CFC replacement HCFC-133a (CF3CH2Cl) as a convenient precursor for the synthesis of chlorodifluorovinyl-metal derivatives of main group and transition metal elements: The first X-ray structural characterisation of chlorodifluorovinyl-containing organometallic complexes

Article abstract of DOI:10.1016/S0022-328X(00)00565-9

The one-pot reaction of the CFC replacement 1-chloro-2,2,2-trifluoroethane (CF₃CH₂Cl, HCFC-133a) with two equivalents of butyllithium in diethylether at -78°C followed by the addition of main group or transition metal halides results in good yields of the metal-chlorodifluorovinyl-containing compounds R₃Sn(CCl=CF₂) {R = Me, Et, Bu}, Sn(CCl=CF₂)₄, Sb(CCl=CF₂)₃, Hg(CCl=CF₂)_n Cl_(2-n) (n = 1,2), trans-[Ni(CCl=CF₂)₂(PBu₃)₂], trans-[Pd(CCl=CF₂)₂(PBu₃)₂] and [Au(CCl=CF₂)(PPh₃)]. The molecular structures of Hg(CCl=CF₂)Cl, trans-[Pd(CCl=CF₂)₂(Bu₃P)₂] and [Au(CCl=CF₂)(PPh₃)] have been obtained from single crystal data; these are the first such structural data to be reported for any chlorodifluorovinyl-containing organometallic complexes. The molecular structure of [Au(CF=CF₂)(PPh₃)], which was prepared using a similar method based on CF₃CH₂F (HFC-134a), is also reported and compared with that of the chlorodifluorovinyl analogue.

Full text of DOI:10.1016/S0022-328X(00)00565-9

www.elsevier.nl/locate/jorganchem
Journal of Organometallic Chemistry 616 (2000) 96–105
CFC replacement HCFC-133a (CF3CH2Cl) as a convenient
precursor for the synthesis of chlorodifluorovinyl-metal derivatives
of main group and transition metal elements: the first X-ray
structural characterisation of chlorodifluorovinyl-containing
organometallic complexes
Nicholas A. Barnes, Alan K. Brisdon *, Wendy I. Cross, Joseph G. Fay,
Jacqueline A. Greenall, Robin G. Pritchard, James Sherrington
Department of Chemistry, UMIST, PO Box 88, Manchester, M60 1QD, UK
Received 3 June 2000; accepted 3 August 2000
Abstract
The one-pot reaction of the CFC replacement 1-chloro-2,2,2-trifluoroethane (CF3CH2Cl, HCFC-133a) with two equivalents of
butyllithium in diethylether at −78°C followed by the addition of main group or transition metal halides results in good yields
of the metal-chlorodifluorovinyl-containing compounds R3Sn(CClꢀCF2) {R=Me, Et, Bu}, Sn(CClꢀCF2)4, Sb(CClꢀCF2)3,
Hg(CClꢀCF2)nCl(2−n) (n=1,2), trans-[Ni(CClꢀCF2)2(PBu3)2], trans-[Pd(CClꢀCF2)2(PBu3)2] and [Au(CClꢀCF2)(PPh3)]. The
molecular structures of Hg(CClꢀCF2)Cl, trans-[Pd(CClꢀCF2)2(Bu3P)2] and [Au(CClꢀCF2)(PPh3)] have been obtained from single
crystal data; these are the first such structural data to be reported for any chlorodifluorovinyl-containing organometallic
complexes. The molecular structure of [Au(CFꢀCF2)(PPh3)], which was prepared using a similar method based on CF3CH2F
(HFC-134a), is also reported and compared with that of the chlorodifluorovinyl analogue. © 2000 Elsevier Science B.V. All rights
reserved.
Keywords: HCFC-133a; Chlorodifluorovinyl; Perfluorovinyl; Synthesis; X-ray diffraction
and platinum. In 1967 the complex cis-
1. Introduction
[Pt(CClꢀCF2)2(PEt3)2] was reported from the reaction
of the bromochlorofluorocarbon CF2ꢀCClBr with
Although fluorine is capable of replacing hydrogen in
many organic systems, usually without major structural
butyllithium followed by addition of cis-[PtCl2(PEt3)2].
Shortly
afterwards
the
synthesis
of
trans-
changes, the number of examples of organometallic
complexes containing perfluorinated ligands is very
much more limited than for the comparable perprotio
analogues [1]. There is an even smaller number of
systems which contain organic ligands in which the
hydrogens are replaced by a mixture of fluorine and
chlorine substituents. A case in point is that of the
chlorodifluorovinyl ligand for which just seven metal
complexes have been reported and all but one of these
are based on the Group 10 elements, nickel, palladium
[M2Br2(CClꢀCF2)2(PEt3)2] from [M2Br4(PEt3)2] {M=
Pt,Pd} in a similar fashion was reported, and the
subsequent reaction of the platinum complex with pyri-
dine afforded [PtBr(CClꢀCF2)py(PEt3)] [2]. By com-
parison the complexes trans-[NiCl(CClꢀCF2)(PPh3)2]
[3], and trans-[NiCl(CClꢀCF2)(AsMe2Ph)2] [4] were ob-
tained from the reaction of [Ni(h2-C2H4)(PPh3)2] and
[Ni(AsMe2Ph)4], respectively, with CF2ꢀCCl2. These re-
actions are believed to proceed via initial p-coordina-
tion of the chlorofluoroalkene which subsequently
undergoes a vinyllic rearrangement to generate the
sigma-bound chlorodifluorovinyl-containing nickel
complexes of the type [NiCl(CClꢀCF2)L2].
* Corresponding author.
E-mail address: alan.brisdon@umist.ac.uk (A.K. Brisdon).
0022-328X/00/$ - see front matter © 2000 Elsevier Science B.V. All rights reserved.
PII: S0022-328X(00)00565-9

N.A. Barnes et al. / Journal of Organometallic Chemistry 616 (2000) 96–105
97
In 1968 the lithium reagent LiCClꢀCF2 was prepared
by proton abstraction from CHClꢀCF2 [5] and its ap-
plication in synthetic organic chemistry was investi-
gated; during the course of that work Et3Si(CClꢀCF2)
was prepared. Much more recently it has been shown
that the CFC-replacement 1,1,1,2-tetrafluoroethane
(CF3CH2F, HFC-134a) can act as a suitable precursor
to the perfluorovinyllithium reagent [6] and we have
used this route to synthesise a number of perfluorovinyl
main-group and transition-metal containing organo-
metallic compounds [7,8]. More recently we, and others
[9], have extended our studies to the use of HCFC-
133a (CF3CH2Cl) as a source of the chlorodifluoro-
vinyl ligand and in this paper we report our results on
the use of this industrially available material as a con-
venient precursor to chlorodifluorovinyl-containing
organometallic complexes.
Hz. The appearance of two doublets is consistent with
that expected for two non-equivalent fluorine nuclei of
the anticipated chlorodifluorovinyl moiety and the ob-
served chemical shifts are comparable with those re-
ported for the complex cis-[Pt(PEt3)2(CClꢀCF2)2]
(−80.0 and −92.2 ppm) by Rest et al. [2]. Both of the
doublets exhibit mercury satellite peaks (199Hg, I=1/2,
16.8%) from which the magnitude of J(HgF) is deter-
mined to be 246.9 and 131.6 Hz for the resonances
at −75.5 and −85.5 ppm respectively. Based on
the mercury–fluorine coupling constants the signal at
−75.5 ppm, which displays the larger Hg–F coupling,
is assigned to the fluorine nucleus trans to the mercury
centre, whilst the peak at −85.5 ppm is assigned to the
cis-fluorine. This assignment is consistent with the or-
dering of the two b-fluorine nuclei in the analogous
bis(perfluorovinyl)mercury complex [10].
The extent of substitution can be determined from
the 199Hg mercury NMR spectrum, which, in this case,
shows an overlapping triplet of triplets, centred at 933.5
ppm. The observation of a triplet-based, rather than
doublet-based, pattern confirms that bis-substitution
has occurred. Final confirmation of the identity of the
product as Hg(CClꢀCF2)2 (2) is provided by satisfac-
tory elemental analysis figures.
2. Results and discussion
The addition of two equivalents of n-butyllithium to
a diethyl ether solution of HCFC-133a at −78°C has
been reported to generate the lithium reagent
CF2ꢀCCl−Li+ (1) in good yields [9]. By analogy with
the reaction of CF3CH2F [6], this is believed to occur
via a two-step deprotonation process and concomitant
elimination of one equivalent of lithium fluoride. The
lithium reagent 1 generated in this way is stable in
solution at low temperature for many hours but decom-
poses rapidly when the temperature is increased above
−45°C. Following a number of reactions it has become
obvious that the lifetime of the lithium reagent is
considerably longer in dilute solutions than it is in more
concentrated solutions, when, even at low temperature,
some decomposition is noted and the solution darkens
in colour.
Metal-bound chlorodifluorovinyl-containing com-
plexes were obtained by the slow addition of cold
diethyl ether, or THF, solutions of the appropriate
metal halides to the lithium reagent 1. Thus, when a
cold, dilute THF solution of mercury(II) chloride was
added to a solution of 1 in a 1:2 ratio an immediate
reaction was observed with the formation of an off-
white precipitate. Subsequent slow warming of the solu-
tion to room temperature was followed by the addition
of an excess of hexane which resulted in the precipita-
tion of white material that was removed by filtration.
Elemental analysis of the solid material identified it as a
mixture of lithium halides. After drying the organic
phase the solvents were removed under vacuum to yield
a liquid product, which was purified, by distillation
under reduced pressure, to yield a colourless liquid. The
19F-NMR spectrum of a CDCl3 solution of the product
exhibited a pair of doublets at ca. −75.5 and −85.5
ppm, which show mutual coupling of magnitude 36.5
When a similar reaction was carried out using a 1:1
ratio of mercury(II) chloride to the chlorodifluorovinyl-
lithium reagent the resulting crude product was an oily
solid, the 19F-NMR spectrum of which displayed four
doublets (at −75.5, −78.4, −83.9 and −85.6 ppm),
each of which show coupling to one mercury nucleus.
Two of the doublets (−75.5 and −85.6 ppm) can be
assigned, on the basis of chemical shift position and
coupling constants, to the bis-substituted product 2.
Washing of the crude product with a little hexane
resulted in an insoluble white solid and a hexane-solu-
ble component. According to NMR spectroscopy the
solution contains the bis-substituted compound 2.
Meanwhile, the fluorine NMR spectrum of the hexane-
insoluble product, dissolved in CDCl3, consists of two
doublets at ca. −78.4 and −83.9 ppm which due to
the similarity in chemical shift position and coupling
constants [J(FF)=40.8 Hz] with those found for
Hg(CClꢀCF2)2 suggest that this product is likely to be
the desired mono-substituted complex 3. Once again
Hg–F coupling is observed to both fluorine nuclei
[ꢀJ(HgF)ꢀ=414.4 and 102.5 Hz] and on this basis the
signals at −78.1 and −83.9 ppm are assigned to the
trans- and cis-fluorine nuclei (F1 and F2, see Fig. 1)
respectively.
The 199Hg-NMR spectrum of a saturated CDCl3
solution of this product displays a doublet of doublets
centred at ca. 702 ppm; this is also consistent with the
mono-substituted complex, and confirmation of the
identity of this product as Hg(CClꢀCF2)Cl (3) was
obtained from elemental analysis results.

98
N.A. Barnes et al. / Journal of Organometallic Chemistry 616 (2000) 96–105
The solid-state structure of 3 is composed of parallel
We note that there is a significant increase in the
stacks of HgCl(CClꢀCF2) molecules separated by ca.
magnitude of the trans-Hg–F1 coupling constant in the
mono-substituted complex [J(HgF)=414 Hz] com-
pared with that recorded for the bis-substituted ana-
logue [J(HgF)=217 Hz]. A similar behaviour is
observed for the analogous perfluorovinyl compounds
where the magnitude of the trans-mercury–fluorine
coupling constant increases from 224 Hz in
Hg(CFꢀCF2)2 [10] to 401 Hz in Hg(CFꢀCF2)Cl [11].
This behaviour appears to be consistent with the re-
placement of the chloride ligand of the mono-substi-
tuted product for a, presumably, more electron-
withdrawing fluorovinyl ligand, which would result
in a decrease in the electron density in the mercury–
carbon bond which should be reflected in the reduced
coupling constants of any bond-mediated coupling
via this linkage.
On leaving chloroform solutions of compound 3 to
stand so that the solvent slowly evaporated, small
crystals were obtained. Solution of the X-ray diffrac-
tion data afforded the molecular structure of 3 which is
shown in Fig. 1, and selected bond angles and distances
which are given in Table 1. The data provide the first
structure for any organometallic chlorodifluorovinyl-
containing complex.
,
4.24 A. Within each molecule a near-linear geometry is
adopted at the mercury atom, Cl2–Hg–C1=176.1(8)°,
with Hg–Cl and Hg–C distances of 2.325(7) and
,
2.09(3) A respectively, neither of which is exceptional.
The bond distances and angles determined within the
ligand are more interesting, however because of the
dominance of the X-ray scattering from the single
heavy mercury atom, which is surrounded by much
lighter atoms, there is a relatively large degree of uncer-
tainty in the position of these atoms in the complex 3.
,
The CꢀC distance of 1.23(5) A is considerably shorter
than that found in bis(perfluorovinyl)mercury [1.362(6)
,
A] [10], however, it is not much shorter than that
reported for related complexes such as cis-1,2-difl-
,
uorovinylpentacarbonyl manganese [1.281 A] [12], and
for trans-bis(trimethylphosphine)p-tolyl(trichlorovinyl)-
,
nickel(II) [1.258 A] [13]. However, it is worth noting
that this apparently short distance may be due to a
combination of poor definition of the C1 atom and
librational effects, which cannot be corrected for in this
case. The two C–F distances in compound 3 [1.32(4)
,
and 1.37(4) A] are within experimental limits the same,
a similar situation exists for the C–Cl distances ob-
served for some trichlorovinyl-containing organometal-
lic complexes, such as the 3,4,7,8-tetramethyl-
1-10-phenanthralene adduct of Hg(CClꢀCCl2)2 [14], but
this is not the case in the solid-state structure of
Hg(CFꢀCF2)2 where a small but significant difference
between the two b-C–F bond distances [Dd=0.038(6)
,
A] is observed [10]. However, for compound 3 any such
differences may be masked by the poorer resolution
observed for the derived distances. Finally, we note that
the F1–C2–F2 angle observed for the formally sp2-hy-
bridised CF2-carbon [15] is 109(3)°.
Both compounds 2 and 3 appear to be reasonably
stable to air and moisture, but we have noticed that the
19F-NMR spectrum of solutions containing pure
Hg(CClꢀCF2)Cl which have been left aside for some
time show peaks due to both Hg(CClꢀCF2)2 and
Hg(CClꢀCF2)Cl. This suggests that redistribution of the
chloride and chlorodifluorovinyl ligands occurs. In or-
der to support this interpretation one equivalent of
HgCl2 was added directly to a solution of complex 2
and the composition of this mixture was followed by
NMR spectroscopy. Within 24 h the 19F-NMR spec-
trum displayed equally intense sets of peaks due to both
compounds 2 and 3, demonstrating that redistribution
to generate the symmetrised products does indeed occur
readily for these compounds, as it does for many other
mercury(II) complexes [16].
Fig. 1. The molecular structure of Hg(CClꢀCF2)Cl (3), thermal
ellipsoids are shown at the 30% probability level.
Table 1
,
Bond lengths (A) and angles (°) for Hg(CClꢀCF2)Cl (3)
Cl(2)–Hg(1)
Cl(1)–C(1)
Hg(1)–C(1)
F(1)–C(2)
F(2)–C(2)
C(1)–C(2)
2.325(7)
1.75(3)
2.09(3)
1.32(4)
1.37(4)
1.23(5)
C(1)–Hg(1)–Cl(2)
C(2)–C(1)–Cl(1)
C(2)–C(1)–Hg(1)
Cl(1)–C(1)–Hg(1)
C(1)–C(2)–F(1)
C(1)–C(2)–F(2)
F(1)–C(2)–F(2)
176.1(8)
122(3)
121(2)
116.7(15)
128(3)
124(3)
Previously we have shown that using CF3CH2F
(HFC-134a) it is possible to generate a range of per-
fluorovinyl-containing complexes with relative ease. In
109(3)

N.A. Barnes et al. / Journal of Organometallic Chemistry 616 (2000) 96–105
99
at −78.6 and −84.0 ppm. Both signals show coupling
to a mercury nucleus [ꢀJ(HgF)ꢀ=418 and 105 Hz re-
spectively] and on this basis the signals may be assigned
to the mono-substituted, rather than bis-substituted,
chlorodifluorovinyl mercury complex 3. We are cur-
rently undertaking a wider investigation of the chem-
istry of these compounds.
The potential utility of 1 for the synthesis of transi-
tion-metal chlorodifluorovinyl complexes was also in-
vestigated. Prototypical reactions with the late-
transition-metal halides [NiCl2(PBu3)2], [PdCl2(PBu3)2]
and [AuCl(PPh3)] were undertaken at low temperature.
In each case NMR spectra recorded after work-up
indicated that reaction had occurred. For example the
product of reaction between [NiCl2(PBu3)2] and 1 in a
1:2 stoichiometric ratio after work-up is a low melting-
point, golden-brown coloured solid, which is confirmed
as [Ni(PBu3)2(CClꢀCF2)2] (9) by elemental analysis. The
31P-NMR spectrum of a sample dissolved in CDCl3
exhibits a triplet at 10.3 ppm [J(PF)=6.5 Hz] which is
shifted slightly from that observed for the starting
metal complex. The 19F-NMR spectrum shows a dou-
blet [J(FF)=75 Hz] of triplets [J(PF)=6.5 Hz] at
−72.8 ppm and a doublet [J(FF)=75 Hz)] at −96.6
ppm. The observation of triplet-based signals in the
fluorine NMR spectrum due to phosphorus–fluorine
coupling suggests that the two phosphines are equiva-
lent. The fact that P–F coupling is observed on only
one of the two signals in the fluorine NMR spectrum is
used to assign the resonance at −72.8 ppm to the
fluorine nucleus trans to the metal-coordinated phos-
phine (F1).
A similar reaction was performed between 1 and
trans-[PdCl2(PBu3)2] in a 2:1 ratio which, after work-
up, yielded a solid yellow product. The 31P-NMR spec-
trum of this compound exhibited a multiplet at ca. 9.4
ppm, which is, again, only slightly shifted from that of
the starting complex at l=11.1 ppm. This multiplet
appears to be a quintet, with a measured coupling
constant of 5 Hz, which presumably arises from cou-
pling to both the F1 and F2 fluorine nuclei of two
chlorodifluorovinyl groups with coupling constants of
similar magnitude. The 19F-NMR spectrum shows two
resonances at −77.5 and −95.8 ppm. Both of these
signals show mutual F–F coupling of 72 Hz and a
smaller triplet coupling of 5 Hz corresponding to
J(PF). These observations are consistent with a bis-sub-
stituted product which is confirmed as [Pd-
(CClꢀCF2)2(PBu3)2] (10) by elemental analysis.
Scheme 1. Reactions of chlorodifluorovinyllithium with a variety of
metal halides.
order to determine the extent with which HCFC-133a
may be used as a general way of introducing the
–CClꢀCF2 moiety into organometallic complexes, reac-
tions with a variety of main-group and transition-metal
starting materials were investigated, as shown in
Scheme 1.
The reactions of 1 with R3SnCl (R=Me, Et, Bu),
SnCl4 and SbI3 were undertaken as representatives of
main-group organometallic precursors. In all cases after
addition of the halide, but prior to work-up, a sample
was withdrawn from the reaction mixture and 19F-
NMR spectra were recorded. These spectra showed
that the signal at ca. −73 ppm due to CF3CH2Cl had
been replaced with two doublets within the chemical
shift range −65 to −90 ppm. The positions of these
signals and the magnitude of the coupling between
them are similar to that observed for compounds 2 and
3, and are indicative that chlorodifluorovinyl complexes
have been prepared. After work-up, mobile liquid prod-
ucts were isolated in yields of 70%, or better, which
analysed as R3Sn(CClꢀCF2) [R=Me (4), Et (5), Bu
(6)], Sn(CClꢀCF2)4 (7) and Sb(CClꢀCF2)3, (8). For
compound 7 the elemental analysis figures suggest that
a small amount of solvent is associated with the final
compound, and this is confirmed by the observation of
low-intensity signals in the proton NMR spectrum; this
corresponds to hexane. Even after redistillation of this
material a small amount of solvent appears to remain
associated with this compound. Although it was not
possible to obtain crystallographic data for any of these
liquid materials, unequivocal identification of all the
compounds was obtained by a combination of spectro-
scopic methods and elemental analyses.
In all cases the products obtained appear to be
neither air nor moisture sensitive. Indeed, after shaking
with water the starting compound can be recovered.
The tin-containing materials also appear to be able to
act as effective transfer reagents of the chlorodifl-
uorovinyl group. For example, when HgCl2 was treated
with one equivalent of Bu3Sn(CClꢀCF2) in ethanol the
subsequent NMR data confirmed that a reaction had
taken place. The signals due to Bu3Sn(CClꢀCF2) were
no longer present, instead two doublets were observed
Slow evaporation of the solvents from a mixed sol-
vent (dichloromethane and hexane) solution of 10 re-
sulted in crystals suitable for X-ray diffraction studies.
Solution of the data resulted in the molecular geometry
shown in Fig. 2 and selected bond distances and angles
are presented in Table 2. A trans-arrangement of the
pairs of ligands in an essentially square-planar geome-

100
N.A. Barnes et al. / Journal of Organometallic Chemistry 616 (2000) 96–105
try at the palladium centre is observed with C1–Pd–
P1=90.8(3)°. This is consistent with the interpretation
of the solution-phase NMR data. The Pd–P and Pd–C
bond lengths were determined as 2.294(3) and 2.036(10)
,
A respectively. These distances are not very different
from those found in related systems; the Pd–P distance
is a little shorter than the average of a series of trans-
phosphine palladium(II) complexes from the Cam-
,
bridge Structural database at 2.331 A, and the Pd–C
distance is also just below the average palladium(II)–C
,
distance of 2.058 A [17]. Within the chlorodifluorovinyl
ligand disorder is observed around the metal-bound
carbon atom (C1) with two alternative orientations for
the CF2 and Cl parts of this ligand. This resulted in two
Fig. 3. The molecular structure of [Au(CClꢀCF2)PPh3] (11), thermal
elipsoids are shown at the 30% probability level and hydrogen atoms
have been omitted for clarity.
equally populated sites for the C2 and Cl atoms and
subsequently two alternative sites for each of the
fluorine atoms. Similar disorder problems have been
reported previously within the vinyl section of the
related systems trans-[PtCl(PEt2Ph)2(CHꢀCH2)] [18]
and trans-[Ni(CClꢀCCl2)2(PMe2Ph)2] [19]. Within ex-
perimental limits the bond lengths obtained from the
chlorodifluorovinyl groups in the two sites are the
same, 1.26(2) [d(CꢀC)], 1.846(17) [d(C–Cl)] and 1.29(2)
,
[d(C2–F1)] and 1.35(3) A [d(C2–F2)]. Although the
two C–F distances are different this is only just signifi-
cant at the 2| level. The CꢀC and C–Cl distances are
both slightly longer than those observed for compound
3 and the F1–C–F2 angle is again significantly smaller
than 120° [109.9(18)°].
Fig. 2. The molecular structure of trans-[Pd(CClꢀCF2)2(PBu3)2] (10),
thermal ellipsoids are shown at the 30% probability level and hydro-
gen atoms have been omitted for clarity. Disorder was observed for
the chlorodifluorovinyl ligand resulting in two equally populated sites
for atoms C2, F1, F2 and Cl1, for clarity only one site for each
moiety is shown.
The reaction between 1 and [AuCl(PPh3)] was also
undertaken. After work-up a light brown coloured solid
was obtained which is light sensitive, but otherwise
stable. Elemental analysis suggests a formulation of this
complex as [Au(CClꢀCF2)(PPh3)] (11), and this is sup-
ported by NMR data. The 31P-NMR spectrum consists
of a doublet at 42.5 ppm (J=18 Hz) and the 19F-NMR
spectrum exhibits a doublets of doublets at −79.1 ppm
and a doublet at −89.9 ppm. Both signals show mu-
tual fluorine–fluorine (J=53 Hz) whilst only one of
them (the signal at −79.1 ppm) shows additional cou-
pling (J=18 Hz) to the phosphorus nucleus. Crystals
of the product suitable for X-ray diffraction studies
were obtained by slow evaporation of the solvent from
a chloroform–hexane solution in the dark. Solution of
the data resulted in a gold complex exhibiting a two-co-
ordinate, near-linear [P–Au–C=178.7(3)°] structure as
shown in Fig. 3. The Au–P distance, Table 3, is
Table 2
,
Selected bond lengths (A) and angles (°) for trans-[Pd-
(CClꢀCF2)2(PBu3)2] (10) a
Pd(1)–C(1)
Pd(1)–P(1)
Cl(1)–C(1)
C(2)–C(1)
C(2)–F(1)
C(2)–F(2)
2.036(10)
2.294(3)
1.846(17)
1.26(2)
1.29(2)
1.35(3)
C(1)–Pd(1)–C(1)c1
C(1)–Pd(1)–P(1)
C(1)c1–Pd(1)–P(1)
C(1)–C(2)–F(1)
180.000
90.8(3)
89.2(3)
134(2)
C(1)–C(2)–F(2)
115(2)
F(2)–C(2)–F(1)
109.9(18)
109.2(13)
134.3(12)
116.4(6)
C(2)–C(1)–Cl(1)
C(2)–C(1)–Pd(1)
Cl(1)–C(1)–Pd(1)
,
,
2.266(2) A and the Au–C distance is 2.026(10) A. Once
again these distances are slightly shorter than the aver-
a Symmetry transformations used to generate equivalent atoms:
,
c1 −x+2, −y, −z+2.
age of those found in related complexes; 2.277 A for the

N.A. Barnes et al. / Journal of Organometallic Chemistry 616 (2000) 96–105
101
,
1.731(10), 1.341(12) and 1.319(14) A respectively and
the F1–C–F2 angle is 106.9(10)°. The two C–F dis-
tances are just within 2| of each other, although it is
interesting to note that in this case it is the C–F bond
trans to the metal which is the longer; this is the reverse
of the situation found for the mercury complexes
Hg(CClꢀCF2)Cl and Hg(CFꢀCF2)2 [10] but is consis-
tent with the distances obtained for 10.
average gold(I)–triphenylphosphine distance and 2.097
,
A for the average Au–C distance [17]. Within the
chlorodifluorovinyl ligand the CꢀC bond length is
,
1.281(15) A, the C–Cl and C–F distances are
Table 3
,
Selected bond distances (A) and angles (°) for [Au(CClꢀCF2)PPh3]
(11)
These variations in C–F distances and other geomet-
ric parameters led us to try to obtain crystals for
analogous perfluorovinyl complexes of the chlorodi-
fluorovinyl complexes reported here with which direct
comparisons may be made. The perfluorovinyl-contain-
ing analogues were synthesised using a similar method
to that described above, but based on CF3CH2F (HFC-
134a) from which perfluorovinyl lithium may be gener-
ated. Unfortunately most of the analogous materials
were obtained as liquids or oily solids; it was, however,
possible to obtain crystals suitable for structural char-
acterisation for the perfluorovinyl-containing gold com-
plex by the slow evaporation of the solvent from a
dilute solution of [Au(CFꢀCF2)PPh3] (12) in chloro-
form and hexane.
Au(1)–C(1)
Au(1)–P(1)
Cl(1)–C(1)
F(1)–C(2)
F(2)–C(2)
C(1)–C(2)
2.026(10)
2.266(2)
1.731(10)
1.341(12)
1.319(14)
1.281(15)
C(1)–Au(1)–P(1)
C(2)–C(1)–Cl(1)
Cl(1)–C(1)–Au(1)
C(1)–C(2)–F(2)
C(1)–C(2)–F(1)
F(2)–C(2)–F(1)
178.7(3)
114.5(8)
120.6(6)
125.2(10)
127.9(12)
106.9(10)
Solution of the data resulted in a similar structure to
that found for the comparable chlorodifluorovinyl-con-
taining complex 11. That is a near-linear geometry is
observed at the gold atom [C1–Au–P=178.1(3)°] with
Au–C1 and Au–P bond distances of 2.028(9) and
,
2.272(3) A respectively (Fig. 4, Table 4). In this respect
there appears to be little difference in the trans-effect of
the chlorodifluorovinyl and perfluorovinyl ligands.
Within the perfluorovinyl ligand the CꢀC distance is
,
1.297(14) A and the three C–F distances are 1.350(11)
,
[C1–F3], 1.304(13) [C2–F2] and 1.342(15) [C2–F1] A.
There is, therefore, no significant difference between the
CꢀC and C–F bond distances found in the chlorodi-
fluorovinyl and perfluorovinyl moieties in complexes 11
and 12. The two b-C–F distances differ significantly
from each other, and this behaviour is consistent with
the parameters obtained for Hg(CFꢀCF2)2 in both the
solid and gas phases via single-crystal X-ray diffraction
and gas-phase electron diffraction studies [10]. How-
ever, for the gold complexes the effects are reversed; for
Hg(CFꢀCF2)2 the carbon–fluorine bond cis to the
Fig. 4. The molecular structure of [Au(CFꢀCF2)PPh3] (12), thermal
elipsoids are shown at the 30% probability level and hydrogen atoms
have been omitted for clarity.
Table 4
,
Selected bond distances (A) and angles (°) for [Au(CFꢀCF2)PPh3]
(12)
metal
is
the
longer
whereas
for
both
Au(1)–C(1)
Au(1)–P(1)
F(3)–C(1)
F(1)–C(2)
F(2)–C(2)
C(1)–C(2)
2.028(9)
2.272(3)
1.350(11)
1.342(15)
1.304(13)
1.297(14)
[Au(CFꢀCF2)(PPh3)] and [Au(CClꢀCF2)(PPh3)] the op-
posite situation exists.
Currently it is not possible to assess, in detail, the
factors that give rise to the variation in C–F bond
lengths within the fluorovinyl groups. This is principally
due to the very limited number of examples of such
compounds which have been structurally characterised
to date. It is, however, obvious on close inspection of
the data that there do not appear to be any particularly
close intermolecular interactions which may account for
these observations. It would also appear, from the data
C(1)–Au(1)–P(1)
C(2)–C(1)–Cl(1)
Cl(1)–C(1)–Au(1)
C(1)–C(2)–F(2)
C(1)–C(2)–F(1)
F(2)–C(2)–F(1)
178.1(3)
113.5(9)
117.2(7)
125.7(12)
125.4(10)
108.9(10)

102
N.A. Barnes et al. / Journal of Organometallic Chemistry 616 (2000) 96–105
Table 5
Crystallographic data for complexes 3, 10, 11 and 12
Compound
Hg(CClꢀCF2)Cl (2) Trans-[Pd(CClꢀCF2)2
[Au(CClꢀCF2)
PPh3] (11)
[Au(CFꢀCF2)
PPh3] (12)
(PBu3)2] (9)
Formula
M
C2Cl2F2Hg
333.51
C28H54Cl2F4P2Pd
705.95
C20H15AuClF2P
556.71
C20H15AuF3P
540.26
Crystal system
Triclinic
Monoclinic
P2(1)/n
Monoclinic
P2(1)/c
Orthorhombic
P212121
(
P1
Unit cell parameters
,
a (A)
4.239(3)
6.146(2)
11.219(7)
100.06(3)
95.93(5)
96.24(4)
283.8(3)
2
9.4885(5)
8.544(5)
22.173(11)
90
98.93(5)
90
1775.8(16)
2
1.320
8.7800(19)
13.081(4)
16.631(4)
90
102.105(17)
90
1867.6(8)
4
1.980
11.403(13)
11.736(14)
13.433(17)
90
90
90
1797.6(4)
4
1.996
,
b (A)
,
c (A)
h (°)
i (°)
k (°)
3
,
U (A )
Z
D (g cm−3
)
3.903
27.964
v(Mo–Ka) (mm−1
)
0.799
8.124
8.301
Crystal size (mm)
No. data collected (q range)
F(000)
0.20×0.10×0.05
718 (3.39–24.45)
288
0.15×0.10×0.10
3320 (1.86–24.98)
736
0.15×0.10×0.10
3509 (2.00–24.96)
1056
0.30×0.25×0.20
1806 (2.30–24.96)
1024
Maximum, minimum transmission
Temperature (K)
0.992, 0.953
293(2)
0.9243, 0.8895
203(2)
0.4972, 0.3754
203(2)
0.2876, 0.1897
203(2)
R1, wR2 [I\2|(I)]
R1 (all data)
0.0725, 0.1686
0.0804
0.0687, 0.1179
0.1717
0.0447, 0.1068
0.0651
0.0315, 0.0.623
0.0411
Maximum, minimum residual electron density
3.034, −2.252
0.594, −0.660
1.646 and −1.440
1.029 and −0.984
−3
,
(e A
)
obtained for the gold complexes, that the substitution
of the halogen on the a-carbon of the fluorovinyl group
does not have a major impact on the structure and
bond lengths within the remainder of the molecule.
In conclusion, we report a high-yielding, one-pot
synthetic method based on CF3CH2Cl (HCFC-133a)
for the synthesis of chlorodifluorovinyl-containing
organometallic complexes via the intermediacy of the
lithium reagent Li+CClꢀCF2 (1). This method provides
a reasonably straightforward route to these complexes
and we have shown that this route appears to be of
general utility. We have prepared and characterised a
number of main group and transition metal chlorodi-
fluorovinyl-containing complexes and report the single-
crystal X-ray derived structures of three of these; this
provides the first crystallographic data for complexes of
this type.
wire for ca. 1 day and subsequently refluxed over
sodium–benzophenone under a dinitrogen atmosphere.
Hexane was stored over sodium wire prior to use.
Fluorine, proton, phosphorus and mercury NMR spec-
tra were recorded on solutions of samples on a Bruker
DPX200 instrument at 188.3, 200.2, 81.0 and 35.8 MHz
and referenced against CFCl3, TMS, 85% H3PO4 and
HgMe2 respectively using the high-frequency-positive
convention. Elemental analyses were performed by the
analytical service in the department. IR spectra were
recorded on a Nicolet PC5 or Nexus instrument. All
reactions were carried out under anaerobic and anhy-
drous conditions. All glassware was flame dried prior to
use, and moisture-sensitive reagents were handled under
an argon atmosphere in a dry-box (Belle Technology,
UK). All recorded reaction temperatures are uncor-
rected temperatures measured using an internal
thermocouple.
3. Experimental
3.1. X-ray data for compounds 10, 11 and 12
HCFC-133a, HFC-134a (both ICI Klea), butyl-
lithium (2.5 M in hexane, Aldrich) and metal halides
were used as supplied after verification of purity. Plat-
inum, palladium and gold starting complexes were pre-
pared from the appropriate halometallate salt (Johnson
Matthey) based on literature methods [20]. Diethyl
ether and THF were dried by standing over sodium
X-ray data for compounds 9, 11a and 11b were
collected at low temperature on a Nonius MACH3
diffractometer, details are given in Table 5. The data
for complex 3 were collected at room temperature at
the EPSRC X-ray service using a Nonius Kappa CCD
area detector mounted on a Nonius FR591 diffrac-
tometer. The data for all the compounds was solved

N.A. Barnes et al. / Journal of Organometallic Chemistry 616 (2000) 96–105
103
using direct methods (SHELXS [21]). The structures were
refined by full-matrix least-squares on F2o, using the
program SHELX-97. All data used were corrected for
Lorentz-polarisation factors and subsequently for ab-
sorption by the c-scan method, except for 3, where the
multscan method was used. The non-hydrogen atoms
were refined with anisotropic thermal parameters. All
hydrogen atoms in the compounds were included in
idealised positions and were refined isotropically. For
compound 10 disorder was observed in the chlorodi-
fluorovinyl part of the complex and this was modelled
by including two equally occupied split sites for the
ligand and refining this isotropically.
−80°C), was added slowly to the reaction mixture
which was held at −110°C. After the reaction was
complete it was worked up as previously described to
yield an oily solid material which was washed with
hexane to leave a white solid. Yield 2.14 g, 87%. m.p.
(uncorrected) 150°C dec. (Anal. Found: C, 7.6; H, 0.0;
Cl, 20.8. Hg(CClꢀCF2)Cl requires: C, 7.2; H, 0.0;
Cl, 21.3%.) lF(CDCl3), −78.4 [d, J(FF)=40.8,
J(HgF)=416.0 Hz], −83.9 [d, J(FF)=40.8, J(HgF)=
102.5 Hz]. l199Hg (CDCl3), 701.8 [dd, J(HgF1)=
414.4, J(HgF2)=100.7 Hz. wmax (cm−1
solution) 1699 (CꢀC), 1273, 995 (C–F).
) (CHCl3
3.2. Preparation of bis(chlorodifluoro6inyl)mercury (2)
3.4. Trimethyl(chlorodifluoro6inyl)tin (4)
A 250 ml three-neck flask held at −80°C was
equipped with a magnetic stirrer, a rubber septum and
a silicon oil bubbler and maintained under a positive
flow of dinitrogen gas. The flask was charged with
pre-cooled diethyl ether (200 ml, −50°C) and liquid
HCFC-133a (d=1.389 g cm−3, 0.80 cm3, 9.36 mmol)
was added. Butyllithium (2.5 M in hexane, 7.52 cm3,
18.8 mmol) was added slowly to the stirred solution
over a 0.5 h period. Once addition of the BuLi was
complete the reaction temperature was maintained be-
tween −55°C and −80°C for 2 h to ensure generation
of CF2ꢀCF−Li+. After lowering the reaction mixture
temperature to −110°C mercury(II) chloride (1.26 g,
4.64 mmol), dissolved in THF (200 ml, −80°C), was
added slowly. The reaction was left to warm to −35°C
over a period of a few hours and then allowed to attain
room temperature by removal of the slush bath. The
solvent volume was reduced in vacuo to half and
hexane (300 ml) was added resulting in the precipitation
of lithium salts, from the organic phase. The mixture
was allowed to settle before being carefully filtered
through a No. 4 sinter under a dinitrogen atmosphere.
The solvent was then removed in vacuo to leave a
brown liquid which was subsequently washed with hex-
ane (4×10cm3), and vacuum distilled (180–185°C, 10
mmHg) to afford 2 as a clear liquid. Yield 1.61 g, 88%.
(Anal. Found: C, 12.9; H, 0.2; Cl, 20.1. Hg(CClꢀCF2)2
requires: C, 12.2; H, 0.0; Cl, 18.0%.) l19F(CDCl3),
−75.5 [d, J(FF)=44, J(HgF)=247 Hz], −85.2 [d,
J(FF)=44, J(HgF)=131 Hz]. l199Hg (CDCl3), 701.8
In a similar procedure to that outlined above, cold
liquid HCFC-133a (0.86 cm3, 10.0 mmol) in hexane
(150 ml) was added to butyllithium (2.5 M in hexane,
8.0 cm3, 20.0 mmol) over a 0.5 h period. After 2 h the
temperature of the reaction mixture was lowered to
−100°C and trimethyltinchloride (1 M in CH2Cl2, 10.0
cm3, 10.0 mmol) was added slowly. After warming to
room temperature the solvent volume was reduced and
hexane was added. The mixture was filtered under a
dinitrogen atmosphere and the volatile organic solvents
were removed in vacuo. Trap to trap distillation under
reduced pressure (0.2 mmHg) resulted in the separation
of the product as a clear liquid. Yield 2.12 g, 81%.
Mass spec., EI, m/z=262 (M+, 5%), 165 (Me3Sn+,
10%). l19F (CDCl3), −72.0 [d, J(FF)=42 Hz,
J(SnF)=6.4 Hz], −87.2 [d, J(FF)=42 Hz]. l1H
(CDCl3), 0.39 [s, J(117SnH)=56.1 Hz, J(119SnH)=58.6
Hz]. wmax (cm−1) (neat liquid) 1685 (CꢀC), 1244, 982
(C–F).
3.5. Triethyl(chlorodifluoro6inyl)tin (5)
A preparative route similar to that for 4 outlined
above was used. BuLi (2.5 M, 8.4 cm3, 21 mmol) was
added to a solution of HCFC-133a (0.90 cm3, 10.5
mmol) in diethyl ether (60 cm3, −80°C). After 2 h
Et3SnBr (3.00 g, 10.5 mmol) dissolved in diethyl ether
(60 cm3, −80°C) was added slowly to the reaction
mixture which was held at −110°C. The reaction was
left to warm slowly to room temperature overnight.
Hexane (100 cm3) was added and the solution was
filtered. The solvents were removed on a rotary evapo-
rator to leave a light brown coloured liquid. Distillation
at 182°C/0.2 mmHg resulted in the separation of the
product as a clear liquid. Yield 2.68 g, 84%. (Anal.
Found: C, 33.7; H, 5.3. Et3Sn(CClꢀCF2) requires: C,
31.7; H, 5.0%.) l19F (CDCl3), −72.2 [d, J(FF)=46
Hz], −87.2 [d, J(FF)=46 Hz, J(SnF)=14 Hz]. wmax
(cm−1) (neat liquid) 1686 (CꢀC), 1243, 980 (C–F).
[dd, J(HgF1)=251, J(HgF2)=135 Hz]. wmax (cm−1
(neat liquid) 1686 (CꢀC), 1258, 992 (C–F).
)
3.3. Chloro(chlorodifluoro6inyl)mercury (3)
In an analogous method to that described for 2, BuLi
2.5 M (5.9 cm3, 14.75 mmol) was slowly added to a
solution of HCFC-133a (0.63 cm3, 7.38 mmol) in di-
ethyl ether (180 ml, −80°C). After approximately 2 h
HgCl2 (2.00 g, 7.36 mmol), dissolved in THF (250 ml,

104
N.A. Barnes et al. / Journal of Organometallic Chemistry 616 (2000) 96–105
3.6. Tributyl(chlorodifluoro6inyl)tin (6)
Yield 1.72 g, 69%. (Anal. Found: C, 18.0; H 0.0;
Cl, 25.9%. Sb(CClꢀCF2)3 requires: C, 17.4; H, 0.0;
Cl, 25.7%.) l19F (CDCl3), −67.8 [d, J(FF)=23 Hz],
−79.5 [d, J(FF)=23 Hz]. wmax (cm−1) (neat liquid)
1680 (CꢀC), 1271, 992 (C–F).
A preparative route similar to that for 4 outlined
above was used. BuLi (2.5 M, 12.0 cm3, 30 mmol) was
added to a solution of HCFC-133a (1.30 cm3, 15.2
mmol) in diethyl ether (60 cm3, −80°C). After 2 h
Bu3SnCl (5.00 g, 15.4 mmol) dissolved in diethyl ether
(100 cm3, −80°C) was added slowly to the reaction
mixture which was held at −110°C. The reaction was
left to warm slowly to room temperature overnight.
Hexane (100 cm3) was added and the solution was
filtered. The solvents were removed on a rotary evapo-
rator to leave a light brown coloured liquid. Distillation
at 242°C/0.2 mmHg resulted in the separation of the
product as a clear liquid. Yield 4.88 g, 84%. (Anal.
Found: C, 44.5; H, 6.9; Cl, 8.1. Bu3Sn(CClꢀCF2) re-
quires: C, 43.4; H, 7.0; Cl, 9.2%.) l19F (CDCl3), −72.2
[d, J(FF)=47 Hz], −87.2 [d, J(FF)=47, J(SnF)=20
Hz]. wmax (cm−1) (neat liquid) 1680 (CꢀC), 1240, 980
(C–F).
3.9. [bis(Tributylphosphine)bis(chlorodifluoro6inyl)nickel]
(9)
[bis(Tributylphosphine)nickeldichloride] was pre-
pared by the reaction of nickel dichloride with two
equivalents of tributylphosphine in ethanol. In a 250 ml
round bottom flask equipped as described previously
BuLi (2.5 M, 1.6 cm3, 4 mmol) was added to a solution
of HCFC-133a (0.17 cm3, 2 mmol) in diethyl ether (180
cm3, −80°C). After 2 h [NiCl2(PBu3)2] (0.53 g, 1 mmol)
dissolved in THF(25 cm3, −80°C) was added slowly to
the reaction mixture the temperature of which was held
at −110°C. The reaction was left to warm slowly to
room temperature overnight. Hexane (100 cm3) was
added and the solution was filtered. The solvents were
removed on a rotary evaporator to leave a light yellow
coloured oil. Yield 0.24g, 36%, m.p. (uncorrected) 84–
88°C, dec. (Anal. Found: C, 51.7; H, 8.4; F, 10.9.
[Ni(PBu3)2(CClꢀCF2)Cl] requires: C, 51.1; H, 8.2; F,
11.5%.) lP{1H}(CDCl3), 10.3 ppm [d, J(PF)=6.5 Hz].
l19F (CDCl3) −72.8 [dt, J= 6.5, 75 Hz], −96.6 [d,
J= 75 Hz]. wmax (cm−1) (nujol mull) 1684 (CꢀC), 1197,
954 (C–F).
3.7. Tetrakis(chlorodifluoro6inyl)tin (7)
A preparative route similar to that for 4 outlined
above was used. BuLi (2.5 M, 63.6 cm3, 159 mmol) was
added over a 3 h period to a solution of HCFC-133a
(6.8 cm3, 79.7 mmol) in diethyl ether (150 cm3,
−80°C). After 2 h SnCl4 (5.00 g, 19.2 mmol) dissolved
in THF (100 cm3, −80°C) was added slowly to the
reaction mixture which was held at −110°C. The reac-
tion was left to warm slowly to room temperature
overnight. Hexane (100 cm3) was added and the solu-
tion was filtered. The solvents were removed on a
rotary evaporator to leave a light brown coloured
liquid. Distillation at 120°C/0.2 mmHg resulted in the
separation of the product as a clear liquid. Yield 7.52 g,
77%. (Anal. Found: C, 20.1; H, 0.1; Cl, 26.9.
Sn(CClꢀCF2)4 requires: C, 18.9; H 0.0; Cl, 27.9%.) l19F
(CDCl3), −66.8 [d, J(FF)=20 Hz, J(SnF)=69 Hz],
−79.2 [d, J(FF)=20 Hz]. wmax (cm−1) (neat liquid)
1678 (CꢀC), 1271, 1001 (C–F).
3.10. trans-[bis(Tributylphosphine)bis-
(chlorodifluoro6inyl)palladium] (10)
Trans - [bis(tributylphosphine)palladiumdichloride]
was prepared by the reaction of disodium tetra-
chloropalladate with two equivalents of tributylphos-
phine in water. In a 250 ml round bottom flask
equipped as described previously BuLi (2.5 M, 8.8 cm3,
22 mmol) was added to a solution of HCFC-133a (0.95
cm3, 11 mmol) in diethyl ether (180 cm3, −80°C). After
2 h [(PBu3)2PdCl2] (3.00 g, 5.2 mmol) dissolved in THF
(25 cm3, −80°C) was added slowly to the reaction
mixture which was held at −110°C. The reaction was
left to warm slowly to room temperature overnight
after which hexane (100 cm3) was added and the solu-
tion was filtered. The solvents were removed on a
rotary evaporator to leave a light yellow–brown
coloured solid. Yield 2.75 g, 75%, m.p. (uncorrected)
91–93°C, dec. (Anal. Found: C, 47.9; H, 7.6; F, 10.4.
[Pd(PBu3)2(CClꢀCF2)2] requires: C, 47.6; H, 7.7; Cl,
10.0; F, 10.8%.) lP{1H}(CDCl3), 9.4 [quintet, J=5
Hz]. l19F (CDCl3), −77.5 [ddt, J=72, 25, 5 Hz],
−95.8 [ddt, J=72, 25, 5 Hz]. wmax (cm−1) (nujol mull)
1686 (CꢀC), 1202, 955 (C–F).
3.8. tris(Chlorodifluoro6inyl)antimony (8)
A preparative route similar to that for 4 outlined
above was used. BuLi (2.5 M, 14.5 cm3, 36.3 mmol)
was added to a solution of HCFC-133a (1.55 cm3, 18.2
mmol) in diethyl ether (60 cm3, −80°C). After 2 h SbI3
(3.00 g, 5.98 mmol) dissolved in THF (60 cm3, −80°C)
was added slowly to the reaction mixture which was
held at −110°C. The reaction was left to warm slowly
to room temperature overnight. Hexane (80 cm3) was
added and the solution was filtered. The solvents were
removed on a rotary evaporator to leave a brown
coloured liquid. Distillation at 195°C/40 mmHg re-
sulted in the separation of the product as a clear liquid.

N.A. Barnes et al. / Journal of Organometallic Chemistry 616 (2000) 96–105
105
3.11. [Triphenylphosphine(chlorodifluoro6inyl)gold] (11)
Acknowledgements
Triphenylphosphinegold chloride was prepared by the
reaction of sodium tetrachloroaurate with triphenylphos-
phine in ethanol [20]. In a 250 ml round bottom flask
equipped as described previously BuLi (2.5 M, 4.9 cm3,
12.3 mmol) was added to a solution of HCFC-133a (0.52
cm3, 6.1 mmol) in diethyl ether (180 cm3, −80°C). After
2 h [AuCl(PPh3)] (3.02 g, 6.1 mmol) dissolved in THF
(25 cm3, −80°C) was added slowly to the reaction
mixture which was maintained at −110°C. The reaction
was left to warm slowly to room temperature overnight.
Hexane (100 cm3) was added and the solution was
filtered. The solvent2s were removed on a rotary evapo-
rator to leave a light brown coloured solid. Yield 2.54
g, 75%. (Anal. Found: C, 43.0; H, 3.0; F, 6.7. [Au(PPh3)-
(CClꢀCF2)] requires: C, 43.2; H, 2.7; F, 6.8%.) l31P-
{1H}(CDCl3), 42.5 ppm [m]. l19F (CDCl3), −79.1 [dd,
J= 53, 18 Hz], −89.9 [d, J=53 Hz]. wmax (cm−1) (nujol
mull) 1670 (CꢀC), 1219, 963 (C–F).
We acknowledge ICI Klea for the donation of CFC-
replacements, Johnson Matthey for the loan of metal
salts, the Royal Society for a research grant and the
EPSRC for the provision of grants for UMIST’s NMR
(GR/L52246) and FT-IR/Raman (GR/M30135) facili-
ties, also for access to the X-ray structural database,
and to the national X-ray Crystallography Data Collec-
tion Service, for compound 3.
References
[1] R.P. Hughes, Adv. Organomet. Chem. 31 (1990) 183.
[2] A.J. Rest, D.T. Rosevear, F.G.A. Stone, J. Chem. Soc. A
(1967) 66.
[3] J. Ashley-Green, M. Green, F.G.A. Stone, J. Chem. Soc. A
(1969) 3019.
[4] J. Browning, M. Green, F.G.A. Stone, J. Chem. Soc. A (1971)
453.
[5] F.G. Drakesmith, R.D. Richardson, O.J. Stewart, P. Tarrant,
J. Org. Chem. 33 (1968) 286.
3.12. [Triphenylphosphine(perfluoro6inyl)gold] (12)
[6] J. Burdon, P.L. Coe, I.B. Haslock, R.L. Powell, Chem. Com-
mun. (1996) 46.
[7] K.K. Banger, A.K. Brisdon, A. Gupta, Chem. Commun.
(1997) 139.
[8] K.K. Banger, A.K. Brisdon, J. Organomet. Chem. 582 (1999)
301.
[9] J.M. Bainbridge, S.J. Brown, P.N. Ewing, R.R. Gibson, J.M.
Percy, J. Chem. Soc. Perkin Trans. 1 (1998) 2541.
[10] K.K. Banger, A.K. Brisdon, P.T. Brain, S. Parsons, D.W.H.
Rankin, H.E. Robertson, B.A. Smart, M. Bu¨hl, Inorg. Chem.
38 (1999) 5894.
[11] K.K. Banger, PhD Thesis, UMIST, 1998.
[12] F.W.B. Einstein, H. Luth, J. Trotter, J. Chem. Soc. A (1967)
89.
[13] J. Chen, Y. Kai, N. Kasai, M. Wada, M. Kumazoe, Bull.
Chem. Soc. Jpn. 64 (1991) 2802.
[14] N.A. Bell, I.W. Nowell, Acta Crystallogr. Sect. B 36 (1980)
447.
This compound was prepared in a similar way to that
described above for 12. BuLi (2.5 M, 5.2 cm3, 12 mmol)
was added to a solution of HFC-134a (d=1.21 g cm−1
,
0.5 cm3, 6 mmol) in diethyl ether (180 cm3, −80°C).
After 2 h [AuCl(PPh3)] (2.97 g, 6.0 mmol) dissolved in
THF (25 cm3, −80°C) was added slowly to the reaction
mixture which was held at −110°C. The reaction was
left to warm slowly to room temperature overnight,
hexane (100 cm3) was added and the solution was filtered.
The solvents were removed on a rotary evaporator to
leave an off-white coloured solid. Yield 1.95 g, 60%, m.p.
(uncorrected) 86–88°C, dec. (Anal. Found: C, 44.4; H,
4.1; F, 10.9. [Au(PPh3)(CFꢀCF2)] requires: C, 44.5; H,
2.8; F, 10.6%.) l31P{1H} (CDCl3) 42.2 ppm [m]. l19F
(CDCl3), −95.5 [ddd, J=14, 35, 87 Hz], −127.3 [ddd,
J=4, 87, 99 Hz], −177.0 ppm [ddd, J=35, 69, 99 Hz].
wmax (cm−1) (nujol mull) 1698 (CꢀC), 1248, 1100, 986
(C–F).
[15] A.H. Lowrey, P. D’Antonio, C. George, J. Chem. Phys. 64
(1976) 2884.
[16] A.J. Bloodworth, in: C.A. McAuliffe (Ed.), The Chemistry of
Mercury, MacMillan Press, London, 1977.
[17] The United Kingdom Chemical Database Service, D.A.
Fletcher, R.F. McMeeking, D. Parkin, J. Chem. Inf. Comput.
Sci. 36 (1996) 746. O. Kennard, Chem. Des. Automation News
8 (1993) 1.
4. Supplementary material
[18] C.J. Cardin, K.W. Muir, J. Chem. Soc. Dalton Trans. (1977)
1593.
[19] J. Carvajal, G. Muller, J. Sales, X. Solans, C. Miravittles,
Organometallics 3 (1984) 996.
[20] K.A. Jensen, Z. Anorg. Allg. Chem. 229 (1936) 242. A.K.
Al-Sa’ady, C.A. McAuliffe, R.V. Parish, J.A. Sandbank, Inorg.
Synth. 23 (1985) 191.
[21] G.M. Sheldrick, SHELXTL, Siemens Analytical X-ray Instru-
ments, Madison, WI, 1995. G.M. Sheldrick, SHELX-97, Institu¨t
fu¨r Anorganische Chemie der Universita¨t Go¨ttingen, Go¨ttin-
gen, Germany, 1998.
Crystallographic data (excluding structure factors)
for the structures reported in this paper have been
deposited with the Cambridge Crystallographic Data-
base as supplementary publication numbers CCDC
144498–144501. Copies of the data can be obtained free
of charge on application to The Director, CCDC,
12 Union Road, Cambridge, CB2 1E2, UK (Fax: +44-
1223-336033; e-mail: deposit@ccdc.cam.ac.uk or www:
http://www.ccdc.cam.ac.uk).
.