Iminium ion catalysis: The enantioselective Friedel-Crafts alkylation-acetalization cascade of naphthols with α,β-unsaturated cyclic ketones

Article abstract of DOI:10.1039/c2cc35582e

The first enantioselective Friedel-Crafts alkylation-acetalization-cascade of naphthols with α,β-unsaturated cyclic ketones and 1H-inden-1-ones was realized. 9-Amino(9-deoxy)epi-quinine A was the catalyst of choice for the realization of polycyclic structures with high enantiocontrol.

Full text of DOI:10.1039/c2cc35582e

Electronic Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2012
Iminium ion catalysis: the enantioselective Friedel‐Crafts
alkylation‐acetalization cascade of naphthols with α,β‐
unsaturated cyclic ketones
Enrico Paradisi, Paolo Righi, Andrea Mazzanti, Silvia Ranieri and Giorgio Bencivenni*
Dipartimento di Chimica Organica “A. Mangini”, Alma Mater Studiorum – Università di Bologna, Viale
del Risorgimento 4, 40136 Bologna, Italy.
E‐mail: giorgio.bencivenni2@unibo.it
Phone: +390512093624
Fax: +390512093654
Supplementary Information
S1

Electronic Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2012
Table of Contents
General Information
S3
S3
S4
Materials
Determination of diastereomeric ratios and enantiomeric purity
Optimization data for ketone 1a and β‐naphthol 2a
S5
Conformational analysis and absolute configuration determination S8
Crystallographic data for compound 10de
Crystallographic data for compound 6da
S20
S21
Relative Configuration assignment of ent‐5da and ent‐6da
Relative Configuration assignment of ent‐5ca
Experimental procedures
S22
S29
S31
1
13
Copy of H‐ and C‐NMR spectra
S42
S67
HPLC traces
S2

Electronic Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2012
General informations.
1
13
The H and C NMR spectra were recorded on a Varian inova 300, at 300 MHz and 75 MHz
respectively, Varian mercury 400, at 400 MHz and 100 MHz respectively, or Varian inova 600, at 600
1
13
MHz and 150 MHz respectively. The chemical shifts (δ) for H and C are given in ppm relative to
residual signals of the solvents (CDCl , DMSO‐d ). The following abbreviations are used to indicate the
multiplicity: s, singlet; d, doublet; t, triplet; q, quartet; m, multiplet; bs, broad signal. CDCl was passed
over a short pad of alumina before use. Coupling constants are given in Hz. When 2D‐NMR were not
3
6
3
13
performed, the carbon types were determined from DEPT C NMR experiments. NOE spectra were
1
recorded using the DPFGSE‐NOE sequence using a mixing time of 2.00 s and “rsnob” 20 ‐ 50 Hz wide
selective pulses, depending on the crowding of the spectra region. High Resolution Mass spectra
(
HRMS) were obtained from the Department of Organic Chemistry “A. Mangini” Mass Spectroscopy
facility, on a Thermo‐Finnigan MAT 95 XP spectrometer. X‐ray data were acquired at the Department
of Physical and Inorganic Chemistry X‐ray Crystallography facility, on a Bruker APEX‐2 difractometer.
t
D
Optical rotations were measured on a Perkin‐Elmer 341 polarimeter and reported as follow: [α] (c in
g per 100 mL, solvent). Thin Layer Chromatography (TLC) was performed on commercially available
Fluka TLC plates on aluminium or PET foils with fluorescent indicator at 254 nm, using UV light as the
visualizing agent and an acidic mixture of ceric ammonium molybdate or basic aqueous potassium
permangante (KMnO ), and heat as developing agents.
4
Purification of the products was carried out by flash chromatography (FC) on silica gel (Aldrich, 230‐
2
4
00 mesh) according to the method of Still . Organic solutions were concentrated under reduced
pressure on a Büchi rotary evaporator.
Materials
All the commercially available reagents and solvents were used without any further purifications;
3
otherwise, where necessary, they were purified as recommended . Chiral primary amine catalysts 9‐
amino(9‐deoxy)epi‐quinine A and its pseudo‐enantiomer 9‐amino(9‐deoxy)epi‐quinidine ent‐A were
4
synthesized according to literature procedures .
1
(a) K. Stott, J. Stonehouse, J. Keeler, T.‐L. Hwand and A. Shaka, J. Am. Chem. Soc. 1995, 117, 4199. (b) K. Stott, J. Keeler, Q.
N. Van and A. J. Shaka, J. Magn. Resonance 1997, 125, 302. (c) Q. N. Van, E. M. Smith and A. J. Shaka, J. Magn. Resonance
999, 141, 191. (d) See also: Claridge, T.D.W. High Resolution NMR Techniques in Organic Chemistry; Pergamon:
1
Amsterdam, 1999.
2
W. C. Still, M. Kahn and A. J. Mitra, J. Org. Chem. 1978, 43, 2923.
W. L. F. Armarego and D. D. Perrin, In Purification of Laboratory Chemicals, 4th ed.; Butterworth Heinemann: Oxford,
3
1
996.
4
(a) S. H.; McCooey and S. J.; Connon Org. Lett. 2007, 9, 599; (b) W. Chen, W. Du, Y.‐Z. Duan, Y. Wu, S.‐Y. Yang and Y.‐C.
Chen, Angew. Chem. Int. Ed. 2007, 46, 7667.
S3

Electronic Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2012
All the ketones and the naphthols were purchased from Sigma‐Aldrich or Alfa Aesar and used as
5
6
received. Compounds 1c‐d and 9a‐d were prepared following the literature procedures Tert‐butyl
7
(
3‐hydroxynaphthalen‐2‐yl)carbamate 2d was synthesized according to the literature procedure .
Tert‐butyl (4‐hydroxynaphthalen‐1‐yl)carbamate 4d was synthesized as follows:
In a flame‐dried flask equipped with a condenser and a magnetic stirring bar, triethylamine (0.344 ml,
1
eq.) was added to a 0.5 M solution of 4‐amino‐1‐naphthol hydrochloride (2.56 mmol, 0.5 g, 1 eq.) in
anhydrous THF (5.11 ml). The reaction mixture was left to stir at r.t. for 5 min. Then di‐tert‐butyl
dicarbonate (0.558 g, 1 eq.) was added as a solid and the reaction mixture was stirred and refluxed for
three days. Then it was cooled to room temperature and the solvent was removed in vacuo. The
crude product was purified through flash chromatography on silica gel (Hex/EtOAc 70:30) to give a
violet solid, which was crystallized from Et
2
O/Hex to afford 4d in 57% yield as a pink solid.
) : δ 1.58 (s, 9H), 6.41 (d, 1H, J = 7.8 Hz), 6.49 (bs, 1H), 6.71 (bs, 1H), 7.22 (d,
H, J = 8.1 Hz), 7.40 (t, 1H, J = 8.2 Hz), 7.51 (dt, 1H, J = 8.2 Hz, J = 1.5 Hz), 7.84 (d, 1H, J = 8.4 Hz), 8.05
d, 1H, J = 8.4 Hz).
1
8
H NMR (300 MHz, CDCl
3
1
(
a
b
Determination of diastereomeric ratios and enantiomeric purity.
1
Diastereomeric ratios was determined by H NMR spectroscopy of the crude product. Enantiomeric
excesses were determined, after purification, through HPLC analysis on chiral stationary phase
performed on an Agilent 1100‐series instrumentation using Daicel Chiralpak AD‐H, Daicel Chiralpak
AS‐H, Daicel Chiralcel OD‐H, Daicel Chiralcel OJ‐H, Phenomenex Lux‐Amilose 2 and Phenomenex Lux‐
Cellulose 2 columns. Racemic samples of compounds 3ab, 3ac, 5ab, 5ac, 10aa were obtained
performing the reaction with p‐anisidine 30 mol% and 5‐nitrosalicylic acid 60 mol% as catalyst
combination. All the other racemic samples were prepared by mixing the two product antipodes
obtained performing the reaction with catalyst 9‐amino(9‐deoxy)epi‐quinine A and its pseudo‐
enantiomer 9‐amino(9‐deoxy)epi‐quinidine ent‐A separately.
5
(a) B.‐D. Chong, Y.‐I. Ji, S.‐S. Oh, J.‐D. Yang, W. Baik and S. Koo J. Org. Chem. 1997, 62, 9323; (b) Y. Ergün, N. Bayraktar, S.
Patir and G. Okay, J. Hetrocyclic Chem. 2000, 37, 11.
6
a) L. Minuti, A. Taticchi, E. Gacs‐Baitz and A. Marrocchi, Tetrahedron, 1995, 51, 8953; b) E. Zimmerman and V.
Suryanarayan, Eur. J. Org. Chem. 2007, 4091.
7
S. Kumar, D. Hernandez, B. Hoa, Y. Lee, J.‐S. Yang and A. McCurdy, Org. Lett., 2008, 10, 3761.
Bachir Latli, J. Label Compd. Radiopharm. 2004; 47, 847.
8
S4

Electronic Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2012
Optimization data for ketone 1a and β‐naphthol 2a
Table 1. Catalyst screening
Entry
1a:2a
1.1:1
1.1:1
1.1:1
1.1:1
1.1:1
Cat. (mol%)
A (20)
Acid (mol%)
TFA (40)
TFA (40)
TFA (40)
-
t (h)
96
48
96
48
48
48
96
48
Yield %
ee %
49
45
45
-
-
-
45
45
1
2
3
4
5
87
12.5
89
0
0
-
B (20)
C (20)
-
-
TFA (40)
-
TFA (40)
TFA (40)
1
.1:1
ent-A (20)
ent-A (20)
ent-C (20)
7
8
1.1:1
1.1:1
42
60
9
1.1:1
3:1
ent-A (20)
TFA (60%)
TFA (40)
96
96
96
63
33
53
10
A (20)
O
N
O
N
OH
H
N
N
H
H
N
NH2
H N
NH2
2
N
ent-A
B
A
9
-amino(9-deoxy)epi-
9-amino-9-deoxy-6'-
hydroxyepiquinine
9
-amino(9-deoxy)epi-
quinidine
quinine
O
O
H
N
N
H
NH2
H N
2
N
N
C
ent-C
9
-amino(9-deoxy)epi- 9-amino(9-deoxy)epi-
dihydroquinine dihydroquinidine
S5

Electronic Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2012
Table 2. Acid screening
Entry
Acid
benzoic acid
p-nitrobenzoic acid
3,5-dinitrobenzoic acid
o-nitrobenzoic acid
Yield (%)
ee (%)
32
71
56
74
74
57
82
76
73
78
19
n.d.
-
1
2
3
4
5
6
7
8
9
29
58
28
82
45
61
78
81
61
58
43
12
0
o-hydroxybenzoic acid
o-fluorobenzoic acid
2-hydroxy-5-nitrobenzoic acid
1-hydroxy-2-naphtoic acid
2-hydroxy-3-nitrobenzoic acid
2-hydroxy-1-naphtoic acid
diphenyl hydrogen phosphate
(R)-BINOL-hydrogenphosphate
p-TsOH
10
11
12
13
14
15
2-mercaptobenzoic acid
Anthranilic acid
57
63
53
0
S6

Electronic Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2012
Table 3. Solvent screening
Entry
Solvent
Toluene
dry Toluene
Water
t (d)
3
3
3
3
3
3
3
2
2
2
2
2
2
3
3
3
Yield (%)
ee (%)
82
84
46
n.d.
-
1
2
3
4
5
6
7
8
9
78
82
43
6
0
16
0
52
18
12
5
Et
2
O
MeOH
EtOAc
THF
n.d.
-
CHCl
3
80
52
n.d.
n.d.
-
80
80
68
79
76
73
Hexane
dioxane
MTBE
1
1
1
1
1
1
1
1
1
0
1
2
3
4
5
6
7
8
HFIP
Fluorobenzene
DCM
p-Xylene
Clorobenzene
1,2-DCE
0
46
47
43
89
24
83
3
3
Toluene/Brine 1:1
S7

Electronic Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2012
Conformational analysis and absolute configuration determination.
Compounds 10
Good crystals suitable for X‐ray diffraction were obtained for compound 10de by slow
evaporation of a methanol solution. The anomalous scattering determination of the absolute
configuration was possible thanks to the presence of the chlorine atom. The S configuration was
determined for the selected crystal, and its relationship to the major enantiomer obtained with ent‐A
catalyst, was confirmed by means of enantioselective HPLC analysis of the very same crystal used for
X‐ray analysis (this was not straightforward, since the crystals were obtained from a 81% ee mixture
of enantiomers). The crystal cell contained two conformations of the S enantiomers, that were
different in the orientation of the OMe group on the naphthalene ring (see below for refinement
details).
Figure S1: X‐Ray structure of 10de. Two different conformations with the same S absolute
configuration represent the asymmetric unity.
Since the prepared compounds belong to four different classes, a straight relationship of the
stereochemical course of the reaction is likely, but in principle it could not be safely assumed.
However, despite many attempts, enantiopure crystals of other compounds containing a suitable
heavy atom could not be obtained. In the remaining cases, the X‐ray analysis could not determine the
absolute configuration. For these reason we switched to a different approach based on
conformational analysis and chirooptical methods.
The determination of the absolute configurations (AC) of chiral molecules using chiroptical
techniques like optical rotation (OR), electronic circular dichroism (ECD), and vibrational circular
dichroism (VCD) has gained feasibility and reliability because of the development of reliable methods
for the prediction of these properties based on density functional theory (DFT) and on its Time‐
9
Dependent formalism (TD‐DFT). In the present case the theoretical calculation of ECD spectra was
9
For recent examples: H. Hussain, K. Krohn, I. Ahmed, S. Draeger, B. Schulz, S. Di Pietro and G. Pescitelli Eur. J. Org. Chem.
2012, 1783; X‐F Hou, S. Yao, A. Mándi, T. Kurtán, C‐P. Tang, C‐Q. Ke, X‐Q. Li and Y. Ye. Org. Lett. 2012, 14, 460; Y‐S. Cai, T.
Kurtán, Z‐H. Miao, A. Mándi, I. Komáromi, H‐L. Liu, J. Ding, and Y‐W Guo J. Org. Chem. 2011, 76, 1821; M, Woźnica, A.
Butkiewicz, A. Grzywacz, P. Kowalska, M. Masnyk, K. Michalak, R. Luboradzki, F. Furche, H. Kruse, S. Grimme and J. Frelek.
S8

Electronic Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2012
selected for the absolute configuration assignment. From a conformational point of view, the rigidity
of the scaffold of compounds 3, 5, 10 and 11 reduces the number of conformations to be considered
10
and thus simplify the conformational analysis. In addition to this, the configuration assignment by
chirooptical methods of a sample compound of the 10 series can provide information on the reliability
of this method applied to the present compounds.
A preliminary conformational search on 10aa was carried out using Monte Carlo method together
with the MMFF94 molecular mechanics force field (as implemented in Titan 1.0.5, Wavefunction inc.).
All the conformations within a 10 kcal/mol window were then optimized using DFT at the B3LYP/6‐
1011
3
1G(d) level , the harmonic vibrational frequencies of each conformation were calculated at the
same level to confirm their stability (no imaginary frequencies were observed) and to evaluate the
free energy of each conformation.
Figure S2: 3D view of the two most stable conformations of 10aa, calculated at the B3LYP/6‐31G(d)
level. Energy differences are in kcal/mol and represent ZPE–corrected free energies in standard
conditions.
After DFT minimization only two conformations were found to be enclosed in the 10 kcal/mol
J. Org. Chem. 2011, 76, 3306. For reviews see: G. Bringmann, T. Bruhn, K. Maksimenka and Y. Hemberger. Eur. J. Org.
Chem. 2009, 2717. T.D. Crawford; M.C. Tam and M.L. Abrams, J. Chem. Phys. A 2007, 111, 12057. For a review on
conformational analysis for the AC determination see: A. Mazzanti and D. Casarini, D. WIREs Comput. Mol. Sci. 2012, 2,
6
13.
10
P.L. Polavarapu; E.A. Donahue; G. Shanmugam; G. Scalmani; E.K. Hawkins; C. Rizzo; I. Ibnusaud; G. Thomas; D. Habel and
D. Sebastian; J Phys Chem A 2011, 115, 5665.
11
Program Gaussian 09, rev A.02. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G.
Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino,
G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O.
Kitao, H. Nakai, T. Vreven, J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V.
N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N.
Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O.
Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P.
Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, Ö. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski and D. J. Fox,
Gaussian, Inc., Wallingford CT, 2009.
S9

Electronic Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2012
window (Figure S2). These correspond to two conformational diastereoisomers due to the rotation on
the indane ring that is almost perpendicular to the napthol ring in the ground states. The electronic
excitation energies and rotational strengths have been calculated in the gas phase for the two
conformations of 10aa using TD‐DFT with four different methods (functionals) to ascertain if different
calculations provide different shapes of the simulated spectra. The simulation were performed with
12
13
the hybrid functionals BH&HLYP and M06‐2X, the Long‐range Correlated LC‐B97XD that includes
14
empirical dispersion, and CAM‐B3LYP that includes long range correction using the Coulomb
1
5
Attenuating Method. . All the calculations employed the 6‐311++G(2d,p) basis set, that proved to be
16
sufficiently accurate at a reasonable computational cost.
Figure S3: ECD simulations for the two conformations of 10aa, obtained with the four different
functionals and the same 6‐311++G(2d,p) basis set.
The rotational strengths were calculated in both length and velocity representation with the
resulting values being very similar. For this reason the errors due to basis set incompleteness should
17
be considered very small, or negligible. All the calculations were performed supposing S Absolute
Configuration, with the results shown in Figure S3: ECD simulations for the two conformations of
1
0aa, obtained with the four different functionals and the same 6‐311++G(2d,p) basis set. The eight
simulated spectra are divided into two set of opposite shaped spectra (reddish lines and
12
13
14
15
16
HF
LSDA
Becke88
LYP
In Gaussian 09 the BH&HLYP functional has the form: 0.5*E
Y. Zhao and D.G. Truhlar, Theor. Chem. Acc. 2008, 120, 215.
X
+ 0.5*E
X
+ 0.5*ΔE
X
+ E
C
J‐D. Chai and M. Head‐Gordon. Phys. Chem. Chem. Phys. 2008, 10, 6615.
T. Yanai; D. Tew and N. Handy. Chem. Phys. Lett. 2004, 393, 51.
G. Cera, M. Chiarucci, A. Mazzanti, M. Mancinelli and M. Bandini. Org. Lett. 2012, 14, 1350‐1353. A. Mazzanti, T. Calbet,
M. Font‐Bardia, A. Moyano and R. Rios. Org. Biomol. Chem. 2012,10,1645‐1652. S. Duce, F. Pesciaioli, L. Gramigna, L.
Bernardi, A. Mazzanti and G. Bencivenni. Adv. Synth. Cat. 2011, 353, 860.
17
P.J. Stephens, D.M. McCann, F.J. Devlin, J.R. Cheeseman and M.J Frisch J.Am.Chem.Soc. 2004, 126, 7514.
S10

Electronic Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2012
bluish/greenish lines), each one corresponding to one of the two conformations of 10aa. The opposite
18
pattern can be attributed to the excitonic coupling to the two chromophores, i.e. the naphtyl and
the indane ring, that have opposite helicity in the two conformations (R*M* in conformation a and
R*P* in conformation b). Therefore the shape of the simulated spectrum that should be compared
with the experimental one strongly depends on the population ratio employed. Unfortunately, the
calculated energies of the two conformations are very similar (∆G° =0.12 kcal/mol), thus a correct
19
evaluation of the populations is completely unreliable. This imply that the AC assignment is
unfeasible without further experimental data.
1
Figure S4: variable temperature spectra ( H at 400 MHz in DMSO‐d
6
) of compound 10aa. The signal of
the CH of the indanone ring is showed. The asterisk marks a solvent impurity.
However, the conformational analysis of 10aa showed that the indane ring is forced to be
perpendicular to the naphtol ring by the steric hindrance exerted by the OH and the peri hydrogen (H‐
18
a) N. Harada and K. Nakanishi, Acc. Chem. Res. 1972, 5, 257; (b) N. Harada and K. Nakanishi, Circular Dichroic
Spectroscopy: Exciton Coupling in Organic Stereochemistry; University Science Books: Mill Valley, CA, 1983; (C) K. Nakanishi
andN. Berova, In Circular Dichroism: Principle and Applications; Berova, N.; Nakanishi, K.; Woody, R.W. Eds; VCH: New
York, 2000; Chapter 12, p 337.
19
G. Mazzeo, E. Giorgio, R. Zanasi, N. Berova andC. Rosini J. Org. Chem. 2010, 75, 4600.
S11

Electronic Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2012
8
) If the interconversion barrier is sufficiently high, two observable conformational diastereoisomers
1
could be generated. This is what actually happened. H NMR showed two sets of signals in a 69:31
ratio due to the frozen rotation around the C ‐CH bond. When the temperature is raised the two
α
multiplets broadens and show a single peak at +120°C, when the rotation is fast in the NMR time
scale, (Figure S4). An energy barrier of 17.8 kcal/mol was derived at the coalescence temperature
(
+90°C).
NOE spectra where then acquired to determine which conformational diastereoisomer is the more
populated. In the a diastereoisomer the CH hydrogen of indane is close to the peri H‐8 hydrogen of
the naphthol ring (1.87 Å), whereas in the b conformation the CH points towards the OH and it is far
20
from H‐8. DPFGSE‐NOE spectra obtained on saturation of the two CH signals showed that the major
conformation has the CH close to H‐8 and the minor has the CH close to OH (Figure S5: DPFGSE‐NOE
spectra of 10aa (600 MHz in DMSO‐d6). Bottom: control spectrum. Middle trace. NOE obtained on
saturation of the major CH. Top: NOE spectrum obtained on saturation of the minor CH. (in both NOEs
the small inverted peak of the second diastereoisomer was due to saturation transfer effects)..
6
Figure S5: DPFGSE‐NOE spectra of 10aa (600 MHz in DMSO‐d ). Bottom: control spectrum. Middle
trace. NOE obtained on saturation of the major CH. Top: NOE spectrum obtained on saturation of the
20
(a) K. Stott, J. Stonehouse, J. Keeler, T. L. Hwand and A. J. Shaka, J. Am. Chem. Soc. 1995, 117, 4199; (b) K. Stott, J. Keeler,
Q. N. Van and A. J. Shaka, J. Magn. Resonance 1997, 125, 302. See also: T. D. W. Claridge, High Resolution NMR Techniques
in Organic Chemistry; Pergamon: Amsterdam, 1999.
S12

Electronic Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2012
minor CH. (in both NOEs the small inverted peak of the second diastereoisomer was due to saturation
transfer effects).
In the present case the integration of the NMR spectra and the NOE spectra provide the exact ratio
of the two conformational diastereoisomers to be used in the simulation of the experimental ECD
spectrum. The simulated ECD spectrum was thus obtained for each model of calculation by taking into
account the 69:31 populations ratio experimentally determined by NMR. All the four simulations
obtained (Figure S6. Simulations of the experimental ECD spectrum of 10aa (black traces) obtained
with different methods of calculation (functionals). Each simulated spectrum (red, blue, green and
purple lines) was obtained starting from the spectra obtained for the two conformations weighted
using the experimental NMR data. The experimental spectrum of 10aa was obtained in acetonitrile
solution (1 10‐4 M, 0.2 cm path length).  are expressed in Mol L‐1 cm‐1. The simulated spectra were
vertically scaled to match the experimental intensity, and red shifted to match the experimental peak
at 238 nm.) display now a good agreement with the experimental spectrum and the best simulation
was obtained with the BH&HLYP functional. Provided the experimental ratio of the two
21
conformations, the ECD simulations showed to be able to tackle the absolute configuration of 10aa.
21
D. Casarini, L. Lunazzi, M. Mancinelli, A. Mazzanti and P. Scafato Chirality 2009, 21, 16.
S13

Electronic Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2012
Figure S6. Simulations of the experimental ECD spectrum of 10aa (black traces) obtained with
different methods of calculation (functionals). Each simulated spectrum (red, blue, green and purple
lines) was obtained starting from the spectra obtained for the two conformations weighted using the
experimental NMR data. The experimental spectrum of 10aa was obtained in acetonitrile solution (1
‐
4
‐1
‐1
1
0 M, 0.2 cm path length).  are expressed in Mol L cm . The simulated spectra were vertically
scaled to match the experimental intensity, and red shifted to match the experimental peak at 238
nm.
Compounds 11
No compound of the 11 series could be crystallized as single crystals and the assignment of
1
1bb was performed by means of same chirooptical methods used for 10aa.
As Before, also in the case of 11bb the two conformations found by conformational analysis
and subsequent DFT optimizations have the CH pointing towards the OH (a conformation) and the CH
in position 3 (b conformation), respectively. In this case, however, the steric hindrance to the rotation
1
of the indane ring is smaller and the H NMR spectrum recorded at room temperature does not show
the signals corresponding to the two conformational diastereoisomers. However, since calculations
estimated a very similar energy they should be both populated. For this reason, variable temperature
NMR spectra were recorded down to ‐100°C. Below ‐60°C the signal of the CH broadened and split at ‐
1
00°C into two signals with a 90:10 ratio. Following the trend observed for 10aa, the lower field signal
S14

Electronic Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2012
(
90%) can be attributed to the conformation in which the CH hydrogen is close to OH. Comfortably,
this is also the lowest energy conformation, and also the evaluation of the energy difference matched
#
well the experimental ratio (calculated ∆G : 0.56 kcal/mol; experimental: 0.75 kcal/mol at ‐100°C). By
applying Boltzmann statistics, the ratio at room temperature is 78:22.
Figure S7: 3D view of the two stable conformations of 11bb, calculated at the B3LYP/6‐31G(d) level.
Energy differences are in kcal/mol and represent ZPE–corrected free energies in standard conditions.
The rotational strengths and electronic excitation energies were calculated in the gas phase for the
two conformations of 11bb using TD‐DFT and the same four different methods used for 10aa. All the
calculations employed the 6‐311++G(2d,p) basis set and supposing S Absolute Configuration (Figure
S8. TD‐DFT simulations for the two conformations of 11bb, obtained with the four functionals and the
6
‐311++G(2d,p) basis set.).
As for 10aa, the opposite helicity generated by the out‐of‐plane disposition of the indane ring
yields two set of calculated ECD spectra with opposite pattern. However, when the final simulated
spectra are obtained using the population ratio determined by low‐temperature NMR, the agreement
with the experimental trace is very good (Figure S9), and the S absolute configuration can be assigned
to compound 11bb. In addition to this, the agreement of the four methods and the similarity of the
spectra enhances the reliability of the assignment.
S15

Electronic Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2012
Figure S8. TD‐DFT simulations for the two conformations of 11bb, obtained with the four functionals
and the 6‐311++G(2d,p) basis set.
Figure S9. Simulations of the experimental ECD spectrum of 11bb (black traces) obtained with
different methods of calculation. Each simulated spectrum (red, blue, green and purple lines) was
obtained starting from the spectra obtained for the four conformations weighted using Boltzmann
‐
4
statistics at +25°C. The experimental spectrum was obtained in acetonitrile solution (1 10 M, 0.2 cm
‐1
‐1
path length).  are expressed in Mol L cm . The simulated spectra were vertically scaled to match
the experimental intensity, and red shifted to match the experimental maxima.
S16

Electronic Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2012
Compounds 5
As reported in the text, compounds 5 correspond to the hemiacetalic form. From a conformational
point of view, these structures are completely blocked, and only one conformation is supposed to be
populated. This correspond to the chair conformation of the cyclohexane part of the bicyclic system.
Compound 5da, containing a phenyl group in position 4 was selected for the spectroscopic analysis
because the presence of a second chromophore that could couple with the naphthalene ring to cause
a stronger ECD spectrum.
1
The analysis of the H spectrum of compound 5da showed that the phenyl ring in position 4 occupies
the equatorial position. The signal of benzylic CH shows a trans‐diaxial coupling constant of 12.8 Hz
and an axial‐equatorial constant of 4.6 Hz. The 2R*,4S*,6S* relative configuration can be therefore
assumed in the following discussions (Figure S10. Left: 3D view of the optimized conformation of 5da,
calculated at the B3LYP/6‐31G(d) level.). It is worth to note that the second diastereoisomer produced
by the reaction (6da) does not evolve to the hemiacetalic form because in this compound the phenyl
ring should occupy the axial position. Although the absolute configuration of 6da could not be
established by anomalous scattering, the X‐ray structure confirmed that the two chiral carbons have
opposite chirality (thus 3R*,5S*).
Figure S10. Left: 3D view of the optimized conformation of 5da, calculated at the B3LYP/6‐31G(d)
1
level. The chair‐shaped cyclohexane ring is shaped. Center: H multiplet of the CH in position 4
showing the trans‐diaxial and axial‐equatorial coupling constants. Right: X‐ray structure of 6da.
To exclude the presence of other low‐energy conformations, the conformational search by MM
methods was performed as above, but no other conformations were found. As in the previous cases,
DFT optimization at the B3LYP/6‐31G(d) level provided the final geometry to be used in the TD‐DFT
simulations. The rotational strengths and electronic excitation energies were calculated in the gas
phase by TD‐DFT and the same four different methods used for 10aa and 11bb. All the calculations
employed the 6‐311++G(2d,p) basis set and supposed S Absolute Configuration at the reaction center.
(thus 2R, 4S, 6S) (Figure S11). In this case the simulation of the experimental spectrum is
straightforward, since no conformational averaging is needed and the four calculated spectra are very
S17

Electronic Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2012
similar. As shown in Figure S11, the experimental trace is well matched by all the simulation obtained
for the 2R, 4S, 6S absolute configuration.
Figure S11. TD‐DFT simulations for 5da, obtained with the four different functionals and the 6‐
3
11++G(2d,p) basis set. The black line corresponds to the experimental spectrum obtained in
‐
4
‐1
‐1
acetonitrile solution (1 10 M, 0.05 cm path length).  are expressed in Mol L cm . The simulated
spectra were vertically scaled to match the experimental intensity, and red shifted to match the
experimental maxima.
Compounds 3
Compounds 3 has the same hemiacetalic scaffold of compounds 5, thus a second chiral carbon is
generated with complete stereocontrol during the cyclization. With respect to the cyclohexane
scaffold the two substituents must occupy the 1‐3 diaxial positions (Figure S12). Compound 3ab was
selected for the stereochemical assignment.
As in the case of 5, the bicyclic structure is very rigid, and only a single conformation was found by
MM search. In this case, however, a second conformation with a boat‐shaped cyclohexane could not
be excluded in principle. When optimized by DFT, this conformation showed to be about 5 kcal/mol
higher in energy with respect to the conformation found by conformational search, and it can be
neglected in the ECD simulations. Compound 3ab lacks a second chromophore on the cyclohexane
ring and the experimental ECD is rather weak. However, the TD‐DFT simulations performed assuming
the 1S, 5R absolute configuration satisfactorily followed the experimental trend (Figure S12). It should
be pointed out that the same S configuration at the reaction center was assigned as in the three
previous cases.
S18

Electronic Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2012
Figure S12 3D view of the optimized conformation of 3ab, calculated at the B3LYP/6‐31G(d) level.
Figure S13. TD‐DFT simulations for 3ab, obtained with the four different functionals and the same 6‐
11++G(2d,p) basis set. The black line corresponds to the experimental spectrum obtained in
3
‐
5
‐1
‐1
acetonitrile solution (5 10 M, 0.2 cm path length).  are expressed in Mol L cm . The simulated
spectra were vertically scaled to match the experimental intensity, and red shifted to match the
experimental maxima
S19

Electronic Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2012
Crystal data for 10de
Molecular formula: C20
H
15ClO
4
, MW 338.77. Monoclinic, space group P2
1
, a = 8.6325(13), b =
3
–3
8
.6693(13), c = 22.017(3), β = 92.146(2). V = 1646.6(4) Å , T = 298(2) °K, Z = 4, 
c
= 1.367 g cm ,
–
1
F(000) = 2762, graphite-monochromated MoK radiation ( = 0.71073 Å), (MoK) = 0.247 mm ,
colorless sticks (0.40 × 0.15 × 0.10 mm ), empirical absorption correction with SADABS (transmission
3
factors: 0.9078 – 0.9758), 2400 frames, exposure time 20 s, 1.85 ≤  ≤ 27.40, –11≤ h ≤ 11, –11 ≤ k ≤
1
1, –28 ≤ l ≤ 28, 7373 reflections collected, 5444 independent (Rint = 0.0277), solution by direct
2
methods (SHELXS97) and subsequent Fourier syntheses, full-matrix least-squares on F
o
(SHELX97),
hydrogen atoms refined with a riding model except for the hydroxyl hydrogen that was experimentally
2
2
located; data / restraints / parameters = 7373/ 1 / 441, S(F ) = 1.044, R(F) = 0.0634 and wR(F ) =
2
0
.1258 on all data, R(F) = 0.0454 and wR(F ) = 0.1124 for 5444 reflections with F
0
> 4( F
)/3, largest difference
0
),
2
2
2
2
o
2
c
weighting scheme w = 1/[ (F
o
) + (0.0646P) + 0.0000P] where P = (F
+ 2F
–
3
peak and hole 0.186 and –0.293 e Å . Flack parameter: 0.02(6). The unit cell contains two different
conformation belonging to the same chirality, that are different because of the different disposition of
the OMe group. CCDC-893970 CIF file contains the supplementary crystallographic data for this
paper. These data can be obtained free of charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
S20

Electronic Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2012
Crystal data for 6da
2
Molecular formula: C22H20ClO , MW 316.38. Monoclinic, space group C2, a = 19.582(3), b =
3
–
6
c
.4957(10), c = 15.898(2), β = 123.3630(10). V = 1688.9(4) Å , T = 298(2) °K, Z = 4,  = 1.244 g cm
3
–1
,
F(000) = 672, graphite-monochromated MoK radiation ( = 0.71073 Å), (MoK) = 0.078 mm ,
colorless plates (0.40 × 0.40 × 0.20 mm ), empirical absorption correction with SADABS (transmission
3
factors: 0.9845 – 0.9694), 2400 frames, exposure time 15 s, 1.53 ≤  ≤ 27.49, –25≤ h ≤ 25, –8 ≤ k ≤ 8,
–
20 ≤ l ≤ 20, 9738 reflections collected, 3846 independent (Rint = 0.0201), solution by direct methods
2
(SHELXS97) and subsequent Fourier syntheses, full-matrix least-squares on F
o
(SHELX97), hydrogen
atoms refined with a riding model except for the hydroxyl hydrogen that was experimentally located;
2
2
data / restraints / parameters = 3846/ 1 / 221, S(F ) = 1.025, R(F) = 0.0440 and wR(F ) = 0.0879 on all
2
data, R(F) = 0.0360 and wR(F ) = 0.827 for 3296 reflections with F
0
> 4( F
0
), weighting scheme w =
2
2
2
2
2
c
1
/[ (F
o
) + (0.0379P) + 0.3168P] where P = (F
o
+ 2F
)/3, largest difference peak and hole 0.127
–
3
and –0.125 e Å . Flack parameter: -0.6(11). CCDC-894320 CIF file contains the supplementary
crystallographic data for this paper. These data can be obtained free of charge from The Cambridge
Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
S21

Electronic Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2012
Relative Configuration assignment of compounds ent‐5da and ent‐6da
The relative stereochemistry of compounds ent‐5da and ent‐6da was established by means of NOE
experiments. It was found that in ent‐5da the phenyl group is trans to the naphthalene moiety, thus
adopting an equatorial position and that in compound ent‐6da the phenyl group is cis to the
naphthalene moiety adopting again an equatorial position. This may also explain why structure ent‐
6
da, which has the cis relative disposition, is only obtained as the “open form”. In fact, in the
corresponding “closed form” the cis‐phenyl group would give rise to a strained all‐axial structure
where the two aromatic moieties will be in close contact.
H
H
H
H
H
a
Hb
H
O
H
O
H
H
H
H
OH
H
H
H
H
b
H
O
H
a
O
H
OH
H
ent-6da,
"
closed form"
ent-5da
ent-6da
(
not observed)
Compound ent‐5da
Irradiation of axial proton H
a
generated a visible NOE effect on the two equatorial protons of the
vicinal carbons of the cyclohexane ring and on the ortho‐protons of the phenyl group.
3
Figure S14. DPFGSE‐NOE spectra obtained for ent‐5da (25°C, 600 MHz, CDCl ) on saturation of proton
H . Full spectrum.
a
S22

Electronic Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2012
Figure S15. DPFGSE‐NOE spectra obtained for ent‐5da (25°C, 600 MHz, CDCl ) on saturation of proton
3
a
H . Detail of the aliphatic region.
3
Figure S16. DPFGSE‐NOE spectra obtained for ent‐5da (25°C, 600 MHz, CDCl ) on saturation of proton
H . Detail of the aromatic region.
a
S23

Electronic Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2012
Irradiation of equatorial proton H produced a NOE effect on all four protons of the two adjacent
b
methylene groups on the cyclohexane ring and on the C3H of the naphthalene ring.
3
Figure S17. DPFGSE‐NOE spectra obtained for ent‐5da (25°C, 600 MHz, CDCl ) on saturation of proton
H .
b
3
Figure S18. DPFGSE‐NOE spectra obtained for ent‐5da (25°C, 600 MHz, CDCl ) on saturation of proton
H . Detail of the aliphatic region.
b
S24

Electronic Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2012
Figure S19. DPFGSE‐NOE spectra obtained for ent‐5da (25°C, 600 MHz, CDCl
3
) on saturation of proton
b
H . Detail of the aromatic region.
S25

Electronic Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2012
Compound ent‐6da
Irradiation of axial proton H
a
at 3.13 ppm generated a strong NOE effect on H
b
, thus confirming the
1
,3‐cis‐diaxial relative disposition of these two protons, and additional NOE effects on the equatorial
protons only of the vicinal methylene groups of the cyclohexane ring and on the ortho‐protons of the
phenyl group.
Figure S20. DPFGSE‐NOE spectra obtained for ent‐6da (25°C, 600 MHz, CDCl
. Full spectrum.
3
) on saturation of proton
H
a
3
Figure S21. DPFGSE‐NOE spectra obtained for ent‐6da (25°C, 600 MHz, CDCl ) on saturation of proton
H . Detail of the aliphatic region.
a
S26

Electronic Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2012
Figure S22. DPFGSE‐NOE spectra obtained for ent‐6da (25°C, 600 MHz, CDCl ) on saturation of proton
3
H
a
. Detail of the aromatic region.
Irradiation of the axial proton H
b
at 3.66 ppm produced visible NOE effect on proton H , on the
a
equatorial protons of the vicinal methylene groups of the cyclohexane ring, and on OH and C3H
protons of the naphthalene moiety.
3
Figure S23. DPFGSE‐NOE spectra obtained for ent‐6da (25°C, 600 MHz, CDCl ) on saturation of proton
H . Full spectrum.
b
S27

Electronic Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2012
3
Figure S24. DPFGSE‐NOE spectra obtained for ent‐6da (25°C, 600 MHz, CDCl ) on saturation of proton
b
H . Detail of the aliphatic region.
Figure S25. DPFGSE‐NOE spectra obtained for ent‐6da (25°C, 600 MHz, CDCl
. Detail of the aromatic region.
3
) on saturation of proton
H
b
S28

Electronic Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2012
Relative Configuration assignment of compound ent‐5ca
The relative stereochemistry of compound ent‐5ca was established by means of
NOE experiments. Irradiation of axial proton H generated a strong NOE effect on
geminal proton H as well as a visible NOE on H , on one of the two protons on
C3 and on the methylene of the ethyl group. As a control, irradiation on
equatorial proton H in addition to the strong NOE effect on geminal proton H
produced only an additional visible NOE effect on H . Neither H nor H show
a
b
c
b
a
,
c
a
b
visible NOE effects on C5H. According to these experiments the C5 ethyl group
should be in an axial position, trans with respect to the C6‐Ar bond.
Figure S26. DPFGSE‐NOE spectra obtained for ent‐5ca (25°C, 600 MHz, CDCl
. Detail of the aliphatic region.
3
) on saturation of proton
H
a
S29

Electronic Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2012
Figure S27. DPFGSE‐NOE spectra obtained for ent‐5ca (25°C, 600 MHz, CDCl ) on saturation of proton
3
b
H . Detail of the aliphatic region.
S30