O-(trifluoromethyl)aryl interactions and stabilization in hypervalent germanium compounds

Article abstract of DOI:10.1021/om980430m

A series of o-(trifluoromethyl)phenyl-substituted germanes have been synthesized to study the potential for weak Ge-F bonding interactions. ¹H-¹⁹F coupling constants in these compounds range between 2 and 5 Hz. These coupling magnitudes are consistent with previously reported literature examples in ¹⁹F and multinuclear NMR spectra for derivatives of the [2,4,6-(CF₃)₃C₆H₂]- ligand. Solid-state structural characterization of these germanes consistently reveals Ge-F contacts at 2.8-2.9 A?, midway between the longest single bond and the longest expected van der Waals contact. For the simple unsymmetrical catenate, [3,5-(CF₃)₂C₆H₃] ₂Ge(H)Ge(H)[2,4,6-(CF₃)₃C₆H ₂]₂ (4), there are close contacts to both germanium atoms from the ortho-CF₃ groups. 4 is not otherwise sterically constrained to force distal, intramolecular contacts, and thus a weak Ge-F bonding interaction is invoked for the entire series of o-(trifluoromethyl)aryl-substituted tetravalent Ge compounds. The synthetic approaches to these germanes make extensive use of Ge(II) precursors, employing both catalytic hydrogenation and insertion reactions of germylenes.

Full text of DOI:10.1021/om980430m

5166
Organometallics 1998, 17, 5166-5171
o-(Tr iflu or om eth yl)a r yl In ter a ction s a n d Sta biliza tion in
Hyp er va len t Ger m a n iu m Com p ou n d s
J ohn E. Bender IV,† Mark M. Banaszak Holl,*,‡ Amy Mitchell,§
Norman J . Wells,§ and J eff W. Kampf‡
Department of Chemistry, Brown University, Providence, Rhode Island 02912, Department of
Chemistry, The University of Michigan, Ann Arbor, Michigan 48109-1055, and Department of
Chemistry, Baldwin-Wallace College, Berea, Ohio 44017
Received May 26, 1998
A series of o-(trifluoromethyl)phenyl-substituted germanes have been synthesized to study
the potential for weak Ge-F bonding interactions. 1H-19F coupling constants in these
compounds range between 2 and 5 Hz. These coupling magnitudes are consistent with
previously reported literature examples in 19F and multinuclear NMR spectra for derivatives
of the [2,4,6-(CF3)3C6H2]- ligand. Solid-state structural characterization of these germanes
consistently reveals Ge-F contacts at 2.8-2.9 Å, midway between the longest single bond
and the longest expected van der Waals contact. For the simple unsymmetrical catenate,
[3,5-(CF3)2C6H3]2Ge(H)Ge(H)[2,4,6-(CF3)3C6H2]2 (4), there are close contacts to both germa-
nium atoms from the ortho-CF3 groups. 4 is not otherwise sterically constrained to force
distal, intramolecular contacts, and thus a weak Ge-F bonding interaction is invoked for
the entire series of o-(trifluoromethyl)aryl-substituted tetravalent Ge compounds. The
synthetic approaches to these germanes make extensive use of Ge(II) precursors, employing
both catalytic hydrogenation and insertion reactions of germylenes.
In tr od u ction
and multidentante pincer ligands,4 often employed to
stabilize main group complexes. The only carbon σ-bond-
ed germanium derivative of this ligand, [2,4,6-
(CF3)3C6H2]2Ge, was first noted in solution by Edelmann1e
and subsequently isolated and structurally character-
ized by Bender et al.5 The germylene was noted to have
all four of the ortho-CF3 groups positioned such that a
fluorine from each was well within the sum of the van
der Waals radii of germanium and fluorine. Two Ge-F
contacts were noted at ∼2.6 Å, and two longer contacts
were also present at ∼2.8 Å. We wished to further
explore the nature of these close contacts in tetravalent
germanium species, particularly in germane catenates,
where contacts could be made to Ge atoms not directly
bonded to the [2,4,6-(CF3)3C6H2]- substituent. Reac-
tions of digermanes have generated significant interest
recently.6,7
Simpler phenyl analogues of [2,4,6-(CF3)3C6H2]- con-
taining only one or two CF3 groups have also been
studied, though in much less experimental detail. For
the 2-(trifluoromethyl)phenyl and bis(2,6-trifluorometh-
yl)phenyl groups,8,9 there is a body of experimental
evidence showing weak intramolecular bonding interac-
tions between the central metal atom and the ligand
The tris(2,4,6-trifluoromethyl)phenyl ligand has been
employed to stabilize a wide variety of unusual, σ-bond-
ed main group and transition metal organometallic
compounds.1 Several compounds of the group 14 ele-
ments with this ligand have appeared.2 The combina-
tion of unusually short M-F contacts observed in the
solid state between the fluorines of the CF3 groups and
the central metal atom to which this ligand is attached
and the highly electron-withdrawing character of the
ligand appears to be the major contributor to the
stabilizing properties typically observed. The ligand
provides an interesting foil to the more common classes
of sterically bulky and electron-rich σ-bonded ligands3
† Brown University.
‡ The University of Michigan.
§ Baldwin-Wallace College.
(1) (a) Carr, G. E.; Chambers, R. D.; Holmes, T. F.; Parker, D. G. J .
Organomet. Chem. 1987, 325, 13. (b) Scholz, M.; Roesky, H. W.; Stalke,
D.; Keller, K.; Edelmann, F. T. J . Organomet. Chem. 1989, 366, 73. (c)
Schluter, R. D.; Cowley, A. H.; Atwood, D. A.; J ones, R. A.; Bond, M.
R.; Carrano, C. J . J . Am. Chem. Soc. 1993, 115, 2070. (d) Edelmann,
F. T. in Inorganic Fluorine Chemistry, Toward the 21st Century;
Thrasher, J . S., Strauss, S. H., Eds.; ACS Symposium Series 555; 1994;
p 309. (e) Edelmann, F. T. Main Group Metal Chem. 1994, 17, 67.
(2) (a) Grutzmacher, H.; Pritzkow, H.; Edelmann, F. T. Organome-
tallics 1991, 10, 24. (b) Brooker, S.; Buijink, J .-K.; Edelmann, F. T.
Organometallics 1991, 10, 25. (c) Lay, U.; Pritzkow, H.; Grutzmacher,
H. J . Chem. Soc., Chem. Commun. 1992, 260. (d) Buijink, J .-K.;
Noltemeyer, M.; Edelmann, F. T. J . Fluor. Chem. 1993, 61, 51. (e)
Braddock-Wilking, J .; Schieser, M.; Brammer, L.; Huhmann, J .;
Shaltout, R. J . Organomet. Chem. 1995, 499, 89. (f) Klinkhammer, K.
W.; Fassler, T. F.; Grutzmacher, H. Angew. Chem., Int. Ed. Engl. 1998,
37, 124.
(4) For example see: (a) Rietveld, M. H. P.; Grove, D. M.; van Koten,
G. New J . Chem. 1997, 21, 751. (b) Carre, F.; Chuit, C.; Corriu, R. J .
P. Organometallics 1995, 14, 2754.
(5) Bender, J . E.; Banaszak Holl, M. M.; Kampf, J . W. Organome-
tallics 1997, 16, 2743.
(6) (a) Pigarev, S. D.; Bravo-Zhivotovskii, D. A.; Kalokhman, I. D.;
Vyazankin, N. S.; Voronkov, M. G. J . Organomet. Chem. 1989, 369,
29. (b) Castel, A.; Riviere, P.; Satge, J .; Desor, D.; Ahbala, M.;
Abdenadher, C. Inorg. Chim. Acta 1993, 212, 51. (c) Chaubon-
Deredempt, M. A.; Escudie, J .; Couret, C. J . Organomet. Chem. 1994,
467, 37.
(3) (a) Tokitoh, N.; Matsuhashi, Y.; Shibata, K.; Matsumoto, T.;
Suzuki, H.; Saito, M.; Manmaru, K.; Okazaki, R. Main Group Metal
Chem. 1994, 17, 55. (b) Lappert, M. F. Main Group Metal Chem. 1994,
17, 183. (c) Driess, M.; Grutzmacher, H. Angew. Chem., Int. Ed. Engl.
1996, 35, 828.
(7) Baines, K. M.; Stibbs, W. G. Coord. Chem. Rev. 1995, 145, 157.
10.1021/om980430m CCC: $15.00 © 1998 American Chemical Society
Publication on Web 10/16/1998

o-(Trifluoromethyl)aryl Interactions
Organometallics, Vol. 17, No. 23, 1998 5167
CF3 group. Butters et al. and Fernandez et al. noted
short M-F contacts in their structural studies of the
[2-(CF3)C6H4]- ligand.8d,h Multinuclear NMR experi-
ments have been used to characterized this interaction,
and there have been two papers applying 119Sn Moss-
bauer spectroscopy.8a,9a A large literature on the long-
range coupling behavior of 19F-X (>4 bonds) indicates
that the magnitudes can be large (>2.0 Hz); however,
they are very sensitive to interatomic separation and
orientation. The sensitivity of the couplings to direct
F-X contacts has lead some workers in the field to
describe the couplings as a “through-space” interac-
tion.10 Braddock-Wilking and co-workers conducted a
detailed multinuclear NMR study of two silicon mol-
ecules closely related to the species reported in this
paper: [2,4,6-(CF3)3C6H2]2SiH2 and [2,4,6-(CF3)3C6H2]2-
SiFH. The nominal five-bond H-F couplings between
the CF3 groups and Si-H moieties were found to be 5.0
and 4.6 Hz, respectively, perhaps larger than would be
expected for coupling nuclei separated by five bonds that
are not otherwise fixed close in space.2e,10 Yet, these
2
couplings are considerably smaller than the J H-F
coupling of 51 Hz observed for the F-Si-H of [2,4,6-
(CF3)3C6H2]2SiFH. In a report on the use of the [2,4,6-
(CF3)3C6H2] substituent to stabilize the first known
iminoarsine complex [2,4,6-(CF3)3C6H2]AsdN[2,4,6-
(CF3)3C6H2], Roesky and co-workers found 19F-19F
coupling constants of 4.2 Hz.11 The coupling 19F nuclei
of the ortho-CF3 groups are nominally nine bonds apart,
yet the six-bond ortho-para CF3 coupling is not seen.
Empirically, 19F NMR spectra for compounds of [2,4,6-
(CF3)3C6H2]- often show coupling magnitudes and
temperature dependence consistent with a through-
space, not through-bond, interaction. The evidence to
date suggests against classifying couplings such as those
described above as purely classical through-bond inter-
actions. However, no consistent alternative description
has been introduced into the literature.
Here we present a system of germanium small
molecules, synthesized as structural models for the
development of ligand-based approaches to the physical
properties of group 14 polymers. A number of methods
for synthesizing germanium organometallics is demon-
strated, based mostly on novel strategies for the use of
the Ge(II) state as a Ge(IV) synthon. The emphasis is
on o-(trifluoromethyl)aryl substituents that bring a CF3
group in close proximity to the group 14 element atom
F igu r e 1. o-(trifluoromethyl)arylgermanes 1, 2, and 4.
F igu r e 2. Synthesis of 1 via catalytic hydrogenation of
[2,4,6-(CF3)3C6H2]2Ge.
(Figure 1). The CF3 groups exhibit strong influences
that are electronically inductive, sterically hindering,
and weakly bonding. Ortho-CF3 substitution can pro-
vide opportunities for such hypercoordination and
potentially create novel tertiary structures in polyger-
manes through distal, intramolecular Ge-F contacts.
Resu lts
[2,4,6-(CF3)3C6H2]2GeH2 (1) was synthesized via an
application of our previously published metal-catalyzed
hydrogenation method (Figure 2).12 In this particular
instance, Ni(COD)2 cannot be used as a catalyst, since
it forms an insoluble precipitate upon reaction with
[2,4,6-(CF3)3C6H2]2Ge. Phosphine substituents must be
used to keep the Ni catalyst in solution. The reaction
is thus somewhat slower than for other germylenes, but
is still efficient.12 [2-(CF3)C6H4]GeH3 (2) and [2-(CF3)-
C6H4]GeBr3 (3) were synthesized by established routes
from Grignard reagents and GeBr4.13 Much like [3,5-
(CF3)2C6H3]2GeH2,14 the overall yields of germane 2
from GeBr4 were low (<25%). It was initially feared
that such CF3-substituted compounds may not give
stable Grignard agents, particularly [2-(CF3)C6H4]MgBr,
which is intramolecularly positioned for the elimination
(8) (a) Eaborn, C.; Hornfeld, H. L.; Walton, D. R. M. J . Organomet.
Chem. 1967, 10, 529. (b) Chivers, T.; Sams, J . R. J . Chem. Soc. (A)
1970, 928. (c) Chivers, T. Can. J . Chem. 1970, 48, 3856. (d) Butters,
T.; Scheffler, K.; Stegmann, H. B.; Weber, U.; Winter, W. J . Organomet.
Chem. 1984, 272, 351. (e) Sihler, R.; Werz, U.; Brune, H.-A. J .
Organomet. Chem. 1989, 368, 213. (f) Akianiec, B. C.; Young, G. B.
Polyhedron 1991, 10, 1411. (g) Huber, F.; Preut, H.; Scholz, D.;
Schurmann, M. J . Organomet. Chem. 1992, 441, 227. (h) Fernandez,
J . M.; Lembrino-Canales, J . J . Monatsh. Chem. 1995, 126, 129. (i)
Driess, M.; Pritzkow, H.; Reisgys, M. Chem. Ber. 1996, 129, 247.
(9) (a) Bigwood, M. P.; Corvan, P. J .; Zuckerman, J . J . J . Am. Chem.
Soc. 1981, 103, 7643. (b) Escudie, J .; Couret, C.; Ranaivonjatovo, H.;
Lazraq, M.; Satge´, J . Phosphorus, Sulfur, Silicon, Relat. Elem. 1987,
31, 27. (c) Kodaira, K.; Okuhara, K. Bull. Chem. Soc. J pn. 1988, 61,
1625.
(10) (a) Emsley, J . W.; Phillips, L.; Wray, V. Fluorine Coupling
Constants; Pergamon Press: Elmsford, NY, 1977. (b) Schaefer, T.;
Sebastian, R.; Veregin, R. P.; Laatikainen, R. Can. J . Chem. 1982, 61,
29. (c) Burgess, D. A.; Finn, M. F.; J ordan, D.; Rae, I. D.; Walters, B.
D. Aust. J . Chem. 1984, 37, 1769. (d) Kira, M.; Tokura, S. Chem. Lett.
1994, 1459.
(12) Litz, K. E.; Bender, J . E.; Banaszak Holl, M. M.; Kampf, J . W.
Angew. Chem., Int. Ed. Engl. 1997, 36, 496.
(13) Tabern, D. L.; Orndorff, W. R.; Dennis, L. M. J . Am. Chem.
Soc. 1925, 47, 2039.
(11) Ahleman, J . T.; Kunzel, A.; Roesky, H. W.; Noltemeyer, M.;
Markovskii, L.; Schmidt, H.-G. Inorg. Chem. 1996, 35, 6644.
(14) Bender, J . E.; Litz, K. E.; Giarikos, D.; Wells, N. J .; Banaszak
Holl, M. M.; Kampf, J . W. Chem. Eur. J . 1997, 3, 1793.

5168 Organometallics, Vol. 17, No. 23, 1998
Bender et al.
of MgBrF. However, the low yields obtained in some
steps of these preparations are consistent with yields
normally obtained for Grignard alkylations of GeBr4
with unfunctionalized aliphatic groups.13,14 Previous
work with fluorinated organometallics generally sup-
ports the chemical robustness of the aryl-CF3 function-
ality observed here;1b,8i,15 however, there have been
recent reports of CF3 group degradation in silicon
compounds.2d,e,16
Insertion reactions of both stable and transient ger-
mylenes to form new Ge-Ge and Ge-Si bonds have
been studied by several groups recently.17 We have
successfully employed this chemistry in the synthesis
of unsymmetrical digermanes 4 and 5. The reactions
were carried out using the known, stable germylene
complexes [2,4,6-(CF3)3C6H2]2Ge and [(Me3Si)2CH]2Ge
as substrates.5,18 Neither 4 nor 5 exhibited photosen-
sitivity under ambient conditions.
CF3 and CH3 groups,19 is also instructive. This complex
has the identical through-bond coupling path except for
the replacement of germanium by carbon. The observa-
tion of a greatly reduced magnitude for the coupling
constant further supports the argument that the Ge-F
close contacts play a key role in the observed coupling
constants for the germane complexes. Given the large
effect the Ge-F interactions appear to cause in the
magnitude of the coupling constants, we believe that
assigning this interaction as a five-bond coupling is
misleading. Therefore, we will use the previously
adopted practice of simply referring to these couplings
2
as J F-H
.
Compound 1 has four CF3 groups present, and a 13-
line spectrum is observed (J F-H ) 5.2 Hz) centered at
5.79 ppm. Not only is maximum coupling observed, but
also rapid exchange by all groups capable of interacting.
Using VT NMR in toluene-d8, an attempt was made to
observe the temperature dependence of the 19F-1H
coupling constants for molecules 1, 2, and 4. However,
upon cooling to -80 °C, the same multiplicities and
magnitude of coupling constants were observed. Similar
room-temperature NMR spectra were observed for
analogous silicon compounds synthesized by Edelmann
et al. and Braddock-Wilking et al.: [2,4,6-(CF3)3C6H2]2-
SiH2 and [2,4,6-(CF3)3C6H2]2SiF2.2d,e These compounds
showed the same rapid exchange behavior as well,
averaging the four ortho-CF3 groups to give 13-line
spectra for the proton or fluorine nuclei to which the
CF3 groups are coupled.
Discu ssion
A primary characterization tool for this series of
o-(trifluoromethyl)aryl germanes has been multinuclear
(19F and 1H), variable-temperature NMR. Over this
range of compounds, certain general trends are observed
in the NMR spectra. The number of CF3 groups
coupling to the Ge-H units are always the maximum
possible for the system. Complex 2, which contains only
a single ortho-CF3 group, gives a quartet for the GeH3
protons (J F-H ) 5.2 Hz) very similar in magnitude to
the couplings previously reported for similar silicon
compounds by Braddock-Wilking et al.2e Once again,
the question arises regarding the mechanism of the
coupling. Is an interaction of this magnitude consistent
The multinuclear VT NMR study of Grutzmacher et
al. on [2,4,6-(CF3)3C6H2]2Sn showed a weak temperature
dependence for the magnitude of the 19F-117Sn and 19F-
119Sn couplings.2a The magnitude of the change seen
for the 1J Sn-F coupling of [2,4,6-(CF3)3C6H2]2Sn was only
0.5%/10 °C, an amount we could not reliably resolve for
the much smaller magnitude J F-H in 1, 2, or 4. This
study provides additional direct evidence that the
coupling magnitudes in these [2,4,6-(CF3)3C6H2]-sub-
stituted compounds are sensitive to interatomic separa-
tion (through-space or weak F-E bond) and are not
simply classical through-bond couplings. A simulta-
neous differentiation between the four CF3 groups
present on [2,4,6-(CF3)3C6H2]2Sn was not reported in the
VT experiment; Schluter et al. failed to freeze out
conformations in an attempt at low-temperature 19F
NMR studies of In and Ga derivatives of [2,4,6-
(CF3)3C6H2]-.1c
5
with a classical J F-H coupling constant? Perhaps it
should be considered as a coupling mediated by the
weak Ge-F interaction and therefore better described
2
as a J F-H coupling constant. Comparison to other
coupling constants present in the same molecule is
informative. The aromatic protons ortho, meta, and
para to the CF3 group give rise to the possibility of
4
5
6
additional J F-H, J F-H, and J F-H couplings all within
the same molecule. This internal comparison is par-
ticularly useful because the aromatic protons in question
are geometrically constrained so that no close contacts
can occur with the CF3 group; however, they are also
located on the same aromatic ring system, so electronic
communication of the coupling should be at least as good
as that to germane hydrogen. Although some fine
structure exists on the quartet observed for the CF3
group in the 19F NMR spectrum, the coupling constants
between the fluorine and the aromatic ring hydrogen
atoms are <0.5 Hz, suggesting that the Ge-F close
contacts likely play an important role in the 5.2 Hz
coupling constant observed between CF3 and GeH3
groups. Comparison to o-(trifluoromethyl)toluene, which
exhibits a quartet (5J F-H ) 1 Hz) coupling between the
3
The J H-H coupling values for the unsymmetrically
substituted digermanes 4 and 5 (4.2 and 7.6 Hz,
3
respectively) are smaller than the J HSiGeH coupling of
11 Hz previously reported by Baines et al.20 The
couplings for the Ge-H resonances of 4 are observed
as broad multiplets (expected triskadectet of doublets),
which we have not been able to resolve sufficiently by
selective decoupling. A further comparison of the long-
range coupling magnitudes can be made using the CF3
groups in the meta positions in complexes 4 and 5. The
meta-CF3 groups form no close intramolecular contacts,
and the nominal five and six through-bond couplings
are < 0.1 Hz. Similar to the case described for 2, the
(15) Porwisiak, J .; Schlosser, M. Chem. Ber. 1996, 129, 233.
(16) Weidenbruch, M.; Will, P.; Marsmann, H.; Peters, K.; von
Schering, H. G. Z. Anorg. Allg. Chem. 1998, 624, 15, and references
therein.
(17) (a) Haberle, K.; Drager, M. J . Organomet. Chem. 1986, 312,
155. (b) Baines, K. M.; Cooke, J . A.; Vittal, J . J . J . Chem. Soc., Chem.
Commun. 1992, 1484.
(18) Fjeldberg, T.; Haaland, A.; Schilling, B. E. R.; Lappert, M. F.;
Thorne, A. J . J . Chem. Soc., Dalton Trans. 1986, 1551.
(19) Tordeux, M.; Langlois, B.; Wakselman, C. J . Chem. Soc., Perkin
Trans. 1990, 2293.
(20) Baines, K. M.; Groh, R. J .; J oseph, B.; Parshotam, U. R.
Organometallics 1992, 11, 2176.

o-(Trifluoromethyl)aryl Interactions
Organometallics, Vol. 17, No. 23, 1998 5169
F igu r e 3. ORTEP of 1 (50% probability). Bond lengths
(Å) and angles (deg) for molecule 2: Ge2-C28 ) 2.000(5);
Ge2-C19 ) 2.003(5); C19-Ge2-C28 ) 112.5(2); Ge2-F27
) 2.841(3); Ge2-F30 ) 2.841(3); Ge2-F36 ) 2.943(3);
Ge2-F21 ) 2.955(3). Average bond length lengths (Å) and
angles (deg) for all three molecules: Ge-C ) 1.998(5);
C-Ge-C ) 113.4(8).
F igu r e 4. ORTEP of 4 (50% probability). Bond lengths
(Å) and angles (deg): Ge1-Ge2 ) 2.4352(3); Ge2-C19 )
1.942(2); Ge2-C27 ) 1.952(2); Ge1-C1 ) 2.012(2); Ge1-
C10 ) 2.001(2); C19-Ge2-C27 ) 109.04(10); C1-Ge1-
C10 ) 112.42(9); C27-Ge2-Ge1-C10 ) 17.58(10); C19-
Ge2-Ge1-C1 ) 98.23(10); Ge1-F10 ) 2.837(2); Ge1-F18
) 3.024(2); Ge1-F2 ) 3.017(1); Ge2-F2 ) 2.947(1).
Ta ble 1. Cr ysta llogr a p h ic Da ta for 1 a n d 4
1
4
formula
a, Å
b, Å
C
18H6F18Ge
C34H12F30Ge2
21.7553(2)
10.90780(10)
16.5229(2)
90
102.0390(10)
90
3834.69(7)
4
21.2602(9)
18.0529(7)
16.4813(7)
90
106.3010(10)
90
c, Å
a, deg
b, deg
g, deg
V, Å3
Z
which is 99.95(10)° in the germylene and 113.4(8)° in
1.23 Steric repulsions in this Ge(IV) species easily
explain the deviation from 109.4°. This angle has only
opened up slightly for (Ph3P)2NiGe[2,4,6-(CF3)3C6H2]2
to 104.3°, consistent with the expected retention of Ge-
(II) character in this molecule.18,24
6071.4(4)
12
fw
3696
2200
space group
T, K
P21/c
133
P21/c
148
l, Å
0.710 73
2.090
0.710 73
1.967
r
calc, g cm-3
Like 1, 4 is also tetravalent, with a C-Ge-C angle
of 112.42(9)° on the side of the molecule derived from
[2,4,6-(CF3)3C6H2]2Ge (C1-Ge1-C10) and 109.04(10)°
on the side derived from [3,5-(CF3)2C6H3]2GeH2 (C19-
Ge2-C27). Ge-F contacts are still seen, but in com-
pletely different form (Figure 4). Ge1 has a close contact
at 2.836(2) Å (F10) and three others at 3.013(2), 3.017-
(1), and 3.025(2) Å. Ge2 experiences a distal Ge-F
contact, from an ortho-CF3 group leaning over to contact
it at 2.947(1) Å (F2), in preference to the expected Ge-F
contact at Ge1 based on the other structurally charac-
terized Ge derivatives described in this paper. The
closest contact of F2 to Ge1 is 3.017(1) Å. The hydrogen
atoms were located in 4 and lie in the expected tetra-
hedral positions on Ge1 and Ge2. The closest approach
of any fluorine to the hydrogens is 2.44 Å, right on the
edge for a weak van der Waals contact (2.65 Å).22 Two
weak intermolecular F-H-Ge contacts are observed at
2.82(3) (H1-F28) and 2.73(3) (H2-F25) Å and five weak
intramolecular F-H-Ge contacts of 2.53(3) (H1-F8),
2.64(3) (H1-F17), 2.44(3) (H1-F18), 2.89(3) (H2-F2),
and 2.62(3) (H2-F3) Å. Finally, the dihedral angle
defined by C27-Ge2-Ge1-C10 is only 17.6°, very close
to eclipsed, again suggesting that steric interactions are
being overcome to satisfy a lower electronic energy state
arising from the Ge-F contact. The two different Ge-F
contacts seen in 4 are consistent with the solution NMR
data, which also show H-F couplings of differing
magnitudes. However, it is also clear from the solution
R1
wR2
0.0773
0.1630
0.0393
0.0907
nominal five-bond couplings are of sufficient magnitude
to detect when a Ge-F close contact effectively shortens
the path to a two-bond interaction. However, classical
five-bond couplings are not observed in the absence of
the Ge-F close contact.
Single-crystal X-ray structure determinations were
performed on compounds 1 and 4 (see Table 1). For the
previously reported [2,4,6-(CF3)3C6H2]2Ge,5 four Ge-F
contacts were seen, two at 2.6 Å and two at 2.8 Å. So
far, this is the only germanium derivative of the [2,4,6-
(CF3)3C6H2]- ligand for which we have seen four close
contacts and the only one for which contacts at 2.6 Å
are seen. In compound 1 (Figure 3), a tetravalent
analogue of [2,4,6-(CF3)3C6H2]2Ge with the smallest
possible additional groups attached (H), only two con-
tacts are retained at 2.8 Å. In another compound
synthesized in our laboratory, (Ph3P)2NiGe[2,4,6-
(CF3)3C6H2]2,21 the [2,4,6-(CF3)3C6H2]2Ge germylene ex-
ists in a dative bonding mode, albeit a very sterically
crowded one, and it also retains only two close Ge-F
contacts at 2.8 Å. Two additional contacts for 1 and
(Ph3P)2NiGe[2,4,6-(CF3)3C6H2]2, which are also within
the sum of the van der Waals radii (3.66 Å),22 are seen
at 2.943 and 2.957 Å and 3.003 and 3.057 Å, respec-
tively. Another major comparison between the Ge(II)
[2,4,6-(CF3)3C6H2]2Ge and 1 is the C-Ge-C angle,
(21) Bender, J . E.; Litz, K. E.; Banaszak Holl, M. M.; Kampf, J . W.
Submitted.
(22) Bondi, A. J . Phys. Chem. 1964, 68, 441.
(23) The average value for the three independent molecules in the
unit cell is given.
(24) Grev, R. S.; Schaefer, H. F. Organometallics 1992, 11, 3489.

5170 Organometallics, Vol. 17, No. 23, 1998
Bender et al.
NMR data that these couplings are in systems where
the CF3 interactions are averaged, so the fixed bond
lengths obtained in the structure determination of 4
cannot be quantitatively related to the NMR data.
Se dimer, being homonuclear and essentially a covalent
molecule, is most relevant. 4 and [2,4,6-(CF3)3C6H2]Se-
Se[2,4,6-(CF3)3C6H2] certainly have many other confor-
mations available to them that would not involve M-F
contacts. There are several other examples of such
distal contacts in the polynuclear literature of the [2,4,6-
(CF3)3C6H2]- substituent, though these remaining ex-
amples only involve single metal centers.30
Simple force field calculations (CAChe-Mechanics and
Spartan-Merck Molecular Force Field) performed on
1, 2, and 4 indicate that the spatial constraints imposed
by placing a CF3 group in the ortho position bring the
Ge and F to within ∼3.1 Å. This value can be used as
an upper bound to rule out any bonding character in
contacts of that length or greater which appear other-
wise to be van der Waals contacts. In a recent review,
Plenio empirically derived 3.0 Å as an upper limit for
an assignable Ge-F contact.25 For 1 and 4, this
qualifies several contacts at <3.0 Å as possible bonding
interactions. In particular, the preference for a contact
between Ge2 and F2 in 4 at 2.947 Å cannot be explained
by steric hindrance or crystal-packing arguments. The
very acute dihedral angle seen in 4 between C27-Ge2-
Ge1-C10 is predicted by force field calculations to be
35°, twice as great as observed experimentally. As is
clearly the case for the more Lewis acidic [2,4,6-
(CF3)3C6H2]2Ge, both 1 and 4 place their CF3 groups
closer to the Ge centers than is necessary by steric
requirements alone. We do not believe that weak F-H
intermolecular contacts play a significant role in deter-
mining the observed dihedral angle. The contacts
appear too long to be considered significant bonding
interactions, particularly when compared to the short
Ge-F contacts and the presence of comparitively shorter
F-H-Ge intramolecular contacts.
The literature of weak-bonding interactions of fluorine-
containing molecules has become quite active recently.25
A system of significant relevance is that of the stabilized
Ziegler-Natta catalysts, though here the metal centers
are cationic and so have enhanced Lewis acidity toward
fluorine.26 Numerous mononuclear compounds contain-
ing the [2,4,6-(CF3)3C6H2]- unit have been synthesized,
though there are few structurally characterized ex-
amples with relevance to 4. A series of papers published
by Roesky and co-workers describes multinuclear de-
rivatives of [2,4,6-(CF3)3C6H2]- similar to 4. [2,4,6-
(CF3)3C6H2]Se-Se[2,4,6-(CF3)3C6H2] was reported, and
the crystal data reveal that this compound contains a
distal M-F contact.27 This contact is in fact the shortest
Se-F contact in the molecule, at 3.07 Å. Such distal
contacts are also seen in the compounds [2,4,6-
(CF3)3C6H2]S-Tl‚dioxane (Tl-F at 3.123 and 3.174 Å)28
and [2,4,6-(CF3)3C6H2]S-Na‚2THF (Na-F at 2.434,
2.436, 2.486, and 2.571 Å).29 These last two compounds
crystallize as ladder polymers, where the contacts
appear to provide considerable stabilization of the
chains. These compounds also leave the CF3 group with
no choice as to which Lewis basic site it may choose to
interact with; the relevance here is the ability to conquer
conformational flexibility to form the M-F contact. The
Con clu sion s
The weak Ge-F interaction highlighted by the set of
complexes presented herein is quite intriguing. For
cases such as 4, the steric demands of the [2,4,6-
(CF3)3C6H2]- substituent alone cannot account for such
contacts. These data are also consistent with a large
amount of previously published work on o-(trifluoro-
methyl)aryl-substituted compounds showing that such
groups stabilize unusual low-valent main group com-
plexes, complexes that are not otherwise stable in the
absence of ortho-CF3 groups. The series of new ger-
manes presented have F-H coupling constant magni-
tudes that are highly dependent upon the presence or
absence of close Ge-F contacts. Although the couplings
mediated by the contacts are significantly weaker than
analogous couplings involving more standard covalent
bonds, direct comparisons within the set of complexes
also show the couplings to be signicantly greater than
expected for a classical through-bond pathway. As a
prelude to polymerization studies using (trifluoro-
methyl)aryl-substituted polygermanes, we wished to
make a concerted investigation of small molecule model
compounds that fills in some of the knowledge gaps
existing in the literature of (trifluoromethyl)aryl ger-
manes. The ortho-CF3 group has clearly demonstrated
its ability to interact with both the germanium atom to
which the ligand is attached as well as neighboring
germanium atoms. These interactions hold out the
promise of interesting properties in related polyger-
manes. The excellent thermal stability of these mol-
ecules may foreshadow a similar stability being im-
parted to chains. We are currently pursuing higher
level calculations and electronic spectroscopy to examine
the electronic effects contributing to the tendency for
Ge-F close contacts in these small molecules and in
polygermanes.
Exp er im en ta l Section
All air-free manipulations were carried out using standard
vacuum line and drybox techniques. All solvents were distilled
from Na/benzophenone just prior to use. GeBr4,31 GeCl2‚
dioxane,32 [2,4,6-(CF3)3C6H2]2Ge,5 (Et3P)2NiGe[N(SiMe3)2]2,12
Ge[CH(SiMe3)2]2,18 and [3,5-(CF3)2C6H3]2GeH214 were prepared
according to literature procedures. 2-(CF3)C6H4Br and Mg
(30) (a) Bertel, N.; Noltemeyer, M.; Roesky, H. W. Z. Anorg. Allg.
Chem. 1990, 588, 102. (b) Labahn, D.; Brooker, S.; Sheldrick, G. M.;
Roesky, H. W. Z. Anorg. Allg. Chem. 1992, 610, 163. (c) Edelmann,
A.; Brooker, S.; Bertel, N.; Noltemeyer, M.; Roesky, H. W.; Sheldrick,
G. M.; Edelamnn, F. T. Z. Naturforsch. 1992, 47b, 305. (d) Labahn,
D.; Bohnen, F. M.; Herbst-Irmer, R.; Pohl, E.; Stalke, D.; Roesky, H.
W. Z. Anorg. Allg. Chem. 1994, 620, 41. (e) Voelker, H.; Pieper, U.;
Roesky, H. W.; Sheldrick, G. M. Z. Naturforsch. 1994, 49b, 255. (f)
Dillon, K. B.; Gibson, V. C.; Howard, J . A. K.; Sequeira, L. J .; Yao, J .
W. Polyhedron 1996, 15, 4173.
(25) Plenio, H. Chem. Rev. 1997, 97, 3363.
(26) For example, see: Karl, J .; Erker, G.; Frohlich, R. J . Am. Chem.
Soc. 1997, 119, 11165, and references therein.
(27) Bertel, N.; Roesky, H. W.; Edelmann, F. T.; Noltemeyer, M.;
Schmidt, H. G. Z. Anorg. Allg. Chem. 1990, 586, 7.
(28) Labahn, D.; Pohl, E.; Herbst-Irmer, R.; Stalke, D.; Roesky, H.
W.; Sheldrick, G. M. Chem. Ber. 1991, 124, 1127.
(29) Brooker, S.; Edelmann, F. T.; Kottke, T.; Roesky, H. W.;
Sheldrick, G. M.; Stalke, D.; Whitmire, K. H. J . Chem. Soc., Chem.
Commun. 1991, 144.
(31) Curtis, M. D.; Wolber, P. Inorg. Chem. 1972, 11, 431.
(32) Viktorov, N. A.; Gar, T. K.; Nikishina, I. S.; Nosova, V. M.;
Ivashchenko, D. A.; Mironov, V. F. Zh. Obshch. Khim. 1986, 325, 13.

o-(Trifluoromethyl)aryl Interactions
Organometallics, Vol. 17, No. 23, 1998 5171
turnings were purchased from Aldrich and used as received.
1H NMR spectra were recorded at 400 MHz on a Varian Inova
400 spectrometer and referenced to the residual protons in
C6D6 at 7.15 ppm, in THF-d8 at 3.78 ppm and in toluene-d8 at
2.09 ppm. 13C NMR spectra were recorded at 100.581 MHz
and referenced to the natural abundance of 13C in C6D6 or
THF-d8 at 128.00 and 67.57 ppm, respectively. 19F NMR
spectra were recorded at 376.321 MHz and referenced exter-
nally to CF3C6H5 in C6D6 at -63.73 ppm. IR spectra were
recorded on a Nicolet 5DXB spectrometer. UV-vis spectra
were recorded on an HP model 8425A diode-array spectropho-
tometer. Mass spectra were collected on an VG Analytical
model 70-S spectrometer. Elemental analyses were performed
in-house on a Perkin-Elmer model 2400 analyzer.
(12.2 g, 0.0312 mol) over a period of 1 h and stirred a further
24 h. A brown solution resulted with formation of a salt
precipitate. The ether was removed under vacuum, and the
crude solids were extracted with 160 mL of hexane. The
hexane was removed under vacuum to give 8.4 g of a crude
brown oil (59% crude yield). Vacuum distillation at 60 °C gave
a clear, colorless liquid in analytical purity (53.5% yield). 1H
NMR (C6D6): δ 7.75 (d, 1H, J H-H ) 7.2 Hz, H-Ar), 7.12 (d, 1H,
J H-H ) 7.6 Hz, H-Ar), 6.75 (t, 1H, J H-H ) 7.2 Hz, H-Ar), 6.68
(t, 1H, J H-H ) 7.6 Hz, H-Ar). 19F NMR (C6D6): δ -56.80 (s,
3F, CF3). 13C{1H} NMR (C6D6): δ 134.93 (s), 132.88 (s), 130.19
2
1
(q, J F-C ) 31 Hz, C-CF3), 127.17 (s), 123.63 (q, J F-C ) 273
Hz, CF3), 120.14 (s). 13C{1H} NMR (CDCl3): δ 134.41 (s),
133.20 (s), 132.77 (q, 2J F-C ) 32.6 Hz, C-CF3), 132.19 (s), 128.14
3
1
(q, J F-C ) 4.6 Hz, CH-C-CF3), 123.94 (q, J F-C ) 274 Hz,
CF3). EIMS (70 eV): 439 (M•+), 377 (M - Br.+). IR (neat,
cm-1): 3071 (C-H), 1180/1138 (C-F). Anal. Calcd for C7H4-
Br3F3Ge: C, 18.4; H, 0.8. Found: C, 18.28; H, 0.97.
[2,4,6-(CF 3)3C6H2]2GeH2 (1). A 300 mg (0.473 mmol)
sample of [2,4,6-(CF3)3C6H2]2Ge and 30 mg (0.043 mmol) of
(Et3P)2NiGe[N(SiMe3)2]2 were placed in a small, round-bottom
flask. A 10 mL sample of benzene was frozen into the flask
under vacuum and allowed to thaw under 10+ equiv of H2 gas
at 1 atm. The thawed solution was allowed to stir 16 h, giving
a dark brown color. The flask was recharged with H2 at 1 atm
and allowed to stir an additional 16 h. The solvent was then
removed under vacuum, taking care not to overdry, as the solid
product is highly sublimable at 10-3 Torr. Colorless crystals
were isolated from the crude solids by sublimation at 25 °C
onto a water-cooled probe (165 mg, 55% yield). A crystal
suitable for X-ray diffraction was selected from the sublimed
material. 1H NMR (C6D6): δ 7.76 (s, 4H, m-H-Ar), 5.79
[3,5-(CF 3)2C6H3]2Ge(H)Ge(H)[2,4,6-(CF 3)3C6H2]2 (4).
A
208 mg (0.328 mmol) sample of [2,4,6-(CF3)3C6H2]2Ge and 160
mg (0.328 mmol) of [3,5-(CF3)2C6H3]2GeH2 were refluxed in
benzene for 15 h. The solvent was distilled off under vacuum,
giving a waxy solid. Side products were sublimed away at 95
°C for 12 h followed by a second sublimation at 120 °C to a
water-cooled probe over 2 days, yielding a colorless micro-
crystalline solid (175 mg, 67% yield). Crystals suitable for
X-ray diffraction were grown from slow evaporation of a dilute
benzene solution. 1H NMR (C7D8): δ 7.81 (s, 4H o-C-H), 7.74
(s, 4H, m-C-H), 7.62 (s, 2H, p-C-H), 6.38 (1H, m, H-Ge[2,4,6-
(CF3)3C6H2]2), 5.72 (1H, m, H-Ge[3,5-(CF3)2C6H3]2). 19F NMR
2
(triskadectet, 2H, J F-H ) 5.1 Hz, H-Ge). 1H NMR (THF-d8):
δ 8.41 (s, 4H, m-H-Ar), 5.85 (triskadectet, 2H, 2J F-H ) 5.1 Hz,
1
2
(C7D8): δ -57.14 (dd, 12F, J F-H ) 5.6 Hz, J F-H ) 1.9 Hz,
o-CF3), -64.46 (s, 12F, m-CF3), -65.02 (s, 6F, p-CF3). Elem.
Anal. Calcd for C34H12F30Ge2: C, 36.0; H, 1.1. Found: C, 36.1;
H, 1.1.
2
H-Ge). 19F NMR (THF-d8): δ -55.51 (t, 12F, J H-F ) 5.1 Hz,
o-CF3), -61.95 (s, 6F, p-CF3). IR (KBr, cm-1): 2186 (Ge-H).
CIMS (methane): m/e 637 (M - H)+, 619 (M - F)+, 357 (M -
C9H2F9)+. Anal. Calcd for C18H6F18Ge: C, 33.9; H, 0.95.
Found: C, 34.0; H, 1.1.
Str u ctu r e d eter m in a tion for 4. A crystal was selected
with dimensions of 0.20 × 0.32 × 0.38 mm, space group P21/c,
a ) 21.7553(2) Å, b ) 10.90780(10) Å, c ) 16.5229(2) Å, â )
102.0390(10)°, V ) 3834.69(7) Å3, Z ) 4, µ(Mo KR) ) 1.742
mm-1, 40 515 reflections measured, 2θmax ) 29.57°, T ) 148
K, empirical absorption correction (SADABS), 9906 unique
reflections, refined in full matrix on F2. All non-hydrogen
atoms were anisotropically refined, with H atoms in idealized
positions. R1 ) 0.0393, wR2 ) 0.0907 (I > 2σI); R1 ) 0.0511,
wR2 ) 0.0980 (all data).
Str u ctu r e d eter m in a tion for 1. A crystal was selected
with dimensions of 0.08 × 0.16 × 0.18 mm, space group P21/c,
a ) 21.2602(9) Å, b ) 18.0529(7) Å, c ) 16.4813(7) Å, â )
106.3010(10)°, V ) 6071.4(4) Å3, Z ) 12, µ(Mo KR) ) 1.684
mm-1, 56 886 reflections measured, 2θmax ) 29.48°, T ) 133
K, empirical absorption correction (SADABS), 15 374 unique
reflections, refined in full matrix on F2. All non-hydrogen
atoms were anisotropically refined, with H atoms in idealized
positions. R1 ) 0.0773, wR2 ) 0.1630 (I > 2σI); R1 ) 0.1353,
wR2 ) 0.1937 (all data).
[(Me3Si)2CH]2Ge(H)Ge(H)[3,5-(CF 3)2C6H3]2 (5). A 235
mg sample of Ge[CH(SiMe3)2]2 (0.60 mmol) and 265 mg of [3,5-
(CF3)2C6H3]2GeH2 (0.53 mmol) were allowed to stir at 20 °C
in hexane for 16 h. The solvent was removed under vacuum;
the solids were recrystallized from cold pentane (-78 °C). 5
was obtained as a white, powdery solid after vacuum-drying
(305 mg, 68% yield). 1H NMR (C6D6): δ 8.12 (s, 4H, o-CH),
[2-(CF 3)C6H4]GeH3 (2). A suspension of LiAlH4 (1.00 g,
0.026 mol) in 50 mL of diethyl ether was generated by allowing
a pellet to slowly break up with stirring over a period of 4 h
at 20 °C under Ar. To this solution was added 2-(CF3)C6H4-
GeBr3 (7.25 g, 0.016 mol) in 10 mL of ether over a period of 30
min. The volume of solution was reduced to 10 mL total and
then extracted into a new flask with hexane (2 × 50 mL). The
hexane and ether were then removed under vacuum, and the
crude oil was further purified by condensing onto a coldfinger
under vacuum at -78 °C. 2 was obtained as a colorless, clear
liquid in 39% yield (1.34 g). 1H NMR (C6D6): δ 7.37 (d, 1H,
J H-H ) 3.9 Hz, H-Ar), 7.23 (d, 1H, J H-H ) 4.0 Hz, H-Ar), 6.87
3
7.73 (s, 2H, p-CH), 5.31 (d, 1H, J H-H ) 7.6 Hz, H-Ge[3,5-
3
3
(CF3)2C6H3]2), 4.83 (dt, 1H, J HGeGeH ) 7.6 Hz, J HGeCH ) 2.4
Hz, H-Ge[CH(SiMe3)2]2), 0.04 (s, 18H, Si(CH3)3), 0.00 (s, 18H,
Si(CH3)3), -0.19 (d, 2H, J HCGeH ) 2.4 Hz, H-C(SiMe3)2). 19F
3
NMR (C6D6): δ -67.44 (s, 12F, m-CF3). 13C {1H} NMR
(C6D6): δ 140.46 (s, i-Ar-CH), 135.5 (m, o-Ar-CH), 131.91 (q,
2
1
C-CF3, J F-C ) 33.1 Hz), 123.81 (q, CF3, J F-C ) 273.1 Hz),
2
(m, 2H, H-Ar), 4.36 (q, J F-H ) 5.6 Hz, GeH3). 19F NMR
3
123.31 (q, Ar-C-C-CF3, J F-C ) 4.2 Hz), 4.67 (s, CH(SiMe3)2),
2
(C6D6): δ -61.48 (q, 3F, J H-F ) 5.6 Hz, CF3). 13C{1H} NMR
3.20 (s, (CH3)3Si), 2.48 (s, (CH3)3Si). IR (Nujol, cm-1): 2052,
2003, 1982 (Ge-H). Elem. Anal. Calcd for C30H46F12Ge2Si4:
C, 40.4; H, 5.2. Found, C, 40.0; H, 5.0.
(C6D6): δ 138.26 (s), 134.43 (q, 2J F-C ) 30.5 Hz, C-CF3), 131.16
(s), 130.7 (q, 3J F-C ) 5 Hz, C(GeH3), 129.37 (s), 126.08, (q, 3J F-C
1
) 5.3 Hz, CH-C-CF3), 125.32 (q, J F-C ) 274 Hz, CF3). IR
(neat, cm-1): 3065 (C-H), 2095 (Ge-H), 1173/1124 (C-F).
EIMS (70 eV): 221 (M - H•+). CIMS (methane, relative
intensity): 238 (M + CH4•+, 9.1). Anal. Calcd for C7H7F3Ge:
C, 38.1; H, 3.2. Found: C, 37.81; H, 3.18. Overall yield from
GeBr4: 23%.
Ack n ow led gm en t. J .E.B. thanks the NSF for the
support of a graduate fellowship. A.M. thanks the NSF
for support through the REU program. K. E. Litz and
M. D. Curtis are thanked for many helpful discussions.
[2-(CF 3)C6H4]GeBr 3 (3). 2-(CF3)C6H4MgBr was generated
by slowly adding 6.94 g of 2-(CF3)C6H4Br (0.0308 mol) to a
suspension of Mg turnings (0.800 g, 0.033 mol) in 50 mL of
diethyl ether. The majority of the Mg reacted, and the solution
became opaque brown after 3 h. This Grignard solution was
then slowly added to a 100 mL diethyl ether solution of GeBr4
Su p p or tin g In for m a tion Ava ila ble: Complete lists of the
crystallographic data for compounds 1 and 4 (27 pages). See
any current masthead page for ordering information and Web
access instructions.
OM980430M