Design and controlled synthesis by dual polymerization of new organic-inorganic hybrid material for photonic devices

Article abstract of DOI:10.1039/c4ra01059k

Organic-inorganic hybrid material was synthesized by double polymerization processes i.e. a sol-gel process and organic polymerization respectively. For this study, hybrid monomer, 4-vinyl ether-phenyltriethoxysilane (VEPTES) was used as starting building block. First, the silica matrix with tunable ratio of siloxane and silanol units was synthesized by a sol-gel process under acidic conditions and the organic network was formed by cationic photopolymerization of vinyl ether groups. Mineral and organic polymerization kinetics were respectively monitored by liquid ²⁹Si-NMR and IR spectroscopy. The effect of the silicate backbone on the organic photopolymerization process was studied and elucidated. The optical performance of this new hybrid material has been studied using the near-infrared spectroscopy.

Full text of DOI:10.1039/c4ra01059k

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Design and controlled synthesis by dual
polymerization of new organic–inorganic hybrid
Cite this: RSC Adv., 2014, 4, 17210
material for photonic devices
Saly Yaacoub, ab Sylvie Calas-Etienne,a Jihane Jabbour,b Kassem Amro,a Rabih Tauk,b
*
Antonio Khoury,b Ahmad Mehdic and Pascal Etiennea
Organic–inorganic hybrid material was synthesized by double polymerization processes i.e. a sol–gel
process and organic polymerization respectively. For this study, hybrid monomer, 4-vinyl ether-
phenyltriethoxysilane (VEPTES) was used as starting building block. First, the silica matrix with tunable
ratio of siloxane and silanol units was synthesized by a sol–gel process under acidic conditions and the
organic network was formed by cationic photopolymerization of vinyl ether groups. Mineral and organic
polymerization kinetics were respectively monitored by liquid 29Si-NMR and IR spectroscopy. The effect
of the silicate backbone on the organic photopolymerization process was studied and elucidated. The
optical performance of this new hybrid material has been studied using the near-infrared spectroscopy.
Received 6th February 2014
Accepted 25th March 2014
DOI: 10.1039/c4ra01059k
www.rsc.org/advances
from the photolysis of diaryliodonium or triarylsulfonium salts
Introduction
bearing weakly coordinating anions suchꢀas hexauoroarsenate
ꢀ
anion AsF6 or hexauorophosphate PF6
.
Organic–inorganic hybrid materials are of central attention in
many elds since they combine both characteristics of inor-
ganic and organic materials.1–7 ORganically MOdied SILlicates
(ORMOSILs) materials are generally obtained from a molecular
precursor R0Si(OR)3 which contains two reactive parts and offers
two polymerization (inorganic and organic) processes. The
inorganic network is obtained by hydrolysis and poly-
condensation of trialkoxysilyl groups Si(OR)3 (sol–gel process)
while the organic network is generated by the photo-
polymerization of the reactive organic function (vinyl, allyl,
epoxy.) of the R0 part. In contrast to radical polymerization,
cationic polymerization is not inhibited by oxygen, but can be
affected by the presence of hydroxyl groups which act as
nucleophiles.8
Among cationically cured systems, vinyl ether is the most
reactive monomer known in this type of polymerization.9 Using
an appropriate initiator, cationic polymerization of vinyl ether
is dened as an additional polymerization reaction mediated by
propagation carbocation derived from a vinyl ether monomer
initiated by strong Bronsted acid. Common cationic photo-
initiators are based on diaryliodonium or triarylsulfonium
salts.10–12 These strong protonic acids are generally obtained
Organic–inorganic hybrid materials are particularly attrac-
tive for photonics and integrated optics13–15 but the main factor
limiting the development of these devices is the propagation
losses. In fact, in organic–inorganic hybrid materials synthe-
sized by a sol–gel process,16 the presence of OH groups under-
lies the attenuation at the second and the third
telecommunication windows. In addition, the presence of CH
aliphatic groups, in the precursor, can contribute to the
absorption at some wavelengths.17
The most powerful experimental technique for studying
silicon-based sol–gel chemistry remains the nuclear magnetic
resonance (29Si-NMR).3,18 An 29Si-NMR of a sol–gel mixture
allows the identication and quantication of all species
present in solution. Hydrolysis and condensation reactions of
VEPTES-based sol have been studied as a function of several
synthesis parameters (temperature, time, pH and concentra-
29
tion) using Si-NMR spectroscopy and reported recently.19 We
have shown clearly the inuence of these different parameters
on the kinetic reactions in order to obtain a complete hydrolysis
of all alkoxide units with the highest possible condensation
rate.
¨
Besides the mineral polymerization we investigate, in this
paper, the photoinitiating polymerization behaviors of vinyl
ether units according to a cationic mechanism. The quantitative
study of the vinyl ether conversion was based on infrared
spectra collected in the range (400–4000 cmꢀ1).
Versace et al. have studied the relation between organic and
inorganic network polymerizations.20 They reported that an
increase of inorganic condensation degree induces a decrease
aCharles Coulomb Laboratory, University of Montpellier 2, Place E. Bataillon, UMR
5221, cc074, 34095 Montpellier Cedex 5, France. E-mail: saly.yaacoub@
univ-montp2.fr
bPlatform for Research in Nanosciences and Nanotechnology, Lebanese University,
Campus Pierre Gemayel, Fanar-Metn, BP90239-Lebanon
c
´
Institute Charles Gerhardt, Chimie Moleculaire et Organisation du solide, UMR 5253,
`
cc1701, Place Eugene Bataillon, 34095 Montpellier Cedex 5, France
17210 | RSC Adv., 2014, 4, 17210–17217
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
of organic conversion rate. For these reasons, synthesis condi- 4-bromophenylvinylether 2, a slightly yellow oil. Yield: 76%. 1H
tions have to be carefully chosen to ensure both a sufficient NMR (200 MHz, CDCl3) d ppm: 7.46 (d, J ¼ 8 Hz, 2H, Ar), 6.92 (d,
degree of mineral condensation and an important organic J ¼ 8 Hz, 2H, Ar), 6.63 (dd, 1H, HC ¼ ), 4.83 (d, J ¼ 8 Hz, 1H,
conversion rate.
H2C]), 4.52 (d, J ¼ 8 Hz, 1H, H2C]); 13C NMR (50 MHz, CDCl3)
This paper deals with the structural characterization of d ppm: 155.84 (Ar), 147.72 (Ar), 132.57 (Ar), 118.84 (Ar), 115.63
double polymerization process for the two reactive parts of a (HC]), 95.93 (H2C]).
new hybrid precursor. In order to reduce the amount of groups
involved in the attenuation, the new synthesized precursor
contains a reduced number of aliphatic CH groups in compar-
ison with other hybrid precursors.19,21 In parallel, mineral
polymerization has been followed through 29Si-NMR spectros-
copy in order to reduce the presence of OH groups. In this
respect, near infrared spectroscopy was used to study the
performance of this new hybrid material at the optical trans-
mission windows.
Synthesis of 4-vinyletherphenyltriethoxysilane (VEPTES)
A solution of 4-bromophenylvinylether 2 (55.0 g, 0.276 mol) in
THF (115 mL) was dropped at room temperature into a three
neck ask containing (8.6 g, 0.360 mol) of magnesium and 70
mL of dried THF. The resulting mixture was heated under reux
for 2 hours. Aer being cooled to room temperature, the ashen
solution was dropped into a 1 L two neck ask containing a
solution of tetraethoxysilane (229.0 g, 1.10 mol) in THF (300 mL)
and cooled at ꢀ30 ꢁC. Aer 48 hours of stirring at room
temperature, the mixture was ltrated and the ltrate was
concentrated under vacuum to give yellow oily curd crude. Aer
distillation under vacuum, 4-vinyletherphenyltriethoxysilane
(VEPTES) was obtained as a colorless oil. Yield: 71%. Bp: 90–92
ꢁC at 0.04 mmHg. 1H NMR (200 MHz, CDCl3) d ppm: 7.70 (d, J ¼
8 Hz, 2H, Ar), 7.07 (d, J ¼ 8, 2H, Ar), 6.72 (dd, 1H, HC ¼ ), 4.87
(dd, 1H, H2C ¼ ), 4.52 (dd, 1H, H2C ¼ ), 3.92 (q, 6H, CH2O), 1.30
(t, 9H, CH3); 13C NMR (50 MHz, CDCl3) d ppm: 158.62 (Ar),
147.48 (Ar), 136.57 (Ar), 124.97 (Ar), 116.34 (HC]), 95.88
(H2C]), 58.72 (CH2O), 18.12 (CH3); 29Si NMR (40 MHz, CDCl3) d
ppm: ꢀ57.3.
Experimental section
Materials
All reagents were purchased from Alfa Aesar. Tetrahydrofuran
was distilled from sodium benzophenone immediately before
use. The 1H, 13C and 29Si-NMR spectra were recorded on a
Bruker Advance 200 DPX spectrometer. The bifunctional
precursor 4-vinyletherphenyltriethoxysilane (VEPTES) was used
as starting monomer. It was prepared by silylation of 4-bro-
mophenylvinylether according to Scheme 1.
Synthesis of 4-bromophenylvinylether
Igracure 250 (iodonium, (4-methylphenyl)[4-(2-methyl-
propyl)-phenyl]-hexauorophosphate) and Darocure ITX
(mixture of 2-isopropyltrioxanthone and 4-isopropyl-
thioxanthone) were purchased from Ciba Speciality Chemicals
Inc. The initiator concentrations were between 1 and 5 wt%
with respect to the vinyl ether monomer mass. A thioxanthone
photosensitizer was used to expand the spectral region over
which the cationic initiator is not effective. Typically, photo-
sensitizers make it possible to initiate the photopolymerization
using near-UV or even visible wavelengths of light.22–25 Other
than iodonium-based PAG, (tris[4-(4-acetylphenyl)sulfanyl-
In a 1 L two neck ask, (318 g, 1.7 mol) of dibromoethane and
(109 g, 0.80 mol) of potassium carbonate were mixed in 500 mL
of acetonitrile. A solution of 4-bromophenol (67 g, 0.38 mol) in
ꢁ
acetonitrile (70 mL) was slowly dropped in the mixture at 0 C
(ice bath) which was then heated under reux for one night.
Once the solution was cooled down to room temperature,
ltered and concentrated, an orange solid substance was
obtained. The solid was crystallized from methanol to give 50 g
(0.18 mol) of 1-bromo-4-(2-bromoethoxy) benzene. This last one
was dissolved in 70 mL of dried tetrahydrofuran (THF) and a
solution of potassium tertbutoxide (27 g, 0.24 mol) in THF
(80 mL) at 0 ꢁC under argon atmosphere. Aer stirring at room
temperature for one night, the mixture was ltered and the
obtained residue was dissolved in pentane and extracted
with water (3 ꢂ 100 mL). The combined extracts were dried
over MgSO4, and concentrated under vacuum to give
phenyl]sulfonium
tris(triuoromethylsulfonyl)methide),
GSID26-1, is used in this study. These two photoactives
compositions, Irgacure 250/Darocure ITX and PAG GSID26-1,
are used to carry out the cationic photopolymerization of vinyl
monomers with long-wavelength UV light at 365 nm. Molecular
structures of the cationic photoinitiator and photosensitizer are
shown in Fig. 1.
Sol synthesis
Based on previous works on phenyltriethoxysilane26 and in
order to obtain a complete hydrolysis of the three alkoxide
units with the highest possible condensation rate, it is
important to work in an acidic medium with a high hydrolysis
ꢀ
ꢁ
H2O
PhSiðOEtÞ3
rate R ¼
. Under these conditions, hydrolysis is
fast compared to condensation reaction, most of alkoxy groups
are converted to hydroxyl groups leading to the formation of a
highly condensed silicate polymers.27–29
Scheme 1 Schematic pathway for synthesis of VEPTES.
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 17210–17217 | 17211

View Article Online
RSC Advances
Paper
pulse program, spectral width of 240 ppm and 12k scans. All
signals were referenced to tetramethylsilane (TMS). However,
in order to obtain a quantitative and good spectrum, chro-
mium(III) acetylacetonate Cr(acac)3 must be added to aid in
the relaxation of the 29Si nuclei, thus allowing quantitative
spectral accumulation with the 3.0 s delay. Mah et al.30 and
Hook31 have used Cr(acac)3 as the non-polar paramagnetic
relaxation agent showing no effect on sol–gel reactions,
resulting only in a shortening of the delay time. The species
formed during sol–gel process were followed according to
their chemical shi.
In the sol–gel reaction of the precursor, the ethoxy silyl
groups are transformed into hydroxyl groups and siloxane
bonds by hydrolysis and polycondensation, respectively. Clas-
sical Tji notation is used for the different trifunctional silicate
species depending on the number of oxygen bridging atoms
surrounding the central silicon atom. i represents the number
of siloxanes and j the number of silanols.
Fig.
1 Molecular structures of used cationic photoinitiators and
photosensitizer.
Structural investigation of VEPTES UV-cured lms using
FTIR spectroscopy. Silicon wafers with h100i orientation were
supplied from SIL'TRONIX Silicon Technologies. The studied
lms were spin coated onto bare silicon wafers that are trans-
parent to infrared light.
The starting sol corresponds to the precursor hydrolysis rate
of 8 in aqueous HCl solution of 10ꢀ2 M (see Fig. 2). A ratio
(16 : 1) of (ethyl alcohol (99%) : VEPTES) is necessary to obtain a
homogeneous solution. For this, two solutions (VEP-
TES : ethanol: 1 : 8) and (HCl : ethanol: 8 : 8) were prepared.
The obtained HCl–ethanol solution was dropped under stirring
to the VEPTES–ethanol.
IR analyses were performed to investigate the conversion
rate of vinyl ether monomers under UV irradiation. Trans-
mission FTIR spectroscopy is the mostly used method to
monitor the photopolymerization process of UV-cured coatings.
Thick lms are shown to be unsuitable for structural investi-
gation using transmission IR spectroscopy, saturation bands
quickly appears for such lms. To avoid this problem, the lm
thickness should be around 1 mm. FTIR spectra were recorded
on a Nicolet FTIR-spectrometer 510p in the medium infrared
range of 4000–400 cmꢀ1. The sample was inserted into a slide
frame and placed in the compartment of the spectrophotom-
eter. The experiments were conducted at room temperature in
absorption mode at 8 cmꢀ1 resolution with 32 scans per spec-
trum to reduce the noise.
Aer the complete addition of the acidic solution, the
ꢁ
mixture was continuously stirred for 1, 2 and 3 days at 60 C.
Aerwards, the solvent was removed by rotavapor at 30 ꢁC under
vacuum (42 mbar) to obtain a dry extract value about 55%.
Finally, the appropriate photoinitiator and photosensitizer were
added to the obtained sol.
Characterization
Structural characterization of the sol by liquid 29Si-NMR
spectroscopy. Liquid state NMR spectra of 29Si were recorded
at 79.49 MHz on a spectrometer (Bruker Avance 400). The
accumulations parameters are 10 s of relaxation delay, zgig
The evolution of vinyl ether conversion a with time, t, was
calculated by the following equation:32
Ct
a ¼ 1 ꢀ
(1)
C0
Ct and C0 are the concentrations of vinyl ether group at time t
and t ¼ 0. Thereby, in order to follow the absorption of the IR
beam by the organic function, the absorbance has been intro-
duced through the Beer–Lambert law:
A ¼ 3lC
(2)
3: molar extinction coefficient (L molꢀ1 cmꢀ1), l: path length
(cm), C: molar concentration (mol Lꢀ1).
Using eqn (1) and (2) the degree of curing of vinyl ether
groups is determined at any time according to the following
equation:
À
ꢂ
Avinyl Aref
ꢂ
Á
t
À
Á
a ¼ 1 ꢀ
(3)
Avinyl Aref
0
Fig. 2 Protocol of sols synthesis (a: 1 day, b: 2 days, c: 3 days at 60 ꢁC).
17212 | RSC Adv., 2014, 4, 17210–17217
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
Avinyl and Aref correspond to the area of the characteristic vinyl
ether and reference peak respectively.
The evolution of the vinyl ether absorbance is compared to
a reference vibration band, independent from the curing
process.
Results and discussion
Mineral polymerization
Liquid 29Si nuclear magnetic resonance spectroscopy was used
to characterize the structural evolution of the mineral network
during the sol–gel synthesis.19
For the following spectra, chemical shis for the species
formed during these reactions are listed in Table 1.
Fig. 3 presents 29Si-NMR spectra of concentrated sols a, b
and c.
Table 1 Chemical shifts and structural formulas of species obtained
from hydrolysis–condensation of VEPTES
Fig. 3 Liquid 29Si-NMR spectra of sols a, b and c after concentration.
Chemical shi
Structural formula
of species (ppm)
We have already shown that working at 60 ꢁC is the best way
to enhance the sol–gel reaction.19 In each spectrum described in
Fig. 3, the hydrolysis species disappear, leaving place to the
condensation ones. T1, T2 and T3 species and their concentra-
tion are summarized in Table 2.
0
ꢀ57 T0
0
ꢀ54 T1
The concentration of the sol aer 1 day induces a
decrease of 11% of T1 species for the benet of T2 and T3
species which increase respectively by 5% and 15% respec-
tively. The same evolution was observed for both sols b and
c. It is worth noting that, the sol concentration induces a
decrease of T1 species for the benet of T2 and T3 species
and subsequently an increase in the condensation rate of
12% and 11% respectively. This evolution highlights
the effect of the sol concentration on the condensation of
silanol groups (Si–OH) and the formation of siloxane units
(Si–O–Si).
0
ꢀ53 T2
0
ꢀ51 T3
1
Example T0
ꢀ61 to ꢀ65 T1
2
Example T0
Besides the synthesis of mineral network, cationic photo-
polymerization process allows the building of organic network.
ꢀ69 to ꢀ71 T2
Table 2 The condensation rate before and after concentration of
solvent for different aging time at 60 ꢁCa
3
Aging time
Example T0
Sol a
Sol b
Sol c
Species
(%)00 (%)000 (%)00 (%)000 (%)00 (%)000
0
T2
1
8
35
53
3
—
—
24
58
18
65
1
1
34
54
10
—
—
15
64
21
0.4
0.6
28
55
16
62
—
—
7
67
26
73
0
ꢀ78 T3
T3
T1
T2
T3
ꢀ
ꢁ
T1
3
50
57.3 69
Condensation rate
i
(%)
a
00: % before concentration, 000: % aer concentration.
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 17210–17217 | 17213

View Article Online
RSC Advances
Paper
In order to choose the optimized sol, the correlation between
the formation of inorganic and organic phases was studied for
the last three solutions.
Organic polymerization
This part is dedicated to the investigation of UV curing behav-
iors using the FTIR spectroscopy. The samples were prepared by
spin-coating a VEPTES photoactive based solution on a silicon
wafer.
Fig. 4 shows the spectrum of the lm before UV curing in
middle infrared region. The characteristic absorption bands of
interest are reported in Table 3.
The conversion rate was calculated using eqn (3). In this
case, the double bond conversion rate during the cationic
photopolymerization of VEPTES was determined according to
the characteristic absorption band of vinyl ether centered at
1645 cmꢀ1 and to the aromatic double bond C]C at 1509 cmꢀ1
used as reference for the normalization of vinyl ether peak
absorbance. The Avinyl and Aref for t ¼ 0 were determined from
the spectrum recorded just aer the deposition of the lm on
the silicon substrate.
Fig. 5 Vinyl ether conversion curves of sols a, b and c in function of
irradiation time. Inset: conversion rate versus condensation rate.
Fig. 5 shows the conversion rate versus exposure time for UV-
cured lms deposited using sols a, b and c.
For all sols, the conversion rate increased with the exposure
time. A similar behavior has been observed previously for vinyl
ether-based photosensitive lms.34
In addition, it was clearly shown that the organic and inor-
ganic polymerizations are related (see inset Fig. 5). There was a
signicant slowing of the vinyl ether conversion rate with
increasing sol aging. The vinyl ether conversion decreased from
72% to 29%, when the inorganic condensation degree increased
from 65% to 73%.
The mobility of the polymerizable groups is reduced by the
increase of the condensation rate of mineral network. In other
words, the reactive species of organic network are xed by the
inorganic network which prevents their propagation to other
active monomers. The similar effect has been already observed
in the literature in the case of epoxy polysilsesquioxane
resins.20,35
Inuence of the aging time of the sol
Structural investigation of UV-cured lms based on VEPTES
begins by the characterization of vinyl ether-conversion rate
versus the aging time of the starting sol. The photo-
polymerization process of vinyl ether groups has been studied
using an UV LED33 lamp with an emission spectrum at 365 ꢃ 5
nm and with an illumination equal to 25 mW cmꢀ2 ꢃ 10%.
To make sol a, b and c sensitive at 365 nm, Irgacure 250/
Darocure ITX was added by respective weight percentages 5%
and 1% of VEPTES.
From these observations, sol a seems to be the most suitable
sol to obtain the minimum residual silanol (Si–OH) groups and
in the same time the highest possible condensation rate. The
amount of OH groups is necessary to ensure a good adhesion of
the coating adhesion on silicon substrate.
Inuence of the photoactive composition
The conversion rate has been also studied as a function of the
photoactive composition. For this investigation, two systems
were tested for their efficiency to start the curing reaction of
VEPTES-based resins at 365 nm. Fig. 6 presents the vinyl ether
conversion curves versus irradiation time for VEPTES UV-cured
lms using two photopolymerizable compositions. For the rst
one, the photoinitiator PAG GSID26-1 was added to the
concentrated sol a at a 5% wt of VEPTES.
Fig. 4 Middle infrared spectrum of sol–gel film before irradiation.
Table 3 Main absorption bands in middle infrared region
For the second one, the couple Irgacure 250/Darocure ITX
was added by respective weight percentages 5% and 1% of
VEPTES.
It was observed for both compositions, that the conversion
rate increased with the exposure time with better reactivity for
the second composition (Irgacure 250/Darocure ITX). With the
Wavenumber (cmꢀ1
)
Characteristics vibration
1000–1130
1645
1509/1595
3390
Si–O asymmetric stretching of Si–O–Si
C]C stretching of vinyl ether bond
C]C stretching of aromatic bond
OH stretching
17214 | RSC Adv., 2014, 4, 17210–17217
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
Fig. 7 Vinyl ether conversion curves versus irradiation time for VEPTES
UV-cured films for different SB times at 60 ꢁC.
Fig. 6 Vinyl ether conversion curves versus irradiation time for VEPTES
UV-cured films with two photoactive compositions.
rst composition, the conversion reached 26% aer 60 s, while
the couple Irgacure 250/Darocure ITX reached 70%. Then, it
continues to grow and reached 81% aer 145 s, instead of 40%
for the rst one.
It is well known, with UV irradiation, that diaryliodonium
and triarylsulfonium salts undergo a photodissociation which
¨
leads to the generation of a strong Bronsted acid. The produced
strong acid is responsible of the initiation of cationic poly-
merization by the direct protonation of the monomer. The
strength of the acid depends on the character of the anion
present on the starting onium salt. Maximum rates of poly-
merization are achieved when the anion is nonnucleophilic.36
Since, the strength of the acid is related to conjugate formed-
base. Thereby, the low reactivity of GSID26-1 in comparison to
that of Irgacure 250/Darocure ITX-based composition is related
to the strongeꢀr nucleophilic character of (CF3SO2)3Cꢀ compared
Fig. 8 Conversion rate a versus the SB annealing time.
to that of PF6
.
conversion rate reached 81%, 75%, 64% and 61% for layers
respectively aer 0, 1, 3 and 5 minutes at 60 C.
Based on these observations, the couple Irgacure 250/Dar-
ocure ITX was chosen for the next study.
ꢁ
The relation between the conversion rate and the annealing
time SB were summarized in Fig. 8 for two exposure time at
365 nm.
Inuence of photolithography parameters
This gure proves the inuence of the SB annealing time on
According to the standard process of so lithography, aer the conversion rate and subsequently the inuence of the
deposition, the photoresist coating is subjected to a so-bake. crosslinking of mineral network. When the crosslinking of
The so-bake SB is done to drive away the solvent from the inorganic network is more advanced, the conversion of organic
liquid lm in order to improve the adhesion on the silicon species is less important. This supports our previous
wafer.
Fig. 7 presents the conversion rate of the photoresist coating
observations.
Associated with a high conversion rate, the rate of poly-
as a function of exposure time for different baking times at merization reaction is a key parameter in photolithography
60 ꢁC. Four photoresist lms deposited on silicon substrate process. Fig. 9 represents the variation of the polymerization
underwent a so bake respectively for 0, 1, 3 and 5 min at 60 ꢁC. rate with the SB annealing time. The rate of polymerization
Aer recording the spectrum at t ¼ 0 as a reference, each of reaction is calculated from the slope of each curve in Fig. 7
these layers is studied as a function of exposure time at 365 nm between 0 s and 50 s.
using IR spectroscopy in the middle range.
This study shows that the rate of polymerization reaction
It was observed that the conversion rate increased with the decreases with the increase of SB annealing time. Hence, SB
exposure time for any SB annealing time (see Fig. 7). However, annealing time has a signicant inuence on the crosslinking
the evolution is not proportional. The conversion rate decreases of the inorganic network, which obviously affects the conver-
with the increase of the annealing time SB. Aer 145 s, the sion rate of the vinyl ether function and also the rate of
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 17210–17217 | 17215

View Article Online
RSC Advances
Paper
Fig. 11 Absorption spectrum for VEPTES-based film.
Fig. 9 Polymerization rate as a function of SB annealing time.
Table 4 Attribution of the absorption bands of the near infrared
spectrum of VEPTES-based coating
polymerization reaction. A high conversion rate must be
obtained by a short SB annealing time.
Till now, we have studied the evolution of the conversion rate
during the rst stage of the photolithography process, mainly
the So-Bake (SB). It is important to study the evolution
throughout the process, from the exposure step and Post-
Exposure Bake (PEB) until the densication treatment, Hard-
Bake (HB).
l (nm)
Vibration mode
1369
1380–1450
1615
1680
1915
OH of Si–OH (elongation)
2n CH (CH2 or CH3) + d CH (CH2 or CH3)
2n CH (–CH]CH2)
2n CH (CH2 or CH3)
OH of H2O exterior (deformation and elongation)
The evolution of vinyl ether-conversion rate at every step of
the photolithographic process is represented in Fig. 10. We
chose to perform this study on a layer without the rst
annealing SB. The circle dots correspond to the exposure step,
the square dots to the PEB step carried out at 80 ꢁC and the
Aer exposure for 60 seconds, we studied the effect of each
annealing, PEB and HB. The conversion rate continues to
increase up to be 92% aer 5 min at 80 ꢁC and 100% just aer
ꢁ
ꢁ
the rst een minutes at 100 C (during the HB).
triangle one for the HB stage performed at 100 C.
Fig. 10 shows clearly the impact of each step in the photo-
lithography process on the conversion rate of organic function.
At rst, a increases very rapidly with exposure time. It increases
from 7% aer 5 s to 72% aer one minute of exposure. This
evolution shows that an increase of the exposure time enhances
greatly the polymerization of vinyl ether functions. This part of
the curve is similar to that observed on Fig. 6 for the second
photoactive composition Irgacure 250/Darocure ITX. Thus, the
result shows the excellent reproducibility of the solution
behavior.
Performance of VEPTES-based coating at 1310 and 1550 nm
In order to use this material for integrated optical devices,
optical performances were studied in the near infrared region.
The near-infrared spectral range (800 to 2500 nm) contains the
two optical transmission windows located at 1310 and 1550 nm
allowing the absorption study of the material.
The lms were deposited on glass microscope slides, which
have been previously treated with piranha solution.
Fig. 11 presents the spectrum obtained in the near infrared
region for VEPTES-based coating.
This spectrum shows the presence of several absorption
bands (overtones and combination bands of fundamental
molecular vibrations), the main ones are shown in Table 4. The
various absorption bands due to overtones and combinations
bands of CH and OH groups.
The centered band at 1680 nm corresponds to the rst
harmonic (2n) of CH bonds in CH2 and CH3 groups. The band at
1615 nm corresponds to the rst harmonic (2n CH) linked to the
vinylic CH groups. The band between 1380 and 1450 nm is due
to many possible combinations between the rst harmonic (2n
CH) and the bending (d CH) of CH aliphatic groups. Then, the
presence of these groups contributes slightly to the absorptions
at 1310 and 1550 nm (observed at the foot of the band).
Therefore, VEPTES-based lm has low level of absorptions at
1310 and 1550 nm.
Fig. 10 Conversion rate throughout the photolithography process.
17216 | RSC Adv., 2014, 4, 17210–17217
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
12 Y. Yagci and I. Reetz, Prog. Polym. Sci., 1998, 23, 1485.
Conclusions
13 H. Krug, F. Tiefensee, P. W. Oliveira and H. K. Schmidt, Sol–
Gel Optics II, Proc. SPIE, 1992, vol. 1758, p. 448.
14 C. Sanchez and P. Gomez-Romero, Functional hybrid
materials, WILEY-VCH, Weinheim, 2004.
15 H. Schmidt and M. Popall, Sol–gel optics, Proc. SPIE, 1990,
vol. 1328, p. 249.
16 M. Oubaha, P. Etienne, S. Calas, P. Coudray, J. M. Nedelec
and Y. Moreau, J. Sol–Gel Sci. Technol., 2005, 33, 241.
17 M. Oubaha, P. Etienne, P. Boutinaud, J. M. Nedelec,
P. Coudray and Y. Moreau, Opt. Mater., 2006, 28, 502.
18 F. Babonneau and J. Maquet, Polyhedron, 2000, 19, 315.
19 S. Yaacoub, S. Calas-Etienne, J. Jabbour, R. Tauk, A. Khoury,
A. Mehdi and P. Etienne, J. Sol–Gel Sci. Technol., 2013, 67,
384.
In this work, the dual polymerization of VEPTES resins was
investigated. The resulting mineral network was synthesized
under acidic conditions via the sol–gel process, followed by an
organic photopolymerization in presence of a diaryliodonium
salt. Sol–gel reactions were monitored by liquid 29Si-NMR
spectroscopy and the photocuring behavior of vinyl ether based
monomers was investigated using infrared spectroscopy in the
middle range.
The correlation between mineral and organic polymerization
has been studied to ensure both a sufficient degree of poly-
merization. A signicant slowdown of vinyl ether conversion
rates with increasing sol aging was observed.
Finally, the optical performance of VEPTES-based resin has
been studied using the near-infrared spectroscopy. Results
show a promising transmission level for the use of VEPTES-
based lm at the second and the third telecommunication
windows. Optical losses measurements are under investigation.
´
20 D. L. Versace, A. Chemtob, C. Croutxe-Barghorn and
S. Rigolet, Macromol. Mater. Eng., 2010, 295, 355.
21 J. Jabbour, S. Calas-Etienne, M. Smaihi, S. Gatti, R. Kribich,
G. Pille, Y. Moreau and P. Etienne, Appl. Surf. Sci., 2007, 253,
8032.
22 J. V. Crivello, Adv. Polym. Sci., 1984, 62, 1.
23 W. D. Cook, S. Chen, F. Chen, M. U. Kahveci and Y. J. Yagci,
J. Polym. Sci., Part A: Polym. Chem., 2009, 47, 5474.
24 J. D. Cho and J. W. Hong, J. Appl. Polym. Sci., 2005, 97, 1345.
25 J. V. Crivello and U. Bulut, J. Polym. Sci., Part A: Polym. Chem.,
2005, 43, 5217.
Acknowledgements
The authors would like to acknowledge the nancial support
provided by the Lebanese National Council for Scientic
Research CNRS (via the Grant Research program) and the social
foundation Azm and Saade (via a Program of cooperation with
the Lebanese university).
26 Y.-S. Li, Y. Wang and S. Ceesay, Spectrochim. Acta, Part A,
2009, 71, 1819.
27 C. J. Brinker, K. D. Keefer, D. W. Schaefer and C. S. Ashley,
J. Non-Cryst. Solids, 1982, 48, 47.
Notes and references
28 R. A. Assink and B. D. Kay, Annu. Rev. Mater. Sci., 1991, 21,
491.
1 C. Sanchez, B. Julian, P. Belleville and M. Popall, J. Mater.
Chem., 2005, 15, 3559.
29 L. W. Kelts, N. J. Effinger and S. M. Melpolder, J. Non-Cryst.
Solids, 1986, 83, 353.
2 C. Sanchez, P. Belleville, M. Popall and L. Nicole, Chem. Soc.
Rev., 2011, 40, 696.
30 S. K. Mah and I. J. Chung, J. Non-Cryst. Solids, 1995, 183, 252.
31 R. J. Hook, J. Non-Cryst. Solids, 1996, 195, 1.
32 E. G. Karayannidou, D. S. Achilias and I. D. Sideridou, Eur.
Polym. J., 2006, 42, 3311.
3 P. Judeinstein and C. Sanchez, J. Mater. Chem., 1996, 6, 511.
4 B. Arkles, MRS Bull., 2001, 26, 402.
5 G. Schottner, Chem. Mater., 2001, 13, 3422.
´
6 E. Besson, A. Mehdi, C. Reye, R. Corriu and A. Gibaud, Silicon
33 http://www.kloe.fr/en/our-solutihttp://www.kloe.fr/fr/nos-
solutions-lithographie/gamme-uv-kub, Gamme UV-KUB
2013.
Based Polym., Springer Science, 2008, p. 223.
7 A. Mehdi, C. Reye and R. Corriu, Chem. Soc. Rev., 2011, 40,
563.
´
34 A. Chemtob, C. Belon, C. Croutxe-Barghorn, S. Rigolet,
8 A. D. Jenkins and A. Ledwith, Reactivity, mechanism and
structure in polymer chemistry, John Wiley and Sons,
London, 1974.
´
L. Vidal, J. Brendle, J. Mandel and N. Blanchard, Polym.
Eng. Sci., 2011, 51, 1466.
35 J. Chen and M. D. Soucek, J. Appl. Polym. Sci., 2003, 90,
9 C. Decker, C. Bianchi, D. Decker and F. Morel, Prog. Org.
Coat., 2001, 42, 253.
2485.
36 J. V. Crivello, Nucl. Instrum. Methods Phys. Res., Sect. B, 1999,
151, 8.
10 J. V. Crivello, J. Polym. Sci., Part A: Polym. Chem., 1999, 37, 4241.
11 J. V. Crivello and J. H. W. Lam, Macromolecules, 1977, 10, 1307.
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 17210–17217 | 17217