Convergent (De)Hydrogenative Pathways via a Rhodium α-Hydroxylalkyl Complex

Article abstract of DOI:10.1021/acs.organomet.7b00158

We report the convergent reaction pathways between [RhH(PPh₃)₄] and POP ketone (1) and alcohol (2) ligands that terminate in the formation of an α-hydroxylalkyl rhodium(I) complex (3), representing two halves of a formal reduction/oxidation pathway between 1 and 2. In the case of hydride transfer to 1, the formation of the α-hydroxylalkyl rhodium(I) complex (3) proceeds via a rare hydrido(²-carbonyl) complex (4). C-H activation in 2 at the proligand's central methine position, rather than O-H activation of the hydroxy motif, followed by loss of dihydrogen also generates the α-hydroxylalkyl rhodium(I) complex (3). The validity of the postulated reaction pathways is probed with DFT calculations. The observed reactivity supports α-hydroxylalkyl complexes as competent intermediates in ketone hydrogenation catalyzed by rhodium hydrides and suggest that ligands 1 and 2 may be noninnocent coligands in reported hydrogenation catalyst systems in which they are utilized.

Full text of DOI:10.1021/acs.organomet.7b00158

Article
pubs.acs.org/Organometallics
Convergent (De)Hydrogenative Pathways via a Rhodium
α‑Hydroxylalkyl Complex
Simon Sung,† Jie Kang Boon,† Johnathan J. C. Lee,† Nasir A. Rajabi,‡ Stuart A. Macgregor,‡
Tobias Kramer,‡,* and Rowan D. Young†,*
̈
†Department of Chemistry, National University of Singapore, 3 Science Drive 3, 117543 Singapore
‡Institute of Chemical Sciences, Heriot-Watt University, Edinburgh EH14 4AS, United Kingdom
S
* Supporting Information
ABSTRACT: We report the convergent reaction pathways
between [RhH(PPh3)4] and POP ketone (1) and alcohol (2)
ligands that terminate in the formation of an α-hydroxylalkyl
rhodium(I) complex (3), representing two halves of a formal
reduction/oxidation pathway between 1 and 2. In the case of
hydride transfer to 1, the formation of the α-hydroxylalkyl
rhodium(I) complex (3) proceeds via a rare hydrido(η2-
carbonyl) complex (4). C−H activation in 2 at the proligand’s
central methine position, rather than O−H activation of the
hydroxy motif, followed by loss of dihydrogen also generates
the α-hydroxylalkyl rhodium(I) complex (3). The validity of
the postulated reaction pathways is probed with DFT
calculations. The observed reactivity supports α-hydroxylalkyl
complexes as competent intermediates in ketone hydrogenation catalyzed by rhodium hydrides and suggest that ligands 1 and 2
may be “noninnocent” coligands in reported hydrogenation catalyst systems in which they are utilized.
The mechanism by which a metal hydride is transferred to a
bound ketone or aldehyde for cases operating via alkoxide
intermediates is well-studied spectroscopically in situ and
computationally (Scheme 1, A).5 However, the transfer of a
metal hydride to generate an α-hydroxylalkyl intermediate has
less precedent despite their inference in catalytic hydro-
genation3a,b and hydroformylation6 reactions.
Indeed, structurally characterized examples mapping hydride
migration to either electrophilic or nucleophilic positions of a
bound organocarbonyl are unknown. Intermediates involved in
such transitions are of great importance to a wide range of
carbonyl reductions but until this study have only been
interrogated in silico or observed spectroscopically in situ.6
α-Hydroxylalkyl complexes are typically unstable with
respect to β-hydrogen elimination, so examples of isolated
complexes are rare.7 In pioneering work, Gladysz et al. (and
later Garralda et al.) demonstrated the formation of α-
hydroxylalkyl complexes in constrained environments based
on hydride migration to o-(diphenylphosphino)benzaldehyde
ligands.7a,b
INTRODUCTION
■
The transfer of hydrogen from metals to ketones and, through
reversibility, from alcohols to metals is of fundamental
importance to (de)hydrogenation reactions mediated by
metal catalysts.1 Such reactions involve metal hydride
intermediates and can proceed via two distinct pathways
involving hydride transfer to either the electrophilic carbon or
the nucleophilic oxygen of the carbonyl group (Scheme 1).
Scheme 1. Ketone Hydrogenation Proceeding via an
Alkoxide Intermediate (Route A) or an α-Hydroxylalkyl
Intermediate (Route B)
If α-hydroxylalkyl complexes lie on the reaction pathway of
ketone hydrogenation, then they should also be accessible
through the C−H activation of an alcohol. Such selective
activation of C−H bonds in the presence of O−H bonds is of
great interest regarding simple alcohol functionalization, with
Thus, basic metal monohydrides tend to form metal alkoxide
intermediates with ketones (Scheme 1, route A); H2 addition
then gives the alcohol product and regenerates the metal
hydride.2,3 In contrast, hydrogenation of aldehydes and ketones
with acidic metal hydrides has been observed to proceed via α-
hydroxylalkyl intermediates in acidic media (Scheme 1, route
B).4
Received: February 28, 2017
© XXXX American Chemical Society
A
DOI: 10.1021/acs.organomet.7b00158
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
a
Scheme 2. Formation of 3 from Addition of 1 or 2 to [RhH(PPh3)4]
a
Geometry of compound 6 is discussed below.
the single-step C−H activation and functionalization of alcohol
Crystals of compound 4 were grown upon layering a toluene
solution of 4 with hexane at low temperature (253 K). The
molecular structure of 4 (Figure 1) supports the coordination
geminal C−H positions remaining a contemporary chemical
challenge.8 However, such selective C−H activation is
unknown, so this approach is nontrivial due to the presence
of several alternative reaction outcomes.
Herein we describe the controlled hydrogen transfer to and
from the ketone and alcohol moieties of the diphosphine POP
ligands 19 and 210 mediated by hydridotetrakis-
(triphenylphosphine)rhodium(I), [RhH(PPh3)4]. The results
demonstrate a convergent pathway to a common α-
hydroxylalkyl complex that is accessible from both ketone
and alcohol precursors using a common ligand platform. Within
this we demonstrate the formation of an α-hydroxylalkyl
species direct from a rare isolated metal-hydride/η2-ketone
precursor, as well as the geminal C−H activation of an alcohol.
DFT calculations are utilized to probe the mechanistic details of
these processes which are shown to map out fully route B
shown in Scheme 1. The results also highlight the potential
noninnocence of these POP ligands, which are commonly used
in asymmetric hydrogenation catalysis.9
Figure 1. Molecular structure of 4. Phenyl groups and hydrogens
except H1 are omitted for clarity; 50% thermal ellipsoids. H1 was
located in a Fourier difference map. Selected distances (Å) and angles
(deg): Rh1−C1, 2.118(7); Rh1−O1, 2.187(5); Rh1−P1, 2.382(2);
Rh1−P2, 2.318(2); Rh1−P3, 2.262(2); C1−O1, 1.339(8); P2−Rh1−
P3, 153.19(8); P1−Rh1−C1, P1, 152.3(2).
RESULTS AND DISCUSSION
■
Addition of POP ketone 1 to [RhH(PPh3)4] in benzene-d6
resulted in the loss of 3 equiv of PPh3 and the formation of the
α-hydroxyl complex 3 over 12 h. Monitoring the reaction at
shorter time intervals revealed that 1 and [RhH(PPh3)4]
initially formed hydrido(η2-carbonyl) 4 (within minutes) that
was converted into 3 over a matter of hours at room
temperature (Scheme 2). Formation of 4 was evident by the
of the carbonyl to rhodium, observed spectroscopically in
solution. The geometry around rhodium is best described as
pseudotrigonal bipyramidal, with numerous examples of
analogous pentacoordinate rhodium complexes subtended by
η2-olefin ligands exhibiting such geometry.13 Significant
elongation of the CO bond (1.339(8) Å) from the free
ligand 1 (1.213(3) Å)9 is observed indicating a significant
degree of π-retrodonation. However, the coordination is
consistent with a bound carbonyl, falling within the range of
previously reported rhodium η2-carbonyl complexes.14 Notably,
the molecular structure of 4 reveals the hydrido ligand to be
trans to the oxygen in the coordinated carbonyl, representing a
barrier for hydride to carbonyl migration.
1
presence of a rhodium-bound carbon (δC 137.8, JRhC = 9.2
Hz). The chemical shift and rhodium−carbon coupling
constant deviate notably from that of the proligand carbonyl
(δC 197.3) and imply a bonding mode lying between the
extreme cases of η2-carbonyl and metallaepoxide (defined by
the Dewar−Chatt−Duncanson model) and exemplified by
recently reported analogues [1-Ni(PPh3)]11 and [L1IrX]12 (L1
= κ3-P,(η2-C,O),P′-bis(5-(diisopropylphosphino)3,4-benzo[b]-
thiophenyl)-methanone, X = Cl or OH). A degree of π-
retrodonation to the carbonyl is supported by relatively small
In solution, α-hydroxylalkyl complex 3 is generated from 4
upon the transfer of hydrogen from rhodium to oxygen. The
1H NMR spectrum of compound 3 exhibits a hydroxyl
resonance at 3.15 ppm (4JPH = 3.5 Hz). Selective decoupling
of a phosphorus signal at δP 37.5 resolves the signal at δH 3.15
one-bond rhodium−phosphorus coupling constants (1JRhP
=
130.6, 108.9 Hz) in 4 indicating an electron-poor rhodium
center. FTIR spectroscopy could not provide support for
carbonyl coordination with failure to identify a specific CO
stretching band. However, a strong Rh−H stretch was observed
at 1969 cm−1 (cf. calcd value of 1993 cm−1, see Supporting
Information).
1
into a singlet, indicative of long-range H−31P coupling. The
addition of D2O to a solution of 3 resulted in the disappearance
of this signal, while other NMR signals remained unaffected.
The carbon−rhodium bond in 3 is characterized by a doublet
B
DOI: 10.1021/acs.organomet.7b00158
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
of doublet of triplets signal in the 13C NMR spectrum at δC
9.87 with a signal at δC 69.3. These data suggest the identities of
the signals at δH 9.87 and 1.39 to be the methine CH and OH
signals of bound 2 respectively. In the upfield region of the H
1
106.1 (1JRhC = 25.2 Hz), with a typical JRhC for a rhodium α-
hydroxylalkyl moiety.7b,h,i
1
X-ray diffraction study confirms 3 to be an α-hydroxylalkyl
complex (Figure 2). In agreement with solution data, 3 assumes
NMR spectrum of 5, a broad doublet is observed at δH −8.58
(2JPH = 90 Hz), 0.52 ppm upfield of the hydride signal of
1
[RhH(PPh3)4] (δH −8.06, JRhH = 12.7 Hz), suggesting
fluxional behavior. The 31P NMR spectrum of 5 further
revealed the dynamic behavior of 5, with two broad doublets
present at δP 34.3 (1JRhP = 170 Hz) and 31.0 (1JRhP = 131 Hz)
with a combined integral of three phosphorus nuclei relative to
free PPh3 (integration: 3P).
Analysis by 31P NMR spectroscopy of a solution of 5 and
liberated PPh3 generated from [RhH(PPh3)4] and 2 in toluene-
d8 at 223 K revealed the presence of at least three separate
phosphorus environments on rhodium (integration: 4P) with
complex coupling patterns in addition to free PPh3
1
(integration: 2P). At this temperature, the H spectrum of 5
1
revealed that fluxional processes were still occurring on the H
NMR time scale. The hydridic signal remained broad. However,
it had shifted upfield to −12.22 ppm, and the 2JPH for this signal
had increased to 110 Hz. Concurrently, the methine CH signal
in bound 2 had shifted upfield from δH 9.87 to 9.20. Overall,
the NMR data imply that 5 exists in equilibrium with its PPh3
adduct, 5·PPh3, and that the adduct may be preferred at lower
temperatures.
Figure 2. Molecular structure of 3. Phenyl groups, benzene solvent
molecule, and hydrogens except H11 are omitted for clarity; 50%
thermal ellipsoids. H11 was located in a Fourier difference map.
Selected distances (Å) and angles (deg): Rh1−C1, 2.103(12); C1−
O1, 1.458(14); Rh1−P1, 2.231(3); Rh1−P2, 2.254(3); Rh1−P3,
2.315(3); P1−Rh1−P2, 132.45(13); P3−Rh1−C1, 166.4(4); O1−
C1−Rh1, 106.4(8).
A downfield shift for C−H bonds in the vicinity of d8 metals
has been observed in bisphosphino methylene ligands related to
2 (that also undergo C−H activation) and has previously been
assigned as an anagostic interaction.15 Assignment based purely
on NMR spectroscopic evidence has recently been reported to
be misleading; however, we cautiously assign the C−H−Rh
interaction as anagostic with supporting computational analysis
(see below).16
a distorted square-planar geometry, with P1−Rh1−P2 and C1−
Rh1−P3 angles deviating greatly from linear (132.45(13) and
166.4(4)°, respectively). The significantly reduced average Rh−
P bond distances in 3 as compared to those in 4 point to a
more electron-rich Rh center in the former. The formation of 3
1
from 4 was monitored with H and 31P NMR spectroscopy
Further insight into the C−H activation of POP alcohol
ligand 2 was obtained from its reaction with [RhCl(COD)-
PPh3] that led to the formation of the hydridochloride 7
across a range of temperatures and in the presence of varying
quantities of added PPh3 (see Supporting Information).
Although it was apparent that free PPh3 accelerated the
reaction, the exact reaction order relative to [PPh3] could not
be precisely determined, but it was found to be between 0 and
1. This may be indicative of a nontrivial mechanism. Thus,
although accurate activation parameters from these collected
data could not be derived, they are discussed in the Supporting
Information.
1
alongside free 1,5-cyclooctadiene (Scheme 3). H NMR data
a
Scheme 3. Reaction of 2 and [RhCl(COD)PPh3] to Form 7
Compound 3 could also be generated by the addition of
alcohol proligand 2 to [RhH(PPh3)4] in benzene-d6 with
concomitant loss of H2 (Scheme 2). The formation of 3 from 2
and [RhH(PPh3)4] completes the (de)hydrogenation reaction
pathway between 1 and 2 mediated by [RhH(PPh3)4].
Monitoring the production of 3 from the combination of
either 1 or 2 with [RhH(PPh3)4] reveals that once a maximum
concentration of 3 has been achieved very small quantities of 4
are still observed. In the presence of >1 equiv of PPh3, this
equilibrium is established in a matter of days but takes weeks to
establish in the absence of PPh3. The ratio of 3/4 after
equilibrium is established is ca. 20:1 (see Supporting
Information), suggesting a ΔG of −1.8 kcal mol−1.
a
Subsequent treatment of 7 with Li[N(SiMe3)2] produces 4.
support compound 7 being an α-hydroxylalkyl complex, with a
hydroxyl signal located at δH 7.57. This signal appears as a
En route to compound 3 from 2 and [RhH(PPh3)4],
compound 5 is observed. Although 5 is transient at room
temperature, at 280 K it can be spectroscopically characterized
and is distinguished by the appearance of new signals in the 1H
4
doublet with long-range coupling to phosphorus (d, JPH = 7.5
Hz); selective 31P decoupling at δP 23.7 collapses this signal to a
singlet. The addition of a small quantity of D2O also resulted in
the disappearance of the signal while other NMR data remain
unaffected. The 1H NMR spectrum also reveals the appearance
3
NMR spectrum at δH 9.87 (br s) and 1.39 (d, JHH = 3.3 Hz)
that correlate to one another in a COSY 2D NMR experiment.
A HSQC experiment provided no correlation for the signal at
δH 1.39 to any 13C atoms but rather correlated the signal at δH
1
of an upfield hydride shift at δH − 16.22 (dtd, JRhH = 22.1 Hz
(d), JPH = 14.3 Hz (t), JPH = 9.2 Hz (d)).
2
2
C
DOI: 10.1021/acs.organomet.7b00158
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
1
A molecular structure determination of 7 (Figure 3) reveals
its geometry with the PCP ligand adopting a mer configuration
a doublet of doublets (2JPC = 95.2 Hz, JRhC = 28.6 Hz). After
warming and degassing of the sample, compound 3 was
quantitatively reformed. These data are indicative of the
formation of a Rh(III) center at low temperature in dynamic
equilibrium with 3 and molecular hydrogen. Given these
spectroscopic data, 6 is assigned as a cis dihydride featuring a
facially coordinated PCP ligand (Scheme 4).15b,e Selected NMR
spectroscopic data for compound 3−7 are shown in Table 1.
Scheme 4. Hydrogenation of 3 Generates Cis Dihydride 6
Figure 3. Molecular structure of 7. Phenyl groups and hydrogens
except H11 and H12 are omitted for clarity; 50% thermal ellipsoids.
H11 and H12 were located in a Fourier difference map. Selected
distances (Å) and angles (deg): Rh1−C1, 2.177(4); Rh1−Cl1,
2.500(1); Rh1−P1, 2.4019(12); Rh1−P2, 2.3189(13); Rh1−P3,
2.2971(12); P2−Rh1−P3, 153.03(4); P1−Rh1−C1, 174.46(12).
Confirmation that the C−H methine of 2 is activated in
reactions with rhodium (as opposed to O−H activation
followed by rearrangement) is confirmed by employing
isotopologues 2a and 2b (Scheme 5).C−H activation is
expected on a dehydrogenation pathway that invokes an α-
hydroxylalkyl intermediate. When [RhH(PPh3)4] is reacted
with isotopologue 2a, 3a is generated (Scheme 5), which is
spectroscopically identical to 3, except that the RhCOH signal
at δH 3.15 was diminished and a signal at 3.25 ppm was located
after C−H oxidative addition to the rhodium center. It is also
observed that the hydroxyl hydrogen (H11), located in a
Fourier difference map, is hydrogen-bonded to the proximal
chloride ligand (Cl1−H11dist = 2.273 Å). Induced elimination
of HCl from 7 by treatment with 1 equiv of Li[N(SiMe3)2]
results in the formation of compound 4, which then transforms
to 3. This stands in contrast to the reaction between 2 and
[RhH(PPh3)4] that generates 3 without any observation of 4,
signifying that H2 loss occurs via a cis-dihydride intermediate
rather than through elimination of H2 from a trans-dihydride
analogue of 7 (i.e., a trans-RhIII(H)2(COH)P2 fragment).
To investigate the possible identity of cis-dihydride
intermediate 6, dihydrogen (4 atm) was introduced into an
NMR sample tube containing 3 in toluene-d8 solution. At room
2
in the H NMR spectrum of 3a, signifying the deuteration of
the hydroxyl position. Conversely, when 2b is reacted with
[RhH(PPh3)4], compound 3 is produced with loss of HD.
Addition of 2b to [RhCl(COD)PPh3] resulted in the
production of 7b, identical to 7 (by NMR spectroscopy) with
the exception of substitution of the hydride ligand with
deuterium, as evident by the absence of a hydridic signal in the
1H NMR spectrum.
To elucidate the mechanistic details of the convergent
pathways that convert both the ketone (4) and alcohol (5)
precursors into the α-hydroxylalkyl product 3, a computational
analysis of the associated free energy surfaces was carried out
using DFT calculations at the B97-D3/BS2//BP86/BS1 level
of theory corrected for benzene solvent (see Supporting
Information for computational details). The most accessible
computed pathways at 298 K for both processes are detailed in
Figure 4. The optimized structure of 4 agrees well with the
crystallographic data. In particular the CO (calcd: 1.35 Å,
exp.: 1.339(8) Å), Rh−C (calcd: 2.16 Å, exp.: 2.118(7) Å), and
Rh−O (calcd: 2.22 Å, exp.: 2.187(5) Å) distances are well-
reproduced, along with the trans-P−Rh−P angle of the POP
ligand (calcd: 154.0°, exp.: 153.19(8)°).
1
temperature, H NMR spectroscopy reveals the formation of a
broad signal at δH −2.8 (integration: 2H). In addition, the
signal for free H2 (expected at δH 4.50) is not observed. The
hydroxyl signal originally at δH 3.15 is broadened and observed
to shift downfield to δH 3.65 (integration: 1H). The 31P NMR
spectrum of this sample reveals resonances at δP 45.6 (dd, 2H,
1JRhP = 125.1 Hz, 2JPP = 20.2 Hz) and 41.7 (dt, 1H, 1JRhP = 93.2
Hz, 2JPP = 20.2 Hz), displaying a similar chemical shift to 3, but
reduced coupling constants. As the temperature is lowered to
233 K, the hydridic signal resolves into a broad doublet at δH −
6.8 (2JPH = 140 Hz), indicative of a single trans phosphorus-
hydride environment. T1 measurements at various temperatures
excluded the identity of 6 as a dihydrogen complex (see
Supporting Information). At this temperature, the hydroxyl
signal is observed at δH 4.25 as a doublet (4JPH = 4.4 Hz), and
free hydrogen is observed as a broad signal at δH 4.5 that
Analysis of the natural bond orbitals (NBOs) in 4 shows that
CO coordination to the metal center is governed by πCO
→
Rh donation, reinforced by substantial π*CO ← Rh back-
donation (see Figure S65). The elongation of the CO bond
arises due to notable population of the π*CO orbital (0.79 e−)
and depopulation of the πCO orbital (1.82 e−). The partial
reduction of the double-bond character of the CO bond is
also reflected in the Wiberg bond index (1.18), lying in
between those for the CO double bond in 1 (1.68) and the
C−O(H) single bond in 2 (0.91). In contrast, the indices for
the Rh···C (0.39) and Rh···O (0.24) interactions are smaller
compared to those found for the Rh−C (0.47) and Rh−H
(0.56) bonds in 3 and 4, respectively, which serve as a reference
1
sharpens at lower temperatures. A H{31P} NMR spectrum
with a decoupling window centered at δP 40.0 collapses both
the hydridic and hydroxyl signals into singlets.
The 31P NMR spectrum of the sample at 233 K shows a
broad doublet at δP 45.6 (1JRhP = 86.5 Hz) and a doublet of
2
triplets at δP 41.7 (1JRhP = 84.5 Hz, JPP = 14.9 Hz). HMBC
experiment at 233 K (optimized for JCH = 10 Hz) exhibits a
correlation between the hydroxyl proton at δH 4.25 and a signal
at δC 95.8. A 1D 13C NMR experiment revealed this signal to be
D
DOI: 10.1021/acs.organomet.7b00158
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Table 1. Selected NMR Spectroscopic Data for Compounds 3-7
compound
δC (Rh−C)
2JPC (Hz)
1JRhC (Hz)
JRhP (Hz)
JPP (Hz)
3
4
5
6
7
106.1
137.8
N/A
4.8 (t), 73.5 (d)
25.2
9.2
189, 120
131, 109
28.3
10.1
a
a
23.7 (d)
b
N/A
N/A
170, 131
c
ac
,
c
c
c
95.8
95.2 (d)
28.6
87, 85
14.9
24.0
a
99.4
94.1 (d)
20.8
120, 79
a
b
c
Coupling to cis phosphorus nuclei was not observed. 280 K. 233 K.
The optimized structure of 3 reproduces the distorted
square-planar geometry around Rh seen experimentally: trans-
P−Rh−P (calcd: 132.9°, exp.: 132.45(13); trans-C−Rh−P
(calcd: 164.1°, exp.: 166.4(4)). The Rh−C distance is reduced
from 2.16 Å in 4 to 2.13 Å in 3, and this is paralleled by an
increase in the calculated Rh−C isotropic spin−spin coupling
Scheme 5. Isotopomers 2a, 2b React with [RhH(PPh3)4] to
Generate Isotopologues 3a and 7b, Respectively
1
1
constant (4: JRhC = −9.1 Hz; 3: JRhC = −17.4 Hz).
The computed mechanism for the formation of 3 from 4
starts with an initial isomerization of 4, with a calculated barrier
of 17.7 kcal mol−1 proceeding via TS(4−Int1). The POP ligand
undergoes isomerization from a mer-κ3-P,(CO),P to a fac-κ3-P,
(CO),P binding mode (∠P−Rh−P 109.1°). Concomitantly,
the hydride moves from its equatorial coordination site trans to
oxygen into the opening axial position. The distortion of the
ligand in TS(4−Int1) also causes the CO unit to move away
from Rh (Rh−O: 2.52 Å; Rh−C: 2.38 Å), thereby decreasing π-
backbonding from the metal center and restoring more double-
point. Thus, in accordance with the experimental findings the
{RhCO} unit is best described as a Rh-bound carbonyl.
Figure 4. Computed profile (B97-D3/BS2//BP86/BS1) for the transformation of 4 into 3 (left) and 5 into 3 (right). The gray profile is associated
with adduct 5P, featuring H-bonded PPh3. Relative Gibbs free energies (298 K, kcal mol−1), corrected for benzene solvent, are given along with key
bond metric data (Å, deg). Double-arrows indicate energy spans ΔG(1), ΔG(2), and ΔG(2)′.
E
DOI: 10.1021/acs.organomet.7b00158
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
bond character (CO: 1.28 Å). The cis arrangement of the
{Rh(CO)(H)} moiety in Int1 (G° = +0.5 kcal mol−1) allows
for insertion of the ketone into the Rh−H bond through
TS(Int1−3) at 21.1 kcal mol−1 to form the α-hydroxylalkyl in
3. By inspection, TS(Int1−3) also defines the overall energetic
span17 (ΔG(1) = +21.1 kcal mol−1) for the transformation of 4
into 3, with the computed barrier being consistent with the
slow process seen experimentally. The hydrogen transfer in
TS(Int1−3) is accompanied by an isomerization of the ligand
to its distorted mer-κ3-P,C,P form (∠P−Rh−P 138.8°).
Complex 3 is energetically stabilized by a mere 1.8 kcal
mol−1 relative to 4. The marginal exergonicity of this process is
in line with the establishment of an equilibrium between these
two species, as confirmed by experiment.
The lowest-energy pathway for the formation of 3 from 5 is
shown on the right side of Figure 4. Precursor complex 5
features an approximately square-planar geometry around the
RhI center, with the POP ligand adopting a cis-κ2-P,P
arrangement (∠P−Rh−P 100.9°) and computed Rh···Ha and
Rh···Ca distances of 2.48 and 3.50 Å, respectively, to the central
C−H bond of the ligand. Computed AIM and NBO parameters
suggest that the Rh···Ha−Ca interaction is of closed-shell
electrostatic nature, in line with a weak anagostic inter-
action.16,18 Oxidative addition of the Ca−Ha bond across the
PPOP−Rh−PPh3 vector occurs with a barrier of 26.4 kcal mol−1
to yield intermediate 6 (−5.1 kcal mol−1). Activation of the C−
H bond in TS(5−6) is accompanied by movement of the PPh3
ligand into the axial position. We have also considered other
possibilities for this reaction step, none of which presented a
feasible alternative. Oxidative addition across the PPPh3−Rh−H
vector in the alternative trans-isomer 5′ shifts the energy profile
upward by ∼10 kcal mol−1 (see Supporting Information). A
search for a concerted C−H activation step via σ-bond
metathesis proved unsuccessful. Experimentally there is an
excess of PPh3 present in solution, so we also scrutinized the
possible effect of C(H)OH···PPh3 H-bonding on the C−H
activation. Under these circumstances, the barrier is notably
lowered (ΔG°⧧ = 18.5 kcal mol−1, relative to 5P, gray profile in
Figure 4). The optimized structure of the corresponding
transition state TS(5P-6P) is shown in Figure 5. The optimized
bond parameters in TS(5P−6P) closely resemble those in
TS(5−6), with Rh···Ca, Rh···Ha, and Ca−Ha distances of 2.45,
1.61, and 1.43 Å, respectively. The (O)H···PPh3 distance is 2.45
Å, similar to the H-bond in complex 5P. Complex 6 exhibits an
octahedral coordination geometry around the RhIII center,
featuring a fac-κ3-P,C,P tridentate ligand with the Rh−C bond
trans to PPh3. Facile reductive elimination of the cis-hydrides in
6 proceeds with a barrier of 11.4 kcal mol−1 readily generating
the final product 3 upon loss of H2. The overall reaction rate is
determined by the initial oxidative addition step at 18.5 kcal
mol−1, and this reduced barrier is consistent with the
observation that 5 is transient at room temperature. Note
that although the formation of 3 is computed to be slightly
endergonic relative to 6 (ΔG° = +3.3 kcal mol−1
)
experimentally species 3 and 6 are in equilibrium which can
be driven to 3 by removal of H2 upon degassing the solution.
Experimentally the NMR characterization of 5 points to it
being fluxional in solution and, moreover, that temperature-
dependent coordination of PPh3 to the Rh center also occurs.
Computationally, the trigonal-pyramid and trans-κ2-P,P isomers
of 5 lie ∼10 kcal mol−1 above the cis-isomer, so these may be
kinetically accessible in potential H/PPh3 exchange pathways.19
Formation of a trigonal-bipyramidal 18-electron complex, 5·
PPh3, via axial addition of PPh3 to 5 was computed to be
energetically strongly favored (ΔG° = −7.6 kcal mol−1), even
when the basis set superposition error (BSSE) was taken into
account. This value runs counter to experimental evidence
suggesting a dynamical equilibrium in which PPh3 reversibly
binds to 5 (i.e., a thermoneutral process with ΔG° ≈ 0 kcal
mol−1). Although dispersion-corrected DFT can predict
phosphine−metal binding energies with good accuracy,20 in
the present example the metal−ligand bond strength appears to
be strongly overestimated by the calculations. Of a range of
functionals that were tested, B3LYP-D3 performs well for the
phosphine binding energy (see Table 2).20e However, with this
Table 2. Summary of the Functional Dependence of
Phosphine Binding Energies According to 5 + PPh3 ⇆ 5·
PPh3 (ΔGbind+BSSE, in kcal mol−1) as Well as Key Energy
a
Spans for the Overall Reaction Profile in Figure 4
functional
ΔGbind+BSSE
ΔG(1)
ΔG(2)
ΔG(2)′
BP86-D3
B97-D3
B3LYP-D3
M06
−13.6
−7.6
+0.1
22.2
21.1
25.6
24.9
22.6
26.4
29.9
33.7
14.6
18.5
21.5
24.4
+8.6
a
See Figure 4for the definition of ΔG(1), ΔG(2) ΔG(2)′).
approach the overall barriers linking 4 to 3 and 5 to 3 are in
excess of 25 kcal mol−1, rather too high for these room
temperature processes. It seems that no single functional can
provide balanced energetics for the various ligand binding and
bond activation steps in this system. Nonetheless, our
conclusions regarding the mechanism for the convergent
formation of 3 from 4 and 5, respectively, obtained with the
B97-D3/BS2//BP86/BS1 protocol are qualitatively in good
agreement with the experimental observations.
It is plausible that dissociation of PPh3 from 5·PPh3 must
occur prior to C−H bond activation to give 6. Indeed, a
stepwise relaxed scan of the Rh···Ca distance in 5·PPh3 induces
dissociation of one PPh3 ligand, restoring the square-planar
geometry of 5 before accessing TS(5−6). The computed
intrinsic reaction coordinate clearly confirms that TS(5−6)
connects 6 to 5 providing support for 5 as an intermediate
along the reaction profile.
Figure 5. Optimized geometry of TS(5P−6P). Bond distances are in Å.
F
DOI: 10.1021/acs.organomet.7b00158
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Hz), 131.0 (s), 128.5 (t, JCP = 5.6 Hz), 127.6 (d, JCP = 8.9 Hz), 126.5
CONCLUSIONS
■
2
1
(s) 106.1 (ddt, JPC = 73.5 (d), 4.8 (t) Hz, JRhC = 25.2 Hz). 31P{1H}
1
2
[RhH(PPh3)4] reacts with both the POP ketone (1) and POP
alcohol (2) proligands to produce α-hydroxylalkyl (3) through
convergent pathways. A number of key intermediates for both
branches of this reactivity were either isolated and fully
characterized or characterized in situ by NMR spectroscopy. In
particular, reaction with 1 gives intermediate 4, a rare example
of a trapped η2-ketone hydrido complex that subsequently
undergoes insertion. With 2 the reaction proceeds via C−H
activation geminal to the hydroxyl group. Independent
synthesis of hydrido chloride complex 7 provided evidence of
the feasibility of this novel C−H activation.
The underlying mechanisms were further validated by DFT
calculations. These show that the formation of 3 from 4
involves initial mer-fac-isomerization of the ligand followed by
rate-limiting insertion. For the generation of 3 from POP
alcohol precursor 5, the initial C−H oxidative addition is rate-
limiting, and this process is facilitated by the presence of PPh3
which H-bonds to the C(H)OH moiety of the ligand.
The observed reactivity supports α-hydroxylalkyl complexes
as competent intermediates in ketone hydrogenation catalyzed
by rhodium hydrides and suggests that 1 and 2 may be
“noninnocent” ligands in reported hydrogenation catalyst
systems. This work demonstrates a new strategy via ketone
insertion to access PCsp3P pincer complexes for metals that
disfavor C−H activation. Additionally, the demonstration of
C−H activation of a geminal hydroxyl position provides insight
into the selective C−H activation in the presence of an alcohol
functionality. Ongoing research is being undertaken to
determine the reactivity of compound 3 and its analogues,
particularly in regards to developing new synthetic routes to
PCsp3P and PCsp2P pincer complexes and their potential in
catalysis.
NMR (162 MHz, C6D6) = δP 37.5 (dt, JRhP = 120.4 (d), JPP = 28.3
1
2
(t) Hz, 1P), 43.3 (dd, JRhP = 189.2 (d), JPP = 28.3 (d) Hz, 2P).
HRMS (ESI-TOF) m/z: [M − H]+ Calcd for C55H43OP3Rh 915.1582;
Found 915.1546; [M − OH]+ Calcd for C55H43P3Rh 899.1633; Found
899.1595.
Synthesis of Complex 4. Method A. Benzene (5 mL) was added
to a mixture of 1 (55.1 mg, 0.100 mmol) and [RhH(PPh3)4] (115.3
mg, 0.100 mmol), and the resultant orange solution stirred at room
temperature for 30 min, after which the solution was filtered. The
filtrate was evaporated to give an orange residue and n-hexane (15
mL) was added. After trituration of the mixture for 5 min, the solid
was filtered and washed with diethyl ether (2 × 2 mL) and then n-
hexane (5 × 10 mL). After drying in vacuo, the product was isolated as
an orange solid (64 mg, 70%).
Method B. Li[N(SiMe3)2] (11.7 mg, 0.07 mmol) was added to
solution of 7 (66.7 mg, 0.07 mmol) in C6H6 (5 mL) at room
temperature and stirred for 15 min. The solution was rapidly
evaporated under vacuum, and then diethyl ether (5 mL) was
added. After trituration of the mixture for 5 min, the solid was filtered
and washed with diethyl ether (2 mL) and then n-hexane (2 × 10 mL).
After drying in vacuo, product 4 was isolated as an orange solid (35
mg, 55%). 1H NMR (500 MHz, C6D6) = δH −13.27 (ddt, 1JRhH = 21.2
2
2
(d), JPH = 10.3 (d), JPH = 4.2 (d) Hz, 1H, Rh-H), 6.64−7.04 (m,
31H, Ar-H), 7.48−7.55 (m, 6H, Ar-H), 7.59−7.67 (m, 4H, Ar-H), 8.05
(d, 3JH−H = 8.0 Hz, 2H, Ar-H). 13C{1H} NMR (126 MHz, C6D6) = δC
1
2
156.8 (t, JPC = 13.0 Hz), 137.8 (dd, JRhC = 23.7 (d), JPC = 9.2 (d)
Hz), 137.1 (m), 134.7 (d, JPC = 13.5 Hz), 133.8 (s), 133.4 (dt, JPC
20.0 (d), 6.9 (t) Hz), 129.1 (s), 128.9 (s), 128.4 (s), 128.2 (d, JPC
=
=
28.7 Hz), 127.4 (d, JPC = 8.3 Hz), 127.3 (t, JPC = 4.1 Hz), 126.0 (m),
1
125.4 (s). 31P{1H} NMR (162 MHz, C6D6) = δP 31.0 (dt, JRhP
=
=
2
1
2
108.9 (d), JPP = 10.1 (t) Hz, 1P), 40.4 (dd, JRhP = 130.6 (d), JPP
10.1 (d) Hz, 2P). IR (nujol mull): 1969 [ν(Rh−H)] cm−1; HRMS
(ESI-TOF) m/z: [M + H]+ Calcd for C55H45OP3Rh 917.1738; Found
917.1736.
Synthesis of Complex 7. Benzene (2 mL) was added to a mixture
of 2 (27.6 mg, 0.05 mmol) and [RhCl(PPh3) (COD)] (25.4 mg, 0.05
mmol), and the resultant solution was stirred at room temperature for
1 h. The solution was concentrated under vacuum to give a residue,
and then n-hexane (5 mL) was added to produce a precipitate. The
solid was filtered and washed with n-hexane (3 × 5 mL). After drying
in vacuo, the product was isolated as yellow-brown solid (31 mg, 65%).
EXPERIMENTAL SECTION
■
General Information. All manipulations were carried out under
nitrogen using a glovebox and/or Schlenk techniques. All reactions
were performed in glassware that was oven-dried for at least 12 h.
Benzene was distilled over sodium and benzophenone under a
nitrogen atmosphere and stored over 4 Å molecular sieves prior to use.
Diethyl ether and n-hexane were dried over activated alumina using an
LC Technology Solution Inc. SP-1 Solvent Purification System and
deoxygenated prior to use. C6D6 used was stirred over CaH2 at room
temperature under a nitrogen atmosphere overnight prior to
distillation under reduced pressure and storage over 4 Å molecular
sieves. Toluene-d8 was deoxygenated and stored over 4 Å molecular
sieves prior to use. [RhH(PPh3)4] and ligands 1 and 2 were prepared
according to reported methods.21
1
1H NMR (500 MHz, C6D6) = δH −16.22 (1 H, dtd, JRhH = 22.1 Hz
(d), 2JPH = 14.3 Hz (t), 2JPH = 9.2 Hz (d), Rh-H), 7.57 (1 H, d, JPH
=
7.5 Hz, OH), 8.2−6.6 (43 H, m, Ar-H). 13C{1H} NMR (126 MHz,
C6D6) = δC 163.9 (t, JPC = 15 Hz), 141.9 (t, JPC = 25 Hz), 136.4 (t, JPC
= 24 Hz), 135.1 (s (br)), 134.4 (t, JPC = 5.6 Hz), 133.6 (t, JPC = 5.6
Hz), 133.0 (s), 130.0 (s (br)), 129.2 (s), 129.1 (s), 128.9 (s), 128.3
(s), 127.3 (d, JPC = 8.4 Hz), 124.4 (s (br)), 99.4 (dd, JPC = 94.1 Hz,
JRhC = 20.8 Hz). 31P{1H} NMR (202 MHz, C6D6) = δ 39.7 (dd, JRhP
=
120.0 Hz, JPP = 24.0 Hz, 2 P), 23.7 (dt, JRhP = 79.0 Hz, JPP = 24.0 Hz, 1
P). IR (nujol mull): 2079 [ν(Rh−H)], 3303 [ν(O−H)] cm−1. HRMS
(ESI-TOF) m/z: [M − Hydride]+ Calcd for C55H44ClOP3Rh
951.1349; Found 951.1335; [M − Cl]+ Calcd for C55H45OP3Rh
917.1738; Found 917.1703.
NMR spectroscopy data were obtained using Bruker AV-300, AV-
400, and AV-500 spectrometers. HRMS (ESI-TOF) spectra were
obtained using an Agilent Technologies 6230 TOF LC/MS. IR
spectroscopy data were obtained using Bruker ALPHA FTIR
spectrometers.
In Situ Characterization of Complex 5. 2 (10 mg, 0.018 mmol)
and [RhH(PPh3)4] (21 mg, 0.018 mmol) were added into a NMR
tube under N2 atmosphere. The components were dissolved in C6D6
(0.6 mL) and then analyzed soon after. The NMR spectroscopic data
were obtained at 280 K to reduce the rate of conversion from
intermediate 5 to complex 3. 1H NMR (500 MHz, C6D6) = δH −8.58
(1 H, d (br), 2JPH = 90 Hz, Rh-H), 1.39 (1 H, d, 3JHH = 3.3 Hz, OH),
Synthesis of Complex 3. 1 (10 mg, 0.018 mmol) and
[RhH(PPh3)4] (21 mg, 0.018 mmol) were added to a NMR tube
under N2 atmosphere. The components were dissolved in C6D6 (0.6
mL) to form an orange solution immediately, which turned green
overnight. NMR analyses showed the reaction to be virtually
quantitative in the formation of complex 3. 1H NMR (300 MHz,
C6D6) = δH 3.15 (d, 4JPH = 3.5 Hz, 1H, OH), 6.68−7.09 (m, 27H, Ar-
1
8.0−6.2 (43 H, m, Ar-H), 9.87 (1 H, s (br), C(OH)-H). H−1H
3
H), 7.23−7.47 (m, 15H, Ar-H), 7.59 (d, JHH = 7.8 Hz, 2H, Ar-H).
COSY NMR shows strong correlation between signals at δH 9.87 and
1.39. The doublet at δH 1.39 is not resolved upon broadband 31P
decoupling. Selected 13C NMR data (from HMBC/HSQC) = δC 69.3
(C), 147.4 (C). 31P{1H} NMR (202 MHz, C6D6) = δP 31.0 (d (br),
13C{1H} NMR (126 MHz, C6D6) = δC 162.7 (t, JPC = 18.5 Hz), 145.0
(dd, JPC = 19 Hz, 8.5 Hz), 144.6 (d, JPC = 25.3 Hz), 140.0 (d, JPC
25.2 Hz), 138.9 (d, JPC = 44.2 Hz), 138.7 (d, JPC = 25.7 Hz), 137.3 (t,
JPC = 16.0 Hz), 135.1 (s), 134.3 (t, JPC = 11.0 Hz), 133.8 (q, JPC = 6.9
=
1
1JRhP = 131 Hz), 34.3 (d (br), JRhP = 170 Hz).
G
DOI: 10.1021/acs.organomet.7b00158
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
In Situ Characterization of Complex 6. To a sample of 3
prepared from 1 and [RhH(PPh3)4] in toluene-d8 (0.6 mL) as
described above was applied a pressure of hydrogen gas (4 atm). The
sample was then analyzed using VT-NMR spectroscopic experiments.
Selected NMR spectroscopic data for 6 at 233 K: 1H NMR (500 MHz,
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
2
We thank the National University of Singapore and the
Singapore Ministry of Education for financial support (WBS R-
143-000-586-112 and R-143-000-666-114) and Heriot-Watt
University for the award of a James Watt Scholarship.
C6D6) = δH −6.8 (2 H, d (br), JPH = 140 Hz, Rh-H), 4.25 (1 H, d,
4JPH = 4.4 Hz, OH). 13C NMR data (from HMBC) = δC 95.8 (1 C, dd,
1
2JPC = 95.2 Hz, JRhC = 28.6 Hz, Rh-C−OH). 31P{1H} NMR (202
MHz, C6D6) = δP 45.6 (d (br), 1JRhP = 86.5 Hz), 41.7 (dt, 1JRhP = 84.5
2
Hz, JPP = 14.9 Hz).
Preparation of Deuterium-Labeled Ligand 2a. A 1:1 mixture
of D2O/THF was added to 2 followed by evaporation to dryness.
Approximately 81% deuteration of the hydroxyl position at 2.25 ppm
was determined by 1H NMR spectroscopy. 1H NMR (400 MHz,
C6D6) = δH 6.89 (td, J = 7.5, 1.4 Hz, 2H, Ar-H), 6.94−7.06 (m, 14H,
Ar-H), 7.19−7.24 (m, 2H, Ar-H), 7.27−7.37 (m, 8H, Ar-H), 7.50−
7.56 (m, 2H, Ar-H), 7.76 (t, J = 6.2 Hz, 1H, C(OH)-H). 31P{1H}
NMR (162 MHz, C6D6) = δP −17.3 (s, 2 P).
Preparation of Deuterium-Labeled Ligand 2b. Part A. (2-
Bromophenyl)diphenylphosphene (2.00 g, 5.9 mmol) was dissolved in
diethyl ether (25 mL). The solution is then treated dropwise with n-
BuLi in hexane (6 mL, 1.6 M, 9.6 mmol) at −78 °C and stirred for 30
min. Dimethylformamide-d7 (3 mL, 38.6 mmol) was added thereafter
at −78 °C. The mixture was then allowed to come to room
temperature and was stirred overnight. Dilute aqueous HCl solution
(20 mL) was added, and then the aldehyde product was extracted
using DCM (3 × 30 mL). The combined extractions were dried with
Na2SO4 and then evaporated. The crude product was then
recrystallized using methanol to give deutero 2-(diphenylphosphino)-
benzaldehyde (0.901 g, 52%).
REFERENCES
■
(1) (a) Birch, A. J.; Williamson, D. H. In Organic Reactions; John
Wiley & Sons, Inc.: Hoboken, NJ, 2011. (b) de Vries, J. G.; Elsevier, C.
J. The Handbook of Homogeneous Hydrogenation; Wiley-VCH:
Weinheim, 2007.
(2) (a) Schrock, R. R.; Osborn, J. A. J. Chem. Soc. D 1970, 567.
(b) Malacea, R.; Poli, R.; Manoury, E. Coord. Chem. Rev. 2010, 254,
729. (c) Samec, J. S. M.; Backvall, J.-E.; Andersson, P. G.; Brandt, P.
̈
Chem. Soc. Rev. 2006, 35, 237. (d) Clapham, S. E.; Hadzovic, A.;
Morris, R. H. Coord. Chem. Rev. 2004, 248, 2201. (e) Gregorio, G.;
Pregaglia, G.; Ugo, R. Inorg. Chim. Acta 1969, 3, 89. (f) Abdur-Rashid,
K.; Clapham, S. E.; Hadzovic, A.; Harvey, J. N.; Lough, A. J.; Morris, R.
H. J. Am. Chem. Soc. 2002, 124, 15104. (g) Wigfield, D. C. Tetrahedron
1979, 35, 449.
(3) Related systems that partake in ligand-assisted ionic hydro-
genations typically rely upon the ligand to deliver a protic hydrogen,
while the metal hydride transfers to the electrophilic carbonyl position
(i.e., the metal-hydride also acts as a nucleophile as in route A). See
(a) Bullock, R. M. Chem. - Eur. J. 2004, 10 (10), 2366. (b) Clapham, S.
E.; Hadzovic, A.; Morris, R. H. Coord. Chem. Rev. 2004, 248 (21−24),
2201. (c) Wang, D.; Astruc, D. Chem. Rev. 2015, 115 (13), 6621.
(4) (a) Simpson, M. C.; Cole-Hamilton, D. J. Coord. Chem. Rev.
1996, 155, 163. (b) MacDougall, J. K.; Simpson, M. C.; Green, M. J.;
Cole-Hamilton, D. J. J. J. Chem. Soc., Dalton Trans. 1996, 1161.
(c) Sola, M.; Ziegler, T. Organometallics 1996, 15, 2611.
(d) Cheliatsidou, P.; White, D. F. S.; Slawin, A. M. Z.; Cole-
Hamilton, D. J. Dalton Trans. 2008, 2389. (e) MacDougall, J. K.;
Simpson, M. C.; Green, M. J.; Cole-Hamilton, D. J. J. J. Chem. Soc.,
Dalton Trans. 1996, 1161. (f) Fahey, D. R. J. Am. Chem. Soc. 1981,
103, 136. (g) Milstein, D. J. Am. Chem. Soc. 1986, 108, 3525.
(5) For examples, see (a) Peterson, E.; Khalimon, A. Y.; Simionescu,
R.; Kuzmina, L. G.; Howard, J. A. K.; Nikonov, G. I. J. Am. Chem. Soc.
Part B. (2-Bromophenyl)diphenylphosphine (0.423 g, 1.24 mmol)
was dissolved in diethyl ether (10 mL) and treated with n-BuLi (2.9
mL, 1.6 M, 1.78 mmol) at 0 °C. The reaction mixture was stirred for
30 min; thereafter, deutero 2-(diphenylphosphino)benzaldehyde
(0.519 g, 1.78 mmol) was added. The reaction mixture was then
stirred for an additional hour. The mixture is then allowed to come to
room temperature, and degassed dilute aqueous HCl solution was
added. Then, the product was extracted with diethyl ether (3 × 20
mL). The solvent was removed under vacuum and the crude product
was recrystallized using methanol to give ligand 2b as a white solid
1
(0.136 g, 20%). H NMR (400 MHz, C6D6) = δH 2.25 (t, J = 1.4 Hz,
1H, OH), 6.89 (td, J = 7.5, 1.4 Hz, 2H, Ar-H), 6.94−7.09 (m, 14H, Ar-
H), 7.18−7.24 (m, 2H, Ar-H), 7.26−7.37 (m, 8H, Ar-H), 7.49−7.56
(m, 2H, Ar-H). 31P{1H} NMR (162 MHz, C6D6) = δP −17.3 (s, 2 P).
2009, 131, 908. (b) Sieffert, N.; Buhl, M. J. Am. Chem. Soc. 2010, 132,
̈
8056. (c) Bosson, J.; Poater, A.; Cavallo, L.; Nolan, S. P. J. Am. Chem.
Soc. 2010, 132, 13146. (d) Khalimon, A. Y.; Ignatov, S. K.; Simionescu,
R.; Kuzmina, L. G.; Howard, J. A. K.; Nikonov, G. I. Inorg. Chem.
2012, 51, 754. (e) Wang, W.; Gu, P.; Wang, Y.; Wei, H.
Organometallics 2014, 33, 847. (f) Iron, M. A.; Sundermann, A.;
Martin, J. M. L. J. Am. Chem. Soc. 2003, 125, 11430.
(6) (a) Pruett, R. L. Ann. N. Y. Acad. Sci. 1977, 295, 239. (b) Feder,
H. M.; Rathke, J. W. Ann. N. Y. Acad. Sci. 1980, 333, 45. (c) Fahey, D.
R. J. Am. Chem. Soc. 1981, 103, 136. (d) Bradley, J. S. J. Am. Chem. Soc.
1979, 101, 7419. (e) Dombek, B. D. J. Am. Chem. Soc. 1980, 102,
6855. (f) Keim, W.; Berger, M.; Schlupp, J. J. Catal. 1980, 61, 359.
(g) Daroda, R. J.; Blackborow, J. R.; Wilkinson, G. J. Chem. Soc., Chem.
Commun. 1980, 1098. (h) Daroda, R. J.; Blackborow, J. R.; Wilkinson,
G. J. Chem. Soc., Chem. Commun. 1980, 0, 1101. (i) Paxson, T. E.;
Reilly, C. A.; Holecek, D. R. J. Chem. Soc., Chem. Commun. 1981, 618.
(j) Knifton, J. F. J. Chem. Soc., Chem. Commun. 1981, 188. (k) Backvall,
J. E.; Akermark, B.; Ljunggren, S. O. J. Am. Chem. Soc. 1979, 101, 2411.
(l) Roth, J. A.; Orchin, M. J. Organomet. Chem. 1979, 172, C27.
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.organo-
met.7b00158.
Experimental methods, formation and isolation details,
NMR monitoring details and spectra, HRMS spectra, X-
ray crystallography data, and DFT calculations (PDF)
Crystallographic information files for 3, 4, and 7 (CIF)
Cartesian coordinates for 4 (XYZ)
AUTHOR INFORMATION
■
Corresponding Authors
*E-mail: t.kraemer@hw.ac.uk.
*E-mail: rowan.young@nus.edu.sg.
(m) Sisak, A.; Sam
Ungvary, F.; Palyi, G. Organometallics 1989, 8, 1096.
(7) (a) Vaughn, G. D.; Gladysz, J. A. J. Am. Chem. Soc. 1981, 103
(18), 5608. (b) Garralda, M. A.; Hernandez, R.; Ibarlucea, L.; Pinilla,
́ ́ ́ ́
par-Szerencses, E.; Galamb, V.; Nemeth, L.;
́
́
ORCID
́
Simon Sung: 0000-0002-7864-7694
Stuart A. Macgregor: 0000-0003-3454-6776
E.; Torres, M. R.; Zarandona, M. Organometallics 2007, 26, 1031.
(c) Van Voorhees, S. L.; Wayland, B. B. Organometallics 1985, 4, 1887.
(d) Fu, X.; Basickes, L.; Wayland, B. B. Chem. Commun. 2003, 520.
(e) Fu, X.; Wayland, B. B. J. Am. Chem. Soc. 2005, 127, 16460.
Tobias Kramer: 0000-0001-5842-9553
̈
Rowan D. Young: 0000-0001-7437-8944
H
DOI: 10.1021/acs.organomet.7b00158
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
(f) Vaughn, G. D.; Strouse, C. E.; Gladysz, J. A. J. Am. Chem. Soc. 1986,
108, 1462. (g) Vaughn, G. D.; Gladysz, J. A. J. Am. Chem. Soc. 1986,
108, 1473. (h) Brockaart, G.; El Mail, R.; Garralda, M. A.; Hernan
R.; Ibarlucea, L.; Santos, J. I. Inorg. Chim. Acta 2002, 338, 249. (i) El
Mail, R.; Garralda, M. A.; Hernandez, R.; Ibarlucea, L.; Pinilla, E.;
́
dez,
́
Torres, M. R. Organometallics 2000, 19, 5310.
(8) For current state-of-the-art, see (a) Nguyen, K. D.; Herkommer,
D.; Krische, M. J. J. Am. Chem. Soc. 2016, 138, 14210. (b) Geary, L.
M.; Glasspoole, B. W.; Kim, M. M.; Krische, M. J. J. Am. Chem. Soc.
2013, 135, 3796. (c) Patman, R. L.; Chaulagain, M. R.; Williams, V.
M.; Krische, M. J. J. Am. Chem. Soc. 2009, 131, 2066.
(9) Jing, Q.; Sandoval, C. A.; Wang, Z.; Ding, K. Eur. J. Org. Chem.
2006, 2006, 3606.
(10) Nakamura, Y.; Yoshikai, N.; Ilies, L.; Nakamura, E. Org. Lett.
2012, 14, 3316.
(11) Saes, B. W. H.; Verhoeven, D. G. A.; Lutz, M.; Klein Gebbink,
R. J. M.; Moret, M.-E. Organometallics 2015, 34, 2710.
(12) (a) Doyle, L. E.; Piers, W. E.; Borau-Garcia, J. J. Am. Chem. Soc.
2015, 137, 2187. (b) Doyle, L. E.; Piers, W. E.; Borau-Garcia, J.; Sgro,
M. J.; Spasyuk, D. M. Chem. Sci. 2016, 7, 921.
(13) For examples, see (a) Bruce, M. I.; Hambley, T. W.; Snow, M.
R.; Swincer, A. G. J. Organomet. Chem. 1982, 235, 105. (b) Akkerman,
F. A.; Lentz, D. Angew. Chem., Int. Ed. 2007, 46, 4902. (c) Choi, J.-C.;
Sarai, S.; Koizumi, T.; Osakada, K.; Yamamoto, T. Organometallics
1998, 17 (10), 2037. (d) Nishihara, Y.; Yoda, C.; Itazaki, M.; Osakada,
K. Bull. Chem. Soc. Jpn. 2005, 78, 1469. (d1) Bianchini, C.; Meli, A.;
Peruzzini, M.; Vizza, F.; Frediani, P.; Ramirez, J. A. Organometallics
́ ́
1990, 9, 226. (e) Tejel, C.; Geer, A. M.; Jimenez, S.; Lopez, J. A.;
Ciriano, M. A. Organometallics 2012, 31, 2895. (f) Carr, S. W.; Shaw,
B. L.; Thornton-Pett, M. J. Chem. Soc., Dalton Trans. 1987, 1763.
(14) (a) van der Boom, M. E.; Zubkov, T.; Shukla, A. D.;
Rybtchinski, B.; Shimon, L. J. W.; Rozenberg, H.; Ben-David, Y.;
Milstein, D. Angew. Chem., Int. Ed. 2004, 43, 5961. (b) Daugulis, O.;
Brookhart, M. Organometallics 2004, 23, 527. (c) Pawley, R. J.;
Huertos, M. A.; Lloyd-Jones, G. C.; Weller, A. S.; Willis, M. C.
Organometallics 2012, 31, 5650.
(15) (a) Moxham, G. L.; Randell-Sly, H.; Brayshaw, S. K.; Weller, A.
S.; Willis, M. C. Chem. - Eur. J. 2008, 14, 8383. (b) Arras, J.; Speth, H.;
Mayer, H. A.; Wesemann, L. Organometallics 2015, 34, 3629.
(c) Barthes, C.; Lepetit, C.; Canac, Y.; Duhayon, C.; Zargarian, D.;
Chauvin, R. Inorg. Chem. 2013, 52, 48. (d) Lepetit, C.; Poater, J.;
̀
Alikhani, M. E.; Silvi, B.; Canac, Y.; Contreras-García, J.; Sola, M.;
Chauvin, R. Inorg. Chem. 2015, 54, 2960. (e) Lesueur, W.; Solari, E.;
Floriani, C.; Chiesi-Villa, A.; Rizzoli, C. Inorg. Chem. 1997, 36, 3354.
(f) Logan, J. R.; Piers, W. E.; Borau-Garcia, J.; Spasyuk, D. M.
Organometallics 2016, 35, 1279.
(16) Scherer, W.; Dunbar, A. C.; Barquera-Lozada, J. E.; Schmitz, D.;
Eickerling, G.; Kratzert, D.; Stalke, D.; Lanza, A.; Macchi, P.; Casati, N.
P. M.; Ebad-Allah, J.; Kuntscher, C. Angew. Chem., Int. Ed. 2015, 54,
2505.
(17) Kozuch, S.; Shaik, S. Acc. Chem. Res. 2011, 44, 101.
(18) Zhang, Y.; Lewis, J. C.; Bergman, R. G.; Ellman, J. A.; Oldfield,
E. Organometallics 2006, 25, 3515.
(19) Goodman, J.; Grushin, V. V.; Larichev, R. B.; Macgregor, S. A.;
Marshall, W. J.; Roe, D. C. J. Am. Chem. Soc. 2010, 132, 12013.
(20) (a) Minenkov, Y.; Occhipinti, G.; Jensen, V. R. J. Phys. Chem. A
2009, 113, 11833. (b) Zhao, Y.; Truhlar, D. G. Org. Lett. 2007, 9,
1967. (c) Sieffert, N.; Buhl, M. Inorg. Chem. 2009, 48, 4622.
̈
(d) McMullin, C. L.; Jover, J.; Harvey, J. N.; Fey, N. Dalton Trans.
2010, 39, 10833. (e) Ahlquist, M. S. G.; Norrby, P.-O. Angew. Chem.,
Int. Ed. 2011, 50, 11794.
(21) (a) Ahmad, N.; Levison, J. J.; Robinson, S. D.; Uttley, M. F.;
Wonchoba, E. R.; Parshall, G. W. Inorg. Synth. 1990, 28, 81−83.
(b) Jing, Q.; Sandoval, C. A.; Wang, Z.; Ding, K. Eur. J. Org. Chem.
2006, 2006, 3606−3616. (c) Nakamura, Y.; Yoshikai, N.; Ilies, L.;
Nakamura, E. Org. Lett. 2012, 14, 3316−3319.
I
DOI: 10.1021/acs.organomet.7b00158
Organometallics XXXX, XXX, XXX−XXX