Synthetic, structural and theoretical studies of amidinate and guanidinate stabilised germanium(I) dimers

Article abstract of DOI:10.1039/b610645e

The neutral germanium(I) dimers, [{Ge(Piso)}₂] and [{Ge(Giso)}₂], Piso = [(ArN)₂CBu^t]^-, Giso = [(ArN)₂CNPrⁱ₂]^-, Ar = C ₆H₃Prⁱ

Full text of DOI:10.1039/b610645e

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Synthetic, structural and theoretical studies of amidinate and
guanidinate stabilised germanium(I) dimers{
Shaun P. Green,a Cameron Jones,*a Peter C. Junk,b Kai-Alexander Lipperta and Andreas Staschab
Received (in Cambridge, UK) 25th July 2006, Accepted 24th August 2006
First published as an Advance Article on the web 8th September 2006
DOI: 10.1039/b610645e
The neutral germanium(I) dimers, [{Ge(Piso)}2] and
[{Ge(Giso)}2], Piso
= = [(ArN)2-
[(ArN)2CBut]2, Giso
CNPri2]2, Ar = C6H3Pri2-2,6, which are stabilised by bulky
amidinate and guanidinate ligands respectively, have been
prepared by reduction of the corresponding germanium(II)
chlorides, [Ge(Piso)Cl] and [Ge(Giso)Cl]; theoretical studies
suggest that the Ge–Ge bonds of [{Ge(Piso)}2] and
[{Ge(Giso)}2] are associated with their HOMOs, whilst their
LUMOs have substantial Ge–Ge p-bonding character.
The heavier group 14 analogues of alkynes have attracted much
interest in recent years.1 A number of compounds of the type
REER (E = Si,2 Ge,3,4 Sn5 or Pb;6 R = bulky monodentate
organyl) have now been prepared, mainly by reduction of REX,
X = halide. When less bulky substituents are employed in these
reductions, oligomeric cage compounds, (RE)n, are often formed.7
The nature of the E–E bonds in REER, especially with respect to
multiple bonding, has been analysed in many theoretical studies.8
Unlike alkynes, these compounds display trans-bent geometries
with decreasing CEE angles as the molecular weight of the group
14 element increases. In the case of digermynes, recent theoretical
studies point towards Ge–Ge interactions having bond orders of
ca. 2 and partial singlet diradicaloid character,8a the latter of which
is consistent with the results of digermyne reactivity studies.9
Distannynes and diplumbynes, on the other hand, probably have
E–E bond orders close to 1, little or no diradicaloid character and
LUMOs corresponding to empty E–E p-bonding orbitals.8a
Indeed, when distannynes are doubly reduced, their Sn–Sn
interactions can shorten.10
Scheme 1 Reagents and conditions: i, GeCl2?dioxane, 2 LiCl; ii, excess
K, 2 KCl.
their solid state structures. That for 1 is depicted in Fig. 1{ and
shows it to be monomeric with a chelating, delocalised amidinate
ligand and a Cl–heterocycle plane angle of 104.6u (cf. 2 102.8u).
The structures of 1 and 2 are reminiscent of related b-diketiminate
complexes, e.g. [GeCl{[(Ar)NC(Me)]2CH}].13
Reductions of 1 and 2 with potassium mirrors in toluene at
room temperature over 3–4 hours afforded deeply coloured
solutions of the germanium(I) dimers, 3 and 4, respectively.
We have had recent successes in using bulky amidinate and
guanidinate ligands in the stabilisation of group 13 metal(I)
compounds.11 It seemed that such ligands may also have utility in
the stabilisation of novel dimeric germanium(I) compounds,
analogous to digermynes, RGeGeR. Our preliminary efforts in
this direction are reported herein.
The reactions of the lithium amidinate and guanidinate salts,
Li[Piso] and Li[Giso],11 Piso = [(ArN)2CBut]2, Giso = [(ArN)2-
CNPri2]2, Ar = C6H3Pri2-2,6, with GeCl2?dioxane yielded the
germanium(II) complexes, 1 and 2,12 in good yields (Scheme 1).{
The spectroscopic data for both complexes are consistent with
aCentre for Fundamental and Applied Main Group Chemistry, School of
Chemistry, Main Building, Cardiff University, Cardiff, UK CF10 3AT.
E-mail: jonesca6@cardiff.ac.uk
Fig. 1 Molecular structure of 1 (hydrogen atoms omitted). Selected
˚
bSchool of Chemistry, Monash University, Melbourne, PO Box 23,
Victoria 2800, Australia
bond lengths (A) and angles (u): Ge(1)–Cl(1) 2.174(2), Ge(1)–N(1)
2.005(3), Ge(1)–N(2) 2.003(4), N(1)–C(1) 1.334(5), N(2)–C(1) 1.356(5);
N(1)–C(1)–N(2) 106.9(3), N(2)–Ge(1)–N(1) 65.25(14), N(2)–Ge(1)–Cl(1)
101.83(18), N(1)–Ge(1)–Cl(1) 98.21(18).
{ Electronic supplementary information (ESI) available: Full synthetic
details for 1–4, spectroscopic data for 1 and 2, molecular structures of 2
and 4, and details of the theoretical study. See DOI: 10.1039/b610645e
3978 | Chem. Commun., 2006, 3978–3980
This journal is ß The Royal Society of Chemistry 2006

View Article Online
Recrystallisation of the crude products from these reactions
afforded dichromic, green–red crystals of 3, and lime-green crystals
of 4, in low yield (Scheme 1). They are both remarkably thermally
stable, with decomposition temperatures . 200 uC, but are
extremely sensitive towards oxygen and moisture. It is of note that
attempts to prepare Si, Sn and Pb analogues of 3 and 4 by similar
routes have so far been unsuccessful.
The NMR spectra§ of the complexes indicate that their solid
state structures are retained in solution. Of most note are their 1H
and 13C{1H} NMR spectra which display signals for four
chemically inequivalent Ar methyl, and two chemically inequi-
valent Ar methine centres. As both compounds have closely
related structures, only that for 3 is shown here (Fig. 2). As in 1, its
amidinate ligand is apparently delocalised but it has longer Ge–N
interactions than in that compound, presumably because of the
expected larger covalent radii of its Ge(I) centres. The Ge–Ge bond
˚
length of 3 (2.6380(8) A, cf. 2.6721(13) A for 4) is in the normal
˚
14
region for single Ge–Ge interactions (2.61 A mean ) but
˚
significantly longer than is typical for digermenes, R2GeGeR2
14
(2.21–2.51 A ) and the two structurally characterised digermynes
˚
3
4
˚
˚
(2.2850(6) A and 2.2060(8) A ). These observations suggest there
is no Ge–Ge multiple bond character in 3 and 4, as do their rather
acute C(1)–Ge–Ge angles of 100.7u and 100.8u respectively
(cf. 128.7u mean in Ar*GeGeAr*, Ar* = C6H3(Ar)2-2,63). In fact,
these angles resemble more the CPbPb angles (94.3u mean) in the
singly Pb–Pb bonded compound, Ar0PbPbAr0, Ar0 = C6H3(Trip)2-
2,6; Trip = C6H2Pri3-2,4,6.6
In order to shed light on the nature of the Ge–Ge interactions in
3 and 4, DFT calculations were performed on the model complex
[{Ge[(Ar9N)2CMe]}2] (Ar9 = C6H3Me2-2,6).{ The optimised
structure is in good agreement with the geometries of 3 and 4,
Fig. 3 Representations and energies of (a) the LUMO, (b) the HOMO
and (c) the HOMO-4 of the model system [{MeC(NAr9)2Ge}2] (Ar9 =
2,6-Me2C6H3).
though with slightly overestimated Ge–Ge and Ge–N bond
˚
˚
lengths (2.679 A; 2.133–2.137 A respectively). An NBO analysis
of the model showed the Ge–Ge bond to be associated with the
HOMO (Fig. 3b) and to be largely derived from Ge p-orbital
overlap (s 17.43%, p 81.74%). Therefore, this bond is almost
exclusively s in character, as is the case for the Pb–Pb bonds in
diplumbynes.8a In the LUMO, empty p-orbitals at the Ge centres
combine to form a Ge–Ge p-bonding component (Fig. 3a), which
is anti-bonding toward the Ge–ligand interactions. Similar
p-bonding LUMOs have been calculated for distannynes.8a
The lone pairs at the Ge centres (Fig. 3c) are high in s-character
(sp0.31/0.30) but have some directionality, whilst the Ge–N bonds
are heavily polarised (Ge +0.54, N 20.69 mean) towards the
Fig. 2 Molecular structure of 3 (hydrogen atoms omitted). Selected
˚
bond lengths (A) and angles (u): Ge(1)–Ge(1)9 2.6380(8), Ge(1)–N(1)
2.032(2), Ge(1)–N(2) 2.049(2), N(1)–C(1) 1.339(3), N(2)–C(1) 1.339(3);
N(2)–C(1)–N(1) 106.9(2), N(1)–Ge(1)–N(2) 63.61(9), N(1)–Ge(1)–Ge(1)9
96.74(7), N(2)–Ge(1)–Ge(1)9 97.70(7). Symmetry operation9: 2x + 2, y,
2z + K.
This journal is ß The Royal Society of Chemistry 2006
Chem. Commun., 2006, 3978–3980 | 3979

View Article Online
Dc = 1.182 g cm23, F(000) = 1148, m(Mo-Ka) = 1.041 mm21, 150(2) K,
5573 unique reflections [Rint 0.0960], R (on F) 0.0761, wR (on F2) 0.1116 (I
. 2s(I)). CCDC 616057–616060. For crystallographic data in CIF or other
electronic format see DOI: 10.1039/b610645e
nitrogen atoms (97% mean N-electron contribution). In addition, a
Wiberg bonding analysis revealed a Ge–Ge bond order of 0.8525
and an average Ge–N bond order of 0.439. Finally, the HOMO–
LUMO gap (1.913 eV; equiv. to lmax 647 nm) is of a similar energy
to an absorption band observed in the visible spectrum of 3 (lmax
568 nm).
§ Selected data for 3: Yield: 16%. Mp . 300 uC. 1H NMR (400 MHz,
C6D6, 298 K): d 0.41 (d, 3JHH = 6.8 Hz, 12 H, CH(CH3)2), 0.96 (s, 18 H,
C(CH3)3), 1.18 (d, 3JHH = 6.8 Hz, 12 H, CH(CH3)2), 1.37 (d, 3JHH = 6.8 Hz,
3
12 H, CH(CH3)2), 1.62 (d, JHH = 6.8 Hz, 12 H, CH(CH3)2), 3.56 (sept,
3
3JHH = 6.8 Hz, 4 H, CH(CH3)2), 3.79 (sept, JHH = 6.8 Hz, 4 H,
In light of the p-bonding LUMO of the model germanium
dimer and the fact that heavier group 14 alkyne analogues can be
singly or doubly reduced,10 DFT calculations were carried out on
the anionic species, [{Ge[(Ar9N)2CMe]}2]1 or 22. For the optimised
singly reduced model complex, the Ge–Ge bond length shortened
˚
by 0.09 A, while one Ge–N bond at each metal centre lengthened
˚
by ca. 0.5 A. This is to be expected since a population analysis
CH(CH3)2), 6.81–7.21 (m, 12 H, ArH); 13C{1H} NMR (75.6 MHz, C6D6,
298 K): d 23.5 (CH(CH3)2), 23.9 (CH(CH3)2), 24.4 (CH(CH3)2), 28.1
(CH(CH3)2), 28.4 (CH(CH3)2), 28.8 (CH(CH3)2), 29.9 (C(CH3)3), 42.5
(C(CH3)3), 123.6, 123.7, 126.1, 141.5, 144.2, 145.0 (ArC), 165.5 (backbone
CN2); UV/Vis: lmax/nm (c = 0.006 mol l21): 348 (intense), 438 (e = 501), 568
(e = 103); MS (EI 70 eV), m/z (%): 493.4 (K M+, 36), 420.4 (PisoH+, 6),
244.3 (PisoH+ 2 ArNH, 100); IR n/cm21 (Nujol): 1616 (s), 1586 (m), 1322
(m), 1260 (m), 1211 (m), 1171 (m), 1028 (m), 799 (s), 760 (m); 4: Yield: 13%.
Mp 200–203 uC (decomp.); 1H NMR (400 MHz, C6D6, 298 K): d 0.70 (d,
showed the molecular orbital energies and ordering to be close to
those of the neutral dimer, but with the additional electron
populating the LUMO which possesses bonding character between
the Ge centres and anti-bonding character between the Ge centres
and the amidinate ligands. Optimisation was unsuccessful for the
doubly reduced species, but it may be reasonable to propose
further strengthening of the Ge–Ge bond and destabilisation of the
Ge–ligand interactions. This is in line with our experimental
observation that reduction of 3 and 4 leads to their decomposition
to mixtures of products, including elemental germanium, K[Piso]
or K[Giso].
3
3JHH = 6.8 Hz, 12 H, CH(CH3)2), 0.93 (d, JHH = 6.8 Hz, 24 H,
3
CH(CH3)2), 1.33 (d, JHH = 6.8 Hz, 12 H, CH(CH3)2), 1.53 (d, JHH
3
=
6.8 Hz, 12 H, CH(CH3)2), 1.78 (d, 3JHH = 6.8 Hz, 12 H, CH(CH3)2), 3.89
3
3
(sept, JHH = 6.8 Hz, 4 H, CH(CH3)2), 4.03 (sept, JHH = 6.8 Hz, 4 H,
CH(CH3)2), 4.21 (sept, 3JHH = 6.8 Hz, 4 H, CH(CH3)2), 6.98–7.38 (m, 12 H,
ArH); 13C{1H} NMR (100.6 MHz, C6D6, 298 K): d 20.4 (CH(CH3)2), 20.7
(CH(CH3)2), 21.3 (CH(CH3)2), 23.0 (CH(CH3)2), 26.6 (CH(CH3)2), 26.7
(CH(CH3)2), 27.5 (CH(CH3)2), 46.5 (NCH(CH3)2), 121.4, 122.0, 126.5,
138.6, 144.5, 147.3 (ArC), 152.3 (backbone CN2); IR n/cm21 (Nujol): 1614
(m), 1581 (m), 1330 (s), 1280 (s), 1125 (s), 1046 (m), 800 (s), 756 (s); MS (EI
70 eV), m/z (%): 536.3 (K M+, 1), 420.3 (K M+ 2 Ge 2 Pri, 100); Acc.
Mass. EI: calc. for K M+: C31H48N3Ge1: 536.3055, found 536.3059.
In summary, we have prepared novel germanium(I) dimers
which are stabilised by bulky amidinate and guanidinate ligands.
Theoretical studies on a model complex suggest their Ge–Ge
bonds show no multiple bond character but their LUMOs are
p-bonding in nature, as is the case for distannynes. We are
currently exploring the further chemistry of germanium(I) dimers
and comparing it with that of ‘‘diradicaloid’’ digermynes.
We gratefully acknowledge financial support from the
Leverhume Trust (postdoctoral fellowship for AS), the EPSRC
(partial studentship for SPG) and the ERASMUS scheme of the
European Union (travel grant for K.-A.L). Thanks also go to the
EPSRC Mass Spectrometry Service.
1 See for example: (a) P. P. Power, Appl. Organomet. Chem., 2005, 19,
488; (b) P. P. Power, Chem. Commun., 2003, 2091; (c) M. Weidenbruch,
Angew. Chem., Int. Ed., 2003, 42, 2222; (d) P. Jutzi, Angew. Chem., Int.
Ed., 2000, 39, 3797 and references therein.
2 A. Sekiguchi, R. Kinjo and M. Ichinohe, Science, 2004, 305, 1755.
3 M. Stender, A. D. Phillips, R. J. Wright and P. P. Power, Angew.
Chem., Int. Ed., 2002, 41, 1785.
4 Y. Sugijama, T. Sasamori, Y. Hosoi, Y. Furukawa, N. Takagi,
S. Nagase and N. Tokitoh, J. Am. Chem. Soc., 2006, 128, 1023.
5 A. D. Phillips, R. J. Wright, M. M. Olmstead and P. P. Power, J. Am.
Chem. Soc., 2002, 124, 5930.
6 L. Pu, B. Twamley and P. P. Power, J. Am. Chem. Soc., 2000, 122, 3524.
7 e.g., A. Sekiguchi, T. Yatabe, H. Kamatani, C. Kabuto and H. Sakurai,
J. Am. Chem. Soc., 1992, 114, 6260.
8 See for example: (a) Y. Jung, M. Brynda, P. P. Power and M. Head-
Gordon, J. Am. Chem. Soc., 2006, 128, 7185; (b) N. Takagi and
S. Nagase, Organometallics, 2001, 20, 5498; (c) Y. Chen, M. Hartmann,
M. Diedenhofen and G. Frenking, Angew. Chem., Int. Ed., 2001, 40,
2052.
Notes and references
{ Crystal data for 1: C29H43ClGeN2, M = 527.69, orthorhombic, space
˚
group Pna21, a = 18.515(4), b = 8.3709(17), c = 18.448(4) A, V =
3
2859.2(10) A , Z = 4, Dc = 1.226 g cm23, F(000) = 1120, m(Mo-Ka) =
9 C. Cui, M. M. Olmstead, J. C. Fettinger, G. H. Spikes and P. P. Power,
J. Am. Chem. Soc., 2005, 127, 17530.
˚
1.183 mm21, 150(2) K, 4919 unique reflections [Rint 0.0845], R (on F)
0.0548, wR (on F2) 0.1133 (I . 2s(I)); 2: C31H48ClGeN3, M = 570.76,
10 L. Pu, A. D. Phillips, A. F. Richards, M. Stender, R. S. Simons,
M. M. Olmstead and P. P. Power, J. Am. Chem. Soc., 2003, 125, 11626.
11 (a) C. Jones, P. C. Junk, J. A. Platts and A. Stasch, J. Am. Chem. Soc.,
2006, 128, 2206; (b) C. Jones, P. C. Junk, J. A. Platts, D. Rathmann and
A. Stasch, Dalton Trans., 2005, 2497.
12 Compound 2 has been used as a reagent in a recent report, though no
structural or spectroscopic data were given: S. P. Green, C. Jones,
K.-A. Lippert, D. P. Mills and A. Stasch, Inorg. Chem., 2006, 45, 7242.
13 Y. Ding, H. W. Roesky, M. Noltemeyer and H.-G. Schmidt,
Organometallics, 2001, 20, 1190.
˚
monoclinic, space group Cc, a = 17.620(4), b = 12.441(3), c = 14.419(3) A,
b = 100.69(3)u, V = 3106.1(11) A , Z = 4, Dc = 1.221 g cm23, F(000) =
3
˚
1216, m(Mo-Ka) = 1.095 mm21, 150(2) K, 6076 unique reflections [Rint
0.0401], R (on F) 0.0643, wR (on F2) 0.1349 (I . 2s(I)); 3?(THF):
C62H94Ge2N4O, M = 1056.59, monoclinic, space group C2/c, a = 19.051(4),
3
˚
˚
b = 16.770(3), c = 19.911(4) A, b = 95.77(3)u, V = 6329(2) A , Z = 4, Dc =
1.109 g cm23, F(000) = 2264, m(Mo-Ka) = 0.989 mm21, 150(2) K, 6871
unique reflections [Rint 0.0488], R (on F) 0.0504, wR (on F2) 0.1112 (I .
2s(I)); 4: C62H96Ge2N6, M = 1070.63, monoclinic, space group P21/n, a =
˚
3
˚
11.818(2), b = 14.892(3), c = 17.589(4) A, b = 103.72(3)u, V = 3007.3(10) A ,
Z = 2,
14 As determined by a survey of the Cambridge Crystallographic
Database.
3980 | Chem. Commun., 2006, 3978–3980
This journal is ß The Royal Society of Chemistry 2006