An investigation of Cu(II) and Ni(II)-catalysed hydrolysis of (di)imines

Article abstract of DOI:10.1016/j.ica.2010.05.025

The reactions of six diimine ligands with Cu(II) and Ni(II) halide salts have been investigated. The diimine ligands were Ph₂CN(CH ₂)_nNCPh₂ (n = 2 (Bz₂en, 1a), 3 (Bz₂pn, 1b), 4 (Bz₂bn, 1c)), N,N′-bis-(2-tert- butylthio-1-ylmethylenebenzene)-2,2′diamino-biphenyl (2), N,N′-bis-(2-chloro-1-ylmethylenebenzene)-1,3-diaminobenzene (3) and N,N′-bis-(2-chloro-1-ylmethylenebenzene)-1,2-ethanediamine (4). Reactions of 1a-c, 2-4 with CuCl₂·2H₂O in dry ethanol at ambient temperature led to complete or partial hydrolysis of the diimine ligands to ultimately form copper diamine complexes. The non-hydrolyzed complexes of 1b and 1c, [Cu(L)Cl₂] (L = 1b, 1c), could be isolated when the reactions were carried out at low temperatures, and the half-hydrolyzed complex [Cu(Bzpn)Cl₂] could also be identified via X-ray crystallography. Similarly, reactions of 1a or 1b with NiCl₂·6H₂O or [NiBr₂(dme)] led to rapid hydrolysis of the imines and Ni complexes containing half-hydrolyzed 1a (Bzen; [trans-[Ni(Bzen)₂Br ₂]) and 1b (Bzpn; [Ni(Bzpn)Br₂] could be isolated and identified via single crystal X-ray analysis. Kinetic studies were made of the hydrolyses of 1a, 1b in THF and 2 in acetone, in the presence of Cu(II), and of 1a in acetonitrile, in the presence of Ni(II). Activation parameters were determined for the latter reaction and for the copper-catalyzed hydrolysis of 2; the relatively large negative activation entropies clearly indicate rate-determining steps of an associative nature.

Full text of DOI:10.1016/j.ica.2010.05.025

Inorganica Chimica Acta 363 (2010) 3102–3112
Contents lists available at ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
An investigation of Cu(II) and Ni(II)-catalysed hydrolysis of (di)imines
Miklós Czaun a,1, Simphiwe M. Nelana b,2, Ilia A. Guzei c, Catrin Hasselgren d, Mikael Håkansson e,
Susan Jagner d, George Lisensky f, James Darkwa b, Ebbe Nordlander a,
*
a Inorganic Chemistry Research Group, Chemical Physics, Center for Chemistry and Chemical Engineering, Lund University, Box 124, SE-221 00 Lund, Sweden
b Department of Chemistry, University of Johannesburg, P.O. Box 524, Auckland Park 2006, South Africa
c Department of Chemistry, University of Wisconsin, 1101 University Ave., Madison, WI 53706, USA
d Department of Chemical and Biological Engineering, Chalmers University of Technology, SE-412 96 Gothenburg, Sweden
e Department of Chemistry, University of Gothenburg, Kemivägen 10, SE-412 96 Göteborg, Sweden
f Department of Chemistry, Beloit College, 700 College St., Beloit, WI 53511, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
The reactions of six diimine ligands with Cu(II) and Ni(II) halide salts have been investigated. The diimine
ligands were Ph2C@N(CH2)nNC@Ph2 (n = 2 (Bz2en, 1a), 3 (Bz2pn, 1b), 4 (Bz2bn, 1c)), N,N0-bis-(2-tert-butyl-
thio-1-ylmethylenebenzene)-2,20diamino-biphenyl (2), N,N0-bis-(2-chloro-1-ylmethylenebenzene)-1,3-
diaminobenzene (3) and N,N0-bis-(2-chloro-1-ylmethylenebenzene)-1,2-ethanediamine (4). Reactions
of 1a–c, 2–4 with CuCl2ꢀ2H2O in dry ethanol at ambient temperature led to complete or partial hydrolysis
of the diimine ligands to ultimately form copper diamine complexes. The non-hydrolyzed complexes of
1b and 1c, [Cu(L)Cl2] (L = 1b, 1c), could be isolated when the reactions were carried out at low temper-
atures, and the half-hydrolyzed complex [Cu(Bzpn)Cl2] could also be identified via X-ray crystallography.
Similarly, reactions of 1a or 1b with NiCl2ꢀ6H2O or [NiBr2(dme)] led to rapid hydrolysis of the imines and
Ni complexes containing half-hydrolyzed 1a (Bzen; [trans-[Ni(Bzen)2Br2]) and 1b (Bzpn; [Ni(Bzpn)Br2]
could be isolated and identified via single crystal X-ray analysis. Kinetic studies were made of the hydro-
lyses of 1a, 1b in THF and 2 in acetone, in the presence of Cu(II), and of 1a in acetonitrile, in the presence
of Ni(II). Activation parameters were determined for the latter reaction and for the copper-catalyzed
hydrolysis of 2; the relatively large negative activation entropies clearly indicate rate-determining steps
of an associative nature.
Received 19 February 2010
Received in revised form 10 May 2010
Accepted 12 May 2010
Available online 24 May 2010
Dedicated to Animesh Chakravorty
Keywords:
Diimine
Hydrolysis
Catalysis
Copper
Nickel
Ó 2010 Elsevier B.V. All rights reserved.
1. Introduction
the opsin family of photoreceptor pigment proteins, in which reti-
nal is covalently bound as a chromophore to the e-amino group of
Symmetrical diimines formed by Schiff base condensation of
aldehydes or ketones with diamines constitute a well-known class
of ligands in coordination chemistry. The formation of correspond-
ing unsymmetrical mono-condensation products is more difficult.
Methods by which such asymmetric imine–amine ligands can be
prepared include protecting one of the amino groups of the dia-
mine starting material or partial hydrolysis of the diimine [1–3].
Hydrolysis of symmetrical diimines are catalyzed by special en-
zymes, as in the case of wild-type rhodopsin [4]. In general, imine
formation and hydrolysis play important roles in the function of
a lysine residue via an imine linkage. It has been found that metal
ions (e.g. Ca(II), Mg(II)) provide cooperative assistance during the
hydrolysis of the retinylidene chromophore by enhancing the posi-
tive charge on the imine N atom [5].
There are several known examples of partial hydrolysis of
(poly)imine ligands by Cu(II) salts [6–8]. Ray et al. [7] have ob-
served the rearrangement of an unsymmetrical imine ligand into
a symmetrical one when the unsymmetrical ligand is coordinated
to Cu(II). The presence of either free Cu(II) or H+ was found to be
necessary for the rearrangement. The susceptibility of unsymmet-
rical Cu(II) complexes towards hydrolysis is influenced by the ste-
ric strain in two adjacent chelate rings formed upon the
coordination of the imine to the metal ion [7]. Here, we wish to
present the synthesis of a number of new diimine ligands, and a
study of the Cu(II) and Ni(II) catalyzed hydrolysis or partial hydro-
lysis of these ligands. To our knowledge, this is the first kinetic
investigation of copper and nickel ion catalyzed imine hydrolysis
in organic solvents.
* Corresponding author. Tel.: +46 46 222 8118; fax: +46 46 222 4119.
E-mail address: Ebbe.Nordlander@chemphys.lu.se (E. Nordlander).
1
Present address: Loker Hydrocarbon Research Institute, University of Southern
California, University Park, Los Angeles, CA 90089-1661, USA.
2
Present address: Departement of Chemistry, Vaal University of Technology,
Andries Potgietr Blvrd., Vanderbijlpark 1911, South Africa.
0020-1693/$ - see front matter Ó 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2010.05.025

M. Czaun et al. / Inorganica Chimica Acta 363 (2010) 3102–3112
3103
Table 2. Although the imine groups are situated in a trans configu-
ration in the solid state and it has been shown that the ligand can
coordinate as a monodentate ligand (via one of the imine nitro-
gens) in this configuration [10], 1a is expected to coordinate to
Cu(II) as a bidentate ligand as has been found for the distorted tet-
rahedral Cu(I) complex [Cu(Bz2en)2]+ [11] (cf. Table 2). Both the
imine C@N bond distance(s) and the C–N bond distance(s) of the
ligand backbone in 1a are slightly shorter than those found in
the Bz2en ligands of the aforementioned complex, C@N: 1.270(5)
versus 1.285(6)ave Å; C–N: 1.462(6) versus 1.484(7)ave Å. The elon-
gation of these distances in the coordinated ligand is a reflection of
2. Results and discussion
2.1. Syntheses and characterization of diimine ligands
Six diimines, viz. N,N0-bis(diphenylmethylene)-1,2-ethanedia-
mine (Bz2en, 1a), N,N0-bis(diphenylmethylene)-1,3-propanediamine
(Bz2pn, 1b), N,N0-bis(diphenylmethylene)-1,4-butanediamine (Bz2
bn, 1c), N,N0-bis-(2-tert-butylthio-1-ylmethylenebenzene)-2,20dia-
mino-biphenyl (2), N,N0-bis-(2-chloro-1-ylmethylenebenzene)-1,
3-diaminobenzene (3) and N,N0-bis-(2-chloro-1-ylmethyleneben-
zene)-1,2-ethanediamine (4) (Fig. 1) were synthesized and charac-
terized by IR, 1H NMR spectroscopy, mass spectrometry and, in the
case of 1a and 2 by single crystal X-ray diffraction (cf. Section 4).
Complete formation of imine C@N bonds in all the products was
identified by loss of the N–H stretching vibrations and of the
stretches for the C@O bonds at 1660 cmꢁ1 for 1a–c or CH@O bonds
at 1700 cmꢁ1 for 2–4, as well as the appearance of characteristic
C@N stretching bands at approximately 1620 cmꢁ1 in the IR spectra.
The mass spectra were in complete agreement with the expected
formulae, with peaks corresponding to the molecular ion being
observed in all cases.
the relative strain on the ligand as well as the
to the metal.
r electron donation
Single crystals of N,N0-bis-(2-tert-butylthio-1-ylmethyleneben-
zene)-2,20diamino-biphenyl (2) were obtained by recrystallization
of the crude product from ethanol at 5 °C. The molecular structure
of 2 determined by X-ray crystallography is shown in Fig. 3 and se-
lected bond distances, angles and torsion angles are collated in Ta-
ble 2. The two phenyl rings bearing tert-butylthio moieties are
almost parallel but the biphenyl backbone does not show planar-
ity; the dihedral angle between the planes defined by C27–C3–
C49 and C15–C6–C26 is 13.0° while the torsion angle determined
by C20–C28–C29–C17 is 46.8°. Although the tert-butyl groups are
situated on the opposite sides of the phenyl rings in the solid state,
2 is expected to coordinate to Cu(II) with a distorted planar geom-
etry in solution. Rotation around the C24–C22 and C38–C25 bonds
and the flexibility of the biphenyl backbone permits 2 to coordi-
nate to Cu(II) with different geometry from that found in the free
ligand.
Single crystals of Bz2en (1a) could be grown from methanol at
room temperature by slow evaporation of the solvent and the
structure was determined by X-ray crystallography. The molecular
structure of 1a is shown in Fig. 2, crystallographic data are summa-
rized in Table 1 and relevant bond distances and angles are listed in
(CH2)n
N
N
2.2. Reaction of diimine ligands with Cu(II) and Ni(II) salts
Ph
Ph
N
N
Ph
Ph
SBut ButS
In attempts to synthesize complexes of the general formula
[Cu(diimine)Cl2], equimolar amounts of the appropriate diimine
(1a–c, 2–4) and CuCl2ꢀ2H2O were allowed to react in ethanol solu-
tion at ambient temperature. In the cases of ligands 1a–c, the reac-
tions resulted in dark green solutions from which after standing at
ꢁ10 °C microcrystalline products could be isolated. In the case of
Bz2en (1a), blue and green crystals were isolated after recrystalli-
zation; the blue crystals were identified by X-ray crystallography
as the hydrolysis product [Cu(en)Cl2] (en = 1,2-ethanediamine),
the structure of which has been published previously [12]. This is
in agreement with the observation by Chowdhury et al. that the
reaction of 1a with Cu(ClO4)2ꢀ6H2O (in 2:1 M ratio) leads to the for-
mation of a green complex, presumably [Cu(Bz2en)2](ClO4)2, that
could not be isolated due to its ready hydrolysis to form the violet
[Cu(en)2](ClO4)2 complex [11]. While none of the green crystals
were of sufficient quality for a structure determination, mass spec-
trometry suggests the formula [Cu(Bz2en)2]Cl2.
1a n=2; Bz2en
1b n=3; Bz2pn
2
1c
n=4; Bz2bn
N
N
N
N
Cl
Cl
Cl
Cl
3
4
Fig. 1. Schematic structures of the imine ligands used in this study.
In the case of the ligand Bz2pn (1b), crystals with slightly differ-
ent morphologies and slightly different colours (different shades of
turquoise) were isolated and identified by single crystal X-ray dif-
fraction as [Cu(Bz2pn)Cl2] (5) and the partial hydrolysis product
[Cu(Bzpn)Cl2] (6), respectively (vide infra). A low temperature
(107 K) EPR spectrum of 5 (Figs. S1 and S2, Supplementary mate-
rial) confirmed the essentially square planar coordination environ-
ment for the Cu(II) ion and suggested the presence of two nitrogen
donors. For Bz2bn (1c), reaction of CuCl2ꢀ2H2O with 1c in ethanol
followed by immediate cooling (cf. Section 4) led to the formation
of
a complex that was identified by mass spectrometry as
[Cu(Bz2bn)Cl2] but no crystallographic results could be obtained.
Stirring the complex of 1c and CuCl2ꢀ2H2O in ethanol at room tem-
perature gave a benzophenone precipitate in high yield, again indi-
cating hydrolysis of the diimine.
The molecular structures of 5 and 6 are shown in Fig. 4
and selected bond lengths and angles are compiled in Table 2.
Fig. 2. An iORTEP [9] plot of the molecular structure of Bz2en, 1a, showing the atom
labelling scheme. Thermal ellipsoids are drawn at the 50% probability level except
for hydrogen atoms which are shown at the 10% level.

3104
M. Czaun et al. / Inorganica Chimica Acta 363 (2010) 3102–3112
Table 1
Crystallographic data for compounds 1a, 2, 5, 6 and 7.
Bz2en (1a)
28H24N2
2
[Cu(Bz2pn)Cl2] (5)
[Cu(Bzpn)Cl2] (6)
[Ni(Bzen)2Br2] (7)
Chemical formula
Mw (g/mol)
Space group
a (Å)
b (Å)
c (Å)
C
C34H36N2S2
536.81
P1 (No. 2)
C29H26Cl2CuN2
536.99
C16H18Cl2CuN2
372.77
C30H32Br2N4Niꢀ2CHCl3
905.86
388.49
P21/n (No. 14)
12.537(1)
5.579(2)
16.037(1)
90
104.417(8)
90
1086.4(4)
2
ꢀ
ꢀ
C2/c (No. 15)
32.403(2)
6.261(1)
21.734(2)
90
130.96(2)
90
3329.8(12)
8
P21/n (No. 14)
10.0158(5)
10.2255(5)
18.2047(9)
90
103.121(1)
90
1815.79(16)
2
P1 (No. 2)
9.849(2)
12.852(3)
13.235(3)
66.32(3)
86.39(3)
89.51(3)
1530.9(5)
2
10.1405(1)
10.5402(2)
12.3653(3)
80.172(1)
76.040(1)
87.223(1)
1263.74(4)
2
a
(°)
b (°)
d (°)
V (Å3)
Z
T (K)
293(2)
1.54178
1.188
293(2)
0.71073
1.164
293(2)
0.71073
1.411
293(2)
0.71073
1.487
100(2)
0.71073
1.657
k (Å)
Dcalc (g cm3)
l
(mmꢁ1
)
0.53
0.198
1.092
1.627
3.206
Measured reflections
Unique reflections
1683
1600
22105
9090
14047
4618
5938
2926
29473
3718
No of reflections Io > 2
r(Io)
843
4128
2978
2010
3432
Data/restraints/parameters
R1 [I > 2r(I))]
wR2
1600/0/143
0.058
9090/0/318
0.051
4618/0/307
0.0504
2926/0/192
0.0367
3718/0/205
0.0198
0.154a
0.142
0.103b
0.0801c
0.0996
0.1174d
0.0748
0.0494e
R1(all reflections)
0.142
0.0231
wR2 (all reflections)
0.250
0.24, ꢁ0.26
0.121
0.40, ꢁ0.33
0.0969
0.33, ꢁ0.27
0.1373
0.49, ꢁ0.42
0.0519
0.513, ꢁ0.389
Final residual peaks (max, min, e Åꢁ3
)
a
b
c
w = 1/[
w = 1/[
w = 1/[
w = 1/[
w = 1/[
r
r
r
r
r
2(Fo2) + (0.1167P)2 + 0.5732P], P = (Fo2 + 2Fc2)/3.
2(Fo2) + (0.0450P)2 + 0.0000P], P = (Fo2 + 2Fc2)/3.
2(Fo2) + (0.0285P)2 + 0.5222P], P = (Fo2 + 2Fc2)/3.
2(Fo2) + (0.100P)2], P = (Fo2 + 2Fc2)/3.
d
e
2(Fo2) + (0.0252P)2 + 1.4163P], P = (Fo2 + 2Fc2)/3.
Table 2
Selected distances and angles.
Bz2pn (1a)
[Cu(Bz2pn)Cl2] (5)
Cu(Bzpn)Cl2 (6)
Cu–N4
Cu–N5
Cu–Cl2
Cu–Cl3
N4–C12
N5–C24
N4@C8
N5@C19
2.028(3)Å
2.031(3)
2.217(1)
2.260(1)
1.478(4)
1.469(5)
1.287(4)
1.301(5)
Cu–N1
Cu–N2
Cu–Cl1
Cu–Cl2
N2–C3
N1–C1
N2@C4
1.994(4)
2.009(4)
2.246(1)
2.254(1)
1.474(6)
1.470(6)
1.276(6)
N–C
N@C
1.462(6)
1.270(5)
N4–Cu–N5
N4–Cu–Cl3
N5–Cu–Cl2
Cl2–Cu–Cl3
N4–Cu–Cl2
N5–Cu–Cl3
86.38(12)°
87.44(9)
92.07(9)
95.05(4)
176.92(9)
155.32(9)
N1–Cu–N2
N1–Cu–Cl2
N2–Cu–Cl1
Cl1–Cu–Cl2
N1–Cu–Cl1
N2–Cu–Cl2
91.01(16)
91.18(12)
89.30(12)
97.03(5)
158.45(12)
156.61(12)
(2)
[Cu(Bz2en)2]ClO4 [11]
[Ni(Bzen)2Br2] (7)
Ni(Bzpn)Br2 (8)
S1–C27
S1–C19
S2–C26
S2–C13
N1–C21
N2–C23
N1@C24
N2@C38
1.790(2) Å
1.854(2)
1.793(2)
1.858(2)
1.420(2)
1.426(2)
1.272(2)
1.274(2)
Cu–N11
Cu–N14
Cu–N21
Cu–N24
N11–C12
N14–C13
N21–C22
N24–C23
N11@C10
N14@C15
N21@C20
N24@C25
2.014(4)
Ni–N1
Ni–N2
Ni–Br
2.0783(13)
2.1918(13)
2.58324(19)
Ni–N1
Ni–N2
1.994(8)
1.992(9)
2.358(3)
2.390(2)
1.468(12)
1.450(15)
1.305(10)
2.183(4)
2.065(4)
2.077(5)
1.486(7)
1.491(7)
1.479(6)
1.478(6)
1.284(6)
1.279(6)
1.282(6)
1.294(6)
Ni–Br1
Ni–Br2
N1–C14
N2–C16
N1@C1
N1–C1
N2–C2
N2@C3
1.477(2)
1.477(2)
1.284(2)
N2–Ni–N1
N2–Ni–N10
N1–Ni–Br
N10–Ni–Bt
N2–Ni–Br
N20–Ni–Bt
80.82(5)
99.18(5)
93.55(4)
86.45(4)
86.23(4)
93.77(4)
N2–Ni–N1
N2–Ni–Br1
N1–Ni–Br1
Br1–Ni–Br2
N2–Ni–Br2
N1–Ni–Br2
94.6(4)
98.8(3)
131.5(2)
122.26(11)
104.2(3)
98.7(2)
C26–S2–C13
C27–S1–C19
C38–N2–C23
C24–N1–C21
105.14(9)°
102.02(9)
116.88(14)
120.17(15)
Bond distances are typical with C@N shorter than C–N (5:
C@Nave = 1.294(5), C–Nave = 1.474(5) Å; 6: C@N = 1.276(6), C–Nave
1.472(6) Å). The Cu–N distances in the non-hydrolyzed complex
Cu–Nave = 2.002(4) Å), possibly because of steric hindrance. There
is no significant difference in the Cu–N bond distances for the coor-
dinated imine nitrogen and the amine nitrogen in complex 6. In
both structures the copper coordination deviates from planarity.
=
(5) are slightly larger than in 6 (5: Cu–Nave = 2.030(3) Å; 6:

M. Czaun et al. / Inorganica Chimica Acta 363 (2010) 3102–3112
3105
When N,N0-bis-(2-chloro-1-ylmethylenebenzene)-1,2-ethane-
diamine (4) was reacted with CuCl2ꢀ2H2O in ethanol, the initial
green colour immediately turned blue and a blue solid precipi-
tated. After filtration, the solid was extracted with hexane in a
Soxhlet extractor. Recrystallization gave single crystals that were
identified as the [Cu(en)Cl2] hydrolysis product [12] by X-ray
crystallography.
The characterization of the products from the above-mentioned
reactions shows that, in all cases, at least partial hydrolysis of the
diimine ligand takes place. In the case of the reaction of Bz2pn (1b)
with the copper salt, it was possible to isolate and crystallograph-
ically characterize the non-hydrolyzed complex (5) as well as the
half-hydrolyzed complex 6 (vide supra). The fact that hydrolysis of
the ligands in this study is slow in the absence of a copper salt
suggests that the hydrolytic reaction is catalyzed by the presence of
the transition metal. In related work, Walters and coworkers [8]
have shown that the symmetric thienyl-diimine ThCH@N(CH2)2-
N@CHTh (Th = 2-thienyl) is rapidly hydrolysed in the presence of
Cu(II) ions; this occurs via a two-step process involving the initial
formation of the half-hydrolyzed ligand.
Divalent metal ions such as Ni(II), Cu(II) and Zn(II) are good Le-
wis acids that tend to lower the pKa of bound water molecules and
thus promote hydrolytic reactions. In order to study whether the
acceleration of hydrolysis was applicable to other metal salts than
CuCl2, the reactions of the ligands Bz2en (1a) and 1b with various
Ni(II) salts were studied. Reaction of 1a with NiCl2ꢀ6H2O in dichlo-
romethane, or with [NiBr2(dme)] in acetonitrile (dme = dime-
thoxyethane) were undertaken. In both cases, the reactions led to
an immediate colour change to purple, and the precipitation of
purple solids. Fast Atom Bombardment (FAB+) and ESI mass spectra
of the purple solids revealed that the complexes [Ni(Bz2en)X2]
(X = Cl, Br) were formed, but the presence of peaks attributable
to the half-hydrolyzed ligand Bzen, as well as benzophenone, sug-
gested that partial hydrolysis of the ligands was taking place. The
Fig. 3. An iORTEP plot of N,N0-bis-(2-tert-butylthio-1-ylmethylenebenzene)-
2,20diamino-biphenyl (2). Thermal ellipsoids are drawn at the 50% probability level
except for hydrogen atoms which are shown at the 10% level.
The angle between the N–Cu–N plane and the Cl–Cu–Cl plane is
23.9° for 5 and 30.7° for 6. For the more sterically hindered
[Cu(Bz2en)2]ClO4 the angle between the two N–Cu–N planes is
63.9° [11].
An ethanol solution of one equivalent of CuCl2ꢀ2H2O added to a
dichloromethane solution of N,N0-bis-(2-tert-butylthio-1-ylmeth-
ylenebenzene)-2,20diamino-biphenyl (2) resulted in an immediate
colour change to dark red and precipitation of green microcrystals
after a few hours. The IR and mass spectra of these microcrystals
were found to be consistent with the product being [Cu(2,20-diami-
nobiphenyl)Cl2], which was also obtained independently by react-
ing 2,20-diaminobiphenyl with copper(II) dichloride dihydrate in
ethanol.
IR spectra of the complexes showed
m(C@N) bands at 1600–
1620 cmꢁ1, confirming the presence of imine ligands, but the
NMR spectra were broad, suggesting that the complexes have close
to tetrahedral symmetry.
The reaction between N,N0-bis-(2-chloro-1-ylmethyleneben-
zene)-1,3-diaminobenzene (3) and CuCl2ꢀ2H2O led to the forma-
While the complexes [Ni(Bz2en)X2] were found to be stable in
the solid state, attempts to crystallize them led only to hydrolysis
despite the fact that care was taken to use dry solvents and dry
atmosphere. Thus, pale blue crystals isolated from the attempted
crystallization of [Ni(Bz2en)X2] in CHCl3 were identified by X-ray
crystallography as the complex [Ni(Bzen)2Br2] (7), containing two
half-hydrolyzed ligands. The molecular structure of 7 is shown in
Fig. 5 and selected bond lengths and bond angles for 7 are listed
tion of an uncharacterized dark brown solid. In
a control
experiment, CuCl2ꢀ2H2O was reacted with 1,3-diaminobenzene in
ethanol; this reaction resulted in the formation of the same type
of dark brown, insoluble material, which may be a polymeric
CuCl2–1,3-diaminobenzene complex. Evaporation of the solvent
from the filtrate led to the formation of an oil that was identified
as the hydrolysis product 2-chloro-benzaldehyde by 1H NMR.
Fig. 4. iORTEP plots of the molecular structures of (a) [Cu(Bz2pn)Cl2] (5) and (b) [Cu(Bzpn)Cl2] (6). Thermal ellipsoids are drawn at the 50% probability level except for
hydrogen atoms which are shown at the 10% level.

3106
M. Czaun et al. / Inorganica Chimica Acta 363 (2010) 3102–3112
Fig. 5. An iORTEP plot of the molecular structure of [Ni(Bzen)2Br2] (7) showing the atom labelling scheme. Thermal ellipsoids are drawn at the 50% probability level except for
hydrogen atoms which are shown at the 10% level.
in Table 2. Two chloroform molecules per Ni(II) are included in the
crystal lattice without any binding interaction with the complex. In
complex 7, the Ni(II) ion is coordinated in an approximately octa-
hedral geometry, with N–Ni–N bond angles of 80.82(5)° on the car-
bon-bridged side and 99.18(5)° on the open side. The two half-
hydrolyzed Bzen ligands are coordinated in a head-to-tail fashion
in one plane, i.e. the amine and imine functionalities of the two li-
gands are trans to each other. The aryl rings bound to the imine
carbons lie nearly perpendicular to the plane formed by the metal
and the coordinated nitrogen atoms. Bond distances are typical
with C@N shorter than C–N (C@N = 1.284(2), C–Nave = 1.477(2) Å).
In the case of the analogous reaction of 1b with NiBr2ꢀ6H2O, the
formation of the half-hydrolyzed product [Ni(Bzpn)Br2] (8), con-
taining one imine–amine ligand, was confirmed by X-ray crystal-
lography (Fig. 6, cf. Section 4.6 and Supplementary material). In
keeping with the observed paramagnetism of the putative parent
compound [Ni(Bz2pn)Br2], the metal in 8 possesses a distorted tet-
rahedral coordination geometry with the angle between the N–Ni–
N plane and the Br–Ni–Br plane being 80.5°. The greater bite angle
of the Bzpn ligand relative to the analogous Bzen ligand may ex-
plain the formation of the tetrahedral complex rather than the
octahedral complex [Ni(Bzpn)2Br2], analogous to 7. In contrast to
what was observed for 7, the Ni–N bond distances for both amine
and imine functionalities are almost the same – Ni–Namine =
1.992(9) Å; Ni–Nimine = 1.994(8) Å. There is a slight difference in
the Ni–Br bond distances with Ni(1)–Br(2) [2.392(3) Å] longer than
Ni(1)–Br(1) [2.358(3) Å].
2.3. Kinetic studies of the transition metal catalyzed diimine
hydrolyses
In order to gain further insight into the nature of the transition
metal catalyzed hydrolysis reactions, kinetic data were obtained
for the hydrolyses of ligands 1a, 1b and 2 in the presence of
CuCl2ꢀ2H2O, as well as for the hydrolysis of 1b in the presence of
NiCl2ꢀ2H2O.
2.3.1. Kinetic investigation of the copper-catalyzed C@N bond
hydrolysis in 2
Acetone solutions of 2 and CuCl2ꢀ2H2O were mixed in a thermo-
stated quartz cell. The concentration of an intermediate complex,
possibly [Cu(2)Cl2], was determined by UV–Vis spectroscopy by
measuring the absorbance of the reaction mixture at 473.0 nm
since the products did not show absorption in this region. Fig. 7
shows the time-dependent variation of the concentration of
‘‘[Cu(2)Cl2]” in a typical experimental run.
As illustrated in Fig. 7, it was generally possible to monitor both
the formation and the subsequent hydrolysis of [Cu(2)Cl2] under
the conditions used (cf. Table S1). The initial rise is faster than
the second process, which is taken to represent the rate-determin-
ing step of the hydrolysis reaction (k1 > k2, vide infra). The actual
addition of water to the imine bonds may take place in the coordi-
nation sphere of Cu(II)-ions, in accordance with the following sche-
matic mechanism (Scheme 1).
The reaction rates (see Table 3) for formation of the initial cop-
per diimine complex, k1, and the hydrolysis reaction rate, k2, are
both first order in CuCl2ꢀ2H2O as seen by the linear dependence
on concentration in Fig. 8.
Fig. 6. An iORTEP plot of the molecular structure of [Ni(Bzpn)Br2] (8) showing the
atom labelling scheme. Thermal ellipsoids are drawn at the 50% probability level
except for hydrogen atoms which are shown at the 10% level.

M. Czaun et al. / Inorganica Chimica Acta 363 (2010) 3102–3112
3107
Table 3
Observed rate constants for the hydrolysis of diimine ligands in their reaction with
CuCl2ꢀ2H2O.
Ligand
k1 (Mꢁ1 sꢁ1
)
10ꢁ4 ꢂ k2 (sꢁ1
)
2
1.316
–
–
6.167
3.322
0.068a
1a (Bz2en)
1b (Bz2pn)
a
Rate constant is determined from initial rate.
Fig. 7. The time dependence of absorbance at 473.0 nm. [CuCl2ꢀ2H2O] = [2] =
1.475 ꢂ 10ꢁ3 M, T = 25 °C, acetone.
The reaction rate of the second step was measured at tempera-
tures varying between 15 °C and 50 °C, and the Eyring equation
(cf. Table S1 for kinetic data and Fig. S3 for an Eyring plot) was used
to derive the following activation parameters for the hydrolysis
of 2:
(20 °C) = 95.6 4.2 kJ molꢁ1
value for
D
Hà = 59.4 2.6 kJ molꢁ1
,
D
Sà = ꢁ123.8 5.5 J molꢁ1 Kꢁ1
, D
Gà
.
The relatively large and negative
D
Sà indicates that the rate-determining step (k2) is of a
associative nature: it may involve the coordination of a water mol-
ecule to the metal, which is followed by a deprotonation step to
form a metal-bound hydroxide that is the active nucleophile in
the hydrolysis reaction.
Fig. 8. Initial rate of the first (squares) and second (circles) step plotted against the
initial concentration [CuCl2ꢀ2H2O] = [2], T = 25 °C, acetone.
asymmetric and symmetric m(NH) resonances of the coordinated
propylenediamine at 3243 and 3207 cmꢁ1, respectively.
The second method for studying the hydrolysis reactions also
involved monitoring the appearance of the hydrolysis product
benzophenone, the concentration of which was determined by
gas chromatography using naphthalene as an internal standard.
In order to speed up the reactions and shorten the data acquisition
time, the hydrolysis data was collected at 50 °C. Fig. 10 shows the
variation in the benzophenone concentration as a function of time
for the copper-catalyzed hydrolysis of 1a and 1b at 50 °C, and these
concentrations were used to calculate the corresponding variations
of the concentrations of [Cu(Bz2en)Cl2] and [Cu(Bz2pn)Cl2]. The
rate constant for the hydrolysis of 1a (k2 = 3.322 ꢂ 10ꢁ4 sꢁ1) was
determined from the slope of the plot of log[Cu(Bz2en)Cl2] versus
time. In the case of 1b, the reaction rate was not calculated because
the time dependence of [Cu(Bz2pn)Cl2] showed that there was an
induction period but an estimate of k2 calculated by the initial rate
method is listed in Table 3.
2.3.2. Kinetic investigation of the copper-catalyzed C@N bond
hydrolysis in 1a and 1b
The copper-catalyzed hydrolysis of 1b was studied for compar-
ison with that of 2. A stoichiometric amount of 1b was placed with
CuCl2ꢀ2H2O in THF in a thermostated glass reactor with an inlet for
sampling via a syringe. This hydrolytic reaction was not amenable
to monitoring by UV–Vis spectroscopy because of strong solvent
absorbance in the region where the complexes absorb (200–
300 nm). Therefore, two other methods of monitoring the reactions
were used. In the first method, the disappearance of the ligand
infrared band at 1622 cmꢁ1
the 1663 cmꢁ1 hydrolysis product band (
(m
(C@N), 1b) and the appearance of
m(C@O), benzophenone)
were used to follow the reaction. The ratio between the absorbance
of the band at 1622 cmꢁ1 (A1) and the band at 1663 cmꢁ1 (A2) is
plotted against time in Fig. 9. The hydrolytic cleavage of the imine
double bond in 1b could also be monitored by the appearance of
"
"
k1
k2
H2O
N
N
.
N
N
Cu
H2N
Cl
NH2
CuCl2 H2O
(slow)
Cu
Cl
Cl
Cl
acetone
(fast)
SBut
ButS
H2O
+
SBut
ButS
O
2 x
ButS
Scheme 1.

3108
M. Czaun et al. / Inorganica Chimica Acta 363 (2010) 3102–3112
Fig. 9. The ratio between the 1622 cmꢁ1 (A1) and 1663 cmꢁ1 bands. [Bz2pn] =
5.000 ꢂ 10ꢁ3 M, [CuCl2ꢀ2H2O] = 5.000 ꢂ 10ꢁ3 M, T = 25 °C, THF.
Fig. 11. Time course for hydrolysis of [Ni(Bz2en)Cl2] monitored at 511 nm and
plot of ln[Ni(Bz2en)Cl2]. [Bz2en]0 = [NiCl2ꢀ2H2O]0 = 3.554 ꢂ 10ꢁ3, [H2O]0 = 1.837 ꢂ
10ꢁ1 M, T = 50 °C, CH3CN.
Fig. 10. Concentration of the hydrolysis product benzophenone measured by GC for
Bz2en (1a) (s) and Bz2pn (1b) (d), [1a]0 = [1b]0 = [CuCl2ꢀ2H2O]0 = [naphthalene] =
7.0 ꢂ 10ꢁ3 M, T = 50 °C, acetone.
Fig. 12. Plot of reaction rate versus the initial water concentration. [Bz2en]0
[NiCl2ꢀ2H2O]0 = 3.554 ꢂ 10ꢁ3, T = 60 °C, CH3CN.
=
2.3.3. Kinetic investigation of the nickel-catalyzed C@N bond
hydrolysis in 1a and 1b
ꢁd½NiðBz2enÞCl2ꢃ ꢁ2d½H2Oꢃ
¼
¼ kobs½NiðBz2enÞCl2ꢃ
ð2Þ
dt
dt
Since the coordination chemistry of Ni(II) and Cu(II) is similar,
and attempts to prepare Ni(II) complexes of the ligands 1a and
1b led to imine hydrolysis, we decided to study the Ni(II) cata-
lyzed hydrolysis of 1a in order to be able to compare it to the
analogous Cu(II) catalyzed reaction. The kinetics of the hydrolysis
was followed by determining the concentration of [Ni(Bz2en)Cl2]
in the reaction solution as a function of time by UV–Vis spectros-
copy (cf. Section 4). A general rate law for the hydrolysis of
[Ni(Bz2en)Cl2] is given in Eq. (1).
The hydrolysis was monitored by following the decrease in the
absorbance for [Ni(Bz2en)Cl2] at 511 nm. A plot of ln[Ni(Bz2en)Cl2]
versus time (Fig. 11) is linear for approximately 1.5 half-lives
(t1/2 = 26 min), indicating that the rate of the reaction is first-order
with respect to the complex concentration. Similarly, kinetic mea-
surements of the rate with respect to water concentration under
pseudo-first order conditions indicated first-order dependence. A
plot of reaction rate versus initial water concentration gave a
straight line (Fig. 12).
Experiments were carried out at different initial [Ni(Bz2en)Cl2]
concentrations for investigation of the reaction rate dependence
on the complex concentration, and the reaction rate as a function
of initial [Ni(Bz2en)Cl2] at low and at high excess of water is shown
in Fig. S5 (Supplementary material). In summary, the kinetic data
indicate that the total order of the hydrolysis of [Ni(Bz2en)Cl2] is
2, i.e. it is first order with respect to both [Ni(Bz2en)Cl2] and H2O
ꢁd½NiðBz2enÞCl2ꢃ ꢁ2d½H2Oꢃ
x
y
¼
¼ k½NiðBz2enÞCl2ꢃ ½H2Oꢃ
ð1Þ
dt
dt
In order to determine the rate dependence on the various reac-
tants, hydrolysis reactions were performed at different water and
complex concentration under pseudo-first order conditions. Thus,
in the presence of a large excess of water, Eq. (1) is reduced to
Eq. (2), kobs being the pseudo-first-order rate constant.

M. Czaun et al. / Inorganica Chimica Acta 363 (2010) 3102–3112
3109
(x = y = 1 in Eq. (1)) and a mean value for the second order reaction
constant (k) of 1.707 ꢂ 10ꢁ4 Mꢁ2 sꢁ1 at 60 °C was obtained.
The activation parameters for the hydrolysis reaction were
determined from the temperature dependence of the rate constant
(k). The Eyring plot of ln(k/T) versus 1/T was linear (Fig. 7) and
priate drying agents and dried over molecular sieves prior to the
reactions. The UV–Vis spectra were collected on a Varian Cary
300 UV–Vis spectrophotometer equipped with a thermostated
sample compartment controlled by a Varian temperature control-
ler with an accuracy of 0.3 °C. Infra-red spectra were obtained
on a Nicolet Avatar 360 spectrometer, and 1H NMR spectra were
recorded on Varian Unity 300 MHz and Inova 500 MHz spectrom-
eters. Fast Atom Bombardment (FAB) mass spectra were obtained
on a JEOL SX-102 mass spectrometer using 3-nitrobenzylalcohol
as matrix and CsI as calibrant. The ligands N,N0-bis(diphenylm-
ethylene)-1,2-ethanediamine (Bz2en, 1a) and N, N0-bis(2-chloro-
ylmethylenebenzene)-1,2-ethanediamine (4) were obtained using
literature procedures for the condensation of 1,2-ethanediamine
with benzophenone [10,11] or 2-chloro-benzaldehyde, respec-
tively [13]. The starting material [NiBr2(dme)] was prepared by a
literature method [14].
gave
D
D
Hà = 26.10 2.89 kJ molꢁ1
,
D
Sà = ꢁ219.71 20.8 J molꢁ1 Kꢁ1
,
Gà = 90.30 9.92 kJ molꢁ1. Although activation parameters are not
discriminating factors in recognizing the reaction pathway, the neg-
ative entropy of activation (
D
Sà) clearly indicates an associative
mode of activation in the rate-determining step. As proposed for
the Cu(II)-hydrolyzed reactions (vide supra), this step may involve
the initial coordination of water to the Ni(II) ion.
3. Summary and conclusions
In this study, Cu(II) and Ni(II) halides were shown to signifi-
cantly enhance the rate of hydrolysis of the six studied diimines
1a–c, 2–4. It is likely that in all cases complexes of the general for-
mula [M(diimine)X2] (M = Ni, Cu; X = Cl, Br) are initially formed.
Thus, the complex [Cu(Bz2pn)Cl2] (5) could be isolated from the
low temperature reaction of Bz2pn (1b) with CuCl2ꢀ2H2O, and spec-
troscopic evidence suggests that the analogous Cu(II) complex of
Bz2bn (1c) is also formed. Considering the strong preference of
the Cu(II) ion for a square planar coordination geometry, it is likely
that all [Cu(diimine)X2] adopt (distorted) square planar geometries
while the analogous Ni(II) complexes are expected to have pseudo-
tetrahedral coordination geometries in order to eliminate steric
strain.
However, the [M(diimine)X2] complexes are not stable in solu-
tion and will readily hydrolyse in the presence of traces of water.
As shown previously by Walters and coworkers [8], these hydro-
lyses are most likely two-step reactions with the initial formation
of a coordinated half-hydrolyzed amine–imine ligand, followed by
the final hydrolysis step to form the metal complex of the analo-
gous diamine ligand. This postulated reaction pathway is indirectly
supported by the isolation and characterization of benzophenone
and the complexes [Cu(Bzpn)Cl2] (6), [Ni(Bzen)2Br2] (7) and
[Ni(Bzpn)Br2] (8).
4.2. Ligand syntheses
4.2.1. Synthesis of N,N0-bis(diphenylmethylene)-1,3-propanediamine
(Bz2pn, 1b)
This ligand was synthesized in accordance with a previously
published procedure [10]. A total of 10.87 g (60.0 mmol) benzo-
phenone and 2.5 mL (30.0 mmol) freshly distilled 1,3-propanedi-
amine were refluxed in 50 mL of anhydrous methanol for
10 days. Three pieces of anhydrous calcium sulfate (Drierite) were
added to the solution every two days. The yellow solution was fil-
tered to remove calcium sulfate. After half the solvent was evapo-
rated, a white solid was recovered by filtration. Recrystallisation of
the crude product (4.6 g) from 55 mL boiling heptane gave fine
white crystals. Yield: 3.1 g (25.7%), m.p.: 82–85 °C, MS (m/z): 402
(M+, 100%), 1H NMR (CDCl3, 400 MHz): 2.05 (quintet, 2H), 3.46 (t,
4H), 7.09–7.55 (m, 20H), IR (KBr, cmꢁ1): 1619 (C@N), 1596, 1574,
779, 767 (Ar), 1444, 2918, 2849 (CHal), 3055, 3024 (CHAr).
4.2.2. Synthesis of N,N0-bis(diphenylmethylene)-1,4-butanediamine
(Bz2bn, 1c)
A total of 10.87 g (60 mmol) benzophenone, 2.64 g (30 mmol)
1,4-butanediamine, and 9 g anhydrous CaSO4 were refluxed in
50 mL of anhydrous methanol for approximately three weeks with
occasional addition of more CaSO4, until the product 1617 cmꢁ1
imine peak was larger than the carbonyl peak of benzophenone
at 1658 cmꢁ1. A white solid was recovered by filtration and rinsed
with methanol. Soxhlet extraction of the solid with 150 mL hep-
tane gave fine white crystals. Yield: 7.75 g (62.1%), m.p.: 115–
6 °C. 1H NMR (CDCl3, 200 MHz) 1.74 (m, 4H) 3.35 (t, 4H), 7.1–7.6
(m, 20H); IR (KBr, cmꢁ1): 1617 (C@N), 1596, 1575, 779, 767 (Ar),
1444, 2910, 2853 (CHal).
Kinetic studies could establish overall observed rates for the Cu-
catalyzed hydrolyses of ligands 1a, 1b and 2, and for the Ni-cata-
lyzed hydrolysis of 1a. For solubility reasons, the copper and nickel
catalyzed hydrolytic reactions were carried out in different sol-
vents; consequently, direct comparison of reaction rates obtained
for the hydrolysis of 1a and 1b using CuCl2ꢀ2H2O and NiCl2ꢀ2H2O
as catalysts is not possible. However, it may be concluded that
the transition metal catalysed hydrolyses of these Schiff bases
are not unique to copper(II) since they readily take place in the
presence of nickel chloride and the obtained activation parameters
are similar for hydrolyses involving the two metal ions (vide supra).
While it was not possible to discern specific reaction steps in the
hydrolysis reactions, these activation parameters clearly suggest
4.2.3. Synthesis of N,N0-bis-(2-tert-butylthio-1-ylmethylenebenzene)-
2,20diamino-biphenyl (2)
A total of 1.69 g (8.70 mmol) o-tert-butylthio-benzaldehyde
[15] was added dropwise into a solution of 0.8 g (4.35 mmol)
2,20-diamino-biphenyl [16] in 15 mL of ethanol at room tempera-
ture. The solution was stirred at 50 °C for 2 h. Yellow precipitate
was filtered and washed with small amount of cold ethanol. Yield:
2.1 g (90%), the yellow solid was recrystallized from 30 mL of boil-
ing ethanol. After cooling, the crystalline material was filtered and
the mother liquor was stored overnight at ꢁ30 °C resulting in more
precipitate that was isolated by filtration. Total yield: 2.05 g (88%),
m.p.: 93 °C. 1H NMR (ppm): 1.06 (But, 18H), 6.90–6.97; 7.10–7.32;
7.71–7.76 (Ar, 16H), 8.93 (N@CH, 2H) IR (KBr)/(cmꢁ1): 1350, 1360
(tert-Bu), 1583 (Ar), 1618 (C@N), 2861–2975 (CHtert-Bu), 3019, 3059
(CHAr). FAB-MS: 537 (M+, 73.4%), 479 (M+–But, 100%), 423 (M+–
2But). Single crystals obtained from ethanol were found to be
appropriate for X-ray diffraction.
an associative rate-determining step (large and negative
D
Sà), pos-
sibly involving metal-coordination of water to form a metal-bound
hydroxide that acts as nucleophile in the hydrolytic reactions. This
correlates well with the known Lewis acidity of Cu(II) and Ni(II) –
in contrast, complexes of Bz2en and Bz2pn involving coordination
to the less acidic Cu(I), Pd(II) and Pt(II) metal ions appear to be sta-
ble in solution [10,11].
4. Experimental
4.1. General
All reactions were carried out without the rigorous exclusion of
air, unless otherwise stated. All solvents were distilled over appro-

3110
M. Czaun et al. / Inorganica Chimica Acta 363 (2010) 3102–3112
4.2.4. Synthesis of N,N0-bis-(2-chloro-1-ylmethylenebenzene)-1,3-
diaminobenzene (3)
itated 3 h later. Yield: 0.0232 g (70%), IR (KBr)/(cmꢁ1): 3211,
3287(m
NH), MS (m/z): 318 (M+, 45%). The IR and mass spectrometry
A total of 2.16 g (20.0 mmol) 1,3-phenylenediamine in 60 mL of
ethanol was added dropwise to a solution of 4.7 mL (21.0 mmol) o-
chloro-benzaldehyde in 80 mL of ethanol under nitrogen. During
the addition, a yellow solid precipitated. The suspension was stir-
red at 60 °C for 1h, and was then filtered and washed with
3 ꢂ 5 mL of ethanol, and dried under vacuum. Yield: 3.0 g
(42.5%), m.p.: 80–85 °C. MS (m/z): 353 (M+, 100%), 1H NMR
(ppm): 8.92 (N@CH, 2H), IR (KBr)/(cmꢁ1): 1619 (C@N), 1583 (Ar),
3058 (CHAr).
data are consistent with the product being [Cu(2,20-diaminobi-
phenyl)Cl2], which was obtained independently by reacting 2,20-
diaminobiphenyl with copper(II) dichloride dihydrate in ethanol.
4.3.5. Reaction between N,N0-bis-(2-chloro-1-ylmethylenebenzene)-
1,3-diaminobenzene (3) and CuCl2ꢀ2H2O
A total of 0.5 g of 3 was suspended in 20 mL ethanol and was
added to a solution of 0.559 g CuCl2ꢀ2H2O in 5 mL ethanol. The
resultant dark brown solid was washed with hexane three times.
Yield: 0.344 g (brown powder). The filtrate was evaporated and
the resultant yellow oil was identified as 2-chloro-benzaldehyde
with 1H NMR.
4.3. Reaction of diimine ligands with copper dichloride dihydrate
4.3.1. Reaction between N,N0-bis(diphenylmethylene)-1,2-
ethanediamine (Bz2en, 1a) and CuCl2 2H2O
4.3.6. Reaction between N,N0-bis-(2-chloro-1-ylmethylenebenzene)-
1,2-ethanediamine (4) and CuCl2 2H2O
A total of 0.132 g (0.77 mmol) CuCl2ꢀ2H2O dissolved in 5 mL
EtOH was added to a solution of 0.30 g (0.77 mmol) 1a in 15 mL
EtOH. Upon mixing the solution color turned darker. Blue and
green crystals appeared after keeping the mixture in refrigerator
for a week. Yield: 0.173 g green crystals, 0.179 g blue crystals.
The structure of the blue crystals was identified by X-ray crystal-
lography as [Cu(en)Cl2], the structure of which has been published
previously [12]. Combining the same amount of pre-chilled
solutions (4 °C) and storing at 4 °C gave 0.276 g (0.30 mmol) of
[Cu(Bz2en)2Cl2] as green crystals – FAB-MS: 839 (Cu(Bz2en)2,
35%), 486 (Cu(Bz2en)Cl, 28%), 451 (Cu(Bz2en), 100%) and 389
(Bz2en+H), 94%).
A total of 1.0 g of 4 was dissolved in 40 mL ethanol and was
added to solution of 0.241 g CuCl2ꢀ2H2O in 6 mL ethanol. The green
color turned into blue immediately and a blue solid precipitated.
After filtration the solid was extracted with hexane in a Soxhlet
extractor. Yield: 0.733 g. Single crystals were grown from a con-
centrated DMSO solution and were identified as [Cu(en)Cl2] [12]
by X-ray crystallography.
4.4. Reaction of diimine ligands with Ni(II) halide salts
4.4.1. Reaction between N,N0-bis(diphenylmethylene)-1,2-
ethanediamine (Bz2en, 1a) and NiCl2ꢀ6H2O
4.3.2. Reaction between N,N0-bis(diphenylmethylene)-1,3-
propanediamine (Bz2pn, 1b) and CuCl2ꢀ2H2O
A total 0.300 g (0.77 mmol) Bz2en was dissolved in 20 mL of
acetonitrile and 0.184 g (0.77 mmol) NiCl2ꢀ6H2O was added. The
colour of the solution turned into purple immediately and a purple
solid precipitated. The slurry was stirred for 2 h, then the reaction
mixture was filtered, and the solid was washed twice with hexane.
Yield: 0.28 g (70%). MS (m/z): 481 (3.5%), 445 (8%), 389 (Bz2en+,
100%), 154 (45%), 136 (30%), IR (KBr)/(cmꢁ1): 1617 (C@N), (Ar),
2953, 2923, 2851 (CHal), 3058.6, (CHAr).
A total of 0.128 g (0.75 mmol) CuCl2ꢀ2H2O dissolved in 5 mL
ethanol was added to a solution of 0.30 g (0.75 mmol) 1b in
15 mL ethanol, whereupon the colour turned dark green and a
microcrystalline precipitate appeared immediately. The crystals
were filtered off and washed with ether. Yield: 0.365 g (91%). Sin-
gle crystals were grown from the mother liquor at 4 °C over a cou-
ple of days. Crystals with two types of morphologies and slightly
different colours were obtained. They were identified by single
crystal X-ray diffraction as [Cu(Bz2pn)Cl2], 5, and [Cu(Bzpn)Cl2],
6, respectively. Combining the same amount of pre-chilled solu-
tions (ꢁ20 °C) and storing at ꢁ20 °C gave 0.360 g product with
no NH peaks in the IR spectrum, while the room temperature reac-
4.4.2. Reaction between N,N0-bis(diphenylmethylene)-1,3-
propanediamine (Bz2pn, 1b) and NiCl2ꢀ6H2O
A total of 0.300 g (0.75 mmol) Bz2pn was dissolved in 20 mL of
acetonitrile and 0.177 g (0.75 mmol) NiCl2ꢀ6H2O was added. The
colour of the solution turned into purple immediately and a purple
solid precipitated. The suspension was stirred for 4 h, then it was
filtered and the solid was washed twice with hexane. Yield:
0.15 g (40%) IR (KBr)/(cmꢁ1): 1613 (C@N), (Ar), 2961, 2943, 2926
(CHal), 3024, 3056 (CHAr).
tion gave NH peaks at 3243 and 3206 cmꢁ1
.
4.3.3. Reaction between N,N0-bis(diphenylmethylene)-1,4-
butanediamine (Bz2bn, 1c) and CuCl2ꢀ2H2O
A total of 0.123 g (0.72 mmol) of CuCl2ꢀ2H2O in 5 mL ethanol
was added to solution of 0.30 g (0.72 mmol) 1c in 100 mL warm
ethanol and was immediately chilled in a 4 °C refrigerator. Small
crystals were isolated by filtration and washed with ether. Yield:
0.095 g. FAB-MS (m/z) 549 ((Bz2bn)CuCl2, 2%), 515 ((Bz2bn)CuCl,
4%), 479 ((Bz2bn)Cu, 6%), 417 (Bz2bn+H, 100%). IR (KBr, cmꢁ1):
1604 (C@N), 1575 (Ar), 1445 (CH2) with no observable resonance
4.4.3. Reaction between N,N0-bis(diphenylmethylene)-1,2-
ethanediamine (Bz2en, 1a) and [NiBr2(dme)]
To a suspension of [NiBr2(dme)] (0.10 g, 0.32 mmol) in 15 ml
CH2Cl2 was added a solution of Bz2en (1a, 0.13 g, 0.32 mmol) in
15 ml CH2Cl2. The mixture turned from grey to purple colour with-
in 5 min. This purple solution was stirred for 5 h under nitrogen
and the solvent was removed on a rotary evaporator to give a pur-
ple powder. The powder was dissolved in chloroform and allowed
to slowly evaporate to give light blue crystals of [Ni(Bzen)2Br2] (7)
suitable for X-ray analysis. Yield: 0.14 g (71%). ESI-MS (m/z): 527
(5%) [M+–Br], 445 (48%) [M+–2Br], 389 (100%) [M+–NiBr2 or
at 1658 (mC@O for benzophenone). The sample decomposed before
it could be analysed by X-ray crystallography. Repetition of the
reaction at 40 °C led to the precipitation of a virtually quantitative
amount of benzophenone that was identified on the basis of 1H
NMR and IR spectroscopy.
1a+H], 225 (20%) [1aꢁC13H12]. IR (KBr), (cmꢁ1):
m(C@N) = 1601.
4.3.4. Reaction between N,N0-bis-(2-tert-butylthio-1-
ylmethylenebenzene)-2,20diamino-biphenyl (2) and CuCl2ꢀ2H2O
A total of 0.56 g N,N0-bis-(2-tert-butylthio-1-ylmethyleneben-
zene)-2,20diamino-biphenyl was dissolved in 3 mL CH2Cl2 and
0.018 g CuCl2ꢀ2H2O in 1.5 ml of ethanol was added drop by drop.
The colour turned dark red immediately and small crystals precip-
4.4.4. Reaction between N,N0-bis(diphenylmethylene)-1,2-
propanediamine (Bz2pn, 1b) and [NiBr2(dme)]
To a suspension of [NiBr2(dme)] (0.10 g, 0.32 mmol) in 15 ml
CH2Cl2 was added a solution of Bz2pn (1b, 0.13 g, 0.32 mmol) in
15 ml CH2Cl2. The mixture turned from grey to purple color within

M. Czaun et al. / Inorganica Chimica Acta 363 (2010) 3102–3112
3111
10 min. The purple solution was stirred for 5 h under nitrogen and
the solvent was removed on a rotary evaporator to give a purple
powder. The powder was dissolved in chloroform and over a period
of 4 days gave purple crystals suitable for X-ray analysis. Yield:
0.16 g (76%). FAB-MS (m/z): 615 (11%) [M+ꢁ2], 539 (4%) [M+–Br],
403 (13%) [M+–NiBr2 or Bz2pn+H], 239 (100%) [Bz2pn–C13H12]. IR
imate dimensions, 0.1 ꢂ 0.1 ꢂ 0.02 mm3, and a blue-green needle
of 6, with approximate dimensions, 0.3 ꢂ 0.1 ꢂ 0.1 mm3 were
mounted in glass capillaries. Data were recorded for 3.5 < 2h <
50.7° (5) and 6.5 < 2h < 50° (6), using the
x/2h scan technique.
Monitoring of three reflections at intervals of 120 reflections mea-
sured indicated stability for both crystals. Correction was made for
Lorentz and polarization effects; no corrections were made for
absorption. For 5 a total of 14047 reflections were measured yield-
ing 4618 unique reflections (Rint = 0.069) and 2978 reflections for
(KBr), (cmꢁ1):
m(C@N) = 1604.
4.5. Kinetic investigation of the nickel-catalyzed C@N bond hydrolysis
in Bz2en (1a) and Bz2pn (1b)
which I > 2
yielding 2926 unique reflections (Rint = 0.040) and 2010 reflections
for which I > 2 (I)). Both structures were solved by direct methods
r(I)). For 6 a total of 5938 reflections were measured
The kinetics of the hydrolysis was followed by determining
the concentration of [Ni(Bz2en)Cl2] in the reaction solution as a
function of time by UV–Vis spectroscopy. A stock solution of
[Ni(Bz2en)Cl2] in dry CH3CN was prepared in a volumetric flask
and 3.0 mL of the stock solution was transferred into a 1 cm path
length quartz cell. Water was added directly into the cell and after
mixing, the cell was placed in the thermostated cell holder. The
rate constants were determined using conventional pseudo-first or-
der kinetics and were reproducible to within 2%.
r
(
SHELXS-97) and refined using full-matrix least-squares calculations
on F2 (SHELXL-97) [17] operating in the WINGX program package [18].
Anisotropic thermal displacement parameters were refined for all
the non-hydrogen atoms; hydrogen atoms were allowed to ride
on the appropriate carbon atom.
4.6.4. Compound 7
A
pale blue crystal of
7
with dimensions 0.39 ꢂ 0.29 ꢂ
0.23 mm3 was selected under oil under ambient conditions and at-
tached to the tip of a nylon loop. The crystal was mounted in a
stream of cold nitrogen at 100(2) K and centred in the X-ray beam
by using a video camera. The crystal evaluation and data collection
4.6. Crystallographic data collection and structure analyses
Crystallographic and refinement data for compounds 1a, 2, 5, 6
and 7 are summarised in Table 1.
were performed on a Bruker CCD-1000 diffractometer with Mo K
a
(k = 0.71073 Å) radiation and a diffractometer to crystal distance of
4.9 cm. Data were collected by using the full sphere data routine to
survey the reciprocal space to a resolution of 0.80 Å. A total of
29473 data were harvested by collecting six sets of frames with
4.6.1. Compound 1a
A colourless irregular-shaped crystal of 1a, with the approxi-
mate dimensions 0.30 ꢂ 0.25 ꢂ 0.20 mm3, was mounted in a glass
capillary and transferred to a Rigaku AFC6R diffractometer. Dif-
fracted intensities were measured at 293(2) °C, using graphite-
monochromated Cu K
rotating anode operated at 9 kW (50 kV; 180 mA). Data were re-
corded for 8 < 2h < 120°, using the /2h scan technique. Correction
was made for Lorentz and polarization effects; no correction was
made for absorption. Monitoring of three reflections at intervals
of 150 reflections measured indicated crystal stability throughout
the data collection. A total of 1683 reflections were measured
yielding 1600 unique reflections (Rint = 0.055) and 843 reflections
0.25° scans in
x with an exposure time 5 s per frame. These highly
redundant datasets were corrected for Lorentz and polarization ef-
fects. The absorption correction was based on fitting a function to
the empirical transmission surface as sampled by multiple equiva-
lent measurements [19]. The structure was solved by direct meth-
ods. All non-hydrogen atoms were refined with anisotropic
displacement coefficients, and hydrogen atoms were allowed to
ride on the neighbouring atoms with relative isotropic displace-
ment coefficients [19]. The Ni complex resides on a crystallo-
graphic inversion centre. There are two solvate molecules of
chloroform per Ni complex in the lattice.
a (k = 1.54178 Å) radiation from a RU200
x
for which I > 2r(I)). The structure was solved by direct methods
(
SHELXS-97) and refined using full-matrix least-squares calculations
on F2
(SHELXL-97) [17] operating in the WINGX programme package
4.6.5. Compound 8
[18]. Anisotropic thermal displacement parameters were refined
for all the non-hydrogen atoms; hydrogen atoms were allowed to
ride on the appropriate carbon atom.
The identity of 8 was established by means of a preliminary
crystal-structure determination, the quality of the crystal being
insufficient to permit otherwise. A purple fragment, with the
approximate dimensions 0.3 ꢂ 0.3 ꢂ 0.3 mm3, was cut from a larger
crystal that had formed in an NMR tube, resulting in disintegration
of the rest of the crystalline material. The results have been depos-
ited with the CCDC (see below) and only a brief description is given
here. [Ni(Bzpn)Br2], C16H18Br2N2Ni, (8), Mw = 456.85, crystallises
4.6.2. Compound 2
A pale yellow plate shaped crystal with the dimensions 0.25 ꢂ
0.25 ꢂ 0.1 mm3 was used for the data collection at 293(2) °C with
a Bruker Smart CCD system using
and a rotating anode with Mo K
x
a
-scans, ꢁ0.3° and 20 s per frame,
ꢀ
radiation (k = 0.71073 Å). The
in space group P1 (No. 2), with a = 9.6476(5), b = 10.086(11),
detector distance was set to 4.0 cm. The first 50 frames were recol-
lected at the end of the data collection to check for decay, none being
observed. All reflections were merged and integrated using SAINT
(Bruker AXS, 1995) and the intensities were corrected for Lorentz
and polarisation effects. A total of 22105 reflections were measured
yielding 9090 unique reflections (Rint = 0.050) and 4128 reflections
c = 11.894(15) Å,
a
= 89.28(15)°, b = 68.00(10)°,
c
a
= 70.25(12)°, V =
1001.3(2) Å3, Z = 2, Dc = 1.515 g cmꢁ3, and
l(Mo K
) = 4.955 mmꢁ1
.
The crystal fragment was mounted in a glass capillary and trans-
ferred to a Rigaku R-AXIS IIc image plate system. Diffracted intensi-
ties were measured at ambient temperature, 293(2) K, using
graphite-monochromated Mo K
a radiation (k = 0.71073 Å) from a
with I > 2
97) and refined using full matrix least squares calculations on F2
SHELXL-97). Anisotropic thermal displacement parameters were re-
r
(I). The structure was solved by direct methods (SHELXS
-
RU-H3R rotating anode. Ninety oscillation photographs with a rota-
tion angle of 2° were collected and processed using the CRYSTALCLEAR
software package. An empirical correction was applied for the ef-
(
fined for all the non-hydrogen atoms.
fects of absorption (l
= 4.955 mmꢁ1) using the REQAB program
under CRYSTALCLEAR. The structure was solved and refined as for 1a.
4.6.3. Compounds 5 and 6
Data/restraints/parameters = 3529/0/191. Final R: 0.082 (wR2 =
Diffraction data for compounds 5 and 6 were measured at
293(2) K using an Enraf–Nonius CAD-4 diffractometer with Mo
0.17) based on 1505 reflections for which Io > 2r(Io); R = 0.103
(wR2 = 0.18) for all 3529 reflections. Max.; min. residual electron
Ka
(k = 0.71073 Å) radiation. A blue-green flake of 5, with approx-
density: 1.0; ꢁ1.0 e Åꢁ3
.

3112
M. Czaun et al. / Inorganica Chimica Acta 363 (2010) 3102–3112
[3] M.D. Shelley, L. Hartley, R.G. Fish, P. Groundwater, J.J.G. Morgan, D. Mort, M.
Acknowledgements
Mason, A. Evans, Cancer Lett. 135 (1999) 171.
[4] A.Th. Chaviara, P.J. Cox, K.H. Repana, A.A. Pantazaki, K.T. Papazisis, A.H.
Kartsaris, D.A. Kyriakidis, G.St. Nikolov, C.A. Bolos, J. Inorg. Biochem. 99 (2005)
467.
[5] A.K. Singh, N. Majumdar, J. Photochem. Photobiol. B: Biol. 39 (1997) 140.
[6] X.R. Bu, C.R. Jackson, D. Van Derveer, X.Z. You, Q.J. Meng, R.X. Wang,
Polyhedron 16 (1997) 2991.
[7] M. Sinha Ray, R. Bhattacharya, S. Chaudhuri, L. Righi, G. Bocelli, G.
Mukhopadhyay, A. Ghosh, Polyhedron 22 (2003) 617.
[8] A.C. Braithwaite, C.F.E. Rickard, T.N. Walters, J. Chem. Soc., Dalton Trans. (1975)
2149.
We thank Prof. Åke Oskarsson for assistance with the crystallo-
graphic data collection for and structure solution of 2, and we are
indebted to the late Prof. Hans Toftlund for recording the EPR spec-
trum of 5. This research has been supported by the Swedish Re-
search Council (VR), the European Union (Marie Curie Training
Site Metals in Biological Systems (MEBIOS)) and the Swedish Inter-
national Development Agency (SIDA)/National Research Founda-
tion (NRF, South Africa) via the Swedish Research Links program.
[9] G. Lisensky, ‘‘Chemistry Software.” iORTEP
– A Crystallographic Plotting
Program. Beloit College <http://chemistry.beloit.edu/classes/programs/>.
[10] S.M. Nelana, J. Cloete, G. Lisensky, E. Nordlander, I.A. Guzei, S.F. Mapolie, J.
Darkwa, J. Mol. Catal. A 285 (2008) 72.
[11] S. Chowdhury, G.K. Patra, M.G.B. Drew, N. Chattopadhyay, D. Datta, J. Chem.
Soc., Dalton Trans. (2000) 235.
[12] D.A. Harvey, C.J.L. Lock, Acta Crystallogr., Sect. C 42 (1986) 799.
[13] (a) J. Bergman, A. Brynolf, Tetrahedron Lett. 30 (1989) 2979;
(b) A.E. Frost, H.H. Freedman, J. Org. Chem. 24 (1959) 1905.
[14] E. Rule, I.M. Han, C.J. Elsevier, K. Vrieze, P.W.N.M. Van Leeuwen, C.F. Roobeek,
M.C. Zoutberg, Y.F. Wand, C.H. Stam, Inorg. Chim. Acta 169 (1990) 5.
[15] Y. El-Ahmad, P. Reynaud, J. Het. Chem. 25 (1988) 711.
Appendix A. Supplementary material
CCDC 773044, 773045, 773046, 773047, 773048 and 773049
contains the supplementary crystallographic data for 1a, 2, 5, 6, 7
and 8. These data can be obtained free of charge from The Cam-
bridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif. Supplementary data associated with this article
can be found, in the online version, at doi:10.1016/j.ica.2010.
05.025.
[16] S. Flanagan, J. Dong, K. Haller, S. Wang, W.R. Scheidt, R.A. Scott, T.R. Webb, D.M.
Stanbury, L.J. Wilson, J. Am. Chem. Soc. 119 (1997) 8857.
[17] G.M. Sheldrick, SHELX97 (Release 97-2). Institut für Anorganische Chemie,
Tammanstrasse 4, D-3400 Göttingen, Germany, 1998.
[18] L.J. Farrugia, J. Appl. Crystallogr. 32 (1999) 837.
[19] Bruker-AXS. (2000–2003) SADABS V.2.05, SAINT V.6.22, SHELXTL V.6.10 and SMART
5.622, Software Reference Manuals. Bruker-AXS, Madison, Wisconsin, USA.
References
[1] K. Kurita, Prog. Polym. Sci. 26 (2001) 1921.
[2] W.-P. Deng, K.A. Wong, K.L. Kirk, Tetrahedron Asymm. 13 (2002) 1135.