Solvatochromism of catechol derivatives - Solute/solvent interactions

Article abstract of DOI:10.1002/poc.3003

The solvatochromism of nine push-pull substituted catechol derivatives has been studied in a set of 39 various solvents. The influence of successive methyl substitution at the catechol OH groups on the extent of the solvatochromic shift has been investiga

Full text of DOI:10.1002/poc.3003

Research Article
Received: 05 March 2012,
Revised: 13 June 2012,
Accepted: 20 June 2012,
Published online in Wiley Online Library: 2012
(wileyonlinelibrary.com) DOI: 10.1002/poc.3003
Solvatochromism of catechol derivatives –
solute/solvent interactions
Franziska Riedela and Stefan Spangea*
The solvatochromism of nine push–pull substituted catechol derivatives has been studied in a set of 39 various solvents.
The influence of successive methyl substitution at the catechol OH groups on the extent of the solvatochromic shift has
been investigated. The positive solvatochromism of 2-(3,4-dihydroxybenzylidene)-2H-indene-1,3-dione amounts
4360 cm–1, which ranges from toluene to hexamethyl-phosphoric triamide. To the best of our knowledge, it is one of
the largest positive solvatochromic extent measured for a positive solvatochromic dye, comparable with Brooker’s
thiobarbituric acid with an extent of 4400 cm–1. The detailed analyses of the solvatochromism were carried out by
alternatively using the Kamlet–Taft and Catalán solvent parameters to achieve information of dipolarity versus
polarizability effects of solvent upon solvatochromic properties. In solvents with high b values such as alcohols
(0.66 < b < 0.90), amides (0.48 < b < 0.80), dimethyl sulfoxide (b = 0.76), tetramethyl urea (b = 0.80) and hexamethyl-
phosphoric triamide (b = 1.05) UV–Vis absorption spectra show two separate lmax, which are caused by a deprotonation
reaction. The solvatochromic behaviour of the anionic species is compared with those of the catechol derivatives.
Copyright © 2012 John Wiley & Sons, Ltd.
Supporting information may be found in the online version of this paper
Keywords: catechol; donor–acceptor systems; hydrogen bonds; linear solvation energy correlation analysis; solvatochromism
parameters were determined from an average of measurements
using numerous data.
INTRODUCTION
Solvatochromism is used to describe the pronounced change
in position (~nmax) of an UV–Vis absorption band with a change in
solvent polarity and has been established as a valuable tool to
investigate the versatile properties of organic solvents.[1–10]
Suitable solvatochromic probes for analytical application are highly
dipolar push–pull aromatic systems or other types of those dyes,
which contain polar or ionic groups.[1] Overall solvent polarity
cannot be described quantitatively by a single physicochemical
parameter. The extent of solvation is a function of several
intermolecular forces between solute and solvent molecules.[5]
These include nonspecific, dipole/induced dipole forces, dipole/
dipoleforces and ion/dipole forces and specific forces, hydrogen
bond donating (HBD), hydrogen bond accepting (HBA), or charge
transfer interactions.[7] Kamlet and colleagues have developed an
LSE (linear solvation energy) model for characterizing manifold
interactions between solutes and solvents.[11–17] The simplified
Kamlet–Taft relationship, which is applied to solvatochromic
~
Catalán et al. published an array of alternative empirical solvent
parameter scales relating to the Kamlet–Taft scales.[18–20] However,
for this task well-defined reference systems were used to
determine the solvent parameters. For this purpose a suitable pair
of homomorph solvatochromic probes were used for each
parameter.[18–20] In this context SA and SB correspond to the
Kamlet–Taft parameter a and b. The specific advantage of the
new approach is the separate consideration of the polarizability
SP and dipolarity SdP (Eqn (2)).
~
~
nmax ¼ nmax;0 þ aꢀSA þ bꢀSB þ cꢀSP þ dꢀSd P
(2)
The extreme sensitivity of the visible absorption spectrum to
small changes in the surrounding medium has made several dyes
as useful molecular probes. The extent of solvatochromism Δ~n is
the difference between the measured ~nmax values of the dye in
two solvents of widely different polarity. This actuality is attributed
to the larger difference in the dipole moment between the
electronic ground and the first excited state.
In many technically important dyes the amino group or
dialkylamino group is employed as a donor substituent.[21] Thus,
4-nitroaniline derivatives have been established as important
UV–Vis shifts nmax, is shown in Eqn (1)[11–17]
:
~nmax ¼ ~nmax;0 þ aꢀa þ bꢀb þ sꢀpꢁ
(1)
~nmax;0 is the solute property of a reference system, the
nonpolar solvent cyclohexane, a represents the HBD ability,[11–14]
b describes the HBA ability[14,15] and p* describes the dipolarity/
polarizability[14,16,17] of the solvents. The contribution of the
different solvent effects on the UV–Vis absorption shift ~nmax is
expressed with solvent-independent correlation coefficients a,
b and s. Disadvantageously, Kamlet–Taft’s p* parameter does
not distinguish between dipolarity and polarizability. Further-
more, the whole empirical solvent parameter set is not based
on well-defined reference processes because the solvent
* Correspondence to: Stefan Spange, Department of Polymer Chemistry, Institute
of Chemistry, Faculty of Natural Science, University of Technology Chemnitz,
Strasse der Nationen 62, D-09111 Chemnitz, Germany.
E-mail: stefan.spange@chemie.tu-chemnitz.de
a F. Riedel, S. Spange
Department of Polymer Chemistry, Institute of Chemistry, Faculty of Natural
Science, University of Technology Chemnitz, Strasse der Nationen 62,
D-09111, Chemnitz, Germany
J. Phys. Org. Chem. (2012)
Copyright © 2012 John Wiley & Sons, Ltd.

F. RIEDEL AND S. SPANGE
solvatochromic dyes for measuring solvent polarity in terms of
the Kamlet–Taft’s p* scale.[14] In contrast, in natural dyes the amino
group serves as an electron-donating substituent only in few
examples. Here, the more important substituent is the catechol
moiety because of its ability to undergo deprotonation and
simultaneous stabilisation of the negative charge by intramolecu-
lar hydrogen bonds (Scheme 1).[22] Deprotonation reactions are
of interest in optical systems, for example, in their application
as acid–base indicators.[23,24] In this case, dye molecules convert
a chemical interaction into an optically detectable signal.
Deprotonation of a hydroxy moiety in push–pull substituted
chromophores induces a large bathochromic spectral shift
because of electron delocalization of the phenoxide into the
chromophore. Importantly, the phenolate oxygen atom is one of
the strongest electron-donating substituent actually whose
The objective of this study is the investigation of solvent effects on
the position of the UV–Vis absorption band of a series of catechol
derivatives to show which type of solvent interaction induces a
significant shift on the position of the UV–Vis absorption band.
RESULTS AND DISCUSSION
Synthesis
Compounds 2–4 were synthesized from 3,4-dihydroxybenzaldehyde,
4-hydroxy-3-methoxy-benzaldehyde and 3,4-dimethoxybenzalde-
hyde 1a–c, respectively, with malononitrile, indandione and
dicyanomethylene indanone in a condensation reaction accord-
ing to Scheme 2. The catechol derivatives were received as
yellow, orange and red solids in good yields.
+
strength can be quantitatively described by its Hammett sp
substituent constant. That is ꢂ4.27 compared with ꢂ1.70 of the
N,N-dimethylamino substituent.[25]
Thermochromism
Catechol derivatives are also of importance for surface
functionalization.[26,27] Therefore, the understanding of the effect
of the environment polarity on the UV–Vis absorption property is
also of importance to adjust the molecular structure for specific
application. Systematic studies on the solvatochromism of cate-
chol derivatives are still lacking in spite of the importance of the
catechol moiety in naturally occurring dyes. So far, 4-nitrocatechol
and its silicon complexes and a phenolate betaine dye were only
studied.[28,29]
From thermodynamic investigations it is known that isoequilibrium
behaviour can elucidate certain interaction mechanisms.[30,31]
The temperature dependence was investigated with chromo-
phore 3a, because of the most pronounced solvatochromic extent.
At room temperature, compound 3a shows two UV–Vis absorption
maxima at l1 = 429nm (n max = 23.31ꢀ103 cm–1) and l2 = 505nm
~
(~nmax = 19.80ꢀ103 cm–1) in methanolic solution. The UV–Vis
absorption maximum at n max = 19.80ꢀ103 cm–1 is caused by the
~
anionic compound (3a)–. On increasing temperature, the intensity
of the UV–Vis absorption at l1 = 429 nm (n max = 23.31ꢀ103 cm–1)
~
decreases while the intensity of the UV–Vis absorption at
l2 = 505 nm (n max = 19.80ꢀ103 cm–1) increases showing an
~
isosbestic point at 461 nm (n~max = 21.69ꢀ103 cm–1, c(3a)=6.1ꢀ10-5
mol L–1). These changes are reversible and correspond
to the temperature-dependent deprotonation equilibrium,
3a ⇄ (3a)–. Shifts of the UV–Vis absorption maxima and
changes in the band shape cannot be observed as a function
of temperature.
Scheme 1. Protonation/deprotonation equilibrium of acceptor-substituted
catechols (X = acceptor groups)
Scheme 2. Synthesis of catechol-based dyes 2–4
wileyonlinelibrary.com/journal/poc
Copyright © 2012 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. (2012)

SOLVATOCHROMIC CATECHOL DERIVATIVES
~
~
In 1,4-dioxane and chloroform no temperature dependent
shifts of the UV–Vis absorption maximum of 3a is observed.
Therefore, temperature dependent measurements give no
indication on isokinetic relationships.
(2b)=385nm (n max = 25.97ꢀ103 cm–1), lmax(3b) = 436 nm (n max
=
=
22.94ꢀ103 cm–1) and formamide
l
max(4b)=526nm (n~max
3
–1
~
19.01ꢀ10 cm ). From this a solvatochromic shift of Δn =1450 (3b)
to 2540 cm–1 (4b) can be observed. Compounds with CH3-
protected OH-group 2c, 3c and 4c show that the UV–Vis absorption
maximum is blue-shifted and the solvatochromic shift becomes
very small (Table 1) because hydroxy groups are not available for
interactions with HBA solvents. The most hypsochromic shifts are
Solvent effects on the UV–Vis absorption spectra
The solvatochromism of 2–4 was studied in 39 various solvents of
different HBD ability, HBA ability and dipolarity/polarizability.UV–Vis
spectroscopic investigation of the 3,4-dihydroxybenzaldehyde
1a,b shows one UV–Vis absorption band in nonpolar solvents
such as toluene. With increasing solvent polarity a bathochro-
mic shift and a splitting of the UV–Vis absorption band, caused
by hydrogen bond formation and deprotonation, is observed.
This illustrates very complex solvation behaviour of the catechol
moiety as a function of the surrounding medium. The position
of the UV–Vis absorption bands are independent of the dye
concentration in the range of 10–4 to 10–5 mol L–1, which is an
indication that dye aggregation has no noticeable effect on
the solvatochromic properties.
observed in hexane, lmax(2c) = 360 nm (n max = 27.78ꢀ103 cm–1),
~
l
~
max(3c) = 408 nm (~nmax = 24.51ꢀ103 cm–1) and lmax(4c) = 455 nm
(n max = 21.98ꢀ103 cm–1). In benzonitrile, lmax(2c) = 380 nm (n max
=
~
26.32ꢀ103cm–1), and HMPA, lmax(3c) = 420 nm ( n max =23.81ꢀ103cm–1),
~
l
max(4c) = 480 nm (n max = 20.83ꢀ103 cm–1), the most bathochromic
~
UV–Vis absorption maxima are observed. It is clearly seen that
an extension of the conjugated p-system from 2 to 4 is caused
in a bathochromic shift of the UV–Vis absorption maxima.
Catechol-based chromophores 2–4a and compounds 2–4b
containing one hydroxy group show in alcohols, carbonic acid
amides, DMSO, TMU and HMPA two separated UV–Vis
absorption maxima. Exemplary, Fig. 1 shows the UV–Vis absorp-
tion spectra of 3a in selected solvents.
Table 1 shows the UV–Vis absorption maxima of the
~
compounds 2–4 and the extent of the solvatochromic shift Δn
for selected solvents (the whole UV–Vis data set is given in
supporting information). The catechol derivatives show the
shortest wavelength in nonpolar solvents such as chloroform
lmax(2a) = 374 nm (nmax = 26.74ꢀ103 cm–1), benzene and toluene
~
lmax(3a) = 379 nm (n max = 26.39ꢀ103 cm–1), and dichloromethane
~
lmax(4a) = 450 nm (n max = 22.22ꢀ103 cm–1). The largest bathochro-
~
mic shift of the UV–Vis absorption maximum is observed in
polar solvents dimethyl sulfoxide (DMSO), lmax(2a) = 408 nm
(n max = 24.51ꢀ103 cm–1) and hexamethyl-phosphoric triamide
~
(HMPA) lmax(3a) = 454 nm ( n max = 22.03ꢀ103 cm–1), lmax(4a) =
~
525 nm (n max = 19.05ꢀ103 cm–1). These shifts of the UV–Vis ab-
~
sorption bands correspond to a positive solvatochromism with
solvatochromic ranges of Δ~n = 2230 (2a) to 4360 cm–1 (3a). The
latter one is very high and comparable with the largest so far
recorded red shift for a thiobarbituric acid derivative with
4400 cm–1 [32,33]
.
For compounds with 4-hydroxy-3-methoxy moieties (2b, 3b, 4b)
the most hypsochromic UV–Vis absorption maxima are
observed in benzene lmax(2b) = 363 nm (n max = 27.55ꢀ103 cm-1),
~
lmax(3b)=410nm(n max =24.39ꢀ103 cm–1) and 1,1,2,2-tetrachloroethane
~
lmax(4b) = 464 nm (n max = 21.55ꢀ103 cm–1). The largest bathochromic
~
Figure 1. UV–Vis absorption spectra of 3a in selected solvents. In DMSO
and HMPA two absorption maxima are observed (see text)
shift is found in polar solvents tetramethyl urea (TMU) lmax
Table 1. UV–Vis absorption maxima of compounds 2–4 in various solvents and the extent of the solvatochromic shift
Solvent
max[103 cm–1]
3b
~
n
2a
2b
2c
3a
3c
4a
4b
4c
Dichloromethane
Toluene
Diethyl ether
1,4-Dioxane
26.41
26.81
27.47
26.88
26.88
26.11
26.11
1570
26.67
27.03
27.17
26.95
26.74
26.53
1460
25.51
23.98
24.15
24.45
24.39
24.27
23.87
23.81b
700
22.22c
21.14
21.32
21.60
21.51
21.32
20.96
20.92
1150
a
a
a
26.39c
24.21
a
a
26.74
26.39
24.63
24.51b
2230
21.05
21.46
19.38
19.46
3170
21.37
20.70
19.42
19.46
2540
24.63
22.88
22.94
4360
23.81
22.99
22.99
1450
Dimethylformamide
DMSO
–1
~
Δn[cm ]
aInsoluble.
bSolvent with the highest bathochromic shift.
cSolvent with the highest hypsochromic shift.
J. Phys. Org. Chem. (2012)
Copyright © 2012 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/poc

F. RIEDEL AND S. SPANGE
It could be proved that the longer wavelength absorption
band is associated with the monoanionic species of the dyes
by means of UV–Vis spectroscopic investigations and nuclear
magnetic resonance (NMR) experiments using 1,8-diazabicyclo
[5.4.0]-undec-7-ene (DBU) as base for deprotonation. The reaction
between the dyes 2–4a,b and DBU in toluene was used as the
reference system esteeming a first indication of the existence of
deprotonated species in other solvents (supporting information).
latter occurrence is associated with an enhancement of the
push–pull character of the dye. Otherwise, the coefficient a would
be negative if the mesomeric structure II with the negatively
charged acceptor group is predominant. In this case the HBD
attack would increase the acceptor strength of this group.
However, the parameter of more significance for these
compounds is b. The distinct magnitude of the coefficient of the
b term is caused by the specific solvation of the catechol moiety
by hydrogen bonding to the HBA-site of the solvent. Compounds
2a–c, 3b,c and 4c formally show no dependence on solvent acidity
(a = 0). HBD solvents can interact both with oxygen atoms of the
hydroxy groups (donor) and carbonyl/cyano groups (acceptor).
Preferentially, protic solvents act with the oxygen atom of the
carbonyl groups and/or the cyano groups (a < 0). Interactions of
these kinds of solvents with the electron donor fragment caused
in a hypsochromic shift of the UV–Vis absorption maxima (a > 0),
the interactions with the electron acceptor moiety lead to a
bathochromic shift. If both effects provide a similar contribution
the coefficient a can be zero. The interactions of HBD solvents
with the acceptor moieties reduce their electron density,
consequentially the push–pull characters of the dyes increase.
Compounds 3a and 4a,b have a negative sign for coefficient a.
Thus, HBD solvents preferentially interact with the electron
acceptor moiety of the dye. Furthermore, the correlation analyses
show a negative sign of the coefficient b, which means that HBA
solvents interact with the acidic hydrogen atom of the hydroxy
groups. The solvent basicity provides the largest contributions.
Linear solvation energy correlation analyses
~
The influence of the solvent properties on the shift ofnmax has been
determined by means of multiple correlation analyses using
the solvent polarity parameter sets of Kamlet–Taft and Catalán.
The best correlation results are summarized in Tables 2 and 3.
The correlation coefficients r are about 0.90 for LSE relationships,
which indicates a high validity of the multiparameter equations
and allow significant conclusions. Specific solvent–solute interac-
tions through hydrogen bonding are expressed by the solvent
acidity a or SA, and by the solvent basicity b or SB.
The preferential location of the negative charge can be evalu-
ated and discussed by knowledge of the solvent-independent
coefficients, which are calculated by means of multiple correlation
analyses. The algebraic sign of coefficient a provides information
about the solvent interactions with the appropriate donor
fragment of the respective dye. A positive sign indicates a
hypsochromic shift and a negative sign a bathochromic shift. The
Table 2. Values of the solvent-independent correlation coefficients a, b and s for the Kamlet–Taft Eqn (1); wavenumber of the
~
reference system cyclohexane nmax;0, number of solvents n, correlation coefficient r, standard deviation sd, and significance f for
compounds 2–4
~
nmax;0
a
b
s
n
r
sd
f
2a
2b
2c
3a
3b
3c
4a
4b
4c
27.797
27.775
27.621
27.002
24.853
24.618
22.851
23.051
21.859
0
0
0
ꢂ2.669
ꢂ1.215
0
ꢂ3.317
ꢂ1.260
ꢂ0.264
ꢂ3.510
ꢂ4.975
ꢂ0.656
ꢂ1.494
ꢂ0.929
ꢂ1.190
ꢂ1.807
ꢂ0.936
ꢂ0.576
ꢂ0.731
ꢂ1.465
ꢂ0.496
28
29
36
28
29
36
33
29
36
0.874
0.884
0.933
0.911
0.967
0.932
0.944
0.990
0.938
0.434
0.205
0.116
0.484
0.108
0.073
0.385
0.218
0.101
< 0.0001
< 0.0001
< 0.0001
< 0.0001
< 0.0001
< 0.0001
< 0.0001
< 0.0001
< 0.0001
ꢂ0.464
0
0
ꢂ0.427
ꢂ0.865
0
Table 3. Values of the solvent-independent correlation coefficients a, b, c and d for Catalán Eqn (2); wavenumber of a reference system
(gas phase) ~nmax;0, number of solvents n, correlation coefficient r, standard deviation sd, and significance f for compounds 2–4
~
nmax;0
a
b
c
d
c/d
n
r
sd
f
2a
2b
2c
3a
3b
3c
4a
4b
4c
30.236
27.895
28.821
27.361
25.894
25.075
25.950
23.769
22.976
0
ꢂ3.242
ꢂ1.129
0
ꢂ3.327
ꢂ1.264
ꢂ0.171
ꢂ4.088
ꢂ2.982
ꢂ0.744
ꢂ2.121
0
ꢂ1.922
ꢂ1.164
ꢂ0.841
ꢂ2.355
ꢂ0.655
ꢂ0.507
ꢂ0.624
ꢂ0.863
ꢂ0.308
1.10
0
2.37
0
2.61
1.45
6.22
2.51
5.30
26
29
34
27
28
34
31
27
34
0.922
0.979
0.944
0.949
0.924
0.943
0.943
0.941
0.945
0.342
0.093
0.109
0.377
0.159
0.068
0.379
0.295
0.095
< 0.0001
< 0.0001
< 0.0001
< 0.0001
< 0.0001
< 0.0001
< 0.0001
< 0.0001
< 0.0001
+0.339
0
0
0
ꢂ1.989
0
ꢂ1.710
ꢂ0.737
ꢂ3.881
ꢂ2.168
ꢂ1.633
0
ꢂ1.790
ꢂ0.961
ꢂ0.179
wileyonlinelibrary.com/journal/poc
Copyright © 2012 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. (2012)

SOLVATOCHROMIC CATECHOL DERIVATIVES
Compounds 2c, 3c and 4c do not have classical hydrogen bond
donors, which results in a smaller amount of coefficient b. The
influence regarding the interaction of compounds 2c, 3c and 4c
with HBA solvents is negligible because the catechol moiety is
protected. The UV–Vis absorption maxima are also shifted to
higher wavelength with increasing solvent dipolarity and
polarizability (s < 0).
Altogether, the compounds 2–4 show positive solvatochromism
with regard to the parameter p*, which correlates with a higher
dipole moment of the electronically excited state.
mixture of solvent polarizability SP and dipolarity SdP in a relation
of 2 : 1.[19] This relation is not given for the discussed
chromophores, showing the c/d-value in Table 3, and can explain
the differences between the LSER of Kamlet–Taft and Catalán. In
comparison with catechol-based chromophores 2–4a, it is striking
that for 3a, with the least polarizable carbonyl group, the largest
solvatochromic extent and the largest d coefficient has been
measured. With increasing of absolute value of coefficient
c (influence of polarizability) from 3a!2a!4a the absolute values
of coefficient d (influence of dipolarity) decreases.
The correlation analyses, which use the solvent parameter of
Catalán, provide regression results with almost the same parame-
ter coefficient signs but allow a differentiation of the solvent
polarity in the two terms SP and SdP.
The plot of calculated wavenumbers as a function of measured
wavenumbers for 3a using the parameter set of Catalán is shown
in Fig. 2.
As mentioned above, in some solvent compounds with
hydroxy groups two UV–Vis absorption bands can be observed.
Generally, for catechol derivatives two aspects with regard to
solvatochromism and reaction with solvent molecules have to be
taken into account. On the one hand, the solvent basicity can cause
a deprotonation and on the other hand the acceptor strength of
the chromophore is important, because the rate of possible
deprotonation in a respective solvent increases with increasing
electron accepting ability. The longer wavelength UV–Vis absorp-
tion maxima of the anionic species are summarized in Table 4.
UV–Vis experiments of the compounds using the base DBU in
toluene clearly show that the UV–Vis absorption maximum after
dye deprotonation is observed in the same region of the longer-
wavelength absorption of the dyes as measured in DMSO, for
example (supporting information). Generally, for the anionic
species a smaller solvatochromic effect on ~nmax is observed.
The linear multiple regression analyses of the monoanions show
also high correlation coefficients r (Table 5), which allows
significant conclusions about the solvatochromic behaviour of
these compounds.
The regression analyses with the more recent parameter set of
Catalán were received with better correlation coefficients r, shown
in Table 3. The coefficient a shows a positive value for compound
2b; coefficient a is negative for 4a–c. Hence, the enhancement of
the push–pull character is due to both the preferential solvation
of the acceptor substituent by HBD solvents and the catechol
moiety by HBA solvents. For compounds 2a, 2c and 3a–c the term
a is formally zero, which shows that the preferential solvation of
the oxygen atom of the catechol and dimethoxy groups,
respectively, or of the carbonyl/cyano groups by HBD solvents is
equally. The substituent change from hydroxy groups to methoxy
groups leads gradually in smaller b values because with the
introduction of methoxy groups there are no more hydrogen bond
donor fragments in the molecules. Because of separated consider-
ation of polarizability SP and dipolarity SdP, more detailed informa-
tion about nonspecific interactions between solute and solvent
can be drawn. A bathochromic shift of lmax is observed caused
by the induced polarizability of the p-system through the solvent
molecules. It can be seen that the magnitude of the polarizability
SP (coefficient c) is higher than that of the dipolarity SdP
(coefficient d), excluding compounds 2b and 3a (c = 0). Otherwise,
negative signs for the terms c and d were observed, showing a
higher dipole moment of the electronically excited state. Catalán
et al. figured out that the Kamlet–Taft parameter p* reflects a
For anionic species a positive sign is determined for the
coefficient a. The UV–Vis absorption maximum undergoes a
hypschromic shift with increasing solvent acidity. Hydrogen
bond donating and electron pair accepting solvents interact
preferentially with the chromophore donor moiety (a > 0).
Figure 2. Results of the multiple correlation analyses of 3a (■) and (3a)– (○) using Catalán’s solvent parameter set
J. Phys. Org. Chem. (2012)
Copyright © 2012 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/poc

F. RIEDEL AND S. SPANGE
Table 4. UV–Vis absorption maxima of the anionic species of 2–4a,b in solvents in which two separated UV–Vis absorption bands
are observed
Solvent
~nmax[103 cm–1]
(3a)–
(2a)–
(2b)–
(3b)–
(4a)–
(4b)–
Methanol
Ethanol
1-Propanol
2-Propanol
1-Butanol
Water
1,2-Ethanediol
HMPA
Formamide
NMF
22.42
23.32
22.22
22.22
22.17
22.32
22.47
22.08
21.88
21.74
19.80
19.57
19.49
19.49
19.69
19.42
19.34
19.38
17.73
17.70
17.83
17.79
17.73
17.73
17.64
17.64
a
21.79
19.53
19.31
17.79
a
a
a
a
a
a
a
a
22.32
22.37
19.46
18.73
17.86
a
a
22.52
17.61
a
22.03
21.93
21.69
21.55
21.69
21.69
1760
19.42
20.37
19.34
19.31
19.42
19.27
19.31
1100
17.27
17.79
17.70
17.61
17.61
17.57
17.54
310
a
a
a
Dimethylformamide
DMSO
DMAA
21.60
21.51
21.51
21.60
1020
19.16
19.01
19.05
19.23
1080
17.30
17.30
17.45
17.51
550
TMU
Δ~n[cm–1]
aSolvent is not considered in multiple regression analysis (Table 5).
Table 5. Values of the solvent-independent correlation coefficients a, b, c and d for Catalán Eqn (2); wavenumber of a reference
system nmax;0, number of solvents n, correlation coefficient r, standard deviation sd, and significance f for anionic species (2–4a,b)–
~
~nmax;0
a
b
c
d
n
r
sd
f
(2a)–
(2b)–
(3a)–
(3b)–
(4a)–
(4b)–
22.246
23.802
21.170
16.506
18.637
17.351
+0.974
+1.000
+0.438
+1.670
+0.184
+0.389
ꢂ1.756
ꢂ0.687
ꢂ0.413
0
ꢂ1.339
ꢂ0.978
ꢂ1.916
+3.522
ꢂ2.039
+0.281
0
12
11
12
10
11
11
0.977
0.986
0.964
0.953
0.936
0.953
0.066
0.065
0.065
0.102
0.075
0.029
< 0.0001
< 0.0001
< 0.0001
< 0.0001
< 0.0001
< 0.0001
ꢂ0.407
ꢂ0.308
0
0
0
ꢂ0.593
0
Solvents with hydrogen bond acceptor and electron pair
donating ability show interactions with the hydroxy group in
(2–4a)– (b < 0). The negative charge of the phenoxide is stabilized
through an intramolecular hydrogen bond and a fast proton
exchange. With the deprotonated hydroxy group in 4-position
and 3-OMe-functionalization(2–4b)–, expectedly, the absolute
value of b is zero or at least smaller compared with that of the
catechol derivatives. The comparison of polarizability SP
(coefficient d) and dipolarity SdP (coefficient e) shows, that SdP
only has a marginal influence. Figure 2 presents the result of the
determined regression plot of (3a)– (○).
separation of these effects. Advantageously, the square correlation
analyses of the solvatochromism of the catechol derivatives using
Catalán’s solvent parameter show that the influence of dipolarity
rather than polarizability of solvent on solvatochromic probe are
crucial that a large solvatochromic extend is achieved. Thus, the
recent four empirical solvent scales of Catalán allow a more com-
prehensive interpretation of solvent effects on solvatochromism
compared with the established Kamlet–Taft solvent parameters.
UV–Vis absorption measurements of 3a in methanol show temper-
ature dependence caused by the protonation/deprotonation
equilibrium between 3a and (3a)–.
As can be seen, the coefficient a (influence of HBD solvents) of
the anionic species ((2ꢂ4a,b)–) is obtained with a positive value,
which indicates that the negative charge concentration is mostly
located on the catechol moiety of the dyes.
CONCLUSIONS
Various donor–acceptor p-conjugated compounds bearing
the catechol moiety were presented. The evaluation of the
solvatochromic properties of these catechol derivatives has been
carried out using alternatively the Kamlet–Taft and Catalán solvent
parameter scales. It can be clearly shown that Catalán’s improved
empirically determined solvents parameter set allows a differentia-
tion of polarizability versus dipolar effects on solvatochromism of
complex dyes and seem to be a valuable tool for quantitative
EXPERIMENTAL SECTION
General
Solvents for UV–Vis measurements were dried and distilled according to
standard procedures prior to use. Compounds 2–4 were synthesized in a
wileyonlinelibrary.com/journal/poc
Copyright © 2012 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. (2012)

SOLVATOCHROMIC CATECHOL DERIVATIVES
3
Knoevenagel condensation reaction using the aldehydes 1. NMR-
spectroscopical data for compounds 2a–c were published elsewhere.[34–36]
2-(3-Oxo-2,3-dihydroinden-1-ylidene)-malononitrile was prepared according
to a known procedure.[37] All other compounds were commercially available
and used as received. NMR spectra were recorded with a Bruker Avance 250
(Bruker Biospin GmbH, Silberstreifen 4, 76287 Rheinstetten, Germany) spec-
trometer using the solvent residue as the internal standard. Infrared (IR)
spectra were obtained with a BioRad FT-IR Spectrometer FTS 165. Elemental
analyses were measured using a Vario EL from Elementaranalysensysteme
GmbH Hanau. UV–Vis spectra were obtained with a MCS 400 diode-array
spectrometer from Carl Zeiss Jena with freshly prepared dye solutions
(c ꢃ 10–5 mol L–1). Multiple regression analyses were performed with the
statistical program ORIGINPRO 8G (OriginLab Corporation, One Roundhouse
Plaza, Northampton, MA 01060 USA). For temperature-dependent measure-
ments 40 mL of a methanolic solution of compound 3a (6.1ꢀ10–5 mol L–1)
was kept in a thermostated bath at different temperatures (267–333 K).
The UV–Vis spectrum at each temperature was measured using a UV–Vis
immersion cell with a path length of d=0.5nm.
1118 cm–1 1 H NMR (d, d6-DMSO, 250 MHz): 6.91 (1H, d, J = 8.4 Hz, ArH),
.
7.64 (1H, dd, 3J = 8.4 Hz, 4J = 2.0 Hz, ArH), 7.93 (4H, m, ArH), 8.29
(1H, s, CH), 8.46 (1H, d, 3J = 7.5 Hz, ArH), 9.63 (1H, bs, OH), 10.66
(1H, s, OH). 13 C NMR (d, d6-DMSO, 62.9 MHz): 69.4, 114.7, 114.9, 115.9,
121.1, 123.8, 124.2, 124.8, 125.5, 130.6, 135.1, 135.5, 136.9, 139.0, 145.3,
147.7, 153.4, 163.0, 186.3. e(DCM, lmax = 450 nm): 36,500 L mol–1 cm–1
.
2-(2-(4-Hydroxy-3-methoxybenzylidene)-3-oxo-2,3-dihydroinden-1-
yli-dene)malono-nitrile, 4b
Wine solid, Yield: 84%. Mp: 239–240 ꢄC. C20H12N2O3 (328.32) Anal. Found:
C73.29; H 3.70; N 8.70. Calcd: C 73.16; H 3.68; N 8.53. IR: 3401 n(OH), 3088
n(CꢂH), 2217, 2205 n(CN), 1698 n(C=O), 1555 n(C=C), 1503, 1293,
1144 cm–1 1 H NMR (d, d6-DMSO, 250 MHz): 3.89 (3H, s, OCH3), 6.95
.
(1H, d, 3J = 8.3 Hz, ArH), 7.57 (1H, dd, 3J = 8.3 Hz, 4J = 1.6 Hz, ArH), 7.91
(4H, m, H-2, ArH), 8.34 (1H, s, CH), 8.48(1H, d, 3J = 7.8 Hz, ArH), 10.82
(1H, bs, OH). 13 C NMR (d, d6-DMSO, 62.9 MHz): 56.0, 83.3, 102.6, 114.7,
115.1, 115.8, 116.2, 117.5, 124.1, 126.5, 135.4, 136.0, 137.4, 139.5, 147.7,
154.3, 163.6, 174.8, 177.7, 187.0. e(DCM, lmax = 473 nm): 38,000 L mol–1 cm–1
.
Preparation
2-(2-(3,4-Dimethoxybenzylidene)-3-oxo-2,3-dihydroinden-1-ylidene)-
malononitrile, 4c
An ethanolic solution of one equivalent of the acceptor was slowly added
to a stirred solution of one equivalent of the respective aldehyde. The
mixture was refluxed for 2 h. In case of compounds 2a–c, two drops of
piperidine were used as a catalyst. After precipitation the dye was
filtered, washed with cold ethanol and dried under reduced pressure.
Wine solid, Yield: 79%. Mp: 222–224 ꢄC. C21H14N2O3 (342.35) Anal. Found:
C 73.67; H 4.11; N 8.18. Calcd: C 73.68; H 4.12; N 8.18. IR: 3098, 3004
n(CꢂH), 2961, 2831 n(CH3), 2218, 2207 n(CN), 1692 n(C=O), 1588, 1553
n(C=C), 1503, 1433, 1352, 1283, 1235, 1154 cm–1 1 H NMR (d, d6-DMSO,
.
250 MHz): 3.88 (3H, s, OCH3), 3.91 (3H, s, OCH3), 7.19 (1H, d, 3J = 8.8 Hz,
ArH), 7.70 (1H, dd, 3J = 8.8 Hz, 4J = 2.3 Hz, ArH), 7.95 (4H, m, ArH), 8.36
(1H, s, CH), 8.51 (1H, d, 3J = 7.8 Hz, ArH). 13 C NMR (d, d6-DMSO,
62.9 MHz): 56.1, 56.3, 90.2, 110.6, 115.7, 116.1, 117.5, 122.4, 123.7, 124.1,
127.4, 135.9, 136.3, 139.4, 140.6, 145.7, 154.8, 164.6, 176.8, 176.3, 187.0.
2-(3,4-dihydroxybenzylidene)-2H-indene-1,3-dione, 3a
Yellow solid, Yield: 74%. Mp: 259–261ꢄC. C16H10O4 (266.25) Anal. Found: C
72.35; H 3.83. Calcd: C 72.18; H 3.79. IR: 3460 n(OH) (H-bonds), 3235 n(OH),
3093, 3041 n(CꢂH), 1721 n(C=O), 1665, 1594 n(C=C), 1551, 1530, 1381,
1295, 1244, 1185, 1161, 1094 cm–1 1 H NMR (d, d6-DMSO, 250 MHz): 6.90
.
e(DCM, lmax = 469 nm): 38600 L mol–1 cm–1
.
(1H, d, 3J = 8.4Hz, ArH), 7.65 (1H, s, CH), 7.84 (1H, dd, 3J = 8.4Hz, 4J= 2.0Hz,
ArH), 7.90 (4H, m, ArH), 8.34 (1H, d, 4J= 2.0Hz, ArH), 9.57 (1H, bs, OH),
10.49 (1H, bs, OH). 13 C NMR (d, d6-DMSO, 62.9 MHz): 116.1, 121.0, 122.9,
125.1, 125.4, 130.6, 135.6, 135.7, 139.5, 141.9, 145.5, 147.2, 152.9, 175.2,
SUPPORTING INFORMATION
189.3, 190.3. e(DCM, lmax = 392 nm): 38,000 L molꢂ1 cm–1
.
Kamlet–Taft’s and Catalán’s solvent parameters, UV–Vis absorption
maxima of compound 2–4, UV–Vis absorption spectra of 2–4a,b in
toluene, dimethyl sulfoxide and after addition of 1,8-diazabicyclo
[5.4.0]-undec-7-ene, UV–Vis absorption maxima of 3a in methanol
at different temperatures.
2-(4-hydroxy-3-methoxybenzylidene)-2H-indene-1,3-dione, 3b
Yellow solid, Yield: 85%. Mp: 209–210 ꢄC. C17H12O4 (280.27) Anal. Found:
C 73.35; H 4.33. Calcd: C 72.85; H 4.32. IR: 3091 n(CꢂH), 1721 n(C=O), 1680,
1573 n(C=C), 1509, 1437, 1389, 1279, 1177, 1158, 1019 cm–1 1 H NMR
.
(d,d6-DMSO, 250 MHz): 3.19 (3H, s, OCH3), 6.93 (1H, d, 3J = 8.3 Hz, ArH),
7.74 (1H, s, CH), 7.91 (5H, m, ArH), 8.70 (1H, d, 4J = 1.8 Hz, ArH), 10.61
(1H, bs, OH). 13 C NMR (d, d6-DMSO, 62.9 MHz): 55.2, 116.5, 121.4, 123.2,
125.9, 126.4, 130.1, 135.3, 135.9, 139.6, 142.1, 145.4, 147.3, 153.3, 175.6,
Acknowledgement
Financial support from the Fonds der Chemischen Industrie is
gratefully acknowledged.
189.8, 191.3. e(DCM, lmax = 417 nm): 39,300 L mol–1 cm–1
.
REFERENCES
2-(3,4-dimethoxybenzylidene)-2H-indene-1,3-dione, 3c
[1] C. Reichardt, Chem. Rev. 1994, 94, 2319.
Yellow solid, Yield: 87%. Mp: 203–204 ꢄC. C18H14O4 (294.30) Anal. Found:
C73.51; H 4.82. Calcd: C 73.46; H 4.79. IR: 3087, 3010 n(CꢂH), 2975, 2846
n(CH3), 1713 n(C=O), 1673, 1561 n(C=C), 1509, 1428, 1275, 1248, 1144,
[2] E. Grundwald, S. Winstein, J. Am. Chem. Soc. 1948, 70, 846.
[3] E. M. Kosower, J. Am. Chem. Soc. 1958, 80, 3253.
[4] E. M. Kosower, J. A. Skorcz, W. M. Schwarz, J. W. Patton, J. Am. Chem.
Soc. 1960, 82, 2188.
[5] U. Mayer, V. Gutmann, W. Gerger, Monatsh. Chem. 1975, 106, 1235.
[6] V. Gutmann, E. Wychera, Inorg. Nucl. Chem. Lett. 1966, 2, 257.
[7] C. Reichardt, Green Chem. 2005, 7, 339.
[8] K. Dimroth, C. Reichardt, T. Siepmann, F. Bohlmann, Lieb. Ann. Chem.
1963, 661, 1.
[9] C. Reichardt, E. Harbusch-Görnert, Lieb. Ann. Chem. 1983, 721.
[10] Y. Marcus, Chem. Soc. Rev. 1993, 22, 409.
1019 cm–1 1 H NMR (d, d6-DMSO, 250 MHz): 3.89 (3H, s, OCH3), 3.90 (3H, s,
.
OCH3), 7.16 (1H, d, 3J= 8.5Hz, ArH), 7.80 (1H, s, CH), 7.94 (4H, m, ArH), 8.04
(1H, dd, 3J= 8.5Hz, 4J = 1.8Hz, ArH), 8.69 (1H, d, 4J= 1.8Hz, ArH). 13 C NMR
(d, d6-DMSO, 62.9 MHz): 55.8, 56.8, 117.1, 122.9, 123.0, 125.2, 125.3, 131.9,
135.6, 136.0, 139.4, 141.9, 147.1, 147.6, 153.5, 176.1, 189.5, 190.2. e(DCM,
l
max = 414 nm): 39,000 L mol–1 cm–1
.
[11] R. W. Taft, M. J. Kamlet, J. Am. Chem. Soc. 1976, 98, 2886.
[12] R. W. Taft, M. J. Kamlet, J. Chem. Soc., Perkin Trans. 2 1979, 349.
[13] R. W. Taft, M. J. Kamlet, J. Chem. Soc. Perkin Trans. 2 1979, 1723.
[14] M. J. Kamlet, J.-L. M. Abboud, M. H. Abraham, R. W. Taft, J. Org. Chem.
1983, 48, 2877.
[15] R. W. Taft, M. J. Kamlet, J. Am. Chem. Soc. 1976, 98, 377.
[16] M. J. Kamlet, J.-L. M. Abboud, R. W. Taft, J. Am. Chem. Soc. 1977, 99, 6027.
2-(2-(3,4-Diyhdroxybenzylidene)-3-oxo-2,3-dihydroinden-1-ylidene)-
malononitrile, 4a
Red solid, Yield: 86%. Mp: 252–254 ꢄC. C19H10N2O3 (314.29) Anal. Found:
C72.65; H 3.32; N 8.93. Calcd: C 72.61; H 3.21; N 8.91. IR: 3318 n(OH),
3083 n(CꢂH), 2228, 2214 n(CN), 1703 n(C=O), 1527 n(C=C), 1505, 1297,
J. Phys. Org. Chem. (2012)
Copyright © 2012 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/poc

F. RIEDEL AND S. SPANGE
[17] M. J. Kamlet, T. H. Hall, J. Bodkin, W. R. Taft, J. Org. Chem. 1979, 44, 2599.
[18] J. Catalán, H. Hopf, Eur. J. Org. Chem. 2004, 4694.
[19] J. Catalán, J. Phys. Chem. B 2009, 113, 5951.
[20] J. Catalán, in: Handbook of Solvents (Eds: G. Wypych), ChemTec
Publishing, Toronto, New York, 2001, p. 583.
[28] F. Riedel, S. Spange, J. Phys. Org. Chem. 2009, 22, 203.
[29] O. Ya. Neiland, I. L. Kraupsha, I. Ya. Gudele, Chem. Heterocycl. Comp.
1993, 29, 1428.
[30] W. Linert, B. Strauss, E. Herlinger, C. Reichardt, J. Phys. Org. Chem.
1992, 5, 275–284.
[21] P. C. Beaumont, D. G. Johnson, B. J. Parsons, J. Chem. Soc. Faraday
Trans. 1993, 89, 4185.
[22] H. Schweppe, Handbuch der Naturfarbstoffe. Vorkommen - Verwendung
- Nachweis, Hüthig Jehle Rehm, Landsberg, 1992.
[31] W. Linert, R. F. Jameson, J. Chem. Soc. Perkin Trans. 2 1993,
1415–1421.
[32] L. G. S. Brooker, A. C. Craig, D. W. Heseltine, P. W. Jenkins, L. L. Lincoln,
J. Am. Chem. Soc. 1965, 87, 2443.
[23] R. W. Sabnis, Handbook of Acid–base Indicators, Taylor & Francis
Group, CRC Press, Boca Raton, 2008.
[24] K. Yumura, H. Otani, J. Mizuguchi, Dyes and Pigments 1998, 38, 227.
[25] A. Oehlke, A. A. Auer, K. Schreiter, K. Hofmann, F. Riedel, S. Spange,
J. Org. Chem. 2009, 74, 3316.
[33] C. Reichardt, in: Solvents and Solvent Effects in Organic Chemistry
(Eds: C. Reichardt), Wiley-VCH, Weinheim, 2003, p. 335.
[34] A. Gazit, P. Yaish, C. Gilon, A. Levitzki, J. Med. Chem. 1989, 32, 2344.
[35] R. Trotzki, M. M. Hoffmann, B. Ondruschka, Green Chem. 2008, 10,
767.
[26] M. D. Shultz, J. U. Reveles, S. N. Khanna, E. E. Carpenter, J. Am. Chem.
Soc. 2007, 129, 2482.
[27] J.-Y. Wach, B. Malisova, S. Bonazzi, S. Tosatti, M. Textor, S. Zürcher,
K. Gademann, Chem. Eur. J. 2008, 14, 10579.
[36] M. Mantri, O. de Graaf, J. van Veldhoven, A. Göblyös, J. K. von Frijtag
Drabbe Künzel, T. Mulder-Krieger, R. Link, H. de Vries, M. W. Beukers,
J. Brussee, A. P. Ijzerman, J. Med. Chem. 2008, 51, 4449.
[37] H. Junek, Monatsh. Chem. 1964, 95, 1201.
wileyonlinelibrary.com/journal/poc
Copyright © 2012 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. (2012)