A non-linear organic reaction of malonate derivative as a base amplifier to generate imidazoles without producing gas

Abstract

A malonate-type base amplifier that decomposes autocatalytically to generate imidazoles has been designed. This is the first example of a base amplifier that does not produce gas as a by-product.

---

UV curing is a key technology, and is used to produce fine electronic devices and coating films quickly.¹ Photoreactive materials used in the UV curing process generally consist of resins and photoinitiators that generate radical, cationic, and/or anionic species. Radical UV curing is a conventional process, although it has some problems of oxygen inhibition and volume contraction during the curing reactions. Cationic UV curing using photoacid generators (PAGs)^2,3 can be used to circumvent these problems, but metallic substrates are eroded by acidic species. Anionic UV curing with photobase generators (PBGs) has recently been featured, and much effort has been made to overcome the relatively low photosensitivity of the materials by improving the quantum yield of PBGs.^4,5

By contrast, as an alternative approach, our group has proposed to introduce base proliferation reactions in anionic UV curing systems.⁶ In this approach, base amplifiers (BAs) that decompose autocatalytically to generate basic species are used with PBGs. Even if a tiny amount of base is generated from PBGs, it works sufficiently as a trigger of the following autocatalytic decomposition reactions of BAs. As a result, basic species in the system “proliferate” in a non-linear manner, leading to apparently high quantum yields and high photosensitivity of the photoreactive materials. Several types of BAs, such as 9-fluorenylmethyl,^6a,c phenylsulfonylethyl,^6b nitropentanyl,⁷ and oxidized thioxanthenyl carbamates,⁸ have been designed and applied in various kinds of photoreactive materials.^9–14 However, these BAs have a serious drawback: gas generation that causes bubbles or cracks in the cured products.

In this communication, we report that a malonate derivative 1 amplifies basic species, as shown in Scheme 1. The α-proton of 1 is relatively acidic (pK_a = 13),¹⁵ and it is abstracted by a base. Subsequently, β-elimination is initiated to release an imidazole. Imidazoles are weak bases, and can act as cross-linking points in epoxy resins at relatively lower temperatures.¹⁶ Compound 1 has two bulky chiral alkyl groups, which ensures its thermal stability and its solubility in organic media. Because the decomposition mechanism is fundamentally different from carbamate-type BAs, compound 1 does not generate carbon dioxide. Decomposition behaviours of 1 were examined using spectral measurements, to ensure that 1 acts as a BA in some photosensitive materials.

Scheme 1 Autocatalytic decomposition reaction of 1.

Compound 1 was synthesized in three reaction steps in 20% total yield (see ESI†). The last step of the synthesis is a Michael reaction between an intermediate 1′ and 4-phenylimidazole. The product was assigned by ¹H–¹³C-NMR, FT-IR, and HR-MS measurements. It was found that the decomposition point of 1 is 121.1 °C by TG-DTA measurements.

BAs should fulfill the following three requirements.^6a First, BAs should be decomposed by a base catalyst to generate basic species, which results in the autocatalytic decomposition reactions. Second, BAs should be thermally stable, unless basic trigger species are generated in the reaction system. Third, the basic species from BAs should catalyse subsequent chemical reactions, which leads to a non-linear chemical transformation of resins. The decomposition behaviours of 1 were then examined in solution. Fig. 1(a) shows the decomposition behaviours of 1 with 13 mM of 4-phenylimidazole at 100 °C in toluene-d₈, monitored by ¹H-NMR. The doublet peak of the α-methine proton of 1 at 4.8 ppm disappeared, while the singlet peak of olefin proton of 1′ appeared at 8.4 ppm during heating. We considered that 4-phenylimidazole acts as a trigger of the decomposition reaction, and ca. 80% of 1 was decomposed within 300 min. The by-product of the decomposition reaction, 1′, was also generated quantitatively (Fig. 1(c), closed circles). The ¹H-NMR spectra indicate that no other by-products are generated, and that a simple reaction as shown in Scheme 1, a retro Michael reaction, proceeds under the conditions. By contrast, 1 is thermally stable at least for 200 min under the same conditions (Fig. 1(b) and (c), closed diamonds). After an induction period, 1 was decomposed in a non-linear manner. These results indicate that 1 fulfills the first and second requirements mentioned above, and that 1 could work as a BA.

Fig. 1 ¹H-NMR spectral changes of 1 (69 mM) in toluene-d₈ at 100 °C (a) in the presence of 4-phenylimidazole (13 mM), and (b) in the absence of 4-phenylimidazole. (c) The consumption of 1 (■) and the yield of 1′ (●) as a function of heating time in the presence of 4-phenylimidazole, as well as the conversion of 1 in the absence of 4-phenylimidazole (◆).

A similar decomposition behaviour was observed with 1 at 100 °C in a poly(tetramethylene ether glycol) (PTMG) matrix, monitored by FT-IR (Fig. 2). An increase of peak intensity at 1625 cm⁻¹ was observed that is assigned to the olefin moiety of 1′ (see ESI†). In the presence of a trigger base molecule, 4-phenylimidazole, 1, was decomposed drastically to give the by-product 1′. By contrast, in the scenario without base, the conversion of 1′ was sigmoidal with an induction period of 10 min. Similarly, it was confirmed that 1 could be used as a BA, and the results encouraged us to apply 1 for photoreactive materials with a PBG.

Fig. 2 The conversion of 1′ (40 wt% of 1 towards PTMG) (●) as a function of heating time in the presence of 4-phenylimidazole (30 mol% towards (1)), and the conversion of 1′ in the absence of 4-phenylimidazole (■) during heating at 100 °C in a PTMG (0.05 g) matrix.

At first, we evaluated the photosensitivity of poly(glycidyl methacrylate) (PGMA) films containing a PBG¹⁷ and 1 (see ESI†). We found that the sensitivity was four times improved by using 1 with PBGs, compared with the corresponding system using PBG alone. This finding shows that 1 also satisfied the third requirement. Because its characteristic is to avert the generation of low-molecular-weight gasses, 1 appears to be more effective in fabrication of UV-cured coating films. Subsequently, we sought to fabricate UV-cured coating films with the same PBG, a tetrafunctional epoxy resin (4,4′-methylenebis(N,N-diglycidylaniline, NN)), and 1. The hardness of the resulting cured films was evaluated using the pencil-hardness test (Japanese Industrial Standard JIS K5400).¹⁸ The hardness of the pencils is arranged as follows: 6B (softest), 5B, 4B, 3B, 2B, B, HB, F, H, 2H, 3H, 4H, 5H, 6H, 7H, 8H, 9H (hardest). The results are shown in Fig. 3. A value of HB was obtained, and a hard UV-cured coating film was successfully fabricated after 5000 mJ cm⁻² of 365 nm light irradiation and subsequent post-baking at 110 °C for 60 min. By contrast, the resin was not cured sufficiently without UV irradiation, even when it contained PBG and 1 under the same post-baking condition, and the hardness test result was 6B. FT-IR spectra measured at the same time show the difference in consumption behaviour. Under UV irradiation, half of the epoxy groups were consumed, indicating that phototriggered base proliferation reactions proceeded under these conditions, to cure the resins using base-catalysed addition reactions between the epoxy resins and “proliferated” 4-phenylimidazoles. However, only 30% of epoxy groups were consumed without UV irradiation, because the trigger bases are only generated by the thermal decomposition of 1, and the delayed initiation of proliferation reactions result in the inadequate curing.

Fig. 3 Results of pencil-hardness tests of UV-cured films of NN containing a PBG and 1 with/without 5000 mJ cm⁻² of UV irradiation and subsequent post-baking at 110 °C. The consumption behaviours of epoxy groups of the resin with (●) and without (■) the UV irradiation, monitored by FT-IR, are also shown in the same figure.

Finally, 1 was mixed with the same resin, which was baked between glass plates at 110 °C for 60 min. Although thermal decomposition of 1 would be caused, there was no remarkable bubble or crack after baking (Fig. 4, left). By contrast, in the corresponding sample of a conventional BA having an Fmoc group,^6a bubbles were clearly observed because of the generation of carbon dioxide gas following the thermal decomposition (Fig. 4, right).

Fig. 4 Generation of bubbles in NN films containing (left) 1 or (right) a conventional BA after baking at 110 °C.

Conclusions

In conclusion, a novel practical BA 1 that is decomposed autocatalytically into an imidazole without gas generation has been designed. We constructed photoreactive materials integrating this BA, demonstrating that it could even be used in a closed system free from bubble generation. It is also known that imidazoles cause chain polymerization at higher temperatures,^16a,19 and we will study and report on such applications in due course.

Notes and references

1. C. Decker, Polym. Int., 1998, 45, 133.
2. J. V. Crivello, J. Photopolym. Sci. Technol., 2009, 22, 575.
3. M. Shirai and M. Tsunooka, Bull. Chem. Soc. Jpn., 1998, 72, 2483.
4. M. Tsunooka, K. Suyama, H. Okamura and M. Shirai, J. Photopolym. Sci. Technol., 2006, 19, 65.
5. (a) T. Ida and K. Arimitsu, Polym. Prepr., 2011, 60, 1173; (b) K. Arimitsu and R. Endo, Chem. Mater., 2013, 25, 4461.
6. (a) K. Arimitsu, M. Miyamoto and K. Ichimura, Angew. Chem., Int. Ed., 2000, 39, 3425; (b) M. Miyamoto, K. Arimitsu and K. Ichimura, J. Photopolym. Sci. Technol., 1999, 12, 315; (c) K. Arimitsu, M. Miyamoto and K. Ichimura, J. Photopolym. Sci. Technol., 1999, 12, 317.
7. K. Arimitsu, Y. Ito, T. Gunji, Y. Abe and K. Ichimura, J. Photopolym. Sci. Technol., 2005, 18, 227.
8. K. Arimitsu, H. Kitamura, R. Mizuochi and M. Furutani, Chem. Lett., 2015, 44, 309.
9. T. Sasaki and H. Yaguchi, J. Polym. Sci., Part A: Polym. Chem., 2009, 47, 602.
10. (a) K. Aoki and K. Ichimura, Polym. J., 2009, 41, 988; (b) K. Aoki, T. Hashimoto and K. Ichimura, J. Photopolym. Sci. Technol., 2014, 27, 529.
11. K. H. Chae, C. S. Lee and J. H. Kim, Polym. Adv. Technol., 2011, 22, 1427.
12. H. Mohapatra, K. M. Schmid and S. T. Phillips, Chem. Commun., 2012, 48, 3018.
13. M. He, X. Huang, Z. Zeng and J. Yang, Macromolecules, 2013, 46, 6402.
14. K. Arimitsu, Y. Kumazawa and M. Furutani, J. Polym. Sci., Part A: Polym. Chem., 2015, 53, 2440.
15. J. McMurry, Organic Chemistry, Brooks/Cole, California, 5th edn, 2000, p. 913.
16. (a) A. Farbas and P. F. Strohm, J. Appl. Polym. Sci., 1968, 12, 159; (b) M. S. Heise and G. C. Martin, Macromolecules, 1989, 22, 99.
17. N. Ž. Knežević, B. G. Trewyn and V. S. Y. Lin, Chem.–Eur. J., 2011, 17, 3338.
18. JIS K5400, defined by Japanese Industrial Standards, is a simple method to test the scratch hardness of coatings. In this test, pencils in the range from 6B to 9H hardness grade are used. The pencil is moved scratching over the surface of the coating at a 45° angle with a constant force.
19. (a) A. Omrani, L. C. Simon, A. A. Rostami and M. Ghaemy, Int. J. Chem. Kinet., 2008, 40, 663; (b) K. R. Hart, N. R. Sottos and S. R. White, Polymer, 2015, 67, 174; (c) K. Kudo, M. Furutani and K. Arimitsu, ACS Macro Lett., 2015, 4, 1085.

---

Footnote

† Electronic supplementary information (ESI) available: Materials, equipment, synthesis of the base amplifier, characterisation, and experimental methods for applications. See DOI: 10.1039/c6ra04328c

---