Chiral oxazoline route to enantiomerically pure biphenyls: Magnesio and copper mediated asymmetric hetero- and homo-coupling reactions

Article abstract of DOI:10.1016/j.tet.2004.01.095

A series of chiral biphenyls were prepared via asymmetric reactions involving chiral oxazolines. One series of chiral biphenyls was reached by the magnesium mediated coupling of aryl bromides (via their Grignard reagents) with o-methoxyaryl oxazolines. In

Full text of DOI:10.1016/j.tet.2004.01.095

Tetrahedron 60 (2004) 4459–4473
Chiral oxazoline route to enantiomerically pure biphenyls:
magnesio and copper mediated asymmetric hetero- and
homo-coupling reactions
b
c
d
e
A. I. Meyers,a Todd D. Nelson, Henk Moorlag, David J. Rawson and Anton Meier
,
*
aDepartment of Chemistry, Colorado State University, Fort Collins, CO 80523-1872, USA
bMerck Process Research, Wayne, PA 19087, USA
cRoche Colorado Corporation, Boulder, CO 80301, USA
dPfizer Laboratories, Sandwich, Kent CT13 9NJ, UK
eBASF, Antwerp, Belgium
Received 17 October 2003; revised 16 January 2004; accepted 21 January 2004
Abstract—A series of chiral biphenyls were prepared via asymmetric reactions involving chiral oxazolines. One series of chiral biphenyls
was reached by the magnesium mediated coupling of aryl bromides (via their Grignard reagents) with o-methoxyaryl oxazolines. In this case
only the o-methoxy group was replaced by the ArMgBr. The initially formed biphenyl adducts were obtained in de’s as high as 92:8. These
adducts could be manipulated to various other chiral biphenyls. Another series of chiral biphenyls were obtained via an asymmetric Ullmann
reaction, which was shown to be thermodynamically controlled. The de’s of this copper mediated process were also in the range of 90% and
could be utilized to reach various derivatives. Racemization, thermal stability, and atropisomerization characteristics were also studied.
q 2004 Elsevier Ltd. All rights reserved.
1. Introduction
were obtained containing a variety of substituents.6 This
readily accessible route to biphenyl derivatives inspired us
to examine the use of chiral oxazolines that may lead to
chiral biphenyls 3. In 1985, we reported7 our first attempts
to prepare non-racemic biphenyls (þ)-3. Although they
were obtained in generally good yields (60–95%), the de’s
varied widely between 0 and 92%. Nevertheless, much was
learned regarding the nature and position of the aryl
substituents. In 1990, we reported8 the asymmetric total
synthesis of (2)-schizandrin 5 (Scheme 2), a dibenzo-
cyclooctane lignin utilizing the chiral 4-methoxymethyl-5-
phenyloxazoline 6. When this was coupled with the
Grignard reagent derived from arylbromide 7, methoxy
displacement occurred giving the appropriate biphenyl in
.97% de. This was ultimately taken on to the biphenyl
lignin, (2)-5.
It has been 30 years, since we first introduced chiral
oxazolines as valuable auxiliaries in asymmetric C–C bond
forming reactions.1 Since then there has been a large
number of advances using oxazolines, and these have been
summarized for chiral nucleophiles,2,3 chiral electrophiles,2,3
and chiral ligands.3b,4
Within the context of this special issue, we will describe our
results leading to chiral, configurationally stable, biphenyls
wherein the oxazoline has been a key component in the
success leading to these goals. Although much of these
studies were reported between 1992 and 1996 as short
communications, further details, including atropisomeriza-
tion and experimental procedures, are presented herein.
In 1975, we reported5 that Grignard reagents displaced
o-methoxy groups from aryl oxazolines 1 at or near room
temperature (Scheme 1). When the Grignard reagent was
derived from an aryl halide, good yields of the biphenyl 2
In a related study, we also earlier reported,10 in brief form,
the copper mediated Ullmann coupling of o-bromoaryl-
oxazolines 8 to give symmetrically substituted biphenyls, 9.
This route to chiral biphenyls also proved to be both
thermodynamically controlled as well as efficient
(Scheme 3).
Keywords: Chiral biphenyls; Atropisomerism; Chiral oxazolines; Ullmann
coupling; Grignard reagents.
The two routes described above will be discussed in the
following sections.
*
Corresponding author. Tel.: þ1-970-491-7060; fax: þ1-970-491-1801;
e-mail address: aimeyers@lamar.colostate.edu
0040–4020/$ - see front matter q 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2004.01.095

4460
A. I. Meyers et al. / Tetrahedron 60 (2004) 4459–4473
Scheme 1.
Scheme 2.
Scheme 3.
2. Magnesio-mediated couplings
nucleophilic substitution leading to 12 was very slow (6%
after 21 h). At 65 8C, the methoxy displacement proceeded
much faster, and the biphenyls were formed in 67–90%
yields, with diastereomeric ratios of varying amounts
(Table 1). Furthermore, a number of different substituents
on the aryl bromide 11 gave, as expected, different ratios of
diastereomeric biphenyls. Examining Table 1, it is immedi-
ately clear that, although the coupling yields are generally
independent of the nature of R1, the de’s varied significantly
(Scheme 4).
As mentioned above in the Section 1, the aromatic
nucleophilic displacement of MeO or F groups, ortho to
the oxazoline group, produced high yields of biphenyls. In
the asymmetric variant7 of this reaction, we subsequently
employed a less substituted chiral oxazoline, readily
accessible from inexpensive amino acids, for example,
valine. The latter was reduced9 to valinol and transformed
into the aryl oxazoline 10. The aryl bromides 11, prepared
via known methods11 from m-anisaldehyde, were converted
into their respective magnesio derivatives using the
entrainment method (BrCH2CH2Br)12 and the oxazoline
10, in THF, was added. At room temperature the
This effect by the substituent R1 in 11 appeared to be
dependent on the availability of the oxygen or oxygens to
act as a ligand or a donor to the intermediate magnesium

A. I. Meyers et al. / Tetrahedron 60 (2004) 4459–4473
4461
Table 1. Variation of substituent on aryl Grignard 11 and the effect of
stereoselectivity on 12
species 14A, 14B (Scheme 5). Thus, complexation by the
oxygen atoms may control the stereochemistry of the
biphenyl products, 12. Since nucleophilic entry from the a-
face (13a) may be hindered by the steric bulk of the pendant
i-propyl or t-butyl (R2) on the oxazoline ring, entry should
occur predominantly from the b-face (13B). Free rotation
around the newly formed s intermediates (14A, 14B), prior
to elimination of the magnesium salt, may therefore be
responsible for the S axial stereochemistry observed in the
biphenyl 12. It is, at this stage of the reaction, that steric and
electronic effects of the substituents ortho to the magnesium
come into the picture. If the s intermediate 14A rear-
omatizes with the ortho-methoxy group complexed to the
magnesium (in place of solvent), the axial (S)-biphenyl will
form. However, if the R1 substituent also contains oxygen
lone pairs, then 14B may also be present at this stage of the
reaction, and elimination of magnesium salt from 14B will
lead to the axial (R)-biphenyl. With this assumption in
place, we found it readily explains the variations of de’s
with R1 substituents in Table 1. The poor selectivities are
seen in all cases (entries 1–6) where the oxygen lone pairs
are accessible for complexing with the magnesium in 14B,
thus competing with 14A for dominance in the formation of
the s intermediates. When R1¼Me, there is no possibility
that 14B will be present to any meaningful extent, thus the
observed diastereomeric ratios are rather good (90:10)
Entry
R1
(11)
R2
i-Pr
i-Pr
i-Pr
i-Pr
Biphenyl
12
Yield
(%)
Diastereomeric
ratio
(S: R)-12a
1
2
3
4
12a
12b
12c
12d
90
78
79
77
20:80
60:40
32:68
34:66
5
6
7
8
9
10
CH2OCH2C6H5
CH2OCH3
CH3
CH2OTBDMS
CH2OTBDMS
CH2OTIPS
i-Pr
i-Pr
i-Pr
i-Pr
t-Bu
t-Bu
12e
12f
12g
12h
12i
12j
80
75
79
73
67
70
42:58
40:60
90:10
93:7
92:8
93:7
a
Refers only to axial configuration.
Scheme 4.
Scheme 5.

4462
A. I. Meyers et al. / Tetrahedron 60 (2004) 4459–4473
favoring axial (S)-12. The S,S-notation in 12 (Scheme 5)
refers to oxazoline as well as axial stereochemistry.
Additionally, in entries 8–10, the oxygen donating ability
is diminished by the silicon atom, thus again favoring 14A
as the s intermediate. Although the bulky siloxy substi-
tuents were originally introduced for steric effects in the
transition state, it was found that the smaller methyl group
(entry 7) was almost as effective in providing good axially
chiral biphenyl ratios. Thus, the electronic effects (avail-
ability of lone pair on oxygen) in the magnesio-assisted
couplings appear to be more important than the size of the
substituent.
chances were high they could be synthetically manipulated
without atropisomeric destruction of their stereochemical
integrity. The transformation of (S,S)-12h, one of the better
results presented in Table 1, was used as a typical example
as shown in Scheme 6. Thus, the biphenyl (S,S)-12h,
purified by radial chromatography15 to remove the 6–7% of
the SR diastereomer (Table 1), was converted into the stable
diester amide (S,S)-15 with aqueous trifluoroacetic acid
followed by acetylation of the intermediate ammonium salt
with acetic anhydride. Reduction of 15 with lithium
aluminum hydride furnished the biscarbinol (S)-16 as a
1
single enantiomer. This was confirmed by the H NMR
spectrum of the Mosher ester and compared with the
spectrum of the racemic Mosher ester.16
These results, using oxazolines derived from valinol or
t-leucinol, 10 (R2¼i-Pr, t-Bu) agreed well with an earlier
study13 using a more substituted chiral oxazoline containing
a methoxymethyl group (2)-6. Thus, the chelating methoxy
group on the oxazoline ring appeared not to have any
significant effect, that is, only the alkoxy groups on the
aromatic ring (14A, 14B) appeared to be important.
Others14 have drawn the same conclusion for the factors
controlling the stereochemistry in these magnesium
mediated couplings. The synthesis of these biphenyl
oxazolines should contribute to the growing interest in
various chiral C2-symmetric and non-symmetric biphenyls
that have more recently appeared in the literature.3b,4
It is noteworthy that (S)-16 was totally stable toward
atropisomerization (racemization) showing no change in
[a]D after 5 days at room temperature. This is undoubtedly
due to the two sp3 carbons in the ortho positions.
Oxidation17 of the biscarbinol 16 gave the dialdehyde
(S)-19 in quantitative yield. This material was unstable to
racemization and, after heating at 90 8C for 24–30 h, lost all
of its [a]D value. However, it was quite stable to
racemization under milder conditions (room temperature).
The transformation to the bisphenol (S)-20 was
accomplished in moderate yield (15%) via Baeyer–Villiger
oxidation. The racemization of the dimethoxy bisphenol 20
was also examined and appeared to be stable at room
temperature, but racemized readily at 64 8C (THF) after
48 h. The poor conversion in the Baeyer–Villiger oxidation
of (S)-19 to (S)-20 may be due to the poor migratory
In order to demonstrate the synthetic value of these biphenyl
oxazolines 12, a study was initiated to transform them into
biphenyls containing other functional groups. It is important
to note that by having four ortho substituents in 12, the
Scheme 6. (a) Na2SO4, TFA, THF, H2O; (b) Ac2O, pyridine, DCM; (c) LAH, THF; (d) CrO3, pyridine, DCM; (e) MCPBA, NaHCO3, DCM; (f) K2CO3,
MeOH, H2O; (g) Pd/C (10%). H2, TFA, MeOH; (h) BBr3, DCM.

A. I. Meyers et al. / Tetrahedron 60 (2004) 4459–4473
4463
Scheme 7. (a) Br2 (Ref. 19); (b) NaBH4, ZrCl4; (c) TBSCI, TEA; (d) NaOMe, Cu (Ref. 19); (e) COCl2, (S)-valinol; SOCl2; K2CO3; (f) Mg, BrCH2CH2Br, heat.
aptitude of the m-methoxyphenyl unit. The lack of sufficient
electron-density at the migrating center allows the hydrogen
to shift preferentially. This result is quite consistent with
earlier reports25 of benzaldehyde derivatives where the aryl
ring is a poor competitor to the H in Baeyer–Villiger
reactions.
Thus, the biphenyl oxazoline (S,S)-25 was transformed into
the bisbenzyl alcohol (S)-26, which proceeded via oxidation
with oxalyl chloride/DMSO to the bisbiphenylaldehyde, 27.
Studies were initiated on the atropisomerization of both the
dibenzyl alcohol 268 and the biphenyl diol 28. The latter
was smoothly prepared by a Baeyer–Villiger oxidation of
the aldehyde 27 in 83%, a significant improvement over the
aldehyde oxidation to the diol, (S)-20. The reason for the
hexamethoxy biphenylaldehyde 27 proceeding much more
efficiently than the earlier described dimethoxy derivative
19 may be due to the higher electron density in the
trimethoxybenzene, which would make the aryl group the
preferred migratory group over the weakly nucleophilic m-
methoxy phenyl moiety (in 19).
To overcome the poor yield of bisphenol 20 which would
make a potentially useful new chiral ligand, we found that
the benzylic hydroxy groups in (S)-16 were readily
hydrogenolyzed. Using Pd-charcoal in 2–5% TFA in
ethanol, the hydrogenolysis proceeds under a balloon of
hydrogen to afford the dimethylbiphenyl (S)-17 in quanti-
tative yield. The desired bisphenol (S)-18 was obtained in
95% yield after methoxyl cleavage of (S)-17 with BBr3. The
latter was stable at room temperature toward racemization.
Upon heating (S)-18 at 110 8C (toluene) for 9 h, no
racemization was observed. However, heating to 140 8C
(xylene) for 24 h caused the loss of half the optical activity.
The atropisomerization of both the biscarbinol 26 and the
bisphenol 28 were examined and the rates of racemization
determined by heating and plotting changes in the [a]D
values. As expected, 26 was stable for months at room
temperature as indicated by no change in the [a]D value.
When heated in tetrahydrofuran (66 8C) for 72 h, there was
also no measurable change in [a]D. Once again, the two o-
substituents containing sp3 carbons was considered to be the
factor responsible for this stability. However, when 28 was
heated to reflux (66 8C) in THF, racemization appeared to
begin after several minutes. The original [a]D of 48.9 (2.3
CHCl3) began to linearly decrease in value until after 60 h
of heating, the [a]D had decreased to 24–25 (2.3 CHCl3).
Thus, half of the enantiomeric purity dropped in 60 h. When
28 was heated at 110 8C (toluene), half of its [a]D had
dropped after only 1 h. At room temperature, however, 28
was found to be stable for at least three months.
Since (S)-18 has been previously reported18 via classical
resolution, we were able to assign absolute configuration of
18 as S, which was expected based on our assumed
mechanism in Scheme 5. It should be noted that no axial
racemization was detectable during the sequence from 12 to
18. Based on the above, the major diastereomer formed via
the magnesium coupling to 12 must be S for the oxazoline
stereocenter and S for the axial portion of the molecule.
Furthermore, we were able to prepare multigram quantities
of (S)-18, enantiomerically pure, and in 50–55% overall
yield from 10. This compares with the 15% overall yield for
18 reported earlier18 using a combination of Ullmann
coupling and resolution.
Finally, it should be stated that the use of valinol to convert
the benzoic acid 23 into the oxazoline (S)-24 (Scheme 7)
appeared to possess optimal characteristics. First of all, the
use of tert-leucinol, which would have provided the 4-t-
butyl group in place of the 4-isopropyl group, was found to
be insignificantly more efficient in the magnesio coupling
step to 25 (98:2 for t-Bu vs 97:3 for i-Pr, Table 1, entries 9
and 10). Secondly, the very high cost of tert-leucine, as the
precursor to t-leucinol, did not justify its use for the
relatively small difference in the observed
diastereoselectivity.
Another closely related series of chiral, non-racemic
biphenyls were also reached by a sequence similar to that
described above. In the event, the trimethoxy benzene 22
was coupled via its Grignard reagent to the tetramethox-
yphenyloxazoline (S)-2419 to give an excellent ratio of
axially diastereomeric biphenyls (S,S)-25 and (S,R)-25 as a
98:2 mixture in 89% yield20 (Scheme 7).
In a similar fashion, the synthesis of the C2-symmetric
hexamethoxy bisphenol (S)-28 was prepared (Scheme 8).

4464
A. I. Meyers et al. / Tetrahedron 60 (2004) 4459–4473
Scheme 8. (a) TFA, Na2SO4, THF, H2O; (b) Ac2O, pyridine, DCM; (c) LAH, THF; (d) oxalyl chloride, DMSO, Et3N; (e) MCPBA, NaHCO3, DCM.
3. Ullmann couplings
Using a chiral oxazoline as the chiral control element, there
had been only one other report of an asymmetric Ullmann in
the literature.24 We performed the Ullmann reaction of an
aryl oxazoline10 involving the copper mediated coupling in
DMF for 24 h at 110 8C. The biphenyl, 29,10 obtained in
58–60% yield, showed an axial diastereomeric ratio of SSS
to SSR of 94:6 (Scheme 10).
The Ullmann reaction, known21 since 1901, has attracted
the attention of many investigators trying to prepare chiral
biphenyls. The requirement for maintaining enantiomeric
integrity involves having at least 3 groups in the ortho
positions adjacent to the aryl–aryl bond.22a,b These would
be necessary to inhibit rotation of the aryl–aryl bond and
allow atropisomers to be isolated. The degree of stability in
these systems would, of course, depend on the nature of the
ortho substituents when this asymmetric Ullman coupling,
shown in Scheme 9,23 was carried out.
After isolation of the biphenyl (SSS)-29, the product was
transformed into the previously prepared bisbenzyl alcohol
(S)-26 and then to the previously prepared bisaldehyde 27
and bisphenol 28 (Scheme 8). In all cases, the compounds
prepared from the Ullmann route were identical to those
prepared by the magnesium mediated coupling in Scheme 8.
There has, in the past few years, been a number of reports
utilizing the above chiral Ullmann couplings, and the
stereochemical results obtained by these investigators were
variable leading some to separate the oxazoline biphenyls
into single diastereomers.26 Further studies will need to be
performed to asses the effects of the substituents on the de
observed in the biphenyl. A study done in this laboratory
Scheme 9.
Scheme 10. Cu, DMF, heat; (b) trituration from Et2O/hexanes; (c) TFA, Na2SO4, THF, H2O; (d) Ac2O, pyridine, DCM; (e) LAH, THF.

A. I. Meyers et al. / Tetrahedron 60 (2004) 4459–4473
4465
Scheme 11. (a) Oxalyl chloride, DMF, t-leucinol, SOCl2; (b) Cu powder, DMF, 72 h, reflux.
involving only one methoxy group (Scheme 11), rather than
the three seen in Scheme 10, gave the biphenyl 32 as a 95:5
mixture of diastereomers in 62% yield. The S-axial-
diastereomer was again the major product. It should be
noted that the copper coupling in Scheme 11 was carried out
on the 4-t-butyloxazoline (S)-31 derived from t-leucinol and
2-bromo-3-methoxybenzoic acid 30. The added bulk on the
oxazoline chiral controlling group, may also have had an
effect on the de of the product. This is perhaps due to the
intermediate copper species forcing the two aromatic rings
into close proximity. In this case, t-butyl may be more
selective than i-propyl. A similar result was previously
observed27 in the coupling of naphthalene oxazolines.
aliquots of the mixture were examined at different time
periods. After work-up, NMR examination of the dia-
stereotopic isopropyl resonances in 2910 afforded a measure
(^3%) of the diastereoselectivity for the process. For a
more accurate measure of diastereoselectivity, the crude
bisoxazoline mixtures were transformed in a three-step
process (Scheme 10) to the dicarbinol 26, which was
assayed by chiral high performance liquid chromatography
(HPLC).29 Comparison of the de of 29 (via NMR
spectroscopy) and HPLC of 26 showed them to be virtually
identical. It can be seen from Figure 1 that after 1 h, the de
of 29 was poor (62:38), indicating that there is a slight
preference for the (SSS)-stereochemistry, even in the early
stages. Continuing the heating of the DMF–Cu mixture
containing the bromooxazoline afforded higher ratios of
atropisomers with increasing time, until, after 60 h, the
mixture (SSS)93:(SSR)7, remained constant.
The most interesting aspect of the Ullmann coupling
involving chiral aryloxazoline halides is the course of the
reaction. In Scheme 10, the coupling of (S)-2-bromo-3,4,5-
trimethoxyphenyloxazoline to S-29 was time dependant
with regard to the % de observed (Fig. 1). In order to
demonstrate that a thermodynamically controlled resolution
was responsible for the observed final diastereoselectivity of
29 (93:7),10 the diastereomeric ratio of the two atropisomers
was monitored over 160 h.
The equilibrium noted above may accurately referred be to
as a first-order asymmetric transformation30a or a dynamic
resolution.30b This involves establishing an equilibrium
between the two diastereomers (SSS-29, SSR-29) such that
the undesired one would be converted into the other under
specified reaction conditions. In this manner, rather than
obtain a 50% yield of enantiomerically pure product from
classical or kinetic resolution, the theoretical yield would be
Thus, a mixture containing bromotrimethoxyoxazoline,10
dry DMF, and copper powder28 was heated to reflux and
Figure 1.

4466
A. I. Meyers et al. / Tetrahedron 60 (2004) 4459–4473
100%. In other words, a 1:1 mixture would be potentially
converted to a diastereomerically enriched or pure product
under the equilibrating conditions. The results of heating the
bisoxazoline 29 such that it changed from a 2:1 ratio to the
final 13:1 ratio of SSS over SSR is clearly an example of this
behavior.
or III) is involved, it can bring together, in close proximity,
the two oxazoline rings (B), which would otherwise be
further apart with minimum non-bonded interactions. Thus,
there is, under the conditions given, a 6:1 turnaround in the
stereochemistry of 33 in the presence or absence of a
chelated metal.34 Recently, Wulff,35 Andrus,36 and Ikeda37
have noted the importance of Cu species in affecting the
stereochemical ratios of biphenyl oxazolines with regard to
the deracemization, stereochemical integrity, and catalytic
efficiency.
The factors responsible for this reasonably effective first
order asymmetric synthesis can be presented with some
degree of confidence. First, although the biaryl 29 is
tetrasubstituted about the chiral axis, under the heated
reaction conditions (refluxing DMF), rotation is still
possible about the biaryl axis. Second, the final dia-
stereomeric ratio may be the result of chelation control
utilizing a Cu(I),31 Cu(II),32 or Cu(III)26b species A, B
(Fig. 1) under equilibrating conditions. Thus, although only
a small diastereomeric excess was initially observed, the
ratio increased steadily until equilibrium for this process
was established. The diastereomeric Cu complexes A, B
indicate how the copper might be complexed between the
two oxazoline moieties. Severe non-bonded interactions
between the isopropyl groups in B result in the equilibrium
being favored to lie towards A, which then led to (SSS)-29.
To further demonstrate how acquisition of chiral biphenyl
oxazolines could serve as precursors to other axially chiral
biphenyl derivatives, we examined some routes to the
dibenzoazepine, 34 (Scheme 13). The latter systems have
been prepared earlier by Hawkins.38 By using the previously
prepared bisbenzyl alcohol (S)-26, we utilized the Hawkins
route to prepare dibenzoazepine 34. The product was
formed, albeit in 28% yield. More significantly, however,
was the fact that there was no observable atropisomerization
(racemization) of 34 after refluxing in THF for 12 h.
This was performed by periodically monitoring the optical
rotation ([a]D¼246.3[2.0 CH2Cl2]), which held constant.
The secondary amine 34 could also be converted, by known
means to the tertiary amine 35 (R1¼R2¼OMe) or the latter
could be directly prepared from the bisaldehyde 27 (Scheme
14) by reductive amination in excellent yield.39 Similarly, the
bisaldehyde 19 was transformed into the corresponding
dibenzoazepine 35 (R1¼R2¼H). The above sequence is
anotherexampleoftheproductsfromtheasymmetriccoupling
being obtained in high enantiomeric purity when the coupling
is carried out under substrate dependent conditions.40
In a further attempt to assess the role of the copper species in
this equilibration behavior, a 3:1 mixture of biphenyl
oxazolines 33 (75% SSR, 25% SSS) was obtained from the
corresponding bromooxazolines.33 The mixture was heated
in DMF (160 8C) for 2 days in the absence of any copper
catalyst and slowly changed to the diastereomer (SSR:SSS,
3:1). On the other hand, when Cu(I) and (II) were added to
the above 3:1 mixture and heated for 2 days, no change from
the SSR:SSS, 3:1 starting mixture occurred. (The absolute
configuration assignments at the biphenyl axis have
changed due to priority of the methyl groups in 33 vs the
methoxyl in 29.) These results clearly support the Cu
chelation shown in Figure 1 (Scheme 12). In the absence of
Cu, there is transformation to the thermodynamically more
favored biaryl-S axis, albeit by 3:1. However, when Cu(I, II,
Finally, it should be stated that the chiral bisphenol 28 has
been shown to be a rather useful ligand for LiAlH4
reductions of various ketones.41 Furthermore, bisphenol
18 prepared in high enantiomeric purity via the magnesio-
mediated coupling (Scheme 6) has already been reported by
Scheme 12.
Scheme 13. (a) PBr3; (b) trifluoroacetamide; (c) K2CO3.

A. I. Meyers et al. / Tetrahedron 60 (2004) 4459–4473
4467
Scheme 14. (a) Ethyl formate; (b) LAH; (c) MeNH2·HCl; (d) NaBH3CN.
Suda42 to also serve as an efficient chiral ligand for hydride
reductions of prochiral ketones. The bisphenol used in the
latter work was prepared by resolution of the racemic 2,20-
dihydroxy-6,60-dimethyl biphenyl.
distillation under argon from an appropriate drying
agent. Dichloromethane, benzene, toluene, triethylamine,
diisopropylamine, and acetonitrile were dried via
distillation from calcium hydride and maintained under
an inert atmosphere (Ar or N2). Tetrahydrofuran and
diethyl ether were dried via distillation from sodium
benzophenone ketyl, and maintained under an inert
atmosphere. Pyridine was distilled from calcium hydride
4. Experimental
4.1. General
˚
and stored over 4 A molecular sieves. Dimethylsulfide,
packaged under nitrogen in 1 L Sure/Seal bottles, was
purchased from Aldrich.
1H NMR spectra were recorded at 300 or 400 MHz and data
were reported as follows: chemical shifts, in parts per
million referenced to the internal chloroform (CHCl3). 13C
NMR spectra were recorded at 75 or 100 MHz, with
chemical shifts referenced to the central peak of the CDCl3
triplet (77.0 ppm). Infrared absorption spectra were
recorded on a Perkin–Elmer model PE 1600 spectro-
photometer. Low resolution mass spectra (GC-MS) were
obtained with a Hewlett–Packard model 5890 instrument
equipped with a Hewlett–Packard 5970B mass selective
detector (ionization potential 70 eV). Elemental analyses
were performed by Atlantic Microlab Inc., Norcross,
Georgia. Optical rotations were determined with a
Rudolph Research Autopol III instrument and were
referenced to the sodium D line (598 nm). Melting points
were measured in open Pyrex capillary tubes and are
uncorrected. High performance liquid chromatography
(HPLC) was performed with Regis Pirkle Covalent and
Characil OD chiral columns. Thin layer chromatography
(TLC) and flash chromatography were performed with
E. Merck or Amicon Matrix silica gel (230–400 mesh),
4.2. 2-(2,3-Dimethoxyphenyl)-4-i-propyloxazoline (10)
R25i-Pr)
Oxalyl chloride (1.91 g, 15 mmol) was added to a solution
of 2,3-dimethoxybenzoic acid (1.82 g, 10 mmol) in CH2Cl2
(25 mL) at room temperature. Five drops of DMF were
added, causing effervescence. After stirring for 3 h, the
volatiles were removed in vacuo. The crude yellow acid
chloride was dissolved in CH2Cl2 (20 mL) and added to a
stirred solution of (S)-valinol (1.13 g, 11 mmol) and
triethylamine (1.21 g, 12 mmol) in CH2Cl2 (20 mL) at
0 8C over 30 min. Stirring was continued for another 8 h,
and the reaction was quenched with 1 N HCl and extracted
from the mixture with CH2Cl2 (3£50 mL). The organic
layers were dried (Na2SO4) and concentrated in vacuo. The
crude intermediate amide was dissolved in CH2Cl2 (30 mL)
and SOCl2 (2.9 mL, 40 mmol) was added dropwise. Stirring
was continued until effervescence ceased, and the mixture
was poured into a 2 N NaOH solution. Extraction with
CH2Cl2, drying (Na2SO4), and concentration under reduced
pressure, gave the oxazoline 10 as a yellow oil (2.1 g, 85%
yield). 1H NMR (300 MHz, CDCl3): d 1.22 (d, 3H,
J¼6.8 Hz), 1.29 (d, 3H, J¼6.8 Hz) 2.15 (octet, 1H,
J¼6.8 Hz), 4.13 (s, 3H), 4.25 (s, 3H), 4.30–4.41 (m, 2H),
4.60–4.75 (m, 1H), 7.25–7.36 (m, 2H), 7.65 (m, 1H). 13C
NMR (75.5 MHz, CDCl3): d 16.8, 17.2, 31.8, 54.5, 59.8,
68.5, 71.2, 112.5, 120.4, 121.9, 122.9, 147.2, 151.9, 160.7.
˚
(60 A). TLCs of compounds were visualized by UV light
(254 nm), or by using potassium permanganate or vanillin
stain.
All nonaqueous reactions were conducted under an argon
atmosphere in a flame dried apparatus. Reaction tempera-
tures are reported as the temperature of the bath surrounding
the vessel. Concentrations were performed under reduced
pressure with a rotary evaporator.
4.2.1. 2-Bromo-3-methoxybenzaldehyde (11)
(R15CHO). n-BuLi (31 mmol, 1.6 N solution in hexanes)
Solvents were dried according to established protocols by

4468
A. I. Meyers et al. / Tetrahedron 60 (2004) 4459–4473
was added to a solution of N,N0, N0-trimethylenediamine11
(4.2 mL, 33 mmol) in toluene (60 mL) at 0 8C. After 15 min
at room temperature, m-anisaldehyde (3.7 mL, 30 mmol)
was added. The mixture was stirred at room temperature for
15 min and phenyllithium (50 mL of a 1.8 M solution in
cyclohexane/ether, (Aldrich) 90 mmol) was added while
maintaining room temperature. The solution was stirred at
room temperature for 8 h. While cooling the mixture to
278 8C, THF (60 mL) was added. 1,2-Dibromotetrafluoro-
ethane (22 mL, 180 mmol) was added dropwise at 278 8C.
The mixture was allowed to warm to room temperature and
poured into a cold 10% HCl solution (150 mL) and
extracted with toluene (2£100 mL). The combined organic
layers were dried over MgSO4. Concentration of the organic
layer under reduced pressure, gave a yellow oil, which was
purified by column chromatography to afford 11 as a
(d, J¼8.1 Hz, 1H), 7.18–7.43 (m, 4H). 13C NMR
(75.5 MHz, CDCl3): d 25.85, 25.47, 18.18, 18.31, 18.54,
25.91, 32.69, 55.76, 62.74, 70.03, 72.49, 108.93, 112.72,
117.56, 121.82, 122.93, 125.15, 127.96, 128.36, 130.32,
141.55, 156.45, 156.82, 163.68.
4.3.2. Diester amide (S,S)-15. A solution of biphenyl
oxazoline (S,S)-12h (1.77 g, 3.77 mmol) in THF (40 mL)
was treated with Na2SO4 (27 g, 0.19 mol), water (3.4 mL,
19 mmol) and trifluoroacetic acid (1.46 mL, 19 mmol). The
suspension was stirred at room temperature for 18 h.
Anhydrous Na2SO4 (7.5 g) was then added. The solids
were filtered and washed with THF (30 mL). The filtrate
was concentrated in vacuo at a bath temperature not to
exceed 25 8C. The remaining ammonium salt was dissolved
in CH2Cl2 (70 mL) and cooled to 0 8C. Acetic anhydride
(12.9 mL, 0.14 mmol) and pyridine (20.2 mL, 0.25 mol)
were added, and the solution was allowed to warm to room
temperature. The mixture was washed with cold 3 N HCl
(3£100 mL), followed by a saturated aqueous NaHCO3
solution (100 mL). After drying over Na2SO4, the solvent
was removed in vacuo, and the residue was purified by
radial chromatography (SiO2; hexanes/EtOAc 9:1 to 1:1) to
afford the diester amide 15 as a colorless oil (1.52 g, 88%).
1H NMR (300 MHz, CDCl3): d 0.72 (d, J¼6.8 Hz, 3H), 0.77
(d, J¼6.8 Hz, 3H), 0.99 (m, J¼6.8 Hz, 1H), 1.88 (s, 3H),
1.92 (s, 3H), 3.68 (s, 3H), 3.61–3.71 (m, 1H), 3.71 (s, 3H),
3.92 (dd, J¼3.0 Hz, J¼11.7 Hz, 1H), 4.22 (dd, J¼3.0 Hz,
J¼11.7 Hz, 1H), 4.73 (dd, J¼12.7 Hz, 2H), 5.19 (d,
J¼9.3 Hz, 1H), 6.91 (d, J¼8.2 Hz, 1H), 7.04–7.11 (m,
2H), 7.30–7.42 (m, 3H), 7.64 (d, J¼7.8 Hz, 1H). 13C NMR
(75.5 MHz, CDCl3): d 19.20, 19.53, 20.73, 23.09, 27.77,
53.04, 55.60, 56.05, 64.45, 65.55, 110.46, 114.60, 119.96,
123.21, 124.54, 126.19, 128.24, 128.00, 132.42, 132.62,
156.83, 156.89, 167.80, 169.60, 170.63.
1
crystalline compound (6.2 g, 96%). Mp: 67.5–68.5 8C. H
NMR (300 MHz, CDCl3): d 3.95 (s, 3H), 7.13 (dd, 1H), 7.38
(m, 1H), 7.53 (dd, 1H), 10.4 (s, 1H). 13C NMR (75.5 MHz,
CDCl3): d 56.91, 117.16, 117.38, 121.67, 128.58, 135.04,
156.52, 192.52.
4.3. General procedure biphenyl 12a–12j (Table 1)
A mixture of aryl bromide 11a–11j11 (1 mmol) and Mg
turnings (2 mmol) in THF (5 mL) was heated at reflux and
treated dropwise with a solution of 1,2-dibromoethane
(1.3 mmol) in THF (1 mL). A vigorous reaction was
observed. The mixture was heated for 1–2 h (the formation
of the Grignard reagent was followed by GC) and cooled to
room temperature. A solution of oxazoline 10a (R2¼I-Pr
0.6 mmol) in THF (4 mL) was added. The reaction mixture
was heated at reflux for 5–18 h and quenched with saturated
aqueous NH4Cl solution (10 mL). The mixture was
extracted with ether (2£10 mL). The combined organic
layers were dried over Na2SO4 and concentrated under
reduced pressure to afford biphenyl 12a–12j as a mixture of
diastereomers. The crude biphenyl was purified by radial
chromatography (SiO2; hexanes/EtOAc). A typical example
is described below.
4.3.3. (S)-2,20-(Dihydroxymethyl)-6,60-dimethoxy-1,10-
biphenyl (16). A solution of diester amide (S,S)-15
(1.56 g, 3.41 mmol) in THF (60 mL) was cooled to 0 8C,
and a solution of LiAlH4 in THF (10.2 mL of a 1.0 M
solution) was added. The mixture was allowed to warm to
room temperature and stirred for 2 h. To the mixture was
added was Na2SO4, 10H2O (2.0 g), 1.0 mL of an aqueous
40% NaOH solution, anhydrous Na2SO4 (4.0 g), and the
mixture was stirred for 6 h. The salts were filtered and
washed with THF, and the filtrate was concentrated in vacuo
to afford an oil that crystallized on standing. The product
was purified by radial chromatography (SiO2; hexanes/
EtOAc 9:1) to yield 16 as a colorless crystalline compound
(0.82 g, 88%). [a]D¼296.5 (c¼1.00, CHCl3). Mp 136–
136.5 8C. 1H NMR (300 MHz, CDCl3): d 2.64 (s, 2H), 3.67
(s, 6H), 4.19 (dd, J¼11.7 Hz, 2H), 6.93 (dd, J¼8.2 Hz, 2H),
7.13 (d, J¼7.5 Hz), 7.37 (t, J¼7.9 Hz, 2H). 13C NMR
(75.5 MHz, CDCl3): d 55.87, 63.27, 110.63, 121.79, 124.23,
129.12, 140.91, 156.63.
4.3.1. Biphenyl (S,S)-12h. A mixture of aryl bromide 11h
(3.0 g, 9.1 mmol) and Mg turnings (0.44 g, 18 mmol) in
THF (25 mL) was heated at reflux and treated dropwise with
a solution of 1,2-dibromoethane (1.0 mL, 11.8 mmol) in
THF (4 mL) over 30 min. A vigorous reaction was
observed. The mixture was heated at reflux for 1 h after
which all Mg had reacted. The solution was cooled to room
temperature, and a solution of oxazoline 10a (R¼i-Pr,
1.37 g, 5.5 mmol) in THF (10 mL) was added. The reaction
mixture was stirred overnight at room temperature, poured
into an aqueous saturated NH4Cl solution (40 mL) and
extracted with ether (2£50 mL). The combined organic
layers were dried over Na2SO4 and concentrated in vacuo to
afford a yellow oil. The crude product, consisting of a
mixture of diastereomers (93:7) was purified by radial
chromatography to remove a trace of the minor isomer
(SiO2; hexanes/EtOAc 9:1 to 4:1). Biphenyl (S,S)-12h was
isolated as a colorless oil (1.89 g, 73%). 1H NMR
(300 MHz, CDCl3): d 20.05 (s, 6H), 0.71 (d, J¼6.7 Hz,
3H), 0.75 (d, J¼6.7 Hz, 3H), 0.86 (s, 9H), 1.50 (septet,
J¼6.7 Hz, 1H), 3.62 (s, 3H), 3.69 (s, 3H), 3.61–3.99 (m,
3H), 4.36 (dd, J¼14 Hz, 2H), 6.77 (d, J¼8.0 Hz, 1H), 7.02
4.3.4. (S)-2,20-(Dimethyl)-6,60-dimethoxy-1,10-biphenyl
(17).
A
suspension of biscarbinol (S)-16 (0.18 g,
0.65 mmol), Pd(OH)2 (20 mg) and trifluoro-acetic acid (5
drops) in EtOH (abs. 10 mL) was stirred under an
atmosphere of hydrogen (balloon). After stirring for 36 h,
the catalyst was removed by filtration, and the solvent was
removed in vacuo. The crude product was purified using

A. I. Meyers et al. / Tetrahedron 60 (2004) 4459–4473
4469
radial chromatography (SiO2; hexanes/EtOAc 6:1) to yield
17 as a colorless solid in quantitative yield (0.157 g).
[a]D¼7253.0 (c¼1.00, CHCl3). Mp 85.9–86.3 8C. 1H
NMR (300 MHz, CDCl3): d 1.95 (s, 6H), 3.70 (s, 6H),
6.82 (d, J¼8.2 Hz, 2H), 6.91 (d, J¼7.5 Hz, 2H), 7.24 (m,
2H). 13C NMR (75.5 MHz, CDCl3): d 19.56, 55.77, 108.31,
122.19, 126.19, 127.87, 138.18, 156.93.
CDCl3): d 3.70 (s, 6H), 5.00 (s, 2H), 6.55 (d, J¼8.0 Hz, 2H),
6.55 (d, J¼7.9 Hz, 2H), 7.24 (t, J¼8.1 Hz, 2H). 13C NMR
(75.5 MHz, CDCl3): d 55.99, 103.46, 106.97, 109.14,
130.51, 155.21, 157.95.
4.4. 2-Bromo-1-[(tert-butyldimethylsiloxy)methyl]-3,4,5-
trimethoxybenzene (22) (According to literature
procedure)8
4.3.5. 2,20-(Dihydroxyl)-6,60-dimethyl-1,10-biphenyl (18).
A
solution of dimethoxybiphenyl (S)-17 (0.245 g,
Alternatively, the silylating system of TBDMSOTf/Et3N
was substituted with TBDMSCl/Et3N/4-DMAP to afford
silyl ether 22 in a similar fashion. For a more convenient and
efficient reduction of acid 21, the following was employed.
1.01 mmol) in CH2Cl2 (4–6 mL) was cooled to 278 8C,
and BBr3 (0.57 g, 2.26 mmol) was added. The mixture was
allowed to warm to room temperature and stirred for 1 h.
The mixture was carefully poured into ice water, stirred for
30 min, saturated with NaCl and extracted with CH2Cl2.
Drying (Na2SO4) and evaporation of the solvent in vacuo
yielded the bisphenol 18 as a colorless crystalline compound
(0.206 g, 95%. [a]D¼260.5 (c¼1.00, CHCl3). Mp 159–
160 8C (lit. 159–160 8C).18 1H NMR (300 MHz, CDCL3): d
1.99 (s, 6H), 4.66 (s, 2H), 6.86–6.92 (m, 4H), 7.21–7.27
(m, 2H). 13C NMR (75.5 MHz, CDCl3): d 19.49, 113.16,
119.49, 122.62, 130.15, 138.94, 154.81.
4.4.1. 2-Bromo-3,4,5-trimethoxybenzyl alcohol. To a
solution of 3.00 g (12.9 mmol, 1.25 equiv.) or zirconium
tetrachloride in 46 mL of anhydrous THF at 25 8C under Ar
was added 1.95 g (51.5 mmol, 5.0 equiv.) of sodium
borohydride with immediate evolution of gas. After
10 min, to this cream-colored mixture was added a solution
of 3.00 g (10.3 mmol) of acid 21 in 10 mL of anhydrous
THF via cannula. (The flask was rinsed with 4.0 mL of THF,
and this was also added via cannula.) After stirring for 12 h,
the reaction was cooled to 0 8C and slowly quenched with
25 mL of H2O. The mixture was then extracted with CH2Cl2
(1£250 mL, 1£200 mL), and the combined organic portions
were dried over MgSO4. Filtration and removal of the
solvent afforded 2.86 g (10.3 mmol, 100%) of an off
colorless solid that was homogeneous by GC and whose
physical characteristics were identical with the previously
reported data.8
4.3.6. (S)-2,20-(Diformyl)-6,60-dimethoxy-1,10-biphenyl
(19). Chromium trioxide (0.54 g, 5.4 mmol) was added to
a solution of pyridine (0.85 g, 10.8 mmol) in CH2Cl2
(15 mL). The solution was stirred for 15 min, and a solution
of biscarbinol (S)-16 in CH2Cl2 (10 mL) was added. A black
deposit formed. After stirring for 15 min, the solution was
decanted from the deposit that was washed twice with
CH2Cl2. The combined organic layers were washed with 5%
NaOH solution (3£100 mL), 5% HCl solution (3£100 mL),
5% NaHCO3 solution (100 mL), and brine (100 mL).
Drying (Na2SO4) and evaporation of the solvent in vacuo
yielded the crude product as a slightly yellow oil, which was
purified by radial chromatography (SiO2; hexanes/EtOAc,
9:1) to give dialdehyde 19 as a colorless oil in quantitative
yield (0.97 g). [a]D¼2291 (c¼1.30, CHCl3). 1H NMR
(300 MHz, CDCl3): d 3.71 (s, 6H), 7.18–7.66 (m, 6H), 9.64
(s, 2H). 13C NMR (75.5 MHz, CDCl3): d 56.59, 115.85,
119.71, 124.54, 129.81, 136.14, 143.32, 191.64.
4.4.2. Phenyl oxazoline (S)-24. To a solution of 7.05 g
(29.1 mmol) of acid 238 in 70 mL of anhydrous CH2Cl2 was
added 5.2 mL (59.6 mmol) of oxalyl chloride and then 5
drops of DMF. The reaction mixture was stirred overnight
under Ar at 25 8C. The CH2Cl2, DMF, and excess oxalyl
chloride were removed in vacuo to afford the acid chloride
as a light yellow solid. This was then dissolved in 30 mL of
anhydrous CH2Cl2 and added over 15 min to a solution of
4.00 g (38.8 mmol) of S-valinol and 9.0 mL of Et3N in
50 mL anhydrous CH2Cl2 cooled to 0 8C. The mixture was
then stirred at room temperature for 6 h at which time the
solvent was evaporated, and the yellow, oily residue was
dissolved in 50 mL of ethyl acetate. The precipitated
trimethylamine hydrochloride was filtered away, and
concentration of the filtrate gave the amide as a yellow
oil. This was dissolved in 120 mL of anhydrous CH2Cl2, and
11 mL of thionyl chloride were added. After stirring at
25 8C, the dark mixture was cooled to 0 8C and carefully
quenched with 40 mL of H2O and then 40 mL of 4 aq.
NaOH. After separation, the aqueous layer was extracted
with CH2Cl2 (2£150 mL), the organic portions combined,
dried over MgSO4, filtered, and the solvent removed. On
occasion, this was contaminated with the aliphatic chloride;
therefore, to induce ring closure, this mixture was refluxed
with K2CO3 in (10:1, CH3CN–H2O) (monitored by TLC),
cooled, the CH3CN removed, and the aqueous portion
extracted with CH2Cl2. This additional process did not
affect the overall yield of oxazoline 24. Purification by flash
chromatography (SiO2, 33% ethyl acetate/hexane) afforded
6.56 g (21.2 mmol, 73%) of oxazoline 24 as a light brown
oil: [a]2D6¼231.5 (c¼1.06, THF). 1H NMR (300 MHz,
4.3.7. 2,20-(Dihydroxy)-6,60-dimethoxy-1,10-biphenyl
(20). To a solution of dialdehyde 19 (0.45 g, 1.67 mmol)
in CH2Cl2 (25 mL) was added NaHCO3 (2.2 g, 26.6 mmol)
and m-chloroperbenzoic acid (80%, 2.3 g, 13.3 mmol). The
mixture was stirred at room temperature for 3 days.
Aqueous 10% Na2S2O3 solution (30 mL) was added, and
stirring was continued for another 30 min. The solution was
diluted with CH2Cl2 and saturated with NaCl. The layers
were separated, and the aqueous layer was extracted with
CH2Cl2. Drying (Na2SO4) and evaporation of the solvent in
vacuo gave a yellow oil. 1H NMR indicated formation of the
desired formate. To the mixture was added a suspension of
K2CO3 (6.2 g) in MeOH (27 mL) and water (13 mL) that
had been degassed in an ultrasonic bath for 1 h. The mixture
was stirred for 1 h, quenched with 1 N HCl, and extracted
with CH2Cl2. Drying (Na2SO4) and evaporation of the
solvent in vacuo yielded a yellow oil. TLC indicated the
presence of other compounds that were not identified.
The crude oil was purified by radial chromatography (SiO2;
hexanes/EtOAc, 9:1 to 5:1) to give bisphenol 20 (60 mg,
15%). [a]D¼2144 (c¼0.77, CHCl3).8 1H NMR (300 MHz,

4470
A. I. Meyers et al. / Tetrahedron 60 (2004) 4459–4473
CDCl3): d 0.92 (d, J¼9.1 Hz, 3H), 1.01 (d, J¼9.1 Hz, 3H),
1.82–1.89 (m, 1H), 3.82 (s, 3H), 3.83 (s, 3H), 3.89 (s, 3H),
3.90 (s, 3H), 4.07–4.12 (m, 2H), 4.32–4.40 (m, 1H), 7.03
(s, 1H). 13C NMR (75.5 MHz, CDCl3): d 18.01, 18.77,
32.70, 56.11, 61.11, 61.24, 61.65, 69.76, 72.44, 105.36,
107.85, 116.85, 147.53, 147.64, 149.03, 161.75; IR (thin
film) 2958, 2938, 2872, 1651, 1493, 1464, 1409, 1231,
1213, 1170, 1119, 1069, 1027, 1009 cm21. Anal. Calcd for
C16H23O5N: C, 62.12; H, 7.49; N, 4.53. Found: C, 62.45; H,
7.44; N, 4.13.
and the solvent removed in vacuo. The crude acylamide was
dissolved in 1 L THF and cooled to 0 8C and 19 g of LiAlH4
was added in portions and then stirred at 0 8C for 4 h and
then at 25 8C for 14 h. The mixture was again cooled to 0 8C
and slowly quenched with 3 N aq. NaOH, dried with 750 g
of Na2SO4, filtered, and the solvent removed. The residue
was dissolved in 350 mL of CH2CL2 and washed with N
HCl (2£100 mL), the organic portions combined, dried over
MgSO4, and the solvent evaporated. This was purified by
flash column chromatography (SiO2, 10–70% EtOAc/
hexane) to afford 20.8 g (52.8 mmol, 70%) of biscarbinol
(S)-26 as a colorless solid. The physical characteristics were
identical with the previously reported data.8
4.4.3. Biaryl oxazoline (S,S)-25. To a solution of 7.31 g
(18.7 mmol) of aryl bromide 228 in 40 mL of anhydrous
THF was added 0.91 g (37.4 mmol) of Mg turnings, and
then the reaction was brought to reflux under Ar. A solution
of 3.52 g (18.7 mmol) of 1,2-dibromoethane in 5 mL of
THF was added in portions over 1 h. Refluxing was
continued until there was complete disappearance of
bromide 22 as monitored by GC (ca. 2 h). To this refluxing
solution, a solution of 2.70 g (9.3 mmol) of oxazoline 24 in
20 mL of THF was added, and reflux was maintained until
oxazoline 24 was completely consumed, as monitored by
GC analysis (ca. 5–15 h). The reaction was cooled to room
temperature, quenched with 30 mL of sat. NH4Cl, separ-
ated, the organic portion was washed with brine (2£30 mL),
dried over MgSO4, and concentrated to obtain a crude
brown oil. An aliquot of this oil was quickly passed through
a plug of silica gel to remove material other than the two
coupled biaryl diastereomers. The diastereomeric ratio, as
determined by analytical HPLC (3% i-propanol/hexane,
flow rate 2 mL/min), was determined to be 98:2 (retention
times¼5.24 and 4.08 min, respectively). The crude oil was
purified by flash chromatography (SiO2, 20–70% EtOAc) to
afford 4.73 g (8.0 mmol, 86%) of biaryl (S)-25 as analyti-
cally homogeneous light yellow viscous oil, which could be
crystallized from Et2O/hexane to give a colorless solid:
4.4.5. (S)-Dialdehyde (27). To a solution of 0.40 mL
(4.40 mmol) of oxalyl chloride in 10 mL of anhydrous
CH2Cl2, cooled to 255 8C under Ar, was added a solution of
0.65 mL (8.40 mmol) of DMSO in 2.5 mL of anhydrous
CH2Cl2. After stirring for 2 min, a solution of 500 mg
(1.26 mmol) of biscarbinol (S)-26 in 5 mL of CH2Cl2 was
added dropwise. After 25 min of stirring at 255 8C,
2.65 mL (19.1 mmol) of triethylamine was added; stirring
continued at 255 8C for 5 min, and then the reaction
mixture was warmed to 25 8C. The solution was quenched
with 18 mL of H2O and then diluted with 100 mL of
CH2Cl2. This was washed with 50 mL of H2O, the aqueous
layer extracted with CH2Cl2 (2£50 mL), the combined
organic layers were dried over MgSO4, and the solvent
evaporated. Purification by flash chromatography (SiO2,
33% EtOAc/hexane) afforded 470 mg (1.21 mmol) of
dialdehyde (S)-27 as
a
colorless oil: [a]2D6¼281.7
1
(c¼1.17, CHCl3). H NMR (300 MHz, CDCl3): d 3.60 (s,
6H), 3.96 (s, 6H), 3.97 (s, 6H), 7.38 (s, 2H), 9.57 (s, 2H). 13C
NMR (75.5 MHz, CDCl3): d 56.14, 60.69, 61.07, 105.69,
124.41, 130.47, 147.27, 151.60, 153.88, 190.13; IR (thin
film) 2941, 1689, 1587. Anal. Calcd for C20H22O8: C, 61.53;
H, 5.68. Found: C, 61.31; H, 5.73.
1
[a]2D6¼215.9 (c¼2.52, CH2Cl2). Mp 83–85 8C. H NMR
(300 MHz, CDCl3): d 20.04 (s, 3H), 20.03 (s, 3H), 0.74 (d,
J¼6.7 Hz, 3H), 0.78 (d, J¼6.7 Hz, 3H), 0.88 (s, 9H), 1.53–
1.67 (m, 1H), 2.61 (s, 3H), 3.65 (s, 3H), 2.66–3.80 (m, 2H),
3.82 (s, 3H), 3.88 (s, 3H), 3.89 (s, 3H), 3.92 (s, 3H), 4.03
(dd, J¼8.0, 9.4 Hz, 1H), 4.27 (AB quartet, JAB¼14 Hz, 1H),
4.37 (d, J¼14 Hz, 1H), 6.94 (s, 1H), 7.18 (s, 1H). 13C NMR
(75.5 MHz, CDCl3): d 25.37, 18.23, 18.29, 18.75, 25.90,
29.67, 32.78, 55.73, 56.00, 60.48, 60.63, 62.67, 170.19,
72.49, 76.58, 77.00, 77.43, 104.22, 108.63, 120.27, 123.44,
124.30, 136.09, 139.90, 144.20, 150.92, 151.54, 152.43,
152.54, 163.56. Anal. Calcd for C31H45O8NSi: C, 63.13; H,
8.03; N, 2.37. Found: C, 62.97; H, 7.98; N, 2.33.
4.4.6. (S)-2,20-Dihydroxy-4,5,6,40,50,60-hexamethoxy-
biphenyl (28). A round-bottomed flask was charged with
0.365 g (0.94 mmol) of dialdehyde 27, 0.472 g (5.62 mmol)
of NaHCO3, 0.607 g (2.81 mmol) of m-chloroperbenzoic
acid (ca. 80%), and 50 mL of dry CH2Cl2. After stirring at
room temperature for 7 h, the reaction was quenched with
30 mL of 10% sodium thiosulfate, and stirring was
continued for 30 min. The reaction mixture was diluted
with 100 mL of CH2Cl2 and then washed with brine
(2£50 mL). The aqueous layer was re-extracted with
CH2Cl2 (2£70 mL). The combined organic layers were
dried over MgSO4, and the solvent was removed in vacuo to
obtain the bisformate ester as a colorless oil. In another
flask, potassium carbonate (3.5 g) was dissolved in a
mixture of 15 mL MeOH and 7 mL H2O. The solution
was degassed in an ultrasonic bath for 1 h. This was then
added to the crude bisformate, and the mixture was stirred
for 10 min, after which 30 mL of 1 N aq. HCl was added and
the mixture extracted with CH2Cl2 (3£80 mL). After drying
over MgSO4, the solvent was removed, and the crude
product purified by silica gel chromatography (50% EtOAc/
hexane) afforded 0.278 g (0.76 mmol, 83%) of diol (S)-28 as
a colorless solid (Mp 155–156 8C. [a]2D6¼248.9 (c¼2.27,
4.4.4. (S)-Biscarbinol (26). A 3-L round-bottomed flask
was charged with 44.49 g (75.4 mmol) of biaryl 25, 1.2 L of
THF, 40 mL of TFA, 101 mL of H2O, and 736 g of Na2SO4.
This was stirred with a mechanical overhead stirrer for
13.5 h. An additional 250 g of Na2SO4 was added and then
filtered. The sulfate was rinsed with THF (3£250 mL) and
the solvent removed in vacuo, affording a yellow oil. This
oil was dissolved in 400 mL of CH2Cl2, and to this was
added 120 mL of pyridine and immediately placed in an ice
bath. To this, 100 mL of Ac2O was added and the mixture
stirred for 5.5 h, diluted with 350 mL of CH2Cl2 washed
with cold 1 N HCl (1£100 mL, 2£200 mL), washed with
sat. aq. NaHCO3 (1£300 mL), dried over MgSO4, filtered,
1
CHCl3). H NMR (300 MHz, CDCl3): d 3.67 (s, 6H), 3.80
(s, 6H), 3.82 (s, 6H), 5.40 (s, 2OH), 6.45 (s, 2H). 13C NMR

A. I. Meyers et al. / Tetrahedron 60 (2004) 4459–4473
4471
(75.5 MHz, CDCl3): d 55.80, 61.17, 96.94, 105.10, 136.21,
150.61, 151.64, 154.34; IR (thin film) 3600–3000, 2940,
1604; Anal. Calcd for C18H22O8: C, 59.01; H, 5.98. Found:
C, 59.01; H, 5.85.
Fraction 2:2.28 g (30.4%) diastereomeric ratio (94.5:5.5).
The minor diastereomer (SSR)-32 was also isolated (0.32 g,
4.3%). The total yield of 32 was 62.7% and, if combined,
had a ratio of diastereomers of approximately 95:5. Further
characterization of 32 is not reported since the biscarbinol
16 derived from it was identical to 16 obtained in Scheme 6.
4.4.7. 2-Bromo-3-methoxy benzoic acid (30). A round-
bottomed flask was charged with 8.38 g of 2-bromo-3-
methoxybenzaldehyde 11 (R¼CHO) (38.9 mmol) and
20.4 g of KMnO4 (0.13 mol) in 65 mL of acetone and
5 mL of H2O and equipped with a reflux condenser. The
mixture was heated at 55 8C overnight (during this time, the
reaction became exothermic and briefly refluxed). An
additional 10 mL of H2O and 11.2 g of KMnO4 was added
and heated at 55 8C overnight. After cooling, the mixture
was filtered through a plug of silica gel with a solution of
MeOH/EtOAc/CH2Cl2 and acetone. The solvent was
removed in vacuo, and the crude acid was used without
further purification.
4.6. Amine (S)-35 from (S)-34 (R15R25OMe)
A round-bottomed flask was charged with 59.1 mg
(0.157 mmol) of the amine (S)-3438 and 5.0 mL of ethyl
formate. This solution was then heated to reflux for 3 h,
cooled, and the solvent removed in vacuo. The residue was
dissolved in 3 mL of THF followed by the addition of
0.460 mL of 1 M LiAlH4 (in THF). After stirring overnight
at room temperature, the mixture was quenched with a few
drops of 4 N NaOH, diluted with THF and dried over
Na2SO4. Purification by radial chromatography (1 mm
SiO2, 20% EtOAc/hexane to pure EtOAc to 50% EtOAc/
MeOH) afforded 42.7 mg (0.11 mmol, 69%) of the
N-methyl amine (S)-35 as a yellow oil. The physical data
for this compound is reported in the following experimental
procedure.
4.4.8. 2-Bromo-3-methoxyphenyl oxazoline (31). To the
crude acid above was added 100 mL of CH2Cl2, 20 mL of
oxalyl chloride and 4 drops of DMF and stirred under Ar at
room temperature overnight. The solvent was removed in
vacuo and then added to 200 mL of CH2Cl2. This mixture
was added to a solution of 5.02 g of (S)-tert-leucinol
(Degussa) and 27 mL of Et3N in 50 mL of CH2Cl2 and
stirred overnight at room temperature. To this, a solution of
8.5 mL of SOCl2 in 20 mL of CH2Cl2 was added, and the
reaction was stirred for 30 min, cooled in an ice bath,
quenched with H2O and made basic with 4 N NaOH. The
two-phase system was separated, the organic portion was
washed with H2O, dried (MgSO4), and the solvent removed.
The residue was dissolved in 100 mL of toluene and 10 mL
of Et3N and heated at reflux overnight. The solvent was
removed in vacuo and filtered through silica gel with EtOAc
to give 7.86 g of the oxazoline 31 (32.3 mmol, 83%) as a
light yellow oil that solidified upon standing. Mp 76.5–
78.5 8C. This material was used without further purification.
4.6.1. Amine (S)-35 from dialdehyde (S)-27
(R15R25OMe). A flame-dried round-bottomed flask was
charged with 202 mg (0.514 mmol) of dialdehyde (S)-27,
287 mg (4.25 mmol, 8.3 equiv.) of methylamine hydro-
˚
chloride, 5 mL of MeOH, and 3 A molecular sieves. After
stirring for 4.5 h at room temperature, the yellow reaction
mixture was cooled to 0 8C, and 213 mg (3.34 mmol) of
NaBH3CN in 2.0 mL MeOH was added over 2 min. The
cream-colored mixture was stirred at room temperature for
16 h, and then the solvent was removed in vacuo. The
residue was acidified with 20 mL of 0.3 N aq. HCl, stirred
for 1 h and then made basic with NaOH pellets. This was
extracted with CH2Cl2 (4£75 mL), concentrated, passed
through a plug of silica gel with EtOAc/MeOH, and the
solvent removed leaving a yellow foam. The product, 35,
was purified by radial chromatography (2 mm SiO2, EtOAc
to 50% MeOH/EtOAc) to give 198.9 mg (0.511 mmol,
99%) of the N-methyl amine (S)-35 (R1¼R2¼OMe) as a
4.5. (S,S,S)-Dimethoxybiphenyl oxazoline (32)
In a round-bottomed flask, the bromo oxazoline 31 (7.86 g,
32.3 mmol) was azeotropically dried twice with benzene
and then held under high vacuum overnight. This was
dissolved in 10 mL of dry DMF, and then 4.7 g of freshly
activated Cu powder was added. The flask was equipped
with a reflux condenser under Ar and then placed in a
preheated (105 8C) sand bath for 2 h. The reaction was then
refluxed overnight. An additional 15 mL of DMF was
added, and the mixture was refluxed an additional 3 days.
After cooling, the mixture was diluted with CH2Cl2 and then
washed with NH4OH until all the copper was removed,
washed with H2O, dried over MgSO4, filtered, and the
solvent removed in vacuo. Purification by silica gel
chromatography (hexane to 1:1 EtOAc/hexane) afforded
two separate fractions of the bis(oxazoline) 32. The
diastereomeric ratios of each were determined by direct
conversion to the biscarbinol (S)-16 (described above)
followed by bis(Mosher ester) formation and integration of
the appropriate resonances. The biscarbinol (S)-16 was
obtained in 2 fractions.
1
yellow oil/foam [a]D¼260.9 (c¼1.28, CHCl3). H NMR
(300 MHz, CDCl3): d 2.32 (s, 3H), 3.05 (AB quarlet,
JAB¼12.3 Hz, 2H), 3.32 (AB quarlet, JAB¼12.3 Hz, 2H),
3.67 (s, 6H), 3.87 (s, 6H), 3.88 (s, 6H), 6.62 (s, 2H). 13C
NMR (75.5 MHz, CDCl3): d 43.09, 56.07, 57.24, 60.70,
61.02, 108.10, 122.70, 130.36, 141.64, 151.41, 152.77).
4.6.2. Amine (S)-35 from (S)-19 (R15R25H). A solution
of dialdehyde (S)-19 (86 mg, 0.32 mmol) and methylamine
hydrochloride (172 mg, 2.55 mmol) in MeOH (3 mL) was
˚
stirred at room temperature for 5 h in the presence of 3 A
sieves. The solution was cooled to 0 8C, and a solution of
NaCNBH3 (130 mg, 2.07 mmol) in MeOH (1 mL) was
added. The resulting mixture was stirred at room tempera-
ture for 15 h. The solvent was removed in vacuo, and the
residue was acidified with 10 mL of 0.3 N HCl. After
stirring for 1 h, NaOH (0.11 g, 2.75 mmol) was added.
Extraction with CH2Cl2 (3£10 mL), drying (Na2SO4)
followed by evaporation of the solvent afforded N-methyl
amine (S)-35 (R1¼R2¼H) as a slightly yellow oil.
Purification using radial chromatography (SiO2; EtOAc to
Fraction 1: 2.35 g (31.3%), diastereomeric ratio (100: 0).

4472
A. I. Meyers et al. / Tetrahedron 60 (2004) 4459–4473
subsequently found that the silylated derivative 12h could also
be separated, and this was, therefore, carried forward to (S)-15.
16. The diastereotopic benzylic protons in the Mosher esters of 16
provided an excellent probe for determining the enantiomeric
purity when examined at 300 MHz.
EtOAc/MeOH 1:1) yielded the product as a colorless oil
(77 mg, 90%). [a]D¼þ70.9 (c¼0.57, CHCl3). 1H NMR
(300 MHz, CDCl3): d 2.62 (s, 3H), 3.41 (d, J¼12.2 Hz, 2H),
3.75 (d, J¼12.2 Hz), 4.09 (s, 6H), 7.21–7.27 (m, 4H), 7.56–
7.61 (m, 2H). 13C NMR (75.5 MHZ, CDCl3): d 42.84,
55.70, 56.73, 110.78, 121.78, 125.18, 128.60, 135.61,
161.60.
17. Ratcliffe, R.; Rodehorst, R. J. Org. Chem. 1970, 35, 4000.
18. Kanoh, S.; Tamura, N.; Motoi, M.; Suda, H. Bull. Chem. Soc.
Jpn 1987, 60, 2307.
19. Meyers, A. I.; Flisak, J. R.; Aitken, R. A. J. Am. Chem. Soc.
1987, 109, 5446.
Acknowledgements
20. Reported in preliminary form, see Ref. 13 above.
21. Ullmann, F.; Bielecki, J. Ber. 1901, 34, 2174. For recent
reviews see (a) Hassan, J.; Sevigoan, M.; Gazzi, C.; Schulz, E.;
Lemaire, M. Chem. Rev. 2002, 102, 1359. (b) Nelson, T. D.;
Crouch, R. D. Org. React. 2004, 63, 265–555.
The authors are grateful to the National Institutes of Health
for financial support. A DFG fellowship (to A.M.) is also
gratefully acknowledged.
22. (a) Adams, R.; Yuan, H. C. Chem. Rev. 1933, 12, 261; (b) A
recent report describes stereochemical stability in biphenyls
containing only two ortho substituents, but both capable of
metal coordination by chiral oxazolines to Cu(I). See Ref. 36
below.
References and notes
23. Nelson, T. D.; Meyers, A. I. Tetrahedron Lett. 1993, 34, 3061.
24. Miyano, S.; Tobita, M.; Suzuki, S.; Nishikawa, Y.; Hashimoto,
H. Chem. Lett. 1980, 1027.
1. Meyers, A. I.; Knaus, G.; Kamata, K. J. Am. Chem. Soc. 1974,
96, 268.
2. Meyers, A. I.; Lutomski, K. A. Asymmetric synthesis via
chiral oxazolines. In Asymmetric synthesis III; Morrison, J. D.,
Ed.; Academic: London, 1984.
25. Migration of aryl groups in aldehydes is only seldom seen. In
some rare cases, aromatic aldehydes will rearrange to phenyl
formates. (a) Godfrey, I. M.; Sargent, M. V.; Elix, J. A.
J. Chem. Soc., Perkin Trans. 1 1974, 1353. (b) Osata, Y.;
Sawaki, Y. J. Org. Chem. 1969, 34, 3985.
3. (a) Reuman, M.; Meyers, A. I. Tetrahedron Rep. 1985, 41,
837. (b) Gant, T. G.; Meyers, A. I. Tetrahedron Rep. 1994, 50,
2297.
26. For some representative examples, see; (a) Lin, G. Q.; Zhong,
M. Tetrahedron Lett. 1996, 37, 3015. (b) Andrus, M. B.;
Asgari, D.; Sclafani, J. A. J. Org. Chem. 1997, 62, 9365.
(c) Solladie, G.; Hugele, P.; Bartsch, R. J. Org. Chem. 1998,
63, 3895.
4. (a) Bolm, C. Angew. Chem., Int. Ed. 1991, 30, 542. (b) Doyle,
M. P. In Catalytic asymmetric synthesis; Ojima, I., Ed.; VCH:
New York, 1993; p 103. (c) Pfaltz, A. Acc. Chem. Res. 1993,
26, 339. (d) Togni, A.; Venanzi, L. M. Angew Chem., Int. Ed.
1994, 33, 542. (e) Braunstein, P.; Naud, F. Angew. Chem., Int.
Ed. 2001, 40, 680. (f) Ghosh, A. K.; Mathivanan, P.;
Cappiello, J. Tetrahedron: Asymmetry 1998, 9, 1.
5. Meyers, A. I.; Mihelich, E. D. J. Am. Chem. Soc. 1975, 97,
7383.
27. Nelson, T. D.; Meyers, A. I. J. Org. Chem. 1994, 59, 2655.
28. Fuson, R. C.; Cleveland, E. A. Org. Synth. Coll. 1955, 3, 339.
29. Chiracel OD column, hexane/2-propanol (85:15), 1.0 mL/min;
l¼254 nm, Rt, R-29¼13 min; S-29¼18 min.
30. (a) Izumi, Y.; Tai, A. Stereo-differentiating reactions—the
nature of asymmetric reactions; Academic: New York, 1977.
¨
(b) Huerta, F. F.; Minidis, A. B.; Backvall, M.-E. Chem. Soc.
6. Meyers, A. I.; Flanagan, M. E. Org. Syn. 1993, 71, 107.
7. Meyers, A. I.; Himmelsbach, R. J. J. Am. Chem. Soc. 1985,
107, 682.
Rev. 2001, 30, 321. (c) Ward, R. Tetrahedron: Asymmetry
1995, 6, 1475. (d) Ebbers, E. J.; Ariaans, G. J. A.; Houbiers,
J. P. M.; Bruggink, A.; Zwanenburg, B. Tetrahedron 1997, 53,
9417.
8. Warshawsky, A. M.; Meyers, A. I. J. Am. Chem. Soc. 1990,
112, 8090.
9. Dickman, D. A.; Meyers, A. I.; Smith, G. A.; Gawley, R. E.
Org. Syn. Coll. VII 1990, 530. McKennon, M. J.; Meyers, A. I.;
Drauz, K.; Schwarm, M. J. Org. Chem. 1993, 58, 3568.
10. Nelson, T. D.; Meyers, A. I. J. Org. Chem. 1994, 59, 2577.
11. (a) Comins, D. L.; Brown, J. D. J. Org. Chem. 1989, 54, 3730;
(b) The aryl bromides 11 (entries 1–10 in Table 1) were all
prepared from the corresponding benzaldehyde 11 (R1¼CHO)
by well-accepted procedures8 for making acetals, alcohols,
TBDMS, and TIPS derivatives.
31. Ziegler, F. E.; Chliwner, I.; Fowler, K. W.; Kanfer, S. J.; Kuo,
S. J.; Sinha, N. D. J. Am. Chem. Soc. 1980, 102, 790.
32. Xi, M.; Bent, B. E. J. Am. Chem. Soc. 1993, 115, 7426.
33. RS-33 was obtained by the usual copper mediated Ullmann
coupling of 2-bromo-3,5-dimethyl oxazoline and, in this case,
gave only a 3:1 mixture of (SSS)-33 and (SSR)-33. It is
noteworthy that the R (axial configuration only) diastereomer
prevailed slightly under these conditions. This is another
example where the nature of the aromatic substituents affected
coupling stereochemistry. It also may be that Cu-p interaction
is somehow involved in determining stereochemistry.
34. This result was initially reported in 1994: Nelson, T. D.;
Meyers, A. I. Tetrahedron Lett. 1994, 35, 3259.
12. Lai, Y. H. Synthesis 1981, 585.
13. Meyers, A. I.; Meier, A.; Rawson, D. J. Tetrahedron Lett.
1992, 33, 853. For a preliminary report on this study, see:
Moorlag, H.; Meyers, A. I. Tetrahedron Lett. 1993, 6989 and
6993.
14. Rizzacasa, M. A.; Chau, P.; Czuba, I. R.; Bringmann, G.;
Gulden, K. P.; Schaffer, M. J. Org. Chem. 1996, 61, 7101.
15. Desilylation of the 93:7 mixture of diastereomers in 12h was
effected by 2% HF in 10% acetonitrile–water for 1 h at room
temperature. The corresponding hydroxy product was com-
pletely separated from its diastereomer by radial chromato-
graphy on silica (hexane–ethyl acetate). However, it was
35. Wulff, W. D.; Zhang, Y.; Man-Sui, Y.; Hongqiao, W.; Heller,
D. P.; Wu, C. Org. Lett. 2003, 5, 1813.
36. Andrus, M. B.; Asgari, D. Tetrahedron 2000, 56, 5775.
37. Imai, Y.; Zhang, W.; Kida, T.; Nakatsuji, Y.; Ikeda, I.
Tetrahedron Lett. 1997, 38, 2681.
38. Hawkins, J. M.; Fu, G. C. J. Org. Chem. 1986, 51, 2820.

A. I. Meyers et al. / Tetrahedron 60 (2004) 4459–4473
4473
39. Adapted from a procedure reported by Fray, A. H.; Augeri,
D. J.; Kleinman, E. F. J. Org. Chem. 1988, 53, 69.
in the Ullmann coupling: Qiu, L.; Qi, J.; Pai, G.; Chan, S.;
Zhou, Z.; Choi, M.; Chan, A. S. C. Org. Lett. 2002, 4, 4599.
41. Meyers, A. I.; Rawson, D. J. Chem. Soc., Chem. Commun.
1992, 491.
40. A recent report by Chan has also demonstrated that other chiral
substrates (chiral 1,3-dioxanes) on the aryl halide can be
utilized to form moderately diastereo-enriched biphenyls (7:2)
42. Suda, H.; Kanoh, S.; Umeda, N. Chem. Lett. 1984, 899.