Spectroscopic and structural insights into N-substituted pyridinium-4-aldoximes and their pentacyanoferrate(II) complexes

Article abstract of DOI:10.1016/j.poly.2012.07.077

Comparative kinetic and equilibrium studies on the formation and dissociation of three mono- and one bis-pyridinium-4-aldoxime pentacyanoferrate(II) complexes have been carried out in aqueous solutions at 25 °C and I = 0.1 M. The synthesis, spectroscopic and thermal characterization of a new N-methylpyridinium-4-aldoxime pentacyanoferrate(II) complex is presented. The obtained values for the equilibrium constants, identified as apparent formation constants (β_f/M^-1) along with kinetic parameters, the formation (k_f/M^-1 s^-1) and dissociation (k_d/s^-1) rate constants indicated the behaving of all protonated pyridinium-4-aldoximes as weak π-acceptors. The pH-dependence of the dissociation rates has been analyzed in terms of ionization abilities of the coordinated ligands. The magnitude of the dissociation rates suggested that both protonated and deprotonated ligand forms are effective σ-donors that bind to the [Fe(CN)₅]^3- moiety through the nitrogen atom. The deprotonation of the coordinated aldoxime group leads to the reduced lability of the complexes due to increased σ-donor capability of aldoximato nitrogen causing the strengthened of the iron(II)-nitrogen bond. The values of dissociation activation parameters, ΔH^? and ΔS^?, are found to be consistent with the S _N1 dissociative type of mechanism. The spectroscopic data (FT-IR, NMR and UV-Vis) of the isolated coordinated pentacyanoferrates(II) conforms with the weak π-acceptor properties of the pyridinium-4-aldoxime ligands. A detailed structural characterization of the iodide and chloride salt of N-methylpyridinium-4-aldoxime was also presented using NMR, FT-IR and UV-Vis spectroscopy, as well as X-ray diffraction.

Full text of DOI:10.1016/j.poly.2012.07.077

Polyhedron 52 (2013) 733–742
Contents lists available at SciVerse ScienceDirect
Polyhedron
journal homepage: www.elsevier.com/locate/poly
Spectroscopic and structural insights into N-substituted
pyridinium-4-aldoximes and their pentacyanoferrate(II) complexes
a,
a,
a
a
b
⇑
⇑
ˇ
´
´
´
´
Blazenka Foretic , Igor Picek , Vladimir Damjanovic , Danijela Cvijanovic , Ivana Pulic ,
b
b
ˇ
´
´
Boris-Marko Kukovec , Dubravka Matkovic-Calogovic
a Department of Chemistry and Biochemistry, Faculty of Medicine, University of Zagreb, Šalata 3, 10000 Zagreb, Croatia
b Department of Chemistry, Faculty of Science, University of Zagreb, Horvatovac 102a, 10000 Zagreb, Croatia
a r t i c l e i n f o
a b s t r a c t
Article history:
Available online 3 August 2012
Comparative kinetic and equilibrium studies on the formation and dissociation of three mono- and one
bis-pyridinium-4-aldoxime pentacyanoferrate(II) complexes have been carried out in aqueous solutions
at 25 °C and I = 0.1 M. The synthesis, spectroscopic and thermal characterization of a new N-methylpyrid-
inium-4-aldoxime pentacyanoferrate(II) complex is presented. The obtained values for the equilibrium
constants, identified as apparent formation constants (bf/Mꢀ1) along with kinetic parameters, the forma-
tion (kf/Mꢀ1 sꢀ1) and dissociation (kd/sꢀ1) rate constants indicated the behaving of all protonated pyrid-
Dedicated to Alfred Werner on the 100th
Anniversary of his Nobel Prize in Chemistry
in 1913.
inium-4-aldoximes as weak
in terms of ionization abilities of the coordinated ligands. The magnitude of the dissociation rates sug-
gested that both protonated and deprotonated ligand forms are effective -donors that bind to the
[Fe(CN)5]3ꢀ moiety through the nitrogen atom. The deprotonation of the coordinated aldoxime group
leads to the reduced lability of the complexes due to increased -donor capability of aldoximato nitrogen
causing the strengthened of the iron(II)–nitrogen bond. The values of dissociation activation parameters,
Hà and Sà, are found to be consistent with the SN1 dissociative type of mechanism. The spectroscopic
data (FT-IR, NMR and UV–Vis) of the isolated coordinated pentacyanoferrates(II) conforms with the weak
-acceptor properties of the pyridinium-4-aldoxime ligands. A detailed structural characterization of the
p-acceptors. The pH-dependence of the dissociation rates has been analyzed
Keywords:
Pyridinium-4-aldoxime derivatives
Pentacyanoferrates(II)
Aqueous equilibria
Kinetics
r
r
Spectroscopy
D
D
p
iodide and chloride salt of N-methylpyridinium-4-aldoxime was also presented using NMR, FT-IR and
UV–Vis spectroscopy, as well as X-ray diffraction.
Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction
oximes has been well documented [3], describing the immense
structural diversity and properties of the complexes with monoa-
The mono- and bis-pyridinium oximes as quaternized deriva-
tives of pyridyl oximes are known as pharmacologically important
agents. Explicitly the ionization of oxime group produces powerful
nucleophilic oximato form, widely used in the hydrolytic cleavage
of either phosphoric and carboxylic acid esters or amides [1]. The
positive charge present in the pyridinium moiety is responsible
for the fact that pKa of oxime group is 3–5 pKa units lower when
compared to a parent pyridyl oxime (10 6 pKa 6 12). Besides, the
coordination of pyridyl oximes to metal increases the acidity in
the coordinated ligands making them important as hydrolytic cat-
alysts in physiological conditions [2]. Investigations of the acidity–
nucleophilicity relationship revealed that the reactivity of com-
plexed pyridyl oximates depends not only on the nature of the
bound metal ion, but also on the structure of the oxime ligand. Un-
like the pyridinium oximes, the coordination chemistry of pyridyl
nions of simple pyridyl oximes that exhibit many distinct coordi-
nation modes. Currently, only formation of the pyridinium oxime
complex with palladium(II) [4a] and, of particular relevance to
our investigations, the complexes with the pentacyanoferrate(II)
moiety were reported [4b–h], but only limited characterization
data were presented at the time. A more detailed studies of the
structure and reactivity toward the aquapentacyanoferrate(II),
[Fe(CN)5(H2O)]3ꢀ, of some N-phenacyl- and N-benzoylethylpyridi-
num-4-aldoximes were recently reported [5]. Acidic properties and
nucleophilicity of the mono- and bis-pyridinium aldoximes make
them well known as effective antidotes against poisoning by
organophosphates which cause long-term inactivation of acetyl-
cholinesterase (AChE, E.C.3.1.1.7) [6]. The positive charge in the
pyridinium moiety site-directs reactivation of the enzyme catalytic
site providing the cleavage of a serine-bounded phosphate by the
aldoximato moiety. Many of pyridinium aldoximes have also
shown other significant and versatile bioactivities [7] which are
evidently closely related to the interplay of their structure, acidic
properties and coordination ability. Therefore, their reactivity
⇑
Corresponding authors. Tel.: +385 1 4566 760; fax: +385 1 4590 236 (B. Foretic´),
tel.: +385 1 4566 755; fax: +385 1 4590 236 (I. Picek).
E-mail addresses: bforetic@mef.hr (B. Foretic´), ipicek@mef.hr (I. Picek).
0277-5387/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.poly.2012.07.077

´
B. Foretic et al. / Polyhedron 52 (2013) 733–742
734
and coordination properties still represent an important area of re-
search in the view to recognize and rationalize relationships be-
tween the electronic/structural properties of the binding metal
centers and the types of reactions they induce at the ligated
aldoximes.
sition. The constant pH values of the aqueous solutions were main-
tained by using Britton and Robinson buffers, prepared by mixing
100 mL of phosphoric, boric and acetic acid solution (all 0.04 M)
with different volumes of 0.2 M sodium hydroxide solution. A
constant ionic strength was adjusted by the addition of sodium
chloride. Redistilled water was used throughout. A pH-meter
(Mettler Toledo) with an open junction combination polymer elec-
trode was used for pH measurements accurate to 0.01 pH units.
Elemental analysis was performed with a LECO elemental analyzer,
using the ASTM D 5291 method. Melting points were determined
by a Stuart SMP3 apparatus with a temperature resolution of
0.1 °C. Thermal studies (TGA–DTA–DSC) were carried out in a N2
atmosphere with the heating rate 5 °C minꢀ1. TA Instrument, Mod-
el SDT 2960 with a temperature range from 25 to 1600 °C was used
for the TG–DT analysis. The DSC measurements were performed on
TA Instrument, Model DSC 2910 with a temperature range from
ꢀ150 to 725 °C.
The
low-spin
pentacyano(ligand)ferrate(II)
complexes,
[Fe(CN)5(L)]nꢀ, have been the subject of numerous investigations.
They were found to be suitable as models for active sites in biolog-
ical macromolecules [8], as representatives of supramolecular sys-
tems [9] and recently as electron donor groups for nonlinear optics
[10]. Since the chemical properties of the [Fe(CN)5(L)]nꢀ are
strongly dependent on the nature of L, the extensive characteriza-
tion of N-heterocyclic and other N- and S-containing pentacyano-
ferrates(II) were summarized and reviewed [11]. Although the
kinetics of a ligand substitution reactions with a wide variety of
N-, S- and O-donor ligands were systematically and comprehen-
sively studied providing the binding affinity of specific donors to-
wards iron(II) [12], oximic ligands have never been among them.
Thus, a comparative study concerning the coordination ability of
the pyridinium-4-aldoxime containing ligands toward the penta-
cyanoferrate(II) moiety was performed, with a special regard to
the N-methylpyridinium-4-aldoxime along with the two mono-
pyridinium ligands and one bis-pyridinium ligand (Scheme 1). A
detailed characterization of the N-methylpyridinium-4-aldoxime
and its pentacyanoferrate(II) complex was performed using UV–
Vis, NMR (1H and 13C), FT-IR spectral and thermal analysis, as well
as X-ray diffraction in the case of the ligand. To recognize the
chemical properties of the pyridinium-4-aldoxime pentacyanofer-
The UV–Vis measurements were performed on either a UNICAM
UV 4 or a Varian Cary Bio 100 spectrophotometer with thermo-
stated cell holders and 1-cm silica-glass cells. FT-IR spectra were
recorded on a Perkin Elmer Spectrum GX, Series R spectrometer
in the range of 4000–400 cmꢀ1 using KBr pellets. The 1H and 13C
NMR spectra were recorded at room temperature with a Bruker
AV-600 spectrometer equipped with 5 mm inverse detection or
dual probes respectively, operating at 600.133 MHz for the 1H nu-
cleus and 150.917 MHz for the 13C nucleus. The spectra were re-
corded in D2O with tetramethylsilane as the internal standard.
rate(II) complexes as well as the
of the cited ligands, equilibrium and kinetic studies have been car-
r-donor and p-acceptor potential
2.2. TMB4-2Br and synthesis of H[Fe(CN)5(TMB4)]ꢁ2H2O
ried out.
The commercially available TMB4-2Br was purified by recrys-
tallization from ethanol according to known procedure [13]. The
obtained pale yellow precipitate was analyzed by FT-IR and NMR
2. Experimental
spectroscopy. FT-IR (cmꢀ1):
1645 (vs);
(NO), 1017 (vs), 984 (vs).
1H NMR (D2O, d/ppm): 8.44 (s, H-1), 8.30 (d, J = 6.78 Hz, H-3(30),
m
(O–H)oxime, 3437 (vs);
m(C@N)oxime,
2.1. Materials and instruments
m(C–C, C–N)pyridinium ring, 1611 (vs), 1570 (m), 1521 (s);
m
The starting materials for the synthesis of PAM4-I, BPA4-ClꢁH2O
and FEPA4-ClꢁH2O were reagent grade products and employed as
purchased. TMB4-2Br was obtained from Aldrich and purified
before use. BPA4-ClꢁH2O and FEPA4-ClꢁH2O were synthesized
according to a general procedure [5b]. Sodium amminepentacy-
anoferrate(II), Na3[Fe(CN)5(NH3)]ꢁ3H2O (Sigma–Aldrich) was used
after recrystallization from a concentrated ammonia solution.
Solutions of [Fe(CN)5(H2O)]3ꢀ were obtained by aquation of
[Fe(CN)5(NH3)]3ꢀ and were freshly prepared at room temperature
and kept in the dark to minimize thermal and photolytic decompo-
H-6(60)), 8.98 (d, J = 6.78 Hz, H-4(40), H-5(50)), 4.89 (t, J = 7.77 Hz,
2H-7(70)), 2.91 (quin, J = 7.80 Hz, 2H-8). 13C NMR (D2O, d/ppm):
146.3 (C-1(10)), 149.2 (C-2(20)), 125.9 (C-3(30), C-6(60)), 144.8 (C-
4(40), C-5(50)), 57.8 (C-7(70)), 31.7 (C-8). The chemical shifts were
found to be in a good agreement with those published for TMB4-
2Br in DMSO-d6 [14].
The solid dark blue TMB4-pentacyanoferrate(II) complex was
isolated as a precipitate from the aqueous mixture containing
[Fe(CN)5(H2O)]3ꢀ and
a fivefold excess of aldoxime at the
Scheme 1. Assignment of the ligands: N-methylpyridinium-4-aldoxime chloride (PAM4-Cl) and iodide (PAM4-I), N-benzylpyridinium-4-aldoxime chloride (BPA4-Cl), N-
phenacylpyridinium-4-aldoxime chloride (FEPA4-Cl) and N,N0-bis(pyridinium-4-aldoxime)trimethylene dibromide (TMB4-2Br).

´
735
B. Foretic et al. / Polyhedron 52 (2013) 733–742
concentrations above 10ꢀ3 M and purified as described previously
[4d]. Anal. Calc. for H[Fe(CN)5(C15H18N4O2)]ꢁ2H2O: C, 47.16; H,
4.55; N, 24.75; Fe, 10.97. Found: C, 47.24; H, 4.59; N, 24.60; Fe,
10.86%.
was used for absorption effects. The structures were solved by
direct methods with SHELXS-97 [17]. Refinement procedure by the
full-matrix least squares method based on F2 values against all
reflections was performed. Anisotropic displacement parameters
for all non-hydrogen atoms were included.
FT-IR (cmꢀ1):
2108 (m), 2044 (vs);
vs) + 1637 (sh, vs); d(HOH), 1616 (s);
(s), 1576 (s), 1517 (vs); (NO)free
m
(O–H)water
+
m
(O–H)oxime, 3402 (br, s);
m
(CN)cyano
(C@N)oxime bonded, 1643 (br,
(C–C, C–N)pyridinium ring, 1611
,
m
(C@N)oxime free
+
m
The positions of hydrogen atoms belonging to the carbon atoms
Csp2 and methyl Csp3 atoms were geometrically optimized in all
three structures applying the riding model [Csp2–H, Csp3
(methyl)–H, 0.93 Å and 0.96 Å, respectively; Uiso(H) = 1.2Ueq(C)
for Csp2 and 1.5Ueq(C) for Csp3(methyl)]. The hydrogen atom
belonging to the oxime O1 atom in PAM4-I-o was not found in
the difference Fourier map, nor could its position be modeled by
applying the appropriate riding model. In the structure of PAM4-
ClꢁH2O, all hydrogen atoms were located in the difference Fourier
maps however those on the C-atoms were geometrically opti-
mized. Those on the oxime O1 and on the water molecule were
placed as found in the map and were refined isotropically. In
PAM4-I-m the oxime hydrogen atom was calculated as an ideal-
ized hydroxyl group with the torsion angle from the electron den-
sity. Calculations were performed with SHELXL-97 [17], PLATON [18]
and PARST [19]. The molecular graphics were done with PLATON and
MERCURY (Version 1.4.2) [20]. Crystal data and details of the struc-
ture determination for PAM4-I polymorphs and PAM4-ClꢁH2O are
given in Table 1.
m
m
+
m
(NO)bounded, 1013 (br, vs).
2.3. Synthesis of PAM4-I and PAM4-ClꢁH2O single crystals
PAM4-I was prepared according to a general procedure by
mixing ethanol solution of pyridine-4-aldoxime (Fluka) with iodo-
methane (Sigma–Aldrich) present in threefold excess. The resulting
mixture was stirred at room temperature for 1 day. The produced
crystalline precipitate was filtered off and purified by recrystalliza-
tion from ethanol. Yellow, water-soluble single crystals of PAM4-I
were obtained (m.p.: 170.5 °C, decomp.). Yield, 47%. Anal. Calc. for
C7H9N2OI: C, 31.84; H, 3.44; N, 10.61. Found: C, 31.89; H, 3.41; N,
10.70%. FT-IR (cmꢀ1):
(s); (C–C, C–N)pyridinium ring, 1611 (s), 1573 (w), 1520 (m);
1000 (vs).
m
(O–H)oxime, ꢂ3150 (s);
m
(C@N)oxime, 1644
m
m
(NO),
1H NMR (D2O, d/ppm): 8.51 (s, H-1), 8.32 (d, J = 6.60 Hz, H-3, H-
6), 8.91 (d, J = 6.66 Hz, H-4, H-5), 4.52 (s, 3H-7). 13C NMR (D2O, d/
ppm): 145.3 (C-1), 148.4 (C-2), 124.3 (C-3, C-6), 145.0 (C-4, C-5),
47.9 (C-7).
2.4. Synthesis of Na(PAM4)[Fe(CN)5(PAM4)]ꢁH2O
PAM4-I was converted into corresponding chloride by treatment
with methanolic solution of hydrogen chloride in accordance with a
convenient method of anion exchange in quaternary salts [15].
PAM4-I was dissolved in freshly prepared methanolic solution satu-
rated with anhydrous HCl at room temperature. The solution was
concentrated by evaporation under reduced pressure followed by
addition of acetone and allowed to cool in a refrigerator (at 4 °C).
After 2 days, the produced transparent crystals of monohydrate salt,
PAM4-ClꢁH2O, were filtered off, washed with acetone and dried in
vacuum(m.p.: 169.1 °C, decomp.). Yield, 42%. Anal. Calc. for C7H11N2-
O2Cl: C, 44.10; H, 5.81; N, 14.70. Found: C, 44.09; H, 5.61; N,
PAM4-I, dissolved in the minimal volume of ethanol–water
solution (in volume ratio of approx. 1) was added to the concen-
trated aquated [Fe(CN)5(NH3)]3ꢀ solution in fivefold excess. The in-
stantly produced dark blue reaction mixture was kept in the
refrigerator overnight (at 4 °C). Afterwards, the precipitation was
initiated by addition of ethanol. The produced dark-blue powder
was filtered off, washed with ethanol and dried in vacuum. Yield,
26%. Anal. Calc. for Na(C7H9N2O)[Fe(CN)5(C7H9N2O)]ꢁH2O: C,
45.53; H, 4.02; N, 25.15. Found: C, 45.10; H, 4.39; N, 24.71%. TGA
analysis supported the loss of one water molecule upon heating.
Unfortunately, attempts to obtain crystals of complex suitable for
X-ray structural analysis remained unsuccessful. FT-IR (cmꢀ1):
14.71%. FT-IR (cmꢀ1):
m(O–H)water
+ m(O–H)oxime, 3356 (s), 3216 (s);
m
(C@N)oxime, 1639 (s); m(C–C, C–N)pyridinium ring, 1607 (s), 1567 (m),
1519 (vs);
m(NO), 1008 (vs).
m
(O–H)water
(vs); (C@N)oxime
C–N)pyridinium ring, 1614 (vs), 1573 (s), 1522 (vs);
(NO)bounded, 1009 (br, vs).
The 1H and 13C NMR spectra were taken immediately after dis-
+
m
(O–H)oxime, 3400 (br, vs);
(C@N)oxime bounded, 1643 (br, vs);
(NO)free + -
m
(CN)cyano, 2111 (s), 2057
1H NMR (D2O, d/ppm): 8.46 (s, H-1), 8.30 (d, J = 6.54 Hz, H-3, H-
6), 8.92 (d, J = 6.54 Hz, H-4, H-5), 4.54 (s, 3H-7). 13C NMR (D2O, d/
ppm): 145.9 (C-1), 147.7 (C-2), 124.2 (C-3, C-6), 145.0 (C-4, C-5),
47.6 (C-7).
m
+
m
m
(C–C,
free
m
m
solving the complex in D2O. In the 1H NMR spectrum only the
broad signals with the absence of the expected splitting patterns
were observed. 13C NMR shifts of coordinated and uncoordinated
PAM-4 were distinguishable except for the signals corresponding
to the C-2 and C-4, C-5 atoms. 13C NMR (D2O, d/ppm) of the com-
plex anion: 149.6 (C-1), 128.2 (C-3, C-6), 48.1 (C-7), 173.2 (cis-
C„N), 169.0 (trans-C„N).
2.3.1. X-ray single-crystal structure determination
Diffraction data were collected from suitable single crystals of
PAM4-I-o (orthorhombic polymorph) and PAM4-ClꢁH2O on an Ox-
ford Diffraction Xcalibur single-crystal diffractometer with Xcalibur
Sapphire 3 CCD detector and Mo K
a radiation. The CrysAlis Soft-
ware system, Version 171.32.24 [16] was used for data collection
and reduction. Data for PAM4-I-o were collected at room tempera-
ture (296 K) while those for PAM4-ClꢁH2O were collected at low
temperature (150 K). Later, the crystals of PAM4-I obtained directly
from the reaction were remeasured at 150 K (same diffractometer)
but it was found to be a different polymorph (monoclinic) and will
be referred to as PAM4-I-m. All trials to obtain again the ortho-
rhombic polymorph have failed so far. It was difficult to find a single
crystal of PAM4-I-m because of twinning. The measured crystal also
had 7% of the twin component (twin law (001)[104]) however
since the structure refined well the small twin component was
not used in calculations. The unit cell of PAM4-I-m was checked
at room temperature to test if there is a possible phase transition
due to the change in temperature. The obtained unit cell
(a = 7.112(6) Å, b = 10.403(6) Å, c = 13.303(14) Å, b = 95.58(7)°)
shows no transformation to PAM4-I-o. The multi-scan procedure
2.5. Electronic absorption studies
2.5.1. Determination of the PAM4-I and PAM4-Cl ionization constants
The spectrophotometric titrations of the 4 ꢃ 10ꢀ5 M solutions of
PAM4-I and PAM4-Cl were performed at 25 °C and I = 0.1 M. The
spectral characteristics of the respective ionic forms and the
ionization constants were derived from the absorbance versus pH
curves by a non-linear regression as described previously [5]
(Fig. S1).
2.5.2. Determination of the apparent formation constants for
[Fe(CN)5(HnL)](3ꢀn)ꢀ complexes
Equilibrium constants specified as apparent formation con-
stants, bf, have been evaluated by non-linear regression of the

´
B. Foretic et al. / Polyhedron 52 (2013) 733–742
736
Table 1
Crystal data and details of the structure determination for PAM4-I-o and PAM4-I-m polymorphs and PAM4-ClꢁH2O.
Compound
PAM4-I-o
PAM4-I-m
PAM4-ClꢁH2O
Empirical formula
Mr
C7H9IN2O
264.06
C7H9IN2O
264.06
C7H11ClN2O2
190.63
Color and habit
Crystal system
Space group
Crystal dimensions (mm)
Unit cell parameters
a (Å)
yellow, prism
orthorhombic
Pnam
0.70 ꢃ 0.39 ꢃ 0.30
yellow, prism
monoclinic
P21/c
0.51 ꢃ 0.37 ꢃ 0.25
colorless, prism
orthorhombic
P212121
0.44 ꢃ 0.40 ꢃ 0.23
13.2260(12)
10.4618(8)
7.0852(7)
90
7.0204(3)
10.3580(4)
13.2217(5)
90
6.80810(10)
8.9236(2)
14.5295(3)
90
b (Å)
c (Å)
a
(°)
b (°)
90
90
98.150(4)
90
951.74(7)
0.71073
90
90
882.71(3)
0.71073
4
c
(°)
V (Å3)
980.36(15)
0.71073
4
1.789
3.218
Radiation, Mo Ka (Å)
Z
4
Dcalc (g cmꢀ3
)
1.843
3.315
1.434
0.394
l
(mmꢀ1
)
h range for data collection (°)
h, k, l range
Number of measured reflections
Number of independent reflections (Rint
3.80–26.99
ꢀ16:16; ꢀ13:13, ꢀ9:9
21902
1155 (0.0357)
1081
4.23–29.00
ꢀ9:9; ꢀ12:14; ꢀ18:18
4786
2481 (0.0139)
2246
4.57–30.00
ꢀ4:9; ꢀ10:12; ꢀ20:16
4482
2555 (0.0113)
2511
)
Number of observed reflections, I P 2r(I)
Number of refined parameters
68
126
126
R,a wRb [I P 2
r(I)]
0.0497, 0.1426
0.0544, 0.1464
0.0758, 1.0496
–
0.0431, 0.1083
0.0470, 0.1100
0.0243, 7.9293
–
0.0243, 0.0626
0.0252, 0.0636
0.0403, 0.0566
0.00(4)
R, wR [all data]
g1, g2 in wc
Flack parameter
Goodness of fit on F2, Sd
1.210
1.199
0.989
Maximum, minimum electron density (e Åꢀ3
)
1.028, ꢀ0.368
ꢀ1.080, 2.149
0.204, ꢀ0.181
Maximum
Absorption correction type
Range of transmission factors minimum, maximum
D/r
<0.001
<0.001
<0.001
multi-scan
0.240, 0.380
multi-scan
0.744, 1.000
multi-scan
0.921, 1.000
a
b
c
R =
wR = [
w = 1/[
R
||Fo| ꢀ |Fc||/
R|Fo|.
(Fo ꢀ Fc2)2/
R .
w(Fo2)2]1/2
2
R
r
2(Fo2) + (g1P)2 + g2P] where P = (Fo2 + 2Fc2)/3.
S =
R
[w(Fo ꢀ Fc2)2/(Nobs ꢀ Nparam)]1/2
.
d
2
experimental data to the hyperbolic Eq. (1) valid for the spectro-
A
1
is the maximal absorbance reached during the course of
photometric mole ratio method [21].
the reaction. In all the kinetic solutions the concentration of
[Fe(CN)5(H2O)]3ꢀ was 5 ꢃ 10ꢀ5 M. The data were collected under
pseudo-first-order conditions by the use of a 20- to 150-fold molar
excess of the ligand. All measurements were performed in tripli-
cate. The second-order formation rate constant, kf, was obtained
as the slope of a plot of kobs versus the concentration of the ligand
(Fig. S3).
1
2
1
AkðMLCTÞ
¼
Alim
ꢁ
1 þ z þ
f
b ꢁ cð½FeðCNÞ ðH2OÞꢄ3ꢀÞ
5
1
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
!
u
u
t
2
1
C
A
ꢀ
1 þ z þ
ꢀ 4 ꢁ z
ð1Þ
b ꢁ cð½FeðCNÞ ðH2OÞꢄ3ꢀÞ
f
5
The dissociation kinetic study of [Fe(CN)5(HnL)](3ꢀn)ꢀ complexes
was performed under pseudo-first-order conditions by following
the absorbance decrease at the wavelengths corresponding to the
MLCT maximum of the formed complexes after addition of large
excess of DMSO as a [Fe(CN)5(H2O)]3ꢀ scavenger. The quantitative
formation of [Fe(CN)5(HnL)](3ꢀn)ꢀ in initial reaction mixtures was
Absorbances collected at the metal-to-ligand charge transfer
(MLCT) maximum of the formed complexes at 25 °C, I = 0.1 M and
pH 6.04 0.02 were analyzed (Fig. S2). The mole ratio of ligand to
metal analytical concentrations is defined by parameter z with the
constant total analytical concentration of [Fe(CN)5(H2O)]3ꢀ
(2 ꢃ 10ꢀ4 M) through measurements and Alim is theoretical absor-
bance corresponding to quantitative reaction:
ensured by the addition of 25-fold excess of ligands to
a
5 ꢃ 10ꢀ5 M [Fe(CN)5(H2O)]3ꢀ solutions. The pseudo-first-order rate
constant, identified as kd, was determined from the dependence of
ln(At) versus time. All measurements were performed in triplicate
and pH range from 5.0 to 11.5. The thermal activation parameters
for the dissociation reactions were determined at pH 6.0 from the
Eyring linear plots of ln[(kd/T)/(sꢀ1 Kꢀ1)] versus 1/T within the tem-
perature range 10–30 °C.
cðHnLnþÞ
z ¼
;
cð½FeðCNÞ ðH2OÞꢄ3ꢀÞ
5
Alim
¼
e
ð½FeðCNÞ ðHnLÞꢄð3ꢀnÞꢀÞ ꢁ cð½FeðCNÞ ðH2OÞꢄ3ꢀÞ:
5
5
2.5.3. Kinetic measurements
The formation rates of [Fe(CN)5(HnL)](3ꢀn)ꢀ complexes were
measured spectrophotometrically at the wavelengths correspond-
ing to the MLCT bands of the complexes at 25.0 0.5 °C, I = 0.1 M
and pH 6.0 0.2. The pseudo-first-order rate constants, kobs, were
3. Results and discussion
3.1. The crystal structures of PAM4-ClꢁH2O, PAM4-I-o and PAM4-I-m
obtained from the slopes of the linear least-squares fits of ln(A
ꢀ At) versus time plots, where At is the absorbance at time t and
ORTEP view of the molecular structures of PAM4-ClꢁH2O and
1
PAM4-I-m is depicted in Fig. 1. That of PAM4-I-o is given in the

´
737
B. Foretic et al. / Polyhedron 52 (2013) 733–742
Fig. 1. ORTEP drawing of (a) PAM4-ClꢁH2O and (b) PAM4-I-m with the atom numbering scheme. Thermal ellipsoids are shown at the 50 probability level at 150 K.
Fig. S4. The crystal structures of PAM4-I polymorphs are given in
the Fig. S5. The asymmetric unit of PAM4-ClꢁH2O contains a N-
methylpyridinium-4-aldoxime cation, a chloride ion and a water
molecule of crystallization (Fig. S5). The selected molecular geom-
etry parameters for all three structures are listed in Table 2 and
hydrogen bond geometry in the Table S1. The bond distances and
angles in the structure of PAM4-I polymorphs and PAM4-ClꢁH2O
correspond well to the values found in the literature for similar
structures containing the aldoxime moiety bonded to the pyridini-
um ring in the position 4 (Table 2) [5a,22,23]. The arrangement of
substituents around the C@N bond in all three structures shows
that the aldoxime has the E-configuration (Fig. 1 and Fig. S4). The
value of the endocyclic bond angle C4–N2–C5 deviates from the
ideal value for the six-membered ring by 0.2(8)° (PAM4-I-o),
1.0(5)° (PAM4-I-m) and 0.8(1)° (PAM4-ClꢁH2O). This deviation is
significantly smaller than the ones found for pyridoxal oxime
methyl iodide (2.5(3)°) and for pyridoxal oxime methyl chloride
(2.2(1)°) [23a]. The sum of the endocyclic bond angles is 720° as
it is expected for an unpuckered aromatic ring. The conformation
of the aldoxime moiety is antiperiplanar, while the torsion angle
O1–N1–C1–C2 is 179.23(9)° in PAM4-ClꢁH2O, 180° in PAM4-I-o
and 179.7(6)° in PAM4-I-m. The pyridinium ring in PAM4-I-o does
not deviate from planarity since the whole molecule (including the
iodide anion) lies on a crystallographic mirror plane, while in
PAM4-I-m the molecule is in the general position and the greatest
deviation from the plane through the pyridinium ring is 0.151(6) Å
of atom O1. In PAM4-ClꢁH2O the greatest deviation is 0.124(1) Å of
the N1 atom. There is a hydrogen bond O1–Hꢁ ꢁ ꢁI between the N-
methylpyridinium-4-aldoxime cation and the iodide anion in the
crystal structures of the two PAM4-I polymorphs (Fig. S5). Similar
O–Hꢁ ꢁ ꢁBr hydrogen bond was found between N-benzylpyridinium-
4-aldoxime cation and bromide anion in the crystal structure of N-
benzylpyridinium-4-aldoxime bromide [23b]. In the structure of
PAM4-ClꢁH2O the water molecule links the chloride anions and
PAM4 cations into endless chains parallel to [100]. Two hydrogen
bonds of the O–Hꢁ ꢁ ꢁCl type are observed (Fig. S5 and Table S1) with
a water molecule being a bridging hydrogen bond donor toward
two chloride anions. In addition, the water molecule is an acceptor
of a hydrogen bond from the aldoxime hydroxyl group. In the three
structures there are also weak intermolecular hydrogen bonds of
the C–Hꢁ ꢁ ꢁO type. Furthermore, the crystal packing of PAM4-I
and PAM4-ClꢁH2O is stabilized by intermolecular C–Hꢁ ꢁ ꢁI and
C–Hꢁ ꢁ ꢁCl contacts, respectively. These interactions in the structure
of PAM4-I link the N-methylpyridinium-4-aldoxime cations and
iodide anions into a network parallel to (001). These networks
are further assembled in the [001] direction only by weak van
der Waals interactions.
3.2. Spectroscopic and thermal studies
3.2.1. Spectroscopic studies of PAM4-I and PAM4-ClꢁH2O
The electronic absorption spectra of PAM4-Cl and PAM4-I were
recorded in aqueous solutions in the pH range of 5.0–11.5. The two
intensive pH-dependent bands at about 280 and 340 nm were ob-
Table 2
Selected bond distances (Å) and angles (°) for PAM4-I-o and PAM4-I-m polymorphs
and PAM4-ClꢁH2O.
served, typical for the
p ?
p⁄ transitions within the protonated
Compound
PAM4-I-o
PAM4-I-m
PAM4-ClꢁH2O
and deprotonated pyridinium-4-aldoxime system, respectively, as
was previously found for the solutions of BPA4-Cl, FEPA4-Cl and
TMB4-2Br [5b,24]. However, the low intensity band centered on
240 nm attributed to the deprotonated pyridinium-4-aldoxime
chromophore, was observed only in the absorption spectra of
PAM4-Cl. In the spectra of PAM4-I this band was covered by inten-
sive, pH-independent charge-transfer band with a maximum
around 225 nm. Like in a similar electronic system, present in N-
methylpyridinium iodide, the band arose as a consequence of the
PAM4+Iꢀ association complex formation predominantly driven by
electrostatic forces [25]. To eliminate the possible influence of such
association on ionization properties and reactivity of the pyridini-
um-4-aldoxime group the ionization constants of both, PAM4-Cl
and PAM4-I, were determined and the 1H and 13C NMR spectra
in D2O were correlated. The comparative UV–Vis spectroscopic
characterization of the respective ionic forms and the ionization
constants of the pyridinium-4-aldoxime group in the examined li-
gands are presented in Table 3. In view of the fact that the ioniza-
tion of the carbonyl group in FEPA4 ion was negligible in specified
pH range [5], it was treated as mono-pyridinium-4-aldoxime.
N1–C1
C1–C2
C2–C3
C2–C6
C3–C4
C4–N2
C5–N2
C5–C6
C7–N2
N1–O1
1.354(16)
1.456(14)
1.376(12)
1.369(15)
1.373(13)
1.312(10)
1.319(11)
1.358(14)
1.467(11)
1.287(13)
1.275(8)
1.476(8)
1.361(8)
1.398(9)
1.383(8)
1.325(7)
1.344(7)
1.376(9)
1.474(7)
1.375(7)
1.2802(14)
1.4669(15)
1.4008(14)
1.3990(14)
1.3760(15)
1.3518(13)
1.3446(13)
1.3767(15)
1.4779(14)
1.3816(12)
O1–N1–C1
N1–C1–C2
C3–C2–C1
C6–C2–C1
C3–C2–C6
C4–C3–C2
N2–C4–C3
N2–C5–C6
C5–C6–C2
C5–N2–C4
C4–N2–C7
C5–N2–C7
105.0(14)
114.9(9)
118.8(9)
123.6(8)
117.6(9)
119.1(8)
121.6(8)
121.0(8)
120.4(8)
120.2(8)
120.9(8)
119.0(8)
111.0(5)
120.1(6)
120.8(6)
120.3(5)
118.8(6)
119.8(5)
120.8(5)
120.4(5)
119.2(5)
121.0(5)
119.5(5)
119.5(5)
112.40(9)
117.23(9)
120.83(10)
121.19(10)
117.97(9)
119.84(10)
120.59(10)
120.95(10)
119.80(10)
120.76(9)
120.91(9)
118.31(9)

´
B. Foretic et al. / Polyhedron 52 (2013) 733–742
738
Table 3
The ionization constants and molar absorption coefficients of the predominant ionic forms of the pyridinium-4-aldoximic compounds in aqueous solutions at 25 °C and
c = 4 ꢃ 10ꢀ5 M.
Compound
kmax (nm)
e
(H2L2+
)
(Mꢀ1 cmꢀ1
)
e
(HL+)max (Mꢀ1 cmꢀ1
)
e
(L)max (Mꢀ1 cmꢀ1
)
pKa1(aldoxime)
pKa2(aldoxime)
max
PAM4-Ia
279
337
278
336
283
342
283
345
282
345
18330
15990
16194
23330
3440
25860
2800
8.56 0.05
PAM4-ClꢁH2Oa
BPA4-ClꢁH2Ob
FEPA4-ClꢁH2Ob
TMB4-2Brc
8.57 0.05
8.76 0.02
8.72 0.07
7.50 0.08
22270
24280
27800
5600
54130
41200
29890
22660
8.90 0.08
a
b
c
This work: I = 0.1 M.
Data from Ref. [5b]: I = 0.1 M.
Data from Ref. [24]: I = 0.05 M.
The identical values of ionization constants for PAM4-I and
significant contribution of coordinated PAM4 as extended chromo-
phore. The low intensity band due to 1A1 ? 1E(1) d–d transition,
usually positioned in the range from 264 to 444 nm [11a,12h],
which has been found to be sensitive to the nature of the sixth
ligand for a large number of pentacyano(ligand)ferrate(II) com-
plexes was completely masked. In the visible region, characteristic
broad, intensive maximum centered in the range from 560 to
570 nm was compatible with the metal-to-ligand charge transfer,
PAM4-Cl (close to the literature pKa value of 8.44 [26]) as well as
the negligible differences in the NMR chemical shifts were estab-
lished. In addition, the NMR chemical shifts were in a good agree-
ment with those obtained for the TMB4-2Br (see Section 2), BPA4-
Cl and FEPA4-Cl [5b]. This clearly indicated a negligible impact of
the PAM4+Iꢀ association complex formation to PAM4 reactivity to-
wards the [Fe(CN)5]3ꢀ moiety.
The similar acidities of the pyridinium-4-aldoxime moieties in
the examined compounds were found to be in accordance with
the similar positions and intensities of the aldoximic C@N and
N–O stretching frequencies in the IR spectra (see Section 2). The
higher frequencies of N–O stretching accompanied by lower C@N
frequencies when compared with the frequencies in pyridine-4-
aldoxime revealed the considerable resonance contribution of
@C–N@O+–H [27]. Evidently, the ionization abilities of mono-
and bis-pyridinium-4-aldoximes were essentially independent of
the substituent attached to the N-methylenepyridinium moiety.
Although the mean pKa value of 8.2 that has been established for
the bis-pyridinium-4-aldoxime, TMB4, fits well, the difference be-
t
2g ? L , whose energy, and only slightly intensity, was pH-sensi-
p⁄
tive (
e
= 4830 530 Mꢀ1 cmꢀ1). The MLCT band shifted about
10 nm toward lower energy upon increasing the pH of the medium
over 9, due to the equilibrium:
ð2Þ
tween pKa1 and pKa2
(D
pKa ꢂ 1.4) implies that ionization of the
where R represents the pyridinium part of the ligand. Similar red-
shifts in the MLCT band together with small intensity changes were
noted previously for the complexes produced with BPA4 and FEPA4
[5b] and were also observed for the aqueous reaction mixtures of
the TMB4-pentacyanoferrate(II). These decreases in the MLCT en-
ergy are attributable to increases in the electron density at the iro-
n(II) center caused by deprotonation of the coordinated ligand.
The 13C NMR spectrum of the PAM4-pentacyanoferrate(II) con-
tained two sets of signals corresponding to C-1, C-3,6 and C-8 nu-
clei of both the coordinated and uncoordinated PAM4.
Furthermore, two cyanide resonance signals, with a ratio of rela-
tive intensities of 4:1, were assigned to four cis- (173.2 ppm) and
one trans-configured cyano groups (169.0 ppm) and are consistent
with the trans-configuration of PAM4 in the octahedral PAM4-
pentacyanoferrate(II). The upfield shifts of the cis- and trans-cyano
groups in respect to hexacyanoferrate(II) ion (177 ppm) [11a], also
observed for the BPA4-pentacyanoferrate(II) [5b] were evident.
The shielding of the cyano carbon atoms is most probably the re-
first aldoxime group influences the ionization of the second. The
pKa difference in the range of 0.6–0.7, based solely on statistical
factors, was shown earlier for some symmetric bis-pyridinium
aldoximes [26] and was in agreement with a lack of direct elec-
tronic communication between the two pyridinium aldoximic do-
nor sites. Therefore, in symmetrically structured TMB4, without
possibility of an electronic communication between the two pyrid-
inium-4-aldoxime rings, the indicated effect may be associated
only with the steric disturbances.
3.2.2. Spectroscopic studies of Na(PAM4)[Fe(CN)5(PAM4)]ꢁH2O
The solids of the pentacyanoferrates(II) complexed with mono-
pyridinium-4-aldoximes, PAM4 and formerly described BPA4 and
FEPA4, have been water-soluble to some extent, while the complex
with bis-pyridinium-4-aldoxime, TMB4, has been insoluble. The
pH-dependent UV–Vis spectra of the solutions of the newly syn-
thesized Na(PAM4)[Fe(CN)5(PAM4)]ꢁH2O solid are presented in
Fig. 2. The UV region was dominated by the intensive superim-
sult of increased
p-back donation from iron(II) to cyano ligands
posed intraligand absorption bands due to
p
?
p⁄ transitions
due to pronounced r
-induction of electron density from coordi-
within the uncoordinated and coordinated ligand (kmax ꢂ 282 nm,
nated PAM4 towards iron(II). This is strongly supported by the con-
siderable deshielding of aldoxime carbon atom as a consequence of
the increased polarization of the aldoxime C@N bond which is con-
(e
e
(HL+) +
(L) +
e
([Fe(CN)5(HL)]2ꢀ))max = 30900 Mꢀ1 cmꢀ1; kmax ꢂ 340 nm,
(
e
([Fe(CN)5(L)]3ꢀ))max = 35550 Mꢀ1 cmꢀ1), whose intensity
was pH-dependent. These pH-dependent spectral changes exhib-
ited excellent isobesticity. The dependence of absorbances at these
maxima on pH resulted in clear titration curves (inset in Fig. 2)
whose inflection points revealed the pKa value higher than
8.6 established for free PAM4, and thus proving the expected
sistent with the poor p-acceptor ability of the coordinated PAM4.
The FT-IR spectra of PAM4-ClꢁH2O, PAM4-I and TMB4-2Br, sim-
ilar to the previously characterized BPA4-ClꢁH2O and FEPA4-ClꢁH2-
O, exhibit strong bands due to aldoxime C@N and N–O stretchings
at ca. 1643 cmꢀ1 and 1000 cmꢀ1, respectively. Generally, upon

´
739
B. Foretic et al. / Polyhedron 52 (2013) 733–742
Fig. 2. pH-dependent UV–Vis spectra of the isolated Na(PAM4)[Fe(CN)5(PAM4)]ꢁH2O in aqueous solutions at 25 °C, c = 8 ꢃ 10ꢀ5 M, I = 0.1 M.
Fig. 3. TGA and DTA curves for Na(PAM4)[Fe(CN)5(PAM4)]ꢁH2O.
coordination of the oxime nitrogen to a metal ion the C@N
stretches shift to lower frequencies, while N–O stretches shift to-
ward higher frequencies [28]. This is not pronounced in the spectra
of the pyridinium-4-aldoxime pentacyanoferrate(II) complexes.
Despite the strong absorption of the uncoordinated mono-pyridini-
um-4-aldoximes (PAM4, BPA4, FEPA4) present as cationic
counterparts, as well as the uncoordinated N-methylenepyridini-
um-4-aldoxime part in the bis-pyridinium-4-aldoxime (TMB4),
only the broadening of the N–O stretching bands in the examined
pentacyanoferrate(II) complexes points to the coordination.
Therefore, only the cyanide stretching modes, in the 2100–
2000 cmꢀ1 range, has been found as diagnostically valuable. Their
similar positions and the splitting pattern are found in the spectra
of all examined pyridinium-4-aldoxime pentacyanoferrates(II)
(this work: PAM4, TMB4; in Ref. [5b]: BPA4, FEPA4). They are dis-
tinctive for the octahedral pentacyano(ligand)ferrate(II) complexes
[11a,29] and strongly indicate the analogous mode of coordination
emphasizing the same nature of the formed iron(II)–pyridinium-4-
aldoxime bond. The striking similarity between the spectra of the
complexes to those of free ligands suggest that
r- and p-bonding
effects seem nearly to cancel out, as was the case for pentacyano-
ferrates(II) complexed with pyridyl amide derivatives [30].
3.2.3. Thermal behavior of Na(PAM4)[Fe(CN)5(PAM4)]ꢁH2O
The TGA, DTA and DSC curves (Fig. 3 and Fig. S6) revealed mul-
tistage decomposition of the PAM4-pentacyanoferrate(II) in a sim-
ilar manner as observed for other pentacyano(ligand)ferrates(II)
[31]. The decomposition started around 30 °C and ended at
750 °C yielding a final stable product with 16.72% of residual
weight. In initial stage, upon heating the sample up to 100 °C,
where generally the dehydration occurs [31], the endothermic pro-
cess was observed with the total weight loss of 3.70%. This is in

´
B. Foretic et al. / Polyhedron 52 (2013) 733–742
740
Table 4
for N-donor cationic ligands [12e] but in agreement with the re-
ported composite formation rates of 15 and 11 Mꢀ1 sꢀ1 for BPA4-
and FEPA4-pentacyanoferrates(II) at 22 °C, pH 7.90, I = 0.05 M
[4c]. The moderately higher formation rate established for the
ambidentate bis-pyridinium-4-aldoxime (TMB4) in comparison
with mono-pyridinium-4-aldoximes (PAM4, BPA4 and FEPA4)
was in accordance with the presence of two donor groups and
the overall higher positive charge of +2. The unexpectedly slow for-
mation, otherwise characteristic for the anionic ligands [11b],
could be speculatively associated with the fact that the charge
within the protonated pyridinium-4-aldoxime cation is rather dis-
persed and/or with a conformation effect related to a steric hin-
drance of E-configurated aldoxime donor group as was suggested
for some pyridyl aldoximes [33].
Formation kinetics and equilibrium constants of [Fe(CN)5(HnL)](3ꢀn)ꢀ at 25.0 0.5 °C
and I = 0.1 M.
Complex
kMLCT (nm) log(bf/Mꢀ1
)
kf (Mꢀ1 sꢀ1
)
log[(kf/kd)/Mꢀ1
]
[Fe(CN)5(PAM4)]2ꢀ 560
[Fe(CN)5(BPA4)]2ꢀ 574
[Fe(CN)5(FEPA4)]2ꢀ 580
1.91
2.54
2.16
2.62
3.9 0.3
7.6 0.2
4.1 0.4
19.7 0.5
2.44
2.74
2.53
3.15
[Fe(CN)5(TMB4)]ꢀ
575
accordance with the release of one water molecule (theoretical:
3.60%) and confirms that within the isolated complex mole ratio
Na(PAM4)[Fe(CN)5(PAM4)]/H2O equals to 1. The next decomposi-
tion stages in the temperature ranges of 94–220 °C and 270–
520 °C, evidenced by two exothermic peaks in DTA and DSC curves,
clearly indicated the simultaneous decomposition reactions which
could not be resolved. Thus, the clear release of coordinated and
uncoordinated PAM4 was not observed. The total weight loss of
83.28% was in good agreement with the weight loss corresponding
to the loss of one water molecule, two PAM4 cations and five cya-
nide groups (theoretical: 84.26%) and strongly supported the pro-
posed molecular formula of the isolated complex.
According to a general knowledge, the rate of pyridinium-4-
aldoxime substitution in the formed [Fe(CN)5(HnL)](3ꢀn)ꢀ com-
plexes, specified as dissociation rate (reverse reaction in Eq. (3)),
is strongly dependent on the
r-donor and p-acceptor ability of
coordinated ligand [11,12]. Thus a comparative analysis of the
kinetics (i.e. lability) of the formed pentacyano(ligand)ferrates(II)
was performed by studies of a pyridinium-4-aldoxime substitution
from the [Fe(CN)5(HnL)](3ꢀn)ꢀ as a function of pH. The rate of disso-
ciation was found to be pH-dependent due to the deprotonation of
the bounded aldoxime group. The pH profiles of the dissociation
rates are shown in Fig. 4. Consequently, for the examined pyridini-
um-4-aldoxime pentacyanoferrate(II) complexes with the dissoci-
ation reaction scheme involved the following general steps:
3.3. Equilibrium and kinetic studies
The substitution of water in the [Fe(CN)5(H2O)]3ꢀ with the
pyridinium-4-aldoximic ligands is presented by the following gen-
eral equation with a related equilibrium constant:
Ka
½FeðCNÞ ðHnLÞꢄð3ꢀnÞꢀ ¢ Hþ þ ½FeðCNÞ ðHnꢀ1LÞꢄð4ꢀnÞꢀ
ð4Þ
ð5Þ
ð6Þ
5
5
k
kf
ꢀ1
½FeðCNÞ ðHnLÞꢄð3ꢀnÞꢀ þ H2O !½FeðCNÞ ðH2OÞꢄ þ HnLnþ
½FeðCNÞ ðHnꢀ1LÞꢄð4ꢀnÞꢀ þ H2O !½FeðCNÞ ðH2OÞꢄ3ꢀ þ H
3ꢀ
½FeðCNÞ ðH2OÞꢄ þ HnLnþ ¢ ½FeðCNÞ ðHnLÞꢄð3ꢀnÞꢀ þ H2O
3ꢀ
5
5
5
5
kd
k
ꢀ2
ðnꢀ1Þþ
L
½½FeðCNÞ ðHnLÞꢄð3ꢀnÞꢀ
ꢄ
nꢀ1
5
5
5
bf ¼
ð3Þ
3ꢀ
½½FeðCNÞ ðH2OÞꢄ ꢄ ꢁ ½HnLnþꢄ
5
where Ka is the ionization constant of the coordinated pyridinium-
4-aldoxime. The rate constant is:
where n is an integer ranging from 1 to 0 for PAM4, BPA4 and
FEPA4, and from 2 to 0 for TMB4.
For all examined ligands the spectrophotometric mole ratio
method, performed at pH 6.04 0.02, confirmed the formation of
k
ꢀ1 ꢁ ½Hþꢄ þ kꢀ2 ꢁ Ka
kd ¼
ð7Þ
½Hþꢄ þ Ka
a
complex of 1:1 molar composition and formation of the
Dissociation rate parameters for the release of protonated (kꢀ1) and
deprotonated (k pyridinium-4-aldoxime ligand from the
corresponding pentacyanoferrate(II) with a protonated pyridini-
um-4-aldoxime ligand, obviously bounded to the iron center via
aldoximic nitrogen atom. Deduced equilibrium constants, specified
as apparent formation constants, bf, are presented in Table 4. De-
spite the smaller stabilities of the BPA4-, FEPA4- and TMB4-penta-
cyanoferrate(II) complexes that we have determined, when
compared with the previous literature [4c,d], the apparent stabili-
ties of the formed complexes are in good correlation with their ki-
netic properties. The comparison of stability constants with the
literature values for various pentacyanoferrates(II) indicated the
)
ꢀ2
[Fe(CN)5(HnL)](3ꢀn)ꢀ complex and the ionization constant (Ka) were
obtained by the non-linear least-squares fit of the experimental
data to Eq. (7) and are presented in Table 5. The apparent formation
constants, bf, for the reaction involving protonated pyridinium-4-
aldoximic ligands (Eq. (1)), were deduced from ratio of kinetic
parameters, kf/kd, (Table 4) and were found to be in reasonable
agreement with the values obtained by the mole ratio method.
The activation parameters determined for the dissociation reactions
confirmed limited SN1 mechanism for ligand substitution [11].
The magnitude of the dissociation rates suggested that both
protonated and deprotonated forms of PAM4, BPA4, FEPA4 and
behaving of protonated pyridinium-4-aldoximes as weak p-accep-
tors [12a–c,g,h,j,k]. Similar MLCT energies of the formed pentacy-
anoferrates(II) suggested the analogous changes in electron
distribution within the pyridinium ring upon coordination.
TMB4 are effective
r
-donors that bind to the [Fe(CN)5]3ꢀ moiety
through the nitrogen atom. Furthermore, the dissociation rate val-
ues were consistent with the values deduced for a labile pentacy-
The ligand substitution reactions of water in labile [Fe(CN)5(H2-
O)]3ꢀ with a wide variety of entering ligands (forward reaction in
Eq. (3)) have been shown to follow a dissociative mechanism
[11]. Consequently, the formation rate has been found to be influ-
enced by the charge of the entering ligand, the protonation of
[Fe(CN)5(H2O)]3ꢀ itself at one of the cyanide ligands (pK(25 °C,
I = 0.10 M) = 2.63 0.12 [32]) and by deprotonation of water in
[Fe(CN)5(H2O)]3ꢀ above pH 13 [11b]. Thus, in this study under
specified reaction conditions, the formation rates of the pyridini-
um-4-aldoxime pentacyanoferrates(II) were only depended on
the ionic form of the ligand. The obtained kf values, presented in
Table 4, have been considerably smaller in comparison with values
ano(ligand)ferrates(II) containing the exclusively
r-donor or poor
p-acceptor ligands [12g–i]. The reduced lability of the examined
pentacyano(ligand)ferrates(II) caused by the deprotonation of the
coordinated aldoxime group indicated the influence of the charge
distribution within the pyridinium-4-aldoxime system on the
strength of iron(II)–nitrogen bond. Evidently, increased basicity
of aldoximic nitrogen, i.e.
on(II)–nitrogen bond.
r-donor capability strengthened the ir-
According to the literature data the N-coordination of oxime li-
gand to the metal center (M2+; M = Cu, Pt, Ni, Zn, Fe) often lead to
dramatically increased acidity of the oxime group, and was primar-

´
741
B. Foretic et al. / Polyhedron 52 (2013) 733–742
16
14
12
10
8
kd·103 / s-1
[
Fe(CN)5(PAM4)]2–
6
[Fe(CN)5(BPA4)]2–
[Fe(CN)5(FEPA4)]2–
[Fe(CN)5(TMB4)]–
4
2
4
5
6
7
8
pH
9
10
11
12
Fig. 4. pH-dependence of the dissociation rate constant for PAM4-, BPA4-, FEPA4- and TMB4-pentacyanoferrate(II) complexes.
Table 5
Rate and activation parameters for the dissociation of pyridinium-4-aldoxime pentacyanoferrate(II) complexes.
[Fe(CN)5(PAM4)](3ꢀn)ꢀ [Fe(CN)5(BPA4)](3ꢀn)ꢀ [Fe(CN)5(FEPA4)](3ꢀn)ꢀ
[Fe(CN)5(TMB4)](3ꢀn)ꢀ
Dissociation rate parameters (rate constants in sꢀ1; 25.0 0.5 °C; I = 0.1 M)
pH 0.02
kd ꢃ 103
pH 0.03
kd ꢃ 103
pH 0.02
kd ꢃ 103
pH 0.03
kd ꢃ 103
5.03
6.05
7.52
8.23
14.4 0.9
14.3 1.1
14.6 0.1
13.8 0.1
10.9 0.4
6.6 0.4
5.8 0.3
5.1 0.6
5.99
7.78
8.19
8.95
13.8 1.8
13.6 1.8
13.2 1.4
11.9 1.3
7.2 0.6
5.8 0.6
5.6 0.7
5.0 0.4
5.05
6.05
7.54
8.24
9.04
9.90
10.49
11.05
12.1 1.6
12.0 1.2
11.9 1.2
10.4 1.6
7.2 1.3
5.3 1.8
5.1 1.8
4.6 1.8
5.75
6.60
7.15
7.55
8.45
8.80
9.55
10.15
15.2 1.1
14.8 0.7
14.8 0.8
13.3 0.6
10.5 0.7
8.7 0.5
3.6 0.4
3.2 0.3
8.99
9.91
10.06
10.61
11.34
10.21
10.44
10.61
k-1 ꢃ 103
14.5 0.7
pKa
k-2 ꢃ 103
5.0 0.5
k-1 ꢃ 103
13.9 1.6
pKa
k-2 ꢃ 103
4.6 0.6
k-1 ꢃ 103
12.0 1.3
pKa
k-2 ꢃ 103
4.5 1.8
k-1 ꢃ 103
15.0 0.9
k-2 ꢃ 103
2.5 0.4
pKa
coordinated PAM4
coordinated BPA4
coordinated FEPA4
coordinated TMB4
9.30 0.5
9.39 0.4
9.01 0.5
8.69 0.6
Activation parameters (
D
Hà in kJ molꢀ1
;
D
Sà in J Kꢀ1 molꢀ1; I = 0.1 M)
pH 6.05 0.02
pH 5.99 0.03
pH 6.05 0.03
pH 6.25 0.03
t
0.1 (°C)
kd ꢃ 103
t
0.1 (°C)
kd ꢃ 103
t
0.1 (°C)
kd ꢃ 103
t
0.1 (°C)
kd ꢃ 103
15.9
20.2
25.0
30.3
4.5 0.4
8.6 0.5
14.3 1.1
30.7 1.0
15.3
20.1
25.0
29.7
2.9 0.1
5.9 0.2
13.8 1.8
20.1 0.1
15.2
20.3
25.0
29.8
3.3 0.3
5.5 0.2
12.0 1.6
19.9 0.3
9.8
1.7 0.4
3.6 0.7
7.9 0.5
15.5 1.1
14.8
20.5
25.0
D
Hà
D
Sà
D
Hà
D
Sà
D
Hà
D
Sà
D
Hà
D
Sà
88.6 2.3
21.8 0.6
94.3 2.2
34.0 7.2
94.3 3.0
34.5 10.5
95.0 3.0
40.0 15.0
ily assigned to the nature of the M–L bond and to lesser extent to the
ligand structure [3]. Interestingly, in contrast to our earlier interpre-
tation that suggested a reduction of the pKa values of the coordi-
nated aldoxime groups in FEPA4 and BPA4 by more than one pH
unit1 [5b] the significant change in acidity of examined ligands upon
coordination has not been observed (Tables 3 and 5). De facto the pKa
value of the coordinated pyridinium-4-aldoximes in the range from
9.0 to 9.5 is undoubtedly supported by the MLCT bands shift toward
lower energy upon increasing the pH of the medium over 9. The insig-
nificant change in acidity upon coordination reveals the canceling ef-
fect between ligand-to-metal -donation and metal-to-ligand -back
donation, and is strongly supported by the established spectroscopic
characteristics (FT-IR and NMR) of the pentacyanoferrates(II) com-
plexed with the pyridinium-4-aldoxime type of ligand.
r
p
4. Conclusion
The crystalline monohydrate of N-methylpyridinium-4-aldox-
ime chloride (PAM4-Cl) and two polymorphs of anhydrous iodide
(PAM4-I) were prepared and characterized. Given that N-meth-
ylpyridinium-4-aldoxime cation represents a structural foundation
of the biologically active mono- and bis-pyridinium-4-aldoximes:
chlorides of N-benzylpyridinium-4-aldoxime (BPA4-Cl) and
1
Previously, equilibrium measurements were performed in the reaction mixtures
that contained equimolar ligand to [Fe(CN)5(H2O)]3ꢀ ratio, which probably resulted in
the formation of more than one complex in neutral and slightly alkaline media,
leading to the misinterpretation of data.

´
B. Foretic et al. / Polyhedron 52 (2013) 733–742
742
N-phenacylpyridinium-4-aldoxime (FEPA4-Cl), as well as N,N0-
bis(pyridinium-4-aldoxime)trimethylene dibromide (TMB4-2Br),
their chemical properties in aqueous media are compared. The
similar ionization abilities of their aldoxime groups in aqueous
solution have proven the insignificant effect of the substituent at-
tached to the N-methylenepyridinium moiety. Both, the mono- and
bis-pyridinium-4-aldoxime ligands react with aquapentacyanofer-
rate(II) ion by forming mononuclear complexes through the coor-
dination of nitrogen to the iron center, of the either protonated
and deprotonated aldoxime group. The produced pyridinium-4-
aldoxime pentacyanoferrates(II) are thermodynamically unstable
and kinetically labile at 25 °C and I = 0.1 M. The pentacyanofer-
rate(II) moiety does not induce a significant change in acidity of
the pyridinium-4-aldoxime ligands upon coordination. This con-
forms with the results of FT-IR and 13C NMR and UV–Vis spectral
[8] See for example: E. Ilkowska, K. Lewin´ ski, R. Van Eldik, G. Stochel, J. Biol. Inorg.
Chem. 4 (1999) 302.
[9] See for example: H. Winnischofer, F.M. Engelmann, H.E. Toma, K. Araki, H.R.
Rechenberg, Inorg. Chim. Acta 338 (2002) 27.
[10] (a) B.J. Coe, S.P. Foxon, E.C. Harper, J. Raftery, R. Shaw, C.A. Swanson, I.
Asselberghs, K. Clays, B.S. Brunschwig, A.G. Fitch, Inorg. Chem. 48 (2009) 1370;
(b) B.J. Coe, J.L. Harries, L.A. Jones, I. Asselberghs, K. Clays, B.S. Brunschwig, J.A.
Harris, J. Garin, J. Orduna, J. Chem. Soc. 128 (2006) 12192.
[11] (a) D.H. Macartney, Rev. Inorg. Chem. 9 (1988) 101;
(b) H.E. Toma, A.A. Batsta, H.B. Gray, J. Am. Chem. Soc. 104 (1982) 7509.
[12] (a) J.B. Luiz, G.J. Leigh, F.S. Nunes, Polyhedron 21 (2002) 2137;
(b) Y. Baran, A. Ulgen, Int. J. Chem. Kinet. 30 (1998) 415;
(c) C. Chen, M. Wu, A. Yeh, T.Y.R. Tsai, Inorg. Chim. Acta 267 (1998) 81;
(d) S.da S.S. Borges, A.L. Coelho, I.S. Moreira, Polyhedron 13 (1994) 1015;
(e) D.H. Macartney, L.J. Warrack, Can. J. Chem. 67 (1989) 1774;
(f) D.H. Macartney, A. McAuley, Inorg. Chem. 20 (1981) 748;
(g) D.H. Macartney, A. McAuley, J. Chem. Soc., Dalton Trans. (1981) 1780;
(h) H.E. Toma, J.M. Martins, E. Giesbrecht, J. Chem. Soc., Dalton Trans. (1978)
1610;
(i) H.E. Toma, J.M. Malin, Inorg. Chem. 13 (1974) 1772;
analysis which revealed weak
p-back bonding capability of the
(j) H.E. Toma, J.M. Malin, E. Giesbrecht, Inorg. Chem. 12 (1973) 2084;
(k) H.E. Toma, J.M. Malin, Inorg. Chem. 12 (1973) 2080;
(l) H.E. Toma, J.M. Malin, Inorg. Chem. 12 (1973) 1039.
pyridinium-4-aldoxime ligands. Spectroscopic, elemental and ther-
mal analyses of the newly synthesized complex were in optimal
accordance with the formula Na(PAM4)[Fe(CN)5(PAM4)]ꢁH2O.
[13] I. Hagedorn, I. Stark, K. Schoene, H. Schenkel, Arzneim.-Forsch./Drug Res. 28
(1978) 2055.
[14] L. Leitis, E. Liepinis, D. Jansone, M.V. Shimanskaya, Latv. PSR Zinat. Akad. Vestis.
Kim. Ser. 6 (1984) 719.
[15] J.J. Kaminski, K.W. Knutson, N. Bodor, Tetrahedron 34 (1978) 2857.
[16] Oxford Diffraction, CrysAlis CCD and CrysAlis RED, Versions 171.32.24, 2008.
[17] G.M. Sheldrick, Acta Crystallogr., Sect. A 64 (2008) 112.
[18] A.L. Spek, J. Appl. Crystallogr. 36 (2003) 7.
Acknowledgments
ˇ ´
´
We thank Mrs. S. Matecic-Mušanic (Brodarski Institute, Zagreb,
Croatia) for the TGA, DTA and DSC measurements.
This work was supported by the Ministry of Science, Education
and Sports of the Republic of Croatia (Grant Nos. 108-1193079-
3070 and 119-1193079-1084).
[19] M. Nardelli, J. Appl. Crystallogr. 28 (1995) 659.
[20] (a) L.J. Farrugia, J. Appl. Crystallogr. 30 (1997) 565;
(b) C.F. Macrae, P.R. Edgington, P. McCabe, E. Pidcock, G.P. Shields, R. Taylor, M.
Towler, J. van de Streek, J. Appl. Crystallogr. 39 (2006) 453.
[21] M. Boccio, A. Sayago, G. Asuero, Int. J. Pharm. 318 (2006) 70.
[22] F.H. Allen, O. Kennard, D.G. Watson, L. Brammer, A.G. Orpen, J. Chem. Soc.,
Perkin Trans. II (1987) S1.
Appendix A. Supplementary data
ˇ
ˇ
´
´
´
[23] (a) R. Odzak, I. Halasz, S. Tomic, D. Matkovic-Calogovic, Acta Crystallogr., Sect.
E 62 (2006) o2423;
CCDC 882946, 883098 and 883099 contain the supplementary
crystallographic data for this paper. These data can be obtained
free of charge via http://www.ccdc.cam.ac.uk/conts/retriev-
ing.html, or from the Cambridge Crystallographic Data Centre, 12
Union Road, Cambridge CB2 1EZ, UK; fax: (+44) 1223-336-033;
or e-mail: deposit@ccdc.cam.ac.uk. Supplementary data associated
with this article can be found, in the online version, at http://
dx.doi.org/10.1016/j.poly.2012.07.077.
´
´
´
ˇ
(b) M. Jukic, A. Hergold-Brundic, M. Cetina, A. Nagl, J. Vorkapic-Furac, Struct.
Chem. 14 (2003) 597;
(c) C.D. Bustamante, R.J. Staples, Z. Kristallogr. NCS 214 (1999) 141;
(d) W. Van Havere, A.T.H. Lenstra, H.J. Geise, G.R. Van Den Berg, H.P. Benschop,
Acta Crystallogr., Sect. B 38 (1982) 1635;
(e) W. Van Havere, A.T.H. Lenstra, H.J. Geise, G.R. Van Den Berg, H.P. Benschop,
Acta Crystallogr., Sect. B 38 (1982) 2516.
´
[24] B. Foretic, N. Burger, Monatsh. Chem. 135 (2004) 261.
[25] (a) J. Gi Jee, O.C. Kwun, Bull. Korean Chem. Soc. 5 (1984) 44;
(b) R.A. Lalancette, W. Furey, J.N. Costanzo, P.R. Hemmes, F. Jordan, Acta
Crystallogr., Sect. B 34 (1978) 2950;
(c) P. Hemmes, J.N. Costanzo, F. Jordan, J. Phys. Chem. 82 (1978) 387.
[26] K. Schoene, E.M. Strake, Biochem. Pharmacol. 20 (1971) 1041.
References
ˇ
´
[27] (a) V. Mohacek Grošev, B. Foretic, O. Gamulin, Spectrochim. Acta A 78 (2011)
1376;
[1] F. Mancin, P. Tecilla, U. Tonellato, Eur. J. Org. Chem. (2000) 1045.
[2] (a) F. Mancin, P. Tecilla, U. Tonellato, Langmuir 16 (2000) 227;
(b) A.K. Yatsimirsky, P. Gomez-Tagle, S. Escalante-Tovar, L. Ruiz-Ramirez,
Inorg. Chim. Acta 273 (1998) 167;
(c) R. Breslow, D. Chipman, J. Am. Chem. Soc. 87 (1965) 4195.
[3] C.J. Milios, T.C. Stamatatos, S. Perlepes, Polyhedron 25 (2006) 134.
(b) T. Stepanenko, L. Lapinski, M.J. Novak, L. Adamowicz, Vib. Spectrosc. 26
(2001) 65.
[28] R. Middleton, J.R. Thornback, G. Wilkinson, J. Chem. Soc., Dalton Trans. (1980)
174.
[29] K. Nakamoto, Infrared and Raman Spectra of Inorganic and Coordination
Compounds, fifth ed., John Wiley and Sons, New York, 1997. p. 266.
[30] P.J. Morando, V.I.E. Bruyere, Miguel A. Blesa, Transition Met. Chem. 8 (1983)
99.
ˇ
´
´
´
[4] (a) Z. Koricanac, K. Kariljkovic-Rajic, B. Stankovic, Talanta 37 (1990) 535;
(b) N. Burger, V. Hankonyi, Anal. Lett. 25 (1992) 1355;
´
(c) N. Burger, V. Hankonyi, Z. Smeric, Z. Phys. Chem. (Leipzig) 271 (1990) 787;
(d) N. Burger, V. Hankonyi, Polyhedron 5 (1986) 663;
(e) N. Burger, Z. Smeric´, V. Hankonyi, Acta Pharm. Jugosl. 36 (1986) 15;
(f) N. Burger, V. Karas-Gašparec, Talanta 31 (1984) 169;
(g) N. Burger, V. Karas-Gašparec, Talanta 28 (1981) 323;
(h) V. Hankonyi, V. Ondrušek, V. Karas-Gašparec, Z. Binenfeld, Z. Phys. Chem.
(Leipzig) 251 (1972) 280.
[31] (a) R.B. Lanjewar, M.R. Lanjewar, Int. J. Knowl. Eng. 3 (2012) 61;
(b) D.B. Soria, P.J. Aymonino, Spectrochim. Acta A 55 (1999) 1243;
(c) R.B. Lanjewar, S. Kawata, S. Kitagawa, M. Katada, J. Therm. Anal. 50 (1997)
375;
(d) R.B. Lanjewar, S. Kawata, T. Nawa, S. Kitagawa, A.N. Garg, M. Katada,
Thermochim. Acta 287 (1996) 111;
(e) E.E. Sileo, M.G. Garcia, A. Posse, P.J. Morando, M.A. Blesa, Polyhedron 6
(1987) 1757;
[5] (a) B. Foretic´, I. Picek, V. Damjanovic´, D. Cvijanovic´, D. Milic´, J. Mol. Struct.
1019 (2012) 196;
(b) B. Foretic´, I. Picek, I. Ðilovic´, N. Burger, Inorg. Chim. Acta 363 (2010) 1425.
(f) M.M. Chamberlain, A.F. Greene Jr., J. Inorg. Nucl. Chem. 25 (1963) 1471.
[32] J.M. Malin, R.C. Koch, Inorg. Chem. 17 (1978) 752.
[33] N.Y. Murakami Iha, H.E. Toma, Inorg. Chim. Acta 81 (1984) 181.
´
[6] M. Jokanovic, M. Prostran, Curr. Med. Chem. 16 (2009) 2177.
[7] (a) E. Abele, R. Abele, E. Lukevics, Chem. Heterocycl. Compd. 39 (2003) 847;
ˇ
ˇ
(b) S.K. Bognar, B. Foretic´, Z. Vukelic´, T. Gulin, D. Jezek, Croat. Chem. Acta 81
(2008) 67.