A new class of Rh(III) catalyst containing an aminoalcohol tethered to a tetramethylcyclopentadienyl group for asymmetric transfer hydrogenation of ketones

Article abstract of DOI:10.1016/j.tetlet.2003.11.024

The synthesis and application to asymmetric reduction of acetophenone, of a novel class of Rh(III) catalyst containing a tether between the cyclopentadienyl group and a homochiral aminoalcohol, is described. The complex is a highly active catalyst for asymmetric ketone reduction, however it appears to be unstable to the extended reaction conditions. The well-defined stereochemical structure of the catalyst offers potential for significant improvement and 'fine tuning' towards specific substrates.

Full text of DOI:10.1016/j.tetlet.2003.11.024

Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 843–846
A new class of Rh(III) catalyst containing an aminoalcohol tethered
to a tetramethylcyclopentadienyl group for asymmetric transfer
hydrogenation of ketones
David J. Cross,a Ian Houson,b Aparecida M. Kawamotoa,ꢀ and Martin Willsa,*
aDepartment of Chemistry, The University of Warwick, Coventry CV4 7AL, UK
bAvecia Huddersfield Works, PO Box A38, Leeds Road, Huddersfield, Yorkshire HD2 1FF, UK
Received 12 September 2003; revised 24 October 2003; accepted 6 November 2003
Abstract—The synthesis and application to asymmetric reduction of acetophenone, of a novel class of Rh(III) catalyst containing a
tether between the cyclopentadienyl group and a homochiral aminoalcohol, is described. The complex is a highly active catalyst for
asymmetric ketone reduction, however it appears to be unstable to the extended reaction conditions. The well-defined stereo-
chemical structure of the catalyst offers potential for significant improvement and ꢁfine tuningꢂ towards specific substrates.
Ó 2003 Elsevier Ltd. All rights reserved.
During the course of a number of investigations into the
development of Rh(III) pentamethylcyclopentadienyl
catalysts for the asymmetric transfer hydrogenation of
ketones, we have focused on the use of homochiral 1,2-
aminoalcohols such as 1 and monotosylated diamines
such as 2 (Noyoriꢂs ligand) to control the enantioselec-
tivity of the reaction.1;2 An active catalyst is formed by
the combination of either ligand with [RhCl2(C5Me5)]2
in an inert solvent with a small amount of base. In this
process an 18-electron ꢁpre-catalystꢂ 3 is formed, which
undergoes subsequent elimination of HCl to form 4,
which in turn abstracts two hydrogen atoms from a
solvent molecule (isopropanol or formic acid depending
on the conditions employed) to give intermediate 5. The
transfer of these two hydrogen atoms to the substrate
results in formation of the alcohol product and regen-
eration of 4, thus completing a catalytic cycle.3
catalytic cycle. Through this modification, the cyclo-
pentadienyl ring would be unable to rotate and there
would now be a potential to introduce selectively
directing groups at specific positions. This would permit
the synthesis of ꢁfine-tunedꢂ reduction catalysts for use
with individual classes of substrate. In addition, the
ꢁthree-pointꢂ attachment of the ligand to the metal was
expected to improve the overall stability of the ligand.
OH
Ph
Ph
NHTs
NH2
NH2
1S,2R-1
RR-2
Rh
Rh
Rh
H
Although these complexes are capable of generating
very high asymmetric inductions in a range of reduction
processes, we believed that they could be modified in a
productive manner through the tethering of the ligand
components, which remain on the metal throughout the
H
Cl
X
X
X
H
H
ligand
backbone
N
N
N
H
H
5 X=O or NTs
4 X=O or NTs
3 X=O or NTs
Towards this end, we wished to establish methodology
for the synthesis of 6, a complex in which the cyclopen-
tadienyl group is attached to L-norephedrine, an amine,
which has been successfully employed as a ligand in this
application. X-ray crystallographic analyses of the
* Corresponding author. Tel.: +44-24-7652-3260; fax: +44-24-7652-
4112; e-mail: m.wills@warwick.ac.uk
ꢀ
On leave from the Brazilian Space Aeronautical Institute––Praca
Mal. Eduardo Gomes 50, Vila das Acacias, SJ Campos, SP, Brazil.
0040-4039/$ - see front matter Ó 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2003.11.024

844
D. J. Cross et al. / Tetrahedron Letters 45 (2004) 843–846
isoelectronic ruthenium(II) complexes 7 have revealed
that the favoured diastereoisomer is that in which the
phenyl group adjacent to the oxygen atom (in the ligand)
lies trans to the Ru–Cl bond in the complex equivalent to
3.2d;e;i An analogous structural relationship is found in
Rh(III) complexes of monotosylated diamine 2.2j;k
Through this diastereoisomeric complexation, the metal
is rendered chiral. Assuming that this stereochemistry is
retained throughout the catalytic cycle (as demonstrated
by Noyori for Ru(II) complexes),2i then the metal con-
figuration in turn controls the approach of the ketone to
the metal hydride species as depicted in Figure 1; the p-
interaction of the phenyl ring with the complexed ring
being a significant control element.4
Br
iii)
H2N
Me
OH
Ph
i) ii)
N
O
75%
42%
Me
8
Ph
HO
iv)
65%
NH OH
Me
N
O
Me
9
Ph
10
Ph
Rh
O
Although the tetramethylcyclopentadienyl group is dif-
ferent in structure from an arene ligand, the same p-face
stabilisation effect is known to operate through methyl
groups on the arene ligand, for example, in the cases of
p-cymene and hexamethylbenzene.4 On this basis, it
is likely that stereochemical control of ketone reduc-
tion using Cpꢀ/Rh(III) complexes proceeds in a similar
manner.
Cl
O
v)
99%
N
H
H
Ph
Me
H
6
11
Scheme 1. Reagents and conditions: (i) o-BrC6H4CH2Br, Et3N, DCM,
94%. (ii) (CH3)2C(OCH3)2, TsOH, 91%. (iii) t-BuLi, Et2O, )78 °C,
then 10, Et2O, )78 °C–rt, o/n. (iv) 2 M H2SO4, Et2O, rt, o/n. (v)
RhCl3ÆH2O, MeOH, reflux, 48 h.
We reasoned that the synthesis of complex 6, as the
single diastereoisomer illustrated, could be achieved
upon reaction of precursor 10 with 1 equiv of rho-
dium(III) trichloride. The synthesis of 6 was thus
completed as illustrated in Scheme 1.5 The reaction of
addition the complex was of the correct mass as deter-
mined by high resolution mass spectrometry. Since we
have not yet been able to confirm the stereochemistry of
6 by X-ray crystallography, we have assigned the rela-
tive stereochemistry on the basis of the structures of
ortho-bromobenzyl bromide with
L-(1R,2S)-norephe-
drine followed by treatment of the product with 2,2-
dimethoxypropane furnished the key intermediate 8 in
excellent yield. Bromine/lithium exchange in 8 upon
treatment with 1 equiv of t-BuLi at )78 °C was followed
by addition of 2,3,4,5-tetramethylcyclopentenone 11.
This resulted in the formation of 9 in reasonable yield as
a mixture of isomers. In the next step 9 was treated with
strong acid overnight, resulting in both hydrolysis of the
aminal ring and dehydration to form the cyclopenta-
diene group in 10. This intermediate exhibited a very
closely related complexes derived from
its derivatives.2
L-ephedrine and
Complex 6 proved to be an effective catalyst for the
asymmetric reduction of acetophenone in isopropanol, a
useful test substrate, which allows comparisons with
other systems (Scheme 2, Table 1).7 It also furnished a
reduction product of R configuration, which would be
predicted based on the stereochemical argument above,
with the assumption that the tethering does not
adversely effect the control by the ligand. A small
amount of base is required to activate the catalyst by
promoting the elimination of HCl in order to form a 16-
electron intermediate, which then enters the catalytic
cycle.
1
complex H NMR spectrum as would be expected due
to cyclopentadiene isomerisation. Upon treatment with
1 equiv of RhCl3 for 48 h the exchange reaction resulted
in direct formation of the required catalyst 6 in essen-
tially quantitative conversion.6
Complex 6, a red/brown crystalline solid, was charac-
1
Perhaps the most striking feature of the reduction
reactions, which were studied using a 0.1 M concentra-
tion of ketone at the 1 and 5 mol % catalyst levels, was
the very high activity demonstrated (selected results are
given in Table 1). At the 1 mol % level the reduction was
50% complete within 10 min, whilst at the 5 mol % level
the reaction was 95% complete after this time. Unfor-
tunately, at the lower catalyst loading, the reaction
terised by H NMR spectroscopy, which exhibited the
four singlet signals expected of four diastereomerically
distinct methyl groups on the cyclopentadienyl ring. In
Favourable
'edge-face'
π-interaction
Rh
Ru
Ru
HO
H
Cl
O
O
O
O
H
Cl
N
N
N
i)
Ph
Me
H
Ph
Me
H
H
O
R
Ph
Me
H
H
H
Me
Me
Me
H
H
H
6
7
Scheme 2. Reagents and conditions: (i) 1–5 mol % tethered catalyst 6,
2.5 mol % KOt-Bu, i-PrOH, 4 h, rt.
Figure 1.

D. J. Cross et al. / Tetrahedron Letters 45 (2004) 843–846
845
Table 1. Asymmetric reduction of acetophenone using 6
our racemisation test, which employed fresh catalyst, did
not fully reproduce the conditions at the end of the
reaction.
Catalyst
(mol %)
Time
(min)
Conversion
(%)
Enantiomeric
excess (%)
1
1
1
1
1
1
18.5
41.2
50.4
62.1
86.2
74.9
72.1
68.6
55.8
41.6
In conclusion, we have successfully completed the syn-
thesis and testing of the first example of a Rh(III) cat-
alyst for transfer hydrogenation in which both ligand
components are tethered. Whilst this is a highly active
catalyst for ketone reduction, and furnishes a product of
the predicted configuration (R) based on the design
principles, it does not remain stable under the reaction
conditions. Studies are currently underway to improve
the stability of the catalyst in order to deliver a finely-
tunable reagent for targeted use on specific substrates.
5
10
60
o/n
5
5
5
5
5
1
78.2
92.7
95.0
96.4
98.2
73.5
70.5
68.0
64.2
62.5
5
10
60
o/n
Conversions were followed by 1H NMR and enantiomeric excesses by
chiral HPLC.
Acknowledgements
failed to go to completion and the rate of conversion
dropped dramatically after the first few minutes. Even at
the higher catalyst loading the reaction failed to go fully
to completion, even when left overnight. In a control
We thank the EPSRC and Avecia for support of a
CASE studentship (D.J.C.). A.M.K. thanks the Brazil-
ian iAe for study leave. Professor D. Games and Dr. B.
Stein of the EPSRC National Mass Spectroscopic ser-
vice (Swansea) are thanked for HRMS analysis of cer-
tain compounds. We also acknowledge the generous
loan of ruthenium and rhodium salts by Johnson–
Matthey limited and the use of the EPSRC Chemical
Database Service at Daresbury.8
reaction, the combination of N-benzyl-L-ephedrine with
rhodium trichloride did not result in the formation of a
catalyst capable of acetophenone reduction. The use of
5 mol % of complex 6 in formic acid/triethylamine (o/n,
rt) resulted in the formation of racemic alcohol in 41%
conversion. This result was not unexpected, as amino
alcohol ligands are not generally compatible with formic
acid/triethylamine conditions.
The enantiomeric excesses also demonstrated an inter-
esting trend. After a promising high selectivity early in
the reaction, the ees dropped in the later stages of the
reaction. That this drop was not due to significant
reversibility of the reaction was demonstrated by treat-
ment of R-1-phenylethanol of 97% ee with 5 mol % of the
catalyst in isopropanol and 1 equiv of acetone to repro-
duce the environment generated at the end of the reac-
tion. The reaction mixture was followed at various times
up to 24 h after the addition of KOH and at no time was
the alcohol observed to racemise. Having ruled out sig-
nificant racemisation of the product by catalyst, we
reasoned that the loss of enantioselectivity may be
accounted for by the transient formation of a slow-
reacting decomposition product, which served to reduce
the remaining ketone in a racemic manner. This would
result in a lower overall observed enantioselectivity. Such
an intermediate could be 12, formed by breakage of the
O–Rh bond, which then subsequently decomposes
further to an unreactive material under the reaction
conditions. However the 5 mol % catalyst figures in
Table 1 suggest that there must also be some racemisa-
tion taking place (i.e., 1.4% conversion increase from 10
to 60 min would not give an ee drop of 3.8% if the new
product was entirely racemic). It is perhaps possible that
References and notes
1. (a) Palmer, M. J.; Wills, M. Tetrahedron: Asymmetry 1999,
10, 2045; (b) Noyori, R.; Hashiguchi, S. Acc. Chem. Res.
1997, 30, 97; (c) Wills, M.; Gamble, M.; Palmer, M. J.;
Smith, A. R. C.; Studley, J. R.; Kenny, J. A. J. Mol. Catal.
A: Chem. 1999, 146, 139; (d) Kenny, J. A.; Palmer, M. J.;
Smith, A. R. C.; Walsgrove, T.; Wills, M. Synlett 1999,
1615; (e) Wills, M.; Palmer, M. J.; Smith, A. R. C.; Kenny,
J. A.; Walsgrove, T. Molecules 2000, 5, 1; (f) Kenny, J. A.;
Versluis, K.; Heck, A. J. R.; Walsgrove, T.; Wills, M. Chem.
Commun. 2000, 99; (g) Kawamoto, A. M.; Wills, M.
Tetrahedron: Asymmetry 2000, 11, 3257; (h) Kawamoto,
A. M.; Wills, M. J. Chem. Soc., Perkin Trans. 1 2001, 1916;
(i) Hannedouche, J.; Kenny, J. A.; Walsgrove, T.; Wills, M.
Synlett 2002, 263; (j) Kenny, J. A.; Palmer, M. J.;
Walsgrove, T.; Kawamoto, A. M.; Wills, M. J. Chem.
Soc., Perkin Trans. 1 2002, 416; (k) Cross, D. J.; Kenny, J.
A.; Houston, I.; Campbell, L.; Walsgrove, T.; Wills, M.
Tetrahedron: Asymmetry 2001, 12, 1801; (l) Alcock, N. W.;
Mann, I.; Peach, P.; Wills, M. Tetrahedron: Asymmetry
2002, 13, 2485.
2. (a) Takehara, J.; Hashiguchi, S.; Fujii, A.; Inoue, S.-I.;
Ikariya, T.; Noyori, R. Chem. Commun. 1996, 233; (b)
Palmer, M.; Walsgrove, T.; Wills, M. J. Org. Chem. 1997,
62, 5226; (c) Nordin, S. J. M.; Roth, P.; Tarnai, T.;
Alonso, D. A.; Brandt, P.; Andersson, P. G. Chem. Eur. J.
2001, 7, 1431; (d) Everaere, K.; Mortreaux, A.; Bulliard,
M.; Brussee, J.; van der Gen, A.; Nowogrocki, G.;
Carpentier, J.-F. Eur. J. Org. Chem. 2001, 275; (e)
Hennig, M.; Puntener, K.; Scalone, M. Tetrahedron:
Asymmetry 2000, 11, 1849; (f) Blacker, A. J.; Mellor, B. J.
(Avecia/Zeneca), WO9842643 A1, 1998; (g) Fujii, A.;
Hashiguchi, S.; Uematsu, N.; Ikariya, T.; Noyori, R. J.
Am. Chem. Soc. 1996, 118, 2521; (h) Hashiguchi, S.; Fujii,
A.; Takehara, J.; Ikariya, T.; Noyori, R. J. Am. Chem. Soc.
Rh
H
N
HO
Ph
OH
H
12
Me

846
D. J. Cross et al. / Tetrahedron Letters 45 (2004) 843–846
1995, 117, 7562; (i) Haack, K.-J.; Hashiguchi, S.; Fujii, A.;
1.76 (3H, s, CpMe), 1.65 (3H, s, CpMe), 1.53 (3H, s,
CpMe), 0.91 (3H, d, J ¼ 6:4 Hz, CHMe). 13C NMR
(300 MHz, CDCl3): d ¼ 9:3 (CH3), 9.5 (CH3), 12.2 (CH3),
10.5 (CH3), 12.5 (CH3), 54.8 (CH2), 59.2 (CH), 72.8 (CH),
125.8, 126.1, 126.5, 127.2, 128.5, 128.7, 128.9, 129.0, 129.5,
129.7, 130.5, 131.9, 133.6, 137.6, 142.0 (Ar–C). IR
(neat) ¼ 3882, 3226, 2978, 2359, 2342, 2211, 2190, 2084,
2064, 1968, 1634, 1447, 1373, 1266, 1197, 1151, 1098, 1022,
968, 749, 730, 700, 262. MS (FAB+): 498 (MHþ, 15%), 374
(21%), 307 (24%), 154 (100%), 136 (75%), 148 (35%), 132
(22%), 106 (14%), 91 (23%), 84 (100%). HRMS (FAB) calcd
for C25H29 ClNORh ¼ 498.10254, found 498.10664.
Ikariya, T.; Noyori, R. Angew. Chem., Int. Ed. 1997, 36, 285;
(j) Mashima, K.; Abe, T.; Tani, K. Chem. Lett. 1998, 1199;
(k) Murata, K.; Ikariya, T.; Noyori, R. J. Org. Chem. 1999,
64, 2186.
3. (a) Alonso, D. A.; Brandt, A.; Nordin, S. J. M.; Andersson,
P. G. J. Am. Chem. Soc. 1999, 121, 9580; (b) Yamakawa,
M.; Ito, H.; Noyori, R. J. Am. Chem. Soc. 2000, 122, 1466;
(c) Yamakawa, M.; Yamada, I.; Noyori, R. Angew. Chem.,
Int. Ed. Engl. 2001, 40, 2818.
4. Yamakawa, M. I.; Yamada, I.; Noyori, R. Angew. Chem.,
Int. Ed. 2001, 40, 2818.
5. For an example of a tethered Cp/Ru system see: Brunner,
H.; Valerio, C.; Zabel, M. New J. Chem. 2000, 24,
275.
7. N-2-(2,3,4,5-Tetramethylcyclopentadienyl)benzyl-(L)-nor-
ephedrine rhodium(III) chloride 6 catalysed transfer hydro-
genation of acetophenone; typical procedure (5 mol %).
N-2-(2,3,4,5-Tetramethylcyclopentadienyl)benzyl-(L)-nore-
6. N-2-(2,3,4,5-Tetramethylcyclopentadienyl)benzyl-(l)-nor-
ephedrine rhodium(III) chloride 6. Rhodium(III) chloride
hydrate (0.197 g, 0.748 mmol) was added to a stirred
solution of N-2-(2,3,4,5-tetramethylcyclopentadienyl)ben-
phedrine rhodium(III) chloride 6 (0.025 g, 0.05 mmol) was
stirred in acetophenone (0.120 g, 1 mmol) under nitrogen in
a flame dried Schlenk tube for 30 min. Degassed propan-2-
ol (10 mL) was added and the reaction mixture was stirred
for 2 h. Sodium tert-butoxide (1.0 mL, 0.1 M solution in
propan-2-ol) was added. Samples were filtered through
silica, concentrated under reduced pressure and analysed by
NMR and HPLC. After 2 min the conversion was 78.2%
and ee 73.5% by HPLC analysis [Chiracel OD, hexane/
ethanol ¼ 95:5 (1.0 mL/min)], R isomer 8.30 min, S isomer
9.35 min.
zyl-(L)-norephedrine 8 (0.27 g, 0.748 mmol) in methanol
(15 mL) at room temperature. The reaction mixture was
heated under reflux and stirred for 72 h. The reaction
mixture was cooled to room temperature and concentrated
under reduced pressure to give the crude product, which
was washed with DCM to give the product 6 as a red/brown
solid (0.37 g, 99%). 1H NMR (300 MHz, CDCl3):
d ¼ 7:07–7:95 (1H, m, ArH), 7.59–7.13 (8H, m, ArH),
5.20 (1H, br s, NH), 4.81 (1H, d, J ¼ 13:1 Hz, CHPh), 4.21
(1H, m, CH2Ar), 2.9 (1H, m, CHMe), 1.84 (3H, s, CpMe),
8. Fletcher, D. A.; McMeeking, R. F.; Parkin, D. J. Chem. Inf.
Comput. Sci. 1996, 36, 746.