Entrapment of THF-Stabilized Iridacyclic IrIII Silylenes from Double H?Si Bond Activation and H2 Elimination

Article abstract of DOI:10.1002/chem.201804048

The reaction of H₃SiR (R=Ph, nBu) with cationic η₅-C₅Me₅- (Cp*) and benzo[h]quinolinyl-based iridacycle [1 b]⁺ gives rise to new [(IrH)→SiRH₂]⁺ adducts. In the presence of THF these adducts readily undergo elimination of H₂ gas at subambient temperature to form THF-stabilized metallacyclic Ir^III silylene complexes, which were characterized in situ by NMR spectroscopy, trapped in minute amounts by reactive crystallization, and structurally characterized by XRD. Theoretical investigations (static DFT-D reaction-energy profiling, ETS-NOCV) support the promoting role of THF in the H₂ elimination step and the consolidation of the Ir-to-Si interaction in the spontaneous (ΔG<0) formation of Ir silylenes in the presence of THF. Mechanistic insights indicate that the Ir silylene species arising from the [1 b]⁺/phenylsilane system are relevant catalytic species in the hydrodefluorination of fluoroalkanes.

Full text of DOI:10.1002/chem.201804048

A Journal of
Accepted Article
Title: Entrapment of THF-stabilized Iridacyclic Ir(III)-“Silylenes” from
Double H-Si Bond Activation and H2 Elimination
Authors: Dang Ho Binh, Mustapha Hamdaoui, Diane Fischer Krauser,
Lydia Karmazin, Corinne Bailly, and Jean-Pierre Djukic
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.201804048
Link to VoR: http://dx.doi.org/10.1002/chem.201804048
Supported by

10.1002/chem.201804048
Chemistry - A European Journal
ARTICLE
Entrapment of THF-stabilized Iridacyclic Ir(III)-“Silylenes” from
Double H-Si Bond Activation and H2 Elimination
Dang Ho Binh[a]‡, Mustapha Hamdaoui[a] ‡, Diane Fischer-Krauser[a], Lydia Karmazin[a], Corinne Bailly[a]
and Jean-Pierre Djukic*[a]
Dedication ((optional))
Abstract: The reaction of H3SiR (R=Ph, nBu) with a cationic Cp*-and
benzo[h]quinolinyl- based iridacycle [1b]+ gives rise to new [(Ir-
H)→SiRH2]+ adducts. In the presence of tetrahydrofuran (THF) the
latter adducts readily convert by H2 gas elimination at subambient
temperature into new THF-stabilized metallacyclic Ir(III)-“silylene”
complexes that were characterized in situ by NMR spectroscopic
analyses, trapped in minute amount by reactive crystallization and
their structures established by X-ray diffraction analysis. Theoretical
investigations (static DFT-D reaction energy profiling, ETS-NOCV)
support the promoting role of THF in the H2 elimination step and the
consolidation of the Ir-to-Si interaction in the spontaneous (G <0)
formation of Ir-silylenes in the presence of THF. Mechanistic
insights point the Ir-silylene species arising from the
[1b]+/phenylsilane system as a relevant catalytic species in the
hydrodefluorination of fluoroalkanes.
The main methods[1a] for preparing metal-silylenes are
based either on the “anionic substituent abstraction”[4f] [11] from a
metal-hydrosilyl precursor or on the sequential H-Si activations[4c,
5a, 12]
from the reaction of a R—metal precursor (e.g., R = alkyl,
aryl) with a hydrosilane.[13] If examples of direct formation of
metal-silylenes by double H-Si bond activations[14] [15] are legion,
those cases where spontaneous elimination of H2 follows the
double Si-H bond activation are relatively scarce.[3a, 4c, 5a, 12, 16]
Recently, Tilley et al.[5a] reported that
L3(H)Ru=SiRR’, can be obtained by reversible H2 elimination
a Ru-silylene, i.e
from
a
L3(H)Ru(3-H2SiRR’) intermediate, which later
demonstrated catalytic turnover in ketone hydrosilylation.[5a]
According to Corriu et al. metal-silylene complex formation can
be promoted by exogenous donor ligands.
Corriu et al
demonstrated that iron-silylene complexes can be obtained by
[4b, 4c, 16a]
reaction of Fe(CO)5 with a hydrosilane
under photolytic
conditions or in the presence of tetrahydrofuran (THF)[4b]; the
latter, by bonding the Si centre, allegedly induces an increase of
the “basicity of the H-Si bond”, thus facilitating H2 elimination.[4b]
It was proposed by Tilley et al. in an extensive account[1a] that
conversion of hydrosilanes into silylene complexes (namely by
Si-H activiation and -hydrogen migration) could constitute an
excellent basis for new catalytic applications. Having this
proposal as aim we investigated the reactivity of hydrosilanes
towards iridacycles [1a,b][BArF24] (Scheme 1).
In transition-metal-“silylenes” M-SiR2 the peculiarity of the
back bonding from the metal-centered d orbital to the high lying
Si-centered p orbital has attracted sustained interest, for it opens
a large field of exploration of the reactivity of such electrophilic
and electropositive Si centres.[1],[2] If a number of transition
metal silylenes were isolated as M-SiR2 complexes, the
electrophilicity of metal-silylenes particularly in cationic
complexes has often led to the isolation of donor-acceptor
complexes M-SiR2•B,[1a] wherein the base B is bonded to the
electrophilic Si centre.[1a, 3] The well-documented reactivity of
transition metal-silylenes[1a] is often a consequence of the high
electrophilic character of the Si centre[1a, 3a, 3b, 4], particularly in
Scheme 1. The reactions of [1a,b]+ with triethylsilane, phenylsilane and n-
butylsilane.
cationic complexes[4d]: reactivity towards carbonyls[4e, 5], nitriles[4e,
4f, 6]
4j], alkynes[4e], isocyanates[4g], alkenes[4d,
and amines[5c] is
 HSiEt3
a
H
N
CD2Cl2, -60 °C
well documented. Quite interestingly the occurence in catalysis
of transition metal-“silylenes” prepared by activation of
hydrosilanes remains limited[1a, 4d, 4f, 7] [5c, 8] while at the same time
coinage metal-mediated transfer of formal silylenes[9] from
silacyclopropanes to unsaturated organic substrates is
increasingly steering interest.[10]
b
Et3Si
*Cp
Our work, 2016,2017
Ir
-MeCN
quantitative
25 min
,[BArF24
]
[2a]-or-[2b][BArF24
]
(with either
[1a]+ or [1b]+)
*Cp
H3SiR (excess)
CD2Cl2, -60 °C
-MeCN
quantitative
H
N
Ir
H
R
Si
,[BArF24
]
H
~ 25 min
route 1
route 2
[2c][BArF24
[2d][BArF24
]
R= Ph
This work
a
MeCN
*Cp
R= nBu
N
]
(with [1b]+)
[a]
D. Ho Binh, Dr M. Hamdaoui, D. Krauser-Fischer, Dr L. Karmazin,
C. Bailly and Dr J.-P. Djukic
Institut de Chimie de Strasbourg UMR 7177
CNRS-Université de Strasbourg
4 rue Blaise Pascal, F-67000 Strasbourg France
E-mail: djukic@unistra.fr
b
Ir
+ THF (excess),
[-60 °Crt], 1 h
 H2
,[BArF24
]
[1a]-or-[1b][BArF24
]
Ir-"silylene" formation
H3SiR (excess)
*Cp
N
F3C
F3C
d8-THF, rt
‡
H
R
Ir
these authors contributed equally to this work
 MeCN,  H2
quantitative
~ 2 h
[BArF24
]
=
Si
thf
B
,[BArF24
]
]
Supporting information for this article is given via a link at the end of
the document. CCDC 1826970 and 1843843 contain the
supplementary crystallographic data for this paper. These data can
be obtained free of charge via
www.ccdc.cam.ac.uk/data_request/cif, or by emailing
data_request@ccdc.cam.ac.uk, or by contacting The Cambridge
Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ,
UK; fax: +44 1223 336033.
[3][BArF24
]
R= Ph
R= nBu
[4][BArF24
The latter (Scheme 1) are efficient precatalysts for the
alcoholysis of silanes[17] and the hydrosilylation of carbonyls and
Most importantly, the products of the reaction of
[1a,b][BArF24] with HSiEt3, i.e [2a,b][BArF24] are neither so-
nitriles[18]
.
1
This article is protected by copyright. All rights reserved.

10.1002/chem.201804048
Chemistry - A European Journal
ARTICLE
19]
called -complexes[3f,
nor the products of the oxidative
Bu) (cf SI). The reaction of a CD2Cl2 solution of [1b][BArF24]
(1eq.) with H3SiPh (4 eq.) was monitored by NMR spectroscopy
(at 60 °C). Quantitative and selective formation of product
[2c][BArF24] besides unreacted H3SiPh in a respective ~1:3 ratio
(route 1, scheme 1; cf. Suppl. Mat.) was noticed. Worthy to note,
there was no sign of hydrosilylation of the released MeCN[17-18]
(singlet at δ 1.98 ppm). [2c][BArF24] is characterized by a typical
[3f, 19a, 19b, 20]
addition of the Si-H bond to the Ir metal-centre
or
any of its “arrested state”[21] but donor-acceptor [(Ir-H)→SiEt3]+
complexes (Scheme 1) resulting from the concerted transfer of
hydride from the silane to the electrophilic Ir(III) centre and the
capture of the labile silylium moiety.[17-18,
The present
22]
investigation of the reaction of [1b][BArF24] with phenylsilane
and n-butylsilane, reveals that not only such [(Ir-H)→SiRH2]+
adducts do form in solution but that tetrahydrofurane (THF)
assists actively their transformation into highly unstable cationic
Ir(III)-silylenes, of which two iridacyclic examples were
successfully trapped by “reactive crystallization”. Remarkably,
the silylene complex arising from the reaction of [1b][BArF24]
with H3SiPh is found to promote the hydro-de-fluorination (HDF)
of fluoroalkanes.
1
1H NMR Ir-H resonance at δ 11.5 ppm for which the H,29Si
HMQC NMR shows a weak cross-peak correlation with a 29Si
resonance at δ 35.6 ppm; the latter being ~ 21 ppm downfield
from the value found for “free” H3SiPh (i.e., δ 56.7 ppm).
1H,29Si HMQC NMR of [2c]+ (Figure 1a) also shows that the 29Si
resonance at δ 35.6 ppm correlates with several other 1H
resonances of the Ir-bound Cp* (s,
δ 1.79 ppm) and
1
PhSi(Hα)(Hα') [Hα‘: δ 3.84 ppm, JH-Si= 197 Hz; Hα: δ 4.40 ppm,
1JH-Si= 224 Hz; Ph: δ 6.19, 6.46, 6.65 ppm] and to one of the
“aromatic” hydrogens of the {C,N}-ligand (δ 7.83 ppm). The Hα
and Hα’ signals appear as doublet of doublets because of the
two distinct couplings Hα–Si–Hα’ (1JH-H= 12 Hz) and Hα/Hα‘–Si–Ir–
H [3J(Hα‘-H)= 3.6; 3J(Hα-H)= 2.6 Hz] couplings. The 29Si,1H
cross-peak at δ {35.6; 7.82} ppm correlates proton HAr (labelled
as HAr in Figure 1a, δ 7.82 ppm) with the Si atom, revealing
scalar coupling from the Si atom through Ir and the carbanionic
C
ipso atom of the benzo[h]quinolinyl ligand. Theory suggests that
the nature of the bonding of the Si centre to the hydridoiridacycle
in [2c]+ (Figure 1a) is subject to secondary interactions with the
neighboring hydrido ligand and the aromatic chelating ligand.
DFT investigations revealed the existence of two low lying
conformational isomers of the (Ir-H)-silylium complex [2c]+, i.e.
[2c-r1-r2]+ (Figure 1b), among which [2c-r1]+ is the lowest lying
and the one for which the computed absolute value of the scalar
coupling constant J(Si-HAr) is the highest. Interestingly, [2c-r2]+,
i.e. the conformational isomer displaying the shortest HIr-Si
distance, has the highest energy and the lowest computed[23-24]
J(Si-HAr) coupling constant.
Bader’s QTAIM analysis of the structure of [2c-r1]+ did not reveal
any Si-CAr bonding interaction inspite of a rather short Cipso-Si
distance of 2.819 Å (3.019 in [2c-r2]+); the weak Wiberg[25] indice
for the Cipso-Si interaction of 0.12 (0.09 in [2c-r2]+) seemingly
rules out the hypothesis that HAr-Si coupling outlined in Figure
3
1b might arise from a J scalar coupling resulting from a weak
covalent interaction between the Ir-bound Cipso and the Si centre
rather than from a 4J coupling.
Figure 1. a) Excerpt of the 1H,29Si HMQC NMR spectrum of [2c][BArF24
]
(CD2Cl2, 213 K) optimized for medium-low values of J(1H,29Si): the circled
cross-peak correlation corresponds to scalar-coupled Si and HAr nuclei . b)
Computed low lying rotamers of [2c]+ and relevant NMR coupling constants
compared to [2b]+[18];EZPE is the relative energy including zero point energy
contribution; Ni/N is the Boltzmann weight; scalar coupling constants were
computed by the DFT-based CPL[23] method. The J(Si-HCp*
) coupling
constants for the 15 1H nuclei of the Cp* ligand in the static DFT model of [2c-
r1]+ were estimated to span -0.5 Hz and 0 Hz; the significant coupling of the
Cp* protons in a) is consistent with a time-averaged 4J scalar coupling.
Figure 2. ETS-NOCV analysis of the interaction of silylium [SiH2Ph]+ with the
hydrido-iridium residue of [2c-r1]+. The deformation density isosurface (0.007
e/bohr3) 1 shows that most of the donation of electron density operates
mainly from a delocalized interaction of the Si centre with the Ir, hydridic H and
carbanionic Cipso centres to build up a rather diffuse 4 centre interaction where
the Ir-Si component dominates. Red and blue lobes are associated to
donating and accepting orbital components respectively.
Results and Discussion
In this study one shall focus on the H3SiPh/[1b]+ interaction.
Nonetheless, identical behavior was noticed when using H3Si(n-
2
This article is protected by copyright. All rights reserved.

10.1002/chem.201804048
Chemistry - A European Journal
ARTICLE
This “secondary interaction” becomes relevant if one looks at the
extended transition state- natural orbital for chemical valence[26]
(ETS-NOCV, Figure 2) analysis of the interaction of cation
[H2SiPh]+ with the neutral hydrido-iridium residue in [2c-r1]+.
Large electron density transfer to the rather diffuse Ir-Si bond
occurs from three donor orbital contributors located at the Ir
atom, the Ir-HIr bond and atom Cipso; the Ir-Si interaction being
seemingly the strongest with a Wiberg[25] bond indice of 0.44
(w(HIr-Si)= 0.18). The sum of Wiberg indices for Si-Cipso, Si-HIr
and Si-Ir intaeraction (w= 0.74) suggests that secondary
interactions of HIr and Cipso with the Si centre contribute to the
cohesion of [2c-r1]+. Complex [2d][BArF24] was generated
quantitatively by addition of H3SinBu (3.5 eq.) to a solution of
[1b][BArF24] (1 eq.) in CD2Cl2 at -40 °C. Subsequent NMR
analysis revealed features consistent with a donor-acceptor
structure similar to [2c][BArF24] (cf. exptl part and SI for NMR
details). The Ir-bound hydrido ligand resonates at -11.6 ppm
with no visible cross-peak correlation with the Si nucleus at δ -
31.6 ppm.
THF promotion of the formation of Ir-silylenes and their
entrapment by reactive crystallization
Repeated attempts to grow crystals of either [2c][BArF24] or
[2d][BArF24] failed. However, addition of minute amounts of THF
induced the formation of new yellow crystals, the X-ray
diffraction analysis of which revealed the structure of THF-bound
silylene complexes [3][BArF24] and [4][BArF24] (Scheme1).
Owing to their high reactivity all attempts to isolate pure
[3][BArF24] and [4][BArF24] at a preparative scale failed, the
reaction medium delivering after workup either starting complex
[1b][BArF24] or a hydrido iridacyclic dimer already reported in
a previous article[18] 1H NMR monitoring revealed that when the
.
reaction of [1b][BArF24] with H3SiPh was initiated in the presence
of excess THF at 60 °C, warming to room temperature induced
the concomitant production of H2 and [3][BArF24] (Scheme 1). It
was furthermore found that complexes [3][BArF24] and
[4][BArF24] may be obtained either sequentially from [1b]+ by
treatment of [2c][BArF24] and [2d][BArF24] with THF (route 1,
Scheme1) or directly by reacting [1b][BArF24] with H3SiR in pure
d8-THF (route 2, scheme 1). Furthermore 1H NMR monitoring of
the reaction of [2c][BArF24] (formed in situ by reaction of a 85
mM CD2Cl2 solution of [1b][BArF24] with 5 eq. of H3SiPh at -
60°C) with ca. 5 equiv. of THF at 25°C in a J-Young sealed tube
showed rapid release of H2 gas in solution and full conversion
into [3][BArF24] within about 20 min (Figure 3a).
Worthy to note, in an attempt to characterize adducts [2c-d]+ by
electrospray mass spectroscopy (heater temperature: 200°C) in
the positive ion mode, analysis of the total ion current revealed
in both cases
a
main peak corresponding to the
[(benzo[h]quinolinyl-C,N)IrCp*]+ cation at m/z 506 da.
Nonetheless, relevant minor signals could also be detected and
assigned to a new ion resulting from [2c-d]+ having formally
gained one oxygen atom. The most illustrative example is the
one obtained from a fresh CH2Cl2 solution of [2d]+ injected
directly in the ionization chamber without any preliminary
treatment, which presented reasonable signal to noise ratio
(Figure 3b). It is speculated that this ion might be an iridium-
silylene complex having captured or reacted with a molecule of
water. The structures of [3][BArF24] and [4][BArF24] were
obtained from crystals grown in a narrow tube by diffusion of a
heptane solution of silane into a solution of [1b]+ in a mixture of
CH2Cl2 and THF through a buffer layer of dry benzene[18], a
“reactive crystallization” which proved to be the only way to trap
crystals of the key Ir-silylene complexes. The structures
displayed in figures 4a-b reveal that Si-Ir interatomic distances
remain somewhat in the range of those reported for exogenous
donor-devoid Ir(III)-silylene complexes[6, 15a, 27](Ir-Si ~2.25 Å), of
the triflate-bonded Ir-silylene of Klei, Tilley and Bergman[11] (Ir-Si
~2.30 Å) and other known neutral donor-bound metal-silylene
complexes[3c, 3e, 14a, 28] (M-Si(donor)RR ~2.26-2.32 Å). The Ir-Si
distances are by ~ 0.2 Å consistently shorter in [3]+ and [4]+ than
in the structures of [2a,b]+.[17-18]
Figure 3. a) Focussed view on typical signals of the 1H NMR spectrum
monitoring as a function of time the conversion of a freshly prepared solution
of [2c]+ (made from [1b]+, cf. dashed arrow) into [3]+ upon addition of THF in
the CD2Cl2 solution at 25°C; spectra were acquired at a 3 min time interval,
time t=0 corresponds to the first spectrum that was recorded around 3-4 min
after introduction of the sealed J-Young NMR sample tube containing the
mixture in the spectrometer’s probe. b) Isotopic signature of one of the minor
products of fragmentation of [2d][BArF24] injected as a solution in CH2Cl2 in the
ionisation chamber of an ESI-MS (positive ion mode) spectrometer.
3
This article is protected by copyright. All rights reserved.

10.1002/chem.201804048
Chemistry - A European Journal
ARTICLE
NOCV) analysis, a fragment-based method that allows the
symmetry-wise decomposition of inter-fragment orbital
interactions, their energetic contribution to the overall orbital
interaction energy and an intuitive visualization of how electron
density is redistributed in the wide variety of molecular natural
orbitals and bonds of the product.
Figure 4. a) ORTEP-type drawing at the 30% probability level of the structure
of [3]+ determined by structural X-ray diffraction analysis; [BArF24]- anion and
atoms corresponding to occupational disorder were omitted for the sake of
clarity. Selected interatomic distances (Å) and angles (deg) for [3][BArF24]:
Si1—Ir1 2.278(4), C1—Ir1 2.06(3), Si1—H1 1.3(2), O1—Si1 1.883(15), N1—
Ir1 2.08(4), C1—Ir1—N1 79.1(9), C1—Ir1—Si1 89.6(7), N1—Ir1—Si1 88.5(9).
b) ORTEP-type drawing at the 30% probability level of the structure of [4]+
determined by structural X-ray diffraction analysis; [BArF24]-, benzene and
atoms corresponding to occupational disorder were omitted for the sake of
clarity. Selected interatomic distances (Å) and angles (deg) for [4][BArF24]:
Si1—Ir1 2.2882(12), C1—Ir1 2.060(5), Si1—H1 1.40(7), O1—Si1 1.825(4),
N1—Ir1 2.084(4), C1—Ir1—N1 79.3(2), C1—Ir1—Si1 91.02(12), N1—Ir1—Si1
87.25(10).
Figure 5. a) Main deformation density isosurface plots (0.005 e/bohr3)
resulting from the ETS-NOCV analysis of the interaction of
a neutral
“prepared” silylene Si(H)Ph fragment with the cationic iridacyclic residue in
THF-devoid [3]+. b) Main deformation density isosurface plots (0.005 e/bohr3)
resulting from the ETS-NOCV analysis of the interaction of
a neutral
“prepared” THF-bound silylene Si(H)Ph(thf) fragment with the cationic
iridacyclic residue in [3]+. The double arrows indicates which bond is virtually
disrupted in the fragmentation. Red and blue colored isosurfaces correspond
to donating and accepting orbitals respectively. Orbital interaction energies
Ek associated to a deformation density isosurface are mentioned underneath
each isosurface plot. Selected distances (Å) for the computed model of [3]+: Ir-
Si 2.302, O-Si 1.877. Selected distances (Å) for the computed model of THF-
devoid [3]+: Ir-Si 2.246.
Theoretical Investigations of Ir-Si bonding
The bond structure in transition metal-“silylene” complexes has
attracted a great deal of interest.[29][30] Several investigations
have particularly focused on the -donor/-acceptor character of
silylenes and its tuning by the choice of substituents.[31] Also
addressed were the effects of the nature of the metal on the
dominant electronic configuration and its related effect on the -
30d]
accepting properties.[30c,
Several tools and methods[29-30]
were used to address these questions in the past and decisive
Assuming the bonding of a neutral :SiRH(thf) moiety (Figure 5)
to the cationic iridacyclic residue, molecular fragment-based
extended transition state-natural orbital for chemical valence
(ETS-NOCV) analysis[26a] indicates that despite the THF-Si
interaction a significant -accepting character (2 in Figure 5b)
resides at the Si centre in [3]+ (Wiberg[25] Ir-Si bond indice –or
wbi = 0.68) and [4]+ (wbi= 0.67). This residual -accepting
character is revealed by the metal-to-Si -backdonation
deformation density plot 2 (Figure 5b) that represents about
[30c,
conclusions arose from ab initio[32]
30d]and DFT studies[33]
,
which stressed that metal-silylenes of late transition metals
generally had a -accepting character significantly lower than
the parent (Fischer) carbene ligands. Of particular interest is the
work of Ziegler et al.[33] who made widespread use of an energy
decomposition analysis (EDA)-derived approach to weight the -
donating vs. -accepting contributions.
In this study we resorted to both the EDA and the extended
transition state-natural orbital for chemical valence[26] (ETS-
4
This article is protected by copyright. All rights reserved.

10.1002/chem.201804048
Chemistry - A European Journal
ARTICLE
10% of the total interfragment orbital interaction energy.
Deformation densities 1 in Figures 5a and 5b materialize
properly the electron density transfers occurring upon build up of
the Si-to-Ir  bond according to the chosen interaction scheme.
Noteworthy, the ETS-NOCV[26a] analysis of the THF-devoid
analog of [3]+ (Figure 5a) confirms the -accepting property of
the :SiPh(H) moiety and the consolidation of the Ir-Si bond(wbi
(Ir-Si) = 0.83) that is materialized by deformation density 2. In
this case the metal-Si -backdonation orbital interaction (Figure
5a) represents about 20% of the interfragment orbital interaction
energy.
Figure 6. Main deformation density isosurface plots (0.005 e/bohr3) resulting
from the ETS-NOCV analysis of the interaction of a neutral “prepared” silylene
SiPh2 fragment with the cationic Ir(III) residue of an Ir-silylene complex
reported by Calimano and Tilley.[6] The double red arrow suggests which
bond(s) is(are) cleaved in the fragmentation mode. 1 corresponds to the
establishment of the Si-to-Ir -bond.
2 corresponds to Ir-to-Si 
backbonding, which associated interaction energy Ek represents about 14%
of the total interfragment orbital interaction energy. Red and blue-colored
isosurfaces correspond to donating and accepting orbitals respectively. The
Wiberg bond indice for the Ir-Si interaction amounts 0.81.
Figure 7. a) Proposed Gibbs energy profile for the THF-assisted formation of
[3]+ as computed by DFT (Gibbs enthalpies in kcal/mol) assuming Coulomb
interaction-supported H-atoms interaction in TS-I-thf; the Gibbs enthalpy
(298.15 K, 1 atm) for the [1b]+ → [3]+ reaction is ca. 13 kcal/mol. b) Detailed
description of the Wiberg bond indices (in blue fonts) and (NPA) natural
charges (in red fonts) at the core motif undergoing bond cleavage in transition
states TS-I-thf and TS-I.
A pertinent reference case reported by Calimano and Tilley[6] is
the cationic Ir(III)-silylene complex devoid of any donor-acceptor
interaction with the Si centre shown in Figure 6, for which the
ETS-NOCV analysis carried out assuming also the interaction
between a neutral :SiPh2 silylene and a cationic Ir-centred
residue indicates that Ir-to-Si -backdonation represents only
14% of the total inter-fragment orbital interaction energy, which
is less than for the fictitious THF-devoid [3]+ and most probably a
consequence of the weaker donating properties of the ligand set
in the PNP complex (Figure 6).
Investigation of natural atomic charges gives some interesting
hints about the driving force of this reaction triggered by THF
coordination and outlines the “protean character” of the Ir-bound
hydrido ligand capable of behaving either as H- in key
hydrosilylation steps or as H+ like shown here. Bonding of THF
to the Si centre induces indeed[4b] a net increase of the formal
natural[34] negative charge of the Si-bound H atoms and a 50%
increase of the natural charge difference q between the HSi1
and HIr atoms in TS-I-thf as compared to TS-I (q(HIr-HSi)TS-I-thf
=
Gibbs Energy Profile of Formation of the THF Bound Ir-
Si(H2)Ph Silylene Complex
0.26, q(HIr-HSi)TS-I= 0.17) (Figure 7b). The formation of H2 is
essentially assisted by coulombic interactions between HSi1 and
HIr. In other words the relative “acidity” of the latter Ir –bound
hydrido ligand HIr and the hydridic character of HSi1 enhanced by
the interaction of THF facilitate H2 release. This conclusion was
The computed reaction energy profile confirmed the key role
played by THF in the H2-elimination step (Figure 7). The
formation of [2c]+ is exoergonic at 298K (Figure 7a). In the
absence of THF, the concerted coupling of the hydridic Si-bound
HSi atom with the Ir-bound HIr atom requires a prohibitive
activation energy G°≠ of ~ +30 kcal/mol (via transition state TS-
I, Figure 7b), the overall process leading to the THF-free silylene
complex being endoergonic (G°~ +11 kcal/mol). In turn, in the
presence of 1 molecule of THF (perfect gas conditions in
vacuum), the formation of van der Waals complex [vdW-2c]+
requires ca. +17 kcal/mol.
This “barrier” that may appear unfavorable at first glance, is
practically to vanish if THF is in large excess or the solvent of
the reaction, like shown experimentally, which makes the
intermediacy of [vdW-2c]+ plausible. The conversion of [vdW-
2c]+ into THF-stabilized “silylene” complex [3]+ requires ca. +14
kcal/mol of activation barrier via TS-I-thf with an exoergonicity
G°([2c]++ THF→ [3]+) of 4 kcal/mol, confirming that in pure
THF the formation of [3]+ is to be spontaneous.¶
already drawn by Corriu and coworkers4b] for
observation on Fe complexes.
a similar
ETS-NOCV[26a] analysis (Figure 8) further suggests that the
transformation of [2c]+ into [3]+ implies a major change within the
Ir-Si interaction. Indeed, to gain further insight into the electron
density redistribution resulting from the departure of H2 an ETS-
NOCV analysis was carried out by fragmenting TS-I-thf in two
fragments consisting namely of H2 and [3]+ in their static so-
called “prepared geometry”. This fragmentation allows the
intuitive visualization of the density transfers occurring
particularly into the complex Ir-HIr-HSi-Si motif. By virtue of the
microreversibility principle one can state from deformation
density 1 (Figure 8) that the departure of H2 entails: 1) electron
density donation to the Ir centre and the Ir-Si  bond from the 
Ir-HIr and Si-HSi bonds, 2) minor donation from the Si-bound THF
5
This article is protected by copyright. All rights reserved.

10.1002/chem.201804048
Chemistry - A European Journal
ARTICLE
oxygen atom, and 3) crucial transfer of density to HIr, which
rebalances the charge density between this hydrogen atom and
disclose that Ir-silylene species might well be responsible for the
hydrodefluorination capabilities that are disclosed below.
H
Si in the process of H2 formation.
NMR properties of the THF-bound Ir-Silylenes
1H NMR analysis (25 °C, CD2Cl2) of [3][BArF24] disclosed a
1
singlet at δ 5.73 ppm with a JH-Si= 99 Hz assigned to the Si-
bound H atom, as well as the typical signals of the Ir-bound Cp*
1
(δ 1.74 ppm) and {C,N}-ligand (δ 7.88-9.02 ppm). H,29Si HMQC
NMR experiment recorded at 60 °C (CD2Cl2) displayed a cross-
peak correlation between the 29Si satellites of the broad 1H
resonance of the Si–H bond at δ 5.59 ppm and the 29Si
resonance at δ 72.8 ppm. Although 1H signals of free THF
overlapped those of the Si-bound THF at room temperature, at
60 °C the latter signals sharpened as four broad peaks
integrating for ca. 2H each at δ 0.65, 0.85, 2.05, and 2.30 ppm
respectively as the consequence of the coordination of THF to
the stereogenic Si centre. The existence of
a
minor
Figure 8. Plot of deformation density 1 (0.005 e/bohr3) arising from the
ETS-NOCV analysis of the fictitious interactions of the prepared fragments of
H2 and THF-bound Ir-silylene cation “leading” to transition state TS-I-thf (the
double red arrow indicates which “bonds” are cleaved in the fragmentation
scheme). The red and blue isosurfaces (0.005 e/bohr3) are associated with
electron density donating and accepting orbitals respectively. The interaction
of H2 with prepared [3]+ suggests a strong polarisation of the H-H bond and
donation to the Ir-HIr and HSi1-Si  bonds from HIr, Ir centres and from Si-Ir -
bond. In the reverse transformation, that is the H2 release and its dissociation
from [3]+, orbitals depleted in electron density are now those colored in blue
and the accepting orbitals are colored in red.
diastereomer of [3][BArF24] was inferred from the presence of a
second set of such THF-related peaks. 1H,1H COSY NMR
indicated cross-peak correlations between the four signals of the
Si-bound THF. 1H,1H NOESY NMR showed intense and positive
cross-peak correlations between the latter and the two signals of
free THF thus confirming chemical exchange (cf. SI).
Noteworthy, the δ (29Si) shift of resonance of ca. +108 ppm on
going from [2c][BArF24] to [3][BArF24] is consistent with a
substantial depletion of electron density at the Si centre.[4b] 1H
NMR analysis (-40 °C, d8-THF) of [4][BArF24] disclosed a triplet
1
at δ 4.78 ppm with a JH-Si = 95 Hz for the Si-bound H atom, as
well as the typical signals of Si-coordinated THF (four broad
peaks at δ 0.65, 0.83, 2.07, 2.31 ppm). The 1H,29Si HMQC NMR
experiment (-40 °C, d8-THF) showed the presence of a cross-
1
peak correlation between the H resonance of the Si–H bond at
δ 4.78 ppm and the 29Si resonance at δ 71.3 ppm (cf. exptl part
and SI for more details).
Hydrodefluorination (HDF)
The carbon-fluorine (C–F) bond is amongst the strongest single
bonds that exist in nature.[35] The hydrodefluorination (HDF)
method, which consists of simply replacing fluorine by hydrogen,
has emerged as a popular strategy towards selective and
catalytic C-F activation.[36]
There are numerous studies
Figure 9. Extrusion of H2 from TS-I-thf (left handside) redistributes electron
density in such a way (red arrows) that the Ir-Si bond is consolidated, leading
to [3]+ (right handside) where the exact nature of the Si centred ligand remains
ambiguous in spite of its significant residual -acidic character (dashed bond).
describing catalytic HDF based on main-group Lewis acids (R3B,
R3Al, R3Si+) in combination with either main-group or transition
metal hydrides as reducing source.[36a, 36b, 37] In 2000, Jones et
al. showed for the first time that the HDF reaction of alkyl
fluorocarbons can be catalyzed by a homogeneous transition
Figure 9 depicts the electron density (red arrows) redistribution
in the H2-release process according to the ETS-NOCV analysis,
metal-based catalyst ([ZrH2Cp*2]/H2)[38]
Crabtree et al. and Richmond et al. were the first to describe the
, whereas in 1996
which entails evolution from
a
donor-acceptor Ir→Si
interaction[17-18] (wbi~ 0.44, “silylicity” [22] related to the
[SiPhH2]+ moiety ~1.3) in [2c]+ to either a reversed “dative” Ir←Si
or a coordinative Ir-Si bond bearing some Ir→Si  character in
[3]+. Similar conclusions can be drawn for [2d]+ and [4]+
(“silylicity” related to [SinBuH2]+  ~1.3). It is worthy to note that
such high “intrinsic silylicities”[22] were observed essentially for
cationic metal-silane adducts displaying catalytic activity in
hydrosilylation reactions.
stoichiometric versions of this reaction by using respectively
[FeCp*2]/Zn[39] and [TiCp2F2]/Al/HgCl2
combinations. Many
[40]
other complexes of Co[41], Ru[42], Pd[43], Rh[44], Fe[45], and Cu[46]
were applied as homogeneous catalysts for the HDF reaction.
On the other hand, few iridium complexes have been employed
for the catalytic C–F bond activation of fluorocarbons. Brookhart
et al. reported the first example[47] followed by more recent work
[48]
by Weaver et al.
and by Kayaki et al.[49]
.
Because
Klei, Tilley and Bergman[4e] have already pointed out the
ambiguous properties of cationic metal silylene complexes that
display a rather weak -acidity,[33a] while their reactivity is much
intermediates [2c-d][BArF24] can be viewed as a source of both
“Et3Si+” and “H–“, it was initially wondered whether they could be
applied to the catalytic HDF reaction of fluorocarbons, since it
was shown in several occurrences that electrophilic silylium
cation may be used in HDF.[37, 50]
more consistent with that of (metalla)silyliums due to the strong
7]
charge depletion residing at the Si centre.[4a,
Therein, we
6
This article is protected by copyright. All rights reserved.

10.1002/chem.201804048
Chemistry - A European Journal
ARTICLE
Preliminary investigation on the catalytic HDF reaction
Initial attempts were unsuccessful, as illustrated with the
reaction of 1-fluoroheptane 1-FH (1 eq.) with HSiEt3 (1.2 eq.) in
the presence of a solution of [3a][BArF24] (2 mol % relative to 1-
FH) in CD2Cl2 (0.5 mL), which proved to lead mainly to
decomposition of the Ir catalyst (Scheme 2a; for more details
related section in SI).
silane upon heating the reaction mixture to 80 °C for 23 h
(Scheme 2d). No conversion was observed with HSiEt3 or
polymethylhydrosiloxane (PMHS) as silane (Table S2, entries 5–
6). Similar results were obtained with the [1b][BArF24]/1-
FH/H3SiPh/PhF catalytic mixture (molar ratio ~ 1/47/142/372), i.e.
complete conversion of 1-FH upon heating to 80 °C for 24 h
(Scheme 2d). Again, no conversion was observed when the
latter reaction (Scheme 2d) was carried out in the absence of
catalyst (“blank run”; Scheme 2e).
Scheme 2. Hydrodefluorination of 1-fluoroheptane (1-FH) to n-heptane, using
as catalyst systems either (a) [1a][BArF24]/HSiEt3 or (b-e) [1a-
b][BArF24]/H3SiPh [a]
Table 1. Comparison of HSiEt3 and H3SiPh in their reaction with different
substrates catalyzed by [1a-b][BArF24]: acetophenone (HSA), benzonitrile
(HSB) and 1-FH (HDF).[a]
HSiEt3 (1.2 eq.)
[1a][BArF24] (2.0 mol % )
HSA yield
HSiEt3
99%
HSB yield
HSiEt3
99%
HDF (1-FH conv.)
(CH2)5CH3
F
(CH2)5CH3
a)
b)
H
CD2Cl2, (0.5 mL), [rt, 24 h]
1%b
1-FH (1 eq.)
H3SiPh
10%
H3SiPh
1%
HSiEt3
1%
H3SiPh
99%
H3SiPh (1.2 eq.)
[1a][BArF24] (2.0 mol % )
(CH2)5CH3
1-FH (1 eq.)
F
(CH2)5CH3
H
 CD2Cl2 (0.2 mL), [rt, 20 h]
[a]
or
HSA: acetophenone (0.2 mL, 1.71 mmol), silane (1.89 mmol), [1a][BArF24
]
 C2H4Cl2 (0.5 mL), [50 °C, 24 h]
(2.4 mg, 1.71 μmol, 0.1 mol %), CH2Cl2 (0.5 mL), rt, 30 min. HSB: benzonitrile
(0.2 mL, 1.94 mmol), silane (4.23 mmol), [1b][BArF24] (27.3 mg, 19.4 μmol, 1.0
mol %), CH2Cl2 (0.5 mL), 80 °C, 24h. HDF: 1-FH (0.1 mL, 0.68 mmol), silane
(0.82 mmol), [1a][BArF24] (20.0 mg, 14 μmol, 2.0 mol %), PhF (0.5 mL), 80 °C,
23 h.
5%b
H3SiPh (1.2 eq.)
[1a][BArF24] (2.0 mol % )
H
(CH2)5CH3
(CH2)5CH3
c)
F
PhF (0.5 mL), [80 °C, 23 h]
1-FH (1 eq.)
99%b
H3SiPh (3 eq.)
Scheme 3. Other HDF application of the [1b][BArF24]/H3PhSi catalytic
system.[a]
[1b][BArF24] (2.0 mol % )
H
(CH2)5CH3
(CH2)5CH3
(CH2)5CH3
d)
e)
F
PhF (0.2 mL), [80 °C, 24 h]
1-FH (1 eq.)
99%b
H3SiPh (3 eq.)
H
F
H3SiPh (3 eq.)
[1b][BArF24] (2 mol %)
neat, [80 °C, 24 h]
99% conv.
(CH2)5CH3
H
F
PhF (0.5 mL), [80 °C, 24 h]
a)
b)
1-FH (1 eq.)

1%b
F-Cy (1 eq.)
H-Cy
[a] Cf. Exp. Part and SI for details. [b] Conversion of 1-FH was estimated by
NMR spectroscopy using internal reference standards (Cf. Exp. Part and SI).
H3SiPh (1 eq.)
H
F
[1b][BArF24] (1 mol %)
OBn
OBn
PhF (0.5 mL), [80 °C, 20h]
13% conv.
Influence of the nature of chlorinated solvents.
FCyOBn
(1 eq.)
HCyOBn
We suspected a detrimental influence of the solvent on the
catalytic efficiency. In fact, as supported by results obtained
with HSiEt3 (Table S1; SI) we could not exclude that CD2Cl2
interferes with the HDF reaction since its C–Cl bond activation is
Bn = CH2Ph
H3SiPh (1 eq.)
[1b][BArF24] (5.0 mol % )
PhCF2H (80%)
+
c)
PhCF3
(1 eq.)
PhCH3 (20%)
neat, [100 °C, 20 h]
31% conv.
a
plausible side reaction, as previously observed by M.
Brookhart et al.[47] We further carried out the same study using
H3SiPh instead of HSiEt3 to check whether the nature of the
silane rather than the solvent itself was responsible for the low
efficiency of the catalysis (Table S2; SI). Whereas no reaction
was observed in the absence of [1a][BArF24] while using CD2Cl2
as the solvent, only 5% of conversion could be obtained in the
presence of 2 mol % of [1a][BArF24] using the two chlorinated
solvents CD2Cl2 and C2H4Cl2 (Scheme 2b). Hence, these
results suggested that aliphatic chlorinated solvents were
detrimental to the HDF reaction. Our choice of solvent turned to
fluorobenzene (PhF).
[a]
Conditions: a) F-Cy (0.05 mL, 0.46 mmol), H3SiPh (0.17 mL, 1.37 mmol),
[1b][BArF24] (13.0 mg, 9.2 μmol; 2.0 mol %); b) FCyOBn (30 mg, 0.14 mmol),
H3SiPh (0.02 ml, 0.14 mmol), [1b][BArF24] (2.0 mg, 1.4 μmol; 1.0 mol %); c)
PhCF3 (0.12 mL, 0.97 mmol), H3SiPh (0.12 ml, 0.97 mmol), [1b][BArF24] (73.0
mg, 52 μmol; 5.0 mol %).
Conversions were determined by NMR
spectroscopy against 1,3,5-trimethylbenzene and hexafluorobenzene as 1H
and 19F internal references respectively (cf. SI for full details).
Comparison of HSiEt3 and H3SiPh in other reactions
The fact that H3SiPh was the only active silane under the
selected conditions strengthened suspicion of the involvement of
an Ir-“silylene” related to [3][BArF24] in the mechanism of the
HDF catalysis. For comparison, Table 1 summarizes the
catalytic performance of H3SiPh and HSiEt3 in HDF and other
reactions reported by us in previous studies.[17-18] These data
clearly show that HSiEt3 performs better than H3SiPh in the
hydrosilylation of acetophenone (HSA) and auto-tandem double
hydrosilylation of benzonitrile (HSB) (Table 1) [and O-silylation
of benzyl alcohol (not shown here)]; the mechanisms of which
were proposed to involve [2a-b][BArF24] (scheme 1) as
intermediates.
Beneficial effect of the H3SiPh/PhF combination.
PhF was further used as the solvent because of its inertness
towards the HDF reaction (as shown with a blank test where
PhF was used as substrate and solvent; cf. Scheme S1a), its
relatively high boiling point (85 °C at 1 atm), and its ability to
dissolve [1a,b][BArF24]. In a first test using PhF (0.5 mL) as
solvent and 2 mol % of [1a][BArF24], we studied the influence of
the nature of silane on the conversion of 1-FH to n-heptane
(Table S2, entries 4-6). It was found that total conversion of 1-
FH (1 eq.) was obtained when H3SiPh (1.2 eq.) was used as
Indeed, whereas quantitative yields were obtained for HSiEt3 in
all reactions but HDF, using H3SiPh dropped the yields to 5% for
7
This article is protected by copyright. All rights reserved.

10.1002/chem.201804048
Chemistry - A European Journal
ARTICLE
HSA and 0% for HSB while leading to quantitative conversion for
the HDF of 1-FH. Thus, data listed in Table 1 clearly indicate
that H3SiPh does not perform as efficiently as HSiEt3 for any of
the reactions where the intervention of an electrophilic “[SiR3]+”
moiety is established, i.e. HSA and HSB, hence supporting
again the idea of a mechanism involving a “silylene” related to
[3][BArF24] as a plausible intermediate for the [2c][BArF24]/1-
FH/H3SiPh catalytic system. Mechanistic studies also show that
this hypothesis is viable (vide infra).
observation confirm earlier observation by Tilley on the effect of
a positive charge on the electrophilicity of the electropositive Si
centre.[1a] Our main objective remains the synthesis of more
manageable metallacyclic Ir-silylene complexes that would be
persistent enough in the solid state to allow comprehensive
reactivity screening.
The disclosed results suggest that
[3][BArF24] is active in the C—F bond activation. However,
elucidating the actual mechanism of the HDF reaction operating
in the absence of THF requires to probe first the reactivity of
more stable cationic metallacyclic Ir-silylene complexes. Further
investigations aiming at establishing the exact nature of the key
intermediates in HDF entail the investigation of the capability of
PhF or any fluoroalkane to promote the formation of transient
highly reactive Ir-“silylenes”. These tasks that fall out of the
scope of this article will be addressed in due time.
Hydro-de-fluorination of other substrates.
The [1b][BArF24]/H3PhSi catalyst system previously used for the
HDF reaction of 1-FH was also applied to some selected
substrates:
fluorocyclohexane
(F-Cy),
[(2-
fluorocyclohexyl)oxy](phenyl)methane
(2-FCyOBn),
and
trifluoromethylbenzene (PhCF3) (Scheme 3). For F-Cy, total
conversion to cyclohexane H-Cy (besides cyclohexene as main
by-product; cf. SI) could be readily achieved under solvent-free
conditions (Scheme 3a). However, as previously shown with 1-
FH, no reaction was observed either at room temperature or in
the absence of the precatalyst [1b][BArF24]. As shown in
Scheme 3b, the HDF of FCyOBn to its defluorinated product
HCyOBn proved to be less efficient than for 1-FH and F-Cy
(13% conversion). Finally, the HDF of PhCF3 to PhCH3 proved
to be quite challenging. In fact, using 1 eq. of H3SiPh only
31% of PhCF3 was converted to a mixture containing
mainly PhCF2H and PhCH3 in a 1:0.2 ratio (Scheme 3c
and SI)
Experimental Section
General
All experiments were conducted under a dry argon atmosphere using
standard Schlenk and glovebox techniques (unless otherwise indicated).
All glassware was oven-dried prior to use. All solvents were distilled over
sodium or CaH2 under argon before use. Deuterated solvents (including
PhF) were dried over sodium or CaH2, filtered over activated neutral
alumina, and stored under argon before use. The following compounds
were purchased from the indicated companies and used as such (unless
otherwise stated): HSiEt3 (Aldrich, 97%), H3SiPh [(Alfa-Aesar, 97%),
(purified by bulb-to-bulb distillation over CaH2 and stored under argon
over MS-3Å)], PMHS (Alfa-Aesar), 1-fluoroheptane (1-FH) (Aldrich,
98%), fluorocyclohexane (F-Cy) (TCI-Chemicals, > 98%). Ligand name
abbreviations: Cp*, (1,2,3,4,5-pentamethyl-η5-cyclopentadienyl); PhPy, 2-
(2-pyridyl)phenyl, B[h]Q, benzo[h]quinolinyl. [1a-b][X] [X= BArF24 , BPh4
(a-b); X= BF4 (a); X= OTf (b)] were prepared following published
procedures.[17-18] 1H (300, 400, 500, and 600 MHz), 13C (75 and 126
MHz), 19F (282 MHz), and 29Si (119 MHz) NMR spectra were measured
on Bruker DPX 300 and 400, Avance I 500, and Avance III 600
spectrometers. All chemical shifts () are expressed in parts per million
(ppm). For 1H and 13C NMR, values of  are reported relative to Me4Si as
an external reference standard and referenced against peaks of solvents
Involvement of THF-bound Ir-Silylene in the Hydro-de-
fluorination of fluoro-alkanes.
To gain mechanistic insights, the reaction of [3][BArF24] with
fluorocyclohexane (~1:7 ratio) was carried out at 25°C in a
sealed NMR tube and monitored by 1H NMR spectroscopy.
After 10 min of reaction, ~14% of fluorocyclohexane was
converted into cyclohexane and cyclohexene (~1:2.5 ratio), and
~52% after 2 h (~1:3 ratio). If almost all of the initial amount of
[3][BArF24] was still present after 10 min, only ~25% remained
after 2 h (cf. SI).
1
(only partially deuterated solvent is visible in H) as secondary reference
standards. For other nuclei, the reported values of  are referenced to
external reference standards (CF3C6H5 in CDCl3 for 19F, Me4Si in CDCl3
for 29Si). All ESI-MS spectra were acquired in the positive mode on a
Bruker MicroTOF II spectrometer.
Conclusion
In summary, this study confirms that the reaction of Et3SiH
and H3SiR (R= Ph, nBu) with [1b]+ entails the disruption of a Si-H
bond leading to intermediates [2b-d]+. NMR data and DFT
calculations suggest that the cohesion of [2c-d]+ might rely on
secondary interactions of the Si atom with the Ir-bound hydrogen
atom and the vicinal chelating ligand. The formation of THF-
stabilized silylene complexes of Cp*-based iridacycles
represents an unexpected advance for this class of
metallacycles[51] that promote diverse behaviour with
hydrosilanes like shown here. The production of silylene
complexes by a sequential double Si-H activation promoted by
THF and leading to the irreversible extrusion of H2 brings to the
forefront the role of the “protean” Ir-bound hydrido ligand, which
is seemingly capable of adopting a Lewis acid, base or even a
radical character if one extrapolates the conclusion drawn by Yu
and Norton[52] for related Rh-H complexes. The promising
potential of reactive complexes [3]+ and [4]+ is currently being
explored, having in sight further possible role as reactive
transients in catalysis. From a fundamental view point our
X ray diffraction analyses
X-Ray diffraction data collection for [3][BArF24] and [4][BArF24] were
carried out on a Bruker APEX II DUO Kappa-CCD diffractometer
equipped with an Oxford Cryosystem liquid N2 device, using Cu-Kα
radiation (λ = 1.54178 Å) for [3][BArF24] and Mo-Kα radiation (λ =
0.71073 Å) for [4][BArF24]. The crystal-detector distance was about 40
mm. The cell parameters were determined (APEX2[53] software) from
reflections taken from three sets of 20 frames, each at 10s exposure. The
structures were solved by Direct methods using the program SHELXS-
2014.[54] The refinement and all further calculations were carried out
using SHELXL-2014.[55] For [3][BArF24], the H-atom H1 bound to Si1 was
added in calculated position and then refined without the “AFIX”
instruction. For [4][BArF24], the H-atom H1 was located by Fourier
difference. 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 semi-empirical absorption correction was applied using
SADABS in APEX2.[53] In [3][BArF24], the benzoquinolinyl ligand is
disordered over two positions, the THF molecule and the phenyl group
linked to Si1 are disordered over two positions (all in all), the fluorine
atoms F4, F5, F6, F14, F15, F19, F20,F21, F22, F23 and F24 from the
8
This article is protected by copyright. All rights reserved.

10.1002/chem.201804048
Chemistry - A European Journal
ARTICLE
counter-anion are disordered over two positions. In [4][BArF24], the C-
atoms of the THF molecule, and the fluorine atoms F7, F8, F9, F10, F11,
F12, F22, F23 and F24 are disordered over two positions. The fluorine
Table S5 and S6 list acquisition and refinement data for [3][BArf24] and
[4][BArF24].
BArF24), 117.6 (m, C-Hpara in BArF24), 103.1 (Cp-Me5), 8.69 (Cp-Me5).
29Si-DEPT NMR (119 MHz, 213 K, CD2Cl2): δ = -35.64 (s, [Ir-H]-[SiH2Ph]).
Data for “free” H3SiPh are as follows: 1H NMR (600 MHz, 213 K, CD2Cl2):
δ = 7.58-7.59 (m, 1H, PhSiH3), 7.44 (m, 2H, PhSiH3), 7.35-7.37 (m, 2H,
1
PhSiH3), 4.13 (s, 3H, PhSiH3, JH-Si = 200.1 Hz. 29Si-DEPT NMR (119
MHz, 213 K, CD2Cl2): δ = -58.75 (s, PhSiH3). Data for “free” CH3CN are
as follows: 1H NMR (600 MHz, 213 K, CD2Cl2): δ = 1.98 (s, 3H, CH3CN);
13C (151 MHz, 233 K, CD2Cl2): δ = 117.7 (NCMe), 2.33 (NCMe).
Computational Details
Computations were performed with the methods of density functional
theory using the SCM-ADF[56]2016.01 package considering the
molecules in the gas phase. The PBE[57] functional augmented with
In situ Formation of [2d][BArF24].
Grimme’s DFT-D3(BJ) implementation of dispersion with
a
In a glovebox, H3SinBu (20.0 μL, 77.0 μmol) was added to a solution of
[1b][BArF24] (30 mg, 21.2 μmol) in CD2Cl2 (0.45 mL). The resulting
solution was shaken and transferred into a J. Young NMR sample tube
which was subsequently tightly sealed for analysis. After 25 minutes of
reaction, the NMR tube was frozen at < -40 °C. Multinuclear NMR
analysis (at -40 °C) of the reaction revealed the total conversion of
[1b][BArF24] to [2d][BArF24], along with an unknown Ir-based side product
(based on Cp*), the released CH3CN (from [1b][BArF24]) and excess of
HSinBuH2 with a relative ~1:1:6 ratio (see NMR spectra below). Isolation
of pure [2d][BArF24] was attempted in many circumstances by varying the
conditions of the reaction and the methods of purification, all of which led
Becke−Johnson (BJ) damping function[58] was used in all geometry
optimizations. Geometry optimizations by energy gradient minimization
were carried out in all cases with integration grid accuracy comprised
between 4.5 and 6.5, an energy gradient convergence criterion of 10−3
au, and a tight to very tight SCF convergence criterion. Counterpoise
correction for basis set superposition error (BSSE) was neglected
throughout this study. All computations were carried out using scalar
relativistic corrections within the zeroth order regular approximation for
relativistic effects with ad hoc all-electron (abbreviated ae) single
polarization function triple-ζ Slater type basis sets (TZP).[59] Vibrational
modes were analytically computed to verify that the optimized geometries
were related to energy minima or to transition states. The geometries of
[2c], [2d], [3]+, [4]+ and all other intermediates, transition states and
reactants were computed at the ZORA-PBE-D3(BJ)/ae-TZP level.
Intrinsic interaction energies, ETS-NOCV analyses, Natural population
analyses (NPA) as well as Wiberg index determinations were performed
with the gas-phase geometry relaxed at the (ZORA) PBE-D3(BJ) level
using all-electron TZP basis sets with the GENNBO[60] 6.0 module of
ADF. NMR coupling constants were computed with the ADF CPL
module[23-24] at the ZORA-SAOP[61]/all-electron TZP basis sets. Drawings
of molecular structures and isosurfaces were produced with
ADFview2013. A geometry optimized at the ZORA-PBE0[62]-DdsC[63]/all
electron TZP level was computed to draw plots of the Laplacian of the
density for selected molecular planes of [2c-r1]+ that can be found in the
SI.
to decomposition products of unidentified nature.
Therefore, the
characterization of [2d][BArF24] was only possible in situ (solution NMR
solvent). However, appropriate identification of all 1H nuclei could not be
realized due to the formation of a unidentified side-product [much likely of
the formula[IrCp*(B[h]Q)X][BArF24], where X may presumably be a
solvent and certainly not NCMe or H)]. Nonetheless, 1H and 29Si
chemical shifts that are diagnostic of the structure of [2d][BArF24] could
be unambiguously identified (vide infra). 13C nuclei could not be
assigned with certainty to the product of interest due to the complex
nature of the reaction mixture and of the resulting NMR spectrum. 1H
NMR data (600 MHz, 233 K, CD2Cl2): δ = 8.76 (broad m, 1H, HAr B[h]Q),
8.43 (m, 1H, HAr B[h]Q), 7.93 (m, 1H, HAr B[h]Q), 7.86 (m, 1H, HAr B[h]Q),
7.81 (m, 1H, HAr B[h]Q), 7.73 (m, 8H, Hortho BArF24), 7.70 (m, 1H, HAr
B[h]Q), 7.61 (m, 2H, HAr B[h]Q), 7.54 (m, 4H, Hpara BArF24), 3.65 (broad
m, 1H, [Ir-H]-[SiH2nBu]), 3.34 (broad m, 1H, [Ir-H]-[SiH2nBu]), 1.74 (s, 15H,
Cp-Me5), −11.55 (bs, 1H, [Ir-H]-[SiH2nBu]). 29Si [indirect detection from
1H,29Si HMQC; 600/119 MHz, 233 K, CD2Cl2]: δ = -31.60 (s, [Ir-H]-
[SiH2nBu]).
In situ Formation of [2c][BArF24].
In a glovebox, H3SiPh (10.0 μL, 81.1 μmol) was added to a solution of
[1b][BArF24] (29 mg, 20.6 μmol) in CD2Cl2 (0.65 mL). The resulting
solution was shaken and transferred into a J. Young NMR sample tube
which was subsequently tightly sealed for analysis. After 25 minutes of
reaction, the NMR tube was frozen at < -60 °C. Multinuclear NMR
analysis (at -60 °C) of the reaction revealed the total and exclusive
conversion of [1b][BArF24] to [2c][BArF24], along with the released
CH3CN (from [1b][BArF24]) and excess of H3SiPh in a ratio of ~1:1:3 (cf.
SI for NMR spectra). 1H NMR (600 MHz, 213 K, CD2Cl2): δ = 8.71 (d, 1H,
HAr B[h]Q, J = 5.4 Hz), 8.22 (d, 1H, HAr B[h]Q, J = 7.9 Hz), 7.81-7.85 (m,
2H, HAr B[h]Q), 7.75 (m, 8H, Hortho BArF24), 7.68 (d, 1H, HAr B[h]Q, J = 8.8
Hz), 7.55 (m, 4H, Hpara BArF24), 7.51 (d, 1H, HAr B[h]Q, J = 8.0 Hz), 7.50
(d, 1H, HAr B[h]Q, J = 8.0 Hz), 7.44 (d, 1H, HAr B[h]Q, J = 8.7 Hz), 6.65 (t,
1H, [Ir-H]-[SiH2Ph], J = 7.5 Hz), 6.46 (t, 2H, [Ir-H]-[SiH2Ph], J = 7.5 Hz),
In situ Generation of [3][BArF24]; route 1(scheme 1)
In a glovebox, H3SiPh (10.0 μL, 81.0 μmol) was added to a solution of
[1b][BArF24] (15 mg, 10.6 μmol) in CD2Cl2 (0.55 mL). The resulting
solution was shaken and transferred into a J. Young NMR sample tube
which was subsequently tightly sealed for analysis. After 25 minutes of
reaction, the NMR tube was frozen at < -60 °C. 1H NMR analysis (at -
60 °C) of the reaction was carried to verify the total and exclusive
conversion of [1b][BArF24] to [2c][BArF24]. Addition of THF (10 μL, 123.3
μmol) to the latter mixture resulted in immediate effervescence of the
yellow solution that indicated H2 gas evolution. To raise up the
temperature from -60 °C to ~ 20 °C, it was found necessary to let the
latter solution standing in the glovebox at room temperature for ~1-2 h.
1H NMR analysis (at -60 °C) indicated complete consumption of
[2c][BArF24] and formation of [3][BArF24] as the major Ir species along
with H2 (singlet at ca. δ 4.60 ppm) in a respective ca. 1:0.2 ratio (see
NMR spectra below). A complex pattern of peaks of low-intensity were
also observed in the range spanning δ 5.10-5.15 ppm (integrating
roughly for 2H relative to [3][BArF24]); part of the peaks, i.e., a singlet at δ
5.15 ppm, is assigned to the Si-bound hydrogen of an unidentified
organosilicon derivative “PhxSiyHz” (on the basis of weak 29Si satellites;
vide infra). This procedure was found to be reproductible over (at least)
≥ 5 experiments.
6.19 (d, 2H, [Ir-H]-[SiH2Ph], J = 7.5 Hz), 4.40 (dd, 1H, [Ir-H]-[SiH2Ph], J1
H-
1
= 11.8 Hz, JH-Si = 197.0 Hz; J2
= 2.6 Hz), 3.84 (dd, 1H, [Ir-H]
H-H
H
→[SiH2Ph], J1
= 11.8 Hz, JH-Si = 224.0 Hz; J2
= 3.6 Hz), 1.79 (s,
H-H
1
H-H
15H, Cp-Me5), −11.50 (bs, 1H, [Ir-H]-[SiH2Ph]). The reaction described
above was repeated (-40 °C) for recording 13C and 2D related NMR
spectra. 13C (151 MHz, 233 K, CD2Cl2): δ = 161.8 (1:1:1:1 quartet, 1JC−B
= 49.5 Hz, C-B in BArF24), 155.8 (B[h]Q), 152.0 (H-C=N, B[h]Q), 144.6
(B[h]Q), 141.5 (B[h]Q), 138.7 (C-H, B[h]Q), 138.1 (B[h]Q), 136.5 (C-H,
B[h]Q), 135.2 (B[h]Q), 134.2 (C-H, IrSiPhH2), 134.7 (m, C-Hortho in
ortho
BArF24), 133.0 (C-H, HAr
in IrSiPhH2), 130.8 (C-H, B[h]Q), 130.3
2
3
(B[h]Q), 128.8 (qq, JC−F = 32.8 Hz, JC−F = 2.9 Hz, CAr-CF3 in BArF24),
128.8 (C-H, HAr
in IrSiPhH2 [this 13C signal could not be directly
para
In situ Generation of [3][BArF24];route 2, (scheme 1)
detected as it overlapped with one quadruplet signal assigned to the
In a glovebox, H3SiPh (10.0 μL, 81.0 μmol) was added to a solution of
[1b][BArF24] (15 mg, 10.6 μmol) in d8-THF (0.55 mL). The resulting
solution was shaken and transferred into a J. Young NMR sample tube
which was subsequently tightly sealed for analysis. After 15 minutes of
reaction at room-temperature, 1H NMR analysis showed full consumption
BArF24 anion [13C-19F coupling]; its presence was inferred from 2D 1H,13
C
meta
HSQC/HMBC NMR), 126.8 (C-H, HAr
in IrSiPhH2), 128.3 (C-Si,
IrSiPhH2), 125.5 (q, 1JC−F = 272.2 Hz, CF3 in BArF24), 124.5 (C-H, B[h]Q),
1
124.1 (C-H, B[h]Q), 123 (C-H, B[h]Q), 123.7 (q, JC−F = 272.8 Hz, CF3 in
9
This article is protected by copyright. All rights reserved.

10.1002/chem.201804048
Chemistry - A European Journal
ARTICLE
of [1b][BArF24], but only partial conversion of [2c][BArF24] to [3][BArF24] (~
50%; cf. spectrum 1 in Figure S28). After 1 h, this conversion was
estimated to amount ~75% (cf. spectrum 2 in Figure S28). Full
conversion was obtained after 2 h of reaction; cf. spectrum 3 in Figure
consumption of [1b][BArF24] and selective formation of [4][BArF24] along
with H2 (singlet at ca. δ 4.55 ppm) in a respective ca. 1:0.3 ratio.
A
complex pattern of unresolved and unidentified peaks of low-intensity
were also observed in the range spanning δ 4.59-4.66 ppm (integrating
1
S28.
consumption of [2c][BArF24] and selective formation of [3][BArF24] along
with H2 (singlet at ca. δ 4.55 ppm) in a respective ca. 1:0.3 ratio.
Indeed, the latter 1H NMR spectrum indicated complete
roughly for 2H relative to [3][BArF24]). H (600 MHz, 233 K, d8-THF): δ =
9.01 (d, 1H, HAr B[h]Q, J = 5.4 Hz), 8.49 (d, 1H, HAr B[h]Q, J = 7.9 Hz),
7.98 (d, 1H, HAr B[h]Q, J = 8.7 Hz), 7.82 (d, 1H, HAr B[h]Q, J = 8.7 Hz),
A
complex pattern of peaks of low-intensity were also observed in the
range spanning δ 5.00-5.25 ppm (integrating roughly for 6H relative to
[3][BArF24]. Also detected are the two peaks resonating at δ 1.53 and
3.33 ppm, which are assigned to the product of polymerization of THF
inititated by the cationic [Cp*Ir(C,N)(PhHSi•thf]+ species.17 Data recorded
at 298 K for [3][BArF24]: 1H NMR (400 MHz, 298 K, CD2Cl2): δ = 9.01 (d,
1H, HAr B[h]Q, J = 5.3 Hz), 8.24 (d, 1H, HAr B[h]Q, J = 8.0 Hz), 8.02 (d,
1H, HAr B[h]Q, J = 7.0 Hz), 7.80 (m, 1H, HAr B[h]Q), 7.75 (m, 8H, Hortho
BArF24), 7.67-7.71 (m, 3H, HAr B[h]Q), 7.57 (m, 4H, Hpara BArF24), 7.52 (m,
7.86 (m, 8H, Hortho BArF24), 7.76 (d, 1H, HAr B[h]Q, J = 7.8 Hz), 7.66 (m,
1
4H, Hpara BArF24), 7.60-7.65 (m, 3H, HAr B[h]Q), 4.78 (t, 1H, Ha, JH-Si
=
1
95.4 Hz, JH-H = 2.8 Hz), 2.31 (bs, 2H, Si-thf), 2.07 (bs, 2H, Si-thf), 1.88
(bs, 15H, Cp-Me5), 1.34-1.44 [m, 4H, Si-thf (overlap with Hc,d peaks of
free silane)], 0.83 (bs, 2H, Hg or Hg’), 0.75 (broad m, 2H, Hd), 0.60-0.69
[broad m, 4H, (Hg or Hg’) + Hc], 0.53 (t, 3H, He, J = 8.7 Hz), -0.1 (m, 2H,
Hb). 13C (151 Hz, 233 K, d8-THF): δ = 163.0 (1:1:1:1 quartet, 1JC−B = 49.7
Hz, C‒B in BArF24), 158.2 (B[h]Q), 153.3 (broad, H-C=N, B[h]Q), 142.6
(B[h]Q), 137.2 (C-H, B[h]Q), 135.8 (B[h]Q), 135.6 (m, B[h]Q + C‒Hortho in
BArF24), 134.7 (B[h]Q), 131.0 (C-H, B[h]Q), 130.8 (C-H, B[h]Q), 130.1 (qq,
1H, HAr B[h]Q), 7.24-7.26 (bs, 5H, [SiHPh]), 5.73 (s, 1H, [SiHPh], 1JH-Si
=
3
99.3 Hz), 1.74 (s, 15H, Cp-Me5). Two broad peaks of THF were detected
at respectively δ 3.66 and 1.79 ppm, indicating fast-exchange of bulk
THF molecules with those coordinated to the Si centre of [3][BArF24] (cf.
Figure S18). 13C (126 Hz, 298 K, CD2Cl2): δ = 162.3 (1:1:1:1 quartet,
1JC−B = 49.9 Hz, C-B in BArF24), 158.1 (B[h]Q), 152.1 (broad, H-C=N,
B[h]Q), 141.9 (B[h]Q), 136.6 (C-H, B[h]Q), 135.2 (B[h]Q), 134.6 (broad,
C-H, B[h]Q), 135.4 (m, C-Hortho in BArF24), 130.88 (C-H, B[h]Q), 130.85
(C-H, B[h]Q), 130 (C-H, B[h]Q), 129.5 (qq, 2JC−F = 31.9 Hz, 3JC−F = 2.9 Hz,
CAr-CF3 in BArF24), 128.88 (B[h]Q), 128.82 (C-Si, IrSiPhH2), 128.80
(broad, C-H, IrSiPhH2), 128.09 (B[h]Q), 124.6 (C-H, B[h]Q), 122.4 (C-H,
2JC−F = 31.0 Hz, JC−F = 2.8 Hz, CAr-CF3 in BArF24), 128.6 (B[h]Q), 125.6
1
(q, JC−F = 272.6 Hz, CF3 in BArF24), 125.2 (C-H, B[h]Q), 123.1 (C-H,
B[h]Q), 121.3 (C-H, B[h]Q), 118.5 (m, C-Hpara in BArF24), 94.7 (Cp-Me5),
14.5 (broad m, Cb), 13.9 (s, Ce), 9.0 (Cp-Me5), [remaining 13C resonances
of IrSinBu(thf) could not be assigned with certainty as their corresponding
13C resonances, being weak in intensity, overlapped with peaks of
excess H3SinBu]. 29Si [obtained indirectly by 1H,29Si HMQC cross-peak
correlation with the 1H triplet at 4.78 ppm; 600/119 Hz, 233 K, d8-THF]: δ
= 71.3 ppm.
1
B[h]Q), 124.6 (q, JC−F = 272.4 Hz, CF3 in BArF24), 118.1 (m, C-Hpara in
Hydrodefluorination Catalysis ; General procedure
BArF24), 94.7 (Cp-Me5), 9.0 (Cp-Me5), [remaining 13C resonances of
IrSiPhH2 in the 1H-13C-HSQC spectrum could not be assigned with
certainty as their corresponding 1H resonances overlapped with some
peaks of the {C,N}-ligand and those of the BArF24 anion]. 29Si [obtained
In a dry Schlenk tube under argon, the fluorocarbon substrate (1 eq.) and
silane (1.2-3 eq.) were mixed together prior to the addition of a solution of
[1a,b][X] (0.5-5 mol %; X = BArF24, BPh4, BF4) in the given solvent (0.2-
0.5 mL).
Then this mixture was vigorously stirred at the given
1
1
indirectly by H,29 Si HMQC cross-peak correlation with the H singlet at
5.73 ppm; 500/99 MHz, 298 K, CD2Cl2)]: δ = 73.1 ppm. Data recorded at
213 K for [3][BArF24]: 1H NMR (600 MHz, 213 K, CD2Cl2): δ = 9.12 (bs,
1H, HAr B[h]Q), 8.23 (d, 1H, HAr B[h]Q, J = 8.0 Hz), 8.0 (bs, 1H, HAr
temperature for the given time. Prior to NMR analysis, 15 μL of
hexafluorobenzene and/or mesitylene (used as 19F and 1H NMR
reference standards, respectively) was added to the reaction mixture as
a solution in 0.1-0.2 mL of the NMR solvent (cf. SI for full details).
B[h]Q), 7.91 (bs, 1H, HAr B[h]Q), 7.76 (m, 8H, Hortho BArF24), 7.57 (m, 4H,
1
Hpara BArF24), 7.54 (m, 1H, HAr B[h]Q), 5.61 (bs, 1H, [SiHPh], JH-Si
=
100.4 Hz), 2.31 (bs, 2H, Si-THF), 2.07 (bs, 2H, Si-THF), 1.68 (bs, 15H,
Cp-Me5), 0.86 (bs, 2H, Si-THF), 0.67 (bs, 2H, Si-THF). The new 1H NMR
Acknowledgements
spectrum recorded at 213
K allowed the detection of peaks
characterizing the Si-coordinated THF. However, some peaks of the
{C,N}-ligand and those of the Ir-bound “PhSiH2” moiety could not be
The authors would like to thank the CNRS, the Agence
Nationale de la Recherche ANR project COCOORDCHEM, the
LABEX “Chimie des Systèmes Complexes” and the University of
Strasbourg for funding.
clearly detected in this new experiment (due to overlapping with peaks of
−
[BArF24
]
and of the Ph group belonging to excess HPhSiH2). Due to
chiral Si centre, [3][BArF24] is one of two possible diastereoisomers.
Thus, the typical peaks of Si-coordinated THF of the other
diastereoisomer (found to be the minor one) were also detected by the 1H
NMR experiment recorded at 213 K (as also confirmed by 1H,1H COSY
and NOESY NMR spectra; cf. spectra section). Selected data are as
follow: 1H NMR (600 MHz, 213 K, CD2Cl2): δ = 2.96 (bs, 2H, Si-THF),
2.72 (bs, 2H, Si-THF), 1.23 (bs, 2H, Si-THF), 1.06 ppm (bs, 2H, Si-THF).
Keywords: Iridium
• Si ligands • Silanes • Metallacycles • Reactive
intermediates
Notes
¶ Like suggested by one referee, a plausible proton shuttling by THF might
further contribute in lowering the activation barrier. At this stage this
scenario cannot be ruled out.
g,g'
f,f'
g,g'
thf
References
O
N
f,f'
Ir
*Cp
Si
[1] a) R. Waterman, P. G. Hayes, T. D. Tilley, Acc. Chem. Res. 2007, 40,
712-719; b) H. Ogino, Chem. Rec. 2002, 2, 291-306; c) P. D. Lickiss,
Chem. Soc. Rev. 1992, 21, 271-271.
H
a
b
c
d
[2] a) N. J. Hill, R. West, J. Organomet. Chem. 2004, 689, 4165-4183; b) S.
Inoue, Wiley-VCH Verlag, Weinheim, Germany, 2014, pp. 243-273; c) P.
Jutzi, A. Möhrke, Angew. Chem. Int. Ed. Eng. 1990, 29, 893-894.
[3] a) H. Handwerker, C. Leis, R. Probst, P. Bissinger, A. Grohmann, P.
Kiprof, E. Herdtweck, J. Blümel, N. Auner, C. Zybill, Organometallics 1993,
12, 2162-2176; b) C. Leis, D. L. Wilkinson, H. Handwerker, C. Zybill, G.
Müller, Organometallics 1992, 11, 514-529; c) D. A. Straus, T. D. Tilley, A.
L. Rheingold, S. J. Geib, J. Am. Chem. Soc. 1987, 109, 5872-5873; d) C.
Zybill, G. Müller, Angew. Chem. Int. Ed. 1987, 26, 669-670; e) K. Ueno,
M. Sakai, H. Ogino, Organometallics 1998, 17, 2138-2140; f) J. Y. Corey,
Chem. Rev. 2016, 116, 11291-11435; g) M. Okazaki, E. Suzuki, N.
Miyajima, H. Tobita, H. Ogino, Organometallics 2003, 22, 4633-4635; h)
K. Ueno, A. Masuko, H. Ogino, Organometallics 1999, 18, 2694-2699; i)
L. K. Woo, D. A. Smith, V. G. Young, Organometallics 1991, 10, 3977-
3982.
e
, BArF24
[4][BArF24
]
In-situ Generation of [4][BArF24]; route 1 (scheme 1)
Compound [4][BArF24] was generated in solution using the procedure 2
previously described for [3][BArF24] (cf. previous section): H3SinBu (20.0
μL, 77.0 μmol) was added to a solution of [1b][BArF24] (30 mg, 21.2
μmol) in d8-THF (0.45 mL). The resulting solution was shaken and
transferred into a J. Young NMR sample tube which was subsequently
tightly sealed for analysis at -40 °C. Full conversion was obtained after 2
h of reaction, as verified by 1H NMR analysis (cf. spectra in the next sub-
1
10
section). Indeed, the latter H NMR spectrum indicated complete
This article is protected by copyright. All rights reserved.

10.1002/chem.201804048
Chemistry - A European Journal
ARTICLE
[4] a) E. Calimano, T. D. Tilley, J. Am. Chem. Soc. 2008, 130, 9226-9227;
b) R. J. P. Corriu, B. P. S. Chauhan, G. F. Lanneau, Organometallics
1995, 14, 1646-1656; c) R. J. P. Corriu, G. F. Lanneau, C. Priou, B. P. S.
Chauhan, N. Auner, H. Handwerker, E. Herdtweck, Organometallics 1995,
14, 1657-1666; d) P. B. Glaser, T. D. Tilley, J. Am. Chem. Soc. 2003, 125,
13640-13641; e) S. R. Klei, T. D. Tilley, R. G. Bergman, Organometallics
2002, 21, 4648-4661; f) M. C. Lipke, A. L. Liberman-Martin, T. D. Tilley,
Angew. Chem. Int. Ed. 2017, 56, 2260-2294; g) M. C. Lipke, T. D. Tilley,
J. Am. Chem. Soc. 2013, 135, 10298-10301; h) M. Okazaki, H. Tobita, H.
Ogino, Dalton Trans. 2003, 493-506; i) D. A. Straus, C. Zhang, G. E.
Quimbita, S. D. Grumbine, R. H. Heyn, T. D. Tilley, A. L. Rheingold, S. J.
Geib, J. Am. Chem. Soc. 1990, 112, 2673-2681; j) T. Watanabe, H.
Hashimoto, H. Tobita, J. Am. Chem. Soc. 2006, 128, 2176-2177; k) H.
Handwerker, M. Paul, J. Blümel, C. Zybill, Angew. Chem. Int. Ed. 1993,
32, 1313-1315; l) H. Sakaba, H. Oike, Y. Arai, E. Kwon, Organometallics
2012, 31, 8172-8177.
[32] A. Marquez, J. F. Sanz, J. Am. Chem. Soc. 1992, 114, 2903-2909.
[33] a) H. Jacobsen, T. Ziegler, Organometallics 1995, 14, 224-230; b) H.
Jacobsen, T. Ziegler, Inorg. Chem. 1996, 35, 775-783.
[34] E. D. Glendening, J. K. Badenhoop, A. E. Reed, J. E. Carpenter, J. A.
Bohmann, C. M. Morales, C. R. Landis, F. Weinhold, Theoretical
Chemistry Institute, University of Wisconsin, Madison, 2013.
[35] D. O’Hagan, Chem. Soc. Rev. 2008, 37, 308-319.
[36] a) M. K. Whittlesey, E. Peris, ACS Catal. 2014, , 4, 3152-3159; b) T.
Stahl, H. F. T. Klare, M. Oestreich, ACS Catal. 2013, 21, 1578-1587; c) G.
Meier, T. Braun, Angew. Chem. Int. Ed. 2009, 48, 1546-1548; d) J. L.
Kiplinger, T. G. Richmond, C. E. Osterberg, Chem. Rev. 1994, 94, 373-
431.
[37] C. Douvris, O. V. Ozerov, Science 2008, 321, 1188-1190.
[38] B. M. Kraft, R. J. Lachicotte, W. D. Jones, J. Am. Chem. Soc. 2000, 122
8559-8560.
[39] J. Burdeniuc, R. H. Crabtree, J. Am. Chem. Soc. 1996, 118, 2525-2526.
[40] J. L. Kiplinger, T. G. Richmond, J. Am. Chem. Soc. 1996, 118 1805-1806.
[41] J. Li, T. Zheng, H. Sun, X. Li, Dalton Trans. 2013, 42, 13048-13053.
[42] T. Stahl, H. F. T. Klare, M. Oestreich, J. Am. Chem. Soc. 2013, 135,
1248-1251.
[5] a) M. C. Lipke, T. D. Tilley, J. Am. Chem. Soc. 2014, 136, 16387-16398;
b) N. Schneider, M. Finger, C. Haferkemper, S. Bellemin-Laponnaz, P.
Hofmann, L. H. Gade, Angew. Chem. Int. Ed. 2009, 48, 1609-1613; c) E.
Calimano, T. D. Tilley, Organometallics 2010, 29, 1680-1692.
[6] E. Calimano, T. D. Tilley, J. Am. Chem. Soc. 2009, 131, 11161-11173.
[7] P. G. Hayes, C. Beddie, M. B. Hall, R. Waterman, T. D. Tilley, J. Am.
Chem. Soc. 2006, 128, 428-429.
[43] S. Sabater, J. Mata, E. Peris, Nat. Commun. 2013, 4, 2553.
[44]R. J. Lindup, T. B. Marder, R. N. Perutz, A. C. Whitwood, Chem. Commun.
2007, 95 3664.
[8] P. Gigler, B. Bechlars, W. A. Herrmann, F. E. Kühn, J. Am. Chem. Soc.
2011, 133, 1589-1596.
[9] a) P. Jiang, D. Trieber, P. P. Gaspar, Organometallics 2003, 22, 2233-
2239; b) T. G. Driver, K. A. Woerpel, J. Am. Chem. Soc. 2004, 126,
9993-10002.
[45] J. Vela, J. M. Smith, Y. Yu, N. A. Ketterer, J. Flaschenriem, R. J.
Lachicotte, P. L. Holland, C. J. Flaschenriem, J. Am. Chem. Soc. 2005,
127, 7857−7870.
[46] H. Lv, Y.-B. Cai, J.-L. Zhang, Angew. Chem.- Int. Ed. 2013, 52, 3203-
3207.
[10] a) C. Z. Rotsides, K. A. Woerpel, Dalton Trans. 2017, 46, 8763-8768; b)
Z. Nevárez, K. A. Woerpel, Org. Lett. 2007, 9, 3773-3776; c) T. G. Driver,
in Silver in Organic Chemistry (Ed.: M. Harmata), John Wiley & sons,
2010.
[11] S. R. Klei, T. D. Tilley, R. G. Bergman, Organometallics 2002, 21, 3376-
3387.
[47] J. Yang, M. Brookhart, J. Am. Chem. Soc. 2007, 129, 12656-12657.
[48] S. M. Senaweera, A. Singh, J. D. Weaver, J. Am. Chem. Soc. 2014, 136
3002-3005.
[49] a) A. Matsunami, S. Kuwata, Y. Kayaki, J. Synth. Org. Chem. Jpn 2018,
76, 315-324; b) A. Matsunami, Y. Kayaki, S. Kuwata, T. Ikariya,
Organometallics 2018, 37 1958-1969; c) A. Matsunami, S. Kuwata, Y.
Kayaki, ACS Catal. 2016, 6, 5181-5185
[12]D. G. Gusev, F. G. Fontaine, A. J. Lough, D. Zargarian, Angew. Chem. Int.
Ed. 2003, 42, 216-219.
[13] P. W. Smith, T. D. Tilley, J. Am. Chem. Soc. 2018, 140, 3880-3883.
[14] a) M. A. Rankin, D. F. MacLean, G. Schatte, R. McDonald, M. Stradiotto,
J. Am. Chem. Soc. 2007, 129, 15855-15864; b) B. V. Mork, T. D. Tilley, J.
Am. Chem. Soc. 2004, 126, 4375-4385.
[15] a) J. D. Feldman, J. C. Peters, T. D. Tilley, Organometallics 2002, 21,
4065-4075; b) M. E. Fasulo, P. B. Glaser, T. D. Tilley, Organometallics
2011, 30, 5524-5531.
[16] a) R. J. P. Corriu, G. F. Lanneau, B. P. S. Chauhan, Organometallics
1993, 12, 2001-2003; b) J. S. Price, D. J. H. Emslie, J. F. Britten, Angew.
Chem. Int. Ed. 2017, 56, 6223-6227; c) K. Dannappel, R. Nienhaus, M.
Schürmann, B. Costisella, K. Jurkschat, Z. Anorg. Allg. Chem. 2009, 635,
2126-2134.
[17] M. Hamdaoui, M. Ney, V. Sarda, L. Karmazin, C. Bailly, N. Sieffert, S.
Dohm, A. Hansen, S. Grimme, J.-P. Djukic, Organometallics 2016, 35,
2207-2223.
[50] a) C. Douvris, C. M. Nagaraja, C.-H. Chen, B. M. Foxman, O. V. Ozerov,
J. Am. Chem. Soc. 2010, 132, 4946-4953; b) V. J. Scott, R. Çelenligil-
Çetin, O. V. Ozerov, J. Am. Chem. Soc. 2005, 127, 2852-2853.
[51] C. Michon, K. MacIntyre, Y. Corre, F. Agbossou-Niedercorn,
ChemCatChem 2016, 8, 1755-1762.
[52] Y. Hu, J. R. Norton, J. Am. Chem. Soc. 2014, 136, 5938-5948.
[53] M86-E01078 APEX2 User Manual, Bruker AXS Inc., Madison, Wisconsin,
USA, 2006.
[54] G. M. Sheldrick, Acta Cryst. 1990, A46, 467-473.
[55] G. M. Sheldrick, Acta Cryst. 2008, A64, 112-122.
[56] G. te Velde, F. M. Bickelhaupt, E. J. Baerends, C. Fonseca Guerra, S. J.
A. van Gisbergen, J. G. Snijders, T. Ziegler, J. Comp. Chem. 2001, 22,
931-967.
[57] J. P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 1996, 77, 3865-
3868.
[18] M. Hamdaoui, C. Desrousseaux, H. Habbita, J.-P. Djukic,
Organometallics 2017, 36, 4864-4882.
[58] a) S. Grimme, J. Antony, S. Ehrlich, H. Krieg, J. Chem. Phys. 2010, 132,
154104; b) S. Grimme, S. Ehrlich, L. Goerigk, J. Comp. Chem. 2011, 32,
1456-1465.
[59] a) E. van Lenthe, E. J. Baerends, J. G. Snijders, J. Chem. Phys. 1993,
99, 4597-4610; b) E. van Lenthe, E. J. Baerends, J. G. Snijders, J. Chem.
Phys. 1994, 101, 9783-9792; c) E. van Lenthe, A. Ehlers, E.-J. Baerends,
J. Chem. Phys. 1999, 110, 8943-8953.
[19] a) J. Y. Corey, J. Braddock-Wilking, Chem. Rev. 1999, 99, 175-292; b) J.
Y. Corey, Chem. Rev. 2011, 111, 863-1071; c) J. Yang, P. S. White, C. K.
Schauer, M. Brookhart, Angew. Chem. Int. Ed. 2008, 2008, 4141-4143;
d) R. N. Perutz, S. Sabo-Etienne, Angew. Chem. Int. Ed. 2007, 46, 2578 -
2592.
[20] J. C. Green, M. L. H. Green, G. Parkin, Chem. Commun. 2012, 48,
11481-11503.
[60] F. Weinhold, in Encyclopedia of Computational Chemistry, Vol. 3 (Eds.: P.
vonRagué-Schleyer, N. L. Allinger, T. Clark, J. Gasteiger, P. A. Kollman,
H. F. Schaefer, P. R. Schreiner), John Wiley & Sons, Chichester, UK,
1998, pp. 1792-1811.
[61] O. V. Gritsenko, P. R. T. Schipper, E. J. Baerends, Chem. Phys. Lett.
1999, 302, 199-207.
[62] M. Ernzerhof, G. E. Scuseria, J. Chem. Phys. 1999, 110, 5029-5036.
[63] S. N. Steinmann, C. Corminboeuf, J. Chem. Theor. Comp. 2011, 7, 3567-
3577.
[21] a) G. I. Nikonov, Adv. Organomet. Chem. 2005, 53, 217-309; b) W.
Scherer, P. Meixner, J. E. Barquera-Lozada, C. Hauf, A. Obenhuber, A.
Brück, D. J. Wolstenholme, K. Ruhland, D. Leusser, D. Stalke, Angew.
Chem. Int. Ed. 2013, 52, 6092-6096; c) C. Hauf, J. E. Barquera-Lozada,
P. Meixner, G. Eickerling, S. Altmannshofer, D. Stalke, T. Zell, D. Schmidt,
U. Radius, W. Scherer, Z. Anorg. Allg. Chem. 2013, 639, 1996-2004.
[22] D. Ho Binh, M. Milovanovic, J. Puertes-Mico, M. Hamdaoui, S. D. Zaric,
J.-P. Djukic, Chem. Eur. J. 2017, 23, 17058–17069
[23] J. Autschbach, T. Ziegler, J. Chem. Phys. 2000, 113, 936-947.
[24] J. Autschbach, T. Ziegler, J. Chem. Phys. 2000, 113, 9410-9418.
[25] K. B. Wiberg, Tetrahedron 1968, 24, 1083-1096.
[26] a) M. P. Mitoraj, A. Michalak, T. Ziegler, J. Chem. Theory Comput. 2009,
5, 962-975; b) M. P. Mitoraj, A. Michalak, T. Ziegler, Organometallics
2009, 28, 3727-3733.
[27] J. C. Peters, J. D. Feldman, T. D. Tilley, J. Am. Chem. Soc. 1999, 121,
9871-9872.
[28] a) H. Sakaba, T. Hirata, C. Kabuto, K. Kabuto, J. Organomet. Chem.
2007, 692, 402-407; b) H. Sakaba, T. Hirata, C. Kabuto, K. Kabuto,
Organometallics 2006, 25, 5145-5150.
[29] H. Nakatsuji, J. Ushio, T. Yonezawa, J. Organomet. Chem. 1983, 258,
C1-C4.
[30] a) I. A. Abronin, N. A. Avdyuhina, L. V. Morozova, G. K.-I. Magomedov, J.
Mol. Struct. (Theochem) 1991, 228, 19-25; b) S. D. Grumbine, T. D. Tilley,
F. P. Arnold, A. L. Rheingold, J. Am. Chem. Soc. 1993, 115, 7884-7885;
c) T. R. Cundari, M. S. Gordon, J. Phys. Chem. 1992, 96, 631-636; d) T.
R. Cundari, M. S. Gordon, Organometallics 1992, 11, 3122-3129.
[31] a) F. P. Arnold, Organometallics 1999, 18, 4800-4809; b) S.
Grumbine, T. D. Tilley, J. Am. Chem. Soc. 1994, 116, 5495-5496.
K. 11
This article is protected by copyright. All rights reserved.

10.1002/chem.201804048
Chemistry - A European Journal
ARTICLE
Entry for the Table of Contents
ARTICLE
Treatment of a Cp*-bound
iridacycle with phenylsilane in the
presence of THF yields a
Dang Ho Binh, Mustapha Hamdaoui,
Diane Fischer-Krauser, Lydia Karmazin,
Corinne Bailly and Jean-Pierre Djukic*
reactive cationic Ir(III)-silylene
complex displaying catalytic
properties in hydro-de-
Page No. – Page No.
Entrapment of THF-stabilized
Iridacyclic Ir(III)-“Silylenes” from
Double H-Si Bond Activation and H2
Elimination
fluorination of fluoroalkanes.
12
This article is protected by copyright. All rights reserved.