Synthesis of diastereomeric bicyclo[3.3.1]nonane dibenzoyl esters and study of their chiroptical properties

Article abstract of DOI:10.1002/chir.22075

A synthesis of diastereomeric bicyclic dibenzoyl esters derived from enantiomerically pure (1S,5S)-bicyclo[3.3.1]nonane-2,6-dione was accomplished. Molecules containing two benzoyl chromophores with different configuration in the bicyclic framework were obtained. Chiroptical properties of the synthesized enantiomerically pure molecules were studied. Diastereomeric esters exhibited exciton couplets in the circular dichroism (CD) spectra because of transannular interaction between non-conjugated benzoate chromophores. The conformational effects and solvent impact on the exciton coupling were examined by CD spectroscopy. Theoretical computation of the CD spectra of diastereomers correctly reproduced the sign of the exciton couplets in the studied molecules, however, no major solvent dielectric constant influence and conformational effects per se on the exciton coupling was observed. Copyright

Full text of DOI:10.1002/chir.22075

CHIRALITY (2012)
Synthesis of Diastereomeric Bicyclo[3.3.1]nonane Dibenzoyl Esters
and Study of Their Chiroptical Properties
¯
ˇ
*
*
G. BAGDŽIUNAS, E. BUTKUS, AND S. STONCIUS
Department of Organic Chemistry, Vilnius University, Vilnius, Lithuania
ABSTRACT
A synthesis of diastereomeric bicyclic dibenzoyl esters derived from enantio-
merically pure (1S,5S)-bicyclo[3.3.1]nonane-2,6-dione was accomplished. Molecules containing
two benzoyl chromophores with different configuration in the bicyclic framework were obtained.
Chiroptical properties of the synthesized enantiomerically pure molecules were studied. Diaste-
reomeric esters exhibited exciton couplets in the circular dichroism (CD) spectra because of
transannular interaction between non-conjugated benzoate chromophores. The conformational
effects and solvent impact on the exciton coupling were examined by CD spectroscopy. Theoretical
computation of the CD spectra of diastereomers correctly reproduced the sign of the exciton
couplets in the studied molecules, however, no major solvent dielectric constant influence and
conformational effects per se on the exciton coupling was observed. Chirality 00:000–000, 2012.
© 2012 Wiley Periodicals, Inc.
KEY WORDS: absolute configuration; CD spectroscopy; DFT calculations; exciton coupling;
enantiospecific synthesis
INTRODUCTION
conformationally flexible molecules depend on the contribu-
tion of all the populated conformers, theoretical methods
became indispensable tools for the elucidation of the confor-
mational factors on the overall shape of the CD spectra.8
Herein, conformational analysis of the diastereomeric
dibenzoates 3, 4, and 9 was performed, and dependence of
the CD spectra on the conformational ambiguity was studied
by using TDDFT methodology.
Chiral molecules having two chromophores with strong
electric-dipole allowed transitions couple to each other through
space (exciton coupling).1,2 This phenomenon was intensively
studied over the years by the circular dichroism (CD) spec-
troscopy, and many examples are available in the literature
demonstrating employment of the exciton chirality method
for stereochemical studies. It was used to set the absolute con-
figuration3 and to study solvent influence upon the intensity of
the bands in the CD spectra.4 CD spectroscopy is not only a tool
for studying the absolute configuration of organic molecules
but also provides means to examine interactions between
chromophores. In majority of examples known in the literature,
appropriate polychromophoric molecular structures were cho-
sen to study the impact of the spatial orientation of chromo-
phores, as well as the conformational and solvent effects, on
the exciton coupling.
Because of a proper molecular geometry, bridged bicyclic
compounds are suitable model structures for studying chirop-
tical properties and interaction between the non-conjugated
chromophores.5–7 Herein, we present the synthesis of chiral
diastereomeric bicyclo[3.3.1]nonane dibenzoyl esters 3, 4,
9 and the respective monoesters 7 and 8 (Scheme 1), and a
study of their CD spectra. The benzoate chromophores with
an intense p-p* absorptions are classical excitons that give
rise to a characteristic pair of CD bands with opposite signs
originating from the 1La transitions. In addition, conformation-
ally labile bicyclo[3.3.1]nonane framework that may adopt the
chair-chair and chair-boat conformation provides a good
model to study the contribution of conformers to the CD spec-
tra8 and, at the same time, more complicated picture of the
exciton coupling. Recent advances on calculation of chiropti-
cal properties have greatly enhanced its value in determining
absolute configuration of chiral molecules. In particular,
the time-dependent density functional theory (TDDFT)5 has
become an important tool in theoretical CD spectroscopy
because of a reasonable balance between accuracy and
computational efficiency. As the chiroptical properties of
© 2012 Wiley Periodicals, Inc.
MATERIALS AND METHODS
Instrumentation
Proton and carbon nuclear magnetic resonance spectra were recorded
in deuterated chloroform if not stated otherwise on a Varian Unity INOVA
spectrometer (Varian, Palo Alto, CA, USA) at 300 MHz for protons and
75 MHz for carbon, and chemical shifts are reported in parts per million
relative to tetramethylsilane as an internal standard. The CD and UV
spectra were recorded with a JASCO J-815 spectrometer (Jasco, Tokyo,
Japan) at 20 ꢀC in a 0.1 cm cell. The stock solutions were prepared by
weighting compound into volumetric flask and diluting with UV-grade sol-
vents. The CD spectra were measured in millidegrees and normalized into
emax [l mol–1 cm–1]/l[nm] units. Optical rotations were measured on a
KRÜSS P3001RS automatic digital polarimeter at 589 nm, and [a]20 values
are given in 10ꢁ1 deg cmꢁ1 gꢁ1, and concentrations are given in units of
g/100 cm3. IR spectra were recorded in KBr pellets on a PerkinElmer
Spectrum BX spectrometer (PerkinElmer, Waltham, MA, USA). Elemen-
tal analysis was performed by the analytical laboratory of the Chemistry
Faculty, Vilnius University. Thin-layer chromatography was carried out
on Kieselgel 60 F254 (Merck) sheets coated with silica gel, and Kieselgel
60 silica gel (0.040–0.063 mm, Merck, Darmstadt, Germany) was used for
column chromatography. Melting points were recorded with a Koefler
melting apparatus and are not corrected.
Contract grant sponsor: European Social Fund under the Global Grant
measure; Contract grant number: VP1-3.1-ŠMM-07-K-01-030).
*Correspondence to: Sigitas Stoncˇius or Eugenijus Butkus, Department of
Organic Chemistry, Vilnius University, Naugarduko 24, 03225 Vilnius,
Lithuania. E-mail: sigitas.stoncius@chf.vu.lt or Eugenijus.butkus@chf.vu.lt
Received for publication 1 December 2011; Accepted 17 April 2012
DOI: 10.1002/chir.22075
Published online in Wiley Online Library
(wileyonlinelibrary.com).

¯
BAGDŽIUNAS ET AL.
(m, 4H), 5.20 (ddd, J = 11.2, 6.4, 4.9 Hz, 2 H), 2.20–1.57 (m, 8H), 1.77
(s, 2H), 1.62–1.57 (m, 2H). 13C NMR: d 166.0, 133.0, 130.9, 129.7,
128.5, 75.7, 32.3, 31.0, 27.9, 23.8; Anal. Calcd for C23H24O4: C 75.80; H
6.64. Found: C 75.67; H 6.73.
Exo, Exo-(+)-(1S,2S,5S,6S)-bicyclo[3.3.1]nonane-2,6-diyl
dibenzoate (4)
A solution of 2 (100 mg, 0.641 mmol), PhCO2H (626 mg, 5.13 mmol),
and Ph3P (1.34 g, 5.13 mmol) in 10 ml of dry THF was cooled to 0 ꢀC,
and DIAD (1.0 ml, 5.13 mmol) was added under an argon atmosphere.
The reaction mixture was stirred at room temperature for 1 h and then
stirred in a closed vessel for 48 h at 60 ꢀC under an argon atmosphere.
Celite was added to the cooled reaction mixture and evaporated to dryness.
The residue was purified by column chromatography (eluent ethyl acetate
and petroleum ether 1:20) to yield 105 mg of 4 (45%).
[a]D = +70.2 (0.57, CHCl3); mp 114–115 ꢀC; IR nmax/cmꢁ1 2927, 1708, 1450,
1277, 714; 1H NMR: d 8.06–7.98 (m, 4H), 7.52–7.49 (m, 2H), 7.47–7.19
(m, 4H), 5.22–5.16 (m, 2H), 2.17–1.75 (m, 10H), 1.74–1.45 (m, 2H). 13C
NMR: d 165.9, 132.9, 131.1, 129.7, 128.5, 74.3, 35.5, 31.8, 26.4, 23.8; Anal.
Calcd for C23H24O4: C 75.80; H 6.64. Found: C 75.79; H 6.69.
Scheme 1. Reagents: a) NaBH4, MeOH, 0 ꢀC; b) PhCOCl, DIPEA, DMAP,
DCM, 0 ꢀC ! Δ; c) PhCO2H, Ph3P, DIAD, THF, 0 ! 60 ꢀC; d) TsOHꢂH2O,
acetone, 40 ꢀC.
Exo-(+)-(1S,2S,5S)-6,6-ethylenedioxybicyclo[3.3.1]
nonane-2-yl benzoate (7) and Endo-(+)-(1S,2R,5S)-6,
6-ethylenedioxybicyclo[3.3.1]nonane-2-yl benzoate (8)
Synthesis
The synthesis of starting enantiomerically pure (+)-(1S,5S)-bicyclo
The monoesters 7 and 8 were synthesized from monoacetal 6 following
the same procedures described for diesters 3 and 4 earlier. The crude
products were purified by column chromatography (eluent ethyl acetate
and petroleum ether 1:10) to yield 78 and 74% of 7 and 8, respectively.
7:[a]D = +34.6 (1.3, CHCl3); mp 81–83ꢀC; IR nmax/cmꢁ1 2942, 1710, 1451,
1280, 1103, 721; UV (EtOH), lmax, nm (eꢂ10ꢁ4, dm3 molꢁ1 cmꢁ1): 230 (1.1),
273 (0.099), 280 (0.082). 1H NMR: d 8.08–8.05 (m, 2H), 7.59–7.53 (m, 1H),
7.47–7.42 (m, 2H), 5.16 (m, 1H), 4.02–3.87 (m, 4H), 2.11–1.78 (m, 13H).
13C NMR: d 165.9, 132.9, 129.7, 128.5, 110.9, 74.2, 64.5, 64.4, 36.5, 31.8,
31.5, 26.4, 26.2, 25.6, 22.1; Anal. Calcd for C18H22O4: C 71.50; H 7.33. Found:
C 71.57; H 7.29.
[3.3.1]nonane-2,6-dione
1 and 6,6-ethylenedioxybicyclo[3.3.1]nonane-
2-one 5 was performed following the procedures described earlier.9,10
(+)-(1S,2R,5S,6R)-bicyclo[3.3.1]nonane-2,6-diol 2 and endo-2-ol 6 were
prepared from the diketone 1 and monoprotected ketone 5, respectively,
by a reduction with sodium borohydride.
Endo, Endo -(+)-(1S,2R,5S,6R)-bicyclo[3.3.1]nonane-2,
6-diol (2)
Sodium borohydride (375 mg, 9.87 mmol) was slowly added to a cooled
0 ꢀC solution of 1 (500 mg, 3.29 mmol) in 20 ml of dry methanol. The reac-
tion mixture was stirred for 2 h at 0 ꢀC, then Celite was added and the
mixture was evaporated to dryness. The residue was purified by column
chromatography (eluent ethyl acetate) to yield 470 mg of 2 (92%).
[a]D = +59 (1.0, EtOH) (lit [a]D = +55 (1.0, EtOH)11 and [a]D = +59.0
(0.77, EtOH)12); 1H NMR (CD3OD): d 3.81 (ddd, J= 12.3, 6.0, 4.8 Hz, 2H),
2.09 (dd, J = 12.3, 6.0 Hz, 2H), 1.88–1.73 (m, 4H), 1.72–1.39 (m, 6H). 13C
NMR (CD3OD): d 73.1, 32.9, 32.1, 30.2, 22.5.
8:[a]D = +32.5 (2.0, CHCl3); mp 62–64ꢀC; IR nmax/cmꢁ1 2938, 1716, 1452,
1279, 1114, 713; UV (EtOH), lmax, nm (eꢂ10ꢁ4, dm3 molꢁ1 cmꢁ1): 229
(1.1), 273 (0.096), 280 (0.079). 1H NMR: d 8.07–8.04 (m, 2H), 7.58–7.52
(m, 1H), 7.46–7.41 (m, 2H), 5.20 (ddd, J= 10.5, 6.5, 4.8 Hz, 1H), 4.02–3.87
(m, 4H), 2.12–1.61 (m, 13H). 13C NMR: d 166.0, 132.9, 130.9, 129.7, 128.4,
110.9, 75.8, 64.5, 64.4, 35.9, 33.0, 31.1, 30.9, 26.5, 24.5, 23.2; Anal. Calcd for
C18H22O4: C 71.50; H 7.33. Found: C 71.54; H 7.45.
Exo, Endo-(+)-(1S,2S,5S,6R)-bicyclo[3.3.1]nonane-2,
(+)-(1S,2R,5S)-6,6-ethylenedioxybicyclo[3.3.1]
6-diyl dibenzoate (9)
nonane-2-ol (6)
A solution of 8 (53 mg, 0.175 mmol) and p-toluenesulfonic acid mono-
hydrate (10 mg, 0.053 mmol) in 5 ml of acetone was stirred at 40 ꢀC for
1 h. The reaction mixture was evaporated to dryness, dissolved in dichlor-
omethane, washed with sat. NaHCO3, brine, dried over Na2SO4, filtered,
and evaporated. The obtained residue was used without further purifica-
tion and converted to the corresponding monobenzoate of 2 via reduction
with sodium borohydride as described earlier for 2. Purification by column
chromatography (eluent ethyl acetate and petroleum ether 1:2) afforded
39mg (86%) of the intermediate monobenzoate of bicyclo[3.3.1]nonane-2,6-
diol 2. 1H NMR: d 8.07–8.04 (m, 2H), 7.58–7.53 (m, 1H), 7.45–7.41 (m, 2H),
5.22 (ddd, J= 11.7, 6.8, 5.0Hz, 1H), 3.93 (ddd, J = 16.0, 6.0, 5.0 Hz, 1H),
2.18–1.53 (m, 12H). 13C NMR: d 166.1, 132.9, 130.9, 129.7, 128.4, 76.0,
72.9, 33.8, 32.5, 31.4, 31.0, 27.9, 23.9, 22.6.
The hydroxy monoacetal 6 was synthesized from acetal 5 by reduction
with sodium borohydride using the same procedure as for diol 2. The
product was purified by column chromatography (eluent ethyl acetate
and petroleum ether 1:2) to yield 89% of 6.
[a]D = +42.1 (1.40, CHCl3); mp 75–77 ꢀC; 1H NMR: d 3.97–3.81 (m, 5H),
2.06–1.47 (m, 12H). 13C NMR: d 111.1, 72.9, 64.4, 64.3, 36.0, 34.0, 33.0,
31.1, 30.1, 24.7, 22.0.13
Endo, Endo (+)-(1S,2R,5S,6R)-bicyclo[3.3.1]nonane-2,
6-diyl dibenzoate (3)
A solution of 2 (100 mg, 0.641 mmol) and DMAP (10 mg, 0.082 mmol)
in 5 ml of dry dichloromethane was cooled to 0 ꢀC under an argon atmo-
sphere. DIPEA (0.45 ml, 2.56 mmol) was added to the reaction mixture,
followed by a slow addition of PhCOCl (0.30 ml, 2.56 mmol). The reaction
mixture was stirred at room temperature for 1 h and then refluxed for
12 h. The reaction mixture was diluted with dichloromethane and washed
with sat. NaHCO3, brine, dried over Na2SO4 and filtered. Celite was added
to the filtrate, and the mixture was evaporated to dryness. The residue was
purified by column chromatography (eluent ethyl acetate and petroleum
ether 1:20) to yield 128 mg of 3 (55%).
The diester 9 was synthesized from bicyclo[3.3.1]nonane-2,6-diol
monobenzoate following the same procedure as described earlier for
diester 4. The crude product was purified by column chromatography
(eluent ethyl acetate and petroleum ether 1:20) to yield 30 mg (47%) of 9.
[a]D = +3.7 (1.35, CHCl3); mp 118–119 ꢀC; IR nmax/cmꢁ1 2934, 1712,
1
1452, 1276, 1110, 718; H NMR: d 8.14–8.09 (m, 4H), 7.63–7.57 (m, 2H),
7.52–7.45 (m, 4H), 5.33–5.28 (m, 2H), 2.32–1.45 (m, 12H). 13C NMR: d
165.99, 165.85, 133.0, 132.9, 131.0, 130.8, 129.7, 129.0, 128.5, 75.7, 74.3,
31.3, 31.3, 27.8, 27.4, 26.7, 26.3, 21.3; Anal. Calcd for C23H24O4: C 75.80;
H 6.64. Found: C 75.89; H 6.76.
[a]D = +89.8 (1.28, CHCl3); mp 60–61 ꢀC; IR nmax/cmꢁ1 2920, 1711, 1451,
1277, 711; 1H NMR: d 8.01–7.98 (m, 4H), 7.51–7.49 (m, 2H), 7.48–7.34
Chirality DOI 10.1002/chir

EXCITON CHIRALITY OF BICYCLIC BENZOATES
Computational Details
This molecule is not of C2 symmetry what is well seen from
the signals of protons adjacent to the benzoyl group in the 1H
NMR spectrum. Two sets of signals correspond to the exo-
protons and endo-protons at the C2 and C6 bearing benzoyl
groups.
Conformational search of 3–4 and 9 was performed with SPARTAN’1014
using Monte Carlo method and MMFF94 force field. The minimum energy
conformers found by molecular mechanics have been further optimized
using Gaussian software15 at DFT/B3LYP/6-31G* level, followed by calcu-
lations of their harmonic vibrational frequencies to verify their stability
and to calculate conformational free energies; population percentages were
calculated using ΔG and applying Boltzmann statistics at T = 298.15 K.
The electronic CD calculations of the conformers have been carried out
by means of the Gaussian software15 employing the TDDFT approach, the
B3LYP functional and the 6-31G* basis set. The rotational strength calcula-
tions have been carried out both in velocity (Rvel) and length formalism
(Rlen); the results in the two formalisms were almost identical. The CD spec-
tra of the individual conformers were simulated by overlapping Gaussian
functions for each transition16 using SpecDis.17 The half-width at 1/e of
the peak maximum (s) value of 0.26eV and Rvel was used in this work.
The calculated spectra of 3–4 and 9 were Boltzmann averaged according
to the population percentages of individual conformers and were red-shifted
by 5 nm in relation to the experimental ones to match the experimental lmax
value of the long wavelength Cotton effect (CE). Additionally, for compound
3, the geometry of the conformers in ethanol solution was optimized using
PCM model at B3LYP/6-31G* level and electronic circular dichroism
(ECD) spectra were calculated at the same level of theory. Boltzmann aver-
aged ECD spectrum of 3 calculated at PCM/DFT/6-31G* level was simu-
lated using s 0.26 eV, and Rvel and was not wavelength corrected.
Conformational Analysis, CD Spectra and Theoretical
Computations
The benzoate chromophores with intense p-p* absorptions
are well-studied excitons that give rise to a characteristic pair
of CD bands with opposite signs originating from the 1La
transitions. In the synthesized diastereomeric esters, these
chromophores have different stereochemical orientation in a
conformationally labile bicyclo[3.3.1]nonane molecule. The
bicyclic framework may adopt the chair-chair and chair-boat
conformations, thus giving a good model of the contribution
of all conformers to the CD spectra and, at the same time,
more complicated picture of the exciton coupling.
We performed lowest energy conformational search of 3,
4, and 9 with SPARTAN’10 using Monte Carlo method and
MMFF94 force field. Three or four low-energy conformers
within 1 kcal/mol energy window were located in each case
and were further optimized using Gaussian and the DFT/
B3LYP/6-31G* method. The relative free energies and
populations of conformers are listed in Table 1.
RESULTS AND DISCUSSION
Synthesis
In general, for all diastereomeric dibenzoates MMFF94
calculations predict three (four in the case of 9) stable
chair-chair conformations, whereas corresponding chair-boat
conformers are much higher in energy (5.97, 3.89 and
4.14 kcal/mol for 3, 4, and 9, respectively) and thus insignif-
icantly populated. The benzoate moieties adopt conformation
typical for alkyl benzoates,8 that is, s-trans conformation
around the benzoate (O)C–O bond with the ester carbonyl
group syn with respect to bicyclic methine H–C2,6 hydrogens.
At the B3LYP/6-31G* level, low-energy chair-chair confor-
mers of 3, 4, and 9 are comparably populated and differ
principally with regard to the conformation around HC2,6–O
bond, as expressed by a torsion angle (O)C–O–C–H (Table 1).
For 3, C1 symmetric conformer (a) with the negative and
positive torsion angles (O)C–O–C–H was found to be the
most stable, whereas the two less stable C2 symmetric confor-
mers (b) and (c) were found to be almost equally populated.
In the case of 4, the C2 and C1 symmetric conformers (a) and
(b) were found to be quite close in energy and both signifi-
cantly populated. Conformer (a) with both positive torsion
angles (O)C–O–C–H was found to be the most stable in the
case of endo, exo-diastereomer 9, whereas the least stable
conformers (c) and (d) were characterized by the negative/
positive and positive/negative torsion angles (O)C–O–C–H
of the benzoate moieties, respectively. The notation of confor-
mers of the diesters 3–4 and 9 is presented in Figure 1.
In the experimental CD spectra of the dibenzoates 3, 4,
The (+)-(1S,5S)-bicyclo[3.3.1]nonane-2,6-dione 1 was obtained
in an enantiomerically pure form by kinetic resolution of the
corresponding racemic diketone by using Baker’s yeast.9
Synthesis of the respective monoprotected acetal 6,6-ethylene-
dioxybicyclo[3.3.1]nonane-2-one 5 was performed following
the procedure described earlier.10 The reduction of carbonyl
groups in 1 as well as in 5 with sodium borohydride afforded
exclusively endo-isomer as was proved for the racemic com-
pound,18 leading to endo, endo-(+)-(1S,2R,5S,6R)-2,6-diol 2
and endo-2-ol 6, respectively.
Endo, endo-(+)-(1S,2R,5S,6R)-bicyclo[3.3.1]nonane-2,6-diyl
dibenzoate 3 was obtained by a standard benzoylation proce-
dure (Scheme 1). The inversion of the endo configuration of
substituents in the bicyclo[3.3.1]nonane molecule has not
been described previously to our knowledge. The exo, exo-
dibenzoate 4 was synthesized from diol 2 by a Mitsunobu
reaction with diisopropyl azodicarboxylate via adducts with
triphenylphosphine and inversion of the configuration. The
1H NMR spectrum of dibenzoate 3 is in correspondence with
the C2 symmetry and resonance signal of the C2,6 exo-protons
appears as doublet of doublet of doublets at 5.20 ppm with
large spin-spin coupling constants to protons at C3,7 and C1,5
(11.2, 6.4 and 4.9 Hz, respectively). For a comparison, in the
1H NMR spectrum of 4, the resonance signal of protons
adjacent to the benzoyl group appears as a broad unresolved
multiplet, and small coupling constants of the endo-proton are
characteristic to this configuration.
1
and 9, the exciton couplet located at the chromophore La
transition wavelength near 230 nm was observed. In concor-
dance with the exciton-type coupling, the strong absorption
maxima at the same wavelength are observed in the UV
spectra. For diendo and diexo benzoates 3 and 4, the positive
Synthesis of the endo, exo-diester 9 involved the reduction of
monoacetal 5 with NaBH4, followed by selective benzoylation
of the intermediate endo-alcohol 6 under Mitsunobu or stan-
dard conditions to give the corresponding exo-monobenzoates
and endo-monobenzoates 7 and 8, respectively. Subsequent
deprotection of the endo-monobenzoate 8 with p-TsOH in
acetone afforded the corresponding carbonyl derivative, the
reduction of which followed by Mitsunobu reaction yielded
endo, exo-dibenzoate 9 (Scheme 1).
1
intense exciton couplet due to coupling between the two La
transitions was observed, whereas the endo, exo-benzoate 9
exhibited couplet of the opposite sign and virtually the same
intensity. Unequivocally, the exciton through-space coupling
was confirmed by a comparison with the respective mono
Chirality DOI 10.1002/chir

¯
BAGDŽIUNAS ET AL.
TABLE 1. Relative free energies, populations, symmetries, and torsion angles for low-energy conformers of 3, 4, and 9, calculated at
the B3LYP/6-31G* level
Conformer
(symmetry)
ΔG,
kcal/mol
Population,
%
Torsion angle
Angle between transition
moments, (ꢀ)
Ester
(O)C–O–C–H, (ꢀ)
Diendo-3
a (C1)
b (C2)
c (C2)
a (C2)
b (C1)
c (C2)
a (C1)
b (C1)
c (C1)
d (C1)
0.00
0.43
0.53
0.00
0.18
0.49
0.00
0.32
0.41
0.45
52.8
25.7
21.5
46.1
33.9
20.0
39.3
22.8
19.7
18.2
ꢁ31.3, 32.3
ꢁ32.7
73.3
46.6
86.2
37.7
63.8
30.9
Diexo-4
ꢁ36.4
32.2, ꢁ35.2
34.6
79.3
Endo,exo-9
34.2, 34.3
ꢁ32.3, ꢁ34.6
ꢁ32.7, 33.9
34.9, ꢁ33.9
ꢁ68.6
ꢁ107.5
ꢁ91.5
ꢁ93.7
The electronic excitation energies and rotational strengths
in the gas phase have been calculated for the conformers of
diesters 3, 4, and 9 using TDDFT with the 6-31G* basis set
and the B3LYP model. The calculated rotatory strengths (Rvel
)
were simulated into the CD curves assuming a Gaussian band
shape with a half-width at 1/e of the peak maximum (s) value
of 0.26 eV, approximately matching the broadening of the CD
bands in the experimental spectra. For a comparison with the
experimental CD data, the spectra of individual conformers
calculated at B3LYP/6-31G*//B3LYP/6-31G* level were red-
shifted to match the experimental lmax value of the long
wavelength CE and are presented in Figures 3–5.
The predicted spectra for the individual conformers of die-
sters 3 and 4 were very similar regarding the pattern, absolute
and relative intensity of the Cotton effects. In each case, a
1
strong positive couplet due to exciton coupling between La
transitions was predicted (Figs. 3 and 4). Theoretical calcula-
tions19 have shown that the intensity of the split Cotton effect
for the vicinal dibenzoates is maximal at a dihedral angle of
ca. 70ꢀ and is inversely proportional to the square of the inter-
chromophoric distance. As the difference of the latter for con-
formers of the same diasteromeric ester 3 (or 4) is negligible,
Fig. 1. Geometry of low-energy conformers of 3–4 and 9 calculated at the
B3LYP/6-31G* level (H atoms omitted for clarity).
congeners 7 and 8, the CD spectra of which were of a very
low intensity (Fig. 2).
The sign of the exciton coupling observed in the CD spec-
tra is in accordance with the orientation of the chromophores
(Fig. 1), that is, all conformers of 3 (or 4) are characterized
by a positive helicity of the interacting transition electric
dipole moments, which are polarized along the axis of the
benzoate chromophore, almost parallel to the C–O bond of
the respective hydroxyl compound. Analogously, for all the
conformers of endo, exo-benzoate 9 negative twist defined
by interacting transition moments is in agreement with the
observed negative couplet in the CD spectrum.
The solvent effect (solvatochromism) on the CD spectra of
the dibenzoates 3, 4, and 9 was also studied. The CD spectra
were recorded in solvents of different polarity—ethanol,
dichloromethane, cyclohexane, 2,2,2-trifluoroethanol, however,
no significant effect on the intensity and position of the respec-
tive CEs was observed (Table 2). This indicates that the
dielectric constant of media has no major effect on the
conformational distribution (Table 1) or conformers exhibit
very similar CD spectra in terms of intensity and position.
The experimental spectrum is the weighted-average
spectrum of the respective conformers, thus the theoretical
computation of the CD spectra of the individual conformers
was performed.
Fig. 2. CD spectra of 3 (—), 4 (ꢁ ꢁ), 7 (ꢁꢂꢂꢁꢂꢂ), 8 (ꢁꢂꢁꢂ), and 9 (ꢂꢂꢂ) and
representative UV spectra of 3 (—) and 8 (ꢁꢂꢁꢂ).
Chirality DOI 10.1002/chir

EXCITON CHIRALITY OF BICYCLIC BENZOATES
TABLE 2. The CD and UV spectral data for compounds 3, 4, and 7–9
UV, lmax, nm (e 10ꢁ4, dm3 molꢁ1 cmꢁ1
)
CD, lmax, nm (Δe, dm3 molꢁ1 cmꢁ1
)
.
Compound
Solvent
Diendo-3
Diendo-3
Diendo-3
Diendo-3
Diexo-4
CH3CH2OH
CH2Cl2
Cyclo-C6H12
CF3CH2OH
CH3CH2OH
CH2Cl2
Cyclo-C6H12
CF3CH2OH
CH3CH2OH
CH2Cl2
Cyclo-C6H12
CF3CH2OH
230 (2.2), 273 (0.16), 280 (0.13)
230 (2.6), 273 (0.21), 280 (0.17)
228 (2.2), 272 (0.17), 280 (0.14)
230 (2.7), 270 (0.029)
230 (2.1), 273 (0.15), 280 (0.12)
230 (2.4), 273 (0.20), 280 (0.17)
228 (2.5), 272 (0.19), 280 (0.15)
231 (2.6), 271 (0.050), 281 (0.0072)
230 (2.2), 273 (0.19), 280 (0.16)
230 (2.4), 273 (0.20), 280 (0.16)
229 (2.7), 273 (0.24), 280 (0.21)
231 (2.4), 270 (0.011)
222 (ꢁ6.0), 237 (+14.2)
222 (ꢁ5.3), 236 (+16.6)
222 (ꢁ3.6), 233 (+14.6), 236 (+13.9)
221 (ꢁ7.5), 238 (+20.0)
221 (ꢁ4.5), 235 (+12.7),
Diexo-4
221 (ꢁ2.2), 236 (+14.5)
Diexo-4
Diexo-4
Exo,endo-9
Exo,endo-9
Exo,endo-9
Exo,endo-9
221 (ꢁ2.6), 233 (+15.7), 235 (+15.6)
221 (ꢁ6.7), 238 (+18.2)
221 (+5.7), 236 (ꢁ11.0), 238 (ꢁ11.3)
220 (+5.8), 222 (+5.7), 235 (ꢁ11.9), 237 (ꢁ11.9)
220 (+6.9), 233 (ꢁ12.3), 237 (ꢁ12.7)
221 (+6.1), 237 (ꢁ14.3)
Fig. 3. Top panel: CD spectra for individual conformers of 3 (a, ꢁ ꢁ; b, —;
c, ꢁꢂꢁꢂ), calculated at B3LYP/6-31G* level; bottom panel: Boltzmann aver-
aged CD spectrum of 3 calculated at B3LYP/6-31G* (ꢁꢂꢁꢂ) and PCM/
B3LYP/6-31G* (ꢁ ꢁ) levels; experimental CD spectrum of 3 (—) in ethanol.
All spectra calculated at B3LYP/6-31G* level were wavelength corrected.
Fig. 5. Top panel: CD spectra for individual conformers of 9 (a, —; b, ꢂꢂꢂꢂ;
c ꢁ ꢁ; d, ꢁꢂꢁꢂ;), calculated at B3LYP/6-31G* level; bottom panel: Boltzmann
averaged CD spectrum of 9, calculated at B3LYP/6-31G* level and experimen-
tal CD spectrum of 9 (—) in ethanol. All calculated spectra were wavelength
corrected.
the calculated amplitude of the exciton couplets correlates with
the calculated angle between the corresponding transition
moments (Table 1) in the dichromophoric molecules. Thus,
in the case of 3, the most intense exciton couplets (A = +50.5
and 43.8) are predicted for the conformers (c) and (a), for
which the calculated distances and angles between transition
moments are 0.979 nm/86.2ꢀand 0.983 nm/73.3ꢀ, respectively.
In comparison, for conformer (b), where the angle and distance
between transition moments are 46.6ꢀ and 0.998 nm, respec-
tively, substantially smaller exciton coupling amplitude (+35.7)
is predicted. The calculated intensity of the exciton couplets
A = +26.4, 38.5, and 48.5, respectively for the conformers a–c
of dibenzoate 4 follows the same trend (calculated distances be-
tween transition moments are 1.015, 0.989, and 0.963 nm, re-
spectively; for angles between transitions, see Table 1).
Analogously, strong negative exciton couplets were pre-
dicted for all conformers of endo, exo-diasteromer 9 (Fig. 5).
Qualitatively, CD spectra of 3–4 and 9, averaged on the
basis of the Boltzmann population of the respective confor-
mers, are in a good agreement with the experimental data,
that is, the sign of the exciton couplet and relative intensity
of the corresponding CEs are correctly reproduced. B3LYP/
6-31G* calculations predict small rotational strengths at
Chirality DOI 10.1002/chir
Fig. 4. Top panel: CD spectra for individual conformers of 4 (a, ꢁꢂꢁꢂ;
b ꢁ ꢁ; c, —;), calculated at B3LYP/6-31G* level; bottom panel: Boltzmann
averaged CD spectrum of 4 calculated at B3LYP/6-31G* level and experimen-
tal CD spectrum of 4 (—) in ethanol. All calculated spectra were wavelength
corrected.

¯
BAGDŽIUNAS ET AL.
1
250–260 nm corresponding to Lb transition of the benzoate
chromophore, resulting in a net weak long-wavelength region
CE in the Boltzmann-averaged CD of 3–4 and 9. This is in
accordance with the experimentally observed small CEs for
1
benzoate Lb transition in these esters. The CD spectra of
3–4 and 9, calculated using semi-empirical ZINDO/S
method, were in much worse agreement with the experimen-
tal data.20 Although the sign of the exciton couplets was
correctly reproduced, the respective CEs were substantially
(ca. 20 nm) blue-shifted; in addition, long-wavelength region
CEs of comparable intensity corresponding to the benzoate
1
chromophore Lb transition were also predicted.
Molecular orbital analysis of investigated diastereomeric
esters allows us to confirm that the coupling mechanism
between two benzoate chromophores is mainly excitonic.
The exciton couplet in the calculated CD spectra of diastereo-
meric benzoates 3–4 and 9 originates from the two intense,
close-lying transitions with opposite signs of the rotational
strengths. Analysis of Kohn–Sham molecular orbitals (MO)
revealed that single-electron excitations from highest occupied
molecular orbital (HOMO) and occupied MOs HOMO-3,
HOMO-2, HOMO-1 to lowest unoccupied molecular orbital
(LUMO) and LUMO + 1 contribute to the two absorption bands
of the exciton couplet. For instance, for the most stable con-
former of endo, endo-diastereomer 3, the electronic excitations
MO 95 ! MO 98 and MO 94 ! MO 99, polarized parallel to
the long axis of the benzoate chromophore, play a dominant role
(contribution 42.4 and 19.9%) in the positive rotatory strength lo-
cated at 225.5 nm (Rvel =346ꢂ10–40 cgs). Contribution of the other
single-electron excitations MO 96 ! MO 99 and MO 97 ! MO 98
to the positive rotatory strength at 225.5 nm is 7 and 6.2%,
respectively. The negative transition of the exciton
couplet is located at 223.9 nm (Rvel = ꢁ308ꢂ10ꢁ40 cgs),
and the contributions in terms of one-electron excitations
are MO 94 ! MO 99 (41.3%), MO 95 ! MO 98 (22.6%),
MO 96 ! MO 99 (13.8%), and MO 97 ! MO 98 (7.1%).
The molecular orbitals involved in the key transitions in
the calculated CD for the conformer (a) of dibenzoate 3
are shown in Figure 6.
In general, the excitation energies calculated at B3LYP/
6-31G* level were consistently underestimated in comparison
with the experimental spectra. The blue shift of the calculated
CD spectra relative to the experimental ones is not uncom-
mon21,22 and can be as a result of the errors in the density
functional and the neglect of solvent and vibrational effects.23
Conformers of 3 were reoptimized with the use of the
Polarizable Continuum Model (PCM) of solvent (ethanol) in
conjunction with B3LYP/6-31G*, followed by calculations of
CD spectra at the PCM/B3LYP/6-31G* level of theory. The
respective structures reoptimized at the PCM/DFT/6-31G*
level were similar to those obtained with the use of B3LYP/
6-31G* method, except for small changes in the bond lengths
and values of (O)C–O–C–H torsion angles; the computed rel-
ative populations of the conformers (79.1, 10.7, and 10.2% for
conformers a, b, and c, respectively; cf. Table 1) were affected
to a small extent. Analogously to the calculations in vacuum,
PCM/B3LYP/6-31G* predicts intensive positive exciton
couplets for all the conformers of 3. The position of the CEs
in the Boltzmann-averaged CD spectrum of 3 is in excellent
agreement with the experimentally observed ones (Fig. 3).
The use of the PCM model, however, required substantially
larger computational costs; thus, gas-phase calculations at
B3LYP/6-31G* level, which provide acceptable compromise
Chirality DOI 10.1002/chir
Fig. 6. Molecular orbitals involved in key transitions in the calculated CD
for the conformer (a) of dibenzoate 3 at B3LYP/6-31G* level.
between the accuracy and the computational costs, were used
for qualitative analysis of the CD spectra.
CONCLUSIONS
Synthesis of diastereomeric bicyclo[3.3.1]nonane dibenzoyl
esters derived from enantiomerically pure (1S,5S)-2,6-dione
was accomplished, and molecules containing two identical
chromophores with different spatial arrangement in the bicy-
clic ring system were obtained. CD spectra of the synthesized
chiral structures were studied. The exciton coupling was
observed in the CD spectra regardless of the orientation of
the chromophores in dichromophoric molecules. The confor-
mational effects and solvent impact on the interaction of the
chromophores were examined by CD spectroscopy. TDDFT
calculated CD spectra of diastereomers were in a good agree-
ment with the experimental data, though showing no major
conformational effect on the exciton coupling. In addition,
the solvent did not affect the conformational distribution
significantly and had negligible effect on the CD spectra.
ACKNOWLEDGMENTS
SS and GB acknowledge the financial support by the
European Social Fund under the Global Grant measure
(VP1-3.1-ŠMM-07-K-01-030). We are grateful to Dr. V.
Šikšnys, Institute of Biotechnology, Vilnius University, for
providing access to the CD spectrometer.
LITERATURE CITED
1. Berova N, Nakanishi K. Exciton chirality method: principles and application.
In: Nakanishi K, Berova N, Woody RW, editors. Circular dichroism: princi-
ples and applications, 2nd ed. New York: Wiley-VCH; 2000. p 337–382.
2. Gawronski J, Skowronek P. Electronic circular dichroism for chiral
analysis. In: Busch, KW, Busch MA, editors. Chiral analysis. Oxford:
Elsevier; 2006. p 397–459.

EXCITON CHIRALITY OF BICYCLIC BENZOATES
3. Berova N, Borhan B, Dong JG, Guo J, Huang X, Karnaukhova E,
15. Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, Robb MA,
Cheeseman JR, Montgomery JA Jr, Vreven T, Kudin KN, Burant JC,
Millam JM, Iyengar SS, Tomasi J, Barone V, Mennucci B, Cossi M,
Scalmani G, Rega N, Petersson GA, Nakatsuji H, Hada M, Ehara M,
Toyota K, Fukuda R, Hasegawa J, Ishida M, Nakajima T, Honda Y,
Kitao O, Nakai H, Klene M, Li X, Knox JE, Hratchian HP, Cross JB,
Bakken V, Adamo C, Jaramillo J, Gomperts R, Stratmann RE, Yazyev O,
Austin AJ, Cammi R, Pomelli C, Ochterski JW, Ayala PY,
Morokuma K, Voth GA, Salvador P, Dannenberg JJ, Zakrzewski VG,
Dapprich S, Daniels AD, Strain MC, Farkas O, Malick DK, Rabuck AD,
Raghavachari K, Foresman JB, Ortiz JV, Cui Q, Baboul AG, Clifford S,
Cioslowski J, Stefanov BB, Liu G, Liashenko A, Piskorz P, Komaromi I,
Martin RL, Fox DJ, Keith T, Al-Laham MA, Peng CY, Nanayakkara A,
Challacombe M, Gill PMW, Johnson B, Chen W, Wong MW, Gonzalez
C, Pople JA. Gaussian 03, Revision C.02. Wallingford CT: Gaussian,
Inc.; 2004.
Kawamura A, Lou J, Matile S, Nakanishi K, Rickman B, Su J, Tan Q, Zanze
I. Solving challenging bioorganic problems by exciton coupled CD. Pure
Appl Chem 1998;70:377–383.
4. Funaioli T, Cavazza M, Zandomeneghi M, Pietra F. On the chiroptical
behavior of conjugated multichromophoric compounds of
a new
pseudoaromatic class: bicolchicides and biisocolchicides. PLoS One
2010;5:e10617. DOI: 10.1371/journal.pone.0010617.
5. Stephens PJ, McCann DM, Butkus E, Stoncˇius S, Cheeseman JR, Frisch
MJ. Determination of absolute configuration using concerted ab initio
DFT calculations of electronic circular dichroism and optical rotation:
bicyclo[3.3.1]nonane diones. J Org Chem 2004;69:1948–1958.
6. Stoncˇius S, Berg U, Butkus E. Chiral bicyclic keto lactones: determination
of the absolute configuration by the study of chiroptical properties and
chemical correlation. Tetrahedron: Asymmetry 2004;15:2405–2413.
7. Butkus E, Berg U, Malinauskiene˙ J, Sandström J. J Org Chem 2000;65:
1353–1358.
16. Stephens PJ, Harada N. ECD cotton effect approximated by the Gaussian
curve and other methods. Chirality 2010;22:229–233.
8. Pesticelli G, DiBari L, Berova N. Conformational aspects in the studies of
organic compounds by electronic circular dichroism. Chem Soc Rev
2011;40:4603–4625.
9. Wallentin CJ, Orentas E, Butkus E, Warnmark K. Baker’s yeast for sweet
dough enables large scale synthesis of enantiomerically pure bicyclo
[3.3.1]nonane-2,6-dione. Synthesis 2009;864–867.
10. Stoncˇius S, Bagdžiu¯nas G, Malinauskiene˙ J, Butkus E. A study of planar
chromophores in dichromophoric molecules by circular dichroism spec-
troscopy. Chirality 2008;20:337–343.
11. Naemura K, Matsumura T, Komatsu M, Hirose Y, Chikamatsu H. Facile
preparation of optically active bicyclo[3.3.1]nonane-2,6-diol and 3,3,7,7-
tetramethylbicyclo[3.3.1] nonane-2,6-diol by enzyme-catalyzed hydrolysis,
and enantiomer recognition behavior of crown ethers and podands having
these diols as a chiral subunit. Bull Chem Soc Jpn 1989;62:3523–3530.
12. Gerlach H. Racematspaltung und Bestimmung der absoluten Konfigura-
tion von 2,6-disubstituierten Bicyclo[3.3.1]nonanen. Helv Chim Acta
1978;61:2773–2776.
13. Bishop R, Parker W, Stevenson JR. Intramolecular reactions involving
positions C(2), C(6) and (C2), C(7) in the bicyclo[3.3.l]nonane ring
system. J Chem Soc Perkin Trans 1 1981;565–573.
17. Bruhn T, Hemberger Y, Schaumlöffel A, Bringmann G. SpecDis, Version
1.51. Germany: University of Wuerzburg; 2011.
18. Butkus E, Malinauskiene J, Kadziauskas P. Zur stereochemie der
Reduktion von Bicyclo[3.3.1]nonan-2,6-dion. Z Chem 1980;20:103.
19. Harada N, Lai Chen S-M, Nakanishi K. Quantitative definition of exciton
chirality and the distant effect in the exciton chirality method. J Am Chem
Soc 1975;97:5345–5352.
20. Humam M, Christen P, Munoz O, Hostettmann K, Jeannerat D. Absolute
configuration of tropane alkaloids bearing Two a,b-unsaturated ester
functions using electronic CD spectroscopy: application to (R,R)-trans-
3-hydroxysenecioyloxy-6-senecioyloxytropane. Chirality 2008;20:20–25.
21. Giorgio E, Tanaka K, Verotta L, Nakanishi K, Berova N, Rosini C.
Determination of the absolute configurations of flexible molecules:
synthesis and theoretical simulation of electronic circular dichroism/
optical rotation of some pyrrolo[2,3-b]indoline alkaloids - a case study.
Chirality 2007;19:434–445.
22. Chochrek P, Frelek J, Kwit M, Wicha J. New insight into chiroptical
properties of 1,2-diols cyclic sulfites. J Org Chem 2009;74:7300–7308.
23. McCann DM, Stephens PJ. Determination of absolute configuration
using density functional theory calculations of optical rotation and
14. SPARTAN ’10 for Windows Version 1.0.1. 1840 Von Karman Avenue,
Suite 370, Irvine, CA 92612: Wavefunction, Inc.; 2010.
electronic circular dichroism: chiral alkenes.
J Org Chem 2006;71:
6074–6098.
Chirality DOI 10.1002/chir