The Blue‐LED‐Sensitive Naphthoquinone‐Imidazolyl Derivatives as Type II Photoinitiators of Free Radical Photopolymerization

In this work, new naphthoquinone‐based photoinitiators are synthesized and applied for the first time in free radical photopolymerization. In the presence of acrylate monomers, these PIs display excellent photoinitiation capabilities under blue‐light LED@405 nm irradiation in a two‐component photoinitiating system in which the additive is a common iodonium salt (bis‐(4‐tert‐butylphenyl)iodonium hexafluorophosphate ‐ Iod) or a tertiary amine (ethyl 4‐dimethylaminobenzoate ‐ EDB). In controlled tests, the polymerization performance of dye4/Iod is equal to or better than that of common commercial Type II photoinitiators (2‐isopropylthioxanthone and benzophenone). Meanwhile, only a trace amount of photoinitiator is required (as low as 0.05%) to initiate the photopolymerization. It is worth noting that some of these compounds can also be used in one‐component photoinitiating system (monocomponent Type II behavior) and exhibit high photoinitiation ability. A series of complementary characterization analyses of two new naphthoquinone dyes (i.e. dye 1 and dye 4) with very similar chemical structures are carried out to investigate the relevant photochemical mechanism. Finally, 3D printing experiments are carried out by a simple and convenient Direct Laser Write (DLW) technique.

most free-radical photoinitiators have a certain sensitivity in the UV and visible regions. [11,12] Under the irradiation of UV or visible light at a certain wavelength, photoinitiators absorb the light energy and then produce free radicals capable to initiate the polymerization of monomers and to form cross-linked network structures when multifunctional monomers are used. [13,14] Thermoplastics can also be prepared by photopolymerization using monofunctional monomers. FRP techniques are of great commercial value in the field of photocuring applications since photoinitiated radical polymerization has the advantages of being a relatively rapid reaction under mild irradiation conditions and to be energy efficient, so that photoresists, printing inks, and curing coatings of different materials have been developed. [15][16][17] The way in which photoinitiators generate radicals is distinguished by the different structures of photoinitiators, and FRP photoinitiators can be divided into two main categories according to the different mechanisms of generating radicals, i.e., Type I and Type II. [18] Upon light irradiation, Type I photoinitiators are directly cleaved to generate initiating radicals, whereas Type II photoinitiators generate radicals through an electron/proton transfer process between hydrogen acceptor (the PI in an excited state) and hydrogen donor compounds (usually called coinitiators). [19][20][21] Type II initiators are mainly aromatic ketones, such as benzophenone (BP), thianthrone, quinone, and their derivatives, etc. [22,23] Among them, BP, thianthrone, and their derivatives have good light absorption properties and high initiation efficiency in the UV region, which makes them very popular photoinitiators in the field of UV curing. [24][25][26] Unfortunately, ketyl radicals formed by these photoinitiators after photolysis can initiate chain termination reactions, so that some iodonium salts, bromine substitutes, and other additives can be added into the photoinitiating system in order to prevent chain termination reactions by inducing oxidation or bromination reactions with ketyl radicals. [27,28] In addition, the aryl radicals generated by the reaction of iodonium salts with ketyl radicals can initiate the polymerization reaction. Therefore, these additives can strongly improve the polymerization reaction rate in general. For efficient polymerization reactions, intermolecular hydrogen capture reactions must compete with other side reactions, such as non-reactive quenching of the excited state initiator caused by monomers or oxygen. [29,30] A coinitiator is usually a hydrogen donor that can form reactive radicals after a photochemical reaction. Amines, [31] ethers, [32] sulfides, [33] and mercaptans [34] can be used as hydrogen donors, but amines are the most widely used in light curing because of their low price and their ability to effectively overcome the oxygen inhibition effect. [35] However, amine additives are often toxic. Due to this major drawback, many researchers have developed a series of low-toxicity, high-initiation efficiency photoinitiators that can be used as monocomponent systems. An effort has also been done in order to expand their scope of application to the visible range so that low-consumption LED sources can be used. [36] For instance, it has been reported [37] that the introduction of acrylic acid and morpholine into the molecular structure of 4-methylbenzophenone can lead to a new one-component photoinitiator, MBPAcMP, with a slight redshift of its absorption toward the visible light and exhibiting a molar extinction coefficient of 18 800 L/mol cm. The one-component photoinitiating system could act both as a Type II photoinitiator and a hydrogen donor with a high initiation efficiency and do not require the addition of additional co-initiators in the process of initiating monomer polymerization. [38][39][40] There are numerous studies and reports [41,42] on imidazolebased photoinitiators. Imidazole is a five-membered ring two nitrogen-containing heterocyclic compound which is a popular base in Organic Chemistry. The unshared electron pair of the 1-position nitrogen atom in the imidazole ring participates to the conjugation, and the electron density of the nitrogen atom decreases, making the hydrogen on this nitrogen atom easy to leave as a hydrogen ion. [43,44] Hexaalkylbisimidazoles (HABIs) and their derivatives o-Cl-HABI and p-HOH-HABI were effective photoinitiators in thiol-alkene reactions. Under UV or short-wavelength visible light irradiation, HABI compounds can be interesting for thiol-alkene photopolymerization. [45] However, in contrast, reports on naphthoquinone-based photoinitiators were notably rare. It has been reported [46] that two naphthoquinone derivatives (5-hydroxy-1,4-naphthoquinone (5HNQ) and 2-hydroxy-1,4-naphthoquinone (2HNQ)), combined with different additives, were used for the radical or cationic polymerizations. Although the two resulting naphthoquinone derivatives showed excellent performance in cationic photopolymerization, they performed slightly less well in free radical polymerization.
Markedly, naphthoquinone and its derivatives are widely found in nature, such as chlorophyll quinone, carob quinone, and vitamin K, among others. [47] Various derivatives of diazonaphthoquinone were also used as photoresists based on linear phenolic resins, although these compounds were rather old-fashioned photoinitiators. It is based on the generation of nitrogen gas upon illumination, followed by reaction with water (which is usually produced from the substrate) to produce soluble 3-carboxyindene. [48] However, it was demonstrated that the diazo group rather than the naphthoquinone group was playing an important role. Therefore, it is extremely interesting and promising to further investigate the photoinitiation ability of naphthoquinones and their corresponding derivatives. In particular, the combination of imidazole with naphthoquinone to generate a series of naphthoquinone-imidazolyl derivatives can bring a whole new possibility in the research of photoinitiators.
In this research, 10 new naphthoquinone-imidazolyl derivatives, which have never been reported before, were synthesized and successfully applied in free radical photoinitiating systems with two different additives, namely bis(4-tert-butylphenyl)iodonium hexafluorophosphate and ethyl 4-dimethylaminobenzoate under the light irradiation of a LED@405 nm in air. Noticeably, the possibility to initiate polymerization processes in the absence of additives (i.e., used in one-component Type II photoinitiating system) was studied. Polymerization of trimethylolpropane triacrylate (TMPTA) was monitored using real-time Fourier transform infrared spectroscopy (RT-FTIR). The photochemical mechanism could be elucidated using steady-state photolysis and fluorescence quenching experiments, cyclic voltammetry, fluorescence decay lifetime, and electron spin resonance (ESR) spin trapping (ESR-ST) experiments. Finally, regular 3D patterns with excellent resolutions were obtained by 3D printing with direct laser write (DLW).

Synthesis of Naphthoquinone-Imidazolyl Derivatives
Ten new naphthoquinone-imidazolyl derivatives (See Scheme 1) never reported before in the literature were synthesized according to the procedure described in detail in Supporting Information (SI). NMR and HRMS characterizations of the different compounds are also given in SI.

Other Chemical Compounds Used in This Work
The iodonium salt (i.e., bis(4-tert-butylphenyl)iodonium hexafluorophosphate, noted Iod or Speedcure 938) and the tertiary amine (ethyl 4-dimethylaminobenzoate, EDB) used in the different photoinitiating system, as well as isopropylthioxanthone (ITX) and benzophenone (BP) used as benchmark Type II PIs in the control experiments, were purchased from Lambson Ltd (Leeds, UK). Trimethylolpropane triacrylate (TMPTA) was obtained from Allnex (Frankfurt am Main, Germany). The above reagents were used without further purification. Chemical structures of the above compounds are shown in Scheme 2.

Photopolymerization Experiments
First, novel naphthoquinone-imidazolyl derivatives were dissolved in TMPTA with Iod or EDB to prepare the photosensitized formulations. The obtained formulations were added dropwise to the pre-prepared molds (8 mm diameter and 4 mm height/thickness) for the photocuring processes. Formulations were exposed to a LED@405 nm (I 0 = 110 mW cm -2 ) to initiate the free radical polymerization (FRP) at room temperature under air. During the photopolymerization processes, the IR peak at ≈6120 cm -1 , corresponding to a characteristic peak of TMPTA, was continuously monitored using the real-time FTIR spectroscopy (JASCO FTIR 4100). The polymerization profiles were established using the following equation [13,49] : where A 0 is the initial peak area before irradiation and A t is the peak area after irradiation for a given time t.

Steady-State Photolysis
Steady-state photolysis experiments were monitored by UVvisible absorption spectroscopy (JASCO V730 UV-visible spectrometer). The solutions containing naphthoquinone-imidazolyl derivatives in chloroform were irradiated in a cuvette at room temperature using a LED@405 nm.

Fluorescence Quenching and Fluorescence Lifetime
Fluorescence quenching experiments were performed using a JASCO FP-6200 spectrometer. After the analyses of the fluorescence quenching spectra, the Stern-Volmer coefficients (K sv ) could be determined by calculating the slope values of the Stern-Volmer treatment. [50] Fluorescence-excited state lifetimes were characterized by a HORIBA PPD-850. Conditions of the experiments were as follows: The excitation wavelength (λ ex ) = 367 nm, and the pulse duration <1.4 ns.

Electrochemical Measurements
The oxidation (E ox ) and reduction (E red ) potentials were determined by electrochemistry as follows. The naphthoquinone-imidazolyl derivatives were co-dissolved with tetrabutylammonium hexafluorophosphate (as the supporting electrolyte) in a mixture of solvents: chloroform/acetonitrile (50%/ 50%: v/v). The singlet state excitation energy level (E s1 ) of the PIs was determined from the crossing point of the normalized UV-vis absorption and fluorescence spectra. The free energy changes from the singlet state (ΔG S1 Iod or ΔG S1 EDB ) for the electron transfer reaction were determined with E ox , E red , and E s1 . The specific calculation formula is detailed in Equations (2) and (3). [51] In the same way that Equations (4) and (5) show the specific calculations regarding the free energy change from the triplet state (ΔG et ). The value of the triplet state energy level (E T1 ) was calculated by molecular modeling (Gaussian 16 suite of programs). In the literature, the reduction potential of the iodonium salt is -0.7 V, [52] while the oxidation potential of EDB is 1.0 V. [53] G E 0.7 E Iod S1 ox S1 ( )

Electron Spin Resonance -Spin Trapping (ESR-ST)
The test solution was made by dissolving the PI/Iod or PI/EDB in tert-butylbenzene (Iod = 2 mg/mL, EDB = 2 mg/mL and PI = 0.2 mg/mL). After nitrogen saturation at room temperature, the solution was irradiated using a light source of LED@405 nm. N-tert-butylbenzonitrone (PBN) was used as a spin trap (PBN = 2 mg/mL). The free radicals formed could be observed by an X-band spectrometer (Bruker EMXplus). The ESR spectral simulation was performed by the PES WINSIM software. [50]

Direct Laser Write/3D Printing
The mixed resin solution containing the photoinitiator was added to the homemade tank. Direct laser write (DLW) experiments were carried out using a laser diode@405 nm under air. The intensity of the laser was 110 mW. The final 3D patterns were observed and photographed using the numerical optical microscope (OLYMPUS DSX-HRSU).

Light Absorption Properties
Molar extinction coefficients of the investigated naphthoquinone-imidazolyl derivatives in chloroform were determined by UV-visible spectroscopy (see Figure 1). The specific values corresponding to the spectra are detailed in Table 1. As shown in Figure 1,   Table 1. Light absorption properties of the different naphthoquinoneimidazolyl derivatives: maximum absorption wavelengths λ max ; extinction coefficients at λ max (ε max ) and extinction coefficients at the emission wavelength of the LED@405 nm(ε @405 nm ). LED light sources. At the maximum absorption wavelength, respectively, dye 10 exhibited the highest molar extinction coefficient (ε max = 19 370 M -1 •cm -1 ) at λ max = 413 nm. It can be confidently assigned to the polyaromaticity of this dye comprising a 5,12-dihexyl-6,7-diphenyl-5,12-dihydroindolo[3,2-a]carbazole moiety. In contrast, dye 4 exhibited the lowest molar extinction coefficient (ε max = 1300 M -1 cm -1 ) at λ max = 380 nm, despite the similarity of structure with the rest of the dyes. It can be tentatively assigned to the ortho-substitution of the electron donor, inducing an internal torsion to the molecule and adversely affecting the electronic delocalization between the electron donor and the acceptor. At the same time, the molar extinction coefficient at λ = 405 nm was also taken into account. Similar to the results for ε max , dye 10 and dye 4 showed the best and worst light absorption properties at 405 nm, with a molar extinction coefficient ε @405 nm = 19 000 M -1 •cm -1 of dye 10 and a molar extinction coefficient of ε @405 nm = 1000 M -1 cm -1 for dye 4, respectively.

Steady-State Photolysis of the Naphthoquinone-Imidazolyl Derivatives-Based Photoinitiating Systems
In this section, the steady state photolyzes of the two-component and three-component systems (dyes/Iod, dyes/EDB, and dyes/Iod/EDB) prepared with the different naphthoquinoneimidazolyl derivatives were studied. Dye 1 and dye 4 were selected as the relevant structures to discuss the PI photolysis behavior.
As shown in Figure 2, the absorption peak of dye 1 at 414 nm did not change significantly after light irradiation during 100 s, indicating that dye 1 possesses a good photostability when only dye 1 was present in the solution. In the dye 1/Iod two-component system, dye 1 was hardly consumed during the photolysis process (the absorption peak showed almost no decrease), and the chemical reaction between dye 1 and Iod was very weak. In contrast, the dye 1/EDB system showed some reactivity. Specifically, dye 1 reacted slowly with EDB (taking 300 s), and the absorption peak showed a really slight redshift at the end of the irradiation. It is noteworthy that in the three-component system of dye 1/Iod/EDB, the reaction rate of dye 1 was greatly enhanced compared with the two-component systems, and the light absorption of dye 1 changed significantly when the irradiation lasted for 20 s. Thus, the interaction of dye 1 with Iod and EDB greatly promoted the comprehensive reaction of the threecomponent system.
Noticeably, as observed for dye 1, no significant photolysis occurred for dye 4 alone under irradiation, and the peak near 320 nm did not change (see Figure 3a). Meanwhile, dye 4 Adv. Mater. Interfaces 2023, 10, 2202352 exhibited a more pronounced reactivity in two-component systems. The photolysis reaction of the two-component dye 4/Iod system was relatively more rapid (Figure 3b), since the reaction was completed within the 60s. Conversely, chemical reaction between dye 4 and EDB (Figure 3c) could also be detected during the photolysis experiments but required more time to occur. Thus, a clear and obvious photolysis could be detected after 300 s of irradiation. During the photolysis of the final three-component dye 4/Iod/EDB system (see Figure 3d), a behavior similar to that observed for the two-component dye 4/ Iod system was detected. Thus, after 60 s of irradiation, modification of the intensity of the absorption peak was no longer detectable. It can be therefore clearly seen that the interactions between the different components in the three-component system were helpful for the enhancement of the photochemical reactivity of the three-component system. Parallel to the photolysis experiments done on dye 4 and dye 10, the steady-state photolysis of the well-known and benchmark photoinitiator (ITX) were carried. Results of the photolysis experiments are shown in Figure S1, Supporting Information. ITX could efficiently react with both Iod and EDB. Under blue light irradiation, ITX had a good photolysis rate in both two-and three-component systems. These characteristics could also be evidenced for dye 4. Noticeably, the reaction rates of the different ITX and dye 4 systems prepared with Iod and EDB were very similar (100 s, 300 s of photolysis). It, therefore,, demonstrates that dye 4 can compete with the well-established ITX in terms of reactivity with iodonium salts and amines under irradiation.

Chemical Mechanism of Electron Transfer Reactions
In photopolymerization, efficiency of the photoinitiating system is not only related to the light absorption properties of the system but also to other parameters such as their redox processes and excited state lifetimes or energies. To some extent, investigation of their fluorescence quenching and fluorescence lifetimes can give an idea of their chemical reactivity in photopolymerization. Therefore, some parameters such as the electron transfer reactions in the photoinitiating systems, the singlet excited state energy (E s1 ), the Gibbs free energy change (ΔG), the triplet state energy (calculated at DFT), and the Stern-Volmer coefficients of dye 1 and dye 4 can furnish interesting information on the photoinitiating systems and the different values were calculated and listed in Table 2.
The singlet excited state energy (E s1 ) was obtained by calculating the intersection of the normalized UV-Vis absorption and fluorescence spectrum of the naphthoquinone-imidazolyl  derivatives (see Figure 4). By means of cyclic voltammetry (see Figure S2, Supporting Information), the oxidation and reduction potentials of the naphthoquinone-imidazolyl derivatives could be determined. Values of the oxidation potentials were 0.33 and 0.07 V, and the values of the reduction potentials were -0.83 and -0.9 V, for dye 1 and 4, respectively. The free energy changes of the naphthoquinone-imidazolyl derivatives in the photoinitiating systems were calculated by the equation given in Section 2. For ΔG S1 Iod and ΔG S1 EDB with dye 1 and dye 4, their values were negative, which indicated the possibility of electron transfer reactions with Iod and EDB. It is worth noting that the ΔG S1 Iod and ΔG S1 EDB values of dye 4 were lower than that of dye 1, which theoretically suggested that dye 4 was more likely to react with Iod and EDB compared to dye 1. The calculations of the triplet state energy level E T1 (Table 2) also show favorable ΔG T1 Iod and ΔG T1 EDB (Equations 4 and 5); therefore, the singlet state pathway is probably the major one (high Φet - Table 2) but triplet state pathway cannot be completely ruled out.
Fluorescence quenching experiments were performed in chloroform for dye 1 and dye 4 (see Figure 5). Fluorescence quenching processes for the two-component dye 1/Iod and dye 4/Iod systems were not recorded. Indeed, upon the addition of Iod into the solutions of dyes, an intense photoluminescence ascribed to the generation of photoproducts appeared, suggesting a good interaction between the PIs and the additives. Conversely, fluorescence quenching processes of the two-component systems composed with EDB were fully observed and the Stern-Volmer coefficients (K sv ) could be calculated by the slopes of Stern -Volmer treatment. The emission peaks of dye 1 and dye 4 at 540 and 510 nm, respectively decreased continuously with the addition of EDB to the solution. The results showed that the interaction of dye 4 and EDB was effective from the singlet excited state. On this basis, the electron transfer quantum yield (Φ et ) in the singlet excited state could be calculated, and the interaction between dye 4 and EDB (Φ et = 0.96) was slightly more efficient than that of dye 1 (Φ et = 0.94). The obtained results were consistent with the free energy change results mentioned above. Finally, fluorescence lifetimes of dye 1 and dye 4 were measured (see Figure S3, Supporting Information). Unfortunately, the calculated and analyzed values of dye 1 and dye 4 were shorter than the response time of the apparatus (1.4 ns).

Free Radical Photopolymerization
Polymerization kinetics of TMPTA in two-component systems under LED@405 nm irradiation could be determined by monitoring the conversion of acrylate functional groups using RT-FTIR. As shown in Figure 6, we first carried out the polymerization tests with the two-component PI/Iod and PI/ EDB photoinitiating systems. Both the dye 4/Iod and dye 6/Iod systems showed relatively good photoinitiation performance at room temperature and under air (acrylate functions conversion ≈78%, 72%), especially for the dye 4/Iod system, where the photopolymerization rate was fast (see Figure 6a). In contrast, the two two-component dye/EDB systems furnished lower acrylate conversions than their two-component dyes/Iod counterparts with a decrease of the monomer conversion of ≈10% (see Figure 6b). Besides, the two two-component dyes/EDB systems clearly outperformed the dyes/Iod systems in terms of polymerization rate.
To further verify our expectation that the interaction of dyes with the additives in three-component system could improve the photoinitiation performance of the resulting systems, dye 4, dye 6, and dye 8 furnished the best two-component systems were selected to form the three-component systems together with Iod and EDB. As shown in Figure 7, it could be clearly seen from the polymerization profiles that the polymerization rate of monomers was greatly improved for the three-component dye/Iod/EDB systems compared to the two-component systems based on Iod or EDB. For the dye 6/Iod/EDB and dye 8/Iod/EDB systems, improvement of the initiation ability was Adv. Mater. Interfaces 2023, 10, 2202352 Table 2. Some parameters concerning the electron transfer reactions of dye 1 and dye 4 in chloroform.
Iod ΔG S1 more obvious than for the dye4-based photoinitiating systems. Compared to the two-component systems, the conversions of acrylate functional groups were greatly enhanced while the polymerization rates also increased. To determine the real contribution of the dyes in the three-component photoinitiating systems, the monomer conversion was determined for the two-component Iod/EDB system that does not comprise any dye. This experiment was used as a blank control group (an initiation is expected due to the photosensitive charge transfer complex formed between Iod and EDB). The content of Iod and EDB was the same as that used for the abovementioned three-component system. It is worth noting that although the conversion of TMPTA after the photopolymerization reaction (≈85%) in the absence of added dyes was slightly higher than that obtained for the three-component systems, besides, the conversion rate of monomers in the dye 4/Iod/EDB system was Adv. Mater. Interfaces 2023, 10, 2202352 much faster than that of the blank group. When comparing the time required to "reach the maximum conversion," the blank group without dye required more time than the three-component systems, evidencing the crucial role of the dye in the initiation step. It could be seen that the addition of dyes led to a significant increase of the polymerization rate. However, this performance has some drawbacks, notably due to the fact that a faster polymerization rate induces an increase of the polymerization stress. [54] As can be seen from the photographs of the samples obtained after polymerization with the three-component systems, the polymerization shrinkage stress caused the samples to rupture, whereas for the samples prepared with the two-component systems, the resulting samples were smooth and homogeneous on the surface and could maintain a good structural integrity. In view of the excellent photoinitiation performance of the dye 4-based systems, we chose ITX and BP as benchmark photoinitiators for comparison (see Figure 8). Surprisingly, in the two-component system with Iod, the initiation ability of dye 4 was similar to that of ITX (see Figure 8a), while the ability of dye 4 when combined with EDB was slightly lower than that of ITX, but still much higher than BP (see Figure 8b). Both dye 4 and ITX were able to maintain a high photoinitiation ability at either 0.05% or 0.1%, while BP was completely ineffective. The low concentration of initiators was more in line with the purpose of environmental protection and green chemistry. Pictures of all samples obtained by photopolymerization are detailed in Table S1, Supporting Information.

EDB
To further demonstrate the interest in dye 4 as a photoinitiator, both ITX and dye 4 were used in one-component photoinitiating systems, and attempts to initiate the polymerization of TMPTA were carried out. As shown in Figure 9, although the polymerization rate of the monomers was not as fast as that of the two-component photoinitiating systems, the final conversion of the acrylate functional group reached 60% (dye 4 at 300 s). Noticeably, in the same conditions, dye 1 exhibits almost no initiation ability.

ESR Spin-Trapping Experiments
As shown in Figure 10a,b, the spectra of the dye 4/Iod system before and after irradiation (LED@405 nm) and the simulated spectra after irradiation were given. After 30 s of irradiation, one free radical adduct was successfully detected which could be assigned to a carbon or an oxygen-centered radical [55] (values of hyperfine coupling constants: α N = 13.50 G and α H = 1.76 G).
For the two-component dye 4/EDB system (see Figure 10c,d), after an extended irradiation time, the values of hyperfine coupling constants for one adduct (72.4%) were α N = 13.41 G and α H = 1.70 G which could be assigned to a carbon or oxygen centered radical while another generated radical was aminoalkyl radical (α N = 14.16 G and α H = 2.69 G, 23.3%). The rest could be attributed to the by-product of the reaction, which is the oxidation product of PBN [56,57] (PBN ox , α N = 7.93 G, 4.3%). The one-component photoinitiating system consisting only of dye 4 was also studied. As can be seen from Figure S4, Supporting Information, the carbon or oxygen-centered radical was also successfully observed (α N = 13.49 G and α H = 1.73 G), which fully demonstrated that dye 4 can be used as a photoinitiator for one-component photoinitiating systems for free radical polymerizations.
Therefore, the different ESR experiments do not allow to establish of the exact structures of all the generated radicals but unambiguously showed that radicals are generated upon irradiation in full agreement with the photoinitiator behavior of these dyes.

Direct Laser Write (DLW) Experiments (3D Printing)
Application of the newly developed photoinitiating systems for 3D printing and the free radical photopolymerization of TMPTA with a laser diode@405 nm as the light source was carried out (see Figure 11). The resulting 3D printed patterns showed clear contours and flat surfaces with a high resolution in the field of view of the 3D Laser Scanning Microscope. It was obvious that the dye 4-based system produced a more regular and better-structured product compared to ITX, which reflected the Adv. Mater. Interfaces 2023, 10, 2202352 consistently excellent photoinitiation performance of dye 4 (even alone in a monocomponent Type II behavior - Figure 11a).

Photoinitiation Mechanism of Free Radical Polymerization
According to the analysis of the experimental results depicted above, for the free radical photopolymerization process initiated by the three-component photoinitiator system, the concomitant presence of two different reaction processes could be determined. Thus, for the dye/Iod system, the dye can act as an electron donor for the iodonium salt. Conversely, for the dye/EDB system, the dye reacts as an electron acceptor with the amine acting as an electron donor. The two reactions are shown in Scheme 3 (r1-r3).

Conclusions
In summary, 10 completely new and never reported naphthoquinone-imidazolyl derivatives were synthesized and successfully applied in free radical photopolymerization. In order to elucidate the relevant photochemical mechanisms, a series of complementary experiments such as steady-state photolysis, cyclic voltammetry, fluorescence quenching, and fluorescence lifetimes were carried out. The generated free radicals from the studied photoinitiating systems were further clarified by ESR experiments. Interestingly, dye 4 demonstrated the best photoinitiation ability among this new series of compounds, both in two-component and in one-component systems. Dye 4 with additives (dye 4/Iod or dye 4/EDB) exhibited the highest photoinitiating ability for free radical polymerization of TMPTA under irradiation of 405 nm LED light (conversion of 78% and 71%, respectively). When no additives are present(dye 4 alone), the final conversion of TMPTA monomer is ≈60%. Compared with the previously reported naphthoquinone derivatives 5-hydroxy-1,4-naphthoquinone (5HNQ) and 2-hydroxy-1,4-naphthoquinone (2HNQ) based three-component photoinitiating system, the free-radical photopolymerization initiation properties of the twocomponent photoinitiating system composed of dye 4/additive were superior. (conversion: 46%, 5HNQ/Iod/NPG, TMPTA, LED@410 nm). By comparing with other commonly used photoinitiators (BP or ITX), dye 4 showed a similar photoinitiating efficiency. In addition, the new photoinitiating systems required only extremely mild and environmentally friendly conditions for the photopolymerization reaction to proceed, and its blue light-sensitive property allowed it to effectively utilize LED light sources. The successful application in 3D printing gave the new naphthoquinone-imidazolyl derivatives a broader prospect for future research in green polymerization processes.

Supporting Information
Supporting Information is available from the Wiley Online Library or from the author.