Iridacycles as Catalysts for the Autotandem Conversion of Nitriles into Amines by Hydrosilylation: Experimental Investigation and Scope

Article

Cite This: Organometallics XXXX, XXX, XXX −XXX

Iridacycles as Catalysts for the Autotandem Conversion of Nitriles

into Amines by Hydrosilylation: Experimental Investigation and

Scope

Mustapha Hamdaoui, Camille Desrousseaux, Houda Habbita, and Jean-Pierre Djukic*

́

Institut de Chimie de Strasbourg, UMR 7177 CNRS, Universite de Strasbourg, 4 rue Blaise Pascal, F-67000 Strasbourg, France

S

* Supporting Information

ABSTRACT: The set of iridacycles [{C,N}Cp*Ir^III-Cl] ({C,N} =

benzo[h]quinoline, dibenzo[f,h]quinoline) containing the

(pentamethylcyclopentadienyl)iridium(III) unit were synthesized

and derivatized into cations [{C,N}Cp*Ir-NCMe]⁺associated

with BArF-type anions. The latter salts were benchmarked for

their potential catalytic properties toward HSiEt₃in a H₂-releasing

test reaction. The best-performing BArF-type salts demonstrated

the capability to promote with a low catalytic load of ca. 0.5−1

mol % the autotandem hydrosilylation of acetonitrile, propioni-

trile, and a series of arylnitrile substrates. Mechanistic inves-

tigations conﬁrmed the preliminary formation of a silane−

iridacycle adduct by electrophilic and heterolytic activation of

the Si−H bond. The molecular structure of a new example of such an adduct was resolved by X-ray diﬀraction analysis.

Theoretical considerations support a donor−acceptor [{C,N}Cp*Ir^III-H]→[SiEt₃]⁺({C,N} = benzo[h]quinolinyl) formulation

where the cationic silyl moiety acting as a Z ligand binds both Ir and H centers. Under the conditions of the catalysis, the latter

adduct is assumed to transfer readily the electrophilic [SiEt₃]⁺moiety to the nucleophilic nitrile substrate to form a N-

silylnitrilium cation and the neutral [{C,N}Cp*Ir-H]. The latter reduces the N-silylnitrilium into the corresponding N-silylimine,

which undergoes further N-silylation and reduction to yield the ﬁnal N,N-disilylamine. Under optimal conditions of low catalyst

load (70 °C, 0.5 mol %) the autotandem hydrosilylation of arylnitriles produces the silylated amines in yields >80% in 24 h.

based¹³organometallic catalysts. Other systems based on main-

INTRODUCTION

■

The catalytic reduction of nitriles is a highly valuable¹reaction,

and mild as well as selective methods of reduction to either

imines or amines are rare.²Although alkyl- and arylnitriles can

be reduced stoichiometrically with alumino- and borohydrides³

or catalytically hydrogenated with H₂,⁴the lack of selectivity

associated with high potential hazards are prohibitive obstacles

to industrial scale-up. Therefore, catalyzed hydrosilylations of

nitriles into N,N-bis(silyl)amines with minimized operational

risks are particularly sought. If one relies on the few studies

published in the ﬁeld, mild and selective catalytic hydro-

silylation of nitriles might be considered as a rather diﬃcult

transformation to achieve. The main diﬃculty seemingly arises

from the strong CN bond (bond dissociation energy 179.3

kcal/mol, 750.0 kJ/mol)⁵and from the moderate electrophilic

character of the sp carbon. In 1961, Calas et al. reported the

ﬁrst transition-metal-based catalytic hydrosilylation of nitriles

using ZnCl₂.⁶Other studies that followed this seminal report

successfully applied cobalt-,⁷rhenium-,⁸rhodium-,⁹ruthe-

nium-,¹⁰and zinc-based¹¹complexes as precatalysts. However,

in most cases heating at 100 °C combined with high catalyst

loadings was necessary to achieve a rather limited substrate

scope and selectivity. Signiﬁcant improvement in the

autotandem double hydrosilylation of nitriles was achieved by

using, for example, ruthenium-based (Scheme 1)^10,12and iron-

group metals and ﬂuoride salts were also found to catalyze this

transformation.¹⁴⁻¹⁶

To the best of our knowledge, the hydrosilylation of nitriles

catalyzed by iridium complexes has not yet been reported. The

only barely related study was reported by Bergman et al. in

2002.¹⁷In the present article, we report on the capability of

readily accessible Cp*Ir-based iridacycles to promote the

double hydrosilylation of a series of nitrile group containing

organic substrates following a so-called autotandem^13,19

catalyzed process. This report is a continuation of our eﬀort

to develop the use of such iridacycles as potential catalysts of

“one-pot” autotandem¹⁸reactions.¹⁹This endeavor has

widened the ﬁeld of organic reactions that could be catalyzed

by various formulations of the iridacyclic catalysts, which

consist of either the neutral chlorido or cationic solvato

complexes.²⁰All have shown outstanding performance in the

transformation of imines, alkynes, alcohols, acetals, and amides

and other carbonyl-containing compounds.²⁰In a recent report,

we disclosed the peculiar mode of activation of HSiEt₃by a

cationic iridacycle.^19bIt was shown that the reaction of a

solvato iridacycle with HSiEt₃mainly led to the formation of a

Received: October 9, 2017

A

DOI: 10.1021/acs.organomet.7b00749

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

Scheme 1. Chosen Examples of Early Studies on the Hydrosilylation of Benzonitrile Catalyzed by Transition-Metal Complexes

new adduct in which the Ir-bound silyl group had the

properties of a Z-type silylium ligand rather than that of a X-

type silyl ligand, according to Green’s ligand class²¹formal-

ism.²²This led us to propose that any catalysis which ensues

should follow a heterolytic pathway where the activation of

silane operates by the concerted transfer of a hydride from the

silane to the iridium center combined with the capture of the

silylium moiety: this mechanism relates to the concept of

“electrophilic activation” outlined by Tilley et al.²³in a recent

review. In the present report a highly active iridacyclic

precatalyst is identiﬁed, which displays excellent capability to

promote the autotandem double hydrosilylation of a variety of

nitrile-containing substrates as well as the hydrosilylation of

carbonyl-containing substrates, which is considered here as a

reference reaction.

Scheme 2. Synthesis of Precatalysts [3a−c][X] by

Cycloiridation of 2-Phenylpyridine (1a), Benzo[h]quinoline

(1b), and Dibenzo[f,h]quinoline (1c) Followed by Chloride

Abstraction and Replacement by Coordinated Acetonitrile

RESULTS AND DISCUSSION

■

Synthesis and Structural Characterization of New

Precatalysts. To address the capabilities of iridacycles in

promoting the hydrosilylation of nitriles, we chose to synthesize

a few new precatalysts diﬀering either in the nature of the

{C,N} chelating ligand or in the nature of the counteranion X,

assuming that slight backbone variation and proper choice of

anion could lead to an optimal catalyst. The synthesis of the

chlorido-bound iridacycles (Scheme 2) was performed by

applying a standard procedure consisting of stirring a mixture of

below, only cation [3b]⁺was prepared associated with the

counteranions OTf, BPh₄, BArF₂₀, and BArF₂₄for the purpose

of comparing the possible eﬀect of the anion on the catalytic

performance. Cation [3a]⁺was associated with BArF₂₀and

BArF₂₄, whereas [3c]⁺was only associated with BArF₂₄. All new

complexes were characterized analytically and spectroscopically

by conventional methods.

ligands 1a−c,²⁴NaOAc, and [Ir(μ-Cl)ClCp*]₂in CH₂Cl₂at

25

room temperature for over 16 h, aﬀording complexes 2a−c

with yields of between 60% (2c) and 94%. Complexes 2a−c

were subsequently all converted into acetonitrile solvato salts

3a−c by the chlorido-abstraction procedures described earlier

that use as a driving force the precipitation of either NaCl

(sodium tetraarylborates) or AgCl (AgOTf).^19bAs shown

¹H and ¹³C NMR spectra (recorded in CD₂Cl₂at 298 K,

unless otherwise stated) of all the ionic iridiacycles disclosed in

this paper showed that the chemical shifts of the NCMe ligand

can slightly vary with the nature of the {C,N} ligand and of the

counteranion X (Table 1). The case of [3b][BPh₄] deviates

B

DOI: 10.1021/acs.organomet.7b00749

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

Table 1. Selected Solution ¹H and ¹³C NMR Chemical Shifts

a

of Ionic Iridacycles

δ (ppm) of [Ir]-NCMe

δ (ppm) of [Ir]-C₅Me₅

¹³C

compound

¹H

NC

Me

¹H

C₅

Me₅

[3a][BArF₂₄]

[3a][BArF₂₀]

[3b][BArF₂₄]

[3b][BArF₂₀]

[3b][BPh₄]

[3b][OTf]

2.23

2.27

2.12

2.15

1.59

2.28

2.41

118.3

118.4

118.5

119.0

118.3

4.2

4.3

4.2

4.6

3.4

4.1

3.4

1.62

1.68

1.73

1.75

1.74

1.79

1.86

91.8

91.6

92.1

91.5

91.1

92.3

8.9

9.1

9.6

9.2

9.0

b

[3c][BArF₂₄]

10.0

a

b

Recorded in CD₂Cl₂(unless otherwise stated) at 298 K. In d₆-

acetone.

from the general trend, for the methyl ¹H and ¹³C nuclei of the

metal-bound NCMe resonate at higher ﬁeld in comparison to

other salts (¹H, ∼2.2 ppm; ¹³C, ∼4.1 ppm), i.e. at respectively

1.59 and 3.4 ppm. Single-crystal X-ray diﬀraction analyses

produced the molecular structures of salts [3a][BArF₂₀],

[3b][BPh₄], [3b][BArF₂₀], [3b][BArF₂₄], and [3c][BArF₂₄]

(cf. the Supporting Information and CIF ﬁles). For the sake of

conciseness only the structures of [3b][BArF₂₄] and [3c]-

[BArF₂₄] are displayed in Figure 1 (cf. the Supporting

Information and CIF ﬁles). The structures did not reveal any

signiﬁcant variations of distances that could be related to any

possible electronic eﬀect induced by the change of {C,N}

chelating ligand on going from 3a to the polyaromatic 3b and

3c. For instance, the Ir−N(MeCN) bond distance was found to

be invariant within the limits imposed by estimated standard

deviations. Quite interestingly, the shortest distance of

separation between the Ir atom of the cation and the B atom

of the counteranion in the crystal lattice of all investigated

complexes was found to remain at around 7.4 Å. For salts of

[3b]⁺this distance was found to slightly increase from 7.315(4)

to 7.411(2) and 7.617(2) Å on going from [3b][BPh₄] to

[3b][BArF₂₀] and [3b][BArF₂₄], which might be related to the

increase in the steric volume of the anion.

Figure 1. ORTEP-type (thermal ellipsoid) drawings of the molecular

structures of [3b][BArF₂₄] (a) and [3c][BArF₂₄] (b) drawn at the

30% probability level. Hydrogen atoms and solvent molecules are

omitted for the sake of clarity. F atoms were found with a 50%

positional disorder at some CF₃groups: only one position was

retained in this graphical representation (cf. the CIF ﬁles for details).

Selected interatomic distances (Å) and angle (deg) for [3b][BArF₂₄]:

C11−Ir1 2.073(3), N1−Ir1 2.099(3), N2−Ir1 2.034(3), C24−N2

1.153(5), C11−Ir1−N1 79.15(12) . Selected interatomic distances

(Å) and angle (deg) for [3c][BArF₂₄]: C1−Ir1 2.051(5), N1−Ir1

2.107(4), N2−Ir1 2.061(4), C1−Ir1−N1 78.23(18).

1

Furthermore, a close analysis of the H NMR spectrum of a

pure sample of [3b][BPh₄] left to stand in solution (CD₂Cl₂)

at room temperature for less than 1 h revealed the presence of

several new signals assigned to the products of decomposition

of [3b][BPh₄]. One of the products was unambigously

identiﬁed to be free benzene (singlet at 7.36 ppm;²⁶ratio

[3b][BPh₄]:benzene ≈ 1:0.2), suggesting the cleavage of the

C−B bond of one of the four phenyls of the [BPh₄]⁻anion.

Other untraceable product(s) were distinguishable from weak

signals appearing at 7.45 (t, 6H relative to benzene) and 7.61

ppm (m, 6H relative to benzene). It is worth noting that free

NCMe could also be detected (broad singlet at 1.97 ppm, 3H

relative to benzene).

reaction solution is captured at high sampling rates by an

adequately chosen absolute pressure−voltage linear transducer

interfaced to an analog−logic converter and a computer. Since

reproducible determinations of kinetic parameters were

achieved, this setup was used for precatalyst discovery. Indeed,

the fact that catalyzed O-dehydrosilylation of alcohols and

hydrosilylation of carbonyls, imines, etc. were assumed to

involve common key silane activation steps and intermediates,

this setup could potentially help identifying the most promising

precatalysts for the challenging duty of achieving the hydro-

silylation of nitriles.

Therefore, our eﬀorts mainly focused on the evaluation of

BArF-based (BArF₂₀and BArF₂₄) salts, knowing that they

generally produce the highest catalytic activity for this

reaction.^19bTable 2 gathers the main results collected for the

comparative evaluation of ﬁve precatalyst salts: i.e., [3a]-

[BArF₂₀], [3b][BArF₂₀], [3a][BArF₂₄], [3b][BArF₂₄], and

[3c][BArF₂₄]. Triethylsilane was favored here, for it gave the

fastest reactions and the highest conversions in comparison to

other considered silanes.

Benchmark of Iridacyclic Precatalysts. In a previous

report,^19bwe presented a straightforward setup for the

qualitative evaluation of the performance of iridacyclic catalysts

in the dehydrosilylation of alcohols by silanes. This reaction,

which releases a signiﬁcant amount of dihydrogen gas within a

few seconds, was carried out in a thermo-regulated sealed glass

reactor and monitored by piezometry. In the typical

experimental setup (cf. Scheme S1 in the Supporting

Information) the variation of internal reactor pressure caused

Entries 1−5 in Table 2 give experiments carried out using a

1:1 ratio of benzyl alcohol and triethylsilane (eq 1), whereas

19

by the release of H₂in the overhead volume above the

C

DOI: 10.1021/acs.organomet.7b00749

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

Table 2. Benchmark of BArF Salts of Iridacycles [3a−c]⁺through O-Silylation of Benzyl Alcohol at 20 °C

a

b

c

d

entry

precatalyst (amt (μmol))

t (s)

yield (%)

v_i(esd) (10⁻²M s⁻¹

)

TON (%)

TOF_t(h⁻¹

)

e

1

[3a][BArF₂₀] (4.8)

[3b][BArF₂₀] (4.8)

[3a][BArF₂₄] (5.0)

[3b][BArF₂₄] (4.4)

[3c][BArF₂₄] (4.8)

[3a][BArF₂₄] (4.8)

37

81

100

85

13.2(3)

57(5)

10(1)

42(9)

42(2)

8(6)

812

1002

818

0.08 × 10⁶

e

2

1.6

34

2 0.2 × 10⁶

0.09 × 10⁶

2.9 × 10⁶

1.6 × 10⁶

0.2 × 10⁶

1.6 × 10⁶

e

3

e

4

1.4

2.3

16

100

1093

1002

1093

e

5

f

6

f

7

[3b][BArF₂₄] (4.4)

2.5

40(2)

a

b

c

Yield in evolved H₂. initial rate v_i= −(d[Et₃SiH]/dt)_iinferred from the rate of H₂gas release. TON: turnover number, calculated at the highest

d

conversion as the molar ratio of inferred silylated alcohol/initial precatalyst. TOF_t: turnover frequency calculated at time t corresponding to the

maximum of conversion calculated as the ratio TON/time. Conditions unless speciﬁed otherwise: benzyl alcohol (0.50 mL, 4.8 mmol), HSiEt₃(0.8

mL, 5.0 mmol), precatalyst (ca. 0.1 mol %, ca. 5 μmol) in a respective 962:1000:1 ratio in 1,2-dichloroethane (0.5 mL) at 20 0.1 °C. 1.9 equiv of

e

f

benzyl alcohol was used (1 mL, 9.6 mmol): i.e., a respective 1920:1000:1 ratio in 1,2-dichloroethane (0.5 mL) at 20 0.1 °C.

entries 6 and 7 are associated with experiments using a ca. 2:1

benzyl alcohol HSiEt₃ratio.

In all cases, data in Table 2 suggest that salts based on [3a]⁺

the largely exothermic double hydrosilylation of the released

MeCN catalyzed by [4]⁺(eq 2).^19bAs far as the eﬀect of BArF

are far slower and less eﬃcient in comparison to those based on

[3b]⁺and [3c]⁺; the turnover frequencies (TOF_t) are 1 order

counteranions is concerned, Table 2 reveals equal eﬃciencies

of magnitude lower for [3a]⁺. Even though total conversion can

for BArF₂₀and BArF₂₄salts: the TOF_tand initial rate v_ivalues

be reached over reaction times much longer than those

considered in this benchmark procedure, the turnovers for

[3a]⁺-based salts are 20% lower than for [3b]⁺- and [3c]⁺-

based precatalysts. The order of catalytic eﬃciency is [3b]⁺>

[3c]⁺≫ [3a]⁺. Quite stunning is the time required to produce

100% of the expected H₂gas amount with [3b]⁺and [3c]⁺

salts: less than 3 s after the injection of the ClCH₂CH₂Cl

solution of precatalyst in the mixture of benzyl alcohol and

HSiEt₃.

The short induction period of ca. 1 s noticeable in Figure 2

for almost all cases of precatalysts can be rationalized by the

activation step necessary to convert the precatalyst into the

active catalyst, which is the silylium-bound hydrido-iridium

intermediate [4]⁺(eq 2), one example of which was reported

previously.^19bThis transformation requires the endoergonic

departure of MeCN followed by the activation of HSiEt₃and

for [3b][BArF₂₀] and [3b][BArF₂₄] are of the same magnitude,

i.e. around 2.5 × 10⁶h⁻¹and 48 × 10⁻²M s⁻¹, respectively.

Finally, the eﬀect of the alcohol:silane ratio is better evaluated

by v_i, which depicts the rate of H₂gas release in the initiation

time period, than by TOF_t. TOF_t, which gives an average

estimate of the rate of the catalysis, suggests contradictory

conclusions. For [3b][BArF₂₄], excess alcohol seemingly

decreases the TOF_tvalue by ca. 10⁶h⁻¹, which could be

interpreted as a sign of coordinative inhibition of the catalyst by

the alcohol. However, the opposite observation is made for

[3a][BArF₂₄], which seemingly increases its TOF_twhen a 2:1

alcohol:silane ratio is used. When initial rates v_iare considered

instead for both [3a][BArF₂₄] and [3b][BArF₂₄], one can

notice that the values are, if not constant, only slightly lowered

when the amount of alcohol increases, which rather suggests

that no coordinative inhibition of the catalysis by the alcohol

takes place in the early stages of the catalysis.

In fact, the most important aspect here is that, given the

limited amount of ClCH₂CH₂Cl used, the polarity of the

solution is mostly governed by the change in composition of

the medium as the dehydrosilylation reaction develops. Indeed,

the steady replacement of alcohol by the silyl ether over time

induces a signiﬁcant drop of polarity of greater extent for a 1:1

alcohol:silane starting mixture than for a 2:1 starting mixture.

This change in polarity is expected to induce a constant change

in ion-pair separation, which may aﬀect the rate of the reaction.

In summary, this benchmark clearly points to BArF salts of

[3b]⁺as the most active in the dehydrosilylation of benzylic

alcohol and places them among the possible leading

precatalysts for the hydrosilylation of nitriles.

Postulated Mechanism of Autotandem Double

Hydrosilylation of Nitriles. Little experimental evidence is

known in support of mechanisms at play in the catalytic

hydrosilylation of nitriles in general.^10,11b,27However, most

experimental and theoretical studies of hydrosilylation reactions

have suggested rather divergent mechanisms, depending on

Figure 2. Variation of the relative pressure of hydrogen (bar) vs time

(s) in the dehydro-O-silylation of benzyl alcohol by HSiEt₃(1:1 ratio)

catalyzed by [3a−c][X] (X= BArF₂₀, BArF₂₄).

D

DOI: 10.1021/acs.organomet.7b00749

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

each studied system.²⁸In 2010, Nikonov et al. proposed a

mechanism of hydrosilylation of nitriles catalyzed by [[Ru]-

(NCMe)(η²-HSiR₃)][B(C₆F₅)₄] intermediates ([Ru] = Ru-

(ⁱPr₃P)Cp), which supposedly are formed upon reaction of

silanes with [[Ru](NCMe)₂][B(C₆F₅)₄] precursors.¹⁰The

crucial step was proposed to be the abstraction of the [SiR₃]⁺

moiety from [[Ru](NCMe)(η²-HSiR₃)][B(C₆F₅)₄] by the

nitrile substrate to form [[Ru](NCMe)(H)] and a transient

silylnitrilium species. This step, invoked in previous DFT-based

studies for the carbonyl hydrosilylation reaction, was

considered as predominant over the dissociative Ojima-type

pathway.²⁹More recently, a DFT-supported study conducted

by Wu et al. reached similar conclusions.^27aThe authors

proposed an outer-sphere silylium abstraction pathway, with

the involvement of a “S_N2-Si” transition state as a plausible

“interaction picture” between the substrates and the Ru catalytic

intermediate.

bridged bis Ir complex [7]⁺or with any potential donor ligand

present in the medium (nitrile, imine, amine). Under the

conditions of the catalysis (T > 298 K, high concentration of

RCN substrate, vide infra) this catalytic cycle is a reasonable

framework that nonetheless raises a few questions: can the large

excess of silane react with highly electrophilic intermediates

such as activated nitrile, imine, and the elusive [SiEt₃]⁺species,

as suggested by recent reports published by Oestreich et al.,³²

Chang et al.,¹⁴and Stephan et al.,¹⁵and compete with the

hydride transfer that is expected to operate from 5? Do dimeric

species [7]⁺have a role diﬀerent from that of somewhat

sequestrating portions of active Ir species away from relevant

catalysis, as reported previously for the O-silylation of

alcohols?^19bFor convenience, all of the mechanistic inves-

tigations deal with in situ characterizations of elementary steps

of the catalysis involving either precatalyst [3b]⁺or [3a]⁺

regardless of the nature of the BArF anion; BArF₂₀is used in

place of BArF₂₄in some cases for technical reasons.

Scheme 3 depicts the proposed idealized mechanism for the

catalyzed autotandem double hydrosilylation of nitriles based

Hydrosilylation of Acetonitrile. The starting point was to

check whether the hydrosilylation of MeCN with HSiEt₃was

possible under catalytic loadings of [3a][BArF₂₄], using

acetonitrile itself as the solvent (“solvent-free” conditions). A

blank control experiment consisting of the reaction of MeCN

(1 equiv) with HSiEt₃(2 equiv) expectedly resulted in no

perceptible conversion (Table 3, entry 1). The same reaction

was subsequently conducted in the presence of 1 mol % of

[3a][BArF₂₄] (relative to MeCN) at 50 °C for 44 h (Table 3,

Scheme 3. Proposed Mechanism of the Autotandem Double

Hydrosilylation of Nitriles Catalyzed by Cationic Iridacycles

1

entry 2): the monitoring of the reaction mixture by H NMR

spectroscopy (CD₂Cl₂, 298 K) revealed the full conversion of

MeCN and only a partial conversion of HSiEt₃(∼65%). In

addition to the typical signals of the expected product of the

double hydrosilylation of MeCN, i.e. N,N-bis(triethylsilyl)-

ethylamine EtN(Et₃Si)₂(characterized by its typical quartet for

the Et group at δ 2.86 ppm, J = 7.1 Hz), other signals belonging

mainly to two side products were also observed with

characteristic peaks of the Et group respectively at δ 3.1 ppm

(q, J = 7.1 Hz) and δ 2.76 ppm (qt, J = 7.2 Hz). Electrospray

ionization (ESI) mass spectrometry analysis of the reaction

mixture conﬁrmed the presence of EtN(SiEt₃)₂along with side

product 9H₂(vide infra, see spectra in Figures S54 and S55 in

the Supporting Information). When the temperature of the

reaction medium was raised from 50 to 80 °C, the same

mixture of products was noticed even after 100 h of reaction

(Table 3, entry 3). Similar features were observed with

[3b][BArF₂₄] as precatalyst under either solvent-free or

cosolvent (CH₂Cl₂) conditions (Table 3, entries 4−7). In

addition, the reaction between acetonitrile and other silanes

such as HSiPh₃(Table 3, entry 8) and H₃SiPh (Table 3, entry

9) under cosolvent conditions (CH₂Cl₂) proved to give mainly

the same mixture of side products in comparison to HSiEt₃

(Table 3, entry 7). NMR spectra related to the above results

can be found in the Supporting Information.

on our earlier report on the activation of [3a]⁺leading to silane

adduct [4a]⁺.^19b,22Step 1 is the activation step that converts the

solvato−Ir precatalyst into adduct [4]⁺by consumption of 3

equiv of HSiEt₃per molecule of precatalyst. This crucial step

leads to adduct [4]⁺, which holds an Ir-bound silyl group of

high electrophilicity, capable of being abstracted by any Lewis

base present in the medium. Step 2 corresponds to the

abstraction of the SiEt₃“silylium” group, which can be captured

by the nitrile substrate in the early stages of the catalysis and by

the in situ formed N-silylimine intermediate in the later stages

of the catalysis to aﬀord N-silylnitrilium and N-silyliminium

cations, respectively.

In step 3, the latter cations undergo the transfer of hydride

from 5 (5b,³⁰5a³¹) to produce the resulting N-silylimine and

the N,N-disilylamine alongside elusive cation [6]⁺. The latter

can react either with HSiEt₃(step 4) to regenerate adduct [4]⁺

or, as shown farther (step 5), with 5 to produce the μ-hydrido-

We decided to further explore the reaction conducted under

solvent-free conditions, for which the results are described in

1

entries 2 and 3 of Table 3. H NMR spectroscopy analysis

indicated that the HSiEt₃:MeCN:EtN(Et₃Si)₂ratio changed

over time from 23:6:1 to 2.2:0:1. This suggests that MeCN was

consumed not only by its expected reaction with HSiEt₃to give

EtN(Et₃Si)₂but also by a secondary reaction with at least one

other molecule. It became obvious that at least one of the side

products observed throughout the reaction could originate

from a coupling reaction between MeCN and another

component of the mixture. To solve this issue, the separation

E

DOI: 10.1021/acs.organomet.7b00749

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

Table 3. Investigation of the Hydrosilylation of Acetonitrile and Propionitrile

a

b

yield (%) [TON]

entry

R

HSiR′₃

[cat.] (amt (mol %))

solvent (0.5 mL)

T (°C)

t (h)

RCH₂CH₂N(SiR′₃)₂

side products

c

1

H

HSiEt₃

HSiPh₃

H₃SiPh

HSiEt₃

HSiPh₃

H₃SiPh

80

50

80

60

80

24

44

100

24

72

24

72

24

<1

10 [10]

14 [14]

13[13]

2 [20]

7 [70]

8 [8]

<1

88

86

53

13

3

c

2

H

[3a][BArF₂₄] (1)

[3b][BArF₂₄] (1)

[3b][BArF₂₄] (0.1)

[3b][BArF₂₄] (1)

[3b][BArF₂₄] (1 mol %)

[3b][BArF₂₄] (1)

c

3

H

d

4

H

e

5

H

f

6

H

CH₂Cl₂

g

7

H

10

78

2

g

8

H

<1

g

9

H

2 [2]

h

10

CH₃

41 [41]

95 [95]

57 [57]

2 [2]

<1

h

11

i

12

i

13

<1

a

b

1

Yields are based on the alkylnitrile and were determined by H NMR spectroscopy using 1,3,5-tri-tert-butylbenzene as internal reference. TON:

c

turnover number, calculated at the highest conversion as the molar ratio of product to initial precatalyst. Conditions: acetonitrile (0.5 mL, 9.7

mmol), HSiEt₃(3.1 mL, 19.4 mmol), [3a][BArF₂₄] (134.4 mg, 97 μmol), molar ratio nitrile:silane:precatalyst = 100:200:1. Conditions: acetonitrile

(0.2 mL, 3.8 mmol), HSiEt₃(1.34 mL, 8.4 mmol), [3b][BArF₂₄] (53.8 mg, 38 μmol), molar ratio nitrile:silane:precatalyst = 100:221:1. Conditions

similar to those in entry 4, except that 5.4 mg of [3b][BArF₂₄] was used (3.8 μmol), molar ratio nitrile:silane:precatalyst = 1000:2210:1. Conditions:

d

e

f

acetonitrile (1 mL, 19.2 mmol), HSiEt₃(6.73 mL, 42.1 mmol), precatalyst (27 mg, 19 μmol), molar ratio nitrile:silane:precatalyst = 1010:2216:1.

g

Conditions: acetonitrile (0.1 mL, 1.9 mmol), HSiR′₃(4.2 mmol), [3b][BArF₂₄] (27 mg, 19 μmol), molar ratio nitrile:silane:precatalyst = 100:221:1.

h

Conditions: propionitrile (0.2 mL, 2.8 mmol), HSiEt₃(1 mL, 6.2 mmol), [3b][BArF₂₄] (39.5 mg, 28 μmol), molar ratio nitrile:silane:precatalyst =

i

100:221:1. Conditions: propionitrile (0.1 mL, 1.4 mmol), HSiR′₃(0.5 mL, 3.1 mmol), [3b][BArF₂₄] (20 mg, 14 μmol), molar ratio

nitrile:silane:precatalyst = 100:221:1.

of the reaction mixture components was attempted. Bulb-to-

bulb distillation under vacuum of the latter reaction mixture

yielded two main fractions (named respectively fractions 1 and

2). Fraction 1 (obtained at ∼70 °C/10 mmHg) contained a

1

coupling product which was structurally identiﬁed by H and

¹³C NMR spectroscopy as 9H₂, i.e. the product of the

desilylative hydrolysis of the coupling product 9 (Figure 3).

Figure 4. ORTEP-type diagrams of 9H₂(right) and 9′H₂(left) in the

Figure 3. N-ethyl-ethanimidamide 9H₂, its tautomer 9′H₂, and 10H₂.

asymmetric unit drawn at 30% probability. Selected interatomic

lengths (Å) and angles (deg): C1−C2 1.5103(19), C2−N1

1.2872(18), C2−N2 1.3540(16), C3−N1 1.4644(18), C3−C4

1.515(2), C5−C6 1.5142(19), C6−N4 1.3014(18), C6−N3

1.3394(18), C7−N3 1.4518(18), C7−C8 1.492(3), C2−N1−C3

117.15(11), C6−N3−C7 121.55(12).

Whereas typical signals for 9 were usually observed in the

hydrosilylation reaction mixture at respectively δ(¹H, CD₂Cl₂)

3.12 ppm (q, J = 7.1 Hz, −NCH₂−CH₃) and 2.01 ppm (s,

−(N)C−CH₃), those of isolated 9H₂resonate at slightly

diﬀerent chemical shifts, i.e. at respectively δ(¹H, CDCl₃) 3.16

ppm (q, J = 7.0 Hz, −NCH₂−CH₃) and 1.94 ppm (s,

−(N)C−CH₃). The molecular structure of 9H₂was further

conﬁrmed by X-ray diﬀraction analysis of single crystals grown

by slow evaporation of a pure sample of fraction 1 at −30 °C

for around 3 months. Due to the instability of the latter crystals,

their elemental analysis aﬀorded three inconsistent values, the

ﬁrst one being the closest to the expected theoretical value (see

the Experimental Section for characterization data as well as the

related NMR spectra depicted in the Supporting Information).

The crystal lattice was found to consist of compound 9H₂and

its tautomer 9′H₂(Figure 3). The molecular structure of both

compounds is depicted in Figure 4 (vide infra and the

Supporting Information for further details). ESI mass

spectrometry analysis of fraction 1 indicated the presence of

9H as the major product: i.e., the product of the desilylative

hydrolysis of one N−Si bond in 9 (m/z [9H + H]⁺= 201.2 Da,

cf. Figure S56 in the Supporting Information). The second

fraction (obtained at ∼100 °C; 10 mmHg) contained a 1.3:1

mixture of mainly the two desilylated coupling products 9H₂

and 10H₂, respectively (Figure 3). Compound 10H₂was

identiﬁed by ¹H and ¹³C NMR spectroscopy (cf. Figures S50 −

S52 in the Supporting Information) as the coupling product of

9′H₂with acetonitrile. Single crystals were grown by slow

evaporation of pure sample of fraction 2 at −30 °C for around

3 months. X-ray diﬀraction analysis of the latter single crystals

F

DOI: 10.1021/acs.organomet.7b00749

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

revealed structures identical with those obtained from fraction

1: i.e., 9H₂and 9′H₂. One portion of the reaction mixture

which was not submitted to the procedure of distillation was

kept for ca. 1 month at −30 °C, after which a precipitate was

(HSiR₃)][B(C₆F₅)₄] (Scheme 1b) under solvent-free con-

ditions (23 h, ambient temperature), the authors noticed that

95% of MeCN was converted into the desired product

CH₃(H)CNSiPhMe₂(10%) and a coupling product

identiﬁed to be CH₃(H)CNCH(CH₃)N(SiMe₂Ph)₂(60%)

and other unknown products. The authors proposed that

CH₃(H)CNCH(CH₃)N(SiMe₂Ph)₂might result from the

“N−Si addition across the CN bond”. Unfortunately, no

further evidence was provided.

1

collected. H spectroscopy analysis of the latter solid indicated

a spectroscopic signature similar to that of 9H₂(cf. Figure S53

in the Supporting Information).

Crystals of each fraction were submitted to X-ray diﬀraction

analysis. A crystal of the ﬁrst fraction was found to belong to

space group C2c with three independent molecules within the

asymmetric unit: that is, N-ethylethanimidamide 9H₂and its

tautomer 9′H₂in a 2:1 ratio (Figure 3). The crystal of the

second fraction that matched space group P2₁/c contained 9H₂

and 9′H₂in a 1:1 ratio. Figure 4 displays the content of the

asymmetric unit of the latter crystal.

Methods for the catalytic coupling reactions of amines with

nitriles are well established.³³For example, Bochkarev et al.^33a

described the amination of acetonitrile by various amines, such

as i-PrNH₂and n-PrNH₂, which was found to be promoted by

lanthanide iodides LnI₂(Ln = Pr, Nd, Dy, Tm). Interestingly,

the X-ray structure obtained by the authors for the coupling

product of MeCN with ⁱPrNH₂, i.e. MeC(NH)NHⁱPr, is very

similar to the structure obtained for 9H₂(and its tautomer

9′H₂).

We postulate that 9 results from the coupling reaction of

MeCN with the product of double hydrosilylation of MeCN:

i.e., EtN(Et₃Si)₂(Scheme 4). The latter process might be

In addition, the protons of the Me group (i.e. MeC(

Scheme 4. Proposed Mechanism for the Formation of 9

1

NH)NHⁱPr) resonate at δ 1.94 ppm (s) in the H NMR

spectrum (CDCl₃, 200 MHz). This is consistent with the data

obtained for 9H₂, for which a chemical shift at δ 1.94 ppm (s)

was observed for the Me protons (CDCl₃, 500 MHz) (see full

NMR data provided in the Experimental Section). In the course

of the present study, it was found that d₆-9H₂can also be

formed by hydrosilylation of CD₃CN with HSiEt₃at room

temperature and even at very low temperature (−50 °C)

(CD₂Cl₂, [3a][BArF₂₄] (10 mol % relative to CD₃CN)) (vide

infra).

Hydrosilylation of Propionitrile (EtCN). Upon heating at

80 °C for 24 h a reaction mixture containing EtCN (1 equiv),

HSiEt₃(2.2 equiv), and [3b][BArF₂₄] (1 mol % relative to

EtCN), the desired product 1,1,1-triethyl-N-propyl-N-

(triethylsilyl)silanamine, i.e. PrN(SiEt₃)₂, was formed in 41%

yield (Table 3, entry 10). However, when the heating was

promoted by salts [3a,b][BArF₂₄], which may activate MeCN

by coordination toward the nucleophilic attack by EtN(SiEt₃)₂.

However, at this stage one cannot exclude alternative

mechanisms, which would involve as coupling partners either

free or coordinated MeCN and ethylamine: i.e., the desilylated

form of NEt(SiEt₃)₂. Nikonov et al. reported similar

observations in 2010.¹⁰When acetonitrile and HSiPhMe₂

were subjected to 0.4 mol % of [CpRu(PiPr₃)(MeCN)-

Table 4. Investigation of the Hydrosilylation of Benzonitrile by Various Silanes

a

b

yield (%) [TON]

entry

HSi[R]

[cat.] (amt (mol %))

solvent (0.5 mL)

t (h)

ix

nx

c

1

HSiEt₃

HSiPh₃

H₃SiPh

[3a][BArF₂₄] (1)

[3b][BArF₂₄] (0.001)

[3b][BArF₂₄] (1)

20

24

72

0.1

48

24

>99 [100]

<1

d

2

>99 [100]

<1

e

3

CH₂Cl₂

<1

f

4

>99 [100]

30 [30]

<1

f

5

70 [70]

>99 [100]

11 [11]

<1

g

6

g

7

42 [42]

<1

g

8

a

b

Yields are based on PhCN and were determined by ¹H NMR spectroscopy using 1,3,5-tri-tert-butylbenzene as internal reference. TON: turnover

c

number, calculated at the highest conversion as the molar ratio of product to initial precatalyst. Conditions: benzonitrile (1 mL, 9.7 mmol), HSiEt₃

(3.1 mL, 19.4 mmol), [3a][BArF₂₄] (134.4 mg, 97 μmol), 50 °C, molar ratio nitrile:silane:precatalyst = 100:200:1. Reaction conducted under

d

e

similar conditions and in the same vessel in comparison to entry 1, except that the temperature was raised from 50 to 80 °C. Conditions:

benzonitrile (1 mL, 9.7 mmol), HSiEt₃(3.4 mL, 19.4 mmol), [3b][BArF₂₄] (0.5 mL from a 2 × 10⁻⁴M solution in CH₂Cl₂), 70 °C, molar ratio

f

nitrile:silane:precatalyst = 100:200:1. Conditions: benzonitrile (0.18 mL, 1.77 mmol), HSiEt₃(0.62 mL, 3.9 mmol), [3b][BArF₂₄] (25 mg, 17.7

μmol), 40 °C, molar ratio nitrile:silane:precatalyst = 100:220:1. Conditions: benzonitrile (0.2 mL, 1.94 mmol), HSi[R] (4.23 mmol), [3b][BArF₂₄]

g

(27.3 mg, 19.4 μmol), 80 °C, molar ratio nitrile:silane:precatalyst = 100:218:1.

G

DOI: 10.1021/acs.organomet.7b00749

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

continued at 80 °C for an additional 48 h (72 h in total), the

NMR-based yield in PrN(SiEt₃)₂was ∼95% (Table 3, entry

11). Even though PrN(SiEt₃)₂, was not isolated, its identity was

assessed by its characteristic ¹H NMR spectrum, which is

available in Figures S65 and S66 in the Supporting Information.

It is worth noting that only the desired product was formed in

the case of propionitrile, with the exclusion of any coupling side

product. One plausible explanation is that the ethyl group of

propionitrile, being a more sterically demanding group than the

methyl group of MeCN, makes the formation of side products

less favorable. Finally, in comparison to HSiEt₃(Table 3, entry

11), other silanes such as HSiPh₃(Table 3, entry 12) and

H₃SiPh (Table 3, entry 13) gave the expected products of

autotandem double hydrosilylation of propionitrile in respec-

tively ∼57% yield for HSiPh₃and ∼2% yield for H₃SiPh.

Hydrosilylation of Benzonitrile. As shown in entries 1

and 2 of Table 4, benzonitrile reacted more readily and

selectively with HSiEt₃in comparison to MeCN under the

same solvent-free catalytic conditions (Table 3, entry 2). In fact,

when the hydrosilylation of 1 equiv of benzonitrile with 2 equiv

of HSiEt₃was conducted in the presence of 1 mol % of

[3a][BArF₂₄] (relative to benzonitrile) upon heating at 50 °C

of 100 (Table 4, entry 6). Under similar conditions, the

hydrosilylation of benzonitrile with HSiEt₃(Table 4, entry 6)

turned out to be more eﬃcient than that of propionitrile (Table

3, entry 10). This observation is consistent with the expected

electron-withdrawing mesomeric activation of arylnitriles in

comparison to alkylnitriles that is expressed by an enhanced

reactivity of the former toward hydrides. NMR data and

assigned NMR spectra related to the results described above are

provided in the Supporting Information.

Optimization of the Precatalyst Load for the Hydro-

silylation of Benzonitrile with HSiEt₃. The inﬂuence of the

proportion of precatalyst [3b][BArF₂₄] on the course of the

hydrosilylation of benzonitrile was investigated. We ﬁrst looked

at the inﬂuence of the proportion of the precatalyst at room

temperature and then raised the temperature to 70 °C. The

proportion of precatalyst was varied from 0.05 to 1 mol %

relative to the concentration of benzonitrile. The conditions for

each catalytic run are described in the Experimental Section.

This study is summarized in the 3D plot displayed in Figure 5,

1

for 20 h, H NMR spectroscopic analysis (CD₂Cl₂, 298 K) of

the reaction mixture showed the quantitative conversion of

benzonitrile (>99%) and the exclusive formation of the product

of monohydrosilylation of benzonitrile: i.e., N-(triethylsilyl)-

1

benzylimine (i1). The identity of i1 was assessed by H NMR

spectroscopy, which showed the presence of a singlet peak at δ

9.06 ppm belonging to the proton of the imino group (i.e.,

Ph(CH)N(SiEt₃)) (our NMR data matched those already

reported by Stephan et al.¹⁵). When the temperature of the

reaction mixture was raised from 50 to 80 °C, analysis of the

reaction medium after 24 h by ¹H NMR spectroscopy (CD₂Cl₂,

298 K) showed the complete conversion of benzonitrile and

the selective and quantitative formation of the product of

double hydrosilylation of benzonitrile: i.e., N,N-bis-

(triethylsilyl)benzylamine, (n1) (Table 4, entry 2). Product

n1 is characterized by a singlet peak at δ 4.20 ppm, which is

typically the NMR signature of a proton bound to an sp³

carbon atom positioned α relative to a nitrogen atom (i.e.,

Ph(CH₂)N(SiEt₃)₂) (our NMR data matched those already

reported by Stephan et al.¹⁵). Similar results were obtained

under cosolvent conditions (CH₂Cl₂). While no reaction

occurred with a 0.001 mol % loading of [3b][BArF₂₄] (Table

4, entry 3), using 1 mol % of precatalyst allowed the selective

and quantitative formation of i1 upon vigorous stirring and

moderate heating of the reaction medium to 40 °C for few

minutes (reaction conducted in a sealed J. Young NMR tube)

(Table 4, entry 4).

When the same reaction mixture (described in Table 4, entry

4) was left to stand at room temperature for additional 48 h, a

mixture of i1 and n1 was formed in a respective ratio of ∼1:2.5

(Table 4, entry 5). The latter result indicates that the room-

temperature hydrosilylation of i1 into n1 is kinetically less

favorable than the hydrosilylation of benzonitrile into i1. As a

consequence, it seemed mandatory to heat the reaction mixture

to at least 70 °C to reduce the reaction time required for the

quantitative and exclusive formation of n1. This fact was further

conﬁrmed during the optimization study described in the next

section. In comparison to HSiPh₃(Table 4, entry 7) and

H₃SiPh (Table 4, entry 8), HSiEt₃appeared to be the most

reactive silane for the hydrosilylation of benzonitrile, giving

selectively the desired product in quantitative yield with a TON

Figure 5. Plot of the yield of n1 (%) over time (in h) as a function of

the loading (mol %) in [3b][BArF₂₄] in the hydrosilylation of

benzonitrile (0.1 mL, 0.97 mmol) by HSiEt₃(0.35 mL, 2.13 mmol) at

70 °C.

which is a plot of the yield (%) of the product of

bishydrosilylation of benzonitrile n1 (y axis) vs time (h) (x

axis) as a function of the proportion (mol %) of [3b][BArF₂₄]

(z axis).

It was found that, independently of the initial loading of

[3b][BArF₂₄], no conversion of the substrates occurred at

room temperature (see Table S2 in the Supporting

Information). However, when the temperature of the reaction

mixture was raised to 70 °C, the products of monohydrosily-

lation (i1) and bishydrosilylation (n1) of benzonitrile began to

appear in the reaction mixture. Through optimization of the

proportion of precatalyst [3b][BArF₂₄] and the time required

for quantitative (>99%) and selective formation of the product

n1, it was found that a loading of 0.5 mol % of precatalyst

(relative to benzonitrile) and a reaction time of 24 h constitute

optimal conditions for further substrate scope study (vide

H

DOI: 10.1021/acs.organomet.7b00749

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

infra). It is worth noting that using 1 mol % of precatalyst

allowed quantitative and selective formation of the product n1

after only 6 h of reaction at 70 °C (see Table S2 in the

Supporting Information for more details). In addition, when

only 0.1 mol % of [3b][BArF₂₄] was used, ¹H NMR

spectroscopy analysis (C₆D₆, 298 K) after 24 h of reaction

showed total conversion of the starting material into n1 as the

major product (∼80% yield), along with an unknown

compound for which a characteristic singlet peak appeared at

δ 3.55 ppm (see the Supporting Information). The identity of

the latter compound is tentatively assigned to the amino

product derived from the desilylative hydrolysis of N−Si bonds

in n1.

Isolation and Structural Characterization of Adduct

[4b][BArF₂₄] (Step 1, Scheme 3). The procedure previously

adopted for the in solutio formation of adduct [4a][BArF₂₄]^19b

was applied for the attempted formation of [4b][BArF₂₄]. A

solution of [3b][BArF₂₄] in CD₂Cl₂placed in an NMR sample

tube equipped with a J. Young valve at −60 °C was treated with

1

4.4 equiv of HSiEt₃. The H NMR spectrum revealed the

presence of [4b]⁺, [7b]⁺, EtN(SiEt₃)₂, and unreacted HSiEt₃in

a respective 1:0.1:1.2:1.2 ratio. The hydridic signal of the Ir−H

motif of [4b]⁺was located at −11.4 ppm surrounded by weak

1

satellites assigned to a H−²⁹Si coupling constant of 20.0 Hz.

The Cp* ligand produced a signal at 1.69 ppm. The Et groups

of the SiEt₃moiety appeared at 0.56 ppm as a triplet (Si−

CH₂CH₃) and as a set of multiplets of doublets at 0.29 and 0.22

ppm (Si−CH₂CH₃) respectively. ²⁹Si NMR of the same sample

produced a signal at 6.92 ppm for the Ir-bound SiEt₃moiety.

2D ¹H−²⁹Si HMQC NMR analysis showed a clear cross-

correlation peak between signals of the Ir-bound hydride (¹H

−11.4 ppm) and the SiEt₃group (²⁹Si 6.92 ppm) (cf. Figure

S40 in the Supporting Information). Cation [7b]⁺was ﬁrmly

identiﬁed from its typical hydridic resonance at −22.6 ppm (see

the Experimental Section and the Supporting Information).

Isolation of a pure sample of [4b]⁺, which was attempted in

several instances, was eventually forsaken due to its high

reactivity. Crystal growth by diﬀusion of layered solvents in a

narrow sample tube was also a tricky challenge, for solutions of

preformed [4b][BArF₂₄] tended to decompose invariably to

give red crystals of [7b][BArF₂₄] at +4 °C. Crystals of

[4b][BArF₂₄] were successfully obtained by a modiﬁcation of

the crystal growth procedure. Adduct [4b][BArF₂₄] was

synthesized directly in a narrow so-called diﬀusion glass tube

at 4 °C from a solution of [3b][BArF₂₄] in 1,2-dichloroethane,

over which was carefully deposited a small layer of dry and

degassed benzene itself topped with a solution of excess HSiEt₃

in n-hexane. This procedure that used benzene as a solvent

diﬀusion buﬀer obviously had the advantage of retarding the

early formation of [7b][BArF₂₄] and ensuring the steady

growth of crystals of [4b][BArF₂₄] that were collected for X-ray

diﬀraction analysis.

Figure 6. (a) ORTEP-type diagram of cation [4b]⁺drawn at 30%

probability. The SQUEEZE procedure was applied to residual electron

density assigned in the lattice to n-hexane. For the sake of clarity, the

counteranion [BArF₂₄]⁻and the position-disordered C and H atoms

(50% occupation probability) of the SiEt₃moiety are omitted. The

position of the hydridic H1A atom was localized from a diﬀerence

Fourier map and reﬁned. Selected interatomic lengths (Å) and angles

(deg): C11−Ir1 2.079(6), N1−Ir1 2.106(5), Si1−Ir1 2.573(3), Ir1−

H1A 1.49, Si1−H1A 1.60, Si1−Ir1−H1A 35.0, N1−Ir1−Si1

81.65(17), N1−Ir1−H1A 47.3, C11−Ir1−N1 78.4(2). (b) ADFview³⁴

plots of noncovalent interaction (NCI) regions materialized by

reduced density gradient isosurfaces (cutoﬀ value s = 0.02 au, ρ =

0.05 au) colored according to the sign of the signed density λ₂ρ for the

gas-phase relaxed singlet ground state model of [4b]⁺: red and blue are

associated with negatively (attractive) and positively (repulsive or van

der Waals) signed terms, respectively. Selected interatomic lengths (Å)

and angles (deg) in the computed model of [4b]⁺: C_Ar−Ir 2.120, N−Ir

2.090, Si−Ir 2.503, Ir−H 1.583, Si−H 1.928, Si−Ir−H 39.2, N−Ir−Si

97.0, N−Ir−H 79.3, C_Ar−Ir−N 77.2.

The resolved structure of the latter salt displayed similarities

with that of [4a][BArF₂₄] (Figure 6a, see also the CIF ﬁles and

Table S10 in the Supporting Information for acquisition and

reﬁnement data).⁵³The hydridic hydrogen atom was localized

from a diﬀerence Fourier map, aﬀording Ir−H and H−Si

interatomic distances of ca. 1.49 and 1.60 Å, respectively. The

more accurate Ir−Si distance of 2.573(3) Å matches the value

already reported for [4a][BArF₂₄].^19b

gas-phase geometry slightly diﬀerent from that resolved by X-

ray diﬀraction analysis, displaying expectedly a slightly shorter

Si−Ir distance and longer C_Ar(C11)−Ir and N−Ir distances

(Figure 6b). It is worth noting that the Ir−H (1.58 Å) and Si−

H (1.93 Å) interatomic distances are signiﬁcantly diﬀerent from

those suggested by X-ray diﬀraction data. This expected

discrepancy stems not only from the inaccurate localization

Geometry optimization (cf. the Supporting Information for

computational details) at the ZORA-PBE0-dDsC³⁵(all-

electron triple-ζ single polarized basis set) level provided a

I

DOI: 10.1021/acs.organomet.7b00749

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

of the hydridic H atom from the Fourier map but also from the

weakness of the Ir−Si interaction that might allow slight

structural distortions as a response to strains caused by crystal

packing.

We ﬁrst attempted to produce this cation by the reaction of

TfO−SiEt₃with MeCN in CD₂Cl₂. Careful scrutiny of the H

1

NMR spectrum of the resulting solution did not reveal the

characteristic signal of the Si-bound acetonitrile.^42bFurther

addition of 5b to this solution did not produce any of the

products expected for the reduction of MeCN. Alternatively, we

opted for a more drastic procedure entailing the production of

a strongly electrophilic [SiEt₃]⁺moiety by the abstraction of the

hydride from HSiEt₃using the BArF₂₀tritylium salt: i.e.,

[Ph₃C][BArF₂₀] (Scheme 5).^32a,43In a ﬁrst experiment, 1 equiv

The strength of the donor−acceptor (Ir−H)→[SiEt₃]⁺

interaction^19b,22in the model of [4b]⁺was evaluated by

computing the intrinsic interaction energy ΔE_intassuming a

fragmentation of the molecule into the neutral “prepared”

fragment 5b and the “prepared” [SiEt₃]⁺moiety, the term

“prepared” meaning that the original geometry of each

fragment was kept as in cation [4b]⁺. ΔE_intin the latter

complex was found to amount to −119 kcal/mol, which is

exactly the same value as that found for [4a]⁺.^19b,22Further

investigation of the electron density topology using the

quantum theory of atoms in molecules^36,37(QTAIM) revealed

that two bond critical points (3,−1) and bond paths exist for

Ir−Si and the Ir−H (Figure 7).^19b,22This situation is akin to

Scheme 5. Electrophilic Activation of MeCN by [SiEt₃]⁺

Followed by Hydride Transfer from 5b

of the latter salt was mixed with ca. 4 equiv of MeCN dissolved

in dry CD₂Cl₂at 0 °C. The mixture was placed in an NMR

sample tube equipped with a J. Young valve, and ca. 2 equiv of

HSiEt₃was added. The ¹H NMR spectrum measured

immediately after homogenization of the solution revealed

the quantitative formation of Ph₃CH, the presence of unreacted

HSiEt₃and a minor amount of O-silylated isopropyl alcohol in

a 1:0.6:0.5 ratio, the latter probably originating from a

contamination of the NMR sample tube by acetone. It was

indeed shown by Oestreich et al.^32athat the tritylium/silane

combination could eﬃciently promote the catalytic reduction of

imines. Forcefully, we assume that the O-silated isopropyl

alcohol results from such a catalyzed reaction. Nonetheless, the

presence of a new species formulated as [(MeCN)_nSiEt₃]⁺(n ≥

1) could not be ﬁrmly established, for its expected signals in

Figure 7. Contour plot of the Laplacian of the density ∇²ρ(r)³⁶for the

singlet ground state geometry of [4b]⁺(ZORA-PBE0-dDsC³⁵/all

electron TZP level, dashed contours correspond to ∇²ρ(r) < 0)

determined in the Ir−Si−H plane and overlaid with a 3D

representation of the QTAIM analysis displaying bond critical points

(BCP, red dots), ring critical points (green dots), and bond paths

(colored by density). BCP (3,−1) #1: ρ = 0.0721 au, ∇²ρ = −0.0557

au. BCP(3,−1) #2: ρ = 0.1603 au, ∇²ρ = 0.0427 au.

1

²⁹Si and H NMR were not detected. However, a group of

signals at δ 2.13 (broad singlet, CH₃), 0.93 (t, 9H, CH₃−

CH₂Si), and 0.52 (q, 6H, CH₃−CH₂Si) ppm were assigned to

the methyl group of protonated acetonitrile and to the ethyl

groups of the SiEt₃moiety of a disiloxane that was already

reported in another study by Oro et al.⁴⁴To support this

assertion, ²⁹Si NMR of the same solution gave a new singlet at

8.92 ppm that can readily be assigned to O(SiEt₃)₂.⁴⁴It is

worth noting that a broad peak attributed to a solvated proton

was also observed at 11.6 ppm⁴⁵(Figure S73 in the Supporting

Information), along with a singlet peak assigned to dissolved H₂

at ca. 4.6 ppm (Figure S74 in the Supporting Information). The

origin of the latter signals and of the disiloxane can be explained

by the presence of residual water contamination in either of the

used reagents or solvent. This solution was subsequently

treated at 0 °C with a solution of 5b in CD₂Cl₂, and the sealed

NMR tube was submitted to NMR analyses. The latter clearly

revealed the quantitative consumption of 5b and its conversion

into [3b][BArF₂₄] and [7b][BArF₂₄] in a 1:1.9 ratio.

Furthermore, the typical signals at δ 3.65, 2.83, and 2.75 ppm

indicated the formation of EtN(SiEt₃)₂and coupling derivatives

(cf. the Experimental Section and Supporting Information). No

signals pertaining to the N-silyl imine MeCHNSiEt₃were

perceptible at this stage of the experiment, suggesting that the

second hydrosilylation step implying the formal reduction of

the N-silyliminium [MeCHN(SiEt₃)₂]⁺was much too fast to

allow detection of the imine. It is worth noting that the

that already reported for [4a]⁺, for which it was established that

the Si−H interaction was evanescent.^19bThe Wiberg bond

indices³⁸for the Ir−H (0.47), Ir−Si (0.38), and Si−H (0.23)

interactions are consistent with the results of the QTAIM

analysis, suggesting a weak Si−H interaction.

Hydride Transfer to the N-Silylnitrilium Intermediate

(Step 3, Scheme 3). We already established in a previous

report that the adduct [4a][BArF₂₄] was a potential source of

electrophilic silyl group^32b,39or “silylium”responsible for

the double hydrosilylation of the MeCN ligand in the

precatalyst activation step. It is therefore assumed that under

catalytic conditions (70 °C, large excess of ArCN) the [SiEt₃]⁺

moiety of either [4a]⁺or [4b]⁺could readily be captured by the

arylnitrile substrate to generate a transient electrophilic N-

silylnitrilium. In this set of experiments, the formation of the

putative N-silylnitrilium was attempted to verify the pertinence

of step 3 (Scheme 3) that entails the transfer of hydride from

5b^30,40to the nitrilium and the formation of a N-silylimine.

Such an organic cation, i.e. [RCN-SiEt₃]⁺, can be considered as

a solvated form of the silylium cation. According to the

literature,⁴¹[SiR₃]⁺cations can bind up to two nitrile donor

ligands, giving rise to the pentacoordinate Si species [RCN−

Si(R₃)−NCR]⁺.⁴²

J

DOI: 10.1021/acs.organomet.7b00749

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

presence of unreacted HSiEt₃left after the reaction with

[Ph₃C][BArF₂₀] and moreover the absence of any trace of

EtN(SiEt₃)₂before the addition of 5b suggest that the silane is

not a source of hydride of suﬃcient reactivity to compete with

5b in the ﬁrst hydrosilylation of the nitrile function.

The formation of the [(MeCN)_nSiEt₃]⁺adduct was

attempted de novo by using a [Ph₃C][BArF₂₀]:HSiEt₃ratio

1

of 1:0.7 at 0 °C in CD₂Cl₂. In this new experiment, H NMR

analysis of the resulting solution at −30 °C revealed a ﬁrst

group of signals at 0.82 (q, J = 8 Hz), 0.98 (t, J = 8 Hz) ppm

and at δ 1.05 (m) ppm in a rough 2:1 ratio that were assigned

to two diﬀerent [(MeCN)_nSiEt₃]⁺species (cf. Figure S78 and

S79 in the Supporting Information). These signals were

correlated by a HMQC experiment with two ²⁹Si NMR signals

at δ 45 and 42 ppm (respective ratio ca. 2:1, cf. Figure S81 in

the Supporting Information). Edlund, Cremer, et al.^42bstressed

that the chemical shift of ²⁹Si was strongly environment

sensitive, displaying a wide span of chemical shifts depending

on the number of Si-bound donor molecules and on the

solvents used in the NMR experiments.⁴⁶However, the

published ²⁹Si chemical shifts for triethylsilylnitrilium salts in

various solvents are often reported above δ +30 ppm.^42b,46The

Figure 8. ORTEP-type diagram of cation [7b]⁺drawn at 30%

probability. For the sake of clarity, the counteranion [BArF₂₄]⁻was

omitted. The hydridic H1A atom was localized from a diﬀerence

Fourier map and reﬁned. Selected interatomic lengths (Å) and angles

(deg): C11−Ir1 2.078(4), N1−Ir1 2.088(7), Ir1−H1A 1.86, Ir2−H1A

1.79, C34−Ir2 2.040(7), N2−Ir2 2.090(7), Ir2−H−Ir1, 131.5(1).

Cp*−Ir bonding is shown, for the sake of clarity, as a green bond

between the white spheres in the center of each cyclopentadienyl ring

and the Ir centers.

1

associated methyl H NMR signal of the Si-bound MeCN

molecule was located at δ 2.30 ppm.^46bAgain at around δ +11.4

1

ppm the H NMR spectrum suggested the presence of some

solvated proton species (solvated proton:[(MeCN)_nSiEt₃]⁺≈

1:08). To this solution was added a substoichiometric amount

1

of 5b (relative to HSiEt₃) at −30 °C. The H NMR spectrum

of the resulting solution revealed the full consumption of 5b,

the formation of [3b]⁺, and the absence of any of the signals

previously assigned to [(MeCN)_nSiEt₃]⁺(cf. Figure S80 in the

Supporting Information). New signals at δ 0.48 (m), 0.88 (t),

2.04 (broad s), and 4.8 (s) ppm could not be assigned with

certainty. We hypothesize that the latter signals could be related

to some form of H₂-based “encounter complex” putatively

engaged in an interaction with MeCN and [SiEt₃]⁺.⁴⁷No

acetonitrile hydrosilylation product was detected in this

experiment.

Scheme 6. Reactivity of 5a,b with Acetonitrile Salts

a

[3a,b][X]

Reversible Formation of [7]⁺(Step 5, Scheme 3). As

mentioned above, crystals of [7b][BArF₂₄] were obtained from

the ﬁrst failed attempts to grow crystals of [4b][BArF₂₄]. The

structure of the cation of the former salt is displayed in Figure

8. This μ-hydrido-bridged bis-iridium complex has the relative

R_Ir*,R_Ir* conﬁguration at each stereogenic Ir center (Figure 8).

Although from a crystallographic standpoint no C₂symmetry

axis formally exists, in solution such a symmetry element that

bisects perpendicularly the Ir1−Ir2 segment through atom H1A

1

is plausible, as suggested by the H NMR spectrum of [7b]⁺

that reveals a single set of aromatic proton signals (cf. the

Experimental Section).

a

1

Ratios determined by H NMR.

Owing to its reactivity and to the lack of stability in the dry

solid state, the quest for an analytically pure sample of

[7b][BArF₂₄] was not further pursued. However, several

experiments indicate that either [7a]⁺or [7b]⁺should be

better considered here as a resting state of the autotandem

double hydrosilylation reaction (Scheme 6 and Figures S82 −

S88 in the Supporting Information). It was found that the

reaction of a slight excess of [3a][BArF₂₄] (2.5 equiv) with 5a

at room temperature in CD₂Cl₂(J. Young NMR sample tube)

produced quantitatively [7a][BArF₂₄] and free acetonitrile in a

1:1 ratio. Addition of a large excess of MeCN (ca. 50 equiv)

converted ca. 50% of [7a][BArF₂₄] back into 5a (Scheme 6).

Similar results were obtained for the reaction of [3b][BArF₂₄]

with 5b (Scheme 6). To model closely the intervention of the

coordinatively unsaturated cation [6]⁺in the formation [7]⁺, a

solution of 5a was separately treated with 0.5 equiv of

[Ph₃C][BArF₂₀] used as a hydride abstraction agent capable of

releasing in situ the elusive [6a]⁺intermediate (eq 3). The

experiment carried out in a sealed NMR tube at 20 °C in d₅-

PhCl indicated the quantitative conversion of 5a into [7a]⁺

(Figure S89 in the Supporting Information).

Low-Temperature Hydrosilylation of MeCN. The

reversibility of the formation of the μ-hydrido dimer [7]⁺was

further conﬁrmed by carrying out the treatment of CD₃CN (10

K

DOI: 10.1021/acs.organomet.7b00749

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

Table 6. Hydrosilylation of Acetophenone at Room

Temperature

a

b

entry

precatalyst

yield (%)

TON (%)

1

2

3

4

5

[3a][BArF₂₀]

[3b][BArF₂₀]

[3a][BArF₂₄]

[3b][BArF₂₄]

[3c][BArF₂₄]

>99

970

equiv) with HSiEt₃(20 equiv) in the presence of nearly

catalytic amounts of [3a][BArF₂₄] (1 equiv) in CD₂Cl₂at −50

°C in a J. Young NMR sample tube. The reaction mixture that

a

1

Conditions: acetophenone (0.10 mL, 0.86 mmol), HSiEt₃(0.15 mL,

0.94 mmol), precatalyst (8.6 μmol, 0.1 mol %), CH₂Cl₂(0.5 mL),

room temperature, ratio acetophenone:silane:precatalyst =

1000:1000:1, 30 min. TON: turnover number, calculated at the

was left to react for 60 min was subsequently analyzed by H

NMR, revealing the formation of large amounts of CD₃CH₂N-

(SiEt₃)₂in addition to trace amounts of other organic

compounds such as protio and deuterio MeCN (CD₃CH₂N-

(SiEt₃)₂:MeCN ≈ 82:1). Among the organometallic species,

one could identify [7a]⁺and the new iridacyclic complex [8a]⁺

(Scheme 3, with L = CD₃CN) present in a 1:1.7 ratio on the

basis of the integration of their respective Cp* methyl signals.

Minor amounts of 5a were also perceptible. The same

experiment carried out at room temperature for 30 min

resulted in the complete conversion of CD₃CN into

CD₃CH₂N(SiEt₃)₂and trace amounts of a mixture of organic

side products displaying ²H NMR signals at 2.09 and 1.09 ppm.

Compound 5a was the only organometallic species identiﬁed in

b

highest conversion as the molar ratio of product to catalyst.

Table 7. Eﬀect of the Nature of the Counteranion of [3b]⁺

on the Double Hydrosilylation of Benzonitrile with HSiEt₃

a

entry

precatalyst

yield (%)

1

2

3

[3b][BArF₂₄]

[3b][OTf]

>99

5

[3b][BPh₄]

1

a

All reactions were conducted at 70 °C: benzonitrile (0.1 mL, 0.97

mmol), HSiEt₃(0.35 mL, 2.1 mmol), precatalyst (9.7 μmol, 1 mol %),

CH₂Cl₂(0.5 mL).

1

the medium by H NMR spectroscopy.

Hydrosilylation of Benzonitrile and Acetophenone.

BArF salts of [3a−c]⁺were further evaluated for the

hydrosilylation of acetophenone and benzonitrile, for which

reaction conditions were optimized. In all cases the hydro-

silylation of acetophenone was completed within 30 min at

room temperature using HSiEt₃. In turn, the exhaustive

hydrosilylation of benzonitrile required an optimal temperature

of 70 °C to produce almost quantitatively in 24 h the

corresponding N,N-bis(triethylsilyl)benzylamine. The amount

of precatalyst in all cases was lower than 1 mol % relative to

acetophenone and benzonitrile. Tables 5 and 6 give the

untouched. The low performance of [3b][OTf] is putatively

assigned to the formation of Et₃SiOTf, which might not display

a suﬃcient electrophilicity toward PhCN to produce eﬃciently

the putative [Et₃Si(PhCN)_n]⁺key intermediate. It is hypothe-

sized that a lack of stability in solution might explain the low

performance of [3b][BPh₄] (vide supra).

Hydrosilylation of Nitriles Catalyzed by [3b][BArF₂₄]:

Scope. The scope of the catalysis of the double hydrosilylation

of nitriles was explored using exclusively arylnitriles containing

either electron-withdrawing or -donating substituents as well as

potential donor ligands susceptible of challenging the catalysis

by catalyst poisoning (Scheme 7, products n1−n14). Scheme 7

displays the targets that were pursued under the standardized

conditions deﬁned for benzonitrile: i.e., 70 °C and 0.5 mol % of

catalyst relative to the starting arylnitrile. Almost all nitrile

substrates were converted into the corresponding N,N-

disilylamines in yields spanning 86−99% (n1−n5, n7−n11,

n13, and n14) without the formation of any side product. For

instance the arylnitrile substrate related to n6, i.e. 4-amino-

benzonitrile, was left untouched with no N-silylation of the

amino group or any sign of monohydrosilylation of the nitrile

Table 5. Hydrosilylation of Benzonitrile at 70 °C

a

b

entry

precatalyst

yield (%)

TON (%)

1

2

3

4

5

[3a][BArF₂₀]

[3b][BArF₂₀]

[3a][BArF₂₄]

[3b][BArF₂₄]

[3c][BArF₂₄]

93

>99

188

200

a

Conditions: benzonitrile (0.10 mL, 0.97 mmol), Et₃SiH (0.35 mL, 2.2

mmol), precatalyst (4.8 μmol, ∼0.5 mol % relative to benzonitrile),

CH₂Cl₂(0.5 mL), 70 °C, molar ratio nitrile:silane:precatalyst =

1

being detected in the H NMR spectrum of the crude reaction

b

202:458:1, 24 h. TON: turnover number, calculated at the highest

mixture. It is speculated that 4-aminobenzonitrile inhibits the

catalysis by binding the only coordination site available at the Ir

center of [3b]⁺for the activation of the silane.

conversion as the molar ratio of product to catalyst.

respective TONs for each type of reaction. The values for the

hydrosilylation of acetophenone are roughly 4.5 times greater

than those of the double hydrosilylation of benzonitrile for

nearly total conversions of the substrates into the respective

silyl ether and N,N-disilylamine. Quite interestingly all BArF

salts promoted the double hydrosilylation of benzonitrile

equally at 70 °C, producing TONs of ca. 200 when a 1:2

benzonitrile:silane ratio was used. It is worth noting, however,

that other salts, such as [3b][OTf] and [3b][BPh₄], did not

perform well at all in the hydrosilylation of benzonitrile (Table

7, entries 2 and 3). Experiments run under identical conditions

did not yield more than 5% of the double-hydrosilylation

product, the starting material being recovered mostly

Such catalyst poisoning is most certainly responsible also for

the lack of conversion of 3-cyanopyridine into n12, the pyridyl

moiety diverting greater amounts of Ir complex into a

catalytically inert form. It is interesting to note that 4-

hydroxybenzonitrile was eﬃciently converted into n11 in a

process involving the joint dehydro O-silylation of the phenolic

hydroxyl substrituent and the double hydrosilylation of the

nitrile function. Similarly 4-acetylbenzonitrile was nonselec-

tively converted into n13 by the joint hydrosilylation of the

carbonyl and nitrile functions. However, side dehydrogenation

occurring at the ketyl function was also observed under certain

conditions (Figure S125 in the Supporting Information). It is

worth noting that n14, a potential precursor of a pincer ligand,

L

DOI: 10.1021/acs.organomet.7b00749

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

Scheme 7. Autotandem Double Hydrosilylation of

a

Arylnitriles: Scope

Another experiment was performed with a 1:1:1 mixture of

acetophenone, benzonitrile, and HSiEt₃at room temperature

with ca. 0.5 mol % of [3b][BArF₂₄] (eq 5). The reaction, which

was monitored by removing aliquots of the reaction medium

1

for H NMR spectroscopic investigations at diﬀerent reaction

times, revealed that the hydrosilylation reaction was predom-

inant with benzaldehyde, producing a yield of about 53% after

only 15 min. The minor improvement in the yield after 60 min

(64% yield) of reaction suggested that benzonitrile might be

inhibiting the catalysis and that slightly forcing experimental

conditions would be required to achieve its hydrosilylation.

The potential of [3b][BArF₂₄] in the hydrosilylation of

ketones and aldehydes was also investigated on a series of

protypical aromatic and aliphatic substrates. The corresponding

results are available in Schemes S4 and S5 in the Supporting

Information. These results show that the hydrosilylation of

carbonyls operates with precatalyst loads of 0.05−2 mol %,

producing the resulting triethylsilyl ethers in yields >90% in

most cases, this without the use of any additive.⁴⁸

CONCLUSION

■

Far from being poisoned by nitriles, the cationic Ir(III)

metallacycles used in this study perform rather well as catalysts

for the autotandem double hydrosilylation of acetonitrile,

propionitrile, and a variety of other arylnitriles. In dilute

solution, the cationic precatalysts [3a,b]⁺undergo an energeti-

cally unfavorable reaction with HSiEt₃, leading to adducts

[4a,b]⁺by a speciﬁc concerted hydride transfer reaction and

capture of the electrophilic [SiEt₃]⁺moiety by the newly

formed hydrido-iridium species: the released MeCN molecules

are converted into the N,N-disilylated ethylamine. Under

solventless conditions, i.e. where the concentration of nitrile

is overwhelming in comparison to that of the Ir catalyst, the

binding of [SiEt₃]⁺to the Ir is most certainly swiftly quenched

by reaction with the nitrile substrate, giving rise to highly

electrophilic forms of reactive nitrilium cations that may hence

undergo the nucleophilic transfer of hydride from transient

intermediate 5a,b. Whether the transfer of hydrido ligand

occurs once the silylnitrilium is released in the solution bulk or

in a somewhat concerted manner upon nucleophilic attack of

the Ir-bound silyl moiety by MeCN remains unsettled.

Experimental evidence of the noninterference of HSiEt₃as a

potential competing source of hydride in the reduction of

nitriles also outlines the central role of the iridium-hydrido

species 5a,b as the main source of nucleophilic hydride in this

catalytic process indeed. However, owing to the experimental

diﬃculties in observing unambiguously the conversion of

[(MeCN)_nSiEt₃]⁺into the expected products of hydro-

silylation, further investigations seem mandatory both exper-

imentally and theoretically. For the sake of conciseness,

theoretical support of the mechanism proposed in Scheme 3

in the form of a full Gibbs catalysis energy proﬁle was not

included to the present paper and will be disclosed elsewhere.

Species [7b]⁺, i.e. the analogue of [7a]⁺, which was considered

a

Conditions: (a) nitrile (1.1 mmol), HSiEt₃(0.35 mL, 2.1 mmol),

[3b][BArF₂₄] (6.8 mg, 4.8 mmol [0.5 mol %] or 13.7 mg, 9.7 mmol [1

mol %]), 70 °C; (b) arylnitrile (0.8 mmol), HSiEt₃(0.3 mL, 1.8

mmol), [3b][BArF₂₄] ([0.5 mol %] or [1 mol %]), 70 °C; (c)

conditions similar to those in (b), except that instead of 2.2 equiv of

HSiEt₃, 4.4 equiv (0.60 mL, 3.7 mmol) was used; (d) conditions

similar to those in (b), except that instead of 2.2 equiv of HSiEt₃, 8.8

equiv (1.20 mL, 7.5 mmol) was used. Yields were determined by H

NMR spectroscopy using 1,3,5-tri-tert-butylbenzene as internal

reference.

1

was also prepared in high yield by the double hydrosilylation of

two nitrile groups.

Investigation of Catalyst Response to Competitive

Hydrosilylation Conditions. We gauged the catalytic

potential of [3b][BArF₂₄] and its reactivity toward aldehydes,

ketones, and nitriles. The latter precatalyst was placed under

competitive reaction conditions consisting of equimolar

mixtures of diﬀerent substrates and HSiEt₃. These sets of

experiments were expected to produce a qualitative preferential

order of reactivity. A ﬁrst experiment consisting of a reaction of

a 1:1:1 mixture of benzaldehyde, acetophenone, and HSiEt₃

was carried out at room temperature in CH₂Cl₂in the presence

of 0.1 mol % of [3b][BArF₂₄] (eq 4). The reaction produced

predominantly the benzyl silyl ether (61% yield determined by

¹H NMR using 1,3,5-tri-tert-butylbenzene as internal refer-

ence), suggesting a net preference of the catalytic pot for the

hydrosilylation of the aldehyde.

M

DOI: 10.1021/acs.organomet.7b00749

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

performed with Thermo Scientiﬁc FLASH 2000 CHNS/O analyzers.

Refer to the Supporting Information for computational details and

information on structural X-ray diﬀraction analyses.

as a catalytically inactive diversion of valuable iridacycle in the

room-temperature dehydro O-silylation of alcohols, should be

considered as a resting state under the conditions of the present

nitriles’ double-hydrosilylation catalysis. The main feature of

this catalytic system remains the peculiar mode of formation of

adducts [4a,b]⁺, the consistent formulation of which as a (Ir−

H)→[SiEt₃]⁺Ir(III) donor−acceptor complex opens upa new

perspective for iridacycles in chemical processes where the

elusive highly electrophilic [SiR₃]⁺moiety is central.^32,39,41a,49

Recent theoretical investigations tend to suggest that similar

bonding situations may well exist in Brookhart’s POCOP-Ir

catalyst as well as in a number of other situations depicted by

Nikonov et al. and Oestreich et al. for example.²²The potential

of this class of iridacycles as multifaceted Lewis acids capable of

not only eﬃciently binding conventional donor ligands but also

activating Si−H bonds justiﬁes our quest for “multi-competent”

catalysts^19acapable of promoting complex autotandem type

chemical transformations. As iridacycles containing the Cp*Ir

moiety are rather easy to synthesize and derive into solvato

MeCN cations, their inherent chirality at Ir raises the question

of their potential in asymmetric catalysis, which motivates

current research that will be disclosed in due time.

Procedure for the Synthesis of [Cp*Ir(DB[f,h]Q)Cl] (2c)

(Scheme 2). A Schlenk ﬂask was loaded with the ligand 1c (150

mg, 0.65 mmol), the iridium precursor [Ir(μ-Cl)ClCp*]₂(287 mg,

0.36 mmol), and anhydrous NaOAc (160 mg, 1.95 mmol). To this

mixture was added ∼15 mL of CH₂Cl₂. After vigorous stirring at 50 °C

for 48 h, the reaction mixture was ﬁltered through Dicalite and the

ﬁltrate stripped of solvent under reduced pressure. The resulting

residue was then washed with n-pentane. The product 2c was

recovered as a yellow-orange powder (231 mg, 60% yield). Anal. Calcd

for C₂₇H₂₅ClIrN·1/10CH₂Cl₂: C, 54.57; H, 4.25; N, 2.37. Found: C,

1

54.76; H, 4.43; N, 2.37. H NMR (500 MHz, 298 K, CDCl₃): δ 8.97

(d, 1H, J = 5.3 Hz, H−CN DB[f,h]Q), 8.78 (d, 1H, J = 8.4 Hz, H_Ar

DB[f,h]Q), 8.64 (d, 1H, J = 8.4 Hz, H_ArDB[f,h]Q), 8.54 (d, 1H, J =

8.4 Hz, H_ArDB[f,h]Q), 8.16 (d, 1H, J = 8.0 Hz, H_ArDB[f,h]Q), 8.05

(d, 1H, J = 7.2 Hz, H_ArDB[f,h]Q), 7.72 (t, 1H, J = 7.2 Hz, H_Ar

DB[f,h]Q), 7.62−7.67 (m, 2H, H_ArDB[f,h]Q), 7.52 (d, 1H, J = 8.1

Hz, H_ArDB[f,h]Q), 7.50 (d, 1H, J = 8.1 Hz, H_ArDB[f,h]Q), 1.75 (s,

15H, Cp-Me₅). ¹³C NMR (126 MHz, 298 K, CDCl₃): δ 161.9

(DB[f,h]Q), 158.4 (DB[f,h]Q), 149.8 (H−CN, DB[f,h]Q), 140.3

(DB[f,h]Q), 133.5 (C−H, DB[f,h]Q), 132.1 (DB[f,h]Q), 132.0

(DB[f,h]Q), 131.2 (C−H, DB[f,h]Q), 128.5 (C−H, DB[f,h]Q),

128.1 (DB[f,h]Q), 127.1 (C−H, DB[f,h]Q), 126.1 (DB[f,h]Q),

124.1 (C−H, DB[f,h]Q), 123.4 (C−H, DB[f,h]Q), 122.0 (C−H,

DB[f,h]Q), 115.7 (C−H, DB[f,h]Q), 88.7 (Cp-Me₅), 8.9 (Cp-Me₅).

HRMS-ESI: calcd for C₂₇H₂₅ClIrN ([2c]^•+) (m/z) 591.1292, found

591.1320. HRMS-ESI: calcd for C₂₇H₂₅IrN ([2c − Cl]⁺) (m/z)

556.1612, found 556.1638.

General Procedure for the Synthesis of [3a−c][X] (Scheme

2). In a Schlenk ﬂask an equimolar mixture of 2a−c and [MX] (MX =

K[BArF₂₀], Na[BArF₂₄], Na[BPh₄], or Ag[OTf]) was dissolved in

acetonitrile (∼7−10 mL). The resulting mixture was vigorously stirred

at room temperature for a period not exceeding 4 h, unless otherwise

stated. The resulting suspension was ﬁltered through a pad of Dicalite,

and the solvent was removed from the ﬁltrate under reduced pressure.

The solid was either recrystallized in a mixture of CH₂Cl₂and n-

hexane or n-pentane or washed with n-hexane or n-pentane, to aﬀord

an analytically pure compound.

EXPERIMENTAL SECTION

■

General Considerations. All experiments were conducted under a

dry argon atmosphere using standard Schlenk and glovebox

techniques. All glassware was oven-dried prior to use. All solvents

were distilled over sodium or CaH₂under argon before use.

Deuterated solvents were dried over sodium or CaH₂, ﬁltered over

activated neutral alumina, and stored under argon before use. Ligands

1a−c, benzyl alcohol, carbonyls, and nitriles were purchased from

Sigma-Aldrich, Alfa Aesar, TCI, and Fluka. IrCl₃was purchased from

Pressure Chemical Co. Dicalite 4158 was purchased from Carlo Erba

Reagents. Ligands 1b,c were used as received, while ligand 1a was

puriﬁed by bulb-to-bulb distillation over a minimal amount of

molecular sieves (MS-3 Å) and stored over MS-3 Å in a Schlenk

tube under argon. Benzyl alcohol was puriﬁed by bulb-to-bulb

distillation over a minimal amount of CaH₂and stored over MS-3 Å

under argon. Nitriles were used as received, whereas liquid carbonyls

were puriﬁed by bulb-to-bulb distillation over a minimal amount of

CaH₂and stored under argon before use. Ligand name abbreviations:

Cp*, 1,2,3,4,5-pentamethyl-η⁵-cyclopentadienyl; PhPy, 2-(2-pyridyl)-

phenyl, B[h]Q, benzo[h]quinolinyl; DB[f,h]Q, dibenzo[f,h]quinolinyl.

[Ir(μ-Cl)ClCp*]₂was synthesized from IrCl₃and 1,2,3,4,5-pentam-

ethylcyclopenta-1,3-diene following the procedure reported in the

literature.^25bCompounds 2a,b were prepared following the procedure

reported in the literature.^24a,50Compounds 5a,b were prepared

according to the procedure reported by Templeton et al.,^31band their

NMR data were found to be similar to those already reported by

Norton et al.³⁰[3a][BArF₂₄] was prepared following a published

procedure.^19bThe so-called trityl salt [Ph₃C][BArF₂₀] was prepared

according to the modiﬁed procedure (originally reported by Ozerov et

al.⁵¹) described by Heinekey et al.⁵²¹H (300, 400, 500, and 600

MHz), ¹¹B (128 MHz), ¹³C (75 and 126 MHz), ¹⁹F (282 MHz), and

²⁹Si (119 MHz) NMR spectra were measured on Bruker DPX 300 and

400, Avance I 500, and Avance III 600 spectrometers. All chemical

[Cp*Ir(PhPy)(NCMe)][BArF₂₀] ([3a][BArF₂₀]) (Scheme 2). This

compound was prepared following the general procedure described

above: 2a (300 mg, 0.580 mmol), K[BArF₂₀] (138.6 mg, 0.193 mmol),

MeCN (7 mL), room temperature, 2 h. The raw material was puriﬁed

by recrystallization with a mixture of CH₂Cl₂and n-pentane, and the

resulting salt was washed with n-pentane and dried under reduced

pressure. [3a][BArF₂₀] was recovered as a yellow powder in 90% yield

(629 mg). Single crystals were grown by slow diﬀusion of hexane into

a solution of [3a][BArF₂₀] in 1,2-dichloroethane (−30 °C). Anal.

Calcd for C₄₇H₂₆BF₂₀IrN₂·CH₂Cl₂: C, 44.81; H, 2.19; N, 2.18. Found:

1

C, 45.04; H, 2.08; N, 2.25. H NMR (500 MHz, 295 K, CD₂Cl₂): δ

8.62 (d, 1H, J = 5.7 Hz, H-CN PhPy), 7.94 (d, 1H, J = 8.0 Hz, H_Ar

PhPy), 7.88 (td, 1H, J₁= 7.8 Hz, J₂= 1.4 Hz, H_ArPhPy), 7.75 (t, 2H, J

= 7.9 Hz, H_ArPhPy), 7.29 (td, 1H, J₁= 7.5 Hz, J₂= 1.4 Hz, H_ArPhPy),

7.25 (td, 1H, J₁= 6.6 Hz, J₂= 1.4 Hz, H_ArPhPy), 7.20 (td, 1H, J = 7.5

Hz, J₂= 1.2 Hz, H_ArPhPy), 2.27 (s, 3H, Ir-NCMe), 1.68 (s, 15H, Cp-

Me₅). ¹³C NMR (126 MHz, 295 K, CD₂Cl₂): δ 168.0 (PhPy), 156.6

(PhPy), 151.9 (H-CN, PhPy), 148.5 (C-F, d (br), ¹J = 240.0 Hz, 2C

of C₆F₅in BArF₂₀), 145.1 (PhPy), 139.7 (C−H, PhPy), 137.6 (C-F, m,

1C of p-C₆F₅in BArF₂₀), 136.0 (C−H, PhPy), 135.7 (C-F, m, ¹J = 24.0

Hz, 2C of C₆F₅in BArF₂₀), 132.1 (C−H, PhPy), 124.9 (C−H, PhPy),

124.5 (C−H, PhPy), 124.0 (C−H, PhPy), 120.3 (C−H, PhPy), 118.4

(Ir−NCMe), 91.8 (Cp-Me₅), 8.9 (Cp-Me₅), 4.3 (Ir-NCMe), [the

1:1:1:1 quartet characterizing the C-B bond in C₆F₅(BArF₂₀) escaped

detection]. ¹⁹F NMR (282 MHz, 298 K, CD₂Cl₂): δ −133.50 (m, 8F,

o-C₆F₅in BArF₂₀), −164.04 (t, 4F, ³J_F−F= 20.7 Hz, p-C₆F₅in BArF₂₀),

−167.82 (m, 8F, m-C₆F₅in BArF₂₀). ¹¹B NMR (128 MHz, 298 K,

CD₂Cl₂): δ −16.7 (s). HRMS-ESI: calcd for C₂₁H₂₃IrN ([3a −

NCMe]⁺) (m/z) 482.1455, found 482.1495.

1

shifts (δ) are expressed in parts per million (ppm). For H and ¹³C

NMR, values of δ are reported relative to Me₄Si as an external

reference standard and referenced against peaks of solvents (only

1

partially deuterated solvent is visible in H) as secondary reference

standards. For other nuclei, the reported values of δ are referenced to

external reference standards (CF₃C₆H₅in CDCl₃for ¹⁹F, Me₄Si in

CDCl₃for ²⁹Si, and NaBH₄in D₂O for ¹¹B). Abbreviations used for

NMR data: s, singlet; d, doublet; t, triplet; q, quartet; qt, quintet; sept,

septet; m, multiplet; br, broad; J, coupling constant; Ar, aromatic. Mass

spectra were run on a MicroTOF Bruker Daltonics spectrometer,

using a TOF-ESI coupling analysis system. Elemental analyses were

N

DOI: 10.1021/acs.organomet.7b00749

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

[Cp*Ir(B[h]Q)(NCMe)][BArF₂₄] ([3b][BArF₂₄]) (Scheme 2). This

compound was prepared following the general procedure described

above: 2b (150 mg, 0.277 mmol), Na[BArF₂₄] (245.7 mg, 0.277

mmol), MeCN (7 mL), room temperature, 3 h. After recrystallization

with a mixture of CH₂Cl₂and n-pentane the product was washed with

n-pentane and the resulting yellow powder of [3b][BArF₂₄] was

isolated in 93% yield (363 mg). Single crystals were grown by slow

diﬀusion of a mixture of hexane and benzene into a solution of

[3b][BArF₂₄] in 1,2-dichloroethane. Anal. Calcd for C₅₇H₃₈BF₂₄IrN₂·

1/10CH₂Cl₂: C, 48.35; H, 2.71; N, 1.98. Found: C, 48.56; H, 2.72; N,

CDCl₃): δ 156.8 (B[h]Q), 154.4 (B[h]Q), 151.6 (C−H, B[h]Q),

142.1 (B[h]Q), 137.7 (C−H, B[h]Q), 134.2 (B[h]Q), 132.7 (C−H,

B[h]Q), 130.7 (C−H, B[h]Q), 129.9 (C−H, B[h]Q), 127.3 (B[h]Q),

123.9 (C−H, B[h]Q), 123.3 (C−H, B[h]Q), 121.9 (C−H, B[h]Q),

119.0 (Ir−NCMe), 91.1 (Cp-Me₅), 9.0 (Cp-Me₅), 4.1 (Ir-NCMe).

¹⁹F{¹H} (282 MHz, 298 K, CDCl₃): δ −78.9 (CF₃, OTf). HRMS-ESI

(m/z): calcd for C₂₅H₂₆IrN₂([3b]⁺) 547.1721, found 547.1655.

HRMS-ESI (m/z): calcd for CF₃O₃S ([OTf]⁻) 148.9515, found

148.9530.

[Cp*Ir(B[h]Q)(NCMe)][BArF₂₀] ([3b][BArF₂₀]) (Scheme 2). This

compound was prepared following the general procedure described

above: 2b (200 mg, 0.370 mmol), K[BArF₂₀] (264 mg, 0.368 mmol),

MeCN (7 mL), room temperature, 2 h. After recrystallization from a

mixture of CH₂Cl₂and n-pentane the resulting powder was washed

with n-pentane and dried under reduced pressure to aﬀord

[3b][BArF₂₀] as a yellow powder in 86% yield (387 mg). Single

crystals were grown by slow diﬀusion of hexane into a solution of

[3b][BArF₂₀] in dichloromethane (−30 °C). Anal. Calcd for

C₄₉H₂₆BF₂₀IrN₂·CH₂Cl₂: C, 45.82; H, 2.15; N, 2.14. Found: C,

45.96; H, 1.99; N, 2.38. ¹H NMR (400 MHz, 298 K, CD₂Cl₂): δ 8.91

(dd, 1H, J₁= 5.4 Hz, J₂= 1.3 Hz, H−CN B[h]Q), 8.39 (dd, 1H, J₁=

8.0 Hz, J₂= 1.4 Hz, H_ArB[h]Q), 8.02 (dd, 1H, J₁= 6.9 Hz, J₂= 1.3 Hz,

H_ArB[h]Q), 7.93 (d, 1H, J = 8.8 Hz, H_ArB[h]Q), 7.74 (dd, 1H, J₁=

7.9 Hz, J₂= 1.2 Hz, H_ArB[h]Q), 7.72−7.66 (m, 2H, H_ArB[h]Q), 7.63

(dd, 1H, J₁= 8.1 Hz, J₂= 5.3 Hz, H_ArB[h]Q), 2.15 (s, 3H, Ir-NCMe),

1.75 ppm (s, 15H, Cp-Me₅). ¹³C NMR (126 MHz, 295 K, CD₂Cl₂): δ

158.0 (C-N, B[h]Q), 154.7 (B[h]Q), 150.6 (H-CN, B[h]Q), 148.9

1

1.99. H NMR (500 MHz, 298 K, CD₂Cl₂): δ 8.89 (dd, 1H, J₁= 5.3

Hz, J₂= 1.2 Hz, H-CN B[h]Q), 8.36 (dd, 1H, J₁= 8.0 Hz, J₂= 1.1

Hz, H_ArB[h]Q), 8.01 (dd, 1H, J₁= 6.9 Hz, J₂= 0.9 Hz, H_ArB[h]Q),

7.92 (d, 1H, J = 8.7 Hz, H_ArB[h]Q), 7.73 (m, 9H, 1H_Arin B[h]Q +

8H_orthoin BArF₂₄), 7.67−7.70 (m, 2H, H_ArB[h]Q), 7.56 (m, 4H_parain

BArF₂₄), 7.61 (dd, 1H, J₁= 7.9 Hz, J₂= 5.3 Hz, H_ArB[h]Q), 2.12 (s,

3H, Ir-NCMe), 1.73 (s, 15H, Cp-Me₅). ¹³C NMR (126 MHz, 298 K,

1

CD₂Cl₂): δ 161.5−162.7 (1:1:1:1 quartet, J_C−B= 49.8 Hz, C−B in

BArF₂₄), 157.6 (C-N, B[h]Q), 154.2 (CB[h]Q), 150.2 (H-CN,

B[h]Q), 142.3 (B[h]Q), 138.4 (C−H, B[h]Q), 135.2 (m, C−H_orthoin

BArF₂₄), 134.8 (B[h]Q), 133.1 (C−H, B[h]Q), 131.4 (C−H, B[h]Q),

2

3

130.6 (C−H, B[h]Q), 128.9−129.6 (q, J_C−F= 31.5 Hz, J_C−B= 3.0

Hz, C-CF₃in BArF₂₄), 128.1 (B[h]Q), 125.0 (q, ¹J_C−F= 272.5 Hz, CF₃

in BArF₂₄), 124.1 (C−H, B[h]Q), 122.9 (C−H, B[h]Q), 122.5 (C−H,

B[h]Q), 118.5 (Ir−NCMe), 117.9 (m, C−H_parain BArF₂₄), 91.6 (Cp-

Me₅), 9.1 (Cp-Me₅), 4.2 (Ir-NCMe). ¹⁹F NMR (282 MHz, 298 K,

CD₂Cl₂): δ −63.8 (s, CF₃in BArF₂₄). ¹¹B NMR (128 MHz, 298 K,

CD₂Cl₂): δ −6.6 (bs). HRMS-ESI: calcd for C₂₅H₂₆IrN₂([3b]⁺) (m/

z) 547.1721, found 547.1716. HRMS-ESI: calcd for C₃₂H₁₂BF₂₄

([BArF₂₄]⁻) (m/z) 863.0649, found 863.0656.

1

(C-F, d (br), J = 240.0 Hz, 2C of C₆F₅in BArF₂₀), 142.7 (B[h]Q),

138.8 (C-F, d (br), ¹J = 240.0 Hz, 1C of p-C₆F₅in BArF₂₀), 138.8 (C−

H, B[h]Q), 136.2 (C−F, d (br), ¹J = 240.0 Hz, 2C of C₆F₅in BArF₂₀),

135.3 (B[h]Q), 133.6 (C−H, B[h]Q), 131.8 (C−H, B[h]Q), 131.0

(C−H, B[h]Q), 128.5 (B[h]Q), 124.5 (C−H, B[h]Q), 123.3 (C−H,

B[h]Q), 122.9 (C−H, B[h]Q), 119.0 (Ir−NCMe), 92.1 (Cp-Me₅), 9.6

(Cp-Me₅), 4.6 (Ir−NCMe), [the 1:1:1:1 quartet characterizing the C-B

bond in C₆F₅(BArF₂₀) escaped detection]. ¹¹B NMR (128 MHz, 298

K, CD₂Cl₂): δ −16.6 (s). HRMS-ESI: calcd for C₂₃H₂₃IrN ([3b -

NCMe]⁺) (m/z) 506.1455, found 506.1462.

[Cp*Ir(B[h]Q)(NCMe)][BPh₄] ([3b][BPh₄]) (Scheme 2). This com-

pound was prepared following the general procedure described above:

2b (200 mg, 0.37 mmol), Na[BPh₄] (126 mg, 0.37 mmol), MeCN (10

mL), 40 °C, 24 h. After recrystallization from a mixture of CH₂Cl₂and

n-pentane the resulting powder was washed with n-pentane and dried

under reduced pressure to aﬀord [3b][BPh₄] as a yellow powder in

78% yield (244 mg). Single crystals were grown by slow diﬀusion of

hexane and benzene into a solution of [3b][BPh₄] in 1,2-dichloro-

ethane. Anal. Calcd for C₄₉H₄₆BIrN₂·1/3CH₂Cl₂: C, 66.26; H, 5.26; N,

3.13. Found: C, 66.49; H, 5.24; N, 2.96. ¹H (500 MHz, 298 K,

CD₂Cl₂): δ 8.87 (dd, 1H, J₁= 5.3 Hz, J₂= 1.2 Hz, H_ArB[h]Q), 8.31

(dd, 1H, J₁= 8.0 Hz, J₂= 1.1 Hz, H_ArB[h]Q), 8.01 (dd, 1H, J₁= 7.1

Hz, J₂= 0.8 Hz, H_ArB[h]Q), 7.93 (d, 1H, J = 8.8 Hz, H_ArB[h]Q), 7.74

(d, 1H, J = 7.8 Hz, H_ArB[h]Q), 7.69 (t, 2H, J = 7.0 Hz, H_ArB[h]Q),

7.26 (m, 9H, 1H_Arin B[h]Q + 8H_Arin BPh₄), 6.95 (t, 8H, J = 7.2 Hz,

H_Arin BPh₄), 6.82 (m, 4H, H_Arin BPh₄), 1.74 (s, 3H, Ir-NCMe), 1.59

(s, 15H, Cp-Me₅). ¹³C NMR (126 MHz, 298 K, CD₂Cl₂): δ 164.5

[Cp*Ir(DB[f,h]Q)(NCMe)][BArF₂₄] ([3c][BArF₂₄]) (Scheme 2). This

compound was prepared following the general procedure described

above: 2c (100 mg, 0.169 mmol), Na[BArF₂₄] (150.0 mg, 0.169

mmol), MeCN (10 mL), 50 °C, 5 h. After recrystallization with a

mixture of CH₂Cl₂and n-pentane (∼−30 °C) the resulting powder

was washed with n-pentane and dried under reduced pressure to aﬀord

[3c][BArF₂₄] as an orange-yellow powder in 70% yield (172 mg).

Anal. Calcd for C₆₁H₄₀BF₂₄IrN₂·4/5CH₂Cl₂: C, 48.58; H, 2.74; N,

1

1.83. Found: C, 48.22; H, 2.70; N, 1.59. H (500 MHz, 298 K, d₆-

acetone): δ 9.33 (d, 1H, J = 8.3 Hz, H_ArDB[f,h]Q), 9.28 (d, 1H, J =

5.5 Hz, H_ArDB[f,h]Q), 8.86 (t, 2H, J = 7.6 Hz, H_ArDB[f,h]Q), 8.46

(d, 1H, J = 8.0 Hz, H_ArDB[f,h]Q), 8.13 (d, 1H, J = 7.3 Hz, H_Ar

DB[f,h]Q), 7.85−7.91 (m, 2H, H_ArDB[f,h]Q), 7.79 (m; 1H_Arin

DB[f,h]Q + 8H_orthoin BArF₂₄), 7.73 (t, 1H, J = 7.5 Hz, H_ArDB[f,h]Q),

7.68 (m, 4H_parain BArF₂₄), 2.41 (s, 3H, Ir-NCMe), 1.86 (s, 15H, Cp-

Me₅). ¹³C NMR (126 MHz, 298 K, d₆-acetone): δ 162.6 (1:1:1:1

1

(1:1:1:1 quartet, J_C−B= 49.2 Hz, C−B in BPh₄), 157.5 (B[h]Q),

154.3 (B[h]Q), 150.2 (H-C=N, B[h]Q), 142.3 (B[h]Q), 138.5 (H-C,

3

B[h]Q), 136.2 (1:1:1:1 quartet, J_C−B= 1.4 Hz, C−H_orthoin BPh₄),

134.8 (B[h]Q), 133.1 (C−H, B[h]Q), 131.4 (C−H, B[h]Q), 130.6

(C−H, B[h]Q), 128.7 (B[h]Q), 126.0 (1:1:1:1 quartet, ²J_C−B= 2.8 Hz,

C−H_metain BPh₄), 124.2 (C−H, B[h]Q), 123.1 (C−H, B[h]Q), 122.5

(C−H, B[h]Q), 122.1 (s, C−H_parain BPh₄), 119.0 (Ir−NCMe), 91.5

(Cp-Me₅), 9.2 (Cp-Me₅), 3.4 (Ir-NCMe). ¹¹B (128 MHz, 298 K,

CDCl₃): δ −6.5 (m). HRMS-ESI: calcd for C₂₅H₂₆IrN₂([3b]⁺) (m/z)

547.1721, found 547.1758. HRMS-ESI: calcd for C₂₄H₂₀B ([BPh₄]⁻)

(m/z) 319.1657, found 319.1655.

1

quartet, J_C−B= 49.6 Hz, C-B in BArF₂₄), 158.7 (DB[f,h]Q), 156.3

(DB[f,h]Q), 152.4 (H-C=N, DB[f,h]Q), 141.3 (DB[f,h]Q), 135.5

(C−H, DB[f,h]Q), 134.6 (C−H, DB[f,h]Q), 134.6 (C−H, DB[f,h]Q),

133.1 (DB[f,h]Q), 132.2 (m, DB[f,h]Q), 130.4 (C−H, DB[f,h]Q),

130.1 (C−H, DB[f,h]Q), 130.0 (q, ²J_C−F= 32.3 Hz, C-CF₃in BArF₂₄),

128.8 (C−H, DB[f,h]Q), 128.7 (DB[f,h]Q), 127.4 (DB[f,h]Q), 125.4

[Cp*Ir(B[h]Q)(NCMe)][OTf] ([3b][OTf]) (Scheme 2). This com-

pound was prepared following the general procedure described above:

2b (93.5 mg, 0.173 mmol), Ag[OTf] (342.2 mg, 0.173 mmol), MeCN

(7 mL), room temperature, 2 h. After the resulting residue was washed

with n-pentane, [3b][OTf] was recovered as a yellow powder in 76%

yield (113 mg). Anal. Calcd for C₂₆H₂₆F₃IrN₂O₃S·1/10CH₂Cl₂: C,

1

(q, J_C−F= 271.5 Hz, CF₃in BArF₂₄), 125.0 (C−H, DB[f,h]Q), 124.9

(C−H, DB[f,h]Q), 124.4 (C−H, DB[f,h]Q), 118.4 (m, C−H_parain

BArF₂₄), 118.3 (Ir−NCMe), 92.3 (Cp-Me₅), 10.0 (Cp-Me₅), 3.4 (Ir-

NCMe). ¹⁹F{¹H} NMR (282 MHz, 298 K, d₆-acetone): δ −64.2 (s,

CF₃in BArF₂₄). ¹¹B NMR (128 MHz, 298 K, d₆-acetone): δ −6.54

(m). HRMS-ESI: calcd for C₂₉H₂₈IrN₂([3c]⁺) (m/z) 597.1887, found

597.1914. HRMS-ESI: calcd for C₃₂H₁₂BF₂₄([BArF₂₄]⁻) (m/z)

863.0649, found 863.0649.

1

44.51; H, 3.75; N, 3.98. Found: C, 44.81; H, 3.77; N, 4.03. H (400

MHz, 298 K, CDCl₃): δ 9.26 (d, 1H, J = 5.3 Hz, H−CN B[h]Q),

8.32 (d, 1H, J = 8.1 Hz, H_ArB[h]Q), 7.97 (d, 1H, J = 6.7 Hz, H_Ar

B[h]Q), 7.87 (d, 1H, J = 8.8 Hz, H_ArB[h]Q), 7.77 (dd, 1H, J₁= 8.0

Hz, J₂= 8.0 Hz, H_ArB[h]Q), 7.63−7.70 (m, 3H, H_ArB[h]Q), 2.28 (s,

3H, Ir-NCMe), 1.79 (s, 15H, Cp-Me₅). ¹³C (126 MHz, 298 K,

Procedure for Piezometric Monitoring of the Catalytic O-

Silylation of Benzyl Alcohol (eq 1, Table 2). Benzyl alcohol (1.0

mL, 9.6 mmol or 0.5 mL, 4.8 mmol) and HSiEt₃(0.8 mL, 5.0 mmol)

O

DOI: 10.1021/acs.organomet.7b00749

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

were introduced in a small-volume double-walled Schlenk vessel (∼28

mL) thermostated at 20 °C. After the resulting mixture was slowly

stirred for a few minutes, a Mykrolis A332984−007 absolute pressure-

to-voltage transducer was airtightly screwed to the neck of the glass

vessel and connected to a Pico-Log analog−logic converter/data-

logger board interfaced to a computer. This allowed the measurement

of the pressure of hydrogen (computed from a linear absolute

pressure−output voltage relationship) developed in the known

overhead volume of the sealed reactor as a function of time. The

absence of leaks was preliminarily checked by applying a moderate

pressure of argon gas. The iridium precatalyst ([3a][BArF₂₄], 6.7 or

6.9 mg, 4.8 or 5.0 μmol; [3b][BArF₂₄], 6.2 mg, 4.4 μmol;

[3c][BArF₂₄],7.0 mg, 4.8 μmol; [3a][BArF₂₀], 5.8 mg, 4.8 μmol;

[3b][BArF₂₀], 5.9 mg, 4.8 μmol; 0.1 mol % relative to HSiEt₃)

dissolved in 1,2-dichloroethane (0.5 mL) was swiftly injected into the

vessel via a lateral glass valve, which was tightly closed immediately

upon injection to avoid gas leaks. A blank control experiment carried

out without any precatalyst did not produce any noticeable hydrogen

gas pressure. A picture of the setup used here is given in Figure S31 in

the Supporting Information.

time. For ¹H NMR analysis, tri-tert-butylbenzene was added as internal

reference to the reaction mixture, after which an aliquot was taken

from this mixture and mixed with CD₂Cl₂or C₆D₆. For further details,

see the Supporting Information.

General Procedure for the [3b][BArF₂₄] Load Optimization

in the Catalytic Hydrosilylation of Benzonitrile with HSiEt₃

(Figure 5). Using the general procedure described above (Table 4),

the conditions for each catalytic run were as follows: benzonitrile (0.1

mL, 0.97 mmol), HSiEt₃(0.35 mL, 2.13 mmol), [3b][BArF₂₄] (0.48

μmol, 0.05 mol %; 0.1 μmol, 0.1 mol %; 1.9 μmol, 0.2 mol %; 4.8

μmol, 0.5 mol %; 9.7 μmol, 1 mol %), CH₂Cl₂(0.5 mL). The reaction

progress was monitored at diﬀerent times by using ¹H NMR

spectroscopy. For further details, see Table S2 in the Supporting

Information.

Reaction of [3b][BArF₂₄] with HSiEt₃(Step 1, Scheme 3). In a

glovebox, HSiEt₃(∼10 μL, 62.6 μmol) was added to a solution of

[3b][BArF₂₄] (20 mg, 14.2 μmol) in CD₂Cl₂(0.55 mL) placed in a

vial equipped with a septum. The resulting solution was vigorously

shaken and transferred into a J. Young NMR sample tube, which was

subsequently tightly sealed for analysis. After ∼25 min the NMR tube

was frozen and introduced in the NMR spectrometer’s probe.

Multinuclear NMR analysis (at −60 °C) of the reaction mixture

revealed the total conversion of [3b][BArF₂₄] into mainly [4b]-

[BArF₂₄], [7b][BArF₂₄], and EtN(Et₃Si)₂along with unconsumed

HSiEt₃in a respective ratio of ∼1:0.1:1.2:1.2 (see spectra in the

General Procedures for the Catalytic Hydrosilylation of

Acetonitrile and Propionitrile (Table 3). Procedure 1 (Solvent-

Free Conditions, Entries 1−5). To a solution of [3a][BArF₂₄] or

[3b][BArF₂₄] (0.1 or 1 mol %) in MeCN (0.5 or 1 mL, 9.7 or 19.4

mmol) was added HSiEt₃(3.1 or 6.73 mL, 19.4 or 42.1 mmol). The

reaction mixture was heated at the given temperature for the given

1

Supporting Information). Data for [4b][BArF₂₄] are as follows. H

1

NMR (600 MHz, 213 K, CD₂Cl₂): δ 8.67 (d, 1H, J = 5.3 Hz, H_Ar

B[h]Q), 8.38 (d, 1H, J = 8.0 Hz, H_ArB[h]Q), 7.90 (d, 1H, J = 8.8 Hz,

H_ArB[h]Q), 7.8 (t, 2H, J = 7.4 Hz, H_ArB[h]Q), 7.73 (m, 8H, H_ortho

BArF₂₄), 7.70 (t, 1H, J = 8.3 Hz, H_ArB[h]Q), 7.71 (m, 1H, H_Ar

B[h]Q), 7.67 (d, 1H, J = 8.7 Hz, H_ArB[h]Q), 7.59 (dd, 1H, J₁= 7.9

Hz, J₂= 7.9 Hz, H_ArB[h]Q), 7.53 (s, 4H, H_paraBArF₂₄), 1.69 (s, 15H,

Cp-Me₅), 0.56 (t, 9H, J = 7.9 Hz, [Ir]-Si(CH₂CH₃)₃), 0.29 (m, 3H

[Ir]-Si(CH₂CH₃)₃), 0.22 (m, 3H, [Ir]-Si(CH₂CH₃)₃), − 11.44 (bs,

1H, J_H−Si= 20 Hz, [Ir]-H). The reaction described above was repeated

time. The reaction mixture was monitored using H NMR spectros-

copy at diﬀerent intervals of time. After 44 h of reaction at 50 °C

(entry 2), the temperature of the reaction mixture was raised to 100

°C (entry 3). For ¹H NMR spectroscopy analysis, tri-tert-butylbenzene

was added as an internal reference to the reaction mixture, after which

an aliquot was taken from this mixture and mixed with CD₂Cl₂or

C₆D₆. For spectroscopic details, see the Supporting Information.

Procedure 2 (with CH₂Cl₂as the Solvent, Entries 6−13). To a

solution of alkylnitrile (1 equiv) and silane (2.2 equiv) was added

[3b][BArF₂₄] (0.1 or 1 mol %) in CH₂Cl₂(0.5 mL). The reaction

mixture was heated at either 60 °C for 72 h (entry 6) or at 80 °C for

1

again for recording of the ¹³C spectrum (and additional 2D H,¹³C

spectra): ¹³C NMR (126 MHz, 300 K, CD₂Cl₂): δ 161.5 (1:1:1:1

1

quartet, J_C−B= 49.6 Hz, C-B in BArF₂₄), 156.5 (B[h]Q), 151.5 (H−

24 h (entries 7−13). For H NMR analyses, tri-tert-butylbenzene was

CN, B[h]Q), 143.9 (B[h]Q), 140.8 (B[h]Q), 138.6 (C−H, B[h]Q),

added as an internal reference to the reaction mixture, after which an

aliquot was taken from this mixture and mixed with C₆D₆. For

spectroscopic details, see the Supporting Information.

135.0 (B[h]Q), 134.9 (C−H, B[h]Q), 134.4 (s, C−H_orthoin BArF₂₄),

2

130.3 (C−H, B[h]Q), 130.2 (C−H, B[h]Q), 128.4 (q, J_C−F= 30.6

Hz, C-CF₃in BArF₂₄), 128.1 (B[h]Q), 124.2 (q, ¹J_C−F= 272.8 Hz, CF₃

in BArF₂₄), 124.0 (C−H, B[h]Q), 123.3 (C−H, B[h]Q), 122.7 (C−H,

B[h]Q), 117.3 (m, C−H_parain BArF₂₄), 101.3 (Cp-Me₅), 8.8 (Cp-

Me₅), 8.6 ([Ir]-Si(CH₂CH₃)₃), 5.8 ([Ir]-Si(CH₂CH₃)₃), 5.3 ([Ir]-

Si(CH₂CH₃)₃). ²⁹Si-DEPT NMR (119 MHz, 213 K, CD₂Cl₂): δ 6.93

(s, [Ir]-SiEt₃), 6.90 (s, [Ir]-SiEt₃). Data for EtN(Et₃Si)₂are as follows:

¹H NMR (600 MHz, 213 K, CD₂Cl₂): δ 2.81 (q, 2H, J = 7.0 Hz, N−

CH₂CH₃), 0.98 (t, 3H, J = 7.0 Hz, N−CH₂CH₃), 0.90 (t, 18H, J = 8.0

Hz, SiCH₂CH₃), 0.56 (q, 12H, J = 8.0 Hz, SiCH₂CH₃,). ²⁹Si-DEPT

NMR (119 MHz, 213 K, CD₂Cl₂): δ 10.9 ((Et₃Si)₂NEt). Data for

Data for 9H₂and Its Tautomer 9′H₂(from Fraction 1) (Figure 3).

Anal. Calcd for C₄H₁₀N₂: C, 55.78; H, 11.70; N, 32.52. Found: C,

54.34; H, 11.31; N, 31.71. ¹H NMR (500 MHz, 298 K, CDCl₃): δ 3.16

(q, 2H, J = 7.0 Hz, −CH₂−), 1.94 (s, 3H, CH₃−CN−), 1.16 (t, 3H,

J = 7.1 Hz, CH₃−CH₂−). ¹³C NMR (126 MHz, 298 K, CDCl₃): δ

161.3 (−CN−, because not directly visible, its presence was inferred

1

from H,¹³C HMBC 2D-correlation spectrum), 36.7 (−CH₂−), 23.9

(CH₃−CN−), 14.8 (CH₃−CH₂−).

Data for 10H₂(from Fraction 2). The sample was obtained as a

mixture with 9H₂in a respective ratio of ∼0.8:1. ¹H NMR (500 MHz,

298 K, CDCl₃): δ 3.25 (q, 2H, J = 7.3 Hz, −CH₂−), 2.00 (s, 6H,

CH₃−CN−), 1.11 (t, 3H, J = 7.3 Hz, CH₃−CH₂−). ¹³C NMR (126

MHz, 298 K, CDCl₃): δ 172.5 (−CN−), 34.5 (−CH₂−), 22.7

(CH₃−CN−), 15.0 (CH₃−CH₂−).

1

unconsumed and “free” HSiEt₃are as follows. H NMR (600 MHz,

1

213 K, CD₂Cl₂): δ 3.50 (sept, 1H, J_H−Si= 173.6 Hz, H−SiEt₃), 0.89

(t, 9H, J = 8.0 Hz, SiCH₂CH₃,), 0.52 (q, 6H, J = 8.0 Hz, SiCH₂CH₃,).

²⁹Si-DEPT NMR (119 MHz, 213 K, CD₂Cl₂): δ 0.0 (s, HSiEt₃). A ²⁹Si

1

signal of signiﬁcant intensity (see H−²⁹Si HMQC and ²⁹Si-DEPT

General Procedures for the Catalytic Hydrosilylation of

Benzonitrile (Table 4). Procedure 1 (Solvent-Free Conditions,

Entries 1 and 2). To a solution of [3a][BArF₂₄] (135 mg, 97 μmol, 1

mol %) in benzonitrile (1 mL, 9.7 mmol) was added HSiEt₃(3.1 mL,

19.5 mmol). The reaction mixture was heated at 50 °C for 20 h (entry

1) and then at 80 °C for 24 h (entry 2). The reaction mixture was

monitored using ¹H NMR spectroscopy at these two intervals of time.

For each analysis, tri-tert-butylbenzene was added as internal reference

to the reaction mixture, after which an aliquot from the reaction

mixture was taken and mixed with CD₂Cl₂for NMR analysis. For

spectroscopic details, see the Supporting Information.

spectra in the Supporting Information) at δ 9.5 ppm (s) could not be

assigned with certainty to any known molecule or side product of the

acetonitrile hydrosilylation.

Procedure for the Attempted in Situ Generation of

[Et₃Si(NCMe)_n][OTf] Followed by Its Stoichiometric Reaction

with either HSiEt₃or 5b (Step 3, Scheme 3). MeCN (2.5 μL, 126

μmol) was added to a solution of TfO-SiEt₃(14 μL, 63 μmol) in

CD₂Cl₂(∼0.5 mL) placed in a 2 mL vial equipped with a septum. The

reaction mixture was vigorously shaken at room temperature for a few

seconds and subsequently transferred into a J. Young NMR sample

tube that was then tightly sealed. The NMR tube was subsequently put

in the probe of the spectrometer (set at 25 °C) for spectral acquisition.

The resulting ¹H NMR spectrum showed the presence of the

characteristic peaks of both free acetonitrile (δ 1.99 (bs)) and free

Procedure 2 (with CH₂Cl₂as the Solvent, Entries 3−8). To a

mixture of benzonitrile (1 equiv) and HSiEt₃(2.2 equiv) was added a

solution of [3b][BArF₂₄] (0.001−1 mol %) in CH₂Cl₂(0.5 mL). The

reaction mixture was heated at the given temperature for the given

P

DOI: 10.1021/acs.organomet.7b00749

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

TfO-SiEt₃(δ 1.09 (t, 9H, J = 8.0 Hz, CH₃CH₂Si−); 0.97 (q, 6H, J =

8.0 Hz, CH₃CH₂Si)). The NCMe:TfO-SiEt₃ratio was estimated to be

∼2:1. To this solution mixture was then added a CD₂Cl₂solution

(∼0.2 mL) of either HSiEt₃(10 μL, 62.6 μmol) or 5b (32 mg, 63.2

μmol). The resulting solution was shaken at room temperature for a

few seconds before a new NMR spectrum was recorded. In the case of

the experiment run with HSiEt₃, the resulting NMR spectrum showed

no noticeable change in the characteristic chemical shifts of

respectively free MeCN, TfO-SiEt₃, and HSiEt₃. However, in the

case of the experiment run with 5b, the resulting NMR spectrum

showed quantitative conversion of 5b to new iridium species. Among

the latter, the ionic salt [3b][OTf] was found to be the major iridium

species on the basis of the identiﬁcation of its typical signals in

comparison to those of an authentic sample. While the typical signals

of TfO-SiEt₃were no longer detected, typical signals of free MeCN

and coordinated MeCN ([3b][OTf]) were observed respectively at δ

1.99 ppm (bs) and δ 2.19 ppm (s). Finally, characteristic peaks of the

products of hydrosilylation of acetonitrile were not observed. Although

typical signals of the ethyl moiety were indeed observed at δ 0.94 ppm

(t, 9H) and δ 0.53 ppm (q, 6H)), the fate of the silylium cation

[SiEt₃]⁺could not be traced.

slow diﬀusion of a n-hexane/benzene solution of excess HSiEt₃into a

solution of [3b][BArF₂₄] in 1,2-dicholoremethane.

NMR Data for [{Cp*Ir(B[h]Q)}₂(μ-H)][BArF₂₄] ([7b][BArF₂₄]

and [7b′][BArF₂₄]) (Step 5, Scheme 3). Since [7b][BArF₂₄] and

1

[7b′][BArF₂₄] could not be successfully isolated, their H NMR data

were obtained from their in situ formation (in a J. Young NMR sample

tube) by reaction of [3b][BArF₂₄] with 5b (see the procedure below).

¹H NMR data for [7b][BArF₂₄] (major product) (400 MHz, 298 K,

CD₂Cl₂): δ 8.07 (d, 2H, J = 7.3 Hz, H−CN B[h]Q), 7.67−7.75 (m,

4H, H_Arof B[h]Q overlapped with H_orthoof BArF₂₄), 7.54 (d, 2H, H_Ar

of B[h]Q overlapped with H_paraof BArF₂₄), 7.36 (dd, 2H, J₁= 6.9 Hz,

J₂= 1.9 Hz, H_ArB[h]Q), 7.28 (d, 1H, J = 8.8 Hz, H_ArB[h]Q), 6.10−

6.13 (m, 4H, H_ArB[h]Q), 1.43 (s, 30H, Cp-Me₅), −22.63 (s, 1H, Ir-

H). ¹H NMR data for [7b′][BArF₂₄] (minor product) (400 MHz, 298

K, CD₂Cl₂): δ 7.77 (m, 2H, H−CN H_Arof B[h]Q overlapped with

H_paraof BArF₂₄), 7.41 (d, 2H, J = 8.7 Hz, H_ArB[h]Q), 7.25 (d, 2H,

H_Arof B[h]Q overlapped with H_paraof BArF₂₄), 7.01 (t, 2H, J = 7.5

Hz, H_ArB[h]Q), 6.86 (dd, 2H, J₁= 8.0 Hz, J₂= 8.0 Hz, H_ArB[h]Q),

6.78 (d, 2H, J = 7.3 Hz, H_ArB[h]Q), 6.10−6.13 (m, 4H, H_ArB[h]Q),

1.43 (s, 30H, Cp-Me₅), −22.79 (s, 1H, Ir-H).

Reaction of [3a][BArF₂₄] with 5a (Step 5, Scheme 3; Scheme

6). In a glovebox, a solution of [3a][BArF₂₄] (35.9 mg, 25.9 μmol) in

CD₂Cl₂(∼0.6 mL) was placed in a vial containing 5a (5.0 mg, 10.4

μmol). The resulting red solution was shaken at room temperature for

few seconds and transferred into a J. Young NMR sample tube, which

was tightly sealed. The NMR tube was left at room temperature for 15

Procedure for the in Situ Generation of [Et₃Si(NCMe)_n]-

[BArF₂₀] Followed by Its Stoichiometric Reaction with 5b (Step

3, Scheme 3; Scheme 5). In a 2 mL vial equipped with a septum,

MeCN (2.5 μL, 48.7 μmol) was added to a solution of [Ph₃C]-

[BArF₂₀] (10.0 mg, 10.8 μmol) in CD₂Cl₂(∼0.5 mL). The reaction

mixture was cooled to 0 °C and transferred into a J. Young NMR

sample tube, which was subsequently tightly sealed. The NMR tube

1

min before recording the H NMR spectrum. The NMR tube was

subsequently left to stand for ∼2.5 h at room temperature before

excess MeCN (53 μL, 1.04 mmol) was added. After 15 min, a new

NMR spectrum was recorded. See the Supporting Information for

assigned NMR spectra.

1

was left at 0 °C for ∼15 min before recording the ﬁrst H NMR

spectrum (the spectrometer’s probe temperature was set to 0 °C). The

resulting ¹H NMR spectrum revealed characteristic peaks of both

MeCN (δ 1.99 (bs)) and [Ph₃C][BArF₂₀] (δ 8.25 (t, 3H, J = 7.4 Hz,

H_para); 7.86 (t, 6H, J = 7.4 Hz, H_meta); 7.66 (d, 6H, J = 7.6 Hz, H_ortho))

accompanied by some residual acetone impurity (δ 2.13 (bs)) in a

respective ratio of 4:1:0.3 (see the NMR spectra in the Supporting

Information). To this solution mixture was subsequently added HSiEt₃

(3.5 μL, 22 μmol). The resulting orange solution was shaken at 0 °C

for a few seconds before a new NMR analysis was recorded. The

resulting NMR spectrum showed the quantitative conversion of

[Ph₃C][BArF₂₀] and MeCN into respectively Ph₃CH (¹H: δ 7.29 (t,

6H, J = 7.3 Hz, H_meta); 7.22 (t, 3H, J = 7.3 Hz, H_para); 7.12 (d, 6H, J =

7.3 Hz, H_ortho)), solvated proton, and O(SiEt₃)₂(¹H: δ 2.13 (bs, nH,

−NCMe); 0.93 (t, 9H, CH₃CH₂Si−); 0.52 (q, 6H, CH₃CH₂Si--); ²⁹Si:

δ 8.92 (s, Et₃Si−)) accompanied by excess HSiEt₃(¹H: δ 3.59 (sept,

1H, J = 3.0 Hz, H−Si−); 0.96 (t, 9H, J = 7.9 Hz, CH₃CH₂Si−); 0.52

(qd, 6H, J₁= 8.0 Hz, J₂= 3.0 Hz, CH₃CH₂Si−); ²⁹Si: δ 0.53 (s,

Et₃Si−)) and the product of hydrosilylation of the residual acetone, i.e.

[CH₃CHO(SiEt₃)] (¹H: δ 4.71 (sept, 1H, J = 6.2 Hz, CH₃CH−); the

signals of the SiEt₃group overlapped with other similar groups). The

Ph₃CH:O(SiEt₃)₂:HSiEt₃:[CH₃CHO(SiEt₃)] ratio was estimated to

be ∼1:0.5:0.6:0.5. To this solution mixture was then added a solution

of 5b (5.5 mg, 10.8 μmol) in CD₂Cl₂(∼0.5 mL). The resulting red

solution was shaken at 0 °C for a few seconds before a new NMR

analysis was recorded. The resulting NMR spectrum revealed the

quantitative conversion of 5b (¹H: δ 1.90 (s, 15H, Cp-Me₅); −15.28

(s, 1H, Ir-H)) into [7b][BArF₂₀] (¹H: δ 1.42 (s, 30H, Cp-Me₅);

−22.80 (s, 1H, Ir-H)) and [3b][BArF₂₀] (¹H: δ 2.13 (s, 3H, Ir-

NCMe), 1.73 (s, 15H, Cp-Me₅)). The spectrum also revealed the

formation of the three main hydrosilylation products EtNR₂(R = SiEt₃

H, N(SiEt₃)C(Me)N(SiE₃)₂), as suggested by the presence of the

characteristic CH₃CH₂N− quadruplet resonance signal at respectively

δ(¹H) 3.65 (J = 6.8 Hz); 2.83 (J = 7.0 Hz), and 2.75 (J = 7.0 Hz) (see

the NMR spectra in the Supporting Information for details). The

chemical shift at δ 2.83 ppm is assigned to the product of double

hydrosilylation of MeCN: i.e., EtN(SiEt₃)₂. The Ph₃CH:[7b][BArF₂₀]:

[3b][BArF₂₀]:EtNR₂ratio was estimated to be ∼1.9:1:2.

Reaction of [3b][BArF₂₄] with 5b (Step 5, Scheme 3; Scheme

6). In a glovebox, a solution of [3b][BArF₂₄] (18.1 mg, 13 μmol) in

CD₂Cl₂(∼0.3 mL) was placed in a vial containing a solution of 5b

(5.0 mg, 10 μmol) in CD₂Cl₂(∼0.3 mL). The resulting red solution

was shaken at room temperature for a few seconds and transferred into

a J. Young NMR sample tube, which was tightly sealed. The NMR

tube was left to stand at room temperature for 15 min before recording

the ¹H NMR spectrum. The NMR tube was subsequently left to stand

for ∼4 h at room temperature before excess MeCN (25 μL, 774.0

μmol) was added. After 15 min, a new NMR spectrum was recorded.

See the Supporting Information for assigned NMR spectra.

Reaction of [Ph₃C][BArF₂₀] with 5a (Step 5, Scheme 3; eq 3).

In a glovebox, a solution of [Ph₃C][BArF₂₀] (9.5 mg, 10.4 μmol) in

d₅.PhCl (∼0.3 mL) was placed in a vial containing a solution of 5a (10

mg, 20.7 μmol) in d₅-PhCl (∼0.3 mL). The resulting red solution was

shaken at room temperature for a few seconds and transferred into a J.

Young NMR sample tube, which was tightly sealed. The NMR tube

was left to stand at room temperature for 15 min before recording the

¹H NMR spectrum. The resulting ¹H NMR spectrum showed

quantitative conversion of the starting material into Ph₃CH and 7a.

See the Supporting Information for the assigned NMR spectrum.

1

2

Procedure for H and H NMR Monitoring of the Catalytic

Hydrosilylation of CD₃CN Catalyzed by [3a][BArF₂₄] (Steps 3

and 5, Scheme 3). This experiment was ﬁrst conducted at room

temperature and then repeated at −50 °C, as described by the

following general procedure: in a glovebox, CD₃CN (9.2 μL, 180

μmol) was added to a solution of [3a][BArF₂₄] (25 mg, 18 μmol) in

CD₂Cl₂(0.5 mL) placed in a vial equipped with a septum. HSiEt₃(58

μL, 360 μmol) was then added dropwise. The resulting solution was

shaken and transferred into a J. Young NMR sample tube, which was

subsequently tightly sealed for analysis. After 30 min of reaction at

room temperature (or 60 min at −50 °C), the NMR tube was

introduced in the NMR spectrometer’s probe (298 or 223 K). ¹H and

²H NMR spectra were subsequently recorded (see spectra in

Supporting Information).

Procedure for the Catalytic Hydrosilylation of Benzonitrile

at 70 °C (Table 5). To a mixture of HSiEt₃(0.35 mL, 2.2 mmol) and

benzonitrile (0.1 mL, 0.97 mmol) was added a solution of the iridium

precatalyst ([3a][BArF₂₄], 6.7 mg; [3b][BArF₂₄], 6.8 mg; [3c]-

Procedure for the Crystallization of [{Cp*Ir(B[h]Q)}₂(μ-

H)][BArF₂₄] ([7b][BArF₂₄]) (Step 5, Scheme 3). Red single crystals

of [7b][BArF₂₄] suitable for X-ray diﬀraction analysis were grown by

Q

DOI: 10.1021/acs.organomet.7b00749

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

[BArF₂₄], 7.0 mg; [3a][BArF₂₀], 5.8 mg; [3b][BArF₂₀], 5.9 mg, 4.8

μmol; 0.5 mol % relative to benzonitrile) in CH₂Cl₂(0.5 mL). The

reaction mixture was heated at 70 °C for 24 h, after which tri-tert-

butylbenzene was added as internal NMR standard. An aliquot from

this solution was taken and mixed with C₆D₆for further NMR analysis.

Procedure for the Catalytic Hydrosilylation of Acetophe-

none at Room Temperature (Table 6). To a mixture of HSiEt₃

(0.15 mL, 0.94 mmol) and acetophenone (0.1 mL, 0.86 mmol) was

added a solution of the iridium precatalyst ([3a][BArF₂₄], 1.2 mg;

[3b][BArF₂₄], 1.2 mg; [3c][BArF₂₄], 1.3 mg; [3a][BArF₂₀], 1.0 mg;

[3b][BArF₂₀], 1.1 mg, 8.6 μmol; 0.1 mol % relative to acetophenone)

in CH₂Cl₂(0.5 mL). The reaction was allowed to run at room

temperature for 30 min, after which tri-tert-butylbenzene was added as

internal NMR standard. An aliquot of this solution was removed and

mixed with C₆D₆for further NMR analysis.

Cambridge Crystallographic Data Centre, 12 Union Road,

Cambridge CB2 1EZ, UK; fax: +44 1223 336033.

AUTHOR INFORMATION

■

Corresponding Author

*J.-P.D.: e-mail, djukic@unistra.fr; ph, +33(0)3 68 85 15 23.

ORCID

Mustapha Hamdaoui: 0000-0002-2486-7261

Jean-Pierre Djukic: 0000-0003-3196-4921

Notes

The authors declare no competing ﬁnancial interest.

General Procedure for the Study of the Eﬀect of the Nature

of the Counteranion of [3b]⁺on the Double Hydrosilylation of

Benzonitrile with HSiEt₃(Table 7). All of the reactions were

conducted for 24 h at 70 °C following the general procedure 2

described above (Table 4): benzonitrile (0.1 mL, 0.97 mmol), HSiEt₃

(0.35 mL, 2.13 mmol), precatalyst (9.7 μmol, 1 mol %), CH₂Cl₂(0.5

mL).

ACKNOWLEDGMENTS

■

The authors thank the CNRS, the Agence Nationale de la

Recherche (ANR) project COCOORDCHEM, the LABEX

̀

“Chimie des Systemes Complexes”, and the University of

Strasbourg for funding. The authors gratefully thank Lydia

Karmazin and Corinne Bailly for the crystallographic resolution

of the structures disclosed therein.

General Procedure for the Catalytic Hydrosilylation of

Nitriles (Scheme 7). To a mixture of silane (1.1−8.8 equiv) and

nitrile (1.0 equiv) was added a solution of the precatalyst in CH₂Cl₂

(0.5 mL). The reaction was allowed to take place upon heating at 70

°C for 24 h, after which tri-tert-butylbenzene was added as internal

NMR standard. An aliquot from this solution was taken and mixed

with C₆D₆for NMR analysis. Further details are provided in the

Supporting Information, which include the conditions used for each

REFERENCES

■

(1) (a) Martin, S. F. Pure Appl. Chem. 2009, 81, 195−204.

(b) Lawrence, S. A. Amines: synthesis, properties, and applications;

Cambridge University Press: Cambridge, U.K., 2004.

(2) Das, S.; Zhou, S.; Addis, D.; Enthaler, S.; Junge, K.; Beller, M.

Top. Catal. 2010, 53, 979.

1

catalytic run, characterization data of new products, and assigned H

NMR spectra of all products.

(3) Amundsen, L. H.; Nelson, L. S. J. Am. Chem. Soc. 1951, 73, 242−

General Procedure for Competition Experiments (eqs 4 and

5). In a J. Young NMR sample tube was placed an equimolar ratio of

the two substrates (benzaldehyde [2.13 or 0.426 mmol], acetophe-

none [2.13 mmol], or benzonitrile [0.426 mmol] and HSiEt₃(2.13 or

0.426 mmol). Then a solution of [3b][BArF₂₄] (2.13 μmol, 0.1−0.5

mol % relative to substrate) in CH₂Cl₂(0.5 mL) was added. The

reaction was allowed to run at room temperature for a given time, after

which a small aliquot was removed, mixed with a solution of tri-tert-

butylbenzene in C₆D₆(∼0.2 mL) used as internal NMR standard, and

analyzed by NMR spectroscopy.

244.

(4) (a) Bagal, D. B.; Bhanage, B. M. Adv. Synth. Catal. 2015, 357,

883−900. (b) Gomez, S.; Peters, J. A.; Maschmeyer, T. Adv. Synth.

Catal. 2002, 344, 1037−1057. (c) Tokmic, K.; Jackson, B. J.; Salazar,

A.; Woods, T. J.; Fout, A. R. J. Am. Chem. Soc. 2017, 139, 13554−

13561. (d) Mukherjee, A.; Srimani, D.; Ben-David, Y.; Milstein, D.

ChemCatChem 2017, 9, 559−563. (e) Saito, Y.; Ishitani, H.; Ueno, M.;

Kobayashi, S. ChemistryOpen 2017, 6, 211−215. (f) Bornschein, C.;

Werkmeister, S.; Wendt, B.; Jiao, H.; Alberico, E.; Baumann, W.;

Junge, H.; Junge, K.; Beller, M. Nat. Commun. 2014, 5, 4111.

(5) Luo, Y.-R. Comprehensive Handbook of Chemical Bond Energies;

CRC Press: Boca Raton, FL, 2007.

General Procedure for the Catalytic Hydrosilylation of

Carbonyls. See the Supporting Information for substrate scope. To

a mixture of silane (1.1−4.0 equiv) and carbonyl (1.0 equiv) substrates

was added a solution of the precatalyst in 0.5 mL of CH₂Cl₂. The

reaction was allowed to take place at room temperature for a given

time, after which tri-tert-butylbenzene was added as internal NMR

standard. An aliquot from this solution was taken and mixed with

C₆D₆for NMR analysis. Further details are provided in the Supporting

Information, which include the conditions used for each catalytic run,

(6) Calas, R.; Frainnet, E.; Bazouin, A. C.R. Acad. Sci. 1961, 252,

420−422.

(7) Murai, T.; Sakane, T.; Kato, S. Tetrahedron Lett. 1985, 26, 5145−

5148.

(8) Cabrita, I.; Fernandes, A. C. Tetrahedron 2011, 67, 8183−8186.

(9) (a) Huckaba, A. J.; Hollis, T. K.; Reilly, S. W. Organometallics

2013, 32, 6248−6256. (b) Itagaki, S.; Sunaba, H.; Kamata, K.;

Yamaguchi, K.; Mizuno, N. Chem. Lett. 2013, 42, 980−982.

(c) Campos, J.; Rubio, M.; Esqueda, A. C.; Carmona, E. J. Labelled

Compd. Radiopharm. 2012, 55, 29−38. (d) Caporusso, A. M.; Panziera,

N.; Pertici, P.; Pitzalis, E.; Salvadori, P.; Vitulli, G.; Martra, G. J. Mol.

Catal. A: Chem. 1999, 150, 275−285.

1

¹H NMR data of new products, and assigned H NMR spectra of all

products.

ASSOCIATED CONTENT

■

S

* Supporting Information

(10) Gutsulyak, D. V.; Nikonov, G. I. Angew. Chem., Int. Ed. 2010, 49,

7553−7556.

The Supporting Information is available free of charge on the

ACS Publications website at DOI: 10.1021/acs.organo-

met.7b00749.

(11) (a) Rit, A.; Zanardi, A.; Spaniol, T. P.; Maron, L.; Okuda, J.

Angew. Chem., Int. Ed. 2014, 53, 13273−13277. (b) Boone, C.;

Korobkov, I.; Nikonov, G. I. ACS Catal. 2013, 3, 2336−2340.

Extensive experimental, structural, and computational

details (PDF)

Cartesian coordinates for the calculated structures (XYZ)

(12) Wubbolt, S.; Maji, M. S.; Irran, E.; Oestreich, M. Chem. - Eur. J.

̈

2017, 23, 6213−6219.

(13) Ito, M.; Itazaki, M.; Nakazawa, H. ChemCatChem 2016, 8,

3323−3325.

Accession Codes

(14) Gandhamsetty, N.; Jeong, J.; Park, J.; Park, S.; Chang, S. J. Org.

Chem. 2015, 80, 7281−7287.

CCDC 1578556−1578565 contain the supplementary crystal-

lographic data for this paper. These data can be obtained free of

charge via www.ccdc.cam.ac.uk/data_request/cif, or by email-

ing data_request@ccdc.cam.ac.uk, or by contacting The

́

(15) Perez, M.; Qu, Z.-W.; Caputo, C. B.; Podgorny, V.; Hounjet, L.

J.; Hansen, A.; Dobrovetsky, R.; Grimme, S.; Stephan, D. W. Chem. -

Eur. J. 2015, 21, 6491−6500.

R

DOI: 10.1021/acs.organomet.7b00749

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

(16) Bornschein, C.; Werkmeister, S.; Junge, K.; Beller, M. New J.

Chem. 2013, 37, 2061.

(17) Klei, S. R.; Tilley, T. D.; Bergman, R. G. Organometallics 2002,

T.; Balasubramanian, T. M.; Liu, G. K.; Kinsley, S. A.; Duckworth, C.

A.; Poteruca, J. J.; Brown, P. S.; Miller, J. L. J. Org. Chem. 1987, 52,

1017−1021.

(34) (a) te Velde, G.; Bickelhaupt, F. M.; Baerends, E. J.; Fonseca

Guerra, C.; van Gisbergen, S. J. A.; Snijders, J. G.; Ziegler, T. J.

Comput. Chem. 2001, 22, 931−967. (b) ADFview; SCM, Theoretical

Chemistry, Vrije Universiteit, Amsterdam, The Netherlands, 2016.

(35) Steinmann, S. N.; Corminboeuf, C. J. Chem. Theory Comput.

2011, 7, 3567−3577.

(36) Popelier, P. L. A. Coord. Chem. Rev. 2000, 197, 169−189.

(37) Bader, R. F. W. In Atoms in Molecules: A Quantum Theory;

Clarendon: Oxford, U.K, 1990.

21, 4648−4661.

(18) (a) Fogg, D. E.; dos Santos, E. N. Coord. Chem. Rev. 2004, 248,

2365−2379. (b) Shindoh, N.; Takemoto, Y.; Takasu, K. Chem. - Eur. J.

2009, 15, 12168−12179.

(19) (a) Iali, W.; Paglia, F. L.; Goff, X.-F. L.; Sredojevic, D.; Pfeffer,

M.; Djukic, J.-P. Chem. Commun. 2012, 48, 10310−10312.

(b) Hamdaoui, M.; Ney, M.; Sarda, V.; Karmazin, L.; Bailly, C.;

Sieffert, N.; Dohm, S.; Hansen, A.; Grimme, S.; Djukic, J.-P.

Organometallics 2016, 35, 2207−2223.

(38) Wiberg, K. B. Tetrahedron 1968, 24, 1083−1096.

(39) (a) Grossekappenberg, H.; Reissmann, M.; Schmidtmann, M.;

Mueller, T. Organometallics 2015, 34, 4952−4958. (b) Gaspar, P. P.

Science 2002, 297, 785−786. (c) Mueller, T. Struct. Bonding (Berlin,

Ger.) 2013, 155, 107−162.

(40) Nava, M.; Reed, C. A. Organometallics 2011, 30, 4798−4800.

(41) (a) Hosomi, A.; Miura, K. In Comprehensive Organometallic

Chemistry III; Crabtree, R. H., Mingos, D. M. P., Eds.; Elsevier:

Oxford, U.K., 2007; pp 297−339. (b) Lambert, J. B.; Zhang, S.; Ciro,

S. M. Organometallics 1994, 13, 2430−2443. (c) Kessler, M.; Knapp,

C.; Sagawe, V.; Scherer, H.; Uzun, R. Inorg. Chem. 2010, 49, 5223−

5230.

(20) (a) Corre, Y.; Rysak, V.; Trivelli, X.; Agbossou-Niedercorn, F.;

Michon, C. Eur. J. Org. Chem. 2017, 2017, 4820−4826. (b) Corre, Y.;

Trivelli, X.; Capet, F.; Djukic, J.-P.; Agbossou-Niedercorn, F.; Michon,

C. ChemCatChem 2017, 9, 2009−2017. (c) Corre, Y.; Rysak, V.;

Capet, F.; Djukic, J.-P.; Agbossou-Niedercorn, F.; Michon, C. Chem. -

Eur. J. 2016, 22, 14036−14041. (d) Corre, Y.; Werle, C.; Brelot-

Karmazin, L.; Djukic, J.-P.; Agbossou-Niedercorn, F.; Michon, C. J.

Mol. Catal. A: Chem. 2016, 423, 256−263. (e) Michon, C.; MacIntyre,

K.; Corre, Y.; Agbossou-Niedercorn, F. ChemCatChem 2016, 8, 1755−

1762.

(21) (a) Green, M. L. H. J. Organomet. Chem. 1995, 500, 127−148.

(b) Green, M. L. H.; Parkin, G. Dalton Trans. 2016, 45, 18784−18795.

(c) Green, M. L. H.; Parkin, G. J. Chem. Educ. 2014, 91, 807−816.

(d) Green, J. C.; Green, M. L. H.; Parkin, G. Chem. Commun. 2012,

48, 11481−11503.

(22) Ho Binh, D.; Milovanovic, M.; Puertes-Mico, J.; Hamdaoui, M.;

Zaric, S. D.; Djukic, J.-P. Chem. - Eur. J. 2017, DOI: 10.1002/

chem.201703373.

(42) (a) Boberski, W. G.; Allred, A. L. J. Organomet. Chem. 1975, 88,

65−72. (b) Arshadi, M.; Johnels, D.; Edlund, U.; Ottosson, C.-H.;

Cremer, D. J. Am. Chem. Soc. 1996, 118, 5120−5131.

(43) (a) Lambert, J. B.; Kania, L.; Zhang, S. Chem. Rev. 1995, 95,

1191−1201. (b) Lambert, J. B.; Zhang, S. J. Chem. Soc., Chem.

Commun. 1993, 383−384.

(44) Garces

́ ́

, K.; Fernandez-Alvarez, F. J.; Polo, V.; Lalrempuia, R.;

(23) Lipke, M. C.; Liberman-Martin, A. L.; Tilley, T. D. Angew.

Chem., Int. Ed. 2017, 56, 2260−2294.

́

Perez-Torrente, J. J.; Oro, L. A. ChemCatChem 2014, 6, 1691−1697.

(45) (a) Haiges, R.; Baxter, A. F.; Goetz, N. R.; Axhausen, J. A.;

Soltner, T.; Kornath, A.; Christe, K. O. Dalt. Trans. 2016, 45, 8494−

8499. (b) Reed, C. A. Acc. Chem. Res. 2013, 46, 2567−2575.

(46) (a) Bahr, S. R.; Boudjouk, P. J. Am. Chem. Soc. 1993, 115,

4514−4519. (b) Kira, M.; Hino, T.; Sakurai, H. Chem. Lett. 1993, 22,

153−156.

(24) (a) Li, L.; Brennessel, W. W.; Jones, W. D. J. Am. Chem. Soc.

2008, 130, 12414−12419. (b) Scheeren, C.; Maasarani, F.; Hijazi, A.;

Djukic, J.-P.; Pfeffer, M.; Zaric, S. D.; Le Goff, X.-F.; Ricard, L.

Organometallics 2007, 26, 3336−3345. (c) Park-Gehrke, L. S.;

Freudenthal, J.; Kaminsky, W.; DiPasquale, A. G.; Mayer, J. M. Dalton

Trans. 2009, 1972−1983.

(47) (a) Reißmann, M.; Schafer, A.; Jung, S.; Muller, T.

̈

(25) (a) Toennemann, J.; Risse, J.; Grote, Z.; Scopelliti, R.; Severin,

K. Eur. J. Inorg. Chem. 2013, 2013, 4558−4562. (b) White, C.; Yates,

A.; Maitlis, P. M.; Heinekey, D. M. Inorg. Synth. 1992, 29, 228−234.

(26) Fulmer, G. R.; Miller, A. J. M.; Sherden, N. H.; Gottlieb, H. E.;

Nudelman, A.; Stoltz, B. M.; Bercaw, J. E.; Goldberg, K. I.

Organometallics 2010, 29, 2176−2179.

(27) (a) Yang, Y. F.; Chung, L. W.; Zhang, X.; Houk, K. N.; Wu, Y.

D. J. Org. Chem. 2014, 79, 8856−8864. (b) Khalimon, A. Y.;

Shirobokov, O. G.; Peterson, E.; Simionescu, R.; Kuzmina, L. G.;

Howard, J. A. K.; Nikonov, G. I. Inorg. Chem. 2012, 51, 4300−4313.

(28) (a) Komuro, T.; Begum, R.; Ono, R.; Tobita, H. Dalt. Trans.

2011, 40, 2348−2357. (b) Zhang, X.-H.; Chung, L. W.; Lin, Z.; Wu,

Y.-D. J. Org. Chem. 2008, 73, 820−829. (c) Watanabe, T.; Hashimoto,

H.; Tobita, H. J. Am. Chem. Soc. 2006, 128, 2176−2177.

(29) Gutsulyak, D. V.; Vyboishchikov, S. F.; Nikonov, G. I. J. Am.

Chem. Soc. 2010, 132, 5950−5951.

(30) Hu, Y.; Li, L.; Shaw, A. P.; Norton, J. R.; Sattler, W.; Rong, Y.

Organometallics 2012, 31, 5058−5064.

(31) (a) Ngo, A. H.; Ibanez, M.; Do, L. H. ACS Catal. 2016, 6,

2637−2641. (b) Turlington, C. R.; Harrison, D. P.; White, P. S.;

Brookhart, M.; Templeton, J. L. Inorg. Chem. 2013, 52, 11351−11360.

(c) Hu, Y.; Li, L.; Shaw, A. P.; Norton, J. R.; Sattler, W.; Rong, Y.

Organometallics 2012, 31, 5058−5064.

(32) (a) Muether, K.; Mohr, J.; Oestreich, M. Organometallics 2013,

32, 6643−6646. (b) Klare, H. F. T.; Oestreich, M. Dalton Trans. 2010,

39, 9176−9184.

(33) (a) Bochkarev, M. N.; Balashova, T. V.; Maleev, A. A.; Fagin, A.

A.; Fukin, G. K.; Baranov, E. V. Inorg. Chim. Acta 2007, 360, 2368−

2378. (b) Balashova, T. V.; Khoroshenkov, G. V.; Kusyaev, D. M.;

Eremenko, I. L.; Aleksandrov, G. G.; Fukin, G. K.; Bochkarev, M. N.

Russ. Chem. Bull. 2004, 53, 825−829. (c) Forsberg, J. H.; Spaziano, V.

Organometallics 2013, 32, 6736−6744. (b) Herrington, T. J.; Ward,

B. J.; Doyle, L. R.; McDermott, J.; White, A. J. P.; Hunt, P. A.; Ashley,

A. E. Chem. Commun. 2014, 50, 12753−12756. (c) Scott, D. J.;

Fuchter, M. J.; Ashley, A. E. Angew. Chem., Int. Ed. 2014, 53, 10218−

10222.

(48) Corre, Y.; Rysak, V.; Trivelli, X.; Agbossou-Niedercorn, F.;

Michon, C. Eur. J. Org. Chem. 2017, 2017, 4820−4826.

(49) (a) Noedling, A. R.; Muether, K.; Rohde, V. H. G.; Hilt, G.;

Oestreich, M. Organometallics 2014, 33, 302−308. (b) Kim, K.-C.;

Reed, C. A.; Elliott, D. W.; Mueller, L. J.; Tham, F.; Lin, L.; Lambert, J.

B. Science 2002, 297, 825−827. (c) Khalimon, A. Y.; Lin, Z. H.;

Simionescu, R.; Vyboishchikov, S. F.; Nikonov, G. I. Angew. Chem., Int.

Ed. 2007, 46, 4530−4533. (d) Hoffmann, S. P.; Kato, T.; Tham, F. S.;

Reed, C. A. Chem. Commun. 2006, 767−769.

(50) Davies, D. L.; Al-Duaij, O.; Fawcett, J.; Giardiello, M.; Hilton, S.

T.; Russell, D. R. Dalton Trans. 2003, 2, 4132−4138.

(51) Scott, V. J.; C

2005, 127, 2852−2853.

(52) Connelly, S. J.; Kaminsky, W.; Heinekey, D. M. Organometallics

2013, 32, 7478−7481.

̧elenligil-Çetin, R.; Ozerov, O. V. J. Am. Chem. Soc.

(53) Note that two structures were acquired and resolved for [4b]

[BArF₂₄], to which were applied the SQUEEZE procedure to restrain

residual electron density pertaining to solvent molecules present

within the crystal lattice: i.e., benzene and n-hexane molecules

respectively. Figure 6 depicts the structure containing unresolved n-

hexane.

S

DOI: 10.1021/acs.organomet.7b00749

Organometallics XXXX, XXX, XXX−XXX

Article Doi

Iridacycles as Catalysts for the Autotandem Conversion of Nitriles into Amines by Hydrosilylation: Experimental Investigation and Scope

DOI: 10.1021/acs.organomet.7b00749

Source and publish data:

Authors:

Article abstract of DOI:10.1021/acs.organomet.7b00749

Full text of DOI:10.1021/acs.organomet.7b00749

Products guided by the article

R&D Labs maybe for 1312544-55-0

Relevant to this article

Hot Product