Reversible Silylium Transfer between P-H and Si-H Donors

Article abstract of DOI:10.1002/anie.202011372

The Mo=PR₂ π* orbital in a Mo phosphenium complex acts as acceptor in a new P^III-based Lewis superacid. This Lewis acid (LA) participates in electrophilic Si-H abstraction from E₃SiH to give a Mo-bound secondary phosphine ligand, Mo-PR₂H. The resulting Et₃Si⁺ ion remains associated with the Mo complex, stabilized by η¹-P-H donation, yet undergoes rapid exchange with an η¹-Si-H adduct of free silane in solution. The equilibrium between these two adducts presents an opportunity to assess the role of this new LA in catalytic reactions of silanes: is the LA acting as a catalyst or as an initiator? Preliminary results suggest that a cycle including the Mo-bound phosphine-silylium adduct dominates in the catalytic hydrosilylation of acetophenone, relative to a putative cycle involving the silane-silylium adduct or “free” silylium.

Full text of DOI:10.1002/anie.202011372

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Reversible silylium transfer between P-H and Si-H donors
Authors: Roman G. Belli, Dimitrios A. Pantazis, Robert McDonald, and
Lisa Rosenberg
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: Angew. Chem. Int. Ed. 10.1002/anie.202011372
Link to VoR: https://doi.org/10.1002/anie.202011372

10.1002/anie.202011372
Angewandte Chemie International Edition
RESEARCH ARTICLE
Reversible silylium transfer between P-H and Si-H donors
Roman G. Belli,[a] Dimitrios A. Pantazis,[b] Robert McDonald,[c] and Lisa Rosenberg*[a]
[a]
R. G. Belli, L. Rosenberg
Department of Chemistry, University of Victoria, P.O. Box 1700 STN CSC, Victoria, British Columbia, V8W 2Y2, Canada
E-mail: lisarose@uvic.ca
[b]
[c]
D. A. Pantazis
Max-Planck-Institut für Kohlenforschung, Kaiser-Wilhelm-Platz 1, 45470 Mülheim an der Ruhr, Germany
R. McDonald
X-ray Crystallography Laboratory, Department of Chemistry, University of Alberta, Edmonton, Alberta T6G 2G2, Canada
Supporting information for this article is given via a link at the end of the document.
Abstract: The Mo=PR2 π* orbital in a Mo phosphenium complex acts
as acceptor in a new P(III)-based Lewis superacid. This Lewis acid
(LA) participates in electrophilic Si-H abstraction from E3SiH to give a
Mo-bound secondary phosphine ligand, Mo-PR2H. The resulting
Et3Si+ ion remains associated with the Mo complex, stabilized by h1-
P-H donation, yet undergoes rapid exchange with an h1-Si-H adduct
of free silane in solution. The equilibrium between these two adducts
presents an opportunity to assess the role of this new LA in catalytic
reactions of silanes: is the LA acting as a catalyst or as an initiator?
Preliminary results suggest that a cycle including the Mo-bound
phosphine-silylium adduct dominates in the catalytic hydrosilylation of
acetophenone, relative to a putative cycle involving the silane-silylium
adduct or “free” silylium.
LA
+
SiR3
−
+
Catalysis
Initiation
!
!
LAH−
+
“SiR3
+
LA
H
SiR3
”
H
Scheme 1. Two roles of Lewis acids in catalysis involving Si-H activation.
The Lewis acidity of free phosphenium ions is well
established,[7i,14,15] and recently examples of electrophilic Si-H
abstraction by phosphenium ions have been reported.[2b,2f,7g,16]
However, despite evidence for the electrophilicity of metal-
coordinated phosphenium ligands,[17,18] their Lewis acidity has not
been explicitly assessed, and their interactions with silanes
remain unexplored. We describe here a simple, cationic metal
phosphenium complex that acts as a remarkably strong P(III) LA.
Its reaction with Et3SiH leads to a P-H bond-stabilized silylium
adduct that reversibly transfers the Et3Si+ ion to the Si-H bond of
free silane in solution, an unusual equilibrium that represents the
critical bifurcation between catalyst and initiator in LA-mediated
reactions of silanes.
Introduction
The electrophilic activation of Si-H bonds by Lewis acids (LAs) is
increasingly exploited as a fundamental step in a range of
catalytic reactions of silanes,[1,2] a development that has been
complemented by an explosion of interest in the discovery of new
main group LAs.[3] This exciting class of compounds now extends
well beyond neutral group 13 reagents[4] to include low-coordinate,
cationic boron and aluminum species[5] and diverse new
structures based on elements from groups 14,[2c,2h,6] 15,[2e,7] and
16.[8] While the LAs themselves are of fundamental interest and
importance for the variety of electronic structures that permit
acceptor behaviour,[9] the specific role of silane-LA interactions in
catalysis has also come under increasing scrutiny (Scheme 1).
Does the LA remain associated with reactive intermediates during
catalysis, as has been demonstrated for the well-studied,
electrophilic borane B(C6F5)3?[10] This now-archetypal LA catalyst
maintains a close, polarizing interaction with the Si-H bond that
induces silylium character in the substrate silane and allows rapid
hydride transfer to the cationic intermediates resulting from
nucleophilic attack at R3Si+. Or, does the LA simply initiate the
catalysis through complete hydride abstraction and release of
"free" silylium (generally either solvent- or silane-stabilized),[11]
which can itself act as a catalyst?[12] This latter behavior is typified
by the addition of trityl cation, Ph3C+, to silanes.[2a,10a,13]
X
X
PR2H
CO
[Ph3C][X]
– Ph3CH
OC
OC
CO
OC
OC
CO
PR2H
cis-Mo(CO)4(PR2H)2
Mo
Mo
+
CO
1
P
P
X = [B(m-C6H3Cl2)4]–
R = Ph (a), Tolp (b)
R
R
R
R
trans-2
cis-2
95:5
Scheme 2. Synthesis of phosphenium complex cations.
Results and Discussion
Dark green complex cations 2a-b were prepared via hydride
abstraction from bis(secondary phosphine) complexes 1a-b
(Scheme 2). Cis- and trans-2a-b show diagnostic downfield
31P{1H} chemical shifts for signals due to the new phosphenium
ligand (d440-441 ppm). The observed preference for trans versus
cis geometry in 2a-b is consistent with the established strong π-
acidity of phosphenium ligands,[15,17b,18a,19,20] since
a trans
structure ensures that PR2+ does not compete with CO ligands for
π-backbonding at a Mo-based d-orbital.[21] X-ray diffraction
analysis of crystals of 2b confirmed the solid state structure of the
trans isomer (Figure 1), with bond angles at the phosphenium
(P2) consistent with planarity at this sp2-hybridized centre, and a
1
This article is protected by copyright. All rights reserved.

10.1002/anie.202011372
Angewandte Chemie International Edition
RESEARCH ARTICLE
short Mo-P2 distance of 2.29 Å, relative to the Mo-phosphine (P1)
distance of 2.52 Å. These distances are similar to or slightly longer
than those previously reported for other Mo phosphenium and
phosphine complexes.[22] Scalar relativistic DFT calculations with
the BP86 functional[23] combined with D4 dispersion corrections[24]
(ZORA-BP-86_D4) and all-electron SARC-ZORA-TZVP basis
sets[25] reproduce the bond shortening for Mo-P2 relative to Mo-
P1 in trans-2b, and calculated bond order metrics clearly indicate
double bond character in the Mo-phosphenium bonds in the cis
and trans isomers of 2a-b, as is generally observed for metal-
coordinated phosphenium ligands.[15,18,20,26] As expected based
on π-backbonding considerations, this is slightly reduced in the
cis relative to the trans isomers. NBO analysis of the Mo-
phosphenium interaction in trans-2b indicates one s-bond
(polarized 64% toward P) and one π-bond (polarized 66% toward
Mo) (Figure 2a,b). The π* LUMO (Figure 2c) is delocalized but
shows significant contribution from a p-type lobe at P, the likely
site of potential donor-acceptor interactions. This sets these
diarylphosphenium complexes apart from the more commonly
studied bis(amido)-substituted phosphenium complexes (and
most other P-based LAs) for which calculations show LUMOs of
s*-character.[9,15a,17e,18a]
another indicator of the Lewis acidity of 2.[28] We assessed
complexes 2a-b by three complementary LA scaling methods.
Gutmann-Beckett experiments[29] performed in CD2Cl2 show very
large changes in d31P values for O=PEt3 adducts of 2 relative to
the free phosphine oxide, with Dd31P ranging from +44.7 ppm for
trans-2b to +48.7 ppm for cis-2a. The cis isomers are apparently
stronger LAs than their respective trans isomers, which is
consistent with diminished phosphenium π-backbonding from Mo
in the cis complexes (vide supra). These chemical shift changes
are larger than that reported for the very strong cationic, P-based
LA [P(C6F5)3F]+ (+40.4 ppm).[9d] Calculated Global Electrophilicity
Indices (GEI)[30] for the isomers of 2a-b are high, ranging from
4.54 eV to 5.19 eV. We used a long-range corrected DFT method
(LC-BLYP)[31] to address challenges in using DFT to define EHOMO
and ELUMO in this calculation, so we also (re)calculated GEIs
recently reported for some cationic main group LAs[32] for
comparison (see Supporting Information). The values for 2 are
+
lower than those for trityl cation (5.81 eV) and free PPh2 (6.98
eV) and straddle that for [P(C6F5)3F]+ (5.00 eV). Likewise, we
calculated fluoride ion affinity (FIA) values at the ZORA-BP86-D4
level for some established LAs as well as the isomers of 2a-b.
FIAs for 2a-b of 681-719 kJ/mol place the P(III) centres in these
complexes among established “Lewis superacids”,[33] since they
are higher than the benchmark value for SbF5 (473 kJ/mol),
although by this metric all four isomers of 2a-b are weaker LAs
than [P(C6F5)3F]+ (743 kJ/mol). The P(III)-based Lewis acidity of
free diaryl phosphenium PPh2+ (793 kJ/mol) is notably reduced on
coordination to Mo, but 2a-b are stronger LAs than the free N-
heterocyclic
phosphenium
derived
from
bis(N-
methyl)diazaphospholene (659 kJ/mol). The unexpectedly strong
acceptor nature of 2 probably derives from the inductive effect of
the Mo(CO)4 fragment in these complexes. This points to the
intriguing potential for tuning Lewis acidity through modification of
the coordination environment at the "ancillary metal" in
phosphenium complexes.
Figure 1. Structure of [trans–Mo(CO)4{PTolp2H}{PTolp2}]+ (trans-2b), showing
the atom labelling scheme. Non-hydrogen atoms are represented by Gaussian
ellipsoids at the 30% probability level. The hydrogen atom attached to P1 is
shown with an arbitrarily small thermal parameter; all other hydrogens are not
shown. Selected interatomic distances (Å) and bond angles (°): Mo-P1
2.5216(14), Mo-P2 2.2878(13), P1-H1P 1.30(7), P1-Mo-P2 177.96(5), Mo-P2-
C31 125.87(16), Mo-P2-C41 125.86(18), C31-P2-C41 108.3(2).
[Mo(CO)4(PR2H)(PR2)]+
gel
2
[NBu4][PF6]
THF
< 1 h
py
+
+
PR2H
PR2H
PR2H
CO
Mo
OC
OC
CO
OC
OC
CO
OC
OC
Mo
Mo
CO
PR2F
CO
PR2
O
CO
a)
b)
c)
R2P
N
+ [NBu4]+, PF5
Scheme 3. Reactions of phosphenium complex 2 in CD2Cl2.
The dark green colour of 2a-b in CD2Cl2 changes to yellow
with the addition of Et3SiH, and the sharp 31P{1H} doublets due to
2 are replaced by broad, overlapping resonances stretching from
+15 to –15 ppm (Figure 3a bottom), consistent with one or more
Figure 2. Mo-PR2 (a) s- and (b) π-bonding orbitals for 2b from NBO analysis,
and LUMO (π*) from the canonical DFT orbitals
1
dynamic processes occurring on the NMR time scale. H NMR
signals for the phosphine P-H and silane Si-H are also broad and
shifted relative to those for 2a-b and free Et3SiH, respectively
(Figure 3b bottom). Electrospray ionization mass spectrometry
(ESI-MS) of the reaction mixtures shows the presence of cations
with m/z and isotope patterns matching the formula [2•Et3SiH]+,
which we propose are the h1-Si-H phosphenium adducts 3a-b
The NMR-scale reactions shown in Scheme 3 demonstrate
the strong acceptor properties of the phosphenium ligand in 2. It
readily abstracts F– from the PF6– anion, and forms P-N and P-O
adducts, respectively, with pyridine and THF.[27] In the latter case,
the solution NMR sample becomes a gel within one hour,
suggesting ring-opening polymerization of the THF has occurred,
2
This article is protected by copyright. All rights reserved.

10.1002/anie.202011372
Angewandte Chemie International Edition
RESEARCH ARTICLE
(Scheme 4). Although a number of LAs that participate in
electrophilic Si-H activation show facile, reversible silane
binding,[34] MS/MS experiments on 3a-b show only the sequential
loss of two CO ligands before decomposition to unidentified
fragments at high collision voltages, which indicates unusually
high stability of the putative P-H-Si interaction in these complexes,
with respect to loss of free silane. This suggests that silane
dissociation is not important in the RT exchange process
observed by NMR. Accordingly, a DFT estimate of the energy of
association of silane with 2 provides a lower limit of ~19 kcal mol–
1 for the barrier to silane dissociation from 3. The very low rate of
this reverse reaction is also supported by the addition of two equiv
of Et3SiD to 2. This gives an equimolar mixture of two d1-
isotopomers of 3, as determined by 31P{1H} COSY NMR for the
mixture at 183 K (vide infra). These differ in location of the
deuterium label (bridging P-D-Si versus terminal P-D), which
points to an exchange of the Et3Si+ group between P-D/H bonds
in d1-3 (vide infra). Critically, we have evidence for only trace
amounts of d2-3 in this reaction mixture; its formation requires
dissociation of Et3SiH from the P-H-Si isotopomer of d1-3 to give
d1-2, which could then react with an equiv of Et3SiD (see
Supporting Information for details).
intensity (2JPP ~23 Hz) that can be assigned to inequivalent
phosphorus centres in cis-h1-Si-H adducts 3a-b, and a broad
singlet that corresponds to the bis(phosphine) precursor complex
cis-1 (Figure 3a top). Three distinct P-H 1H signals (Figure 3b top)
correlate with these 31P signals in the1H/31P HSQC spectra for the
reactions of 2 with silane. Low temperature 1H/29Si HMQC for
these mixtures show correlations between Si-Et 1H signals (0.81,
0.91 ppm) and a 29Si signal at ~39 ppm, which we have assigned
to the SiEt3 group in 3a-b. However, neither of the two distinct P-
H signals assigned to 3a-b shows 1H/29Si HMQC correlation with
this 29Si signal, 1H COSY correlations with these Si-Et signals, or
29Si satellites. This loss of scalar coupling information is
consistent with the rapid exchange of Et3Si+ between the two P-H
bonds in 3, as suggested by the deuterium labelling experiment
described above. It also prevents assignment of the distinct
1
31P{1H} signals for 3.[35] JPH values for these signals are very
similar to each other (345-351 Hz) and are intermediate to those
observed for the secondary phosphine ligands in 1 (331 Hz) and
2 (360 Hz), suggesting that both retain significant P-H bond
character. Collectively, the NMR and MS data suggest that
complex 3 is best considered as h1-P-H adduct of the silylium ion
Et3Si+, as opposed to an h1-Si-H adduct of the phosphenium
ligand in 2.
Well characterized examples of such h1-E-H stabilized
silylium ions are rare, despite the increasing interest in
electrophilic Si-H bond activation (vide supra), for which such
structures represent important intermediates. Silane adducts
CO
+
+ Et3SiH
OC
OC
CO
[Mo(CO)4(PR2H)(PR2)]+
Mo
R2P
PR2H
– Et3SiH
2
R = Ph (a), Tolp (b)
H
SiEt3
comparable to
3 have been reported for a fluorinated
cis-3
+ Et3SiH
boraindene,[34b] a cationic iridium fragment,[36] and a series of
silylium ions.[37] Phosphine adducts of silylium ions containing
direct P-Si donor-acceptor interactions are known,[7g,38] but we
have found no previous examples of silylium ions stabilized by h1-
P-H interactions, although a similar structure has been proposed
as an intermediate in oxidative insertion of the phosphenium-like
cation [Cp*PH]+ into a Et3Si-H bond.[7g] DFT-calculated structures
of 3a-b show approximately linear P-H-Si linkages in these
complexes, with P-H and Si-H bond lengthening consistent with
delocalized bridge bonding, at least in these gas-phase models
(see Supporting Information).
– Et3SiH
CO
OC
Mo
OC
CO
+
+
SiEt3
Et3Si
H
PR2H
4
PR2H
1
Scheme 4. Reaction of phosphenium complex 2 with Et3SiH in CD2Cl2.
(a)
(b)
3b
CDHCl2
Complexes 1 and 3 are in equilibrium, based on VT-NMR and
low temperature 1H EXSY NMR experiments (vide infra).
Conversion of 3 to 1 involves loss of the highly reactive Et3Si+
ion,[39] which occurs sufficiently easily that attempts to crystallize
adducts 3a-b inevitably result in their decomposition to 1a-b.
Heating samples of the equilibrium mixture in CD2Cl2/C7D8 to 338
K also causes irreversible conversion to 1. The surprisingly robust
equilibrium between 3 and 1 in solution at ambient and low
temperature appears to involve the rapid exchange of Et3Si+
between 3 and the cationic silane adduct [Et3Si-H-SiEt3]+, 4
(Scheme 4), which shows a diagnostic 29Si signal at 59.3 ppm[37a]
in low temperature 29Si DEPT and 1H/29Si HMQC experiments
(see Supporting Information). We can prepare a comparable
equilibrium mixture (as determined by RT NMR) by the direct
addition of 4 (generated in situ from Ph3C+ and two equiv Et3SiH)
to the bis(phosphine) complex 1. Low temperature 1H EXSY
experiments with relatively long mixing times (≥ 50 ms) show
correlations between signals for all three P-H groups in
complexes 1 and 3, but shorter mixing times (5 to 12 ms) give
correlations only between 1 and each of the two P-H signals for 3,
not between the two P-H signals for 3. This indicates that the
mutual exchange of P-H sites within complex 3 results from
1b
3b
3b
1b
3b
3b
1b
*
*
*
183 K
*
218 K
258 K
HP
298 K
ppm
6.5
6.0
5.5 ppm
10
0
-10
Figure 3. Variable temperature (a) 31P{1H} and (b) 1H NMR spectra for a mixture
of 2b and two equiv Et3SiH in CD2Cl2. "*" marks unidentified products that do
not participate in exchange processes (see Supporting Information for details).
The broad RT NMR signals decoalesce when mixtures of 2a-
b and 2 equiv of Et3SiH in CD2Cl2 are cooled to 183 K (Figure 3).
The low temperature 31P{1H} spectra show two doublets of equal
3
This article is protected by copyright. All rights reserved.

10.1002/anie.202011372
Angewandte Chemie International Edition
RESEARCH ARTICLE
intermolecular transfer of Et3Si+ between the silane adduct 4 and
either of the two P-H donors in 3, as opposed to intramolecular
"hopping" of the Et3Si+ between the two P-H bonds in 3.
Lineshape analysis of the VT-31P{1H} NMR spectra gave first
resulting from nucleophilic attack of ketone at the silylium ion in 3.
The B(C6F5)3-catalyzed reaction of benzophenone gives a small
amount of over-reduction (Table 1), probably due to slightly less
steric congestion at the C-O bond in the silyl ether product.
Complex 2 gives more over-reduction of this substrate under
identical conditions, but the product distribution is still not
consistent with simple initiation as for the trityl cation.
order rate constants for the conversion of 3 to 1 (kobs
=
k[Et3SiH][40]), which allowed the extraction of activation
parameters for this process. These confirm a low barrier for
silylium exchange between the two E-H donors (DG‡(300K) =
12.5-12.6 kcal mol–1 for [Et3SiH] = 0.14 M) and the expected
associative nature of such an exchange (DS‡ = –111 eu).
Et3Si
Et3Si
H
H
LA
Et3SiH
OSiEt3
Ph
Catalysis
OSiEt3
LA
H
Me
Et3SiH
Me
Ph
H
+
O
LA
Reversible exchange of silylium ions between Lewis bases is
well known,[38,41] but very few examples involve h1-E-H bond
adducts like 3 and 4. Silane adduct 4 itself undergoes self-
exchange with free silane in solution.[37a] The resting state for an
iridium system active for the catalytic reduction of ethers by
silanes involves a well-characterized equilibrium that transfers
R3Si+ between an Ir-H donor and the substrate ether; the
equilibrium lies toward the silyloxonium ether adduct.[42] A similar
transfer of B-H-stabilized silylium to substrate is critical to the
catalytic activity of B(C6F5)3 in reactions of silanes, but the
reversibility of the process is not evident under catalytic conditions
because the resulting silylorganocation is rapidly quenched by
H
H
Et3SiH
Me
Ph
OSiEt3
Ph
Initiation
Me
Ph
Et3Si
H
+
Me
–
LA
H
O(SiEt3)2
[Mo]
(C6F5)3B
H
Ph3C
H
LA
H
P
R2
1
Scheme 5. Whether a LA acts as catalyst (top) or initiator (bottom) for the
hydrosilylation of acetophenone with Et3SiH depends on how effectively LA-H
competes with Et3Si-H as a hydride donor to the carbocationic intermediate.
hydride
transfer
from
the
associated
hydridoborate,
Table 1. Product distributions for LA-mediated reactions of ketones with
[HB(C6F5)3)]–.[10a,b] However, in the absence of nucleophilic
substrate, B(C6F5)3 does mediate H/D exchange between silanes,
which implicitly involves silylium transfer between distinct Si-H(D)
and B-H(D) donors.[10a,b] This is the only other example we have
found of such transfer between two distinct E-H bonds.
Et3SiH.[a]
Lewis acid
R
RC(=O)Ph
Et3SiOCH(R)Ph
RCH2Ph[b]
[2][B(ArCl)4]
B(C6F5)3
Me
Me
Me
Ph
Ph
Ph
0
1
1
0
The above examples highlight a balance of hydride donor
abilities that is relevant to the mechanisms of Lewis acid
participation in catalytic reactions of silanes (cf. Scheme 5). For
the Ir and B(C6F5)3 systems, the “LA-H” that results from
electrophilic Si-H activation both stabilizes a reactive silylium ion
and competes effectively with free silane to quench reactive
silylorganocation intermediates. For other LAs, most notably trityl
cation, the LA-H generated is a poorer hydride donor than
substrate silane; these LAs initiate catalysis by generating silylium
ions, which are actually responsible for catalytic turnover, since it
is the silane Si-H that quenches the intermediate carbocation,
generating a new silylium. Seminal mechanistic studies of the
participation of B(C6F5)3 in hydrosilylation of carbonyl groups
relied on the catalytic hydrosilylation of acetophenone by Et3SiH
as a control experiment to diagnose the hydride donor role of
transient hydridoborate in catalysis.[10a] Under identical reaction
conditions, B(C6F5)3 gave exclusively the silyl ether resulting from
a single reduction event, while trityl cation and isolated “Et3Si+”
0
0
[Ph3C][B(C6F5)4]
[2][B(ArCl)4]
B(C6F5)3
0.5
0.29
0.17
0.5
0
0.5
0.29
0.17
0.5
0.42
0.67
0
[Ph3C][B(C6F5)4]
[a]Reactions were performed at RT in CD2Cl2 using a 1:1 ratio of ketone:Et3SiH
and 10 mol% of the LA, and were complete within the time taken for mixing and
obtaining spectra (~15 min). Et3SiH was added to 2a-b before ketone to ensure
formation of 3a-b. Et3SiH and ketone were premixed before addition to B(C6F5)3
and [Ph3C][B(C6F5)4]. [b]Accompanied by an equimolar amount of Et3SiOSiEt3.
Conclusion
These results are particularly interesting in the context of two
other observations. First, we can generate catalytic mixtures
containing 3 through the addition of excess silane to 2 in CD2Cl2,
but comparable pre-mixing of either B(C6F5)3 or [Ph3C][B(C6F5)4]
with silane in CD2Cl2 leads to catalytic inactivity, presumably due
to decomposition of in situ-generated silylium in the chlorinated
solvent. 1:1 mixtures of silane and ketone must be added to these
LAs to give catalysis in this solvent. This speaks to the importance
of the equilibrium between 3 and 4 in stabilizing the Et3Si+ ion in
this solvent. Second, the ability of complex 1 to out-compete
silane as a hydride donor in this catalysis is consistent with the
relative amounts of 3 and 1 (~3-5:1) we observe at low
temperature by NMR, but 1 is a weaker hydride donor than
[HB(C6F5)3]– - B(C6F5)3 does not react with 1. This points to the
delicate interplay of thermodynamic (Lewis acidity) and kinetic
(degree of association of LA-H with transient silylorganocation)
participated in “over-reduction”
- facile deoxygenation of
acetophenone to alkane and disiloxane (Scheme 5). This
divergent behaviour was attributed to low steric hindrance
associated with silane as the hydride donor, as opposed to the
bulky [HB(C6F5)3)]–, which stays closely associated with the
silylium/carbocation species in the borane catalysis. Recently,
product analysis for this same control reaction was used to probe
the extent to which Lewis acidic fluorophosphonium cations act
as catalysts or initiators.[12a] The hydrosilylation of acetophenone
mediated by complex 2 gives solely the silyl ether product
resulting from a single reduction event (Table 1), similar to
B(C6F5)3. This result suggests that equilibrium mixture of [3 +
Et3SiH] and [1 + 4] is indeed catalysing the reduction, presumably
via participation of an intermediate or transition state involving
close association of the hydride donor 1 with the carbocation
4
This article is protected by copyright. All rights reserved.

10.1002/anie.202011372
Angewandte Chemie International Edition
RESEARCH ARTICLE
[6]
a) J. C. L. Walker, H. F. T. Klare, M. Oestreich, Nat. Rev. Chem. 2020,
4, 54-62; b) A. E. Waked, S. S. Chitnis, D. W. Stephan, Chem. Commun.
2019, 55, 8971-897; c) M. Wiesemann, B. Hoge, Chem. Eur. J. 2018, 24,
16457-16471; d) R. Maskey, M. Schädler, C. Legler, L. Greb, Angew.
Chem. Int. Ed. 2018, 57, 1717-1720; Angew. Chem. 2018, 130, 1733-
1736; e) A. L. Liberman-Martin, R. G. Bergman, T. D. Tilley, J. Am. Chem.
Soc. 2015, 137, 5328-5331.
parameters that allows a LA to catalyze, as opposed to simply
initiate, reactions of silanes.
We continue to explore the scope of catalytic reactivity using
this new, unusually strong LA system, whose potentially tuneable
ancillary metal fragment provides intriguing prospects for further
development of the catalytic reactions of silanes.
[7]
a) J. Ramler, K. Hofmann, C. Lichtenberg, Inorg. Chem. 2020, 59, 3367-
3376; b) A. Brand, A. Hentschel, A. Hepp, W. Uhl, Eur. J. Inorg. Chem.
2020, 2020, 361-369; c) J. Zhou, L. L. Liu, L. L. Cao, D. W. Stephan,
Angew. Chem. Int. Ed. 2019, 58, 5407-5412; Angew. Chem. 2019, 131,
5461-5466; d) M. Olaru, D. Duvinage, E. Lork, S. Mebs, J. Beckmann,
Chem. Eur. J. 2019, 25, 14758-14761; e) P. Mehlmann, T. Witteler, L. F.
B. Wilm, F. Dielmann, Nat. Chem. 2019, 11, 1139-1143; f) J. C. Gilhula,
A. T. Radosevich, Chem. Sci. 2019, 10, 7177-7182; g) J. Zhou, L. L. Liu,
L. L. Cao, D. W. Stephan, Chem 2018, 4, 2699-2708; h) M. Yang, D.
Tofan, C.-H. Chen, K. M. Jack, F. P. Gabbaï, Angew. Chem. Int. Ed. 2018,
57, 13868-13872; Angew. Chem. 2018, 130, 14064-14068; i) J. M.
Bayne, D. W. Stephan, Chem. Soc. Rev. 2016, 45, 765-774; j) S. S.
Chitnis, A. P. M. Robertson, N. Burford, B. O. Patrick, R. McDonald, M.
J. Ferguson, Chem. Sci. 2015, 6, 6545-6555.
Acknowledgements
We thank NSERC of Canada for funding (Discovery Grant to LR,
PGS-D to RB) and Sofia Donnecke and Anuj Joshi, from the
group of Prof. J. Scott McIndoe at the University of Victoria, for
ESI-MS experiments. D.A.P. gratefully acknowledges support by
the Max Planck Society.
Keywords: hydrosilylation • Lewis acids • phosphenium
complex • Si-H activation • silylium adduct
[8]
[9]
a) M. Hejda, D. Duvinage, E. Lork, R. Jirásko, A. Lyčka, S. Mebs, L.
Dostál, J. Beckmann, Organometallics 2020, 39, 1202-1212; b) F. A.
Tsao, A. E. Waked, L. Cao, J. Hofmann, L. Liu, S. Grimme, D. W.
Stephan, Chem. Commun. 2016, 52, 12418-12421.
[1]
[2]
M. C. Lipke, A. L. Liberman-Martin, T. D. Tilley, Angew. Chem. 2017, 129,
2298-2335; Angew. Chem. Int. Ed. 2017, 56, 2260-2294.
Recent reviews and reports of the participation of electrophilic Si-H
activation in catalysis include: a) A. Roy, V. Bonetti, G. Wang, Q. Wu, H.
F. T. Klare, M. Oestreich, Org. Lett. 2020, 22, 1213-1216; b) T. Lundrigan,
E. N. Welsh, T. Hynes, C.-H. Tien, M. R. Adams, K. R. Roy, K. N.
Robertson, A. W. H. Speed, J. Am. Chem. Soc. 2019, 141, 14083-14088;
c) D. Hartmann, M. Schädler, L. Greb, Chem. Sci. 2019, 10, 7379-7388;
d) T. Hackel, A. N. McGrath, Molecules 2019, 24, 432-461; e) S.
Balasubramaniam, S. Kumar, A. P. Andrews, B. Varghese, E. D. Jemmis,
A. Venugopal, Eur. J. Inorg. Chem. 2019, 2019, 3265-3269; f) K. G.
Pearce, A. M. Borys, E. R. Clark, H. J. Shepherd, Inorg. Chem. 2018, 57,
11530-11536; g) V. Fasano, J. H. W. LaFortune, J. M. Bayne, M. J.
Ingleson, D. W. Stephan, Chem. Commun. 2018, 54, 662-665; h) V.
Fasano, J. E. Radcliffe, L. D. Curless, M. J. Ingleson, Chem. Eur. J. 2017,
23, 187-193; i) 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; j) J. Chen, L. Falivene, L.
Caporaso, L. Cavallo, E. Y. X. Chen, J. Am. Chem. Soc. 2016, 138, 5321-
5333; k) M. Oestreich, J. Hermeke, J. Mohr, Chem. Soc. Rev. 2015, 44,
2202-2220; l) B. Pan, F. P. Gabbaï, J. Am. Chem. Soc. 2014, 136, 9564-
9567; m) T. Robert, M. Oestreich, Angew. Chem. Int. Ed. 2013, 52, 5216-
5218; Angew. Chem. 2013, 125, 5324-5326; n) M. Pérez, L. J. Hounjet,
C. B. Caputo, R. Dobrovetsky, D. W. Stephan, J. Am. Chem. Soc. 2013,
135, 18308-18310; o) M. Iglesias, P. J. Sanz Miguel, V. Polo, F. J.
Fernández-Alvarez, J. J. Pérez-Torrente, L. A. Oro, Chem. Eur. J. 2013,
19, 17559-17566; p) W. E. Piers, A. J. V. Marwitz, L. G. Mercier, Inorg.
Chem. 2011, 50, 12252-12262.
The acceptor character of many recent examples of strong main group
Lewis acids has been shown to arise from diverse, low-lying s* orbitals
and/or s-hole behaviour. See references 2e, 2f, 7f, 7h, 8b and a) S. S.
Chitnis, F. Krischer, D. W. Stephan, Chem. Eur. J. 2018, 24, 6543-6546;
b) S. Benz, A. I. Poblador-Bahamonde, N. Low-Ders, S. Matile, Angew.
Chem. Int. Ed. 2018, 57, 5408-5412; Angew. Chem. 2018, 130, 5506-
5510; c) M. Yang, F. P. Gabbaï, Inorg. Chem. 2017, 56, 8644-8650; d)
C. B. Caputo, L. J. Hounjet, R. Dobrovetsky, D. W. Stephan, Science
2013, 341, 1374-1374.
[10] a) D. J. Parks, J. M. Blackwell, W. E. Piers, J. Org. Chem. 2000, 65,
3090-3098; b) S. Rendler, M. Oestreich, Angew. Chem. Int. Ed. 2008, 47,
5997-6000; Angew. Chem. 2008, 120, 6086-6089; c) K. Sakata, H.
Fujimoto, J. Org. Chem. 2013, 78, 12505-12512.
[11]
C. A. Reed, Acc. Chem. Res. 1998, 31, 325-332.
[12] Recent examples of Lewis acids that have been shown to act as initiators
in catalytic reactions of silanes include references 2l, 6b, and a) L. Süsse,
J. H. W. LaFortune, D. W. Stephan, M. Oestreich, Organometallics 2019,
38, 712-721.
[13] a) K. Müther, J. Mohr, M. Oestreich, Organometallics 2013, 32, 6643-
6646; b) K. Müther, M. Oestreich, Chem. Commun. 2011, 47, 334-336;
c) V. J. Scott, R. Çelenligil-Çetin, O. V. Ozerov, J. Am. Chem. Soc. 2005,
127, 2852-2853; d) J. B. Lambert, Y. Zhao, H. Wu, J. Org. Chem. 1999,
64, 2729-2736; e) J. B. Lambert, Y. Zhao, J. Am. Chem. Soc. 1996, 118,
7867-7868.
[14] a) J. M. Slattery, S. Hussein, Dalton Trans. 2012, 41, 1808-1815; b) B.
D. Ellis, P. J. Ragogna, C. L. B. Macdonald, Inorg. Chem. 2004, 43, 7857-
7867.
[3]
[4]
S. A. Weicker, D. W. Stephan, Bull. Chem. Soc. Jpn. 2015, 88, 1003-
1016, and references therein.
[15] a) D. Gudat, Coord. Chem. Rev. 1997, 163, 71-106; b) A. H. Cowley, R.
A. Kemp, Chem. Rev. 1985, 85, 367-382.
a) E. A. Patrick, W. E. Piers, Chem. Commun. 2020, 56, 841-853; b) C.
P. Manankandayalage, D. K. Unruh, C. Krempner, Dalton Trans. 2020,
49, 4834-4842; c) J. L. Carden, A. Dasgupta, R. L. Melen, Chem. Soc.
Rev. 2020, 49, 1706-1725; d) D. Tanaka, Y. Kadonaga, Y. Manabe, K.
Fukase, S. Sasaya, H. Maruyama, S. Nishimura, M. Yanagihara, A.
Konishi, M. Yasuda, J. Am. Chem. Soc. 2019, 141, 17466-17471; e) A.
Ben Saida, A. Chardon, A. Osi, N. Tumanov, J. Wouters, A. I. Adjieufack,
B. Champagne, G. Berionni, Angew. Chem. Int. Ed. 2019, 58, 16889-
16893; Angew. Chem. 2019, 131, 17045-17049; f) J. F. Kögel, A. Y.
Timoshkin, A. Schröder, E. Lork, J. Beckmann, Chem. Sci. 2018, 9,
8178-8183; g) J. F. Kögel, D. A. Sorokin, A. Khvorost, M. Scott, K. Harms,
D. Himmel, I. Krossing, J. Sundermeyer, Chem. Sci. 2018, 9, 245-253;
h) A. Y. Houghton, V. A. Karttunen, W. E. Piers, H. M. Tuononen, Chem.
Commun. 2014, 50, 1295-1298.
[16] N. Đorđević, R. Ganguly, M. Petković, D. Vidović, Inorg. Chem. 2017, 56,
14671-14681.
[17] a) R. C. King, S. Nilewar, B. T. Sterenberg, J. Organomet. Chem. 2019,
880, 68-74; b) C. M. Feil, T. D. Hettich, K. Beyer, C. Sondermann, S. H.
Schlindwein, M. Nieger, D. Gudat, Inorg. Chem. 2019, 58, 6517-6528; c)
S. Nilewar, A. Jayaraman, B. T. Sterenberg, Organometallics 2018, 37,
4699-4710; d) A. Jayaraman, S. Nilewar, T. V. Jacob, B. T. Sterenberg,
ACS Omega 2017, 2, 7849-7861; e) M. Gediga, C. M. Feil, S. H.
Schlindwein, J. Bender, M. Nieger, D. Gudat, Chem. Eur. J. 2017, 23,
11560-11569; f) A. Jayaraman, B. T. Sterenberg, Organometallics 2016,
35, 2367-2377.
[18] a) H. Nakazawa, Adv. Organomet. Chem. 2004, 50, 108-108; b) H.
Nakazawa, J. Organomet. Chem. 2000, 611, 349-363.
[5]
a) P. Eisenberger, C. M. Crudden, Dalton Trans. 2017, 46, 4874-4887;
b) K. Jakobsson, T. Chu, G. I. Nikonov, ACS Catal. 2016, 6, 7350-7356;
and references therein.
[19] M. W. Bezpalko, A. M. Poitras, B. M. Foxman, C. M. Thomas, Inorg.
Chem. 2017, 56, 503-510.
5
This article is protected by copyright. All rights reserved.

10.1002/anie.202011372
Angewandte Chemie International Edition
RESEARCH ARTICLE
[20] L. Rosenberg, Coord. Chem. Rev. 2012, 256, 606-626.
[21] This argument is consistent with cis/trans isomer ratios reported for other
Mo phosphenium complexes containing CO ligands. a) H. Nakazawa, Y.
Yamaguchi, T. Mizuta, K. Miyoshi, Organometallics 1995, 14, 4173-
4182; b) H. Nakazawa, Y. Yamaguchi, K. Miyoshi, J. Organomet. Chem.
1994, 465, 193-198.
on the NMR timescale. This demonstrates the impact of decreasing
silane conentration on the rate of this bimolecular exchange reaction.
[41] a) C. A. Roselli, M. R. Gagné, Org. Biomol. Chem. 2018, 16, 2318-2323;
b) J. Chu, T. G. Carroll, G. Wu, J. Telser, R. Dobrovetsky, G. Ménard, J.
Am. Chem. Soc. 2018, 140, 15375-15383; c) J. B. Lambert, S. Zhang, S.
M. Ciro, Organometallics 1994, 13, 2430-2443; d) S. R. Bahr, P.
Boudjouk, J. Am. Chem. Soc. 1993, 115, 4514-4519; e) M. Kira, T. Hino,
H. Sakurai, J. Am. Chem. Soc. 1992, 114, 6697-6700.
[22] Representative crystallographic studies show Mo=PR2 distances of 2.20-
2.25 Å and Mo-PR3 distances of 2.42-2.53 Å. See reference 21a and a)
D. Gudat, A. Haghverdi, M. Nieger, J. Organomet. Chem. 2001, 617-618,
383-394; b) A. H. Cowley, D. M. Giolando, C. M. Nunn, M. Pakulski, D.
Westmoreland, N. C. Norman, J. Chem. Soc., Dalton Trans. 1988, 2127-
2134; c) L. D. Hutchins, H. U. Reisacher, G. L. Wood, E. N. Duesler, R.
T. Paine, J. Organomet. Chem. 1987, 335, 229-237; d) V. W. Day, I.
Tavanaiepour, S. S. Abdel-Meguid, J. F. Kirner, L. Y. Goh, E. L.
Muetterties, Inorg. Chem. 1982, 21, 657-663.
[42] J. Yang, P. S. White, M. Brookhart, J. Am. Chem. Soc. 2008, 130, 17509-
17518.
[23] (a) A. D. Becke, Phys. Rev. A 1988, 38, 3098-3100; (b) J. P. Perdew,
Phys. Rev. B 1986, 33, 8822-8824.
[24] E. Caldeweyher, C. Bannwarth, S. Grimme, J. Chem. Phys. 2017, 147,
034112.
[25] (a) J. D. Rolfes, F. Neese, D. A. Pantazis, J. Comput. Chem. 2020, 41,
1842-1849; (b) D. A. Pantazis, X. Y. Chen, C. R. Landis and F. Neese,
J. Chem. Theory Comput. 2008, 4, 908-919; (c) F. Weigend and R.
Ahlrichs, Phys. Chem. Chem. Phys. 2005, 7, 3297-3305.
[26] Recent examples of N-heterocyclic phosphenium complexes exhibiting
double bond character include references 17b, 17e, 19, and a) M. Gediga,
S. H. Schlindwein, J. Bender, M. Nieger, D. Gudat, Angew. Chem. Int.
Ed. 2017, 56, 15718-15722; Angew. Chem. 2017, 129, 15924-15928; b)
M. W. Bezpalko, B. M. Foxman, C. M. Thomas, Inorg. Chem. 2015, 54,
8717-8726.
[27] Complex 2 also reacts with acetonitrile and methanol to give complex
mixtures of unidentified products.
[28] Other reports of ring-opening polymerization of THF mediated by main
group Lewis acids include references 2l, 7c, 8b, and 9c.
[29] a) M. A. Beckett, G. C. Strickland, J. R. Holland, K. Sukumar Varma,
Polymer 1996, 37, 4629-4631; b) U. Mayer, V. Gutmann, W. Gerger,
Monatsh. Chem. 1975, 106, 1235-1257.
[30] R. G. Parr, L. v. Szentpály, S. Liu, J. Am. Chem. Soc. 1999, 121, 1922-
1924.
[31] H. Iikura, T. Tsuneda, T. Yanai, K. Hirao, J. Chem. Phys. 2001, 115,
3540-3544.
[32] a) A. R. Jupp, T. C. Johnstone, D. W. Stephan, Inorg. Chem. 2018, 57,
14764-14771; b) A. R. Jupp, T. C. Johnstone, D. W. Stephan, Dalton
Trans. 2018, 47, 7029-7035.
[33]
L. Greb, Chem. Eur. J. 2018, 24, 17881-17896.
[34] Reports of facile, reversible h1-silane binding at Lewis acids include
references 2g, 2h, 10a, 10b, and a) J. Chen, E. Y. X. Chen, Angew.
Chem. Int. Ed. 2015, 54, 6842-6846; Angew. Chem. 2015, 127, 6946-
6950; b) A. Y. Houghton, J. Hurmalainen, A. Mansikkamäki, W. E. Piers,
H. M. Tuononen, Nat. Chem. 2014, 6, 983-983; c) J. Yang, M. Brookhart,
J. Am. Chem. Soc. 2007, 129, 12656-12657.
[35] The downfield chemical shift for one of the 31P{1H} signals for 3 (e.g. -9.5
ppm for 3b) may reflect proximity to the silylium cation of the phosphorus
participating in the P-H-Si interaction. The upfield shift of the other signal
(d6-8 ppm) is closer to that observed for the secondary phosphines in
neutral complex 1 (d16-17 ppm).
[36] J. Yang, P. S. White, C. K. Schauer, M. Brookhart, Angew. Chem. Int.
Ed. 2008, 47, 4141-4143; Angew. Chem. 2008, 120, 4209-4211.
[37] a) S. J. Connelly, W. Kaminsky, D. M. Heinekey, Organometallics 2013,
32, 7478-7481; b) M. Nava, C. A. Reed, Organometallics 2011, 30, 4798-
4800; c) S. P. Hoffmann, T. Kato, F. S. Tham, C. A. Reed, Chem.
Commun. 2006, 767-769.
[38] A. Gudz, P. R. Payne, M. R. Gagné, Organometallics 2017, 36, 4047-
4053.
[39] The ability of trialkylsilylium ions to abstract halide from halogenated
solvents, and to scavenge water from solvents and even the surface of
glassware is well established. See reference 11.
[40] Room temperature 1H NMR spectra for experiments involving the
addition of 2, 1, and 0.9 equiv silane to 2a show increasing line widths
for the (averaged) P-H signals, consistent with slowing of the exchange
6
This article is protected by copyright. All rights reserved.

10.1002/anie.202011372
Angewandte Chemie International Edition
RESEARCH ARTICLE
Entry for the Table of Contents
No MoPing around: A new P(III) Lewis superacid with an ancillary molybdenum carbonyl fragment reacts with Et3SiH to give a P-H
stabilized silylium ion that participates in catalytic hydrosilylation.
Institute and/or researcher Twitter usernames: @chemistryatuvic, @lisarose_ltpb, @belli_roman
7
This article is protected by copyright. All rights reserved.