Carbon-carbon bond formation by electrochemical catalysis in conductive microemulsions

Article abstract of DOI:10.1021/jo9608477

Bicontinuous microemulsions made from oil, water, and surfactants were examined as substitutes for organic solvents in carbon-carbon bond-forming reactions. Conjugated additions of primary alkyl iodides 3a-c to 2-cyclohexen-1-one (4) to give 3-alkyl cyclohexanones 5a-c and cyclization of 2-(4-bromobutyl)-2-cyclohexen-1-one (9) to 1-decalone (10) were mediated by the Co(I)L complex vitamin B_12s generated at carbon cloth electrodes under mild conditions. Reaction of the Co(I)L nucleophile with the alkyl halides gives a Co-alkyl complex. Cleavage of the Co-alkyl complexes by using an electrode potential of -0.85 V (all vs SCE) and irradiation with visible light, or a potential of -1.45 V in the dark, were compared. Addition of the resulting alkyl radicals to the activated double bonds gave comparable yields of 3-alkylcyclohexanone 5a-c (70-80% using -0.85 V + light) and 1-decalone (90%, both cleavage modes) 10 in microemulsions and in DMF. Microemulsions containing hexadecyltrimethylammonium bromide (CTAB) gave remarkable stereoselectivity for the trans isomer of 10, while homogeneous DMF and a sodium dodecylsulfate (SDS) microemulsion gave little stereoselectivity.

Full text of DOI:10.1021/jo9608477

5972
J . Org. Chem. 1996, 61, 5972-5977
Ca r bon -Ca r bon Bon d F or m a tion by Electr och em ica l Ca ta lysis in
Con d u ctive Micr oem u lsion s
J ianxin Gao, J ames F. Rusling,* and De-ling Zhou
Department of Chemistry, Box U-60, University of Connecticut, Storrs, Connecticut 06269-4060
Received May 8, 1996X
Bicontinuous microemulsions made from oil, water, and surfactants were examined as substitutes
for organic solvents in carbon-carbon bond-forming reactions. Conjugated additions of primary
alkyl iodides 3a -c to 2-cyclohexen-1-one (4) to give 3-alkyl cyclohexanones 5a -c and cyclization
of 2-(4-bromobutyl)-2-cyclohexen-1-one (9) to 1-decalone (10) were mediated by the Co(I)L complex
vitamin B12s generated at carbon cloth electrodes under mild conditions. Reaction of the Co(I)L
nucleophile with the alkyl halides gives a Co-alkyl complex. Cleavage of the Co-alkyl complexes
by using an electrode potential of -0.85 V (all vs SCE) and irradiation with visible light, or a
potential of -1.45 V in the dark, were compared. Addition of the resulting alkyl radicals to the
activated double bonds gave comparable yields of 3-alkylcyclohexanone 5a -c (70-80% using -0.85
V + light) and 1-decalone (90%, both cleavage modes) 10 in microemulsions and in DMF.
Microemulsions containing hexadecyltrimethylammonium bromide (CTAB) gave remarkable
stereoselectivity for the trans isomer of 10, while homogeneous DMF and a sodium dodecylsulfate
(SDS) microemulsion gave little stereoselectivity.
In tr od u ction
In electrochemical catalytic reductions, electrons are
delivered from electrodes to reactants by a mediator.
Kinetic studies of the electrochemically mediated forma-
tion of olefins from alkyl vicinal dibromides showed that
rate enhancement can be achieved by reactant precon-
centration at the electrode surface4c or by adjusting the
formal potential of the mediator.7 Benzyl radical coup-
ling mediated by vitamin B12 was recently achieved in a
microemulsion.8
Alternatives to organic solvents for synthesis are
currently receiving attention because of environmental
concerns. One alternative is microemulsions,1,2 which we
have been exploring for electrochemical catalysis.3-8
Microemulsions are clear, thermodynamically stable
fluids made from water, oil, and surfactants. They are
less toxic and often less expensive than alternative
organic solvents.1-3
Conductive fluids suitable for electrochemical synthesis
include oil-in-water (o/w) microemulsions (µEs) featuring
surfactant-coated oil droplets in a continuous water phase
and bicontinuous microemulsions composed of dynamic
intertwined microscopic networks of oil and water with
surfactant at the interfaces.1-3 Bicontinuous microemul-
sions have good solubility for polar and nonpolar com-
pounds with faster mass transport for both types of
solutes.4b Tuning reactivity by controlling microemulsion
composition is also possible.7 Reduction potential win-
dows are extended by adsorption of surfactant on the
electrode, which partly blocks the reduction of water.9
Co(I)L forms of vitamin B12,7,8,10-14 cobaloximes,15a,16 Co-
(salen),15a,17 Co(salophen),15 and other cobalt complexes18
react with alkyl halides to give alkylcobalt complexes.
The carbon-cobalt bonds can be cleaved by visible light,
electrolysis, or reducing agents to give carbon-centered
radicals that can add in-situ to activated alkenes to form
carbon-carbon bonds (eq 2).10-18 Such cobalt-complex-
mediated bond formation has been used for the syntheses
of a variety of natural products.10,11,19,20 Syntheses of
(9) Texter, J . In Electrochemistry in Colloids and Dispersions;
Mackay, R. A., Texter, J ., Eds.; VCH Publishers: New York, 1992; pp
3-21.
* To whom correspondence should be addressed. Tel.: (860) 486-
4909. Fax: (860) 486-2981. E-mail: Jrusling@nucleus.chem.uconn.edu.)
X Abstract published in Advance ACS Abstracts, August 1, 1996.
(1) Bourrel, M.; Schechter, R. S. Microemulsions and Related
Systems; Marcel Dekker: New York, 1988.
(2) (a) Fendler, J . H. Membrane Mimetic Chemistry; Wiley: New
York, 1982. (b) O’Connor, C. J .; Lomax, T. D.; Ramage, R. E. Adv.
Colloid Interface Sci. 1984, 20, 21-97.
(10) Scheffold, R.; Rytz, G.; Walder, L. In Modern Synthetic Methods;
Scheffold, R., Ed.; Wiley: New York, 1983; Vol. 3, pp 355-439.
(11) Scheffold, R.; Abrecht, S.; Orlinski, R.; Ruf, H.-R.; Stamonli,
P.; Tinembart, O.; Walder, L.; Weymuth, C. Pure Appl. Chem. 1987,
59, 363-372.
(12) (a) Dodd, D.; J ohnson, M. D. Organomet. Chem. Rev. 1973, 52,
1-232. (b) Pratt, J . M.; Craig, P. J . Adv. Organomet. Chem. 1973, 11,
331.
(3) (a) Rusling, J . F. In Modern Aspects of Electrochemistry; Conway,
B. E., Bockris, J . O’M., Eds.; Plenum Press: New York, 1994; No. 26,
pp 49-104. (b) Rusling, J . F. In Electroanalytical Chemistry; Bard, A.
J ., Ed.; Marcel Dekker: New York, 1994; Vol. 18, pp 1-88.
(4) (a) Owlia, A.; Wang, Z.; Rusling, J . F. J . Am. Chem. Soc. 1989,
111, 5091-5098. (b) Iwunze, M. O.; Sucheta, A.; Rusling, J . F. Anal.
Chem. 1990, 62, 644-649. (c) Kamau, G. N.; Hu, N.; Rusling, J . F.
Langmuir 1992, 8, 1042-1044.
(13) (a) Scheffold, R.; Dike, M.; Dike, S.; Herold, J .; Walder, L. J .
Am. Chem. Soc. 1980, 102, 3642-3644. (b) Scheffold, R. Chimia 1985,
39, 203. (c) Scheffold, R.; Orlinski, R. J . Am. Chem. Soc. 1983, 105,
7202-7203. (d) Ruhe, A.; Walder, L.; Scheffold, R. Helv. Chim. Acta
1985, 68, 1301.
(14) (a) Zhou, D.-L.; Walder, P.; Scheffold, R.; Walder, L. Helv. Chim.
Acta 1992, 75, 995. (b) Tinembart, O.; Walder, L.; Scheffold, R. Ber.
Bunsenges. Phys. Chem. 1988, 92, 1225-1231.
(15) (a) Pattenden, G. Chem. Soc. Rev. 1988, 17, 361-382 and
references cited therein. (b) Bhandal, H.; Howell, A. R.; Patel, V. F.;
Pattenden, G. J . Chem. Soc., Perkin Trans. 1 1990, 2709-2714.
(16) (a) Inokochi, T.; Tsuji, M.; Kawafuchi, H.; Torri, S. J . Org. Chem.
1991, 56, 5945-5948. (b) Torri, S.; Inokochi, T.; Yukama, T. J . Org.
Chem. 1985, 50, 5875-5877.
(5) Iwunze, M. O.; Hu, N.; Rusling, J . F. J . Electroanal. Chem. 1992,
333, 353-361.
(6) (a) Zhang, S.; Rusling, J . F. Environ. Sci. Technol. 1993, 27,
1375-1380. (b) Schweizer, S.; Rusling, J . F.; Huang, Q. Chemosphere
1994, 28, 961-970. (c) Huang, Q.; Rusling, J . F. Environ. Sci. Technol.
1995, 29, 98-103. (d) Zhang, S.; Rusling, J . F. Environ. Sci. Technol.
1995, 29, 1195-1199.
(7) Zhou, D.-L.; Gao, J .; Rusling, J . F. J . Am. Chem. Soc. 1995, 117,
1127-1134.
(8) Zhou, D.-L.; Carrero, H.; Rusling, J . F. Langmuir 1996, 12,
3067-3074.
(17) Fry, A. J .; Singh, A. H. J . Org. Chem. 1994, 59, 8172-8177.
(18) Ozaki, S.; Nakanishi, T.; Sugiyama, M.; Miyamoto, C.; Ohmori,
H. Chem. Pharm. Bull. 1991, 39, 31-35.
(19) Auer, L.; Weymuth, C; Scheffold, R. Helv. Chim. Acta 1993,
76, 810-818.
S0022-3263(96)00847-X CCC: $12.00 © 1996 American Chemical Society

Electrochemical Catalysis in Conductive Microemulsions
J . Org. Chem., Vol. 61, No. 17, 1996 5973
Ta ble 1. Rea ction Med ia , Sp ecific Con d u ctivity, a n d
F or m a l P oten tia l of Co(II)L/Co(I)L Red ox Cou p le of
vita m in B12
specific
conductivity
medium
composition
Ω-1‚cm-1
-E°′ a
CTAB µEb CTAB:1-pentanol:tetradecane:
H2O ) 17.5:35:12.5:35 (wt %)
2.0 × 10-3
7.0 × 10-3
0.79
SDS µE
SDS:1-pentanol:tetradecane:
brinec ) 13.3:26.7:8:52 (wt %)
0.2 M TBAB/DMF
0.73
DMF
3.5 × 10-3
0.71d
a
b
Co(II)L/Co(I)L redox couple of vitamin B12
.
CTAB µE: cetyl-
trimethyl ammounium bromide microemulsion. c 0.1 M NaCl.
Data from ref 7.
d
olefins can be achieved directly via mediated reductive
elimination of vicinal dihalides (eq 1).11,21,22
To our knowledge, dechlorinations of organic halides
to hydrocarbons are thus far the only synthetic scale
electrochemical catalytic reactions in microemulsions
that have been reported.6a,b,d,8 Microemulsions should be
useful for reactions of more general synthetic utility. In
this paper, we compare carbon-carbon bond-forming
reactions in microemulsions and DMF. Comparable
yields were obtained in DMF and microemulsions for all
reactions. Remarkable stereoselectivity was obtained for
an intramolecular cyclization in a microemulsion made
with a quaternary alkylammonium salt surfactant.
Resu lts
Micr oem u lsion s. While several microemulsions were
originally explored, two fluids with high conductivities
(Table 1) were found to be superior for a test reaction,
reductive elimination of trans-1,2-dibromocyclohexane 1
(DBCH). For example, reductive elimination of DBCH
to cyclohexene 2 at a carbon cloth electrode at -0.85 V
vs SCE gave yields of 98% in the CTAB microemulsion
and 97% in DMF/0.2 M tetrabutylammonium bromide
(TBAB) with consumption of about 2 F/mol of electricity.
Specific conductivities of the microemulsions are simi-
lar to that of 0.2 M TBAB in DMF, commonly used for
electrochemical reactions. These microemulsions were
previously characterized as bicontinuous.23 Formal po-
tentials of the Co(II)L/Co(I)L redox couple of vitamin B12
are also listed in Table 1.
Volta m m etr y. Cyclic voltammetry (CV) was used to
choose appropriate electrode potentials for electrosyn-
thesis. The vitamin B12 Co(II)L/Co(I)L redox couple gave
a reversible CV at relatively low scan rates in all media
(Figure 1a). A large increase in the reduction peak for
Co(II)L after addition of DBCH was observed (Figure 1a),
along with the disappearance of the oxidation peak. This
is consistent with the catalytic debromination of DBCH.7
F igu r e 1. Cyclic voltammograms at 0.1 V s-1 on a glassy
carbon electrode in CTAB microemulsions: (a) (---) 1 mM
vitamin B12 + 1 mM DBCH, (s) 1 mM vitamin B12 alone; (b)
(---) 1 mM vitamin B12 alone, (s) 1 mM vitamin B12 + 6 mM
n-dodecyl iodide; (c) (---) 1 mM vitamin B12 alone, (s) 1 mM
vitamin B12 + 1 mM 2-cyclohexen-1-one (4); (d) (---) background
of CTAB microemulsion, (s) 6 mM n-dodecyl iodide.
(20) Hemamalini, S.; Scheffold, R. Helv. Chim. Acta 1995, 78, 447-
451.
(21) Lexa, D.; Saveant, J . M.; Schafer, H. J .; Su, K.-B.; Vering, B.;
Wang, D. L. J . Am. Chem. Soc. 1990, 112, 6162-6177.
(22) Lexa, d.; Saveant, J . M.; Su, K.-B.; Wang, D. L. J . Am. Chem.
Soc. 1987, 109, 6464-6470.
(23) (a) Ceglie, A.; Das, K. P.; Lindman, B. Colloids Surf. 1987, 28,
29-40. (b) Georges, J .; Chen, J . W. Colloid Polym. Sci. 1986, 264, 896-
902.
Oxidative addition of the alkyl iodides 3a -c to B12
Co(I)L gives an alkylcobalt complex that shows a new

5974 J . Org. Chem., Vol. 61, No. 17, 1996
Gao et al.
Ta ble 2. Red u ction P oten tia ls of Su bstr a tes a n d
Alk ylcoba lt Com p lexes in Micr oem u lsion s a n d DMF
direct reduction concn of vit B12
,
substrate medium Ep,a
V
potential,b V
mM
3a
3b
3c
3c
3c
9
9
9
4
CTAB µE -1.51
-2.37
-2.31
-2.40
c
1.0
0.9
0.6
1.0
1.1
1.0
0.72
1.0
0.96
CTAB µE -1.47
CTAB µE -1.47
DMF
SDS µE
SDS µE
CTAB µE -1.45
DMF
-1.42
-1.48d
-1.45d
c
c
-1.82e
-2.18e
-1.71e
-1.40
CTAB µE
a
Peak potential of reduction of Co-C bond measured by CV
b
at scan rate 100 mV/s on glassy carbon electrode. Measured by
CV at scan rate 100 mV/s vs SCE on glassy carbon electrode. The
direct reduction potential is out of the potential window. This
reduction peak is overlapped with a background peak. The
c
d
e
reduction potential of the conjugated carbonyl group.
peak in the range -1.40 to -1.50 V, with disappearance
of the oxidation peak for Co(I)L (Figure 1b). The new
irreversible reduction peak represents cleavage of the
alkyl-cobalt bond.
Photolytic cleavage after formation of the alkyl-Co
complex gave much higher yields of addition products
(Table 3, entries 1, 3, and 5). The number of electrons
consumed per mole of iodide (n) was close to 1 in this
mode. Yields of 5c were similar in microemulsions and
in DMF. In a control experiment with no catalyst (Table
3, entry 7), only a small amount of 5c was formed by
photolytic cleavage of the carbon-iodine bond of 3c and
subsequent radical addition to 4.
After activated alkene 4 was added to microemulsions
containing only vitamin B12, cathodic and anodic peaks
of Co(II)L/Co(I)L were unchanged (Figure 1c). The
irreversible peak at -1.71 V represents direct reduction
of 4. Peaks for direct, irreversible reduction of alkyl
iodides in the CTAB microemulsion were at -2.3 to -2.4
V (Figure 1d and Table 2).
A large mole ratio (m) of 4 to alkyl iodide was necessary
to obtain good yields of coupling product 5 (Table 4).
Alkene 7c was the major product from mediated elec-
trolysis of 3c at m ) 1, but the amount formed was small
at m ) 300. A mole ratio of m ) 40 was sufficient for
g70% yields of addition products in CTAB microemul-
sions. In 0.2 M TBAB/DMF with m ) 300, a yield of 5c
similar to those in microemulsions was found using the
photolytic cleavage mode (Table 3, entry 9).
In the absence of vitamin B12 in the microemulsions,
9, the starting material for the cyclization, gave an
irreversible peak at -1.82 V from reduction of the
conjugated carbonyl group (Table 2). In DMF, reduction
potentials of alkyl-Co complexes were slightly more
positive than in the microemulsion.
In ter m olecu la r Ad d ition . On the basis of the above
results, photolytic cleavage of alkyl-Co complexes was
done by using carbon cloth cathodes at an applied
potential of -0.85 V, where formation of alkyl-Co
complexes occurs following generation of Co(I)L. Direct
electrochemical cleavage was done at -1.45 V. In the
presence of vitamin B12 and 4, alkyl iodides 3a -c gave
addition products 5a -c, alkanes 6a -c and 8a -c, and
alkenes 7a -c (eq 2 and Table 3).
Smaller yields of 5a -c were obtained when electro-
chemical cleavage of the alkyl-Co complexes at -1.45 V
was used (Table 3, entries 2, 4, and 6). The amount of
6c and the n value increased compared to photolytic
cleavage. This suggests that reductive dehalogenation
competes with the addition reaction at the negative
potentials required to cleave the alkyl-Co complex.
Ta ble 3. In ter m olecu la r Ad d ition in Micr oem u lsion s a n d in DMF
productsb (mol %)
entry
1
iodide
medium
-E, V vs SCE
F/mol 3a -ca (n)
time (h)
4
5a -c
6a -c
7a -c
8a -c
3a
CTAB µE
0.85
0.8
81 (5a )
c
c
c
with light
1.45
0.85
with light
1.45
0.85
with light
1.45
0.85
2
3
3a
3b
CTAB µE
CTAB µE
45 (5a )
68 (5b)
c
c
c
c
c
c
1.0
3
2
4
5
3b
3c
CTAB µE
CTAB µE
39 (5b)
71 (5c)
c
c
c
0.8
1.8
19 (6c)
9 (7c)
2 (8c)
6
3c
3c
CTAB µE
CTAB µE
3
4
42 (5c)
14 (5c)
42 (6c)
5 (6c)
9 (7c)
1 (7c)
3 (8c)
7d
with light
no catalyst
0.75
with light
0.85
8
9
3c
3c
SDS µE
0.9e
1.1e
5.6
1.5
72 (5c)
69 (5c)
14 (6c)
13 (6c)
6 (7c)
e1 (8c)
e1 (8c)
0.2 M
15 (7c)
TBAB/DMF
with light
a
b
After background current was substracted; mole ratio m ) 40 unless otherwise noted. GC yields based on 3a -c. c Volatile byproducts
not trapped for reactions of 3a ,b. 80% of 3c was recovered. e Mole ratio m ) 300.
d

Electrochemical Catalysis in Conductive Microemulsions
J . Org. Chem., Vol. 61, No. 17, 1996 5975
Ta ble 4. P r od u ct Distr ibu tion s of In ter m olecu la r C-C
Ad d ition a t Differ en t Mole Ra tios of 4/3c in CTAB
Micr oem u lsion a
Discu ssion
Results demonstrate that microemulsions are suitable
fluid media for synthetic applications using electrochemi-
cal catalysis. Yields of olefin from DBCH and of the more
complex intermolecular carbon-carbon addition reactions
were similar in microemulsions and DMF. The latter
reaction gave higher yields by using electrochemical
Co(I)L formation and photolytic alkyl-Co bond cleavage
compared to the purely electrochemical approach.
For the intramolecular cyclization, remarkable stereo-
selectivity was obtained in the CTAB microemulsion
compared to DMF or the SDS microemulsion. The trans:
cis ratio of 14 for the formation of 10 in the CTAB
microemulsion, obtained by both the electrochemical/
photochemical or pure electrochemical method, was much
larger than that in any other fluid. Results suggest that
the stereoselectivity is related to a property of CTAB, as
will be discussed below.
productsc (mol %)
mole ratio F/mol of time
entry
of 4/3c, m
3c,b n
(h)
5c
6c
7c
8c
1
2
3
4
5
1
5
10
40
300
0.8
0.9
0.8
1.0
1.1
4
5
4
2
4
14
44
50
71
74
24
23
20
19
13
61
29
26
9
1
2
1
2
0
4
a
At -0.85 V vs SCE under the irradiation of light. b Background
current was subtracted. c GC yields based on 3c.
In tr a m olecu la r Cycliza tion . Here, 9 was the start-
ing material and photolytic and electrochemical alkyl-
Co cleavage modes were tested in microemulsions and
DMF. Both methods gave comparable results.
Reduction potentials of the radicals formed by cleavage
of the Co-alkyl bond can be used to predict whether the
reactions have radical or anionic pathways. Reduction
potentials of primary alkyl radical R• to carbanion R-
cannot be measured directly by voltammetry of the
corresponding organic halide, which gives a single peak
for two-electron reduction leading to the cabanion.24
However, a standard potential of -1.3 to -1.42 V vs SCE
was estimated for the primary butyl radical in DMF by
Saveant et al.25a This is much more negative than
potentials used in our experiments with photolytic cleav-
age. Comparing the direct reduction potentials for n-
butyl iodide of -2.37 V (Table 2) in CTAB microemulsion
and -2.33 V25b in DMF suggests that the difference in
the redox potential of R•/R- in CTAB microemulsion and
DMF may be small.
On the basis of the above and previous work,17,25c the
photochemical cleavage mode with E ) -0.85 V should
cleave LCo(III)-R to give R• radicals. First, nucleophile
Co(I)L is produced by reducing vitamin B12a at the
electrode (Scheme 1). Oxidative addition then gives
RCo(III)L. Photoactivated [RCo(III)L]* A, going through
a transition state [R••Co(II)L], leads to radical B (eq 7)
or dehydrocobaltation (eq 6), as reported previously,15 to
form alkene. Alkyl radical B reacts rapidly with acti-
vated alkene C (Z ) electron withdrawing group) to form
a new radical D, which is an electrophile because a
carbonyl group is attached to the radical center.26 H-
atom abstraction by D from some component in the
medium yields the final addition products 5a -c.
The reactive radical B also abstracts H• from a donor
in the fluid medium to form 6a -c (eq 10). Radical B also
can disproportionate to form 6a -c and 7a -c (eq 9)24b and
dimerize to form 8a -c (eq 11). An E2 elimination
The main products were trans and cis isomers of 10
(Table 5). Less than 2% of 11 was obtained. Two GC
peaks had GC/MS ion peaks at m/ e ) 150, but no
fragment ion peak at m/ e ) 95 that would be charac-
teristic of a 2-alkyl-2-cyclohexen-1-one, such as 11. These
peaks may represent 12 and 13.
Product distributions in the CTAB microemulsion were
similar for both alkyl-Co bond cleavage modes, with a
trans:cis ratio of 14 for 10 (Table 5). In DMF, this ratio
was nearly one without acid, but increased to 2.5 when
acetic acid was added. The ratio was 1.6 in the SDS
microemulsion.
In DMF, the n value was doubled after acetic acid
(Table 5, entry 4) was added. Without acid more 12 and
13 were formed (Table 5, entry 3). Similar results in
DMF were reported by Scheffold et al.13a for the same
reaction in DMF with 0.1 N LiClO4, and 0.05 N NH4Br
at a Hg electrode at -1.54 V. In that work, 95% yields
of 10 with a trans:cis ratio of unity were obtained. When
CH3OH was the solvent, the trans isomer predominated.
Ta ble 5. In tr a m olecu la r Cycliza tion in CTAB Micr oem u lsion a n d DMF
productsb (mol %)
entry
medium
F/mol of 9,a
n
time (h)
-E, V
1.45
10 (trans:cis)
11
12 + 13c
1
2
CTAB µE
CTAB µE
2.1
0.9
3.4
5
90 (14:1)
89 (14:1)
2
2
5
4
0.85
with light
0.85
with light
0.85
with light
0.85
with light
3
4
5
DMF/0.2 M TBAB
0.9
1.8
0.8
1.5
2.6
4.0
74 (2.5:1)
88 (1.0:1)
75 (1.6:1)
1
1
0
17
6
DMF/0.2 M TBAB/0.12 M acetic acid
SDS µE
17
a
b
Background current was subtracted. GC analysis results based on 9. c Identities uncertain (see text), amounts estimated using the
same GC response factor as 10.

5976 J . Org. Chem., Vol. 61, No. 17, 1996
Gao et al.
Sch em e 1
involving the starting alkyl iodides (eq 12) may possibly
contribute a small amount of 7a -c.24
than in entry 3 (Table 5) may involve protonation of a
second intermediate complex.14a Similarly, Scheffold et
al.14b reported that the radical •CH2COOMe was not
reduced to the respective carbanion at -1.4 V in DMSO,
and a one-electron pathway was favored during the
cleavage of the C-Co bond. In the presence of acetic acid,
a two-electron pathway was favored, leading to CH3-
COOMe.
High stereoselectivity in the cyclization of 9 was found
only in the CTAB microemulsion (Table 5). While the
reasons are under further study, the results can be
rationalized on the basis of kinetic control of addition of
a H-atom to the intermediate radical D (Scheme 1). A
similar rationale may be invoked in a two-electron
pathway (Scheme 2, discussed below), where the key
reaction is proton donation to an anion similar to D. In
either case, a slower reaction would give a preference for
the more stable trans isomer of 10. If CTAB stabilizes
the intermediate radical or anion, the trans isomer would
be preferred. Very fast kinetics would lead to equal
amounts of cis and trans isomers, as found in the other
media. Also, specific interfacial influences may be in-
volved in the CTAB microemulsion.
In the electrochemical cleavage mode (-1.45 V) in the
CTAB microemulsion, the results (Table 3, entries 2, 4,
and 6; Table 5, entry 1) are consistent with a two-electron
pathway (Scheme 2).11 At the potentials used, interme-
diate radicals derived from alkyl halide reactants are
likely to be reduced to anions. Lower yields of 5a -c are
consistent with the formation of carbanion R- and its
subsequent fast protonation, which leads to more 6c
(Table 3, entry 6). The yield of 10 (Table 5, entry 1) may
not be affected by formation of R- because intramolecular
cyclization is much faster.
There may also be competition between eqs 6 and 7. If
the mole ratio m is large, radical B can be trapped by C
quickly, eq 7 may predominate, and more addition
product 5c may be formed (Table 4). Since the amount
of 6c and 8c did not change much when the ratio m
increases from 1 to 300, other radical followup reactions
(eqs 9-11) are probably not much faster than the desired
radical trapping (eq 8). If m is small, fewer radicals B
may be trapped by C. Then dehydrocobaltation may
yield more alkene (eq 6). Results in Table 3 (entries 1,
3, 5, 8, and 9) are also consistent with Scheme 1.
Radical cyclization is typically faster than intermo-
lecular radical addition.26 Thus, higher yields of in-
tramolecular cyclization products were obtained. Other
competitive radical and elimination reactions (eqs 6 and
9-11) were minimized in the CTAB microemulsion
(Table 5, entry 2). Radical disproportionation, self-
coupling, and E2 elimination products were not observed
here. The n value of 0.9 is consistent with the one-
electron pathway in Scheme 1.
The results of photolytic/electrochemical cyclizations
in DMF and SDS microemulsions (Table 5, entries 3 and
5) are also consistent with Scheme 1. Lower yields of
10 may be caused by the lack of a good H-atom donor in
the media. In the presence of acid in DMF (Table 5, entry
4), an 88% yield of 10 may indicate that acetic acid is a
better H-atom donor. Consumption of more electrons
(24) (a) Cleary, J . A.; Mubarak, M. S.; Vieira, K. L.; Anderson, M.
R.; Peters, D. G. J . Electroanal. Chem. 1986, 198, 107-124. (b) Vieira,
K. L.; Peters, D. G. J . Org. Chem. 1986, 51, 1231-1239.
(25) (a) Andrieux, C. P.; Gallardo, I.; Saveant, J . M. J . Am. Chem.
Soc. 1989, 111, 1620-1626. (b) Andrieux, C. P.; Gallardo, I.; Saveant,
J . M.; Su, K.-B. J . Am. Chem. Soc. 1986, 108, 638-647. (c) Zhou, D.-
L.; Tinembart, O.; Scheffold, R.; Walder, L. Helv. Chim. Acta 1990,
73, 2225-2241.
In summary, the results reported here show that
conductive microemulsions are synthetically useful media
for carbon-carbon bond formation mediated by electro-
(26) Giese, B. Radicals in Organic Synthesis: Formation of Carbon-
Carbon Bonds; Pergamon: Oxford, 1986.
chemically generated Co(I)L from vitamin B12a
. Pho-

Electrochemical Catalysis in Conductive Microemulsions
J . Org. Chem., Vol. 61, No. 17, 1996 5977
Sch em e 2
spectra were recorded on a Hewlett-Packard GC/MS that had
a 5970 quadrupole mass spectrometric detector attached to a
HP 5890 GC operated at the same conditions as above for every
product. The GC response factors of 1, 2, 3a -c, 5a -c, 6c,
7c, 8c, 9, 10, and 11 were measured for determining the mol
% of these products.
Con tr olled P oten tia l Electr olyses. A PARC Model 273
potentiostat and a H-type divided cell with water jacket were
used for controlled potential electrolyses. Working and counter
electrodes were 2 × 1 cm carbon cloth (National Electrical
Carbon Corp.). The reference electrode was an SCE. A Kratos
Xenon SS 1000 lamp was used as a light source, with the
power supply set at 500 W. Electrolyses were done at 25 °C
with magnetic stirring and a stream of argon passing through
the cathode solution. When visible light was used, the lamp
was positioned 20 cm away from the cell and filtered by a
Schott glass filter (Melles Griot), with below 480 nm wave-
length cutoff.
Reductive elimination of DBCH was done with 122.2 mg
(0.5 mmol) of DBCH (99%), 34.57 mg (0.025 mmol) of vitamin
B12a (98%) in 10 mL of CTAB microemulsion, or 0.2 M TBAB/
DMF at -0.85 V. Reaction mixtures were passed though a
silica gel column with toluene as eluent. A volume of 100 mL
was collected and analyzed by GC or GC/MS.
For intermolecular additions, 3c is given as an example.
The reaction was performed with 30.23 mg (0.1 mmol) of 3c
(98%), 28.22 mg (0.02 mmol) of vitamin B12a at a variety of
mole ratios 4/3c in 10 mL microemulsions, or 10 mL of 0.2 M
TBAB/DMF at -0.85 V with light or at -1.45 V without light.
The reaction mixtures were passed through a silica gel column
using n-hexane/ethyl acetate (19:1, v/v) as eluent. After
concentration to 10 mL under vacuum, the eluent was ana-
lyzed by GC or GC/MS.
tolytic cleavage of C-Co bonds gave higher yields of
intermolecular addition products than electrochemical
bond cleavage. Cyclizations give excellent yields of 10
for both C-Co bond cleavage methods. The CTAB
microemulsion provided excellent stereoselectivity for
trans isomer of 10. While products were not generally
isolated, a workup involving treatment of the product
mixture by passing twice through silica gel columns and
evaporating the solvent (see the Experimental Section)
gave 97% pure 5c.
Exp er im en ta l Section
Ch em ica ls. Dimethylformamide (J . T. Baker) was purified
by passing through a neutral activated alumina (Aldrich)
column.27 Tetrabutylammonium bromide (Aldrich) used for
CV measurements was recrystallized from ethyl acetate (Al-
drich). Other commercially available chemicals were used as
received. Distilled water was purified with a Barnstead
Nanopure system to a specific resistance >15 MΩ cm. A YSI
Model 3400 cell and a YSI Model 35 Conductance Meter were
used for conductance measurements. Cyclic voltammetry was
done with a BAS-100B electrochemical analyzer (Bioanalytical
Systems) as described previously.6 All potentials are referred
to aqueous saturated calomel electrode (SCE).
2-(4-Bromobutyl)-2-cyclohexen-1-one (9),28 2-butyl-2-cyclo-
hexen-1-one (11),28 and 5a -c29 were prepared according to the
literature. 9 and 11 were purified by silica gel (EM Science,
70-230 mesh) chromatography with methylene chloride as
eluent. 5a -c were purified by silica gel chromatography with
n-hexane/ethyl acetate (19:1, v/v) as eluent. Microemulsions
were prepared by mixing the components in the desired ratios
(Table 1), as described previously.23 For the SDS microemul-
sion, tetradecane was used instead of dodecane.23b
P r od u ct An a lysis. A Hewlett-Packard 5890 gas chro-
matograph (GC) with a thermal conductivity detector (TCD)
was used for product analysis. The following operating
conditions were employed: column, methyl silicone (10 m ×
0.53 mm i.d.); carrier gas, helium.
For intermolecular addition products, tR′ of 5a is 7.0 min,
tR′ of 5b is 10.7 min (temperature programmed from 50 to 250
°C at 15 °C/min), and tR′ of 5c is 10.2 min (temperature
programmed from 80 to 250 °C at 15 °C/min). For intramo-
lecular cyclization products, tR′ of 10 is 5.4 min (trans) and
5.6 min (cis) (temperature programmed from 70 to 250 °C at
10 °C/min).
To isolate 5c, the eluant from the first silica gel column was
concentrated and then passed through a second silica gel
column, and the solvent was removed completely from the
second effluent under vacuum to yield a light yellow oil, which
was 97% 5c by GC/MS.30
Intramolecular additions were done with 23.81 mg (0.1
mmol) of 9 (97%) and 28.22 mg (0.02 mmol) of vitamin B12a
with the same conditions as for the intermolecular additions
except ether was used as eluent.
Ack n ow led gm en t. This work was supported by
Grant No. CTS-9306961 from NSF. The authors thank
J ames M. Bobbitt for helpful discussions. Registr y n os.
p r ovid ed by th e a u th or : 5a , 113829-98-4; 5b, 57242-
85-0; 5c, 138695-42-8; 11, 34737-39-8; 9, 74056-05-6.
Products were identified by comparing their GC retention
times and mass spectra with those of authentic samples. Mass
J O9608477
(27) Rusling, J . F.; Connors, T. F. Anal. Chem. 1983, 55, 776-781.
(28) Taber, D. F. J . Org. Chem. 1976, 41, 2649-2650.
(29) Whitmore, F. C.; Pedlow, G. W., J r. J . Am. Chem. Soc. 1941,
63, 759-760.
(30) Surfactant, excess 4, and vitamin B12 are retained on the silica
columns.