Cationic (η5-C5Me4R)RhIII Complexes with Metalated Aryl Phosphines Featuring η4-Phosphorus plus Pseudo-Allylic Coordination

Article abstract of DOI:10.1021/acs.organomet.7b00688

In this contribution we study experimentally and computationally some electrophilic cationic (η⁵-C₅Me₄R)Rh^III complexes containing a cyclometalated bis(aryl) phosphine, PR′Ar₂. The phosphine Ar groups feature methyl substituents at the 2- and 6-positions of the aromatic rings, allowing the formation of the complexes [(η⁵-C₅Me₄R)Rh(C^{^}P)]⁺ (3⁺), where the metalated phosphine exhibits η⁴ coordination to rhodium through phosphorus and the carbon atoms of the adjoining pseudo-allylic functionality. The solution and solid-state structures of complexes 3⁺ have been studied by NMR and X-ray methods and their electronic properties investigated with the aid of DFT calculations. The Lewis acid behavior of complexes 3⁺ has also been addressed, concentrating on reactivity toward CO, H₂,and hydrosilanes. Some catalytic Si-H/Si-D exchange and hydrosilylation reactions are also reported.

Full text of DOI:10.1021/acs.organomet.7b00688

Article
Cite This: Organometallics XXXX, XXX, XXX−XXX
Cationic (η5‑C5Me4R)RhIII Complexes with Metalated Aryl Phosphines
Featuring η4‑Phosphorus plus Pseudo-Allylic Coordination
María F. Espada, Ana C. Esqueda,† Jesus Campos, Miguel Rubio, Joaquín Lopez-Serrano,
́
́
́
Eleuterio Alvarez, Celia Maya, and Ernesto Carmona*
Instituto de Investigaciones Químicas (IIQ) and Departamento de Química Inorgan
Americo Vespucio 49, 41092 Sevilla, Spain
†Escuela de Nivel Medio Superior de Leon
, Universidad de Guanajuato, Hermanos Aldama y Blvd. Torres Landa s/n, Leon
Guanajuato, Mexico CP 37480
́
ica, CSIC and Universidad de Sevilla, Avda.
́
́
́
,
́
S
* Supporting Information
ABSTRACT: In this contribution we study experimentally and computationally some
electrophilic cationic (η5-C5Me4R)RhIII complexes containing a cyclometalated bis(aryl)
phosphine, PR′Ar2. The phosphine Ar groups feature methyl substituents at the 2- and 6-
positions of the aromatic rings, allowing the formation of the complexes [(η5-
C5Me4R)Rh(C∧P)]+ (3+), where the metalated phosphine exhibits η4 coordination to
rhodium through phosphorus and the carbon atoms of the adjoining pseudo-allylic
functionality. The solution and solid-state structures of complexes 3+ have been studied
by NMR and X-ray methods and their electronic properties investigated with the aid of
DFT calculations. The Lewis acid behavior of complexes 3+ has also been addressed,
concentrating on reactivity toward CO, H2 ,and hydrosilanes. Some catalytic Si−H/Si−D
exchange and hydrosilylation reactions are also reported.
This uncommon structural motif is thought to be responsible
for their unusual reactivity properties, including intermolecular
H−H and Si−H activation,4 as well as intramolecular C−H
oxidative cleavage and C−C bond formation.5a
INTRODUCTION
■
Complexes of Rh(III) and Ir(III) built on a (η5-C5Me5)M
framework are the subject of numerous experimental and
computational studies, owing to their capacity to participate in
a plethora of chemical transformations, comprising many
catalytic reactions.1,2 In recent years we have investigated
diverse complexes of this type, particularly those that possess a
monoanionic cyclometalated phosphine ligand, C∧P. It has long
been known that the cyclometalation reaction is considerably
facilitated by steric hindrance and by the formation of a stable
five-membered ring.3 Accordingly, we chose aryl phosphines
fulfilling these requirements, principally some mono- and bis-
xylyl phosphines, PR2Xyl and PRXyl2, where R is an alkyl group
and Xyl stands for 2,6-Me2C6H3. Other closely related PRAr2
phosphines were also studied.4−7
With the existing information on the interesting reactivity
reported for the parent [(η5-C5Me5)M(Me)(PMe3)-
(ClCH2Cl)]+ (M = Ir, Rh) complexes,8 particularly in C−H
bond activation reactions, the initial objective of this work was
to ascertain whether the related metallacyclic complexes [(η5-
C5Me5)M(C∧P)(ClCH2Cl)]+, incorporating an M−C σ bond
and a P-donor atom within the chelating C∧P ligand, would
behave similarly. None of the new compounds prepared were
reactive under mild conditions toward intermolecular C−H
bond activation, although those derived from the PMeXyl2
ligand exhibited interesting dynamic and chemical behavior.
Instead of the foreseen CH2Cl2 adducts, both the rhodium4 and
the iridium5a,6 complexes of this ligand featured a metalated
phosphine with an unusual η4 coordination involving the
phosphorus atom and the contiguous pseudo-allylic fragment.
The formally tridentate, η4 coordination mode of the
metalated PMeXyl2 ligand could only be authenticated by X-
ray crystallography for one rhodium complex and was
preliminarily communicated along with highly efficient catalytic
hydrogen isotope exchanges in a variety of hydrosilanes (Si−H,
Si−D, and Si−T).4 The same type of ligand coordination was
unequivocally established by NMR methods in some related
iridium complexes but could not be substantiated by X-ray
studies, as single crystals could not be grown.5a,6 Moreover,
although this coordination cannot be viewed as exceptional, for
after all it combines P and η3-benzylic bonding, surprisingly it
finds no precedent among the compounds registered in the
Cambridge Structural Database (CSD).9 It is additionally
puzzling that such a structure appears not to be accessible for
closely related Rh and Ir complexes of the mono(xylyl)
phosphines PR2Xyl (R = Me, Ph, iPr, Cy).7 For rhodium, stable
CH2Cl2 adducts [(η5-C5Me5)Rh(C∧P)(ClCH2Cl)]+ derived
from PMe2Xyl and PiPr2Xyl were observed, whereas iridium
yielded similar adducts on coordination to the less bulky
phosphines PMe2Xyl and PPh2Xyl and hydride alkylidene
complexes resulting from α-H elimination when the bulkier
PiPr2Xyl and PCy2Xyl were utilized.7
Received: September 8, 2017
© XXXX American Chemical Society
A
DOI: 10.1021/acs.organomet.7b00688
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Considering the above, we pursued additional work on the
occurrence and stability of the η4, P plus pseudo-allylic
cyclometalated structure. Herein we report experimental and
computational studies on a variety of (η5-C5Me5R)RhIII
complexes of metalated PR′Ar2 ligands (R′ = Me, Et; Ar =
aryl group; vide infra). The structures of the chloride precursors
are schematically represented in Figure 1. We focus attention
Scheme 1. Synthesis of Chloride Complex 2e
characteristic NMR signals were unambiguously identified in
mixtures with the corresponding chlorides. Furthermore,
complex 1a was synthesized independently from the reaction
of 2a with NaBH4, although due to its marked reactivity toward
oxygen and chlorinated solvents, analytically pure samples
could not be obtained (see the Experimental Section for
details). We, however, managed to isolate microanalytically
pure samples of the hydride complex 1i derived from PMeXyl2
but containing C5Me4SiMe3 as the capping ligand and provide
full characterization data in the accompanying Supporting
Information. All of the chloride complexes 2a−2g were
obtained as a single diastereoisomer with the stereochemistry
shown in Scheme 1 for 2e, with the only exception being 2c,
which formed as a 3:2 mixture of stereomers. Compounds 2a−
2g were fully characterized by microanalysis and NMR
spectroscopy. A comprehensive set of NMR data can be
found in the Supporting Information. In addition, the solid-
state molecular structures of compounds 2a, 2b, and 2g were
determined by single-crystal X-ray studies. Figures S2 and S3 in
the Supporting Information collect ORTEP perspective views
of the molecular structures of complexes 2b and 2g,
respectively. The structure of 2a is shown in Figure 2 (left)
to facilitate comparison with that of the corresponding cationic
complex 3a+ (vide infra).
Figure 1. Rh(III) chloride complexes 2 utilized in this work and
labeling scheme. As indicated in the text, ArBut represents 3,5-But2-
2
C6H3.
on the rhodium−cyclometalated phosphine linkage exhibiting
η4 coordination and discuss its solution dynamic behavior, X-
ray metrical parameters, and reactivity toward H2, silanes, and
catalytic hydrosilylation. Computational studies are also
provided. As noted briefly, part of this work has been
communicated.4,10
RESULTS AND DISCUSSION
■
Neutral Hydride and Chloride Complexes. To gain
access to the target cationic complexes [(η5-C5Me4R)Rh-
(C∧P)]+ containing a cyclometalated PR′Ar2 phosphine, the
corresponding neutral chloride precursors were needed (Figure
1)
Cationic Complexes Resulting from Chloride Abstrac-
tion from Compounds 2. Reactions of the cyclometalated
chloride complexes 2 with NaBArF (BArF = B(3,5-
C6H3(CF3)2)4) in the presence of a Lewis base L gave the
expected cationic adducts [(η5-C5Me5)Rh(C∧P)(L)]+ (3·L+),
which were isolated as salts of the BArF anion. In general,
compounds 3·L+ formed as a mixture of syn and anti
diastereomers, with the former being predominant. Carbonyl
adducts 3·CO+ were generated from all complexes 2 except 2g.
For the C5Me4But system the acetonitrile derivative 3f·NCMe+
was also isolated (Scheme 2), while from the parent compound
2a, 3·L+ cations were also obtained for L = NCMe, NH3, PMe3,
CNXyl, C2H4. Here we concentrate our attention on the
carbonyl complexes 3a·CO+−3f·CO+; characterization data for
other 3·L+ compounds are given in the Supporting Information.
While comparable iridium complexes were readily prepared
from the iridium dimer [(η5-C5Me5)IrCl2]2 and the corre-
sponding PRAr2 phosphine under appropriate experimental
conditions,5,6 for rhodium the analogous reaction proved
unsatisfactory and gave rise to a manifold of products
containing only low yields of the desired complex. Accordingly,
complexes 2a−2g were obtained by a somewhat more elaborate
procedure that yielded first the related hydrides 1a−1g
accompanied by small amounts of the desired chlorides. The
hydrides formed in a two-step reaction of [RhCl(C2H4)2]2, first
with the phosphine at −40 °C and then with Zn(C5Me5)2 as
cyclopentadienyl transfer reagent.11 The resulting mixture was
stirred for about 5 h while it was warmed to −20 °C and then
worked up at room temperature. Direct cyclometalation was
observed with formation of hydrides 1a−1g, along with minor
amounts of chlorides 2a−2g, the formation of the latter being a
consequence of the high reactivity of the hydrides toward
chlorinating agents. Indeed, addition of CH2Cl2 or CHCl3 to
the crude reaction mixtures produced exclusively the chloride
complexes 2a−2g, which were isolated in high yields (≥80%).
Scheme 1 outlines the synthesis of 2e, derived from the
phosphine PEtXyl2.
Some of these adducts, viz. 3a·CO+, 3a·NCMe+, and 3a·C2H4 ,
+
were additionally studied by X-ray crystallography (see below
and also the Supporting Information).
Carbonyl complexes [3·CO][BArF] were obtained as yellow
or orange solids. In the 13C{1H} NMR spectra, the carbonyl
ligand resonates at around 189 ppm as a doublet of doublets,
1
2
with JCRh ≈ 75 Hz and JCP ≈ 20 Hz. The IR spectra display
the expected carbonyl stretching vibrations in the short interval
2055−2050 cm−1, except for the F-substituted complex 3d·
CO+, where it is slightly shifted to higher wavenumbers (2060
cm−1). The high ν(CO) values hint at back-donation of
somewhat smaller magnitude in comparison to that in the
Ir(III) analogues,12 which feature bands at 2030−2035 cm−1
(2040 cm−1 in the fluoro-substituted counterpart of 3d·CO+).
No attempts were made to fully characterize hydride
complexes 1a−1g because they were considered of secondary
importance for the objectives of this work. Nevertheless, their
B
DOI: 10.1021/acs.organomet.7b00688
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Figure 2. ORTEP perspective views of the molecules of 2a and 3a+. Hydrogen atoms are excluded for clarity, and thermal ellipsoids are set at 50%.
Selected bond distances (Å) and angles (deg) for 2a: Rh1−P1 = 2.255(1); Rh1−C26 = 2.101(4); Rh1−Cl1 = 2.390(1); P1−C20 = 1.821(3); C26−
C21 = 1.488(5); C20−C21 = 1.391(5); P1−Rh1−C26 = 81.5(1); Rh1−P1−C11 = 115.7(1); Cl1−Rh1−C26 = 87.9(1); Cl1−Rh1−P1 = 90.05(3).
Selected bond distances (Å) and angles (deg) for 3a+: Rh1−P1 = 2.260(1); Rh1−C17 = 2.254(5); Rh1−C12 = 2.201(5); Rh1−C11 = 2.135(5);
P1−C17 = 1.788(5); C17−C12 = 1.458(7); C12−C11 = 1.439(7); P1−C17−C12 = 111.8(4); C17−C12−C11 = 117.6(4).
Scheme 2. Synthesis of the Cationic Adducts 3f·CO+ and 3f·
Scheme 3. Synthesis of Cationic Complex 3d+ as a
Representative Example of a Cyclometalated Phosphine with
η3-Benzylic Coordination
NCMe+
due to the expected8 adduct 3a·CH2Cl2 (vide infra) that
+
This electronic picture is common for cationic carbonyl
complexes of the transition metals,13 including other late
transition metals such as Pt(II) and Au(I). Also in accordance
with these arguments, 3a·CNXyl+ exhibits the ν(CN) band
at 2150 cm−1, which is approximately 35 cm−1 higher in energy
than in the free CNXyl. Clearly, the cationic Rh(III) center of
the [(η5-C5Me5)Rh(C∧P)]+ fragments investigated has a weak
capability to act as a π donor toward π-acid ligands and,
moreover, the electronic properties do not change notably with
the constitutional modifications that were introduced in the
cyclopentadienyl and phosphine ligands employed in this work.
In contrast to previous results for the related rhodium and
iridium complexes [(η5-C5Me5)M(Me)Cl(PMe3)],8 CH2Cl2
solutions of the chlorides 2 reacted at room temperature with
NaBArF (or AgPF6) to afford the solvent-free cationic
complexes [(η5-C5Me5)Rh(C∧P)]+ (3+), in which the cyclo-
metalated phosphine is bonded to rhodium in an η4 fashion,
through the phosphorus and the three carbon atoms of the
adjoining benzylic fragment (Scheme 3). Low-temperature
31P{1H} NMR monitoring of this reaction (2a, CD2Cl2, −80
°C to room temperature) revealed that at −80 °C complex 3a+
was the major solution species, though it was accompanied by
minor amounts of another product responsible for a doublet
resonance at 48.3 ppm (1JPRh = 160 Hz). The latter is probably
converted into the pseudo-allylic species 3a+ above −30 °C.
Since as explained above effecting the reactions in the presence
of NCMe furnished adducts 3·NCMe+, it is evident that the
weakly coordinating CH2Cl2 is unable to change the
coordination of the benzylic terminus from η3 to η1. It is also
apparent that an equilibrium is established between complexes
3+ and 3·L+ (Scheme 4). The former feature a characteristic
31P{1H} NMR resonance in the vicinity of −15 ppm (−4.4
ppm for the PEtXyl2-derived complex 3e+), whereas the latter
resonate at markedly higher frequencies (near 40 ppm; for 3e·
Scheme 4. Equilibria between Complexes 3+ and 3·L+ in the
Presence of the Lewis Base
C
DOI: 10.1021/acs.organomet.7b00688
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
CO+ at 54.2 ppm). Thus, equilibria like that represented in
Scheme 4 can be readily analyzed by 31P{1H} NMR
spectroscopy. Poor σ donors such as NCMe, MeOH, EtOH,
and Me2C(O) (neat solvents) formed adducts 3·L+ of fleeting
existence, identifiable by resonances in the region ca. 38−50
ppm that reverted to the corresponding base-free cations 3+
upon solvent removal by action of vacuum. Predictably,
stronger donors (CO, CNXyl, NH3, PMe3) originated stable
3·L+ adducts, while 3a·C2H4+ was found to be stable as a solid,
although it lost readily C2H4 in solution.
In contrast to homologous iridium compounds that, as briefly
noted, could not be isolated as single crystals, samples of several
rhodium complexes 3+ suitable for X-ray studies were procured.
Figure 2 depicts the molecular structure of complex 3a+, while
structural data for complexes 3d+−3f+ can be found in Figure
S4 in the Supporting Information. For comparative purposes
the structure of the neutral chloride 2a has been included in
Figure 2
Figure 3. Molecular graph of complexes 3a+ (left) and A+ (right)
showing bond critical points (bcp) and bond paths connecting the
rhodium atom and the cyclometalated phosphine. Relevant bcps and
bond paths are highlighted with arrows.
Differences in the spatial distribution and metal coordination
of the cyclometalated PMeXyl2 moiety of complexes 2a and 3a+
are evident at first sight (Figure 2). In contrast to compound
2a, where only the terminal P and C atoms of the chelating
ligand are bound to rhodium, in 3a+ the two aromatic ring
carbon atoms of the benzylic unit are also metal-bonded, giving
rise to an η3-benzylic end that complements P binding. The
aromatic ring that is part of the metallacyclic structure of 2a
forms an angle of ca. 64.4° with the η5-C5Me5 ring, while the
dihedral angle between the free Xyl substituent and the C5Me5
is ca. 20.6° and is almost perpendicular to the other (∼77.5°).
In contrast, the P and the benzylic carbon atoms of cation 3a+
(P1, C17, C11, and C12) form an almost planar arrangement,
practically parallel to the C5Me5 ring (dihedral angle 7.3°).
Consequently, this distribution of the donor atoms leads to a
mixed-sandwich structure, somehow related to that of the
decamethylrhodocenium complex.14 In fact, the mean Rh−Cring
distance in the latter complex of 2.180(5) Å is comparable to
the Rh−C5Me4R mean distances found for complexes 3+ (e.g.,
2.194(5) Å in 3a+). Although the Rh−P distances in complexes
3+ are similar (2.257(1)−2.269(7) Å) and do not diverge
significantly from those in compounds 2, the two aromatic
carbon atoms of the metalated moiety in complexes 3+
approach rhodium to well within bonding distances, somewhat
less for the adjacent carbon atom (C17 in Figure 2, 2.254(4) Å)
but significantly more for the central benzylic carbon (C12 in
Figure 3, 2.200(5) Å). The fluctuation experienced by the
metalated unit upon conversion of 2 into 3+ causes the free Xyl
radical to dispose itself nearly perpendicular to the η4-
P,C,C′,C′′ and C5Me5 planes, pertinent angles in 3a+ being
87.4 and 85.4°, respectively. As can be seen in Figures 2 and S4,
one of the Me groups of the free Xyl ring (C26 in Figure 2;
C26, C28, and C30 for 3d+−3f+, respectively, in Figure S4) and
the corresponding ipso carbon atom hover over the open space
of the wedge created by the Rh−P and Rh−Cterminal bonds
within the η4-(C∧P) ligand. These Rh···Me contacts are far
from bonding (ca. 3.876(5) Å for 3a+); nevertheless, the
existence of this ring permits attainment of a kinetically facile
equilibrium with the higher energy species A+ (see computa-
tional discussion next) that exhibits a δ CH3⇀Rh agostic
interaction15 and κ2 coordination of the metalated xylyl ring.
Despite lying largely on the side of complexes 3+, this
equilibrium appears to facilitate the observed η4-phosphine
structure and above all the H2 and hydrosilane activation
reactions (vide infra).
Computational studies of the electronic structures of these
complexes were undertaken. Analysis of the calculated electron
density of 3a+ (DFT-optimized geometry in the gas phase, see
Computational Details) was carried out within the quantum
theory of atoms in molecules formalism.16 Accordingly, three
bond critical points (bcp) and bond paths connect Rh1 and the
P1, C17, and C12 atoms (no bcp was located between any of
the hydrogen atoms on C26 and the rhodium atom; see Figure
3) and account for the linkage between the Rh atom and the
cyclometalated phosphine. The Laplacian of the electron
density (∇2ρb) is positive at these points, which is diagnostic
of closed-shell interactions. In addition, the values of the local
potential (Vb) and kinetic (Gb) energies and their ratios (1 < |
Vb|/Gb < 2) are consistent with metal−ligand interactions
(Table 1).17 Interestingly, DFT calculations located a second
Table 1. Selected Data of the Electron Density of 3a+ and A+
at Relevant Bond Critical Points of the Rh Linkage
a
b
b
c
bcp connecting atoms
ρ
Gb
Vb
∇2ρ
3a+
A+
Rh1−P1
0.0910
0.0692
0.0961
0.0973
0.1047
0.0485
0.0661
0.0634
0.0734
0.0670
0.0674
−0.0977
−0.0778
−0.1039
−0.1012
−0.1036
−0.0589
0.1383
0.1962
0.1714
0.1314
0.1242
0.1986
Rh1−C17
Rh1−C11
Rh1−P1
Rh1−C11
C26−H⇀Rh
0.0543
c
a
b
In units of e bohr−3. In units of hartree. In units of e bohr−5.
minimum for the species A+ (Figures 3 and 4), 5 kcal mol−1 less
stable than 3a+ (ΔE in dichloromethane), which features κC,κP
bidentate binding of the cyclometalated phosphine, as indicated
by two bcps and bond paths connecting the rhodium and the
P1 and C11 atoms (and the absence of bcps between Rh1 and
C12 or C17). The loss of the pseudo-allylic interaction between
Rh and the phosphine in A+ is partially offset by the
establishment of a δ-agostic interaction15 between the rhodium
atom and one of the methyl substituents of the pendant xylyl
group of the phosphine, which is demonstrated by a new bcp
connecting Rh1 and one of the hydrogen atoms on C26
(Figure 3). The role of A+ in the reactivity of 3a+ will be
discussed below. Localized molecular orbital analysis (NBO)18
of 3a+ and A+ offers a complementary description of the
bonding around the rhodium atom. In 3a+, a donor−acceptor
D
DOI: 10.1021/acs.organomet.7b00688
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
quite different chemical shifts, specifically 3.06 (Hα) and 1.34
ppm (Hβ). The former appears as a doublet of triplets (2JHH
=
2
3
4.5 Hz, JHRh = JHP = 1.5 Hz) and the latter as a doublet of
3
doublets (2JHRh = 1.5 Hz, JHP = 10.0 Hz). The associated 13C
signal is a doublet of doublets centered at 42.1 ppm (1JCRh = 15,
2JCP = 2 Hz) and exhibits a one-bond 13C−1H coupling
constant of 160 Hz. It is also pertinent to mention that the
aromatic carbon atoms participating in the η3-benzylic
interaction (C1 and C2 in Figure S1 in the Supporting
Information) are considerably shielded in comparison with the
corresponding signals in the free Xyl substituent, for they
resonate at 104.3 and 77.6 ppm, respectively (see Figure S1)
while, for instance, the ipso carbon atom C5 appears at 120.8
ppm (vs the 77.6 ppm value found for C2).
The 1H NMR spectrum of 3a+ contains several broad signals
that denote solution dynamic behavior. Variable-temperature
NMR studies (10−75 °C) revealed that, in addition to rotation
of the free Xyl ring around the P−C bond exchanging Meβ and
Meγ in Figure 5, the metalated and the noncoordinated xylyl
units commute their roles: that is, upon metalation of the latter
the former becomes disengaged. This dynamic behavior is
similar to that disclosed for an iridium complex analogue.5a 2D
EXSY NMR experiments performed in the interval 25−45 °C
showed transfer of magnetization between the methyl groups
Meβ and Meγ of the nonmetalated ring as well as between those
and Meα and Rh−CH2 groups, although at a slower rate. We
were able to determine the first-order rate constant for both
exchange processes as 2.1 and 1.2 × 10−2 s−1, corresponding to
Figure 4. Structural representation for some relevant computed
intermediates.
interaction is established between a π orbital localized on the
C17 and C12 atoms and an empty d orbital of the rhodium
atom. This interaction is absent in A+, but in turn the new δ-
agostic interaction arises from a donor−acceptor interaction
between a σ C−H bond and an empty σ* Rh−C(Cp).
Additional data on the AIM and NBO analyses can be found
inFigures S11 and 12 and Table S19 in the Supporting
Information.
ΔG298 K values of 18 and 21 kcal mol−1, respectively. These
⧧
barriers are significantly higher than those measured for the
iridium counterpart (ΔG⧧ = 13 and 17.3 kcal mol−1,
respectively),5a in accordance with an exchange implicating, in
both cases, a doubly metalated M(V) hydride intermediate.19
Rather surprisingly, DFT calculations predict that rotation of
the nonmetalated Xyl ring of 3a+ has a barrier that exceeds 40
kcal mol−1. This result can be reconciled with the experimental
data if it is assumed that 3a+ is in equilibrium with the
aforementioned intermediate A+ (ΔE = 5 kcal mol−1). The
necessary coordination change of the cyclometalated phosphine
from η4 to the more common κC,κP bidentate binding has a
barrier (ΔE⧧ in dichloromethane) of 11.8 kcal mol−1 (Figure
6). In the analogous iridium system,5a both the barrier and
relative energy are significantly lower (6.4 and 0.9 kcal mol−1,
respectively). Now, rotation of the Xyl ring has an energy
barrier (see Figure S14 in the Supporting Information) of ca. 10
kcal mol−1 from A+, 15 kcal mol−1 overall from 3a+, which is in
good agreement with the experiments. Calculations also reveal
that the higher energy (23.8 kcal mol−1) Rh(V) intermediate B+
resulting from oxidative cleavage of the agostic C−H⇀Rh
bond (Figures 4 and 6) makes possible the exchange of the two
fragments after a barrier of 25.5 kcal mol−1 is overcome, once
again the barrier being significantly higher than for iridium
(17.3 kcal mol−1).5a These results are in agreement with the
lower accessibility of the M(V) oxidation state for rhodium,
although some stable Rh(V) complexes are known.19 Recently,
an unusual Rh(IV) coordination compound was reported by
Crabtree and co-workers.20 The computed energy barrier is in
good agreement with the NMR measurements described above.
Reactions of Complexes 3a+−3e+ with Dihydrogen
and Hydrosilanes. Catalytic C−H/C−D and Si−H/Si−D
Exchanges and Hydrosilylation Reactions. Studying the
reactivity of the η4 complexes 3+ toward dihydrogen and
hydrosilanes, as substrates that possess only σ-bonding
NMR data for complexes 3+ are in full agreement with the
solid-state structure presented in the preceding paragraphs and
highlight in addition an interesting dynamic behavior. Using as
a reference the parent complex 3a+, the 31P{1H} NMR
spectrum contains a shielded doublet at −14.7 ppm (1JPRh
=
138 Hz). The 1H NMR spectrum features methyl resonances at
4
1.74 (C5Me5, d, JHP = 2.5 Hz), 2.66, 2.59, and 2.12 ppm, the
last three appearing as singlets and being attributable to the
xylyl group substituents (Meα, Meβ, and Meγ in Figure 5). The
Rh−CH2 protons of the pseudo-allyl moiety resonate with
1
Figure 5. H NMR spectrum and labeling scheme for complex 3a+
1
utilized for assignment of H and 13C NMR signals.
E
DOI: 10.1021/acs.organomet.7b00688
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
1H resonances located at around −10 and 0 ppm, which
undergo mutual exchange. The more shielded of the two,
though broad, becomes sufficiently resolved as a doublet of
doublets, with 1JHRh = 18.5 and 2JHP = 37 Hz, and has the spin−
lattice relaxation time T1 = 300 ms (data for 4a+), clearly
corresponding to a Rh−H bond.21 The exchanging, also broad,
singlet in the proximity of 0 ppm has relative intensity
corresponding to 3H atoms and is assigned to a Rh↽H3C−
agostic interaction.15 The exchange process likely occurs by
oxidative cleavage of the agostic C−H bond, as shall be
described below.
A final comment in regard to the H2-promoted equilibria of
Scheme 5 pertains to the relative stabilities of complexes 3+ and
4+. As briefly noted, the putative agostic hydride 4c+ of the
OMe-substituted aryl phosphine could not be detected.
Changing the p-OMe substituent to a methyl group led to a
4b+:3b+ ratio of roughly 2:98, whereas PMeXyl2 (i.e., H at the
4-position of the rings) and the F-containing PMeAr2
phosphine gave rise to improved 4a+:3a+ and 4d+:3d+ ratios
of 10:90 and 17:83, respectively. For the PEtXyl2 derivatives
4e+ and 3e+, the observed ratio was ca. 5:95. Although the
differences are relatively small, they correlate with the ν(CO)
changes discussed earlier for the carbonyl complexes 3·CO+
and reveal that the relative stability of the cationic agostic
hydride complexes increases in the order OMe < Me < H < F,
that is, with a decrease in the overall donor capacity of the
PMeAr2 phosphine utilized and a concomitant increase in the
Lewis acidity of the 16-electron cationic Rh(III) center.
In addition, no apparent reactions were observed when the
above complexes 3+ were treated with SiEt3H and other
hydrosilanes. Although reactivity similar to that represented in
Scheme 5 for H2 was reasonably expected, cationic agostic silyl
complexes analogous to hydrides 4+ did not form in detectable
concentrations. DFT calculations on the reactions with H2 and
SiMe3H (used as a model silane) were developed and suggested
the key intermediacy of both σ-H2 and HSiMe3 complexes in
the corresponding reactions (Figure 7). Thus, syn and anti
Figure 6. Calculated energy profile accounting for the dynamic
behavior of 3a+ in solution and DFT-optimized geometry of the
Rh(V) intermediate B+. Gibbs energy and potential energy (in
parentheses) data are given in kcal mol−1.
electrons, was an attractive target.21 The iridium complex
analogous to 3a+ was found to react with H2 under ambient
conditions with quantitative formation of a bis(hydride)-
iridium(V) species. However, the reaction was reversible, H2
removal under vacuum yielding back the starting Ir(III)
derivative.5a In contrast to this observation, no reaction was
detected when complexes 3a+−3e+ were stirred at room
temperature under 1 bar of H2. Nevertheless, as briefly reported
for 3a+,4 upon cooling at −90 °C minor quantities of the new
cationic complex 4a+, in equilibrium with 3a+, were detected by
NMR spectroscopy (Scheme 5). All complexes 3+ studied
Scheme 5. Reaction of Complexes 3a+, 3b+, 3d+, and 3e+ with
H2
+
+
diasteromers of 3a·H2 and 3a·HSiMe3 were located as
minima on the potential energy surface; however, their
+
formation from 3a+ is endergonic in all cases. syn-3a·H2 lies
7.2 kcal mol−1 above 3a+ + dihydrogen (ΔG in dichloro-
methane) and can be described as an η2-σ-H2 complex.
Similarly, syn-3a·HSiMe3 lies 4.6 kcal mol−1 above the
+
separated fragments, but then it is best described as an η1-
silane complex (dRh−H = 1.81 Å, dRh−Si = 3.27 Å).22 Rh(III)
hydride 4a+ was also modeled as a relatively stable species and
is only 1.9 kcal mol−1 above 3a+ + H2 at 295 K, which agrees
with 4+ cations being detected by NMR at low temperature.
These species interconvert through an energy barrier of 8.1 kcal
mol−1 from 3a·H2 . The corresponding transition state has
+
Rh(V) character, as indicated by the Rh−H distance of 1.57 Å
(Figure 7). Nevertheless, at variance with the related Ir
system,5a no Rh(V) bis(hydride) intermediate was located in
+
this case. 3a·HSiMe3 also interconverts with the correspond-
ing agostic silyl complex 5a+ in an analogous manner. In this
case, formation of 5a+ is endergonic by 6.2 kcal mol−1, which
justifies why the related species were never detected
experimentally. It should be emphasized that both of the
reversible exchange processes that have just been described can
place hydrogen atoms from the H2 and silane molecules onto
the phosphine ligand of the metal complex.
behaved similarly except for the OMe-substituted 3c+ (see
Figure 1), for which no observable product was detected. The
new complexes 4+ were identified as cationic Rh(III) hydrides
featuring a δ-agostic interaction15 involving one of the Me
substituents of the phosphine aryl rings. Although compre-
hensive NMR data for cations 4+ are provided in the
Supporting Information, it is pertinent to mention that they
exhibit a 31P{1H} doublet resonance notably shifted to higher
frequencies (δ 27−30, 1JPRh ≈ 140 Hz), as well as two shielded
To complement the anterior computational findings, it
should be mentioned that exposure of solutions of complexes
F
DOI: 10.1021/acs.organomet.7b00688
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Similarly, conversion for Si(SiMe3)3H was lower (ca. 30%) in
all cases investigated. Reactions were performed at room
temperature for 16 h under otherwise analogous experimental
conditions.
In noticeable contrast with these results, the hydro-
silylation23,24 of the aromatic ketones ArC(O)Me (Ar = Ph,
1-naphthyl) was found to be markedly dependent on the
phosphine substituents (Scheme 7 and Table 2). Thus, the p-
Scheme 7. Hydrosilylation of Aromatic Ketones Catalyzed
by Complexes 3b+−3e+
a
Table 2. Hydrosililation Reaction
entry
catalyst
Ar
Ph
T (°C)
t (h)
conversn (%)
1
2
3a+
3b+
3c+
3d+
3e+
3a+
3b+
3c+
3d+
3e+
25
25
25
25
25
50
50
50
50
50
14
14
14
14
14
24
24
24
24
24
≥99
30
0
Ph
Figure 7. Calculated ΔG profiles in dichloromethane (kcal mol−1) for
the reactions of 3a+ with H2 (upper trace) and Me3SiH (lower trace).
The first transition state in each profile (insets depict their DFT-
optimized geometries) connects the σ-3a+·L complex and the
corresponding hydride and silyl cations 4a+ and 5a+. The second
transition state corresponds to rotation of the Rh↽H3C− methyl,
which accounts for H/D exchange (see the Supporting Information for
details). Data in parentheses are ΔE values in dichloromethane.
3
Ph
4
Ph
≥99
0
5
Ph
6
1-Naph
1-Naph
1-Naph
1-Naph
1-Naph
10
0
7
8
0
9
15
0
10
a
Conditions: 0.1−0.5 mmol of substrate, 2.2 equiv of SiEt3H, CD2Cl2
3a+−3e+ to D2 led to fast H/D exchange at all the benzylic CH3
and metalated CH2 sites. These observations allowed
developing a catalytic synthesis of deuterated silanes4 and its
exploitation for several-gram-scale preparations of SiEt3D,
SiMe2PhD, and SiPh2D2, using D2 (0.5 bar) as the source of
deuterium10b and the parent complex 3a+ as the catalyst at a
loading of ca. 0.01 mol %. Tritiated silanes were also produced,4
and since 3a+ catalyzed with high efficacy the hydrosilylation of
carbonyl compounds10a and of the CN bonds of α,β-
unsaturated nitriles,10b a highly convenient one-flask, two-step
catalytic procedure for deuterio- and tritiosilylation of the
mentioned substrates was developed.10
Considering the similar chemical behavior of all complexes
3+, we have completed these studies with the investigation of
the catalytic activity of complexes 3b+−3e+ in related work
involving D2 and SiEt3H and Si(SiMe3)3H as the silanes
(Scheme 6). No significant differences were found among the
new complexes and the original catalyst 3a+, so that for instance
quantitative Si−H/Si-D exchanges were disclosed for SiEt3H
(≥99%) in CD2Cl2, at 50 °C after 5 h, under 0.5 bar of D2,
regardless of the nature of the original phosphine ligand.
(0.5 mL).
methyl-substituted precursor 3b+ had significantly diminished
activity in comparison to 3a+, whereas both the p-methoxy
analogue 3c+ and the PEtXyl2 derivative 3e+ were inactive
under similar conditions. Somewhat rewardingly, the fluoro-
containing complex 3d+ exhibited activity comparable if not
superior to that of 3a+ (Table 2). To gain mechanistic insights,
some additional experiments were performed.
In parallel tests, we monitored by low-temperature 31P{1H}
NMR spectroscopy (CD2Cl2, −80 °C) the interaction of
complex 3a+ with a 100-fold excess of SiEt3H and acetone. As
anticipated, no reaction was detected for the silane while a
70:30 mixture of 3a+ and the corresponding acetone adduct, 3a·
S+, was obtained in the second instance. Clearly, under our
catalytic reaction conditions (Scheme 7), 3a+ is the catalyst
resting state and is in equilibrium with lower concentrations of
the ketone adduct. This is in dissimilarity with findings by
Brookhart and co-workers on related hydrosilylation reactions
also catalyzed by a cationic Ir(III) pincer complex, where the
catalyst resting state was a ketone adduct, in equilibrium with
low concentrations of an η1-σ-silane.22,24b An ionic mechanism
implying R3Si+ transfer to the ketone to form the oxocarbenium
ion R3SiOCR2′+, subsequently reduced by a neutral Ir(III)
dihydride, was proposed, in close analogy with the ionic
mechanism previously postulated by Piers when B(C6F5)3 was
employed as the catalyst.25
Scheme 6. Deuteration of Hydrosilanes Catalized by
Complexes 3b+−3e+
Kinetic studies on the Ir(III) system also disclosed that the
mechanism was zero order in ketone and first order in silane.
Once more at variance with these observations, our Rh(III)
catalytic reaction featured first-order dependence on both
SiEt3H and the ketone (Figure 8). In addition, for
G
DOI: 10.1021/acs.organomet.7b00688
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Figure 8. Kinetic studies for the reaction of acetophenone and triethylsilane at 25 °C in CD2Cl2 catalyzed by 1+: (A) pseudo-first-order consumption
of silane (conditions: [acetophenone]0 = 1.85 M; [HSiEt3]0 = 0.18 M; [1+] = 1.4 × 10−4 M); (B) pseudo-first-order consumption of acetophenone
(conditions: [acetophenone]0 = 0.09 M; [HSiEt3]0 = 2.14 M; [1+] = 1.4 × 10−4 M); (C) plot of the observed rate constant, kobs, of acetophenone
consumption vs silane concentration (excess silane); (D) plot of the initial rate, Vi, of acetophenone consumption vs ketone concentration (<10%
conversion, excess silane).
spectra were referenced to external H3PO4. Spectral assignments were
made by routine one- and two-dimensional NMR experiments where
acetophenone hydrosilylation a kinetic isotope effect (KIE) of
kH/kD = 1.9( 0.1) was measured employing SiEt3H and
SiEt3D.
On the basis of these results and taking into consideration
prior literature reports,26,27 along with knowledge emanated
from the experimental and computational studies already
discussed, acetone hydrosilylation by SiEt3H could follow a
variant of the Chalk−Harrod mechanism like that presented in
Scheme S1 in the the Supporting Information.
t
appropriate. Metal complexes Zn(η5-C5Me5)2, Zn(C5Me4 Bu)2, Zn-
[C5Me4(3,5-tBu2C6H3)]2, and [RhCl(C2H4)]2 as well as NaBArF
(BArF = B[(C6H3-3,5-(CF3)2)4])28 and the phosphines of the type
PMe(Xyl′)2 and PEt(Xyl)2 were prepared according to literature
procedures.11
Synthesis and Characterization of Complex 1a. A solution of
the corresponding phosphine (0.5 mmol) in 3 mL of THF was added,
at −40 °C, to a solution of [RhCl(C2H4)2]2 (0.25 mmol) in 3 mL of
THF. The reaction mixture was stirred for 3 h at this temperature.
Then, a solution of Zn(η5-C5Me5)2 (0.25 mmol) in 2 mL of THF was
added and the mixture was stirred for 5 h at −25 °C. At this
temperature the solvent was removed under vacuum, the residue was
extracted with diethyl ether, and the resulting solution was evaporated
to dryness. The dark yellow solid obtained contained a mixture of
complexes 1a and 2a. To the solid mixture was added a NaBH4 (0.26
mmol) solution in ethanol/THF (3/1 mixture, 5 mL), and this
mixture was stirred at room temperature for 10 min. Analytically pure
samples of complex 1a could not be obtained. Thus, characterization
of 1a was carried out in solution by NMR spectroscopy of the crude
SUMMARY AND CONCLUSIONS
■
In summary, this work shows that cationic Rh(III) complexes
of type [(η5-C5Me4R)Rh(C∧P)]+ (3+) that contain a cyclo-
metalated bis(aryl)phosphine PR′Ar2, featuring 2,6-Me2 sub-
stitution at the P-bonded aryl groups, strongly favor η4
coordination of the phosphine through the phosphorus atom
and the adjacent metalated pseudo-allylic unit. The η4 structure
is labile and is responsible for the facile intramolecular exchange
between the metalated and the free phosphine Ar substituents,
as well as for the interesting H−H and Si−H bond activation
reactivity of the complexes. Changing the nature of the 4-
substituent R′′ of the PMeAr2 phosphines (Ar = 2,6-Me2-4-R′′-
C6H2 and R′′ = H, Me, OMe, F) has a small, albeit detectable,
effect on the electronic properties of the cationic Rh(III) center
and also a scant influence on the chemical reactivity.
1
reaction product. H NMR (400 MHz, C6D6, 25 °C): δ 7.45 (d, 1 H,
5
5
Ha), 7.06 (td, 1 H, JHP = 2.8 Hz, Hb), 6.94 (td, 1 H, JHP = 1.6 Hz,
2
He), 6.84, 6.73, 6.71 (m, 1 H each, Hd, Hf, Hc), 3.59 (d, 1 H, JHH
=
13.6, RhCHH), 3.00 (dd, 1 H, 2JHRh = 4.2 Hz, RhCHH), 2.36, 1.63 (s,
3 H each, Meβ, Meγ), 1.93 (d, 3 H, 2JHP = 9.0 Hz, PMe), 1.89 (s, 3 H,
Meα), 1.69 (d, 15 H, 4JHP = 1.6 Hz, C5Me5), −13.77 (dd, 1 H, 1JRhH
=
2
3
46.0, JHP = 37.0 Hz, RhH). All aromatic JHH couplings were ca. 7.5
Hz. 13C{1H} NMR (100 MHz, C6D6, 25 °C): δ 158.6 (d, JCP = 33
2
Hz, C1), 141.9 (dd, 1JCP = 54, 2JCRh = 4 Hz, C2), 141.9, 139.0 (d, 2JCP
=
EXPERIMENTAL SECTION
General Considerations. All operations were performed under an
argon atmosphere using standard Schlenk techniques, employing dry
solvents and glassware. Microanalyses were performed by the
■
8 Hz, C4, C6), 138.8 (C3), 134.0 (d, 1JCP = 27 Hz, C5), 129.9, 129.8 (d,
3JCP = 7 Hz, CHd, CHf), 129.2 (CHb), 128.7 (CHe), 127.3 (d, JCP
=
3
3
1
2
16 Hz, CHa), 127.1 (d, JCP = 7 Hz, CHc), 96.9 (dd, JCRh = JCP = 3
Hz, C5Me5), 21.7 (d, 1JCP = 38 Hz, PMe), 25.4, 22.4 (d, 3JCP = 4, 3JCP
=
́
Microanalytical Service of the Instituto de Investigaciones Quimicas
1
2
9 Hz, Meβ, Meγ), 22.4 (dd, JCRh = 26, JCP = 4 Hz, RhCH2), 20.9 (d,
(Sevilla, Spain). The NMR instruments were Bruker DRX-500, DRX-
400, and DRX-300 spectrometers. Spectra were referenced to external
SiMe4 (δ 0 ppm) using the residual proton solvent peaks as internal
standards (1H NMR experiments) or the characteristic resonances of
the solvent nuclei (13C NMR experiments), while 31P{1H} NMR
3JCP = 3 Hz, Meα), 10.6 (C5Me5). 31P{1H} NMR (160 MHz, C6D6, 25
1
°C): δ 49.9 (d, JPRh = 156 Hz).
Synthesis of Complexes 2. A solution of the corresponding
phosphine (0.5 mmol) in 2 mL of THF was added, at −40 °C, to a
H
DOI: 10.1021/acs.organomet.7b00688
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
solution of [RhCl(C2H4)2]2 (0.25 mmol) in 3 mL of THF. The
reaction mixture was stirred for 3 h at this temperature. Then, a
solution of the corresponding Zn(η5-C5Me4R)2 (0.25 mmol) in 1 mL
of THF was added and the mixture was stirred for 5 h at −25 °C. After
the mixture was warmed to room temperature, the solvent was
removed under vacuum, the residue was extracted with diethyl ether,
and the resulting solution was evaporated to dryness. The solid
obtained was dissolved in 5 mL of CHCl3 and stirred for 3 h at room
temperature. The solvent was removed under vacuum and the crude
product washed with pentane to yield complexes 2a−g as red-orange
solids in 70−80% yields. These complexes can be recrystallized from
diethyl ether. Data for complex 2b are as follows. Anal. Calcd for
C29H39ClPRh: C, 62.5; H, 7.1. Found: C, 62.7; H, 6.9. 1H NMR (500
MHz, 25 °C, CDCl3): δ 7.07 (s, 1 H, Ha), 6.93 (s, 1 H, Hc/d), 6.70 (s,
4JHP = 2.7 Hz, C5Me5), 1.29 (ddd, 1 H, 3JHP = 13.9, 2JHH = 4.0, 2JHRh
=
0.7 Hz, RhCHβ). 13C{1H} NMR (100 MHz, 25 °C, CDCl3): δ 144.1
(C4), 143.2 (C8), 141.9, 140.9 (C5, C7), 137.3 (C3), 134.7 (CHb),
130.4, 130.3 (CHc, CHd), 125.1 (d, 3JCP = 8 Hz, CHa), 116.7 (d, 1JCP
=
58 Hz, C6), 104.6 (dd, 1JCRh = 14, 2JCP = 4 Hz, C1), 98.8 (dd, 1JCRh = 6,
1
2
2JCP = 2 Hz, C5Me5), 74.4 (d, JCP = 25 Hz, C2), 41.5 (dd, JCP = 15,
1JCRh = 2 Hz, RhCH2), 22.5 (Meα), 21.7, 21.5 (Meγ/ε), 21.1 (Meβ),
20.9 (Meδ), 13.0 (d, 1JCP = 33 Hz, PMe), 8.9 (C5Me5). 31P{1H} NMR
1
(160 MHz, 25 °C, CDCl3): δ −15.7 (d, JPRh = 138 Hz).
X-ray Structure Analysis of Complexes 2a, 2b, 2g, 3a+, and
3d+−3f+. A summary of the crystallographic data and structure
refinement of these new crystalline compounds is given in the
Supporting Information. A single crystal of suitable size, coated with
dry perfluoropolyether (Fomblin Y H-VAC 140/13), was mounted on
a glass fiber and fixed under a cold nitrogen stream (T = 100(2) K) to
the goniometer head. Data collection was performed on a Bruker-
Nonius X8APEX-II CCD diffractometer, using monochromatic
radiation (λ(Mo Kα1) = 0.71073 Å) by means of ω and φ scans
with a width of 0.5°. The data were reduced (SAINT)30 and corrected
for Lorentz−polarization effects and absorption by multiscan methods
applied by SADABS.31 The structures were solved by direct methods
(SIR-2002)32 and refined against all F2 data by full-matrix least-squares
techniques (SHELXTL-6.12).33 All of the non-hydrogen atoms were
refined with anisotropic displacement parameters. The hydrogen
atoms were included from calculated positions and refined riding on
their respective carbon atoms with isotropic displacement parameters.
Computational Details. Geometry optimizations were carried out
at the DFT level with Gaussian 0929 using the meta-GGA functional
M06.30 The C, H, and P atoms were described with the 6-31G(d,p)
basis set,31 and the Rh atoms were represented by the Stuttgart/
Dresden effective core potential and the associated basis set (SDD).32
Optimizations were made in the gas phase without restrictions. The
stationary points of the potential energy surface and their nature as
minima or saddle points (TS) were characterized by vibrational
analysis, which also gave gas-phase enthalpies (H), entropies (S), and
Gibbs energies (G). The minima connected by a given transition state
were determined by intrinsic reaction coordinate (IRC) calculations or
by perturbing the transition states along the TS coordinate and
optimizing to the nearest minimum. The solvent effects (dichloro-
methane) were modeled with the SMD continuum model by single-
point calculations on gas-phase-optimized geometries.33 Atoms in
molecules analysis of the electron density was performed with the
Multiwfn program34 on wave functions calculated for species
reoptimized at the DFT, M06, and 6-311g(d,p)35 + SDD level. The
same level of theory was used to construct the natural bonding
orbitals, which were analyzed with the NBO6.0 suite.36
1 H, Hd/c), 6.63 (s, 1 H, Hb), 3.53 (dt, 1 H, 2JHH = 12.9, 2JHRh = 3JHP
=
2
3
3.0 Hz, RhCHH), 3.40 (dd, 1 H, JHH = 12.9, JHRh = 2.5 Hz,
RhCHH), 2.56 (s, 3 H, Meγ/ε), 2.26 (s, 3 H, Meδ), 2.56 (s, 3 H, Meβ),
2
2.17 (d, 3 H, JHP = 10.4 Hz, PMe), 1.92 (s, 3 H, Meα), 1.47 (s, 3 H,
4
Meε/γ), 1.41 (d, 15 H, JHP = 2.3 Hz, C5Me5). 13C{1H} NMR (125
MHz, 25 °C, CDCl3): δ 157.0 (d, 2JCP = 33 Hz, C1), 142.0 (d, 2JCP = 9
2
4
Hz, C6/8), 140.0 (d, JCP = 8 Hz, C6/8), 139.6, 139.4 (d, JCP = 2 Hz,
1
2
C4, C7), 138.9 (C3), 135.6 (dd, JCP = 55, JCRh = 2 Hz, C2), 131.1,
130.8 (d, 3JCP = 8 Hz, CHc, CHd), 128.8 (d, 3JCP = 7 Hz, CHb), 128.7
1
3
1
(d, JCP = 34 Hz, C5), 127.7 (d, JCP = 17 Hz, CHa), 98.2 (t, JCRh
=
2JCP = 4 Hz, C5Me5), 34.2 (dd, JCRh = 24, 2JCP = 7 Hz, RhCH2), 25.9
1
(d, 3JCP = 5 Hz, Meγ/ε), 23.6 (d, 3JCP = 8 Hz, Meγ/ε), 21.2, 21.1 (Meβ,
3
1
Meδ), 20.6 (d, JCP = 3 Hz, Meα), 19.9 (d, JCP = 34 Hz, PMe), 8.8
(C5Me5). 31P{1H} NMR (200 MHz, 25 °C, CDCl3): δ 44.1 (d, 1JPRh
=
157 Hz).
Synthesis of Complexes 3·CO+. To a solid mixture of complexes
2a−2f (0.08 mmol) and NaBArF (0.08 mmol), placed in a thick-
walled ampule, was added 5 mL of CH2Cl2, and the reaction mixture
was stirred for 10 min at room temperature under 1.5 bar of CO. After
filtration, the solvent was evaporated under reduced pressure to obtain
orange (3c·CO+) and yellow powders (3a·CO+, 3b·CO+, 3d·CO+−3f·
CO+) in ca. 95% yield. These complexes can be recrystallized by slow
diffusion at −20 °C of pentane into a CH2Cl2 solution (2:1 by vol.).
Data for complex 3b·CO+ are as follows. IR (Nujol): 2050 cm−1. Anal.
Calcd for C62H51BF24OPRh: C, 52.7; H, 3.6. Found: C, 52.5; H, 3.7.
1H NMR (400 MHz, 25 °C, CD2Cl2): δ 7.20 (s, 1 H, Ha), 7.13 (s, 1 H,
2
2
Hc/d), 6.94 (s, 2 H, Hd/c, Hb), 3.53 (dd, 1 H, JHH = 12.5, JHRh = 2.0
2
2
Hz, RhCHH), 3.32 (dd, 1 H, JHH = 12.5, JHRh = 4.2 Hz, RhCHH),
2.52 (s, 3 H, Meγ/ε), 2.37 (s, 3 H, Meβ/δ), 2.36 (d, 3 H, 2JHP = 9.7 Hz,
4
PMe), 2.35 (s, 3 H, Meδ/β), 2.04 (s, 3 H, Meα), 1.71 (d, 15 H, JHP
=
2.8 Hz, C5Me5), 1.53 (s, 3 H, Meε/γ). 13C{1H} NMR (100 MHz, 25
1
2
°C, CD2Cl2): δ 188.7 (dd, JCRh = 73, JCP = 19 Hz, CO), 152.1 (d,
2JCP = 29 Hz, C1), 142.8, 142.6 (d, 4JCP = 3 Hz, C4, C7), 142.1 (d, 1JCP
= 10 Hz, C6/8), 140.3 (d, 2JCP = 8 Hz, C8/6), 139.8 (t, 2JCP = 3JCRh = 3
ASSOCIATED CONTENT
■
Hz, C3), 132.6 (dd, 1JCP = 61, 2JCRh = 3 Hz, C2), 131.8, 131.6 (d, 3JCP
=
S
* Supporting Information
3
3
9 Hz, CHc, CHd), 130.8 (d, JCP = 8 Hz, CHb), 126.8 (dd, JCP = 17,
3JCRh = 2 Hz, CHa), 121.8 (d, 1JCP = 42 Hz, C5), 106.1 (dd, 1JCRh = 4,
2JCP = 2 Hz, C5Me5), 28.4 (d, 1JCRh = 21 Hz, RhCH2), 25.5 (d, 3JCP = 6
Hz, Meγ/ε), 25.2 (d, 1JCP = 40 Hz, PMe), 23.2 (d, 3JCP = 8 Hz, Meε/γ),
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.organo-
met.7b00688.
20.6 (Meβ, Meδ), 20.1 (d, JCP = 4 Hz, Meα), 8.6 (C5Me5). 31P{1H}
3
Analytical and spectroscopic data for complexes 1·SiMe3,
+
+
+
2c−2g, 3a·C2H4 , 3a·CNXyl+, 3a·NH3 , 3a·PMe3 , and
3c+−3f+, 13C{1H} and 13C NMR spectrum of 3a+, low-
temperature reaction of 3+ with H2, spectroscopic data of
agostic hydrides 4+, general procedure for Si−H/Si−D
exchanges, general method for the hydrosilylation of C−
O multiple bonds, kinetic studies, X-ray structure analysis
1
NMR (160 MHz, 25 °C, CD2Cl2): δ 36.1 (d, JPRh = 126 Hz).
Synthesis of Complexes 3+. To a mixture of 2a−2f (0.15 mmol)
and NaBArF (0.15 mmol) was added 5 mL of CH2Cl2 under argon.
The reaction mixture was stirred for 10 min at room temperature, after
which the solution turned from orange to red. The resulting
suspension was filtered and the solvent evaporated under reduced
pressure to obtain compounds 3a+−3f+ as red solids in ca. 95% yield.
For further purification, the complexes can be recrystallized by slow
diffusion at −20 °C of pentane into a CH2Cl2 solution (2/1 v/v). Data
for complex 3b+ are as follows. Anal. Calcd for C61H51BF24PRh: C,
of 2a, 2b, 2g, 3a+, 3d+−3f+, 3a·CO+, 3a·NCCH3 , and
+
+
3a·C2H4 , and additional figures and tables with
computational details (PDF)
1
52.9; H, 3.7. Found: C, 52.6; H, 3.7. H NMR (400 MHz, 25 °C,
Cartesian coordinates of calculated structures (XYZ)
CDCl3): δ 7.34 (s, 1 H, Hb), 6.88, 6.76 (s, 1 H, Hc/d), 6.60 (s, 1 H,
Ha), 2.91 (dt, 1 H, 2JHH = 4.0, 2JHRh = 3JHP = 1.3 Hz, RhCHα), 2.51 (s,
3 H, Meγ/ε), 2.44 (s, 3 H, Meα), 2.36 (s, 3 H, Meβ), 2.20 (s, 3 H, Meδ),
2.11 (d, 3 H, 2JHP = 13.0 Hz, PMe), 1.97 (s, 3 H, Meε/γ), 1.61 (d, 15 H,
Accession Codes
CCDC 1566023−1566028, 1571395, and 1571397 contain the
supplementary crystallographic data for this paper. These data
I
DOI: 10.1021/acs.organomet.7b00688
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
(13) Goldman, A. S.; Krogh-Jespersen, K. J. Am. Chem. Soc. 1996,
118, 12159−12166.
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.
(14) Buchholz, D.; Zsolnai, L.; Huttner, G.; Astruc, D. J. Organomet.
Chem. 2000, 593−594, 494−496.
(15) (a) Brookhart, M.; Green, M. L. H. J. Organomet. Chem. 1983,
250, 395−408. (b) Brookhart, M.; Green, M. L. H.; Parkin, G. Proc.
Natl. Acad. Sci. U. S. A. 2007, 104, 6908−6914. (c) Green, J. C.; Green,
M. L. H.; Parkin, G. Chem. Commun. 2012, 48, 11481−11503.
(16) Bader, R. F. W. Atoms in Molecules: A Quantum Theory; Oxford
University Press: Oxford, UK, 1995.
(17) (a) Varadwaj, P. R.; Varadwaj, A.; Marques, H. M. J. Phys. Chem.
A 2011, 115, 5592−5601. (b) Varadwaj, P. R.; Cukrowski, I.; Marques,
H. M. J. Phys. Chem. A 2008, 112, 10657−10666.
(18) Weinhold, F. J. Comput. Chem. 2012, 33, 2363−2379.
(19) (a) Fernandez, M. J.; Bailey, P. M.; Bentz, P. O.; Ricci, J. S.;
Koetzle, T. F.; Maitlis, P. M. J. Am. Chem. Soc. 1984, 106, 5458−5463.
(b) Duckett, S. B.; Haddleton, D. M.; Jackson, S. A.; Perutz, R. N.;
Poliakoff, M.; Upmacis, R. K. Organometallics 1988, 7, 1526−1532.
(20) Sinha, S. B.; Shopov, D. Y.; Sharninghausen, L. S.; Vinyard, D. J.;
Mercado, B. Q.; Brudvig, G. W.; Crabtree, R. H. J. Am. Chem. Soc.
2015, 137, 15692−15695.
(21) (a) Crabtree, R. H. Angew. Chem., Int. Ed. Engl. 1993, 32, 789−
805. (b) Kubas, G. J. Metal Dihydrogen and Sigma-Bond Complexes:
Structure Theory and Reactivity; Kluwer Academic: New York, 2001.
(22) Yang, J.; White, P. S.; Schauer, C. K.; Brookhart, M. Angew.
Chem., Int. Ed. 2008, 47, 4141−4143.
(23) (a) Marciniec, B. Silicon Chem. 2002, 1, 155−174. (b) Roy, A. K.
Adv. Organomet. Chem. 2007, 55, 1−59. (c) Malacea, R.; Poli, R.;
Manoury, E. Coord. Chem. Rev. 2010, 254, 729−752. (d) Morris, R. H.
Chem. Soc. Rev. 2009, 38, 2282−2291. (e) Marciniec, B. Comprehensive
Handbook on Hydrosilylation; Pergamon: Oxford, U.K., 1992.
(24) (a) Calimano, E.; Tilley, T. D. J. Am. Chem. Soc. 2009, 131,
11161−11173. (b) Park, S.; Brookhart, M. Organometallics 2010, 29,
6057−6064. (c) Yang, J.; White, P. S.; Brookhart, M. J. Am. Chem. Soc.
2008, 130, 17509−17518. (d) Tondreau, A. M.; Lobkovsky, E.; Chirik,
P. J. Org. Lett. 2008, 10, 2789−2792. (e) Tondreau, A. M.; Atienza, C.
C. H.; Weller, K. J.; Nye, S. A.; Lewis, K. M.; Delis, J. G. P.; Chirik, P.
J. Science 2012, 335, 567−570. (f) Yang, J.; Tilley, T. D. Angew. Chem.,
Int. Ed. 2010, 49, 10186−10188. (g) Blackwell, J. M.; Morrison, D. J.;
Piers, W. E. Tetrahedron 2002, 58, 8247−8254. (h) Buchan, Z. A.;
Bader, S. J.; Montgomery, J. Angew. Chem., Int. Ed. 2009, 48, 4840−
4844. (i) Tran, B. L.; Pink, M.; Mindiola, D. J. Organometallics 2009,
28, 2234−2243.
(25) (a) Parks, D. J.; Piers, W. E. J. Am. Chem. Soc. 1996, 118, 9440−
9441. (b) Parks, D. J.; Piers, W. E.; Parvez, M.; Atencio, R.;
Zaworotko, M. J. Organometallics 1998, 17, 1369−1377. (c) Parks, D.
J.; Blackwell, J. M.; Piers, W. E. J. Org. Chem. 2000, 65, 3090−3098.
(26) (a) Ojima, I.; Kogure, T.; Kumagai, M.; Horiuchi, S.; Sato, T. J.
Organomet. Chem. 1976, 122, 83−97. (b) Duckett, S. B.; Perutz, R. N.
Organometallics 1992, 11, 90−98.
(27) Perutz, R. N.; Sabo-Etienne, S. Angew. Chem., Int. Ed. 2007, 46,
2578−2592.
(28) (a) Yakelis, N. A.; Bergman, R. G. Organometallics 2005, 24,
3579−3581. (b) Blom, R.; Boersma, J.; Budzelaar, P. H. M.; Fischer,
B.; Haaland, A.; Volden, H. V.; Weidlein, J. Acta Chem. Scand. 1986,
40a, 113−120. (c) Giordano, G.; Crabtree, R. H. Inorg. Synth. 1979,
19, 218−220.
(29) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,
B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H.
P.; Iz-maylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.;
Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin,
K. N.; Staroverov, V. N.; Keith, T.; Kobayashi, R.; Normand, J.;
Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.;
Cossi, M.; Rega, N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.;
Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.;
Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.;
AUTHOR INFORMATION
Corresponding Author
*E.C.: e-mail, guzman@us.es; fax, (+34)95-446-0165.
■
ORCID
Jesus Campos: 0000-0002-5155-1262
́
Joaquín Lopez-Serrano: 0000-0003-3999-0155
́
́
Eleuterio Alvarez: 0000-0002-8378-8459
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
Financial support (FEDER support) from the Spanish
Ministerio de Ciencia e Innovacion
42501-P) is gratefully acknowledged. The use of computational
facilities of the supercomputing center of Galicia, CESGA, and
■
́
(Project No. CTQ2013-
́
of the Centro de Servicios de Informatica y Redes de
Comunicaciones (CSIRC) of the Universidad de Granada
(UGR) is also acknowledged. M.F.E. thanks the Universidad de
Sevilla for a research grant.
REFERENCES
■
(1) (a) Song, G.; Wang, F.; Li, X. Chem. Soc. Rev. 2012, 41, 3651−
3678. (b) Albrecht, M. Chem. Rev. 2010, 110, 576−623. (c) Kuwata,
S.; Ikariya, T. Dalton Trans. 2010, 39, 2984−2992. (d) Liu, Z.; Sadler,
P. J. Acc. Chem. Res. 2014, 47, 1174−1185. (e) Bergman, R. G. Nature
2007, 446, 391−393.
(2) (a) Sharninghausen, L. S.; Campos, J.; Manas, M. G.; Crabtree, R.
H. Nat. Commun. 2014, 5, 5084. (b) Díaz-Morales, O.; Hersbach, T. J.
P.; Hetterscheid, D. G. H.; Reek, J. N. H.; Koper, M. T. M. J. Am.
Chem. Soc. 2014, 136, 10432−10439. (c) Kumaran, E.; Leong, W. K.
Organometallics 2012, 31, 4849−4853. (d) Wang, C.; Villa-Marcos, B.;
Xiao, J. Chem. Commun. 2011, 47, 9773−9785.
(3) (a) Cheney, A. J.; Mann, B. E.; Shaw, B. L.; Slade, R. M. J. Chem.
Soc. D 1970, 0, 1176−1177. (b) Cheney, A. J.; Mann, B. E.; Shaw, B.
L.; Slade, R. M. J. Chem. Soc. A 1971, 0, 3833−3842. (c) Cheney, A. J.;
Shaw, B. L. J. Chem. Soc., Dalton Trans. 1972, 754−763.
́ ́
(4) Campos, J.; Esqueda, A. C.; Lopez-Serrano, J.; Sanchez, L.;
́
Cossio, F. P.; de Cozar, A.; Alvarez, E.; Maya, C.; Carmona, E. J. Am.
Chem. Soc. 2010, 132, 16765−16767.
(5) (a) Campos, J.; Lopez-Serrano, J.; Alvarez, E.; Carmona, E. J. Am.
́
Chem. Soc. 2012, 134, 7165−7175. (b) Campos, J.; Alvarez, E.;
Carmona, E. New J. Chem. 2011, 35, 2122. (c) Campos, J.; Esqueda, A.
C.; Carmona, E. Chem. - Eur. J. 2010, 16, 419−422.
(6) Campos, J.; Espada, M. F.; Lopez-Serrano, J.; Carmona, E. Inorg.
Chem. 2013, 52, 6694−6704.
(7) Campos, J.; Carmona, E. Organometallics 2015, 34, 2212−2221.
(8) (a) Burger, P.; Bergman, R. G. J. Am. Chem. Soc. 1993, 115,
10462−10463. (b) Taw, F. L.; Mellows, H.; White, P. S.; Hollander, F.
J.; Bergman, R. G.; Brookhart, M.; Heinekey, D. M. J. Am. Chem. Soc.
2002, 124, 5100−5108.
(9) Groom, C. R.; Bruno, I. J.; Lightfoot, M. P.; Ward, S. C. Acta
Crystallogr., Sect. B: Struct. Sci., Cryst. Eng. Mater. 2016, B72, 171−179.
(10) (a) Rubio, M.; Campos, J.; Carmona, E. Org. Lett. 2011, 13,
5236−5239. (b) Campos, J.; Rubio, M.; Esqueda, A. C.; Carmona, E. J.
Labelled Compd. Radiopharm. 2012, 55, 29−38.
(11) Esqueda, A. C.; Conejero, S.; Maya, C.; Carmona, E.
Organometallics 2010, 29, 5481−5489.
(12) (a) Shriver, D. F. Acc. Chem. Res. 1970, 3, 231−238. (b) Angelici,
R. J. Acc. Chem. Res. 1995, 28, 51−60.
J
DOI: 10.1021/acs.organomet.7b00688
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador,
P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.;
Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09,
Revision E.01; Gaussian, Inc., Wallingford, CT, 2013.
(30) Zhao, Y.; Truhlar, D. Theor. Chem. Acc. 2008, 120, 215−241.
(31) (a) Hehre, W. J.; Ditchfield, R.; Pople, J. A. J. Chem. Phys. 1972,
56, 2257. (b) Hariharan, P. C.; Pople, J. A. Theor. Chim. Acta. 1973,
28, 213. (c) Francl, M. M.; Pietro, W. J.; Hehre, W. J.; Binkley, J. S.;
Gordon, M. S.; Defrees, D. J.; Pople, J. A. J. Chem. Phys. 1982, 77,
3654.
(32) Andrae, D.; Haussermann, U.; Dolg, M.; Stoll, H.; Preuss, H.
Theor. Chem. Acc. 1990, 77, 123−141.
(33) Marenich, A. V.; Cramer, C. J.; Truhlar, D. G. J. Phys. Chem. B
2009, 113, 6378−6396.
(34) Lu, T.; Chen, F. W. J. Comput. Chem. 2012, 33, 580−592.
(35) (a) McLean, A. D.; Chandler, G. S. J. Chem. Phys. 1980, 72,
5639−5648. (b) Krishnan, R.; Binkley, J. S.; Seeger, R.; Pople, J. A. J.
Chem. Phys. 1980, 72, 650−654.
(36) Glendening, E. D.; Badenhoop, J. K.; Reed, A. E.; Carpenter, J.
E.; Bohmann, J. A.; Morales, C. M.; Landis, C. R.; Weinhold, F. NBO
6.0; Theoretical Chemistry Institute, University of Wisconsin,
Madison, WI, USA, 2013.
K
DOI: 10.1021/acs.organomet.7b00688
Organometallics XXXX, XXX, XXX−XXX