A zinc - Aldimine complex

Article abstract of DOI:10.1016/S0020-1693(99)00535-6

Dimethylaminobenzaldimine - zinc-bis(pentafluorothiophenolate) was obtained from the reaction of dimethylaminobenzaldehyde with ammonia-contaminated zinc-bis(pentafluorothiophenolate) and characterized by a structure determination. Zinc effects the cataly

Full text of DOI:10.1016/S0020-1693(99)00535-6

www.elsevier.nl/locate/ica
Inorganica Chimica Acta 300–302 (2000) 181–185
A zinc–aldimine complex
B. Mu¨ller, H. Vahrenkamp *
Institut fu¨r Anorganische und Analytische Chemie der Uni6ersita¨t Freiburg, Albertstr. 21, D-79104 Freiburg, Germany
Received 23 August 1999; accepted 29 October 1999
Dedicated to Professor Dirk Walther on the occasion of his 60th birthday.
Abstract
Dimethylaminobenzaldimine–zinc-bis(pentafluorothiophenolate) was obtained from the reaction of dimethylaminobenzalde-
hyde with ammonia-contaminated zinc-bis(pentafluorothiophenolate) and characterized by a structure determination. Zinc effects
the catalytic formation and stabilization of the aldimine by complexation. © 2000 Elsevier Science S.A. All rights reserved.
Keywords: Zinc complexes; Aldimine complexes; Crystal structures
1. Introduction
We came across this class of complexes during our
studies of zinc aldehyde complexes in the context of the
modelling of the zinc enzyme alcohol dehydrogenase [13].
We had modelled the coordination environment of the
zinc ion in this enzyme by the 2-dimethylaminobenzalde-
hyde-zinc-thiolate complex 1. Complex 2 is the aldimine
analogue of 1. This paper describes the synthesis and
molecular structure of 2. 2 is the first fully characterized
zinc complex of an aldimine, and it has for the first time
been possible to compare the structures of an aldehyde
and the corresponding aldimine complexes.
Aldehyde imines, the simplest of all Schiff bases, are
not stable at room temperature, undergoing condensa-
tion and polymerization reactions [1]. Blocking the
unshared electron pair on nitrogen should lower their
lability, typically by coordination to a transition metal.
In accord with this, it was observed many years ago by
Pfeiffer, in the developing phase of the salen ligands, that
chelating aldehyde ligands react with ammonia to form
aldimine complexes. The most typical of these are derived
from substituted salicylic aldehydes [2], but the reaction
was also observed for coordinated pyrrolyl aldehyde [3].
Subsequently, nickel and copper complexes of 3-alkoxy-
salicyl-aldimines were structurally characterized and
used for the formation of dinuclear complexes [4–8], and
one copper complex of pyrrolyl-2-aldimine [9] was sub-
jected to crystal structure determinations. In the field of
zinc chemistry the electronic spectra of a few salicyl-
aldimine complexes have been reported [10,11], and
copper and nickel salicylaldimine complexes were used
as ‘ligands’ for zinc salts [12]. These few reports justify
the statement that very little use has been made of the
complex-stabilized aldimines, neither in organic chem-
istry nor in coordination chemistry.
2. Results and discussion
2.1. Preparation
The aldimine complex 2 was found accidentally when
preparing complex 1 from zinc-bis(pentafluorothiophe-
nolate) and dimethylaminobenzaldehyde [13]. Since wa-
ter is a better ligand for zinc than aldehydes, completely
* Corresponding author. Tel.: +49-761-203 6120; fax: +49-761-
203 6001.
E-mail address: vahrenka@uni-freiburg.de (H. Vahrenkamp)
0020-1693/00/$ - see front matter © 2000 Elsevier Science S.A. All rights reserved.
PII: S 0 0 2 0 - 1 6 9 3 ( 9 9 ) 0 0 5 3 5 - 6

182
B. Mu¨ller, H. Vahrenkamp / Inorganica Chimica Acta 300–302 (2000) 181–185
Fig. 1. Molecular structures of 2. (a): ring-parallel conformers in 2a, 2b and 2c(1). (b): ring-separated conformer in 2c(2).
water-free conditions had to be ensured. This was done
for the zinc thiolate by preparing it from zinc-bis-
(bistrimethylsilylamide) and pentafluorothiophenol [14],
overlooking originally that the product obtained is
contaminated with ammonia and silylamide components
[15] resulting from partial thiolysis of Zn[N(SiMe3)2]2.
Consequently the combination of the aldehyde and the
zinc thiolate produced a mixture of 1 and 2.
a result three different polymorphs of 2 (2a–c) were found
and subjected to structure determinations. In addition the
orthorhombic crystals of 2c contain two formula units
per asymmetric unit, bringing the number of independent
molecular structures to four. Thus the structure determi-
nation of 2 offered a unique opportunity to compare
crystallographic and ‘real’ standard deviations of bond
lengths and angles.
After the successful isolation of pure 2 it could be
verified that careful treatment of pure 1 [13] with
water-free ammonia yields 2, i.e. the conversion of the
aldehyde to the imine can take place in the ligand sphere
of zinc. This, however, is not a favourable process for the
preparation of 2 due to the efforts involved in obtaining
pure 1 and separating the product mixture resulting from
1 and NH3. Instead we found an easy way to separate
1 and 2 by using tetrahydrofuran as a solvent. The latter
is a better ligand for zinc than dimethylaminobenzalde-
hyde and allowed for the crystallization of uncontami-
nated 2 from THF/CH2Cl2 solutions.
The molecular shapes of 2 in crystals of 2a, 2b and one
independent unit of 2c are practically identical, as
displayed in Fig. 1(a). The other independent unit in 2c
is another conformer, differing by a rotation about the
ZnꢁS axes, as shown in Fig. 1(b). While in the latter the
two C6F5 rings of the complex do not interact with each
other (but with phenyl rings of neighbouring molecules),
in the three other conformers they are parallel, showing
a stacking interaction with ring-to-ring distances of
,
approximately 3.3 A.
The bond lengths in the four independent mole-
cular structures (Table 1) show a remarkable uniformity,
with only those for 2c, having the crystals of lowest
quality, differing more than three standard devia-
tions from the average values. This demonstrates that
packing forces have little influence on the bond distances
for this complex. In contrast, there is a larger spread
of the ‘softer’ bond angles, specifically of the NꢁZnꢁS
angles and the ZnꢁNꢁC angles. Yet even here the total
variation of the values is small. This applies also to the
two different conformers in 2c. In particular the ZnꢁS,
SꢁC, and CꢀN bonds as well as the NꢁZnꢁN and
NHꢁCꢁC angles are practically constant throughout the
series.
Complex 2 was characterized by its IR absorptions (in
KBr) at 3208 cm−1 for the NH stretch and at 1627 cm−1
1
for the CꢀN unit, both of medium intensity. In the H
NMR spectrum (for details see Section 3) the aldimine
function shows up in the form of two characteristic
doublets at 11.19 and 8.87 ppm for the NH and CH units,
respectively. In the 13C NMR spectrum the aldimine
carbon resonance is observed at 176.7 ppm. A heteronu-
clear COSY experiment with indirect 15N detection confi-
rmed the presence of the imine nitrogen.
2.2. Structure
Typical molecular features are the large SꢁZnꢁS and
small NꢁZnꢁN angles, as observed in many other com-
Our original doubts about the identities of 1 and 2 led
to many attempts at separation and crystallization. As

B. Mu¨ller, H. Vahrenkamp / Inorganica Chimica Acta 300–302 (2000) 181–185
183
Table 1
Bond lengths (A) and angles (°) in the crystals of 2
weak ligands like THF or acetone replace the aldehyde
ligand, 2 can be transferred as a solid into the open
atmosphere and was isolated from THF solution. Ex-
posure to water in solution, however, also leads to the
decomposition of 2.
,
2a
2b
2c (1)
2c (2)
ZnꢁN(NH)
ZnꢁN(NMe2)
ZnꢁS1
2.023(1) 2.007(4)
2.149(2) 2.142(4)
2.286(1) 2.302(1)
2.281(1) 2.283(1)
1.275(2) 1.273(6)
1.452(2) 1.454(5)
1.762(1) 1.757(4)
1.756(1) 1.754(4)
2.073(10)
2.118(10)
2.283(3)
2.021(9)
2.166(9)
2.283(3)
2.275(3)
1.273(14)
1.487(13)
1.778(11)
1.751(12)
The molecular shapes of 1 and 2 (the three ring-par-
allel conformers, Fig. 1(a) are practically superimpos-
able. The orientation of the SC6F5 groups and the
folding of the six-membered chelate rings are nearly
identical. The only noteworthy differences lie in the
ZnꢁN and ZnꢁO bond distances. The ZnꢁN(NH) bond
ZnꢁS2
2.280(3)
CꢀN(NH)
CꢁN(NMe2)
S1ꢁC
1.254(15)
1.498(15)
1.763(12)
1.750(12)
S2ꢁC
,
in 2 is about 0.08 A shorter than the ZnꢁO(aldehyde)
S1ꢁZnꢁS2
NꢁZnꢁN
120.9(1) 115.8(1)
125.1(1)
88.8(4)
122.9(1)
88.4(4)
87.3(1)
89.8(1)
bond in 1, in contrast to the increase in atomic radii
from O to N. This bond strengthening is compensated
by a bond weakening for the ZnꢁN(NMe2) bond which
S1ꢁZnꢁN(NH)
S2ꢁZnꢁN(NH)
S1ꢁZnꢁN(NMe2)
S2ꢁZnꢁN(NMe2)
ZnꢁNHꢁC
ZnꢁNMe2ꢁC
NHꢁCꢁC
114.9(1) 121.5(1)
111.8(1) 117.0(1)
109.9(1) 107.6(1)
106.4(1) 102.7(1)
121.1(1) 124.2(3)
107.6(1) 111.7(3)
124.8(1) 125.8(4)
113.5(3)
113.0(3)
103.0(3)
105.3(2)
120.4(9)
104.6(7)
124.3(12)
112.4(2)
112.2(3)
108.8(3)
106.2(3)
123.6(8)
109.2(6)
125.6(11)
,
in 2 is about 0.045 A longer than that in 1. Taking the
large variability of ZnꢁN bond lengths (even when
maintaining tetrahedral coordination) the observed
changes from 1 to 2 are not dramatic. They are in
accord with the increased chemical stability of 2, but
cannot provide the sole explanation for it. We would
therefore invoke the ZnN2S2 coordination, which is
highly favored in the classical and biological coordina-
tion chemistry of zinc, as the main reason of stability.
Thereby the inherent lability of the unsubstituted imine
function is overcome, firstly by coordination to a metal
and secondly by the advantageous coordination mode.
plexes with ZnN2S2 coordination [16]1. The ZnꢁS bond
lengths are normal for tetrahedral zinc with terminal
thiolate [16]. The ZnꢁN(imine) and ZnꢁN(amine) bond
lengths differ significantly due to the different radii of
sp2 and sp3 nitrogen. The ZnꢁN(imine) distances are
quite short, even for tetrahedral zinc and also when
compared to ZnꢁN distances for heterocyclic nitrogen
ligands or Schiff-base complexes. This indicates a
strong coordination ability of the aldimine function. In
contrast the ZnꢁN(amine) bonds are rather long. This
corresponds to our experience that tertiary amines are
weak ligands for zinc and bind well only in a fa-
vourable chelating situation, an example with a similar
bonding situation being a complex with a chelating
pyrrolidine donor [17].
The chelating aldimine ligand offers no unusual
bonding features. Bond lengths and angles in the
CꢁCHꢀNH unit are comparable to those in the three
other structurally characterized aldimine complexes [4–
9]. Coordination geometries and chelating situations
cannot be compared as all three reported complexes are
square planar and derived from different aldimines.
3. Experimental
The general experimental methods and measuring
techniques were as in Ref. [13,18]. 2-Dimethyl-
aminobenzaldehyde [19] and Zn[N(SiMe3)2]2 [20] were
prepared as described. Zn(SC6F5)2 [21] was synthesized
by Bochmann’s method [14] from Zn[N(SiMe3)2]2 and
C6F5SH.
3.1. Preparation of 2
A total of 5.08 g (10.96 mmol) of Zn(SC6F5)2 were
dissolved in 25 ml of diethyl ether and a few drops of
THF. 1.64 g (10.99 mmol) of 2-dimethylaminoben-
zaldehyde in 5 ml of diethyl ether was added dropwise
with stirring. Upon slow addition of 200 ml of
petroleum ether (b.p. 60–70°C) an orange–yellow pre-
cipitate formed consisting of an approximately 1:1 mix-
ture of 1 and 2. After filtration the precipitate was
dissolved in a minimum amount of THF/CH2Cl2 and
the solution layered with petroleum ether. Within 1 day
yellow crystals of 2 were formed which were filtered off,
washed with petroleum ether and dried in vacuo. 2.35 g
2.3. Comparison of the aldehyde and aldimine
complexes
The aldimine complex 2 is more stable than the
aldehyde complex 1. This is not only obvious from its
spontaneous formation but also from its behaviour in
solution and in the solid state. While 1 is decomposed
by traces of moisture (e.g. the atmosphere), and even
1
(35%) of 2, m.p. 130°C (dec.), were obtained. H NMR
(acetone-d6): 3.02 [s, 6H, CH3(NMe2)], 7.42 [m, 1H,
1 Comparisons were made with structures deposited in the Cam-
bridge Crystallographic Data Centre.
H5], 7.63 [m, 1H, H3], 7.68 [m, 2H, H4, H6], 8.87 [d,
3
3J=14.7 Hz, 1H, CH(Imine)], 11.19 [d, J=14.7 Hz,

184
B. Mu¨ller, H. Vahrenkamp / Inorganica Chimica Acta 300–302 (2000) 181–185
Table 2
Crystallographic details
2a
2b
2c
Formula
C21H12F10N2S2Zn
611.82
THF/CH2Cl2
0.5×0.5×0.4
C
611.82
THF/hexane
1.0×0.8×0.7
P1
21H12F10N2S2Zn
C21H12F10N2S2Zn
611.82
THF/CH2Cl2/hexane
0.5×0.1×0.07
Pbca
16
Molecular weight
Crystallization from
Crystal size (mm)
Space group
Z
(
(
P1
2
2
,
a (A)
8.770(3)
7.492(3)
14.238(2)
12.650(2)
49.857(9)
90
90
90
8980(3)
1.81
1.38
2.5–24.5
−165h50
05k514
−585l50
7590
7587
4051
,
b (A)
12.043(3)
12.189(3)
111.24(2)
105.48(3)
97.14(3)
1120.8(5)
1.81
1.38
12.039(3)
13.180(5)
83.78(3)
74.50(3)
79.94(2)
1125.6(7)
1.81
1.37
3.0–26.5
−95h50
−155k514
−165l516
5062
,
c (A)
h (°)
i (°)
k (°)
3
,
V (A )
Dcalc (g cm−3
)
v (Mo Ka) (mm−1
q Range (°)
)
2.5–28.0
hkl Ranges
−115h50
−155k515
−155l516
5740
5392
4987
No. reflections measured
No. independent reflections
No. reflections observed (I\2|)
Parameters ref.
4698
4361
325
325
649
R (observed reflections)
R (all reflections)
Residual electron density (e A
R1=0.023, wR2=0.061
R1=0.027, wR2=0.064
+0.4
R1=0.058, wR2=0.210
R1=0.062, wR2=0.218
+0.9
R1=0.077, wR2=0.170
R1=0.206, wR2=0.221
+3.0
−3
,
)
−0.5
−1.1
−1.6
1H, NH(Imine)]. 13C NMR (acetone-d6): 47.4
[CH3(NMe2)], 121.2, 126.5, 127.4 [s, C1, C3, C5], 135.9,
137.2 [s, C4, C6], 135.5, 140.3, 145.1, 146.3, 150.5 [m,
C6F5], 176.7 [CN(Imine)]. 19F NMR (acetone-d6):
S[w(Fo 2)2]}1/2. Drawings were produced with SCHAKAL
[23]. Table 2 lists the crystallographic data.
4. Supplementary material
3
3
−133.0 [d, J=23.0 Hz, 4F, F2,6], −164.2 [t, J=
3
20.7 Hz, 2F, F4], −165.5 [t, J=20.7 Hz, 4F, F3,5].
The crystallographic data of the structures described
in this paper were deposited with the Cambridge Crys-
tallographic Data Centre as supplementary publication
No. CCDC 134723–134725. Copies of these data are
available free of charge from the following address: The
Director, CCDC, 12 Union Road, Cambridge CB2
1EZ, UK (fax: +44-1223-336033; e-mail: teched@
chemcrys.cam.ac.uk).
Anal. Calc. for C21H12F10N2S2Zn (611.9): C, 41.07; H,
1.97; N, 4.22; Zn, 10.74. Found: C, 41.23; H, 1.98; N,
4.58; Zn, 10.69%.
3.2. Structure determinations
Diffraction data were taken by the ꢀ/2q technique
on a Nonius CAD4 diffractometer using graphite-
filtered Mo Ka radiation. They were treated without an
absorption correction. The structures were solved with
direct methods and refined anisotropically with the
SHELX program suite [22]. Hydrogen atoms were in-
cluded with fixed distances and isotropic temperature
factors 1.2 times those of their attached atoms. Parame-
ters were refined against F2. In the case of 2c the low
quality of the crystals yielded a somewhat unsatisfac-
tory data set resulting in a high R value and large
standard deviations. The R values are defined as
R1=SꢀFoꢀ−ꢀFcꢀ/SFo and wR2={S[w(Fo 2−Fc 2)2]/
Acknowledgements
This work was supported by the Deutsche
Forschungsgemeinschaft.
References
[1] G.M. Robertson, in: A.R. Katritzky, O. Meth-Cohn, C.W. Rees
(Eds.), Comprehensive Organic Functional Group Transforma-
tions, vol. 3, Pergamon, Oxford, 1995, pp. 403–423.

B. Mu¨ller, H. Vahrenkamp / Inorganica Chimica Acta 300–302 (2000) 181–185
185
[2] P. Pfeiffer, E. Buchholtz, J. Prakt. Chem. 129 (1930) 163.
[3] P. Pfeiffer, T. Hesse, J. Prakt. Chem. 149 (1937) 217.
[4] B. Kamenar, B. Kaitner, A. Stefanovic, T.N. Waters, Acta
Crystallogr., Sect. C 46 (1990) 1627.
[13] B. Mu¨ller, A. Schneider, M. Tesmer, H. Vahrenkamp, Inorg.
Chem. 38 (1999) 1900, and Refs. cited therein.
[14] M. Bochmann, K.J. Webb, M.B. Hursthouse, M. Mazid, J.
Chem. Soc., Dalton Trans. (1991) 2317.
[5] J.P. Costes, F. Dahan, J.P. Laurent, Inorg. Chem. 33 (1994)
2738.
[15] H. Gru¨tzmacher, M. Steiner, H. Pritzkow, L. Zsolnai, G. Hut-
tner, A. Sebald, Chem. Ber. 125 (1992) 2199.
[6] J.P. Costes, F. Dahan, J.P. Laurent, Inorg. Chem. 34 (1995)
3102.
[16] R. Burth, H. Vahrenkamp, Inorg. Chim Acta 282 (1998) 193,
and Refs. cited therein.
[7] J.P. Costes, J.P. Laurent, P. Chabert, G. Commenges, F. Dahan,
Inorg. Chem. 36 (1997) 656.
[8] J.P. Costes, F. Dahan, A. Dupuis, J.P. Laurent, Inorg. Chem. 36
(1997) 3429.
[9] R. Tewari, R.C. Srivastava, Acta Crystallogr., Sect. B 27 (1971)
1644.
[17] A. Abufarag, H. Vahrenkamp, Inorg. Chem. 34 (1995) 3279.
[18] M. Fo¨rster, R. Burth, A.K. Powell, T. Eiche, H. Vahrenkamp,
Chem. Ber. 126 (1993) 2643.
[19] D.J. Gale, J.F.K. Wilshire, Aust. J. Chem. 28 (1975) 2447.
[20] H. Bu¨rger, W. Sawodny, U. Wannagat, J. Organomet. Chem. 3
(1965) 113.
[10] A.C. Brathwaite, T.N. Waters, J. Inorg. Nucl. Chem. 35 (1973)
3223.
[11] A.C. Brathwaite, P.E. Wright, T.N. Waters, J. Inorg. Nucl.
Chem. 37 (1975) 37.
[21] M.E. Peach, Can. J. Chem. 46 (1968) 2659.
[22] G.M. Sheldrick, SHELXS-86, Universita¨t Go¨ttingen, Germany,
1986; SHELXL-93, Universita¨t Go¨ttingen, Germany, 1993.
[23] E. Keller, SCHAKAL for Windows, Universita¨t Freiburg, Ger-
many, 1999.
[12] C.M. Dani, A.K. Das, Indian J. Chem., Sect. A 23 (1984) 1041.
.