2782 PEACOCK : ROTATORY POWER AND REFRACTIVITP. PART I.CCLXL-Rotatory Powe?. and Refractivity. Pa9.t I.The Rotatory Powers, Refiactivities and MoleculaySolution-volumes of Cinchonicine and Boyneol inCertain Solvents.By DAVID HENRY PEACOCK.THE close connexion between the physical expressions deduced onthe electromagnetic theory of light for the optical properties of asubstance has long been noticed. I n practice it is found that sucha molecular structure as is connoted by the term “conjugateddouble bonds ” produces an exaltation of the magnetic and opticalrotatory powers axid of the refractivity. An investigation of therotatory powers, rtfractivities, and molecular solution-volumes wascommenced by the author in order t o discover if there was any* In Koblrausch’s expeiiments the maximum specific conductivity occurred at aconcentration of 2 5 N PEACOCK : ROTATORY POWER AND REFRACTIVITY.PART I. 2783connexion between the two optical properties. While this workwas in progras a series of papers appeared by Livens (Phil. Nag.,1913, [vi], 25, 817 et seq.), in which on theoretical grounds anexpression was deduced connecting refractivity and rotatory power.If n is the refractive index of a solution of density d containing asolut’e in concentration c, then:7ll2 - 1 --- = ?‘C + s(d - c) . . . .a(n2 - 1) + Iniid if [a’] =specific rotation per :Init leiigtli for light of frequeucyg, then :where T and 9” and a. are constants depending on the solute, a i ds is a constant depending on the solvent. Then we have:Tfp%[ Q f ] = 7 (d - 1) ( r 2 - 1 + l/a) 0 .. . L (3)These equations are all given in Livens’ paper. From this lastequation it will be seen that specific rotatory power should vary inthe same manner as the expression ( d - l)(n2- 1 + l/a), No datahave been hitherto available to test this, and in the present paper afew simple cases are considered; the work is still in progress, andmore data are being collected.From the first equation it will be seen that i t should be possibleto calculate the constant, a from the refractivity alone. It is, how-ever, more easily calculated from equation (3). I f we put, c=O inequation (1) then we get:an equation connecting a with the constants of the pure solvent.On the other hand, putting c=lC)O per cent.in (1) we get anexpression for the refractive index of the solute a t 100 per cent.concentration. The quantities can also be deduced from the rotatorypowers, since we have from (3):Therefore if we know the value of [a,] a t a particular concentra-tion c and of m,, the refractive index a t the same concentration,then [a,] or n, a t a concentration x can be found if either is known.I n very dilute solutions it is difficult accurately t o measure [a],and so from the above equation it can be deduced as the refractiveindex is susceptible of careful measurement ; thus trustworthyvalues of [a] a t infinite dilution can be obtained. Similarly, fromthe value of [a] in concentrated solutions the value a t zero dilution8 ~ 2784 PEACOCK: ROTATORY POWER AND REFRACTIVITY.PART I.can be found by extrapolation, and from this the value of ?z underthe same conditions deduced. This quantity may aitain someusefulness, as hitherto the refractive indices of dissolved substanceshave been almost entirely deduced from the additive formula:d nZO - 1d, n2s -I- 2 do nZ0 + 2- --__- M n2 - 1 Jf -I- I n2sq- I - = -__- -~where iK + 1 is the weight of solution containing the gram-molecularweight M of the solute, I is the weight of the solvent of density doand refractive index t h o , and d , and I t , are those quantities f o r thesolvent.From equation (5) it will also be seen that the specific rotatorypower can only be independent of the concentration in cases wheretha refractive index of the solution is independent of the concen-tration ; in other cases there should be variation with concentration,although the amount of this variation cannot at present be pre-dicted.Pope and Gibson (T., 1912, 101, 1702) describe certain deriv-atives of d-ax.-butylamine which show practically no rotatorydispersion. It would be interesting to examine also the refractionconstants of these derivatives in the light of equation (3) givenabove.According to the theory on which the equations given arededuced, the effect' of concentration on specific rotatory power isexplained solely by the variation in the velocity of transmissionof light waves in the solution.When a beam of plane polarisedlight is sglit up on entering an active solution into two beams ofoppositely circularly polarised light travelling a t different velocitiesthe retardation, t o which the rotation of the plane of polarisationis due, is shown by this theory to depend on the optical propertiesof the solvent, and not merely on the active solute.Previoustheories have explained the effects of concentration and of changeof solvent' by (1) electrolytic dissociation, (2) formation of molecularaggregates, and (3) changes in molecular symmetry due to theeft'ect of the internal pressure of the molecule. I n the presentpaper the data are not' very numerous in the cases where a changeof specific rotatory power with concentration occurs, but it may bepointed out that all the above theories ascribe the change toan actual variation of the forces acting within the active molecule,that is t o say, to a change in r' (equation 3). I f , therefore, theeffect of concentration is due t o the above causes, then equation (3)will no longer hold as r' will depend on c, and therefore [a] willnot be proportional to the quantity (122- l ) ( n 2 - 1 + 1 /a).Theresults given in the present communication show that for the solu-tions examined [ a] is, within reasonable limits, proportional t o thPEACOCK: ROTATORY POWER AND REFRACTIVITY. PART I. 8785refraction expression, and therefore in those cases where [a] dependson concentration the dependence is probably not due t o the mole-cular changes usually invoked to explain this dependence, but isdue simply "to the: variation in the velocity of transmission oflight."The rotatory powers of the solutions examined were measured in2-dcm.tabes a t 25O, sodium light being used. The solutions weremade up in 25 C.C. flasks a t 25O. The refractive indices of thesolutions were measured in a Zeiss total reflection refractometer, asis used for examining oils technically. The densities were measuredin pyknometers holding about 10 C.C.Borneol, C,,H,80.This was recrystallised from light petroleum. As also found byVanstone (T., 1909, 95, 600), i t was partly racemised. This doesnot alter the deductions in the present papw, as only ratios ofrota t o.ryP.1.01622.7363.5865-2907.5419.96724.33035.070P1.03022.98055.07309.994015.322powers are dealt with.TABLE I.AlcoJtol Solutions.a".0.275"1.121.462.153.114.1510-6515.70ny.1.36131.36351.36361.36591.36821.37111.38721-3993(156.0.78890.79210.79340.79440.79980.80400.82780.8466[U]':.(27.4') - 1.0915(72; - 1) (%a - 1 +/In).25.8 1.09225-6 1-09325.5 1.09525.7 1.09925.9 1.10226-4 1-11726.4 1.124p=conceiitration in grams per 100 grams.TABLE 11.Acetone Solutions.8 v,n.C.C.[a]%. (Pi - 1) (1L;4- 1 + l / o ) d2 34 .,ZLiD .as60.45" 1.3583 0.7860 158 27.6" - 0.933D '1.26 1.3613 0.7925 160 26.3 0-9342.31 1.3623 0.7960 158 28.6 0.9354.62 1.3675 0.8035 158 25.7 0.9376-73 1.3739 0.8113 159 27.0 0.9418-73 1,3791 0.8184 (175) 27-1 0.943 19.65237.697 17.47 1-4005 0.8493 '159' 27.2 0.942786 PEACOCK : ROTATORY POWER AND REFRACTIVITY. PART 1.&? 5 P.D .0.8699 0.44"2.2461 1.124.433 2-258.951 4.5417.663 9.1826.301 13-72,a 5 P. D *0.9238 0.43'2.3214 1.244.886 2.3611.348 5.78TABLE 111.Ethyl A cetate Solutions.1 L : j . di j. C.C. [a]:. (78; - 1) ( I L 6 - 1 + l / O ? ) .'V*.1.3711 0.8953 161 28.2" -0.9311.3719 0.8956 162 28.4 0.93 11.3761 0.8976 160 28.2 0.9331.3790 0.9004 161 28.1 0.9341.3891 0.9063 160 28.6 0.9371.3983 0.9127 160 28.5 0.938TABLE IV.B e n z e n e Sol& ions.a vnt.1.4995 0.8737 154 (26.6') 4-0.4661.4992 0.8747 159 30.5 0.4641.4985 0.8765 160 29.3 0.4601.4973 0.8813 161 28.8 0.454C.C. [a]:5. (4 -- 1) (92; - 1 + lh]. ,>?. di.5.16.940 8-59 1.4960 0-8864 161 28.6 0.44822.455 11.30 1.4953 0.8909 161 28.2 0.445The solvents used were not specially purified, but were preparedin such quantities that the same salnple could be used throughouta wries of experiments. The benzene and acetone were preparedfrom technical material by drying and fractional distillation.Thefollowing are the constants :TABLE V.Solvent d l i . n2'.Alcohol ........................ 0.7878 1-3602Acetone ........................ 0-7881 1.3571Ethyl acetate ............... 0.8947 1.3701Benzene ........................ 0.8728 1.5004The specific rotatory power of borneol is seen to depend only toa small extent on concentration or solvent; the expression( n 2 - 1)(n2 - 1 + 1 /a), whilst varying with concentration in the Bitmeway as does [a], depends to a very great extent on the solvent; inthe case of benzene which has its refractivity diminished by a solu-tion of bornool the expression has a sign opposite to that for theother solvents examined.Experiments now in progress show thatthis is not a general effect.. Livens (Zoc. cit.) leaves it apparentlyan open question as to the dependence of a on the solvent. Themresults show that n decidedly depends on the nature of the solvent;there is no very apparent connexion between the value of a and therefractivity of the solvent.It is, however, quite well established that for the siinplesb casePEACOCK: ROTATORY POWER AND REFHACTIVITY. PART I. 2787namely, that in which [a] varies very little with concentration, thevalue of (n2 - 1) ( n 2 - 1 + 1 / a ) also shows practically no variation,although over the range of concentrations used the values of n 2 - 1and n2 - 1 In2 + 2 vary considerably.--Cinchomkine.This was prepared according to Miller and Rohde's method (Ber.,1895, 28, 1056).It was purified first by crystallisation of theoxalats from water and then by crystallisation of the base from dryether. Only sufficient was available for a limited examination.TABLE V I .Alcohol Solzctions.C. af".0.1232 0.125'0-6240 0.590.9528 0.902.8900 2.674-4612 4-098.5548 7.699.3000 8.328 13%.71;s. d?"". C.C. [a!:'. (mi - 1) (72: - 1 -i- I / u ) .1.3607 0.7887 75 50.5' -0.4131-3617 0.7910 179 47-3 0.4131.3630 0.7920 222 47.2 0.41 11.3667 0.7984 238 46-2 0.4071-3695 0.8032 243 45.8 0-4041.3796 0.8169 247 44.9 0.3921.3800 0.8186 259 44.2 0.390C.a;6 *0.6160 0.58'2.1136 2.174.1592 4.056.0348 5.928.8712 8.60TABLE VLI.Acetone Solutions.8 vWVn;f". d i e , C.C. [a;']. (ni -- 1) (n: - 1 + lia).1.3579 0.7898 249 56.2' -0.4381.3614 0.7951 244 50.2 0.4351.3662 0.8015 253 48.7 0.4301.3711 0.8100 238 49.0 0-4251.3774 0.8152 260 48.5 0.418c=concentration in grams per 100 C.C.Rques (Compt. rend., 1895, 120, 1170) gives [a]= in alcohol forcinchonicine as 48*25O, but gives no account of a dependence onconcentration. The above results show that in both alcohol andacetone [a] decreases with increasing concentration. I n both cases[a] and (122- 1)(n2- 1 + l l a ) are very closely proportional; thus forthis simple case of dependence on concentration Livens' suggestion,that ths dependence is due simply to the variation in optical proper-ties of thel medium, seems t o be true.If the dependence had beendue to dynamic isomerism or a similar cause then [a] andwould not have varied together as has been explained in theintroduction.(n2- l)(n2- 1 + I / a 2788 PEACOCK : ROTATOKY POWER AND REFRACTIVlTY. PART I.For alcohol solutions the valum of sV,,,, increase with concentra-tion. I f this were a real and not an apparent phenomenon theni t would be expected that the corresponding deformation of themolecule would lead t o considerable variation in [a]; but a deforma-tion of the molecule would lead to a variation in 7" (see equation 3),and [a] and ( tz2 - 1) ( t z 2 - 1 + 1 / a ) would not vary together, whilst inactual experiment they are very closely proportIona1.The varia-tion in S V m is therefore probably, at least in part, fictitious, anddue t o alterations in the volume occupied by the alcohol molecules.Thus by this method $he effect of an active solute on the molecularvolume of a solvent can be at any rate qualitatively examined.Since at present there does not seem to: be any mathematical theoryf o r the mutual effects on density of solvent and solute, this indirectmethod seems the only mode of attack; the theory of solution inthis respect lags far behind the optical theory.,Passing from alcohol t o acetone both [a] and(122-l)(n2-1+l/a)increase.A reference to table X shows that a for these two solu-tions is approximately constant. I n the case of acetone, althoughthe variation of [a] with concentration is as great as that foralcohol solutions, the values of SVm for acetone solutions show nosuch great dependence on concentration.~enzoylci?zchoniciize.Solutions of this substance were examineld in order to see whethera or (n2 - 1) (n2 - 1 + 1 / a ) showed any constitutive effect.c.0.94482.52804.158462128C.0.97042.1028aI,.0.736"1.933-254.86a,.0.88"1.80TABLE VIII.Acetone Solutions.s J7m.? I D . d ? . [a],,. C.C. (n: - 1) (n;", - 1 t l/a).1.3581 0.7916 38.9" 322 -0.8031.3623 0.7973 (38.2) 322 0-8041.3651 0-8033 39.1 322 0.8041.3726 0.8102 39.2 325 0.806TABLE IX.L41cohol Solutions.s I.',.77 1).dt6. [a],. C . C . (?lt - 1) (n; - 1 +- lh).1.3621 0.7911 45.3" 329 -0.9321.3647 0.7984 42.8 320 0.88PEACOCK : ROTATORY POWER AND REFRACTIVITY. PART I. 2789TABLE X.Values of a for the Solutions Examined.Substance. Solvent.Cinchonicine ............ Alcohol9 ) ............ AcetoneBenzoylcinchonicine ... Alcohol1 9 ... AcetoneBorneo1 .................. AlcoholAcetoneEthyl acetateBenzenea.- 0.747-0.733- 1.036- 0.557- 0.468-0.513-0.615- 1.1411 /a.- 1.337- 1.363- 0.964- 1.795-2.132- 1.948- 1.938- 0.876Beiizoylcinchoiiicine shows a decrease in both [a] and(d- l)(nz- 1 + l/a)on passing from alcohol to acetone, the reverse of 'the case forcinchonicine.There is no very apparent constitutive connexionbetween the value of n o r (n.2- l)(n2- 1 + l / a ) for cinchonicine andits benzoyl derivative ; both of these quantities depend too greatlyon the nature of the solvent. However, the results obtained withthe acetone solutions confirm the interdependence of [a] and(722- 1 ) ( d - 1 + 1 /a). I n passing from alcohol to acetone the valuesof [a] for benzoylcinchonicine decrease considerably, whilst thevalues of ,V, show no very great variation.For the cases examined in the present paper confirmation hasbeen obtained of the theory that where variations in [a] occur theyare due to variations in the velocity of transmission of light withinthe medium, and not t o causes directly affecting the degree ofasymnletry of the molecule. It has, furthermore, been shown howevidence can be obtained as to the effect of a solute on the densityof a solvent.Further work is in progress on these lines, and with the addi-tional object of obtaining a value for the refractivity of an activesolute.I n conclusion, the author wishes to express his thanks to theNobel's Explosive Company, Ardeer Factory, and t'o Mr. W.Rintoul, F.I.C., Manager of the research section, for the facilitiesafforded f o r carrying out this work and permission t o publish theresults.THE RESEARCH SECTION,NOBEL'S EXPLOSIVES COMPA~ Y,A RDEKI