Dialkylzinc additions with a chiral osmaimidazolidine ligand from asymmetric diamination of olefins

Article abstract of DOI:10.1016/S0040-4039(03)00581-1

A new chiral ligand was prepared in a convenient two-step procedure starting from an asymmetric diamination reaction. Subsequent treatment of the resulting osmaimidazolidine with a phenyl Grignard reagent and titanium tetra(iso-propoxide) furnished a complex that catalyses asymmetric dialkylzinc additions to aromatic aldehydes.

Full text of DOI:10.1016/S0040-4039(03)00581-1

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 3547–3549
Dialkylzinc additions with a chiral osmaimidazolidine ligand
from asymmetric diamination of olefins
Kilian Mun˜iz*
Kekule´-Institut fu¨r Organische Chemie und Biochemie, Universita¨t Bonn, Gerhard-Domagk-Str. 1, D-53121 Bonn, Germany
Received 27 January 2003; revised 28 February 2003; accepted 1 March 2003
Abstract—A new chiral ligand was prepared in a convenient two-step procedure starting from an asymmetric diamination
reaction. Subsequent treatment of the resulting osmaimidazolidine with a phenyl Grignard reagent and titanium tetra(iso-prop-
oxide) furnished a complex that catalyses asymmetric dialkylzinc additions to aromatic aldehydes. © 2003 Elsevier Science Ltd.
All rights reserved.
We have recently described a first stereoselective diami-
nation of olefins.1 This one-step process employs known
achiral imidoosmium(VIII) reagents2–4 and acrylic esters
bearing chiral non-racemic (−)-8-phenyl menthol. The
products are obtained as diastereomerically enriched
osma(VI)imidazolidine complexes that are stable under
a variety of conditions. A second asymmetric reaction of
this type has employed commercially available bis(men-
thyl) fumarate 1 as substrate and both the bis- and the
trisimido complexes 2 and 3 underwent diamination
reactions with high diastereomeric excesses to yield the
symmetrically substituted osmaimidazolidines 4a and 4b
with 76 and 90% de, respectively (Scheme 1).1,5
ing ester group that is believed to act as a stabilising
element within the osmaimidazolidines, the resulting
complex did not display high stability and underwent
loss of the osmium moiety. However, the corresponding
complex 4b bearing an additional imido ligand at
osmium led to the product 5 which was sufficiently
stable for isolation and complete characterisation.5 The
difference in stability of the two products is noteworthy,
but can be rationalised from the difference in electron
density at the osmium metal. While in the case of the
labile dioxo derivative there is no other stabilising
moment than the lone pairs of the two nitrogens of the
metallaimidazolidine ring, the corresponding reaction
starting from 4b generates an osmaimidazolidine with
an additional imido moiety at the central Os atom.
Apparently this additional electronic factor enhances
the stability of 5. This result is in good agreement
with a report by Schrock on related osmaimidazolidi-
nes containing arylimido ligands6 and is remini-
scent to related discussions on nitrogen donor
stability in osmium complexes for olefin functionaliza-
tion.7
Within the attempts to remove the chiral auxiliary and
thus generate enantiopure osmaimidazolidines it was
anticipated that reduction of the ester moieties with
suitable carbon nucleophiles should furnish tertiary
alcohols. While addition of phenyl lithium to a solution
of 4a led to a complex mixture of products, use of the
related phenyl magnesiumchloride gave a clean reaction.
Unfortunately, due to the loss of the electron-withdraw-
Scheme 1. Asymmetric diamination and subsequent Grignard addition [R*OH=(−)-menthol].
* Fax: +49 228 73 5813; e-mail: kilian.muniz@uni-bonn.de
0040-4039/03/$ - see front matter © 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0040-4039(03)00581-1

3548
K. Mun˜iz / Tetrahedron Letters 44 (2003) 3547–3549
Interestingly, the structure of 5 with the nitrogen
chelated osmium atom displays a certain similarity to
Seebach’s TADDOL compounds (Fig. 1) that represent
successful ligands for transition metal mediated and
catalysed transformations.8
Table 1. Dialkylzinc addition to aromatic aldehydes with
a catalyst derived from complex 5 (Scheme 2)11
Entry
Rest
Rest R
Yield (%)a,b
Ee (%)c
1
C6H5
Et
Et
Et
Me
Me
Me
Me
Me
Me
Me
87 (99)
93 (99)
83 (95)
92 (99)
94 (99)
91 (99)
94 (96)
91 (97)
90 (96)
91 (98)
87
2d
3d
4
4-NO2-C6H4
4-OMe-C6H4
C6H5
C6H5
C6H5
4-MeO-C6H4
Naphthyl
Naphthyl
Ferrocenyl
91
In view of the successful chelating ability of TADDOL
for a variety of transition metals, a metal complex from
5 and titanium tetra(iso-propoxide) was considered a
promising target. Catalytic amounts of related com-
plexes on the TADDOL basis are known to promote
several asymmetric reactions, among which is catalytic
addition of dialkylzinc to aldehydes (Scheme 2, Eq.
(1)).9,10 Thus, when 5 was treated in situ with titanium
tetra(iso-propoxide), formation of the bimetallic com-
plex 6 could be anticipated which was then employed in
the prototypical diethylzinc addition to benzaldehyde.10
As already described by Seebach for his Ti-TADDOL-
ate catalysts, use of a catalytic amount of 6 alone had
no decisive effect since the respective reaction pro-
ceeded at very low rate and the crude reaction mixture
contained various products. However, the reaction pro-
ceeded efficiently in the presence of an equimolar
amount of the achiral titanium reagent, and 2-phenyl
propanol of 87% ee was isolated from a diethylzinc
addition to benzaldehyde when 20 mol% of 5 had been
added previously (Table 1, entry 1).11
85e
91
5d
6f
7
90
75
88e
92e
92e
91
8
9d
10d,g
a Isolated yields after standard work-up and column chromatogra-
phy.
b Yields in brackets refer to % conversion as estimated from the
proton NMR of the crude reaction mixture.
c Determined by HPLC.
d A 0.5-M catalyst stock solution prepared from 5 and Ti(O-i-Pr)4 in
toluene was used.
e Determined by NMR analysis after conversion to the respective
Mosher ester.
f With a sample of 6 containing 82% ee (see text).
g Reaction was carried out at −20°C.
excesses in the range of 88–92%. As judged by compari-
son with literature data, the products were S-configu-
rated in all cases. This indicates an identical reaction
course via Si-face addition and is in agreement with
those transition states assumed for related catalysts
derived from (R,R)-configurated TADDOL. While 6
contains two metal centres, the osmium centre does not
participate in catalysis: because of its electron donating
imido ligand it is formally an electronically saturated
metal centre with no Lewis acidity. For the same rea-
son, the nitrogen groups of the osmaimidazolidine
backbone do not participate in the formation of alter-
native, catalytically active species. These observations
confirm that 5 indeed behaves as a TADDOL-analogue
in the present reaction, although the sterical bulk from
the tert-butylimido ligand renders the observed enan-
tioselectivities lower than those for reactions with
TADDOL itself.10 In view of the rather tedious purifi-
cation of the osmaimidazolidine, an alternative one-pot
procedure for the preparation of a stock solution of 6
was pursued. Thus, treatment of 4b with 4.8 equiv. of
the phenyl Grignard reagent followed by methanol
addition, filtration through celite, evaporation of the
solvents and addition of 2.6 equiv. of Ti(O-i-Pr)4 in
toluene allowed for the preparation of a 0.5-M toluene
stock solution of the catalyst that was sufficiently stable
when stored under argon at −24°C. The results from
reactions using this catalyst solution are given in Table
1, entries 2, 3, 5, 9 and 10, and show that this procedure
gives nearly identical results. However, the exact course
of these reactions remains to be elucidated since the
prepared stock solution of 6 may still contain addi-
tional salt complexes generated from the excess Grig-
nard reagent and the liberated mentholate, respectively.
However, even if such species are present, their catalytic
reactivity should be low rendering their influence on the
stereochemical reaction course rather small.
Other aldehydes can be submitted to alkylation under
identical reaction conditions (Table 1). For diethylzinc
additions, 4-nitro benzaldehyde gave the best result
(91% ee) while the electron-rich 4-methoxy benzalde-
hyde was rather problematic. Dimethylzinc worked
equally well as alkylating reagent and the respective
secondary alcohols were isolated with enantiomeric
Figure 1. Structural relation of 5 and TADDOL.
Scheme 2. In situ formation of the assumed catalyst 6 for
dialkylzinc addition to aldehydes.

K. Mun˜iz / Tetrahedron Letters 44 (2003) 3547–3549
3549
Since the initial diamination reaction in the formation
of 4b proceeds with high, albeit not with complete
diastereoselectivity, it was possible to obtain
diastereomeric mixtures for the preparation of enan-
tioenriched ligands. For example, when catalyst 6 was
prepared by the in situ protocol from a sample of 4b
with 82% diastereomeric excess, a dimethylzinc addition
to benzaldehyde yielded a product with 75% ee. This
value points to a linear relation between the ee of the
catalyst and the ee of the product. Therefore, the actual
catalyst can be considered monomeric in nature, a fact
that is evidenced from the steric properties of 5 and
6.12,13
3. Nugent, W. A.; Harlow, R. L.; McKinney, R. J. J. Am.
Chem. Soc. 1979, 101, 7265–7268.
4. Danopoulos, A. A.; Willkinson, G.; Hussain-Bates, B.;
Hursthouse, M. B. J. Chem. Soc., Dalton Trans. 1991,
269–275.
5. Full account on osmaimidazolidine structures and com-
plexes including 5: Mun˜iz, K.; Iesato, A.; Nieger, M.
manuscript in preparation.
6. Schofield, M. H.; Kee, T. P.; Anhaus, J. T.; Schrock, R.
R.; Johnson, K. H.; Davis, W. M. Inorg. Chem. 1991, 30,
3595–3604.
7. (a) Hanessian, S.; Meffre, P.; Girard, M.; Beaudoin, S.;
Sance´au, J.-Y.; Bennani, Y. J. Org. Chem. 1993, 58,
1991–1993; (b) Donohoe, T. J.; Blades, K.; Moore, P. R.;
Waring, M. J.; Winter, J. J. G.; Helliwell, M.; New-
combe, N. J.; Stemp, G. J. Org. Chem. 2002, 67, 7946–
7956 and literature cited therein; (c) Corey, E. J.; Sarshar,
S.; Azimioara, M. D.; Newbold, R. C.; Noe, M. C. J.
Am. Chem. Soc. 1996, 118, 7851–7854; (d) Tomioka, K.;
Nakajima, M.; Koga, K. J. Am. Chem. Soc. 1987, 109,
6213–6215; (e) Corey, E. J.; DaSilva Jardine, P.; Virgil,
S.; Yuen, P.-W.; Connell, R. D. J. Am. Chem. Soc. 1989,
111, 9243–9244.
8. Review on TADDOL chemistry: Seebach, D.; Beck, A.
K.; Heckel, A. Angew. Chem., Int. Ed. 2001, 40, 92–138.
9. (a) Noyori, R. Asymmetric Catalysis in Organic Synthe-
sis; Wiley: New York, 1994; (b) Noyori, R.; Kitamura,
M. Angew. Chem., Int. Ed. Engl. 1991, 30, 49–69; (c)
Soai, K.; Niwa, S. Chem. Rev. 1992, 92, 833–856; (d)
Knochel, P.; Singer, R. D. Chem. Rev. 1993, 93, 2117–
2188; (e) Pu, L.; Yu, H.-B. Chem. Rev. 2001, 101, 757–
824.
This successful first application of osmaimidazolidine 5
to asymmetric catalysis demonstrates that diamination
reactions with the bisimido reagents 2 and 3 can furnish
synthetically valuable intermediates even for those cases
where the tert-butyl substituents at nitrogen are not
removed afterwards. This feature had previously been
regarded as the most significant drawback in the diami-
nation of olefins by means of tert-butylimido osmium
reagents.1,2 However, in the present case one might
speculate that due to their steric bulk these moieties
enforce a rigid configurational catalyst environment.
In summary, we have described a selective modification
of stereochemically defined osmaimidazolidines that
can be obtained in a one-step procedure from a chiral
commercially available olefin and easily accessible
osmium reagents. The application of such a modified
complex 5 in asymmetric catalytic C–C bond formation
has been demonstrated and represents the first example
of an osmaimidazolidine ligand in asymmetric catalysis.
Current work is aimed to enhance the application of
this catalyst and related ones in other areas of asym-
metric catalysis, especially in that of oxidative
transformations.
10. (a) Seebach, D.; Beck, A. K.; Schmidt, B.; Wang, Y. M.
Tetrahedron 1994, 50, 4363–4384; (b) see also Refs 8 and
9e.
11. General reaction conditions for dialkylzinc additions: 5
(20 mol%) and Ti(O-i-Pr)4 (1.45 equiv.) were dissolved in
toluene, cooled to −30°C and treated with ZnR2 (1.8
equiv.) and the respective aldehyde (1.0 equiv). The
resulting solution was stirred for 20–24 h, followed by
addition of a saturated solution of NH4Cl and extraction
with dichloromethane. The alcohols were isolated by
conventional column chromatography on silica gel. In the
case of reactions with preformed catalyst, 5 was replaced
by 20 mol% of 6; otherwise the procedure was identical.
12. (a) Girard, C.; Kagan, H. B. Angew. Chem., Int. Ed.
1998, 37, 2922–2959; (b) Kagan, H. B. Adv. Synth. Catal.
2001, 343, 227–233; (c) Avalos, M.; Babiano, R.; Cintas,
P.; Jime´nez, J. L.; Palacios, J. C. Tetrahedron: Asymmetry
1997, 8, 2997–3017.
Acknowledgements
K.M. is grateful to Prof. Dr. K. H. Do¨tz for generous
support of this work and to A. Schneider for HPLC
analyses. A Liebig stipend from the Fonds der Chemis-
chen Industrie is gratefully acknowledged.
References
13. For a discussion on catalyst preparation regarding non-
linear effects: Luukas, T. O.; Fenwick, D. R.; Kagan, H.
B. C.R. Chimie 2002, 5, 487–491.
1. Mun˜iz, K.; Nieger, M. Synlett 2003, 211–214.
2. Chong, A. O.; Oshima, K.; Sharpless, K. B. J. Am. Chem.
Soc. 1977, 99, 3420–3426.