Synthesis and application of arylmonophosphinoferrocene ligands: ultrafast asymmetric hydrosilylation of styrene.

Article abstract of DOI:10.1021/jo0257804

A short and efficient synthetic route to a novel class of atropisomeric and planar chiral 2-aryl-1-diphenylphosphanylferrocene ligands is presented. The modular design of the ligands allows a synthetic approach in which both the aromatic moiety and the phosphino substituent can be varied easily. This permits fine-tuning of steric and electronic properties. The ligands have been tested in the asymmetric hydrosilylation of styrene where enantioselectivities up to 90% are obtained. Optimization of the palladium-to-ligand ratio resulted in hitherto unparalleled turnover frequencies (TOF) exceeding 180 000 h(-1). The absolute stereochemistry of the ligands was determined from an X-ray structure. 2D NMR experiments in combination with ab initio calculations were used to assign the conformation of the atropisomeric biarylic scaffold.

Full text of DOI:10.1021/jo0257804

Syn th esis a n d Ap p lica tion of Ar ylm on op h osp h in ofer r ocen e
†
Liga n d s: Ultr a fa st Asym m etr ic Hyd r osilyla tion of Styr en e
Henriette Lodberg Pedersen and Mogens J ohannsen*
Department of Chemistry, Building 201, Technical University of Denmark,
DK-2800 Kgs. Lyngby, Denmark
mj@kemi.dtu.dk
Received March 31, 2002
A short and efficient synthetic route to a novel class of atropisomeric and planar chiral 2-aryl-1-
diphenylphosphanylferrocene ligands is presented. The modular design of the ligands allows a
synthetic approach in which both the aromatic moiety and the phosphino substituent can be varied
easily. This permits fine-tuning of steric and electronic properties. The ligands have been tested in
the asymmetric hydrosilylation of styrene where enantioselectivities up to 90% are obtained.
Optimization of the palladium-to-ligand ratio resulted in hitherto unparalleled turnover frequencies
-
1
(
TOF) exceeding 180 000 h . The absolute stereochemistry of the ligands was determined from an
X-ray structure. 2D NMR experiments in combination with ab initio calculations were used to assign
the conformation of the atropisomeric biarylic scaffold.
In tr od u ction
limitations imparted in this ligand type. The problems
arise in reactions where a vacant site on the metal is
needed for substrate coordination in the course of the
catalytic cycle. Due to the strong preference for bidentate
coordination, bisphosphines are in some instances unable
to provide the required coordination site, and hence, they
Asymmetric catalysis is a continuously expanding field
where the development of novel chiral ligands is essential
for ensuring an elaboration of the nearly unendless
possibilities that lie within this area. The ultimate goal
within this field seems to be construction of a universal
ligand, i.e., one that is broadly applicable and substrate
independent. However, a consideration of the diversity
of the ligands unveiled so far and of the reactions in
which they are effective suggests this task to be unrealiz-
able. Thus, the focus of research is shifted toward design
of modular ligand classes that are easily synthesized and
modified to meet the demands of a particular substrate
and reaction.
Among the selection of chiral ligands, optically pure
tertiary phosphines have emerged as extremely versatile
due to their outstanding ability to form highly active and
selective complexes with late-transition metals in par-
ticular. Hence, efficient catalysts have been achieved for
a large number of reactions where the products are
provided in excellent yields and with high enantiomeric
excess. Since the development of the DIOP ligand by
Kagan and co-workers, the design of chiral phosphines
has for a long period been dominated by chelating
6
either inhibit the reaction or give low turnover.
Consequently, chiral monophosphine ligands are highly
important since they show activity and selectivity in
catalytic reactions where the bidentate phosphine-based
ligands have failed. Recent examples have also shown
that easily accessible monophosphine derivatives can
display reactivity and selectivity comparable to or ex-
ceeding bidentate ligands.7 A distinguished reaction
exemplifying the significance of monophosphines is the
palladium-catalyzed asymmetric hydrosilylation of alk-
enes. Herein, palladium-bisphosphine complexes display
no catalytic activity, whereas monophosphine ligands
(
3) For important contributions to this area, see: (a) Marquarding,
D.; Klusacek, H.; Gokel, G.; Hoffmann, P.; Ugi, I. J . Am. Chem. Soc.
970, 92, 5389. (b) Hayashi, T.; Yamamoto, K.; Kumada, M. Tetrahe-
1
dron Lett. 1974, 15, 4405. (c) Hayashi, T.; Mise, T.; Fukushima, M.;
Kagotani, M.; Nagashima, M.; Hamada, Y.; Matsumoto, A.; Kawakami,
S.; Konishi, M.; Yamamoto, K.; Kumada, M. Bull. Chem. Soc. J pn.
1980, 53, 1138. (d) Togni, A.; Breutel, C.; Schnyder, A.; Spindler, F.;
Landert, H.; Tijani, A. J . Am. Chem. Soc. 1994, 116, 4062. For reviews
and accounts, see e.g.: (e) Hayashi, T.; Kumada, M. Acc. Chem. Res.
1
bisphosphine derivatives possessing a chiral scaffold.
1
982, 15, 395. (f) Hayashi, T. Pure Appl. Chem. 1988, 60, 7. (g)
2
3
4
Prominent examples are the BINAP, ferrocene, Trost,
Ferrocenes. Homogeneous Catalysis, Organic Synthesis, Materials
5
Science; Togni, A., Hayashi, T., Eds.; VCH: Weinheim, Germany, 1995.
and Duphos ligands (Figure 1).
(
i) Sawamura, M.; Ito, Y. Chem. Rev. 1992, 92, 857.
In spite of the enormous potential of bidentate phos-
phines, the past decades of research have revealed some
(4) For early contributions, see: (a) Trost, B. M.; Van Vranken, D.
L. Angew. Chem., Int. Ed. Engl. 1992, 31, 228. (b) Trost, B. M.; Van
Vranken, D. L.; Bingel, C. J . Am. Chem. Soc. 1992, 114, 9327. For
overviews, see: (c) Trost, B. M. Acc. Chem. Res. 1996, 29, 355. (d) Trost,
B. M.; Van Vranken, D. L. Chem. Rev. 1996, 96, 395.
†
Synthesis and Application of Arylferrocenyl (Pseudo-Biarylic)
Complexes. 2. For part 1, see ref 13a.
(5) For early contributions, see: (a) Burk, M. J .; Feaster, J . E.;
Harlow, R. L. Organometallics 1990, 9, 2653. (b) Burk, M. J .; Feaster,
J . E.; Harlow, R. L. Tetrahedron: Asymmetry 1991, 2, 569. (c) Burk,
M. J .; Harlow, R. L. Angew. Chem., Int. Ed. Engl. 1990, 29, 1462. (d)
Burk, M. J . J . Am. Chem. Soc. 1991, 113, 8518. For a historical review
and an account, see: (e) Burk, M. J . ChemTracts: Org. Chem. 1998,
11, 787. (f) Burk, M. J . Acc. Chem. Res. 2000, 33, 363.
(
1) (a) Dand, T. P.; Kagan, H. B. Chem. Commun. 1971, 481. (b)
Kagan, H. B.; Dang, T. P. J . Am. Chem. Soc. 1972, 94, 6429.
2) For the first contribution, see: (a) Miyashita, A.; Yasuda, A.;
(
Takaya, H.; Toriumi, K.; Ito, T.; Souchi, T.; Noyori, R. J . Am. Chem.
Soc. 1980, 102, 7932. For an account, see e.g.: (b) Noyori, R.; Takaya,
H. Acc. Chem. Res. 1990, 23, 345.
1
0.1021/jo0257804 CCC: $22.00 © 2002 American Chemical Society
Published on Web 10/24/2002
7
982
J . Org. Chem. 2002, 67, 7982-7994

Synthesis of Arylmonophosphinoferrocene Ligands
ments can be attained via this synthetic strategy (Figure
).
2
F IGURE 2. Monophosphine ligands.
Among axially chiral monophosphines, the MOP ligands
developed by Hayashi and co-workers are outstanding
in terms of reactivity and enantioselectivity. Undoubt-
F IGURE 1. Bisphosphine ligands.
9
edly, the axially chiral binaphthylic scaffold found in, e.g.,
create active palladium species able to afford the product
in high yield.6
a
(8) (a) Hayashi, T.; Tajika, M.; Tamao, K.; Kumada, M. J . Am. Chem.
Soc. 1976, 98, 3718. (b) Hayashi, T.; Tamao, K.; Katsuro, Y.; Nakae,
I.; Kumada, M. Tetrahedron Lett. 1980, 21, 1871. (c) Hayashi, T.;
Konishi, M.; Hioki, T.; Kumada, M.; Ratajczak, A.; Niedbala, H. Bull.
Chem. Soc. J pn. 1981, 54, 3615. (d) Hayashi, T.; Konishi, M.;
Fukushima, M.; Mise, T.; Kagotani, M.; Tajika, M.; Kumada, M. J .
Am. Chem. Soc. 1982, 104, 180. (e) Hayashi, T.; Konishi, M.; Ito, H.;
Kumada, M. J . Am. Chem. Soc. 1982, 104, 4962. (f) Hayashi, T.;
Hagihara, T.; Katsuro, Y.; Kumada, M. Bull. Chem. Soc. J pn. 1983,
Despite the versatility of chiral monophosphines, the
reports on novel ligands of this types are still limited.
The most efficient chiral monophosphine ligands reported
up to this point are based on either a planar chiral
ferrocene scaffold or an axially chiral biaryl scaffold. The
planar chiral ferrocenyl monophosphines developed by
Kumada and Hayashi4 stems from stereoselective de-
rivatization of Ugi’s amine4 and hence contain a chiral
center adjacent to the 1,2-disubstituted ferrocene. The
original dimethyl amino substituent from Ugi’s amine
can be replaced by a variety of heteroatomic nucleophiles
or removed entirely. Consequently, a variety of ligands
with different coordinating ability and steric require-
5
6, 363. (g) Hayashi, T.; Kabeta, K.; Yamamoto, T.; Tamao, K.;
,8
Kumada, M. Tetrahedron Lett. 1983, 24, 5661. (h) Hayashi, T.; Kabeta,
K. Tetrahedron Lett. 1985, 26, 3023. (i) Hayashi, T.; Konishi, M.;
Okamoto, Y.; Kabeta, K.; Kumada, M. J . Org. Chem. 1986, 51, 3772.
(j) Hayashi, T.; Hayashizaki, K.; Kiyoi, T.; Ito, Y. J . Am. Chem. Soc.
a
1
988, 110, 8153. (k) Hayashi, T.; Matsumoto, Y.; Morikawa, I.; Ito, Y.
Tetrahedron: Asymmetry 1990, 1, 151. (l) Han, J . W.; Tokunaga, N.;
Hayashi, T. J . Am. Chem. Soc. 2001, 123, 12915.
(9) For the first contribution, see: (a) Uozumi, Y.; Hayashi, T. J .
Am. Chem. Soc. 1991, 113, 9887. For later papers on the use of MOP-
ligands, see e.g.: (b) Uozumi, Y.; Tanahashi, A.; Lee, S.-Y.; Hayashi,
T. J . Org. Chem. 1993, 58, 1945. (c) Uozumi, Y.; Suzuki, N.; Ogiwara,
A.; Hayashi, T. Tetrahedron 1994, 50, 4293. (d) Hayashi, T.; Niizuma,
S.; Kamikawa, T.; Suzuki, N.; Uozumi, Y. J . Am. Chem. Soc. 1995,
117, 9101. (e) Uozumi, Y.; Kitayama, K.; Hayashi, T.; Yanagi, K.;
Fukuyo, E. Bull. Chem. Soc. J pn. 1995, 68, 713. (f) Hayashi, T.;
Iwamura, H.; Naito, M.; Matsumoto, Y.; Uozumi, Y.; Miki, M.; Yanagi,
K. J . Am. Chem. Soc. 1994, 116, 775. (g) Uozumi, Y.; Lee, S.-Y.;
Hayashi, T. Tetrahedron Lett. 1992, 33, 7185. (h) Uozumi, Y.; Hayashi,
T. Tetrahedron Lett.1993, 34, 2335. (i) Kitayama, K.; Uozumi, Y.;
Hayashi, T. J . Chem. Soc., Chem. Commun. 1995, 1533. (j) Kitayama,
K.; Tsuji, H.; Uozumi, Y.; Hayashi, T. Tetrahedron Lett. 1996, 37, 4169.
(k) Hayashi, T.; Iwamura, H.; Uozumi, Y. Tetrahedron Lett. 1994, 35,
4813. (l) Hayashi, T.; Kawatsura, M.; Iwamura, H.; Yamaura, Y.;
Uozumi, Y. J . Chem. Soc., Chem. Commun. 1996, 1767. (m) Kawatsura,
M.; Uozumi, Y.; Ogasawara, M.; Hayashi, T. Tetrahedron 2000, 56,
2247. (n) Hayashi, T.; Kawatsura, M.; Uozumi, Y. Chem. Commun.
1997, 561. (o) Hayashi, T.; Kawatsura, M.; Uozumi, Y. J . Am. Chem.
Soc. 1998, 120, 1681. (p) Matsumoto, Y.; Naito, M.; Hayashi, T.
Organometallics 1992, 11, 2732. (q) Matsumoto, Y.; Naito, M.; Uozumi,
Y.; Hayashi, T. J . Chem. Soc., Chem. Commun. 1993, 1468. (r) Tsuji,
Y.; Kusui, T.; Kojima, T.; Sugiura, Y.; Yamada, N.; Tanaka, S.; Ebihara,
M.; Kawamura, T. Organometallics 1998, 17, 4835. (s) Imada, Y.; Fujii,
M.; Kubota, Y.; Murahashi, S.-I. Tetrahedron Lett. 1997, 38, 8227. (t)
Oppolzer, W.; Kuo, D. L.; Hutzinger, M. W.; Le’ger, R.; Durand, J .-O.;
Leslie, C. Tetrahedron Lett. 1997, 38, 6213. (u) Hanzawa, Y.; Tabuchi,
N.; Saito, K.; Noguchi, S.; Taguchi, T. Angew. Chem., Int. Ed. 1999,
38, 2395. (v) RajanBabu, T. V.; Nomura, N.; J in, J .; Radetich, B.; Park,
H.; Nandi, M. Chem. Eur. J . 1999, 5, 1963. (w) Sato, Y.; Nishiyama,
T.; Mori, M. J . Org. Chem. 1994, 59, 6133. (x) Stara’, I. G.; Stary, I.;
Kolla’rovic, A.; Teply, F.; Vyskocil, S.; Saman, D. Tetrahedron Lett.
1999, 40, 1993. (y) Sakai, M.; Ueda, M.; Miyaura, N. Angew. Chem.,
Int. Ed.1998, 37, 3279. (z) Hayashi, T.; Ishigedani, M. J . Am. Chem.
Soc. 2000, 122, 976. (a′) Hayashi, T.; Hirate, S.; Kitayama, K.; Tsuji,
H.; Torii, A.; Uozumi, Y. Chem. Lett. 2000, 1272. (b′) Hayashi, T. Catal.
Today 2000, 62, 3. (c′) Hayashi, T.; Hirate, S.; Kitayama, K.; Tsuji,
H.; Torii, A.; Uozumi, Y. J . Org. Chem. 2001, 66, 1441. (d′) Hayashi,
T.; Han, J . W.; Takeda, A.; Tang, J .; Nohmi, K.; Mukaide, K.; Tsuji,
H.; Uozumi, Y. Adv. Synth. Catal. 2001, 343 (3), 279. (e′) Han, J . W.;
Hayashi, T. Chem. Lett. 2001, 976. (g′) Shimada, T.; Mukaise, K.;
Shinohara, A.; Han, J . W.; Hayashi, T. J . Am. Chem. Soc. 2002, 124,
1584. (h′) Han, J . W.; Hayashi, T. Tetrahedron: Asymmetry 2002, 13,
325. For review and accounts on the use of MOP ligands, see: (a)
Hayashi, T. Acta Chem. Scand. 1996, 50, 259. (b) Hayashi, T. Acc.
Chem. Res. 2000, 33, 354.
(
6) Some of the most prominent examples of this are as follows: (a)
The asymmetric hydrosilylation: Uozumi, Y.; Hayashi, T. J . Am. Chem.
Soc. 1991, 113, 9887. (b) The asymmetric hydrovinylation: RajanBabu,
T. V.; Nomura, N.; J in, J .; Radetich, B.; Park, H.; Nandi, M. Chem.
Eur. J . 1999, 5, 1963. (c) The asymmetric arylation of imines: Hayashi,
T.; Ishigedani, M. J . Am. Chem. Soc. 2000, 122, 976. (d) The
asymmetric Grignard cross-coupling: Hayashi, T.; Hayashizaki, K.,
Kiyoi, T.; Ito, J . J . Am. Chem. Soc. 1988, 110, 8153. See also: (e)
Matsumoto, Y.; Naito, M.; Uozumi, Y.; Hayashi, T. Chem. Commun.
993, 1468. (f) Tsuji, Y.; Kusui, T.; Kojima, T.; Sugiura, Y.; Yamada,
N.; Tanaka, S.; Ebihara, M.; Kawamura, T. Organometallics 1998, 17,
835. (g) Imada, Y.; Fujii, M.; Kubota, Y.; Murahashi, S.-I. Tetrahedron
Lett. 1997, 38, 8227. (h) Star a´ , I. G.; Stary, I.; Koll a´ rovic, A.; Teply,
F.; Vyskocil, S.; Saman, D. Tetrahedron Lett. 1999, 40, 1993. (i) Sato,
Y.; Nishiyama, T.; Mori, M. J . Org. Chem. 1994, 59, 6133. (j) Oppolzer,
W.; Kuo, D. L.; Hutzinger, M. W.; L e´ ger, R.; Durand, J .-O.; Leslie, C.
Tetrahedron Lett. 1997, 38, 6213. (k) Hanzawa, Y.; Tabuchi, N.; Saito,
K.; Noguchi, S.; Taguchi, T. Angew. Chem., Int. Ed. Engl. 1999, 38,
1
4
2
395. (l) Hayashi, T. J . Organomet. Chem. 2002, 653 (1-2), 41. (m)
Han, J . W.; Tokunaga, N.; Hayashi, T. Synlett 2002, 871. (n) For the
asymmetric arylation of aldehydes, see: Sakai, M.; Ueda, M.; Miyaura,
N. Angew. Chem., Int. Ed. 1998, 37, 3279.
(7) For examples where monophosphines give higher selectivity than
the usual bisphosphines, see e.g.: (a) Feringa, B. L.; Pineschi, M.;
Arnold, L. A.; Imbos, R.; de Vries, A. H. M. Angew. Chem., Int. Ed.
Engl. 1997, 36, 2620. For examples where monophosphines give similar
selectivities as bisphosphines, see e.g.: (b) Reetz, M. T.; Mehler, G.
Angew. Chem., Int. Ed. 2000, 39, 3889 (c) Kitayama, K.; Uozumi, Y.;
Hayashi, T. Chem. Commun. 1995, 1533. (d) Tye, H.; Smyth, D.;
Eldred, C.; Wills, M. Chem. Commun. 1997, 1053. (e) Graf, C. D.;
Malan, C.; Harms, K.; Knochel, P. J . Org. Chem. 1999, 64, 5581. (f)
Zhu, G.; Chen, Z.; J iang, Q.; Xiao, D.; Cao P.; Zhang, X. J . Am. Chem.
Soc. 1997, 119, 3836. (g) Bartels, B.; Helmchen, G. Chem. Commun.
1
999, 741. For examples where monophosphines display superior
reactivity, see e.g.: (h) Old, D. V.; Wolfe, J . P.; Buchwald, S. L. J . Am.
Chem. Soc. 1998, 120, 9722. (i) Wolfe, J . P.; Buchwald, S. L. Angew.
Chem., Int. Ed. Engl. 1999, 38, 2413. (j) Aranyos, A.; Old, D. W.;
Kiyomori, A.; Wolfe, J . P.; Sadighi, J . P.; Buchwald, S. L. J . Am. Chem.
Soc. 1999, 121, 4369. (k) Wolfe, J . P.; Singer, R. A.; Yang, B. H.;
Buchwald, S. L. J . Am. Chem. Soc. 1999, 121, 9550. (l) Fox, J . M.;
Huang, X.; Chieffi, A.; Buchwald, S. L. J . Am. Chem. Soc. 2000, 122,
1
360.
J . Org. Chem, Vol. 67, No. 23, 2002 7983

Pedersen and J ohannsen
BINOL,10 BINAP, and MOP ligands creates chiral sur-
roundings that are highly efficient exemplified by their
application in a range of catalytic reactions covering most
of the field of asymmetric synthesis.11 The elaboration
of this ligand class from diol over bisphosphine to
monophosphine elegantly exemplifies how modifications
on the chiral scaffold can lead to ligands that meet the
demands of different reaction types. Within the MOP
ligands, variation of the substituent in the 2′-position and
on the phosphorus atom demonstrates a spectacular fine-
tuning of the ligands toward a variety of substrates in
the palladium-catalyzed asymmetric hydrosilylation9
for fine-tuning reactivity and selectivity. Furthermore,
it should be noted that the modular design imparts the
possibility of constructing families of ligands with varying
coordination number and atoms in two to three steps,
e.g., P-N, P-O, and P-P derivatives.
Here, the methodology has been applied to the syn-
thesis of the aryl-monophosphinoferrocene (MOPF)
ligands. Within this particular ligand array, the steric
and electronic properties of the aryl moiety have been
varied. The importance of these modifications has been
examined in the palladium-catalyzed asymmetric hy-
drosilylation of styrene. The ligands are shown to be
extremely efficient with respect to rate and afford enan-
tioselectivities comparable to the best results previously
reported.
a′-h′
(Figure 2).
Herein, we wish to present a class of monophosphine
ligands 1 containing a novel pseudo-biarylic scaffold,
which is constituted of an arylferrocene framework
(Figure 3). The phosphino moiety is attached to the
Liga n d Syn th esis
ferrocene adjacent to the aryl group rendering the ligands
planar chiral. Depending on the substitution pattern on
the aryl group restricted rotation around the arylfer-
rocene bond can lead to atropisomery (pseudoaxial chiral-
The modular approach to the MOPF ligands requires
a common ferrocene scaffold that by simple operations
can be modified into an array of different ligands.
Additionally, the chiral auxiliary on the ferrocene should
be removed entirely to allow for direct attachment of both
the aromatic moiety and the phosphino group onto the
ferrocene. A suitable precursor for this purpose was found
in the easily accessible ferrocenyl sulfoxide 2. By
diastereoselective ortho-lithation of 2 and concomitant
tert-butyllithium-mediated sulfoxide cleavage, the opti-
cally pure arylferrocenylmonophosphines should be avail-
able in two steps (Scheme 1).
ity), consequently adding to the steric diversity of the
ligands.12
1
4
SCHEME 1
F IGURE 3. General structure of (S)-aryl-MOPF ligands.
Although the efficiency of both planar chiral and
axially chiral ligands is well established, the combination
of those two systems is still unexplored.13 We envisioned
that having an aryl substituent directly linked to a
monophosphinoferrocene could lead to ligands that dis-
play reactivity and selectivity similar to other biarylic
systems. However, opposed to axially chiral binaphthylic
derivatives, the chirality of the ligands presented here
does not rely on restricted rotation around the biarylic
bond and virtually any aromatic group can be introduced,
retaining the inherent planar chirality of the 1,2-disub-
stituted ferrocene. Consequently, derivatization in the
pseudo-biarylic system can easily be accomplished. The
short synthetic routes described facilitate preparation of
arrays of ligands, rendering this approach highly useful
To construct the pseudo-biarylic system, we chose to
focus on the Negishi cross-coupling. None of the aryl
groups introduced in the reaction sequence are sensitive
toward the rather basic reaction conditions present in
this reaction. Furthermore, the Negishi cross-coupling
does not require the synthesis of a precursor and the
ferrocenyl sulfoxide can be used directly, hence lowering
the number of synthetic steps. The initial attempts to
synthesize the aryl-MOPF ligands involved the known
intermediate 3 (Scheme 2).1 Treatment of this with tert-
butyllithium removed the sulfoxide quantitatively, but
all attempts to use the lithiated species in a Negishi
cross-coupling failed and only the hydrolyzed cleavage
product was isolated. Apparently, the steric bulk of the
(10) For early contributions, see: (a) Kyba, E. B.; Koga, K.; Sousa,
L. R.; Siegel, M. G.; Cram, D. J . J . Am. Chem. Soc. 1973, 95, 2692. (b)
Norori, R.; Tomino, I.; Tanimoto, Y. J . Am. Chem. Soc. 1979, 101, 3129.
For overviews, see e.g.: (c) Pu, L. Chem. Rev. 1998, 98, 2405. (d) Rosini,
C.; Franzini, L.; Raffaelli, A.; Salvadori, P. Synthesis 1992, 503.
4a
(11) Asymmetric Synthesis; Hayashi, T., Tomioka, K., Yonemitsu,
O., Eds.; Gordon and Breach Science Publishers: Tokyo, 1998. See also
refs 2, 9, and 10.
(12) The term axial chiral should preferably be avoided in this
context since these compounds are not axial chiral but rather rotational
diastereomers. Note that the two conformers of the fragment containing
the naphthyl and ferrocene molecule with, e.g., the naphthyl group in
an exo and endo position are not enantiomorphic, e.g., super imposable
mirror images of each other.
(14) This ortho-lithiation chemistry was originally developed by
Kagan et al.: (a) Riant, O.; Argourch, G.; Guillaneux, D.; Samuel, O.;
Kagan, H. B. J . Org. Chem. 1998, 63, 3511. See also: (b) Hua, D. H.;
Lagneau, N. M.; Chen, Y.; Robben, P. M.; Clapham, G.; Robinson, P.
D. J . Org. Chem. 1996, 61, 4508.
(13) Part of this work has been communicated: (a) Pedersen, H. L.;
J ohannsen, M. Chem. Commun. 1999, 2517. For the synthesis of
another chiral arylferrocenyl derivative, see: (b) Bringmann, G.;
Hinrichs, J .; Peters, K.; Peters, E.-V. J . Org. Chem. 2001, 66, 629.
7
984 J . Org. Chem., Vol. 67, No. 23, 2002

Synthesis of Arylmonophosphinoferrocene Ligands
SCHEME 2
borane-protected diphenylphosphino group prevents
trans-metalation to palladium. In one case, the anion was
trapped with chlorotrimethylsilane giving, after removal
of the borane protecting group, ligand 4. Attempts to
introduce more bulky silyl electrophiles at this stage were
unsuccessful, confirming the reduced access to this
position when situated adjacent to the sterically demand-
ing borane adduct.
Finally, a third approach based on electrophilic addition
of chloro diphenylphosphine oxide and reduction was
examined (Table 4).
As seen in Table 2 (entries 1 and 2), chlorodiphen-
ylphosphine reacted smoothly with the quantitatively
formed lithiated ferrocenyl species, and using strictly
anhydrous workup procedures the phosphines 1a and 1b
could be isolated in good yields without oxidation to the
corresponding phosphine oxide. Introduction of an o- or
p-methoxy substituent on the aromatic moiety resulted
in a more sensitive reaction with varying amounts of
oxidized byproduct (Table 2, entries 3 and 4). Surpris-
ingly, the reaction with 5e was very prone to oxidation,
and the corresponding phosphine oxide was the major
product in all instances. On the basis of these results, it
seems that the oxidation potential of the phosphino
moiety in the aryl-MOPF ligands is strongly influenced
by the electronic and steric properties of the aryl sub-
stituent on the ferrocene.
To overcome this obstacle, the order of events was
reversed. First, the biarylic scaffold was constructed by
a palladium-catalyzed Negishi cross-coupling between the
metalated ferrocenyl sulfoxide and an aryl iodide. Suc-
cessfully, a range of electron-rich and -poor aromatics
could be introduced in high yields and with complete
diastereo- and atroposelectivity (Table 1). No other
diastereoisomer or rotamer could be detected according
1
to H NMR, indicating a diastereoselectivity of >98%
(vide infra). Examination of different phosphine ligands
for palladium revealed that 2-trifurylphosphine gave the
highest yield in the coupling although the difference was
not decisive.
TABLE 2. Dir ect Syn th esis of th e P h osp h in es (S)-1a -d
TABLE 1. Dia ster eoselective Or th o-Lith ia tion -Negish i
Ar yla tion of (S)-2 Yield in g (SP ,SS)-5a -e
MOPF
ligand
stereochemistry
of liganda
entry
sulfoxide
yield/%
1
2
3
4
5a
5b
5c
5d
1a
1b
1c
1d
S
S
S
S
77
70
34-76
38
b
stereochemistry reaction
a
The stereochemistry is based on an X-ray structure of 1a and
by chemical correlation with 2 (vide infra). b The yield is dependent
a
b
entry product
of product
time/h
T /°C yield/%
1
2
3
4
5
5a
5b
5c
5d
5e
SP,SS
SP,SS
SP,SS
SP,SS
SP,SS
48
96
48
96
96
rt
rt
reflux
rt
rt
77
60
90
95
88
on the amount of phosphine oxide formed.
Next, the approach where the phosphines were pro-
tected in situ as the borane adducts was examined. This
procedure turned out to be the most efficient for the
synthesis of most of the ligands. After the in situ
protection, the derivatives 6b-e could be isolated in good
yields (Table 3).
a
The subscripts “P” and “S” designate the descriptor of planar
chirality and the chirality at sulfur, respectively. b Ortho-lithiation
and trans-metalation were carried out at -78 °C.
To complete the synthesis of the aryl-MOPF ligands,
the phosphino group had to be introduced in the second
step. Three approaches were evaluated. Direct entrap-
ment of the ferrocenyl anion with chlorodiphenylphos-
phine would give immediate access to the desired product
Interestingly, the borane group in adduct 6b was
relatively labile due to the moderately more electron poor
phosphino moiety. The borane group was partly released
during aqueous work up of 6b affording 44% of the
desired product in combination with 32% of the deborated
and oxidized derivative (Table 3, entry 1). More electron
rich aryl groups on the ferrocene led to good yields of the
desired products (Table 3, entries 2-4). After facile
deprotection with DABCO, the pure phosphines could be
(Table 2). However, knowing that phosphines are ex-
tremely prone to oxidation, in situ protection with borane
could be necessary to afford stable adducts that could be
purified and then deprotected prior to use (Table 3).
J . Org. Chem, Vol. 67, No. 23, 2002 7985

Pedersen and J ohannsen
TABLE 3. Syn th esis of MOP F Liga n d s 1b-e via th e Bor a n e Ad d u cts
entry
sulfoxide
MOPF ligand
stereochemistry of liganda
yield of 6b-e/%
yield of 1b-e/%
1
2
3
4
5b
5c
5d
5e
1b
1c
1d
1e
S
S
S
S
44 (76)b
73
63
93
75
96
88
86
a
The stereochemistry is based on an X-ray structure of 1a and by chemical correlation with 2 (vide infra). b 32% of the deborated and
oxidized compound was also isolated.
TABLE 4. Syn th esis of (S)-1a , (S)-1c, a n d (S)-1e via th e P h osp h in e Oxid es
entry
sulfoxide
MOPF ligand
method
stereochemistry of liganda
yield of 7/%
yield of 1/%
1
2
3
5a
5c
5e
1a
1c
1e
A, 1
B, 1
B, 2
S
S
S
82
75
67
73
71
85
a
The stereochemistry is based on an X-ray structure of 1a and by chemical correlation with 2 (vide infra).
isolated in high yields. This procedure also turned out
to be efficient for the synthesis of aryl-MOPF ligand 1e,
with an acceptable overall yield of 76%.
3 4
protocol applying CeCl in combination with LiAlH
turned out to be successful for this particular phosphine
oxide affording the product 1e in 85% yield corresponding
to an overall yield of 57% starting from 5e.15 Although
the phosphines 1a , 1c, and 1e can be synthesized in
acceptable yields via reduction of the corresponding
phosphine oxides, the reaction sequence involving the
borane adduct was preferable due to the easy handling
and purification of these compounds.
Finally, synthesis of the phosphine oxides followed by
reduction was tested (Table 4). Potentially, this method
could improve the yield since chlorodiphenylphosphine
oxide is expected to be a more reactive electrophile. This
was, however, only the case for the simple phenyl
derivative 7a (compare entry 1 in Tables 2 and 4). In the
case of the two substrates 7c and 7e, the reactivity of
the electrophile was overruled by the steric properties
of the biarylic system, and the sterically more demanding
phosphine oxide electrophile afforded low yields (<50%)
in both reactions. The most effective way to synthesize
the phosphine oxides finally turned out to be the in situ
oxidation of the initially formed phosphine. This could
easily be accomplished by entrapment of the ferrocenyl
anion with chlorodiphenylphosphine and basic workup
affording the phosphine oxides 7c and 7e in good yields
The phosphines 1a -e are all stable when stored cold
and in the dark. Exposure to a diluted basic reaction
media though leads to various amounts of oxidation.
Str u ctu r a l a n d Con for m a tion a l An a lysis
The assigned stereochemistry for ligand 1a is con-
firmed by an X-ray crystal structure (Figure 4). It is
revealed from this structure that the dihedral angle
between the substituted Cp-ring and the phenyl sub-
stituent is 38°.
(
Table 4, entries 2 and 3). Reduction of 7a and 7c under
standard conditions furnishes the MOPF ligands 1a and
c in 60% and 53% overall yield, respectively. The
naphthyl-substituted phosphine oxide 7e was reduced
slowly in the presence of SiHCl and Et N providing only
1
(
15) (a) Henson, P. D.; Naumann, K.; Mislow, K. J . Am. Chem. Soc.
1
969, 91, 5645. (b) Imamoto, T.; Kikuchi, S.; Miura, T.; Wada, Y. Org.
3
3
Lett. 2001, 3, 87. (c) Imamoto, T.; Takeyama, T.; Kusumoto, T. Chem.
Lett. 1985, 1491. (d) J ohnson, C. R.; Imamoto, T. J . Org. Chem. 1987,
small amount of the product. Prolonged reaction times
led to decomposition of the substrate making optimiza-
tion of this procedure difficult. An alternative reduction
5
1
2, 2170. (e) Koide, Y.; Sakamoto, A.; Imamoto, T. Tetrahedron Lett.
991, 32, 3375. (f) Imamoto, T.; Hikosaka, T. J . Org. Chem. 1994, 59,
6753.
7
986 J . Org. Chem., Vol. 67, No. 23, 2002

Synthesis of Arylmonophosphinoferrocene Ligands
F IGURE 6.
F IGURE 4. View of phenyl-MOPF (S)-1a . Displacement
1
6
ellipsoids are drawn at the 50% probability level.
To clarify the behavior of these novel ligands in
solution and possibly get some valuable help to design
novel ligands, a study of the ligands dynamic behavior
was undertaken.
F IGURE 7.
As a starting point, ab initio calculations on the simple
phenyl-substituted ferrocene (Figure 5) were performed.
The energy of activation for rotation around the phenyl-
respectively could be expected to have a higher barrier
to rotation, which might give rise to atropisomers (Fig-
ures 6 and 7). A favored orientation would, in turn, be
ferrocene bond was found to be less than for ethane (E
10 vs 12 kJ /mol). This corresponds to completely free
a
)
1
revealed in the NMR spectra of the ligands. The H NMR
rotation around the biarylic bond and indicates that the
repulsion between the lower Cp-ring and the phenyl
group is negligible. In the calculated lowest energy
conformation, the dihedral angle was found to be 35°,
which is close to the value found in the solid state.
spectrum of 1c showed that one proton is considerably
more deshielded compared to the rest of the spin system.
Assignment by 1D-TOCSY and NOE difference estab-
d
lished that this absorption stems from H (Figure 6a).
d
An orientation of H below the upper Cp-ring would lead
to a downfield shift (vide supra).17 A NOESY experiment
confirms this orientation since dipolar coupling is ob-
d
served between H and the lower Cp-ring protons (H′).
The methoxy protons on the other hand did not show any
dipolar coupling to H′, indicating no close proximity to
the lower Cp-ring.
The phenyl groups on phosphorus are believed to be
positioned similar to the orientation seen in the crystal
structure of 1a (Figure 6b). The phenyl group pointing
F IGURE 5.
Introduction of the 1-diphenylphosphino substituent on
the ferrocene scaffold might lead to restricted rotation
around the biarylic bound, due to steric interactions
between the two substituents. If the phenyl group were
incapable of rotating freely within the NMR time scale,
the protons on the phenyl substituent located above the
upper Cp-ring would no longer be chemically equivalent
to the protons located below the upper Cp-ring. The
protons located below the Cp-ring should be relatively
more deshielded due to anisotropic effects from the
cyclopentadienide ligand and the iron atom. The actual
o
out of the plane has NOE between the ortho protons (H )
and protons on both Cp rings, verifying the conformation
in Figure 6b. The other phenyl group is coupled through
space to ring protons of the anisole substituent.
The same series of experiments were conducted for 1e
and 7e. For both derivatives, the NOESY experiment
verified the rotamer depicted in Figure 7. Moreover, the
phenyl groups on phosphorus showed the same dipolar
coupling found for 1c corresponding to an orientation
similar to 1c (Figure 6b) and the solid-state structure of
1a .
1
H NMR shows a complex higher order spectrum where
two equivalent protons are deshielded relative to the
three remaining protons. Based on the coupling pattern
The NMR data obtained for the ligands 1c and 1e
heavily suggest that they exist as an atropisomer with
1
3
and correlation with C NMR, we assign the least
shielded signal in the spectrum to the two ortho protons
on the phenyl group. Accordingly, no restricted rotation
around the phenyl-ferrocene bond is indicated, and we
expect the aryl group in the simple MOPF ligand 1a to
rotate with a low barrier of activation.
(16) Cr ysta l d a ta : C28H23Fe P, M ) 446.28, monoclinic, a ) 9.332-
(
2) Å, b ) 11.638(2) Å, c ) 10.114(2) Å, â ) 106.22(3)°, U ) 1054.6(4)
3
1
Å , T ) 120(2) K, space group P2 (no. 4), Z ) 2, µ(Mo KR) ) 0.803
-1
mm , 7514 reflections measured, 4770 unique (Rint ) 0.041) and 3720
reflections with I > 2σ(I) that were used in all calculations. The Flack
factor was -0.05(2), indicating that the absolute configuration is
2
correct. The final R1 was 0.0496 and wR(F ) was 0.1138 (all data).
In contrary to the ligand 1a , the ligands 1c and 1e
carrying the o-MeO phenyl and 1-naphthyl substituent
(17) For a related observation, see: Uemura, M. J . Org. Chem. 1996,
61, 1375.
J . Org. Chem, Vol. 67, No. 23, 2002 7987

Pedersen and J ohannsen
SCHEME 3. Hyd r osilyla tion a n d Oxid a tion of
Styr en e
TABLE 5. Asym m etr ic Hyd r osilyla tion of Styr en e w ith
Liga n d 1a -e
mol % Pd/
mol %
ligand
reaction reaction conversion ee of
a
b,c
entry ligand
temp/°C time/h
to 8 /%
9 /%
1
2
3
4
5
1a
1b
1c
1d
1e
1/2
1/2
1/2
1/2
1/2
rt
rt
rt
rt
rt
1.5
8
0.5
1.6
1.5
100
100
100
100
100
76 (S)
25 (S)
68 (S)
86 (S)
79 (S)
the sterically demanding ortho substituent located above
the ferrocene system. This implies a more rigid biarylic
system where the rotational freedom of the aryl subtitu-
ent is restricted compared to ligand 1a .
a
1
b
Conversion determined by H NMR. ee determined by HPLC
on a DAICEL OD-H column. c Absolute configuration determined
by optical rotation.
Asym m etr ic Hyd r osilyla tion of Styr en e
TABLE 6. Asym m etr ic Hyd r osilyla tion of Styr en e w ith
Catalytic asymmetric functionalization of alkenes is an
important task in organic chemistry. The palladium-
catalyzed asymmetric hydrosilylation of double bonds is
a highly potent hydrometalation reaction that displays
excellent regioselectivity for aryl-substituted olefins.
Among the various transformations organosilanes can
undergo, the Tamao-Flemming oxidation is extremely
important.18 This stereospecific oxidation converts chiral
silanes into alcohols with complete retention of stereo-
chemistry. Hence, the overall reaction sequence of hy-
drosilylation and oxidation is equivalent to an enantio-
selective Markovnikov hydration of an olefin (Scheme 3).
Until recently, only Hayashi’s MOP ligands have
displayed high efficiency with respect to reactivity and
selectivity in the hydrosilylation of olefins. Thus, the
palladium-catalyzed hydrosilylation of styrene with MOP
ligand 10 affords upon oxidation 1-phenylethanol with
Liga n d s 1c-e
mol % Pd/
mol %
ligand
reaction reaction conversion ee of
a
b,c
entry ligand
temp/°C time/h
to 8 /%
9 /%
1
2
3
1c
1d
1e
0.1/0.2
0.1/0.2
0.1/0.2
rt
rt
rt
4
5.5
3.5
100
100
100
71 (S)
90 (S)
85 (S)
a
1
b
Conversion determined by H NMR. ee determined by HPLC
on a DAICEL OD-H column. c Absolute configuration determined
by optical rotation.
5
, entry 2), indicating that a relative electron-rich aryl
substituent is beneficial for the reactivity. However, a
comparison with the reaction rate for the electronically
similar p-MeO-Ph-MOPF 1d (Table 5, entry 4) reveals
that the electronic properties are not exclusively decisive
for the conversion rate. Apparently, the steric require-
ments of the ligands likewise influence the reactivity.
Tamao oxidation of the organosilanes affords (S)-1-
phenylethanol with an enantiomeric excess ranging from
9
a′,c′
9
8% enantiomeric excess.
Earlier this year, we dis-
closed the palladium-catalyzed hydrosilylation of aro-
matic alkenes with Feringa’s chiral phosphoramidite
ligand 11.19 Here, hydrosilylation and oxidation of styrene
affords 1-phenylethanol with 99% enantiomeric excess,
which is the highest ever reported.
2
3
5% to 86%. The 3,5-(CF )-Ph-MOPF ligand 1b only
induces a small selectivity, and combined with the
relatively low reactivity of this ligand, it seems that this
aromatic moiety is a poor second component for the
pseudo-biarylic scaffold in the ligands (Table 5, entry 2).
As seen in entries 1, 3, 4, and 5 in Table 5, there is no
obvious correlation between the observed enantioselec-
tivies and the electronic and steric properties of the
aromatic moiety in aryl-MOPF ligands 1a , c, d , and e.
The presumed rigidity of the pseudo-biarylic scaffold in
o-MeO-Ph-MOPF 1c seems to be unfavorable providing
(S)-1-phenylethanol with only 68% ee. Similarly, the
relatively rigid biarylic scaffold in 1-Naph-MOPF 1e has
only a slightly positive effect on the selectivity compared
to the simple and more flexible Ph-MOPF 1a , 79% ee and
The aryl-MOPF ligands were screened in the pal-
ladium-catalyzed asymmetric hydrosilylation of styrene
in order to explore the influence of the aromatic moiety
on the reactivity and selectivity. Commencing with 1 mol
7
6% ee, respectively. The highest enantioselectivity is
provided by p-MeO-Ph-MOPF 1d , hence suggesting that
a combination of an electron-rich aryl substituent with
steric flexibility is optimal.
Prompted by the initial results, the investigation of the
aryl-MOPF ligands was continued. In the following
experiments, the catalyst loading was decreased to 0.1
mol % palladium and 0.2 mol % of ligand 1c, d , or e
%
palladium and 2 mol % aryl-MOPF ligand 1a -e
styrene was smoothly hydrosilylated in the presence of
excess of trichlorosilane (Table 5). In all cases, 100%
conversion to 1-phenyl(trichlorosilyl)ethane was observed
within a short time (0.5-8 h). The o-MeO-Ph-MOPF 1c
results in the highest conversion rate (Table 5, entry 3),
(Table 6). Again, all the catalyses proceeded efficiently.
3
whereas 3,5-(CF )-Ph-MOPF 1b gives the lowest (Table
A 2-8-fold increase in reaction times and a noteworthy
increase in asymmetric induction was observed. The
highest selectivity was observed for ligand 1d , where (S)-
(
18) Tamao, K. Oxidative Cleavage of the Silicon-Carbon Bond:
Development, Mechanism, Scope and Limitations. In Advances in
Silicon Chemistry; Larson, G. L., Ed.; J AI Press: Greenwich, CT, 1996;
Vol. 3, pp 1-62.
1
-phenylethanol was obtained with 90% ee upon oxida-
tion (Table 6, entry 2). Remarkably, the reaction time in
(
19) J ensen, J . F.; Svendsen, B. Y.; la Cour, T. V.; Pedersen, H. L.;
J ohannsen, M. J . Am. Chem. Soc. 2002, 124, 4558.
the catalysis with 1e is increased only by approximately
7
988 J . Org. Chem., Vol. 67, No. 23, 2002

Synthesis of Arylmonophosphinoferrocene Ligands
SCHEME 4. Ch a lk -Ha r r od Mech a n ism for th e
Hyd r osilyla tion of Styr en e
TABLE 7. Hyd r osilyla tion of Styr en e w ith Liga n d s 1d ,e
a n d P P h 3
mol % Pd/ reaction reaction conversion ee of
to 8a (%)
9 /%
b,c
entry ligand mol %ligand temp/°C
time
1
2
3
4
5
6
7
8
9
0
1
1e
1e
1e
1e
1e
1d
1d
PPh
PPh
PPh
PPh
0.1/0.16
0.1/0.14
0.1/0.12
0.1/∼0.10
0.1/0.06
0.1/0.14
0.1/∼0.1
1/2
1/1.5
1/1.2
1/∼1
rt
rt
rt
rt
rt
rt
rt
rt
rt
rt
rt
5 min
100
100
100
100
100
100
100
54
63 (S)
69 (S)
62 (S)
80 (S)
79 (S)
79 (S)
86 (S)
d
20 s
20 s
d
18 h
25 h
15 min
24 h
1 h
1 h
1 h
1 h
3
3
3
3
84
94
35
1
1
a
1
b
Conversion determined by H NMR. ee determined by HPLC
c
on a DAICEL OD-H column. Absolute configuration determined
by optical rotation. d The reaction proceeds for 20 s, but the total
time from addition of silane to precipitation of Pd is 2 min.
of styrene with 2 equiv of aryl-MOPF ligand per pal-
ladium is a yellow homogeneous solution, and no color
change is observed upon addition of the reagents. The
observed color change might indicate the formation of a
different Pd-species in the reactions where only a slight
excess of the ligand is applied. Immediately after the
color change, a highly exothermic reaction is initiated.
This is maintained for ∼20 s, after which time palladium
a factor 2 when the catalyst loading is decreased by a
factor 10 (Table 6, entry 3).
The hydrosilylation of styrene in the presence of a
palladium-monophosphine complex is assumed to pro-
ceed via the Chalk-Harrod mechanism depicted in
Scheme 4.20
Presumably, the first step in the catalysis with ligand
1
1
e where a phosphine needs to dissociate to open a
black precipitates. H NMR of the reaction before forma-
coordination site on palladium for the incoming substrate
is a rate-limiting process. Lowering the concentration of
palladium and phosphine ligand should affect this equi-
librium, pushing it toward the alkene-coordinated pal-
ladium complex 14. Hence, a decrease in catalyst loading
can be revealed in an increase in turnover frequency. The
origin of the enhanced selectivity could probably be found
in the same relative rate enhancement. Background
reactions leading to racemic 1-phenyl(trichlorosilyl)-
ethane, 2-phenyl-1-(trichlorosilyl)ethane, and ethylben-
zene will be suppressed when the monophosphine-
palladium complex is the dominating catalyst. Consequent-
ly, a relative decrease of the inactive period in the
catalytic system decreases the relative amount of product
formed via unwanted side reactions. The consequences
of lowering the catalyst loading are most profound for
ligand 1e, although the catalyses applying 1d and 1c
show similar tendencies.
tion of the red complex shows only small amounts of the
product, whereas complete conversion is observed after
precipitation of palladium. This strongly indicates that
the reaction mainly takes place in the period between
the color change and the precipitation. Due to lack of
precise kinetic measurements no absolute TOF number
can be calculated, but based solely on the reaction time
of 20 s the catalyst must turn over with a frequency of
-1 21
at least 180 000 h
. Importantly, the enantioselectivity
is largely maintained during the ultrafast reactions, and
only a small decrease is affected. This is properly due to
the highly exothermic nature of the reaction.
The reaction becomes more problematic when ap-
proaching a 1:1 ratio of palladium and ligand 1e. Due to
oxidation of small amounts of the phosphine in the
reaction media, it is not possible to have exactly one
phosphine per palladium. Consequently, either a fast
reaction indicating a slight excess of the phosphine or a
heterogeneous reaction indicating an excess of palladium
is attained. Surprisingly, the latter reaction proceeds,
although slowly, and full conversion to the product is
observed after 18 h (Table 7, entry 4). Even with a 0.1:
0.06 ratio of palladium and 1e, a catalytically active
species is generated affording 100% conversion to the
product after 25 h (Table 7, entry 5).
To gain more information on the relative rate enhance-
ment observed when decreasing the palladium and ligand
concentration, a series of experiments were conducted
where the palladium/ligand quotient was varied (Table
7
). A palladium concentration of 0.1 mol % was retained,
but the concentration of the phosphine ligand was varied.
The first experiment performed with 0.16 mol % 1e
revealed a highly interesting observation (Table 7, entry
To explore the generality of the observed rate enhance-
ment, we examined the aryl-MOPF ligand 1d as well as
PPh3 under similar conditions. An analogous behavior
was observed when applying 0.14 mol % of 1d and 0.1
1
1
). H NMR after 2 min showed only traces of styrene,
and after 5 min, full conversion was obtained. This
indicates that the palladium-to-ligand ratio is a decisive
factor for the conversion time. Further lowering the
amount of ligand 1e to 0.14 and 0.12 mol %, respectively,
gave the fastest turnover observed (entries 2 and 3, Table
(21) At present, we can only give an underestimate of the actual
turnover frequency based on the complete reaction time, i.e., 20 s for
each catalytically active species to turn over 1000 molecules (0.1 mol
7
). In both experiments, the addition of trichlorosilane
%
catalyst) (i.e., 180 000 turnovers per hour). The closest match we
led to a profound color change within the first 30 s,
furnishing a bright red solution. A typical hydrosilylation
have found is in a recently published paper by Hayashi et al. By
modifying the standard MOP ligand to a fluoro derivative they
increased the TOF from 83 at 0 °C to around 1000 at -24 °C. See:
Hayashi, T.; Hirate, S.; Kitayama, K.; Tsuji, H.; Torii, A.; Uozumi, Y.
J . Org. Chem. 2001, 66, 1441.
(20) Chalk, A. J .; Harrod, J . F. J . Am. Chem. Soc. 1965, 87, 16.
J . Org. Chem, Vol. 67, No. 23, 2002 7989

Pedersen and J ohannsen
SCHEME 5
mol % Pd. Although not as fast as Pd-1e catalyst, the
reaction still finished within 15 min (Table 7, entry 6).
A lower activity of the Pd-1d complex compared to Pd-
upon coordination of a phosphine forms 16. Finally,
reductive elimination releases the silane, and regenerates
Pd(0)P . In cycle B, an alkene is coordinated to complex
2
1
e was already established in the catalysis under stan-
dard conditions (entries 2 and 3, Table 6). Palladium-
PPh complexes display a profound lower activity in the
hydrosilylation of styrene, and the catalyses are con-
ducted with 1 mol % palladium. Applying 2 mol % PPh
15, yielding complex 18. Reductive elimination from 18
again reduces palladium, this time in the form of complex
17. Oxidative addition of trichlorosilane to 17 reinitiates
the catalytic cycle by forming 14. Contemplating, that
phosphines are superior to alkenes as ligands for pal-
ladium, it appears reasonable, that cycle A will dominate
when the palladium-to-phosphine ratio is 1:2. Accord-
ingly, 16 will be the prevailing species formed from 15,
trapping the catalyst in cycle A. Lowering the amount of
phosphine facilitates the generation of complex 18,
favoring cycle B. Furthermore, it has been postulated,
that reductive elimination occurs faster from a palladium
complex bearing only one phosphine ligand, which could
add to the speed of cycle B, if this step turned out to be
rate determining under these conditions.23
Summarizing the above, two factors could influence the
observed relative rate of the hydrosilylation. The first
factor is the unfavored ligand exchange between complex
13 and 14 when 2 equiv of phosphine ligand is present
(slow A cycle) vs the generation of complex 18 when less
than 2 equiv of phosphine is used (fast B cycle). With
similar reasoning, we expect the aryl-MOPF ligands to
be superb in terms of stabilizing low-valent palladium
intermediates as, e.g., 17, thereby facilitating turnovers
via the fast cycle B. This stabilization could be steric or
due to pseudo-bidentate coordination to the aryl group
and the phosphine in the ligand. The second factor could
be the presumed faster reductive elimination from com-
plex 18 compared to the reductive elimination from
complex 16. It should be stressed that the mechanistic
considerations presented here is our working hypothesis.
At present we do not have kinetic measurements or any
exact knowledge about the rate-determining step to
support our conclusions.
3
3
affords 54% conversion within 1 h (Table 7, entry 8). As
demonstrated by entries 9 and 10 (Table 7), lowering the
amount of phosphine ligand to 1.5 and 1.2 mol %,
respectively, did indeed increase the reaction rate, al-
though the rates still are far from those obtained with
the aryl-MOPF ligands. A conversion of 84% and 94%,
respectively, was attained within 1 h (Table 7, entries 9
and 10). The catalysis with a 1:1 palladium triphen-
ylphosphine ratio provided the lowest rate with 35%
conversion after 1 h, which is similar to the effect
observed for the aryl-MOPF ligands although not as
profound (Table 7, entry 11).
Consequently, it seems like a general observation that
simply lowering the phosphine-to-palladium ratio en-
hances the rate of hydrosilylation. A large difference in
absolut reaction rate between the different Pd-ligand
complexes, however, persists, stating that the rate also
is greatly influenced by the electronic and steric proper-
ties of the ligand.
To account for the drastic effect observed upon de-
creasing the amount of phosphine, we invoke the com-
petitive action of two catalytic cycles (A and B, Scheme
5
). Cycle A commences with the oxidative addition of
trichlorosilane to the Pd(0)P species 12, affording the
6-electron complex 13 having two coordination sites
2
1
occupied by phosphine ligands. A ligand exchange reac-
tion must follow, allowing coordination of the alkene. This
equilibrium is expected to be far to the left due to the
high affinity of palladium for coordinating phosphines,
thereby favoring complex 13 over 14.22 It is well-known
that bisphosphines are poor ligands for the palladium-
catalyzed hydrosilylation. The lack of reactivity can be
explained by their reluctance to liberate a vacancy on the
metal for the incoming alkene. Migratory insertion from
Con clu sion
A class of novel arylmonophosphinoferrocene ligands
has been introduced to the field of catalytic asymmetric
synthesis. The modular design of the ligands was dem-
onstrated to be highly successful for the construction of
1
4 affords the transient 14-electron complex 15, which
(22) For other data supporting the importance of monophosphine
complexes in hydrosilylations, see: Marinetti, A. Tetrahedron Lett.
1
994, 35, 5861.
(23) Gillie, A.; Stille, J . K. J . Am. Chem. Soc. 1980, 102, 4937.
7
990 J . Org. Chem., Vol. 67, No. 23, 2002

Synthesis of Arylmonophosphinoferrocene Ligands
an array of aryl-MOPF ligands with different electronic
and steric properties. The short synthesesstwo to three
steps starting from the optically pure ferrocenyl sulfoxide
solution of Pd
2
(dba)
3
3
‚CHCl (0.09 g, 0.09 mmol, 4 mol % Pd)
and 2-trifurylphosphine (0.09 g, 0.38 mmol, 9.25 mol %) in
anhyd THF (3 mL). The reaction was stirred at rt for 48 h,
and after acidic workup the crude reaction mixture was
purified by flash chromatography (pentane/ethyl acetate/
dichloromethane 69/22/9) to afford 1.28 g (3.19 mmol, 77%) of
2
smakes the design outstandingly effective for accessing
a wide variety of ligands.
We have shown that the new class of aryl-MOPF
ligands disclosed in this paper is among the best ligands
so far developed for the asymmetric hydrosilylation of
styrene affording products with up to 90% ee. Further-
more, the extremely high turn over frequency effectuated
by the MOPF ligands brings the hydrosilylation in to a
new era where industrial application becomes an option.24
We are currently pursuing the extension of our design
to other ligand types as well as the optimization of the
MOPF ligands. Initial experiments show that the aryl
group on the ferrocene scaffold has significant influence
on the rate of â-elimination contra the rate of reductive
elimination. This is currently being explored in the
hydrosilylation of aliphatic terminal alkenes. Due to the
excellent reactivity displayed by the MOPF ligands, we
expect them to show interesting results in a variety of
palladium-catalyzed bond-constructing reactions and are
currently pursuing this task. This work will be reported
in due course.
P S
(S ,S )-5a as a yellow amorphous powder together with 0.29
g (0.88 mmol, 21%) of recovered (S)-2: H NMR (300 MHz) δ
7.80 (d, J ) 7.5 Hz, 2H), 7.71 (d, J ) 7.5 Hz, 2H), 7.33 (m,
1
5
(
1
H), 4.71 (m, 1H), 4.43 (m, 1H), 4.16 (s + m, 5H + 1H), 2.43
s, 3H); 13C NMR (75 MHz) δ 141.5 (2C), 136.0, 130.0 (2C),
29.5 (2C), 128.4 (2C), 127.4, 125.9 (2C), 90.0, 72.4 (2C), 71.4
+
(5C), 69.7 (2C), 21.8; MS (EI) m/ z 400 [M ]; HRMS (ES, m/z)
+
calcd for C23
H20FeOS (M ) 400.0584, found 400.0584.
(
S
P
S
,S )-2-[3,5-Bis(tr iflu or om eth yl)ph en yl-1-(p-tolylsu lfi-
n yl)fer r ocen e (5b). Compound 5b was prepared according
to the general procedure. Purification by flash chromatography
(pentane/ethyl acetate 88/12) afforded 60% of (S
P S
,S )-5b as a
red amorphous powder: 1H NMR (500 MHz) δ 8.16 (bs, 2H),
7
4
3
3
2
.69 (s, 1H), 7.41 (d, J ) 7.5 Hz, 2H), 7.15 (d, J ) 7.5 Hz, 2H),
.73 (m, 1H), 4.61 (m, 1H), 4.57 (m, 1H), 4.34 (s, 5H), 2.33 (s,
H); 13C NMR (75 MHz) δ 141.0, 140.3, 139.5, 131.2 (q, J )
3 Hz, 2C), 130.1 (2C), 129.4 (2C), 124.9 (2C), 123.5 (q, J )
73 Hz, 2C), 120.6, 94.4, 85.4, 72.8, 71.9 (5C), 70.0 (2C), 21.4;
+
MS (EI) 536 [M ]; HRMS (ES, m/ z) calcd for C H F FeOS
(M ) 536.0332, found 536.0325.
2
5
18
6
+
(S ,S )-2-(o-An isole)-1-(p-t olylsu lfin yl)fer r ocen e (5c).
P
S
Compound 5c was prepared according to the general proce-
dure. Purification by flash chromatography (pentane/ethyl
acetate 67/33) afforded 90% of (S ,S )-5c as a red-orange
P S
Exp er im en ta l Section
1
1
amorphous powder: H NMR (300 MHz) δ 8.02 (dd, J ) 7.4
Gen er a l In for m a tion . All moisture- and air-sensitive
manipulations were carried out under an argon atmosphere
using standard Schlenk techniques. Glassware was flame- or
oven-dried prior to use. Argon gas was dried by passage
Hz, J 2 ) 1.1 Hz, 1H), 7.64 (d, J ) 8.0 Hz, 2H), 7.29 (m, 3H),
7
.06 (dd, J ) J 2 ) 7.4 Hz, 1H), 6.85 (d, J ) 8.2 Hz, 1H), 4.77
1
(
m, 1H), 4.44 (m, 1H), 4.23 (m, 1H), 4.18 (s, 5H), 3.73 (s, 3H),
2
1
7
.43 (s, 3H); 13C NMR (75 MHz) δ 157.6, 141.1, 141.0, 133.9,
through P
under nitrogen. Benzene, dichloromethane, and toluene were
distilled from CaH . t-BuLi (1.5 or 1.7 M solution in pentane),
n-BuLi (1.6 M solution in hexane), ZnCl (0.5 M solution in
THF), BH ‚THF (1 M solution in THF), and SiHCl were used
2 5
O . THF was distilled from sodium/benzophenone
29.3 (2C), 128.9, 125.6 (2C), 123.8, 120.4, 111.1, 94.1, 86.3,
+
4.1, 71.2 (5C), 69.2, 68.4, 55.6, 21.7; MS (EI) m/ z 430 [M ];
2
+
HRMS (ES, m/z) calcd for C24
30.0609.
,S )-2-(p-An isole)-1-(p-tolylsu lfin yl)fer r ocen e (5d ).
H
22FeO
2
S (M ) 430.0609, found
2
4
3
3
(
S
P
S
as received without further purification. LDA (0.5 M THF
solution) was prepared immediate prior to use.
Compound 5d was prepared according to the general proce-
dure. Purification by flash chromatography (pentane/diethyl
TLC analyses were performed on Merck silica gel 60 F254
plates, using UV light and a 5% solution of phosphormolybdic
acid in ethanol for visualization. Ferrocenes could be seen
directly as yellow-brown spots. Purification by flash chroma-
tography was performed using Merck silica gel 60 (0.035-
ether 50/50) afforded 95% of (S
P
,S
S
)-5d as a yellow amorphous
2
0
1
powder: [R]
D
) -29 (c ) 1.0, CHCl ); H NMR (300 MHz) δ
3
7
2
3
.69 (m, 4H), 7.29 (d, J ) 6.0 Hz, 2H), 6.87 (d, J ) 6.0 Hz,
H), 4.63 (m, 1H), 4.35 (m, 1H), 4.11 (s, 5H), 4.07 (m, 1H),
.81 (s, 3H), 2.40 (s, 3H); 13C NMR (75 MHz) δ 159.1, 141.5
0
.070 mm or 0.040-0.063 mm).
1
1
3
(2C), 131.1 (2C), 129.5 (2C), 128.1, 125.9 (2C), 113.9 (2C), 90.2,
H NMR (300 MHz) and C NMR (75 MHz) spectra were
1
13
72.1 (2C), 71.3 (5C), 69.3 (2C), 55.5, 21.7; MS (EI) m/ z 430
recorded at 30 °C. H NMR (500 MHz), C NMR (125 MHz),
+
+
3
1
[M ]; HRMS (ES, m/z) calcd for C24
H
22FeO
2
S (M ) 430.0609,
and P NMR (202 MHz) spectra were recorded at 30 °C.
Chemical shifts are reported in ppm using CHCl (δ 7.27) and
CDCl
found 430.0609.
(S ,S )-2-(1-Na p h th yl)-1-(p-tolylsu lfin yl)fer r ocen e (5e).
P S
Compound 5e was prepared according to the general proce-
dure. Purification by flash chromatography (pentane/ethyl
3
1
13
3
(δ 77.0) as internal standard for H and C, respec-
3
3
1
tively. H
4
PO (85%) was used as external standard for P.
Coupling constants are reported in hertz.
Gen er a l P r oced u r e: Syn th esis of (S ,S )-2-P h en yl-1-
P S
acetate 75/25) affords 88% of (S
P
,S
S
)-5e as a yellow amorphous
3
2
0
1
powder: [R]
D
) 41 (c ) 1.0, CHCl ); H NMR (500 MHz) δ
(
p -t olylsu lfin yl)fer r ocen e (5a ) b y Dia st er eoselect ive
8
.24 (d, J ) 7.3 Hz, 1H), 7.82 (d, J ) 8.1, 1H), 7.77 (d, J ) 8.1
1
Or t h o-Met a la t ion a n d Con com it a n t Negish i Cr oss-
Cou p lin g of F er r ocen yl Su lfoxid e (S)-2. To a stirred
suspension of sulfoxide (S)-2 (1.34 g, 4.13 mmol) in anhyd THF
Hz, 1H), 7.56 (dd, J ) 8.1 Hz, J 2 ) 7.3 Hz, 1H), 7.45 (d, J )
2
8
.1 Hz, 1H), 7.35 (dd, J 1 ) 8.1 Hz, J ) 7.3 Hz, 1H), 7.26 (d,
J ) 7.9 Hz, 2H), 7.20 (ddd, J ) 8.1 Hz, J 2 ) 7.3 Hz, J ) 1.3
1
3
(
15 mL, 0.3 M) at -78 °C under argon was added LDA (9.90
Hz, 1H), 6.92 (d, J ) 7.9 Hz, 2H), 4.64 (m, 1H), 4.57 (m, 1H),
mL, 0.5 M, 4.95 mmol) dropwise via syringe. The resulting
4
1
1
7
.53 (m, 1H), 4.38 (s, 5H), 2.19 (s, 3H); 13C NMR (75 MHz) δ
40.6, 140.4, 133.5, 133.4, 131.5, 131.4, 128.9 (2C), 128.3,
28.1, 125.9, 125.8, 125.5, 125.0, 124.6 (2C), 96.5, 89.7, 74.1,
orange-red solution was stirred at -78 °C for 30 min before
ZnCl (9.10 mL, 0.5 M, 4.54 mmol) was added slowly via
2
syringe. After being stirred for 30 min at -78 °C, the reaction
was slowly warmed to rt over 1 h. To the resulting yellow
solution of the trans-metalated ferrocenyl sulfoxide was added
iodobenzene (2.11 g, 10.32 mmol) followed by a preformed
+
1.3 (5C), 69.1, 68.7, 21.4; MS (EI): m/z) 450 [M ]; HRMS
+
(ES, m/z) calcd for C27
H
22FeOS (M ) 450.0741, found 450.0741.
Syn th esis of (S)-2-P h en yl-1-(d ip h en ylp h osp h a n yl)fer -
r ocen e (1a ) by Selective Rem ova l of th e Su lfoxid e a n d
Dir ect In tr od u ction of th e P h osp h in o Gr ou p . To a stirred
solution of (S ,S )-5a (0.10 g, 0.25 mmol) in anhyd THF (5 mL,
P S
0.05 M) under argon at -78 °C was added t-BuLi (0.18 mL,
1.7 M, 0.30 mmol) dropwise. The reaction was stirred for 5
(
24) To make a reaction industrially relevant, TOF of at least 10 000
is required: Blaser, H.-U.; Spindler, F. In Comprehensive Asymmetric
Catalysis; J acobsen, N. J ., Pfaltz, A., Yamamoto, H., Eds.; Springer:
Heidelberg, 1999; Chapter 41.1
J . Org. Chem, Vol. 67, No. 23, 2002 7991

Pedersen and J ohannsen
min at -78 °C before Ph
2
PCl (122 mg, 0.55 mmol) was added
mL, 1.7 M, 1.20 mmol) dropwise. The reaction was stirred for
5 min at -78 °C before Ph PCl (0.40 g, 1.81 mmol) was added
2
dropwise to the dark red solution of the lithiated ferrocene.
dropwise to the dark red solution of the lithiated ferrocene.
After being stirred for 30 min at -78 °C, the reaction was
heated quickly to rt, diluted with 5 mL of diethyl ether, filtered
through silica gel, and concentrated under reduced pressure.
The resulting residue was flash chromatographed (pentane/
diethyl ether 98/2) to afford 0.09 g (1.93 mmol, 77%) of the
After the mixture was stirred at -78 °C for 30 min, BH ‚THF
3
(3.60 mL, 1.0 M, 3.60 mmol) was added slowly to the reaction
via syringe. The reaction was heated quickly to rt and stirred
at that temperature for 1 h followed by quenching with 2 M
NaOH and extraction with diethyl ether. The organic layer
1
product as a yellow amorphous powder: H NMR (500 MHz)
δ 7.65 (d, J ) 7.0 Hz, 2H), 7.60 (m, 2H), 7.41 (m, 3H), 7.25-
was washed with saturated brine, dried over MgSO , filtered,
4
7
.16 (m, 8 H), 4.76 (m, 1H), 4.42 (m, 1H), 4.03 (s, 5H), 3.83
and concentrated under reduced pressure. The resulting
residue was filtered through a short column of silica gel to
afford 0.19 g (0.32 mmol, 44% yield) of the product as an orange
1
3
(m, 1H); C NMR (125 MHz) δ 140.7 (d, J C-P ) 11.4 Hz, 1C),
1
2
2
38.8, 138.7 (d, J C-P ) 12.4 Hz, 1C), 136.1 (d, J C-P ) 21.4 Hz,
C), 132.9 (d, J C-P ) 17.4 Hz, 2C), 130.0 (d, J C-P ) 6.9 Hz,
C), 129.8, 128.8 (m, 4C), 128.5 (2C), 128.4, 127.1, 94.0 (d, J C-P
amorphous powder together with 0.05 g (0.25 mmol) of Ph
2
P-
)H: H NMR (300 MHz) δ 7.71-7.63 (m, 4H), 7.60 (bs,
H), 7.53-7.40 (m, 6H), 7.38-7.26 (m, 2H), 4.68 (m, 1H), 4.60
1
(BH
3
)
21.4 Hz, 1C), 75.4 (d, J C-P ) 11.5 Hz, 1C), 73.0 (m, 2C),
1
31
7
4
4
1.3 (bs, 5C), 70.5; P NMR (202 MHz) δ -20.3; MS (EI) m/z
13
(
(
1
m, 1H), 4.40 (s, 5H), 4.19 (m, 1H), 1.40-1.18 (m, 3H); C NMR
75 MHz) δ 139.4, 133.1 (d, J C-P ) 9.2 Hz, 2C), 131.9 (2C),
+
+
46 [M] ; HRMS (ES, m/z) calcd for C28
69.0784, found 469.0786.
H
23FePNa (M )
31.4, (d, J C-P ) 11.8 Hz, 2C), 130.9 (2C), 130.0 (d, J C-P
)
)
(
S)-2-[3,5-Bis(tr iflu or om eth yl)ph en yl]-1-(diph en ylph os-
64.4 Hz, 1C), 129.3 (d, J C-P ) 70.4 Hz, 1C), 129.2 (d, J C-P
p h a n yl)fer r ocen e (1b). Compound 1b was synthesized ac-
cording to the procedure described for 1a Purification by flash
chromatography (pentane/diethyl ether 98/2) afforded the
10.4 Hz, 2C), 128.7 (d, J C-P ) 9.6 Hz, 2C), 120.8, 91.0 (d, J C-P
) 10.0 Hz, 1C), 75.5 (m, 2C), 71.6 (s, 5C), 70.9 (d, J C-P ) 7.2
Hz, 1C), 70.6 (d, J C-P ) 61.5 Hz, 1C).
1
product as a yellow amorphous powder: yield 70%; H NMR
(
S)-2-(o-An isole)-1-[d ip h en ylp h osp h a n yl(bor a n e)]fer -
(
3
3
300 MHz) δ 7.95 (bs, 2H), 7.57 (s, 1H), 7.55 (m, 2H), 7.38 (m,
r ocen e (6c). The compound was synthesized from (S ,S )-5c
P
S
H), 7.16 (m, 5H), 4.71 (m, 1H), 4.42 (m, 1H), 3.98 (s, 5H),
according to the procedure described for 6b. Purification by
flash chromatography (pentane/diethyl ether 95/5) afforded the
1
3
.86 (m, 1H); C NMR (75 MHz) δ 141.7, 138.8 (d, J C-P ) 10.0
Hz, 1C), 137.0 (d, J C-P ) 10.1 Hz, 1C), 135.3 (d, J C-P ) 21.5
Hz, 2C), 132.7 (d, J C-P ) 19.1 Hz, 2C), 131.1 (q, J C-F ) 32.6
Hz, 2C), 129.6, 129.2 (bs, 2C), 128.6 (d, J C-P ) 7.3 Hz, 4C),
2
0
product as a yellow amorphous powder: yield 73%; [R]
D
)
1
-47 (c ) 0.92, CHCl
3
); H NMR (300 MHz) δ 7.67-7.60 (m,
1 2
2
)
4
0
)
5
1
)
H), 7.50-7.30 (m, 7H), 7.23-7.10 (m, 3H), 6.70 (ddd, J ) J
1
2
23.6 (q, J C-F ) 272.2 Hz, 2C), 120.0 (bs, 2C), 89.8 (d, J C-P
)
7.5 Hz, J 3 ) 0.9 Hz, 1H), 6.56 (d, J ) 7.5 Hz), 4.58 (m, 1H),
1.0 Hz, 1C), 76.4 (d, J C-P ) 12.3 Hz, 1C), 73.5 (d, J C-P ) 3.9
.54 (m, 1H), 4.44 (m, 1H), 4.33 (s, 5H), 3.44 (s, 3H), 1.53-
Hz, 1C), 72.6 (d, J C-P ) 2.9 Hz, 1C), 71.2 (5C), 70.7; HRMS
.93 (m, 3H); 13C NMR (75 MHz) δ 158.1, 134.5, 133.3 (d, J C-P
+
(
6
ES) m/z calcd for C30
05.0533.
S)-2-(o-An isole)-1-(diph en ylph osph an yl)fer r ocen e (1c).
21 6
H F FePNa (M ) 605.0532, found
5.5 Hz, 2C), 133.2 (d, J C-P ) 5.7 Hz, 2C), 132.1 (d, J C-P
9.2 Hz, 1C), 130.7 (d, J C-P ) 58.0 Hz, 1C), 130.6 (d, J C-P
.2 Hz, 1C), 130.5 (d, J C-P ) 0.8 Hz, 1C), 128.8, 128.2 (d, J C-P
10.1 Hz, 2C), 128.0 (d, J C-P ) 10.7 Hz, 2C), 124.2, 119.7,
)
)
(
Compound 1c was synthesized according to the procedure
described for 1a . Purification by flash chromatography (pen-
tane/diethyl ether 98/2) afforded the product as an orange
1
7
1
4
09.4, 92.0 (d, J C-P ) 9.4 Hz, 1C), 76.3 (d, J C-P ) 6.7 Hz, 1C),
4.1 (d, J C-P ) 10.6 Hz, 1C), 71.1 (5C), 70.3 (d, J C-P ) 7.4 Hz,
1
amorphous powder: yield 34-76%; H NMR (300 MHz) δ 8.00
+
C), 69.1 (d, J C-P ) 64.5 Hz, 1C), 54.5; MS (EI) m/z 490 [M] ,
(
6
d, J ) 8.0 Hz, 1H), 7.61 (m, 2H), 7.39 (m, 3H), 7.16 (m, 6H),
+
76 [M - BH
3
].
1
2
3
.96 (ddd, J ) J ) 8.0 Hz, J ) 1.0 Hz, 1H), 6.74 (d, J ) 8.0
(
S)-2-(p-An isole)-1-[d ip h en ylp h osp h a n yl(bor a n e)]fer -
Hz, 1H), 4.71 (m, 1H), 4.45 (m, 1H), 4.03 (s, 5H), 3.92 (m, 1H),
3
Hz, 1C), 139.1 (d, J C-P ) 9.8 Hz, 1C), 135.6 (d, J C-P ) 21.8
Hz, 2C), 133.7 (d, J C-P ) 6.7, 1C), 132.4 (d, J C-P ) 17.5 Hz,
2
)
J
5
1
3
r ocen e (6d ). The compound was synthesized from (S ,S )-5d
P
S
.49 (s, 3H); C NMR (75 MHz) δ 157.8, 140.6 (d, J C-P ) 9.8
according to the procedure described for 6b. Purification by
flash chromatography (pentane/diethyl ether 95/5) afforded the
1
product as a yellow amorphous powder: yield 63%; H NMR
C), 129.2, 128.3 (d, J C-P ) 4.1 Hz, 2C), 128.2, 127.8 (d, J C-P
5.8 Hz, 2C), 127.4, 125.8, 120.0, 110.6, 91.7 (m, 2C), 74.3 (d,
(
2
500 MHz) δ 7.59 (m, 2H), 7.42 (m, 6H), 7.26 (m, 2H), 7.11 (m,
H), 6.58 (m, 2H), 4.59 (m, 1H), 4.51 (m, 1H), 4.31 (s, 5H),
C-P ) 3.5 Hz, 1C), 71.2 (d, J C-P ) 4.3 Hz, 1C), 70.5 (5C), 69.6,
1
3
3
1
+
4.21 (m, 1H), 3.74 (s, 3H), 1.51-1.20 (m, 3H); C NMR (125
MHz) δ 159.3, 134.0 (d, J C-P ) 9.1 Hz, 2C), 133.6 (d, J C-P
)
5.1; P NMR (202 MHz) δ -20.3; MS (EI) m/z 476 [M] ;
+
HRMS (ES, m/z) calcd for C29H25FeOPNa (M ) 499.0890, found
4
99.0890.
9.9 Hz, 2C),132.7 (2C), 131.9 (d, J C-P ) 57.8 Hz, 1C), 131.4 (d,
C-P ) 3.0 Hz, 1C), 131.3 (d, J C-P ) 2.3 Hz, 1C), 131.0 (d, J C-P
57.0 Hz, 1C), 128.9 (d, J C-P ) 4.5 Hz, 2C), 128.8 (d, J C-P
.5 Hz, 2C), 128.7, 113.4 (2C), 94.6 (d, J C-P ) 8.4 Hz, 1C), 75.8
J
)
4
(
S)-2-(p-An isole)-1-(diph en ylph osph an yl)fer r ocen e (1d).
)
Compound 1d was synthesized according to the procedure
described for 1a . Purification by flash chromatography (pen-
tane/diethyl ether 98/2) afforded the product as a yellow
amorphous powder: yield 38%; H NMR (300 MHz) δ 7.61-
7
7
(d, J C-P ) 6.9 Hz, 1C), 75.3 (d, J C-P ) 9.9 Hz, 1C), 71.6 (5C),
1
70.7 (d, J C-P ) 7.6 Hz, 1C), 69.4 (d, J C-P ) 61.4 Hz, 1C), 55.9;
+
MS (EI) m/z ) 476 [M - BH
3
].
.54 (m, 2H), 7.56 (d, J ) 8.7 Hz, 2H), 7.41-7.38 (m, 3H),
.22-7.13 (m, 5H), 6.78 (d, J ) 8.7 Hz, 2H), 4.70 (m, 1H), 4.38
(S)-2-(1-Naph th yl)-1-[diph en ylph osph an yl(bor an e)]fer -
r ocen e (6e). The compound was synthesized from (S ,S )-5e
1
3
(m, 1H), 4.01 (s, 5H), 3.79 (m, 1H), 3.77 (s, 1H); C NMR (75
P
S
MHz) δ 158.5, 140.3 (d, J C-P ) 10.1 Hz, 1C), 138.2 (d, J C-P
)
according to the procedure described for 6b. Purification by
1
1
0.8 Hz, 1C), 135.6 (d, J C-P ) 20.6 Hz, 2C), 132.4 (d, J C-P
)
flash chromatography (pentane/diethyl ether 98/2) afforded the
1
7.7 Hz, 2C), 130.5 (d, J C-P ) 6.2 Hz, 2C), 130.3, 129.3, 128.3
product as a yellow amorphous powder: yield 86%; H NMR
(
J
d + s, J C-P ) 6.2 Hz, 2C + 2C), 127.9, 113.5 (2C), 93.6 (d,
C-P ) 22.1 Hz, 1C), 74.6 (d, J C-P ) 10.9 Hz, 1C), 72.2 (d, J C-P
3.6 Hz, 1C), 72.1 (d, J C-P ) 4.1 Hz, 1C), 70.7 (5C), 69.8,
(300 MHz) δ 7.61-7.51 (m, 3H), 7.46-7.39 (m, 3H), 7.26-7.16
(m, 4H), 7.13-7.00 (m, 5H), 6.90-6.84 (m, 2H), 4.61 (m, 1H),
1
3
)
4.56 (m, 1H), 4.53 (m, 1H), 4.40 (s, 5 H), 1.5-1.0 (m, 3H);
C
3
1
+
5
5.4; P NMR (202 MHz) δ -20.4; MS (EI) m/ z 476 [M ];
NMR (75 MHz) δ 133.6 (d, J C-P ) 10.4 Hz, 2C), 132.7 (d, J C-P
) 9.2 Hz, 2C), 132.5 (d, J C-P ) 84.2 Hz, 1C), 130.7, 130.2 (d,
J C-P ) 154.1 Hz, 1C), 128.3-127.6 (m, 9C), 126.0, 125.4 (d,
+
HRMS (ES, m/ z) calcd for C29H25FeOPNa (M ) 499.0890,
found 499.0889.
Syn th esis of (S)-2-[3,5-Bis(tr iflu or om eth yl)p h en yl]-1-
d ip h en ylp h osp h a n yl(bor a n e)]fer r ocen e (6b). To a stirred
J
J
C-P ) 8.3 Hz, 2C), 125.1, 93.5 (d, J C-P ) 6.4 Hz, 1C), 76.2 (d,
C-P ) 5.8 Hz, 1C), 74.6 (d, J C-P ) 12.0 Hz, 1C), 71.4 (d, J C-P
[
solution of (S
P
,S
S
)-5b (0.39 g, 0.72 mmol) in anhyd THF (14
) 63.0 Hz, 1C), 74.3 (5C), 70.4 (d, J C-P ) 7.8 Hz, 1C); MS (EI)
m/z 510 [M] , 496 [M - BH ].
+
+
mL, 0.05 M) under argon at -78 °C was added t-BuLi (0.55
3
7
992 J . Org. Chem., Vol. 67, No. 23, 2002

Synthesis of Arylmonophosphinoferrocene Ligands
Syn th esis of (S)-2-[3,5-Bis(tr iflu or om eth yl)p h en yl]-1-
4
with brine, dried over MgSO , filtered, and concentrated under
(
6
d ip h en ylp h osp h a n yl)fer r ocen e (1b) via Debor a tion of
b. A 0.19 g (0.317 mmol) portion of 6b was dissolved in anhyd
reduced pressure. The resulting residue was purified flash
chromatography (pentane/ethyl acetate 50/50) to afford 0.08
g (0.16 mmol, 76%) of the product as an orange-red amorphous
powder: 1H NMR (300 MHz) δ 8.04 (d, J ) 7.5 Hz, 1H), 7.76
toluene (19 mL, 0.02 M). DABCO (0.18 g, 1.59 mmol) was
added, and the mixture was heated to 60 °C for 30 min. After
concentration under reduced pressure, the crude residue was
purified by flash chromatography (pentane/diethyl ether 98/
(m, 2H), 7.45 (m, 6H), 7.25 (m, 1H), 7.14 (m, 1H), 7.04 (ddd,
1
) J 2 ) 7.5 Hz, J 3 ) 1.8 Hz, 1H), 6.54 (d, J ) 8.1 Hz, 1H),
J
2
) to afford 0.17 g (0.30 mmol, 93%) of (S)-1b as a red-orange
amorphous powder.
S)-2-(o-An isole)-1-(diph en ylph osph an yl)fer r ocen e (1c).
4.71 (m, 1H), 4.47 (m, 1H), 4.38 (s, 5H), 4.07 (m, 1H), 3.51 (s,
3H); 13C NMR (75 Mz) δ 157.4, 135.1 (d, J C-P ) 105.5 Hz, 1C),
1
1
(
34.7, 134.2 (J C-P ) 107.0 Hz, 1C), 131.8 (J C-P ) 9.5 Hz, 2C),
31.7 (J C-P ) 9.6 Hz, 2C), 131.3 (J C-P ) 2.3 Hz, 1C), 130.9
C-P ) 1.7 Hz, 1C), 128.5, 128.1 (J C-P ) 11.7 Hz, 2C), 127.7
(
Compound 6c was deborated according to the procedure
described for 1b. Purification of the crude residue by flash
chromatography (pentane/diethyl ether 98/2) afforded (S)-1c
as a red-orange amorphous powder in 75% yield.
J
J
(
C-P ) 12.2 Hz, 2C), 124.7, 120.0, 109.8, 89.9 (J C-P ) 11.2
Hz, 1C), 75.6 (J C-P ) 10.2 Hz, 1C), 74.4 (J C-P ) 15.1 Hz, 1C),
7
1
4.1 (J C-P ) 112.7 Hz, 1C), 71.1 (5C), 69.8 (J C-P ) 11.3 Hz,
(
S)-2-(p-An isole)-1-(diph en ylph osph an yl)fer r ocen e (1d).
31
+
C), 55.1; P NMR (202 MHz) δ 29.1; MS (EI) m/ z 492 [M ];
Compound 6d was deborated according to the procedure
described for 1b. Purification of the crude residue by flash
chromatography (pentane/diethyl ether 98/2) afforded (S)-1d
as a yellow amorphous powder in 96% yield.
+
HRMS (ES, m/z) calcd for C29
92.0942.
S)-2-(1-Na p h t h yl)-1-(d ip h en ylp h osp h in yl)fer r ocen e
7e). Compound 7e was synthesized according to the procedure
2
H25FeO P (M ) 492.0942, found
4
(
(
(
S)-2-(1-Na p h t h yl)-1-(d ip h en ylp h osp h a n yl)fer r ocen e
described for 7c. Purification by flash chromatography (pen-
(
1e). Compound 6e was deborated according to the procedure
tane/ethyl acetate 60/40) afforded the product as a yellow
described for 1b. Purification of the crude residue by flash
chromatography (pentane/diethyl ether 98/2) afforded (S)-1e
as a yellow amorphous powder in 88% yield: H NMR (500
MHz) δ 8.38 (d, J ) 7.3 Hz, 1H), 7.91 (d, J ) 8.0 Hz, 1H), 7.74
(
1
amorphous powder: yield 68%; H NMR (500 MHz) δ 8.34 (dd,
1
) 7.3 Hz, J 2 ) 1.3 Hz, 1H), 7.77 (m, 2H), 7.70 (d, J ) 8.5
1
J
Hz, 1H), 7.66 (d, J ) 7.3, 1H), 7.57 (d, J ) 8.1 Hz, 1H), 7.50
1
2
(m, 1H), 7.43 (m, 2H), 7.37-7.25 (m, 3H), 7.21 (m, 2H), 6.95
m, 2H), 7.61 (m, 2H), 7.54 (dd, J ) J ) 8.0 Hz, 1H), 7.42
(
m, 1H), 6.81 (m, 2H), 4.76 (m, 1H), 4.59 (m, 1H), 4.52 (s, 5H),
(
m, 3H), 7.33 (m, 2H), 7.12 (m, 5H), 4.76 (m, 1H), 4.54 (m,
1
3
H), 4.15 (s, 5H), 3.92 (m, 1H) 13C NMR (75 Mz) δ 139.4 (d,
4.18 (m, 1H); C NMR (75 MHz) δ 134.3 (d, J C-P ) 105.2 Hz,
1
J
J
1
1
(
J
1
C), 132.9, 132.7, 132.6 (d, J C-P ) 106.7 Hz, 1C), 132.2, 131.6,
31.3 (d, J C-P ) 9.3 Hz, 2C), 131.1 (d, J C-P ) 3.6 Hz, 1C), 131.0
C-P ) 11.2 Hz, 1C), 138.6 (d, J C-P ) 10.4 Hz, 1C), 135.3 (d,
C-P ) 21.4 Hz, 2C), 133.9, 133.6, 133.5, 132.2 (d, J C-P ) 17.4
d, J C-P ) 9.8 Hz, 2C), 130.2 (d, J C-P ) 2.6 Hz, 1C), 127.9 (d,
Hz, 2C), 131.1, 131.0, 129.2, 128.3 (m, 2C), 127.8 (d, J C-P
)
C-P ) 4.1 Hz, 2C), 127.8, 127.3, 127.1, 126.9, 125.3, 125.1,
24.8 (d, J C-P ) 4.7 Hz, 2C), 92.3 (d, J C-P ) 10.8 Hz, 1C), 76.4
6
1
1
7
4
5
.2 Hz, 2C), 127.6 (d, J C-P ) 8.7 Hz, 2C), 126.1, 125.5, 125.4,
25.0, 94.2 (d, J C-P ) 23.6 Hz, 1C), 78.9 (d, J C-P ) 9.5 Hz,
C), 74.1 (d, J C-P ) 2.9 Hz, 1C), 71.3 (d, J C-P ) 4.3 Hz, 1C),
(
d, J C-P ) 111.5 Hz, 1C), 75.1 (d, J C-P ) 9.8 Hz, 1C), 74.3 (d,
3
1
J C-P ) 11.4 Hz, 1C), 70.9 (5C), 69.4 (d, J C-P ) 10.8 Hz, 1C),
0.8 (5C), 69.6; P NMR (202 MHz) δ -21.5; MS (EI) m/ z
31
+
+
+
53.3; P NMR (202 MHz) δ 27.8; MS (EI) m/ z 512 [M ]; HRMS
96 [M ]; HRMS (ES, m/z) calcd for C32
H
25FePNa (M )
+
(
ES, m/z) calcd for C32
Syn th esis of (S)-2-P h en yl-1-(d ip h en ylp h osp h a n yl)fer -
r ocen e (1a ) by Red u ction of 7a . To a stirred solution of 7a
0.20 g, 0.43 mmol) in anhyd toluene (22 mL, 0.02 M) were
added Et N (1.49 mL, 10.75 mmol) and SiHCl (5.97 mmol).
H25FeOP (M ) 512.0992, found 512.0993.
19.0941, found 519.0941.
(
S)-2-P h en yl-1-(d ip h en ylp h osp h in yl)fer r ocen e (7a ). To
a stirred solution of (S ,S )-5a (0.08 g, 0.20 mmol) in anhyd
THF (4 mL, 0.05 M) under argon at -78 °C was added t-BuLi
0.16 mL, 1.5 M, 0.24 mmol) dropwise. The reaction was stirred
for 5 min at -78 °C before Ph P(O)Cl (0.12 g, 0.50 mmol) was
P
S
(
3
3
(
The reaction was heated to reflux for 24 h. After being cooled
to rt, the reaction was diluted with diethyl ether and quenched
2
added dropwise to the dark red solution of the lithiated
ferrocene. After being stirred for 30 min at -78 °C, the reaction
was heated quickly to rt, quenched with 1 N HCl, and
extracted with diethyl ether. The organic layer was washed
with saturated NaHCO and brine, dried over MgSO , filtered,
3 4
and concentrated under reduced pressure. The crude residue
was purified by flash chromatography (pentane/ethyl acetate
with a small amount of saturated NaHCO
was separated off, dried over MgSO , filtered, and concentrated
3
. The organic layer
4
under reduced pressure. Purification of the crude residue by
flash chromatography (pentane/diethyl ether 98/2) afforded
0
.07 g (0.146 mmol, 73%) of 1a as a yellow amorphous powder.
(
S)-2-(o-An isole)-1-(diph en ylph osph an yl)fer r ocen e (1c).
Compound 7c was reduced according to the procedure de-
scribed for 1a . Purification by flash chromatography (pentane/
diethyl ether 98/2) afforded 1c in 83% yield.
(S)-2-(1-Na p h t h yl)-1-(d ip h en ylp h osp h a n yl)fer r ocen e
3
(1e). CeCl (1.5 equiv) was suspended in anhyd THF (0.01 M)
under argon. Compound 7e (1 equiv) dissolved in a small
amount of anhyd THF was added followed by slow addition of
4
a THF solution of LiAlH (1 M, 4 equiv). The reaction mixture
was heated 40 °C for 3 h. After being cooled to rt, the reaction
was diluted with toluene and quenched by slow addition of
water. The organic layer was washed with water and filtered
through a short column of silica gel. Concentration afforded
the crude product, which was purified by flash chromatography
7
0/30) to afford 0.08 g (0.17 mmol, 82%) of the product as a
1
yellow amorphous powder: H NMR (300 MHz) δ 7.77 (m, 3H),
7
5
Hz, 1C), 134.0 (J C-P ) 105.4 Hz, 1C), 132.4, 131.9 (J C-P ) 1.8
Hz, 2C), 131.8 (J C-P ) 2.4 Hz, 2C), 131.5, 131.3, 130.1 (2C),
1
(
1
(
.48 (m, 4H), 7.20 (m, 8H), 4.78 (m, 1H), 4.43 (m, 1H), 4.36 (s,
H), 3.88 (m, 1H); 13C NMR (75 MHz) δ 136.3 (J C-P ) 128.3
28.2 (J C-P ) 10.8 Hz, 2C), 128.1 (J C-P ) 11.7 Hz, 2C), 127.8
2C), 126.8, 92.5 (J C-P ) 9.7 Hz, 1C), 76.5 (J C-P ) 15.8 Hz,
C), 73.4 (J C-P ) 111.7 Hz, 1C), 72.8 (J C-P ) 9.3 Hz, 1C), 71.3
3
1
5C), 70.4 (J C-P ) 11.8 Hz, 1C); P (202 MHz) δ 29.8; MS (EI)
+
+
m/ z 462 [M ]; HRMS (ES, m/z) calcd for C28
62.0836, found 462.0836.
S)-2-(o-An isole)-1-(diph en ylph osph in yl)fer r ocen e (7c).
To a stirred solution of (S ,S )-5c (0.09 g, 0.21 mmol) in anhyd
THF (4 mL, 0.05 M) under argon at -78 °C was added t-BuLi
0.17 mL, 1.5 M, 0.25 mmol) dropwise. The reaction was stirred
for 5 min at -78 °C before Ph PCl (0.12 g, 0.53 mmol) was
H23FeOP (M )
4
(
(pentane/diethyl ether 98/2) to yield 85% of 1e.
P
S
Gen er a l P r oced u r e for Hyd r osilyla tion of Styr en e.
(
The MOPF ligand and [ClPd(π-C
3 5 2
H )] were dissolved in
styrene (1 equiv) under argon, and the solution was stirred at
2
added dropwise to the dark red solution of the lithiated
ferrocene. After being stirred for 30 min at -78 °C, the reaction
was heated quickly to rt, and 4 mL of 2 M NaOH was added.
The two-phase system was vigorously stirred for 15 min until
TLC showed full oxidation to the phosphine oxide. After
extraction with diethyl ether, the organic layer was washed
3
rt for 30 min before SiHCl (1.2 equiv) was added. The reaction
was monitored by 1H NMR. The product was purified by
Kugelrohr distillation (95 °C/0.5 mm).
Gen er a l P r oced u r e for Oxid a tion . The silane (1 equiv)
dissolved in anhyd benzene was added slowly to a suspension
of KF (6 equiv) and KHCO
3
(6 equiv) in MeOH (8 mL/mmol
J . Org. Chem, Vol. 67, No. 23, 2002 7993

Pedersen and J ohannsen
silane) and THF (8 mL/mmol silane). After the mixture was
stirred at rt for 15 min, H (35%, 30 equiv) was added slowly.
The reaction mixture was stirred vigorously for 48 h and then
quenched with water. The mixture was extracted five times
with diethyl ether, and the combined organic phases were
performing the ab initio calculation, The Danish Tech-
nical Research Council, Leo Pharmaceutical Products,
Familien Hede Nielsens Fond, Direktør Ib Henriksens
Fond, and the Lundbeck Foundation for generous
financial support.
2
O
2
washed twice with saturated Na
2 2 3
S O and once with saturated
NaHCO , dried over Na SO , filtered, and concentrated under
3
2
4
Su p p or tin g In for m a tion Ava ila ble: Copies of H and 13C
spectra of new compounds and X-ray crystal structure data
for (S)-1a . This material is available free of charge via the
Internet at http://pubs.acs.org.
1
reduced pressure. The crude residue was purified by flash
chromatography (pentane/ethyl acetate 85/15) to afford the
alcohol 9 in 75-98% yield.
Ack n ow led gm en t. We thank Dr. Inger Søtofte for
performing the X-ray analysis, Dr. Per-Ola Norrby for
J O0257804
7
994 J . Org. Chem., Vol. 67, No. 23, 2002