Catalytic asymmetric deprotonation of a phosphine borane: comparison of two-ligand and one-ligand catalysis

Article abstract of DOI:10.1016/j.tetasy.2009.10.002

The catalytic asymmetric deprotonation of tert-butyldimethylphosphine borane using s-BuLi or n-BuLi and sub-stoichiometric amounts of (-)-sparteine under one-ligand and two-ligand manifolds has been investigated. Using s-BuLi, slightly higher enantioselec

Full text of DOI:10.1016/j.tetasy.2009.10.002

Tetrahedron: Asymmetry 20 (2009) 2407–2412
Contents lists available at ScienceDirect
Tetrahedron: Asymmetry
journal homepage: www.elsevier.com/locate/tetasy
Catalytic asymmetric deprotonation of a phosphine borane: comparison
of two-ligand and one-ligand catalysis
b
Steven J. Canipa a, Peter O’Brien a, , Sven Taylor
*
a Department of Chemistry, University of York, Heslington, York YO10 5DD, UK
b F. Hoffmann-La Roche Ltd, CH-4070, Basel, Switzerland
a r t i c l e i n f o
a b s t r a c t
Article history:
The catalytic asymmetric deprotonation of tert-butyldimethylphosphine borane using s-BuLi or n-BuLi
and sub-stoichiometric amounts of (À)-sparteine under one-ligand and two-ligand manifolds has been
investigated. Using s-BuLi, slightly higher enantioselectivity was obtained using two-ligand catalysis
(use of sub-stoichiometric (À)-sparteine in the presence of a stoichiometric amount of a second achiral
ligand) compared to one-ligand catalysis (use of sub-stoichiometric (À)-sparteine only). With n-BuLi,
two-ligand catalysis using LiDMAE (DMAE = dimethylaminoethanol) as the stoichiometric ligand was
the only method for obtaining good yield and enantioselectivity. In this case, one-ligand catalysis failed
as the (À)-sparteine was not turned over.
Received 8 September 2009
Accepted 2 October 2009
Available online 28 October 2009
Ó 2009 Elsevier Ltd. All rights reserved.
1. Introduction
hindered to minimise the background deprotonation). Thus, N-Boc
pyrrolidine 4 and O-alkyl carbamate 5 require what we refer to as
The chiral base obtained by combining s-BuLi or n-BuLi and (À)-
sparteine is particularly effective for the asymmetric deprotona-
tion of O-alkyl carbamates,1 N-Boc heterocycles,2 medium-ring
epoxides,3 ferrocene amides4 and phosphine boranes and sulfides.5
One limitation of the use of such chiral bases in synthesis is their
reliance on a stoichiometric amount of chiral ligand. With this in
mind, our group became interested in developing catalytic variants
of existing asymmetric deprotonation reactions in which a sub-
stoichiometric amount of (À)-sparteine is used. It is well estab-
lished that complexing a diamine to an organolithium reagent in-
creases its reactivity. From this, it follows that as long as there is
little racemic (background) deprotonation by the uncomplexed
organolithium reagent then it should be possible to carry out cat-
alytic asymmetric deprotonation with high enantioselectivity. In-
deed, Hodgson demonstrated this in the asymmetric lithiation-
rearrangement of cyclooctene oxide 13 whilst we have also re-
ported successful examples with ferrocene amide 26 and phos-
phine sulfide 37 (Scheme 1). We refer to these examples as one-
ligand catalysis.
two-ligand catalysis for successful catalytic asymmetric deprotona-
tion (Scheme 1). More recently, we have obtained high enantiose-
lectivity using lithiated dimethylaminoethanol (LiDMAE) in two-
ligand catalysis with N-Boc pyrrolidine 4.12
On inspection of our original disclosures on two-ligand and
one-ligand catalysis, we reported the catalytic asymmetric
One-ligand catalysis successful
O
Me
S
P
Me
Ph
N
O
Fe
Me
Me
tBu
Et
1
2
3
Two-ligand catalysis required
O
N
N
In contrast, the attempted one-ligand catalytic asymmetric
deprotonation of N-Boc pyrrolidine 4 or O-alkyl carbamate 5 led
to low yields and poor enantioselectivity.8 Suspecting that the
problems lay with the poor turn-over of the chiral ligand, we intro-
duced a second ligand into the system. The best results with N-Boc
pyrrolidine 48–10 or O-alkyl carbamate 58,11 were obtained using
achiral bispidine 6 (whose s-BuLi complex is sufficiently sterically
Ph
O
NiPr2
N
Boc
4
5
6
H
N
N
H
BH3
P
OLi
Me
tBu
Me2N
LiDMAE
Scheme 1. One-ligand and two-ligand catalytic asymmetric deprotonation.
Me
7
(–)-sparteine
* Corresponding author. Tel.: +44 01904 532535; fax: +44 01904 432516.
E-mail address: paob1@york.ac.uk (P. O’Brien).
0957-4166/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2009.10.002

2408
S. J. Canipa et al. / Tetrahedron: Asymmetry 20 (2009) 2407–2412
s
deprotonation of phosphine borane 7 using both approaches
(Scheme 2).13 For example, using one-ligand catalysis, lithiation–
oxygenation of phosphine borane 7 with s-BuLi and 0.2 equiv
(À)-sparteine gave hydroxy phosphine borane (R)-8 in 57% yield
with 77:23 er.6 Slightly higher enantioselectivity was obtained in
the two-ligand catalytic asymmetric deprotonation: lithiation of
7 using s-BuLi, 0.2 equiv (À)-sparteine and 1.2 equiv bispidine 6
and trapping with benzophenone-delivered hydroxy phosphine
borane (S)-9 in 67% yield and 83:17 er.8
1. 1.1 eq. BuLi
0.2 eq. (–)-sparteine
0 or 1.2 eq. ligand
BH3
P
BH3
P
SiMe2Ph
tBu
Me
Me
tBu
Me
–78 °C, Et2O, 3 h
2. PhMe2SiCl
7
(R)-10
Scheme 3. s-BuLi-mediated one-ligand and two-ligand catalytic asymmetric
deprotonation of phosphine borane 7.
Using one-ligand catalysis (1.1 equiv s-BuLi and 0.2 equiv (À)-
sparteine), adduct (R)-10 was isolated in 76% yield and 74:26 er
(Table 1, entry 1). This is in lower enantioselectivity than the stoi-
chiometric reaction (92:8 er) and can be attributed to an apprecia-
ble degree of background (racemic) lithiation by s-BuLi. Slightly
better results were obtained when 1.2 equiv bispidine 6 was
added, under otherwise identical conditions: adduct (R)-10 was
formed in 78% yield with 79:21 er (Table 1, entry 2). This result
clearly indicates the beneficial effect of adding bispidine 6 as the
second ligand and is consistent with the trend shown in Scheme 2.
As part of this study, we also investigated two-ligand catalysis
using LiDMAE and N-methyl morpholine as the second, regenerat-
ing ligands since both had given satisfactory results with N-Boc
pyrrolidine 4.10,12 The use of LiDMAE (formed in situ by adding
an extra 1.2 equiv s-BuLi to 1.2 equiv dimethylaminoethanol) gave
adduct (R)-10 in 59% yield and 81:19 er (Table 1, entry 3), similar
enantioselectivity to that obtained with bispidine 6. In contrast,
poor enantioselectivity (65:35 er) was observed using N-methyl
morpholine (Table 1, entry 4). Overall, the s-BuLi-mediated two-li-
gand catalytic asymmetric deprotonation of phosphine borane 7
proceeds with higher enantioselectivity using either bispidine 6
or LiDMAE (prepared in situ from commercially available DMAE)
than the corresponding one-ligand catalysis.
The preliminary conclusion from the results in Scheme 2 is that
two-ligand catalysis provides a slight improvement in enantiose-
lectivity over one-ligand catalysis. However, since there are differ-
ences between these two examples (e.g., amount of s-BuLi used
and different electrophile trapping procedures), we have now car-
ried out a systematic comparison of the s-BuLi-mediated one-li-
gand and two-ligand catalytic asymmetric deprotonation of
phosphine borane 7. The study includes a wider range of achiral,
regenerating ligands and examples with n-BuLi which behaves in
a different fashion to s-BuLi. Herein, we report our results and pro-
vide some guidelines on the scope and limitations of the catalytic
asymmetric deprotonation of phosphine boranes.
s
1. 1.1 eq. BuLi
One-ligand
catalysis
0.2 eq. (–)-sparteine
–78 °C, Et2O, 3 h
BH3
P
BH3
P
57%
77:23 er
OH
tBu
Me
(R)-8
Me
tBu
Me
2. O2, –78 °C, 1 h
then rt, 16 h
7
s
1. 1.3 eq. BuLi
0.2 eq. (–)-sparteine
1.2 eq. bispidine 6
–78 °C, Et2O, 3 h
Two-ligand
catalysis
BH3
P
BH3 OH
P
tBu
Me
As Kann14 and ourselves have noted,6–8 it is possible to access
either enantiomer of phosphine boranes such as 8, 9 and 10 using
(À)-sparteine or (+)-sparteine surrogate.15 In one-ligand and two-
ligand catalysis with s-BuLi, we have previously noted that that
higher enantioselectivity was obtained using the (+)-sparteine sur-
rogate compared to (À)-sparteine under the same conditions.6,8
This was rationalised by higher reactivity of the s-BuLi/(+)-sparte-
ine surrogate complex, as noted in the deprotonations of N-Boc
pyrrolidine 49 and O-alkyl carbamate 5.11 To confirm this, a compe-
tition experiment was carried out (Scheme 4).
Me
Ph
tBu
67%
83:17 er
Ph
Me
2. Ph2CO
3. HCl(aq)
7
(S)-9
Scheme 2. Catalytic asymmetric deprotonation of phosphine borane 7.
2. Results and discussion
For the purposes of comparing one-ligand and two-ligand catal-
ysis, the lithiation reactions presented in Tables 1 and 2 utilise
s
Table 1
1. 2.4 eq. BuLi
s-BuLi-mediated one-ligand and two-ligand catalytic asymmetric deprotonation of
phosphine borane 7?(R)-10 (Scheme 3)
BH3
P
1.2 eq. (–)-sparteine
1.2 eq. (+)-sparteine surrogate
BH3
P
SiMe2Ph
56%
Me
Me
tBu
tBu
Entry
Ligand
Yielda (%)
er (R:S)b
Me
–78 °C, Et2O, 3 h
2. PhMe2SiCl
7
(S)-10 74:26 er
1
2
3
4
—
76
78
59
67
74:26
79:21
81:19
65:35
Bispidine 6
Me
H
N
H
LiDMAEc
N
N
N-Me morpholine
N
a
b
c
H
Yield after purification by flash column chromatography.
Enantiomer ratio (er) determined by chiral HPLC.
1.2 equiv LiDMAE was formed in situ by deprotonation with an extra 1.2 equiv
(+)-sparteine
surrogate
(–)-sparteine
s-BuLi.
Scheme 4. Competition experiment between s-BuLi/(À)-sparteine and s-BuLi/(+)-
sparteine surrogate.
1.1 equiv s-BuLi or n-BuLi and PhMe2SiCl for the electrophilic trap
(thus generating silyl phosphine borane (R)-10). Using 1.1 equiv s-
BuLi alone, lithiation–silylation gave adduct rac-10 in 70% yield
indicating a considerable background lithiation rate by s-BuLi in
Et2O at À78 °C. Carrying out the same reaction in the presence of
stoichiometric (À)-sparteine (1.2 equiv) generated (R)-10 in 74%
yield and 92:8 er.7 This is the benchmark result for the one-ligand
and two-ligand catalysis study, the results of which are presented
in Scheme 3 and Table 1.
Thus, 2.4 equiv s-BuLi was combined with 1.2 equiv of each of
(À)-sparteine and the (+)-sparteine surrogate in Et2O at À78 °C.
Phosphine borane 7 was then added to this mixture and, after
3 h, the lithiated intermediates were trapped with PhMe2SiCl. This
generated a 56% yield of (S)-10 with 74:26 er. Since the major
product is formed with the (+)-sparteine surrogate sense of induc-
tion, we conclude that the s-BuLi/(+)-sparteine surrogate complex
is indeed more reactive in the deprotonation of phosphine borane
7 than s-BuLi/(À)-sparteine.

S. J. Canipa et al. / Tetrahedron: Asymmetry 20 (2009) 2407–2412
2409
1. 1.1 eq. nBuLi
Next, we turned our attention to the use of n-BuLi. Using n-BuLi in
Et2O at À78 °C, we have been unable to lithiate phosphine borane 7
(despite numerous attempts and the use of different electrophiles).
This reflects the lower basicity of n-BuLi compared to s-BuLi and is
significant for catalysis since there should be no background racemic
lithiation using n-BuLi. Under stoichiometric conditions (1.1 equiv
s-BuLi and 1.2 equiv (À)-sparteine), lithiation occurred satisfactorily
and, after trapping with PhMe2SiCl, adduct (R)-10 was obtained in
76% yield and 89:11 er.7 For comparison with this stoichiometric re-
sult, one-ligand catalysis and two-ligand catalysis (with four regen-
erating ligands) were explored (Scheme 5 and Table 2).
0-1.2 eq. (–)-sparteine
Temp, Et 2O, 3 h
BH3
P
BH3 OH
P
tBu
Me
Me
Ph
tBu
Ph
Me
2. Ph2CO
7
(S)-9
Scheme 6. n-BuLi-mediated one-ligand catalytic asymmetric deprotonation of
phosphine borane 7.
Table 3
Temperature effect on the n-BuLi-mediated one-ligand catalytic asymmetric depro-
tonation of phosphine borane 7?(S)-9 (Scheme 6)
Table 2
Entry
(À)-Sparteine (equiv)
1.2
0.2
0
0.2
0
0.2
0
Temperature (°C)
Yielda (%)
er (R:S)b
n-BuLi-mediated one-ligand and two-ligand catalytic asymmetric deprotonation of
phosphine borane 7?(R)-10 (Scheme 5)
1
2
3
4
5
6
7
8
9
À78
À78
À78
À60
À60
À50
À50
À40
À40
À40
56
19
0
42
0
50
11
84
69
29
88:12
87:13
—
87:13
—
81:19
—
84:16
78:22
—
Entry
Ligand
Yielda (%)
er (R:S)b
1
2
3
4
5
—
21
52
48
22
83
84:16
59:41
82:18
87:13
51:49
Bispidine 6
LiDMAEc
N-Me morpholine
PMDETAd
1.2
0.2
0
a
b
c
Yield after purification by flash column chromatography.
Enantiomer ratio (er) determined by chiral HPLC.
1.2 equiv LiDMAE was formed in situ by deprotonation with an extra 1.2 equiv
10
a
Yield after purification by flash column chromatography.
Enantiomer ratio (er) determined by chiral HPLC.
b
s-BuLi.
d
PMDETA = pentamethyldiethylenetriamine.
ane 7 using n-BuLi, we decided to investigate whether increasing
the lithiation temperature might facilitate a turn-over of the (À)-
sparteine under a one-ligand catalytic protocol. At higher temper-
atures, however, there may be reduced enantioselectivity and
background lithiation may become an issue. Nevertheless, some
interesting results were obtained from a temperature study on
the n-BuLi/(À)-sparteine-mediated one-ligand catalytic deprotona-
tion of phosphine borane 7. In these examples, benzophenone was
used as the electrophilic trap (Scheme 6 and Table 3).
1. 1.1 eq. nBuLi
0.2 eq. (–)-sparteine
0 or 1.2 eq. ligand
BH3
P
BH3
P
SiMe2Ph
tBu
Me
Me
tBu
Me
–78 °C, Et2O, 3 h
2. PhMe2SiCl
7
(R)-10
Scheme 5. n-BuLi-mediated one-ligand and two-ligand catalytic asymmetric
deprotonation of phosphine borane 7.
The stoichiometric lithiation trapping at À78 °C (1.1 equiv s-
BuLi and 1.2 equiv (À)-sparteine) afforded a 56% yield of (S)-9 with
88:12 er (Table 3, entry 1). Use of 0.2 equiv (À)-sparteine at À78 °C
gave a 19% yield of (S)-9 of 87:13 er (Table 3, entry 2), indicating no
turn-over of (À)-sparteine and consistent with the result obtained
using PhMe2SiCl (21% yield of (R)-10 of 84:16 er; Table 2, entry 1).
However, by raising the temperature to À60 °C, ligand turn-over
was facilitated as the adduct (S)-9 of high er was formed in >20%
yield: 42% yield of (S)-9 of 87:13 er (Table 3, entry 4). At À60 °C,
there was no background lithiation (Table 3, entry 5) but an
increasing amount of background lithiation was observed at higher
temperatures: using n-BuLi alone gave an 11% yield of rac-9 at
À50 °C (Table 3, entry 7) and a 29% yield of rac-9 at À40 °C (Table 3,
entry 10). As a result, the catalytic reactions performed at À50 °C
and À40 °C resulted in a drop in enantioselectivity (Table 3, entries
6 and 9). Stoichiometric lithiation-trapping was carried out at
À40 °C and gave (S)-9 in 84% yield with 84:16 er (Table 3, entry
8) indicating a lower enantioselectivity compared to that carried
out at À78 °C (88:12 er; Table 3, entry 1). Overall, higher temper-
atures did allow (À)-sparteine to be turned over and, at À50 °C, (S)-
9 of 81:19 er was isolated in 50% yield using n-BuLi and 0.2 equiv
(À)-sparteine (Table 3, entry 6).
Under one-ligand catalytic conditions (1.1 equiv n-BuLi and
0.2 equiv (À)-sparteine), we isolated adduct (R)-10 of 84:16 er
but in only 21% yield (Table 2, entry 1). The low yield and high
enantioselectivity indicate that the (À)-sparteine chiral ligand is
not being turned over in a catalytic fashion. This is different to that
observed with s-BuLi indicating that ligand turn-over depends on
the organolithium reagent employed. Since two-ligand catalysis
is a way of recycling the chiral diamine, we wondered whether a
two-ligand catalytic system could be developed for the n-BuLi-
mediated lithiations.
With bispidine 6 (Table 2, entry 2) and PMDETA (Table 2, entry
5), yields in excess of 20% were obtained, but the enantioselectivity
was poor (59:41 er and 51:49 er, respectively). These results could
be explained by a significant amount of background lithiation by n-
BuLi/6 and n-BuLi/PMDETA in Et2O at À78 °C. This was confirmed
for PMDETA: lithiation trapping using n-BuLi/PMDETA (Et2O,
À78 °C) gave adduct rac-10 in 69% yield. Using N-methyl morpho-
line, there was no turn-over of (À)-sparteine and no background
lithiation: adduct (R)-10 was formed in 22% yield and 87:13 er (Ta-
ble 2, entry 4). As with the s-BuLi results in Table 1, the highest
yield and enantioselectivity with n-BuLi were obtained using LiD-
MAE in two-ligand catalysis. In this way, adduct (R)-10 was iso-
lated in 48% yield and with 82:18 er (Table 2, entry 3). This
result indicates some turn-over of (À)-sparteine and the loss of
er can be attributed to some background lithiation by n-BuLi/LiD-
MAE which was confirmed in a separate experiment. Lithiation
trapping of phosphine borane 7 with n-BuLi/LiDMAE alone in
Et2O at À78 °C gave rac-10 in 13% yield.
3. Conclusion
In conclusion, we have provided definitive proof that two-ligand
catalytic asymmetric deprotonation of phosphine borane 7 using s-
BuLi gives higher enantioselectivity than the corresponding one-li-
gand catalysis. The best result with s-BuLi was obtained using LiD-
MAE as the second ligand (59% yield of (R)-10 with 81:19 er)
although bispidine 6 also worked well (78% yield of (R)-10 with
Given the rather disappointing results with the one-ligand and
two-ligand catalytic asymmetric deprotonation of phosphine bor-

2410
S. J. Canipa et al. / Tetrahedron: Asymmetry 20 (2009) 2407–2412
79:21 er). With n-BuLi at À78 °C, one-ligand catalysis was unsuc-
cessful as (À)-sparteine was not turned over and two-ligand catal-
ysis only worked in the presence of LiDMAE: adduct (R)-10 of 82:18
er was isolated in 48% yield. Alternatively, one-ligand catalysis with
n-BuLi in the temperature range of À60 to À40 °C provided satisfac-
tory results: at À50 °C, alcohol (S)-9 of 81:19 er was formed in 50%
yield. If a simple catalytic variant is to be employed for the depro-
tonation of phosphine boranes such as 7, we advocate the use of
n-BuLi in a one-ligand system at À50 °C or use of s-BuLi in a two-li-
gand system with bispidine 6 or LiDMAE (formed from commer-
cially available DMAE). We believe that these catalytic protocols
are useful for the synthesis of chiral phosphine ligands since a num-
ber of such ligands have been prepared via asymmetric deprotona-
tion using a stoichiometric amount of (À)-sparteine.16
(3 mL) at À78 or À40 °C under Ar. After stirring for 15 min at
À78 or À40 °C, solution of phosphine borane (80 mg,
a
7
0.61 mmol) in Et2O (3 mL) was added dropwise over 10 min via a
cannula. The resulting solution was stirred at À78 or À40 °C for
3 h. Then, a solution of benzophenone (1.1 equiv) in Et2O (3 mL)
was added via a syringe and the resulting solution was allowed
to warm to rt over 16 h. 1 M HCl(aq) (5 mL) and then EtOAc
(5 mL) were added. The two layers were separated and the aqueous
layer was extracted with EtOAc (3 Â 5 mL). The combined organic
layers were washed with 1 M HCl(aq) (5 mL), water (5 mL) and
brine (5 mL), dried (MgSO4) and evaporated under reduced pres-
sure to give the crude product as a white solid.
4.4. General procedure C: one-ligand catalytic asymmetric
deprotonation of phosphine borane 7 using n-BuLi/(À)-
sparteine
4. Experimental
4.1. General
n-BuLi (2.4 M solution in hexane, 1.1 equiv) was added drop-
wise to a stirred solution of (À)-sparteine (0.2 equiv) in Et2O
(10 mL) at À78, À60, À50 or À40 °C under Ar. After stirring for
15 min at À78, À60, À50 or À40 °C, a solution of phosphine borane
7 (290 mg, 2.20 mmol) in Et2O (5 mL) was added dropwise over
10 min via a cannula. The resulting solution was stirred at À78,
À60, À50 or À40 °C for 3 h. Then, a solution of benzophenone
(1.1 equiv) in Et2O (5 mL) was added via a syringe and the resulting
solution was allowed to warm to rt over 16 h. Next, 1 M HCl(aq)
(10 mL) and then EtOAc (10 mL) were added. The two layers were
separated and the aqueous layer was extracted with EtOAc
(3 Â 10 mL). The combined organic layers were washed with 1 M
HCl(aq) (10 mL), water (10 mL) and brine (10 mL), dried (MgSO4)
and evaporated under reduced pressure to give the crude product
as a yellow oil.
Water used is distilled water. Et2O was freshly distilled from
benzophenone ketyl. Brine is a saturated aqueous solution. (À)-
Sparteine, bispidine 6, DMAE and PMDETA were distilled from
CaH2 before use. Petrol refers to the fraction of petroleum ether
with a boiling range of 40–60 °C. All reactions were carried out un-
der O2-free Ar using oven-dried and/or flame-dried glassware.
Flash column chromatography was carried out using Fluka Silica
Gel 60 (0.035–0.070 mm particle size). Thin layer chromatography
was carried out using Merck F254 alumina-backed silica plates. Pro-
ton (400 MHz) and carbon (100.6 MHz) NMR spectra were re-
corded on a Jeol ECX-400 instrument with internal deuterium
lock. Chemical shifts are quoted as parts per million and referenced
to CHCl3 (7.27). Carbon NMR spectra were recorded with broad-
band proton decoupling and were assigned using DEPT experi-
ments. Infra-red spectra were recorded on an ATI Matteson
Genesis FT-IR spectrometer. Optical rotations were recorded at
room temperature (20 °C) on a Jasco DIP-370 polarimeter (using
4.5. General procedure D: background lithiation of phosphine
borane 7 using n-BuLi
At first, n-BuLi (2.4 M solution in hexane, 1.1 eq) was added
dropwise to a stirred solution of phosphine borane 7 (80 mg,
0.61 mmol) in Et2O (3 mL) at À78, À60, À50 or À40 °C under Ar.
The resulting solution was stirred at À78, À60, À50 or À40 °C for
3 h. Then, a solution of benzophenone (1.1 equiv) in Et2O (3 mL)
was added via a syringe and the resulting solution was allowed
to warm to rt over 16 h. Next, 1 M HCl(aq) (5 mL) and then EtOAc
(5 mL) were added. The two layers were separated and the aqueous
layer was extracted with EtOAc (3 Â 5 mL). The combined organic
layers were washed with 1 M HCl(aq) (5 mL), water (5 mL) and
brine (5 mL), dried (MgSO4) and evaporated under reduced pres-
sure to give the crude product as a white solid.
the sodium D line; 259 nm) and [a] measurements are given in
D
units of 10À1 deg cm2 gÀ1. Melting points were measured on a Gal-
lenkamp melting point apparatus. Chiral stationary phase HPLC
was performed on an Agilent 1200 series instrument and a multi-
ple wavelength, UV/Vis diode array detector.
4.2. General procedure A: two-ligand catalytic asymmetric
deprotonation of phosphine borane 7
s-BuLi (1.2 M solution in cyclohexane, 1.1 or 2.3 equiv) or n-BuLi
(2.4 M solution in hexane, 1.1 or 2.3 equiv) was added dropwise to a
stirred solution of (À)-sparteine (0.2 equiv) and a second ligand
(1.2 equiv) in Et2O (10 mL) at À78 °C under Ar. After stirring for
15 min at À78 °C, a solution of phosphine borane 7 (290 mg,
2.20 mmol) in Et2O (5 mL) was added dropwise over 10 min via a can-
nula. The resulting solution was stirred at À78 °C for3 h. Then, Me2Ph-
SiCl (1.1 equiv or 2.3 equiv) was added via a syringe and the resulting
solution was allowed to warm to rt over 16 h. Next, 1 M HCl(aq)
(10 mL) and then EtOAc(10 mL)were added. The two layerswere sep-
arated and the aqueous layer was extracted with EtOAc (3 Â 10 mL).
The combined organic layers were washed with 1 M HCl(aq) (10 mL),
water (10 mL) and brine (10 mL), dried(MgSO4) and evaporated under
reduced pressure to give the crude product as a yellow oil.
4.6. (R)-tert-Butylmethylphosphineborane(methyl)-
dimethylphenylsilane (R)-10
Table 1, entry 2: Using general procedure A, (À)-sparteine
(97 mg, 0.41 mmol), bispidine 6 (522 mg, 2.48 mmol) and s-BuLi
(1.91 mL of a 1.19 M solution in cyclohexane, 2.28 mmol) in Et2O
(10 mL), phosphine borane 7 (273 mg, 2.07 mmol) in Et2O (5 mL)
and Me2PhSiCl (0.38 mL, 2.28 mmol) gave the crude product. Puri-
fication by flash chromatography on silica with petrol–Et2O (98:2)
as eluent gave adduct (R)-10 (432 mg, 78%, 79:21 er by chiral
HPLC) as a colourless oil which slowly crystallised to a white solid,
mp 40–42 °C; [
IR (NaCl) 2960, 2375, 1426, 1265, 1114, 1060, 904, 887, 822, 738,
704 cmÀ1 1H NMR (400 MHz, CDCl3) d: 7.59–7.54 (m, 2H, o-Ph),
7.40–7.38 (m, 3H, Ph), 1.12 (d, JPH = 13.5 Hz, 9H, PCMe3), 1.06–
0.97 (m, 2H, PCH2), 1.00 (d, JPH = 10.0 Hz, 3H, PMe), 0.56 (s, 3H,
SiMe), 0.50 (s, 3H, SiMe); 13C NMR (100.6 MHz, CDCl3) d: 138.4
a]D = +6.3 (c 1.10, CHCl3), RF (4:1 petrol–EtOAc) 0.3;
4.3. General procedure B: stoichiometric asymmetric
deprotonation of phosphine borane 7 using n-BuLi/(À)-sparteine
;
n-BuLi (2.4 M solution in hexane, 1.1 equiv) was added drop-
wise to a stirred solution of (À)-sparteine (1.2 equiv) in Et2O

S. J. Canipa et al. / Tetrahedron: Asymmetry 20 (2009) 2407–2412
2411
(d, JPC = 4.5 Hz, ipso-Ph), 133.5 (Ph), 129.4 (Ph), 128.0 (Ph), 28.3 (d,
JPC = 33.0 Hz, PCMe3), 24.5 (d, JPC = 2.5 Hz, PMe), 7.6 (d,
JPC = 21.5 Hz, PCH2), 7.2 (d, JPC = 7.0 Hz, PCMe3), À0.4 (d,
JPC = 2.5 Hz, SiMe), À1.5 (SiMe); MS (ESI) m/z 289 [(M+Na)+]; HRMS
(ESI) m/z calcd for C14H28BPSi (M+Na)+ 289.1683, found 289.1683
(0.1 ppm error). HPLC: Daicel Chiracel OD, 1:99 v/v iPrOH-hexane,
0.1 mL minÀ1, 254 nm, 52.7 min [(R)-10], 56.3 min [(S)-10].
Table 1, entry 3: Using general procedure A, (À)-sparteine
(91 mg, 0.39 mmol), dimethylaminoethanol (208 mg, 2.33 mmol)
silica with petrol–Et2O (98:2) as eluent gave adduct (R)-10
(131 mg, 22%, 87:13 er by chiral HPLC) as a colourless oil which
slowly crystallised to a white solid, [a]D = +4.8 (c 1.00, CHCl3).
Table 2, entry 5: Using general procedure A, (À)-sparteine
(91 mg, 0.39 mmol), PMDETA (404 mg, 2.33 mmol) and n-BuLi
(0.89 mL of a 2.40 M solution in cyclohexane, 2.14 mmol) in Et2O
(10 mL), phosphine borane 7 (256 mg, 1.94 mmol) in Et2O (5 mL)
and Me2PhSiCl (0.36 mL, 2.14 mmol) gave the crude product. Puri-
fication by flash chromatography on silica with petrol–Et2O (98:2)
as eluent gave adduct (R)-10 (427 mg, 83%, 51:49 er by chiral
HPLC) as a colourless oil which slowly crystallised to a white solid,
and s-BuLi (3.75 mL of
a
1.19 M solution in cyclohexane,
(256 mg,
4.46 mmol) in Et2O (10 mL), phosphine borane
7
1.94 mmol) in Et2O (5 mL) and Me2PhSiCl (0.75 mL, 4.46 mmol)
gave the crude product. Purification by flash chromatography on
silica with petrol–Et2O (98:2) as eluent gave adduct (R)-10
(304 mg, 59%, 81:19 er by chiral HPLC) as a colourless oil which
[a]D = +0.3 (c 1.05, CHCl3).
4.7. (S)-P-(2,2-Diphenyl-2-hydroxyethyl)-P-methyl-tert-
butylphosphine (S)-9
slowly crystallised to a white solid, [a]D = +6.3 (c 1.00, CHCl3).
Table 1, entry 4: Using general procedure A, (À)-sparteine
Table 3, entry 1: Using general procedure B, (À)-sparteine
(171 mg, 0.73 mmol) and n-BuLi (0.32 mL of a 2.10 M solution in
hexane, 0.67 mmol) in Et2O (2 mL), phosphine borane 7 (80 mg,
0.61 mmol) in Et2O (3 mL) and benzophenone (122 mg,
0.67 mmol) in Et2O (3 mL) at À78 °C gave the crude product. Puri-
fication by flash chromatography on silica with petrol–EtOAc
(19:1) as eluent gave adduct (S)-9 (106 mg, 56%, 88:12 er by chiral
(102 mg, 0.44 mmol), N-methyl morpholine (264 mg, 2.61 mmol)
and s-BuLi (2.01 mL of
2.39 mmol) in Et2O (10 mL), phosphine borane
a
1.19 M solution in cyclohexane,
(287 mg,
7
2.18 mmol) in Et2O (5 mL) and Me2PhSiCl (0.40 mL, 2.39 mmol)
gave the crude product. Purification by flash chromatography on
silica with petrol–Et2O (98:2) as eluent gave adduct (R)-10
(390 mg, 67%, 65:35 er by chiral HPLC) as a colourless oil which
HPLC) as
a white solid, [a]
D = +20.4 (c 1.05, CHCl3) {lit.,14b
slowly crystallised to a white solid, [
a
]
D = +3.2 (c 1.10, CHCl3).
[
a
]
D = À14.7 (c 0.47, CHCl3) for (R)-9 of 96:4 er}; RF (4:1 petrol–
Scheme 4, competition experiment: s-BuLi (0.98 mL of a 1.29 M
solution in cyclohexane, 1.26 mmol) was added dropwise to a stir-
red solution of (À)-sparteine (148 mg, 0.63 mmol) and (+)-sparte-
ine surrogate (123 mg, 0.63 mmol) in Et2O (2 mL) at À78 °C
under Ar. After stirring for 15 min at À78 °C, a solution of phos-
phine borane 7 (70 mg, 0.53 mmol) in Et2O (3 mL) was added drop-
wise over 10 min via a cannula. The resulting solution was stirred
at À78 °C for 3 h. Then, Me2PhSiCl (0.21 mL, 1.27 mmol) was added
via a syringe and the resulting solution was allowed to warm to rt
over 16 h. 1 M HCl(aq) (5 mL) and then EtOAc (5 mL) were added.
The two layers were separated and the aqueous layer was ex-
tracted with EtOAc (3 Â 5 mL). The combined organic layers were
washed with 1 M HCl(aq) (5 mL), water (5 mL) and brine (5 mL),
dried (MgSO4) and evaporated under reduced pressure to give
the crude product as a yellow oil. Purification by flash chromatog-
raphy on silica with petrol–Et2O (98:2) as eluent gave adduct (S)-
10 (79 mg, 56%, 74:26 er by chiral HPLC) as a colourless oil which
EtOAc) 0.3; 1H NMR (400 MHz, CDCl3) d: 7.51–7.46 (m, 4H, Ph),
7.35–7.29 (m, 4H, Ph), 7.26–7.20 (m, 2H, Ph), 4.58 (s, 1H, OH),
2.88 (app t, JPH = JHH = 14.5 Hz, 1H, PCHAHB), 2.67 (dd, JHH = 14.5 Hz,
JPH = 6.5 Hz, 1H, PCHAHB), 1.17 (d, JPH = 13.5 Hz, 9H, CMe3), 0.74 (d,
JPH = 10.0 Hz, 3H, PMe), 0.88–0.23 (m, 3H, BH3); 13C NMR
(100.6 MHz, CDCl3) d: 147.7 (d, JPC = 8.5 Hz, ipso-Ph), 145.3 (d,
JPC = 1.5 Hz, ipso-Ph), 128.3 (Ph), 128.2 (Ph), 127.2 (Ph), 125.3
(Ph), 34.2 (d, JPC = 28.5 Hz, PCH2), 28.0 (d, JPC = 36.0 Hz, CMe3),
24.7 (d, JPC = 2.5 Hz, CMe3), 6.5 (d, JPC = 34.5 Hz, PMe) (some aro-
matic signals not resolved and one carbon overlaps with CDCl3);
HPLC: Daicel Chiracel OD, 1:19 v/v iPrOH–hexane, 0.5 mL minÀ1
,
254 nm, 11.1 min [(R)-9], 13.5 min [(S)-9]. Spectroscopic data were
consistent with those reported in the literature.14b
Table 3, entry 2: Using general procedure C, (À)-sparteine
(107 mg, 0.46 mmol) and n-BuLi (1.06 mL of a 2.37 M solution in
hexane, 2.51 mmol) in Et2O (10 mL), phosphine borane
7
(301 mg, 2.28 mmol) in Et2O (5 mL) and benzophenone (457 mg,
2.51 mmol) in Et2O (5 mL) at À78 °C gave the crude product. Puri-
fication by flash chromatography on silica with petrol–EtOAc
(19:1) as eluent gave adduct (S)-9 (139 mg, 19%, 87:13 er by chiral
HPLC) as a white solid.
Table 3, entry 3: Using general procedure D, n-BuLi (0.27 mL of a
2.45 M solution in hexane, 0.67 mmol) in Et2O (3 mL), phosphine
borane 7 (80 mg, 0.61 mmol) in Et2O (2 mL) and benzophenone
(121 mg, 0.67 mmol) in Et2O (3 mL) at À78 °C gave the crude prod-
uct. No evidence of adduct rac-9 was observed in the 1H NMR spec-
trum of the crude product.
slowly crystallised to a white solid, [
a
]
D = À4.7 (c 1.05, CHCl3).
Table 2, entry 2: Using general procedureA, (À)-sparteine (116 mg,
0.50 mmol), bispidine 6 (622 mg, 2.96 mmol) and n-BuLi (1.13 mL of
a 2.40 M solution in cyclohexane, 2.72 mmol) in Et2O (10 mL), phos-
phine borane 7 (326 mg, 2.47 mmol) in Et2O (5 mL) and Me2PhSiCl
(0.46 mL, 2.72 mmol) gave the crude product. Purification by flash
chromatography on silica with petrol–Et2O (98:2) as eluent gave ad-
duct (R)-10 (344 mg, 52%, 59:41 er by chiral HPLC) as a colourless oil
which slowly crystallised to a white solid, [a]D = +2.1 (c 1.05, CHCl3).
Table 2, entry 3: Using general procedure A, (À)-sparteine
(107 mg, 0.46 mmol), dimethylaminoethanol (244 mg, 2.74 mmol)
Table 3, entry 4: Using general procedure C, (À)-sparteine
and n-BuLi (2.19 mL of
5.25 mmol) in Et2O (10 mL), phosphine borane
a
2.40 M solution in cyclohexane,
(301 mg,
(109 mg, 0.47 mmol) and n-BuLi (1.08 mL of a 2.37 M solution in
hexane, 2.56 mmol) in Et2O (10 mL), phosphine borane 7
7
2.28 mmol) in Et2O (5 mL) and Me2PhSiCl (0.88 mL, 5.25 mmol)
gave the crude product. Purification by flash chromatography on
silica with petrol–Et2O (98:2) as eluent gave adduct (R)-10
(290 mg, 48%, 82:18 er by chiral HPLC) as a colourless oil which
(307 mg, 2.33 mmol) in Et2O (5 mL) and benzophenone (467 mg,
2.56 mmol) in Et2O (5 mL) at À60 °C gave the crude product. Puri-
fication by flash chromatography on silica with petrol–EtOAc
(19:1) as eluent gave adduct (S)-9 (306 mg, 42%, 87:13 er by chiral
slowly crystallised to a white solid, [
a]
D = +6.5 (c 1.05, CHCl3).
HPLC) as a white solid, [a]D = +18.7 (c 1.00, CHCl3).
Table 2, entry 4: Using general procedure A, (À)-sparteine
Table 3, entry 5: Using general procedure D, n-BuLi (0.44 mL of a
2.10 M solution in hexane, 0.92 mmol) in Et2O (2 mL), phosphine
borane 7 (110 mg, 0.83 mmol) in Et2O (3 mL) and benzophenone
(167 mg, 0.92 mmol) in Et2O (3 mL) at À60 °C gave the crude prod-
uct as a white solid. No evidence of adduct rac-9 was observed in
the 1H NMR spectrum of the crude product.
(103 mg, 0.44 mmol), N-methyl morpholine (267 mg, 2.64 mmol)
and n-BuLi (1.01 mL of
2.42 mmol) in Et2O (10 mL), phosphine borane
a
2.40 M solution in cyclohexane,
(290 mg,
7
2.20 mmol) in Et2O (5 mL) and Me2PhSiCl (0.41 mL, 2.42 mmol)
gave the crude product. Purification by flash chromatography on

2412
S. J. Canipa et al. / Tetrahedron: Asymmetry 20 (2009) 2407–2412
Table 3, entry 6: Using general procedure C, (À)-sparteine
(121 mg, 0.52 mmol) and n-BuLi (1.20 mL of a 2.37 M solution in
hexane, 2.84 mmol) in Et2O (10 mL), phosphine borane
(340 mg, 2.58 mmol) in Et2O (5 mL) and benzophenone (517 mg,
2.84 mmol) in Et2O (5 mL) at À50 °C gave the crude product. Puri-
fication by flash chromatography on silica with petrol–EtOAc
(19:1) as eluent gave adduct (S)-9 (402 mg, 50%, 81:19 er by chiral
(135 mg, 0.74 mmol) in Et2O (3 mL) at À40 °C gave the crude prod-
uct. Purification by flash chromatography on silica with petrol–
EtOAc (19:1) as eluent gave adduct rac-9 (61 mg, 29%) as a white
solid.
7
Acknowledgements
HPLC) as a white solid, [
a]
D = +19.0 (c 1.10, CHCl3).
We thank the EPSRC and Hoffmann-La Roche for funding.
Table 3, entry 7: Using general procedure D, n-BuLi (0.39 mL of a
2.14 M solution in hexane, 0.83 mmol) in Et2O (2 mL), phosphine
borane 7 (100 mg, 0.76 mmol) in Et2O (3 mL) and benzophenone
(152 mg, 0.83 mmol) in Et2O (3 mL) at À50 °C gave the crude prod-
uct. Purification by flash chromatography on silica with petrol–
EtOAc (19:1) as eluent gave adduct rac-9 (27 mg, 11%) as a white
solid.
Table 3, entry 8: Using general procedure B, (À)-sparteine
(188 mg, 0.80 mmol) and n-BuLi (0.34 mL of a 2.14 M solution in
hexane, 0.73 mmol) in Et2O (2 mL), phosphine borane 7 (88 mg,
0.67 mmol) in Et2O (3 mL) and benzophenone (134 mg,
0.73 mmol) in Et2O (3 mL) at À40 °C gave the crude product. Puri-
fication by flash chromatography on silica with petrol–EtOAc
(19:1) as eluent gave adduct (S)-9 (177 mg, 84%, 84:16 er by chiral
References
1. Hoppe, D.; Hintze, F.; Tebben, P. Angew. Chem., Int. Ed. Engl. 1990, 29, 1422.
2. Beak, P.; Kerrick, S. T.; Wu, S.; Chu, J. J. Am. Chem. Soc. 1994, 116, 3231.
3. Hodgson, D. M.; Lee, G. P.; Marriott, R. E.; Thompson, A. J.; Wisedale, R.;
Witherington, J. J. Chem. Soc., Perkin Trans. 1 1998, 2151.
4. (a) Tsukazaki, M.; Tinkl, M.; Roglans, A.; Chapell, B. J.; Taylor, N. J.; Snieckus, V.
J. Am. Chem. Soc. 1996, 118, 685; (b) Metallinos, C.; Szillat, H.; Taylor, N. J.;
Snieckus, V. Adv. Catal. Synth. 2003, 345, 370.
5. Muci, A. R.; Campos, K. R.; Evans, D. A. J. Am. Chem. Soc. 1995, 117, 9075.
6. Genet, C.; Canipa, S. J.; O’Brien, P.; Taylor, S. J. Am. Chem. Soc. 2006, 128, 9336.
7. Gammon, J. J.; Canipa, S. J.; O’Brien, P.; Kelly, B.; Taylor, S. Chem. Commun. 2008,
3750.
8. McGrath, M. J.; O’Brien, P. J. Am. Chem. Soc. 2005, 127, 16378.
9. McGrath, M. J.; Bilke, J. L.; O’Brien, P. Chem. Commun. 2006, 2607.
10. Bilke, J. L.; O’Brien, P. J. Org. Chem. 2008, 73, 6452.
11. McGrath, M. J.; O’Brien, P. Synthesis 2006, 2233.
HPLC) as a white solid, [a]D = +13.8 (c 1.10, CHCl3).
12. Bilke, J. L.; Moore, S. P.; O’Brien, P.; Gilday, J. Org. Lett. 2009, 11, 1935.
13. For a review on catalytic asymmetric routes to P-stereogenic compounds, see:
Glueck, D. S. Chem. Eur. J. 2008, 14, 7108.
14. (a) Johansson, M. J.; Schwartz, L. O.; Amedjkouh, M.; Kann, N. C. Eur. J. Org.
Chem. 2004, 1894; (b) Johansson, M. J.; Schwartz, L. O.; Amedjkouh, M.; Kann,
N. Tetrahedron: Asymmetry 2004, 15, 3531.
Table 3, entry 9: Using general procedure C, (À)-sparteine
(83 mg, 0.36 mmol) and n-BuLi (0.93 mL of a 2.16 M solution in
hexane, 2.00 mmol) in Et2O (10 mL), phosphine borane
7
(240 mg, 1.82 mmol) in Et2O (5 mL) and benzophenone (364 mg,
2.00 mmol) in Et2O (5 mL) at À40 °C gave the crude product. Puri-
fication by flash chromatography on silica with petrol–EtOAc
(19:1) as eluent gave adduct (S)-9 (396 mg, 69%, 78:22 er by chiral
15. (a) Dearden, M. J.; Firkin, C. R.; Hermet, J.-P. R.; O’Brien, P. J. Am. Chem. Soc.
2002, 124, 11870; (b) O’Brien, P. Chem. Commun. 2008, 655.
16. (a) Imamoto, T.; Watanabe, J.; Wada, Y.; Masuda, H.; Yamada, H.; Tsuruta, H.;
Matsukawa, S.; Yamaguchi, K. J. Am. Chem. Soc. 1998, 120, 1635; (b) Gridnev, I.
D.; Yamanoi, Y.; Higashi, N.; Tsuruta, H.; Yasutake, M.; Imamoto, T. Adv. Synth.
Catal. 2001, 343, 118; (c) Yamanoi, Y.; Imamoto, T. J. Org. Chem. 1999, 64, 2988;
(d) Johansson, M. J.; Andersson, K. H. O.; Kann, N. J. Org. Chem. 2008, 73, 4458;
(e) Imamoto, T.; Sugita, K.; Yoshida, K. J. Am. Chem. Soc. 2005, 127, 11934.
HPLC) as a white solid, [a]D = +8.5 (c 1.10, CHCl3).
Table 3, entry 10: Using general procedure D, n-BuLi (0.35 mL of
a 2.14 M solution in hexane, 0.74 mmol) in Et2O (3 mL) phosphine
borane 7 (89 mg, 0.68 mmol) in Et2O (2 mL) and benzophenone