Combined use of three forms of chiroptical spectroscopies in the study of the absolute configuration and conformational properties of 3- phenylcyclopentanone, 3-phenylcyclohexanone, and 3-phenylcycloheptanone

Article abstract of DOI:10.1016/j.tet.2013.10.016

Three forms of chiroptical spectroscopies, electronic circular dichroism (ECD), vibrational circular dichroism (VCD), and optical rotatory dispersion (ORD) have been employed to study the configuration and conformational properties of the three molecules:

Full text of DOI:10.1016/j.tet.2013.10.016

Tetrahedron 69 (2013) 10752e10762
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Combined use of three forms of chiroptical spectroscopies in the
study of the absolute configuration and conformational properties of
3-phenylcyclopentanone, 3-phenylcyclohexanone, and
3-phenylcycloheptanone
,
Patrizia Scafato a, Francesca Caprioli a y, Laura Pisani a, Daniele Padula b,
,
d
Fabrizio Santoro c, Giuseppe Mazzeo d, Sergio Abbate d , France Lebon ,
*
Giovanna Longhi d
a
ꢀ
Dipartimento di Scienze, Universita della Basilicata, Via dell’Ateneo Lucano, 85100 Potenza, Italy
b
ꢁ
Institute of Organic Chemistry and Biochemistry, Academy of Sciences, Flemingovo namĕstí 2, 16610 Prague, Czech Republic
c CNR, Consiglio Nazionale delle Ricerche, Istituto di Chimica dei Composti Organometallici (ICCOM-CNR), UOS di Pisa, Area della Ricerca, via G. Moruzzi
1, 56124 Pisa, Italy
d
ꢀ
Dipartimento di Medicina Molecolare e Traslazionale, Universita di Brescia, Viale Europa 11, 25123 Brescia, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 26 July 2013
Received in revised form 23 September 2013
Accepted 7 October 2013
Available online 21 October 2013
Three forms of chiroptical spectroscopies, electronic circular dichroism (ECD), vibrational circular di-
chroism (VCD), and optical rotatory dispersion (ORD) have been employed to study the configuration
and conformational properties of the three molecules: (S)-3-phenylcyclopentanone, (S)-3-
phenylcyclohexanone, and (S)-3-phenylcycloheptanone (including (S)-3-phenylcyclopentanone-2,2,5,5-
d4 and (S)-3-phenylcyclohexanone-2,2,6,6-d4). ECD and VCD spectra in the mid-IR for the three mo-
lecular systems are marginally dependent on fine conformational details, as interpreted in terms of
standard DFT computational methods, with common spectroscopic features to the three systems clearly
We dedicate this work to the memory of
Carlo Rosini, who suggested undertaking
this study before his premature death
identified. Accounting for vibronic coupling mechanisms reproduces the structuring of ECD n/
The ORD curves are quite similar for the three types of molecules, but their interpretation highlights
crucial role played by conformations of the cycloalkanone ring in the case of (S)-3-
p band.
*
a
Keywords:
ORD
ECD
phenylcycloheptanone. The same conclusions are reached by considering the VCD spectra in the CH-
stretching region.
Ó 2013 Elsevier Ltd. All rights reserved.
VCD
Vibronic features
Absolute configuration
Conformations
Phenyl hindered rotation
1. Introduction
chirality. Through the investigation of chiroptical properties,1 like
optical rotation (OR), optical rotatory dispersion (ORD), electronic
In the last 15e20 years, the increased availability of quantum-
mechanical and computational methods, coded in easy-to-use
and largely available packages, has enabled more and more scien-
tists to better understand chiral molecules by predicting and
comparing several experimental properties with computational
analyses, disclosing new opportunities in the research field of
circular dichroism (ECD) and vibrational circular dichroism (VCD),2
it is possible nowadays to perform accurate structural analyses and
to arrive at the safe determination of the molecular absolute con-
figuration (AC). In many studies, chiroptical data of molecules
having either known or unknown AC were correctly predicted.
More recently the absolute configuration has been assigned mainly
by the concerted use of these techniques.3 The combined use of all
these powerful chiroptical tools allows the AC determination of
organic molecules to receive a safer solution. However, in spite of
such significant progress, some relevant problems still remain: in
particular, the treatment of molecules endowed with high confor-
mational flexibility requires accurate calculation of the relative
* Corresponding author. E-mail address: abbate@med.unibs.it (S. Abbate).
y
Present address: Stratingh Institute for Chemistry, Faculty of Mathematics and
Natural Sciences, University of Groningen, Nijenborgh 4, 9747AG Groningen, The
Netherlands.
0040-4020/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.tet.2013.10.016

P. Scafato et al. / Tetrahedron 69 (2013) 10752e10762
10753
population and chiroptical response for each conformer in order to
obtain the weighted average over the conformer populations of
each property.
Experimental ECD spectra were obtained by a JASCO 815SE appa-
ratus from 400 to 180 nm under the following experimental con-
ditions: integration time 1 s, scan speed 100 nm/min, bandpass
1 nm, 20 accumulations. Compounds 1e3 were measured in hex-
ane solutions in 0.1 mm pathlength quartz cuvette. Concentrations
were 0.0048 M, 0.0036 M, and 0.0058 M for 1, 2, and 3, respectively.
IR and VCD spectra were collected on a JASCO FVS4000 FTIR
equipped with a liquid N2-cooled MCT detector, 2000 accumula-
tions were averaged in the 900e1800 cmꢀ1 region at 4 cmꢀ1 res-
olution, respectively. The spectra were obtained in CCl4 solutions, in
As pointed out above, the use of different chiroptical methods
enables one to get an unambiguous and consistent description of
the important structural and electronic molecular features. Since
the conformational aspect may influence quite differently the var-
ious chiroptical data, on one hand one needs to pay attention in
order to obtain an unequivocal response from all kinds of spec-
troscopies, on the other hand, a satisfactory matching between
experiments and theory gives more information than the simple AC
assignment. The systems herein studied are quite simple and for
this reason they are a good benchmark to test computation po-
tentiality. As we will show in this paper, despite improvements of
theory and computation in the last years, the most primitive among
chiroptical data, namely OR data still remain the most difficult ones
to calculate.4 We compare experimental and computational results
from three techniques (ORD, ECD, and VCD) on three related model
molecular systems, namely 3-phenylcycloalkanones 1e3 (Chart 1)
synthesized by catalytic asymmetric addition of phenylboronic acid
to the corresponding cycloalkenones.5
100 mm pathlength BaF2 cells for 1, 2, and 3 for 0.42 M, 0.35 M, and
0.21 M solutions, respectively. The VCD spectra in the CH-stretching
region (2000e3200 cmꢀ1) were obtained on the same solution
contained in the same cells, using the same apparatus with an InSb
detector and with 5000 accumulations with 8 cmꢀ1 resolution.
2.3. Computational details
The preliminary conformational analysis was performed on (3S)
1e3 by using the Spartan026 package with the MMFF94s molecular
mechanics (MM) force field adopting Monte Carlo (MC) and Sys-
tematic options as search method. Geometries within 10 kcal/mol
energy window have been re-optimized by Density Functional
Theory (DFT) adopting the B3LYP functional and the TZVP basis set
using the Gaussian09 package.7 All investigated (ground-electronic
state) conformers are real minima, no imaginary vibrational fre-
quencies were found. Free energies were calculated and used to
determine the Boltzmann populations of the conformers at
298.15 K.
Calculations of ORD and ECD spectra were performed by the TD-
DFT approach using the Coulomb-attenuated CAM-B3LYP func-
tional8 and aug-cc-pVDZ as basis set. Theoretical ORD, ECD, and
VCD spectra were obtained as averages weighted on the Boltzmann
populations (VCD spectra were calculated at the B3LYP/TZVP level,
as done for the conformer determination). The theoretical ab-
sorption aꢀn1d VCD spectra were simulated with Lorentzian bands
with 4 cm half-width at half maximum (HWHM) for the mid-IR
and 16 cmꢀ1 for the CH-stretching region. To best compare with
experimental spectra, the computed frequencies have been scaled
by a factor 0.985 in the mid-IR, and 0.97 in the CH-stretching re-
gion. The purely electronic ECD spectra were obtained as the sum of
Gaussian functions centered at the calculated wavelength of each
transition with the band area equal to the calculated rotational
strength of the transition and the half-width at half maximum
(HWHM) equal to 0.2 eV and elaborated using SpecDis v1.53.9
Vibronic features of ECD and absorption UV spectra were ob-
tained in FrankeCondon (FC) approximation. In the excited state
(ES) two minima characterized by a pyramidal arrangement of the
CCC]O moiety are found, and they interconvert through a planar
transition state. A rigorous treatment of the effect of the ES double-
Chart 1. Structures of studied compounds 1e3.
2. Materials and methods
2.1. Synthesis and purification
3-Phenylcycloalkanones 1e3 were synthesized in 94e96% ee, as
reported in Ref. 5a, by catalytic asymmetric addition of phenyl-
boronic acid using (S)-BINAP/rhodium(I) complex, in the case of 3-
phenylcyclopentanone 1,5b or a tropos phosphoramidite/rhodium(I)
complex for cycloalkanones 2 and 3.5a 3-Phenylcycloalkanones-d4
10 and 20 were prepared in 94% isotopic purity by refluxing the
corresponding ketones in deuterium oxide, in presence of potas-
sium carbonate (0.11 equiv), and monitoring the conversions by
GCeMS analyses (see SI-part 1 for a detailed description of the
experimental procedure and characterization of deuterated com-
pounds). 1H (400 MHz) NMR spectra were recorded in CDCl3 on
a Varian 400 NMR spectrometer, using tetramethylsilane (TMS) as
internal standard. GCeMS analyses were performed on a Hewlett
Packard 6890 gas chromatograph equipped with an HP-5973 mass
detector and an HP-5MS capillary column.
minimum profile along the carbonyl C pyramidalization (g) would
be challenging and require the following steps: (i) representation of
the normal modes in internal coordinates; (ii) separation (if pos-
sible) of
lation of the anharmonic states supported by the ES energy profile
along . Such an accurate simulation is beyond the scope of the
g from the bath of the other harmonic modes, (iii) calcu-
g
present work, and here we resorted to a simpler approach, con-
sidering as ES reference geometry the transition state in which the
CCC]O moiety is constrained to be planar, as it is at the ground-
state (GS) equilibrium geometry. This strategy has been already
adopted in similar cyclopentanone systems,10,11 showing that it
delivers results in reasonable agreement with experiment as far as
the vibronic progressions along high-frequency modes are con-
cerned; moreover, though the choice of the real frequency to be
assigned to the pyramidalization mode is arbitrary (we simply
considered the absolute value of the imaginary frequency), such
2.2. ORD, ECD, and VCD spectra acquisition
ORD spectra were recorded with Jasco DIP370 digital polarim-
eter at four different wavelengths (589, 546, 435, 405 nm) in
hexane for 1 and 2 and in chloroform for 3 at concentration of 1 g/
100 mL (0.063 M, 0.057 M, and 0.053 M, respectively).

10754
P. Scafato et al. / Tetrahedron 69 (2013) 10752e10762
strategy is safer than trying to describe the ES vibronic states rel-
evant for the spectrum, i.e., those whose population density covers
the planar region, assuming that the ES PES is characterized by
a single harmonic well centered at one of the two pyramidal ge-
ometries. As a matter of fact, simulations performed according to
this later recipe, not reported, lead to completely structureless and
extremely broad spectra. Vibronic calculations were performed by
the FCClasses code,12 in harmonic approximation,13 employing the
so-called Adiabatic Hessian approach (FCjAH). We adopted both an
effective time-independent (TI) method based on a partition of the
possible transitions into classes and on a prescreening technique
for the individuation of the major stick bands14 and a time-
dependent (TD) method ensuring full convergence even at room
temperature.15 Convoluted spectra were obtained employing
a Gaussian bandshape with HWHM¼0.03 eV. Optimized geome-
tries, normal modes and the corresponding frequencies of the
electronic ground and excited states were obtained at DFT/B3LYP/
TZVP level while the excitation energies and circular dichroism
rotatory strengths were calculated at TDDFT/B3LYP/TVZP using
Gaussian09.
3. Results and discussion
3.1. Conformational analysis
Preliminary MM based conformational analysis (with S AC on
carbon 3) provided 4 low energy stable conformers for 1, 5 for 2,
and 16 for 3. The three sets of conformations lie within 10 kcal/mol
and differ mostly by relative axialeequatorial phenyl position and
by the cycloalkanone ring twist; phenyl hindered rotation and
cycloalkanone ring twist interact as investigated in Supplementary
data (Fig. SI-1) for (3S)-3 for conformers 3a and 3b (vide infra),
where the potential energy surface (PES) plots are given. Sets of
conformations sorted by MM were then fully optimized at DFT/
B3LYP/TZVP level of theory and provided the most likely con-
formers presented in Fig. 1. According to DFT results, basically
compounds 1 and 2 exist in almost a single phenyl-equatorial
conformation (more than 92% of the overall population). Higher
flexibility of the seven-membered ring in 3, resulted in eight DFT
optimized conformers with a relative energy value less than 3 kcal/
mol with respect to the most stable phenyl-equatorial conformer
3a. Contrary to what happens in 1 and 2, axial conformations of 3
do not appreciably contribute to overall population (the first most
populated axial structure has 0.3% population, being 3.12 kcal/mol
above the most stable conformer).
Cyclopentanone ring has a half-chair M-twisted conformation,
cyclohexanone assumes the stable chair conformation and cyclo-
heptanone is in a twisted pseudo-chair conformation (see Ref. 16,
pp 174e181). Using the nomenclature introduced in Ref. 16 for (R)-
3-methylcycloheptanone, in the case of molecule 3, conformer 3a is
described as TC7(ꢀ) and conformer 3b as TC2(ꢀ); no other conformer
with appreciable population factor has been found. Phenyl moiety
is in eclipsed conformation with respect to the hydrogen on ster-
eogenic carbon 3.
Fig. 1. Most stable conformers for molecules 1e3 as calculated at DFT/B3LYP/TZVP
level of theory; 1a (91.9%), and 1b (6.6%), 2a (96.7%), and 2b (2.2%), 3a (51.5%), 3b
(44.0%). Fixed AC is (3S)-1e3.
CAM-B3LYP/aug-cc-pVDZ level of theory was used in ORD pre-
diction. In Fig. 2 we show comparison of experimental and calcu-
lated curves for 1e3.
As shown in Fig. 2, TDDFTcalculations predict experimental ORD
curves of 1 and 2 in trend, sign, and magnitude (the latter charac-
teristic is somewhat overestimated though). The case of 3 requires
more attention. TDDFT ORD prediction at this level of theory, is not
a 100% satisfactory even if the negative curve trend is accounted for.
Calculated ORD is approximately five times smaller than the ex-
perimental (Table SI-1). This is indeed unexpected, in view of the
very good consistency between experimental and calculated curves
for 1 and 2 (we also notice that (R)-3-methylcycloheptanone gave
an approximately enantiomeric ORD with respect to that of 316,17).
Since the experimental ORD curves are similar for the three com-
pounds and their absolute values have the same order of magni-
tude, we considered a few possible sources of errors for compound
3. As previously shown, compound 3 has two main conformers,
both with the phenyl in equatorial conformation, similar in relative
populations (51.5% for conformer 3a and 44.0% for 3b) but differing
in cycloheptanone ring conformation (Fig. SI-2) (It is reassuring to
learn that also for (R)-3-methylcycloheptanone, important con-
former dependence of ECD spectra was observed.16,18). The con-
formation, which best fits experimental data is 3b, for this reason
we thought it was possible that the employed level of theory did
not properly describe the two conformations relative energies. We
thus increased basis sets level using (with B3LYP functional) aug-
cc-pVDZ and cc-pVTZ. Use of different basis sets did not lead to
significant changes (Table SI-2). Moreover use of aug-cc-pVDZ basis
set promoted conformer 3a to 62.1% (Fig. 3).
We only considered as input geometries for chiroptical proper-
ties calculations the conformations with relative energies below
2 kcal/mol depicted in Fig. 1. Unlike 1 and 2, the second populated
conformer for 3 has a population factor comparable to the first one,
being 0.85 of that.
3.2. Analysis of ORD data
Experimental ORD curves of 1e3 are very similar in sign, trend,
and approximate magnitude of OR values measured at four differ-
ent wavelengths (589, 546, 435, and 405 nm). Theoretical pre-
dictions failed somewhat to follow such simple similarity. A TDDFT/
Also adoption of the IEF-PCM19 solvent model in the optimiza-
tions (chloroform) at DFT/B3LYP/TZVP level reveals no significant
change in the relative population of conformers (Table SI-2). We
also performed ORD calculation with PCM model (chloroform) at

P. Scafato et al. / Tetrahedron 69 (2013) 10752e10762
10755
Fig. 3. Comparison of calculated (TDDFT/CAM-B3LYP/aug-cc-pVDZ on DFT/B3LYP/
TZVP input geometries) ORD curves for 3a (dashed blue line), 3b (dotted blue line) and
experimental (solid red line, chloroform).
TDDFT/CAM-B3LYP/aug-cc-pVDZ level (Table SI-3). The ORD curve
of 3a is still positive except for the [
a
]
value, which turns out
405
slightly negative (ꢀ3.56).
Another source of discrepancy between experimental and cal-
culated ORD curves could be the quite flat energy dependence from
the phenyl dihedral angle, which makes multiple orientations of
the phenyl group thermally accessible. Then a correct prediction of
ORD possibly requires those structures to be analyzed and com-
puted in order to be considered in the averaged property. We
performed a relaxed scan along the phenyl torsion C9eC8eC3eH3
(4) for both 3a and 3b (at DFT/B3LYP/TZVP level), with a stepsize of
20ꢁ of dihedral angle (Fig. 4).
Fig. 4. Numbering of dihedral angle 4 considered in performing optimizations scan of
1e3.
In a section of Supplementary data dedicated to the analysis of
ORD for (S)-3-phenylcycloheptanone, we have evaluated and
compared the energy and ORD dependence on 4 and compared
results for 3a and 3b with those for 1 and 2 (Table SI-4, Figs. SI-
3e5).
3.3. Analysis of ECD spectra
UV and ECD spectra were recorded in hexane and CCl4. ECD
spectra of 1e3 (Fig. 5) all show a negative and vibrationally resolved
(vide infra) Cotton effect (CE) centered at 300 nm, which is ascribed
Fig. 2. Comparison of calculated (dashed blue lines, TDDFT/CAM-B3LYP/aug-cc-pVDZ
on DFT/B3LYP/TZVP input geometries) and experimental (solid red lines, (1) and (2)
hexane, (3) chloroform) ORD curves of 1e3. Fixed AC are (3S)-1e3.
to ne
p carbonyl transition (this happens also for other substituted
*
cycloalkanones11,16). The vibronic structure fades away with in-
creasing ring dimensionality. Two positive CEs centered at 220 nm

10756
P. Scafato et al. / Tetrahedron 69 (2013) 10752e10762
and 196 nm (allied to 1La and 1Bb benzene transitions, respectively)
are present for all three molecules. ECD spectrum of 3 shows an
additional weak positive band at 230 nm ascribed to 1Lb transition
and a negative CE at about 186 nm is present in the ECD of 1. These
bands are allied to
pep benzene transitions. ECD calculations
*
were performed at TDDFT/CAM-B3LYP/aug-cc-pVDZ level of theory
both in gas phase and according to a hexane IEF-PCM calculation,
taking into account the first 30 excited states. Length and velocity
formalisms led to similar rotational strengths prediction; then only
velocity rotatory strengths based ECD shapes are displayed. Not-
withstanding a quite good prediction based on gas phase calcula-
tions (Fig. SI-6) we report in Fig. 5 ECD calculations based on
solvation model, allowing to better predict in magnitude the ex-
perimental spectra in the 190e200 nm range. Equatorial confor-
mations, of 1 and 2, drive negative sign of the ne
p CE, which is
*
consistent with the experimental ones and with the octant rule.16
The two equatorial conformations 3a and 3b possess opposite
ne
p sign and consistent with octant rule. Conformer 3a, which is
*
the most populated one, has a negative CE as in the experimental
spectrum (vide infra). The existence of multiple conformers was
postulated in Refs. 16e18 for (R)-3-methylcycloheptanone and was
demonstrated by running temperature dependent ECD spectra,
where better resolved vibronic features were evidenced at lower
temperatures.18 Calculations also exhibit weak positive CE associ-
ated with 1Lb transition at 240 nm, which are not easily visible in 1
and 2 experimental spectra. On the other hand positive CE of 3,
related to 1Lb transition, is predicted even if, due to the Boltzmann
averaging, its rotatory strength is very weak (þ0.17). It is important
to note that the nice matching between calculated and experi-
mental UV profiles leads to reliable and safe assignment of corre-
sponding ECD bands. Comparison of theoretical and experimental
ECD spectra of 1e3 shows good agreement with a full prediction of
experimental CEs in sign, position and intensities (Fig. 5).
On the side of ECD calculation it is important to note that con-
formers 3a and 3b have again an enantiomeric behavior (Fig. 6) and
the most populated conformer 3a is the one with the correct ECD
signs (cf. the results for (R)-3-methylcycloheptanone18) but the
correct intensity is obtained only considering both conformers.
Fig. 5. Experimental ECD (top red lines, hexane), experimental UV (bottom red lines,
hexane), and calculated ECD (top blue lines), UV (bottom blue lines) comparison
spectra of 1 (plot a), 2 (plot b), and 3 (plot c). Calculated ECDs were performed at
TDDFT/CAM-B3LYP/aug-cc-pVDZ (IEF-PCM, hexane) on DFT/B3LYP/TZVP input geom-
etries, 30 first excited states, 0.2 eV of bandwidth. All calculated spectra are 5 nm red
shifted, calculated UV and ECD of 1 have been divided by 2, those for 2 and 3 by 3 (see
text).
Fig. 6. Calculated ECD spectra comparison for conformers 3a (solid red line) and 3b
(solid blue line) at TDDFT/CAM-B3LYP/aug-cc-pVDZ-(IEF-PCM, hexane) level, on DFT/
B3LYP/TZVP-(IEF-PCM, hexane) calculated geometries, 30 first excited states, 0.2 eV
bandwidth.
In contrast to the ORD case, the results of ECD calculations are
off with respect to experiment by a smaller factor except for 3, since

P. Scafato et al. / Tetrahedron 69 (2013) 10752e10762
10757
the ECD spectrum is due to cancellation of two oppositely signed
ECD spectra for conformers 3a and 3b and the ECD spectrum of the
more populated conformer 3a better matches the experimental
data. We checked that phenyl rotation of þ/ꢀ10ꢁ does not affect
ECD calculation confidence, despite a difference in relative in-
tensities of rotatory strengths for the bands below 190 nm while
the effect on ORD for the same changes is quite dramatic (Fig. SI-7).
an impact on such spectral features. With this limitation in mind,
we can notice that the results obtained with harmonic approxi-
mation still match a number of experimental findings. Let us focus
on the spectra computed at 300 K. In agreement with experiment,
the spectra of the three species show a characteristic progression,
observed in experiments, that is due to the CO stretching. Two
additional features agree with experiment: (i) moving from 1 to 2
and 3 the peaks of the CO vibronic progression are less and less
‘resolved’, finally reducing to a weak modulation of a broad and
unstructured background in 3a/3b; (ii) the maxima of the 3a and 3b
spectra are blueshifted with respect to those of 1 and 2. The former
phenomenon (i) is, at least partially, due to the fact that the normal
The vibronic structure of ne
p
*
ECD bands in (R)-3-
methylcycloalkanones were discussed at length in Refs. 16 and 18.
In Fig. 7 we compare to experiment the calculated vibronically
resolved bands for 1e3 allied to the ne
p
*
low lying transition (we
considered only equatorial conformations of 1, 2, and 3 and, for the
Fig. 7. Comparison of experimental (left panels) and theoretical (right panels) absorption UV spectra of nep electronic transition of 1, 2 and 3 in a range from 3.5 to 5 eV calculated
*
within FCjAH approximation (see text).
latter systems, both 3a and 3b species). Before presenting the re-
sults, it should be emphasized that the ECD spectra of 1, 2, and 3
compounds mainly differ in width and resolution, and that the
specific properties of the manifold of the anharmonic states along
the CCC]O pyramidalization, here neglected, would probably have
modes that are related to the CO stretching exhibit an increased
Duschinsky mixing moving from 1 to 3, so that while a single
dominant progression is seen in 1, a larger number of (weaker)
vibronic transitions, with close but different energies, contribute to
the 3a/3b ECD lineshape. The same mechanism explains also the

10758
P. Scafato et al. / Tetrahedron 69 (2013) 10752e10762
blueshift of the 3a/3b maxima, but in this case also thermal exci-
tation plays an essential role and in fact the blueshift is not ob-
served at 10 K. The existence of two conformers of comparable
population represents an additional factor leading to a blurring of
the 3 spectrum. This is shown in the bottom panel of Fig. 7 com-
paring the ECD spectra of the 3a and 3b with their Boltzmann av-
erage (notice however that the computed Boltzmann populations
cannot be considered free of errors).
configurational assignment. Accordingly DFT calculations, which
provide good simulation of these bands in sign and magnitude,
allow to attribute the negative feature to phenyl CeH in-plane
bendings coupled to C
counterpart is associated to CH2 bending mode of units 4 for 1 and
units 4 and 5 for 2 in the cycloalkanone ring, coupled to little C eH
*
eH in-plane bending, whilst the positive
*
bending. This feature is also present in the deuterated d4 (Chart 2)
derivatives (10 and 20) of 1 and 2, which possess the same vibra-
tional normal modes of their non-deuterated counterparts (In Fig. 8
we show that both VA and VCD calculated spectra for 10 compare
also favorably with experiments.).
3.4. Analysis of VCD spectra
Proceeding with the analysis of VA/VCD spectra of 1, we notice
that the intense band numbered as 33 in the VA spectrum in Fig. 8 is
calculated at slightly different frequencies for the two conformers,
close to the weak features #34 and 35 and corresponds to the
multiple bands group in the experimental spectrum (from here on
we speak of band # and normal mode # interchangeably, when no
Experimental vibrational absorption (VA) and VCD spectra,
measured in CCl4 solutions were analyzed and compared with
calculated ones at DFT/B3LYP/TZVP level of theory considering (3S)-
1e3 as fixed AC. In Figs. 8e10 we compare experimental and pre-
dicted (for each most stable conformation and Boltzmann average)
VA and VCD spectra in the mid-IR region (900e1600 cmꢀ1). As
Fig. 8. Comparison of experimental and calculated VA (bottom panels) and VCD (top panels) spectra for 1 and deuterated 10 (see Chart 2). Experimental spectra (exp. traces) are
measured in CCl4. Calculated spectra of equatorial conformations 1a and 10a (deuterated equatorial conformer), axial conformations 1b and 10b (deuterated axial conformer) and
Boltzmann’s averaged spectra (avg traces) are computed at DFT/B3LYP/TZVP level.
a general qualitative comparison there is a clear matching between
experimental and calculated spectra, which are mainly determined
by the major conformer. In particular, we observe that a bisignate
confusion is caused, i.e., when one conformer is prevalent).
According to DFT calculation, the normal modes associated to band
#33 are mainly the CeC in-plane anti-symmetric stretching of the
CC bond pair in the alkanone ring sharing the carbonyl carbon
atom. The modes are also coupled to some in-plane benzene group
CeH bendings. It is interesting to note that bands 28 and 32, which
are present in the VCD of the axial conformer 1b, do not appear in
negativeepositive feature at 1450e1457 cmꢀ1 for 1,
2 and
1440e1453 cmꢀ1 for 3 appears in the VCD spectra, including the
deuterated counterparts of 1 and 2: for this reason we think that
this couplet has the characteristics of
a marker signal for

P. Scafato et al. / Tetrahedron 69 (2013) 10752e10762
10759
Fig. 9. Comparison of experimental and calculated VA (bottom panels) and VCD (top panels) spectra for 2 and deuterated 20 (see Chart 2). Experimental spectra (exp. traces) are
measured in CCl4. Calculated spectra of equatorial conformations 2a and 2’a (deuterated equatorial conformer), axial conformations 2b and 2’b (deuterated axial conformer) and
Boltzmann’s averaged spectra (avg traces) are computed at DFT/B3LYP/TZVP level.
the measured spectrum, confirming low experimental contribution
of axial conformation but with not as sufficient finesse as pre-
viously done by He et al.,20 where (R)-3-methylcyclopentanone was
investigated and the relative ratio of axial and equatorial con-
formers was determined.
belonging to 3a and 3b, respectively, and 44e46, for 3a and 3b,
respectively. The corresponding experimental VCD spectrum shows
very weak features in that range. At higher wavenumbers, the
calculated average VCD spectrum shows a bisignate couplet (at ca.
1350 cmꢀ1) arising from modes 54 and 56 almost exclusively from
In Fig. 9 comparison of experimental and calculated VA and VCD
conformer 3a. The normal modes involved are due to C eH bending
*
spectra of 2 and 20 is reported.
and CH2 twistings in methylene units of the cycloheptanone ring.
Even if very weak, in the experimental VCD spectrum, we note
a non-conservative bisignate couplet at 1250 cmꢀ1 ascribed to CH2
twistings. The positive band, referred to mode 48 results from 3a
conformation whilst weaker negative 47 band is from conformer
3b. Negative CE numbered 30 (rock CH2 bendings and cyclo-
heptanone ring deformation) is also well predicted. We have
checked the sensitivity of the VCD technique to phenyl moiety
Main VA and VCD experimental features are well predicted. The
predicted VCD intense negative band 43 at 1230 cmꢀ1 for the axial
conformation (related to some CH2 wagging on the cyclohexanone
ring) does not appear in experimental spectrum, confirming too
low contribution of axial conformation to be detected, within the
limits of the quality of our experiments. Possibly also normal modes
# 47 (þVCD) and # 48 (ꢀVCD) of the axial conformer of the d4
homologue manifest themselves in the experimental VCD spectra.
However these conclusions are only qualitative: we notice that in
the analogous case of (R)-3-methyl cyclohexanone, the use of mid-
IR VCD allowed to quantitatively define the relative content of axial
versus equatorial conformation. In this case we have a lower con-
tribution of axial conformer than the one at 10% as found by Devlin
and Stephens.21
rotation of 3 by considering structures 3
a comparison of experimental VA and VCD spectra with the cal-
culated spectra for separate conformations 3a, 3 and 3 and their
Boltzmann average is presented. Both 3 and 3 VA spectra appear
quite similar but broadening of band group 38e40 manifests in 3
which relates to a CeC stretching affecting the stereogenic carbon
atom. In the VCD trace of 3 a positive band 41 allied to C eH
a and 3b. In Fig. SI-8
a
b
a
b
b
,
b
*
The VA spectrum of 3 in the 900e1300 cmꢀ1 region (Fig. 10)
presents weak signals. Anyway in correspondence of experimental
bands at ca. 1150 and 1200 cmꢀ1 it is possible to find in the cal-
culated average spectrum the couples of bands 41 and 42,
bending creeps in. It appears that vibrational local modes affecting
the chiral carbon atom may somewhat change VCD response.
Anyway the overall matching with the VA and VCD data for most
stable 3a is acceptable.

10760
P. Scafato et al. / Tetrahedron 69 (2013) 10752e10762
Fig. 10. Comparison of experimental and calculated VA (bottom panel) and VCD (top panel) spectra for 3. Experimental spectra (exp. traces) are measured in CCl4. Calculated spectra
of equatorial conformations 3a and 3b and Boltzmann’s averaged spectra (avg traces) are computed at DFT/B3LYP/TZVP level.
even though it was the focus of the first VCD investigations.22,23 In
Figs. 11e13 comparison of calculated and experimental spectra for
(S)-1, 2, and 3, respectively, is reported. VCD features in the range of
2900e3000 cmꢀ1 are generated by cycloalkanones methylene CeH
stretchings. In compound 1 (Fig. 11), according to DFT/B3LYP/TZVP
calculation, the experimental VCD positive feature at 2900 cmꢀ1
arises from CeH stretching of axial hydrogen in C5 position of 1a.
The negative experimental CE at 2970 cmꢀ1 corresponds to the
negative band of conformer 1a related to symmetric stretchings of
axial hydrogen at C4 and equatorial hydrogen at C5. Axial con-
former 1b provides enantiomeric VCD behavior to equatorial con-
former 1a, and this had been recognized some time ago on a semi-
Chart 2. Structures of deuterated homologues of compounds 1 and 2.
empirical basis:24,25 since the experimental VCD spectrum is jus-
tified by 1a data, we have another proof of the minor importance of
the axial conformer.
Finally we investigated the CeH stretching fundamental region
(2800e3100 cmꢀ1), a region, which has been recently overlooked,
Also in the case of 2 (Fig. 12) the experimental VCD spectrum
reveals a (ꢀ,þ,ꢀ) structure between 2900 and 2970 cmꢀ1, which

P. Scafato et al. / Tetrahedron 69 (2013) 10752e10762
10761
Fig. 12. Comparison of experimental and calculated VA (bottom panel) and VCD (top
panel) spectra for 2 in the CeH stretching region (2800e3100 cmꢀ1). Experimental
spectra (exp. traces) are measured in CCl4. Calculated spectra of equatorial and axial
conformations 2a and 2b respectively and Boltzmann’s averaged spectra (avg traces)
are computed at DFT/B3LYP/TZVP level.
Fig. 11. Comparison of experimental and calculated VA (bottom panel) and VCD (top
panel) spectra for 1 in the CeH stretching region (2800e3100 cmꢀ1). Experimental
spectra (exp. traces) are measured in CCl4. Calculated spectra of equatorial and axial
conformations 1a and 1b respectively and Boltzmann’s averaged spectra (avg traces)
are computed at DFT/B3LYP/TZVP level.
has been known for 30 years for (R)-3-methylcyclohexanone24
(as opposite) and attributed to concerted normal modes involving
the stretching of C eH and the anti-symmetric stretchings of the
*
two CH2 units 4 and 5 near the stereogenic carbon atom. Later this
situation was further studied and called as vibrational excitons.25
The three units are mirror images of each other in axial (2b) and
equatorial (2a) conformers and this justifies the calculated VCD
spectra being opposite. Correspondence of experimental and cal-
culated VCD for 2a rules out the presence of 2b (as already obtained
by mid-IR VCD, ECD, and ORD).
In the case of 3 ((S)-3-phenylcycloheptanone) (Fig. 13) the VCD
behavior in this region for 3a and 3b is enantiomeric. Since the
population factors of 3a and 3b are similar, this leads to almost
a cancellation of the two VCD traces. However 3a is slightly more
populated than 3b and this allows to predict right signs for the VCD
spectrum; and also justifies why one does not observe the usual
spectrum with alternating signs.24,25 Together with ORD, VCD in
the CH-stretching region provides the most compelling evidence of
the conformational equilibria in the cycloheptanone ring. Further
data on d4-isotopomers in the CD stretching region for 1 and 2 are
presented in Supplementary data.
4. Conclusions
In the present work we have investigated the configuration and
conformational properties of (S)-3-phenylcyclopentanone (1), (S)-
3-phenylcyclohexanone (2), and (S)-3-phenylcycloheptanone (3)
and some deuterated species thereof. Indeed we have noticed that,
notwithstanding the copious literature on the synthesis of these
compounds, no complete analysis, based on advanced chiroptical
Fig. 13. Comparison of experimental and calculated VA (bottom panel) and VCD (top
panel) spectra for 3 in the CeH stretching region (2800e3100 cmꢀ1). Experimental
spectra (exp. traces) are measured in CCl4. Calculated spectra of equatorial confor-
mations 3a and 3b and Boltzmann’s averaged spectra (avg traces) are computed at
DFT/B3LYP/TZVP level.

10762
P. Scafato et al. / Tetrahedron 69 (2013) 10752e10762
and conformational studies, exists.26 All chiroptical data for 1, 2,
and 3, with one exception, look similar, namely experimental ECD
spectra are almost identical for 1, 2, and 3; experimental VCD and
VA spectra in the mid-IR for 1, 2, and 3 show common features and
ORD curves are quite similar for 1, 2, and 3; only in the CH-
stretching region the VCD data for 3 look special. From this we
conclude that the configuration is the same for the three molecules,
however the conformational properties are similar, though not
identical, as predicted at DFT level and confirmed by good com-
parison with experiments. Indeed the ECD and VCD spectra are
correctly interpreted by DFT calculations in terms of a prevalent
percentage of the equatorial conformer of the phenyl moiety with
the ring being half-chair for 1, stable chair for 2, and twisted
pseudo-chair conformers for 3. However, while in the former two
cases the prevalent conformer has a major importance being about
95% of the total, in the last case there is a second conformer with
inferior but similar statistical weight and opposite chiroptical
properties. The ORD and VCD data in the CH-stretching region are
the most sensitive to these conformational details.
507e510; (l) Marchesan, D.; Coriani, S.; Forzato, S.; Nitti, P.; Pitocco, G.; Ruud, K.
J. Phys. Chem. A 2005, 109, 1449e1453; (m) Cheng, M.; Li, Q.; Lin, B.; Sha, Y.; Ren,
J.; He, Y.; Wang, Q.; Hua, H.; Ruud, K. Tetrahedron: Asymmetry. 2006, 17,
179e183; (n) Voloshina, E.; Fleischhauer, J.; Kraft, P. Helv. Chim. Acta 2005, 88,
194e209; (o) Sebert, S.; Konig, G. M.; Voloshina, E.; Raabe, G.; Fleischhauer, J.
Chirality 2006, 18, 413e418; (p) Mazzeo, G.; Santoro, E.; Andolfi, A.; Cimmino,
A.; Troselj, P.; Petrovic, A. G.; Superchi, S.; Evidente, A.; Berova, N. J. Nat. Prod.
2013, 76, 588e599.
4. (a) Stephens, P. J.; Pan, J. J.; Devlin, F. J.; Cheeseman, J. R. J. Nat. Prod. 2008, 71,
285e288; (b) Mazzeo, G.; Giorgio, E.; Zanasi, R.; Berova, N.; Rosini, C. J. Org.
Chem. 2010, 75, 4600e4603; (c) Mazzeo, G.; Giorgio, E.; Rosini, C.; Fabris, F.;
Fregonese, E.; Toniolo, U.; Lucchi, O. D. Chirality 2009, 21, E86eE97.
5. (a) Scafato, P.; Caprioli, F.; Rosini, C. Tetrahedron: Asymmetry 2011, 22, 558e561;
(b) Takaya, Y.; Ogasawara, M.; Hayashi, T.; Sakai, M.; Miyaura, N. J. Am. Chem.
Soc. 1998, 120, 5579e5580.
6. SPARTAN 02. Wavefunction: Irvine, CA.
7. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.;
Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; Na-
katsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng,
G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.;
Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery,
J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K.
N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A.;
Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, J. M.; Klene, M.;
Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.;
Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.;
Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dan-
Acknowledgements
€
nenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.; Foresman, J. B.; Ortiz, J. V.;
Cioslowski, J.; Fox, D. J. Gaussian 09, Revision A.02; Gaussian, Inc.: Wallingford,
CT, 2009.
G.M. and G.L. wish to thank Cariplo Foundation, Milan, Italy, for
funding a post-doc position. F.S. acknowledges support from MIUR
(PRIN 2010e2011 prot. 2010ERFKXL).
8. (a) Yanai, Y.; Tew, D. P.; Handy, N. C. Chem. Phys. Lett. 2004, 393, 51e57; (b)
€
Paterson, M. J.; Christiansen, O.; Paw1owski, F.; Ju˛rgensen, P.; Hattig, C.; Hel-
gaker, T.; Sa1ek, P. J. Chem. Phys. 2006, 124, 054322e054332; (c) Peach, M. J. G.;
Helgaker, T.; Sa1ek, P.; Keal, W.; Lutns, O. B.; Tozer, D. J.; Handy, N. C. Phys. Chem.
Chem. Phys. 2006, 8, 558e562.
Supplementary data
€
9. Bruhn, T.; Hemberger, Y.; Schaumloffel, A.; Bringmann, G. SpecDis Version 1.53;
University of Wuerzburg: Wuerzburg, Germany, 2011.
10. Santoro, F.; Improta, R.; Lami, A.; Bloino, J.; Barone, V. J. Chem. Phys. 2008, 128,
224311e224328.
11. Lin, N.; Solheim, H.; Ruud, K.; Nooijen, M.; Santoro, F.; Zhao, X.; Kwit, M.;
Skowronek, M. Phys. Chem. Chem. Phys. 2012, 14, 3669e3680.
12. Santoro, F. FCclasses a Fortran 77 code. See: http://village.pi.iccom.cnr.it [last
accessed 10.07.13]; (b) Cerezo, F. “TDspectrum” a routine for TD calculations
within FCclasses.
13. Avila Ferrer, F. J.; Santoro, F. Phys. Chem. Chem. Phys. 2012, 14, 13549e13563.
14. Santoro, F.; Barone, V. Int. J. Quantum Chem. 2010, 110, 624e636.
15. Tatchen, J.; Pollack, E. J. Chem. Phys. 2008, 128, 164303e164317.
16. Lightner, D. A.; Gurst, J. E. Organic Conformational Analysis and Stereochemistry
from Circular Dichroism Spectroscopy; Wiley-VCH: New York, NY, 2000.
17. Djerassi, C.; Krakower, G. W. J. Am. Chem. Soc. 1959, 81, 94e101.
18. Lightner, D. A.; Docks, E. L. Tetrahedron 1979, 35, 713e720.
19. Mennucci, B.; Tomasi, J. J. Phys. Chem. A 2002, 106, 6102e6113.
20. He, J.; Petrovic, A. G.; Polavarapu, P. L. J. Phys. Chem. B 2004, 108, 20451e20457.
21. Devlin, F. J.; Stephens, P. J. J. Am. Chem. Soc. 1999, 121, 7413e7414.
22. (a) Holzwarth, G.; Hsu, E. C.; Mosher, H. S.; Faulkner, T. R.; Moscowitz, A. J. Am.
Chem. Soc. 1974, 96, 251e252; (b) Nafie, L. A.; Cheng, J. C.; Stephens, P. J. J. Am.
Chem. Soc. 1975, 97, 3842e3843; (c) Nafie, L. A.; Keiderling, T. A.; Stephens, P. J. J.
Am. Chem. Soc. 1976, 98, 2715e2723.
Supplementary data associated with this article can be found in
the online version, at http://dx.doi.org/10.1016/j.tet.2013.10.016.
References and notes
1. The general theory of optical activity is given by (a) Rosenfeld, L. Z. Phys. 1928,
52, 161e171; (b) Condon, E. U. Rev. Mod. Phys. 1937, 9, 432e457; (c) Buckingham,
A. D. Adv. Chem. Phys. 1967, 12, 107e142.
2. For general discussion of the ab initio calculation of chiroptical properties: (a)
Comprehensive Chiroptical Spectroscopy: Applications in Stereochemical Analysis
of Synthetic Compounds, Natural Products, and Biomolecules; Berova, N., Pola-
varapu, P. L., Nakanishi, K., Woody, R. W., Eds.; John Wiley & Sons: Hoboken, NJ,
USA, 2012; (b) Autschbach, J. Chirality 2009, 21, E116eE152; (c) Autschbach, J.;
Nitsch-Velasquez, L.; Rudolph, M. Top. Curr. Chem. 2011, 298, 1e98; (d) Yang, G.;
Xu, Y. Top. Curr. Chem. 2011, 298, 189e236.
3. (a) Stephens, P. J.; McCann, D. M.; Butkus, E.; Stoncius, S.; Cheeseman, J. R.;
Frisch, M. J. J. Org. Chem. 2004, 69, 1948e1958; (b) Stephens, P. J.; McCann, D.
M.; Devlin, F. J.; Cheeseman, J. R.; Frisch, M. J. J. Am. Chem. Soc. 2004, 126,
7514e7521; (c) Stephens, P. J.; McCann, D. M.; Devlin, F. J.; Flood, T. C.; Butkus,
E.; Stoncius, S.; Cheeseman, J. R. J. Org. Chem. 2005, 70, 3903e3913; (d) Ste-
phens, P. J.; McCann, D. M.; Devlin, F. J.; Smith, A. B., III. J. Nat. Prod. 2006, 69,
1055e1064; (e) Stephens, P. J.; Pan, J. J.; Devlin, F. J.; Krohn, K.; Kurtan, T. J. Org.
Chem. 2007, 72, 3521e3536; (f) Stephens, P. J.; Pan, J. J.; Devlin, F. J.; Urbanova,
M.; Hajicek, J. J. Org. Chem. 2007, 72, 2508e2524; (g) Polavarapu, P. L.; He, J.;
Crassous, J.; Ruud, K. ChemPhysChem 2005, 6, 2535e2540; (h) Petrovic, A. G.;
He, J.; Polavarapu, P. L.; Xiao, L. S.; Armstrong, D. W. Org. Biomol. Chem. 2005, 3,
1977e1981; (i) Petrovic, A. G.; Polavarapu, P. L.; Drabowicz, J.; Zhang, Y.;
McConnell, O. J.; Duddeck, H. Chem.dEur. J. 2005, 11, 4257e4262; (j) Zuber, R.;
Goldsmith, M.; Hopkins, T. D.; Beratan, D.; Wipf, P. Org. Lett. 2005, 23,
5269e5272; (k) Zuber, R.; Goldsmith, M.; Beratan, D.; Wipf, P. Chirality 2005, 17,
23. (a) Longhi, G.; Abbate, S.; Gangemi, R.; Giorgio, E.; Rosini, C. J. Phys. Chem. A
2006, 110, 4958e4968; (b) Abbate, S.; Burgi, L. F.; Gangemi, F.; Gangemi, R.;
Lebon, F.; Longhi, G.; Pultz, V. M.; Lightner, D. A. J. Phys. Chem. A 2009, 113,
11390e11405.
24. Laux, L.; Pultz, V. M.; Abbate, S.; Havel, H. A.; Overend, J.; Moscowitz, A.;
Lightner, D. A. J. Am. Chem. Soc. 1982, 104, 4276e4278.
25. Gangemi, R.; Longhi, G.; Lebon, F.; Abbate, S.; Laux, L. Monatsh. Chem. 2005, 136,
325e345.
26. (a) Hayashi, T.; Yamazaki, K. Chem. Rev. 2003, 103, 2829e2844; (b) Iuliano, A.
Tetrahedron: Asymmetry 2010, 21, 1943e1958.