Mechanisms of Photoreactions in Solution. II. Reduction of

Chemical Laboratory of Iowa State University, Ames, Iowa]. Mechanisms of Photoreactions in Solution. II. Reduction of Benzophenone by Toluene and Cume...
0 downloads 0 Views 656KB Size
REDUCTION OF BENZOPHESONE BY TOLUENE AND CUMENE

July 5 , 1961

2795

[CONTRIBUTIOS FROM THE GATESAND CRELLIN LABORATORIES OF CHEMISTRY, CALIFORNIA INSTITUTE O F TECHNOLOGY (NO. 2639), PASADENA, CALIFORNIA, AND FROM THE CHEMICAL LABORATORY O F IOWA STATE UNIVERSITY, AhfES, I O \ V A ]

Mechanisms of Photoreactions in Solution. 11. Reduction of Benzophenone by Toluene and Cumene BY GEORGEs. HAMMOND, WILLIAM P. BAKER AND WILLIAM &I. MOORE RECEIVED NOVEMBER 28, 1960 The photoreduction of benzophenone by toluene and cumene has been analyzed quantitatively. Paramagnetic metal chelate compounds have powerful quenching effects. The data are interpreted in terms of the reactivity of the triplet state of benzophenone, and rate constants for all processes are estimated. Preliminary niensurements of the reactivity of other ketones have been made using toluene as the reductant.

The photoreduction of ketones by a variety of hydrogen donors has been reviewed by Schonberg and Mustafa.’ Included among the donors are various hydrocarbons. I n most instances, in which qualitative observations indicate that the reaction “goes well,” the donor has a structure such that i t should yield a relatively stable free radical upon loss of a hydrogen atom. A full material balance usually has not been carried out but qualitative observations suggest that an excited state of the carbonyl compound extracts hydrogen from the reductant and that the ultimate reaction products are formed by coupling reactions of the radicals produced in the first step.

\ CEO* / 2

\.

+ RH+

COH+--‘2-C-

/

\.

COH

/

I I

OH 2R.

/ I

+ R.

(1)

(2)

OH

R-R

\. I R * + COH +R-C-OH / I

(3)

(4)

We have carried out quantitative studies of several reactions of this type. The quantum yields of the various possible processes have been measured and the quenching technique, developed in the accompanying paper,2 was used to make a preliminary estimate of the lifetime of the triplet state of benzophenone in toluene and cumene. Results and Discussion Large batches of benzophenone in toluene and in cumene were photolyzed by prolonged exposure to sunlight. The mixtures were then worked up by conventional techniques. It was found that benzpinacol (I) and benzyldiphenylcarbinol (11) accounted quantitatively for the benzophenone photolyzed in the toluene mixture. Bibenzyl, which should have been formed in amount equivalent to benzpinacol, was detected as a reaction product but was not isolated quantitatively. Separation of the products formed in cumene gave benzpinacol, the tertiary carbinol (111), and 1,2-diphenyl-1, 2-dimethylbutane (bicumyl). Quantitative analyses of photolysis mixtures were carried out by spectrophotometric determination of residual benzophenone and/or by determination of benzpinacol either gravimetrically or (1) A. Schijnberg a n d A. M u s t a f a , Ckcm. Rem., 46, 181 (1947). (2) G. S. H a m m o n d , W. M . Moore a n d R. P . Foss, J . A m . Ckcm. S O L ,83, 2789 (1961).

CeHs CsHs

C6H5

I / C~H~-C---C-CGTTS I 1 OH

I I

C~HS-C-CH~C~H~ OH I1

OH

I

C6Hs CH3

1 I

C6H5-c-c-c&

1 t

OH CHB I11

by lead tetraacetate oxidation. Lead tetraacetate determination could not be used in runs with cumene as i t was found that the tertiary alcohol, 111, was cleaved rapidly by the reagent. This interesting observation will be reported in more detail elsewhere. Quantum yields were determined by reference to the uranyl oxalate actinometer. Studies by two workers, using different filter systems, gave slightly different values for the quantum yields in the benzophenone-toluene reaction. These are identified as belonging to series A and series B in Table I. In the work with ketones the concentration conditions were such that all of the light was absorbed in all experiments. This condition will have no influence on the results if the mechanism of equations 1-4 is correct; however, if some unexpected chain reaction occurs, the results might vary with the steady-state concentrations of the radicals in the solution. In order to check the possible consequences of gradients in the radical concentrations in the solution, wire screens were interposed between the light source and the reaction cell. Although precision of such runs was not as good as usual, the results indicate that the quantum yields are not influenced by this operation. Quantum yields in the benzophenone-toluene system were measured both a t 25’ and a t 52.5’. Since the results were essentially identical, it was surmised that reaction 1 has no more than a small activation energy. Therefore, the waterbath used originally was replaced with an air-bath designed only to hold the temperature close t o room temperature. Table I presents representative data for the reduction of benzophenone in toluene, in cumene and in benzene solutions of toluene and diphenylmethane. Table I1 shows data for the reaction of other ketones with toluene. The low-quantum yields obtained with filter system 1 are attributable to the fact that this filter passed light down to (3) P.A. Leighton and G.S. Forbes, ibid., 62,3139 (1930).

G. S. HAMMOND, W. P. B.IKER AND U'. 11.MOORE

"96

(2UAN.l rlf \'IEI.I)S

Ruu no.

a

156-la 159-10 173-lb 175-lb 173-2b 168-3 171-2 177-lC 177-2' 55-2 61-4 32-1 32-2 59-4 101 102 103 104 105 106c 113 114 116< 117' 118C 6UI-1 64hI-2 T = 26".

IN

TABLE I PHOTOREDUCTIOS OF BENZOPHENONE B Y HYDROCARBONS Initial benzophenone

Filter system

Series

A

Vol. 83

Reductant

x IO',

moles/l.

% Cpnversion of benzophenone

*KetonS

% Yield of pinacoi

Toluene 101 14 0.19 67 Toluene 101 18 .21 64 Toluene 7% 100 20 .19 Toluene 100 22 .21 70 Toluene 67 100 31 ,22 Toluene 300 16 .19 (i3 Toluene .22 72 300 22 Toluene ti5 100 23 .21 Toluene 100 2