1H,1H,2H,2H-Perfluoroalkyl-functionalization of Ni(II), Pd(II), and Pt(II) mono- and diphosphine complexes: Minimizing the electronic consequences for the metal center

Article abstract of DOI:10.1021/ic025770l

A series of fluorous derivatives of group 10 complexes MCl₂(dppe) and [M(dppe)₂](BF₄)₂ (M = Ni, Pd or Pt; dppe = 1,2-bis(diphenylphosphino)ethane) and cis-PtCl₂(PPh₃)₂ was synthe

Full text of DOI:10.1021/ic025770l

Inorg. Chem. 2003, 42, 2115−2124
1H,1H,2H,2H-Perfluoroalkyl-Functionalization of Ni(II), Pd(II), and Pt(II)
Mono- and Diphosphine Complexes: Minimizing the Electronic
Consequences for the Metal Center
Elwin de Wolf,† Ad J. M. Mens,‡ Onno L. J. Gijzeman,‡ Joop H. van Lenthe,§
,
Leonardus W. Jenneskens,| Berth-Jan Deelman,* and Gerard van Koten†
Department of Metal-Mediated Synthesis, Debye Institute, Utrecht UniVersity, Padualaan 8,
3584 CH Utrecht, The Netherlands, Department of Inorganic Chemistry, Debye Institute,
Utrecht UniVersity, Sorbonnelaan 16, 3584 CA Utrecht, The Netherlands, Department of
Theoretical Chemistry, Debye Institute, Utrecht UniVersity, Padualaan 8, 3584 CH Utrecht,
The Netherlands, Department of Physical Organic Chemistry, Debye Institute, Utrecht UniVersity,
Padualaan 8, 3584 CH Utrecht, The Netherlands, and ATOFINA Vlissingen B.V., P.O. Box 70,
4380 AB Vlissingen, The Netherlands
Received June 5, 2002
A series of fluorous derivatives of group 10 complexes MCl2(dppe) and [M(dppe)2](BF ) (M ) Ni, Pd or Pt; dppe
4 2
) 1,2-bis(diphenylphosphino)ethane) and cis-PtCl2(PPh3)2 was synthesized. The influence of para-(1H,1H,2H,2H-
perfluoroalkyl)dimethylsilyl-functionalization of the phosphine phenyl groups of these complexes, as studied by
NMR spectroscopy, cyclovoltammetry (CV), XPS analyses, as well as DFT calculations, points to a weak steric
and no significant inductive electronic effect. The steric effect is most pronounced for M ) Ni and leads in the case
of NiCl2(1c) (3c) and [Ni(1c)2](BF ) (7c) (1c ) [CH P{C H (SiMe2CH CH C F )-4}2]2) to a tetrahedral distortion
4 2
2
6
4
2
2 6 13
from the expected square planar geometry. The solubility behavior of NiCl2[CH P{C H (SiMe3-b(CH CH C F2x+1)b)-
2
6
4
2
2 x
4}2]2 (3: b ) 1−3; x ) 6, 8) in THF, toluene, and c-C F CF was found to follow the same trends as those
6
11
3
observed for the free fluorous ligands 1. A similar correlation between the partition coefficient (P) of complexes 3
and free 1 was observed in fluorous biphasic solvent systems, with a maximum value obtained for 3f (b ) 3, x )
6, P ) 23 in favor of the fluorous phase).
Introduction
perfluoroalkyl groups are usually attached to the ligand
system, according to the like-dissolVes-like principle. Since
perfluoroalkyl groups are bulky3 and highly electron with-
drawing,4 perfluorotail-functionalization of the ligand system
is often not trivial, because it can influence the properties
of the ligand, which can result in altered reactivity and
selectivity. Therefore, several insulating groups have been
used to minimize the direct inductive electronic effect that
a perfluoroalkyl tail can exert on the metal center.5 However,
small electron-withdrawing effects were still observed for
most spacers, e.g., as reflected in a change of νCO in the IR
spectrum of trans-[MCl(CO)(L)2] (M ) Rh, Ir; L )
Preferential solubility of a transition metal complex in
perfluorocarbon solvents is of interest because it allows its
separation from nonfluorous components, a feature especially
interesting for the recycling of homogeneous catalysts.1,2 To
achieve preferential fluorous phase solubility, one or more
* To whom correspondence should be addressed. E-mail: berth-jan.
deelman@atofina.com.
† Department of Metal-Mediated Synthesis, Debye Institute.
‡ Department of Inorganic Chemistry, Debye Institute.
§ Department of Theoretical Chemistry, Debye Institute.
| Department of Physical Organic Chemistry, Debye Institute.
ATOFINA Vlissingen B.V.
(1) Cornils, B.; Herrmann, W. A. In Applied Homogeneous Catalysis with
Organometallic Compounds; Cornils, B., Herrmann, W. A., Eds.; VCH
Verlagsgesellshaft mbH, Weinheim, 1996; Chapter 3.1, p 575.
(2) (a) de Wolf, E.; van Koten, G.; Deelman, B.-J. Chem. Soc. ReV. 1999,
28, 37. (b) Hope, E. G.; Stuart, A. M. J. Fluorine Chem. 1999, 100,
75. (c) Horva´th, I. T. Acc. Chem. Res. 1998, 31, 641. (d) Curran, D.
P. Angew. Chem., Int. Ed. 1998, 37, 1174. (e) Cornils, B. Angew.
Chem., Int. Ed. Engl. 1997, 36, 2057.
(3) Kloubek, J. Colloids Surf. 1991, 55, 191.
(4) Huheey, J. E. J. Phys. Chem. 1965, 69, 3284.
(5) (a) Horva´th, I. T.; Ra´bai, J. Science 1994, 266, 72. (b) Smith, D. C.;
Stevens, E. D.; Nolan, S. P. Inorg. Chem. 1999, 38, 5277. (c)
Bhattacharyya, P.; Croxtall, B.; Fawcett, J.; Fawcett, J.; Gudmunsen,
D.; Hope, E. G.; Kemmitt, R. D. W.; Paige, D. R.; Russell, D. R.;
Stuart, A. M.; Wood, D. R. W. J. Fluorine Chem. 2000, 101, 247.
10.1021/ic025770l CCC: $25.00 © 2003 American Chemical Society
Published on Web 02/22/2003
Inorganic Chemistry, Vol. 42, No. 6, 2003 2115

de Wolf et al.
monodentate (fluorous) phosphine).6 Only a few studies were
reported in which the influence of a perfluoroalkyl group
on the metal center of transition metal complexes was
investigated in more detail, using electrochemical tech-
niques,7 density functional theory (DFT) calculations8 and
X-ray photoelectron spectroscopy (XPS).9
Scheme 1
We recently reported on several fluorous derivatives of
1,2-bis(diphenylphosphino)ethane (dppe) (1a and 1c-f)10
and fluorous monophosphines such as 2b11 containing a
-SiMe3-b(CH2CH2-)b-spacer. The metal (predominantly
Rh(I)) complexes containing these fluorous ligands were used
as homogeneous catalysts for the hydrogenation of 1-alkenes
and 4-octyne.12 Interestingly, these catalysts displayed similar
activity compared to their nonfluorous counterparts.
was used as a reference compound, two new ligands
containing a para-SiMe2CH2CH2CF3 (2a) and a para-
SiMe2C8H17 (1b) substituent, respectively, were prepared
which allow the separate study of electronic (2a) and steric
properties (1b) of a para-SiMe2R substituent. Also, the
solubility behavior and partitioning of the various fluorous
nickel complexes NiCl2(1) in a fluorous biphasic solvent
system are presented.
Results
Synthesis. The neutral fluorous complexes MCl2(1)
(M ) Ni (3), Pd (4), Pt (5)), cis-PtCl2(2) (6), and the
dicationic complexes [M(1)2](BF4)2 (M ) Ni (7), Pd (8))
were synthesized analogously to published procedures13,14
with some modifications (see Experimental Section) (Scheme
1). In the case of 3d and 3f, the diphosphines (1d and 1f)
were added as solution in CF3C6H5 (BTF). For the prepara-
tion of highly fluorous 3f, also a BTF/MeOH solvent mix-
ture had to be used to dissolve the starting compound
[Ni(H2O)6]Cl2 as well as to prevent immediate precipita-
tion of the reactants and (side)-products. Similarly, 4f was
prepared in a fluorous biphasic system consisting of
CH2Cl2 and c-C6F11CF3 (PFMCH).
In this paper, a quantitative investigation of the effects of
para-(1H,1H,2H,2H-perfluoroalkyl)silyl-functionalization of
the phosphine phenyl groups on the steric and electronic
properties of d8-metal(II) complexes is presented using the
new transition metal complexes MCl2(1) and [M(1)2](BF4)2
(M ) Ni, Pd or Pt) and cis-PtCl2(2) (Scheme 1).12b Apart
from the previously reported para-SiMe3 derivative 1a, which
(6) (a) Alvey, L. J.; Meier, R.; Soo´s, T.; Bernatis, P.; Gladysz, J. A. Eur.
J. Inorg. Chem. 2000, 1975. (b) Klose, A.; Gladysz, J. Tetrahedron:
Asymmetry 1999, 10, 2665. (c) Smith, D. C.; Stevens, E. D.; Nolan,
S. P. Inorg. Chem. 1999, 38, 5277. Bhattachrayya, P.; Corxtall, B.;
Fawcett, J.; Facett, J.; Gudmunsen, D.; Hope, E. G.; Kemmitt, R. D.
W.; Paige, D. R.; Russell, D. R.; Stuart, A. M.; Wood, D. R. W. J.
Fluorine Chem. 2000, 101, 247.
(7) Hughes, R. P.; Trujillo, H. A. Organometallics 1996, 15, 286.
(8) (a) Banet Osuma, A. M.; Chen, W.; Hope, E. G.; Kemmitt, R. D. W.;
Paige, D. R.; Stuart, A. M.; Xiao, J.; Xu, L. J. Chem. Soc., Dalton
Trans. 2000, 4052. (b) Szla´vik, Z.; Ta´rkay´i, G.; Go¨mo¨ry, A.; Tarczay,
G.; Ra´bai, J. J. Fluorine Chem. 2001, 108, 7.
(9) (a) Goll, J. G.; Moore, K. T.; Ghosh, A.; Therien, M. J. J. Am. Chem.
Soc. 1996, 118, 8344. (b) Moore, K. T.; Fletcher, J. T.; Therien, M.
J. J. Am. Chem. Soc. 1999, 121, 5196.
(10) de Wolf, E.; Richter, B.; Deelman, B.-J.; van Koten, G. J. Org. Chem.
2000, 65, 5424.
1
All complexes were characterized by H and 31P NMR
spectroscopy as well as elemental analyses. In addition,
3c-d, 4c, 5c, 7c, and 8c were characterized by 19F NMR
spectroscopy. The 1H NMR spectra of complexes containing
the highly fluorous ligand 1e or 1f displayed broad peaks
(multiplicities were not resolved).15 All 31P and 195Pt NMR
1
chemical shifts and JPtP coupling constants of the fluorous
compounds are close to the values measured for complexes
containing the corresponding nonsubstituted as well as
trimethylsilyl-substituted ligands (Table 1 and Experimental
Section) indicating similar (cis-)square planar structures for
(11) Richter, B.; de Wolf, E.; van Koten, G.; Deelman, B.-J. J. Org. Chem.
2000, 65, 3885.
(12) (a) de Wolf, E.; Spek, A. L.; Kuipers, B. W. M.; Philipse, A. P.;
Meeldijk, J. D.; Bomans, P. H. H.; Frederik, P. M.; Deelman, B.-J.;
van Koten, G. Tetrahedron 2002, 58, 3911. (b) Richter, B.; de Wolf,
A. C. A.; van Koten, G.; Deelman, B.-J. (ATOFINA) PCT Int. Appl.
WO 0018444, 2000. (c) Richter, B.; Spek, A. L.; van Koten, G.;
Deelman, B.-J. J. Am. Chem. Soc. 2000, 122, 3945. (d) van den Broeke,
L. J. P.; Goetheer, E. L. V.; Verkerk, A. W.; de Wolf, E.; Deelman,
B.-J.; van Koten, G.; Keurentjes, J. T. F. Angew. Chem., Int. Ed. 2001,
40, 4473.
(13) (a) Booth, G.; Chatt, J. J. Chem. Soc. 1965, 3238. (b) Broadwood-
Strong, G. T. L.; Chaloner, P. A.; Hitchcock, P. B. Polyhedron 1993,
12, 721. (c) Drew, D.; Doyle, J. R.; Shaver, A. G. Inorg. Synth. 1972,
13, 47.
(14) Miedaner, A.; Curtis Haltiwanger, R.; DuBois, D. L. Inorg. Chem.
1991, 30, 417.
(15) No significant changes in line broadening were observed upon heating
to 50 °C.
2116 Inorganic Chemistry, Vol. 42, No. 6, 2003

Perfluoroalkyl-Functionalized Metal Complexes
Table 1. 31P NMR Data of PtCl2(1) and cis-PtCl2(2) and Their
Nonsubstituted Analogues
further, whereas the solubility in toluene decreased and no
solubility in PFMCH could be detected. This behavior has
been observed for the free ligand (1c) as well.10 When the
number of perfluoroalkyl tails was further increased from 4
to 8 or 12 per complex (e.g., 3e or 3f), both the solubility in
THF and toluene decreased, and now, most importantly, the
solubility in PFMCH increased dramatically. Similar trends
in solubility have been observed previously for the free
ligands 1a,c,e-f.10
Not withstanding the described general trends, a large
difference in the absolute solubility of free 1d and 3d was
observed. Whereas 1d was hardly soluble in THF, toluene,
and PFMCH, its nickel complex 3d displayed high solubility
in THF, moderate solubility in toluene, and limited solubility
in PFMCH. By considering both the solubility behavior and
the weight percentage of fluorine of the various complexes,
the fluorous character of 3d can be placed between that of
3c and 3e. This is corroborated by the earlier observation
that an increase of the number of perfluoroalkyl tails is more
effective for obtaining good solubility in fluorous solvents
than an increase of the length of the tail.10,11,12c Although
not studied quantitatively, similar solubility behavior was
observed for the fluorous complexes 4-8, i.e., high solubility
in polar organic solvents and insolubility in fluorous solvents
for complexes containing 1a, and low solubility in organic
solvents accompanied with high solubility in fluorous
solvents for complexes containing 1e or 1f.
compd
δ (31P) (ppm)a
1JPtP (Hz)
δ (195Pt) (ppm)a
PtCl2(dppe)b
40.9
41.8
41.9
42.0d
15.2
14.1
14.1
3618
3612
3604
3593
3666
3665
3663
-4572
-4573
-4570
e
-4431
-4418
-4420
5ac
5c
5e
cis-PtCl2(PPh3)2
6a
6b
a In CDCl3 unless noted otherwise. b 31P NMR data in (CD3)2CO: δ 42.8
(1JPtP ) 3594 Hz). c 31P NMR data in (CD3)2CO: δ 47.4 (1JPtP ) 3588
Hz). d In CDCl3/CF3C6H5 1:1 (v/v). e Not soluble in CDCl3.
Figure 1. van’t Hoff plot of variable temperature 31P NMR measurements
on 7c.
The most important parameter for recycling of the
complexes by fluorous biphasic separation techniques is their
partition coefficient P in fluorous biphasic solvent combina-
tions (P ) cfluorous phase/corganic phase). The partition coefficients
of 3 and those of free 1 parallel the solubility properties (vide
supra); i.e. an increase of the weight percentage of fluorine
in the ligand or complex results in a higher affinity for the
fluorous phase. As observed for the free fluorous ligand, a
maximum partition coefficient was observed for complex 3f
with the highest fluorine content (61.8 wt % F; P ) 23).
Although its fluorophilicity17 is still modest, it makes 3f the
most interesting candidate as an alternative for NiCl2(dppe)
in fluorous biphasic separation strategies. Ligand 1f was also
recently employed in rhodium-catalyzed hydrogenation
where recycling efficiencies better than 99.9% have been
obtained.12b
Electronic and Conformational Study. The 31P NMR
chemical shift of platinum phosphine complexes can be
unreliable as an indicator for electronic or steric influences
exerted by the ligands.18 The 1JPtP coupling constant, which
is also easily obtained, is a more reliable probe for quantify-
ing such influences.19 In particular, the 1JPtP coupling constant
is known to correlate well with the Hammett constant of the
para-substituent on the aryl ring20 and in the case of
the fluorous compounds and their nonfluorous analogues.
The only exceptions were the ionic bis(diphosphine) nickel
complexes 7; whereas a single resonance in the 31P NMR
spectra of [Ni(dppe)2](BF4)2 and 7a was observed, two singlet
resonances were detected for 7c at δ ) 54.9 and 55.0 ppm
(3:1 ratio),16 probably indicating that at least two different
isomers of 7c are present in solution (vide infra). The 3:1
ratio increased upon increasing the temperature. From the
different ratios of these two peaks in 31P NMR obtained upon
heating gradually from 25 to 80 °C, the thermodynamic
parameters (∆H° ) 2.2 kJ/mol and ∆S° ) 1.7 J/K mol) were
obtained from a van’t Hoff plot (Figure 1) which also
confirmed that the two isomers of 7c are indeed in equilib-
rium. The modest value of ∆H° also suggests that the two
isomers are structurally closely related (vide infra).
Solubility and Partitioning. For immobilization and
fluorous phase extraction of fluorous complexes in a fluorous
phase, their solubility characteristics and partitioning behavior
in a fluorous biphasic system are important properties. To
this end, the solubility and partitioning of NiCl2(dppe) and
complexes 3 in a biphasic toluene/PFMCH (1:1) solvent
mixture were studied (Table 2). It was found that the
solubility of NiCl2(dppe) in THF and toluene increased by
an order of magnitude upon para-trimethylsilyl-functional-
ization of the dppe ligand (3a). Interestingly, when 4 para-
SiMe2CH2CH2C6F13 groups were attached (i.e., as in 3c), the
solubility of the nickel complex in THF increased even
(17) Garelli, N.; Vierling, P.; Fischel, J. L.; Milano, G. Eur. J. Med. Chem.
1993, 28, 235.
(18) Atherton M. J.; Fawcett, J.; Hill, A. P.; Holloway, J. H.; Hope, E. G.;
Russell, D. R.; Saunders: G. C.; Stead, R. M. J. J. Chem. Soc., Dalton
Trans. 1997, 1137.
(16) Despite several attempts to grow single crystals of 7c, no crystals
suitable for X-ray crystal structure determination were obtained, which
might possibly be caused by disorder and the presence of more than
one structure in the solid state.
(19) (a) Pregosin, P. S. Coord. Chem. ReV. 1982, 44, 247. (b) Goel, R. G.
Inorg. Nucl. Chem. Lett. 1979, 15, 437.
(20) Cobley, C. J.; Pringle, P. G. Inorg. Chim. Acta 1997, 265, 107.
Inorganic Chemistry, Vol. 42, No. 6, 2003 2117

de Wolf et al.
Table 2. Solubility and Partition Coefficients of NiCl2(dppe) and NiCl2[{CH2P(C6H4-SiMe3-b(CH2CH2CxF2x+1)b)2}2]
solubilitya
THF
toluene
(mol/L)
PFMCH
compd
x
b
(g/L)
(mol/L)
(g/L)
(g/L)
(mol/L)
Pb
NiCl2(dppe)
3
130
816
212
91
0.008
0.154
0.381
0.083
0.026
0.002
<0.2
96
28
8
5
<0.0005
0.114
0.013
0.003
0.001
c
c
c
c
c
0.0
1.6
3.7
8
3a
3c
3d
3e
3f
0
1
1
2
3
6
8
6
6
<0.2
4
124
>400d
<0.00009
0.002
0.036
9
<0.2
<0.00004
>0.083d
23
a Expressed as the amount of phosphine which can be dissolved in 1 L of pure solvent at 25 °C. Determined by gravimetric methods (PFMCH )
perfluorocyclohexane). b In a 1:1 (v/v) mixture of PFMCH and toluene at 0 °C (P ) cfluorous phase/corganic phase). Determined by gravimetric methods. The
estimated error is (1 in the last digit. c Not determined. d Compound 3f is completely miscible with PFMCH.
Table 3. CV Dataa for p-Silyl-Substituted Derivatives of
diphosphine ligands also with the P-Pt-P bite angle.21 In
[M(dppe)2](BF4)2 (M ) Ni, Pd)
addition, the 1JPtP value may be sensitive to the presence of
bulky groups, in analogy to observations for the JSeP
coupling constant.19,22
E1/2
Eonset
E1/2
Eonset
1
compd
(-1/-2)b
(-1/-2)c
(0/-1)b
(0/-1)b
[Ni(dppe)2](BF4)2
-0.73 (80)
-0.56
-0.55
-0.56
-0.31
-0.93 (84)
-0.69
-0.66
-0.72
-0.51
The para-trimethylsilyl substituted phosphine derivative
7a
7b
7c
d
d
d
-0.94 (90)
1
-d
d
of PtCl2(dppe) (5a) (Table 1) shows a slightly lower JPtP
coupling constants compared to its nonsubstituted derivative.
1
A further small decrease in the JPtP coupling constant was
E
1/2 (0/-2) Eonset (2/1)c
E
1/2 (1/0)b
E
onset (1/0)b
found for fluorous 5c and 5e containing 4 or 8 perfluoroalkyl
tails, respectively, although an effect of the different solvent
needed to dissolve 5e cannot be excluded. A similar trend
was observed when the 1JPtP coupling constants of nonfluo-
rous cis-PtCl2(PPh3)2 and fluorous 6a-b were compared
which may well correspond to a progressively increasing
electron-withdrawing character of the para-silyl-substituent.
It is clear that the effect is modest, however.
[Pd(dppe)2](BF4)2 -1.04 (35)e
f
d
d
8a
8c
d
d
-0.58
-0.81
-0.99
-1.07
a All peaks (reported in V) are referenced to the ferrocene/ferrocinium
couple in MeCN/CH2Cl2 1:1. Voltammograms were measured at 50 mV/s.
b Numbers in brackets indicate the peak to peak separation (∆E) at 50 mV/s
in mV. c Similar values were obtained for scan rates of 20 and 100 mV/s.
d Irreversible reduction. e Compound is insoluble in MeCN/CH2Cl2 1:1. Data
from ref 14 in DMSO, Eonset not given. f Not observed.
Also, the metal chemical shifts of transition metal com-
plexes are known to be very sensitive probes for changes in
the steric and electronic environment.23 Considering the broad
chemical shift window of 195Pt NMR and the high sensitivity
for small structural or electronic variations, the nearly
identical 195Pt NMR chemical shifts found for both the
diphosphine complexes 5a,c,e and bis(monophosphine)
complexes 6a-b compared to their nonsubstituted analogues
further demonstrate that para-trimethylsilyl- or para-
(1H,1H,2H,2H-perfluoroalkyl)dimethylsilyl-functionaliza-
tion does not significantly affect the electronic and/or
geometric properties of the platinum metal center.
As a further technique to study the electronic environment
of the metal center, the electrochemical behavior of the nickel
and palladium complexes 7 and 8 was investigated by cyclic
voltammetry (CV) (Table 3). Two reduction waves were
observed for all para-silyl-substituted nickel complexes 7,
similar to what was reported for [Ni(dppe)2](BF4)2, which
is again indicative of the closely related electronic properties
of the metal centers.14 However, whereas the first reduction
wave of [Ni(dppe)2]2+ to [Ni(dppe)2]+ is quasireversible at
high scan rates,14 an irreversible first reduction was observed
for the para-silyl-substituted derivatives. The second reduc-
tion to the corresponding Ni(0)-species is reversible only for
the nonsubstituted and trimethylsilyl-substituted compounds.
For 7b-c also the second reduction was irreversible, showing
that apparently attachment of a large group rather than
specific attachment of a perfluoroalkyl tail affects the
reversibility. Possibly, the size of the molecule affects the
reversibility of the reduction process, although subsequent
reaction of the formed Ni(I)-radical with a functional group
(e.g., C-Si, C-F) cannot be excluded. Although CV is often
used to monitor electronic effects of substituents, the
(ir)reversibility of the reductions studied also depends on
the chemical reactivity of the complexes generated. There-
fore, the different reversibility of the reductions of the silyl-
substituted complexes compared to their nonsubstituted
analogues is less important for the study of the electronic
and steric influence of a (fluorous) silyl-substituent in the
starting complexes.
Since the onset values of the reduction waves were found
to be independent of experimental conditions (scan rate,
potential window, etc.), the onset reduction potentials can
be used for comparison of the substituted fluorous and
nonfluorous compounds with their nonsubstituted parent
complex. All nickel complexes, except 7c, displayed similar
onset values for both the first and the second reduction wave,
respectively. The less negative reduction potentials found
for 7c show that the fluorous complex is more easily reduced,
which could correspond to a lower electron density on the
metal center.
(21) Anderson, G. K.; Davies, J. A.; Schoeck, D. L. Inorg. Chim. Acta
1983, 76, L251.
(22) Allen, D. W.; Taylor, B. F. J. Chem. Soc., Dalton Trans. 1982, 51.
(23) (a) Loggin, P. L.; Goodfellow, R. J.; Haddock, S. R.; Taylor, B. F.;
Marshall, I. R. H. J. Chem. Soc., Dalton Trans. 1976, 459. (b) Pregosin,
P. S. Coord. Chem. ReV. 1982, 44, 247. (c) Hope, E. G.; Levason,
W.; Powell, N. A. Inorg. Chim. Acta 1986, 115, 187.
2118 Inorganic Chemistry, Vol. 42, No. 6, 2003

Perfluoroalkyl-Functionalized Metal Complexes
Figure 2. XPS spectra of NiCl2(dppe) (9), 3a ([), and 3c (1). Spectra were referenced to C1s ) 285.0 eV. The intensities of the 2p3/2 peaks were scaled
to 1000 au.
Table 4. XPS Dataa of PdCl2(1) (4), [Pd(1)2](BF4)2 (8), PtCl2(1) (5),
Two reduction processes were observed in the cyclic
and Their Nonsubstituted Analogues
voltammograms of both 8a and 8c, whereas only a single
two-electron reversible reduction wave to [Pd(dppe)2]0 was
binding energy (eV)
compd
P2p
Pd3d
Pd3d
3/2
reported for [Pd(dppe)2]2+.14 Furthermore, these two reduc-
tion waves observed for 8a and 8c turned out to be
irreversible, in constrast to the reduction behavior of
[Pd(dppe)2]2+. The completely different reduction behavior
of the para-silyl-substituted palladium complexes 8a and 8c
is most probably caused by different reduction kinetics or a
different reactivity of the species resulting from their
reduction compared to the species resulting from reduction
of their nonsubstituted analogue [Pd(dppe)2]2+ (vide supra).
To further monitor the possible occurrence of an electron-
withdrawing effect of the perfluoroalkyl groups on the metal
center,24 XPS measurements were performed on 4, 5, and 8
(Table 4) and on 3 and 7 (Table 5).25,26 Very similar metal
as well as phosphorus electron binding energies were found
for the complexes containing nonsubstituted dppe, para-
trimethylsilyl-substituted 1a, and para-(1H,1H,2H,2H-per-
fluorooctyl)dimethylsilyl-substituted 1c, respectively.
In the XPS spectra of nickel complexes, both the observed
electron binding energies and the shape of the signals are
important, because of the possible appearance of shake-up
satellites.27 The XPS spectra for nonfluorous NiCl2(dppe) and
3a are very similar (Table 5 and Figure 2). This observation
5/2
PdCl2(dppe)
132.0
132.1
132.2
131.9
132.3
132.3
338.4
338.4
338.2
338.7
338.7
338.7
343.6
343.6
343.4
343.9
344.0
344.0
4a
4cb
[Pd(dppe)2](BF4)2
8a
8cb
P2p
Pt
Pt
4f7/2
4f5/2
PtCl2(dppe)
132.0
132.3
132.3
73.3
73.6
73.4
76.5
76.8
76.3
5a
5cb
a Peaks referenced to C1s ) 285.0 eV. b Extra C1s peak observed at 292.0
eV for (CF2)x.
Table 5. XPS Dataa of NiCl2(1) (3), [Ni(1)2](BF4)2 (7), and Their
Nonsubstituted Analogues
binding energy (eV)
b
compd
P2p
Ni2p
SAT2p
Ni2p
SAT2p
1/2
3/2
3/2
1/2
NiCl2(dppe)
132.0 854.9 (2.7)
132.1 854.5 (3.5)
132.8 855.8 (8.2)d
862.1
861.6
872.5
872.7
874.0
872.6
872.3
873.3
880.4
880.5
881.0
878.6
878.1
881.0
3a
3cc
862.3d
858.4
e
[Ni(dppe)2](BF4)2 132.0 855.4 (2.2)
7a
131.8 855.1 (2.9)
7cc
132.2 854.6 (6.8)d
861.4d
a Peaks referenced to C1s ) 285.0 eV. b Full width at half-maximum
given in parentheses. c Extra C1s peak observed at 292.0 eV for (CF2)n.
d Values obtained after deconvolution of overlapping 2p3/2 peak and its
satellite into two peaks. e Not observed.
(24) (a) Leigh, G. J.; Bremser, W. J. Chem. Soc., Dalton Trans. 1972, 1216.
(b) Andreocci, M. V.; Mattogno, G.; Zanoni, R.; Giannoccaro, P.;
Vasapollo, G. Inorg. Chim. Acta 1982, 63, 225. (c) Peuckert, M.; Keim,
W.; Storp, S.; Weber, R. S. J. Mol. Catal. 1983, 20, 115. (d) Gassman,
P. G.; Winter, C. H. J. Am. Chem. Soc. 1986, 108, 4228. (e) Gassman,
P. G.; Deck, P. A.; Winter, C. H.; Dobbs, D. A.; Cao, D. H.
Organometallics 1992, 11, 959.
(25) Nefedov, V. I.; Porai-Koshits, M. A. Mater. Res. Bull. 1972, 7, 1543.
(26) Since 7b is an oil, no XPS data from this compound can be obtained.
(27) A shake-up electron with an apparent higher binding energy is observed
when apart from normal ionization from a 2p orbital, another electron
is excited to a higher lying orbital. For nickel, this extra excitation is
ascribed to a 3d-4s transition. Since this transition is forbidden for
diamagnetic square planar d8 complexes, usually low shake-up satellite
intensity is observed for such complexes. See: Matienzo, L. J.; Yin,
L. I.; Grim, S. O.; Swartz, W. E. Inorg. Chem. 1973, 12, 2762.
is also valid for their cationic analogues 7. However, the
shapes of the signals of both fluorous 3c and 7c are very
different from their respective nonfluorous analogues con-
taining nonsubstituted dppe or trimethylsilyl-substituted 1a;
i.e., relatively high shake-up satellite intensity was observed
for the fluorous compounds (Figure 2). Since no or low
shake-up satellite intensity is usually observed for square
planar nickel complexes,27,28 the high shake-up satellite
intensity observed in the XPS spectra of fluorous 3c and 7c
Inorganic Chemistry, Vol. 42, No. 6, 2003 2119

de Wolf et al.
Table 6. Mulliken Populations of
populations of the -SiH3 and -SiMe2CH2CH2CxF2x+1
(x ) 1, 6) moieties.
[NiCl2{(4-R-C6H4)2PCH2CH2P(C6H4-4-R)2}] Calculated by DFTa
Mulliken populations
Hammett
To further study the relative energies of possible geometric
distortions that were observed by XPS for 3c and 7c,
constrained, tetrahedrally distorted structures of [Ni(dppe)2]2+
were calculated by DFT. The energies of the highest occupied
molecular orbital (HOMO) and lowest unoccupied molecular
orbital (LUMO) and the relative, calculated total energies
are shown in Figure 3a,b. The energy of the HOMO
gradually increases upon tetrahedral distortion, whereas the
energy of the LUMO only decreases after a maximum at
slightly distorted structures ( trans-P-Ni-P ) 176 °). For
this conformation, the total energy shows a minimum, in
agreement with extended Hu¨ckel calculations on [Ni-
(PH3)4]2+.14 Subsequently, the total energy gradually in-
creases upon tetrahedral distortion showing that only one
minimal structure is favored for [Ni(dppe)2]2+, in line with
31P NMR spectroscopic data (vide supra) and extended
Hu¨ckel calculations on [Ni(PH3)4]2+.14
R
param (σp)
Si
Cl
CF3
F
H
0.53
0.06
0.00
0.00b
-0.41
-0.42
-0.43
-0.43
-0.43
-0.43
-0.44
-0.45
SiH3
SiMe2CH2CH2CF3
SiMe2CH2CH2C6F13 (3c)
0.62
1.43
1.43
OMe
NH2
-0.27
-0.66
a Mulliken populations of Ni (0.75) and P (0.55) similar for all
compounds. b Hammett parameter of -SiMe3.
is most likely caused by tetrahedral distortion of these
complexes. Furthermore, deconvolution of the overlapping
2p3/2 peak and its shake-up satellite shows that also the full
width at half-maximum (fwhm) (∆E1/2) is much larger for
fluorous compounds 3c and 7c. This indicates that both of
the fluorous nickel compounds adopt at least two geometries
in the solid state.16 Further deconvolution for 7c of the main
peak into two peaks, in line with the 31P NMR spectrum of
7c, gave peaks at 851.9 and 855.3 eV (fwhm ) 5.8 and 5.0
eV, respectively) in a 1:2 ratio.
Discussion
The electronic and conformational studies described in the
preceding section have been performed to investigate the
steric and electronic influence of para-SiMe2CH2CH2C6F13-
functionalization of the ligands on the metal center in a range
of catalytically relevant group 10 metal complexes. The
similar electron binding energies for fluorous and nonfluo-
rous complexes found by XPS and the similar Ni, P, and Cl
Mulliken populations calculated for NiCl2(dppe) and 3c by
DFT show that any inductive electronic effect of the
(1H,1H,2H,2H-perfluoroalkyl)dimethylsilyl-substituent ex-
erted on the metal center is negligible. Differences in the
onsets of the reduction potentials observed by cyclovolta-
mmetry and in the shake-up satellite intensity in the XPS
spectra were found for fluorous and nonfluorous nickel
complexes. These observations, however, can be explained
by steric rather than inductive electronic effects:
It is known that nickel complexes are configurationally
more flexible than the corresponding palladium and platinum
complexes due to lower crystal field stabilization energy.
For example, X-ray crystallographic structures of both
tetrahedral and square planar forms of NiCl2(PPh3)2 have
been reported.29 It was also mentioned that, for diamagnetic
d8 nickel complexes containing four phosphorus ligands, a
variety of structures can be expected as a result of com-
promises between steric and electronic influences.14 This
makes the nickel complexes more sensitive toward non-
bonding interactions with bulky substituents when compared
to the corresponding palladium and platinum complexes,
which in turn have strong preferences for the square planar
conformation.
Finally, a possible inductive electronic influence of para-
(1H,1H,2H,2H-perfluoroalkyl)dimethylsilyl-functionaliza-
tion was studied using density functional theory (DFT)
calculations on para-substituted derivatives of NiCl2(dppe)
and determination of the atomic population distributions
(Table 6). Similar slightly distorted, square planar structures
( trans-Cl-Ni-P ) 175 °) were found for all compounds.
Interestingly, electron-withdrawing (CF3, F) or electron-
donating (OMe, NH2) substituents influence the Mulliken
populations on the anionic chloride ligands rather than the
populations of the nickel (0.75) and phosphorus atoms (0.55),
which hardly varied over the range of compounds studied.
Except for a para-F-substituent, a linear relationship between
the Hammett parameter (σp) and the Mulliken population
on the chlorine atoms was found. It seems that changes of
the population on the phosphorus and nickel atoms are
compensated for by variation of the electron populations on
the chloride ligands. Most importantly, the Mulliken popula-
tions on the P, Ni, and Cl atoms of the -SiH3-, -SiMe2-
CH2CH2CF3-, and -SiMe2CH2CH2C6F13-substituted nickel
complexes are essentially identical. It was previously shown
by DFT calculations8 and by IR spectroscopy on a series of
trans-[MCl(CO)(L)2] complexes6 (M ) Rh, Ir; L ) mono-
dentate (fluorous) phosphine) that para-CxF2x+1- and para-
(CH2)nCxF2x+1-substituents (n ) 2, 3) directly attached to the
aryl ring exert a (small) electron-withdrawing effect on the
phosphorus donor atom. The results of the calculations
presented here are further support for the idea that the
electron-withdrawing effect of the perfluoroalkyl tail is
compensated for by the silicon atom of the -SiMe2CH2CH2-
spacer. The electron-withdrawing effect of the perfluoroalkyl
tail only becomes apparent when comparing the Si Mulliken
Tetrahedral distortion, possibly occurring upon attachment
of para-SiMe2CH2CH2C6F13 substituents, would also be in
(29) (a) Brammer, L.; Stevens, E. D. Acta Crystallogr. 1989, 45, 400. (b)
Corain, B.; Longato, B.; Angeletti, R. Inorg. Chim. Acta 1985, 104,
15. (c) Bruins Slot, H. J.; Havere, W. K. L.; Noordik, J. H.; Beurskens,
P. T.; Royo, P. J. Crystallogr. Spectrosc. Res. 1984, 14, 623. (d)
Garton, G.; Henn, D. E.; Powell, H. M.; Venanzi, L. M. J. Chem.
Soc. 1963, 3625.
(28) Perera, J. S. H. Q.; Frost, D. C.; McDowell, C. A. J. Chem. Phys.
1980, 72, 5151.
2120 Inorganic Chemistry, Vol. 42, No. 6, 2003

Perfluoroalkyl-Functionalized Metal Complexes
upon tetrahedral distortion of [Ni(dppe)2]2+ by DFT calcula-
tions. However, Figure 3 shows that the energy surface upon
tetrahedral distortion of [Ni(dppe)2]2+ is relatively flat,
allowing thermals distortions with trans-P-Ni-P angles up
to 165° (4RT), demonstrating the geometrical flexibility of
the nickel(II)-center. In contrast to [Ni(dppe)2]2+, 7c contains
8 large, rigid perfluoroalkyl groups. Possibly, steric constraint
might arise in 7c due to the relatively small space available
for the 8 rigid p-(1H,1H,2H,2H-perfluoroalkyl)dimethylsilyl-
substituents. This might result in an energy surface for tet-
rahedral distortion which is different from the one calculated
for [Ni(dppe)2]2+. Tetrahedral distortion, resulting in a more
spherical distribution of the para-substituents, could relieve
some of this steric constraint. When combined with the high
flexibility of the nickel geometry, this could give rise to a
second minimal, tetrahedrally distorted structure of 7c.
It remains intriguing that only fluorous 7c and 3c and not
nonfluorous 7b, all three of them containing a large p-
alkyldimethylsilyl-substituent, show differences in XPS and
CV compared to their nonsubstituted analogues. This might
be related to steric constraint arising from the larger size of
a fluorine atom compared to a hydrogen atom and/or the
lower flexibility of perfluoroalkyl tails compared to perpro-
tioalkyl tails,2 leading to distortion from planarity of the
geometry around the nickel atom.
Although no significant differences between fluorous and
nonfluorous palladium and platinum complexes were ob-
served by XPS and 195Pt NMR, slightly lower 1JPtP coupling
constants were observed upon both trimethylsilyl- and
(1H,1H,2H,2H-perfluoroalkyl)silyl-functionalization (vide
supra). Inductive electronic effects as well as severe tetra-
hedral distortion are not likely to occur for platinum
complexes, and no indications for these phenomena were
found by XPS or 195Pt NMR. It seems therefore that the
somewhat lower 1JPtP coupling constants are most probably
caused by subtle steric effects, e.g., causing a widening of
the intervalence angles around the phosphorus atom (distor-
tion to sp2 hybridization), and/or by changes in the P-Pt-P
bite angle. Therefore, the Hammett parameters for para-
SiMe2CH2CH2CF3 and para-SiMe2CH2CH2C6F13 (0.08 and
Figure 3. a. Energy levels of HOMO (b) and LUMO (9) of constrained,
tetrahedrally distorted structures of [Ni(dppe)2]2+, calculated by DFT. b.
Total energies of constrained, tetrahedrally distorted structures of
[Ni(dppe)2]2+, relative to the nonconstrained minimum (-12.82283 MJ/
mol), calculated by DFT.
agreement with the difference in the CV data obtained for
[Ni(dppe)2](BF4)2 and [Ni(1c)2](BF4)2 (7c). Figure 3 shows
that the energy of the LUMO of [Ni(dppe)2]2+ is lowered
upon tetrahedral distortion. It was found that attachment
of p-(1H,1H,2H,2H-perfluorooctyl)dimethylsilyl-groups to
[Ni(dppe)2]2+ results in a reduction potential which is 0.25
V less negative compared to that of [Ni(dppe)2]2+. If it is
assumed that this corresponds to an energy decrease of the
LUMO of 0.25 eV and that the energies of the LUMO of 7c
follow a similar path upon tetrahedral distortion as calculated
for [Ni(dppe)2]2+, then this would correspond by interpolation
in Figure 3a to a tetrahedrally distorted structure with trans-
P-Ni-P angles of 158°.
A tetrahedral distortion upon (1H,1H,2H,2H-perfluoro-
alkyl)dimethylsilyl-functionalization would also explain the
higher shake-up satellite intensity observed in the XPS
spectra of 3c and 7c, since the 3d-4s transition becomes
more allowed in that case.27 Consequently, mixtures consist-
ing of square planar and tetrahedrally distorted geometries
could be present in both the solid state (as in XPS
measurements) and in solution (as in CV and 31P NMR
measurements). This would also be in agreement with the
larger fwhm found in the XPS spectra of the fluorous nickel
compounds.
1
0.10, respectively), that can be determined from the JPtP
coupling constants of cis-[PtCl2(2)] by interpolation in a
1
linear plot of the Hammett constants and the JPtP coupling
constant for several cis-PtCl2[P-(C6H4-4-R)3]2 complexes,20
seem to be unreliable and are not in agreement with the DFT
calculations on 3 (Hammett parameter σp ) 0.00 for both
substituents).
Conclusions
Fluorous derivatives of dppe and PPh3, containing the
para-SiMe3-b(CH2CH2-)b-spacer, show similar complex-
ation behavior toward Ni, Pd, and Pt compared to their
nonsubstituted analogues. Using the fluorous diphosphines,
both neutral and ionic group 10 metal complexes with
preferential solubility in perfluorinated solvents were syn-
thesized. NMR spectroscopy, XPS data, DFT calculations,
and to a lesser extent cyclic voltammetry indicate that the
para-(1H,1H,2H,2H-perfluoroalkyl)dimethylsilyl-function-
In agreement with the single structure found for
[Ni(dppe)2]2+ by 31P NMR, no local minimum was found
Inorganic Chemistry, Vol. 42, No. 6, 2003 2121

de Wolf et al.
using the SSF weighting scheme with screening37 and MHL angular
grid pruning.36b The gradient of the energy is evaluated without
considering the gradient of the quadrature weights and grid points.
The basis set was the double zeta basis double due to Schafer, Horn,
and Ahlrichs;38 Cartesian Gaussians were used throughout.
1,2-Bis[bis{4-((dimethyloctyl)silyl)phenyl}phosphino]-
ethane (1b). A solution of 0.74 g of 1,2-bis[bis(4-bromophenyl)-
phosphino]ethane (1.04 mmol) in THF (20 mL) was cooled to -90
°C, and 5.5 mL of a solution of t-BuLi (1.5 M; 8.3 mmol) was
added. After the mixture was stirred for 30 min while the
temperature was kept lower than -60 °C, 0.86 g (4.15 mmol) of
C8H17SiMe2Cl was added. After the mixture was heated to room
temperature, all volatiles were evaporated. The residue was dis-
solved in CH2Cl2/degassed water. The organic layer was separated,
dried on MgSO4, and evaporated to dryness, yielding 1.04 g (93%)
of a clear light brown oil. Anal. Calcd for C66H112P2Si4: C 73.41,
H 10.45. Found: C 69.67, H 10.93. 1H NMR (200 MHz, CDCl3):
δ 0.28 (s, 24 H), 0.77 (m, 8 H), 0.92 (t, 3JHH ) 3.4 Hz, 12 H), 1.29
(m, 48 H), 2.16 (ps t, J ) 3.5 Hz, 4H), 7.35-7.54 (m, 16 H). 31P
NMR (CDCl3, 81.0 MHz): δ -12.0.
alization of the phosphine ligand does not influence the
electronic properties of the metal center in these group 10
complexes. It can thus be concluded that, by using a silicon
spacer, substantial fluorophilicity has been introduced in
group 10 metal diphosphine complexes with minimal con-
sequences for the electronic properties of the metal center.
So far this has been difficult to achieve for diphosphines
complexes. The metal complexes are potentially interesting
for fluorous biphasic separation strategies.
Experimental Section
General. All experiments were performed in a dry dinitrogen
atmosphere using standard Schlenk techniques. Solvents were stored
over sodium benzophenone ketyl or CaH2 and distilled before use.
Fluorinated solvents (c-C6F11CF3 (Lancaster), FC-72 and CF3C6H5
(Acros)) were degassed and stored under dinitrogen atmosphere.
Derivatives of dppe (1a-f),10 2b,12c [Ni(MeCN)6](BF4)2‚1/2MeCN,30
[Pd(MeCN)4](BF4)2,31 PdCl2(COD),32 and PtCl2(COD)33 were
prepared according to previously reported procedures. Other
chemicals were obtained from commercial suppliers (Acros, Ald-
rich, Lancaster) and used as delivered. Microanalyses were carried
out by H. Kolbe, Mikroanalytisches Laboratorium, Mu¨lheim an der
Ruhr. 19F NMR spectra were externally referenced against CFCl3
(δ ) 0 ppm), 31P NMR spectra against H3PO4 (δ ) 0 ppm), and
195Pt NMR spectra against a 1 M solution of NaPtCl6 in D2O (δ )
0 ppm).
The XPS data were obtained with a Vacuum Generators XPS
system, using a CLAM-2 hemispherical analyzer for electron
detection. Nonmonochromatic Al KR X-ray radiation was used for
exciting the photoelectron spectra using an anode current of 20
mA at 10 keV. The pass energy of the analyzer was set at 50 eV.
The survey scan was taken with a pass energy of 100 eV.
Electrochemical measurements were performed in an electro-
chemical cell where a nitrogen atmosphere was maintained. The
measurements were performed with a conventional three-electrode
configuration using platinum electrodes and a Ag/AgCl reference
electrode. Cyclovoltammetric measurements were carried out in
MeCN/CH2Cl2 (1:1 v/v) containing the supporting electrolyte (0.05
M Bu4NPF6). The reference electrode was separated from the
solution by a glass frit. Peaks were referenced to the Fc/Fc+ couple
in MeCN/CH2Cl2 (1:1 v/v). An EG&G potentiostat/galvanostat
model 263A was used, connected to model 270/250 Research
Electrochemistry Software (version 4.23).
Tris[4-{dimethyl(3,3,3-trifluoropropyl)silyl}phenyl]phos-
phine (2a). To a solution of tris(4-bromophenyl)phosphine (0.56
g; 1.12 mmol) in Et2O (20 mL) was added t-BuLi (6.7 mmol; 4.5
mL 1.5 M solution in pentane) at -78 °C. After stirring for 10
min, the reaction mixture was quenched with chlorodimethyl(3,3,3-
trifluoropropyl)silane (0.64 g; 3.36 mmol). After the reaction
mixture stirred for 3 h at room temperature, degassed water was
added. After phase separation, drying of the organic layer on
MgSO4, filtration, and evaporation of the solvent, 0.70 g (86%) of
a white solid was obtained. Anal. Calcd for C33H42F9PSi3: C 54.68,
1
H 5.84, P 4.27. Found: C 54.82, H 5.74, P 4.36. H NMR (200
MHz, CDCl3): δ 0.31 (s, 18H), 0.95 (m, 6H), 1.95 (m, 6H), 7.30
(m, 6H), 7.46 (m, 6H). 19F NMR (282 MHz, CDCl3): δ -69.1 (t,
3JHF ) 9.9 Hz). 31P NMR (81 MHz, CDCl3): δ -4.9.
General Procedure for Synthesis of 3. A solution of the dppe
derivative in an appropriate solvent was added to NiCl2‚6H2O
dissolved in EtOH or EtOH/CF3C6H5 (3f). All volatiles were
evaporated in vacuo, and the product was washed with EtOH (3c,
3d, and 3e) or purified in a FC-72/EtOH biphasic system (3f).
Compound 3a was not purified any further.
3a. Compound 1a (0.30 g; 0.44 mmol) dissolved in EtOH/CH2-
Cl2 (50 mL; 4:1 v/v) and NiCl2‚6H2O (0.10 g; 0.44 mmol) yielded
0.35 g of a red solid (94%). Anal. Calcd for C38H56Cl2NiP2Si4: C
1
55.88, H 6.91, P 7.58. Found: C 55.72, H 6.83, P 7.63. H NMR
(200 MHz, CDCl3): δ 0.30 (s, 36H), 2.10 (m, 4H), 7.61 (m, 8H),
All calculations were perfomed using Gamess-UK SGI version
6.3.34 The DFT is invoked using the following keywords: DFT
B3LYP. The keyword B3LYP selects the hybrid exchange-
correlation energy functional due to Becke.35 The quadrature grids
are designed to obtain a relative error of less than 1.0 × 10-6 in
the number of electrons per atom. These grids are constructed from
the logarithmic radial grid36 and Gauss-Legendre angular grid,
7.91 (m, 8H). 31P NMR (81 MHz, CDCl3): δ 57.9.
3c. Compound 1c (0.56 g; 0.28 mmol) dissolved in EtOH/CH2-
Cl2 (50 mL; 4:1 v/v) and NiCl2‚6H2O (66 mg; 0.28 mmol) yielded
0.51 g of a red solid (85%). Anal. Calcd for C66H60Cl2F52NiP2Si4:
C 36.96, H 2.82, P 2.89. Found: C 37.02, H 2.94, P 3.02. 1H NMR
(200 MHz, CDCl3): δ 0.36 (s, 24H), 1.02 (m, 8H), 2.06 (m, 12H),
7.61 (m, 8H), 7.97 (m, 8H). 31P NMR (81 MHz, CDCl3): δ 58.1.
19F NMR (282.4 MHz, CDCl3): δ -126.6 (m, 8F), -123.5 (m,
8F), -123.3 (m, 8F), -122.4 (m, 8F), -116.3 (m, 8F), -81.3 (m,
12F).
(30) Sen, A.; Ta-Wang, L. J. Am. Chem. Soc. 1981, 103, 4627.
(31) Ittel, S. D. In Inorganic Synthesis; MacDarmid, A. G., Ed.; Mcgraw-
Hill: New York, 1977; Vol. 17, p 117.
(32) Frye, H.; Kuljian, E.; Viebrock, J. Inorg. Chem. 1965, 4, 1499.
(33) Drew, D.; Doyle, J. R. Inorg. Synth. 1990, 28, 346.
(34) Guest, M. F.; van Lenthe, J. H.; Kendrick, J.; Scho¨ffel, K.; Sherwood,
P.; Harrison, R. J. GAMESS-UK Version 6.3, a package of ab initio
programs; 2001. This was produced with contributions from Amos,
R. D.; Buenker, R. J.; Dupuis, M.; Handy, N. C.; Hillier, I. H.;
Knowles, P. J.; Bonacic-Koutecky, V.; von Niessen, W.; Saunders:
V. R.; Stone, A. J. It is derived from the original GAMESS code:
Dupuis, M.; Spangler, D.; Wendolowski, J. GAMESS; NRCC Software
Catalog, 1980; Vol. 1, Program QG01.
3d. Compound 1d (0.96 g; 0.40 mmol) dissolved in CF3C6H5
and NiCl2‚6H2O (95 mg; 0.40 mmol) yielded 0.91 g of a red solid
(36) (a) Mura, M. E.; Knowles, P. J. J. Chem. Phys. 1996, 104, 9848. (b)
Murray, C. W.; Handy, N. C.; Laming, G. J. Mol. Phys. 1993, 78,
997.
(37) Stratmann, R. E.; Scuseria, G. E.; Frisch, M. J. Chem. Phys. Lett.
1996, 257, 213.
(35) Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
(38) Schafer, A.; Horn, H.; Ahlrichs, R. J. Phys. Chem. 1992, 97, 2571.
2122 Inorganic Chemistry, Vol. 42, No. 6, 2003

Perfluoroalkyl-Functionalized Metal Complexes
(89%) Anal. Calcd for C74H60Cl2F68NiP2Si4: C 34.92, H 2.38, F
50.76. Found: C 35.11, H 2.51, F 50.55. H NMR (200 MHz,
Found C 31.04, H 1.89, F 54.29. 1H NMR (200 MHz, C6D6/FC-72
1:1 v/v): δ 0.00 (s, 12H), 0.83 (br, 16H), 1.84 (br, 16H), 7.15 (br,
8H), 7.96 (br, 8H).
cis-PtCl2[P(4-C6H4-SiMe2CH2CH2CF3)3]2 (6a). Compound 2a
(0.25 g; 0.35 mmol) and PtCl2(COD) (64.6 mg; 0.17 mmol) yielded
0.24 g (82%) of a white solid. Anal. Calcd for C66H84Cl2F18P2-
PtSi6: C 46.20, H 4.93, P 3.61. Found: C 46.16, H 4.98, P 3.68.
1H NMR (200 MHz, CDCl3): δ 0.33 (s, 36H), 0.96 (m, 12H), 1.96
(m, 12H), 7.28 (m, 12H), 7.45 (m, 12H).
cis-PtCl2[P(4-C6H4-SiMe2CH2CH2C6F13)3]2 (6b). Compound 2b
(0.53 g; 0.36 mmol) and PtCl2(COD) (67.0 mg; 0.18 mmol) yielded
0.53 g (91%) of a white solid. Anal. Calcd for C96H84Cl2F78P2-
PtSi6: C 35.85, H 2.63, P 1.93. Found: C 35.70, H 2.71, P 1.85.
1H NMR (200 MHz, CDCl3): δ 0.29 (s, 36H), 0.94 (m, 12H), 1.92
(m, 12H), 7.46 (m, 24H).
General Procedure for Synthesis of 7. A solution of 1 in CH2-
Cl2 was added to a solution of [Ni(MeCN)6](BF4)2‚1/2MeCN in
MeCN. After the reaction mixture stirred for 2 h, all volatiles were
evaporated in vacuo. The residue was dissolved in CH2Cl2, filtered,
and evaporated to dryness.
1
CDCl3): δ 0.37 (s, 24H), 1.03 (m, 8H), 2.07 (m, 12H), 7.61 (m,
8H), 7.98 (m, 8H). 31P NMR (81 MHz, CDCl3): δ 58.1. 19F NMR
(282.4 MHz, CDCl3): δ -126.6 (m, 8F), -123.5 (m, 8F), -123.2
(m, 8F), -122.4 (m, 24 F), -116.4 (m, 8F), -81.3 (m, 12 F).
3e. Compound 1e (2.29 g; 0.69 mmol) dissolved in EtOH/
CF3C6H5 (100 mL; 1:1 v/v) and NiCl2‚6H2O (163 mg; 0.69 mmol)
yielded 2.13 g of a red waxy solid (89%). Anal. Calcd for C94H64-
Cl2F104NiP2Si4: C 32.51, H 1.86, P 1.78. Found: C 32.38, H 1.94,
1
P 1.88. H NMR (200 MHz, CDCl3): δ 0.34 (br, 12H), 1.06 (br,
16 H), 2.02 (br, 20 H), 7.4-8.2 (br, 16H). 31P NMR (81 MHz,
CDCl3): δ 58.0.
3f. Compound 1f (1.68 g; 0.36 mmol) dissolved in CF3C6H5 and
NiCl2‚6H2O (85 mg; 0.36 mmol) yielded 1.50 g of a red oil (87%).
Anal. Calcd for C122H68Cl2F156NiP2Si4: C 30.52, H 1.43, P 1.29.
Found: C 30.32, H 1.36, P 1.42. 1H NMR (200 MHz, C6D6/FC-72
1:1 v/v): δ 1.01 (br, 24H), 1.96 (br, 28H), 7.2-8.2 (br, 16H). 31
NMR (81 MHz, CDCl3/CF3C6H5 1:1 (v/v)): δ 57.8.
P
General Procedure for Synthesis of 4. A solution of 1 in CH2-
Cl2 (1a, 1c) or PFMCH (1f) was added to a solution of PdCl2(COD)
in CH2Cl2. After the reaction mixture stirred overnight, all solvent
was evaporated in vacuo.
7a. Compound 1a (0.10 g; 0.14 mmol) and [Ni(MeCN)6](BF4)2‚
1/2MeCN (35 mg; 70 µmol) yielded 0.10 g (89%) of a yellow solid.
Anal. Calcd for C76H112B2F8NiP4Si8: C 56.82, H 7.03, P 7.71.
Found: C 56.70, H 6.92, P 7.56. 1H NMR (200 MHz, CD3CN): δ
0.36 (s, 72H), 2.71 (m, 8H), 7.06 (m, 16H), 7.51 (m, 16H). 31P
NMR (81 MHz, CD3CN): δ 54.5.
4a. Compound 1a (0.16 g; 0.22 mmol) and PdCl2(COD) (64 mg;
0.22 mmol) yielded 0.16 g (81%) of a yellow solid. Anal. Calcd
for C38H56Cl2P2PdSi4: C 52.80, H 6.53, P 7.17. Found: C 52.55,
1
H 6.46, P 7.25. H NMR (200 MHz, CDCl3): δ 0.27 (s, 24H),
7b. Compound 1b (0.28 g; 0.26 mmol) and [Ni(MeCN)6](BF4)2‚
1/2MeCN (65 mg; 0.13 mmol) yielded 0.27 g (87%) of an orange
oil. Anal. Calcd for C132H224B2F8NiP4Si8: C 66.28, H 9.44, P 5.18.
Found: C 66.20, H 9.35, P 5.29. 1H NMR (200 MHz, CDCl3): δ
0.24 (s, 48H), 0.73 (m, 16H), 0.87 (br, 16H), 1.25 (br, 96H), 2.92
(m, 8H), 7.1-7.5 (m, 32H). 31P NMR (81 MHz, CDCl3): δ 54.9.
7c. Compound 1c (0.68 g; 0.34 mmol) and [Ni(MeCN)6](BF4)2‚
1/2MeCN (85 mg; 0.17 mmol) yielded 0.67 g (93%) of a yellow
solid, which was recrystallized from MeCN and hexane. Anal. Calcd
for C132H120B2F112NiP4Si8: C 37.19, H 2.84, P 2.91. Found: C
2.44 (m, 4H), 7.59 (m, 8H), 7.82 (m, 8H). 31P NMR (81 MHz,
CDCl3): δ 64.7.
4c. Compound 1c (0.27 g; 0.13 mmol) and PdCl2(COD) (38 mg;
0.13 mmol) yielded 0.27 g (95%) of a yellow solid. Anal. Calcd
for C66H60Cl2F52P2PdSi4: C 36.15, H 2.76, P 2.83. Found: C 36.26,
1
H 2.70, P 2.99. H NMR (200 MHz, CDCl3): δ 0.35 (s, 24H),
1.01 (m, 8H), 2.02 (m, 8H), 2.50 (m, 4H), 7.61 (m, 8H), 7.90 (m,
8H). 31P NMR (81 MHz, CDCl3): δ 64.5. 19F NMR (282.4 MHz,
CDCl3): δ -126.7 (m, 8F), -123.6 (m, 8F), -123.4 (m, 8F),
-122.5 (m, 8F), -116.4 (m, 8F), -81.4 (m, 12F).
1
37.09, H 2.85, P 3.04. H NMR (200 MHz, CD3CN): δ 0.36 (s,
4f. Compound 1f (0.21 g; 44 µmol) and PdCl2(COD) (13 mg;
44 µmol) yielded 0.21 g (90%) of a yellow oil. Anal. Calcd for
48H), 1.05 (m, 16H), 2.08 (m, 16H), 2.71 (m, 8H), 6.9-7.6 (m,
32H). 31P NMR (81 MHz, CD3CN): δ 54.5 (s, 3P). 19F NMR (282
MHz, CD3CN): δ -150.7 (m, 8F), -125.7 (m, 16F), -122,4 (m,
32F), -121.4 (m, 16F), -115.1 (m, 16F), -80.6 (m, 24F).
General Procedure for Synthesis of 8. A solution of 1 in CH2-
Cl2 was added to a solution of [Pd(MeCN)4](BF4)2 in MeCN. After
the reaction mixture for 2 h, all volatiles were evaporated in vacuo.
The residue was dissolved in CH2Cl2, filtered, and evaporated to
dryness.
8a. Compound 1a (217 mg; 0.30 mmol) and [Pd(MeCN)4](BF4)2
(67 mg; 0.15 mmol) yielded 240 mg (97%) of a light yellow solid.
Anal. Calcd for C76H112B2F8P4PdSi8: C 55.18, H 6.82, P 7.49.
Found: C 54.94, H 6,73, P 7.58. 1H NMR (200 MHz, CD3CN): δ
0.20 (s, 72 H), 3.02 (m, 8H), 7.20 (m, 16H), 7.41 (m, 16H). 31P
NMR (81 MHz, CDCl3): δ 56.5.
C
122H68Cl2F156P2PdSi4: C 30.22, H 1.41, P 1.28. Found: C 30.25,
1
H 1.49, P 1.41. H NMR (200 MHz, C6D6/PFMCH 1:1 (v/v)): δ
0.86 (br, 24H), 1.85 (br, 28H), 6.8-8.0 (br, 16H). 31P NMR (81
MHz, C6D6/PFMCH 1:1 (v/v)): δ 62.2.
General Procedure for Synthesis of 5. A solution of 1 equiv
of 1 or 2 equiv of monophosphine in CH2Cl2 was added to a solution
of 1 equiv of PtCl2(COD) in CH2Cl2. After the reaction mixture
stirred overnight, all solvent was evaporated in vacuo.
5a. Compound 1a (0.15 g; 0.22 mmol) and PtCl2(COD) (81 mg;
0.22 mmol) yielded 0.19 g (92%) of a white solid. Anal. Calcd for
C38H56Cl2P2PtSi4: C 47.88, H 5.92, P 6.50. Found: C 47.90, H
1
6.05, P 6.39. H NMR (200 MHz, CDCl3): δ 0.30 (s, 36H), 2.37
(m, 4H), 7.60 (m, 8H), 7.81 (m, 8H).
5c. Compound 1c (0.45 g; 0.22 mmol) and PtCl2(COD) (84 mg;
0.22 mmol) yielded 0.46 g (93%) of a white solid. Anal. Calcd for
C66H60Cl2F52P2PtSi4: C 34.75, H 2.65, P 2.72. Found: C 34.61, H
8c. Compound 1c (294 mg; 0.146 mmol) and [Pd(MeCN)4](BF4)2
(32.4 mg; 73 µmol) yielded 292 mg (93%) of a light yellow solid.
Anal. Calcd for C132H120B2F112P4PdSi8: C 36.78, H 2.81, P 2.87.
Found: C 36.91, H 2.90, P 2.97. 1H NMR (200 MHz, CDCl3): δ
0.29 (s, 48H), 1.07 (m, 16H), 2.10 (m, 16H), 2.92 (m, 8H), 6.9-
7.6 (m, 32H). 31P NMR (81 MHz, CDCl3): δ 56.8. 19F NMR (282
MHz, CD3CN): δ -155.2 (m, 8F), -130.1 (m, 16F), -126.8 (m,
32F), -125.9 (m, 16F), -199.5 (m, 16F), -85.1 (m, 24F).
Solubility Studies of NiCl2(dppe) and 3. Saturated solutions
were prepared by stirring the nickel complex in the appropriate
solvent for 30 min at 25 °C. A sample was taken after the mixture
1
2.70, P 2.80. H NMR (200 MHz, CDCl3): δ 0.34 (s, 24H), 1.00
(m, 8H), 2.02 (m, 8H), 2.36 (m, 4H), 7.56 (m, 8H), 7.85 (m, 8H).
19F NMR (282.4 MHz, CDCl3): δ -126.7 (m, 8F), -123.6 (m,
8F), -123.4 (m, 8F), -122.5 (m, 8F), -116.4 (m, 8F), -81.4 (m,
12F).
5e. Compound 1e (0.41 g; 0.12 mmol) and PtCl2(COD) (46 mg;
0.12 mmol) yielded 0.39 g (91%) of a waxy white-yellow solid.
Anal. Calcd for C94H64Cl2F104P2PtSi4: C 31.28, H 1.79, F 54.74.
Inorganic Chemistry, Vol. 42, No. 6, 2003 2123

de Wolf et al.
was allowed to settle and the mass of the sample was determined.
All solvent was removed in vacuo, and the residue was kept under
vacuum (0.1 mbar) for 15 h upon which the weight was constant
within (1 mg and the weight of the residue was determined.
Determination of Partition Coefficients. A known amount of
3 (between 9 and 33 µmol) was dissolved in a biphasic system
containing PFMCH (2.000 ( 0.002 mL) and toluene (2.000 ( 0.002
mL). The mixture was stirred until all of the compound had
dissolved and then equilibrated at 0 °C. When two clear layers were
obtained, an aliquot (500 ( 2 µL) was removed from each layer
by syringe. This was evaporated to dryness, and the residue was
kept under vacuum (0.1 mbar) for 15 h upon which the weight
was constant within (1 mg. Subsequently, the weight of the residue
was determined.
Acknowledgment. ATOFINA Vlissingen B.V. and the
Dutch Ministry for Economic Affairs (SENTER/BTS) are
gratefully acknowledged for their financial support. The
Stichting Nationale Computerfaciliteiten (National Comput-
ing Facilities Foundation, NCF) and NWO are acknowledged
for the use of supercomputer facilities. Also, the staff of
SARA is thanked for their help with the DFT calculations.
Supporting Information Available: Gamess output file listing
energy values, z-matrix, orbital energies, and Mulliken populations
for geometry-optimized [Ni(dppe)2]2+. This material is available
free of charge via the Internet at http://pubs.acs.org.
IC025770L
2124 Inorganic Chemistry, Vol. 42, No. 6, 2003