Evidence of a Donor-Acceptor (Ir-H)→SiR3 Interaction in a Trapped Ir(III) Silane Catalytic Intermediate

Article

pubs.acs.org/Organometallics

Evidence of a Donor−Acceptor (Ir−H)→SiR₃Interaction in a Trapped

Ir(III) Silane Catalytic Intermediate

Mustapha Hamdaoui,^†Marjolaine Ney,^†Vivien Sarda,^†Lydia Karmazin,^†Corinne Bailly,^†

Nicolas Sieﬀert,^‡Sebastian Dohm,^§Andreas Hansen,^§Stefan Grimme,^§and Jean-Pierre Djukic*^,†

^†Institut de Chimie de Strasbourg, UMR 7177 CNRS, Universite

^‡Universite

Grenoble Alpes & CNRS, DCM UMR 5250, F-38000 Grenoble, France

^§Mulliken Center for Theoretical Chemistry, Institut fur Physikalische und Theoretische Chemie, Universitat Bonn, Beringstraße 4,

́

de Strasbourg, 4 rue Blaise Pascal, F-67000 Strasbourg, France

́

̈

D-53115 Bonn, Germany

S

* Supporting Information

ABSTRACT: The ionic iridacycle [(2-phenylenepyridine-κN,κC)-

IrCp*(NCMe)][BArF₂₄] ([2][BArF₂₄]) displays a remarkable capa-

bility to catalyze the O-dehydrosilylation of alcohols at room

temperature (0.4 × 10³< TON < 10³, 8 × 10³< TOF_i< 1.9 × 10⁵

h⁻¹for primary alcohols) that is explained by its exothermic reaction

with Et₃SiH, which aﬀords the new cationic hydrido-Ir(III)-silylium

species [3][BArF₂₄]. Isothermal calorimetric titration (ITC) indicates

that the reaction of [2][BArF₂₄] with Et₃SiH requires 3 equiv of the

latter and releases an enthalpy of −46 kcal/mol in chlorobenzene.

Density functional theory (DFT) calculations indicate that the

thermochemistry of this reaction is largely dominated by the

concomitant bis-hydrosilylation of the released MeCN ligand. Attempts

to produce [3][BF₄] and [3][OTf] salts resulted in the formation of a

known neutral hydrido-iridium(III) complex, i.e. 4, and the release of

Et₃SiF and Et₃SiOTf, respectively. In both cases formation of the cationic μ-hydrido-bridged bis-iridacyclic complexes [5][BF₄]

and [5][OTf], respectively, was observed. The structure of [5][OTf] was established by X-ray diﬀraction analysis. Conversion of

[3][BArF₂₄] into 4 upon reaction with either 4-N,N-dimethylaminopyridine or [nBu₄][OTf] indicates that the Ir center holds a

+III formal oxidation state and that the Et₃Si⁺moiety behaves as a Z-type ligand according to Green’s formalism. [3][BArF₂₄],

which was trapped and structurally characterized and its electronic structure investigated by state-of-the-art DFT methods (DFT-

D, EDA, ETS-NOCV, QTAIM, ELF, NCI plots and NBO), displays the features of a cohesive hydridoiridium(III)→silylium

donor−acceptor complex. This study suggests that the fate of [3]⁺in the O-dehydrosilylation of alcohols is conditioned by the

nature of the associated counteranion and by the absence of Lewis base in the medium capable of irreversibly capturing the

silylium species.

IHI (interligand hypervalent interactions) and SISHA⁸

INTRODUCTION

■

(secondary interaction between a silicon and a hydrogen

atom) were proposed to rationalize the cohesion and

spectroscopic properties of the M−Si−H motif. The SISHA

concept, rather imprecise in its deﬁnition of the actual nature of

the key secondary interaction, embraces a large number of

bonding situations where the Si···H interaction within the M−

Si−H motif is weak but not negligible.⁸The complexity that

may lie behind the cohesion of the M−Si−H motif⁹was

addressed theoretically in the past,¹⁰although without formal

consideration for noncovalent interactions. Recent recourse to

X-ray diﬀraction multipolar reﬁnement methods with the

support of quantum theory of atoms in molecules¹¹(QTAIM)

has fuelled the quest for a uniﬁed picture of the electronic

Metal−silane complexes are central to many chemical trans-

formations that aim for the synthesis of high-value organic

molecules and materials.¹The most documented^1,2types of

metal−silane adducts are the σ-complexes (η^1,2-R₃Si-H)Mⁿ(n =

formal oxidation state) arising from the isohypsic³(i.e., n =

constant; by deﬁnition the isohypsic term refers to reactions

occurring at a given reactive center with no change in its formal

oxidation state) metal coordination of silane⁴and R₃Si-Mⁿ⁺²-H

complexes arising from oxidative addition of the Si−H bond at

Mⁿ.⁵However, intermediary situations considered as so-called

“arrested states” toward the Si−H bond cleavage by oxidative

addition were also pointed out and raised sustained attention.⁶

M−silane adducts are commonly categorized according to the

bonding relationships existing within the M−Si−H motif, i.e.

two-center−two-electron and three-center−two-electron inter-

actions.⁷Subcategories of stabilizing interactions referred to as

Received: March 27, 2016

A

DOI: 10.1021/acs.organomet.6b00248

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

Scheme 1. Synthesis of [2][X] (X = BArF₂₄, BF₄, OTf) and the Latter’s Anion-Sensitive Reactivity toward Et₃SiH

structure of metal−silane complexes.¹²Iridacycles of the

RESULTS AND DISCUSSION

■

formula [(LC)IrCp*L′]^q(where q is the charge 0 or +1

In the present study, the reaction of Et₃SiH with a variety of

alcohols, benzylic or aliphatic, was investigated using a 2-

phenylpyridine-derived solvato iridacycle prepared from

complex 1 (Scheme 1). The corresponding salts [2][BArF₂₄],¹³

[2][BF₄], and [2][OTf]²⁰were synthesized in acetonitrile by

abstraction of the chlorido ligand with Na[BArF₂₄] or with the

help of AgBF₄or AgOTf (Scheme 1). The structures of

[2][BArF₂₄] and [2][BF₄] were further established by X-ray

diﬀraction analysis (Figure 1). Table 1 gives all pertinent

structural X-ray diﬀraction acquisition and reﬁnement data

reported in this article. The structure of [2][BArF₂₄] displays

the longest cation−anion separation (d_Ir−B≈ 11.1 Å), arbitrarily

deﬁned here as the shortest distance between the central atoms

of each ion, i.e. the Ir atom for the cation and the B atom for

depending on the nature of L′, LC is a heterochelating ligand,

and Cp* is pentamethylcyclopentadienyl) belong to the class of

coordinatively saturated complexes that require the L′ ligand to

be displaced to become catalytically relevant and active. They

catalyze the hydrosilylation of imines¹³in the presence of the

−

BArF₂₄anion, i.e. tetrakis[3,5-bis(triﬂuoromethyl)phenyl]-

borate, for instance (cf. Scheme 1). The high potential of this

class of organoiridium complexes for hydrogen atom transfer

catalysis has also been documented by the groups of Pfeﬀer¹⁴

and Xiao.¹⁵Furthermore, under all-in-one-pot conditions, such

iridacycles are capable of orderly and selectively achieving the

hydroamination of alkynes and the subsequent hydrosilylation

of imine intermediates in methanol without interference of the

latter solvent.¹⁶In this article, we report on the structure of a

new (R₃Si)(H)Ir reactive intermediate arising from the reaction

of the 18-electron precursor¹⁷[2][BArF₂₄] with a silane

(Scheme 1). The investigation of the interaction of the 18-

electron complex [2][X] (X= BArF₂₄, BF₄, OTf) with silanes

and alcohols was carried out by focusing on a model catalytic

reaction that is widely documented: i.e., the O-dehydrosilyla-

tion of alcohols.¹⁸The aforementioned (R₃Si)(H)Ir inter-

mediate can reasonably be considered as the key catalytic

species of the fast room-temperature O-dehydrosilylation of

alcohols^18c,19and certainly of other hydrosilylation reactions

catalyzed by iridacycles of formula [(LC)IrCp*L′]^q.^13,16

Inspection of its electronic structure reveals a peculiar bonding

relationship within the M−Si−H motif.

−

the anion, within the crystal lattice. The BF₄salt displays

(Figure 1) the typical cation−anion separation already reported

for [2][OTf]²⁰(O as the central atom of the anion) and

[2][PF₆]²¹(P as central atom of the anion), which lies within

the 6.82−5.06 Å range.

¹H and ¹⁹F NMR diﬀusion-ordered spectroscopic (DOSY)

analyses of the structurally characterized [2][BArF₂₄] and

[2][BF₄] were carried out in d₅-PhCl (ε ≈ 5.6²²) and CD₂Cl₂

(ε ≈ 8.9²²) solutions at 298 K. Hydrodynamic diﬀusion

coeﬃcients were determined accurately by an analytical method

and from 2D-DOSY plots (cf. the Experimental Section and

Supporting Information). Both methods aﬀorded identical

values of cation and anion hydrodynamic diﬀusion coeﬃcients

D, which were based on ¹H and ¹⁹F NMR experiments,

respectively. For the measures carried out in d₅-PhCl, the

D

_cation/D_anionratio for [2][BArF₂₄] was found to be only slighly

higher than the theoretical value of 1 for an ideal tight ion pair

B

DOI: 10.1021/acs.organomet.6b00248

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

[2]⁺[BF₄]⁻and 75% in [2]⁺[BArF₂₄]⁻. This diﬀerence in

Coulombic contribution is consistent with the larger delocaliza-

−

tion of the charge density mostly at the F atoms in the BArF₂₄

23

−

anion in comparison to BF₄. The dispersion and orbital

interaction energy terms contribute respectively to 3% and 13%

of the attractive term ∑ΔE_attractivein [2]⁺[BArF₂₄]⁻and 16%

and 8% in [2]⁺[BF₄]⁻. The lower value of ΔE_int(solvation term

included) for [2]⁺[BArF₂₄]⁻(ΔE_int= −72 kcal/mol) in

comparison to [2]⁺[BF₄]⁻(ΔE_int= −99 kcal/mol) is consistent

with the former’s ability to dissociate more readily in

moderately polar solvents.

Catalytic Activity. Three [2][X] salts (X = BArF₂₄, BF₄,

OTf) were probed for their catalytic activity in the O-

dehydrosilylation of alcohols (Table 2 and Figure 3; cf. Figure

S37 in the Supporting Information). Et₃SiH was used

throughout this study. It is worth noting that other silanes

such as Ph₃SiH, PhSiH₃, and polymethylhydrosiloxane

(PMHS) proved much less reactive than Et₃SiH under the

conditions of catalysis (20 °C, small amounts of 1,2-

dichloroethane) (cf. Figure S34 in the Supporting Informa-

tion). All reactions were generally performed with 2/1 mixtures

of neat alcohol and Et₃SiH to which a limited amount (0.5 mL)

of 1,2-dichloroethane (DCE) was added to allow the full

solubilization of the catalyst and ensure homogeneity of the

reaction medium throughout the experiments. It is worth

noting that the use of MeCN instead of DCE at +20 °C fully

inhibited the production of H₂and the silylation of alcohol: no

signs of hydrosilylation of MeCN were perceptible under the

conditions of the catalysis. In general, addition of amounts as

low as 1 mol % (relative to silane) of a reasonable Lewis base

(i.e., any imine or aniline) would also inhibit alcohol O-

dehydrosilylation. The O-dehydrosilylation of primary benzylic

alcohols and methanol was achieved quantitatively within a few

seconds up to a few minutes at 20.0 0.1 °C with loadings in

[2][BArF₂₄] as low as 0.1 mol % (relative to Et₃SiH) (Table 2,

cf Figure S37). It is worth noting that [2][BF₄] and 1 proved to

be nearly catalytically inactive (Figure 3). As shown below, this

lack of catalytic activity of [2][BF₄] is not related to the ion-

pair separation. [2][OTf] displayed a catalytic activity similar to

that of [2][BArF₂₄] with TOF_ivalues higher than 10⁵h⁻¹for

PhCH₂OH and (4-MeO-Ph)CH₂OH (Table 2).

A partial second order in [2][BArF₂₄] for the reaction

between Et₃SiH and 2 equiv of benzyl alcohol was inferred

from a set of ﬁve catalytic runs carried out with diﬀerent

concentrations of [2][BArF₂₄] by the piezometric monitoring

of the pressure of H₂(or HD) developed in the overhead

volume of a sealed thermostated reactor as a function of time: a

linear curve ﬁt of ln ν_ivs ln c₂plot²⁸aﬀorded an order of 1.6(2)

(R²= 0.95) and a nonlinear curve ﬁt of a ν_ivs c₂plot (taking ν_i

= ac₂ⁿas a model, where n is the partial order in [2][BArF₂₄])

aﬀorded an order of 2.2(2) (R²= 0.99) in catalyst: ν_iis the

initial reaction rate and c₂the initial concentration in

[2][BArF₂₄] (cf. Figures S35 and S36 in the Supporting

Information).

Figure 1. Ellipsoid-type diagrams drawn at the 30% probability level

with partial atom numbering of the structures of (a) [2][BF₄] and (b)

[2][BArF₂₄]. The lattice CH₂Cl₂molecule was omitted for clarity in

(a). Selected interatomic distances (Å) for [2][BF₄]: Ir1−N1

2.096(2), Ir1−N2 2.044(3), B1−Ir1 5.088(2). Selected interatomic

distances (Å) for [2][BArF₂₄]: Ir1−N1 2.091(3), Ir1−N2 2.049(4),

B1−Ir1 11.079(3). The depicted anions are the ones nearest to the

considered cation in the crystal lattice.

model (analytical D_cation/D_anion= 1.11(1), 2D-DOSY plot

D

_cation/D_anion= 1.1), whereas for [2][BF₄] the ratio was closer

to 1 (D_cation/D_anion= 1.01(1), 2D-DOSY plot D_cation/D_anion

=

1.0).²³In CD₂Cl₂, a slightly more polar solvent, the qualitative

values of D_cation/D_anionratios extracted from 2D-DOSY plots

were 1.2 and 0.9 for [2][BArF₂₄] and [2][BF₄], respectively. A

¹⁹F−¹H HOESY experiment carried out with [2][BArF₂₄] at

298 K in CD₂Cl₂displayed no mutual through-space

correlation between the signals of the two ions, which indicates

that within the time scale of NMR spectroscopy the ion pair is

indeed dissociated according to Pregosin’s criteria.²³These

−

results indicate that the ion pairings in the latter BArF₂₄and

−

BF₄salts are not largely diﬀerent in a solvent of moderate

polarity such as d₅-PhCl and that ion-pair dissociation is only

slightly enhanced in a more polar solvent such as CD₂Cl₂

because the values of the D_cation/D_anionratio remain close to 1.

Static DFT calculations of idealized ion pair dissociation

enthalpies²⁴ΔH_cp→∞(the cp index stands for contact ion pair

and ∞ stands for the dissociated pair) for [2][BArF₂₄]

(ΔH_cp→∞= +15, ΔG_cp→∞= −1 kcal/mol) and [2][BF₄]

(ΔH_cp→∞= +14; ΔG_cp→∞= +2 kcal/mol) with a conventional

conductor-like screening model of solvation²⁵(COSMO) in

PhCl (ε = 5.62,²²T = 298.15 K) merely²⁶suggest that the

slightly more favored dissociation of BArF₂₄⁻should be entropy

Comparative catalytic runs carried out with MeOD(or H),

Et₃SiH, and [2][BArF₂₄] were also monitored by piezometry to

determine the kinetic isotope eﬀect (KIE²⁹) k_H/k_D(based on

initial reaction rates; cf. Figure S38 in the Supporting

Information). A value of 2.7 0.2 for the KIE suggests that

the rate-determining step is the RO−H bond cleavage in the

last step of formation of MeOSiEt₃that most likely implies

reaction of a [RO(H)(SiEt₃)]⁺species with an Ir-bound

hydrido ligand (vide infra).^18cSimilar experiments carried out

−

grounded in comparison to BF₄(Figure 2). If one considers

the total cation−anion interaction energy ΔE_intas deﬁned by

the energy decomposition analysis²⁷scheme, i.e. ΔE_int

=

+

(ΔE_Pauli

)

+ (ΔE_coul+ ΔE_orb+ ΔE_disp

)

= ΔE_Pauli

repulsive

attractive

∑ΔE_attractive, most of the attractive energy contribution in the

ion pair arises from the Coulombic (ΔE_coul) energy term: 83%

(ΔE_coul× 100/∑ΔE_attractive) of the attractive term in

C

DOI: 10.1021/acs.organomet.6b00248

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

Table 1. Structural X-ray Diﬀraction Acquisition and Reﬁnement Data

[2][BF₄]

[2][BArF₂₄]

[3][BArF₂₄]

[5][OTf]

formula

mol wt

habit

C₂₃H₂₆IrN₂BF₄·CH₂Cl₂

694.39

C₃₂H₁₂BF₂₄·C₂₃H₂₆IrN₂

1385.88

C₃₂H₁₂BF₂₄·C₂₇H₃₉IrNSi ·0.5C₆H₁₄C₄₂H₄₇Ir₂N₂·CF₃O₃S ·2CH₂Cl₂

1504.19

1283.14

yellow prism

0.30 × 0.25 × 0.20

monoclinic

yellow plate

yellow block

0.38 × 0.35 × 0.30

triclinic

orange prism

0.40 × 0.35 × 0.30

monoclinic

P2₁/c

cryst size (mm)

cryst syst

space group

a (Å)

0.22 × 0.20 × 0.06

triclinic

P2₁/c

P1

̅

P1

̅

14.6676(5)

12.8467(5)

13.2327(5)

17.0526(6)

91.546(1)

97.457(1)

107.091(1)

2740.94(18)

2

12.614(5)

13.698(5)

19.789(5)

81.064(5)

71.685(5)

79.438(5)

3173.5(19)

2

10.9779(9)

23.2754(19)

21.0040(13)

b (Å)

12.8597(4)

c (Å)

15.8707(4)

α (deg)

β (deg)

γ (deg)

V (Å³)

Z

119.400(2)

120.899(3)

2608.02(14)

4605.1 (6)

4

D_x(Mg m⁻³

)

1.768

1.679

1.574

1.851

T (K)

173(2)

θ_max, θ_min(deg)

μ (mm⁻¹

32.0, 1.6

31.1, 1.6

32.1, 1.5

34.0, 2.3

)

5.37

2.56

2.23

6.11

h,k,l range

−21/21, −19/17, −23/23

−10/18, −19/12, −24/24

−18/18, −20/20, −29/29

−14/17, −36/35, −33/13

no. of measd rﬂns

no. of indep rﬂns

no. of rﬂns with I > 2σ(I)

no. of params

R_int

R(F²> 2σ(F²))

R_w(F²)

35077

9049

7606

313

42350

17435

14177

739

83067

22017

19600

820

74913

18826

15330

555

0.025

0.030

0.025

0.047

0.023

0.033

0.041

a

b

c

d

0.063

0.124

0.079

0.083

S

1.11

1.02

1.04

1.15

a

b

c

2

w = 1/[σ²(F_o) + (0.0174P)²+ 6.0988P], where P = (F_o²+ 2F_c²)/3. w = 1/[σ²(F_o) + (0.0693P)²+ 4.0848P], where P = (F_o²+ 2F_c²)/3. w = 1/

d

[σ²(F_o) + (0.0415P)²+ 2.3317P], where P = (F_o+ 2F_c²)/3. w = 1/[σ²(F_o) + (0.021P)²+ 18.0829P], where P = (F_o+ 2F_c²)/3.

2

Figure 2. COSMO (PhCl) singlet ground state geometries (ZORA-PBE-D3(BJ)/ae-TZP) of ion pairs (a) [2][BF₄] and (b) [2][BArF₂₄] (left) with

associated maps of electrostatic potential (MEP) drawn on the isosurface (0.05 e bohr⁻³) of the self-consistent ﬁeld electron density (right).

Interatomic distance Ir−B for [2][BF₄]: 5.341 Å. Shortest F−Ir distance for [2][BF₄]: 4.667 Å. Interatomic distance Ir−B for [2][BArF₂₄]: 8.322 Å.

Shortest F−Ir distance for [2][BArF₂₄]: 5.654 Å.

with Et₃Si-D (or -H) produced an inverse KIE k_H/k_Dof 0.73

0.01, the origin of which was not assigned at the present stage

of our investigations (cf. Figure S39 in the Supporting

Information). Such inverse KIE in metal-promoted Si−H

bond activation of silanes²⁹has been assigned in one case to the

Si−H bond activation assistance by remote groups.³⁰

Investigation of the Reaction of [2][X] with Et₃SiH.

The fate of [2][BArF₂₄] in catalysis, which in principle requires

the departure of the bound CH₃CN molecule to allow Et₃SiH

to interact with the Ir(III) center, was investigated by various

1

methods: i.e., H and ²⁹Si NMR spectroscopic monitoring and

31

1

isothermal titration calorimetry (ITC ) at 20.0 °C. H NMR

D

DOI: 10.1021/acs.organomet.6b00248

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

a solution of [2][BArF₂₄]. With 2 equiv of Et₃SiH/Ir the ratio

(2.6:1) was again favorable to [3][BArF₂₄], the presence of

(Et₃Si)₂NEt was noted, and traces of unreacted [2][BArF₂₄]

were detected. These results suggested at ﬁrst that [5][BArF₂₄]

might be a resting state possibly resulting from the reaction of

transient 4 with [2][BArF₂₄] (eq 1) or from a reaction of

[3][BArF₂₄] with [2][BArF₂₄] (eq 2).

Table 2. Catalytic Performance of [2][BArF₂₄] in the

Dehydrosilylation of Selected Alcohols with Et₃SiH

a

yield

b

c

d

alcohol

T (min)

(%)

v_i(10⁻²M s⁻¹

)

TON TOF_i(h⁻¹

)

PhCH₂OH

1

100

8.39

8.87

1040

832

163636

192100

(4-MeO-Ph)

CH₂OH

[2]⁺+ 4 → [5]⁺+ MeCN

[2]⁺+ [3]⁺→ [5]⁺+ [MeCN‐SiEt₃]⁺

(4-iPr-Ph)

1

100

2.90

934

58004

(1)

CH₂OH

MeOH

10

100

43

1

2.69

3.00

0.25

1008

450

12

38504

8266

(2)

n-C₁₀H₁₁OH

1-(naphth-2-

yl)ethan-1-

ol

Isolation of [5][BArF₂₄] was attempted by crystallization

from a reaction aliquot at low temperature, which aﬀorded two

sets of moisture-sensitive red-orange ([5][BArF₂₄]; Scheme 1

and the Supporting Information) and pale yellow crystals

([3][BArF₂₄]; vide infra). Owing to the low quality of the

grown crystals, only a preliminary structure of [5][BArF₂₄]

could be obtained, showing a ca. 3.2 Å separation of the two Ir

centers (cf. the Supporting Information). As shown below,

[5][X] is a side product in all of the reactions of [2][X] with

Et₃SiH in CD₂Cl₂.

a

Yield in evolved H₂for the following conditions: 9.5 mmol of alcohol,

4.8 mmol of Et₃SiH, 0.1 mol % of [2][BArF₂₄] (6.7 mg), 0.5 mL of

b

c

1,2-dichloroethane, 20

turnover number, calculated at the highest conversion as the molar

ratio silylated alcohol:catalyst. TOF_i: TON/time or turnover

frequency calculated at a reaction time corresponding to 10% of

0.1 °C. v_i= −(d[Et₃SiH]/dt)_i. TON:

d

conversion.

It must be noted that addition of 2 equiv of 4-N,N-

dimethylaminopyridine (DMAP) to a solution containing

[3][BArF₂₄] and [5][BArF₂₄] in a 2.6:1 ratio resulted in the

formation of 4 and in the increase of the relative amount of

[5][BArF₂₄] (4:[5][BArF₂₄] ≈ 1:1, see Figure S11a in the

Supporting Information). Similarly, addition of excess amounts

of [nBu₄][OTf] to a 7.7:1 mixture of [3][BArF₂₄] and

[5][BArF₂₄] in CD₂Cl₂caused the full consumption of

[3][BArF₂₄] to yield a new mixture containing 4 and

[5][BArF₂₄] in a 50:1 ratio and Et₃SiOTf (see Figure S11b

in the Supporting Information). These two experiments

indicate that the triethylsilylium moiety can be readily

abstracted from [3]⁺by Lewis bases.

¹H NMR monitoring of the reaction of 2 equiv of Et₃SiH

with [2][BF₄] (cf. Figures S21 −S28 in the Supporting

Information) in CD₂Cl₂indicated the swift formation of

complexes 4 and [5][BF₄] in a 2.6:1 ratio and the formation of

Et₃SiF, the NMR signature of which matched literature data³⁵

(δ(²⁹Si) 33.1 ppm, δ(¹⁹F) −176.9 ppm, J_Si−F= 287 Hz). The

formation of 4 can be explained^18c,36by the abstraction of one

Figure 3. Pressure of hydrogen (bar) as a function of time (s):

comparison of the catalytic performance of DCE (0.5 mL) solutions of

[2][BArF₂₄], [2][BF₄], [2][OTf], [5][OTf], and 1 (4.8 μmol) in the

O-dehydrosilylation of benzyl alcohol (1 mL, 9.5 mmol) with Et₃SiH

(0.8 mL, 5.0 mmol). The molar ratio catalyst:alcohol:Et₃SiH was

1:2000:1000 in all experiments.

ﬂuoride from BF₄by the Lewis acidic³⁷silylium fragment

−

Et₃Si⁺of putative transient [3][BF₄]. Very similar behavior was

observed in the reaction of [2][OTf]²⁰with 2 equiv of Et₃SiH.

¹H NMR spectroscopic monitoring (cf. Figures S29 and S30 in

the Supporting Information) of the reaction revealed the rapid

formation of 4 and [5][OTf] in a 1.1:1 ratio and the presence

of Et₃SiOTf, the spectroscopic signature of which was ﬁrmly

established by comparison with that of an authentic sample.

Rather interesting is the catalytic performance of [5][OTf]

displayed in Figure 3, which is comparable to that of reference

compound 1 and much lower by far than that of [2][OTf].

Treatment of a CD₂Cl₂solution of [5][OTf] with excess

amounts of Et₃SiH (6 equiv/Ir) in a sealed NMR sample tube

monitoring of the reaction of [2][BArF₂₄] in CD₂Cl₂with

excess Et₃SiH (6 equiv/Ir) revealed the swift and exothermic in

situ formation of (Et₃Si)₂N-Et:³²i.e., the product of the double

hydrosilylation of CH₃CN.³³More importantly, the H NMR

1

spectrum revealed the complete conversion of [2][BArF₂₄]

mainly into [3][BArF₂₄] (Scheme 1), which is characterized by

a hydridic signal at −11.7 ppm showing up about 3.5 ppm

downﬁeld of the typical signal of the neutral hydrido complex 4

reported by Norton and co-workers.²¹¹H and ²⁹Si HMQC

NMR experiments (cf. Figures S12 −S17 in the Supporting

Information) conﬁrmed the coupling existing between the Si

nucleus (δ(²⁹Si) 7.0 ppm) and the hydridic proton (δ(¹H)

−11.7 ppm) of [3][BArF₂₄] with a Si−H coupling constant

J_Si−Hof 20 Hz, suggesting a rather weak H−Si interaction.^1b,34

It was also noted that another new species was present in

addition to [3][BArF₂₄] (Scheme 1), i.e. [5][BArF₂₄],

displaying a typical hydridic proton signal at around δ−23

ppm. Under these conditions (6 equiv of Et₃SiH/Ir) the

[3][BArF₂₄]:[5][BArF₂₄] ratio was 1.5:1. The latter ratio

reversed to 1:1.3 when only 3 equiv of Et₃SiH was added to

1

led according to H NMR spectroscopy to the formation of

very low amounts of 4, the [5][OTf]:4 ratio being

approximately 4:1 (ca. one-ninth of the iridacyclic content of

[5]⁺converted into 4). It was also noted that the characteristic

38

1

Si-H H NMR signal of Et₃SiH at ca. δ 3.65 ppm vanished.

New peaks assigned to dissolved H₂and possible Et₄Si³⁹were

observed at δ 4.59 (singlet) ppm and at δ 0.94 (triplet) and

0.54 (quartet) ppm, respectively. It is speculated that upon

reaction of excess Et₃SiH with [5]⁺the release of Et₃SiOTf

E

DOI: 10.1021/acs.organomet.6b00248

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

(and 4) promotes the formation of the putative transient

[(Et₃Si)₂H][OTf], a stabilized form of the highly electrophilic

silylium Et₃Si⁺cation according to Heinekey^37cand Reed,^37a,b

which is responsible for the transformation of Et₃SiH into

Et₄Si.^37c

Structure and Bonding in [5]⁺. Compound [5][OTf] was

synthesized by reaction of [2][OTf] with excess Et₃SiH and

fully analytically characterized (cf. Figures S31 −S33 in the

Supporting Information). Its structure was resolved by X-ray

diﬀraction analysis. Figure 4 displays the ellipsoid-type diagram

Figure 5. (a) ETS-NOCV analysis of the bonding interaction between

closed-shell fragments 4 and [6]⁺in their prepared geometry in [5]⁺.

The deformation density isosurface plot Δρ₁(0.005 e bohr⁻³) is

associated with an orbital interaction energy of −48 kcal/mol. Electron

density transfer operates from the red to the blue volumes upon

interaction. (b) ADFview2013 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 a singlet ground state model of [5]⁺. Red

(attractive) and blue (Pauli repulsion or nonbonded van der Waals)

colors are associated with negatively and positively signed terms,

respectively.

Bonding within cation [5]⁺was further analyzed assuming an

equivalent [4_H→6_Ir]⁺donor−acceptor pair formulation, where

6 is the 16-electron cationic iridacycle resulting from the

departure of MeCN from [2]⁺and H and Ir indexes specify

atoms involved in the main bonding interactions taking place

between fragments 4 and [6]⁺. This fragmentation scheme,

although artiﬁcial, is of great use to outline the peculiar

propensity of 4 to form donor−acceptor complexes⁴¹(vide

infra). Extended transition state-natural orbital for chemical

valence⁴²(ETS-NOCV) analysis was performed to visualize

electron density transfer upon orbital interaction between the

latter two fragments in their so-called prepared geometry: that

is, the geometry preceding the establishment of interfragment

bonds leading to [5]⁺. The two fragments were deﬁned as

resulting from the cleavage of the longest Ir−H distance (1.743

Å) in [5]⁺. It was found that the main orbital interaction energy

component (−48 kcal/mol) arises from the interaction

materialized by density deformation Δρ₁(Figure 5a), which

represents 61% of the interfragment orbital interaction energy

(ΔE_orb= −77 kcal/mol). The electron density transfer

associated with Δρ₁entails donation from d orbitals at Ir and

from the s orbital at the hydrido H atom. Furthermore, natural

bonding orbital (NBO) analysis of the Lewis structure of [5]⁺

provided an explicit formulation for one Ir−H natural bonding

orbital as a linear combination of atomic orbital components.

The electron occupation of the latter fell above the standard

threshold value of 1.5 electrons, below which bonding

interactions are not considered within the NBO valence

scheme. This natural bonding orbital located on the shortest

interatomic H−Ir segment is formulated as ψ_Ir−H(1.545 e) =

0.71(s)_H+ 0.70(sd^2.59)_Ir.

Figure 4. Ellipsoid-type diagram of the structure of [5]⁺drawn at the

30% probability level with partial atom numbering. Note that the

hydridic H1A atom was located from Fourier diﬀerence maps and its

position reﬁned. Dichloromethane molecules and TfO⁻anion were

omitted for clarity. Selected interatomic distances (Å) and angles

(deg): Ir1−H1 1.63(6), Ir2−H1 1.79(6), C1−Ir1 2.037(4), N1−Ir1

2.082(3), C22−Ir2 2.040(4), N2−Ir2 2.079(3), Ir1−H1−Ir2 155.1(4).

of the structure of cation [5]⁺. In this complex the position of

the bridging hydrido ligand was determined from Fourier

diﬀerence maps and reﬁned. The Ir1−H1 and Ir2−H1

distances are slightly diﬀerent and amount respectively to

1.63(6) and 1.79(6) Å. This apparent diﬀerence nonetheless

lies within the limits of accuracy deﬁned by standard deviation

and could be an artifact of the localization from Fourier

diﬀerence maps. In a perfectly symmetric molecule a C₂

rotation axis would be positioned at the bridging hydrogen

atom somewhat parallel to the Ir−N bonds and would bisect

perpendicularly the Ir1−Ir2 segment. The structure drawn in

Figure 4 indicates π−π stacking interactions with a pyridyl−

pyridyl intercentroid distance of ca. 3.4 Å, a structural feature

similar to that reported previously for another catalytically

inactive μ-hydrido-bridged bis-ruthenacycle.⁴⁰A rationale

similar to that proposed for the formation of [5][BArF₂₄] can

be made for [5][OTf]; the latter stems from a reaction of 4

with [2][OTf] with the concomitant displacement of MeCN

and its capture by Et₃SiOTf in the early stages of the reaction.

Symmetry-unconstrained DFT geometry optimization car-

ried out with a gas-phase singlet ground state model of [5]⁺

(ZORA-PBE-D3(BJ)/ae-TZP level) produced a more sym-

metric geometry in which the Ir−H−Ir motif displays Ir−H

distances of the same order (1.743 and 1.737 Å, Figure 5). The

π−π interaction is locally materialized by blue NCI isosurfaces

in Figure 5b, depicting nonbonded van der Waals contacts.

Proposed Catalytic Cycle. The results disclosed indicate

complex [3]⁺to be the most direct product of the activation of

[2]⁺by Et₃SiH. As such, its capability to readily release the

Et₃Si⁺cation is central to the catalysis; for the O-silylation of

alcohols to take place, the silylium cation must be transferred

from [3]⁺in the early stages of the reaction in a way similar to

that depicted by Brookhart and co-workers for the catalyzed

F

DOI: 10.1021/acs.organomet.6b00248

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

Scheme 2. Proposed Catalytic Cycle for the O-Dehydrosilylation of Alcohols Promoted by [2][BArF₂₄]

Figure 6. Calorigram of the isothermal (T = 293.15 K) titration (ITC) of [2][BArF₂₄] in PhCl (c = 1.0 mmol L⁻¹) by sequential injections (v = 2

μL, time delay between injections 1500 s) of Et₃SiH (c = 78.4 mmol L⁻¹).

cleavage of alkyl ethers with Et₃SiH.⁴³Whether the transfer

occurs within the coordination sphere of the Ir center in [3]⁺

with the concerted deprotonation of the [R-O(H)-SiR′₃]⁺

intermediate or by intermolecular means in the solution bulk

has not been settled. Scheme 2 depicts a simpliﬁed outer-sphere

mechanism⁴⁴that accounts for the KIE observed with

MeOD(H). The renewal of [3][BArF₂₄] in the cycle results

from reaction of the putative transient [6][BArF₂₄] with

Et₃SiH. [6][BArF₂₄] can be produced by the reaction of 4 with

the Brønsted acid [R-O(H)-SiR′₃]⁺. Complex [5][BArF₂₄] can

therefore arise from a reaction of 4 with the putative transient

[6][BArF₂₄] or its solvated form [6·S][BArF₂₄] (S = alcohol,

alkoxytriethylsilane, etc.). Even though [5]⁺was shown to

produce low amounts of 4 in the presence of Et₃SiH, the poor

catalytic activity of this dinuclear species seemingly disqualiﬁes

it for any major contribution to the O-dehydrosilylation

reaction (Figure 3).

here) devoid of [3]⁺involving the reaction of the in situ formed

Et₃SiOTf with the alcohol can be envisaged.

Enthalpy of Activation of [2][BArF₂₄]. Isothermal

titration calorimetry (ITC) experiments were carried out in

PhCl at 20 °C. The measurements were performed by the

sequential addition of small volumes of a PhCl solution of

Et₃SiH into 1 mL of a solution of [2][BArF₂₄]. Under such

conditions the side formation of [5][BArF₂₄] was expected to

be minimized. Three experiments were run independently; they

all indicated that the thermochemical stationary state was

reached with 3 equiv of Et₃SiH with an associated (exothermic)

enthalpy of reaction of −46

3 kcal/mol (Figure 6). This

value of enthalpy can be decomposed as the sum of three main

thermochemical contributions (Scheme 3): (1) the reaction of

[2][BArF₂₄] with Et₃SiH leading to [3][BArF₂₄], MeCN,

[5][BArF₂₄] and the possible partial conversion of [3]⁺into 4

and the silylium-chlorobenzene cation^37c[(Et₃Si)(PhCl)]⁺

(ΔH₁), (2) the catalyzed hydrosilylation of the released

MeCN into (Et₃Si)NC(H)Me (ΔH₂), and (3) the hydro-

silylation of the latter into (Et₃Si)₂NEt (ΔH₃).

The origin of the aforementioned kinetic second order in

[2][BArF₂₄] determined from initial reaction rates cannot be

assigned with certainty at this stage. Finally, in the case of

[2][OTf] a similar outer-sphere catalytic cycle (not shown

Computation of the thermochemistry²⁶of the last two steps

was carried out at the B3LYP-NL/def2-QZVP level with zero-

G

DOI: 10.1021/acs.organomet.6b00248

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

Scheme 3. Dominant ΔH₂and ΔH₃Contributions among

the Three Thermochemical Contributions Identiﬁed for the

Activation of [2][BArF₂₄] by Et₃SiH in PhCl, According to

COSMO-RS/DFT-D Computations

point energy and solvation corrections (see Computational

Details) using the COSMO-RS⁴⁵procedure. The computed

values of ΔH₂(−16.5 kcal/mol) and ΔH₃(−28.1 kcal/mol)

suggest a dominant thermochemical contribution of the

hydrosilylation of MeCN to the total enthalpy of the reaction.

For convenience, the enthalpy of the single reaction of

[2][BArF₂₄] with Et₃SiH leading to [3][BArF₂₄] and MeCN,

i.e. ΔH₁*, was computed by substituting [BArF₂₄]⁻by [BF₄]⁻.

Assuming contact ion pairs for [2][BF₄] and [3][BF₄], ΔH₁*

was found to be slightly endothermic with a value of +1.2 kcal/

mol (ΔG₁* = +4.5 kcal/mol); the total computed reaction

enthalpy amounted to ΔH_tot= ΔH₁* + ΔH₂+ ΔH₃= −43.5

kcal/mol (ΔG_{298 K}= −24 kcal/mol). It must be noted that

performing the calculation without inclusion of the counter-

anion had a minor eﬀect on the value of ΔH₁*. According to

DFT calculations, it can be stated that the catalyzed

hydrosilylation of MeCN assists the departure of MeCN

from [2]⁺and enables the formation of [3]⁺, which is

thermodynamically not favored according to the computed

positive value of ΔG₁*.

Figure 7. (a) Ellipsoid-type diagram of the structure of [3]⁺drawn at

the 30% probability level with partial atom numbering. Note that the

hydridic H1A atom was located from Fourier diﬀerence maps and its

position frozen during reﬁnement. A pentane molecule and the

BArF₂₄⁻anion were omitted for clarity. Selected interatomic distances

(Å) and angles (deg): Ir1−Si1 2.5008(8), Ir1−H1A 1.47, Si1−H1A

2.10, N1−Ir1 2.0754(19), C11−Ir1 2.089(2), Ir1−C13 2.226(2), Ir1−

B1 10.077(2), N1−Ir1−Si1 99.78(6), C11−Ir1−Si1 80.70(6), N1−

Ir1−H1A 76.2, N1−Ir1−C11 77.60(8). (b) Gas-phase singlet ground

state geometry of [3]⁺computed at the ZORA-PBE0-dDsC/ae-TZP

level. Selected interatomic distances (Å): Ir−Si 2.514, Ir−H 1.583, Si−

H 1.913, C_Ar−Ir 2.056, N−Ir 2.069.

Structural Characterization of [3]⁺. Owing to its high

reactivity and sensitivity to moisture, [3][BArF₂₄] could not be

isolated and fully characterized analytically. Nonetheless, as

already mentioned above, crystals were grown by the solvent

diﬀusion technique at −20 °C directly from reaction aliquots,

thus aﬀording samples suitable for X-ray diﬀraction analysis.

Close analysis of diﬀraction data of two diﬀerent crystals

suggested that [3]⁺exists in two distinct diastereomeric forms

that diﬀer in the position of the SiEt₃group and the hydridic H

atom relative to the iridacycle. Those two isomers are

isoergonic (298.15 K) within 1 kcal/mol according to gas-

phase singlet ground state DFT calculations. According to

NMR data that display a single signal for the hydridic proton

around δ −11.7 ppm, it is speculated that the two isomers [3]⁺

are either engaged in a fast mutual exchange or that their

distance of ca. 1.47 Å is akin to that reported for complex 4

(1.52 Å) by Norton and co-workers.²¹

Investigation of the Structure of [3]⁺by DFT Methods.

To establish the bonding relationship within the critical Ir−Si−

H motif of [3]⁺, DFT investigations were carried out using

DFT-D functionals⁴⁷that reproduce the eﬀect of nonlocal

dispersion interactions.^40,48Computations carried out with

generalized gradient approximation (GGA), meta-GGA, and

hybrid functionals (ZORA PBE-D3(BJ), TPSS-D3(BJ), and

PBE0-dDsC, respectively) associated with a Slater-type basis set

(TZP) produced geometries displaying a good match with the

X-ray diﬀraction based structure (Figure 7b). Geometry

optimizations using GGA and hybrid functionals (PBE-D3

and PBE0-D3, respectively) associated with an eﬀective core

potential Gaussian-type basis set (ECP SDD) performed

satisfactorily as well. The map of electrostatic potential

(MEP) shown in Figure 8a clearly shows that the Si atom

carries the greatest charge density depletion in [3]⁺, which

according to natural population analysis⁴⁶(NPA; Figure 9)

assigns to the Si atom the highest partial positive natural charge.

Figure 8b displays noncovalent interaction (NCI) regions

1

respective H NMR spectra coincide. Figure 7a displays the

structure of [3][BArF₂₄] above its computed model (Figure

7b), which displays a geometry consistent with the X-ray

diﬀraction data. Incomplete structural data of the other

diastereomer of [3]⁺(named as [3′][BArF₂₄]) are provided

in the Supporting Information. The coordination geometry of

[3]⁺suggests a heptacoordinate Ir center displaying the main

feature of a SISHA-type adduct:⁸i.e., an experimental Si−H

distance (ca. 2.10 Å) slightly longer than the generally accepted

limiting value of 1.9 Å.^1bThe Ir−Si distance of ca. 2.50 Å is

only slightly longer than that reported by Bergman, Tilley, and

co-workers for [(Ph₃Si)(Me₃P)IrCp*]⁺(2.41 Å).⁵The Ir−H

H

DOI: 10.1021/acs.organomet.6b00248

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

Figure 9. (top) Wiberg bond indexes and NPA⁴⁶charges at atoms of

interest computed at the ZORA-PBE-D3(BJ)/ae-TZP level. (bottom)

ADFview2013 isosurface plot of deformation density Δρ₁(0.005 e

bohr⁻³) produced by an ETS-NOCV⁴²analysis according to the

realistic fragmentation scheme chosen for [3]⁺. Electron density

transfer operates from the red to the blue volumes upon interaction.

Figure 8. (a) Map of electrostatic potential (MEP) drawn on the

isosurface (0.05 e bohr⁻³) of the self-consistent ﬁeld electron density

of a gas-phase singlet ground state model of [3]⁺. (b) ADFview2013

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. Red

(attractive) and blue (Pauli repulsive or nonbonded van der Waals)

isosurfaces are associated with negatively and positively signed terms,

respectively.

at BCPs of Ir−H and Ir−Si bonds (Table 3) is a known

behavior of QTAIM for heteropolar bonds.⁵¹It is also generally

considered that the Laplacian for such bonds can adopt either a

positive or negative value.^48cThe delocalization index δ(A,B),⁵²

which provides a measure of electron delocalization probability

between two atomic basins, follows the trend of Wiberg indexes

(Figure 8 and Table 3) for the Ir−Si−H motif: δ(Ir,H) >

δ(Ir,Si) ≫ δ(H,Si) (Table 3). Its value clearly suggests a

stronger Ir−H bond and a weaker Ir−Si interaction. The low

but signiﬁcant value of δ(H,Si) suggests an evanescent

interaction that cannot be dismissed as unimportant to the

cohesion of the complex. The value of the bond ellipticity ε^11,53

for the Ir−H bond (ε ≈ 0, Table 3) is consistent with a

cylindrically symmetric bond and a more localized bonding

interaction directed toward the Ir center. In the case of the Ir−

Si bond the value of the elllipticity (0.5 < ε < 1, Table 3) might

indicate delocalization of the respective bonding electrons. On

the main QTAIM features, all computations suggest a strong

Ir−H bond, a weaker Ir−Si bond, and the absence of a

signiﬁcant covalent H−Si interaction. Finally, Figure 10 displays

the so-called localization domains for two diﬀerent isosurface

plots of the electron localization function⁵⁴(ELF) η(r) (values

of η(r) greater than 0.5 usually denote larger electron

localization).⁵⁵Monosynaptic core (C(Ir), C(Si), and C(C)),

disynaptic and polysynaptic basins (V(C,H), V(Ir,Si), and

V(Ir,H)) are drawn at two values of the ELF (η(r) = 0.55 and

0.70 in Figure 10). At values 0.55 < η(r) < 0.7, the ELF

disynaptic basin located at the Ir−Si segment merges partially

with the Ir,H valence basin. Figure 10 displays the same

disynaptic basin located at the Ir−Si segment for η(r) = 0.7,

which further supports the hypothesis of a dominant dative Ir→

determined on the basis of the reduced density gradient

method.⁴⁹Apart from the blue isosurfaces that depict

nonbonded van der Waals interactions and Pauli repulsion

areas, the typical red NCI isosurface ring that surrounds the Si−

Ir bond and the Si−H interaction within a so-called covalent

trough^48eindicates that the Si center interacts covalently with

the Ir center.

Topological analysis of the electron density following the

quantum theory of atoms in molecules¹¹(QTAIM from the

AIMAll⁵⁰software) was performed at the PBE0-D3 and PBE-

D3 levels with eﬀective core potential (ECP) Gaussian-type

Stuttgart/Dresden (SDD) contracted and diﬀuse basis sets.

Table 3 gives the main topological electron density parameters

for the Ir−Si−H motif: i.e., the electron density ρ_βat bond

critical points (3,−1), the Laplacian of the density ∇²ρ_β, the

bond ellipticity (ε), and the interatomic electron delocalization

index (δ(A,B)). This QTAIM analysis reveals that bond critical

points (BCP (3,−1)) exist for Ir−Si and Ir−H bond paths

(Table 3). However, no BCP (3,−1) and bond path were found

for a Si−H interaction. Identical conclusions were drawn from

QTAIM investigations carried out with hybrid PBE0-dDsC,

meta-GGA TPSS-D3(BJ), or GGA PBE-D3(BJ) functionals

associated with Slater-type basis sets of various sizes and

number of polarization functions (with or without scalar ZORA

correction, using all-electron TZP or TZ2P basis sets ﬁt with

diﬀuse functions in the case of the PBE0 functional). The

observed basis set dependence of the computed value of ∇²ρ_β

I

DOI: 10.1021/acs.organomet.6b00248

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

Table 3. QTAIM and NBO Analysis of [3]⁺Performed on Three Diﬀerent Geometries

a

QTAIM

∇²ρ_β

c

d

b

geometry opt level

analysis level

path

ρ_β

ε

δ(A,B)

NBO WBI

e

ZORA-PBE0-dDsC/ae-TZP

PBE0-D3/SDD/def2-TZVPP

PBE-D3/SDD/def2-TZVPP

ZORA-PBE0-dDsC/ae-TZP

PBE0-D3/SDD/6-311++G**

PBE0-D3/SDD/def2-TZVPP

PBE-D3/SDD/def2-TZVPP

ZORA-PBE0-dDsC/ae-TZ2P

ZORA-PBE0-dDsC/ae-QZ4P

Ir−Si

Ir−H

Si···H

Ir−Si

Ir−H

Si···H

Ir−Si

Ir−H

Si···H

Ir−Si

Ir−H

Si···H

Ir−Si

Ir−H

Si···H

Ir−Si

Ir−H

Si···H

0.0723

0.1618

f

−0.0689

0.0036

f

0.68

0.43

0.85

0.14

0.40

0.87

0.13

0.40

0.86

0.14

0.49

0.84

0.14

g

0.43

0.55

0.21

0.42

0.55

0.20

0.42

0.55

0.22

0.44

0.55

0.19

0.35

0.45

0.26

g

0.05

f

0.0741

0.1646

f

−0.0701

−0.0234

f

0.69

0.08

f

0.0728

0.1610

f

−0.0706

−0.0108

f

0.98

0.08

f

0.0720

0.1589

f

−0.0658

−0.0242

f

0.55

0.08

f

0.0734

0.1639

f

−0.0605

0.0111

f

g

0.0738

0.1632

f

0.0169

0.0041

f

g

a

QTAIM analysis: electron density (ρ_β), Laplacian of the electron density (∇²ρ_β) and bond ellipticity (ε) at bond critical points, delocalization index

b

c

δ(A,B) between the pair of atoms (A,B) (all values are given in au). Wiberg bond index (WBI), as obtained from NBO analysis. Level of theory

d

e

used in the structure optimization. Level of theory used in QTAIM and NBO analyses. Identical results were obtained with a 6-311+G** basis set.

f

g

No value because no bond critical point (3,−1) was localized for this interaction. Not computed.

density isosurface Δρ₁(corresponding to 68% of the total

orbital interaction energy ΔE_orb= −135 kcal/mol) suggests that

upon interaction between prepared geometries of 4 and Et₃Si⁺

electron density donation operates (1) from the Ir center and

to a lesser extent (2) from the Ir-bound H atom and chelate C_Ar

atom. This density donation operates toward a broad volume of

space (blue in Figure 9) surrounding the Ir−Si segment and the

σ Si−C bonds at the Et₃Si fragment.

This comprehensive analysis converges to a rather simpliﬁed

picture of the bonding situation within the Ir−Si−H motif: that

is, a donor−acceptor interaction between the Ir−H unit and the

Si atom with a largely dominant contribution of the Ir−Si

interaction and a minor contribution of the Si−H interaction.

The reactivity of [3]⁺toward Lewis donors serves as a base

to establish the formal oxidation state of the Ir center.

According to Tilley and Bergman, Ir(V) species might play an

important role in C−H and E−H activation reactions.⁵A priori,

[3]⁺could be formulated as an 18-electron [ML₃X₄]⁺Ir(V)

complex (of neutral equivalent formula [ML₂X₅]) according to

Green’s formalism.⁵⁶However, the fact that the silylium

fragment can putatively be abstracted by weak Lewis bases

Figure 10. Isosurface plots of the ELF⁵⁴of [3]⁺with values for η(r) of

0.55 (top) and 0.77 (bottom) with an enlarged view of the localization

basins for the Ir−Si−H motif.

57

such as BF₄and OTf⁻and also by the stronger DMAP to

give 4 is clear evidence that the Et₃Si moiety in [3]⁺is a Z

ligand and not an X ligand.⁵⁸The silylium Et₃Si⁺is indeed

isoelectronic with neutral boranes BR₃that are prototypical Z-

type ligands. Therefore, a [ML₃X₃Z] formulation would be

more appropriate for [3]⁺.^58aFurthermore, the electron density

donation from the Ir center toward the Si center in [3]⁺is too

weak to consider the Ir center as holding a +V formal oxidation

state according to DFT calculations. The diﬀerence of the sum

of atomic natural charges at the Et₃Si moiety in [3]⁺and at the

−

Si component within the slightly delocalized (Ir−H)→Si

interaction.

Extended transition state-natural orbital for chemical valence

(ETS-NOCV) analysis⁴²was carried out considering the

fragmentation scheme displayed in Figure 9. The latter

fragmentation scheme is not artiﬁcial though: it was chosen

on the basis of the experimentally established reactivity of [3]⁺

(vide supra; DMAP and TfO⁻quenching experiments) that

behaves as a source of electrophilic Et₃Si⁺moiety. This

“realistic” scheme entails a donor−acceptor pair formulation

such that [3]⁺≡ [4_Ir→SiEt₃]⁺. A plot of the deformation

+

“gas-phase” silylium, i.e. Δ∑q(SiEt₃) = [(∑q_i)_[3]− 1], where

q_istands for the natural charge of atom i in the Et₃Si moiety

and 1 is the charge of the gas-phase silylium, provides a hint to

the extent of the charge density transfer from the Ir toward the

Et₃Si moiety. It must be noted that natural charges (NPA)

J

DOI: 10.1021/acs.organomet.6b00248

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

follow properly the trend of the charge density and can be used

to characterize the overall electron density enrichment or

depletion at given molecular fragments.^48a,59The value of

piano stool subunits that are related by an inversion center

located at the bridging Ag atom are four-legged and the Ir

centers are formally heptacoordinated. Even though Templeton

et al. did not localize the two hydrido ligands in the published

X-ray diﬀraction structure, the position of the Ag atom in this

heterotrimetallic complex leaves no doubts about their

respective positions close to the silver center.

+

Δ∑q(SiEt₃) amounts to −0.16 ((∑q_i)_[3]= +0.84) for the

Et₃Si moiety in [3]⁺, which means that only one-sixth of the

positive charge of [3]⁺is borne by fragment 4 in [3]⁺. The

conversion of [2]⁺into [3]⁺can therefore be considered as

isohypsic at Ir: i.e., as taking place at a constant formal +III

oxidation state. Consequently, it is proposed that the

mechanism of formation of [3]⁺implies the hydride transfer

from Et₃SiH to the Ir center of transient [6]⁺and the concerted

bonding of the silylium cation to the new hydrido-iridium unit

(eq 3), rather than a conventional oxidative addition step, a

Having in scope the emerging property of Ir(III) complex 4

as a peculiar Lewis donor, we undertook a DFT study of

Templeton’s complex, looking for possible similarities with

[3]⁺. The gas-phase singlet ground state geometry of [4→Ag←

4]⁺was optimized at the ZORA-PBE-D3(BJ)/ae-TZP level

with C_isymmetry. Figure 11a displays the corresponding

geometry decorated with a NCI region plot. The main

interatomic distances match the structural X-ray diﬀraction

analysis data. More interesting are the Ir−H distances of ca.

1.66 Å and the hydridic H−Ag distances of 1.90 Å, which

corroborate Templeton’s report of a signiﬁcant J_H−Agcoupling

41

1

constant in the H NMR spectrum of [4→Ag←4]⁺. The

attractive NCI domain is located in the close vicinity of the

Ag−Ir segment; part of it contains a covalent trough that

surrounds the H−Ag and Ir−H segments, whereas the

remaining plain NCI isosurface somewhat eclipses the Ir−C_Ar

bond of the neighboring iridacycle.

hypothesis that is currently under investigation. The polar

character of the Si−H bond in Et₃SiH characterized by marked

negative natural charge (NPA computed at the ZORA-PBE-

D3(BJ)/ae-TZP level, q(Si-H) = −0.19) at the (Si-bound) H

atom and positive partial natural charge at the Si atom (NPA,

q(Si-H) = +1.22) creates a predisposition to hydride transfer

that is well documented.⁶⁰

Hydridoiridium(III) Complex 4, an Ambident Lewis

Donor. Our results echo recent claims by Templeton et al. on

the Lewis donor character of 4.⁴¹The latter authors reported

the adventitious formation of [4→Ag←4]⁺from the reaction of

4 with silver(I) salts. Several other examples of donor−acceptor

metal−silver and metal−metalloid complexes have been

reported by Braunschweig et al.⁶¹The crystal structure of

[4→Ag←4]⁺(Cambridge Structural Database refcode BIQ-

GOZ) reveals geometric features akin to those of [3]⁺: the

QTAIM analysis (ZORA-PBE0-DdSc/ae-TZ2P) indicates no

BCP (3,−1) and bond path in the Ir−Ag segments (Wiberg

bond index w = 0.10). Within the Ir−H−Ag motif the only

existing BCPs (3,−1) are for Ir−H (w = 0.50, ρ_β= 0.134, ∇²ρ_β

= 0.072) and Ag−H (w = 0.15, ρ_β= 0.069, ∇²ρ_β= 0.140) bond

paths. The last two interactions match the criteria of relatively

strong and weak polar bonds, respectively. Analysis of the NPA

natural charges (q_Ir= +0.47, q_H= −0.08, q_Ag= +0.65) indicates

that a moderate portion of the positive charge of this cation is

distributed over the two fragments 4 (+0.35, i.e. a charge of

+0.17 on each fragment 4), the remaining charge residing at the

silver(I) center. Standard ETS-NOCV⁴²analysis was performed

by assuming fragmentation of the molecule into the Ag⁺ion

and a ﬁctitious fragment made of the two molecules of 4 in

Figure 11. (a) Gas-phase singlet ground state geometry of Templeton’s [4→Ag←4]⁺complex optimized at the ZORA-PBE-D3(BJ)/ae-TZP level

decorated with an ADFview2013 plot of noncovalent interaction (NCI) regions (ZORA-PBE0-dDsC/ae-TZP) within [4→Ag←4]⁺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. Red (attractive)

and blue (Pauli repulsive or nonbonded van der Waals) isosurfaces are associated with negatively and positively signed terms, respectively. Selected

interatomic distances (Å) and angles (deg): Ir−H 1.661, H−Ag 1.902, Ir−Ag 2.736 (exptl ∼2.716), Ir−C_Ar2.058 (exptl 2.055), Ir−N 2.079 (exptl

2.031), Ag−Ir−CAr 67.8 (exptl 71.3), Ag−Ir−N 103.3 (exptl 108.2). (b) ADFview2013 isosurface plots of deformation densities Δρ₁, Δρ₂, and Δρ₃

(0.002 e bohr⁻³) produced by an ETS-NOCV⁴²analysis according to the depicted fragmentation scheme chosen for [4→Ag←4]⁺. Electron density

transfer operates from the red to the blue volumes upon interaction.

K

DOI: 10.1021/acs.organomet.6b00248

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

their prepared geometry and position in [4→Ag←4]⁺(Figure

11b). A density deformation plot of Δρ₁(33% of the total

orbital interaction energy ΔE_orb= −121 kcal/mol) corresponds

to a density donation mostly from the two Ir−H motifs toward

a broad area surrounding the silver atom. Δρ₂(8% of ΔE_orb)

corresponds to back-donation from a d orbital at silver toward

two areas located in the Ag−H segments, whereas Δρ₃(12% of

ΔE_orb) corresponds to density donation from a d orbital at Ir

atoms toward a broad area that embraces the Ir−Ag and H−Ag

segments. Combined theoretical analyses suggest indeed a

delocalized and NCI-supported donor−acceptor (Ir−H)→

Ag⁺←(Ir−H) interaction of overall weak covalent charac-

ter,^48b,c,62wherein the Ir−Ag interactions seem to be

dominantly noncovalent according to QTAIM analysis (no

BCPs, no bond paths). Having in sight the electronic structures

of [5]⁺, [3]⁺, and [4→Ag←4]⁺, and in full support of earlier

statements by Templeton et al.,⁴¹complex 4 can indeed be

considered as an ambident Lewis donor capable of binding Z-

type Lewis acceptors (Z = fragment [6]⁺in [5]⁺, Et₃Si⁺in [3]⁺,

and Ag⁺in [4→Ag←4]⁺) by the Ir atom, the hydridic hydrogen

atom, or both atoms of the Ir−H motif (Scheme 4).

contributes to cohesion. The fact that the silylium cation can

readily be transferred to a Lewis base (DMAP or TfO⁻) from

intermediate [3]⁺sheds a new light on the possible role of

similar Ir intermediates in the catalytic hydrosilylation of imines

promoted by similar iridacycles.^13,16Related to the Lewis pair

character of [3]⁺is the issue of the nature of the associated

anion. Quite tangible is the assumption that [3]⁺also forms in

the early stages of the reaction of [2][BF₄] or [2][OTf] with

Et₃SiH. The highly electrophilic silylium fragment of putative

[3][OTf] can be captured by any Lewis donors present in the

medium (TfO⁻, MeCN, alcohol, and any Lewis base) and at

least by the triﬂate anion itself if no better base is present,

leading to a still catalytically eﬀective 4/Et₃SiOTf combination.

Catalysis can be inhibited if abstraction of the Et₃Si⁺moiety

leads to inert X-SiR₃species devoid of the electrophilic

character (such as F-SiEt₃in the case of [2][BF₄]) necessary for

the activation of so-called organic hydride acceptors (ketones

and imines for example).⁶⁴Finally, the formation of [5]⁺is a

major downside of the considered catalytic system, for it diverts

portions of the iridacycle into a catalytically inactive form. It is

speculated that this issue can be addressed by proper molecular

engineering. Further catalytic applications of [2][BArF₂₄] and

similar iridacycles are being investigated and will be disclosed in

due time.

Scheme 4. Binding Sites at 4 for a Z Acceptor

EXPERIMENTAL SECTION

■

General Considerations. All experiments were conducted under a

dry argon atmosphere using standard Schlenk lines and dry glovebox

techniques. The following compounds were purchased from Sigma-

Aldrich: benzyl alcohol (99%), 2-phenylpyridine (abbrev. PhPy, 98%),

silver tetraﬂuoroborate (AgBF₄, 98%), d-triethylsilane (Et₃SiD, 97%

atom D), and triethylsilane (Et₃SiH, 99%). The following compounds

were purchased from Alfa Aesar: 4-isopropylbenzyl alcohol (90%), 4-

methoxybenzyl alcohol (98%), 1-(2-naphthyl)ethanol (97%), 1,2,3,4,5-

pentamethylcyclopentadiene (abbreviated Cp*, 94%), and sodium

tetrakis[3,5-bis(triﬂuoromethyl)phenyl]borate (abbreviated Na-

[BArF₂₄], 97%). Anhydrous sodium acetate (NaOAc, 99%), d-

methanol (MeOD, >99.5% D atom), 4-N,N-dimethylaminopyridine

(DMAP, 99%), and 1-decanol (97%) were purchased from Fluka. IrCl₃

was purchased from Pressure Chemical Co. Dicalite 4158 was

purchased from Carlo Erba Reagents. The iridium(III) complexes

CONCLUSION

■

65

[Cp*Ir(μ-Cl)Cl]₂and 1⁶⁶were prepared as reported in the

In summary, [2][BArF₂₄] is an eﬃcient catalyst of the O-

dehydrosilylation of primary alcohols, a reaction that can be

readily quenched by addition of Lewis bases. This result brings

a new light to our earlier observations of the lack of major

interference of the reaction solvent, i.e. methanol, in a tandem

(sequential) hydroamination−hydrosilylation of terminal al-

kynes promoted by a catalyst structurally similar to 1.¹⁶The

activation of [2][BArF₂₄] by Et₃SiH entails the formation of a

highly reactive hydrido-Ir(III)-silylium adduct, i.e. [3][BArF₂₄],

which putatively results from hydride transfer from the silane to

the iridium center with no change in the formal +III oxidation

state of the latter. The intermediate [3][BArF₂₄] has a bifaceted

behavior; it is a source of a highly electrophilic silylium and a

hydride donor. ITC proved to be a valuable analytical tool for

establishing not only the stoichiometry in Et₃SiH required for

the activation of [2][BArF₂₄] but also the enthalpy of activation

of [2][BArF₂₄] in PhCl. According to COSMO-RS/DFT

calculations, the activation of [2][BArF₂₄] is enabled by the

largely exothermic catalyzed consumption of the released

MeCN by a double-hydrosilylation process. The metallacyclic

heptacoordinate Ir(III) complex [3]⁺can be formulated as a

Lewis donor−acceptor pair established between the hydridoiri-

dium complex 4²¹(donor) and the triethylsilylium cation⁶³(Z-

type acceptor), wherein a rather weak Si^...H interaction

literature. 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. All of the alcohols used in catalysis were puriﬁed by a bulb-

to-bulb distillation over a minimal amount of CaH₂and stored under

argon before use. ¹H (300, 400, 500, and 600 MHz), ¹¹B (128 MHz),

¹³C (75 and 126 MHz), ¹⁹F (282 MHz), ³¹P (121 and 162 MHz), and

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

400, Avance I 500, and Avance III 600 spectrometers. Chemical shifts

(expressed in parts per million) were referenced against solvent peaks

or external reference standards (CF₃C₆H₅in CDCl₃for ¹⁹F, Me₄Si in

CDCl₃for ²⁹Si, and NaBH₄in D₂O for ¹¹B). Mass spectra were run on

a MicroTOF Bruker Daltonics spectrometer, using a TOF-ESI

coupling analysis system. Elemental analyses were achieved with

Thermo Scientiﬁc FLASH 2000 CHNS/O analyzers.

Structural X-ray Diﬀraction Analyses. X-ray diﬀraction data

collection was carried out on a Bruker APEX II DUO Kappa-CCD

diﬀractometer equipped with an Oxford Cryosystem liquid N₂device,

using Mo Kα radiation (λ = 0.71073 Å). The crystal−detector distance

was 38 mm. The cell parameters were determined (APEX2 software)⁶⁷

from reﬂections taken from 3 sets of 12 frames, each at 10 s exposure.

The structure was solved by direct methods using the program

SHELXS-2013.⁶⁸The reﬁnement and all further calculations were

carried out using SHELXL-2013.⁶⁹For compound [3][BArF₂₄], the

hydride H1A was located from Fourier diﬀerence maps and ﬁxed with

L

DOI: 10.1021/acs.organomet.6b00248

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

the AFIX 1 constraint. For compound [5][OTf], the hydride H1 was

located from Fourier diﬀerence maps and its position reﬁned. The

other H atoms were included in calculated positions and treated as

riding atoms using SHELXL default parameters. The non-H atoms

were reﬁned anisotropically, using weighted full-matrix least squares on

F². A semiempirical absorption correction was applied using SADABS

in APEX2;⁶⁷transmission factors T_min/T_max= 0.6270/0.7463, 0.6387/

0.7462, 0.6650/0.7463, and 0.6242/0.7467 for compounds [2][BF₄],

[2][BArF₂₄], [3][BArF₂₄], and [5][OTf], respectively. For compound

[3][BArF₂₄], the SQUEEZE instruction in PLATON⁷⁰was applied.

The residual electron density was assigned to two molecules of water.

Computational Details. Computations were performed with the

methods of density functional theory using the SCM-ADF2013.01⁷¹

and Gaussian09⁷²(revision D01) packages. The PBE functional⁷³and

the TPSS⁷⁴functional implemented in the Amsterdam Density

Functional package⁷⁵(ADF2013 version) and augmented with

Grimme’s DFT-D3(BJ) implementation of dispersion with a Becke−

Johnson (BJ) damping function were used in all geometry

optimizations.^47a,76Corminboeuf’s PBE0⁷⁷-dDsC^47bhybrid functional

(ADF) was also used for geometry optimizations as well as for

QTAIM and NCI analyses. Geometry optimizations by energy

gradient minimization were carried out in all cases with integration

grid accuracy comprised between 4.5 and 7.5, an energy gradient

convergence criterion of 10⁻³au, and a tight to very tight SCF

convergence criterion. Counterpoise correction for basis set super-

position error (BSSE) was neglected throughout this study. Within the

PBE scheme, electron correlation was treated within the local density

approximation (LDA) in the PW92⁷⁸parametrization. With

ADF2013.01, unless otherwise stated, all computations were carried

out using scalar relativistic corrections within the zeroth order regular

approximation for relativistic eﬀects⁷⁹with ad hoc all-electron

(abbreviated ae) single and double polarization function triple-ζ Slater

type basis sets (TZP and TZ2P). ETS-NOCV⁴²analyses as well as

calculations of vibrational modes were performed with optimized

geometries using ADF2013 subroutines. Vibrational modes were

analytically computed to verify that the optimized geometries were

related to energy minima: statistical thermodynamic data at T = 298.15

K and P = 1 atm were extracted for further determination of enthalpies

and variations of Gibbs free enthalpies by conventional methods. The

COSMO model²⁵of solvation was used for the computation of ion-

dissociation thermochemistries. The geometries of ion pairs were all

computed at the ZORA-PBE-D3(BJ)/ae-TZP level and their

minimum energy nature conﬁrmed by the absence of any imaginary

frequency above 50i cm⁻¹in their computed vibrational modes.

Natural population analyses (NPA) as well as Wiberg index

determinations were performed with geometries of models relaxed at

the ZORA-PBE-D3(BJ) level using all-electron TZP basis sets with the

GENNBO⁴⁶6.0 module of ADF. QTAIM^11,52band NCI⁴⁹region plot

analyses were carried out using the modules embedded within

ADF2013. Representations of molecular structures and isosurfaces

were produced with ADFview2013. QTAIM analyses were performed

on three geometries of [3]⁺, optimized at diﬀerent levels. Figure 7b

depicts the geometry computed at the PBE0-dDsC/ae-TZP level. The

complex was also reoptimized at the PBE0-D3/SDD and PBE-D3/

SDD levels for comparison. These levels consist of the PBE or the

PBE0 functionals, respectively, augmented with the “-D3” correction.

Three basis sets were considered: def2-TZVPP,⁸⁰6-311+G**, and 6-

311++G**,⁸¹on all elements except Ir. The Stuttgart/Dresden

(SDD)⁸²eﬀective core potential (ECP) and its associated basis set

were employed on Ir. The wave functions were then calculated by

performing single-point calculations on these geometries to generate

the “.wfx” ﬁles required for subsequent QTAIM analyses. The electron

density (ρ_β), the Laplacian of the electron density (∇²ρ_β), and the

bond ellipticity (ε) at bond critical points were computed, along with

delocalization indexes (δ(A,B)) between selected pairs of atoms.

These calculations were performed with the AIMAll package (version

15.09.27).⁵⁰QTAIM analyses were also carried out with and without

scalar relativistic corrections for basis sets, starting from geometries

optimized at the (ZORA) PBE0-dDsC level with all-electron TZP and

TZ2P basis sets except in the case of [4→Ag←4]⁺, for which QTAIM

analysis was performed by a single point calculation with a geometry

optimized at the ZORA-PBE-D3(BJ)/ae-TZP level. All DFT

calculations of reaction enthalpies were performed with the

TURBOMOLE 7.0 suite of programs.⁸³The geometry optimizations

were carried out using the PBEh-3c density functional composite

method.⁸⁴Single-point energies were obtained at the B3LYP⁸⁵level

associated with the extended quadruple-ζ basis set def2-QZVP.⁸⁶The

resolution-of-identity (RI) approximation for the Coulomb integrals⁸⁷

with matching default auxiliary basis sets⁸⁸was applied. For integration

of the exchange-correlation contribution, the numerical quadrature

grid m4 was employed for B3LYP. For DFT calculations with the

PBEh-3c functional, the D3 dispersion-correction scheme,⁷⁶applying

the Becke−Johnson (BJ) damping,^47a,89was used. For evaluation of

the single-point energies the nonlocal (NL) van der Waals functional

approach by Vydrov and van Voorhis⁹⁰DFT-NL⁹¹was employed non-

self-consistently (i.e., calculating the NL energy once in a post-SCF

manner) for B3LYP. Computations of the harmonic vibrational

frequencies were performed at the PBEh-3c level. Thermal corrections

from energy to enthalpy were obtained by a coupled rigid-rotor−

harmonic-oscillator approximation for each molecule in the gas

phase.⁹²The PBEh-3c vibrational frequencies were scaled by a factor

of 0.95. The COSMO-RS continuum solvation model^45,93was used as

implemented in COSMOtherm 2016⁹⁴to obtain all solvation

enthalpies. For this purpose single-point calculations employing the

default BP86⁹⁵/def-TZVP⁹⁶level of theory were performed on the

optimized gas-phase geometries for each molecule. The solvation

contribution was then added to the gas-phase enthalpies.

NMR Diﬀusion-Ordered Spectroscopy (DOSY NMR). Measure-

ments of hydrodynamic diﬀusion coeﬃcients were performed on a

Bruker Avance III 600 MHz spectrometer, equipped with a BBFO z

gradient probe, developing a pulse ﬁeld gradient of 5 G (A cm)⁻¹. The

samples were analyzed in a 2.5 mm sample NMR tube thermostated at

298 K. Diﬀusion NMR data were acquired using a stimulated echo

pulse sequence with bipolar z gradients. Limited Eddy current delay

was ﬁxed to 5 ms, and the gradient recovery delay was set to 500 μs. A

total recycling delay of 3 s was applied between scans. The gradient

strength varied linearly between 1 and 34 G cm⁻¹in 34 experiments.

The diﬀusion time and the duration of the sinusoidal gradients were

optimized for each sample. Typically the diﬀusion time was set

between 90 and 150 ms and the half-gradient delay was set between

850 and 1300 μs. Hydrodynamic diﬀusion coeﬃcients were

determined by two diﬀerent methods either from direct reading of

2D-DOSY plots (Table S1 in the Supporting Information) or

analytically as described below. 2D-DOSY plots were generated by

the DOSY module of the software NMRNotebook, using inverse

Laplace transform (ILT) driven by maximum entropy, to build the

diﬀusion dimension. The analytical method corresponds to the graphs

depicted in Figures S9 and S10 in the Supporting Information, which

show the variation of ln(I/I₀) as a function of G²(relative intensities

(I/I₀) were measured for a set of 34 gradient strength values G). By

applying a linear ﬁtting to each correlation curve,^23,97one can extract

the value of the hydrodynamic diﬀusion coeﬃcient D according to

2

ln(I/I₀) = γ_Xδ²(Δ − δ/3 − τ/2)DG², where γ_Xis the gyromagnetic

ratio of the X nucleus, δ is the length of the gradient pulse, G is the

gradient strength, Δ is the delay between the midpoints of the

gradients, τ is the delay between two pulses, and D is the

hydrodynamic diﬀusion coeﬃcient. Under our experimental con-

ditions, the parameters used were as follows: γ _H(BArF₂₄⁻) =

1

98

−

8

γ _H(BF₄) = 2.67522128 × 10 rad s⁻¹T⁻¹; γ (BArF₂₄⁻) =

1

19

F

98

γ (BF₄) = 2.518148 × 10 rad s⁻¹T⁻¹; δ (BArF₂₄⁻) = 2.6 × 10⁻³

−

8

19

F

s; δ _H(BArF₂₄⁻) = 2.3 × 10⁻³s; Δ (BArF₂₄) = 0.17 s; Δ _H(BArF₂₄

)

−

1

19

1

F

−

−3

−

1

19

1

= 0.15 s; δ _H= δ (BF₄) = 2.2 × 10 s; Δ _H(BF₄) = 0.13 s;

F

−

−1

19

Δ (BF₄) = 0.14 s; G in T μm (34 pulse experiments between

F

1.749 × 10⁸and 3.323 × 10⁷with an incremental separation range of

9.6 × 10⁻²); τ _F(BArF₂₄) = τ _F(BF₄⁻) = 5.285 × 10⁻²s;

−

19

−

−3

1

τ _H(BArF₂₄) = τ _H(BF₄) = 5.250 × 10 s. Plotting ln(I/I₀)/C as a

2

function of G²(where C = γ_Xδ²(Δ − δ/3 − τ/2)) gives a negative

slope, the absolute value of which is the hydrodynamic diﬀusion

coeﬃcient D expressed in μm²s⁻¹: ln(I/I₀)/C = −DG².

M

DOI: 10.1021/acs.organomet.6b00248

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

Isothermal Titration Calorimetry. All measurements were

carried out with a Waters-SAS nanoITC device equipped with two

stainless steel Hastealloy cells of 1 mL volume each (measurement and

reference cells) placed in a dry glovebox ﬁlled with argon. Solutions of

[2][BArF₂₄] were prepared by sonication of suspensions of the

complexes in dry and degassed PhCl and were subsequently

thoroughly degassed under moderate reduced pressure. In a typical

titration, a solution of Et₃SiH was placed in the servo-controlled

syringe and injected sequentially into a less concentrated solution of

complex [2][BArF₂₄] in PhCl. The volume injected was set so that the

heat release did not reach the saturation level of the device. Every

injection of silane solution required a time delay to allow the relaxation

of the cell to thermal equilibrium prior to subsequent injection. Time

delays below 1000 s generally produced a baseline drift due to

uncompensated heat. Such a drift indicative of a slow chemical process

could be reduced by increasing the time delay to 1500 s. The enthalpy

of reaction was calculated by summing the integration of each thermal

response over the molar content after subtraction of the dilution

response produced in the last section of the calorigram (Figure 6),

which corresponds to the addition of further volumes of Et₃SiH

solution. The reported value of reaction enthalpy results from three

independent ITC experiments.

¹⁰BF₄). ¹¹B NMR (128 MHz, 298 K, CD₂Cl₂): δ −1.1. HRMS-ESI

calcd for C₂₃H₂₆IrN₂([2]⁺) (m/z): 523.1721. Found: 523.1764.

HRMS-ESI calcd for C₂₁H₂₃IrN ({[2] − CH₃CN}⁺): 482.1455.

Found: 482.1490.

Modiﬁed Procedure for the Synthesis of [2][OTf]. In a Schlenk

ﬂask an equimolar mixture of 1 (200 mg, 0.387 mmol) and Ag[OTf]

(99.4 mg, 0.387 mmol) was dissolved in acetonitrile (5 mL), which

was then vigorously stirred at room temperature for a period not

exceeding 2 h. The resulting yellow suspension was ﬁrst ﬁltered

through a pad of Dicalite, and the solvent was removed from the

ﬁltrate under reduced pressure. The solid was ﬁnally washed with cold

pentane to aﬀord an analytically pure compound as yellow crystals

(218 mg, 84%). Analytical data matched those previously reported by

Sau et al.²⁰

General Procedure for Catalysis Monitoring with a

Piezometric Setup. The alcohol (1 equiv) and Et₃SiH (0.5 equiv)

were introduced in a small-volume double-walled Schlenk vessel

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

checked by applying a moderate pressure of argon gas. The iridium

catalyst (0.1 mol % relative to Et₃SiH) dissolved in 1,2-dichloroethane

(0.5 mL) was injected swiftly into the vessel via a lateral glass valve,

which was tightly sealed immediately upon injection to avoid gas leaks.

Blank control experiments without catalyst were systematically carried

out with all alcohols. After each catalytic run the crude reaction

Synthesis of [2][BArF₂₄]. In a Schlenk ﬂask an equimolar mixture

of 1 (507.2 mg, 0.981 mmol) and Na[BArF₂₄] (869.3 mg, 0.981

mmol) was dissolved in acetonitrile (7 mL). The resulting mixture was

vigorously stirred at room temperature within a period not exceeding 4

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

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

solid was recrystallized in a mixture of CH₂Cl₂and n-hexane to aﬀord

an analytically pure compound. [2][BArF₂₄] was isolated as a yellow

powder in 83% yield (1.129 g). Monocrystals were grown by diﬀusion

of diisopropyl ether vapors into a solution of [2][BArF₂₄] in acetone.

Anal. Calcd for C₅₅H₃₈BF₂₄IrN₂: C, 47.67; H, 2.76; N, 2.02. Found: C,

47.61; H, 2.74; N, 1.98. ¹H NMR (500 MHz, 298 K, CD₂Cl₂): δ 8.61

(d, 1H, J = 6 Hz, H-CN PhPy), 7.91 (m, 1 H, H_ArPhPy), 7.85 (m, 1

H, PhPy), 7.74 (m, 10 H, 2 H_ArPhPy + 8 H_ArBArF₂₄), 7.57 (m, 4 H,

BArF₂₄), 7.27−7.30 (m, 1H, PhPy), 7.18−7.23 (m, 2 H, PhPy), 2.23

(s, 3H, CH₃-CN-Ir), 1.62 (s, 15H, Cp-Me₅). ¹³C NMR (126 MHz, 298

K, CD₂Cl₂): δ 166.0 (C, PhPy), 161.5−162.7 (1:1:1:1 quartet, C-B

1

mixture was analyzed by H NMR spectroscopy, and the identity of

the alkoxysilane was assessed by comparison with data from the

literature: PhCH₂OSiEt₃,⁹⁹(4-iPr-Ph)CH₂OSiEt₃,¹⁰⁰(4-MeO-Ph)-

CH₂OSiEt₃,¹⁰¹MeOSiEt₃,¹⁰²nC₁₀H₁₁OSiEt₃,¹⁰³and [1-(naphthalen-

2-yl)ethoxy]triethylsilane.¹⁰⁴Kinetic isotopic eﬀects were computed

from initial rates of Et₃SiH consumption in solution, which were

inferred from an ideal gas law treatment of the developed H₂(or HD)

gas pressure.

1

BArF₂₄, J_C−B= 50 Hz), 156.6 (C, PhPy), 151.9 (H-CN, PhPy),

Reaction of Et₃SiH with [2][BArF₂₄]. In a glovebox, Et₃SiH (5.8

μL, 36.3 μmol) was added to a solution of [2][BArF₂₄] (25 mg, 18.0

μmol) in CD₂Cl₂(0.5 mL). The resulting solution was shaken and

transferred into a J. Young NMR sample tube which was subsequently

tightly sealed for analysis. Low-temperature (−60 °C) NMR analysis

of the reaction of Et₃SiH with [2][BArF₂₄] revealed spectroscopic

evidence of the formation of both the adducts [3][BArF₂₄] and

[5][BArF₂₄] and (Et₃Si)₂NEt. The presence of (Et₃Si)₂NEt was

conﬁrmed by mass spectroscopy analysis of the reaction mixture. Data

for [3][BArF₂₄] are as follows. ¹H NMR (600 MHz, 213 K, CD₂Cl₂):

δ 8.43 (d, 1 H PhPy, J = 5.3 Hz), 7.98 (m, 1 H PhPy), 7.92 (m, 1 H

PhPy), 7.82 (m, 2 H), 7.77 (m, 8 H_orthoBArF₂₄), 7.59 (m, 4 H_para

BArF₂₄), 7.30−7.35 (m, 2 H, PhPy), 7.23−7.27 (m, 1 H PhPy), 1.71

145.0 (C, PhPy), 139.6 (C-H, PhPy), 136.0 (C-H, PhPy), 135.2 (m, C-

H_orthoBArF₂₄), 132.1 (C-H, PhPy), 128.9−129.6 (q, C-CF₃BArF₂₄,

²J_C−F= 31 Hz), 124.9 (C-H, PhPy), 124.5 (C-H, PhPy), 124.0 (C-H,

PhPy), 124.03 (q, C-CF₃BArF₂₄, ¹J_C−F= 272 Hz), 120.3 (C-H PhPy),

118.3 (Ir-NCMe), 117.9 (m, C-H_paraBArF₂₄), 91.8 (Cp-Me₅), 8.9 (Cp-

Me₅), 4.2 (Ir-NC-Me). ¹⁹F NMR (282 MHz, 298 K, CD₂Cl₂): δ − 63.8

(s, CF₃BArF₂₄). ¹¹B NMR (128 MHz, 298 K, CD₂Cl₂): δ −6.6.

HRMS-ESI calcd for C₂₃H₂₆IrN₂([2]⁺) (m/z): 523.1721. Found:

523.1738. HRMS-ESI calcd for C₃₂H₁₂BF₂₄([BArF₂₄]⁻) (m/z):

863.0649. Found: 863.0723.

Synthesis of [2][BF₄]. In a Schlenk ﬂask an equimolar mixture of 1

(230 mg, 0.445 mmol) and Ag[BF₄] (87 mg, 0.445 mmol) was

dissolved in acetonitrile (5 mL). The resulting mixture was vigorously

stirred at room temperature, and the resulting suspension was ﬁltered

through a pad of Dicalite. The solvent was removed from the ﬁltrate

under reduced pressure, and the resulting solid was washed with cold

pentane to aﬀord an analytically pure compound. [2][BF₄] was

isolated as orange crystals in 80% yield (214 mg). Monocrystals were

obtained by slow diﬀusion of n-heptane into a solution of [2][BF₄] in

CH₂Cl₂. Anal. Calcd for C₂₃H₂₆B₁F₄IrN₂: C, 45.32; H, 4.30; N, 4.60.

Found: C, 45.30; H, 4.34; N, 4.25. ¹H NMR (500 MHz, 298 K,

CD₂Cl₂): δ 8.74 (d, 1 H, H-CN PhPy, ³J = 5.6 Hz), 7.94−7.89 (m,

2 H, PhPy), 7.78−7.74 (m, 2 H PhPy), 7.34 (m, 1 H, PhPy), 7.28 (m,

1 H, PhPy), 7.19 (m, 1 H, PhPy), 2.32 (s, 3 H, CH₃-CN-Ir), 1.70 (s,

15 H, Cp-Me₅). ¹³C NMR (126 MHz, 298 K, CD₂Cl₂): δ 167.6 (C

PhPy), 156.9 (C PhPy), 152.6 (H-CN, PhPy), 145.3 (C, PhPy),

139.6 (C-H, PhPy), 136.1 (C-H, PhPy), 131.8 (C-H, PhPy), 124.8 (C-

H, PhPy), 124.3 (C-H, PhPy), 124.2 (C-H, PhPy), 120.1 (C-H, PhPy),

3

(s, 15 H Cp-Me₅), 0.72 (t, 9 H Ir-H[Si(CH₂CH₃)], J_H−H= 7.8 Hz),

0.42−0.52 (m, 6 H Ir-H[Si(CH₂CH₃)₃]), −11.75 (s, 1 H Ir-H(SiEt)₃,

J_H−Si= 20.1 Hz). ²⁹Si coupled-¹H dimension of HMQC ²⁹Si−¹H (213

K, CD₂Cl₂): δ −11.7 (d, 1H, Ir-H(SiEt₃), J_Si−H= 20 Hz). ²⁹Si-DEPT

NMR (119 MHz, 213 K, CD₂Cl₂): δ 7.1 (d, Ir-H(SiEt₃)). Data for

1

N(Et₃Si)₂Et are as follows. H NMR (600 MHz, 213 K, CD₂Cl₂): δ

3

2.81 (q, 2 H N-CH₂CH₃, J_H−H= 7 Hz), 0.98 (t, 3 H N-CH₂CH₃,

³J_H−H= 7 Hz), 0.90 (t, 18 H SiCH₂CH₃, ³J_H−H= 8 Hz), 0.56 (q, 12 H

3

SiCH₂CH₃, J_H−H= 8 Hz). ²⁹Si NMR (119 MHz, 213 K, CD₂Cl₂): δ

10.9. HRMS-ESI calcd for C₁₄H₃₆N₁Si₂([M + H]⁺) (m/z): 274.2381.

Found: 274.2362.

Interaction of 4-N,N-Dimethylaminopyridine with [3]-

[BArF₂₄]. In a glovebox, Et₃SiH (4.61 μL, 28.9 μmol) was added to

a solution of [2][BArF₂₄] (20 mg, 14.4 μmol) in CD₂Cl₂(0.4 mL).

The resulting mixture was shaken and transferred into a standard

NMR tube, which was then tightly sealed with a septum (t = 0 s or t₀)

and analyzed at t₀+ 10 min by ¹H NMR spectroscopy at room

temperature to verify the presence of [3][BArF₂₄]. A solution of N,N-

118.9 (Ir−NCMe), 91.7 (Cp-Me₅), 8.9 (Cp-Me₅), 4.2 (Ir-NC-Me). ¹⁹

NMR (282 MHz, 298 K, CD₂Cl₂): δ − 153.8 (s, ¹¹BF₄), − 153.8 (s,

F

N

DOI: 10.1021/acs.organomet.6b00248

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

dimethylaminopyridine (DMAP) (3.53 mg, 29.0 μmol) in CD₂Cl₂

yield). Anal. Calcd for C₄₃H₄₇F₃Ir₂N₂O₃S: C, 46.39; H, 4.26; N, 2.52.

Found: C, 46.13; H, 4.25; N, 2.51. ¹H NMR (500 MHz, 298 K,

CD₂Cl₂): δ 7.84 (d, J = 7.5 Hz, 2 H H-CN PhPy), 7.42 (d, J = 7.8

Hz, 2 H PhPy), 7.23−7.33 (m, 4 H PhPy), 7.14 (d, J = 8.0 Hz 2 H

PhPy), 7.05 (m, 2 H PhPy), 6.54−6.64 (m, 4 H PhPy), 1.46 (s, 30 H,

Cp-Me₅), −22.86 (s, 1 H Ir(μ-H)Ir). ¹³C NMR (126 MHz, 298 K,

CD₂Cl₂): δ 166.3 (C PhPy), 159.4 (C PhPy), 151.8 (C-H PhPy),

143.7 (C PhPy), 137.1 (C-H PhPy), 136.7 (C-H PhPy), 130.1 (C-H

PhPy), 125.1 (C-H PhPy), 122.8 (C-H PhPy), 121.6 (C-H PhPy),

117.3 (C-H PhPy), 91.7 (Cp-Me₅), 9.3 (Cp-Me₅). ¹⁹F NMR (282

MHz, 298 K, CD₂Cl₂): δ −79.9 (s, 3F CF₃O₃S). HRMS-ESI (m/z)

calcd for C₄₂H₄₇Ir₂N₂([5]⁺): 965.2997. Found: 965.3018. HRMS-ESI

(m/z) calcd for CF₃O₃S ([OTf]⁻): 148.9515. Found: 148.9527.

(0.3 mL) was added via a syringe into the NMR tube, and the resulting

1

solution was vigorously shaken and a new H NMR spectrum was

acquired (t₀+ 25 min). Before addition of DMAP, the ratio

[3][BArF₂₄]:[5][BArF₂₄] was 2.6:1 (cf. Figure S11a in the Supporting

Information). After addition of DMAP the typical signals of

[3][BArF₂₄] were absent. Only the signals of 4, [5][BArF₂₄], and

traces of [2][BArF₂₄] were observed with a 4:[5][BArF₂₄] ratio of

1.1:1 (cf. Figure S11a).

Preparation of a Dry Solution of [nBu₄N][OTf]. In a glovebox,

dry n-tetrabutylammonium bromide [nBu₄N][Br] (23 mg, 0.072

mmol) was added to a solution of silver triﬂuoromethanesulfonate

Ag[OTf] (18.5 mg, 0.072 mmol) in dry and degassed CD₂Cl₂(0.5

mL). The resulting mixture was vigorously shaken to induce the

quantitative precipitation of AgBr, which was removed by ﬁltration

over a pad of dry Dicalite. The collected ﬁltrate containing

[nBu₄N][OTf] was used as prepared without further puriﬁcation.

Interaction of [nBu₄N][OTf] with [3][BArF₂₄]. In a glovebox,

Et₃SiH (11.5 μL, 72 μmol) was added to solution of [2][BArF₂₄] (20

mg, 14 μmol) in CD₂Cl₂(0.5 mL). The resulting mixture was shaken

and transferred into a standard NMR tube, which was then tightly

sealed with a septum (t = 0 s or t₀) and analyzed at t₀+ 15 min by ¹H

NMR spectroscopy at room temperature to assess the formation of

[3][BArF₂₄]. A solution of [nBu₄N][OTf] (72 μmol, vide supra) in

CD₂Cl₂(0.4 mL) was added via a syringe into the NMR sample tube,

ASSOCIATED CONTENT

* Supporting Information

■

S

The Supporting Information is available free of charge on the

ACS Publications website at DOI: 10.1021/acs.organo-

met.6b00248.

Experimental details, analytical data, acquisition and

reﬁnement parameters and the associated geometrical

parameters for all relevant structures disclosed herein,

energies of all singlet ground state geometries of relevant

species, computed vibrational modes, and QTAIM and

ETS-NOCV diagrams (PDF)

1

the resulting solution was vigorously shaken, and a new H NMR

spectrum was acquired (t₀+ 25 min). Before addition of [nBu₄N]-

[OTf], the ratio [3][BArF₂₄]:[5][BArF₂₄] was 7.7:1 (cf. Figure S11b

in the Supporting Information). After addition of [nBu₄N][OTf] the

typical signals of [3][BArF₂₄] were absent. Only the signals of 4,

[5][BArF₂₄], and traces of Et₃SiH and Et₃SiOTf (partially overlapped

with the signals of nBu₄N⁺) and [nBu₄N][BArF₂₄]) were observed

with a 4:[5][BArF₂₄] ratio of 50:1 (cf. Figure S11b).

Crystallographic data (CIF)

All computed molecule Cartesian coordinates (XYZ)

Partial Structural data for [3′][BArF₂₄] and [5][BArF₂₄]

(PDB)

Reaction of Et₃SiH with [2][BF₄]. In a glovebox, Et₃SiH (5.6 μL,

35.1 μmol) was added to a solution of [2][BF₄] (10 mg, 16.4 μmol) in

CD₂Cl₂(0.5 mL). The resulting solution was shaken and transferred

into a J. Young NMR sample tube, which was subsequently tightly

sealed for analysis. Room-temperature NMR spectroscopic analysis

showed spectroscopic evidence of the formation of 4, [5][BF₄], F₃B·

N(SiEt₃)₂Et, and Et₃SiF. The presence of 4 in the reaction mixture was

conﬁrmed by comparing the ¹H NMR spectrum of the reaction

mixture with that of an authentic sample. The latter compound was

synthesized following the literature procedure.²¹The formation of

Et₃SiF was conﬁrmed by its characteristic ¹⁹F³⁵and ²⁹Si NMR signals,

which both displayed a doublet with a high coupling constant of 287

Hz. The identity of BF₃·N(SiEt₃)₂Et was conﬁrmed by a quartet in the

¹¹B spectrum with a J_B−Fcoupling constant matching the value

observed in the 1:1:1:1 quartet of the ¹⁹F spectrum. Complex

[5][BF₄] displayed a hydride signal at around δ −23 ppm (¹H NMR)

characteristic of Ir(μ-H)Ir dimers (cf. the synthetic procedure of

[5][OTf] and Figure S22 in the Supporting Information).

Reaction of Et₃SiH with [2][OTf]. In a glovebox, Et₃SiH (8.0 μL,

50.1 μmol) was added to a solution of [2][OTf] (19.7 mg, 29.2 μmol)

in CD₂Cl₂(0.5 mL). The resulting solution was shaken and transferred

into a J. Young NMR sample tube, which was subsequently tightly

sealed for analysis. Room-temperature NMR spectroscopic analysis

displayed spectroscopic evidence of the formation of 4, [5][OTf], and

Et₃SiOTf. The identity of the last species was assessed by comparison

of the ¹H NMR spectra of authentic samples (Figures S29 and S30 in

the Supporting Information).

AUTHOR INFORMATION

Corresponding Author

*E-mail for J.-P.D.: djukic@unistra.fr.

Notes

■

The authors declare no competing ﬁnancial interest.

ACKNOWLEDGMENTS

■

Lionel Allouche and Bruno Vincent are thanked for their

expertise in DOSY NMR and Thomas de Oliveira for his

technical contribution. Christophe Michon, Francine Agbos-

́

sou-Niedercorn (ENSCL-Universite de Lille), and Michel

Pfeﬀer (University of Strasbourg) are thanked for fruitful

discussions. N.S. thanks the University of Grenoble Alpes, the

CNRS, and the ICMG FR 2607 for ﬁnancial support and

CIMENT/PCECIC for computer resources. J.-P.D. thanks the

Alexander von Humboldt foundation for ﬁnancial support. The

Centre National de la Recherche Scientiﬁque, the Agence

Nationale de la Recherche, the Deutsche Forschungsgemein-

̀

schaft, and the LABEX “Chimie des Systemes Complexes” are

gratefully acknowledged for their ﬁnancial support.

REFERENCES

■

(1) (a) Corey, J. Y.; Braddock-Wilking, J. Chem. Rev. 1999, 99, 175−

292. (b) Corey, J. Y. Chem. Rev. 2011, 111, 863−1071.

(2) 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.

Optimized Synthesis of [5][OTf]. Et₃SiH was added dropwise

(70 μL, 0.44 mmol) to a solution of [2][OTf] (87 mg, 0.13 mmol) in

CH₂Cl₂(3 mL). A color change from yellow to deep red was observed

immediately after addition of Et₃SiH. This mixture was stirred at room

temperature for 22 h. The resulting solution was concentrated to ca. 1

mL, and benzene (2 mL) and pentane (3 mL) were added. This

mixture was left to stand at +4 °C overnight to induce the slow

crystallization of [5][OTf], which was collected after removal of the

supernatant by suction and dried overnight at room temperature under

reduced pressure to aﬀord an analytically pure compound (40 mg, 56%

(3) Burns, N. Z.; Baran, P. S.; Hoffmann, R. W. Angew. Chem., Int. Ed.

2009, 48, 2854−2867.

(4) Yang, J.; White, P. S.; Schauer, C. K.; Brookhart, M. Angew.

Chem., Int. Ed. 2008, 47, 4141−4143.

(5) Klei, S. R.; Tilley, T. D.; Bergman, R. G. J. Am. Chem. Soc. 2000,

122, 1816−1817.

O

DOI: 10.1021/acs.organomet.6b00248

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

(6) (a) Perutz, R. N.; Sabo-Etienne, S. Angew. Chem., Int. Ed. 2007,

46, 2578−2592. (b) Calimano, E.; Tilley, T. D. Dalton Trans. 2010, 39,

9250−9263. (c) Martin, M.; Sola, E.; Torres, O.; Plou, P.; Oro, L. A.

Organometallics 2003, 22, 5406−5417.

(31) (a) Martinho Simoes, J. A.; Minas da Piedade, M. E. Molecular

Energetics Condensed-Phase Thermochemical Techniques; Oxford Uni-

versity Press: New York, 2008. (b) Krell, T.; Lacal, J.; Garcia-Fontana,

C.; Silva-Jimenez, H.; Rico-Jimenez, M.; Lugo, A. C.; Darias, J. A. R.;

Ramos, J.-L. Methods Mol. Biol. 2014, 1149, 193−203.

(7) Green, J. C.; Green, M. L. H.; Parkin, G. Chem. Commun. 2012,

48, 11481−11503.

́

(32) Perez, M.; Qu, Z.-W.; Caputo, C. B.; Podgorny, V.; Hounjet, L.

(8) Lachaize, S.; Sabo-Etienne, S. Eur. J. Inorg. Chem. 2006, 2006,

2115−2127.

J.; Hansen, A.; Dobrovetsky, R.; Grimme, S.; Stephan, D. W. Chem. -

Eur. J. 2015, 21, 6491−6500.

(9) (a) Schubert, U.; Ackermann, K.; Worle, B. J. Am. Chem. Soc.

1982, 104, 7378−7380. (b) Nikonov, G. I. Organometallics 2003, 22,

1597−1598. (c) Nikonov, G. I. Adv. Organomet. Chem. 2005, 53, 217−

309. (d) Vyboishchikov, S. F.; Nikonov, G. I. Chem. - Eur. J. 2006, 12,

8518−8533.

(33) (a) Gandhamsetty, N.; Jeong, J.; Park, J.; Park, S.; Chang, S. J.

Org. Chem. 2015, 80, 7281−7287. (b) Yang, Y.-F.; Chung, L. W.;

Zhang, X.; Houk, K. N.; Wu, Y.-D. J. Org. Chem. 2014, 79, 8856−8864.

(34) Dubberley, S. R.; Ignatov, S. K.; Rees, N. H.; Razuvaev, A. G.;

Mountford, P.; Nikonov, G. I. J. Am. Chem. Soc. 2003, 125, 642−643.

(35) Vela, J.; Smith, J. M.; Yu, Y.; Ketterer, N. A.; Flaschenriem, C. J.;

Lachicotte, R. J.; Holland, P. L. J. Am. Chem. Soc. 2005, 127, 7857−

7870.

(10) Lin, Z. Chem. Soc. Rev. 2002, 31, 239−245.

(11) Bader, R. F. W. In Atoms in Molecules: A Quantum Theory;

Clarendon: Oxford, U.K., 1990.

(12) (a) Scherer, W.; Meixner, P.; Barquera-Lozada, J. E.; Hauf, C.;

(36) Bassindale, A. R.; Stout, T. Tetrahedron Lett. 1984, 25, 1631−

1632.

Obenhuber, A.; Bruck, A.; Wolstenholme, D. J.; Ruhland, K.; Leusser,

̈

(37) (a) Hoffmann, S. P.; Kato, T.; Tham, F. S.; Reed, C. A. Chem.

Commun. 2006, 767−769. (b) Nava, M.; Reed, C. A. Organometallics

2011, 30, 4798−4800. (c) Connelly, S. J.; Kaminsky, W.; Heinekey, D.

M. Organometallics 2013, 32, 7478−7481.

D.; Stalke, D. Angew. Chem., Int. Ed. 2013, 52, 6092−6096. (b) Hauf,

C.; Barquera-Lozada, J. E.; Meixner, P.; Eickerling, G.; Altmannshofer,

S.; Stalke, D.; Zell, T.; Schmidt, D.; Radius, U.; Scherer, W. Z. Z.

Anorg. Allg. Chem. 2013, 639, 1996−2004.

(38) Chen, J.; Chen, E. Y. X. Angew. Chem., Int. Ed. 2015, 54, 6842−

6846.

(13) Corre, Y.; Iali, W.; Hamdaoui, M.; Trivelli, X.; Djukic, J. P.;

Agbossou-Niedercorn, F.; Michon, C. Catal. Sci. Technol. 2015, 5,

1452−1458.

(39) Einholz, W.; Gollinger, W.; Haubold, W. Z. Naturforsch., B:

Chem. Sci. 1990, 45, 25−30.

́

(40) Petrovic, P.; Djukic, J.-P.; Hansen, A.; Bannwarth, C.; Grimme,

(14) Pannetier, N.; Sortais, J.-B.; Issenhuth, J.-T.; Barloy, L.; Sirlin,

C.; Holuigue, A.; Lefort, L.; Panella, L.; de Vries, J. G.; Pfeffer, M. Adv.

Synth. Catal. 2011, 353, 2844−2852.

S. In Non-Covalent Interactions in the Synthesis and Design of New

Compounds; Maharramov, A. M., Mahmudov, K. T., Kopylovich, M.

N., Pombeiro, A. J. L., Eds.; John Wiley & Sons: 2016; p 115−143.

(41) Turlington, C. R.; Harrison, D. P.; White, P. S.; Brookhart, M.;

Templeton, J. L. Inorg. Chem. 2013, 52, 11351−11360.

(42) (a) Mitoraj, M. P.; Michalak, A.; Ziegler, T. Organometallics

2009, 28, 3727−3733. (b) Mitoraj, M. P.; Michalak, A.; Ziegler, T. J.

Chem. Theory Comput. 2009, 5, 962−975.

(43) Yang, J.; White, P. S.; Brookhart, M. J. Am. Chem. Soc. 2008,

130, 17509−17518.

(44) (a) Metsanen, T. T.; Hrobarik, P.; Klare, H. F. T.; Kaupp, M.;

Oestreich, M. J. Am. Chem. Soc. 2014, 136, 6912−6915. (b) Iglesias,

M.; Fernandez-Alvarez, F. J.; Oro, L. A. ChemCatChem 2014, 6, 2486−

2489.

(15) (a) Wu, X.; Liu, J.; Li, X.; Zanotti-Gerosa, A.; Hancock, F.;

Vinci, D.; Ruan, J.; Xiao, J. Angew. Chem., Int. Ed. 2006, 45, 6718−

6722. (b) Wang, C.; Pettman, A.; Basca, J.; Xiao, J. Angew. Chem., Int.

Ed. 2010, 49, 7548−7552. (c) Wang, C.; Chen, H.-Y. T.; Bacsa, J.;

Catlow, C. R. A.; Xiao, J. Dalton Trans. 2013, 42, 935−940. (d) Wei,

Y.; Wang, C.; Jiang, X.; Xue, D.; Liu, Z.-T.; Xiao, J. Green Chem. 2014,

16, 1093−1096. (e) Chen, H. Y. T.; Wang, C.; Wu, X.; Jiang, X.;

Catlow, C. R. A.; Xiao, J. Chem. - Eur. J. 2015, 21, 16564−16577.

(16) Iali, W.; La Paglia, F.; Le Goff, X.-F.; Sredojevic, D.; Pfeffer, M.;

Djukic, J.-P. Chem. Commun. 2012, 48, 10310−10312.

(17) Gutsulyak, D. V.; Vyboishchikov, S. F.; Nikonov, G. I. J. Am.

Chem. Soc. 2010, 132, 5950−5951.

(18) (a) Cardoso, J. M. S.; Lopes, R.; Royo, B. J. Organomet. Chem.

(45) Klamt, A. J. Phys. Chem. 1995, 99, 2224−2235.

́ ́

2015, 775, 173−177. (b) Esteruelas, M. A.; Olivan, M.; Velez, A. Inorg.

(46) Weinhold, F. In Encyclopedia of Computational Chemistry;

Schleyer, P. v. R., Allinger, N. L., Clark, T., Gasteiger, J., Kollman, P.

A., Schaefer, H. F., Schreiner, P. R., Eds.; Wiley: Chichester, U.K.,

1998; Vol. 3, pp 1792−1811.

Chem. 2013, 52, 12108−12119. (c) Luo, X. L.; Crabtree, R. H. J. Am.

Chem. Soc. 1989, 111, 2527−2535. (d) Sattler, W.; Parkin, G. J. Am.

Chem. Soc. 2012, 134, 17462−17465.

́

(19) (a) Labouille, S.; Escalle-Lewis, A.; Jean, Y.; Mezailles, N.;

(47) (a) Grimme, S.; Ehrlich, S.; Goerigk, L. J. Comput. Chem. 2011,

32, 1456−1465. (b) Steinmann, S. N.; Corminboeuf, C. J. Chem.

Theory Comput. 2011, 7, 3567−3577.

LeFloch, P. Chem. - Eur. J. 2011, 17, 2256−2265. (b) Cardoso, J. M.

S.; Lopes, R.; Royo, B. J. Organomet. Chem. 2015, 775, 173−177.

(20) Sau, Y.-K.; Yi, X.-Y.; Chan, K.-W.; Lai, C.-S.; Williams, I. D.;

Leung, W.-H. J. Organomet. Chem. 2010, 695, 1399−1404.

(21) Hu, Y.; Li, L.; Shaw, A. P.; Norton, J. R.; Sattler, W.; Rong, Y.

Organometallics 2012, 31, 5058−5064.

(48) (a) Hyla-Kryspin, I.; Grimme, S.; Djukic, J.-P. Organometallics

2009, 28, 1001−1013. (b) Schwabe, T.; Grimme, S.; Djukic, J.-P. J.

Am. Chem. Soc. 2009, 131, 14156−14157. (c) Grimme, S.; Djukic, J.-P.

Inorg. Chem. 2010, 49, 2911−2919. (d) Grimme, S.; Djukic, J.-P. Inorg.

(22) Reichardt, C. In Solvents and Solvent Eﬀects in Organic Chemistry;

Wiley-VCH: Weinheim, Germany, 2004; pp 471−507.

(23) Pregosin, P. S. Pure Appl. Chem. 2009, 81, 615−633.

(24) Vanka, K.; Chan, M. S. W.; Pye, C. C.; Ziegler, T.

Organometallics 2000, 19, 1841−1849.

́

Chem. 2011, 50, 2619−2628. (e) Werle, C.; Karmazin, L.; Bailly, C.;

Ricard, L.; Djukic, J.-P. Organometallics 2015, 34, 3055−3064.

(49) (a) Contreras-Garcia, J.; Johnson, E. R.; Keinan, S.; Chaudret,

R.; Piquemal, J. P.; Beratan, D. N.; Yang, W. J. Chem. Theory Comput.

2011, 7, 625−632. (b) Johnson, E.; Keinan, S.; Mori-Sanchez, P.;

Contreras-Garcia, J.; Cohen, A. J.; Yang, W. J. Am. Chem. Soc. 2010,

132, 6498−6506.

(25) Klamt, A.; Schuurmann, G. J. Chem. Soc., Perkin Trans. 2 1993,

799−805.

́ ́

(26) Hansen, A.; Bannwarth, C.; Grimme, S.; Petrovic, P.; Werle, C.;

(50) Keith, T. A. AIMall; TK Gristmill Software, Overland Park, KS,

USA, 2016.

(51) Jabłonski, M.; Palusiak, M. J. Phys. Chem. A 2010, 114, 12498−

12505.

Djukic, J.-P. ChemistryOpen 2014, 3, 177−189.

(27) Ziegler, T.; Rauk, A. Inorg. Chem. 1979, 18, 1755−1759.

(28) Espenson, J. H. Chemical Kinetics and Reaction Mechanisms;

McGraw-Hill: New York, 1981.

(52) (a) Fradera, X.; Poater, J.; Simon, S.; Duran, M.; Sola, M. Theor.

Chem. Acc. 2002, 108, 214−224. (b) Bader, R. F. W.; Streitwieser, A.;

Neuhaus, A.; Laidig, K. E.; Speers, P. J. Am. Chem. Soc. 1996, 118,

4959−4965.

(29) Gom

4963.

́

ez-Gallego, M.; Sierra, M. A. Chem. Rev. 2011, 111, 4857−

(30) Nolin, K. A.; Krumper, J. R.; Pluth, M. D.; Bergman, R. G.;

Toste, F. D. J. Am. Chem. Soc. 2007, 129, 14684−14696.

P

DOI: 10.1021/acs.organomet.6b00248

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

(53) Lopez, C. S.; de Lera, A. R. Curr. Org. Chem. 2011, 15, 3576−

(78) Perdew, J. P.; Wang, Y. Phys. Rev. B: Condens. Matter Mater.

Phys. 1992, 45, 13244−13249.

3593.

(79) (a) van Lenthe, E.; Ehlers, A.; Baerends, E.-J. J. Chem. Phys.

1999, 110, 8943−8953. (b) van Lenthe, E.; Baerends, E. J.; Snijders, J.

G. J. Chem. Phys. 1994, 101, 9783−9792. (c) van Lenthe, E.; Baerends,

E. J.; Snijders, J. G. J. Chem. Phys. 1993, 99, 4597−4610.

(80) Weigend, F.; Ahlrichs, R. Phys. Chem. Chem. Phys. 2005, 7,

3297−3305.

(54) Savin, A.; Nesper, R.; Wengert, S.; Fassler, T. F. Angew. Chem.,

̈

Int. Ed. Engl. 1997, 36, 1808−1832.

(55) Fowe, E. P.; Therrien, B.; Suss-Fink, G.; Daul, C. Inorg. Chem.

2008, 47, 42−48.

(56) Parkin, G. J. Chem. Educ. 2006, 83, 791.

(57) (a) Gutsulyak, D. V.; van der Est, A.; Nikonov, G. I. Angew.

Chem., Int. Ed. 2011, 50, 1384−1387. (b) Osakada, K. Angew. Chem.,

Int. Ed. 2011, 50, 3845−3846.

(81) (a) Clark, T.; Chandrasekhar, J.; Spitznagel, G. W.; Schleyer, P.

v. R. J. Comput. Chem. 1983, 4, 294−301. (b) McLean, A. D.;

Chandler, G. S. J. Chem. Phys. 1980, 72, 5639−5648. (c) Krishnan, R.;

Binkley, J. S.; Seeger, R.; Pople, J. A. J. Chem. Phys. 1980, 72, 650−654.

(58) (a) Green, M. L. H. J. Organomet. Chem. 1995, 500, 127−148.

(b) Amgoune, A.; Bourissou, D. Chem. Commun. 2011, 47, 859−871.

(59) (a) Djukic, J.-P.; Fetzer, L.; Czysz, A.; Iali, W.; Sirlin, C.; Pfeffer,

M. Organometallics 2010, 29, 1675−1679. (b) Djukic, J.-P.; Iali, W.;

Pfeffer, M.; Le Goff, X.-F. Chem. - Eur. J. 2012, 18, 6063−6078.

(60) (a) Carey, F. A.; Tremper, H. S. J. Am. Chem. Soc. 1968, 90,

2578−2583. (b) Carey, F. A.; Tremper, H. S. J. Org. Chem. 1969, 34,

4−6. (c) Carey, F. A.; Hsu, C.-L. W. J. Organomet. Chem. 1969, 19,

29−41. (d) Carey, F. A.; Tremper, H. S. J. Org. Chem. 1971, 36, 758−

761. (e) Chojnowski, J.; Wilczek, L.; Fortuniak, W. J. Organomet.

Chem. 1977, 135, 13−22. (f) Mayr, H.; Basso, N.; Hagen, G. J. Am.

Chem. Soc. 1992, 114, 3060−3066. (g) Voskoboynikov, A. Z.;

Parshina, I. N.; Shestakova, A. K.; Butin, K. P.; Beletskaya, I. P.;

Kuz’mina, L. G.; Howard, J. A. K. Organometallics 1997, 16, 4041−

4055.

(82) Andrae, D.; Haußermann, U.; Dolg, M.; Stoll, H.; Preuß, H.

̈

Theor. Chim. Acta 1990, 77, 123−141.

(83) Furche, F.; Ahlrichs, R.; Hattig, C.; Klopper, W.; Sierka, M.;

̈

Weigend, F. WIREs Comput. Mol. Sci. 2014, 4, 91−100.

(84) Grimme, S.; Brandenburg, J. G.; Bannwarth, C.; Hansen, A. J.

Chem. Phys. 2015, 143, 054107.

(85) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648−5652. (b) Lee,

C.; Yang, W.; Parr, R. G. Phys. Rev. B: Condens. Matter Mater. Phys.

1988, 37, 785−789. (c) Stephens, P. J.; Devlin, F. J.; Chabalowski, C.

F.; Frisch, M. J. J. Phys. Chem. 1994, 98, 11623−11627.

(86) Weigend, F.; Ahlrichs, R. Phys. Chem. Chem. Phys. 2005, 7,

3297−3305.

̈

(87) Eichkorn, K.; Treutler, O.; Ohm, H.; Haser, M.; Ahlrichs, R.

̈

Chem. Phys. Lett. 1995, 242, 652−660.

(61) (a) Braunschweig, H.; Gruss, K.; Radacki, K. Angew. Chem., Int.

Ed. 2007, 46, 7782−7784. (b) Braunschweig, H.; Brunecker, C.;

Dewhurst, R. D.; Schneider, C.; Wennemann, B. Chem. - Eur. J. 2015,

21, 19195−19201. (c) Braunschweig, H.; Dewhurst, R. D.; Hupp, F.;

Schneider, C. Chem. Commun. 2014, 50, 15685−15688. (d) Braunsch-

weig, H.; Dewhurst, R. D.; Hupp, F.; Kaufmann, C.; Phukan, A. K.;

Schneider, C.; Ye, Q. Chem. Sci. 2014, 5, 4099−4104. (e) Bauer, J.;

Bertermann, R.; Braunschweig, H.; Gruss, K.; Hupp, F.; Kramer, T.

Inorg. Chem. 2012, 51, 5617−5626.

(88) Weigend, F. Phys. Chem. Chem. Phys. 2006, 8, 1057−1065.

(89) (a) Becke, A. D.; Johnson, E. R. J. Chem. Phys. 2005, 123,

154101. (b) Johnson, E. R.; Becke, A. D. J. Chem. Phys. 2005, 123,

024101.

(90) Vydrov, O. A.; vanVoorhis, T. J. Chem. Phys. 2010, 133, 244103.

(91) Hujo, W.; Grimme, S. J. Chem. Theory Comput. 2011, 7, 3866.

(92) Grimme, S. Chem. - Eur. J. 2012, 18, 9955−9964.

(93) Eckert, F.; Klamt, A. AIChE J. 2002, 48, 369−385.

(94) Eckert, F.; Klamt, A. In COSMOtherm, Version C3.0, Release

16.01; COSMOlogic, Leverkusen, Germany, 2016.

́

(62) Werle, C.; Bailly, C.; Karmazin-Brelot, L.; Le Goff, X. F.; Pfeffer,

M.; Djukic, J. P. Angew. Chem., Int. Ed. 2014, 53, 9827−9831.

(63) Lambert, J. B.; Zhang, S.; Stern, C. L.; Huffman, J. C. Science

1993, 260, 1917−1918.

(95) (a) Becke, A. D. Phys. Rev. A: At., Mol., Opt. Phys. 1988, 38,

3098−3100. (b) Perdew, J. P. Phys. Rev. B: Condens. Matter Mater.

Phys. 1986, 33, 8822−8824.

(64) (a) Metsanen, T. T.; Hrobarik, P.; Klare, H. F. T.; Kaupp, M.;

̈

(96) Schafer, A.; Huber, C.; Ahlrichs, R. J. Chem. Phys. 1994, 100,

̈

Oestreich, M. J. Am. Chem. Soc. 2014, 136, 6912−6915. (b) Iglesias,

M.; Miguel, P. J. S.; Polo, V.; Fernandez-Alvarez, F. J.; Perez-Torrente,

J. J.; Oro, L. A. Chem. - Eur. J. 2013, 19, 17559−17566.

(65) O’Connor, J. M. Sci. Synth. 2001, 7, 617−744.

(66) 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.

5829−5835.

(97) Wu, D. H.; Chen, A. D.; Johnson, C. S. J. Magn. Reson., Ser. A

1995, 115, 260−264.

(98) Harris, R. K.; Becker, E. D.; De Menezes, S. M. C.; Goodfellow,

R.; Granger, P. Pure Appl. Chem. 2001, 73, 1795−1818.

(99) Ison, E. A.; Trivedi, E. R.; Corbin, R. A.; Abu-Omar, M. M. J.

Am. Chem. Soc. 2005, 127, 15374−15375.

(67) M86-E01078 APEX2 User Manual; Bruker AXS: Madison, WI,

USA, 2006.

(100) Wright, A.; West, R. J. Am. Chem. Soc. 1974, 96, 3214−3222.

(101) Kennedy-Smith, J. J.; Nolin, K. A.; Gunterman, H. P.; Toste, F.

D. J. Am. Chem. Soc. 2003, 125, 4056−4057.

(102) Caseri, W.; Pregosin, P. S. Organometallics 1988, 7, 1373−

1380.

(103) Ito, H.; Watanabe, A.; Sawamura, M. Org. Lett. 2005, 7, 1869−

1871.

(104) Diez-Gonzalez, S.; Escudero-Adan, E. C.; Benet-Buchholz, J.;

Stevens, E. D.; Slawin, A. M. Z.; Nolan, S. P. Dalton Trans. 2010, 39,

7595−7606.

(68) Sheldrick, G. M. Acta Crystallogr., Sect. A: Found. Crystallogr.

1990, 46, 467−473.

(69) Sheldrick, G. M. Acta Crystallogr., Sect. A: Found. Crystallogr.

2008, 64, 112−122.

(70) Spek, A. L. J. Appl. Crystallogr. 2003, 36, 7−13.

(71) ADF2013.01; Theoretical Chemistry, Vrije Universiteit,

Amsterdam, The Netherlands, 2013.

(72) Pople, J. A. In Gaussian 09; Gaussian, Inc., Pittsburgh, PA, 2009

(the full reference is given in the Supporting Information).

(73) Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. Rev. Lett. 1996, 77,

3865−3868.

(74) Tao, J.; Perdew, J. P.; Staroverov, V. N.; Scuseria, G. E. Phys.

Rev. Lett. 2003, 91, 146401.

(75) 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.

(76) Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H. J. Chem. Phys.

2010, 132, 154104.

(77) (a) Ernzerhof, M.; Scuseria, G. E. J. Chem. Phys. 1999, 110,

5029−5036. (b) Adamo, C.; Barone, V. J. Chem. Phys. 1999, 110,

6158.

Q

DOI: 10.1021/acs.organomet.6b00248

Organometallics XXXX, XXX, XXX−XXX

Article Doi

Evidence of a Donor-Acceptor (Ir-H)→SiR3 Interaction in a Trapped Ir(III) Silane Catalytic Intermediate

DOI: 10.1021/acs.organomet.6b00248

Source and publish data:

Authors:

Article abstract of DOI:10.1021/acs.organomet.6b00248

Full text of DOI:10.1021/acs.organomet.6b00248

Products guided by the article

R&D Labs maybe for 1384179-05-8

Relevant to this article

Hot Product