Spectroscopic and electron-transfer reactivity studies of bulky bis(N-cycloalkyl-3,5-tBu2-salicylaldiminato)copper(II) complexes: Generation of uncoordinated and coordinated phenoxyl radicals

Article abstract of DOI:10.1016/j.saa.2010.02.016

This work summarizes the results of our studies on the spectral, magnetic, electrochemical and chemical redox properties of N-cycloalkyl-3,5-^tBu₂-salicylaldimine ligands [cycloalkyl = cyclo-C₅H₉(HL₁), cyclo-C₆H₁₁ (HL₂), cyclo-C₇H₁₃ (HL₃), cyclo-C₈H₁₅ (HL₄), 1-adamantyl (HL₅), 2-adamantyl (HL₆)] and their copper(II) complexes (1-6). The compounds have been characterized by IR, ¹H NMR, UV-vis, EPR spectroscopy, electrochemical and magnetic susceptibility measurements. The geometry of 1-6, according to their EPR (g_II and g_II/A_II) and visible spectral data, exhibit a significant amount distortion from slightly distorted square-planar to pseudo-tetrahedral. The cyclic voltammetric studies of 1-6 reveal that as the extent of the tetrahedral distortion of Cu^II center increases on going from 1 to 5, the values of Cu^II/Cu^I potentials became more negative. The compounds have been oxidized electrochemically and chemically and the generated relatively stable uncoordinated phenoxyl [HL_x]^{{radical dot}+} and coordinated Cu(II)-phenoxyl radical [1-6]^{{radical dot}+} species have been characterized by UV/vis and EPR spectroscopy.

Full text of DOI:10.1016/j.saa.2010.02.016

Spectrochimica Acta Part A 76 (2010) 99–106
Contents lists available at ScienceDirect
Spectrochimica Acta Part A: Molecular and
Biomolecular Spectroscopy
journal homepage: www.elsevier.com/locate/saa
Spectroscopic and electron-transfer reactivity studies of bulky
bis(N-cycloalkyl-3,5-tBu2-salicylaldiminato)copper(II) complexes:
Generation of uncoordinated and coordinated phenoxyl radicals
Veli T. Kasumova,∗, Fevzi Köksalb, Ays¸ egül Kutluaya
a Department of Chemistry, Harran University, Osmanbey, 63300 S¸ anlıurfa, Turkey
b Department of Physics, Ondokuz Mayıs University, 55139 Kurupelit, Samsun, Turkey
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 2 January 2010
Received in revised form 10 February 2010
Accepted 15 February 2010
This work summarizes the results of our studies on the spectral, magnetic, electrochemical and
chemical redox properties of N-cycloalkyl-3,5-tBu2-salicylaldimine ligands [cycloalkyl = cyclo-C5H9(HL1),
cyclo-C6H11 (HL2), cyclo-C7H13 (HL3), cyclo-C8H15 (HL4), 1-adamantyl (HL5), 2-adamantyl (HL6)] and
their copper(II) complexes (1–6). The compounds have been characterized by IR, 1H NMR, UV–vis,
EPR spectroscopy, electrochemical and magnetic susceptibility measurements. The geometry of 1–6,
according to their EPR (gII and gII/AII) and visible spectral data, exhibit a significant amount distor-
tion from slightly distorted square-planar to pseudo-tetrahedral. The cyclic voltammetric studies of
1–6 reveal that as the extent of the tetrahedral distortion of CuII center increases on going from 1
to 5, the values of CuII/CuI potentials became more negative. The compounds have been oxidized
Keywords:
Bis[N-cycloalkyl-3,5-tBu2-
salicylaldiminato]Cu(II)
complexes
Phenoxyl radicals
Spectroscopy
Redox reactivity
electrochemically and chemically and the generated relatively stable uncoordinated phenoxyl [HLx]•
+
and coordinated Cu(II)-phenoxyl radical [1–6]• species have been characterized by UV/vis and EPR
+
spectroscopy.
© 2010 Elsevier B.V. All rights reserved.
1. Introduction
ization or a variety of asymmetric oxidation, epoxidation and other
reactions.
In our previous studies [1–4] on coordination chemistry of ster-
ically hindered salicylaldimine (SA), N-2,6-tBu2-1-hydroxyphenyl-
X-salicylaldimines, we have found that chemical oxidation of their
Co(II), Ni(II), Pd(II) and Zn(II) complexes readily generate periph-
erally bounded stable phenoxyl radical complexes. However, in
the oxidation of Cu(II) complexes with these ligands, without
detection of radical species, the formation of unprecedented oxida-
tive C–C coupled tetradentate salicylaldimine copper(II) complexes
have been observed [2,3]. One of the unique properties of above
complexes is that they easily generate peripherally bonded stable
M(II)–phenoxyl radical complexes upon chemical oxidation [1,4].
The discovery of Cu(II)–phenoxyl radical centers in the active site
of CuII-containing enzymes (galactose and glyoxal oxidases) has
stimulated the development of various transition metal chelates
capable to form directly coordinated MII,III–phenoxyl radical com-
plexes [5–7]. In addition, recent studies have revealed that bulky
tert-butylated bi-, tri- and tetradentate phenoxyimine type ligand
complexes of group 4 metals [8,9] and p-block metals [10–12] are
one of the versatile and highly active catalysts for olefin polymer-
To continue our study on the radical ligand complexes, recently
we have interested in the complexation behavior of some bulky N-
aryl(alkyl)-3,5-tBu2-salicylaldimine (3,5-DTBSA) ligands and redox
chemistry of their transition metal(II) chelates [13–15]. Our studies
demonstrated that some complexes of Cu(II), Pd(II) and Co(II) with
the above bulky ligands upon chemical oxidation affords directly
coordinated phenoxyl radical complexes stable at ambient tem-
perature. While the redox chemistry of various bis(SA)Cu(II) is well
studied, the contradictory relationships for the CuII/CuI couples in
the series of bis(N-alkyl-SA)Cu(II) and bis(N-cycloalkyl-SA)Cu(II)
complexes were reported in the literature [16].
In the present paper we report the synthesis, spectroscopic,
magnetic and redox behaviors of a new series of bis[N-cycloalkyl-
3,5-tBu2-salicylideneiminato]copper(II) complexes, Cu(Lx)2 (X),
derived from 3,5-di-tert-butylsalicylaldehyde and cycloalky-
lamines [RNH2, R = cyclo-C5H9 (HL1), cyclo-C6H11 (HL2), cyclo-
C7H13 (HL3) and cyclo-C8H15 (HL4)]. The Cu(Lx)2 with x = 1, 2, 3
and 4 are abbreviated as 1, 2, 3 and 4, respectively. To investi-
gate the dependence of the CuII/CuI potentials on the bulkiness
of cycloalkyl groups, we also prepared recently reported bulkier
N-(1-adamantyl)-3,5-DTBSA and N-(2-adamantyl)-3,5-DTBSA lig-
ands Cu(II) complexes [15], abbreviated as 5 and 6, respectively
(Scheme 1).
∗
Corresponding author. Tel.: +90 414 3440020x1265; fax: +90 414 3440051.
E-mail address: vkasumov@harran.edu.tr (V.T. Kasumov).
1386-1425/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.saa.2010.02.016

100
V.T. Kasumov et al. / Spectrochimica Acta Part A 76 (2010) 99–106
2. Experimental
2.1. Methods
acteristics
for
N-cyclohexyl-3,5-di-tert-butylsalicylaldimine
(HL2), N-(1-adamantyl)-3,5-tBu2-salicylaldimine (HL5) and N-
(2-adamantyl)3,5-tBu2-salicylaldimine (HL6) ligands and their
Cu(II) complexes have been reported previously [15]. HL1: yield:
88%. M.p. 97 ◦C. C20H31NO (301.47): calcd. C, 79,68; H, 10.37; N,
4.64%. Found: C, 79.27; H, 9.54; N, 4.34%. IR (KBr pellet, ꢀ [cm−1]):
The C, H, N elemental analyses were performed on a LECO
CHNS-932 model microanalyses. UV–vis spectra were recorded
on a PerkinElmer Lambda 25 spectrometer. IR spectra obtained
on a Perkin-Elmer FTIR spectrometer using KBr pellet. 1H NMR
spectra were recorded on a Bruker Spectro spin Avance DPX-400
Ultra Shield Model NMR with Me4Si as an internal standard in
CDCl3 solutions. The room temperature magnetic susceptibility
was measured by using a Sherwood Scientific magnetic balance
and the diamagnetic corrections were evaluated using Pascal’s
constant. X-Band ESR spectra were taken on a Varian E109 C
spectrometer at 300 and 173 K. The field and frequency calibra-
tion were made using DPPH (g = 2.0037). Cyclic voltammograms
(CV) of all complexes have been recorded in dimethylformamide
(DMF) solutions containing ca.10−3 M 1–6 and 0.05 M tetraethy-
lammonium tetrafloroborate (Et4NBF4) as supporting electrolyte.
A three-electrode cell was used which was equipped with a plat-
inum working and counter electrodes and an Ag/AgCl reference
electrode. The potentials are referred to Ag/AgCl in the potential
range +2.0 to −2.0 V. All solutions were purged with N2 for about
3 min prior to each measurement.
t
2867–2963 (C–H of –CH2– and Bu), 1627 (CH N), 1594 (–C C–),
around 2600 (intramolecularly H-bonding OH). HL3: yield: 78%.
M.p. 73–74 ◦C. C22H35NO (329.52) calcd. C, 80.19; H, 10.71; N,
4.25%. Found: C, 80.35; H, 10.68; N, 4.42%. IR (KBr pellet, ꢀ [cm−1]):
2862–2967 (C–H of –CH2 and tBu), 1631 (CH N), 1593 (–C C–),
at ca. 2635 (intramolecularly H-bonding OH). HL4: yield: 86%. M.p.
64–65. C23H37NO (343.55): calcd. C, 80.41; H, 10.86; N, 4.08%.
Found: C, 79.86; H, 11.04; N, 3.95%. IR (KBr pellet. ꢀ [cm−1]): 1627
(CH N), 1595 (–C C–), 2620 (OH, a broad band), 2906–2970 (C–H
of –CH2– and tBu).
2.4. Synthesis of complexes: 1, 3, 4
To
a
hot ethanol solution (80 cm3) of HLx (1.5 mmol),
Cu(OAc)2·H2O (0.75 mmol) in 10 cm3 methanol was added. The
mixture was boiled under reflux for ca. 2 h and the reaction volume
reduced to 25–30 cm3 and left for crystallization at room tem-
perature (r.t.). The precipitate green microcrystals were collected
by filtration, washed with cold methanol, n-hexane and dried at
60–70 ◦C. 1: yield: 67%. M.p. 183–185 ◦C, elemental Anal.: calcd.
for C40H60N2O2Cu(664.46): C, 72,31; H, 9.10; N, 4.23%. Found: C,
71.12; H, 9.41; N, 4.3 1%. IR (KBr pellet, ꢀ [cm−1]): 2867–2963
2.2. Materials
All commercial chemicals and solvents were of spectroscopic
and HPCL grade and were used without further purification.
N,N-Dimethylformamide (DMF), dimethylsulfoxide (DMSO),
acetonitrile (MeCN), chloroform, ethanol, methanol, 2,4-di-tert-
butylphenol, copper(II) acetate monohydrate, cyclo-pentylamine,
t
(C–H of –CH2– and Bu), 1621 (CH N), new bands at 1542, 1533
(C C–C O), 449, 492, 536, 568 (Cu–N, Cu–O). Magnetic moment,
ꢁeff = 1.76 ꢁB at 285 K. 3: yield: 69%. M.p.188–189 ◦C; Calcd. for
C
44H68N2O2Cu (720.57): C, 73,34; H, 9.51; N, 3.88%. Found: C,
74.13; H, 9.31; N, 4.16%. IR (KBr pellet, ꢀ [cm−1]): 2867–2963
cyclo-hexylamine,
cyclo-heptylamine,
cyclo-octylamine,
1-
t
adamantylamine and 2-adamantylamine hydrochloride obtained
from Aldrich Chemical Co, Inc., and were used without further
purification. The 3,5-tBu2-salicylaldehyde was synthesized by the
reported procedure [15]. The HLx ligands and their 1–6 complexes
were synthesized as described before [13–15].
(C–H of –CH2– and Bu), 1622 (CH N), new bands at 1549, 1533
(C C–C O), 449, 497, 540, 569, 578 (Cu–N, Cu–O). ꢁeff = 1.91 ꢁB
at 287 K. 4: m.p. 144–146 ◦C, Calcd. for C46H72N2O2Cu (748.62):
C, 73,80; H, 9.69; N, 3.74%. Found: C, 73.62; H, 9.74; N, 3.65%. IR
(KBr pellet, ꢀ [cm−1]): 2864–2956 (C–H of –CH2– and Bu), 1620
t
(CH N), new bands at 1548, 1533 (C C–C O), 497, 415, 539, 568,
579 (Cu–N, Cu–O). ꢁeff = 1.83 ꢁB at 289 K.
2.3. Synthesis of ligands: HL1, HL3, HL4
N-Cyclopentyl-3,5-di-tert-butylsalicylaldimine
(HL1),
3. Results and discussion
N-cycloheptyl-3,5-di-tert-butylsalicylaldimine (HL3) and N-
cyclooctyl-3,5-di-tert-butylsalicylaldimine (HL4) were prepared
as previously reported [13]. All salicylaldimines, HLx, were
prepared in high yields (80–96%) via condensation of the 3,5-
di-t-butylsalicylaldehyde with the corresponding cycloalkyl
amines (1:1 molar ratio) in methanol/ethanol at reflux for
8–10 h in the presence of 3–5 drops of formic acid and
recrystallized from methanol, and air-dried. The crude yel-
low viscous oil of HL4, on standing at 5–10 ◦C for 10 days,
was converted into crystal state. Analytical and spectral char-
The complexes 1, 3 and 4 and the corresponding ligands (HL1,
HL2 and HL3) were synthesized for the first time, whereas 2, 5 and
6 have been reported recently by us [13,15]. All of the N-cycloalkyl-
3,5-tBu2-salicylaldimines were prepared in high yields as a yellow
viscous oils (HL4) or crystalline solids (HL1, HL2, HL3, HL5, HL6).
Satisfactory elemental analyses were obtained for all of HL1, HL3,
HL4 and their 1, 3, 4 chelates. We also tried to prepare NiII, MnII,
CoII, VOIV and ZnII complexes with HLx by using the method similar
to that for preparation of CuII. However, in spite of our repeated
attempts, we have failed to prepare their complexes under above
conditions.
3.1. IR and 1H NMR spectra
The selected IR spectral data of HL1, HL3, HL4 and correspond-
ing 1, 3 and 4 compounds and their assignments are presented
in Section 2. The IR spectra of all ligands and complexes exhibit
sharp peaks in the range of 2860–2970 cm−1 attributable to the
C–H stretching frequencies (ꢀsym and ꢀasym) of the methylene and
methyl groups. The IR spectrum of HL4, obtained as a yellow viscous
oil, exhibits strong sharp peaks at 1684, 1632 and ca. 2630 cm−1
assigned to C O (in quinod tautomer forms of HL4), CH N and
,
Scheme 1. Structural presentations of the 1–6 complexes.

V.T. Kasumov et al. / Spectrochimica Acta Part A 76 (2010) 99–106
101
Table 1
Electronic spectral data of HLx ligands and 1–6 complexes in various solvents. Molar extinction coefficients (log ε) are given in parenthesis.
Compound
Solvent
Electronic spectral ꢂ/nm (log ε, M−1 cm−1
)
HL1
EtOH
CH3CN
EtOH
CH3CN
EtOH
CH3CN
CHCl3
EtOH
CH3CN
DMF
234(4.3), 260(4.3), 317(3.8), 326(3.9), 412(2.2)
225(4.50), 261(4.22), 326(3.71), 410(2.05)
225(4.22), 231(sh) (4.19), 260(3.97), 327(3.51), 415(2.24)
224(4.43), 260(4.13), 326(3.65), 414(2.16)
226(4.3), 260(4.1), 328(3.6), 415(2.4)
HL2
HL3
1
260(3.97), 325(3.27), 412(2.00)
273a(4.5), 326(4.2), 376(4.1), 480a(3.1), 660a(2.3)
233(4.6), 247(4.5), 276(4.3), 326(3.9), 378(3.9), 480a(2.7), 660a(2.3)
274(4.52), 326(4.24), 376(4.17), 490a(2.83), 650a(2.40)
277(4.87), 328(4.65), 380(4.66), 480a(3.01), 650a(2.43)
278a, 319, 380, 722
DMF + oxid.
2
3
4
CHCl3
EtOH
CH3CN
DMF
DMF + oxid.
277a, 328(4,3), 380(4.3), 482(2.83), 650a(2.3)
233(4.3), 247(4.6), 276(4.2) 328(3.9), 379(3.1), 480a(3.9), 665(1.9)
274(4.52), 326(4.24), 376(4.17), 490a(2.83), 660a(2.40)
276(4.33), 329(4.03), 381(4.05), 480a, 660(2.37)
279a, 313, 375, 714
CHCl3
EtOH
CH3CN
DMF
DMF + oxid.
268a, 278a, 328(4.2), 379(4.1), 660a(2.3)
233(4.3), 247(4.6), 276(4.2) 328(3.9), 379(3.1), 480a(3.9), 665(1.9)
274(4.34), 327(4.08), 377(4.00), 480a(2.79), 660a(2.28)
276(4.58), 329(4.23), 380(4.26), 480a(2.88), 660a(2.38)
280a, 318, 377, 719
CHCl3
EtOH
CH3CN
DMF
DMF + oxid.
268a, 278a, 328(4.2), 379(4.1), 487(2.82), 660a(2.3)
233(4.6), 247(4.5), 276(4.2), 328(3.9), 381(3.9), 480a(2.8), 670a(2.3)
275(4.49), 328(4.22), 376(4.17), 477a(2.85), 660a(2.34)
276(4.33), 330(4.03), 381(4.02), 480a(2.82), 670a(2.40)
279a, 314, 376, 722
5
6
CHCl3
CH3CN
DMF
262(5.0), 280a, 335(3.9), 445a, 526(2.5), 800a(2.2)
263, 329(4.1), 382(3.9), 526(2.5), 800a(2.3)
271(4.44), 278a, 337(3.81), 385 (3.78), 536(2.69), 800a(2.19)
ꢂmax (A) shifted from 748(0.54b) to 798(0.41)
DMF + oxid.
CHCl3
CH3CN
DMF
268a, 331(4.7), 372(4.6), 480a(2.9), 680a(2.3)
275(4.49), 328(4.22), 376(4.17), 520a(2.47), 690(2.40)
274(4.27), 332(3.95), 371(3.99), 388a, 490a(2.69), 680(2.26)
285c, 310c, 330c, ꢂmax (A) shifted from 745 (0.19b) to 795(0.20)
DMF + oxid.
a
Shoulder.
Absorbance.
Very weak shoulder.
b
c
OH (intramolecularly H-bonded) stretching frequencies, respec-
tively, suggesting the existence of both enol and keto-tautomer
forms in oil samples [17,18]. On the other hand, IR spectrum of
crystalline samples of this ligand exhibit sharp peaks at 1627 and
1595 cm−1, assigned to ꢀ (CH N) and benzene ꢀ (C C) vibrations,
respectively, which are typical for enol-imine tautomer forms. The
ꢀ (CH N) of 1, 3 and 4 appear at 1620–1622 cm−1 and these bands
are shifted by about 6–9 cm−1 to lower energy regions compared
to the free ligand (HLx). A broad shoulder, observed in the range
2450–2800 cm−1 due to intramolecularly H-bonded ꢀ (OH), and
sharp peaks at 1593–1597 cm−1, detected in the spectra of HLx free
ligands, disappears on the complexation. These changes suggest
that the ligands are coordinated via imine nitrogen and deproto-
nated OH group oxygen atoms to the Cu(II) center. New bands,
which are absent in the spectra of all free ligands, were observed at
1547 and 1533, 1549 and 1533, 1548 and 1532 cm−1, in the spectra
of 1, 3 and 4, respectively, attributable to phenolic C–O links which
attains a considerable amount of partial double bond character in
these complexes [19]. These IR spectral paterns were detected in
the spectra of all 1–6 complexes. The appearance of medium inten-
sity a new bands in the ranges 536–569 and 449–515 cm−1, which
are absent in the spectra of free ligands, may be assigned to the ꢀ
(Cu–O) and ꢀ (Cu–N) modes [17,19], respectively. The 1H NMR spec-
tra of HL1–HL6 ligands showed multiplets in the ranges 1.53–1.88
and 3.14–3.28 ppm which are assigned to –CH2– and –CH– groups
protons of the cyclic ring backbone, respectively. The CH N and
O–H groups protons resonances appeared as singlet peaks at ca.
8.30 and 14.15 ppm, respectively. Two doublets, observed at around
7.36 and 7.07, are assigned to metha-coupled salicylidene rings
C4-H and C6-H protons resonances. The protons of –C(CH3)3 groups
appeared as two singlet signals at 1.30 and 1.45 ppm. Thus, the
formation of the presented new Schiff bases and their copper(II)
complexes along with microanalytical data also was confirmed by
IR and 1H NMR spectroscopic examination.
3.2. Electronic spectra
To examine the influence of polarity of solvents on the chem-
ical oxidation reactivity, the electronic spectra of 1–6 complexes
are recorded in various coordinated and non-coordinated solvents.
The electronic spectral data of HLx and 1–6 are summarized in
Table 1. The absorption spectral data obtained for HLx in EtOH and
CH3CN are very similar to each other and exhibit bands appeared
in the 224–327 and 410–415 nm regions which can be assigned to
intraligand ␲ → ␲* and n → ␲* transitions in phenol-imine struc-
tures, and n → ␲* transitions in dipolar zwitterionic or keto-amine
tautomer structures, respectively [15,18,20]. The electronic spectra
of all 1–4 and 6 complexes in coordinating and less coordinating
solvents (CHCl3, EtOH, MeCN and DMF) in UV and visible regions
are practically identical (Fig. 1) indicating that the possible adduct
formation even in strong coordinating DMF did not take place. As
can be seen from Table 1 and Fig. 1, their spectra besides similar
intraligand bands in the 268–388 nm region, also exhibit identi-
cal absorptions at ca. 460–480 nm (ε = 500–700 M−1 cm−1), which
can be assigned to phenolate (p )-to-Cu(d ) or to copper(d )-to-
␲
␲
␲
penolate CT transitions [21,22]. The low energy band appeared as

102
V.T. Kasumov et al. / Spectrochimica Acta Part A 76 (2010) 99–106
Fig. 1. UV–vis spectra for 1–6 complexes in DMF; 4ꢀ-spectrum of diluted sample of 4, inset: UV spectra of 1, 4 and 5 complexes in DMF.
Table 2
Solid state and solution EPRa spectral data for 1–6 complexes.
Complex
Solid state spectra
Solution spectra
b
c
1–6
1
2
3
4
gII (AII)
g
giso
g
gII
Aiso
AII
A
⊥
gII/AII cm
136
147
151
159
G
⊥
⊥
gav = 2.052
2.226(149)
2.226(160)
gav = 2.059
2.286 (118)
2.248 (140)
2.118
2.121
2.126
2.124
2.132
2.121
2.061
2.059
2.063
2.067
2.065
2.055
2.232
2.244
2.246
2.257
2.275
2.236
63.2
67.3
63.5
61.7
61.4
61.0
164.4
153.2
149.5
142.1
119.6
145.7
12.7
24.4
20.5
21.5
31.6
18.7
3.9
4.2
4.1
3.9
4.4
4.7
2.047
2.045
5
6
2.062
2.062
190
154
a
A values are in 10−4 cm−1
.
b
c
Estimated error in g = 0.001.
Estimated error in A = 0.5 × 10−4 cm−1
.
a broad shoulder around 660 nm (ε = 80–269 M−1 cm−1) (Table 1
and Fig. 1) in the spectra of 1–4 and 6, assigned to the enve-
is typical for axial symmetric monomeric copper(II) complexes
exhibiting a tetragonal geometry and indicating that the unpaired
2
2
lope of three possible d–d transitions, 2A1(dx
–
2) → A (dz2),
2
2
electron to lie in the dx
–
ground state in these complexes
y
y
1
2
2
2
2
2
2A1(dx
–
)
→ E1(dxz yz
,
) and 2A1(dx
–
y
2) → B (dxy), in tetra-
[24,26,28]. No half-field ꢃMs = 2 signals typical for binuclear
triplet-state species were observed. The EPR spectra of polycrys-
talline samples 1 and 4 exhibit only one exchange-averaged broad
isotropic signal (Fig. 2a) with gav = 2.052 and gav = 2.059 and line
width ꢃHpp of 216 and 82 G, respectively, which is attributable to
dipolar interaction between crystallographycally non-equivalent
Cu(II) centers in the solid [28]. This line broadening is probably due
to insufficient spin-exchange narrowing toward the coalescence
of four CuII(I = 3/2) hyperfine lines to a single line, meaning that
intermolecular spin-exchange interaction energies (J) are as small
y
2
hedrally distorted square-planar geometry of Cu(II) compounds
[21,22]. For 5, which bears more bulkier 1-adamantyl group, the
d–d bands exhibit significantly red shifts to 536 (490) and 800 (sh,
220) nm, indicating that the arrangement of Cu(II) center in this
complex is a pseudo-tetrahedral geometry [23].
3.3. Magnetic measurements
The r.t. ꢁeff values for 1, 2, 3 and 4 are 1.76, 1.82, 1.91 and
1.83 ꢁB. These values are close to those obtained for 5 (1.81 ꢁB)
and 6 (1.83 ꢁB) complexes bearing bulky 3,5-di-tert-butylated
salicylic moiety and bulkier adamantyl groups attached to the
imine nitrogen [13,15]. The ꢁeff values of 1, 2, 4–6 bearing
larger volume N-cycloalkyl (adamantly) and bulky 3,5-di-tert-
butyl substituents are smaller than those (1.90–1.92 ꢁB) founded
for the N-branched-alkylsalicylaldiminato copper(II) complexes
which exhibit increasing distortion from planar towards pseudo-
tetrahedral geometry [23–26]. These results contradict with the
criterion stating that the degree of tetrahedral distortion from pla-
narity in bis-bidentate Schiff base copper(II) complexes increases as
the size of substituents increases on the imine nitrogen [23,24,26].
Similar situations have been also observed in other related com-
plexes [27].
3.4. EPR investigation
3.4.1. Powder spectra
The polycrystalline powder EPR spectral spin Hamiltonian
parameters of 1–6 complexes, estimated from r.t. experimental
Fig. 2. X-Band EPR spectra powder samples of 4 (a), 3 (b) and 2 (c) at r.t.
spectra, are listed in Table 2. The observed g-values, gII > g > 2.04
⊥

V.T. Kasumov et al. / Spectrochimica Acta Part A 76 (2010) 99–106
103
Measurements of powder EPR spectrum of 5 over the temperature
range 298–140 K revealed that without any changes in line width
and g-factor only a little increase in the intensity of EPR signal was
detected. However, significantly increase in the intensity of this
spectrum was detected on cooling below ca.135 K (Fig. 4), which
suggests the existence of ferromagnetic interactions between adja-
cent Cu(II) centers at very lowered temperatures.
3.4.2. Solution spectra
The r.t. EPR spectra of 1–6 in CHCl3 solutions generally show
three resolved patterns of four copper(II) hyperfine lines, where
first and second low-field components are overlapped to one
broad peak. No 14N-shfs resolutions on the high-field copper
components were observed. The frozen-solution EPR spectra of
1–6 at 173 K show typical axial pattern with gII > g > 2.04, typi-
⊥
2
2
cal for copper(II) complexes with a dx
–
ground state (S = 1/2)
y
[28] (Table 2). As expected for 5, the lowest AII (119.6 G) and
the highest gII (2.286) values, due to considerable tetrahedral
distortion of Cu(II) geometry as a result of steric interactions
between bulky N-1-adamantyl and di-tert-butyl groups salicylic
moieties, were observed. As the size of the cycloalkyl groups on the
imines nitrogen increase, along with the distortion from square-
planar to pseudo-tetrahedral geometry, the increase of gII and
gII/AII values were also detected (Table 2). The exchange interac-
tion parameter, G, empirical tetrahedral distortion index, gII/AII,
and in-plane sigma bonding parameter, ˛2, values for all 1–6
complexes are calculated. G = (gII − 2.0023)/(g⊥ − 2.0023), which
measures the exchange interaction between the copper centers
in a polycrystalline solid, have been calculated for all spectra.
According to Hathaway [28], for axially symmetric spectra with
the lowest g-value > 2.04, if G ≥ 4, the interexchange coupling
between the Cu(II) centers will be negligible, but a value of G < 4
indicates the existence of exchange interaction in the solid cop-
per(II) complexes. The calculated G values, 3.9–4.7, suggest that
the exchange coupling effects between copper(II) centers are not
operative in the present 1–6 complexes [28,31]. It is interest-
ing that with the increasing of tetrahedral distortion in Cu(II)
geometry on going from 1 to 5, decreasing in the value of in-
plane ␴-bonding parameter, ˛2 (0.75–0.67), evaluated using the
expression ˛2 = −AII/0.036 + (gII − 2.0023) + 3/7(g⊥ − 2.0023) + 0.04
[32], have been founded. For presented complexes, an increase in
tetrahedral distortion was accompanied by a concomitant decrease
in covalency ˛2, related to structural deviations from square-planar
geometry towards distorted tetrahedral symmetry (as supported
by ESR spectra), rather than increased metal-ligand covalency [33].
Another convenient empirical index of tetrahedral distortion,
f(˛) = gII/AII, is regarded as an extent of deviation from idealized
planar geometry. It is well known that the values of gII/AII, ranging
from 105 to 135 cm are indicative for square-planar structures and
above this are typical for out-of-plane distortions towards tetrahe-
dral geometry [34]. The values obtained for the present complexes,
gII/AII = 136, 147, 151, 154, 159 and 190 cm observed for 1, 2, 3, 6,
4 and 5, respectively (Table 2), suggest a small to moderately dis-
torted geometry for 1, 2, 3, 6 and 4 complexes and considerable
tetrahedrally distorted symmetry for 5.
Fig. 3. EPR spectra of powdered samples of 6 at 250 K.
as the order of 0.1 cm−1 [29]. The powder spectra of all other 2, 3,
5 and 6 complexes showed three of the four AII hyperfine compo-
nents (gII = 2.226–2.286 and AIICu = 118–161 × 10−4 cm−1) (Table 2),
while the fourth one overlapped with the g features as depicted
⊥
in Fig. 2b and c for powder spectra of 3 and 2, respectively. This
fact indicates that these complexes are monomeric in crystals with
|J| ≤ 0.02 cm−1 [30], indicating that spin exchange between adja-
cent molecules is not fast enough to average the A anisotropy in
these complexes. The exchange coupling term G, estimated from
the G = (gII − 2.0023)/(g⊥ − 2.0023) [28], for 2, 3, 5 and 6 were 5.0,
5.23, 4.75 and 4.2, respectively, suggesting that the exchange cou-
pling effects are not operative for these complexes in the solid
state [28]. While the powder spectrum of 6 at 253 K exhibits nearly
isotropic signal (Fig. 3a), the tenfold enlarged feature (Fig. 3b) of
this spectrum clearly displays the existence of unequally spaced
well resolved five components pattern in the gII region. The anal-
ysis of these spectra gII = 2.254, AII = 135 × 10−4 m−1 and gII = 2.181,
AII ∼ 140 × 10−4 cm−1 parameters are estimated for A and Aꢀ com-
plexes, respectively. When 6 was cooled from 298 to 133 K, only
the intensity of the spectrum slightly increased and the line width
and g-factors remained unchanged. EPR spectra of powder samples
of 5 were recorded in the temperature range 298–113 K (Fig. 4).
3.5. Chemical oxidations of 1–6 complexes
An EPR investigation of the oxidative behavior of 1–4 revealed
that when they are treated with 3–4-fold excess of Ce(IV) in CHCl3
at r.t., along with a color change from dark brown to greenish
orange, without detection of radical signals, a slightly decrease
in intensity of their ESR spectra were observed. The EPR exami-
nation of the oxidative reactivity of 5 and 6 complexes revealed
that upon addition of 3–4-fold excess of Ce(IV) to their CHCl3
solution along with a color change from dark brown to green, a
Fig. 4. Temperature dependence of EPR spectra of powder samples of 5.

104
V.T. Kasumov et al. / Spectrochimica Acta Part A 76 (2010) 99–106
•
Fig. 5. EPR spectra of the phenoxyl radical species at r.t. in CHCl3 and under aerobic conditions: spectrum of 5 + Ce(IV) system (a); spectra of uncoordinated (HL5 +) (b) and
coordinated Co(III)–phenoxyl radicals (c) generated by one-electron oxidation of HL5 and bis(N-cyclohexyl-3,5-DTBSA)Co., respectively.
decrease in the intensity of their EPR spectra and the appearance
of a new radical signals with g = 2.0098 and g = 2.0066, respec-
tively, were observed (Fig. 5a). The decrease in the ESR signal
intensity of the oxidized complexes probably results from an anti-
ferromagnetic coupling between an S = 1/2 Cu(II) center and an
S = 1/2 coordinated radical ligand in CuII–•OPh radical species. The
formation of M–phenoxyl radical complexes have been reported
recently for bis(N-aryl(alkyl)-3,5-tBu-salicylaldiminato)M(II) com-
plexes where M = Cu(II), Co(II) and Pd(II), and for various salen
and salan type ligands M(II)/M(III) complexes bearing di-tert-butyl
protecting •groups on the ligands [5–7,35]. Note that the oxidized
ance of a new broad maximum peak centered at ca. 700–730 nm
was detected in the spectra of the reaction mixture (Fig. 6 and
Table 2). Similar spectral changes were also observed upon two-
electron oxidation for all 1–6 complexes. The spectra of the diluted
samples of oxidized [1–6]•+ species also exhibit very similar absorp-
tions around 280, 318, 380 and 400 nm in DMF and MeCN (Fig. 6 and
Table 2). The absorption bands, appeared at around 380, 400 and
770 nm in the electronic spectra of oxidized [1–6]•+ radical species,
is very similar to those of the phenoxyl radical species which are
chemically or electrochemically generated from various bulky phe-
nolate bearing complexes [5–7,35–37]. Thus, while we have not
been able to detect the EPR signal of the oxidized 1–4 species,
their in situ UV/vis spectra are very similar with those reported for
phenoxyl radical species [35–38]. The similarity of the absorption
spectra observed in the oxidation of complexes and ligands lead us
to the conclusion that the oxidation process in 1–6 is predominantly
ligand centered.
•
+
HL1 +–HL4 species are unstable and their ESR signals can be
•
+
detected at 253 K, while the ESR spectra of the oxidized HL5 and
•
+
HL6 samples in CHCl3 display narrow (ꢃHpp = 5–6 G) unresolved
strong singlets at g = 2.0036 and 2.0046 (Fig. 5b) in air and at r.t.,
respectively, which is characteristic for phenoxyl radicals [36]. The
generation of stable M–phenoxyl radical complexes in these sys-
tem also has been demonstrated upon one-electron oxidation of
bis(N-cyclohexyl-3,5-DTBSA)Co(II) by Ce(IV) in CHCl3, in which the
EPR spectrum exhibit octet of doublets with g = 2.0027, ACo = 9.25
G and AH = 3.65 G at r.t. (Fig. 5c). The observed lower values of the
radical species as well as narrow nature of the signals suggest that
the unpaired electron is in a delocalized orbital which has almost
ligand character.
It is interesting that upon chemical oxidation of all HLx ligands
with one equivalent of Ce(IV) in MeCN and at r.t., their light yellow
color simultaneously turns to deep green, and along with the dis-
appearance of the bands at 412 nm, a new broad maximum peaks
appear at 700–720 nm (Fig. 7). While the one-electron oxidized
[HLx]• (x = 1–4) species are unstable and their dark green color
+
after about a few minutes turned to yellowish brown, the oxidized
[HL5]• species is relatively stable under aerobic conditions and
+
The oxidation process was further investigated by in situ UV/vis
spectral measurements. The UV/vis spectral changes observed at
r.t. during the chemical oxidation of 5 and 6 are shown in Fig. 6.
The spectral changes for 1–6 + Ce(IV) system were monitored by
UV/vis spectroscopy during the progress of the oxidation of 1–6 in
CH3CN and DMF solutions. Upon one-electron oxidation of 1–6 by
one equivalent of Ce(IV) along with the disappearance of a broad
shoulder at ca. 660–800 nm of the parent complexes, the appear-
at r.t. (Fig. 7). Upon one-electron oxidation of HL5 in MeCN by one
equivalent Ce(IV), its light yellow color instantly turns to dark green
and the intensity of observed broad maximum at 715 nm is rapidly
decreased at r.t. (Fig. 7). Its decay constant was estimated to be
kobs = 2.57 × 10−3 s−1 (t1/2 = 270 s). However, [5]• phenoxyl radical
+
complex is relatively stable at ambient temperature, but gradually
decomposes obeying first-order kinetics with kdec = 5.78 × 10−4 s−1
•
•
+
+
Fig. 6. Absorption spectral changes of the 5 (a) and 6 (b) Cu–phenoxyl radicals generated by the oxidation of 5 and 6 complexes, respectively, with two equiv of Ce(IV) at
r.t. in MeCN. Inset: spectra of the dilute solution samples of 5• + (a) and 6• + (b) radicals recorded in MeCN solution. The arrow indicates the direction of decreasing absorbance.

V.T. Kasumov et al. / Spectrochimica Acta Part A 76 (2010) 99–106
105
•
•
•
+
+
+
Fig. 7. Absorption spectra of HL5 (a), HL1 and HL1 (b) after one-electron oxidation with 1 equiv. of Ce(IV) at r.t. in MeCN. Inset: diluted spectrum of HL5 in MeCN.
Table 3
Cyclic voltammetric redox potentials (vs. Ag/AgCl) for 1–6 complexes in DMF.
Compound 1–6
Ligand centered oxidation Epa/Epc (V)
CuII/IIIIcouple Epa
/
pc
(V)
E1/2 (V)
CuII/I couple Epa
/
pc
(V)
CuI/o couple Epa
/
pc
(V)
1
2
3
4
5
6
1.115
1.128
1.099
1.128
1.094
1.125
–
–
−0.695/−0.745
−0.293/−0.806
−0.351/−0.812
−0.585/−0.828
−0.608/−0.967
−0.318/−0.849
−0.727/−1.355
−0.722/−1.229
−0.795/−1.196
−1.101/−1.348
–
0.503/0.183
0.425/0.161
0.313/0.168
0.525/0.276
0.466/0.252
0.343
0.293
0.241
0.491
0.359
−1.147/−1.283
(t1/2 = 1200 s) decay constant. Thus, the presented bulky ligands and
their Cu(II) complexes undergo a facile one-electron oxidation. The
fact that one-electron oxidation of ligands and their complexes
accompanied by similar spectral changes with the appearance of
absorption bands typical for phenoxyl radical species and observa-
tion of the ESR radical signals for some oxidized species, suggest
that the complexes undergo ligand centered oxidation. The prod-
ucts of the one-electron oxidation of 1–6 can be considered as
far from unity, suggesting the quasi-reversible character of these
redox-couples. As can be seen from Table 3, the reduction poten-
tial of CuII(Lx)2/[CuI(Lx)2]−1 couple, E1pc, becomes more negative in
the order 1 < 2 < 3 < 4 < 6 < 5, i.e. in a sequence of an increasing bulk-
iness of cycloalkyl groups. These results indicate that the reduction
of Cu(II) to Cu(I) in 1–6 complexes became more difficult as the sub-
stituent on imine nitrogen becomes more voluminous. Thus, the
obtained results indicate that the E1 (CuII/CuI) values of the 1–6
pc
[CuLxLx]• Cu–phenoxyl radicals.
became more negative as the bulkiness of cycloalkyl substituents
are increased from 1 to 5. This trend is in contrast with those
reported by Patterson and Holm [38], who observed that as the
size of alkyl groups increases, the values of Epc (CuII/CuI) became
more positive in a series of bis(N-alkyl-SA)copper(II) complexes.
On the other hand, our results are well agree with those reported
by Martinez and co-workers who found that the Epc became more
negative as the size of cycloalkyl groups increases in a series of
bis(N-cycloalkyl-SA)Cu(II) complexes [16,27].
+
3.6. Electrochemical behavior of 1–6
The electrochemistry of 1–6 complexes were studied by cyclic
voltammograms (CVs) in the scan rate of 0.2 V s−1in DMF solu-
tion containing 0.05 M Et4NBF4 supporting electrolyte. The CV was
recorded in the scale +2.0 V to −2.0 V vs. Ag/AgCl at r.t. under
N2. In the positive range, from 0 to +2.0 V, a quasi-reversible
Cu(II)/Cu(III) redox-couples with of E1/2 at 0.241–0.359 V and irre-
versible ligand centered oxidation occurring from 1.094 to 1.125 V
vs. Ag/AgCl were observed (Table 3). The CV scanning between 0
and −2.0 V permitted to the study of copper(II) centered reductions.
The CV recorded in the potential range 0.0 to −2.0 V vs. Ag/AgCl
for 1–6 show a similar electrochemical behavior (Fig. 8). The CV
of these complexes, runned at the scan rate of 0.2 V s−1 toward
the cathodic direction, display first quasi-reversible reductive weak
shoulder (E1pc) in the potential range −0.795 to −1.10 V and sec-
ond quasi-reversible maximum peak from −1.228 to −1.360 V
(E2pc) vs. Ag/AgCl (Table 3). Reductive responses are assigned
to CuII(Lx)2/[CuI(Lx)2]−1 and [CuI(Lx)2]−1/[Cu0(Lx)2]−2 (Lx present
coordinated HLx ligands) processes, respectively. In all cases, the
corresponding reversal anodic waves in the range ∼ −1.150 to ∼
−0.7 V were very poorly defined weak curves, indicating that the
[CuI(Lx)2]−1 species are unstable and decomposed to a Cu+ ion,
subsequently reduced to copper metal [27]. For the complexes
1–6, the anodic–cathodic peak separation value, ꢃE = Epa − Epc, in
the cathodic region, was about 150–700 mV and ipa/ipc ratio was
Fig. 8. Cyclic voltammograms recorded at a platinum electrode in DMF solu-
tion containing 0.05 M Et4NBF4: (a) complex 1 (3.3 × 10−3 M) and (b) complex 6
(1.5 × 10−3 M). Scan rate = 0.2 V/s.

106
V.T. Kasumov et al. / Spectrochimica Acta Part A 76 (2010) 99–106
4. Conclusion
[2] M.K. Guseyinova, A.A. Medjidov, V.T. Kasumov, J. Struct. Chem. 23 (1982)
115–119.
[3] A.A. Medjidov, V.T. Kasumov, H.S. Mamedov, Dokl AN USSR 283 (1981)
866–869.
[4] V.T. Kasumov, A.A. Medjidov, Russ. J. Coord. Chem. 21 (1995) 783–787.
[5] B.A. Jazdewski, W.B. Tolman, Coord. Chem. Rev. 200–202 (2000) 633–685.
[6] J. Stubbe, J.W.A. van der Donk, Chem. Rev. 98 (1998) 705–762, and refs. cited
therein.
[7] J.W. Whittaker, Chem. Rev. 103 (2003) 2347–2364.
[8] K. Kawai, T. Fujita, Top. Organomet. Chem. 26 (2009) 3–41.
[9] L. Canali, D.C. Sherrington, Chem. Soc. Rev. 28 (1999) 85–93.
[10] T. Katsuki, Adv. Synth. Catal. 344 (2002) 131–148.
[11] V.C. Gibson, S.K. Spitzmesser, Chem. Rev. 103 (2003) 283–315.
[12] D.A. Atwood, A.R. Hutchson, Y.Z. Zang, Structure and Bonding, vol. 105,
Springer-Verlag, Berlin, 2003, pp. 167–201.
In conclusion, a series of copper(II) complexes with sterically
hindered N,O-donor N-cycloalkyl-3,5-di-tert-butylaslicylaldimine
ligands have been prepared and fully characterized using various
spectroscopic techniques. The preliminary examination revealed
that while the above ligands readly form Cu(II) and Pd(II) com-
plexes, no complexation of Co(II), Ni(II), Mn(II), VO(II) and Zn(II)
ions takes place with these ligands. The results of EPR and electronic
spectral studies demonstrated that on going from cyclopentyl
to 1-adamantyl bearing complex
a distortion from square-
planar to the pseudo-tetrahedral geometry takes place. It has
been shown that N-cycloalkyl-3,5-di-tert-butyl-salicylaldimines
are redox-noninnocent ligands which can be exest in different oxi-
dation levels. Their Cu(II) complexes can be used for generation
of stable Cu(II)-phenoxyl radical species. Electrochemical studies
on the CuII(Lx)2 demonstrated that as the bulkiness of cycloalkyl
group on the imine nitrogen increases, the tetrahedral distortion
in the CuII(Lx)2 enhances and the higher distorted 1-adamantyl
bearing complex exhibits the more negative Cu(II)/Cu(I) reduction
potential. This result is in contrast to that reported for their N-alkyl-
salicylaldamine copper(II) counterparst, where it was found that
the reduction shifts to more positive values as the distortion from
planarity is enhanced.
[13] V.T. Kasumov, F. Köksal, A. Sezer, Polyhedron 24 (2005) 1203–1211.
[14] V.T. Kasumov, F. Köksal, Y. Zeren, Spectrochim. Acta Part
330–336.
A 63 (2006)
[15] V.T. Kasumov, A.I. Öztürk, F. Köksal, Polyhedron 26 (2007) 3129–3135.
[16] M.A. Martinez, R.S. Augliar, N.M. Ruvalcaba, R.C. Rosado, A.N. Vazquez, V.G.
Vidales, A.Z. Dehesa, J.M. Fernandez-G, S.J.M. Fernandez-G, J. Chem. Soc., Dalton
Trans. (2001) 2346–2352.
[17] D. Heinert, A.E. Martel, J. Am. Chem. Soc. 84 (1962) 3257–3263.
[18] P.W. Alexander, R.J. Sleet, Aust. J. Chem. 23 (1970) 1183–1190.
[19] S.J. Gruber, C.M. Harris, E. Sinn, J. Inorg. Nucl. Chem. 30 (1968) 1805–1830.
[20] G.L. Estiu, A.H. Jubert, J. Costamagna, J. Vargas, J. Molec. Struct. (Theochem.) 367
(1996) 97–110.
[21] G. Fraser, B. Bosnich, Inorg. Chem. 33 (1994) 338–346.
[22] E.I. Solomon, Inorg. Chem. 45 (2006) 8012–8025.
[23] L. Sacconi, M. Ciampolini, U. Campigli, Inorg. Chem. 4 (1965) 407–409.
[24] H. Yokoi, A.W. Addison, Inorg. Chem. 16 (1977) 1341–1349.
[25] H. Yokoi, M. Chikira, J. Am. Chem. Soc. 97 (1975) 3975–3978.
[26] H. Yokoi, Bull. Chem. Soc. Jpn. 47 (1974) 3037–3040.
Acknowledgement
[27] J.M. Fernandez-G, E.A. Arauz, R.C. Rosado, R.A. Toscano, N.M. Ruvalcaba, M.A.
Martinez, Trans. Met. Chem. 24 (1999) 18–24.
[28] B.J. Hathaway, A.A.G. Tomlinson, Coord. Chem. Rev. 5 (1970) 1–43.
[29] K. Ono, M. Ohtsuka, J. Phys. Soc. Jpn. 13 (1958) 206–209.
[30] A. Takeuchi, S. Yamada, H. Yokoi, Bull. Chem. Soc. Jpn. 58 (1985) 2987–2989.
[31] B.J. Hathaway, Chem. Soc., Dalton Trans. (1972) 1196–1199.
[32] A.H. Maki, B.R. McGarvey, J. Chem. Phys. 29 (1958) 35–39.
[33] E.W. Ainscough, A.G. Bingham, A.B. Brodie, J.M. Husbands, J.E. Polwman, J.
Chem. Soc., Dalton Trans. (1981) 1701–1707.
We thank the Harran University Research Council for financial
support and the reviewers for their useful comments.
Appendix A. Supplementary data
[34] U. Sakaguchi, A.W. Addison, J. Chem. Soc., Dalton Trans. (1979) 600–608.
[35] A. Sokolovsky, J. Müller, T. Weyhermüller, R. Schnepf, P. Hilderbant,
K. Hildenbrand, E. Bothe, K. Wieghardt, J. Am. Chem. Soc. 119 (1997)
8889–8900.
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.saa.2010.02.016.
[36] E.R. Altwicker, Chem. Rev. 67 (1967) 475–510.
References
[37] I.J. Rhile, T.F. Markle, H. Nagao, A.G. DiPasquale, O.P. Lam, M.A. Lockwood, K.
Rotter, J.M. Mayer, J. Am. Chem. Soc. 128 (2006) 6075–6088.
[38] G.S. Patterson, R.H. Holm, Bioinorg. Chem. 4 (1975) 257–275.
[1] A.A. Medjidov, V.T. Kasumov, H.S. Mamedov, Russ. J. Coord. Chem. 7 (1981)
66–71.