Photophysical behaviour of 4-hexyloxysalicylaldimies and their copper(II) complexes

Article abstract of DOI:10.1016/j.jphotochem.2014.01.012

Salicylaldimines are popular ligands used in the synthesis of anisotropic complexes which can form liquid crystal phases (metallomesogens). They tautomerise via exchange of the phenol proton, giving rise to solvent-dependent photophysical behaviour. In th

Full text of DOI:10.1016/j.jphotochem.2014.01.012

Journal of Photochemistry and Photobiology A: Chemistry 279 (2014) 52–58
Contents lists available at ScienceDirect
Journal of Photochemistry and Photobiology A:
Chemistry
journal homepage: www.elsevier.com/locate/jphotochem
Photophysical behaviour of 4-hexyloxysalicylaldimies and their
copper(II) complexes
William A. Greenbank, Kathryn M. McGrath∗
MacDiarmid Institute for Advanced Materials and Nanotechnology, School of Chemical and Physical Sciences, Victoria University of Wellington, Wellington,
New Zealand
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 2 July 2013
Received in revised form 22 January 2014
Accepted 24 January 2014
Available online 5 February 2014
Salicylaldimines are popular ligands used in the synthesis of anisotropic complexes which can form liq-
uid crystal phases (metallomesogens). They tautomerise via exchange of the phenol proton, giving rise
to solvent-dependent photophysical behaviour. In this study the photophysical behaviour of a family
of 4-alkoxysalicylaldimine ligands and their corresponding copper(II) complexes were examined using
electronic absorption spectroscopy and density functional theory (DFT). The solvent-dependent photo-
physical behaviour exhibited by the ligands is due to changes in the position of the tautomeric equilibrium
as the degree of hydrogen-bonding character of the solvent is modified. Furthermore, the ligands bind
in an intermediate state in the copper(II) complex, with the possible existence of a third tautomer. The
electronic absorption properties of these compounds are strongly affected by changes in the environment
around the molecule.
Keywords:
Alkoxysalicylaldimine
Enol-imine tautomers
Photochromism
Electronic absorption spectroscopy
Density functional theory
© 2014 Elsevier B.V. All rights reserved.
1. Introduction
Here the photophysical properties of salicylaldimine ligands,
differing only in the length of the hydrocarbon chains and the cor-
Salicylaldimines (Scheme 1) are popular ligands in transi-
tion metal chemistry. In particular, the square planar complexes
they form are frequently employed in the synthesis of metal-
lomesogens (metal-containing liquid crystals) as the complexes
are highly anisotropic and the ligands are easily modified. Tran-
sition metal-salicylaldimine metallomesogens exhibit a range
of structure-dependent mesophases, detailed in the review by
Hoshino [1].
One well-known feature of salicylaldimines is their ability to
tautomerise via transfer of the phenol proton (Scheme 1), pre-
senting as the enol-imine or keto-amine tautomer. The enol-imine
tautomer has the lower energy. Tautomerization arises due to pho-
toexcitation (photochromism) [2], variation of solvent/complex
interactions [3], and thermal excitation [2].
Although the relationships between metallomesogens’ chem-
ical structure and their phase behaviour have been thoroughly
investigated, comparatively few studies have been made of their
physical properties. Previous studies show that the physical prop-
erties of the metallomesogen in the mesophase are markedly
different from those observed in the crystalline or isotropic states
and are highly dependent on the structure of the mesophase [4].
Understanding these relationships is critical to finding new appli-
cations for liquid crystals.
responding copper(II) complexes (Scheme 2) were examined using
solution phase electronic absorption spectroscopy and compared
to results obtained from DFT simulations. The structure of the core
of these molecules is commonly used in metallomesogens; vari-
ants of these complexes with differing alkyl chain lengths were
investigated by Paschke et al. for their liquid crystal properties
[5], and the mesogenic properties of the N-phenyl variants (vide
infra) are reviewed by Hoshino [1]. The electronic properties of
these complexes, however, are not well understood, in particular
how these properties respond to structural changes and changes in
the environment around the molecule; these are the focus of this
paper.
In particular we focus on the ability of the solvent to alter the
position of the tautomeric equilibrium and how the ligands bind
to the copper. This has not been investigated previously. Under-
standing the electronic behaviour of the ligands when bound to
copper and how the environment around the ligands affects their
structure and electronic properties is critical in understanding any
self-assembly dependent electronic interactions.
2. Methods
2.1. Synthesis
The ligands were synthesised in two steps (Scheme 2). In
the first step the alkoxy chain is formed by reaction of 4-
hydroxysalicyaldehyde with n-alkylbromide in the presence of
∗
Corresponding author. Tel.: +64 4 463 5963.
E-mail address: kathryn.mcgrath@vuw.ac.nz (K.M. McGrath).
1010-6030/$ – see front matter © 2014 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.jphotochem.2014.01.012

W.A. Greenbank, K.M. McGrath / Journal of Photochemistry and Photobiology A: Chemistry 279 (2014) 52–58
53
for 90 hours. The reaction mixture was then filtered through
diatomaceous earth. The crude products were purified by flash
chromatography, using a 19:1 mixture of petroleum ether/ethyl
acetate as the eluent. Yield: 62% of a pale yellow oil. Rf: 0.6 (4:1
petroleum ether/ethyl acetate). 1H NMR (500 MHz, CDCl3): ı 11.48
(s, 1H); 9.70 (s, 1H); 7.41 (d, 8.8 Hz, 1H); 6.52 (dd, 8.8 Hz, 2.3 Hz,
1H); 6.41 (d, 2.2 Hz, 1H); 4.00 (t, 6.6 Hz, 2H); 1.79 (quin., 7.1 Hz,
2H); 1.45 (m, 2H); 1.2–1.4 (m, 4H); 0.91 (t, 7.1 Hz, 3H).
Scheme 1. The tautomerization equilibrium between the enol-imine and keto-
amine tautomers of a salicylaldimine.
potassium bicarbonate. In the second step, the imine bond is formed
via condensation with either 1-aminoalkane to form the L1 series
of ligands (N-alkyl), or 4-alkylaniline to form the L2 ligand series
(N-phenyl).
Complexation with copper was achieved by reacting the
appropriate ligand with copper(II) acetate. The ligands were
characterised using infrared, ultraviolet, and 13C and 1H NMR spec-
troscopies, while the complexes were characterised by infrared and
ultraviolet spectroscopies, and elemental analysis.
Syntheses done under inert atmospheres were carried out by
degassing the reaction vessel in nitrogen gas and then fitting the
vessel with a balloon of the relevant gas (all supplied by BOC New
Zealand).
The synthetic protocols were adapted from: Yelamaggad et al.
[6], van Deun and Binnemans [7], Paschke et al. [5], and López de
Murillas [8]. A family of ligands and complexes were synthesised
(Scheme 2). The hexyl members of the two families are highlighted
as exemplars of the ligands and complexes investigated.
2.1.2. N-hexyl 4-hexyloxysalicylaldimine (L1(6,6))
4-Hexyloxysalicylaldehyde, acetic acid (one equivalent, Pure
Science, AR grade–glacial), and 1-aminohexane (1.2 equivalent,
Merck, 98%) were dissolved in absolute ethanol (15 mL, Pure Sci-
ence, ≥98.85%) under a nitrogen atmosphere. The solution was
refluxed for four hours under nitrogen. The solvent was removed
in vacuo, and the crude product was purified by flash chromatog-
raphy, using a 9:1 mixture of petroleum ether/ethyl acetate as the
eluent. Yield: 77% of a viscous yellow oil. Rf: 0.5 (4:1 petroleum
ether/ethyl acetate). 1H NMR (500 MHz, CDCl3): ı 14.15 (s, 1H);
8.09 (s, 1H); 7.06 (d, 8.5 Hz, 1H); 6.37 (d, 2.2 Hz, 1H); 6.33 (dd,
8.7 Hz, 2.3 Hz, 1H); 3.96 (t, 6.6 Hz, 2H); 3.52 (t, 6.8 Hz, 2H); 1.78
(quin., 7.1 Hz, 2H); 1.67 (quin., 7.1 Hz, 2H); 1.26–1.48 (m, 12H); 0.91
(t, 6.6 Hz, 3H); 0.90 (t, 6.9 Hz, 3H). 13C NMR (500 MHz, CDCl3): ı
167.7; 163.5; 163.1; 132.5; 111.8; 106.7; 102.0; 68.0; 57.1; 31.6;
31.5; 30.8; 29.0; 26.7; 25.7; 22.6; 22.6; 14.0; 14.0. IR ␯/cm−1: 2955
(C H); 2928 (C H); 2855 (C H); 2800 (O H); 1624 (C N); 1514
(C OH).
2.1.1. 4-Hexyloxysalicylaldehyde
To a solution of 2,4-dihydroxybenzaldehyde (1.08 g, 7.77 mmol,
Sigma–Aldrich, 98%) in dry acetone (20 mL, Panreac, ≥99%, ≤0.01%
H2O) were added potassium bicarbonate (789 mg, 7.88 mmol, May
and Baker Ltd., ≥98%), and n-hexylbromide (7.80 mmol, Merck,
98%). The mixture was refluxed under an argon atmosphere
2.1.3. N-(4-butylphenyl) 4-hexyloxysalicylaldimine (L2(4,6))
The same procedure as above was used, except that 4-
butylaniline (1.2 equivalent, Sigma–Aldrich, ≥97%) was used
instead of 1-aminohexane. Yield: 94% of a yellow crystalline solid
that melts slightly above room temperature to form a cloudy yellow
(i)
OH
OCnH2n+1
O
O
HO
HO
(ii)
OCnH2n+1
OCnH2n+1
O
R
N
L1(m,n)/L2(4,6)
HO
HO
OCnH2n+1
R
N
(iii)
Cu
O
O
OCnH2n+1
Cu1(m,n)/Cu2(4,6)
N
R
R
N
H2n+1CnO
HO
L1(m,n)/Cu1(m,n):R=C mH2m+1 (n=4,6,8;m=6,8,10,12)
C4H9
(n=6)
L2(4,6)/Cu2(4,6):R=
Scheme 2. Reaction scheme: (i) CnH2n+1Br, KHCO3, acetone; (ii) CnH2n+1NH2/4-butylaniline, AcOH, EtOH; and (iii) Cu(OAc)2, EtOH. L1(6,6), L2(4,6) and their corresponding
copper(II) complexes Cu1(6,6) and Cu2(4,6) are discussed in the text as examples.

54
W.A. Greenbank, K.M. McGrath / Journal of Photochemistry and Photobiology A: Chemistry 279 (2014) 52–58
Table 1
2.0
1.5
1.0
0.5
0.0
C
B
A
Solvents used in electronic absorption experiments.
Solvent
Supplier
Purity
Methanol
Ethanol
Acetonitrile
Hexane
Acetonitrile
Ethanol (absolute)
n-Hexane
Panreac
Pure Science
Labscan
99.9%
≥98.85%
99%
Methanol (dry)
Panreac
99.8%, ≤0.005% H2O
fluid. Rf: 0.5 (4:1 petroleum ether/ethyl acetate). 1H NMR (500 MHz,
CDCl3): 13.94 (s, 1H); 8.52 (s, 1H); 7.24 (d, 8.3 Hz, 1H); 7.21 (d,
8.5 Hz, 2H); 7.18 (d, 8.0 Hz, 2H); 6.49 (s, 1H); 6.47 (d, 8.5 Hz, 1H);
4.00 (t, 6.6 Hz, 2H); 2.63 (t, 7.8 Hz, 2H); 1.80 (quin., 7.5 Hz, 2H);
1.61 (quin., 7.6 Hz, 2H); 1.3–1.5 (m, 8H); 0.94 (t, 7.6 Hz, 3H); 0.92
(t, 7.0 Hz, 3H). 13C NMR (500 MHz, CDCl3): ı 164.0; 163.5; 160.6;
145.9; 141.3; 133.3; 129.3; 120.8; 113.0; 107.5; 101.6; 68.2; 35.2;
33.7; 31.6; 29.0; 25.7; 22.6; 22.3; 14.0; 14.0. IR ␯/cm−1: 2958 (C H);
2956 (C H); 2859 (C H); 2750 (O H); 1618 (C N); 1598 (C N);
1514 (C OH).
42000
37000
32000
Wavenumber (cm−1
27000
22000
)
Fig. 1. The electronic absorption spectra of L1(6,6) (representative of L1 ligands) in
a range of solvents with bands A, B, and C annotated.
compounds previously [10]. Charge distributions were determined
by the natural bond orbital (NBO) method [11]. Selected extracted
data are supplied in the electronic supplementary data.
2.1.4. Bis (N-hexyl 4-hexyloxysalicylaldimine) copper(II)
(Cu1(6,6))
A solution of L1(6,6) (2.0 mmol) in absolute ethanol (10 mL) was
added dropwise to a stirring solution of copper(II) acetate monohy-
drate (1.0 mmol, May and Baker Ltd., ≥98.5%) in absolute ethanol
(10 mL) at 60 ◦C. The mixture was refluxed for five hours before
being allowed to cool to room temperature overnight. Following
this, the solvent was removed in vacuo and the resulting green solid
was dissolved in dichloromethane (Panreac, ≥99.9%) and filtered
through diatomaceous earth. The dichloromethane was removed
in vacuo and the remaining solid was recrystallised from absolute
ethanol. Yield: 80% of a golden crystalline solid. IR ␯/cm−1: 2957
(C H); 2932 (C H); 2857 (C H); 1605 (C N); 1527 (C OCu). Anal.
Calc. for CuC38H60N2O4: C 67.87, H 8.99, N 4.17; Found: C 68.18, H
9.31, N 4.20%.
3. Results and discussion
3.1. Photophysical properties of the ligands
The solution phase electronic absorption spectrum of the L1
(N-alkyl) ligands show significant variation with solvent (Fig. 1),
but negligible variation with N-alkyl and alkoxy chain lengths.
The largest difference is observed for the spectra in hexane and
methanol where the band labelled C (Fig. 1) is significantly blue
shifted with a decreased intensity. The band labelled B is blue
shifted with an increase in intensity. Band A is very weak in hexane
and very strong in methanol.
This solvent effect is also seen when ethanol is used (Fig. 1), but
is present to a far smaller extent when acetonitrile is used. These
data indicate that the effect arises due to the hydrogen-bonding
character of the solvents, and not their polarity.
2.1.5. Bis (N-(4-butylphenyl) 4-hexyloxysalicylaldimine)
copper(II) (Cu2(4,6))
The same procedure as outlined in 2.1.4 was used, with L2(4,6)
(2.0 mmol) replacing L1(6,6). Yield: 54% of a golden crystalline solid.
IR ␯/cm−1: 2954 (C H); 2931 (C H); 2858 (C H); 1608 (C N); 1587
(C N); 1523 (C OCu). Anal. Calc. for CuC46H60N2O4: C 71.89, H
7.87, N 3.65; Cu 8.27; Found: C 72.32, H 8.01, N 3.65; Cu 8.3%.
This dependence of the data on the hydrogen-bonding character
of the solvent suggests that the solvent effects are the result of a
tautomeric equilibrium being established in solution (analogous
to that shown in Scheme 1). This is supported by the observa-
tion of a highly deshielded 1H NMR resonance of the phenolic
proton (14.15 ppm) and the peak broadening brought about by cou-
pling to the nitrogen indicating a high degree of intra-molecular
hydrogen-bonding between the two. For the enol-imine (O H N)
2.2. Instrumentation
NMR spectra were obtained in deuterated chloroform
(Sigma–Aldrich, 99.8% atom D, 17.6 ppm H2O) using a Varian
Unity Inova 500 spectrometer. Infrared spectra were obtained
using a Bruker Tensor 27 FTIR spectrometer in attenuated total
reflectance (ATR) mode and elemental analysis was carried out
by the Campbell Microanalytical Laboratory at the University of
Otago.
Electronic absorption data were obtained using a Varian Cary
100, dual beam, baseline corrected UV–vis spectrophotometer in
transmission mode. Solutions were prepared quantitatively. The
solvents used, their suppliers and purities are shown in Table 1.
˚
the hydrogen bond is 1.725 A and for the keto-amine (O H N)
˚
1.714 A, based on the ground state DFT data.
Hammud et al. proposed that the solvent dependence is due
to hydrogen-bonding solvents increasing the acidity of the phe-
nol proton [12], therefore shifting the position of the tautomeric
equilibrium more towards the keto-amine tautomer. If this is the
driving factor for the L1 ligands, it is expected that features due to
the presence of both tautomers in solution would be evident in the
spectrum and that quantitative analysis would provide the equilib-
rium ratios as set by the hydrogen-bonding character of solvent.
Of the solvents shown in Fig. 1, hexane has no hydrogen-bonding
character, hence in solution only the enol-imine tautomer will
be present. Conversely, methanol has the most hydrogen-bonding
character of the solvents and would therefore contain the largest
proportion of the keto-amine tautomer. Comparison of the spectra
obtained for the L1(6,6) ligand in methanol vs. hexane reveals few,
if any, common features (it is noted that the breadth of the peaks
may mask particular features). When one considers the spectrum
obtained in ethanol there is a shoulder at 35,500 cm−1 potentially
2.3. Computational methods
Gaussian 09 was used to perform all computations [9]. Geometry
optimisation was carried out using DFT with a B3LYP functional
and a 6-31G++ basis set in the case of L1(6,6) and 6-31G** in the
case of Cu1(6,6). Excited state calculations were performed using
time dependent DFT (TD-DFT) with the same functional and basis
set. TD-DFT has been used to calculate the excited states of similar

W.A. Greenbank, K.M. McGrath / Journal of Photochemistry and Photobiology A: Chemistry 279 (2014) 52–58
55
L1 (keto-amine)
L1 (enol-imine)
Cu1
44000
39500
35000
30500
26000
Wavenumber (cm−1
)
Fig. 3. Simulated electronic absorption spectra of the enol-imine and keto-amine
tautomers of the L1(6,6) ligand, and Cu1(6,6) calculated with TD-DFT using Gaussian
09 [9]. Note that the intensities of bands are not comparable between spectra.
Fig. 2. Titration study of L1(6,6) ligand where the solvent is varied from 100% (v/v)
ethanol to 100% (v/v) hexane.
because of its strong dependence on the hydrogen-bonding ability
of the solvent.
The data for acetonitrile support these conclusions, with its very
weak capacity to hydrogen bond. Hence the equilibrium lies pre-
dominantly towards the enol-imine and the intensity of band A is
very weak though readily observed.
corresponding to the enol-imine tautomer band C. To investigate
this further, a titration experiment was performed where the sol-
vent was varied from 100% v/v ethanol to 100% v/v hexane (Fig. 2).
The growth of one peak at the expense of another in the band C
region, as well as the presence of multiple isosbestic points, shows
that these data result from the presence of two separate chemical
species. If the solvent effects were due to the solvent affecting the
energy of the ground or excited states of the transition in band
C, then a gradual progression of the peak from 35,700 cm−1 to
39,200 cm−1 would be observed, not the growth of one peak at the
expense of the other as seen here. These observations can be ration-
alised if band C is assigned to a ␲–␲* transition delocalised over
the conjugated system. The reduction in wavelength and molar
absorption coefficient would be consistent with the formation of
the keto-amine tautomer as this would reduce the extent of conju-
gation. This assignment is consistent with reported studies of other
salicylaldimine-based compounds [12–14].
The behaviour of band B is, at first glance, harder to rationalise
by this means. Peaks in this region and having this intensity in the
spectra of salicylaldimine compounds are generally assigned to a
␲–␲* transition localised on the imine C N chromophore [12–14].
This peak would therefore be expected to decline in intensity as the
proportion of the enol-imine tautomer is reduced by the addition
of a hydrogen-bonding solvent. This is not consistent with the data
shown in Figs. 1 and 2, in which the peak increases significantly in
intensity, and shifts to lower wavelength with increasing ethanol
fraction.
The inconsistency between the predicted decline in band B and
the experimental data can potentially be explained when the chro-
mophores present in the keto-amine tautomer, in particular the
ketone formed at the 2 position on the ring are considered. If the
␲–␲* transition of the ketone is more strongly absorbing than
the imine in the enol-imine tautomer, and has a peak wavelength
slightly below that of the imine, then the two absorbances would
appear as a single peak, especially when the peaks are broad. The
apparent growth of the peak with increasing ethanol volume frac-
tion would then be as expected.
Turning now to band A. This band is strongly affected by the
hydrogen-bonding character of the solvent and is therefore related
to the tautomerism of the ligands. There has been a number of
studies of the photochromism of salicylaldimine-based compounds
[2,12] in which tautomerization from the enol-imine tautomer to
the higher energy keto-amine tautomer is induced by irradiation
with ultraviolet radiation. Band A is therefore consistent with such
a photochromic charge transfer as it is similar in wavelength to
photochromic charge transfer bands in similar compounds [2], and
The electronic absorption of the two tautomers of the L1(6,6)
ligand was modelled with TD-DFT (Fig. 3, see ESI for full data
on molecular orbital energies, symmetries and transitions con-
tributing to the calculated peaks) using the Gaussian 09 software
package [9]. The calculated UV–vis spectrum of the enol-imine tau-
tomer shows bands that can be correlated to bands B and C in
the spectrum of the L1(6,6) ligand in hexane at 33,700 cm−1 and
37,000 cm−1 respectively. Likewise, in the calculated spectrum of
the keto-amine tautomer, bands at 27,900 cm−1, 35,000 cm−1 and
41,500 cm−1 can be correlated to bands A, B and C (respectively)
in the measured spectrum of the L1(6,6) ligand in methanol. This
further supports the hypothesis that the solvent effects observed in
the experimental spectra are due to differences in the equilibrium
levels of the two tautomers in solvents with different hydrogen-
bonding character.
As shown in Fig. 4, the UV–vis spectrum of L2(4,6) contains
the same features as those observed and fully discussed above for
the L1 ligands. For L2(4,6) the bands are red shifted and there are
some differences in intensity. This is consistent with the extended
conjugation in the L2(4,6) ligand as compared with the L1 series.
Another expected consequence of the extra phenyl ring in L2(4,6) is
that bands B and C no longer show the strong solvent dependence
exhibited by L1 ligands. This is because of the increased stability
of the enol-imine tautomer due to the greater conjugation, hence
the tautomeric equilibrium is shifted much further towards the
C
B
A
2.5
2.0
1.5
1.0
0.5
0.0
Methanol
Acetonitrile
Hexane
L1
35000
30000
25000
20000
Wavenumber (cm−1
)
Fig. 4. Electronic absorption spectra of L2(4,6) in a range of solvents with bands A,
B and C annotated. The spectrum of the L1(6,6) ligand in acetonitrile is shown for
reference (dashed curve).

56
W.A. Greenbank, K.M. McGrath / Journal of Photochemistry and Photobiology A: Chemistry 279 (2014) 52–58
Table 2
C
A
B
5.0
4.0
3.0
2.0
1.0
0.0
Bond lengths determined from DFT simulations of the L1(6,6) ligand and Cu1(6,6)
complex.
˚
˚
Amine/imine C N/A
Ketone/enol C O/A
L1
Cu1
L1(6,6) (keto-amine)
Cu1(6,6)
L1(6,6) (enol-imine)
1.326
1.301
1.287
1.270
1.302
1.345
spectra of the Cu1 series is most likely closer to the position of the
component of band B in the L1 series spectra that arises from the
keto-amine tautomer.
In addition to the ligand-centred bands, an additional very low
intensity, broad peak is evident at 16,100 cm−1 (130 L mol−1 cm−1).
This is due to the metal-centred d–d transition and is consistent
with other similar compounds [14].
The TD-DFT simulation of the electronic absorption spectrum of
the Cu1(6,6) complex (Fig. 5) contains features that can be corre-
lated to bands A, B and C in the measured spectrum. The simulated
spectrum contains features from both tautomers, indicating the
ligand is in a state that is intermediate to the two tautomers when
bound. Moreover, the bond lengths of the amine/imine C N and
ketone/enol C O bonds in the DFT optimised structure of Cu1(6,6)
(Table 2), are between the values for the keto-amine and enol-imine
tautomers of L1(6,6).
The hypothesis that the ligand binds to copper(II) in an interme-
diate state to the two tautomers is also supported by the plot of the
HOMO of the DFT optimised Cu1(6,6) complex (Fig. 6). This shows
that the aromatic ␲ system is partially delocalised to include both
the enol/ketone oxygen and the nitrogen. In addition, the HOMO
plot shows that the alkoxy oxygen is included, suggesting involve-
ment of a third tautomer that was not observed in the free ligand.
The calculated charge distribution in the complex (calculated using
42000
37500
33000
28500
A
24000
Wavenumber (cm−1
)
C
B
5.0
4.0
3.0
2.0
1.0
0.0
L1
Cu1
42000
37500
33000
28500
24000
Wavenumber (cm−1
)
Fig. 5. Representative electronic absorption spectra of the Cu1 complexes in ace-
tonitrile (top) and methanol (bottom), shown here for Cu1(6,6). Bands A, B and C are
labelled and the spectra of L1(6,6) are shown for reference, data for the C6 ligand
and complex are shown.
enol-imine tautomer. As such solvent effects are significantly
reduced. The greatest effect is seen for band A, but this is still
significantly reduced compared with the L1 series.
3.2. Photophysical properties of the complexes
Complexation of L1 ligands to copper(II) have two main effects
on the electronic absorption spectrum (Fig. 5). The first is that the
molar absorption coefficient of all peaks is significantly increased.
An increase in the molar absorption coefficients is expected, since
the ligands form a 2:1 complex with copper, meaning that the
number of chromophores per mole is effectively doubled upon
complexation. What is unusual, though, is that for some of the peaks
the intensity more than doubles upon complexation. This effect has
been reported in previous studies of similar complexes by Paschke
et al. [5], Bhattacharjee et al. [13], and Vafazadeh et al. [14], but no
attempt was made to explain the results in these reports.
The other noticeable effect is the blue shifts of all the peaks upon
complexation. This is reminiscent of the blue shifted peaks of the
L1 ligands discussed in Section 3.1. Comparing the methanol spec-
tra of the L1(6,6) ligand and the corresponding copper(II) complex,
Cu1(6,6) (Fig. 5), shows that the wavelength of band C matches, and
the wavelength of band B is only blue shifted by 1100 cm−1 upon
complexation. This suggests that the L1 ligands actually favour the
keto-amine tautomer when bound to copper in solution, not the
enol-imine tautomer which is favoured when not bound.
This also explains the apparent significant increase in the molar
absorption coefficient as the intensity of band C in the spectrum
of a pure or enriched solution of the keto-amine tautomer of an L1
ligand would be higher than band C in the spectrum of a methanol
solution of an L1 ligand as the latter is a mixture of both tautomers
with a slight preference for the enol-imine tautomer. The shift in
band B can be explained when one considers that band B in the
spectrum of an L1 ligand in methanol is a composite of two peaks,
one from each tautomer. Therefore the position of band B in the
Fig. 6. (Top) A plot of the HOMO of the DFT optimised structure of Cu1(6,6) and (bot-
tom) the charge distribution in the DFT optimised structure of Cu1(6,6) calculated
using the NBO method [11].

W.A. Greenbank, K.M. McGrath / Journal of Photochemistry and Photobiology A: Chemistry 279 (2014) 52–58
57
how the ligands would behave electronically in a mesophase, either
as ligands, mesogens or dopants.
C
B
A
6.0
4.5
3.0
1.5
0.0
This study has also used electronic absorption spectroscopy and
TD-DFT simulations to demonstrate for the first time the bind-
ing mode of the ligands to copper. The bands of the Cu1 and Cu2
complexes are blue shifted from those of the free ligands. The
wavelengths of the ligand-centred bands indicate that the ligands
in either complex bind to the metal centre in solution in a form
that exhibits features of the keto-amine tautomer, despite the free
ligand preferring the lower energy enol-imine tautomer. TD-DFT
simulations confirm that the ligands are in a state that is interme-
diate to the two tautomers, and that aspects of both are present in
the electronic absorption spectra of the Cu1 complexes.
The electronic absorption properties of these compounds
are strongly affected by changes in the environment around
the molecule. This will prove important in understanding any
self-assembly dependent photophysical properties that these com-
pounds exhibit.
L2
Cu2
Cu1
42000
37000
32000
Wavenumber (cm−1
27000
22000
)
Fig. 7. The electronic absorption spectrum of Cu2(4,6) in acetonitrile with bands A,
B and C labelled. The spectra of L2(4,6) and Cu1(6,6) in acetonitrile are shown for
reference.
the natural bond orbital method [11]) indicates that this oxygen has
a lower charge than the enol/ketone oxygen, suggesting that this
tautomer is less favoured than the other two.
Acknowledgements
The blue shift of band A upon complexation of an L1 ligand to
copper(II) is much greater than that for bands B and C. This can be
justified when one considers that in the L1 ligands, band A results
from a photochromic charge transfer, exciting the molecule into
the high energy keto-amine tautomer from the more stable enol-
imine tautomer. It is evident that, without a proton to transfer, this
tautomerism is no longer possible. Further evidence for this comes
from the solvent dependence that was observed for the L1 ligand
series which is no longer observed on complexation to copper(II).
Since band A in the Cu1 complex spectrum does not originate from
photochromism, it must be a new band resulting from complex-
ation to copper. Given its similarity to band A in the L1 ligands
spectra, it most likely still arises from a charge transfer, but this
time involving the metal rather than a phenol proton, i.e. a ligand to
metal charge transfer, LMCT. This is consistent with the lack of sol-
vent dependency as the transition does not depend on the lability
of a phenol proton as in the case of the ligand. This also agrees with
the studies done by Vafazadeh et al. [14] and Tas et al. [15] who
reported that bands in this region in similar complexes resulted
from LMCT.
The changes in the L2(4,6) data upon complexation to copper(II)
are relatively similar to those observed following complexation of
the L1 ligands to copper(II) with regard to the blue shift and increase
in molar absorption coefficient for peaks B and C for the complex
vs. the ligand (Fig. 7). This is consistent with the ligands of Cu2(4,6)
binding to copper(II) in an intermediate state to the two tautomers,
in the same manner as the ligands of Cu1. The molar absorption
coefficient of the LMCT band A in Cu2(4,6) is significantly increased
compared with the corresponding band in the Cu1 spectra and is red
shifted by 2200 cm−1. Since the LMCT transition causes a substan-
tial redistribution of charge in the molecule, increasing the extent
of conjugation would be expected to have both of these effects as
it would raise the probability of the transition and better stabilise
the excited state via resonance.
The authors thank Ivan Welsh and Zhi Xiang Wong for their help
with Gaussian 09. WG was financially supported by a J.L. Stewart
scholarship and a Curtis-Gordon scholarship.
Appendix A. Supplementary data
Supplementary data associated with this article can be found,
in the online version, at http://dx.doi.org/10.1016/j.jphotochem.
2014.01.012.
References
[1] N. Hoshino, Liquid crystal properties of metal–salicylaldimine complexes.
Chemical modifications towards lower symmetry, Coordination Chemistry
Reviews 174 (1998) 77.
[2] J. Zhao, B. Zhao, J. Liu, W. Xu, Z. Wang, Spectroscopy study on the pho-
tochromism of Schiff Bases N,Nꢀ-bis(salicylidene)-1,2-diaminoethane and
N,Nꢀ-bis(salicylidene)-1,6-hexanediamine, Spectrochimica Acta Part A: Molec-
ular and Biomolecular Spectroscopy 57 (2001) 149.
[3] G. Türkog˘lua, H. Berber, H. Dal, C. Ög˘retirb, Synthesis, characterization,
tautomerism and theoretical study of some new Schiff base derivatives, Spec-
trochimica Acta, Part A 79 (2011) 1573.
[4] J.L. Serrano, Metallomesogens: Synthesis, Properties, and Applications, VCH,
Weinheim, 1996.
[5] R. Paschke, S. Liebsch, C. Tschierske, M.A. Oakley, E. Sinn, Synthesis and
mesogenic properties of binuclear copper(II) complexes derived from salicy-
laldimine schiff bases, Inorganic Chemistry 42 (2003) 8230.
[6] C.V. Yelamaggad, R. Prabhu, G. Shanker, D.W. Bruce, Optically active, mesogenic
lanthanide complexes: design, synthesis and characterization, Liquid Crystals
36 (2009) 247.
[7] R. Van Deun, K. Binnemans, Mesomorphic lanthanide complexes with azome-
thine ligands, Journal of Alloys and Compounds 303–304 (2000) 146.
[8] D. Lopez de Murillas, R. Pinol, M. Blanca Ros, J.L. Serrano, T. Sierra, M. Rosario de
la Fuente, Structure–activity studies of ferroelectric and antiferroelectric imine
ligands and their palladium(II) complexes. An antiferroelectric metallomeso-
gen, Journal of Materials Chemistry 14 (2004) 1117.
[9] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman,
G. Scalmani, V. Barone, B. Mennucci, G.A. Petersson, H. Nakatsuji, M. Caricato,
X. Li, H.P. Hratchian, A.F. Izmaylov, J. Bloino, G. Zheng, J.L. Sonnenberg, M. Hada,
M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O.
Kitao, H. Nakai, T. Vreven, J. Montgomery, J.E. Peralta, F. Ogliaro, M. Bearpark,
J.J. Heyd, E. Brothers, K.N. Kudin, V.N. Staroverov, R. Kobayashi, J. Normand, K.
Raghavachari, A. Rendell, J.C. Burant, S.S. Iyengar, J. Tomasi, M. Cossi, N. Rega,
J.M. Millam, M. Klene, J.E. Knox, J.B. Cross, V. Bakken, C. Adamo, J. Jaramillo,
R. Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W.
Ochterski, R.L. Martin, K. Morokuma, V.G. Zakrzewski, G.A. Voth, P. Salvador,
J.J. Dannenberg, S. Dapprich, A.D. Daniels, Ö. Farkas, J.B. Foresman, J.V. Ortiz, J.
Cioslowski, D.J. Fox, Gaussian 09, Revision B.01 ed., Gaussian, Inc., Wallingford,
2009.
4. Conclusions
In this study we have observed that the L1 and L2 ligand series
exhibit strong solvent dependence in their electronic absorption
spectra. Through the use of electronic absorption spectroscopy, and
TD-DFT simulations, this was shown to be due to the ligands exist-
ing in a tautomeric equilibrium. This enables their photophysical
properties to be strongly affected by their environment, particu-
larly the presence of any hydrogen-bonding, which is relevant to
[10] Y.P. Tong, S.L. Zheng, Synthesis, structure, spectroscopic properties,
DFT and TDDFT investigations of copper(II) complex with 2-(2-
hydroxyphenyl)benzimidazole, J. Mol. Struct. 841 (2007) 34.
[11] E.D. Glendening, A.E. Reed, J.E. Carpenter and F. Weinhold, NBO, Version 3.

58
W.A. Greenbank, K.M. McGrath / Journal of Photochemistry and Photobiology A: Chemistry 279 (2014) 52–58
[12] H.H. Hammud, A. Ghannoum, M.S. Masoud, Spectral regression and correlation
coefficients of some benzaldimines and salicylaldimines in different solvents,
Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 63
(2006) 255.
[13] C.R. Bhattacharjee, G. Das, P. Mondal, Liquid-crystalline oxovanadium(IV) com-
plexes accessed from bidentate [N, O] donor salicylaldimine Schiff-base ligands,
Journal of Coordination Chemistry 64 (2012) 3273.
[14] R. Vafazadeh, V. Hayeri, A.C. Willis, Synthesis, crystal structure and electronic
properties of bis(N-2-bromophenyl-salicydenaminato)copper(II) complex,
Polyhedron 29 (2010) 1810.
[15] E. Tas, A. Kilic, N. Konak, I. Yilmaz, The sterically hindered salicylaldimine
ligands with their copper(II)metal complexes: synthesis, spectroscopy, electro-
chemical and thin-layer spectroelectrochemical features, Polyhedron 27 (2008)
1024.