Divergent reactivity in tandem reduction - Michael ring closures of five- and six-membered cyclic enones

Article abstract of DOI:10.1002/jhet.111

(Chemical Equation Presented) In this study, methyl (±)-1-(2- nitrobenzyl)-4-oxo-2-cyclohexene-1-carboxylate and methyl (±)-(2- nitrobenzyl)-4-oxo-2-cyclopentene-1-carboxylate were prepared and subjected to reductive cyclization under dissolving metal conditions. The two reactants showed divergent behavior with the six-ring substrate reacting at the ester carbonyl and the five-ring substrate closing on the enone double bond. The difference in reactivity is attributed to the conformational flexibility, relative reactivity, and steric environment of C4-substituted six- and five-membered cyclic enones.

Full text of DOI:10.1002/jhet.111

854
Divergent Reactivity in Tandem Reduction–Michael Ring
Closures of Five- and Six-Membered Cyclic Enones
Vol 46
Richard A. Bunce,* Baskar Nammalwar, and LeGrande M. Slaughter
Department of Chemistry, Oklahoma State University, Stillwater, Oklahoma 74078-3071
*E-mail: rab@okstate.edu
Received November 6, 2008
DOI 10.1002/jhet.111
Published online 26 August 2009 in Wiley InterScience (www.interscience.wiley.com).
In this study, methyl (ꢀ)-1-(2-nitrobenzyl)-4-oxo-2-cyclohexene-1-carboxylate and methyl (ꢀ)-(2-
nitrobenzyl)-4-oxo-2-cyclopentene-1-carboxylate were prepared and subjected to reductive cyclization
under dissolving metal conditions. The two reactants showed divergent behavior with the six-ring sub-
strate reacting at the ester carbonyl and the five-ring substrate closing on the enone double bond. The
difference in reactivity is attributed to the conformational flexibility, relative reactivity, and steric envi-
ronment of C4-substituted six- and five-membered cyclic enones.
J. Heterocyclic Chem., 46, 854 (2009).
INTRODUCTION
with easier purification of the product. Ketoester 4 was
prepared as previously described [8]. Alkylation of 3
and 4 with 2-nitrobenzyl bromide [9] using potassium
carbonate and catalytic 18-crown-6 in acetonitrile under
anhydrous conditions [10] gave products 5 and 6,
respectively. Reduction of the enone carbonyls in 5 and
6 with sodium borohydride in the presence of cerium(III)
chloride [11], followed by treatment with aqueous acid
resulted in 1,3-carbonyl transposition to give substrates
7 and 8.
The reductive cyclization of 2-nitrobenzyl ketones
under dissolving metal conditions is well established as
a route to the synthesis of indoles [1,2]. Earlier work
from our laboratory studied a tandem reduction–Michael
addition variant of this reaction as a route to the synthe-
sis of 1,2,3,4-tetrahydroquinoline-2-acetic esters [3], and
we have recently used this reaction to synthesize
1,2,3,9-tetrahydro-4H-carbazol-4-one [4]. In this investi-
gation, we sought to expand the scope of the tandem
reduction–Michael sequence to access functionalized lin-
ear-fused tricyclic systems. For this study, we prepared
six- and five-membered cyclic enones substituted at C4
by a methyl ester and a 2-nitrobenzyl group, and sub-
jected each to mild reduction using iron in acetic acid.
To our surprise, divergent reactivity was observed from
the cyclohexenone and cyclopentenone substrates, result-
ing in two relatively uncommon ring systems. In addi-
tion, a mechanistically novel competitive ester reduction
process was observed. Thus, we report our findings in
this area.
The results of our reduction-cyclization study are out-
lined in Scheme 2. In each case, the reaction was
Scheme 1 [a]
RESULTS AND DISCUSSION
The syntheses of our cyclization substrates are sum-
marized in Scheme 1. Ketoester 3 was prepared from
1,3-cyclohexanedione (1) by Lewis acid-catalyzed enol
ether formation to give 2 [5] followed by kinetic depro-
tonation [6] and reaction with methyl cyanoformate [7].
In this case, we found that methyl cyanoformate gave
better yields of the ketoester than methyl chloroformate
C
V
2009 HeteroCorporation

September 2009
Divergent Reactivity in Tandem Reduction–Michael Ring Closures
of Five- and Six-Membered Cyclic Enones
855
Scheme 3
Scheme 2
of the ring junction was confirmed by the conversion of
11 to its solid N-benzoyl derivative 12 and single crystal
X-ray analysis (Fig. 1).
Examination of molecular models provides some
insight into the observed difference in reactivity. Fol-
lowing reduction of the nitro function in 7, alignment of
the amino group for addition to the enone would result
in steric repulsion between the C5 methylene of the
cyclohexenone and the aromatic ring as in A (Scheme
3). Rotation about the benzylic bond to minimize this
interaction would then lead to conformation B, which is
more prone to react at the ester carbonyl. By compari-
son, similar steric interference is not present in cyclo-
pentenone 8. Furthermore, the five-membered cyclic
enone should be more reactive due to strain. Eclipsing
interactions that develop in the five-membered ring of
10 during addition should not significantly deter cycliza-
tion since the starting enone also possesses considerable
torsional strain. The eclipsing in the cyclized product is
clearly visible in the X-ray structure of 12 (Fig. 1).
The preference for the cis stereochemistry of the ring
junction in 10 is in accord with both strain and stereo-
electronic considerations. The cis-fused stereochemistry
would be expected based on strain arguments, with the
cis-fused ring junction preferred over the more strained
trans [12]. Stereoelectronically, it is well established
that the cis-fused isomer is strongly favored in nucleo-
philic ring closures on pre-existing rings via an axial
attack that permits a chair-like transition state [13].
Although, a true chair transition state is not possible due
to the aromatic sp2 carbons, pseudoaxial attack would
still be expected to afford a cis product.
complete in 30 min and led predominantly to a single
product. For cyclohexenone 7, the expected reduction-
Michael addition was not observed, but instead, reduc-
tion of the nitro group was followed by addition of the
aniline nitrogen to the ester to give the spiro-fused 3,4-
dihydro-2(1H)-quinolinone derivative 9 in 95% yield.
For cyclopentenone 8, the reduction–Michael sequence
proceeded as planned, but was accompanied by reduc-
tion of the ester to afford 10 in 76% yield. Extended
reaction times (4 h) led to further acylation of the pri-
mary alcohol in 10 to give 11. The cis stereochemistry
The reduction of the ester group in 8 is also an inter-
esting observation. The reduction is analogous to the
classical Bouveault-Blanc reaction [14], but would not
be expected to occur with iron as the electron source
[15]. In our substrates, the a,b-unsaturated ketone is the
functional group most susceptible to reduction under
dissolving metal conditions [16], and we believe that
this moiety is the key to reduce the ester.
To explore this process without interference from the
amino group, methyl (ꢀ)-1-benzyl-4-oxo-2-cyclohexene-
1-carboxylate (13) and methyl (ꢀ)-1-benzyl-4-oxo-2-
cyclopentene-1-carboxylate (14) were prepared using the
method described for the nitro-bearing substrates. Treat-
ment of 13 with iron in acetic acid for 24 h yielded a
Figure 1. Molecular structure of compound 12, with thermal ellipsoids
drawn at the 50% probability level. Hydrogen atoms on C10 and on
the aromatic rings have been removed for clarity.
Journal of Heterocyclic Chemistry
DOI 10.1002/jhet

856
R. A. Bunce, B. Nammalwar, and L. M. Slaughter
Vol 46
Scheme 5
33:67 mixture (by NMR) of starting material 13 and the
double bond reduction product 15. This ratio varied lit-
tle with longer reaction times or increased amounts of
iron. Similar reaction of 14 gave more interesting
results, and the reaction was considerably faster. Expo-
sure of 14 to iron in refluxing acetic acid gave nearly
complete conversion to alcohol 18 in 15 min. Prolonged
treatment (2 h) under the same conditions gave a 67:33
mixture (by NMR) of 18:20, as the acetates, in 95%
yield. These results are summarized in Scheme 4.
Mechanistically, the reduction of 13 and 14 is initi-
ated by protonation of the enone carbonyls followed by
the addition of two electrons to each conjugated system
[16] to give anions 21 and 22, respectively (Scheme 5).
In 21, the six-membered ring is conformationally flexi-
ble making the ester at C4 less accessible to attack by
the anionic center at C3. Thus, protonation and tautomeri-
zation occur to give 15. In the more rigid structure 22,
the C3 anion is closer to the C4 ester and cyclization
occurs to afford the strained cyclopropanone hemiketal
23. Under the acidic conditions of the reaction, 23
would undergo rapid proton and enol-assisted three-ring
opening [17], as in 24, followed by the loss of methanol
to give aldehyde 25. Further reduction of 25 would then
afford alcohol 18 and, eventually, 20.
The systems resulting from these ring closures have
minimal precedent in the literature. The 3,3-dialkyl-3,4-
dihydro-2(1H)-quinolinone scaffold of 9 is found in
some antidepressants [18], but spiro-fused compounds
have not been extensively investigated [19]. The
2,3,3a,4,9,9a-hexahydro-1H-cyclopenta[b]quinoline sys-
tem has been reported [20] and is known to exhibit
some antipsychotic activity [21], but structures with the
functional group arrangement of 10 are unknown.
CONCLUSIONS
Divergent behavior has been observed in the dissolv-
ing metal reduction–Michael reaction of two substrates
differing only in the size of the ring incorporating the
Michael acceptor. The disparate reaction pathways can
be attributed to the differences in strain and steric envi-
ronment of the enone acceptor as well as the alignment
of the reacting functionality in the two systems. The
reaction is clean and offers an efficient route to a rela-
tively rare ring skeleton from each substrate. The reduc-
tion of the ester functionality in the five-membered ring
substrate is novel and likely involves the participation of
the enone moiety.
Scheme 4
EXPERIMENTAL
Commercial reagents and solvents were used as received.
Tetrahydrofuran was dried over potassium hydroxide pellets
and distilled from lithium aluminum hydride before use. The
hydrochloric acid (3M), ammonium chloride (saturated), so-
dium bicarbonate (saturated), and sodium chloride (saturated)
used in workup procedures refer to aqueous solutions. All
reactions were run under dry nitrogen in oven-dried glassware.
Reactions were monitored by thin layer chromatography on
silica gel GF plates (Analtech 21521). Preparative separations
were performed using flash column chromatography [22] on
silica gel (grade 62, 60–200 mesh) mixed with ultraviolet-
Journal of Heterocyclic Chemistry
DOI 10.1002/jhet

September 2009
Divergent Reactivity in Tandem Reduction–Michael Ring Closures
of Five- and Six-Membered Cyclic Enones
857
active phosphor (Sorbent Technologies No. UV-5) or thin layer
chromatography on 20 cm ꢁ 20 cm silica gel GF plates (Anal-
tech 02015); band elution was monitored using a hand-held
ultraviolet lamp. Hexanes used in chromatography had a boil-
ing range of 65–70ꢂC. Melting points were uncorrected. Infra-
red spectra were run as thin films on sodium chloride disks
funnel, a reflux condenser, and a magnetic stir bar was charged
with 35 mL of dry acetonitrile, 4.87 g (35.2 mmoles) of anhy-
drous potassium carbonate, and 10 mg of 18-crown-6. Stirring
was initiated and a solution of 2.16 g (11.7 mmoles) of 3 in
10 mL of acetonitrile was added dropwise at 22ꢂC. The result-
ing blue solution was stirred for 10 min and a solution of 2.80
g (13.0 mmoles) of 2-nitrobenzyl bromide [9] in 10 mL of
acetonitrile was added dropwise. The reaction was refluxed for
18 h at which time thin layer chromatography indicated com-
plete consumption of 3. The crude reaction mixture was
cooled, diluted with ether, vacuum filtered, and concentrated
under vacuum. The resulting dark yellow oil was purified by
flash chromatography on a 100 cm ꢁ 2.5 cm silica gel column
eluted with 20–30% ether in hexanes to give 3.30 g (88%) of
5 as a light yellow oil. IR: 1729, 1660, 1609, 1526, 1384,
1
and were referenced to polystyrene. H- and 13C-NMR spectra
were measured in deuteriochloroform at 300 MHz and 75
MHz, respectively, using tetramethylsilane as the internal
standard; coupling constants (J) are reported in Hertz. Low-re-
solution mass spectra (electron impact/direct probe) were run
at 30 eV.
3-Methoxy-2-cyclohexen-1-one (2). The procedure of Sabi-
tha et al. [5] was modified. A mixture of 20.0 g of silica gel
(Alfa-Aesar, 220–440 mesh) and 3.60 g (9.60 mmoles) of cer-
ium(III) chloride heptahydrate in 60 mL of dry acetonitrile
was stirred for 12 h at 22ꢂC. The acetonitrile was removed
under vacuum and a solution of 5.00 g (44.6 mmoles) of 1 in
20 mL of methanol was added. The mixture was stirred for 72
h at 22ꢂC and filtered with ethyl acetate. The filtrate was con-
centrated under reduced pressure, and the crude product was
purified by flash chromatography on a 30 cm ꢁ 2.5 cm silica
gel column eluted with 50% ethyl acetate in hexanes to give
3.70 g (90%) of 2 as a colorless oil. IR: 1671, 1645, 1605
1
1350 cmꢃ1; H-NMR: d 7.82 (dd, 1H, J ¼ 8.2, 6.6), 7.47 (td,
1H, J ¼ 7.5, 1.5), 7.36 (td, 1H, J ¼ 7.9, 1.6), 7.35 (d, 1H, J ¼
7.0), 5.41 (d, 1H, J ¼ 1.5), 3.87 (d, 1H, J ¼ 14.1), 3.68 (s,
3H), 3.67 (s, 3H), 3.57 (d, 1H, J ¼ 14.1), 2.59 (m, 1H), 2.39
(m, 1H), 2.27 (m 1H), 1.77 (m, 1H); 13C-NMR: d 193.9,
177.7, 171.1, 150.9, 133.4, 133.3, 131.5, 127.9, 124.7, 101.9,
57.0, 55.8, 52.7, 34.7, 28.3, 26.3; ms: m/z 319 (Mþ). Anal.
Calcd. for C16H17NO6: C, 60.19; H, 5.33; N, 4.39. Found: C,
60.29; H, 5.36; N, 4.35.
1
cmꢃ1; H-NMR: d 5.37 (s, 1H), 3.70 (s, 3H), 2.42 (t, 2H, J ¼
Methyl (ꢀ)-4-methoxy-1-(2-nitrobenzyl)-2-oxo-3-cyclopen-
tene-1-carboxylate (6). This compound (3.00 g, 84%) was
obtained as a light yellow oil. IR: 1740, 1699, 1596, 1526,
6.4), 2.35 (t, 2H, J ¼ 6.9), 1.98 (quintet, 2H, J ¼ 6.6); 13C-
NMR: d 199.4, 178.4, 102.1, 55.4, 36.5, 28.6, 21.0.
1
Methyl (ꢀ)-4-methoxy-2-oxo-3-cyclohexene-1-carboxylate
(3). To a stirred solution of 2.88 g (4.00 mL, 28.6 mmoles) of
diisopropylamine in 30.0 mL of tetrahydrofuran at ꢃ78ꢂC,
was slowly added 17.0 mL of 1.75M n-butyllithium in hexanes
(30.0 mmoles). After 30 min, a solution of 3.00 g (23.8
mmoles) of 2 in 20.0 mL of tetrahydrofuran was added drop-
wise and stirring was continued at ꢃ78ꢂC for 30 min. A solu-
tion of 3.40 g (40.0 mmoles) of methyl cyanoformate in 10
mL of tetrahydrofuran was then added dropwise and the reac-
tion was stirred for 1 h at ꢃ78ꢂC. The reaction mixture was
slowly warmed to 22ꢂC, stirred for 30 min, cautiously added
to saturated ammonium chloride and extracted with ether. The
ether extracts were washed with saturated sodium bicarbonate
(one time), water (one time), and saturated sodium chloride
(one time), and then dried (magnesium sulfate) and concen-
trated under vacuum. The crude product was purified by flash
chromatography on a 100 cm ꢁ 2.5 cm silica gel column
eluted with 30% ethyl acetate in hexanes to give 0.60 g (20%)
of 2 and 3.20 g (73%) of 3. The yield of 3 was 91% based on
1359 cmꢃ1; H-NMR: d 7.86 (dd, 1H, J ¼ 8.1, 1.5), 7.47 (td,
1H, J ¼ 7.5, 1.5), 7.38 (td, 1H, J ¼ 7.9, 1.6), 7.35 (dd, 1H, J
¼ 7.7, 1.5), 5.26 (t, 1H, J ¼ 1.1), 3.80 (s, 3H), 3.75 (d, 1H, J
¼ 14.6), 3.75 (s, 3H), 3.52 (d, 1H, J ¼ 14.6), 3.17 (dd, 1H, J
¼ 17.9, 1.1), 2.48 (dd, 1H, J ¼ 17.9, 1.1); 13C-NMR: d 200.3,
191.0, 170.8, 150.5, 132.7, 132.2, 131.4, 128.0, 124.7, 102.0,
59.8, 59.1, 53.1, 37.0, 33.7; ms: m/z 305 (Mþ). Anal. Calcd.
for C15H15NO6: C, 59.02; H, 4.92; N, 4.59. Found: C, 59.13;
H, 4.96; N, 4.53.
Representative procedure for 1,3-carbonyl tranposition:
Methyl (ꢀ)-1-(2-nitrobenzyl)-4-oxo-2-cyclohexene-1-carbox-
ylate (7). The procedure of Luche was modified [11]. A 250-
mL three-necked, round-bottomed flask equipped with a reflux
condenser and a magnetic stir bar was charged with 20 mL of
methanol followed by 4.60 g (12.4 mmoles) of cerium(III)
chloride heptahydrate. The mixture was stirred for 10 min and
a solution of 2.50 g (7.84 mmoles) of 5 in 10 mL of methanol
was added dropwise. After 5 min, 1.25 g (32.9 mmoles) of so-
dium borohydride was added in small portions over a period
of 20 min. (Note: Frothing is a problem if the added portions
of sodium borohydride are too large). The reaction mixture
was stirred for 15 min at which time 12 mL of 3M hydrochlo-
ric acid was added. After 20 min, the mixture was concen-
trated under vacuum to one-third its volume and extracted
with ether (three times). The combined ether extracts were
washed with water (three times) and saturated sodium chloride
(one time), then dried (magnesium sulfate) and concentrated
under vacuum. The resulting dark brown liquid was purified
by flash chromatography on a 50 cm ꢁ 2.5 cm silica gel col-
umn eluted with 20–30% ether in hexanes to give 1.78 g
(79%) of 7 as thick yellow oil. IR: 1732, 1685, 1609, 1528,
recovered starting material. IR: 1740, 1656, 1606 cmꢃ1 1H-
;
NMR: d 5.41 (s, 1H), 3.76 (s, 3H), 3.72 (s, 3H), 3.35 (dd, 1H,
J ¼ 9.2, 5.3), 2.57 (m, 1H), 2.46 (m, 1H), 2.36 (m, 1H), 2.17
(m, 1H); 13C-NMR: d 193.5, 178.4, 170.6, 101.6, 55.8, 52.2,
52.0, 27.0, 24.0; ms: m/z 184 (Mþ); Anal. Calcd. for C9H12O4:
C, 58.70; H, 6.52. Found: C, 58.77; H, 6.55.
Methyl (ꢀ)-4-methoxy-2-oxo-3-cyclopentene-1-carboxylate
(4). This compound was prepared by the method of Fuchs
and McGarrity [8]. The spectral data matched those
reported.
Representative procedure for alkylation of 3 with 2-
nitrobenzyl bromide: Methyl (ꢀ)-4-methoxy-1-(2-nitroben-
zyl)-2-oxo-3-cyclohexene-1-carboxylate (5). The general pro-
cedure of Makosza and Tyrala [10] was used. A 100-mL
three-necked, round-bottomed flask equipped with an addition
1
1351 cmꢃ1; H-NMR: d 7.92 (dd, 1H, J ¼ 8.1, 1.5), 7.55 (td,
1H, J ¼ 7.5, 1.3), 7.43 (td, 1H, J ¼ 8.1, 1.5), 7.27 (dd, 1H, J
¼ 7.7, 1.5), 6.82 (d, 1H, J ¼ 10.4), 6.02 (d, 1H, J ¼ 10.4),
Journal of Heterocyclic Chemistry
DOI 10.1002/jhet

858
R. A. Bunce, B. Nammalwar, and L. M. Slaughter
Vol 46
3.70 (s, 3H), 3.66 (d, 1H, J ¼ 13.8), 3.50 (d, 1H, J ¼ 13.8),
2.45 (m, 2H), 2.36 (m, 1H), 2.06 (m, 1H); 13C-NMR: d 197.8,
172.8, 150.2, 149.0, 133.0, 132.7, 130.3, 129.7, 128.5, 125.2,
52.8, 48.6, 39.5, 34.4, 30.9; ms: m/z 289 (Mþ). Anal. Calcd.
for C15H15NO5: C, 62.28; H, 5.19; N, 4.84. Found: C, 62.40;
H, 5.24; N, 4.76.
of 8 and 812 mg (14.6 mmoles) of iron powder in 30 mL of
acetic acid. After 4 h at 115ꢂC, workup and flash chromatogra-
phy on a 25 cm ꢁ 2 cm silica gel column using 15% ether in
hexanes gave 400 mg (85%) of 10 as a tan oil. IR: 3394, 1740
cmꢃ1
;
1H-NMR: d 7.05 (m, 2H), 6.66 (td, 1H, J ¼ 7.5, 1.3),
6.50 (dd, 1H, J ¼ 7.8, 0.8), 4.10 (br s, 1H), 4.08 (d, 1H, J ¼
11.4), 4.02 (d, 1H, J ¼ 11.4), 3.92 (t, 1H, J ¼ 6.4), 2.80 (d,
1H, J ¼ 16.5), 2.70 (d, 1H, J ¼ 16.5), 2.70 (m, 1H), 2.27 (m,
2H), 2.21 (dd, 1H, J ¼ 5.4, 1.1), 2.06 (s, 3H); 13C-NMR: d
214.9, 170.8, 141.4, 129.7, 127.6, 117.7, 116.8, 113.5, 67.6,
52.4, 46.3, 46.0, 39.6, 31.5, 20.8; ms: m/z 259 (Mþ). Anal.
Calcd. for C15H17NO3: C, 69.50; H, 6.56; N, 5.41. Found: C,
69.52; H, 6.55; N, 5.39.
Methyl (ꢀ)-1-(2-nitrobenzyl)-4-oxo-2-cyclopentene-1-car-
boxylate (8). This compound (1.65 g, 73%) was obtained as
yellow crystals, mp 103–105ꢂC. IR: 1712, 1679, 1608, 1526,
1
1352 cmꢃ1; H-NMR: d 7.86 (dd, 1H, J ¼ 8.2, 1.3), 7.52 (td,
1H, J ¼ 7.5, 1.3), 7.47 (d, 1H, J ¼ 5.7), 7.39 (td, 1H, J ¼ 8.1,
1.5), 7.29 (dd, 1H, J ¼ 7.7, 1.5), 6.08 (d, 1H, J ¼ 5.7), 3.65
(s, 3H), 3.50 (d, 1H, J ¼ 13.7), 3.17 (d, 1H, J ¼ 13.7), 2.36
(d, 1H, J ¼ 18.7), 2.18 (d, 1H, J ¼ 18.7); 13C-NMR: d 208.7,
167.3 (2C), 150.1, 134.9, 133.4, 132.6, 131.2, 128.1, 125.0,
67.7, 52.3, 42.6, 36.5; ms: m/z 275 (Mþ). Anal. Calcd. for
C14H13NO5: C, 61.09; H, 4.73; N, 5.09. Found: C, 61.13; H,
4.74; N, 5.09.
(ꢀ)-(3aR*, 9aR*)-9a-Acetoxymethyl-4-benzoyl-1,3,3a,4,9,
9a-hexahydro-2H-cyclopenta[b]quinolin-2-one (12). To
a
stirred solution of 200 mg (0.77 mmoles) of 11 and 85.6 mg
(0.85 mmoles) of triethylamine in 20 mL of dichloromethane,
a solution of 120 mg (0.85 mmoles) of benzoyl chloride in 1
mL of dichloromethane was slowly added over a period of 5
min. The reaction mixture was stirred at 22ꢂC for 2 h at which
time thin layer chromatography confirmed the absence of start-
ing material. The reaction mixture was poured into cold water
and the dichloromethane layer was separated. The organic
phase was washed with cold water (two times), dried (magne-
sium sulfate) and concentrated under vacuum. The resulting
residue was passed through a small plug of silica gel with
30% ether in hexanes to give 260 mg (93%) of 12 as a light
(ꢀ)-10,40-Dihydrospiro[2-cyclohexene-1,30(20H)-quinoline]-
20,4-dione (9). The procedure of Bunce et al. was used [1]. A
mixture of 500 mg (1.73 mmoles) of 7, 25 mL of acetic acid,
and 773 mg (13.8 mmoles, 8.0 equiv) of iron powder (>100
mesh) was heated with stirring at 115ꢂC (oil bath) until thin
layer chromatography indicated complete consumption of start-
ing material (ca. 30 min). The reaction mixture was cooled,
diluted with 50 mL of water, and extracted with ether (three
times). The combined ether layers were washed with water (one
time), saturated sodium bicarbonate (three times), saturated so-
dium chloride (one time), then dried (magnesium sulfate) and
concentrated under vacuum to give 373 mg (95%) ofꢃ91 as a
1
yellow solid, mp 108–110ꢂC. IR: 1744, 1643 cmꢃ1; H-NMR:
d 7.38–7.21 (complex, 6H), 7.07 (td, 1H, J ¼ 7.5, 1.1), 6.93
(t, 1H, J ¼ 7.5), 6.49 (d, 1H, J ¼ 7.9), 5.19 (dd, 1H, J ¼ 9.0,
4.4), 4.28 (d, 1H, J ¼ 10.9), 4.20 (d, 1H, J ¼ 10.9), 3.04 (ddd,
1H, J ¼ 19.4, 9.0, 1.8), 2.94 (d, 1H, J ¼ 14.1), 2.73 (d, 1H, J
¼ 14.1), 2.29 (ddd, 1H, J ¼ 19.4, 4.4, 1.8), 2.20 (dd, 1H, J ¼
18.5, 1.8), 2.10 (s, 3H), 2.06 (dd, 1H, J ¼ 18.5, 1.8); 13C-
NMR: d 214.1, 170.7, 169.8, 138.1, 134.8, 130.8, 130.6,
129.1, 129.0, 128.1, 127.4, 126.9, 126.1, 71.1, 57.4, 47.9, 45.6,
45.2, 34.6, 20.7; ms: m/z 363 (Mþ). Anal. Calcd. for
C22H21NO4: C, 72.73; H, 5.79; N, 3.86. Found: C, 73.71; H,
5.79; N, 3.88.
X-ray crystallographic analysis of 12. Flat, elongated rods
of 12 were obtained by slow diffusion of pentane into an ether
solution of the compound. A sample measuring 0.4 mm ꢁ 0.4
mm ꢁ 0.1 mm, which was cut from a longer rod, was
immersed in polyisobutylene and placed in a nylon loop under
a nitrogen cold stream. X-ray intensity data were measured at
115(2) K on a Bruker SMART Apex II diffractometer. Graph-
pale white solid, mp 212–215ꢂC. IR: 3195, 1667 cm
;
1H-
NMR: d 8.69 (br s, 1H), 7.21 (complex, 2H), 7.05 (td, 1H, J ¼
7.5, 1.3), 6.82 (obscured, 1H) 6.81 (d, 1H, J ¼ 10.3), 6.14 (d,
1H, J ¼ 10.3), 3.13 (d, 1H, J ¼ 15.9), 2.99 (d, 1H, J ¼ 15.9),
2.73 (ddd, 1H, J ¼ 17.1, 8.4, 4.9), 2.49 (ddd, 1H, J ¼ 17.1, 8.4,
4.9), 2.34 (ddd, 1H, J ¼ 13.4, 8.4, 4.9), 1.98 (ddd, 1H, J ¼
13.4, 8.4, 4.9); 13C-NMR: d 198.2, 172.1, 148.9, 136.1, 130.7,
128.6, 128.2, 123.7, 121.1, 115.2, 42.6, 36.6, 33.6, 29.4; ms:
m/z 227 (Mþ). Anal. Calcd. for C14H13NO2: C, 74.01; H, 5.73;
N, 6.17. Found: C, 74.00; H, 5.71; N, 6.20.
(ꢀ)-(3aR*, 9aR*)-9a-Hydroxymethyl-1,3,3a,4,9,9a-hexahy-
dro-2H-cyclopenta[b]quinolin-2-one (10). The procedure
used to prepare 9 was followed using 200 mg (0.73 mmoles)
of 8 and 325 mg (5.84 mmoles) of iron powder in 12 mL of
acetic acid. After 30 min at 115ꢂC, workup and preparative
thin layer chromatography using 40% ether in hexanes gave
120 mg (76%) of 10 as a light yellow oil. IR: 3395, 1733
˚
ite-monochromated Mo-Ka radiation (k ¼ 0.71073 A, fine-
1
cmꢃ1; H-NMR: d 7.01 (td, 1H, J ¼ 7.3, 1.2), 6.98 (dd, 1H, J
focus sealed tube) was used with the CCD detector placed 6.0
cm from the sample. Data frames were collected in a series of
/ and x scans with 0.5ꢂ scan widths and 30-s exposure times.
Data integration used the Bruker SAINT software package
[23]. Data were corrected for absorption effects using the mul-
tiscan technique (SADABS) [24]. The structure was solved by
direct methods and refined by full-matrix least-squares on F2
using the Bruker SHELXTL software suite [25]. Non-hydrogen
atoms were assigned anisotropic temperature factors. Hydrogen
atoms were placed in calculated positions based on the geome-
try at carbon (riding model). Refined formula: C22H21NO4, M
¼ 7.5, 0.8), 6.64 (td, 1H, J ¼ 7.3, 1.2), 6.48 (dd, 1H, J ¼ 7.9,
0.8), 3.92 (br s, 1H), 3.94 (t, 1H, J ¼ 6.4), 3.59 (d, 1H, J ¼
10.9), 3.55 (d, 1H, J ¼ 10.9), 2.74 (d, 1H, J ¼ 16.7), 2.69
(obscured, 2H), 2.65 (d, 1H, J ¼ 16.7), 2.37 (d, 1H, J ¼ 18.7),
2.17 (dd, 1H, J ¼ 18.7, 1.6), 2.20 (m, 1H); 13C-NMR: d
216.6, 141.7, 129.6, 127.4, 117.5, 117.4, 113.4, 66.6, 52.0,
46.4, 45.9, 41.3, 31.3; ms: m/z 217 (Mþ). Anal. Calcd. for
C13H15NO2: C, 71.89; H, 6.91; N, 6.45. Found: C, 71.96; H,
6.95; N, 6.40. This reaction also gave 26 mg (14%) of com-
pound 11.
˚
(ꢀ)-(3aR*, 9aR*)-9a-Acetoxymethyl-1,3,3a,4,9,9a-hexahy-
dro-2H-cyclopenta[b]quinolin-2-one (11). The procedure
used to prepare 9 was followed using 500 mg (1.82 mmoles)
¼ 363.40, monoclinic, space group P21/n, a ¼ 11.1983(2) A,
ꢂ
˚
˚
b ¼ 8.18310(10) A, c ¼ 19.7453(3) A, b ¼ 101.7010(10) ,ꢃU1
3
˚
¼ 1771.80(5) A , Z ¼ 4, Dc ¼ 1.362 g/cm, l ¼ 0.094 mm
,
Journal of Heterocyclic Chemistry
DOI 10.1002/jhet

September 2009
Divergent Reactivity in Tandem Reduction–Michael Ring Closures
of Five- and Six-Membered Cyclic Enones
859
T ¼ 115(2) K, 2ymax ¼ 50.6ꢂ, completeness to 2ymax
100.0%, 13372 total reflections, 3227 independent (Rint
¼
¼
mixture of 13:15. The spectral data for 15 were: IR: 1727,
1702 cmꢃ1 1H-NMR: d 7.36–7.27 (complex, 3H), 7.12 (m,
;
0.0248), 2619 observed [I > 2r(I)]. Final R1 [I > 2r(I)] ¼
0.0336, wR2 (all data) ¼ 0.0833, largest difference peak and
2H), 3.77 (s, 3H), 2.96 (s, 2H), 2.56–2.33 (complex, 6H), 1.78
(m, 2H); 13C-NMR: d 210.4, 174.8, 129.4, 128.1, 127.9, 126.6,
51.6, 47.5, 45.6, 38.1, 33.1. The use of more iron and longer
reaction times failed to significantly alter this product ratio.
Reduction of 14 with iron and acetic acid: (ꢀ)-4-Benzyl-
4-hydroxymethyl-2-cyclopenten-1-one (18). The procedure
used to prepare 9 was followed using 500 mg (2.17 mmoles)
of 14 and 969 mg (17.3 mmoles) of iron powder in 35 mL of
acetic acid. After 15 min, workup and preparative thin layer
chromatography gave 320 mg (72%) of 18 as a colorless oil.
hole 0.225 and ꢃ0.194 eAꢃ3. CCDC 692896 contains the sup-
˚
plementary crystallographic data for compound 12. These data
can be obtained free of charge from the Cambridge Crystallo-
graphic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
Methyl (ꢀ)-4-benzyl-2-oxo-3-cyclohexene-1-carboxylate
(13). The procedure used to prepare 5 was followed using
1.08 g (5.85 mmoles) of 3 and 1.11 g (6.5 mmoles) of benzyl
bromide. Following flash chromatography, 1.52 g (95%) of
methyl
boxylate was isolated as a white solid, mp 68–70ꢂC. IR: 1729,
(ꢀ)-1-benzyl-4-methoxy-2-oxo-3-cyclohexene-1-car-
IR: 3416, 1711, 1677 cmꢃ1
;
1H-NMR: d 7.53 (d, 1H, J ¼
5.7), 7.30–7.21 (complex, 3H), 7.10 (d, 2H, J ¼ 6.8), 6.11 (d,
1H, J ¼ 5.7), 3.60 (d, 1H, J ¼ 10.7), 3.58 (d, 1H, J ¼ 10.7),
2.96 (d, 1H, J ¼ 13.5), 2.81 (d, 1H, J ¼ 13.5), 2.62 (br s, 1H),
2.29 (d, 1H, J ¼ 18.6), 2.26 (d, 1H, J ¼ 18.6); 13C-NMR: d
209.5, 168.7, 136.3, 134.4, 130.1, 128.2, 126.7, 67.0, 51.8,
42.9, 41.3; ms: m/z 111 (Mþ-C7H7). Anal. Calcd. for
C13H14O2: C, 77.23; H, 6.93. Found: C, 77.29; H, 6.97. This
reaction also afforded 10% of recovered 14.
1
1660, 1610 cmꢃ1; H-NMR: d 7.28–7.17 (complex, 3H), 7.14
(m, 2H), 5.41 (d, 1H, J ¼ 1.2), 3.71 (s, 3H), 3.66 (s, 3H), 3.32
(d, 1H, J ¼ 13.7), 3.23 (d, 1H, J ¼ 13.7), 2.63 (m, 1H), 2.29
(m, 2H), 1.80 (m, 1H); 13C-NMR: d 194.4, 177.8, 171.5,
136.5, 130.4, 128.1, 126.7, 101.9, 56.9, 55.7, 52.4, 40.0, 28.1,
26.3; ms (30 eV): m/z 274 (Mþ); Anal. Calcd. for C16H18O4:
C, 70.07; H, 6.57. Found: C, 70.12; H, 6.59.
Reduction and carbonyl transposition were carried out as
described for the preparation of 7 using 1.52 g (5.56 mmoles)
of the benzylated product from the earlier procedure. Follow-
ing flash chromatography, 1.22 g (90%) of 13 was isolated as
Upon prolonged heating for 2 h, the reaction gave 450 mg
of an inseparable 67:33 mixture of 18:20 as the acetates. The
1
spectral data for 18 (acetate): IR: 1743, 1715 cmꢃ1; H-NMR:
d 7.46 (d, 1H, J ¼ 5.7), 7.33–7.17 (complex, 3H), 7.09 (m,
2H), 6.13 (d, 1H, J ¼ 5.7), 4.17 (d, 1H, J ¼ 10.9), 4.00 (d,
1H, J ¼ 10.9), 2.96 (d, 1H, J ¼ 13.7), 2.84 (d, 1H, J ¼ 13.6),
2.31 (s, 2H), 2.06 (s, 3H); 13C-NMR: d 207.6, 170.6, 166.7,
135.5, 134.6, 130.0, 128.4, 127.0, 67.7, 49.4, 43.0, 42.0, 20.7.
a colorless oil. IR: 1732, 1682 cmꢃ1 1H-NMR: d 7.32–7.22
;
(complex, 3H), 7.09 (dd, 1H, J ¼ 7.8, 1.8), 6.95 (d, 1H, J ¼
10.3), 6.02 (d, 1H, J ¼ 10.3), 3.69 (s, 3H), 3.08 (s, 2H), 2.47
(m, 2H), 2.40 (m, 1H), 2.07 (m, 1H); 13C-NMR: d 198.3,
173.2, 150.3, 135.4, 129.8, 129.2, 128.4, 127.2, 52.3, 49.0,
44.5, 34.5, 30.5; ms: m/z 244 (Mþ); Anal. Calcd. for
C15H16O3: C, 73.77; H, 6.56. Found: C, 73.81; H, 6.60.
The spectral data for 20 (acetate): 1743 cmꢃ1 1H-NMR: d
;
7.34–7.18 (complex, 3H), 7.10 (m, 2H), 3.98 (d, 1H, J ¼
11.1), 3.89 (d, 1H, J ¼ 11.1), 2.79 (s, 2H), 2.39–2.21 (com-
plex, 2H), 2.33 (s, 2H), 2.10 (s, 3H), 1.93 (m, 2H); 13C-NMR:
d 217.5, 170.7, 136.6, 130.0, 128.3, 126.7, 68.4, 47.0, 43.6,
42.3, 36.3, 30.2, 20.8.
Methyl (ꢀ)-4-benzyl-2-oxo-3-cyclopentene-1-carboxylate
(14). The procedure used to prepare 5 was followed using 0.99
g (5.85 mmoles) of 4 and 1.11 g (6.50 mmoles) of benzyl bro-
mide. Following flash chromatography, 1.46 g (96%) of methyl
(ꢀ)-1-benzyl-4-methoxy-2-oxo-3-cyclopentene-1-carboxylate
was isolated as white solid, mp 96–98ꢂC. IR: 1741, 1699,
Acknowledgment. B. N. thanks the Department of Chemistry
at Oklahoma State University for a teaching assistantship.
L.M.S. thanks the NSF for a Career Award (NSF-0645438).
Funding for the 300 MHz NMR spectrometer of the Okla-
homa Statewide Shared NMR Facility was provided by NSF
(BIR-9512269), the Oklahoma State Regents for Higher Edu-
cation, the W. M. Keck Foundation, and Conoco, Inc.
Finally, the authors thank the OSU College of Arts and Sci-
ences for funds to upgrade our departmental FTIR and GC-
MS instruments.
1597 cmꢃ1 1H-NMR: d 7.27–7.17 (complex, 3H), 7.13 (m,
;
2H), 5.18 (s, 1H), 3.76 (2s, 6H), 3.30 (d, 1H, J ¼ 14.1),
3.25 (d, 1H, J ¼ 14.1), 3.11 (dd, 1H, J ¼ 17.7, 1.0), 2.61
(dd, 1H, J ¼ 17.7, 1.0); 13C-NMR: d 200.6, 190.3, 171.0,
136.2, 130.0, 128.2, 126.9, 102.3, 60.1, 59.0, 52.9, 39.1,
36.5; ms: m/z 260 (Mþ). Anal. Calcd for C15H16O4: C,
69.23; H, 6.15. Found: C, 69.29; H, 6.17.
Reduction and carbonyl transposition were carried out as
described for the preparation of 7 using 1.46 g (5.62 mmoles)
of the benzylated product from the earlier procedure. Follow-
ing flash chromatography, 1.16 g (90%) of 14 was isolated as
REFERENCES AND NOTES
1
a colorless oil. IR: 1710, 1677 cmꢃ1; H-NMR: d 7.54 (d, 1H,
[1] Bunce, R. A.; Randall, M. H.; Applegate, K. G. Org Prep
Proced Int 2002, 34, 493.
J ¼ 5.7), 7.32–7.20 (complex, 3H), 7.13 (m, 2H), 6.14 (d, 1H,
J ¼ 5.7), 3.62 (s, 3H), 2.97 (d, 1H, J ¼ 13.3), 2.84 (d, 1H, J
¼ 13.3), 2.29 (s, 2H); 13C-NMR: d 209.0, 168.2 (2C), 136.4,
134.7, 130.2, 128.3, 126.8, 67.3, 51.7, 43.0, 41.5; ms: m/z 230
(Mþ). Anal. Calcd. for C14H14O3: C, 73.04; H, 6.09. Found:
C, 73.10; H, 6.12.
Reduction of 13 with iron and acetic acid: Methyl 1-ben-
zyl-4-oxocyclohexane-1-carboxylate (15). The procedure used
to prepare 9 was followed using 500 mg (2.05 mmoles) of 13
and 915 mg (16.4 mmoles) of iron powder in 35 mL of acetic
acid. After 24 h, workup gave 450 mg of an inseparable 33:67
[2] Augustine, R. L.; Gustavsen, A. J.; Wanat, S. F.; Pattison,
I. C.; Houghton, K. S.; Koletar, G. J Org Chem 1973, 38, 3004.
[3] Bunce, R. A.; Herron, D. M.; Ackerman, M. L. J Org
Chem 2000, 65, 2847.
[4] Bunce, R. A.; Nammalwar, B. J Heterocyclic Chem 2009,
46, 172.
[5] Sabitha, G.; Reddy, M.; Sudhakar, K.; Yadav, J. S. Lett
Org Chem 2005, 2, 763.
[6] Lee, R. A.; McAndrews, C.; Patel, K. M.; Reusch, W. Tet-
rahedron Lett 1973, 965.
Journal of Heterocyclic Chemistry
DOI 10.1002/jhet

860
R. A. Bunce, B. Nammalwar, and L. M. Slaughter
Vol 46
[7] (a) Crabtree, S. R.; Mander, L. N.; Sethi, S. P. Organic
Reactions in Nonaqueous Systems; Marcel Dekker: New York, 1970;
pp 179–189.
Syntheses; Wiley: New York, 1998; Coll. Vol. 9, p 619; (b) Mander,
L. N.; Sethi, S. P. Tetrahedron Lett 1983, 24, 5425.
[17] (a) Wiberg, K. B.; Kass, S. R. J Am Chem Soc 1985, 107,
988; (b) Dewar, M. J. S.; Healy, E. F.; Ruiz, J. M. J Chem Soc Chem
Commun 1987, 943; (c) Koch, W.; Liu, B.; von Rague Schleyer, P. J
Am Chem Soc 1989, 111, 3479.
[8] Fuchs, R.; McGarrity, J. F. Synthesis 1992, 373.
[9] Collins, D. J.; Drygala, P. F.; Swan, J. M. Aust J Chem
1983, 36, 209.
[10] Makosza, M.; Tyrala, A. Synth Commun 1986, 16, 419.
[11] Luche, J.-L. J Am Chem Soc 1978, 100, 2226.
[12] Eliel, E.; Wilen, S. H.; Mander, L. N. Stereochemistry of
Organic Compounds; Wiley-Interscience: New York, 1994; pp 774–
776.
[18] (a) Beadle, C. D.; Boot, J.; Camp, N. P.; Dezutter, N.; Find-
lay, J.; Hayhurst, L.; Masters, J. J.; Penariol, R.; Walter, M. W. Bioorg
Med Chem Lett 2005, 15, 4432; (b) Cho, S. S. Y.; Gregory, T. F.;
Guzzo, P. R.; Howard, H. R., Jr.; Nikam, S. S.; Surman, M. D.;
Walters, M. A. Int. Pat. WO 2005066165 (2005); Chem Abstr 2005,
143, 153404; (c) Andreana, T. L.; Cho, S. S. Y.; Graham, J. M.; Greg-
ory, T. F.; Howard, H. R., Jr.; Kornberg, B. E.; Nikam, S. S.; Pflum,
D. A. Int. Pat. WO 2004026864 (2004); Chem Abstr 2004, 140,
287412.
[13] Oare, D. A.; Heathcock, C. A. In Topics in Stereochemis-
try; Eliel, E. L., Wilen, S. H., Eds.; Wiley-Interscience: New York,
1989; Vol. 19, p 227. See, in particular, pp 372–374.
[14] Hudlicky, M. Reductions in Organic Chemistry, 2nd ed.;
American Chemical Society: Washington, DC, 1996; ACS Monograph
188, p 213.
[19] Conley, R. T.; Frainier, L. J. J Org Chem 1962, 27, 3844.
[20] Bunce, R. A.; Herron, D. M.; Lewis, J. R.; Kotturi, S. V.;
Holt, E. M. J Heterocycl Chem 2003, 40, 101.
[15] (a) Iron in acetic acid spiked with nickel chloride hexahy-
drate has been used to reduce aldehydes, but not esters, see Clarke,
H. T.; Dreger, E. E. Organic Syntheses; Wiley: New York, 1932;
Coll. Vol. 1, pp 304–306; (b) A control experiment run by treating
methyl 1-methyl-1-cyclopentanecarboxylate (prepared according to
Fadel, A.; Salaun, J. Tetrahedron 1985, 41, 1267) with iron in reflux-
ing acetic acid for 60 min resulted in >96% recovery of the starting
material.
[21] Ramamoorthy, P. S.; McDevitt, R. E. Int. Pat. WO
2003091251 (2003); Chem Abstr 2003, 139, 364966.
[22] Still, W. C.; Kahn, M.; Mitra, A. J Org Chem 1978, 43, 2923.
[23] Bruker AXS. SAINT, Version 7.23A; Bruker AXS: Madi-
son, WI, 2006.
[24] Sheldrick, G. M. SADABS, Version 2.10; Bruker AXS Inc.:
Madison, WI, 2000.
[16] (a) House, H. O.; Huber, L. E.; Umen, M. J. J Am Chem
Soc 1972, 94, 8471; (b) Mann, C. K.; Barnes, K. K. Electrochemical
[25] Sheldrick, G. M. SHELXTL, Version 6.14; Bruker AXS
Inc.: Madison, WI, 2000.
Journal of Heterocyclic Chemistry
DOI 10.1002/jhet