F19 Chemical Shifts of the ... - ACS Publications

D. R. Eaton and. William A. Sheppard. Vol. 85. Table VI. The Rate Constants of. Copolymerization of. Styrene. Derivatives with. Dianion of. «-Methyls...
5 downloads 0 Views 479KB Size
D. R. EATOX ASD ~VILLIAM A. SHEPPARD

1310

Vol. 85

this trend is not clear. The lower reactivity of 2,4dimethylstyrene could be anticipated in view of the DERIVATIVES WITH DIASIONOF a-METHYLSTYRENE TETRAMER enhanced inductive effect of two methyl substituents. The pattern observed in the rate constants of coSa-, -C( CH3)(Pli)CHzCH?C(CH,)(Ph)C(C&)( Ph)CH?CH2Cpolyinerization of styrene derivatives with living poly(CH,)(Pli-), S a - + M + K a + , -C(CH,)(Ph)CH?CH,C(CH,)styrene is reflected in the respective rate constants of (Ph)C(CH,)(Ph)CH2CH~C(CH3)(Ph)&lvl, Sa+ addition to the living a-methylstyrene tetramer (Na+, Solvent, T H F ; T , 25' -C(CH3) (Ph)CHzCHl [C (CH,) (Ph) ]2CH?CH2C(CH3)Ii i v . (Ph)-,Ka+. The pertinent results are collected in ends], [AI], x 10' x 103 Conv., 1. mole-1 Table YI] and it is interesting to notice t h a t all the RIonomer m./l. m/I, 7% sec. values given in this table are lower by approximately a Styrene 3.0 1.4 ... 500" factor of 2 than the corresponding values related to the Styrece 3.i 2.4 ... 520" addition to living polystyrene. On the other hand, the p-Methylstyrene 3.3 2.1 6-15 105 addition of styrene to poly-p-methylstyrene- or poly-pp-Meth ylstyrene 5.9 4.9 12-20 82 methoxystyrene- (see Table V) proceeds as fast as the p-MethosystyrenP 2.6 2 .4 23-57 28' addition to the living polystyrene. I t seems, therefore, that the steric factor is mainly responsible for the lower These data, calculated from the Y.P.C. analysis of the residual rate constants of addition to the tetramer of a-methylstyrene, are lower than those obtained previously (800-1000). We feel t h a t the spectrophotometric technique used in the earlier styrene. investigations could be less reliable, since the tetramer solution Finally, i t should be pointed out t h a t the results acquired a slightly yellow color on standing in air. ' This syscollected in Table V indicate t h a t the polar factors are tem was studied in a stirred-flow reactor. less important when they operate on the anion than when they act on the monomer (compare the rate conin the ordinary anionic polymerization this step apstants of copolymerizatim of the pairs : styrene-pparently involves a negative center. niethylstyrene and styrene-p-methoxystyrene) . They The low reactivities of a- and P-methylstyrenes reflect become, however, significant when the polarity of the the effect of steric hindrance due to the presence of a anion is changed considerably, e.g., in the addition of methyl group located close to the center of reaction. styrene to the living polyvinylpyridine. This effect is superimposed upon the inductive effect which also reduces the rate of copolymerization. Acknowledgment.-iVe wish to acknowledge the The higher reactivity of o-methylstyrene when comfinancial support of this investigation by the Kational pared with its p-isomer is unexpected and the cause of Science Foundation through Grant G-14393. TABLE YI

THEK A T E

CONSTANTS O F COPOLYMERIZATIOX O F S T Y R E S E

kij,

-1

[CONTRIBUTIOS S o . 731

FROM THE

CESTRALRESEARCH DEPARTMEST, EXPERIMENTAL STATIOS, E. I. DC POSTDE SEJIOURS AND Co., ~ ~ I L h I I S G T O98, N DEL.]

Flg

Chemical Shifts of the Sulfur Pentafluoride, Trifluoromethoxy, Trifluoromethylthio and Trifluoromethylsulf onyl Groups in Aromatic Compounds BY D. R . EATOX AND WILLIAM A . SHEPPARD RECEIVED OCTOBER18, 1962 T h e F19 n.m.r. spectra have been measured for a series of arylsulfur pentafluorides and aryl trifluoromethyl ethers, sulfides and sulfones substituted in the ineta or para position. The twenty-one line pattern for the sulfur pentafluoride group was analyzed as an AB4 system, and the chemical shift and spin-spin coupling constants determined. The coupling constant of 146 C.P.S. between the base and apex fluorine is independent of the substituent. The F19chemical shifts of these groups have been correlated with Hammett u-factors and with Taft's inductive and resonance parameters. T h e magnitudes of the changes in chemical shift with substituent and the correlations with reactivity parameters are discussed.

Introduction

X number of correlations between chemical reactivity and physical properties, such as vibrational band frequencies and intensities, polarographic halfwave potentials and n.m.r. chemical shifts, have been observed and explored in recent years. Of these physical properties, the study of the F19 n . m . r . chemical shifts is particularly attractive because of the high precision of the measurements and the relatively direct relationship to chemical bonding. Gutowsky, McCall, McGarvey and Meyerl first noted the correlation of the fluorine chemical shifts with Hammett 0-constants for a series of substituted fluorobenzenes. T a f t 1 h , 2 ,has 3 extended and elaborated this correlation by interpreting the shifts in terms of inductive and resonance effects. Theoretical treatments of the problem have been presented recently by several groups. (I) ( a ) H . S. Gutowsky, 11. W. hIcCall, B. R . 3 l c G a r v e y a n d L. H. I f e y e r , J A r i l . C h e w S o r , 7 4 , 4809 ( 1 9 3 2 ) ; ( h ) J. A . P a p l e , \V. G Schneider a n d H. J . Bernstein, "High-Resolution S u c l e a r I I a g n e t i c Resunance," SIcGra>v-Hill Book C o . , I n c . , S e w Y o r k , S. Y , 193'3, 11. 3 2 3 (2) R . \V. 'raft, J r . , J A r r i . C'hrrii. S.,c., 79, 104.5 flS.i7). (J) R LV. ' r a f t , J r , J . P h y s C'heru., 6 4 , 1805 , l ~ l f ~ 0 ) .

There is, however, very little experimental data available on the chemical shifts of fluoroaromatics other than the fluor~benzenes.~It is the purpose of the present study to investigate the possible correlation of the F19 chemical shifts of the SF6, OCF3, SCF3 and S03CF3 groups and the spin-spin coupling constant of the SF5 group with substituent parameters. Such data should provide a further test of the validity of current theories of aromatic substitution effects and also should give some information on the transmission of inductive and mesomeric interactions through oxygen and sulfur atoms.6 A concurrent study has been made of the [ A ) (a) A I . Karplus a n d T . P. Das, J . C h e m P h y s . , 34, l f i 8 3 ( 1 9 f i l ) , ( b ) 1'.Yunezawa, K . Fukui, H . K a t o , H. K i l a n o , S. H a t t o r i a n d S. \ I a t suoka. H i ~ l ! .C'lze,li, .5oc. J a p a i z , 34, 707 (1961); (c) R . W . T a f t . I:. Prosser, I.. Goodman a n d G .'I Ilavis, J . Chem. P h y s . , 38, 380 (19fXj). ( 3 ) 4 limited a m o u n t of chemical s h i f t d a t a has been reported un b e n z w trifluorides'" a n d on aryl trifluoromethyl e t h e r s , sulfides a n d sulfones; 1,. 31. Yagupolsky, V. F. Bystrov a n d E . Z . U t y a n s k a y a , P, oc. A c n d . Sci. r S S R , P h y s a i ! d C ' h u f , l . Sect. ( E n g l . Trans1 ), 135, 1039 (19fi0), a n d OPCirs nirci s p e c l i o s c o p y , 10, li8 (191jl). Correlationsof proton chemical shift with H a m m e t t o-parameters for t h e methyl group of irl- a n d fi-substituted anisoles were reJ . C h e i i i . , 40, 1%; (1962). cently reported by C. Heathcock, C Q ~ Z lli) 'l'he substituent parameters f o r t h e groups along with t h e synthetic

F19 CHEMICAL SHIFTSIN SULFUR AROMATICS

May 5, 1963

1311

contact interaction shifts of the same groups substituted on the ligands of paramagnetic Ni(I1) aminotroponeimines.' T h e chemical shift measurements are indicative of the transmission of charge density, whereas the contact shifts are determined by the transmission of spin density through the conjugated system so that the two studies are in many ways complementary.

F19

Experimental The compounds used in the present study have the structures 1-117. SF6 OCF3 7CF3 so,CFa

i

I

Q x

I1

I11

x I

Q x IV

where X is a meta or pura substituent ( H , NO*, KHz, OH, CHI and halogen). The synthesis of these compounds is described elsewhere.6J The spectra were obtained with a Varian high resolution instrument operating a t 40 and 56.4 Mc./sec. Reagent grade carbon tetrachloride was used as a solvent and 1,2difluorotetrachloroethane (Freon 112) as an internal calibrant. The trifluoromethyl sulfones IT.'were also examined in methanol solution to investigate the possible effects of hydrogen bonding. Calibration w-as accomplished b y superimposing an audiofrequency on the sweep field t o prodlice side band peaks of the reference. I n most cases measurements were made at four different concentrations and chemical shifts were obtained by extrapolation to infinite dilution. The corrections involved were of the order of 2-3 C.P.S. in most cases. The m- and p-hydroxy derivatives of I and IV were only slightly soluble in carbon tetrachloride, and the chemical shifts were obtained directly for t h e spectra of the saturated solutions. The precision of the measurements is estimated to be f l C.P.S. so t h a t the chemical shift measurements are meaningful for even the OCF, and SOZCR groups where the range in shifts is only 30 to 40 C.P.S. Analysis of the Spectra.-Compounds 11, I11 and IV have only a single F19 resonance, and the chemical shift is obtained directly. Compound I has a twenty-one line pattern typical.of a n .AB4 system with the apex fluorine ( A ) appearing to low field of the basal fluorines (B4). This pattern is consistent with a tetragonal-pyramidal configuration for the sulfur pentafluoride group as ascribed to alkylsulfur pentafluorides and other derivatives of sulfur h e x a f l ~ o r i d e . ~The present spectra were analyzed using the perturbation theory of and satisfactory agreement was obtained between calculated and observed spectra measured both a t 56.4 and 40 Mc./sec.

Results The chemical shifts and spin-spin coupling constants for compounds I are presented in Table I . It is apparent that there is no significant change (less than 2 c.P.s.) of JFF with substituent. The mean value of 146 C.P.S. for the spin-spin coupling constant is very close to the value of 145 C.P.S. reported by Muller, Lauterbur and Svatosga for perfluoroalkylsulfur pentafluoride derivatives. Chemical shift data for compounds 11, I11 and IV in carbon tetrachloride and for compound IV in methanol are given in Table 11. There is a marked dependence of the chemical shift on the substituent for all four series of compounds. Furthermore, in each case there is good correlation with substituent parameters. Gutowsky, McCall, McGarvey and Meyer'" originally correlated the chemical shifts of the fluorobenzenes with the Hammett U-values. Taft has pointed out that the relative importanceof the inductive andresonance contributions may not be the same for F19chemical shifts as it is for acidity and reactivity functions, and for the fluoro*v3

chemistry have been described a n d discussed, W. A. Sheppard, J . A n t . Chem. soc., 84, 3064, 3072 (1962); 83, 4800 (1901), and publications i n preparation. (7) Publication i n preparation. (8) L. M. Yagupolsky a n d M. S. Marenets, J . Gen. C h e m . C S S R (Engl. Transl.), 24, 885 (19541, a n d 26, 99 (1950). (9) ( a ) K. Muller, P. C. Lauterbur and G. F. Svatoe, J . A m . Chevz. Soc., 79, 1043 (1957); see, ref. l b , p , 339; (b) R.K . Harris a n d K. J. Packer, J. C h e m SOL.,4730 (1901); (c) C. I. Merrill, S. M. Williamson, G. H. Cady a n d D. F. Eggers, Jr., 1noi.g. Cheirz., 1, 21.5 (19ti2). (10) W. A . Anderson, Phys. Rezb., 102, I:jl (19.5ti); see also ref. I b , p. 13:

- 4 Y '

" -0.3

"

"

'

0

"

+ OI S

"

'

"

" +LO

HAMMETT PARAMETER r,

Fig. 1.-Correlation plot of 6 for apex fluorine us. Hammett parameter u for substituted arylsulfur pentafluorides (6 = -0.50 4.33u).

+

benzenes he obtained better correlations with chemical shift data by utilizing separate inductive (UI) and resonance (UR) parameters. Krueger and Thompson'' TABLE I CHEMICAL SHIFTS AND SPIN-SPIN COUPLING CONSTANTS FOR F19 IS ARYLSULFUR PENTAFLUORIDES ( XCsH4SFs)AT INFINITE DILUTIONI N CARBON TETRACHLORIDE Substituent

x

--Apex YA.

c.p.s.a

fluorine8.b p.p.m.b

--Basal fluorines-VB, SB, c . p . ~ . ~ p.p.m.b

JFF,

cycles/sec.

-6035 0.00 -5163 0.00 +147 - .40 +146 -51T9 $2.85 -5921 - , ia +147 -5193 +2.T3 -5926 m-SO2 -2.25 f146 -5253 -3.15 -6161 p-NHz -0.48 +147 -1.23 -5182 -6084 m-NH2 -1.10 +145 -5207 +0.30 p-Br -6023 -0.75 +146 -5193 $1.13 -5990 m-Br -1.03 +146 -5204 +0.45 -601i p-c1 -5245 -1.85 -2.05 (+149) -6109 p-OH -0.63 t146 -5188 -0.35 -6049 m-OH Kecorded a t 40 Mc./sec. relative to CFClXFClZ (internal). b 6 relative t o phenylsulfur pentafluoride as zero and positive for shift to higher field. H

--

p-NOn

0

have treated their results on infrared frequencies and intensities in a similar way, The present results for compounds I1 and I11 and the basal fluorines of compounds I give satisfactory correlations by plotting the meta shifts against (TI and the difference between the para and meta shifts against U R in the manner descrjbed by Taft. (Values used for u , UI, U R and UR' are given in Table 11.) Examples of these plots are shown in Fig. 2 and 3. The chemical shifts for compounds I V and the axial fluorine of compounds 1 plot equally satisfactorily against Hammett u directly as shown in Fig. 1 and 4. These results, together with those of T a f t on the fluorobenzenes, may be summarized by the following equations. The equations (with the exception of eq. lib) were determined by least squares best fits, 6 = bo blu~ b z ~ ~ " . For ' ~ these equations U R O was found more appropriate than UR (note, U R / U R ' 1.5 to 2.O).3 For comparison, all results have been expressed in terms of (TI and ~11". The two cases which correlate directly with Hammett u ) namely, SFb axial and S02CF3,have the coefficient of UI of the same order In the remaining cases the dominance of the as U R O .

+

+

(11) P. J. Krueger a n d H. W . Thompson, Pvoc. R o y . SOL. ( L o n d o n ) , A260, 22 (1959). (12) W. A. Pavelick and R. W. T a f t , J . A m . Chern Soc., 79, 4935 (1957). We are indebted t o Professor T a f t f o r advice a n d assistance in carrying o u t these correlations,

1312

D.

K.EATONASD

A. SHEPPARU

KILLIm

I

I

I

I

1

I

I

I

I

I

1

I

I

I

I

1

I

I

I

1

I

-

-

+06 l

\'ol S.5

SCF3

to4-

- 0.8 0

- 0.2 -0.4

+1.0,,

-E

,

,

I

I

I

I

I

1

,

I

I

I

,

I

r1 -

o.6

-0.8

t

~

O-

a

NO2 I

I

I

I

I

I

+ 0.5

0

1

I

I

INDUCTIVE PARAMETER rI,

b"

w

-10-

x

Fig. sa.-Correlation plot of 6 for F I 9 of 111-substituteclSCsH4OCFI, XC6H4SCF3and S C 6 H , S O ~ C F'L".T. , inductive parameter U I in CCI,.

LD

I

c9

-2.0

-

1

1

1

1

1

-1.0

I

- 0.5

1

1

1

1

1

1

'

0 RESONANCE PARAMETER C R .

"

1

+ 0.5

+ 0.2

Fig, 2b.-Correlation plot of 6 - 0 . 4 5 2 ~ for 1 basal fluorine t ' s . the resonance parameter U R for p-substituted arylsulfur pentafluorides (6,) = -0.28 - 0,45u1 2 . 5 0 ~ ~ ) .

Br

0

- 0.2

+

- 0.4 I

term in the p-substituted compounds destroys the simple dependence on Hammett (T.

I

I

I

I

I

I

1

1

-

URO

F

$0.20 $0.80 SFs apex 6", 0.00 -0.09 basal 6,LF -0.01 = -0.10 OCF, 6mF = - 0 . 0 4 6pF = S o , 04 SCF,, 6,F = - 0 . 0 3 SPF = + 0 . 0 4 SOzCF, (in eel,) 6,F = -0.01 6pF = 0.00 (in CH,OH) amF = -0.02 = +0.01 ",6

= = = =

SCF3

++ + +

1

+ 1.2 -

- 5.83~1 - 5 . 8 3 ~ 1- 1 8 . 8 u U ~ ( 1 ) 3.7901 3 . 4 7 U R ' 3.54u1 6.99~~' - 1.3861 0 . 6 5 ~ ~ ' - 0.99~1 3 . 9 8 ~ ~ ' 0 . 5 6 ~ 1 0.10UR' 0.3 O U I - 1.35UR0 - 0.79~1 0.00~~' - 1.3201 - 3 . 7 6 ~ ~ ' - 1 . 3 1 ~1 0.45~~' - 1.04U' - 1.51UR' - 1.09~1 0.82~~' - 0 . 8 7 ~1 1.82~~'

+1.4

++ + ++ -+

+1.0 +OB -

+0.6+0.4

-

+0.2

-

0-0.2 -0.4

-0.6

-

" I

Discussion I t seems clear from the analysis of Karplus and D ~ that s ~fluorine ~ chemical shifts are dominated by the second-order paramagnetic term and hence by the nature of the fluorine bonding. There are, however, a variety of inductive, resonance and field effects p ~ s s i b l e , ' and ~ an extensive accumulation of various kinds of physical data will probably be necessary before the contributions of the various factors can be disentangled. Ring current effects in these aromatic systems may also have to be considered, but it seems unlikely that these would vary significantly with substituents. In the case of the sulfur pentafluoride compounds in particular, small changes in the hybridization of the sulfur from compound to compound and differe b q q in the proportion of d-orbital contribution to the apex and basal bonds may also be significant. Several comments can, however, be made from the results of the present study as embodied in eq. 2-6. The (I:() 1

l!llj2).

51. J .

S. Ilewar a n d P. J . Grisdale, J . .4 J U ( ' h e m . S o i . , 84, : % i 4 0 , :3548

I

I

I

I

I

1

I

I

1

I

I

I

I

I

1

2 + 0.8

-2 ++ b"

w

u)

-s

0.6 0.4 +0.2

o -0.2 1

.

I

-0.5 0 RESONANCE PARAMETER u R ,

+

Fig. 3b.-Correlatiori plot of 6 X U I for Flg of p-substituted SC6H40CF.4,XCsH4SCF3and XC&I,SO&F:j V T . the resonance parameter UR in CCI,.

observed correlations demonstrate that substituent effects are indeed transmitted to atoms remote from the ring. As might be anticipated, the magnitude of the effects falls off with increasing number of intervening bonds in the series : fluorobenzenes (one bond), SF5 compounds (two bonds), and OCF3, SCF3 and S02CFa compounds (three bonds). Conjugation through -0-, -S->and -SO2- has recently been demonstrated by the

FI9 CHEMICAL SHIFTSIN SULFURAROMATICS

May 5 , 1963

1313

TABLE I1 CHEMICAL SHIFTS FOR

F'9 I N

ARYL

,-Substituent Substituent

0

TRIFLUOROMETHTL ETHERS, SVLFIDES A S D SULFOKES CHLORIDE AND METHANOL --XCsHaOCFs-?-In CCla--

parsmetera-OR

UI

o

~

O

AT I S F I N I T E

--XC6HkXFa----In CCh--

C . P . S . ~6 , ~ . p . r n . Y~, p.p.m.'

Y,

CARBON

TETRA-

XC~HISOZCF~ CClr--In CHaOH--

--In

6 , p.p.rn.'

DILUTION IN

p . P . m . * 6, P . P . m C

v 8 P.P.m

*

6 , P.P.m.'

-377 0.00 - 984 0.00 $458 0.00 +465 0.00 ,78 -376 .03 -1034 -1.25 +425 - .83 $437 - . 7 1 nieta ,71 -366 .28 -1010 -0.55 +422 - .9O +434 - .78 S H ? parad - .66 .10 .76 - .48 -351 .65 -917 +1.68 $482 .60 $499 .80 mete - , 16 -379 - .05 -992 -0.20 +459 .02 $473 .20 CH3 para - .17 - .05 .13 - .10 -370 .18 -966 .45 meta - .07 -382 .13 -985 .03 $487 .53 OH para - .36 ,25 .61 - .40 -350 .68 -937 $1.18 mela - .02 -374 .08 -991 -0.18 $450 -0.20 $469 .09 Br para .23 ,45 - .22 - .19 -363 .35 -983 .Oi $450 - . 2 1 +459 - .16 $440 .46 $449 - .41 meta .39 -370 .18 C1 pura .23 .47 .24 - .20 -361 .4O -977 .18 mefa .37 -369 .20 -999 - .38 F para .06 .50 .44 - .35 -350 .68 meta .34 -369 .20 OCFI para .35 .5l .13 -360 .43 mela .40 -366 .28 I'alues of u from H . H. Jaff6, Chem. Rev., 53, 222 (1953); UI and UR from ref. Ib, 2, 3; URO from R . W. Taft, S. Ehrenson, I. C. Recorded at 40 Mc./sec. relative to CFC12CFC12 (internal). Relative Lewis and R. E. Glick, J . A m . Chent. Soc., 81, 5352 (1959). to unsubstituted aryl compound as zero and positive for shift to higher field. d ortho NH2 for OCFl recorded as -388 C.P.S. with 6 -0.28 p.p.m.

H S O Y para

0

0 0.15

0 0.63

0 0.08

-

-

-

transmission of spin density through these groups14; the non-zero values of the ~ J R O coefficients in eq. 4, 5 and 6 provide further evidence for this effect. The relative values of the coefficients also are reasonable with -SO?-being the least effective conjugating group and -S- being significantly more effective than -0-. It is interesting that the signs of the U R coefficients alternate on transmission through one bond (negative), two bonds (positive) and three bonds (negative). Whether this is a general effect or a peculiarity of the particular groups studied remains to be established. The different signs of the UI coefficients for the OCF3 and SCF3 may also be of significance. Differences in the substitutent constants for these groups have been reported previously6 and attributed to participation of the d-orbitals in the sulfur compound. There is also a reversal in the signs of these coefficients between the basal and apex fluorines of the SFSgroup. It is possible that the basal fluorines may experience a large direct interaction with the ring n-system, giving a negative contribution to the upcoefficient (as in the fluorobenzenes) which opposes the intrinsic positive contribution reflected in the apex fluorine dependence. In the two cases where the chemical shifts correlate directly with Harnmett a-parameter (S02CF3 and apex SFb) the fluorines are screened from the a-system of the ring by the SO2 and SFd groups, respectively. It is possible t h a t these groups may have an insulating effect in preventing direct interaction of the fluorine with the a-system of the ring, and the electronic effect transmitted by field and u- and a-inductive mechan i s m ~is' ~now the major factor influencing the chemical shifts. The results for the S02CF3 compounds in carbon tetrachloride and in methanol are very similar in their substituent dependence. I n each case there is a shift of around 10-20 C.P.S. to high field in methanol compared with carbon tetrachloride. I t is deduced from this that although hydrogen bonding with the -SO2- group can ( 1 4 ) I).I