Silver compounds in synthetic chemistry. Part 3. 4-Tetrafluoropyridyl silver(I), AgC5F4N - A reagent for redox transmetallations with group 12-14 elements

Article abstract of DOI:10.1016/j.jorganchem.2005.09.017

4-Tetrafluoropyridyl silver(I), AgC₅F₄N, has been obtained from Me₃SiC₅F₄N and AgF in nearly quantitative yield. Reactions with bis(triphenylphosphoranyliden)ammonium chloride, [PNP]Cl, gave crystalli

Full text of DOI:10.1016/j.jorganchem.2005.09.017

Journal of Organometallic Chemistry 691 (2006) 514–522
www.elsevier.com/locate/jorganchem
Silver compounds in synthetic chemistry. Part 3.
4-Tetrafluoropyridyl silver(I), AgC5F4N – A reagent for redox
q
transmetallations with group 12–14 elements
*
Wieland Tyrra , Said Aboulkacem, Ingo Pantenburg
Institut fu¨r Anorganische Chemie, Universita¨t zu Ko¨ln, Greinstrasse 6, D-50939 Ko¨ln, Germany
Received 30 August 2005; accepted 6 September 2005
Available online 20 October 2005
Abstract
4-Tetrafluoropyridyl silver(I), AgC5F4N, has been obtained from Me3SiC5F4N and AgF in nearly quantitative yield. Reactions with
bis(triphenylphosphoranyliden)ammonium chloride, [PNP]Cl, gave crystalline material of the composition [PNP][Ag(C5F4N)2]. Redox
transmetallations of AgC5F4N and group 12–14 elements Zn, Cd, Hg, Ga, In, Sn yielded the corresponding 4-tetrafluoropyridyl element
compounds. The molecular structures of [PNP][Ag(C5F4N)2], Hg(C5F4N)2, Ga(C5F4N)3 Æ EtCN Æ H2O, In(C5F4N)3 Æ 2EtCN and
Sn(C5F4N)4 are discussed.
ꢀ 2005 Elsevier B.V. All rights reserved.
Keywords: 4-Tetrafluoropyridyl; Silver; Synthesis; Crystal structure
1. Introduction
2. Results and discussion
The tetrafluoropyridyl group has already attracted
considerable interest in transition metal [1,2] as well as
main group metal chemistry [3]. On the basis of
theoretical calculations [4], the electron-withdrawing
effect of the C5F4N ligand should be considered to be
higher than that of the isolobal C6F5 group. Together
with the knowledge that perfluoroorgano silver(I)
compounds with strong electron-withdrawing organic
ligands can be synthesized conveniently and are
versatile tools for redox transmetallations [5–8], we stud-
ied the reactions of AgC5F4N and elements of groups
12–14.
2.1. Syntheses of 4-tetrafluoropyridyl silver(I), AgC5F4N
and bis(triphenylphosphoranyliden)ammonium bis
(4-tetrafluoropyridyl)argentate, [PNP][Ag(C5F4N)2]
4-Tetrafluoropyridyl silver(I), AgC5F4N, is conveniently
accessible via the reactions of the corresponding silane, Me3-
SiC5F4N, and silver(I) fluoride as already outlined for other
perfluorinated groups [5–8]. In propionitrile solution – other
solvents are also suitable [6] – the reaction proceeds selec-
tively to give AgC5F4N in nearly quantitative yield (Eq. (1)).
F
N
F
F
F
N
F
F
EtCN
r.t.
SiMe 3
Ag
+ AgF
+ Me3SiF
F
F
q
For Part 2, see M.M. Kremlev, W. Tyrra, D. Naumann, Yu.L.
ð1Þ
Yagupolskii, J. Fluorine Chem. 126 (2005) 1327.
*
Exchange processes in solution between a ‘‘neutral’’ and an
Corresponding author. Tel./fax: +49 221 4703276.
E-mail address: tyrra@uni-koeln.de (W. Tyrra).
‘‘ionic’’ species appear to be too fast for the NMR time
0022-328X/$ - see front matter ꢀ 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2005.09.017

W. Tyrra et al. / Journal of Organometallic Chemistry 691 (2006) 514–522
515
scale, i.e., only two well-resolved signal groups are ob-
served in the 19F NMR spectra irrespective of the solvent
(Eq. (2)). These experiences are in contrast to results
known for perfluoroalkyl silver(I) derivatives [9].
Zn(C5F4N)2 and Cd(C5F4N)2 were isolated as waxy solids
presumably as 1:2 adducts [14] with EtCN. Both com-
pounds are extremely sensitive to moisture. They react
spontaneously with water under formation of tetrafluoro-
pyridine. By contrast, Hg(C5F4N)2 is not affected by mois-
ture. It has been obtained free of co-ordinating solvent
molecules as crystalline material (cf. Section 2.4).
In a similar manner reactions proceed with the group 13
elements gallium and indium. Due to the passivation of
aluminium, no reaction was observed with aluminium
shots, while with elemental thallium the oxidation is fin-
ished on the oxidation step of thallium(I), because the oxi-
dation potential of AgC5F4N is too low [8]. As a
consequence, compounds of general formula [TlI Æ nEtC-
N][Ag(C5F4N)2] were formed.
2AgC5F4N ꢀ Agþ þ ½AgðC5F4NÞ ꢀꢁ
ð2Þ
2
Unfortunately, single crystals of AgC5F4N could not be
obtained. As previously outlined for two related examples
[7,10], the reaction of AgC5F4N and bis(triphenylphospho-
ranyliden)ammonium chloride, [PNP]Cl, in a stoichiome-
tric ratio of 2:1 gave crystalline [PNP][Ag(C5F4N)2]
suitable for XRD measurements (Eq. (3)).
F
F
F
F
N
N
+ AgCl
Ag
2 AgC5F4N + [PNP]Cl
[PNP]
3AgC5F4N þ M ! MðC5F4NÞ3 ꢂ nEtCN þ 3Ag
F
F
F
F
M ¼ Ga ðn ¼ 1Þ; In ðn ¼ 2Þ
ð5Þ
ð3Þ
Results of the single crystal structure analyses of Ga(C5F4-
N)3 Æ EtCN Æ H2O and In(C5F4N)3 Æ 2EtCN are presented
in Sections 2.5 and 2.6, respectively.
Elemental tin is oxidized by AgC5F4N in a primary step
to a compound presumed to be ‘‘Ag[SnII(C5F4N)3]’’ [15].
This intermediate is converted into SnIV(C5F4N)4 and ele-
mental silver in a sublimation apparatus after extended
heating to 230 ꢁC at 0.1 mbar.
Results of mass spectrometric investigations (negative ESI)
gave evidence of the complex equilibria in solution. For re-
dox transmetallations, AgC5F4N was prepared according
to Eq. (1) and used without further purification.
2.2. Molecular structure of [PNP][Ag(C5F4N)2]
[PNP][Ag(C5F4N)2] (1) crystallizes in the monoclinic
space group P21/c (no. 14) with 4 molecules per unit cell
(Table 1; Fig. 1). The lattice is built up by well-separated
ion pairs of the [PNP] cations and the argentate anions.
The anions are layered in cavities formed by the framework
of the bulky [PNP] cations without significant F–H con-
tacts (>241 pm). The silver atom is quasi-linear co-ordi-
nated as found for all perfluoroorganosilver(I) derivatives
crystallographically characterized so far. This is obvious
in the structures of homoleptic [8,10–12] as well as hetero-
leptic derivatives [7,11,13]. The planar rings are tilted at
83.5 0.5ꢁ with respect to each other.
The C–Ag–C bond angle of 173.0(1)ꢁ only slightly devi-
ates from linearity; Ag–C bond lengths are found to be
210.8(4) and 211.2(4) pm. These values are in good agree-
ment with reported data [7,8,10–12] implying similar elec-
tronic distributions around the silver atoms irrespective
of the perfluoroorgano group.
4AgC5F4N þ Sn ! SnðC5F4NÞ4 þ 4Ag
ð6Þ
Unfortunately, low solubility of Sn(C5F4N)4 in common
organic solvents prevented from a complete characteriza-
tion in solution. The results of XRD measurement are sum-
marized in Section 2.7.
Reactions with elemental germanium remained unsuc-
cessful. Rising the temperature up to 60 ꢁC for 48 h did
not give even spectroscopic evidence for the formation of
GeC5F4N derivatives. Reactions with elemental lead pro-
ceeded in a similar manner as with thallium to give
‘‘PbII[Ag(C5F4N)2]2’’; however, thermal decomposition of
the raw material did not yield Pb(C5F4N)4, as expected
[8], but 4,40-octafluorobipyridyl.
2.4. Molecular structure of Hg(C5F4N)2
Hg(C5F4N)2 (2) crystallizes in the monoclinic space
group P21/c (no. 14) with 8 molecules per unit cell. The
molecular structure of Hg(C5F4N)2 is depicted in Fig. 2,
the packing scheme in Fig. 3. The compound crystallizes
in parallel infinite zig-zag chains of two crystallographi-
cally independent Hg(C5F4N)2 molecules. Each mercury
atom shows one significant Hg–N contact (284.8(7) and
287.0(8) pm) to one of the ring nitrogen atoms of adjacent
Hg(C5F4N)2 moieties within the chain. As a consequence,
mercury atoms are T-shaped co-ordinated with N–Hg–C
angles of 89.1(3)–98.3(3)ꢁ. This co-ordination appears to
be a characteristic structural motif of many perfluoroorg-
ano mercury compounds in the solid state [16].
2.3. Redox transmetallations of AgC5F4N and elements of
groups 12–14
AgC5F4N shows excellent oxidizing abilities in reactions
with elemental zinc, cadmium, mercury (Eq. (4)), gallium,
indium (Eq. (5)), and tin (Eq. (6)). All reactions proceed
more or less indiscriminately to give the corresponding 4-
tetrafluoropyridyl element compounds and elemental silver
beside traces of tetrafluoropyridine.
2AgC5F4N þ M ! MðC5F4NÞ2 ꢂ nEtCN þ 2Ag
M ¼ Zn; Cd ðn ¼ 2Þ; Hg ðn ¼ 0Þ
ð4Þ

516
W. Tyrra et al. / Journal of Organometallic Chemistry 691 (2006) 514–522
Table 1
Crystal data and structure refinement parameters for [PNP][Ag(C5F4N)2] (1), Hg(C5F4N)2 (2), Ga(C5F4N)3 Æ EtCN Æ H2O (3), In(C5F4N)3 Æ 2EtCN (4),
and Sn(C5F4N)4 (5)
Complex
1
2
3
4
5
Empirical formula
Formula mass (g/mol)
C46H30N3F8P2Ag
946.54
C10N2F8Hg
500.71
C18H7N4F12OGa
593.00
C21H10N5F12In
675.16
C20N4F16Sn
718.93
Data collection
Diffractometer
Radiation
STOE IPDS II
Mo Ka (graphite monochromator, k = 71.073 pm)
T (K)
170(2)
170(2)
150(2)
150(2)
293(2)
Index range
ꢁ12 6 h 6 12,
ꢁ24 6 k 6 23,
ꢁ30 6 l 6 30
ꢁ12 6 h 6 12,
ꢁ6 6 k 6 7,
ꢁ41 6 l 6 41
ꢁ16 6 h 6 15,
ꢁ11 6 k 6 11,
ꢁ26 6 l 6 25
ꢁ12 6 h 6 12,
ꢁ11 6 k 6 11,
ꢁ35 6 l 6 33
ꢁ15 6 h 6 15,
ꢁ15 6 k 6 14,
ꢁ9 6 l 6 9
Rotation angle range
0ꢁ 6 x 6 180ꢁ;
w = 0ꢁ
0ꢁ 6 x 6 180ꢁ;
w = 0ꢁ
0ꢁ 6 x 6 180ꢁ;
w = 0ꢁ
0ꢁ 6 x 6 180ꢁ;
w = 0ꢁ
0ꢁ 6 x 6 180ꢁ;
w = 0ꢁ
0ꢁ 6 x 6 14ꢁ;
w = 90ꢁ
0ꢁ 6 x 6 86ꢁ;
w = 90ꢁ
0ꢁ 6 x 6 158ꢁ;
w = 90ꢁ
0ꢁ 6 x 6 94ꢁ;
w = 0ꢁ
0ꢁ 6 x 6 126ꢁ;
w = 90ꢁ
Increment
Dx = 2ꢁ
97
12
Dx = 2ꢁ
133
7
Dx = 2ꢁ
169
4
Dx = 1ꢁ
274
7
Dx = 2ꢁ
153
4
Number of images
Exposure time (min)
Detector distance (mm)
2h Range (ꢁ)
Total data collected
Unique data
120
140
120
120
120
1.9–54.8
33700
9036
5508
0.0579
1.6–50.5
18567
3945
2827
0.0605
1.9–54.8
30785
4830
3727
0.1456
1.9–54.8
29497
5494
4362
0.0476
1.9–54.8
14180
1264
965
Observed data
Rmerg
0.0367
Absorption correction
Minimum/maximum
transmission
Numerical, after crystal shape optimization [44,45]
0.8251/0.9336
0.1012/0.2583
none
0.7582/0.8963
0.7742/0.8749
Crystallographic data [43]
Crystal size (mm)
Colour
0.2 · 0.2 · 0.15
Colourless
Polyhedron
Monoclinic
P21/c (no. 14)
983.0(1)
1890.1(1)
2403.8(2)
114.24(1)
4.0724(7)
4
0.2 · 0.2 · 0.1
Colourless
Plate
0.3 · 0.3 · 0.2
Colourless
Block
0.25 · 0.25 · 0.05
Colourless
Plate
0.2 · 0.05 · 0.05
Colourless
Column
Habit
Crystal system
Space group
a (pm)
b (pm)
c (pm)
Monoclinic
P21/c (no. 14)
1079.5(1)
593.1(1)
3491.1(5)
90.46(1)
2.2370(5)
8
Monoclinic
P21/c (no. 14)
1245.6(7)
872.5(4)
2014.0(12)
92.47(5)
2.1870(2)
4
Monoclinic
P21/c (no. 14)
988.0(1)
902.9(1)
2793.6(1)
97.17(1)
2.4725(1)
4
Tetragonal
ꢀ
P421c ðno. 114Þ
1222.7(1)
751.6(1)
b (ꢁ)
V (nm3)
Z
1.1235(2)
2
2.125
1.289
2904
qcalc (g cmꢁ3
)
1.544
0.648
1904
2.973
13.864
1808
1.801
1.379
1160
1.814
1.067
1312
l (mmꢁ1
F(000)
)
Structure analysis and refinementa
Structure determination
Number of variables
SHELXS-97 [46] and SHELXL-93 [47]
662
380
336
355
94
R indexes [I > 2rI]
R1 = 0.0461,
wR2 = 0.1072
R1 = 0.0322,
wR2 = 0.0742
R1 = 0.0438,
wR2 = 0.1023
R1 = 0.0466,
wR2 = 0.1263
R1 = 0.0307,
wR2 = 0.0838
R indexes (all data)
R1 = 0.0865,
wR2 = 0.1208
R1 = 0.0514,
wR2 = 0.0799
R1 = 0.0570,
wR2 = 0.1130
R1 = 0.0568,
wR2 = 0.1317
R1 = 0.0412,
wR2 = 0.0890
Goodness of fit (Sall
)
0.938
0.890
1.051
1.045
1.073
Largest difference map in
hole/peak (e 10ꢁ6 pmꢁ3
ꢁ1.034/1.755
ꢁ1.390/1.574
ꢁ0.415/0.572
ꢁ1.194/3.683
ꢁ0.415/0.532
)
P
P
P
P
P
R1 = iFo| ꢁ |Fci/ |Fo|, wR2 = [ w(|Fo|2 ꢁ |Fc|2)2/ w(|Fo|2)2]1/2
,
S2 = [ w(|Fo|2 ꢁ |Fc|2)2/(n ꢁ p)]1/2
,
with w = 1/[r2(Fo)2 + (0.0669 · P)2] for 1,
w = 1/[r2(Fo)2 + (0.0523 · P)2] for 2, w = 1/[r2(Fo)2 + (0.0464 · P)2 + (0.9379 · P)] for 3, w = 1/[r2(Fꢁo)12=4+ (0.0939 · P)2] for
4
and w =
2
1/[r2(Fo)2 + (0.0579 · P)2 + (0.1387 · P)] for 5, where P ¼ ðF 2o þ 2F 2c Þ=3. F ꢃc ¼ kF c½1 þ 0:001 ꢄ jF cj k3= sinð2hÞꢀ
.
a
H atoms for 1 as well as the protons of the water molecule in 3 were taken from the difference Fourier map at the end of the refinement; protons of
EtCN in 3 and 4 were placed in idealized positions and constrained to ride on their parent atom.

W. Tyrra et al. / Journal of Organometallic Chemistry 691 (2006) 514–522
517
Fig. 1. Molecular structure of the anion in [PNP][Ag(C5F4N)2] (1) showing the labeling scheme (thermal ellipsoids at the 50% probability level).
Interatomic distances in pm and angles in degrees (with estimated standard deviations in parentheses): Ag1–C11 210.8(4), Ag1–C21 211.2(4), C11–Ag1–
C21 173.0(1).
Fig. 2. View of Hg(C5F4N)2 (2) showing the labeling scheme and the coordination environments of the mercury atoms (thermal ellipsoids at the 50%
probability level). Interatomic distances in pm and angles in degrees (with estimated standard deviations in parentheses): Hg1–C111 209.3(9), Hg1–C121
209.9(9), Hg1–N224 287.0(8), Hg2–C211 207.3(11), Hg2–C221 206.1(10), Hg2–N114 284.8(7), C111–Hg1–C121 175.6(3), C211–Hg2–C221 172.1(3).
It must be noted that torsion or twist angles in both
crystallographically independent Hg(C5F4N)2 moieties
deviate significantly from each other. The one has a value
of 4.5 0.5ꢁ, i.e., both aromatic rings may be regarded as
co-planar, while a value of 60.0 0.5ꢁ is found for the
other. In the case of approximately co-planar units, Hg–C
bond lengths are shorter (206.1(10) and 207.3(11) pm) com-
pared with those of twisted rings (209.3(9) and
209.9(9) pm). In both crystallographically independent
units, C–Hg–C angles only slightly deviate from linearity
(172.1(3)ꢁ and 175.6(3)ꢁ). These results fall in the good
scope of structural data for homoleptic [17,18] as well as
heteroleptic [19] polyfluorinated phenyl mercury deriva-
tives. Differences especially with respect to the twist angles
and Hg–C bond length cannot be explained by differences
in the electronic natures of the ligands and should therefore
be attributed to packing considerations.
2.5. Molecular structure of Ga(C5F4N)3 Æ EtCN Æ H2O
Examples of structurally characterized gallium
derivatives with strong electron withdrawing organic
ligands are rare and limited to salts with the tetrakis(penta-
fluorophenyl)gallate anion, [Ga(C6F5)4]ꢁ [20–23], the

518
W. Tyrra et al. / Journal of Organometallic Chemistry 691 (2006) 514–522
Fig. 4. View of Ga(C5F4N)3 Æ EtCN Æ H2O (3) showing the labeling
scheme and the coordination environment of the gallium atom (thermal
ellipsoids at the 50% probability level). Interatomic distances in pm and
angles in degrees (with estimated standard deviations in parentheses):
Ga1–C11 199.5(3), Ga1–C21 199.4(3), Ga1–C31 199.9(3), Ga1–N41
227.0(3), Ga1–O51 221.9(2), C11–Ga1–C21 121.1(1), C11–Ga1–C31
116.4(1), C11–Ga1–N41 90.0(1), C11–Ga1–O51 91.9(1), C21–Ga1–C31
122.4(1), C21–Ga1–N41 88.4(1), C21–Ga1–O51 92.1(1), C31–Ga1–N41
89.6(1), C31–Ga1–O51 88.0(1), N41–Ga1–O51 177.4(1).
motifs (221.9(2) pm versus ca. 200 pm (Ga–O) [25–27]).
There was found no evidence for intermolecular contacts.
Neither the O–Ga–N angle (177.4(1)ꢁ) deviates signifi-
cantly from linearity nor the sum of C–Ga–C angles (mean
value 120.0ꢁ (116.4(1)ꢁ, 121.1(1)ꢁ, 122.4(1)ꢁ)) from planar-
ity exhibiting real trigonal bi-pyramidal geometry for
Ga(C5F4N)3 Æ EtCN Æ H2O. The unusual penta-coordina-
tion might be interpreted in terms of great electron defi-
ciency at the gallium centre which is compensated by
co-ordination of EtCN and one additional water molecule
acting as donors.
Fig. 3. The packing diagram for Hg(C5F4N)2 (2) viewed along the
crystallographic b-axis.
anions l-fluorobis[tris(pentafluorophenyl)gallate], {[Ga-
(C6F5)3]2(l-F)}ꢁ [24] and l-hydroxybis[tris(pentafluor-
ophenyl)gallate], {[Ga(C6F5)3]2(l-OH)}ꢁ [25] as well as
the 1:1 complexes of Ga(C6F5)3 with THF [26,27] and
Et2O [27].
All spectroscopic data of the product derived from the
reaction of AgC5F4N with elemental gallium in EtCN
prove that the primary formed compound Ga(C5F4-
2.6. Molecular structure of In(C5F4N)3 Æ 2EtCN
Examples of crystal structures of polyfluoroaryl indium
compounds are to our knowledge rare and limited to three
examples: In(C6F5)Br2 Æ 2 THF [28], In(C6F5)2(CH2)3-
NMe2 [29], and [PNP][In(C6F5)4] [30].
N)3 Æ EtCN is contaminated by
a
salt with the
[Ga(C5F4N)4]ꢁ anion. Single crystals were grown upon
storing an n-pentane solution of the raw material at
ꢁ21 ꢁC over a period of two months.
In(C5F4N)3 Æ 2EtCN (4) crystallizes in the monoclinic
space group P21/c (no. 14) with 4 molecules per unit cell
(Table 1, Fig. 5). It exhibits well separated trigonal bipyra-
midal In(C5F4N)3 Æ 2EtCN units without any significant
intermolecular contacts. The aromatic rings occupy equa-
torial positions, while the propionitrile ligands reside in ax-
ial sites. In–C bond lengths vary from 216.5(4) to
217.6(3) pm and are in good agreement with known values
for pentafluorophenyl compounds [28–30]. The mean value
of the angles C–In–C of 119.7ꢁ only slightly deviates from
Ga(C5F4N)3 Æ EtCN Æ H2O (3) (Table 1, Fig. 4) crystal-
lizes in the monoclinic space group P21/c (no. 14) with 4
molecules per unit cell in a similar manner as In-
(C5F4N)3 Æ 2EtCN (Section 2.6). The tetrafluoropyridyl li-
gands reside in equatorial positions of a distorted trigonal
bi-pyramid; Ga–C bond lengths do not deviate from those
of covalent Ga–C (2c–2e) bonds [20–27]. Nitrogen and
oxygen bonds of the linear, apical motif are elongated by
approximately 10% compared with those in tetrahedral

W. Tyrra et al. / Journal of Organometallic Chemistry 691 (2006) 514–522
519
considerations. The first coordination sphere around the
tin atom only slightly deviates from an ideal tetrahedron
(111.0(1)ꢁ (4·), 106.5(2)ꢁ (2·)) and is in accordance with
that of SnPh4 [34] and Sn(C6F5)4 [36].
Fig. 5. Central projection of the asymmetric unit of In(C5F4N)3 Æ 2EtCN
(4) showing the atom numbering scheme (thermal ellipsoids at the 50%
probability level). H atoms have been omitted for clarity. Interatomic
distances in pm and angles in degrees (with estimated standard deviations
in parentheses): In1–C11 217.2(4), In1–C21 217.6(3), In1–C31 216.5(4),
In1–N41 238.1(4), In1–N51 244.6(4), C11–In1–C21 117.6(2), C21–In1–
C31 121.2(1), C11–In1–C31 120.5(1), N41–In1–N51 176.0(1), C11–In1–
N41 95.1(1), C11–In1–N51 87.4(1), C21–In1–N41 93.4(1), C21–In1–N51
88.2(1), C31–In1–N41 90.0(1), C31–In1–N51 86.1(1).
Fig. 6. View of Sn(C5F4N)4 (5) showing the labeling scheme and the
coordination environment of the tin atom (thermal ellipsoids at the 50%
probability level). Symmetry-related atoms are drawn as empty ellipsoids.
Interatomic distances in pm and angles in degrees (with estimated
standard deviations in parentheses): Sn1–C1 215.5(4) (4·), C1–Sn1–C1
111.0(1) (4·), C1–Sn1–C1 106.5(2) (2·).
the ideal value of 120ꢁ. Approximate linearity is found for
the N–In–N unit (176.0(1)ꢁ). As a consequence, the mean
value of the C–In–N angles (90.0ꢁ) does not deviate from
ideal relations although values vary from 86.1(1)ꢁ to
95.1(1)ꢁ. In–N contacts of 238.1(4) and 244.6(4) pm are
longer than the sum of covalent radii (220 pm [31]) but
significantly shorter than the sum of van der Waals radii
(ꢅ353 pm [32]). The surrounding on the indium atom
matches well with that determined for In(S-2,4,6-
(i-Pr3)3C6H2)3 Æ 2MeCN [33].
2.7. Molecular structure of Sn(C5F4N)4
Sn(C5F4N)4 (5) crystallizes in the tetragonal space group
ꢀ
ðP421c; Z ¼ 2Þ as tetraphenyltin [34] and tetrakis(3,5-
dimethylphenyl)tin [35]. A view on the molecule is depicted
in Fig. 6, while a view on the unit cell is given in Fig. 7. In
the same manner as several related examples, such quasi-
spherical molecules crystallize in tetragonal space groups
allowing close packing.
Sn–C bonds of 215.5(4) pm (4·) are slightly elongated in
comparison with Sn(C6F5)4 (212.6(8) pm [36]) but in good
agreement with those reported for Sn(C6Cl5)4 [37] and fall
absolutely into the range of bond lengths for SnAr4 com-
pounds structurally characterized so far (e.g. [34–38, and
literature cited therein]). Differences in bond lengths appear
to be independent of the electron withdrawing character of
the aromatic system but might be attributed to packing
Fig. 7. The packing diagram for Sn(C5F4N)4 (5).

520
W. Tyrra et al. / Journal of Organometallic Chemistry 691 (2006) 514–522
3. Experimental
checked with the help of an IP-diffractometer (STOE IPDS
II) [44,45]. The same device was used to collect the reflec-
tion data of the respective best specimen. The positions
of the heavy atoms were extracted from these data using
the direct methods provided by the program SHELXS-97
[46]. The light atoms were localised during the refinement
(SHELXL-93) by difference Fourier syntheses [47]. Tilted an-
gles were calculated with the program PARST-97 [48].
Schlenk techniques were used throughout most of the
manipulations. NMR spectra were recorded on Bruker
AC 200 (1H, 19F, 13C) and AVANCE 400 (199Hg,
13C{19F}) spectrometers. External standards were used in
all cases (1H, 13C: Me4Si; 19F: CCl3F; 199Hg: Me2Hg). Ace-
tone-d6 was used as an external lock (5 mm tube) in reac-
tion control measurements while an original sample of
the reaction mixture was measured in a 4 mm insert. Neg-
ative ESI-mass spectra in MeCN solutions were run on a
Finnigan MAT 900 apparatus with a flow rate of 2 ll/
min. EI mass spectra were recorded with a Finnigan
MAT 95 spectrometer (20 eV). Intensities are referenced
to the most intense peak of a group. Isotope patterns were
calculated for comparison with the program ISOPRO [39].
Visible decomposition points were determined using the
HWS Mainz 2000 apparatus. C, H and N analyses were
carried out with a HEKAtech Euro EA 3000 apparatus.
Details of crystal data and structure refinement parameters
for [PNP][Ag(C5F4N)2] (1), Hg(C5F4N)2 (2), Ga(C5F4-
N)3 Æ EtCN Æ H2O (3), In(C5F4N)3 Æ 2EtCN (4), and
Sn(C5F4N)4 (5) are summarised in Table 1.
3.2. 4-Tetrafluoropyridylsilver propionitrile adduct,
AgC5F4N Æ EtCN
By analogy to [8], 0.25 g (2.0 mmol) AgF were suspended
in 5 mL EtCN at ambient temperature. 0.46 g (2.1 mmol)
Me3SiC5F4N were added and the mixture was stirred for
2 h. The colourless solution was taken off by a pipette and
all volatile compounds were distilled off in vacuo at ambient
temperature. Colourless, amorphous material was obtained
in nearly quantitative yield which was sensitive to moisture
and light.
19F NMR (188.3 MHz, EtCN): d = ꢁ98.9 (m, 2F,
F-2,6); ꢁ114.1 (m, 2F, F-3,5).
13C NMR (50.3 MHz, CD2Cl2/CD3CN): d = 152.0 (t,
1
4H-C5F4N was prepared according to [40] at 0 ꢁC,
[PNP]Cl according to [41]. AgF (Apollo) and all metals
were used as received. EtCN was purchased from Acros
Chemicals and used without further purification.
2JF,C ꢅ 76 Hz, C-4); 145.8 (dm, JF,C ꢅ 228 Hz, C-2,6);
142.5 (dm, 1JF,C ꢅ 247 Hz, C-3,5); 121.3 (broad,
CH3CH2CN); 10.7 (tq, CH3CH2CN); 10.1 (qt, CH3CH2-
CN).
Negative ESI-MS (MeCN): m/z (rel. int.): 924
[Ag3(C5F4N)4]ꢁ (1), 665 [Ag2(C5F4N)3]ꢁ (100), 407
[Ag(C5F4N)2]ꢁ (76).
3.0.1. 4-Trimethylsilyl-2,3,5,6-tetrafluoropyridine,
Me3SiC5F4N
14.5 g C5HF4N (97 mmol) were dissolved in 80 mL Et2O
at ꢁ60 ꢁC. 48.5 mL of a 2 M solution of n-BuLi (97 mmol)
in n-pentane were dropped into the well-stirred solution.
After stirring for 1 h, temperature was increased to
ꢁ40 ꢁC and 12.3 mL (97 mmol) of freshly distilled Me3SiCl
were dropped to the reaction mixture. After gentle warm-
ing to room temperature, LiCl was filtered off, and low
boiling compounds were distilled at a bath temperature
not exceeding 116 ꢁC (Et2O, n-pentane, C5HF4N) on a
rotation evaporator. The remaining brown, oily residue
was stored over Na2SO4 overnight, and finally the product
was condensed in vacuo. 16.8 g Me3SiC5F4N (77% yield)
were isolated as a colourless liquid.
3.3. Bis(triphenylphosphoranyliden)ammonium bis
(4-tetrafluoropyridyl)argentate, [PNP][Ag(C5F4N)2]
To a solution of ca. 2.0 mmol AgC5F4N prepared as
described above, 0.64 g (1.0 mmol) of [PNP]Cl dissolved
in EtCN were added. Directly after combining both solu-
tions, AgCl precipitated which was filtered off. The result-
ing solution was evaporated to dryness. After washing with
n-pentane, colourless crystals of [PNP][Ag(C5F4N)2] were
grown on storing a CH2Cl2/Et2O mixture for several days
at ꢁ20 ꢁC. Colourless, crystalline [PNP][Ag(C5F4N)2] was
obtained in 96% yield. 19F and 13C NMR and negative
ESI spectra matched with those of AgC5F4N Æ EtCN.
Results of XRD measurements are described in Section 2.2.
Anal. Calc. for C46H30F8N3AgP2: C, 58.36; H, 3.19; N,
4.44. Found: C, 57.64; H, 3.44; N, 3.55%.
19F NMR (188.3 MHz, neat): d = ꢁ94.2 (m, 2F, F-2,6);
ꢁ131.3 (m, 2F, F-3,5).
1H NMR (200.1 MHz, neat): d = 0.5 (s, Me3Si).
13C NMR (50.3 MHz, neat): d = 143.8 (dm,
1
1JF,C ꢅ 251 Hz, C-2,6); 142.5 (dm, JF,C ꢅ 247 Hz, C-3,5);
3.4. Transmetallations with group 12 metals
2
1
132.2 (t, JF,C ꢅ 61 Hz, C-4); ꢁ2.3 (q, JC,H ꢅ 121 Hz,
Me3Si).
3.4.1. Reactions with elemental zinc and cadmium
Analytical data matched with the literature values [42].
Transmetallations of AgC5F4N with zinc and cadmium
were terminated after 16 h of heating to 75 ꢁC in EtCN.
After all AgC5F4N had been consumed, 19F NMR spectra
exclusively exhibited the signals of Zn(C5F4N)2 (d = ꢁ98.2,
ꢁ123.4) and Cd(C5F4N)2 (d = ꢁ97.6, ꢁ119.2), respec-
tively. With the knowledge that Zn(Rf)2 and Cd(Rf)2 form
3.1. Crystal structure analyses
All compounds 1–5 [43] form colourless single crystals
which were sealed in glass capillaries. Suitability was

W. Tyrra et al. / Journal of Organometallic Chemistry 691 (2006) 514–522
521
1:2 complexes with donor molecules [14] together with the
fact that Zn(C5F4N)2 and Cd(C5F4N)2 are known from the
literature [49], no further efforts were made to obtain crys-
talline derivatives from the waxy raw materials.
Complete conversion of the silver compound was
controlled by 19F NMR spectroscopic means. After a total
reaction time of 3 h, all volatile components were distilled
off in vacuo giving a metallic grey residue which was
extracted with CH2Cl2 in a Soxhlet-apparatus. In(C5-
F4N)3 Æ 2EtCN (m.p. 126–130 ꢁC (loss of EtCN)) was ob-
tained as a colourless solid in 35% yield (0.94 g). Crystals
suitable for XRD measurements were grown from concen-
trated EtCN solutions at ꢁ28 ꢁC (Section 2.6).
3.4.2. Bis(4-tetrafluoropyridyl)mercury, Hg(C5F4N)2
The reaction of AgC5F4N and a manifold excess of ele-
mental mercury proceeded selectively at room temperature
within a few minutes to give Hg(C5F4N)2. Starting from
2.0 mmol AgF and 2.1 mmol Me3SiC5F4N, 0.45 g
Hg(C5F4N)2 (45% yield) were isolated after distilling off
all volatile compounds and subliming the residue in vacuo
at ca. 150 ꢁC. Colourless single crystals of Hg(C5F4N)2
were obtained from CH2Cl2 or n-pentane solutions. The
melting point (195–196 ꢁC) agreed with the literature val-
ues of 193 ꢁC [49] and 201–202 ꢁC [50].
19F NMR (188.3 MHz, (CD3)2CO); d = ꢁ95.3 (m, 2F,
F-2,6); ꢁ122.3 (m, 2F, F-3,5).
1H NMR (200.1 MHz, (CD3)2CO): d = 2.45 (q,
3
3JHH = 7.7 Hz, 2H, CH3CH2CN); 1.24 (t, JHH = 7.7 Hz,
3H, CH3CH2CN).
13C NMR (50.3 MHz, (CD3)2CO): d = 145.6 (dm,
1
1JF,C ꢅ 247 Hz, C-2,6); 143.6 (dm, JF,C ꢅ 247 Hz, C-3,5);
19F NMR (188.3 MHz, CD3CN): d = ꢁ93.8 (m,
121.6 (broad, CH3CH2CN and C-4); 10.8 (tq,
CH3CH2CN); 10.7 (qt, CH3CH2CN).
3
4JHg,F ꢅ 108 Hz, 2F, F-2,6); ꢁ123.2 (m, 2F, JHg,F
ꢅ
371 Hz, F-3,5).
EI-MS (20 eV): m/z (rel. int.): 565 [In(C5F4N)3]+ (26),
415 [In(C5F4N)2]+ (100), 151 [C5HF4N]+ (17).
Anal. Calc. for C21H10N5F12In: C, 36.64; N, 9.46; H,
1.28. Found: C, 37.36; N, 10.37; H, 1.49%.
13C NMR (50.3 MHz, CD3CN): d = 153.3 (t,
2JF,C ꢅ 46 Hz, JHg,C ꢅ 1734 Hz, C-4); ꢅ 145 (overlapping
1
dm, C-2,6 and C-3,5).
199Hg NMR (71.7 MHz, CD3CN): d = ꢁ1029 (qiqi,
EI-MS (20 eV): m/z (rel. int.): 501 [Hg(C5F4N)2]+ (100),
351 [Hg(C5F4N)]+ (25), 150 [C5F4N]+ (50).
Anal. Calc. for C10F8N2Hg: C, 31.68; N, 7.39. Found:
C, 32.58; N, 7.28%.
4
3JHg,F ꢅ 370 Hz, JHg,F ꢅ 109 Hz).
3.6. Tetrakis(4-tetrafluoropyridyl)tin, Sn(C5F4N)4
To a solution of AgC5F4N prepared from 4.0 mmol AgF
and 4.2 mmol Me3SiC5F4N in EtCN elemental tin was
added in a manifold excess. The reaction mixture was stirred
for 16 h at ambient temperature. After this period, signals of
AgC5F4N were no longer detected in the 19F NMR spectrum
of the reaction mixture. All volatile components were dis-
tilled off in vacuo. The raw material was sublimed at ca.
230 ꢁC/0.1 mbar for several hours giving a colourless mate-
rial which was recrystallized from boiling EtCN/n-hexane
in an open beaker. 0.15 g (20% yield) Sn(C5F4N)4 were ob-
tained as colourless needles with a melting point of 275 ꢁC.
The very low solubility in common organic solvents pre-
vented from recording 13C and 119Sn NMR spectra.
19F NMR (188.3 MHz, EtCN): d = ꢁ91.4 (m, 2F, F-
2,6); ꢁ123.7 (m, 2F, F-3,5).
3.5. Transmetallations with group 13 metals
3.5.1. Reactions with elemental gallium and indium
Reactions of AgC5F4N and elemental aluminium re-
mained unsuccessful due to the passivation of the alumin-
ium shot used. Reactions with elemental thallium
proceeded with formation of a silver mirror but stopped
on the thallium oxidation state of +I due to the too low
oxidation potential of AgI. In reactions with excess of ele-
mental gallium, signals of Ga(C5F4N)3 (d = ꢁ97.0, ꢁ127.1)
and the corresponding gallate, [Ag Æ nEtCN][Ga(C5F4N)4]
(d = ꢁ95.4, ꢁ127.8) were found in the 19F NMR spectrum
(integrative ratio = 2.4:1). Single crystals were grown from
the raw material in n-pentane solutions over a period of
two months at ꢁ21 ꢁC in NMR tubes sealed with Para-
filmꢂ. XRD measurements proved the formation of the
gallium species Ga(C5F4N)3 Æ EtCN Æ H2O (Section 2.5).
EI-MS (20 eV): m/z (rel. int.): 519 [Ga(C5F4N)3]+ (100),
369 [Ga(C5F4N)2]+ (62), 243 (?) (26), 151 [C5HF4N]+ (90).
NMR data cannot be provided due to very low solubil-
ity of the crystals in common organic solvents.
EI-MS (20 eV): m/z (rel. int.): 720 [Sn(C5F4N)4]+ (100),
570 [Sn(C5F4N)3]+ (42), 151 [C5HF4N]+ (16).
Anal. Calc. for C20N4F20Sn: C, 33.41; N, 7.79. Found:
C, 33.59; N, 7.82%.
Acknowledgements
Generous support in all fields of this work by Prof. Dr.
Dieter Naumann is gratefully acknowledged. We are in-
debted to our technicians for their great help. Thanks to
Dr. Mathias Scha¨fer (Institute of Organic Chemistry) for
recording the ESI mass spectra.
3.5.2. Tris(4-tetrafluoropyridyl)dipropionitrile indium,
In(C5F4N)3 Æ 2EtCN
To a solution of AgC5F4N prepared from 4.0 mmol
AgF and 4.2 mmol Me3SiC5F4N in EtCN elemental in-
dium (shots, 1–3 mm) was added in large excess. The reac-
tion mixture was stirred for 3 h at ambient temperature.
References
´
[1] (a) For example: M. Ferrer, M. Mounir, L. Rodrıguez, O. Rossell, S.
´
´
Coco, P. Gomez-Sal, A. Martın, J. Organomet. Chem. 690 (2005) 2200;

522
W. Tyrra et al. / Journal of Organometallic Chemistry 691 (2006) 514–522
(b) D. Noveski, T. Braun, B. Neumann, A. Stammler, H.-G. Stammler, [22] A.H. Cowley, C.L.B. Macdonald, J.S. Silverman, J.D. Gorden, A.
Dalton Trans. (2004) 4106;
Voigt, Chem. Commun. (2001) 175.
(c) N.A. Jassim, R.N. Perutz, A.C. Whitwood, T. Braun, J. Izundu, B.
Neumann, S. Rothfeld, H.-G. Stammler, Organometallics 23 (2004)
6140;
(d) B.V. Nguyen, D.J. Burton, J. Fluorine Chem. 67 (1994) 205;
(e) E.J. Soloski, W.E. Ward, C. Tamborski, J. Fluorine Chem. 2 (1973)
361;
[23] B.L. Kaafarani, H. Gu, A.A. Pinkerton, D.C. Neckers, J. Chem.
Soc., Dalton Trans. (2002) 2318.
[24] M.-C. Chen, J.A.S. Roberts, T.J. Marks, Organometallics 23 (2004)
932.
[25] F. Conrad, Dissertation, Universita¨t zu Ko¨ln, 1996.
[26] Cf. Ref [22], therein [5]; CCDC 137250.
(f) J. Cooke, M. Green, F.G.A. Stone, Inorg. Nucl. Chem. Lett. 3
(1967) 47;
(g) B.L. Booth, R.N. Haszeldine, J. Organomet. Chem. 6 (1966) 570.
[27] (a) C. Meyer, Dissertation, Universita¨t Dusseldorf, 1994;
¨
(b) F. Conrad, Z.-H. Choi, K. Ludovici, W. Tyrra, in: Proceedings of
the 11th European Symposium on Fluorine Chemistry, Bled, Slove-
nia, 1995, Abstract 116.
[2] A.O. Miller, V.I. Krasnov, D. Peters, V.E. Platonov, R. Mietchen,
Tetrahedron Lett. 41 (2000) 3817.
[3] (a) For example: R.G. Kultyshev, G.K.S. Prakash, G.A. Olah, J.W.
Faller, J. Parr, Organometallics 23 (2004) 3184;
(b) V.V. Bardin, L.S. Pressman, G.G. Furin, J. Organomet. Chem.
448 (1993) 55.
[28] W. Tyrra, M.S. Wickleder, J. Organomet. Chem. 677 (2003) 28.
[29] H. Schumann, O. Just, Th.D. Seuß, F.H. Go¨rlitz, R. Weimann, J.
Organomet. Chem. 466 (1994) 5.
[30] Z.-H. Choi, W. Tyrra, A. Adam, Z. Anorg. Allg. Chem. 625 (1999)
1287.
[4] B. Hoge, C. Tho¨sen, T. Herrmann, P. Panne, I. Pantenburg, J.
Fluorine Chem. 125 (2004) 831.
[31] J. Emsley, The Elements, third ed., Clarendon Press, Oxford, 1998.
[32] A. Bondi, J. Phys. Chem. 68 (1964) 441;
[5] W.E. Tyrra, J. Fluorine Chem. 112 (2001) 149.
[6] W. Tyrra, Heteroatom. Chem. 13 (2002) 561.
[7] W. Tyrra, D. Naumann, J. Fluorine Chem. 125 (2004) 823.
[8] W. Tyrra, M.S. Wickleder, Z. Anorg. Allg. Chem. 628 (2002) 1841.
[9] D. Naumann, W. Wessel, J. Hahn, W. Tyrra, J. Organomet. Chem.
547 (1997) 79.
S.C. Nyburg, C.H. Faerman, Acta Crystallogr., Sect. B 41 (1985) 274.
[33] R.D. Schluter, G. Krauter, W.S. Rees Jr., J. Cluster Sci. 8 (1997) 123,
CA 126:324477; ROLREP.
[34] L.M. Engelhardt, L.-P. Leung, C.L. Raston, A.H. White, Aust. J.
Chem. 35 (1982) 2383.
´
´
[35] I. Wharf, F. Belanger-Gariepy, Acta Crystallogr., Sect. E 59 (2003)
[10] S. Aboulkacem, W. Tyrra, I. Pantenburg, J. Chem. Crystallogr., in
press.
[11] R.R. Burch, J.C. Calabrese, J. Am. Chem. Soc. 108 (1986) 5359.
[12] P.M. Jeffries, S.R. Wilson, G.S. Girolami, J. Organomet. Chem. 449
(1993) 203.
m661.
[36] A. Karipides, C. Forman, R.H.P. Thomas, A.T. Reed, Inorg. Chem.
13 (1974) 811.
[37] I. Ara, J. Fornies, M.A. Garcıa-Monforte, A. Martın, B. Menjon,
´
´
´
´
Chem. Eur. J. 10 (2004) 4186.
´
´
[13] R. Uson, A. Laguna, A. Uson, P.G. Jones, K. Meyer-Ba¨se, J. Chem.
[38] I. Wharf, A.-M. Lebuis, Acta Crystallogr., Sect. E 59 (2003) m794.
[39] M. Senko, ISOPRO 3.0, Shareware, Sunnyvale, CA.
[40] S.S. Laev, V.D. Shteingarts, Tetrahedron Lett. 38 (1997) 3765.
[41] J.K. Ruff, W.J. Schlientz, Inorg. Synth. 15 (1974) 84.
[42] R.E. Banks, R.N. Haszeldine, E. Philips, I.M. Young, J. Chem. Soc.
C (1967) 2091.
[43] Crystallographic data for the structures have been deposited with the
Cambridge Crystallographic Data Centre as supplementary publica-
tion nos. CCDC 272520 for 1, CCDC 272522 for 2, CCDC 274915 for
3, CCDC 272521 for 4, and CCDC 272519 for 5. Copies of the data
can be obtained, free of charge, on application to CCDC, 12 Union
Road, Cambridge CB2 1EZ, UK (fax: +44 1223 336 033 or e-mail:
deposit@ccdc.cam.ac.uk).
Soc., Dalton Trans. (1988) 341.
[14] (a) For Zn: H. Lange, D. Naumann, J. Fluorine Chem. 26 (1984) 435;
(b) D. Naumann, C. Schorn, W. Tyrra, Z. Anorg. Allg. Chem. 625
(1999) 827;
(c) For Cd: H. Lange, D. Naumann, J. Fluorine Chem. 26 (1984) 1;
(d) D. Naumann, K. Glinka, W. Tyrra, Z. Anorg. Allg. Chem. 594
(1991) 95.
[15] S. Aboulkacem, W. Tyrra, unpublished results.
[16] (a) F. Schulz, I. Pantenburg, D. Naumann, Z. Anorg. Allg. Chem.
629 (2003) 2312;
(b) D. Naumann, F. Schulz, I. Pantenburg, W. Tyrra, Z. Anorg.
Allg. Chem. 630 (2004) 529.
[17] (a) G.B. Deacon, C.M. Forsyth, D.M.M. Freckmann, G. Meyer, D.
Stellfeldt, Z. Anorg. Allg. Chem. 626 (2000) 540;
[44] X-RED 1.22, Stoe Data Reduction Program ꢀ 2001 Stoe & Cie
GmbH Darmstadt.
(b) D.L. Wilkinson, J. Riede, G. Muller, Z. Naturforsch. B 51 (1991)
¨
285;
[45] X-Shape 1.06, Crystal Optimisation for Numerical Absorption
Correction ꢀ 1999 STOE & Cie GmbH Darmstadt.
(c) D.S. Brown, A.G. Massey, D.A. Wickens, J. Organomet. Chem.
194 (1980) 131.
[46] G.M. Sheldrick, SHELXS-97: Program for Structure Analysis, Univer-
sity of Go¨ttingen, Go¨ttingen, Germany, 1998.
[18] D. Naumann, F. Schulz, Z. Anorg. Allg. Chem. 631 (2005) 122.
[19] G.B. Deacon, P.C. Junk, J. Chem. Crystallogr. 33 (2003) 605.
[20] K.-F. Tebbe, T. Gilles, F. Conrad, W. Tyrra, Acta Crystallogr., Sect.
C 52 (1996) 1663.
[21] I.A. Guzei, S. Dagorne, R.F. Jordan, Acta Crystallogr., Sect. C 56
(2000) e134.
[47] G.M. Sheldrick, SHELXL-93: Program for Crystal Structure Refine-
ment, University of Go¨ttingen, Go¨ttingen, Germany, 1993.
[48] M. Nardelli, PARST-97, J. Appl. Crystallogr. 28 (1995) 659.
[49] P. Sartori, H. Adelt, J. Fluorine Chem. 3 (1973/74) 275.
[50] R.D. Chambers, F.G. Drakesmith, J. Hutchinson, W.K.R. Mus-
grave, Tetrahedron Lett. 18 (1967) 1705.