of decomposition is not the dihydroxy-compound, as in the caw of benzidine and tolidine. I am investigating this at present.-[J. C. C.]CAMPHORSULPHONIC ACID (REYCHLER). 1441 As regards the remaining substances which do not conform to the law, we can of course only very roughly compare them among them- selves. We may say that the stability of the tetrazo-salt from diztnisidine is of the same magnitude as that from dichlorobenaidine; then comes the tetrazo-salt from benzidine, about as stable as diazosulphanilic acid, and, lastly, the tetrazo-salt from tolidine is about five times as stable as diazobenzene chloride. We are at present engaged in studying the decomposition of diazo- salts of the naphthalene series. We wish to thank Messrs. Levinstein, Ltd., Manchester, for kindly supplying us with the pure aminoacetanilide and dichlorobenzidine used in this work. CHEMICAL LABORATORY, MUNICIPAL TECHNICAL SCHOOL, BURP, LANCASHIRE.