Examination of oxygen atom transfer reactivity of heteroscorpionate dioxo-Mo(VI) complexes: Geometric isomers, solvent effect, intermediates, and catalytic oxidation

Article abstract of DOI:10.1016/j.ica.2016.03.035

Heteroscorpionate-based [(L10O)MoO₂Cl] and [(L3S)MoO₂Cl] complexes containing an interchangeable third heteroatom donor have been utilized for the systematic investigation of oxygen atom transfer (OAT) reactivity. The detection of phosphoryl intermediates and products in the reaction pathway were probed by UV-Vis, mass spectrometry, and ³¹P NMR spectroscopy. The OAT reactivity of the metal complexes toward PPh₃ were monitored by UV-Vis spectroscopy under pseudo-first order conditions. The sterically encumbered (L10O) ligand gives rise to isolable trans and cis isomers of [(L10O)MoO₂Cl] allowing investigation into the role of geometry on OAT reactivity. The OAT reactivity of the cis isomer of (L10O)MoO₂Cl demonstrated a dramatic solvent dependence, in which the reaction proceeded at a measureable rate only in pyridine. However, the trans counterpart reacted in all solvents and at much faster rates. The catalytic oxidation of PPh₃ to OPPh₃ by trans-[(L10O)MoO₂Cl] and cis-[(L3S)MoO₂Cl] complexes using DMSO as an oxygen donor was monitored by ³¹P NMR in DMF at 30 °C with rates, k_cat = 4.26 × 10^-5 s^-1 and 5.28 × 10^-5 s^-1, respectively.

Full text of DOI:10.1016/j.ica.2016.03.035

Inorganica Chimica Acta 447 (2016) 45–51
Contents lists available at ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
Examination of oxygen atom transfer reactivity of heteroscorpionate
dioxo-Mo(VI) complexes: Geometric isomers, solvent effect,
intermediates, and catalytic oxidation
⇑
Ba L. Tran, Amy Arita, Andrew L. Cooksy, Carl J. Carrano
Department of Chemistry and Biochemistry, San Diego State University, San Diego, CA 92182-1030, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 6 January 2016
Received in revised form 22 March 2016
Accepted 24 March 2016
Available online 31 March 2016
Heteroscorpionate-based [(L10O)MoO2Cl] and [(L3S)MoO2Cl] complexes containing an interchangeable
third heteroatom donor have been utilized for the systematic investigation of oxygen atom transfer
(OAT) reactivity. The detection of phosphoryl intermediates and products in the reaction pathway were
probed by UV–Vis, mass spectrometry, and 31P NMR spectroscopy. The OAT reactivity of the metal com-
plexes toward PPh3 were monitored by UV–Vis spectroscopy under pseudo-first order conditions. The
sterically encumbered (L10O) ligand gives rise to isolable trans and cis isomers of [(L10O)MoO2Cl] allow-
ing investigation into the role of geometry on OAT reactivity. The OAT reactivity of the cis isomer of
(L10O)MoO2Cl demonstrated a dramatic solvent dependence, in which the reaction proceeded at a mea-
sureable rate only in pyridine. However, the trans counterpart reacted in all solvents and at much faster
rates. The catalytic oxidation of PPh3 to OPPh3 by trans-[(L10O)MoO2Cl] and cis-[(L3S)MoO2Cl] complexes
using DMSO as an oxygen donor was monitored by 31P NMR in DMF at 30 °C with rates, kcat = 4.26 ꢀ
10ꢁ5 sꢁ1 and 5.28 ꢀ 10ꢁ5 sꢁ1, respectively.
Keywords:
Oxygen atom transfer
Molybdenum complexes
Heteroscorpionate ligands
Ó 2016 Published by Elsevier B.V.
1. Introduction
the enzymes [6–18]. Mechanistically a variety of computational
studies have proposed that the oxygen atom transfer performed
Oxygen atom transfer (OAT) is an important class of chemical
transformation in biology and chemistry [1–3]. Mononuclear
molybdenum enzymes are ubiquitously involved in this two elec-
tron process by shuttling a terminal oxo from the metal center to
an acceptor and vice versa with the molybdenum center inter-
changing between the +6 and +4 oxidation states [4]. These
enzymes have been classified into three major families: Sulfite Oxi-
dase (SO), Xanthine Oxidase (XO), and DMSO Reductase (DMSOR),
in which all the active sites contain conserved pterin-dithiolene
ligation to the molybdenum [5]. The proper regulation and func-
tion of these enzymes affect the environment, biological metabo-
lism, and homeostasis. For instance, mutation on the structural
gene of SO causes complications from dysfunctional lipid metabo-
lism to fatal neurological abnormalities [5].
by these molybdoenzymes follows an inner sphere electron trans-
fer mechanism proceeding through the associative formation of a
bridging intermediate that subsequently decomposes to products
via a dissociative pathway [19]. Utilizing the Tp⁄ system, Basu
and co-workers provided direct evidence of this predicted interme-
diate with the crystal structure of the phosphoryl adduct complex
[Tp⁄MoOCl(OPMe3)] [20–24]. We have observed similar species via
ESI-MS and UV–Vis spectroscopy utilizing the sulfur rich
[TmMeMoO2Cl] complex [25].
Prompted by Basu’s work and our own results, we have
expanded on this research by investigating our family of
heteroscorpionate complexes of the type [LMoO2Cl], where L =
(L10O) and (L3S) shown in Scheme 1. Over the last several years,
we have demonstrated that these dioxo-Mo(VI) complexes signifi-
cantly influenced OAT reactivity based on contribution of the third
heteroatom, i.e. alkoxide, phenol, carboxylate, or thiolate [13]. In
addition, the introduction of steric bulk into the phenolate ligand
has allowed for the isolation of both cis and trans geometric iso-
mers and we have studied the isomerization rates from kinetically
labile isomer to the thermodynamically favored product for both
Mo(V) and Mo(VI). The effect of geometry on OAT reactivity of
the dioxo-Mo(VI) complexes has also been superficially studied
The utility of facially tridentate scorpionate and heteroscorpi-
onate model systems has provided important spectroscopic and
OAT mechanistic insights into the function and mechanism of
mononuclear molybdoenzymes despite their incorporating pyra-
zolate donors rather than the sulfur rich environment native to
⇑
Corresponding author. Tel.: +1 619 594 1617; fax: +1 619 594 4634.
E-mail address: ccarrano@mail.sdsu.edu (C.J. Carrano).
http://dx.doi.org/10.1016/j.ica.2016.03.035
0020-1693/Ó 2016 Published by Elsevier B.V.

46
B.L. Tran et al. / Inorganica Chimica Acta 447 (2016) 45–51
1.2
-
-
S
S
N
1.0
0.8
0.6
0.4
0.2
0.0
S
N
N
N
N
N
N
N
N
N
B
N
B
N
H
H
-
(TmMe
)
(Tp*)-
N
N
N
N
340
360
380
400
420
440
460
480
500
N
O
S
N
Wavelength (nm)
N
N
2.0
1.5
1.0
0.5
0.0
H
H
(L10O)-
Scheme 1. Ligands referred to in this study.
(L3S)-
[12,14,16,18]. Finally, we demonstrated a metal ion dependence
between Mo(VI) and W(VI) on both isomerization kinetics and
OAT reactivity [15,26]. Herein, utilizing complexes of [(L10O)
MoO2Cl] and [(L3S)MoO2Cl], we explore the effect of solvent on
OAT reactivity along with the detection of [(L10O)MoOCl(OPPh3)]
and [(L3S)MoOCl(OPPh3)] intermediates by mass spectrometry
and 31P NMR spectroscopy. Moreover, utilizing 31P NMR spec-
troscopy, we have monitored the ability of these complexes to
mediate catalytic OAT between DMSO and triphenylphosphine.
340
360
380
400
420
440
460
480
500
Wavelength (nm)
Fig. 1. Plot shows rapid formation of trans-[(L10O)MoOCl(OPPh3)] indicated by the
increase at 400 nm in acetonitrile (top). After several minutes, decomposition of the
intermediate occurs with isosbestic point at 370 nm.
2. Results
2.1. trans-[(L10O)MoO2Cl]
was confirmed by ESI-MS recorded in positive mode in acetonitrile
which revealed a cluster centered at 812.8 m/z, (Fig. 2) consistent
with the simulated isotope pattern expected for the bridging inter-
mediate (Fig. S1). Furthermore, monitoring the OAT reaction in
benzene by 31P NMR revealed three peaks, assignable to PPh3
(ꢁ4 ppm), OPPh3 (26 ppm), and the [(L10O)MoOCl(OPPh3)] at
51 ppm. The subsequent exponential decay of the 400 nm band
represents the dissociation of the OPPh3 product from the primary
coordination sphere. This is consistent with the solvent assisted
dissociative interchange (Id) mechanism proposed for a related sys-
tem [25].
Isolation of the pure trans isomer of [(L10O)MoO2Cl] remains
difficult due to its rapid conversion to the thermodynamically
more stable cis isomer under a variety of conditions. The mecha-
nism of isomerization at the metal center is proposed to occur
via a trigonal twist mechanism as the bromide–chloride exchange
was not observed by mass spectrometry [16]. Optimal synthetic
efforts have yielded a product with an 11:1 trans/cis ratio based
on proton NMR [15].
Examination of the OAT reactivity of trans-[(L10O)MoO2Cl] in
acetonitrile, DMF, benzene, and pyridine under pseudo-first order
conditions with 50-fold excess of phosphine was monitored by
UV–Vis spectroscopy. On mixing of acetonitrile solution of trans-
[(L10O)MoO2Cl] and phosphine, the optical spectra depicted in
Fig. 1 (top) showed a rapid increase of the band at 400 nm and
simultaneous appearance of the weak, broad, d–d transition band
in the 600–800 nm region (data not shown) as the direct result
of the two electron reduction of Mo(VI) to Mo(IV). This reaction
proceeded rapidly leading to a bright green solution. After several
minutes in acetonitrile and as shown in Fig. 1 (bottom), the band at
400 nm decreased exponentially with concomitant growth of a
new peak at 360 nm with an isosbestic point at 370 nm. The d–d
band at 600–800 nm also decreased slightly in intensity. These
UV–Vis spectral features are identical to those reported during
the OAT chemistry of the related, [TmMeMoO2Cl]. The rapid growth
of the 400 nm peak corresponds to the formation of the intermedi-
ate [(L10O)MoOCl(OPPh3)]. The assignment of 400 nm peak
observed in the UV–Vis as being due to the phosphine adduct
The effect of pyridine on the reaction run in DMF or acetonitrile
was probed through UV–Vis spectroscopy by allowing steady-state
formation of the phosphine adduct to age for ꢂ10 min followed by
addition of an aliquot of pyridine. Introduction of pyridine resulted
in the more rapid decay of the 400 nm shoulder with concomitant
growth of bands at 500 nm and in the 600–800 nm region. The
final spectrum is identical to the previously isolated and character-
ized trans-[(L10O)MoOCl(py)]. Probing the same sample via ESI
mass spectrometry also detected [(L10O)MoOCl(py)] at 613 m/z
(Fig. 3) confirmed by the simulated isotope pattern (data not
shown). When the OAT was performed in pure pyridine, rapid for-
mation of the [(L10O)MoOCl(OPPh3)] intermediate was again wit-
nessed at 400 nm along with the rise of the broad Mo(IV) d–d
transition band at 600–800 nm (Fig. 4). The stability of phosphine
bound intermediate in pyridine is much less than that in DMF, ace-
tonitrile, and benzene however as indicated by the near simultane-
ous appearance of
a shoulder at 500 nm corresponding to
formation of both the trans-[(L10O)MoOCl(py)] complex and the

B.L. Tran et al. / Inorganica Chimica Acta 447 (2016) 45–51
47
812.8
100
95
90
85
80
75
70
65
60
55
50
45
40
35
30
25
20
15
10
5
810.8
809.8
811.8
814.7
813.7
808.8
806.8
815.7
807.8
816.7
817.7
818.7
819
819.7
820
805.1
820.5
806.0
806
0
805
807
808
809
810
811
812
813
m/z
814
815
816
817
818
821
Fig. 2. ESI-MS in positive mode of trans-[(L10O)MoOCl(OPPh3)]+ in acetonitrile.
613.4
100
95
90
85
80
75
70
65
60
55
50
45
40
35
30
25
20
15
10
5
611.4
615.4
612.4
610.4
609.5
607.4
614.4
616.4
608.5
617.5
619.5
618.5
620.6
0
607
608
609
610
611
612
613
614
m/z
615
616
617
618
619
620
Fig. 3. ESI-MS spectrum of [(L10O)MoOCl(pyridine)]+.
intermediate. Fig. 5 shows the deconvoluted individual UV–Vis
spectra in the OAT reaction profile. Nevertheless the solvation of
the phosphine oxide bound intermediate by pyridine is still
quite slow relative to that seen with [TmMeMoOCl(OPPh3)] and
[Tp⁄MoOCl(OPPh3)] likely due to the steric protection of the inter-
mediate afforded by the biphenyl rings of the (L10O) ligand.
unchanged UV–Vis spectrum over many days. Moreover, analysis
of the OAT reaction mixture by ESI-MS revealed no detectable
phosphine adduct and only the starting dioxo complex was
observed. However, when the OAT reaction of cis-[(L10O)MoO2Cl]
was performed in pyridine as a solvent, OAT occurred and could
be followed by the appearance of the Mo(IV) d–d transition band
at 600–800 nm and the intense absorption at 500 nm reminiscent
of the optical spectrum of trans-[(L10O)MoOCl(py)] (Fig. S2). Thus
in contrast to trans-[(L10O)MoO2Cl] which underwent OAT reac-
tion in all solvents (vide supra), the reaction with the cis isomer
proceeded only in pyridine. In addition, and consistent with our
previously reported OAT results, even in pyridine the cis isomer
reacted at least 20 times slower than the trans isomer (Fig. S3)
[12]. The redox potential contribution to the substantial variation
2.2. cis-[(L10O)MoO2Cl]
The pure cis isomer of [(L10O)MoO2Cl] was synthesized accord-
ing to published protocol by heating a crude isomeric mixture in
DMF to assure the complete conversion of all the trans to the cis
isomer. Under identical OAT conditions in acetonitrile and DMF,
the cis isomer appears completely unreactive based on the

48
B.L. Tran et al. / Inorganica Chimica Acta 447 (2016) 45–51
600
500
400
300
200
100
0
Scheme 2. Assignments of cis and trans OAT pathways.
Table 1
400
500
600
700
800
900
B3LYP energies and free energies of activation and pseudo-first order rate constants.
k1 (sꢁ1
0
)
Wavelength (nm)
Step 1
Ea (kcal/mol)
Ga (kcal/mol)
trans c
trans t
cis
7.3
11.9
15.0
20.2
24.3
26.9
2 ꢃ 10ꢁ4
2 ꢃ 10ꢁ7
3 ꢃ 10ꢁ9
Fig. 4. UV–Vis spectrum in pyridine shows rapid formation of trans-[(L10O)MoOCl
(OPPh3)] by the increase at 400 nm. The 500 nm peak represents the simultaneous
formation of [(L10O)MoOCl(pyridine)].
k2 (sꢁ1
)
0
Step 2
Ea (kcal/mol)
Ga (kcal/mol)
trans c
14.7
29.6
3 ꢃ 10ꢁ9
1600
1400
1200
1000
800
600
400
200
0
which is roughly 90 degrees from O1 and O2 (trans c path). The
pseudo-first order rate constants in Table 1 were obtained by com-
bining zero-point and thermal energy corrections to obtain pre-
dicted free energies of activation Ga. The computational results
given in Table 1 correctly predict that the trans c reaction pathway
is the most likely, and that step 1 tends to be slower than step 2.
However, the agreement is only qualitative, with rate constants
much too low predicted for k2 and for the cis k1 . Rate constants
are difficult to predict quantitatively for complex systems, where
a 1 kcal/mol error in Ga corresponds to more than a factor of 5 in
reaction rate. For step 1, the computed entropic contribution to
Ga alone is nearly 13 kcal/mol. Solvent effects have also been
neglected in the present calculations.
0
0
400
500
600
700
800
900
Natural bond order (NBO) analyses were carried out on the step
1 reactants and transition states in an attempt to explain the
greater relative stability of the trans c transition state. Selected
results are summarized in Table 2. Before the reaction, the elec-
tron-deficient metal pulls electron density from the ligands in
the complex. In the cis reactant, each of the three oxygens is
aligned along a different axis in the octahedral complex, and this
allows maximum donation into the empty d orbitals of the metal,
accounting for the relatively strong Mo–O1 bond (R = 1.971 Å for
cis, 2.018 Å for trans) and the slight relative stability of the cis to
Wavelength (nm)
Fig. 5. Spectral deconvolution of the OAT reaction between trans-[(L10O)MoO2Cl]
and PPh3 in pyridine. The biphasic OAT kinetic parameters extracted at 500 nm
assumes an A ? B ? C kinetic scheme. The solid line represents the spectra of the
starting trans-[(L10O)MoO2Cl], the dashed line the phosphine adduct, and the
dotted line trans-[(L10O)MoOCl(py)].
in the two observed rates of OAT reactivity appears to be inade-
quate because the difference in the Mo(VI/V) redox potential
between the isomers is quite small (<100 mv) [15], which cannot
account for the variation in the two observed rates. More interest-
ingly, the lack of OAT reactivity by the cis isomer in acetonitrile and
DMF compared to pyridine suggests the crucial effect of solvent.
The 1H NMR of cis-[(L10O)MoO2Cl] in d5-pyridine revealed only
the cis isomer, eliminating the possibility that rapid isomerization
of the cis to the trans isomer in pyridine could account for its reac-
tivity in pyridine but not in ACN or DMF [27].
the trans overall, with
DG (cis–trans) = 2.3 kcal/mol. When the
phosphine attaches to either of the two oxo’s, electron density
donated by the phosphorus increases the electron density at the
metal, which weakens the Mo–O1 bond. In all three transition
states, the Mo–O1 bond distance has increased to 2.045 Å, but this
is a greater weakening of the bond in the cis structure than in the
trans, and therefore the cis has the highest barrier to reaction.
In order to better understand the effect of geometry at the dioxo
center with OAT rate we conducted computational studies (for
details see Table S1). The two reaction steps i.e. binding to phos-
phine and displacement of the phosphine oxide by solvent, were
computationally modeled for the [(L10O)MoO2Cl] complexes in
order to determine the origin of the observed stereospecificity.
Transition states for three pathways were identified for step 1 of
the reaction (Scheme 2): binding of the phosphine to either oxo
group of the cis reactant (cis path); binding to the trans reactant
at the oxo group O2, which is trans to the ligand oxygen atom O1
(trans t path); and binding to the trans at the cis oxo group O3,
Table 2
Selected NBO charges and bond lengths.
Reactants
TS1 complexes
trans
cis
trans c
trans t
cis
O1 charge (e)
O2,O30 charge (e)
O3 charge (e)
R(Mo–O1) (Å)
R(Mo–O2,O30) (Å)
R(Mo–O3) (Å)
ꢁ0.68
ꢁ0.45
ꢁ0.39
2.019
1.733
1.719
ꢁ0.66
ꢁ0.41
ꢁ0.41
1.971
1.721
1.722
ꢁ0.71
ꢁ0.48
ꢁ0.59
2.045
1.728
1.825
ꢁ0.72
ꢁ0.67
ꢁ0.44
2.045
1.864
1.708
ꢁ0.71
ꢁ0.45
ꢁ0.63
2.046
1.710
1.846

B.L. Tran et al. / Inorganica Chimica Acta 447 (2016) 45–51
49
Of the two oxo’s in the trans path, O3 is the better electron
acceptor (charge = ꢁ0.39e for O3, ꢁ0.45e for O2) because it is able
to donate more electron density than O2, which shares an axis with
O1. This favors O3 for attachment of the phosphine, and is consis-
tent with the smaller change in Mo–O3 bond distance upon forma-
tion of the transition state in the trans c path than the trans t
addition of pyridine had no effect. Based on published results and
our other observations reported here either the phosphine oxide
intermediate or the solvent adduct would be expected to react with
pyridine to yield the mononuclear species, i.e. [(L3S)MoOCl(py)].
Immediate analysis of the OAT reaction mixture in acetonitrile by
mass spectrometry led to the detection of the [(L3S)MoOCl(OPPh3)]
intermediate isotopic cluster at 704 amu (Fig. S5) however also
detected was the presence of [(L3S)MoOCl]2O at 867 amu
(Fig. S6). Analysis after one day showed that the phosphine inter-
(DR = 0.106 Å for trans c and DR = 0.131 Å for trans t). During reac-
tion, the HOMO of the PMe3 preferentially acts as a sigma donor to
the lowest unoccupied orbital of one of the oxygens. As shown in
Fig. 6, the LUMOs in each of the three step 1 transition states span
a metal d orbital and an Mo–O p⁄ orbital ending at the attachment
site of the phosphine.
mediate had decomposed, but the l-oxo dimer cluster at 867 m/z
remained. A reasonable explanation for these observations is that
the less bulky methyl groups on the thiolate ligand do not prevent
l
-oxo dimerization of the [(L3S)MoOCl(solvent)] species formed
after OAT.
In contrast, mixing of the metal complex and phosphine in pyr-
2.3. cis-[(L3S)MoO2Cl]
idine generated [(L3S)MoOCl(OPPh3)] followed by the growth of
the 500 nm band and d–d transition band from Mo(IV) at 800 nm
with a tight isosbestic point at 405 nm indicating only two absorb-
ing species in solution (Fig. S7). Fig. 7 shows the individual decon-
voluted UV–vis spectra in the OAT reaction profile. The rate of
formation of [(L3S)MoOCl(OPPh3)] extracted at 500 nm is
k1obs = 6.96 ꢀ 10ꢁ4 sꢁ1 (Table 3). Upon reaching steady state, the
phosphoryl adduct decomposed by releasing OPPh3 as monitored
The UV–Vis spectrum for the OAT reaction of [(L3S)MoO2Cl]
with phosphine in DMF and acetonitrile shows an absorbance
increase at 500 nm and between 600–900 nm producing a red solu-
tion after ꢂ10 min (Fig. S4). The non-exponential kinetics and the
lack of isosbestic points suggest a competing elementary reaction.
The red solution was found to persist for weeks. We postulated that
the red solution did not contain the phosphine adduct or [(L3S)
MoOCl(solvent)] species, but rather it was the l-oxo dimer, as the
Fig. 6. HOMO for trans c transition state and LUMOs for all three step 1 transition states (after removing PMe3).

50
B.L. Tran et al. / Inorganica Chimica Acta 447 (2016) 45–51
1800
1600
1400
1200
1000
800
600
400
200
0
2.0
1.5
1.0
0.5
0.0
0
5000
10000 15000 20000 25000 30000 35000 40000 45000
Time (s)
400
500
600
700
800
900
Wavelength (nm)
Fig. 7. OAT reaction of PPh3 with cis-[(L3S)MoO2Cl] in pyridine. Right: the biphasic OAT kinetics extracted at 500 nm (right) where the points are the data and the solid line a
fit to an A ? B ? C kinetic scheme. Left: Spectral deconvolution of the OAT reaction profile where the solid line represents the spectra of the starting cis-[(L3S)MoO2Cl], the
dashed line the phosphine adduct, and the dotted line [(L3S))MoOCl(py)].
Table 3
Kinetic Parameters for OAT.
3. Experimental
k1 (sꢁ1
)
k2 (sꢁ1
)
3.1. Preparation of compounds
Metal complex
Solvent
Trans-(L10O)MoO2Cl
ACN
DMF
4.18 ꢀ 10ꢁ2
3.80 ꢀ 10ꢁ4
6.08 ꢀ 10ꢁ3
All syntheses were performed under a dinitrogen or argon
atmosphere using standard Schlenk or drybox techniques. Acetoni-
trile, ether, dichloromethane and THF were freshly purified using
an Innovative Technology solvent purification system. All other
solvents were dried with appropriate drying agents and distilled
prior to use. Silica gel (SilicAR Silica Gel 150, 60–200 mesh 60–
200 mesh) used in adsorption chromatography was obtained from
Mallinckrodt while the filtering agent Celite was obtained from
Aldrich. The heteroscorpionate ligands (L10OH, L3SH) utilized in
this study was synthesized according to previous reports. The syn-
thesis and detailed characterization of [(L10O)MoO2Cl] stereoiso-
mers and [(L3S)MoO2Cl], are reported elsewhere [13,15].
DMF/pyridine
Pyridine
Pyridine
Pyridine
ACN and DMFa
1.99 ꢀ 10ꢁ4
1.82 ꢀ 10ꢁ4
9.11 ꢀ 10ꢁ6
5.92 ꢀ 10ꢁ5
cis-(L10O)MoO2Cl
cis-(L3S)MoO2Cl
6.96 ꢀ 10ꢁ4
a
The k1 and k2 could not be quantitatively measured due to competing
dimer reaction.
l-oxo
by the decreased absorption intensity at 500 nm to complete the
oxygen atom transfer. The decomposition rate of intermediate to
products is k2obs = 5.92 ꢀ 10ꢁ5 sꢁ1
.
3.2. Physical methods
2.4. Catalytic oxygen atom transfer
Electrospray mass spectra (ESI-MS) were recorded on a Finnigan
LCQ ion-trap mass spectrometer equipped with an ESI source
(Finnigan MAT, San Jose, CA) in positive ion mode. A PC with
Navigator software version 1.2 (Finnigan Corp., 1995–1997), was
used for data acquisition and plotting. Isotope distribution patterns
were simulated using the program Molecular Weight Calculator
6.42. GC–MS detection of dimethylsulfide was performed on the
Varian Chrompack Saturn 2000R GC–MS containing auto sampler
operated under the Varian MS Workstation System Control soft-
ware version 6.6. UV–vis spectra were recorded using a Cary 50
UV–Vis spectrophotometer under PC control using the Cary WinUV
The catalytic OAT of trans-[(L10O)MoO2Cl] and cis-[(L3S)MoO2-
Cl] were monitored in 50:50 v/v DMF and d6-DMSO with 10-fold
excess of PPh3 by 31P NMR. The cis isomer of [(L10O)MoO2Cl]
was not examined due to its extremely slow OAT rate in any sol-
vent other than pure pyridine and pyridine has been shown to
inhibit the catalytic cycle [25]. The reaction (Me2S = O + PPh3 ?
Me2S + OPPh3) was followed by the growth of phosphine oxide
(27 ppm) over time. The rate of catalytic OAT by trans-[(L10O)
MoO2Cl] is kcat = 4.26 ꢀ 10ꢁ5 sꢁ1. Upon completion of the reaction,
precipitation of a white solid ensued which was isolated by vac-
uum filtration and washed with copious amounts of hexane to
remove excess solvents and OPPh3. IR characterization of the white
solid confirmed its identity as [(L10O)MoO2Cl], thus providing
direct evidence that the mononuclear complex is the active cata-
lyst. cis-[(L3S)MoO2Cl] also demonstrated catalytic conversation
of phosphine (-5 ppm) to phosphine oxide (27 ppm) as shown by
31P NMR with kcat = 5.28 ꢀ 10ꢁ5 sꢁ1. A stoichiometric OAT reaction
of cis-[(L3S)MoO2Cl] with PPh3 monitored by 31P NMR in DMF
showed 100% conversion to OPPh3 followed by rapid coloration
software. Infrared spectra were collected KBr disks on
a
ThermoNicolet Nexus 670 FT-IR spectrometer under PC control
and are reported in wavenumbers.
3.3. Oxygen atom transfer kinetics
A 2.5 mM solution of LMoO2Cl in acetonitrile, DMF, benzene,
and pyridine was reacted with an excess of 50-fold PPh3 at 30
(1) °C using a jacketed quartz cell attached to a circulating bath
controlling the temperature. Fresh samples were prepared for each
kinetic run and solutions were preequilibrated at 30 °C in a water
bath prior to any measurements. Tight isosbestic points were
observed throughout the reaction indicating the presence of only
two absorbing species. All reactions were monitored using a Cary
50 UV–Vis spectrometer under PC control. Rate constants for OAT
to dark purple red indicating
l-oxo dimer formation. The experi-
ment indicates that the mononuclear complex transferred an oxy-
gen atom to the phosphine acceptor, but dimerization occurred
soon after as the ligand lacks steric bulk to protect the Mo center
from dimerization.

B.L. Tran et al. / Inorganica Chimica Acta 447 (2016) 45–51
51
were extracted at 700 nm or 500 nm and where appropriate global
References
fitting of the data were performed with OLIS Spectral Works
version 3.0.8 kinetic software.
[1] R.A. Sheldon, J.K. Kochi, Metal-catalyzed Oxidation of Organic Compounds,
Academic, 1981.
[2] R.H. Holm, Chem. Rev. 87 (1987) 1401.
[3] L.K. Woo, Chem. Rev. 93 (1993) 1125.
3.4. Catalytic OAT studies
[4] J.H. Enemark, J.J.A. Cooney, J.J. Wang, R.H. Holm, Chem. Rev. 104 (2004) 1175.
[5] R. Hille, Chem. Rev. 96 (1996) 2757.
[6] Z. Xiao, C.G. Young, J.H. Enemark, A.G. Wedd, J. Am. Chem. Soc. 114 (1992)
9194.
[7] W.E. Cleland, K.M. Barnhart, K. Yamanochi, D. Collison, F.E. Mabbs, R.B. Ortega,
J.H. Enemark, Inorg. Chem. 1017 (1987) 26.
[8] S.A. Roberts, C.G. Young, C.A. Kipke, W.E. Cleland, K. Yamanouchi, M.D.
Carducci, J.H. Enemark, Inorg. Chem. 29 (1990) 3650.
[9] F.E. Inscore, R. McNaughton, B.L. Westcott, M.E. Helton, R. Jones, I.K. Dhawan, J.
H. Enemark, M.L. Kirk, Inorg. Chem. 38 (1999) 1401.
[10] S.A. Roberts, C.G. Young, W.E. Cleland, R.B. Ortega, J.H. Enemark, Inorg. Chem.
27 (1988) 3044.
[11] Z. Xiao, M.A. Bruck, J.H. Enemark, C.G. Young, A.G. Wedd, Inorg. Chem. 35
(1996) 7508.
[12] J.T. Hoffman, S. Einwaechter, B.S. Chohan, P. Basu, C.J. Carrano, Inorg. Chem. 43
(2004) 7573.
In a typical experiment, 6.5 mg (1.26 ꢀ 10ꢁ5 mol) of the metal
complex and a 10-fold excess of PPh3 (6.29 ꢀ 10ꢁ3 mol) were dis-
solved in a 50:50 v/v mixture of DMF and d6-DMSO with a total
volume of 1.0 mL in an NMR tube. The 31P NMR kinetic experiment
was performed on a 400 MHz Varian FT-NMR spectrometer with
the automated temperature controller set at 30 °C. Prior to kinetic
measurements the NMR sample was equilibrated at 30 °C and
spectra were recorded every 15 min under PC control. The detec-
tion of dimethyl sulfide (DMS) by GC–MS confirmed the reduction
of DMSO via oxo abstraction from Mo(IV) intermediate. In control
experiments without metal complex under identical conditions,
OAT conversion of PPh3 to OPPh3 was not observed.
[13] B.S. Hammes, B.S. Chohan, J.T. Hoffman, S. Einwaechter, C.J. Carrano, Inorg.
Chem. 43 (2004) 7800.
[14] S.R. Davie, N.D. Rubie, B.S. Hammes, C.J. Carrano, M.L. Kirk, P. Basu, Inorg.
Chem. 40 (2001) 2632.
3.5. Computational methods
[15] J.T. Hoffman, B.L. Tran, C.J. Carrano, Dalton Trans (2006) 3822.
[16] C.J. Carrano, B.S. Chohan, B.S. Hammes, B.W. Kail, V.N. Nemykin, P. Basu, Inorg.
Chem. 42 (2003) 5999.
[17] K. Peariso, B.S. Chohan, C.J. Carrano, M.L. Kirk, Inorg. Chem. 42 (2003) 6194.
[18] B. Kail, V.N. Nemykin, S.R. Davie, C.J. Carrano, B. Hammes, P. Basu, Inorg. Chem.
41 (2002) 1281.
[19] M.A. Pietsch, M.B. Hall, Inorg. Chem. 35 (1996) 1273.
[20] P.D. Smith, A.J. Millar, C.G. Young, A. Ghosh, P. Basu, J. Am. Chem. Soc. 122
(2000) 9298.
[21] V.N. Nemykin, J. Laskin, P. Basu, J. Am. Chem. Soc. 126 (2004) 8604.
[22] V.N. Nemykin, P. Basu, Inorg. Chem. 44 (2005) 7494.
[23] A.J. Millar, C.J. Doonan, P.D. Smith, V.N. Nemykin, P. Basu, C.G. Young, Chem.
Eur. J. 11 (2005) 3255.
Full geometry optimizations, vibrational calculations, and
natural bond order (NBO) analyses were carried out with Gaussian
03 [28], using the B3LYP density functional method [29,30], with
Dunning’s cc-pVDZ basis set [31] for the main group atoms and
the CEP-121G effective core potential and basis set of Stevens
et al. [32] for the metal center. In the computations, PMe3 was used
in place of PPh3 for simplicity.
Acknowledgements
[24] B.W. Kail, L.M. Perez, S.D. Zaric, A.J. Millar, C.G. Young, M.B. Hall, P. Basu, Chem.
Eur. J. 12 (2006) 7501.
This work was supported in part by grants CHE-0313865 (CJC)
and CHE-0202535 (ALC) from the NSF. The NSF-MRI program
grants CHE-0320848 and CHE-0216563 are acknowledged for sup-
port of the X-ray diffraction and computational facilities, respec-
tively, at San Diego State University. We thank Dr. LeRoy Lafferty
for assistance with the 31P NMR kinetic studies.
[25] B.L. Tran, C.J. Carrano, Inorg. Chem. 46 (2007) 5429.
[26] B.L. Tran, C.J. Carrano, Inorg. Chim. Acta 2006 (1961) 360.
[27] J.T. Hoffman, Heteroscorpionate Model Complexes of Relevance to the Active
Sites of Various Molybdenum and Zinc Metalloenzymes (MS thesis), San Diego
State University, San Diego, 2005.
[28] M.J. Frish et al., Gaussian 03, Revision B.02, Gaussian, Inc, Wallingford, CT,
2004.
[29] A.D. Becke, J. Chem. Phys. 98 (1993) 5648.
[30] C. Lee, W. Yang, R.G. Parr, Phys. Rev. B 37 (1988) 785.
[31] T.H. Dunning Jr., J. Chem. Phys. 90 (1989) 1007.
[32] W.J. Stevens, M. Krauss, H. Basch, P.G. Jasien, Can. J. Chem. 70 (1992) 612–630.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.ica.2016.03.035.