4016
RORYA . MOREO'FERRALL A N D SIDNEY I. MILLER
bital theory showed that here the coupling constants could be calculated by considering contributions from magnetic dipolar and electron-orbital terms only. 30 These are likely t o be important factors in any analysis of coupling involving only one fluorine also. The aPParent dependence of both fluorine-fluorine and Pro(30) H . M. McConnell, J . Chem. P h y s . , 24, 460 (1956).
(CONTRIBUTION FROM T H E
DEPARTMENT OF CHEMISTRY,
Vol. 86
ton-fluorine couplings on spatial factors suggests that a mathematical treatment of either phenomenon must also aid our understanding of the other. Achowledgment.-The author is indebted to Professor M . Karplus, Columbia University, for invaluable correspondence and discussion on sundry theoretical aspects of coupling mechanisms.
ILLINOIS ISSTITUTE O F
TECHXOLOGY, CHICAGO 16, ILL.]
The Transmission of Electronic Effects. The p-p Relation in the Reaction of Phenylpropionic Acids with Diphenyldiazomethane' BY RORYA . MOREO'FERRALL ASD SIDSEYI. MILLER RECEIVED FEBRUARY 18, 1964 Rate d a t a for the reactions of ten phenylpropionic acids (p-(CHa)?CH,p-CH,, p-CH30, 3,4-(CH10)2,H,
p-F, WHO, p-Br, m-C1, p-NO?) with diphenyldiazomethane ( D D M ) in ethanol have been obtained. The entropies of activation ( - A S * = 6-15 e . u . ) are proportional to theenthalpies of activation ( A H * = 12.7-15.4 kcal. mole-') in this series with a slope of CQ. 260'. Acceptable Hammett lines have been obtained for k in I. mole-' min.-' a t 26.05", log k = 0.1% - 0.226 and a t 35.35", log k = 0 . 1 7 ~ 0.1105. Taking p from the dissociation of the acids RGCOOH and p' from the rate constants of these acids with D D M , one obtains satisfactory test comparisons in the form p / p ' = constant. Alternatively, the attenuation factor for phenylpropionic acids or esters relative to corresponding benzoic derivatives is E = 0.22 for four different reactions. Despite their utility, i t does not yet seem possible to provide a rigorous theoretical basis for these p-p relations.
+
In this paper we continue our inquiry into the factors which affect the relay of electronic influences from a substituent R to a reaction site S in a species R-GiCOS. Previously, it was shown t h a t two substantially equivalent p-p relations PR-GCOOH~PR-GCOS = 7
PR-GCOS!
PRC~HCOS -
E
(1) (2)
were useful in examining the alterations in the Hammett p as the reaction type changed but the interposed group G remained the same (eq. l), or as the reaction type remained unchanged but the interposed group G varied (eq. 2).2-5 T is a measure of the ability of a substituent to communicate electronic effects to a given site for a given reaction (e.g., ester hydrolysis compared to a standard reaction, i.e., acid dissociations in water a t 26") and 6 is a measure of the attenuation of an electronic effect through G as compared with the group -C6H4-. 4-6 Together with acid dissociation of acids RGCOOH or the basic hydrolysis of their esters, the reactions of the acids with diphenyldiazomethane (DDM) have attained the status of a standard or probe process in structure reactivity s t ~ d i e s . ~ - ~T. o' test eq. 1 and 2 we have accumulated rate data on several acid families (1) Supported b y t h e U. S. Army Research Office ( D u r h a m ) . (2) H . H. Jaff6, Chem. R P V . 63, , 191 (1953). (3) R . A hI. O'Ferrall a n d S. I. Miller. J . A m Chem. Soc., 8 6 , 2440 (1963). (-1) J. D . S. R i t t e r a n d S . I. hfiller, ibid., 86, 1507 (1964). (5) 11'. K . K w o k . R . A . h l o r e O'Ferrall, a n d S. I. Miller. T e l r a h e d r o n , 2 0 , 1913 (1964). (6) M. C h a r t o n , J . Or#, C h e m . , 2 6 , 735 (1961). (7) J . D . R o b e r t s a n d R. A . C a r b o n i , ' J . A m . Chem. Soc., 77, 5551 (1955), I. Solomon a n d R . Filler, ibid.,86, 3492 (1963). J . Hine a n d U'. L. Bailey, ibid.. 81, 2075 ( 1 9 5 9 ) ; W. F. Little a n d R . Eisenthal, ibid., 83,4936 (1961); C K Hancock a n d J . S. Westmoreland, ibid.. 80,545 (1958), R . Benkeser, C E . de Boer, K . E Robinson, a n d D. h l . S a u v e , ibrd , 78, 682 (1956), R . U'. T a f t , J r . , a n d L). J. S m i t h , i b i d , 81, 207.5 (1959); K . B. C h a p m a n . J . S h o r t e r , and J. H . P. Utley, J . C k e m . Soc., 1823 (1962). A . Buckle)., N. B C h a p m a n . a n d J. Shorter. ibid., 178 (1963). 0 . E x n e r , Tetrahedroir I . e l l e i s . 815 (1963); A. T a v l i k . P. Z u m a n , a n d 0. E x n e r , Coilcclton Czuch. Chem ~ o m n i i c ~ i2. 9. , 1266 (19fi4J
in this reaction
Here rate data are reported for several phenylpropionic acids in process 3 and the problem of attenuation of the substituent effect by G is considered. Because its reaction site is "insulated" from the substituent, the phenylpropionic system is of particular interest in theories of the influence of structure on reactivity. Results and Discussion Rate data for the reactions of the phenylpropionic acids with DDM are given in Table I. For a wide range TABLE I
RATECONSTAXTS
AND dCTIVATION PARAMETERS FOR THE
REACTION:
-Airs* i 3.0, 1. mole-' min. - l a , b AH* & 1.0, cal. deg. -1 mole - 1 26.05 i 0 1' 36.35 i 0.1' kcal. mole-'
-k,
0 52 .i4
1.18 14.4 9 3 14.9 56 1.11 12 7 7 3 55 1.22 15.0 11 5 58' 1 21 13.7 .63 5.7 .54 1.22 15.4 6 4 15.3 .64 1.44 7 7 .68 1.50 15.0 12.7 .s5 1.77 13 6 h -4verage of three runs a Average deviation less than 2 5 except a s indicated c Average of four runs
of substituents the total variation in the rates is small a t 1 6 As in the DDM reactions with the phenyl-
acetic3 and t r ~ n s - c i n n a m i cacids ~ the activation entropies and enthalpies are proportional; here the slope
Oct. 5, 1964
PHENYLPROPIONIC ACIDSWITH DIPHENYLDIAZOMETHANE
4017
SS(r, ethanol, and reaction of DD-M with the acids in ethanol-leads to €-values of 0.41, 0.61, and 0.49, respectively3; K.W. theory leads to €-valuesof 0.52,0.77, and 0.77, r e s p e c t i ~ e l y . ~ ~ " ) ~ ~ ~ ~ The discrepancies between the experimental fall-off factor contained in the p-p relation and those produced by theory are not trivial but certainly no larger than what one has learned to expect from K.W. theory.1° TABLE I1 When a resonance interaction between the substituent LINEARCORRELATIOSS FOR THE REACTIOSS OF and the reaction site is possible, e.g., in the cinnamic PHESYLPROPLOSIC ACIDSWITH DIPHENYLDIAZOMETHANE acids, the discrepancy is somewhat larger.4 Even T e m p . , 'C "c"* P log kLn Sb IC fld when the theory was modified to correct for systems 26 05 u 0 18 -0 226 0 0030 0 96 10 with conjugation, l 3 the results were not wholly satisUn .21 - 243 0034 95 8 factory.4 35 35 u 17 1105 0029 .96 8 Both K.IV-.theory and the Dewar-Grisdale elaboraUn 22 859 ,0013 .98 8 tion of it give approximate estimates of the transmission a Intercept of Hammett equation. Standard deviation. S u m b e r of compounds used in corof electronic effects in chemical systems. In taking c Correlation coefficient. relation. into account detailed changes in molecular geometry ( r , 0) and medium ( D E ) ,K.W. theory predicts t h a t for Both correlations are adequate although a Hammett any given G in RGCOS the attenuation factor ( E ) plot, not shown, reveals typical scatter. The p based should vary with the medium.l0 On the other hand, on the present D D M data holds up well when tested the p-p relations, eq. 1 and 2 , were derived on the basis in eq. 1 ; figures showing the development of this p-p rethat E does not vary with the reaction or the medium.3 lation are given e l ~ e w h e r e . ~ -Relative ~ t o the standAlthough these two approaches are clearly inconsistent, ard p-values for dissociation of the acids in water a t it does not seem possible to estimate how important So, the efficiency of transmission of electronic effects the difference may be in any given case. The availto the site of the D D M reaction in ethanol a t 30" is ca. able experimental data indicate that the families RG9056, or PRGCOS PRGCOOH E 0.9, COS, with few exceptions, fit the p-p re1ati0ns.I~ This Otherwise. eq. 2 can be used. Making the compariof course justifies the judicious application of p-p relason between the phenylpropionic and the benzoic systions until such time as they have outlived their usefultems in the four reactions-acid dissociation in water, ness. acid dissociation in 50% ethanol, basic hydrolysis of the ethyl esters in 88YG ethanol a t 30°, and acid and Experimental D D M in ethanol a t 30°2 9-0ne obtains fArCHgCH2COS The procedure used in studying the reaction of DDM with the phenylpropionic acids and the treatment of the kinetic d a t a have PArCOS = 0.212, 0.23. 0.20, and 0 . 2 2 , respectively. T h a t been d e ~ c r i b e d . ~T,h~e properties of the acids are given in Table is, in any reaction, the two methylene groups of the phenylpropionic system have cut the transmission t o TABLE 111 22% of t h a t in the benzoic acids, or E = 0.22. Either THEPHENYLPROPIOSIC ACIDS viewpoint of eq. 1 or 2 means of course t h a t G is damp--Equiv wt -M P " C -__Substituent Calcd Found Llt Found ing out the relay of electronic effects in a completely p-Br 229 1 232 1 136b 137-138 predictable manner. Clearly, the phenylpropionic 213 7 3,4-(CH30)2 210 2 98h 97-98 system is now close to the lower limit of transmission m-HO 166 2 lllC 163 7 113-1 14 efficiency among available fa mi lie^^,^; by contrast, H 150 2 150 8 48 5-49 5" 48 5-49 5 the high end of the p-scale is found in the furoic acid 164 2 165 3 116 5-117" 115 5-116 5 P-CHa series.j P-i-CaHi 192 2 192 6 73d 74-75 d p a r t from the experimental approach of eq. 1 and 2, 168 2 169 1 85 5-87 9l b P-F i t would be desirable to estimate attenuation factors m-C1 184 6 185 7 73-74' 71 5-72 5 ( E ) independently. If field-effect theory is to be app-so, 195 2 166-167" 199 6 168 5-170 plicable a t all, i t should apply to series in which con180 2 180 1 107-1 08" 100 5-102 P-CHaO jugation between the substituent and the reaction site E. K. Trachtenberg and G. Odian, J . A m . Chem. Soc., 80,401 (1958). * I(. Kindler and T. Li, Ber., 74, 321 (1941). c F. is precluded. The relation between attentuation and Tiemann and R. Ludwig, ihid.,1 5 , 2043 (1882) G. H . Slotta the Kirkwood-m'estheimer (K.m'.) theory1@takes the and H . Heller, ihid., 6 3 , 3029 (1930). e J . Kenner and E form4 \\'itham, J . Chem. SOC.,119, 1452 (1921). is ca. 260". Our determination of the activation parameters, which were not of primary interest, derives from two temperatures, so the entropy-enthalpy relation can only be tentative. The results of correlating log k with U-values based on the dissociation of the benzoic acidssa or with "resonance-free" u' valuessb are given in Table 11.
-
-4 comparison of p-values for the phenylacetic and phenylpropionic families in three reaction types-acid dissociation in water, basic hydrolysis of ethyl esters in (8 ( a ) D. H. llcL3aniel a n d H . C. Brown, J . O y g Chem., 23, 420 (1958); ( b ) H. van B e k k u m , P . E . Yerkade, a n d B . .If. U'epster, Rec. ~ ~ Q Zchim., I . 78, 815 (1959). (9) R. Fuchs, C. A. K a p l a n , J J. Bloomfield. a n d L F. H a t c h , J . Org. C h e m . , 2 7 , 733 (1962); J . J Bloomfield a n d R Fuchs. ibid.,28, 910 (1963). (10) ( a ) J. G Kirkwood a n d F. Westheimer, J . Chem. P h y s . , 6, 506, 513 (1938); ( b ) F. Westheimer. J . A m . Chrm Sac., 61, 1977 (1939), 62, 1892 (1940) icj C T a n f o r d , i b i d , 79, 5348 11957), ( d ) C . D. Ritchie a n d E. S. Lewis. ibid 84, 5 9 1 (1962). ~
(11) For a critical appraisal of K . W . theory see ref. l a c , f o r t h e problem of solvation of substituents and K.W. theory see ref. 10d. T h e problem of uncertain geometry in t h e phenylpropionic acid side chain is no longer a m a t t e r of concern T h e acidities of P-substitilted propionic acids were recently compared with those of a bicyclic system, 3- e n d a - or 3 ~ e x o ~ K ~ b i c y c l o [2.2 l]hept-5-ene-2-endo-carboxylic acid, in which the substituent ( R ) a n d t h e carboxyl group were fixed either in a ris or l v a x s structure.12 Plots of the p K values of t h e propionic acids were linearly related t o corresponding acids in t h e ~ Y Q P Z series S with a slope = 0 90 & 0.1 b u t did not correlate with those of t h e cis series l 2 T h i s strongly suggests t h a t t h e trans conformation is favored by t h e phenylpropionic acids (12) H . Hogeveen a n d F M o n t a n a r i . J . Chem. Soc., 4864 (1963). (13) M J. S. Dewar a n d P. J . Grisdale, J. A m . Chem. Soc.. 8 4 , 3518 (1962 j. (14) T h i s s t a t e m e n t applies t o families covered by t h e H a m m e t t equation Unpublished work shows t h a t attenuation f a c t o r s obtained f o r families covered by t h e T a f t equation are much more variable.
40 18
M. J. KAMLET, H. G. ADOLPH,AND J. C. HOFFSOMMER
111. A Beckman DK spectrophotometer was used in this work. Kinetic d a t a were taken at 35.35 =t0.1" and a t different ternperatures held to i 0 . l " in the range 25.5-26.2'. Activation energy plots permitted the adjustment of the rate constants t o a common temperature 26.05 ZIC 0.1" given in Table I ; in no case was the correction greater than 4%. The DDM concentration a t CU. 0.0022 M was normally in the range t o 1/80 of the
[CONTRIBUTIOS FROM T H E ORGANIC CHEMISTRY
Vol. 86
acid so that pseudo-first-order kinetics applied; two runs a t a ratio of l / 6 are still acceptable for the pseudo-first-order treatment since only ca. 60c2 of the DDM reacts to give the ester (see eq. 3). -411 of the kinetic data are given in Table I.
Acknowledgment.-We wish to thank J. Belluzzi for preparing several of the acids.
DIVISION,u.s. NAVAL
ORDNANCE
LABORATORY, WHITE
OAK, SILVER SPRING, M D . ]
Steric Enhancement of Resonance. 11. Absorption Spectra of N-Alkyl- and N,N-Dialkyl-2,4-dinitroanilines 1,2 BY MORTIMER J. KAMLET, HORSTG . = ~ D O L P H ,A N D
JOHN
C. HOFFSOMMER
RECEIVED APRIL 27, 1964
Close comparison of the ultraviolet spectra of 2,4-dinitroaniline ( I ) and its N-alkyl and X,N-dialkyl derivatives allows one to discern the spectral effects of steric enhancement of ( +R2N=Cl+C4=riOz-) resonance as resonance. T h e existence of another hitherto undescribed well as steric inhibition of ( *R&=C1+C2=NO2-) phenomenon, the electronzc buttresszng efecl, is suggested t o explain some of the spectral variations.
Steric diminution of electronic suppression of resonance interaction or, more succinctly, steric enhancement of resonance may occur in a system such as the following where, in Ingold's notation,3 A is a +M or +I substituent and X and Y are --M substituents. A measure
3
A+
X
X-
II
of resonance interaction between A and X (or between C1 and X where A is a + I substituent) is the electronic transition energy of the ( +A=Cl-.C4=X-) band in the ultraviolet. This transition energy depends strongly on the ground-state electron density a t A, A,, shifting to the red with increasing electron density a t A, to the blue with decreasing electron density. The ground-state electron density a t A is, in turn, a strong function of the nature of Y and its degree of coplanarity with the ring. Coplanar Y effects both mesomeric and inductive electron withdrawal from A, thereby suppressing resonance interaction between A and X ; noncoplanar Y exerts only its inductive effect,4 also suppressing A+X resonance interaction, but to a lesser extent. I t follows that as progressively increasing steric requirements of A force Y from coplanarity, groundstate electron withdrawal from A by Y will decrease and resonance interaction between A and X will increase. As concerns A-cY resonance interaction, this is the classical situation of steric inhibition of resonance. In the case of A+X resonance interaction, however, the newly reported phenomenon,' steric (1) P a r t I M ,J. Ktrmlet, J. C. Hoffsommer, a n d H. G. Adolph, J . A m C h e m . Sac. 84, 3926 (1962). ( 2 ) V . Baliah a n d M. U m a [Telrahedvon Leltevs, No. 2 6 , 21 (1960)l had used t h e t e r m , steric enhancement of Yesonance, t o describe t h e effect of a 2-alkyl group in increasing t h e probability of alkoxy or alkylthio groups approaching coplanarity in such compounds a s 2-methyl-4 nitroanisole or methyl 2 - m e t h ~ l - 4 - n i t r o p h e n y lsulfide. Although t h e s a m e t e r m adequately describes both sets of phenomena, t h e phenomena themselves are somewhat different, We regret t h a t Baliah and Uma's paper had n o t come t o o u r a t t e n t i o n prior t o t h e publication of P a r t I . (3) C. K . Ingold, "Structure a n d Mechanism in Organic Chemistry," Cornell University Press, I t h a c a , N. Y . , 1953. Section 7. (4) R. W. T a f t a n d I. C. Lewis, J . A m . C h e m . Sac., 80, 2436 (1958)
enhancement of resonance, is being brought into play. It is an important corollary of the above that steric enhancement of resonance in one molecular a x i s i s alw a y s accompanied by steric inhibition of resonance in another molecular axis. We have reported that steric enhancement of resonance is evidenced by progressive bathochromic spectral displacements in the series [2,4,6-trinitrotoluene, l-ethyl-2,4,6-trinitrobenzene, l-isopropyl-2,4,6trinitrobenzene, l-t-b~tyl-2,4,6-trinitrobenzene]~ and [2,4-dinitrotoluenej l-ethyl-2,4-dinitrobenzene,l-isopropyl-2,4-dinitrobenzene, l-t-butyl-2,4-dinitrobenzene].5 We wish now to demonstrate that this phenomenon also obtains with the N,N-dialkyl-2,4dinitroanilines and may be discerned on detailed examination of the ultraviolet spectra of these and related compounds. Band Assignments.-The spectrum of 2,4-dinitroaniline (I) above 280 r n p comprises a high intensity N -+ V band, Amax 336 mp, upon whose longer wave length edge is superimposed a pronounced shoulder, h -390 mp (Table I , Fig. 1). The spectral envelope appears to result from the fusion of two bands : (1) Amax 333-336 mp ( E 13,500-14,000) ; (2) Amax 380-300 mp ( E 40OO-j00O). Since higher absorption intensities are generally associated with longer transition moments, it is reasonable to couple band 1 with the (+H*N=C1+ C4=N02-) electronic transition, band 2 with the ( +H2N=C1+C2=N02-) transition. Comparison of the spectrum of I with those of other mono- and polynitroanilines lends strength to these band assignments. An ( +HzN=C~-+CZ=NO~-) electronic transition in 2-nitroaniline leads to , , ,A 404 mp ( e 5200) ; 2,6-dinitroaniline, with electronic transitions of this type in two mutually equivalent molecular axes, shows Amax 412 mp ( E 9200). 4-Nitroaniline, on the other hand, shows higher intensity absorption characteristic of an ( + H ~ N = C ~ + C F = N O ~ - ) electronic transition, Amax 371 mfi ( E 15,900). 2,4,6Trinitroaniline, with one para electronic transition and two mutually equivalent ortho transitions. shows two distinct bands, Amax 318 mp ( E 12,000),and Xmax 408 rnp
(E
7800).
( 5 ) P a r t 111: M , J. K a m l e t , H. G. Adolph, a n d B. Johnson, manuscript in preparation,
