Platinum Complexes with Chelating Acyclic Aminocarbene Ligands Work as Catalysts for Hydrosilylation of Alkynes

Abstract

This work describes the preparation of a series of platinum–aminocarbene complexes [PtCl{C(N=C^a(C₆R²R³R⁴R⁵CON^b))=N(H)R¹}(CNR¹)]^a–b (8–19, 65–75% isolated yield) via the reaction of cis-[PtCl₂(CNR¹)₂] (R¹ = Cy 1, t-Bu 2, Xyl 3, 2-Cl-6-MeC₆H₃4) with 3-iminoisoindolin-1-ones HN=C^a(C₆R²R³R⁴R⁵CON^bH) (R²–R⁵ = H 5; R³ = Me, R², R⁴, R⁵ = H 6; R³, R⁴ = Cl, R², R⁵ = H 7). New complexes 17–19 were characterized by elemental analyses (C, H, N), ESI⁺-MS, Fourier transform infrared spectroscopy (FT-IR), one-dimensional (¹H, ¹³C{¹H}), and two-dimensional (¹H,¹H correlation spectroscopy (COSY), ¹H,¹³C heteronuclear multiple quantum correlation (HMQC)/¹H,¹³C heteronuclear single quantum coherence (HSQC), ¹H,¹³C heteronuclear multiple bond correlation (HMBC)) NMR spectroscopy, and authenticity of known species 8–16 was confirmed by FT-IR and ¹H and ¹³C{¹H} NMR. Complexes 8–19 were assessed as catalysts for hydrosilylation of terminal alkynes with hydrosilanes to give vinyl silanes, and complex [PtCl{C(N=C^a(C₆H₃(5-Me)CON^b))=N(H)(2-Cl-6-MeC₆H₃)}{CN(2-Cl-6-MeC₆H₃)}]^a−b (18) showed the highest catalytic activity. The catalytic system proposed operates at 80–100 °C for 4–6 h in toluene and with catalyst loading of 0.1 mol %, enabling the reaction of a number of terminal alkynes (PhC≡CH, t-BuC≡CH, and 4-(t-Bu)C₆H₄C≡CH) with hydrosilanes (Et₃SiH, Pr₃SiH, i-Pr₃SiH, and PhMe₂SiH). Target vinyl silanes were prepared in 48–95% yields (as a mixture of α/β isomers) and with maximum turnover number of 8.4 × 10³. Hydrosilylation of internal alkynes (PhC≡CPh, Me(CH₂)₂C≡C(CH₂)₂Me, and PhC≡CMe) with hydrosilanes (Et₃SiH, PhMe₂SiH) led to the corresponding trisubstituted silylated alkenes in 86–94% yields. Initial observations on the mechanism of the catalytic action of platinum–ADC catalysts 8–19 suggested a molecular catalytic cycle.

Introduction

Metal complexes with acyclic diaminocarbenes ([M]ADCs) have proven to be a valuable alternative to metal-N-heterocyclic carbene ([M]NHCs) catalysts in contemporary transition metal catalysis.¹ Both ADCs and NHCs possess a powerful combination of strong donor abilities with a wide range of steric properties and akin binding characteristics upon coordination.² From the synthetic perspective, a general approach to [M]NHCs involves the direct coordination of the in situ generated free NHCs to a metal center, whereas the most versatile route to [M]ADCs involves metal-mediated nucleophilic addition to isocyanides (Scheme 1).^1a,2a The latter template approach allows for an upfront installation of a variety of functionalities within the generated acyclic aminocarbene ligand, resulting in an efficient variation of electronic and steric properties, denticity of the carbene ligand, and eventually the catalytic properties of metal–ADCs.^1b,1c This methodology can also be used for the preparation of [M]NHCs, although in this case, the application of prefunctionalized isocyanides³ and specific nucleophiles^2c is required.

Scheme 1. Routes to [M]NHCs and [M]ADCs.

Insofar as generation of [M]ADCs is concerned, different types of nucleophiles have been successfully added to metal-bound isocyanides, including amines, hydrazines (sp³-N), alcohols (sp³-O), imines (sp²-N), imidines, amidines, hydrazides, and hydrazones (mixed sp³-N/sp²-N).^1a,1b,2c,4 Coupling with mixed sp³-N/sp²-N nucleophiles led to unconventional amino(imino)- or amino(hydrazido)carbenes.⁵ It is noteworthy that application of these easily accessible and modular ADC species in catalysis allowed the discovery of new application fields, where metal–ADC catalysts were never used previously. For instance, recently prepared palladium–amino(hydrazido)carbenes derived from the addition of hydrazides exhibit an excellent activity in the Suzuki–Miyaura reaction conducted in aqueous medium^5b and also in Sonogashira coupling performed at room temperature (RT).^5h At the same time, we reported on the first use of platinum–ADCs in catalysis, viz., for the catalytic hydrosilylation of terminal alkynes with hydrosilanes.^5h

Whereas catalytic properties of different palladium and gold–ADCs are well-documented, the application of platinum–ADCs in catalysis is restricted to only a few reports on catalytic alkyne hydrosilylation.^1b,5h We have recently revealed that amino(imino)carbene–palladium complexes derived from the addition of 3-iminoisoindolin-1-ones to palladium-bound isocyanides demonstrate a high efficiency in the Suzuki–Miyaura coupling (yields up to 81–99%, turnover numbers (TONs) up to 7.6 × 10⁴).^5d

Inspired by these results, we prepared corresponding platinum–aminocarbene derivatives and assessed their catalytic properties in a substantially more demanding reaction such as hydrosilylation of terminal alkynes.

Results and Discussion

Synthesis and Structural Characterization of Aminocarbene Complexes

Reaction of platinum(II)-isocyanides cis-[PtCl₂(CNR¹)₂] (R¹ = cyclohexyl (Cy) 1, t-Bu 2, 2,6-Me₂C₆H₃ (Xyl) 3, 2-Cl-6-MeC₆H₃4) with 3-iminoisoindolin-1-ones HN=C^a(C₆R²R³R⁴R⁵CON^bH) (R²–R⁵ = H 5; R³ = Me, R², R⁴, R⁵ = H 6; R³, R⁴ = Cl, R², R⁵ = H 7) in CHCl₃ under reflux conditions for 8 h gave platinum–aminocarbene complexes [PtCl{C(N=C^a(C₆R²R³R⁴R⁵CON^b))=N(H)R¹}(CNR¹)]^a−b (8–19) that were isolated in 65–75% yield (Scheme 2).

Scheme 2. Preparation of Platinum–ADC Complexes via Coupling of 3-Iminoisoindolin-1-ones (5–7) with cis-[PtCl₂(CNR¹)₂] (1–4).

Complexes 17–19 were characterized using elemental analyses (C, H, N), ESI⁺-MS, Fourier transform infrared (FT-IR), and ¹H and ¹³C{¹H} NMR spectroscopy, whereas the authenticity of known^5a species 8–16 was established using FT-IR and ¹H and ¹³C{¹H} NMR. Satisfactory C, H, and N elemental analyses were obtained for 17–19 and ESI⁺-MS for those showed peaks due to protonation of molecular ion [M + H]⁺. In the FT-IR spectra of 17–19, bands due to ν(C_carbene–N) and ν(N–H) appeared within 1523–1521 and 3248–3227 cm^–1 ranges, respectively. A strong ν(C≡N) stretch due to the presence of the unreacted isocyanide ligand was found at ca. 2193 cm^–1, whereas the corresponding bands due to overlapped ν(C=O) and ν(C=N) of the 3-iminoisoindolin-1-one moiety emerged between 1734 and 1611 cm^–1. The ¹H NMR spectra of carbene complexes 17–19 displayed a broad peak in the range of δ 9.7–11.4, assigned to the Pt-{C_carbene–N(H)R} proton, whereas the corresponding signal of the carbene carbon in the ¹³C{¹H} spectra resonated at ca. 200–207 ppm. Gradient-enhanced ¹H,¹H correlation spectroscopy (COSY), ¹H,¹³C heteronuclear multiple quantum correlation (HMQC)/¹H,¹³C heteronuclear single quantum coherence (HSQC), and ¹H,¹³C heteronuclear multiple bond correlation (HMBC) spectra aided the ¹H and ¹³C signal assignment in 17–19. All spectroscopic and nonspectroscopic features of 17–19 match those of other platinum and palladium complexes, i.e., [MCl{C(N=C^a(C₆R²R³R⁴R⁵CON^b))=N(H)R¹}(CNR¹)]^a−b reported previously by some of us.^5a,5d

Application of Platinum–Aminocarbene Complexes as Catalysts for Hydrosilylation of Alkynes

Hydrosilylation of terminal alkynes has become one of the most powerful ways for the preparation of organosilane species,⁶ which are useful intermediates in several synthetic transformations, including Hiyama, Hosomi–Sakurai-type allylation, and others.⁷ Despite being extensively studied, only a few types of catalysts are generally used for this process, in particular for industrial applications. Two commonly employed catalysts for the hydrosilylation reaction in the industry include Speier’s catalyst (H₂PtCl₆ in i-PrOH) or Karstedt’s catalyst Pt₂(dvtms)₂ (dvtms: divinyltetramethylsiloxane).⁸ Despite the popularity and the broad application scope, these catalysts suffer from a number of disadvantages and, over the past years, alternative platinum catalysts for alkyne hydrosilylation were described. These include platinum complexes with NHC ligands that work in toluene at 100 °C with 1.0 mol % catalyst loading⁹ or mixed platinum complexes with NHC and dvtms ligands that operate in o-xylene at 80 °C.¹⁰ Recently, we reported on the first application of the [Pt]ADC species as catalysts for alkyne hydrosilylation,^5h and in the development of this project, we assessed complexes 8–19 for this purpose.

Reaction of phenylacetylene with triethylsilane affording a mixture of vinyl silanes was chosen as a model hydrosilylation system (Scheme 3). Under the catalytic conditions specified below, the formation of only triethyl(1-phenylvinyl)silane (α product) and (E)-triethyl(styryl)silane (β-(E) product) was observed and no (Z)-triethyl(styryl)silane (β-(Z) product) was detected.

Scheme 3. Model System for Hydrosilylation of Terminal Alkynes with Silanes.

Although the formation of three isomers is possible in a such reaction, the majority of previous reports indicate that platinum-catalyzed hydrosilylation of alkynes typically brings about the formation of two isomers, namely, α and β-(E) products. Three isomers are also expected for the process involving radical pathways.^9a,11

From the industrial perspective, it is important to develop a catalytic system that can run in undried solvents in air. Therefore, we started this project by assessing the effects of solvent and the presence of moisture and air on the course of the hydrosilylation process. Conditions previously reported by some of us for other platinum–ADC catalysts were used (catalyst loading 0.1 mol %, 80 °C for 6 h and/or 100 °C for 4 h).^9a Several dry and undried solvents were used in air or under a dinitrogen atmosphere, and the results for runs in air are presented in Table 1.

Table 1. Effect of Solvent on the Reaction Course [Yields and Isomeric Content (α/β Ratio)]^a,b.

entry	solvent	yield and isomeric ratio at 80 °C/6 h	yield and isomeric ratio at 100 °C/4 h
1–2	toluene (dried)	90 (19/81)	86 (48/52)
3–4	toluenec (undried)	82 (56/44)	85 (48/52)
5–6	p-xylene	67 (22/78)	88 (28/72)
7–8	ethanol	<5	<5
9–10	water	<5	<5
11–12	none	65 (41/59)	64 (48/52)

^aPhC≡CH (5.0 × 10^–4 mol, 1 equiv), Et₃SiH (5.0 × 10^–4 mol, 1 equiv), catalyst 18 (5.0 × 10^–7 mol); solvent (0.5 mL).

^bYields of products were determined by ¹H NMR spectroscopy using 1,2-dimethoxyethane as standard, whereas the isomeric content was determined on the basis of the alkene coupling constants.

^cWater content in this solvent was 0.1% (v/v) based upon Karl Fischer coulometry.

No marked difference was observed between dried and undried toluene as the reaction medium (entries 1–4), and overall yields of the product of 82–90% were achieved at both 80 and 100 °C. Use of p-xylene (entries 5 and 6) resulted in the decrease of the product yield to 67% at 80 °C, whereas the reaction course at 100 °C seemed to be unaffected. We believe the major reason for this is the insufficient solubility of catalysts and substrates at lower temperatures. Application of ethanol or water as the solvent (entries 7–10) was unsuccessful, and only traces of the vinyl silanes were detected. In this case, a mixture of starting hydrosilanes alongside the products of their hydrolysis or alcoholysis (formed upon reaction with water or EtOH, correspondingly)¹² was generated. Finally, fairly surprising results were obtained without any solvent (entries 11 and 12). Moderate product yields (ca. 64–65%) were subsequently achieved, showing the potential of development of this system into a solvent-free one. When reaction in toluene was attempted under a dinitrogen atmosphere under the conditions of Table 1, no change in the product yield was detected, showing that the system is not influenced by the presence of air. For further studies in this report, all catalytic runs were performed in air with undried toluene as the solvent.

As the next step, we verified the effect of temperature (60–100 °C range) on the model reaction (Table 2). At RT, no catalytic reaction occurred and only the mixture of starting materials was recovered after 48 h. At 60 °C, the maximum yield of 37% were achieved after 4 h and no further accumulation of the products was observed even after 12 h of heating (entries 1, 4, 7, and 10). We observed that in this case the reaction stops at conversion of ca. 40% and the unreacted starting material remains in the reaction mixture. At 80 °C, nearly quantitative conversion of the starting material was achieved to afford silylated products with yields up to 90% after 6 h (entry 8). The same process at 100 °C led to the maximum yield of 85% achieved after 4 h. Increasing the reaction time (entries 9, 11, and 12) did not further improve the yield but resulted in the partial decomposition of the silylated products; this process occurred faster at 100 °C when compared to that at 80 °C. Monitoring of this reaction by ¹H NMR and gas chromatography–mass spectrometry (GC-MS) allowed the detection of styrene and triethyl(vinyl)silane. We believe that these species are formed in a similar fashion as reported by Marciniec¹³ and Seki,¹⁴ who showed that metal hydride complexes, that can be generated in situ from the hydrosilanes and metal source, can catalyze the conversion of vinyl silanes to styrene. Taking these results into account, all further catalytic tests were undertaken at two sets of temperature/reaction time, i.e., at 80 °C/6 h and 100 °C/4 h.

Table 2. Effect of Temperature on the Reaction [Yields and Isomeric Content (α/β Ratio)]^a,b.

entry	time, h	temperature, °C	yield and isomeric content
1	2	60	29 (30/70)
2		80	56 (21/79)
3		100	78 (52/48)
4	4	60	37 (29/71)
5		80	82 (20/80)
6		100	85 (47/53)
7	6	60	35 (23/77)
8		80	90 (19/81)
9		100	77 (25/75)
10	12	60	25 (41/59)
11		80	80 (51/49)
12		100	58 (39/61)

^aPhC≡CH (5.0 × 10^–4 mol, 1 equiv), Et₃SiH (5.0 × 10^–4 mol, 1 equiv), catalyst 18 (5.0 × 10^–7 mol); toluene (0.5 mL).

^bYields of products were determined by ¹H NMR spectroscopy using 1,2-dimethoxyethane as the standard, whereas the isomeric content was determined on the basis of the alkene coupling constants.

A comparison of the catalytic activity of all prepared aminocarbene complexes 8–19 in the model hydrosilylation system is given in Table 3. All catalysts studied are scarcely selective for Markovnikov versus anti-Markovnikov addition to alkyne. Complex 18 was the most active (entries 21 and 22; 85–90% products yield), closely followed by complexes 15 and 16 (entries 15–18; 81–83%), 14 (entries 13–14; 49–82%), 17 (entries 19–20; 42–80%), and 19 (entries 23–24; 41–75%). For 14–19, a nearly quantitative conversion of the starting material was achieved at 100 °C; the same was also observed with catalysts 15, 16, and 18 at 80 °C. For all other catalysts and conditions, an incomplete conversion of the starting materials occurs and the reaction stops presumably due to the deactivation of the catalyst.

Table 3. Comparison of the Catalytic Activity of 8–19 in the Model Hydrosilylation System [Yields and Isomeric Content (α/β Ratio)]^a,b.

entry	catalyst	yield and isomeric ratio at 80 °C/6 h	yield and isomeric ratio at 100 °C/4 h
1–2	8	83 (67/33)	25 (60/40)
3–4	9	64 (35/65)	60 (28/72)
5–6	10	66 (38/62)	60 (27/73)
7–8	11	<5	14 (57/43)
9–10	12	7 (49/51)	44 (66/34)
11–12	13	5 (44/56)	83 (67/33)
13–14	14	49 (52/48)	82 (35/65)
15–16	15	81 (62/38)	82 (60/40)
17–18	16	81 (60/40)	83 (55/45)
19–20	17	42 (41/59)	80 (42/58)
21–22	18	90 (20/80)	85 (48/52)
23–24	19	41 (54/46)	75 (43/57)

^aPhC≡CH (5.0 × 10^–4 mol, 1 equiv), Et₃SiH (5.0 × 10^–4 mol, 1 equiv), catalyst 8–19 (5.0 × 10^–7 mol); toluene (0.5 mL).

^bYields of products were determined by ¹H NMR spectroscopy using 1,2-dimethoxyethane as the standard, whereas the isomeric content was determined on the basis of the alkene coupling constants.

All of the most efficient catalysts 14–19 are derived from the coupling of aryl isocyanides in cis-[PtCl₂(CNR¹)₂] [R¹ = Xyl 3, 2-Cl-6-MeC₆H₃4] with 3-iminoisoindolin-1-ones, whereas less active catalysts 8–13 are derived from the addition to the aliphatic isocyanides. These results are also in agreement with those previously observed for the corresponding palladium complexes with amino(imino)carbenes derived from the addition of 3-iminoisoindolin-1-ones to palladium-bound isocyanides that were studied as catalysts for the Suzuki–Miyaura reaction.^5d We believe that in this case both steric and electronic factors might be responsible for this effect. From the steric perspective, planar aromatic groups in both XylCN and 2-Cl-6-MeC₆H₃NC exhibit lower hindrance upon oxidative addition of silane substrates, particularly at a lower temperature. From the electronic viewpoint, moderate electron donation from the aminocarbenes derived from the aromatic isocyanides to the metal center makes it more susceptible to reductive elimination of the silylated product. It is also known that many carbene ligands with aromatic substituents exhibit superior stability when compared to that of the aliphatic counterpart that might also account for a plausible reason for their superior catalytic efficiency. More examples of the catalyst structure/catalytic activity relationship for metal–ADCs are required before more definitive conclusions can be drawn.

We assessed the scope of our system (Table 4) using representative catalyst 18. A selection of terminal alkynes (PhC≡CH, 4-(t-Bu)C₆H₄C≡CH, and t-BuC≡CH) and silanes (Et₃SiH, Pr₃SiH, PhMe₂SiH, and i-Pr₃SiH) with different steric hindrances and electronic properties was used, thus attesting the versatility of our system for the preparation of corresponding vinyl silane products.

Table 4. Hydrosilylation of the Terminal Alkynes Employing Catalyst 18 [Yields and Isomeric Content (α/β Ratio)]^a,b.

entry	substrates	PhC≡CH	4-(t-Bu)-C6H4C≡CH	t-Bu-C≡CH
1–3	Et3SiH	90 (19/81) (6 h)	86 (19/81) (3 h)c	61/52 (7/93) (6 h)
4–6	Pr3SiH	88 (65/35) (3 h)c	80 (63/37) (3 h)c	77/70 (7/93) (1 h)c
7–9	i-Pr3SiH	70 (75/25) (6 h)c	89 (81/19) (3 h)c	48/40d (51/48) (1 h)c
10–12	PhMe2-SiH	96 (29/71) (4 h)c	95 (25/75) (6 h)	83/78 (5/95) (6 h)

^aAlkyne (5.0 × 10^–4 mol, 1 equiv), silane (5.0 × 10^–4 mol, 1 equiv), catalyst 18 (5.0 × 10^–7 mol); toluene (0.5 mL), reaction temperature: 80 °C.

^bYields of products were determined by ¹H NMR spectroscopy using 1,2-dimethoxyethane as the standard, whereas the isomeric content was determined on the basis of the alkene coupling constants.

^cReaction temperature: 100 °C.

Yields of vinyl silanes were up to 96%. For aliphatic terminal alkyne t-BuC≡CH, moderate product yields were achieved presumably due to its lower reactivity when compared to that of aromatic alkynes. For some of the alkyne/silane pairs, the variation of the reaction time and temperature gave higher product yields when compared to those under the standard conditions of 80 °C/6 h and 100 °C/4 h.

To extend the application scope for our system, we also assessed the hydrosilylation of internal alkynes (Table 5).¹⁵ To our satisfaction, we observed that symmetric disubstituted acetylenes (entries 1–4) react with silanes with a high degree of stereoselectivity to afford target E-silylated alkenes in 86–92% yields after 4 h at 100 °C. For asymmetrically disubstituted alkynes (entries 5 and 6), an expected mixture of α/β-E-silylated alkenes is generated. For all of these examples, the amount of corresponding Z-silylated products formed was below 1%.

Table 5. Hydrosilylation of the Internal Alkynes Employing Catalyst 18 [Yields and Isomeric Content (α/β Ratio)]^a,b.

entry	alkyne	silane	yield and isomeric ratio at 100 °C/4 h
1	PhC≡CPh	Et3SiH	92
2	(R6, R10 = Ph)	PhMe2SiH	88
3	Me(CH2)2C≡C(CH2)2Me	Et3SiH	86
4	(R6, R10 = Me(CH2)2)	PhMe2SiH	91
5	PhC≡CMe	Et3SiH	87 (81/19)
6	(R6 = Ph, R10 = Me)	PhMe2SiH	94 (45/55)

^aAlkyne (5.0 × 10^–4 mol, 1 equiv), silane (5.0 × 10^–4 mol, 1 equiv), catalyst 18 (5.0 × 10^–7 mol); toluene (0.5 mL).

^bYields of products were determined by ¹H NMR spectroscopy using 1,2-dimethoxyethane as the standard, whereas the isomeric content was determined on the basis of matching of chemical shifts against the authentic compounds.

The effect of catalyst loading has been assessed in the model hydrosilylation reaction with terminal alkynes (Table 6). The maximum catalyst turnover number (TON) of 8.4 × 10³ was achieved with 0.01 mol % loading of 18 within 6 h of reaction at 80 °C (entry 5). With higher catalyst loadings (entries 1–4, 0.1–1.0 mol %), no considerable improvement in the yield was achieved, whereas with lower catalyst loadings (entries 7–10, 0.001–0.0001 mol %), only traces of products were detected.

Table 6. Effect of the Catalyst Loading on the Reaction Efficiency [Yields, Isomeric Content (α/β Ratio) and TONs]^a,b.

		yields, isomeric ratio, and TON under selected conditions
entry	catalyst loading, mol %	80 °C/6 h	100 °C/4 h
1–2	1	83 (17/83) 83	82 (17/83) 82
3–4	0.1	90 (19/81) 900	85 (48/52) 850
5–6	0.01	85 (68/32) 8500	77 (68/32) 7700
7–8	0.001	9 (64/36) 9000	7 (62/38) 7000
9–10	0.0001	<5	<5

^aPhC≡CH (5.0 × 10^–4 mol, 1 equiv), Et₃SiH (5.0 × 10^–4 mol, 1 equiv), catalyst 18; toluene (0.5 mL).

^bYields of products were determined by ¹H NMR spectroscopy using 1,2-dimethoxyethane as the standard, whereas the isomeric content was determined on the basis of the alkene coupling constants.

Finally, we undertook additional tests to shed some light on the mechanism of catalytic action of 8–19. A catalytic run in the presence of metallic mercury (mercury drop test¹⁶) showed similar results to those in the absence of mercury, and no change in the reaction rate or product yields was observed. Furthermore, the accumulation of the products in the initial period followed a nearly linear time dependence and no induction period in the system was evident. These observations clearly indicated that no formation of platinum nanoparticles occurs during the reaction; catalytic system operates under typically homogeneous conditions, and the mechanistic cycle should involve molecular compounds. Several platinum and gold systems were reported to operate through the formation of catalytically competent ligand-free metal clusters.¹⁷

In our case, some of the plausible intermediates of the catalytic cycle were detected using electrospray ionization mass spectrometry (ESI-MS) with catalyst 17 and phenylacetylene/propyl silane as substrates (Figures 1 and S1). Intermediates A and B (m/z: 673.14 and 685.14) represent a product of an oxidative addition of the H–Si bond of the hydrosilane to the catalyst core. The composition of these intermediates suggests that the bidentate carbene ligand remains attached to the metal, whereas the isocyanide and chloride are lost presumably upon the catalyst activation step. Intermediate C (m/z: 751.22) represents a product of the 1,2-alkyne insertion into the Pt–H bond of the catalyst core already containing the silyl fragment immediately before the reductive elimination step to afford the target vinyl silane product.

Figure 1.

Plausible intermediates of the catalytic cycle detected.

Although the exact mechanism of the catalytic action of platinum–ADCs remains unclear and requires further elucidation, our initial observations are consistent with other catalytic applications of metal–ADC species reported previously.^1b,5h,18

Final Remarks

Results of this study can be considered from both synthetic and catalytic perspectives. From the synthetic point of view, we prepared a series of 12 platinum complexes, containing aminocarbene ligands, namely, [PtCl{C(N=C^a(C₆R²R³R⁴R⁵CON^b))=N(H)R¹}(CNR¹)]^a−b via the reaction of cis-[PtCl₂(CNR¹)₂] with 3-iminoisoindolin-1-ones. All new complexes were characterized by elemental analyses (C, H, N), ESI⁺-MS, FT-IR, and ¹H and ¹³C{¹H} NMR spectroscopy. From the catalytic perspective, we evaluated the properties of platinum–aminocarbenes prepared as catalysts for hydrosilylation of terminal alkynes with hydrosilanes. The highest catalytic efficiency was achieved with complex [PtCl{C(N=C^a(C₆H₃(5-Me)CON^b))=N(H)(2-Cl-6-MeC₆H₃)}{CN(2-Cl-6-MeC₆H₃)}]^a−b (18) that is derived from platinum-mediated coupling of aromatic isocyanide CN(2-Cl-6-MeC₆H₃) with HN=C^a(C₆R²R³R⁴R⁵CON^bH) (R³ = Me, R², R⁴, R⁵ = H 6). The designed catalytic system operates at 80–100 °C for 4–6 h in toluene with a typical catalyst loading of 0.1 mol % and allows the transformation of a variety of hydrosilanes (Et₃SiH, Pr₃SiH, i-Pr₃SiH, and PhMe₂SiH) and terminal alkynes (PhC≡CH, t-BuC≡CH, and 4-(t-Bu)C₆H₄C≡CH) into the respective vinyl silanes in yields of up to 96%. Lowering the catalyst loading to 0.01 mol % increased the TON to 8.4 × 10³. Hydrosilylation of internal alkynes PhC≡CPh, Me(CH₂)₂C≡C(CH₂)₂Me, and PhC≡CMe with hydrosilanes Et₃SiH and PhMe₂SiH gave corresponding trisubstituted silylated alkenes in 86–94% yields. The results of the mercury drop test, absence of an induction period, and the linear initial kinetics are supportive of a molecular catalytic cycle that operates under homogeneous conditions. It is notable that although amino(imino)carbene catalysts from this study are just slightly less active than amino(hydrazido)carbenes reported previously,^5h one of the major advantages of the former species is their substantial thermal stability. It enables the catalytic applications at temperatures up to 150 °C and nearly indefinite storage at RT. It was recently shown that platinum and iridium complexes with parent isocyanides exhibit outstanding and unexpected catalytic properties in the cross-linking of silicone rubbers at high temperature.¹⁹ Although use of platinum–ADCs for this tremendously important industrial process is not yet revealed, our current catalysts might serve as logical candidates for this application. Future studies aiming at expanding the field of catalytically relevant metal–ADCs and at understanding the mechanism of their catalytic action are currently underway in our group.

Experimental Section

Materials and Instrumentation

Solvents, K₂[PtCl₄], and all isocyanides were obtained from commercial sources and used as received, apart from chloroform that was purified by conventional distillation over calcium chloride. The starting cis-[PtCl₂(CNR¹)₂] (R¹ = Cy 1, t-Bu 2, Xyl 3, 2-Cl-6-MeC₆H₃4) complexes,²⁰ substituted and unsubstituted 3-iminoisoindolin-1-ones,²¹ and aminocarbene complexes 8–16^5a were prepared as previously reported. C, H, and N elemental analyses were carried out by Microanalytical Service of the Instituto Superior Técnico. ESI⁺ mass spectra were recorded on Thermo Scientific LCQ Fleet and QqTOF Impact II mass spectrometers in MeOH. Infrared spectra (4000–400 cm^–1) were measured on a Bruker Vertex-70 instrument in KBr pellets. One-dimensional (¹H, ¹³C{¹H}) and two-dimensional (¹H,¹H COSY, ¹H,¹³C HMQC/¹H,¹³C HSQC, and ¹H,¹³C HMBC) NMR spectra were recorded on Bruker Avance II+ 300, 400, and 500 MHz (UltraShield Plus Magnet) spectrometers at ambient temperature using solvent resonances as a reference.

Synthetic Work

Coupling of cis-[PtCl₂(C≡NR¹)₂] (R¹ = Cy, t-Bu, Xyl, 2-Cl-6-MeC₆H₃) with 3-Iminoisoindolin-1-one

The corresponding solid 3-iminoisoindolin-1-one (5–7) (0.20 mmol) was added to a solution of cis-[PtCl₂(C≡NR¹)₂] (0.20 mmol) in 10 mL of CHCl₃. The reaction mixture was subsequently refluxed for ca. 8 h; during the reaction course the color of the mixture changed from yellow to bright yellow-orange. After cooling down, the reaction mixture was evaporated to dryness at RT under a stream of dinitrogen and the solid residue was extracted with two 5 mL portions of CHCl₃. The bright yellow solution formed was filtered off to remove some insoluble material. The filtrate was evaporated to dryness under a stream of dinitrogen at RT to give a yellow precipitate, which was washed with five 5 mL portions of i-Pr₂O, one 1 mL portion of cold (5 °C) Et₂O, and again with five 5 mL portions of i-Pr₂O and dried in vacuo at RT. Yields of 8–19 were 65–75%. The authenticity of known species 8–16 was established upon comparison of their FT-IR and ¹H and ¹³C{¹H} NMR spectra with those previously reported.^5a

[PtCl{C(N=C(C₆H₄CON))=N(H)(2-Cl-6-MeC₆H₃)}{C≡N(2-Cl-6-MeC₆H₃)}] (17)

Anal. calcd for C₂₄H₁₇N₄Cl₃OPt: C, 42.46; H, 2.52; N, 8.25. Found: C, 41.90; H, 2.41; N, 8.55. ESI⁺-MS, m/z: 678 [M + H]⁺. IR (KBr, selected bands, cm^–1): 3248 w ν(N–H), 2968 mw, 2922 w, 2863 w ν(C–H), 2192 s ν(C≡N), 1734 m, 1676 mw, 1617 w ν(C=N) + ν(C=O), 1522 s ν(C_carbene–N), 776 s δ(C–H from Ar). ¹H NMR (DMSO-d₆, δ): 9.89 (s, br, 1H, NH), 7.83–7.76 (m, 4H) (aryls from isoindolin-1-one moiety), 7.34–7.13 (m, 6H, aryls), 2.59 and 2.52 (s, 6H). ¹³C{¹H} NMR (DMSO-d₆, δ): 206.8 (C_carbene–N), 187.4 (C=O), 184.4 (C=N), 134.8 and 133.1 (C–Cl), 132.5, 131.4, 130.0, 129.8, 129.2, 129.1, 128.2, 127.6, 127.0, 126.5, 123.4 (aryls), 19.4 and 18.4 (Me).

[PtCl{C(N=C(C₆H₃(5-Me)CON))=N(H)(2-Cl-6-MeC₆H₃)}{C≡N(2-Cl-6-MeC₆H₃)}] (18)

Anal. calcd for C₂₅H₁₉N₄Cl₃OPt: C, 43.34; H, 2.76; N, 8.09. Found: C, 42.95; H, 2.65; N, 7.95. ESI⁺-MS, m/z: 692 [M + H]⁺. IR (KBr, selected bands, cm^–1): 3246 w ν(N–H), 2968 mw, 2923 w, 2862 w ν(C–H), 2194 s ν(C≡N), 1735 mm, 1677 mw, 1617 w ν(C=N) + ν(C=O), 1523 s ν(C_carbene–N), 776 s δ(C–H from Ar). ¹H NMR (CDCl₃, δ): 9.67 (s, br, 1H, NH), 7.78–7.58 (m, 3H) (aryls from isoindolin-1-one moiety), 7.40–7.12 (m, 6H, aryls), 2.55, 2.53 (s, 6H, Me), 2.44 (s, 3H, Me from isoindolin-1-one moiety). ¹³C{¹H} NMR (CDCl₃, δ): 199.9 (C_carbene–N), 180.0 (C=O), 173.6 (C=N), 138.1 and 134.9 (C–Cl), 133.9, 131.4, 131.2, 130.0, 129.3, 128.1, 127.7, 126.4, 123.7, 123.5, 123.2, 122.9 (aryls), 22.8 (Me from the isoindolin-1-one moiety), 19.1 and 18.7 (Me).

[PtCl{C(N=C(C₆H₂(Cl₂)CON))=N(H)C₆H₃(2-Cl-6-Me)}{C≡NC₆H₃(2-Cl-6-Me)}] (19)

Anal. calcd for C₂₄H₁₅N₄Cl₅OPt: C, 38.55; H, 2.02; N, 7.49. Found: C, 38.20; H, 2.15; N, 7.30. ESI⁺-MS, m/z: 748 [M + H]⁺. IR (KBr, selected bands, cm^–1): 3227 w ν(N–H), 2967 mw, 2922 w, 2865 w ν(C–H), 2193 s ν(C≡N), 1736 m, 1675 mw, 1611 w ν(C=N) + ν(C=O), 1521 s ν(C_carbene–N), 775 s δ(C–H from Ar). ¹H NMR (DMSO-d₆, δ): 11.41 (s, br, 1H, NH), 8.10–7.98 (m, 2H) (aryls from isoindolin-1-one moiety), 7.53–7.29 (m, 6H, aryls), 2.39 and 2.34 (s, 6H, Me). ¹³C{¹H} NMR (DMSO-d₆, δ): 201.4 (C_carbene–N), 185.0 (C=O), 167.7 (C=N), 138.0, 137.6, 137.0, and 133.1 (C–Cl), 130.0, 129.7, 129.5, 129.1, 128.1, 127.6, 127.4, 127.0, 126.0 (aryls), 19.8 and 18.2 (Me).

General Procedure for Catalytic Hydrosilylation of Alkenes with Hydrosilanes (Specific Conditions are Provided in Tables 1–5)

Alkyne (5.0 × 10^–4 mol), silane (5.0 × 10^–4 mol), selected catalysts 8–19 (5.0 × 10^–7 mol), toluene (1 mL), and a poly(tetrafluoroethylene)-coated magnetic bar were placed in a 5 mL vial. The vial was closed with a septum, sealed with an aluminum crimp seal with an open top, and vented using a needle through several cycles of vacuum/dinitrogen flow. The vial was kept at 60–100 °C for 2–12 h (see Tables 1–5 for details) and cooled down to RT, and the reaction mixture was evaporated to dryness under a stream of dinitrogen. The contents of the vial were extracted with three 0.20 mL portions of CDCl₃; all fractions were combined, followed by the addition of 1,2-dimethoxyethane (1 equiv, used as an NMR internal standard), and then analyzed by ¹H NMR spectroscopy. The isomeric content was determined on the basis of the alkene coupling constants in the ¹H NMR spectra (e.g., for the model reaction of phenylacetylene and triethylsilane leading to a mixture of triethyl(1-phenylvinyl)silane (α product, J²_HH = ca. 3 Hz) and (E)-triethyl(styryl)silane (β-(E) isomer, J³_HH = ca. 19 Hz))^9a,15a,22 or the analysis and matching of chemical shifts for products against authentic samples. Quantifications were performed upon integration of the selected peaks of the product against peaks of 1,2-dimethoxyethane.

Supporting Information Available

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsomega.7b01688.

• Spectroscopic data for prepared hydrosilylation products and proposed structures of the catalytic intermediates detected using mass spectrometry (PDF)

The authors declare no competing financial interest.