Dichroic and spectral properties of anthraquinone-based azo dyes for PVA polarizing film

Article abstract of DOI:10.1016/j.dyepig.2011.06.024

Dichroic dyes with broad absorption bands are advantageous for the production of neutral gray polarizing films. For this purpose, anthraquinone-based azo dyes were synthesized to have a conjugated bichromophore design and this was applied to PVA polarizin

Full text of DOI:10.1016/j.dyepig.2011.06.024

Dyes and Pigments 92 (2011) 737e744
Contents lists available at ScienceDirect
Dyes and Pigments
journal homepage: www.elsevier.com/locate/dyepig
Dichroic and spectral properties of anthraquinone-based azo dyes for PVA
polarizing film
,
Jae Bok Chang a, Sim Bum Yuk a, Jong S. Park b, Jae Pil Kim a
*
a Department of Materials Science and Engineering, Seoul National University, Seoul 151-744, Republic of Korea
b Department of Polymer & Textile Industry, Dong-A University, Busan 604-714, Republic of Korea
a r t i c l e i n f o
a b s t r a c t
Article history:
Dichroic dyes with broad absorption bands are advantageous for the production of neutral gray polar-
izing films. For this purpose, anthraquinone-based azo dyes were synthesized to have a conjugated
bichromophore design and this was applied to PVA polarizing films. The spectral and dichroic properties
of the polarizing films were examined and compared to the films produced with a general disazo dichroic
dye. The polarizing films with anthraquinone-based azo dyes had wider absorption bands, but lower
dichroic ratios compared to the films with a general disazo dichroic dye. It was found that the alignment
between the transition moments of the anthraquinone and azo chromophores is an important factor for
the dichroic properties of anthraquinone-based azo dyes.
Received 19 March 2011
Received in revised form
26 May 2011
Accepted 2 June 2011
Available online 6 July 2011
Keywords:
Anthraquinone
Bichromophore
Charge transfer transition
Dichroic ratio
Crown Copyright Ó 2011 Published by Elsevier Ltd. All rights reserved.
Polarizing films
Transition moment
1. Introduction
matching for neutral gray becomes complicated when the number
of dyes used increases. Therefore, broadening of absorption ranges
The polarizing films are linear polarizers that absorb light in
one particular vibration direction. Commercial polarizing films for
liquid crystal displays (LCDs) are manufactured from oriented
poly(vinyl alcohol) films containing polarizing materials, such as
iodine or dichroic dyes. Iodine types have superior transmittance
and polarizing efficiency but they are vulnerable to heat and
humidity [1e3]. In contrast, dye types have inferior optical prop-
erties compared with iodine types but they have higher resistance
to heat and humidity [4e6]. As the flat panel display market grows
and the demand for highly durable polarizing film increases, dye
types are gaining more industrial attention.
of dyes is helpful for easier color matching and higher color
reproducibility. However, recent studies have mostly been focused
on the improvement of transmittance and polarizing efficiency of
dye type polarizing film [4e8].
In this study, anthraquinone-based azo dyes, which have an
anthraquinone and two azo chromophores, were designed and
synthesized for dichroic dyes with broader absorption spectra than
the dyes used in our previous work [9]. PVA polarizing films were
produced with the dyes, and their spectral and dichroic properties
were examined and compared with those of the disazo dye (DA).
Poly azo dyes are the most widely used class of dyes for polar-
izing films because they are advantageous in adsorption on PVA
and they achieve higher dichroism due to their linear structure and
anisotropic transition moment. Generally, poly azo dyes have nar-
rower light absorption bands compared to iodine anions (Iꢀ3 and I5ꢀ)
[7,8]. Therefore, various dyes of different colors should be appro-
priately blended for the even absorption of visible light. This color
2. Experimental
2.1. Materials and instrumentation
1,5-Diamino-anthraquinone and 2,6-diamino-anthraquinone
were purchased from SigmaeAldrich and 7-amino-4-hydroxy-2-
naphthalenesulfonic acid (J-acid) and 4-amino-acetophenone
were obtained from TCI. All other chemicals used in this study were
of synthesis grade. Poly(vinyl alcohol) film was supplied by Kuraray
Co. Ltd., with degree of polymerization of 2400, a degree of
* Corresponding author. Tel.: þ82 2 880 7187; fax: þ82 2 880 7238.
E-mail address: jaepil@snu.ac.kr (J. P. Kim).
saponification of 99.9% and its thickness was 75 mm.
0143-7208/$ e see front matter Crown Copyright Ó 2011 Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.dyepig.2011.06.024

738
J.B. Chang et al. / Dyes and Pigments 92 (2011) 737e744
O
O
O
N+N
NH2
HO3S
2
NH2
nitrosylsulfuric acid
22~25
+
N N+
H2N
OH
O
HO3S
NH2
O
O
N
N
OH
pH 8~9
not excess 35
OH N
N
H2N
SO3H
Scheme 1. Synthesis of 26B.
The 1H NMR spectra were recorded on a Bruker Avance 500
2.2. Synthesis of dyes
spectrometer using DMSO-d6 and TMS as the solvent and
the internal standard, respectively. The FT-IR spectra were
recorded on a Thermo Scientific Nicolet 6700 using ATR method.
The mass spectra were recorded in fast atom bombardment
(FAB) ionization mode using a JEOL JMS-AX505WA/HP 6890
Series II Gas Chromatography-Mass Spectrometer. The absorp-
tion spectra were measured on an HP 8452A spectrophotometer
which was equipped with a Glan-Thompson polarizer. The color
differences of the polarizing films were analyzed using a Scinco
Colormate.
2.2.1. Synthesis of anthraquinone-based azo dyes
Anthraquinone-based azo dyes were synthesized as shown in
Scheme 1. 2,6-Diamino-anthraquinone (1.43 g, 0.006 mol) were
slowly introduced into concentrated sulfuric acid (10 mL) and
temperature of the mixture was kept below 40 ꢁC. After stirring for
30 min at room temperature, 40% nitrosyl sulfuric acid (2.6 mL) was
added dropwise maintaining the mixture at room temperature. The
mixture was stirred for 4 h at room temperature and then it was
quenched with ice water (140 mL). After stirring for 30 min, the
N
HO3S
NH2
HO3S
NH2
HO3S
N
HCl / NaNO2
0~5
+
OH
OH
OH
pH 8~9 0~5
NH2
HO3S
N
NH2
N
N
N
HCl / NaNO2
0~5
HO3S
OH
+
O
O
OH
NH2
HO3S
N
N
HO3S
OH
pH 8~9
0~5
N
N
OH
O
Scheme 2. Synthesis of DA.

J.B. Chang et al. / Dyes and Pigments 92 (2011) 737e744
739
NH2
OH
H2N
O
SO3H
HO
SO3H
α
β
N
N
N
N
O
N
O
N
O
N
N
HO3S
NH2
HO3S
OH
OH
15B
15A
NH2
HO3S
OH
HO3S
NH2
O
O
N
N
O
N
N
α
β
NH2
OH
NH2
N
N
OH
N
N
O
H2N
SO3H
HO
SO3H
26B
26A
Fig. 1. Structures of anthraquinone-based azo dyes.
diazonium salts were obtained by filtration of the mixture. The
diazonium salts were added to 30 mL of water with 5 mL of 2 M
hydrochloric acid and the mixture was stirred for 30 min.
The ensuing mixture was dropped into a solution of 7-amino-4-
hydroxy-2-naphthalenesulfonic acid (2.87 g, 0.012 mol) dissolved
in water (100 mL) for coupling reaction. In the course of dropping,
the temperature and pH of the mixture were maintained at 0e5 ꢁC
and 9e10, respectively. The reaction mixture was stirred for 2 h at
0e5 ꢁC and then another 5 h at room temperature. The mixture was
acidified with 2 M hydrochloric acid and the precipitate was
collected by filtration. The crude product (26B) was washed with
brine and ethanol and dried in a vacuum oven. The other
anthraquinone-based azo dyes were prepared in a similar manner.
2.2.2. Synthesis of disazo dye
A disazo dye (DA) was synthesized as shown in Scheme 2. 7-
Amino-4-hydroxy-2-naphthalenesulfonic acid (2.65 g, 0.01 mol)
was dissolved in water (50 mL) at room temperature and pH of the
solution was adjusted to 7.0 by adding 10% sodium carbonate
solution. The solution was cooled to 0e5 ꢁC and 2.5 mL of 2 M aq
sodium nitrite was added dropwise. The ensuing solution was
dropped into 2 M hydrochloric acid (50 mL) and stirred for 1 h,
after which time, a small portion of sulfamic acid was added as
a nitrous acid scavenger. The resulting diazonium salt solution was
added to a coupling component solution of 7-amino-4-hydroxy-2-
naphthalenesulfonic acid (2.65 g, 0.01 mol) dissolved in water
(80 mL), while the temperature and pH of the mixture were
Table 1
Yield, 1H NMR and mass data of the synthesized dyes.
Dye Yield (%) 1H NMR (ppm, DMSO-d6)
Mass
1
(m zꢀ , 100%, [M þ H]þ
)
15A 81.3
15B 82.4
26A 83.1
26B 81.6
DA 67.8
7.08 (s, 2H, NH), 7.95 (s. 2H, ArH), 8.01 (d, J ¼ 8.5 Hz, 2H, ArH), 8.28 (d, J ¼ 8.6 Hz, 2H, ArH), 8.31 (t, J ¼ 8.1 Hz, 2H, ArH), 8.40 738.08
(s, 2H, ArH), 8.45 (d, J ¼ 7.4 Hz, 2H, ArH), 8.65 (d, J ¼ 7.8 Hz, 2H, ArH), 10.13 (s, 2H, OH), 13.92 (s, 2H, NH)
6.89 (s, 4H, NH2), 7.55 (s, 2H, ArH), 7.66 (d, J ¼ 8.4 Hz, 2H, ArH), 7.74 (d, J ¼ 8.6 Hz, 2H, ArH), 7.94 (s, 2H, ArH), 8.18
(t, J ¼ 8.0 Hz, 2H, ArH), 8.21 (d, J ¼ 7.5 Hz, 2H, ArH), 8.33 (d, J ¼ 7.9 Hz, 2H, ArH), 15.81 (s, 2H, OH)
7.17 (s, 2H, NH), 7.45 (s, 2H, ArH), 7.65 (d, J ¼ 8.3 Hz, 2H, ArH), 7.73 (d, J ¼ 8.5 Hz, 2H, ArH), 7.98 (s, 2H, ArH), 8.15
(s, 2H, ArH), 8.29 (d, J ¼ 7.4 Hz, 2H, ArH), 8.30 (d, J ¼ 7.7 Hz, 2H, ArH), 10.89 (s, 2H, OH), 12.54 (s, 2H, NH)
6.67 (s, 4H, NH2), 7.56 (s, 2H, ArH), 7.69 (d, J ¼ 8.3 Hz, 2H, ArH), 7.78 (d, J ¼ 8.4 Hz, 2H, ArH), 8.01 (s, 2H, ArH), 8.18
(s, 2H, ArH), 8.26 (d, J ¼ 7.5 Hz, 2H, ArH), 8.45 (d, J ¼ 7.8 Hz, 2H, ArH), 16.01 (s, 2H, OH)
738.08
738.08
738.08
635.08
6.55 (s, 2H, NH2), 6.66 (s, lH, ArH), 6.69 (d, J ¼ 8.6 Hz, lH, ArH), 7.29 (s, 1H, ArH), 7.51 (s, 1H, ArH), 7.80 (s, 1H, ArH), 7.82
(d, J ¼ 8.4 Hz, 2H, ArH), 7.87 (d, J ¼ 8.1 Hz, 1H, ArH), 7.93 (d, J ¼ 9.5 Hz, 1H, ArH), 7.98 (d, J ¼ 8.7 Hz, 1H, ArH), 8.24
(d, J ¼ 9.1 Hz, 2H, ArH), 15.7 (s, lH, OH), 15.9 (s, lH, OH)

740
J.B. Chang et al. / Dyes and Pigments 92 (2011) 737e744
Table 2
O
O
O
D
Spectral properties of the synthesized dyes.
D
a
a
b
Dye
lmax
3max
Half-band
lmax
Half-band
(nm)
(L molꢀ1 cmꢀ1
)
widtha (nm)
(nm)
widthb (nm)
15A
15B
26A
26B
DA
534
528
526
538
532
21,000
22,000
21,000
23,000
31,000
234
230
228
205
95
576
548
580
534
556
172
176
154
158
90
O
a
E
In water.
In PVA film.
b
O
O
D
O
O
maintained at 0e5 ꢁC and 8e9, respectively, in the course of
the addition. The reaction mixture was stirred for 2 h and the
precipitate was filtered, washed with brine and dried in a vacuum
oven. The intermediate was heated in ethanol for 2 h under reflux,
hot filtered, washed with hot ethanol and then dried in a vacuum
oven.
D
Fig. 2. Charge transfer transitions of 1,5- and 2,6-substituted anthraquinone.
A mixture of 4-amino-acetophenone (0.54 g, 0.004 mol), 2 M aq
hydrochloric acid (36 mL) and sodium nitrite (0.28 g, 0.004 mol)
was stirred for 1 h at 0e5 ꢁC. To this solution, a small portion of
sulfamic acid was added as a nitrous acid scavenger. The resulting
diazonium salt solution was added to a coupling component solu-
tion where the intermediate (2.14 g, 0.004 mol) was dissolved in
water (150 mL) while the temperature and pH of the mixture were
maintained at 0e5 ꢁC and 8e9, respectively, in the course of the
addition. After coupling, the solution was stirred for 2 h and the
precipitate was filtered, washed with brine and dried in a vacuum
oven. The crude product (DA dye) was refluxed in ethanol for 2 h,
hot filtered, washed with hot ethanol and then dried in a vacuum
oven.
the conjugate gradient minimizer algorithm [12]. CONFLEX
conformational search procedure was used for finding low-energy
conformations of the dye molecules [13]. The semi-empirical
method, PM5, was also examined with respect to geometry opti-
mization. On optimization of conformers, the dye molecules were
assumed to have azo linkages in the trans forms.
Aspect ratios (l/d) of dye molecules were calculated from the
geometry of the most stable conformer optimized by CONFLEX/
MM3 calculation in consideration of van der Waals radius, where l
and d represent the length of the long axes and diameter of the
circumscribed cylinders of the dye molecules, respectively. Direc-
tions of the transition dipole moments of the dichroic dyes were
examined by INDO/1 in the ZINDO package [14] using the dye
structures optimized by the above-mentioned calculation. Config-
uration interaction (CI) calculation up to 676 configuration functions
was performed by adapting the occupied orbitals of (HOMO ꢀ 25 to
HOMO) and the vacant orbitals of (LUMO to LUMO þ 25). Each
configuration function was generated through the single excitation
(S-CI) from the ground-state function [15,16].
2.3. Preparation of polarizing films
Dye baths were prepared with a dye (0.5 g), Na2SO4 (1.5 g) and
distilled water (1 L). PVA films (50 ꢂ 70 mm) were treated in the
dyeing solution for 6 min at 40 ꢁC. The dyed PVA films were drawn
6 times in 2wt% of boric acid solution at 50 ꢁC. The stretched films
were washed with water and dried in an oven for 10 min.
2.4. Investigation of optical properties of polarizing films
3. Results and discussion
The optical properties of the polarizing films were measured
with a UVevis spectrophotometer equipped with a Glan-Thompson
polarizer. Dichroic ratio (R), single-piece transmittance (Tsp) and
polarizing efficiency (PE) were evaluated at the absorption
maximum of the polarizing films according to Eqs. (1)e(3) [10].
3.1. Structure of synthesized dyes
Anthraquinone-based azo dyes have azo groups conjugated with
anthraquinone, as shown in Fig. 1. 15A and 15B were synthesized by
using 1,5-diamino-anthraquinone as a diazo component and J-acid
as a coupling component. The position of coupling depended on the
pH of the coupling reaction mixture. The coupling of 15A was
induced to take place at the ortho position of the amino group
R ¼ A =At
(1)
(2)
k
ꢀ
ꢁ.
Tsp
¼
T þ Tt
k
2
(a-position) of J-acid by coupling it under acidic conditions. The
coupling reaction of 15B proceeded in alkaline conditions to induce
coupling at ortho position of the hydroxyl group (b-position). For
nꢀ
ꢁ.ꢀ
ꢁo
1=2
PE ¼
Tt ꢀ T
Tt þ T
(3)
k
k
where A and At are parallel and perpendicular absorbance to the
k
Table 3
drawing direction, respectively. T and Tt were calculated from the
Aspect ratio and optical properties of the polarizing films at the absorption
maximum of each dye.
k
absorbance (T ¼ 10ꢀA).
Dye
Aspect ratio
R
TSP (%)
PE (%)
2.5. Molecular orbital (MO) calculation
15A
15B
26A
26B
DA
2.46
2.71
2.40
3.16
2.70
14.12
14.29
15.21
15.14
24.68
36.14
35.43
38.13
36.34
39.52
98.80
99.09
98.33
99.04
99.65
CAChe 6.1.8 software package [11] was used in order to optimize
the geometry of dye structures by using molecular mechanics
MM3, conducting the iterative energy-minimizing routines with

J.B. Chang et al. / Dyes and Pigments 92 (2011) 737e744
741
Fig. 3. (a) Dichroic ratio spectra of anthraquinone-based azo dyes and dichroic ratio (black line) and absorption (red line) spectra of (b) DA, (c) 15A, (d) 15B, (e) 26A and (f) 26B. (For
interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.).
26A and 26B, the diazonium salt of the 2,6-diamino-anthraquinone
was coupled to the and -position of the J-acid, respectively.
DA was prepared as reference for comparison with
anthraquinone-based azo dyes. DA has a disazo structure which is
common for dichroic dyes and its azo linkages are arranged linearly
3.2. Spectral properties of the dyes and PVA polarizing films
a
b
a
The spectral properties of synthesized dyes in water and PVA
film are presented in Table 2. The dyes in water had maximum
absorption wavelengths of 526e538 nm with magenta shades.
Anthraquinone-based azo dyes exhibited similar molarꢀ1extinction
coefficients ranging from 21,000 to 23,000 L molꢀ1 cm , whereas
DA had higher tinctorial strength (3max ¼ 31,000 L molꢀ1 cmꢀ1).
Half-band widths of anthraquinone-based azo dyes in water were
more than 200 nm, whereas that of DA was only 95 nm.
by coupling at the
b-position of the J-acid (Scheme 2). Also,
a terminal acetophenone moiety is introduced in DA to make it
similar to the chromophore unit of anthraquinone.
The azo linkages of the synthesized dyes existed predominantly
in the hydrazone form due to intramolecular hydrogen bonding
with an adjacent amino or hydroxyl group. This was confirmed by
the down fielded proton peaks (12.5e16.0 ppm) of amino and
hydroxyl groups in the 1H NMR data of Table 1 [17].
The spectral properties of anthraquinone-based azo dyes in PVA
film showed a clear dependence on the structure of the dye. 15A
and 26A exhibited bathochromic shifts of more than 40 nm in the

742
J.B. Chang et al. / Dyes and Pigments 92 (2011) 737e744
Fig. 4. Schematic diagrams of the orientations of the transition moments of the anthraquinone and azo chromophores in 15B and 26B.
PVA film whereas 15B and 26B had relatively small spectral
changes. While the absorption maxima of dyes in the PVA film
depended on the position of the coupling of the J-acid, half-band
widths were influenced by the position of coupling at the anthra-
quinone. 15A and 15B had relatively wider half-band widths than
26A and 26B.
The absorption maxima of dyes appear to be determined
primarily by the charge transfer of the azo chromophore, which
consists of a donor (J-acid), a conjugated system (azo linkage) and
an acceptor (anthraquinone). This fact is supported by the
distinction between the absorption maxima of the dyes coupled at
3.3. Dichroic properties of anthraquinone-based azo dyes
in PVA polarizing films
The linearity of dye molecules could be used as an index for the
comparison of their dichroic properties when dyes have chromo-
phores of similar structures, as reported in our previous work [9].
However, the linearity of anthraquinone-based azo dyes, denoted
as aspect ratio, seems to bear no relation to the dichroic ratio at the
absorption maxima, as shown in Table 3. In the case of DA, the
comparison of its dichroic ratio to other dyes is meaningless
because of the differences in the structures of chromophores. When
dyes have different chromophores, their dichroic ratios are more
greatly influenced by other factors, such as the strength and
direction of transition moment, than by their aspect ratios.
However, this cannot explain the disagreement between the aspect
ratio and dichroic ratio in anthraquinone-based azo dyes. The
reason for this unexpected result could be accounted for by
studying the dichroic ratios of the PVA polarizing film throughout
the whole visible region.
Dichroic ratios of anthraquinone-based azo dyes in a PVA
polarizing film throughout the whole visible region are depicted in
Fig. 3(a). For all anthraquinone-based azo dyes, their maximum
dichroic ratios were obtained at longer wavelengths than their
absorption maxima. The maximum dichroic ratios of 15A, 15B, 26A
and 26B were 15.46,17.57,15.52 and 18.35, respectively. When these
values were compared to those of their aspect ratio, they were in
good agreements. The wavelengths for the maximum dichroic ratio
and their difference with the absorption maxima were 598, 586, 590
the a-position and at the b-position. An azo chromophore absorbs
a longer wavelength, as the electron-donating power of the adja-
cent substituent increases [18]. Therefore, 15A and 26A were more
bathochromic than 15B and 26B due to the stronger electron-
donating power of the amino group compared to the hydroxyl
group. Unlike the donor part, the difference of the acceptor part
seems to have relatively little influence on the absorption maxima
of the dyes. This is because anthraquinones act as acceptor parts
with similar levels of electron-withdrawing ability. Carbonyl
groups of anthraquinone in 1,5-disubstituted dyes are located on
the ortho or meta positions relative to the azo linkage, and carbonyl
groups in 2,6-disubstituted dyes are located on the meta or para
positions relative to the azo linkage. It is known that electron-
withdrawing groups exert a similar inductive effect on the color
of the azo chromophore when they are situated on the ortho or
para position relative to the azo linkage [18].
Anthraquinone in this study acted as a chromophore as well as
an acceptor for the azo chromophore. However, an anthraquinone
chromophore typically has a lower tinctorial strength than an azo
chromophore [18]. Also, the anthraquinone chromophores of the
synthesized dyes are supposed to absorb around the mid-400 nm
wavelength range while the absorption band of azo chromophore
lies around the mid-500 nm wavelength range. Therefore, the
anthraquinone chromophores did not change the absorption
maxima of the dyes, which were determined by the azo chromo-
phores, but they broadened the absorption spectra of the dyes by
additional absorption in the shorter wavelength range.
As shown in Fig. 2, 1,5- and 2,6-disubstituted anthraquinones
had different charge transfer transitions and 1,5-disubstituted
anthraquinones are known to be generally more bathochromic
and tinctorially stronger than 2,6-disubstituted anthraquinones
[18]. Therefore, additional absorption by 1,5-disubstituted anthra-
quinones could be higher and closer to the absorption maxima
compared to that achieved by 2,6-disubstituted anthraquinones. As
Fig. 5. Charge distribution at HOMO (blue and green electron clouds) and LUMO (red
and yellow electron clouds) level of 26B. (For interpretation of the references to color
in this figure legend, the reader is referred to the web version of this article.).
a
result, the dyes have distinctively wide half-band widths
depending on the coupling position of the anthraquinone.

J.B. Chang et al. / Dyes and Pigments 92 (2011) 737e744
743
ratio absorbs the shorter wavelength region and the azo chromo-
phore with a higher dichroic ratio absorbs in the longer wavelength
region. Within the overlapping absorption region of the anthra-
quinone and azo chromophore, the absorbance can increase but the
dichroic ratio will decrease since absorbances are obtained by the
sum of each contribution, but dichroic ratios are calculated by
averaging their values. Therefore, the dichroic ratio at the absorp-
tion maxima would be lower than the dichroic ratio of the azo
chromophore itself. For these reasons, 15-substituted dyes had the
discordance of absorption maxima and maximum dichroic ratios.
The dichroic ratio curve of the 2,6-substituted dyes were more
similar to their absorbance curve in the shortwave region owing to
the parallel transition moment of anthraquinone. However, there is
still some discrepancy in the wavelengths of absorption maxima
and maximum dichroic ratios. This result suggests that the dichroic
characteristic of the anthraquinone-based azo dyes is further
affected by factors other than the overlapping of the absorption of
anthraquinone and azo chromophore.
Iodine
15B
26B
DA
100
90
80
70
60
50
40
30
20
10
0
300
400
500
600
700
800
900
1000
wavelength (nm)
Fig. 6. Cross transmittances of the polarizing films.
Fig. 5 shows the charge density distribution of 26B as prepared
by molecular modeling. Blue and green clouds on the J-acid are the
electron rich parts in the ground state and the red and yellow
clouds on anthraquinone are the electron rich parts in the excited
state of the dye. When anthraquinone acts as an acceptor of the azo
chromophore, the resonance energy transfer pathway of the
excited electron will branch off into two directions. This pathway
would depend on the energy state of the absorbed light and thus
the direction of the transition moment should be subject to change
with the wavelengths of absorption. In anthraquinone-based azo
dyes, the transition moment beyond the absorption maxima is
believed to have a smaller deviation angle relative to the molecular
long axis than the angle at the absorption maxima. Accordingly, the
maximum dichroic ratio was observed in longer wavelengths than
the absorption maxima.
and 592 nm and 22, 38,10 and 58 nm, respectively, for 15A,15B, 26A
and 26B. This discrepancy between the absorption maxima and
maximum dichroic ratio is noticeable since the dichroic ratios of
general dichroic dyes, such as DA, are almost consistent with their
absorption spectra, as shown in Fig. 3(b).
To investigate the transition moment of each chromophore, the
maximum heights of the dichroic ratio spectrum and absorption
spectrum were set equal and they are depicted in Fig. 3(c)e(f). In
the wavelength region below the absorption maxima, 1,5-
substituted dyes had lower dichroic ratios compared to absor-
bance, while 2,6-sustituted dyes had similar values. This difference
is believed to be caused by the disagreement of the directions of the
transition moments of anthraquinones. As mentioned before in
Fig. 2, the direction of charge transfer in 1,5- and 2,6-sustituted
anthraquinone chromophores are different. The different charge
transfers lead to changes in the direction of the transition moment.
Consequently, the direction of the transition moment of the
anthraquinone chromophore would be perpendicular to the
molecular axis in 1,5-substituted dyes and parallel in 2,6-
substituted dyes as shown in Fig. 4. The deviated transition
moment of the anthraquinone chromophore in 1,5-substituted
dyes resulted in the increase of perpendicular absorbance and
thus lowered the dichroic ratio in the shortwave region with
respect to absorbance.
3.4. Advantage for neutral gray
The cross transmittance of a pair of polarizing films was
measured for films dyed with 15B, 26B, DA and iodine. As shown in
Fig. 6, the iodine type absorbed evenly throughout the visible range
but the dye types absorbed in limited regions within the visible
range. The absorption regions of the anthraquinone-based azo dyes
were broader than that of DA.
The broadness in the longer wavelength region is likely due to
the acceptor effect of anthraquinone. When an anthraquinone acts
as an acceptor of the azo chromophore, their two-way electron-
The absorptions of 1,5-substituted dyes are caused by two
chromophores: anthraquinone chromophore with a lower dichroic
Fig. 7. Color difference on L*a*b* color space CIE1976 and luminance of the polarizing films.

744
J.B. Chang et al. / Dyes and Pigments 92 (2011) 737e744
withdrawing effect that was explained previously would allow
various excited state -electrons distributions and this will lead to
increases in the bathochromic absorption of the dye.
Difference of transmittance in the shortwave range between 15B
and 26B is caused by the different tinctorial strength of anthra-
quinone chromophores. The wavelengths ranging from 400 nm to
450 nm leaked more on 26B than they did on 15B. This different
light leakage related to the shades of the dyes, as shown in Fig. 7. In
the L*a*b* color space, 15B and 26B had closer color coordinates to
neutral gray than DA.
References
p
[1] Lyoo WS, Yeum JH, Ghim HD, Park JM, Lee SJ, Kim JH, et al. Effect of the
molecular weight of poly(vinyl alcohol) on the water stability of a syndiotactic
poly(vinyl alcohol)/iodine complex film. Colloid & Polymer Science 2003;281:
416e22.
[2] Shin EJ, Lee YH, Choi SC. Study on the structure and processability of the
iodinated poly(vinyl alcohol). Journal of Applied Polymer Science 2004;91:
2407e15.
[3] Shin EJ, Lyoo WS, Lee YH. Making polyvinyl alcohol films iodinated as solution
state before casting and its application. Journal of Applied Polymer Science
2008;109:1143e9.
[4] Song DH, Yoo HY, Lee JJ, Kim JP. Polarizing films based on oriented poly(vinyl
alcohol)-dichroic dyes. Molecular Crystals and Liquid Crystals 2006;445:
355e60.
4. Conclusions
[5] Beekman E, Kocher C, Kokil A, Zimmermann S, Weder C. UV polarizers based
on oriented poly(vinyl alcohol)eChrysophenineeCongo red blend films.
Journal of Applied Polymer Science 2002;86:1235e9.
The introduction of the conjugated bichromophore system in
a dichroic dye was found to be very effective for increasing the
absorption bandwidth. The broad absorption spectra of anthraqui-
none-based azo dyes are influenced not only by merging different
absorptions of anthraquinone and azo chromophores but also by
the acceptor effect of anthraquinone. The anthraquinone acted as an
acceptor of the azo chromophore and led to broader and bath-
ochromic absorption of azo chromophore due to its various tran-
sient energy states. Accordingly, anthraquinone-based azo dyes had
a favorable color shade for producing neutral gray polarizing films.
Anthraquinone-based azo dyes had lower dichroic ratios in
comparisonwith general polyazo dichroic dyes such as DA due to the
misalignment of transition moments depending on absorption band.
The difference between the transition moments of anthraquinone
and azo chromophores reduces dichroic ratios in overlapping
absorption bands. The transition moment of azo chromophore
is attenuated by acceptor effect of anthraquinone because two
electron-withdrawing groups of anthraquinone share the charge
density of the excited state. These facts could be confirmed by
analysis of the discrepancy in the wavelengths between the
absorption maxima and maximum dichroic ratio. The findings of this
work are expected to provide a useful information about the corre-
lation between dichroic ratio and transition moment.
[6] Lyoo WS, Yeum JH, Park JM, Kwak JW, Kim JH, Kim SS, et al. Role of molecular
weight of atactic poly(vinyl alcohol) (PVA) in the polarizing efficiency of PVA/
Azo dye complex film with high durability. Journal of Applied Polymer Science
2005;96:967e74.
[7] Song DH, Kim JP. Effect of transition moments and orientational behavior of
dichroic dyes on the optical anisotropy of poly(vinyl alcohol) polarizing films.
Dyes and Pigments 2009;80:219e25.
[8] Oh SY, Oh JH, Shin DY. Preparation and polarization characteristics of polar-
izing film using a dichroic azo dye with UV absorbing moiety. Journal of the
Korean Institute of Chemical Engineers 2001;39:411e6.
[9] Chang JB, Hwang JH, Park JS, Kim JP. The effect of dye structure on the dyeing
and optical properties of dichroic dyes for PVA polarizing film. Dyes and
Pigments 2011;88:366e71.
[10] Dirix Y, Tervoort TA, Bastiaansen C. Optical properties of oriented polymer/
dye polarizers. Macromolecules 1995;28:486e91.
[11] CAChe 6.1.8. Beaverton, Oregon: Fujitsu Systems Business of America Inc.
[12] Lye J, Freeman HS, Hinks D. Molecular modeling in dye chemistry: studies
involving two disperse dyes. Textile Research Journal 1999;69:583e90.
[13] Goto H, Osawa E. Viewpoint 11 e approaches to the global minimum problem.
Journal of Molecular Structure: THEOCHEM 1993;285:157e68.
[14] Ridley JE, Zerner MC. An intermediate neglect of differential overlap tech-
nique for spectroscopy: pyrrole and the azines. Theoretica Chimica Acta 1973;
32:111e34.
[15] Kubo Y, Yoshida K, Adachi M, Nakamura S, Maeda S. Experimental and
theoretical study of near-infrared absorbing naphthoquinone methide dyes
with a nonplanar geometry. Journal of the American Chemical Society 1991;
113:2868e73.
[16] Hanemann T, Böhm MC, Haase W. Calculation of UV/VIS absorption spectra of
liquid crystals and dye molecules: an INDO MO approach. Liquid Crystals
1992;11:917e27.
[17] Skrabal P, Zollinger H. Mechanism of azo coupling reactions: part XXXV. pH-
dependence and ortho/para ratio in coupling reactions of amino-
hydroxynaphthalenesulfonic acids. Dyes and Pigments 1988;9:201e7.
[18] Gordon PF, Gregory P. Organic chemistry in colour. Springer-Verlag Berlin
Heidelberg; 1987. p. 126e95.
Acknowledgment
This work was supported by a research grant from Ministry of
Knowledge Economy (MKE, No. 10030016).