Photopolymerization of thick films and in shadow areas: A review for the access to composites

Abstract

The photopolymerization processes are currently associated with thin samples for which the light penetration is good enough to activate the photoinitiator or the photoinitiating system for the entre sample’s thickness. The photopolymerization of very thick films and in shadow areas where the light penetration is inhibited (e.g in filled, pigmented, dispersed samples) remains a huge challenge (e.g. for the access to composites). In the present paper, an overview of the different strategies for the photopolymerization of thick samples is reported. First, strategies based on optimization of the photonic (light intensity, excitation wavelength…) or chemical (efficiency/reactivity/bleaching of the photoinitiating systems…) parameters are presented resulting in a full temporal and spatial control. Then, the main strategies based on propagation/diffusion mechanisms of latent species for the curing beyond the irradiated areas are given (partial loss of spatial resolution and access to shadow areas). Also, dual systems (thermal/photochemical or photochemical/redox) are described. The state of the art for the access to thick samples by photopolymerization processes as well as some perspectives are provided.

Full text

Polymer
Chemistry
REVIEW
Cite this: Polym. Chem., 2017, 8,
7088
Received 21st October 2017,
Accepted 7th November 2017
DOI: 10.1039/c7py01778b
rsc.li/polymers
Photopolymerization processes of thick films
and in shadow areas: a review for the access to
composites
Patxi Garra,
a
Céline Dietlin,
a
Fabrice Morlet-Savary,
a
Frédéric Dumur,
b
Didier Gigmes,
b
Jean-Pierre Fouassier
a
and Jacques Lalevée *
a
The photopolymerization processes are currently associated with thin samples for which the light pene-
tration is good enough to activate the photoinitiator or the photoinitiating system for the entre sample’s
thickness. The photopolymerization of very thick films and in shadow areas where the light penetration is
inhibited (e.g. in filled, pigmented, and dispersed samples) remains a huge challenge (e.g. for the access
to composites). In the present paper, an overview of the different strategies for the photopolymerization
of thick samples is reported. First, strategies based on the optimization of the photonic (light intensity,
excitation wavelength, etc.) or chemical (efficiency/reactivity/bleaching of the photoinitiating systems,
etc.) parameters are presented that result in a full temporal and spatial control. Then, the main strategies
based on propagation/diffusion mechanisms of latent species for the curing beyond the irradiated areas
are given (partial loss of spatial resolution and access to shadow areas). Also, dual systems (thermal/
photochemical or photochemical/redox) are described. The state of the art for the access to thick
samples by photopolymerization processes as well as some perspectives are provided.
I. Introduction
The photopolymerization reactions are currently restricted to
the field of coatings, clear coats or varnishes where the light
penetration is high enough for these thin samples (a few μm
to 50 μm typically) to ensure a fast and efficient curing. For
thicker samples and/or in the presence of fillers or pigments
or in dispersed media, the light penetration is not good
enough to ensure an efficient polymerization process in depth.
The presence of shadow areas is also a classical drawback of
the photopolymerization approach vs. classical techniques
(thermal or redox initiated polymerization). Therefore, the
development of new strategies for efficient curing in depth
and/or with shadow areas is a huge challenge to extend the
advantages of the photopolymerization (reaction at room
temperature, potential spatial and temporal control, no VOC
involved, the use of safe and cheap irradiation devices (LEDs),
etc.) to new fields with the access to composites or polymeriz-
ation in pigmented or dispersed media.
It is not relevant to compare photopolymerization depths
(also denoted the depth of cure (DoC)) or extensions in
shadow areas for all the existing reports, patents or publi-
cations as they depend on too many variables: optical den-
sities, light scattering, additives, monomers, the photobleach-
ing phenomenon, photopolymer applications, concentrations,
irradiances, irradiation times, sample masses, one com-
ponent/two component approaches, resin stability, etc.
Therefore, this review will propose a state of the art concerning
the various aspects involved in thick film and shadow area
photopolymerization: (part I) a rapid overview of photo-
polymerization principles and the associated light penetration
issues; (part II) highlights of interesting reports showing in
depth photopolymerization strategies based on low optical
densities, photobleaching and NIR upconversion strategies
(full time/spatial control) and finally (part III) a broad pano-
rama of the strategies based on propagation/diffusion mecha-
nisms (latent species, heat, etc.) for partial to complete curing
beyond the irradiated areas.
I.1 Photopolymerization and photoinitiating systems (PIS)
Light induced reactions are of great interest as they require
small energy inputs with limited emission of volatile organic
compounds and excellent time and spatial controls. The reac-
tion can be carried out at room temperature using eco-friendly,
compact, cheap and safe irradiation devices like Light
Emitting Diodes (LEDs). Among them, photopolymerization
has emerged as an elegant method to produce a broad range
of polymeric materials.
1–8
Photoinitiating systems (PIS) are
a
Institut de Science des Matériaux de Mulhouse IS2M, UMR CNRS 7361, UHA, 15,
rue Jean Starcky, 68057 Mulhouse Cedex, France. E-mail: jacques.lalevee@uha.fr
b
Aix-Marseille Univ, CNRS, ICR, F-13397 Marseille, France
7088 |Polym. C hem.,2017,8,7088–7101 This journal is © The Royal Society of Chemistry 2017
Published on 08 November 2017. Downloaded by Universite Pierre et Marie Curie on 01/12/2017 07:05:50.
View Article Online
View Journal
| View Issue

irradiated by an actinic source which leads –through different
chemical mechanisms –to the formation of initiating species
(e.g. free radicals, cations, anions, acids, bases, etc.)
(Scheme 1). These initiating species –as in thermal polymeriz-
ation –react with monomers and/or oligomers in order to
form the required polymer through radical, cationic or anionic
polymerization. The design of the photopolymerizable resin is
always related to the required application and the final end-
user properties of the polymers (mechanical, flexibility,
adhesion properties, etc.). The dual propagation of active
species is also possible, for example, in interpenetrating
polymer network (IPN) synthesis
9
through the polymerization
of radical/cationic monomer blends.
Photopolymerization has many applications in varnishes,
paints, coatings, adhesives, graphic arts, medicine, microelec-
tronics, microlithography, 3D machining, optics, etc. Many
actinic sources can be used: xenon lamps, mercury arc lamps,
doped lamps, microwave lamps, excimer lamps, light emitting
diodes (LED), pulsed light sources, laser sources, the sun,
household lamps, UV plasma sources, etc. The most con-
venient way to characterize a light source is to give its irradi-
ance i.e. the energy intensity per unit of time per unit of
surface (J s
−1
cm
2
or W cm
−2
) as photopolymerizable resins are
irradiated during a certain time (s) on a spatially limited
surface (cm
2
) exposed to a light source. Indeed, broad range
emission sources such as mercury lamps (e.g. 300–500 nm)
can lead to some energy losses i.e. only a selected wavelength
range is matching with the PIS absorption spectrum in order
to form active species.
2
In contrast, LEDs having narrow emis-
sion wavelengths allow optimum photochemical processes
with limited energy losses with a good match with the PIS i.e.
in the ideal case, the maximum emission wavelength of the
selected LED must be close to the maximum absorption wave-
length of the photoinitiator.
10
The power input/emission
output rate of the actinic source has to be also taken into
consideration.
Historically, the applications of photopolymerization are
related to thin samples (<50 µm) as light penetration in thick
samples is limited (the PIS have generally high to very high
light absorptivities
1
). Accordingly, currently, the major indus-
trial applications are in the field of coatings or adhesives
requiring such thin films. Recently, access to thick samples
and/or shadow areas was proposed in order to strongly extend
the potential of the photopolymerization applications.
However, the photopolymerization strategies involved have to
face light penetration issues as a minimum irradiance in
depth is required to initiate the polymerization process for a
given PIS. This will be developed in parts I-2, I-3 and I-4.
I.2 Inner filter effect in photopolymerization for clear
formulations
It is generally well known that the light penetration decreases
with the optical density that depends on the molar concen-
tration of the absorbing species, their molar absorptivity and
the length of the optical path through the Beer–Lambert law.
11
It is of major importance for in depth curing of thick samples:
irradiance will decrease throughout the entire photopolymeriz-
able medium (that contains absorbing PIS) and thus the pro-
duction of active species allowing curing. Therefore, the depth
of cure is strongly related to optical densities, irradiance of the
actinic source(s) and irradiation times. It can also be influ-
enced by the photobleaching phenomenon (see part II.2). The
depth of cure in photopolymerization was well exemplified by
Lee et al.
12
(Fig. 1). Their system contains a Norrish type I
photoinitiator (2-benzyl-2-N,N-(dimethylamino)-1-(4-morpholi-
nophenyl)-1-butanone DBMP) irradiated by an actinic source
having monochromatic irradiance at 325 nm (fully absorbed
by DBMP). When multiplying irradiation time per irradiance,
one can obtain an energy dose (in J cm
−2
) that is a good way to
account for both parameters. First, the larger the DBMP
amount, the higher the optical density and thus the higher the
irradiation attenuation, which leads to a thinner depth of cure.
Also, too low DBMP amounts lead to a weakening of the photo-
polymerization rate: indeed, the energy absorbed by the PIS
(I
abs
) is of major importance for efficient photopolymerization
rates. As a result of all these effects, there is not an optimum
concentration of DBMP that will result in the deepest photo-
Fig. 1 Surface topology of the calculated curing space reproduced
from Lee et al.
12
For a 2-benzyl-2-N,N-(dimethylamino)-1-(4-morpho-
linophenyl)-1-butanone (DBMP) photoinitiator receiving an E
m
energy
dose.
Scheme 1 General photopolymerization chemical mechanisms for a
photoinitiating system (PIS).
Polymer Chemistry Review
This journal is © The Royal Society of Chemistry 2017 Polym. Chem.,2017,8,7088–7101 | 7089
Published on 08 November 2017. Downloaded by Universite Pierre et Marie Curie on 01/12/2017 07:05:50.
View Article Online

polymer for any energy dose. The DBMP concentration has to
be re-assessed as a function of the energy dose received and
can be particularly low for a good in depth curing. For
example, when the energy dose is 5 J cm
−2
, the DoC will be
2.4 mm when [DBMP] ≈0.08 mM.
In depth photopolymerization can be followed using thermo-
couples present at a fixed depth in the photopolymerizable
media, but more conveniently by thermal imaging
13
and also
by more exotic means such as 2D electron spin resonance
(ESR) methods
14–16
or electroresistance monitoring.
17
For
thermal follow-ups of the photopolymerization, one should
carefully separate exothermicities related to monomer conver-
sions and other heat generating phenomena such as non-
radiative excited state relaxations (when high irradiances are
involved) or direct heating from high irradiance polychromatic
lamps containing infrared radiation.
17,18
I.3 Other light penetration issues: pigments, fillers and
dispersed media
In photopolymerizable media, light absorption can occur due
to species that are not the PIS leading to an inner filter effect
that strongly decreases both the polymerization efficiency and
the DoC. Rarely, the monomers
19,20
used can absorb the
actinic light (almost only under UV irradiations <300 nm).
Therefore, huge light penetration issues are usually associated
with pigments and fillers (in functional photopolymers) which
will be in competition with the PIS for the light absorption.
Pigments are defined as insoluble particles (of sizes <10 µm)
strongly changing the reflected or transmitted color interacting
with the suspension containing them. Pigments are particu-
larly useful for inks or painting applications and generally
show very high light absorptions, particularly in the near UV
to visible wavelengths.
21
Due to their own absorption spectra,
the pigments usually absorb a large part of the light intensity
i.e. there is usually an excitation wavelength overlap between
PIS absorption (in actinic light wavelengths) and pigment
absorption, a very strong light penetration issue will occur due
to the inner filter effect of the pigments.
22
For inks, the curing
is usually restricted to very thin samples (1–5μm) for which
the remaining light intensity (not absorbed by the pigment)
can still activate the PIS to initiate the polymerization process.
The curing of a thicker sample in the presence of pigments is
much more difficult i.e. the absorption of the pigments avoid-
ing a sufficient penetration of the light.
When fillers are inserted into the photopolymerizable
media, the problem is even more complex.
23
First, light scatter-
ing phenomena occur according to the size and shapes of the
fillers (Mie scattering, Rayleigh scattering, multiple scattering
effects, fiber homogeneity, etc.). As an example case study
(Fig. 2),
24
the scattering coefficient for composite dental
materials (containing fillers) is 3.1 mm
−1
under visible light
irradiation (470 nm) while it is lower for near infrared
irradiation (NIR): 1.2 mm
−1
. Second, fillers themselves can
show elevated light absorption coefficients throughout variable
wavelength ranges of the UV-NIR spectra; for example, carbon
black shows broad range absorption (200–2500 nm).
25
As a
general rule (that can show exceptions), NIR irradiations
should have a better penetration in applied photopolymers
containing pigments or fillers.
Therefore, in photopolymerization, the search for red
shifted (up to NIR ranges) active PIS is very intense
26–33
even
if: (i) the inner filter effect is still present at NIR wavelengths
and (ii) NIR photochemical processes are not easy due to the
reduced energy of the photon. Indeed, a photon absorbed at
900 nm is three times less energetic than a UV one at 300 nm,
which partly explains why photopolymerization has historically
started under UV irradiations:
1,2,34
this lower photon energy
leads to harder photochemical processes when the excited
state energy of the photoinitiator decreases.
35–37
I.4 Shadow areas in photopolymerization
Some applications require photopolymerization to be carried
out in shadow areas. Shadow areas are defined as areas where
photopolymerizable resins are present but light obstacles
prevent the light irradiation and no irradiance can reach
them.
28,38
Curing of these shadow areas is necessary for many
applications: indeed, experimental setups (during actual appli-
cations) are not as perfect as in the laboratory and the actinic
user does not perform optimum irradiation of all the photo-
polymerizable media. Also, for example, in dental restorative
composites, some areas are actually light hindered and it is
not possible to irradiate them. Next, some fillers and pigments
(see part I.3) can also completely prevent light penetration
(no irradiance). For all these applications, it should be very
interesting to develop PIS that can be active in the irradiated
areas and also diffuse active species (called latent species) from
the irradiated zone to the shadow areas in order to cure them.
Curing of shadow areas will necessarily imply partial to
complete loss of spatial (and sometimes temporal) control of
the photopolymerization process.
Fig. 2 Wavelength dependence of the scattering coefficient in dental
composites (commercial glass fillers); reproduced from Uo et al.
24
Review Polymer Chemistry
7090 |Polym. Chem.,2017,8,7088–7101 This journal is © The Royal Society of Chemistry 2017
Published on 08 November 2017. Downloaded by Universite Pierre et Marie Curie on 01/12/2017 07:05:50.
View Article Online

II. Strategies for thick samples based
on direct optimization of the photonic
parameters for the primary actinic
source/PIS/polymerizable medium
systems
The following part will discuss strategies based on the optimi-
zation of the photonic parameters for the three key features of
photopolymerization: (i) the actinic source, (ii) the PIS, and
(iii) the polymerizable medium. Allegedly, such strategies
allow good spatial and temporal control of the photo-
polymerization even if, to a certain extent, propagating effects
might be present, for example, heat diffusion for thick
samples or heating effects through high irradiance NIR lasers.
II.1 Low optical density strategies
When considering a simple Beer–Lambert
11
model for the
irradiance penetration of a LED@405 nm (irradiance of 100
mW cm
−2
, narrow emission spectra) the penetration of light
can be simplified using eqn (1) where I(x) is the irradiance at a
distance of x(in cm) from the surface; I(0 cm) is the irradiance
at the surface by the LED@405 nm; and Abs@405 is the optical
density of the photopolymerizable resin at 405 nm (1 cm cell).
IðxÞ¼Ið0cmÞ10xAbs@405 ð1Þ
For example, in Fig. 3, at the beginning of a LED@405 nm
irradiation of a resin having an optical density (at 405 nm) of
1, the penetrating irradiance is very low beyond 2 cm (only
0.9 mW cm
−2
remaining at 2 cm). In contrast, there is still
12.3 mW cm
−2
irradiance remaining at 10 cm distance when
O.D@405 nm is 0.1. As a result, when designing PIS concen-
trations aiming at the photopolymerization of very thick
samples, one could target very low optical densities at the
irradiation wavelengths (the use of high dye
39
concentrations
is usually not recommended). Also, the PIS involved should be
very efficient to still be active under very low irradiances (in
depth). This strategy was particularly adapted for the photo-
polymerization of 8.5 cm thick samples using charge transfer
complexes (CTC).
38
CTC were formed through an amine/
iodonium equilibrium and the quantity of CTC absorbing
species (and their molar absorptivity) was low. Nevertheless,
no measurement of the optical densities in the resin was
performed.
New PIS, to be published by Garra et al., are also involving
charge transfer complexes (CTC) between N-aromatic amines,
iodonium salts and phosphines and were able to polymerize
rapidly a methacrylate resin under soft LED irradiation. This
system showed very low optical densities at the excitation wave-
lengths (O.D@405 nm = 0.13 for 1 cm in a methacrylate resin),
which enabled a fast 9 cm photopolymerization under a soft
irradiance (230 mW cm
−2
) LED@405 nm (see Fig. 4). Next, the
same resin enabled complete curing of 31 cm photopolymer
under a higher irradiance LED@405 nm (1.1 W cm
−2
). This
size, to the best of our knowledge, is outranking previously
reported extremely thick photopolymerizations.
The issues of optical densities are also reported in holo-
graphic PIS
40,41
and numerical modeling studies.
12,42
To a
certain extent, the commercial camphorquinone/amine PIS is
based on a low optical density strategy: the molar absorptivity
of camphorquinone is low under blue irradiation (about
40 M
−1
cm
−1
at 470 nm) and quite thick filled samples can be
obtained when optimal concentrations and fillers are
used.
24,43,44
Finally and obviously, higher irradiances and
irradiation times will be preferable in order to have better
irradiation doses in depth curing.
II.2 Photochemical bleaching of the primary PIS
After the initial light penetration issue (caused by the initial
optical density of the PIS), the photochemical bleaching
Fig. 3 Theoretical penetrating irradiance vs. depth (cm) from a
LED@405 nm for optical densities at 405 nm (O.D@405 nm) of photo-
polymerizable resins of 0.1 and 1.
Fig. 4 Relationship between the calculated irradiance penetration (for
LED@405 nm, eqn (1)) and photopolymerization propagation time as a
function of the distance from the surface (9 cm photopolymerization
using a low optical density strategy). To be published.
Polymer Chemistry Review
This journal is © The Royal Society of Chemistry 2017 Polym. Chem.,2017,8,7088–7101 | 7091
Published on 08 November 2017. Downloaded by Universite Pierre et Marie Curie on 01/12/2017 07:05:50.
View Article Online

phenomenon can occur during irradiation. Generally speak-
ing, photolysis is occurring when photochemical reactions are
leading to changes in the chemical nature of the absorbing
species. When photolysis products are absorbing at lower
wavelengths, the degradation of the absorbing species is com-
monly referred to as photobleaching.
45
In thick sample photo-
polymerization, this effect was well covered by Miller et al. who
carried out exhaustive simulations of the problem.
46
They
highlighted that many parameters can influence the photo-
polymerization in depth (a thickness of 1 cm was considered)
such as molar absorptivity, photobleaching rate and initial PIS
concentrations and also the possibility of photolysis by-
products. Previously, during the 90s, mono and bis phosphine
oxides appeared as the only family of photoinitiators being
able to polymerize clear, pigmented, filled or glass reinforced
fiber acrylate coatings.
3,5,47–50
Thanks to the particular bleach-
ing properties of these compounds, the photocuring of 3 cm
thick filled materials and ∼10–15 cm thick clear varnishes has
been claimed. Nevertheless long irradiation times and high
energy UV irradiation sources were necessary for very thick
sample curing thanks to phosphine oxide photobleaching (e.g.
25 minutes to cure 8.5 cm of a clear acrylate formulation
17
).
Asmussen et al.
51
have studied the photobleaching of the
commercial camphorquinone (CQ)/amine system under
various conditions. The results are different: CQ consumption
(bleaching) close to the surface rapidly slows down photo-
polymerization rates at these depths when in-depth bleaching
is not very intense as a result of high light attenuation. For
that system, very long irradiation times (>1000 s) were necess-
ary for significant bleaching of CQ (50% bleached throughout
the entire medium). PIS with high photobleaching rates are of
high interest for dental composites as proposed by Bouzrati-
Zerelli et al.
52
Very recently, new classes of Type I photoinitia-
tors such as silylglyoxylates showed excellent photobleaching
properties (see Fig. 5) under blue light irradiations; 6 mm
clear samples could be obtained using these PIS.
53
Altogether, photobleaching strategies are not very user
friendly and each case will show different kinetics. The main
issues are that high photobleaching rates are particularly
evidenced when PIS exhibit high molar absorptivities. This last
point leads to a high optical density in resin, which prevents
light penetration in thick samples (see part II.1). In general,
efficient photobleaching strategies for thick samples require
very long irradiation times or very high irradiances
46,54
(for
example, 3 cm thicknesses after 3000 s UV irradiation
55
).
II.3 Secondary light emission strategy: upconversion particles
as a way to extend the irradiated areas
Conventional luminescent materials re-emit photons with a
lower energy than the incident ones (thus with higher wave-
length from the usual Stokes shift)
56
in classical luminescent
materials such as fluorescent proteins or dyes,
57,58
in materials
composed of (noble) metals
59–61
or even in metal-free lumines-
cent photopolymers.
62
In contrast, upconversion particles
generate anti-Stokes photoluminescence
56
(see Scheme 2),
which makes them interesting in a photopolymerization
process –though the yield of upconversion is usually low –
requiring high energy sources. From the light absorbed at a
higher wavelength (such as NIR, which generally shows better
in-depth penetration), near UV or blue light is re-emitted and
can be absorbed by existing PIS
63
(Scheme 3).
This approach was elegantly applied by Soga and co-
workers
24,64
for the photopolymerization of filled samples for
dental applications where the primary light emission source
was a NIR laser (980 nm, 900 mW; the size of the surface was
not reported) and the visible light PIS was an Irgacure 784
(Type I photoinitiator). Thick filled samples up to 6 mm
(40 wt% dental fillers) were obtained.
Stepuk et al.
65
also used a similar strategy using a NIR laser
operating at a very high irradiance of 90 W cm
−2
, resulting –
through upconversion mechanisms –in a blue light re-emitted
Scheme 2 Upconversion particles mechanism and emission spectrum
from Y
2
O
3
:1Tm,5Yb under NIR excitation (980 nm), reproduced from
Soga and coworkers.
64
Fig. 5 Steady state photolysis of a 1.4 mm methacrylate film of in the
presence of 2 wt% tert-butyl (tert-butyldimethylsilyl) glyoxylate (DKSi)
upon irradiation with a dental LED at 477 nm (300 mW cm
−2
). 1.4 mm
thickness. Reproduced from Macromolecules.
53
Review Polymer Chemistry
7092 |Polym. Chem.,2017,8,7088–7101 This journal is © The Royal Society of Chemistry 2017
Published on 08 November 2017. Downloaded by Universite Pierre et Marie Curie on 01/12/2017 07:05:50.
View Article Online

of 1 mW cm
−2
(absorbed by a camphorquinone/amine system)
i.e. a 1/90 000 upconversion yield. 7 mm thick doped (20 wt%
upconversion) polymers were obtained. The temperature
elevation of the sample was also considered using thermo-
couples placed at 1 and 2 mm from the irradiation source.
More recently, the work of Liu et al. allowed extremely deep
clear thiol–ene photopolymerization
66
(8.4 cm, irradiance of
24.5 W cm
−2
) and clear acrylate photopolymerization
18
(13.7 cm, irradiance of 9.4 W cm
−2
) using high power NIR
(980 nm) light excitation. Again, the visible light PIS was an
Irgacure 784; this latter initiator was characterized by high
photochemical reactivity. Very interestingly, the deep photo-
polymerization was followed by thermal imaging, which allows:
(i) estimation of the blank elevation of temperature without
visible light PIS and (ii) estimation of the photopolymerization
propagation times. For the deepest curing, 0.3 wt% upconver-
sion particles were used (optimum concentration with the low
NIR inner filter effect). Very recent research on upconversion
particles indicates that a UV light (350–365 nm) can also be
obtained from a NIR light irradiation (974 nm (ref. 67) and
980 nm (ref. 68)) of upconversion particles. They were therefore
applied to UV-induced radical
68
(PIS = benzophenone/amine)
and cationic
67
(PIS = 2-isopropylthioxanthone (ITX) in combi-
nation with iodonium salt) polymerizations.
Altogether, the use of NIR upconversion particles is promis-
ing in order to extend the irradiated areas but some limit-
ations still prevent their widespread use: (i) the high ir-
radiances (9–90 W cm
−2
)–thus high energy consumptions –
of the NIR lasers currently used, (ii) the use of rare earth
metals, (iii) no estimation of the current spatial control with
such systems, (iv) the inner filter effect at NIR wavelengths
caused by the upconversion particles themselves (as they
absorb light), and more importantly (v) the low yield of the
upconversion mechanism.
III. Strategies based on latent species
and access to shadow areas
In this part, strategies that can occur without light will be dis-
cussed. They appear thanks to: latent species produced in
radical photopolymerization (part III.1) or in cationic polymeriz-
ation (part III.2); but also the underlying slow polymerization
process (e.g. redox polymerization) in part III.3. All of them
exhibit partial to complete losses of spatial and temporal control
of the photopolymerization which is of high interest for photo-
polymerizable media with very high light penetration issues.
III.1 Latent species in radical photopolymerization
III.1.1 Rose bengal/ferrocenium salt/amine/CHP for thick
pigmented samples. A four-component photoinitiating system
for the photopolymerization of thick (300–400 µm) pigmented
coatings was developed by Grotzinger et al.:
22
it was proposed
that three first components (rose bengal/ferrocenium salt/
amine) initiate the photopolymerization but also lead to the
formation of an in situ reduced form of the initiator (latent
species being the reduced iron arene salt) that can react with a
fourth component (cumene hydroperoxide (CHP)) in a redox
(dark) process. The light intensity used was particularly low
(25 mW cm
−2
at λ> 530 nm).
Diffusion from irradiated zones to deep photopolymeriz-
able resin (with almost no irradiance) was very useful in the
case of thick pigmented samples.
III.1.2 Phototriggered latent base generation. More recently,
a report using photolatent amines
69
for epoxide ring opening
polymerization and thiol-epoxide interpenetrating network
synthesis showed interesting but moderate polymerization
results (>20 minutes to reach mild conversions). The approach
to release latent amines (or base) from light stimuli is more
and more arising under various terms: photolatent amines,
photobase generator,
70
phototriggered base proliferation,
photocaged (super) bases.
Very interestingly (Scheme 4), under UV irradiation, photo-
caged bases release bases that can react with base amplifiers
to generate latent amines which further initiate a redox
polymerization process with a peroxide inherently present in the
polymerizable media. This approach was used by He et al.
71
for
the free radical polymerization of acrylates. Only the center of the
polymerizable medium was irradiated and excellent polymeriz-
ation propagation was observed for the entire polymerizable
medium (2.5 cm from the center). Interestingly, thermal imaging
also enabled following the propagation front but some questions
remain: the temperatures reached throughout the entire medium
were very high (167 to 210 °C) and these temperatures are well
above the benzoylperoxide (BPO) self-accelerating decomposition
into radical species
72,73
(see part III.3.2).
The report was followed by other applications such as thiol–
ene polymerization,
74
photolatent anionic polymerization at high
temperature (heated media at 120 °C),
75
the use of other per-
oxides for FRP
76
or the use of advanced photocaged bases.
77,78
The main limitations of these approaches are: (i) the
harmful low penetration UV irradiation; (ii) the stability of the
peroxides in the resin; (iii) the solvent-dependent base prolifer-
Scheme 3 General path for free radical photopolymerization under
NIR lights using upconversion particles (UP) and a visible light PIS (here
Irgacure 784) in the medium.
24,64
Scheme 4 General path for free radical latent photopolymerization
under UV light (365 nm) irradiation of the photocaged base which will
result in the production of 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU).
DBU reacts with some base amplifier in order to produce latent amines
active in the redox polymerization with peroxides (here benzoylperoxide
(BPO)).
Polymer Chemistry Review
This journal is © The Royal Society of Chemistry 2017 Polym. Chem.,2017,8,7088–7101 | 7093
Published on 08 November 2017. Downloaded by Universite Pierre et Marie Curie on 01/12/2017 07:05:50.
View Article Online

ation; (iv) the apprehension of the mass (thus exothermicity)
effects on the proliferation reaction. Indeed, very high exo-
thermicities (>150 °C) were often reported.
III.1.3 Latent radical-initiated polymerization via 2e
−
/1H
+
transfer. Next, Aguirre-Soto et al.
79
reported an advanced study
defining the lateral photopolymerization. A metal-free three-
component system composed of methylene blue (MB), N,N-
diisopropylethylamine (DIPEA) and diphenyliodonium (DPI
+
).
The photoredox catalytic cycle is depicted as follows
(Scheme 5): under visible light (500–800 nm), MB was excited
to a triplet state and further quenched by DIPEA; the charge
transfer exciplex formed released protons, amine by-products
and a leuco form of the dye (LMB). This latter can diffuse
throughout the entire polymerizable medium and slowly react
with DPI
+
in order to generate radical initiating species (R
•
).
A 1.2 cm thick clear photopolymer was obtained under soft
irradiation conditions (3.4 mW cm
−2
). Also, lateral photo-
polymerization was proposed as a concept and achieved (3.7 ±
0.7 mm polymerization beyond light exposure): usually, axial
photopolymerization setups are composed of a light source
vertically irradiating a polymerizable medium containing the
PIS. If a mask is present horizontally preventing light absorp-
tion, lateral propagation is obviously related to dark curing
mechanisms involving the diffusion of latent species from the
irradiated area to the shadow monomer areas.
Very importantly, in this study, lateral photopolymerization
was performed in 500 µm thick layers which discard the possi-
bility of heat propagation as a possible way to propagate photo-
polymerization in shadow areas. Also, very low irradiances
were used (3–30 mW cm
−2
), which validates the latent photo-
chemical mechanism proposed: a slow diffusion of latent
species is more favored under low irradiances compared to a
rapid stiffening of the polymer network under high ir-
radiances.
80
The main issues of the current system are: (i) the
slow propagation (∼30 minutes); (ii) the low viscosity mono-
functional monomers such as 2-hydroxyethyl methacrylate
(HEMA) and glyceroldimethacrylate (GDMA) which allow good
diffusion but have to be blended with crosslinkers (and/or
additives) in many of the current photopolymerization
applications.
III.1.4 Latent ROOH species in situ formed through oxygen
inhibition: lateral photopolymerization as a way to access 9 cm
thick filled samples. More recently, Garra et al.
80
reproduced
the lateral photopolymerization setup proposed by Aguirre-
Soto et al.
79
(500 µm thick) for a three-component system com-
posed of the following (Scheme 6): a highly active Cu(I) photo-
redox catalyst,
81
iodonium co-initiator (Iod, for the first gene-
ration of radicals in the irradiated zone) and tin(II) 2-ethylhex-
anoate (tin(II), a reducing agent active in copper catalytic
cycles). In the irradiated areas, the well-known Cu(I)*/Iod
reaction
81–83
(Cu(II) being a by-product) allows the first gene-
ration of radicals, which classically produces a chain growing
polymerization of methacrylates. Due to the very mild
irradiation conditions (LED@405 nm, 4 mW cm
−2
), the radical
production rate is not fast enough and many growing macro-
radicals are quenched by the oxygen dissolved in the low vis-
cosity resin (production of peroxyl radicals R′OO
•
). As is gener-
ally accepted,
84,85
the inhibited macroradicals (R′OO
•
)are
resulting in R′OOH hydroperoxides (in situ generated) in the
media. R′OOH are latent species i.e. they can easily diffuse
throughout the entire polymerizable area, especially in the
shadow area (where no light is present). In these shadow
areas, Cu(I)/R′OOH redox reaction
86
(Cu(II) being a by-product)
allows the generation of R′O
•
active species that allow a lateral
propagation (in these areas, oxygen inhibition results in R′OO
•
inhibited species that result in R′OOH production, etc.). Tin(II)
finds use in both the irradiated and shadowed areas in order
to regenerate Cu(I)fromCu(II) by-products: photoredox (Cu(I)*/
Iod/Tin(II)) and redox (Cu(I)/R̲′O̲O̲H̲i̲n̲s̲i̲t̲u̲/Tin(II)) catalytic cycles
are proposed to take place (Scheme 6). The system was capable
of full 29 mm lateral polymerization (see Fig. 6) i.e. the entire
sample beyond the irradiated area was cured.
Also, the usefulness of strategies based on latent species
diffusion was clearly demonstrated for: (i) sunlight induced
photopolymerization (very low irradiances <5 mW cm
−2
)oflow
viscosity methacrylates; (ii) a 9 cm thick homogeneous photo-
polymer (full sample cured) and, even more interestingly, (iii) a
9 cm thick filled sample (45 wt% fillers, full sample cured after
4 minutes) under soft LED@405 nm irradiation (230 mW cm
−2
)
with low copper catalyst contents (0.1 wt%). The main issues
of the current system are related to the metal-based PIS and its
very high reactivity which requires fast methacrylate resin
Scheme 5 Latent radical-initiated polymerization via 2e
−
/1H
+
transfer
of the MB/DIPEA/DPI
+
system.
79
Leuco MB (LMB) is the latent specie
diffusing in the medium.
Scheme 6 A) Photoredox (Cu(I)*/Iod/Tin(II)) and redox (Cu(I)/R̲’O̲O̲H̲/
Tin(II)) catalytic cycles involved in the production of active radical
species (R
•
). (B) Scheme of the experimental set-up used for lateral
photopolymerization (beyond light exposure). Scheme reproduced from
Macromolecules, 2017.
80
Review Polymer Chemistry
7094 |Polym. Chem.,2017,8,7088–7101 This journal is © The Royal Society of Chemistry 2017
Published on 08 November 2017. Downloaded by Universite Pierre et Marie Curie on 01/12/2017 07:05:50.
View Article Online

application (full conversion in 90 seconds under sunlight
exposure). Also, the copper catalyst is not yet commercially
available. Finally, both composites and clear samples (∼9 cm)
were entirely cured; the setup should therefore be upgraded in
order to record a limit of photopolymerizable sample size.
III.1.5 Dark polymerization in controlled radical photo-
polymerization. Controlled polymerization is of huge impor-
tance in polymer science.
87–91
Recently, light was introduced
into controlled polymerization processes
92–96
as it allows a sig-
nificant reduction of the reaction temperatures, of the energy
consumption and of the environmental costs. Nevertheless
when light absorbing species are involved, the same inner filter
effect (as in classical photopolymerization) is present and
upscaling controlled radical photopolymerization can be
difficult. Light can be useful in controlled radical photo-
polymerization (photoCRP): (i) for the first generation of active
species (through PIS); (ii) for the control of the propagation or
(iii) for the re-activation of dormant species. Indeed, classically,
many authors demonstrate the light control of the polymeriz-
ation by typical light ON/light OFF experiments
97,98
where,
without light, no significant monomer conversion is achieved.
Out of this trend, Boyer and coworkers recently
99
proposed
the interesting idea of generating latent species from a first
irradiation that can be active for dark (irradiation OFF) con-
trolled polymerization: they associated this with energy storage
as first energetic stimuli implied an active process even after
shutting down the light (and energy) source (Scheme 7). The
proposed mechanisms are involving singlet oxygen formed
from photoredox catalyst quenching by molecular oxygen
(under light exposure); this singlet oxygen can result –thanks
to reducing agents –in latent H
2
O
2
produced in the reduction
media. Slowly, without light, H
2
O
2
can be consumed by redu-
cing agents such as ascorbic acid in order to generate active
species. As a result and thanks to latent species, controlled
polymerization experiments can be carried out under air, in
ultralow volumes and using so-called light energy storage.
100
Also, in similar references,
101
light was maintained continu-
ously on and oxygen inhibition produced similar hydroperox-
ides which were consumed through latent redox mechanisms.
These proposed latent mechanisms for photoCRP confirm the
trend observed in conventional photopolymerization that light
propagation issues can be overcome thanks to diffusion of
active species from the irradiated area to inner layers.
III.2 Dark curing in cationic photopolymerization
Compared to free radical polymerization, cationic polymeriz-
ation shows –in general –slower polymerization kinetics (and
sometimes higher induction periods) for the propagation of
cationic species.
102
Intrinsically, cationic species show much
better diffusion constants than radicals as they have higher
lifetimes (e.g. H
+
can be stable). As a result, dark curing in cat-
ionic polymerization is a well-known phenomenon where a
slow post-curing process is occurring thanks to the slow
diffusion of active species in the polymer network.
103–105
Dark curing is also present in cationic
photopolymerization
4,106–109
which –as for other latent
species strategies –results in partial loss of spatial resolution
for the cationic photopolymerization. One way to prevent the
diffusion of latent species in cationic photopolymerization is
to develop highly efficient PIS and epoxy monomers (i.e. acti-
vated monomers such as (3,4-epoxycyclohexane)methyl 3,4-
epoxycyclohexylcarboxylate (Uvacure 1500)): doing so, cationic
photopolymerization shows very good spatial control, for
example, in 3D printing applications.
110–112
The highly
efficient PISs lead to the rapid formation of a rigid polymer
network where latent species diffusion –thus dark curing –is
usually prevented. In contrast, using the same highly reactive
monomer, and during 8 hours at 50 °C; it was only possible to
produce a 5 mm extension thanks to dark curing.
113
Altogether, in cationic photopolymerization, the diffusion
coefficients of the active species (for dark curing) are low and
polymerizations of thick samples –if possible at all –would
take very long times and require high temperatures.
113–117
More interestingly, Yang et al.
115
proposed a two-tempera-
ture approach in order to produce 45 mL (roughly 5 cm) of
thick pigmented composite (the mechanism in Scheme 8)
from the cationic photopolymerization of unactivated epoxy
monomers: the study takes advantage of the long induction
period for the cationic photopolymerization at low temperature
(−20 °C set during 10 minutes) in order to produce a signifi-
cant amount of latent species (oxonium cations) close to the
irradiation source (UV lamp >200 nm, 1 mW cm
−2
). The
medium is stirred, and latent species can spread throughout
the entire polymerizable medium. Then, the temperature was
set to 25 °C and latent oxonium cations –at this temperature –
were very active for the polymerization of the entire polymeriz-
able medium (that takes 10 to 20 minutes).
The main disadvantage of this strategy is the experimental
setup that is not likely to be industrially applicable (cooling,
stirring, re-heating, etc.); nevertheless, the underlying idea of
Fig. 6 Polymerization beyond the irradiated area (protocol from
Aguirre-Soto et al.
79
) Observed for CQ = 0.4%; EDB = 2.2% (dark yellow,
almost no extension); Cu(I) = 0.17 wt%, Iod = 2.0 wt% with 1.3 wt% Tin(II)
(purple) and 8.0 wt% Tin(II) (red). “JACS 2014”: MB/DIPEA/DPI
+
.
79
Figure reproduced from Macromolecules, 2017.
80
Polymer Chemistry Review
This journal is © The Royal Society of Chemistry 2017 Polym. Chem.,2017,8,7088–7101 | 7095
Published on 08 November 2017. Downloaded by Universite Pierre et Marie Curie on 01/12/2017 07:05:50.
View Article Online

latent species (taking advantage of induction times) is very
interesting for cationic photopolymerization.
III.3 Dual-curing systems: light induced acceleration of slow
polymerization processes
At the end of the present review, systems showing light
induced accelerations of slow underlying polymerization pro-
cesses will be considered. First, light induced accelerations/
improvements of redox polymerization systems will be con-
sidered. Then, an open discussion will be held about the high
temperatures present during (photo)polymerization of thick
samples as a result of the conversion of high masses of mono-
mers (i.e. more than 100 °C reached for redox initiated FRP of
4mm–2g–samples
86
). High temperatures can indeed easily
propagate and help in the polymerization propagation
throughout the entire polymerizable medium as shown in
frontal polymerization. Quite importantly, UV lamps such as
xenon lamps or mercury arc lamps emit significant heat which
can enhance the efficiencies in many cases, particularly for
long irradiation times.
III.3.1 A dark curing system is a redox polymerization
process. Redox polymerization is a two component approach
where a reducing agent (or system
118
) is mixed with an oxidiz-
ing agent (or system) generating active species that allow the
polymerization of the surrounding media.
119,120
Many of the
current oxidizing agents show low bond dissociation energies
which allows reactions to occur at room temperature, contrary
to thermal polymerization. One of the most efficient references
for redox polymerization is based on reaction of N-aromatic
amines with benzoyl peroxide (BPO) and was discovered in the
1950s!
121,122
It is still applied to many polymerization
processes
119,120,123–126
and can find use in many high-tech
Scheme 7 Light induced energy storage in controlled radical photopolymerization resulting in a dark polymerization process as presented by
Boyer and coworkers.
99
PC: photocatalyst. Reproduced from Macromolecules.
Scheme 8 Proposed two-temperature strategy for the cationic photo-
polymerization of a thick filled pigmented sample of 1,4-butanediol
diglycidyl ether (BDE).
Review Polymer Chemistry
7096 |Polym. Chem.,2017,8,7088–7101 This journal is © The Royal Society of Chemistry 2017
Published on 08 November 2017. Downloaded by Universite Pierre et Marie Curie on 01/12/2017 07:05:50.
View Article Online

applications requiring in-depth polymerization.
127–131
Nevertheless, for this reference and for many redox systems,
three strong issues are still remaining: (i) first, a thick oxygen
inhibited layer (e.g. >70 µm for the 4-N,N-trimethylaniline/BPO
redox system
38
) is present for many redox systems (Scheme 9);
(ii) once the two component mixing is done, no fast/slow time
control is possible and (iii) mechanical properties of the redox
polymer can be improved.
132,133
A hybrid strategy between photopolymerization (Scheme 10)
and redox polymerization can therefore be proposed and is
referred to as photoactivated redox polymerization. Indeed,
light activation of redox systems thanks to PIS is interesting as:
(i) the efficiency of the PIS should highly increase surface con-
version (where the light absorption is maximum); (ii) an accel-
eration of the slow redox process is possible on demand and
(iii) the global conversion can be improved (more active
species generated).
In dental materials, dual-curing systems (mixing redox and
photoinitiating systems) were studied by Mehmood et al.
132
It
appears that for 5 out of 10 commercial resins, the flexural pro-
perties of the final dental composite were highly enhanced by
the light activation (no significant changes for the remaining
5) compared to the redox curing system. Similarly, for sub-
sequent researchers, light activations in dual curing systems
have generally positive but variable effects on the final pro-
perties of the dental material.
133–135
The usefulness of dual redox/PIS strategies was also studied
by Allonas et al.
136
who showed the possibility to generate
thick samples in free radical acrylic polymerization (20 mm
thick, 50 g). Diphenyl(2,4,6-trimethylbenzoyl)phosphine oxide
(TPO) was the PIS and a redox cobalt/peroxide system was
used. Without light, the system exhibits a mild exothermicity
of 100 °C after 150 minutes but under light irradiation
(420 mW cm
−2
UV irradiation of a Hg/Xe lamp filtered at
365 nm), a completely more efficient (210 °C) curing was poss-
ible after 35 minutes. Another report
38
indicates the use of
amine/phosphines/iodonium charge transfer complexes as
visible light PIS and amine/BPO as a redox system. In this
report, the underlying slow redox system took about 350
seconds to reach 84% CvC methacrylate conversion when a
light (LED@405 nm, 110 mW cm
−2
) accelerated system
reached 89% CvC conversion in about 50 seconds. Raman
confocal microscopy also enabled the characterization of the
improved curing at the very surface.
Another redox initiating system developed by Garra et al.
28
involved the following two-component redox system: a 2-di-
phenylphosphino benzoic acid (2dppba) system mixed with
common copper acetylacetonate (Cu(acac)
2
). Remarkably, the
Cu(acac)
2
/2dppba/Iod dual-curing system could be photoacti-
vated under visible or even NIR light. In the same study, the
Cu(acac)(phen)/2dppba dual-curing system was highly
enhanced by light activation especially at the surface (tack-free
for the dual curing system vs. strongly inhibited layer (>60 µm)
for the dark curing system).
In contrast, light activation (PIS = Cu(I)*/Iod) of a recently
developed Cu(I)/ascorbic acid/benzoyl peroxide redox system
showed very different results with about 82 °C reached for the
light activated system and still 70 °C for the dark curing
systems (rather similar times to reach full exothermicity).
86
The mild effect of light was attributed to a very poor light
penetration throughout the 4 mm sample (the Cu(I) used
showing high extinction coefficients).
Dual-curing thanks to light activation of underlying redox
processes is very promising but the efficiency of the light acti-
vation can be very different according to the: (i) PIS, (ii) redox
systems and (iii) characteristics of the polymerizable media.
First, the PIS/redox systems must be stable in two separate
components; second, redox polymerization systems must be
slow enough for the application; and finally, light penetration
during light activation is not guaranteed throughout the entire
polymerizable medium, which results in variable light acti-
vation improvements.
III.3.2 Light induced temperature increases allowing spatial
propagation of polymerization processes. High irradiances can
result in local temperature increases that are strongly related
to the experimental setup. Then, heat dissipation is inherently
a propagation phenomenon which results in increasing temp-
erature throughout the polymerizable medium. Also, high
polymerizing masses lead to large temperature increases.
106
In
general, temperature increases enable the acceleration of
Scheme 9 Simplified strategies for redox initiated polymerization from
the two component mixing of reducing and oxidizing systems.
Scheme 10 The main features of photopolymerization, photoactivated REDOX polymerization (this study) and redox polymerization under air.
Reproduced from Macromolecules.
28
Polymer Chemistry Review
This journal is © The Royal Society of Chemistry 2017 Polym. Chem.,2017,8,7088–7101 | 7097
Published on 08 November 2017. Downloaded by Universite Pierre et Marie Curie on 01/12/2017 07:05:50.
View Article Online

polymerization processes and are used in frontal
polymerization.
137–142
Thermal initiators in the polymerizable
media strongly help in frontal polymerizations.
For example, Zhang et al.
13,143
proposed a laser induced
thermal frontal polymerization strategy (Scheme 11): a 10 s
short laser (C-60 infrared laser, a high irradiance of 157 W
cm
−2
) leads to a very high temperature increase at the surface
(∼118 °C). In the polymerizable media, an amine/ammonium
persulfate system (not highly active at RT) is turned ON by this
thermal stimuli and radicals (R
•
) are generated, which leads to
the (exothermic) polymerization of the area close to the
surface. The exothermicity of this reaction is dissipated to
deeper polymerizable layers that contain an amine/ammonium
persulfate system activated by the high temperature and so on.
Again, the size of the sample is crucial: 15 mm diameter
induces high masses and polymerization exothermicity will be
very high, which enhances heat generation and propagation.
Also, thermal imaging was used in these studies: it is a very
convenient way to follow polymerization propagation.
Frontal polymerization through heat propagation has also
been described in the literature for pyridinium/hydroperoxide
systems in cationic polymerization.
144,145
Hydroperoxides are
proposed to be used as weak acids in these studies. Six type K
thermocouples were placed in the 50 mm polymerizable
media (12 mm diameter) for vinylether cationic polymeriz-
ations in order to see the polymerization propagation.
145
Three
thermocouples were used for the 20 mm frontal polymeriz-
ation of activated epoxides (Uvacure 1500, 12 mm diameter
samples)
144
under a UV irradiation (700 mW cm
−2
). Again,
high masses of the samples used enhanced the thermal
frontal propagation. The ability of increasing temperatures to
decompose thermal initiators has been discussed in many
studies.
71,76,77
Interestingly, in photodynamic therapies,
146
new strategies involve NIR light irradiation (805 nm, 1 W cm
−2
)
absorbed by a NIR dye (cypate), which generates temp-
erature increases responsible for H
2
O
2
cleavage: the radicals
generated decompose polymer vesicles at the designated areas
(such as tumors) for targeted drug releases. Light induced
temperature increases are generally very convenient for
increasing the thermal generation of radicals.
Altogether, photoinduced thermal frontal polymerization is
averyefficient way to generate photopolymerization propagation
in depth (thus extremely thick samples). Nevertheless a very
careful engineering choice of the masses involved has to be
carried out for each application: too high masses will lead to
final polymers with high temperature damage,
71
and obviously,
heat dissipation in low masses or thicknesses of samples will not
allow the frontal propagation. Also, the introduction of fillers in
composites is very likely to dissipate the heat generated, thus
shutting down the polymerization front. Finally, many of the
thermal initiators used in thermal frontal polymerization (i.e.
BPO) are not stable for long term storage in resins at RT.
IV. Conclusion
The photopolymerization processes are currently associated
with thin samples exhibiting good light penetration (coatings,
very thin inks or paints, etc.). However, the photo-
polymerization of very thick films or even in shadow areas –
where limited light (in depth) or even no light is present –
remains a huge challenge with important academic and indus-
trial potentials (e.g. for the access to filled, pigmented, and dis-
persed samples in composites). In the present review, different
strategies were presented. We propose to divide them into (i)
strategies showing complete spatial and temporal control and
(ii) strategies showing partial to complete loss of spatial and/or
temporal control. For some of these strategies, important devel-
opments are taking place, for example: (i) NIR photoinitiating
systems as these wavelengths show a better penetration in com-
posites; (ii) redox photoactivated systems (two components) or
light induced latent species diffusion (one component) with
redox mechanisms generated in situ; (iii) a soft use of the
thermal propagating effects in photopolymerization reactions.
As a general conclusion, the last 10 years have seen more and
more reports interested in increasing the photopolymerization
thicknesses from coatings (<50 µm) up to tens of centimeters.
One can foresee that during the next 10–15 years, implemen-
tation of the developed strategies in the polymer industry will
extend the applications of photopolymerization to brand new
classes of products in dentistry, composites, 3D printing, bio-
materials, plastic manufacturing, etc.
Conflicts of interest
There are no conflicts to declare.
Acknowledgements
Authors wish to thank the Agence Nationale de la Recherche
for the grants “PhotoRedox”and “FastPrinting”.
Scheme 11 Laser ignited thermal frontal polymerization (heat (Δ)
propagation) as described by Zhang et al.
13,143
for the amine/ammonium
persulfate (APS) thermal system.
Review Polymer Chemistry
7098 |Polym. Chem.,2017,8,7088–7101 This journal is © The Royal Society of Chemistry 2017
Published on 08 November 2017. Downloaded by Universite Pierre et Marie Curie on 01/12/2017 07:05:50.
View Article Online

References
1 J. Lalevée and J.-P. Fouassier, Dyes and Chomophores in
Polymer Science, John Wiley & Sons, 2015.
2 J.-P. Fouassier and J. Lalevée, Photoinitiators for Polymer
Synthesis: Scope, Reactivity, and Efficiency, John Wiley &
Sons, 2012.
3 W. Rutsch, K. Dietliker, D. Leppard, M. Köhler, L. Misev,
U. Kolczak and G. Rist, Prog. Org. Coat., 1996, 27, 227–239.
4 J. V. Crivello and K. Dietliker, Photoinitiators for Free
Radical Cationic & Anionic Photopolymerisation, Wiley,
1999.
5 K. Dietliker, A Compilation of Photoinitiators Commercially
Available for UV Today, SITA Technology Limited, 2002.
6 S. P. Pappas, Radiation Curing: Science and Technology,
Springer Science & Business Media, 2013.
7 S. Dadashi-Silab, S. Doran and Y. Yagci, Chem. Rev., 2016,
116, 10212–10275.
8 W. A. Green, Industrial Photoinitiators: A Technical Guide,
CRC Press, 2010.
9 M.-A. Tehfe, J. Lalevée, S. Telitel, E. Contal, F. Dumur,
D. Gigmes, D. Bertin, M. Nechab, B. Graff, F. Morlet-
Savary and J.-P. Fouassier, Macromolecules, 2012, 45, 4454–
4460.
10 C. Dietlin, S. Schweizer, P. Xiao, J. Zhang, F. Morlet-
Savary, B. Graff, J.-P. Fouassier and J. Lalevée, Polym.
Chem., 2015, 6, 3895–3912.
11 A. Beer, Ann. Phys., 1852, 162,78–88.
12 J. H. Lee, R. K. Prud’Homme and I. A. Aksay, J. Mater.
Res., 2001, 16, 3536–3544.
13 Z.-F. Zhou, C. Yu, X.-Q. Wang, W.-Q. Tang, C.-F. Wang and
S. Chen, J. Mater. Chem. A, 2013, 1, 7326.
14 P. Levêque, J. G. Leprince, S. Bebelman, J. Devaux,
G. Leloup and B. Gallez, J. Magn. Reson., 2012, 220,45–53.
15 F. M. Salomão, B. L. S. Vicentin, E. F. R. Contreras,
M. G. Hoeppner and E. Di Mauro, Mater. Res., 2015, 18,
1023–1028.
16 B. L. S. Vicentin, F. M. Salomão, M. G. Hoeppner and
E. Di Mauro, Appl. Magn. Reson., 2016, 47, 211–222.
17 Y. Cui, J. Yang, Z. Zeng, Z. Zeng and Y. Chen, Eur. Polym.
J., 2007, 43, 3912–3922.
18 R. Liu, H. Chen, Z. Li, F. Shi and X. Liu, Polym. Chem.,
2016, 7, 2457–2463.
19 P. Xiao, J. Lalevée, J. Zhao and M. H. Stenzel, Macromol.
Rapid Commun., 2015, 36, 1675–1680.
20 J. Lalevée, F. Dumur, C. R. Mayer, D. Gigmes, G. Nasr,
M.-A. Tehfe, S. Telitel, F. Morlet-Savary, B. Graffand
J. P. Fouassier, Macromolecules, 2012, 45, 4134–4141.
21 J. Fouassier, X. Allonas and D. Burget, Prog. Org. Coat.,
2003, 47,16–36.
22 C. Grotzinger, D. Burget and J. P. Fouassier, Trends
Photochem. Photobiol., 2003, 71–84.
23 C. F. Bohren and D. R. Huffman, Absorption and Scattering
of Light by Small Particles, John Wiley & Sons, 2008.
24 M. Uo, E. Kudo, A. Okada, K. Soga and Y. Jogo,
J. Photopolym. Sci. Technol., 2009, 22, 551–554.
25 D. Han, Z. Meng, D. Wu, C. Zhang and H. Zhu, Nanoscale
Res. Lett., 2011, 6, 457.
26 S. Shanmugam, J. Xu and C. Boyer, Angew. Chem., Int. Ed.,
2016, 55, 1036–1040.
27 N. Corrigan, J. Xu and C. Boyer, Macromolecules, 2016, 49,
3274–3285.
28 P. Garra, F. Dumur, F. Morlet-Savary, C. Dietlin,
J. P. Fouassier and J. Lalevée, Macromolecules, 2016, 49,
6296–6309.
29 C. Schmitz, A. Halbhuber, D. Keil and B. Strehmel, Prog.
Org. Coat., 2016, 100,32–46.
30 T. Brömme, C. Schmitz, N. Moszner, P. Burtscher,
N. Strehmel and B. Strehmel, ChemistrySelect, 2016, 1,
524–532.
31 M. Jin, J. Xie, J.-P. Malval, A. Spangenberg, O. Soppera,
D.-L. Versace, T. Leclerc, H. Pan, D. Wan, H. Pu,
P. Baldeck, O. Poizat and S. Knopf, J. Mater. Chem. C,
2014, 2, 7201–7215.
32 P. Xiao, F. Dumur, B. Graff, D. Gigmes, J. P. Fouassier and
J. Lalevée, Macromol. Rapid Commun., 2013, 34, 1452–
1458.
33 P. Garra, F. Dumur, F. Morlet-Savary, C. Dietlin,
D. Gigmes, J. P. Fouassier and J. Lalevée, J. Polym. Sci.,
Part A: Polym. Chem., 2017, 55, 3646–3655.
34 D. C. Neckers and W. Jager and SITA Technology
Limited, Chemistry & Technology for UV & EB Formulation
for Coatings, Inks & Paints, Photoinitiation for
Polymerization: UV & EB at the Millenium, Wiley,
Chichester, 1999.
35 N. J. Green, S. M. Pimblott and M. Tachiya, J. Phys. Chem.,
1993, 97, 196–202.
36 N. J. Turro, Modern Molecular Photochemistry, University
Science Books, U.S.A., Sausalito, Calif., New edn, 1991.
37 K. J. Laidler, Chemical Kinetics of Excited States, ASIN:
B00JCVJ2QG, 1955.
38 P. Garra, F. Morlet-Savary, C. Dietlin, J. P. Fouassier and
J. Lalevée, Macromolecules, 2016, 49, 9371–9381.
39 J.-P. Fouassier, F. Morlet-Savary, J. Lalevée, X. Allonas and
C. Ley, Materials, 2010, 3, 5130–5142.
40 R. T. Ingwall and D. Waldman, in Holographic data
storage, Springer, 2000, pp. 171–197.
41 Y.-N. Hsiao, K. Y. Hsu and S. H. Lin, Opt. Mem. Neural
Netw., 2009, 18, 253–259.
42 A. K. O’Brien and C. N. Bowman, Macromolecules, 2003,
36, 7777–7782.
43 T. Guimarães, L. F. Schneider, R. R. Braga and
C. S. Pfeifer, Dent. Mater., 2014, 30, 1274–1279.
44 L. F. J. Schneider, L. M. Cavalcante, S. A. Prahl,
C. S. Pfeifer and J. L. Ferracane, Dent. Mater., 2012, 28,
392–397.
45 D. Axelrod, D. E. Koppel, J. Schlessinger, E. Elson and
W. W. Webb, Biophys. J., 1976, 16, 1055–1069.
46 G. A. Miller, L. Gou, V. Narayanan and A. B. Scranton,
J. Polym. Sci., Part A: Polym. Chem., 2002, 40, 793–808.
47 V. Narayanan and A. B. Scranton, Trends Polym. Sci., 1997,
5, 415–419.
Polymer Chemistry Review
This journal is © The Royal Society of Chemistry 2017 Polym . Chem.,2017,8,7088–7101 | 7099
Published on 08 November 2017. Downloaded by Universite Pierre et Marie Curie on 01/12/2017 07:05:50.
View Article Online

48 V. Narayanan, K. K. Baikerikar and A. B. Scranton, A study
of initiating systems for photopolymerization of composites,
Radtech Int Northamerica, Northbrook, 1998.
49 C. Decker, K. Zahouily, D. Decker, T. Nguyen and T. Viet,
Polymer, 2001, 42, 7551–7560.
50 J.-P. Fouassier and J. F. Rabek, Radiation Curing in Polymer
Science and Technology, Springer Science & Business
Media, 1993.
51 S. Asmussen, G. Arenas, W. D. Cook and C. Vallo, Eur.
Polym. J., 2009, 45, 515–522.
52 M. Bouzrati-Zerelli, N. Zivic, F. Dumur, D. Gigmes,
B. Graff, J. P. Fouassier and J. Lalevée, Polym. Chem., 2017,
8, 2028–2040.
53 M. Bouzrati-Zerelli, J. Kirschner, C. P. Fik, M. Maier,
C. Dietlin, F. Morlet-Savary, J. P. Fouassier, J.-M. Becht,
J. E. Klee and J. Lalevée, Macromolecules, 2017, 50, 6911–
6923.
54 M. W. Grabowski, K. M. Vogelhuber, J. H. Chen and
R. R. McLeod, Absorption and bleaching dynamics of
initiator in thick photopolymer exposed to Gaussian illumi-
nation, ed. S. Orlic, Proc. SPIE 7053, 2008, p. 70530D.
55 N. Hayki, L. Lecamp, N. Désilles and P. Lebaudy,
Macromolecules, 2010, 43, 177–184.
56 J. Zhou, Q. Liu, W. Feng, Y. Sun and F. Li, Chem. Rev.,
2015, 115, 395–465.
57 R. Y. Tsien, Annu. Rev. Biochem., 1998, 67, 509–544.
58 H. Kobayashi, M. Ogawa, R. Alford, P. L. Choyke and
Y. Urano, Chem. Rev., 2010, 110, 2620–2640.
59 J.-C. G. Bünzli, Chem. Rev., 2010, 110, 2729–2755.
60 C. P. Montgomery, B. S. Murray, E. J. New, R. Pal and
D. Parker, Acc. Chem. Res., 2009, 42, 925–937.
61 J.-C. G. Bünzli and C. Piguet, Chem. Soc. Rev., 2005, 34,
1048–1077.
62 A. Al Mousawi, F. Dumur, P. Garra, J. Toufaily, T. Hamieh,
F. Goubard, T.-T. Bui, B. Graff, D. Gigmes, J. Pierre
Fouassier and J. Lalevée, J. Polym. Sci., Part A: Polym.
Chem., 2017, 55, 1189–1199.
63 S. Wu, J. P. Blinco and C. Barner-Kowollik, Chem. –Eur. J.,
2017, 23, 8325–8332.
64 K. Soga, A. Okada and M. Yamada, J. Photopolym. Sci.
Technol., 2006, 19,45–48.
65 A. Stepuk, D. Mohn, R. N. Grass, M. Zehnder,
K. W. Kramer, F. Pelle, A. Ferrier and W. J. Stark, Dent.
Mater., 2012, 28, 304–311.
66 R. Liu, X. Zou, Y. Xu, X. Liu and Z. Li, Chem. Lett., 2016,
45, 1054–1056.
67 Z. Chen, D. Oprych, C. Xie, C. Kütahya, S. Wu and
B. Strehmel, ChemPhotoChem, 2017, 1, 499–503.
68 M. K. Darani, S. Bastani, M. Ghahari, P. Kardar and
E. Mohajerani, Prog. Org. Coat., 2017, 104,97–103.
69 H. Salmi, X. Allonas, C. Ley, D. Marechai and A. Ak,
J. Photopolym. Sci. Technol., 2012, 25, 147–151.
70 M. Bouzrati-Zerelli, M. Frigoli, F. Dumur, B. Graff,
J. P. Fouassier and J. Lalevée, Polymer, 2017, 124, 151–156.
71 M. He, X. Huang, Z. Zeng and J. Yang, Macromolecules,
2013, 46, 6402–6407.
72 J. Sun, Y. Li and K. Hasegawa, J. Loss Prev. Process Ind.,
2001, 14, 331–336.
73 Arkema Inc., Luperox® Organic Peroxides Safe Handling
Guide - their-safe-handling-and-use, open access technical
data, 2007.
74 R. Xu, X. Guan, M. He and J. Yang, RSC Adv., 2017, 7, 914–
918.
75 M.-H. He, M.-L. Su, Z.-H. Yu, G.-X. Chen, Z.-H. Zeng and
J.-W. Yang, Chin. Chem. Lett., 2015, 26,21–25.
76 M. He, S. Jiang, R. Xu, J. Yang, Z. Zeng and G. Chen,
J. Polym. Sci., Part A: Polym. Chem., 2014, 52, 1560–1569.
77 M. He, G. Chen, X. Huang, R. Xu, Z. Zeng and J. Yang,
Polym. Chem., 2014, 5, 2951.
78 M.-H. He, R.-X. Xu, G.-X. Chen, Z.-H. Zeng and J.-W. Yang,
Chin. Chem. Lett., 2014, 25, 1445–1448.
79 A. Aguirre-Soto, C.-H. Lim, A. T. Hwang, C. B. Musgrave
and J. W. Stansbury, J. Am. Chem. Soc., 2014, 136, 7418–
7427.
80 P. Garra, F. Dumur, D. Gigmes, A. Al Mousawi, F. Morlet-
Savary, C. Dietlin, J. P. Fouassier and J. Lalevée,
Macromolecules, 2017, 50, 3761–3771.
81 P. Xiao, F. Dumur, J. Zhang, J. P. Fouassier, D. Gigmes
and J. Lalevée, Macromolecules, 2014, 47, 3837–3844.
82 A. Baralle, L. Fensterbank, J.-P. Goddard and C. Ollivier,
Chem. –Eur. J., 2013, 19, 10809–10813.
83 A. A. Mousawi, A. Kermagoret, D.-L. Versace, J. Toufaily,
T. Hamieh, B. Graff, F. Dumur, D. Gigmes, J. P. Fouassier
and J. Lalevée, Polym. Chem., 2016, 8, 568–580.
84 B. Husár, S. C. Ligon, H. Wutzel, H. Hoffmann and
R. Liska, Prog. Org. Coat., 2014, 77, 1789–1798.
85 S. C. Ligon, B. Husár, H. Wutzel, R. Holman and R. Liska,
Chem. Rev., 2014, 114, 557–589.
86 P. Garra, A. Kermagoret, A. Mousawi, F. Dumur,
D. Gigmes, F. Morlet-Savary, C. Dietlin, J. P. Fouassier and
J. Lalevée, Polym. Chem., 2017, 8, 4088–4097.
87 K. Matyjaszewski and J. Xia, Chem. Rev., 2001, 101, 2921–
2990.
88 W. A. Braunecker and K. Matyjaszewski, Prog. Polym. Sci.,
2007, 32,93–146.
89 M. Szwarc, Nature, 1956, 178, 1168–1169.
90 O. W. Webster, Science, 1991, 251, 887–893.
91 C. Barner-Kowollik, T. P. Davis, J. Heuts, M. H. Stenzel,
P. Vana and M. Whittaker, J. Polym. Sci., Part A: Polym.
Chem., 2003, 41, 365–375.
92 M. A. Tasdelen, M. Çiftci, M. Uygun and Y. Yagci, in
Progress in controlled radical polymerization: mechanisms
and techniques, ACS Publications, 2012, pp. 59–72.
93 C.-H. Lim, M. D. Ryan, B. G. McCarthy, J. C. Theriot,
S. M. Sartor, N. H. Damrauer, C. B. Musgrave and
G. M. Miyake, J. Am. Chem. Soc., 2017, 139, 348–355.
94 N. J. Treat, B. P. Fors, J. W. Kramer, M. Christianson,
C.-Y. Chiu, J. R. de Alaniz and C. J. Hawker, ACS Macro
Lett., 2014, 3, 580–584.
95 B. Wenn, M. Conradi, A. Demetrio Carreiras,
D. M. Haddleton and T. Junkers, Polym. Chem., 2014, 5,
3053–3060.
Review Polymer Chemistry
7100 |Polym. Chem.,2017,8,7088–7101 This journal is © The Royal Society of Chemistry 2017
Published on 08 November 2017. Downloaded by Universite Pierre et Marie Curie on 01/12/2017 07:05:50.
View Article Online

96 X. Pan, M. A. Tasdelen, J. Laun, T. Junkers, Y. Yagci and
K. Matyjaszewski, Prog. Polym. Sci., 2016, 62,73–125.
97 B. Cabannes-Boué, Q. Yang, J. Lalevée, F. Morlet-Savary
and J. Poly, Polym. Chem., 2017, 8, 1760–1770.
98 Q. Yang, F. Dumur, F. Morlet-Savary, J. Poly and J. Lalevée,
Macromolecules, 2015, 48, 1972–1980.
99 S. Shanmugam, J. Xu and C. Boyer, Macromolecules, 2017,
50, 1832–1846.
100 S. Shanmugam, J. Xu and C. Boyer, Macromolecules, 2016,
49, 9345–9357.
101 J. Yeow, R. Chapman, J. Xu and C. Boyer, Polym. Chem.,
2017, 8, 5012–5022.
102 R. Bongiovanni and M. Sangermano, Encyclopedia of
Polymer Science and Technology, 2014.
103 E. W. Nelson, J. L. Jacobs, A. B. Scranton, K. S. Anseth and
C. N. Bowman, Polymer, 1995, 36, 4651–4656.
104 V. Sipani and A. B. Scranton, J. Photochem. Photobiol., A,
2003, 159, 189–195.
105 U. Bulut and J. V. Crivello, Macromolecules, 2005, 38,
3584–3595.
106 B. Falk, S. M. Vallinas and J. V. Crivello, J. Polym. Sci., Part
A: Polym. Chem., 2003, 41, 579–596.
107 B. Golaz, V. Michaud, Y. Leterrier and J.-A. E. Månson,
Polymer, 2012, 53, 2038–2048.
108 V. Sipani and A. B. Scranton, J. Polym. Sci., Part A: Polym.
Chem., 2003, 41, 2064–2072.
109 M. Sangermano, Pure Appl. Chem., 2012, 84, 2089–2101.
110 A. Al Mousawi, F. Dumur, P. Garra, J. Toufaily, T. Hamieh,
B. Graff, D. Gigmes, J. P. Fouassier and J. Lalevée,
Macromolecules, 2017, 50, 2747–2758.
111 A. Al Mousawi, D. M. Lara, G. Noirbent, F. Dumur,
J. Toufaily, T. Hamieh, T.-T. Bui, F. Goubard, B. Graff,
D. Gigmes, J. P. Fouassier and J. Lalevée, Macromolecules,
2017, 50, 4913–4926.
112 A. Ambrosi and M. Pumera, Chem. Soc. Rev., 2016, 45,
2740–2755.
113 B. A. Ficek, A. M. Thiesen and A. B. Scranton, Eur. Polym.
J., 2008, 44,98–105.
114 M. Sangermano, S. Pegel, P. Pötschke and B. Voit,
Macromol. Rapid Commun., 2008, 29, 396–400.
115 L. Yang, J. Yang, J. Nie and X. Zhu, RSC Adv., 2017, 7,
4046–4053.
116 C. C. Hoppe, B. A. Ficek, H. S. Eom and A. B. Scranton,
Polymer, 2010, 51, 6151–6160.
117 G. Terrones and A. J. Pearlstein, Macromolecules, 2004, 37,
1565–1575.
118 P. Garra, F. Dumur, A. A. Mousawi, B. Graff, D. Gigmes,
F. Morlet-Savary, C. Dietlin, J. P. Fouassier and J. Lalevée,
Polym. Chem., 2017, 8, 5884–5896.
119 G. S. Misra and U. D. N. Bajpai, Prog. Polym. Sci., 1982, 8,
61–131.
120 A. S. Sarac, Prog. Polym. Sci., 1999, 24, 1149–1204.
121 L. Horner and K. Scherf, Justus Liebigs Ann. Chem., 1951,
573,35–55.
122 J. Lal and R. Green, J. Polym. Sci., 1955, 17, 403–409.
123 H. Yang, L. Zhang, W. Ma, H. Pu and F. Gong, J. Appl.
Polym. Sci., 2015, 132, 41567.
124 A. Zoller, D. Gigmes and Y. Guillaneuf, Polym. Chem.,
2015, 6, 5719–5727.
125 Y. Zhang, B. Gao, L. Gu and X. Zhao, Acta Polym. Sin.,
2012, 264–271.
126 M. Worzakowska, J. Therm. Anal. Calorim., 2011, 105, 987–
994.
127 T.-Y. Kwon, R. Bagheri, Y. K. Kim, K.-H. Kim and
M. F. Burrow, J. Investig. Clin. Dent., 2012, 3,3–16.
128 I. D. Sideridou, D. S. Achilias and N. C. Kostidou, J. Appl.
Polym. Sci., 2008, 109, 515–524.
129 I. D. Sideridou, D. S. Achilias and O. Karava,
Macromolecules, 2006, 39, 2072–2080.
130 D. S. Achilias and I. Sideridou, J. Macromol. Sci., Part A:
Pure Appl.Chem., 2002, 39, 1435–1450.
131 G. O. Wilson, J. W. Henderson, M. M. Caruso,
B. J. Blaiszik, P. J. McIntire, N. R. Sottos, S. R. White and
J. S. Moore, J. Polym. Sci., Part A: Polym. Chem., 2010, 48,
2698–2708.
132 H. Lu, A. Mehmood, A. Chow and J. M. Powers, J. Prosthet.
Dent., 2005, 94, 549–554.
133 L. Feng and B. I. Suh, J. Biomed. Mater. Res., Part B, 2006,
76, 196–202.
134 M. Okuma, M. Nakajima, K. Hosaka, S. Itoh, M. Ikeda,
R. M. Foxton and J. Tagami, Dent. Mater. J., 2010, 29, 642–
648.
135 L. A. Jongsma, N. de Jager, C. J. Kleverlaan, P. Pallav and
A. J. Feilzer, Dent. Mater., 2012, 28, 1080–1088.
136 X. Allonas, A. Ibrahim, V. Charlot, M. Retailleau, F. Karasu
and C. Croutxé-Barghorn, J. Photopolym. Sci. Technol.,
2015, 28,25–29.
137 J. A. Pojman, G. Curtis and V. M. Ilyashenko, J. Am. Chem.
Soc., 1996, 118, 3783–3784.
138 N. M. Chechilo, E. Ns and K. Ry, Dokl. Akad. Nauk SSSR,
1972, 204, 1180.
139 M. Bazile, H. A. Nichols, J. A. Pojman and V. Volpert,
J. Polym. Sci., Part A: Polym. Chem., 2002, 40, 3504–3508.
140 M. F. Perry, V. A. Volpert, L. L. Lewis, H. A. Nichols and
J. A. Pojman, Macromol. Theory Simul., 2003, 12, 276–
286.
141 X. Cai, S. Chen and L. Chen, J. Polym. Sci., Part A: Polym.
Chem., 2008, 46, 2177–2185.
142 Y. Fang, L. Chen and S. Chen, J. Polym. Sci., Part A: Polym.
Chem., 2009, 47, 1136–1147.
143 W. Zhang, S. Yang, C.-F. Wang and S. Chen,
Macromolecules, 2015, 48, 5543–5549.
144 M. Lecompère, X. Allonas, D. Maréchal and A. Criqui,
Polym. Chem., 2017, 8, 388–395.
145 M. Lecompère, X. Allonas, D. Maréchal and A. Criqui,
Macromol. Rapid Commun., 2017, 1600660.
146 J. Li, K. Wei, S. Zuo, Y. Xu, Z. Zha, W. Ke, H. Chen and
Z. Ge, Adv. Funct. Mater., 2017, 27, 1702108.
Polymer Chemistry Review
This journal is © The Royal Society of Chemistry 2017 Polym. Chem.,2017,8,7088–7101 | 7101
Published on 08 November 2017. Downloaded by Universite Pierre et Marie Curie on 01/12/2017 07:05:50.
View Article Online