English

• Photopolymerization is currently a well-developed technology that has assumed an increasing relevance in a wide range of applications, including packaging, coatings, photoresist, 3D printing, and microelectronics [1-4]. Compared to conventional UV light, visible light and near-infrared light (NIR) present advantages such as biosafety, deeper penetration, and reduced light scattering, making them a better option for photopolymerization [5-8]. In particular, NIR has an appealing prospect for biomedical applications, photocatalysis, and additive manufacturing [9-12]. Photoinitiators (PIs), which can produce active species to initiate polymerization when exposed to light, is the central component of a photopolymerization formulation [13, 14]. However, the migration of PIs and their cleavage products from cured materials has been a major safety concern with photopolymerization [15-18]. Therefore, to address this limitation, several strategies have been proposed such as developing PIs with large molecular weight [19], Macro-PIs [20], polymerizable PIs [21-23], macrocyclic PIs [24], and coupling PIs to polyhedral oligomeric silsesquioxanes (POSS) [25]. Among them, Macro-PIs have been an important topic due to their advantages of low cost, ease of synthesis, and good migration stability.

  Click reactions [26], living radical polymerization [27, 28], copolymerization [29], ring opening polymerization [30], and other techniques are currently the primary methods used to create Macro-PIs. These Macro-PIs can be classified into free radical and cationic types based on how polymerization is initiated, and current research on Macro-PIs mostly concentrates on free radical-type Macro-PIs [31, 32]. CPs have been developed extensively due to their advantages including low shrinkage [33], oxygen immunity [34], and dark curing [35]. Developing high-efficient and multi spectral applicable cationic Macro-PIs possesses significant research value. On the one hand, multi-component cationic Macro-PIs occupy the main part of cationic Macro-PIs, which can undergo free radical promoted cationic polymerization (FRPCP, Scheme 1a) when mixed with iodonium salts [36]. However, the photo-yellowing effect [37], back electron transfer [38], and low efficiency [39, 40] caused by interactions between multiple components limit their applications. On the other hand, the development of one-component cationic Macro-PIs is still relatively slow and facing great challenges including low wavelength absorption, poor solubility, and low efficiency because of insufficient repetition of effective units (PIs). Yagci et al. [41, 42] reported one-component cationic Macro-PIs constructed by copolymerizing styrene and 2-vinylpyridine before the synthesis of pyridinium salts. In the above strategy, it is difficult to ensure the number of PIs on the macromolecular chain, the low proportion of PIs on the chain leads to low efficiency of the Macro-PIs mentioned above, and the restrictions of chromophore selection limit their absorption wavelength. Coumarin and derivates possess good photosensitive properties [43], such as a high fluorescence quantum yield [44] and molar extinction coefficient [45], and they exhibit the ability to be chemically modified to make highly active and visible light responsive PIs [46-48].

  Scheme 1

  

  Scheme 1.  Schematic diagram showing the previous work about cationic Macro-PIs and P-CSS in this work.

  DownLoad: Full-Size Img PowerPoint

  In this work, we adopted the strategy of copolymerizing small molecule PI (CSS) and MMA (Scheme 1b) to synthesize P-CSS, which increased the number of effective units on the side chain and subsequently enhanced photoinitiation efficiency. The longest absorption wavelength of P-CSS is redshifted by about 50 nm compared to CSS, and the photoinitiation activity showed that P-CSS can not only induce CP, FRP, and HP under UV and visible light, but also be utilized with UCPs to induce CP and FRP under NIR. Notably, the migration stability of P-CSS was evaluated by polymer sample immersion experiments, which was 1.25% of CSS in TMPTA and 1.96% in EPOX. A new avenue was opened up for the development of multifunctional Macro-PIs.

  The overall process for synthesizing P-CSS is summarized in Scheme S1 (Supporting information), and the structures of each step product were confirmed by the nuclear magnetic resonance hydrogen spectrum (¹H NMR, Figs. S10-S13 in Supporting information). ¹H NMR validated that CSS has successfully undergone copolymerization with MMA, and P-CSS was obtained. During copolymerization, deionization of sulfonium salt occurs after active free radicals oxidated by sulfonium salt (Scheme S2 in Supporting information), so that the side chains of coumarin phenyl sulfide also exist [49]. P-CSS has an average molecular weight of 12, 143 g/mol and a PDI of 2.9 (Fig. S1 in Supporting information). According to ¹H NMR, the ratio of each repeating unit in the macromolecular segment is x: y: z = 3.08:2.97:1 (x: y: z = c/3:d: a, Fig. S13 in Supporting information).

  Photochemical and photophysical characteristics of P-CSS were studied (Fig. 1) and the absorption properties of CSS and P-CSS are provided in Table S1 (Supporting information). It can be observed that P-CSS has a much higher molar extinction coefficient (116, 186 L mol⁻¹ cm⁻¹) than CSS (6850 L mol⁻¹ cm⁻¹) at 335 nm. This corresponds to the chromophores on the side chains of P-CSS aggregating, resulting in a favorable overlap with the emission spectra of UCPs. In order to evaluate the response of P-CSS under various light sources, the photolysis experiments (Figs. 1b-d) showed that P-CSS is significantly reduced in the 275-400 nm range when exposed to LED@365 nm, LED@405 nm, and Laser@980 nm (with UCPs). These indicate that P-CSS may cleavage when exposed to these light sources. A series of curve graphs of A₀ at λ_max = 335 nm and illumination time were summarized based on the photolysis data (Fig. 1f). Photoacid generation is one of the important indicators of cationic PIs. The increased absorbance at 550 nm, which belongs to the discolored Rhodamine B (Rh B), evidently confirms the generation of Brønsted acid (Fig. 1e, Fig. S2 in Supporting information).

  Figure 1

  

  Figure 1.  (a) UV–vis absorption spectrum of CSS and P-CSS in acetonitrile (1.18 × 10^-5 mol/L, ε=A/cL) and the emission spectrum of UCPs under Laser@980 nm, steady-state photolysis of P-CSS (6.0 × 10^-6 mol/L) in acetonitrile under irradiation of (b) LED@365 nm (100 mW/cm²), (c) LED@405 nm (100 mW/cm²), (d) Laser@980 nm (18 W/cm², UCPs = 10 wt%), (e) LED@365 nm (100 mW/cm²), Rh B (2.3 × 10^-5 mol/L) is added as an acid indicator. (f) Changes of the absorption at the wavelength of maximum absorption (λ_max = 335 nm): A₀ is the absorption of the maximum absorption wavelength before irradiation. A_t is the absorption of the maximum absorption wavelength at a certain irradiation time.

  DownLoad: Full-Size Img PowerPoint

  Photopolymerization kinetics investigations of P-CSS for FRP, CP, and HP were explored (Fig. 2, concentrations of P-CSS were selected in Fig. S3 (Supporting information), and the structures of monomers showed in Scheme S3 (Supporting information). As the kinetic profiles demonstrated, high conversions for FRP (LED@365 nm: HDDA, 84.8%, TPGDA, 84.5%, TMPTA, 70.1%; LED@405 nm: HDDA, 82.2%, TPGDA, 74.6%, TMPTA, 47.6%), CP (LED@365 nm: EPOX, 64.0%; LED@405 nm: EPOX, 48.5%), and HP (LED@365 nm: EPOX, 59.2%, TMPTA, 64.9%; LED@405 nm: EPOX, 57.0%, TMPTA, 39.9%) were attained under the irradiation of LED@365, 405 nm. It is worth noting that the speeds of the photopolymerization processes under LED@405 nm were slower than under LED@365 nm, which caused by P-CSS possessing a higher molar extinction coefficient at 356 nm. Photopolymerizations in solution were conducted to extend the application of P-CSS under 980 nm at room temperature (water bath with temperature monitor, Fig. S4 in Supporting information) and the conversions of monomers were calculated by weight, the molecular weight (M_n) and PDI of the polymerization products were obtained by gel permeation chromatography (GPC). Under constant temperature reaction at 980 nm, the final conversion rates of cyclohexene oxide (CHO) and MMA are 83.9% (30 min) and 51.1% (120 min), respectively (Table 1).

  Figure 2

  

  Figure 2.  Photopolymerization profiles of (a, b) cationic monomer (EPOX), (c, d) free radical monomers (HDDA, TPGDA, and TMPTA), and (e, f) the mixture of TMPTA and EPOX (1: 1) in the presence of P-CSS (1.6 × 10^-6 mol/g for free radical systems, 4.1 × 10^-7 mol/g for cationic systems, and 8.2 × 10^-7 mol/g for hybrid systems) under LED@365 nm (100 mW/cm²) and under LED@405 nm (100 mW/cm²) irradiation.

  DownLoad: Full-Size Img PowerPoint

  Table 1

  

  Table 1.  The photopolymerization results of CHO and MMA in dichloromethane (DCM) solution under Laser@980 nm (18 W/cm²) irradiation.

  DownLoad: CSV

  Monomer	CHOa	MMAb
  PIc (6.5 × 10-7 mol/L)	P-CSS	P-CSS
  Time (min)	30	120d
  Conversion (%)	83.9	51.1
  Mn (g/mol)	4471	595320
  PDI	1.4	4.9
  a The concentration of CHO was 3.35 mol/L. b The concentration of MMA was 5 mol/L. c UCPs existed at a proportion of 10 wt%. d Reaction in nitrogen.

  1, 1-Diphenylethylene (C₁₅H₁₄, m/z 195, Fig. S5 in Supporting information) was added to the solvent (DCM, deoxygened) of P-CSS and exposed to LED@365 nm (100 mW/cm²) for 20 min before the GC-MS test. According to GC-MS data, 1, 1-diphenylbutan (C₁₆H₁₈, m/z 210, Fig. S5) was detected, which can be attributed to the reaction between 1, 1-diphenylethylene and ethyl radicals. The initiation mechanism of P-CSS is speculated based on GC-MS and photoacid generation experiment (Scheme 2). After absorbing photons, P-CSS transitions to the excited state, followed by homolysis to produce sulfur free radical cations that can produce Brønsted acid (Fig. 1e) through hydrogen abstraction reaction to initiate CP and ethyl free radicals that can induce FRP.

  Scheme 2

  

  Scheme 2.  Proposed initiation mechanism for FRP and CP by using P-CSS.

  DownLoad: Full-Size Img PowerPoint

  Thermal gravimetric analysis (TGA) and differential scanning calorimetry (DSC) experiments suggest that P-CSS has excellent thermal stability and formula stability. The degradation temperature of P-CSS at 5% decomposition is 255 ℃ (Fig. S6 in Supporting information). In EPOX, the thermal polymerization temperature is 155.4 ℃ (Fig. S7a in Supporting information), and in TMPTA, it is 200.0 ℃ (Fig. S7b in Supporting information). Furthermore, two weeks of storage stability can be maintained through mixing P-CSS with TPGDA resin (Fig. S8 in Supporting information). The thermal stability, formula stability, and storage stability of P-CSS indicate that P-CSS offers a lot of promise for industrial applications.

  To compare the migration stability of P-CSS and CSS in polymers cured by TMPTA and EPOX, the relationships between concentration and fluorescence emission intensity of P-CSS and CSS in acetone solvent were calibrated (Fig. S9 in Supporting information), and the migration concentrations of PIs and their cleavage products from polymer powders to acetone were investigated and calculated (Fig. 3). It can be seen that the migration concentrations of P-CSS in the polymers of TMPTA and EPOX are 194.45 and 108.39 µmol/L, while the migration concentrations of CSS are 15498.78 and 5525.39 µmol/L, respectively. Migration concentrations of P-CSS in the polymer are only 1.25% (TMPTA) and 1.96% (EPOX) of CSS. According to the aforementioned findings, P-CSS demonstrates strong migration stability and significantly reduces the migration rate of PIs during the polymerization process.

  Figure 3

  

  Figure 3.  Concentrations of P-CSS and CSS extracted with acetone for polymers prepared by TMPTA and EPOX polymerization ([PIs] = 2 × 10^-5 mol/g).

  DownLoad: Full-Size Img PowerPoint

  In conclusion, a novel type of highly efficient cationic Macro-PI (P-CSS) was synthesized by copolymerization with MMA and characterized by ¹H NMR and GPC. P-CSS demonstrates a much higher molar extinction coefficient (116, 186 L mol⁻¹ cm⁻¹, 335 nm) and a redshift of 50 nm at the longest absorption wavelength than CSS possibly due to the chromophores aggregating on the side chain. Photopolymerization kinetics experiments have shown that P-CSS not only has high efficiency for FRP, CP, and HP under UV and visible light but is also applicable under NIR. GC-MS and photoacid generation experiments demonstrated that active species can be produce through homolysis of P-CSS. P-CSS also has good thermal stability, formula stability, and storage stability, which make it has great application prospects in industrial production. More importantly, the migration of P-CSS is significantly reduced compared to CSS (1.25% in TMPTA, 1.96% in EPOX), making it safer for processing polymer materials. In general, we have built an efficient, stable, and low migration cationic PI (P-CSS), which has great potential in biosafety and environmentally friendly applications.

  Declaration of competing interest

  The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

  CRediT authorship contribution statement

  Ying Chen: Writing – original draft, Visualization, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Lun Li: Methodology, Software, Data curation. Guohao Han: Formal analysis, Data curation. Ren Liu: Resources, Funding acquisition. Guanghui An: Writing – review & editing, Conceptualization. Yi Zhu: Writing – review & editing, Resources, Supervision, Project administration, Funding acquisition.

  Acknowledgments

  The authors acknowledge the financial support by the National Natural Science Foundation of China (Nos. 22301107, 52373057), Nature Science Foundation of Jiangsu Province (No. BK20242080), Key Laboratory of Synthetic and Biological Colloids, Ministry of Education, Jiangnan University (No. 1042050205243170/008) and State Key Laboratory of Molecular Engineering of Polymers, Fudan University (No. K2024-39).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2024.110458.

  
  
• 1. [1]

     Y. Yagci, S. Jockusch, N. Turro, Macromolecules 43 (2010) 6245–6260. doi: 10.1021/ma1007545

  2. [2]

     L. Pierau, C. Elian, J. Akimoto, et al., Prog. Polym. Sci. 127 (2022) 101517.

  3. [3]

     C. Yu, J. Schimelman, P. Wang, et al., Chem. Rev. 120 (2020) 10695–10743. doi: 10.1021/acs.chemrev.9b00810

  4. [4]

     F. Han, S. Gu, A. Klimas, et al., Science 378 (2022) 1325–1331. doi: 10.1126/science.abm8420

  5. [5]

     Y. Zhang, Y. Xu, A. Simon-Masseron, et al., Chem. Soc. Rev. 50 (2021) 3824–3841. doi: 10.1039/d0cs01411g

  6. [6]

     Y. Zhu, D. Xu, Y. Zhang, et al., Angew. Chem. Int. Ed. 60 (2021) 16917–16921. doi: 10.1002/anie.202104531

  7. [7]

     W. Wang, Y. Zhang, Y. Gao, et al., Eur. Polym. J. 208 (2024) 112874.

  8. [8]

     Y. Pang, S. Fan, Q. Wang, et al., Angew. Chem. Int. Ed. 59 (2020) 11440–11447. doi: 10.1002/anie.202004413

  9. [9]

     C. Han, B. Kundu, Y. Liang, et al., Adv. Mater. 36 (2024) 2307759. doi: 10.1002/adma.202307759

  10. [10]

      Z. Chen, X. Wang, S. Li, et al., ChemPhotoChem 3 (2019) 1077–1083. doi: 10.1002/cptc.201900007

  11. [11]

      Z. Chen, S. He, H. Butt, et al., Adv. Mater. 27 (2015) 2203–2206. doi: 10.1002/adma.201405933

  12. [12]

      Z. Li, X. Zou, F. Shi, et al., Nat. Commun. 10 (2019) 3560. doi: 10.1038/s41467-019-11522-0

  13. [13]

      P. Garra, J. Fouassier, S. Lakhdar, et al., Prog. Polym. Sci. 107 (2020) 101277.

  14. [14]

      Y. Zhu, J. Zhang, Y. Zhang, et al., Chin. Chem. Lett. 35 (2024) 109573.

  15. [15]

      Y. Peng, J. Zhang, C. Peng, et al., Food Packag. Shelf Life 32 (2022) 100853.

  16. [16]

      J. Scarsella, N. Zhang, T. Hartman, Molecules 24 (2019) 3592. doi: 10.3390/molecules24193592

  17. [17]

      R. Liu, S. Mabury, Environ. Sci. Technol. Lett. 6 (2019) 702–707. doi: 10.1021/acs.estlett.9b00626

  18. [18]

      L. Zheng, C. Lv, W. Cai, et al., Chin. Chem. Lett. (2024) 109922.

  19. [19]

      Y. Wu, J. Ke, C. Dai, et al., Eur. Polym. J. 175 (2022) 111380.

  20. [20]

      J. Zhou, X. Allonas, A. Ibrahim, et al., Prog. Polym. Sci. 99 (2019) 101165.

  21. [21]

      W. Qiu, J. Zhu, K. Dietliker, et al., ChemPhotoChem 4 (2020) 5296–5303. doi: 10.1002/cptc.202000146

  22. [22]

      S. Liang, Y. Yang, H. Zhou, et al., Prog. Org. Coat. 110 (2017) 128–133.

  23. [23]

      Y. Wu, R. Li, J. Wang, et al., J. Polym. Sci. 60 (2022) 52–61. doi: 10.1002/pol.20210586

  24. [24]

      S. Zhao, T. Xue, D. Pei, et al., Org. Lett. 23 (2021) 1709–1713. doi: 10.1021/acs.orglett.1c00131

  25. [25]

      Y. Han, F. Wang, C. Lim, et al., ACS Appl. Mater. Interfaces 9 (2017) 32418–32423. doi: 10.1021/acsami.7b08399

  26. [26]

      T. Li, Z. Su, H. Xu, et al., Chin. Chem. Lett. 29 (2018) 451–455.

  27. [27]

      E. Kandirmaz, E. Gençoğlu, N. Apohan, Macromol. Res. 27 (2019) 756–763. doi: 10.1007/s13233-019-7107-3

  28. [28]

      E. Kandirmaz, N. Apohan, E. Gençoğlu, Prog. Org. Coat. 119 (2018) 36–43.

  29. [29]

      J. Wei, R. Lu, F. Liu, J. Polym. Res. 18 (2011) 1001–1008. doi: 10.1007/s10965-010-9500-7

  30. [30]

      Z. Uyar, A. Öncel, J. Polym. Res. 25 (2018) 245. doi: 10.1007/s10965-018-1638-8

  31. [31]

      J. Jin, G. Lu, J. Nie, et al., Eur. Polym. J. 202 (2024) 112625.

  32. [32]

      W. Liao, M. Jin, Eur. Polym. J. 187 (2023) 111899.

  33. [33]

      X. Zhang, S. Keck, Y. Qi, et al., ACS Sustainable Chem. Eng. 8 (2020) 10959–10970. doi: 10.1021/acssuschemeng.0c03587

  34. [34]

      D. Kojic, K. Ehrmann, R. Wolff, et al., Polym. Chem. 14 (2023) 4809–4818. doi: 10.1039/d3py00787a

  35. [35]

      L. Zhang, L. Li, Y. Chen, et al., Polymers 15 (2023) 2524. doi: 10.3390/polym15112524

  36. [36]

      R. Taschner, T. Koch, R. Wolff, et al., ACS Appl. Polym. Mater. 5 (2023) 3023–3033. doi: 10.1021/acsapm.3c00191

  37. [37]

      Y. Yagci, Y. Durmaz, B. Aydogan, Chem. Rec. 7 (2007) 78-90. doi: 10.1002/tcr.20110

  38. [38]

      S. Silab, H. Bildirir, R. Dawson, et al., Macromolecules 47 (2014) 4607-4614. doi: 10.1021/ma501001m

  39. [39]

      Z. Uyar, M. Durgun, M. Yavuz, et al., Polymer 123 (2017) 153-168. doi: 10.1016/J.POLYMER.2017.07.019

  40. [40]

      Z. Uyar, F. Turgut, U. Arslan, et al., Eur. Polym. J. 119 (2019) 102–113. doi: 10.1016/J.EURPOLYMJ.2019.07.028

  41. [41]

      O. Karal, A. Önen, Y. Yaǧcı, Polymer 35 (1994) 4694–4696.

  42. [42]

      O. Taskin, I. Erel-Goktepe, M. Khan, et al., J. Photochem. Photobiol. A: Chem. 285 (2014) 30–36. doi: 10.1016/J.JPHOTOCHEM.2014.03.018

  43. [43]

      F. Dumur, Eur. Polym. J. 163 (2022) 110962.

  44. [44]

      S. Wet-osot, T. Pewklang, C. Duangkamol, et al., J. Mol. Struct. 1260 (2022) 132840. doi: 10.1021/acscombsci.6b00055

  45. [45]

      G. Knorr, K. Hotzel, A. Chettri, et al., J. Mater. Chem. A 11 (2023) 23260–23269. doi: 10.1039/d3ta04450e

  46. [46]

      Y. Zhu, L. Li, Y. Zhang, et al., Prog. Org. Coat. 174 (2023) 107272.

  47. [47]

      F. Petko, A. Świeży, M. Jankowska, et al., Polym. Chem. 14 (2023) 3018–3034. doi: 10.1039/d3py00359k

  48. [48]

      Y. Zhang, Z. Liu, T. Borjigin, et al., Green Chem. 25 (2023) 6881–6891. doi: 10.1039/d3gc02004e

  49. [49]

      K. Kaya, J. Kreutzer, Y. Yagci, J. Polym. Sci. A: Polym. Chem. 56 (2018) 451–457. doi: 10.1002/pola.28918

• 

  Scheme 1  Schematic diagram showing the previous work about cationic Macro-PIs and P-CSS in this work.

  下载: 全尺寸图片 幻灯片
  

  Figure 1  (a) UV–vis absorption spectrum of CSS and P-CSS in acetonitrile (1.18 × 10^-5 mol/L, ε=A/cL) and the emission spectrum of UCPs under Laser@980 nm, steady-state photolysis of P-CSS (6.0 × 10^-6 mol/L) in acetonitrile under irradiation of (b) LED@365 nm (100 mW/cm²), (c) LED@405 nm (100 mW/cm²), (d) Laser@980 nm (18 W/cm², UCPs = 10 wt%), (e) LED@365 nm (100 mW/cm²), Rh B (2.3 × 10^-5 mol/L) is added as an acid indicator. (f) Changes of the absorption at the wavelength of maximum absorption (λ_max = 335 nm): A₀ is the absorption of the maximum absorption wavelength before irradiation. A_t is the absorption of the maximum absorption wavelength at a certain irradiation time.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  Photopolymerization profiles of (a, b) cationic monomer (EPOX), (c, d) free radical monomers (HDDA, TPGDA, and TMPTA), and (e, f) the mixture of TMPTA and EPOX (1: 1) in the presence of P-CSS (1.6 × 10^-6 mol/g for free radical systems, 4.1 × 10^-7 mol/g for cationic systems, and 8.2 × 10^-7 mol/g for hybrid systems) under LED@365 nm (100 mW/cm²) and under LED@405 nm (100 mW/cm²) irradiation.

  下载: 全尺寸图片 幻灯片
  

  Scheme 2  Proposed initiation mechanism for FRP and CP by using P-CSS.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  Concentrations of P-CSS and CSS extracted with acetone for polymers prepared by TMPTA and EPOX polymerization ([PIs] = 2 × 10^-5 mol/g).

  下载: 全尺寸图片 幻灯片

  Table 1.  The photopolymerization results of CHO and MMA in dichloromethane (DCM) solution under Laser@980 nm (18 W/cm²) irradiation.

  Monomer	CHOa	MMAb
  PIc (6.5 × 10-7 mol/L)	P-CSS	P-CSS
  Time (min)	30	120d
  Conversion (%)	83.9	51.1
  Mn (g/mol)	4471	595320
  PDI	1.4	4.9
  a The concentration of CHO was 3.35 mol/L. b The concentration of MMA was 5 mol/L. c UCPs existed at a proportion of 10 wt%. d Reaction in nitrogen.

  下载: 导出CSV
•