238 RICHARDS AND LOWRY THE ROTATORY DISPERSIVE XL. - The Rotatory Dispersive Power of Organic Compoundg. Part X I V . Simple Dispersion in l-Methylc yclohexylidene-4-acetic Acid. By EVAN MATTHEW RICHARDS and THOMAS M~ELTIN LOWRY. SPECIAL interest attaches to the rotatory dispersion of centro-asymmetric compounds which (according to the ordinary definition) contain no asymmetric atom. Through the kindness of Professor Pope we have recently had the opportunity of studying from this point of view his own specimens (compare J. 1909,95,1789) of the d- and 2-forms of the well-known acid, On Professor Pope's recommendation the measurements were made with solutions in methylal. The volatility of this solvent made it necessary to express the results in the form of dispersion-ratios, instead of by means of specific or molecular rotations since the solutions were originally cloudy and their concentrations were altered appreciably by filtration ; but we had the very great advan-tage of being able (except in the photographic region) to read a dextro against a l ~ v o solution under precisely similar conditions, without being compelled to make use of zero-readings.In these circumstances a very satisfactory agreement was obtained between the results recorded in two complete series of independent readings. The data are set out in Table I together wifh the corresponding dispersion-ratios calculated from the equation "/a5461 = 0.2422/ (A2 - 0.056). Since the value of 45461 was about 20" for the visual and 10" for the photographic observations an error of 0.001 in the dispersion-ratios corresponds with an error of 0.02" in the visual readings and of 0-01" in the photographic readings.The largest visual error was therefore about 0.2" (for a dBcult dark-blue line) and the largest photographic error about 0-3". The average error in the visual ratios is 0.003 and in the photographic ratios 0.02; the negative and positive errors are moreover dis-tributed in small groups and afford no evidence of systematic deviations from the formula such as were observed in the case of octyl oxalate. The data can be expressed over the range of wave-lengths covered by our observations and up to the existing limits of experi-mental accuracy by one term of Drude's equation. Moreover, since an exposure of more than an hour was required to give the last photographic reading it is clear that the measurements had been carried right up to the limit of the region of transparency withi POWER OF ORGANIC COMPOUNDS.PART XIV 239 TABLE I. Dispersion-ratios of 1 -MethylcycZohexylidene-4-acetic Acid in Methylal (about 6 gms. per 100 c.c.) at 20°.* Mol. wt. 154.17. a/a5461 = 0*2422/(h2 - 0.056). Sum of d and I w d First Secind -, Dispersion-ratios rotations. a la54 61' Line. series. series. series. series. Mean. Calc. Diff. Li 6708 12.81 11.00 0.611 0.602 0.607 0.615 -0.008 Zn 6362 14.53 12.56 0.694 0.688 0.691 0.694 -0.003 CU 5782 18.24 15.91 0.870 0.871 0.871 0.870 +O-001 Cd 6438 14-20 12-31 0.678 0.674 0.676 0.676 rt Na 5893 17.41 15.20 0.531 0.832 0.832 0.832 i Hg 5780 18.23 15-85 0.870 0.870 0.870 0.871 -0.001 CU 5700 18.82 16.45 0.898 0.901 0.900 0.901 -0.001 Hg 5461 20.95 18-26 1.000 1.000 1.000 1.000 * Cu 5218 23.48 20.47 1.121 1.121 1.121 1.120 +O-OOl CU 5153 24-29 21-19 1.160 1-160 1.160 1-156 +0-004 Cu 5106 24.89 21.64 1.188 1.185 1.187 1.183 +0.004 Cd 5086 25.19 21.85 1.202 1.196 1.199 1.195 +0*004 Zn 4811 28.87 - 1.378 - 1.378 1.381 -0.003 Cd 4800 29.14 25-32 1.391 1.386 1.389 1.389 3I Zn 4722 30.48 26.60 1.455 1.456 1.456 1.450 +0*006 Zn 4680 31.05 27.18 1.482 1.488 1.485 1.486 -0.001 Cd 4678 31.49 27.23 1.503 1.491 1.497 1.487 +O.OlO Hg 4368 37.92 32.96 1.810 1.805 1.808 1.808 & Photographic Series (a5461 = 9-54').Dispersion ratio. Line. Rotation. Obs. Cali. Diff. Fe 4376 17.11 1.79 1.79 f Fe 4337 17.36 1.82 1-83 -0.01 Fe 4308 17.61 1-85 1.87 - 0.02 Fe 4271 18.11 1.90 1.92 - 0.02 Fe 4261 18.36 1.92 1.93 -0.01 Fe 4236 18-61 1.95 1.96 - 0.01 Fe 4199 18.98 1.99 2.01 - 0.02 Fe 4187 19.23 2.02 2-03 - 0.01 Fe 4144 20.23 2.12 2.09 + 0.03 Fe 4132 20.36 2-13 2-1 1 + 0.02 Fe 4119 20.61 2.16 2.13 + 0.03 Fe 4072 21-50 2.23 2.21 + 0.02 * The temperature 20° was accidentally omitted from the corresponding tables for octyl alcohol and octyl oxalate in Part XI1 of this series (J.1924, 125 1595 1596). The opportunity may also be taken of correcting an error in Part XI (ibid. p. 1466 two lines from the bottom) where a shallow minimum " at 5 per cent." should be " a t 50 per cent." of ethyl tartrate. which alone Drude's formula is valid. As any deviations disclosed by pushing the observations beyond this limit would be irrelevant to the present discussion it appears that the simplicity of the dis-persion is not likely to be disproved by direct experimental measure-ments.Complexity could therefore be established only by the use of indirect tests such as that described in the previous pape 240 BRISCOE ROBINSON AND STEPHENSON : (p. 2511). In the present instance however as in the analogous case of octyl alcohol this test gives no clear answer to our question since the characteristic frequency of the dispersion-equation falls in a region in which absorption-bands are very difficult to detect. Thus the dispersion-constant A2 = 0.056 of Pope's acid corresponds to a wave-length X = 2364 A.U. only a little longer than that of the last strong line in the iron arc spectrum.Direct measurements of the molecular extinction coefficient of the acid in this region (Table 11) show however that the " general absorption '' (log Q = 3.9) of the acid at this wave-length is already more than 100 times greater than the maximum selective absorption (log C= 1.5) of ctimphor and of camphorquinone at the head of their absorption bands. It was therefore quite impossible to establish the existence of a band of selective absorption a t the wave-length indicated. In our opinion, however a negative result of this character does not provide valid evidence that the dispersion is complex; and until some positive evidence to the contrary is available we propose to dispersion of the compound as simple.TABLE 11. Molecular Extinction Coefficients. A = 2470 2440 2411 2375 2348 2338 log E = 3.31 3.61 3.78 3.91 4.01 4.09 describe the 2327 4.15 The acid contains only one unsaturated group namely the con-jugated system >C=CH-CqOH 0 . The characteristic frequency of this is perhaps given by the dispersion-constant of our equation. We desire to express our thanks to the Department of Scientific and Industrial Research for a maintenance grant to one of the authors (E. M. R.). UNIVERSITY CHEMICAL LABORATORY, CAMBRIDGE . [Received September 20th 1924. 238 RICHARDS AND LOWRY THE ROTATORY DISPERSIVE XL. - The Rotatory Dispersive Power of Organic Compoundg. Part X I V . Simple Dispersion in l-Methylc yclohexylidene-4-acetic Acid. By EVAN MATTHEW RICHARDS and THOMAS M~ELTIN LOWRY.SPECIAL interest attaches to the rotatory dispersion of centro-asymmetric compounds which (according to the ordinary definition) contain no asymmetric atom. Through the kindness of Professor Pope we have recently had the opportunity of studying from this point of view his own specimens (compare J. 1909,95,1789) of the d- and 2-forms of the well-known acid, On Professor Pope's recommendation the measurements were made with solutions in methylal. The volatility of this solvent made it necessary to express the results in the form of dispersion-ratios, instead of by means of specific or molecular rotations since the solutions were originally cloudy and their concentrations were altered appreciably by filtration ; but we had the very great advan-tage of being able (except in the photographic region) to read a dextro against a l ~ v o solution under precisely similar conditions, without being compelled to make use of zero-readings.In these circumstances a very satisfactory agreement was obtained between the results recorded in two complete series of independent readings. The data are set out in Table I together wifh the corresponding dispersion-ratios calculated from the equation "/a5461 = 0.2422/ (A2 - 0.056). Since the value of 45461 was about 20" for the visual and 10" for the photographic observations an error of 0.001 in the dispersion-ratios corresponds with an error of 0.02" in the visual readings and of 0-01" in the photographic readings. The largest visual error was therefore about 0.2" (for a dBcult dark-blue line) and the largest photographic error about 0-3".The average error in the visual ratios is 0.003 and in the photographic ratios 0.02; the negative and positive errors are moreover dis-tributed in small groups and afford no evidence of systematic deviations from the formula such as were observed in the case of octyl oxalate. The data can be expressed over the range of wave-lengths covered by our observations and up to the existing limits of experi-mental accuracy by one term of Drude's equation. Moreover, since an exposure of more than an hour was required to give the last photographic reading it is clear that the measurements had been carried right up to the limit of the region of transparency withi POWER OF ORGANIC COMPOUNDS.PART XIV 239 TABLE I. Dispersion-ratios of 1 -MethylcycZohexylidene-4-acetic Acid in Methylal (about 6 gms. per 100 c.c.) at 20°.* Mol. wt. 154.17. a/a5461 = 0*2422/(h2 - 0.056). Sum of d and I w d First Secind -, Dispersion-ratios rotations. a la54 61' Line. series. series. series. series. Mean. Calc. Diff. Li 6708 12.81 11.00 0.611 0.602 0.607 0.615 -0.008 Zn 6362 14.53 12.56 0.694 0.688 0.691 0.694 -0.003 CU 5782 18.24 15.91 0.870 0.871 0.871 0.870 +O-001 Cd 6438 14-20 12-31 0.678 0.674 0.676 0.676 rt Na 5893 17.41 15.20 0.531 0.832 0.832 0.832 i Hg 5780 18.23 15-85 0.870 0.870 0.870 0.871 -0.001 CU 5700 18.82 16.45 0.898 0.901 0.900 0.901 -0.001 Hg 5461 20.95 18-26 1.000 1.000 1.000 1.000 * Cu 5218 23.48 20.47 1.121 1.121 1.121 1.120 +O-OOl CU 5153 24-29 21-19 1.160 1-160 1.160 1-156 +0-004 Cu 5106 24.89 21.64 1.188 1.185 1.187 1.183 +0.004 Cd 5086 25.19 21.85 1.202 1.196 1.199 1.195 +0*004 Zn 4811 28.87 - 1.378 - 1.378 1.381 -0.003 Cd 4800 29.14 25-32 1.391 1.386 1.389 1.389 3I Zn 4722 30.48 26.60 1.455 1.456 1.456 1.450 +0*006 Zn 4680 31.05 27.18 1.482 1.488 1.485 1.486 -0.001 Cd 4678 31.49 27.23 1.503 1.491 1.497 1.487 +O.OlO Hg 4368 37.92 32.96 1.810 1.805 1.808 1.808 & Photographic Series (a5461 = 9-54').Dispersion ratio. Line. Rotation. Obs. Cali. Diff. Fe 4376 17.11 1.79 1.79 f Fe 4337 17.36 1.82 1-83 -0.01 Fe 4308 17.61 1-85 1.87 - 0.02 Fe 4271 18.11 1.90 1.92 - 0.02 Fe 4261 18.36 1.92 1.93 -0.01 Fe 4236 18-61 1.95 1.96 - 0.01 Fe 4199 18.98 1.99 2.01 - 0.02 Fe 4187 19.23 2.02 2-03 - 0.01 Fe 4144 20.23 2.12 2.09 + 0.03 Fe 4132 20.36 2-13 2-1 1 + 0.02 Fe 4119 20.61 2.16 2.13 + 0.03 Fe 4072 21-50 2.23 2.21 + 0.02 * The temperature 20° was accidentally omitted from the corresponding tables for octyl alcohol and octyl oxalate in Part XI1 of this series (J.1924, 125 1595 1596). The opportunity may also be taken of correcting an error in Part XI (ibid. p. 1466 two lines from the bottom) where a shallow minimum " at 5 per cent." should be " a t 50 per cent." of ethyl tartrate. which alone Drude's formula is valid. As any deviations disclosed by pushing the observations beyond this limit would be irrelevant to the present discussion it appears that the simplicity of the dis-persion is not likely to be disproved by direct experimental measure-ments.Complexity could therefore be established only by the use of indirect tests such as that described in the previous pape 240 BRISCOE ROBINSON AND STEPHENSON : (p. 2511). In the present instance however as in the analogous case of octyl alcohol this test gives no clear answer to our question since the characteristic frequency of the dispersion-equation falls in a region in which absorption-bands are very difficult to detect. Thus the dispersion-constant A2 = 0.056 of Pope's acid corresponds to a wave-length X = 2364 A.U. only a little longer than that of the last strong line in the iron arc spectrum. Direct measurements of the molecular extinction coefficient of the acid in this region (Table 11) show however that the " general absorption '' (log Q = 3.9) of the acid at this wave-length is already more than 100 times greater than the maximum selective absorption (log C= 1.5) of ctimphor and of camphorquinone at the head of their absorption bands.It was therefore quite impossible to establish the existence of a band of selective absorption a t the wave-length indicated. In our opinion, however a negative result of this character does not provide valid evidence that the dispersion is complex; and until some positive evidence to the contrary is available we propose to dispersion of the compound as simple. TABLE 11. Molecular Extinction Coefficients. A = 2470 2440 2411 2375 2348 2338 log E = 3.31 3.61 3.78 3.91 4.01 4.09 describe the 2327 4.15 The acid contains only one unsaturated group namely the con-jugated system >C=CH-CqOH 0 . The characteristic frequency of this is perhaps given by the dispersion-constant of our equation. We desire to express our thanks to the Department of Scientific and Industrial Research for a maintenance grant to one of the authors (E. M. R.). UNIVERSITY CHEMICAL LABORATORY, CAMBRIDGE . [Received September 20th 1924.