Novel photoinitiator with a radical quenching moiety for confining radical diffusion in two-photon induced photopolymerization

Article abstract of DOI:10.1039/c0jm04025h

We designed and synthesized a novel two-photon induced photoinitiator with radical quenching moiety, 3,6-bis[2-(4-nitrophenyl)-ethynyl]-9-(4-methoxybenzyl) -carbazole (BNMBC), and successfully demonstrated its optical properties, high initiating efficiency in two-photon induced polymerization (TPIP) and capability to fabricate 3D structures with high resolution. The high resolution in TPIP using BNMBC was achieved as compared to the reported photoinitiator 3,6-bis[2-(4-nitrophenyl)-ethynyl]-9-benzylcarbazole (BNBC). BNMBC was confirmed to have a large two-photon absorption cross-section of 2367 GM by Z-scan measurement. We investigated the photopolymerization properties of resins R ₁-R₃ in which BNBC, BNBC/phenyl methyl ether (PhOMe) and BNMBC were used as photoinitiators, respectively. The TPIP fabrication experiments exhibited that BNMBC possessed high TPIP initiating efficiency and an effective radical quenching effect. The volumes of polymer fibers fabricated by the TPIP of the photoresist using BNMBC as the photoinitiator were decreased to 20%-30% of those synthesised using BNBC as a photoinitiator. The introduction of a radical quenching group to the photoinitiator resulted in the effective confining effect of radical diffusion compared to the addition of the same molar ratio of radical quencher. This kind of photoinitiator with radical quenching group should benefit the fabrication of precise structures. The Royal Society of Chemistry.

Full text of DOI:10.1039/c0jm04025h

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links <
C
Journal of
Materials Chemistry
Cite this: J. Mater. Chem., 2011, 21, 5650
www.rsc.org/materials
PAPER
Novel photoinitiator with a radical quenching moiety for confining radical
diffusion in two-photon induced photopolymerization
a
a
a
a
Wei-Er Lu,ab Xian-Zi Dong, Wei-Qiang Chen, Zhen-Sheng Zhao and Xuan-Ming Duan
*
*
Received 21st November 2010, Accepted 16th February 2011
DOI: 10.1039/c0jm04025h
We designed and synthesized a novel two-photon induced photoinitiator with radical quenching
moiety, 3,6-bis[2-(4-nitrophenyl)-ethynyl]-9-(4-methoxybenzyl)-carbazole (BNMBC), and successfully
demonstrated its optical properties, high initiating efficiency in two-photon induced polymerization
(TPIP) and capability to fabricate 3D structures with high resolution. The high resolution in TPIP
using BNMBC was achieved as compared to the reported photoinitiator 3,6-bis[2-(4-nitrophenyl)-
ethynyl]-9-benzylcarbazole (BNBC). BNMBC was confirmed to have a large two-photon absorption
cross-section of 2367 GM by Z-scan measurement. We investigated the photopolymerization
properties of resins R1–R3 in which BNBC, BNBC/phenyl methyl ether (PhOMe) and BNMBC were
used as photoinitiators, respectively. The TPIP fabrication experiments exhibited that BNMBC
possessed high TPIP initiating efficiency and an effective radical quenching effect. The volumes of
polymer fibers fabricated by the TPIP of the photoresist using BNMBC as the photoinitiator were
decreased to 20%–30% of those synthesised using BNBC as a photoinitiator. The introduction of
a radical quenching group to the photoinitiator resulted in the effective confining effect of radical
diffusion compared to the addition of the same molar ratio of radical quencher. This kind of
photoinitiator with radical quenching group should benefit the fabrication of precise structures.
from both the viewpoint of fabrication technique and photoresist
design. As the results, the resolution of TPIP has been achieved
in nanometre scale.19–23
1. Introduction
The two-photon induced photopolymerization (TPIP) technique
has been extensively studied and established as a powerful and
unique tool for micro/nanostructure fabrication in the past
decade,1–4 which provides a unique opportunity for making
structures of almost any complexity due to its capability for
intrinsic three-dimensional (3D) processing. The fabrication of
various micromachines and microdevices, such as microgears,5,6
microactuators,7,8 microlens arrays,9 microemitters10 and 3D
photonic crystals,11,12 have been demonstrated with conventional
photoresists and functional polymers by TPIP for applications in
micro/nano-electro mechanical systems (M/NEMS).2–18
The most often used TPIP technique in the above-mentioned
studies is that of radical polymerization, in which the radicals are
first generated by the two-photon absorption (TPA) of photo-
initiators, that then, initiate radical polymerization as normal.
The resolution of TPIP is influenced by the radicals produced at
the focal spot of the laser as well as their diffusion in chain
growth radical polymerization. In the development of the
fabrication technique, decreasing laser power and exposure time
and using a large numerical aperture (NA) objective lens are very
useful in confining the two-photon absorption region.24,25
Recently, Li et al. successfully realized the control of two-photon
induced radicals region by a photo-induced deactivation tech-
nique.23 Reducing the radical diffusion region in the polymeri-
zation process is another route to improve the resolution of
TPIP. The deactivation of radicals in the polymerization of
photoresists mainly comes from radical termination by the
combination of radicals and quenching molecules. TPIP fabri-
cation regions were well confined by adding radical quenchers to
a resin SCR500.19,26 In that case, the radical quencher can
deactivate the photoinduced radical around the focused spot,
which results in chain reaction termination and prevents polymer
growth. However, in this reported approach a large amount of
radical quencher, compared to the amount of photoinitiator
As a potential technique of advanced photolithography,
improving the resolution of TPIP is desirable. After Kawata et al.
demonstrated TPIP fabrication results beyond the diffraction
limit,3 a resolution of 120 nm with a laser wavelength of 780 nm,
many efforts have been made to improve the resolution further
aLaboratory of Organic NanoPhotonics and Key Laboratory of Functional
Crystals and Laser Technology, Technical Institute of Physics and
Chemistry, Chinese Academy of Sciences, No.2, Beiyitiao,
Zhongguancun, Haidian District, Beijing, 100190, China. E-mail:
chenwq7315@mail.ipc.ac.cn; xmduan@mail.ipc.ac.cn; Fax: +86-10-
82543597; Tel: +86-10-82543596
bGraduate School of Chinese Academy of Sciences, No.2, Beiyitiao,
Zhongguancun Road, Haidian District, Beijing, 100190, China
5650 | J. Mater. Chem., 2011, 21, 5650–5659
This journal is ª The Royal Society of Chemistry 2011

View Article Online
used, was needed to obtain a significant confining effect.
Consequently, photoinitiators with the ability of radical-diffu-
sion control are desirable in TPIP nanofabrication since the
development of this technique easily reaches its limit.
2. Experimental part
2.1 Materials
2-Methyl-3-butyn-2-ol was bought from Acros Organics.
Dimethylformamide (DMF), sodium hydride (60% in mineral
oil, NaH), 9-H-carbazole, sodium hydroxide (NaOH), copper
iodide (CuI), toluene, tetrahydrofuran (THF), triethylamine
(Et3N), triphenylphosphine (PPh3), 4-nitrobromobenzene,
phenyl methyl ether (PhOMe), p-methoxy benzyl alcohol,
methylacrylic acid (MAA) and thionyl chloride were purchased
from the Beijing Chemical Reagent Company. Palladium bis
(triphenylphosphine) dichloride [Pd(PPh3)2Cl2],38 3,6-dibromo-
9H-carbazole,39 p-methoxybenzyl chloride39 and BNBC36 were
synthesized using literature procedures. Toluene and THF were
dried over Na and freshly distilled before use. Et3N was dried
over calcium hydride and freshly distilled before use. Other
chemicals and solvents were used as received without further
purification. Pentaerythritol triacrylate (PE-3A) was obtained
from Kyoeisha Chemical Co. Ltd., Japan.
An intramolecular radical-quench reaction can happen more
rapidly and efficiently than an intermolecular one, as the
quenching of the radicals is controlled by local molecules.27,28 An
intramolecular radical termination reaction may be more effi-
cient and result in a well-confined polymerization range. We
suppose that the shorter distance between monomer radicals and
the radical quenching group will bring more quenching chances if
the radical quenching group is introduced into a highly efficient
photoinitiator. Consequently, the more radical termination
processes, the smaller the residual feature sizes.
Developing highly efficient TPIP photoinitiators by opti-
mizing the traditional UV photoinitiator system29,30 and
designing novel TPIP photoinitiators31–37 is necessary to improve
the sensitivity of resins, which is also benefit from control of the
excitation region by using lower laser power. In our previous
studies,34–36 we designed and synthesized a series of TPIP pho-
toinitiators with C2V symmetrical molecular structure. These
photoinitiators possessed large two-photon absorption cross-
sections and were successfully used in TPIP 3D microfabrication
with high initiating efficiency. Thus, the molecules combined
with a high efficiency TPIP photoinitiator and radical quenching
group can be expected to play an important role in TPIP
microfabrication with high resolution through confining radical
diffusion by deactivating radicals in a tiny region. In the present
work, we designed and synthesized a C2V symmetrical molecule,
3,6-bis[2-(4-nitrophenyl)-ethynyl]-9-(4-methoxy-benzyl)-carba-
zole (BNMBC, Scheme 1) in which an intramolecular charge
transfer (ICT) conjugate system was introduced for high TPA
cross section and p-methoxybenzyl group for radical quenching.
We investigated the optical properties and photoinitiating
properties of BNMBC in comparison to its analogous, 3,6-bis-
[2-(4-nitrophenyl)-ethynyl]-9-benzyl-carbazole (BNBC, Scheme
1), which was reported as a highly efficient TPIP photoinitiator.36
The evaluation with Z-scan method revealed that BNMBC had
a large TPA cross-section. The TPIP fabrication experiments
showed that the volumes of fabricated suspended fibers were
significantly decreased when BNMBC was used as photoinitiator
compared to the comparable photoinitiating systems studied
here. It was indicated that the introduction of the p-methoxy-
benzyl group into BNMBC led to a radical quenching effect in
TPIP. Meanwhile, a fine 3D microstructure with high TPIP
resolution was fabricated using the photoresist with BNMBC as
the photoinitiator. Finally, a mechanism of the confining effect
on radical diffusion in TPIP was proposed for different photo-
initiating systems.
2.2 Synthesis
Chromophore BNMBC was synthesized in a facile way by
Sonogashira coupling reactions40 as shown in Scheme 2.
Synthesis of A. To a 100 mL three-necked flask, 3,6-dibromo-
9H-carbazole (6.50 g, 20 mmol), CuI (191 mg, 1.0 mmol),
PdCl2(PPh3)2 (350 mg, 0.50 mmol), PPh3 (262 mg, 1.0 mmol),
2-methyl-3-butyn-2-ol (3.36 g, 40 mmol) and triethyl amine
(60 mL) were added and bubbled with N2 for 30 min. The
resulting mixture was allowed to reflux for 24 h. After cooling,
the solvent was removed. The residue was poured into water and
extracted with ethyl acetate (200 mL ꢀ 3). The organic layers
were combined, dried over MgSO4, and concentrated. To the
residue was added toluene (150 mL) and filtration gave a pale
ꢁ
1
yellow powder (5.10 g, 77% yield). Mp. 194–196 C. H NMR
(400 MHz, CDCl3, d ppm): 1.68 (s, 12 H), 2.11 (s, 2 H), 7.35 (d, 2
H, J ¼ 8.4 Hz), 7.49 (d, 2 H, J ¼ 8.4 Hz), 8.12 (s, 2 H), 8.21 (s, 1 H).
Synthesis of B. To a solution of 3,6-bis(3,3-dimethyl-3-
hydroxy-1-butynyl)-9H-carbazole (3.31 g, 10 mmol) in toluene
(140 mL), was added sodium hydroxide (4.80 g, 120 mmol). Then
the resulting mixture was refluxed for 6 h. After filtration and
concentration, the brown residue was purified by recrystalliza-
tion using ethyl acetate–petroleum ether to give a light yellow
Scheme 1 The chemical structure of photoinitiators.
This journal is ª The Royal Society of Chemistry 2011
Scheme 2 Synthesis of BNMBC.
J. Mater. Chem., 2011, 21, 5650–5659 | 5651

View Article Online
ꢀ
ꢁꢀ
ꢁꢀ ꢁꢀ
ꢁ
1
powder (1.34 g, 62% yield). Mp. 212–214 ꢁC. H NMR (400
ArðlrÞ IrðlrÞ n2s
Fs
Fr
Fs ¼ Fr
(1)
AsðlsÞ IsðlsÞ n2r
MHz, CDCl3, d ppm): 3.08 (s, 2 H), 7.37 (d, 2 H, J ¼ 8.4 Hz), 7.57
(d, 2 H, J ¼ 8.4 Hz). 8.21 (s, 2 H), 8.25 (s, 1 H).
where F is the quantum yield, n is the refractive index, I(l) is the
relative intensity of exciting light at wavelength l, and A(l) is the
absorbance of solution at the exciting wavelength l, and F is
the integrated area under the emission spectrum. Subscripts s and
r refer to the sample and reference solution, respectively.
Two-photon excited fluorescence (TPEF) spectra were
recorded on SD2000 spectrometer (Ocean Optical) with exci-
tation with a femtosecond laser (Tsunami, Spectra-Physics) with
100-fs pulse width and 82 MHz repetition rate. TPA cross
sections were determined by the TPEF method.43 It is assumed
that the quantum efficiencies after two-photon excitation are the
same as those after one-photon excitation. The TPA cross
sections are obtained by calibration against fluorescein with
known Fd value in aqueous NaOH solution (pH ¼ 11) at
concentrations of 1.0 ꢀ 10ꢂ4 M. The samples were dissolved in
solvents at concentrations of about 1.0 ꢀ 10ꢂ4 M. The error in
TPEF measurements is about 15%. To ensure that the measured
signals were solely due to TPA, the dependence of TPEF on the
incident intensity was verified in each case to be quadratic. Then
the TPA cross sections d were calculated on the basis of the
following expression:
Synthesis of C. To a 100 mL three-necked flask, 3,6-diethynyl-
9H-carbazole (430 mg, 2.0 mmol), CuI (19.0 mg, 0.1 mmol),
PdCl2(PPh3)2 (35.0 mg, 0.05 mmol), PPh3 (26.2 mg, 0.1 mmol), 4-
nitrobromobenzene (1212 mg, 6.0 mmol) and triethyl amine
(60 mL) were added and bubbled with N2 for 30 min. The
resulting mixture was allowed to reflux for 24 h. After cooling,
the solvent was removed and recrystallized from ethyl acetate to
give a dark yellow powder (795 mg, 87% yield). Mp. >300 ꢁC.1H
NMR (400 MHz, CDCl3, d ppm): 7.48 (d, 2 H, J ¼ 8.4 Hz), 7.67
(d, 2 H, J ¼ 8.4 Hz), 7.71 (d, 4 H, J ¼ 8.8 Hz), 8.26 (d, 4 H, J ¼
8.8 Hz), 8.32 (s, 2 H), 8.36 (s, 1 H).
Synthesis of BNMBC. To a flask containing NaH (60% in
mineral oil, 24.0 mg, 1.0 mmol) in DMF (10 mL), a solution of
3,6-bis[2-(4-nitrophenyl)-ethynyl]-9H-carbazole
(229
mg,
0.5 mmol) in DMF (25 mL) was added dropwise; after one hour,
p-methoxy benzyl chloride (94 mg, 0.6 mmol) was added slowly.
The resulting suspension was stirred for another 6 h and carefully
poured into water. After filtration and recrystallization using
ethyl acetate–petroleum ether, 3,6-bis[2-(4-nitrophenyl)-
ethynyl]-9-(4-methoxybenzyl)-carbazole
(BNMBC)
was
Cr nr Fs Fr
obtained as a yellow solid (191 mg, yield: 66%). Mp. 274–275
ꢁC.1H NMR (400 MHz, CDCl3, d ppm): 3.76 (s, 3 H), 5.50 (s, 2
H), 6.82 (d, 2 H, J ¼ 8.7 Hz), 7.08 (d, 2 H, J ¼ 8.6 Hz), 7.41 (d, 2
H, J ¼ 8.5 Hz), 7.64 (dd, 2 H, J1 ¼ 8.5 Hz, J2 ¼ 1.3 Hz), 7.70 (d, 4
H, J ¼ 8.8 Hz), 8.25 (d, 4 H, J ¼ 8.8 Hz), 8.35 (d, 2 H, J ¼ 1.3
Hz),.13C NMR (100 MHz, CDCl3, ppm): 46.6, 55.4, 86.9, 96.3,
109.8, 113.5, 113.9, 114.5, 122.9, 123.8, 124.8, 127.8, 128.1, 129.6,
130.5, 130.9, 132.2, 141.3, 146.9,159.5. FT-IR (KBr, cmꢂ1): 2918,
2203, 1587, 1512, 1365, 1106, 853, 748. HR-MS (C36H23N3O5):
calcd: 577.1638; obsd: 577.1699.
ds ¼ dr
(2)
Cs ns Fr Fs
where d is TPA cross section, C and n are the concentration and
refractive index of the sample solution, and F is the integrated
area under the TPEF spectrum.
The TPA cross sections at 780 nm were further measured by
open aperture Z-scan method.44,45 A concentration of 2.5 ꢀ 10ꢂ3
M in THF solution for the chromophores was used. The laser
beam was divided into two parts. One was used as the intensity
reference and monitored by a 5020PE laser power and energy
meter (Genetec.). The other was used for transmittance
measurement, the laser beam was focused by passing through
a lens (f ¼ 10 cm) and a quartz cell with a sample thickness of
1 mm. The position of the sample cell could be varied along the
laser-beam direction (z-axis), so the local power density within
the sample cell could be changed under a constant laser power
level. The transmitted laser beam from the sample cell was then
detected by the same power meter as used for reference moni-
toring. The on-axis peak intensity of the incident pulses at the
focal point, I0, ranged from 40 to 60 GW cmꢂ2. Assuming
a Gaussian beam profile, the nonlinear absorption coefficient
b can be obtained by curve fitting to the observed open-aperture
traces with the eqn (3).
2.3 Measurements
1H NMR and 13C NMR spectra were recorded on a Varian
Jemini-300/Brucker AV 400 spectrometer using CDCl3 as
a solvent and all shifts are referenced to tetramethylsilane (TMS).
The chemical shift (s ¼ singlet, d ¼ doublet, t ¼ triplet, m ¼
multiplet) is shown in ppm. Melting points were measured on an
X-4 microscopic melting-point apparatus produced by Beijing
Focus Instrument Company. High resolution mass spectra
(HRMS) were measured on Microflex (Bruker Germany).
Infrared spectra were recorded on an Excalibur 3100 (Varian,
America).
À
Á
UV–Visible spectra were recorded at the concentration of 1 ꢀ
10ꢂ5 M in THF on a UV 3100PC Scanning Spectrophotometer
(Shimadzu, Japan) with slit widths of 5 nm. One-photon excited
fluorescence (OPEF) spectra were recorded at the concentrations
of 2.5 ꢀ 10ꢂ6 M in THF with both the excited and emission slit
widths of 5 nm on an RF5300PC spectrofluorophotometer
(Shimadzu, Japan). The fluorescence quantum yields were
obtained with quinine sulfate dihydrate in 0.05 M sulfuric
acid aqueous solution as a reference standard (Fr ¼ 0.51) by
eqn (1).41,42
bI0 1 ꢂ eꢂa l
0
ꢂ
ꢃ
TðzÞ ¼ 1 ꢂ
(3)
2
2a0 1 þ ðz=z0Þ
where a0 is the linear absorption coefficient, l is the sample length
and z0 is the diffraction length of the incident beam. We obtained
the nonlinear absorption coefficient b at the corresponding
wavelength, where linear absorption is negligible and the
condition of a0l ꢃ 1 is satisfied. When a0l ꢃ 1, eqn (3) can be
simplified as eqn (4).
5652 | J. Mater. Chem., 2011, 21, 5650–5659
This journal is ª The Royal Society of Chemistry 2011

View Article Online
wavelength of 780 nm, a pulse width of 100 fs, and a repetition
rate of 82 MHz. The laser beam was tightly focused by a 100 ꢀ
oil-immersion objective lens with a high numerical aperture (N.
A. ¼ 1.45, Olympus). We applied the laser power around 5 mW
(measured on the front of the objective lens) which is equal to
a peak power density of laser focus about 59 GW cmꢂ2. The
fabrication process began after the resin was drop-casted onto
a cover glass substrate for 30 min. The laser beam was focused
into the liquid photopolymerisable resin which was placed on the
cover glass above a 3D piezostage controlled by a computer. The
TPIP cured suspended lines were fabricated by scanning the laser
beam between two parallel rectangular supports with a space of
2.3 mm and a height of 3 mm on the glass substrate. The final
structures were obtained after washing to remove the unpoly-
merized resins with ethanol. The images of the fabricated struc-
tures were recorded by a field-emission scanning electron
microscope (SEM, S-4300, Hitachi, Japan). The width and
height of the suspended lines were measured from the top and 45ꢁ
view from SEM images, respectively. The polymerization volume
VTPIP of fabricated suspended lines were calculated using eqn (7).
bI0l
ꢂ
ꢃ
TðzÞ ¼ 1 ꢂ
(4)
2
2 1 þ ðz=z0Þ
The TPA cross sectionsd of BNMBC and BNBC (in units of
1 GM ¼ 10ꢂ50 cm4s/photon) can be determined by eqn (5):
dNAd ꢀ 10ꢂ3
b ¼
(5)
hn
where NA is the Avogadro constant, d is the concentration of
compounds in solution, h is the Planck constant, and n is the
frequency of the incident laser beam.
2.4 Photopolymerization
2.4.1 Sample preparation of photoresists. The photoresists
were prepared by mixing appropriate amounts of methyl acrylic
acid as the monomer, PE-3A as the cross-linker and one of three
photoinitiating systems, BNBC, BNBC/PhOMe (molar ratio
1 : 1) and BNMBC. The components of the photoresists with the
three different photoinitiating systems were listed in Table 1. The
photoresists were designed to investigate the effect of radical
diffusion by the three photoinitiating systems.
1
VTPIP
¼
lwL
(7)
4
2.4.2 One-photon induced photopolymerization. A Fourier
transform real-time infrared spectrometer (Nicolet 5700 FT-IR)
adapted with a horizontal sample holder was used to monitor the
one-photon induced photopolymerization (OPIP) properties of
the photoresists. The sample was coated on a KBr disk and
placed in the sample holder, whereupon the sample was simul-
taneously exposed to a UV-light source (Rolence-100 UV) and
an IR analyzing light beam. The light intensity on the sample was
Where l and w were the height and width of the fiber, respectively,
and L was the length of the suspended fibers, here L ¼ 2.3 mm.
3
Results and discussion
3.1 Synthesis
The synthesis of BNMBC used 3,6-dibromo-9H-carbazole as
a starting materials which was obtained by the bromination of
carbazole as shown in Scheme 2. After the first Sonogashira
reaction, we introduced a triple bond at the 3,6-position in the
carbazole and obtained compound A in 77% yield by simple
filtration, which easily precipitated from cool toluene. The diol A
was treated with NaOH in toluene and converted into diyne B in
62% yield. A further coupling reaction between the diyne and p-
bromonitrobenzene gave compound C in 87% yield by recrys-
tallization from ethyl acetate. Compound C was converted into
its sodium salt and was finally reacted with p-methoxy benzyl
chloride to give BNMBC in 66% yield. In this route, the products
were purified only by recrystallization and filtration instead of
flash chromatography. BNBC was synthesized according to our
previous study.
206 mW cmꢂ2 measured by a Honle UV meter (Germany). The
€
IR spectrometer must be set in the absorbance mode and the
detection wavelength fixed at a value where the double bond
exhibited a discrete and intense absorption from 819 to 838 cmꢂ1
for the CH2¼CH twisting mode.46 The degree of conversion
(DC) can be expressed by eqn (6).
À
Á
A0 ꢂ At
DC % ¼
ꢀ 100
(6)
A0
Where A0 is the initial peak area before irradiation and At is the
peak area of the double bond at time t.
2.4.3 Two photon induced photopolymerization. We per-
formed the TPIP experiments by employing a mode-locked Ti :
Sapphire laser system (Tsunami, Spectra-Physics) with a central
3.2 One-photon absorption and fluorescence
As depicted in Fig. 1a, both of BNBC and BNMBC in THF had
two absorption bands in the UV-Vis spectra. The absorption
band at 384 nm was the result of the electronic transition from
the ground state to the ICT excited state, and the band at 290 nm
assigned to the typical localized p–p* transition. Besides these
two bands, there was a shoulder peak at 376 nm, which
contributed to by the coupling of two ICT branches. Beyond
475 nm, there was no absorption. Their OPEF peaks appeared at
546 nm with the excitation wavelength of 360 nm. The fluores-
cence quantum yield for BNMBC was confirmed as 0.17 and for
BNBC as 0.077. The photophysical data of BNBC and BNMBC
are summarized in Table 2.
Table 1 Components of resins used in photopolymerizaiton experiment
Resin component (wt%)
Resins Photoinitiator
MMA Photointiator PE-3A M.R.a (%)
R1
R2
R3
R4
R5
R6
BNBC 80.02 0.47
BNBC/PhOMe (1 : 1) 79.95 0.57
19.51 0.1
19.49 0.1
BNMBC
BNBC
80
80
0.5
0.1
0.1
0.1
19.5
19.9
19.9
19.9
0.1
0.02
0.02
0.02
BNBC/PhOMe(1 : 1) 80
BNMBC 80
a
Molar ratio of the photoinitiator.
This journal is ª The Royal Society of Chemistry 2011
J. Mater. Chem., 2011, 21, 5650–5659 | 5653

View Article Online
Using the TPEF method, TPA spectra were obtained as shown in
Fig. 1b. Both of them showed a maximum TPA cross section (d),
at 770 nm, 500 GM for BNBC and 262 GM for BNMBC.
Although the TPA cross section of BNBC was about twice as
that of BNMBC, the two-photon action cross section (Fd) of the
two compounds were very close, 38.5 GM for BNBC and 44.5
GM for BNMBC, considering about 15% experimental error.
Consequently, the big difference in TPA cross section of the two
compounds mainly came from the difference of their fluorescence
quantum yields. In order to decrease the error from the fluo-
rescence quantum yield, we further investigated the TPA cross-
sections of BNMBC and BNBC using the Z-scan method at
a wavelength of 780 nm, which was used in our TPIP experi-
ments. The obtained Z-scan experimental data of BNBC and
BNMBC is shown in Fig. 1c. After data fitting, we obtained the
nonlinear absorption coefficients, b, which measured 0.1350 GW
cmꢂ2 for BNBC and 0.1398 GW cmꢂ2 for BNMBC. The TPA
cross sections (d) of BNBC and BNMBC were obtained as 2285
GM and 2367 GM as shown in Table 2, which indicates that the
two compounds possessed large TPA properties.
3.4 One-photon induced photopolymerization
To confirm the photopolymerization of the designed photo-
initiating systems, a real-time near FT-IR technique (RT-IR) was
employed to evaluate the efficiency of the photoinitiators of the
three designed photoresists in one-photon induced photo-
polymerization (OPIP).46,47 We prepared the photoresists R1, R2
and R3 in which the molar ratios of photoinitiators are 0.1% (as
shown in Table 1), of which only R1 included the photoinitiator
BNBC, R2 had BNBC with PhOMe as the radical quencher, and
R3 used the photoinitiator with radical quenching moiety
BNMBC. The recorded RT-IR polymerization profiles of the
photoresists are shown in Fig. 2. The DC of R1 increased sharply
in the time range of 1.5–3 min and finally reached to 80%, which
indicated a self-accelerated polymerization process. However,
the DC of R2 and R3 in the same time range increased almost
linearly, such that the final conversion of R2 and R3 were 70%
and 65%, respectively. The magnified polymerization profiles of
times up to 2 min are shown in the inset of Fig. 2. The DC value
of R3 was very similar to that of R2 and a little higher than R1
after 1.5 min. In the case of R1, the small amount of oxygen
dissolved in the photoresist quenched the photo-generated
radicals at the initiating period, which was similar to normal
radical polymerization. The additional quenchers or quenching
groups in the photoinitiating systems of R2 and R3 showed an
average radical quenching effect which led to the linear
increase of DC. The introduction of the p-methoxybenzyl moiety
into the photoinitiator changed the photopolymerization from
Fig. 1 (a) Normalized absorption (black), one-photon induced fluores-
cence spectra (red, excited at 380 nm) and two-photon fluorescence
spectra (blue, excited at 780 nm) of BNBC and BNMBC. (b) TPA spectra
of BNBC and BNMBC in the wavelength range of 740 to 870 nm. (c)
Open-aperture Z-scan traces of BNBC and BNMBC at a concentration
of 2.5 mM in THF. The solid lines are the best fitted curves of experi-
mental data.
3.3 TPEF and TPA spectra
The TPEF spectra of two compounds excited at 780 nm are
shown in Fig. 1a, which exhibited a maximum at 545 nm. The
TPEF spectra were in accordance with their OPEF spectra.
Table 2 Optical properties of BNBC and BNMBC in THF at room temperature
3b (Mꢂ1 cmꢂ1
)
lemc (nm)
De (cmꢂ1
)
F f
dh (GM)
d
lem (nm)
g
dmax (GM)
Compounds
labsa (nm)
BNBC
BNMBC
383
384
4.23 ꢀ 104
3.23 ꢀ 104
546
546
545
545
7794
7726
0.077
0.17
500
262
2285
2367
a
The wavelength of absorption maximum. b The extinction coefficient. c The wavelength of one-photon emission maximum. d The wavelength of two-
photon emission maximum. Stokes shift. Fluorescence quantum yield. The maximum TPA cross section at around 770 nm measured by TPEF.
e
f
g
h
The TPA cross section measured by open-aperture Z-scan method at 780 nm, 1 GM ¼ 10ꢂ50 cm4 s photonꢂ1
.
5654 | J. Mater. Chem., 2011, 21, 5650–5659
This journal is ª The Royal Society of Chemistry 2011

View Article Online
Fig. 2 Polymerization profiles of resins R1, R2 and R3 under UV light
irradiation (I0 ¼ 206 mW cmꢂ2) monitored by real-time FT-IR. Inset
showed the magnified polymerization profiles up to 2 min.
a self- accelerated polymerization process to a linear increment
process. The results revealed that the photoinitiating capability
of BNMBC was very similar to that of BNBC/PhOMe and a little
higher than BNBC after 1.2 min. As TPIP microfabrication is
usually performed in a short irradiation time range, BNMBC
with its high photoinitiating efficiency as well as its radicals-
quenching ability could be expected to act as confining radical
diffusion in TPIP.
3.5 Two-photon induced photopolymerization
The results of OPIP promoted us to investigate if the confining
radical effect of BNMBC existed in TPIP. We performed the
TPIP by fabricating the suspended fibers of R1, R2 and R3 at
a laser power of 3.67 mW, and then measured the sizes of the
fibers. Their relevant SEM images are shown in Fig. 3a. The
TPIP fabricated fibers thinned down when the laser scan speed
was accelerated along the direction indicated by the arrow. The
height and width of the suspended fibers were plotted against
exposure time for R1, R2 and R3 in Fig. 3b. Each fiber hada
larger height than width because of the elliptical laser focal spot.
The height for R1 increased from 720 to 949 nm and the width
from 296 to 466 nm; while R2 was smaller than R1 in the same
conditions. The sizes for R3 decreased further and ranged from
244 to 572 nm for height and from 118 to 328 nm for width. The
plots of their VTPIP upon exposure time were summarized in
Fig. 3c. The VTPIP of fibers for R1–R3 at the same exposure time
ranks in the order of R1 > R2 > R3. The VTPIP values increased
from 0.283 mm3 at an exposure time of 6.3 ms to 0.541 mm3 at
38 ms for R1. When PhOMe was added, the VTPIP of the fibers
for R2 significantly decreased to 0.175 and 0.413 mm3 at 6.3 and
38 ms, respectively. Moreover, the volume of fibers fabricated
from R3 in which BNMBC was used as the photoinitiator,
decreased further to 0.052 and 0.240 mm3 at 6.3 and 38 ms,
respectively. In order to give a clear comparison, we calculated
the relative polymerization volumes of R2 and R3 compared to
R1, the values are approximate 0.7 for R2 to R1 and 0.3 for R3 to
R1 (Fig. 3d). Just as the results from OPIP, the photoresists R2
and R3, with radical quenchers and/or quenching groups
increased the chances of a radical termination reaction, such that
the volume of photopolymerization was confined to a smaller
Fig. 3 (a) SEM images of suspended fibers for R1, R2 and R3 fabricated
with a laser power of 3.67 mW. The images on the left for the top view
and the images on right for the 45ꢁ view. The laser scan speed was
accelerated along the arrow direction is shown. Scale bars are 2 mm. (b)
Plots of sizes of suspended fibers for R1, R2 and R3 as the function of the
exposure time. (c) VTPIP versus the exposure time. (d) Relative VTPIP
versus exposure time.
range. Moreover, photoinitiator BNMBC exhibited a more
efficient radical quenching effect than the BNBC/PhOMe system.
The p-methoxybenzyl moiety in BNMBC resulted in intra-
molecular collision with radicals in R3 during the fabrication,
This journal is ª The Royal Society of Chemistry 2011
J. Mater. Chem., 2011, 21, 5650–5659 | 5655

View Article Online
while PhOMe in R2 contributed to intermolecular quenching of
radicals. Since the intramolecular radical-quench reaction would
be more efficient than an intermolecular one, the radicals were
quenched faster in R3 than in R2 during TPIP fabrication.
We further decreased the molar ratio of photoinitiator to
0.01% in the resins (R4–R6 in Table 1) to study the radical
quenching effects. The suspended fibers of R4–R6 were fabricated
with a laser power of 5.45 mW by increasing the laser scan speed.
The arrow direction in Fig. 4a indicates the scan speed from low
to high. The sizes of fabricated fibers for R6 were significant
reduced as compared to R4 and R5 (Fig. 4b). The height for R6
increased from 283 to 400 nm and width from 183 to 236 nm;
while for R5 the height increased to 516 to 752 nm and width
ranged from 165 to 274 nm; for R4 the height and width further
increased to 1129 and 329 nm at exposure time of 38 ms,
respectively. As depicted in Fig. 4c, the VTPIP of fibers for R6
increased slowly from 0.070 to 0.170 mm3 with prolonged expo-
sure time, which were much less than those of R4 and R5. The
VTPIP relative to R4 were about 0.5 for R5 and 0.2 for R6(as
shown in Fig. 4d). Furthermore, as depicted in Fig. 4e, the aspect
ratio (l/w) of fibers for R4 ranged in 3.34–4.18; those for R5 was
less at 2.71–3.16; while the l/w for R6 was even lower at 1.54–2.08
with an average value of 1.80. The usage of BNMBC improved
not only the size of fabricated fibers (smaller fibres were
Fig. 4 (a) SEM images of suspended fibers for R4, R5 and R6 fabricated with a laser power of 5.45 mW. The images in left are for the top view and the
images in right are for the 45ꢁ view. Laser scan speed accelerated along the arrow direction as shown. Scale bars: 2 mm. (b) Plots of sizes of suspended
fibers for R4, R5 and R6 as a function of the exposure time. (c) VTPIP versus the exposure time. (d) Relative VTPIP versus exposure time. (e) The plot of
aspect ratio (l/w) versus exposure time.
5656 | J. Mater. Chem., 2011, 21, 5650–5659
This journal is ª The Royal Society of Chemistry 2011

View Article Online
obtained), but also the lower aspect ratio. The results provided
further evidence that the addition of BNMBC was effective to
confine the radical diffusion in polymerization due to the radical
quenching effect of the p-methoxybenzyl moiety. It was also
demonstrated that the intramolecular radical quenching became
more effective when the molar ratio of photoinitiator decreased,
which resulted in the formation of finer polymer features.
Besides, we investigated the polymerization threshold power
(Pth) for R4–R6. Pth was defined as the lowest laser energy density
before the objective lens with which the polymer fiber can be
fabricated at a scan speed of 10 mm sꢂ1. The laser power
decreased from 9.38 to 4.02 mW as the arrow direction in Fig. 5
indicates. The Pth for R4, R5 and R6 were obtained as 4.02, 4.28
and 4.53 mW, respectively, which were equal to the laser focus
peak power density of 48, 51, and 54 GW cmꢂ2 after attenuation
by the object lens, respectively. With the extra methoxy group,
BNMBC exhibited a very similar effect on the photoinitiating
process in comparison to BNBC. This was in accordance with the
small difference between their two-photon absorption cross
sections. Therefore, BNMBC showed the effects of not only
excitation-region control due to its high photoinitiating effi-
ciency but also the radical-diffusion control due to its p-
methoxybenzyl moiety.
Fig. 6 Schematic illustration of the mechanism on confining radical
diffusion in TPIP using different photoinitiators: BNBC, BNBC/PhOMe
(molar ratio 1 : 1) and BNMBC.
because the radical quencher, PhOMe, was independently
dispersed in the photoresist, such that the radical diffusion
distances would be limited due to the deactivation of radicals by
quenchers, which resulted in the resolution improvement of
TPIP. Further, when the photoinitiator with quenching moiety,
BNMBC, was introduced into photoresist, higher efficiencies of
radicals quenching and chain terminating could be obtained due
to the higher chance of collision between radicals and quenching
group of BNMBC in shorter distances than that using BNBC/
PhOMe, which led to higher TPIP resolution. As a result, the
smallest voxel size was obtained for the photoresist with
BNMBC as photoinitiator out of the three photoinitiating
system studied here.
In order to provide a better understanding of the confining
effect on radical diffusion in TPIP with the different photo-
initiating systems used in this study, a schematic mechanism is
illustrated in Fig. 6. The density of two-photon induced radicals
was distributed depending on the intensity of the laser focused
beam when the photoresist was exposed to the laser. Then, the
radicals initiated the TPIP to form the polymeric 3D networks
via the crosslinking reaction between monomers and cross-
linkers, which led to cured polymers formation. In the normal
photoinitiating system such as that used in BNBC, the radicals
were terminated by their combination during the chain growth
reactions after some average distance of radical diffusion.
However, in the photoinitiating system of BNBC/PhOMe,
Additionally, we demonstrated the fabrication of arbitrary
complex 3D structures using the photoresist with our designed
photoinitiator BNMBC. Fig. 7 shows SEM images of the sample
of a fly fabricated using R6 with a laser power of 7.70 mW and
a scan speed of 66 mm sꢂ1. The fly is 6.47 mm in hieght, 20.90 mm
Fig. 5 SEM images of suspended fibers for R4, R5 and R6 (top view)
fabricated with reducing laser power along the arrow direction (the
detailed values were: 9.38, 8.96, 8.54, 8.13, 7.73, 7.34, 6.99, 6.63, 6.33, 6.02
5.71, 5.45, 5.22, 4.98, 4.77, 4.53, 4.28, 4.02 mW) under a laser scan speed
Fig. 7 SEM images of a fly fabricated with a laser power of 7.70 mW and
a scan speed of 66 mm sꢂ1 on the photoresist R6. Scale bars is 5mm.
of 10 mm sꢂ1
.
This journal is ª The Royal Society of Chemistry 2011
J. Mater. Chem., 2011, 21, 5650–5659 | 5657

View Article Online
10 C. F. Li, X. Z. Dong, F. Jin, W. Jin, W. Q. Chen, Z. S. Zhao and
X. M. Duan, Appl. Phys. A: Mater. Sci. Process., 2007, 89, 145.
11 K. Kaneko, H. B. Sun, X. M. Duan and S. Kawata, Appl. Phys. Lett.,
2003, 83, 2091.
12 X.-Z. Dong, Q. Ya, X.-Z. Sheng, Z.-Y. Li, Z.-S. Zhao and Xuan-
Ming Duan, Appl. Phys. Lett., 2008, 92, 231103.
13 Q. Ya, X.-Z. Dong, W.-Q. Chen, S. Shoji, S. Nakanishi, X.-M. Duan
and Satoshi Kawata, Appl. Phys. A: Mater. Sci. Process., 2008, 93,
393.
14 W. Zhou, S. M. Kuebler, K. L. Braun, T. Yu, J. K. Cammack,
C. K. Ober, J. W. Perry and S. R. Marder, Science, 2002, 296, 1106.
15 J. Serbin, A. Egbert, A. Ostendorf, B. N. Chichkov, R. Houbertz,
in long and 17.61 mm in wide was fully supported by six legs with
the diameter of about 650 nm. It was indicated that the radical
quenching effect of BNMBC did not decrease the cured polymer
strength for a photoresist with high concentration of crosslinker,
which could form high crosslinking 3D polymer networks. The
fine striates on the wings of the fly were clearly reproduced at
a scale smaller than 100 nm by TPIP on the photoresist con-
taining our novel photoinitiator, BNMBC. The high resolution
was contributed to by the confining radical diffusion of BNMBC
in TPIP. Thus, the photoresist consisting of the photoinitiator
with radical quenching moiety and the appropriate photoresist
components would lead to high TPIP resolution and fine struc-
ture fabrication by confining radical diffusion.
€
G. Domann, J. Schulz, C. Cronauer, L. Frohlich and M. Popall,
Opt. Lett., 2003, 28, 301.
16 Z. B. Sun, X. Z. Dong, S. Nakanishi, W. Q. Chen, X. M. Duan and
S. Kawata, Appl. Phys. A: Mater. Sci. Process., 2007, 86(4),
427–431.
17 Z. B. Sun, X. Z. Dong, W. Q. Chen, S. Nakanishi, X. M. Duan and
S. Kawata, Adv. Mater., 2008, 20, 914.
18 M. P. Joshi, H. P. Pudavar, J. Swiatkiewicz, P. N. Prasad and
B. A. Reianhardt, Appl. Phys. Lett., 1999, 74, 170.
19 K. Takada, H.-B. Sun and S. Kawata, Appl. Phys. Lett., 2005, 86,
071122.
20 J.-F. Xing, X.-Z. Dong, W.-Q. Chen, X.-M. Duan, N. Takeyasu,
T. Tanaka and Satoshi Kawata, Appl. Phys. Lett., 2007, 90, 131106.
21 X. Z. Dong, Z. S. Zhao and X. M. Duan, Appl. Phys. Lett., 2008, 92,
091113.
22 D. F. Tan, Y. Li, F. J. Qi, H. Y. Yang, Q. H. Gong, X. Z. Dong and
X. M. Duan, Appl. Phys. Lett., 2007, 90, 071106.
23 L. J. Li, R. R. Gattass, E. H. Hwang and J. T. Tourkas, Science, 2009,
324, 910.
24 K. S. Lee, D. Y. Yang, S. H. Park and R. H. Kim, Polym. Adv.
Technol., 2006, 17, 72.
25 H. B. Sun, K. Takada and S. Kawata, Appl. Phys. Lett., 2003, 83,
1104.
26 S. H. Park, T. W. Lim, D. Y. Yang, R. H. Kim and K. S. Lee,
Macromol. Res., 2006, 14, 559.
Conclusion
In conclusions, we designed and synthesized a novel two-photon
induced initiator with radical quenching moiety, BNMBC, and
successfully demonstrated its optical properties, high initiating
efficiency in two-photon induced polymerization, capability to
fabricate 3D structures with high resolution. BNMBC was
designed by introducing a p-methoxybenzyl moiety to the
reported TPIP photoinitiator BNBC as a radical quenching
moiety. BNMBC was confirmed to have a large two-photon
absorption cross-section of 2367 GM by Z-scan measurement,
which was similar to BNBC. The TPIP fabrication exhibited that
BNMBC possessed high TPIP initiating efficiency and effective
radical quenching effect. The volumes of polymer fibers fabri-
cated by TPIP of the photoresist using BNMBC as the photo-
initiator were decreased to 20%–30% of those using BNBC as the
photoinitiator. The introduction of a radical quenching group
into photoinitiator exhibited an effective confining effect of
radical diffusion compared to the addition of same molar ratio of
radical quencher. This kind of photoinitiator with radical
quencher should benefit the fabrication of precise structure.
ꢁ
ꢁ
27 E. Hideg, T. Kalai, P. B. Kos, K. Asada and K. Hideg, Photochem.
Photobiol., 2006, 82, 1211.
28 T. Tansakul, E. Lilie, E. D. Walter, F. Rivera III, A. Wolcott,
J. Z. Zhang, G. L. Millhauser and R. Braslau, J. Phys. Chem. C,
2010, 114, 7793.
29 P. J. Campagnola, D. M. Delguidice, G. A. Epling, K. D. Hoffacker,
A. R. Howell, J. D. Pitts and S. L. Goodman, Macromolecules, 2000,
33, 1511.
30 M. Farsari, G. Filippidis, K. Sambani, T. S. Drakakis and C. Fotakis,
J. Photochem. Photobiol., A, 2006, 181, 132.
31 B. A. Reinhardt, L. L. Brott, S. J. Clarson, A. G. Dillard, J. C. Bhatt,
R. Kannan, L. Yuan, G. S. He and P. N. Prasad, Chem. Mater., 1998,
10, 1863.
ꢁ
32 M. Albota, D. Beljonne, J. L. Bredas, J. E. Ehrlich, J. Y. Fu,
A. A. Heikal, S. E. Hess, T. Kogej, M. D. Levin, S. R. Marder,
Acknowledgements
The authors are grateful to the NSFC (Grant Nos. 50773091,
50973126, 60907019, 61077028), 973 (2010CB934103) and ICP
programs of MOST (2008DFA02050, S2010GR0980) for
financial support.
€
D. McCord-Maughon, J. W. Perry, H. Rockel, M. Rumi,
G. Subramaniam, W. W. Webb, X. L. Wu and C. Xu, Science,
1998, 281, 1653.
33 Y. Ren, X. Q. Yu, D. J. Zhang, D. Wang, M. L. Zhang, G. B. Xu,
X. Zhao, Y. P. Tian, Z. S. Shao and M. H. Jiang, J. Mater. Chem.,
2002, 12, 3431.
Notes and references
34 J.-F. Xing, W.-Q. Chen, X.-Z. Dong, T. Tanaka, X.-Y. Fang, X.-
M. Duan and Satoshi Kawata, J. Photochem. Photobiol., A, 2007,
189, 398.
35 J. Gu, Y.-L. Wang, W.-Q. Chen, X.-Z. Dong, X.-M. Duan and
S. Kawata, New J. Chem., 2007, 31, 63.
36 J. F. Xing, W. Q. Chen, J. Gu, X. Z. Dong, N. Takeyasu, T. Tanaka,
X. M. Duan and S. Kawata, J. Mater. Chem., 2007, 17, 1433.
37 D. Beljonne, W. Wenseleers, E. Zojer, Z. Shuai, H. Vogel,
1 S. Maruo, O. Nakamura and S. Kawata, Opt. Lett., 1997, 22, 132.
2 B. H. Cumpston, S. P. Ananthavel, S. Barlow, D. L. Dyer,
J. E. Ehrlich, L. L. Erskine, A. A. Heikal, S. M. Kuebler,
I. Y. S. Lee, D. McCord-Maughou, J. Qin, H. Rockel, M. Rumi,
X. L. Wu, S. R. Marder and J. W. Perry, Nature, 1999, 398, 51.
3 S. Kawata, H. B. Sun, T. Tanaka and K. Takada, Nature, 2001, 412,
697.
4 X.-M. Duan, H.-B. Sun and Satoshi Kawata, J. Photopolym. Sci.
Technol., 2004, 17, 393.
5 P. Galajda and P. Ormos, Appl. Phys. Lett., 2001, 78, 249.
6 X. Z. Dong, Z. S. Zhao and X. M. Duan, Appl. Phys. Lett., 2007, 91,
124103.
ꢁ
S. J. K. Pond, J. W. Perry, S. R. Marder and J.-L. Bredas, Adv.
Funct. Mater., 2002, 12, 631.
38 L. Brandsma, S. F. Vasilevsky and H. D. Verkruijsse, Application of
Transition Metal Catalysts in Organic Synthesis, Springer, Berlin
and New York, 1999, ISBN: 3540655506.
39 K. R. J. Thomas, J. T. Lin, Y. T. Tao and C. W. Ko, J. Am. Chem.
Soc., 2001, 123, 9404.
7 S. Maruo and H. Inoue, Appl. Phys. Lett., 2006, 89, 144101.
8 Z. Xiong, X. Z. Dong, W. Q. Chen and X. M. Duan, Appl. Phys. A:
Mater. Sci. Process., 2008, 93, 447.
ꢀ
40 K. Sonogashira, Y. Tohda and N. Hagihara, Tetrahedron Lett., 1975,
16, 4467.
9 M. Malinauskas, H. Gilbergs, A. Zukauskas, V. Purlys, D. Paipulas
and R. Gadonas, J. Opt., 2010, 12, 035204.
5658 | J. Mater. Chem., 2011, 21, 5650–5659
This journal is ª The Royal Society of Chemistry 2011

View Article Online
41 G. A. Crosby and J. N. Demas, J. Phys. Chem. A, 1971, 75, 991.
42 R. A. Velapoldi and H. H. Tonnesen, J. Fluoresc., 2004, 14, 465.
43 C. Xu and W. W. Webb, J. Opt. Soc. Am. B, 1996, 13, 481.
44 M. Sheik-Bahae, A. A. Said, T. H. Wei, D. J. Hagan and E. W. Van
Stryland, IEEE J. Quantum Electron., 1990, 26, 760.
45 M. Sheik-Bahae, A. A. Said and E. W. Van Stryland, Opt. Lett., 1989,
14, 955.
46 C. Decker and K. Moussa, Macromolecules, 1989, 22, 4455.
47 H. J. Avens and C. N. Bowman, J. Polym. Sci., Part A: Polym. Chem.,
2009, 47, 6083.
This journal is ª The Royal Society of Chemistry 2011
J. Mater. Chem., 2011, 21, 5650–5659 | 5659