DEOXYGENATION OF ARYL KETONES
2929
Upon completion of the reaction, the aryl alkane could be isolated after suitable
workup as described in the experimental section. All of the aryl alkanes prepared
in this study are known. Therefore, product identities were confirmed by comparison
of NMR and mass spectral data with authentic samples or literature spectra. In some
instances, trace amounts of other reduction products were observed in addition to
the deoxygenated aromatic compound. In these cases, the aromatic ring moiety
underwent various degrees of reduction. As these products represented a very minor
component of the product mixture (<5%), this did not cause yields to suffer on the
reaction time scales we were using.
To demonstrate that the presence of Raney nickel was indeed required, a con-
trol experiment was run in which benzophenone (0.506 g, 2.78 mmol) was refluxed in
2-propanol (15 mL) for 3 h (no Raney nickel added). No conversion to diphenyl-
methane was observed. By comparison, in the presence of Raney nickel (2.5 g), after
1 h of heating, nearly quantitative conversion to diphenylmethane had been realized.
This established that the presence of Raney nickel is necessary for the reaction to
occur.
To verify that the mechanism of reduction was indeed CTH, an experiment was
set up to verify that acetone was being generated as a by-product of the redox reac-
tion. Hexanophenone (0.512 g, 2.64 mmol) was refluxed in a magnetically stirred sus-
pension of Raney nickel (2.5 g) in 2-propanol (15 mL) for 0.5 h with the container
open to the atmosphere. A Dean–Stark trap was used to collect distillate over the
course of the reaction. The distillate was tested for the presence of acetone by mixing
several drops of a 3% solution of 2,4-dinitrophenylhydrazine with a few drops of dis-
tillate. The formation of a yellow precipitate suggested a positive test for the presence
of acetone. Examination of the melting point of the precipitate revealed a melting
point identical to that of acetone 2,4-dinitrophenylhydrazone (126–128 ꢀC).[12]
A
source has reported that this test easily detects 1 part of acetone in 500 to 1000 parts
of 2-propanol.[13]
In our study, a 1:5 (w=w) substrate-to-catalyst ratio was used to deoxygenate
the aryl ketones. Because this is a relatively large excess of Raney nickel, we were
interested in determining whether the Raney nickel could be recycled. To verify that
the deoxygenation reaction was indeed catalytic with respect to the Raney nickel, an
experiment was set up to test whether the catalyst could be recycled without any loss
of activity. We found that a given batch of Raney nickel catalyst could be used
repeatedly (at least six times) without any loss of activity. Prior to reuse, the catalyst
was washed with 2-propanol (3 Â 10 mL). In a given run, benzophenone (0.50 g,
2.7 mmol) was refluxed in 2-propanol (10 mL) for 1 h. Yields for the reaction were
greater than 93% for each run.
Since other Raney catalysts have been observed to be effective at reducing
benzylic alcohols,[11] we were curious to see if these would also be effective at reduc-
ing aryl ketones. The two additional Raney catalysts explored were Raney cobalt
and Raney copper. For each catalyst, a reaction was set up in which benzophenone
(0.25 g, 1.4 mmol) was refluxed in a magnetically stirred suspension of the Raney
catalyst (1.25 g) in 2-propanol (10 mL) for 1 h with the container open to the atmos-
phere. No change was observed with either catalyst. The experiments were repeated
twice, once using acid catalyst (1 drop conc. HCl) and once using base catalyst (1 pel-
let solid NaOH), but again, in each case, unreacted starting material was recovered.