Free Radical-Mediated Ketone Synthesis from Alkyl Iodides via Sequential Radical Acylation Approach

Full text of DOI:10.1021/ja9710316

5982
J. Am. Chem. Soc. 1997, 119, 5982-5983
the result indicates that 1a is somewhat less reactive than 3a,
we conceived that this unexpected result would be due to the
steric bulkiness of two phenylsulfonyl groups in 1a.12 Thus,
we next examined the reactivity of 1b.8 In addition to steric
effects, 1b would be more reactive than 1a due to the slightly
lower energy of the LUMO of 1b relative to 1a. As predicted,
when the reaction was carried out with 1b under the similar
conditions, the reaction was much faster and complete within
4 h, yielding 3b in 82% yield along with a small amount of 4
(9%) as shown in eq 1.13 However, in the preparation of more
hindered ketone 6b, the second step involving intermolecular
Free Radical-Mediated Ketone Synthesis from Alkyl
Iodides via Sequential Radical Acylation Approach
Sunggak Kim* and Joo-Yong Yoon
Department of Chemistry, Korea AdVanced
Institute of Science and Technology
Taejon 305-701, Korea
ReceiVed April 1, 1997
Although numerous reports on the synthesis of ketones have
appeared to date,1 a free radical-mediated ketone synthesis is
fairly uncommon, and only two methods are presently available.
Free radical carbonylation approach to a ketone synthesis has
been reported by Ryu et al.2 We have recently reported a novel
free radical acylation approach utilizing radical reactions of
phenylsulfonyl oxime ethers.3 However, the synthesis of
ketones based on the sequential radical acylation approach has
not been reported and seemed a conceptual advance of our
previous findings.3,4 We now report a general approach for the
synthesis of acyclic and cyclic ketones from alkyl iodides with
bis-sulfonyl oxime ether 1 under radical conditions. To our
knowledge, 1 appears to be the first example of a carbonyl
equivalent geminal radical acceptor.5,6
addition of cyclohexyl radical to 3b turned out to be inefficient,
yielding ketone 6b only in 13% yield along with 3b (60%).
Therefore, we turned our attention to somewhat less sterically
hindered 1c.8,14 Gratifyingly, 1c gave much better results as
shown in eq 2. Thus, remaining reactions were carried out with
Before we began our study with 1, standard AM1 calculations
of LUMO energies were performed to predict the reactivity of
1 relative to 3 and indicated that 1 would be more reactive
toward nucleophilic radicals than 3.7 Our initial experiment
was carried out with 1a.8 When 4-phenoxybutyl iodide was
treated with 1a, hexamethylditin, and acetone as a sensitizer in
ethanol at 300 nm for 15 h,9,10 an initial result was discouraging,
yielding 3a in low yield (28%) along with 4 (58%).11 Since
1c. The synthesis of ketones was normally carried out by a
three-step, one-pot procedure. Treatment of an alkyl iodide with
1c, hexamethylditin (1.2 equiv), and acetone (5 equiv) as a
sensitizer in EtOH (0.3 M in the iodide) and irradiation at 300
nm for 3 h followed by addition of another alkyl iodide and
hexamethylditin (1.2 equiv) with an additional irradiation for 7
h at 300 nm afforded ketoxime 5. This ketoxime was further
hydrolyzed with 30% HCHO solution in THF (1:1) containing
a small amount of HCl to yield unsymmetrical ketone 6.
Table 1 summarizes the experimental results and illustrates
the efficiency and scope of the present method. The reaction
worked well with primary alkyl iodides but somewhat less
efficiently with secondary alkyl iodides. It is noteworthy that
stable allylic and benzylic radicals reacted smoothly with 1c.
Acetal, ester, alcohol, and carbamate moieties were all tolerated,
as would be expected from the unique nature of the radical
reactions. In addition, the syntheses of ketones 6i and 6j
demonstrates the mildness of the present approach.
(1) (a) O’Neill, B. T. In ComprehensiVe Organic Synthesis; Trost, B.
M., Fleming, I., Eds.; Pergamon Press: New York, 1991; Vol. 1. Chapter
1.13. (b) Davis, B. R.; Garratt, P. J. In ComprehensiVe Organic Synthesis;
Trost, B. M., Fleming, I., Eds.; Pergamon Press: New York, 1991; Vol. 2.
Chapter 3.6. (c) Larock, R. C. ComprehensiVe Organic Transformations;
VCH Publishers, Inc.: 1989; p 583-818.
(2) For reviews, see: (a) Ryu, I.; Sonoda, N. Angew. Chem., Int. Ed.
Engl. 1996, 35, 1050. (b) Ryu, I.; Sonoda, N.; Curran, D. P. Chem. ReV.
1996, 96, 177.
(3) Kim, S.; Lee, I. Y.; Yoon, J-Y.; Oh, D. H. J. Am. Chem. Soc. 1996,
118, 5138.
(4) Kim, S.; Yoon, J-Y.; Lee, I. Y. Synlett 1997, in press.
(5) Open circles represent radical acceptors.
(6) For the carbonyl radical acceptor synthons, (a) Acylgermanes: Curran,
D. P.; Palovich, M. Synlett 1992, 631. Kiyooka, S.; Kaneko, Y.; Matsue,
H.; Hamada, M.; Fujiyama, R. J. Org. Chem. 1990, 55, 5562. (b) Thio-
and selenoesters: Kim, S.; Jon, S. Y. J. Chem. Soc., Chem. Commun. 1996,
1335. Dowd, P.; Wilk, B. K. J. Am. Chem. Soc. 1992, 114, 7949. (c)
Nitriles: Griller, D.; Schmid, P.; Ingold, K. U. Can. J. Chem. 1979, 49,
1313. Clive, D. L. J.; Beaulieu, P. L.; Set, L. J. Org. Chem. 1984, 49,
1313.
(12) MOPAC low-energy conformations of 1a, 1b, and 1c are shown in
the Supporting Information.
(13) Intermediate 3b is at the oxidation level of a carboxylic acid.
Conversion of 3b into an ester and an amide under mild conditions would
be feasible. We thank a referee for helpful suggestions.
(14) The preparation of 1c (mp 75 °C) is as follows. We had difficulties
in the preparation of less bulky methyl oxime ether (R2 ) Me in 1c) due
to low boiling points of intermediates shown below.
(7) MOPAC calculations (AM1 method, version 6.3) give the following
LUMO energies in eV: 1a (-0.7715), 1b (-1.0098), 1c (-1.1394), 3 (R1
) Ph, R3 ) Me) (-0.5869). Further computational results on HOMO
energies and charges at iminyl carbon atoms are contained in the Supporting
Information.
(8) See the Supporting Information for an experimental procedure and
spectral data.
(9) 1a, 1b, and 1c are slightly soluble in most solvents (C6H6, EtOAc,
and THF) and soluble in alcoholic solvents (MeOH and EtOH). Furthermore,
the use of MeOH as solvent led to the formation of blackish colored
materials in the reaction mixture. This prevented effective radiation, thereby
decreasing the rate or stopping the reaction halfway.
(10) Harendza, M.; Junggebauer, J.; Lebmann, K.; Neuman, W. P.; Tews,
H. Synlett 1993, 286.
(11) Without acidic workup, the oxime ether group was removed during
the reaction.
S0002-7863(97)01031-7 CCC: $14.00 © 1997 American Chemical Society

Communications to the Editor
J. Am. Chem. Soc., Vol. 119, No. 25, 1997 5983
Table 2. Preparation of Cyclic Ketones
Table 1. Preparation of Ketones from Alkyl Iodidesa
a The yield was not optimized. b Reaction time: 6 h. c Bu3SnH/AIBN
was used for the second step.8
15
was obtained in good yield (eq 5).
A similar result was
a The yield was based on R3-I and was not optimized.
realized with 16b.
We next studied sequential inter- and intramolecular acylation
approach leading to cyclic ketones (eq 3). When a mixture of
1,4-diiodobutane (7), 1c (1.2 equiv), hexamethylditin (2.2 equiv),
and acetone (10 equiv) in ethanol (0.3 M in iodide) was
irradiated at 300 nm for 5 h, cyclopentanone benzyl oxime ether
(9) was isolated in 87% yield. As shown in Table 2, similar
results were obtained with other diiodides. Using
allyl bromide and aryl iodide as radical precursors, function-
alized cyclic ketones could be prepared, although the use of
aryl iodide required the standard radical conditions (Bu3SnH/
AIBN).
In conclusion, we have demonstrated the first successful
sequential radical acylation approach, which appears to be highly
useful for the synthesis of various carbocyclic ketones as well
as acylic ketones under mild conditions. Further studies on the
synthesis of polycarbonyl compounds, lactones, and lactams
using 1c and relating reagents are in progress.
Further synthetic utility of the present approach is shown in
eqs 4 and 5. Treatment of 11 with 1c and hexamethylditin in
Acknowledgment. This paper was supported by NON DIRECTED
RESEARCH FUND (01D0374), Korea Research Foundation, 1996.
We thank Dr. Sung-eun Yoo and Ms. Soo Kyung Kim of KRICT for
their help in using the SYBYL program.
Supporting Information Available: Experimental procedures for
the preparation of reagents (1a, 1b, 1c), 6a, 9, 10a, 10f, and 19a as
well as spectral data (1H NMR, 13C NMR, IR, and HRMS) for the
reaction products, computational results for the energies of HOMO and
LUMO of 1a, 1b, 1c, and 3, and low energy conformations of 1a, 1b,
and 1c (19 pages). See any current masthead page for ordering and
Internet access instructions.
JA9710316
EtOH at 300 nm for 4 h and subsequent acidic workup afforded
15 in 65% yield. Apparently, the reaction proceeded via
intermediate 12 and 13. Furthermore, when the reaction was
carried out with 16a under the similar conditions, enone 19a
(15) Hydrolysis of 18 with aqueous HCHO/HCl was unsuccessful and
18 was recovered unchanged. Thus, 18 was hydrolyzed into 19 by treatment
with pyruvic acid and NaOAc in AcOH-H2O (2:1) at 120 °C for 6 h.16
(16) Hershberg, E. B. J. Org. Chem. 1948, 13, 542.