Synthesis and structural characterization of carbon-centered tris(pentafluorophenyl)silyl derivatives

Article abstract of DOI:10.1016/j.jorganchem.2005.04.058

A general method for the synthesis of carbon-centered tris(pentafluorophenyl)silyl derivatives (RSi(C₆F₅) ₃) by reaction of trichlorosilanes (RSiCl₃) with pentafluorophenylmagnesium bromide was described. The cr

Full text of DOI:10.1016/j.jorganchem.2005.04.058

Journal of Organometallic Chemistry 690 (2005) 3680–3689
www.elsevier.com/locate/jorganchem
Synthesis and structural characterization
of carbon-centered tris(pentafluorophenyl)silyl derivatives
a,
b
,c
a
*
Alexander D. Dilman *, Dmitry E. Arkhipov , Alexander A. Korlyukov
,
a
Valentine P. Ananikov , Vitalij M. Danilenko , Vladimir A. Tartakovsky
b
a
N. D. Zelinsky Institute of Organic Chemistry, 119991 Moscow, Leninsky prosp. 47, Russian Federation
b
Moscow Chemical Lyceum, 109033 Moscow, Tamozhennyi pr. 4, Russian Federation
c
A. N. Nesmeyanov Institute of Organoelement Compounds, 119991 Moscow, Vavilov str. 28, Russian Federation
Received 7 February 2005; received in revised form 7 April 2005; accepted 7 April 2005
Available online 11 July 2005
Abstract
A general method for the synthesis of carbon-centered tris(pentafluorophenyl)silyl derivatives (RSi(C6F5)3) by reaction of trichloro-
silanes (RSiCl3) with pentafluorophenylmagnesium bromide was described. The crystal structures of obtained compounds were
studied by X-ray diffraction analysis (7 structures). The peculiarities of crystal packing were analyzed by means of DFT calculations.
ꢀ 2005 Elsevier B.V. All rights reserved.
Keywords: Organosilicon chemistry; Tris(pentafluorophenyl)silyl derivatives; Trichlorosilanes; Structural studies
1. Introduction
tered groups. It may be expected that owing to signifi-
cant differences in the nature of silicon-heteroatom
and silicon–carbon bonds, in the latter case new reactiv-
ity patterns may emerge.
Organosilicon compounds bearing at the silicon atom
three heteroatomic substituents (RSiX3, X = Hal, OR0)
have found widespread use in organometallic chemistry
[1], in materials science [2], and as reagents in organic
synthesis [3]. The key feature of these derivatives corre-
sponds to their facile associative interaction with nucle-
ophiles affording species with hyper-coordinate silicon
[4]. Even week nucleophiles such as dimethyl ether
may form five-coordinate complex with trichlorosilane
[5]. Similarly, trichloro- and trialkoxysilanes easily un-
dergo hydrolysis reaction, which likely proceeds through
five- or six-valent intermediates [6]. At the same time, it
would be interesting to consider compounds bearing at
the silicon atom three electron withdrawing carbon cen-
Among many electron acceptors the pentafluoro-
phenyl group is particularly attractive, since its ability
for the modification of Lewis acidic or basic properties
of different elements has been noted and extensively
exploited [7]. In addition, pentafluorophenyl organo-
metallic reagents are easily accessible and, contrary
to many other fluorinated carbanions, quite stable
[8]. Accordingly, we decided to synthesize tris(penta-
fluorophenyl)silyl (TPFS) derivatives and investigate
their structural characteristics.
Several representatives of this class have been de-
scribed in the literature [9], but no general approach
for their synthesis has been proposed. Herein, we pres-
ent a convenient procedure for the preparation of var-
ious compounds containing TPFS group and discuss
on their structures studied by X-ray diffraction
analysis.
*
Corresponding author. Tel.: +7 095 938 35 60; fax +7 095 135 53
28.
E-mail addresses: adil25@mail.ru (A.D. Dilman), alex@xrlab.
ineos.ac.ru (A.A. Korlyukov).
0022-328X/$ - see front matter ꢀ 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2005.04.058

A.D. Dilman et al. / Journal of Organometallic Chemistry 690 (2005) 3680–3689
3681
Table 1
Synthesis of TPFS derivativesa
2. Results and discussion
Entry
1
2
Yield of 2, %b
2.1. Synthesis of TPFS derivatives
1
2
1a
1b
2a
2b
74
67
Me SiCl3
For the preparation of TPFS group, we selected the
approach involving coupling of trichlorosilanes with
three molecules of pentafluorophenyl carbanion. From
practical point of view, the most convenient way to
the pentafluorophenyl nucleophile consists of formation
of Grignard reagent directly from bromopentafluoro-
benzene and magnesium in diethyl ether [8a]. Surpris-
ingly, when the obtained dark brown solution was
combined with organotrichlorosilane 1 no reaction was
observed. However, subsequent addition of tetrahydro-
furan in the amount equal to that of ether caused rapid
reaction as evidenced by the precipitation of magnesium
salts. Presumably, significant rate acceleration exhibited
by tetrahydrofuran may be connected with better solva-
tion of magnesium leading to the deaggregation of
organomagnesium reagent. Employment of ether/tetra-
hydrofuran solvent system allows for the preparation
of various TPFS derivatives (Scheme 1, Table 1). Silanes
containing synthetically valuable functionalities such as
allyl, benzyl, phenyl, vinyl, and alkynyl can be readily
synthesized as stable and distillable compounds.
Despite steric bulkiness created in the resulting TPFS
fragment, the introduction of three pentafluorophenyl
groups proceeds quite easily [10]. It is believed that after
the introduction of pentafluorophenyl group the remain-
ing chlorines become more reactive than in the parent
trichlorosilane [11].
The present method is advantageous in comparison
with the approach employing organolithium reagent.
Thus, compounds 2h [9a,12] and 2j [9b] were previously
prepared from corresponding trichlorosilanes and pen-
tafluorophenyllithium in 67% and 38% yield, respec-
tively, while as shown in Table 1 (entries 8 and 10),
yields of 75% and 68% can be achieved using our ap-
proach. Furthermore, the procedure reported herein is
easier to perform and can be applied to larger scale syn-
theses, as tested for the preparation of 80 g of 2h in a
single run.
Cl
SiCl3
3
4
5
1c
1d
1e
2c
2d
2e
85
93
85
Cl
SiCl3
SiCl3
SiCl3
SiCl3
6
7
1f
2f
65
70
Ph
Ph
1g
2g
SiCl3
Me
Ph SiCl3
8
9
1h
1i
1j
2h
2i
2j
75
72
68
SiCl3
10
SiCl3
SiCl3
Ph
Ph
11
1k
2k
67
a
The ratio 1:Mg:C6F5Br = 1:3.1:3.1.
b
Isolated yield.
2.2. Structural studies
To date only one carbon-centered TPFS derivative,
namely (C6F5)4Si, has been structurally characterized
[14]. In this regard, the examination of the series of com-
pounds differing only in the organic substituent (R) is
expected to provide meaningful information on the
properties of the TPSF group in the solid state.
The crystal and molecular structures of 2a, b, d, f, g, i,
k were investigated by X-ray diffraction analysis (Figs.
1–7 and Tables 2 and 3). The presence of fluorine atoms
in all ortho positions of phenyl rings of 2 imparts signif-
icant steric bulk to the TPFS fragment that is reflected in
the crystal structures. For example, in compounds 2a, b,
d, f, g where TPFS group is attached to sp3 carbon, the
It should be pointed out that compounds 2a–k are
absolutely insensitive towards water, thus allowing for
very convenient handling and storage. Moreover, aque-
ous work-up can be employed upon their isolation. The
inertness of carbon-centered TPFS derivatives stands in
sharp contrast with the behavior of oxygen-centered
counterparts, ROSi(C6F5)3, which are moisture sensitive
[13].
1. R SiCl3
2. THF
1a-k
Mg, Et2O
C6F5Br
C6F5MgBr
R
Si(C6F5)3
Fig. 1. Molecular structure of one of three independent molecules 2a
presented by thermal ellipsoids at 50% probability. Hydrogen atoms
are omitted for clarity. CCDC 262523.
2a-k
Scheme 1.

3682
A.D. Dilman et al. / Journal of Organometallic Chemistry 690 (2005) 3680–3689
Fig. 5. Molecular structure of 2g presented by thermal ellipsoids at 50%
probability. Hydrogen atoms are omitted for clarity. CCDC 262527.
Fig. 2. Molecular structure of 2b presented by thermal ellipsoids at
50% probability. Hydrogen atoms are omitted for clarity. CCDC
262524.
Fig. 3. Molecular structure of 2d presented by thermal ellipsoids at
50% probability. CCDC 262525.
Fig. 6. Molecular structure of 2i presented by thermal ellipsoids at 50%
probability. CCDC 262528.
bond distance Csp3 –SiTPFS increases with the increase of
˚
steric requirements of the substituent, from 1.856 A in
˚
2a (R = Me) to 1.887 A in 2g (R = a-phenylethyl). On
the other hand, the electron withdrawing influence of
three pentafluorophenyl groups tends to reduce the
bond length between TPFS moiety and adjacent carbon
atom. Thus, in compound 2k (R = PhC„C) the dis-
˚
tance Csp–SiTPFS of 1.798 A was observed, which is
˚
shorter than typical values of 1.82–1.85 A for silyl alky-
nes (analysis of 178 structures from Cambridge data
base). Similar phenomenon has recently been encoun-
tered for oxygen-centered TPFS derivatives, ROTPFS,
that exhibited the shortest oxygen–silicon bond lengths
[13]. However, the distances between carbon and silicon
in 2a, b, d, f, g, i lie within the typical range suggesting
that the extent of bond reduction depends on the nature
of substituent.
Fig. 4. Molecular structure of 2f presented by thermal ellipsoids at
50% probability. Hydrogen atoms are omitted for clarity. CCDC
262526.

A.D. Dilman et al. / Journal of Organometallic Chemistry 690 (2005) 3680–3689
3683
with more bulky groups such as benzyl (2f) or a-phenyl-
ethyl (2g) the distortion of tetrahedral configuration be-
comes noticeable with the C–Si–C angles ranging from
103ꢁ to 114ꢁ (Table 3).
Concerning carbon atoms, of particular note is the
bending of silylacetylene fragment in compound 2k with
the angle Si–C„C of 171.04ꢁ. Such deviation results in
close contact between an acetylene carbon and a fluorine
˚
atoms, Cꢀ ꢀ ꢀF 2.94 A (Fig. 7).
Relative position of the organic substituent (R) at sil-
icon and perfluorinated phenyl rings deserves special
comments. In most of studied compounds the substitu-
ent and one C6F5 group are in distorted gauche confor-
mation (torsion angles vary from 38.5ꢁ to 71.2ꢁ). The
only exception is 2i (R = vinyl) where the relative posi-
tions of vinyl with respect to C6F5 ring may be described
as eclipsed (angle C(11)Si(1)C(1)C(2) = 3.24ꢁ). Such
conformation results in proximate arrangement of
Fig. 7. Molecular structure of 2k presented by thermal ellipsoids at
50% probability. Hydrogen atoms are omitted for clarity. CCDC
262529.
˚
C(11)ꢀ ꢀ ꢀH(1), 2.73 A, that may be regarded as weak
intramolecular C–Hꢀ ꢀ ꢀpi interaction [15,16]. In the
structure of 2f (R = PhCH2), the interplanar angle be-
tween planes of phenyl and one pentafluorophenyl
groups is 21.2ꢁ with the interatomic distance
The silicon atom in 2 has slightly distorted tetrahe-
dral geometry. In the structure of 2a the configuration
of silicon is almost ideal tetrahedral while in compounds
˚
C(2)ꢀ ꢀ ꢀC(11) being equal to 3.066 A. In addition, the
Table 2
Crystallographic parameters of studied compounds
2a
2b
2d
2f
2g
2i
2k
Molecular formula
Formula weight
Melting point (ꢁC)
Dimension (mm)
Crystal system
C
19H3F15Si
C19H2ClF15Si C21H5F15Si
570.34
48–51
C25H7F15Si
620.40
133–141
C26H9F5Si
634.42
182–188
C20H3F15Si
556.31
117–119
C26H5F15Si
630.39
170–175
544.30
89–90
578.75
119–122
0.2 · 0.2 · 0.2 0.3 · 0.2 · 0.03 0.2 · 0.1 · 0.1 0.3 · 0.3 · 0.3 0.1 · 0.07 · 0.02 0.3 · 0.2 · 0.08 0.1 · 0.1 · 0.02
Trigonal
ꢀ
Monoclinic
P21/c
Monoclinic
P21/n
Orthorhombic Monoclinic
P212121
Monoclinic
P21/c
Triclinic
ꢀ
Space group
Unit cell dimensions
P3
P21/n
P1
˚
a (A)
15.5358(17)
15.5358(17)
13.678(2)
90
9.8335(17)
28.807(6)
7.3241(14)
90
7.3223(5)
14.4292(10)
19.1591(15)
90
7.348(3)
16.777(6)
18.310(6)
90
11.159(3)
17.070(5)
12.725(4)
90
16.196(7)
10.361(4)
12.138(5)
90
11.914(10)
12.545(10)
15.945(13)
90.068(16)
92.255(16)
101.424(16)
2334(3)
˚
b (A)
˚
c (A)
a (ꢁ)
b (ꢁ)
c (ꢁ)
90
120
109.352(7)
90
97.430(4)
90
90
90
99.338(6)
90
107.632(8)
90
3
˚
V (A )
2859.0(6)
6
1.897
1957.5(6)
4
1.964
2007.3(3)
4
1.887
2257.1(14)
4
2391.8(12)
4
1.762
1941.0(14)
4
1.904
Z
4
1.794
qcalc g cmꢁ1
Temperature (K)
Min/Max H, (ꢁ)
Scan type
1.826
120
2.51/29.39
120
120
1.41/29.96
120
1.77/30.06
120
2.25/29.59
120
2.69/29.99
120
2.48/29.69
1.51/30.02
w-scan
0.71073
2.73
˚
Radiation k(Mo Ka) (A)
Linear absorption (l) (cmꢁ1
Tmin/Tmax
)
4.05
2.64
2.44
2.32
2.71
0.201/0.744
1088
2.37
0.976/0.995
1240
0.947/0.947
1596
33677(0.025) 7092(0.040)
0.888/0.988
1128
0.628/0.932
1120
13358(0.014) 18040(0.029) 16489(0.032)
0.935/0.935
1224
0.977/0.995
1256
F(000)
Total reflection (Rint
)
22069(0.053)
5580
18672(0.070)
10011
Number of independent
reflections
5511
4156
3979
3006
5800
4706
6571
5122
6816
4155
Number of independent
reflections with I>2(r)
Parameters
3265
6404
328
325
354
370
380
337
757
wR2
0.0989
0.0443
0.974
0.0917
0.0382
0.981
0.1168
0.0402
1.037
0.1059
0.0509
0.997
0.1108
0.0507
0.944
0.0695
0.0356
0.941
0.1542
0.0662
0.933
R1
Goodness-of-fit
ꢁ3
˚
qmax/qmin (e A
)
0.63/ꢁ0.23
0.39/ꢁ0.26
0.58/ꢁ0.25
0.38/ꢁ0.26
0.43/ꢁ0.27
0.31/ꢁ0.26
0.62/ꢁ0.51

3684
A.D. Dilman et al. / Journal of Organometallic Chemistry 690 (2005) 3680–3689
Table 3
The principal experimental and calculated structural parameters of studied compounds
2aa
2b 2d 2f
Experimental parameters
2g
2i
2k
Si(1)–C(1)
Si(1)–C(11)
1.856(3)
1.886(2)
1.886(2)
1.886(2)
1.875(2)
1.882(2)
1.886(2)
1.880(3)
1.872(1)
1.886(1)
1.891(1)
1.896(1)
1.499(2)
1.881(3)
1.881(3)
1.890(3)
1.885(3)
1.508(4)
1.887(2)
1.893(2)
1.893(2)
1.894(2)
1.527(3)
1.844(2)
1.886(2)
1.883(2)
1.897(2)
1.315(3)
1.798(5)
1.886(4)
1.882(4)
1.879(4)
1.216(4)
Si(1)–C(21)
Si(1)–C(31)
C(1)–C(2)
C(1)Si(1)C(11)
C(1)Si(1)C(21)
C(1)Si(1)C(31)
C(11)Si(1)C(21)
C(11)Si(1)C(31)
C(21)Si(1)C(31)
C(11)Si(1)C(1)C(2)
110.1(2)
108.0(1)
112.4(1)
105.9(1)
111.4(1)
112.8(1)
106.3(1)
41.2(2)
105.78(6)
112.00(6)
112.58(6)
110.29(9)
109.9(1)
106.3(1)
62.5(1)
103.1(1)
111.6(1)
114.4(1)
110.4(1)
111.2(1)
106.3(1)
38.4(2)
103.8(10)
112.1(1)
110.94(9)
111.5(1)
110.94(9)
104.44(9)
57.5(2)
107.82(8)
111.80(8)
110.63(8)
109.45(8)
112.85(7)
104.32(7)
3.5(2)
107.8(2)
110.1(2)
110.1(2)
109.0(2)
109.0(2)
109.0(2)
112.8(2)
106.7(2)
108.9(2)
111.5(2)
108.7(2)
71.2(2)
Calculated parametersb
Si(1)–C(1)
Si(1)–C(11)
1.852/1.854
1.874/1.874
1.889/1.874
1.890/1.891
1.844/1.846
1.799/1.798
1.891/1.878
1.891/1.884
1.891/1.884
1.889/1.879
1.890/1.886
1.889/1.881
1.494/1.487
107.30/107.0
113.2/112.0
113.8/111.9
110.4/109.6
110.2/110.2
101.8/106.2
70.01/59.7
15.36
1.887/1.876
1.893/1.886
1.890/1.884
1.500/1.500
104.1/105.8
112.2/115.4
110.7/114.4
111.1/111.1
110.4/111.4
103.5/98.9
42.9/43.1
1.891/1.883
1.892/1.886
1.894/1.886
1.507/1.504
107.5/104.8
112.4/112.1
114.0/114.1
108.8/109.5
110.8/110.6
103.2/105.7
63.8/57.3
1.877/1.880
1.882/1.877
1.887/1.888
1.343/1.342
108.7/110.2
112.3/113.0
112.4/111.5
109.1/108.6
111.3/108.8
103.0/104.5
2.9/4.5
1.887/1.888
1.879/1.888
1.886/1.899
1.234/1.239
107.0/118.1
113.0/108.3
107.1/110.9
109.3/103.3
111.0/107.2
108.9/108.7
73.9/159.9
20.13
Si(1)–C(21)
Si(1)–C(31)
N(1)–C(2)
C(1)Si(1)C(11)
C(1)Si(1)C(21)
C(1)Si(1)C(31)
C(11)Si(1)C(21)
C(11)Si(1)C(31)
C(21)Si(1)C(31)
C(11)Si(1)C(1)C(2)
Binding energy
a
110.5/110.5
110.5/109.1
110.5/107.2
108.4/108.6
108.4/111.6
108.4/110.5
17.6
21.38
22.02
16.91
The averaged values of three independent molecules are shown.
b
The bond lengths and bond angles for molecules in crystal and isolated state are separated by ‘‘/’’ character.
distortion of the angle between the plane of pentafluor-
ophenyl ring C(11)ꢀ ꢀ ꢀC(16) and silicon–carbon bond
Si(1)–C(11) may be clearly seen from Fig. 8 and
amounts to 11.2ꢁ, while the same angles at the rest of
C6F5 groups do not exceed 3–4ꢁ. Taken together the no-
ticed features allow one to suggest the realization of
piꢀ ꢀ ꢀpi intramolecular interaction in the structure of 2f.
The proposed intramolecular C–Hꢀ ꢀ ꢀpi and piꢀ ꢀ ꢀpi
interactions in 2i and 2f may be caused by steric re-
straints or forces of crystalline packing and may not
be favored in isolated molecules.
In order to understand the role of crystalline packing
in stabilization of observed molecular structures we car-
ried out the quantum chemistry calculations using peri-
odic boundary conditions at DFT level. For consistency
only molecules with the same type of atoms, 2a, d, f, g, i,
k, were considered, while calculations of 2b (R =
CH2Cl) were not performed. To mimic the isolated mol-
ecule all but one molecule were removed from the crystal
cell and to avoid interactions between periodic images
the parameters of the crystal cell were increased (see
computational details). By using the same level of theory
for calculation, the comparison of total energies of crys-
tal and isolated molecules gave us the binding energy,
which is useful for estimation of influence of crystalline
packing on geometry of molecules.
The optimization of atomic positions reproduced
the bond lengths and angles with sufficient accuracy.
˚
The main differences (up to 0.03 A) from experimental
values are observed in the case of C–F bonds and for
C(1)–C(2) bond in vinyl group of compound 2i. The
Fig. 8. The distortion of C(11) atom valence environment in 2f.

A.D. Dilman et al. / Journal of Organometallic Chemistry 690 (2005) 3680–3689
3685
Si–C and C–C bond lengths were reproduced within
˚
0.015 A. In general, both in crystal and in isolated
jected onto C6F5-group. The calculated and experimen-
tal interatomic distances between centroid of former
ring and C(14) atom of the latter one constitute 3.40
molecules the relative positions of substituent R and
C6F5 rings predicted by quantum chemistry calcula-
tions are close to those in experiment. The consider-
able differences between crystal and isolated
molecules are observed in the case of compounds
where the presence of piꢀ ꢀ ꢀpi intramolecular interac-
tion were proposed (vide supra). In isolated molecule
of 2f the interplanar angle and Cꢀ ꢀ ꢀC interatomic dis-
˚
and 3.41 A, respectively, while interplanar angles in
both cases are equal to 7.1ꢁ. The stacking interaction
of similar type was found in perfluorinated tolanes
[18]. The values of interplanar angles and interplanar
˚
separation were in the range of 5–9ꢁ and 3.4–3.5 A,
correspondingly.
Despite shortening of Fꢀ ꢀ ꢀF and Hꢀ ꢀ ꢀF interatomic
distances the total number of intermolecular contacts
changes insignificantly, so one may believe that calcu-
lated binding energies should be reliable for character-
ization of studied structures. Also it should be noted
that the presence of short interatomic distances between
pair of atoms is not a sufficient criterion for attractive
interaction [19]. In order to study the nature and
strength of Fꢀ ꢀ ꢀF and Hꢀ ꢀ ꢀF contacts, we plan in future
to investigate the electron density distribution function
in terms of BaderÕs ‘‘Atoms in molecules’’ theory [20].
The calculated binding energies (Table 3) correspond
to sublimation energies at 0 K and therefore may corre-
late with the magnitude of crystal melting (Table 2).
Analysis of these values did not reveal the exact correla-
tion, presumably because melting point is defined not
only by enthalpy of sublimation. Nevertheless, one
may see that the lowest value of binding energy is found
for compound 2d (R = allyl) which is characterized by
low melting point. Anomalously low melting point of
2d may also be explained by the presence of flexible allyl
group that is in agreement with corresponding values for
2e (R = 2-methallyl) and 2j (R = trans-PhCH@CH) –
they also have flexible groups and rather low melting
points.
Without studies of electron density it is difficult to
completely investigate the influence of intermolecular
interactions on the structure and physical properties of
compounds of such class. On qualitative level one may
explain the observed tendency by different number of
Fꢀ ꢀ ꢀF and C–Hꢀ ꢀ ꢀF contacts in crystal packing of inves-
tigated compounds. In the case of methyl (2a) and vinyl
(2i) derivatives the number of C–Hꢀ ꢀ ꢀF intermolecular
contacts is significantly lower than that in compounds
with more bulky R groups. Probably, an increase of
number of hydrogen atoms in R group leads to corre-
sponding increase of number of C–Hꢀ ꢀ ꢀF contacts,
which are stronger than Fꢀ ꢀ ꢀF ones. On the other hand
the enlargement of R causes crystalline packing to be
more ‘‘loose’’. This is in agreement with density values,
the largest value corresponds to R = Me (2a) and vinyl
(2i), while the least one to R = a-phenylethyl (2g).
From experimental and theoretical structural studies
the following generalizations may be formulated: the
TPFS group possesses considerable steric bulk, which
is further augmented by the tendency to shorten the car-
bon–silicon bond C–Si(C6F5)3, and that the structure of
˚
tances are changed to 47ꢁ and 3.63 A, respectively, as
compared to crystal. Also, the angle between plane
of C(11)ꢀ ꢀ ꢀC(16) pentafluorophenyl ring and Si(1)–
C(11) bond is decreased to 3.4ꢁ in comparison with
calculated values in crystal (12.8ꢁ). So, we may con-
clude that the influence of crystal field led to the par-
allel orientation of phenyl and C6F5 rings and
proposed piꢀ ꢀ ꢀpi interaction between them is not favor-
able. On the other hand, in isolated molecule of com-
pound 2i the interatomic distance C(11)ꢀ ꢀ ꢀH(1) is
nearly the same in experimental and calculated crystal
structures, a fact which can be attributed to the forma-
tion of attractive C–Hꢀ ꢀ ꢀpi contact.
The presence of many fluorine atoms causes the
appearance of C–Hꢀ ꢀ ꢀF and Fꢀ ꢀ ꢀF contacts in crystal
structures of 2a, b, d, f, g, i, k [17]. The closest Fꢀ ꢀ ꢀF
˚
contacts (2.65 A) in experimental crystal structure were
observed in 2a (R = Me), as well as the strongest
˚
C–Hꢀ ꢀ ꢀF contact (2.37 A) occurred in 2f (R = CH2Ph).
The optimization of atomic positions in general led to
some shortening of Fꢀ ꢀ ꢀF and Hꢀ ꢀ ꢀF interatomic dis-
˚
tances by approximately 0.1 A (Table 3). Such errors
in definition of weak intermolecular contacts are typical
for modern DFT functionals which cannot account for
nonlocal dispersion interactions. At the same time the
calculations reproduced well the intermolecular stacking
interaction in 2g (Fig. 9), where the phenyl ring is pro-
Fig. 9. Stacking interaction in crystal structure of 2g. The molecule A
is drawn with dashed lines obtained by following symmetry operations:
ꢁ0.5 + x, 0.5 ꢁy, ꢁ0.5 + z.

3686
A.D. Dilman et al. / Journal of Organometallic Chemistry 690 (2005) 3680–3689
4
(sept, JC–F = 3.1, CH3), 105.1 (t, JC–F = 27.4,
2
TPFS unit in the solid state may be affected by the
fourth substituent and by forces of crystalline packing.
1
i-C6F5), 137.6 (dm, JC–F = 254, CF), 143.6 (dm,
C
1
1JC–F = 258, CF), 149.3 (dm, JC–F = 245, CF). NMR
19F, d: ꢁ160.1 (m, meta), ꢁ147.7 (tt, JF–F = 4.4, 19.8,
3. Conclusion
para), ꢁ127.2 (d, JF–F = 17.0, ortho). IR (KBr, cmꢁ1
)
1645 (w), 1523 (m), 1469 (st), 1380 (w), 1292 (w), 1093
(m), 972 (m), 807 (w), 514 (w), 422 (w). Anal. Calc.
for C19H3F15Si: C, 41.93; H, 0.56. Found: C, 42.01; H,
0.42%.
Herein we presented a convenient procedure for the
synthesis of TPFS derivatives. The ready availability
of this compounds opens up opportunity for investiga-
tion of their synthetic applications. Given the signifi-
cant steric requirements of the TPFS group, the
diminished reactivity of the adjacent carbon atoms
may be expected. At the same time, the presence of
three electron withdrawing pentafluorophenyl rings
can make the silicon atom susceptible to nucleophilic
activation. Studies along these lines are in progress
in our laboratories.
Chloromethyltris(pentafluorophenyl)silane (2b). B.p.
130–134 ꢁC/0.2 Torr. Distilled product was recrystal-
lized from hexane. M.p. 119–122 ꢁC (hexane). NMR
1H (CDCl3), d: 3.69 (s, 2H, CH2); NMR 13C, d:
4
2
26.9 (sept, JC–F = 3.4, CH2), 102.5 (t, JC–F = 29.5,
C
1
i-C6F5), 137.6 (dm, JC–F = 254, CF), 144.0 (dm,
1
= 260, CF), 149.4 (dm, JC–F = 247, CF).
1JC–F
NMR 19F, d: ꢁ159.5 (m, meta), ꢁ146.3 (tt, JF–F
=
4.4, 20.1, para), ꢁ126.2 (d, JF–F = 18.4, ortho). IR
(KBr, cmꢁ1) 1646 (w), 1521 (m), 1476 (st), 1383 (w),
1292 (w), 1102 (m), 1089 (m), 973 (m). Anal. Calc.
for C19H2ClF15Si: C, 39.43; H, 0.35. Found: C,
39.33; H, 0.43%.
4. Experimental
Bromopentafluorobenzene was purchased from P&M
Invest and was used as received. Silanes 1b [21], 1d,e
[22], 1f [23], 1g [24], 1j [25], 1k [26] were synthesized
according to literature procedures; others were commer-
cial products (Aldrich, Acros).
(3-Chloropropyl-1)tris(pentafluorophenyl)silane (2c).
B.p. 140–154 ꢁC/0.4 Torr. M.p. 106–111 ꢁC (hexane).
1
NMR H (CDCl3), d: 1.78–1.88 (m, 4H, CH2CH2Si),
3.58 (t, 2H, J = 6.0, CH2Cl), NMR 13C, d: 12.8 (br,
CH2–Si), 26.6 (CH2CH2Si), 46.5 (CH2Cl), 104.0 (t,
Bromopentafluorobenzene (3.25 mL, 26.0 mmol) was
added dropwise to a suspension of magnesium turnings
(0.625 g, 26.0 mmol) in diethyl ether (14 ml) at such a
rate to maintain gentle reflux. After complete addition
the mixture was refluxed for additional 1 h and cooled
with ice/water bath. Trichlorosilane 1 (8.4 mmol) and
THF (14 mL) were successively added dropwise. The
cooling bath was removed and the mixture was refluxed
for 1 h. After cooling to room temperature, the mixture
was diluted with hexane (14 mL, for 2a–f, h, i) or with
toluene (14 mL, for 2g, j, k), and filtered. The filter cake
was washed with hexane (for 2a–f, h, i) or with toluene
(for 2g, j, k). The combined filtrate was washed with
water (2 · 20 mL), dried (MgSO4) and concentrated un-
der vacuum to give crude product which was distilled in
vacuum.
Distillation of 2c–f, k afforded pure substances, while
for 2a, b, g–j the recrystallization was necessary. The
yields given in Table 1 correspond to pure material.
In large scale syntheses extra care should be taken
after the addition of THF, since removal of cooling bath
and warming causes the spontaneous exothermic reac-
tion accompanying by the precipitation of magnesium
salts. At this point the external cooling has to be ap-
plied. After calming down the reaction the mixture
was refluxed for 1 h as in standard procedure.
1
2JC–F = 28.2, Ci-C6F5), 137.6 (dm, JC–F = 255, CF),
1
143.7 (dm, JC–F = 258, CF), 149.3 (dm, JC–F = 246,
1
CF). NMR 19F, d: ꢁ160.0 (m, meta), ꢁ147.5 (tt, JF–F
=
4.1, 19.8, para), ꢁ126.8 (d, JF–F = 17.7, ortho). IR (KBr,
cmꢁ1) 1645 (m), 1520 (st), 1474 (st), 1384 (w), 1293 (w),
1090 (st), 973 (st), 702 (w), 514 (w), 422 (w). Anal. Calc.
for C21H6ClF15Si: C, 41.57; H, 1.00. Found: C, 41.34,
1.30%.
3-Propenyltris(pentafluorophenyl)silane (2d). B.p.
124–127 ꢁC/0.25 Torr. M.p. 48–51 ꢁC. NMR 1H
(CDCl3), d: 2.64 (d, 2H, J = 7.8, CH2Si), 4.93–4.99 (m,
2H, CH2 = CH), 5.69–5.80 (m, 1H, CH); NMR 13C
2
(CDCl3), d: 22.2 (br., CH2Si), 104.1 (t, JC–F = 29.8,
C
i-C6F5), 118.1 (CH), 129.6 (CH2@), 137.6 (dm,
1
1JC–F =255, CF), 143.7 (dm, JC–F = 259, CF), 149.2
(dm, JC–F = 245, CF); NMR 19F (CDCl3), d: ꢁ160.7
1
(m, meta), ꢁ148.2 (tt, J = 4.5, 20.3, para), ꢁ127.0 (d,
J = 18.1, ortho). IR (KBr, cmꢁ1) 1644 (w), 1521 (m),
1469 (st), 1381 (w), 1294 (w), 1091 (st), 975 (st) 787
(w), 518 (w), 435 (w). Anal. Calc. for C21H5F15Si: C,
44.22; H, 0.88. Found: C, 44.01, 0.88%.
(2-Methylpropenyl-3)tris(pentafluorophenyl)silane (2e).
B.p. 135–145 ꢁC/0.27 Torr. M.p. 74–78 ꢁC. NMR 1H
(CDCl3), d: 1.69 (s, 3H, CH3), 2.76 (s, 2H, CH2–Si),
4.50 (s, 1H), 4.66 (s, 1H) (@CH2). NMR 13C, d:
2
Methyltris(pentafluorophenyl)silane (2a). B.p. 127–
135 ꢁC/0.5 Torr. Distilled product was recrystallized
from hexane. M.p. 89–90 ꢁC. NMR 1H (CDCl3), d:
24.0 (CH2Si), 104.8 (t, JC–F = 28.7, Ci-C6F5), 113.0
1
(CH2@), 137.7 (dm, JC–F = 255, CF), 137.7 (C@CH2),
143.7 (dm, JC–F = 258, CF), 149.5 (dm, JC–F = 245,
1
1
1.17 (sept, 3H, JC–F = 1.5, CH3); NMR 13C, d: ꢁ0.2
CF). NMR 19F, d: ꢁ160.7 (m, meta), ꢁ148.4 (br,
5

A.D. Dilman et al. / Journal of Organometallic Chemistry 690 (2005) 3680–3689
3687
2
para), ꢁ126.1 (d, JF–F = 18.0, ortho). IR (KBr, cmꢁ1
)
104.3 (t, JC–F = 27.3, Ci-C6F5), 115.7 (CHASi), 127.3,
128.8, 129.9 (3 CHPh), 136.6 (Ci-Ph), 137.6 (dm, JC–F
1
1645 (w), 1519 (m), 1471 (st), 1381 (w), 1290 (w),
1090 (m), 970 (m), 531 (w), 432 (w). Anal. Calc. for
C22H7F15Si: C, 45.22; H, 1.21. Found C, 45.27; H,
1.23%.
1
= 255, CF), 143.7 (dm, JC–F = 259, CF), 149.3 (dm,
1JC–F = 246, CF), 151.1 (CH@CHSi). NMR 19F, d:
ꢁ159.9 (m, meta), ꢁ147.4 (tt, JF–F = 4.1, 19.8, para),
ꢁ126.0 (d, JF–F = 18.4, ortho). IR (KBr, cmꢁ1) 1645
(w), 1519 (w), 1470 (st), 1382 (w), 1092 (m), 973 (m).
Anal. Calc. for C26H7F15Si: C, 49.38; H, 1.12. Found:
C, 49.40, 1.35%.
Benzyltris(pentafluorophenyl)silane (2f). B.p. 133–
1
141 ꢁC/0.07 Torr. M.p. 132–134 ꢁC (hexane). NMR H
(CDCl3), d: 3.27 (s, 3H, CH2–Si), 6.90–6.93 (m, 2H,
Ph), 7.08–7.15(m, 3H, Ph); NMR 13C, d: 23.9 (sept.,
2
4JC–F = 2.5, CH2–Si), 104.3 (t, JC–F = 29.6, Ci-C6F5),
126.2, 128.2, 128.7 (CHPh), 134.6 (Ci-Ph), 137.5 (dm,
Phenylethynyltris(pentafluorophenyl)silane (2k). B.p.
170–175 ꢁC (bath temperature)/0.1 Torr. M.p. 130–
133 ꢁC (hexane). 1H NMR (CDCl3), d: 7.39 (t, 2H,
1
1JC–F = 255, CF), 143.5 (dm, JC–F = 259, CF), 149.0
(dm, JC–F = 246, CF). NMR 19F, d: ꢁ160.0 (m, meta),
3J = 7.5), 7.45 (t, 1H, J = 7.5), 7.59 (d, 2H, J = 7.5).
1
3
3
2
ꢁ147.5 (tt, JF–F = 4.1, 20.0, para), ꢁ125.6 (d, JF–F
=
NMR 13C, d: 81.9 (C„C), 103.7 (t, JC–F = 26.3, Ci-
17.7, ortho). IR (KBr, cmꢁ1) 1644 (w), 1519 (m), 1475
(st), 1379 (w), 1293 (w), 1093 (m), 972 (m), 741 (w),
521 (w). Anal. Calc. for C25H7F15Si: C, 48.40; H, 1.14.
Found: C, 48.13, 1.31%.
C6F5), 111.6 (C„C), 121.1 (Ci-Ph), 128.5, 130.3, 132.5 (3
CHPh), 137.7 (dm, JC–F = 254, CF), 144.0 (dm, JC–F
1
1
=
259, CF), 149.5 (dm, JC–F = 246, CF). NMR 19F, d:
ꢁ160.1 (m, meta), ꢁ147.1 (tt, JF–F = 4.4, 19.8, para),
ꢁ126.5 (d, JF–F = 17.7, ortho). IR (KBr, cmꢁ1) 2174
(m), 1645 (m), 1518 (st), 1478 (st), 1382 (m), 1294 (m),
1095 (st), 973 (st), 763 (w), 521 (w) 438 (w). Anal. Calc.
for C26H5F15Si: C, 49.54; H, 0.80. Found: C, 49.40; H,
0.79%.
1
(1-Phenylethyl-1)tris(pentafluorophenyl)silane (2g).
Sublimation 160–164 ꢁC (bath temperature)/0.4 Torr.
Sublimed product was recrystallized from dichloroeth-
1
ane. M.p. 182–188 ꢁC. NMR H (CDCl3), d: 1.56 (d,
3H, J = 7.6, Me), 3.89 (q, 1H, J = 7.6, CHSi), 6.92–
6.96 (m, 2H, Ph), 7.12–7.19 (m, 3H, Ph); NMR 13C, d:
14.3 (Me), 26.9 (CHSi), 104.2 (t, JC–F = 30.0, Ci-C6F5),
Computational details.
2
The quantum chemistry calculations were carried out
using CPMD 3.7.2 [27] density functional (DFT) code.
For the optimizations of atomic position (started from
experimental crystal structures) in crystal simulated
annealing technique was used followed by BFGS mini-
mization of total energy. VanderbiltÕs ultrasoft pseupo-
tentials [28] have been applied to account of core
electrons while valence electrons were approximated
by plane-wave expansion with 25 Ry cutoff. Exchange
and correlation terms of total energy were described
by LDA approximation. Kohn-Sham equations were
integrated using C-point approximation. We believe that
such approximation is sufficient because of rather large
crystal cells. DFT does not take into account dispersion
interactions, so calculated cell parameters may be sys-
tematically overestimated or underestimated up to 5%.
Thus, the experimental values of cell parameters were
used in calculations. Atomic displacements converged
better than 10ꢁ4 a.u., as well as energy variations were
less than 10ꢁ6 a.u.
126.6, 127.0, 128.6 (CHPh), 139.6 (Ci-Ph), 137.5 (dm,
1
1JC–F = 255, CF), 143.4 (dm, JC–F = 259, CF), 148.9
1
(dm, JC–F = 246, CF). NMR 19F, d: ꢁ160.0 (m, meta),
ꢁ147.8 (t, JF–F = 19.8, para), ꢁ124.4 (d, JF–F = 18.4,
ortho). IR (KBr, cmꢁ1) 1644 (m), 1520 (m), 1477 (st)
1470 (st), 1378 (w), 1293 (m), 1089 (st), 975 (m), 521
(w). Anal. Calc. for C26H9F15Si: C, 49.22; H, 1.43.
Found: C, 49.12, 1.57%.
Phenyltris(pentafluorophenyl)silane (2h). B.p. 185–
187 ꢁC/0.5 Torr. Distilled product was recrystallized
from hexane. M.p. 135–136 ꢁC [9a].
Vinyltris(pentafluorophenyl)silane (2i). B.p. 140–
150 ꢁC (bath temperature)/0.2 Torr. Distilled product
was recrystallized from hexane. M.p. 117–119 ꢁC.
1
NMR H (CDCl3), d: 5.84 (d, 1H, J = 19.7), 6.39 (dd,
1H, J = 2.3, 14.4), 6.73 (ddm, 1H, J = 14.4, 19.7);
2
NMR 13C d: 127.9, 138.8 (CH2@CH), 103.9 (t, JC–F
= 29.8, Ci-C6F5), 137.6 (dm, JC–F = 253, CF), 143.8
1
1
1
(dm, JC–F = 257, CF), 149.3 (dm, JC–F = 244, CF).
NMR 19F, d: ꢁ161.2 (m, meta), ꢁ148.6 (tt, JF–F = 4.4,
19.5, para), ꢁ126.9 (d, JF–F = 18.2, ortho). IR (KBr,
cmꢁ1) 1644 (m), 1520 (st), 1478 (st), 1469 (st), 1382
(m), 1294 (m), 1087 (st), 975 (st), 702 (w), 557 (w), 527
(w), 454 (w). Anal. Calc. for C21H3F15Si: C, 43.18; H,
0.54. Found: C, 43.24, 0.60%.
The isolated molecules were simulated utilizing the
same theoretical background, basis sets and convergence
criteria by quantum chemistry calculation of single mol-
˚
ecule in cubic box with side 15 A. The structures of
isolated molecules were tested on stability by calculation
of vibrational frequencies.
Trans-Phenylethenyltris(pentafluorophenyl)silane (2j).
B.p. 169–174 ꢁC/0.4 Torr. Distilled product was treated
with hexane/i-PrOH mixture to cause crystallization.
M.p. 72–75 ꢁC. NMR 1H (CDCl3), d: 6.92 (d, 1H,
J = 18.9), 7.01 (d, 1H, J = 18.9) (CH@CH), 7.36–7.42
(m, 3H, Ph), 7.49–7.52 (m, 2H, Ph). NMR 13C, d:
5. Supplementary materials
Crystallographic data are deposited with the Cam-
bridge Crystallographic Data Centre and are available
free of charge at CCDC, 12 Union Road, Cambridge

3688
A.D. Dilman et al. / Journal of Organometallic Chemistry 690 (2005) 3680–3689
(c) E.J. Harper Jr., E.J. Soloski, C. Tamborski, J. Org. Chem. 29
(1964) 2385.
CB2 1EZ, UK, fax: +44 0 1223 336 033, e-mail: deposit
@ccdc.cam.ca.uk. CCDC ref codes: 262523 (2a), 262524
(2b), 262525 (2d), 262526 (2f), 262527 (2g), 262528 (2i),
262529 (2k).
[9] (a) PhSi(C6F5)3, H.-J. Frohn, A. Lewin, V.V. Bardin, J. Organo-
met. Chem. 568 (1998) 233;
(b) PhCH = CHSi(C6F5)3 and CH2 = C(Ph)Si(C6F5)3 T. Brennan,
H. Gilman, J. Organomet. Chem. 16 (1969) 63;
(c) Si(C6F5)4, see ref 8a.
Acknowledgments
[10] Reaction of ortho-tolyltrichlorosilane with excess of ortho-meth-
oxyphenylmagnesium bromide affords product of substitution of
only two chlorines, see: I.I. Lapkin, F.G. Mukhina, N.F. Kirilov,
L.P. Sheshina, M.S. Kozhevnikova, Zh. Obsch. Khim. 54 (1984)
2039 (Russ. J. Org. Chem. 54 (1984) 1820).
This work was performed at the Scientific educational
canter for young chemists and supported by the Minis-
try of Science (Project MK-2352.2003.03), Russian
Foundation for Basic Research (Project # 00-15-
97359), and INTAS (Project # 2003-55-1185). We also
thank Prof. S.L. Ioffe for start-up funding and V. Levin
for experimental assistance.
[11] (a) The acceleration of nucleophilic attack at the silicon atom with
the introduction of pentafluorophenyl group was documented:
A.L. Klebanskii, Yu.A. Yuzhelevskii, E.G. Kagan, G.A. Niko-
laev, A.V. Harlamova, Zh. Obsch. Khim. 38 (1968) 914 (Russ. J.
Gen. Chem. 38 (1968) 878);
(b) For pertinent discussion on reaction of pentafluorophe-
nylmagnesium bromide with tetraethoxysilane see: A. Whit-
tingham, A.W.P. Jarvie, J. Organomet. Chem. 13 (1968)
125.
References
[12] The yield of 80% for the synthesis of 2h from PhSiCl3 and C6F5Li
was reported: M. Schmeisser, N. Wessal, M. Weidenbruch, Chem.
Ber. 101 (1968) 1897, However, in this reaction we could not
achieve the yield greater than 60%.
[1] (a) M. Nakash, M. Goldvaser, J. Am. Chem. Soc. 126 (2004)
3436;
(b) R.J.P. Corriu, M. Granier, G.F. Lanneau, J. Organomet.
Chem. 562 (1998) 79;
[13] (a) A.D. Dilman, P.A. Belyakov, A.A. Korlyukov, V.A. Tarta-
kovsky, Tetrahedron Lett. 45 (2004) 3741;
(c) R. Pietschnig, F. Belaj, J. Tirree, J. Organomet. 23 (2004)
4897.
(b) V.V. Levin, A.D. Dilman, P.A. Belyakov, A.A. Korlyukov,
M.I. Struchkova, V.A. Tartakovsky, Eur. J. Org. Chem. (2004)
5141.
[2] B. Boury, R.J.P. Corriu, in: Z. Rappoport, Y. Apeloig (Eds.),
The Chemistry of Organic Silicon Compounds, vol. 3, Wiley,
Chichester, UK, 2001, pp. 565–640.
[14] CCDC ref code PFPSIL: A. Karipides, B. Foerst, Acta Crystal-
logr., Sect. B 34 (1978) 3494.
[15] To avoid systematic errors in single crystal X-ray analysis C–H
[3] (a) S. Kobayashi, M. Sugiura, C. Ogawa, Adv. Synth. Catal. 346
(2004) 1023;
˚
bonds were normalized to 1.08 A obtained from single crystal
(b) S.E. Denmark, R.A. Stavenger, Acc. Chem. Res. 33 (2000)
432;
(c) H. Sakurai, Synlett (1989) 1;
neutron studies, see: F.H. Allen, O. Kennard, D.G. Watson, L.
Brammer, A.G. Orpen, R. Taylor, J. Chem. Soc., Perkin Trans. 2
(1987) S1.
(d) S. Kobayashi, K. Nishio, J. Org. Chem. 59 (1994) 6620;
(e) A.V. Malkov, L. Dufkova, L. Farrugia, P. Kocovsky, Angew.
Chem., Int. Ed. Engl. 42 (2003) 3674;
[16] Search in Cambridge data base revealed 35 ordered struc-
tures of vinyl silanes with terminal CH2 group, from which
˚
(f) K. Kubota, J.L. Leighton, Angew. Chem., Int. Ed. Engl. 42
(2003) 946.
only one compound has shorter Cꢀ ꢀ ꢀH contact of 2.64 A;
CH2@C(Ph)SiPh3, ref code VOLZAX, L.N. Lewis, K.G. Sy,
G.L. Bryant, P.E. Donahue, Organometallics 10 (1991)
3750.
[4] (a) C. Chuit, R.J.P. Corriu, C. Reye, J.C. Young, Chem. Rev. 93
(1993) 1371;
(b) D. Kost, I. Kalikhman, in: Z. Rappoport, Y. Apeloig (Eds.),
The Chemistry of Organic Silicon Compounds, vol. 2, Wiley,
Chichester, UK, 1998, p. 1339.
[17] According to Rowland and Taylor Fꢀ ꢀ ꢀF and Hꢀ ꢀ ꢀF intermolec-
˚
ular contacts should be as much as 2.90 and 2.54 A, respectively.
R.S. Rowland, R. Taylor, Phys. Chem. 100 (1996) 7384.
[18] C.E. Smith, P.S. Smith, R.Ll. Thomas, E.G. Robins, J.C.
Collings, C. Dai, A.J. Scott, S. Borwick, A.S. Batsanov, S.W.
Watt, S.J. Clark, C. Viney, J.A.K. Howard, W. Clegg, T.B.
Marder, J. Mater. Chem. 14 (2004) 413.
[5] A.J. Blake, S. Cradock, E.A.V. Ebsworth, K.C. Franklin, Angew.
Chem., Int. Ed. Engl. 29 (1990) 76.
[6] A.R. Bassindale, S.J. Glynn, P.G. Taylor, in: Z. Rappoport, Y.
Apeloig (Eds.), The Chemistry of Organic Silicon Compounds,
vol. 2, Wiley, Chichester, UK, 1998, pp. 495–511.
[7] (a) Pentafluorophenylboron derivatives: K. Ishihara, H. Yamam-
oto, Eur. J. Org. Chem. (1999) 527;
[19] (a) Several papers describe such examples: A.A. Korlyukov,
K.A. Lyssenko, M.Yu. Antipin, A.G. Shipov, O.A. Zamyshlya-
eva, E.P. Kramarova, Vad.V. Negrebetsky, S.A. Pogozhikh,
Yu.E. Ovchinnikov, Yu.I. Baukov, Russ. Chem. Bull. 9 (2004)
1846;
(b) J.M. Blackwell, W.E. Piers, R. McDonald, J. Am. Chem. Soc.
124 (2002) 1295;
(c) T. Schwier, M. Rubin, V. Gevorgyan, Org. Lett. 6 (2004) 1999;
(d) Pentafluorophenylphosphorous derivatives:A.L. Fernandez,
M.R. Wilson, A. Prock, W.P. Giering, Organometallics 20 (2001)
3429;
(b) N.R. Mitzel, U. Losehand, A. Wu, D. Cremer, D.W.H.
Rankin, J. Am. Chem. Soc. 122 (2000) 4471;
(c) K.A. Lyssenko, M.Yu. Antipin, D.Yu. Antonov, Chem.
Phys. Chem. 4 (2003) 817.
(e) H.J.V. Barros, M.L. Ospina, E. Arguello, W.R. Rocha, E.V.
Gusevskaya, E.N. dos Santos, J. Organomet. Chem. 671 (2003) 150;
(f) For examples of pentafluorophenyl derivatives of Ge, Sn, Pb, Ti,
and Zr, see:C. Tamborski, E.J. Sloloski, S.M. Dec, J. Organomet.
Chem. 4 (1965) 446.
[20] R.F.W. Bader, Atoms in Molecules.
Clarendron Press, Oxford, 1990.
A Quantum Theory,
[21] C.-T. Huang, P.-J. Wang, Acta Chim. Sinica 22 (1956) 123
(Chem. Abstr. 52 (1958) 5285b).
[22] N. Furuya, T. Sukawa, J. Organomet. Chem. 96 (1975) C1.
[23] R.A. Benkeser, J.M. Gaul, W.E. Smith, J. Am. Chem. Soc. 91
(1969) 3666.
[8] (a) L.A. Wall, R.E. Donadio, W.J. Pummer, J. Am. Chem. Soc.
82 (1960) 4846;
(b) P.L Coe, R. Stephens, J.C. Tatlow, J. Chem. Soc. (1962) 3227;
[24] Y. Uozumi, H. Tsuji, T. Hayashi, J. Org. Chem. 63 (1998) 6137.

A.D. Dilman et al. / Journal of Organometallic Chemistry 690 (2005) 3680–3689
3689
[25] V.F. Mironov, V.V. Nepomnina, Izv. Akad. Nauk SSSR Otd.
Khim Nauk (1960) 1419 (Bull. Acad. Sci. USSR Div. Chem. Sci.
(1960) 1318).
[27] J. Hutter, P. Ballone, M. Bernasconi, P. Focher, E. Fois, S.
Goedecker, M. Parrinello, Tuckerman (1995–2004) CPMD v.
3.7.2, MPI fur Festko¨rperforschung and IBM Zurich Research
¨
[26] A.D. Petrov, L.L. Schukovskaya, Zh. Obsh. Khim. 25 (1955) 1128
(Russ. J. Gen. Chem. 25 (1955) 1083).
Laboratory.
[28] D. Vanderbilt, Phys. Rev. B 41 (1985) 7892.