A Cp-based Molybdenum Catalyst for the Deoxydehydration of Biomass-derived Diols

Article abstract of DOI:10.1002/cctc.202001115

Dioxo-molybdenum complexes have been reported as catalysts for the deoxydehydration (DODH) of diols and polyols. Here, we report on the DODH of diols using [Cp^*MoO₂]₂O as catalyst (Cp^*=1,2,3,4,5-pentamethylcyclopentadienyl). The DODH reaction was optimized using 2 mol % of [Cp^*MoO₂]₂O, 1.1 equiv. of PPh₃ as reductant, and anisole as solvent. Aliphatic vicinal diols are converted to the corresponding olefins by [Cp^*MoO₂]₂O in up to 65 % yield (representing over 30 turnovers per catalyst) and 91 % olefin selectivity, which rivals the performance of other Mo-based DODH catalysts. Remarkably, cis-1,2-cyclohexanediol, which is known as quite a challenging substrate for DODH catalysis, is converted to 30 % of 1-cyclohexene under optimized reaction conditions. Overall, the mass balances (up to 79 %) and TONs per Mo achievable with [Cp^*MoO₂]₂O are amongst the highest reported for molecular Mo-based DODH catalysts. A number of experiments aimed at providing insight in the reaction mechanism of [Cp^*MoO₂]₂O have led to the proposal of a catalytic pathway in which the [Cp^*MoO₂]₂O catalyst reacts with the diol substrate to form a putative nonsymmetric dimeric diolate species, which is reduced in the next step at only one of its Mo-centers before extrusion of the olefin product.

Full text of DOI:10.1002/cctc.202001115

The European Society Journal for Catalysis
Supported by
Accepted Article
Title: A Cp-based Molybdenum Catalyst for the Deoxydehydration of
Biomass-derived Diols
Authors: Jing Li, Martin Lutz, and Robertus Johannes Klein Gebbink
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemCatChem 10.1002/cctc.202001115
Link to VoR: https://doi.org/10.1002/cctc.202001115
01/2020

10.1002/cctc.202001115
ChemCatChem
FULL PAPER
A Cp-based Molybdenum Catalyst for the Deoxydehydration of
Biomass-derived Diols
Jing Li [a], Dr. Martin Lutz [b], and Prof. Dr. Robertus J. M. Klein Gebbink *[a]
[a]
J. Li, Prof. Dr. R. J. M. Klein Gebbink
Organic Chemistry and Catalysis
Debye Institute for Nanomaterials Science
Utrecht University
Universiteitsweg 99, 3584CG, Utrecht (The Netherlands)
E-mail: R.J.M.KleinGebbink@uu.nl
Dr. Martin Lutz
[b]
Crystal and Structural Chemistry
Bijvoet Center for Biomolecular Research
Faculty of Science, Utrecht University
Padualaan 8, 3584 CH, Utrecht (The Netherlands)
Supporting information for this article is given via a link at the end of the document.
Abstract: Dioxo-molybdenum complexes have been reported as
catalysts for the deoxydehydration (DODH) of diols and polyols. Here,
we report on the DODH of diols using [Cp*MoO2]2O as catalyst (Cp* =
1,2,3,4,5-pentamethylcyclopentadienyl). The DODH reaction was
optimized using 2 mol% of [Cp*MoO2]2O, 1.1 equiv. of PPh3 as
reductant, and anisole as solvent. Aliphatic vicinal diols are converted
to the corresponding olefins by [Cp*MoO2]2O in up to 65% yield
(representing over 30 turnovers per catalyst) and 91% olefin
selectivity, which rivals the performance of other Mo-based DODH
catalysts. Remarkably, cis-1,2-cyclohexanediol, which is known as
quite a challenging substrate for DODH catalysis, is converted to 30%
of 1-cyclohexene under optimized reaction conditions. Overall, the
mass balances (up to 79%) and TONs per Mo achievable with
[Cp*MoO2]2O are amongst the highest reported for molecular Mo-
based DODH catalysts. A number of experiments aimed at providing
insight in the reaction mechanism of [Cp*MoO2]2O have led to the
proposal of a catalytic pathway in which the [Cp*MoO2]2O catalyst
reacts with the diol substrate to form a putative nonsymmetric dimeric
diolate species, which is reduced in the next step at only one of its
Mo-centers before extrusion of the olefin product.
catalysts derived from these latter metals for the DODH of
(biomass-derived) diols and polyols.
For the homogeneous molybdenum-catalyzed DODH of diols and
polyols, a relatively limited number of examples have been
reported,[24] mainly including the use of molybdate salts[15–17,21] or
dioxo-Mo(VI) complexes.[14,18-20] Ammonium heptamolybdate
((NH4)6Mo7O24, AHM), a commercially available molybdate salt,
has been reported as a catalyst for the deoxydehydration of diols.
The first example was reported by Fristrup and co-workers in
2014.[21] Using 0.5 mol% of AHM in a reactive distillation set-up,
16% of 1-hexene was formed as well as 5% of 2-hexanone
(dehydration product of 1,2-hexanediol) in neat 1,2-hexanediol,
after heating the mixture at 190–220 °C. Under these conditions,
1,2-hexanediol serves as substrate, solvent, as well as sacrificial
reductant. Based on a density functional theory (DFT) study, the
authors proposed a catalytic cycle that involves the condensation
of two diols with a Mo(VI)-trioxo species to form an oxo-Mo(VI)
bisdiolate intermediate, subsequent oxidative cleavage of one of
the diolate ligands resulting in a reduced oxo-Mo(IV) monodiolate
species, formaldehyde and an aldehyde (for example
formaldehyde and pentanal through oxidative cleavage of 1,2-
hexanediol), followed by a final olefin extrusion step that also
forms back the trioxo-Mo(VI) species (Scheme 1, a).[17] Operating
through this proposed mechanism, this catalytic system forms
carbonyl compounds either via the oxidative cleavage of a diol or
via the dehydration of the diol. These carbonyl compounds can
further react with the diol substrate to generate ketals (Scheme 2).
For this reason, the olefin product selectivity of the DODH
catalyzed by AHM will be substantially lower compared to the Re-
catalyzed DODH of diols.
Introduction
Deoxydehydration (DODH), known as
a
combination of
deoxygenation and dehydration, provides an efficient means for
removing vicinal hydroxyl groups from diols and polyols to form
the corresponding olefins.[1] DODH reactions allow for high
oxygen-content starting materials, such as cellulosic biomass
derivatives, to be converted into useful olefins. Next to a number
of heterogeneous catalysts, soluble metal complexes based on
rhenium,[1–4] vanadium,[5–13] and molybdenum,[14–23] have been
reported to catalyze the DODH of diols and polyols. In terms of
activity and selectivity, rhenium complexes so far outperform
homogeneous catalysts based on vanadium and molybdenum.
However, the scarcity and high cost of rhenium asks for the
availability of alternative catalysts based on less expensive and
more abundant metals, such as Mo and V. It is for this reason that
there is a current interest in the development of homogeneous
The secondary alcohol isopropanol was then used as the
sacrificial reductant in the AHM-catalyzed deoxydehydration
reaction.[16] In this case, the reaction proceeded at 240-250 oC in
a pressurized autoclave, and the olefin yield from simple aliphatic
diols was as high as 77% when NBu4OH was added to the
reaction (46% of 1-hexene was formed from 1,2-hexanediol when
no base was applied). The authors claimed that the use of this
particular base would not only affect the alkene yield but also
change the product distribution. The use of a number of other
bases did not lead to an overall improvement in catalysis, which
1
This article is protected by copyright. All rights reserved.

10.1002/cctc.202001115
ChemCatChem
FULL PAPER
the authors explained as a result of the high solubility of the
(Bu4N)2MoO4 salt, while other molybdate salts are quite insoluble.
On the basis of DFT calculations, the reduction step under these
conditions was found to be most favorable when both a diolate
and an isopropoxide ion are coordinated to the molybdenum
center (Scheme 1, b). The computed catalytic cycle consists of
the condensation of the diol with MoO3 in two steps to form the
dioxo-Mo(VI) diolate, followed by coordination of the isopropoxide
and reduction to form an oxo-Mo(IV) diolate, water and acetone,
and finally olefin extrusion. These calculations furthermore
showed that the reduction of the metal center prior to extrusion of
the alkene is rate-limiting under these conditions. Not only
alcohols were investigated as reductant for AHM-catalyzed
DODH, Na2SO3 was also applied for this reaction. John and co-
workers reported on the Na2SO3-mediated AHM-catalyzed DODH
of diols using 5 mol% of AHM as catalyst and 1.5 equiv. of Na2SO3
as reductant. This allowed for 23% formation of styrene from 1-
phenylethane-1,2-diol.[21] For all these AHM-catalyzed DODH
systems, the TON per “MoO2” unit was quite low, at maximum
less than 5.
Only four examples have been reported of diol DODH catalyzed
by ligand-supported “MoO2” complexes. In 2013, a dioxo-Mo(VI)
complex bearing an acylpyrazolonate ligand was reported as a
DODH catalyst.[14] This complex is able to form 10% of styrene
from 60% conversion of 1-phenylethane-1,2-diol using a catalyst
loading of 2 mol% and 1.1 equiv. of PPh3 as the sacrificial
reductant. The reaction can be carried out at a reaction
temperature of only 110 oC, which is quite low for a Mo-catalyzed
DODH reaction. Octahedral dioxo-Mo(VI) complexes derived
from (OSSO)-type bis(phenolate) ligands were reported as DODH
catalyst precursors by Okuda and co-workers in 2016.[18] 1,4-
Anhydroerythritol was successfully converted into 2,5-
dihydrofuran with 49% product yield and 89% substrate
conversion using 5 mol% of Mo complex and 3-octanol as
reductant within 1 h at 200 oC under microwave irradiation.
According to the authors, microwave irradiation prevents the use
of high reaction temperatures and long reaction times. Last year,
MoO2(acac)2 catalyzed DODH reactions were reported by De Vos
et al.[19] It was shown that the addition of the β-diketone 2,2,6,6-
tetramethylheptanedione (TMHDH) leads to a strong increase in
yield of the desired olefin product. The excess amount of the
diketone ligand (4.0 equiv. on basis of substrate) was claimed to
stabilize the Mo complexes and inhibit catalyst decomposition
through oligomerization of Mo, which is the proposed
decomposition pathway of Mo DODH-catalysts. Very recently,
Kilyanek et al. reported a dioxo-molybdenum complex supported
by a dianionic ONO pincer ligand, which catalyzes the DODH of
diols using PPh3, Na2SO3, Zn, C, 3-octanol or 2-propanol as
reductant.[20] For these ligand-supported “MoO2” DODH catalysts
similarly low TONs per “MoO2” were found as for the AHM
systems; the highest TON being around 9.
O
R
R
HO
a
Mo
OH
O
O
H2O
R
R
O
O
Mo
O
O
Mo
O
O
O
R
HO
O
H
H
R
O
O
OH
+
O
Mo
O
H
O
H2
O
R
R
O
OH
O
b
O
Overall, there has not been a lot of activity in the research field of
homogeneous Mo-based DODH-catalysis. Next to the overall
number of reported studies, in particular the number of ligands
that have been applied in this field is rather limited.. From our
experience in Re-catalyzed DODH-reactions, substituted
OH
O
Mo
O
O
O
O
Mo
HO Mo
O
O
OH
O
O
O
O
cyclopentadienyl (Cp) ligands represent
a versatile ligand
platform for the development of different trioxo-Re complexes and
their application in DODH-catalysis.[25–27] Previous studies have
shown that the catalytic properties of CpReO3 type complexes in
DODH reactions are affected by modifications of the Cp ligand.
We were therefore curious to investigate if the corresponding Cp-
based Mo-complexes would also show catalytic activity towards
the deoxydehydration of diols and polyols.
H2
O +
O
OH
O
Mo
O
O
O
H2O
O
Mo
O
O
O
O
O
OH
Scheme 1. Proposed catalytic cycles for AHM-catalyzed DODH of diols based
on DFT studies: a) using the diol itself as reductant,[17] b) using iPrOH as
reductant.[16]
Results and Discussion
OH
O
O
ꢀꢁꢂꢃ
+
+
+
H2O
2
R
OH
R
R
H
H
H
For our studies on Cp-based molybdenum-oxo complexes in
DODH catalysis, we have considered the use of a number of
different complexes. The dinuclear oxo-bridged complex
[Cp*MoO2]2O[28] was first reported by Herberhold and Thewalt et
al. in 1985.[29] This complex can be synthesized either by direct
oxidation of suitable low-valent precursors (typically carbonyl-
based precursors), or by hydrolysis of easily accessible high-
valent precursors. [28] This oxophilic complex has earlier been
reported as a catalyst for the epoxidation of olefins.[30,31] The
bulkier [CpttMoO2]2O complex, derived from the 1,3-di-tert-
butylcyclopentadienyl ligand, was also synthesized according to
a literature procedure starting from Mo(CO)6.[32,33] The substituted
side reactions:
O
OH
OH
ꢀꢁꢂꢃ
+
H2O
H2O
R
R
OH
ꢀꢁꢂꢃ
O
+
OH
R
R
R2
R1
OH
R2
O
+
+
H2O
O
OH
O
R
R1
R
Scheme 2. Mo-catalyzed DODH of diols and side reactions observed when
using the diol itself as sacrificial reductant.[15]
2
This article is protected by copyright. All rights reserved.

10.1002/cctc.202001115
ChemCatChem
FULL PAPER
cyclopentadiene preligand CpttH was refluxed with Mo(CO)6 in p-
xylene overnight under nitrogen to form [CpttMo(CO)2]2, which
was oxidized by stirring under air for 3 days to obtain the target
dinuclear [CpttMoO2]2O complex as yellow needle-like crystals.
Previous work discussed above has shown that the active site of
Mo-based DODH-catalysts is based on the “MoO2” motif.
Accordingly, these bridged Cp-based MoO2 dimers could be
interesting catalyst candidates for DODH reactions.
molybdenum species under aerobic conditions. Last but not the
least, 30% conversion of 1,2-octanediol was observed in the
absence of any catalyst under the reaction conditions, albeit
without olefin formation. This result indicates that thermal
substrate degradation can take place to a substantial extent
during the reaction. So far, we have not been able to track down
which organic products form during this thermal reaction.
In conclusion, from all the Mo catalysts we have tested here,
[Cp*MoO2]2O gave the highest 1-octene yield as well as alkene
product selectivity. Accordingly, [Cp*MoO2]2O was used in further
optimization studies. Its higher reactivity compared to that of the
mononuclear trioxo complex NBu4[Cp*MoO3] could suggest that it
operates via a dinuclear mechanism, instead of breaking up in
monomeric entities during catalysis (vide infra). If so, the reactivity
of [Cp*MoO2]2O is significantly higher than that of MoO2(acac)2
per catalytic entity. The low to moderate alkene product selectivity
indicates that side reactions take place when using [Cp*MoO2]2O
as catalyst. Except 1-octene, 2,4-dihexyl-1,3-dioxolane and 2-
heptyl-4-hexyl-1,3-dioxolane were detected by GC and GC-MS
(vide infra). These two products indicate the formation of 1-
heptanal (oxidative C–C bond cleavage product of 1,2-octanediol)
and 1-octanal (dehydration product of 1,2-octanediol).
Accordingly, the side reactions for AHM-catalyzed DODH reaction
described by Fristrup15 also take place for the [Cp*MoO2]2O
catalyst. In addition, triphenylphosphine oxide (OPPh3) was also
detected by GC and GC-MS, which indicates that PPh3 can also
act as reductant for this catalyst. Accordingly, both PPh3 and 1,2-
octanediol may act as the reductant in the deoxydehydration of
1,2-octanediol by [Cp*MoO2]2O.
Besides, considering that the Cp-based trioxo-rhenium catalysts
are mononuclear species, the [Cp*MoO3]– anion would be a close
structural analogue of the CpRe-catalysts. On basis of a literature
report, NBu4[Cp*MoO3] was synthesized by slowly adding a
NBu4OH solution into a [Cp*MoO2]2O solution in THF, to form
colorless crystals of NBu4[Cp*MoO3] after recrystallization in THF
o
at –30 C.[34] In this case, it would be interesting to investigate
which effect the difference in overall charge of the complex will
have on catalytic activity. In addition, no trioxo-molybdenum
complexes have so far been investigated for DODH reactions.
Alternatively, the Cp-based ligand could be replaced by a different
and neutral 6e– spectator ligand, for example Me3tacn (1,4,7-
trimethyl-1,4,7-triazacyclononane). The complex (Me3tacn)MoO3
has been synthesized on the basis of a literature procedure
reported by McGowan et al. from 2001.[35]
Accordingly, we started our investigations on Mo-based DODH-
catalysis by screening a number of reported Mo-complexes for
their activity in the DODH reaction of 1,2-octanediol to form 1-
octene, using PPh3 as the reductant and anisole as solvent. The
reaction was performed at 200 oC under an N2 atmosphere for 15
h in a closed pressure tube with a catalyst loading of 4% based
on Mo (Table 1). When [Cp*MoO2]2O and [CpttMoO2]2O were used
as catalyst, they gave almost the same 1-octene yield (55%, 51%)
Table 1. Mo-catalyzed DODH of 1,2-octanediol with different Mo complexes.[a]
at full substrate conversion (entries
1 and 2). For the
entry
[Mo]
yield [b]
(%)
conversion[b]
(%)
alkene
selectivity
(%)
corresponding mononuclear species, NBu4[Cp*MoO3], the 1-
octene yield was 38% at full substrate conversion (entry 3). Yet,
changing catalyst to the trioxo-complex (Me3tacn)MoO3 gave a 1-
octene yield of only 9% at 38% of substrate conversion (entry 4).
This observation suggests that the anionic Cp-ligand is preferred
over a neutral 6e– ligand like Me3tacn in enabling DODH reactivity
with Mo. Remarkably, when the commercially available
MoO2(acac)2 complex was applied under our reaction conditions,
the 1-octene yield was 47% with full substrate conversion, which
was also quite acceptable (entry 5). The commercially available
AHM was also tested here and 38% of 1-octene had formed from
87% of 1,2-octanediol conversion (entry 6).
1
2
[Cp*MoO2]2O
[CpttMoO2]2O
NBu4[Cp*MoO3]
(Me3tacn)MoO3
MoO2(acac)2
55
51
38
9
>99
>99
>99
38
55
51
38
24
47
44
38
38
3
4
5
47
38
38
38
>99
87
6
(NH4)6Mo7O24·4H2O
[Cp*MoO2]2O
7[c]
8[d]
>99
>99
[Cp*MoO2]2O
[a] Reaction conditions: 1,2-octanediol (0.5 mmol), PPh3 (0.55 mmol, 1.1
equiv.), Mo complex (4 mol% on the basis of Mo), anisole (5 mL), 200 oC
(temperature of oil bath), 15 h, N2, closed pressure tube. Reported values are
the average of three independent runs. [b] Yield of 1-octene, determined by gas
chromatography using mesitylene (0.5 mmol) as an internal standard. [c] 6
mol% of NBu4OH was added as a 1 M solution of NBu4OH in methanol. [d]
Under aerobic conditions.
Among the molybdenum complexes we have tested,
[Cp*MoO2]2O gave the highest olefin yield as well as alkene
product selectivity, but still not as high as in some rhenium-
catalyzed DODH reactions. Since the use of NBu4OH was
reported to significantly improve the olefin yield for AHM-
catalyzed DODH,[16] we examined the effect of this additive. Yet,
the use of NBu4OH did not increase the olefin yield when
[Cp*MoO2]2O was used as catalyst (entry 7). The olefin yield and
substrate conversion in this case were the same as when using
NBu4[Cp*MoO3] as the catalyst (entry 3). This result is in line with
the visual changes while setting up the reaction: the reaction
mixture changed colour from bright yellow to colourless when
adding NBu4OH, which indicates the in situ formation of
NBu4[Cp*MoO3]. Furthermore, 38% of 1-octene was formed when
[Cp*MoO2]2O was used as catalyst under aerobic condition (entry
8). This finding is suggestive of the formation of trioxo-
Continuing, different solvents were investigated in the DODH
reaction of 1,2-octanediol with [Cp*MoO2]2O as catalyst (Table 2).
Under neat conditions, 26% of 1-octene formed at 35% substrate
conversion and an alkene product selectivity of 74% (entry 1).
When substituted benzenes were used as the solvent, the 1-
octene yield varied from 36% to 65% (entries 2-10), while very
high substrate conversions of >90% could be reached for several
solvents. The highest 1-octene yield (65%, representing over 30
turnovers per catalyst) was obtained by using m-dichlorobenzene
as the solvent (entry 4), while the highest alkene selectivity (91%)
was achieved by using 1,2,4-trichlorobenzene as solvent (entry
3
This article is protected by copyright. All rights reserved.

10.1002/cctc.202001115
ChemCatChem
FULL PAPER
7). In the latter solvent, a somewhat lower substrate conversion
of 44% was found after a reaction time of 15 h. In order to reach
higher substrate conversion in this solvent, the reaction time was
prolonged to 30 h, which gave an overall conversion of 59% and
a 1-octene yield of 52%, leading to a slightly decreased alkene
product selectivity of 88% (entry 12). This result indicates that
catalyst decomposition is likely to occur during prolonged reaction
times. When either anisole or mesitylene were used as the
reaction solvent, full substrate conversion resulted in 55% and
52% 1-octene yield, respectively (entries 9 and 10).
agents were tested for the DODH-reaction of 1,2-octanediol with
[Cp*MoO2]2O as the catalyst (Table 3). Interestingly, when the
more reducing PnBu3 was used as reductant, a lower 1-octene
yield (39%) was obtained compared to the use of PPh3 (entries 1
and 2). The less reducing triethylphosphite gave a low 9% yield,
albeit at full substrate conversion (entry 3). The use of indoline,
Na2SO3, 3-octanol, or zinc powder as reductant gave 1-octene
yields in the range of 20-32%. Without an additional reductant, i.e.
only using the diol itself as reductant, 18% of 1-octene was formed
(entry 8). Considering that the maximum yield of 1-octene is 50%
in this case, the alkene product selectivity was 36%. On the basis
of these findings, PPh3 was selected as the reductant for further
studies.
Although the halogenated benzene solvents m-dichlorobenzene
and 1,2,4-trichlorobenzene gave the highest alkene yield and the
highest alkene selectivity, respectively, these solvents were
disregarded in our further catalytic studies because of the
For most of the above reactions, very high substrate conversions
implications of their high halogen-content on practical applications. were observed, indicating that other products than 1-octene
Alternatively, the Innovative Medicines Initiative (IMI)-Chem21, a
public–private partnership of pharmaceutical companies,
universities and small-to-medium enterprises supporting research
dominate for most reducing agents. As described in the
introduction, this phenomenon is not only observed for our Cp-
based molybdenum system. Without excess amount of ligand or
additives, low to moderate alkene product selectivities are
typically obtained for Mo-catalysts. The use of excesses of
organic ligands has been found to affect the alkene product
selectivity. In case of the MoO2(acac)2-catalyzed DODH of 1,2-
hexanediol, the yield of 1-hexene dramatically increased when 40
equiv. of TMHDH (on the basis of MoO2(acac)2) were added. The
alkene product selectivity was 93% and the TON per Mo was 9.[19]
By tuning the ligand on molybdenum increased alkene product
selectivities can also be obtained. The alkene product selectivity
is 70% with (OSSO)-type bis-phenolate ligands in the DODH of
1,4-anhydroerythritol, and 60% with a dianionic ONO pincer-type
ligand in the DODH of 1,2-octanediol. The TON per Mo in these
cases is 11 and 6, respectively.[18,20] In our case, the application
of the Cp* ligand also gives a quite good alkene product selectivity
(55%) and TON per catalyst (27; i.e. 14 per Mo).
into
sustainable
pharmaceuticals
manufacturing
(http://www.chem21.eu/), has listed anisole as a recommended
solvent on the basis of safety (S), occupational health (H),
environment (E), quality (risk of impurities in the drug substance),
industrial constraints (e.g. boiling point, freezing temperature,
density, recyclability), and cost.[36,37] Accordingly, we set out to
evaluate anisole as the solvent of choice for further catalyst
optimization.
Table 2. [Cp*MoO2]2O-catalyzed DODH of 1,2-octanediol in different solvents.[a]
entry
solvent
yield[b]
(%)
26
conversion[c]
selectivity[e]
b.p.[f]
(oC)
(%)
35
94
95
92
(%)
74
39
38
70
1
2
3
4
--
PhCl
--
37
132
111
172-
173
178-
180
138
214
169
154
163-
166
285
214
toluene
36
m-
65
Table 3. [Cp*MoO2]2O-catalyzed DODH of 1,2-octanediol with different
reductants. [a]
dichlorobenzene
o-
5
46
70
66
entry
reductants
yield[b]
(%)
conversion[b]
(%)
alkene
selectivity
(%)
dichlorobenzene
p-xylene
trichlorobenzene
tBuPh
6
7
36
40
39
55
52
97
44
40
91
48
55
52
1
2
PPh3
PnBu3
55
39
9
>99
>99
>99
77
55
39
9
8
82
9
10[d]
anisole
>99
>99
3
P(OEt)3
indoline
Na2SO3
3-octanol
Zn
mesitylene
4
26
24
20
32
18
34
24
26
34
36
5
98
11
12[g]
sulfolane
5
28
59
18
88
trichlorobenzene
52
6
76
7
8[c]
94
[a] Reaction conditions: 1,2-octanediol (0.5 mmol), PPh3 (0.55 mmol, 1.1
equiv.), [Cp*MoO2]2O (0.01 mmol, 2 mol%), solvent (5 mL), 200 oC (temperature
of oil bath), 15 h, N2, closed pressure tube. Reported values are the average of
1,2-octanediol
>99
[a] Reaction conditions: 1,2-octanediol (0.5 mmol), reductant (0.55 mmol, 1.1
equiv.), [Cp*MoO2]2O (0.01 mmol, 2 mol%), anisole (5 mL), 200 oC (temperature
of oil bath), 15 h, N2, closed pressure tube. Reported values are the average of
three independent runs. [b] Determined by gas chromatography using
mesitylene (0.5 mmol) as an internal standard. [c] 3-Octanol (5 mL) was used
as solvent instead of anisole.
three independent runs.
[b] Yield of 1-octene, determined by gas
chromatography using mesitylene (0.5 mmol) as an internal standard. [c]
Determined by gas chromatography using mesitylene (0.5 mmol) as an internal
standard. [d] Determined by 1H NMR using dibromomethane (0.5 mmol) as an
internal standard. [e] Alkene product selectivity. [f] All solvents were purchased
from Sigma-Aldrich, boiling points are reported on the basis of corresponding
Safety Data Sheet. [g] Reaction time: 30 h.
Given the fact that full substrate conversions were observed in
many of the reactions described in Tables 1-3, the high reaction
temperature that was employed could be the reason for the lower
alkene product selectivities. Accordingly, different reaction
temperatures were investigated (Table 4). No conversion was
As mentioned above, both PPh3 and 1,2-octanediol can act as
sacrificial reductant in the DODH reaction catalyzed by
[Cp*MoO2]2O. In order to enhance the formation of 1-octene, both
the dehydration of 1,2-octanediol and the oxidative cleavage of
1,2-octanediol should be limited. The use of a more competitive
reductant will then benefit alkene formation. Different reducing
o
obtained at temperatures below 170 C. When the reaction was
o
performed at 170 C, only 9% of 1-octene was formed while the
4
This article is protected by copyright. All rights reserved.

10.1002/cctc.202001115
ChemCatChem
FULL PAPER
substrate conversion was also 9%, which means that the alkene
product selectivity was 100% albeit at a TON per Mo of only 2.
When the reaction temperature was increased to 185 C, the 1-
forming ketone and aldehyde byproducts.[15,16] Furthermore,
ketals can be formed as by-products in reactions where carbonyl
compounds are formed (Scheme 3). For our [Cp*MoO2]2O-
catalyzed DODH reactions, ketals were indeed detected by a
combined GC and GC-MS analysis (Table 5), which is in line with
findings by the Fristrup group.[15] Two different ketals actually
formed during these reaction: one derived from 1,2-octanediol
and heptanal (the oxidative cleavage product of 1,2-octanediol)
and one derived from 1,2-octanediol and octanal (the dehydration
product of 1,2-octanediol).
o
octene yield increased from 9% to 13%, while the 1,2-octanediol
conversion increased from 9% to 36%, so the alkene product
selectivity dropped dramatically from 100% to 36%. When the
o
reaction temperature was further raised to 190 C, the 1-octene
yield was 24%, the 1,2-octanediol conversion was 62%, and the
alkene product selectivity of 39% was very close to the one at 185
o
oC. At an elevated reaction temperature of 225 C, the same 1-
octene yield was achieved compared to the reaction carried out
o
at 200 C. All the reactions described here were performed in a
In order to increase the alkene product selectivity, or to avoid the
formation of by-products, several bases were tested by Fristrup
and co-workers. The use of NBu4OH did significantly increase the
yield of olefin as well as change the distribution of products for the
AHM catalyst.[16] We did also use NBu4OH as an additive for our
reaction system, but did not notice an increase in the olefin yield
(Table 2, entry 7). However, the use of NBu4OH seems to change
the distribution of product and by-products (Table 5, entry 3). Most
importantly, less formation of 1-octene was noticed after applying
this base (from 55% to 38%). In addition, there are some
differences in the product distribution between directly using
NBu4[Cp*MoO3] as catalyst and forming NBu4[Cp*MoO3] in situ,
i.e. by applying a slight excess amount of base (compare Table 5,
entries 2 and 3). It seems that less diol dehydration occurs,
leading to a different ketal distribution; 1-octene formation is the
same though in these cases. Based on the ratio of different ketals
formed, we are not able to distinct the operation of a dinuclear
catalyst from that of a mononuclear catalyst, even though a
significantly higher amount of olefin is formed when starting from
a dinuclear catalysis.
closed pressure tube and the temperatures are based on the
temperature of the oil bath, which means neither the actual
temperature nor the pressure in the reaction tube could be
monitored in this experimental setup. Table 4 clearly shows that
an oil bath temperature of 200 oC is most optimal under the
current reaction conditions. Comparing entry 4 and entry 5, olefin
formation more than doubled by a 10 oC change in temperature.
Moreover, the alkene product selectivity increased when the
temperature was increased from 190 oC to 200 oC. An explanation
for the latter observation could be that ketal formation during the
reaction is reversible and that ketal formation is less favored at
higher temperatures. Accordingly, a control experiment was done
by using the ketal as substrate, but no ketal conversion was
observed (Scheme 3), likely due to the lack of water in this control
experiment. Water is required in the diol-ketal equilibrium to form
the diol from the ketal.
Table 4. [Cp*MoO2]2O-catalyzed DODH of 1,2-octanediol at different
temperature. [a]
entry
temperrature
yield[b]
(%)
conversion[b]
(%)
alkene selectivity
(%)
Table 5. Product profile in the DODH of 1,2-octanediol using different Mo
catalysts. [a]
1
2
3
4
5
170 oC
185 oC
190 oC
200 oC
225 oC
9
9
100
36
entry
[Mo]
conversion
%
mass
balance
%
13
24
55
55
36
C6H13
O
C7H15
O
C6H13
62
39
O
O
C6H13
C6H13
>99
>99
55
%
%
%
55
1
2
3[b]
[Cp*MoO2]2O
NBu4[Cp*MoO3]
[Cp*MoO2]2O
55
38
38
6
7
6
8
4
>99
>99
>99
79
68
68
[a] Reaction conditions: 1,2-octanediol (0.5 mmol), PPh3 (0.55 mmol, 1.1
equiv.), [Cp*MoO2]2O (0.01 mmol, 2 mol%), anisole (5 mL), 15 h, N2, in closed
pressure tube. Reported values are the average of three independent runs. [b]
Determined by gas chromatography using mesitylene (0.5 mmol) as an internal
standard.
11
[a] Reaction conditions: 1,2-octanediol (0.5 mmol), PPh3 (0.55 mmol, 1.1 equiv.),
[Mo] (4 mol% on basis of Mo atom), anisole (5 mL), 200 oC, 15 h, N2, closed
pressure tube. Reported values are the average of three independent runs.
Products were determined by gas chromatography using mesitylene (0.5 mmol)
as an internal standard. [b] 6 mol% of NBu4OH was added as a 1 M solution of
C6H13
[Cp*MoO2]2O 2 mol%
NBu4OH in methanol. [c]. mass balance = ([yield
+ 2*yield
(2,4-dihexyl-1,3-
(1-octene)
O
+
PPh3
no conversion
O
anisole, 200 oC, N2, 15 h
dioxolane) + 2*yield (2-heptyl-4-hexyl-1,3-dioxolane)] / conversion)*100%.
C6H13
1.1 equiv.
Scheme 3. Use of 2,4-dihexyl-1,3-dioxolane as a DODH substrate to form 1-
octene.
The quantification of ketal by-products provides a clearer view of
the overall mass balance of these reactions. Although far from
being perfect, a mass balance close to 80% was found for the
[Cp*MoO2]2O catalyst. Since we observed 30% of ‘thermal’ 1,2-
octanediol conversion, i.e. in the absence of catalyst, the
observed mass balances are quite acceptable. Last but not least,
2-octanone, i.e. the dehydration product of 1,2-octanediol, was
difficult to detect because of peak overlap with solvent in the GC
spectra. On the basis of Fristrup’s work, an approximately equal
amount of 2-decanone (detected as 2-decanone and 2-decanol)
and decanal (detected as 1-decanol) was obtained.[16] This result
As mentioned before, poor to moderate alkene product
selectivities are generally obtained for Mo-catalyzed DODH
reactions. For these reactions, it is believed that reduction of Mo
via oxo-abstraction and via deformylation are competitive,[20,23]
which could explain why the alkene product selectivity is not very
high. Also, it is known from literature that dehydration of the diol
takes place in Mo-catalyzed DODH reactions as a side-reaction,
5
This article is protected by copyright. All rights reserved.

10.1002/cctc.202001115
ChemCatChem
FULL PAPER
suggests that not only octanal but also 2-octanone will form from
1,2-octanediol while using a Mo-based catalyst.
[Cp*MoO2]2O did completely dissolve, resulting in a biphasic
mixture during the reaction. Normally, a color changing from bright
yellow to yellow-orange and ending with light brown was observed
for this [Cp*MoO2]2O-catalyzed DODH. But in case of the reaction
with glycerol, the anisole phase containing the catalyst
maintained a bright yellow color, which indicated that no catalysis
took place in this case. Neat conditions were also applied to this
reaction, i.e. glycerol was explored as solvent, substrate, and
reductant. [Cp*MoO2]2O did dissolve in glycerol at higher
temperatures, but because of the reflux of glycerol in the top part
of the reaction tube, a black residue was obtained in the bottom
of the reaction tube without formation of allyl alcohol.
Table 6. [Cp*MoO2]2O-catalyzed DODH of diols using PPh3 as reductant.[a]
[b]
entry
substrate
Yieldalkene
(%)
conversion[b]
(%)
alkene
selectivity
(%)
OH
1
2
3
4
5
6
36
30
22
56
51
47
93
39
OH
OH
>99
>99
>99
>99
>99
30
22
56
51
47
OH
OH
The above study shows that the [Cp*MoO2]2O complex can be
used as a catalyst for the DODH of aliphatic diols using PPh3 as
reductant. In previous work of our group on related Re-based
DODH catalysts, dissociation of the Cp-ligand was observed
when using Cp’ReO3 type complexes as catalyst,[27] in addition to
an induction period in the case of using CptttReO3 and Cp*ReO3
as catalyst.[26] Accordingly, we were wondering if ligand
dissociation accompanied by an induction period would also be
observed when using [Cp*MoO2]2O as catalyst. A time course
profile experiment was carried out by using 1,2-octanediol as the
substrate, 5 mol% of [Cp*MoO2]2O as catalyst, 1.1 equiv. of PPh3
as reductant, and deuterated toluene as the solvent in a J-Young
OH
C4H9
OH
OH
C8H17
OH
OH
C10H21
OH
OH
O
7
8
5
--
--
--
--
HO
OH
OH
--
HO
OH
o
NMR tube under N2 atmosphere at 200 C. Progression of the
[a] Reaction conditions: diol (0.5 mmol), PPh3 (0.55 mmol, 1.1 equiv.),
[Cp*MoO2]2O (0.01 mmol, 2 mol%), anisole (5 mL), 15 h, N2, closed pressure
tube. Data in this table is based on a single catalytic run. [b] Determined by gas
chromatography using mesitylene (0.5 mmol) as an internal standard.
reaction over time was monitored by 1H NMR. In time intervals of
10 to 20 minutes, the reaction mixture was taken out of the
heating bath and cooled to room temperature to record an 1H
NMR spectrum, after which the mixture was re-heated to allow the
reaction to progress further. As shown in Figure 1, the reaction
showed a zeroth-order rate profile without an induction period.
After 10 min, 2% of 1-octene was detected, while after 260 min
44% of 1-octene had formed. Because of the way the reaction
was carried out, the final 1-octene yield was not as high as
Next, the substrate scope of the [Cp*MoO2]2O catalyst was
investigated by applying different diols to the optimized reaction
conditions. Both linear and cyclic aliphatic diols were tested
(Table 6). In the case of linear diols, when 1,2-hexanediol was
used as substrate, the 1-hexene yield was only 22% (entry 3).
Considering 1-hexene is quite a volatile chemical, we assume that
the actual yield of 1-hexene could be higher. Other aliphatic diols
with longer carbon chain lengths, such as 1,2-decanediol and 1,2-
dodecanediol, were therefore tested (entry 4 and 5). The yield of
1-decene and 1-dodecene in these cases is very close to the
olefin yield obtained from 1,2-octanediol under the same reaction
conditions (55%). When cis-1,2-cyclohexanediol and cis-1,2-
cyclooctanediol were used as substrate, the corresponding olefin
yield was 36% and 30%, respectively. Remarkably, for the
CptttReO3-catalyzed DODH of cis-1,2-cyclohexanediol, only 10%
of 1-cyclohexene is formed.[25] In case of using 1,4-
anhydroerythritol as substrate, only 5% of 2,5-dihydrofuran was
formed, mainly because of the poor solubility of 1,4-
anhydroerythritol in anisole. In short, the [Cp*MoO2]2O catalyst
system can be applied to the deoxydehydration of linear and
cyclic aliphatic diols to form the corresponding olefins with
moderate yields. Attempts to extend the substrate scope to
aromatic diols lead to 47% of styrene formation when starting from
1-phenyl-1,2-ethanediol, but in case of (1R,2R)-1,2-
diphenylethane-1,2-diol and (1R,2S)-1,2-diphenylethane-1,2-diol
as substrate, only benzaldehyde was formed as a detectable
product. This observation indicates that Mo-catalyzed cleavage of
aromatic diols apparently outperforms DODH pathways.[38,23]
Additionally, biomass-derived glycerol was investigated under the
optimized reaction conditions (entry 8). No allyl alcohol was
detected, likely due to the poor solubility of glycerol in anisole. No
glycerol was dissolved in anisole during the reaction while
1
reported in Table 1. No free ligand was detected by H NMR in
this time course experiment. On the other hand, free ligand was
detected by GC-MS analysis at the end of the experiment. These
observations suggest that the Cp* ligand does not dissociate in
the course of the DODH reaction, but does dissociate during the
GC-measurement.
Figure 1. Time profile of the [Cp*MoO2]2O catalyzed DODH of 1,2-octanediol.
For our [Cp*MoO2]2O catalyst system both mononuclear as well
as dinuclear catalytic pathways can be considered.
A
mononuclear pathway similar to the one put forward for the
Cp’ReO3/PPh3 system would involve the formation of either the
[Cp*MoO2]+ cation or the [Cp*MoO3]– anion as the active species.
These species could form simultaneous upon cleavage of the
6
This article is protected by copyright. All rights reserved.

10.1002/cctc.202001115
ChemCatChem
FULL PAPER
Mo–O–Mo linkage in [Cp*MoO2]2O. Either active species would
then engage in a sequence of diol condensation and Mo=O
reduction (in either order), followed by olefin extrusion. However,
when the mononuclear complex NBu4[Cp*MoO3] (as well as the
related neutral complex (Me3tacn)MoO3) were explored as
catalyst for the DODH of 1,2-octanediol, a lower 1-octene yield
was obtained (Table 1, entry 3 and 4), compared to the ones using
dinuclear complexes as catalyst (Table 1, entry 1 and 2). In
addition, O-atom abstraction from the [Cp*MoO2]+ cation is
expected to be energetically rather unfavourable, which rules out
this mononuclear dioxo cation as the active species. These
considerations suggest that the catalytic cycle going through a
mononuclear pathway is not favourable for the [Cp*MoO2]2O
catalyst system.
conditions. In addition, these experiments clearly show that unlike
in the case of Cp’ReO3 catalyzed DODH reactions, reduction of
[Cp*MoO2]2O does not take place readily. These observations are
in line with the work of Fristrup, which showed that the reduction
of Mo=O is the rate limiting step in catalytic DODH reactions using
AHM as catalyst and isopropanol as reductant.[16]
R
Red.
Cp*
Cp*
O
Cp*
O
Cp*
O
O
ORed.
Mo
Mo
Mo
Mo
O
R
O
O
O
O
R
H2O
Cp*
Cp*
O
O
Pathway A
Pathway B
Mo
Mo
O
O
R
HO
OH
Dinuclear catalytic pathways for [Cp*MoO2]2O could either start by
an initial reduction step or an initial diol condensation step. In view
of the mechanism put forward for the related Cp’ReO3/PPh3
catalyst system, in which reduction precedes diol condensation,
we considered pathways in which the [Cp*MoO2]2O catalyst is
initially reduced to the symmetric MoVMoV dimer [Cp*MoO2]2
(Scheme 4). In pathway A, [Cp*MoO2]2O is reduced to form
[Cp*MoO2]2, which will then condense with the diol to form
[Cp*Mo(diolate)](µ-O)[Cp*MoO2], followed by extrusion of the
olefin product to regenerate [Cp*MoO2]2O. In this case, half of the
[Cp*MoO2]2O complex is ‘untouched’ during catalysis. In pathway
B, [Cp*MoO2]2O is reduced in two steps to initially form
[Cp*MoO2]2 and subsequently [Cp*Mo]2(µ-O)3. This kind of
“Mo2(µ-O)3” moiety commonly exists in molybdates.[39] The doubly
reduced [Cp*Mo]2(µ-O)3 will then condense with the diol to form
an unsymmetrical [Cp*Mo(diolate)](µ-O)[Cp*MoO] complex,
followed by olefin product extrusion to reform [Cp*MoO2]2. Within
this binuclear reaction manifold, the singly reduced [Cp*MoO2]2
complex connects pathways A and B. Pathway A proceeds
through a MoVIMoVI « MoVMoV redox cycle including the non-
symmetric mixed-valent MoVIMoIV species [Cp*Mo(diolate)](µ-
Cp*
O
Cp*
O
Mo
Mo
O
O
O
Mo
Cp*
O
O
Mo
Cp*
O
R
R
HO
Red.
H2
O
OH
ORed.
Scheme 4. Proposed dinuclear pathways A and B for DODH-catalysis by
[Cp*MoO2]2O.
For the reduction of Mo=O units, a vicinal diol can also be used
as the reductant according to literature reports.[15,17] When heating
a 1:2 ratio of [Cp*MoO2]2O and 1,2-octanediol at 120 oC, 1H NMR
analysis showed the appearance of signals at 1.83 ppm and 1.79
ppm, again indicating the formation of [Cp*MoO2]2. In addition, no
indication for the formation of 1-octene was obtained for this
reaction. In a similar reaction, equimolar amounts of [Cp*MoO2]2O
and 1-phenyl-1,2-ethanediol were reacted in toluene-D8. By
increasing the temperature from room temperature to 100 oC, new
1
signals were detected by H NMR. After keeping the mixture at
100 oC overnight, 6% of styrene had formed, while no
acetophenone or aldehydes were detected and a significant
amount of 1-phenyl-1,2-ethanediol was still present.
O)[Cp*MoO2], while pathway B proceeds through a MoVMoV
«
These reduction experiments, combined with our results for the
reductant screening (Table 3, entry 1 and entry 8), as well as the
formation of ketals in our system, indicate that phosphines as well
as diols can be used as reductant for catalytic DODH reactions
using [Cp*MoO2]2O. In addition, the experiments confirm the
formation of [Cp*MoO2]2 as a reduction product from [Cp*MoO2]2O.
Next, our experiments focused on the reactivity of [Cp*MoO2]2. If
[Cp*MoO2]2 is indeed the reduced product of [Cp*MoO2]2O
involved in the proposed catalytic cycle, it should be active in
DODH catalysis. Therefore, we investigated [Cp*MoO2]2 as
catalyst for the deoxydehydration of 1,2-octanediol under
standard reaction conditions. Analysis of this reaction indicated
that only 8% of 1-octene had formed at 77% substrate conversion
(Scheme 5a). Since it is known that [Cp*MoO2]2 is oxidized to
[Cp*MoO2]2O under aerobic conditions,[28] we performed the same
reaction under aerobic conditions (Scheme 5b). In this case, 30%
of 1-octene formed at full substrate conversion, indicating that
[Cp*MoO2]2O formed in situ under these aerobic conditions is
responsible for catalysis. These experiments cast doubts on the
overall involvement of [Cp*MoO2]2 as an intermediate in DODH
catalysis by [Cp*MoO2]2O, and, furthermore, seem to rule out
pathway B, since in that case [Cp*MoO2]2O would act as a pre-
catalyst and catalytic turn-over could be achieved through
[Cp*MoO2]2 without [Cp*MoO2]2O being involved at all.
MoIVMoIV redox cycle without the involvement of a mixed-valence
intermediate. Likewise, for pathway A the starting complex
[Cp*MoO2]2O can be considered as the actual catalyst, while in
pathway B it would be a pre-catalyst. The absence of an induction
period in the time course study may not support pathway B from
this point of view. Since the dinuclear MoV complex [Cp*MoO2]2 is
pivotal is both mechanisms, our attention focussed on addressing
the possible role of [Cp*MoO2]2 in DODH catalysis.
First, we considered the formation of [Cp*MoO2]2 through the
reduction of [Cp*MoO2]2O. Upon treatment of [Cp*MoO2]2O with
PPh3 (the reductant used in DODH catalysis), the progression of
the reaction was very slow at 120 oC; only 27% PPh3 conversion
occurred after 2 h, with the colour changing from bright yellow to
brown. Remarkably, the reduction of Cp’ReO3 by PPh3 happens
instantly at room temperature. Treatment of [Cp*MoO2]2O with 2.0
o
equiv. of the stronger reductant P(nBu)3 in toluene-D8 at 80 C
under N2 overnight gave a brown solution, and 55% of P(nBu)3
was consumed based on 31P NMR. Monitoring the reaction by
means of 1H NMR showed the appearance of new signals at 1.83
ppm and 1.79 ppm. A red-orange crystal separated from the
mixture and a solution of the crystal showed a signal at 1.79 ppm
in 1H NMR. The identity of the red-orange crystal could be traced
back to [Cp*MoO2]2 after comparison with data from the
literature.[40] These experiments lend credit to the formation of
[Cp*MoO2]2 as the reduction product of [Cp*MoO2]2O under DODH
7
This article is protected by copyright. All rights reserved.

10.1002/cctc.202001115
ChemCatChem
FULL PAPER
In a next, stochiometric experiment [Cp*MoO2]2 was reacted with
1 equiv. of the more reactive diol substrate 1-phenyl-1,2-
H2O
R
HO
o
OH
ethanediol at 120 C to result in full conversion of 1-phenyl-1,2-
O
Mo
Cp*
Cp*
ethanediol, as well as formation of styrene (Scheme 5c). No
aldehydes were detected in this reaction. The outcome of this
stoichiometric reaction indicates that [Cp*MoO2]2 is able to
engage in a condensation reaction with a diol substrate and in a
subsequent olefin extrusion step to form an olefin at a relatively
low temperature. Yet, catalytic turnover starting from [Cp*MoO2]2
seems very limited (vide supra), possibly through the formation of
oxidized species other than [Cp*MoO2]2O that do not allow for the
regeneration of [Cp*MoO2]2 upon reduction. These considerations
also seem to rule out pathway A.
O
Mo
O
O
O
O
Cp*
Cp*
O
Cp*
Cp*
O
O
O
O
Mo
Mo
Mo
Mo
R
Red.
pathway C
O
O
O
O
Red.
Cp*
Cp*
O
R
ORed.
Mo
Mo
O
O
O
R
O
Scheme 6. Proposed catalytic pathway C for [Cp*MoO2]2O-catalyzed DODH
reactions involving [Cp*MoO2]2 is an off-cycle species. Red. = PPh3 or diol.
ORed. = OPPh3 or carbonyl compounds.
OH
a)
[Cp*MoO2]2 2 mol%
200 oC, 15 h, N2, anisole
+
PPh3
+
OPPh3
+
H2O
C6H13
OH
C6H13
yield: 8%
conversion: 77%
Conclusion
OH
OH
b)
c)
[Cp*MoO2]2 2 mol%
200 oC, 18.5 h, air, anisole
+
PPh3
+
OPPh3
+
H2O
C6H13
OH
OH
C6H13
Exploratory studies in this work have shown that the dinuclear
complex [Cp*MoO2]2O is able to catalyze the DODH of diols using
PPh3 as reductant and anisole as solvent. The reaction requires
a high reaction temperature (200 oC) to reach high substrate
conversion. Under these optimized reaction conditions, the
[Cp*MoO2]2O catalyst can be applied to the DODH of aliphatic
vicinal diols to form the corresponding olefins with moderate
yields up to 55% yield. Product yields and selectivities have also
always been an issue in Mo-catalyzed DODH reaction, with the
exception of using excess amounts of ligands or additives. Also
for our current catalyst system, the dehydration of diol substrate
and its oxidative cleavage seem to be part of the overall reactions
that take place. Yet, [Cp*MoO2]2O catalyst rivals other Mo-based
catalysts in its DODH performance, with mass balances (up to
79%) and TONs per Mo (up to 27) amongst the highest reported
for molecular Mo-based DODH catalysts. Interestingly, the rather
challenging substrate cis-1,2-cyclohexanediol can be converted
to 30% 1-cyclohexene by [Cp*MoO2]2O, thereby outperforming
the typical CptttReO3-catalyzed DODH reaction.
Mechanistic studies have detected [Cp*MoO2]2 as one of the
reduction products arising from [Cp*MoO2]2O. While, [Cp*MoO2]2
is able to engage in a condensation reaction with a diol substrate
and a subsequent olefin extrusion step to form an olefin, the
catalytic activity of [Cp*MoO2]2 is much lower than that of
[Cp*MoO2]2O. Along with the observation of zero-order reaction
kinetics and the formation of other reaction species when treating
[Cp*MoO2]2O with diols, we have proposed an reaction pathway
in which [Cp*MoO2]2 is an off-cycle species and where catalysis
proceeds via initial diol condensation to [Cp*MoO2]2O. Further
studies aim at identifying the diolate intermediate and at shedding
further light on the DODH reaction mechanism, to ultimately aid
further catalyst development.
yield: 30%
conversion: >99%
+
[Cp*MoO2]2
120 oC, o.n., N2, toluene-D8
1.0 eq.
1.0 eq.
full conversion
Scheme 5. Experiments probing the role of [Cp*MoO2]2 in DODH catalysis: a.
use of [Cp*MoO2]2 as catalyst under inert conditions; b. use of [Cp*MoO2]2 as
catalyst under aerobic conditions; c. stochiometric reaction of 1-phenyl-1,2-
ethanediol and [Cp*MoO2]2 under inert conditions.
Overall, our data do not provide strong evidence for either a
mononuclear catalytic mechanism for DODH reactions with
[Cp*MoO2]2O or for the dinuclear pathways A and B. The next
mechanism to consider would be a dinuclear mechanism in which
diol condensation precedes Mo=O reduction. In experiments to
show the formation of [Cp*MoO2]2 upon reduction of [Cp*MoO2]2O
with an excess of diol, additional Mo-based species were
observed by NMR analysis (vide supra). Although not conclusive,
we tentatively assign these species as diolates of either
[Cp*MoO2]2O or [Cp*MoO2]2. On the basis of this tentative
assignment, we propose the putative mechanism C shown in
Scheme 6 for DODH reactions catalyzed by [Cp*MoO2]2O. In this
mechanism, [Cp*MoO2]2O is either reduced to [Cp*MoO2]2 or
reacts with a diol to form a nonsymmetric diolate. As shown in our
mechanistic studies, [Cp*MoO2]2 cannot be considered as a
kinetically competent species involved in catalysis. Instead,
catalysis in pathway C evolves through the reduction of the
nonsymmetric diolate formed from [Cp*MoO2]2O, followed by
olefin extrusion to yield the react product and form [Cp*MoO2]2O
back to close the cycle. Accordingly, we propose that
[Cp*MoO2]2O is the actual DODH catalyst and that it operates via
a formal MoVIMoVI « MoVIMoIV redox cycle. Pathway C seems to
agree with our observation that catalysis proceeds without an
induction period. Further mechanistic studies will have to shed
further light on the viability of pathway C, e.g. with respect to the
identity of the diolate intermediate, the number of diols that can
condense to [Cp*MoO2]2O, and the energy profile of the individual
reaction steps and the overall catalytic reaction. Efforts to this
extent are ongoing in our labs.
Experimental Section
All chemicals including solvents were degassed by either freeze-pump-
thaw cycles or degasification under vacuum. Triphenylphosphine was
crystallized from ethanol and dried under vacuum. [Cp*MoO2]2O,[32,33]
(tacn)MoO3,[35] and NBu4[Cp*MoO3][34] was prepared according to a known
literature procedure. Unless otherwise stated, all other commercial
chemicals were used without further purification. NMR spectra were
8
This article is protected by copyright. All rights reserved.

10.1002/cctc.202001115
ChemCatChem
FULL PAPER
recorded on a Varian VNMRS400 (400 MHz) instrument at 298 K. ESI-MS
spectra were recorded using a Waters LCT Premier XE instrument. GC
measurements were performed using a Perkin Elmer Clarus 500 Gas
Chromatograph equipped with a Perkin Elmer Elite-5 column (Length: 30m,
I.D.: 0.32 mm, Film thickness: 0.25 μm, 5% phenyl-95%
methylpolysiloxane), and with FID detector. GC method: 40 °C, 5 min;
3 °C/min to 55 °C; 20 °C/min to 250 °C; 250 °C, 10 min; 20 °C/min to
300 °C; 300 °C, 5 min. All olefinic products are known compounds and
were calibrated against mesitylene for quantification.
[28]
[29]
R. Poli, Coord. Chem. Rev. 2008, 252, 1592-1612.
M. Herberhold, W. Kremnitz, A. Razavi, H. Schöllhorn, U. Thewalt,
Angew Chem. Int. Ed. Engl. 1985, 24, 601-611.
[30] P. Sözen-aktas, E. Manoury, F. Demirhan, R. Poli, Eur. J. Inorg. Chem.
2013, 15, 2728-2735.
[31] C. Bhaumik, E. Manoury, J. Daran, P. Sözen-aktas, F. Demirhan, R. Poli,
J. Organomet. Chem. 2014, 760, 115-123.
[32] M. Scheer, K. Schuster, K. Schenzel, E. Herrmann, Z. anorg. allg. Chem.
1991, 600, 109-119.
[33]
[34]
K. Umakoshi, K. Isobe, J. Organomet. Chem. 1990, 395, 47-53.
J. Sundermeyer, U. Radius, C. Burschka, Chem. Ber. 1992, 125, 2379-
2384.
Acknowledgements
[35]
P. C. McGowan, T. J. Podesta, M. Thornton-Pett, Inorg. Chem. 2001,
40, 1445-1453.
JL acknowledges financial support through the Chinese
Scholarship Council and Utrecht University. The X-ray
diffractometer has been financed by the Netherlands
Organization for Scientific Research (NWO).
[36]
[37]
[38]
[39]
T. Welton, R. Soc. Publ. 2015, 471, 1-26.
D. Prat, J. Hayler, A. Wells, Green Chem. 2014, 16, 4546-4551.
A. M. Khenkin, R. Neumann, Adv. Synth. Catal. 2002, 344, 1017-1021.
S. T. Pritam, S. Harbans, (Miss) Parveen. Bull. Chem. Soc. Jpn. 1983,
56, 1536-1538.
[40]
H. Arzoumanian, A. Baldy, M. Pierrot, J.-F. Petrignani, J. Organomet.
Chem. 1985, 294, 327-331.
Keywords: deoxydehydration • homogeneous catalysis •
molybdenum complexes • biomass conversion
[1]
S. Raju, M.-E. Moret, R. J. M. Klein Gebbink, ACS Catal. 2015, 5, 281-
300.
[2]
[3]
[4]
J. R. Dethlefsen, P. Fristrup, ChemSusChem 2015, 8, 767-775.
K. A. Denike, S. M. Kilyanek, R. Soc. open sci. 2019, 6, 1-16.
L. J. Donnelly, S. P. Thomas, J. B. Love, Chem. Asian J. 2019, 14, 3782-
3790.
[5]
S. C. C. van der Lubbe, C. Fonseca Guerra, Chem. - A Eur. J. 2017,
23, 10249-10253.
[6]
[7]
[8]
[9]
E. Steffensmeier, K. M. Nicholas, Chem. Commun. 2018, 54, 790-793.
G. Chapman, K. M. Nicholas, Chem. Commun. 2013, 49, 8199-8201.
T. V. Gopaladasu, K. M. Nicholas, ACS Catal. 2016, 6, 1901-1904.
Y. Y. Jiang, J. L. Jiang, Y. Fu, Organometallics 2016, 35, 3388-3396.
[10] A. Galindo, Inorg. Chem. 2016, 55, 2284-2289.
[11] L. C. De Vicente Poutás, M. Castiñeira Reis, R. Sanz, C. Sliva López, O.
Nieto Faza, Inorg. Chem. 2016, 55, 11372-11382.
[12] A. R. Petersen, L. B. Nielsen, J. R.Dethlefsen, P. Fristrup, ChemCatChem
2018, 10, 769-778.
[13] K. M. Kwok, C. K. S. Choong, D. S. W. Ong, J. C. Q. Ng, C. G. Gwie, L.
Chen, A. Borgna, ChemCatChem 2017, 9, 2443-2447.
[14] L. Hills, R. Moyano, F. Montilla, A. Pastor, A. Galindo, E. Álvarez, F.
Marchetti, C. Pettinari, Eur. J. Inorg. Chem. 2013, 19, 3352-3361.
[15]
J. R. Dethlefsen, D. Lupp, B.-C. Oh, P. Fristrup, ChemSusChem 2014,
7, 425-428.
[16] J. R. Dethlefsen, D. Lupp, A. Teshome, L. B. Nielsen, P. Fristrup, ACS
Catal. 2015, 5, 3638-3647.
[17] D. Lupp, N. J. Christensen, J. R. Dethlefsen, P. Fristrup, Chem. - A Eur.
J. 2015, 21, 3435-3442.
[18] K. Beckerle, A. Sauer, T. P. Spaniol, J. Okuda, Polyhedron 2016, 116,
105-110.
[19]
M. Stalpaert, D. De Vos, ACS Sustain. Chem. Eng. 2018, 6, 12197-
12204.
[20]
[21]
R. Tran, S. M. Kilyanek, Dalt. Trans. 2019, 48, 16304-16311.
C. A. Navarro, A. John, Inorg. Chem. Commun. 2019, 99, 145-148.
[22] S. C. A. Sousa, A. C. Fernandes, Coordination Chemistry Reviews, 2015,
284, 67-92.
[23]
N. García, R. Rubio-Presa, P. García-García, M. A. Fernández-
Rodríguez, M.R. Pedrosa, F. J. Arnáiz, R. Sanz, Green Chem. 2016, 18,
2335-2340.
[24]
[25]
A. R. Petersen, P. Fristrup, Chem. - A Eur. J. 2017, 23, 10235-10243.
S. Raju, J. T. B. H. Jastrzebski, M. Lutz, R. J. M. Klein Gebbink,
ChemSusChem 2013, 6, 1673-1680.
[26]
[27]
S. Raju, C. A. M. R. Van Slagmaat, J. Li, M. Lutz, J. T. B. H. Jastrzebski,
M.-E. Moret, R. J. M. Klein Gebbink, Organometallics 2016, 35, 2178-
2187.
J. Li, M. Lutz, M. Otte, R. J. M. Klein Gebbink, ChemCatChem 2018, 10,
4755-4760.
9
This article is protected by copyright. All rights reserved.

10.1002/cctc.202001115
ChemCatChem
FULL PAPER
Entry for the Table of Contents
OH
[Cp*MoO2]2O 2 ꢂꢃꢄꢅ
anisole, 200 oC, 15 h, N2
PPꢀꢁ
OPPꢀꢁ
+
H2O
+
+
R
OH
R
yield up to 56%
vicinal diols
[Cp*MoO2]2O is reported as catalyst for the deoxydehydration of (biomass-derived) vicinal diols, providing reasonable yields and
mass balances.
10
This article is protected by copyright. All rights reserved.