Enantioselective [6π]-photocyclization reaction of an acrylanilide mediated by a chiral host. Interplay between enantioselective ring closure and enantioselective protonation

Article abstract of DOI:10.1021/jo026602d

The [6π]-photocyclization of the anilides 1a and 5 was studied in the absence and in the presence of the enantiomerically pure chiral lactam 4. The relative configuration of the products was unambiguously established by single-crystal X-ray crystallography and by NMR spectroscopy. A significant enantiomeric excess was observed upon reaction of compound 1a to its photocyclization products at -55 °C employing lactam 4 as a chiral complexing agent in toluene as the solvent (66% yield). The trans product ent-3a was obtained in 57% ee, and the minor diastereoisomer (trans/ cis = 73/27), cis product ent-2a, was obtained in 30% ee. DFT calculations were conducted modeling the complexation of intermediates 8 and ent-8 to host 4. In agreement with steric arguments concerning the conrotatory ring closure of la, the formation of ent-8 is favored leading to the more stable complex 4·ent-8 as compared to 4·8. Whereas the enantioselectivity in the photocyclization to trans compound ent-3a increased upon reduction in the reaction temperature, the enantiomeric excess in the formation of cis compound ent-2a went through a maximum at -15 °C (45% ee) and decreased at lower temperatures. Deuteration experiments conducted with the pentadeuterated analogue of 1a, d⁵-1a, revealed that the protonation of the intermediates 8 and ent-8 is influenced by chiral amide 4. In the formation of ent-3a/3a, both the enantioselective ring closure and the enantioselective protonation by amide 4 favor the observed (6aS,10aS)-configuration of the major enantiomer ent-3a. In the formation of ent-2a/2a, the enantioselective ring closure (and the subsequent diastereoselective protonation) favors the (6aR,10aS)-configuration that is found in compound 2a. Contrary to that, the enantioselective protonation by amide 4 shows a preference for ent-2a with the (6aS,10aR)-configuration.

Full text of DOI:10.1021/jo026602d

En a n tioselective [6π]-P h otocycliza tion Rea ction of a n Acr yla n ilid e
Med ia ted by a Ch ir a l Host. In ter p la y betw een En a n tioselective
Rin g Closu r e a n d En a n tioselective P r oton a tion
Thorsten Bach,*,† Benjamin Grosch,† Thomas Strassner,‡,§ and Eberhardt Herdtweck‡,|
Lehrstuhl fu¨r Organische Chemie I, Technische Universita¨t Mu¨nchen, D-85747 Garching, Germany, and
Lehrstuhl fu¨r Anorganische Chemie, Technische Universita¨t Mu¨nchen, D-85747 Garching, Germany
thorsten.bach@ch.tum.de
Received October 24, 2002
The [6π]-photocyclization of the anilides 1a and 5 was studied in the absence and in the presence
of the enantiomerically pure chiral lactam 4. The relative configuration of the products was
unambiguously established by single-crystal X-ray crystallography and by NMR spectroscopy. A
significant enantiomeric excess was observed upon reaction of compound 1a to its photocyclization
products at -55 °C employing lactam 4 as a chiral complexing agent in toluene as the solvent
(66% yield). The trans product ent-3a was obtained in 57% ee, and the minor diastereoisomer (trans/
cis ) 73/27), cis product ent-2a , was obtained in 30% ee. DFT calculations were conducted modeling
the complexation of intermediates 8 and ent-8 to host 4. In agreement with steric arguments
concerning the conrotatory ring closure of 1a , the formation of ent-8 is favored leading to the more
stable complex 4‚ent-8 as compared to 4‚8. Whereas the enantioselectivity in the photocyclization
to trans compound ent-3a increased upon reduction in the reaction temperature, the enantiomeric
excess in the formation of cis compound ent-2a went through a maximum at -15 °C (45% ee) and
decreased at lower temperatures. Deuteration experiments conducted with the pentadeuterated
analogue of 1a , d5-1a , revealed that the protonation of the intermediates 8 and ent-8 is influenced
by chiral amide 4. In the formation of ent-3a /3a , both the enantioselective ring closure and the
enantioselective protonation by amide 4 favor the observed (6aS,10aS)-configuration of the major
enantiomer ent-3a . In the formation of ent-2a /2a , the enantioselective ring closure (and the
subsequent diastereoselective protonation) favors the (6aR,10aS)-configuration that is found in
compound 2a . Contrary to that, the enantioselective protonation by amide 4 shows a preference
for ent-2a with the (6aS,10aR)-configuration.
In tr od u ction
plied to the synthesis of naturally occurring hetero-
cycles.5,6 Under oxidative conditions, the formation of a
central pyridone ring E is facile especially if both rings
I and II are aromatic.7
The reaction of acrylanilides to the corresponding 3,4-
dihydroquinolin-2-ones was first described by Chapman
et al.8 and extensively studied by Ninomiya and co-
workers.9 The reactions proceed cleanly and with high
Acryl- and aroylenamides of the general structure A
undergo a photochemically allowed conrotatory [6π]-
cyclization1,2 reaction. If one of the two rings, I or II, is
an arene or a hetarene, a subsequent tautomerization of
intermediate B occurs regioselectively and yields either
product C or product D.3 Many variants of this useful
transformation have been explored4 and have been ap-
(4) Examples: (a) Ninomiya, I.; Naito, T.; Mori, T. J . Chem. Soc.,
Perkin Trans. 1 1973, 505-509. (b) Lenz, G. R. J . Org. Chem. 1974,
39, 2846-2851. (c) Kanaoka, Y.; Itoh, K.; Hatanaka, Y.; Flippen, J .
L.; Karle, I. L., Witkop, B. J . Org. Chem. 1975, 40, 3001-3003. (d)
Lenz, G. R. J . Heterocycl. Chem. 1979, 16, 433-437. (e) Kanaoka, Y.;
San-nohe, K. Tetrahedron Lett. 1980, 21, 3893-3896. (f) Ninomiya, I.;
Yamamoto, O.; Kiguchi, T.; Naito, T. J . Chem. Soc., Perkin Trans. 1
1980, 203-207. (g) Ninomiya, I.; Kiguchi, T.; Naito, T. J . Chem. Soc.,
Perkin Trans. 1 1983, 208-211. (h) Bates, R. B.; Kane, V. V.; Martin,
A. R.; Mujumdar, R. B.; Ortega, R.; Hatanaka, Y.; San-nohe, K.;
Kanaoka, Y. J . Org. Chem. 1987, 52, 3178-3180. (i) Fodor, L.; Szabo´,
J .; Berna´th, G.; MacLean, D. B.; Smith, R. W.; Ninomiya, I.; Naito, T.
J . Heterocycl. Chem. 1989, 26, 333-337. (j) Yamaguchi, S.; Yokoi, T.;
Yamada, M.; Arai, H.; Uchizo, Y.; Kawase, Y. J . Heterocycl. Chem.
1990, 27, 1003-1005.
† Lehrstuhl fu¨r Organische Chemie I.
‡ Lehrstuhl fu¨r Anorganische Chemie.
§ To whom inquiries about the DFT calculations should be ad-
dressed.
| To whom inquiries about the X-ray analyses should be addressed.
(1) General reviews: (a) Laarhoven, W. H. Org. Photochem. 1989,
10, 163-308. (b) Laarhoven, W. H. Org. Photochem. 1987, 9, 129-
224. (c) Mallory, F. B.; Mallory, C. W. Org. React. 1984, 30, 1-456. (d)
Laarhoven, W. H. Recl.: J . R. Neth. Chem. Soc. 1983, 102, 241-254.
(e) Laarhoven, W. H. Recl.: J . R. Neth. Chem. Soc. 1983, 102, 185-
204. (f) Tominaga, Y.; Castle, R. N. J . Heterocycl. Chem. 1996, 33, 523-
538.
(2) For a review on asymmetric photoreactions of R,â-unsaturated
carboxylic acid derivatives and enones: Pete, J .-P. Adv. Photochem.
1996, 21, 135-216.
(3) Reviews: (a) Ninomiya, I.; Naito, T. Heterocycles 1981, 15, 1433-
1462. (b) Lenz, G. R. Synthesis 1978, 489-518. (c) Zeller, K.-P. In
Houben-Weyl Methods of Organic Chemistry; Mu¨ller, E., Ed.; Thieme:
Stuttgart, 1975; Vols. 4 and 5, pp 542-546.
(5) Reviews: (a) Ninomiya, I.; Naito, T. In The Alkaloids; Brossi,
A., Ed.; Academic Press: San Diego, 1983; Vol. 22, pp 189-279. (b)
Ninomiya, I.; Yamamoto, O.; Kiguchi, T.; Naito, T.; Ishii, H. Hetero-
cycles 1977, 6, 1730-1734. Ninomiya, I. Heterocycles 1974, 2, 105-
123.
10.1021/jo026602d CCC: $25.00 © 2003 American Chemical Society
Published on Web 01/04/2003
J . Org. Chem. 2003, 68, 1107-1116
1107

Bach et al.
SCHEME 1
yields in a variety of solvents (e.g., ether, benzene, and
methanol). The diastereomeric ratio depends on the
solvent. In an aprotic solvent, the tautomerization occurs
to a large extent via a thermally allowed suprafacial
hydrogen 1,5-shift and yields preferentially the trans
product. In protic solvents or in the presence of Brønsted
acids, the proton is delivered intermolecularly to inter-
mediates of type B and the cis isomer prevails. Enantio-
selective solid-state [6π]-photocyclization reactions have
been studied by Toda et al.10 Inclusion crystals11 of
acrylanilides with tartaric acid-derived 1,4-dioxaspiro-
[4.4]nonanes and -[4.5]decanes yielded predominantly the
trans products in high enantiomeric excess (41-70%
yield, up to 98% ee). In a detailed study it was demon-
strated that the enantiomeric excess is a result of the
chiral conformation in the crystalline clathrate environ-
ment provided by the chiral host. The C-C bond forma-
tion step is decisive for the absolute configuration of the
product. The formation of enantiomerically enriched
photocyclization products in solution was reported by
Ninomiya et al.12 They used an external chiral proton
source, (+)-di(p-toluoyl)tartaric acid (DTTA), to achieve
an enantioselective protonation of the intermediate zwit-
terion. The best results are summarized in Scheme 2. The
photocyclization reaction of anilides 1 to tetrahydrophen-
athridinones 2 and 3 was conducted at 5-10 °C in a
solvent mixture. The protonation step and not the C-C
bond formation step is selectivity determining. The ee
value is low for the major cis isomer 2 and only slightly
higher for the minor trans isomer 3. The intermolecular
protonation by the chiral acid occurs at carbon atom C-6a
and is responsible for the enantiomeric excess. An
explanation for the face discrimination has been pro-
posed.12
SCHEME 2
We became interested in the [6π]-photocyclization in
connection with our work13 on enantioselective photo-
(6) For recent applications, see: (a) Rigby, J . H.; Maharoof, U. S.
M.; Mateo, M. E. J . Am. Chem. Soc. 2000, 122, 6624-6628. (b) Bois,
F.; Gardette, D.; Gramain, J .-C. Tetrahedron Lett. 2000, 41, 8769-
8772. (c) Schultz, A. G.; Holoboski, M. A.; Smyth, M. S. J . Am. Chem.
Soc. 1996, 118, 6210-6219. (d) Naito, T.; Yuumoto, Y.; Kiguchi, T.;
Ninomiya, I. J . Chem. Soc., Perkin Trans. 1 1996, 281-288. (e)
Heathcock, C. H.; Norman, M. H.; Dickman, D. A. J . Org. Chem. 1990,
55, 798-811.
(7) (a) Kanaoka, Y.; Itoh, K. Synthesis 1972, 36. (b) Winterfeldt, E.;
Altmann, H. J . Angew. Chem., Int. Ed. Engl. 1968, 7, 466. (c) Kanaoka,
Y.; San-nohe, K.; Hatanaka, Y.; Itoh, K.; Machida, M.; Terashima, M.
Heterocycles 1977, 6, 29-32. (d) Yamaguchi, S.; Oh-hira, Y.; Yamada,
M.; Michitani, H.; Kawase, Y. Bull. Chem. Soc. J pn. 1990, 63, 952-
954.
(8) (a) Cleveland, P. G.; Chapman, O. L. Chem. Commun. 1967,
1064-1065. (b) Chapman, O. K.; Adams, W. R. J . Am. Chem. Soc. 1968,
90, 2333-2342. (c) See also: Ogata, Y.; Takai, K.; Ishino, I. J . Org.
Chem. 1971, 36, 3975-3979.
(9) (a) Ninomiya, I.; Yamauchi, S.; Kiguschi, T.; Shinobara, A.; Naito,
T. J . Chem. Soc., Perkin Trans. 1 1974, 1747-1751. (b) Ninomiya, I.;
Kiguchi, T.; Yamauchi, S.; Naito, T. J . Chem. Soc., Perkin Trans. 1
1976, 1861-1865. (c) Ninomiya, I.; Kiguchi, T.; Naito, T. Heterocycles
1978, 9, 1023-1029. (d) Ninomiya, I.; Kiguchi, T.; Yamauchi, S.; Naito,
T. J . Chem. Soc., Perkin Trans. 1 1980, 197-202. (e) Ninomiya, I.;
Hashimoto, C.; Kiguchi, T.; Naito, T. J . Chem. Soc., Perkin Trans. 1
1983, 2967-2971.
chemical reactions14 in solution. Host compounds such
as lactam 415 (Figure 1) have been established as
versatile complexing agents for achieving moderate to
high enantioselectivities in photocycloaddition,16 Nor-
rish-Yang cyclization,17 and [4π]-photocyclization
reactions16c of lactams. The hosts interact with lactams
by two-point hydrogen binding, which facilitates the
(13) Initial studies: (a) Bach, T.; Bergmann, H.; Harms, K. J . Am.
Chem. Soc. 1999, 121, 10650-10651. (b) Bach, T. Synlett 2000, 1699-
1707. (c) Bach, T.; Bergmann, H.; Brummerhop, H.; Lewis, W.; Harms,
K. Chem. Eur. J . 2001, 7, 4512-4521.
(14) Reviews: (a) Everitt, S. R. L.; Inoue, Y. In Molecular and
Supramolecular Photochemistry: Organic Molecular Photochemistry;
Ramamurthy, V., Schanze, K. S., Eds.; Dekker: New York, 1999; Vol.
3, pp 71-130. (b) Inoue, Y. Chem. Rev. 1992, 92, 741-770. (c) Rau, H.
Chem. Rev. 1983, 83, 535-547.
(15) Bach, T.; Bergmann, H.; Grosch, B.; Harms, K.; Herdtweck, E.
Synthesis 2001, 1395-1405.
(16) (a) Bach, T.; Bergmann, H.; Harms, K. Angew. Chem., Int. Ed.
2000, 39, 2302-2304. (b) Bach, T.; Bergmann, H. J . Am. Chem. Soc.
2000, 122, 11525-11526. (c) Bach, T.; Bergmann, H.; Harms, K. Org.
Lett. 2001, 3, 601-603. (d) Bach, T.; Bergmann, H.; Grosch, B.; Harms,
K. J . Am. Chem. Soc. 2002, 124, 7982-7990.
(17) (a) Bach, T.; Aechtner, T.; Neumu¨ller, B. Chem. Commun. 2001,
607-608. (b) Bach, T.; Aechtner, T.; Neumu¨ller, B. Chem. Eur. J . 2002,
8, 2464-2475.
(10) (a) Tanaka, K.; Kakinoki, O.; Toda, F. Chem. Commun. 1992,
1053-1054. (b) Toda, F.; Miyamoto, H.; Kanemoto, K.; Tanaka, K.;
Takahashi, Y.; Takenaka, Y. J . Org. Chem. 1999, 64, 2096-2102. (c)
Ohba, S.; Hosomi, H.; Tanaka, K.; Miyamoto, H.; Toda, F. Bull. Chem.
Soc. J pn. 2000, 73, 2075-2085.
(11) Reviews: (a) Toda, F. Acc. Chem. Res. 1995, 28, 480-486. (b)
Tanaka, K.; Toda F. Chem. Rev. 2000, 100, 1025-1074.
(12) Naito, T.; Tada, Y.; Ninomiya, I. Heterocycles 1984, 22, 237-
240.
1108 J . Org. Chem., Vol. 68, No. 3, 2003

Enantioselective [6π]-Photocyclization
TABLE 1. P r ep a r a tion of Ra cem ic Com p ou n d s r a c-2a , r a c-3a , r a c-6, a n d r a c-7 by P h otocyliza tion u n d er Va r iou s
Con d ition s
c
temp
(°C)
yield
(%)b
entry
substrate
(mM)
solvent
rac-product
trans/cisa
1
2
3
4
5
6
1a
1a
1a
1a
5
4
4
4
4
10
10
toluene
toluene
toluene
Et2O
toluene
toluene
35
-15
-55
35
35
-15
3a /2a
3a /2a
3a /2a
3a /2a
7/6
36/64
36/62
46/54
46/54
54/46
77
66
65
82
61
66
c
5
7/6
-
a
b
Diastereomeric ratio was determined by GC analysis of the crude product mixture. Yield of isolated product (mixture of the two
diastereoisomers). c Not determined.
F IGURE 2. Structure of photocyclization products rac-6 and
rac-7.
F IGURE 1. Structure of host 4 employed in the enantiose-
lective [6π]-photocyclization of compounds 1a and 5.
In ether as the solvent, the simple diastereoselectivity
decreased (entry 4 in Table 1). Surprisingly, a decrease
in the simple diastereoselectivity was also observed while
decreasing the reaction temperature (entries 1-3). This
issue will be discussed in more detail later.
positioning of the substrates in a chiral environment.
Acyclic amides had not yet been tested in this context,
and we wondered whether these substrates would be
amenable to an enantioselective reaction in the presence
of host 4. We selected the previously used anilide 1a as
the substrate, which exhibits a hydrogen bond acceptor
(NH) and a hydrogen bond donor (CdO) site. Since a
cisoid arrangement of the amide is required for a suc-
cessful photocyclization, we were optimistic that binding
to host 4 and a face differentiation could be achieved. For
comparison, the analogous pyridine 5 was also studied.
It was suspected, however, that the basic pyridine
nitrogen atom would interfere with the hydrogen bonding
and that its reaction would consequently be less selective.
In this account, we report on the results of the study
that focused mainly on the [6π]-cyclization of substrate
1a . Contrary to our expectations, it turned out that the
enantioselectivity depends not only on the binding to the
host and the enantioface differentiation but also on the
ability of the host to act as a Brønsted acid. The study
was accompanied by deuteration experiments, which
shed light on the mechanism of the reaction, and by DFT
calculations, which give a more detailed picture of the
binding to hosts of type 4.
It turned out that compounds rac-2a and rac-3a are
separable by HPLC (reverse-phase column YMC ODS-
A; eluent, 30/70 f 60/40 MeCN/H2O). Moreover, all
stereoisomers (enantiomers and diastereoisomers) could
be separated by chiral GC (Suppelco â-Dex 325, 160 f
200 °C at 0.5 K/min), which in turn facilitated the
determination of ee values. The assignment of the rela-
tive configuration was based on reported data9a and
supported by the NMR data of the individual compounds.
The coupling constant (3J HH) of the two protons at C-6a
and C-10a is 4.8 Hz for the cis isomer rac-2a and 14.1
Hz for trans isomer rac-3a . In addition, we obtained
crystals of both diastereoisomers that were suitable for
single-crystal X-ray crystallography. The molecular struc-
tures of the starting material 1a , of the minor diastereo-
isomer rac-2a , and of the major trans diastereoisomer
rac-3a are found in the Supporting Information.
In a second set of experiments, the irradiation of
substrates 1a and 5 was conducted in the presence of
host 4. The experimental setup was identical to the setup
used in the racemic case. Upon determination of the
simple diastereoselectivity, the host was removed by
column chromatography and the purified mixture of
diastereoisomers was studied by chiral GC. The choice
of Vycor glass as the filter was a consequence of the
strong decrease in host recovery upon using quartz glass
(90% transmission at 250 nm). With duran glass (5%
transmission at 250 nm), the reaction remained incom-
plete. The recovery yields are also provided in Table 2.
In all cases in which host 4 and substrate 1a were used
(entries 1-9), the major enantiomer detected was ent-
3a for the isolated trans diastereoisomer and ent-2a for
the isolated cis diastereoisomer (vide infra). The deter-
mination of the absolute configuration was not possible
for the products derived from substrate 5 (entries 10 and
11). Compared to the earlier results obtained in the
absence of host 4 (Table 1), the trans/cis ratio shifted in
Resu lts a n d Discu ssion
Photocyclization and Assessment of the Product Con-
figuration. To obtain racemic compounds for comparison,
amides 1a and 5 were irradiated in an immersion
apparatus with a mercury high-pressure lamp (Original
Hanau TQ 150). A Vycor filter (90% transmission at 330
nm, 45% transmission at 250 nm) was used to cut off
short-wavelength UV light. Phenyl derivative 1a yielded
the racemic cis and trans products rac-2a and rac-3a ,
respectively (Table 1). Pyridyl derivative 5 gave the
corresponding cis and trans products rac-6 and rac-7,
respectively (Figure 2). Whereas cis product rac-2a was
favored in the former case (entries 1-4), a slight prefer-
ence for trans isomer rac-7 (entry 5) was observed in the
latter case.
J . Org. Chem, Vol. 68, No. 3, 2003 1109

Bach et al.
TABLE 2. P r ep a r a tion of En a n tiom er ica lly En r ich ed Com p ou n d s en t-2a , en t-3a , 6/en t-6, a n d 7/en t-7 by
P h otocycliza tion in th e P r esen ce of Host 4
camide
(mM)
chost
(mM)
temp
(°C)
time
rec. host
(%)b
yield
(%)c
eetrans
eecis
(%)e
entry
substrate
solvent
(min)a
trans/cisd
(%)e
1
2
3
4
5
6
7
8
9
1a
1a
1a
1a
1a
1a
1a
1a
1a
5
4
30
4
4
4
4
4
4
4
toluene
toluene
toluene
toluene
toluene
toluene
toluene
Et2O
4.8
36
35
35
35
-15
-15
-55
-55
35
120
120
120
135
135
180
180
180
180
90
86
-
75
-
52/48
52/48
51/49
55/45
58/42
71/29
73/27
58/42
58/42
57/43
55/45
35
35
38
39
45
49
57
22
28
25
8
34
39
37
36
45
28
30
28
34
10
13
f
f
9.6
4.8
9.6
4.8
9.6
4.8
9.6
14.4
6
86
75
78
76
74
84
84
77
79
75
63
70
60
66
82
82
79
90
Et2O
toluene
toluene
35
35
-15
10
11
6
2.5
5
135
a
b
Time after which the conversion was complete. Yield of recovered host after the enantioselective photochemical reaction. c Yield of
d
isolated product (mixture of the two diastereoisomers). The diastereomeric ratio was determined by GC analysis of the crude product
mixture. e The enantiomeric excess was determined by chiral GC analysis. f Not determined.
favor of the trans diastereoisomer. In addition, the trans
preference increased significantly at lower temperatures
(entries 6 and 7).
) +53.9 (c 0.36, CHCl3) and that enantiomerically pure
compound ent-3b is levorotatory with [R]D ) -169.2 (c
0.50, CHCl3).12 The compounds we isolated from the
photocyclization of substrate 1a were levorotatory. On
the basis of measurements conducted with diastereo-
merically pure products (HPLC separation), we deter-
mined the specific rotation of ent-3a with 47% ee to be
[R]2D0 ) -68.7 (c 0.42, CHCl3) and that of ent-2a with
40% ee to be [R]2D0 ) -19.7 (c 0.23, CHCl3). The magni-
tude of the specific rotation for 3a /ent-3a was conse-
quently in line with previous data, whereas the value we
had determined for 2a /ent-2a was higher than expected.
To confirm the direction of the specific rotation, quantum
chemical calculations were conducted.18 The calculation
did indeed confirm the assignment, i.e., that both ent-2a
and ent-3a are levorotatory.
An additional line of evidence for the configuration
assignment was derived from N-methylation of enantio-
merically enriched products ent-2a and ent-3a (Scheme
3). N-Methylation of the levorotatory trans diastereoiso-
mer ent-3a gave a mixture of trans product ent-3b, which
was levorotatory, and cis diastereoisomer 2b, which was
dextrorotatory. The enantiomeric excess (47% ee) re-
mained unchanged. The cis diastereoisomer can only be
formed by epimerization at carbon atom C-6a. Cis dias-
tereoisomer ent-2b obtained from enantiomerically en-
riched cis compound ent-2a was levorotatory. The result
unambiguosly proved that the absolute configuration at
carbon atom C-6a is responsible for the direction of the
Irradiation experiments conducted at room tempera-
ture in toluene revealed significant enantioselectivities
(entries 1-3) for both the trans and cis products. Neither
the absolute substrate concentration nor the relative
concentration of substrate/host altered the enantioselec-
tivity. In all earlier experiments,16,17 an increased host
concentration (relative to the substrate) had led to a
significant increase in the product ee. It was even more
surprising that the major enantiomers ent-3a and ent-
2a had identical absolute configurations at the stereo-
genic carbon atom C-6a but differed in the absolute
configuration at the stereogenic carbon atom C-10a,
which is formed in the C-C bond-forming step. In the
more polar solvent diethyl ether (entries 8 and 9), the
enantioselectivity was lower, which was in line with our
expectations. Pyridine derivative 5 gave the correspond-
ing products 6/ent-6 and 7/ent-7 in very modest enantio-
selectivities (entry 10). The baseline separation (GC) was
not complete for trans diastereoisomer 7/ent-7, and its
ee value was not accurately determined. In this respect,
the ee value obtained at lower temperatures (entry 11)
was not fully accurate either. As both values remained
low, substrate 5 was not studied more closely. Lowering
the reaction temperature in the irradiation of compound
1a had an astonishing effect (entries 4-7). Whereas the
enantiomeric excess in which enantiomer ent-3a of the
trans diastereoisomer was formed increased up to 57%
ee, the enantiomeric excess for the minor diastereoisomer
went through a maximum at -15 °C (entry 5) and
decreased significantly at -55 °C (entries 6 and 7). The
results were fully reproducible and raised the question
as to which factors had an influence on the enantio-
selectivity.
(18) Quantum chemical calculations were performed with the TUR-
BOMOLE suite of programs.18a Structures of 2a and 3a were fully
optimized at the density functional (DFT) level employing the B3LYP
hybrid functional18b and a Gaussian AO basis of valence-double-ú
quality, including polarization functions on nonhyrdogen atoms (SV-
(P)).18c Structures were used in subsequent calculations of the frequency-
dependent optical rotation ([R]D) at the sodium D-line wavelength.
These calculations were performed in the framework of time-dependent
DFT18d as described in detail.18e.f In the TDDFT approach, aug-cc-pVDZ
basis sets,18g the BP86 density functional,18h and the RI approximation18f
were used. The results are +133 and +224 for 2a and 3a , respectively.
(a) Ahlrichs, R.; Ba¨r, M.; Ha¨ser, M.; Horn, H.; Ko¨lmel, C. Chem. Phys.
Lett. 1989, 162, 165. Current version: see http://www.turbomole.de.
(b) Becke, A. D. J . Chem. Phys. 1993, 98, 5648. Stephens, P. J .; Devlin,
F. J .; Chabalowski, C. F.; Frisch, M. J . J . Phys. Chem. 1994, 98, 11623.
(c) Scha¨fer, A.; Horn, H.; Ahlrichs, R. J . Chem. Phys. 1992, 97, 2571.
(d) Bauernschmitt, R.; Ahlrichs, R. Chem. Phys. Lett. 1996, 256, 454.
(e) Grimme, S. Chem. Phys. Lett. 2001, 339, 380. (f) Grimme, S.;
Furche, F.; Ahlrichs, R. Chem. Phys. Lett. 2002, 361, 321. (g) Dunning,
T. H. J . Chem. Phys. 1993, 98, 7059. Kendall, R. A.; Dunning, T. H.;
Harrison, R. J . J . Chem. Phys. 1992, 96, 6796. (h) Becke, A. D. Phys.
Rev. A. 1988, 38, 3098. Perdew, J . B. Phys. Rev. B. 1986, 33, 8822.
Before this question is addressed, it should be men-
tioned how the absolute configuration of compounds 2a ,
ent-2a , 3a , and ent-3a was assigned. In previous work,
Ninomiya et al. had found that enantiomerically pure
compound 2a is dextrorotatory with a reported specific
rotation [R]D ) +22.9 (c 0.48, CHCl3) and that enantio-
merically pure compound ent-3a is levorotatory with [R]D
) -158.8 (c 0.16, CHCl3). In addition, it was shown that
in the N-methyl series, enantiomerically pure compound
2b is dextrorotatory with a reported specific rotation [R]D
1110 J . Org. Chem., Vol. 68, No. 3, 2003

Enantioselective [6π]-Photocyclization
SCHEME 3
F IGURE 3. Complexation of substrate 1a to host 4 and
subsequent conrotatory photocyclization to complex 4‚ent-8.
the amide is located directly opposite the tetrahydronaph-
thalene moiety of the host. The phenyl group of the
anilide is not in the immediate vicinity of the sterically
bulky environment. Upon conrotatory ring closure, the
cyclohexene ring can either turn in the direction of the
tetrahydronaphthalene or it can move away and in doing
so open its Si-face to attack from the benzene ring.
According to simple molecular models, the latter move-
ment, as schematically drawn in Figure 3, appears to be
favored for steric reasons. As a consequence, the product
ent-8 is formed as the major product, which is still bound
to the host.
SCHEME 4
By the same reasoning, product 4‚ent-8 should also be
thermodynamically favored over the alternative complex
4‚8. In the former complex, the hydrogen atom at carbon
atom C-10a points away from the tetrahydronaphthalene
and the cyclohexane is essentially unstrained. In the
latter complex, a more strained arrangement is expected.
As the diastereotopic complexes 4‚ent-8 and 4‚8 can be
envisaged as the closest accessible models for the dias-
tereotopic transition states of the photocyclization, we
tried to determine their relative energy by appropriate
calculations.
Density functional theory calculations (Becke3LYP/6-
31G*)20,21 were employed to study the system. The
interaction energy for the complexation of 1a by the host
molecule 4 is calculated to be -10.4 kcal/mol. The bond
lengths for the two NH‚‚‚O bonds are 1.89 and 1.84 Å,
showing that the hydrogen bonds and therefore the
coordination of 1a to 4 is weaker in this case compared
to other previously studied lactam guest compounds.16d
specific rotation. The major enantiomers ent-3a and ent-
2a have different absolute configurations at C-10a, i.e.,
the stereogenic center that is formed in the ring closure
step. Qualitatively, the specific rotations reported by
Ninomiya et al. were confirmed.
Con r ota tor y Rin g Closu r e. The degree of face dif-
ferentiation in the photocyclization step can be evaluated
from the data recorded in Table 2. The absolute config-
uration at carbon atom C-10a is established in this step
and cannot be inverted in the further course of the
reaction. Given zwitterionic intermediates 8 and ent-8
as primary products of a conrotatory [6π]-photocycliza-
tion, it is apparent (Scheme 4) that the products ent-3a
and 2a are formed from intermediate ent-8 and that the
products 3a and ent-2a are formed from intermediate 8.
Since the product distribution is known from experimen-
tal data (Table 2), the enantiomeric ratio with which ent-8
is formed relative to 8 can be calculated. The data are
implemented in Scheme 4 for irradiation experiments
that were conducted with 2.4 equiv of host (entries 3, 5,
and 7 in Table 2) in toluene as the solvent. At low
temperatures (-55 °C), a significant enantioselective
differentiation occurs in the presence of host 4 that leads
to the preferential formation of intermediate ent-8 (ent-
8/8 ) 67/33).
(19) The equilibrium between the cis and trans rotamers of com-
pound 1a is in favor of the trans rotamer, which complicates any
attempt to determine the complex formation quantitatively. NMR
titration experiments were not conclusive. In addition, as one referee
suggested, the photoexcited cis rotamer (being a stronger acid and a
stronger base after photoexcitation) might be trapped in complex 4‚
1a and consequently end up as 4‚ent-8.
(20) (a) Becke, A. D. J . Chem. Phys. 1993, 98, 5648-5652. (b) Lee,
C.; Yang, W.; Parr, R. G. Phys. Rev. B: Condens. Matter 1988, 37, 785-
789. (c) Vosko, S. H.; Wilk, L.; Nusair, M. Can. J . Phys. 1980, 58, 1200-
1211. (d) Stephens, P. J .; Devlin, F. J .; Chabalowski, C. F.; Frisch, M.
J . J . Phys. Chem. 1994, 98, 11623-11627.
(21) Frisch, M. J .; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J . R.; Zakrzewski, V. G.; Montgomery, J . A.,
J r.; Stratmann, R. E.; Burant, J . C.; Dapprich, S.; Millam, J . M.;
Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J .;
Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo,
C.; Clifford, S.; Ochterski, J .; Petersson, G. A.; Ayala, P. Y.; Cui, Q.;
Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J . B.; Cioslowski, J .; Ortiz, J . V.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.;
Fox, D. J .; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.;
Gonzalez, C.; Challacombe, M.; Gill, P. M. W.; J ohnson, B. G.; Chen,
W.; Wong, M. W.; Andres, J . L.; Head-Gordon, M.; Replogle, E. S.;
Pople, J . A. Gaussian 98, revision A.7; Gaussian, Inc.: Pittsburgh, PA,
1998.
To explain the differentiation of enantiotopic faces, we
invoke a binding of amide 1a in its cis configuration to
host 4.19 In this arrangement, the cyclohexene moiety of
J . Org. Chem, Vol. 68, No. 3, 2003 1111

Bach et al.
F IGURE 4. Calculated structure of cyclization product ent-8
and transition state ent-8TS that would lead to ent-8 on the
S0-hypersurface
Due to the size of the system, it has not been possible
to study the diastereotopic transition states for the
formation of 8 and ent-8 coordinated to host 4. In
addition, the photochemical reaction proceeds via excited
states and does not proceed on the S0-hypersurface. Still,
the conrotatory transition state for the formation of
enantiomer ent-8 from 1a shown in Figure 4, which is
calculated on the S0-hypersurface, can be considered a
model of the real transition state on the excited surface.
The conformation of the calculated transistion state ent-
8
TS provides a rough picture for the topologic change upon
going from 1a to 8/ent-8. The forming bond in the
transition state with a distance of 2.20 Å connects two
carbon centers, which are right between sp2- and sp3-
hybridization (Figure 4).
The complexation of 8 and ent-8 to 4 leads to diaster-
eomeric complexes 8‚4 and ent-8‚4, as shown in Figure
5. On the basis of the previous argument that these
complexes are the closest approximations to the corre-
sponding transition states, their relative energy should
reflect the outcome of the [6π]-cyclization. Indeed, it could
be shown that complex 4‚8 is disfavored as compared to
4‚ent-8. In agreement with the experimental results, the
latter complex was calculated to be more stable by 0.2
kcal/mol. From the calculation of the transition states
without the host molecule 4 and the energy difference
between the diastereotopic complexes, it is reasonable to
assume that the host-guest interaction will influence the
diastereotopic transition states. Indeed, Figure 5 nicely
complements Figure 3 by showing the disfavored inter-
action of the cyclohexene ring with the tetrahydronaph-
thalene backbone in 4‚8 and by demonstrating that the
repulsion between the phenyl ring and the tetrahy-
dronaphthalene is marginal in 4‚ent-8.
Concluding this chapter, the enantioselective formation
of intermediate ent-8 in the [6π]-photocyclization reaction
can be rationalized by steric interactions in the ring
closure step. The preferred formation of ent-8 is in line
with the observed preference for the major trans enan-
tiomer ent-3a . It is apparent, however, that the proto-
nation of intermediates 8 and ent-8 is involved in this
reaction as a second selection step. A protonation must
occur preferentially from the Re-face (relative to carbon
atom C-6a) of intermediate 8 to account for the preference
of ent-2a over 2a . Deuteration experiments were under-
taken to study the influence of possible proton sources
more closely. Given the use of imides related to compound
4 in enantioselective protonation experiments,22 its po-
tential influence was to be looked at.
F IGURE 5. Diastereotopic complexes 4‚8 and 4‚ent-8, shown
in two orientations.
[6π]-photocyclization determines the absolute configura-
tion at carbon atom C-10a of the product. The configu-
ration at carbon atom C-6a is established in the proto-
nation step, which can occur intramolecularly (1,5-H
shift) or intermolecularly. To distinguish between an
inter- and an intramolecular protonation, the deuterated
analogue d5-1a of compound 1a was prepared (Scheme
5). Acylation of cyclohexene carboxylic acid 9 with d5-
aniline gave the desired product. Its photocyclization
yielded photocyclization products, the deuterium content
of which was determined by 1H NMR experiments.
Irradiation experiments with compound d5-1a were con-
ducted at room temperature and at -55 °C in the absence
and in the presence of host 4. The most relevant results
of this study are summarized in Table 3. The deuterium
content in trans product 3a (d5-3a vs d4-3a ) remained
unchanged in all runs and was roughly 65%. The
deuterium content in cis product 2a was roughly 10% in
the absence of the host and 0% in its presence. The
observed selectivities (trans/cis ratio, ee) did not signifi-
cantly differ from the values recorded for the nondeu-
terated starting material (Tables 1 and 2). The trans/cis
ratio was ca. 35/65 in the absence of the host at room
temperature and at low temperatures (entries 1 and 3).
(22) (a) Yanagisawa, A.; Kikuchi, T.; Watanabe, T.; Yamamoto, H.
Bull. Chem. Soc. J pn. 1999, 72, 2337-2343. (b) Yanagisawa, A.;
Kikuchi, T.; Kuribayashi, T.; Yamamoto, H. Tetrahedron 1998, 54,
10253-10264. (c) Potin, D.; Williams, K.; Rebek, J ., J r. Angew. Chem.,
Int. Ed. Engl. 1990, 29, 1420-1422.
Deuteration Experiments. As previously discussed
(schemes 1 and 4; Figure 3), the ring closure step of the
1112 J . Org. Chem., Vol. 68, No. 3, 2003

Enantioselective [6π]-Photocyclization
TABLE 3. [6π]-P h otocycliza tion of Deu ter a ted Su bstr a te d 5-1a u n d er Va r iou s Con d ition s w ith Tolu en e a s th e Solven t
camide
mM)
chost
(mM)
temp
(°C)
time
yield
(%)b
eetrans
(%)d
Dtrans
(%)e
eecis
(%)d
Dcis
entry
substrate
(min)a
trans/cisc
(%)f
1
2
3
4
d5-1a
d5-1a
d5-1a
d5-1a
4
4
4
4
0
9.6
0
35
35
-55
-55
120
120
210
210
81
84
89
74
33/67
49/51
39/61
62/38
0
39
0
63
64
63
67
0
33
0
8
0
9
0
9.6
64
26
a
b
Time after which the conversion was complete. Yield of isolated product (mixture of the two diastereoisomers). c Diastereomeric
ratio was determined by GC analysis of the crude product mixture. Enantiomeric excess was determined by chiral GC. e Deuterium
d
content given in percent for racemic or enantiomerically enriched d5-3a /(d4-3a + d5-3a ). The deuterium content was determined by 1H
NMR spectroscopy. f Deuterium content given in percent for racemic or enantiomerically enriched d5-2a /(d4-2a + d5-2a ). The deuterium
content was determined by 1H NMR spectroscopy.
SCHEME 5
SCHEME 6
content and the trans/cis ratio. From this, the ratios
between 1,5-H-shift (x) and intermolecular protonation
processes (y and z) were determined to be 21/79 at 30 °C
and 25/75 at -55 °C. The facial diastereoselectivity of
the intermolecular protonation is represented by the
quotient y/z, with y and z representing the protonations
to the trans and cis compounds, respectively. It was found
to be 15/85 at 30 °C and 19/81 at -55 °C in favor of the
formation of the cis compound.
The interplay between enantioselective ring closure
and enantioselective protonation can be elucidated from
the product distribution of hosted irradiations (2.4 equiv
of 4). GC analyses provided the amounts of d4-2a and
ent-d4-2a , whereas only the total amounts of d4- and d5-
3a and of d4- and d5-ent-3a were obtainable by this
means. However, the total amount of pentadeuterated
d5-3a and d5-ent-3a could be determined by 1H NMR
measurements. Estimating a single parameter in the
reaction scheme allows the product distribution to be
calculated from these analyses. In our view, it is reason-
able to assume that the 1,5-H-shift on the face that is
not shielded by the host (x′′) occurs with the same
probability as it does in the unhosted reaction, i.e., x′′ )
x ) 0.21 at 30 °C and x′′ ) x ) 0.25 at -55 °C. From x′′
and the amount of d5-8 calculated from total 3a and ent-
d4-2a , the amount of d5-3a was derived. Subsequently,
the product distribution was easily obtained and is
graphically depicted in Scheme 7.
The ratio was altered in the experiments conducted in
the presence of host 4 (entries 2 and 4). It was 50/50 at
room temperature and 62/38 at -55 °C. Additionally,
modest to good ee values were recorded in these photo-
cyclization reactions.
What conclusions can be drawn from the deuteration
data regarding the mechanism of the enamide cyclization
and its control by host 4? Pentadeuterated trans product
rac-d5-3a is formed by an intramolecular 1,5-D-shift,23
whereas the tetradeuterated trans and cis compounds
rac-d4-3a and rac-d4-2a result from intermolecular proton-
transfer processes. The small percentage (8-9%) of
pentadeuterated cis compound rac-d5-2a produced from
irradiations in the absence of host (nonhosted irradia-
tions) was presumably formed by intermolecular deute-
rium transfer. In the presence of host 4, rac-d5-2a was
not formed at all. It was concluded that the intermolecu-
lar deuterium transfer is a minor pathway, and it was
therefore neglected in the mechanistic considerations.
The product distribution of nonhosted irradiations as
depicted in Scheme 6 was derived from the deuterium
As in the undeuterated case, the cyclization to the
intermediate d5-8/ent-d5-8 occurs with no enantioselec-
tivity (2% ee) at 30 °C and moderate enantioselectivity
(30% ee) at -55 °C. In intermediate ent-d5-8, the 1,5-D-
shift is facile (x′ ) 0.41 at 30 °C and x′ ) 0.51 at -55
°C). At the same time, the host acts as a proton source
and influences the protonation selectivity y/z. Enhanced
(23) For a previous study in which deuterium labeling was used to
prove the intramolecular nature of the 1,5-H-shift, see: Lenz, G. R. J .
Org. Chem. 1976, 41, 2201-2207.
J . Org. Chem, Vol. 68, No. 3, 2003 1113

Bach et al.
SCHEME 7
F IGURE 6. Dissociation of the NH‚‚‚O bond in complex 4‚8
and potential transition state for the protonation of 8 associ-
ated to host 4.
nature of this complex formation as determined by
microcalorimetry for the complex of 2-quinolone with host
4.16d In addition, the stabilization of the required cis
conformation of the amide in the transition state of the
cyclization has a greater effect on the relative velocities
of hosted and unhosted cyclizations at low temperatures.
The control of the protonation step is similarly effective
at 30 and -55 °C. Evidently, host 4 acts as a chiral acid
with similar selectivity, regardless of whether it is
involved in a defined complex by hydrogen bonds.
The fact that the mechanistic discussion is valid both
for the deuterated and the undeuterated case can be
concluded qualitatively from comparison of the product
distributions. The D-substitution does not significantly
change either the ratio of 8/ent-8, the trans/ cis ratio or
the enantiomeric excesses. The protonation selectivity of
the intermolecular H-transfer (y/z) must remain the same
in the undeuterated case, because no D is involved. With
the amount of cis product 2a being similar, y and
consequently the ratio of 1,5-H-shift and intermolecular
H transfer (x/y) must also be comparable to the undeu-
terated case. Moreover, the variation of the ratio of
diastereoisomers (trans/cis) in the hosted and unhosted
cyclizations at different temperatures is nicely reflected
by the consideration described and is in full agreement
with the mechanistic picture presented above.
protonation of d5-8 from the Re-face (z′′) leads to an
improved selectivity of y′′/z′′ ) 13/87 and 8/92 at 30 and
-55 °C, respectively. The same effect reduces the selec-
tivity of the protonation of ent-d5-8 to y′/z′ ) 43/57 at 30
°C and even causes a reversion to 56/44 at -55 °C. This
phenomenon is responsible for the astonishing variation
of the observed enantiomeric excesses at different tem-
peratures. At 30 °C, ent-d4-3a /ent-d5-3a (39% ee) and ent-
d4-2a (33% ee) are favored because of enhanced proto-
nation from the shielded face (y′ and z′′). At lower
temperatures, the control of the protonation is ac-
companied by an increased control of the C-C bond-
forming cyclization. In the case of the trans product, the
effects complement each other because ent-d4-3a /ent-d5-
3a are formed from the preferred intermediate ent-d5-8
upon protonation from the shielded face (y′) or by 1,5-D-
shift (x′). This leads to an increased enantioselectivity of
64% ee. The opposite development holds true for the cis
compound. Lactam ent-d4-2a is formed from the dis-
favored intermediate d5-8 by the favored protonation from
the shielded face (z′′), thereby reducing its enantio-
selectivity to 26% ee.
If one takes into account a coordination of the inter-
mediates 8 and ent- 8 to the host 4, the protonation
selectivity in favor of a Re-face attack is apparent.
Dissociation of the guest, which is a necessary require-
ment for protonation by the host, opens the Re-face for a
potential attack. While the NH‚‚‚O bond is broken, the
O‚‚‚HN might still be intact. A potential transition state
that is accessible for any intermediate 8 is schematically
drawn in Figure 6. From complex 4‚8 the transition state
is accessed by rotation about the O‚‚‚HN hydrogen bond
as depicted in Figure 6. The same argument holds true
for the analogous complex 4‚ent-8 (Figure 3).
Con clu sion
In summary, the individual steps involved in the
enantioselective [6π]-photocyclization of substrate 1a
have been elucidated. Initially, the relative and absolute
configuration of all possible stereoisomeric products 2a ,
ent-2a , 3a , and ent-3a was established using spectro-
scopic (X-ray crystallography, polarimetry, and NMR
spectroscopy) and computational methods. The preferred
formation of compound ent-3a in the presence of host 4
could be attributed to the preferred formation of inter-
mediate ent-8 upon conrotatory ring closure. Steric
arguments that were supported by DFT calculations
favor complex 4‚ent-8 vs 4‚8. Finally, by irradiation of
deuterated amide d5-1a , the variation of the enantio-
selective excesses with the temperature was fully ex-
plained as a result of host 4 controlling the cyclization
and acting as a chiral proton source. We are aware that
the mechanistic discussion of the deuteration experi-
ments has not included potential primary and secondary
isotope effects. Although it appears as if the effects are
not relevant for the stereochemical outcome, and argu-
ments why this is the case have been given, we are
The increased control of the cyclization at low temper-
atures can be rationalized by a higher association con-
stant of amide 1a and host 4. This is due to the exergonic
1114 J . Org. Chem., Vol. 68, No. 3, 2003

Enantioselective [6π]-Photocyclization
data: Rf ) 0.44 (2:1 Et2O/P); IR (KBr) ν˜ ) 3195 cm-1 (s), 3063
(s), 2933 (s), 2846 (s), 1673 (vs), 1591 (s), 1491 (s), 1391 (s),
1295 (s), 1252 (s), 867 (s), 834 (s), 751 (s).
currently trying to extract additional mechanistic infor-
mation from appropriate isotope labeling.
2a a n d en t-2a : 1H NMR (400 MHz, CDCl3) δ 1.46-1.74 (m,
Exp er im en ta l Section
3
3
3
8 H), 2.80 (virt. q, J = J = J ) 4.8 Hz, 1 H), 2.94 (virt. dt,
3J ) 10.4 Hz, J = J ) 4.8 Hz, 1 H), 6.73 (d, J ) 7.7 Hz, 1
3
3
3
Gen er a l In for m a tion . All reactions involving water-sensi-
tive chemicals were carried out in flame-dried glassware with
magnetic stirring under Ar. Pyridine was distilled from
calcium hydride. Common solvents [diethyl ether (Et2O),
pentane (P), methanol, ethyl acetate (EtOAc), and CH2Cl2]
were distilled prior to use. Anhydrous NEt3 was distilled from
CaH2. All other reagents and solvents were used as received.
TLC was performed on aluminum sheets (0.2 mm silica gel
60 F254) with detection by UV (254 nm) or by coloration with
ceric ammonium molybdate (CAM). Flash chromatography was
performed on silica gel 60 (230-400 mesh) (ca. 50 g per 1 g of
material to be sparated) with the indicated eluent. 1H and 13C
NMR spectra were recorded at 303 K. Chemical shifts are
reported relative to tetramethylsilane as an internal reference.
Apparent multiplets that occur as a result of accidental
equality of coupling constants to those of magnetically non-
equivalent protons are marked as virtual (virt.). The multi-
plicities of the 13C NMR signal were determined by DEPT
experiments.
P r ep a r a tion of Sta r tin g Ma ter ia ls. The chiral host
compound 4 was prepared as previously described15 and
employed in enantiomerically pure form (>95% ee), Rf ) 0.17
(2:1 Et2O/P). Amides 1a 9a and 59b were synthesized according
to reported procedures. Deuterated amide d5-1a was prepared
analogously to the following procedure. Cyclohexene-1-car-
boxylic acid (600 mg, 4.76 mmol) was heated under reflux in
SOCl2 (5 mL) for 2.5 h. The SOCl2 was removed in vacuo, and
benzene (8 mL) was added. This solution was added to a
solution of d5-aniline (C6D5NH2, 477 µL, 514 mg, 5.24 mmol)
and NEt3 (1.46 mL, 10.47 mmol) in benzene (13 mL) with ice-
cooling. The resulting solution was heated under reflux for 2
h. The solution was diluted with benzene (20 mL) and washed
with 1 M HCl (5 mL), water (5 mL), and brine (5 mL). The
combined aqueous layers were extracted with CH2Cl2 (10 mL).
The organic layers were dried over NaSO4. After filtration, the
solvents were removed in vacuo. Purification of the crude
product by flash chromatography (1:1 Et2O/P) gave a colorless
solid (894 mg, 91%): Rf ) 0.58 (1:1 Et2O/P ); mp 175-177 °C;
1H NMR (250 MHz, CDCl3) δ 1.57-1.76 (m, 4 H), 2.15-2.24
(m, 2 H), 2.30-2.37 (m, 2 H), 6.69-6.74 (m, 1 H), 7.48 (s, b, 1
H); 13C NMR (62.9 MHz, CDCl3) δ 21.3 (CH2), 21.9 (CH2), 24.2
(CH2), 25.5 (CH2), 113.8 (CarD), 133.7 (CdCH), 135.7 (CarD),
145.6 (Car), 150.2 (CdCH), 167.4 (CO); IR (KBr) ν˜ 3270 cm-1
(s), 2934 (s), 2857 (w), 1659 (s), 1630 (s), 1564 (s), 1504 (vs),
1385 (s), 1328 (s), 1302 (s), 1251 (s), 922 (m), 886 (w), 828 (w),
759 (m), 695 (m); MS (70 eV, EI) m/z (%) 206 (64) [M]+, 109
(100) [M - C6D5NH]+, 81 (55) [M - C6D5NHCO]+; HRMS (EI)
calcd for C13H10D5NO 206.14674, found 206.14698.
3
3
H), 6.99 (virt. t, J = J ) 7.7 Hz, 1 H), 7.13-7.17 (m, 2 H),
8.00 (s, b, 1 H); 13C NMR (90.6 MHz, CDCl3) δ 22.7 (CH2), 24.5
(CH2), 25.1 (CH2), 29.7 (CH2), 39.1 (CH), 40.7 (CH), 115.1
(CarH), 123.2 (CarH), 127.3 (CarH), 127.4 (CarH), 136.3 (Car),
172.6 (CO); MS (70 eV, EI) m/z (%) 201 [M+] (72), 172 (22) [M
- CONH]+, 159 (31), 146 (100) [MH - C4H8]+, 130 (17); HRMS
(EI) calcd for C13H15NO 201.11537, found 201.11482. en t-2a :
GC (160 f 200 °C at 0.5 K/min) tR ) 54.84 min; [R]20D ) -49.1
(c 0.23, CHCl3) [40% ee] (lit.12 [R]20 ) -22.9 (c 0.48, CHCl3)).
D
2a : GC (160 f 200 °C at 0.5 K/min) tR ) 55.47 min.
3a a n d en t-3a : 1H NMR (400 MHz, CDCl3) δ 1.23-1.44
(m, 4 H), 1.91-1.97 (m, 2 H), 2.06 (ddd, 3J ) 14.1 Hz, 3J )
3
11.4 Hz, J ) 3.7 Hz, 1 H), 2.38-2.44 (m, 1 H), 2.46-2.51 (m,
3
3
3
1 H), 2.61 (ddd, J ) 14.1 Hz, J ) 10.6 Hz, J ) 3.7 Hz, 1 H),
6.77 (d, 3J ) 7.7 Hz, 1 H), 7.03 (vt, 3J = J ) 7.6 Hz, 1 H),
3
7.16-7.24 (m, 2 H), 8.20 (s, b, 1 H); 13C NMR (90.6 MHz,
CDCl3) δ 25.2 (CH2), 25.2 (CH2), 26.1 (CH2), 28.7 (CH2), 37.9
(CH), 43.2 (CH), 115.1 (CarH), 123.0 (CarH), 124.3 (CarH), 127.3
(CarH), 128.3 (Car), 136.7 (Car), 173.5 (CO); MS (70 eV, EI) m/z
(%) 201 [M+] (100), 172 (29) [M - CONH]+, 159 (18), 146 (15)
[MH - C4H8]+, 130 (35); HRMS (EI) calcd for C13H15NO
201.11537, found 201.11546. en t-3a : GC (160 °C f 200 °C at
0.5 K/min) tR ) 64.34; [R]20 ) -146.1 (c 0.42, CHCl3) [47%
D
ee] (lit.12 [R]20 ) -158.8 (c 0.16, CHCl3)). 3: GC (160 f 200
D
°C at 0.5 K/min) tR ) 63.41 min.
6a ,7,8,9,10,10a -Hexa h yd r oben zo[c][1,6]n a p h th yr id in -
6(5H)-on e (6, en t-6, 7, en t-7).9b Previously unreported ana-
lytical data: Rf ) 0.30 (96:4 EtOAc/MeOH); IR (KBr): ν˜ ) 3045
cm-1 (m), 2934 (s), 2847 (m), 1682 (vs), 1584 (s), 1494 (s), 1381
(s), 1261 (s), 948 (m), 830 (s); HRMS (EI) calcd for C12H14N2O
202.11061, found 202.11058.
1
6 a n d en t-6: H NMR (360 MHz, CD3OD) δ 1.25-1.65 (m,
3
3
3
8 H), 2.77 (virt. q, J = J = J ) 4.8 Hz, 1 H), 3.01 (virt. dt,
3J ) 10.4 Hz, J = J ) 4.8 Hz, 1 H), 6.76 (d, J ) 5.5 Hz, 1
H), 8.12-8.16 (m, 2 H); 13C NMR (90.6 MHz, CD3OD) δ 23.8
(CH2), 25.6 (CH2), 26.5 (CH2), 31.3 (CH2), 37.7 (CH), 41.8 (CH),
111.6 (CarH), 146.7 (Car), 147.0 (Car), 148.8 (CarH), 149.5 (CarH),
174.9 (CO); MS (70 eV, EI) m/z (%) 202 [M+] (44), 173 (35) [M
- CONH]+, 160 (30), 147 (100) [MH - C4H8]+, 118 (15). en t-
6: GC (190 f 215 °C at 0.5 K/min) tR ) 77.74 min. 6: GC (190
f 215 °C at 0.5 K/min) tR ) 78.55 min.
3
3
3
1
7 a n d en t-7: H NMR (360 MHz, CD3OD) δ 1.25-1.65 (m,
3
3
4 H), 1.81-1.91 (m, 2 H), 2.06 (ddd, J ) 13.4 Hz, J ) 11.1
Hz, 3J ) 3.6 Hz, 1 H), 2.24-2.27 (m, 1 H), 2.46-2.51 (m, 1 H),
3
2.59-2.64 (m, 1 H), 6.78 (d, J ) 5.5 Hz, 1 H), 8.18-8.24 (m,
2 H); 13C NMR (90.6 MHz, CD3OD) δ 26.4 (CH2), 26.5 (CH2),
27.4 (CH2), 29.6 (CH2), 38.0 (CH), 44.5 (CH), 111.5 (CarH),
146.1 (CarH), 146.5 (Car), 147.0 (Car), 149.6 (CarH), 175.34 (CO);
MS (70 eV, EI) m/z (%) 202 [M+] (100), 173 (75) [M - CONH]+,
160 (44), 147 (53) [MH - C4H8]+, 131 (34), 118 (16). en t-7: GC
(190 f 215 °C at 0.5 K/min) tR ) 87.58 min. 7: GC (190 f 215
°C at 0.5 K/min) tR ) 87.00.
Gen er a l Ir r a d ia tion P r oced u r e. A solution of amide 1a ,
5, or d5-1a (c ) 4 × 10-3 M) and of the chiral host 4 in toluene
was irradiated in quartz tubes (light source for the reactions
at 30 °C, Rayonet RPR 3000, λ ) 300 nm; light source for the
reactions at -15 and -55 °C, Original Hanau TQ 150, Vycor
glass filter). The solution was not degassed, and the volume
varied from 3 to 30 mL depending on the amount of amide to
be reacted. After complete conversion (2-5 h), the solvent was
removed in vacuo and the residue was purified by flash
chromatography (irradiation products of 1a and d5-1a , 2:1 f
3:1 Et2O/P; irradiation products of 5, 10:1 EtOAc/MeOH). The
enantiomeric excess was determined by chiral GC (2,3-di-O-
methyl-6-O-TBDMS-â-cyclodextrin column). Diastereoisomers
2a /3a and their deuterated derivatives were separated by
reverse-phase HPLC (YMC ODS-A, 250 × 20 mm i.d., gradient
) 30:70 f 60:40 CH3CN/H2O over 30 min), and diastereoiso-
mers 6/7 could not be separated.
6a,7,8,9,10,10a-Hexah ydr o-1,2,3,4-tetr adeu ter o-ph en an t-
r id in -6(5H)-on e (d 4-2a , en t-d 4-2a , d 4-3a , en t-d 4-3a ) a n d
7,8,9,10,10a-P en tah ydr o-1,2,3,4,6a-pen tadeu ter o-ph en an t-
r id in -6(5H)-on e (d 5-2a , en t-d 5-2a , d 5-3a , en t-d 5-3a ): Rf )
0.44 (2:1 Et2O/P); IR (KBr) ν˜ 3181 cm-1 (s), 3058 (s), 2934 (s),
1674 (vs), 1567 (s), 1387 (s), 1286 (m), 840 (m), 656 (m), 626
(m), 598 (m); HRMS (EI) calcd for C13H11D4NO 205.14047,
found 205.14049; calcd for C13H10D5NO 206.14674, found
206.14679.
d 4-2a a n d en t-d 4-2a (92%) a n d d 5-2a a n d en t-d 5-2a (8%):
1H NMR (400 MHz, CDCl3) δ 1.36-1.84 (m, 8 H), 2.80 (virt. q,
3
3
3J = 3J = 3J ) 4.4 Hz, 0.92 H), 2.95 (virt. dt, J ) 10.2 Hz, J
6a ,7,8,9,10,10a -Hexa h yd r op h en a n tr id in -6(5H)-on e (2a ,
=
3J ) 5.1 Hz, 1 H), 8.00 (s, b, 1 H); 13C NMR (90.6 MHz,
CDCl3) δ 22.7 (CH2), 24.4 (CH2), 25.1 (CH2), 29.7 (CH2), 39.0
en t-2a , 3a , en t-3a ).9a,24 Previously unreported analytical
J . Org. Chem, Vol. 68, No. 3, 2003 1115

Bach et al.
1
(CH), 40.7 (CH), 114.7 (t, J (C, D) ) 23.9 Hz, CarD), 122.7 (t,
were solved by direct methods27,28 and refined with standard
difference Fourier techniques.29 All non-hydrogen atoms of the
asymmetric unit were refined with anisotropic thermal dis-
placement parameters. All hydrogen atoms were found in the
difference Fourier maps and refined freely with individual
isotropic thermal displacement parameters, except those
located at the disordered atoms, which were placed in calcu-
lated positions (riding model). Full-matrix least-squares re-
finements were carried out by minimizing Σw(Fo2 - Fc2)2 with
the SHELXL-97 weighting scheme and stopped at maximum
shift/err < 0.001. Crystallographic data (excluding structure
factors) for the structures reported in this paper have been
deposited with the Cambridge Crystallographic Data Centre
as supplementary publication nos. CCDC-191141 (1a ), CCDC-
191139 (2a ), and CCDC-191140 (3a ). Copies of the data can
be obtained free of charge upon application to CCDC, 12 Union
Road, Cambridge CB2 1EZ, UK (fax, (+44)1223-336-033;
e-mail, deposit@ccdc.cam.ac.uk).
1J (C, D) ) 24.1 Hz, CarD), 126.9 (t, 1J (C, D) ) 24.7 Hz, CarD),
126.9 (t, 1J (C, D) ) 25.4 Hz, CarD), 128.4 (Car), 136.2 (Car),
172.7 (CO); MS (70 eV, EI) m/z (%): 206 [d5-M+] (14), 205
[d4-M+] (59), 177 (5) [d5-M - CONH]+, 176 (20) [d4-M -
CONH]+, 163 (30), 151 (24) [d5-MH - C4H8]+, 150 (100)
[d4-MH - C4H8]+, 134 (17). en t-d 4/d 5-2a : GC (160 f 200 °C
at 0.5 K/min) tR ) 53.36 min. d 4/d 5-2a : GC (160 f 200 °C at
0.5 K/min) tR ) 53.79 min.
d 4-3a a n d en t-d 4-3a (33%) a n d d 5-3a a n d en t-d 5-3a
(67%): 1H NMR (500 MHz, CDCl3) δ 1.27-1.45 (m, 4H), 1.94-
3
3
3
1.99 (m, 2H), 2.10 (ddd, J ) 14.0 Hz, J ) 11.5 Hz, J ) 4.3
Hz, 0.33 H), 2.41-2.46 (m, 1 H), 2.51-2.54 (m, 1 H), 2.62-
2.68 (m, 1 H), 7.87 (s, b, 1 H); 13C NMR (90.6 MHz, CDCl3) δ
25.2 (CH2), 26.1 (CH2), 26.2 (CH2), 28.8 (CH2), 37.8 (COCDCH),
37.9 (COCHCH), 42.9 (t, 1J (C, D) ) 18.7 Hz, COCDCH), 43.3
1
1
(COCHCH), 114.6 (t, J (C, D) ) 23.8 Hz, CarD), 122.5 (t, J
1
(C, D) ) 25.1 Hz, CarD), 124.0 (t, J (C, D) ) 24.4 Hz, CarD),
126.9 (t, 1J (C, D) ) 24.4 Hz, CarD), 128.4 (Car), 136.6 (Car),
173.2 (CO); MS (70 eV, EI) m/z (%) 206 [d5-M+] (100), 205
[d4-M+] (48), 177 (23) [d5-M - CONH]+, 176 (21) [d4-M -
CONH]+, 163 (22), 151 (18) [d5-MH - C4H8]+, 150 (18)
[d4-MH - C4H8]+, 135 (49). en t-d 4/d 5-3a : GC (160 f 200 °C
at 0.5 K/min) tR ) 62.49. d 4/d 5-3a : GC (160 f 200 °C at 0.5
K/min) tR ) 61.59 min.
Com p ou n d 3a . C13H15NO, Mr ) 201.26 g/mol, colorless
fragment, size 0.25 × 0.33 × 0.69 mm, triclinic, space group
P1h (No. 2), a ) 6.3016(3) Å, b ) 8.7289(4) Å, c ) 10.1131(5) Å,
R ) 104.554(2)°, â ) 105.514(2)°, γ ) 96.646(3)°, V ) 508.73-
(4) Å3, Z ) 2, Fcalcd ) 1.314 g/cm3, µ(Mo KR) ) 0.083 mm-1
,
F(000) ) 216, T ) 173 K. The cyclohexane part of the molecule
is disordered over two positions (90.5:9.5). Hydrogen atoms of
the major part were found and refined freely. The final model
(213) parameters were refined to wR2 ) 0.1167 based on all
1873 data, R1 ) 0.0425 based on 1600 data with I g 2σ(I),
GOF ) 1.064, and min/max residual electron density ) -0.18/
0.18 e/Å3.
N-Meth ylla cta m en t-2b by N-Meth yla tion of La cta m
en t-2a . To a solution of lactam ent-2a (4.1 mg, 0.02 mmol, 40%
ee) in dry DMF (1 mL) was added KOtBu (4.5 mg, 0.04 mmol).
After the mixture was stirred for 15 min at room temperature,
a solution of MeI (2.5 µL, 5.7 mg, 0.04 mmol) in dry DMF (0.5
mL) was added. Following overnight stirring, the mixture was
diluted with 1 N HCl (10 mL). Extraction with CH2Cl2 (2 ×
10 mL) followed by washing of the organic extracts with water
(2 × 10 mL) and brine (10 mL), drying with Na2SO4, and
removal of the solvent in vacuo gave a light yellow oil. The
crude product was purified by HPLC (YMC ODS-A, 250 × 20
mm i.d.; gradient ) 50:50 f 80:20 CH3CN/H2O over 30 min,
tR ) 16.49 min) to give a colorless oil (2.3 mg, 53%): Rf ) 0.72
(1:1 Et2O/P); 1H NMR (500 MHz, CDCl3) δ 1.21-1.65 (m, 8
Com p u ta tion a l Deta ils. The density functional/Hartree-
Fock hybrid model Becke3LYP20 implemented in GAUSSIAN-
9821 has been used throughout this study together with the
6-31G(d) basis set. All geometries have been fully optimized.
Frequency calculations for 1a , 4, ent-8TS, 8, and ent-8 ensured
that they either correspond to minima (NIMAG 0) or transition
states (NIMAG 1) on the potential energy surface. Complete
details can be found in Supporting Information.
3
H), 2.77 (m, 1 H), 2.89 (m, 1 H), 3.37 (s, 3 H), 6.96 (d, J ) 8.2
Hz, 1 H), 7.03 (vt, 3J = 3J ) 7.4 Hz, 1 H), 7.16 (d, 3J ) 7.4 Hz,
1 H), 7.23-7.26 (m, 1 H). en t-2b: GC (125 f 180 °C at 0.5
Ack n ow led gm en t. We thank Prof. Stefan Grimme
(Universita¨t Mu¨nster) for calculating the specific rota-
tion of compounds 2a and 3a . This work was supported
by the Fonds der Chemischen Industrie and by the
Deutsche Forschungsgemeinschaft. B.G. is grateful to
the Fonds der Chemischen Industrie for a Kekule´
fellowship.
K/min) tR ) 71.64 min; [R]20 ) -48.5 (c 0.11, in CHCl3) [40%
D
ee] (lit.12 [R]20 ) -53.9 (c 0.36, CHCl3)). 2b: GC (125 f 180
D
°C at 0.5 K/min) tR ) 72.45 min. All other spectroscopic data
were in accordance with reported data.9a
N-Meth ylla cta m en t-3b by N-Meth yla tion of La cta m
en t-3a . Lactam ent-3a (5.2 mg, 0.025 mmol, 47% ee) was
converted into the N-methyl compound by the aforementioned
procedure. Chiral GC-analysis of the crude product showed a
9:1 mixture of trans compound ent-3b (47% ee) and cis isomer
2b (46% ee). Compound ent-3b was purified by HPLC (YMC
ODS-A, 250 × 20 mm i.d.; gradient ) 50:50 f 80:20 CH3CN/
H2O in 30 min, tR ) 17.96 min) to give a colorless oil (4.0 mg,
72%): Rf ) 0.75 (1:1 Et2O/P); 1H NMR (360 MHz, CDCl3) δ
Su p p or tin g In for m a tion Ava ila ble: Cartesian coordi-
nates of all calculated structures and crystal data, details of
the structure in the solid state together with PLATON30
graphics, and X-ray crystallographic files for compounds 1a ,
2a , and 3a . This material is available free of charge via the
Internet at http://pubs.acs.org.
3
1.21-1.41 (m, 4 H), 1.88-1.92 (m, 2 H), 2.00 (ddd, J ) 15.0
3
3
J O026602D
Hz, J ) 11.6 Hz, J ) 3.8 Hz, 1 H), 2.37-2.45 (m, 2 H), 2.55
3
3
3
(ddd, J ) 14.1 Hz, J ) 10.7 Hz, J ) 3.8 Hz, 1 H), 3.36 (s, 3
3
3
3
H), 6.98 (d, J ) 7.3 Hz, 1 H), 7.03 (t, J ) J ) 7.5 Hz, 1 H),
7.23-7.27 (m, 2 H). en t-3b: GC (125 f 180 at 0.5 K/min) tR
88.34 min; [R]20D ) -193.9 (c 0.16, CHCl3) [47% ee] (lit.12 [R]20
(24) Masamune, T.; Takasugi, M.; Suginome, H.; Yokohama, M. J .
Org. Chem. 1964, 29, 681-685.
)
D
(25) Hooft, R.; Nonius, B. V. COLLECT, Data Collection Software
for Nonius KappaCCD Devices; Nonius: Delft, The Netherlands, 1998.
(26) Otwinowski, Z.; Minor, W. In Processing of X-ray Diffraction
Data Collected in Oscillation Mode; Carter, C. W., J r., Sweet, R. M.,
Eds.; Academic Press: New York, 1997; Vol. 276, pp 307-326.
(27) Altomare, A.; Cascarano, G.; Giacovazzo, C.; Guagliardi, A.;
Burla, M. C.; Polidori, G.; Camalli, M. J . Appl. Crystallogr. 1994, 27,
435-441.
) -158.8 (c 0.16, CHCl3)). 3b: GC (125 f 180 °C at 0.5 K/min)
tR ) 88.83 min. All other spectroscopic data were in accordance
with reported data.9a
X-r a y Cr ysta llogr a p h y. Gen er a l Rem a r k s. Crystals
suitable for diffraction experiments were selected in perflu-
orinated ether. Preliminary examination and data collection
were carried out at the window of a rotating anode X-ray
generator and graphite monochromated Mo KR radiation (λ
) 0.71073 Å), controlled by the COLLECT software package.25
Collected images were processed using Denzo. Decay effects
were corrected during the scaling procedure.26 The structures
(28) Sheldrick, G. M. SHELXS-86; Universita¨t Go¨ttingen: Go¨ttin-
gen, Germany, 1997.
(29) Sheldrick, G. M. SHELXL-97; Universita¨t Go¨ttingen: Go¨ttin-
gen, Germany, 1998.
(30) Spek, A. L. PLATON, A Multipurpose Crystallographic Tool;
Utrecht University: Utrecht, The Netherlands, 2001.
1116 J . Org. Chem., Vol. 68, No. 3, 2003