Rh(II) catalysts with 4-hydroxyproline-derived ligands

Article abstract of DOI:10.1016/j.tetlet.2010.07.115

Three new chiral Rh(II) catalysts with 4-hydroxyproline-derived ligands have been synthesised through a short and efficient synthetic route. The catalysts give good yields and ees in C-H insertion and cyclopropanation reactions, and their properties indicate an all-up reactive conformation of proline- and 4-hydroxyproline-derived Rh(II) catalysts.

Full text of DOI:10.1016/j.tetlet.2010.07.115

Tetrahedron Letters 51 (2010) 5375–5377
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Rh(II) catalysts with 4-hydroxyproline-derived ligands
*
Hanne Therese Bonge, Massoud Kaboli, Tore Hansen
Department of Chemistry, University of Oslo, Sem Sælands vei 26, N-0315 Oslo, Norway
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 29 April 2010
Revised 6 July 2010
Accepted 21 July 2010
Available online 29 July 2010
Three new chiral Rh(II) catalysts with 4-hydroxyproline-derived ligands have been synthesised through
a short and efficient synthetic route. The catalysts give good yields and ees in C–H insertion and
cyclopropanation reactions, and their properties indicate an all-up reactive conformation of proline-
and 4-hydroxyproline-derived Rh(II) catalysts.
Ó 2010 Elsevier Ltd. All rights reserved.
Carbenoid C–H insertion and cyclopropanation reactions
catalysed by chiral dirhodium(II) catalysts represent extremely
powerful methods in asymmetric synthesis, often giving high
chemo-, regio- and diastereoselectivity along with excellent
enantioselectivity.1 The development of new chiral Rh(II) catalysts,
and the understanding of their mode of asymmetric induction, is
therefore of great importance.
plished through high temperature ligand exchange5 with commer-
cially available Rh2(OAc)2, giving catalysts 1–3 in good to excellent
yields.
The Rh(II)-HYP catalysts have advantageous solubility in organic
solvents. With the sizeable hydrophobic side-chains of the HYP-li-
gands, the 4-dodecylphenyl substituent used to aid solubility in
Rh2(DOSP)4 (4) is no longer necessary. For instance, the commer-
cially available Rh2(TBSP)4 (5) is a catalyst that displays near identi-
cal reactivity and selectivity to 4,6 but with a 4-t-butylphenyl group
instead of 4-dodecylphenyl, it is less soluble in organic solvents and
has therefore found little use. However, the 4-t-butyl-phenyl group
works well in the Rh(II)-HYP catalysts, showing how the range of
possible aryl groups is broadened compared to the catalysts with
proline-derived ligands.
The three new catalysts 1–3 were tested in cyclopropanation
reactions with styrene and C–H insertion reactions with adaman-
tane (Table 1). As methyl aryldiazoacetate is known to give good
results in reactions catalysed by Rh2(DOSP)4,7 this diazo compound
became our test substrate. In order to lower the ee compared to the
optimal results, thus facilitating comparison of the different cata-
lysts, the reactions were performed at room temperature, a higher
reaction temperature than is optimal for 4.6 The Rh(II)-HYP
Proline-derived Rh(II) carboxylates, in particular Rh2(DOSP)4 (4,
Fig. 1), are catalysts that have found widespread use in carbenoid
reactions.2 The merits of 4 include excellent enantioselectivities in
C–H insertion and cyclopropanation reactions with Davies’ aryl-
and vinyl-substituted diazo compounds. Our group has recently
developed methodology for selective O-acylation of 4-hydroxypro-
line with acyl chlorides.3 This methodology offers a short and prac-
tical route to chiral, 4-hydroxyproline-based ligands (HYP-ligands)
for new Rh(II) catalysts. While proline is a well-knownchiral compo-
nent in Rh(II) ligands, HYP-ligands are, to the best of our knowledge,
yet to be used. They do, however, offer certain advantages over
proline-derived ligands, such as a second stereogenic centre, and a
side-chain that can be used for tuning properties such as solubility
and steric bulk. Additional factors that make 4-hydroxyproline a
highly appealing building block for new chiral ligands are those of
economy and availability: trans-4-hydroxyproline is commercially
available at a low cost and can, in a facile manner,3,4 be cleanly
converted into the more costly cis-isomer.
O
O
R
R
O
O
Rh
Rh
Herein, we report the synthesis of three new Rh(II)-HYP cata-
lysts (1–3, Fig. 1) through a short, convenient synthetic route
(Scheme 1) in good overall yields.
O
O
N
SO2
O
O
Rh
Rh
O
O
Rh
Rh
Following the previously described method,3 trans- and
cis-4-hydroxyprolines were O-acylated with lauroyl chloride or
cyclohexylcarbonyl chloride in good to excellent yields without
the need for chromatography. N-Sulfonation with 4-t-buty-
lphenylsulfonyl chloride, in good yields, completed the synthesis
of the ligands. Coordination of the ligands to dirhodium was accom-
N
SO2
N
SO2
4
R
4
4
t-Bu
4: Rh2(
R=C12H25
5: Rh2( -TBSP)4
R= -Bu
S-DOSP)4
t-Bu
1: R=C12H25
2: R=C12H25
3: R=cyclohexyl
S
t
* Corresponding author. Tel.: +47 22 85 53 86; fax: +47 22 85 54 41.
E-mail address: t.hansen@kjemi.uio.no (T. Hansen).
Figure 1. New catalysts 1–3 and catalysts with proline-derived ligands 4 and 5.
0040-4039/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2010.07.115

5376
H. T. Bonge et al. / Tetrahedron Letters 51 (2010) 5375–5377
each other. It is known from results with Rh2(S-DOSP)4 and
HO
O
HO
2. HCl, H2O
1. Ac2O
Rh2(R-DOSP)4 that the enantioselectivities of proline-derived cata-
lysts are governed by the stereochemistry at C2 of proline, and as a
starting material cis-4-hydroxyproline is generated from trans-4-
hydroxyproline through inversion at C2, catalysts 1 and 2 have
opposite stereochemistry at this centre.3,4
CO2H
CO2H
N
H
N
H
O
CF3CO2H
CF3CO2H
R
O
Cl
R
Cl
The Rh(II)-HYP catalysts are also useful tools in the ongoing
pursuit for understanding the mode of asymmetric induction in
the chiral Rh(II) catalysts. The exact nature of the stereochemical
induction achieved using the proline-based carboxylate ligands
in catalysts 4 and 5 has yet to be explored. The prevalent hypoth-
esis is that the ligands in chiral Rh(II) catalysts possess ‘blocking
groups’ that, pointing either up or down in a total of four possible
catalyst conformations (Fig. 2), favour one enantiotopic trajectory
of the substrate towards the carbenoid over the other.2,6 Davies
et al. have postulated a D2-symmetric (up, down, up, down) orien-
tation of the aryl groups in 4 and similar catalysts, resulting in two
equivalent catalyst faces.2 This has its basis in that catalysts with
two different faces should be ineffective, as the less sterically
encumbered face appears achiral. Recently, however, Fox and co-
workers performed a computational study of Rh2(S-PTTL)4, another
frequently employed chiral Rh(II) catalyst, with results indicating
an all-up, ‘chiral crown’ conformation with C4-symmetry.8 The
proposition that the catalyst has a reactive chiral face and an unre-
active achiral face represents a novel way of perceiving these cat-
alysts, and has sparked a renewed interest in understanding their
mode of chiral induction. The group of Charette has further inves-
tigated this theory, finding strong experimental evidence in favour
of a reactive all-up conformation for certain catalysts similar to
Rh2(S-PTTL)4.9 An important question which then arises is if this
conformation is specific only for the catalysts studied by Charette
and Fox, or if an all-up orientation of the ligands is in fact the active
conformation of other chiral Rh(II) catalysts as well, such as
Rh2(DOSP)4 (4) and its analogues.
The observed high level of similarities between the Rh(II)-HYP
catalysts and the proline-derived catalysts in the C–H insertion
and cyclopropanation reactions, both in yields and enantioselectiv-
ity, has interesting implications for understanding of the mode of
stereoselectivity for proline- and hydroxyproline-derived catalysts.
Our findings show that the outcome of the C–H insertion and
cyclopropanation reactions is not affected by the presence of a
large substituent on C4 of the proline ring. Both yields and ee re-
main largely unchanged in going from unsubstituted 4 and 5 to
lauroyl-substituted 1 and 2 or cyclohexylcarboxyl-substituted 3,
and changing the relative stereochemistry of the ligands from trans
in 1 to cis in 2 also has minor effects. Two implications follow from
these results: (i) the active conformation of the catalysts must have
the ligands oriented in such a way that the side-chains are remote
from the reactive site of the carbenoid, so that they do not
O
O
O
R
R
CO2H
CO2H
N
N
H. HCl
H. HCl
Na2CO3
or KHCO3
Na2CO3
or KHCO3
R' SO2Cl
R' SO2Cl
O
O
R
O
O
R
CO2H
CO2H
N
N
SO2
R'
SO2
R'
Rh2(OAc)4
Rh2(OAc)4
Rh2(cis-HYP)4
Rh2(trans-HYP)4
R=C12H25 or cyclohexyl
R'=4- -Bu-Ph
t
Scheme 1. General synthesis of Rh(II)-HYP catalysts.
Table 1
Cyclopropanation and C–H insertion reactions with catalysts 1–5
Ph CO2Me
Ph
Ph
Ph
CO2Me
Ph
N2
(1.5 equiv)
Catalyst (1 mol%)
i-hexane, r.t., 1 h
CO2Me
Ph
CO2Me
(1.5 equiv)
Catalyst (1 mol%)
N2
i-hexane, r.t., 1 h
Entry
Substrate
Catalyst
Yield (%)
ee
1
2
3
4
5
6
7
8
9
Styrene
Styrene
Styrene
Styrene
Adamantane
Adamantane
Adamantane
Adamantane
Adamantane
4
1
2
3
4
5
1
2
3
80
76
78
74
56
55
56
65
67
80
84a
90b
80b
86
85a
92a
90b
93b
The ee was measured by chiral HPLC, directly for entries 5–9, and after LiAlH4
reduction of the ester group to the corresponding alcohol for entries 1–4.
a
O
O
O
O
Same absolute configuration as with 4.
Rh
Rh
Rh
Rh
b
O
O
O
O
Opposite absolute configuration to that with 4. The de in the cyclopropanation
O
O
O
O
O
O
O
O
reactions, measured by 1H NMR spectroscopy of crude product, was in all instances
ꢀ20:1.
D2
C4 (all-up)
O
O
catalysts gave very good results in both the cyclopropanation reac-
tion and the C–H insertion reaction. The yields with catalysts 1–3
in all the reactions were very similar to those obtained with 4
and 5. The level of enantiocontrol with the new catalysts also
matched that observed with the proline-derived catalysts; the
ees obtained with 1–3 were similar to the ees with 4 and 5. As ex-
pected, catalysts 1 and 2 displayed opposite enantioselectivity to
O
O
O
O
Rh
Rh
O
O
Rh
Rh
O
O
O
O
O
O
O
O
C2
C1
Figure 2. Four possible conformations of the chiral Rh(II) catalysts. The spheres
represent arylsulfonyl groups.

H. T. Bonge et al. / Tetrahedron Letters 51 (2010) 5375–5377
5377
take part in a dynamic process in coordination of a carbenoid to
the catalyst, giving rise to an induced fit with a chiral pocket
around the carbenoid, the arylsulfonyl groups blocking one of the
enantiotopic faces of the carbenoid.
In summary, we have developed a facile protocol for the synthe-
sis of new 4-hydroxyproline-derived chiral Rh(II) catalysts. The
Rh(II)-HYP catalysts possess a side-chain that aids solubility in or-
ganic solvents, and they have been shown to give similar yields and
enantioselectivities in cyclopropanation and C–H insertion reac-
tions in comparison to the widely used Rh2(DOSP)4 and its ana-
logues. Studies on the Rh(II)-HYP catalysts imply an all-up
conformation with a reactive chiral face and an unreactive achiral
face for Rh(II) catalysts with proline-derived ligands such as
Rh2(DOSP)4 and Rh2(TBSP)4, along with the new Rh(II)-HYP cata-
lysts. Further studies will be reported in due course.
S
O
O
O
O
O
Rh
Rh
N
O
O
O
O
O
Figure 3. Illustration of the wide range of movement of acyl substituents (red).
Only one substituent is drawn for clarity. The orientations of the proline and aryl
rings are based on computational optimisation of a simplified DOSP-ligand.10
influence the reaction and (ii) this orientation of the ligands must
be the same for all catalysts 1–5, as indicated by their very similar
enantioselectivities. Of the four catalyst conformations shown in
Figure 2, the only one that clearly fits these criteria is the all-up,
C4-symmetric conformation. Only if the reaction takes place at
the top face of a catalyst with all the arylsulfonyl groups pointing
up can the side-chains, at all times, be sufficiently far away from
the reaction site not to be expected to influence the reaction. The
large acyl substituents have a wide range of movement, as illus-
trated in Figure 3, and if either of the ligands were to have the aryl-
sulfonyl group pointing down, the side-chain could come into
proximity to the carbenoid.
An all-up conformation gives rise to two different catalyst faces:
a bottom face which is not expected to induce stereocontrol, and
which in the case of 4 and 5 is completely unshielded, and a steri-
cally congested chiral top face, where the reaction must take place.
We hypothesise, along the lines of Charette and co-workers,9 that
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.tetlet.2010.07.115.
References and notes
1. (a) Davies, H. M. L.; Antoulinakis, E. G.. In Organic Reactions; Wiley, 2001; Vol.
. p 1–326; (b) Singh, V. K.; DattaGupta, A.; Sekar, G. Synthesis 1997, 137–149;
(c) Doyle, M. P.; Duffy, R.; Ratnikov, M.; Zhou, L. Chem. Rev. 2010, 110, 704–724;
(d) Davies, H. M. L.; Manning, J. R. Nature 2008, 451, 417–424; (e) Davies, H. M.
L.; Beckwith, R. E. J. Chem. Rev. 2003, 103, 2861–2903.
2. Hansen, J.; Davies, H. M. L. Coord. Chem. Rev. 2008, 252, 545–555.
3. Kristensen, T. E.; Hansen, F. K.; Hansen, T. Eur. J. Org. Chem. 2009, 387–395.
4. Dalla Croce, P.; La Rosa, C. Tetrahedron: Asymmetry 2002, 13, 197–201.
5. Pirrung, M. C.; Zhang, J. Tetrahedron Lett. 1992, 33, 5987–5990.
6. Davies, H. M. L.; Bruzinski, P.; Hutcheson, D. K.; Kong, N.; Fall, M. J. J. Am. Chem.
Soc. 1996, 118, 6897–6907.
7. Davies, H. M. L.; Walji, A. M. In Modern Rhodium-catalyzed Organic Reactions;
Evans, P. A., Ed.; Wiley-VCH, 2005; pp 301–340.
8. DeAngelis, A.; Dmitrenko, O.; Yap, G. P. A.; Fox, J. M. J. Am. Chem. Soc. 2009, 131,
7230–7231.
9. Lindsay, V. N. G.; Lin, W.; Charette, A. B. J. Am. Chem. Soc. 2009, 131, 16383–
16385.
10. Nowlan, D. T.; Gregg, T. M.; Davies, H. L. M.; Singleton, D. A. J. Am. Chem. Soc.
2003, 125, 15902–15911.
p-stacking interactions between the arylsulfonyl groups and the
carbenoid substituents may be the reason why formation of the
most sterically hindered carbenoid is favoured. The scope of 4 gives
credence to this theory: catalyst 4 induces high ee in reactions with
aryl- and vinyldiazoacetates,2,7 which have substituents capable of
p
-stacking, but gives poor enantiocontrol with others such as ethyl
11. Davies, H. M. L.; Bruzinski, P. R.; Fall, M. J. Tetrahedron Lett. 1996, 37, 4133–
4136.
12. Bonge, H. T.; Pintea, B.; Hansen, T. Org. Biomol. Chem. 2008, 6, 3670–3672.
diazoacetate11 and halodiazoacetates.12 Thus, our findings indicate
that the arylsulfonyl groups are not merely blocking groups, but