Novel chiral biferrocene ligands for palladium-catalyzed allylic substitution reactions

Article abstract of DOI:10.1021/jo016249w

Eleven novel aminophosphine ligands have been synthesized, all of which contain a chiral 2,2″-bridged biferroceno unit as part of a biferrocenoazepine substructure. The efficiency of these compounds as chiral auxiliaries in palladium-mediated allylic subs

Full text of DOI:10.1021/jo016249w

2206
J . Org. Chem. 2002, 67, 2206-2214
Novel Ch ir a l Bifer r ocen e Liga n d s for P a lla d iu m -Ca ta lyzed Allylic
Su bstitu tion Rea ction s
Li Xiao,† Walter Weissensteiner,*,† Kurt Mereiter,‡ and Michael Widhalm*,†
Institute of Organic Chemistry, University of Vienna, Wa¨hringer Strasse 38, A-1090 Wien, Austria, and
Department of Chemistry, Vienna University of Technology, Getreidemarkt 9, A-1060 Wien, Austria
m.widhalm@univie.ac.at
Received October 30, 2001
Eleven novel aminophosphine ligands have been synthesized, all of which contain a chiral 2,2′′-
bridged biferroceno unit as part of a biferrocenoazepine substructure. The efficiency of these
compounds as chiral auxiliaries in palladium-mediated allylic substitution reactions has been
investigated. Depending on the degree of (steric) fit between proper ligands and cyclic or noncyclic
substrates, reactions with 46-87% ee were achieved. The molecular structure of a palladium
dichloride complex of one of the ligands was determined by X-ray diffraction and compared to its
binaphthyl analogue. In the solid state, the azepine substructure of these two complexes adopts
totally different conformations with either local C2 (binaphthyl) or local C1 (biferrocene derivative)
symmetry. These structural changes are well-reproduced by empirical force field calculations and
are also reflected in significantly different behavior in asymmetric catalysis.
In tr od u ction
tion has been attributed to the predominance of one of
the allyl complexes in conjunction with a sterically or
electronically substantiated regioselectivity of the nu-
cleophilic attack. Irrespective of the origin of the asym-
metric induction, the stereoelectronic substrate-ligand
interactions are known to play an important role but still
need to be optimized empirically. A wide variety of
bidentate ligands have been investigated along with
various types of nucleophiles, and some ligands show
excellent enantioselectivities for differently substituted
cyclic and noncyclic substrates.5 Moreover, several ap-
plications in natural product syntheses have been re-
ported.6
Enantiopure bidentate ligands of transition metal
complexes with phosphorus and nitrogen donor sites have
been the subject of considerable interest since their
potential in asymmetric catalysis was discovered.1 To
date, bidentate PN ligands have been applied mainly in
carbon-carbon and carbon-heteroatom bond-forming
reactions. In general, for enantioselective catalysts, mod-
ular ligand structures have proved to be advantageous,
since ligand fine-tuning by varying structural fragments
or substituents is expected to allow their optimal adjust-
ment to the requirements of a specific reaction mecha-
nism, reagent, or substrate. Among carbon-carbon bond-
forming processes, palladium-mediated allylic substitutions
have found broad application as test reactions, mainly
because many mechanistic details are relatively well-
understood.2,3 The catalytic cycle starts with an oxidative
addition of a racemic allyl substrate to a palladium(0)
precursor. In this step, several isomeric palladium(II)
allyl complexes may be generated either directly or
through a η3-η1-η3 interconversion mechanism. A sub-
sequent nucleophilic attack at one of the terminal allyl
carbon atoms leads to η2-coordinated palladium olefin
complexes with a stereogenic carbon center adjacent to
the coordinated double bond. Mechanistic investigations
with different ligands and substrates have been reported
and early and late transition states have been suggested
for this step.4 The enantioselectivity of the overall reac-
Am in op h osp h in es 1-3 belong to a group of ligands
that contain the 2,2′-bridged binaphthyl entity as part
of a dihydrodinaphthazepine substructure7 in which the
(3) Selected papers dealing particularly with mechanistic aspects:
(a) Auburn, P. R.; Mackenzie, P. B.; Bosnich, B. J . Am. Chem. Soc.
1985, 107, 2033. (b) Mackenzie, P. B.; Whelan, J .; Bosnich, B. J . Am.
Chem. Soc. 1985, 107, 2046. (c) Baeckvall, J . E.; Granberg, K. L.;
Heumann, A. Isr. J . Chem. 1991, 31, 17. (d) Togni, A.; Burckhardt,
U.; Gramlich, V.; Pregosin, P. S.; Salzmann, R. J . Am. Chem. Soc. 1996,
118, 1031. (e) Blo¨chl, P. E.; Togni, A. Organometallics 1996, 15, 4125.
(f) Burckhardt, U.; Gramlich, V.; Hofmann, P.; Nesper, R.; Pregosin,
P. S.; Salzmann, R.; Togni, A. Organometallics 1996, 15, 3496. (g)
Togni, A.; Burckhardt, U.; Gramlich, V.; Pregosin, P. S.; Salzmann, R.
J . Am. Chem. Soc. 1996, 118, 1031. (h) Oslob, J . D.; Åkermark, B.;
Helquist, P.; Norrby, P.-O. Organometallics 1997, 16, 3015. (i) Lloyd-
J ones, G. C.; Stephen, S. C. Chem. Eur. J . 1998, 4, 2539. (j) Ramdeehul,
S.; Dierkes, P.; Aguado, R.; Kamer, P. C. J .; Van Leeuwen, P. W. N.
M.; Osborn, J . Angew. Chem., Int. Ed. 1998, 37, 3118. (k) Trost, B.
M.; Bunt, R. C. J . Am. Chem. Soc. 1998, 120, 70. (l) Hagelin, H.;
Svensson, M.; Åkermark, B.; Norrby, P.-O. Organometallics 1999, 18,
4574. (m) Svensson, M.; Bremberg, U.; Hallman, K.; Cso¨regh, I.;
Moberg, C. Organometallics 1999, 18, 4900. (n) Branchadell, V.;
Moreno-Man˜as, M.; Pajuelo, F.; Pleixats, R. Organometallics 1999, 18,
4934. (o) Canal, J . M.; Gome´z, M.; J ime´nez, F.; Rocamora, M.; Muller,
G.; Dun˜ach, E.; Franco, D.; J ime´nez, A.; Cano, F. H. Organometallics
2000, 19, 966. (p) Delbecq, F.; Lapouge, C. Organometallics 2000, 19,
2716. (q) Liu, S.; Mu¨ller, J . F. K.; Neuburger, M.; Schaffner, S.;
Zehnder, M. Helv. Chim. Acta 2000, 83, 1256.
* Corresponding author. Fax: ++43-1-42779521.
† University of Vienna.
‡ Vienna University of Technology.
(1) For a compilation of ligands, see: (a) Ojima I., Ed.; Catalytic
Asymmetric Synthesis, Wiley-VCH: New York, 2000; pp 802-856. (b)
Brunner H.; Zettlmeier W. Handbook of Enantioselective Catalysis with
Transition Metal Compounds; VCH: Weinheim, 1993; Vol. 1, p 2.
(2) (a) Trost, B. M.; Lee, C. Asymmetric Allylic Alkylation Reactions,
in Catalytic Asymmetric Synthesis; Ojima I., Ed.; Wiley-VCH: New
York, 2000; pp 593-650. (b) Trost, B. M.; van Vranken, D. L. Chem.
Rev. 1996, 96, 395. (c) Helmchen, G.; Kudis, S.; Sennhenn, P.;
Steinhagen, H. Pure Appl. Chem. 1997, 69, 513.
(4) Cf. (a) Sprinz, J .; Kiefer, M.; Helmchen, G.; Reggelin, M.;
Huttner, G.; Walter, O.; Zsolnai, L. Tetrahedron Lett. 1994, 35, 1523.
(b) Brown, J . M.; Hulmes, D. I.; Guiry, P. J . Tetrahedron 1994, 50,
4493.
10.1021/jo016249w CCC: $22.00 © 2002 American Chemical Society
Published on Web 03/08/2002

Novel Chiral Biferrocene Ligands
J . Org. Chem., Vol. 67, No. 7, 2002 2207
the benzyl positions of the binaphthyl core9 showed a
marked influence on the asymmetric induction. This
effect was attributed to different steric interactions in
the diastereomeric allyl intermediates, which results in
either significantly different complex concentrations or
changes of their individual reactivity and site selectivity.
Unfortunately, structural changes that enhanced the
ligand-substrate interaction were often accompanied by
a loss of reactivity.9
One obvious shortcoming of the dinaphthoazepine
substructure was the rather small biaryl angle of only
63°. With PN coordination in a square planar transition
metal complex, a dicoordinated substrate will receive only
a weak interaction with the dissymmetric unit. A ligand
developing its steric dissymmetry closer to the substrate
but without steric hindrance of the substrate coordination
would therefore be desireable, more so if small substrates
are of interest. Originally, we thought that replacing the
binaphthyl unit of ligands such as 1-3 by biferrocene
would allow the local C2 symmetry of the azepine
substructure to be maintained while, at the same time,
changing its spatial requirements. We therefore aimed
to synthesize a group of related biferrocene ligands with
a modular architecture that would enable a stepwise
adoption of the ligand structure with respect to electronic
and steric demands of the substrates. Such an approach
should allow a better fit into the chiral pocket.
F igu r e 1. Binaphthyl ligands 1-3.
tertiary nitrogen is situated in an environment of local
C2 symmetry (Figure 1). During investigations into the
catalytic efficiency of these systems in allylic substitution
reactions, we noticed that modification of the P-N
tether7b-d,8 and also the introduction of substituents in
(5) Selected recent papers: (a) Danjo, H.; Tanaka, D.; Hayashi, T.;
Uozumi, Y. Tetrahedron 1999, 55, 14341. (b) Cahill, J . P.; Cunneen,
D.; Guiry, P. J . Tetrahedron: Asymmetry 1999, 10, 4157. (c) Yonehara,
K.; Hashizume, T.; Mori, K.; Ohe, K.; Uemura, S. J . Org. Chem. 1999,
64, 9374. (d) Hallman, K.; Macedo, E.; Nordstrom, K.; Moberg, C.
Tetrahedron: Asymmetry 1999, 10, 4037. (e) Chelucci, G.; Deriu, S.;
Pinna, G. A.; Saba, A.; Valenti, R. Tetrahedron: Asymmetry 1999, 10,
3803. (f) Hilgraf, R.; Pfaltz, A. Synlett 1999, 1814. (g) Chelucci, G.;
Culeddu, N.; Saba, A.; Valenti, R. Tetrahedron: Asymmetry 1999, 10,
3537. (h) Trost, B. M.; Ariza, X. J . Am. Chem. Soc. 1999, 121, 10727.
(i) Gomez, M.; J ansat, S.; Muller, G.; Panyella, D.; Van Leeuwen, P.
W. N. M.; Kamer, P. C. J .; Goubitz, K.; Fraanje, J . Organometallics
1999, 18, 4970. (j) Evans, P. A.; Brandt, T. A. Org. Lett. 1999, 1, 1563.
(k) Ito, K.; Kashiwagi, R.; Iwasaki, K.; Katsuki, T. Synlett 1999, 1563.
(l) Schleich, S.; Helmchen, G. Eur. J . Org. Chem. 1999, 5. (m) Fuji, K.;
Kinoshita, N.; Tanaka, K. J . Chem. Soc., Chem. Commun. 1999, 1895.
(n) Yan, Y.-Y.; Widhalm, M. Monatsh. Chem. 1999, 130, 873. (o) Zhang,
W.; Shimanuki, T.; Kida, T.; Nakatsuji, Y.; Ikeda, I. J . Org. Chem.
1999, 64, 6247. (p) Trost, B. M.; Schroeder, G. M. J . Am. Chem. Soc.
1999, 121, 6759. (q) Saitoh, A.; Misawa, M.; Morimoto, T. Tetrahe-
dron: Asymmetry 1999, 10, 1025. (r) Trost, B. M.; Toste, F. D. J . Am.
Chem. Soc. 1999, 121, 4545. (s) Moberg, C.; Bremberg, U.; Hallman,
K.; Svensson, M.; Norrby, P.-O.; Hallberg, A.; Larhed, M.; Csoregh, I.
Pure Appl. Chem. 1999, 71, 1477. (t) Fairlamb, I. J . S.; Lloyd-J ones,
G. C. J . Chem. Soc., Chem. Commun. 2000, 2447. (u) Kodama, H.; Taiji,
T.; Ohta, T.; Furukawa, I. Tetrahedron: Asymmetry 2000, 11, 4009.
(v) Shintani, R.; Lo, M. M.-C.; Fu, G. C. Org. Lett. 2000, 2, 3695. (w)
Chelucci, G.; Saba, A.; Sanna, G.; Soccolini, F. Tetrahedron: Asymmetry
2000, 11, 3427. (x) Rabeyrin, C.; Nguefack, C.; Sinou, D. Tetrahedron
Lett. 2000, 41, 7461. (y) Trost, B. M.; Surivet, J .-P. Angew. Chem., Int.
Ed. 2000, 39, 3122. (z) Kaiser, N.-F. K.; Bremberg, U.; Larhed, M.;
Moberg, C.; Hallberg, A. J . Organomet. Chem. 2000, 603, 2. (aa) Mino,
T.; Tanaka, Y.; Sakamoto, M.; Fujita, T. Heterocycles 2000, 53, 1485.
(bb) Deerenberg, S.; Schrekker, H. S.; Van Strijdonck, G. P. F.; Kamer,
P. C. J .; Van Leeuwen, P. W. N. M.; Fraanje, J .; Goubitz, K. J . Org.
Chem. 2000, 65, 4810. (cc) Liu, S.; Muller, J . F. K.; Neuburger, M.;
Schaffner, S.; Zehnder, M. Helv. Chim. Acta 2000, 83, 1256. (dd)
Takada, H.; Oda, M.; Oyamada, A.; Ohe, K.; Uemura, S. Chirality 2000,
12, 299. (ee) McCarthy, M.; Guiry, P. J . Polyhedron 2000, 19, 541. (ff)
Okuyama, Y.; Nakano, H.; Hongo, H. Tetrahedron: Asymmetry 2000,
11, 1193. (gg) Robert, F.; Delbecq, F.; Nguefack, C.; Sinou, D. Eur. J .
Inorg. Chem. 2000, 351. (hh) Sauthier, M.; Fornies-Camer, J .; Toupet,
L.; Reau, R. Organometallics 2000, 19, 553. (ii) Mino, T.; Ogawa, T.;
Yamashita, M. Heterocycles 2001, 55, 453. (jj) Uozumi, Y.; Shibatomi,
K. J . Am. Chem. Soc. 2001, 123, 2919. (kk) Mino, T.; Shiotsuki, M.;
Yamamoto, N.; Suenaga, T.; Sakamoto, M.; Fujita, T.; Yamashita, M.
J . Org. Chem. 2001, 66, 1795.
Resu lts a n d Discu ssion
On the basis of the ideas described above, a group of
11 structurally related biferrocene ligands, 4-14, was
prepared, all of which have a topology that is entirely
different from previously investigated aminophosphines
(Figure 2). A systematic investigation of electronic (4/5/
6) or steric (4/10/11, 7/12, 8/13, 9/14) factors as well as
the influence of the chelate ring size (4/7/8 or 9, respec-
tively, 10/12) on reactivity and enantioselectivity was
undertaken. In several cases, the existence of diastere-
omers with matched and mismatched stereogenic units
was also considered (8/9, 10/11, 13/14).
Syn th esis of Liga n d s. The synthesis of the ligands
4-14 was achieved by the same general route: (i)
formation of an enantiopure biferrocene intermediate, (ii)
reaction of this entity with a suitable bromo- or iodo-
substituted amino derivative in order to build up the
ligand skeleton, and (iii) replacement of the bromide or
iodide by an appropriate phosphino group. All of the
target ligands, 4-14, were accessible from three enan-
tiopure biferrocene intermediates: ligands 4-9 (without
methyl substituents in the pseudo-benzyl position) from
2310 (Schemes 1-3), the methyl-substituted derivatives
10 and 12-14 from 2811 (Scheme 4), and 11 from 38
(Scheme 5).
Biferrocene 23, the key intermediate in the preparation
of ligands 4-9, was easily accessible in two steps from
enantiopure 15 (Scheme 1). In a homocoupling reaction,
2-iodo-1-dimethylaminomethylferrocene, 15, was trans-
formed either under Ullmann conditions (Cu, 110 °C,
37%) or oxidatively [BuLi, Fe(acac)3, 35%] into 2,2′′-bis-
(6) (a) Trost, B. M.; and Shi Z. J . Am. Chem. Soc., 1996, 118, 3039.
(b) Trost, B. M.; Chupak, L. S.; Lubbers T. J . Am. Chem. Soc., 1998,
120, 1732. (c) Trost, B. M.; Asakawa, N. Synthesis 1999 (special issue),
1491. (d) Trost, B. M.; Toste, F. D. J . Am. Chem. Soc. 2000, 122, 11262.
(e) Trost, B. M.; Tang, W.; Schulte, J . L. Org. Lett. 2000, 2, 4013. (f)
Tietze, L. F.; Schirok, H.; Wohrmann, M.; Schrader, K. Eur. J . Org.
Chem. 2000, 2433.
(7) (a) Kubota, H.; Koga, K. Tetrahedron Lett. 1994, 35, 6689. (b)
Wimmer, P.; Widhalm, M. Tetrahedron: Asymmetry 1995, 6, 657. (c)
Widhalm, M.; Mereiter, K.; Bourghida; M. Tetrahedron: Asymmetry
1998, 9, 2983. (d) Widhalm, M.; Nettekoven, U.; Mereiter, K. Tetra-
hedron: Asymmetry 1999, 10, 4369.
(9) Bourghida, M.; Widhalm, M. Tetrahedron: Asymmetry 1998, 9,
1073.
(10) For racemic compounds 18 and 21 see: Marr, G.; Moore, R. E.;
Rockett, B. W. Tetrahedron 1969, 25, 3477.
(11) Spescha, M.; Duffy, N. W.; Robinson, B. H.; Simpson, J .
Organometallics 1994, 13, 4895.
(8) Wimmer, P.; Widhalm, M. Monatsh. Chem. 1996, 127, 669.

2208 J . Org. Chem., Vol. 67, No. 7, 2002
Xiao et al.
F igu r e 2. Novel biferrocene ligands 4-14.
Sch em e 1a
Sch em e 2
enantiopure iodo derivatives 16-18, as we described
previously.12 The diastereomers (Sc,Sp)-16 and (Sc,Rp)-
16 were actually obtained as intermediates in a resolu-
tion procedure of racemic 15 with the use of phenyleth-
ylamine,12 and these compounds could be readily separated
by column chromatography. Enantiopure 17 and 18, on
the other hand, were obtained through highly diastereo-
selective o-lithiation/iodations of N-ferrocenylmethyl-O-
methylprolinol13 (17) or N-ferrocenylmethyl-O-methyl-
ephedrine (18), an auxiliary recently developed in our
group.14
In an attempt to shorten the synthesis, we tried to use
these precursors of 15 as alternative starting materials
for the synthesis of intermediate 24. Indeed, iodides
16-18 could be transformed into biferrocenes 20-22 [20,
Cu, 136 °C, 72%; n-BuLi, Fe(acac)3, 44%; 21, Cu, 130 °C,
64%; 22, Cu, 130 °C, 58%], and these were subsequently
reacted with methyl iodide and 2-bromobenzylamine
hydrochloride to give the azepine intermediate 24 (yields
from 20, 41%; 21, 40%; 22, 63%).
a
Reagents: (a) n-BuLi, Fe(acac)3 or Cu, 110 °C; (b) CH3I,
acetonitrile; (c) Cu, 130-136 °C; (d) 2-bromobenzylamine hydro-
chloride, triethylamine, acetonitrile, reflux.
(dimethylaminomethyl)-1,1′′-biferrocene, 19, which, on
treatment with methyl iodide, gave 23 in 92% yield.
Reaction of 23 with bromo-substituted amines (2-bro-
mobenzylamine, 2-bromoaniline, and 2-bromoferrocenyl-
methylamine) or, more conveniently, with their hydro-
chlorides afforded the azepine precursors 24-27 (Schemes
1 and 3) in 52% (27) to 90% (24) yield. In a final step,
ligands 4-9 were obtained in 55-71% yield by reacting
each bromo derivative 24-27 with n-BuLi, followed by
addition of an excess of the appropriate diarylchloro-
phosphine (Schemes 2 and 3).
Although compound 24 was obtained in all four syn-
thetic pathways from starting materials 15-18 in com-
parable overall yields (28, 30, 26, and 37%, respectively),
(12) Xiao, L.; Mereiter, K.; Weissensteiner, W.; Widhalm, M. Syn-
thesis 1999, 1354.
(13) Ganter, C.; Wagner, T. Chem. Ber. 1995, 128, 1157.
(14) (a) Kitzler, R.; Xiao, L.; Weissensteiner, W. Tetrahedron:
Asymmetry 2000, 11, 3459. (b) Kitzler, R.; Xiao, L.; Weissensteiner,
W. J . Org. Chem. 2001, 66, 8912.
The synthesis of 23 required the enantiomerically pure
starting material 15, and this was easily accessible from

Novel Chiral Biferrocene Ligands
J . Org. Chem., Vol. 67, No. 7, 2002 2209
Sch em e 3a
a
Reagents: (a) 2-bromoaniline hydrochloride or (R)- or (S)-2-bromoaminomethylferrocene, triethylamine, acetonitrile, reflux; (b) n-BuLi
(-40 °C), Ph2PCl (-78 °C).
Sch em e 4a
Sch em e 5a
a
Reagents: (a) n-BuLi, I2 (-78 °C); (b) KOtBu, DMSO, rt; (c)
n-BuLi, Fe(acac)3; (d) CH3I, acetonitrile; (e) 2-bromobenzylamine
hydrochloride, triethyamine, acetonitrile, reflux; (f) n-BuLi (-40
°C), Ph2PCl (-78 °C).
a
Reagents: (a) 2-bromobenzylamine hydrochloride or 2-bro-
moaniline hydrochloride or (R)-2-bromoaminomethylferrocene or
(S)-2-iodoaminomethylferrocene, triethylamine, acetonitrile, re-
flux; (b) n-BuLi (-40 °C), Ph2PCl (-78 °C).
The azepinium salt 2811 (for 10, 12-14) and its
diastereomer 38 (for 11) served as the central precursors
for the synthesis of the 3,5-dimethyl-substituted dihy-
droazepine ligands 10-14 (Schemes 4 and 5). Both
diastereomers were prepared from enantiopure (R)-33.
the route 18 f 22 f 24 proved to be most economical
and, moreover, appeared to be the least time-consuming.

2210 J . Org. Chem., Vol. 67, No. 7, 2002
Xiao et al.
F igu r e 3. Molecular structure of 4‚PdCl2‚3CHCl3 [(Rp,Rp)-
40‚3CHCl3]. The solvate molecules and hydrogen atoms have
been removed for clarity.
According to a published procedure,11 o-lithiation of 33
with n-BuLi followed by homocoupling with Fe(acac)3 and
subsequent treatment of the biferrocene intermediate
with methyl iodide gave 28. The diastereomer 38 was
made in a similar two-step sequence from 36, which in
the first step was reacted with n-BuLi followed by
treatment with Fe(acac)3 to give biferrocene 37 in 67%
yield. In the second step, treatment with methyl iodide
led to the azepinium salt 38 (99%). Enantiopure iodide
36 was obtained in three steps from 33 by adapting a
published procedure for analogous derivatives.15
It is interesting to note that the applicability of the
two homocoupling methods, either with Fe(acac)3 or
under Ullmann conditions, appeared to be complemen-
tary. While the oxidative coupling worked in a satisfac-
tory manner with R-methylated ferrocenylamines 33 or
36, the unsubstituted diastereomers 16 and 17 afforded
almost no coupling product at all. Only 19 was formed
from 15 (albeit in low yield, 35%), but this was ac-
companied by a considerable amount of 2-aminomethyl-
2′′-formyl-1,1′′-biferrocene and starting material. In con-
trast, the Ullmann coupling of 16 and 17 afforded the
biferrocenes 20 and 21 in 72 and 64% yield, respectively.
Starting from the azepinium salts 28 or 38, the benzyl-
methylated aminophosphine ligands 10-14 were ob-
tained in enantiomerically pure form in 23-64% overall
yield by applying the same two-step sequence as de-
scribed above for the synthesis of 4-9 from 23.
X-r a y Str u ctu r e An a lysis. Complex (Rp,Rp)-40, ob-
tained by treating a solution of (Rp,Rp)-4 in benzene with
Pd(CH3CN)2Cl2, crystallizes in the form of a chloroform
solvate, 4‚PdCl2‚3CHCl3, in the triclinic space group P1
(no. 1) with one formula unit per cell. An ORTEP plot of
the molecular structure is shown in Figure 3. Crystal-
lographic data are given in the Experimental Section, and
further details such as bond lengths and angles are given
in the Supporting Information. The general structural
features of 40 are as expected for this type of complex.
However, in contrast to the molecular structure of the
palladium allyl complex of binaphthyl ligand 216 (Figure
1, n ) 1, m ) 0, R ) H), in which the azepine subunit is
F igu r e 4. Biaryl- and biferrocenoazepines used as models in
empirical force field calculations (A, C). Different spatial
requirements of binaphthyl (B) or bisferroceno (C) azepine
subunits.
in a rather strainless twist conformation of local C2
symmetry with a biaryl angle of 63°, the connected Cp
rings of 40 are found in a nearly coplanar arrangement
(dihedral angle between planes C21-C25 and C31-C35
) 14.4°) that forces the seven-membered ring system into
an envelope-like conformation of local C1 symmetry.
Em p ir ica l F or ce F ield Ca lcu la tion s. To rationalize
these strong structural differences, we carried out em-
pirical force field calculations.17 Minimum energy struc-
tures of a number of differently substituted biferro-
cenoazepines and their related biaryl derivatives, including
the palladium dichloride complex 40 and its biphenyl
analogue, were searched for [Figure 4, R ) CH3, CH2Ph,
CH2(2-Ph2P)Ph]. To make a better comparison, calcula-
tions on biphenyl rather than binaphthyl derivatives
were carried out. The reason for this is that 6,6′-
substituents are expected to prevent a biaryl system from
adopting a coplanar arrangement of the aryl groups, an
arrangement that is necessary in an envelope-like azepine
structure. Details on the calculations are given in the
Supporting Information.
Qualitatively, the following results were obtained: (i)
in the case of all biphenyl derivatives, only minimum
energy structures with an azepine subunit of local C2 (or
very close to local C2) symmetry and with a rather
strainless twist conformation were located, while (ii) all
biferrocene ligands can adopt conformers with both the
twist and the envelope-like azepine subunit. Further-
more, in the envelope-like structure, the substituents at
the nitrogen atom can be located either in an axial or an
equatorial position.
The order of minimum energy structures for all bifer-
rocene ligands was as follows: independent of the sub-
stituents at nitrogen, the envelope-like conformer with
an axial substituent was always of lowest energy, fol-
(16) Widhalm, M.; Nettekoven, U.; Kalchhauser, H.; Mereiter, K.;
Calhorda, M.-J .; Fe´lix, V. Organometallics 2002, 21, 315.
(17) PCMODEL V 7.0; Serena Software, Box 3076, Bloomington, IN
47402-3076.
(15) Hayashi, T.; Mise, T.; Fukushima, M.; Kagotani, M.; Na-
gashima, N.; Hamada, Y.; Natsumoto, A.; Kawakami, S.; Konishi, M.;
Yamamoto, K.; Kumada, M. Bull. Chem. Soc. J pn. 1980, 53, 1138.

Novel Chiral Biferrocene Ligands
J . Org. Chem., Vol. 67, No. 7, 2002 2211
Sch em e 6
tion on the configuration of the biferrocene unit was
observed for noncyclic substrates. Products with the (R)-
configuration were formed in reaction 1 with a moderate
to good level of enantioselectivity. An exception to this
trend concerns ligand 14, which gave the product with
the (S)-configuration and with only 5% ee. Reaction 2
generally proceeded with a low level of enantioselectivity
in the range 3-46%. In most cases the product configu-
ration changed to (R), probably due to the lack of π
stacking interactions between substrate and ligand phen-
yl groups. The best result (46% ee) was obtained with
ligand 9, which contains a further chiral ferrocene moiety
of matching configuration. When the mismatched ligand
8 was used, the enantioselectivity dropped to 16%. A
similar effect of matching/mismatching carbon and fer-
rocene configuration can be seen from the results for
ligands 4, 10, and 11. For example, ligand 4swithout a
methyl-substituted azepine ring and with a low degree
of steric interaction in close proximity to the nitrogen
coordination sitesgave 15% ee, while the mismatched
ligand 10 led to a drop in the ee to 4%. The matched
ligand 11, however, led to a product with 26% ee but with
the opposite configuration (R).
In the case of cyclic substrates (reactions 3-5), the
metallocene configuration of the PN-bridging ferrocene
unit proved to have a predominant effect on the enanti-
oselectivity and product configuration. On using ligand
8 instead of ligand 9, there is a change from 53%,
57%, and 74% ee and (S)-configuration to 59%, 65%, and
77% ee and (R)-configuration. A similarly pronounced
relationship was found between ligands 13 and 14. The
substrate cycloheptenyl acetate (reaction 5) proved to
be most sensitive to structural changes of the ligands.
Comparison of 4, 7, and 8 or 4, 7, and 9 shows that
the effect of the size of the chelate ring is less pro-
nounced and is dominated by configurational differences
in 8 and 9.
lowed by the structure with an equatorial substituent
[2.7, 6.6, and 4.3 kcal/mol for R ) CH3, CH2Ph, and CH2-
(2-PPh2)C6H4, respectively]. The conformers with an
azepine subunit of local C2 symmetry were always found
to be of highest energy. The same result was obtained
for the palladium dichloride complex of 4 (40) and its
biphenyl analog: for 40 a molecular structure similar to
that found in the solid state was calculated to be of lowest
energy (∼5.8 kcal/mol more stable than the C2-type
structure), while the azepine unit of the biphenyl ana-
logue was calculated to adopt a twist conformation
comparable to that of 2 in the solid state.16
On the basis of the results of the calculations, we
conclude that in the case of the biferrocene derivatives
an envelope-like azepine conformation is more stable
than a twisted arrangement. There are two reasons for
this conclusion: (i) as compared to the six-membered aryl
rings, bond angles within the Cp ring are smaller, while
the bond angles within the azepine moiety necessarily
increase. This enables a nearly coplanar arrangement of
the ferrocene units, a situation that is even more favored
by the fact that the steric interactions of ortho-substit-
uents are less pronounced in the five-membered Cp rings
as compared to six-membered aryl rings. (ii) Steric
interactions between substituents at the nitrogen atom
and one of the ferrocenyl units are reduced in an
envelope-like conformation.
In contrast to steric effects, the influence of electronic
effects on the enantioselectivity was found to be small
and without a clear trend. In only one case, when
cyclopentenyl acetate was used as the substrate (reaction
3), was the asymmetric induction improved. In this case
a value of 27% ee was improved to 58% ee by changing
the electron-withdrawing phosphino group of 5 to an
electron-donating phosphine in 6.
Allylic Su bstitu tion . Asymmetric allylic substitution
reactions with typical cyclic and noncyclic substrates
were performed (Scheme 6). Experimental details and
tables of all results are given in the Supporting Informa-
tion. Table 1 lists the best results obtained in allylic
substitution reactions 1-6 with ligands 4-14. Reaction
1 was used to optimize the reaction conditions with
dimethylmalonate and N,O-bis(trimethylsilyl)acetamide
(BSA)/KOAc as the base. In a comparative study, all
reactions were conducted on a 1 mmol scale with 1 mol
% of catalyst prepared in situ from the appropriate
ligands and [Pd(allyl)Cl]2 (1:0.5) in CH2Cl2 at room
temperature. A reaction time of 1-2 h for complete
conversion was sufficient, apart from ligands 9, 12, and
14, where longer periods were required. Longer reaction
times were also required for reaction 2 with an aliphatic
substrate. The reaction was generally faster with cyclic
substrates, and isolated yields were excellent (92-99%).
In contrast, the amination of 1,3-diphenyl-2-propenyl
acetate with benzylamine proceeded slowly, and isolated
yields after a reaction time of 48 h varied considerably.
A clear dependence of the sense of asymmetric induc-
Con clu sion s
The use of a modular and stereoselective approach has
enabled a new class of aminophosphine ligands, all of
which contain a biferroceno azepine subunit, to be
synthesized. Eleven ligands were prepared for use as
auxiliaries in asymmetric allylic substitution reactions.
On the basis of the results of empirical force field
calculations and on an X-ray diffraction study, we
conclude that the biferrocenoazepine subunit, both in the
free ligands and as part of a palladium dichloride
complex, prefers an envelope-like conformation of the
seven-membered ring rather than a twisted arrangement
of local C2 symmetry. The latter situation has been found
experimentally for binaphthyl analogues.
In allylic substitution reactions, two noncyclic and
three cyclic substrates (reactions 1-5) were alkylated
enantioselectively with dimethyl malonate to give prod-
ucts with 46-87% ee. In addition, the amination of 1,3-
diphenylprop-2-ene-1-yl acetate with benzylamine pro-

2212 J . Org. Chem., Vol. 67, No. 7, 2002
Xiao et al.
Ta ble 1. Resu lts of Allylic Alk yla tion a n d Am in a tion Rea ction s 1-6 (ee) w ith Liga n d s 4-14
enantiomeric excess (configuration)a
ligand
(Rp,Rp)-4
(Rp,Rp)-5
(Rp,Rp)-6
(Rp,Rp)-7
(Rp,Rp,Rp)-8
(Rp,Rp,Sp)-9
(Sc,Sc,Rp,Rp)-10
(Rc,Rc,Rp,Rp)-11
(Sc,Sc,Rp,Rp)-12
(Sc,Sc,Rp,Rp,Rp)13
(Sc,Sc,Rp,Rp,Sp)-14
rxn 1
rxn 2
rxn 3
rxn 4
rxn 5
rxn 6
87 (R)
86 (R)
79 (R)
70 (R)
42 (R)
69 (R)
64 (R)
79 (R)
60 (R)
59 (R)
5 (S)
15 (S)
16 (S)
11 (S)
39 (S)
16 (S)
46 (S)
4 (S)
26 (R)
6 (R)
3 (S)
33 (S)
27 (S)
58 (S)
41 (S)
53 (S)
59 (R)
11 (S)
12 (S)
55 (S)
42 (S)
11 (R)
50 (S)
49 (S)
49 (S)
61 (S)
57 (S)
65 (R)
22 (S)
1 (S)
77 (S)
77 (S)
69 (S)
81 (S)
74 (S)
77 (R)
23 (S)
1 (S)
86 (S)
75 (S)
82 (S)
24 (S)
2 (R)
74 (S)
80 (S)
65 (S)
77 (S)
41 (S)
20 (S)
43 (S)
46 (S)
36 (R)
70 (S)
65 (S)
1 (R)
24 (S)
a
Boldfaced figures indicate maxium e.e. for this reaction.
room temperature and was stirred overnight. The reaction was
quenched with saturated NaHCO3 solution, and the organic
phase was separated, washed with brine, and dried with
Na2SO4. The solvent was evaporated and the residue was
purified by chromatography on silica gel. Elution with PE/
Et2O/Et3N (40:5:1) afforded 4 (611 mg, 62%) as a yellow solid:
ceeded with 86% ee (reaction 6). The enantioselectivity
with noncyclic substrates was found to be mainly con-
trolled by the configuration of the biferrocene moiety but
modulated by methyl substituents on the azepine ring
as well as by the configuration of the P-N linking
ferroceno group. In contrast, for cyclic substrates the
product configuration depends primarily on the config-
uration of the ferroceno group located in the central
region between the P and N coordination sites. From both
the structural and the catalytic point of view, the
investigated biferrocenoazepines constitute a new class
of ligands that are not directly comparable to their biaryl
(binaphthyl) analogues.
1
mp 73-75 °C; H NMR δ 3.35 (d, 2H, J ) 15.6 Hz, CH2),
3.57 (d, 2H, J ) 15.6 Hz, CH2), 3.88 (m, 2H, Cp-H), 3.90 (s,
10H, Cp-H), 3.98 (d, 1H, J ) 13.6 Hz, CH2), 4.07 (t, 2H, J )
2.5 Hz, Cp-H), 4.17 (dd, 1H, J ) 13.6, 2.5 Hz, CH2), 4.34 (m,
2H, Cp-H), 6.96-7.00 (m, 1H, Ph), 7.17-7.37 (m, 13 H, Ph);
13C NMR δ 54.38 (CH2), 57.38 (d, J ) 16.7 Hz, CH2), 65.96
(Cp-CH), 66.24 (Cp-CH), 67.35 (Cp-CH), 69.82 (Cp), 82.46
(Cp-C), 85.23 (Cp-C), 127.12 (Ph-CH), 128.13, 128.17, 128.30
(Ph-CH), 128.30 (d, J ) 11.4 Hz, Ph-CH), 129.27 (d, J ) 5.3
Hz, Ph-CH), 133.47 (d, J ) 19.0 Hz, Ph-CH), 133.66 (d, J )
19.0 Hz, Ph-CH), 134.41 (Ph-CH), 137.96, 137.12, 137.21,
137.29 (Ph-C), 138.65 (d, J ) 11.4 Hz, Ph-C), 144.57 (d, J )
22.8 Hz, Ph-C); 31P NMR δ -14.01; MS (210 °C) m/z (rel %) )
Exp er im en ta l Section
Gen er a l Meth od s. Melting points were determined on a
Kofler melting point apparatus and are uncorrected. NMR
spectra were recorded at 400.132 MHz (1H) and 100.624 MHz
(13C) in CDCl3 if not stated otherwise. Chemical shifts δ are
20
685 (66, M+), 500 (7), 410 (100); [R]λ ) +553 (589 nm), +603
(578 nm), +838 (546 nm), (c ) 0.50); CD λmax (∆ꢀ) ) 280 nm
(21.5), 298 nm (-3.80), 306 nm (-3.65), 343 (5.01), 455 nm
(2.84) (c ) 9.83 × 10-4 mol/L). Anal. Calcd for C41H36Fe2NP:
C, 71.85; H, 5.29; N, 2.04. Found: C, 71.57; H, 5.53; N, 2.01.
1
reported in ppm relative to CHCl3 (7.26 and 77.00 ppm for H
and 13C), CH2Cl2 (5.31 and 53.7 ppm for 1H and 13C), DMSO
1
(2.50 and 39.5 ppm for H and 13C), and 85% H3PO4 (31P). Italic
(Rp,Rp)- an d (Sp,Sp)-2,2′′-Bis(N,N-dim eth ylam in om eth yl)-
1,1′′-bifer r ocen e (19). Meth od A. To a solution of (Rp)- or
(Sp)-1-iodo-2-dimethylaminomethylferrocene (15) (1.68 g, 4.55
mmol) in dry ether (8 mL) was added n-butyllithium in hexane
(3.4 mL, 1.6 M, 5.46 mmol). The mixture was stirred at room
temperature for 2 h and then cooled to 0 °C. A solution of iron-
(III) acetylacetonate (2.47 g, 7.00 mmol) in degassed benzene
(10 mL) was added rapidly to the reaction mixture. The rust-
colored thick sludge was stirred at 0 °C for 10 min and then
for an additional 22 h at room temperature. The reaction was
hydrolyzed with 10% NaOH and filtered over a plug of Celite,
and the cake was washed with CH2Cl2 and water. The aqueous
phase was extracted with CH2Cl2 (3 × 50 mL), and the
combined organic extracts were dried with MgSO4. The solvent
was removed and the residue was chromatographed on silica
gel. Elution with PE/Et2O/Et3N (1:3:1) gave biferrocene 19 in
the last band as a yellow oil (380 mg, 35%).
Meth od B. To a solution of (Rp)- or (Sp)-1-iodo-2-dimeth-
ylaminomethylferrocene (15) (3.30 g, 8.9 mmol) in CH2Cl2 (5
mL) was added activated copper bronze (15 g) and the sol-
vent was evaporated. The resulting solid was heated under
an argon atmosphere to 110 °C for 12 h. Then the mixture
was cooled to room temperature and was washed repeatedly
with CH2Cl2. The solvent was removed and the residue was
purified by column chromatography on silica gel. Elution with
PE/Et2O/Et3N (10:10:1) gave 960 mg of starting material 15
in the first band, followed by biferrocene 19 as a red oil (562
mg, 37%).
13C data indicate that the signals could not be assigned
unambiguously. Mass spectra were EI at 70 eV, if not stated
otherwise. Optical rotations were measured in CHCl3 at 20
°C. CD spectra were recorded on a dichrograph in CH2Cl2 at
20 °C. Elemental analyses were performed at the Mikroana-
lytisches Laboratorium der Universita¨t Wien (Mag. J . Thei-
ner). All reactions requiring inert conditions were conducted
under Ar using standard Schlenk techniques. Ether was
distilled from LiAlH4, acetonitrile and benzene were distilled
from calcium hydride, and THF was dried over potassium prior
to use. Chromatographic separations were performed under
gravity either on silica gel (40-63 µm) or alumina (activity
II-III, 0.063-0.200 mm).
Iron(III) acetylacetonate was dried overnight at 100 °C
under high vacuum. Copper was activated according to ref 18.
(Rp)- And (Sp)-1-iodo-2-dimethylaminomethylferrocene (15),12
(Sc,Rp)- or (Sc,Sp)-1-iodo-2-(1-phenylethylaminomethyl)ferro-
cene (16),12 (Rc,Rp)-1-iodo-2-[(2-methoxymethylpyrrolidin-1-yl)]-
methylferrocene (17),12 (1Rc,2Sc)-1-iodo-2-[N-(1-methoxy-1-phen-
ylprop-2-yl)-N-methylaminomethyl]ferrocene (18),14 (Rc,Rc,Sp,Sp)-
N,N-dimethyl-3,5-dimethyl-3,5-dihydro-4H-diferroceno[c,e]aze-
pinium iodide (28),11 and (Rc,Sp)-2-(1-N,N-dimethylamino)-
ethyl-1-trimethylsilylferrocene (34)15 were synthesized as
described in the literature.
(Rp ,Rp )-N-(2-Dip h en ylp h osp h in ob en zyl)-3,5-d ih yd r o-
4H -d ifer r ocen o[c,e]a zep in e (4). To a solution of azepine
(Rp,Rp)-24 (660 mg, 1.14 mmol) in THF (20 mL) was added
n-BuLi (0.88 mL, 1.6 M, 1.41 mmol) at -40 °C. The reaction
mixture was stirred for 2 h, and Ph2PCl (385 mg, 1.71 mmol)
was added at -78 °C. The mixture was allowed to warm to
(Rp,Rp)-19: 1H NMR δ 1.95 (s, 12H, 4 CH3), 3.15 (d, 2H, J
) 13.0 Hz, CH2), 3.23 (d, 2H, J ) 13.0 Hz, CH2), 4.23 (m, 2H,
Cp-H), 4.25 (s, 10H, Cp-H), 4.29 (m, 2H, Cp-H), 4.40 (m, 2H,
Cp-H); 13C NMR δ 45.10 (CH3), 57.14 (CH2), 66.41 (Cp-CH),
69.61 (Cp), 70.03 (Cp-CH), 71.02 (Cp-CH), 85.03 (Cp-C), 85.10
(18) Fuson, R. C.; Cleveland, E. A. In Organic Synthesis; Wiley: New
York, 1995; Collect. Vol. 3, p 339.

Novel Chiral Biferrocene Ligands
J . Org. Chem., Vol. 67, No. 7, 2002 2213
(Cp-C); MS (90 °C) m/z (rel %) ) 484 (44, M+), 439 (32), 395
(24), 242 (4); [R]λ20 ) +1855 (589 nm), +2063 (578 nm), +2463
(546 nm), (c ) 1.00); CD λmax (∆ꢀ) ) 285 nm (10.1), 342 nm
(-1.92), 470 nm (5.07) (c ) 1.05 × 10-3 mol/L). Anal. Calcd
for C26H32Fe2N2: C, 64.49; H, 6.66; N, 5.78. Found: C, 64.43;
H, 6.81; N, 5.61.
10-4 mol/L). Anal. Calcd for C29H26Fe2NBr: C, 60.04; H, 4.52;
N, 2.41. Found: C, 59.45; H, 4.70; N, 2.36.
(Sp,Sp)-24: yellow solid; mp 127-129 °C; [R]λ20 ) -660 (589
nm), -699 (578 nm), -949 (546 nm), (c ) 0.52).
(Rp ,Rp )-[N-(2-Dip h en ylp h osp h in oben zyl)-3,5-d ih yd r o-
4H-d ifer r ocen o[c,e]a zep in e-N,P ] Dich lor op a lla d iu m (II)
(40). A solution of azepine 4 (68.5 mg, 0.1 mmol) in dry
benzene (2 mL) was degassed and transferred into a solution
of (CH3CN)2PdCl2 (24.6 mg, 0.095 mmol) in benzene (2 mL).
The mixture was stirred overnight at room temperature under
argon. The resulting precipitate was filtered off and dried
under vacuum to give complex 40 as a red solid (66 mg, 77%):
mp 165 °C (dec); 1H NMR (250 MHz, CD2Cl2, 240 K) δ 3.35
(d, 1H, J ) 13.0 Hz, CH2), 3.73 (s, 1H, Cp-H), 3.76 (bs, 5H,
Cp-H), 3.85 (d, 1H, J ) 15.5 Hz, CH2), 3.97 (s, 5H, Cp-H), 4.07-
4.19 (m, 3H, Cp-H and CH2), 4.28 (s, 1H, Cp-H), 4.56 (bs, 1H,
Cp-H), 4.61 (s, 1H, Cp-H), 5.07 (pt, 1H, J ) 11.5 Hz, CH2),
5.45 (d, 1H, J ) 15.3 Hz, CH2), 6.02 (d, 1H, J ) 15.3 Hz, CH2),
6.67 (bs, 1H, Ph), 6.86 (m, 1H, Ph), 7.42-7.90 (m, 12H, Ph);
13C NMR (250 MHz, CD2Cl2, 240 K) δ 56.83 (d, J ) 10.6 Hz,
CH2), 57.85 (CH2), 63.58 (CH2), 65.95 (Cp-H), 66.03 (Cp-CH),
66.45 (Cp-CH), 66.60 (Cp-H), 67.36 (Cp-CH), 69.86 (Cp), 70.15
(Cp), 70.38 (Cp-H), 78.27 (Cp-C), 79.66 (Cp-C), 82.15 (Cp-C),
83.26 (Cp-C), 123.94 (d, J ) 52.6 Hz, Ph-C), 124.51 (d, J )
49.8 Hz, Ph-C), 126.84 (d, J ) 70.4 Hz, Ph-C), 128.09 (d, J )
12.6 Hz, Ph-CH), 129.52 (d, J ) 10.9 Hz, Ph-CH), 129.98 (d, J
) 9.2 Hz, Ph-CH), 131.23-134.50 (several m, Ph-CH), 138.66
(d, J ) 15.7 Hz, Ph-C); 31P NMR δ 25.97; MS (FD) m/z (rel %)
(Sp,Sp)-19: [R]λ20) -1813 (589 nm), -2056 (578 nm), -2426
(546 nm), (c ) 1.00); CD λmax (∆ꢀ) ) 285 nm (-10.1), 340 nm
(1.90), 471 nm (-5.05) (c ) 1.03 × 10-3 mol/L).
(1R c ,2S c ,1R ′c ,2S ′c ,R p ,R p )-2,2′′-B is [N -(1-m e t h o x y -1-
p h e n ylp r op -2-yl)-N -m e t h yla m in om e t h yl]-1,1′′-b ife r -
r ocen e (22). Biferrocene 22 was synthesized from (1Rc,2Sc)-
1-iodo-2-[N-(1-methoxy-1-phenylprop-2-yl)-N-methylamino-
methyl]ferrocene (18) using method B at 130 °C, as described
for the synthesis of 19. Crystallization of the residue afforded
compound 22 as an orange solid in 58% yield:
1
mp 213 °C; H NMR δ 0.99 (d, 6H, J ) 6.6 Hz, CH3), 1.96
(s, 6H, CH3), 2.76 (m, 2H, CH), 3.22 (s, 6H, OCH3), 3.27 (d,
2H, J ) 13.6 Hz, CH2), 3.44 (d, 2H, J ) 13.6 Hz, CH2), 4.10
(m, 2H, Cp-H), 4.14 (m, 4H, 2 CH and 2 Cp-H), 4.20 (s, 10H,
Cp-H), 4.34 (m, 2H, Cp-H), 7.22-7.37 (m, 10H, Ph); 13C NMR
δ 9.42 (CH3), 36.80 (CH3), 52.10 (CH2), 56.57 (CH3), 64.68 (CH),
66.02 (Cp-CH), 69.38 (Cp-CH), 69.43 (Cp-CH), 70.89 (CH),
84.68 (Cp-H), 86.52 (br s, Cp-C), 126.87 (Ph-H), 126.94 (Ph-
CH), 127.99 (Ph-CH), 141.75 (Ph-C); MS (220 °C) m/z (rel %)
) 752 (15, M+), 631 (41), 574 (100); [R]λ ) +463 (589 nm),
20
+516 (578 nm), +822 (546 nm) (c ) 0.37). Anal. Calcd for
C
44H52Fe2N2O2: C, 70.22; H, 6.96; N, 3.72. Found: C, 70.04;
20
) 827.9 (50, M - Cl); [R]λ ) +218 (589 nm), +242 (578 nm),
H, 6.91; N, 3.73.
+369 (546 nm), (c ) 0.12); CD λmax (∆ꢀ) ) 293 nm (8.15), 298
nm (-10.5), 304 nm (74.8), 309 (-25.4), 355 nm (7.58), 456
nm (1.68). (c ) 9.97 × 10-4 mol/L).
Bism eth yl Iod id e of (Rp ,Rp )- a n d (Sp ,Sp )-2,2′′-Bis(N,N-
d im eth yla m in om eth yl)-1,1′′-bifer r ocen e (23). Biferrocene
19 (900 mg, 1.86 mmol) was dissolved in acetonitrile (5 mL),
and iodomethane (2.64 g, 18.6 mmol) was added dropwise. A
yellow precipitate formed during the addition of iodomethane.
The mixture was stirred for a further 1 h at room temperature
and then ether (30 mL) was added slowly. The reaction
mixture was stirred for 30 min and the precipitate was filtered
off and washed with ether. The product was dried under
vacuum to give a yellow powder (1.31 g, 92%):
mp 193 °C (dec); 1H NMR (DMSO-d6) δ 2.70 (s, 18H, 6 CH3),
4.45 (s, 10H, Cp-H), 4.50 (d, 2H, J ) 15.2 Hz, CH2), 4.63 (m,
2H, Cp-H), 4.65 (d, 2H, J ) 15.2 Hz, CH2), 4.77 (m, 2H,
Cp-H), 4.93 (m, 2H, Cp-H); 13C NMR (DMSO-d6) δ 51.81 (CH3),
63.77 (CH2), 69.50 (Cp-CH), 70.63 (Cp), 71.46 (Cp-CH), 73.41
(Cp-C), 73.83 (Cp-CH), 85.92 (Cp-C).
X-r a y Cr ysta llogr a p h ic Stu d y of (Rp ,Rp )-40‚3CHCl3.
Crystals of 40 suitable for X-ray structure analysis were
obtained by evaporation crystallization from CHCl3 in the form
of the air-stable solvate (Rp,Rp)-40‚3CHCl3. The compound
(C44H39Cl11Fe2NPPd, Mcalcd ) 1219.77) forms orange-red in-
clined prisms and crystallizes in the acentric and chiral
triclinic space group P1 (no. 1), a ) 10.906(6) Å, b ) 11.205(6)
Å, c ) 11.719(6) Å, R ) 103.44(2)°, â ) 116.36(2)°, γ ) 90.02-
(2)°, V ) 1239.2(11) Å3, Z ) 1, Dcalcd ) 1.636 g/m3, T ) 295(2)
K. A Bruker Smart platform diffractometer with CCD area
detector, Mo-KR radiation, graphite monochromator, and λ )
0.71073 Å was used. A total of 27 190 reflections covering an
entire sphere of the reciprocal space with Θmax ) 30° were
collected, giving 14 014 independent reflections, with correction
for absorption with SADABS19 and Rint ) 0.021. The structure
was solved with direct methods and was refined on F2 with
the program SHELXL9720 using anisotropic temperature fac-
tors for non-hydrogen atoms. Hydrogen atoms were inserted
in idealized positions and were refined riding with the atoms
to which they were bonded. An orientation disorder of the
CHCl3 molecules was taken into account. Final R1 ) 0.047
and wR2 ) 0.093 with 578 parameters and all 14 014 data,
(Rp ,Rp )- a n d (Sp ,Sp )-N-(2-Br om oben zyl)-3,5-d ih yd r o-
4H-d ifer r ocen o[c,e]a zep in e (24). Rou te A. Compound 23
(1.26 g, 1.64 mmol), 2-bromobenzylamine hydrochloride (402
mg, 1.8 mmol), triethylamine (222 mg, 2.2 mmol), and anhy-
drous acetonitrile (20 mL) were mixed in a Schlenk tube. The
degassed mixture was refluxed under an argon atmosphere
for 9 h. The reaction mixture was cooled to room temperature,
the solvent was evaporated under reduced pressure, and the
residue was purified by chromatography on silica gel. Elution
with PE/Et2O/Et3N (60:1:3) afforded azepine 24 (837 mg, 90%).
Rou te B. Biferrocene 22 was first methylated in acetonitrile
with excess iodomethane using the method described for
compound 23. The resulting yellow solid was used without
further purification and was reacted with 2-bromobenzylamine
hydrochloride and triethylamine in anhydrous acetonitrile as
described for route A. Azepine 24 was isolated in an overall
yield of 63%.
and a residual electron density between -0.34 and 0.44 e Å-3
,
Flack absolute structure parameter 0.01(2), and chirality of
the structure in agreement with the chemistry were found.
Selected bond lengths (Å) and angles (deg): Pd-N, 2.145(3);
Pd-P, 2.245(2); Pd-Cl2, 2.275(2); Pd-Cl1, 2.397(2);
N-Pd-P, 93.2(1); N-Pd-Cl2, 173.1(1); N-Pd-Cl1, 91.0(1);
P-Pd-Cl2, 88.8(1); P-Pd-Cl1, 173.2(1); Cl2-Pd-Cl1, 87.7-
(1); C22-C21-C31-C32, 14.1(6). Two of the CHCl3 molecules
are anchored via clear-cut C-H‚‚‚Cl hydrogen bonds to Cl1
(Figure 3).
1
(Rp,Rp)-24: yellow solid; mp 127-129 °C; H NMR δ 3.78
(d, 2H, J ) 15.6 Hz, CH2), 3.93-3.99 (m, 14H, CH2 and 10
Cp-H), 3.99 (bm, 2H, Cp-H), 4.11 (t, 2H, J ) 2.0 Hz, Cp-H),
4.38 (bm, 2H, Cp-H), 7.10-7.14 (m, 1H, Ph), 7.29-7.33 (m,
1H, Ph), 7.53-7.57 (m, 2H, Ph); 13C NMR δ 56.71 (CH2), 59.30
(CH2), 65.86 (Cp-CH), 66.35 (Cp-CH), 67.54 (Cp-CH), 69.90
(Cp), 82.46 (Cp-C), 84.88 (Cp-C), 124.47 (Ph-C), 127.17
(Ph-CH), 128.28 (Ph-CH), 130.44 (Ph-CH), 132.75 (Ph-CH),
138.75 (Ph-C); MS (150 °C) m/z (rel %) ) 581/579 (100/90, M+),
500 (4), 410 (16); [R]λ20 ) +664 (589 nm), +705 (578 nm), +964
(546 nm), (c ) 0.55); CD λmax (∆ꢀ) ) 279 nm (16.7), 286 nm
(-1.04), 295 nm (-3.92), 341 (4.26), 454 nm (2.50) (c ) 9.43 ×
Ack n ow led gm en t. M.W. thanks the Fonds zur
Fo¨rderung der wissenschaftlichen Forschung for finan-
cial support (P11990-CHE). This work was also kindly
supported by O¨ sterreichische Nationalbank (W.W.,
(19) Sheldrick, G. M. SADABS, program for semiempirical absorp-
tion correction from equivalent reflections, University of Go¨ttingen,
Germany, 1996.
(20) Sheldrick, G. M.: SHELXL97, program for crystal structure
refinement, University of Go¨ttingen, Germany, 1997.

2214 J . Org. Chem., Vol. 67, No. 7, 2002
Xiao et al.
4-14; details on empirical force field calculations and output
files for 4 and 40 and their related biaryl derivatives; and
listings of full crystallographic data, atomic coordinates, aniso-
tropic temperature factors, bond lengths, bond angles, and
least-squares planes for (Rp,Rp)-40‚3CHCl3. This material is
available free of charge via the Internet at http://pubs.acs.org.
project 7516). A Ph.D. grant from Bundesministerium
fu¨r Auswa¨rtige Angelegenheiten (Nord-Su¨d-Dialogue
Stipendienprogramm) is kindly acknowledged by L.X.
Su p p or tin g In for m a tion Ava ila ble: Experimental de-
tails for the synthesis of compounds 5-14, 20, 21, 25-32, and
35-39, including spectral and analytical data; experimental
details and full lists of catalytic results from the use of ligands
J O016249W