Facile synthesis of functionalized polysiloxanes with nonconventional fluorescence by oxa-Michael addition reaction

ABSTRACT

Developing a novel methodology for the synthesis of functionalized polysiloxanes is highly desirable because of their extensive applications in engineering, textile, construction, electronics, and medical systems. Herein we introduce the oxa-Michael addition reaction as an efficient methodology to prepare a series of functionalized polysiloxanes from hydroxyalkyl-containing polysiloxanes and vinyl monomers using phosphazene base as the catalyst. The effects of various factors, including reaction time, feed ratio and solvent, were explored in this reaction. It was found that the reaction can be carried out under mild reaction (room temperature in 6 h to 24 h) with moderate to high conversion yields (60% to 99%). In particular, vinyl monomers with strong electron-withdrawing groups (e.g., cyano and sulfone groups) have higher reactivity, and the functionalized polysiloxanes can be obtained quantitatively or near-quantitatively. Interestingly, their molecular weights determined by gel permeation chromatography (GPC) reveal that the Si-O-Si skeletons were attacked by phosphazene base and re-arranged during the reaction, thus leading to improved molecular weights and more uniform molecular weight distributions in the final products than the original polysiloxanes. Moreover, these polysiloxanes exhibited intriguing nonconventional fluorescence due to the presence of unique chromophores, and those containing cyano and sulfone groups particularly exhibit the best fluorescence performance. More organosilicon materials could be developed under this simple strategy.

KEYWORDS:

• Polysiloxane
• oxa-Michael addition reaction
• nonconventional fluorescence

1. Introduction

Polysiloxanes as an important class of organosilicon materials have gained significant attention due to their unique properties, such as high/low temperature resistance, oxidation resistance, well insulation, low surface tension, and biocompatibility, and extensive applications in pharmaceuticals, household chemicals, food, construction, and other fields [1–8]. In particular, functionalized polysiloxanes have attracted specific attention because the introduction of functionality can largely broaden their applications. Functionalized polysiloxanes are commonly prepared by incorporating functional groups into polysiloxanes at the end of the polymer chain or on the side chain by selecting suitable methodologies. At present, the most mature and widely used methodology in laboratory and industry is the hydrosilylation reaction because of facile availability of Si-H containing molecules or polymers in commerce, which has been found for more than 80 years [9]. However, this reaction has some drawbacks. For example, the reliance on rare metal complexes (e.g., Pt and Rh) as catalysts results in the risk of residual toxic heavy metal elements [10]. It is still desirable to develop new synthetic strategies for the synthesis of functionalized polysiloxanes.

In the past decade, several efficient strategies have been developed for the construction of functionalized polysiloxanes by taking the advantage of well-developed organic and polymerization synthetic methodologies, such as click reaction [11–13], Piers-Rubinsztajn reaction [14] and dynamic chemistry [15]. Among these reactions, click technologies have attracted specific interests due to their intriguing advantages, such as mild reaction conditions, fast speed, high yields, and brilliant tolerance of functional groups. Considering the existing raw materials, typical examples include thiol – ene reaction (Scheme 1a-b) and aza-Michael addition reaction (Scheme 1c), which have been extensively applied as efficient methodologies to prepare functionalized polysiloxanes [16,17], as well as organosilicon compounds [18,19], and silicone elastomers [20,21], reported by us and other groups [22]. However, feedstocks containing amino or sulfhydryl groups for these reactions are potentially toxic and possess limited storage stability, which can not meet the requirements of green chemistry [23].

Scheme 1. Novel synthetic methodologies of functionalized polysiloxanes. (a-b) thiol–ene reaction; (c) aza-Michael addition reaction; (d) oxa-Michael addition reaction.

From the viewpoint of reaction mechanism, when the double bond in the starting materials for thiol – ene reaction and aza-Michael addition reaction is linked by the electron-withdrawing groups (EWGs), these two reactions belong to the classical Michael addition reaction and thiol-ene reaction is also named thiol-Michael addition reaction. Similar to them, when the nucleophilic center is an oxygen atom, it is referred to the oxa-Michael addition reaction (Scheme 1d). Compared to other Michael addition reactions, oxa-Michael addition offers significant advantages because the hydroxyl components are commonly stable, odorless and non-toxic. However, it has not been well developed due to the weak reactivity of the hydroxyl group, although it has been widely used in organic synthesis [24,25]. Recent reports demonstrated that this reaction can also possess the feature of click reaction since a green, safe, and efficient catalyst, phosphazene base, was found, and has been used to prepare co-polymers and hyperbranched polymers [23,26]. In addition, it can be also utilized to prepare shape memory polymers based on natural rubber by crosslinking epoxidized natural rubber with zinc diacrylate [27,28]. However, the exploration in the preparation of polymers is still limited. In particular, the utilization in the preparation of organosilicon materials is still not explored.

Herein, we introduce the oxa-Michael addition reaction to prepare a series of functionalized polysiloxanes using hydroxyalkyl-containing polysiloxanes and vinyl-containing compounds as the starting materials and phosphazene base t-BuP₂ as the catalyst. Although previous reports have proved the effectiveness of t-BuP₂ in the oxa-Michael addition reaction [23,26], it remains necessary to optimize reaction condition when preparing functionalized polysiloxanes because of the sensitivity of Si-O and Si-C bonds to strong base. Thus, the reaction condition was optimized and it was found that this reaction can be carried out under mild conditions. The structures of the resulting polysiloxanes were characterized by Fourier transform infrared spectra (FT-IR), ¹H NMR, and ¹³C NMR. The molecular weights and polymer dispersity index (PDI) of the products were determined by GPC. Moreover, it was found that these new polysiloxanes exhibit interesting nonconventional fluorescence.

2. Experimental

2.1. Materials and instrumentation

Bis[[3-(2-hydroxyethoxy)propyl]dimethylsilyl]-terminated polydimethylsiloxane (P1, M_n = 8011 g/mol, determined by GPC) and poly[dimethylsiloxane-co-methyl[3-(2-hydroxyethoxy) propyl]siloxane] (P2, M_n = 2039 g/mol, determined by GPC, 10 mol% of (2-hydroxyethoxy)propyl) were purchased from Dongyue Chemical Co. Ltd, China. Acrylonitrile, methyl vinyl sulfone, phenyl vinyl sulfone, methyl acrylate, and ethyl acrylate were purchased from Energy Chemical Co., China. Phosphazene base (t-BuP₂) solution (2 M in THF) was purchased from Sigma-Aldrich Co., China.

Fourier transform infrared spectra (FT-IR) were performed on a Bruker Tensor 27 infrared spectrophotometer with KBr pellets technique in 4000–400 cm⁻¹ region, with scanning 16 times and a resolution of 4 cm⁻¹. ¹H and ¹³C NMR spectra were recorded on a Bruker Avance 400 MHz spectrometer at 25°C, using CDCl₃ as solvent and without an interior label. Fluorescence emission spectra were determined on a Hitachi F-4500 fluorescence spectrometer using a monochromated Xe lamp as an excitation source. The excitation slit and emission slit are 5 nm or 10 nm with the voltage of 400 V or 700 V. The molecular weights were estimated by an Agilent 1260 infinity II GPC/SEC System, calibrated with standard polystyrene, using THF as the mobile phase at a flow rate of 1 mL/min with a constant column temperature of 30°C. Two chromatographic columns of bead size 5 μm (PL MIXED-C 7.5 × 50 mm, PL MIXED-C 7.5 × 300 mm) were used, monitoring with a differential refractive index detector.

2.2. Synthetic procedures

The typical procedure was described as follows. Hydroxyalkyl-functionalized polysiloxane (P1 or P2), vinyl monomers (1.0 equiv. to -OH group), and t-BuP₂ (5 mol% equiv. to -OH group) were dissolved in a certain amount of dichloromethane (DCM) (2 mL per 1 mmol of vinyl monomers). The resultant mixture was stirred at room temperature for 24 h. Upon the completion of the reaction, excessive acetic acid was added to the mixture to terminate the reaction. After removing DCM by rotary evaporation, the rest was washed with 50% ethanol aqueous solution to remove excess acetic acid and unreacted vinyl monomers, and then dried in a vacuum drying oven, resulting in the final products.

Data of P1T1. P1T1 was synthesized by the reaction of P1 and T1 and obtained as a colorless transparent liquid. Yield: 97%. FT-IR (KBr, cm⁻¹): 2963, 2901, 1704, 1410, 1261, 1020, 798, 697. ¹H NMR (400 MHz, CDCl₃, ppm) δ 0.00 (s, -SiCH₃), 0.45 (t, 4 H, -SiCH₂CH₂CH₂O-), 1.48–1.61 (m, 4 H, -SiCH₂CH₂CH₂O-), 2.55 (t, 4 H, -OCH₂CH₂CN), 3.36 (t, 4 H, -SiCH₂CH₂CH₂O-), 3.52 (td, 4 H, -OCH₂CH₂OCH₂CH₂CN), 3.60 (td, 4 H, -OCH₂CH₂OCH₂CH₂CN), 3.66 (t, 4 H, -OCH₂CH₂CN). ¹³C NMR (100 MHz, CDCl₃, ppm) δ 0.00, 13.02, 17.82, 22.30, 64.93, 68.91, 69.78, 73.30, 116.78. M_n = 10179 g/mol, M_w = 14474 g/mol, PDI = 1.42.

Data of P1T2. P1T2 was synthesized by the reaction of P1 and T2 and obtained as a colorless transparent liquid. Yield: 98%. FT-IR (KBr, cm⁻¹): 2963, 2906, 1550, 1409, 1319, 1260, 1093, 866, 798, 693. ¹H NMR (400 MHz, CDCl₃, ppm) δ 0.00 (s, -SiCH₃), 0.44 (t, 4 H, -SiCH₂CH₂CH₂O-), 1.45–1.58 (m, 4 H, -SiCH₂CH₂CH₂O-), 2.95 (s, 6 H, -SO₂CH₃), 3.15 (t, 4 H, -OCH₂CH₂SO₂-), 3.33 (t, 4 H, -SiCH₂CH₂CH₂O-), 3.50 (td, 4 H, -OCH₂CH₂OCH₂CH₂SO₂-), 3.58 (td, 4 H, -OCH₂CH₂OCH₂CH₂SO₂-), 3.86 (t, 4 H, -OCH₂CH₂SO₂-). ¹³C NMR (100 MHz, CDCl₃, ppm) δ 0.00, 13.10, 22.36, 42.10, 54.31, 63.90, 68.55, 69.67, 73.14. M_n = 13648 g/mol, M_w = 19312 g/mol, PDI = 1.42.

Data of P1T3. P1T3 was synthesized by the reaction of P1 and T3 and obtained as a colorless transparent liquid. Yield: 91%. FT-IR (KBr, cm⁻¹): 2963, 2906, 1409, 1319, 1260, 1093, 866, 798, 693. ¹H NMR (400 MHz, CDCl₃, ppm) δ 0.00 (s, -SiCH₃), 0.41 (t, 4 H, -SiCH₂CH₂CH₂O-), 1.43–1.57 (m, 4 H, -SiCH₂CH₂CH₂O-), 3.26 (t, 4 H, -SiCH₂CH₂CH₂O-), 3.29–3.38 (m, 8 H, -OCH₂CH₂OCH₂CH₂SO₂-), 3.41 (td, 4 H, -OCH₂CH₂OCH₂CH₂SO₂-), 3.77 (t, 4 H, -OCH₂CH₂SO₂-), 7.45–7.51 (m, 4 H, phenyl), 7.54–7.60 (m, 2 H, phenyl), 7.82–7.87 (m, 4 H, phenyl). ¹³C NMR (100 MHz, CDCl₃, ppm) δ 0.00, 13.04, 22.28, 55.19, 63.51, 68.64, 69.55, 73.23, 127.10, 128.08, 132.61, 138.89. M_n = 9702 g/mol, M_w = 11740 g/mol, PDI = 1.41.

Data of P1T4. P1T4 was synthesized by the reaction of P1 and T4 and obtained as a colorless transparent liquid. Yield: 96%. FT-IR (KBr, cm⁻¹): 2963, 2902, 1746, 1409, 1261, 1019, 797, 697. ¹H NMR (400 MHz, CDCl₃, ppm) δ 0.00 (s, -SiCH₃), 0.44 (t, 4 H, -SiCH₂CH₂CH₂O-), 1.45–1.60 (m, 4 H, -SiCH₂CH₂CH₂O-), 2.54 (t, 4 H, -OCH₂CH₂ COO-), 3.36 (t, 4 H, -SiCH₂CH₂CH₂O-), 3.50 (td, 4 H, -OCH₂CH₂OCH₂CH₂COO-), 3.54 (td, 4 H, -OCH₂CH₂OCH₂CH₂COO-), 3.62 (s, 6 H, -COOCH₃), 3.70 (t, 4 H, -OCH₂CH₂COO-). ¹³C NMR (100 MHz, CDCl₃, ppm) δ 0.00, 13.07, 22.34, 33.84, 50.59, 65.54, 68.89, 69.43, 73.22, 171.01. M_n = 14297 g/mol, M_w = 23934 g/mol, PDI = 1.67.

Data of P1T5. P1T5 was synthesized by the reaction of P1 and T5 and obtained as a colorless transparent liquid. Yield: 97%. FT-IR (KBr, cm⁻¹): 2962, 2904, 1743, 1409, 1261, 1097, 796, 698. ¹H NMR (400 MHz, CDCl₃, ppm) δ 0.00 (s, -SiCH₃), 0.44 (t, 4 H, -SiCH₂CH₂CH₂O-), 1.19 (t, 6 H, -COOCH₂CH₃), 1.45–1.60 (m, 4 H, -SiCH₂CH₂ CH₂O-), 2.53 (t, 4 H, -OCH₂CH₂COO-), 3.34 (t, 4 H, -SiCH₂CH₂CH₂O-), 3.50 (td, 4 H, -OCH₂CH₂OCH₂CH₂COO-), 3.54 (td, 4 H, -OCH₂CH₂OCH₂CH₂COO-), 3.69 (t, 4 H, -OCH₂CH₂COO-), 4.07 (q, 4 H, -COOCH₂CH₃). ¹³C NMR (100 MHz, CDCl₃, ppm) δ 0.00, 13.07, 13.17, 22.34, 34.09, 59.41, 65.65, 68.90, 69.42, 73.22, 170.57. M_n = 9430 g/mol, M_w = 14000 g/mol, PDI = 1.48.

Data of P2T1. P2T1 was synthesized by the reaction of P2 and T1 and obtained as a light-yellow transparent liquid. Yield: 98%. FT-IR (KBr, cm⁻¹): 2962, 2971, 1619, 1411, 1260, 1078, 807, 694. ¹H NMR (400 MHz, CDCl₃, ppm) δ 0.00 (s, -SiCH₃), 0.42 (t, 2 H, -SiCH₂CH₂CH₂O-), 1.45–1.60 (m, 2 H, -SiCH₂CH₂CH₂O-), 2.54 (t, 2 H, -OCH₂ CH₂CN), 3.34 (t, 2 H, -SiCH₂CH₂CH₂O-), 3.50 (td, 2 H, -OCH₂CH₂OCH₂CH₂CN), 3.58 (td, 2 H, -OCH₂CH₂OCH₂CH₂CN), 3.64 (t, 2 H, -OCH₂CH₂CN). ¹³C NMR (100 MHz, CDCl₃, ppm) δ 0.00, 12.28, 18.12, 22.31, 65.25, 69.24, 70.09, 73.30, 117.03. M_n = 5771 g/mol, M_w = 6740 g/mol, PDI = 1.17.

Data of P2T2. P2T2 was synthesized by the reaction of P2 and T2 and obtained as a light-yellow transparent liquid. Yield: 96%. FT-IR (KBr, cm⁻¹): 2962, 2868, 1409, 1260, 1020, 801, 695. ¹H NMR (400 MHz, CDCl₃, ppm) δ 0.00 (s, -SiCH₃), 0.42 (t, 2 H, -SiCH₂CH₂CH₂O-), 1.46–1.61 (m, 2 H, -SiCH₂CH₂CH₂O-), 2.94 (s, 3 H, -SO₂CH₃), 3.15 (t, 2 H, -OCH₂CH₂SO₂-) 3.32 (t, 2 H, -SiCH₂CH₂CH₂O-) 3.50 (td, 2 H, -OCH₂CH₂ OCH₂CH₂SO₂-) 3.57 (td, 2 H, -OCH₂CH₂OCH₂CH₂SO₂-) 3.85 (t, 2 H, -OCH₂CH₂SO₂-). ¹³C NMR (100 MHz, CDCl₃, ppm) δ 0.00, 12.23, 22.05, 42.07, 54.29, 63.88, 68.53, 69.64, 72.97. M_n = 5982 g/mol, M_w = 8944 g/mol, PDI = 1.50.

Data of P2T3. P2T3 was synthesized by the reaction of P2 and T3 and obtained as a light-yellow transparent liquid. Yield: 88%. FT-IR (KBr, cm⁻¹): 2962, 2872, 1618, 1448, 1408, 1317, 1260, 1089, 803, 690. ¹H NMR (400 MHz, CDCl₃, ppm) δ 0.00 (s, -SiCH₃), 0.38 (t, 2 H, -SiCH₂CH₂CH₂O-), 1.43–1.56 (m, 2 H, -SiCH₂CH₂CH₂O-), 3.25 (t, 2 H, -SiCH₂CH₂CH₂O-), 3.29–3.38 (m, 4 H, -OCH₂CH₂OCH₂CH₂SO₂-), 3.40 (td, 2 H, -OCH₂CH₂OCH₂CH₂SO₂-), 3.76 (t, -OCH₂CH₂SO₂-), 7.44–7.51 (m, 2 H, phenyl), 7.54–7.60 (m, 1 H, phenyl), 7.82–7.87 (m, 2 H, phenyl). ¹³C NMR (100 MHz, CDCl₃, ppm) δ 0.00, 12.29, 21.97, 55.17, 63.49, 68.58, 69.52, 73.07, 127.08, 128.07, 132.60, 138.87. M_n = 6130 g/mol, M_w = 8708 g/mol, PDI = 1.42.

Data of P2T4. P2T4 was synthesized by the reaction of P2 and T4 and obtained as a colorless transparent liquid. Yield: 95%. FT-IR (KBr, cm⁻¹): 2962, 2895, 1741, 1409, 1261, 1186, 1082, 805, 695. ¹H NMR (400 MHz, CDCl₃, ppm) δ 0.00 (s, -SiCH₃), 0.43 (t, 2 H, -SiCH₂CH₂CH₂O-), 1.46–1.62 (m, 2 H, -SiCH₂CH₂CH₂O-), 2.54 (t, 2 H, -OCH₂ CH₂COO-), 3.34 (t, 2 H, -SiCH₂CH₂CH₂O-), 3.49 (td, 2 H, -OCH₂CH₂OCH₂CH₂COO-), 3.53 (td, 2 H, -OCH₂CH₂OCH₂CH₂COO-), 3.61 (s, 3 H, -COOCH₃), 3.69 (t, 2 H, -OCH₂CH₂COO-). ¹³C NMR (100 MHz, CDCl₃, ppm) δ 0.00, 12.32, 22.03, 33.82, 50.56, 65.57, 68.86, 69.40, 73.04, 170.99. M_n = 7042 g/mol, M_w = 8873 g/mol, PDI = 1.26.

Data of P2T5. P2T5 was synthesized by the reaction of P2 and T5 and obtained as a colorless transparent liquid. Yield: 97%. FT-IR (KBr, cm⁻¹): 2962, 2864, 1744, 1410, 1261, 1083, 804, 696. ¹H NMR (400 MHz, CDCl₃, ppm) δ 0.00 (s, -SiCH₃), 0.42 (t, 2 H, -SiCH₂CH₂CH₂O-), 1.19 (t, 3 H, -COOCH₂CH₃), 1.46–1.63 (m, 2 H, -SiCH₂CH₂ CH₂O-), 2.52 (t, 2 H, -OCH₂CH₂COO-), 3.33 (t, 2 H, -SiCH₂CH₂CH₂O-), 3.49 (td, 2 H, -OCH₂CH₂OCH₂CH₂COO-), 3.53 (td, 2 H, -OCH₂CH₂OCH₂CH₂COO-), 3.69 (t, 2 H, -OCH₂CH₂COO-), 4.07 (q, 2 H, -COOCH₂CH₃). ¹³C NMR (100 MHz, CDCl₃, ppm) δ 0.00, 12.33, 13.16, 22.04, 34.08, 59.37, 65.64, 68.87, 69.40, 73.05, 170.55. M_n = 8218 g/mol, M_w = 10821 g/mol, PDI = 1.76.

3. Results and discussion

3.1. Optimization of the reaction condition

As illustrated in Scheme 2, functionalized polysiloxanes were prepared by the oxa-Michael addition reactions of polysiloxane containing hydroxyalkyl groups at the end of the polymer chain (P1) or on the side chain (P2) with various vinyl monomers (T1~T5). Previous reports have proved that phosphazene base t-BuP₂ is an efficient catalyst for the oxa-Michael addition reaction [23,26]. To determine the suitability of this catalyst, the reaction condition was optimized using the reaction of P1 and methyl acrylate (T4) as the representative example. The results are summarized in Table 1.

Scheme 2. Synthetic routes of functionalized polysiloxanes by the oxa-Michael addition reaction.

Table 1. The optimization of reaction condition of P1T4.

Entry	[C=C]/[−OH]	Equiv of t-BuP2	Temp. (°C)	Time	Conversion (%) ^a	Transesterification A (%) ^b
1	1	0.05	25	3 h	32	13
2	1	0.05	25	6 h	40	9
3	1	0.05	25	9 h	49	6
4	1	0.05	25	12 h	54	4
5	1	0.05	25	24 h	60	6
6	1	0.05	25	7 d	61	6
7	2	0.05	25	24 h	57	13
8	3	0.05	25	24 h	48	17
9^c	1	0.05	25	24 h	61	6

^aCalculated by ¹H NMR; ^bThe ratio of transesterification product A (Scheme 3) is calculated by ¹H NMR; ^cThe reaction was conducted in DCM (0.5 M).

Scheme 3. Plausible mechanism of oxa-Michael addition reaction of P1 and methyl acrylate (T4).

To find the appropriate reaction time, the reaction conversion at 3 h, 6 h, 9 h, 12 h, 24 h, and 7 d were monitored by ¹H NMR, and the results are shown in Table 1 (Entries 1–6). As shown in Figure 1, with an increment of the reaction time, the peaks of P1 at 3.46 ppm and 3.66 ppm (H^e and H^f) attributable to methylene decrease, while the peaks at 3.70 ppm and 2.54 ppm (H^g and H^h) assigned to the methylene in the product P1T4 increase. Comparing the integration ratios of the peaks at 2.54 ppm (H^h) and 0.44 ppm (H^b and H^b’), the result reveals that the reaction conversion yield, which is calculated based on the consumption of – OH group by ¹H NMR, gradually increases to ca. 60% within 24 h and no obvious change was found even after 7 days.

Figure 1. [1] H NMR spectra of P1T4 at different reaction times of 3 h, 6 h, 9 h, 12 h, 24 h, and 7 days.

Based on previous results [23], transesterification reaction also occurred during the oxa-Michael addition reaction of P1 and T4 due to the presence of strong base t-BuP₂. As shown in Scheme 3, when the transesterification reaction occurs, the by-product A with the alkyl chain connecting to the ester group and methanol is generated. The newly generated alkyl chain has a stronger ability to push electrons than the methyl group in T4, and the density of the electron cloud in the ester group will increase. Thus, this side reaction may play a negative impact on the process of nucleophilic addition. The side product A can further react with P1 or methanol, resulting in by-product B. This speculation can be confirmed by the ¹H NMR spectra of the product. As shown in Figure 2, the peaks of double bonds (Hⁱⁱ and Hⁱⁱⁱ) can be obviously observed, although the product has been dried exhaustively under vacuum. Two classes of peaks corresponding to the methylene groups (Hⁱ and H^i*) from the by-products A and B were found. This finding further proves the occurrence of the transesterification reaction. The transesterification ratio of the by-product A for P1T4 were calculated from the integration of the signals at 0.44 ppm (H^b*) and 4.33 ppm (Hⁱ) shown in Figure 2 and summarized in Table 1. In addition, the transesterification reaction also occurred during the reaction of P2 and T4 (Figure S24). Considering that if the R is CH₂CH₂OCH₂CH₂CH₂Si in by-product B, crosslinking between the chain of P2 May occur. However, this phenomenon was not observed. We speculate that the possibility of CH₂CH₂OCH₂CH₂CH₂Si is much lower than that of -CH₃ because the reaction activity between the by-product A and methanol is much higher than that between A and P2.

Figure 2. [1] H NMR spectra of P1T4 with transesterification products.

To further improve the conversion yield of P1T4, excess T4 was attempted with the molar ratios of 1:1, 1:2, and 1:3 (Table 1, Entries 5, 7, and 8). The results show that with an increment of T4, the conversion yield decreases, while the transesterification reaction is easier to occur due to the presence of excessive T4. As a result, partial hydroxyl groups connected by polysiloxane was consumed and more transesterification product A was generated. However, the by-product A has lower reaction activity and hinders the oxa-Michael addition reaction, thus leading to the reduction of conversion yield. In addition, considering some vinyl monomers (e.g., T2 and T3) are not miscible with polysiloxanes. The reaction was also carried out in solvent and DCM was selected as an example (Table 1, Entry 9). The result shows that the addition of a small amount of solvent does not have an obvious effect on the conversion yield.

Therefore, the appropriate reaction condition for the oxa-Michael reaction is using equivalent hydroxyalkyl-containing polysiloxane and vinyl monomers (to – OH group), and 5 mol% of t-BuP₂ (to – OH group), in the presence or absence of solvent at room temperature for 24 h.

3.2. Synthesis of functionalized polysiloxane via oxa-Michael addition reaction

Based on the optimized condition, a series of functionalized polysiloxanes were prepared based on hydroxyalkyl-containing polysiloxanes and various vinyl monomers, including acrylonitrile (T1), methyl vinyl sulfone (T2), phenyl vinyl sulfone (T3), methyl acrylate (T4), and ethyl acrylate (T5) (Scheme 1 and Table 2). The results reveal that the conversion yields are similar when the same monomer reacts with P1 or P2, indicating that the position of the hydroxyalkyl group in the polysiloxane does not have a significant impact on the conversion efficiency.

Table 2. Reaction data for different functionalized polysiloxane.

Entry	Polymers	Vinyl monomers	Temp. (℃)	Time	Conversion (%)
1	P1	(T1)	25	24 h	95
2	P1	(T1)	25	6 h	95
3	P1	(T2)	25	24 h	97
4	P1	(T3)	25	24 h	99
5	P1	(T4)	25	24 h	60
6	P1	(T5)	25	24 h	79
7	P2	(T1)	25	24 h	99
8	P2	(T2)	25	24 h	92
9	P2	(T3)	25	24 h	99
10	P2	(T4)	25	24 h	61
11	P2	(T5)	25	24 h	74

^aThe conversion yields were calculated by ¹H NMR.

As expected, the vinyl monomers with strong electron-withdrawing groups (T1, T2, and T3) have higher reactivity and the functionalized polysiloxanes can be obtained quantitatively or near-quantitatively (the conversion yields are > 95%, Entries 1, 3–4, and 7–9). Moreover, after shortening the reaction time from 24 h to 6 h, the conversion rate of P1T1 is still high (95%, Entry 2). The ester group has a relatively lower electron-withdrawing ability, and the conversion rates are moderate. When using ethyl acrylate (T5), the conversion yields are 79% and 74% for P1T5 and P2T5, respectively. The values are higher than P1T4 (60%) and P2T4 (61%). This finding may be due to the different transesterification degrees.

3.3. Molecular weights of functionalized polysiloxanes

The molecular weights and PDI of functionalized polysiloxanes were determined by GPC (Table S1). As expected, compared to P1 and P2, the molecular weight of the products increased. It is worth noting that the increment is greater than the molecular weight of the functional group introduced by the oxa-Michael addition reaction and the PDIs were reduced after functionalization. This finding can be explained by the re-arrangement of the polysiloxane segments during the oxa-Michael addition reaction. Based on previous reports, phosphazene base has been proven as a kind of efficient catalyst for the ring-opening polymerization of cyclosiloxane [29,30]. Therefore, t-BuP₂ may also attack the Si-O bond, while severing as the catalyst. Subsequently, partial Si-O skeletons were destructed and rearranged, resulting in an improved molecular weight and a more uniform PDI. For example, compared to P1, the increased molecular weight of P1T1 is ca. 108 g/mol in theory due the addition of acrylonitrile. In contrast, the M_n of P1T1 is found to be 10,179 g/mol with the increased molecular weight of ca. 2000 g/mol and narrower PDI of 1.42, in comparison to the M_n of 8011 g/mol and PDI of 1.54 of P1.

3.4. Fluorescent properties of functionalized polysiloxanes

Previous reports have demonstrated that molecules or polymers containing functional groups such as cyano [31], sulfone [32], and carbonyl [33] can exhibit interesting nonconventional fluorescence due to the aggregation of chromophores. Considering the presence of these groups in the obtained polysiloxanes, their fluorescent properties were investigated. As shown in Figure 3, all the functionalized polysiloxanes in dichloromethane (DCM) solutions emit blue fluorescence under UV excitation at 365 nm. This finding can be explained by the charge transfer emission (CTE) mechanism, consistent with previous reports [34]. Induced by the chain entanglement, the chromophores (e.g., cyano and sulfone) can approach each other, and clusters or aggregation was formed. Thus, space electronic communications, including electron overlap between lone pairs and π electrons, dipole – dipole interactions, and n–π interactions were afforded and an extended electronic conjugation was yielded. The newly generated conjugation facilitated excitation and the radiative deactivation, thus generating fluorescence. In addition, P1 and P2 also display weak blue emission, but the intensity is lower than the products. This finding is probably due to the aggregation of -OH groups [35].

Figure 3. Fluorescence emission spectra of P1, P2 and functionalized polysiloxanes (PxTy, x = 1 or 2; y = 1 to 5) in DCM (10 mg/mL, λ_ex = 365 nm).

Among these polysiloxanes, those containing cyano or methylsulfonyl groups (e.g., P1T1 and P1T2) possess higher fluorescence intensity than those containing phenylsulfonyl and ester groups (e.g., P1T3 and P1T4). This phenomenon can be explained in three aspects. The first is higher polarity of cyano and sulfone groups than ester group, thus leading to stronger interactions (e.g., dipole – dipole interactions) when forming the clusters and facilitating the formation of space electronic communications. The second is the lower reactivity of acrylate (e.g., T4 and T5) than cyano or sulfonyl groups (e.g., T1 and T2), resulting in incomplete reaction conversion (e.g., P1T4 and P1T5), and thus there is less content of chromophores at the same concentration with P1T1 and P1T2. The third is ease or difficulty for the formation of aggregation. This speculation could also explain the finding that the polysiloxanes containing phenylsulfonyl groups (P1T3 and P2T3) exhibit lower fluorescence intensity than those containing methylsulfonyl groups (P1T2 and P2T2). Although they all contain sulfone groups in the structures, phenyl groups in P1T3 and P2T3 may partially hinder the aggregation of sulfone groups due to the π-π interactions among phenyl groups, unlike the easy formation of aggregation among methylsulfonyl groups in P1T2 and P2T2.

Considering better fluorescence of cyano- and methylsulfonyl containing polysiloxanes, P2T1 and P2T2 were selected as representative examples and their fluorescent properties were further investigated. As expected, the intensity of nonconventional fluorescence is closely related to the concentration of chromophores in the system. The fluorescence intensities of P2T1 and P2T2 increase as the concentration increases, and a linear correlation between fluorescence intensity and concentration was found (Figure 4a-b). Moreover, P2T1 and P2T2 exhibit significant excitation-dependent fluorescence emissions. As the excitation wavelength increases from 280 nm to 400 nm, the fluorescence intensity decreases significantly and the maximum emission wavelength is red-shifted from 331 nm to 435 nm (Figure 4c) and from 333 nm to 455 nm (Figure 4d) for P2T1 and P2T2, respectively. The possible mechanism is that the nonconventional chromophores can form different clusters with different extents of space conjugations [36].

Figure 4. (a-b) fluorescence emission spectra of P2T1 (a) and P2T2 (b) dissolved in DCM (λ_ex = 365 nm) under different concentrations; (c-d) fluorescence emission spectra of P2T1 (c) and P2T2 (d) in DCM (10 mg/mL) under different excitation wavelength.

In addition, the solvent effect on the fluorescence was also studied. The sample in polar solvents (e.g., DMF) exhibits better fluorescence performance than those dissolved in non-polar solvents (e.g., hexane) at the same concentration (Figure 5). This finding may be due to the poorer affinity between non-polar polysiloxanes and polar solvents, and thus the aggregation of chromophores is easy to form, resulting in the enhanced conjugation degree and fluorescence intensity.

Figure 5. Fluorescence emission spectra of P2T1 (a) and P2T2 (b) in different solvents (10 mg/mL, λ_ex = 365 nm).

4 Conclusions

In conclusion, we have prepared a series of functionalized polysiloxanes by oxa-Michael addition reaction based on hydroxyalkyl-containing polysiloxanes and vinyl monomers. The effects of reaction times, feed ratios and the addition of a small amount of solvent on the conversion rate were investigated by taking the preparation of P1T4 as an example. The results demonstrated that the reaction can be conducted under mild conditions by using 5 mol% of t-BuP₂ as the catalyst in the presence or absence of solvent at room temperature for 24 h and the conversion yields of the products are moderate to high from 60% to 99%. The structures of the products were fully characterized by FT-IR, ¹H NMR, and ¹³C NMR. The molecular weights determined by GPC reveal that the products possess an enhanced molecular weight and a more uniform molecular weight distribution than the original polysiloxanes due to the re-arrangement of the polysiloxane segments induced by the strong base t-BuP₂. Moreover, these polysiloxanes exhibit intriguing non-conventional fluorescence and their fluorescent properties depend on functional group species, solution concentration, excitation wavelength and solvent polarity. These results demonstrate that the oxa-Michael addition reaction is a facile and efficient strategy to prepare functionalized polysiloxanes with unique properties. Compared with traditional hydrosilylation reaction, the oxa-Michael addition can be carried out in the absence of metal catalysts with high efficiency. In future, this reaction can be utilized to prepare various organosilicon materials with new structures and functionality, such as organosilicon block co-polymers, hyperbranched polysiloxanes, and silicone elastomers, and the application scope of organosilicon materials could be expanded by exploring their applications, such as wearable devices, e-skin, sensing and biomedicine. Additionally, further optimization of the reaction conditions may be pursued in future investigations, especially in the crosslinking system, because the residual of strong base t-BuP₂ in the system may cause the un-stability of the materials.

Acknowledgments

This research was supported by Fluorine Silicone Materials Collaborative Fund of Shandong Provincial Natural Science Foundation (ZR2020LFG011 and ZR2021LFG001), National Natural Science Foundation of China (No. 22271175 and 52173102), and Young Scholars Program of Shandong University.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data that support the findings of this study are available from the corresponding author, D. Wang, upon reasonable request.

Additional information

Funding

This work was supported by the National Natural Science Foundation of China [52173102]; National Natural Science Foundation of China [22271175]; Fluorine Silicone Materials Collaborative Fund of Shandong Provincial Natural Science Foundation [ZR2020LFG011]; Fluorine Silicone Materials Collaborative Fund of Shandong Provincial Natural Science Foundation [ZR2021LFG001].