Binding multidentate ligands to Ni2+: Kinetic identification of preferential binding sites

Article abstract of DOI:10.1039/c3dt53359j

The kinetics of the reactions between [Ni(MeOH)₆]²⁺ (hereafter Ni²⁺) and a variety of neutral Schiff base multidentate ligands have been measured in methanol at 25.0°C using stopped-flow spectrophotometry. The ligands cont

Full text of DOI:10.1039/c3dt53359j

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Binding multidentate ligands to Ni2+: kinetic
identification of preferential binding sites†
Cite this: Dalton Trans., 2014, 43,
3372
Dilek Nartop,*a William Clegg,b Ross W. Harrington,b Richard A. Henderson*b and
Corinne Y. Willsb
The kinetics of the reactions between [Ni(MeOH)6]2+ (hereafter Ni2+) and a variety of neutral Schiff base
multidentate ligands have been measured in methanol at 25.0 °C using stopped-flow spectrophotometry.
The ligands contain a variety of different potential donor sites (phenolic OH, imine N, pyridyl N and NH
groups), different structural components and substituents. The kinetic studies explore how systematic
changes to the composition of the ligands affect the rates of binding. The results are consistent with the
Eigen–Wilkins mechanism in which the ligand initially forms an outer-sphere association with Ni2+ prior
to dissociation of a coordinated solvent molecule and binding to the metal ion. The general features that
emerge from these studies are as follows. (i) For ligands with the same donor set, the rates of binding are
all similar irrespective of changes to the ligand framework (bridge and substituents). (ii) Comparison of
structurally analogous ligands shows that the presence of pyridyl or NH groups in the multidentate results
in significantly faster reactions. (iii) With ligands containing multiple NH groups, the rate of ligand binding
increases as the number of NH groups increases. The extent to which these kinetic features can be attri-
buted to preferential binding of particular donor groups is discussed.
Received 28th November 2013,
Accepted 17th December 2013
DOI: 10.1039/c3dt53359j
www.rsc.org/dalton
Introduction
The kinetics of the reactions of multidentate ligands binding
to metal ions can obviously be more complicated than the
reactions of unidentate ligands because of the multi-stage
time courses associated with coordination of the large number
of sites.1,2 An intriguing aspect of the reaction mechanisms
with multidentate ligands containing a variety of different
types of donor groups is whether the metal discriminates
between the various donors and preferentially binds to one
type of site. Recent studies have explored the binding of Cu2+
Fig. 1 Structures of polytopic ligands phenL and pyL and representation
to the polytopic ligands shown in Fig. 1 and observed different of the pathway for incorporation of Cu2+ into the macrocyclic cavity for
phenL.
types of reactivity.3,4 For the phenanthroline-based ligand
(
phenL), Cu2+ initially binds to the phenanthroline residue and
subsequently moves into a macrocyclic cavity. In contrast, for
the analogous pyridyl-based ligand, pyL, the pyridyl-residue
appears to be just a spectator, with the Cu2+ going directly to
the macrocyclic cavity. Similar behaviour, in which a metal ion
initially binds at one site before moving to its final residence,
has been observed with other polytopic ligands.5,6
Important questions which need to be addressed are
whether binding of metal ions to multidentate ligands (con-
taining various types of donors) involves preferential binding
at a specific site, and whether this can be detected using kine-
tics. In this paper we report kinetic studies on the reactions of
[Ni(MeOH)6]2+ (hereafter written as Ni2+) with a variety of
neutral multidentate Schiff base ligands (Fig. 2) and investi-
gate if the differences in rates can be attributed to preferential
aNevşehir University, Nevşehir, Turkey. E-mail: dileknartop@nevsehir.edu.tr;
Fax: +90 384 215 39 48; Tel: +90 384 215 39 00
bNewcastle University, Newcastle upon Tyne, United Kingdom.
E-mail: richard.henderson@ncl.ac.uk; Fax: +44 (0)191 222 6929;
Tel: +44 (0)191 222 6636
†Electronic supplementary information (ESI) available: Tables of spectroscopic
and kinetic data; full crystallographic details in CIF format. CCDC
959697–959699. For ESI and crystallographic data in CIF or other electronic
format see DOI: 10.1039/c3dt53359j
binding of different donors to Ni2+
.
3372 | Dalton Trans., 2014, 43, 3372–3382
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Fig. 3 Synthesis of unsymmetric diimines (H2LH–Cl, H2LMe–Cl,
H2LCl–Cl).
Synthesis of unsymmetrical Schiff bases H2LX–Cl (X = H, Me or Cl):
general procedure
The Schiff bases HLH (or HLMe or HLCl) were synthesized by
adding 50 mmol of 2-aminophenol (or 2-amino-4-methyl-
phenol or 2-amino-4-chlorophenol) to a stirred solution of
1-nitro-2-naphthaldehyde (50 mmol) in ethanol (100 mL) and
heating for 1 hour at 60 °C (Fig. 3). The unsymmetrical Schiff
bases (H2LH–Cl, H2LMe–Cl, H2LCl–Cl) were prepared by using a
two-stage method, as shown in Fig. 3. In the first stage,
2 mmol of the Schiff base HLH (or HLMe or HLCl) was dissolved
in 100 mL ethanol–water solution (1 : 1) at 70 °C. The nitro
group of these Schiff bases was reduced to an amino group
with solid sodium dithionite as reducing agent. In this
reduction process, 5 mmol of solid sodium dithionite was
slowly added to the solution in small portions over the course
of 1 hour, then the solution was stirred for a further 1 hour at
45–50 °C. In this way the amino derivative of the Schiff bases
(A) was obtained in solution. In the second stage, 2 mmol of
2-hydroxy-5-chlorobenzaldehyde in 25 mL ethanol was added
to the solution of the amino derivative and was heated to
reflux for 2 hours. The mixture was evaporated at room temp-
erature in about 1 day. The orange product was treated with
warm water and filtered. The crude product was recrystallized
from ethanol. The elemental analyses and mass spectral data
Fig. 2 Structures of the ligands used in this study and the abbreviations
used.
Schiff bases and their metal complexes have been studied
widely due to their important chemical and biological
applications.7–11 Recently, unsymmetrical Schiff base ligands
(H2LaX) were reported, and herein the similar ligands H2LX–Cl
are reported.12,13
H2LaX, H2LX–Cl, H2LXn and Lpny have structures in which there
are different types of binding sites: imine nitrogen, terminal
phenolic, terminal pyridyl and NH groups. In this paper we
present kinetic studies which indicate that the quadridentate
ligands H2LaX, H2LX–Cl, H2salphen and H2L0X all have very
similar reactivities with only minor modulation of the
reactivity which is attributable to subtle changes in the outer-
sphere association of the ligand with Ni2+, prior to binding.
However, for Lnpy (where a pyridyl formally replaces a phenol
group of H2LnH) binding is about 103 times faster. Furthermore,
introduction of NH groups into either the pyridyl- or salicaldi-
mine-based ligands results in an increase in the rate. The
origin of these labilising effect has been explored.
1
for H2LX–Cl are presented in Table 1 and the H NMR and IR
spectra of the compounds are presented in Table 2.14
Synthesis of unsymmetrical Schiff bases H2LaX (X = H, Me or Cl):
general procedure
Experimental
One of us recently reported the synthesis and characterization
All chemicals were purchased from Aldrich or Merck and used of the unsymmetrical Schiff bases, H2LaX, together with some
without further purification. Elemental analyses were deter- of their metal complexes.11 The unsymmetrical Schiff bases
mined using a Leco CHNS-932 analyzer. IR spectra were were prepared by the method described in the literature and
recorded using a Mattson-5000 FT-IR instrument with KBr shown in Fig. 4. Initially, 2-hydroxy-N-(2-nitrobenzylidene)-
pellets. UV-visible absorption spectra were recorded on a aniline was obtained by reacting 2-hydroxyaniline (50 mmol)
Perkin Elmer spectrophotometer. 1H-NMR spectra were carried with 2-nitro-benzaldehyde (50 mmol) in ethanol (100 mL) as
out on a Bruker Avance 500 MHz or a Bruker Avance 300 MHz described previously.15,16 Subsequently, the nitro group of (B)
instrument. MS analysis was performed using an Agilent was reduced to an amino group using sodium dithionite
Technologies 6410 Triple Quad instrument.
(5 mmol) as reducing agent, in solution. Finally, 1 mmol of
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 3372–3382 | 3373

View Article Online
Paper
Dalton Transactions
Table 1 Elemental analyses and mass spectral data for H2LX–Cl
Elemental analysis (calc.) %
Mass spectrum
Compounda
Formula
C
N
H
m/z
%
Peak
H2LH–Cl
H2LMe–Cl
H2LCl–Cl
C24H17N2O2Cl
C25H19N2O2Cl
C24H16N2O2Cl2
70.05 (71.91)
70.24 (72.38)
65.13 (66.21)
4.22 (4.24)
4.07 (4.58)
4.39 (3.68)
6.37 (6.99)
6.39 (6.76)
6.85 (6.44)
399.2
413.2
433.0
7.9
22.1
18.4
[M]+
[M − H]+
[M − 2H]+
a All compounds are orange.
Table 2 The IR vibration frequencies (cm−1) and 1H-NMR chemical shift (ppm) of the unsymmetrical diimines H2LX–Cla
Compound
IR
1H-NMR
ν(OH)
3411
3406
3367
ν(CH)arom.
3066
ν(CvN)
1624, 1611
1611, n.o.
1608, 1589
ν(CvC)ring
1589–1454
1546–1462
1547–1461
OH
CHvN
Arom.-H
CH3
—
H2LH–Cl
H2LMe–Cl
H2LCl–Cl
13.81 (d) 1H
10.23 (s) 1H
13.87 (s) 1H
9.90 (s) 1H
13.39 (s) 1H
10.07 (s) 1H
9.85 (d) 1H
8.96 (d) 1H
9.57 (s) 1H
8.31 (s) 1H
9.85 (s) 1H
8.97 (s) 1H
6.96–8.41 (m)
6.98–8.41 (m)
6.57–8.45 (m)
3044
2.50 (s)
—
3056
a s = singlet, d = doublet, m = multiplet, n.o. = not observed.
Synthesis of NiII complexes with unsymmetrical diimines H2LXa
and symmetrical diimines H2LnH: general procedure
All the complexes were prepared by the same general pro-
cedure. A solution of the Schiff base ligand (5.0 mmol) in
methanol (30 mL) was prepared in a round-bottom flask. A
solution of NiCl2·6H2O (5 mmol) in methanol (30 mL) was
then added to the ligand solution. The mixture immediately
became a brown/orange colour. The mixture was stirred and
heated for about 1 hour, maintaining the temperature at less
than 75 °C. The solution was then allowed to cool. Volatiles were
removed in vacuo until incipient crystallization. The solid was
removed by filtration, washed with diethyl ether and dried in air.
The solid was recrystallized by dissolving in the minimum
volume of methanol, then layering diethyl ether (about 4–5
times the volume of methanol) on top. The purified products
were all dark orange/brown complexes. The spectroscopic
characterization of these complexes is presented in the ESI.†
Fig. 4 Synthesis of unsymmetrical diimines (H2LaH, H2LaMe, H2LCa l).
2-hydroxybenzaldehyde (or 2-hydroxy-5-methylbenzaldehyde or
2-hydroxy-5-chlorobenzaldehyde) in ethanol (25 mL) was
added to the solution to obtain the unsymmetrical diimines,
H2LaH, H2LaMe or H2LCa l. Spectroscopic data for all H2LaX have
already been reported.11
X-Ray crystallography
Data for [NiCl(Lpy)]Cl were collected on a Bruker SMART 1 K
diffractometer, those for [Ni(HL1H)]Cl·CH3OH and [{NiCl(Lp1y)}2]-
Cl2·2CH3OH·2H2O on an Oxford Diffraction (now Agilent
Synthesis of symmetrical Schiff bases H2LnH (n = 0–3): general
procedure
All H2LnH were prepared by the same general procedure. A solu- Technologies) Gemini A Ultra diffractometer. All measure-
tion of salicaldehyde (4.25 mL; 40 mmol) in ethanol (50 mL) ments were made at 150 K with Mo Kα radiation (λ =
was heated to boiling. The polyamine (20 mmol) was added 0.71073 Å). Semi-empirical absorption corrections were
dropwise to the boiling, stirred solution. The solution turned applied, based on repeated and symmetry-equivalent reflec-
bright yellow and was stirred for a further 5 minutes then left tions. The structures were solved by direct methods, and
to cool. Upon cooling yellow crystals of the product formed. refined on all F2 values. H atoms bonded to ordered hetero-
The crystals were removed by filtration, washed with diethyl atoms were refined freely; a riding model was used for other
ether and dried in air. Yields of the compounds were all in the H atoms. Crystal and refinement data are presented in Table 3.
range 83–92%. The spectroscopic characterization of these Disorder was resolved for the unique methanol solvent mole-
ligands is presented in the ESI.†
cule in [{NiCl(L1py)}2]Cl2·2CH3OH·2H2O. Diffraction for the
3374 | Dalton Trans., 2014, 43, 3372–3382
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Table 3 X-ray crystallographic data
Compound
[Ni(HL1H)]Cl
18H19ClN3NiO2
[NiCl(Lpy)]Cl·CH3OH
[{NiCl(L1py)}2]Cl2·2CH3OH·2H2O
Formula
C
C17H27Cl2N5NiO
447.0
C34H50Cl4N10Ni2O4
922.1
Mr/g mol−1
403.5
Crystal system
Space group
Monoclinic
P21/c
Monoclinic
P21/c
Triclinic
P1
ˉ
a/Å
b/Å
c/Å
α/°
10.8103(12)
14.5972(12)
12.1913(14)
90
114.061(13)
90
1756.6(4)
4
1.530
13.9076(7)
11.3295(5)
14.1620(7)
90
117.8704(8)
90
1972.62(16)
4
1.505
9.2913(5)
10.2003(5)
12.0495(7)
83.850(4)
67.556(5)
82.657(4)
1044.70(10)
1
β/°
γ/°
V/Å3
Z
Dcalc/g cm−3
1.466
1.21
μ/mm−1
1.27
1.27
Min., max. transmission
Reflections measured
Unique reflections, Rint
Refined parameters
R (F, F2 > 2σ)
0.890, 0.950
5795
3100, 0.0613
234
0.0856
0.2425
0.700, 0.800
11 514
4799, 0.0335
252
0.0348
0.0776
0.804, 0.890
10 097
4900, 0.0221
266
0.0236
0.0609
Rw (F2, all data)
Goodness of fit on F2
Max., min. electron density/e Å−3
1.022
2.10, −0.84
1.063
0.38, −0.37
1.053
0.32, −0.36
small and poorly formed crystal of [Ni(HL1H)]Cl was weak,
especially at higher angles; significant residual electron
density is found close to the Ni atom in this structure, and the
R values are relatively high, but the refinement is otherwise
satisfactory. Programs used were standard data collection and
processing software, SHELXTL and SHELXL-2013.17
Kinetics studies
All kinetic studies were conducted using an Applied Photo-
physics SX.18MV stopped-flow spectrophotometer. The temp-
erature was maintained at 25.0 0.1 °C using a Grant LTD6G
recirculating thermostated tank. All solutions were prepared in
air and were transferred into the stopped-flow spectrophoto-
meter using all-glass syringes. The solutions of NiCl2·6H2O
and the various Schiff base ligands were prepared from fresh
stock solutions and used within 2 hours.
All kinetic studies were performed in methanol, with the
concentration of Ni2+ = 0.2 or 0.5 mmol dm−3. Because of
limited solubility, the concentrations of the salicaldimine-
based ligands were in the range 1–4 mmol dm−3. The reactions
of Ni2+ with the various Schiff base ligands were studied at λ =
Fig. 5 (Main) Plot of kobs versus concentration of H2LaCl for the first
phase of the reaction between H2LaCl and Ni2+ in methanol at 25.0 °C.
(Insert) Stopped-flow absorbance–time trace (black curve) for the reac-
tion of H2LaCl (4 mmol dm−3) with Ni2+ (0.5 mmol dm−3) in methanol at
25.0 °C, λ = 460 nm. The curve is fitted to the expression At = 1.90 −
0.24e−1.93t − 0.24e−0.22t (grey curve).
460 nm. At this wavelength there is an increase in absorbance, H2LnH (n = 1–3), the absorbance–time traces could be fitted to
as shown in Fig. 5 (insert). The reactions of Ni2+ with Lpy, Lpny two exponentials. Analysis of the same data applying second
and H2LnH (n = 1–3) were studied under pseudo first order con- order kinetics18 also showed biphasic behaviour with rate con-
ditions. However, the reactions with H2LaX, H2LX–Cl, H2LX0 and stants for both phases in good agreement with those derived
H2salphen were all studied using concentrations which are not using exponential curve fits (examples shown in ESI†). (3) In
strict pseudo first-order conditions so the suitability of expo- selected studies with [Ni2+] = 0.2 mmol dm−3 and [ligand] =
nential curve fits to the data needs to be justified. There are 2–4 mmol dm−3 (where [ligand]/[Ni2+] ≥ 10) the exponential
three lines of evidence which validate the use of exponential curve fits gave observed rate constants in good agreement with
fits for the reactions with H2LaX, H2LX–Cl, H2LX0 and H2salphen. those determined when [Ni2+] = 0.5 mmol dm−3 (ESI†). These
(1) For the reactions of H2LaH, H2LX–Cl (X = H or Me), H2LX0 observations indicate it is appropriate to fit the absorbance–
(X = H, MeO, Me or Cl) and H2salphen the traces were good time traces with exponential curves to obtain the observed
fits to single exponential curves at all concentrations (examples first-order rate constants (kobs). In the reactions with H2LaX,
shown in ESI†). (2) For H2LaX (X = Me or Cl), H2LCl–Cl, and H2LX–Cl, H2LXn and H2salphen, both the absorbance change
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 3372–3382 | 3375

View Article Online
Paper
Dalton Transactions
and the final absorbance increase with increasing concen-
tration of the ligand. This is a consequence of both the ligand
absorbing at λ = 460 nm and the binding of these ligands to
Ni2+ being equilibrium reactions. For the reactions with Lpy
and Lnpy, the absorbance change is independent of the concen-
tration of ligand, but the final absorbance increases with
increase in the concentration of ligand. This behaviour is
because the ligands absorbs at λ = 460 nm but binding of Lpy
or Lnpy to Ni2+ are not equilibrium reactions (spectroscopic
changes are presented in ESI†). That the binding of Ni2+ to
H2LaX, H2LX–Cl, H2LX0 and H2salphen are equilibrium reactions
suggests the reason why the absorbance–time curves are good
fits to exponential curves. In equilibrium reactions, the con-
centrations of Ni2+ and ligand that are added to the solution
are not good indicators of the concentrations which are react-
ing. Using the values of the equilibrium constants for the reac-
tions, Ka = ka/k−a, derived from the kinetic studies (Table 5), we
have calculated for solutions where added concentrations are
[ligand] = 1 mmol dm−3, [Ni2+] = 0.5 mmol dm−3, that the con-
centration of ligand is 4–40 times larger than the ‘reacting con-
centration’ of Ni2+ (ESI†).
The dependences of kobs on the concentrations of the
ligands were determined from plots of kobs versus concen-
tration of Schiff base. For H2LaH, H2LH–Cl, H2LMe–Cl, H2L0X and
H2salphen (traces fit single exponentials), such plots were
straight line graphs with a positive intercept, as shown by the
example in Fig. 5 (main). For the reactions with H2LaMe, H2LCa l,
H2LCl–Cl and H2LnH (traces fit two exponentials), the values of
kobs, for the fast phase, also showed a linear dependence on
the concentration of ligand with a positive intercept. For the
slow phase, the values of kobs were independent of the concen-
tration of ligand. The values of kobs presented in Tables 5 and
6 are the average of at least three duplicate experiments where
the values of kobs all agree within 10%, and this is reflected in
the error bars presented in Fig. 5 and 10.
Fig. 6 Suggested structures of Ni complexes, NiLaMe (left hand side) and
Ni2(LaX)2 (X = H or Cl) (right hand side).
Suggested structures of the complexes NiLaH, NiLMa e and NiLCa l
are shown in Fig. 6.
The further diversity of the structures of the products of the
reactions between Ni2+ and the multidentate ligands is illus-
trated by the X-ray crystal structures for which the cations are
shown in Fig. 7; selected bond lengths and angles describing
the metal coordination geometry are given in Table 4. Although
H2L1H, L1py and Lpy all have five potential donor atoms, only with
Lpy do all donor atoms bind to a single Ni. With H2L1H, and Lp1y
the inflexibility of the ligand results in a dimeric structure
([{NiCl(L1py)}2]Cl2) or a mononuclear species with a pendant arm
([Ni(HL1H)]Cl). The product with Lpy is a mononuclear complex
containing an octahedral Ni in [Ni(Lpy)Cl]+, which is coordinated
by the five donor nitrogens of Lpy and a chloro ligand. In con-
trast, the Ni complex with L1py is
a dimeric dication
{[Ni2(L1py)2Cl2]2+}. The coordination sphere of each Ni is identical.
Each Ni is octahedral, with one L1py acting as a tridentate ligand
and the other acting as a bidentate. The final coordination site
on each Ni is occupied by a chloro ligand. It is impossible for all
five nitrogen donors in L1py to coordinate to a single octahedral
Ni because it is only the central NH group which is capable of
subtending two chelate arms which are 90° to one another.
H2L1H is also too inflexible to be pentadentate to a single Ni.
However, in this case, the geometry at the Ni is square planar.
The four donors to the Ni are one oxygen of a phenolate residue,
the two imine nitrogens and the NH group. The remaining
phenol residue is uncoordinated and remains protonated.
All the structures are of chloride salts. Two of them are sol-
vates, both containing methanol and one of them also water.
The solvent molecules and chloride anions are involved in
hydrogen bonding, with most N–H groups of the ligands also
acting as donors. Full details available in ESI.†
Results and discussion
Products of the reactions of the ligands with Ni2+
Not all the products from the reactions of Ni2+ with the various
ligands shown in Fig. 2 have been isolated. Those that have
been isolated show the structure of the product depends on
the composition of the ligand.
The structure of [Ni(L0H)] (i.e. [Ni(salen)]) has been known
for some time and is square planar with both oxygen and both
We have isolated the products of the reactions of Ni2+ with
L1H, L2H or L3H. Attempts to determine the X-ray crystal structures
for the products of Ni2+ with LH2 and L3H were unsuccessful, but
the 1H NMR spectra of the isolated compounds (ESI†) show
sharp resonances in similar positions to those for [Ni(HL1H)]+
indicating that in these systems similar square-planar com-
plexes are formed, presumably with pendant groups. However,
the 1H NMR spectra also show some broad resonances (par-
ticularly for the product with L3H), indicating paramagnetic pro-
ducts (either octahedral or tetrahedral) are also present,
possibly in equilibrium. It is worth noting that the kinetics of
nitrogen donors coordinated.21 The complexes NiLHa , NiLMa e
,
NiLaCl have previously been characterized by spectroscopic tech-
niques. All of the H2LaX are potential N2O2-type quadridentate
ligands and the room-temperature magnetic properties of the
reaction products of H2LaH and H2LCa l with NiII salts indicate
the complexes are binuclear with Ni–Ni interactions.11 The
dimeric nature of these complexes was further indicated by
mass spectrometry, where the molecular ion was observed at
the position corresponding to a dimer. On the other hand, the
analysis for the H2LaMe product indicates it is mononuclear.12
3376 | Dalton Trans., 2014, 43, 3372–3382
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Table 4 Selected bond lengths (Å) and angles (°) at the Ni atoms in
[Ni(HL1H)]+, [NiCl(Lpy)]+ and [{NiCl(L1py)}2]2+
[Ni(HL1H)]+ (pendant arm)
Ni–N(1)
Ni–N(2)
Ni–N(3)
Ni–O(2)
1.891(6)
1.883(6)
1.837(6)
1.823(4)
N(1)–Ni–N(2)
N(1)–Ni–O(2)
N(2)–Ni–N(3)
N(3)–Ni–O(2)
N(1)–Ni–N(3)
N(2)–Ni–O(2)
86.2(3)
91.6(2)
86.4(3)
96.0(2)
171.4(3)
177.3(2)
[NiCl(Lpy)]+ (mononuclear)
Ni–Cl(1)
Ni–N(1)
Ni–N(2)
Ni–N(3)
Ni–N(4)
Ni–N(5)
2.4634(6)
Cl(1)–Ni–N(1)
Cl(1)–Ni–N(2)
Cl(1)–Ni–N(3)
Cl(1)–Ni–N(4)
Cl(1)–Ni–N(5)
N(1)–Ni–N(2)
N(1)–Ni–N(3)
N(1)–Ni–N(4)
N(1)–Ni–N(5)
N(2)–Ni–N(3)
N(2)–Ni–N(4)
N(2)–Ni–N(5)
N(3)–Ni–N(4)
N(3)–Ni–N(5)
N(4)–Ni–N(5)
91.73(5)
91.26(5)
95.06(5)
171.51(5)
91.82(5)
79.23(7)
160.01(7)
92.56(7)
100.96(7)
81.85(7)
96.74(7)
176.91(7)
83.31(7)
97.60(7)
80.17(7)
2.0948(17)
2.0714(18)
2.1181(19)
2.1277(18)
2.0956(17)
[{NiCl(L1py)}2]2+ (dimeric)
Ni–Cl(1)
Ni–N(1)
Ni–N(2)
Ni–N(3)
Ni–N(4′)
Ni–N(5′)
2.4222(4)
Cl(1)–Ni–N(1)
Cl(1)–Ni–N(2)
Cl(1)–Ni–N(3)
Cl(1)–Ni–N(4′)
Cl(1)–Ni–N(5′)
N(1)–Ni–N(2)
N(1)–Ni–N(3)
N(1)–Ni–N(4′)
N(1)–Ni–N(5′)
N(2)–Ni–N(3)
N(2)–Ni–N(4′)
N(2)–Ni–N(5′)
N(3)–Ni–N(4′)
N(3)–Ni–N(5′)
N(4′)–Ni–N(5′)
90.02(3)
94.64(3)
85.76(3)
94.25(3)
172.57(3)
78.62(4)
159.78(4)
99.77(4)
89.03(4)
82.02(4)
170.96(4)
92.39(4)
100.26(4)
97.60(4)
78.66(5)
2.1014(11)
2.0020(11)
2.1293(11)
2.0726(11)
2.1226(11)
spectroscopic changes and the kinetics of these reactions
(Experimental section) indicate that binding of these ligands
are equilibrium reactions. Furthermore, the reactions of H2LXa
(X = Me or Cl), and H2LCl–Cl with Ni2+ are associated with two
phases in which the second phase occurs at a rate indepen-
dent of the concentration of ligand. For the reactions of H2LaH,
H2LX–Cl (X = H or Me), H2salphen and H2L0X (X = H, Me, MeO
or Cl), and the fast phase of the reactions with H2LaX (X = Me
or Cl) and H2LCl–Cl the rate law is that shown in eqn (1). The
rate law associated with the second phase of the reactions with
H2LaX (X = Me or Cl) and H2LCl–Cl is shown in eqn (2). The
values of ka, k−a and kb are summarised in Table 5. For the
reactions of all the ligands presented in Table 5, ka is the rate
constant for the binding and k−a is the rate constant for the
dissociation of the ligand with Ni2+.
Fig. 7 Crystallographically determined structures of the cations [Ni-
(HL1H)]+, [NiCl(Lpy)]+ and [{NiCl(L1py)}2]2+
. Displacement ellipsoids are
shown at the 40% probability level, and selected atoms are labelled;
primes indicate atoms generated by inversion symmetry in the centro-
symmetric dimeric cation.
the reactions between all the ligands and Ni2+ do not reflect
the composition (nuclearity) of the product.
Kinetics of the reactions of Ni2+ with H2LXa, H2LX–Cl,
Rate ¼ fkꢀa þ ka½ligandꢁg½Ni2þꢁ
Rate ¼ kb½Ni2þꢁ
ð1Þ
ð2Þ
H2salphen or H2L0X
The kinetics of the reactions between Ni2+ and H2LXa, H2LX–Cl,
H2salphen or H2L0X have been studied in methanol. Both the
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 3372–3382 | 3377

View Article Online
Paper
Dalton Transactions
Table 5 Summary of the rate constants in the reactions between Ni2+
and Schiff base ligands measured in methanol at 25.0 °C
Ligand
k−a/s−1
ka/dm3 mol−1 s−1
kb/s−1
Unsymmetrical Schiff bases
H2LaH
0.022 0.005
0.35 0.02
0.43 0.02
0.11 0.02
0.013 0.005
0.22 0.02
19.0 1.0
87.5 10
350 30
5.0 0.3
19.3 1.0
32.5 2.0
—
H2LaMe
0.06 0.01
0.20 0.05
—
H2LaCl
H2LH–Cl
H2LMe–Cl
H2LCl–Cl
—
Fig. 8 The mechanism for the reactions of multidentate ligands with
Ni2+ in methanol (top line). Also included is the conjugate base pathway
(bottom line) observed in the presence of mixtures of NHEt3+ and NEt3.
0.04 0.01
H2L0H
6.5 0.5
5.5 0.5
2.0 0.2
12.0 1.0
4.0 0.5
—
—
—
—
—
Rate ¼ KLk2½Lꢁ½Ni2þꢁ
ð3Þ
ð4Þ
KLk2k3½Lꢁ½Ni2þꢁ
Rate ¼
k
ꢀ2 þ k3
The kinetic data presented in Table 5 relate to the reactions
of both conventional, symmetric types (H2L0X and H2salphen)
and asymmetric types (H2LXa and H2LX–Cl) salicaldimine-based
It is not clear if the observed reversible reactions for the
ligands in Table 5 correspond to the complete attachment of
the ligand to the metal, or just partial attachment. Although
all the ligands are quadridentates, in no case do we observe a
four-step time course. The rate constants presented in Table 5
relate to the first phase, or first and second phases of the wrap-
ping. However, from these kinetics alone, it is not possible to
establish if the first phase {eqn (1)} corresponds to the binding
of the first donor of the ligand or to the first chelation of the
ligand (vide infra). The second phase {eqn (2)} is independent
of the concentrations of the ligands and this is consistent with
an intramolecular chelation of the partially coordinated
ligand. There are several possible reasons why two phases are
observed in some reactions, but only one phase in others. It
could be that for some systems the second phase is too slow to
be observed over the time-scale of the stopped-flow exper-
iments (ca. 100 s). Alternatively, it could be that for some
ligands the second phase is faster than the first phase. Finally,
it could be that, in some systems, the absorbance change of
the second phase is too small to be detectable.
quadridentate ligands. The values of the rate constants {k−a
=
0.01–0.43 s−1, ka = 4.0–350 dm3 mol−1 s−1 and kb = 0.04–0.20
s−1} shows that there is not a large variation in any of the rate
constants. Furthermore, the values of ka and k−a are very
similar to the rate constants observed in many other kinetic
studies on the binding of various N, O or F donor ligands to
Ni2+.24 Thus, changes of the bridging residue (CH2CH2, C6H4
or anthracene), the symmetry of the imine bonds and the sub-
stituent (on the phenolic residue) have little effect on ka, k−a or
kb. The minor differences observed in ka are presumably
attributable to slight variations in KL. The similar reactivities
of H2L0X, H2salphen, H2LaX and H2LX–Cl (Table 5), and particu-
larly that electron-withdrawing and electron-donating substitu-
ents on the phenolic group have similar effects on the
reactivity, indicates that the outer-sphere association (KL) is
rather non-specific.
Kinetic studies on the reactions of Ni2+ with H2LHn and Lpny
(n = 0–3)
The kinetics observed for the reactions of Ni2+ with H2LX0,
H2salphen, H2LaX and H2LX–Cl are consistent with the mechan-
ism shown in the top line of Fig. 8. This is the Eigen–Wilkins
mechanism22–24 which involves an initial outer-sphere associ-
ation between the multidentate ligand and Ni2+ (KL), followed
by the dissociation of a coordinated solvent molecule from
Ni2+ (k2), and the subsequent binding of the first donor atom
of the multidentate. Extensive studies have shown that in the
reactions with Ni2+, the dissociation of the coordinated solvent
occurs at a rate essentially independent of the nature of the
nucleophile.25 Depending on the multidentate ligand, the rate-
limiting step for the reaction can either be the formation of A
{rate law shown in eqn (3), assuming KL[L] < 1} or the for-
mation of B {rate law shown in eqn (4), assuming KL[L] < 1}.
Clearly, the kinetics cannot distinguish between the rate-limit-
ing step being the initial binding of the multidentate to Ni2+
or the subsequent chelation step.
In a further series of studies, the kinetics of the reactions
between Ni2+ and H2LnH (n = 1–3) have been investigated. The
H2LnH ligands have features in common with the Schiff base
ligands presented in Table 5 (i.e. neutral ligands with terminal
salicaldimine residues). However, in H2LnH, a number of NH
groups are introduced into the ligand.
The reactions of all H2LnH (n = 1–3) with Ni2+ in methanol
are associated with two phases with the rate law for the first
phase being that presented in eqn (1) and the rate law for the
second phase being that shown in eqn (2). The rate constants
for these reactions are summarised in Table 6 and Fig. 9. It is
clear that the presence of the NH groups results in signifi-
cantly faster rates (ka ≥ 1 × 103 dm3 mol−1 s−1). Indeed, these
rates are similar to, or faster than, the solvent exchange rate
for [Ni(MeOH)6]2+ which is in the range k = 2 × 102–1 × 103
s−1 26
.
Similar behaviour has been observed (in aqueous
3378 | Dalton Trans., 2014, 43, 3372–3382
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Table 6 Summary of the rate constants in the reactions between Ni2+
and H2LnH measured in methanol at 25.0 °C
atoms (statistical effect). Consistent with this interpretation,
earlier
studies
on
the
reactions
of
Ni2+
with
(NH2CH2CH2NHCH2)2 or (NH2CH2CH2NHCH2CH2)2NH in the
presence of acid showed that the rates of the reactions with
the various protonated forms of these ligands are dominated
by an electrostatic effect, but statistical factors also
contribute.27,28
Studies on another series of ligands show a similar behav-
iour to that of H2LnH. The Lpny ligands are structurally similar to
H2LnH but contain terminal pyridyl residues rather than pheno-
lic groups. For the reactions between Ni2+ and Lpny, both the
spectrophotometric changes and the kinetics indicate that
these reactions go essentially to completion (Experimental
section). The associated rate law for the reactions is eqn (5).
Ligand
na
k−a /s−1
ka/dm3 mol−1 s−1
kb/s−1
Symmetrical Schiff bases
H2L0H
H2L1H
H2L2H
H2L3H
0
1
2
3
0.022 0.005
0.35 0.02
1.1 0.1
6.5 0.5
—
1.0 0.1 × 103
1.1 0.1 × 103
1.3 0.1 × 103
0.04 0.01
0.05 0.01
0.01 0.005
2.0 0.2
Pyridyl Schiff bases
L0py
L1py
L2py
L3py
0
1
2
3
3.2 0.3 × 103
4.3 0.3 × 103
4.8 0.3 × 103
5.3 0.3 × 103
Pyridyl pentadentate
Lpy
3
0.8 0.1 × 103
Rate ¼ ka½Lnpyꢁ½Ni2þꢁ
ð5Þ
a n = number of NH groups.
The values of ka are presented in Table 6 and Fig. 9 (insert).
Two aspects of these data are worthy of comment. First, com-
parison of the data for the analogous ligands H2L0H and Lp0y
shows that L0py reacts nearly 103 times faster than H2L0H, indi-
cating that a pyridyl residue, like an NH group, is a labilising
influence in these reactions. Secondly, it is evident from the
data for Lnpy that the rate of the reaction increases as the
number of NH groups increase, in a similar fashion to that
observed for H2LnH. The rate of the reaction of Ni2+ with the
saturated quinquidentate ligand, Lpy also occurs at a rate
similar to those observed for Lnpy. Finally, it is worth noting
that both Lpy and L1py have the same number, but different
types, of N donors. Comparison of the rates of the reactions of
Ni2+ with these two ligands indicates that the imine N has a
minor or negligible effect on the rate of binding.
That ligands containing either, or both, pyridyl and NH
groups result in fast reactions with Ni2+ requires further con-
sideration. The increased rate of reaction (faster than that of
the solvent exchange rate for [Ni(MeOH)6]2+) could be either
due to both pyridyl and NH groups being involved in ‘internal
conjugate base mechanisms’ or that the reactions being
observed involve rate-limiting chelate formation with the rate
of chelation being rapid because initial coordination of a
pyridyl or NH group affects the lability of the coordinated
methanol. To further address this issue we have studied the
effect of deprotonating a coordinated methanol (by a non-coor-
dinating base, NEt3) has on the rate of reactions between Ni2+
and H2L0H.
Fig. 9 (Main) Variation of rate of reaction between Ni2+ and H2LHn (n =
0–3). The data presented are for the formation pathway {ka in eqn (1)}. A
similar trend is observed for the corresponding dissociation pathway
{k−a in eqn (1)}. (Insert) Variation of rate of reaction between Ni2+ and
Lnpy (n = 0–3).
solution) for the reactions between [Ni(OH2)6]2+ and aliphatic
polyamines.27 It has been suggested, in these earlier studies,
that a so-called ‘internal conjugate base mechanism’ could
operate in these systems. The internal conjugate base mechan-
ism involves hydrogen bonding of a coordinated MeOH to an
NH group in the outer-sphere association between the ligand
and Ni2+. This hydrogen bonding imparts methoxide character
to the methanol ligand which labilises the other Ni-methanol
co-ligands (vide infra).
Effect of NEt3 on the reactions between Ni2+ and H2LH0 : the
conjugate base mechanism
A further feature of the reactivities of H2LnH is evident in
Fig. 9. It is clear that the rate of the reactions increase with the
value of n. However, whilst there is a large increase in rate
between H2L0H and H2L1H, the subsequent increases in the rates
for H2L2H and H2L3H are much smaller. The large difference in
rates between H2L0H and H2LH1 indicates that even a single NH
group leads to a significant increase in the rate. The smaller
increases in rate associated with H2L2H and H2LH3 are a conse-
quence of the increase in the number of potential NH donor
The rate of the reaction between Ni2+ and H2LH0 is affected by
+
the ratio [NHEt3 ]/[NEt3] as shown in Fig. 10. At high [NHEt3+]/
[NEt3] the rate of the reaction is essentially that observed in
the studies between Ni2+ and H2L0H described earlier in this
+
paper. However, at low [NHEt3 ]/[NEt3] the rate increases and
the absorbance–time curves become biphasic with k1obs for the
+
fast phase dependent on [NHEt3 ]/[NEt3] and the slow phase
independent of [NHEt3 ]/[NEt3] (k2obs = 21.0
2.0 s−1). The
+
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 3372–3382 | 3379

View Article Online
Paper
Dalton Transactions
The rate law associated with the mechanism shown in
Fig. 8 is presented in eqn (7), where kc = k2 (if binding of the
first donor is rate-limiting) or kc = k2k3/(k−2 + k3) (if chelation is
rate-limiting) and kd = k4 (if binding of the first donor is rate-
limiting) or kd = k4k5/(k−4 + k5) (if chelation is rate-limiting).
Eqn (7) is derived assuming that both KL and K1 are rapidly
established equilibria and that both species A and C (in Fig. 8)
are steady state intermediates. Comparison of eqn (6) and (7)
yields the values KLkc = 6.0 0.5 dm3 mol−1 s−1, KLkd = (4.8
0.5) × 104 dm3 mol−1 s−1 and K1 = 0.065 0.01. Consideration
of these kinetic parameters for the reactions between Ni2+ and
H2L0H in the presence of mixtures of NEt3 and NHEt3 allow us
+
to calculate pKa = 11.9 0.1 for [Ni(MeOH)6]2+. This value is
reasonable, because previous studies have shown that pKa
=
Fig. 10 Dependence of kobs on the ratio [NHEt3+]/[NEt3] for the reac-
tion of H2L0H (2.5 mmol dm−3) with Ni2+ (0.5 mmol dm−3) in MeOH at
25.0 °C. Data points correspond to [NEt3] = 20 mmol dm−3, [NHEt3+] =
▲
9.2–10.9 for [Ni(OH2)6]2+.32 Using the relationship, pKa(MeOH) =
1.02pKa(H2O) + 0.72,33 we can estimate pKa = 10.1–11.8 for
[Ni(MeOH)6]2+, in good agreement with that measured in
these studies. Furthermore, the ratio kd/kc = (8.0 1.1) × 103 is
a measure of the increased lability of [Ni(OMe)(MeOH)5]+ com-
pared to [Ni(MeOH)6]2+. The best fit value for KLkc = 6.0
0–30 mmol dm−3
(
); [NEt3] = 10 mmol dm−3, [NHEt3+] = 0–30 mmol
dm−3
(
●
). Curve drawn is that defined by eqn (6).
increase in k1obs at low [NHEt3 ]/[NEt3] indicates a deprotona-
+
0.5 dm3 mol−1 −1, derived from this analysis, is in reasonable
s
+
tion is labilising. We presume that at all values of [NHEt3 ]/
agreement with the value KLkc = ka = 6.5 0.5 dm3 mol−1 s−1
measured directly for the reaction of Ni2+ with H2LH0 (Table 5).
The slow phase only becomes evident when k1obs ≥ ca.
50 s−1. It seems likely that the slower phase, k2obs = 21.0
2.0 s−1, corresponds to the chelation step and becomes distinct
[NEt3] the reaction also exhibits a first order dependence on
the concentration of H2L0X. We have shown that when
+
[NHEt3 ]/[NEt3] = 2, the reaction does indeed exhibit a first
order dependence on the concentration of H2L0H (ESI†).
However, it is difficult to demonstrate a similar dependence
only when k1obs > ko2bs
.
on the concentration of H2L0H at low [NHEt3 ]/[NEt3] because
+
the rate of reaction, under these conditions, is close to the
limit of the stopped-flow apparatus. Analysis of the data shown
in Fig. 10 yields the rate law shown in eqn (6), when [H2L0H] =
Kinetics of binding and evidence of preferential binding
The results from the kinetic studies of the reactions between
2.5 mmol dm−3
.
+
Ni2+ and H2LX0 in the presence of mixtures of NEt3 and NHEt3
allow us to make some conclusions concerning: (i) the involve-
ment of an internal conjugate base mechanism in the reac-
tions with ligands containing pyridyl or NH groups and (ii) the
preferential binding of certain groups to Ni2+.
f120 þ 0:23½NHEt3þꢁ=½NEt3ꢁg½Ni2þꢁ
1 þ 15:5½NHEt3þꢁ=½NEt3ꢁ
Rate ¼
ð6Þ
ð7Þ
fKLkd þ ðKLkc=K1Þ½NHEt3þꢁ=½NEt3ꢁg½H2LH0 ꢁ½Ni2þꢁ
The increased rates observed in the reactions with H2LXn
and Lnpy are consistent with an ‘internal conjugate base mech-
anism’ involving the NH groups (pKa ∼ 11).30 However, it
seems unreasonable that the increase in rate observed between
H2L0H and Lp0y is due to the pyridyl groups operating an
‘internal conjugate base mechanism’ because the pyridyl
group is a much weaker base (pKa ∼ 5).30 It seems more likely
that the enhanced rate observed with L0py is consistent with the
mechanism shown in Fig. 8, where chelation is the rate-limit-
ing step. Thus, the increased reactivity observed with L0py is
attributable to the initial coordination of the pyridyl group to
the Ni and the presence of the pyridyl group in the coordi-
nation sphere of the Ni labilises the methanol co-ligands
hence facilitating the chelation step (k3).
Rate ¼
1 þ ð1=K1Þ½NHEt3þꢁ=½NEt3ꢁ
The data in Fig. 10 shows that kobs depends on the ratio
[NHEt3 ]/[NEt3], but not the absolute concentrations of either
+
of these reactants. This behaviour indicates that only a single
deprotonation occurs. There are two candidates for being
deprotonated: either [Ni(MeOH)6]2+ or H2LH0 . In methanol, the
pKa of NHEt3+ is 10.729 and the pKa of phenol is 14.3,30 and so,
over the range of [NHEt3+]/[NEt3] used in these studies essen-
tially all H2L0H remains diprotonated. Thus the dependence on
+
[NHEt3 ]/[NEt3] must be attributable to the formation of the
conjugate base, [Ni(OMe)(MeOH)5]+.
The mechanism consistent with the kinetics in the pres-
ence of [NHEt3 ]/[NEt3], is presented in Fig. 8. The top line of
+
Fig. 8 presents the pathway described earlier (observed in the
absence of base), whilst the lower pathway is the conjugate
base mechanism. In this lower pathway, rapid deprotonation
of a coordinated methanol produces [Ni(OMe)(MeOH)5]+ (K1),
Conclusions
which reacts more rapidly with H2L0H, presumably because The aim of this work was to see if kinetics (specifically the
[Ni(OMe)(MeOH)5]+ is more labile to methanol dissociation rates of reactions) could be used to probe which donor site in
than [Ni(MeOH)6]2+.31
a multidentate ligand binds preferentially to a metal ion. The
3380 | Dalton Trans., 2014, 43, 3372–3382
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
kinetic studies reported herein for the reactions of Ni2+ with a
variety of similar neutral multidentate ligands containing
various types of donors (phenolic OH, imine N, pyridyl N and
NH) certainly indicate that the rates of reactions depend on
the composition of the ligands, and follows the order: NH >
pyridyl N > phenolic OH ∼ imine N. However, only with
ligands containing pyridyl groups is there evidence that this
order reflects preferential initial binding of specific groups to
Ni2+. The rapid reactions observed with ligands containing NH
6 J. Gonzalez, R. Gavara, O. Gadea, S. Blasco, E. Garcıa-
Espana and F. Pina, Chem. Commun., 2012, 48, 1994.
7 (a) T. Punniyamurphy, S. Velusamy and J. Iqbal, Chem. Rev.,
2005, 105, 2329; (b) E. N. Jacobsen, Acc. Chem. Res., 2000,
33, 421.
8 C. M. Da Silva, D. L. Da Silva, L. V. Modolo, R. B. Alves,
M. A. De Resende, C. V. B. Martins and A. De Fatima,
J. Adv. Res., 2011, 2, 1.
9 N. A. Negm and M. F. Zaki, Colloids Surf., A, 2008, 322, 97.
groups could also be due to preferential initial binding of the 10 Y.-F. Ji, R. Wang, S. Ding, C.-F. Du and Z.-L. Liu, Inorg.
NH group to Ni2+, but the NH groups are sufficiently basic to
Chem. Commun., 2012, 16, 47.
possibly be involved in an ‘internal conjugate base mechan- 11 D. Nartop, P. Gürkan, N. Sari and S. Çete, J. Coord. Chem.,
ism’, and this could be the origin of the increased reactivity
observed with ligands containing NH groups.
2008, 61, 3516.
12 For both types of ligands (except for H2LaH and H2LCl–Cl),
these ligands are unsymmetrical in two senses: they are
unsymmetrical with respect to the ‘direction’ of the imine
bond (i.e. –CHvN-aryl-CHvN–) and they are unsymmetri-
cal with respect to the substituents on the terminal pheno-
lic residues. Throughout this paper, reference to
unsymmetrical Schiff base ligands refers to the unsymme-
trical nature of the imine bond.
Studies on ligands containing terminal phenolic groups
(H2L0X, H2LXa, H2LX–Cl) show that the rates of reactions are all
similar, and rather insensitive to substituents on the phenolic
groups, symmetry of the imine bonds and nature of bridge
between phenolic groups. This indicates that for H2L0X, H2LXa,
H2LX–Cl that either the binding of the first donor is rate-limit-
ing (k2) or, that if chelation (k3) is rate-limiting, the binding of
the first donor to Ni has little effect on the lability of the 13 Ö. Güngör and P. Gürkan, Spectrochim. Acta, Part A, 2010,
methanol co-ligands. 77(1), 304.
In the long term it is anticipated that understanding the 14 H2LX–Cl show two strong bands in the region
preferences of different donors binding to metal ions will con-
tribute to understanding the dynamics of metal ions being
encapsulated by biomolecules. In this study we deliberately
chose to study reactions with Ni2+, where the substitution
mechanism is predominantly dissociative. Future studies will
explore the reactions of multidentate ligands with metal ions
where associative substitution mechanisms can operate.
1589–1624 cm−1, attributable to νCvN, and νCvC in the
region 1454–1589 cm−1 19
Two bands are observed because
.
of the two asymmetric imine groups. A similar character-
istic has been reported for H2LaX.11 Because of the different
chemical environments of the unsymmetrical imine
groups, the 1H NMR spectra of all H2LX–Cl show two
signals in the range 8.31–8.97 ppm and 9.57–9.85 ppm
(Table 2). The two phenolic protons of all H2LX–Cl are
observed in the ranges 9.90–13.81 ppm. Finally, the peak
attributable to CH3 in H2LMe–Cl is observed at 2.50 ppm.
The CH3 group in H2LMe is also observed at 2.50 ppm.20
The mass spectroscopy results for H2LX–Cl are presented in
Table 1. The values of the molecular ion peaks and the
fragmentation products are consistent with the proposed
structures of unsymmetrical Schiff bases. The molecular
ion peaks are observed at the predicted values of m/z: 399.2
[M]+ (H2LH–Cl), 413.2 [M − H]+ (H2LMe–Cl) and 433.0 [M −
2H]+ (H2LCl–Cl). The same fragmentation pathways appear
for the highest intensity peaks in each ligand. Thus, peaks
at m/z = 242.3, m/z = 262.1 and m/z = 282.0 for H2LH–Cl,
H2LMe–Cl and H2LCl–Cl, respectively, are attributed to the
loss of the [M − (C7H5NOCl) − 4H]+, [M − (C7H5NOCl) +
2H]+, [M − (C7H5NOCl) + 3H]+ fragments, which is
common to all H2LX–Cl.
Acknowledgements
We thank Mr Thaer Al-Rammahi and Dr Ahmed Al-Saffar for
assistance with some experiments. This work is supported by
the Council of Higher Education (Turkey) and the Research
Foundation of Nevşehir University (BAP 2010/17). We thank
EPSRC (UK) for equipment funding.
Notes and references
1 E. Horst, Coord. Chem. Rev., 1999, 187, 37, and references
therein.
2 R. G. Wilkins, Comments Inorg. Chem., 1983, 2, 187.
3 C. E. Castillo, M. A. Mánez, J. González, J. M. Llinares,
H. R. Jiménez, M. G. Basallote and E. García-España, Chem. 15 R. Pohloudek-Fabini and D. Froehling, Arch. Pharm., 1965,
Commun., 2010, 46, 6081. 298, 423.
4 C. E. Castillo, J. González-García, J. M. Llinares, 16 M. Nakamura, K. Komatsu, Y. Gondo, H. Ohta and Y. Ueda,
M. A. Máñez, H. R. Jimenez, E. García-España and
M. G. Basallote, Dalton Trans., 2013, 42, 6131.
5 L. Siegfried, C. N. McMahon, J. Baumeister, M. Neuburger,
T. A. Kaden, S. Anandaram and C. G. Palivan, Dalton Trans.,
2007, 4797.
Chem. Pharm. Bull., 1967, 15, 585.
17 SMART, SAINT and SADABS software, Bruker AXS Inc.,
Madison, Wisconsin, USA, 2001; CrysAlis Pro, Agilent
Technologies, Oxford, 2009; G. M. Sheldrick, Acta Crystal-
logr., Sect. A: Fundam. Crystallogr., 2008, 64, 112.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 3372–3382 | 3381

View Article Online
Paper
Dalton Transactions
18 J. Espenson, Chemical Kinetics and Reaction Mechanisms, 25 R. G. Wilkins, Acc. Chem. Res., 1970, 3, 409, and references
McGraw-Hill, New York, 1981, p. 16. therein.
19 R. M. Silverstein, G. C. Bassler and T. C. Morrill, Spectro- 26 H. P. Bennetto and E. F. Caldin, J. Chem. Soc. A, 1971, 2198.
photometric Identification of Organic Compounds, Wiley, 27 (a) R. B. Jordan, Reaction Mechanisms of Inorganic and
New York, 1981.
20 E. Erdik, Organik Kimyada Spektroskopik Yontemler, Gazi
Büro Kitapevi, Ankara, 1993, p. 201.
Organometallic Systems, OUP, New York, 1991, p. 84;
(b) D. B. Rorabacher, Inorg. Chem., 1966, 5, 1891;
(c) R. B. Jordan, Inorg. Chem., 1976, 15, 748.
21 M. A. Siegler and M. Lutz, Cryst. Growth Des., 2009, 9, 1194, 28 D. W. Margerum, D. B. Rorabacher and J. F. G. Clarke Jnr.,
and references therein. Inorg. Chem, 1963, 2, 667.
22 M. L. Tobe and J. Burgess, Inorganic Reaction Mechanisms, 29 N. F. Hall and M. R. Spinkle, J. Am. Chem. Soc., 1932, 54, 3469.
Longman, Harlow, 1999, p. 310.
23 R. G. Wilkins, Kinetics and Mechanism of Reactions of
30 F. Rived, M. Roses and E. Bosch, Anal. Chim. Acta, 1998,
374, 309.
Transition Metal Complexes, VCH, Weinheim, 2nd edn, 1991, 31 Ref. 21, p. 152.
ch. 4. 32 Ref. 23, p. 265.
24 J. Burgess, Metal Ions in Solution, Ellis Horwood, Chiche- 33 B. G. Cox, Acids and Bases Solvent Effects on Acid-Base
ster, 1978, ch. 10.
Strength, OUP, Oxford, 2013, p. 131.
3382 | Dalton Trans., 2014, 43, 3372–3382
This journal is © The Royal Society of Chemistry 2014