Application of the Suzuki reaction as the key step in the synthesis of a novel atropisomeric biphenyl derivative for use as a liquid crystal dopant

Article abstract of DOI:10.1021/jo034652s

A heavily functionalized atropisomeric biphenyl derivative (4), which is designed to possess a large lateral dipole moment, has been synthesized with use of a Suzuki coupling as the key step and resolved by chiral HPLC. The final coupling reaction is complicated by rapid hydrolytic deboronation of the sterically hindered, electron poor boronate 22. Rigorously anhydrous conditions are therefore necessary to achieve the coupling.

Full text of DOI:10.1021/jo034652s

Ap p lica tion of th e Su zu k i Rea ction a s th e
Key Step in th e Syn th esis of a Novel
Atr op isom er ic Bip h en yl Der iva tive for Use
a s a Liqu id Cr ysta l Dop a n t
Andrew N. Cammidge* and Karen V. L. Cre´py
Wolfson Materials and Catalysis Centre, School of Chemical
Sciences and Pharmacy, University of East Anglia,
Norwich NR4 7TJ , United Kingdom
With the above criteria in mind it is clear that
atropisomeric biphenyls could fulfill the role of ideal
dopants for smectic liquid crystals leading to chiral
smectic C phases. Indeed the first examples of such
materials have been reported recently by Lemieux,6 who
prepared derivatives of 4,4′-dihydroxy-2,2′-dimethyl-6,6′-
dinitrobiphenyl such as 2. The materials (resolved by
chiral HPLC) were not liquid crystalline but were capable
of inducing spontaneous polarization when doped into
host materials.
One of the most useful and versatile methods for
synthesizing biaryls is the Suzuki reaction7 whereby an
arylboronic acid (or derivative) is coupled with an aryl
halide (for example) under palladium catalysis. The
reactions generally work well for simple substrates but
problems are often encountered when sterically hindered
partners are employed.8,9 We recently reported10 the first
examples of intermolecular asymmetric Suzuki reac-
tions11 leading to atropisomeric binaphthalenes in rea-
sonable yields and ee values, and further examples have
since appeared.12 We reasoned that the Suzuki reaction
could be employed as the key step in the synthesis of
dipolar, unsymmetrical biphenyls for use as liquid crystal
dopants, and that the asymmetric modification would
lead to enantiomerically enriched material.
In this paper we report the synthesis of a heavily
functionalized biphenyl derivative 4. The choice of 4 as
a target molecule stems from the design criteria set out
above. A typical smectic C liquid crystal host material is
difluorobiphenyl 3.5 Elaboration of this structure to give
a molecule with sufficient hindrance to prevent racemi-
zation and bear a lateral dipole moment yields structure
4 (by definition the chiral center of such atropisomers is
on the core). Fluoride and nitrile substituents were
chosen to induce the dipole moment because of their
compatibility with use in liquid crystal devices. Apolar
(methyl) groups were chosen to occupy the 6,6′-positions.
a.cammidge@uea.ac.uk
Received May 15, 2003
Abstr a ct: A heavily functionalized atropisomeric biphenyl
derivative (4), which is designed to possess a large lateral
dipole moment, has been synthesized with use of a Suzuki
coupling as the key step and resolved by chiral HPLC. The
final coupling reaction is complicated by rapid hydrolytic
deboronation of the sterically hindered, electron poor bor-
onate 22. Rigorously anhydrous conditions are therefore
necessary to achieve the coupling.
Restricted rotation around the aryl-aryl bond in
biaryls can lead to the phenomenon of atropisomerism
(sometimes referred to as helical chirality). The chiral
biaryl motif is encountered in some natural products1 but
is most commonly exploited in the field of asymmetric
synthesis2 where derivatives of, for example, binaphtha-
lene find widespread use. The biphenyl unit itself is
commonly encountered as a core in the liquid crystal field
where derivatives have found commercial application
(most notably in displays).3,4 Fast switching ferroelectric
display devices require a smectic liquid crystal that is
chiral and exhibits a room-temperature chiral smectic C
phase. Furthermore, to switch the device the molecules
must possess a significant lateral dipole moment. In real
systems this is most commonly achieved by using mix-
tures whereby a host liquid crystal is “doped” with a
chiral additive. For maximum effect (induction of high
spontaneous polarization with minimal disruption of
liquid crystal properties) the stereopolar unit (which
includes the chiral center and all coupled polar functional
groups contributing to a transverse molecular dipole)
should be as close as possible to the core. Furthermore,
to ensure compatibility the dopant structure should
ideally bear close resemblance to the hosts (for example
polyfluorobiphenyls such as 15).
(6) Lemieux, R. P. Acc. Chem. Res. 2001, 34, 845.
(7) (a) Miyaura, N.; Suzuki, A. Chem. Rev. 1995, 95, 2457. (b) Kotha,
S.; Lahiri, K.; Kashinath, D. Tetrahedron 2002, 58, 9633.
(8) Buchwald has recently reported an improved catalyst for cou-
pling certain sterically hindered substrates: Yin, J .; Rainka, M. P.;
Zhang, X.-X.; Buchwald, S. L. J . Am. Chem. Soc. 2002, 124, 1162.
(9) Fu has recently reported that Negishi coupling can be success-
fully employed to couple hindered substrates but noted that homocou-
pling was a significant side reaction when tetra-o-substituted biphenyls
were synthesized: Dai, C.; Fu, G. C. J . Am. Chem. Soc. 2001, 123,
2719.
(10) Cammidge, A. N.; Cre´py, K. V. L. Chem. Commun. 2000, 1723.
(11) A diastereoselective Suzuki coupling was reported by Nico-
laou: Nicolaou, K. C.; Li, H.; Boddy, C. N. C.; Ramanjulu, J . M.; Yue,
T.-Y.; Natarajan, S.; Chu, X.-J .; Bra¨se, S.; Ru¨bsam, F. Chem. Eur. J .
1999, 5, 2584.
(1) Dewick, P. M. Medicinal Natural Products; Wiley: Chichester,
UK, 2002.
(2) Pu, L. Chem. Rev. 1998, 98, 2405.
(3) Collings, P. J .; Hird, M. Introduction to Liquid Crystalss
Chemistry and Physics; Taylor and Francis: London, UK, 1997.
(4) Goodby, J . W.; Clark, N. A.; Laggerwall, S. T. Ferroelectric Liquid
Crystals: Principles, Properties and Applications; Gordon and Breach:
Philadelphia, PA, 1991.
(5) (a) Gray, G. W.; Hird, M.; Lacey, D.; Toyne, K. J . Mol. Cryst.
Liq. Cryst. 1990, 191, 1. (b) Gray, G. W.; Hird, M.; Lacey, D.; Toyne,
K. J . J . Chem. Soc., Perkin Trans. 2 1989, 2041. (c) Glendenning, M.
E.; Goodby, J . W.; Hird, M.; Lacey, D.; Toyne, K. J . J . Chem. Soc.,
Perkin Trans. 2 2000, 27. (d) Glendenning, M. E.; Goodby, J . W.; Hird,
M.; Lacey, D.; Toyne, K. J . J . Chem. Soc., Perkin Trans. 2 1999, 481.
(12) (a) Yin, J .; Buchwald, S. L. J . Am. Chem. Soc. 2000, 122, 12051.
(b) Castanet, A.-S.; Colbert, F.; Broutin, P.-E.; Obringer, M. Tetrahe-
dron: Asymmetry, 2002, 13, 659.
10.1021/jo034652s CCC: $25.00 © 2003 American Chemical Society
Published on Web 07/26/2003
6832
J . Org. Chem. 2003, 68, 6832-6835

SCHEME 1. Molecu la r a n d Syn th etic Design of
Ta r get Bip h en yl 4
as a minor product. An activating (directing) methoxy
group was therefore introduced by displacing the iodide
by treatment with sodium methoxide and catalytic cop-
per(I) iodide14 giving 18. This reaction resulted in some
reduced product 13, which was recycled. Chloromethyl-
ation of 18 successfully yielded the correct regioisomer.
However, it proved most convenient to introduce the
methyl group directly via a further ortho-lithiation.
Consequently anisole 18 was treated with butyllithium
in cyclohexane/THF15 and then quenched with methyl
iodide to afford 19. The use of carefully controlled
conditions reduced the formation of a side product
resulting from displacement of one fluoride by butyl-
lithium. Hydrolysis of 19 produced the phenol, which was
deoxygenated via triflate 20 to give 21.16 The target
boronic acid 5 was prepared in modest yield (42%) by a
further ortho-lithiation of 21 followed by reaction with
trimethylborate and aqueous workup. Significant quanti-
ties (30%) of unreacted starting material were isolated
from this reaction. The corresponding boronate esters 22
and 23 were prepared by reaction of 5 with the appropri-
ate diol. It is worth noting that direct synthesis of pinacol
boronate 23 from triflate 20 was attempted with Masu-
da’s conditions17 but resulted in no reaction.
SCHEME 2. P la n n ed Syn th esis of Bor on ic Acid 5
Coupling partner 6 was synthesized following the route
shown in Scheme 4, using a statistical desymmetrization
as the key step. 3,5-Dimethylphenol was brominated with
molecular bromine in acetic acid18 and the resulting
bromide 25 was smoothly alkylated to give symmetrical
intermediate 26. Benzylic bromination of 26 was achieved
with use of NBS and catalytic benzoyl peroxide under
irradiation by a 150-W tungsten lamp. By using this
method, dibromide 27 could be obtained in a yield of 45%.
A number of standard methods were employed for the
conversion of benzyl bromide 27 into the corresponding
amine 28. It proved most efficient to use a two-step
procedure whereby the azide was prepared and subse-
quently reduced with lithium aluminum hydride afford-
ing an overall yield of about 80%.19 Oxidative dehydro-
genation of 28 proved troublesome. Standard conditions
employing nickel peroxide20 or sodium hypochlorite21
resulted in the formation of large amounts of aldehyde
(from hydrolysis of the intermediate imine). The best
conditions involved the use of copper(I) chloride/oxygen
in dry pyridine in the presence of molecular sieves22 and
yields of over 80% were possible.
Simple molecular modeling (Hyperchem-AM1) and cal-
culation of interference values13 led to the prediction that
4 would be stable to racemization even at elevated
temperatures.
The key disconnection for our synthesis is also shown
in Scheme 1. It was envisaged that boronic acid 5 would
be easily synthesized from commercially available 3,4-
difluorotoluene 7 according to Scheme 2. 7 was therefore
ortho-lithiated with butyllithium at -78 °C and quenched
with heptanal. Analysis of the crude product clearly
indicated that a mixture of regioisomeric alcohols 8 and
9 had been formed in a ratio of 8:3. This mixture proved
impossible to separate by chromatography and modifica-
tion of the reaction conditions to favor exclusive formation
of the desired regioisomer 8 (lower temperature, t-BuLi)
led to no improvement. The isomeric mixture was there-
fore dehydrated (phosphorus pentoxide) but again the
resulting mixture of regioisomeric alkenes proved in-
separable.
Initial attempts at Suzuki coupling reactions between
boronic acid 5 and bromide 6 were performed with either
Boronic acid 5 and its corresponding pinacol ester 22
were eventually synthesized from difluorobenzene itself
according to Scheme 3. 1,2-Difluoro-3-heptylbenzene was
synthesized via ortho-lithiation following the reported
procedure.5b Direct electrophilic halogenation of 13 was
not possible (it resulted only in formation of 14) so a
second ortho-lithiation was employed and the resulting
aryllithium intermediate was quenched with iodine giv-
ing 15. Treatment of 15 with methoxyacetyl chloride/tin
chloride at room temperature resulted in slow chloro-
methylation with the desired regioisomer being produced
(14) McKillop, A.; Howarth, B. D.; Kobylecki, R. J . Synth. Commun.
1974, 19, 865.
(15) Slocum, D. W.; Dietzel, P. Tetrahedron Lett. 1999, 40, 1823.
(16) (a) Ritter, K. Synthesis 1993, 735. (b) Cabri, W.; De Bernardinis,
S.; Francalanci, F.; Penco, S. J . Org. Chem. 1990, 55, 350. (c) Cacchi,
S.; Ciattini, P. G.; Morera, E.; Orta, G. Tetrahedron Lett. 1986, 27,
5541.
(17) Murata, M.; Watanabe, S.; Masuda, Y. J . Org. Chem. 1997, 62,
5458.
(18) Fuson, R. C.; Corse, J .; Welldon, P. B. J . Am. Chem. Soc. 1941,
63, 2645.
(19) Kobayashi, Y.; Shiozaki, M.; Ando, O. J . Org. Chem. 1995, 60,
2570.
(20) Yamazaki, S.; Yamazaki, Y. Bull. Chem. Soc. J pn. 1990, 63,
301.
(13) (a) Stanley, W. M.; Adams, R. J . Am. Chem. Soc. 1930, 52, 1200.
(b) Kleiderer, E. C.; Adams, R. J . Am. Chem. Soc. 1933, 55, 4219. (c)
Kawano, N.; Okigawa, M.; Hasaka, N.; Kouno, I.; Kawahara, Y.; Fujita,
Y. J . Org. Chem., 1981, 46, 389.
(21) Yamazaki, S. Synth. Commun. 1997, 27, 3559.
(22) (a) Capdevielle, P.; Lavigne, A.; Maumy, M. Synthesis 1989,
453. (b) Capdevielle, P.; Lavigne, A.; Sparfel, D.; Baranne-Lafont, J .;
Cuong, N. K.; Maumy, M. Tetrahedron Lett. 1990, 31, 3305.
J . Org. Chem, Vol. 68, No. 17, 2003 6833

SCHEME 3. Syn th esis of Bor on ic Acid 5 a n d Cor r esp on d in g Bor on a te Ester s
SCHEME 4. Syn th esis of 6
SCHEME 5. Attem p ted Su zu k i Cou p lin g To Give
Ta r get Bip h en yl 4 a n d Mod el Cou p lin gs
preformed Pd(Ph3)4 or a catalyst prepared in situ from
palladium chloride and 2 equiv of triphenyl phosphine
under standard conditions. Not surprisingly, these condi-
tions failed to produce any of the desired biphenyl and
unreacted 6 was reisolated along with 21 (resulting from
hydrolytic deboronation). It is well-known that debor-
onation tends to dominate in reactions of hindered
boronic acids23 and in our previous work on the asym-
metric Suzuki coupling of hindered naphthalenes this
problem was overcome by employing the boronate ester
as a coupling partner.10 These (homogeneous) conditions
were applied to the coupling of 6 with ester 22. Unfor-
tunately, the coupling reaction in either DME or THF
with cesium fluoride gave essentially the same disap-
pointing result, prompting a closer examination of the
reaction to obtain information about the individual
mechanistic steps. To this end, bromide 6 and boronic
acid 5 were coupled with benzene boronic acid and
bromobenzene respectively under standard conditions. In
both cases the coupled product was smoothly obtained
(albeit in modest unoptimized yields of around 40%),
demonstrating that oxidative addition of 6 and trans-
metalation of 5 to palladium are facile. Further experi-
ments were performed to test the susceptibility of 22 to
deboronation. It was found that the rate of deboronation
was dramatically accelerated by addition of palladium
catalyst. In one control experiment 22 was stirred in
dried dioxane with cesium fluoride at 100 °C under
nitrogen. Deboronation was complete within 4 h. As
expected use of carefully dried reagents and nitrogen
slowed the deboronation considerably. When the original
experiment was carried out in the presence of a catalytic
quantity of palladium it was observed that deboronation
was complete within minutes. Further control experi-
ments demonstrated that formation of ate complexes is
required to promote deboronation. For example, no
deboronation was observed after 24 h when 22 was
treated with Pd/C, triphenyl phosphine, and triethyl-
amine in refluxing ethanol. Addition of cesium fluoride
to the reaction mixture resulted in complete deboronation
in under 2 h.
These observations permit some assumptions to be
made about the limiting steps in this coupling process
and they, along with the side reactions, are depicted in
Figure 1. The overall rate-limiting step appears to be
reductive elimination, itself likely to be driven by a
further oxidative addition of 6 via a transition state such
as 31.24 (This triggered reductive elimination is likely to
account for the improvement in the coupling of sterically
(24) (a) Gillie, A.; Stille, J . K. J . Am. Chem. Soc. 1980, 102, 4933.
(b) Widdowson, D. A.; Zhang, Y.-Z. Tetrahedron 1986, 42, 2111.
(23) Watanabe, T.; Miyaura, N.; Suzuki, A. Synlett. 1992, 207.
6834 J . Org. Chem., Vol. 68, No. 17, 2003

F IGURE 1. Catalytic cycle and side reactions.
SCHEME 6
hindered substrates when aryl iodides are used in place
of bromides even in cases where it is known that the
“simple” oxidative addition step is facile.) The reductive
elimination can only occur from the cis-complex and the
slow rate could equally plausibly stem from the unfavor-
able (for such hindered substrates) cis-trans equilibrium.
These factors mean that the diaryl palladium species has
a long lifetime and is susceptible to competing processes
such as disproportionation and hydrolysis. The latter
process clearly dominates in this case. Attempts to
improve the reaction by choice of palladium source (such
as PdCl2, Pd(OAc)2, Pd2(dba)3), ligand (such as PPh3, dppf,
dppe, P(OMe)3), solvent (such as DME, DMF, toluene,
dioxane, THF), and base (K3PO4, CsF) met with no
success and in all cases hydrolytic deboronation domi-
nated.
Successful coupling could only be achieved when the
reaction was performed with rigorous exclusion of mois-
ture.25 To this end all reagents were dried under vacuum
and handled under dried nitrogen. The reaction proved
to be poorly reproducible but gave the target biphenyl 4
in a yield of up to 21% when PdCl2 (3%)/PPh3 (6%) was
used as catalyst in DME with cesium fluoride at 90 °C
for 2 days. Application of our previously established
conditions for performing asymmetric Suzuki couplings
failed to yield any biphenyl. A sample of (()-4 was,
however, separated into its enantiomers by chiral HPLC
to permit characterization of the pure compounds (which
are oils).
as the key step. As with other examples of attempts to
couple sterically hindered partners, this reaction proved
difficult to achieve with, in our case, hydrolytic debor-
onation dominating. Model experiments and a careful
analysis of reaction conditions leads to the conclusion
that the slow reaction step is reductive elimination (or
the trans to cis isomerization required to give the
intermediate capable of reductive elimination). Debor-
onation takes place in the absence of palladium but the
reaction rate is dramatically accelerated with palladium
present. This suggests that deboronation takes place from
a diarylpalladium species, setting up a separate catalytic
cycle that consumes the boronate. Coupling can only be
achieved if the deboronation pathway can be stopped by
rigorous exclusion of moisture.
Ack n ow led gm en t. We thank the EPSRC mass
spectrometry service center at Swansea, UK for mass
spectra and Takeda Chemical Industries Ltd for per-
forming preparative chiral HPLC.
In conclusion, a heavily functionalized biphenyl deriva-
tive has been synthesized with use of a Suzuki coupling
Su p p or tin g In for m a tion Ava ila ble: Full experimental
details and comparison of synthetic methods, NMR spectra,
and HPLC. This material is available free of charge via the
Internet at http://pubs.acs.org.
(25) It is interesting to compare our results to those of Buchwald
[ref 8], who reported that strictly anhydrous conditions were required
to prevent destruction of the bromide coupling partner, an observation
that further emphasizes the mechanistic balance required for success-
ful coupling of hindered substrates.
J O034652S
J . Org. Chem, Vol. 68, No. 17, 2003 6835