Enantioselective [4+2]-Cycloaddition Reaction of a Photochemically Generated o-Quinodimethane: Mechanistic Details, Association Studies, and Pressure Effects

Article abstract of DOI:10.1002/chem.200306049

1,2,3,4-Tetrahydro-2-oxoquinoline-5-aldehyde (2) was prepared from m-aminobenzoic acid and 3-ethoxyacryloyl chloride (4) in 19% overall yield. Compound 2 underwent a photochemically induced [4+2]-cyclo-addition reaction with various dienophiles upon irradiation in toluene solution. The exo product 10 a was obtained with acrylonitrile (9a) as the dienophile, whereas methyl acrylate (9b) and dimethyl fumarate (9c) furnished the endo products 11b and 11c (69-77% yield). The reactions proceeded at -60°C in the presence of the chiral complexing agent 1 (1.2 equiv) with excellent enantioselectivity (91-94% ee). The enantiomeric excess increases in the course of the photocycloaddition as a result of the lower product association to 1. The intermediate (E)-dienol 8 was spectroscopically detected at -196°C in an EPA (diethyl ether/isopentane/ethanol) glass matrix. The association of the substrate 2 to the complexing agent 1 was studied by circular dichroism (CD) titration. The measured association constant (KA) was 589 M^-1 at room temperature (25°C) and normal pressure (0.1 MPa). An increase in pressure led to an increased association. At 400 MPa the measured value of K_A was 703 M^-1. Despite the stronger association the enantioselectivity of the reaction decreased with increasing pressure. At 25°C the enantiomeric excess for the enantioselective reaction 2 + 9a→10a decreased from 68% ee at 0.1 MPa to 58% ee at 350 MPa. This surprising behavior is explained by different activation volumes for the diastereomeric transition states leading to 10a and ent-10a.

Full text of DOI:10.1002/chem.200306049

FULL PAPER
Enantioselective [4+2]-Cycloaddition Reaction of a Photochemically
Generated o-Quinodimethane: Mechanistic Details,Association Studies,and
Pressure Effects
Benjamin Grosch,[a] Caroline N. Orlebar,[b] Eberhardt Herdtweck,[c] Masayuki Kaneda,[d]
Takehiko Wada,[d] Yoshihisa Inoue,*[d] and Thorsten Bach*[a]
Abstract: 1,2,3,4-Tetrahydro-2-oxoqui-
noline-5-aldehyde (2) was prepared
from m-aminobenzoic acid and 3-
ethoxyacryloyl chloride (4) in 19%
overall yield. Compound 2 underwent
a photochemically induced [4+2]-cyclo-
addition reaction with various dieno-
philes upon irradiation in toluene solu-
tion. The exo product 10a was obtained
with acrylonitrile (9a) as the dieno-
phile, whereas methyl acrylate (9b)
and dimethyl fumarate (9c) furnished
the endo products 11b and 11c (69
77% yield). The reactions proceeded
at ꢀ608C in the presence of the chiral
complexing agent 1 (1.2 equiv) with ex-
cellent enantioselectivity (91 94% ee).
The enantiomeric excess increases in
the course of the photocycloaddition as
a result of the lower product associa-
tion to 1. The intermediate (E)-dienol
589mꢀ1 at room temperature (258C)
and normal pressure (0.1 MPa). An in-
crease in pressure led to an increased
association. At 400 MPa the measured
value of KA was 703mꢀ1. Despite the
stronger association the enantioselec-
tivity of the reaction decreased with in-
creasing pressure. At 258C the enantio-
meric excess for the enantioselective
reaction 2 + 9a!10a decreased from
68% ee at 0.1 MPa to 58% ee at
350 MPa. This surprising behavior is
explained by different activation vol-
umes for the diastereomeric transition
states leading to 10a and ent-10a.
8
was spectroscopically detected at
ꢀ1968C in an EPA (diethyl ether/iso-
pentane/ethanol) glass matrix. The as-
sociation of the substrate 2 to the com-
plexing agent 1 was studied by circular
dichroism (CD) titration. The meas-
ured association constant (KA) was
Keywords: Diels Alder reactions ¥
enantioselectivity
¥
high-pressure
chemistry ¥ hydrogen bonds ¥ photo-
chemistry
Introduction
A chemical reaction during which a prochiral substrate is
converted into a chiral product proceeds enantioselectively
if enantiotopic groups or enantiotopic faces of the substrate
are sufficiently differentiated by a chiral entity. The degree
of enantioselectivity is measured by the enantiomeric excess
(ee) of the corresponding product. In conventional thermal
reactions, the entity which delivers the chiral information
has very often been the reagent or the catalyst. In photo-
chemical reactions, the ™reagent∫light has–despite many
attempts–never been successfully employed in a similar
fashion. Results obtained with circularly polarized light
(CPL) as a source of the chiral information have not
reached significant ee values in preparatively useful reac-
tions.[1] There are two alternative approaches towards enan-
tioselective photochemical reactions. One approach is based
on a chiral solid-state environment as the source of chirali-
ty.[2] Homochiral crystals,[3] ionic chiral auxiliary approach,[4]
inclusion complexes,[5] and reactions in zeolites[6] are a few
key concepts to describe this subject in brief. The limitation
[a] Dipl. Chem. B. Grosch, Prof. Dr. T. Bach
Lehrstuhl f¸r Organische Chemie I
Technische Universit‰t M¸nchen
Lichtenbergstrasse 4, 85747 Garching (Germany)
Fax : (+49)89-28913315
E-mail: thorsten.bach@ch.tum.de
[b] M. Chem. C. N. Orlebar
Inorganic Chemistry Laboratory
University of Oxford
South Parks Road, Oxford OX1 3QR (UK)
[c] Dr. E. Herdtweck
Crystal Structure Determination
Lehrstuhl f¸r Anorganische Chemie
Technische Universit‰t M¸nchen
Lichtenbergstrasse 4, 85747 Garching (Germany)
[d] Dr. M. Kaneda, Prof. Dr. T. Wada, Prof. Dr. Y. Inoue
ICORP Entropy Control Project (JST) and
Department of Molecular Chemistry
Osaka University
2-1 Yamada-oka, Suita 565-0871 (Japan)
2179
Chem. Eur. J. 2004, 10, 2179 2189
DOI: 10.1002/chem.200306049
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

FULL PAPER
to intramolecular reactions is a major disadvantage associat-
ed with this approach. A major advantage is the fact that
the choice of substrates is not restricted provided that chiral
crystalline material can be obtained. The other approach is
based on the use of chiral complexing agents (templates) in
solution-phase photochemistry.[1a,7] By definition, chiral
complexing agents interact with prochiral substrates by non-
covalent interactions in a fast, reversible equilibrium. If the
reaction in the template substrate complex is enantioselec-
tive, a high equilibrium constant in favor of complex forma-
tion guarantees high enantioselectivity. Enantioselective cat-
alysis is possible if the reaction proceeds faster in the com-
plex than in the unbound substrate. The binding motif most
commonly encountered in nature for noncovalent associa-
tion is hydrogen bonding.[8] Following this precedence we
Scheme 1. Association behavior of a lactam substrate in the presence of
complexing agent 1.
Despite the simple binding motif, the equilibrium constant
(KA) in favor of the complex (1¥substrate) is sufficiently
high to facilitate reactions which proceed in up to 95% ee at
ꢀ608C.[13c] Since the face differentiation is independent
from the mode of attack, the complexing agent can be ap-
plied to inter- and intramolecular reactions. The study we
describe in this account is concerned with a thermal [4+2]-
cycloaddition reaction which is photochemically initiated.
The results of the cycloaddition studies have been reported
in preliminary form.[14] We now provide details on the prep-
aration of the starting material, the structural elucidation of
the products, on some mechanistic work, and on the pres-
sure dependence of the reaction. The last of these aspects
turned out to be particularly intriguing. We observed an en-
hanced association and a decrease of the enantioselectivity
with increasing pressure.
have devised complexing agents which exhibit hydrogen-
11]
bond donor and hydrogen-bond acceptor sites.[9
The ap-
proach appeared particularly promising for enantiocontrol
in photochemical reactions, since light does normally not in-
terfere with hydrogen bonds. Our most successful complex-
ing agent is lactam 1 (Scheme 1) and its enantiomer. The
compounds are available in multigram quantities[12] and
have been employed for several enantioselective photo-
chemical reactions.[13]
Abstract in German: 1,2,3,4-Tetrahydro-2-oxochinolin-5-al-
dehyd (2) wurde ausgehend von m-Aminobenzoes‰ure und
3-Ethoxyacryloylchlorid (4) in f¸nf Schritten und einer Ge-
samtausbeute von 19% hergestellt. Die Verbindung lie˚ sich
in Toluol als Lˆsungsmittel mit verschiedenen Dienophilen in
einer photochemisch induzierten [4+2]-Cycloaddition umset-
zen (69 77% Ausbeute), wobei als Hauptprodukt mit Acryl-
nitril (9a) das exo-Produkt 10a entstand. Methylacrylat (9b)
und Dimethylfumarat (9c) lieferten die endo-Produkte 11b
and 11c. In Gegenwart des chiralen Komplexierungsreagenz’
1 (1.2 æquiv.) verliefen die Reaktionen mit exzellenter Enan-
tioselektivit‰t (91 94% ee). Der Enantiomeren¸berschu˚
nahm im Verlauf der photochemischen Umsetzung zu, was
man auf die relativ zum Substrat 2 niedrigere Assoziation
des Produkts zur¸ckf¸hren kann. Das intermedi‰r gebildete
(E)-Dienol 8 wurde spektroskopisch in einer EPA (Ether/i-
Pentan/Ethanol) Glasmatrix bei ꢀ1968C nachgewiesen. Die
Assoziation des Substrats 2 an das Komplexierungsreagenz 1
wurde durch CD-Titration genauer untersucht. Die Assozia-
tionskonstante (KA) wurde bei Zimmertemperatur (258C)
und Normaldruck (0.1 MPa) zu 589mꢀ1 bestimmt. Bei hˆher-
em Druck beobachtete man eine verst‰rkte Assoziation und
bei 400 MPa wurde eine Assoziationskonstante von KA =
703mꢀ1 bestimmt. Trotz der st‰rkeren Assoziaion nahm die
Enantioselektivit‰t mit wachsendem Druck ab. Bei 258C
sank der Enantiomeren¸berschu˚ der enantioselektiven Rea-
tion 2 + 9a!10a von 68% ee bei 0.1 MPa auf 58% ee bei
350 MPa. Dieses ¸berraschende Verhalten l‰˚t sich mˆgli-
cherweise durch die unterschiedlichen Aktivierungsvolumina
f¸r die ‹bergangszust‰nde erkl‰ren, die zu 10a und ent-10a
f¸hren.
Results and Discussion
Upon photochemical excitation, o-alkylated aromatic alde-
hydes of type A (Figure 1) undergo a hydrogen abstraction
which leads to two diastereomeric dienols. The correspond-
Figure 1. Structures of an o-substituted benzaldehyde A, the (E)-dienol B
derived from A, and the aldehyde 2 used in this study.
ing Z dienol undergoes rapid tautomerization to the starting
material, whereas (E)-dienol B lives long enough to be trap-
ped by an alkene in a Diels Alder reaction.[15,16] The reac-
tion has proven to be useful for the construction of chiral
tetralins and has been frequently applied in synthesis.[17,18]
Substrate 2 contains an o-alkylbenzaldehyde moiety re-
quired for the photochemically induced Diels Alder reac-
tion and a lactam binding site required for enantioselective
differentiation by complexing agent 1. It was therefore con-
sidered as a useful probe to assess the chances for an enan-
tioselective [4+2] cycloaddition in solution.[19]
2180
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 2179 2189

Enantioselective o-Quinodimethane Photocycloadditions
2179 2189
Preparation of the substrate: Commercially available ethyl
3,3-diethoxypropionate (3) was converted into 3-ethoxy-
acryloyl chloride (4) by
a known two-step procedure
(Scheme 2).[20] Acylation of m-aminobenzoic acid with chlo-
ride 4 led to an anilide (75% yield) which was cyclized
under acidic conditions to an unsaturated quinolone (89%
yield).[21] The previously reported hydrogenation of the qui-
nolone[21] to dihydroquinolone 5 was conducted at 608C
under 2 MPa (20 bar) H2 pressure with Pd/C as the catalyst
(10 mol%). Under these conditions the hydrogenation pro-
ceeded in 75% yield. The activation of acid 5 towards a che-
moselective reduction was achieved with ethoxy chlorofor-
mate and triethylamine in tetrahydrofuran (THF). After fil-
tration the resulting solution was directly reduced with
sodium borohydride. Although the yields achieved in the ac-
tivation/reduction sequence were only moderate, we did not
optimize the reaction any further. The alcohol was eventual-
ly oxidized to the desired aldehyde 2 with pyridinium chloro-
chromate (PCC). The overall yield of substrate 2 was repro-
ducibly 13% in seven steps starting from ester 3.
Scheme 3. Enantioselective photoinduced [4+2] cycloaddition of (E)-
dienol 8 with the dienophiles 9.
1.2, or 2.5 equivalents of compound 1. The other reaction
conditions remained unchanged. Most of the results have
been tabulated elsewhere[14] and will not be repeated in
detail. The key issues are that excellent enantiofacial differ-
entiation was achieved at ꢀ608C with only 1.2 equivalents
of the complexing agent 1 (91 94% ee), and that the selec-
tivity increased further upon increasing the concentration of
the complexing agent (up to 97% ee) and decreased upon
raising the temperature.
Typical figures which illustrate the correlation between
enantiomeric excess and complexing agent concentration
were obtained at ꢀ358C: 35% ee in favor of 10a with
0.5 equivalents of 1, 76% ee with 1.2 equivalents, and
92% ee with 2.5 equivalents. Table 1 gives some insight into
the complexation behavior by looking at the enantiomeric
excess of product 10a in the same reaction after different
conversions. The enantiomeric excess increased with increas-
ing conversion.
Scheme 2. Synthesis of substrate 2.
Stereoselectivity: Irradiation experiments were conducted
with a toluene solution of substrate 2 (ca. 1 mm) in the pres-
ence of an excess of the corresponding dienophile
(50 equiv). Under these conditions, a complete conversion
was achieved at room temperature within 5 10 min or at
ꢀ608C within 30 min. For acrylonitrile (9a, Scheme 3) as
the dienophile, the ratio of product diastereoisomers was
roughly 70:30 in favor of the exo product rac-10a, whereas
methyl acrylate (9b) and dimethyl fumarate (9c) gave pre-
dominantly the endo products rac-11b and rac-11c. The
yields were 60 80% in all reactions we conducted. The exo/
endo selectivity was not significantly affected by the reaction
temperature. A solvent change, however, led to a selectivity
reversal in one example. Thus, irradiation of substrate 2
with acrylonitrile (9a) in acetonitrile gave preferentially the
racemic endo product rac-11a [diastereoisomeric ratio
(d.r.)=92:8, 62% yield].
Table 1. Conversion and enantiomeric excess (ee) in the Diels Alder re-
action of compound 2 with dienophile 9a at ꢀ358C in toluene mediated
by complexing agent 1 (1.2 equiv, cf. Scheme 3).
Conversion[a] [%]
8
57
27
64
80
74
100
76
ee (10a)[b] [%]
[a] The conversion was determined by GC analysis. [b] The ee values
were determined by chiral GC analysis (after derivatisation with TMSCl/
hexamethyldisilazane).
Proof of configuration and discussion: The relative configu-
ration of the products could be determined by 1H NMR
spectroscopy and NOE measurements. The general proce-
dure is exemplified for the two major products obtained
from the reaction of substrate 2 with acrylonitrile (9a).
Figure 2 gives the most prominent NMR data for the two di-
astereoisomers 10a and 11a obtained as [4+2]-cycloaddition
products.
Cycloaddition reactions in the presence of the complexing
agent were performed at 308C, ꢀ358C, or ꢀ608C with 0.5,
2181
Chem. Eur. J. 2004, 10, 2179 2189
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Y. Inoue, T. Bach et al.
FULL PAPER
atoms H-5 and H-3a are in an axial position, whereas H-6
and the cyano group are equatorial.
1
The H NMR and NOE data obtained for the major dia-
stereoisomers of the photocycloaddition products derived
from methyl acrylate (9b) and dimethyl fumarate (9c) were
similar to the data of compound 11a. Most significantly, the
NOE contacts between the hydrogen atoms H-3a and H-5
1
3
were strong, and the H NMR J coupling constants for H-5/
H-6 were low (3.4 Hz). The data support a conformation in
which the hydrogen atom H-5 is axially positioned, and a
configuration in which H-5 and H-6 are cis to each other.
Further evidence for the relative configuration was again
obtained
by
single-crystal
X-ray
crystallography
(Figure 4).[23] In this case, endo diastereoisomer rac-11c
3
Figure 2. Prominent NOE data and relevant JH,H coupling constants ob-
served for the exo and endo diastereoisomers 10a and 11a.
could be crystallized.
3
The J coupling constants between the hydrogen atoms at
C-4, C-5, and C-6 in compound 10a provide evidence for
the fact that these hydrogen atoms are axial/equatorial or
equatorial/equatorial to each other. Hydrogen atom H-5
must consequently reside in an equatorial position. Hydro-
gen atom H-3a is positioned axially, as it has a typically
large axial/axial coupling constant to one hydrogen atom at
C-4. In product 11a, the large coupling constant between H-
4 and H-5 supports a conformation in which H-5 resides in
3
an axial position. Based on the small J value between H-5
and H-6, H-6 must be equatorial. The major difference in
the NOE spectra of 10a and 11a concerns the proximity of
hydrogen atoms H-3a and H-5. There is no NOE contact in
the former case but a strong contact in the latter case. In ad-
dition, both hydrogen atoms at C-4 exhibit NOE contacts to
hydrogen atom H-5 in compound 10a, but only one hydro-
gen atom H-4’ interacts with H-5 in 11a. Indeed, the crystal
structure[22] of compound rac-11a (Figure 3) confirmed the
assignment of the relative configuration and the conforma-
tional analysis derived from the NMR data. Hydrogen
Figure 4. A molecule of compound rac-11c in the crystal structure.[23]
The elucidation of the absolute configuration has already
been discussed in our preliminary paper.[14] It rests on an
enantiomerically pure product whose configuration could be
determined by anomalous X-ray diffraction, and on the
comparison of measured and calculated circular dichroism
(CD) spectra. The data are in line with a Re-face attack rel-
ative to the enol carbon atom which ends up as carbon atom
C-6 in the product. As depicted in Figure 5, the absolute
configuration at C-6 and C-3a can be nicely explained by an
Figure 5. Model to explain the face differentiation in the binary complex
1¥8 of template 1 and (E)-dienol 8.
Figure 3. A molecule of compound rac-11a in the crystal structure.[22]
2182
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 2179 2189

Enantioselective o-Quinodimethane Photocycloadditions
2179 2189
approach of the dienophile to the intermediate (E)-dienol 8
bound to complexing agent 1 from the less shielded face.
There is evidence that the face differentiation provided by
the tetrahydronaphthalene shield is close to perfect,[7,13c] and
the enantiomeric ratio therefore corresponds to the ratio of
bound to unbound substrate. In earlier work we have shown
that an association constant of 500mꢀ1 at room temperature
can account for an almost complete complexation at ꢀ608C
and consequently for almost complete selectivity in favor of
one enantiomer.[13]
An additional benefit of the cycloaddition we have now
studied is the result of the apparently lower association con-
stant of the products 10 and 11 relative to 2 and 8. Unfortu-
nately, we could not access KA for 10 or 11 by CD titration
(vide infra) because the induced CD absorption was covered
by the absorption band of the complexing agent. However,
at low template concentration (c) the titration data for 10a
gave a semiquantitative picture. The flat binding isotherm
c(1)/c(10a) was indicative of a weak complexation which in
turn can be nicely explained by the steric bulk of the tetralin
generated in the [4+2]-cycloaddition reaction. This observa-
tion is in agreement with the increase of the product ee with
increasing conversion (Table 1). The lower association of the
product 10a liberates complexing agent 1. The concentra-
tion of 1 relative to 2 increases leading to an increased asso-
ciation 1¥2 and to a higher enantioselectivity.
Figure 6. Detection of E dienol 8 upon irradiation of aldehyde 2 in an
EPA glass matrix at ꢀ1968C. The UV spectra were recorded after irradi-
ating 2 for the indicated period of time.
ation. The new band reached saturation after 440 s. Simulta-
neously, the aldehyde absorption at 324 nm lost intensity,
and an isosbestic point was observed at 344 nm. The absorp-
tion wavelength recorded for the intermediate is in good
agreement with the spectroscopic properties of other dienols
which have been detected so far.[27]
The regioselectivity of the o-quinodimethane [4+2]-cyclo-
addition reactions is in line with previous results.[16,17] There
is also precedence for the endo selectivity in the reaction of
methyl acrylate and dimethyl fumarate with o-quinodime-
thanes.[17c,24] The unexpected exo selectivity observed with
acrylonitrile in the nonpolar solvent toluene is in agreement
with recent computational results by Freccero et al. which
suggest an exo preference in the gas phase and in nonpolar
solvents.[25,26] In polar solvents like acetonitrile, the endo
product should prevail which it indeed did (vide supra).
Pressure dependence: Reactions which proceed via a transi-
tion state with a highly negative activation volume can be
significantly accelerated by applying high pressure.[28] Many
examples of this phenomenon have been reported for Diels
Alder reactions, in which the rate of reaction significantly
increased at high pressure.[29] In addition, a shift in the dia-
stereoselectivity[29,30] in favor of the product generated via
the more compact transition state, often the endo diaster-
eoisomer, has been observed. The enantioselectivity of a
Lewis acid catalyzed hetero-Diels Alder reaction was en-
hanced at high pressure.[31,32] The endo/exo diastereoselectiv-
ity was also noted to be pressure-dependent in photochemi-
cal [2+2]-cycloaddition[33] and photoinduced [4+2]-cycload-
dition[34] reactions. In studies of sensitized enantiodifferenti-
ating E/Z isomerizations, an influence of pressure on the ef-
ficiency and direction of the chiral induction has been
reported.[35] The pressure dependence of the enantioselectiv-
ity is attributed to nonzero differential activation volumes
(DDV°RꢀS) as described in the relationship (1) between pres-
sure (P) and the relative rate of R and S product formation
(kR/kS):
Dienol intermediate: The relative product configuration at
carbon atoms C-3a and C-6, as determined for compounds
11a, 10c, and 11c by X-ray crystallography and for the
other products by NMR spectroscopy, is in line with a ster-
eospecific attack of the dienophile at the (E)-dienol 8
(Scheme 3). However, in mechanistic studies the postulated
intermediate was initially not observed. The dienol absorp-
tion should occur with a bathochromic shift relative to the
aldehyde absorption at 324 nm. The experiments were per-
formed by irradiating a toluene solution of photosubstrate 2
in a temperate cuvette in the absence of the dienophile and
by subsequent UV spectroscopic detection (<3 s after the ir-
radiation was stopped). The expected dienol band at longer
wavelength was not detected at room temperature, ꢀ508C,
or ꢀ808C. It was speculated that the proton-catalyzed tauto-
merization 8!2 is accelerated by the Br˘nsted acidic
lactam unit. To confirm this speculation, substrate 2 was ir-
radiated in an EPA glass matrix (diethyl ether/isopentane/
ethanol, 5:5:2 v/v, ꢀ1968C) in which the intermolecular
proton transfer is suppressed (Figure 6).
lnðkR=kSÞ ¼ ln½ð100 þ % eeÞ=ð100ꢀ% eeÞꢁ ¼
ð1Þ
ꢀðDDVR°ꢀS=RTÞP þ C
It appeared interesting to study the pressure effect on the
reaction 2!10/11 (Scheme 3). We had hoped for an increase
of selectivity based on a higher association. Indeed, two im-
portant parameters of the reaction are altered by pressure.
Changing pressure should influence the relative rate of the
enantiomer formation as a consequence of different activa-
tion volumes as in the sensitized case. Furthermore, the
Indeed, a strong absorption at 422 nm was observed after
30 s, the intensity of which increased upon prolonged irradi-
2183
Chem. Eur. J. 2004, 10, 2179 2189
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Y. Inoue, T. Bach et al.
FULL PAPER
ground-state equilibrium is also affected by the applied pres-
sure. By NMR titration experiments and microcalorimetric
analyses, it was shown that the face differentiation in the
complex with template 1 is almost perfect, whereas only at
low temperature is the association constant (KA) high
enough to ensure the complexation of most substrate mole-
cules.[13c] The direct relationship between the observed enan-
tioselectivities and KA renders this parameter the primary
lever to increase the efficiency of the chiral induction by
complexing agents such as 1. Since hydrogen-bond-mediated
association processes are typically accompanied by a volume
decrease, the application of high pressures should lead to
better association and concomitantly to higher enantiomeric
excesses. We have therefore studied the effect of pressure
on the enantioselectivity and diastereoselectivity of the in-
termolecular [4+2] cycloaddition of the photochemically
generated intermediate (E)-dienol 8 (Scheme 3).
tion between 1 and 2 was derived to be ꢀ1.24 cm3 molꢀ1
(Figure 8).
lnKA ¼ ꢀðDVA=RTÞP þ lnKA;0
ð2Þ
Since two hydrogen bonds are involved in the complex
formation, this DVA value corresponds to a molar volume
decrease of 0.6 cm3 molꢀ1 per NH¥¥¥O hydrogen bond. This is
considerably lower than the values of ꢀ1.3 to
ꢀ4.6 cm3 molꢀ1 volume change per NH¥¥¥O hydrogen bond
To quantify the pressure dependence of KA, titration ex-
periments were performed at high pressures. Addition of
chiral template 1 induced a strong signal at 325 nm in the
CD spectrum of a toluene solution of prochiral substrate 2.
The binding isotherm at normal pressure (0.1 MPa) was ob-
tained by CD spectral titration with c(1):c(2) ratios in the
range of 0.33 32 (Figure 7, see Experimental Section). From
Figure 8. Determination of the volume decrease DVA of the 1¥2 complex
formation from the pressure dependence of KA.
that have been reported in the
literature.[37,38] The low volume
decrease can be interpreted as
an additional indication of an
entropy-driven complex forma-
tion between 1 and 2.[39] Anoth-
er explanation draws on the
close proximity of the two hy-
drogen bonds which leads to a
DVA value similar to that for
one hydrogen bond.
To elucidate the effect of the
enhanced binding on the effica-
cy of the asymmetric induction,
the photoenolization-initiated
Diels Alder reaction of lactam
2 with acrylonitrile (9a) was ex-
amined at high pressures. In all
instances, the products 10a and
11a were isolated as a mixture
of diastereoisomers in compara-
Figure 7. Circular dichroism spectra of substrate 2 induced by titration with template 1 in toluene at 0.1 MPa
and 298 K, and the derived binding isotherm.
these data, KA at 0.1 MPa was determined as 589mꢀ1 by
nonlinear curve fitting. This value is in good accordance
with the data obtained in NMR titration studies for the
binding of 2-quinolone to host 1 (KA =580mꢀ1).
ble yields (60 65%). The determination of the enantiomeric
excess with chiral GC required derivatization with trime-
thylsilyl chloride (TMSCl)/hexamethyldisilazane. The results
are summarized in Table 2.
Analogous CD titrations of substrate 2 with template 1
were conducted at 100, 200, 300, and 400 MPa by using a
0.2-mL cuvette incorporated into a high-pressure vessel
fitted with CD-inactive diamond windows. The association
constants derived from the binding isotherms[36] by nonlinear
curve fitting indeed increased at elevated pressures up to
KA =703mꢀ1 at 400 MPa. By plotting lnKA against P accord-
ing to Equation (2), the reaction volume DVA of the associa-
Unexpectedly, the enantioselectivity decreased significant-
ly at higher pressures. With 1.2 equivalents of template 1,
the enantiomeric excess was reduced from 56% ee at
0.1 MPa to 50% ee at 200 MPa (Table 2, entries 2 and 5).
This trend was confirmed by the series with 2.4 equivalents
of the chiral complexing agent 1. The exo diastereoisomer
10a was obtained with 68% ee at 0.1 MPa, 62% ee at
200 MPa, and only 58% ee at 350 MPa (Table 2, entries 3, 6,
2184
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 2179 2189

Enantioselective o-Quinodimethane Photocycloadditions
2179 2189
Table 2. Enantioselective Diels Alder reaction of compound
acrylonitrile (9a) (cf. Scheme 3) in the presence of the chiral complexing
agent 1 at various pressures (see Experimental Section).
2 with
Conclusion
In summary, the enantioselective photochemically induced
[4+2]-cycloaddition reaction of compound 2 and the dieno-
philes 9 in the presence of complexing agent 1 has provided
us not only with a new method for the construction of chiral
tetralins but it has also allowed us to gain new insight into
important phenomena involved in template-based photo-
chemistry. The temperature dependence of the enantiomeric
excess is in line with previous work[13] and can be explained
by an increased association at low temperature. The ee
values obtained under optimized conditions (toluene,
ꢀ608C, 2.5 equiv of the complexing agent) are excellent
(96 97% ee). Even with 1.2 equivalents of complexing agent
1, an enantiomeric excess above 90% ee is possible at
ꢀ608C or alternatively with 2.5 equivalents of 1 at ꢀ358C.
These numbers surpass all previous ee values we have been
able to obtain. The reason for the superiority of this photo-
chemical reaction compared to previous systems is most
likely not the higher association of 1¥2, but rather the fact
that the products 10 or 11 of the reaction exhibit a much
lower association to the template. The intermediate (E)-
dienol 8 of the reaction could be spectroscopically detected
in an EPA glass matrix. A study of the enantiomeric excess
in the reaction 2 + 9a!10a relative to external pressure re-
vealed a surprising ee decrease with increasing pressure. Al-
though the association constant increased–as expected–
the determined ee decreased from 68% ee (258C, 0.1 MPa)
to 58% ee (258C, 350 MPa). A possible reason for this be-
havior could be the lower (more negative) activation
Entry
P[a]
[MPa]
Equiv[b]
d.r. (10a/
11a)[c]
ee (10a)[d]
[%]
c(1¥2)/c0(2)
[%]
1
2
3
4
5
6
7
0.1
0.1
0.1
200
200
200
350
64/36
71/29
76/24
67/33
65/35
75/25
67/33
1.2
2.4
56
68
43
64
1.2
2.4
2.4
50
62
58
51
73
74
[a] Pressure applied during irradiation at 298 K with a 500-W ultrahigh-
pressure Hg lamp (Wacom Co.) fitted with a UV-33 filter. [b] Equiva-
lents of the chiral complexing agent 1. [c] The diastereomeric ratio (d.r.)
was determined by HPLC analysis of the crude product mixture. [d] The
ee-values of the major diastereoisomer 10a were determined by chiral
GC (after derivatization with TMSCl/hexamethyldisilazane). Margin of
error ꢂ0.5%.
and 7). The enhanced complexation is evidently overcom-
pensated by a controversial competing effect of the chiral
template 1 on the relative rate of enantiomer formation.
This can be illustrated by calculating the relative amount of
complexed substrate c(1¥2)/c0(2) from KA for the employed
pressures (Table 2, column 6). This value represents the
enantiomeric excess at the beginning of the reaction in the
ideal case of perfect face differentiation.[13c] The observation
that the enantiomeric excesses at 0.1 MPa even exceeded
this number is attributed to the weaker binding of the steri-
cally demanding product to the chiral template. The amount
of unbound host rises in the course of the reaction, and so
does the ee value. This was experimentally proved by ee de-
termination at different levels of conversion (vide infra).
The increased KA values at higher pressure translate to a rel-
ative amount of complexed substrate c(1¥2)/c0(2) that is
about 8% (200 MPa) or 10% (350 MPa) higher, giving an
estimate for the expected ee increase.
volume DV° for the transition state leading to the minor
S
enantiomer ent-10a compared to a higher DV° for the
R
major product.
Experimental Section
In search for an explanation of the unprecedented pres-
sure effect observed, we rely on our model for the face dif-
ferentiation in this and related reactions (Figure 5). The de-
cisive parameter is the differential activation volume
General: All reactions involving water-sensitive chemicals were carried
out in flame-dried glassware with magnetic stirring under Ar. Irradiation
experiments were performed in Merck p. a. solvents (Uvasol). Common
solvents (pentane (P), methanol (MeOH), ethanol (EtOH), ethyl acetate
(EtOAc), tetrahydrofuran (THF), diethyl ether (Et2O), CH2Cl2) were dis-
tilled prior to use. All other reagents and solvents were used as received.
1H and 13C NMR spectra were recorded at 303 K. Chemical sifts are re-
ported relative to tetramethylsilane as an internal reference. Apparent
multiplets which occur as a result of accidental equality of coupling con-
stants to those of magnetically nonequivalent protons are marked as vir-
tual (virt.). NOESY contacts are reported as weak (’), medium (’’), or
strong (’’’). Thin-layer chromatography (TLC) was performed on alumi-
num sheets (0.2-mm silica gel 60F254) with detection by UV (254 nm) or
by coloration with ceric ammonium molybdate (CAM). Flash chromatog-
raphy[40] was performed on silica gel 60 (230 400 mesh; ca. 50 g for 1 g of
material to be separated) with the indicated eluent.
DDV°
as described by Equation (1). The attack from the
RꢀS
shielded face, which leads to the S configuration at the car-
binol carbon atom, is severely hampered by the tetrahydro-
naphthalene shield, and gives rise to a sterically congested
situation. As a consequence, the activation volume DV° is
S
lower (more negative) than DV°R. The more compact transi-
tion state to the minor S enantiomer becomes more favora-
ble at higher pressures and the product ee decreases. Hence,
at higher pressures the enhanced complexation is overridden
by the reduced face differentiation of the chiral template,
which is in turn caused by a high-pressure acceleration of
the attack from the shielded face. This could be a general
phenomenon for chiral templates that induce enantiomeric
excesses by providing face differentiation in a noncovalent
complex.
1,2,3,4-Tetrahydro-2-oxoquinoline-5-carboxylic acid (5): 1,2-Dihydro-2-
oxoquinoline-5-carboxylic acid (6.00 g, 31.7 mmol) was dissolved in aque-
ous NaOH (0.3m, 150 mL). Pd on charcoal (10 mol% Pd, 3.38 g,
3.10 mmol Pd) was added. The mixture was stirred in a high-pressure
vessel for 10 d at 608C in an atmosphere of hydrogen (20 bar). After fil-
tration, aqueous HCl (conc.) was added to the filtrate until no further
precipitate was formed. The white precipitate was filtered, dried under
reduced pressure, and recrystallized from MeOH (250 mL) to give tetra-
hydroquinoline 5 (4.54 g, 75%) as a colorless solid. M.p. 3078C (lit.[21]
309 3118C); 1H NMR (500 MHz, [D6]DMSO): d=2.40 (t, 3J=7.7 Hz,
:
2185
Chem. Eur. J. 2004, 10, 2179 2189
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Y. Inoue, T. Bach et al.
FULL PAPER
2H; C3H2), 3.20 (t, 3J=7.7 Hz, 2H; C4H2), 7.03 (d, 3J=8.0 Hz, 1H;
CHar), 7.21 (dd, 3J=8.0 Hz, 3J=7.2 Hz, 1H; C7H), 7.41 (d, 3J=7.2 Hz,
1H; CHar), 10.13 (s, 1H; NH), 12.54 ppm (s, 1H; OH); 13C NMR
(90.6 MHz, [D6]DMSO): d=22.6 (CH2), 29.9 (CH2), 118.6 (CHar), 123.7
(CHar), 124.8 (Cq,ar), 126.9 (CHar), 130.5 (Cq,ar), 139.5 (Cq,ar), 168.5 (CO),
170.2 ppm (CO); IR (KBr): n˜ =3420 (m, OH), 3200 (m, NH), 3070 (m,
acrylonitrile (9a, 14.3 mmol, 0.76 g, 0.94 mL) was irradiated as described
in the general procedure. From the crude product mixture the diastereo-
meric ratio was determined by HPLC (YMC ODS-A, 250î4.6-mm i.d.;
CH3CN/H2O, 5:95!30:70 in 30 min). After separation by flash chroma-
tography, the exo and endo diastereoisomers 10a and 11a, and the tem-
plate 1/ent-1 (>90%, i.e., 2.4 equiv: >217 mg; 1.2 equiv: >109 mg;
0.5 equiv: >45 mg) were isolated. The enantiomeric excess was deter-
mined upon TMS derivatization: 0.5 mg of each diastereoisomer was sep-
arately dissolved in acetone (1 mL) and treated with hexamethyldisila-
zane/TMSCl (2:1 v/v, 0.1 mL) and 3 drops of pyridine. After shaking for
10 min, the suspensions were filtered over silica gel and, after concentra-
tion to about 0.2 mL, these were analyzed by GC with a chiral stationary
phase (2,3-di-O-methyl-6-O-TBDMS-b-cyclodextrin column, 2008C!
2308C at 0.58Cminꢀ1).
CHar), 2924 (s, CHal), 1720 (vs, C=O), 1682 (vs, NHC=O), 1469 (s, CHal),
ꢀ1
ꢀ
1388 (vs), 1290 (s, C N), 755 cm (w, CHar); MS (70 eV, EI): m/z (%):
191 (100) [M+], 163 (42) [M+ꢀCO], 146 (8) [M+ꢀCOOH], 118 (10).
The spectroscopic data were in accordance with reported data.[21]
(1,2,3,4-Tetrahydro-2-oxo-quinolin-5-yl)methanol (6): NEt3 (1.50 mL,
1.09 g, 10.8 mmol) was added slowly to a suspension of acid 5 (1.91 g,
10.0 mmol) in THF (200 mL) under an atmosphere of argon. The mixture
was cooled to 08C. Ethyl chloroformate (1.03 mL, 1.17 g, 10.8 mmol) was
added dropwise, and the resulting mixture was stirred for 1 h at 08C. The
mixture was then allowed to warm to room temperature and was stirred
for 1 h. The precipitate was filtered, and the filtrate was slowly added to
a solution of NaBH4 (0.76 g, 0.20 mmol) in water (200 mL). The solution
was stirred for 4 h, after which the THF was removed in vacuo. The re-
sidual aqueous solution was acidified to pH 2 with aqueous HCl (1m)
and extracted with CH2Cl2 (5î200 mL). The combined organic layers
were dried over Na2SO4. After filtration the solvent was removed in
vacuo to give alcohol 6 (0.80 g, 42%) as a colorless solid. Rf =0.29
(EtOAc/MeOH, 95:5); m.p. 164 1658C; 1H NMR (500 MHz,
[D6]DMSO): d=2.60 (t, 3J=7.6 Hz, 2H; C3H2), 3.04 (t, 3J=7.6 Hz, 2H;
exo diastereomer 10a: Rf =0.15 (EtOAc); m.p. 230 2368C (decomp);
[a]2D0 =+21.9 (c=0.10, MeOH) [96% ee]; 1H NMR (500 MHz, [D6]ace-
tone): d=2.00 (ddd, 2J=12.0 Hz, 3J=13.6 Hz, 3J=3.5, 1H; C4H), 2.13
3
3
(virt.dt, 2J=12.0 Hz, 3Jꢃ J=4.1 Hz, 1H; C4H), 2.29 (virt.t, 2Jꢃ J=
15.8 Hz, 1H; C3H), 2.52 (dd, J=15.8 Hz, J=4.9 Hz, 1H; C3H), 3.16 (m,
2
3
1H; C3aH), 3.26 (virt.q, 3Jꢃ Jꢃ J=3.4 Hz, 1H; C5H), 4.90 (dd, 3J=
3
3
5.5 Hz, J=2.5 Hz, 1H; C6H), 5.07 (d, J=5.5 Hz, 1H; OH), 6.81 (d, 3J=
3
3
3
3
7.9 Hz, 1H; C9H), 7.03 (d, J=7.5 Hz, 1H; C7H), 7.17 (virt. t, J=7.6 Hz,
1H; C8H), 9.18 ppm (s, 1H; NH); NOESY experiment (500 MHz,
[D6]acetone): H (2.00) H (3.26)’, H (2.13) H (3.16)’, H (2.13) H (3.26)’’,
H (2.29) H (2.52)’’’, H (2.29) H (3.16)’, H (2.52) H (3.16)’, H (3.26) H
(5.07)’, H (5.07) H (7.03)’, H (6.81) H (7.17)’’, H (7.03) H (7.17)’’;
13C NMR (90.6 MHz, [D6]acetone): d=26.2 (C4H2), 30.0 (C3aH), 34.1
(C5H), 37.6 (C3H2), 68.7 (CHOH), 116.8 (C9H), 120.8 (Cq,ar), 123.8 (CN),
126.5 (C7H), 129.5 (C8H), 136.1 (Cq,ar), 138.8 (Cq,ar), 173.4 ppm (CO); IR
(KBr): n˜ =3198 (s, NH), 3020 (s, CHar), 2965 (m, CHal), 2927 (m, CHal),
3
3
C4H2), 4.66 (s, 2H; CH2OH), 6.86 (d, J=7.9 Hz, 1H; CHar), 7.10 (d, J=
7.4 Hz, 1H; CHar), 7.18 (dd, J=7.9 Hz, J=7.4 Hz, 1H; C7H), 12.54 ppm
(s, 1H; OH); 13C NMR (90.6 MHz, [D6]DMSO): d=22.5 (C3H2), 31.4
(C4H2), 63.4 (CH2OH), 116.6 (CHar), 123.8 (Cq,ar), 124.5 (CHar), 128.5
(CHar), 139.4 (Cq,ar), 140.3 (Cq,ar), 173.1 ppm (CO); IR (KBr): n˜ =3200
3
3
(m, NH), 2930 (w, CHal), 1698 (vs, C=O), 1593 (s, C=Car), 1471 (s, CHar),
2242 (m, CꢄN), 1654 (vs, NHCO), 1589 (s, C=Car), 1474 (s, CHar), 1290
ꢀ1
ꢀ1
ꢀ
ꢀ
1221 (m, C N), 1075 (m), 1030 cm (m, C O); MS (70 eV, EI): m/z
(%): 177 (72) [M+], 159 (80) [M+ꢀH2O], 128 (60), 110 (100); HRMS:
m/z calcd for C10H11NO2: 177.07898; found: 177.07887.
ꢀ
ꢀ
(m, C N), 1100 (m, C O), 1054 (m), 791 cm (s, CHar); MS (70 eV, EI):
m/z (%): 228 (70) [M+], 175 (72) [M+ꢀCH2CHCN], 151 (30), 128 (23),
100 (100); HRMS: m/z calcd for C13H12N2O2: 228.08987; found:
228.08998.
1,2,3,4-Tetrahydro-2-oxo-quinoline-5-aldehyde (2): Alcohol
6 (0.30 g,
1.69 mmol) was dissolved in CH2Cl2 (300 mL) and treated with pyridini-
um chlorochromate (1.10 g, 5.08 mmol). The solution was stirred at room
temperature for 8 h, and the solvent was removed in vacuo. The residue
was directly subjected to flash chromatography (EtOAc/P, 9:1) to afford
aldehyde 2 (0.26 mg, 88%) as a slightly yellow solid. Rf =0.40 (EtOAc);
m.p. 200 2018C; 1H NMR (500 MHz, [D6]DMSO): d=2.60 (t, 3J=
7.6 Hz, 2H; C3H2), 3.04 (t, 3J=7.6 Hz; 2H, C4H2), 4.66 (s, 2H; CH2OH),
6.86 (d, 3J=7.9 Hz, 1H; CHar), 7.10 (d, 3J=7.4 Hz, 1H; CHar), 7.18 (dd,
3J=7.9 Hz, 3J=7.4 Hz, 1H; C7H), 12.54 ppm (s, 1H; OH); 13C NMR
(90.6 MHz, [D6]DMSO): d=22.5 (C3H2), 31.4 (C4H2), 63.4 (CH2OH),
116.6 (CHar), 123.8 (cq,ar), 124.5 (CHar), 128.5 (CHar), 139.4 (cq,ar), 140.3
(cq,ar), 173.1 ppm (CO); IR (KBr): n˜ =3145 (m, NH), 2746 (m, CHO),
1734 (s, C=O), 1681 (vs, NHC=O), 1468 (s, CHal), 1388 (vs), 1378 (s),
1061 (m), 801 cmꢀ1 (s, CHar); UV/Vis (MeOH): lmax (e)=326.6 nm
(1764); MS (70 eV, EI): m/z (%): 175 (100) [M+], 146 (28) [M+ꢀCHO],
119 (27), 118 (28), 110 (22); HRMS: m/z calcd for C10H9NO2: 175.06332;
found: 175.06336.
endo diastereomer 11a: Rf =0.28 (EtOAc); m.p. 220 2238C (decomp);
[a]2D0 =+39.3 (c=0.10, MeOH) [55% ee]; 1H NMR (500 MHz, [D6]ace-
3
3
tone): d=1.85 (virt.q, 3Jꢃ Jꢃ J=12.6 Hz, 1H; C4H), 2.15 (m, 1H;
3
C4H), 2.26 (virt.t, 2Jꢃ J=16.1 Hz, 1H; C3H), 2.52 (dd, 2J=16.1 Hz, 3J=
3
3
3
5.2 Hz, 1H; C3H), 2.98 (m, 1H; C3aH), 3.18 (virt.dt, J=12.8 Hz, Jꢃ J=
3.0 Hz, 1H; C5H), 4.92 (dd, 3J=3.0 Hz, 3J=6.4 Hz, 1H; C6H), 5.03 (d,
3J=6.4 Hz, 1H; OH), 6.76 (d, 3J=7.9 Hz, 1H; C9H), 6.99 (d, 3J=7.5 Hz,
3
1H; C7H), 7.12 (virt.t, 3Jꢃ J=7.7 Hz, 1H; C8H), 9.11 ppm (s, 1H; NH);
NOESY experiment (500 MHz, [D6]acetone):
H (2.15) H (2.98)’, H
(2.26) H (2.52)’’, H (2.26) H (2.98)’, H (2.52) H (2.98)’, H (2.98) H
(3.18)’’, H (3.18) H (5.03)’’, H (5.03) H (6.99)’’’, H (6.76) H (7.12)’’, H
(6.99) H (7.12)’’; 13C NMR (90.6 MHz, [D6]acetone): d=28.3 (C4H2), 33.1
(C3aH), 34.9 (C5H), 39.0 (C3H2), 67.2 (CHOH), 116.8 (C9H), 122.2 (Cq,ar),
123.9 (CN), 126.3 (C7H), 129.6 (C8H), 137.6 (Cq,ar), 139.0 (Cq,ar),
173.5 ppm (CO); IR (KBr): n˜ =3202 (m, NH), 3101 (w, CHar), 2924 (m,
CHal), 2243 (m, CꢄN), 1667 (vs, C=O), 1590 (s, C=Car), 1474 (s, CHal),
ꢀ1
ꢀ
1374 (s), 1099 (m, C O), 979 (w), 796 (m, CHar), 745 (w), 696 cm (m,
CHar); MS (70 eV, EI): m/z (%): 228 (85) [M+], 175 (100) [M+
ꢀCH2CHCN], 151 (43), 128 (39), 100 (68); HRMS: m/z calcd for
C13H12N2O2: 228.08987; found: 228.09006.
General irradiation procedure at normal pressure: A solution of alde-
hyde 2 and the chiral template 1/ent-1[12] in toluene (Merck Uvasol,
150 mL) was added to a 200-mL irradiation vessel with a cooled lamp
insert (duran glass) and then degassed by purging with argon for 20 min.
In the case of a low-temperature irradiation the solution was cooled by
an acetone/dry ice bath to the desired temperature. To maintain a tem-
perature of ꢀ608C during the irradiation, the lamp insert was thermostat-
ed to ꢀ608C by an external cryostat. After the solution had equilibrated
to the desired temperature, the dienophile was added whilst carefully
maintaining the argon atmosphere. After stirring for 10 min the solution
was irradiated (light source: Original Hanau TQ150, duran filter) to
complete conversion (10 min at 308C, 20 min at ꢀ158C/ꢀ358C, 30 min at
ꢀ608C; GLC control). The solution was allowed to warm to room tem-
perature, and the solvent was removed in vacuo. The product mixture
was separated by flash chromatography (EtOAc/MeOH, 97:3!95:5).
5-Methoxycarbonyl-6-hydroxy-3a,4,5,6-tetrahydro-1H,3H-1-azaphena-
len-2-one (11b, ent-11b): A solution of aldehyde 2 (50.0 mg, 0.29 mmol),
template 1/ent-1 (2.4 equiv: 0.69 mmol, 241 mg; 1.2 equiv: 0.34 mmol,
121 mg), and freshly distilled acrylic acid methyl ester (9b, 14.3 mmol,
1.22 g, 1.29 mL) was irradiated as described in the general procedure.
From the crude product mixture the diastereomeric ratio was determined
by HPLC (YMC ODS-A, 250î4.6-mm i.d.; CH3CN/H2O, 5:95!30:70 in
30 min). After separation by flash chromatography, the endo diaster-
eoisomer 11b and the template 1/ent-1 (>90%, i.e., 2.4 equiv: >217 mg;
1.2 equiv: >109 mg) were isolated. The enantiomeric excess was deter-
mined upon TMS derivatization by chiral GC as described in the afore-
mentioned procedure. Rf =0.22 (EtOAc/MeOH, 97:3); m.p. 218 2208C
1
5-Cyano-6-hydroxy-3a,4,5,6-tetrahydro-1H,3H-1-azaphenalen-2-one
(decomp); [a]2D0 =+67.2 (c=0.61, MeOH) [97% ee]; H NMR (360 MHz,
(10a, ent-10a,11,
ent-11a):
A
solution of aldehyde
2
(50.0 mg,
CD3OD): d=1.65 (dd, 2J=15.0 Hz, 3J=13.2 Hz, 1H; C4H), 1.99 (m, 1H;
C4H), 2.14 (virt.t, 2J ꢃ 3J=15.7 Hz, 1H; C3H), 2.39 (dd, 2J=15.7 Hz,
3J=5.0 Hz, 1H; C3H), 2.65 (virt.dt, 3J=13.2 Hz, 3J ꢃ 3J=3.2 Hz, 1H;
0.29 mmol), template 1/ent-1 (2.4 equiv: 0.69 mmol, 241 mg; 1.2 equiv:
0.34 mmol, 121 mg; 0.5 equiv: 0.14 mmol, 50 mg), and freshly distilled
2186
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 2179 2189

Enantioselective o-Quinodimethane Photocycloadditions
2179 2189
C5H), 2.79 (m, 1H; C3aH), 4.52, (s, 1H; NH), 4.84 (d, 3J=3.4 Hz, 1H;
C6H), 6.62 (d, 3J=7.8 Hz, 1H; C9H), 6.87 (d, 3J=7.0 Hz, 1H; C7H),
nol (5:5:2 v/v) was degassed by purging with argon for 10 min and then
transferred to fill a quartz cuvette (1 cmî1 cmî3 cm) fitted with a 25-cm
glass tube. The cuvette was slowly immersed into a quartz glass Dewar
with a square bottom filled with liquid nitrogen. Inside the Dewar, the
cuvette was irradiated for 10, 20, 30, 80, 110, 440 s with a Hg high-pres-
sure lamp (Pyrex filter), and a UV spectrum was directly recorded. After
20 s of irradiation the cuvette contents turned yellow. As expected, the
colour disappeared when the matrix was allowed to warm.
3
7.00 ppm (virt.t, 3Jꢃ J=7.4 Hz, 1H; C8H); NOESY experiment
(500 MHz, [D6]acetone): H (1.65) H (1.99)’’’, H (1.65) H (2.14)’, H
(1.99) H (2.65)’, H (1.99) H (2.79)’, H (2.14) H (2.39)’’’, H (2.39) H
(2.79)’’, H (2.65) H (2.79)’’, H (2.65) H (4.84)’’, H (4.84) H (6.87)’’, H
(6.62) H (7.00)’’, H (6.87) H (7.00)’’; 13C NMR (90.6 MHz, CD3OD): d=
26.0 (C4H2), 33.1 (C3aH), 33.1 (C3H2), 47.2 (C5H), 52.6 (OCH3), 68.7
(CHOH), 116.3 (C9H), 124.8 (Cq,ar), 126.3 (C7H), 129.2 (C8H), 138.6
(Cq,ar), 138.7 (Cq,ar), 173.8 (CO), 175.1 ppm (CO); IR (KBr): n˜ =3204 (w,
NH), 3063 (w, CHar), 3008 (w, CHar), 2952 (m, CHal), 1711 (vs, C=O),
1674 (vs, NHC=O), 1592 (s, C=Car), 1472 (s, CHar), 1443 (s), 1072 (m),
943 (w), 767 (m, CHar), 732 (m), 691 cmꢀ1 (w, CHar); MS (70 eV, EI): m/z
(%): 261 (95) [M+], 229 (22), 184 (100), 175 (32), 151 (22), 128 (15);
HRMS: m/z calcd for C14H15NO4: 261.10010; found: 261.09998.
General irradiation procedure at high pressure: A solution of aldehyde 2
(2.0 mm, 0.7 mg, 4.0 mmol), acrylonitrile (80 mm, 10.5 mL, 8.5 mg,
0.16 mmol), and complexing agent 1 (2.4 equiv: 4.8 mm, 9.6 mmol, 3.4 mg;
1.2 equiv: 2.4 mm, 4.8 mmol, 1.7 mg) in toluene was degassed by purging
with argon for 10 min. A 2-mL portion of the solution was transferred
into a 2-mL quartz cuvette incorporated into a high-pressure vessel fitted
with a sapphire window for UV irradiation (Teramecs Co., Kyoto) and
thermostated at 298 K by water circulating through the reactor body. The
vessel was pressurized up to 350 MPa (which is the maximum pressure to
be applied when using the 2-mL cuvette), and the solution was irradiated
for 60 min through a UV-33 filter with a 500-W ultrahigh-pressure Hg
lamp (Wacom Co., Tokyo), which led to complete conversion. The solu-
tion was retrieved from the vessel, and the solvent was removed in
vacuo. From the crude mixture, the d.r. (10a:11a) was determined by re-
versed-phase HPLC. The residue was dissolved in MeOH (0.5 mL) and
separated by preparative TLC (CH2Cl2/MeOH, 92:8). The product was
isolated as a mixture of diastereoisomers 10a and 11a with a combined
yield of 60 65%. The enantiomeric excess (error <ꢂ0.5% ee) was de-
termined after TMS derivatization by gas chromatography on a Supelco
b-DEX 325 column (0.2 mm 1î20 m) using a Shimadzu GC-14AM in-
strument equipped with a C-R6A integrator.
4,5-Di(methoxycarbonyl)-6-hydroxy-3a,4,5,6-tetrahydro-1H,3H-1-aza-
phenalen-2-one (10c, ent-10c,11c, ent-11c): A solution of aldehyde 2
(50.0 mg, 0.29 mmol), template 1/ent-1 (2.4 equiv: 0.69 mmol, 241 mg;
1.2 equiv: 0.34 mmol, 121 mg), and dimethyl fumarate (9c, 14.5 mmol,
2.09 g) was irradiated as described in the general procedure. From the
crude product mixture the diastereomeric ratio was determined by
HPLC (YMC ODS-A, 250î4.6-mm i.d.; CH3CN/H2O, 5:95!50:50 in
30 min). After separation by flash chromatography, the exo and endo dia-
stereoisomers 10c and 11c, and the template 1/ent-1 (>90%, i.e.,
2.4 equiv: >217 mg; 1.2 equiv: >109 mg) were isolated. The enantiomer-
ic excess was determined by 1H NMR spectroscopy with template 1 as
chiral shift reagent: 2 5 mg of each diastereoisomer was dissolved sepa-
rately with 2 equiv of 1 in [D6]acetone. The NH signal split into two sig-
nals (dNH,11c =9.58, dNH,ent-11c =9.93, dNH,10c =9.41, dNH,ent-10c =9.69 ppm),
the integral ratio of which corresponds to the enantiomeric ratio.
Pressure dependence of the association constant
endo diastereomer 11c: Rf =0.17 (EtOAc/MeOH, 97:3); m.p. 209 2158C
CD titration at normal pressure: A solution of template 1 (200 mm,
35.2 mg in 0.5 mL toluene) was added stepwise (5, 5, 5, 5, 10, 10, 20, 20,
50, 50, 100, 100, 100 mL) to 3 mL of a solution of aldehyde 2 (1 mm,
4.4 mg in 25 mL toluene) to prepare solutions with c(1):c(2) ratios of
0.333, 0.667, 1.000, 1.333, 2.000, 2.667, 4.000, 5.333, 8.667, 12.000, 18.667,
25.333, 32.000. After each addition the mixture was transferred into a
quartz cuvette (light path 1 cm), and a CD spectrum (Jasco J-720W spec-
tropolarimeter, bandwidth 5.0 nm, response time 4 s, 20 nmminꢀ1, 4 8 ac-
cumulations) was recorded at 298 K. Each spectrum was corrected for
the CD spectrum of pure 2 by subtraction. The difference spectra were
transformed from the induced molar ellipticity Dq into the induced
molar circular dichroic absorption DDe (d=1 cm, c=0.001 molLꢀ1, i.e.,
the volume change resulting from the addition of the template was ne-
glected). The maximum circular dichroic absorption DDe of the band at
326 nm was plotted against the template substrate ratio c(1):c(2). From
this curve the association constant (KA) of 589mꢀ1 was determined by
nonlinear curve regression.
1
(decomp); [a]2D0 =+93.6 (c=0.55, MeOH) [96% ee]; H NMR (360 MHz,
2
3
2
CD3OD): d=2.46 (dd, J=16.0 Hz, J=14.2 Hz, 1H; C3H), 2.57 (dd, J=
16.0 Hz, 3J=5.0 Hz, 1H; C3H), 3.03 3.09 (m, 2H; C3aH, C4H), 3.22 (dd,
3J=11.8 Hz, 3J=3.4 Hz, 1H; C5H), 3.34 (s, 1H; NH), 3.75 (s, 3H;
OCH3), 3.77 (s, 3H; OCH3), 5.07 (d, 3J=3.4 Hz, 1H; C6H), 6.86 (d, 3J=
7.9 Hz, 1H; C9H), 7.08 (d, 3J=7.5 Hz, 1H; C7H), 7.24 ppm (virt.t, 3Jꢃ
3J=7.7 Hz, 1H; C8H); NOESY experiment (500 MHz, [D6]acetone): H
(2.46) H (2.57)’’’, H (2.46) H (3.03 3.09)’’, H (2.57) H (3.03 3.09)’’, H
(3.03 3.09) H (3.22)’’, H (3.22) H (5.07)’’, H (5.07) H (7.08)’’, H (6.86)
H (7.24)’’, H (7.08) H (7.24)’’; 13C NMR (90.6 MHz, CD3OD): d=34.6
(C3H2), 34.7 (C3aH), 41.8 (C4H), 48.5 (C5H), 50.8 (OCH3), 50.9 (OCH3),
66.6 (CHOH), 114.7 (C9H), 120.4 (Cq,ar), 124.0 (C7H), 127.7 (C8H), 135.9
(Cq,ar), 136.6 (Cq,ar), 170.7 (CO), 172.0 (CO), 174.5 ppm (CO); IR (KBr):
n˜ =3241 (w, NH), 2961 (m, CHal), 1734 (vs, C=O), 1672 (vs, NHCO),
ꢀ
1595 (m, C=Car), 1470 (m, CHar), 1440 (w, CHar), 1361 (s), 1251 (w, C N),
ꢀ1
ꢀ
1210 (m, C O), 1163 (s), 798 (w, CHar), 766 cm (w); MS (70 eV, EI):
m/z (%): 319 (45) [M+], 301 (24) [M+ H2O], 269 (22)
ꢀ
[M+ CH3OH H2O], 241 (100), 228 (20), 210 (38), 183 (28), 163 (21);
HRMS: m/z calcd for C16H17NO6: 319.10559; found: 319.10573.
ꢀ
ꢀ
Determination of the pressure-induced volume reduction of toluene: A
0.15-mL high-pressure quartz cuvette (light path 0.2 cm) filled with a sol-
ution of naphthalene (2.6 mg, 0.02 mmol) in toluene (20 mL) was incor-
porated into a high-pressure vessel fitted with sapphire windows for UV
measurement (Teramecs Co., Kyoto) and thermostated at 298 K by water
circulating through the reactor body. The vessel was pressurized to
100 MPa, 200 MPa, 300 MPa, and 400 MPa. At each pressure a UV spec-
trum was recorded (V-550 spectrometer, 280 340 nm, slit width 1 nm).
Analogous measurements were done with pure toluene. These spectra
were subtracted from the spectra of the naphthalene/toluene solution at
the corresponding pressure. The pressure-induced volume reduction was
exo diastereomer 10c: Rf =0.30 (EtOAc/MeOH, 97:3); m.p. 215 2208C
1
(decomp); [a]2D0 =+24.8 (c=0.10, MeOH) [84% ee]; H NMR (360 MHz,
CD3OD): d=2.31 (dd, 2J=15.8 Hz, 3J=4.8 Hz, 1H; C3H), 2.73 (dd, 2J=
15.8 Hz, 3J=14.8 Hz, 1H; C3H), 3.09 (dd, 3J=8.2 Hz, 3J=7.0 Hz, 1H;
C5H), 3.43 (ddd, 3J=14.8 Hz, 3J=7.5 Hz, 3J=4.8 Hz, 1H; C3aH), 3.49
(dd, 3J=8.2 Hz, 3J=7.5 Hz, 1H; C4H), 3.70 (s, 3H; OCH3), 3.72 (s, 3H;
3
3
OCH3), 4.88 (d, J=7.0 Hz, 1H; C6H), 5.48 (s, 1H; NH), 6.82 (virt.t, Jꢃ
3J=4.5 Hz, 1H; C8H), 7.20 7.22 ppm (m, 2H, C7H; C9H); NOESY ex-
periment (500 MHz, [D6]acetone):
H (2.31) H (2.73)’’’, H (3.09) H
determined at the absorption maximum of 290 295 nm [V/Vꢀ
=
(3.49)’, H (3.09) H (4.88)’, H (4.88) H (7.20 7.22)’’, H (6.82) H (7.20
7.22)’’; 13C NMR (90.6 MHz, CD3OD): d=29.6 (C3aH), 33.0 (C3H2), 41.5
(C4H), 48.0 (C5H), 50.4 (OCH3), 50.7 (OCH3), 68.0 (CHOH), 113.8
(C7H), 120.1 (Cq,ar), 121.0 (C9H), 126.7 (C8H), 136.3 (Cq,ar), 137.7 (Cq,ar),
171.4 (CO), 171.8 (CO), 172.8 ppm (CO); IR (KBr): n˜ =3358 (m, NH),
0.1 MPa
0.9038 (100 MPa), 0.8597 (200 MPa), 0.8247 (300 MPa), 0.8044
(400 MPa)].
CD titrations at high pressures: Analogous titrations were performed by
employing a 0.15-mL high-pressure quartz cuvette (light path 0.2 cm) in-
corporated into a high-pressure vessel fitted with diamond windows for
CD measurements (Teramecs Co., Kyoto) and thermostated at 298 K by
water circulating through the reactor body. The vessel was pressurized to
100 MPa, 200 MPa, 300 MPa, and 400 MPa. Each spectrum was corrected
for the CD spectrum of pure 2 at the corresponding pressure by subtrac-
tion. The difference spectra were transformed from the induced molar el-
lipticity Dq into the induced molar circular dichroic absorption DDe (d=
2954 (m, CHal), 1732 (vs, C=O), 1708 (vs, C=O), 1673 (vs, NHCO), 1591
ꢀ1
ꢀ
ꢀ
(s, C=Car), 1227 (s, C N), 1213 (s, C O), 788 (w, CHar), 743 cm (w); MS
(70 eV, EI): m/z (%): 319 (32) [M+], 301 (28) [M+ꢀH2O], 287 (28) [M+
ꢀCH3OH], 269 (22) [M+ꢀCH3OHꢀH2O], 241 (100), 228 (10), 210 (58),
183 (40); HRMS: m/z calcd for C16H17NO6: 319.1055; found: 319.10511.
Detection of the intermediate dienol in the EPA glass matrix: A solution
of aldehyde 2 (0.1 mm, 50.0 mL, 0.9 mg, 5 mmol) in Et2O/isopentane/etha-
2187
Chem. Eur. J. 2004, 10, 2179 2189
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Y. Inoue, T. Bach et al.
FULL PAPER
0.2 cm). The concentration was corrected for the pressure-induced
volume reduction (vide supra). The volume change resulting from the ad-
dition of the template was neglected. For each pressure the maximum cir-
cular dichroic absorption DDe of the band at 326 nm was plotted against
the template substrate ratio c(1):c(2). From this curve the association
constants KA(P) were determined by nonlinear curve regression to be
[12] T. Bach, H. Bergmann, B. Grosch, K. Harms, E. Herdtweck, Synthe-
sis 2001, 1395 1405.
[13] a) T. Bach, H. Bergmann, K. Harms, Org. Lett. 2001, 3, 601 603;
b) T. Bach, T. Aechtner, B. Neum¸ller, Chem. Commun. 2001, 607
608; c) T. Bach, H. Bergmann, B. Grosch, K. Harms, J. Am. Chem.
Soc. 2002, 124, 7982 7990; d) T. Bach, T. Aechtner, B. Neum¸ller,
Chem. Eur. J. 2002, 8, 2464 2475; e) T. Bach, B. Grosch, T. Strass-
ner, E. Herdtweck, J. Org. Chem. 2003, 68, 1107 1116.
KA(100 MPa)=604mꢀ1
, , ,
KA(200 MPa)=649mꢀ1 KA(300 MPa)=668mꢀ1
and KA(400 MPa)=703mꢀ1
.
[14] B. Grosch, C. Orlebar, Y. Inoue, E. Herdweck, W. Massa, T. Bach,
Angew. Chem. 2003, 115, 3822 3824; Angew. Chem. Int. Ed. 2003,
42, 3693 3696.
[15] N. C. Yang, C. Rivas, J. Am. Chem. Soc. 1961, 83, 2213.
[16] Reviews: a) A. C. Weedon in The Chemistry of Enols (Ed.: Z. Rap-
poport), Wiley, New York, 1990, pp. 591 638; b) P. G. Sammes, Tet-
rahedron 1976, 32, 405 422.
[17] Reviews on o-quinodimethane Diels Alder reactions: a) J. L.
Segura, N. Martin, Chem. Rev. 1999, 99, 3199 3246; b) H. Nemoto,
K. Fukumoto, Tetrahedron 1998, 54, 5425 5464; c) J. L. Charlton,
M. M. Alauddin, Tetrahedron 1987, 43, 2873 2889; d) G. Quinkert,
H. Stark, Angew. Chem. 1983, 95, 651 669; Angew. Chem. Int. Ed.
Engl. 1983, 22, 637 655.
Acknowledgement
This work was supported by the Deutsche Forschungsgemeinschaft and
by the Fonds der Chemischen Industrie (FCI). B.G. is grateful to the FCI
for a Kekulÿ Scholarship and acknowledges further support by the Stu-
dienstiftung des deutschen Volkes and by the Deutscher Akademischer
Austauschdienst (research fellowship in Japan).
[18] Examples: a) B. J. Arnold, S. M. Mellows, P. G. Sammes, J. Chem.
Soc. Perkin Trans. 1 1973, 1266 1270; b) G. Quinkert, W.-D. Weber,
U. Schwartz, H. Stark, H. Baier, G. D¸rner, Liebigs Ann. Chem.
1981, 2335 2371; c) G. Quinkert, U. Schwartz, H. Stark, W.-D.
Weber, F. Adam, H. Baier, G. Frank, G. D¸rner, Liebigs Ann.
Chem. 1982, 1999 2040; d) M. B. Glinski, T. Durst, Can. J. Chem.
1983, 61, 573 575; e) D. I. Macdonald, T. Durst, J. Org. Chem. 1986,
51, 4749 4750; f) J. L. Charlton, G. L. Plourde, K. Koh, A. S. Secco,
Can. J. Chem. 1989, 67, 574 579; g) J. L. Charlton, K. Koh, J. Org.
Chem. 1992, 57, 1514 1516; h) J. L. Charlton, G. L. Plourde, K.
Koh, A. S. Secco, Can. J. Chem. 1989, 67, 574 579; i) K. C. Nico-
laou, D. Gray, Angew. Chem. 2001, 113, 783 785; Angew. Chem.
Int. Ed. 2001, 40, 761 763; j) K. C. Nicolaou, D. Gray, J. Tae,
Angew. Chem. 2001, 113, 3787 3790; Angew. Chem. Int. Ed. 2001,
40, 3675 3678; k) K. C. Nicolaou, D. Gray, J. Tae, Angew. Chem.
2001, 113, 3791 3795; Angew. Chem. Int. Ed. 2001, 40, 3679 3683.
[19] Examples of selectivity control by hydrogen bonds in thermal Diels
Alder reactions: a) A. Wittkopp, P. R. Schreiner, Chem. Eur. J. 2003,
9, 407 414; b) C. Palomo, M. Oiarbide, J. M. Garcia, A. Gonzalez,
A. Lecumberri, A. Linden, J. Am. Chem. Soc. 2002, 124, 10288
10289; c) E. J. Corey, T. W. Lee, Chem. Commun. 2001, 1321 1329;
d) T. Schuster, M. Kurz, M. W. Gˆbel, J. Org. Chem. 2000, 65,
1697 1701; e) A. Robertson, D. Philp, N. Spencer, Tetrahedron
1999, 55, 11365 11384, and refs. cited therein.
[1] Reviews: a) B. Grosch, T. Bach in CRC Handbook of Photochemis-
try and Photobiology, 2nd ed. (Eds.: W. M. Horspool, F. Lenci),
CRC Press, Boca Raton, 2004, pp. 61.1 14; b) S. R. L. Everitt, Y.
Inoue in Molecular and Supramolecular Photochemistry: Organic
Molecular Photochemistry, Vol. 3 (Eds.: V. Ramamurthy, K. S.
Schanze), Dekker, New York, 1999, pp. 71 130; c) Y. Inoue, Chem.
Rev. 1992, 92, 741 770; d) H. Rau, Chem. Rev. 1983, 83, 535 547.
[2] Reviews: a) Y. Ito, Synthesis 1998, 1 32; b) F. Toda, Acc. Chem.
Res. 1995, 28, 480 486.
[3] Review: M. Sakamoto, Chem. Eur. J. 1997, 3, 684 689.
[4] Reviews: a) J. R. Scheffer, Can. J. Chem. 2001, 79, 349 357; b) J. N.
Gamlin, R. Jones, M. Leibovitch, B. Patrick, J. R. Scheffer, J. Trot-
ter, Acc. Chem. Res. 1996, 29, 203 209.
[5] Review: F. Toda, Aust. J. Chem. 2001, 54, 573 582.
[6] Reviews: a) J. Sivaguru, A. Natarajan, L. S. Kaanumalle, J. Shailaja,
S. Uppili, A. Joy, V. Ramamurthy, Acc. Chem. Res. 2003, 36, 509
521; b) A. Joy, V. Ramamurthy, Chem. Eur. J. 2000, 6, 1287 1293;
b) J. C. Scaiano, H. Garcia, Acc. Chem. Res. 1999, 32, 783 793.
[7] B. Grosch, T. Bach in Chiral Photochemistry (Eds.: Y. Inoue, V.
Ramamurthy), Dekker, New York, in press.
[8] G. A. Jeffrey, W. Saenger, Hydrogen Bonding in Biological Struc-
tures, Springer, New York, 1991.
[9] a) T. Bach, H. Bergmann, K. Harms, J. Am. Chem. Soc. 1999, 121,
10650 10651; b) T. Bach, H. Bergmann, K. Harms, Angew. Chem.
2000, 112, 2391 2393; Angew. Chem. Int. Ed. 2000, 39, 2302 2304;
c) T. Bach, H. Bergmann, J. Am. Chem. Soc. 2000, 122, 11525
11526; d) T. Bach, H. Bergmann, H. Brummerhop, W. Lewis, K.
Harms, Chem. Eur. J. 2001. 7, 4512 4521.
[10] Previous work on hydrogen bonding as control element in regio-
and stereoselective photochemical reactions: a) S. McN. Sieburth,
P. V. Joshi, J. Org. Chem. 1993, 58, 1661 1663; b) L. K. Sydnes, K. I.
Hansen, D. L. Oldroyd, A. C. Weedon, E. J˘rgensen, Acta Chem.
Scand. 1993, 47, 916 924; c) C. Zhang, X.-C. Guo, Synth. Commun.
1994, 24, 3157 3165; d) K. Mori, O. Murai, S. Hashimoto, Y. Naka-
mura, Tetrahedron Lett. 1996, 37, 8523 8526; e) M. T. Crimmins,
A. L. Choy, J. Am. Chem. Soc. 1997, 119, 10237 10238; f) S. McN.
Sieburth, K. F. McGee Jr. , T. H. Al-Tel, J. Am. Chem. Soc. 1998,
120, 587 588.
[11] Selected recent examples related to hydrogen bonding in regio- and
stereoselective photochemical reactions: a) W. Adam, K. Peters,
E. M. Peters, V. R. Stegmann, J. Am. Chem. Soc. 2000, 122, 2958
2959; b) A. Yokoyama, K. Mizuno, Org. Lett. 2000, 2, 3457 3459;
c) D. M. Bassani, V. Darcos, S. Mahony, J.-P. Desvergne, J. Am.
Chem. Soc. 2000, 122, 8795 8796; d) A. G. Griesbeck, S. Bondock,
J. Am. Chem. Soc. 2001, 123, 6191 6192; e) W. Adam, V. R. Steg-
mann, R. Synthesis 2001, 1203 1214; f) D. F. Cauble, V. Lynch,
M. J. Krische, J. Org. Chem. 2003, 68, 15 21; g) W. G. Skene, E.
Couzignÿ, J.-M. Lehn, Chem. Eur. J. 2003, 9, 5560 5566; h) N. D.
McClenaghan, C. Absalon, D. M. Bassani, J. Am. Chem. Soc. 2003,
125, 13004 13005.
[20] I. I. Kolodkina, K. V. Levshina, S. I. Sergievskaya, A. I. Kravchenko,
Zh. Org. Khim. 1966, 2, 63 69.
[21] M. Tominaga, E. Yo, H. Ogawa, S. Yamashita, Y. Yabuuchi, K. Na-
kagawa, Chem. Pharm. Bull. 1986, 34, 682 693.
[22] X-ray data for compound rac-11a: C13H12N2O2, Mr =228.25, crystal
size 0.23î0.30î0.38 mm, monoclinic, space group P21/n (No. 14),
a=14.8445(2), b=9.2941(2), c=15.2297(3) ä, b=90.6006(6)8, V=
2101.07(7) ä3, Z=8, 1calcd =1.443 gcmꢀ3, F000 =960, m=0.099 mmꢀ1
,
Nonius k-CCD diffractometer, rotating anode, l=0.71073 ä, T=
173 K, Vmax =25.488, 37843 integrated (h:ꢂ17, k:ꢂ11, l:ꢂ18),
3866 independent and 3338 observed reflections [Io >2s(Io)],
3866 reflections used for refinement, 412 parameters, R1=0.0414
(obs. data), wR2=0.1106 (all data), residual electron density 0.36/
ꢀ0.21 eäꢀ3, direct methods, hydrogen atoms refined. The asymmet-
ric unit cell contains two crystallographically independent molecules.
The crystal is slightly twinned [BASF: 0.0265(6)]. Strong hydrogen
bonds built up double strands along [011]. Both hydroxyl groups
appear to be disordered over two positions (50:50). CCDC 226238
(rac-11a) and CCDC-226239 (rac-11c) contain the supplementary
crystallographic data for this paper. These data can be obtained free
of charge via www.ccdc.cam.ac.uk/conts/retrieving.html (or from the
Cambridge Crystallographic Data Centre, 12, Union Road, Cam-
bridge CB21EZ, UK; fax: (+44)1223-336-033; or deposit@ccdc.
cam.ac.uk).
[23] X-ray data for compound rac-11c: C16H17NO6¥H2O, Mr =337.32,
crystal size 0.23î0.33î0.43 mm, monoclinic, space group P21/n
(No. 14), a=6.9796(1), b=9.8708(1), c=22.2769(2) ä, b=
2188
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 2179 2189

Enantioselective o-Quinodimethane Photocycloadditions
2179 2189
91.5307(3)8, V=1534.20(3) ä3, Z=4, 1calcd =1.460 gcmꢀ3, F000 =712,
m=0.115 mmꢀ1, Nonius k-CCD diffractometer, rotating anode, l=
0.71073 ä, T=173 K, Vmax =25.368, 31668 integrated (h:ꢂ8, k:ꢂ11,
l:ꢂ26), 2816 independent and 2613 observed reflections [Io >
2s(Io)], 2816 reflections used for refinement, 293 parameters, R1=
0.0301 (obs. data), wR2=0.0783 (all data), residual electron density
0.21/ꢀ0.21 eäꢀ3, direct methods, hydrogen atoms refined. Strong hy-
drogen bonds build up a three-dimensional network between the in-
terstitial water and the target molecules.[22b]
[31] L. F. Tietze, C. Ott, K. Gerke, M. Buback, Angew. Chem. 1993, 105,
1536 1538; Angew. Chem. Int. Ed. Engl. 1993, 32, 1485 1486.
[32] Other examples: a) J. Knol, A. Meetama, B. L. Feringa, Tetrahe-
dron: Asymmetry 1995, 6, 1069 1072; b) M. Malinowska, P. Salan-
ski, J.-C. Caille, J. Jurczak, Synthesis 2002, 2707 2710; c) P. Kwiat-
kowski, M. Asztemborska, J.-C. Caille, J. Jurczak, Adv. Synth. Catal.
2003, 345, 506 509.
[33] M. Buback, J. B¸nger, L. F. Tietze, Chem. Ber. 1992, 125, 2577
2582.
[24] T. J. Conolly, T. Durst, Tetrahedron 1997, 53, 15969 15982.
[25] C. Di Valentin, M. Freccero, M. Sarzi-Amadõ, R. Zanaletti, Tetrahe-
dron 2000, 56, 2547 2559.
[26] For solvent effects in the Diels Alder reactions, see: C. Cativiela,
J. I. GarcÏa, J. A. Mayoral, L. Salvatella, Chem. Soc. Rev. 1996,
##25##209 218.
[34] W. S. Chung, N. J. Turro, J. Mertes, J. Mattay, J. Org. Chem. 1989, 54,
4881 4887.
[35] a) Y. Inoue, E. Matsushima, T. Wada, J. Am. Chem. Soc. 1998, 120,
10687 10696; b) M. Kaneda, S. Asaoka, H. Ikeda, T. Mori, T.
Wada, Y. Inoue, Chem. Commun. 2002, 1272 1273.
[36] The CD intensity was corrected for the volume contraction of tolu-
ene at elevated pressure (see Experimental Section). The contrac-
tion of toluene was determined by the pressure dependence of the
UV absorption of naphthalene in toluene and was found to be
almost identical to the known literature value for n-hexane: D. W.
Brazier, G. R. Freeman, Can. J. Chem. 1969, 47, 893 899.
[37] For molar volume change of hydrogen bonds in general, see: a) S. N.
Vinogradov, R. H. Linnell, Hydrogen Bonding, New York, 1971;
b) G. C. Pimentel, A. L. McClellan, The Hydrogen Bond, W. H.
Freeman, San Francisco, 1960.
[38] a) E. Fishman, H. G. Drickamer, J. Chem. Phys. 1956, 24, 548 553;
b) K. Suzuki, M. Tsuchiya, Bull. Chem. Soc. Jpn. 1975, 48, 1701
1704; c) C. Josefiak, G. M. Schneider, J. Phys. Chem. 1979, 83,
2126 2128; d) C. Josefiak, G. M. Schneider, J. Phys. Chem. 1980, 84,
3004 3007; e) R. B. Thompson, J. R. Lakowicz, Biochim. Biophys.
Acta 1984, 790, 87 90.
[39] KA is one order of magnitude greater than KDim of the dimerization
of 2-quinolones (KDim in toluene=40 50mꢀ1) and the association
constant that can be expected for the lactam lactam binding motif
according to the Schneider Jorgenson model. This difference can be
accounted for by toluene molecules solvating substrate and template
that are set free in the course of the complex formation. This effect
would entropically favor the complex formation over the dimeriza-
tion and at the same time reduce the molar volume decrease: a) J.
Sartorius, H.-J. Schneider, Chem. Eur. J. 1996, 2, 1446 1452; b) L. J.
Prins, D. N. Reinhoudt, P. Timmerman, Angew. Chem. 2001, 113,
2446 2492; Angew. Chem. Int. Ed. 2001, 40, 2382 2426; c) R.
Ruloff, U. P. Seelbach, A. E. Merbach, F.-G. Kl‰rner, J. Phys. Org.
Chem. 2002, 15, 189 196.
[27] a) E. F. Zwicker, L. I. Grossweiner, N. C. Yang, J. Am. Chem. Soc.
1963, 85, 2671 2672; b) G. Porter, M. F. Tchir, J. Chem. Soc. A
1971, 3772 3777; c) R. Haag, J. Wirz, Helv. Chim. Acta 1977, 60,
2595 2607; d) J. Gebicki, A. Krantz, J. Chem. Soc. Perkin Trans. 1
1984, 1623 1627; e) W. C. Agosta, R. A. Caldwell, J. Jay, L. J. John-
ston, B. R. Venepalli, J. C. Scaiano, M. Singh, S. Wolff, J. Am. Chem.
Soc. 1987, 109, 3050 3057.
[28] Reviews: a) G. Jenner, Tetrahedron 2002, 58, 5185 5202; b) F.
Wurche, F.-G. Kl‰rner in High Pressure Chemistry, (Eds.: R. van El-
dik, F.-G. Kl‰rner), Wiley-VCH, Weinheim, 2002, pp. 41 96; c) F.-
G. Kl‰rner, F. Wurche, J. Prakt. Chem. 2000, 342, 609 636; d) N.
Katagiri, M. Yamaguchi, C. Kanenko, Heterocycles 1998, 48, 1023
1043; e) R. Winter, J. Jonas, High Pressure Chemistry, Biochemistry
and Material Science, Kluwer, Dordrecht, 1993; f) G. Jenner, Angew.
Chem. 1975, 87, 186 194; Angew. Chem. Int. Ed. Engl. 1975, 14,
137 143.
[29] Reviews: a) L. F. Tietze, P. L. Steck in High Pressure Chemistry
(Eds.: R. van Eldik, F.-G. Kl‰rner), Wiley-VCH, Weinheim, 2002,
pp. 239 283; b) L. F. Tietze, G. Kettschau, J. A. Gewert, A. Schuf-
fenhauer, Curr. Org. Chem. 1998, 2, 19 62; c) G. Jenner, Tetrahe-
dron 1997, 53, 2669 2694; d) N. S. Isaacs, Tetrahedron 1991, 47,
8463 8497; e) F.-G. Kl‰rner, Chem. Unserer Zeit 1989, 23, 53 63;
f) R. van Eldik, T. Asano, W. J. le Noble, Chem. Rev. 1989, 89, 549
688.
[30] Examples: a) L. F. Tietze, T. H¸bsch, C. Ott, G. Kuchta, M. Buback,
Liebigs Ann. 1995, 1 7; b) Y. Araki, T. Konoike, J. Org. Chem.
1997, 62, 5299 5309; c) L. F. Tietze, M. Henrich, A. Niklaus, M.
Buback, Chem. Eur. J. 1999, 5, 297 304, and references therein;
d) K. Goldenstein, T. Fendert, P. Proksch, E. Winterfeldt, Tetrahe-
dron 2000, 56, 4173 4185; e) H. Al-Badri, N. Collignon, J. Maddalu-
no, S. Masson, Chem. Commun. 2002, 1191 1192.
[40] W. C. Still, M. Kahn, A. J. Mitra, J. Org. Chem. 1978, 43, 2923 2925.
Received: December 19, 2003 [F6049]
2189
Chem. Eur. J. 2004, 10, 2179 2189
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim