Extended mechanistic aspects on photoinitiated polymerization of 1,6-hexanediol diacrylate by hexaarylbisimidazoles and heterocyclic mercapto compounds

Article abstract of DOI:10.1039/c3pp50379h

Chlorine substituted hexaarylbisimidazole (o-Cl-HABI) efficiently initiates radical polymerization of multifunctional acrylic esters in the presence of a heterocyclic mercapto compound if the latter can form its tautomeric thione. Exposure of o-Cl-HABI results in lophyl radicals, which efficiently add to the thione in the first step while the second step releases a highly reactive thiyl radical from this intermediate. LC-MS and CID-MS measurements support this reaction scheme. Furthermore, photo-DSC experiments applying UV light between 320 and 380 nm showed that mercaptotriazole and phenylmercaptotriazole exhibited the best reactivity in the monomer 1,6-hexanediol diacrylate (HDDA) while alkyl substituted mercaptotriazoles showed less reactivity. Change of the triazole heterocycle by mercaptoimidazole resulted in a significant decrease of photoinitiation efficiency. This heterocycle does not form the corresponding thione in HDDA as shown by NMR measurements. Replacement of mercaptotriazole by an alkylthiol leads to a system showing the lowest photoinitiation efficiency in this series. Formation of thione structure in the case of heterocyclic mercapto compounds may cause higher reactivity of the heterocyclic mercapto compounds with the lophyl radical in the monomer chosen. This journal is

Full text of DOI:10.1039/c3pp50379h

Photochemical &
Photobiological Sciences
View Article Online
View Journal | View Issue
PAPER
Extended mechanistic aspects on photoinitiated
polymerization of 1,6-hexanediol diacrylate
by hexaarylbisimidazoles and heterocyclic
mercapto compounds†
Cite this: Photochem. Photobiol. Sci.,
2014, 13, 789
Stefan Berdzinski,a Nadine Strehmel,b Heike Lindauer,c Veronika Strehmel*a and
Bernd Strehmel*a
Chlorine substituted hexaarylbisimidazole (o-Cl-HABI) efficiently initiates radical polymerization of multifunc-
tional acrylic esters in the presence of a heterocyclic mercapto compound if the latter can form its tautomeric
thione. Exposure of o-Cl-HABI results in lophyl radicals, which efficiently add to the thione in the first step while
the second step releases a highly reactive thiyl radical from this intermediate. LC-MS and CID-MS measurements
support this reaction scheme. Furthermore, photo-DSC experiments applying UV light between 320 and
380 nm showed that mercaptotriazole and phenylmercaptotriazole exhibited the best reactivity in the monomer
1,6-hexanediol diacrylate (HDDA) while alkyl substituted mercaptotriazoles showed less reactivity. Change of the
triazole heterocycle by mercaptoimidazole resulted in a significant decrease of photoinitiation efficiency. This
heterocycle does not form the corresponding thione in HDDA as shown by NMR measurements. Replacement
of mercaptotriazole by an alkylthiol leads to a system showing the lowest photoinitiation efficiency in this series.
Formation of thione structure in the case of heterocyclic mercapto compounds may cause higher reactivity of
the heterocyclic mercapto compounds with the lophyl radical in the monomer chosen.
Received 31st October 2013,
Accepted 21st February 2014
DOI: 10.1039/c3pp50379h
www.rsc.org/pps
1. Introduction
The efficiency of this reaction depends on the surrounding
Substituted hexaarylbisimidazoles (HABI) have been known in solvent, and how efficiently L• can escape from the solvent
photoinitiating systems for a long time.1–3 These materials
result in formation of lophyl radicals (L•) upon exposure by
homolytic cleavage of the precursor L–L resulting in formation
of purple colored relatively stable lophyl radicals (L•) as inter-
mediates in ordinary solvents, eqn (1).
cage.4–6 Thus, the lifetime of the lophyl radical ranges between
several minutes to hours in ordinary solvents down to milli-
seconds or microseconds in ionic liquids.4–6 Nevertheless,
the study of this photoreaction has received the attention
of researchers with a focus on both optical properties,
such as photochromism,7–18 and magnetic properties, using
ESR to explore the properties of the radicals generated
and the cleavage mechanism. This can occur through a
radical triplet-pair.19–25 Furthermore, the efficient initiation of
initiator systems containing HABI derivatives directed the
use of such materials in holographic applications based on
photopolymers.26–28
ð1Þ
Thus, hexaarylbisimidazoles have been widely applied in
photoinitiating systems in radical polymerization, in which
they either function as a radical initiator in combination with
a H-donating compound1–3,29 or in combination with a sensi-
tizer, which donates an electron from its excited state to L–L
resulting in formation of the lophyl radical (L•), lophin (L–H)
and the oxidized form of the sensitizer.19,21,26–28,30–37 It has
also been accepted that such sensitized systems require the
addition of a H-donor, which is usually a heterocyclic mer-
capto compound. The generated reactive radical (L•) possesses
aDepartment of Chemistry and Institute for Coatings and Surface Chemistry,
Niederrhein University of Applied Sciences, Adlerstr. 32, D-47798 Krefeld, Germany.
E-mail: bernd.strehmel@hsnr.de, veronika.strehmel@hsnr.de;
Fax: +49 2151 8224195; Tel: +49 2151 822 4075
bLeibniz Institute of Plant Biochemistry, Weinberg 3, D-06120 Halle, Germany
cThuringian Institute of Textile and Plastics Research e.V, Breitscheidstraße 97,
D-07407 Rudolstadt, Germany
†Electronic supplementary information (ESI) available. See DOI:
10.1039/c3pp50379h
This journal is © The Royal Society of Chemistry and Owner Societies 2014
Photochem. Photobiol. Sci., 2014, 13, 789–798 | 789

View Article Online
Paper
Photochemical & Photobiological Sciences
the capability to abstract hydrogen from RS–H.3 Further different compared to a photoinitiating system comprising a
photosensitizer, a radical builder, and a radical initiator. Fur-
thermore, we found that thiol–thione tautomerism has not
received the necessary attention in previous studies, which
should also influence the formation of initiating species.
Therefore, we additionally studied the mechanism in more
detail applying UPLC-ESI-QTOF-MS (ultra performance liquid
chromatography electrospray ionization quadrupole time-of-
flight mass spectrometry) to explore the structure of the com-
pounds formed upon irradiation of the L–L/RSH system. We
chose this hyphenation technique because UPLC provides the
best opportunity to separate compounds from a complex
mixture. The hybrid mass spectrometer ensures a high
resolution in mass and the recording of CID-MS (collision-
induced dissociation mass spectrometry). For this purpose,
the use of a chlorine containing HABI gives us the opportu-
nity to draw conclusions on whether the photoproduct con-
tains chlorine or not. NMR experiments were carried out to
obtain a deeper insight into the thiol–thione equilibrium
of heterocyclic RS–H. This equilibrium depends on the
surrounding medium, and it strongly affects the reactivity of
the L–L/RS–H system. This was studied by photo-DSC
mechanistic discussions also include a mechanism based on
energy transfer from a sensitizer to HABI. The triplet state of
HABI lies relatively low (283 kJ mol−1) and is located in the
region of typical triplet sensitizers (isopropylthioxanthone:
272 kJ mol−1, Michler’s ketone: 260 kJ mol−1).38 Thus, triplet–
triplet energy transfer would be possible in principle using an
appropriate triplet sensitizer.38 Furthermore, ref. 21 critically
compares both mechanisms relating either to energy transfer
or electron transfer. The same authors investigated a sensi-
tized HABI-system comprising three components: a sensitizer,
a radical builder – the HABI – and a radical initiator RS–H –
the mercapto compound.
Furthermore, lophyl radicals are extremely slowly added to
acrylic monomers. As mentioned above, initiation of photo-
polymerization generally requires the addition of a hydrogen
donor (R′–H) resulting in formation of initiating radicals (R′•).
Eqn (2) shows the principal occurring steps, which have been
accepted for several decades.1,2
L• þ RSH ! ꢀ ꢀ ꢀ ! L–H þ RS•
ð2Þ
H-abstraction of the lophyl radical from RS–H was dis- measurements upon exposure to the UV. The combination
cussed as the major reaction.21,26,32–35,39–48 The RS• radical of these experiments extends the consideration to such
finally formed possesses a high reactivity to react with acrylic photoinitiating systems.
monomers resulting in initiation of photoinduced polymeri-
zation. Nowadays, there is no doubt that a hydrogen donor
must be added to HABI upon exposure to obtain initiating
2. Experimental section
2.1 Materials
radicals R• in high yield.
Direct exposure of L–L results in L•1,2 with a quantum yield
of 1.961,2,38 because absorption of one light quant by L–L
2,2′-Bis(2-chlorophenyl)-4,4′,5,5′-tetraphenyl-1,1′-bi-1H imida-
results in formation of two L• radicals. The higher extinction
zole, also known as o-Cl-HABI (L–L), was obtained as a scienti-
fic gift from Hampford Research. The thiols 4H-1,2,4-triazole-
3-thiol (Ia), 4-methyl-4H-1,2,4-triazole-3-thiol (Ib), 5-phenyl-4H-
1,2,4-triazole-3-thiol (Ic), 5-t-butyl-4H-1,2,4-triazole-3-thiol (Id),
and 5-mercapto-1-phenyl-1H-tetrazol (Ie) were kindly donated
by FEW Chemicals GmbH. The photolysis product 2,4,5-triphe-
nylimidazol was available in the lab. Deuterated solvents were
purchased from ARMAR. 2-Mercapto-1-methylimidazol (If),
1-dodecanethiol (II), and the monomer 1,6-hexanediol diacry-
late (HDDA) were purchased from Sigma-Aldrich. All other lab
chemicals needed for the measurements were obtained from
Sigma-Aldrich.
coefficient in the UV compared to many commercial UV-
initiators49 could be seen as one advantage to apply such com-
pounds as photoinitiators. Furthermore, photolytic formation
of lophine (L–H) results in a significant hypsochromic shifted
absorption in the presence of a hydrogen donating coinitiator
(RSH). L–H shows no absorption at λ > 350 nm where HABI
derivatives absorb. This makes HABI interesting for curing
coatings requiring a large depth profile because it can comple-
tely bleach above 350 nm.
Furthermore, different mercapto compounds were selected
to initiate photopolymerization choosing various photosensi-
tizers and L–L32 to discuss the relationship between reactivity
and RS–H bond dissociation energy. All reactions discussed
2.2 Photolysis experiments
in ref. 32 start from the excited state of the sensitizer. The A Hg–Xe 200 W UV lamp (LOT-Oriel) in combination with a
thiyl radicals are generated by different reactions: direct reac- water filter and a 320–380 nm broadband filter was used to
tion of the excited sensitizer with a mercapto compound on generate lophyl radicals by irradiation of o-Cl-HABI dissolved
the one hand and reaction of the excited sensitizer with o-Cl- in CD3CN. The light entered a fiber connected with a tempera-
HABI resulting in lophyl radicals, which react with the mer- ture-controlled cuvette holder QPOD (Ocean Optics) equipped
capto compound in a secondary reaction on the other hand. with a Peltier element and a TC125 temperature controller
Therefore, the thiyl radical concentration is influenced by (Ocean Optics). Exposure measurements were carried out in a
various reactions, and taking a decision about the contri- 2 mm cuvette in which 1.8 mg o-Cl-HABI and 12 mg of the
bution of a single reaction becomes difficult. In this contri- 4H-1,2,4-triazole-3-thiol (Ia) were dissolved in 0.7 mL CD3CN
bution, we discuss the reaction of thiyl radical formation by and exposed at 298 K for 10 h. The solutions exposed were
direct excitation of L–L without any sensitizer. This may be transferred to LC-MS experiments.
790 | Photochem. Photobiol. Sci., 2014, 13, 789–798
This journal is © The Royal Society of Chemistry and Owner Societies 2014

View Article Online
Photochemical & Photobiological Sciences
Paper
2.3 Photo-DSC experiments
II electrospray ion source (Bruker Daltonics, Billerica, MA,
USA). All mass spectra were acquired in centroid mode
and recalibrated on the basis of lithium formate cluster ions.
Reaction products were extracted by visual inspection of
base peak chromatograms (isolation width: 0.02 Da) and
identified on the basis of their accurate mass and their col-
lision-induced dissociation (CID) mass spectrum. For acqui-
sition of CID mass spectra quasi-molecular cluster ions were
isolated at the Q1 (isolation width: 3) and fragmented inside
the collision cell using argon as a collision gas. Product ions
were detected as already described. More details can be found
in the ESI.†
A DSC Q-300 from TA Instruments was synchronized with an
Omnicure Series 2000-XLA with appropriate y-fiber optics,
which equally directed the light on both sensors using the
PCA-kit available from TA-Instruments. Insertion of a broad-
band filter results in exposure at 320–380 nm using a mid-
pressure mercury lamp while the main part of the emission
originates from the 365 nm Hg emission line. An amount of
0.1 mol% o-Cl-HABI and 0.67 mol% of the corresponding co-
initiator RSH was dissolved in HDDA at a temperature of
50 °C. The time required to dissolve the mercapto compounds
depended on their structure (Ia: 1 h; Ib: 2 h; Ic:10 h; Id: 3 h; Ie:
4 h; If: 2 h; II: 1 h). The samples were placed on an uncovered
Tzero aluminium pan having a weight of 5 mg under nitrogen
purge of 50 mL min−1. Equilibration of each sample was com-
plete 15 min prior to exposure. The samples were irradiated
for 5 minutes under a constant exposure power of 25 mW
cm−2 at 25 °C. After irradiation, residual heat flow was moni-
tored in a time frame of an additional 10 minutes. The heat
flow (in J s−1) measured was divided by both the molar
polymerization heat (ΔHp) of HDDA (167 kJ mol−1; ref. 50)
and the quantity of the monomer nm (in mol) resulting in
the polymerization rate (dx/dt in s−1) as a function of time,
eqn (3).
P1 (3-[2-(4,5-diphenyl-1H-imidazol-2-yl)phenyl]-4,5-dihydro-
1H-1,2,4-triazole-5-thione). ESI(+)-CID-MS/MS of m/z 396.13,
[M + H]+, CE = 40 eV; m/z (rel. int. (%)) = 327.0936 (100,
+
C21H15N2S+), 293.1058 (17.4, C21H13N2 ), 250.0534 (2.8,
C15H10N2S•+), 249.0469 (2.1, C15H9N2S+), 224.0507 (6.7,
C14H10NS+), 223.0437 (5.4, C14H9NS•+), 218.0823 (6.7,
+
C15H10N2•+), 217.0755 (2.5, C15H9N2 ), 193.0876 (45.5,
C14H11N•+), 167.0846 (28.7, C13H11+), 165.0693 (11.3, C13H9 ),
+
152.0615 (6.1, C12H8•+).
P2
(2-(2-chlorophenyl)-4,5-diphenyl-1H-imidazole). ESI
(+)-CID-MS/MS of m/z 331.1, [M + H]+, CE = 40 eV; m/z (rel.
+
+
int. (%)) = 331.0979 (2, C21H16ClN2 ), 295.1217 (3.7, C21H15N2 ),
194.0944 (49.2, C14H12N+), 193.0874 (100, C14H11N•+), 177.0681
(8, C14H9+), 176.0614 (4.5, C14H8•+), 167.0841 (84.7, C13H11+),
˙
dx
dt
Q
1
˙
¼ x ¼
ð3Þ
+
ΔHp nm
165.0691 (79.3, C13H9 ), 152.0612 (68.1, C12H8•+), 125.014
(2.4, C7H6Cl+), 117.0557 (12.1, C8H7N•+), 116.0483 (20.6,
C8H6N+).
In the next step, these data were numerically integrated
with respect to the reaction time resulting in the fractional
degree of polymerization x(t) as a function of time, eqn (4).
Some samples were poured into MeOH to determine the
amount of insoluble crosslinked polymer.
ð
P3 (2,4,5-triphenyl-1H-imidazole). ESI(+)-CID-MS/MS of m/z
297.14, [M + H]+, CE = 40 eV; m/z (rel. int. (%)) = 297.1375 (1.5,
+
C21H17N2 ), 193.0877 (100, C14H11N•+), 192.0795 (10.3,
C14H10N+), 177.0684 (6.6, C14H9 ), 176.0609 (4.8, C14H8•+),
+
+
167.0837 (63.3, C13H11+), 165.0688 (72.5, C13H9 ), 152.0611
˙
xðtÞ ¼ x dt
ð4Þ
+
(72.6, C12H8•+), 151.0531 (5.1, C12H7+), 141.0685 (2.3, C11H9 ),
0
117.0554 (13.5, C8H7N•+), 116.0478 (20.7, C8H6N+), 104.0473
(2.1, C7H6N+).
2.4 NMR study of tautomerism
P4 (2-(2-chlorophenyl)-4,5-diphenyl-imidazol-1-ylium). ESI
(+)-CID-MS/MS of m/z 329.08, [M
+ = 40 eV;
H]+, CE
A Bruker Fourier-300 MHz spectrometer was used for all NMR
measurements. The mercapto compound was dissolved in
HDDA at a concentration of 0.05 M. The resulting solution was
transferred to the NMR-tube containing a capillary filled with
CDCl3 being necessary to obtain the log-signal. H-NMR spec-
troscopy was applied to study the thiol–thione tautomer equili-
+
m/z (rel. int. (%)) = 329.0822 (25.1, C21H14ClN2 ), 294.1142
(6.1, C21H14N2•+), 192.0784 (6.1, C14H10N+), 165.071 (100,
+
C13H9 ).
1
P5a
(1-[2-(2-chlorophenyl)-4,5-diphenyl-2H-imidazol-2-yl]-
2,4,5-triphenyl-1H-imidazol-3-ium). ESI(+)-CID-MS/MS of m/z
625.22, [M + H]+, CE = 40 eV; m/z (rel. int. (%)) = 625.2125
brium in HDDA.
+
+
(100, C42H30ClN4 ), 488.2102 (3.7, C35H26N3 ), 432.127 (4,
2.5 LC-MS analysis of photolysis products
+
+
C28H19ClN3 ), 329.0817 (3.1, C21H14ClN2 ), 194.0965 (10.4,
Reaction products were analyzed by ultra performance liquid C14H12N+).
chromatography coupled to electrospray ionization quadrupole P5b (1-[2-(2-chlorophenyl)-4,5-diphenyl-2H-imidazol-2-yl]-
time-of-flight mass spectrometry (UPLC-ESI-QTOF-MS). On 2,4,5-triphenyl-1H-imidazol-3-ium). ESI(+)-CID-MS/MS of m/z
this account, diluted samples were injected onto an Acquinity 625.22, [M + H]+, CE = 40 eV; m/z (rel. int. (%)) = 625.2145
+
UPLC system (Waters, Eschborn/Germany) mounted with a (100, C42H30ClN4 ), 488.2109 (63.1, C35H26N3+), 472.194
HSS T3 column and separated using a binary gradient. Eluting (3.5, C35H24N2•+), 432.1344 (7.6, C30H16N4•+), 419.1346
compounds were detected in positive and negative ionization (4.4, C28H20ClN2 ), 396.1543 (2.7, C27H23ClN+), 385.1685
+
modes from m/z 50–1000 using a MicroTOF-Q I hybrid quadru- (39.2, C28H21N2 ), 322.1392 (12.3, C21H21ClN+), 306.118 (8,
+
+
pole time-of-flight mass spectrometer equipped with an Apollo C22H14N2•+), 295.1143 (7, C20H13N3•+), 207.0897 (2, C14H11N2 ),
This journal is © The Royal Society of Chemistry and Owner Societies 2014
Photochem. Photobiol. Sci., 2014, 13, 789–798 | 791

View Article Online
Paper
Photochemical & Photobiological Sciences
194.0956 (17.8, C14H12N+), 193.0876 (7.7, C14H11N•+), 180.0832 density in the heterocycle, and bond dissociation energy of the
(29.7, C13H10N+), 167.0864 (18.3, C13H11+), 165.0694 (2.7, S–H bond.
C13H9+), 152.063 (4.4, C12H8•+).
L–L (1H-imidazole, 2-(2-chlorophenyl)-1-[2-(2-chlorophenyl)-
4,5-diphenyl-2H-imidazol-2-yl]-4,5-diphenyl). ESI(+)-CID-MS/
MS of m/z 659.18, [M + H]+, CE = 40 eV; m/z (rel. int. (%)) =
+
329.082 (100, C21H14ClN2 ), 294.1126 (4.5, C21H14N2•+),
226.0405 (23.8, C14H9ClN+), 192.0792 (11.5, C14H10N+),
191.0713 (5.9, C14H9N•+), 165.0692 (28.2, C13H9+), 122.998
(5.4, C7H4Cl+).
Abbreviation
R1
Y
Z
Ia. ESI(−)-CID-MS/MS of m/z 100, [M − H]−, CE = 10 eV; m/z
(rel. int. (%)) = 99.997 (100, C2H2N3S−), 71.989 (2, C2H2NS−).
Ia
Ib
Ic
Id
Ie
If
H
CH3
H
H
Phenyl
CH3
N
N
N
N
N
CH
CH
CH
C-Phenyl
C-t-Butyl
N
2.6 Electrochemical experiments
CH
The electrochemical investigations were carried out with a
three-electrode setup using platinum disks as a working
respectively auxiliary electrode and Ag/AgCl as a reference elec-
trode. For all measurements, ferrocene was used as an internal
The general structure shown in I exhibits one possible tau-
tomeric form, which is the thiol in this case. Principally, the
thiol coexists in an equilibrium together with the tautomeric
thione, eqn (6). This equilibrium depends on the surrounding
solvent. Thus, one needs to ask which tautomeric form signifi-
cantly affects the initiation of photopolymerization. The thiol
should favor a mechanism based on hydrogen abstraction
while the thione may result in a different one. 1H-NMR spectra
exhibit the ratio of thiol/thione in the monomer used, which is
HDDA in our experiments.
standard and 0.1
M
tetrabutylammonium hexafluoro-
phosphate (Bu4NPF6) as a supporting electrolyte. The half
wave potentials of the internal standard ferrocene usually
amounted to 0.51 eV, which was checked prior to each series
of measurements. The compounds studied were dissolved
in absolute acetonitrile (water content ≤0.001%, from Sigma-
Aldrich) with a concentration of 0.003 M. All solutions
were purged with nitrogen prior to each measurement.
Cyclic voltammetry was performed with a Solartron 1285
Potentiostat at a scan rate of 15 mV s−1. It started with measur-
ing an entire cycle of the respective compound. The measure-
ments of the oxidation and reduction potentials of the
compounds were carried out separately from each other. The
measured cyclic voltammograms were evaluated using self-
developed software. The HOMO and LUMO energies were
calculated from the respective half wave potentials according
to eqn (5):
ð6Þ
We also chose the aliphatic mercapto compound II, in
which no thione exists. Thus, II represents a system for a
hydrogen transfer mechanism with no alternatives.
EHOMO=LUMO ¼ ½ꢁðEonset ðvs: Ag=AgClÞ ꢁ Eonset ðFc=Fc vs: Ag=AgClÞÞꢂ
þ
The HDDA monomer efficiently crosslinks. We chose a photo-
DSC setup to study the efficiency of the photoinitiating system
comprising L–L and the mercapto compound selected from
either I or II.
ꢁ 4:8 eV
ð5Þ
In the cases where no oxidation or reduction peak was
found, the HOMO and LUMO energies were calculated by the
optical gap ΔE from the UV-Vis spectra using the tangent method.
3.2 Reactivity of the HABI-mercapto systems
Photo-DSC measurements easily provide information regard-
ing the reactivity of photoinitiating systems containing cross-
linking monomers. The maximum rate of polymerization vmax
and the time necessary to reach vmax empirically relate to the
reactivity of the initiator system in the crosslinking material.
3. Results and discussion
3.1 Selection of materials
Excitation of L–L quantitatively results in formation of two The higher the vmax the shorter must be tmax. A large vmax cor-
lophyl radicals L• with a quantum yield of ≈2 with respect to responds to a highly reactive and efficient photoinitiating
L•.38 The lophyl radical formed efficiently reacts with hetero- system. Fig. 1 represents one possible example of photo-
cyclic mercapto compounds I, which were selected from induced crosslinking of HDDA initiated by the initiator combi-
5-membered heterocycles derived from triazole (Ia–d), tetrazole nation L–L and Ia showing the reactivity parameters vmax and
(Ie), and imidazole (If). The different materials compiled tmax.. The limiting conversion (x∞) typically depends on avail-
in the general structure I possess different redox potentials, able free volume, reaction temperature, and functionality of
and therefore, different HOMO–LUMO energies, electron the monomer chosen.
792 | Photochem. Photobiol. Sci., 2014, 13, 789–798
This journal is © The Royal Society of Chemistry and Owner Societies 2014

View Article Online
Photochemical & Photobiological Sciences
Paper
similar tendency in comparison with Ie, Ia and If in our
directly photolysed HABI/RS–H systems. These results indicate
that a discussion of reactivity cannot be exclusively explained
by the hydrogen abstraction mechanism.3 Further reactions
may compete with H-abstraction in the HABI/RSH system. This
is supported by comparison with data of the aliphatic thiol II
exhibiting the lowest efficiency of all coinitiators. Structure II
allows only hydrogen abstraction by a precursor radical.
The tautomeric equilibrium may provide alternative reaction
routes.
Moreover, the reactivity of L• with RSH also strongly
depends on the surroundings. Irradiation of L–L in HDDA
showed a fast loss of L• absorbance after addition of Ia while II
resulted in a slower reaction between L• and II (see ESI† for
more details). However, a faster reactivity of II with L• was
observed in DMSO showing that H-abstraction can also
compete by change of the surrounding. This would be impor-
tant for selection of the monomer. Nevertheless, analysis of
the sample exposed in the photo-DSC setup showed an almost
quantitative conversion of soluble HDDA into a crosslinked
polymer using Ia as a coinitiator while II gives only 15% in-
soluble polymer under similar conditions (more details in
ESI†). This demonstrates again the higher reactivity of Ia in
comparison with II in HDDA.
Fig. 1 Photopolymerization of HDDA (99.4 wt%) initiated by the system
L–L (0.1 mol%) and Ia (0.67 mol%) using the photo-DSC setup
described.
Table 1 Determined maximum polymerization rates vmax and the time
tmax to reach vmax during photopolymerization of HDDA using the
initiator system consisting of o-Cl-HABI (0.1 mol%) and the investigated
co-initiators I and II (0.67 mol%). As a reference, photoinitiated poly-
merization was studied without any coinitiator
Coinitiator
vmax (10−3 s−1
)
tmax (s)
Ia
Ib
Ic
Id
Ie
If
II
162
40
144
72
182
5
2.2
18.3
1.6
7.8
1.4
88.9
284
3.3 Tautomerism of the mercapto compounds
NMR experiments focused on studying the thiol–thione equili-
brium existing in compounds of the general structure I. We
chose two compounds with high and low reactivity, that is, Ia
and If, respectively. Both only differ in the Y-position of I.
Fig. 2 shows the NMR-spectrum of Ia in HDDA. It exhibits
typical protons assigned to the tautomeric thione-form. Inte-
2
Without
1
1859
Table 1 shows the results obtained for photopolymerization
of HDDA using different L–L/RSH systems. The aliphatic mer-
capto compound II, which cannot form any thione, exhibits
the lowest reactivity in this series. The reactivity of II is close to
a system without any mercapto compound. It has been well
accepted that HABIs without any coinitiator do not function as
initiators in radical photopolymer systems.1–3
Data compiled in Table 1 allow us to notice a drawback
regarding the reactivity of the HABI/RSH systems in HDDA.
Thus, one can write the following series that generally splits
into three groups:
1
gration of the overall H-NMR spectrum yields the amount of
each proton. In this experiment we found that 79% thione
exists, though this depends on the polarity of the solvent.
Thus, the spectrum in DMSO-d6 results in almost quantitative
formation of the thione form. Therefore, a more detailed
understanding requires us to take the NMR spectrum in the
monomer chosen for radical polymerization. The insertion of
CDCl3 in fused capillaries in the NMR-tube allows us to obtain
the necessary log-signal.
Ie > Ia > Ic ≫ Id > Ib ≫ If > II > without.
The mercapto compounds Ie, Ia, and Ic exhibit the highest
reactivity derived from heterocycles with a distinct electron
density in the heterocyclic ring. Nevertheless, the alkyl substi-
tuted triazoles Id and Ib result in a reactivity that is signifi-
cantly lower compared to Ie, Ia, and Ic. There is no clear
correlation between the substitution pattern of the triazole
moiety and photopolymerization reactivity measured. More-
over, the imidazole compound If exhibits the lowest reactivity
in the series of heterocyclic mercapto compounds although it
possesses the lowest S–H bond dissociation energy (322 kJ
mol−1) compared to Ia and Ie showing both a S–H bond dis-
sociation energy of 365 kJ mol−1 32
investigated in sensitized HABI/RSH systems32 fairly shows a Fig. 2 1H-NMR spectrum of Ia dissolved in HDDA.
.
Nevertheless, the reactivity
This journal is © The Royal Society of Chemistry and Owner Societies 2014
Photochem. Photobiol. Sci., 2014, 13, 789–798 | 793

View Article Online
Paper
Photochemical & Photobiological Sciences
Fig. 3 Signal of the proton bound at the nitrogen-atom of the thione-
tautomer of If in DMSO-d6 (red) and in HDDA (black) in the 1H-NMR
spectrum.
Fig. 4 UPLC-ESI(+)-QTOF-MS base peak chromatogram (m/z
100–700) of an irradiated mixture comprising Ia and L–L (elemental
C , P3: C21H17N2 , P4:
23H18N5S+, P2: C21H16ClN2
+
+
composition: P1:
+
C21H14ClN2+, P5a: C42H30ClN4+, P5b: C42H30ClN4+, L–L: C42H29Cl2N4
more details in ESI†).
,
Almost no thione was found in the case of If dissolved in
HDDA (Fig. 3) although it possesses a large amount of thione
in DMSO-d6 (≈100%). The more reactive Ia shows a higher con-
tribution of thione in HDDA while the less reactive If does not
exhibit any thione in the HDDA monomer measured by
1H-NMR spectroscopy. If exhibits a reactivity comparable with
the reference compound II, which cannot form any thione
tautomer. Thus, it is obvious that a certain amount of thione
significantly increases photopolymerization reactivity.
On the other hand, no clear correlation exists between
thione content in HDDA and photopolymerization reactivity
(Ia > Id > lb). 1H-NMR measurements indicated high thione
contents in alkyl substituted triazole derivatives (Ia: 64%; Id:
100%, lb: 90%). Furthermore, determination of thione was not
possible in the phenyl-substituted heterocycles Ic and Ie
because H-exchange may occur between more available
tautomers.
expected isotope ratio (more details in ESI†). Thus, chlorine
must be released somehow during photolysis. The identity of
P3 was confirmed by measuring the in-house available refer-
ence substance. As a proposal, photolysis product P1 can be
explained as an addition product of L• and Ia, because the
CID-MS shows a loss of a triazole unit (C2H3N3) resulting in a
fragment at 327.09 amu. This exhibits the elemental compo-
sition C21H15N2S+ and bears a sulfur at the triphenylimidazol
skeleton. Furthermore, P1 exhibits a similar CID-MS compared
to P3. These are the fragments found at 193.09, 167.08, 165.07
and 152.06 amu confirming the identity of a triphenylimida-
zole skeleton.
Furthermore, chlorine-containing P2 was assigned as a
further photolysis product. It exhibits the typical isotope ratio
of a chlorine containing material (more details in ESI†). P4
was identified as a cation. The fragments found in the CID-MS
indicate the presence of a triphenylimidazole skeleton. Pro-
ducts P5a and P5b were most likely formed by cross coupling
of a chlorine-bearing and a chlorine-free triphenylimidazolyl
radical. Both are isomers with the same mass (compare with
the caption of Fig. 4).
3.4 Photolysis products
The reaction mixture obtained after exposure of L–L and Ia
was investigated by UPLC-ESI-QTOF-MS to gain more infor-
mation about the photolysis products formed. This analysis is
necessary because the occurring mechanism appears to be
more complicated as previously assumed.3 Fig. 4 shows a
typical chromatogram of the reaction mixture. L–L (appearing
at 900 s) nearly completely converted into photolysis products
from which the main products P1–P5 resulted.
3.5 Mechanistic aspects
Based on the literature,29,51 one could propose a mechanism
describing electron transfer of RS–H into the hole of photo-
excited HABI (L–L*) explaining formation of the lophine P2,
eqn (7). Such mechanistic discussions of o-Cl-HABI were pre-
viously described in L–L/leuco dye systems.1,2
L–L* þ RS–H ! L–Lꢁ• þ RS–Hþ•
ð7Þ
This would be thermodynamically possible resulting in oxi-
dation of the heterocyclic mercapto compound (RS–H+•) and
Ia only shows a mass spectrum after ionization in the negative formation of the reduced HABI (L2−•). RS–H+• could form a
mode (injection peak at 30 s). This signal clearly diminishes thiyl radical by release of a proton. The redox potentials
after reaction (data not shown).
measured for o-Cl-HABI and the mercapto compounds would
Interestingly, the photolysis products P1 and P3 do not support such a proposal only for some of the heterocyclic
contain any chlorine because the MS does not exhibit the compounds I, Table 2. L–L possesses an oxidation potential of
794 | Photochem. Photobiol. Sci., 2014, 13, 789–798
This journal is © The Royal Society of Chemistry and Owner Societies 2014

View Article Online
Photochemical & Photobiological Sciences
Paper
Table 2 Oxidation potential (Eox) and respective HOMO energies of I
measured in acetonitrile. ΔE represents the energy difference between
the HOMO of L–L and the HOMO of I
Reactivity
scale
Compound
Eox (V)
HOMO (eV)
ΔEa (eV)
Ie
Ic
Id
Ib
Ia
If
1.89
1.54
1.43
1.06
0.94
0.83
−5.99
−5.65
−5.37
−5.23
−5.06
−4.98
0.18
−0.16
−0.44
−0.58
−0.75
−0.83
1
3
4
5
2
6
a ΔE = (HOMOL–L–HOMORS–H).
Fig. 5 Proposal for the extended mechanistic formation of initiator
radicals (Cl• and In•) during exposure of L–L in the presence of the
thione of Ia.
1.58 V corresponding to a HOMO energy of −5.81 eV. According
to the data of I compiled in Table 2, all thiols/thiones except for
Ie should follow this proposal, because their HOMO energies
are located higher compared to L–L. On the other hand, Ie
exhibited the best photoinitiation reactivity but had the lowest
oxidation capability. Furthermore, the imidazole derivative If
showed a low initiation efficiency but the best oxidation capa-
bility. If electron transfer would dominate according to eqn
(7), one should observe a clear correlation between the free
energy of electron transfer and photoinitiation efficiency. This
is not really supported by our data. In addition, a mechanism
in which L• would be reduced by the mercapto compound
can be excluded due to its reduction potential; that is, Ered
(L•/L−) = −0.15 V (measured against the ferrocene/ferrocenium
potential).52
The formation of chlorine free photoproducts P1 and P3 as
well as mixed photoproducts P5a and P5b shows that alterna-
tive reactions presumably explain the formation of thiyl radi-
cals in our system. Therefore, a mechanistic discussion
should more or less focus on the electron density distribution
comparing Ia and If. According to its electronic properties, L•
belongs to a nucleophilic radical.53 Most of its unpaired elec-
tron density is located in the p-orbitals of the SOMO (compare
with Fig. S1 in ESI†). In addition, nucleophilic radicals possess
between L• and If, which exists in HDDA only as a thiol. This
agrees with our experimental findings where L• exhibits low
reactivity with II. On the other hand, remarkable electron
density exists for the C-atoms at the 5-position of Ia consider-
ing the thione tautomer. This must lead to a better interaction
of the heterocycle Ia with L• resulting in an increased reactiv-
ity. HDDA favors formation of the thione tautomer of Ia as
shown by NMR.
An addition elimination mechanism as shown in Fig. 5
would explain formation of chlorine free photoproducts, i.e.
P1 and P3 as well as the mixed photoproducts P5a and P5b,
and formation of initiating radicals In•. Both, P1 and P2 are
major products. Formation of P2 can alternatively occur
according to H-abstraction.1,2,38 The release of highly reactive
chlorine concluded from MS analysis of photoproducts may
explain the high reactivity of o-Cl-HABI, which can compete
with photoinitiators grouped in the IRGACURE series.
4. Conclusions
a low capability to abstract hydrogen from appropriate donors, Results obtained would require us to extend the mechanism
which are for example aliphatic thiols. This would require regarding the formation of initiator radicals in the case of a
unpaired electron density in the s-orbitals of the SOMO photoinitiator system comprising L–L and a heterocyclic mer-
because H-abstraction results in formation of a σ-bond. Our capto compound. Mercapto compounds, which are able to
experiments showed a low reactivity of L• with II, which experi- form the tautomeric thione, exhibit a higher reactivity. This
mentally proves the nucleophilic properties of L•. Thus, H- tautomeric form reacts fast with the lophyl radical resulting in
abstraction should occur with lower efficiency as experi- generation of initiating thiyl radicals although competitive
mentally shown by data in Table 1.
reactions such as H-abstraction or electron transfer according
Nucleophilic radicals prefer to interact with the LUMO of a to eqn (7) may occur with lower efficiency leading to the same
conjugated system, that is, in our case the reaction of L• with initiating species – the thiyl radical. The surrounding matrix
the heterocycle I. The larger the atomic orbital coefficients in influences these competing reactions. Furthermore, the rela-
the LUMO, the larger the interaction energy, and thus, the tively low sensitivity of the lophyl radical against oxygen may
reactivity according to the perturbation theory.53 This refers be a further benefit of this system explaining the high photo-
for example to the C-atoms in the 4 and 5 position of If and initiating reactivity.
the 5-position of Ia (compare with the patterns in ESI†). MO-
Further research on systems comprising L–L should focus
calculations of If showed for the LUMO nearly no electron on multifunctional heterocyclic thiol/thione compounds
density at 4 and 5 positions if this exists as a thiol. Thus, the keeping in mind that the thione mainly affects photoini-
low reactivity of If is therefore a result of the low interaction tiation efficiency in acrylic monomers. Such multifunctional
This journal is © The Royal Society of Chemistry and Owner Societies 2014
Photochem. Photobiol. Sci., 2014, 13, 789–798 | 795

View Article Online
Paper
Photochemical & Photobiological Sciences
heterocycles could be easily built up by the reaction of a multi-
functional amine and thiosemicarbazide. This may lead to
higher crosslinking densities of the crosslinked acrylate
because formation of the initiating thiyl radical occurs at a
higher functional moiety. Triazoles based on structure Ic rep-
resent a reasonable basis of such an approach because even
this structure shows a high reactivity in HDDA while alkyl sub-
stituted compounds based on Id would result in a decrease of
initiation efficiency.
8 S. Hatano, T. Horino, A. Tokita, T. Oshima and J. Abe,
Unusual Negative Photochromism via
a Short-Lived
Imidazolyl Radical of 1,1′-Binaphthyl-Bridged Imidazole
Dimer, J. Am. Chem. Soc., 2013, 135, 3164–3172.
9 K. Fujita, S. Hatano, D. Kato and J. Abe, Photochromism of
a radical diffusion-inhibited hexaarylbiimidazole derivative
with intense coloration and fast decoloration performance,
Org. Lett., 2008, 10, 3105–3108.
10 K. Mutoh, E. Nakano and J. Abe, Spectroelectrochemistry of
Solubility of o-Cl HABI in common coatings can be seen as
one issue involved in making this system more attractive for a
broader range of applications. This can be achieved by
a Photochromic [2.2]Paracyclophane-Bridged Imidazole
Dimer: Clarification of the Electrochemical Behavior of
HABI, J. Phys. Chem. A, 2012, 116, 6792–6797.
incorporation of substituents improving the solubility. Never- 11 A. Kikuchi, T. Iyoda and J. Abe, Electronic structure of
theless, these changes should not influence the reactivity of
HABI in combination with thiol/thione compounds.
light-induced lophyl radical derived from a novel hexa-
arylbiimidazole with pi-conjugated chromophore, Chem.
Commun., 2002, 1484–1485.
Further mechanistic studies should also clarify the function
of chlorine in the HABI skeleton. Perhaps, a modification with 12 H. Sato, K. Kasatani and S. Murakami, Magnetic field
more halogens at the ring would bring new impetus into this
field since the release of chlorine as reactive and additional
effects on the photochromism of hexaphenylbiimidazolyl,
Chem. Phys. Lett., 1988, 151, 97–101.
initiating species has been discussed as a matter of fact in this 13 Y. Sakaino, Structures and chromotropic properties of imi-
contribution. Thus, we believe that further mechanistic studies
will bring new progress in developing efficient photoinitiating
systems. This requires an extended mechanistic concept.
dazole derivatives produced from 3,6-bis(4,5-diphenyl-2H-
imidazol-2-ylidene)cyclohexa-1,4-diene, J. Chem. Soc., Perkin
Trans. 1, 1983, 1063–1066.
14 Y. Sakaino, H. Kakisawa and T. Kusumi, Synthesis of
new photochromic compounds from the dimer of aryl-
phenanthroimidazoles, J. Heterocycl. Chem., 1975, 12,
953–956.
Acknowledgements
VS gratefully acknowledges the Federal Ministry of Education 15 K. Maeda and T. Hayashi, Mechanism of photochromism,
and Research for financial support within the “Professorinnen-
programm” (grant #01FP09231 C).
thermochromism, and piezochromism of dimers of triaryl-
imidazolyl, Bull. Chem. Soc. Jpn., 1970, 43, 429–438.
16 T. Shida, K. Maeda and T. Hayashi, Optical and
E.S.R. studies on triphenylimidazolyl radicals produced by
photolysis and radiolysis at low temperatures, Bull. Chem.
Soc. Jpn., 1969, 42, 3044.
Notes and references
1 R. Dessauer, Photochemistry, History and Commercial Appli- 17 M. A. J. Wilks and M. R. Willis, Kinetics of the photo-
cations of Hexaarylbiimidazoles: All about HABIs, Elsevier,
2006.
chromic decay reaction of 2,2′,4,4′,5,5′-hexaphenyl-
biimidazolyl, J. Chem. Soc. B, 1968, 1526–1529.
2 R. Dessauer, The invention of Dylux instant-access imaging 18 T. Hayashi, K. Maeda and T. Kanaji, The mechanism of the
materials and the development of HABI chemistry – a
personal history, Adv. Photochem., 2005, 28, 129–261.
3 B. M. Monroe and G. C. Weed, Photoinitiators for free-
photochromism and thermochromism of 2,2′,4,4′,5,5′-hexa-
phenyl-1,1′-biimidazolyl in a solid state, Bull. Chem. Soc.
Jpn., 1965, 38, 857.
radical-initiated photoimaging systems, Chem. Rev., 1993, 19 J. V. Caspar, I. V. Khudyakov, N. J. Turro and G. C. Weed,
93, 435–448.
ESR Study of Lophyl Free Radicals in Dry Films, Macro-
molecules, 1995, 28, 636–641.
4 V. Strehmel, J. V. Whishart, D. A. Polyansky and
B. Strehmel, Recombination of Photogenerated Lophyl 20 X. Z. Qin, A. Liu, A. D. Trifunac and V. V. Krongauz,
Radicals in Imidazolium-Based Ionic Liquids, Chem-
PhysChem, 2009, 10, 3112–3118.
Photodissociation of hexaarylbiimidazole. 1. Triplet-state
formation, J. Phys. Chem., 1991, 95, 5822–5826.
5 S. Berdzinski, J. Horst, P. Strassburg and V. Strehmel, 21 A. D. Liu, A. D. Trifunac and V. V. Krongauz, Photo-
Recombination of Lophyl Radicals in Pyrrolidinium-Based
dissociation of hexaarylbiimidazole. 2. Direct and sensi-
Ionic Liquids, ChemPhysChem, 2013, 24, 1751–1760.
tized dissociation, J. Phys. Chem., 1992, 96, 207–211.
6 V. Strehmel, Radicals in ionic liquids, ChemPhysChem, 22 T. Matsuura, Y. Ito and I. Saito, Photoinduced reactions.
2012, 13, 1649–1663.
LXVIII. Photochemical dehydrogenation of imidazolines to
7 K. Mutoh, M. Sliwa and J. Abe, Rapid Fluorescence Switch-
imidazoles, Bull. Chem. Soc. Jpn., 1973, 46, 3805–3809.
ing by Using a Fast Photochromic [2.2]Paracyclophane- 23 T. Hayashi and K. Maeda, Mechanism of reversible photo-
Bridged Imidazole Dimer, J. Phys. Chem. C, 2013, 117,
4808–4814.
chemical color change of bis(triphenylimidazolyl) at low
temperature, Bull. Chem. Soc. Jpn., 1967, 40, 2990.
796 | Photochem. Photobiol. Sci., 2014, 13, 789–798
This journal is © The Royal Society of Chemistry and Owner Societies 2014

View Article Online
Photochemical & Photobiological Sciences
Paper
24 J. Abe, T. Sano, M. Kawano, Y. Ohashi, M. M. Matsushita 37 T. Urano, H. Ito, K. Takahama and T. Yamaoka, Double-
and T. Iyoda, EPR and density functional studies of light-
induced radical pairs in a single crystal of a hexaaryl-
biimidazolyl derivative, Angew. Chem., Int. Ed., 2001, 40,
580–582.
amplified photoinitiator systems, RadTech’98 North America
UV/EB Conference Proceedings, Chicago, April 19–22, 1998,
1998, 734–745.
38 Q. Q. Zhu, M. Fink, F. Seitz, S. Schneider and W. Schnabel,
On the photolysis of bis[2-(o-chlorophenyl)-4,5-diphenyl-
imidazole] sensitized by 2-isopropylthioxanthone or
Michler’s ketone, J. Photochem. Photobiol., A, 1991, 59,
255–263.
39 M. Jin, J.-P. Malval, D.-L. Versace, F. Morlet-Savary,
H. Chaumeil, A. Defoin, X. Allonas and J.-P. Fouassier,
Two-photon absorption and polymerization ability of intra-
molecular energy transfer based photoinitiating systems,
Chem. Commun., 2008, 6540–6542.
40 X. Allonas, J. P. Fouassier, M. Kaji and Y. Murakami,
Excited state processes in a four-component photosensitive
system based on a bisimidazole derivative, Photochem.
Photobiol. Sci., 2003, 2, 224–229.
41 X. Allonas, J. P. Fouassier, M. Kaji, M. Miyasaka and
T. Hidaka, Two and three component photoinitiating
systems based on coumarin derivatives, Polymer, 2001, 42,
7627–7634.
25 M. Schrödner, H. Schache, H. Lindauer, R. Schrödner and
A. Konkin, Oscillations of the lophyl radical concentra-
tion during the photo-decomposition of o-Cl-hexaaryl-
bisimidazole, J. Photochem. Photobiol., A, 2012, 233,
60–64.
26 A. Ibrahim, C. Ley, X. Allonas, O. I. Tarzi, A. Chan Yong,
C. Carre and R. Chevallier, Optimization of a photopoly-
merizable material based on a photocyclic initiating system
using holographic recording, Photochem. Photobiol. Sci.,
2012, 11, 1682–1690.
27 X. Hu, H. Zhang, G. Zhang, Y. Sheng, Y. Wang and R. Zhao,
Water-resisting compound red-sensitive photopolymer:
mechanism and optimization of components, Proc.
SPIE-Int. Soc. Opt. Eng., 2005, 5636, 635–645.
28 C. Zhang, J. Zhao, J. He, L. Li and Y. Yang, Photoinitiating
systems and photopolymer materials for holography, Proc.
SPIE-Int. Soc. Opt. Eng., 1998, 3559, 81–87.
29 D. F. Eaton, A. G. Horgan and J. P. Horgan, Mechanism of 42 X. Allonas, J. P. Fouassier, M. Kaji and M. Miyasaka, On the
coinitiation of photopolymerization of methyl methacrylate
by hexaarylbiimidazole-hydrogen atom donor combi-
ability of coumarin derivatives to interact with photo-
initiators, J. Photopolym. Sci. Technol., 2000, 13, 237–242.
nations. The role of electron transfer versus direct hydro- 43 T. Urano, Sensitizer dyes and sensitization mechanisms in
gen atom abstraction, J. Photochem. Photobiol., A, 1991, 58,
373–391.
photopolymer coating layer, Proc. SPIE-Int. Soc. Opt. Eng.,
2002, 4659, 420–427.
30 X. Li, Y. Zhao, J. Wu, M. Shi and F. Wu, Two-photon 44 T. Urano, M. Ishikawa, Y. Sato and H. Itoh, Sensitizer dyes
photopolymerization using novel asymmetric keto-
coumarin derivatives, J. Photochem. Photobiol., A, 2007, 190,
22–28.
and sensitization mechanisms in photopolymer coating
layer II, J. Photopolym. Sci. Technol., 1999, 12, 711–716.
45 T. Urano, E. Hino, H. Ito, M. Shimizu and T. Yamaoka,
Study of radical generated from coumarin dye-sensitized
photoinitiator systems in high-speed photopolymer coating
layers using laser flash photolysis, Polym. Adv. Technol.,
1998, 9, 825–830.
31 J. Xue, Y. Zhao, J. Wu and F. Wu, Novel benzylidene
cyclopentanone dyes for two-photon photopolymerization,
J. Photochem. Photobiol., A, 2008, 195, 261–266.
32 J. Lalevee, F. Morlet-Savary, R. M. El, X. Allonas and
J. P. Fouassier, Thiyl radical generation in thiol or disulfide 46 T. Urano, H. Nagasaka, S. Shimizu, H. Ito, M. Shimizu,
containing photosensitive systems, Macromol. Chem. Phys.,
2009, 210, 311–319.
33 S. Suzuki, P. Emilie, T. Urano, S. Takahara and T. Yamaoka,
Three-component photoradical initiating system-the effect
of 2-mercaptobenzothiazole as a coinitiator in polymer
matrix, Polymer, 2005, 46, 2238–2243.
34 S. Suzuki, T. Urano, K. Ito, T. Murayama, I. Hotta,
S. Takahara and T. Yamaoka, Pyrromethene dye sensitized
photopolymer and the application to visible laser direct
imaging, J. Photopolym. Sci. Technol., 2004, 17, 125–
130.
35 S. Suzuki, T. Urano, N. Miyagawa, S. Takahara and
T. Yamaoka, Three-component radical photoinitiating
systems – the effect of the accelerator, J. Photopolym. Sci.
Technol., 2001, 14, 259–262.
S. Takahara and T. Yamaoka, Laser-flash photolysis in
high-speed photopolymer coating layers containing carbo-
nylbiscoumarin dyes and 2,2′-bis(2-chlorophenyl)-4,4′,5,5′-
tetraphenyl-1,1′-bi-1H-imidazole, Bull. Chem. Soc. Jpn.,
1997, 70, 1659–1664.
47 T. Urano, H. Nagasaka, M. Shimizu, S. Takahara and
T. Yamaoka, Laser flash photolysis in high-speed photo-
polymer coating layers: effects of incorporating dimethyl
phthalate as a plasticizer into coating formulation, Bull.
Chem. Soc. Jpn., 1996, 69, 693–700.
48 T. Urano, H. Nagasaka, M. Tsuchiyama, S. Shimizu,
K. Kawazoe, M. Shimizu and T. Yamaoka, Laser flash
photolysis in high-speed photopolymer coating layers, Bull.
Chem. Soc. Jpn., 1995, 68, 1661–1668.
49 W. A. Green, Industrial Photoinitiators: A Technical Guide,
CRC Press, Boca Raton, 2010.
36 T. Urano, H. Ito, K. Takahama and T. Yamaoka, Photo-
polymerization mechanisms of acrylates in poly(methyl 50 H.-J. Timpe and B. Strehmel, Lichtinduzierte Polymer- und
methacrylate) films, Polym. Adv. Technol., 1999, 10,
201–205.
Polymerisationsreaktionen. 44. Zur Kinetik der radi-
kalischen Photopolymerisation mehrfunktionellerAcrylate
This journal is © The Royal Society of Chemistry and Owner Societies 2014
Photochem. Photobiol. Sci., 2014, 13, 789–798 | 797

View Article Online
Paper
Photochemical & Photobiological Sciences
in polymeren Bindemitteln, Makromol. Chem., 1991, 192, 52 E. W. Oliver, D. H. Evans and J. V. Caspar, Electrochemical
779–791.
studies of a hexaarylbiimidazole, J. Electroanal. Chem.,
1996, 403, 153–158.
51 M. Muneer, P. V. Kamat and M. V. George, Electron transfer
reactions. Reaction of nitrogen heterocycles with potassium, 53 I. Fleming and J. Podlech, Molekülorbitale und Reaktionen
Can. J. Chem., 1990, 68, 969–975.
organischer Verbindungen, Wiley-VCH, Weinheim, 2012.
798 | Photochem. Photobiol. Sci., 2014, 13, 789–798
This journal is © The Royal Society of Chemistry and Owner Societies 2014