Coordination chemistry of tetra- and tridentate ferrocenyl polyphosphines: An unprecedented [1,1′-heteroannular and 2,3-homoannular]-phosphorus- bonding framework in a metallocene dinuclear coordination complex

Article abstract of DOI:10.1021/ic7022105

Palladium(II) and nickel(II) halide complexes of the ferrocenyl polyphosphines 1,1′,2,3-tetrakis(diphenylphosphino)ferrocene (1), and 1,1′,2-tris(diphenylphosphino)-4-tert-butylferrocene (5) were prepared and characterized by multinuclear NMR. The metallo-ligand 1, the palladium [Pd ₂Cl₄(1)] (3b) and nickel [NiCl₂(5)] (6) coordination complexes were additionally characterized by X-ray diffraction crystallography. The behavior of 1 toward coordination to nickel and palladium was surprisingly different because the coordination of a second metal center after the initial 1,2-phosphorus-bonding of nickel was markedly difficult. The preference of nickel for 1,2-P coordination on 1,1′-bonding was confirmed by the exclusive formation of 6 from 5. The changes noted between the solid state structure of the ligand 1 and the structure obtained for the dinuclear palladium complex 3b reveal the rotational flexibility of this tetraphosphine. This flexibility is at the origin of the unique framework for a metallocenic dinuclear metal complex in which both coexist a 1,1′-heteroannular chelating P-bonding and a 2,3-homoannular chelating P-bonding with two palladium centers. Some reported specimens of ferrocenyl polyphosphines of constrained geometry have previously revealed that phosphorus lone pair overlap can lead to very intense through-space ³¹P³¹P nuclear spin-spin coupling constants (J_PP) (J. Am. Chem. Soc. 2004, 126 (35), 11077-11087] in solution phase. In these cases, an intemuclear distance between heteroannular phosphorus atoms below 4.9 A, with an adequate orientation of the lone-pairs in the solid state and in solution, was a necessary parameter. The flexibility of the new polyphosphines 1 and 5 does not allow that spatial proximity (intemuclear distances between heteroannular phosphorus above 5.2 A in the solid state); accordingly the expected through-space nuclear spin-spin coupling constants were not detected in any of their coordination complexes nor in 1.

Full text of DOI:10.1021/ic7022105

Inorg. Chem. 2008, 47, 1607-1615
Coordination Chemistry of Tetra- and Tridentate Ferrocenyl
Polyphosphines: An Unprecedented [1,1′-Heteroannular and
2,3-Homoannular]-Phosphorus-Bonding Framework in a Metallocene
Dinuclear Coordination Complex
D. A. Thomas,† V. V. Ivanov,‡ I. R. Butler,*,† P. N. Horton,# P. Meunier,‡ and J.-C. Hierso*,‡
Department of Chemistry, UniVersity of Wales, Bangor, Gwynedd LL57 2UW, Wales, Department
of Chemistry, UniVersity of Southampton, Highfield, Southampton SO17 1BG, U.K., and ICMUB
(Institut de Chimie Moléculaire de l’UniVersité de Bourgogne), UMR CNRS 5260, UniVersité de
Bourgogne, 9 aVenue Alain SaVary, 21078 Dijon, France
Received November 9, 2007
Palladium(II) and nickel(II) halide complexes of the ferrocenyl polyphosphines 1,1′,2,3-tetrakis(diphenylphosphino)ferrocene
(1), and 1,1′,2-tris(diphenylphosphino)-4-tert-butylferrocene (5) were prepared and characterized by multinuclear NMR.
The metallo-ligand 1, the palladium [Pd2Cl4(1)] (3b) and nickel [NiCl2(5)] (6) coordination complexes were additionally
characterized by X-ray diffraction crystallography. The behavior of 1 toward coordination to nickel and palladium was
surprisingly different because the coordination of a second metal center after the initial 1,2-phosphorus-bonding of nickel
was markedly difficult. The preference of nickel for 1,2-P coordination on 1,1′-bonding was confirmed by the exclusive
formation of 6 from 5. The changes noted between the solid state structure of the ligand 1 and the structure obtained
for the dinuclear palladium complex 3b reveal the rotational flexibility of this tetraphosphine. This flexibility is at the origin
of the unique framework for a metallocenic dinuclear metal complex in which both coexist a 1,1′-heteroannular chelating
P-bonding and a 2,3-homoannular chelating P-bonding with two palladium centers. Some reported specimens of ferrocenyl
polyphosphines of constrained geometry have previously revealed that phosphorus lone pair overlap can lead to very
intense “through-space” 31P31P nuclear spin-spin coupling constants (JPP) (J. Am. Chem. Soc. 2004, 126 (35),
11077–11087] in solution phase. In these cases, an internuclear distance between heteroannular phosphorus atoms
below 4.9 Å, with an adequate orientation of the lone-pairs in the solid state and in solution, was a necessary
parameter. The flexibility of the new polyphosphines 1 and 5 does not allow that spatial proximity (internuclear
distances between heteroannular phosphorus above 5.2 Å in the solid state); accordingly the expected through-
space nuclear spin-spin coupling constants were not detected in any of their coordination complexes nor in 1.
Introduction
in metal-catalyzed fine chemicals synthesis has been inten-
sively investigated.3 For instance, carbon-carbon couplings
that result from reaction of organo bromides or chlorides
with various organometallic reagents has excellent synthetic
scope when catalytic systems combining palladium and a
polyphosphine auxiliary are employed in reactions such as
Suzuki-Miyaura arylations3a,e or Sonogashira-Heck
alkynylations.3b,f Among multidentate ligands, ferrocenyl
polyphosphines have been identified as versatile species
Polyphosphine compounds have been fascinating and
useful species since their initial development in the early
1960s. In the recent years, the high-effectiveness of novel
organic and organometallic multidentate phosphine ligands1,2
* To whom correspondence should be addressed. E-mail: hiersojc@
u-bourgogne.fr (J.-C.H.); i.r.butler@bangor.ac.uk (I.R.B.). Phone: +33 3
80 39 61 06(J.-C.H). Fax: +33 3 80 39 36 82 (J.-C.H.).
†
Bangor University.
Université de Bourgogne.
Southampton University.
‡
(2) For recent reviews, see: (a) Hierso, J.-C.; Amardeil, R.; Bentabet, E.;
Broussier, R.; Gautheron, B.; Meunier, P.; Kalck, P. Coord. Chem.
ReV. 2003, 236, 143. (b) Mayer, H. A.; Kaska, W. C. Chem. ReV.
1994, 94, 1239. (c) Cotton, F. A.; Hong, B. Prog. Inorg. Chem. 1992,
40, 179.
#
(1) Hierso, J.-C.; Fihri, A.; Ivanov, V. V.; Hanquet, B.; Pirio, N.;
Donnadieu, B.; Rebière, B.; Amardeil, R.; Meunier, P. J. Am. Chem.
Soc. 2004, 126, 11077.
10.1021/ic7022105 CCC: $40.75
Published on Web 02/02/2008
2008 American Chemical Society
Inorganic Chemistry, Vol. 47, No. 5, 2008 1607

Thomas et al.
amenable to many structural modifications3k and therefore
to the tuning of both chemical features and coordination
properties. Their intrinsic characteristics such as the stability
toward oxygen and moisture permit one to employ them
under nonglove box laboratory conditions and to stock them
under air at ambient temperature for months without troubles
from light or moisture conditions. These features prompted
us to further develop the synthetic routes toward a more
diverse class of ferrocenyl polyphosphines of higher “rank”
than classical ferrocenyl mono or diphosphines.4 The studies
conducted by several other groups for the synthesis of
specifically chiral polysubstituted ferrocenes represent also
very important and complementary contributions in this area.5
Some representatives species of ferrocenyl polyphosphines
of a constrained geometry have allowed the study of the
intriguing and yet poorly known phenomenon of nonbonded
indirect nuclear spin-spin couplings detected by NMR
(“through-space” nuclear spin-coupling, TS). In these com-
pounds, abnormally strong JPP and JCP spin–spin coupling
constants have been observed, and attributed to the spatial
proximity of the NMR-active nucleus.1,3k,4a,b Clear qualita-
tive and a semiquantitative experimental correlations have
been obtained, which links the geometric parameters of the
constrained structures and the intensity of the corresponding
P···P spin coupling constants. The lone-pair overlap theory,
developed for 19F19F and 15N19F through-space couplings in
organic compounds,6 has appeared to be a reliable foundation
to account for our results.1,3k This model was extended to
31P31P spin–spin couplings, showing that only one lone pair
of electrons that interacts with another bonding electron pair
can transfer the 31P31P nuclear spin information through space
between two phosphorus nuclei in a coordination complex.
The exponential dependence of the J coupling intensity on
internuclear distances, initially predicted by modeling and
theoretical means,6h has been experimentally established
through the example of several palladium and nickel tetra-
phosphine mononuclear complexes. As a part of our continu-
ous effort to investigate the TS coupling phenomenon, and
to develop novel multidentate ferrocenyl-based metallo-
ligands, the present paper provides details on the synthesis
and characterization of the original dissymmetric tetraphos-
phine 1,1′,2,3-tetrakis(diphenylphosphino)ferrocene 1. The
coordination chemistry of 1 toward palladium and nickel is
presented. In particular, the behavior of 1 toward coordination
to nickel or palladium was found surprisingly different. A
critical comparison of NMR and X-ray structure between
the new-formed species and already known tetraphosphine
and triphosphine derivatives was conducted to pursue our
efforts at revealing and toward understanding of through-
space 31P31P nuclear spin couplings and their internuclear
distance dependence. The preference of nickel compounds
for 1,2-phosphorus bonding coordination was verified with
the triphosphine parent ligand 1,1′,2-tris(diphenylphosphino)-
4-tert-butylferrocene 5, and its nickel chloride coordination
complex for which an X-ray structure was solved.
The changes noted between the solid-state structure of the
ligand 1 and the structure obtained for its dinuclear palladium
complex 3b revealed the rotational flexibility of this novel
tetraphosphine. This flexibility is at the origin of an unique
framework for a dinuclear complex coordinated by a met-
allocene, in which both coexist a 1,1′-heteroannular P-
bonding and a 2,3-homoannular P-bonding.
(3) (a) Hierso, J.-C.; Fihri, A.; Amardeil, R.; Meunier, P.; Doucet, H.;
Santelli, M.; Donnadieu, B. Organometallics 2003, 22, 4490. (b)
Hierso, J.-C.; Fihri, A.; Amardeil, R.; Meunier, P.; Doucet, H.; Santelli,
M.; Ivanov, V. V. Org. Lett. 2004, 6, 3473. (c) Feuerstein, M.;
Laurenti, D.; Bougeant, C.; Doucet, H.; Santelli, M. Chem. Commun.
2001, 4, 325. (d) Laurenti, D.; Feurstein, M.; Pepe, G.; Doucet, H.;
Santelli, M. J. Org. Chem. 2001, 66, 1633. (e) Feuerstein, M.; Laurenti,
D.; Doucet, H.; Santelli, M. Synthesis. 2001, 4, 2320. (f) Feuerstein,
M.; Berthiol, F.; Doucet, H.; Santelli, M. Org. Biomol. Chem. 2003,
1, 2235. (g) Broussard, M. E.; Juma, B.; Train, S. G.; Peng, W.-J.;
Laneman, S. A.; Stanley, G. G. Science 1993, 260, 1784. (h) Schill,
H.; de Meijere, A.; Yufit, D. S. Org. Lett. 2007, 9, 2617. (i) Kondolff,
I.; Feurstein, M.; Doucet, H.; Santelli, M. Tetrahedron 2007, 63, 9514.
For reviews, see: (j) Hierso, J.-C.; Beaupérin, M.; Meunier, P. Eur.
J. Inorg. Chem. 2007, 3767. (k) Hierso, J.-C.; Smaliy, R.; Amardeil,
R.; Meunier, P. Chem. Soc. ReV. 2007, 36, 1754.
Experimental Section
The reactions were carried out in oven-dried (115 °C) glassware
under an argon atmosphere using Schlenk and vacuum-line
techniques. Palladium and nickel precursors, ferrocene, and FeCl2
from commercial source were used (FeCl2 anhydrous beads, 99.9%,
H2O < 100 ppm). The solvents were distilled over appropriate
1
drying and deoxygenating agents prior to use. H (500.13 MHz),
31P (202.44 MHz), and 13C NMR (125.77 MHz) were performed
in our laboratories (on Bruker DRX 500, or on Bruker 300) in
CDCl3 at 298 K unless otherwise is stated. Mass and elemental
analyses were performed by the analytical services of ICMUB
(Dijon) on a Kratos Concept IS instrument and on Eager 200,
respectively.
(4) (a) Hierso, J.-C.; Ivanov, V. V.; Amardeil, R.; Richard, P.; Meunier,
P. Chem. Lett. 2004, 33, 1296. (b) Ivanov, V. V.; Hierso, J.-C.;
Amardeil, R.; Meunier, P. Organometallics 2006, 25, 989. (c) Butler,
I. R.; Horton, P. N.; Fortune, K. M.; Morris, K.; Greenwell, C. H.;
Eastham, G. R.; Hursthouse, M. B. Inorg. Chem. Commun. 2004, 7,
923. (d) Butler, I. R.; Baker, P. K.; Eastham, G. R.; Fortune, K. M.;
Horton, P. N.; Hursthouse, M. B. Inorg. Chem. Commun. 2004, 7,
1049. (e) Butler, I. R.; Müssig, S.; Plath, M. Inorg. Chem. Commun.
1999, 2, 424. (f) A preliminary communication related to the present
work has been reported, see: Butler, I. R.; Drew, M. G. B.; Greenwell,
C. H.; Lewis, E.; Plath, M.; Müssig, S.; Szewczyk, J. Inorg. Chem.
Commun. 1999, 2, 576.
1,1′,2,3-Tetrakis(diphenylphosphino)ferrocene (1). LDA (1
mol equiv to 1,1′-dibromoferrocene) was prepared in situ from
n-butyllithium and di-isopropylamine at -80 °C in THF. 1,1′-
(6) (a) Mallory, F. B. J. Am. Chem. Soc. 1973, 95, 7747. (b) Mallory,
F. B.; Mallory, C. W.; Fedarko, M.-C. J. Am. Chem. Soc. 1974, 96,
3536. (c) Mallory, F. B.; Mallory, C. W.; Ricker, W. M. J. Am. Chem.
Soc. 1975, 97, 4770. (d) Mallory, F. B.; Mallory, C. W.; Ricker, W. M.
J. Org. Chem. 1985, 50, 457. (e) Mallory, F. B.; Mallory, C. W. J. Am.
Chem. Soc. 1985, 107, 4816. (f) Mallory, F. B.; Mallory, C. W.; Baker,
M. B. J. Am. Chem. Soc. 1990, 112, 2577. (g) Mallory, F. B.; Luzik,
E. D., Jr.; Mallory, C. W.; Carroll, P. J. J. Org. Chem. 1992, 57, 366.
(h) Mallory, F. B.; Mallory, C. W.; Butler, K. E.; Lewis, M. B.; Xia,
A. Q.; Luzik, E. D., Jr.; Fredenburgh, L. E.; Ramanjulu, M. M.; Van,
Q. N.; Francl, M. M.; Freed, D. A.; Wray, C. C.; Hann, C.; Nerz-
Stormes, M.; Carroll, P. J.; Chirlian, L. E J. Am. Chem. Soc. 2000,
122, 4108 and references therein.
(5) For example, see: (a) Iftime, G.; Daran, J.-C.; Manoury, E.; Balavoine,
G. G. A. Organometallics 1996, 15, 4808. (b) Balavoine, G. G. A.;
Daran, J.-C.; Iftime, G.; Manoury, E. J. Organomet. Chem. 1998, 567,
191. (c) Deng, W.-P.; Hou, X.-L.; Dai, L.-X.; Yu, Y.-H.; Xia, W.
Chem. Commun. 2000, 285. (d) Deng, W.-P.; You, S.-L.; Hou, X.-L.;
Dai, L.-X.; Yu, Y.-H.; Xia, W.; Sun, J. J. Am. Chem. Soc. 2001, 123,
6508. (e) Chesney, A.; Bryce, M. R.; Chubb, R. W. J.; Batsanov, A. S.;
Howard, J. A. K. Synthesis 1998, 413. (f) For a recent excellent review,
see Gomez Arrayas, R.; Adrio, J.; Carretero, J.-C. Angew. Chem., Int.
Ed. 2006, 45, 7674.
1608 Inorganic Chemistry, Vol. 47, No. 5, 2008

Coordination Chemistry of Ferrocenyl Polyphosphines
Scheme 1. Four-Step Synthesis of the Tetraphosphine 1
Dibromoferrocene (0.5 mol equiv. to LDA), dissolved in the
minimum quantity of THF required for its complete dissolution,
was then added. The reaction mixture was stirred for 2 h at a
temperature between -50 and -80 °C, with external cooling by a
dry ice/acetone bath. ClPPh2 (1 mol equiv to 1,1′-dibromoferrocene)
was added to get product B (Scheme 1), and the reaction mixture
was allowed to warm to room temperature over 2 h. Water was
added to the reaction mixture (in some cases quantities of diethyl
ether were also added to facilitate phase separation), and the organic
phase was separated; the aqueous phase was further extracted with
diethyl ether before the combined organic phases were dried over
anhydrous magnesium sulfate. Following filtration, silica gel was
added, and the solvent was removed to leave the products adhered
to silica gel. The silica gel fraction was then added as a top layer
to column chromatographic support, and the products were isolated.
The first bright yellow orange fraction eluted with a 3% solution
of diethyl ether in petroleum ether was a low-yield mixture of
unreacted dibromoferrocene and traces of monobromoferrocene. The
reactions products [1,1′-dibromo-2,5-bis(diphenylphosphino)]fer-
rocene B (55%) and [1,1′-dibromo-2,2′-bis(diphenylphosphino)]-
ferrocene (45%) were subsequently eluted by incremental addition
of further quantities of diethyl ether. The isolation procedure can
be simplified since we have observed that mixtures of the previous
compounds, which coelute, crystallize independently upon ether
evaporation, with compound B, forming large well-formed brown-
orange crystals easily separable from the pale yellow powder formed
by the over ferrocene derivatives. Compound B: C34H26Br2FeP2;
m/z 712 (parent ion); calcd C 57.34, H 3.68; found C 57.43, H
3.61; 1H NMR δ 3.74 (s, 2H, Cp), 4.12 (m, 2H, Cp), 4.29 (pt, 2H,
Cp), 7.15–7.41 (series of m, 20H Ph); 31P{1H} NMR δ -22.4. A
crystal structure of the product B is reported in ref 4f.
1,1′,2,3-Tetrakis(diphenylphosphino)ferrocene 1 was obtained in
a high 90% yield from lithiation of B (1 mol equiv) with
n-butyllithium (2.2 mol equiv) at -80 °C in diethyl ether (1 h),
followed by quenching with chlorodiphenylphosphine (2.2 mol
equiv.); the reaction mixture was allowed to warm to room
temperature over 2 h. Water was added to the crude product, and
the organic phase separated was subjected to column chromatog-
raphy for purification. C58H46FeP4; m/z 922 (parent ion); calcd C
75.50, H 5.02; found C 75.60, H 4.88; 1H NMR δ 3.50 (pq, 2H, J
≈ 1.8 Hz), 3.90 (s, 2H), 4.38 (pt, 2H, J ≈ 1.8 Hz), 6.80–7.30 (series
of m, 40H); 31P{1H} NMR δ -24.1 (d, JPP ) 59.7 Hz), -18.7 (s),
-15.0 (t, JPP ) 59.7 Hz); 13C{1H} NMR δ 75.4, 75.8, 75.9, 77.6
(s, 1 each, C-Fc), 78.7 (d, C-Fc), 87.0, 87.1, 87.3, 88.5, 88.7 (s,
br, 1 each C-Fc), 127.5, 127.6, 127.8 (d, J ) 16 Hz, C-Ph), 128.0
(m, C-Ph), 128.2 (s, C-Ph), 133.1 (dd, C-Ph), 134.7, 134.9,
137.0, 138.7, 139.2 (d, C-Ph). Crystallographic data are available
as Supporting Information.
(s, 1H), 4.75 (s, 1H), 6.75–8.85 (series of m, 40H); 31P{1H} NMR
δ -29.2 (s), -18.6 (s), 40.3 (d, JPP ) 10.9 Hz), 52.2 (d, JPP
)
10.9 Hz).
[Pd2Cl4(1)] (3b). The complex was isolated in quantitative yield
and pure form from the addition at RT of [PdCl2(cod)] or
[PdCl2(PhCN)2] (2 mol equiv) to 1 (1 mol equiv) in a minimum of
chloroform after stirring for 3 h and evaporation of solvent and
volatiles. Slow evaporation of a chloroform solution allowed to
obtained single crystals suitable for X-ray diffraction studies:
C58H46Cl4FeP4Pd4; m/z 1277 (parent ion); calcd C 54.54, H 3.63;
found C 54.60, H 3.52; 1H NMR δ 3.08 (s, 2H), 3.28 (s, 2H), 3.60
(s, 2H), 6.75–8.10 (series of m, 40H); 31P{1H} NMR δ 23.6 (d, JPP
) 22.5 Hz), 34.3 (d, JPP ) 9.0 Hz), 44.6 (d, JPP ) 22.5 Hz), 62.3
(d, JPP ) 9.0 Hz). Because of solubility troubles, a proper 13C NMR
assignment was difficult to obtain.
[NiCl2(1)] (4a). The complex was quantitatively obtained by
reaction of 1 with 1 mol equiv of NiCl2DME in chlorinated solvent.
Addition of a 2-10 excess of nickel halide and stirring for five
days did not produce any dinuclear nickel complex; changes of
solvent (such as DMSO) were not more helpful. After filtration
and solvent/volatile evaporation, the complex is obtained as a red-
orange powder: C58H46Cl2FeP4Ni; m/z 1050 (parent ion); calcd C
1
66.20, H 4.41; found C 66.08, H 4.48; H NMR δ 2.00 (s, 1H),
2.31 (s, 1H), 3.01(s, 1H), 3.85 (s, 1H), 4.45 (s, 1H), 4.65 (s, 1H),
6.80–9.30 (series of m, 40H); 31P{1H} NMR: δ -28.6 (s), -18.8
(s), 38.3 (d, JPP ) 69.8 Hz), 49.4 (d, JPP ) 69.8 Hz).
1,1′,2-Tris(diphenylphosphino)-4-tert-butylferrocene (5). The
ligand was synthesized in THF following a reported procedure:4b
to a stirred suspension of FeCl2 (0.53 g, 4.2 mmol) was added at
-40 °C a THF solution of tBuCp(PPh2)2Li (2.18 g, 4.4 mmol)
dropwise. After the mixture was stirred for 2 h, it was treated at
-40 °C with a solution of Cp(PPh2)Li (1.00 g, 3.9 mmol). The
THF was evaporated under vacuum, and the residue was refluxed
in toluene for 3 h. After filtration, the crude product was subjected
to column chromatography (SiO2, toluene hexane 4:1) to provide
2.60 g of 5 (yield 84%): C50H45FeP3 (794.7); calcd C 75.57, H
5.71; found 75.59, H 5.68; m/z 794; 1H NMR δ 1.03 (s, 9H, t-Bu),
4.07 (dd, JPH ) 3.3 Hz, 3JHH ) 1.8 Hz, 2H, Cp), 4.17 (dd, 3JHH
≈
JPH ) 1.8 Hz, 2H, Cp), 4.20 (dd, 2 JPH ≈ 1.5 Hz, 2H, 3,5-Cp),
6.91–7.12 (m, 10H, Ph, 1,2-PPh2), 7.13–7.26 (m, 10H, Ph, 1′-PPh2),
7.33 (m, 6H, m,p-Ph 1,2-PPh2), 7.64 (m, 4H, o-Ph 1,2-PPh2);
31P{1H}(CDCl3) δ -22.7 (d, TSJ ) 2.8 Hz, 1,2-PPh2), -19.3 (t,
TSJ ) 2.8 Hz, 1′-PPh2); 31P{1H}(C6D6) δ -21.9 (d, J ) 5.0 Hz,
1,2-PPh2), -20.0 (t, J ) 5.0 Hz, 1′-PPh2); 13C{1H} δ 30.5 (s,
C(CH3)3), 31.3 (d, JPC ) 1.5 Hz, C(CH3)3), 71.4 (m, 3,5-Fc), 72.0
(m, 2 C-Cp), 74.4 (d, JPC ) 11.0 Hz, 2 C-Cp), 80.2 (d, JPC ) 12.5
Hz, 1′-Fc), 81.9 (m, 1,2-Fc), 106.8 (s, 4-Fc), 127.4 (t, 2JPC ≈ 3.5
Hz, m-Ph), 127.5 (s, p-Ph), 128.0 (d, JPC ) 7.2 Hz, m-Ph), 128.1
(t, 2JPC ≈ 8.8 Hz, m-Ph), 128.3 (s, p-Ph), 128.8 (s, p-Ph), 132.9 (t,
2JPC ≈ 10.3 Hz, o-Ph), 133.8 (d, JPC ) 21.0 Hz, o-Ph), 135.1 (dt,
2JPC ≈ 10.9 Hz, JP′C ) 1.5 Hz, o-Ph), 137.6 (t, 2JPC ≈ 2.9 Hz,
ipso-Ph), 138.8 (t, 2JPC ≈ 4.7 Hz, ipso-Ph), 139.5 (d, JPC ) 14.8
Hz, ipso-Ph).
[PdCl2(1)] (3a). The complex was immediately formed as an
intermediate species, together with a minor amount of 3b, upon
addition at room temperature (RT) of [PdCl2(cod)] or [PdCl2(PhCN)2]
(2 mol equiv) in a minimum of chloroform to 1 (1 mol equiv): 1H
NMR δ 2.05 (s, 1H), 2.42 (s, 1H), 2.94 (s, 1H), 4.08 (s, 1H), 4.50
Inorganic Chemistry, Vol. 47, No. 5, 2008 1609

Thomas et al.
[NiCl2(5)] (6). A mixture of 1,1′,2-tris(diphenylphosphino)-4-
tert-butylferrocene (857 mg, 1.08 mmol) and NiCl2DME (237 mg,
1.08 mmol) was refluxed in acetonitrile for 3 h. The solvent was
evaporated, and the residue was refluxed in toluene for 10 min.
After it cooled, the solid material was filtered off, giving a yield of
729 mg (73%): C50H45Cl2P3FeNi (924.27); calcd. C 64.98, H 4.91;
1
found C 64.83, H, 5.11; H NMR δ 1.31 (s, 9H, tBu), 3.29 (m,
2H, Fc), 3.63 (m, 2H, Fc), 4.35 (s, 2H, 3,5-Fc), 7.03–7.54 (m, 26H,
m, p-Ph), 8.22 (m, 4H, o-Ph); 31P{1H} NMR: δ -20.1 (br, 1′-PPh2),
36.2 (br, 1,2-PPh2); 13C{1H} NMR: δ 32.1 (C(CH3)3), 32.7
(C(CH3)3), 68.5 (s, 2 C-Cp), 72.5 (s, 3,5-Fc), 75.4 (s, 2 C-Cp),
83.9 (s, 1′(or 1,2)-Fc), 86.9 (s, 1,2(or 1′)-Fc), 118.8 (s, 4-Cp), 128.6,
128.7, 128.8, 129.3, 131.1, 131.8, 133.2 (s, CH-Ph), 134.0 (br,
CH-Ph), 135.3 (s, CH-Ph), 138.3 (s, ipso-Ph).
[PdCl2(5)], (7). A mixture of 1,1′,2-tris(diphenylphosphino)-4-
tert-butylferrocene (200 mg, 0.25 mmol) and PdCl2 (47 mg, 0.27
mmol) was refluxed under stirring in THF overnight. The solvent
was evaporated in vacuum, and the residue was dissolved in CH2Cl2.
After addition of ether (20 mL), the reaction mixture was filtered
Figure 1. Tetraphosphine 1,1′,2,2′-tetrakis(diphenylphosphino)-4,4′-di-tert-
butylferrocene 2 exhibits an AA′BB′ 31P NMR spectrum in CDCl3.
groups. On the basis of an analogous ortho-metalation
process, the tetraphosphine 1 can be obtained in four
synthetic steps starting from the easily available 1,1′-
dibromoferrocene (Scheme 1). The first synthesis step leading
to compound A is the more delicate because an accurate
adjustment of the amount of lithium diisopropylamide reagent
(LDA) is necessary to limit the formation of side products
as monolithiated and 2,2′-dilithiathed dibromoferrocenes.4f
Quenching of the mixture with chlorodiphenylphosphine
results in the major formation of compound B, which can
then undergo after purification a selective substitution of
bromides by lithium atoms using nBuLi, in a third step.
Phosphorylation of the resulting compound C, using chlo-
rodiphenylphosphine, quantitatively leads to the dissymmetric
tetraphosphine 1.
The characterization of 1 both in solution and in the solid
state is of great interest for comparison with the unusual
spectroscopic and structural properties of the closely related
tetraphosphine 1,1′,2,2′-tetrakis(diphenylphosphino)-4,4′-di-
tert-butylferrocene (2 in Figure 1) that has been previously
reported.1 The blocked cisoid conformation of 2, resulting
from the presence of encumbering tBu substituents on
cyclopentadienyl (Cp) rings, enforces a close proximity of
the four phosphorus atoms and lead to the existence of
spin–spin nuclear coupling constants via non-bonded interac-
tion (through-space spin coupling). The resulting AA′BB′
spin system (Figure 1, right) comprises a very intense JP1P2
coupling of 59.8 Hz between the two pseudoeclipsed
heteroannular phosphorus P1 and P2, clearly identified as a
through-space coupling arising from close vicinity of the
nucleus (with lone-pair electronic cloud overlap) even in
solution (the distance in the solid state P···P ) 3.728(2) Å).
In the 31P NMR spectrum obtained for 1, the signals
corresponding to the phosphorus, (P1, P3) and P2, on the
same cyclopentadienyl ring (Cp), are found as a doublet and
a triplet at -24.1 and -15.0 ppm, respectively (Figure 2).
A 3JPP coupling constant of ∼60 Hz is observed between P2
and the isochronous nucleus (P1, P3). In contrast to the
spectroscopic features reported for 2 concerning its het-
eroannular phosphorus atoms, the phosphorus P4 on the
second Cp ring did not exhibit any nuclear spin coupling
with the three other phosphorus (singlet at -18.7 ppm,
see Figure 2). On the basis of the lone-pair phosphorus
and evaporated, and the residue was crystallized from
CH2Cl2-hexane mixture to yield 230 mg of (95%):
C50H45Cl2P3FePd (972); calcd C 61.78, H 4.67; found C 61.62, H
4.56; 1H NMR δ 1.35 (s, 9H, t-Bu), 3.35 (dd, JHH ) 1.8 Hz, JPP
a
7
)
3.3 Hz, 2H, Fc), 3.59 (dd, JHH ≈ JPP ≈ 1.8 Hz, 2H, Fc), 4.47 (dd,
JHH ≈ JP-P ≈ 1.2 Hz, 2H, 3,5-Fc), 7.0–7.6 (m, 26H, the rest of Ph
protons), 8.12 (m, 4H, o-Ph); 31P{1H} NMR δ -20.8 (s, 1′-PPh2),
45.3 (s, 1,2-PPh2); 13C{1H} NMR δ 32.1 (m, C(CH3)3), 32.8 (s,
C(CH3)3), 69.2 (m, 3,5-Fc), 72.5 (d, JPC ) 2.3 Hz, 2 C-Cp), 75.6
(d, JPC ) 9.1 Hz, 2 C-Cp), 84.4 (d, JPC ) 13.6 Hz, 1′-Fc), 86.1
(m, 1,2-Fc), 119.7 (m, 4-Fc), 128.7–129.1 (m, CH-Ph), 129.4 (s,
CH-Ph), 130.8 (dd, JPC ) 65.7 Hz, JPC ) 2.3 Hz, ipso-Ph), 131.5
(s, CH-Ph), 132.2 (dd, JPC ) 57.4 Hz, JPC ) 2.3 Hz, ipso-Ph),
132.3 (s, CH-Ph), 132.8 (m, o-Ph), 134.0 (d, JPC ) 21.1 Hz, o-Ph),
135.5 (m, o-Ph), 138.2 (d, JPC ) 13.6 Hz, ipso-Ph).
Crystallographic Data. For 1 and 3b data were collected at the
University of Southampton on a Nonius Kappa CDD diffractometer
with Mo KR radiation (λ ) 0.71073Å, details are available as
Supporting Information). The data for compound 6 (single crystals
[6·2(CHCl3)] were obtained from CHCl3) were collected at 160 K
on a IPDS STOE diffractometer using a graphite-monochromated
Mo KR radiation (λ ) 0.71073 Å), and equipped with an Oxford
Cryosystems Cryostream Cooler Device at the LCC-CNRS Tou-
louse. The final unit cell parameters were obtained by means of a
least-squares refinement; structures were solved by direct methods
using SHELX97 and were refined by means of least-squares
procedures on F2.
Results and Discussion
Tetraphosphine Synthesis and Characterization. An
examination of the literature reveals that a number of efficient
methods are available for the selective formation of 1,2-
substituted ferrocenes via ortho-directing groups (chiral or
nonchiral), for example, nonexhaustively, amine,7 acetal,8
oxazoline,9 amide,10 sulfoxide,11 and phosphine oxide12
(7) Hayashi, T.; Yamamoto, K.; Kumada, M. Tetrahedron Lett. 1974,
4405.
(8) Riant, O.; Samuel, O.; Kagan, H. B. J. Am. Chem. Soc. 1993, 5835.
(9) Sammakia, T.; Latham, H. A.; Schaad, D. R. J. Org. Chem. 1995, 60,
10.
(10) Tsukazaki, M.; Tinkl, M.; Roglans, A.; Chapell, B. J.; Taylor, N. J.;
Snieckus, V. J. Am. Chem. Soc. 1996, 118, 685.
(11) Hua, D. H.; Lagneau, N. M.; Chen, Y.; Robben, P. M.; Clapham, G.;
Robinson, P. D. J. Org. Chem. 1996, 61, 4508.
(12) Nettekoven, U.; Widhalm, M.; Kamer, P. C. J.; van Leeuwen,
P. W. N. M.; Mereiter, K.; Lutz, M.; Spek, A. L. Organometallics
2000, 19, 2299.
1610 Inorganic Chemistry, Vol. 47, No. 5, 2008

Coordination Chemistry of Ferrocenyl Polyphosphines
above 4.90 Å (this value is nevertheless far superior to
the physical data known as the sum of van der Waals radii
for phosphorus,13 3.6 Å).1 Furthermore, we still consis-
tently note that for P···P internuclear distances below 4.00
Å strong coupling constants (above 20 Hz) are observed.
It is noteworthy that in the conformation found in the solid
state the P···P internuclear distances between (P1, P2) and
(P2, P3) are substantially different (about 0.5 Å); the shortest
distances is found between the phosphorus atoms (P2,P3)
for which the lone-pairs are pointing toward each other.
Similarly, the position of P4 is much closer to P3 (P3···P4
) 5.180 Å) than to P1 (P1···P4 ) 6.335 Å), inducing a
general dissymmetry of the tetraphosphine in the preferred
conformation found in the solid state.
Figure 2. 31P NMR spectrum of the tetraphosphine 1,1′,2,3-tetrakis(diphe-
nylphosphino)ferrocene 1 in CDCl3.
Coordination of Tri and Tetraphosphines toward
Group 10 Transition Metals. A correlation between the
internuclear phosphorus atoms distances through-space and
the intensity of 31P nuclear spin couplings could be estab-
lished within the framework of nickel and palladium
coordination complexes of 2 (Figure 1).1 We were, thus,
much interested in the synthesis of some palladium and nickel
coordination complexes incorporating the new ligand 1. One
of the goals was to estimate whether such a correlation could
also exist for these derivatives. On the basis of the conforma-
tion obtained for 1 in the solid state (Figure 3), it was also
interesting to find which kind of coordination species would
be favored. Upon reaction of 1 with 2 equiv of palladium
chloride precursors (PdCl2cod or PdCl2(RCN)2, cod ) 1,5-
cyclooctadiene, R ) Me, Ph, Scheme 2) it was observed
that within the ten first minutes a rapid coordination occurred,
initially at the trisubstituted ring leading in majority to the
formation of the palladium mononuclear complex 3a.
For 3a, the signals corresponding to the three phosphorus
P1, P2, and P3 on the same Cp ring are easily distinguishable:
two doublets at 40.3 and 52.2 ppm are attributed to P1 and
P2 (2JP1P2 ) 11 Hz), respectively. The signal corresponding
to P3 appears as a singlet at -29.2 ppm, no longer showing
any spin coupling to the phosphorus P2. This might be an
indication of a possible contribution of the lone-pairs on the
Figure 3. Plot of metallo-ligand 1 (30% probability ellipsoids). Hydrogen
atoms are omitted for clarity. Selected bond distances (Å): Fe-CNT(1) )
1.637, Fe-CNT(2) ) 1.653. Selected internuclear distances (Å): P1···P2 )
3.805, P2···P3 ) 3.336, P1···P4 ) 6.335, P3···P4 ) 5.180.
interaction model we developed for tetraphosphine fer-
rocenyl derivatives,1 we anticipated that the structure of
the polyphosphine 1 would not enforce a position of close
vicinity for P4 to any of the other phosphorus atoms. Our
hypothesis was a posteriori supported by the solid-state
molecular structure displayed in Figure 3 (crystallographic
data are given in Table 1), in which P4 is facing in a
direction opposite to that of P1, P2, and P3. In the absence
of structural constraint for the rotation of the Cp rings, in
this conformation, a through-space coupling transmitted
via lone-pair electron overlapping interaction is barely
intensity of 3JP1P2 coupling in the ligand 1 (Table 1, 3JP2P3
)
59.7 HZ).14 The chemical shift corresponding to phosphorus
P4, on the other Cp ring, is not affected by the coordination
to palladium and a singlet at -18.7 ppm was still detectable.
Because of the absence of nuclear spin-coupling between
the bonded and the free phosphorus atoms, we definitely
admit the absence of TS J couplings in 3a.
The coordination of a second palladium is a slower
process, which results in a complete conversion of the
mixture of 1 and 3a into the dinuclear palladium complex
3b after three hours. The 31P NMR spectrum observed for
3b (Figure 4) gave the first evidence of the conformational
changes in the ferrocenyl backbone: a rotation of the ring
being required to allow the heteroannular coordination of a
4
conceivable (while a through-bond JPP coupling would
be negligible). Table 2 presents the internuclear through-
space distances found by X-ray diffraction between the
four phosphorus atoms of ligand 1, together with the
corresponding data for ligand 2. The internuclear distances
between heteroannular phosphorus atoms P1···P4, P2···P4,
and P3···P4 (Figure 3), for which no spin coupling was
detected, range between 5.180 and 6.747 Å. This is
consistent with previously reported studies in which
ferrocenyl tetraphosphine derivatives were found to have
no detectable JPP′ that can be observed with P···P′ distances
(13) (a) Bondi, A. J. Phys. Chem. 1964, 68, 441. (b) Pauling, L. The Nature
of the Chemical Bond; Cornell University Press: Ithaca, NY, 1945..
(14) With no certainty, J coupling might be, for example, increased by a
TS contribution resulting from a lateral overlapping of phosphorus
lone pairs in 1, which is absent in 3a.
Inorganic Chemistry, Vol. 47, No. 5, 2008 1611

Thomas et al.
Table 1. Crystal, Collection, and Refinement Parameters for 1 and 3b
1
3b
C58H46Cl4FeP4Pd2
empirical formula
fweight
C58H46FeP4
922.68
1277.28
temp
120(2) K
0.71073 Å
monoclinic
P21/n
a ) 12.2404(10) Å, R ) 90°
b ) 21.4422(11) Å, ꢀ ) 90.914(3)°
c ) 17.6539(14) Å, γ ) 90°
4632.9(6) Å3
120(2) K
0.71073 Å
monoclinic
P21/c
a ) 21.099(17) Å, R ) 90°
b ) 19.036(19) Å, ꢀ ) 112.19(4)°
c ) 17.797(6) Å, γ ) 90°
6619(9) Å3
wavelength
cryst system
space group
unit cell dimensions
vol
Z
4
4
density (calcd)
abs coeff
1.323 Mg/m3
1.282 Mg/m3
-1
-1
0.503 mm
1.043 mm
F(000)
1920
2560
cryst shape, color
cryst size
θ range for data collection
index range
shard, orange
rod, orange
0.24 × 0.22 × 0.14 mm3
0.28 × 0.04 × 0.03 mm3
3.30–27.48°
2.91–27.48°
-14 e h e 15
-24 e h e 252
-22 e k e 2
-21 e l e 20
48 345
-26 e k e 27
-19 e l e 22
reflns collected
44 519
independent reflns
completeness to θ ) 27.48°
abs correction
10 084 [Rint ) 0.1261]
11 234 [Rint ) 0.1810]
95.0%
95.0%
semiempirical from equivalents
0.9329 and 0.8888
full-matrix least-squares on F2
10 084/0/568
semiempirical from equivalents
0.9694 and 0.7589
full-matrix least-squares on F2
11 234 /623/622
1.028
R1 ) 0.1343, wR2 ) 0.3333
R1 ) 0.2609, wR2 ) 0.3821
0.959 and -1.133 e Å-3
max. and min. transm
refinement method
data/restraints/params
GOF on F2
1.054
Final R indices [F2 > 2σ(F2)]
R indices (all data)
largest diff. peak and hole
R1 ) 0.0826, wR2 ) 0.1807
R1 ) 0.1939, wR2 ) 0.2193
1.248 and -0.558 e Å-3
Table 2. Through-Space Distances between Phosphorus Nuclei (from
X-ray Data) and the Corresponding 31P NMR JPP Coupling Constants
modes. The measured bite angles P-Pd-P (nonstandard-
ized)15 are within the range of typical values of the different
modes of coordination (∼96-102° for 1,1′-P bonded and
around 84-88° for 1,2-P bonded ferrocene complexes) with
a value of 102.4° for P3-Pd-P4 and 88.9° for P1-Pd-P2.
These structural data are important with regard to complexes
reactivity, and especially in catalysis.16 The other distances
and angles (Figure 5) are consistent with reported values for
compounds of this family.3a
On the basis of these results, the coordination chemistry
of 1 with nickel was of particular interest because it is known
that the coordination framework with ferrocenyl diphosphines
could be either tetrahedral (in 1,1′-P substituted derivatives
such as with the ubiquitous 1,1′-bis(diphenylphosphino)fer-
rocene dppf)17 or square planar (in 1,2-P substituted deriva-
tives); the ligand 1 appeared as very useful to investigate
the coordination preference of nickel. Upon reaction of 1
with 1 or 2 equiv of nickel halide precursors (NiCl2DME or
NiBr2DME, DME ) 1,2-dimethoxyethane) it was observed,
as in the case of palladium, the rapid coordination of a nickel
dichloride fragment to two adjacent phosphorus atoms of
the trisubstituted ring (Scheme 3).
P···P distancesa
(Å)/coupling constant (Hz)
2b
1
3b
d(P1···P2)/JPP
d(P1···P3)/JPP
d(P4···P2)/JPP
d(P4···P3)/JPP
d(P1···P4)/JPP
d(P2···P3)/JPP
a
3.728/59.8
4.861/ND
4.861/ND
6.633/ND
3.364/74.5
3.364/74.5
3.805/59.7
5.788/ND
6.747/ND
5.180/ND
6.335/ND
3.336/59.7
3.119/9
5.854/ND
6.236/ND
3.570/22
6.862/ND
4.226/ND
The numbers attributed to phosphorus nucleus correspond to those
displayed in Figures 1 and 3 and in Scheme 2. b Data from ref 1. ND ) no
coupling detected.
PdCl2 moiety via P3 and P4 (the closer atoms in an
heteroannular pair). A confirmation of this unprecedented
palladium dinuclear framework was given by the X-ray
diffraction structure solved (Figure 5, Table 2). In the NMR
spectrum of 3b, four different signals are observed as two
pairs of related doublets. The significantly different JPP
coupling constants, 9 and 22 Hz, allowed us to assign the
signals by comparison with 3a; the more shielded phosphorus
nucleus are assumed to occupy external positions.3a There-
fore, two doublets corresponding to P1 and P2 are found at
34.3 and 62.3 ppm, respectively (JPP ) 9 Hz, close to the J
constant found in 3a), and two doublets corresponding to
P3 and P4 are found at 44.6 and 23.6 ppm, respectively (JPP
) 22 Hz).
In the molecular structure of 3b, the palladium centers
both lie in a square planar environment. However, whereas
coordination complexes of metallocenes incorporating either
1,2-homoannular or 1,1′-heteroannular chelating frameworks
are very common, the complex 3b is a unique example of
metallocene derivative incorporating both these coordination
In this case, however, in contrast to palladium, there was
no indication of further coordination of the remaining
heteroannular phosphine atoms toward a dinuclear species.
The 31P NMR characterization of 4a in CDCl3 (Figure 6)
(15) For bite angle definitions, see: Dierkes, P.; van Leeuwen, P. W. N. M
J. Chem. Soc., Dalton Trans. 1999, 1519 and references therein.
(16) Fihri, A.; Meunier, P.; Hierso, J.-C. Coord. Chem. ReV. 2007, 251,
2017.
(17) Butler, I. R.; Cullen, W. R.; Kim, T.-J.; Rettig, S. J.; Trotter, J.
Organometallics 1985, 4, 972.
1612 Inorganic Chemistry, Vol. 47, No. 5, 2008

Coordination Chemistry of Ferrocenyl Polyphosphines
Scheme 2. Formation of the Palladium Complexes 3a and 3b (the Phenyl Groups on Phosphorus Are Omitted for Clarity)
Scheme 3. Formation of the Mononuclear Nickel Complexes 4aa
shows two doublets for the bonded phosphorus centered at
49.4 and 38.3 (2JPP ) 70 Hz). The signal corresponding to
the last phosphorus of the trisubstituted ring was found as a
singlet at -28.6 ppm, and as was the case of the mononuclear
palladium complex 3a, the chemical shift corresponding to
P4 was not affected by the coordination to metal and a singlet
was detected at -18.7 ppm.
With the view to get a dinuclear nickel complex analogous
to 3b, several experiments were conducted both in chlori-
nated solvents and in DMSO. The formation of such complex
would either shift to lower fields the signals found around
a
The postulated dinuclear 4b could not be formed despite the use of
large excess of nickel halide precursors and long-time reactions.
-20 to -30 ppm in 4a or make it disappear (tetrahedral
paramagnetic species). Curiously, it was not possible to
induce or even detect any further coordination when a large
excess of NiBr2(DME) or NiCl2(DME) was added to 4a. The
addition of Pd(PhCN)2Cl2 to 4a, to form a hetero dinuclear
complex, was also unsuccessful. For these reasons, in Scheme
3 (compared to Scheme 2), we suggested a first coordination
of nickel between P2 and P3 (in solid state d(P2···P3) ) 3.336
Å), which might make a subsequent coordination of a metal
center between P1 and P4 (d(P1···P4) ) 6.335 Å)18 more
difficult. Small differences were noted on samples run with
excess nickel several days in DMSO, as shown in Figure 7,
broadening of the signals appeared for the bonded phospho-
rus, possibly paramagnetic in origin. This is attributed to the
effect of bonded DMSO molecules because DMSO-solvated
nickel(II) complexes are known,19 as well as paramagnetic
five-coordinated nickel (II) halide phosphine complexes.20
This preference of nickel complexes for a 1,2-P coordina-
tion (in a square planar environment) on a 1,1′-P coordination
(tetrahedral geometry) was confirmed by the study of the
Figure 4. 31P NMR spectrum of the dinuclear palladium complex 3b.
Figure 5. Plot of dinuclear palladium complex 3b (30% probability
ellipsoids). Hydrogen atoms are omitted for clarity. Selected bond distances
(Å) and angles (deg): Fe-CNT(1) ) 1.627, Fe-CNT(2) ) 1.654. P1-Pd1
) 2.207(4), P2-Pd1 ) 2.247(5), P3-Pd2 ) 2.304(5), P4-Pd2 ) 2.278(6),
Cl1-Pd1 ) 2.355(5), Cl2-Pd1 ) 2.383(4), Cl3-Pd2 ) 2.336(5), Cl4-Pd2
) 2.343(5); P1-Pd1-P2 ) 88.88(18), Cl1-Pd1-Cl2 ) 93.12(17),
P3-Pd2-P4 ) 102.38(19), Cl3-Pd2-Cl4 ) 87.64(19), Cl2-Pd1-P1 )
175.47(18), Cl2-Pd1-P2 ) 87.72(17), Cl1-Pd1-P2 ) 172.31(19),
Cl1-Pd1-P1 ) 90.65(17), Cl4-Pd2-P3 ) 174.3(2), Cl4-Pd2-P4 )
83.3(2), Cl3-Pd2-P4 ) 166.62(19), Cl3-Pd2-P3 ) 86.94(19).
Figure 6. 31P NMR spectrum of the mononuclear nickel complex in CDCl3,
4a.
Inorganic Chemistry, Vol. 47, No. 5, 2008 1613

Thomas et al.
Figure 7. 31P NMR spectrum of the complex obtained in DMSO
[4a(DMSO)x].
Scheme 4. Synthesis of the Ligand 5 and Its Corresponding Nickel
Dichloride Complex 6
Figure 9. Plot of nickel complex 6. Hydrogen atoms are omitted for clarity.
Selected bond distances (Å) and angles (deg): Fe-CNT(1) ) 1.668,
Fe-CNT(2) ) 1.659, P1-Ni1 ) 2.1483(15), P2-Ni1 ) 2.1554(12),
Cl1-Ni1 ) 2.1967(12), Cl2-Ni1 ) 2.1949(15), P1-Ni1-P2 ) 90.20(5),
Cl1-Ni1-Cl2 ) 93.72(4), Cl2-Ni1-P1 ) 177.73(5), Cl2-Ni1-P2 )
87.53(4), Cl1-Ni1-P2 ) 177.86(5), Cl1-Ni1-P1 ) 88.55(5).
Table 3. Crystal, Collection, and Refinement Parameters for
[6·2(CHCl3)]
coordination chemistry of a parent triphosphine ligand 5. This
triphosphine was also expected to allow a certain rotational
flexibility and thus to potentially lead to 1,1′- or 1,2-P
different coordination modes. The synthetic pathway for the
synthesis of 5, which is inherently different from synthesis
for 1, is displayed in Scheme 4.4b
The coordination chemistry of 5 toward nickel halides
(NiCl2DME and NiBr2DME) preferentially lead to the
formation of the chelate complex 6. The flexibility of the
ferrocene backbone was somewhat confirmed by the 31P
NMR spectrum obtained at ambient temperature for 6, which
displays two broad peaks at -20.1 and 36.2 ppm in CDCl3
with a relative integration of 1:2 respectively (Figure 8); in
the low-temperature NMR spectrum, the signal broadness
decreases as the complex lability is slowed, leading to two
fine singlets with no phosphorus spin-spin coupling.
Single crystals of 6 solvated with chloroform were
obtained, from which an X-ray diffraction study was carried
out that consistently, with 31P NMR, establishes the 1,2-P
[6·2(CHCl3)]
empirical formula
fw
temp
wavelength
cryst syst
space group
unit cell dimensions
C52H47Cl8P3FeNi
1162.97
160(2) K
0.71073 Å
Monoclinic
P21/c
a ) 10.00317(8) Å, R ) 90°
b ) 23.3697(2) Å, ꢀ ) 98.943(7)°
c ) 22.4018(6) Å, γ ) 90°
5173.9(3) Å3
vol
Z
4
density (calcd)
abs coeff
F(000)
θ range for data collection
index range
1.493 Mg/m3
-1
1.186 mm
2376
2.96–30.52°
-13 e h e 13
-31 e k e 32
-20 e l e 31
reflns collected
independent reflns
completeness to 2θ ) 61.04°
refinement method
45 905
14 596 [Rint ) 0.0987]
92.5%
full-matrix least-squares on F2
14 596/14/616
data/restraints/params
GOF on F2
0.806
Final R indices [F2 > 2σ(F2)]
R indices (all data)
largest diff. peak and hole
R1 ) 0.0561, wR2 ) 0.0590
R1 ) 0.1696, wR2 ) 0.0772
0.499 and -0.558 e Å-3
bonding preference of the nickel atom (Figure 9 ). The nickel
center lies in a fairly regular square planar environment for
which P1-Ni1-P2 ) 90.2°, Cl1-Ni1-Cl2 ) 93.7°,
Cl2-Ni1-P2 ) 87.5°, and Cl1-Ni1-P1 ) 88.6°; the Ni-X
(X ) Cl, P) bonds range from 2.15 to 2.19 Å. The analogous
palladium chloride complex 7 was also obtained; it displays
(18) The difference of coordination behavior of the ligand depending on
the metal might be then partly attributed to the relative atomic size of
nickel and palladium in the decisive first 1,2-P chelation. Obviously,
this hypothesis assumes that only small changes exist between solid-
state structural features and solution behavior that should be further
substantiated.
Figure 8. 31P NMR spectrum of the complex 6 in CDCl3 at RT.
1614 Inorganic Chemistry, Vol. 47, No. 5, 2008

Coordination Chemistry of Ferrocenyl Polyphosphines
in 31P NMR two singlets at -20.8 and 45.3 ppm with a
relative integration of 1:2 respectively. In this case, 1,2-P-
bonding was exclusively observed.
conformational features of tetraphosphine 1 gave rise to an
unprecedented coordination framework in a metallocene
coordination complex (3b), in which both coexist 1,1′ and
1,2 chelating P-bondings. The preference of nickel complexes
for a 1,2-P coordination (in a square planar environment)
on a 1,1′-P coordination was confirmed by the study of the
coordination chemistry of the triphosphine ligand 5. This
study finally confirmed that in these ferrocenyl polyphosphine
derivatives the correlation of solid state structures and
solution multinuclear NMR is very profitable for the
understanding of both their spectroscopic and coordination
properties.
Summary and Conclusion
Ferrocenyl diphosphines are very well-known and useful
compounds; conversely, only a few tri and tetraphosphines
based on a ferrocene backbone have been developed to date.
The present study details the two different synthetic routes
employed to form the original 1,1′,2,3-tetrakis(diphenylphos-
phino)ferrocene (1) and 1,1′,2-tris(diphenylphosphino)-4-tert-
butylferrocene (5) metallo-ligands. These species originally
produced to observe possible intense 31P31P spin–spin
couplings through-space originating from phosphorus atoms
proximity in solution were shown to be too flexible (i.e.,
not constrained) for that purpose. The coordination behavior
of the tetraphosphine 1 toward palladium and nickel was
found markedly different since both dinuclear and mono-
nuclear palladium complexes were obtained with palladium
([Pd2Cl4(1)], 3b and 3a), while only mononuclear nickel
compounds were produced from nickel ([NiCl2(5)], 6). A
reasonable explanation for that is based on a different
regioselectivity of the first metal coordination, favoring or
hampering the subsequent second metal coordination. The
Acknowledgment. We thank B. Donnadieu for X-ray
diffraction studies of nickel complex 6 and S. Royer and G.
Delmas for technical support. We thank The “Conseil
regional de Bourgogne” and CNRS for a postdoctoral grant
to V.V.I. The cooperation program “Alliance” is also thanked
for financial support of a France-U.K. collaboration between
the Université de Bourgogne (Dijon) and the University of
Wales (Bangor).
Supporting Information Available: Crystal details and data for
1, 3b, and 6, conformation views of 1, and typical 13C, 1H, and 31
P
NMR spectrum. This material is available free of charge via the
Internet at http://pubs.acs.org.
(19) Kristiansson, O.; Persson, I.; Bobicz, D.; Xu, D. Inorg. Chim. Acta
2003, 344, 15.
(20) Hou, J.; Sun, W.-H.; Zhang, S.; Ma, H.; Deng, Y.; Lu, X. Organo-
metallics 2006, 25, 236.
IC7022105
Inorganic Chemistry, Vol. 47, No. 5, 2008 1615