On the two-ligand catalytic asymmetric deprotonation of N-Boc pyrrolidine: Probing the effect of the stoichiometric ligand

Article abstract of DOI:10.1021/jo8010655

(Chemical Equation Presented) To map out the stoichiometric ligand requirements in the two-ligand catalytic asymmetric deprotonation of N-Boc pyrrolidine, 24 different ligands have been evaluated; the highest enantioselectivity (90:10 er) was obtained by using s-BuLi in the presence of 0.3 equiv of (-)-sparteine and 1.3 equiv of a cyclohexanediamine-derived ligand.

Full text of DOI:10.1021/jo8010655

SCHEME 1. Two-Ligand Catalytic Asymmetric
Deprotonation of N-Boc Pyrrolidine 1, using s-BuLi/
(-)-Sparteine or (+)-Sparteine Surrogate 3
On the Two-Ligand Catalytic Asymmetric
Deprotonation of N-Boc Pyrrolidine: Probing the
Effect of the Stoichiometric Ligand†
Julia L. Bilke and Peter O’Brien*
Department of Chemistry, UniVersity of York, Heslington,
York, YO10 5DD, United Kingdom
paob1@york.ac.uk
ReceiVed May 16, 2008
SCHEME 2. Two-Ligand Catalytic Asymmetric
Deprotonation of N-Boc Pyrrolidine 1
To map out the stoichiometric ligand requirements in the
two-ligand catalytic asymmetric deprotonation of N-Boc
pyrrolidine, 24 different ligands have been evaluated; the
highest enantioselectivity (90:10 er) was obtained by using
s-BuLi in the presence of 0.3 equiv of (-)-sparteine and 1.3
equiv of a cyclohexanediamine-derived ligand.
yield of silylated adduct (S)-2 of 90:10 er (Scheme 1).7 A similar
result, with the opposite sense of induction, was obtained by
using the (+)-sparteine surrogate 3. If bispidine 4 was omitted,
there was no turnover of the (-)-sparteine and low yield and
reduced enantioselectivity resulted. This two-ligand strategy was
also successful in Hoppe’s O-alkyl carbamate methodology.9
Bispidine 4 was designed with steric hindrance on the amines
so that its s-BuLi complex would have a low reactivity thus
minimizing background racemic deprotonation.8 To establish
whether such a design criteria was important and to map out
the structural requirements of the stoichiometric ligand, we have
now carried out an extensive ligand variation study. For the
synthesis of the ligands, our criteria were that the synthesis
should be no more than three steps, it must allow multigram
quantities to be prepared, and there should be only one
purification at the end of the synthesis. Herein, we report the
results of an investigation of 24 different stoichiometric ligands
in the s-BuLi-mediated two-ligand asymmetric deprotonation
of N-Boc pyrrolidine 1.
To evaluate the different stoichiometric ligands, the following
conditions were adopted for the deprotonation of N-Boc
pyrrolidine 1: 1.6 equiv of s-BuLi, 0.3 equiv of (-)-sparteine,
and 1.3 equiv of the stoichiometric ligand (either achiral or
racemic). Using bispidine 4 under these conditions gave a 70%
yield of silylated adduct (S)-2 of 95:5 er (Scheme 2), the same
enantioselectivity as we routinely obtain using stoichiometric
(-)-sparteine. This was the benchmark result and we hoped to
identify other ligands that would produce such high enantiose-
lectivity. We designed a series of more sterically hindered
analogues of commonly encountered ligands for organolithium-
mediated deprotonation reactions.
The asymmetric deprotonation of N-Boc pyrrolidine 1 with
s-BuLi/(-)-sparteine, first reported by Kerrick and Beak in
1991,1 is not only a landmark achievement in chiral base
methodology but has also proved to be a very useful approach
for the direct synthesis of chiral pyrrolidines.1b,2 Our group has
a long-standing interest in Beak’s N-Boc pyrrolidine methodol-
ogy: we have used it to evaluate the efficacy of (+)-sparteine
surrogates3–6 and in total synthesis.6 Recently, we reported a
two-ligand catalytic asymmetric deprotonation of N-Boc pyr-
rolidine 1.7,8 Thus, deprotonation of 1 was achieved by using
1.3 equiv of s-BuLi, 0.2 equiv of (-)-sparteine, and 1.2 equiv
of bispidine 4; subsequent trapping with Me3SiCl gave a 76%
† This paper is dedicated to the memory of Professor A. I. Meyers for his
numerous pioneering contributions to organolithium-mediated deprotonation
reactions.
(1) (a) Kerrick, S. T.; Beak, P. J. Am. Chem. Soc. 1991, 113, 9708. (b) Beak,
P.; Kerrick, S. T.; Wu, S.; Chu, J. J. Am. Chem. Soc. 1994, 116, 3231.
(2) (a) Campos, K. R.; Klapars, A.; Waldman, J. H.; Dormer, P. G.; Chen,
C.-Y. J. Am. Chem. Soc. 2006, 128, 3538. (b) Dieter, R. K.; Chen, N. J. Org.
Chem. 2006, 71, 5674. (c) Deng, X.; Neelakandha, S. M. Tetrahedron: Asymmetry
2005, 16, 661. (d) Dieter, R. K.; Chen, N.; Watson, R. T. Tetrahedron 2005,
61, 3221. (e) Watson, R. T.; Gore, V. K.; Chandupatla, K. R.; Dieter, R. K.;
Snyder, J. P. J. Org. Chem. 2004, 69, 6105. (f) Majewski, M.; Shao, J.; Nelson,
K.; Nowak, P.; Irvine, N. M. Tetrahedron Lett. 1998, 39, 6787.
(3) Dearden, M. J.; Firkin, C. R.; Hermet, J.-P. R.; O’Brien, P. J. Am. Chem.
Soc. 2002, 124, 11870.
(4) O’Brien, P.; Wiberg, K. B.; Bailey, W. F.; Hermet, J.-P. R.; McGrath,
M. J. J. Am. Chem. Soc. 2004, 126, 15480.
(5) O’Brien, P. Chem Commun. 2008, 655.
(6) Stead, D.; O’Brien, P.; Sanderson, A. Org. Lett. 2008, 10, 1409.
(7) McGrath, M. J.; O’Brien, P. J. Am. Chem. Soc. 2005, 127, 16378.
(8) McGrath, M. J.; Bilke, J.; O’Brien, P. Chem. Commun. 2006, 2607.
(9) McGrath, M. J.; O’Brien, P. Synthesis 2006, 2233.
6452 J. Org. Chem. 2008, 73, 6452–6454
10.1021/jo8010655 CCC: $40.75 2008 American Chemical Society
Published on Web 07/19/2008

FIGURE 1. Results for the two-ligand catalytic asymmetric deprotonation of N-Boc pyrrolidine 1.
These included analogues of tetramethylethylenediamine
(TMEDA), tetramethyl-1,3-propanediamine (TMPDA), and
pentamethyldiethylenetriamine (PMDETA) as well as TMEDA-
TMPDA hybrid ligands containing a homopiperazine motif
(Figure 1). The syntheses of all of the ligands 5-25 are
described in the Supporting Information. The ligands were
evaluated by using the conditions shown in Scheme 2 (1.3 equiv
of ligand in place of bispidine 4) and the full results are shown
in Figure 1. The yields are those obtained after purification by
chromatography and the er values were determined by using
chiral GC. In three examples, we also carried out the catalytic
reactions using the (+)-sparteine surrogate 3.
By using TMEDA and amino ether 5, racemic silylated
adduct 2 was generated and this reflects the rapid lithiation of
N-Boc pyrrolidine 1 by their s-BuLi complexes.10 Beak had
previously noted a similar result using TMEDA and substo-
ichiometric amounts of (-)-sparteine.1b By increasing the steric
hindrance of the N-substituents (ligands 6-10), better enanti-
oselectivity was obtained (up to 86:14 er with (-)-sparteine).
Broadly speaking, the degree of enantioselectivity mirrored the
steric hindrance of the ligand but, perhaps surprisingly, the
highest er (86:14 er of (S)-2 with ligand 10) was not obtained
with the most sterically hindered ligands 7-9. Similarly, with
morpholines 11 and 13, better results were obtained with the
less sterically hindered ligand 13 (80:20 er). The 80:20 er
observed with ligand 13 is the highest enantioselectivity we have
obtained with a commercially available ligand. In contrast, N,N′-
dimethylpiperazine 12 was a poor stoichiometric ligand, giving
low yield (31%) and enantioselectivity (68:32 er). Investigations
into the use of diamines based on a racemic trans-cyclohex-
anediamine scaffold 14-16 led to the discovery of our best
stoichiometric ligand: rac-16 gave (S)-2 of 90:10 er in 61%
yield.
Collum and co-workers had previously reported that n-BuLi/
TMPDA was less reactive than n-BuLi/TMEDA in the R-lithia-
tion of an imine.11 Our catalytic results with s-BuLi/TMPDA
are consistent with a similar rate difference for the deprotonation
of N-Boc pyrrolidine 1: using TMPDA, (S)-2 of 66:34 er was
obtained (compared to rac-2 with TMEDA). Therefore, we
explored more sterically hindered TMPDA analogues 17-20
with moderate success (up to 79:21 er with ligand 20) and
TMEDA-TMPDA hybrids 21-24 with more success. The
catalytic results with N,N′-dimethylhomopiperazine ligand 21
led to the generation of rac-2 suggesting that it is more TMEDA-
like than TMPDA-like. In line with this, increasing the steric
hindrance in the homopiperazines 22-24 did lead to improved
catalytic results (up to 86:14 er with (-)-sparteine).
Finally, we briefly investigated the use of the triamines
PMDETA and ligand 25, a more sterically encumbered ana-
logue. In these cases, the (-)-sparteine catalytic results were
somewhat disappointing and the best enantioselectivity was 68:
32 er with (-)-sparteine. For three of the ligands rac-16, 24,
and PMDETA, the (+)-sparteine surrogate 3 was compared with
(11) Rutherford, J. L.; Hoffmann, D.; Collum, D. B. J. Am. Chem. Soc. 2002,
124, 264.
(10) Collum, D. B. Acc. Chem. Res. 1992, 25, 448.
J. Org. Chem. Vol. 73, No. 16, 2008 6453

SCHEME 3. Synthesis of the Optimum Stoichiometric
Ligands 10, rac-16, and 24
In summary, a study of 24 different stoichiometric ligands
in the two-ligand catalytic asymmetric deprotonation of N-Boc
pyrrolidine 1 has identified three satisfactory TMEDA-like
diamine ligands: 10, rac-16, and 24. These ligands are easy to
synthesize on a multigram scale and the best result (61% yield,
90:10 er) was obtained with cyclohexanediamine-derived ligand
rac-16. Surprisingly, use of the most sterically hindered ligands
such as 7, 8, 9, 19, and 23 did not lead to the highest
enantioselectivity indicating the subtle and unpredictable nature
of steric effects in organolithium-mediated deprotonation
processes.
Experimental Section
General Procedure for the Catalytic Asymmetric Deproto-
nation of N-Boc Pyrrolidine 1. A solution of (-)-sparteine or the
(+)-sparteine surrogate 3 (0.6 mmol, 0.3 equiv) in Et2O (1 mL)
was added dropwise to a stirred solution of s-BuLi (2.7 mL of a
1.2 M solution in cyclohexane, 3.2 mmol, 1.6 equiv) in Et2O (5
mL) at -78 °C under Ar. Then, a solution of the stoichiometric
ligand (TMEDA, TMPDA, PMDETA or ligands 4-25) (2.6 mmol,
1.3 equiv) in Et2O (1 mL) was added dropwise. After being stirred
for 10 min, a solution of N-Boc pyrrolidine 1 (347 mg, 2.0 mmol,
1.0 equiv) in Et2O (1 mL) was added dropwise via a cannula over
10 min. The resulting solution was stirred at -78 °C for 4 h. Then,
Me3SiCl (0.4 mL, 3.2 mmol, 1.6 equiv) was added dropwise and
the reaction mixture was allowed to warm to rt over 4 h and stirred
at rt for 12 h. H3PO4(aq) (5%, 10 mL) was added and the reaction
mixture was extracted with Et2O (3 × 10 mL). The combined Et2O
extracts were dried (MgSO4) and evaporated under reduced pressure
to give the crude product. Purification by flash column chroma-
tography with 19:1 petrol-Et2O as eluent gave the pure adduct
(S)- or (R)-2 as a colorless oil (chiral GC 30 m × 0.25 mm i.d.
(ꢀ-cyclodextrin), T ) 91 °C isothermal, He carrier gas at 12 psi
constant pressure). Spectroscopic data are consistent with those
reported in the literature.1b
(-)-sparteine in the catalytic reactions. In each case, higher
enantioselectivity was obtained with diamine 3 and the largest
difference was noted with PMDETA: 68:32 er with (-)-
sparteine and 81:19 er with diamine 3. These results fit our
previous suggestion8 that the higher reactivity of the s-BuLi/
diamine 3 complex facilitates more efficient turnover in the two-
ligand catalytic process and hence results in improved enanti-
oselectivity compared to (-)-sparteine.
Our studies have revealed that diamine ligands 10 (86:14 er),
rac-16 (90:10 er), and 24 (86:14 er) are the optimum TMEDA-
like stochiometric ligands for the two-ligand catalytic asym-
metric deprotonation of N-Boc pyrrolidine 1. These diamines
are easy to synthesize and their synthetic routes are summarized
in Scheme 3. Diamine 1012 (97% yield) was prepared in one
step from diamine 26 via Eschweiler-Clarke methylation. For
the synthesis of diamine rac-16, trans-cyclohexanediamine rac-
27 was first converted into diamine rac-28 (via bis-methyl
carbamate formation and LiAlH4 reduction). Then, reductive
amination generated diamine rac-1613 in 62% yield over 3 steps.
A two-step route comprising acylation and LiAlH4 reduction
was used to synthesize diamine 24 from homopiperazine 29
(85% yield over 2 steps).
Acknowledgment. We thank The Leverhulme Trust for
funding (J.L.B.).
Supporting Information Available: Full experimental pro-
cedures, characterization data, and copies of 1H/13C NMR
spectra of novel compounds. This material is available free of
charge via the Internet at http://pubs.acs.org.
(12) Mahadevan, V.; Henson, M. J.; Solomon, E. I.; Stack, T. D. P. J. Am.
Chem. Soc. 2000, 122, 10249.
(13) Kizirian, J.-C.; Cabelo, N.; Pinchard, L.; Caille, J.-C.; Alexakis, A.
Tetrahedron 2005, 61, 8939.
JO8010655
6454 J. Org. Chem. Vol. 73, No. 16, 2008