Synthesis, chiroptical properties and density functional theory calculations of 3,3'-Biphenyl-2,2'-BiTropone

Article abstract of DOI:10.1002/chir.22191

The synthesis of new bitropone derivatives, namely, 3,3'-biphenyl-2,2'- bitropone and 7,7'-biphenyl-2,2'-bitropone, are reported. Isolation of enantiomers arising from restricted rotation around the C-C bond connecting the tropone moieties was attempted by means of chiral high performance liquid chromatography (HPLC). No separation was obtained for 7,7'-biphenyl-2,2'- bitropone. For 3,3'-biphenyl-2,2'-bitropone, difficulties were encountered because of the low separation factor of the peaks and the presence of a rapid racemization process. However, quantitative chiroptical data on the antipodes were obtained by linking a circular dichroism (CD) spectrometer and a UV-vis spectrophotometric detector in series to the HPLC instrument. The analysis of the CD and UV-vis spectra in terms of absolute conformations was done with the help of theoretical calculations performed at the Density Functional Theory (DFT) level. The most stable conformations of the 3,3'-biphenyl-2,2'-bitropone in its ground state were obtained. Starting from these minimum energy conformations, it was possible to compute theoretical CD and UV absorption spectra that fit well with the experimental ones. From this comparison the absolute configuration to the antipodes was assigned. Finally, the effect of the presence of the two lateral phenyl substituents on the structure of the bitropone and hence on the CD spectrum is discussed. Chirality 25:648-655, 2013. 2013 Wiley Periodicals, Inc.

Full text of DOI:10.1002/chir.22191

CHIRALITY 25:648–655 (2013)
Synthesis, Chiroptical Properties and Density Functional Theory
Calculations of 3,3’-Biphenyl-2,2’-BiTropone
1
1
1
1
1
*
MARINO CAVAZZA, MARIO CIFELLI, VALENTINA DOMENICI, TIZIANA FUNAIOLI, BENEDETTA MENNUCCI,
CARLO ALBERTO VERACINI,1 AND MAURIZIO ZANDOMENEGHI1,2
*
1Dipartimento di Chimica e Chimica Industriale, via Risorgimento 35, 56126 Pisa, Italy
2Istituto di Chimica dei Composti OrganoMetallici (ICCOM-CNR), UOS Pisa -Via Moruzzi, 1 56124 Pisa, Italy.
ABSTRACT
The synthesis of new bitropone derivatives, namely, 3,3’-biphenyl-2,2’-bitropone
and 7,7’-biphenyl-2,2’-bitropone, are reported. Isolation of enantiomers arising from restricted
rotation around the C-C bond connecting the tropone moieties was attempted by means of chiral
high performance liquid chromatography (HPLC). No separation was obtained for 7,7’-biphenyl-
2,2’-bitropone. For 3,3’-biphenyl-2,2’-bitropone, difficulties were encountered because of the low
separation factor of the peaks and the presence of a rapid racemization process. However, quan-
titative chiroptical data on the antipodes were obtained by linking a circular dichroism (CD)
spectrometer and a UV–vis spectrophotometric detector in series to the HPLC instrument.
The analysis of the CD and UV–vis spectra in terms of absolute conformations was done with
the help of theoretical calculations performed at the Density Functional Theory (DFT) level.
The most stable conformations of the 3,3’-biphenyl-2,2’-bitropone in its ground state were
obtained. Starting from these minimum energy conformations, it was possible to compute theo-
retical CD and UV absorption spectra that fit well with the experimental ones. From this compar-
ison the absolute configuration to the antipodes was assigned. Finally, the effect of the presence
of the two lateral phenyl substituents on the structure of the bitropone and hence on the CD
spectrum is discussed. Chirality 25:648–655, 2013. © 2013 Wiley Periodicals, Inc.
KEY WORDS: atropoisomerism; HPLC; separation, racemization; CD; DFT; UV–vis; tropone;
enantiomers
INTRODUCTION
transitions dipoles, to which CD is very sensitive.8,10 Hence,
a precise molecular geometry is the prerequisite to calculate
reliable theoretical UV and CD spectra. As a matter of fact,
QM-based computations of the CD spectra of both 4 and 5
based in the most stable QM-conformations resulted in unsat-
isfactory results.8
The intrinsic optical activity due to the (potential)
atropisomerism11 of the bitropone moiety present in 1, as
well in compounds 4, 5, and 6, is an interesting aspect of
all these systems. Up to now the importance of this mecha-
nism of CD generation has been unclear.
In order to investigate these interesting issues from an ex-
perimental as well as theoretical point of view, we decided to
focus on a simpler bitropone derivative to be used as a molec-
ular model. As our attempts to resolve optically active enantio-
mers of bitropone 1, as reported in Figure 1, via chiral high
performance liquid chromatography (HPLC) failed, we
decided to study the 3,3’-biphenyl-2,2’-bitropone 7, a substituted
bitropone system in which bulky substituents, namely, phenyl
groups, were placed in the 3 and 3’ positions of the tropone-
rings connected to each other. Clearly, from the chirality point
Contrary to tropone and its derivatives (troponoids),1,2
which attract interest for their potential in many fields from
agriculture to medicine,1 bitropones 1, 2, and 3 (Fig. 1) have
attracted little attention by chemists, since the first pre-
paration reported by Iyoda et al. in 1985.3 In recent years,
liquid crystals and organogelators with a bitropone core
(of type 3 in Fig. 1) have been synthesized4 and electro-
photographic photoconductors containing bitropone deriva-
tives as electron-transporting agents have been prepared.5,6
An efficient new synthesis of 2,2’-bitropone (1) has been
recently proposed.7
For instance, a new class of asymmetric pseudoaromatic
compounds containing the 2,2’-bitropone core, namely, the
9,9’-biisocolchicide (as an equilibrium mixture of conformers
4 and 5) and 10,10’-bicolchicide (6) (Fig. 2), has been
recently investigated by some of us.8 Interestingly, these
compounds show intense and complex circular dichroism
(CD) spectra in the 450–200 nm spectral region and their
CD resulted in being strongly dependent on the solvent.8 In
these dimeric species (see Fig. 2), the main sources of confor-
mational mobility are localized into the seven-carbon rings in
which the asymmetric carbon is present and in the rotation
angle around the C-C bond interconnecting the colchicides
or isocolchicides moieties. Mobility in these seven-carbon
rings, e.g., in the equilibrium between conformers 4 and 5,
gives rise to mutarotation phenomena due to helicity inver-
sion.8,9 In terms of spectroscopic properties, even a small
deformation of the saturated seven-carbon rings or variation
of the dihedral angle around the linking C-C bond can result
in a large change of the relative orientation/distance of the
© 2013 Wiley Periodicals, Inc.
Additional Supporting Information may be found in the online version of this
article.
*Correspondence to: Valentina Domenici or Maurizio Zandomeneghi,
Dipartimento di Chimica e Chimica Industriale, Università di Pisa,v. Risorgi-
mento 35, 56126 Pisa, Italy.
E-mails: valentin@dcci.unipi.it or maurizio.zandomeneghi@gmail.com.
Received for publication 25 January 2013; Accepted 25 April 2013
DOI: 10.1002/chir.22191
Published online 5 July 2013 in Wiley Online Library
(wileyonlinelibrary.com).

SYNTHESIS AND CHIROPTICAL PROPERTIES OF 3,3’-BIPHENYL-2,2’-BITROPONE
649
Fig. 1. Structures of bitropones 1, 2, and 3.
Fig. 3. Schematic representation of a possible equilibrium between enan-
tiomers of 3,3’-biphenyl-2,2’-bitropone (7).
of view, atropisomerism is possible in these systems, as repre-
sented in Figure 3.
literature12 starting from tropolone. Aldrich preparative silica
gel TLC plates 20 x 20 cm and 2-mm layer thickness were
used.
The selected system 7 resulted in a good compromise
between the need to have a bitropone “model” for a success-
ful and satisfactory computation of its molecular structures
by the available theoretical methods and the experimental
requirement to be able to separate chiral enantiomers by
HPLC and to study their CD/UV spectra.
Instruments
A JascoV-550 UV–vis spectrophotometer and a Jasco J40AS
spectropolarimeter were used for the UV and CD absorption
measurements, respectively.
In this article, we present the synthesis and the chiroptical
behavior of 7. First, we shall prove experimentally that 7 is
given as a racemic mixture of conformational enantiomers
of the type shown in Figure 3. Separation of these enantio-
mers was attempted through chiral HPLC at various tempera-
For HPLC determinations we used a Jasco 880-PU pump,
equipped with a Shimadzu SPD-10A UV detector having a 6
nm spectral bandwidth (SBW) and an 8-mL flow cell with an
optical path of 10 mm. In the case of synchronous CD and
UV instantaneous detection, we connected the HPLC chiral
column out end to the J40AS spectropolarimeter equipped
with a home-made CD flow cell, whose out end was
connected to the above SPD-10A UV detector. Connections
were made by PEEK tubes, 0.25mm internal diameter. The
CD flow cell was obtained from a Jasco front loading cassette
HPLC UV disassemblable analytical cell, standard in the
HPLC Jasco 875-UV detectors, 8 mL capacity and 10 mm opti-
cal path. This cell has been modified by enlarging the 1 mm
cell diameter up to 2 mm and maintaining the 10 mm optical
path. The dead volume resulting was low and was evaluated
as 74 mL in the first PEEK tube (input tube) and 32 mL in
the enlarged cell, so that the estimated time necessary to fill
both the tube and the CD cell is about 8.5 s (2.6 s to transit
the cell) with a flux of 0.75 mL/min. With this enlarged cell
diameter the light intensity of the J40AS machine was suffi-
cient to obtain good CD measurements with acceptable noise
at sensitivity S = 1 cm elongation/millidegree ellipticity, SBW
2 or 4 nm, and absorbance up to 1.5 AU. We successfully
tested this cell in static conditions by measuring the CD of
(À)-2,2’-binaphthol in methanol and in HPLC flow conditions
by injecting (Æ)-2,2’-binaphthol, without detecting any
noxious birefringence or scattering phenomena.
ꢀ
tures, in the range 0–35 C and partial success was obtained
ꢀ
at temperatures lower than 15 C. However, isolation of con-
formers is made difficult by the strong overlapping of the very
large and asymmetric chromatographic peaks and by a rapid
racemization process, occurring even at the lowest tempera-
ꢀ
tures (near 0 C) compatible with the chiral columns used.
The racemization kinetics were also studied, and reliable
chiroptical data on the antipodes were obtained by coupling
a CD spectrometer and a spectrophotometric detector in
series to the HPLC machine, an instrument arrangement that
minimizes the time lag between the exit of 7 from the HPLC
column and its chiral analysis, hence minimizing the effects
of racemization processes. Finally, the interpretation of these
spectra and the stereochemical assignment of the enantio-
mers of 7 was done thanks to the support of Density
Functional Theory (DFT) calculations.
MATERIALS, INSTRUMENTS, AND METHODS
Materials
Solvents used in HPLC measurements were of HPLC qua-
lity from Carlo Erba. Tropolone, Ni(COD)2, thionyl chloride
(SOCl2), phenylboronic and phenylboronic-d5 acids, enantio-
meric and racemic 2,2’-binaphthol were purchased from
Aldrich and used as is. 2-Chloro-3-phenyltropone and 2-
chloro-7-phenyltropone were prepared according to the
The chiral column generally used was a 250 x 4.6 mm
Daicel Chiralcel OJ, with cellulose ester derivative as absor-
bent. The column temperature was controlled by immersion
in a home-made thermostatic bath in the range 0–40 ꢀC. Glass
tubes at ca. –10 ꢀC were used to collect samples from the chro-
matographic peaks. A 250 x 4.6 mm HPLC chiral column Pirkle-
DNPG (absorbent: dinitrobenzoyl phenylglicine ionically
bound to aminopropylsilica) was used too.
For analytical/semipreparative work a 250 x 4.6 mm Wa-
ters RP Spherisorb 5 mM ODS2 column was used. Retention
time for 7 was 29 min, while the retention time of 10 was
20.5 min with a flux 0.5 ml/min and mobile phase CH3CN/
H2O 65: 35.
HPLC analysis of 7was performed also with a Merck 250 x 4 mm
Lichrospher CN column, eluent ethyl acetate / n-hexane 60:40, 1
mL/min flux. In this case the retention time was 6.3 min.
Chirality DOI 10.1002/chir
Fig. 2. Structures of biisocolchicides 4 and 5, and bicolchicide 6.

650
CAVAZZA ET AL.
Substrates Synthesis
7.57 (2H, m). 13C NMR: d 128.4, 129.1, 132.0, 133.0, 134.3,
141.1, 151.0, 153.9. UV(CH3CN), lmax, nm (e Á 10À4):222
(3.6), 275 (1.45), 315 (1.5) with a shoulder at 390 (0.7). ESI-
MS: m/z 363.2 [M + H]+, 385.1 [M + Na]+. Anal. Calcd. for
C26H18O2: C, 86.16; H, 5.01. Found C, 86.15; H 5.0.
3,3’-biphenyl-2,2’-bitropone 7. 2-chloro-3-phenyltropone (8)12
(0.058 g, 0.267 mmol) was added to a solution of Ni(COD)2
(0.0528 g, 0.19 mmol) in 7 mL of DMF under an argon atmo-
sphere (Fig. 4). After 4 h at room temperature (RT), the dark
solution was evaporated under vacuum to give a gummy res-
idue that was subjected to silica gel TLC, eluting with CHCl3/
CH3COCH3 9.5: 0.5. The band at Rf = 0.15 was repeatedly
extracted with CHCl3 to give 7 as a colorless solid, melting
point (mp) 175–178 ꢀC (0.024g, 0.068 mmol, 50% yield).
3-phenyltropone (9)12 was recovered from band at Rf = 0.43
as a colorless solid mp 113ꢀC (0.0168 g, 0.092 mmol, 34.4%
yield).
Chiral HPLC Methodologies
In the case of 7, the mobile phase that resulted in the best
compromise among chiral efficiency and retention times was
the 82.5:17.5 mixture of n-hexane and isopropyl alcohol (IPA).
The best results in terms of resolution of optically actꢀive spe-
cies were obtained at temperatures between 0 and 5 C with
mobile phase flux 0.75 mL/min. These experimental condi-
tions were chosen in the limits of operating flux, temperature,
and pressure of our OJ column.
Spectral data for 7:. 1H NMR (400 MHz CD3COCD3): d 6.74
(2H, d, J = 11.52 Hz), 6.86 (2H, m), 6.86 (2H, ddd, J =11.6,
11,5, 1.12 Hz), 6.99 (2H, d, J = 11.88 Hz), 7.12 (2H, ddd,
J = 11.6, 11.88, 1.12 Hz), 7.24 (4H, m), 7.32 (4H, m). 13C
NMR (50.3 MHz CDCl3): d 127.7, 128.2, 128.6, 132.1, 135.1,
139.4, 141.3, 142.1, 147.7, 162.8. ESI-MS: m/z = 363.2
[M + H]+, 385.4 [M + Na]+. Anal. Calc d. for C26H18O2 : C,
86.16; H, 5.01. Found C, 86.20; H, 5.0.
Semipreparative chromatography was performed collecting
samples from the chromatographic peaks in glass tube at
ꢀ
about À10 C.
The instantaneous measurements of both circular dichro-
ism and absorbance on the chromatographic fluid exiting
the chiral column, at a chosen wavelength, were performed
exploiting the alternative approach previously described.
In the attempt to separate atropisomers of 1 and 10, mobile
phases of widely different composition were used, also forc-
ing the OJ column to work at À2 Cꢀ and pressure up to
50 bar. In any case, we had no separation of stereoisomers
and a single broad peak was obtained. As an example, in
the case of 10, retention times of 41 and 90 min at a flux of
0.65 mL/min with n-hexane/IPA 40:60 and 60:40 were
obtained, respectively.
The 1H and 13C NMR assignments were facilitated by com-
parison with the spectra of 3,3’-biphenyl-2,2’-bitropone-d10
which was prepared following the above procedure for 7
starting from 2-chloro-3-phenyltropone-d5. This compound
was prepared from phenyltropolone-d5 by reaction with
SOCl2.12 7-phenyltropolone-d5 was prepared according to
the method of Piettre et al.13 using phenylboronic-d5 acid in
1
the place of the corresponding light phenylboronic acid. H
In the case of 7, a partial separation of enantiomers was
obtained also operating with a chiral Pirkle column. Optimal
working conditions were slightly different with respect to
the Daicel OJ column. In this case, the best results were
obtained operating at the same temperatures of OJ, but with
80:20 an n-hexane/IPA mixture as mobile phase. With this
column the retention times were considerably longer than
with the Daicel OJ column, without any improvement of the
resolution of optically active species.
NMR (400 MHz CD3COCD3): d 6.74(2H, d, J = 11.52 Hz),
6.86 (2H, ddd, J =11.6, 11,5, 1.12 Hz), 6.99 (2H, d, J = 11.88
Hz), 7.12 (2H, ddd, J = 11.6, 11.88, 1.12 Hz). 13C NMR (50.3
MHz CDCl3): d 132.1, 135.1, 139.4, 141.3, 142.1, 147.7, 162.8.
7,7-biphenyl-2,2-bitropone
(10). 7,7-biphenyl-2,2-bitropone
(10) (Fig. 5) was prepared according to the synthesis of 7.
Starting from 2-chloro-7-phenyltropone12 (0.065 g, 0.30
mmol), Ni(COD)2 (0.059g, 0.21 mmol) 10 was obtained as
an orange material (0.051g, 0141 mmol, 94% yield) after TLC
purification of the reaction mixture (CHCl3/CH3COCH3 9.5:
0.5, Rf 0.60).
Computational Details
All geometries were optimized at the DFT level using the
M06-2X14 functional and the 6-311G(d) basis set. The elec-
tronic transitions and the related spectroscopic properties
were obtained at the TD-DFT level using the same functional
used in the geometry optimization but in combination with a
more extended basis set, namely, 6-311++G(d,p). The simu-
lated UV–vis and CD spectra were obtained by convoluting
the bands due to all the transitions up to 200 nm: each band
was modeled using a Gaussian bandshape with a fixed band-
width (0.3 eV). All calculations were performed with the
Gaussian09 suite of programs.15.
Spectral data for 10. 1H NMR(200 MHz, CDCl3): d 6.90 (2H,
ddd, J = 11, 10, 1 Hz), 6.86 (2H, ddd, J = 11,10, 1 Hz), 7.37
(2H, dd, J = 10, 1 Hz), 7.38 (2H, J = 10, 1 Hz), 7.39 (8H, m),
Fig. 4. Synthesis of 3,3’-biphenyl-2,2’-bitropone (7).
RESULTS AND DISCUSSION
Chirality of 3,3’-Biphenyl-2,2’-bitropone (7)
An example of chiral chromatogram of 7 is shown in
Figure 6. Two very large and asymmetric peaks partially
overlapping were observed with resolution parameter very
low, less than 0.4. Notwithstanding the bad resolution charac-
teristics, we attempted to recover optically active analytes by
collecting a part of each of the two chromatographic peaks.
In order to avoid the overlapping region, fractions flowing
from 31 to 35.5 min and from 50 to 60 min were selected, as
Fig. 5. Molecular structure of 7,7-biphenyl-2,2-bitropone (10).
Chirality DOI 10.1002/chir

SYNTHESIS AND CHIROPTICAL PROPERTIES OF 3,3’-BIPHENYL-2,2’-BITROPONE
651
to be isolated, that racemization processes were only partially
controlled by our handling procedures.
Kinetic Measurements
The kinetics of racemization were followed by measuring
the decrease of the CD elꢀongation at 330 nm and in the tem-
perature range À6 to +6 C of optically active samples from
chiral HPLC as a function of time. We used a temperature-
controlled CD cell of 10 mm optical path. The following first-
order reaction rate constants (s-1) were obtained: 3.9 10-4,
3.0 Á 10-4, 1.3 Á 10-4, and 1.0 Á 10-4 at 6, 2.5, –3 and, –6 ꢀC, respec-
tively. From these data we obtained ΔH# = 18,500 cal mol -1 and
ΔS# = À7.5 e.u. for the racemization reaction of 7 at 25 ꢀC.
Fig. 6. Chromatogram of the preparative separation of enantiomers of 7
obtained as discussed in the text. The colored areas correspond to the col-
lected fractions of the two antipodes.
Chiral Chromatography: Conventional Detection
Figure 8 reports the chromatograms (absorbances at the
UV Shimadzu detector) obtained by injecting similar quanti-
ties of 7 and varying only the HPLC column temperature.
The mobile phase, 82.5/17.5 n-hexane/IPA, was the best
one for resolving the peaks at the lowest temperatures. We
judged the best resolution when the ratio of the first HPLC
peak to the minimum between peaks was maximal. The chro-
matograms here reported have been normalized to the same
area.
shown in Figure 6. The corresponding CD spectra, shown in
Figure 7, were acquired at T = À6 C, after volume reduction
ꢀ
of the samples to that of the CD cell of 1 ml by evaporation
of the mobile phase, at ca. –10 ꢀC. The first eluting peak
was characterized by two CD bands, with a negative maxi-
mum at 330 nm and positive one at 290 nm, while the second
eluted peak revealed a CD spectrum exactly opposite to that
of the first one and having, in the example here reported,
comparable intensity. Both the optically active samples
resulted in chemically pure 7, as shown by spectroscopic
and HPLC analyses. The CD signal was not proportional to
the concentration of 7, this last being about double in the sec-
ond sample. Hence, the optical purity of the samples was
neatly different. The CD of both samples decreased with time
up to zero, the samples becoming indistinguishable except
for their absorbance.
By increasing the column temperature we observed a radi-
cal change of the chromatogram profile, as shown in Figure 8.
ꢀ
In particular, at low temperatures (T < 15 C) the chromato-
gram is characterized by two partially overlapping peaks with
resolution decreasing on increasing the temperature. At 18.8
and 20 ꢀC a “racemate peak” appears between the two “enan-
tiomeric” peaks. Finally, at higher temperatures only a peak is
observed. For other chromatograms of 7 with OJ chiral col-
umn used in different conditions of flux, see the “Supporting
Material” Section.
The body of measurements indicates that optically inactive
7 really consists of a racemic mixture of conformers that can
be partially resolved into two enantiomeric conformations
through chiral chromatography.16,17 Isolation of pure anti-
podes were impossible by the usual manipulation techniques,
i.e., chromatographic isolation of active fractions.
An ample variation of the intensity of the molar Δe spec-
trum of the samples collected with similar injections of 7
was observed. This fact shows, in addition to the variability
due to the more or less different choices of HPLC fractions
Chiral Chromatography: CD and UV Synchronous Detection
In Figure 9 the HPLC profiles obtained with the Daicel
Chiralcel OJ column are reported as measured both at 330
nm with our CD apparatus joined with the absorbance
Shimadzu detector. Note that the scales of both profiles were
conveniently adjusted to compare their shapes. The time axis
is unique, related to the CD detector, and the absorbance
measurements reported have been shifted to take into ac-
count their real retardation with respect to the CD detection.
The red and blue broken lines represent the integrated CD
Fig. 7. CD spectra of the isolated optically active samples measured at
T = À6 ꢀC. The colors correspond to those of the collected fractions by means
of HPLC separation reported in Fig. 6.
Fig. 8. Effect of temperature on the separation of enantiomers of 7, by
means of chiral HPLC. Absorbance was measured at 330 nm. Curves were
normalized to the same area.
Chirality DOI 10.1002/chir

652
CAVAZZA ET AL.
Fig. 11. (a) Molar CD, Δe ( M-1cm-1), as a function of l (nm) and (b) molar
absorbance, Δe (M-1cm-1) as a function of l (nm) for (Sa)-7.
Fig. 9. Chiral HPLC separation of the 7 enantiomers by using the CD de-
than 65 min both CD and UV measurements decrease in
value until losing significance at times greater than 80 min.
In the many attempts of isolating the antipodes of 7, we
obtained, for the first HPLC peak, molar CD varying from
À5 to À22 M-1 cm-1 for the band at 330 nm, as previously
noted. However, these experiments led to a precise CD spec-
trum profile: now we used the minimum value À29 M-1 cm-1
of Δe(330), obtained with our CD/UV apparatus “on-line” with
the chiral column, to scale the CD profile to what we think to
be near the enantiomeric Δe(l) of the first eluting enantiomer
(Sa)-7 (Fig. 11). Note that here and in the following we will
use the Ra and Sa stereodescriptors for the 7 antipodes as
resulting from theoretical computations (vide infra).
tector (red line). Comparison with the synchronous UV detection (blue line)
is reported. The integrals of both curves, from 25 min up to time t are also
shown (dashed lines of the same corresponding color). [Color figure can be
viewed in the online issue, which is available at wileyonlinelibrary.com.]
and absorbance, respectively, starting from their beginning
(27 min) up to the observational instant. We noted that inte-
grated absorbance asymptotic value doubles that observed
at ca. 43 min (the inflection moment of the CD curve that is
near the non-zero minimum of absorbance).
In this case the negative and positive values tend to a
perfect compensation so that the value is zero at very high
retention times.
We used the absorbance value to calculate the concentra-
tion of bitropone 7 and its measured ellipticity at any instant
in order to evaluate the molar CD of enantiomers of 7. In
Figure 10 the value of the molar circular dichroism,
Δe(330), obtained at each instant is reported, together with
the corresponding enantiomeric excess (e.e.). We note the
minimum value of this molar CD, –29.0 M-1 cm-1, at ca 31
min in the initial part of both the CD and UV HPLC curves.
The Δe curve slowly increases up to 36 min, after which the
increasing trend neatly accelerates: the Δe(330) value be-
comes positive and at the maximum value, ca 22 M-1 cm-1, re-
mains almost constant for more than 10 min. Finally, at more
By using the enantiomeric Δe at the wavelength (330 nm)
we could calculate the e.e. at each instant (Fig. 10), and hence
the instantaneous concentration of each antipode. In
Figure 12 the concentration of (Sa)-7 and (Ra)-7 antipodes
together with the HPLC overall concentration are reported.
Looking at Figure 12, we note the presence of racemate
also near the maximum of the first peak and the presence of
(Sa)-7, still present at retention times higher than 60 min in
constant concentration. Racemization in the time elapsed at
RT of the antipodes during their postcolumn migrations is
negligible, given both the very reduced times passed at room
temperature of optically active 7 and the measured racemiza-
tion kinetic constant.
Absolute Configuration of 7 Enantiomers
Now the problem is the identification and determination of
the absolute configuration of the 7 antipodes revealed by
HPLC separation. From the experimental point of view, handy
Fig. 10. Chiral separation of enantiomers 7 detected by the instantaneous
molar Δe (M-1cm-1) measured at 330 nm. In the lower part of the figure the cor-
responding enantiomeric excess is also reported (red line) with the corre-
sponding absorbance intensity as eye guide (black line). [Color figure can
be viewed in the online issue, which is available at wileyonlinelibrary.com.]
Fig. 12. Instantaneous concentrations of the antipodes of the sample 7:
blue curve refers to (Sa)-7 and red curve corresponds to (Ra)-7. Dotted line
indicates the total concentration. [Color figure can be viewed in the online is-
sue, which is available at wileyonlinelibrary.com.]
Chirality DOI 10.1002/chir

SYNTHESIS AND CHIROPTICAL PROPERTIES OF 3,3’-BIPHENYL-2,2’-BITROPONE
653
Fig. 13. Optimized geometry of the enantiomers of 7, computed as reported in the text. The corresponding 1,2,2’,1’ dihedral angles defining the relative geom-
etry between the two pseudoaromatic rings are reported.
solutions are not available, since we do not know other opti-
cally active molecules having similar structures which corre-
late to 7. In addition, we could hardly proceed to chemical
transformations of 7 to obtain some molecule with known
absolute configuration given the high thermal sensibility of
7 antipodes in solution. Thus, we decided to proceed with
theoretical calculations on the most stable conformations of
7 in its ground state. Starting from these minimum energy
conformations, it was possible to compute theoretical CD
and UV absorption spectra to be compared with the experi-
mental ones.
crucial geometrical feature determining the atropisomerism
between the two enantiomers 7.
For both enantiomers, the calculations confirm that the
tropone rings are substantially planar, as observed in experi-
ments on tropone,18 colchicine,19 and isocolchicine.20 In our
systems, the presence of phenyl substituents in positions 3
and 3’ is at the origin of significant hindering and restrictions
with respect to the simpler 2,2’ bitropone (1). The calculated
dihedral angle (ca. 70˚) is a bit larger than that optimized for
1 (ca. 60º) and it corresponds to the angle which maximizes
the noncovalent interactions between the two phenyl groups
which are here properly described by the M06-2X functional.21
The calculated UV–vis and CD spectra for the two enantio-
mers are reported in Figure 14.
As expected, the two enantiomers present an identical UV–
vis spectrum, whereas the corresponding CD are opposite in
sign notwithstanding the small differences in structural pa-
rameters obtained in the geometry optimization process. In
particular, in the low-energy region (240–400 nm), the one ex-
perimentally explored, the Ra-enantiomer shows a clear +/À
couplet which is reverted in the Sa-enantiomer. By comparing
the simulated spectra with the experimental one reported in
Figure 11, we see that the TD-DFT description for the Sa-
enantiomer well reproduces the shape of the measured CD
The optimized geometry of the two minimum energy
enantiomers, determined as described in the Computational
Details, are reported in Figure 13. In the same figure we also
report the dihedral angle 1,2,2’,1’ (see Fig. 1), which is the
Fig. 15. Comparison of the low-energy region of the CD spectra computed
for the two enantiomers (dashed curves) and for the analogs obtained by re-
placing the phenyl groups with hydrogen atoms (solid curves). Blue curves re-
fer to (Sa)-7 and red curves correspond to (Ra)-7. [Color figure can be viewed
in the online issue, which is available at wileyonlinelibrary.com.]
Fig. 14. (below) Computed UV–vis spectrum and (above) CD spectra com-
puted for the two enantiomers: blue curve refers to (Sa)-7 and red curve cor-
responds to (Ra)-7. [Color figure can be viewed in the online issue, which is
available at wileyonlinelibrary.com.]
Chirality DOI 10.1002/chir

654
CAVAZZA ET AL.
spectrum, allowing the identification of the first-eluted
enantiomer.
conclude that this new class of bitropones offers intense
atropisomeric phenomena comparable to those of other well-
established families of atropisomeric compounds.
To better understand the role of the phenyl groups in
determining the shape of the CD spectra we repeated
the calculations for two artificial systems obtained by
replacing the phenyl groups with hydrogen atoms in both
the Ra and Sa enantiomers and keeping all the geo-
metrical parameters fixed. The resulting CD spectra are
reported in Figure 15.
LITERATURE CITED
1. Zhao J. Plant troponoids: Chemistry, biological activity, and biosynthesis.
Curr Med Chem 2007;14:2597–2621.
2. Banwell MG. New methods for the synthesis of troponoid compounds.
Aust J Chem 1991;44:1–36.
As can be seen, the two bands in the low-energy region are
completely reverted in sign with respect to the original sys-
tems, showing that it is exactly the presence of the phenyl
groups that determine the shape of the CD spectra. This
can be explained looking at the characteristics of the involved
transitions in the two systems: we observe a change in the
angle between the electric and the magnetic transition
dipoles which finally leads to a change in the sign of the
corresponding rotational strengths.
3. Iyoda M, Sato K, Oda M. A simple synthesis of bitropones and conversion
of 2,2’-bitropone into dicyclohepta[b,d] furan dication, a novel dication
species. J Chem Soc Chem Commun 1985; 21: 1547.
4. Kubo K, Mori A, Ujiie S, Tschierske C. Synthesis and properties of colum-
nar liquid crystals and organogelator with a bitropone core. J Oleo Sci
2004;53:575–579.
5. Kubo K, Sutoh T, Mori A, Ujiie S. Synthesis, structure and mesomorphic
properties of liquid crystals with a bitropone core. Bull Chem Soc Jpn
2002;75:1353–1358.
6. Okada H. Electrophotographic photoconductor containing bitropone
derivative. Patent. CAN 154:37420 AN 2010:1536624.
7. Funaioli T, Cavazza M. Homocoupling of halotropones promoted by bis
(1,5-cyclooctadiene)nickel in the presence of tris(1-pyrazolyl)methane:
an easy route to [bi-1,3,5-cycloheptatrien-1-yl]-7,7’-diones. Synthetic
Commun 2009;39:1592–1600.
CONCLUSIONS
In this work we demonstrated that the optically inactive
3,3’-biphenyl-2,2’-bitropone really consists of
mixture of conformers, which can be partially resolved into
two antipodes through chiral HPLC, when the separation is
performed at T < 15 C.
a racemic
8. Funaioli T, Cavazza M, Zandomeneghi M, Pietra F. On the chiroptical be-
havior of conjugated multichromophoric compounds of
a new
pseudoaromatic class: bicolchicides and biisocolchicides. PLoS One
2010;5:e10617.
ꢀ
DFT calculations were used to determine the absolute
configurations of the two antipodes, assigning the Sa configu-
ration to the first eluted enantiomer in the HPLC separation.
Moreover, theoretical calculations have also shown the crit-
ical role played by the phenyl substituents in the determina-
tion of spectroscopic properties of 7. In particular, it has
been shown that removing the phenyl rings produces an
inversion of the optical activity, as shown in Figure 15.
In the case of 10, where phenyl groups are present at a
neatly higher distance with respect to 7, we were unable to
obtain any measurable separation of antipodes. This can be
due to the loss of the stabilizing noncovalent interactions
between phenyl substituents that are present in 7.
9. Cavazza M, Zandomeneghi M, Pietra F. (2000) Isolation of atropisomers
in both the isocolchicides and colchicine series of alcaloids and determi-
nation of their chiroptical properties. Tetrahedron Lett 2000;41:9129–9133.
10. Lincoln P, Nordh J, Deinum J, Angstrom J, Norden B, Conformation of
thiocolchicine on 2 B-ring-modified analogs bound to tubulin studied with
optical spectroscopy. Biochemistry 1991;30:1179–1187.
11. Barton DHR. The principles of conformational analysis. Science
1970;169:539–544.
12. Mukai T. Reaction of 2-chloro-3-phenyltropone and 2-chloro-7-
phenyltropone with ammonia and amines. Bull Chem Soc Jpn
1959;32:272–279.
13. Piettre SR, André C, Chanal MC, Ducep JB, Lesur B, Piriou F, Raboisson
P, Rondeau JM, Schelcher C, Zimmermann P, Ganzhorn AJ. Monoaryl-
and bisaryldihydroxytropolones as potent inhibitors of inositol
monophosphatase. J Med Chem 1997;40:4208–4221.
The coupling between the chiral HPLC column, the J40AS
spectropolarimeter, and the standard absorbance detector
allowed us to measure the enantiomeric Δe for the Sa-
antipode (see Fig. 11) at 330 nm within the shortest postcolumn
racemization time possible (a few seconds). This is of capital
importance for the accuracy of the results in cases, as the ac-
tual, where racemization is appreciable even at the lowest tem-
peratures compatible with the experimental apparatuses
(particularly the chiral HPLC columns).
However, as we cannot experimentally exclude that the material
exiting the chiral column contains, even in correspondence to the
maximum CD signal, some racemate, we have to point out that the
measured Δe value, –29 M-1cm-1, was hypothesized to correspond
to the 100% e.e. of the first eluted peak. Thus, the Δe(l) here
reported in Figure 11 may be underestimated.
On the other hand, the measured Δe(l) gives c.a. 0.003 for
g,22 the Kuhn dissymmetry factor Δe/e for the couplet
centered at 315 nm. We remember that in compounds 4–6
the maxima values of g are on the order of 0.001, while for
2,2’-binaphthol, a typical atropisomeric aromatic molecule, g
presents its maximum, ca 0.002, in correspondence to its
CD couplet.23 Many examples of atropisomeric molecules
are reported in Ref. 24 with their Δe and e spectra: rarely does
their asymmetry factor exceed 0.0025. We can hence
Chirality DOI 10.1002/chir
14. Zhao Y, Truhlar DG. The M06 suite of density functionals for main group
thermo- chemistry, thermochemical kinetics, noncovalent interactions,
excited states, and transition elements: two new functionals and system-
atic testing of four M06-class functionals and 12 other functionals. Theor
Chem Acc 2008;120:215–241.
15. Frisch MJ, Trucks GW, Schlegel HB
, Scuseria GE, Robb MA,
Cheeseman JR, Scalmani G, Barone V, Mennucci B, Petersson GA,
Nakatsuji H, Caricato M, Li X, Hratchian HP, Izmaylov AF, Bloino J,
Zheng G, Sonnenberg JL, Hada M, Ehara M, Toyota K, Fukuda R,
Hasegawa J, Ishida M, Nakajima T, Honda Y, Kitao O, Nakai H, Vreven
T, Montgomery J Jr., Peralta JE, Ogliaro F, Bearpark M, Heyd JJ,
Brothers E, Kudin KN, Staroverov VN, Kobayashi R, Normand J,
Raghavachari KR, Rendell A, Burant JC, Iyengar SS, Tomasi J, Cossi M,
Rega N, Millam JM, Klene M, Knox JE, Cross JB, Bakken V, Adamo C,
Jaramillo J, Gomperts R, Stratmann RE, Yazyev O, Austin AJ, Cammi R,
Pomelli C, Ochterski JW, Martin RL, Morokuma K, Zakrzewski VG, Voth
GA, Salvador P, Dannenberg JJ, Dapprich S, Daniels AD, Farkas Ö,
Foresman JB, Ortiz JV, Cioslowski J, Fox DJ. Gaussian 09, Revision
H.11, Wallingford CT: Gaussian Inc.; 2009.
16. Wolf C. Stereolabile chiral compounds: analysis by dynamic chromatogra-
phy and stopped-flow methods. Chem Soc Rev 2005;43:595–608.
17. Nishikawa T, Hayashi Y, Suzuki S, Kubo H, Ohtani H. On-column
enantiomerization of 3-hydroxybenzodiazepines during chiral liquid
chromatography with optical rotation detection.
J Chromatogr A
1997;767:93–100.
18. Veracini CA, Pietra F. Molecular structure of tropone from its 1H nuclear
magnetic resonance spectrum in a nematic solvent. J Chem Soc Chem
Commun 1972, 1262–1263.

SYNTHESIS AND CHIROPTICAL PROPERTIES OF 3,3’-BIPHENYL-2,2’-BITROPONE
655
19. Lessinger L, Margulis TN. The crystal structure of colchicines. A new ap- 22. Kuhn W., Optical rotatory power. Ann Rev Phys Chem 1958;9:417–438.
plication of magic integers to multiple-solution direct methods. Acta Cryst
1978a;B 34:578–584.
23. Cavazza M, Zandomeneghi M, Ouchi A, Koga Y. Photochromism in 1,1’-
Bi-2-naphthols. J Am Chem Soc 1996;118:9990–9991.
20. Lessinger L, Margulis TN. The crystal structure of isocolchicine, an inactive iso-
24. Harada N, Nakanishi K, Berova N. Electronic CD exciton chirality
method: principles and application. In: Comprehensive chiroptical spec-
troscopy, application in stereochemical analysis of synthetic compounds,
natural products and biomolecules, vol. 2. Hoboken, NJ: John Wiley and
Sons; 2012. p 115–166.
mer of the mitotic spindle inhibitor colchicines. Acta Cryst 1978b;B 34:1556–1561.
21. Hohenstein EG, Chill ST, Sherrill CD. Assessment of the performance of the
M05 À 2X and M06 À 2X exchange-correlation functionals for noncovalent
interactions in biomolecules. J Chem Theory Comput 2008;4:1996–2000.
Chirality DOI 10.1002/chir