Versatile Manganese Catalysis for the Synthesis of Poly(silylether)s from Diols and Dicarbonyls with Hydrosilanes

Abstract

Poly(silylether)s are interesting materials because of their degradation property under hydrolytic conditions and have been prepared via hydrosilylation polymerization from dicarbonyl and hydrosilanes, and via dehydrogenative cross-coupling of diols and hydrosilanes under catalytic conditions. Here, we present a manganese–salen compound based on an inexpensive and nontoxic metal that could effectively catalyze both polymerization reactions with hydrosilane. A series of poly(silylether)s containing various aliphatic and aromatic backbones have been synthesized from diol and dicarbonyl substrates. Moderate to high yields of polymers with number-average molecular weights up to 15 kg/mol are obtained. Because of the dual activity of the manganese catalyst, unsymmetrical substrates with mixed functional groups, such as p-hydroxybenzaldehyde, p-hydroxy benzylalcohol, and 3-(4-hydroxyphenyl)-1-propanol, have been employed to afford poly(silylether)s with multiple silicon connectivity in the main chain.

Introduction

Polymers with silicon in the main chain have received much attention in organic and material chemistry because of their various industrial and academic applications.¹⁻⁶ Polysilanes with the Si–Si repeating units are promising for use in electronic devices because of their conductivity.⁷ Poly(siloxane)s with Si–O–Si linkages and their copolymers are extensively used as elastomers and plastics, and in other industrial applications because of their low-temperature flexibility and high-temperature stability.⁸⁻¹⁵ Hydro-sensitive silicon polymers can be employed in medical applications such as controlled release of drugs.¹⁶ Poly(siloxane)s are precursors for the synthesis of silicon oxycarbides (SiOC) via pyrolysis and direct photo-cross-linking methods.^17,18 The SiOC ceramic has been applied in coatings and as electrode materials in lithium batteries.¹⁹⁻²⁴ Recently, there have been renewed interests in silicon-containing polymers, particularly poly(silylether)s with Si–O–C linkages, on account of their potential sustainability.^25,26 Silicon and oxygen are practically inexhaustible because of their abundance, and carbon can be more readily sourced from biomass. The Si–O–C linkages are hydrolytically degradable, different from the typical biodegradable ester linkages that require enzymes for degradation. This makes the poly(silylether)-based materials particularly attractive in short-term and/or single-use applications.²⁷ Importantly, the degradation behavior, along with the thermomechanical properties, can be adjusted by changing the substituent groups on silicon and/or carbon backbones of poly(silylether)s²⁸⁻³⁰ and by copolymerization with other segments.³¹

Poly(silylether)s have been prepared by various methods depending on the nature of the actual linkages. Similar to the synthesis of silylethers, the reaction of dichlorosilanes with diols through polycondensation leads to the formation of poly(silylether)s.³²⁻³⁸ Chlorosilanes can also react with bis(epoxide)s and bis(oxetane)s through additions catalyzed by quaternary onium salts to afford poly(silylether)s.³⁹⁻⁴² However, use of chlorosilanes has limitations.⁴³ Chlorosilanes are moisture-sensitive and usually produce hydrogen chloride and other unwanted byproducts in polymerization reactions, which requires additional methods for separation.⁴⁴ To overcome these problems, chlorosilanes have been replaced with diamino- and dialkoxysilanes as the coupling partners; in particular, diphenoxy- and dianilinosilanes have shown good activity toward the synthesis of high-molecular-weight polymers.^14,45 Still these methods may have limited substrate scope or require maintaining the high-temperature conditions (200–300 °C) throughout the reaction.^14,45−48

Alternatively, hydrosilanes have been deemed as optimal replacements for chlorosilanes because of their stability to air and ease of handling. Reactions with diols through dehydrogenative coupling^28,49−51 and with dicarbonyls through hydrosilylation polymerization^29,52−54 afford a variety of poly(silylether)s (Scheme 1). Notably, these methods are highly atom-economical, producing H₂ as the sole byproduct or no byproduct. In general, these reactions are effected by catalysts derived from precious transition metals, usually ruthenium, palladium, and rhodium.^{49,52,55−57} However, the high cost, low abundance, and high catalyst loading (up to 10 mol %) of such metals are the main drawbacks of these methods.⁵⁸ Other catalysts based on boranes⁵⁹ and alkali metals^25,60 are also known. Recently, the catalytic systems derived from inexpensive, earth-abundant metals such as iron have been reported.^61,62

Scheme 1. Poly(silylether)s from Hydrosilanes.

We have investigated the high-valent transition-metal complexes for their roles in catalytic reduction and silane activation.⁶³⁻⁶⁶ It is observed that an air-stable and easily prepared salen–manganese complex, [MnN(salen-3,5-^tBu₂)]⁶⁷ (1), is an effective catalyst for hydrosilylation of carbonyl compounds and dehydrogenative coupling of hydroxyl compounds with hydrosilanes.^68,69 Encouraged by these results and in connection with our interests in biodegradable materials,⁷⁰⁻⁷² we sought to synthesize poly(silylether)s from a variety of diols and dicarbonyls under manganese catalysis. Furthermore, taking advantage of the dual activity of 1, we have employed substrates with mixed functional groups in the reaction with hydrosilanes to generate poly(silylether)s. As far as we are aware, no metal catalyst has been reported that could catalyze the synthesis of poly(silylether)s from three types of substrates (diols, dicarbonyls, and hydroxyl carbonyls) with hydrosilanes.

Results and Discussion

Optimization Reactions

Our initial experiments started with optimizing suitable reaction conditions for polycondensation of diols and hydrosilanes. 1,4-Benzenedimethanol and diphenylsilane were chosen as the representative coupling partners in the presence of catalytic amount of manganese complex 1 (1 mol %), and selected results are presented in Table 1. According to the conditions in our previous work,⁶⁸ the reaction was performed using acetonitrile as the solvent at reflux temperature (entry 1). As expected, the initial green color of the reaction mixture arising from 1 turned brown and then yellow, indicating onset of the reaction. After 12 h, ¹H NMR spectroscopy showed greater than 95% conversion of diphenylsilane. However, despite the high conversion, the number-average molecular weight (M_n) of the polymer was determined by gel permeation chromatography (GPC) to be 4000 g/mol (M_w/M_n = 1.28). To improve M_n, several solvents were screened next. When acetonitrile was replaced by tetrahydrofuran (THF) (entry 2), low-M_n (1900 g/mol) polymers were obtained with only 50% conversion of diphenylsilane in 12 h. Use of 1,4-dioxane as a solvent resulted in 95% conversion and high molecular weight (M_n = 7900 g/mol) (entry 3), suggesting that the coordination ability of cyclic ethers might not be the main reason for the low conversion observed with THF. Reactions carried out in refluxing toluene led to even higher M_n (9200 g/mol) within 12 h (entry 4). Given the above observations, it seemed that the reaction temperature played a key role in the polymerization reaction. In agreement with this, the reaction performed at ambient temperature in toluene showed no reaction even after 48 h and the green color remained the same during that period (entry 5). However, attempts at achieving high molecular weights using high-boiling aromatic solvents such as xylenes and mesitylene met with limited success (entries 6 and 7). Although the molecular weights obtained under refluxing conditions were decent, they were lower than those obtained with toluene. Finally, as air extrusion seemed to have little effect in our previous dehydrogenative coupling,⁶⁹ a reaction was performed under air in refluxing toluene for convenience (entry 8). However, conversion of monomers remained low (58%) even after extending the reaction time to 48 h and M_n also dropped approximately by half (4500 g/mol). These results suggested the importance of inert conditions for the generation of high-molecular-weight polymers.

Table 1. Dehydrogenative Coupling of 1,4-Benzenedimethanol with Diphenylsilane^a.

entry	solventb	temp.	condition	t (h)	conv%	Mn (g/mol)c	Mw/Mnc
1	CD3CN	reflux (81 °C)	under N2	12	>95	4000	1.28
2	THF	reflux (66 °C)	under N2	12	50	1900	1.36
3d	1,4-dioxane	reflux (101 °C)	under N2	20	>95	7900	1.57
4e	toluene	reflux (110 °C)	under N2	12	>95	9200	1.66
5	toluene	25 °C	under N2	48	0
6f	xylenes	reflux (143 °C)	under N2	12	>95	6000	1.14
7	mesitylene	reflux (165 °C)	under N2	12	>95	7200	1.23
8	toluene	reflux (110 °C)	under air	48	58	4500	1.15

^aReaction conditions: substrate, 0.8–0.9 mmol; silane, 1.0 equiv; and Mn catalyst 1, 1.0 mol %.

^bSolvent used was 2.4–2.8 mL.

^cDetermined by GPC calibrated with polystyrene standard.

^dIsolated yield 75.9%.

^eIsolated yield 75.6%.

^fIsolated yield 70%.

The brown-colored reaction mixtures from entries 3 and 4 were purified by precipitation. The resulting whitish-gray poly(silylether)s were obtained in 75.9 and 75.6% yields, respectively. In the ¹H NMR spectrum, the disappearances of hydroxyl groups at 2.0 ppm and silane hydrogens at 4.84 ppm and the shift of benzylic protons from 4.65 to 4.81 ppm suggested the formation of poly(silylether) of 1,4-benzenedimethanol and diphenylsilane (Figure 1). The ¹³C NMR spectrum also supported the formation of poly(silylether), as only six peaks in the aromatic region (129.71–139.47 ppm, four from silane and two from diol) and one peak at 64.98 ppm (benzylic carbon) were observed. In addition, the absence of both hydroxyl (3500 cm^–1) and SiH groups (2133 cm^–1) in the attenuated total reflection Fourier transform infrared (ATR FT-IR) spectrum further supported the NMR assignments (Figure 2). The characteristic stretching frequencies at ∼1040 cm^–1 indicated the formation of Si–O–C connections.⁵⁰

Figure 1.

¹H (left) and ¹³C (right) NMR spectra of the poly(silylether) from 1,4-benzenedimethanol and diphenylsilane.

Figure 2.

ATR FT-IR spectra of diphenylsilane (bottom), 1,4-benzenedimethanol (middle), and their poly(silylether) (top).

Step-Growth Polymerization

To learn more about the polymerization, we performed a set of reactions between 1,4-benzenedimethanol and diphenylsilane to explore the correlation among the conversion, time, and molecular weight of the polymer. During the reaction, samples were taken at each time interval and then were subjected to analysis by NMR and GPC. Conversion of diol was 19% in 1 h and reached around 50% in 4 h. However, at this point, the molecular weight of the polymer was only 1600 g/mol (M_w/M_n = 1.33), indicating that mostly oligomers were formed in the starting hours. At 8 h, the conversion increased to 91% with M_n of 4600 g/mol, and at 22 h, the conversion was 96% with M_n of 13000 g/mol. The sharp increase in the molecular weight in the later stage of the reaction suggested that the oligomers formed in the initial reaction stage were still active and were later incorporated to form long-chain polymers. The growth behavior of molecular weight (Figure 3) is typical of step-growth polymerization reactions.⁷³ The similar nonlinear growth of the dispersities (M_w/M_n) also supported the fact that these are step-growth-type polymerization reactions.

Figure 3.

Plot of number-average molecular weight (M_n, filled brown circle) and dispersity (M_w/M_n, filled blue square) vs conversion.

Diols Scope

To extend the substrate scope, we first chose a simple aryldiol, 1,4-benzenediol (1,4-hydroquinone), because polymers with π-conjugated groups in the main or side chains are of interest for their prospective applications as electronic, photonic, and ceramic materials (Table 2, entry 2).⁷⁴⁻⁷⁶ The copolymerization reaction between 1,4-benzenediol and a disiloxane (1,3-bis(diethylamino)tetramethyldisiloxane) under melt conditions was reported in the literature; however, the reported molecular weight of the polymer was low (M_n = 3600 g/mol).³³ In our current dehydrogenative coupling process between 1,4-benzenediol and diphenylsilane, >95% conversion of silane was observed in 12 h and a white solid was obtained in 70% yield after purification. The disappearance of hydroxyl and silicon hydrogen peaks in both ¹H NMR and FT-IR spectra confirmed the formation of the desired poly(silylether). In the ²⁹Si NMR spectrum, a single peak at −35.2 ppm was observed. GPC analysis showed a high-molecular-weight polymer with M_n of 15 000 g/mol (M_w/M_n = 2.38). To expand on aryl diols, a dehydrogenative coupling reaction between a highly π-conjugated 4,4′-biphenol and diphenylsilane was performed (entry 3) under the same conditions. The conversion reached 95% in 15 h, and a whitish-gray polymer was obtained in 88% yield after purification. However, the molecular weight obtained was somewhat lower, M_n = 7200 g/mol (M_w/M_n = 1.87).

Table 2. Dehydrogenative Coupling of Symmetrical Diols with Diphenylsilane^a.

^aReaction conditions: substrate, 0.8–0.9 mmol; silane, 1.0 equiv; and MnN catalyst, 1.0 mol %. The reactions were carried out in refluxing toluene under N₂, and the conversions of silane were greater than 95%.

^bDetermined by GPC calibrated with polystyrene standard.

^cIsolated yield.

^dDetermined by differential scanning calorimetry (DSC) in the second heating cycle.

Next, we carried out polycondensation reactions with aliphatic diols, namely, 1,4-cyclohexanedimethanol and 1,4-cyclohexanediol (Table 2, entries 4 and 5). These two substrates can be viewed as the aliphatic analogues of 1,4-benzenedimethanol and 1,4-benzenediol, respectively. The reaction between primary diol 1,4-cyclohexanedimethanol and diphenylsilane was monitored up to 24 h, by which time silane conversion of 95% was achieved. Purification afforded the corresponding poly(silylether) with 62% isolated yield, and the observed M_n was 5100 g/mol (M_w/M_n = 1.4). The reaction with a bulkier secondary diol, 1,4-cyclohexanediol, required a longer reaction time, as only 80% conversion was achieved after 20 h, although the isolated yield was good (70%). Given the slow conversion, it was not surprising that the molecular weight of the resulting polymer was rather low, M_n = 3100 g/mol (M_w/M_n = 1.73). In agreement with the low molecular weight, small peaks at around 5.3–5.4 ppm were observed in the ¹H NMR spectrum (Figure S9), which could be attributed to the tertiary Si–H end group in the polymer. Our previous results showed that diphenylsilane reacted with secondary alcohols in a two-step process, but the second step was considerably slower than the first step.⁶⁸ Similarly here, the steric hindrance plays a vital role in these two polycondensation reactions, as reflected in the extended reaction times and low molecular weights of the polymers.

Conversely, when a sterically less hindered linear monomer, 1,6-hexanediol, was used as the substrate, a high-molecular-weight polymer, M_n = 11400 g/mol, was produced in high yields (84%) (Table 2, entry 6). These results suggest that sterically less hindered monomers react readily to give the high-molecular-weight polymers. On the other hand, when another diol, 1,4-butanediol, was used, cyclic silylether was obtained as the main product instead of poly(silylether)s.^68,77 Obviously a rigid and/or long linker between the two hydroxyl groups of diols is desirable for the formation of poly(silylether)s.

Dicarbonyls

Considering our previous results that manganese complex 1 is also an efficient catalyst for the hydrosilylation of aldehydes and ketones,⁶⁹ we decided to apply the current system to dicarbonyl compounds for the synthesis of poly(silylether)s. In this study, benzene-1,4-dicarboxaldehyde (terephthalaldehyde) was chosen as the reaction partner with diphenylsilane, in part, because the resulting polymer should be the same as the polymer obtained from 1,4-benzenedimethanol (Table 2, entry 1 vs Table 3, entry 1). Under the optimized conditions, the consumption of terephthalaldehyde and the hydrosilylation of the carbonyl group were confirmed by the appearance of benzylic signals at 4.81 ppm in the ¹H NMR spectrum and at 65.0 ppm in the ¹³C NMR spectrum. The conversion of terephthalaldehyde reached 90% in 24 h, and workup by precipitation gave 58% isolated yield of a polymer. The FT-IR spectrum of the polymer showed the disappearance of the C=O stretching frequency at 1663 cm^–1 and of silane hydrogens at 2134 cm^–1, and a single peak at −30.3 was observed in the ²⁹Si NMR spectrum. These spectroscopic data are essentially the same as those resulting from 1,4-benzenedimethanol. The main difference is that the molecular weight is low, M_n = 2400 g/mol (M_w/M_n = 1.86). The low molecular weight is supposed to come from the less active nature of carbonyls in comparison with that of the hydroxyl groups in the reaction. The monitoring of the reaction profile of terephthalaldehyde and 1,4-benzenedimethanol under the same conditions confirms the faster reaction of diol (Figure S36). This difference in reactivity with hydrosilanes has been observed in other systems and is attributed to the stronger nucleophilicity of hydroxyls over that of carbonyls (see below).⁷⁸ To further compare the activity of dicarbonyl monomers with that of diols, reactions were performed with 1,6-hexanedial⁷⁹ and 1,4-cyclohexanedione, the analogous structures of 1,6-hexanediol and 1,4-cyclohexanediol, respectively (Table 3, entries 2 and 3). As expected, the reaction with 1,6-hexanedial took 24 h to reach 90% conversion and the isolated yield was only 40%; it took even longer (48 h) for 1,4-cyclohexanedione to reach 90% conversion with a slightly higher isolated yield (49%). Both reactions resulted in low-molecular-weight polymers, M_n = 1800 and 2100 g/mol, respectively, consistent with the slow rate of reaction. The low yields in these reactions, despite the >90% conversion, might be a result of the increased presence of short-chain oligomers in these reactions that were lost during the precipitation process.

Table 3. Poly(silylether)s of Dicarbonyls, Hydroxy Carbonyls, and Unsymmetrical Diols^a.

^aReaction conditions: substrate, 0.8–0.9 mmol; silane, 1.0 equiv; and MnN catalyst, 1.0 mol %.

^bDetermined by GPC calibrated with polystyrene standard.

^cIsolated yield.

^dDetermined by DSC in the second heating cycle.

Substrates with Mixed Functional Groups

So far all of the substrates used are structurally symmetrical with two identical functional groups. Encouraged by the results that poly(silylether)s can be prepared from both diols and dicarbonyls using the same manganese catalyst, we sought to employ substrates with unsymmetrical, mixed functional groups to further extend the substrate scope. Thus, p-hydroxybenzaldehyde was selected as a monomer to react with diphenylsilane (Table 3, entry 4). Although the reaction time (18 h) required to reach 90% conversion of silane was less than that of the three reactions in entries 1–3, the isolated yield and the molecular weight (M_n = 3800 g/mol, M_w/M_n = 1.73) were greater than those of the three reactions. Relatively speaking, these values fall right in between those obtained from 1,4-benzenediol and terephthalaldehyde, in line with the difference in their respective reactivities toward polymerization reactions under manganese catalysis. As substrate p-hydroxybenzaldehyde has two different termini, it can form polymers with different connectivities, which was evidenced by the presence of multiple peaks for the benzylic signals in both the ¹H and ¹³C NMR spectra (Figures S17 and S18). These observations were consistent with the ²⁹Si NMR spectrum, in which three peaks at −37.1, −33.5, and −30.1 ppm were observed (Figure S30), as predicted from the three different silicon environments resulting from the head-to-head, head-to-tail, and tail-to-tail connections (Figure 4).

Figure 4.

Three silicon centers in the poly(silyl ether) backbone.

Similarly, the reaction of p-hydroxybenzyl alcohol with diphenylsilane (entry 5) yielded a polymer with a slightly higher molecular weight (M_n = 4500 g/mol) that features three ²⁹Si NMR signals (Figure S31), as expected. It should be noted that the relative intensities of the three peaks are somewhat different from the results obtained with p-hydroxybenzaldehyde. The reaction with another unsymmetrical monomer, 3-(4-hydroxyphenyl)-1-propanol, gave comparable results (entry 6): more than 95% conversion and 80% isolated yield were achieved in 19 h. The molecular weight (M_n = 9000 g/mol) was high when compared with that of polymers from dicarbonyls and hydroxyl carbonyl. Again three distinct peaks appeared in the ²⁹Si NMR (Figure S32) spectrum, in agreement with the formation of three different types of connectivities. The high yield and M_n constitute additional evidence that hydroxyl compounds are faster than carbonyls in the polymerization reactions.

From these results and early studies, a plausible mechanism can be proposed for the manganese-catalyzed polymerization of diols and dicarbonyls (Scheme 2). The high-valent Mn(V) species is first reduced by diphenylsilane to a low-valent Mn(II) or Mn(III) species,⁶⁸ which correlates with the color changes during the early stage of the reaction. The reduction of Mn(V) is not unexpected given the oxidizing power of high-valent first-row transition metals. Hydrosilane is then activated by coordination with the metal center via either an η¹- or an η²-SiH adduct,⁸⁰ which may serve as a common intermediate for hydrosilylation and dehydrogenative cross-coupling. Although not directly observed in the present system, such adducts with low-valent manganese are well-documented in the literature.⁸¹⁻⁸³ Nucleophilic attack of the hydroxyl or carbonyl groups on the silicon furnishes the Si–O bonds, which is accompanied by the hetereolytic cleavage of the Si–H bonds. The observation that diols are more active with hydrosilanes than with dicarbonyls is in agreement with the stronger nucleophilicity of hydroxyls over that of carbonyls.⁷⁸ It should be mentioned that the oxidative addition of hydrosilane leading to a classical silyl hydride species cannot be ruled out at this stage;⁸⁴⁻⁸⁶ however, our experimental observation that the present manganese catalyst is ineffective in catalyzing the hydrosilylation of allylcyclohexane is more consistent with a Mn–SiH adduct intermediate.⁷⁸ The exact process following the nucleophilic attack may vary, and a comprehensive investigation is needed to reveal these mechanistic details.

Scheme 2. Plausible Mechanism for Polymerization.

Conclusions

In conclusion, we have demonstrated that a salen–manganese complex can effectively catalyze the synthesis of various poly(silylether)s from hydrosilanes and diols via dehydrogenative cross-coupling and from dialdehyde/diketone via hydrosilylation polymerization. Diol monomers show a higher activity than that of dicarbonyls in these reactions, as reflected in the higher yields and molecular weights obtained under comparable conditions. Furthermore, because of the dual activity of the manganese catalyst, unsymmetrical monomers with mixed functional groups have been successfully incorporated into poly(silylether)s that contain different connectivities around the silicon center. The use of a catalyst derived from an earth-abundant, biocompatible metal via highly atom-economical processes represents a sustainable strategy for the production of hydrolytically degradable materials. Further work on improving the catalytic efficiency and integrating bio-based building blocks is underway in our laboratory.

Experimental Section

Materials and Instrument

All of the solvents and liquid substrates were degasified and dried over activated molecular sieves (4 Å) prior to use. Deuterated solvents were purchased from Cambridge Isotope Laboratory. All ¹H and ¹³C NMR spectra were recorded on a Bruker AVANCE-500 NMR spectrometer and referenced to the residue solvents in CDCl₃ or CD₃CN. GPC analysis was performed on a Varian Prostar system, using the PLgel 5 μm Mixed-D column, a Prostar 355 RI detector, and THF as the eluent at a flow rate of 1 mL/min (20 °C). Polystyrene standards were used for calibration. Infrared spectroscopy results were obtained on a Thermo Scientific Nicolet iS5 FT-IR instrument and analyzed with OMNIC 8.2 software. The samples for IR analysis were directly loaded using an iD5 ATR accessory as a thin film without any support. DSC measurements were performed on a Perkin Elmer Jade differential scanning calorimeter, and the instrument was calibrated using zinc and indium standards. Glass-transition temperatures (T_g) of polymer samples were determined from the second heating cycles with a heating/cooling rate of 20 °C/min under nitrogen atmosphere (20 mL/min). DSC data were analyzed using Pyris V9.0.2 software.

General Procedure for the Synthesis of Poly(silylether)s

All of the reactions were performed under inert conditions unless otherwise mentioned. A Schlenk flask (50–100 mL) was used as the reaction flask and oil baths equipped with digital thermometers and controllers were used to set and read the temperature. The general procedure includes loading the Schlenk flask with 1 mol % catalyst [MnN(salen-3,5-^tBu₂)] (1), stoichiometric equivalents of substrate and diphenylsilane (1:1 molar ratio), followed by the addition of 2.5–3.0 mL of solvent in the glove box. The resulting reaction mixture was taken out of the glove box and connected to a Schlenk line under inert conditions. Then, the reaction mixture was heated at reflux temperature for a specific time (mentioned in the article). The reaction was monitored by periodically taking out small amount of samples for NMR and GPC analyses.

The polymers were purified by the precipitation method. As all of the polymers were soluble in dichloromethane (DCM) and insoluble in methanol (MeOH), these two solvents were used in the precipitation process. At the end of the reaction (>90% conversion), the brown color viscous reaction mixture was first homogenized by the addition of as low as possible amount of DCM (1–2 mL), then MeOH was added portionwise (8–10 mL) until it turned to a biphasic mixture. The top layer, which has the unreacted materials, was taken out by a pipette, and the bottom viscous/solid layer was washed with MeOH a few times (typically two to three times with 6–8 mL of MeOH each time) until it gave a white/light yellow color viscous/solid polymer. The resulting polymer was dried to a constant weight and characterized by ¹H NMR, ¹³C NMR, and FT-IR spectroscopies and GPC and DSC.

Table 2, Entry 1: Poly(silylether) of 1,4-Benzenedimethanol

Scale: catalyst 5.0 mg, substrate 0.89 mmol. Polymer weight 214.0 mg, yield 76%. ¹H NMR (500 MHz, CDCl₃, 298 K, δ): 4.81 (s, 4H, OCH₂), 7.26 (m, 4H, C₆H₄), 7.33 (m, 4H, m-Ph), 7.38 (m, 2H, p-Ph), 7.70 (m, 4H, o-Ph). ¹³C {¹H} NMR (125 MHz, CDCl₃, 298 K, δ): 64.9 (OCH₂). 126.7 (o-C₆H₄), 128.1 (m-Ph), 130.6 (p-Ph), 132.5 (i-Ph), 135.1 (o-Ph), 139.4 (i-C₆H₄). ²⁹Si {¹H} NMR (99 MHz, CDCl₃, 298 K, δ): −30.4. T_g: 22 °C.

Table 2, Entry 2: Poly(silylether) of 1,4-Benzenediol

Scale: catalyst 5.0 mg, substrate 0.89 mmol. Polymer weight 207.0 mg, yield 70%. ¹H NMR (500 MHz, CDCl₃, 298 K, δ): 6.64 (m, 4H, C₆H₄), 7.28 (m, 4H, m-Ph), 7.36 (m, 2H, p-Ph), 7.63 (m, 4H, o-Ph). ¹³C {¹H} NMR (125 MHz, CDCl₃, 298 K, δ): 120.5 (C₆H₄), 128.2 (m-Ph), 131.0 (p-Ph), 131.5 (i-Ph), 135.2 (o-Ph), 148.9 (i-C₆H₄). ²⁹Si {¹H} NMR (99 MHz, CDCl₃, 298 K, δ): −35.2.

Table 2, Entry 3: Poly(silylether) of 4,4′-Biphenol

Scale: catalyst 5.0 mg, substrate 0.89 mmol. Polymer weight 299.0 mg, yield 88%. ¹H NMR (500 MHz, CDCl₃, 298 K, δ): 6.64 (m, 4H, C₁₂H₈), 7.26 (m, 4H, C₁₂H₈), 7.28 (m, 4H, m-Ph), 7.36 (m, 2H, p-Ph), 7.63 (m, 4H, o-Ph). ¹³C {¹H} NMR (125 MHz, CDCl₃, 298 K, δ): 120.2 (C₁₂H₈), 128.0 (m-Ph), 128.3 (C₁₂H₈), 131.1 (p-Ph), 131.4 (i-Ph), 134.8 (C₁₂H₈), 135.3 (o-Ph), 153.4 (i-C₁₂H₈). ²⁹Si {¹H} NMR (99 MHz, CDCl₃, 298 K, δ): −37.4. T_g: 101 °C.

Table 2, Entry 4: Poly(silylether) of 1,4-Cyclohexanedimethanol

Scale: catalyst 5.0 mg, substrate 0.89 mmol. Polymer weight 184.8 mg, yield 62%. ¹H NMR (500 MHz, CDCl₃, 298 K, δ): 0.98 (m, 4H, (CH₂) C₆H₁₀), 1.54 (m, 2H, (CH) C₆H₁₀), 1.86 (m, 4H, (CH₂) C₆H₁₀), 3.77 (m, 4H, OCH₂), 7.41 (m, 6H, m- & p-Ph), 7.66 (m, 4H, o-Ph). ¹³C {¹H} NMR (125 MHz, CDCl₃, 298 K, δ): 29.2 ((CH₂) C₆H₁₀), 40.6 ((CH) C₆H₁₀), 68.7 ((CH₂) C₆H₁₀), 127.9 (m-Ph), 130.3 (p-Ph), 133.5 (i-Ph), 135.1 (o-Ph). ²⁹Si {¹H} NMR (99 MHz, CDCl₃, 298 K, δ): −29.9. T_g: 38 °C.

Table 2, Entry 5: Poly(silylether) of 1,4-Cyclohexanediol

Scale: catalyst 5.0 mg, substrate 0.89 mmol. Polymer weight 222.5 mg, yield 70%. ¹H NMR (500 MHz, CDCl₃, 298 K, δ): 1.41 (m, 4H, (CH₂) C₆H₁₀), 1.84 (m, 4H, (CH₂) C₆H₁₀), 3.91 (m, 2H, (OCH) C₆H₁₀), 5.46 (m, 1H, SiH), 7.35 (m, 4H, m-Ph), 7.39 (m, 2H, p-Ph), 7.64 (m, 4H, o-Ph). ¹³C {¹H} NMR (125 MHz, CDCl₃, 298 K, δ): 30.7, 32.3 ((CH₂) C₆H₁₀), 68.9, 70.1 ((OCH) C₆H₁₀), 127.7, 128.1 (m-Ph), 130.1, 131.3 (p-Ph), 134.3, 134.7 (i-Ph), 135.1, 135.1 (o-Ph). ²⁹Si {¹H} NMR (99 MHz, CDCl₃, 298 K, δ): −35.4, −35.6. T_g: 60 °C.

Table 2, Entry 6: Poly(silylether) of 1,6-Hexanediol

Scale: catalyst 5.0 mg, substrate 0.89 mmol. Polymer weight 222.5 mg, yield 84%. ¹H NMR (500 MHz, CDCl₃, 298 K, δ): 1.32 (m, 4H, OCH₂CH₂CH₂), 1.56 (m, 4H, OCH₂CH₂CH₂), 3.74 (m, 4H, OCH₂CH₂CH₂), 7.33 (m, 4H, m-Ph), 7.38 (m, 2H, p-Ph), 7.63 (m, 4H, o-Ph). ¹³C {¹H} NMR (125 MHz, CDCl₃, 298 K, δ): 25.7 (OCH₂CH₂CH₂), 32.6 (OCH₂CH₂CH₂), 63.3 (OCH₂CH₂CH₂), 127.9 (m-Ph), 130.3 (p-Ph), 133.4 (i-Ph), 135.1 (o-Ph). ²⁹Si {¹H} NMR (99 MHz, CDCl₃, 298 K, δ): −32.6.

Table 3, Entry 1: Poly(silylether) of Terephthalaldehyde

Scale: catalyst 5.0 mg, substrate 0.89 mmol. Polymer weight 178.0 mg, yield 58%. ¹H NMR (500 MHz, CDCl₃, 298 K, δ): 4.81 (m, 4H, OCH₂), 7.26 (m, 4H, C₆H₄), 7.37 (m, 4H, m-Ph), 7.71 (m, 6H, p- & o-Ph). ¹³C {¹H} NMR (125 MHz, CDCl₃, 298 K, δ): 65.0 (OCH₂). 126.7 (p- & o-C₆H₄), 128.1 (m-Ph), 130.6 (p-Ph), 132.5 (i-Ph), 135.1 (o-Ph), 139.4 (i-C₆H₄). ²⁹Si {¹H} NMR (99 MHz, CDCl₃, 298 K, δ): −30.3. T_g: 20 °C.

Table 3, Entry 2: Poly(silylether) of 1,6-Hexanedial

Scale: catalyst 5.0 mg, substrate 0.89 mmol. Polymer weight 105.6 mg. yield 40%. ¹H NMR (500 MHz, CDCl₃, 298 K, δ): 1.26–1.62 (m, 8H, OCH₂CH₂CH₂), 3.82 (m, 4H, OCH₂CH₂CH₂), 7.30–7.40 (m, 10H, Ph). ¹³C {¹H} NMR (125 MHz, CDCl₃, 298 K, δ): 26.82 (OCH₂CH₂CH₂), 33.21 (OCH₂CH₂CH₂), 64.15 (OCH₂CH₂CH₂), 127.8 (m-Ph), 131.1(p-Ph), 133.8 (i-Ph), 135.2 (o-Ph).

Table 3, Entry 3: Poly(silylether) of 1,4-Cyclohexanedione

Scale catalyst 5.0 mg, substrate 0.89 mmol. Polymer weight 130.1 mg, yield 49%. ¹H NMR (500 MHz, CDCl₃, 298 K, δ): 1.38 (m, 4H, (CH₂) C₆H₁₀) 1.81 (m, 4H, (CH₂) C₆H₁₀), 3.87 (m, 2H, (OCH) C₆H₁₀), 7.27 (m, 6H, m- & p-Ph), 7.58 (m, 4H, o-Ph). ¹³C {¹H} NMR (125 MHz, CDCl₃, 298 K, δ): 30.9, 32.0 ((CH₂) C₆H₁₀), 68.8, 70.0 ((OCH) C₆H₁₀), 127.7 (m-Ph), 130.1 (p-Ph), 134.1 (i-Ph), 135.0 (o-Ph). ²⁹Si {¹H} NMR (99 MHz, CDCl₃, 298 K, δ): −35.6, −35.7. T_g: 51 °C.

Table 3, Entry 4: Poly(silylether) of p-Hydroxy Benzaldehyde

Scale: catalyst 5.0 mg, substrate 0.89 mmol. Polymer weight 188.0 mg, yield 69%. ¹H NMR (500 MHz, CDCl₃, 298 K, δ): 4.66–481 (m, 2H, OCH₂), 6.88 (m, 2H, m-C₆H₄), 7.08 (m, 2H, o-C₆H₄), 7.32 (m, 6H, m- & p-Ph), 7.68 (m, 4H, o-Ph). ¹³C{¹H} NMR (125 MHz, CDCl₃, 298 K, δ): 64.7, 65.1 (OCH₂), 119.6, 119.7 (Ph), 128.2, 130.5, 130.7, 131.0, 132.0, 133.7, 134.5, 135.1 (m-, p-, i-, & o-Ph), 153.5 (i-C₆H₄). ²⁹Si {¹H} NMR (99 MHz, CDCl₃, 298 K, δ): −30.1, −33.5, −37.1. T_g: 18 °C.

Table 3, Entry 5: Poly(silylether) of p-Hydroxy Benzylalcohol

Scale: catalyst 5.0 mg, substrate 0.89 mmol. Polymer weight 198.5 mg, yield 73%. ¹H NMR (500 MHz, CDCl₃, 298 K, δ): 4.83–4.60 (m, 2H, OCH₂), 6.97 (m, 2H, C₆H₄), 7.45–7.35 (m, 2H, C₆H₄ and m, 6H, m- & p-Ph), 7.72 (m, 4H, o-Ph). ¹³C{¹H} NMR (125 MHz, CDCl₃, 298 K, δ): 64.7, 65.2 (OCH₂), 119.6, 119.9 (C₆H₄), 128.2, 128.40, 130.6, 130.8, 133.3, 134.6, 134.9, 135.3 (m-, p-, i-, & o-Ph), 153.5, 154.2 (i-C₆H₄). ²⁹Si {¹H} NMR (99 MHz, CDCl₃, 298 K, δ): −30.5, −33.5, −37.7. T_g: 25 °C.

Table 3, Entry 6: Poly(silylether) of 3-(4-Hydroxyphenyl)-1-propanol

Scale: catalyst 5.0 mg, substrate 0.89 mmol. Polymer weight 238.0 mg, yield 80%. ¹H NMR (500 MHz, CDCl₃, 298 K, δ): 1.73 (m, 2H, OCH₂CH₂CH₂C₆H₄), 2.51 (m, 2H, OCH₂CH₂CH2C₆H₄), 3.75 (m, 2H, OCH₂CH₂ CH₂C₆H₄), 6.79 (m, 4H, C₆H₄), 7.30 (m, 6H, m- & p-Ph), 7.65 (m, 2H, o-Ph). ¹³C{¹H} NMR (125 MHz, CDCl₃, 298 K, δ): 31.3, 31.3 (OCH₂CH₂CH₂C₆H₄), 34.1, 34.2 (OCH₂CH₂CH₂C₆H₄), 62.5, 63.0 (OCH₂CH₂CH₂C₆H₄), 119.5, 119.6 (C₆H₄), 128.0, 128.1, 129.5, 130.4, 130.7, 131.7, 132.5, 133.2, 135.1 (m-, p-, i-, & o-Ph), 152.2, 152.6 (i-C₆H₄). ²⁹Si{¹H} NMR (99 MHz, CDCl₃, 298 K, δ): −31.9, −34.5, −37.8. T_g: 21 °C.

Supporting Information Available

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsomega.6b00538.

• Experimental details including characterization data and selected spectra (PDF)

The authors declare no competing financial interest.