High efficiency photoinitiators with extremely low concentration based on furans derivative

Article abstract of DOI:10.1016/j.jphotochem.2020.112994

In this paper, two type II photoinitiators (1E,4E)1,5-di(furan-2-yl) penta-1,4-dien-3-one (DFP) and (E)-1,3-di(furan-2-yl) prop-2-en-1-one (DFP-e) were synthesized via Claisen-Schmidt reaction. The maximum peak of absorption could reach about 360 nm and the molar extinction coefficient as high as 4 × 10⁴ M^?1 cm^?1, it could initiate photopolymerization of HDDA and PEGDA at very low concentration, and the photopolymerization efficiency is much higher than that of ITX at the same concentration. In addition, the synthesized initiators show higher photopolymerization efficiency for PEGDA than that of ITX in the absence of co-initiators as well. On the other hand, the initiators had high rate of photo-bleach exposure upon LED lamp, it could be appropriate for light color and colorless system. With the properties, these initiators could be a promising candidate for UV LED photopolymerization.

Full text of DOI:10.1016/j.jphotochem.2020.112994

Journal Pre-proof
High Efficiency Photoinitiators with Extremely Low Concentration Based
on Furans Derivative
Fan Yang (Conceptualization) (Methodology) (Software) (Data
curation) (Writing - original draft) (Software), Meijuan Zhong,
Xiangchen Zhao, Jingfang Li, Fang Sun (Supervision), Jun Nie
(Supervision), Xiaoqun Zhu (Supervision)
PII:
S1010-6030(20)30791-7
DOI:
https://doi.org/10.1016/j.jphotochem.2020.112994
JPC 112994
Reference:
To appear in:
Journal of Photochemistry & Photobiology, A: Chemistry
Received Date:
Revised Date:
Accepted Date:
3 September 2020
6 October 2020
20 October 2020
Please cite this article as: Yang F, Zhong M, Zhao X, Li J, Sun F, Nie J, Zhu X, High Efficiency
Photoinitiators with Extremely Low Concentration Based on Furans Derivative, Journal of
Photochemistry and amp; Photobiology, A: Chemistry (2020),
doi: https://doi.org/10.1016/j.jphotochem.2020.112994
This is a PDF file of an article that has undergone enhancements after acceptance, such as
the addition of a cover page and metadata, and formatting for readability, but it is not yet the
definitive version of record. This version will undergo additional copyediting, typesetting and
review before it is published in its final form, but we are providing this version to give early
visibility of the article. Please note that, during the production process, errors may be
discovered which could affect the content, and all legal disclaimers that apply to the journal
pertain.
© 2020 Published by Elsevier.

High Efficiency Photoinitiators with Extremely Low Concentration
Based on Furans Derivative
Fan Yang, Meijuan Zhong, Xiangchen Zhao, Jingfang Li, Fang Sun, Jun Nie, Xiaoqun Zhu*
State Key Laboratory of Chemical Resource Engineering, Beijing University of Chemical
Technology, Beijing, 100029, P.R. China
Graphical Abstract
Highlights


DFP and DFP-e are straightforward synthesized via one step.
the absorption wavelength of DFP and DFP-e could be extend to LED light wavelength region
which is appropriate for LED light resource for free radical photopolymerization.
DFP and DFP-e possess higher molar extinction coefficient that it can be used at extremely low
concentration in photopolymerization system.


ABSTRACT: In this paper, two type II photoinitiators (1E,4E)1,5-di(furan-2-yl) penta-1,4-dien-3-one
(DFP) and (E)-1,3-di(furan-2-yl) prop-2-en-1-one (DFP-e) were synthesized via Claisen-Schmidt
reaction. The maximum peak of absorption could reach about 360nm and the molar extinction coefficient
as high as 4×104 M-1cm-1, it could initiate photopolymerization of HDDA and PEGDA at very low
concentration, and the photopolymerization efficiency is much higher than that of ITX at the same
concentration. In addition, the synthesized initiators show higher photopolymerization efficiency for
PEGDA than that of ITX in the absence of co-initiators as well. On the other hand, the initiators had high
rate of photo-bleach exposure upon LED lamp, it could be appropriate for light color and colorless system.
With the properties, these initiators could be a promising candidate for UV LED photopolymerization.
Keywords: Photopolymerization; Photoinitiator; LED light; Molar extinction coefficient.
1. INTRODUCTION
As a green-conditioning technology, UV-LED curing technique has become an important promoting
factor for green transformation and development in the modern world and widely applied in coating,
adhesives, printing plates, dental filling materials, laser-induced 3D printing, integrated circuit plate,
holographic optics element, nanoscale micromechanics.1-7 Due to the Minamata Convention on Mercury
issued by the United Nations, the mercury lamp is further restricted. As one of the priorities in the
photopolymerization system, the photoinitiators which can absorb LED light and generate active species
to initiate the photopolymerization are supposed to appropriate for emission wavelength of LED light
source.8-11 We all know photoinitiators could be classified into two types of radical or cationic, but the
free radical polymerization still is the most widely used.
At last past few years, much of newly development of LED photoinitiators was carried out. For
instance, Lalevee’s group place emphasis on a series of photoinitiators based on D-A(donor-acceptor)

structure to develop LED light cationic initiate polymerization combined with oxidants, such as
diphenyliodonium hexafluorophosphate(Ph2I+·PF6-)12-15. Furthermore, a series of LED initiation system
based on aromatics with condensed rings including anthracene16, pyrene16 were reported, especially
anthracene which represents a novel contribution for effective solutions to resist oxygen inhibition17. It
is particularly noteworthy the photosensitizer based on aromatic is limited by its solubility and expensive
material costs and use in higher concentrations. In such a case, it is difficult to develop volume production
and commercialization. Thus, a LED-compatible photoinititator synthesized by a lower cost, less
complexity synthesis strategy and lower usage of concentration was required urgently.
In latest research of our group, Tang has developed a higher efficiency LED-compatible
photoinitiator 1,3-bis(1-methyl-1H-pyrrol-2-yl) prop-2-en-1-one (BMO) containing two heterocyclic
nitrogen rings, pyrrole, and synthesized by Claisen-Schmidt reaction in one step. It capable of initiating
free radical photopolymerization when exposed to LEDs (365, 385 and 405 nm) and shows higher molar
extinction coefficient at long wavelength. In addition, it can not only initiate radical photopolymerization
by itself, but also cationic photopolymerization with Iod under low light intensity. This kind of structure
improved our interest and we developed a new photoinitiator based on Tang’s research.
In this paper, we synthesized a series of photoinitiators (DFP, DFP-e) by one-step Claisen-Schmidt
condensation reaction. We introduce the conjugated structure containing furan and bridge double bonds.
Through our results of research, it shows a widely absorption arrange from 325nm to 425nm which well-
matched LEDs wavelength, and gives a high extinction coefficient about 4×104 M-1cm-1 so that it can be
used at very low concentration in radical photopolymerization. And meanwhile, the synthesized
photoinitiators shows higher double bonds final conversion than ITX at the same very low concentration
about 10-4 w% with co-initiator EDAB for HDDA. And we found that both of the synthesized initiators

could initiate polymerization in the absence of EDAB for PEGDA, furthermore, they exhibit much higher
initiation efficiency and final conversion than ITX system at the same condition. Such low levels of
initiation would greatly reduce the cost of industrial application and decrease the risks of possibility
migration. Moreover, all of the synthesized photoinitiators exhibit high rate of photo-bleach effect for
polymerization, thus, both of them can be appropriate for photopolymerization which is required delicate
color or colorless. There can be no doubt that these straightforward synthesized initiators would be
promising candidates for commercialization.
2. EXPERIMENTAL DETAILS
2.1 Materials.
Furfural and 2-acetyl furan were purchased from Energy Chemical (Shanghai, China). Acetone,
NaOH, ethanol, and acetonitrile were purchased from Sinopham Group Chemical Reagent Co. (Beijing,
China). The monomer hexamethylene diacrylate (HDDA) and Polyethylene (glycol) Diacrylate (PEGDA)
were donated by Sartomer (Guangzhou, China), co-initiator Ethyl 4–(N,N–dimethylamino)
benzoate(EDAB) was purchased from Hubei Gurun technology company. 2-Isopropylthioxanthone (ITX)
were obtained from JIURI NEW MATERIAILS (Tianjin, China). All reagents were used as received.
Scheme 1. The chemical structures of photointiators, monomers and additives.
2.2 Synthesis of DFP and DFP-e.

The synthesis route of DFP and DFP-e are displayed in scheme1. As general produce for Claisen-
Schmidt Reaction, 50 mmol of furfural, 30 mmol of acetone (for the synthesis of DFP) or 2-acetyl furan
(for the synthesis of DFP-e) and 50mL of ethanol were put into a three-necked glass flask, then 30
mmol of sodium hydroxide dissolved in 9 mL of ultrapure water was added dropwise. The mixture react
at 60oC until the aldehyde compound was completely consumed (TLC analysis), then filter the
precipitation and washed with saturated NH4Cl solution until the aqueous phase was neutral. Purifying
the crude product via recrystallization from ethanol, the yield is 75%.
Scheme 2. Synthesis route of DFP and DFP-e
2.3 Computational procedure.
Density functional theory (DFT) calculations were carried out with the Gaussian 09 software. The
synthesized photoinitiators DFP and DFP-e were fully optimized in ground state by DFT method with
B3LYP/6-31G(d) level of calculation. The molecular orbital properties were carried out and visualized
with Gaussian View 5.0.8 software18.
2.4 Fluorescence experiment.
The fluorescence properties in acetonitrile were investigated with a fluorescence spectrophotometer
(HITACHIF-7000, Japan). The fluorescence quenching experiment was studied by testing the change of
fluorescence intensity with the equation Stern-Volmer (F0/F=1+KSV*[EDAB]), F0 and F represent the

before and after adding EDAB
2.5 Characterization.
1H-NMR (400MHz) spectra were determined by a Bruker 400MHz spectrometer at room
temperature, the chemical shift (s = singlet, bs = broad singlet, d = doublet, t = triplet, m = multiplet) is
located in ppm with an internal standard. Elemental analysis was performed by a CE44. UV-vis spectra
were recorded using Hitachi U-3010 UV spectrophotometer（Hitachi，Japan）. In the steady state
photolysis experiments, the LED lamp at 365 nm was utilized and the light intensity was settled at 70
mW/cm2.
2.6 Photopolymerization.
The photopolymerization kinetics were followed by using a real time Fourier IR (FT-IR)
spectrometer (Nicolet 5700). The experiment was figured out by dropping out a KBr (2±0.1 mm) crystal
for irradiation. The mixture was covered by another transparent KBr crystal at room temperature and
irradiated by LED@365nm. The light intensity was 70mW/cm2 measured by an ultraviolet-visible
radiometer which was produced by Beijing Normal University, China. Photopolymerization profiles
were characterized by calculation of the decrease in the area of acrylate absorption peak around 812cm-
1 of the double bond conversion. The final conversion of double bond in MIR absorbance is calculated
by the following equation:
FC% =（1-At/A0）*100%
(1)
The A0 and At represent the area of the absorption peak of the double bond (812cm-1) before and
after photopolymerization respectively.
2.7 EPR spin trapping (EPR) experiments.
The EPR spectrum was collected by using the Bruker EMX-plus EPR spectrometer (X-Band). The

radicals generated at room temperature upon a halogen lamp exposure under N2 and trapped by phenyl-
N-tbutylnitrone (PBN), dissolved in tbutylbenzene, according to a procedure described elsewhere in ref19.
The concentration of PBN and photoinitiator were 5.5×10-2 and 6×10-3 mol/L, respectively. The αN and
αH stand for the hyperfine coupling constants in PBN radical adducts for the nitrogen and the hydrogen.
The EPR spectra simulations were found in ref20.
3. RESULTS AND DISCUSSION
3.1 Synthesis and Characterization of initiators.
In order to extend the absorption wavelength of initiator, the furan rings were introduced into
acetone or 2-acetyl furan in one step-synthesis by Claisen-Schmidt reaction and the products were very
easy to purify by simple re-crystallization in ethanol. DFP and DFP-e perform good solubility in usual
organic solvents, such as ethyl acetate, methanol ethanol, acetone, acetonitrile, DMSO and THF,
indicating great compatibility with other constituents of the photopolymerization system. The structure
1
of DFP and DFP-e were identified by H-NMR in D-DMSO solution. Elemental analysis showed that
DFP (C%=72.6%, H%=4.5%) and DFP-e (C%=76.1%, H%=4.6%)
(1E,4E)1,5-di(furan-2-yl) penta-1,4-dien-3-one (DFP). The structure was characterized by 1H-NMR
(400MHz, D-DMSO): δ(ppm) = 7.89(d, J = 1.6Hz, 2H), 7.53 (d, J = 16.4Hz, 2H), 7.01 (d, J = 3.6Hz,
2H), 6.97 (d, J = 16.4Hz, 2H), 6.67 (m, 2H).
(E)-1,3-di(furan-2-yl) prop-2-en-1-one (DFP-e). Purification by recrystallization from ethanol. The
structure was characterized by 1H-NMR (400MHz, D-DMSO): δ(ppm) = 8.06(d, J = 1.6Hz, 1H), 7.92(d,
J = 1.6Hz, 1H), 7.68(d, J = 3.6Hz, 1H), 7.58(d, J = 15.6Hz, 1H), 7.35(d, J = 15.6Hz, 1H), 7.09(d, J =
3.6Hz, 1H), 6.78(m, 1H), 6.7(m, 1H).
3.2 Absorption and Photopolymerization.

The UV-Vis spectra of DFP and DFP-e in acetonitrile were displayed in Fig.1, the absorption
maximum(λmax) and the molar extinction coefficients at λmax are summarized in Table 1. As can be seen
from Table 1, The maxima absorption of DFP and DFP-e were at 360, 343nm respectively, due to the π–
π* transition of one bridge double bond in the conjugated structure. In addition, the extinction coefficient
was as high as over 4×104 M-1cm-1, which was attributed to the electron withdrawing effect of carbonyl
group and conjugated structure. The absorption property of DFP and DFP-e were contributed by the
strong delocalization of the highest occupied molecular orbital (HOMO) and the lowest unoccupied
molecular orbital (LUMO), which are involved in the lowest energy transition. Gaussian 09 software was
used to calculate the lowest energy spatial conformation. The HOMO and LUMO of photoinitiators DFP
and DFP-e after structure optimization by density functional theory at the B3LYP/6-31G level were
presented in Fig. 1. The HOMO-LUMO transitions exhibit an intramolecular charge transfer (ICT)
feature aside from the π–π* transition behavior.

Table 1. The photophysical data of DFP and DFP-e
Photoinitiators
λmaxa(nm)
εmaxb(M-1cm-1)
ε
365c(M-1cm-1)
DFP
360
343
48250
47510
21505
DFP-e
42940
a
b
λmax is the maximum absorption wavelength in acetonitrile solution; ε max is the molar
extinction coefficient at λ max; ε365c is the molar extinction coefficient at λ365nm
Fig. 1 (a)UV-visible spectrum of DFP and DFP-e in acetonitrile; (b) Calculated electron
distributions of HOMOs and LUMOs for photoinitiators.
In order to measure the photo-induce efficiency, DFP and DFP-e were used as the photoinitiator for
the polymerization of HDDA with amine co-initiator EDAB, a commercial initiator ITX was use as
comparison. As showed in Fig.2(a), even though the ITX had higher rate of polymerization at beginning
of irradiation, the final double bond conversion of DFP and DFP-e were 48% and 68% respectively,
which is much higher than ITX (34%), it indicated that DFP and DFP-e had higher photopolymerization
efficiency. As we know, these three initiators were type II photoinitiator, which need hydrogen donor to
produce free radical during UV irradiation, HDDA do not have any hydrogen donor group, thus, EDAB
was necessary. On the other hand, we know that PEGDA have hydrogen donor ability because of the

ether chain, when the two initiators were used to induced photopolymerization of PEGDAwithout EDAB,
it showed clearly in Figure 2(b) that the final double conversion for DFP and DFP-e were 77% and 87%
respectively, which is much higher that of ITX (28%), and also the rate of polymerization of DFP and
DFP-e were much higher than that of ITX. This demonstrated that addition of hydrogen like EDAB is
not necessary for the DFP and DFP-e, but not the case of ITX, we all know that the EDAB or other amine
are toxic compounds, this made DFP and DFP-e a good candidate as initiator to reduce the toxicity.
The EPR results of DFP/EDAB and DFP-e/EDAB indicated that the hyperfine slitting constants
(hfs) of PBN adducts were computed as αN = 14.3 G, αH = 2.4 G and αN = 14.3 G, αH = 2.5 respectively,
both of the initiated systems were assigned to the PBN/aminoalkyl radical adducts21 (Fig.2(c) and
Fig.2(d)). In addition, the maximum signal intensity of DFP was less than that of DFP-e. The results
would lead to explain DFP-e possess higher initiation efficiency than DFP. Hydrogen atom transfer
interaction of both DFP and DFP-e with co-initiator occurs on carbonyl group and generated active
radical species to initiating photopolymerization.

Fig.2 (a) Photopolymerization profiles of HDDA use (1)DFP-e/EDAB; (2)DFP/EDAB;
(3)ITX/EDAB as initiator; (b) Photopolymerization profiles of PEGDA use (1)DFP-e; (2)DFP; ITX
as initiator. Light intensity LED@365nm 70mW/cm2, (initiator/coinitiator=0.0625%/1%, w/w);
(c)EPR spin-trapping spectra of DFP/EDAB and (d)DFP-e/EDAB by using PBN in tert-
butylbenzene upon LED lamp irradiation.
Scheme 3. Reaction Mechanism for DFP and DFP-e
Figure 3 showed the polymerization profiles of HDDA initiated by DFP/EDAB and DFP-e/EDAB,
the results indicated that the maximum double bong conversion could be research at
DFP/EDAB=0.075%/1%; DFP-e/EDAB=0.075%/1%, which indicated that at very low concentration of
initiator, the polymerization could reach high value. It is quite different with the commercial initiator,
normal, the concentration commercial initiator should be as high as 2-4wt%22, then could reach high
double bond conversion. These results indicated that the efficiency of DFP and DFP-e were 2.3 times
higher than a commercial initiator, one of the reasons should be the high molar extinction coefficients of
the DFP and DFP-e.

Fig.3 (a)Photopolymerization profiles of HDDA use (1)DFP-e/EDAB (0.075%/1%, w/w);
(2)DFP/EDAB (0.075%/1%, w/w); (3)DFP-e (0.075%, w); (4)DFP (0.075%, w) as initiator.
(b)Photopolymerization profiles of PEGDA use (1)DFP-e/EDAB (0.075%/1%, w/w);
(2)DFP/EDAB (0.075%/1%, w/w); (3)DFP-e (0.075%, w); (4)DFP (0.075%, w) as initiator. light
intensity LED@365nm 70mW/cm2.
3.3 Steady-state photolysis of DFP and DFP-e characterized by UV-visible absorption
spectroscopy.
The steady-state photolysis experiments with DFP and DFP-e in the presence of the amine were
carried out in acetonitrile upon irradiation with a LED@365 nm. The effects of the amine on the
photolysis of DFP and DFP-e are shown in Fig. 4. Obvious and significant decrease of the UV-visible
absorption intensity was observed for the two-component (DFP/amine and DFP-e/amine) photoinitiating
systems during light irradiation. The characteristic absorption peaks of DFP significantly decreased faster
than that of DFP-e due to its higher molar extinction coefficients at 365 nm (e.g. ε365=47510M-1cm-1 for
DFP, ε365=21505M-1cm-1 for DFP-e, respectively; see Table1).

Fig.4 Photolysis of DFP (a) and DFP-e (b) in the presence of an amine(EDAB, 0.5M) upon a
LED@365 nm under air in acetonitrile solvent; consumption of (c) DFP and (d) DFP-e during the
photolysis experiments.
3.4 Consumption of DFP and DFP-e in photolysis reactions.
According to the UV-visible absorption spectra presented above, the consumption of the DFP and
DFP-e vs. the irradiation time can be summarized. The absorption changes were detected examined for
the two-component initiating system (DFP/amine and DFP-e/amine) (Fig. 4). It obvious from Fig. 4 that
the consumption of DFP was much higher than that observed from DFP-e (e.g. the consumption of DFP
within 6s = 46% for DFP/EDAB vs. 30% for DFP-e/EDAB) due to the higher molar coefficient for DFP.
However, it can also cause the polymerization that only occurred on the surface resulting in a low overall
conversion.

3.5 Fluorescence quenching for electron transfer reaction for DFP and DFP-e.
The fluorescence and UV-visible absorption spectra were recorded in acectonitrile for DFP and
DFP-e and the data are presented in Fig. 5. As a matter of fact, the first singlet excited state energy (ES1)
can be determined by crossing point between the absorption of the UV-vis spectra and fluorescence
spectra (e.g. ES1 = 3.06eV for DFP, 3.39eV for DFP-e; Table 2).
Fig.5 Singlet state energy determination in acetonitrile of (a) DFP and (b) DFP-e.
The fluorescence quenching experiments for DFP and DFP-e were performed in acetonitrile to seek
the theoretical feasibility of the interactions between DFP/amine and DFP-e/amine. The results are shown
in Fig. 623,24
.
Table 2. Parameters characterizing the fluorescence properties of DFP and DFP-e in acetonitrile: the
interaction constant (KSV) between DFP/EDAB and DFP-e/EDAB systems calculated by the Stern-
Volmer equation; electron transfer quantum yield (ΦetEDAB) of DFP/EDAB and DFP-e/EDAB interaction
KSV
Φet
EDA0B
ES1(eV)
DFP
157.87
0.49
3.06
DFP-e
105.67
0.39
3.39
The fluorescence experiments of DFP showed a obvious decrease of the fluorescence intensity with
the addition of amines in the solution, demonstrating that the quantity increased of amine can lead to

interaction with the chromophore in the excited singlet state. For both photoinitiator DFP and DFP-e ,
amine performed a great quencher. The interaction constant (KSV) is obtained from the stern-Volmer plot
according to the well-known equation23,24; the electron transfer quantum yields Φet (EDAB) were
calculated from equation (2) (Table 2).
ΦetEDB = KSV
× [EDAB]/(1 + KSV
× [EDAB])
(2)
EADB
EADB
Significantly, such high electron transfer quantum yields (Φet) are calculated (Table 2). For both
photoinitiator DFP and DFP-e generated lots of active species with amine and high electron transfer
quantum yields.
Fig.6 (a) Fluorescence quenching of DFP by amine (EDAB); (b) Stern-Volmer treatment for
DFP/amine fluorescence quenching; (c) fluorescence quenching of DFP-e by amine (EDAB); (d)
Stern-Volmer treatment for DFP-e/amine fluorescence quenching.

3.6 Photobleaching studies.
When DFP and DFP-e were irradiated by 365nm LED, the absorption peak decrease gradually, as
can be seen from Fig.7(a), during decrease of the absorption peak, the maximum absorption peak blue
shift 40nm, which meant there are other photoreaction exception the photodegradation, according to our
previous research, the photo-isomer occurred as well25.26. but in figure 7(b), there are not any absorption
peak shift during irradiation, but a new peak formed in the short wavelength, which indicated that most
of the photoreaction is photodegradation. The results demonstrated that photo-bleach occurred during
irradiation.
Fig.7 The normalized UV-vis spectral changes during irradiation of (a)DFP and (b)DFP-e in
acetonitrile(2×10-5mol/L) upon LED 365nm exposure.
4. CONCLUSION.
In summary, two straightforward synthesized photoinitiators (DFP and DFP-e) by Claisen-Schmidt
reaction were shown, accompanied with high yield, and the products were purified easily. We found that
both of DFP and DFP-e have the characteristic of initiating ability at very low levels concentration around
10-4wt% exposure upon LED lamp, and show higher photoinitiation efficiency than that of ITX system.
Meanwhile, the synthesized initiators have the ability to occur hydrogen transfer reaction with PEGDA’s
ether chain so that they can initiating photopolymerization of PEGDA without any coinitiators and
exhibit higher initiating efficiency and final double bond conversion than ITX at the same condition. All

the facts pointed to that DFP and DFP-e possess strong initiating activity. Besides, photo-bleach effect
and used at such low concentration endow the initiators potential apply to photopolymerization for
colorless system under LED light source. Hence forecast, these synthesized initiators (DFP and DFP-e)
could be promising candidate for commercialization.
CRediT authorship contribution statement
Fan Yang: Conceptualization, Methodology, Software, Data curation, Writing original draft,
Software. Meijuan Zhong: Synthesis. Xiangchen Zhao: Synthesis. Jingfang Li: Synthesis. Fang Sun:
Supervision. Jun Nie: Supervision. Xiaoqun Zhu: Supervision.
Declaration of Competing financial interest
The authors declare no competing financial interest.
Acknowledgments
This study was financially supported by the National Key Research and Development Program of
China (2017YFB0307800) and the National Natural Science Foundation of China (No. 51873008,
51603007). The authors also appreciate the support of the Beijing Laboratory of Biomedical Materials.

REFERENCES
[1]N. S. Allen, J. Photochem. Photobiol., A, 1996, 100, 101–107.
[2] J. Hiller, J. D. Mendelsohn and M. F. Rubner, Nat. Mater.,2002, 1,59–63.
[3]Fouassier JP. Photoinitiator, photopolymerization and photocuring: Fundamentals and applications.
New York: Hanser Publishers; 1995.
[4]J. P. Fouassier and F. Morlet-Savary, Opt. Eng., 1996, 35, 304–312.
[5]J. Jakubiak and J. F. Rabek, Polimery, 1999, 44, 447–461.
[6]Lalevee J, Fouassier JP. Dyes and chromophores in polymer science. John Wiley & Sons; 2015.
[7]Ganster B, Fischer UK, Moszner N, Liska R. Macromolecules 2008;41(7):2394-2400.
[8]Zhang, J.; Lalevée, J.; Mou, X.; Morlet-Savary, F.; Graff, B.; Xiao, P., Macromolecules 2018, 51 (10),
3767-3773.
[9]Tar, H.; Sevinc Esen, D.; Aydin, M.; Ley, C.; Arsu, N.; Allonas, X., Macromolecules 2013, 46 (9),
3266-3272.
[10]Yu, J.; Gao, Y.; Jiang, S.; Sun, F., Macromolecules 2019, 52 (4), 1707-1717.
[11]Xiao P, Dumur F, Bui TT, Goubard F, Graff B, Morlet-Savary F, Fouassier JP, Gigmes D, Lalevee
J. ACS Macro Lett 2013;8:736-40.
[12]Lalevꢀe, J.; Blanchard, N.; Tehfe, M.-A.; Morlet-Savary, F.; Fouassier, J. P., Macromolecules 2010,
43 (24), 10191-10195.
[13]Tehfe, M.-A.; Lalevꢀe, J.; Gigmes, D.; Fouassier, J. P., Macromolecules 2010, 43 (3), 1364-1370.
[14]Tehfe, M.-A.; Dumur, F.; Graff, B.; Morlet-Savary, F.; Fouassier, J.-P.; Gigmes, D.; Lalevée, J.,
Macromolecules 2013, 46 (10), 3761-3770.
[15]Xiao, P.; Dumur, F.; Graff, B.; Gigmes, D.; Fouassier, J. P.; Lalevée, J., Macromolecules 2014, 47
(2), 601-608.

[16]Tehfe, M.-A.; Dumur, F.; Graff, B.; Morlet-Savary, F.; Fouassier, J.-P.; Gigmes, D.; Lalevée, J.,
Macromolecules 2012, 45 (21), 8639-8647.
[17]Mutlu, S.; Watanabe, K.; Takahara, S.; Arsu, N., Journal of Polymer Science Part A: Polymer
Chemistry 2018, 56 (16), 1878-1883.
[18]P. Xiao, F. Dumur, J. Zhang, B. Graff, D. Gigmes, J. P. Fouassier and J. Lalevée, RSC Advances,
2016, 6, 4868448693.
[19]Yang, J.; Xu, C.; Xiong, Y.; Wang, X.; Xie, Y.; Li, Z.; Tang, H., Macromolecular Chemistry and
Physics 2018, 219 (24), 1800256.
[20]Zhao, J.; Lalevée, J.; Lu, H.; MacQueen, R.; Kable, S. H.; Schmidt, T. W.; Stenzel, M. H.; Xiao, P.,
Polymer Chemistry 2015, 6 (28), 5053-5061.
[21]Wu, Z.; Ji, C.; Zhao, X.; Han, Y.; Mullen, K.; Pan, K.; Yin, M., Journal of the American Chemical
Society 2019, 141 (18), 7385-7390.
[22]Gauss, P.; Griesser, M.; Markovic, M.; Ovsianikov, A.; Gescheidt, G.; Knaack, P.; Liska, R.,
Macromolecules 2019, 52 (7), 2814-2821.
[23]Atar, M.; Taspinar, O.; Hanft, S.; Goldfuss, B.; Schmalz, H. G.; Griesbeck, A. G., The Journal of
organic chemistry 2019, 84 (24), 15972-15977.
[24]Atar, M.; Öngel, B.; Riedasch, H.; Lippold, T.; Neudörfl, J.; Sampedro, D.; Griesbeck, A. G.,
ChemPhotoChem 2019, 4 (1), 89-97.
[25]Li, J.; Zhang, X.; Ali, S.; Akram, M. Y.; Nie, J.; Zhu, X., Journal of Photochemistry and Photobiology
A: Chemistry 2019, 384, 112037.
[26] Tang, L.; Nie, J.; Zhu, X.,. Polymer Chemistry 2020, 11 (16), 2855-2863.