Tailor-Made Synthesis of Hydrosilanols, Hydrosiloxanes, and Silanediols Catalyzed by di-Silyl Rhodium(III) and Iridium(III) Complexes

Abstract

Siloxanes and silanols containing Si–H units are important building blocks for the synthesis of functionalized siloxane materials, and their synthesis is a current challenge. Herein, we report the selective synthesis of hydrosilanols, hydrosiloxanes, and silanodiols depending on the nature of the catalysts and the silane used. Two neutral ({MCl[SiMe₂(o-C₆H₄PPh₂)]₂}; M = Rh, Ir) and two cationic ({M[SiMe₂(o-C₆H₄PPh₂)]₂(NCMe)}[BAr^F₄]; M = Rh, Ir) have been synthesized and their catalytic behavior toward hydrolysis of secondary silanes has been described. Using the iridium complexes as precatalysts and diphenylsilane as a substrate, the product obtained is diphenylsilanediol. When rhodium complexes are used as precatalysts, it is possible to selectively obtain silanediol, hydrosilanol, and hydrosiloxane depending on the catalysts (neutral or cationic) and the silane substituents.

Short abstract

Bidentate silicon-based ligands have been used to synthesize two neutral ({MCl[SiMe₂(o-C₆H₄PPh₂)]₂}; M = Rh, Ir) and two cationic ({M[SiMe₂(o-C₆H₄PPh₂)]₂(NCMe)}[BAr_^F4]; M = Rh, Ir) complexes. The four new complexes proved to be catalytically active in the hydrolysis of dihydrosilanes, rendering different products, silanediols, hydrosilanols, or hydrosiloxanes, depending on the catalyst/substrate combination.

Introduction

Silicones (oligo- and polysiloxane materials) are an important class of inorganic materials that find many industrial applications due to their outstanding physicochemical properties (thermal and light stability and resistance to water and oxidation).^1,2 There are well-known methods for the synthesis of siloxane materials. The most used to date are the ring-opening polymerization of cyclic siloxanes and sol–gel processes (hydrolytic condensation of chlorosilanes (Cl–Si) or alkoxysilanes (RO–Si)).³

Nowadays, one of the main challenges in the synthesis of siloxane materials is to find methods to construct highly functionalized siloxanes with a well-defined structure. This could be achieved by using siloxanes, alkoxysilanes, and silanols, containing Si–H moieties as building blocks. Si–H bond is one of the most useful functional groups in silicon chemistry, and there are a large number of well-known reactions to transform this functional group (e.g., via hydrosilylation reactions).^4,5 Moreover, siloxanes with Si–H moieties are also interesting molecules to functionalize organic and inorganic compounds.⁶ Silanols, besides being used in the production of silicon-based polymers,⁷ are important intermediates in organic synthesis (e.g., an efficient organic donor in metal-catalyzed cross-coupling).⁸

In 2010, Kuroda et al. reported a pioneering study on the selective synthesis of hydrosiloxanes through a cross-coupling type reaction catalyzed by BiCl₃. Unfortunately, the product could not be effectively separated from the catalyst.⁹ Recently, more effective synthetic catalytic methods for the formation of hydrosiloxanes have been reported by reacting dihydrosilanes with silanols catalyzed by gold,¹⁰ cobalt,¹¹ Fe,¹² or main-group species.¹³ Alternatively, they can be obtained by dealcoholization reaction between silanols and alkoxyhydrosilanes without the use of catalyst or additives.¹⁴

In the case of the synthesis of silanols,¹⁵ there are well-established methods based on the hydrolysis of chlorosilanes,¹⁶ oxidation of hydrosilanes using strong oxidants,¹⁷ H₂O₂¹⁸ or O₂¹⁹ and the metal catalytic hydrolytic oxidation of hydrosilanes with water (also known as hydrolysis of silanes).²⁰ Reusable heterogeneous catalysts have also been used in the hydrolysis of silanes.²¹ The importance of the hydrolysis of silanes lies in the atom efficiency of this process. Recent studies have shown that silanols can also be obtained by electrochemical hydrolysis²² and enzymatic oxidation²³ of hydrosilanes.

Focusing on the synthesis of hydrosilanols through metal-catalyzed oxidation of dihydrosilanes using H₂O, few selective methods have been reported to date.²⁴ The problem is the presence of two reactive Si–H and the competitive condensation reactions.^{20f,20g,20m,20n} Thus, the hydrolysis of dihydrosilanes might lead to the formation of several products, as shown in Scheme 1, or mixtures thereof.

Scheme 1. Hydrolysis of Dihydrosilanes and Possible Reaction Products.

Herein, we report the catalytic activity of two neutral ({MCl[SiMe₂(o-C₆H₄PPh₂)]₂}; M = Rh, Ir) and two cationic ({M[SiMe₂(o-C₆H₄PPh₂)]₂(NCMe)}[BAr^F₄]; M = Rh, Ir) complexes in the hydrolysis of dihydrosilanes. The difference in the structure of each complex, together with the nature of the dihydrosilane used, allowed us to selectively synthesize different types of related Si–O products (Scheme 1).

Results and Discussion

Synthesis and Characterization of Neutral and Cationic P,Si-Complexes

Reaction of [RhCl(coe)₂]₂ with 4 equiv of the proligand [Si(H)Me₂(o-C₆H₄PPh₂)] in CH₂Cl₂ at room temperature afforded, after 1 h, the air-stable compound {RhCl[SiMe₂(o-C₆H₄PPh₂)]₂} (1 in Scheme 2) in good yield (92%), which was characterized in solution by NMR spectroscopy. In the ¹H NMR spectrum, two singlets are observed for the Si–CH₃ groups (δ −0.07 and δ −0.43) due to the nonequivalence of the methyl groups on each silicon atom upon coordination to rhodium, as previously observed in a related compound.²⁵ The ³¹P{¹H} NMR spectrum shows a unique signal as a doublet at δ 55.6 (J_Rh-P = 120 Hz), which indicates that both ligands are equivalent in solution. An isoelectronic iridium complex, {IrCl[SiMe₂(o-C₆H₄PPh₂)]₂} (2 in Scheme 2), was synthesized by reaction of [IrCl(coe)₂]₂ with 4 equiv of [Si(H)Me₂(o-C₆H₄PPh₂)] under the same reaction conditions (78% yield). The ¹H NMR spectroscopic pattern of complex 2 in solution is similar to that observed for complex 1 (see Experimental Section and Supporting Information, SI for more details). The ³¹P{¹H} NMR shows a singlet at δ 54.4 that indicates equivalent triarylphosphine groups.

Scheme 2. Synthesis of Complexes 1 and 2.

Compounds 1 and 2 were also characterized by single-crystal X-ray structural determination (Figure 1). The resulting molecular structures are in good agreement with the structures deduced from the spectroscopic data in solution. In the solid state, the geometry about the Rh(III) center in complex 1 can be described as a distorted trigonal bipyramid with the apical positions occupied by the phosphorous atoms (P1 and P1_i) and the equatorial positions occupied by the silicon (Si1 and Si1_i) and the chlorine atoms. The rhodium atom is included in the plane Si1Cl1Si1_i. The sum of the three angles (Cl1–Rh1–Si1, Si1–Rh1–Si1_i, and Si1_i–Rh1–Cl1) being 360° and the P1–Rh1–P1_i angle of 177.43(3)° suggest this geometry. Molecular structure of complex 2, in the solid state, is isostructural to that described for compound 1 (Figure 1b).

Figure 1.

Displacement ellipsoids are drawn at a 50% probability level. The hydrogen atoms and solvent molecules are omitted for clarity. (a) Molecular structure of 1. Selected bond lengths (Å) and angles (°): Rh1–Cl1 2.4357(8), Rh1–Si1 2.3099(6), Rh1–P1 2.3234(6), Si1–Rh1–Si1_i 83.37(3), P1–Rh1–P1_i 177.43(3), P1–Rh1–Si1 84.11(2), P1–Rh1–Cl1 91.28(2), and Si1–Rh1–Cl1 138.32(2). (b) Molecular structure of 2. Selected bond lengths (Å) and angles (°): Ir1–Cl1 2.399(4), Ir1–P1 2.318(4), Ir1–P2 2.318(4), Ir1–Si1 2.319(5), Ir1–Si2 2.325(5), P1–Ir1–P2 178.25(15), Si1–Ir1–Si2 83.43(18), Si1–Ir1–P1 83.89(15), Si1–Ir1–P2 95.88(15), Si2–Ir1–P2 84.53(15), Si2–Ir1–P1 94.42(15), Si1–Ir1–Cl1 134.87(18), Si2–Ir1–Cl1 141.70(19), and P1–Ir1–Cl1 90.71(14), P2–Ir1–Cl1 90.17(14).

Treatment of complexes 1 and 2 with an equimolar amount of NaBAr^F₄ in CH₂Cl₂ and in the presence of a small amount of MeCN yielded the Rh(III) cationic complex {Rh[SiMe₂(o-C₆H₄PPh₂)]₂(NCMe)}[BAr^F₄] (1[BAr^F₄]) and a similar cationic complex with two acetonitrile molecules coordinated to the metal center, {Ir[SiMe₂(o-C₆H₄PPh₂)]₂(NCMe)₂}[BAr₄^F] (2[BAr^F₄]), respectively (Scheme 3).

Scheme 3. Synthesis of Complexes 1[BAr^F₄] and 2[BAr^F₄].

Compounds 1[BAr^F₄] and 2[BAr^F₄] were characterized in solution by NMR spectroscopy (see Experimental Section and SI for more details). Both complexes showed in the ¹H NMR spectrum two different signals for the Si–CH₃ groups (δ 0.04 and δ −0.32 for 1[BAr^F₄]; δ −0.10 and δ −0.26 for 2[BAr^F₄]). The signals assigned to coordinated MeCN were observed at 1.78 ppm (integrating by 3H) in the case of 1[BAr^F₄] and at 1.48 (integrating by 6H) in the case of 2[BAr^F₄]. The equivalence of both phosphines in solution is confirmed by the presence of only one signal in the ³¹P{¹H} NMR spectra. They appear as a doublet at 55.5 ppm {J_Rh-P = 120 Hz} ppm for 1[BAr^F₄] and a singlet at 38.1 ppm for 2[BAr^F₄]. The spectroscopic data were consistent with a pentacoordinated Rh(III) complex in 1[BAr^F₄] and a pseudo-octahedral Ir(III) structure in the case of 2[BAr^F₄], as proposed in Scheme 3.

These structures were confirmed by X-ray diffraction (Figure 2). Figure 2a shows the solid-state structure of 1[BAr^F₄]. The formally Rh(III) center adopts a square pyramid geometry with one of the silicon atoms (Si1) located on the apical position. The plane defined by the other silicon atom (Si2), the two phosphorous atoms, and the acetonitrile ligand (Si2P1P2N1), where is included the rhodium atom, forms the base of the pyramid. The sum of the four angles (Si2–Rh1–P1, P1–Rh1–N1, N1–Rh1–P2, and P2–Rh1–Si2) being 360° suggests this geometry. It is important to note that in solution, both SiP chelate ligands are equivalent. This could be due to a rapid interchange of MeCN located trans to Si2 to be placed trans to Si1.

Figure 2.

Displacement ellipsoids are drawn at a 50% probability level. The hydrogen atoms and the counteranion [BAr₄^F] are omitted for clarity. (a) Molecular structure of 1[BAr^F₄]. Selected bond lengths (Å) and angles (°): Rh1–N1 2.163(3), Rh1–Si1 2.341(1), Rh1–Si2 2.311(1), Rh1–P1 2.317(1), Rh1–P2 2.317(1), Si1–Rh1–Si2 86.44(4), Si1–Rh1–N1 107.45(9), Si2–Rh1–N1 165.76(9). (b) Molecular structure of 2[BAr^F₄]. Selected bond lengths (Å) and angles (°): Ir1–N1 2.279(4), Ir1–N2 2.148(3), Ir1–Si1 2.356(1), Ir1–Si2 2.366(1), Ir1–P1 2.331(1), Ir1–P2 2.326(1), Si1–Ir1–Si2 89.57(4), Si1–Ir1–N1 90.99(11), Si2–Ir1–N2 92.93(9), N1–Ir1–N2 86.52(14).

The solid-state structure of 2[BAr₄^F] is showed in Figure 2b. The resulting molecular structure agrees with the structure deduced from the spectroscopic data in solution. The coordination geometry of the Ir(III) atom is pseudo-octahedral. One of the chelate SiP ligands and one of the acetonitrile (Si1, P1, and N1) are located on one of the faces of the octahedron. The three remaining coordination sites are occupied by the other SiP ligand and MeCN (Si2, P2, and N2). The trans-labilizing nature of the silyl group is reflected in long Ir—N bond lengths [N1—Ir1 = 2.536(2) Å and N2—Ir1 = 2.536(2) Å], when compared with other iridium-acetonitrile distances in similar structures.²⁶

Catalytic Hydrolysis of Diphenylsilane

Compounds, 1, 2, 1[BAr^F₄], and 2[BAr^F₄] were tested as precatalysts for the hydrolysis of Ph₂SiH₂ in tetrahydrofuran (THF) under standard reaction conditions (0.2 mol % catalyst, [Ph₂SiH₂] = 0.22 M, 10 equiv H₂O; see SI for more information). The H₂ evolution was monitored through the Man on the Moon X102 kit. The reaction profiles obtained (equiv of H₂ vs time) are shown in Figure 3.

Figure 3.

Reaction profiles (equiv of H₂ generated vs time) for hydrolysis of diphenylsilane using 1 and a 1[BAr^F₄] (left), 2 and 2[BAr^F₄] (right) as precatalysts. Reaction conditions: Silane (0.22 mmol), H₂O (2.2 mmol), 0.2 mol % of catalyst in 1 mL of THF at 25 °C under N₂. The hydrogen production was calculated by continuous monitoring of the pressure evolution using a pressure transducer (Man on the Moon X102 kit).

It is worth highlighting the different reaction profiles obtained when comparing the cationic with the neutral compounds, which points to the integrity of the M–Cl bond through the catalytic process.

When Ir(III)-based catalysts 2 and 2[BAr^F₄] were employed, more than 1.6 equiv of H₂ were liberated in the process, being the product of the double hydrolysis, diphenylsilanediol (3c in Table 1, entries 2 and 4 and Figures S3 and S4 in SI), the only silane-containing product detected by ¹H NMR at the end of the reaction. Moreover, the absence of Ph₂SiH₂ indicates that the reaction is complete. In this process, the cationic 2[BAr^F₄] is more active than the neutral compound 2. The reaction profiles observed, based on hydrogen evolution (Figure 3), did not show an evident two-stepped process, pointing to similar reaction rates for the first and second hydrolysis of the dihydrosilane.

Table 1. Catalytic Hydrolysis of Diphenylsilane^a.

entry	cat.	H2b equiv	time (s)	3ac	3bc	3cc
1	1	1	2500	93	7
2	2	1.6	155,500			>99
3	1[BArF4]	1	50	1	99
4	2[BArF4]	1.75	38,460			>99

^aReaction conditions: Silane (0.22 mmol), H₂O (2.2 mmol), 0.2 mol % of catalyst in 1 mL of THF at 25 °C.

^bHydrogen equivalents evolved.

^cMolar ratio of products calculated by ¹H NMR.

To confirm the sequential formation of 3c, the hydrolysis of diphenylsilane catalyzed by 2[BAr^F₄] was monitored by in situ ¹H NMR spectroscopy (Figure S5 in SI). The time-correlated speciation diagram obtained (Figure 4) shows the sigmoidal formation of diphenylsilanediol (3c), typical of consecutive reactions, being diphenylhydrosilanol (3a) the reaction intermediate. It is worth mentioning that the resemblance of the first and second hydrolysis rate constants hampers the isolation of diphenylhydrosilanol 3a with Ir(III)-based catalyst 2 and 2[BAr₄^F].

Figure 4.

Time-correlated speciation diagram. Hydrolysis of Ph₂SiH₂ catalyzed by 2[BAr^F₄]. Reaction conditions: Ph₂SiH₂ (0.11 mmol), H₂O (1.1 mmol), 2[BAr^F₄] 0.2 mol %, 0.5 mL THF-d₈, 25 °C.

In contrast, when the reaction was catalyzed by rhodium(III)-based complexes 1 or 1[BAr^F₄], the reaction profile reached a plateau after liberating only 1 equiv of H₂ (Figure 3). The cationic complex 1[BAr^F₄] was the most efficient precatalyst, and the process was completed in only 50 seconds (Figure 3 and entry 3 in Table 1). When the neutral complex 1 was used as a precatalyst, 2500 s were required to reach the same conversion (Figure 3 and entry 1 in Table 1). The difference in reaction rate observed depending on the catalyst used (neutral or cationic) may be due to two reasons: (a) the easier accessibility of the substrate to the metal center in the case of the cationic complex; (b) the stronger electrophilicity of the cationic complexes with respect to the neutral ones.

Surprisingly, ¹H NMR inspection of the reaction mixtures at the end of the reaction showed that the product obtained was mainly diphenylhydrosilanol (3a) when 1 was used as a catalyst (Table 1, entry 1 and Figure S1 in SI), but it proceeded toward tetraphenyldihydrosiloxane (3b) with the cationic catalyst 1[BAr^F₄] (Table 1, entry 3 and Figure S2 in SI). Two reaction pathways can be envisaged for the formation of dihydrosiloxane 3b under the reaction conditions: the condensation of two molecules of diphenylhydrosilanol (3a) (liberating one molecule of H₂O) or the nucleophilic attack of a diphenylhydrosilanol (3a) to a diphenyldihydrosilane (evolving 1 equiv of H₂) (Scheme 4).

Scheme 4. Possible Reaction Pathways for the Formation of Diphenyldihydrosiloxane Catalyzed by 1[BAr^F₄].

To shed some light on the origin of the observed dihydrosiloxane when using 1[BAr^F₄], two control experiments were performed by means of ¹H NMR. First, diphenylhydrosilanol 3a (0.11 mmol) was treated with 0.2 mol % of 1[BAr^F₄] in THF-d₈ (0.5 mL) at 25 °C. The ¹H NMR spectrum acquired after 5 min of reaction shows nearly complete consumption of 3a and the clear formation of the dihydrosiloxane 3b (Scheme 5a, Figure S6a in SI). When an equimolecular mixture of diphenylhydrosilanol 3a (0.11 mmol) and Ph₂SiH₂ (0.11 mmol) was reacted under identical conditions, the ¹H NMR spectra acquired after 5 min showed, as in the previous experiment, the formation of the dihydrosiloxane 3b and the total consumption of 3a. Noticeably, Ph₂SiH₂ was only partially (≈50%) consumed, which supports the hypothesis that 3b is formed by a fast condensation of the formed diphenylhydrosilanol 3a. The partial consumption of Ph₂SiH₂ must be attributed to its 1[BAr^F₄]-catalyzed hydrolysis due to the H₂O formed during the condensation reaction. Accordingly, a small signal assigned to the H₂ generated in this process was observed in the ¹H NMR spectra (Scheme 5b and Figure S6b in SI).

Scheme 5. Control Experiments for Diphenyldihydrosiloxane Formation.

The results obtained show that whereas both neutral and cationic rhodium complexes 1 and 1[BAr^F₄] are very effective catalysts for the first hydrolysis of Ph₂SiH₂, in the case of 1[BAr^F₄], the product of the reaction 3a is immediately involved in a metal-mediated condensation process. Noticeably, this condensation reaction is not observed in the case of 1 at short reaction times. These results suggest the need of two coordination vacancies in the metal center, or a highly electrophilic metal center, for the condensation to proceed at competitive rates.

A global analysis of the results obtained with the four catalytic systems studied reveals the network of coexisting reactions in the hydrolysis of diphenylsilane (Scheme 6).²⁷ The different reaction rates for each individual step observed when compared to the different catalysts studied allowed us to selectively obtain the product of monohydrolysis (3a), its condensation product (3b), or the doubly hydrolyzed silane (3c). No other silane-containing products were detected by ¹H NMR in the reaction mixtures when iridium-based catalysts were used in the process, which discards the formation of polysiloxanes with these systems.

Scheme 6. Hydrolysis of Diphenylsilane: Network of Potential Coexisting Reactions.

To confirm whether compound 1 at extended reaction times would catalyze the second hydrolysis or the condensation reaction, a catalytic reaction was conducted with this compound for 24 h. The reaction profile showed that after the initial formation of 3a, a second equivalent of hydrogen evolved but at a much lower rate (Scheme 7 and Figure S7 in SI). ¹H NMR analysis of the final reaction mixture shows that the reaction product is not the diphenylsilanediol (3c), but a complex mixture of (O–SiPh₂–O) species (3e is proposed as one of these species, Figure S9 in SI). To obtain more information, the same reaction was monitored by in situ ¹H NMR spectroscopy (Figure S11 in SI). This experiment shows that after Ph₂SiH₂ is consumed, 3a begins to transform into dihydrosiloxane 3b, which subsequently transforms into another product containing at least one Si–H unit (probably 3d). At longer reaction times, these signals evolve towards (O–SiPh₂–O) species. Similar results were obtained when the hydrolysis of Ph₂SiH₂ was conducted for 24 h with catalysts 1[BAr^F₄] (Figures S8 and S10 in SI). Taken together, these results suggest that most probably, 1 follows a condensation route (via3b) rather than the double hydrolysis of the dihydrosilane (via3c) (k₃ > k₂).

Scheme 7. Hydrolysis of Diphenylsilane Catalyzed by 1 (a) and 1[BAr^F₄] (b) TOFs_1/2 Calculated in Red.

Hydrolysis of Other Secondary Silanes

The fine balance between rate constants observed when catalysts 1, 1[BAr^F₄], 2, and 2[BAr^F₄] were studied in the hydrolysis of Ph₂SiH₂ and the special relevance of hydrosilanols and hydrosiloxanes as valuable synthons for the chemical industry prompted us to extend the study of the rhodium-based catalysts to other commercial secondary silanes (entries 1–8 in Table 2).

Table 2. Catalytic Hydrolysis of Dihydrosilanes^a.

entry	cat.	R1, R2	H2b equiv	TOF1/2 (h–1)	Ac	Bc	Cc
1	1	Ph, Ph	1	1889	93	7
2	1	Ph, Naph	1	78	90	10
3	1	Ph, Me	1	2991	10	90
4	1	Et, Et	0.93	6452	unidentified
5	1[BArF4]	Ph, Ph	1	149,533	1	99
6	1[BArF4]	Ph, Naph	1	13,962	97	3
7	1[BArF4]	Ph, Me	1	472,861	2	98
8	1[BArF4]	Et, Et	1.8	390,087d			99

^aReaction conditions: Silane (0.22 mmol), H₂O (2.2 mmol), 0.2 mol % of catalyst in 1 mL of THF at 25 °C under N₂.

^bHydrogen equivalents released.

^cMolar ratio of products calculated by ¹H NMR.

^dTOF_1/2 has been calculated for the release of 1 equiv of hydrogen.

The reaction of 1-naphtyl(phenyl)dihydrosilane (1-Naph(Ph)SiH₂) with H₂O in the presence of catalytic amounts of the rhodium complexes 1 or 1[BAr^F₄] led to the formation of a mixture of hydrosilanol 3a′ and dihydrosiloxane 3b′ (entries 2 and 6 in Table 2), being the hydrosilanol 3a′ the main product (selectivity > 90%) in both cases. Accordingly, in both reactions, only 1 equiv of H₂ was released. As with Ph₂SiH₂, the reaction catalyzed by complex 1 was slower than the reaction catalyzed by 1[BAr^F₄].

The results obtained in the hydrolysis of methyl(phenyl)dihydrosilane (entries 3 and 7 of Table 2) show a change in the reactivity, obtaining the respective dihydrosiloxane 3b″ as a main product indistinctly when 1 or 1[BAr^F₄] were used as precatalysts (selectivity > 90%). As with Ph₂SiH₂ and 1-Naph(Ph)SiH₂, these reactions only released 1 equiv of H₂ and the hydrolysis performed using the cationic rhodium complex is faster.

Finally, diethyldihydrosilane (Et₂SiH₂) was tested (entries 4 and 8 in Table 2). The hydrolysis of Et₂SiH₂ using 1 as a precatalyst releases 1 equiv of H₂, but the products formed could not be identified. Surprisingly, using this substrate and 1[BAr^F₄] as a precatalyst, almost 2 equiv of H₂ were released, albeit at clearly different rates. The first equivalent was liberated in less than 1 min, and 25 min sufficed to reach completion. The product observed at the end of the reaction was diethylsilanediol (3c‴), presumably formed by hydrolysis of diethylhydrosilanol (3a‴).

The results obtained with the different silanes are compatible with the reaction scheme shown above (Scheme 6). Once the hydrosilanol is formed, its fate will depend on the relative rates of second hydrolysis (k₂) and condensation (k₃) reactions. If both are slow, the product could be isolated (as is the case with Ph₂SiH₂ and to a certain extent with 1-Naph(Ph)SiH₂). If the second hydrolysis is fast, the product will evolve to the corresponding silanediol (C in Table 2). The relative rates for the hydrolysis can be inferred from the calculated turnover frequency (TOFs) based on the liberation of 1 equiv of H₂. According to these values, this reaction is approximately 100 times faster using the cationic rhodium complex 1[BAr^F₄] than when neutral 1 was used, with all the substrates. As discussed above, this enhanced reactivity toward the hydrolysis could be explained by the stronger electrophilicity and/or the better accessibility of the substrate to the catalytic pocket when the cationic catalyst is used. The relative reaction rates observed for the different substrates (Et₂SiH₂ > MePhSiH₂ > Ph₂SiH₂ > 1-Naph(Ph)SiH₂) support the important influence of electronic factors in the activation barrier for this process. This electronic effect could be extrapolated to the hydrolysis of the formed hydrosilanol. Consistently, only in the case of Et₂SiH₂, the diethylsilanediol (3c‴) was obtained. It is worth reminding that highly electrophilic Ir(III) catalyst 2 and 2[BAr^F₄] were required to obtain the corresponding silanediol (3c) from Ph₂SiH₂.

Alternatively, if the second hydrolysis is slow compared to the condensation reaction, the hydrosilanol could be isolated or evolve toward the dihydrosiloxane if involved in a fast metal-catalyzed condensation. According to the results obtained, steric effects determine the rate of the condensation process. When using the largest silane, 1-Naph(Ph)SiH₂, the major product obtained is unreacted hydrosilanol, independently of the catalyst used. In contrast, the product obtained when using Me(Ph)SiH₂ is the dihydrosiloxane.

In view of the excellent catalytic properties of our system, we decided to study the catalytic behavior of complex 1[BAr^F₄] in the alcoholysis of Ph₂SiH₂ and the hydrolysis of other silanes (see Section S9 in SI). Satisfyingly, the alcoholysis of Ph₂SiH₂ with MeOH, EtOH, and iPrOH rendered the corresponding alkoxyhydrosilanes. When primary silanes were subjected to hydrolysis, the corresponding siloxanes were obtained as the main product, whereas tertiary silanes rendered an unidentified mixture of products liberating 2 equiv of H₂.

Scale-Up Experiments

It should be noted that, to the best of our knowledge, few efficient methods for the selective synthesis of hydrosilanols have been reported to date.²⁴

To prove that using precataysts 1 and 1[BAr^F₄], it is possible to obtain selectively the diphenylhydrosilanol (3a) and diphenyldihydrosiloxane (3b) in a gram-scale, the reaction of 5.4 mmol (1 g) of Ph₂SiH₂ with 10 equiv of H₂O under standard catalytic conditions (0.2 mol % of 1 or 1[BAr^F₄] as catalyst, THF as solvent) was performed. When precatalyst 1 was used, this reaction led to the formation of diphenylhydrosilanol (3a) in a 56% of isolated yield (600 mg). Using 1[BAr^F₄] as a precatalyst, diphenyldihydrosiloxane 3b was obtained in 81% of isolated yield (832 mg) (Scheme 8).

Scheme 8. Gram-Scale Synthesis of Diphenylhydrosilanol and Diphenyldihydrosiloxane.

The robustness of the catalytic system derived from 1[BAr^F₄] was also confirmed through successive additions of Ph₂SiH₂. The sequential reaction profiles obtained showed that the catalyst maintained its activity for, at least, 10 successive cycles (Figure S12 in SI). In addition, in the case of the neutral compounds 1 and 2, their structures remain unchanged after the catalytic reaction (Figures S1 and S3 in SI), which also demonstrate the robustness of the catalytic system.

Conclusions

In summary, two neutral {MCl[SiMe₂(o-C₆H₄PPh₂)]₂} (M = Rh, 1; Ir, 2) and two cationic {M[SiMe₂(o-C₆H₄PPh₂)]₂(NCMe)_n}[BAr^F₄] (M = Rh and n = 1, 1[BAr^F₄]; M = Ir and n = 2, 2[BAr^F₄]) complexes were prepared and structurally characterized in solution by NMR and in the solid state by X-ray crystallography. The four new complexes proved to be catalytically active in the hydrolysis of dihydrosilanes, rendering different products depending on the catalyst/substrate combination. These air-stable complexes have shown excellent catalytic properties with low catalyst loadings at room temperature and without the requirement of additives. A rational analysis of the different products obtained allowed us to propose a network of coexisting metal-catalyzed reactions of different nature (hydrolysis of hydrosilanes, condensation of silanols, and nucleophilic attack of silanols on hydrosilanes). Apparently, the latter are not operative with our systems. The results showed that hydrolytic processes are mainly controlled by the electronic nature of both catalysts and substrates, whereas metal-catalyzed condensations are mostly affected by sterics. This subtle balance permitted us to obtain selectively hydrosilanols, silanediols, or dihydrosilanes by an educated selection of the catalyst and substrate. The practical application of this methodology has been demonstrated by gram-scale synthesis experiments using Ph₂SiH₂.

Experimental Section

General Considerations

The preparation of the metal complexes was carried out at room temperature under nitrogen by standard Schlenk techniques. Glassware was oven dried at 120 °C overnight and flamed under a vacuum prior to use. CH₂Cl₂ and THF were distilled over CaH₂ and Na, respectively, degassed by successive freeze–pump–thaw cycles and stored over molecular sieves (3 Å). [RhCl(coe)₂]²⁸ and [IrCl(coe)₂]²⁹ complexes and the proligand o-Ph₂P(C₆H₄)SiMe₂H³⁰ were prepared as previously reported. Microanalyses were carried out with a Leco CHNS-932 microanalyzer. NMR spectra were recorded with Bruker Avance DPX 300, Bruker Avance 400, or Bruker Avance 500 spectrometers at room temperature unless otherwise stated; ¹H and ¹³C{¹H} (residual solvents) and ³¹P{¹H} (H₃PO₄ external standard) spectra were measured from CDCl₃, CD₂Cl₂, THF-d₈ solutions. IR spectra were recorded with a Nicolet FTIR 510 spectrophotometer using KBr pellets.

Synthesis of {MCl[SiMe₂(o-C₆H₄PPh₂)]₂} (M = Rh, 1; M = Ir, 2)

[M(coe)₂Cl]₂ (M = Rh, Ir) (0.14 mmol) was solved in CH₂Cl₂ (4 mL), 4 equiv of o-Ph₂P(C₆H₄)SiMe₂H (180 mg, 0.560 mmol) were added, and it was stirred for 20 min. After this time, solvent was removed under the vacuum, and the resulting solid was washed with 5 mL of n-pentane and 5 mL of methanol and dried under vacuum to obtain a pale-yellow (1) and yellow (2) solids. Yield of 1 164 mg (76%). Yield of 2 201 mg (83%).

1

¹H NMR (400 MHz, CD₂Cl₂): δ 8.10–7.10 (Si(o-C₆H₄PPh₂), 28 H_arom), −0.07 (s, Si–CH₃, 6H), −0.43 (s, Si–CH₃, 6H). ³¹P{¹H} NMR (202 MHz CD₂Cl₂): δ 55.6 (d, J_Rh-P = 120 Hz, 2P). ¹³C{¹H} NMR (101 MHz, CD₂Cl₂): δ 160.0–128.0 (36 C_arom., Si(o-C₆H₄PPh₂)), 7.3 (s, 2C, Si–CH₃), 2.4 (s, 2C, Si–CH₃). Microanalysis (RhClSi₂P₂C₄₀H₄₀·CH₂Cl₂): Requires: C 57.12, H 4.91. Obtained: C 56.95, H 5.02.

2

¹H NMR (400 MHz, CD₂Cl₂): δ 8.02–7.25 (Si(o-C₆H₄PPh₂), 28 H_arom), −0.11 (s, Si–CH₃, 6H), −0.51 (s, Si–CH₃, 6H). ³¹P{¹H} NMR (202 MHz, CD₂Cl₂): δ 54.4 (s, 2P). ¹³C{¹H} NMR (101 MHz, CD₂Cl₂): δ 159.6–126.5 (36 C_arom, Si(o-C₆H₄PPh₂)), 5.5 (s, 2C, Si–CH₃), −0.3 (s, 2C, Si–CH₃). Microanalysis (IrClSi₂P₂C₄₀H₄₀): Requires: C 55.44, H 4.65. Obtained: C 55.55, H 4.58.

Synthesis of {Rh[SiMe₂(o-C₆H₄PPh₂)]₂(NCMe)}[BAr^F₄] (1[BAr^F₄]) and {Ir[SiMe₂(o-C₆H₄PPh₂)]₂(NCMe)₂}[BAr^F₄] (2[BAr^F₄])

To a Schlenk charged with the [M(SiMe₂(o-C₆H₄PPh₂))₂Cl] (M = Rh, Ir) complex (1, 100 mg, 0.13 mmol; 2, 112 mg, 0.13 mmol) and NaBAr^F₄ (127 mg, 0.14 mmol), 4 mL of CH₂Cl₂ was added. After 20 min stirring, the yellow suspension formed was filtered via cannula to remove NaCl. Addition of 1 mL of MeCN gave a pale-yellow solution, and it was left stirring 10 min. After this time, solvent was removed under vacuum to give a pale-yellow solid in both cases. Yield of 1[BAr^F₄] 184 mg (86%). Yield of 2[BAr^F₄] 194 mg (84%).

1[BAr^F₄]

¹H NMR (400 MHz, CDCl₃): δ 7.73 (m, 8H, BAr^F₄), 7.55 (m, 4H, BAr^F₄), 7.70–7.32 (Si(o-C₆H₄PPh₂), 28 H_arom), 1.64 (s, CH₃–CN, 3H), 0.04 (s, Si–CH₃, 6H), −0.32 (s, Si–CH₃, 6H). ³¹P{¹H} NMR (202 MHz, CDCl₃): δ 55.5 (d, J_Rh-P = 120 Hz, 2P). ¹³C{¹H} NMR (101 MHz, CDCl₃): δ 161.8 (q, J_B-C = 50 Hz, BAr^F₄), 134.9 (s, BAr^F₄), 129.0 (q, J_F-C = 12 Hz, BAr^F₄), 124.7 (q, J_F-C = 273 Hz, CF₃), 122.4 (s, 2C, NC–CH₃), 117.6 (m, BAr₄^F), 159.0–128.0 (36 C_arom, Si(o-C₆H₄PPh₂)), 7.5 (s, 2C, Si–CH₃), 2.6 (s, 2C, Si–CH₃), 1.8 (s, 1C, NC–CH₃). Microanalysis (RhSi₂P₂C₇₄H₅₅BNF₂₄): Required: C 54.00, H 3.37, N 0.85. Obtained: C 53.88, H 3.55, N 1.02.

2[BAr^F₄]

¹H NMR (400 MHz, CDCl₃): δ 7.71 (s, 8H, BAr^F₄), 7.53 (s, 4H, BAr^F₄), 7.75–7.70 (Si(o-C₆H₄PPh₂), 28 H_arom), 1.48 (s, CH₃–CN, 3H), −0.10 (s, Si–CH₃, 6H), −0.26 (s, Si–CH₃, 6H). ³¹P{¹H} NMR (202 MHz, CDCl₃): δ 38.1 (s, 2P). ¹³C{¹H} NMR (101 MHz, CDCl₃): δ 161.8 (q, J_B-C = 50 Hz, BAr^F₄), 134.9 (s, BAr^F₄), 129.0 (q, J_F-C = 12 Hz, BAr₄^F), 124.7 (q, J_F-C = 273 Hz, CF₃), 119.3 (s, 2C, NC–CH₃), 117.6 (m, BAr₄^F), 160.0–127.0 (36 C_arom, Si(o-C₆H₄PPh₂)), 4.4 (s, 2C, Si–CH₃), 1.8 (s, 2C, NC–CH₃), 0.9 (s, 2C, Si–CH₃). Microanalysis (IrSi₂P₂C₇₆H₅₈BN₂F₂₄): Requires: C 51.39, H 3.29 N 1.58. Obtained: C 51.02, H 3.39, N 1.29.

X-ray Crystallography

Crystals for 1, 1[BAr^F₄], 2, and 2[BAr^F₄] were obtained by diffusion of pentane over CH₂Cl₂ and were mounted on a glass fiber and used for data collection on a Bruker Apex II with a photon detector equipped with graphite monochromated Mo Kα radiation (λ = 0.71073 Å). Lorentz-polarization and empirical absorption corrections were applied. The structures were solved by direct methods and refined with full-matrix least-squares calculations on F² using the program SHELXT.³¹ Anisotropic temperature factors were assigned to all atoms except for hydrogen atoms, which are riding their parent atoms with an isotropic temperature factor arbitrarily chosen as 1.2 times that of the respective parent. Final R(F), wR(F2), and goodness-of-fit agreement factors, details on the data collection, and analysis can be found in Table S1.

General Procedure for Catalytic Hydrolysis of Silanes

A closed reaction vessel equipped with a pressure transducer (Manonthemoon kinetic kit X102)³² was immersed in a thermostated ethylene glycol/water bath and charged with the catalyst (0.00044 mmol) in 1 mL of distilled THF and H₂O (2.2 mmol). Once the pressure of the system was stabilized, the silane (0.22 mmol) was added, which was considered initial reaction time. The solution was left stirring until the pressure stabilized again, which was indicative that the reaction ended. The quantity of gas evolved was calculated from the measured pressure inside the reaction vessel following the ideal gases law equation (reactor volume 13.2 mL). Then, solvent was removed under vacuum, and the reaction mixture was analyzed by ¹H NMR to determine the molar ratio of products. The major reaction product was isolated through purification by column chromatography on silica gel using Hexane/EtOAc (10:1) as eluents.

Ph₂Si(H)OH, Diphenylsilanol (3a)

¹H NMR (400 MHz, CDCl₃): δ 7.68–7.63 (m, 4H_arom), 7.48–7.39 (m, 6H_arom), 5.53 (s, 1H, Si–H), 2.59 (s, 1H, Si–OH). ¹³C{¹H} NMR (126 MHz, CDCl₃): δ 135.3 (2C), 134.3 (4C), 130.7 (2C), 128.4 (4C) ppm. HMQC (¹H–²⁹Si) NMR (400 MHz, CDCl₃): δ (²⁹Si) −12.9. IR (KBr): 3209 cm^–1 (broad, Si–O–H), 2125 (Si–H) cm^–1.

Ph₂(H)SiOSi(H)Ph₂, Tetraphenyldisiloxane (3b)

¹H NMR (400 MHz, CDCl₃): δ 7.60–7.55 (m, 8H_arom), 7.45–7.40 (m, 4H_arom), 7.39–7.33 (m, 8H_arom), 5.62 (s, 2H, Si–H). ¹³C{¹H} NMR (126 MHz, CDCl₃): δ 135.3 (4C), 134.7 (8C), 130.6 (4C), 128.3 (8C) ppm. HMQC (¹H–²⁹Si) NMR (400 MHz, CDCl₃): δ (²⁹Si) −19.5 ppm. Ir (KBr): 2123 cm^–1 (Si–H), 1087 (Si–O–Si) cm^–1.

Ph₂Si(OH)₂, Diphenylsilanediol (3c)

¹H NMR (400 MHz, THF-d₈): δ 7.68–7.64 (m, 4H_arom), 7.32–7.23 (m, 6H_arom), 6.00 (s, 2H, Si–OH). ¹³C{¹H} NMR (126 MHz, THF-d₈): δ 138.8 (2C), 135.2 (4C), 129.9 (2C), 128.0 (4C). HMQC (¹H–²⁹Si) NMR (400 MHz, THF-d₈): δ (²⁹Si) −33.3 ppm. IR (KBr): 3197 (broad, Si–O–H) cm^–1.

1-NaphPhSi(H)OH, 1-Naphtyl(phenyl)silanol (3a′)

¹H NMR (400 MHz, CDCl₃): δ 8.17 (dm, J = 7.9 Hz, 1H_arom), 7.97 (dm, J = 8.3 Hz, 1H_arom), 7.89 (tm, J = 7.5 Hz, 2H_arom), 7.68 (dm, J = 7.9 Hz, 2H_arom), 7.54–7.42 (m, 4H_arom), 7.39 (dm, J = 7.5 Hz, 2H_arom), 5.90 (s, 1H, Si–H), 2.80 (s, 1H, Si–OH). ¹³C{¹H} NMR (126 MHz, CDCl₃): δ 137.1 (1C), 135.5 (1C), 135.4 (1C), 134.6 (2C), 133.6 (1C), 133.1 (1C), 131.5 (1C), 130.7 (1C), 129.2 (1C), 128.4 (2C), 128.1 (1C), 126.7 (1C), 126.1 (1C), 125.5 (1C). HMQC (¹H–²⁹Si) NMR (400 MHz, CDCl₃): δ (29Si) −13.2. IR (KBr): 3295 cm^–1 (broad, Si–O–H), 2136 (Si–H) cm^–1.

1-NaphPh(H)SiOSi(H)Ph(1-Naph), Di-1-naphtyl(diphenyl)disiloxane (3b′)

¹H NMR (400 MHz, CDCl₃): δ 8.06 (dd, J = 8.4 Hz, J = 4.0, 2H_arom), 7.95 (dd, J = 8.4 Hz, J = 4.0 Hz, 2H_arom), 7.87 (dd, J = 8.4 Hz, J = 3.2 Hz, 2H_arom), 7.84 (tm, J = 7.6 Hz, 2H_arom), 7.60 (tm, J = 7.6 Hz, 4H_arom), 7.49–7.40 (m, 6H_arom), 7.36–7.26 (m, 6H_arom), 6.00 (s, 2H, Si–H). ¹³C{¹H} NMR (126 MHz, CDCl₃): δ 137.0 (2C), 135.6 (2C), 135.5 (2C), 134.5 (4C), 133.5 (2C), 133.0 (2C), 131.4 (2C), 130.5 (2C), 129.0 (2C), 128.3 (6C), 126.4 (2C), 126.0 (2C), 125.4 (2C). HMQC (¹H–²⁹Si) NMR (400 MHz, CDCl₃): δ (²⁹Si) −18.3 ppm. IR (KBr): 2129 cm^–1 (Si–H), 1056 (Si–O–Si) cm^–1.

MePh(H)SiOSi(H)PhMe, Dimethyl(diphenyl)disiloxane (3b″)

¹H NMR (400 MHz, CDCl₃): δ 7.61–7.57 (m, 4H_arom), 7.44–7.37 (m, 6H_arom), 5.18 (m, 2H, Si–H), 0.47 (d, J = 2.82 Hz, 6H, Si–CH₃). ¹³C{¹H} NMR (126 MHz, CDCl₃): δ 137.4 (2C), 133.7 (4C), 130.2 (2C), 128.2 (4C), −0.3 (s, 2C, Si–CH₃). HMQC (¹H–²⁹Si) NMR (400 MHz, CDCl₃): δ (²⁹Si) −11.6. IR (KBr): 2132 cm^–1 (Si–H), 1062 cm^–1 (Si–O–Si).

Et₂Si(OH)₂, Diethylsilanediol (3c‴)

¹H NMR (400 MHz, THF-d₈): δ 4.96 (s, 2H, Si–OH), 0.93 (t, J = 7.9 Hz, 6H, CH₃) 0.45 (q, J = 7.9 Hz, 4H, CH₂). ¹³C{¹H} NMR (126 MHz, THF-d₈): δ 7.5 (2C), 7.1 (2C). HMQC (¹H–²⁹Si) NMR (400 MHz, THF-d₈): δ (²⁹Si) −7.0. IR (KBr): 3162 cm^–1 (broad, Si–O–H).

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.inorgchem.2c03953.

• Additional experimental details, NMR and FTRI spectra, and X-ray crystallographic tables (PDF)

The authors declare no competing financial interest.

The authors declare no competing financial interest.