Avoiding the classical resolution during the synthesis of MeO-BIPHEP and 3,3′-disubstituted derivatives

Article abstract of DOI:10.1016/j.tetlet.2005.03.183

The Ullmann coupling of 1 (R = H) gives a 2:1 mixture of diastereomers 2 (R = H) in 81% yield that are easily separated by silica gel chromatography. This procedure avoids the generally cumbersome and sometimes difficult resolution step with DBTA. Similar Ullmann couplings and separation of the corresponding diastereomers are employed with other derivatives of 1 (R = OtBu, iPr, Ph, and mesityl) ultimately affording a new series of 3,3′-disubsituted-MeO-BIPHEP derivatives. The use of these new derivatives in palladium-catalyzed asymmetric Heck reaction, Pd-catalyzed polyene cyclizations and rhodium-catalyzed hydrogenations is also reported.

Full text of DOI:10.1016/j.tetlet.2005.03.183

Tetrahedron
Letters
Tetrahedron Letters 46 (2005) 3843–3846
Avoiding the classical resolution during the synthesis of
MeO-BIPHEP and 3,30-disubstituted derivatives
Evgueni Gorobets,a Bronwen M. M. Wheatley,a J. Matthew Hopkins,a
Robert McDonaldb and Brian A. Keaya,*
aDepartment of Chemistry, University of Calgary, Calgary, AB, Canada T2N 1N4
bDepartment of Chemistry, University of Alberta, Edmonton, AB, Canada T6G 2G2
Received 7 March 2005; revised 28 March 2005; accepted 28 March 2005
Available online 12 April 2005
Abstract—The Ullmann coupling of 1 (R = H) gives a 2:1 mixture of diastereomers 2 (R = H) in 81% yield that are easily separated
by silica gel chromatography. This procedure avoids the generally cumbersome and sometimes difficult resolution step with DBTA.
Similar Ullmann couplings and separation of the corresponding diastereomers are employed with other derivatives of 1 (R = OtBu,
iPr, Ph, and mesityl) ultimately affording a new series of 3,30-disubsituted-MeO-BIPHEP derivatives. The use of these new deriva-
tives in palladium-catalyzed asymmetric Heck reaction, Pd-catalyzed polyene cyclizations and rhodium-catalyzed hydrogenations is
also reported.
Ó 2005 Elsevier Ltd. All rights reserved.
We recently reported the synthesis, resolution, and
applications of a variety of 3,30-disubstituted BIPHEP
derivatives (3,30 groups = OMe, OiPr, OPiv, and OTol-
yl).1 In order to compare their efficacy in asymmetric
reactions with that of MeO-BIPHEP 11 (Scheme 2),
we embarked on preparing 11 according to the proce-
dure reported by Schmid et al.2,3 but quickly found
the resolution step with dibenzoyl-L-tartaric acid ((ꢀ)-
DBTA) or di-p-toluoyl-L-tartaric acid ((ꢀ)-DTTA) to
be cumbersome. Since then, we have experienced similar
resolution problems with some of our newer 3,30-disub-
stituted MeO-BIPHEP derivatives 3–6 (3,30 groups
= OtBu, iPr, Ph,4 and mesityl) and thus embarked on
designing a new synthesis of MeO-BIPHEP 11 that
would alleviate the need for a classical resolution step.
Herein we report our new synthetic strategy toward
the preparation of phosphine oxides 2–6 that involved
the attachment of a substrate bound chiral auxiliary to
a suitable Ullmann precursor (i.e., 1) (Scheme 1), so that
upon affecting an Ullmann coupling diastereomers
would be formed that could be easily separated by col-
umn chromatography.5 In addition, the corresponding
phosphines of 3–6 are used in some asymmetric
transformations.
We started by redesigning the synthesis toward MeO-
BIPHEP (Scheme 2) to allow for the incorporation of
a chiral auxiliary prior to the Ullmann coupling.
R
R
O
2 R=H
O
OAc
O
3 R=OtBu
4 R=iPr
Cu, DMF
133oC, 4h
diastereomers are easily
separated by silica gel
chromatography
R*O
R*O
PPh2
PPh2
O
PPh2
5 R=Ph
I
1
6 R=Mesityl
O
R
Scheme 1.
Keywords: Ullmann coupling; BIPHEP derivatives.
*
Corresponding author. Tel.: +1 403 220 5340; fax: +1 403 284 1372; e-mail: keay@ucalgary.ca
0040-4039/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2005.03.183

3844
E. Gorobets et al. / Tetrahedron Letters 46 (2005) 3843–3846
OR1
R3
a-d (66%)
OtBu
O
PPh2
e (91%)
O
g
HO
PPh2
MeO
Br
R*O
PPh2
R2O
RO
I
7
I
O
9
I
12 R1=R2=R3=H
8
a,b
d
18 R=TBS (86%)
19 R=H (83%)
13 R1=H, R2=Ac, R3=Br (50%)
14 R1=tBu, R2=Ac, R3=Br (89%)
15 R1=tBu, R2=H, R3=Br (99%)
16 R1=tBu, R2=TBS, R3=Br (94%)
h
f (81%)
c
i
20 R=AcLact
e
(99%)
j
f
17 R1=tBu, R2=TBS, R3=POPh2 (94%)
O
PPh2
PPh2
O
g-i (83%)
MeO
MeO
PPh2
PPh2
R*O
R*O
OtBu
OtBu
O
l
MeO
MeO
PPh2
PPh2
MeO
MeO
PPh2
PPh2
(Rax)-(+)-10
11 MeO-BIPHEP
O
OtBu
(Rax)-(+)-23 (54%)
Scheme 2. Reagents and conditions: (a) n-BuLi, THF, ꢀ78 °C, 1 h,
then Ph2PCl, ꢀ78 °C to rt, 2 h; (b) H2O2, MeOH, rt, 0.5 h (93% two
steps); (c) LDA, THF, ꢀ78 °C, 1.5 h; then I2, THF, ꢀ78 °C to rt, 1 h
(76%); (d) BBr3, DCM, rt, 12 h (93%); (e) (S)-2-acetoxypropanoyl
chloride, TEA, DCM, rt, 1 h (91%); (f) Cu powder, DMF, 150 °C, 3 h;
separation of diastereomers by silica gel chromatography (76% both
diastereomers); (g) KOH, EtOH, rt, 0.5 h; (h) MeI, DCM, H2O,
Adogen, rt, 24 h (94% two steps); (i) Cl3SiH, n-Bu3N, xylene, 145 °C,
24 h (89%).
OtBu
(Sax)-(-)-21 and
(Rax)-(+)-21 R=AcLact (89%)
(Rax)-(+)-22 R=Me (97%)
k
Scheme 3. Reagents and conditions: (a) Ac2O, AcOH, rt, 1 h; (b) Br2,
CHCl3, 0 °C, 1 h; (c) 2-methyl-propene, DCM, TfOH, ꢀ78 °C, 4 h; (d)
KOH, EtOH, rt, 15 min; (e) TBSCl, imid, DMC, rt, 1 h; (f) n-BuLi,
THF, ꢀ78 °C, 1 h; then Ph2PCl, 2 h; then H2O2, MeOH, 0 °C, 20 min;
(g) LDA, THF, ꢀ78 °C, 1.5 h; then I2, ꢀ78 °C to rt, 1.5 h; (h) TBAF,
THF, rt, 15 min; (i) (R)-2-acetoxypropanoyl chloride, DCM, TEA,
DMAP, 0 °C, 30 min; (j) Cu, DMF, 150 °C, 6 h; separation of
diastereomers by silica gel chromatography (89% for both diastereo-
mers); (k) KOH, EtOH, rt, 30 min; then MeI, DCM, Adogen, rt, 12 h;
(l) HSiCl3, xylene, 105 °C, 24 h.
Halogen–metal exchange of bromide 7 with n-BuLi fol-
6
7
lowed by the addition of ClPPh2 and H2O2 gave a
phosphine oxide that when treated with LDA at
ꢀ78 °C in THF followed by iodine gave 8 after removal
of the methyl group with BBr3. Reaction of 8 with (S)-2-
acetoxypropanoyl chloride8 gave 9, which was subjected
to an Ullmann coupling9 to give a 33% de in favor of
diastereomer (Rax)-(+)-10. The diastereomers were eas-
ily separated by silica gel chromatography, with the dia-
stereomer with the Sax configuration always having the
larger Rf. Removal of the chiral auxiliary (KOH,
EtOH), methylation (MeI, Adogen), and trichlorosilane
reduction10 of the phosphine oxides provided MeO-BIP-
HEP 11. Although the sequence is two steps longer than
the original, a resolution step was not necessary and the
diastereomers from the Ullmann coupling were easily
separated.
CHO
d-i (55%)
CH(OMe)2
a-c (62%)
iPr
MeO
PPh2
MeO
RO
PPh2
O
I
O
24
25
26 R=Me
j,k
27 R=AcLact (67%)
l (69%)
iPr
iPr
O
MeO
MeO
PPh2
PPh2
m-o (99%)
AcOLactO
AcOLactO
PPh2
PPh2
O
iPr
iPr
(Rax)-(+)-29
(Rax)-(-)-28
The diastereomeric Ullmann coupling strategy was
applied to the synthesis of ligands 23, 29, and 34. Com-
pound (+)-23 was prepared in twelve steps from
hydroquinone 12 as outlined in Scheme 3. The Ullmann
coupling of 20 provided a 24% de in favor of (Rax)-(+)-
21. The diastereomers of 21 were separated and the Rax-
isomer carried on to compound (Rax)-(+)-23.11 Com-
pound (+)-29 was prepared in 15 steps from 24 (Scheme
4). Ullmann coupling of 27 again gave a 3:2 mixture of
(Rax)-28 and (Sax)-28 that were easily separated by col-
umn chromatography. Finally, Ullmann coupling of
32 gave a 1:1.3 mixture of diastereomers 33 (Scheme 5)
that was eventually converted into compound (Rax)-34
in 12 steps from 2-bromo-4-methoxyaniline (30).12 The
same sequence outlined in Scheme 5 was used to prepare
the 3,30-dimesitylene compound 35. Removal of the chi-
ral auxiliary from (+)-35 and methylation gave (ꢀ)-36.
Reduction of (ꢀ)-36 with Cl3SiH resulted in the reduc-
tion of only one of the two phosphine oxides in 25%
yield (75% unreacted (ꢀ)-36). Changing the reaction
conditions never afforded the desired bis-phosphine.
Interestingly, treatment of (ꢀ)-36 with alane unexpect-
edly gave compound 37.
Scheme 4. Reagents and conditions: (a) t-BuLi, Et2O, ꢀ25 °C, 3 h,
then Ph2PCl, ꢀ78 °C to rt, 12 h; (b) H2O2, MeOH, rt, 1 h (68% two
steps); (c) pTsOH, acetone, rt, 1.5 h (91%); (d) MeMgBr, THF, 5 °C,
1 h; (e) MnO2, acetone, rt, 36 h; (f) MeMgBr, THF, rt, 1 h; (g) Ac2O,
120 °C, 24 h; (h) H2, Pt2O, 1 atm, rt, 6 h; (i) LDA, THF, ꢀ78 °C, 1.5 h;
then I2, THF, ꢀ78 °C to rt, 2 h (55% six steps); (j) BBr3, DCM, rt, 20 h
(93%); (k) (S)-2-acetoxypropanoyl chloride, TEA, DCM, rt, 1 h (67%
two steps); (l) Cu powder, DMF, 133 °C, 4 h; separation of diastereo-
mers by silica gel chromatography (69% both diastereomers); (m)
KOH, EtOH, rt, 1 h; (n) MeI, DCM, H2O, Adogen, rt, 48 h; (o)
Cl3SiH, n-Bu3N, xylene, 135 °C, 48 h (99% three steps).
With (+)-(R)-23, (+)-(R)-29, and (ꢀ)-(R)-34 in hand, we
compared the efficacy of these ligands in the asymmetric
Heck arylation of 2,3-dihydrofuran and compared the
results to those obtained with (+)-MeO-BIPHEP (11,
Table 1). (+)-MeO-BIPHEP 11 provided 65% conver-
sion to products and gave 92% ee of 40, 63% ee of 42
along with a small amount of the conjugated product
41 (Table 1, entry 1). The ratio of 40:41:42 was
83:7:10. The best result was with the 3,30-di-OtBu

E. Gorobets et al. / Tetrahedron Letters 46 (2005) 3843–3846
3845
conversion increased to 60% in a 24 h period. It is note-
worthy that 34 gave 40 not only with a 99% ee but also
with an excellent 40:42 ratio of 99:1 (entries 4 and 5).
This is the highest % ee and product ratio observed to
date in our laboratory using a variety of chiral cata-
lysts.1,15 Work is continuing to improve the % conver-
sion with (ꢀ)-(R)-34.
NH2
Br
Ph
a-c (37%)
MeO
RO
PPh2
30
I
O
31 R=Me
d,e
32 R=AcLact (68%)
f (72%)
Ph
Ph
O
Ligands (+)-(R)-23, (+)-(R)-29, and (ꢀ)-(R)-34 were
then used in the palladium catalyzed polyene cyclization
(43 ! 44) and compared to the results obtained with
(+)-(R)-MeO-BIPHEP (11, Table 2). (+)-(R)-MeO-BIP-
HEP (11) afforded (R)-44 in 72% ee (53% yield). The use
of the 3,30-di-O-tBu ligand (+)-(R)-23 gave a compar-
able yield and slightly lower % ee than with 11; however,
the absolute stereochemistry of the major enantiomer
was opposite of that obtained with (+)-(R)-11. This
result is in line with the results obtained with other
3,30-diOR BIPHEP derivatives reported previously.1
MeO
MeO
PPh2
PPh2
g-i (86%)
AcOLactO
AcOLactO
PPh2
PPh2
O
Ph
Ph
(Rax)-(-)-34
(Rax)-(-)-33
Mes
O
PPh2
PPh2
O
Mes
H
PPh
H
RO
RO
j
MeO
MeO
Mes
Mes
(+)-35 R=AcOLact
(-)-36 R=Me
37
g,h
3,30-Di-Ph-BIPHEP derivative (ꢀ)-(R)-34 gave a low
53% yield and disappointing 24% ee. In this case, the
major enantiomer had the R-configuration, which was
in line with that obtained using (+)-(R)-11. Finally,
(+)-(R)-29 gave (R)-44 in excellent yield (83%) and with
an 80% ee. This is the highest % ee we have observed in
this reaction to date.1,16
Scheme 5. Reagents and conditions: (a) n-C5H11ONO, benzene (or
mesitylene), reflux, 1.5 h (52%); (b) n-BuLi, THF, ꢀ78 °C, 3 h, then
Ph2PCl, ꢀ78 °C to rt, 12 h; then H2O2, MeOH, rt, 1 h (91%); (c) LDA,
THF, ꢀ78 °C, 1.5 h; then I2, THF, ꢀ78 °C to rt, 2 h (78%); (d) BBr3,
DCM, rt, 20 h (93%); (e) (S)-2-acetoxypropanoyl chloride, TEA,
DCM, rt, 1 h (72%); (f) Cu powder, DMF, 145 °C, 4 h; separation of
diastereomers by silica gel chromatography (72% both diastereomers);
(g) KOH, EtOH, rt, 1 h; (h) MeI, DCM, H2O, Adogen, rt, 48 h; (i)
Cl3SiH, n-Bu3N, xylene, 135 °C, 48 h (93% three steps); (j) alane, THF,
reflux, 3 h (78%).
Ligands (+)-(R)-23, (+)-(R)-29, and (ꢀ)-(R)-34 were
then used in catalytic hydrogenations of acid 45
(R = H) and ester 45 (R = Me) and compared to results
obtained with MeO-BIPHEP 11 (Table 3).17 While (+)-
(R)-MeO-BIPHEP 11 performed poorly in the hydro-
genation of 45 (entry 1) an improvement was observed
with ligand (+)-(R)-29. A lower % ee upon using (+)-
(R)-29 with acid 45 (R = H) was surprising given the
excellent results with other ligands; repeating the reac-
tion did not change the % ee. The best results were seen
with ligands (+)-(R)-23 and (ꢀ)-(R)-34 with the latter
giving the best results on hydrogenation with both the
acid and ester of 45.
Table 1. Asymmetric Heck results with ligands 11, 23, 29, and 34
OTf
Ph
Ph
Ph
40
41
3 mol% Pd(OAc)2
6 mol% ligand
O
+
+
benzene, 40 o
iPr2NEt, 24h
C
O
O
+
sealed vial
38
39
42
O
Ligand
% Conversion
Ratio of products
40 (% ee) 41 42 (% ee)
Table 2. Asymmetric Pd-catalyzed polyene cyclization results with
ligands 11, 23, 29, and 34
1
2
3
4
5
6
(+)-(R)-11
(+)-(R)-23
(+)-(R)-29
(ꢀ)-(R)-34
(ꢀ)-(R)-34
65
100
12
83 (92)
96 (89)
66 (51)
99 (>99)
99 (>99)
91 (81)
7
0
0
0
0
0
10 (63)
4 (11)a
34 (25)
1 (>99)a
1 (>99)a,b
9 (61)c
14
60
O
5 mo% Pd2(dba)3,
(R)-BINAP
41
OTf
10 mol% ligand
(R)-44
O
a The major enantiomer of 42 had the R-configuration.
b 10 mol % Pd(OAc)2, 20 mol % (ꢀ)-(R)-34.
+
PMP, toluene, 105 oC,
48h
O
O
43
c Results reported by Hayashi and co-workers.13
O
(S)-44
O
bisphosphine 23 (entry 2). The conversion to products
was 100% with 40 formed in 96% yield with 89% ee. Li-
gand 23 well outperformed (+)-(R)-BINAP in % conver-
sion, ratio of 40:42 and in % ee of 40.13 Both the 3,30-di-
iPr derivative 29 (entry 29) and 3,30-di-Ph derivative 34
(entry 4) gave poor conversions at 3 mol % Pd loading
(12% and 14%, respectively);14 however, when the Pd
loading was increased to 10 mol % with ligand 34, the
Catalyst
% Yield
Ratio of
enantiomers
% ee
(R)-44 (S)-44
1
2
3
4
(+)-(R)-11
(+)-(R)-23
(+)-(R)-29
(ꢀ)-(R)-34
53
56
83
53
86
19
90
62
14
81
10
38
72
62
80
24

3846
E. Gorobets et al. / Tetrahedron Letters 46 (2005) 3843–3846
Table 3. Asymmetric hydrogenation results with ligands 11, 23, 29,
and 34
References and notes
1. Gorobets, E.; Sun, G.-R.; Wheatley, B. M. M.; Parvez,
M.; Keay, B. A. Tetrahedron Lett. 2004, 45, 3597–3601.
2. (a) Schmid, R.; Cereghetti, M.; Heiser, B.; Schonholzer,
P.; Hansen, H.-J. Helv. Chim. Acta 1988, 71, 897; (b)
Schmid, R.; Foricher, J.; Cereghetti, M.; Scho¨nholzer, P.
Helv. Chim. Acta 1991, 74, 370–389.
3. Another reason for preparing 11 is its high cost from
Aldrich Chemical Co. 2005–06 price = $0.86/mg ($Cdn).
4. (a) The synthesis of 3,30-diphenyl MeO-BIPHEP was
recently reported by Zhang and co-workers and involved a
resolution step with (L)-DTTA: Wu, S.; He, Minsheng;
Zhang, X. Tetrahedron: Asymmetry 2004, 15, 2177–2180;
(b) Tang, W.; Chi, Y.; Zhang, X. Org. Lett. 2002, 4, 1695;
(c) Tang, W.; Zhang, X. Chem. Rev. 2003, 103, 3029–3069.
5. We were not concerned about achieving high diastereose-
lectivity in the Ullmann coupling as both enantiomers of
the various BIPHEP derivatives were required for appli-
cation purposes.
O
O
H2 (2 atm),
1mol % Rh(COD)2OTf,
1.1 mol% ligand,
MeOH, r.t., 24h
OR
OR
N
N
H
O
H
O
45
Catalyst
46
R
% Conversion
% ee (config)
1
2
3
4
(+)-(R)-11
(+)-(R)-23
(+)-(R)-29
(ꢀ)-(R)-34
Me
H
100
100
100
100
100
100
100
77
25.6 (R)
19.5 (R)
94.6 (R)
97.7 (R)
80.6 (R)
37.1 (Ra)
98.5 (R)
96.8 (R)
Me
H
Me
H
Me
H
a Reaction attempted twice.
6. Hall, T. J.; Hargis, J. H. J. Org. Chem. 1986, 51, 4185–
4189.
7. Hamashima, Y.; Kanai, M.; Shibasaki, M. Tetrahedron
Lett. 2001, 42, 691–694.
8. (S)-2-Acetoxypropanoyl chloride is available from
Aldrich.
We have shown that a diastereoselective Ullmann cou-
pling using a (+)-2-acetoxypropanoyl group as the chiral
auxiliary provides biaryl systems that are easily sepa-
rated by silica gel chromatography, thereby avoiding
the generally cumbersome and sometimes difficult reso-
lution step with (L)-DBTA. Interestingly, ligands (+)-
(R)-23, (+)-(R)-29, and (ꢀ)-(R)-34 each gave the best %
eeÕs in the Hayashi Heck reaction, palladium-catalyzed
polyene cyclization and catalytic hydrogenation of 45,
respectively, indicating that one specific 3,3-disubstituted
BIPHEP ligand is not ideal for all asymmetric transfor-
mations. Work is continuing to shorten the synthetic
routes to 3,30-disubsituted BIPHEP derivatives and
applying these ligands in other asymmetric reactions.
´
9. Hassan, J.; Sevignon, M.; Gozzi, C.; Schulz, E.; Lemaire,
M. Chem. Rev. 2002, 102, 1359–1469.
10. Takaya, H.; Mashima, K.; Yagi, M.; Kumobayashi, H.;
Taketomi, T.; Akutagawa, S.; Noyori, R. J. Org. Chem.
1986, 51, 629–635.
11. The lower yield of 54% was due to loss of some of the t-
butyl group in the presence of Cl3SiH at 105 °C.
12. Ishibashi, H.; Kobayashi, T.; Machida, N.; Tamura, O.
Tetrahedron 2000, 56, 1469–1473.
13. (a) Ozawa, F.; Kubo, A.; Matsumoto, Y.; Hayashi, T.
Organometallics 1993, 12, 4188; (b) Ozawa, F.; Kubo, A.;
Hayashi, T. J. Am. Chem. Soc. 1991, 113, 1417.
14. Stirring the reaction longer did not increase the %
conversion. We are continuing to investigate why the
reaction appears to stall with ligands 29 and 34.
15. Andersen, N. G. Ph.D. Dissertation, University of Cal-
gary, 2000.
16. (a) Lau, S. Y. W.; Keay, B. A. Synlett 1999, 605–607; (b)
Lau, S. Y. W.; Andersen, N. G.; Keay, B. A. Org. Lett.
2001, 3, 181–184; (c) Che, D.; Andersen, N. G.; Lau, S. Y.
W.; Parvez, M.; Keay, B. A. Tetrahedron: Asymmetry
2000, 11, 1919–1925; (d) Andersen, N. G.; McDonald, R.;
Keay, B. A. Tetrahedron: Asymmetry 2001, 12, 263–269;
(e) Maddaford, S. P.; Andersen, N. G.; Cristofoli, W. A.;
Keay, B. A. J. Am. Chem. Soc. 1996, 118, 10766–10773; (f)
Keay, B. A.; Maddaford, S. P.; Cristofoli, W. A.;
Andersen, N. G.; Passafaro, M. S.; Wilson, N. S.; Nieman,
J. A. Can. J. Chem. 1997, 75, 1163–1171.
Acknowledgements
We thank Merck Frosst (Pointe Claire, PQ), NSERC
CRD program, and the University of Calgary for finan-
cial support. In addition NSERC and the Alberta Inge-
nuity Fund are thanked for postgraduate scholarships
(for B.M.M.W. and J.M.H.).
Supplementary data
Experimental procedures for the diastereoselective Ull-
mann coupling of 9, 20, 27, and 32 are provided. The
X-ray crystal structure of 37 is also provided. Supple-
mentary data associated with this article can be found,
in the online version, at doi:10.1016/j.tetlet.2005.03.183.
17. Li, W.; Zhang, Z.; Xiao, D.; Zhang, X. J. Org. Chem.
2000, 65, 3489–3496.