Synthesis and X-ray crystal structure of bis(germyl)carbodiimides. Spectroscopic characterization of (hydrogeno)(germyl)cyanamides

Article abstract of DOI:10.1016/S0022-328X(98)00546-4

Bis(triethylgermyl)- and bis(trimesitylgermyl)-carbodiimides are easily formed either by transmetallation or transamination reactions. (Hydrogeno)triethyl- or trimesitylgermyl-cyanamides are characterized in the dehydrohalogenation reaction between cyanam

Full text of DOI:10.1016/S0022-328X(98)00546-4

Journal of Organometallic Chemistry 562 (1998) 191–195
Synthesis and X-ray crystal structure of bis(germyl)carbodiimides.
Spectroscopic characterization of (hydrogeno)(germyl)cyanamides
Mohamed Dahrouch a, Monique Rivie`re-Baudet a,*, Heinz Gornitzka b, Guy Bertrand b
a Laboratoire d’He´te´rochimie Fondamentale et Applique´e, UPRES-A 5069 du CNRS, Uni6ersite´ Paul Sabatier, 31062 Toulouse-Ce´dex, France
b Laboratoire de Chimie de Coordination du CNRS, UPR 8241, 205, route de Narbonne, 31077 Toulouse-Ce´dex, France
Received 17 December 1997
Abstract
Bis(triethylgermyl)- and bis(trimesitylgermyl)-carbodiimides are easily formed either by transmetallation or transamination
reactions. (Hydrogeno)triethyl- or trimesitylgermyl-cyanamides are characterized in the dehydrohalogenation reaction between
cyanamide and triethyl- or trimesityl-chlorogermanes. According to
a single X-ray diffraction analysis, bis(trime-
sitylgermyl)carbodiimide is monomeric and the GeNCNGe framework is not linear, in contrast to tin and silicon analogues,
respectively. © 1998 Elsevier Science S.A. All rights reserved.
Keywords: Carbodiimide; Cyanamide; Germanium
1. Introduction
2. Results and discussion
In the last few years, the chemistry of Group 14
substituted cyanamides and carbodiimides has been
significantly developed, mainly because of their poten-
It has been reported that halogermanes react with
lead
or
silver
cyanamides
affording
bis(-
germyl)carbodiimides [1,17]; not surprisingly addition
of triethylchlorogermane to the bis(lithium) salt of
cyanamide affords the carbodiimide 1 (Scheme 1).
It is also known that N-triethylgermyl–diethylamine
reacts with dicyanamide at high temperature (110–200
°C) yielding carbodiimide 1 [1]. Using cyanamide, in-
tial
applications
[1–10].
Although
R3Sn–NH–CN
(hy-
are
drogeno)(stannyl)cyanamides
known [1,11–14], silyl [1,15] and germyl analogues [16]
have never been obtained, probably because of their
lack of stability. Our interest in the preparation of new
germylated polymers has led us to investigate the possi-
bility of preparing stable R3Ge–NH–CN derivatives,
and also to find the easiest way to synthesize bis-
(germyl)carbodiimides. Here we report our results in
the triethylgermyl and trimesitylgermyl series.
Scheme 1.
* Corresponding author. Tel.: +33 05 61558348; fax: +33 05
61558234; e-mail: riviere@ramses.ups-tlse.fr
Scheme 2.
0022-328X/98/$19.00 © 1998 Elsevier Science S.A. All rights reserved.
PII S0022-328X(98)00546-4

192
M. Dahrouch et al. / Journal of Organometallic Chemistry 562 (1998) 191–195
Scheme 3.
stead of its dimer, the transamination reaction occurs at
room temperature (r.t.), quantitatively affording 1. In
the hope of preparing the desired (hydrogeno)
(germyl)cyanamide 2, the sterically hindered N-tri-
ethylgermyl–diisopropylamine was used (Scheme 2).
However, once again, whatever the ratio of the reagents
used, carbodiimide 1 was obtained; in the case of the
1/1 ratio, half of the starting cyanamide was recovered
unchanged. Note that at the beginning of the sixties
[18,19], there was controversy about the structure of
bis-metallated NCN compounds of group 14
[cyanamide (R3Ge)2N–CN versus carbodiimide R3Ge–
NCN–GeR3]. It is now well established that, in solu-
tion, all of them have the carbodiimide structure
[1,4,6,7]; the bulky metal groups cannot be accommo-
dated by the same nitrogen atom [7].
Interestingly, when the reaction of triethylchloroger-
mane with cyanamide, in the presence of triethylamine,
was monitored by spectroscopy, the (hydrogeno)
(triethylgermyl)cyanamide 2 was detected (Scheme 3).
The structure of 2 was clearly established by GC/mass
spectrometry [203: (M+1)+], infrared [3381 (NH),
2193 (CN) cm−1] and 13C-NMR spectroscopy (117
ppm: CN). Without further treatment, the reaction
mixture eventually leads to 1.
We obtained evidence that the formation of 2 is
limited by the reverse cleavage reaction of the Ge–N
bond by the triethylammonium chloride, acting as a
protic species. Indeed, the yield of 2 increased when an
excess of cyanamide was used (ratio 1:6), or by remov-
ing the ammonium salt from the reaction mixture by
precipitation and filtration. We also observed that the
excess of cyanamide in THF solution prevents the
transformation of 2 into 1, so does nitromethane and,
slightly less efficiently, acetonitrile (cf. Section 4, Tables
3 and 4). Two competitive mechanisms (Scheme 4) can
rationalize the formation of 1 from 2: (i) a dispropor-
tionation of 2 giving back cyanamide along with 1; (ii)
the reaction of 2 with the triethylchlorogermane which
is present in the reaction mixture.
Scheme 5.
Since neither nitromethane nor acetonitrile prevented
the transamination reaction between triethylgermy-
lamine and cyanamide (Scheme 2), in which 2 is almost
certainly the intermediate, it is quite likely that these
solvents strongly complex triethylchlorogermane, thus
preventing any nucleophilic attack. A further support
for this hypothesis was found in the fact that no
reaction occurred when reacting triethylchlorogermane
with cyanamide and triethylamine in nitromethane.
In
the
hope
of
isolating
a
(hy-
drogeno)(germyl)cyanamide, trimesitylchlorogermane
was used as the electrophile. However, a mixture of
germylcyanamide 3 and bis(germyl)carbodiimide 4 was
obtained. Attempts to isolate 3 by recrystallization
from a nitromethane solution, which prevents the for-
mation of 4, failed. Of course, the bis(trime-
sitylgermyl)carbodiimide 4 was the only final product
when trimesitylchlorogermane was added to the bis(-
lithium) salt of cyanamide (Scheme 5).
Since the solid state structure of bis(trimethylstan-
nyl)carbodiimide
[20]
and
bis(triphenylsi-
lyl)carbodiimide [21] are known and are quite different,
it was of interest to carry out a single X-ray diffraction
study of the bis(germyl)carbodiimide 4. In the solid
state, the structure of the bis(trimethylstan-
nyl)carbodiimide consists of an infinite helical network
of planar trimethyltin groups linked by linear NCN
units with SnNC angles of 117.6°; all nitrogen atoms
are coordinated to two tin units with crystallographi-
cally identical tin–nitrogen bonds (2.47 A). In other
words, tin atoms are midway between two NCN frag-
ments, which can be seen as carbodiimides or alterna-
tively as cyanamides (NC: 1.24 A). In contrast, the
bis(triphenylsilyl)carbodiimide is monomeric with a
˚
˚
˚
perfectly linear SiNCNSi unit and short SiN (1.70 A)
and CN (1.16 A) bond lengths. These results are consis-
˚
tent with delocalization of nitrogen lone pairs into the
SiC |*-orbitals, and with the interaction of the NC
antibonding y-orbitals with the SiC |-orbitals. The
results of the X-ray diffraction analysis of 4 are shown
in Fig. 1, Tables 1 and 2. Compound 4 is monomeric,
the NCN fragment is nearly linear [176.0(3)°], but in
contrast with the silicon case, the germanium atoms are
out of this chain [GeNC: 154.9(2) and 143.3(2)°]. As
discussed in the gas phase study of germyl pseudo-
halides [22], the GeNC angles do not provide evidence
for the back donation from the NCN fragment to the
Scheme 4.

M. Dahrouch et al. / Journal of Organometallic Chemistry 562 (1998) 191–195
193
Table 1
Crystallographic data for 4
germanium atoms, but the GeN [1.832(2) and 1.836(2)
˚
˚
A] and the NC bond lengths [1.199(3) and 1.206(3)A]
do. Lastly, probably because of the bulky mesityl
groups, there is a GeNNGe dihedral angle of 126.5(3)°.
The influence of solvents on the stability and forma-
tion of 2 is shown in Tables 3 and 4.
Formula
Formula mass
Color
Crystal size (mm)
Temperature (K)
F(000)
C55H66Ge2N2
900.28
Colorless
0.8×0.7×0.4
193(2)
948
Crystal system
Space group
Triclinic
3. Conclusion
(
P1
˚
a [A]
13.294(3)
13.509(3)
15.356(3)
101.39(3)
110.15(3)
100.78(3)
2438.6(9)
2
˚
b [A]
To our knowledge [16], (triethylgermyl)- and (trime-
sitylgermyl)-cyanamide 2 and 3, although not isolated
as pure samples, are the first examples of hydrogeno-
cyanamide ever characterized in the germanium series.
Carbodiimide 4 is monomeric in the solid state, in
contrast to tin analogues, and the GeNCNGe frame-
work is not linear, in contrast to the silicon analogues.
The use of these air-stable germylcarbodiimides as pre-
cursors for inorganic polymers is under active
investigation.
˚
c [A]
h [°]
i [°]
k [°]
V [A ]
3
˚
Z
zcalcd [mg cm3]
1.226
1.269
0.500; −0.405
v [mm−1
]
Max./min. peaks in final difference map
−3
˚
(e A
)
R1 (I\2|(I))
0.0321
0.0853
1.002
wR2 (all data)
Goodness-of-fit on F2
4. Experimental section
All reactions were performed in dry solvents, under
dry nitrogen using standard Schlenk techniques. The
compounds where characterized by usual analytical
techniques: l ppm, 1H-NMR, AC 80 Bruker; 13C-NMR
AC 200; IR: Perkin Elmer 1600 FT; mass spectra,
Ribermag R1010 (DCI, CH4) and HP 5989 (EI). Gas
chromatography: Hewlett Packard 6890 GC (column
HP1, methylsilicon).
Table 2
˚
Selected bond length (A) and angles (°) for 4.
Ge(1)–N(1)
Ge(1)–C(20)
Ge(2)–N(2)
Ge(2)–C(47)
N(1)–C(1)
C(2)–C(7)
1.832(2) Ge(1)–C(2)
1.975(2) Ge(1)–C(11)
1.836(2) Ge(2)–C(38)
1.976(3) Ge(2)–C(29)
1.199(3) C(1)–N(2)
1.407(4) C(2)–C(3)
1.9 63(3)
1.978(3)
1.969(2)
1.977(3)
1.206(3)
1.418(4)
N(1)–Ge(1)–C(2)
C(2)–Ge(1)–C(20)
C(2)–Ge(1)–C(11)
N(2)–Ge(2)–C(38)
C(38)–Ge(2)–C(47) 114.1(1)
C(38)–Ge(2)–C(29) 115.6(1)
105.7(1)
N(1)–Ge(1)–C(20) 100.5(1)
115.0(1)
113.3(1)
100.8(1)
N(1)–Ge(1)–C(11) 105.9(1)
C(20)–Ge(1)–C(11) 114.7(1)
N(2)–Ge(2)–C(47) 108.5(1)
N(2)–Ge(2)–C(29) 103.7(1)
C(47)–Ge(2)–C(29) 112.6(1)
C(1)–N(1)–Ge(1)
C(1)–N(2)–Ge(2)
C(7)–C(2)–Ge(1)
154.9(2)
143.2(2)
120.6(2)
N(1)–C(1)–N(2)
C(7)–C(2)–C(3)
C(3)–C(2)–Ge(1)
176.0(3)
118.3(2)
121.1(2)
C(16)–C(11)–Ge(1) 121.9(2)
C(25)–C(20)–Ge(1) 122.9(2)
C(30)–C(29)–Ge(2) 122.7(2)
C(43)–C(38)–Ge(2) 123.1(2)
C(48)–C(47)–Ge(2) 120.3(2)
C(12)–C(11)–Ge(1) 118.6(2)
C(21)–C(20)–Ge(1) 118.4(2)
C(34)–C(29)–Ge(2) 118.6(2)
C(39)–C(38)–Ge(2) 118.6(2)
C(52)–C(47)–Ge(2) 121.0(2)
Fig. 1. Solid state structure of 4 (anisotropic displacement parameters
are depicted at the 50% probability level) and view along the
GeꢀNꢁCꢁNꢀGe axis.

194
M. Dahrouch et al. / Journal of Organometallic Chemistry 562 (1998) 191–195
Table 3
4.2. Preparation and characterization of (2)
Influence of nucleophilic solvents on the stability of 2 in the reaction
mixture
To a THF solution (4 ml) of cyanamide (0.30 g, 7.14
mmol) was added Et3GeCl (0.23 g, 1.19 mmol). Et3N
(0.120 g, 1.19 mmol) was added to the mixture under
stirring at r.t. A precipitation of Et3N·HCl was ob-
served. After 15 min, GC analysis (Et4Ge as internal
standard) showed Et3GeNHCN 2 (33%)1, Et3GeCl
(67%), 1 (0%). After filtration of the triethylammonium
chloride and evaporation of the solvent under 30
mm Hg, immediate spectroscopic analysis allowed the
characterization of 2: GC/Mass: (M+1)+: 203 (8%),
(M+1)+-Et: 174 (100%); (M+1)+-3Et: 116 (50%); IR
Solvent
Time
Et3GeCl%
2%
1%
Starting time
20’
18 h or 3 days
20’
3 days
3 days
67
33
49
67
67
35
33
25
0
33
33
30
0
CDCl3
42
51
0
0
35
CH3NO2
CH3CN
(CDCl3): 3381 (NH); 2193 (CꢂN) cm−1 1H-NMR
;
(CDCl3, 25°C): 0.88 (m, 15H, C2H5), 4.1 (s, 1H, NH);
13C-NMR (CDCl3, 25°C): 117.00 (N–CꢂN), 7.87
(CH3), 7.39 (CH2).
4.1. Bis(triethylgermyl)carbodiimide (1)
(a) By transmetallation from lithium cyanamide: To
a solution of H2NCN (1.03 g, 24.5 mmol) in 16 ml of
THF was added at −65°C under stirring 49 mmol of
t-BuLi (28.82 ml, 1.7 M in pentane). The reaction
mixture was warmed to r.t. after ca. 30 min. Then
Et3GeCl (9.56 g, 49 mmol) was added under stirring.
After LiCl filtration, spectroscopic analyses indicated
the quantitative formation of 1. A distillation (Eb:
162°C/17 mm Hg; in conformity with Refs. 1 and 17)
afforded 5.56 g of 1 (63% yield); Mass spectrometry
(EI): M+: 360 (1%), M+-Et: 331 (100%), M+-2Et: 302
4.3. Preparation and characterization of
trimesitylgermylcyanamide (3)
Trimesitylgermylcyanamide 3 was prepared in a simi-
lar way as 2, using cyanamide (0.15 g, 3.57 mmol),
Mes3GeCl (0.27 g, 0.59 mmol) and Et3N (0.06 g, 0.59
mmol). After filtration of Et3N·HCl and evaporation
of the solvent under vacuo, the white residue was
analyzed by 1H NMR spectroscopy: Mes3GeCl (31%), 3
(52%), 4 (17%). Spectroscopic data for 3: Mass spec-
trum (DCI, CH4): (M-1)+: 471 (6%), (M-1)+-Mes: 352
(15%), M+-3Et: 273 (10%); IR (CDCl3): 2108 cm−1
:
(was NꢁCꢁN); 1H-NMR (CDCl3) l C2H5: 0.85–0.91
(m); 13C-NMR (CDCl3, 25°C): l 134.44 (NCN), 7.88
(CH3), 7.46 (CH2).
(16%); IR(CDCl3): 3427 (NH), 2258 (N–CꢂN) cm−1
;
1H-NMR (C6D6, 25°C): 2.07 (s, 9H, p-CH3), 2.30 (s,
18H, o-CH3), 6.70 (s, 6H, C6H2); 13C-NMR (CDCl3,
25°C): 21.04 (p-CH3), 23.91 (o-CH3), 134.70 (C1),
143.25 (C2), 129.86 (C3), 139.68(C4), 114.48 (N–CꢂN).
Derivative 3 in THF, benzene, or CDCl3, was slowly
converted at r.t. into 4, and rapidly when heated for 30
(b) By transamination reaction from tri-
ethylgermyldimethylamine: To a solution of H2NCN
(0.05 g, 1.19 mmol) in 3 ml of THF (or CH3NO2) at
20°C, was added under stirring Et3GeNMe2 (0.24 g,
1.19 mmol). The reaction was exothermic with evolu-
1
tion of dimethylamine. According to GC, and H- and
1 By addition of a chloroform solution of Et3N·HCl, the percent-
age of 2 decreased while that of Et3GeCl increased. In contrast,
cooling the solution to −35°C, favored the precipitation of
Et3N·HCl, and the percentage of 2 increased; 2: 42%, Et3GeCl: 58%,
1: 0% (percentages, measured by GC with Et4Ge as internal stan-
dard).
13C-NMR spectroscopy, the reaction was quantitative.
(c) By transamination from triethylgermyldiisopropy-
lamine: in a similar way H2NCN (0.05 g, 1.19 mmol)
and Et3GeNiPr2 (0.21 g, 0.80 mmol) quantitatively led
to 1 with elimination of diisopropylamine.
Table 4
Influence of the Et3GeCl/H2NCN ratio on the formation of 2
Starting relative proportions Et3GeCl/H2NCN
Et3N
CPV analysis (relative percent-
ages)
(solvent THF)
% 2
% 1
1/6
1/4
1/1
4/1
6/1
1/1
3
3
3
3
100
34
14
2
0
100
0
66
86
98
3
(1) 3Et3N (2) after 15 min, +5CH3NO2
100

M. Dahrouch et al. / Journal of Organometallic Chemistry 562 (1998) 191–195
195
References
min at 60°C. In a way similar to that observed for 2,
addition of CH3NO2 prevented the transformation of 3
into 4. All attempts to recrystallize 3 in nitromethane
failed. In C6H6 or ether, after filtration of excess
cyanamide, crystals of 4 were obtained.
[1] A.S. Gordetsov, V.P. Kozyukov, I.A. Vostokov, S.V. Shellud-
dyakova, Yu. I. Dergunov, V.F. Mironov, Usp. Khim. 51 (1982)
848; Russ. Chem. Rev. 51 (1982) 485.
[2] A. Obermeyer, A. Kienzle, J. Weindlein, R. Riedel, A. Simon, Z.
Anorg. Allg. Chem. 620 (1994) 1357.
[3] A. Kienzle, A. Obermeyer, R. Riedel, F. Aldinger, A. Simon,
Chem. Ber. 126 (1993) 2569.
[4] P. Haag, R. Lechler, J. Weidlein, Z. Anorg. Allg. Chem. 620
(1994) 112.
4.4. Synthesis of bis(trimesitylgermyl)carbodiimide
(4) by transmetallation from lithium cyanamide
[5] T. Pasinszki, T. Veszpremi, M. Feher, J. Mol. Struct.
(THEOCHEM) 331 (1995) 289.
[6] G. Veneziani, R. Reau, G. Bertrand, Organometallics 12 (1993)
4289.
[7] R. Reau, G. Bertrand, Rev. Heteroatom. Chem. 14 (1996) 134.
[8] H. Koehler, R. Menzel, F. Mandl, L. Jaeger, Z. Anorg. Allg.
Chem. 551 (1987) 173.
[9] A.S. Gordetsov, A.P. Kosina, S.E. Skobeleva, A.N. Egorochkin,
Metallorg. Khim. 1 (1988) 107.
[10] G.A. Razuvaev, A.S. Gordetsov, A.P. Kozina, et al., J.
Organometal. Chem. 327 (1987) 303.
[11] V.F. Gerega, Yu. I. Dergunov, A.V. Pavlycheva, Yu. I.
Mushkin, Yu. A. Aleksandrov, Z. Obsch. Khim. 40 (1970) 1099.
[12] R.A. Cardona, E.J. Kupchik, J. Organometal. Chem. 34 (1972)
129; 43 (1972) 163.
[13] J.A. Feiccabrino, E.J. Kupchik, J. Organometal. Chem. 56
(1973) 167.
[14] G. Domazetis, R.J. Magee, B.D. James, J. Organometal. Chem.
148 (1978) 339.
[15] L. Birkofer, A. Riter, P. Richter, Tetrahedron Lett. (1962) 195.
[16] P. Rivie`re, M. Rivie`re-Baudet, J. Satge´, Comprehensive
Organometallic Chemistry, vol. 2, Pergamon Press, Oxford,
1982, p. 399 and 1995, p. 137.
[17] J.E. Drake, R.T. Hemmings, H.E. Henderson, J. Chem. Soc.
Dalton Trans. (1976) 366.
[18] E.A. Ebsworth, M.J. Mays, J. Chem. Soc. (1961) 4879; Spec-
trochimica Acta (1963) 1127.
[19] S. Cradock, E.A. Ebsworth, J. Chem.Soc. A (1968) 1423.
[20] R.A. Forder, G.M. Sheldrick, J. Chem. Soc. Chem. Commun.
(1970) 1023; J. Chem. Soc. A (1971) 1107.
[21] G.M. Sheldrick, R. Taylor, J. Organomet. Chem. 101 (1975) 19.
[22] J.D. Murdoch, D.W.H. Rankin, J. Chem. Soc. Chem. Commun.
(1972) 748.
[23] G.M. Sheldrick, Acta Crystallogr. Sect A 46 (1990) 467.
[24] G.M. Sheldrick, SHELXL-93, Program for Crystal Structure
Refinement, Universita¨t Go¨ttingen, 1997.
Lithium cyanamide prepared as for 1 [H2NCN (0.01
g, 0.32 mmol), t-BuLi: (0.64 mmol)], was added into a
THF solution (3 ml) of Mes3GeCl (0.30 g, (0.64
mmol)]. The reaction mixture was heated at 55°C for 2
h. THF was replaced by benzene and LiCl eliminated
by centrifugation. Evaporation of the solvents under
vacuo led to 0.25 g (75% yield) of a white residue
identified as 4; recrystallization from ether afforded
colorless crystals; m.p.: 237–238°C; Mass spectrum
(DCI, CH4) (M+1)+: 901 (30%), (M+1)+-CH4: 885
(12%), (M+1)+-MesH: 781 (18%); IR (CDCl3): 2158;
(nujol mull): 2179 (was N=C=N) cm−1 1H-NMR
.
(CDCl3, 25°C): 2.08 (s, 36H, o-CH3), 2.23 (s, 18H,
p-CH3), 6.70 (s, 12H, C6H2); 13C-NMR (CDCl3, 25°C):
21.02 (p-CH3), 23.74 (o-CH3), 137.23 (C1), 143.47 (C2),
129.27 (C3), 138.33 (C4), 136.44 (N=C=N).
4.5. Crystal and X-ray experimental data for (4)
The crystal data for 4 are presented in Table 1. The
data were collected on a STOE-IPDS diffractometer
˚
with Mo–Kh (u=0.71073 A) radiation using €-scans.
The structure was solved by direct methods using
SHELXS-97 [23] and refined with all data on F2
with a weighting scheme of ꢀ−1=|2(Fo2)+(g1·P)2+
(g2·P) with P=(Fo2 +2F2c)/3 using SHELXL-97.
[24] All non-hydrogen atoms were treated anisotropi-
cally.
.
.