Modular approach to obtaining organic glasses from low-molecular weight dyes using 1,1,1-triphenylpentane auxiliary groups: Nonlinear optical properties

Article abstract of DOI:10.1016/j.dyepig.2013.08.020

A new modular synthetic approach to obtain organic glasses from low-molecular weight azobenzenes and stilbenes has been accomplished using 1,1,1-triphenylpentane crystallization preventing auxiliary groups. Six new structures show excellent solubility in non-polar solvents and thin films with good optical qualities have been obtained using a spin-coating technique. The glass transition temperatures of the new amorphous materials were in the range of 73-108 C. The nonlinear optical activity in thin amorphous films was measured after a corona poling procedure.

Full text of DOI:10.1016/j.dyepig.2013.08.020

Dyes and Pigments 99 (2013) 1044e1050
Contents lists available at ScienceDirect
Dyes and Pigments
journal homepage: www.elsevier.com/locate/dyepig
Modular approach to obtaining organic glasses from low-molecular
weight dyes using 1,1,1-triphenylpentane auxiliary groups: Nonlinear
optical properties
,
a
b
a
Kaspars Traskovskis a , Kristine Lazdovica , Andrejs Tokmakovs , Valdis Kokars ,
*
Martins Rutkis b
a Riga Technical University, Faculty of Materials Science and Applied Chemistry, 14/24 Azenes Street, Riga LV-1048, Latvia
b Institute of Solid State Physics, University of Latvia, 8 Kengaraga Street, Riga LV-1063, Latvia
a r t i c l e i n f o
a b s t r a c t
Article history:
A new modular synthetic approach to obtain organic glasses from low-molecular weight azobenzenes
and stilbenes has been accomplished using 1,1,1-triphenylpentane crystallization preventing auxiliary
groups. Six new structures show excellent solubility in non-polar solvents and thin films with good
optical qualities have been obtained using a spin-coating technique. The glass transition temperatures of
the new amorphous materials were in the range of 73e108 ꢀC. The nonlinear optical activity in thin
amorphous films was measured after a corona poling procedure.
Received 15 May 2013
Received in revised form
16 August 2013
Accepted 19 August 2013
Available online 30 August 2013
Ó 2013 Elsevier Ltd. All rights reserved.
Keywords:
Azobenzene
Indane-1,3-dione
Molecular glass
Non-linear optics
Glass transition temperature
Second harmonic generation
1. Introduction
task, the molecular compounds are much easier to obtain, isolate
and identify. The general structural requirements regarding the
The last decade has witnessed a notable increase in organic
nonlinear optical (NLO) materials performance characteristics,
making these compounds viable for practical applications in
various photonic devices [1,2]. The general structural requirement
for NLO activity in materials is the presence of pushepull type
synthesis of the molecular glasses are still not strictly defined,
although, some molecular properties as the presence of bulky, non-
planar, heavy and rigid fragments are often taken into the consid-
eration to acquire such materials [10,11].
In the case of NLO materials where the active part of the material
is a pushepull type chromophore, the formation of a stable amor-
phous phase is an even harder task. The chromophores used in such
materials are almost always linear, planar and polar molecules
which contradict with the general structural requirements of a good
glass forming compound. To overcome this, the molecular designs
exploiting dendrimeric approach are widely used where the crys-
tallization of compounds is prevented by an increased weight and a
complicated geometry of the molecule [12e14]. The success of this
method is not always predictable and, in addition, with the increase
of molecular weight many dendrimeric compounds, like in the case
of polymers, become hard to purify and characterize.
One of the promising directions regarding obtaining molecular
glasses from small chromophores is the introduction of phase
behavior modifying groups. In particular, groups like mexylamino-
triazine [15] and triphenylamine derivatives [16] had been applied
for such a use. Recently we have presented an approach where the
chromophores with high molecular hyperpolarizability (b) values
[3]. For the practical applications NLO materials need to meet
several criteria such as high optical transparency, good solubility
and high chemical and thermal stability. To achieve this, the
chromophores are often incorporated into guestehost [4,5] and
polymeric [6,7] systems.
The relatively less studied material type is molecular glasses, a
compound class where small-molecular compounds are capable of
forming a stable amorphous phase [8,9]. This material design
approach, however, has several important advantages. Compared
to polymeric compounds, where batch to batch synthetic repeti-
tion, purification and product characterization are often a difficult
* Corresponding author. Tel.: þ371 29148070.
E-mail address: kaspars.traskovskis@rtu.lv (K. Traskovskis).
0143-7208/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.dyepig.2013.08.020

K. Traskovskis et al. / Dyes and Pigments 99 (2013) 1044e1050
1045
Scheme 1. Synthetic routes for organic glasses OGL-(1e5). Example compound OGL-0 showing previously presented structural approach is also given.
crystallization of small molecular chromophores is prevented by
using triphenylmethyl or triphenylsilyl auxiliary groups [17]. From a
structural point of view these triphenyl substituents present several
favorable standalone properties like being chemically inert, non-
polar, non-absorbent in the visible light range and solubility
enhancing. Unfortunately, in the mentioned paper the triphenyl
moieties were bound to the chromophore core by a labile CeO bond
making materials susceptible to an acidic hydrolysis (see the
structure OGL-0 given as an example in Scheme 1). While this
disadvantage can be overcome by simply avoiding the presence of
acids during the processing of materials, this notably limits the
possible synthetic routes during the production of compounds. In
the present work we describe an improved chemical design for the
modular approach introducing 1,1,1-triphenylpentane amorphous
phase formation enhancing auxiliary group, where the aforemen-
tioned disadvantages are overcome. Using the new modifying
moiety, the series of molecular glasses from different low-molecular
chromophores frequently used in different NLO studies have been
synthesized to show the versatility of the molecular design. The
thermoplastic, linear and non-linear optical properties of the ma-
terials have been closely studied.
following compounds were prepared according to known methods:
5,5,5-Triphenylpentan-1-ol (1) [18], 4-trityldiazobenzene-40-dia-
zonium tetrafluoroborate [19], 4-nitrobenzenediazonium tetra-
fluoroborate, 4-tritylbenzenediazonium tetrafluoroborate [17], 1,3-
indandione [20], 2-Dicyanomethylene-3-cyano-4,5,5-trimethyl-
2,5-dihydrofuran (TCF) [21], diazomethane [22], 4-(tert-butyl)
phthalic acid [23]. The solvents used were reagent grade, DMF was
distilled from P2O5 and THF from CaH2 before use. 1H NMR spectra
were obtained using Bruker Avance 300 MHz spectrometer using
TMS as an inner standard. Low-resolution mass spectra were ac-
quired on Waters EMD 1000 MS detector (ESIþ mode, cone voltage
30 V). The elemental analysis was carried out with Costech In-
struments ECS 4010 CHNSeO Elemental Combustion System. The
glass transition temperatures of the compounds were determined
using Mettler Toledo DSC-1/200 W, by DSC thermograms acquired
at heating rate 10 ꢀC/min, initially the heating samples to decom-
position onset temperatures and then cooling at the rate 50 ꢀC/min.
Decomposition temperatures were obtained using a Perkin Elmer
STA 6000 thermal analyzer.
2.2. Synthesis
2. Experimental
2.2.1. 1,1,1-Triphenylpentan-5-iodide (2)
1,1,1-Triphenylpentan-5-ol (14 g, 44.4 mmol), triphenylphos-
phine (17.41 g, 66.4 mmol) and imidazole (6.02 g, 88.6 mmol) were
dissolved in anhydrous THF (100 mL). The solution was cooled to
0 ꢀC and iodine (16.87 g; 66.4 mmol) was added in small portions.
The resulting mixture was stirred at room temperature for 24 h. The
2.1. Starting materials and general procedures
If not further specified, the starting reagents and solvents were
obtained commercially from Alfa Aesar and used as received. The

1046
K. Traskovskis et al. / Dyes and Pigments 99 (2013) 1044e1050
formed precipitate was filtered off and washed with THF
(3 ꢁ 10 mL). The solution was evaporated under reduced pressure
to 30 mL volume and ethanol (70 mL) was added. After the repeated
evaporation to 30 mL volume light-yellow precipitate formed and
was filtered off. The crude product was crystallized from ethanol
2.2.5. 4-((4-Nitrophenyl)diazenyl)-N,N-bis(5,5,5-triphenylpentyl)
aniline (OGL-2)
Compound was prepared according to General procedure A
described above to yield OGL-2 as red powder (0.64 g, 82%). IR (KBr,
y
cmꢂ1): 3054, 2939, 2870, 1515, 1419, 1331, 1131. 1H NMR (
d, CDCl3,
and 2 was obtained as white flakes (13.8 g, 73%). IR (KBr,
y
cmꢂ1):
d, CDCl3, 300 MHz):
300 MHz): 1.03 (4H, m), 1.49 (4H, m), 2.50 (4H, m), 3.06 (4H, t,
3J ¼ 7.3 Hz), 6.48 (2H, d, 3J ¼ 8.5 Hz), 7.05e7.23 (30H, m), 7.79 (2H, d,
3J ¼ 8.7 Hz), 7.85 (2H, d, 3J ¼ 9.0 Hz), 8.23 (2H, d, 3J ¼ 9.0 Hz). MS
(ESIþ) m/z: 840.7 (Mþ, requires 840.1). Elemental analysis. Calcd.
for C58H54N4O2: C, 83.02, H, 6.49, N, 6.68. Found: C, 82.39, H, 6.37, N,
6.33%.
3053, 2951, 1489, 1443, 1262, 1183. 1H NMR (
1.12 (2H, m), 1.76 (2H, quin, 3J ¼ 7.3 Hz), 2.49 (2H, m), 2.99 (2H, t,
3J ¼ 7.3 Hz), 7.06e7.21 (15H, m). Elemental analysis. Calcd. for
C
23H23I: C, 64.80, H, 5.44. Found: C, 65.12, H, 5.50.
2.2.2. N,N-bis(5,5,5-Triphenylpentyl)aniline (3)
Aniline (1.1 g, 12.7 mmol) and compound 2 (12.0 g, 28.2 mmol)
were dissolved in acetone (20 mL). Powdered K2CO3 (4.0 g,
29 mmol) was added and the mixture was heated under reflux for
48 h. The precipitate was filtered, washed with acetone and the
resulting solution was evaporated under reduced pressure. The
obtained yellow oil was purified by column chromatography over
silica gel using toluene/petroleum ether (2/1) as eluant to give 3 as
clear viscous oil (8.50 g, 97%). The excess of reagent 2 was quanti-
2.2.6. N,N-bis(5,5,5-Triphenylpentyl)-4-((4-tritylphenyl)diazenyl)
aniline (OGL-3)
Compound was prepared according to General procedure A
described above to yield OGL-3 as yellow powder (0.79 g, 89%). IR
(KBr,
NMR (
y
cmꢂ1): 3054, 2940, 2870, 1494, 1444, 1394, 1363, 1140. 1H
d, CDCl3, 300 MHz): 1.00 (4H, m), 1.47 (4H, m), 2.49 (4H, m),
3.03 (4H, t, 3J ¼ 7.2 Hz), 6.46 (2H, d, 3J ¼ 8.2 Hz), 7.04e7.23 (45H, m),
7.27 (2H, d, 3J ¼ 8.7 Hz), 7.64 (2H, d, 3J ¼ 8.7 Hz), 7.72 (2H, d,
3J ¼ 7.7 Hz). MS (ESIþ) m/z: 1037.6 (Mþ, requires 1037.4). Elemental
analysis. Calcd. for C77H69N3: C, 89.23, H, 6.71, N, 4.05. Found: C,
89.19, H, 6.69, N, 4.00%.
tatively recovered. IR (KBr,
y
cmꢂ1): 3056, 2942, 2870, 1504, 1492,
1446,1367,1264.1H NMR (
d
, CDCl3, 300 MHz): 0.97 (4H, m),1.42 (4H,
quin, 3J ¼ 7.3 Hz), 2.47 (4H, m), 2.94 (4H, t, 3J ¼ 7.7 Hz), 6.44 (2H, d,
3J ¼ 8.4 Hz), 6.53 (1H, t, 3J ¼ 7.2 Hz), 7.03e7.24 (32H, m). Calcd. for
C
52H51N: C, 90.52, H, 7.45, N, 2.03. Found: C, 90.81, H, 7.20, N, 2.14%.
2.2.7. 2-(4-(bis(5,5,5-Triphenylpentyl)amino)benzylidene)-1H-
indene-1,3(2H)-dione (OGL-4)
2.2.3. 4-(bis(5,5,5-Triphenylpentyl)amino)benzaldehyde (4)
Aldehyde 4 (0.80 g, 1.11 mmol) and indane-1,3-dione (0.20 g,
1.36 mmol) were dissolved in n-butanol (10 mL) and a drop of
piperidine was added. After heating under reflux for 4 h, the product
was precipitated by addition of ethanol (50 mL), filtered, washed
with methanol and dried. After the purification by column chro-
matography over silica gel using DCM as eluant and following pre-
cipitation from DCM with ethanol, compound OGL-4 was obtained
DMF (1.41 g, 18.8 mmol) was cooled in an ice bath and phos-
phoryl chloride (2.89 g, 18.8 mmol) was added dropwise. After 1 h
the solution of compound 3 (10 g, 14.5 mmol) in DMF (5 mL) was
added and the resulting mixture was stirred at 70 ꢀC temperature
for 4 h. The water (2 mL) solution of NaOH (3.01 g, 75.3 mmol) was
then added and stirred for 30 min. The mixture was poured into
water (150 mL), and the product was extracted with toluene
(3 ꢁ 50 mL). The combined extract was dried over anhydrous
Na2SO4 and evaporated under reduced pressure. The crude product
was purified by column chromatography over silica gel using
toluene, then toluene/ethyl acetate (2/1) eluents to yield 4 as white
as orange powder (0.60 g, 64%). IR (KBr,
y
cmꢂ1): 3057, 2941, 2869,
, CDCl3, 300 MHz):
1713,1667,1552,1512,1445,1351,1182.1H NMR (
d
1.02 (4H, m), 1.49 (4H, m), 2.49 (4H, m), 3.07 (4H, t, 3J ¼ 7.7 Hz), 6.48
(2H, d, 3J ¼ 9.0 Hz), 7.07e7.22 (30H, m), 7.84 (2H, m), 7.88 (1H, s), 8.05
(2H, m), 8.60 (2H, d, 3J ¼ 8.3 Hz). MS (ESIþ) m/z: 847.6 (Mþ, requires
847.1). Elemental analysis. Calcd. for C62H55NO2: C, 88.01, H, 6.55, N,
1.66. Found: C, 87.75, H, 6.47, N, 1.68%.
amorphous solid (8.10 g; 78%). IR (KBr,
y
cmꢂ1): 3083, 2941, 2870,
, CDCl3, 300 MHz): 0.99 (4H,
1670, 1542, 1492, 1443, 1164. 1H NMR (
d
m), 1.43 (4H, quin, 3J ¼ 7.3 Hz), 2.47 (4H, m), 2.99 (4H, t, 3J ¼ 7.7 Hz),
6.40 (2H, d, 3J ¼ 9.0 Hz), 7.03e7.21 (30H, m), 7.56 (2H, d, 3J ¼ 8.8 Hz),
9.59 (1H, s). Calcd. for C53H51NO: C, 88.66, H, 7.16, N, 1.95. Found: C,
88.30, H, 7.05, N, 2.06%.
2.2.8. (E)-2-(4-(4-(bis(5,5,5-Triphenylpentyl)amino)styryl)-3-
cyano-5,5-dimethylfuran-2(5H)-ylidene)malononitrile (OGL-5)
Compound 4 (1.38 g, 2.0 mmol) and TCF (0.40 g, 2.0 mmol) were
dissolved in DMF (10 mL) and two drops of piperidine were added
to the mixture. After stirring for 12 h at 80 ꢀC temperature the
solvent was evaporated under reduced pressure till 2 mL volume
and ethanol (50 mL) was added. The blue precipitate was filtered,
dried and purified by column chromatography over silica gel using
DCM/hexane (6/1) as eluant. After the evaporation of the solvent
from the clean fractions, the obtained blue glass was dissolved in
DCM and ethanol was added. DCM was fractionally removed under
reduced pressure and the blue precipitate was filtered to yield OGL-
2.2.4. N,N-bis(5,5,5-Triphenylpentyl)-4-(-(4-(-(4-tritylphenyl)
diazenyl)phenyl)diazenyl) aniline (OGL-1)
General procedure A. Aniline 3 (0.65 g, 0.94 mmol) was dissolved
in acetic acid (5 mL) and 4-trityldiazobenzene-40-diazonium tet-
rafluoroborate (0.76 g, 1.15 mmol) was added to the solution at
room temperature, followed by addition of sodium acetate (0.23 g,
2.3 mmol). After stirring for 12 h ethanol (50 mL) was added to the
solution and a red precipitate formed that was collected by filtra-
tion and air dried. The crude product was purified by column
chromatography over silica gel using DCM/petroleum ether (2/1) as
eluant. After the evaporation of solvent from the clean fractions, the
obtained red glass was dissolved in DCM and ethanol was added.
DCM was fractionally removed under reduced pressure and the red
precipitate was filtered to yield OGL-1 as red powder (0.80 g, 74%).
5 as blue powder (0.40 g, 22%). IR (KBr,
y
cmꢂ1): 3053, 2938, 2871,
, CDCl3,
2220, 1559, 1517, 1448, 1367, 1273, 1161. 1H NMR (
d
300 MHz): 1.02 (4H, m), 1.44 (4H, m), 1.67 (6H, s), 2.49 (4H, m), 3.05
(4H, t, 3J ¼ 7.6 Hz), 6.41 (2H, d, 3J ¼ 9.0 Hz), 6.64 (1H, d, 3J ¼ 15.8 Hz),
7.07e7.21 (30H, m), 7.38 (2H, d, 3J ¼ 8.7 Hz), 7.51 (1H, d,
3J ¼ 15.8 Hz). MS (ESIþ) m/z: 900.8 (Mþ, requires 900.2). Elemental
analysis. Calcd. for C64H58N4O: C, 85.49, H, 6.50, N, 6.23. Found: C,
85.48, H, 6.49, N, 6.07%.
IR (KBr,
NMR (
y
cmꢂ1): 3053, 2914, 2871, 1494, 1445, 1389, 1361, 1142. 1H
d
, CDCl3, 300 MHz): 1.02 (4H, m), 1.48 (4H, m), 2.49 (4H, m),
3.05 (4H, t, 3J ¼ 7.5 Hz), 6.48 (2H, m), 7.06e7.23 (45H, m), 7.33 (2H,
d, 3J ¼ 8.7 Hz), 7.72e7.84 (4H, m), 7.87e7.94 (4H, m). Elemental
analysis. Calcd. for C83H73N5: C, 87.41, H, 6.45, N, 6.14. Found: C,
87.15, H, 6.41, N, 6.52%.
2.2.9. Dimethyl 4-(tert-butyl)phthalate (6)
In a well-ventilated fume hood to the solution of 4-(tert-butyl)
phthalic acid (10 g; 45.0 mmol) in methanol/water (10/1, 100 mL)

K. Traskovskis et al. / Dyes and Pigments 99 (2013) 1044e1050
1047
freshly prepared diazomethane MTBE solution was added drop-
3. Results and discussion
wise. When the mixture had turned light yellow and the release of
gas bubbles was no more observed, the addition was stopped and
the solvent was removed under reduced pressure. The resulting oil
was dissolved in MTBE and washed with 5% NaHCO3 solution. The
organic layer was then dried over anhydrous Na2SO4 and evapo-
rated under reduced pressure. The crude product was purified by
column chromatography over silica gel using DCM as eluant to give
The main task in this study was to improve the chemical sta-
bility of our previously presented compounds that contain triphe-
nylmethyl (trityl) auxiliaries as amorphous phase formation
ensuring groups (see example structure OGL-0) [17]. The logical
synthetic step in this direction was to replace the labile CeO bond
with a more stable CeC bond. Following our structural approach
presented previously this new molecular fragment should be con-
nected to an aniline nitrogen atom to form the basis of an electron
donating part of a chromophore. To achieve this the alkylating
agent 1,1,1-triphenylpentane-5-iodide (2) was synthesized from the
known intermediate 1,1,1-triphenylpentane-5-ol (1) [18] and was
reacted with aniline to obtain the key building block for molecular
glasses N,N-bis(5,5,5-triphenylpentyl)aniline (3) (Scheme 1). The
choice of a four carbon atom long connecting bridge between a
trityl group and nitrogen was not coincidental. While not described
here in the detail, we performed initial synthetic routes involving
propane and butane analogs of compound 2. In the first case the
alkylation did not proceed due to steric hindrance of electrophilic
center by trityl group. The second synthetic approach resulted in
the products with relatively poor solubility and glass formation
ability.
The aniline 3 was azo-coupled with the series of diazonium salts
to yield final products OGL-1,2,3. Compounds OGL-1,3 contain an
additional triphenyl group in the acceptor part of chromophore
what could result in a better phase behavior and thermoplastic
properties. In addition, sterically large groups in close proximity to
the chromophores have been shown to increase the macroscopic
NLO performance of materials [26]. This so called site isolation
principle is based on the prevention of dipoleedipole interactions
of polar chromophore moieties. The NLO activity of chromophore
systems is ensured by ordering these molecules non-
centrosymmetrically by an external electric field poling. If mole-
cules are not sufficiently isolated dipoles tend to spontaneously
align centrosymmetrically leading to a decreased poling efficiency
and stability of a poled system.
6 as clear viscous oil (7.0 g, 62%). IR (KBr,
y
cmꢂ1): 2955, 2907, 2871,
d, CDCl3, 300 MHz): 1.26 (9H,
1725, 1434, 1292, 1281, 1247. 1H NMR (
s), 3.81 (3H, s), 3.84 (3H, s), 7.46 (1H, dd, 3J ¼ 8.1 Hz, 4J ¼ 2.1 Hz), 7.60
(1H, d, 4J ¼ 2.1 Hz), 7.63 (1H, d, 3J ¼ 8.3 Hz). Elemental analysis.
Calcd. for C14H18O4: C, 67.18, H, 7.25. Found: C, 67.25, H, 7.29.
2.2.10. 2-(4-(bis(5,5,5-Triphenylpentyl)amino)benzylidene)-5-(tert-
butyl)-1H-indene-1,3(2H)-dione (OGL-6)
To the compound 6 (7.0 g, 28.0 mmol) small lumps of sodium
(1.21 g, 52 mmol), anhydrous ethanol (0.14 mL) and ethyl acetate
(6.77 mL) were added and the mixture was heated under reflux for
8 h. Ethanol (20 mL) was then added and the mixture was left to stir
for 1 h. The solvent was removed under reduced pressure and the
resulting brown glass was added in small portions to the boiling
water (125 mL). The water solution was then cooled to 70 ꢀC and
after 1 h sulfuric acid/water mixture (1/3, 25 mL) was added
dropwise. The mixture was cooled and extracted with MTBE. The
organic layer was then dried over anhydrous Na2SO4 and evapo-
rated under reduced pressure. The crude product was then dis-
solved in n-butanol (10 mL) along with aldehyde 4 (1 g, 1.40 mmol)
and a drop of piperidine was added. After heating under reflux for
4 h, the product was precipitated by addition of ethanol (50 mL),
filtered, washed with methanol and dried. After the purification by
column chromatography over silica gel using DCM as eluant and
following precipitation from DCM with ethanol, compound OGL-6
was obtained as orange powder (0.48 g). IR (KBr,
y
cmꢂ1): 3055,
2948, 1713, 1670, 1554, 1511, 1445, 1365, 1323, 1173. 1H NMR (
d
,
CDCl3, 300 MHz): 1.02 (4H, m), 1.34 (9H, 2 signals representing E/Z
isomers with integral intensity 1:1), 1.46 (4H, m), 2.50 (4H, m), 3.07
(4H, t, 3J ¼ 7.5 Hz), 6.46 (2H, d, 3J ¼ 9.2 Hz), 7.08e7.23 (30H, m), 7.66
(1H, d), 7.67e7.71 (1H, m), 7.78 (1H, m), 7.87 (1H, m), 8.39 (2H, d,
3J ¼ 8.5 Hz). MS (ESIþ) m/z: 903.6 (Mþ, requires 903.2). Elemental
analysis. Calcd. for C66H63NO2: C, 87.86, H, 7.04, N, 1.55. Found: C,
87.25, H, 7.14, N, 1.42%.
Vilsmeier formylation of compound 3 yielded 4-(bis(5,5,5-
triphenylpentyl)amino)benzaldehyde (4), a building block for sty-
rene dyes. The Knoevenagel condensations of
4 with 1,3-
indandione and TCF yielded products OGL-4,5. To reduce the
intermolecular dipole interactions and improve the NLO perfor-
mance of the material a sterically bulky tert-butyl group in the 5
position of indane-1,3-dione was introduced according to Scheme
2. The esterification of 4-(tert-butyl)phthalic acid (5) by conve-
nient methods was unsuccessful and was accomplished by the use
of diazomethane. After the subsequent cyclization the acquired 5-
tert-butylindane-1,3-dione, without isolation, was condensed with
4 to yield OGL-6. The 1H NMR spectra revealed that a 1:1 mixture of
E/Z isomers was obtained depending on the tert-butyl group
placement in the indanedione ring. It was not separable with both
the flash chromatography and HPLC and the further measurements
were taken for the mixture of isomers.
2.3. Sample film preparation
The typical procedure involved the preparation of a compound
solution with concentration 100 mg/mL. The solution was then
dropped onto indium tin oxide (ITO) covered glass substrates and
spin-coated with a Laurell WS-400B-6NPP/LITE (starting speed e
0 rpm, terminal speed e 300 rpm, acceleration e 200 rpm/s2,
spinning time e 1 min). The resulting films were 1e2
mm thick. The
samples for absorption spectra were 0.05e0.1
mm thick and were
made from 5 times more diluted solutions.
The acquired compounds OGL-(1e6) all show a good glass for-
mation ability and an excellent solubility in non-polar organic
solvents, so the films with a high optical quality can be easily ac-
quired by spin-coating from concentrated solutions using volatile
solvents like chloroform without the use of added plasticizers. The
thermoplastic properties of the compounds were examined by DSC.
The glass transition temperatures (Fig. 1) were acquired on the
second heating cycle after the samples were initially heated to the
temperatures where a decomposition onset was observed (Table 1).
In the all cases no other peaks than Tg were observed during the
repeated heating. Moreover, only samples OGL-2 and OGL-3
revealed the presence of a crystalline phase during the initial
2.4. Linear and non-linear optical measurements
The NLO properties of films were characterized by means of the
Maker fringe technique with an excitation at 1064 nm. The thick-
ness, refractive indexes and absorption coefficients of amorphous
films were determined in advance. The NLO coefficients d33, d31 and
d15 were calculated by Herman and Hayden [24] approach using an
x-cut quartz crystal as a reference (d11 ¼ 0.3 pm/V) and taking into
the account the absorption of film at 532 nm. The more detailed
descriptions of these procedures could be found in our previous
studies [17,25].

1048
K. Traskovskis et al. / Dyes and Pigments 99 (2013) 1044e1050
Scheme 2. Synthetic route for 5-tert-butylindane-1,3-dione fragment containing compound OGL-6.
heating showing melting peaks at 185 and 171 ꢀC representatively.
The acquired Tg values are close to those we measured for the tri-
tiloxy moieties containing compounds in our previous study [17]. A
linear positive correlation between molecular weight and glass
transition temperature can be observed for azobenzene dye con-
taining compounds OGL-(1e3), the same can be said for indane-
1,3-dione fragments containing OGL-4,6 as with the introduction of
tert-butyl substituent the glass transition temperature increases by
15 ꢀC.
bands. These observations of the absorption spectra suggest that
with the increase of chromophore polarity the trityl auxiliary
groups bound at the donor part of the chromophore cannot ensure
sufficient chromophore shielding from dipoleedipole interactions
and further molecular modifications at the acceptor part of the
molecules are necessary.
The measured nonlinear coefficients for compounds are given in
Table 1. To evaluate the relative performance of materials several
Despite the relatively low molecular weight, the highest Tg value
of 108 ꢀC was measured for compound OGL-5. Taking into the
consideration the highest calculated dipole momentum value for
this compound (see Table 1), the increased heat resistance of the
glass can be explained by a strong intermolecular chromophore
pe
p
stacking in the solid phase, a property that is known for the given
dye system [27]. The obtained solid phase UVeVis absorption
spectra reassure this assumption (Fig. 2) as OGL-5 shows a
distinctive pair of absorption bands. The band with a peak at
577 nm is comparable to the absorption of the compound in CH2Cl2
(the maximum value at 575 nm) and likely represents non-
aggregated molecules while bathochromically shifted band at
619 nm can be attributed to the aggregates. For the other com-
pounds no notable multimodality is observed in the absorption
Fig. 2. Absorption spectra and maxima in thin films for azobenzene (a) and stilbene (b)
Fig. 1. Acquired DSC thermograms showing Tg peaks.
chromophores containing compounds.
Table 1
Calculated hyperpolarizabilities and dipole moments; measured thermoplastic, linear and nonlinear optical properties in solid films.
Compound
b(0), 1030 esua
m
, Da
Tdecomp.
,
ꢀC
TSHI50
,
ꢀC
RI532
RI1064
d31(532), pm ꢁ Vꢂ1
d33(532), pm ꢁ Vꢂ1
d33(0), pm ꢁ Vꢂ1
OGL-1
OGL-2
OGL-3
OGL-4
OGL-5
OGL-6
32.7b
40.3b
20.1b
31.0
2.81
9.85
2.59
4.00
331
320
355
298
279
290
92
66
73
74.5
86
2.55
2.15
1.71
1.95
1.04
1.99
1.63
1.61
1.59
1.66
1.72
1.64
11.5
28.7
2.1
32.6
7.5
24.2
117.7
10.8
84.8
24.9
57.4
1.9
11.0
3.0
7.7
5.7
60.1
18.3
27.7c
3.15c
75
25.8
6.2
a
Quantum chemical calculations (RHF with basis set 6e31G(p,d) in the vacuum by Gaussian 09W).
Trans configuration for all azo dyes was energetically preferable, so given values correspond to this conformer.
Both isomers (E/Z) have almost the same molecular hyperpolarizability.
b
c

K. Traskovskis et al. / Dyes and Pigments 99 (2013) 1044e1050
1049
stable organic glasses from small-molecular chromophores. The
obtained compounds show good solubility in non-polar solvents and
good optical quality films can be prepared using a simple spin-
coating procedure. Glass transition temperatures in the range of
73e108 ꢀC were measured depending on the chromophore. Gener-
ally, the heat resistance of the compound increases with the mo-
lecular weight, although the high chromophore dipole momentum
and the resulting partial pep stacking also contribute to increased Tg
values. After the poling procedure the materials showed nonlinear
optical activity. For the chromophores with relatively low hyper-
polarizability the measured NLO performance is in the expected
range, but for more polar chromophores it falls due to the reduced
polar order of chromophores as it was illustratively observed for
compound OGL-5. Additional site isolating groups are suggested to
improve materials performance in these cases.
Acknowledgments
Fig. 3. The temperature induced decay of second harmonic intensity in poled films.
This work has been supported by the European Social Fund
within the project “Support for the implementation of doctoral
studies at Riga Technical University”, and National Research Pro-
gram “Development of Innovative Multifunctional Materials, Signal
Processing and Information Technologies for Competitive Science
Intensive Products”.
factors need to be taken into the consideration. The general
macroscopic NLO activity for a compound is proportional to an
active chromophore hyperpolarizability and accentric chromo-
phore order. Additionally, the measured second harmonic (SH)
generation efficiency is frequency dependent and approaching the
lowest chromophore chargeetransfer transition (CT) the resonance
enhancement takes place and inflates the acquired d33 values. The
extrapolation to the zero frequency d33(0) according to the two level
model [28] is often used to benchmark different chromophores,
however, if SH and CT frequencies are close the precision of this
model falls drastically [29]. Overall, the measured d33 and d33(0)
References
[1] Dalton LR, Sullivan PA, Bale DH. Electric field poled organic electro-optic
materials: state of the art and future prospects. Chem Rev 2010;110:25e55.
[2] Li Z, Li Q, Qin J. Some new design strategies for second-order nonlinear optical
polymers and dendrimers. Polym Chem 2011;2:2723e40.
values show good correlation with calculated
b values with the
notable exceptions being compounds OGL-4 and OGL-5. In the case
of OGL-5 the low measured nonlinearity despite the highest
[3] Luo J, Zhou XH, Jen AKY. Rational molecular design and supramolecular as-
sembly of highly efficient organic electro-optic materials.
2009;19:7410e24.
J Mater Chem
calculated
b can be attributed to the formation of aggregates dis-
[4] Mohajerani E, Gilbert A, Mitchel GR. Stability of NLO chromophores in
doped polymer films during electric field poling. J Phys D Appl Phys 1992;25:
1304e10.
[5] Lee SM, Jahng WS, Lee JH, Rhee BK, Park KH. Second-order nonlinear optical
activity vs. chromophore content in simple organic glass/PMMA system. Chem
Phys Lett 2005;411:496e500.
[6] Chen TA, Jen AKY, Cai Y. A novel class of nonlinear optical side-chain polymer:
polyquinolines with large second-order nonlinearity and thermal stability.
Chem Mater 1996;8:607e9.
[7] Lu JX, Yin J, Deng XX, Shen QS, Cao ZQ. Device-quality second-order nonlinear
optical poly(ester-imide) for electro-optic applications. Opt Mater 2004;25:
17e23.
[8] Strohriegl P, Grazulevicius JV. Charge-transporting molecular glasses. Adv
Mater 2002;14:1439e52.
[9] Shirota Y. Photo- and electroactive amorphous molecular materi-
alsdmolecular design, syntheses, reactions, properties, and applications.
J Mater Chem 2005;15:75e93.
[10] Naito K, Miura P. Molecular design for nonpolymieric organic dye glasses with
thermal stability: relations between thermodynamic parameters and amor-
phous properties. J Phys Chem 1993;97:6240e8.
[11] Salbeck J, Yu N, Bauer J, Weissortel F, Bestgen H. Low molecular organic
glasses for blue electroluminescence. Synth Met 1997;91:209e15.
[12] Gao J, Cui Y, Yu J, Lin W, Wang Z, Qian G. Enhancement of nonlinear optical
activity in new six-branched dendritic dipolar chromophore. J Mater Chem
2011;21:3197e203.
[13] Wu W, Huang L, Song C, Yu G, Ye C, Liu Y, et al. Novel global-like second-order
nonlinear optical dendrimers: convenient synthesis through powerful click
chemistry and large NLO effects achieved by using simple azo chromophore.
Chem Sci 2012;3:1256e61.
[14] Seniutinas G, Tomasiunas R, Czaplicki R, Sahraoui B, Daskeviciene M,
Getautis V, et al. Arylmethylene-1,3-indandione based molecular glasses:
third order optical non-linearity. Dyes Pigm 2012;95:33e40.
[15] Meunier A, Lebel O. A glass forming module for organic molecules: making
tetraphenylporphyrin lose its crystallinity. Org Lett 2010;12:1896e9.
[16] Hohie C, Strohriegl P, Schioter S, Hofmann U, Haarer D. Synthesis and char-
acterization of novel photorefractive low molecular weight glasses. Proc SPIE
1998;3471:29e40.
cussed previously, resulting in lowered accentric chromophore
order. The opposite situation is observed for compound OGL-4
where the d33 value exceeds expected. With a comparable struc-
ture, compound OGL-6 was predicted to perform better due to tert-
butyl group providing improved shielding from dipoleedipole in-
teractions. And if compared with our previously synthesized
indane-1,3-dione containing molecular glasses [17] this was the
case as the increase by close to 50% was achieved regarding ma-
terials NLO performance. However, OGL-4 performs notably better.
A specific structure related anisotropic self assembly and resulting
increase in non-centrosymmetry is proposed as the cause of this,
but additional experiments are planned to examine this feature in
detail. The temperature dependent decay of SH intensity was also
measured (Fig. 3) and the values TSHI50 were obtained (Table 1)
representing the temperatures at which a half of starting SH in-
tensity has vanished. These values should be close to the measured
glass transition temperatures as the steep increase in molecular
motion leads to the loss of chromophore accentric order. However,
the measured TSHI50 values are notably lower than Tg for most
compounds. This can be partly explained by the different glass
formation conditions (spin-coating and melting) and the resulting
structural differences for the amorphous materials and a lower
thermodynamic stability of supercooled system in the poled film.
The highest TSHI50 value of 92 ꢀC was measured for compound OGL-
1 what was expected as it is the compound with the highest mo-
lecular weight.
[17] Traskovskis K, Mihailovs I, Tokmakovs A, Jurgis A, Kokars V, Rutkis M. Tri-
phenyl moieties as building blocks for obtaining molecular glasses with
nonlinear optical activity. J Mater Chem 2012;22:11268e76.
[18] Jensen F, Bedard R. Notes: cleavage of tetrahydrofuran by triphenylme-
thylmagnesium bromide. J Org Chem 1959;24:874e5.
4. Conclusions
We have shown that the phase behavior modifying 1,1,1-
triphenylpentane substituents can be successfully used to obtain

1050
K. Traskovskis et al. / Dyes and Pigments 99 (2013) 1044e1050
[19] Yu BC, Shiraib Y, Tourb JM. Syntheses of new functionalized azobenzenes for
potential molecular electronic devices. Tetrahedron 2006;62:10303e10.
[20] Vanags G, Silaraja R, Glemitis G. Esters of indane-1,3-dione and 2-
chlorindane-1,3-dione 2-carboxylic acid. Chem Abstr 1963;58:483d.
[21] Gopalan P, Katz HE, McGee DJ, Erben C, Zielinski T, Bousquet D, et al. Star-
shaped azo-based dipolar chromophores: design, synthesis, matrix compati-
bility, and electro-optic activity. J Am Chem Soc 2004;126:1741e7.
[25] Jecs E, Kreicberga J, Kampars V, Jurgis A, Rutkis M. Novel azobenzene pre-
cursors for NLO active polyurethanes: synthesis, quantum chemical and
experimental characterization. Opt Mater 2009;31:1600e7.
[26] Luo J, Ma H, Haller M, Jen AKY, Barto RR. Large electro-optic activity and low
optical loss derived from a highly fluorinated dendritic nonlinear optical
chromophore. Chem Commun 2002:888e9.
[27] Li S, Li M, Qin J, Tong M, Chen X, Liu T, et al. Synthesis, crystal structures and
nonlinear optical properties of three TCF-based chromophores. Crys-
tEngComm 2009;11:589e96.
[28] Oudar JL, Chemla DS. Hyperpolarizabilities of the nitroanilines and their re-
lations to the excited state dipole moment. J Chem Phys 1977;66:2664e8.
[29] Bella SD. On the determination of the molecular static first hyper-
polarisability: how reliable are literature data? New J Chem 2002;26:495e7.
[22] Arndt F. Diazomethane. Org Syn Coll 1943;2:165.
[23] Larner BW, Peters AT. New intermediates and dyes. Part II. Preparation and
properties of 4-tert-butylphthalic anhydride. Orientation of its condensation
products with benzenoid hydrocarbons. J Chem Soc 1952:680e6.
[24] Herman WN, Hayden LM. Maker fringes revisited: second-harmonic generation
from birefringent or absorbing materials. J Opt Soc Am B 1995;12:416e27.