Stereodynamics and edge-to-face CH-π aromatic interactions in o-phenethyl-substituted biaryls

Article abstract of DOI:10.1021/jo0520902

As part of a search for systems that exhibit intramolecular aromatic edge-to-face interactions, a series of four biaryl compounds containing a phenethyl side chain were prepared. These compounds exist as a slowly interconverting mixture of two atropisomer

Full text of DOI:10.1021/jo0520902

Stereodynamics and Edge-to-Face CH-π Aromatic Interactions in
o-Phenethyl-Substituted Biaryls
W. Brian Jennings,* Brid M. Farrell, and John F. Malone*,‡
,
†
†,§
Department of Chemistry and Analytical & Biological Chemistry Research Facility, UniVersity College
Cork, Cork, Ireland, and School of Chemistry and Chemical Engineering, The Queen’s UniVersity of
Belfast, Belfast BT9 5AG, Northern Ireland
brianj@ucc.ie
ReceiVed October 5, 2005
As part of a search for systems that exhibit intramolecular aromatic edge-to-face interactions, a series of
four biaryl compounds containing a phenethyl side chain were prepared. These compounds exist as a
slowly interconverting mixture of two atropisomers due to steric hindrance to rotation about the biaryl
bond. The more thermodynamically stable isomer exhibits an abnormal shielding of an ortho-proton in
solution indicative of an edge-to-face CH-π interaction with the terminal phenyl ring on the side chain.
This tilted-T type of geometry was observed in the X-ray crystal structure of one of the compounds. The
edge-to-face conformation in solution is estimated by variable temperature NMR studies to be energetically
-
1
-1
-1
favored by ca. 1.6 kcal mol but entropically disfavored by ca. 5.0 cal K mol .
Introduction
into three categories: T-shaped, tilted-T (subdivided into face
8
and edge tilted), and offset parallel. All of the above arrange-
Edge-to-face orientations between aromatic rings were ini-
tially observed almost 50 years ago in the herringbone structure
of crystalline benzene. However the more general significance
of these relatively weak edge-to-face CH-π aromatic interactions
did not become evident until 1985 when Burley and Petsko,
and Singh and Thornton presented evidence that they played a
significant role in protein structure. Edge-to-face aromatic
interactions can also be important in such diverse areas as crystal
packing and crystal engineering, supramolecular assemblies
-1
ments are believed to be attractive by ca. 1-2 kcal mol , but
more examples are needed in order to provide additional
information.
1
Following previous observations of edge-to-face aromatic
9
interactions in imino systems, we now report the occurrence
of these interesting interactions in twisted biphenyl systems
containing an ortho R-hydroxyphenethyl substituent.
2
3,4
Results and Discussion
5
6
and host-guest binding, drug receptor interactions, and other
areas of molecular recognition. Examples have emerged where
7
Synthesis and Stereodynamics. Synthesis of the hindered
target compounds 3-6 was achieved in just four steps from
intramolecular edge-to-face interactions have a considerable role
in determining the conformation of flexible organic molecules
in solution and in the crystalline state. Molecular orbital
calculations and evidence from some crystal structures indicate
that the geometry of edge-to-face interactions can be classified
(4) Some recent examples: (a) Brayshaw, P. A.; Hall, A. K.; Harrison,
W. T. A.; Harrowfield, J. M.; Pearce, D.; Shand, T. M.; Skelton, B. W.;
Whitaker, C. R.; White, A H. Eur. J. Inorg. Chem. 2005, 6, 1127-1141.
(
2
b) Oh, M.; Stern, C. L.; Mirkin, C. A. Inorg. Chem. 2005, 44, 2647-
653. (c) Abboud, K.; Harrowfield, J. M.; James, B. D.; Skelton, B. W.;
†
‡
§
White, A. H. Inorg. Chim. Acta 2005, 358, 1293-1297. (d) Soudi, A. A.;
Marandi, F.; Ramazani, A.; Ahmadi, E.; Morsali, A. C. R. Chim. 2005, 8,
157-168. (e) Reger, D. L.; Semeniuc, R. F.; Smith, M. D. Cryst. Growth
Des. 2005, 5, 1181-1190. (f) Harrowfield, J.; Matt, D. J. Inclusion Phenom.
Macrocyclic Chem. 2004, 50, 133-150. (g) Emseis, P.; Hibbs, D. E.;
Leverett, P.; Reddy, N.; Williams, P. A. Inorg. Chim. Acta 2004, 357, 3251-
3263. (h) Xu, W.; Lu, Y. X.; Liu, C. M.; Guo, P.; Lan, B. J.; Zhoua, H.
Acta Crystallogr., Sect. E: Struct. Rep. Online 2004, 60, 1049-1050. (i)
Yang, J. S.; Liu, C. P.; Lin, B. C.; Tu, C. W.; Lee, G. H. J. Org. Chem.
2002, 67, 7343-7354.
University College Cork.
The Queen’s University of Belfast.
Present address: GlaxoSmithKline Manufacturing Ltd., Currabinny, Car-
rigaline, Co. Cork, Ireland.
1) Cox, E. G.; Cruickshank, D. W. J.; Smith, J. A. S. Proc. R. Soc.
London 1958, 247, 1-21.
(
(
2) (a) Burley, S. K.; Petsko, G. A. Science 1985, 229, 23-28. (b) Singh,
J.; Thornton, J. M. FEBS Lett. 1985, 191, 1-6. (c) Burley, S.K.; Petsko,
G. A. AdV. Protein Chem. 1988, 39, 125-189.
(
3) For a review see: Nishio, M. CrystEngComm 2004, 6, 130-158.
1
0.1021/jo0520902 CCC: $33.50 © 2006 American Chemical Society
Published on Web 02/21/2006
J. Org. Chem. 2006, 71, 2277-2282
2277

Jennings et al.
CHART 1. Structures and Relative Stereochemistry
SCHEME 1. Synthetic Route
commercially available 2-bromonaphthalene or 2-bromotoluene
(Scheme 1). These aryl bromides were converted to the
corresponding boronic acids via the Grignard reagents. Aryl
coupling to 2-bromoacetophenone was achieved in good yield
(80%) via a Suzuki coupling reaction to afford the biaryl ketones
1
and 2. Addition of benzyl- or 2-methylbenzyl-magnesium
bromide in ether gave the target biaryl carbinols 3-6 in good
yield (60-90%) as a mixture of two diastereomers as shown
by NMR analysis.
TABLE 1. Atropisomer Ratios under Kinetic and Thermodynamic
Controla
kinetic ratio
thermodynamic ratio
The diastereomers are atropisomers arising from slow rotation
on the NMR time-scale around the biaryl bond. This generates
a chiral axis with M or P configuration, which in combination
with the stereogenic carbinol center gives two racemic diaster-
eomers RM/SP and RP/SM (Chart 1). In the case of the naphthyl
compounds 3 and 4 the diastereomers were separated by flash
column chromatography on silica gel. The diastereomers of the
o-tolyl compounds 5 and 6 failed to separate completely, but
pure samples were obtained by recrystallizing partially resolved
chromatography fractions.
compd
a:b
a:b
b
3
4
5
6
25:75
28:72
27:73
33:67
68:32
60:40
64:36
62:38
c
d
d
a
1
b
Measured by integration of H NMR signals. Measured at 83 °C in
1,1,2,2-tetrachlorethane-d2 after equilibration. c Measured at 110 °C in
,1,2,2-tetrachlorethane-d2 after equilibration. d Measured at 61 °C in
1
chloroform-d after equilibration.
TABLE 2. Kinetic Data for Atropisomerization
In each of the compounds the more quickly eluted isomer
a
kbfa a
∆G afb b
q
q
b
(
3
3a-6a) was the minor component of the reaction mixture (25-
T
kafb
∆G bfa
compd (°C)
(s 1)
-
(s
-1
)
(kcal mol )
-1
(kcal mol-1)
3%). The isomer distribution was found to be under kinetic
c
-6
-5
-6
-5
control as the ratio was observed to change markedly on heating
the samples to 90-120 °C in toluene-d8 or 1,1,2,2,-tetrachlo-
roethane-d2. The initial (kinetic) and final equilibrium (ther-
modynamic) isomer ratios are given in Table 1 as determined
3b
5
83
61
2.25 × 10
4.75 × 10
30.2
27.2
29.7
26.9
d
a
1.07 × 10
1.93 × 10
a
Estimated precision ( 10%. b Estimated precision ( 0.1 kcal mol-1.
c
Solvent 1,1,2,2-tetrachloroethane-d2. d Solvent chloroform-d.
1
by integration of the respective signals in the H NMR spectra.
1
vary much with structure on changing the Ar group from
-naphthyl (3 and 4) to o-tolyl (5 and 6) or Ar from phenyl (3
and 5) to o-tolyl (4 and 6). Raising the temperature facilitates
interconversion of the isomers by rotation about the hindered
central biaryl bond.
Interestingly, the more quickly eluted minor diastereomer is
actually the thermodynamically preferred isomer in each case
accounting for 60-68% of the equilibrium mixture (Table 1).
Neither the initial kinetic nor the equilibrium isomer distributions
2
1
(
5) (a) Reger, D. L.; Watson R. P.; Smith, M. D.; Pellechia, P. J.
Compounds 3b and 5a were selected for a kinetic study of
the atropisomerization. The process was followed by H NMR
Organometallics 2005, 24, 1544-1555. (b) Lukyanenko, N. G.; Kirichenko,
T. I.; Lyapunov, A. Y.; Mazepa, A. V.; Simonov, Y. A.; Fonari, M. S.;
Botoshansky, M. M. Chem.- Eur. J. 2004, 11, 262-270. (c) Leonard, M.
S.; Carroll, P. J.; Joullie, M. M. J. Org. Chem. 2004, 69, 2526-2531. (d)
Balog, M.; Grosu, I.; Ple, G.; Ramondenc, Y.; Condamine, E.; Varga, R.
A. J. Org. Chem. 2004, 69, 1337-1345.
1
using the change in relative intensity of the methyl signals. The
kinetics followed the standard pattern for a reversible first-order
approach to equilibrium and the results are given in Table 2.
-1
The free energy of activation is about 3 kcal mol lower for
(6) (a) Sukalovic, V.; Andric, D.; Roglic, G.; Kostic-Rajacic, S.;
compound 5. The interconversion involves passage of the Ar1
ring through the plane of the directly attached ortho-substituted
phenyl ring. The least congested transition state will involve
the 1-naphthyl group in 3, or the o-tolyl group in 5, passing the
ortho-hydrogen atom of the second ring. Clearly it is sterically
easier for an o-tolyl group to adopt the coplanar biaryl geometry
than it is for a 1-naphthyl group. Literature reports on other
systems also indicate that rotational barriers involving a
Schrattenholz, A.; Soskic, V. Eur. J. Med. Chem. 2005, 40, 481-493. (b)
Sukalovic, V.; Zlatovic, M.; Andric, D.; Roglic, G.; Kostic-Rajacic, S.;
Soskic, V. Arzneim.-Forsch. 2005, 55, 145-152. (c) Fattori, D. Drug
DiscoVery Today 2004, 9. 229-238.
(
7) Shu, L. H.; Shi, Y. Tetrahedron Lett. 2004, 45, 8115-8117.
(8) For a review, see: Jennings, W. B.; Farrell, B. M.; Malone, J. F.
Acc. Chem. Res. 2001, 34, 885-894.
(9) (a) Boyd, D. R.; Evans, T. A.; Jennings, W. B.; Malone, J. F.;
O’Sullivan, W. O.; Smith, A. Chem. Commun. 1996, 2269-2270. (b)
Hamor, T. A.; Jennings, W. B.; Proctor, L. D.; Tolley, M. S.; Boyd, D. R.;
Mullan, T. J. Chem. Soc., Perkins Trans. 2 1990, 25-30.
-1
1-naphthyl group are 2-3 kcal mol larger than those involving
2278 J. Org. Chem., Vol. 71, No. 6, 2006

Aromatic Interactions in o-Phenethyl-Substituted Biaryls
FIGURE 1. X-ray structure of 3a showing (a) atom numbering and (b) edge-to-face interaction.
FIGURE 2. Two views of the X-ray structure of 4a, with atom numbering.
an o-tolyl group.10 In each case the thermodynamically preferred
atropisomer a showed in the H NMR spectrum an aromatic
doublet signal at abnormally high field: δ 6.19 (3a), 6.44 (4a),
component naphthyl compounds, 3a and 4a, which provided
1
suitable crystals, were subjected to single-crystal X-ray analysis.
In both compounds all hydrogen atoms were located from
difference Fourier syntheses but were included in the final
calculations at positions calculated from the geometries of the
molecules. For the hydrogen atoms being examined for possible
interaction with a second aromatic ring in an edge-to-face
contact, the hydrogen was positioned to give a C-H bond length
of 1.083 Å, the standard value obtained from neutron diffraction
determination of the position of the hydrogen nucleus, and used
5
.90 (5a), and 6.07 (6a). The kinetic atropisomers 3b-6b did
not show this upfield aromatic signal. The upfield doublet signal
of the a isomers integrated for one proton (relative to the
corresponding methyl signals) and showed evidence of ad-
ditional meta coupling. This signal is assigned to the proton
1
H2 on the Ar ring ortho to the biphenyl bond. This assignment
was confirmed by 500 MHz HMBC, HMQC, COSY and
TOCSY 2D NMR spectra of 3a, 4a, 5a, and 6a. The upfield
shift of aromatic protons in rings devoid of strong electron
donating substituents is characteristic of a proton experiencing
ring current shielding due to an edge-to-face interaction with
the face of a neighboring aromatic ring. The only feasible source
8
in earlier studies as a normalized distance.
Both 3a and 4a have configuration RM/SP (shown as RM in
Figures 1 and 2). Since all four a isomers show similar upfield
1
aromatic signals in their H NMR spectra and are thermody-
namically more stable, it can be reasonably assumed that 5a
and 6a also have the RM/SP configuration. The atom number-
ings for 3a and 4a, respectively, are shown in Figures 1 and
2
of the shielding in the present compounds is the terminal Ar
ring on the side chain.
X-ray Crystal Structures. To investigate the possibility of
an edge-to-face interaction in the solid state, the two minor
2
.
The conformation adopted by 3a in the crystalline state is
8
classified as a face-tilted T-geometry. Thus the ortho hydrogen,
H2, of the naphthyl group is located above the plane of the
(
10) (a) Boyd, D. R.; Al-Showiman, S.; Jennings, W. B. J. Org. Chem.
978, 43, 3335-3339. (b) Jennings, W. B.; Kochanewycz, M. J.; Lunazzi,
L. J. Chem. Soc., Perkin Trans. 2 1997, 2271-2273.
1
2
terminal Ar phenyl ring (C19-C24; Figure 1b). The H to ring-
J. Org. Chem, Vol. 71, No. 6, 2006 2279

Jennings et al.
TABLE 3. Torsion Angles (deg)
and 6a there must be a considerable population of a sice chain
conformation that places the Ar phenyl or tolyl ring in a face-
2
C10-C1-
C11-C12
C11-C12-
C17-C18
C12-C17-
C18-C19
C17-C18-
C19-C20
compd
to-edge orientation with this proton, as observed in the solid-
state structure of 3a. Although the shielding effect on proton
H2 in solution is sizable (ca.1.5 ppm based on a standard
aromatic signal position of ca. δ 7.5), it is less than the shielding
predicted for a proton fixed 2.59 Å above the face of the Ar2
ring (ca. 2.5 ppm using the Johnson-Bovey ring current
3
4
a
a
-102
-100
-114
+49
+67
+56
+87
+185
TABLE 4. Estimated Enthalpy (∆H°) and Entropy (∆S°) Changes
for the Proposed Closed (I) h Open (II) Conformational
Equilibrium
1
1
tables).
H°a
∆S°a
∆G°b
We propose that in solution the edge-to-face “closed”
conformation is in rapid equilibrium with at least one “open”
conformation that does not have a close association between
∆
temp range
(kcal
(cal
(kcal
-
mol-1 K-1
-1
compd
(°C)
δI
δII
mol 1)
)
mol )
2
1 8,9
3
5
a
a
-80 to 20
-50 to 20
4.8
5.0
7.7
7.5
1.5
1.7
4.9
5.0
0.05
0.24
the Ar ring and the ortho proton on Ar . Hence the actual
shift of this proton is a time average of the shifts in the open
and closed conformations weighted according to their relative
a
Estimated precision: ∆H°( 0.3 kcal mol ; ∆S°( 0.5 cal mol K-1
-1
-1
.
b
Measured at 20 °C, precision ( 0.2 kcal mol-1
9
.
populations. This leads to an observed signal position well
upfield of the normal aromatic value of ca. δ 7.5 but short of
the predicted position of ca. δ 5.0 for an edge-to-face interaction
at 2.59 Å. To test this postulate two representative compounds
centroid distance is 2.64 Å. The perpendicular distance from
H2 to the Ar ring is 2.59 Å, and the point of intersection of
2
3a and 5a were subjected to a variable-temperature NMR study.
this perpendicular to the plane is offset by 0.52 Å from the
ring centroid. These distances are among the shortest found in
The relative conformer populations in a dynamic situation should
change with temperature (unless they happen fortuitously to have
equal energy and the same entropy). The temperature effect on
conformational equilibria is greatest in situations where enthalpic
and entropic factors are in opposition, i.e., where the lower
energy conformation has the lowest (less favorable) entropy.
On lowering the temperature, the shielded ortho doublet signal
of the naphthyl compound 3a steadily moved further upfield
from δ 6.20 at 20 °C to δ 5.60 at -50 °C in CD2Cl2 solution.
To extract the thermodynamic factors determining the confor-
mational equilibrium it is necessary to have an estimate of the
shifts of this proton in both the open and closed (edge-to-face)
conformations. Initially these values were taken to be δII 7.5 as
being typical of an aromatic proton ortho to an imino group in
the absence of any ring current effect and δI 5.0 estimated from
the Johnson-Bovey ring current tables for a proton in the closed
8
previous face-tilted aromatic contacts and represent a substantial
edge-to-face interaction. Moreover the dihedral angle of 53°
between the interacting aromatic rings is typical of angles found
in earlier examples of other systems exhibiting an edge-to-face
interaction.8
Surprisingly, compound 4a (Figure 2) adopts a very different
conformation in the solid state. Figures 1a and 2b, respectively,
show molecules 3a and 4a from a similar perspective with
respect to the biaryl group. The differences in molecular
conformation can be quantified by comparing the torsion angles
along the four rotatable bonds between the naphthyl and terminal
phenyl groups (Table 3).
Relative to the conformation of 3a, compound 4a shows a
1
63° rotation about the C12-C17 bond and a 102° rotation
about C18-C19. Thus the naphthyl ortho-proton, H2, is
removed from the vicinity of the Ar ring (Figure 2b), and the
11
edge-to-face conformation using the X-ray data for 3a.
2
However, in the final computer-based analysis both of these
estimates were varied slightly to optimize the agreement between
calculated and observed chemical shift versus temperature curves
edge-to-face contact of H2 indicated by NMR in solution is
absent in the solid state. The closest approach of any aromatic
proton in 4a to a second ring is between the naphthyl peri proton
(Table 4). The resulting values of ∆H° and ∆S° were reasonably
(H9) and the terminal tolyl ring. The H9 to tolyl ring-centroid
robust (within 10%) to the precise values of δI and δII used in
the calculation.
It can be seen from the results in Table 4 that the closed
distance, 3.15 Å, is probably beyond an accepted limit for a
strong edge-to-face interaction. However, the perpendicular
distance, H9 to ring plane, 2.92 Å, with offset 1.18 Å and a
very shallow 18° interplanar angle is suggestive of a weak
distorted offset-stacked edge-to-face interaction. Possibly crystal
packing forces are responsible for the unexpected geometry of
compound 4a in the solid state. Alternatively 4a may have
crystallized in a conformation approximating to the “open”
conformer believed to be in rapid equilibrium with the face-
tilted T conformation in solution (see Solution Dynamics).
Indeed it can be estimated from the small ∆G° for this
equilibrium (Table 4) that there is ca. 40-50% of at least one
alternative conformation present in solution at room temperature,
offering an opportunity for the compound to crystallize in this
geometry.
edge-to-face conformation (I) has significantly lower enthalpy
-1
(
by ca. 1.6 kcal mol ) consistent with the supposed attractive
nature of this interaction. However, entropic factors strongly
favor the open conformation, presumably due to its greater
flexibility. A similar trend was previously observed for edge-
9
to-face interactions in imino compounds. It is possible that there
is more than one significantly populated open conformation.
This would not invalidate the model provided that, as expected,
the chemical shift (δII) of all open conformations lies in the
normal aromatic region (ca. δ 7.5). However significant
contributions from more than one open conformation would
further raise the entropy due to a contribution from entropy of
mixing (RlnN).
Solution Dynamics. In solution compounds 3-6 will have
rapid conformational mobility about all three single bonds in
the phenethyl side chain. Mobility about the biphenyl bond is
severely restricted at ambient temperature, leading to the
separable atropisomers a and b. To create the observed marked
(
11) The ring current shielding of a proton situated 2.59 Å vertically
above the face of a phenyl ring and offset from the center by 0.52 Å (as in
the X-ray structure of 3a) was estimated to be -2.5 ppm using the tabular
data given in Emsley, J. W.; Feeney, J.; Sutcliffe, L. H. High Resolution
Nuclear Magnetic Resonance Spectroscopy; Pergamon Press: Oxford, 1965;
Vol. 1, pp 595-604.
1
shielding of the ortho A ring proton in compounds 3a, 4a, 5a,
2280 J. Org. Chem., Vol. 71, No. 6, 2006

Aromatic Interactions in o-Phenethyl-Substituted Biaryls
Experimental Section
1-(2-Methylphenyl)-2-[2-(1-naphthyl)phenyl]propan-2-ol (4).
A solution of ketone 1 (2.0 g, 8.1 mmol) in dry diethyl ether (15
mL) was added dropwise under nitrogen to an ether solution of
1-[2-(1-Naphthyl)phenyl]ethanone (1). To a stirred solution of
palladium tetrakistriphenylphosphine (97 mg, 0.084 mmol) in
benzene (20 mL) under nitrogen were added 2-bromoacetophenone
2
(
-methylbenzylmagnesium bromide [from magnesium turnings
0.48 g, 20.3 mmol) and 2-methylbenzyl bromide (4.13 g, 22 mmol)
in dry ether (30 mL)]. The reaction mixture was heated at reflux
for 12 h and cooled, and a saturated solution of NH Cl was added.
The aqueous phase was extracted with ether. The organic layer was
washed with aqueous NaHCO and water, dried over magnesium
(0.42 g, 2.11 mmol), a 2 M NaOH solution (2.2 mL), and 1-naphthyl
boronic acid (0.4 g, 2.32 mmol). The reaction mixture was heated
at reflux for 12 h and then cooled to room temperature. Aqueous
sodium hydroxide (30% w/w) was then added, followed by the slow
addition of 30% w/w hydrogen peroxide. The reaction mixture was
stirred for a 1 h followed by the addition of ether. The layers were
separated, and the aqueous layer was further extracted with ether.
The combined ether extracts were dried over magnesium sulfate
and concentrated under reduced pressure to give the crude product.
4
3
sulfate, filtered, and concentrated under reduced pressure to give a
crude product containing a mixture of two alcohol atropisomers
1
4
a and 4b (ratio 1:2. from H NMR analysis). The crude product
was purified by flash chromatography on silica gel eluting with
ether-hexane to give alcohol 4a (0.77 g, 27%) and alcohol 4b (2.03
g, 71%) both as colorless solids. Compound 4a was recrystallized
Flash chromatography on silica gel eluting with ether-hexane (1:
1
9) afforded 1 as a colorless oil (0.41 g, 80%): H NMR (270 MHz;
1
from hexane-chloroform, mp 65-66 °C: H NMR (270 MHz;
CDCl
3
) δ 1.77 (3H, s, CH
) δ 29.6 (CH
CdO); REI-MS m/z (EI) 246 (M , 100%), 231 (92), 202 (96),
3
), 7.31-7.96 (m, ArH); 13C NMR (67.8
), 125.2-141.2 (aromatic signals), 203.0
CDCl
ArCH
3 3
) δ 1.50 (3H, s, CH ), 1.56 (1H, br s, OH), 1.83 (3H, s,
3
MHz; CDCl
(
3
3
2
), 2.92 and 3.04 (each 1H, 2 × d, J ) 13.4, CH
2
), 6.44
+
(
1H, d, J ) 6.8, ArC(2′)H), 6.73 (1H, d, J ) 7.3, o-H of o-tolyl
2
101 (4) and 84 (8). Anal. Calcd for C18
H
14O: C, 87.77; H, 5.73.
13
CH
), 6.90-7.82 (13H, m, ArH); C NMR (67.8 MHz; CDCl
), 30.6 (CH ), 47.1 (CH ), 76.7 (C-OH), 124.4-146.6
aromatic signals). Anal. Calcd for C26 24O: C,88.60, H, 6.86.
Found: C, 88.89, H, 6.95. Compound 4b was recrystallized from
3
) δ
Found: C, 87.66; H, 5.97. HRMS calcd for C18
H
14O 246.1045,
1
(
9.7 (ArCH
3
3
2
found 246.1049. This compound has been mentioned but not
H
characterized in a recent communication.12
1-[2-(1-Methylphenyl)]ethanone (2). Palladium tetrakistriphen-
1
hexane-chloroform, mp 97-98 °C: H NMR (270 MHz; CDCl
δ 1.03 (3H, s, CH ), 1.66 (1H, s, OH), 1.95 (3H, s, ArCH ), 3.05
2H, s, CH ), 6.85-7.84 (15H, m, ArH); C NMR (67.8 MHz;
CDCl ) 19.9 (ArCH ), 29.2 (CH ), 46.3 (CH ), 76.3 (C-OH),
24.4-150.2 (aromatic signals). Anal. Calcd for C26 24O: C, 88.60,
H, 6.86. Found: C, 88.59; H, 6.86).
-Phenyl-2-[2-(2-methylphenyl)phenyl]propan-2-ol (5). A so-
3
)
ylphosphine (1.0 g, 0.8 mmol) in benzene (20 mL), 2-bromoac-
etophenone (4.5 g, 22 mmol), a 2 M NaOH solution (20 mL), and
o-tolylboronic acid 0.4 g, 2.32 mmol) were combined according to
procedures described for the preparation of ketone 1 to give the
crude product. Flash chromatography on silica gel eluting with
ether-hexane (1:9) afforded 2 as a colorless oil (4.28 g, 81%):
3
3
13
(
2
3
3
3
2
1
H
1
-
1
1
V
3
max(film)/cm 1682 (CdO); H NMR (270 MHz; CDCl ) δ 1.98
13
lution of 1-[2-(1-methylphenyl)]ethanone (2) (2.0 g, 9.5 mmol) in
dry ether (15 mL) was added dropwise to an ether solution of
benzylmagnesium chloride [from magnesium turnings (0.55 g, 23
mmol) and benzyl chloride (3.20 g, 25 mmol) in dry ether (30 mL)]
under nitrogen. The reaction mixture was heated at reflux for 12 h
(
(
(
3H, s, CH
67.8 MHz; CDCl
aromatic signals), 202.6 (CdO). HRMS calcd for C15
3
), 2.14 (3H, s, CH
3
), 7.09-7.69 (m, ArH); C NMR
3
) δ 20.0 (CH
3
), 29.6 (CH ), 125.7-141.8
3
14
H O
2
10.1047, found 210.1045. This compound has been recently been
1
3
reported elsewhere.
and cooled, and a saturated solution of NH
aqueous phase was extracted with ether. The organic layer was
washed with aqueous NaHCO and water, dried over magnesium
4
Cl was added. The
1-Phenyl-2-[2-(1-naphthyl)phenyl]propan-2-ol (3). A solution
of 1-[2-(1-naphthylphenyl] ethanone 1 (1.45 g, 5.92 mmol) in dry
ether (10 mL) was added dropwise under nitrogen to an ether
solution of benzylmagnesium chloride [from magnesium turnings
3
sulfate, filtered, and concentrated under reduced pressure to give a
crude product containing a mixture of two alcohol atropisomers
(
0.36 g, 15 mmol) and benzyl chloride (2.06 g, 16 mmol) in dry
ether (30 mL)]. The reaction mixture was heated at reflux for 12 h
and cooled, and a saturated solution of NH Cl was added. The
aqueous phase was extracted with ether. The organic layer was
washed with NaHCO solution and water, dried over magnesium
1
5
a and 5b (ratio 1:2.7 from H NMR analysis). The crude product
was purified by flash chromatography on silica gel eluting with
ether-hexane to give mixed samples of the atropisomers as
colorless solids (2.58 g, 92%). Recrystallization of the initial band-
head sample from chloroform-hexane afforded compound 5a, mp
4
3
sulfate, filtered, and concentrated under reduced pressure to give a
crude product containing a mixture of two alcohol atropisomers
7
1
3-74 °C: 1H NMR (270 MHz; CDCl
3
) δ 1.56 (1H, br s, OH,),
), 2.86 and 3.14 (1H each, AB
), 5.89 (1H, d, J ) 7.3, ArC(6′)H), 6.73
3 3
.59 (3H, s, CH ), 2.03 (3H, s, CH
1
3a and 3b (ratio 1:3 from H NMR analysis). The crude product
2
quartet, J ) 13.1, CH
2
was purified by flash chromatography on silica gel eluting with
ether-hexane to give alcohol 3a (0.33 g, 16%) and alcohol 3b (0.93
g, 46%) both as colorless solids. Compound 3a was recrystallized
3
13
[
(
(
2H, m, J ) 8.2, o-H of PhCH
67.8 MHz; CDCl ) δ 20.7 (ArCH
-C-OH), 124.7-144.6 (aromatic signals). Anal. Calcd for
22O: C, 87.37; H, 7.33. Found: C, 87.09; H, 7.34. Recrys-
2
), 6.81-7.64 (m, ArH); C NMR
3
3
), 30.3 (CH ), 51.8 (CH ), 77.0
3
2
CH
2
1
from chloroform-hexane, mp 106-107 °C: H NMR (270 MHz;
C
22
H
2
CDCl
2.8, PhCH
o-H of PhCH
CDCl ) δ 30.62 (CH
signals). Anal. Calcd for C25
3
) δ 1.52 (3H, s, CH
3
), 2.82 and 3.13 (each 1H, 2 × d, J )
tallization of the band-tail sample gave a pure sample of compound
1
2
), 6.19 (1H, dd, J ) 7.0, 1.3, Ar(2′)H), 6.77 (2H, m,
), 6.95-7.82 (13H, m, ArH); 13C NMR (67.7 MHz;
), 75.9 (C-OH), 121.7-145. 8 (aromatic
22O: C, 88.72; H, 6.55. Found: C,
1
5
b, mp 67-68 °C: H NMR 270 MHz; CDCl
CH ), 1.55 (1H, br s, OH), 1.92 (3H, s, CH ), 2.98 (1H, d, J )
3.5, CH ), 6.81-7.64 (m,
) δ 20.4 (ArCH ), 29.3 (CH ),
,), 75.8 (C-OH), 124.7-144.6 (aromatic signals). Anal.
Calcd for C22 20O: C, 87.37; H, 7.33. Found: C, 87.57; H, 7.46.
3
) δ 1.27 (3H, s,
2
2
3
3
3
3
2
1
A
H
B
), 3.14 (1H, d, J ) 13.5, CH
ArH); C NMR (67.8 MHz; CDCl
0.1 (CH
A B
H
H
13
3
3
3
8
8.75, H, 6.33.
Compound 3b was recrystallized from chloroform-hexane, mp
5
2
H
1
9
4-95 °C: H NMR (270 MHz; CDCl ) δ 1.09 (3H, s, CH ); 1.65
3
3
2
1-(2-Methylphenyl)-2-[2-(2-methylphenyl)phenyl]-propan-2-
ol (6). A solution of 1-[2-(1-methylphenyl)]ethanone 2 (2.0 g, 9.5
mmol) in dry ether (15 mL) was added dropwise under nitrogen to
an ether solution of 2-methylbenzylmagnesium bromide [from
magnesium turnings (0.57 g, 23 mmol) and 2-methylbenzyl bromide
(1H, s, OH), 2.94 and 3.14 (1H each, AB system, 2 × d, JAB 13.5,
13
2 3
CH
), 6.84-7.83 (16H, m, ArH); C NMR (270 MHz; CDCl
), 51.3 (CH ), 77.1 (C-OH), 125.2-146.9 (aromatic
signals). Anal. Calcd for C25 22O; C, 88.72; H, 6.55. Found: C,
8.32; H, 6.56.
) δ
3
0.7 (CH
3
2
H
8
(
4.57 g, 24.7 mmol) in dry ether (30 mL)]. The reaction mixture
was heated at reflux for 12 h and cooled, and a saturated solution
of NH Cl was added. The aqueous phase was extracted with ether.
The organic layer was washed with aqueous NaHCO and water,
(
12) Tobias, J. K.; Knochel, P. Angew. Chem. Int. Ed. 2005, 44, 2947-
952.
13) de Koning, C. B.; Michael, J. P.; Rousseau, A. L. J. Chem. Soc.,
Perkin Trans. 1 2000, 787-797.
4
2
(
3
dried over magnesium sulfate, filtered, and concentrated under
J. Org. Chem, Vol. 71, No. 6, 2006 2281

Jennings et al.
reduced pressure to give a crude product containing a mixture of
two alcohol atropisomers 6a and 6b (ratio 1:2 from H NMR
analysis). The crude product was purified by flash chromatography
on silica gel eluting with ether-hexane to give a sample of mixed
isomers (2.12 g, 70%) all as colorless solids. Recrystallization of
parameters for non-hydrogen atoms; hydrogens located in difference
Fourier but included at positions calculated from the geometry of
the molecule using the riding model, R1 ) 0.043 for 1034 data
1
with F
of fit ) 1.06, ∆Fmin,max ) -0.17/0.15 e Å . Although 4a is racemic
the chiral space group P2 implies that individual crystals are
o
> 4σ(F ), 248 parameters, wR2 ) 0.117 (all data), goodness
o
-3
the initial band-head sample from chloroform-hexane afforded a
1
1
pure sample of 6a, mp 60-62 °C: H NMR (270 MHz; CDCl
3
) δ
), 2.02
), 6.07 (1H, d,
J ) 7.3, ArC(6′)H), 6.67 (1H, d, J ) 7.3, o-H of o-tolyl-CH ),
) 19.8,
), 124.8-144.5
24O: C, 87.30; H, 7.64.
enantiopure. This was confirmed by recording the CD spectra of a
number of randomly selected single crystals, dissolved in aceto-
nitrile. Each crystal was shown to be enantiopure and examples of
each enantiomer were identified. Thus the racemic compound was
a spontaneously resolved conglomerate comprising RM crystals and
SP crystals.
Variable Temperature NMR and Kinetic Studies. Variable-
temperature NMR spectra were recorded at 270 MHz. Probe
temperatures were calibrated using a digital copper-constantan
thermometer with the fine sensor inserted into an NMR tube
containing 0.6 mL of the appropriate solvent. Proton chemical shifts
1.60 (3H, s, CH
3 3
), 1.69 (1H, br s, OH), 1.94 (3H, s, CH
3H, s, CH ), 3.04 (2H, AB quartet, J ) 13.3, CH
2
(
3
2
2
13
6
.84-7.52 (10H, m, ArH); C NMR (67.8 MHz; CDCl
ArCH ) 20.7 (ArCH ), 30.5 (CH ), 47.6 (CH
aromatic signals). Anal. Calcd for C23
3
(
(
3
3
3
2
H
Found: C, 87.56; H, 7.68. Atropisomer 6b could not be obtained
completely free of 6a but a chromatographic fraction highly
enriched in 6b gave the following NMR data: (270 MHz; CDCl
δ 1.23 (CH ), 1.68 (1H, br s, OH), 2.01 (ArCH ), 2.13 (ArCH
.06 (2H, m, CH
MHz; CDCl ) δ 20.9, (ArCH
), 125.1-145.8 (aromatic signals).
Crystal Data for 3a. C25
3
)
3
3
3
),
), 6.84-7.75 (12H, m, ArH); C NMR (67.8
) 21.0 (ArCH ), 29.0 (CH ), 46.5
13
3
2
2
were referenced to the residual CHDCl solvent signal set at δ 5.366
3
3
3
3
at all temperatures. Atropisomer equilibration experiments were
carried out in an NMR tube inserted into a constant-temperature
bath. The tube was removed at appropriate time intervals, cooled
in ice and the isomer ratio determined by integration of the
(CH
2
H22O, M
r
) 338.4, monoclinic, a )
1
0.563(2), b ) 9.546(2), c ) 19.386(4) Å, â ) 105.2(1)°, V )
3
1
1
886.4(11) Å , T ) 293 K, Cu KR radiation, λ ) 1.54178 Å, space
appropriate methyl signals in the H NMR spectrum. Rate constants
-
3
group P2
1
/c, Z ) 4, D
x
) 1.192 g cm , colorless blocks, 0.60 ×
were determined from a standard reversible first-order plot of ln-
-
1
0
.50 × 0.50 mm, µ ) 0.54 mm , F(000) ) 720, Siemens P3/
[X
e
/(X
e
- X)] versus temperature affording a slope of (k
f
+ k
r
).
V2000 diffractometer, ω scan, 10° < 2θ < 110°, measured/
independent reflections 2509/2355, direct methods solution, full
Supporting Information Available: 1H NMR spectra of both
2
atropisomers of compounds 3-6, graphs of observed and calculated
matrix least squares refinement on F
o
, anisotropic displacement
1
H chemical shifts versus temperature for the upfield aromatic signal
parameters for non-hydrogen atoms, hydrogens located in difference
Fourier but included at positions calculated from the geometry of
the molecule using the riding model, R1 ) 0.046 for 1760 data
in 3a and 5a, and crystallographic details for 3a and 4a, including
files in CIF format. This material is available free of charge via
the Internet at http://pubs.acs.org. The CIF files are also available
from the Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge, CB2 1EZ, U.K. [fax: +44(0) 1223 336033 or e-mail
deposit@ccdc.cam.ac.uk]. CCDC reference numbers 284636 (3a)
and 284637 (4a).
with F
of fit ) 1.07, ∆Fmin,max ) -0.13/0.15 e Å
Crystal Data for 4a. C26 24O, M
o
> 4σ(F ), 238 parameters, wR2 ) 0.123 (all data), goodness
o
-
3
.
H
r
) 352.5, monoclinic, a )
7
.526(2), b ) 16.602(3), c ) 7.809(2) Å, â ) 98.9(1)°, V )
3
9
64.1(3) Å , T ) 293(2) K, Cu KR radiation, λ ) 1.54178 Å, space
-
3
group P2
1
, Z ) 2, D
x
) 1.214 g cm , colorless blocks, 0.53 ×
Acknowledgment. We thank Dr Paul Stevenson, Queen’s
University, for helpful discussion relating to the HMBC, HMQC
and TOCSY NMR spectra.
-1
0
.47 × 0.43 mm, µ ) 0.55 mm , F(000) ) 376, Siemens P3/
V2000 diffractometer, ω scan, 10° < 2θ < 110°, measured/
independent reflections 1265/1155, direct methods solution, full
2
matrix least squares refinement on F
o
, anisotropic displacement
JO0520902
2282 J. Org. Chem., Vol. 71, No. 6, 2006