Application of polydentate chiral amines within magnesium-mediated asymmetric deprotonation reactions

Article abstract of DOI:10.1016/j.tetlet.2004.03.113

A set of polydentate secondary amines, each containing two stereogenic centres, were prepared. These were subsequently utilised in magnesium-mediated asymmetric deprotonation reactions and excellent enantiomeric ratios were obtained (up to 94:6 er). Furth

Full text of DOI:10.1016/j.tetlet.2004.03.113

Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 4175–4179
Application of polydentate chiral amines within magnesium-
mediated asymmetric deprotonation reactions
Martin J. Bassindale,a James J. Crawford,a Kenneth W. Hendersonb,*
and William J. Kerra,*
aDepartment of Pure and Applied Chemistry, University of Strathclyde, 295 Cathedral Street, Glasgow G1 1XL, Scotland, UK
bDepartment of Chemistry and Biochemistry, University of Notre Dame, 251 Nieuwland Science Hall, Notre Dame, IN 46556, USA
Received 27 February 2004; accepted 19 March 2004
Abstract—A set of polydentate secondary amines, each containing two stereogenic centres, were prepared. These were subsequently
utilised in magnesium-mediated asymmetric deprotonation reactions and excellent enantiomeric ratios were obtained (up to 94:6 er).
Furthermore, these bases tend to be highly efficient in the absence of strong Lewis base additives, which demonstrates an additional
practical advantage over existing protocols.
ꢀ 2004 Elsevier Ltd. All rights reserved.
Chiral magnesium amides have been shown to be
effective reagents in asymmetric synthesis both as
homo-1;2 and heteroleptic species.3 Our recent studies
into enantioselective deprotonations have, for the most
part, been carried out using the structurally simple and
commercially available (R)-N-benzyl-a-methylbenzyl-
amine (R)-1 as its respective magnesium bisamide (R)-2
(Fig. 1). Having stated this and in a more general sense,
the substantial literature on the lithium amide counter-
parts4 has shown that many of the most selective sys-
tems utilise more complex ligands, which commonly
incorporate side arms capable of chelating to the metal
centre.5 With this in mind, we decided to prepare a set of
polydentate amines for use in our Mg-mediated enan-
tioselective deprotonation processes.
There are three key points to our ligand design: (i) each
ligand has the ability to form a stable five-membered
chelate ring (chelation model 3 shown in Fig. 1); (ii) a
subset of the amides contain yet another heteroatom as
a potential donor site; and (iii) each amine features two
stereocentres in order to investigate the effect of the
second stereogenic centre upon the selectivity of their
respective magnesium bisamides. In this regard we tar-
geted the set of six amines shown in Figure 2.
Ph
N
H
Ph
(R)-1
R2
Y
1
R
N
1
Mg
Ph
N
Ph Mg
2
N
R
R2
Y
3
(R)-2
Figure 1.
Keywords: Asymmetric synthesis; Chiral amines; Enantioselective
deprotonation; Magnesium amides.
* Corresponding authors. Tel.: +1-574-631-8025; fax: +1-574-631-6652
(K.W.H); tel.: +44-141-548-2959; fax: +44-141-548-4246 (W.J.K.);
e-mail addresses: khenders@nd.edu; w.kerr@strath.ac.uk
Figure 2.
0040-4039/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2004.03.113

4176
M. J. Bassindale et al. / Tetrahedron Letters 45 (2004) 4175–4179
The original preparation, and hence application, of
amines of this type has suffered from convoluted syn-
thetic routes involving four or more steps from amino
acid starting materials.5a;6 However, alternatives to the
initially published routes have appeared recently7 and
the ‘one-step’ synthesis from O’Brien and co-workers
has been particularly notable.7c More specifically, this
route starts from optically enriched styrene oxide 10,
which is commercially available, albeit around 1000-fold
more expensive than in its racemic form. Alternatively,
optically enriched styrene oxide may be obtained
through the efficient catalytic hydrolytic kinetic resolu-
tion method of Jacobsen and co-workers.8 Through the
minor modification of doubling the catalyst loading (to
1.6%), in addition to increasing the quantity of the H2O
nucleophile (to 0.7 equiv), we found that (R)-10 is con-
sistently produced in excellent optical purity (>99% ee),
without loss of catalyst activity.
solvents and reagents, without additional purification,
afforded 77% and 80% yields of (R,R)-4 and (S,R)-5,
respectively, showing a 20% improvement over the
previous conditions.
With quantities of the desired set of amines in hand,
initial tests using magnesium bisamide derivatives could
be carried out. We chose the enantioselective deproto-
nation of the benchmark conformationally locked pro-
chiral ketone 4-tert-butylcyclohexanone 15 to form
optically enriched samples of trimethylsilyl enol ether
16. Bisamide formation was carried out using our
established method1 of heating 2 equiv of the amine
together with commercially available Bu2Mg 17 in THF
for 90 min (Scheme 2).
These bases were then used directly in deprotonation
reactions, using the previously optimised conditions for
our magnesium-mediated processes.1 The reactions were
carried out both in the absence and presence of the
Lewis base additives HMPA and DMPU, since we have
previously shown that these can have effects on both
reactivity and selectivity.10 Considering the magnesium
bisamide derived from the piperidine substituted amine
(R,R)-18 we see good conversions and excellent ers of
93:7 observed for all reaction runs (Table 1, entries 1–3).
These results are particularly encouraging in that they
represent the highest levels of enantioselection achieved
to date with this substrate within our magnesium-med-
iated systems.1;10 Moreover, these enhanced levels of
selectivity are achieved even in the absence of a donor
additive. It is also interesting to note that the opposite
With quantities of (R)-10 in hand the synthesis of the
first set of six amines was undertaken, following the
route proposed by O’Brien and co-workers.7c Upon
isolation and purification, pure amines 4–7 were
obtained in 53–60% overall yield,9 ready for application
in Mg-mediated asymmetric processes. Somewhat dis-
appointingly, using the same approach, the amine pos-
sessing the N-methylpiperazine unit (R,R)-8 was isolated
in a low 21% yield. Based on this, an alternative route to
the remaining desired N-methylated amine (S,R)-9 was
attempted. More specifically, using an N-formyl unit
(as a masked methyl group), (R)-10 was treated with
N-formylpiperazine 11, giving the requisite aminoalco-
hols, and the aziridinium ion (S)-12 upon mesylation.
Trapping with the amine nucleophile (S)-13 in a
biphasic mixture, but this time using slightly different
conditions, that is, with THF as co-solvent and an ex-
tended reaction time of 72 h, we obtained the N-formyl
amine (S,R)-14. Subsequent reduction with LiAlH4 gave
the pure targeted amine (S,R)-9 (74%; Scheme 1).
Mg
2
MgBu2
Ph NH
X
17
Ph
N
X
2
N
N
THF, , 90 min
∆
Ph
Ph
X=CH2 (R,R)-18, (S,R)-19
X=O (R,R)-20, (S,R)-21
X=NMe (R,R)-22, (S,R)-23
-2BuH
4-9
In order to test the generality of the latter set of con-
ditions for amine formation, the syntheses of the
piperidine-derived amines (R,R)-4 and (S,R)-5 were
repeated employing THF as a co-solvent and a longer
reaction time of 72 h. Using commercially available
Scheme 2.
Table 1. Enantioselective deprotonation of 15 with Mg-bisamides
18/19
Mg
CHO
CHO
N
N
X
OTMS
Ph
O
2
N
N
, EtOH,
∆
, 3 h
Ph
N
H
1.
O
N
11
(S)-12
TMSCl, additive,
THF, -78°C, 75 min
Ph
Ph
2. MsCl, Et3N, Et2O
t
t
Bu
Bu
(R)-10
Et3N, THF,
H2O, 72 h
15
(R)-16
Ph NH2
Entry
Mg-amide
X
Additive
(0.5 equiv)
Conv. er (R):(S)1
(%)
(S)-13
O
1
2
3
4
5
6
(R,R)-18
(R,R)-18
(R,R)-18
(S,R)-19
(S,R)-19
(S,R)-19
CH2
CH2
CH2
CH2
CH2
CH2
HMPA
DMPU
None
77
90
58
66
88
87
93:7
93:7
LiAlH4
Ph NH
N
Ph NH
Ph
N
N
H
N
93:7
∆
THF, , 2 h
Ph
HMPA
DMPU
None
87:13
88:12
88:12
(60%)
(S,R)-14
(S,R)-9
(74%)
Scheme 1.

M. J. Bassindale et al. / Tetrahedron Letters 45 (2004) 4175–4179
4177
Table 3. Enantioselective deprotonation of 4-substituted cyclohexa-
nones 24–27 with Mg-bisamide (R,R)-18
enantiomer of silyl enol ether (R)-16 was formed pref-
erentially here, compared with the (S)-enantiomer,
which is accessed in excess from the use of the previously
employed Mg-amide bases, for example, (R)-2.1;10 The
conversions obtained with the Mg-systems are also
worthy of note; in the presence of additives a conversion
of up to 90% was achieved (Table 1, entry 2)11 and,
importantly, in the absence of any additive the conver-
sion of 58% was observed (cf. the 33% obtained previ-
ously with our more simple Mg-bisamide system (R)-2
with no additive, over 6 h at ꢀ78 ꢁC).1 Additionally and
overall, these results also compare favourably with those
obtained using this substrate within the analogous chiral
lithium amide base area.4;5;12
Mg
Ph
N
2
OTMS
O
N
Ph
(R,R)-18
TMS-Cl, additive,
THF, -78°C, 75 min
R
R
(R)-28
24: R=Me
25: R=Ph
26: R=nPr
27: R=iPr
(R)-29
(R)-30
(R)-31
Entry
Ketone
Additive
(0.5 equiv)
Conv.
(%)
er (R):(S)1
1
2
3
4
5
6
7
8
24
24
25
25
26
26
27
27
DMPU
None
89
67
91
60
93
76
97
78
92:8
93:7
94:6
94:6
94:6
94:6
94:6
94:6
On examining the effect of changing to the (S,R)-dia-
stereomer of this piperidine-derived base (S,R)-19,
again, consistently high conversions were obtained and
most notably 87% in the absence of donor additive
(Table 1, entry 6). However, in each of these reactions a
small reduction in enantioselectivity was observed. It is
notable that the same (R)-enantiomer of product was
formed preferentially,1;10 indicating that this second
chiral centre has only a minor effect on the selectivity of
these bases.
DMPU
None
DMPU
None
DMPU
None
ers across the reactions attempted (Table 2, entries 7–
12). Interestingly, the enantioselection observed with
(R,R)-20 increased in a stepwise fashion, from having
HMPA to having no Lewis base additive (Table 2,
entries 7–9). Relatively good conversions were also
observed with, again, DMPU providing optimum effi-
ciency (up to 96%; Table 2, entries 8 and 11). Indeed,
this is in line with each of the bases studied (Tables 1 and
2). Therefore, it is clear that use of DMPU as additive
(and even, in some cases, no additive at all) is preferred
to the less acceptable HMPA.
Next, we wished to test whether the presence of an
additional heteroatom within the ligand would afford a
more selective system. The first species investigated were
the two N-methylpiperazine-derived Mg-amides, (R,R)-
22 and (S,R)-23. Good to excellent conversions were
again achieved in all cases, most notably in the presence
of 0.5 equiv of DMPU wherein up to 94% conversion
was obtained after 75 min (Table 2, entry 2). For the
most part, the ers obtained were comparable between
the two stereoisomers (except that shown in entry 6,
with no additive, wherein a lower er was observed).
Overall with (R,R)-22 and (S,R)-23, only slightly lower
levels of selection were obtained than with (R,R)-18,
with the optimum ers of 92:8 being obtained in the
presence of DMPU with either diastereomer (Table 2,
entries 2 and 5).
In order to initially assess the wider utility of these
chelating bases, the species found to give the highest
levels of enantioselection, Mg-bisamide (R,R)-18, was
tested on a series of 4-substituted cyclohexanones 24–27
(Table 3). This base was tested on each substrate both
with no additive and in the presence of DMPU. In each
case, excellent enantioselectivity was obtained. The ers
of the products formed were comparable across each
substrate and, pleasingly, there was, generally, a signif-
icant improvement in selectivity over our other Mg-
amide bases.1;10 Interestingly, the presence of DMPU as
additive seems to have very little, if any, effect on
selectivity, but affords a general increase in conversion
of around 20%. These promising results have prompted
further and ongoing investigations into the wider
application of this, and other, Mg-amide bases.
Moving to test the effect of changing to an oxygen
heteroatom within our ligand, we observed comparable
Table 2. Enantioselective deprotonation of 15 with Mg-bisamides
20–23
Entry
Mg-amide
X
Additive
(0.5 equiv)
Conv. er (R):(S)1
(%)
1
(R,R)-22
(R,R)-22
(R,R)-22
(S,R)-23
(S,R)-23
(S,R)-23
(R,R)-20
(R,R)-20
(R,R)-20
(S,R)-21
(S,R)-21
(S,R)-21
NMe HMPA
NMe DMPU
NMe None
NMe HMPA
NMe DMPU
NMe None
86
94
88
68
92
59
93
96
89
57
83
55
91:9
92:8
2
As part of efforts towards rationalising the various
effects of altering the ligand backbone, we carried out
ab initio molecular calculations on model systems at the
HF/6-31G* level of theory.13 The simple bisamide
complex 32 (Fig. 3), which contains a pair of N-methyl
amino anions and two 5-membered chelate rings,
geometry optimised to an energy minimum as shown.
This demonstrates that it is, indeed, possible to form a
chelate within such Mg-bisamide species, in the manner
proposed.
3
91:9
91:9
4
5
92:8
6
87:13
86:14
88:12
90:10
86:14
89:11
89:11
7
O
O
O
O
O
O
HMPA
DMPU
None
8
9
10
11
12
HMPA
DMPU
None

4178
M. J. Bassindale et al. / Tetrahedron Letters 45 (2004) 4175–4179
4. (a) Simpkins, N. S. Chem. Soc. Rev. 1990, 19, 335; (b) Cox,
P. J.; Simpkins, N. S. Tetrahedron: Asymmetry 1991, 2, 1;
(c) Koga, K. Pure Appl. Chem. 1994, 66, 1487.
5. (a) Shiria, S.; Sato, D.; Aoki, K.; Tanaka, M.; Kawasaki,
H.; Koga, K. Tetrahedron 1997, 53, 5963; (b) Aoki, K.;
Tomioka, K.; Noguchi, H.; Koga, K. Tetrahedron 1997,
53, 13641.
6. Shirai, R.; Aoki, K.; Sato, D.; Kim, H.-D.; Murakata, M.;
Yasukata, T.; Koga, K. Chem. Pharm. Bull. 1994, 42, 690.
7. (a) Saravanan, P.; Singh, V. K. Tetrahedron Lett. 1998, 39,
167; (b) de Sousa, S. E.; O’Brien, P.; Poumellec, P. J.
Chem. Soc., Perkin Trans. 1 1998, 1483; (c) Curthbertson,
E.; O’Brien, P.; Towers, T. D. Synthesis 2001, 693.
8. (a) Tokunaga, M.; Larrow, J. F.; Kakiuchi, F.; Jacobsen,
E. N. Science 1997, 277, 936; (b) Schaus, S. E.; Brandes, B.
D.; Larrow, J. F.; Tokunaga, M.; Hansen, K. B.; Gould,
A. E.; Furrow, M. E.; Jacobsen, E. N. J. Am. Chem. Soc.
2002, 124, 1307.
Figure 3.
Moving to examine the more complex system of a
potentially tridentate ligand (e.g., as would be possible
with the use of amines 6–9), we chose to incorporate
only one ligand onto the magnesium centre, to minimise
any possible steric influences. Based on this, model
complex 33 was constructed, which contains a mor-
pholine-based side arm and a methyl group as the sec-
ond anion on Mg (Fig. 3). Upon geometry optimisation
of 33 it was found that the additional oxygen chelation
is not possible, with the ligand preferring to undergo
donation exclusively through the nitrogen atom. Finally,
a comparative calculation was carried out using complex
34 to investigate the stability of the eight-membered ring
chelate (Fig. 3). Comparing isomers 33 and 34 we see an
increase in energy for the eight-membered chelate of
+9.98 kcal/mol, indicating that the former would be
significantly more stable.
9. All amines prepared exhibited satisfactory analytical and
spectral data.
ꢀ
ꢀ
10. Anderson, J. D.; Garcıa Garcıa, P.; Hayes, D.; Hender-
son, K. W.; Kerr, W. J.; Moir, J. H.; Fondekar, K. P.
Tetrahedron Lett. 2001, 42, 7111.
11. Representative experimental procedure: To a Schlenk
flask, under N2, was added Bu2Mg (0.97 M, 1.03 mL,
1.00 mmol, in heptanes) and the heptanes removed in
vacuo. A solution of (1R)-N-[(1R)-1-phenylethyl]-1-phen-
yl-2-(piperidin-1-yl)ethanamine
(R,R)-4
(617 mg,
2.00 mmol) in THF (10 mL) was then added. The solution
was heated to reflux for 90 min, under a nitrogen atmo-
sphere, then cooled to ꢀ78 ꢁC and then DMPU (60 lL,
0.5 mmol) and TMSCl (0.5 mL, 4 mmol) added. After
stirring for 20 min, a solution of 4-tert-butylcyclohexanone
15 (123 mg, 0.8 mmol, in 2 mL THF) was added over 1 h,
via a syringe pump. The resulting solution was allowed to
stir for a further 15 min at ꢀ78 ꢁC (overall reaction time
1.25 h) and was then quenched by the addition of
saturated aqueous NaHCO3 (5 mL). After warming to rt,
the reaction was extracted with Et2O (40 mL) and washed
with saturated aqueous NaHCO3 (2 · 20 mL). The com-
bined aqueous phase was then extracted with Et2O
(2 · 20 mL) and the combined organics dried over Na2SO4
and the solvent removed in vacuo to yield the crude
product. The reaction conversion was determined as 90%
by GC analysis [CP SIL 19CB fused silica capillary
column; carrier gas H2 (80 kPa); 45 ꢁC (1 min)–190 ꢁC;
temperature gradient: 45 ꢁC/min]; tR ¼ 3:27 min (15);
tR ¼ 3:68 min (16)]. The crude product was then purified
by flash silica column chromatography (eluting with
petroleum ether) to afford the desired product (R)-16 as
a clear, colourless oil (130.6 mg, 72%).1 The enantiomeric
ratio was determined by chiral GC analysis to be 7:93
((S):(R)) {Chirasil-DEX CB capillary column; carrier gas
H2 (80 kPa); 70 ꢁC (1 min)–130 ꢁC; temperature gradient:
1.7 ꢁC/min; tR ¼ 39:20 [(S)-16]; tR ¼ 39:53 min [(R)-16]}.
12. For a comparison with Li base systems, see: (a) Aoki, K.;
Koga, K. Chem. Pharm. Bull. 2000, 48, 571; (b) Graf,
C.-D.; Malan, C.; Knochel, P. Angew. Chem., Int. Ed.
1998, 37, 3014; (c) Graf, C.-D.; Malan, C.; Harms, K.;
Knochel, P. J. Org. Chem. 1999, 64, 5581; (d) Busch-
Petersen, J.; Corey, E. J. Tetrahedron Lett. 2000, 41, 6941.
13. (a) The Gaussian 98 series of programs were used for the
calculations: Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.;
Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.;
Zakrzewski, V. G.; Montgomery, J. A., Jr.; Stratmann,
R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels,
A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.;
Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli,
C.; Adamo, C.; Clifford, S.; Ochterski, J.; Petersson, G.
In summary, a range of chiral amines with the potential
for chelation were targeted and prepared. These were
subsequently tested as their Mg-bisamides within
asymmetric deprotonation reactions and found to be
effective chiral bases that display the highest general
selectivity yet found within these systems. Furthermore,
even in the absence of donor additives, high reactivities
and selectivities were retained. Computational studies
on model compounds support the formation of the
doubly chelated complexes containing a pair of five-
membered chelate rings. However, these calculations
indicate that tridentate chelation to a magnesium centre
is precluded due to conformational constraints within
the ligands and complexes studied.
Acknowledgements
We thank the University of Strathclyde for a University
Studentship (J.J.C.). We also thank the EPSRC for a
grant (GR/M12711; M.J.B.), and the EPSRC Mass
Spectrometry Service, University of Wales, Swansea, for
analyses.
References and notes
1. Henderson, K. W.; Kerr, W. J.; Moir, J. H. Tetrahedron
2002, 58, 4573.
2. Henderson, K. W.; Kerr, W. J. Chem. Eur. J. 2001, 7,
3430.
3. (a) Carswell, E. L.; Hayes, D.; Henderson, K. W.; Kerr,
W. J.; Russell, C. J. Synlett 2003, 1017; (b) Yong, K. H.;
Taylor, N. J.; Chong, J. M. Org. Lett. 2002, 4, 3553; (c)
Yong, K. H.; Chong, J. M. Org. Lett. 2002, 4, 4139.

M. J. Bassindale et al. / Tetrahedron Letters 45 (2004) 4175–4179
4179
A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.; Malick, D. K.;
Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.;
Cioslowski, J.; Ortiz, J. V.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.;
Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.;
Peng, C. Y.; Nanayakkara, A.; Gonzalez, C.; Challa-
combe, M.; Gill, P. M. W.; Johnson, B. G.; Chen, W.;
Wong, M. W.; Andres, J. L.; Head-Gordon, M.; Replogle,
E. S.; Pople, J. A. Gaussian 98, revision A.6; Gaussian,
Inc.: Pittsburgh, PA, 1998; Frequency analysis was used to
ensure that true minima were located. Absolute energies:
32, ꢀ811.2193; 33, ꢀ696.7553; 34, ꢀ696.7393 (Hartrees);
(b) Dill, J. D.; Pople, J. A. J. Chem. Phys. 1975, 62, 2921;
(c) Hariharan, P. C.; Pople, J. A. Theor. Chim. Acta 1973,
28, 213; (d) Hehre, W. J.; Ditchfield, R.; Pople, J. A. J.
Chem. Phys. 1972, 56, 2257.