Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.
Polymer Journal (2025 )Cite this article di 2 ethylhexyl peroxydicarbonate
Among the photoinitiators that can be activated using ultraviolet-visible light-emitting diodes, Type I photoinitiators often contain sulfur, nitrogen, and phosphorus and may affect human health and the environment, whereas Type II photoinitiators typically contain only carbon, hydrogen, and oxygen, as exemplified by anthraquinone derivatives, but require coinitiators. Hence, sulfur-, nitrogen-, and phosphorus-free Type I photoinitiators are highly desirable. In our pursuit of such photoinitiators, we examined the ability of different silyloxyanthraquinones to initiate radical photopolymerization upon irradiation at 405 nm and found that some achieved high conversion in the absence of a coinitiator. The initiation mechanism was probed by analyzing the photolysis products, electron spin resonance spectroscopy, and isotope labeling experiments. The 1-substituted silyloxy compounds acted as Type I photoinitiators, generating isopropyl radicals as the initiating species. These compounds are among the very few known Type I photoinitiators with an anthraquinone skeleton that are sensitive to 405 nm visible light. The findings of this study facilitate the design of clean initiators free of the sulfur, nitrogen, and phosphorus commonly present in other Type I initiators.
Photopolymerization proceeds in the absence of heat and solvent and therefore offers the advantages of safety, ecofriendliness, and low space/energy requirements [1, 2]. Moreover, photopolymerization has numerous applications in 3D/4D printing [3, 4] and the production of adhesives [5] and dental materials [6]. Although photopolymerization reactions in industry are triggered mainly by ultraviolet (UV) irradiation [7], visible (vis) light can also be utilized [8,9,10], as exemplified by photopolymerization initiated by UV–vis light-emitting diodes (LEDs), which are safer, less expensive, and more photoefficient than other light sources [11, 12].
Free radical photopolymerization is initiated by the free radicals generated from photoinitiators upon photoirradiation and is widely used in various industrial applications, such as laser imaging, the radiation curing of inks and coatings, and 3D printing [13]. In terms of the formation mechanism of the initiating radical, radical polymerization photoinitiators can be classified as those relying on (a) intramolecular bond cleavage (Type I) or (b) hydrogen abstraction from a coinitiator (Type II) [13]. UV–vis LED–compatible Type I initiators, e.g., phosphine oxides [14, 15], phosphinates [14, 15], and oxime esters [16, 17], typically contain sulfur, nitrogen, and phosphorus and may therefore seriously affect human health and the environment. In fact, photoinitiators such as diphenyl(2,4,6-trimethylbenzoyl)phosphine oxide (Omnirad TPO), 2-methyl-1-(4-methylthiophenyl)-2-morpholinopropan-1-one (Omnirad 907), and 2-benzyl-2-dimethylamino-4’-morpholinobutyrophenone (Omnirad 369) have been added to the list of substances of very high concern (SVHC), which are regulated by Registration, Evaluation, Authorization and Restriction of Chemicals (REACH) [18].
In contrast, more ecofriendly Type II initiators often contain only carbon, hydrogen, and oxygen but require coinitiators (e.g., iodonium salts, amine derivatives, and thiol crosslinkers), which increases the complexity of the resin formulation by necessitating careful optimization of the coinitiator ratio due to the hydrogen abstraction process.
Anthraquinone derivatives absorb above 350 nm and can act as Type II initiators [19], as exemplified by radical photopolymerization initiated by disubstituted anthraquinones in the presence of amines [20, 21] and multi-hydroxyanthraquinones in the presence of iodonium salts and amines [22, 23]. However, since hydroxyl groups can trap radicals and thus interfere with radical chain reactions [24], they can be masked by conversion into glycidyl and allyl ethers [25] or methacrylic esters [26]. However, the use of coinitiators in Type II systems introduces additional environmental concerns, as these coinitiators (e.g., amines or iodonium salts) may not be ecofriendly. Furthermore, because Type II initiators rely on hydrogen abstraction from the coinitiators, the polymerization rate is lower, resulting in longer processing times. To apply photoinitiators in advanced manufacturing technologies, such as 3D printing, the development of Type I photoinitiators that enable rapid polymerization is crucial. Moreover, sulfur-, nitrogen-, and phosphorus-free Type I photoinitiators, such as those based on anthraquinone derivatives, are both rare and highly desirable [19], as they can address the environmental concerns.
Silanes, which are less toxic than amines, can be used as coinitiators with Type II initiators (e.g., benzophenone) to reduce the inhibitory effects of oxygen through efficient consumption by silyl radicals [27]. Type I initiators using silyl radicals as the initiating species have also been reported [28,29,30] and rely on the generation of Si-centered radicals via irradiation-induced Si–Si bond scission in tris(trimethylsilyl)silyl-substituted 4-hydroxybenzophenones. Silyl compounds are also used in polymerization reactions and hold considerable promise for green chemistry applications [31].
In this study, our goal was to develop novel anthraquinone-based Type I photoinitiators that do not require coinitiators and are free from S, N, and P, thereby increasing both the efficiency and environmental safety. While oxime esters, the current focus of Type I photoinitiator active research, typically require at least two synthetic steps, we successfully synthesized our compounds in a single step. To achieve this goal, we converted the hydroxy groups of hydroxyanthraquinones to silyloxy groups and evaluated the ability of the resulting species to initiate radical polymerization in the absence of a coinitiator upon irradiation at 405 nm. The initiation mechanism was probed by photolysis and isotope labeling experiments, electron spin resonance (ESR) spectroscopy measurements, and theoretical calculations. The 1-substituted silyloxy compounds acted as Type I photoinitiators, generating isopropyl radicals as the initiating species, making them among the few known Type I photoinitiators with an anthraquinone skeleton that are sensitive to 405 nm visible light.
Triisopropylsilyl chloride and various (di)hydroxyanthraquinones (1-hydroxy-, 2-hydroxy-, 1,4-dihydroxy-, 1,5-dihydroxy-, 1,8-dihydroxy-, and 2,6-dihydroxyanthraquinone) were purchased from Tokyo Chemical Industry (TCI, Japan). Trimethylolpropane triacrylate (TMPTA) and totally deuterated methyl methacrylate (MMA-d8) were obtained from TCI and General Science Corporation, respectively. N-tert-Butyl-α-phenylnitrone (PBN) was obtained from Cosmo Bio Corporation. The structures of the employed photoinitiators (the triisopropylsilyloxy-substituted anthraquinones; xyTIPS-AQs or xTIPS-AQs, where x and y denote the positions of the silyloxy substituents on the anthraquinone skeleton) and monomers are shown in Fig. 1.
Chemical structures of the employed photoinitiators and monomers
The TIPS-AQs were synthesized from commercial products in one step as described elsewhere [32]. The desired (di)hydroxyanthraquinone and triisopropylsilyl chloride were dissolved in toluene and triethylamine or imidazole were added with stirring under reflux (40–75 °C) overnight. The solvent was removed by evaporation, and the crude product was purified by column chromatography (ethyl acetate/hexane = 1:40, v/v).
1TIPS-AQ: 1H-nuclear magnetic resonance ((NMR); 500 MHz, chloroform-D) δ 8.24 (ddd, J = 22.1, 7.7, 1.4 Hz, 2H), 7.94 (dd, J = 8.0, 1.1 Hz, 1H), 7.73 (tdd, J = 14.5, 7.2, 1.6 Hz, 2H), 7.58 (t, J = 8.0 Hz, 1H), 7.21 (dd, J = 8.0, 1.1 Hz, 1H), 1.47–1.38 (m, 3H), 1.16 (d, J = 7.4 Hz, 18H). 13C-NMR (125 MHz, chloroform-D) δ 183.8, 182.0, 157.2, 135.7, 135.2, 134.3, 134.1, 133.1, 132.8, 127.5, 127.2, 126.6, 123.4, 120.5, 18.1, 13.5.
14TIPS-AQ: 1H-NMR (500 MHz, chloroform-D) δ 8.15-8.12 (m, 2H), 7.67 (q, J = 3.1 Hz, 2H), 7.08 (s, 2H), 1.43–1.34 (m, 6H), 1.14 (d, J = 7.4 36H). 13C-NMR (125 MHz, chloroform-D) δ 183.2, 151.0, 134.7, 133.0, 128.6, 124.3, 124.1, 18.1, 13.4.
15TIPS-AQ: 1H-NMR (500 MHz, chloroform-D) δ 7.84 (d, J = 7.4 Hz, 2H), 7.52 (t, J = 8.0 Hz, 2H), 7.12 (d, J = 8.0 Hz, 2H), 1.45–1.36 (m, 6H), 1.15 (d, J = 7.4 Hz, 36H). 13C-NMR (125 MHz, chloroform-D) δ δ 182.6, 156.4, 137.7, 134.1, 125.9, 123.2, 120.2, 18.1, 13.4.
18TIPS-AQ: 1H-NMR (500 MHz, chloroform-D) δ 7.83 (d, J = 7.4 Hz, 2H), 7.49 (t, J = 8.0 Hz, 2H), 7.16 (d, J = 8.0 Hz, 2H), 1.47–1.38 (m, 6H), 1.15 (d, J = 7.4 Hz, 36H). 13C-NMR (125 MHz, chloroform-D) δ 184.5, 181.4, 156.4, 135.1, 133.1, 127.3, 125.1, 119.1, 18.1, 13.6.
2TIPS-AQ: 1H-NMR (500 MHz, chloroform-D) δ 8.28 (qd, J = 8.1, 5.7 Hz, 2H), 8.22 (d, J = 8.6 Hz, 1H), 7.23 (dd, J = 8.6, 2.9 Hz, 1.39–1.30 (m, 3H), 1.13 (d, J = 7.4 Hz, 18H). 13C-NMR (125 MHz, chloroform-D) δ 183.4, 182.3, 161.8, 135.6, 134.2, 133.7, 130.0, 127.2, 127.2, 125.8, 125.8, 18.0, 18.0, 12.8.
26TIPS-AQ: 1H-NMR (500 MHz, chloroform-D) δ 8.18 (d, J = 8.0 Hz, 2H), 7.68 (d, J = 2.9 Hz, 2.19 (dd, J = 8.6, 2.9 Hz, 1.38–1.29 (m, 6H), 1.12 (d, J = 7.4 Hz, 36H). 13C-NMR (125 MHz, chloroform-D) δ 182.4, 161.8, 135.9, 135.9, 127.5, 125.3, 117.3, 18.0, 12.8.
The UV–vis spectra (OP-FLAME-S-TR, Ocean Photonics, Japan) of the TIPS-AQs were recorded immediately after irradiation at 1-min intervals (405 nm) in acetonitrile (0.25 mM). NMR measurements (JNM-ECA-500, JEOL Ltd, Japan) were performed at 20 min intervals for the irradiated (LED at 405 nm, SMD5050, iNextStation, Japan) solutions of TIPS-AQs in deuterated chloroform (10 mM) containing dimethyl terephthalate as a reference material. Additionally, an oxygen-, air-, or nitrogen-purged solution of 1TIPS-AQ in acetonitrile (17 mM) was irradiated (LED at 405 nm) for 1 d and subsequently concentrated by evaporation. The residue was dissolved in deuterated chloroform, and the solution was characterized via 1H NMR spectroscopy.
ESR measurements (JES-FA200, JEOL Ltd, Japan) were performed under the following conditions: temperature, ambient temperature; irradiation wavelength, 365 nm; microwave frequency, 9.43 GHz; microwave power, 4.0 mW; center field, 342.5 mT; sweep width, 4.0 mT; modulation width, 0.6 mT or 0.006 mT; sweep time, 1 min; and time constant, 0.03 s. A 365 nm light source was used instead of a 405 nm source because the latter was not installed in our ESR instrument. The TIPS-AQs, which can absorb 365 nm light, were dissolved in benzene to a concentration of 5 mM, and nitrogen was bubbled through the solutions for 30 s. PBN [33] was used as the spin-trapping agent to identify the radical species.
To identify the polymer end groups, we photopolymerized MMA-d8 [34] or MMA using 15TIPS-AQ (10 wt%) as the photoinitiator. Polymerization was performed using an LED at 405 nm (SLD-2101BF, Matsuo Sangyo Co., Ltd, Japan) and an intensity of 300 mW cm–2 for 10 min under nitrogen. The reaction mixture was dissolved in a small amount of THF, and the solution was added dropwise to hexane to precipitate the formed polymer. The precipitate was collected via suction filtration, washed several times with hexane, dried at 60 °C, and characterized via 1H NMR (JNM-ECS400, JEOL Ltd, Japan).
where dH/dt is the maximum measured heat flow and ΔHPtheory = 78.2 kJ mol−1 [39] is the theoretical enthalpy change due to the complete conversion of acrylate double bonds. The monomer conversion (C, %) was determined as
where ΔHt is the heat release at time t obtained by exothermic peak integration and ΔHMtheory is the theoretical heat release for complete conversion. Given that TMPTA has three acrylate residues, ΔHMtheory can be obtained as
where Mm is the molecular weight of TMPTA.
The reaction mechanism was probed by performing theoretical calculations. RDKit and Gaussian 16 were used to perform conformational searches and quantum chemical calculations, including structural optimization and vibrational analysis, respectively. In all calculations, 1000 conformations were generated using the experimental-torsion-knowledge distance geometry (ETKDG) method [40], which was followed by a search for the most stable conformation using the Merck molecular force field (MMFF) [41], structural optimization, and vibrational analysis using density functional theory (DFT) calculations at the B3LYP/6-31G(d) level. Excited-state calculations, namely, ground-state structure optimization at the B3LYP/6-31G(d) level and single-point energy calculations for excited states at the cam-B3LYP/6-311G+(2d,p) level, were performed using time-dependent DFT (TD-DFT) [42, 43]. Bond dissociation energy (BDE) calculations were performed at the UB3LYP/6-31G(d) level.
Table 1 summarizes the maximum absorption wavelengths (λmax values) and molar absorption coefficients at λmax (εmax values) and 405 nm (ε405nm values) of different TIPS-AQs. Figure S1 presents their UV‒vis spectra at a concentration of 0.25 mM in acetonitrile.
The examined photoinitiators featured absorption maxima between 350–409 nm and absorbed at 405 nm. 2TIPS-AQ and 26TIPS-AQ had lower λmax values and hence lower ε405nm values than the 1-substituted compounds did.
From the TD-DFT calculations, the types of transitions were identified, and the main transitions were classified. The ground state (S0) to excited state (S1) transitions for 1TIPS-AQ and 14TIPS-AQ occur from the highest occupied molecular orbital-1 (HOMO-1) to the lowest unoccupied molecular orbital (LUMO). S0 → S1 transitions for 15TIPS-AQ, 18TIPS-AQ, 2TIPS-AQ, and 26TIPS-AQ occur from HOMO-2 to LUMO. Furthermore, the S0 → S1 transitions of all TIPS-AQs were dominated by their nπ* states (Fig. 2).
Molecular orbitals involved in the S0 → S1 transitions of the TIPS-AQs
To assess photoreactivity, we performed steady-state photolysis experiments by recording the UV–vis spectra of the TIPS-AQ solutions in acetonitrile (0.25 mM) upon irradiation at 405 nm. The absorbances of the 1-substituted compounds (1TIPS-AQ, 14TIPS-AQ, 15TIPS-AQ, and 18TIPS-AQ) decreased with increasing irradiation time in air; i.e., these compounds underwent photolysis (Fig. 3). 1-TIPS-AQ, 14-TIPS-AQ, and 15-TIPS-AQ rapidly lost their entire long-wavelength absorption, whereas the photolysis of 18-TIPS-AQ was slow, and the 2-substituted compounds (2-TIPS-AQ and 26-TIPS-AQ) did not undergo photolysis even after 10 min.
Effects of irradiation duration (at 405 nm) on the UV–vis spectra of acetonitrile solutions of A 1TIPS-AQ, B 14TIPS-AQ, C 15TIPS-AQ, D 18TIPS-AQ, E 2TIPS-AQ, and F 26TIPS-AQ exposed to air
To quantify the photolysis rates, we dissolved the photolytically active compounds (1TIPS-AQ, 14TIPS-AQ, 15TIPS-AQ, and 18TIPS-AQ) and a reference material (dimethyl terephthalate) in deuterated chloroform (10 mM) and performed 1H NMR measurements after 0, 20, 40, and 60 min of irradiation (Fig. 4). The concentrations of 15TIPS-AQ and 14TIPS-AQ decreased to <20% of their original concentration after 20 min and were close to zero after 40 min. In contrast, the concentrations of 1TIPS-AQ and 18TIPS-AQ after 40 min were 10% and 90% of their original concentration, respectively. Given that 2TIPS-AQ and 26TIPS-AQ did not undergo photolysis, the photolysis rates followed the order of 15TIPS-AQ > 14TIPS-AQ > 1TIPS-AQ > 18TIPS-AQ >> 2TIPS-AQ = 26TIPS-AQ.
Effects of irradiation duration on the 1H nuclear magnetic resonance (NMR) spectroscopy-determined yields of the TIPS-AQs in deuterated chloroform
To identify the photolysis products, we irradiated a solution of 1TIPS-AQ in acetonitrile exposed to oxygen, air, or nitrogen at 405 nm for 1 d. The results of the 1H NMR, 13C NMR, H-H correlation spectroscopy (HH-COSY), heteronuclear multiple bond correlation (HMBC), and heteronuclear multiple quantum correlation (HMQC) analyses revealed the formation of 1-hydroxyanthraquinone (1OH-AQ) as the main product and a photolytically inactive byproduct (Scheme 1). The 1OH-AQ/byproduct ratio depended on the atmosphere, i.e., the byproduct peak was not observed after exposure to oxygen but were more intense than those of 1OH-AQ after nitrogen exposure (Fig. S2). Thus, irradiating 1TIPS-AQ at 405 nm resulted in the cleavage of Si–O and Si–C bonds.
Outcomes of 1TIPS-AQ photolysis in acetonitrile under different conditions
Radical formation was observed only with the photolytically active compounds, i.e., for 1TIPS-AQ, 14TIPS-AQ, 15TIPS-AQ, and 18TIPS-AQ, but not for 2TIPS-AQ and 26TIPS-AQ (Fig. 5), which suggested that photolysis resulted in radical formation.
Electron spin resonance (ESR) spectra obtained upon 365 nm irradiation of benzene solutions of A 1TIPS-AQ, B 14TIPS-AQ, C 15TIPS-AQ, D 18TIPS-AQ, E 2TIPS-AQ, and F 26TIPS-AQ
The radicals generated by photolysis were identified on the basis of the hyperfine splitting constants of nitrogen (aN) and hydrogen (aH) in the PBN/radical adducts formed upon irradiation in the presence of PBN. The ESR spectra of the 14TIPS-AQ/PBN and 15TIPS-AQ/PBN mixtures are shown in Fig. 6. The aN and aH values obtained herein (14.6 and 2.6 G, respectively) were close to those of the PBN/isopropyl radical adducts (14.67 and 2.59 G, respectively) [44]. In contrast, the PBN/silyl radical adducts featured a high aH of 6.0 G [26, 44,45,46]. Thus, irradiation of 14TIPS-AQ and 15TIPS-AQ concluded by generating isopropyl radicals (via Si–C bond cleavage), which were preferentially trapped by PBN compared with trapping of the concomitantly produced Si-centered radicals.
ESR spectra of the radicals generated in the A 14TIPS-AQ/N-tert-butyl-α-phenylnitrone (PBN) and B 15TIPS-AQ/PBN systems upon irradiation at 365 nm in benzene
To determine whether the radicals observed by ESR spectroscopy were the initiating species attacking the monomer, we examined the end groups of the polymers produced from MMA and MMA-d8 in the presence of 10 wt% 15TIPS-AQ via 1H NMR spectroscopy (Fig. 7). Unlike that of PMMA, the spectrum of PMMA-d8 featured no monomer-derived peaks (a–c) but contained the peaks of an isopropyl group (1.5 ppm (1H) and 0.7 ppm (6H)). No anthraquinone-derived peaks were observed between 7.0–8.0 ppm in either spectrum. Similar results were obtained for PMMA-d8 obtained using 14TIPS-AQ as the initiator (Fig. S3). Thus, we concluded that the irradiation of 1-substituted compounds generated isopropyl radicals, which attacked the monomer as the initiating species.
1H NMR spectra of poly(methyl methacrylate) (PMMA) and PMMA-d8 obtained by photopolymerization (405 nm) in the presence of 15TIPS-AQ
Figure 8 shows the results of the photo-DSC measurements, namely, the time courses of heat flow (solid line) and monomer conversion (dotted line), revealing that all the compounds efficiently initiated the polymerization of TMPTA upon 405 nm irradiation. The RP values (and hence, the photopolymerization rates) followed the order of 15TIPS-AQ (0.956 mmol s−1) > 14TIPS-AQ (0.902 mmol s−1) > 1TIPS-AQ (0.500 mmol s−1) > 18TIPS-AQ (0.294 mmol s−1) > 2TIPS-AQ (0.274 mmol s−1) > 26TIPS-AQ (0.194 mmol s−1), i.e., the RP was higher for 1-substituted compounds than for 2-substituted compounds. The higher RP values for 15TIPS-AQ and 14TIPS-AQ can be attributed to their ability to rapidly undergo photolysis and efficiently generate isopropyl radicals, whereas 18TIPS-AQ, with slower photolysis, has a lower RP, and 2TIPS-AQ and 26TIPS-AQ, which do not undergo photolysis at all, have the lowest RP values. For reference, the commercially available oxime ester-based photoinitiator OXE-01 has an RP of 0.877 mmol s−1 (Fig. S4), indicating that both 15TIPS-AQ and 14TIPS-AQ are more reactive than OXE-01.
Heat flows (solid lines) and monomer conversions (dotted lines) as functions of time during the radical photopolymerization of trimethylolpropane triacrylate in the presence of the TIPS-AQs
Monomer conversion after 5 min followed the order of 14TIPS-AQ (54.9%) > 15TIPS-AQ (53.2%) > 2TIPS-AQ (51.5%) > 26TIPS-AQ (48.8%) > 1TIPS-AQ (47.1%) > 18TIPS-AQ (43.2%). The conversions immediately after irradiation onset were lowest for 2TIPS-AQ and 26TIPS-AQ, whereas the final conversions were lowest for 1TIPS-AQ and 18TIPS-AQ. Thus, initiator performance depended on the substitution pattern and was highest for 14TIPS-AQ and 15TIPS-AQ. Compared with commercially available OXE-01, which achieved 55.0% monomer conversion in 5 min, the TIPS-AQs presented slightly lower final conversions. However, 14TIPS-AQ and 15TIPS-AQ exhibited higher RP values, which rapidly increased the network density, limiting the mobility of free radicals and consequently hindering further polymerization; this explains their lower final conversions. Nevertheless, 14TIPS-AQ and 15TIPS-AQ can be considered efficient alternatives to oxime ester-based initiators such as OXE-01—which typically require at least two steps for synthesis—as they achieve high performance with a more straightforward synthesis process. In the following section, we propose the mechanisms behind the superior performances of 14TIPS-AQ and 15TIPS-AQ and discuss how 2TIPS-AQ and 26TIPS-AQ, despite not generating radicals on their own, were still capable of initiating polymerization.
The Norrish Type I reaction, commonly observed for Type I initiators, is the photoinduced homolytic cleavage of the α-bond of the carbonyl group. The nπ* transition state is an electron-deficient state with one electron in the n orbital of the carbonyl group [47]. The overlap of this n orbital with the σ-bonding orbital of the α-bond weakens the α-bond and facilitates its cleavage. Therefore, such cleavage is more likely to occur in the nπ* transition state than in the ππ* state [48]. The S1 transition states of all TIPS-AQs used in this study were dominated by the nπ* state (Fig. 2). However, only the 1-substituted compounds (1TIPS-AQ, 14TIPS-AQ, 15TIPS-AQ, and 18TIPS-AQ) underwent photolysis, possibly because α-bond cleavage was caused not by overlap with the n orbital of the keto group but by overlap with the Si–O or Si–C bond at neighboring position 1. Photolysis from the triplet state was also possible since the nπ* state was confirmed to be the dominant T1 or T2 transition state for these compounds. The compounds substituted at position 2 (2TIPS-AQ and 26TIPS-AQ) did not undergo photolysis, as the n orbital of the keto group did not efficiently facilitate α-bond cleavage and the Si–O or Si–C bond at position 2 was too far away for the overlap to be efficient.
Table 2 lists the Si–O and Si–C BDEs of the TIPS-AQs, revealing that the latter bonds are easier to cleave. The overlap between the n orbitals of the keto groups of TIPS-AQs and the easily cleaved Si–C bonds at the 1 positions was responsible for isopropyl radical formation upon irradiation. In the nitrogen-purged solvent system without TMPTA, the isopropyl radicals attacked the anthraquinone moiety at position 10 to generate the abovementioned byproduct (Scheme 2). In contrast, the byproduct was not formed in the presence of oxygen, as oxygen trapped the isopropyl radicals. Furthermore, oxygen trapped the silyl radicals produced by Si–C bond dissociation to afford 1OH-AQ as the main product.
Photopolymerization and photoreaction mechanisms of the TIPS-AQs substituted at position 1 with 1TIPS-AQ as an example
The faster photolysis and better initiation properties of 14TIPS-AQ and 15TIPS-AQ than those of the other compounds were attributed to the interactions among the n orbitals of the keto groups at positions 9 and 10 and the Si–C bonds at positions 1 and 4 or 1 and 5, respectively. In contrast, the slower photolysis of 18TIPS-AQ arises from the n orbitals of the keto group at position 10 competing with the Si‒C bonds at positions 1 and 8. This competition reduces the efficiency of radical generation during photolysis, leading to poorer initiation properties than those of 14TIPS-AQ and 15TIPS-AQ.
Photolytically inactive 2TIPS-AQ and 26TIPS-AQ did not produce radicals upon irradiation but induced the polymerization of TMPTA, which, however, was slow (Fig. 8).
Given that quinones, such as anthraquinones and naphthoquinones, can act as Type II initiators [19, 49, 50], we believe that so can 2TIPS-AQ and 26TIPS-AQ. In fact, no radicals were observed by ESR after irradiation of the photoinitiator (2TIPS-AQ and 26TIPS-AQ)-only systems; however, radical formation was observed after 5 min of irradiation with the mixed solutions of 2TIPS-AQ/TMPTA and 26TIPS-AQ/TMPTA (Fig. S5). Thus, we concluded that upon irradiation of these mixtures, TMPTA acted as a coinitiator (hydrogen donor) and was converted into the TMPTA radical (reaction initiator) via hydrogen abstraction.
Position 1 (1TIPS-AQ, 14TIPS-AQ, 15TIPS-AQ and 18TIPS-AQ)
TIPS–AQs → 1,3TIPS–AQs (nπ*) → TIPS–AQs• + isopropyl radical (IP•)
TIPS–AQs → 1,3TIPS–AQs → no radical generation
1.3Tips -AQS + TMPTA -H → Tips -AQS -H • + TMPTA •
Traditionally, anthraquinones have been used as Type II photoinitiators, which require coinitiators such as amines for effective radical generation. However, in this study, we demonstrated the potential of using anthraquinone derivatives as Type I photoinitiators for rapid polymerization, offering a significant advantage by eliminating the need for a coinitiator, thereby simplifying the photoinitiation process. Some of the examined TIPS-AQs, which can be synthesized in a single step, showed potential as standalone initiators upon irradiation at 405 nm, as their photoinitiating performance was strongly influenced by the position of the silyloxy group. Specifically, the 1-substituted silyloxy compounds produced isopropyl radicals, which attacked the monomers to initiate polymerization. All of the TIPS-AQs were dominated by the nπ* transition state: compounds substituted at position 1 dissociated because of the overlap between the n orbital of the keto group and the Si–C bonding orbital at position 1, whereas compounds substituted at position 2 did not dissociate because of the large distance between the abovementioned n orbital and the Si–C bonding orbital at position 2. In particular, 15TIPS-AQ and 14TIPS-AQ had high initiation abilities owing to the effective overlap between their keto group n orbitals and Si‒C bonds. These compounds hold great promise as clean Type I initiators, as they do not require coinitiators such as amines, which are often used with Type II initiators, and do not contain S, N, or P atoms, which are present in other Type I initiators. Future work will concentrate on designing photoinitiators with enhanced radical generation, with a focus on optimizing the overlap between the n orbitals and Si–C bonds, particularly with substitution at the 1,4 or 1,5 positions. Additionally, exploring photoinitiators capable of generating isopropyl radicals as well as other potential radical-generating groups will broaden the scope of development. To assess their commercial viability, the reactivity of these compounds should also be tested in resins containing black pigments, which pose challenges for light penetration.
Supplementary information is available at the Polymer Journal website.
Choi SS, Lee AS, Lee HS, Jeon HY, Baek KY, Choi DH, et al. Synthesis and characterization of UV-curable ladder-like polysilsesquioxane. J Polym Sci Part A Polym Chem. 2011;49:5012–18. https://doi.org/10.1002/pola.24942.
Payne JA, Francis LF, McCormick AV The effects of processing variables on stress development in ultraviolet-cured coatings. J Appl Polym Sci. 1997;66:1267–77. https://doi.org/10.1002/(SICI)1097-4628(19971114)66:7%3C1267::AID-APP6%3E3.0.CO;2-E.
Ligon-Auer SC, Schwentenwein M, Gorsche C, Stampfl J, Liska R. Toughening of Photo-Curable Polymer Networks: A Review. Polym. 2016; 7: 257–86. https://doi.org/10.1039/c5py01631b.
Andreu A, Su P-C, Kim J-H, Ng CS, Kim S, Kim I, et al. 4D printing materials for vat photopolymerization. Additive Manufacturing 2021;44:102024. https://doi.org/10.1016/j.addma.2021.102024.
Decker C, Thi Viet TN, Decker D, Weber-Koehl E. UV-radiation curing of acrylate/epoxide systems. Polymer. 2001;42:5531–41.
Williams PT, Johnson LN. Composite resin restoratives revisited. J Can Dent Assoc. 1993;59:538–43.
Bagheri A, Jin J. Photopolymerization in 3D printing. ACS Appl Polym Mater. 2019;1:593–611. https://doi.org/10.1021/acsapm.8b00165.
Dadashi-Silab S, Doran S, Yagci Y. Photoinduced electron transfer reactions for macromolecular syntheses. Chem Rev. 2016;116:10212–75. https://doi.org/10.1021/acs.chemrev.5b00586.
Article CAS PubMed Google Scholar
Wang J, Stanic S, Altun AA, Schwentenwein M, Dietliker K, Jin L, et al. A highly efficient waterborne photoinitiator for visible-light-induced three-dimensional printing of hydrogels. Chem Commun. 2018;54:920–3. https://doi.org/10.1039/C7CC09313F.
Dadashi-Silab S, Aydogan C, Yagci Y. Shining a light on an adaptable photoinitiator: advances in photopolymerizations initiated by thioxanthones. Polym Chem. 2015;6:6595–615. https://doi.org/10.1039/C5PY01004G.
Fouassier JP, Lalevée J. Photoinitiators for polymer synthesis—scope, reactivity, and efficiency. Weinheim: Wiley-VCH Verlag GmbH & Co KGaA; 2012.
Zhang J, Frigoli M, Dumur F, Xiao P, Ronchi L, Graff B, et al. Design of novel photoinitiators for radical and cationic photopolymerizations under near UV and visible LEDs (385, 395 and 405 nm). Macromolecules. 2014;47:2811–9. https://doi.org/10.1021/ma500612x.
Lalevée J, Telitel S, Tehfe MA, Fouassier JP, Curran DP, Lacôte E. N-heterocyclic carbene boranes accelerate type I radical photopolymerizations and overcome oxygen inhibition. Angew Chem Int Ed. 2012;51:5958–61. https://doi.org/10.1002/anie.201200975.
Wu Y, Ke J, Dai C, Wang J, Huang C, Situ Y, et al. Large-molecular-weight acyldiphenylphosphine oxides as low-mobility type I photoinitiator for radical polymerization. Eur Polym J. 2022;175:111380. https://doi.org/10.1016/j.eurpolymj.2022.111380.
Fairbanks BD, Schwartz MP, Bowman CN, Anseth KS. Photoinitiated polymerization of PEG-diacrylate with lithium phenyl-2,4,6-trimethylbenzoylphosphinate: polymerization rate and cytocompatibility. Biomaterials. 2009;30:6702–7. https://doi.org/10.1016/j.biomaterials.2009.08.055.
Article CAS PubMed PubMed Central Google Scholar
Hammoud F, Giacoletto N, Nechab M, Graff B, Hijazi A, Dumur F, et al. 5,12-Dialkyl-5,12-dihydroindolo[3,2-a]carbazole-based oxime-esters for LED photoinitiating systems and application on 3D printing. Macromol Mater Eng. 2022;307:2200082. https://doi.org/10.1002/mame.202200082.
Qiu W, Zhu J, Dietliker K, Li Z. Polymerizable oxime esters: an efficient photoinitiator with low migration ability for 3D printing to fabricate luminescent devices. ChemPhotoChem. 2020;4:5296–303. https://doi.org/10.1002/cptc.202000146.
[ECHA] European Chemical Agency. Candidate List of substances of very high concern for authorisation. 2024. https://echa.europa.eu/candidate-list-table.
Dumur F. Recent advances on anthraquinone-based photoinitiators of polymerization. Eur Polym J. 2023;191:112039. https://doi.org/10.1016/j.eurpolymj.2023.112039.
Allen NS, Pullen GK, Shah M, Edge M, Weddell I, Swart R, et al. Photochemistry and photoinitiator properties of 2-substituted anthraquinones: 2. Photopolymerization and flash photolysis. Polymer. 1995;36:4665–74. https://doi.org/10.1016/0032-3861(95)96834-U.
Pullen GK, Allen NS, Edge M, Weddell I, Swart R, Catalina F, et al. Anthraquinone photoinitiators for free-radical polymerisation: structure dependance on photopolymerisation activity. Eur Polym J. 1996;32:943–55. https://doi.org/10.1016/0014-3057(96)00035-3.
Zhang J, Lalevée J, Zhao J, Graff B, Stenzel MH, Xiao P. Dihydroxyanthraquinone derivatives: natural dyes as blue-light-sensitive versatile photoinitiators of photopolymerization. Polym Chem. 2016;7:7316–24. https://doi.org/10.1039/C6PY01550F.
Zhang J, Hill N, Lalevee J, Fouassier JP, Zhao J, Graff B, et al. Multihydroxy-anthraquinone derivatives as free radical and cationic photoinitiators of various photopolymerizations under green LED. Macromol Rapid Commun. 2018;39:1800172. https://doi.org/10.1002/marc.201800172.
Caldwell RG, Ihrig JL. The reactivity of phenols toward peroxy radicals. I. Inhibition of the oxidation and polymerization of methyl methacrylate by phenols in the presence of air. J Am Chem Soc. 1962;84:2878–86. https://doi.org/10.1021/ja00874a006.
Sautrot-Ba P, Jockusch S, Malval J-P, Brezová V, Rivard M, Abbad-Andaloussi S, et al. Quinizarin derivatives as photoinitiators for free-radical and cationic photopolymerizations in the visible spectral range. Macromolecules. 2020;53:1129–41. https://doi.org/10.1021/acs.macromol.9b02448.
Breloy L, Brezová V, Barbieriková Z, Ito Y, Akimoto J, Chiappone A, et al. Methacrylated quinizarin derivatives for visible-light mediated photopolymerization: promising applications in 3D-printing biosourced materials under LED@405 nm. ACS Appl Polym Mater. 2022;4:210–28. https://doi.org/10.1021/acsapm.1c01210.
Lalevée j, Field A, El-roz M, Allonas x, fuassier jp. Silanes as new highly efficient co-initiators for radical polymerization in aired media. The macrolecules. 2008;41:2003–1 https://doi.org/10.1021/ma702301x.
Lalevee J, Blanchard N, El-Roz M, Graff B, Allonas X, Fouassier JP. New photoinitiators based on the silyl radical chemistry: polymerization ability, ESR spin trapping, and laser flash photolysis investigation. Macromolecules. 2008;41:4180–6. https://doi.org/10.1021/ma8005125.
Graff B, Klee JE, Fik C, Maier M, Fouassier JP, Lalevée J. Development of novel photoinitiators as substitutes of camphorquinone for the LED induced polymerization of methacrylates: a bis-silyl ketone. Macromol Rapid Commun. 2017;38:1600470. https://doi.org/10.1002/marc.201600470.
Versace DL, Tehfe MA, Lalevée J, Casarotto V, Blanchard N, Morlet-Savary F, et al. Silyloxyamines as sources of silyl radicals: ESR spin-trapping, laser flash photolysis investigation, and photopolymerization ability. J Phys Org Chem. 2011;24:342–50. https://doi.org/10.1002/poc.1762.
Lalevée J, Fouassier JP. Recent advances in sunlight induced polymerization: role of new photoinitiating systems based on the silyl radical chemistry. Polym Chem. 2011;2:1107–13. https://doi.org/10.1039/C1PY00073J.
Numata S, Yamada A, Yokoyama S, Miki Y, Preparation of silyloxy-substituted 9,10-anthraquinone compounds, radical photopolymerization initiators and polymerizable compositions containing them, their polymerization method, and polymerized products. JP2016113392, 2016-06-23.
Xiao P, Lalevée J, Allonas X, Fouassier JP, Ley C, El Roz M, et al. Photoinitiation mechanism of free radical photopolymerization in the presence of cyclic acetals and related compounds. J Polym Sci Part A Polym Chem. 2010;48:5758–66. https://doi.org/10.1002/pola.24383.
Hatada K, Kitayama T, Masuda E. Studies on the radical polymerization of methyl methacrylate in bulk and in benzene using totally deuterated monomer technique. Polym J. 1986;18:395–402. https://doi.org/10.1295/polymj.18.395.
Bachmann J, Gleis E, Schmölzer S, Fruhmann G, Hinrichsen O. Photo-DSC Method for Liquid Samples Used in Vat Photopolymerization. Anal chim acta. 2021; 1153: 338268. https://doi.org/10.1016/j.aca.2021.338268.
Article CAS PubMed Google Scholar
Rusu MC, Block C, Van Assche G, Van Mele B. Influence of temperature and UV intensity on photo-polymerization reaction studied by photo-DSC. J Therm Anal Calorim. 2012;110:287–94. https://doi.org/10.1007/s10973-012-2465-5.
Green WA. Industrial photoinitiators: a technical guide. Boca Raton: CRC Press; 2010. p. 120.
Erdur S, Yilmaz G, Colak DG, Cianga I, Yagci Y. Poly(phenylenevinylene)s as sensitizers for visible light induced cationic polymerization. Macromolecules. 2014;47:7296–302. https://doi.org/10.1021/ma5019457.
Brandrup J, Immergut EH. Polymer handbook. 3rd ed. New York: Wiley; 1989.
Riniker S, Landrum GA. Better informed distance geometry: using what we know to improve conformation generation. J Chem Inf Model. 2015;55:2562–74. https://doi.org/10.1021/acs.jcim.5b00654.
Article CAS PubMed Google Scholar
Halgren TA. Merck molecular force field. I. Basis, form, scope, parameterization, and performance of MMFF94. J Comput Chem. 1996;17:490–519. https://doi.org/10.1002/(SICI)1096-987X(199604)17:5/6%3C490::AID-JCC1%3E3.0.CO;2-P.
Foresman JB, Frisch A. Exploring chemistry with electronic structure methods. 2nd ed. Pittsburgh, PA: Gaussian Inc.; 1996.
Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, Robb MA, Cheeseman JR, et al. Gaussian 16, Revision A.03. Wallingford, CT: Gaussian, Inc.; 2016.
Haire LD, Krygsman PH, Janzen EG, Oehler UM. Correlation of radical structure with EPR spin adduct parameters: utility of the proton, carbon-13, and nitrogen-14 hyperfine splitting constants of aminoxyl adducts of PBN-nitronyl-13C for three-parameter scatter plots. J Org Chem. 1988;53:4535–42. https://doi.org/10.1021/jo00254a022.
Tordo P. In: Atherton NM, Davies MJ, Gilbert BC, editors. Spin-trapping: recent developments and applications. Cambridge: The Royal Society of Chemistry; 1998. p. 16.
Peyrot F, Lajnef S, Versace D-L. Electron paramagnetic resonance spin trapping (EPR–ST) technique in photopolymerization processes. Catalysts. 2022;12:772. https://doi.org/10.3390/catal12070772.
Turro NJ, Ramamurthy V, Scaiano JC. Modern molecular photochemistry of organic molecules. Sausalito, CA: University Science Books; 2010. p. 677
Jockusch S, Landis MS, Freiermuth B, Turro NJ. Photochemistry and photophysics of R-hydroxy ketones. Macromolecules. 2001;34:1619–26. https://doi.org/10.1021/ma001836p.
Strzelczyk R, Podsiadły R. Derivatives of 1,4-naphthoquinone as visible-light-absorbing one-component photoinitiators for radical polymerization. Color Technol. 2015;131:229–35. https://doi.org/10.1111/cote.12144.
Cokbaglan L, Arsu N, Yagci Y, Jockusch S, Turro NJ. 2-Mercaptothioxanthone as a novel photoinitiator for free radical polymerization. Macromolecules. 2003;36:2649–53. https://doi.org/10.1021/ma0214613.
Open Access funding provided by Yokohama National University.
Department of Applied Chemistry, Yokohama National University, Yokohama, Japan
Eunseok Lee & Hiroaki Gotoh
Air Water Performance Chemical, Inc., Research & Development Center, Kawasaki, Japan
Takuya Sekizawa, Yuki Kobayashi-Miyajima, Takaya Hirose, Shunichi Himori & Akihiko Yamada
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
The authors declare no competing interests.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Lee, E., Sekizawa, T., Kobayashi-Miyajima, Y. et al. Silyloxy-substituted anthraquinones as Type I photoinitiators for visible light-induced radical polymerization. Polym J (2025). https://doi.org/10.1038/s41428-024-01001-9
DOI: https://doi.org/10.1038/s41428-024-01001-9
Anyone you share the following link with will be able to read this content:
Sorry, a shareable link is not currently available for this article.
Provided by the Springer Nature SharedIt content-sharing initiative
tert butyl peroxybenzoate Polymer Journal (Polym J) ISSN 1349-0540 (online) ISSN 0032-3896 (print)