Stereodynamics and absolute configuration of stereolabile atropisomers in 2,2-dimethyl-1-aryl-1-indanols

Article abstract of DOI:10.1002/chir.20988

We describe herein the investigation of the stereodynamic processes occurring in a series of 1-aryl-2,2-dimethylindanols, by dynamic NMR. When the aryl moiety is a mesityl or a 2-methyl-1-naphthyl, the rotational barrier exceeds the 25 kcal/mol, so that s

Full text of DOI:10.1002/chir.20988

CHIRALITY 23:768–778 (2011)
Stereodynamics and Absolute Configuration of Stereolabile
Atropisomers in 2,2-Dimethyl-1-aryl-1-indanols
2
2
DANIELE CASARINI,1 MICHELE MANCINELLI, ANDREA MAZZANTI,
AND FRANCESCA BOSCHI2,3
*
*
1Department of Chemistry, University of Basilicata, via dell’Ateneo Lucano 10, Potenza 85100, Italy
2Department of Organic Chemistry ‘‘A. Mangini’’, University of Bologna, viale Risorgimento 4, Bologna 40136, Italy
3COMECER SpA, via Emilia Ponente 390, Castel Bolognese (RA) 48100, Italy
Contribution to the Carlo Rosini Special Issue
ABSTRACT
We describe herein the investigation of the stereodynamic processes occurring
in a series of 1-aryl-2,2-dimethylindanols, by dynamic NMR. When the aryl moiety is a mesityl
or a 2-methyl-1-naphthyl, the rotational barrier exceeds the 25 kcal/mol, so that stable atro-
pisomers are observed. In two cases, all the chiral-atropisomeric species have been separated
by enantioselective HPLC, and the comparison between theoretical and experimental electronic
circular dichroism spectra allowed the absolute configuration assignment of all the isolated spe-
C
VV
cies to be obtained. Chirality 23:768–778, 2011.
2011 Wiley-Liss, Inc.
KEY WORDS: indanols; dynamic NMR; enantioselective HPLC; electronic circular dichroism;
DFT calculations
INTRODUCTION
General Procedure for the Synthesis of the
2,2-Dimethyl-1-aryl-1-indanols (1-7)
Over the past several years, we have been using dynamic
NMR spectroscopy to investigate the conformations and the
stereomutation processes occurring in aryl alcohols deriva-
tives.1–4 In combination with these studies, we get intrigued
in the investigation of 1-aryl-1-indanol derivatives, because a
large number of bioactive and pharmaceutically interesting
molecules contain the indanyl moiety5–8 in the framework. In
contrast to this, only little information is available in the liter-
ature about the conformational preferences of the aryl sub-
stituents and their effect on the chirooptical properties.
n-BuLi 1.6 M in n-hexane (2.5 mmol) was added in 15 min to a solu-
tion of the appropriate aryl halide (2.5 mmol in 25 ml of dry THF),
cooled to 2808C and under N2. The mixture was allowed to react for 20
min, and then a solution of 2,2-dimethyl-1-indanone (2.5 mmol in 5 ml of
dry THF) was added, and the temperature was allowed to rise slowly at
308C. After checking by GC-MS, the end of the reaction, 40 ml of H2O
was added. The mixture was extracted with ether, and the collected or-
ganic layers were dried. Then, the solvent was removed, and the crude
was purified by chromatography on a silica-gel column eluted with petro-
leum ether/acetone 9:1 Analytically pure samples were obtained by
semipreparative HPLC on a C18 column.
MATERIALS AND METHODS
General
1-(3,5-Dimethylphenyl)-2,2-dimethyl-2,3-dihydro-1H-inden-
1-ol (1). 1H NMR (CDCl3, 600 MHz) d 0.65 (3H, s, Me), 1.20 (3H, s,
Me), 1.85 (1H, s, OH), 2.30 (6H, s, 2Me), 2.72 (1H, AB d, J 5 15.5Hz,
CH2), 2.99 (1H, AB d, J 5 15.5 Hz, CH2), 6.92 (1H, s, Ph), 6.94 (2H, br s,
Ph), 7.21 (1H, d, J 5 7.2 Hz), 7.24 (1H, m), 7.30 (2H, m). 13C NMR
(CDCl3, 150.8 MHz) d 21.5 (2CH3), 22.0 (CH3), 26.4 (CH3), 45.4 (CH2),
48.3 (Cq), 88.0 (Cq), 124.8 (CH), 125.2 (2CH), 125.3 (CH), 126.7 (CH),
128.2 (CH), 128.6 (CH), 136.8 (2Cq), 141.9 (Cq), 143.5 (Cq), 147.7 (Cq).
All the solvents were dried over sodium/benzophenone and were dis-
tilled just before the use, once the mixture had turned to a persistent
blue color. The 2,2-dimethyl-1-indanone was prepared according to the
literature,9,10 whereas the other reagents were of commercial grade (pu-
rity > 98%) and used without any further purification. The analytical TLC
was performed on silica gel plates Merck 60 F 254, and the preparative
purification was carried out on a chromatography column with silica gel
Merck 60 (80–230 mesh). Analytical samples were obtained by semipre-
parative HPLC using a Waters Delta-600 pump and a Waters 2487 UV-de-
tector. A Phenomenex Luna C18(2), 250 mm 3 20 mm, 10-lm column
was routinely used for the purifications.
HRMS (ESI-FT-ICR) calcd for
265.15942.
C19H21O
[M-H]2 265.15979; found
1-(3,5-Dimethylbenzyl)-2,2-dimethyl-2,3-dihydro-1H-inden-
1-ol (2). 1H NMR (CDCl3, 600 MHz) d 0.96 (3H, s, Me), 1.26 (3H, s,
Me), 1.85 (1H, s, OH), 2.21 (6H, s, 2Me), 2.63 (1H, AB, J 5 15.5Hz,
CH2Ph), 2.67 (1H, AB, J 5 15.5Hz, CH2), 2.78 (1H, AB, J 5 15.5 Hz,
CH2), 2.93 (1H, AB, J 5 15.5Hz, CH2Ph), 6.37 (1H, d, J 5 7.5Hz), 6.49
(2H, s, Ph), 6.84 (1H, s, Ph), 6.95 (1H, m), 7.14 (2H, m). 13C NMR
3-Methoxybromomesitylene
A solution of 2,4,6-trimethylphenol (20 mmol) in 5 ml of THF was
dropped in 20 min to a suspension of NaH 90% (22 mmol) in 40 ml of
THF kept under N2 and cooled at 58C. The mixture was stirred as long
as the gas evolution was completed, and then methyl sulfate (25 mmol)
was added. After 1 h at room temperature, the reaction resulted com-
pleted, and it was quenched by adding 10 ml of water. THF was removed
at reduced pressure, and the residue was extracted with ether. The col-
lected organic layers were dried, the solvent removed, and the crude
was dissolved in CH2Cl2 and reacted with bromine following a known
procedure.11
Additional Supporting Information may be found in the online version of this
article.
Contract grant sponsors: University of Basilicata (Potenza); MIUR-PRIN
2008, Rome; University of Bologna; Contract grant numbers: RFO 2008, RFO
2009.
*Correspondence to: Andrea Mazzanti, Department of Organic Chemistry
‘‘A. Mangini,’’ University of Bologna, viale Risorgimento 4, Bologna 40136,
Italy. E-mail: mazzand@ms.fci.unibo.it
1H NMR (CDCl3, 300 MHz) d 2.23 (3H, s, Me), 2.34 (3H, s, Me), 2.38
(3H, s, Me), 3.69 (3H, s, OMe), 6.94 (1H, s, CH). 13C NMR (CDCl3, 75.5
MHz) d 15.8 (CH3), 16.8 (CH3), 23.3 (CH3), 60.1 (OCH3), 125.1 (Cq),
129.3 (Cq), 130.0 (CH), 131.3 (Cq), 133.5 (Cq), 155.2 (Cq).
Received for publication 23 February 2011; accepted in revised form 13 May
2011; Accepted 17 May 2011
DOI: 10.1002/chir.20988
Published online 18 August 2011 in Wiley Online Library
(wileyonlinelibrary.com).
C
VV
2011 Wiley-Liss, Inc.

STEREODYNAMICS OF STEREOLABILE ATROPISOMERS
769
(CDCl3, 150.8 MHz) d 21.2 (2CH3, Ph), 22.2 (CH3), 24.1 (CH3), 41.9
(CH2Ph), 45.0 (CH2), 47.8 (Cq), 85.0 (Cq), 124.6 (CH), 125.0 (CH),
125.1 (CH), 127.5 (CH), 127.8 (CH, Ph), 129.0 (2CH, Ph), 136.9 (Cq),
136.9 (2Cq, Ph), 140.3 (Cq, Ph), 146.2 (Cq). HRMS (ESI-FT-ICR) calcd
for C20H23O [M-H]2 279.17544; found 279.17514.
123.9 (CH), 124.6 (CH), 125.0 (CH), 125.1 (CH), 125.4 (CH), 126.1 (CH),
126.7 (CH), 127.8 (CH), 128.1 (CH), 128.5 (CH), 128.8 (CH), 131.7 (Cq),
134.2 (Cq), 138.0 (Cq), 142.6 (Cq), 147.9 (Cq). HRMS (ESI-FT-ICR)
calcd for C21H19O [M-H]2 287.14414; found 287.14380.
2,2-Dimethyl-1-(2-methylnaphthalen-1-yl)-2,3-dihydro-1H-
inden-1-ol (7). Major Atropisomer (70%) 1H NMR (CDCl3, 600
MHz) d 1.06 (3H, s, Me), 1.55 (3H, s, Me), 1.61 (3H, s, Me), 2.18 (1H, s,
OH), 3.02 (1H, AB d, J 5 16.5 Hz, CH2), 3.07 (1H, AB d, J 5 16.5 Hz,
CH2), 7.08 (1H, d, J 5 8.5 Hz, Ar), 7.17 (1H, d, J 5 7.5 Hz, Ar), 7.25 (1H,
t, J 5 7.5 Hz, Ar), 7.26 (1H, m, Ar), 7.31 (1H, dt, J 5 7.0;1.2 Hz, Ar), 7.40
(1H, dt, J 5 6.5; 1.2 Hz, Ar), 7.45 (1H, m, J 5 8.5; 6.5; 1.6 Hz, Ar), 7.63
(1H, d, J 5 8.5Hz, Ar), 7.78 (1H, dd, J 5 8.0;1.5Hz, Ar), 9.01 (1H, d, J 5
8.8Hz, Ar). 13C NMR (CDCl3, 150.8 MHz) d 24.2 (CH3), 27.0 (CH3), 28.0
(CH3), 49.0 (CH2), 49.0 (Cq), 91.5 (Cq), 124.2 (CH), 124.3 (CH), 125.0
(CH), 125.2 (CH), 127.5 (CH), 127.8 (CH), 128.1 (CH), 128.4 (CH), 129.9
(CH), 131.0 (CH), 133.5 (Cq), 133.7 (Cq), 134.4 (Cq), 138.2 (Cq), 141.3
(Cq), 151.0 (Cq).
1-Mesityl-2,2-dimethyl-2,3-dihydro-1H-inden-1-ol (3). 1H NMR
(CDCl3, 600 MHz) d 1.00 (3H, s, Me), 1.35 (3H, s, Me), 1.40 (3H, s, Me),
1.82 (1H, s, OH), 2.23 (3H, s, Me), 2.66 (3H, s, Me), 2.94 (1H, AB d, J 5
16.2 Hz, CH2), 2.97 (1H, AB d, J 5 16.2 Hz, CH2), 6.65 (1H, s, Ph), 6.92
(1H, s, Ph), 7.16–7.23 (3H, m), 7.26 (1H, dt, J 5 7.2; 1.4 Hz). 13C NMR
(CDCl3, 150.8 MHz) d 20.4 (CH3), 23.5 (CH3), 25.8 (CH3), 27.2 (CH3),
27.9 (CH3), 48.6 (Cq), 49.0 (CH2), 90.8 (Cq), 124.9 (CH), 125.2 (CH),
127.3 (CH), 128.3 (CH), 130.9 (CH), 131.9 (CH), 135.3 (Cq), 136.2 (Cq),
138.6 (Cq), 139.1 (Cq), 141.3 (Cq), 151.0 (Cq). HRMS (ESI-FT-ICR)
calcd for C20H23O [M-H]2 279.17544; found 279.17557.
1-(3-Methoxy-2,4,6-trimethylphenyl)-2,2-dimethyl-2,3-dihy-
dro-1H-inden-1-ol (4). Major Atropisomer (54%) 1H NMR (CDCl3,
600 MHz) d 1.02 (3H, s, Me), 1.31 (3H, s, Me), 1.39 (3H, s, Me), 1.77
(1H, s, OH), 2.22 (3H, s, Me), 2.58 (3H, s, Me), 2.95 (2H, s, CH2), 3.73
(3H, s, OCH3), 6.64 (1H, s, Ar), 7.12 (1H, d, J 5 7Hz, Ar), 7.16–7.22 (2H,
m, Ar), 7.26 (1H, t, J 5 7 Hz, Ar). 13C NMR (CDCl3, 150.8 MHz) d 15.8
(CH3), 17.4 (CH3), 22.8 (CH3), 27.4 (CH3), 27.9 (CH3), 48.4 (Cq), 49.5
(CH2), 59.4 (OCH3), 90.7 (Cq), 124.7 (CH), 125.5 (CH), 127.4 (CH),
128.3 (CH), 128.4 (Cq), 131.4 (Cq), 132.3 (Cq), 133.2 (CH), 141.2 (Cq),
141.3 (Cq), 151.0 (Cq), 156.4 (Cq).
Minor Atropisomer (30%) 1H NMR (CDCl3, 600 MHz) d 1.00 (3H, s,
Me), 1.50 (3H, s, Me), 1.68 (1H, br s, OH), 2.90 (3H, s, Me), 3.00 (1H,
AB d, J 5 16.5Hz, CH2), 3.17 (1H, AB d, J 5 16.5Hz, CH2), 6.77 (1H, d, J
5 7.5Hz, Ar), 6.86 (1H, m, J 5 8.5;6.7;1.5Hz, Ar), 7.05 (1H, d, J 5 9.0Hz,
Ar), 7.06 (1H, t, J 5 7.5Hz, Ar), 7.18 (1H, t, J 5 7.2Hz, Ar), 7.28 (1H, dt, J
5 7.5;1.5Hz, Ar), 7.34 (1H, d, J 5 7.7Hz, Ar), 7.36 (1H, d, J 5 8.5 Hz,
Ar), 7.65 (1H, d, J 5 8.5 Hz, Ar), 7.68 (1H, dd, J 5 8.0; 1.6 Hz, Ar). 13C
NMR (CDCl3, 150.8 MHz) d 24.5 (CH3), 26.3 (CH3), 28.5 (CH3), 46.0
(Cq), 47.5 (CH2), 91.2 (Cq), 123.3 (CH), 123.6 (CH), 125.2 (CH), 125.6
(CH), 127.1 (CH), 127.9 (CH), 128.2 (2CH), 128.3 (CH), 132.1 (CH),
132.6 (Cq), 132.9 (Cq), 136.4 (Cq), 138.0 (Cq), 140.7 (Cq), 151.4 (Cq).
Minor Atropisomer (46%) 1H NMR (CDCl3, 600 MHz) d 1.00 (3H, s,
Me), 1.35 (3H, s, Me), 1.37 (3H, s, Me), 1.90 (1H, s, OH), 2.23 (3H, s,
Me), 2.62 (3H, s, Me), 2.94 (1H, JAB 5 16 Hz, CH2), 3.01 (1H, JAB 5 16
Hz, CH2), 3.54 (3H, s, OCH3), 6.92 (1H, s, Ar), 7.12 (1H, d, J 5 7 Hz,
Ar), 7.16–7.22 (2H, m, Ar), 7.26 (1H, t, J 5 7Hz, Ar). 13C NMR (CDCl3,
150.8 MHz) d 15.5 (CH3), 15.8 (CH3), 25.4 (CH3), 27.2 (CH3), 28.1
(CH3), 48.2 (Cq), 48.9 (CH2), 59.4 (OCH3), 90.7 (Cq), 124.8 (CH), 125.0
(CH), 127.3 (CH), 128.1 (CH), 128.2 (Cq), 129.9 (Cq), 132.4 (CH), 134.4
(Cq), 140.6 (Cq), 141.1 (Cq), 151.1 (Cq), 155.4 (Cq). HRMS (ESI-FT-
ICR) calcd for C21H25O2 [M-H]2 309.18600; found 309.18568.
HRMS (ESI-FT-ICR) calcd for
301.15948.
C22H21O
[M-H]2 301.15979; found
NMR Spectroscopy
The spectra were recorded at 600 MHz for 1H and 150.8 MHz for 13C.
The assignments of the 1H and 13C signals were obtained by bi-dimen-
sional experiments (COSY, edited-gHSQC,12,13 and gHMBC14 sequences).
The NOE experiments were obtained by means of the DPFGSE-NOE15
sequence. To selectively irradiate the desired signal, a 20-Hz wide-shaped
pulse was calculated with a refocusing-SNOB shape16 and a pulse width of
92.5 ms. Mixing time was set to 1.8 s. The samples for obtaining spectra
at temperatures lower than 21008C were prepared by connecting to a vac-
uum line the NMR tubes containing the compound and a small amount of
C6D6 (for locking purpose), and condensing therein the gaseous CHF2Cl
and CHFCl2 (4:1 v/v) under cooling with liquid nitrogen. The tubes were
subsequently sealed in vacuum and introduced into the precooled probe
of the spectrometer. The temperature calibrations were performed before
the experiments, using a Cu/Ni thermocouple immersed in a dummy
sample tube filled with isopentane, and under conditions as nearly identi-
cal as possible. The uncertainty in the temperatures was estimated from
the calibration curve to be 628C. The line shape simulations were per-
formed by means of a PC version of the QCPE program DNMR 6 n8 633,
Indiana University, Bloomington, IN.
1-(Benzo[b]thiophen-7-yl)-2,2-dimethyl-2,3-dihydro-1H-inden-
1-ol (5). 1H NMR (CDCl3, 600 MHz) d 0.78 (3H, s, Me), 1.35 (3H, s,
Me), 2.29 (1H, s, OH), 2.79 (1H, AB d, J 5 15.5Hz, CH2), 3.02 (1H, AB d,
J 5 15.5Hz, CH2), 6.64 (1H, d, J 5 7.5Hz, Ar), 7.21 (1H, t, J 5 7.5Hz, Ar),
7.27 (2H, m, Ar), 7.33 (2H, m, Ar), 7.35 (1H, d, J 5 5.6Hz; Ar), 7.48 (1H,
d, J 5 5.6Hz; Ar), 7.75 (1H, d, J 5 7.6Hz; Ar). 13C NMR (CDCl3, 150.8
MHz) d 22.2 (CH3), 26.0 (CH3), 45.2 (CH2), 48.8 (Cq), 87.3 (Cq), 121.7
(CH), 122.1 (CH), 123.3 (CH), 123.8 (CH), 124.0 (CH), 124.5 (CH), 125.5
(CH), 126.9 (CH), 129.0 (CH), 139.6 (Cq), 139.7 (Cq), 142.9 (Cq), 146.7
(Cq), 148.2 (Cq). HRMS (ESI-FT-ICR) calcd for C19H17OS [M-H]2
293.10056; found 293.10028.
2,2-Dimethyl-1-(naphthalen-1-yl)-2,3-dihydro-1H-inden-1-ol
(6). Major Atropisomer (85%) 1H NMR (CDCl3, 600 MHz) d 0.79 (3H,
s, Me), 1.39 (3H, s, Me), 1.53 (1H, s, OH), 2.79 (1H, AB d, J 5 15.5 Hz,
CH2), 2.87 (1H, AB d, J 5 15.5 Hz, CH2), 6.67 (1H, dd, J 5 7.3; 1.2 Hz,
Ar), 7.21 (1H, t, J 5 7.5 Hz, Ar), 7.29–7.31 (3H, m, Ar), 7.34 (1H, m, Ar),
7.46 (1H, m, J 5 8.5;1.5 Hz, Ar), 7.50 (1H, m, J 5 8.5;1.5 Hz, Ar), 7.73
(1H, d, J 5 8.0 Hz, Ar), 7.84 (1H, dd, J 5 8.0; 1.5 Hz, Ar), 9.19 (1H, dd, J
5 9.0, 0.8 Hz, Ar). 13C NMR (CDCl3, 150.8 MHz) d 24.4 (CH3), 26.9
(CH3), 46.0 (CH2), 49.8 (Cq), 92.3 (Cq), 124.2 (CH), 124.8 (CH), 124.9
(CH), 125.0 (CH), 125.1 (CH), 127.1 (CH), 127.5 (CH), 128.4 (CH), 128.6
(CH), 128.7 (CH), 129.4 (CH), 132.7 (Cq), 134.9 (Cq), 138.4 (Cq), 142.4
(Cq), 149.7 (Cq).
Calculations
Geometry optimization was carried out at the B3LYP/6-31G(d) and
CAM-B3LYP/6-31G(d) level by means of the Gaussian 03 and Gaussian
09 series of programs.17,18 The standard Berny algorithm in redundant
internal coordinates and default criteria of convergence were used. The
harmonic vibrational frequencies were calculated for all the stationary
points. For each optimized ground state, the frequency analysis showed
the absence of imaginary frequencies, whereas each transition state
showed a single imaginary frequency. Visual inspection of the corre-
sponding normal mode was used to confirm that the correct transition
state had been found. If not explicitly indicated, the reported energy val-
ues represent the total electronic energies. In general, these give the
best fit with experimental Dynamic-NMR data.19,20 This approach avoids
artifacts that might result from the inevitably ambiguous choice of an
adequate reference temperature, and from the idealization of low-fre-
Minor Atropisomer (15%) 1H NMR (CDCl3, 600 MHz) d 0.75 (3H, s,
Me), 1.29 (3H, s, Me), 1.68 (1H, s, OH), 2.87 (1H, AB d, J 5 15.5Hz,
CH2), 3.22 (1H, AB d, J 5 15.5 Hz, CH2), 7.01 (1H, d, J 5 7.6 Hz, Ar),
7.06 (1H, m, Ar), 7.16 (1H, t, J 5 7.3 Hz, Ar), 7.40 (1H, d, J 5 8.5 Hz,
Ar), 7.42 (1H, d, J 5 8.5 Hz, Ar), 7.59 (1H, t, J 5 8.0 Hz, Ar), 7.82 (1H, d,
J 5 8.0 Hz, Ar), 8.24 (1H, dd, J 5 7.3; 1.2 Hz, Ar). 13C NMR (CDCl3,
150.8 MHz) d 21.4 (CH3), 28.3 (CH3), 45.9 (CH2), 49.8 (Cq), 92.3 (Cq),
Chirality DOI 10.1002/chir

770
CASARINI ET AL.
quency vibrators as harmonic oscillators (very important in the present
cases, where about one third of the calculated frequencies fall below the
500–600 cm21 range). TD-DFT calculations of 3 and 4 were obtained at
the CAM-B3LYP/6-31111G(2d,p)//CAM-B3LYP/6-31G(d) level. TD-
DFT calculations of
7
were obtained at the CAM-B3LYP/6-
3111G(d,p)//B3LYP/6-31G(d) level. To cover the 170–400 nm range,
50–75 transitions were calculated. The electronic circular dichroism
(ECD) spectra were then obtained by applying a 0.55 eV Gaussian band-
shape.21
Enantioselective HPLC Separation
and ECD Spectra
The analytic separation of the racemic mixtures of compounds 3, 4,
and 7 was achieved at 258C on a DAICEL Chiralcel AD-H 250 mm 3 4.6
mm column, at a flow rate of 1.0 mL/min, using hexane/iPr-OH 90:10
v:v as eluent. Semipreparative separations were obtained on a semipre-
parative DAICEL Chiralcel AD-H 250 mm 3 21.2 mm at a flow rate of 20
ml/min, using the same eluent. In the case of compound 4, a Phenom-
enex LUX Amylose-2 (250 mm 3 10 mm, 5 lm, 4 ml/min, hexane/iPr-
OH 98:2 v:v) was also used. UV detection was fixed at 210 nm. UV
absorption spectra were recorded at 258C in acetonitrile on the racemic
mixtures, in the 190–400 nm spectral region. The cell path length was
0.1 cm, concentration was 1.0–1.2 3 1024 mol/L. ECD spectra were
recorded at 258C in acetonitrile solutions, with the same path lengths
and concentrations used for UV spectra, in the range 190–400 nm;
reported De values are expressed as L/mol/cm.
Chart 1. Chemical formula of aryl-indanols 1-7.
methyl groups that protrude from the 2-position on the
indane ring.
The lack of symmetry due to the chiral carbon on the
indanyl structure implies a C1 symmetry in all the investi-
gated molecules, even when the rotation around the Cq-Ar
bond is rapid in the NMR time scale. Furthermore, the
chiral center makes diastereotopic the two adjacent gemi-
nal methyls as well as the methylenic hydrogens on the
X-Ray Diffraction
Crystals of compound 6 suitable for X-ray analysis were obtained by
slow evaporation of an hexane solution. Molecular formula: C21H20O, Mr
5 288.37, monoclinic, space group P21/c (No. 14), a 5 8.823(4), b 5
3
˚
˚
8.564(3), c 5 20.207(8) A, b 5 93.274(5), V 5 1524.3(11) A , T 5 298(2)
1
same ring.24 For this reason, the H and 13C variable tem-
K, Z 5 4, qc 5 1.257 g/cm3, F(000) 5 616, graphite-monochromated
MoKa radiation (k 5 0.71073 A), l(MoKa) 5 0.075 mm–1, colorless brick
˚
perature spectra (at 600 and 150.8 MHz, respectively) can
monitor the aryl rotation in compound 1 only by the split-
ting of the ortho-hydrogens and meta-methyls in the phe-
nyl ring. The local C2 symmetry axis of the 3,5-dimethyl-
phenyl ring implies that when the rotation is frozen the
groups that do not lie on such axis experience a different
environment and the corresponding NMR signals split into
two equally intense lines.
On cooling, a sample of 1 in CDFCl2 below 2308C
broadening of the ortho-hydrogens and of the meta-methyl
signals is observed as a consequence of the restricted
rotation of the phenyl ring that generates two identical
conformers (topomerization). In the 1H spectrum, the sig-
nals reach the coalescence point at 2538C and sharpen
again at 2778C into a 1:1 ratio (Fig. 1) with a separation
of 89 Hz (554 Hz in the 13C spectrum, Supporting Infor-
mation Fig. S1). The line shape simulation allowed the
determination of the rate constant for the dynamic process
at various temperatures; from these values, a rotational
(0.6 3 0.4 3 0.4 mm3), empirical absorption correction with SADABS
(transmission factors: 0.9706–0.9563), 2400 frames, exposure time 10 s,
2.58 ꢀ y ꢀ 28.63, 211 ꢀ h ꢀ 11, 211 ꢀ k ꢀ 11, 227 ꢀ l ꢀ 26, 16301
reflections collected, 3673 independent reflections (Rint 5 0.0556), solu-
tion by direct methods (SHELXS97) and subsequent Fourier syntheses,
2
full-matrix least-squares on Fo (SHELXTL 6.10), hydrogen atoms
refined with a riding model except for the hydroxyl hydrogen that was
experimentally localized and optimized. Data/restraints/parameters 5
3673/0/205, S(F2) 5 1.034, R(F) 5 0.0715, and wR(F2) 5 0.1774 on all
data, R(F) 5 0.0594 and wR(F2) 5 0.1636 for 2910 reflections with I >
2r (I), weighting scheme w 5 1/[r2(Fo2) 1 (0.1130P)2 1 0.2301P]
2
where P 5 (Fo 1 2Fc2)/3, largest difference peak and hole 0.438 and
3
˚
20.268 e/A *
RESULTS AND DISCUSSION
The 1-arylindanols of Chart 1 resemble the sterically hin-
dered tertiary aryl carbinols that are known to display hin-
dered rotation around the Cq-Ar bond when the two remain-
ing substiuents are sufficiently large.22,23 In the present case,
the two substituents are linked together in the rigid indanyl
core, and therefore, the conformational freedom is further
reduced. For this reason, these aryl indanols exhibit a re-
stricted rotation about the C(OH)–Ar bond because of the
steric interaction in the rotational transition state between
the ortho substituents of the 1-aryl moiety and the geminal
1
barrier of 10.4 6 0.2 kcal/mol was derived in both the H
and 13C simulated spectra.
On further cooling below 21008C, the signals of the phe-
nyl moiety remain sharp and keep the same shift, whereas
those belonging to the indanyl framework broaden and
change their chemical shift with the temperature (Fig. 2).
However, below 21208C even the phenyl signals begin to
broaden, so that at 21508C all signals are divided approxi-
mately with a 70:30 ratio, and the spectrum exhibits a com-
plex pattern of overlapped signals. This dynamic behavior
can be attributed to the slowing down of the flipping in the
five-membered ring that generates two diastereomeric con-
formers. Unfortunately below 21508C, the solute precipitated
*Crystallographic data (excluding structure factors) for the structure
reported in this paper have been deposited with the Cambridge Crystallo-
graphic Data Centre as supplementary publication no. CCDC-813948. Copies
of the data can be obtained free of charge on application to CCDC, 12 Union
Road, Cambridge CB21EZ, UK. Fax: (144) 1223-336-033; e-mail: deposit
@ccdc.cam. ac.uk.
Chirality DOI 10.1002/chir

STEREODYNAMICS OF STEREOLABILE ATROPISOMERS
771
preventing from the observation of the complete 1H signal
splitting. A rough estimation, based on the coalescence point
(about 21308C), indicates a value of 6.0 6 0.4 kcal/mol for
the energy barrier of this process.
DFT calculations showed that in the indanyl frame the
five-membered ring is not planar and the CMe2 carbon does
not lie on the same plane of the phenyl. Because of the pres-
ence of the asymmetric carbon in position 1, the two possible
positions of the CMe2 moiety (i.e., over or under the plane of
the phenyl) correspond to two diastereomeric conformations
(see Fig. 3) that can be named as ‘‘syn’’ when the C(Me2) is
on the same side of the hydroxyl group, and ‘‘anti’’ when the
edge of the envelope is on the other side. The calculations
(at the B3LYP/6-31G(d) and CAM-B3LYP/6-31G(d) levels)
indicate that both conformations should be populated, their
energy difference being 0.8 kcal/mol after zero point energy
correction (ZPE). These two conformational diastereoisom-
ers can exchange by the flipping of the five-membered ring,
through a transition state where the cycle is planar (see Fig.
3). This transition state has a calculated barrier of 4.5 kcal/
mol, in reasonable agreement with the experimental data.
The same calculations estimate a barrier of 9.6 kcal/mol for
the rotation of the 3,5-dimethylphenyl, again in good agree-
ment with the experiment. The four stationary points found
by DFT were then recalculated as single point energies at
the ab initio CISD/6-31G(d) level, to verify if an higher level
of calculation could provide a better fit with the experimental
data. The results were almost identical with that obtained
with DFT regarding the energy difference between the two
ground states (0.9 kcal/mol). On the contrary, the transition
state energies were higher with respect to the DFT data and
matched very well the experimental data (10.5 and 7.0 kcal/
Fig. 2. 1H NMR spectra (600 MHz in CDFCl2) of the aromatic zone of
compound 1. At 21508C all the signals are split approximately in a 70:30 ra-
tio (arrows).
mol for the aryl-rotation and for the ring inversion, respec-
tively). These results show that the CISD approach could
provide slightly better data at the price of very large compu-
tational times. On the other hand, the DFT approach should
be considered suitable to tackle the conformation analysis of
the present set of molecules because of its fairly good accu-
racy coupled with a reasonable computational cost.y
The barriers to rotation observed in the 2,2-dimethylinda-
nols (1–7) are largely influenced by the bulkiness of the aryl
substituent25,26,{ as shown in Table 1.
When the ortho-hydrogens of the aryl group are replaced
by methyls (compound 3), the steric hindrance, and conse-
quently, the barrier increases so much that the mesityl rota-
1
tion is already locked at 258C, as deduced by the H spec-
trum where the ortho-methyls and the meta-hydrogens show
pairs of sharp singlets separated by 750 and 160 Hz, respec-
yA single point energy CISD/6-31G(d) calculation on compound 1 takes
about 16 h on a 8-cores job running on Xeon X7355 processors @ 2.93 GHz,
with a scratch file of about 32 Gbytes. CISD calculations scale roughly with
N6 (N being the number of the basis functions, that is, of the orbitals), so
this approach would imply computational times close to a week when dealing
with larger molecules like compound 5, 6, and 7. On the other hand, an opti-
mization with DFT and the subsequent frequency calculation usually takes
2–8 h.
{The steric origin of the barrier to rotation is further supported by the fact
that the insertion of a methylene between the indanyl and the phenyl, com-
pound 2, drastically reduces the barrier at about 6.0 kcal/mol (Supporting In-
formation Figs. S2 and S3) in good agreement with the values measured in
similar compounds.25,26 Actually, the whole rotational process for 2 consists
of two simultaneous rotational processes: indanyl–CH2 and CH2–Ph, which
allow to better accommodate the phenyl bulkiness and thus facilitate the
whole motion by reducing the steric interactions.
Fig. 1. 1H NMR spectra (600 MHz in CDFCl2) of compound 1. (Left) ex-
perimental traces at various temperatures of the meta-methyls of the phenyl
ring. (Right) line shape simulations and rate constants (k) at the same tem-
peratures.
Chirality DOI 10.1002/chir

772
CASARINI ET AL.
TABLE 1. Rotational barriers (DG=) for 1-aryl-2,2-dimethylin-
TABLE 2. Calculated total energies (E8) and free energies
(G8) of the conformations of 3 (in kcal/mol,
CAM-B3LYP/6–31G(d) level)
danols (1–7)
Exp.
Calcd.
Exp DG=
Calcd DE=
Cpd
sp/ap ratio
sp/ap ratio
(kcal/mol)
(kcal/mol)
Cpd. Conformationa
E8
% Pop (DE8)
G8
% Pop (DG8)
1
Topomers
–
10.4
6.2 (inversion)
6.0
9.6,a 10.5b
4.3,a 7.0b
–
3
Syn, 180
Anti, 60
Syn, 60
0.00
0.62
1.79
1.19
65
23
3
0.00
0.25
0.48
0.91
43
28
19
10
2
3
4
5
6
7
Topomers
Topomers
54:46
96:4
89:11
–
–
>22.0
26.95
16.7
18.8
25.5
29.0a
Anti, 180
9
52:48a
97:3d
81:19c
66:34c
29.9a
aThe first descriptor correspond to the relationship between the OH and the
CMe2 group with respect to the plane of the aromatic ring of indane. The sec-
ond descriptor indicates the dihedral angle of the hydroxyl hydrogen with
respect to the mesityl quaternary carbon.
17.0d
19.6c
70:30
26.7c
Populations percentages (Pop.) are calculated assuming Boltzmann statistics
at T 5 258C.
aCAM-B3LYP/6–31G(d).
bCISD/6–31G(d)//B3LYP/6–31G(d).
cB3LYP/6–31G(d).
dCAM-B3LYP/6–31111G(2d,p).
4.5 kcal/mol range, a barrier hardly observable by the
dynamic NMR technique.29–32 Low-temperature spectra of 3
down to 21608C did not show any further broadening ascrib-
able to the slowdown of the ring inversion. The lowering of
the barrier to the ring inversion could be due to the steric
hindrance caused by the mesityl ring that flattens the enve-
lope of the ring and rises the energy of the ground state,
thus reducing the barrier.}
When a sample of 3 was heated in C2D2Cl4, the signals of
the ortho-methyls remained sharp at all the temperatures and
the large chemical shift separations resulted almost
unchanged even at 11408C, confirming that the rotational
barrier of the mesityl is larger than 22 kcal/mol and that it is
too high to be handled by the Dynamic-NMR technique**;
indeed DFT calculations predict a barrier of 29.0 kcal/mol.
Additionally, the rotational process here corresponds to a
topomerization hence there are no atropisomers to separate
in this case. The two residual enantiomers due to the asym-
metric carbon were separated by enantioselective HPLC on a
Chiralcel AD-H column, to determine the absolute configura-
tion by chirooptical methods. In particular, we used the theo-
retical calculation of the ECD spectra by means of the TD-
DFT method, because this technique has emerged as a reli-
able approach to assign the absolute configuration, and it
has been successfully used many times to predict the ECD
spectra of complex organic molecules.33–38
The conformational search carried out by MM methods
had shown that a second conformation, due to the rotation of
the OH group, was available for each of the two conforma-
tional diastereoisomers. All the four resulting conformers are
comprised in a 1 kcal/mol range, and the simulation of the
ECD spectrum must take into account the contribution of all
of them (Table 2). The electronic excitation energies and
rotational strengths needed for the simulation of the ECD
spectra have been calculated in the gas phase for the four
conformations of 3 using TD-DFT with the 6-31111G(2d,p)
basis set and the CAM-B3LYP39 model. The rotational
strengths were calculated with the dipole-length and the
dipole-velocity gauge and the resulting values are very simi-
lar, the average difference being less than 3%. For this rea-
son, the errors due to basis set incompleteness should be
tively. The huge separation between the methyl signals can
be referred to the ring-current effect that shifts upfield the
methyl (1.40 ppm) if it is oriented inward the indanyl moiety.
Consequently, the phenyl must be nearly perpendicular to
the aromatic part of the indane so that the second ortho-
methyl results oriented toward the geminal methyls and
shifted downfield to 2.66 ppm. These remarks are completely
supported by NOE experiments (Fig. 4). Saturation of the
methyl at 2.66 ppm yields a strong NOE on the meta-hydro-
gen at 6.92§ and other significant NOE effects on the OH
and both the geminal methyls at 1.00 and 1.35 ppm. On the
contrary, the irradiation of the methyl at 1.40 ppm shows a
strong NOE effect on the other meta-hydrogen at 6.65 ppm,§
on the H7 of the indane at 7.2 ppm and on the methyl at 1.00
ppm, whereas no effect is observed on the OH signal.
Finally, the irradiation of the methyl at 1.00 ppm show nearly
equivalent NOEs on both the ortho-methyls proving that they
are at similar distances. This implies that the mesityl and the
indanyl moieties must adopt a propeller-like disposition (see
Fig. 4).27,28
DFT calculations suggest that the ground state structure
of compound 3 adopts a propeller-like disposition between
the two aromatic rings. The energy difference between the
two conformational diastereoisomers, generated by the dif-
ferent dispositions of the CMe2 group, is very small (0.6 or
0.25 kcal/mol, if the free energy is considered), thus both
conformations should be populated when the inversion of
the five-membered ring is frozen. NOE data confirm that the
most populated conformation corresponds to the lowest
energy structure, where the OH and the CMe2 are on the
same side of the plane of the aromatic ring in the indane
(syn). In this conformation, the distances (and thus the NOE
enhancements) between the two geminal methyls of CMe2
and between one of the latter with the two ortho-methyl of
mesitilene are very similar, as experimentally observed (top
trace in Fig. 4).
The calculations also suggest that the barrier for the five-
membered ring inversion is considerably smaller (2.0 kcal/
mol) with respect to that calculated in compound 1. Taking
into account the underestimation of the DFT calculations in
the case of compound 1 (about 1.7 kcal/mol), the ring inver-
sion barrier in compound 3 should be comprised in the 3.5–
}The calculated dihedral angle (Cq–Cq–Cq(OH)–CMe2) is 238 in the case
of 1 and 288 for 3. This indicates that in the case of 3 the five membered
ring is flatter and that in the best conformation of 3 the envelope is reversed
(i.e. the OH and the CMe2 are on the same side) with respect to 1.
§This is a ‘‘control’’ NOE effect that can be used as a distance marker for
the calculation of the distance ratio with other moieties.
**At 11408C a rate constant of 10-20 Hz, corresponding to a barrier of 22
kcal/mol, would be clearly visible as a broadening of the two NMR lines.
Chirality DOI 10.1002/chir

Fig. 5. A: calculated ECD spectra for the four conformation of 3. Calcula-
tions were performed at the TD-DFT CAM-B3LYP/6-31111G(2d,p) level. B:
experimental (black line, 1.1 3 1024 M in acetonitrile) and calculated ECD
spectra, weighed on the calculated free energies (DG8, red line) or on the
total energies (DE8, blue line). Simulated spectra were redshifted by 10 nm
and scaled to match the experimental trace. [Color figure can be viewed in
the online issue, which is available at wileyonlinelibrary.com.]
Fig. 3. Center: the two conformational diastereoisomers of 1. Top: Calcu-
lated transitions state for the ring-flip process that interconverts the anti and
syn diastereoisomers. Bottom: transition state for the rotation of 3,5-dimetyl-
phenyl ring, corresponding to a topomerization process. [Color figure can be
viewed in the online issue, which is available at wileyonlinelibrary.com.]
very small, or negligible,40 and the selected basis set should
be considered suitable to tackle the present case. All the cal-
culations were performed supposing the absolute configura-
tion S, with the results shown in Figure 5. All the ECD spec-
tra were generated by applying a 0.55 eV Gaussian shaped
line width,21 and to generate the spectrum in the 170–400 nm
range, 60 transitions were calculated for each conformation.
The patterns of the simulated spectra for the four confor-
mations are quite similar regarding the pattern, but quite dif-
ferent in the absolute intensity and in the relative intensity of
the two Cotton effects. Therefore, any error in the exact
determination of the conformer’s population could noticeably
influence the shape of the final spectrum. The final simulated
ECD spectra were obtained taking into account the
43:28:19:10 or the 65:23:3:9 population ratios determined
assuming Boltzmann statistics from the calculated free ener-
gies, or from the electronic total energies, at the CAM-
B3LYP/6-31G(d) level, respectively (Table 2). In both cases,
the agreement between the calculated and the experimental
spectra obtained for the first eluted enantiomer is very good,
and the ECD simulations suggest that the absolute configu-
ration of this enantiomer of 3 is 1S. It should be noted that
the spectral simulation obtained using the conformers ratio
derived from the total energies is slightly better than that
obtained from the ZPE-corrected energies.
To experimentally determine the rotational barrier of the
mesityl and to achieve the separation of the stable atropisom-
ers, the symmetry axis (Cipso–Cpara) was removed by insert-
ing a methoxy group in the meta position (compound 4). In
the latter case, the molecule bears two chirality sources (i.e.,
the asymmetric carbon and a chirality axis) so that two dia-
stereoisomers (each one corresponding to a pair of enan-
Fig. 4. NOE spectra (600 MHz in CDCl3) of compound 3. Bottom: control
spectrum, the geminal methyls are marked with g. Second and third traces:
irradiation of the aromatic methyls at 2.66 and 1.40 ppm. Top: irradiation of
the geminal methyl at 1.00 ppm. The 3D structure refers to the most stable
conformation. [Color figure can be viewed in the online issue, which is avail-
able at wileyonlinelibrary.com.]

774
CASARINI ET AL.
tiomers) are expected to exist: in one diastereoisomer (atro-
pisomer), the OMe group is closer to the OH and the gemi-
nal methyls, whereas in the other, the OMe is far from OH
and oriented inward the indanyl moiety. At ambient tempera-
1
ture, the H and 13C spectra of 4 show the presence of the
two atropisomers in a 53:47 ratio.
The stereoisomeric mixture was separated at room tem-
perature by means of a semipreparative enantioselective
HPLC column (chiralcel AD-H, see Fig. S4 in the Supporting
Information), but only three of the four expected stereoiso-
mers could be achieved. The first and the third eluted peaks
corresponded to the two enantiomers of the minor diaster-
eoisomer (mesityl CH at 6.92 ppm), whereas the second
peak corresponded to the major diastereoisomer (mesityl
CH at 6.64 ppm), whose enantiomers were poorly separated.
This pair of enantiomers was eventually resolved using a dif-
ferent enantioselective column (Phenomenex LUX Amylose-
2). The absence of any effect due to in-column interconver-
sion in the chromatograms confirmed that the rotational bar-
rier of the mesithyl group is larger than 23 kcal/mol.41–46
NOE experiments taken on the major atropisomer showed
that the irradiation of the ortho-methyl at 1.29 ppm resulted
in a strong enhancement on the adjacent hydrogen of the
ring and in weaker NOE effects on the H7 and CH2 of the
indanyl moiety. Conversely, the irradiation of the ortho-
methyl at 2.54 ppm showed a strong NOE effect on the adja-
cent OMe, marked enhancements on the OH and both the
geminal methyls of the indane, confirming that in the major
atropisomer the OMe is spatially close to the OH (Fig. S5,
Supporting Information). From an alternative point of view,
the major atropisomer corresponds to a syn-periplanar (sp)
disposition of the OH group with respect to the methoxyme-
sityl.1,47 Taking into account the dihedral angle (Arindanyl)C–
C1–Cipso–Cortho(OMe side), the R*,M* relative configuration
can be assigned to the major atropisomer (see Fig. S6, Sup-
porting Information).
Following the same approach already shown for compound
3, the experimental ECD spectra of a pure enantiomer was
acquired in acetonitrile and simulated by the TD-DFT
method. In the case of compound 4, the results of NOE anal-
ysis helped in the assignment of the absolute configuration,
in that the relative configuration was already assigned.
The minor diastereoisomer had been determined to have
the R*,P* configuration, and the ECD spectrum of the
second eluted enantiomer (Fig. 6, third peak on the AD-H
column) was correctly simulated by the calculations taken on
the R,P enantiomer (see Figs. S6 and S7 and Table S1 of the
Supporting Information for tables and spectra of the popu-
lated conformations). This assignment is further supported
by the comparison with the spectrum of the S enantiomer of
compound 3 that exhibits the mirror image spectrum. The
correspondence with the two peaks observed on the LUX
column for the major diastereoisomer was determined by the
kinetic equilibration of the R,P enantiomer into the R,P/R,M
mixture, because during this equilibration, the rotation of the
methoxymesityl group selectively causes the helicity change
of the chirality axis. When subjected to HPLC analysis, the
R,M atropisomer (major) showed to be the first eluted
on the LUX column. Consequently, the second eluted enan-
tiomer had to be the S,M atropisomer.
Fig. 6. Experimental (black line, 1.2 3 1024 M in acetonitrile) and calcu-
lated ECD spectra, weighed on the calculated free energies (DG8, red line)
or on the total energies (DE8, blue line) of the 28 eluted enantiomer of the
minor atropisomer of 4 (on AD-H column). Calculations were performed at
the TD-DFT CAM-B3LYP/6-31111G(2d,p) level. Simulated spectra were
redshifted by 10 nm and scaled to match the experimental trace. [Color fig-
ure can be viewed in the online issue, which is available at wileyonlinelibrary.
com.]
into the more stable atropisomer (k1; Fig. S8, Supporting In-
formation). A classic kinetic treatment gave a rate constant
(k2) of 1.5 3 1024 s21 for the conversion of the more into the
less stable atropisomer to which corresponds a rotational
barrier of 26.95 kcal/mol that is in good agreement with the
calculated value.
Likewise compound 6, where the aryl substituent is the 1-
naphthyl, lacks any element of symmetry coincident with the
rotational axis and, as for 4, the presence of two stable atro-
pisomers at room temperature is highlighted by two sets of
signals in a 89:11 ratio.
NOE experiments on the signals of the major atropisomer
(Fig. S9, Supporting Information) showed that irradiation of
the methyl at 0.67 ppm yielded a marked enhancement on
the doublet at 6.63 ppm attributed to H20 of the naphthyl moi-
ety, whereas the irradiation of the methyl at 1.31 ppm yields
NOE effects on the OH at 3.97 ppm and on the doublet at
9.26 ppm, assigned to the H80 of the naphthalene (peri hydro-
gen). Conversely, when the OH signal is saturated, NOE
effects were observed on the signals at 1.31, 9.26, and 7.30
ppm, the latter being assigned to the peri hydrogen
(H7) of the indanyl moiety. Finally, the irradiation of H80 at
9.26 ppm that produces NOE effects on the OH and on both
the geminal methyls corroborate a geometry where the
naphthyl moiety has the H20 placed above the indanyl
ring and points inward the molecule, whereas the H80 is
located under the indanyl, close to the OH and the geminal
methyl at 1.31 ppm. From an alternative point of view, the
major atropisomer corresponds to a syn-periplanar (sp) dis-
position of the OH group with respect to naphthalene (see
Fig. 7).
The X-ray structure of 6 (crystals grown from n-hexane,
see Fig. 7 and Supporting Information Fig. S10) confirms
that in the solid state the preferred conformation corre-
sponds to the most populated atropisomer in solution. The
naphthyl and indanyl rings are nearly perpendicular, as the
dihedral angle C2ꢁꢁC1ꢁꢁC10ꢁꢁC20 is 93.78. The H20 is
˚
located above the indanyl ring at a distance of 3.39 A from
0
˚
˚
H7 and H8 , only 2.27 A away from OH, and 2.39A from the
closest hydrogen of the geminal methyl under the indanyl
ring. The shape of the five-membered ring corresponds to an
The rotational barrier of the 3-methoxy-mesityl moiety
could be determined by heating at 1808C an NMR sample of
the most retained enantiomer and following the equilibration
Chirality DOI 10.1002/chir

STEREODYNAMICS OF STEREOLABILE ATROPISOMERS
775
The calculations indicate that also in the case of the anti-
periplanar (ap) atropisomer, where the OH is close to H20,
the CMe2 group is far from the OH (ap-anti); whereas the
remaining conformation, in which the envelope is reversed
(ap-syn) and has a much higher energy (3.5 kcal/mol). When
a sample of 6 is heated, all the NMR signals broaden and
eventually coalesce in the range 85–1008C, yielding sharp
single signals above 1408C. From the line shape simulation
of the CH2 signal, an activation energy of 18.8 kcal/mol (Fig.
8) was derived.
Two pathways are possible for the rotation about the piv-
otal C1ꢁꢁC10 bond that allows the interconversion between
the two atropisomers, namely the crossing of H80 or H20 over
the geminal methyls. DFT calculations indicate that the first
path is higher in energy (26.5 kcal/mol) due to the greater
distortion of the indanyl ring required to allow the leapfrog-
ging of the methyl. Instead, the crossing of H20 over the
geminal methyl needs a maximum energy of 19.6 kcal/mol
even if the H80, at the same time, has to climb over H7 in the
indanyl plane. It has to be noted that the lower calculated
barrier (i.e., the threshold transition state for the rotation of
naphthalene) is actually very close to the experimental value.
The same conformational behavior was also observed in
compound 5, where the naphthyl is replaced by the 7-benzo-
thiophene. NOE experiments (Fig. S11, Supporting Informa-
tion) corroborate the hypothesis that the most populated con-
formation is the same as for 6 (i.e., sp). In addition to that
the presence of a five-membered ring and the sulphur atom
in the peri position reduce the steric clashing in the transi-
tion state, thus a smaller barrier is expected to occur. 1H
NMR spectra of 5 at variable temperature showed at 1388C,
the maximum broadening for the methyl at 1.31 ppm and
this allowed to estimate a rotational barrier of 16.7 kcal/mol.
Furthermore, on cooling at 2308C, the same methyl signal
Fig. 7. Calculated conformational diastereoisomers of 6 and experimental
X-ray structure. [Color figure can be viewed in the online issue, which is
available at wileyonlinelibrary.com.]
envelope in which the CMe2 is out of the plane of the phenyl
by 21.88 (Cq–Cq–CH2–CMe2 dihedral angle) and on the
same side of the OH (thus sp-syn). This shape was correctly
simulated by the calculations in the case of compound 1,
where two conformations, resulting from the frozen inversion
of the cycle, were detected. This is a further confirmation of
the accuracy and reliability of the DFT approach. Taking into
account the rotational asymmetry of naphthalene and the two
available dispositions of the five-membered ring, a total of
four conformational diastereoisomers can be envisaged (sp-
syn, sp-anti, ap-syn, ap-anti).
DFT calculations show that the conformation derived by
the NOE experiments, with the OH group close to H80 is the
most stable. The other conformation, due to 1808 rotation of
the naphthalene (OH close to H20, antiperiplanar), is calcu-
lated to be 0.9 kcal/mol less stable, in good agreement with
the experimental ratio (see Table 1). The calculations also
suggest that the best conformation of the syn-periplanar (sp)
atropisomer is sp-anti (see Fig. 7) and this does not corre-
spond to that observed in the solid state. The latter confor-
mation (sp-syn) has a calculated energy of 0.7 kcal/mol
higher that the former (see also Fig. S10 in the Supporting
Information). This mismatch is probably due to the con-
straints imposed by the crystal lattice. However, it is impossi-
ble to derive from the NOE data whether the most populated
conformation in solution corresponds to the X-ray conforma-
tion or to the calculated one. Both conformations are prob-
ably populated and the very fast ring inversion implies the
observation of averaged NOEs effects. For this reason, the
saturation of H80 at 9.26 ppm yields identical NOE enhance-
ments on both the geminal methyls of indane (top trace in
Fig. S9, Supporting Information).yy
Fig. 8. 1H NMR spectra (600 MHz in DMSO-d6) of compound 6. Left:
experimental traces at different temperatures referring to the indanyl CH2.
Right: simulated traces and calculated rate constants (k) at the same tempera-
tures.
yyThese enhancements should be quite different if only one of the two con-
formations were populated.
Chirality DOI 10.1002/chir

776
CASARINI ET AL.
2.92 ppm, the geminal methyl at 1.40 ppm and the 2-Me of
the naphthyl at 1.50 ppm (Fig. S13, Supporting Information).
This suggests a preferred conformation in which the naph-
thyl is turned nearly orthogonal to the indanyl ring and the
2-Me group, on the opposite side of the OH (thus syn-peripla-
nar). Taking into account the dihedral angle between the
naphthyl ring and the indane, the syn-periplanar pair of enan-
tiomers has the R*,M* relative configuration, whereas the
anti-periplanar pair has the R*,P* relative configuration. DFT
calculations indicate the syn-periplanar atropisomer to be
more stable than the anti-periplanar by 0.4 kcal/mol (after
ZPE correction), in good agreement with the experimental
value (0.5 kcal/mol). The rotational barrier of the 2-methyl-
naphtyl was determined at 1708C on the first eluted peak of
the minor atropisomer by a classical kinetic equilibration in
DMSO-d6, from the less to the more stable atropisomer (Fig.
S14, Supporting Information). A rate constant (k1) of 4.0 3
1024 s21 was obtained, corresponding to a rotational barrier
of 25.5 kcal/mol that turned out to be in good agreement
with the DFT calculated value (26.7 kcal/mol).
As in the cases of compounds 3 and 4, the absolute con-
figuration of the atropisomers of 7 was determined by the
simulation of the ECD spectra. For each atropisomer, the
two different conformations of the five-membered ring and
the two orientations of the OH had to be considered, yielding
a total of 8 stereoisomers (see Table S2 in the Supporting In-
formation). DFT optimization (B3LYP/6-31G(d) level) of
each stereoisomer suggested that all of them had to be con-
sidered in the simulation of the ECD spectrum.
The electronic excitation energies and rotational strengths
were calculated in the gas phase for the four conformations
of 7-ap and 7-sp using TD-DFT with the 6-31111G(2d,p) ba-
sis set and the CAM-B3LYP model. All the calculations were
performed supposing the S absolute configuration, and all
the ECD spectra were generated by applying a 0.55 eV Gaus-
sian shaped line width.21 To generate the spectrum in the
170–400 nm range, 75 transitions were calculated for each
conformation (see Fig. S15 of the Supporting Information for
the calculated ECD spectrum of each conformation). The
final simulated ECD spectra (Fig. 9) were obtained taking
into account the population ratios determined assuming
Boltzmann statistics from the calculated free energies or
from the electronic total energies (Table S2 of the Support-
ing Information).
Fig. 9. A: Enantioselective HPLC trace of compound 7 on Chiralcel AD-H
(eluent: hexane/iPrOH 90:10 v/v). B: Experimental (black line, 1.0 3 1024
M in acetonitrile) and calculated ECD spectra for the first eluted enantiomer
of the minor atropisomer. C: Experimental (black line, 1.2 3 1024 M in aceto-
nitrile) and calculated ECD spectra for the second eluted enantiomer of the
major atropisomer (fourth peak). The red lines correspond to the simulation
weighed on the calculated free energies (DG8), blue lines correspond to the
simulations weighed on the calculated total energies (DE8). Simulated spec-
tra were redshifted by 10 nm and multiplied by 1.1 in order to match the ex-
perimental trace. [Color figure can be viewed in the online issue, which is
available at wileyonlinelibrary.com.]
sharpens and a partner peak (population 4%) appears at a
lower frequency with a separation of 43 Hz (Fig. S12, Sup-
porting Information). Most likely, such a large difference
between the two atropisomer populations could be due to an
hydrogen bonding interaction between the OH and the sul-
phur atom that favors the sp-atropisomer.
CONCLUSIONS
Similar to compounds 1 and 3, when the H20 of the naph-
thyl moiety is replaced by a methyl, as in compounds 7, a
larger rotational barrier is expected, because the methyl is
For the investigated 1-arylindanols (1–7), we have shown
that the barrier involved in the rotational process about the
pivotal bond C1(indanyl)ꢁꢁC10 (aryl) cover the wide range
10–26 kcal/mol, and such results are strongly dependent on
the hindrance of the aryl and the indanyl moieties. When the
ortho-aryl hydrogens are replaced with methyls, the rota-
tional process is already locked at room temperature. Fur-
thermore, if the rotational axis is not coincident with the local
symmetry axis of the aryl moiety two diastereoisomers are
generated, each one as a pair of enantiomers, because of the
chiral center on the indane framework. In the cases of 4 and
7, all the four stereoisomers have been separated by enantio-
selective HPLC and their relative molecular geometries have
been assigned by NMR, as well as the rotational barriers for
their interconversion were determined by kinetic equilibra-
tion. These results were further corroborated from detailed
1
again forced to pass over the indanyl moiety. Indeed, H and
13C NMR spectra of 7 at room temperature show two groups
of signals in a 70:30 ratio corresponding to the two stable
atropisomers. Unlike 4, the expected pair of enantiomers for
each diastereoisomer, in the ratio estimated by NMR, have
been nicely separated by enantioselective HPLC on a semi-
preparative chiralcel AD-H column eluted with n-hexane/ iso-
propyl alcohol 95:5 (Fig. 9).
NOE experiments on the major atropisomer of 7 show
that irradiation of the OH at 5.53 ppm yields enhancements
on the H7 at 7.30 ppm, on the methyl at 1.40 ppm and the
H80 of the naphthyl ring at 9.16 ppm. Conversely, irradiation
of the Me at 0.91 ppm give NOE effects only on the CH2 at
Chirality DOI 10.1002/chir

STEREODYNAMICS OF STEREOLABILE ATROPISOMERS
777
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 D.01. Wallingford CT: Gaussian, Inc.; 2005.
computations of the ground and transition states. These
enabled the calculation of the rotational barriers and the sim-
ulation ECD spectra, from which the absolute configuration
of each stereoisomer could be reliably assigned.
18. Frisch M J, Trucks GW, Schlegel HB, Scuseria GE, Robb MA, Cheese-
man JR, Scalmani G, Barone V, Mennucci B, Petersson GA, Nakatsuji H,
Caricato M, Li X, Hratchian HP, Izmaylov AF, Bloino J, Zheng G, Sonnen-
berg JL, Hada M, Ehara M, Toyota K, Fukuda R, Hasegawa J, Ishida M,
Nakajima T, Honda Y, Kitao O, Nakai H, Vreven T, Montgomery JA, Jr,
Peralta JE, Ogliaro F, Bearpark M, Heyd JJ, Brothers E, Kudin KN, Star-
overov VN, Kobayashi R, Normand J, Raghavachari K, Rendell A, Burant
JC, Iyengar SS, Tomasi J, Cossi M, Rega N, Millam NJ, 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,
ACKNOWLEDGMENTS
The authors thank Prof. L. Lunazzi for reading the manu-
script and to CIGAS, University of Basilicata, for running the
ESI-HRMS analyses.
LITERATURE CITED
1. Casarini D, Lunazzi L, Mazzanti A, Foresti E. Conformational studies by
dynamic NMR. 64. Stereomutations of atropisomers and of conforma-
tional enantiomers in ethers of hindered naphthylcarbinols. J Org Chem
1998;63:4746–4754.
¨
Dapprich S, Daniels AD, Farkas O, Foresman JB, Ortiz JV, Cioslowski J,
Fox DJ. Gaussian 09, Revision A.1. Wallingford, CT: Gaussian, Inc.,
2009.
19. Ayala PY, Schlegel HB. Identification and treatment of internal rotation
in normal mode vibrational analysis. J Chem Phys 1998;108:2314–2325.
2. Casarini D, Coluccini C, Lunazzi L, Mazzanti A. Stereolabile and configu-
rationally stable atropisomers of hindered aryl carbinols. J Org Chem
2005;70:5098–5102.
20. Casarini D, Lunazzi L, Mazzanti A. Recent advances in stereodynamics
and conformational analysis by dynamic NMR spectroscopy and theoreti-
cal calculations. Eur J Org Chem 2010;2035–2056.
3. Casarini D, Lunazzi L, Mazzanti A, Mancinelli M. Conformation and ster-
eodynamics of symmetrically ortho-disubstituted aryl carbinols and aryl
ethers. J Org Chem 2007;72:998–1004.
21. Donnington RD, Keith TA, Millam JM. Gaussview 5.0.9, Semichem, 2008.
22. Lomas JS, Luong PK, Dubois JE. Nucleophilic addition of o-tolyllithium
compounds to di-tert-butyl ketone. Thermal and organolithium-catalyzed
4. Casarini D, Lunazzi L, Mazzanti A, Mancinelli M. Structure, conforma-
tions, stereodynamics, dimer formation and absolute configuration of axi-
ally chiral atropisomers of hindered biphenyl carbinols. J Org Chem
2007;72:7667–7676.
isomerization of o-tolyldi-tert-butylcarbinol rotamers.
J Org Chem
1977;42:3394–3399.
23. Lomas JS, Anderson JE. Synthesis and very slow sp2-sp3 rotation of anti-
5. Cheer JC, Johonson CR. The stereoselective rearrangements of confor-
mationally mobile epoxides. J Am Chem Soc 1968;90:178–183.
o-tolyldi(1-adamantyl)methane. J Org Chem 1995;60:3246–3248.
24. Jennings WB. Chemical shift nonequivalence in prochiral groups. Chem
Rev 1975;75:307–322.
6. Palmer M, Walsgrove T, Wills M. (1R,2S)-(1)-cis-1-Amino-2-indanol: an
effective ligand for asymmetric catalysis of transfer hydrogenations of
ketones. J Org Chem 1997;62:5226–5228.
25. Casarini D, Lunazzi L, Mazzanti A. Correlated rotations in benzylfluorene
derivatives: structure, conformation and stereodynamics. J Org Chem
2007;73:2811–2818.
7. Brewster JH, Buta JG. Conformational mobility and optical rotation
effects of aromatic nuclei. J Am Chem Soc 1966;88:2233–2240.
26. Casarini D, Lunazzi L, Mazzanti A. Steric effects which determine the
conformational preferences and stereodynamic processes of aryl fluore-
nyl ketones. Org. Biomol Chem 2009;7:1619–1626.
8. Al-Saadi AA, Wagner M, Laane J. Spectroscopic and computational stud-
ies of the intramolecular hydrogen bonding of 2-indanol. J Phys Chem A
2006;110:12292–12297.
27. Grilli S, Lunazzi L, Mazzanti A, Casarini D, Femoni C. Conformational
studies by dynamic NMR. 78. Stereomutation of the helical enantiomers
of trigonal carbon diaryl-substituted compounds: dimesitylketone, -thio-
ketone and -ethylene. J Org Chem 2001;66:488–495.
9. Pincock JA, Wedge PJ. The photochemistry of conformationally rigid
benzilic esters:2,2-dimethyl-1-indanyl acetates and pivalates. J Org Chem
1995;60:4067–4076.
10. Orliac-le Moing A, Delauny J, LeBouc A, Simonet J. Reduction electrochi-
mique des derives carbonyles insatures. IX. Indanones et tetralones: elec-
trolyses mixtes en presence d’electrophiles. Tetrahedron 1985;41:4483–
4493.
28. Mislow K. Stereochemical consequences of correlated rotation in molecu-
lar propellers. Acc Chem Res 1976;9:26–33.
29. Mazzanti A, Lunazzi L, Ruzziconi R, Spizzichino S, Schlosser M. The tor-
sional barriers of 2-hydroxy- and 2-fluorobiphenyl: small, but measurable.
Chem Eur J 2010;16:9186–9192.
11. Fuson RC, Corse J, Welldon PB. Enediols. VIII. Methoxystilbenediols J
Org Chem 1941;63:2645–2648.
30. Anet FAL, Chmurni GN, Krane J. Ring inversion in cyclohexanone. J Am
Chem Soc 1973;95:4423–4424.
12. Bradley SA, Krishnamurthy K. A modified CRISIS-HSQC for band-selec-
tive IMPRESS. Magn Res Chem 2005;43:117–123.
31. Pawar DM, Miggins SD, Smith SV, Noe EA. Conformational study of cis-
cyclononene by dynamic NMR spectroscopy and computational methods.
J Org Chem 1999;64:2418–2421.
13. Willker W, Leibfritz D, Kerssebaum R, Bermel W. Gradient selection in
inverse heteronuclear correlation spectroscopy. Magn Res Chem
1993;31:287–292.
32. Casarini D, Grilli S, Lunazzi L, Mazzanti A. Conformational studies
by dynamic NMR. 80. Cog-Wheel effect in the stereolabile helical enan-
tiomers of dimesityl sulfoxide and sulfone. J Org Chem 2001;66:2757–
2763.
14. Hurd RE, John BK. Gradient-enhanced proton-detected heteronuclear
multiple-quantum coherence spectroscopy. J Magn Reson 1991;91:648–
653.
15. Stott K, Stonehouse J, Keeler J, Hwand TL, Shaka AJ. Excitation sculpt-
ing in high-resolution nuclear magnetic resonance spectroscopy: Applica-
tion to selective NOE experiments. J Am Chem Soc 1995;117:4199–4200.
33. Bringmann G, Bruhn T, Maksimenka K, Hemberger Y. The assignment
of absolute stereostructures through quantum chemical circular dichro-
ism calculations. Eur J Org Chem 2009:2717–2727.
ˇ
16. Kupce E, Boyd J, Campbell ID. Short selective pulses for biochemical
34. Stephens PJ, Pan JJ, Devlin FJ, Krohn K, Kurta´n K. Determination of the
absolute configurations of natural products via density functional theory
calculations of vibrational circular dichroism, electronic circular dichro-
ism, and optical rotation: The iridoids plumericin and isoplumericin. J
Org Chem 2007;72:3521–3536.
applications. J Magn Reson Series B 1995;106:300–303.
17. Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, Robb MA, Cheese-
man 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,
´
35. Kwit M, Rozwadowska MD, Gawronski J, Grajewska A. Density func-
tional theory calculations of the optical rotation and electronic circular
dichroism: The absolute configuration of the highly flexible trans-isocy-
toxazone revised. J Org Chem 2009;74:8051–8063.
36. Gioia C, Fini F, Mazzanti A, Bernardi L, Ricci A. Organocatalytic asym-
metric formal [3 1 2] cycloaddition with in situ-generated N-carbamoyl
nitrones J Am Chem Soc 2009;131:9614–9615.
Chirality DOI 10.1002/chir

778
CASARINI ET AL.
37. Pescitelli G, Di Pietro S, Cardellicchio C, Annunziata M, Capozzi M, Di Bari
L. Systematic investigation of CD spectra of aryl benzyl sulfoxides inter-
preted by means of TDDFT calculations. J Org Chem 2010;75:1143–1154.
42. Trapp O. unified equation for access to rate constants of first-order reac-
tions in dynamic and on-column reaction chromatography. Anal Chem
2006;78:189–198.
43. Wolf C. Stereolabile chiral compounds: analysis by dynamic chro-
matography and stopped-flow methods. Chem Soc Rev 2005;34:595–608.
38. Bencivenni G, Wu L-Y, Giannichi B, Mazzanti A, Pesciaioli F, Song M-P,
Bartoli G, Melchiorre P. Targeting structural and stereochemical complexity
by asymmetric organocascade: construction of spirocyclic oxindoles having
multiple stereocenters. Angew Chem Int Ed Engl 2009;48:7200–7203.
44. Trapp O, Shellie R, Marriott P, Schurig V. Simulation of elution profiles
for two-dimensional dynamic gas chromatographic experiments. Anal
Chem 2003;75:4452–4461.
39. Yanai T, Tew D, Handy N. A new hybrid exchange–correlation functional
using the Coulomb-attenuating method (CAM-B3LYP). Chem Phys Lett
2004;393:51–57.
45. Trapp O, Schurig V. Stereointegrity of Tro¨ger’s base: gas-chromato-
graphic determination of the enantiomerization barrier. J Am Chem Soc
2000;122:1424–1430.
40. Stephens PJ, McCann DM, Devlin FJ, Cheeseman JR, Frisch MJ. Deter-
mination of the absolute configuration of [32](1,4)barrelenophanedicarbo-
nitrile using concerted time-dependent density functional theory calcula-
tions of optical rotation and electronic circular dichroism. J Am Chem
Soc 2004;126:7514–7521.
46. Alcaro S, Casarini D, Gasparrini F, Lunazzi L, Villani C. Atropisomerism
in hindered naphthyl sulfones investigated by dynamic NMR and
dynamic HPLC techniques. J Org Chem 1995;60:5515–5519.
47. Casarini D, Lunazzi L, Mazzanti A. Conformational studies by dynamic
nuclear magnetic resonance. Part 59. Stereodynamics of conformational
enantiomers in the atropisomers of hindered naphthilcarbinols. J Org
Chem 1997;62:3315–3323.
41. D’Acquarica I, Gasparrini F, Pierini M, Villani C, Zappia G. Dynamic
HPLC on chiral stationary phases: a powerful tool for the investigation of
stereomutation processes. J Sep Sci 2006;29:1508–1516.
Chirality DOI 10.1002/chir