Synthesis and resolution of the planar chirality of ester-functionalised phospharuthenocenes

Article abstract of DOI:10.1039/b401088d

Menthyl phospholide ester anions provide an operationally simple and high yielding entry to the first planar chiral enantiopure phospharuthenocene derivatives.

Full text of DOI:10.1039/b401088d

Synthesis and resolution of the planar chirality of ester-functionalised
phospharuthenocenes
Duncan Carmichael,* Jürgen Klankermayer, Louis Ricard and Nicolas Seeboth
CNRS UMR 7653, DCPH Ecole Polytechnique, 91128 Palaiseau cedex, France; Fax: + 33 1 69333990;
Tel: + 33 1 69334571. E-mail: Duncan.Carmichael@polytechnique.fr
Received (in Cambridge, UK) 23rd January 2004, Accepted 12th March 2004
First published as an Advance Article on the web 14th April 2004
Menthyl phospholide ester anions provide an operationally
simple and high yielding entry to the first planar chiral
enantiopure phospharuthenocene derivatives.
Complexation of [Cp*RuCl]4 to the menthyl-substituted phos-
pholide anions 8a,b gave pale yellow air-stable diastereomeric
phospharuthenocene pairs in chemical yields of ca 90%. The
diastereomers show the expected21 configurational stability (with
Recent years have seen a renaissance in the chemistry of
phosphametallocenes,1 largely as a result of Ganter,2 Fu3–5 and
Hayashi’s6 elegant work in preparing and evaluating enantiopure
phosphaferrocenes as ligands for asymmetric catalysis. Whilst little
explored, this class has already shown its promise through the
exceptional ees observed in Cu-catalysed enantioselective [3 + 2]
cycloadditions,3 Rh-based asymmetric Kinugasa4 and allylic
alcohol isomerisations5 and Pd-catalysed allylic alkylations.6
Arguably, the principal hurdles to the routine use of such
compounds as chiral building blocks are now simple practical
problems concerning a) the large scale resolution planar chiral
Cp*Fe derivatives such as 1a3,7 which, to date, have often provided
the best excesses and b) the well-documented chemical sensitivity8
of some phosphaferrocenes.
Our studies of phosphametallocenes built around late transition
metals (2 M = Ru,9 Co,10 Co+,11 Rh+,11 Ir+,11 Ni,12 Ni+12) suggest
that the substitution of ruthenium for iron will generate novel
complexes showing improved stability.13 However, synthesising
complexes of type 2 has always involved the use of highly
encumbered phospholyl ligands which promote the formation of
Scheme 1 Reagents and conditions: i; [Cp*RuCl]4 (0.25 eq.), THF, rt, 30
min.
5
the desired h -complexes by exerting steric pressure upon the
otherwise kinetically- favoured dimers (e.g. 3),13,14 a method which
implies extended synthetic pathways13 and difficulties in the
resolution of chirality.15 An untried but, in principle, generally
applicable, higher yielding and less synthetically arduous approach
might involve perturbing the electronics of the phospholide anion
to reduce the charge and/or HOMO localisation at P, thus favouring
5
the formation of the h -phosphametallocene. 2-Ester substituted
phospholide anions are easily accessible16 and provide an attractive
class for exploring this hypothesis; in addition to any electronic
modification, the ester group also provides a convenient basis for
the elaboration of the phosphametallocene product.17
Initial non-optimised studies using the simplest methoxycarbo-
nyl substituted phospholides were encouraging; reaction of 416 with
[Cp*RuCl]4 in THF gave, after chromatography on alumina,
acceptable isolated yields of the pale yellow, air-stable phosphar-
uthenocene 5, a late intermediate in the synthesis of the racemic
phospharuthenocenephosphine 1b13 (Scheme 1). Under analogous
conditions, the similarly encumbered phospholide 6b gives > 95%
of the unwanted dimer 3.13
This proof of principle experiment spurred the development of a
‘second generation’ approach involving the additional use of the
ester function to resolve the chirality of the phosphametallocene.18
The cheap commercial availability of both hands of menthyl-
chloroformate, straightforward separation of diastereomeric cyclo-
pentadienyl-19 and pyrrolyl-20 derived metallocene menthyl esters,
and likely increase in chemical yield when using a more hindered
ester functionality,13 made 2-(menthoxycarbonyl)phospholide ani-
ons a clear target. Two readily available phospholides 6a,b were
found to be easily transformed into the new colourless, solid
menthoxycarbonylphospholides 8a,b in excellent yields through a
classical13,16 [1,5] sigmatropic shift protocol linked to a KOtBu
induced deprotonation and subsequent precipitation from pentane
(Scheme 2).†
Scheme 2 Reagents and conditions: i: (2)-MenOCOCl (1 eq.), THF, 0 °C,
5 min. ii: 65 °C, 3 h, then KOt-Bu (1 eq.) THF, 0 °C, 5 min. iii:[Cp*RuCl]4
(0.25 eq.), THF, 30 min. iv: LiAlH4, (2eq) THF, rt, 4 h, v: LiAlH4, (2eq)
THF, 66 °C, 5 h. Isolated yields after crystallisation in parentheses.
1144
C h e m . C o m m u n . , 2 0 0 4 , 1 1 4 4 – 1 1 4 5
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 4

ppm. [a]25 = 2125° (c = 1.0, DCM). 10a: (Et2O, 298 K) 31P NMR d
D
236.2 ppm. 10b: (CDCl3, 298 K) 31P NMR d 225.4 ppm. [a]25 = 76.4°
D
(c = 1.0, DCM).11a [a]25 = 6.4° (c = 0.5, THF). (+)11b (CDCl3, 298 K)
D
31P NMR d 237.1 ppm. [a]25 = 212° (c = 1.0, DCM).
D
‡ Crystal data: 9a: C27H41O2PRu, M = 529.64, orthorhombic, space group
P212121, a = 9.1310(10), b = 21.5780(10), c = 13.2830(10) Å. U =
2617.1(4) Å3. Z = 4, Dc = 1.344 g cm23, F(000) = 1112. Graphite
monochromated X-ray Mo–Ka radiation, l = 0.71069 Å. m = 0.680 cm21
,
T = 150.0(10) K. Of 7586 independent reflections collected on a Kappa
CCD diffractometer from a pale yellow plate of dimensions 0.24 3 0.20 3
0.17 mm over h = 212 to 12, k = 230 to 22°, l = 218 to 18, 6650 having
I > 2s(I) were refined on F2 using direct methods in Shelxl. wR2 = 0.0918,
R1
33H45O2PRu, M = 605.73, orthorhombic, space group P212121, a =
10.085(5), b = 11.771(5), c = 26.420(5) Å. U = 3136(2) Å3. Z = 4, Dc
= 0.0379, GoF = 1.042, Flack’s parameter = 20.06(2). 10b:
C
Fig. 1 Molecular structure of 9a. Selected bond lengths (Å) Ru(1)–P(1),
2.4142(7); Ru(1)–C(1), 2.212(2); Ru(1)–C(2), 2.199(3); Ru(1)–C(3),
2.192(3); Ru(1)–C(4), 2.188(3); P(1)–C(1), 1.773(3); P(1)–C(4), 1.798(3);
C(1)–C(2), 1.407(4); C(2)–C(3), 1.436(4); C(3)–C(4), 1.438(4); C(4)–C(5),
1.475(4).
=
1.283 g cm23, F(000) = 1272. Graphite monochromated Mo–Ka radiation,
l = 0.71069 Å. m = 0.577 cm21, T = 150.0(10) K. Of 9007 independent
reflections from a pale yellow cube of 0.20 3 0.20 3 0.20 mm collected as
above over h = 214 to 14; k = 216 to 16; l = 237 to 37°, 7400 having
I > 2s(I) were refined on F2 using direct methods in Shelxl. wR2 = 0.0911,
R1 = 0.0399, GoF = 1.039, Flack’s parameter = 20.02(2). CCDC 226405
& 226406. See http://www.rsc.org/suppdata/cc/b4/b401088d/ for crystallo-
graphic data in .cif or other electronic format.
1 D. Carmichael and F. Mathey, Top. Curr. Chem., 2002, 220, 27.
2 C. Ganter, L. Brassat and B. Ganter, Tetrahedron Asymmetry, 1997, 8,
2607.
3 R. Shintani and G. C. Fu, J. Am. Chem. Soc., 2003, 125, 10779.
4 G. C. Fu and R. Shintani, Angew. Chem. Int. Ed., 2003, 42, 4082.
5 K. Tanaka and G. C. Fu, J. Org. Chem., 2001, 66, 8177.
6 M. Ogasawara, K. Yoshida and T. Hayashi, Organometallics, 2001, 20,
3913.
7 S. Qiao and G. C. Fu, J. Org. Chem., 1998, 63, 4168.
8 L. Brassat, B. Ganter and C. Ganter, Chem. Eur. J., 1998, 4, 2148; B.
Deschamps, L. Ricard and F. Mathey, J. Organomet.Chem., 1997, 548,
17; R. M. G. Roberts, J. Silver and A. S. Wells, Inorg. Chim. Acta, 1989,
155, 197 see also ref 5.
Fig. 2 Molecular structure of 10b. Selected bond lengths (Å). Ru(1)–P(1),
2.399(1); Ru(1)–C(1), 2.218(3); Ru(1)–C(2), 2.207(3); Ru(1)–C(3),
2.187(3); Ru(1)–C(4), 2.181(3); P(1)–C(1), 1.787(3); P(1)–C(4), 1.784(3);
C(1)–C(2), 1.439(4); C(1)–C(6), 1.484(3); C(2)–C(3), 1.434(3); C(3)–C(4),
1.434(4); C(4)–C(5), 1.478(4).
9 D. Carmichael, L. Ricard and F. Mathey, Chem. Commun., 1994,
1167.
9b and 10b undergoing no detectable decomposition or inter-
conversion in toluene at 110 °C over 1 month) and were obtained in
purities exceeding 99% through a single recrystallisation from
MeOH after multigram separation by chromatography on alumina
(neutral; hexane–dichloromethane 9:1).22 Their absolute configura-
tions were confirmed by X-ray crystallography‡ for 9a and 10b
(Figs. 1 and 2). In each case, reduction with LiAlH4 gave the
desirable13 enantiopure phospharuthenocenemethanol building
blocks 11a,b in near-quantitative yields.
10 C. Burney, D. Carmichael, K. Forissier, J. C. Green, F. Mathey and L.
Ricard, Chem. Eur. J., 2003, 9, 2567.
11 K. Forissier, L. Ricard, D. Carmichael and F. Mathey, Organometallics,
2000, 19, 954.
12 C. Burney, D. Carmichael, K. Forissier, J. C. Green, F. Mathey, L.
Ricard and S. Wendicke, Phosphorus Sulfur Silicon Relat. Elem., 2002,
177, 1999.
13 D. Carmichael, F. Mathey, L. Ricard and N. Seeboth, Chem. Commun.,
2002, 2976.
14 T. Arliguie, M. Ephritikhine, M. Lance and M. Nierlich, J. Organomet.
Chem., 1996, 524, 239.
The phospholide ester methodology described above provides a
simple solution to problems of stability and accessibility of
phosphametallocenes for enantioselection, and allows the prepara-
tion of only the second series of enantiopure planar chiral
phosphametallocene sandwich complexes (after M = Fe). Addi-
tional preliminary results show that such carboxymenthyl-substi-
tuted phospholides may have a much more profound role to play in
the preparation of enantiopure functionalised phosphametallo-
cenes, with reactions of 8b at [MnBr(CO)5] or [Cp*FeCl] centres
again furnishing ring-functionalised phosphaferrocenes and phos-
phacymantrenes whose chirality can be resolved through simple
crystallisation.23 Elaboration of these complexes should facilitate
access to resolved phosphametallocenes and considerably broaden
their availability. A comparison of the behaviour of phosphaferro-
cene and phospharuthenocene ligands will appear in due course.
We thank CNRS and Ecole polytechnique for support. We also
wish to further acknowledge financial support to JK through the
European Community’s Human potential program under the
contract HPRN-CT-2001-00172 (Daccord).
15 For an elegant exception: M. Ogasawara, T. Nagano, K. Yoshida and T.
Hayashi, Organometallics, 2002, 21, 3062.
16 S. Holand, M. Jeanjean and F. Mathey, Angew. Chem., Int. Ed. Engl.,
1997, 36, 98; M. Melaimi, L. Ricard, F. Mathey and P. Le Floch, Org.
Lett., 2002, 4, 1245.
17 Vilsmeier formylation, efficient for phosphaferrocenes, fails with less
electron rich phospholyl complexes, thus making pre-installed function-
ality rather valuable: B. Deschamps, L. Ricard and F. Mathey,
Organometallics, 1999, 18, 5688.
18 For resolutions of other diastereomeric monophosphaferrocenes: see
refs. 2,4 and R. Shintani and G. C. Fu, Org. Lett., 2002, 4, 3699.
Diphosphaferrocenes: A. Klys, R. B. Nazarski and J. Zakrzewski, J.
Organomet. Chem., 2001, 627, 135; A. Klys, J. Zakrzewski, A.
Rybarczyk-Pirek and T. A. Olszak, Tetrahedron Asymmetry, 2001, 12,
533; A. Klys, J. Zakrzewski and L. Jerzykiewicz, Tetrahedron
Asymmetry, 2003, 14, 3343.
19 G. C. Fu, Acc. Chem. Res., 2000, 33, 412.
20 M. Uno, K. Ando, N. Komatsuzaki and T. Takahashi, Chem. Commun.,
1992, 964.
21 T. K. Hollis, Y. J. Ahn and F. S. Tham, Organometallics, 2003, 22,
3062.
22 The chromatographic separation can be omitted, with a moderate yield
reduction. In such cases, both complexes can be obtained by fractional
crystallisation from the mother liquor. 10a was only enriched to 94%
de.
23 Details will be provided elsewhere. For a much longer synthesis of
enantiopure phosphacymantrenes, see ref. 17.
Notes and references
† Selected spectrocopic data: 5a: (THF, 298 K) 31P NMR d 236 ppm. 5b:
(THF, 298 K) 31P NMR d 226 ppm. 8a: (THF, 298 K) 31P NMR d 102 ppm.
8b: (THF, 298 K) 31P NMR d 107 ppm. 9a: (Et2O, 298 K) 31P NMR d 238.6
ppm. [a]25 = 45° (c = 1.0, DCM). 9b: (CDCl3, 298 K) 31P NMR d 228.6
D
C h e m . C o m m u n . , 2 0 0 4 , 1 1 4 4 – 1 1 4 5
1145
Typeset and printed by Black Bear Press Limited, Cambridge, England