β-Carotene: A green, inexpensive, and convenient solvatochromic probe for the determination of solvent polarizability

Article abstract of DOI:10.1016/j.dyepig.2012.07.017

Solvent polarizability has been previously determined by using the solvatochromic probe 3,20-di-tert-butyl-2,2,21,21-tetramethyl-3,5,7,9,11,13,15, 17,19-docosanonaene whose synthesis involves 15 steps. We show here that the natural dye β-carotene, 1,1′-(3,7,12,16-tetramethyl-1,3,5,7,9,11,13, 15,17-octadecanonaene-1,18-diyl)bis[2,6,6-trimethylcyclohexene], can be conveniently employed for the accurate determination of the same solvent property. This conclusion is based on both theoretical calculations and experimental data. The former includes free energies of solvation, and the wavenumber of the longest wavelength (i.e., the solvatochromic) transition. Both quantities for β-carotene correlate linearly with the corresponding values of the docosanonaene, with slopes and correlation coefficients of practically unity. The plot of experimentally calculated solvent polarizability of β-carotene versus that of the docosanonaene was found to be linear for 68 solvents. Previously unknown solvent polarizability values are reported for eight ROCH₂CH₂OH (R = C₁ to C₁₀) and four 1-allyl-3-R-imidazolium chloride ionic liquids (R = C₆ to C₁₀). The dependence of solvent polarizability on the number of carbon atoms in the hydrocarbon chains of several classes of solvents is calculated, it shows the importance of van der Waals interactions in ionic liquids.

Full text of DOI:10.1016/j.dyepig.2012.07.017

Dyes and Pigments 96 (2013) 16e24
Contents lists available at SciVerse ScienceDirect
Dyes and Pigments
journal homepage: www.elsevier.com/locate/dyepig
b-Carotene: A green, inexpensive, and convenient solvatochromic probe for the
determination of solvent polarizability
*
Carina Loffredo, Paulo Augusto R. Pires, Muhammad Imran, Omar A. El Seoud
Institute of Chemistry, University of São Paulo, P. O. Box 26077, 05513-970 São Paulo, SP, Brazil
a r t i c l e i n f o
a b s t r a c t
Article history:
Solvent polarizability has been previously determined by using the solvatochromic probe 3,20-di-tert-
Received 30 May 2012
Received in revised form
18 July 2012
Accepted 19 July 2012
Available online 3 August 2012
butyl-2,2,21,21-tetramethyl-3,5,7,9,11,13,15,17,19-docosanonaene whose synthesis involves 15 steps. We
show here that the natural dye
b
-carotene, 1,10-(3,7,12,16-tetramethyl-1,3,5,7,9,11,13,15,17-
octadecanonaene-1,18-diyl)bis[2,6,6-trimethylcyclohexene], can be conveniently employed for the
accurate determination of the same solvent property. This conclusion is based on both theoretical
calculations and experimental data. The former includes free energies of solvation, and the wavenumber
of the longest wavelength (i.e., the solvatochromic) transition. Both quantities for
linearly with the corresponding values of the docosanonaene, with slopes and correlation coefficients of
practically unity. The plot of experimentally calculated solvent polarizability of -carotene versus that of
b-carotene correlate
Keywords:
Solvatochromism
b
b-Carotene solvatochromism
the docosanonaene was found to be linear for 68 solvents. Previously unknown solvent polarizability
values are reported for eight ROCH2CH2OH (R ¼ C1 to C10) and four 1-allyl-3-R-imidazolium chloride
ionic liquids (R ¼ C6 to C10). The dependence of solvent polarizability on the number of carbon atoms in
the hydrocarbon chains of several classes of solvents is calculated, it shows the importance of van der
Waals interactions in ionic liquids.
2-Alkoxyethanols polarizability
Ionic liquids polarizability
Solvatochromic probes
Solvation free energy relationships
Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction
required is obtained by manipulating the spectra of pairs of
homomorphs (molecules that have the same, or a closely related
The influence of solvents on chemical phenomena (rate and
equilibrium constants; spectroscopic transitions, etc.) are usually
analyzed in terms of linear combinations of solvent parameters,
e.g.:
molecular structures, e.g., 4-nitroaniline and N,N-dimethyl-4-
nitroaniline for the calculation of SB). Fig. 1 shows typical exam-
ples of such solvatochromic probes. Of these, RB is employed for the
determination of the overall empirical solvent polarity scale ET(30),
that is the sum of the three left-hand parameters of Eqn. (1).
In Eqn. (1) the “cavity” term has been omitted, because the
excitation of these probes obeys the FrankeCondon principle.
Another variety of Eqn. (1) includes the term log P, for solvent
lipophilicity (¼partition coefficient of the solvent between water
and 1-octanol, both mutually saturated) [3]. This descriptor is
required in order to describe the solvatochromism of certain
probes, e.g., merocyanines [4].
Effect of solvent ¼ a SA þ b SB þ d=p SðD=PÞ
(1)
where (S) refers to solvent; (A, B, D/P) refer to solvent “acidity”,
“basicity”, and “dipolarity/polarizability”, and (a, b, d/p) are the
corresponding regression coefficients. These represent the
susceptibility of the phenomenon in question to the particular
solvent property [1,2].
The values of the solvent descriptors in Eqn. (1) are most readily
calculated by manipulating the Uvevis spectra, absorption or
emission, of solvatochromic substances (hereafter designated as
“probes”). The spectra of these compounds are responsive to the
properties of the medium. Because these probes are sensitive, to
varying extents, to more than one solvent property, the information
Catalán and Hopf, have shown that it is possible to separate
solvent dipolarity from its polarizability, by using the probe ttbP9.
In this case, the last term of Eqn. (1) should be expanded to
(d SD þ p SP) [5,6]. This is an interesting approach in view of the
importance of van der Waals interactions for diverse phenomena,
inter alia, the properties and solubilization of biopolymers in ionic
liquids, ILs [7e9], and the precipitation of petroleum asphaltenes
[10]. Obtaining ttbP9 is, however, a major synthetic undertaking,
involving an expensive, laborious route, composed of fifteen steps
[11]. For example, the cost of chemicals for obtaining 1 g of certain
* Corresponding author. Tel./fax: þ55 11 3091 3874.
E-mail address: elseoud@iq.usp.br (O.A. El Seoud).
0143-7208/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.dyepig.2012.07.017

C. Loffredo et al. / Dyes and Pigments 96 (2013) 16e24
t-Bu
17
t-Bu
H3C
N
H3C
N
O
O
A
C
B
D
F
t-Bu
CH3
O-
O
O-
O
N+
NH2
N+
N
N+
CH3
O-
O-
O-
CH3
CH3
N+
O
N+
F
N
G
E
O
CH3
CH3
H3C
H3C
H3C
CH3
CH3
H3C
H3C
CH3
CH3
H
H3C
H3C
CH3
CH3
CH3
H3C
CH3
CH3
CH3
CH3
CH3
I
Fig. 1. Typical examples of the probes that are employed for the determination of the solvent descriptors of Eqn. (1). These include the pairs of homomorphs: o-tert-butylstilbazolium
betaine (A) and o,o0-di-tert-butylstilbazolium betaine (B) (SA); 4-nitroaniline (C) and N,N-dimethyl-4-nitroaniline (D) (SB); 2-(N,N-dimethylamino)-7-nitrofluorene (E) and 2-fluoro-7-
nitrofluorene (F) (D/P). The probes 2,6-diphenyl-4-(2,4,6-triphenylpyridinium-1-yl) phenolate (RB) (G); 3,20-di-tert-butyl-2,2,21,21-tetramethyl-3,5,7,9,11,13,15,17,19-docosanonaene,
ttbP9 (H); and (all trans) 1,10-(3,7,12,16-tetramethyl-1,3,5,7,9,11,13,15,17-octadecanonaene-1,18-diyl)bis[2,6,6-trimethylcyclohexene],
the empirical solvent (overall) polarity, and polarizability, vide infra.
b-carotene (I) are employed for the determination of
intermediates (compound 27b from compound 23) is ca. 65 times
the cost of 1 g of a good commercial grade of -carotene [11]. These
difficulties are manifested by the fact that the only experimental
data about (SP) came from Catalán’s and Hopf’s research groups
[5,6,11,12].
calculations that showed perfect linear correlation between values
~
~
and nmax; ttbP9; calculated for 23 solvents.
ꢀcarotene; calculated
b
of nmax;
b
Finally, for solvent homologous series (1-alkanes, ROCH2CH2OH;
ROH; ILs) the dependence of (SP) on the number of carbon atoms
(NC) in the longest carbon-chain has been calculated and analyzed.
Fig. 1 shows that ttbP9 and
b-carotene have similar polyene
systems and should, in principle, be sensitive to solvent polariz-
2. Materials and methods
ability to the same extent. Indeed, solvent effects on the Uvevis
spectrum of
b
-carotene have been the subject of several theoret-
2.1. Chemicals
ical studies [13e17], and experimental ones for both pure solvents
[18,19], and binary mixtures [20,21]. The result of these studies e
that are relevant to the present one e is that the wavenumber of the
visible absorption maximum of the longest wavelength transition,
correlate linearly with the solvent LorenzeLorentz refractive index
(n) function, f(n2) ¼ (n2 ꢀ 1)/(n2 þ 2); deviations occur in the media
containing high proportion of F, Cl or S atoms, having very different
polarizability densities, as compared to (common) H, C, N and O.
With this background, and taking into account that natural
The chemicals were purchased from Alfa-Aeser, Merck, or
D-
PAC2 Química; the solvents were purified as given elsewhere [22]
ꢀ
and stored over activated type 4 A molecular sieves; RB was that
available from a previous study [23]. The purity of the solvents was
established by comparing their densities (DMA4500 digital density
meter, Anton Paar, Graz), and ET(30) with literature values [2,24].
The probe
b
-carotene (Fluka, purity ꢁ 97.0%) was employed
either as received, or after purification by crystallization from
ethanol/toluene. Both samples were divided into small portions,
put into tightly-stoppered glass vials, and stored at ca. ꢀ18 ꢂC until
used. All manipulation of this probe has been carried out under
nitrogen, in absence of direct light.
b-carotene is a green probe, readily available at low cost, we decided
to determine its use as a convenient substitute for ttbP9. We have
expanded the list of solvents for which the values of SPb-carotene are
available by additional 38 solvents, including the industrially impor-
tant 2-alkoxyethanols, ROCH2CH2OH (where R ¼ C1 to C10); and the
ionic liquids, ILs (whose polarizability has been determined for the
first time) 1-R-3-allylimizalolium chloride (where R ¼ C6 to C10).
Our results for several molecular solvents were found to be in
excellent agreement with those reported by Abboud et al.; the
2.2. Synthesis
2.2.1. 2-Alkoxyethanols [25]
The following is a general procedure: to a 3-necked round-bottom
flask, provided with a refluxcondenser, dropping funnel, and inlet for
dry nitrogen was added 50 g (0.8 mol) of dry ethylene glycol. Small
pieces of sodium (0.29 mol, 6.8 g), were slowly added to the glycol
under vigorous magnetic stirring and the mixture was heated to
60 ꢂC until the sodium dissolved. The appropriate 1-bromoalkane
~
~
values of nmax;
and nmax; ttbP9 for 68 solvents, hence the
b
ꢀcarotene
corresponding SP values, were found to correlate linearly with
slope and correlation coefficient, r, of practically unity. That is, both
probes can be conveniently employed in order to determine (SP).
This conclusion is corroborated by the results of theoretical

18
C. Loffredo et al. / Dyes and Pigments 96 (2013) 16e24
(0.25 mol for each RBr, 37.76 g; 41.27 g; 48.28 g; 55.30 g for R ¼ C5; C6;
C8 and C10, respectively) was added slowly (ca. 30 min) and the
solutionwas then kept under reflux for the appropriate time (1 h; 3 h;
and 4 h for R ¼ C5; C6; C8 and C10, respectively). The precipitated NaBr
was filtered off and the residue was washed with water; the upper
layer was separated and dried with anhydrous MgSO4, and the
product purified by fractional distillation.
imidazolium chloride, AlHxImCl; 1-allyl-3-(1-heptyl)imidazolium
chloride, AlHpImCl; 1-allyl-3-(1-octyl)imidazolium chloride AlO-
cImCl; 1-allyl-3-(1-decyl)imidazolium chloride, AlDcImCl. N-(1-alkyl)
imidazole (0.22 mol; 33.49 g, 36.58 g, 39.66 g, 45.8 g for R ¼ C6; C7; C8
and C10) and 3-chloro-1-propene (0.25 mol; 19.13 g) were mixed in
a glass reactor, provided with a reflux condenser and inlet for dry
nitrogen; the reactor was introduced in the cavity of the MW appa-
ratus (Discover model DU-8316; CEM; Matthews) and irradiated
under vigorous stirring for 1 h; 20 W, 80 ꢂC.
The product was dissolved in ethyl acetate (100 mL); the solu-
tion placed in a freezer (phase separation of the IL); the upper layer
discarded; this process was repeated thrice. After removal of ethyl
acetate, the IL was dried at 110 ꢂC, 1 mmHg for 12 h. The yields were
98 ꢃ 1% (light yellow viscous liquid in all cases).
The boiling points and densities of the solvents were in good
agreement with literature values [25]. The 1H NMR data (Bruker
DPX-300 NMR spectrometer, 300 MHz for 1H; CDCl3;
d
in ppm, J in
Hz) of the synthesized 2-alkoxyethanols are listed below, the
numbering of atoms is shown in Fig. 2.
2.2.1.1. 2-Pentyloxyethanol
Colorless liquid: C5H11OCH2CH2OH; yield: 16 g (46%); b.p.:
186e187 ꢂC/753 torr; d240: 0.8917; 1H NMR: 0.91 (3H, t, J ¼ 8.0, H7),
1.29e1.37 (4H, m, H6), 1.61 (2H, t, J ¼ 6.0, H5), 3.51 (4H, m, H2, 4),
3.72 (2H, t, J ¼ 6.0, H3), 2.91 (1H, H1).
For elemental analyses, the (hygroscopic) ILs were transformed
into their (non-hygroscopic) perchlorate counterparts, as follows: to
a solution containing 0.5 g of the IL in 5 mL water was added 5 mL of
an aqueous solution containing 0.5 g NaClO4. The IL perchlorate
(colorless viscous liquids) were suspended in water several times;
the water was separated; and the IL dried at 40 ꢂC, under reduced
pressure, over P4O10 until constant weight. The results of elemental
analyses and the 1H NMR data of AlRImCl (Bruker DRX-500 NMR
2.2.1.2. 2-Hexyloxyethanol
Colorless liquid: C6H13OCH2CH2OH; yield: 22 g (59%); b.p.:
111e113 ꢂC/27 torr; d240: 0.8859; 1H NMR: 0.91 (3H, t, J ¼ 8.0, H7),
1.29e1.36 (6H, bs, H6), 1.60 (2H, t, J ¼ 6.0, H5), 3.52 (4H, m, H2, 4),
3.73 (2H, t, J ¼ 6.0, H3), 2.86 (1H, H1).
spectrometer; 500 MHz for 1H; CDCl3;
d in ppm, J in Hz) are listed
below; Fig. 3 shows the numbering of the hydrogen atoms.
2.2.1.3. 2-Octyloxyethanol
2.2.3.1. 1-Allyl-3-(1-hexyl)imidazolium chloride
Colorless liquid: C8H17OCH2CH2OH; yield: 24 g (55%); b.p.:
130e132 ꢂC/22 torr; d240: 0.8795; 1H NMR: 0.91 (3H, t, J ¼ 8.0, H7),
1.29e1.36 (6H, bs, H6), 1.60 (2H, t, J ¼ 6.0, H5), 3.52 (4H, m, H2, 4),
3.73 (2H, t, J ¼ 6.0, H3), 2.86 (1H, H1).
Light-yellow, viscous liquid. 1H NMR for C12H21N2Cl: 0.88 (3H, t,
J ¼ 7.0, H13),1.27e1.40 (6H, m, H12),1.78e1.97 (2H, m, H11), 4.34 (2H,
t, J ¼ 7.7, H10), 5.07 (2H, d, J ¼ 6.6, H6), 5.46 (1H, d, J ¼ 4.2, H9), 5.49
(1H, s, H8), 6.03 (1H, m, H7) 7.25 (2H, s, H4 & H5), 10.98 (1H, s, H2).
Calcd. for C12H21N2ClO4: C, 49.2%; H, 7.2%; N, 9.6%. Found: C, 49.1%;
H, 7.2%; N, 9.7%.
2.2.1.4. 2-Decyloxyethanol
Colorless liquid: C10H21OCH2CH2OH; yield: 23 g (45%); b.p.:
156e157 ꢂC/20 torr; d240: 0.8715; 1H NMR: 0.92 (3H, t, J ¼ 8.0, H7),
1.23e1.36 (14H, bs, H6), 1.79 (2H, t, J ¼ 6.0, H5), 3.45 (4H, m, H2, 4),
3.58 (2H, t, J ¼ 6.0, H3), 2.17 (1H, H1).
2.2.3.2. 1-Allyl-3-(1-heptyl)imidazolium chloride
Light-yellow, viscous liquid. 1H NMR for C13H23N2Cl: 0.88 (3H, t,
J ¼ 7.0, H13),1.27e1.40 (8H, m, H12),1.82e1.97 (2H, m, H11), 4.34 (2H,
t, J ¼ 7.7, H10), 5.07 (2H, d, J ¼ 6.6, H6), 5.46 (1H, d, J ¼ 4.2, H9), 5.49
(1H, s, H8), 6.03 (1H, m, H7) 7.25 (2H, s, H4 & H5), 10.96 (1H, s, H2).
Calcd. for C13H23N2ClO4: C, 50.9%; H, 7.6%; N, 9.1%. Found: C, 50.9%;
H, 7.6%; N, 9.2%.
2.2.2. N-(1-alkyl)imidazoles
These compounds were obtained from the reaction of the sodium
salt of imidazole and 1-bromoalkanes, as follows: to a 3-necked
round-bottom flask, provided with a reflux condenser, dropping
funnel, and inlet for dry nitrogen was added a solution of imidazole
(7.49 g; 0.11 mol) in dry ethanol (25 mL). Small pieces of sodium
(2.53 g, 0.11 mol), were slowly introduced, the mixture was vigor-
ously stirred until the metal has dissolved and then heated under
reflux for 2 h. The mixture was cooled to room temperature; the
appropriate 1-bromoalkane (0.10 mol; 16.50 g, 17.91 g, 19.31 g,
22.12 g for R ¼ C6; C7; C8 and C10, respectively) was added slowly (ca.
30 min); then the mixture was refluxed for 2 h. The mixture was
cooled to room temperature, the precipitated NaBr was filtered off;
the ethanol evaporated, the residual oil was washed with water, and
then dried with anhydrous MgSO4. The N-(1-alkyl)imidazoles were
purified by fractional distillation (yields before distillation ca. 98%);
their purity was confirmed with 1H NMR spectroscopy (results not
shown).
2.2.3.3. 1-Allyl-3-(1-octyl)imidazolium chloride
Light-yellow, viscous liquid. 1H NMR for C14H25N2Cl: 0.88 (3H, t,
J ¼ 7.0, H13), 1.20e1.39 (10H, m, H12), 1.88e1.96 (2H, m, H11), 4.34
(2H, t, J ¼ 7.4, H10), 5.07 (2H, d, J ¼ 6.5, H6), 5.46 (1H, m, H9), 5.48
(1H, m, H8), 6.03 (1H, m, H7) 7.26 (2H, s, H4 & H5), 10.93 (1H, s, H2).
Calcd. for C14H25N2ClO4: C, 52.4%; H, 7.9%; N, 8.7%. Found: C, 52.5%;
H, 7.9%; N, 8.8%.
2.2.3.4. 1-Allyl-3-(1-decyl)imidazolium chloride
Light-yellow, viscous liquid. 1H NMR for C16H29N2Cl: 0.88 (3H, t,
J ¼ 7.2, H13), 1.19e1.40 (14H, m, H12), 1.92 (2H, dxt, J ¼ 7.4, H11), 4.33
(2H, t, J ¼ 7.6, H10), 5.07 (2H, d, J ¼ 6.5, H6), 5.46 (1H, d, J ¼ 7.6, H9),
5.48 (1H, s, H8), 6.02 (1H, m, H7) 7.20 (2H, s, H4 & H5), 11.03 (1H, s,
H2). Calcd. for C16H29N2ClO4: C, 55.1%; H, 8.4%; N, 8.0%. Found: C,
55.1%; H, 8.5%; N, 8.1%.
2.2.3. 1-Allyl-3-R-imidazolium chlorides for R ¼ C6; C7; C8 and C10
These ILs were obtained by the reaction of the appropriate N-(1-
alkyl)imidazole with allyl chloride (3-chloro-1-propene), under
microwave (MW) irradiation as follows, for 1-allyl-3-(1-hexyl)
H2
H8
H
H6
H7
N+
N CH2CH2(CH2)nCH3
H11 H12
H10
H13
HOCH2CH2OCH2CH2(CH2)nCH3
H5
H4
H9H
H1
H2
H4 H5
H3
H7
H6
Fig. 2. Numbering of the hydrogen atoms of 2-alkoxyethanols.
Fig. 3. Numbering of the hydrogen atoms of the ionic liquid cation.

C. Loffredo et al. / Dyes and Pigments 96 (2013) 16e24
19
2.3. Uvevisible spectroscopic measurements
parameterized data are available in the software package employed
(Gamess US, vide Calculations). The solvent set employed covers
nonpolar and weakly polar aliphatic and aromatic hydrocarbons,
dipolar aprotic solvents, and aliphatic alcohols. Gibbs free energies
These have been carried out at 25 ꢃ 0.1 ꢂC by using a Shimadzu
UV-2500 spectrophotometer, equipped with a model 4029 digital
thermometer (Control Company, Friendsmood), under the
following experimental conditions: each spectrum was recorded at
least thrice, at 140 nm/min; slit width 0.5 nm; sampling interval
0.2 nm. The accuracy of lmax was routinely checked against the
known peaks of a holmium oxide glass filter (model 666-F1, Hellma
Analytics, Müllheim). Values of lmax were determined from the first
derivative of the absorption spectra, calculated by using the
commercial software Grams/32 [26]; the uncertainty in lmax is
ꢃ0.2 nm.
~
~
of solvation, and the values of nmax; ꢀcarotene; Theo and nmax; ttbP9; Theo
b
were compared. The results of these calculations are listed inTable 1.
At the outset, the quality of geometry optimization of both probes
can be shown by the agreement between our calculated bond
lengths and angles with those calculated by other authors, and with
X-ray data, see Tables SM-1 and SM-2. Several points can be deduced
from Table 1: (1) the values of
DGsolvation for both solvents correlate
linearly; (2) in any solvent, the solvation of
b-carotene is more
favorable than that of ttbP9; (3) the values of theoretically calculated
~
and experimental values of nmax are in reasonable agreement.
The first point is clear from Eqn. (4); it indicates that the ener-
gies of solvation of the two probes are similar.
3. Results and discussion
Note: the procedures and equations that have been employed
for all calculations are given in the “Calculations” section.
D
Gsolvation;
¼ ꢀ1:3464 þ 1:1431
ꢀcarotene
D
Gsolvation; ttbP9
b
ðr ¼ 0:9988; sd ¼ 0:1899; Ns ¼ 23Þ
ð4Þ
3.1. Relevance of the purity of
previously published data
b-carotene; agreement with
The second point reflects the more lipophilic character of
-carotene, as shown by the corresponding values of log P [28]; 13.1
b
and 14.8 for ttbP9 and b-carotene, respectively. Because each probe
carries a polyene chain consisting of 9 conjugated all-trans double
bonds, then the difference must be due to the methyl groups on the
The deceptively simple task of obtaining highly pure b-carotene
has been discussed in details elsewhere [27]. We have employed
this probe with and without recrystallization. Impurities, if they are
present or develop during probe manipulation, are not expected to
affect our results, provided that they do not absorb at the same lmax
of the solvatochromic peak. The reason is that it is peak position
~
polyene chain, plus the cyclohexene rings of
b-carotene. This
conclusion is corroborated by calculating Gibbs free energies of
solvation (gas phase / solvent) of the compounds shown in Fig. 4;
these model the termini of both probes.
The calculations showed that the solvation of compound 1 (model
for the termini of b-carotene) is more favorable than that of 2 (model
(i.e., lmax) and not intensity that is employed in calculating nmax
,
hence SP. The following has been done in order to assess whether it
is appropriate to use
b-carotene as received: the values of
for the end groups of ttbP9). The corresponding values of DGsolvation,
~
nmax; ꢀcarotene, calculated from the spectra of recrystallized and
b
non-crystallized samples were correlated, as shown by Eqn. (2),
where r ¼ correlation coefficient; sd ¼ standard deviation;
NS ¼ number of solvents employed:
Table 1
~
DGsolvation, and values of nmax,
Theoretically calculated Gibbs free energy of solvation,
for -carotene and di-tert-butyl-tetramethyl-docosanonaene, ttbP9.
b
~
nmax; ꢀcaroteneðprobe used as receivedÞ ¼ ꢀ754
b
Solvent
b
-carotene
Gsolvation
ttbP9
~
þ 1:03 nmax; ꢀcaroteneðrecrystallized sampleÞ;
(2)
b
a
a
~
~
~
~
%
Dnmax
D
,
nmax
,
%
Dnmax
D
Gsolvation
,
nmax
,
kcal molꢀ1 cmꢀ1
kcal molꢀ1 cmꢀ1
r ¼ 0:9984; sd ¼ 18:4888; Ns ¼ 8
Gas phase
n-Heptane
Cyclohexane
e
22,443 ꢀ0.17
21,215 4.25
21,126 3.82
21,113 2.41
21,081 2.24
21,072 2.46
e
24,767 ꢀ3.30
21,924 ꢀ0.88
21,841 ꢀ1.29
21,833 ꢀ2.56
21,791 ꢀ2.74
21,777 ꢀ2.41
~
Additionally, we have plotted nmax;
(sample used as
ꢀcarotene
b
ꢀ22.31
ꢀ24.20
ꢀ18.09
ꢀ20.01
ꢀ21.57
ꢀ21.05
ꢀ20.60
received) against those published by Abboud et al., [19] (b-carotene
recrystallized from benzeneemethanol). As shown by Eqn. (3), both
sets are in excellent agreement. In subsequent work, therefore,
-carotene has been employed as received.
Dichloromethane ꢀ25.93
Trichloromethane ꢀ25.49
Tetra-
ꢀ25.21
b
chloromethane
1,2-Dichloro
ethane
Benzene
Nitromethane
Acetonitrile
Butanenitrile
Pentanenitrile
1,4-Dioxane
THF
Acetone
DMSO
DMF
Methanol
Ethanol
ꢀ23.03
21,073 2.30
ꢀ19.00
21,785 ꢀ2.50
~
nmax; ꢀcaroteneðprobeusedasreceivedÞ ¼ 237:27
b
ꢀ24.11
ꢀ18.46
ꢀ20.79
ꢀ22.14
ꢀ22.54
ꢀ13.24
ꢀ20.49
ꢀ21.86
ꢀ12.64
ꢀ17.41
ꢀ17.65
ꢀ18.77
ꢀ18.53
ꢀ18.01
ꢀ17.68
ꢀ17.07
ꢀ16.22
21,001 2.53
21,189 e
21,269 3.67
21,223 e
ꢀ19.56
ꢀ15.13
ꢀ17.22
ꢀ18.37
ꢀ18.70
ꢀ10.08
ꢀ16.77
ꢀ18.10
ꢀ9.93
21,696 ꢀ2.46
21,915 ꢀ2.08
22,003 ꢀ1.40
21,954 ꢀ5.11
21,914 ꢀ5.64
21,850 ꢀ1.62
21,871 ꢀ1.94
21,971 ꢀ5.30
21,839 ꢀ3.77
21,809 ꢀ5.83
22,038 ꢀ0.96
21,962 ꢀ1.39
21,909 ꢀ1.19
21,879 ꢀ1.31
21,841 ꢀ1.53
21,818 ꢀ1.68
21,805 ꢀ1.77
~
þ0:9882nmax; ꢀcaroteneðAbboud’sdata; recrystallizedsampleÞ;
b
r ¼ 0:9985; sd ¼ 15:617; Ns ¼ 23
ð3Þ
21,185 e
3.2. Can
solvent polarizability?
b-carotene be reliably employed for the determination of
21,142 2.67
21,152 3.00
21,239 3.49
21,118 1.04
21,092 2.37
21,304 4.18
21,234 3.97
21,184 3.82
21,156 3.71
21,122 2.70
21,102 3.35
21,090 e
In order to answer this question, we have employed two strat-
ꢀ14.10
ꢀ14.40
ꢀ15.34
ꢀ15.09
ꢀ14.61
ꢀ14.27
ꢀ13.71
ꢀ12.92
egies: (i) we compared theoretically calculated energies of solva-
~
tion and ðnmaxÞ, hence SP, of
b-carotene and of ttbP9 for a large set of
1-Propanol
1-Butanol
1-Hexanol
1-Octanol
1-Decanol
representative solvents; (ii) we have compared the corresponding
~
experimental nmax values.
3.2.1. Theoretical calculations
With regard to (i), the following procedure has been employed
for each probe: its geometry in gas phase has been optimized; the
geometry-optimized probe was solvated in 23 solvents, for which
a
~
~
~
~
%
Dnmax ¼ 100 ꢄ ðnmax experimental ꢀ nmax TheoÞ=ðnmax experimentalÞ. The experi-
mental data for
elsewhere [5].
b-carotene are listed in Table 2 below; those for ttbP9 are published

20
C. Loffredo et al. / Dyes and Pigments 96 (2013) 16e24
kcal/mol, are
(
b
-carotene; ttbP9; solvent): ꢀ5.80, ꢀ4.51 (in
benzene); ꢀ4.03, ꢀ2.85 (in 1-octanol); ꢀ3.12, ꢀ2.04 (in 1,4-dioxane).
~
For each probe, plot of theoretical versus experimental nmax
showed correlation coefficients of, after excluding the worst
offenders (22% of the data), 0.8389 and 0.8217, for
ttbP9, respectively.
b-carotene and
A clear indication, however, that both probes are proportionally
sensitive to solvent polarizability can be concluded from Fig. 5 that
~
~
depicts the correlation of nmax;
and nmax; ttbP9; Theo for
b
ꢀcarotene; Theo
23 solvents; the corresponding equation is:
~
~
nmax;
¼ 1574 þ 0:895 nmax; ttbP9; Theo
b
ꢀcarotene; Theo
(5)
ðr ¼ 0:9974; sd ¼ 5:2490; Ns ¼ 23Þ
3.2.2. Experimental data
~
3.2.2.1. Is the value of nmax affected by the experimental conditions?
~
The accuracy of the value of nmax may be slightly affected if the
peak full width at half height (FWHH), or its shape are noticeably
dependent on experimental conditions, in particular the dye
concentration, and the nature of the solvent [17]. Based on curve
deconvolution (see item 5.3 in Calculations), we have found that this
is not the case. For example, the FWHH is the same (uncertainty
~
Fig. 5. Correlation between theoretically calculated transition energies, nmax, of ttbP9
and -carotene.
b
Both results show that our initial expectation is confirmed,
i.e., because the polyene systems are the same, both indicators
respond similarly to solvent polarizability. Therefore b-carotene is
within 3%) for five different concentrations of
b
-carotene in isooctane
[(1.6e16.1)
ꢄ
10ꢀ5 mol/L]. For the same dyes concentration
(1.2 ꢄ 10ꢀ5 mol/L) in 4 distinct solvents (hexane; dichloromethane,
1-hexanol, and THF) both peak shape and FWHH are the same
(uncertainty in FWHH 1e4%). An example of these deconvolutions is
shown in Fig. 6. Therefore, the experimental conditions have no
a convenient probe for SP as argued in Introduction;
(ii) The above conclusion can be corroborated by comparing the
dependences of SP and LorenzeLorentz refractive index
function, vide supra, on the number of carbon atoms, NC, of the
aliphatic chain for series of homologous compounds, e.g.,
alkanes, see Eqns. (8) and (9):
~
bearing on the accuracy of nmax, hence on the calculated values of SP.
~
3.2.2.2. Comparison of the experimental nmax; Exp, and the calculated
SP values
~
~
nmax; ttbP9; Exp are
The experimental values of nmax;
;
b
ꢀcarotene; Exp
SPAlkanes;
¼ 0:46877 þ 0:0263 NC ꢀ 0:0008 NC2
ꢀcarotene
b
listed in Table 2, along with the corresponding SP, and the refractive
index (n).
ðr ¼ 0:99154; sd ¼ 0:00418; NS ¼ 8Þ
Reduced f n2
ð8Þ
ð9Þ
À
Á
Regarding these data, the following is relevant:
¼ 0:49956 þ 0:0296 NC ꢀ 0:00094 NC2
Alkanes
ðr ¼ 0:99494; sd ¼ 0:00345; NS ¼ 8Þ
(i) The agreement between the results of the two probes can be
~
demonstrated from plots of nmax or SP, the latter is shown in
Fig. 7; the corresponding correlations are given by:
1.2
1.0
0.8
0.6
0.4
0.2
0.0
~
~
nmax;
¼ ꢀ1348 þ 1:08 nmax; ttbP9; Exp
b
ꢀcarotene; Exp
(6)
(7)
r ¼ 0:9934; sd ¼ 38:85; Ns ¼ 68
SPbꢀcarotene ¼ ꢀ0:04 þ 1:04 SPttbP9
sd ¼ 0:0105; Ns ¼ 68
;
r ¼ 0:9934;
18000
20000
22000
24000
26000
28000
-1
Wavenumber, cm
Fig. 6. Example of the deconvolution of the band of the longest wavelength transition
of
b
-carotene. [
b
-carotene] ¼ 1.2 ꢄ 10ꢀ5 mol Lꢀ1, in isooctane. The upper curve
represents an overlap between the calculated spectrum (red color; sum of the
deconvoluted bands) and the experimental (black color, not seen because of overlap of
the two curves). (For interpretation of the references to color in this figure legend, the
reader is referred to the web version of this article.)
Fig.
4. Model
compounds
(1,2,3,3-tetramethylcyclohexene
and
2,2,4,4-
tetramethylpentane) employed for the calculation of Gibbs free energy of solvation
of the termini of -carotene and ttbP9, respectively.
b

C. Loffredo et al. / Dyes and Pigments 96 (2013) 16e24
21
Table 2
~
Wavenumbers of the visible absorption maximum of the longest-wavelength transition, nmax of ttbP9 and
b
-carotene, the corresponding solvent polarizability (SPs) and the
refractive index (n).
Entry Solvent
Gas phase
nmax; ttbP9, cmꢀ1a
nmax; ꢀcarotene, cmꢀ1
SPttbP9
SPb
na
~
~
-carotene
b
23,975
24,402c
e
e
1
Aliphatic and alicyclic hydrocarbon
1
n-Pentane
21,880
21,786b
21,793b
21,799
21,564
21,591
21,732
21,646
21,614
21,564
21,520
21,491
21,350
22,271c
22,251d
22,181c
22,202c
21,964c
22,030d
22,157c
22,070d
21,985c
21,953c
21,917c
21,894c
21,788d
0.5935
0.6201
0.6181
0.6164
0.6830
0.6754
0.6354
0.6598
0.6688
0.6830
0.6955
0.7037
0.7436
0.5803
0.5858
0.6048
0.5991
0.6639
0.6460
0.6114
0.6351
0.6582
0.6669
0.6767
0.6830
0.7119
1.358
1.376
1.392
1.375
1.426
1.399
1.387
1.405
1.411
1.422
1.429
1.434
1.475
2
3
4
3-Methylpentane
2,2,4-Trimethylpentane
n-Hexane
5
Cyclohexane
6
7
Methylcyclohexane
n-Heptane
8
n-Nonane
9
n-Decane
10
11
12
13
n-Dodecane
n-Tetradecane
n-Hexadecane
Decalins (mixture)
Halogenated compounds
14
15
16
17
18
19
20
21
Dichloromethane
21,288
21,210
21,265
21,253
21,529
21,035
21,289
21,776
21,634c
21,564c
21,603c
21,570c
21,891c
21,394c
21,697d
22,165d
0.7612
0.7833
0.7677
0.7711
0.6929
0.8329
0.7609
0.6229
0.7538
0.7729
0.7623
0.7712
0.6838
0.8192
0.7367
0.6092
1.424
1.444
1.460
1.445
1.402
1.524
1.465
1.377
Trichloromethane
Tetrachloromethane
1,2-Dichloroethane
1-Chlorobutane
Chlorobenzene
Fluorobenzene
Hexafluorobenzene
Aromatic compounds
22
Benzene
21,176
21,216
21,183
21,079
21,017
20,771
21,546c
21,583c
21,595c
21,401c
21,434d
21,160d
0.7929
0.7816
0.7909
0.8204
0.8380
0.9076
0.7778
0.7677
0.7642
0.8173
0.8083
0.8829
1.501
1.493
1.501
1.516
1.541
1.615
23
Toluene
24
Xylenes
25
Anisole
26
Tetralin
27
1-Methylnaphthalene
Alcohols
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
Methanol
Ethanol
21,829
21,739
21,652
21,739b
21,595
21,660b
21,745b
21,549
21,622b
21,511
21,563b
21,459
21,520b
21,363b
20,935
22,234c
22,111c
22,026c
22,130c
21,971c
22,028c
22,145c
21,936c
21,988c
21,877c
21,942c
21,834c
21,876c
21,705c
21,262c
0.6079
0.6334
0.6581
0.6334
0.6742
0.6558
0.6317
0.6873
0.6666
0.6980
0.6833
0.7127
0.6955
0.7399
0.8612
0.5904
0.6239
0.6471
0.6187
0.6620
0.6465
0.6147
0.6716
0.6574
0.6876
0.6699
0.6993
0.6879
0.7345
0.8551
1.329
1.359
1.384
1.377
1.382
1.397
1.386
1.409
1.406
1.418
1.415
1.429
1.426
1.453
1.540
1-Propanol
2-Propanol
1-Butanol
2- Butanol
tert-Butanol
1-Pentanol
2-Pentanol
1-Hexanol
2-Hexanol
1-Octanol
2-Octanol
Cyclopentanol
Benzyl alcohol
Nitriles
43
Acetonitrile
21,699
22,080c
0.6448
0.6324
1.344
Ethers
44
45
46
47
Diethyl ether
Di-iso-propyl ether
Di-n-butyl ether
1,4-Dioxane
THF
21,798
21,769b
21,603b
21,372
21,455
22,208c
22,183d
21,922c
21,722c
21,806c
0.6167
0.6249
0.6720
0.7374
0.7139
0.5975
0.6043
0.6754
0.7298
0.7070
1.353
1.368
1.399
1.422
1.407
48
Ketones
49
50
Acetone
2-Butanone
Acetophenone
21,677
21,613b
20,981
22,007c
21,971c
21,256c
0.6510
0.6691
0.8482
0.6522
0.6520
0.8568
1.359
1.379
1.532
51
Esters
52
53
Diethyl carbonate
Propylene carbonate
Ethyl acetate
21,633
21,342
21,660
21,996c
21,734d
22,028c
0.6635
0.7459
0.6558
0.6552
0.7266
0.6465
1.384
1.421
1.372
54
Carboxylic acid amides
55
DMF
21,296
21,281
21,604c
21,594c
0.7589
0.7632
0.7620
0.7647
1.431
1.438
56
DMAc
Sulfoxides
57
Pyridines
58
Amines
59
DMSO
21,047
21,004
20,714
21,340c
21,331c
21,039d
0.8295
0.8416
0.9238
0.8339
0.8363
0.9158
1.479
1.510
1.586
Pyridine
Aniline
(continued on next page)

22
C. Loffredo et al. / Dyes and Pigments 96 (2013) 16e24
Table 2 (continued )
~
~
Entry
60
Solvent
Triethylamine
nmax; ttbP9, cmꢀ1a
nmax; ꢀcarotene, cmꢀ1
SPttbP9
0.6603
SPb
na
-carotene
b
21,644
22,039c
0.6439
1.400
Miscellaneous solvents
61
Hexamethylphosphoric acid triamide
21,347b
20,445
21,676
21,529
21,031
21,110
21,470
21,881
21,592d
20,730d
22,102c
21,915c
21,296c
21,415c
21,951c
22,267c
0.7445
1.0000
0.6513
0.6929
0.8340
0.8116
0.7096
0.5932
0.7653
1.0000
0.6264
0.6773
0.8459
0.8135
0.6675
0.5814
1.458
1.627
1.369
1.390
1.529
1.470
1.379
1.363
62
Carbon disulfide
63
Acetic acid
64
Acetic anhydride
65
66
67
68
N-Methylimidazole
1-Methylpyrrolidin-2-one
Nitromethane
Petroleum ether b.p.: 50e70 ꢂC
2-Alkoxyethanols
69
2-Methoxy ethanol
2-Ethoxy ethanol
2-Propoxy ethanol
2-Butoxy ethanol
2-Pentyloxy ethanol
2-Hexyloxy ethanol
2-Octyloxy ethanol
2-Decyloxy ethanol
e
e
e
e
e
e
e
e
21,833c
21,825c
21,820c
21,806c
21,800c
21,781c
21,767c
21,711c
e
e
e
e
e
e
e
e
0.6996
0.7018
0.7032
0.7070
0.7086
0.7138
0.7176
0.7328
1.402e
1.408e
1.414e
1.420e
1.423e
1.429e
1.435e
1.450e
70
71
72
73
74
75
76
Ionic liquids
77
78
79
80
AlHxImCl
AlHpImCl
AlOcImCl
AlDcImCl
e
e
e
e
21,574c
21,545c
21,527c
21,479c
e
e
e
e
0.7702
0.7781
0.7830
0.796
1.5134c
1.5090c
1.5061c
1.4889c
a
From Catalán and Hopf (2004) [5].
From Catalán (2009) [6].
Present work.
From Abe et al. (1998) [19].
From Cooper and Partridge (1950) [25].
b
c
d
e
interactions (on the expense of van der Waals interactions)
may explain the larger response of the ROH series to NC, relative
to the 2-alkoxyethanol counterpart;
The same correlations for aliphatic alcohols, 2-alkoxyethnols,
and ILs are listed in Eqns. (1)e(6) of item 3 of SM. The relevant point
in all these correlations is that the regression coefficients of SP and
f (n2) are similar, a clear indication that
b
-carotene and f (n2)
(iv) Entries 77e80 of Table 2 show (SP) of ILs. Although the
number of ILs tested is small, the results show that the
dependence of SP on NC is relatively large, in agreement with
the view that dipolar interactions are important for ILs with
short side-chain, whereas hydrophobic interactions predom-
inate for ILs with R > C6. That is, ILs and possibly their solu-
tions in some molecular solvents should be regarded as micro-
heterogeneous media, due to the formation of hydrophobic
domains, akin to micellar aggregates [32e35].
respond similarly to solvent polarizability. The advantage of using
the probe is that it reports on its local solubilization site. Conse-
quently, it can be employed for determining polarizability in cases
where f (n2) is either experimentally inaccessible, or has dubious
meaning, e.g., the interfacial region of micellar aggregates; the
surfaces of biopolymers, etc.;
(iii) Table 2 contains previously unreported data for 2-alkoxye-
thanols. In addition to being employed extensively as indus-
trial solvents and in household cleaning formulations [29,30],
the bifunctional molecular structure of ROCH2CH2OH leads to
the formation of several inter- as well as intra-molecular H-
bonded species [31]. The strength of these dipolar head-group
In summary, the results of theoretical calculations and experi-
mental ones point in the same direction, namely b-carotene and
ttbP9 respond similarly to (SP). This similarity, as evidenced by Eqn.
(7), and Fig. 7, argues convincingly, that both probes can be employed
interchangeably for the calculation of solvent polarizability.
1.0
0.9
0.8
0.7
0.6
0.5
4. Conclusions
The green, commercially available, low cost b-carotene can be
conveniently employed for the determination of medium polariz-
ability instead of ttbP9 that is obtained by an elaborate, rather
expensive synthetic route. Any impurity, if it is present, has no
consequence on the spectroscopic parameter of interest, lmax of the
solvatochromic peak. The FWHH and shape of the latter are inde-
pendent of dye concentration, or the nature of solvent. Theoretical
~
calculations of nmaxðNS ¼ 23Þ, and experimental correlation
~
between the values of nmaxðNS ¼ 68Þ have shown that both probes
are sensitive, almost to the same extent, to solvent polarizability. New
SP data are reported for important homologous series of molecular
solvents and ILs. The dependence of (SP) on NC of four homologous
series shows the importance of solvent head-group interactions
(ROEtOH) and that van der Waals interactions are important in ILs.
The removal of the synthesis “hurdle” is important for under-
standing the role of solvent, an important aspect of green chemistry.
0.6
0.7
0.8
SPttbP9
0.9
1.0
Fig. 7. Correlation between the SP values of ttbP9 and b-carotene.

C. Loffredo et al. / Dyes and Pigments 96 (2013) 16e24
23
5. Calculations
Appendix A. Supplementary material
5.1. Geometry optimization of the molecular structures of
carotene, ttbP9, and the model compounds
b
-
Supplementary data related to this article can be found online at
http://dx.doi.org/10.1016/j.dyepig.2012.07.017.
The most stable conformations of the natural dye, s-cis form of
trans- -carotene; ttbP9; 2,2,4,4-tetramethylpentane, and 1,2,3,3-
tetramethylcyclohexene had their geometries optimized by using
DFT calculation that employed B3LYP density functional and 6-
31G(d,p) basis set.
For both probes, the transition energies between the funda-
mental- and excited state were calculated by TD-DFT, using BOP
density functional and DH (Dunning/Hay “double zeta”) basis set
References
b
[1] El Seoud OA. Understanding solvation. Pure Appl Chem 2009;81:697e708.
[2] Reichardt C, Welton T. Solvents and solvent effects in organic chemistry. 4th
ed. Weinheim: Wiley: VCH; 2011.
[3] Leo AJ, Hansch C. Role of hydrophobic effects in mechanistic QSAR. Perspect
Drug Discov Des 1999;17:1e25.
[4] Martins CT, Lima MS, El Seoud OA. Thermosolvatochromism of merocyanine
polarity indicators in pure and aqueous solvents: relevance of solvent lip-
ophilicity. J Org Chem 2006;71:9068e79.
~
[5] Catalán J, Hopf H. Empirical treatment of the inductive and dispersive
components of solute-solvent interactions: the solvent polarizability (SP)
scale. Eur J Org Chem 2004:4694e702.
[6] Catalán J. Toward a generalized treatment of the solvent effect based on four
empirical scales: dipolarity (SdP, a new scale), polarizability (SP), acidity (SA),
and basicity (SB) of the medium. J Phys Chem B 2009;113:5951e60.
[7] Lindman B, Karlström G, Stigsson L. On the mechanism of dissolution of
cellulose. J Mol Liquids 2010;156:76e81.
with the application of long-range correction. Values of nmax of the
excited states with the largest oscillator strength are shown in
Table 1. Solvent effects were simulated by using the Polarizable
Continuum Model (PCM) combined with SMD solvation model. The
solvation free energy, DGsolvation, was calculated from the difference
between the free energy of the probe (ground state) in the solvent
and in the gas phase. All calculations were performed by using the
version 1 OCT 2010 (R3) of Gamess program package [36].
[8] Liu H, Sale KL, Holmes BM, Simmons BA, Singh S. Understanding the inter-
actions of cellulose with ionic liquids: a molecular dynamics study. J Phys
Chem B 2010;114:4293e301.
[9] El Seoud OA, Silva VC, Possidonio S, Casarano R, Arêas EPG, Gimenes P.
Microwave-assisted derivatization of cellulose, 2 e the surprising effect of the
structure of ionic liquids on the dissolution and acylation of the biopolymer.
Macromol Chem Phys 2011;212:2541e50.
5.2. Calculation of solvent SP and reduced f (n2)
[10] Gray MR, Tykwinski RR, Stryker JM, Tan X. Supramolecular assembly model
for aggregation of petroleum asphaltenes. Energy Fuels 2012;25:3116e24.
[11] Klein D, Kiliçkiran P, Mlynek C, Hopf H, Dix I, Jones PG. A general route to fully
terminally tert-butylated linear polyenes. Chem Eur J 2010;16:10507e22.
[12] Catalán J, Hopf H, Klein D, Martus M. On the photophysics of polyenes. 1.
Bathochromic shifts in their S0(11Agꢀ) / S2(11Buþ) electronic transitions
caused by the polarizability of the medium. J Phys Chem A 2008;112:5653e7.
[13] Canuto S, Coutinho K, Trzesniak D. New developments in Monte Carlo/
The solvent polarizability scale (SP) is calculated by Eqn. (10) [5]:
~
~
nmax gas ꢀ nmax
solvent
SP ¼
(10)
~
~
nmax gas ꢀ nmax
CS2
~
where nmax
is the wavenumber of the probe in the gas phase,
gas
quantum mechanics methodology. The solvatochromism of
b-carotene in
~
and nmax
is the correspondent value in CS2. The negligible vapor
CS2
different solvents. Adv Quantum Chem 2002;41:161e83.
pressure of both probes precludes a direct determination of
[14] Martins JBL, Durães JA, Sales MJA, Vilela AFA, Silva GM, Gargano R. Theoretical
investigation of carotenoid ultraviolet spectra. Int J Quantum Chem 2009;109:
739e45.
~
nmax
~
gas. This value was obtained by extrapolating to zero the plot
of nmax versus f (n2) for a homologous series of linear alkanes. For
b
-
[15] Ruiz-Anchondo T, Glossman-Mitnik D. Computational characterization of the
carotene the resulting equation is:
b-carotene molecule. J Mol Struct THEOCHEM 2012;913:215e20.
[16] Renge I, Sild E. Absorption shifts in carotenoidsdinfluence of index of
refraction and submolecular electric fields. J Photochem Photobiol A Chem
2011;218:156e61.
[17] Ouyang S-L, Sun C-L, Zhou M, Li D-F, Wang WW, Qu G-N, et al. Effect of
solution concentration on the structured order and optical properties of
short-chain polyene biomolecules. Sci China Phys Mech Astron 2010;53:
1646e50.
À
Á
nmax
¼ 24402 ꢀ 9624 f n2
ðr ¼ 0:99188;
~
gas
(11)
sd ¼ 19:3666; Ns ¼ 8Þ
The equation for calculating f (n2) is given in Introduction. The
reduced value of this function is given by:
[18] Myers AB, Birge RR. The effect of solvent environment on molecular electronic
oscillator strengths. J Chem Phys 1980;73:5314e21.
À
Á
[19] Abe T, Abboud J-LM, Belio F, Bosch E, Garcia JI, Mayoral JA, et al.
Empirical treatment of solventesolute interactions: medium effects on
f n2
À
À
Á
Reduced f n2
¼
(12)
solvent
Á
f n2
the electronic absorption spectrum of b-carotene. J Phys Org Chem 1998;
11:193e200.
[20] Mancini PM, Pérez AC, Vottero LR. Nonspecific soluteesolvent interactions in
binary solvent mixtures containing an aprotic hydrogen-bond acceptor and
a hydrogen-bond donor: dipolarity/polarizability and refractive index. J Solu-
tion Chem 2001;30:695e707.
CS2
This equation is equivalent to Eqn. (10), because f (n2)gas ¼ 0.
[21] Mancini PM, Adam CG, Fortunato GG, Vottero LR. A comparison of nonspecific
solvent scales. Degree of agreement of microscopic polarity values obtained
by different measurement methods. Arkivoc 2007;xvi:266e80.
[22] Armarego WLF, Chai C. Purification of laboratory chemicals. 5th ed. New York:
Elsevier; 2003.
[23] Silva PL, Pires PAR, Trassi MAS, El Seoud OA. Solvation in pure liquids: what
can be learned from the use of pairs of indicators? J Phys Chem B 2012;112:
14976e84.
5.3. Deconvolution of the Uvevis spectrum of
b
-carotene
-carotene has
The deconvolution of the absorption bands of
b
been carried out by using five individual vibronic features and
assuming they have Gaussian shape, as described elsewhere [37],
with the help of Grams/32 v.5 software [26].
[24] Lide DR. CRC handbook of chemistry and physics. 85th ed. Boca Raton: CRC
Press; 2004.
[25] Cooper FC, Partridge MW. Antituberculous compounds part VI. p-(w-alkox-
yalkoxy)-N-arylbenxamidines and analogues. J.Chem.Soc 1950:459e64.
[26] Galactic Ind. Corp. Grams/32. Version 5. Salem: 1997.
[27] Laughlin RG, Bunke GM, Eads CD, Laidig WD, Shelley JC. Preparation and
Acknowledgments
We thank FAPESP (State of São Paulo Research Foundation) for
financial support, and the following for research fellowships: Pet-
robrás SA (C. Loffredo); CNPq (National Council for Scientific
Technological Research) (O. A. El Seoud); CNPq/The Academy of
Sciences for the Developing World, TWAS (M. Imran). We thank C.
Guizzo for his help.
physical characterization of pure
b-carotene. Chem Phys Lipids 2002;115:
63e76.
[28] ACD Labs. Advanced chemistry development (ACD/Labs) software V11.02.
Web version of SciFinder scholar. Available from: <http://scifinder.cas.org>;
2012 [accessed 24.03.12].
[29] Traynor MJ, Wilkinson SC, Williams FM. Metabolism of n-butoxyethanol in
excised human skin in vitro. Toxicol Lett 2008;177:151e5.

24
C. Loffredo et al. / Dyes and Pigments 96 (2013) 16e24
[30] Ullmann’s encyclopedia of industrial chemistry. Weinheim: Wiley-VCH; 2003.
[31] D’Angelo M, Onori G, Santucci A. Self-association of monohydric alcohols in
water: compressibility and infrared absorption measurements. J Chem Phys
1994;100:3107e13.
[34] Canongia Lopes JN, Padua AGH. Nanostructural organization in ionic liquids.
J Phys Chem B 2006;110:3330e5.
[35] Wang Y, Voth GA. Tail aggregation and domain diffusion in ionic liquids.
J Phys Chem B 2006;110:18601e8.
[32] Urahata SM, Ribeiro MCC. Structure of ionic liquids of 1-alkyl-3-
[36] Schmidt MW, Baldridge KK, Boatz JA, Elbert ST, Gordon MS, Jensen JH, et al.
General atomic and molecular electronic structure system. J Comput Chem
1993;14:1347e63.
methylimidazolium cations:
J Chem Phys 2004;120:1855e63.
[33] Urahata SM, Ribeiro MCC. Single particle dynamics in ionic liquids of 1-
a systematic computer simulation study.
[37] Chen Z, Lee C, Lenzer T, Oum K. Solvent effects on the S0(11Agꢀ) / S2(11Buþ
)
alkyl-3-methylimidazolium cations.
J
Chem Phys 2005;122:24511-
transition of
b-carotene, echinenone, canthaxanthin, and astaxanthin in
1e24511-9.
supercritical CO2 and CF3H. J Phys Chem A 2006;110:11291e7.