Diastereoselectivity in an electrocyclization reaction of cyclopentadienones

Article abstract of DOI:10.1016/j.tetlet.2007.06.011

Two cyclopentadienones were generated and both underwent conrotatory electrocyclization as expected based on Woodward-Hoffmann rules. This result lends support to the idea that these ring-closing reactions are, in fact, pericyclic processes.

Full text of DOI:10.1016/j.tetlet.2007.06.011

Tetrahedron Letters 48 (2007) 5919–5922
Diastereoselectivity in an electrocyclization reaction
of cyclopentadienones
a,
a
b
*
Michael Harmata, Pinguan Zheng, Peter R. Schreiner and
c
Armando Navarro-V a´ zquez
a
Department of Chemistry, University of Missouri-Columbia, Columbia, Missouri 65211, USA
Institute of Organic Chemistry, Justus-Liebig University, Heinrich-Buff-Ring, 58, 35392 Giessen, Germany
b
c
Departamento de Qu ı´ mica Org a´ nica Facultade de Qu ı´ mica, Universidade de Santiago de Compostela,
1
5782 Santiago de Compostela, Spain
Received 2 May 2007; revised 1 June 2007; accepted 4 June 2007
Available online 9 June 2007
Abstract—Two cyclopentadienones were generated and both underwent conrotatory electrocyclization as expected based on Wood-
ward–Hoffmann rules. This result lends support to the idea that these ring-closing reactions are, in fact, pericyclic processes.
Ó 2007 Elsevier Ltd. All rights reserved.
Electrocyclic reactions represent a powerful method for
creating rings efficiently and stereoselectively. A num-
O
O
1
ber of strategies are available for conducting and cata-
lyzing such reactions. Recently, we reported the use of
Br
TEA (3 equiv)
o
TFE, 70 C, 2 days
‘
deantiaromatization’ as a driving force for the electro-
cyclization of a set of cyclopentadienones, in conjunc-
tion with studies ultimately directed at the synthesis of
hamigeran B. For example, treatment of 1 with trieth-
78%
O
O
2
2
1
O
O
ylamine in trifluoroethanol (TFE) for 2 days at 70 °C
resulted in the formation of 4 in 78% yield. Presumably,
cyclopentadienone 2 was generated from 1 by a 1,4-elim-
ination. Cyclization and trapping then proceeded to
afford 4 as shown in Scheme 1.
_
O
O
H
TFE
conrotatory
ring closure
+
As a pericyclic process, this cyclization should proceed
stereospecifically in a conrotatory fashion, as it involves
TFEO
O
O
4
8
p-electrons. To probe this, we prepared two isomeric
3
O
O
precursors and examined their cyclizations to establish
whether this is the case. This Letter contains the details
of that study.
Scheme 1. Cyclization through deantiaromatization.
The two substrates required for this study were prepared
from the halogen–metal exchange reaction between 5
16% (80% based on recovered starting material) yields
after hydrolysis (Scheme 2). The low (unoptimized)
4
and 6 with n-BuLi followed by transmetalation with
yield for 9 can be ascribed to steric effects, but may also
3
5
CeCl . The organocerium reagents were reacted with
be due to the quality (e.g., dryness) of the CeCl . In our
3
3
7
and the products 8 and 9 were obtained in 85% and
hands, drying protocols for this salt do not always pro-
duce consistent results.
6
*
Corresponding author. Tel.: +1 573 882 1097; fax: +1 573 882
With the pure E and Z cyclopentenones in hand, electro-
2754; e-mail: HarmataM@missouri.edu
cyclic reactions were carried out under standard reaction
0040-4039/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2007.06.011

5920
M. Harmata et al. / Tetrahedron Letters 48 (2007) 5919–5922
O
O
1
. 5 or 6, n-BuLi, THF
Me
3.0 eq TEA, TFE
Br
Br
o
-
78 C, 15 min
C= 0.01 M
o
2
. CeCl , THF, -78 C, 1 h
Br
3
o
5
0 C, 7 d
Me
Me
o
3
. 7, THF, -78 C, 24 h
o
or 70 C, 2 d
5
6
(96%
E
)
(100%
Z
)
8
9
(85%, 96%
E
)
8 (96% E)
O
(16%, 100%
Z
)
O
Br
7
O
1
2
4
H
OEt
+
3
Me
Scheme 2. Preparation of starting materials.
Me
F CH CO
3
2
Me
O
1
0
11
7
conditions (50 °C, 7 days or 70 °C, 2 days). When E
8
isomer 8 (96% E; 24:1, E:Z) was treated with triethyl-
Scheme 3. Cyclization of 8.
amine at 50 °C for 7 days, the desired cyclization adduct
1
0 was isolated in 43% yield with nearly complete diaste-
reoselectivity (10:1), accompanied by 33% recovered
starting material and some cyclopentadienone dimer
9
1
1 (Scheme 3). The diastereoselectivity in the formation
of 10 was not complete as the starting material 8 con-
tained some 9 (4%). The change in the product ratio
vis- a` -vis the ratio of starting materials suggested that 9
reacted more efficiently than 8. The ratio of 8, 10, and
1
1 in the crude reaction mixture was 0.39:0.51:0.10 by
1
H NMR.
The amount of dimer increased significantly when 8 was
stirred in TFE at 70 °C for 2 days. The expected product
1
0 was isolated in 39% yield and (8, 10, and dimer:
0
.33:0.48:0.19). The stereochemistry of 10 was deter-
mined by X-ray crystallography and is consistent with
1
0
our proposed conrotatory cyclization mechanism.
NMR data (Fig. 1) in solution are also in agreement
with the X-ray data as evident from the observed
NOE signals between H and the methyl group and also
2
between the H and H protons.
3
4
O
H
Me
F CH CO
3
2
12
Figure 1. NOE correlations for 10 (top) and 12 (bottom). Dashed lines
In contrast to 8, when the Z isomer 9 was treated with
base, the reaction was surprisingly clean. Product 12
was isolated in 56% yield as a single diastereomer,
indicate weak NOEs.
We have investigated the proposed pericyclic processes
by means of DFT computations. An unrestricted bro-
ken-spin-symmetry (UBS) procedure was used for
structures with a biradicaloid character. The structures
7
,11
together with 27% recovered starting material.
Since
1
4
the product could not be crystallized, the stereochemical
assignment of 12 was based on the analysis of NOESY
1
2
15
*
experiments and scalar coupling constants. First, the
were optimized at the B3LYP /6-31G level upon
which single point energies were computed at B3LYP/
3
observed JH3–H4 coupling constant of 2.2 Hz agrees
*
*
better with the computed equatorial–equatorial con-
6-311+G . Solvation effects in trifluoroethanol were
1
3
16
stant, using the Hassnoot–Altona empirical equation
modeled using the continuum PCM model, employing
of 2.0 Hz. Placing the trifluoroethoxy group in the equa-
torial position gives an equatorial axial computed J
ethanol parameters as provided in GAUSSIAN03 but with
3
a relative permitivity e of 26.73 and effective solvent
˚
constant of 4.4 Hz; secondly, no cross peak between
H and H was observed in the NOESY experiment.
radius of 0.3 A; single point energies were computed at
*
*
the PCM-B3LYP/6-311+G level as well. Harmonic
2
4

M. Harmata et al. / Tetrahedron Letters 48 (2007) 5919–5922
5921
L. Barnes for acquisition of X-ray data. A.N. thanks the
Centro de Supercomputaci o´ n de Galicia (CESGA) for
computer time.
O
O
H
R1
H
R1
R2
R2
O
Supplementary data
1
1
5
6
32.4 ( 29.9
35.4 ( 33.9
)
)
17 Š1.9 ( Š8.3
18 Š4.8 ( Š8.0
)
)
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.tetlet.
2007.06.011.
O
O
R2
R1
1
1
3
4
0.0
0.0
H
H
References and notes
. (a) Beaudry, C. M.; Malerich, J. P.; Trauner, D. Chem.
R2
R1
R1
R2
1
Rev. 2005, 105, 4757; (b) Frontier, A. J.; Collison, C.
Tetrahedron 2005, 61, 7577.
1
2
9 26.3 ( 18.8
0 27.8 ( 20.5
)
)
21 18.4 ( 11.9
22 16.8 ( 10.2
)
)
2
. Harmata, M.; Zheng, P.; Schreiner, P. R.; Navarro-
V a´ zquez, A. Angew. Chem., Int. Ed. 2006, 45, 1966.
1
1
3,15,17,19,21 ( R1 = H, R2 = Me)
4,16,18,20,22 ( R1 = Me, R2 = H )
3. Narayanan, B. A.; Bunnelle, W. H. Tetrahedron Lett.
987, 28, 6261.
4. Compound 8: 85% yield; pale yellow solid; mp: 92–93 °C;
1
Scheme 4. Activation and reaction B3LYP/6-311+G**//B3LYP/6-
1G* DG298.15 K free energies. Regular font: gas phase values, italics:
À1
1
3
IR (film): 2908.9, 1707.6, 1613.6, 1433.8, 1184.5 cm ; H
NMR (500 MHz, CDCl ): d 7.56 (d, J = 7.5 Hz, 1H), 7.37
td, J = 7.5, 1.0 Hz, 1H), 7.30 (td, J = 7.5, 1.0 Hz, 1H),
PCM (TFE) data.
3
(
7
.14 (dd, J = 7.5, 1.0 Hz, 1H), 6.19–6.27 (m, 2H), 2.93–
frequencies were computed on all optimized structures
to verify the nature of the stationary points and to ob-
tain thermochemical quantities. Reported free energies
2.95 (m, 2H), 2.71–2.73 (m, 2H), 1.87 (d, J = 5.0 Hz, 3H);
13
C NMR (125 MHz, CDCl
3
): d 202.2, 174.0, 135.6, 134.4,
1
1
2
30.1, 129.7, 128.8, 127.6, 127.4, 126.8, 125.7, 34.2, 33.9,
+
9.5; HRMS Calcd for C14H13BrO M , 276.0144; found,
DG2
were obtained by inclusion of B3LYP/6-
98.15 K
*
76.0137. Compound 9: 16% yield (80% based on recov-
3
1G thermochemical corrections in the B3LYP/6-
ered starting material); solid, mp: 64–65 °C; IR: 3015.2,
*
*
17
3
11+G computations.
À1
1
1
7
7
715.8, 1613.6 cm ; H NMR 300 MHz, CDCl ): d 7.24–
3
.45 (m, 4H), 6.33 (d, J = 11.4, 1H), 5.83 (dt, J = 11.4,
.0 Hz, 1H), 2.93–2.96 (m, 2H), 2.67–2.70 (m, 2H), 1.75
(dd, J = 7.0, 1.7 Hz, 3H); C NMR (75.0 MHz, CDCl ): d
The computations (Scheme 4) help elucidate the some-
what unexpected finding that 9 is better behaved than
1
3
3
8
. This can be rationalized in terms of the relative
201.5, 173.3, 134.9, 134.6, 129.8, 129.0, 128.8, 127.5, 126.8,
energies of the zwitterionic intermediates 21 and 22.
Although the position of the methyl group over the p-
system in transition state 20 results in a slightly higher
activation free energy for the formation of zwitterion
126.7, 124.3, 33.2, 32.3, 14.4; HRMS Calcd for C14
M , 276.0144; found, 276.0137.
. The reactivity of cerium chloride is dependent on the
manner in which the cerium chloride is activated. See also:
H
13BrO
+
5
6
(
a) Conlon, D. A.; Kumke, D.; Moeder, C.; Hardiman,
M.; Hutson, G.; Sailer, L. Adv. Synth. Catal. 2004, 346,
307; (b) Evans, W. J.; Feldman, J. D.; Ziller, J. W. J. Am.
2
2 as compared to 21, formation of the former is less
endothermic, which, as a consequence, results in a faster
overall process, making side reactions less competitive.
Alternatively, the electrocyclization might be reversible.
The computational data suggest reversibility should be
more important for the cyclization of 13 than 14. This
would result in the regeneration of 13, from zwitterion
1
Chem. Soc. 1996, 118, 4581.
. Other methods were used for dehydrating CeCl Æ7H O: (a)
3 2
Takeda, N.; Imamoto, T. Org. Synth. 1999, 76, 228; (b)
Paquette, L. A. Encyclopedia of Reagents for Organic
Synthesis; Wiley: Chichester, 1995 pp 1031.
2
1, allowing it to engage in side reactions and thus
7. Typical electrocyclic reaction procedure: To a trifluoroeth-
anol (10.3 mL) solution of E-8 (28.7 mg, 0.103 mmol) was
added triethylamine (43 lL, 0.31 mmol). The resulting
mixture was heated under certain conditions (method A:
2
lowering the yield of 10. As previously observed, the
conrotatory process for 13 and 14 is kinetically and ther-
modynamically favored over the disrotatory cyclizations
involving transition states 15 and 16, respectively.
5
0 °C, 7 days; method B: 70 °C, 2 days). The reaction was
monitored by TLC. After the completion of the reaction,
trifluoroethanol was evaporated and the residue was
In conclusion, we uncovered new evidence for an
8
dissolved in CH
2 2
Cl and washed with 1 N HCl, saturated
p-electron conrotatory mechanism for the cyclization
NaHCO , H O and brine. The extract was dried over
3
2
reaction of cyclopentadienones. Further studies of the
mechanism of this process and applications to synthesis
are underway.
MgSO4 and the solvent was removed by a rotavap.
Column chromatography (15% EtOAc, 30% EtOAc) gave
the desired product.
. The E/Z ratio was derived from the respective alkenes by
proton NMR.
8
9
. Compound 10 (from 8): Method A: 43% yield (33%,
recovered starting material); method B: 39% yield (27%
recovered starting material). Colorless crystal, mp: 132–
133 °C; IR: m 2917.1, 1687.1, 1670.8, 1593.1, 1270.3,
Acknowledgements
This work was supported by the National Science Foun-
dation to whom we are grateful. Thanks to Dr. Charles
À1
1
1151.8, 1115.1 cm
;
3
H NMR (250 MHz, CDCl ): d

5922
M. Harmata et al. / Tetrahedron Letters 48 (2007) 5919–5922
7
6
3
6
1
.78–7.81 (m, 1H), 7.44–7.51 (m, 2H), 7.29–7.33 (m, 1H),
(d, J = 2.2 Hz, 1H), 3.80–4.06 (m, 3H), 2.67 (dd, J = 18.6,
6.8 Hz, 1H), 2.53–2.59 (m, 1H), 2.35 (dd, J = 18.6, 3.5 Hz,
.44 (d, J = 2.2 Hz, 1H), 4.39 (d, J = 2.4 Hz, 1H), 3.78–
.94 (m, 2H), 3.44–3.54 (m, 1H), 2.75 (dd, J = 18.0,
.4 Hz, 1H), 2.25 (dd, J = 18.0, 4.1 Hz, 1H), 1.95–2.08 (m,
1
3
1H), 0.73 (d, J = 7.1 Hz, 3H); C NMR (75.0 MHz,
CDCl ): d 207.9, 171.5, 133.1, 131.8, 131.5, 129.8, 129.3,
127.0, 126.3, 123.8 (q, JC–F = 279.2 Hz), 82.5, 66.1 (q,
C–F = 34.1 Hz), 38.3, 37.8, 34.7, 10.2; HRMS Calcd
3
1
3
H), 1.29 (d, J = 6.8 Hz, 3H); C NMR (125 MHz,
): d 207.3, 173.6, 135.8, 131.0 130.3, 130.0, 129.7,
27.8, 125.1, 124.0 (q, JC–F = 279.2 Hz), 81.1, 66.2 (q,
C–F = 34.1 Hz), 41.1, 40.5, 40.4, 15.6; HRMS Calcd for
CDCl
1
J
C
3
J
+
for C H F O Na [M+Na] , 319.0916; found, 319.0923.
16 15 3 2
12. Coupling constants were computed using the NMRDev
program [http://desoft03.usc.es/armando/software.html]
+
16
H
15
F
3
O
2
Na [M+Na] , 319.0916; found, 319.0902.
*
1
1
0. Crystallographic data (excluding structure factors) for the
on B3LYP/6-31G optimized structures.
structure of compounds in this Letter has been deposited
with the Cambridge Crystallographic Data Centre as
supplementary publication numbers CCDC 637396 (10).
Copies of the data can be obtained, free of charge, on
application to CCDC, 12 Union Road, Cambridge CB2
13. Hassnoot, C. A. G.; De Leeuw, F. A. A. M.; Altona, C.
Tetrahedron 1980, 36, 2783.
14. (a) Gr a¨ fenstein, E.; Kraka, E.; Filatov, M.; Cremer, D.
Int. J. Mol. Sci. 2002, 3, 360; (b) Schreiner, P. R.;
Navarro-V a´ zquez, A.; Prall, M. Acc. Chem. Res. 2005, 38,
29.
15. (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648; (b) Lee, C.;
Yang, w.; Parr, R. G. Phys. Rev. B 1988, 37, 785.
16. Miertus, S.; Scrocco, E.; Tomasi, J. Chem. Phys. 1981, 55,
11.
17. Frisch, M. J. et al.; GAUSSIAN03, Revision B.05, Gaussian:
Wallingford, CT, 2004.
1EZ, UK [fax: +44
0
1223 336033 or e-mail;
deposit@ccdc.cam.ac.uk].
1. Compound 12 (from 9): Method A: 56% yield (27%
recovered starting material). solid, mp: 55–56 °C; IR: m
2
1
966.1, 2925.3, 1699.4, 1597.2, 1278.5, 1151.8,
À1
1
102.8 cm
;
H NMR (300 MHz, CDCl ): d 7.74–7.77
3
(
m, 1H), 7.37–7.53 (m, 3H), 6.52 (d, J = 1.5 Hz, 1H), 4.41