Synthesis and reactivity of [M(η3-allyl)(η2-amidinato)(CO)2(phosphonium ylide)] (M?=?Mo, W): Investigation of the ligand properties of phosphonium ylides

Article abstract of DOI:10.1016/j.ica.2017.11.019

Phosphonium ylide complexes of Mo and W formulated as [M(η³-allyl){η²-(NPh)₂CH}(CO)₂(CH₂PR₃)] (M = Mo, R = Me: 2a-Mo; M = Mo, R = Ph: 2b-Mo, and M = W, R = Me: 2a-W) were prepared by the reaction of amidinato(pyridine) complex, [M(η³-allyl){η²-(NPh)₂CH}(CO)₂(NC₅H₅)] (M = Mo: 1-Mo and M = W: 1-W), with a phosphonium ylide, CH₂PR₃ (R = Me, Ph), which was generated in situ by the reaction of the corresponding phosphonium salt with ⁿBuLi. These complexes were characterized spectroscopically, as well as by the X-ray diffraction. The phosphonium ylide ligand shows stronger electron donating ability toward the metal than N-heterocyclic carbene or phosphine ligands. This trend is supported by the comparison of the spectroscopic data and the DFT calculations. We also investigated the reactivity of the phosphonium ylide complexes 2-Mo with two-electron donors such as PEt₃ and NHC. In the case of the PPh₃ ylide complex (2b-Mo), the substitution reaction of the ylide ligand for the two-electron donors took place cleanly to yield the corresponding complexes. On the other hand, in the PMe₃ ylide complex (2a-Mo), the substituted complexes formed but the unreacted ylide complex 2a-Mo was also present in the reaction mixture. These results show that the bond strength of the M-C(phosphonium ylide) bond is affected by the substituents on the phosphorus atom.

Full text of DOI:10.1016/j.ica.2017.11.019

Inorganica Chimica Acta 471 (2018) 310–315
Contents lists available at ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
Research paper
Synthesis and reactivity of [M(g g
3-allyl)( 2-amidinato)
(CO)2(phosphonium ylide)] (M = Mo, W): Investigation of the ligand
properties of phosphonium ylides
⇑
Daichi Takaki, Kenichi Ogata, Youji Kurihara, Kazuyoshi Ueda, Toru Hashimoto, Yoshitaka Yamaguchi
Department of Advanced Materials Chemistry, Graduate School of Engineering, Yokohama National University, 79-5 Tokiwadai, Hodogaya-ku, Yokohama 240-8501, Japan
a r t i c l e i n f o
a b s t r a c t
Phosphonium ylide complexes of Mo and W formulated as [M(g g
3-allyl){ 2-(NPh)2CH}(CO)2(CH2PR3)] (M
Article history:
Received 23 September 2017
Received in revised form 9 November 2017
Accepted 11 November 2017
= Mo, R = Me: 2a-Mo; M = Mo, R = Ph: 2b-Mo, and M = W, R = Me: 2a-W) were prepared by the reaction
of amidinato(pyridine) complex, [M(
g
3-allyl){ 2-(NPh)2CH}(CO)2(NC5H5)] (M = Mo: 1-Mo and M = W: 1-
g
W), with a phosphonium ylide, CH2PR3 (R = Me, Ph), which was generated in situ by the reaction of the
corresponding phosphonium salt with nBuLi. These complexes were characterized spectroscopically, as
well as by the X-ray diffraction. The phosphonium ylide ligand shows stronger electron donating ability
toward the metal than N-heterocyclic carbene or phosphine ligands. This trend is supported by the com-
parison of the spectroscopic data and the DFT calculations. We also investigated the reactivity of the
phosphonium ylide complexes 2-Mo with two-electron donors such as PEt3 and NHC. In the case of
the PPh3 ylide complex (2b-Mo), the substitution reaction of the ylide ligand for the two-electron donors
took place cleanly to yield the corresponding complexes. On the other hand, in the PMe3 ylide complex
(2a-Mo), the substituted complexes formed but the unreacted ylide complex 2a-Mo was also present in
the reaction mixture. These results show that the bond strength of the M-C(phosphonium ylide) bond is
affected by the substituents on the phosphorus atom.
Keywords:
Phosphonium ylide
Electron donating ability
Bond strength
DFT calculations
Ó 2017 Elsevier B.V. All rights reserved.
1. Introduction
est in the field of homogeneous catalysts. Recently, phosphonium
ylide-based hybrid-type ligands, i.e., bidentate ylide-phosphine
Electron rich metal centers in metal complexes play an impor-
tant role in the control of metal-mediated processes, namely, catal-
ysis. Therefore, the development of transition metal complexes
bearing strongly electron-donating ligands has attracted much
attention from academia, as well as industry. N-Heterocyclic carbe-
nes (NHCs) are now fully established as an important class of
ligands for organometallic chemistry because they are known to
enhance the electron density at the metal center with a thermally
robust metal-ligand bond [1]. NHCs are categorized as unsaturated
sp2-hybridized carbon coordinated ligands. In comparison with an
NHC ligand, saturated sp3-hybridized carbon ligands, i.e., a phos-
phonium ylide, which can be considered as a phosphine-stabilized
carbene ligand, have been recognized as strong electron donors
toward the metal because their metal complexes have a canonical
zwitterionic structure. Thus, negative and positive charges are
accumulated at the metal center and the phosphorus atom, respec-
tively [2,3]. Therefore, phosphonium ylides have drawn great inter-
and -NHC ligands have been reported, and these complexes act
as effective catalysts for C-C and C-N forming reactions [4] and
hydrosilylation and hydrogenation reactions [5]. Furthermore,
Chauvin and co-workers have reported systematic investigations
on the coordinating properties of phosphonium ylide and NHC
ligands on the late transition metal complexes through C-C chelat-
ing ligands containing two moieties [6]. They have concluded that
phosphonium ylides act as stronger electron-donating ligands than
NHCs. The electronic properties of these chelating ligands are sup-
ported by the DFT calculations [7].
We are interested in the synthesis, structure, and reactivity of
group 6 transition metal complexes, especially molybdenum and
tungsten. Furthermore, we have already reported the preparation
and explored some of the reactivity of amidinato(pyridine) com-
plexes formulated as [M(g g
3-allyl)( 2-amidinato)(CO)2(pyridine)]
(M = Mo; 1-Mo, W; 1-W) [8]. Complex 1 has a labile pyridine
ligand, and, thus, the substitution reaction with two-electron
donors such as PR3 or NHC takes place smoothly to afford the cor-
responding complexes [8a,9]. The complexes in this series of
amidinato complexes have two CO ligands and, thus, the
⇑
Corresponding author.
E-mail address: yamaguchi-yoshitaka-hw@ynu.ac.jp (Y. Yamaguchi).
https://doi.org/10.1016/j.ica.2017.11.019
0020-1693/Ó 2017 Elsevier B.V. All rights reserved.

D. Takaki et al. / Inorganica Chimica Acta 471 (2018) 310–315
311
comparison of the CO stretching frequencies of the complexes
formed is an effective probe for the electron donating ability of
the coordinating ligands. We are interested in not only the
electron donating ability of phosphonium ylide but also the bond
strength of the M-C(phosphonium ylide) bond. In this paper, we
report the preparation of phosphonium ylide complexes of Mo
and W and the electronic properties of phosphonium ylides
comparing to NHC and PR3. Furthermore, we have estimated the
influence of the substituents of the phosphorus moiety on the
bond strength of the M-C(phosphonium ylide) bond by reaction
with two-electron donors such as NHC and PR3.
2H, allyl-CHH), 1.84 (d, J = 13.2 Hz, 9H, PMe3), 3.47 (d, J = 6.6 Hz,
2H, allyl-CHH), 3.87 (tt, J = 9.9, 6.6 Hz, 1H, allyl-CH), 6.84–7.31
(m, 10H, Ph), 8.81 (s, 1H, amidinato-CH). 13C{1H} NMR (CDCl3): d
15.2 (d, J = 55.5 Hz, PMe3), 17.9 (d, J = 36.0 Hz, CH2PMe3), 57.6 (s,
allyl-CH2), 81.9 (s, allyl-CH), 117.5 (s, Ph), 120.8 (s, Ph), 129.1 (s,
Ph), 146.6 (s, Ph), 151.8 (s, NCN), 233.3 (d, J = 3.7 Hz, CO). 31P{1H}
NMR (CDCl3): d 30.4 (s).
2.2.2. Preparation of [Mo(g g
3-allyl){ 2-(NC6H5)2CH}(CO)2(CH2PPh3)]
(2b-Mo)
Complex 2b-Mo was prepared from 1-Mo (57 mg, 0.12 mmol)
and a THF solution of CH2PPh3, which was prepared by the reaction
of [PMePh3]Br (49 mg, 0.14 mmol) with nBuLi (0.10 mL of its 1.5 M
hexane solution, 0.15 mmol) at –78 °C, in the same manner as that
for 2a-Mo. 2b-Mo was isolated as a yellow powder (48 mg, 0.072
mmol, 60%). Anal. Calcd for C37H33MoN2O2P: C, 66.87; H, 5.00; N,
2. Experimental
2.1. General procedures
4.22. Found: C, 66.34; H, 4.78; N, 4.27%. IR (KBr):
m (CO) 1909,
All manipulations involving air- and moisture-sensitive
organometallic compounds were carried out under an atmosphere
of dry nitrogen, which had been purified by SICAPENT (Merck Co.,
Inc.), using standard Schlenk tube or high vacuum techniques. All
solvents were distilled over appropriate drying agents before use.
[PMe4]Br (Tokyo Chemical Industry), [PMePh3]Br (Sigma-Aldrich),
and [PMe4]I (Alfa Aesar) were purchased and used without further
purification. Other reagents employed in this investigation were
commercially available and used without further purification. [M
1820 cmÀ1 1H NMR (CDCl3): d 1.06 (d, J = 13.8 Hz, 2H, CH2PPh3),
.
1.12 (d, J = 9.9 Hz, 2H, allyl-CHH), 3.49 (d, J = 6.6 Hz, 2H, allyl-
CHH), 3.85 (tt, J = 9.9, 6.6 Hz, 1H, allyl-CH), 6.88–7.79 (m, 25H,
Ph), 8.98 (s, 1H, amidinato-CH). 13C{1H} NMR (CDCl3): d 12.1 (d, J
= 30.5 Hz, CH2PPh3), 58.0 (s, allyl-CH2), 81.1 (s, allyl-CH), 117.9 (s,
Ph), 120.9 (s, Ph), 128.9 (s, Ph), 129.1 (s, Ph), 129.2 (d, J = 16.8 Hz,
Ph), 132.4 (d, J = 3.1 Hz, Ph), 133.3 (d, J = 9.1 Hz, Ph), 146.9 (s,
Ph), 152.5 (s, NCN), 230.7 (d, J = 3.1 Hz, CO). 31P{1H} NMR (CH2Cl2):
d 36.4 (s).
(g g
3-allyl){ 2-(NC6H5)2CH}(CO)2(NC5H5)] (M = Mo: 1-Mo; M = W:
1-W) [8a] and IiPrÁBEt3 (IiPr = 1,3-diisopropyllimidazol-2-ylidene)
[10] were prepared according to literature methods.
2.2.3. Preparation of [W(g g
3-allyl){ 2-(NC6H5)2CH}(CO)2(CH2PMe3)]
(2a-W)
The IR spectra were recorded on a HORIBA FT-730 spectrome-
ter. 1H, 13C{1H} and 31P{1H} NMR spectra were recorded on JEOL
EX-270 and BRUKER DRX-300 spectrometers at ambient tempera-
ture. 1H and 13C{1H} NMR chemical shifts were recorded in ppm
relative to Me4Si as an internal standard. The 31P{1H} NMR chem-
ical shifts were recorded in ppm relative to H3PO4 as an external
standard. All coupling constants were recorded in hertz. The mul-
tiplicity is indicated by s (singlet), d (doublet), tt (triplet of triplets),
and m (multiplet). Cyclic voltammograms were recorded on a Huso
Electrochemical System, which is a combination of a potential
coulometry timer (317S), potential scanning unit (321), and digital
universal signal processing unit (326), in CH3CN containing 0.1 M
of nBu4NBF4 as a supporting electrolyte and using a conventional
three-electrode system. In this system, platinum was used as the
working electrode, a platinum wire was used as the counter elec-
trode, and Ag/AgCl was used as the reference electrode. The scan
rate was 100 mV/s. Potentials are given vs. Fc/Fc+. Elemental anal-
yses were performed on a Perkin-Elmer 240C.
Complex 2a-W was prepared from 1-W (83 mg, 0.15 mmol) and
a THF solution of CH2PMe3, which was prepared by the reaction of
[PMe4]I (35 mg, 0.16 mmol) with nBuLi (0.12 mL of its 1.5 M hex-
ane solution, 0.18 mmol) at À78 °C, in the same manner as that
for 2a-Mo. 2a-W was isolated as a yellow powder (45 mg, 0.079
mmol, 53%). Anal. Calcd for C22H27N2O2PW: C, 46.66; H, 4.81; N,
4.95. Found: C, 46.41; H, 4.81, N, 4.78%. IR (KBr):
m (CO) 1904,
1789 cmÀ1 1H NMR (CDCl3): d 0.16 (d, J = 14.5 Hz, 2H, CH2PMe3),
.
1.54 (d, J = 9.2 Hz, 2H, allyl-CHH), 1.87 (d, J = 12.5 Hz, 9H, PMe3),
2.99 (tt, J = 9.9, 6.6 Hz, 1H, allyl-CH), 3.39 (d, J = 6.6 Hz, 2H, allyl-
CHH), 6.87–7.33 (m, 10H, Ph), 9.47 (s, 1H, amidinato-CH). 13C{1H}
NMR (CDCl3): d 15.2 (d, J = 54.8 Hz, PMe3), 23.5 (d, J = 36.9 Hz, CH2-
PMe3), 50.3 (s, allyl-CH2), 76.1 (s, allyl-CH), 117.5 (s, Ph), 121.2 (s,
Ph), 129.0 (s, Ph), 145.2 (s, Ph), 151.7 (s, NCN), 226.6 (d, J = 3.4
Hz, CO). 31P{1H} NMR (CDCl3): d 30.4 (s).
2.3. The reaction of phosphonium ylide complex 2 with two-electron
donors
2.2. Preparation of phosphonium ylide complexes (2)
2.3.1. The reaction of 2b-Mo with IiPrÁBEt3: Isolation of complex 3
Complex 2b-Mo (80 mg, 0.12 mmol), IiPrÁBEt3 (31 mg, 0.12
mmol), and toluene (10 mL) were placed in a Schlenk tube. After
being refluxed for 1 h, complex 3 was isolated in the same manner
as reported previously (45 mg, 0.083 mmol, 69%). The product was
characterized by comparison with the reported 1H NMR data [9].
2.2.1. Preparation of [Mo(g g
3-allyl){ 2-(NC6H5)2CH}(CO)2(CH2PMe3)]
(2a-Mo)
A solution of complex 1-Mo (383 mg, 0.82 mmol) in tetrahydro-
furan (THF) (5 mL) was cooled to –78 °C and then a THF solution of
the CH2PMe3 phosphonium ylide, which was prepared by the reac-
tion of [PMe4]Br (139 mg, 0.81 mmol) with nBuLi (0.51 mL of its
1.6 M hexane solution, 0.82 mmol) at –78 °C, was added. Subse-
quently, the reaction mixture was allowed to warm to room tem-
perature. After 2 h, the volatiles were removed under reduced
pressure. The residual solid was extracted with CH2Cl2 (20 mL Â
1 and 10 mL Â 2) and the filtrate was evaporated. The solid was
washed with MeOH (5 mL Â 3), and then dried in vacuo to give
2a-Mo as a yellow powder (229 mg, 0.48 mmol, 59%). Anal. Calcd
for C22H27MoN2O2P: C, 55.24; H, 5.69; N, 5.86. Found: C, 55.51;
2.3.2. The reaction of 2b-Mo with PEt3: Isolation of complex 4
PEt3 (20 lL, 16 mg, 0.14 mmol) was added to a solution of com-
plex 2b-Mo (80 mg, 0.12 mmol) in toluene (5 mL) at room temper-
ature. After being refluxed for 1 h, complex 4 was isolated in the
same manner as reported previously (54 mg, 0.11 mmol, 92%).
The product was characterized by comparison with the reported
1H NMR data [11].
2.3.3. The reaction of 2a-Mo with IiPrÁBEt3: Formation of complex 3
Complex 2a-Mo (103 mg, 0.22 mmol), IiPrÁBEt3 (55 mg, 0.22
mmol), and toluene (10 mL) were placed in a Schlenk tube. After
H, 5.48, N, 5.80%. IR (KBr):
(CDCl3): d 0.19 (d, J = 13.2 Hz, 2H, CH2PMe3), 1.36 (d, J = 9.9 Hz,
m .
(CO) 1912, 1798 cmÀ1 1H NMR

312
D. Takaki et al. / Inorganica Chimica Acta 471 (2018) 310–315
Table 1
being refluxed for 1 h, the volatiles were removed under reduced
pressure. The residual solid was washed with MeOH and then dried
in vacuo to give complexes 3 and 2a-Mo as a mixture (total yield,
93 mg). The yields of each complex were determined by the proton
integrations based on the 1H NMR spectrum (3; 34 mg, 0.063
mmol, 29% and 2a-Mo; 59 mg, 0.12 mmol, 55%).
Summary of crystal data for complexes 2a-Mo and 2b-Mo.
2a-Mo
2b-Mo
Empirical formula
Formula weight
Crystal color, habit
Crystal size/mm
Crystal system
Space group
Lattice parameters
a (Å)
C
22H27MoN2O2P
C37H33MoN2O2P
664.59
yellow, needle
0.25 Â 0.13 Â 0.13
monoclinic
478.38
yellow, prismatic
0.38 Â 0.25 Â 0.25
monoclinic
P21/c (No. 14)
P21/c (No. 14)
2.3.4. The reaction of 2a-Mo with PEt3: Formation of complex 4
PEt3 (48 lL, 38 mg, 0.32 mmol) was added to a solution of com-
16.490(10)
9.309(7)
16.457(10)
119.85(4)
2191(3)
4
1.450
984.00
6.898
11.003(6)
14.397(10)
20.342(4)
102.12(3)
3151(3)
4
1.401
1368.00
5.021
plex 2a-Mo (150 mg, 0.31 mmol) in toluene (10 mL) at room tem-
perature. After being refluxed for 1 h, the volatiles were removed
under reduced pressure. The residual solid was then subjected to
1H NMR measurements. The ratio of 4:2a-Mo was estimated to
be ca. 1:4. After washing the residual solid with MeOH, complex
2a-Mo was recovered as a yellow solid (83 mg, 0.17 mmol, 55%).
b (Å)
c (Å)
b (°)
V (Å3)
Z
Dc (g cmÀ3
)
F000
l
(Mo-Ka )
) (cmÀ1
Reflections measured
Independent reflections
Rint
No. variables
Reflection/parameter ratio
Residuals: R; wR2
5211
5030
0.0409
256
19.65
0.0577; 0.0891
0.0333
1.028
0.70, À0.54
7606
7239
0.0453
388
18.66
0.1073; 0.1752
0.0579
1.004
1.87, À1.21
2.4. DFT calculations
Density functional theory (DFT) [12] calculations were per-
formed with the Becke’s hybrid three-parameter exchange func-
tional [13] with the Lee-Yang-Parr nonlocal correlation functional
[14] (B3LYP) as implemented in the Gaussian 03 program package
[15]. A double-zeta valence plus polarization (DGDZVP) basis set
[16] was used for all atoms. All geometries were optimized without
constraints, and vibrational frequencies were computed. The
molybdenum atomic charges were estimated using the Mulliken
charges [17], natural population analysis (NPA) [18], and atomic
polar tensor (APT) [19].
Residuals: R1 [I > 2.00
Goodness-of-fit (GOF) on F2
qmax,min (eÀ ÅÀ3
r(I)]
d
)
On treatment of molybdenum complex 1-Mo with CH2PMe3,
which was prepared in situ by the reaction of [PMe4]Br with nBuLi
in THF, the PMe3 ylide complex 2a-Mo was isolated as a yellow
powder in 59% yield. The PPh3 ylide complex of Mo (2b-Mo) and
the tungsten analog of 2a-Mo (2a-W) were also obtained in mod-
erate yield (Scheme 1).
2.5. Experimental procedure for X-ray crystallography
In the IR spectra, the molybdenum complexes showed two CO
stretching bands at 1912 and 1798 cmÀ1 for 2a-Mo and 1909 and
1820 cmÀ1 for 2b-Mo. These results imply that the PMe3 ylide is
a better electron donor than the PPh3 ylide. Tungsten complex
2a-W also showed two CO stretching bands (1904 and 1789
cmÀ1), which are at lower frequencies than those for 2a-Mo. These
observations indicate that two carbonyl ligands are in mutually cis
positions [23]. In the 31P{1H} NMR spectra of complexes 2, only one
singlet was observed, and these were all found in a similar region
(30.4 ppm for 2a-Mo, 36.4 ppm for 2b-Mo, and 30.4 ppm for 2a-
W). The 1H NMR spectrum of 2a-Mo showed symmetrical allylic
signals at 1.36 (anti-CH2), 3.47 (syn-CH2), and 3.87 (central proton)
ppm. The doublet signal assignable to the CH2 moiety of the ylide
ligand was observed at 0.19 ppm with a coupling constant of 13.2
Hz. The doublet signal assignable to the proton atoms of PMe3 was
Suitable single crystals (2a-Mo and 2b-Mo) were obtained by
recrystallization from CH2Cl2/ether and were individually mounted
on glass fibers. Diffraction measurements of 2a-Mo and 2b-Mo
were carried out on a Rigaku AFC-7R automated four-circle diffrac-
tometer by using graphite-monochromated Mo K
0.71069 Å). The data collections were carried out at À50 1 °C
using the -2h scan technique to a maximum 2h value of 55.0°
a radiation (k =
x
for 2a-Mo and 60.0° for 2b-Mo. Cell constants and an orientation
matrix for data collection were determined from 25 reflections
with 2h angles in the range 29.54–29.99° for 2a-Mo and 27.94–
29.86° for 2b-Mo. Three standard reflections were monitored every
150 measurements. In the data reduction, Lorentz and polarization
corrections and an empirical absorption correction (
made.
wscan) were
Crystallographic data and the results of measurements are sum-
marized in Table 1. The structures were solved by direct methods
(SHELXT) [20] for 2a-Mo and 2b-Mo and expanded using Fourier
techniques. Least-square refinements were carried out using
SHELXL [21]. All non-hydrogen atoms were refined anisotropically.
Hydrogen atoms were introduced at the ideal position and were
refined using the riding model. All calculations were performed
using the CrystalStructure crystallographic software package [22].
observed at 1.84 ppm with a coupling constant of 13.2 Hz. The 13
C
{1H} NMR spectrum of 2a-Mo also indicates that complex 2a-Mo
has a symmetrical geometry around the Mo center; that is, allylic
carbon signals appeared at 57.6 ppm (terminal carbons) and 81.9
ppm (central carbon), and a doublet signal arising from two CO car-
bon atoms was seen at 233.3 ppm with a coupling constant of 3.7
Hz. The CH2 carbon atom in the ylide ligand was observed at 17.9
3. Results and discussion
3.1. Synthesis and characterization of phosphonium ylide complexes of
Mo and W
We have already reported that the pyridine ligand in the amid-
inato(pyridine) complex 1 is easily substituted by two-electron
donors such as PR3 and NHC to afford the corresponding complexes
[8a,9]. Therefore, we examined the reaction of the complex 1 with
phosphonium ylides.
Scheme 1. Preparation of phosphonium ylide complexes of Mo and W.

D. Takaki et al. / Inorganica Chimica Acta 471 (2018) 310–315
313
Table 2
ppm as a doublet signal (JPC = 36.0 Hz). These NMR data suggested
Selected bond lengths (Å) and angles (°) for 2a-Mo and 2b-Mo.
that the phosphonium ylide ligand of complex 2a-Mo was trans to
the allyl ligand in solution. Complexes 2b-Mo and 2a-W showed
similar spectroscopic features in their NMR spectra to those of
complex 2a-Mo.
2a-Mo
2b-Mo
Bond lengths
Mo1-N1
Mo1-N2
N1-C1
2.274(3)
2.253(3)
1.322(3)
1.315(5)
1.960(4)
1.156(5)
1.941(3)
1.167(3)
2.277(3)
1.746(3)
1.783(4)
1.795(5)
1.796(4)
Mo1-N1
Mo1-N2
N1-C1
2.301(4)
2.248(4)
1.332(6)
1.305(6)
1.952(4)
1.162(6)
1.946(5)
1.164(6)
2.305(5)
1.765(5)
1.799(5)
1.821(5)
1.786(5)
The structure of complexes 2a-Mo and 2b-Mo were determined
by X-ray analysis. The ORTEP drawings 2a-Mo and 2b-Mo are
shown in Fig. 1. Selected bond lengths and angles are listed in
Table 2. These complexes show pseudo-octahedral geometry
around the Mo center. The amidinato ligand is located at an equa-
torial position and is coplanar with two CO ligands. The phospho-
N2-C1
N2-C1
Mo1-C14
O1-C14
Mo1-C15
O2-C15
Mo1-C16
P1-C16
P1-C17
P1-C18
P1-C19
Mo1-C14
O1-C14
Mo1-C15
O2-C15
Mo1-C16
P1-C16
P1-C17
P1-C23
P1-C29
nium ylide ligand is at an axial position trans to the
The open face of the allyl ligand is directed toward two CO ligands,
which is known to be a favorable orientation for a series of [M(g3
g
3-allyl ligand.
-
allyl)(CO)2] (M = Mo, W) complexes [24]. The geometry of these
complexes revealed by X-ray analysis coincides with the solu-
tion-state geometry indicated by the 1H and 13C{1H} NMR spectra.
The Mo1–C16 bond length for 2a-Mo (2.277(3) Å) is slightly
shorter than that for 2b-Mo (2.305(5) Å). The P1–C16 bond length
for 2a-Mo (1.746(3) Å) is also slightly shorter than that for 2b-Mo
(1.765(5) Å). The Mo1–C16–P1 angle for 2a-Mo (121.09(14)°) is
smaller than that for 2b-Mo (124.8(2)°). Although the electronic
influence of the phosphine cannot be ruled out, these structural
features might arise from the steric hindrance of the phosphine
of the ylide ligand, i.e., the cone angles of these phosphines are
118° for PMe3 and 145° for PPh3 [25]. Additionally, in complex
2a-Mo, the P1–C16 bond length (1.746(3) Å) is slightly shorter
than those of the P1–C(Me) bond lengths (1.783(4)–1.796(4) Å),
whereas the P1–C16 bond is longer than the reported P = C double
bond lengths (1.640(6) Å for Me3P = CH2 [26] and 1.661(8) Å for
Ph3P = CH2 [27]). A similar trend was observed in the P-C bond
lengths in 2b-Mo. The orientation of the ylide ligand is somewhat
interesting. The torsion angles of C15–Mo1–C16–P1 are very small:
À0.3(2) ° for 2a-Mo and 12.3(3)° for 2b-Mo. The Mo1–C16–P1
planes in both complexes do not bisect the C14–Mo1–C15 angle
in the solid state.
Bond angles
N1-Mo1-N2
C14-Mo1-C15
N1-Mo1-C14
N1-Mo1-C15
N2-Mo1-C14
N2-Mo1-C15
N1-Mo1-C16
N2-Mo1-C16
C14-Mo1-C16
C15-Mo1-C16
Mo1-C14-O1
Mo1-C15-O2
Mo1-N1-C1
Mo1-N1-C2
C1-N1-C2
58.16(11)
78.33(14)
164.10(9)
113.74(14)
108.56(12)
169.89(14)
80.96(12)
78.47(10)
87.97(14)
94.68(12)
177.4(3)
176.4(3)
93.8(2)
142.50(16)
123.7(3)
94.96(16)
140.6(2)
N1-Mo1-N2
C14-Mo1-C15
N1-Mo1-C14
N1-Mo1-C15
N2-Mo1-C14
N2-Mo1-C15
N1-Mo1-C16
N2-Mo1-C16
C14-Mo1-C16
C15-Mo1-C16
Mo1-C14-O1
Mo1-C15-O2
Mo1-N1-C1
Mo1-N1-C2
C1-N1-C2
57.93(14)
79.76(19)
165.98(17)
111.22(16)
109.91(17)
166.93(17)
81.71(15)
77.68(17)
88.99(18)
94.12(19)
178.1(4)
178.2(4)
92.7(3)
144.8(3)
122.5(4)
95.9(3)
141.2(3)
122.8(4)
113.3(4)
124.8(2)
Mo1-N2-C1
Mo1-N2-C8
C1-N2-C8
N1-C1-N2
Mo1-C16-P1
Mo1-N2-C1
Mo1-N2-C8
C1-N2-C8
N1-C1-N2
Mo1-C16-P1
124.3(3)
113.1(3)
121.09(14)
showed lower CO stretching frequencies than those of others (3
and 4). This result reflects the fact that the negative charge is accu-
mulated at the Mo center in the phosphonium ylide complexes.
Furthermore, we found that, in the phosphonium ylide complexes,
the substituents on the phosphorus atom have an influence on the
electron donating ability toward the metal center, i.e., the PMe3
ylide is a better electron donor than the PPh3 ylide. To obtain fur-
ther information concerning electronic influence on the Mo center,
we made CV measurements of complexes 2a-Mo, 3 and 4 in a CH3-
CN solution. All complexes exhibited irreversible oxidation waves.
The CV of ylide complex 2a-Mo shows the one-oxidation wave at
À0.05 V vs. Fc/Fc+. This oxidation potential is lower than those of
complexes 3 (0.09 V) and 4 (0.19 V). The second oxidation wave
for 2a-Mo was observed at 0.31 V, which is also lower potential
than those of 3 (0.58 V) and 4 (0.69 V). Based on the oxidation
potentials in those complexes, it is conceivable that the electron
density accumulated at the Mo center increases in the order 4 <
3 < 2a-Mo. We assumed that the highest occupied molecular orbi-
tal (HOMO) levels of these complexes also increase in this order.
Next, we discuss the DFT calculations of the series of Mo
complexes.
3.2. Electronic property of the phosphonium ylides as ligands:
Estimation by experimental and theoretical studies
To estimate the electron donating ability of the phosphonium
ylide as a ligand, we examined a comparison of CO stretching
frequencies of a series of [Mo(g g
3-allyl){ 2-(NC6H5)2CH}(CO)2(L)]
(L = CH2PMe3 (2a-Mo), CH2PPh3 (2b-Mo), IiPr (IiPr = 1,3-diiso-
propylimidazol-2-ylidene) (3), and PEt3 (4)) and the oxidation
potentials of these complexes by cyclic voltammetry (CV). DFT cal-
culations were also carried out. These results are summarized in
Table 3.
In the IR spectra, the mean value of two CO stretching frequen-
cies increased in the order 2a-Mo < 2b-Mo < 3 < 4. Of these com-
plexes, phosphonium ylide complexes (2a-Mo and 2b-Mo)
Geometry optimizations of Mo complexes 2a-Mo, 2b-Mo, 3, and
4 were successful. The optimized molecular structures are shown
in Fig. S1 (Supplementary Material). Selected geometrical parame-
ters of these complexes are summarized in Table S1 (Supplemen-
tary Material). Geometric parameters of 2a-Mo, 2b-Mo, 3, and 4
determined by DFT are in good agreement with those obtained
from the X-ray analysis, although the calculations predicted
slightly longer bond lengths. The calculated CO stretching frequen-
cies for the Mo complexes are summarized in Table 3. These Mo
complexes showed two CO stretching bands and these values
Fig. 1. ORTEP drawings of complexes 2a-Mo (left) and 2b-Mo (right) with thermal
ellipsoids drawn at 30% probability level. All hydrogen atoms are omitted for clarity.

314
D. Takaki et al. / Inorganica Chimica Acta 471 (2018) 310–315
Table 3
Spectroscopic data of [Mo(g g
3-allyl)( 2-amidinato)(CO)2L].
2a-Mo
2b-Mo
3
4
(L = CH2PMe3)
(L = CH2PPh3)
(L = IiPr)
(L = PEt3)
CO stretching band (cmÀ1
mCO (found, KBr)
mCO (av. found)a
mCO (calcd)b
)
1912, 1798
1855
1977, 1892
1934.5
1909, 1820
1864.5
1987, 1912
1949.5
1913, 1829f
1871
1999, 1922
1960.5
1921, 1836g
1878.5
2000, 1933
1966.5
mCO (av. calcd)a
Cyclic voltammetry (V)c
Ep1
Ep2
À0.05
0.31
À5.04
–
–
0.09
0.58
À5.22
0.19
0.69
À5.32
HOMO (eV)b
À4.95
Atomic charges (on Mo)b
Mulliken
0.81
0.97
1.05
0.76
APTd
À0.41
À0.34
À0.47
À0.63
NPAe
À0.71
À0.69
À0.78
À0.92
a
b
c
d
e
f
The mean value of two CO stretching frequencies.
DFT calculations (Gaussian 03, B3LYP/DGDZVP level).
Versus Fc/Fc+.
APT: Atomic polar tensor.
NPA: Natural population analysis.
Ref [9].
g
Ref [11].
gradually increased in the order 2a-Mo < 2b-Mo < 3 < 4. This trend
agrees with the experimental data. The HOMO of 2a-Mo (-5.04 eV)
is slightly lower in energy than that of 2b-Mo (À4.95 eV), whereas
these energy levels are obviously higher than those of 3 (À5.22 eV)
and 4 (À5.32 eV) (Table 3). For the Mo complexes, there is a good
relationship between the HOMO energy and the oxidation poten-
tial. The results obtained by DFT calculations indicate that the
PR3 ylides are more electron donating than IiPr and PEt3.
We considered that the atomic charges at the Mo center are one
of the most useful properties to estimate the electron donating
ability of the ligand. The properties were estimated by three differ-
ent methods: Mulliken charge analysis, APT, and NPA. These
results are listed in Table 3. Regarding 2a-Mo and 2b-Mo, the
molybdenum atomic charges obtained by three methods were rea-
sonable. On the other hand, in the cases of APT and NPA, the atomic
charge at the Mo center increases in the order 2 < 3 < 4. This result
is the reverse trend to that predicted by the experimental results
such as IR spectra and CV measurements. In particular, phosphine
was more electron donating than the phosphonium ylide and NHC
ligands. These reverse trends might originate from the different
elements bonded to the metal (phosphorus or carbon) or the
hybridization of the coordinating atom (sp3 or sp2 carbon atoms).
Furthermore, in the case of complex 3, the Mulliken charge showed
the opposite trend to those of the APT and NPA charges. It is known
that the Mulliken charge is very dependent on the method or basis
sets [28]. Thus, further investigations of the atomic charges at the
metal center as estimated by DFT calculations are needed.
Scheme 2. Reaction of complex 2-Mo with two-electron donors.
stitution reaction took place cleanly, giving NHC complex 3 in 69%
yield. In the reaction with PEt3 under the same conditions, phos-
phine complex 4 was isolated in 92% yield. Next, the reactions of
PMe3 ylide complex 2a-Mo with IiPrÁBEt3 or PEt3 under the same
conditions were examined. In the case of IiPrÁBEt3, NHC complex
3 was formed, whereas complex 2a-Mo remained. These com-
plexes were isolated as a mixture. The yields of each complex were
determined by the proton integrations based on the 1H NMR spec-
tra (2a-Mo; 58%, 3; 29%). With PEt3, the 1H NMR spectrum of the
reaction mixture showed the formation of complex 4, accompanied
by 2a-Mo. The ratio of 2a-Mo:4 was estimated to be ca. 4:1. From
the reaction mixture, unreacted 2a-Mo was recovered in 55% yield.
These results show that the PMe3 ylide acts as a stronger coordi-
nating ligand to the Mo center than the PPh3 ylide and that the
M-C(phosphonium ylide) bond strength is affected by the sub-
stituents on the phosphorus atom.
3.3. The reaction of complex 2-Mo with two-electron donors
As mentioned in Section 3.2, we have demonstrated that the
phosphonium ylides act as one of the most electron-donating
ligands toward transition metals. Next, we were interested in the
reactivity of the phosphonium ylide complex 2-Mo toward two-
electron donors such as NHC (IiPr) and phosphine (PEt3) to obtain
the information about the robustness of the M-C(phosphorus
ylide) bond. Thus, the reactions of complex 2-Mo with IiPr or
PEt3 were investigated (Scheme 2).
4. Concluding remarks
This paper describes the synthesis of phosphonium ylide com-
plexes of Mo and W and a comparison of their ligand-metal prop-
erties to those of NHC or phosphine by experimental and
theoretical investigations. The phosphonium ylides are stronger
electron-donating ligands than NHCs or phosphines, and the elec-
tronic features are tunable by varying the substituent on the phos-
phorus moiety. Furthermore, from the reactivity of complexes
2-Mo with NHC or phosphine, the robustness of the M-C(phospho-
nium ylide) bond is significantly affected by the phosphorus
On treatment of PPh3 ylide complex 2b-Mo with the BEt3
adduct of IiPr (IiPrÁBEt3), which acts as an NHC transfer reagent to
the metal [9,10],under toluene-refluxing conditions for 1 h, a sub-

D. Takaki et al. / Inorganica Chimica Acta 471 (2018) 310–315
315
3991–3993;
substituents. These findings offer fundamental concepts for mak-
ing an electron-rich metal center with thermally robust metal-
ligand bonds and for developing an active homogeneous catalyst
system for organic transformations [29].
(f) N.A. Bokach, S.I. Selivanov, V.Yu. Kukushkin, J. Vicente, M. Haukka, A.J.L.
Pombeiro, Organometallics 21 (2002) 3744–3748;
(g) L.R. Falvello, M.E. Margalejo, R. Navarro, E.P. Urriolabeitia, Inorg. Chim. Acta
347 (2003) 75–85;
(h) A. Spannenberg, P. Arndt, W. Baumann, U. Rosenthal, J. Organomet. Chem.
683 (2003) 261–266.
Author contributions
[4] (a) T. Ohta, H. Sasayama, O. Nakajima, N. Kurahashi, T. Fujii, I. Furukawa,
Tetrahedron Asymmetry 14 (2003) 537–542;
(b) R. Zurawinski, B. Donnadieu, M. Mikolajczyk, R. Chauvin, J. Organomet.
Chem. 689 (2004) 380–386;
(c) Y. Canac, C. Duhayon, R. Chauvin, Angew. Chem. Int. Ed. 46 (2007) 6313–
6315.
D. T. and K. O. contributed equally to this work.
Acknowledgements
[5] R. Zurawinski, B. Donnadieu, M. Mikolajczyk, R. Chauvin, Organometallics 22
(2003) 4810–4817.
The authors are grateful to Ms. Keiko Yamada (Chemical Analy-
sis Division, RIKEN) for the measurements of elemental analyses.
We thank Prof. Hirotaka Nagao (Sophia University) and emeritus
Prof. Takashi Ito (Yokohama National University) for fruitful dis-
cussions. Dr. Morgan L. Thomas (Yokohama National University)
is appreciated for his assistance in the revision of this manuscript.
The computations were performed using the Research Center for
Computational Science, Okazaki, Japan. This work was supported
by JSPS KAKENHI Grant Number JP17K05805 and by the Collabora-
tive Research Program of Institute for Chemical Research, Kyoto
University (grant number 2017-20).
[6] Y. Canac, C. Lepetit, M. Abdalilah, C. Duhayon, R. Chauvin, J. Am. Chem. Soc. 130
(2008) 8406–8413.
[7] Y. Canac, C. Lepetit, Inorg. Chem. 56 (2017) 667–675.
[8] (a) Y. Yamaguchi, K. Ogata, K. Kobayashi, T. Ito, Inorg. Chim. Acta 357 (2004)
2657–2668;
(b) Y. Yamaguchi, K. Ogata, K. Kobayashi, T. Ito, Dalton Trans. (2004) 3982–
3990.
[9] K. Ogata, Y. Yamaguchi, T. Kashiwabara, T. Ito, J. Organomet. Chem. 690 (2005)
5701–5709.
[10] Y. Yamaguchi, T. Kashiwabara, K. Ogata, Y. Miura, Y. Nakamura, K. Kobayashi,
T. Ito, Chem. Commun. (2004) 2160–2161.
[11] Y. Yamaguchi, K. Ogata, K. Kobayashi, T. Ito, Bull. Chem. Soc. Jpn. 77 (2004)
303–309.
[12] R.G. Parr, W. Yang, Density-Functional Theory of Atoms and Molecules, Oxford
University Press, Oxford, 1989.
[13] A.D. Beck, J. Chem. Phys. 98 (1993) 5648–5652.
Appendix A. Supplementary data
[14] C. Lee, W. Yang, R.G. Parr, Phys. Rev. B 37 (1998) 785–789.
[15] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman,
J.A.Jr. Montgomery, T. Vreven, K.N. Kudin, J.C. Burant, J.M. Millam, S.S. Iyengar,
J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G.A.
Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa,
M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J.E. Knox, H.
P. Hratchian, J.B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R.E.
Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W. Ochterski, P.Y.
Ayala, K. Morokuma, G.A. Voth, P. Salvador, J.J. Dannenberg, V.G. Zakrzewski, S.
Dapprich, A.D. Daniels, M.C. Strain, O. Farkas, D.K. Malick, A.D. Rabuck, K.
Raghavachari, J.B. Foresman, J.V. Ortiz, Q. Cui, A.G. Baboul, S. Clifford, J.
Cioslowski, B.B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R.L.
Martin, D.J. Fox, T. Keith, M.A. Al-Laham, C.Y. Peng, A. Nanayakkara, M.
Challacombe, P.M.W. Gill, B. Johnson, W. Chen, M.W. Wong, C. Gonzalez, J.A.
Pople, Gaussian 03, Revision B.04, Gaussian Inc, Pittsburgh, PA, 2003.
[16] (a) N. Godbout, D.R. Salahub, J. Andzelm, E. Wimmer, Can. J. Chem. 70 (1992)
560–571;
Supplementary data associated with this article can be found, in
the online version, at https://doi.org/10.1016/j.ica.2017.11.019.
References
[1] Selected reviews: (a) W.A. Herrmann, Angew. Chem. Int. Ed. 41 (2002) 1290–
1309;
(b) M.C. Perry, K. Burgess, Tetrahedron: Asymmetry 14 (2003) 951–961;
(c) V. César, S. Bellemin-Laponnaz, L.H. Gade, Chem. Soc. Rev. 33 (2004) 619–
636;
(d) R.E. Douthwaite, Coord. Chem. Rev. 251 (2007) 702–717;
(e) L.H. Gade, S. Bellemin-Laponnaz, Coord. Chem. Rev. 251 (2007) 718–725;
(f) E.A.B. Kantchev, C.J. O’Brien, M.G. Organ, Angew. Chem. Int. Ed. 46 (2007)
2768–2813;
(g) F.E. Hahn, M.C. Jahnke, Angew. Chem. Int. Ed. 47 (2008) 3122–3172;
(h)S.P. Nolan (Ed.), N-Heterocyclic Carbenes in Synthesis, WILEY-VCH Verlag
GmbH & Co. KGaA, 2006;
(i)F. Glorius (Ed.), N-Heterocyclic Carbenes in Transition Metal Catalysis,
Springer-Verlag, Berlin Heidelberg, 2007 (Top. Organomet. Chem., 2007, 21);
(j) S. Díez-González, N. Marion, S.P. Nolan, Chem. Rev. 109 (2009) 3612–3676;
(k) M. Poyatos, J.A. Mata, E. Peris, Chem. Rev. 109 (2009) 3677–3707;
(l) J. Tornatzky, A. Kannenberg, S. Blechert, Dalton Trans. 41 (2012) 8215–
8225;
(m) F. Wang, L.-J. Liu, W. Wang, S. Li, M. Shi, Coord. Chem. Rev. 256 (2012)
804–853;
(n) M.N. Hopkinson, C. Richter, M. Schedler, F. Glorius, Nature 510 (2014) 485–
496;
(o) S. Bellemin-Laponnaz, S. Dagorne, Chem. Rev. 114 (2014) 8747–8774;
(p) B. Waters, J.M. Goicoechea, Coord. Chem. Rev. 293–294 (2015) 80–94;
(q) Z. Wang, L. Jiang, D.K.B. Mohamed, J. Zhao, T.S.A. Hor, Coord. Chem. Rev.
293–294 (2015) 292–326.
(b) C. Sosa, J. Andzelm, B.C. Elkin, E. Wimmer, K.D. Dobbs, D.A. Dixon, J. Phys.
Chem. 96 (1992) 6630–6636.
[17] (a) R.S. Mulliken, J. Chem. Phys. 23 (1955) 1833–1840;
(b) R.S. Mulliken, J. Chem. Phys. 23 (1955) 1841–1846;
(c) R.S. Mulliken, J. Chem. Phys. 23 (1955) 2338–2342;
(d) R.S. Mulliken, J. Chem. Phys. 23 (1955) 2343–2346.
[18] (a) A.E. Reed, L.A. Curtiss, F. Weinhold, Chem. Rev. 88 (1988) 899–926;
(b) E.D. Glendening, A.E. Reed, J.E. Carpenter, F. Weinhold, NBO (version 3.1),
Gaussian Inc., Pittsburg, PA, 2003.
[19] J. Cioslowski, J. Am. Chem. Soc. 111 (1989) 8333–8336.
[20] SHELXT Version 2014/5: G.M. Sheldrick, Acta Cryst. A70 (2014) C1437.
[21] SHELXL Version 2014/7: G.M. Sheldrick, Acta Cryst. A 64 (2008) 112–122.
[22] CrystalStructure 4.2.2: Crystal Structure Analysis Package, Rigaku Corporation
(2000–2016). Tokyo 196-8666, Japan.
[23] K. Nakamoto, Infrared and Raman Spectra of Inorganic and Coordination
Compounds, third ed., John Wiley & Sons, New York, 1978.
[24] (a) M.D. Curtis, O. Eisenstein, Organometallics 3 (1984) 887–895;
(b) M.D. Curtis, N.A. Fotinos, J. Organomet. Chem. 272 (1984) 43–54.
[25] C.A. Tolman, Chem. Rev. 77 (1977) 313–348.
[2] Reviews: (a) W.C. Kaska, Coord. Chem. Rev. 48 (1983) 1–58;
(b) H. Schmidbaur, Angew. Chem. Int. Ed. Engl. 22 (1983) 907–927;
(c) D. Steinborn, Angew. Chem. Int. Ed. Engl. 31 (1992) 401–421;
(d) U. Belluco, R.A. Michelin, M. Mozzon, R. Bertani, G. Facchin, L. Zanotto, L.
Pandolfo, J. Organomet. Chem 557 (1998) 37–68;
(e) R. Navarro, E. Urriolabeitia, J. Chem. Soc., Dalton Trans. (1999) 4111–4122;
(f) Y. Canac, R. Chauvin, Eur. J. Inorg. Chem. (2010) 2325–2335.
[3] For examples (a) H. Nakazawa, D.L. Johnson, J.A. Gladysz, Organometallics 2
(1983) 1846–1851;
[26] E.A.V. Ebsworth, T.E. Fraser, D.W.H. Rankin, Chem. Ber. 110 (1977) 3494–3500.
[27] J.C.J. Bart, J. Chem. Soc. B (1969) 350–365.
[28] (a) F.M. Balci, N. Uras-Aytemiz, R. Escribano, P.C. Gómez, Comp. Theor. Chem.
1038 (2014) 71–77;
(b) J.J. Philips, M.A. Hudspeth, P.M. Browne Jr., J.E. Peralta, Chem. Phys. Lett.
495 (2010) 146–150.
[29] PPh3 ylide in complex 2b-Mo acts as a stronger electron-donating ligand than
NHC and PR3. However the substitution reaction of PPh3 ylide with NHC or PR3
easily takes place. Therefore, this result suggests that a stronger electron donor
ligand does not always construct the most robust metal-ligand bond in a given
complex. Further investigations on the relationship between electronic and/or
steric factors of the ligand and the metal-ligand bond character might be
needed. We thank Reviewer #2 for the comment on this point.
(b) H. Scordia, R. Kergoat, M.M. Kubicki, J.E. Guerchais, P. L’Haridon,
Organometallics 2 (1983) 1681–1687;
(c) H. Nakazawa, Y. Ueda, K. Nakamura, K. Miyoshi, Organometallics 16 (1997)
1562–1566;
(d) L.R. Falvello, S. Fernández, R. Navarro, A. Rueda, E.P. Urriolabeitia,
Organometallics 17 (1998) 5887–5900;
(e) A. Spannenberg, W. Baumann, U. Rosenthal, Organometallics 19 (2000)