Chiral photochemistry within a confined space: Diastereoselective photorearrangements of a tropolone and a cyclohexadienone included in a synthetic cavitand

Article abstract of DOI:10.1039/b900017h

The value of a supramolecular assembly to enforce a closer interaction between a chiral auxiliary and a reaction center has been established using photoreactions of tropolone and cyclohexadienone derivatives. Two probe molecules utilized to establish the concept undergo 4 e- electrocyclization and oxa-di-π-methane rearrangement from excited singlet and triplet state, respectively. The chiral auxiliaries investigated here has no/little effect in acetonitrile solution during phototransformations of the probe molecules to yield products with new chiral centers. On the other hand the same ones are able to enforce diastereoselectivities to the extent of ～30% when the reactions occur within the restricted space of a capsule made up of a synthetic cavitand commonly known as octa acid. Extensive NMR studies have been utilized to characterize the guest-host supramolecular structures. The results presented here should be of value in the overall understanding of chiral induction in photochemical reactions. The Royal Society of Chemistry 2009.

Full text of DOI:10.1039/b900017h

View Article Online / Journal Homepage / Table of Contents for this issue
This paper is published as part of a Dalton Transactions themed issue on:
Supramolecular photochemistry
Guest Editors Michael D. Ward and Julia Weinstein
University of Sheffield, UK
Published in issue 20, 2009 of Dalton Transactions
Images reproduced with permission of U. Nickel (left) and Shunichi Fukuzumi (right)
Papers published in this issue include:
Perspective: Sensitised luminescence in lanthanide containing arrays and d–f hybrids
Stephen Faulkner, Louise S. Natrajan, William S. Perry and Daniel Sykes, Dalton Trans., 2009,
DOI: 10.1039/b902006c
[Pt(mesBIAN)(tda)]: A near-infrared emitter and singlet oxygen sensitizer
Aaron A. Rachford, Fei Hua, Christopher J. Adams and Felix N. Castellano, Dalton Trans., 2009,
DOI: 10.1039/b818177b
Self-assembly of double-decker cages induced by coordination of perylene bisimide with a trimeric
Zn porphyrin: study of the electron transfer dynamics between the two photoactive components
Ana I. Oliva, Barbara Ventura, Frank Würthner, Amaya Camara-Campos, Christopher A. Hunter,
Pablo Ballester and Lucia Flamigni, Dalton Trans., 2009, DOI: 10.1039/b819496c
Synthesis, photophysical and electrochemical characterization of phthalocyanine-based poly(p-
phenylenevinylene) oligomers
Juan-José Cid, Christian Ehli, Carmen Atienza-Castellanos, Andreas Gouloumis, Eva-María Maya,
Purificación Vázquez, Tomás Torres and Dirk M. Guldi, Dalton Trans., 2009, DOI:
10.1039/b818772j
Luminescence, electrochemistry and host–guest properties of dinuclear platinum(II) terpyridyl
complexes of sulfur-containing bridging ligands
Rowena Pui-Ling Tang, Keith Man-Chung Wong, Nianyong Zhu and Vivian Wing-Wah Yam,
Dalton Trans., 2009, DOI: 10.1039/b821264c
Visit the Dalton Transactions website for more cutting-edge inorganic and organometallic research
www.rsc.org/dalton

PAPER
www.rsc.org/dalton | Dalton Transactions
Chiral photochemistry within a confined space: diastereoselective
photorearrangements of a tropolone and a cyclohexadienone included
in a synthetic cavitand†
Arun Kumar Sundaresan,a Lakshmi S. Kaanumalle,a Corinne L. D. Gibb,b Bruce C. Gibbb and V. Ramamurthy*a
Received 6th January 2009, Accepted 25th February 2009
First published as an Advance Article on the web 31st March 2009
DOI: 10.1039/b900017h
The value of a supramolecular assembly to enforce a closer interaction between a chiral auxiliary and a
reaction center has been established using photoreactions of tropolone and cyclohexadienone
derivatives. Two probe molecules utilized to establish the concept undergo 4 e- electrocyclization and
oxa-di-p-methane rearrangement from excited singlet and triplet state, respectively. The chiral
auxiliaries investigated here has no/little effect in acetonitrile solution during phototransformations of
the probe molecules to yield products with new chiral centers. On the other hand the same ones are able
to enforce diastereoselectivities to the extent of ~30% when the reactions occur within the restricted
space of a capsule made up of a synthetic cavitand commonly known as octa acid. Extensive NMR
studies have been utilized to characterize the guest–host supramolecular structures. The results
presented here should be of value in the overall understanding of chiral induction in photochemical
reactions.
and shape of this host is compared in Scheme 1 with that of
Introduction
cyclodextrin (CD) and cucurbituril (CB) that are more known to
Several strategies have been employed to effect chiral induction
photochemists. The main difference between OA and the other
in asymmetric photoreactions.1 Photoreactions have been carried
two hosts is that OA forms a capsule surrounding one or two
guest molecules.16–18 Results presented here must be viewed from
the context that even with cyclodextrin the best chiral induction
in photochemical reactions is less than 30% (diastereomeric
excess via chiral auxiliary strategy and enantiomeric excess via
chiral inductors strategy where CD is the chiral inductor).19–24
Usefulness of cucurbiturils in chiral photochemistry has not yet
been established. Further, unlike thermal reactions, rules of chiral
induction in photochemical reactions are yet to be understood and
results presented here, we believe, would help construct a model
that would enable one predict in due course the factors that need
to be controlled to achieve chiral induction in photoreactions. The
current study consists of NMR characterization and photochem-
ical investigation of host–guest complexes.
The two systems, tropolone ether 1 and cyclohexadienone 2, we
have examined in the context of chiral induction within OA capsule
are represented in Scheme 2. In both cases a chiral auxiliary is
covalently placed far removed from the reaction site. The chiral
auxiliary thus placed has very little effect during photoreaction
in solution (diastereomeric excess, de < 3%). Thus the chemistry
in solution was used as the standard to compare the results in
guest–OA complex.
Chiral products from achiral tropolone ether arise via a 4-
electron disrotatory cyclization process. Two of the three p-
bonds of tropolone undergo concerted cyclization reaction to
yield the bicyclic products 3 and 3a shown in Scheme 3. The
cyclization can occur along two different faces of the planar
molecule and can follow paths A or B in Scheme 3. The
newly formed four membered ring is either below (path A) or
above (path B) the plane of the five membered ring and yields
photoproducts 3 and 3a, respectively, in Scheme 3. As illustrated
out in solid phase, solution phase and in organized assemblies.
The chiral sources employed include circularly polarized light,
chiral sensitizers, chiral solvents, chiral substituents, chiral host–
guest assemblies and chiral crystalline environments.2,3 In each
method, chiral information is transferred to a prochiral or a
racemic substrate through non-covalent interactions, allowing
chiral amplification. In recent years we have successfully applied
two techniques, chiral inductor and chiral auxiliary, to achieve
chiral induction in a variety of photoreactions within zeolites.4,5
In the chiral inductor approach, both the chiral molecule and the
substrate are confined to the same reaction site of a supramolecular
host. In the chiral auxiliary approach, the substrate and the chiral
molecule are covalently linked. When the chiral auxiliary is linked
to the substrate, they are forced to be in close contact within the
confined space of a zeolite and hence this method is generally
more effective than the chiral inductor method. Photochemical
reactions using both these methods are reported by several groups
in crystalline state as well as in solid-state host–guest assemblies
with considerable success.6–9
In this manuscript, we highlight results of our studies in host–
guest assemblies in aqueous medium using photocylization of
a tropolone ether and oxa-di-p-methane rearrangement of a
cyclohexadienone.10–13 The host used is a synthetically available
cavitand commonly known as octa acid (OA).14,15 The size
aDepartment of Chemistry, University of Miami, Coral Gables, FL 33124,
USA. E-mail: murthy1@miami.edu
bDepartment of Chemistry, University of New Orleans, New Orleans, LA
70148, USA
† Electronic supplementary information (ESI) available: Additional NMR
spectra. See DOI: 10.1039/b900017h
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 4003–4011 | 4003
©

Scheme 1 Pictorial representation of the structures of octa acid, cucurbit[8]uril and g-cyclodextrin showing the relative dimensions of the host cavity.
Scheme 2 Structures of the substrates investigated and their primary
photoproducts.
in Scheme 4, 3,3,5-trimethyl-4-oxocyclohexa-1,5-dienecarboxylic
acid undergoes oxa-di-p-methane rearrangement to yield bicyclic
products 4 and 4a The reaction proceeding via triplet state
involves formation of 1,5-diradical intermediate. Analogous to
the cyclization reaction of tropolone ethers, the photoproduct can
cyclize along either face of the five-membered cyclic intermediate
Scheme 3 The 4-electron disrotatory cyclization reaction of tropolone
ether yielding two chiral photoproducts.
to yield a pair of enantiomers. In the current examples (1 and 2) we
envision that differences in the interaction between chiral auxiliary
and the two faces of tropolone or cyclohexadienone would either
impede one reaction pathway or facilitate one pathway and thus
yield excess of one of the photoproducts. The belief that within
a confined space one could force stronger interaction between a
4004 | Dalton Trans., 2009, 4003–4011
This journal is
The Royal Society of Chemistry 2009
©

Scheme 4 Mechanism of the oxa-di-p-methane rearrangement of a dienone.
chiral auxiliary and the reaction site than in a free solution has
recognize that in theory the signals from H-b, c, d, f and g should
each appear as two signals because in the capsule the northern
hemisphere and the southern hemisphere of the host are not the
same (they are binding different parts of the guest). Furthermore,
because the guest is chiral, H-a and e are diastereomeric in the
complex; they are non-equivalent. Thus, for example, the two
H-e protons in one hemisphere are non-equivalent, and both
are different from the non-equivalent H-e signals in the other
hemisphere. Thus, for both H-e and H-a there could be four signals.
Sometimes, because of coincidence, these splitting patterns may
not be apparent.
prompted the current investigation.
Results and discussion
NMR characterization of reactant–OA complexes
Octa acid and tropolone ether 1 and cyclohexadienone derivative
2 formed complexes in 2 : 1 ratio in buffered aqueous solution.
1H NMR spectrum, 1D selective total correlation spectroscopy
(TOCSY) NMR spectra, nuclear Overhauser effect spectroscopy
(NOESY) NMR spectrum and diffusion ordered spectroscopy
1D selective TOCSY NMR spectra provided in Fig. 2 and 3 for
OA and OA-1 complex helped in identifying the host signals. In
the TOCSY NMR of free OA, (Fig. 2), selective irradiation of each
1
(DOSY) NMR spectrum of 1@OA2 are shown in Fig. 2–6. H
NMR, COSY and NOESY NMR spectra of OA complex of 2 are
provided in Fig. 8–11. A detailed discussion of the NMR analyses
of OA and tropolone ether 1 complex is provided below followed
by a brief discussion on OA and cyclohexadienone derivative 2.
In 1H NMR spectrum of 1@OA2 shown in Fig. 1, first, consider
the signals of the host. The aromatic signals of OA were split after
addition of 1. Before embarking on the analyses of the spectra we
Fig. 1 1H NMR spectrum (500 MHz, D2O, 2.5 ¥ 10-3 M 1, 5 ¥ 10-3
M
OA, 5 ¥ 10-2 M sodium tetraborate, 298 K) of 1@OA2 complex. Octa
acid signals are labelled (a–j) and guest aliphatic signals are labelled (1–5).
Other guest signals are indicated with *. Residual water signal is indicated
with ᭹.
Fig. 2 Top: 1D selective TOCSY NMR spectra of octa acid (500 MHz,
DMSO-d6, 0.12 s mixing time). Irradiated signals are marked with an
arrow. Refer structure of OA in Scheme 1 for the correlations.
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 4003–4011 | 4005
©

two sets of signals corresponding to Hc and He originate from two
different octa acid molecules.
Aliphatic group signals of the guest in 1@OA2 are shifted below
0 ppm due to the shielding effect of the aromatic rings of the
host. The 3.8 ppm and 2 ppm upfield shift of CH3 signals ‘1’
and ‘3’, respectively, is an illustration of the greater, differential
shielding of the guest at the tapering end of the capsule than its
wider ‘equatorial region. TOCSY NMR spectrum shown in Fig. 4
revealed all the aliphatic signals of the guest. Olefinic signals of
the guest were observed at 5.1 ppm, 4 ppm 3.8 ppm and 3.2 ppm
and are marked with * in Fig. 1.
Fig. 3 1D selective TOCSY NMR spectra of 1@OA2 (500 MHz, D2O,
2.5 ¥ 10-3 M 1, 5 ¥10-3 M OA, 5 ¥ 10-2 M sodium tetraborate 0.12 s mixing
time). Irradiated signals are marked with an arrow. Correlations are based
on TOCSY NMR of octa acid shown in Fig. 2.
Fig. 4 1D selective TOCSY NMR spectrum (top) and 1D NMR
spectrum (bottom) of 1@OA2 (500 MHz, D2O, 2.5 ¥ 10-3 M 1, 5 ¥10-3
M
OA, 5 ¥ 10-2 M sodium tetraborate, 0.12 s mixing time). Irradiated signal
is marked with an arrow.
host signal was carried out. The resultant NMR spectrum showed
the signals of the covalently linked hydrogens of the host. For
example, irradiation at 7.8 ppm (signal a in Fig. 2, corresponding
to Ha of OA) showed one additional signal and was assigned as
Hd, which is present in the same ring as Ha. In the TOCSY NMR
of OA–1 complex shown in Fig. 3, irradiation of the Hd signal at
6.55 ppm showed the Ha signal at 7.6 ppm. Similarly, Hc and He
hydrogen atoms of OA that are in the same ring show TOCSY
correlations and were identified in the NMR of the complex.
Signals of Hb and Hf atoms did not show TOCSY correlations,
although they are in the same ring. By comparing the TOCSY
NMR of free OA and the TOCSY NMR of OA–1 complex, all
the signals of the host were assigned. In the NMR spectrum of the
complex, the diastereotopic Ha signals of the host were spilt into
two, and so were Hb, Hc, and Hf signals. He signals were split into
two doublets in the NMR spectrum of the complex. The additional
observation in the 1D TOCSY correlation of the complex 1@OA2
was that while Hc and He signals showed TOCSY correlations,
no TOCSY correlation was observed between the two Hc signals
(c1 and c2 in Fig. 3). Similarly, two sets of He signals (marked e1
and e2 in Fig. 3) did not show through bond connectivity. That
is, the two sets of hydrogens Hc1 and He1, and Hc2 and He2 were
not covalently linked. Generally, split signals originate from the
two host molecules that constitute the complex. Since they bind
to different portions of the guest, their signals are affected by the
guest in different manner and are observed as split signals. Thus
based on the TOCSY information, it can be confirmed that the
Additionally, 2D NOESY NMR of the complex was performed
to obtain through space interactions between the host and the
guest. NOESY spectrum showed strong correlations between the
more shielded (terminal) methyl hydrogens of the guest (signal 1
in Fig. 5) and only one Hg signal of the host. In addition to host–
guest correlations, intramolecular NOE interactions between the
guests’ signals was evident in the NOESY analysis. In the 1H NMR
spectrum, the methylene signals (signals ‘2’ and ‘5’ in Fig. 1 and 4)
were split into two. NOE correlation between the two methylene
hydrogens of signal 2 was evident in the NOESY spectrum (marked
with dotted lines in Fig. 5).
Diffusion coefficient of the complex was determined by DOSY
NMR analysis (Fig. 6) and was calculated to be 1.26 ¥ 10-10
m2s-1. In the DOSY NMR spectrum of the complex shown in
Fig. 6, both the octa acid signals and the guest signals posses
identical diffusion coefficients confirming that the complex is
stable during the experiment time period. For comparison, the
diffusion coefficient of the water signal (at 4.8 ppm) is much
higher than that of the complex. Since the diffusion coefficient is
a function of the size of the molecule, presence of free (unbound)
guest will be revealed in the DOSY spectrum, as free guest will
have a greater rate of diffusion. Hence, based on DOSY NMR
analysis, it can be concluded that the complex of octa acid and 1
is stable and the complex is either a 2 : 1 or 2 : 2 complex.25–27 Due
to solubility problems, proper titration experiments could not be
carried out both in the case of 1 and 2. However, when the guest
concentration is greater than 0.5 equiv. with respect to the host,
4006 | Dalton Trans., 2009, 4003–4011
This journal is
The Royal Society of Chemistry 2009
©

only signals seen were due to the bound guest. This alone is not
sufficient to generate a Job-plot.
Similar analyses of the complex between OA and 2 revealed
it to be a 2 : 1 complex. The stoichiometry of the complex was
established by integration of the NMR spectrum shown in Fig. 7.
DOSY NMR of 5 ¥ 10-3 M solution of the 2-OA complex showed
it to have a diffusion coefficient of 1.26 ¥ 10-10 m2s-1, comparable to
complex 1@OA2. The stability of the complex was also established
by way of identical diffusion coefficients for both the guest and
the host.
Fig. 7 1H NMR spectrum (500 MHz, D2O, 298 K, 2.5 ¥ 10-3 M 2, 5 ¥
10-3 M OA, 5 ¥ 10-2 M sodium tetraborate) of 2@OA2 complex. Host
signals are labelled (a–j). Guest methyl signals are numbered and the
residual water signal is marked with ᭹.
Fig. 5 NOESY NMR spectrum (500 MHz, D2O, 298 K, 2.5 ¥ 10-3
M
1, 5 ¥ 10-3 M OA, 5 ¥ 10-2 M sodium tetraborate, 0.5 s mixing time) of
the 1@OA2 complex. Correlation between guest methyl and host’s Hg, and
correlation between the two guest H2 signals are highlighted.
Similar to the OA complex with 1, signals of both the host and
the guest were affected upon binding. All the aromatic signals
of OA were split upon binding of the guest. Analysis of the
guest signals of OA bound 2 was performed based on COSY and
NOESY NMR along with 1H NMR chemical shifts. Compound
2 possesses six methyl groups, of which two are geminal methyl
groups. Consequently, only four signals were observed when the
NMR spectrum of 2 was recorded in CDCl3. 1D TOCSY analysis
of the guest signals (spectrum not shown) did not show any
correlations, and were not useful in assigning the guest signals.
COSY spectrum (Fig. 9) helped us to assign the methyl signals of
the guest. 1H NMR of 2@OA2 complex showed six distinct signals
of equal intensity, one for each CH3 group (signals 1–6 in Fig. 7).
Not only were they distinct, chemical shift difference between the
first and the last signal was 2.8 ppm. Even two sets of geminal
dimethyl signals are well separated after binding to OA. Of the
two geminal methyl groups, chemical shift of the most shielded
methyl group (signal 1 in Fig. 7) is -2.9 ppm, whereas chemical
shift of the comparatively less shielded group is -1.4 ppm. Such
a large difference in chemical shift is attributed to the presence of
only one methyl group at the deepest core of the cavity.
Fig. 6 DOSY NMR spectrum (500 MHz, D2O, 2.5 ¥ 10-3 M 1, 5 ¥ 10-3
M
OA, 5 ¥ 10-2 M sodium tetraborate) of 1@OA2 complex. The calculated
diffusion coefficient is 1.26 ¥ 10-10 m2s-1.
In the NOESY NMR spectrum of the complex (Fig. 8 and 9),
correlations existed between all six methyl groups and Hg of the
host. The three most shielded methyl groups (signals 1, 2 and 3)
showed NOE interactions exclusively with Hg, while other three
methyl groups of the guest showed NOE interactions with Hd and
He of the host as well. The more shielded signals are expected to
be deeper in the cavity and NOE interaction specifically with Hg
of the host confirms that the guest exists in the orientation shown
in Fig. 10. Note that only one of the geminal methyl groups is
the clear solution turns turbid and gradual precipitation of free
guest is visibly clear when excess guest is added. This suggested
that the host–guest ratio is 2 : 1 and not 2 : 2. We were not able
to generate a Job-plot to obtain the exact host–guest ratio for the
following reason. Job-plot is an effective method of depicting the
host–guest binding behaviour when there is a gradual change in
the NMR shift with gradual addition of the host. Since the guest
was not water soluble no signal due to free guest was evident and
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 4003–4011 | 4007
©

Fig. 8 Partial COSY NMR spectrum (300 MHz, D2O, 2.5 ¥ 10-3 M 2,
5 ¥ 10-3 M OA, 5 ¥ 10-2 M sodium tetraborate) of 2@OA2. Correlations
between the guest signals are marked with dotted lines.
Fig. 10 Top: partial NOESY NMR spectrum (500 MHz, D2O, 2.5 ¥
10-3 M 2, 5 ¥ 10-3 M OA, 5 ¥ 10-2 M sodium tetraborate, 0.5 s mixing
time) of 2@OA2. Distinct correlations between the guest methyl signals
‘1–6¢ and Hg of the host, and guest methyl signals ‘4–6¢ and Hg of the
host can be seen and is shown in the bottom figure. Intramolecular NOE
interaction between signal ‘6¢ and the olefin signal at 5.7 ppm is connected
with dotted lines.
Fig. 9 NOESY NMR spectrum (500 MHz, D2O, 2.5 ¥ 10-3 M 2,
5 ¥ 10-3 M OA, 5 ¥ 10-2 M sodium tetraborate, 0.5 s mixing time) of
2@OA2. Correlations between the guest methyl signals and host signals
are highlighted.
directed to the center of the cavity, based on the chemical shifts of
the two methyl groups of -2.9 ppm and -1.4 ppm.
Based on the above analyses of the NMR spectra, we visualize
the OA complexes of 1 and 2 to have structures shown in Fig. 11.
In the excited state time scale (excited singlet and triplet) these
structures are expected to be stationary and not undergo assembly
and disassembly of the capsular complex. In addition, guest
molecules within the OA capsule are not expected to have as much
freedom when they are free in the absence of OA in solution.
Fig. 11 Cartoon representation of the 2 : 1 complexes formed between
host OA and guests 1 (left) and 2 (right), respectively.
Photochemistry of guest@OA2 complexes
The cyclization reaction of tropolone ether 1 was carried out in
acetonitrile and as OA complex at 298 and 278 K for 5 min. The
photoproducts were extracted with chloroform and analyzed by
gas chromatography using a Supelco b-dex 325 chiral column.
When irradiated in acetonitrile, the reaction was racemic—
presence of the chiral auxiliary did not have any effect on the
reaction. Octa acid bound 1 yielded photoproducts 3 and 3a
with de of 17% (A) at 298 K and 35% (A) at 278 K (the first
photoproduct signal in GC analysis was arbitrarily assigned as A
and the second signal was assigned B; no attempt was made to
ascertain the absolute configuration of the photoproducts). This
result confirmed that although the substrate tropolone and the
chiral moiety were bound to two different OA molecules that make
4008 | Dalton Trans., 2009, 4003–4011
This journal is
The Royal Society of Chemistry 2009
©

up the capsule, chiral induction is possible. It also suggested that
although the OA capsule is large enough to accommodate the
entire guest and allow movement of the bound guest, yet is small
enough to restrict the motion to force interaction between the
chiral auxiliary and reactant parts of the molecule.
time scale between non-complexed and complexed guest and host.
Irradiation of octa acid solutions of either guests 5 or 6 showed
no selectivity in the cyclization reaction. The observed negative
results could be due to the fact that either 5 and 6 did not
complex to OA or the complexes were extremely weak. It is quite
likely that photoreactions result from non-complexed undissolved
microparticulates in solution. The absence of any NMR signals
for either complexed or free guest molecules are consistent
with this possibility. Since the photochemical results were not
promising no variable temperature NMR experiments were carried
out.
Results obtained with dienone 2 were also encouraging. As
mentioned in Scheme 4, the oxa-di-p-methane rearrangement of 2
occurs to yield two photoproducts. In homogeneous solvents like
acetonitrile, the reaction yields equal amounts of both the products
even in the presence of the chiral auxiliary because of the large
freedom of rotation present in such conditions. Photochemical
reaction was carried out by irradiating 2 in acetonitrile and
as OA complex at 278 K and 298 K. When 2 was irradiated
as acetonitrile solution and the products were analyzed by gas
chromatography using a Supelco b-dex 325 chiral column; the
formation of 3% excess of one diastereoisomer was observed.
When 2@OA2 complex was irradiated and the photoproducts
analyzed, formation of 36% excess of one diastereoisomer at 278 K
(17% at 298 K) was observed.
Both in the case of tropolone ether 1 and dienone 2, chiral
auxiliary does not affect the reaction when it is carried out
in a homogeneous solution. This is because of the rotational
freedom experienced by the chiral auxiliary in isotropic media.
When the molecule is included in a supramolecular host, the
role of the host molecule is to rigidly contain the guest inside
the cavity. Depending on the orientation of the guest, and the
proximity of the chiral auxiliary to the substrate, preference
for cyclization reaction along one face of the molecule can be
achieved.
Another tropolone ether studied was the (-)-menthol ether
of tropolone, 7 (Scheme 5).28 Based on 1H NMR spectra of
the complex, the formation of a 2 : 1 complex between octa
1
acid and 7 was confirmed. H NMR spectrum of the octa acid
complex of 7 recorded at 298 K is shown in Fig. 12 (top).
Characteristic high field shifted guest signals and multiply split
host signals were readily observed. Guest signals were generally
broad and were spread from 0.5 ppm to -1.5 ppm and between
-2.5 to -3.2 ppm. However, unlike most other guests studied
with OA, intensities of the guest signals of 7 were low and
signals were broad. Even upon cooling to 278 K (Fig. 12,
bottom), the characteristics of the guest signals did not improve.
Broadening of the octa acid signals at 278 K when compared
to the spectrum recorded at 298 K suggested that upon cooling,
the conformational dynamics of the guest inside the cavity and
interconversion between several different guest orientations were
slower. The above data is also consistent with the postulate that
the kinetic stability of 7@OA complex is poor compared with
1@OA2.
Negative results of some consequence
Based on the results of 1@OA2 complex, tropolone derivatives
bearing a few more chiral auxiliaries (Scheme 5) were synthesized
and their binding properties with octa acid and their photochem-
istry were explored. NMR titration studies with 5 and octa acid
confirmed that 5 did not form a strong complex with octa acid.
While bound guest signals were observed at 2 : 0.25 (host–guest)
ratio, guest signals were not evident in the NMR spectrum of
the complex in 2 : 1 ratio (see ESI for NMR spectra).† It may
be a result of rapid exchange of the guest or weak binding,
but even the free (unbound) guest signals were not recorded.
Similarly, the ester derivative 6 did not form a strong complex
with octa acid. Only the octa acid signals were observed in the
titration NMR spectra of octa acid–6 mixture (see ESI for NMR
spectra).† Although at host–guest ratio 2 : 0.25 and 2 : 0.5 equiv.,
broadening of Hc and Hd signals of octa acid was observed, guest
signals were not seen. NOESY NMR of the complex recorded
at 278 K (5 ◦C) also showed no host–guest interactions. Thus,
the NMR data is consistent with either a weak complexation
between 5 and 6 with OA or/and rapid equilibrium in the NMR
Fig. 12 1H NMR spectra of 7@OA2 (500 MHz, D2O, 2.5 ¥ 10-3 M 7, 5 ¥
10-3 M OA, 5 ¥ 10-2 M sodium tetraborate) recorded at 298 K (top) and at
278 K (bottom). Intensity of the high field region (spectrum on the right)
is greater than the intensity of the low-field region (spectrum on the left).
Photochemical behaviour of 7 bound to octa acid was similar
to the reaction in acetonitrile. Racemic mixture of photoproducts
was observed. Unlike 5 and 6 that did not bind at all, 7 formed
a relatively stable complex with octa acid. In spite of that
the photoreaction yielded a racemic product mixture. Results
observed with four tropolone ethers (1, 5, 6 and 7) suggest that
for the chiral auxiliary to be effective the host–complex must be
Scheme 5 Structures of tropolone derivatives appended to chiral ester
(5), chiral amide (6) and ether (7).
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 4003–4011 | 4009
©

strong and the structural flexibility of the guest must be restricted
than in solution.
Synthesis of N-((R)-3-methylbutan-2-yl)-2-((1E,3Z,5Z)-7-
oxocyclohepta-1,3,5-trienyloxy)ethanamide (5)/(R)-3-
methylbutan-2-yl 2-((1E,3Z,5Z)-7-oxocyclohepta-1,3,5-
trienyloxy)ethanoate (6)
Conclusion
Tropolone (1 equiv.) was dissolved in acetonitrile and anhydrous
NaY (activated at 500 ◦C for 5 h) and anhydrous potassium
carbonate were added. The mixture was heated to reflux and stirred
at reflux for 60 min. Intermediate 8/9 was added in one portion to
the reaction mixture. After 16 h at reflux, the insoluble content was
filtered and the acetonitrile was distilled. Pure 5/6 was obtained
by column purification (silica gel, hexanes and chloroform).
Moderately diastereoselective photoreactions of 1 and 2 were
observed when they are complexed to a synthetic cavitand. The
observed results suggest that factors like substituent effects and
weak interactions between octa acid and the guests influenced
the outcome of the reaction. Strong binding nature of the two
guests to OA is likely to play a major role in the observed
selectivity. But exact mechanism of chirality transfer is not clear.
Presently, further study is required to establish the role of the
chiral auxiliary and the influence of OA in the diastereoselective
reactions. The results presented here establish that OA has
the ability to confine the guest in its cavity and consequently,
diastereoselective reactions can be carried out using octa acid as
host.
Experimental
Synthesis of compounds 1, and 2, and their photoproducts were
reported earlier.28 The procedure for the synthesis of compounds
5, 6 and 7 is given below. The sequence of reactions followed is
outlined in Schemes 6 and 7.
Scheme 7 Sequence of reactions used to synthesize reactant 7.
Synthesis of tropolone-4-methylbenzenesulfonate (10)
Tropolone (1 equiv.) was dissolved in minimum amount of pyridine
(1 g in 6–8 mL) and recrystallized para-toluenesylfonyl chloride
(TsCl, 1.2 equiv.) was added to the solution. The turbid solution
was stirred for 24 h at room temperature and filtered. The
precipitate was dried and used in the next step as obtained.
Synthesis of (2E,4Z,6Z)-2-((1R,2S,5R)-2-isopropyl-5-
methylcyclohexyloxy)cyclohepta-2,4,6-trienone (7)
(-)-Menthol (1 equiv.) was dissolved in THF and gradually
potassium tertbutoxide (1.1 equiv.) was added with stirring. After
15 min, 10 (1 equiv.) was added as solid in one portion to the
reaction slurry. The clear solution turned dark brown instantly.
It was stirred at room temperature for 24 h. The reaction was
quenched by adding ice-cold water. THF was distilled and the
organics were extracted with three portions of chloroform. The
organic layer was washed with brine, dried with anhydrous sodium
sulfate, and concentrated. Flash column chromatography (silica
gel, hexanes, ethyl acetate) was employed to isolate pure 7.
Scheme 6 Sequence of reactions used to synthesize reactants 5 and 6.
Synthesis of (R)-2-bromo-N-(3-methylbutan-2-yl)ethanamide (8)
and (R)-3-methylbutan-2-yl-2-bromoethanoate (9)
A nitrogen bubbled solution of N-(3-dimethylaminopropyl)-N¢-
ethylcarbodiimide hydrochloride (EDC·HCl) in methylene chlo-
ride was stirred with bromoacetic acid (1 equiv.) and either (R)-
3-methyl-2-butanol or (R)-3-methylbutan-2-amine (1.2 equiv.) at
room temperature for 24 h. The organic layer was washed with
water, aqueous Na2CO3 solution and brine, dried with anhydrous
sodium sulfate and concentrated. The crude ester was used as such
for the next step.
Protocol for preparation and photolysis of complex, and isolation
and analysis of products
General methods. All NMR experiments were carried out
with Bruker Avance Spectrometers at 298 K unless mentioned
otherwise. 1D TOCSY spectra were recorded with 0.12 s mixing
time; NOESY spectra were recorded with 0.5 s mixing time.
Photolysis of the substrates were carried out using a 450 watt
4010 | Dalton Trans., 2009, 4003–4011
This journal is
The Royal Society of Chemistry 2009
©

◦
◦
medium pressure Hg lamp. All solutions were bubbled with
nitrogen for 15 min prior to photolysis.
7. Initial Temperature 120 C for 5 min; ramp 5 C min-1 till
200 ◦C. Retention time of photoproducts 25.6 min and 26.2 min.
Procedure for NMR titration experiments. Octa acid solution
(0.6 mL, 10-3 M OA in buffered D2O) was taken in an NMR sample
tube and its 1H NMR spectrum was recorded. Stock solution of the
guest was prepared in CD3CN. Aliquots of the guest solution were
added to the octa acid solution, usually in 2.5 mL increments, and
the NMR spectrum of the complex was recorded after addition.
Acknowledgements
VR thanks the National Science Foundation, USA (CHE-0213042
and CHE-0531802) and BCG the National Institute of Health
(RO1 GM074031) for financial support.
Preparation of the complex with guests 1, 2, 5 and 7. A stock
solution of the guest was prepared in CD3CN. An aliquot of the
stock solution was transferred to a sample vial and the solvent
was evaporated in a stream of air with gentle warming. Octa acid
solution (in buffered D2O) was added to the guest and the vial
was sonicated for ~5 min. The solution was then transferred to
an NMR tube and sealed. NMR spectrum of the complex was
recorded. The NMR sample solution was used for irradiation
experiments.
References
1 Y. Inoue and V. Ramamurthy, Chiral Photochemistry, Marcel Dekker,
New York, 2004.
2 Y. Inoue, Chem. Rev., 1992, 92, 741–770.
3 H. Rau, Chem. Rev., 1983, 83, 535–547.
4 J. Sivaguru, A. Natarajan, L. S. Kaanumalle, J. Shailaja, S. Uppili, A.
Joy and V. Ramamurthy, Acc. Chem. Res., 2003, 36, 509–521.
5 J. Sivaguru, J. Shailaja and V. Ramamurthy, in Handbook Of Zeolite
Science and Technology, ed. S. M. Auerbach, K. A. Carrado and P. K.
Dutta, Marcel Dekker, New York, 2003, pp. 515–589.
6 J. R. Scheffer and W. Xia, Top. Curr. Chem., 2005, 254, 233–262.
7 J. R. Scheffer, in Chiral Photochemistry, ed. Y. Inoue and V. Rama-
murthy, Marcel Dekker, New York, 2004, pp. 463–483.
8 F. Toda, K. Tanaka and H. Miyamoto, in Understanding and Manipu-
lating Excited-State Processes, ed. V. Ramamurthy and K. S. Schanze,
Marcel Dekker, New York, 2001, vol. 8, pp. 385–425.
9 K. Tanaka and F. Toda, in Organic Solid State Reactions, ed. F. Toda,
Kluwer, New York, 2002, pp. 109–157.
Photolysis experiments. A solution of octa acid–guest complex
was sealed in a Pyrex test tube with a septum and was purged with
nitrogen for 15 min prior to irradiation. The solution was cooled
in an ice-water bath held at 278 K and equilibrated for 15 min. A
medium pressure Hg lamp was used to irradiate the complexes.
All tropolone ether solutions (1, 5 and 7) were irradiated for
2–4 min while solution of 2 was irradiated for 30–45 min. The
irradiation period was determined by comparison with irradiation
of acetonitrile solutions of the guests. After irradiation, octa acid
solutions were extracted with two portions of chloroform. The
chloroform layer was analyzed by GC using a chiral column
(Supelco b-DEX 325 capillary column). GC conditions and
retention times of the photoproducts for each substrate are given
below. The enantiomeric excess (ee) of a reaction is given by the
ratio (A - B/A + B) ¥ 100, where A refers to the area under the
first peak of the product and B refers to the area under the second
peak.
10 W. G. Dauben, K. Koch, S. L. Smith and O. L. Chapman, J. Am. Chem.
Soc., 1963, 85, 2616–2621.
11 O. L. Chapman and D. J. Pasto, J. Am. Chem. Soc., 1960, 82, 3642–
3648.
12 H. Hart, P. M. Collins and A. J. Waring, J. Am. Chem. Soc., 1966, 88,
1005–1011.
13 H. Hart and R. K. J. Murray, J. Org. Chem., 1970, 35, 1535–1542.
14 C. L. D. Gibb and B. C. Gibb, J. Am. Chem. Soc., 2004, 126, 11408–
11409.
15 B. C. Gibb, in Organic Nanostructures, ed. J. L. Atwood and J. W. Steed,
VCH Verlag GmbH & Co., Weinheim, 2008, pp. 291–304.
16 L. S. Kaanumalle, C. L. D. Gibb, B. C. Gibb and V. Ramamurthy, J.
Am. Chem. Soc., 2005, 127, 3674–3675.
17 L. S. Kaanumalle, C. L. D. Gibb, B. C. Gibb and V. Ramamurthy, J.
Am. Chem. Soc., 2004, 126, 14366–14367.
18 A. Natarajan, L. S. Kaanumalle, S. Jockusch, C. L. D. Gibb, B. C.
Gibb, N. J. Turro and V. Ramamurthy, J. Am. Chem. Soc., 2007, 129,
4132–4133.
Analyses conditions for the diastereomeric excess of the pho-
toproducts of 1, 2, 5 and 7. GC: HP 5890 series fitted with
a chiral column (Supelco -dex 325 [phase, non-bonded; 25%
2,3-di-O-methyl-6-O-TBDMS–cyclodextrin in SPB-20 poly(20%
phenyl/80% dimethyl siloxane)]).
1. Initial Temperature 100 ◦C for 2 min; initial ramp 0.5 ◦C
min-1 till 125 ◦C; hold at 125 ◦C for 2 min, ramp at 10 ◦C min-1 till
200 ◦C. Retention time of photoproducts 39.9 min and 40.5 min.
2. Initial Temperature 120 ◦C for 2 min; initial ramp 5 ◦C min-1
till 160 ◦C; hold at 160 ◦C for 5 min, ramp at 2 ◦C min-1 till 200 ◦C.
Retention time of photoproducts 30.6 min and 31.1 min.
5. Initial Temperature 60 ◦C for 1 min; initial ramp 2 ◦C
min-1 till 200 ◦C; hold at 200 ◦C for 10 min. Retention time of
photoproducts 46.6 min and 48 min.
19 S. Koodanjeri and V. Ramamurthy, Tetrahedron Lett., 2002, 43, 9229–
9232.
20 S. Koodanjeri, A. Joy and V. Ramamurthy, Tetrahedron, 2000, 56, 7003–
7009.
21 J. Shailaja, S. Karthikeyan and V. Ramamurthy, Tetrahedron Lett.,
2002, 43, 9335–9339.
22 A. Nakamura and Y. Inoue, J. Am. Chem. Soc., 2003, 125, 966–972.
23 C. Yang, G. Fukuhara, A. Nakamura, Y. Origane, K. Fujita, D.-Q.
Yuan, T. Mori, T. Wada and Y. Inoue, J. Photochem. Photobiol., A,
2005, 173, 375–383.
24 K. Vizvardi, K. Desmet, I. Luyten, P. Sandra, G. Hoornaert and E. V.
der Eycken, Org. Lett., 2001, 3, 1173–1175.
25 C. L. D. Gibb, A. K. Sundaresan, V. Ramamurthy and B. C. Gibb, J.
Am. Chem. Soc., 2008, 130, 4069–4080.
26 A. K. Sundaresan and V. Ramamurthy, Photochem. Photobiol. Sci.,
2008, 7, 1555–1564.
27 A. K. Sundaresan and V. Ramamurthy, Org. Lett., 2007, 9, 3575–3578.
28 A. Joy, L. S. Kaanumalle and V. Ramamurthy, Org. Biomol. Chem.,
2005, 3, 3045–3053.
6. Initial Temperature 80 ◦C for 1 min; initial ram◦p 3 ◦C min-1
till 160 ◦C; hold at 160 ◦C for 30 min, ramp at 2 C min-1 till
200 ◦C. Retention time of photoproducts 53.5 min and 54.3 min.
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 4003–4011 | 4011
©