Hyperfine coupling constants of the 2-chloroethyl and related radicals

Hyperfine coupling constants of the 2-chloroethyl and related radicals

V&me 5, number 9 CIEMXALPHYSICS LETTERS HYPERFINE COUPLING CONSTANTS AND RELATED OF THE 15 June 1570 P-CHLOROETHYL RADICALS A J. BOWLES, ...

261KB Sizes 0 Downloads 21 Views

V&me 5, number

9

CIEMXALPHYSICS LETTERS

HYPERFINE

COUPLING

CONSTANTS

AND

RELATED

OF

THE

15 June 1570

P-CHLOROETHYL

RADICALS

A J. BOWLES, A HUDSON and R A JACKSON School of Molecular Sciences, University of Sussex, BrightonBM

SQJ,

UK

Received 10 April 1970 The electron resonance spectra of several fluorine and chlorine substituted alkyl radicals have been recorded in fluid solution. We have found that p-chlorine hyperfine splittings are often larger than those

of a-chlorines. The Z-chloroethyl radical has a preferred conformation in which the chlorine atom is eclipsed by the p-orbital containing the unpaired electron suggesting an unsymmetrical bridged structure for this

species;

relatively

free

internal

rotation

We wish to draw attention to a number of interesting features which have emerged during our current investigations of fluorine and chlorine containing alkyd radicals. The coupling constants of the new radicals obtained at ca. -60’ are in table 1. They were mostly prepared from the

corresponding

with triethylsilyl

alkyl

bromides

radicals

[l],

by reaction

in a few cases hy-

drogen atoms were abstracted from the parent alkane by t-butoxy radicals [2].

is observed

in the Z-fluoroetbyl

radical.

Table 1 Hyperfine coupling constants (gauss) of some halogen containing alkyl radicals Radical

ag

aa

&Cl,

“Fp

=c1

84.6 c)

10.5

CF3cH2 =)

23.9

-

29.9

CF3eHCl

22.2

-

23.4

3.7

CH2FeH2

22.3

26.9

The 19F splitting of 84.6G in the radical kFC12 obtained from CFCl3 is very close to that found [3] in CHF2 (84.2G) and suggests a substantial deviation from planarity; in addition the chlorine splitting is unusually large. Most (Ychlorine coupling constants in the literature

CH.&H2 CH&HCH2Cl

21.5 21.3

11.5

17.4

11.4

14.2

22.2

-

19.3, 19.7

-

[4,5]

CF+(OSiEtg)CH3

lie in the range

3-4G

and this

is

for the radicals such as CF$HCl and

also

found

CH2(OH)CC12 reported here. A-slightly larger splitting of 6.25G was found in CC13 [2] (a value of ?.‘7G has been reported in the solid state [,6]). FroE its chlorine splitting of 10.5G _we deduce that CFC12 is more pyramidal than CC13. This is in accord with the higher electronegativity of fluorine compared with chlorine. Reduced steric interactions due to the smaller size of fluorine are also likely to be of some significance in determining the structure. Examples of B-chlorine coupling constants in the literature are restricted to small splittings from trichloromethyl substituted nitroxides, a~1 = 2.2G in f-butyl trichloromethyl nitraxide

552

CH30CF2eCl2 CH3OCF2eHCl CH2ClCOeH2 CHzCLC&%r2

CH,(OH)@ CH2(OH)i:HCl

47.1

21.1

4.0

26.9

3.5 0.42

21.5

17.6 20.7

9-1 b) 13.4 18.0

4.9

4.1 2.6

a) An *proton coupling constant of 23.9G has also been reported by Fessenden [lS]. b) d-proton coupling constant. c) @-fluorine coupling constant.

and 1.25G in (CCl3)2NO [8]. The only other reported @-chlorine interaction we are aware of

[?‘I

is a 0.5G splitting

a+ibuted

to two equivalent

chlorines in CH2ClC(OH)CH2Cl. The radical was produced in a flow system by the reaction of hydroxyl radicals with 1,3-dichloroisopropanok [4].

CHEiMICALPHYSICSLETTEFtS

Volume 5, number 9

In view of the large &chlorine splittings found in our present work, we have reinvestigated the ab-

straction of a hydrogen atom from 1,3-dichloroisopropanol by photolysing solutions of the alcoho1 containing

di-t-butyl

peroxide.

We assign

the

spectrum ob&ined (fig. 1) to CH2ClCdH2 formed-by the elimination of HCl from CH2ClC(OH)CH2Cl. Similar eliminations have been observed previously in radicals of this

type [9], Theg-factor

of LOO43_is typical of

acetonyl radicals. As in CH3COCH2 the methylene protons are inequivalent at the temperature of our experiments (-40’). The barrier to rota-

tion about the CH2-CO bond in the acetonyl radical has recently been determined as 39.5 kJ mole-l [lo]. In view of the similarity of our proton splittings and g-factor to the values reported by MSbius et al. [4], we suggest that they ob-

served the same radical in their experiments. small interaction of 0.5G which they assigned to P-chlorines is in fact ay-chlorine

The

coupling of 0.4ZG. The effects of hindered internal rotation on the P-proton [email protected] constants of a-substituted ethyl radicals, CH2CH2X, are well doctimented [ll]. The observed splitting is proportional to ~~ + B2(cos20) where the angular average is

.

15 June 1950

evaluated over the appropriate torsiona wavefunctions; in practice I30 = 0 and, since (CO&~) = $ for the ethyl radical, B2 = 53.76. If the radical has an equilibrium conformation in which

tha odd electron

X eclipses

orbital,

is less

than 26.9G and decreases as the temperature is lowered towards a limiting value of about 13.5G (B. = SO’). In the other possible equilibrium con-

formation X lies in the nodal plane of the z-orbital. Then $-R LS greater than 26.9G and increases as the temperature is lowered towards a value of about 40.5G (00 = 30’). The 11.5G P-proton coupling constant found for

2-chloroethyl

at -SO0 suggests

that this ra-

dical is essentially locked in a eodormation with chlorine eclipsing the odd electron orbital. The observed splitting is slightly lower than the value predicted

some

for

this conformation

deformation

chlorine

moving

and may

of the chloromethyi towards

a bridging

Lnciicate

group with

position

and

the P-protons moving towards the nodal plane of the r-orbital so that 60 > 60°. A similar cordormational preference has recently been found [LZ] for radicals CH2CH2X with X = SiEt3, GeEtg and SnEt3 which have p-proton splittings of ca L7G coupling constant in 2at -1OOO. The P-proton fluoroethyl at -70° is very close tc that predicted for isotropic averaging; its magnitudeincreases

as the temperature is lowered and the fluorine splitting decreases. We therefore deduce that this radical

has a preferred

conformation

with

fluorine in the nodal plane of the ;;-orbitaL This is also the equilibrium position for radicals with X = CH3 1131, CH3CH2 [14], CH2=CH [13], C6H5 [14], OH [15], CN [l], and from our own unpublished work COOH and OCH3. A &eroton splitting of 13.4G indicates that CH2(0H)CC12 has an equilibrium conEormation with the OH group ecl_ipsing the n-orbital whereas in CH2(OH)CH2 the OH group prefers to lie in the nodal

plane

[IS].

A wide variation

p-fluorine

clipi-cnp

AH

-HCI

0

Fig. 1. ESR spectrum observed during the photolysis of a mixture containing 1,34ichloroisopropanol and

di-t-butyl peroxide at -No with an expanded trace of the outer and central lines.

is apparent in the values of coupling constants. The l9F splitting

in CH2FCH2 is substantially larger than that found in CF3CH2 although there is essentiaLIy free rotation of the fluoromethyl group. The introduction of an cy-chloro substituent into CF3CH2 lowers QF from 29.9 to 23.4G whereas introducing an OH group raises it; in CF3CHOH aF is 31.7G 1161, although the p-proton CoupIing

in CHQCHOH(22.2G [15]) is substantiallp lower than in CH3CH2. To obtain a direct comparison between p-fluorine and p-proton splittings we have recorded the ESR spectrum of 553

Volume 5. number 9

CHEMICAL PHYSICS LETTERS

CF$(OSiEt3)CH3 prepared by adding triethylsilyl radicals to trifluoroacetone [l]. The_ ratio

of aF/~is close to that found from CF3CHOH and CH$HOH. The large difference in fluorine splittings found between (CF&NO (8.26G) and (CF3)2CO-

(34.94G)

has been

interpreted

in

terms of an interaction between the lowest unoccupied antibocding orbital on the CF3 group and the antibonrling

orbital

containing

the un-

paired electron [l?]. It seems possible that variations in the energy of the odd electron orbital are also important in determining the extent of electron spin transferred to fluorine in the new radicals reported here. We plan to carry out

calculations to investigate further the mechanism of B-fluorine hyperfi,le interactions which is of some

current

interest

fellow-

ship to AJ.B. and a grant towards the cost of apparatus, LC.L Ltd. for a gift of ‘Fluothane’

(CF3CHClBr) and Abbott Laboratories of ‘Penthrane’ (CH30CF2CHC12).

for a gi’;

RE FERENC ES [lj A. Hudson and R. A. Jackson, 1323.

Chem. Commun. (1969)

[2] A. Hudson and H. A. Hussain. Mol. Phys. 16 (1969) 199.

554

[3] R. W. Fessenden and R. H. Schuler. J. Chem. Phys. 43 (1965) 2704. [4j K. Miibius. K. Hoffmann and M. Plato, Z.Naturforsch. 23a (1968) 1209. [5] R. P.Kohin, J. Chem. Phys. 50 (1969) 5356: A. Hudson, Chem. Phys. Letters 4 (1969) 295; A. L.J.Beckwith and R-0. C. Norman. J. Chem. Sot. B (1969) 400.

[Sl J. Ron&n. Mol. Cry&_ 3 (1967) 117. [7j hf. J. Perkins, P. Ward and A. Hosfield. J. Chem. Sac. B (1970) 395.

[8] H.Sutciiffe and H. W. Wardale, J. Am. Chem. Sot. 89 (1967) 5487. [9] A. L.Buley, R.O. C. Norman and R. J. Pritchett, J. Chem. Sot. B (1966) 849. [lo] G.Golde, K. M6bius and W.Kaminski, Z.Naturforsch.

[ll]

24a (1969) 1214.

E. W. Stone and A. H.Maki. J. Chem. Phys. 37 (1962) 1326; R. W. Fessenden.

[la].

We thank the SRC for a postdoctoral

15 June 1970

J. Chin;. Phys.

61 (1964) 1570:

D. H. Geske, Progr. Phys.Org. Chem. 4 (1967) 125; M. D. Sevilln and G. Vincow. J. Phys. Chem. 72 (1968) 3647.

1121$. J. I&sic and J. K. Kochi, J. Am. Chem. Sot. 91 (1969) 6161. 1131R. W. Fessenden and R. H. Schuler. J. Chem. Phys. 39 (1963) 2147. [14] J. K.Koci and P. J. Krusic. J. Am. Chem. Sot. 91

(1969) 3940. [15] R. Livingston and H. Zeldes, J. Chem. Phys. 44 119661 1245. 1161 A. H&son and H. A. Hussain. J. Chem.Soc. B (1969) 793. [17] K. Morokumn. J. Am. Chem. Sot. 91 (1969) 5412.

[18] D. Kosman and L. M. Stock, J, Am. Chem. Sot. 92 (1970)

409.

1191R. W. Fessenden,

J. Phys. Chem. 71 (1967) 74.