Synthesis of Cyclopentadienyl-Based Trioxo-rhenium Complexes and Their Use as Deoxydehydration Catalysts

Article abstract of DOI:10.1021/acs.organomet.6b00120

The cyclopentadienyl-based trioxo-rhenium complexes Cp^tttReO₃ (Cp^ttt = 1,2,4-tritert-butylcyclopentadienyl) 1b and Cp?ReO₃ (Cp? = pentamethylcyclopentadienyl) 4b) are known to be active catalysts for the deoxydehydration (DODH) of vicinal diols to olefins. Here, we report on the preparation of a series of complexes of the general formula Cp′ReO₃ (Cp′ = 1,3-di-tert-butylcyclopentadienyl (Cp^tt, 2b, 18%), 1,2,3-triisopropylcyclopentadienyl (3b, 4%), 1,2,3-trimethyl-4,5,6,7-tetrahydroindenyl (6b, 36%), and tetramethylcyclopentadienyl (7b, 33%)) in which the electronic and steric properties of the Cp′ ligand are varied. These complexes were synthesized via oxidative decarbonylation from the corresponding Cp′Re(CO)₃ complexes with either H₂O₂ or tBuOOH. An in situ NMR investigation revealed that complexes 2b, 3b, and 6b are unstable under the oxidizing reaction conditions. This information was used to determine the optimal reaction time to isolate the complexes 2b, 3b, and 6b. These piano-stool configurated Cp′ReO₃ complexes were characterized spectroscopically and by single crystal X-ray diffraction. The new complexes 2b, 3b, 6b, and 7b were found to be generally less thermally stable than Cp^tttReO₃ (1b) or Cp?ReO₃ (4b). It appeared that 2b and 6b were better catalysts for the DODH of 1,2-octanediol to octene with PPh₃ as an oxygen-atom acceptor. Remarkably, the less substituted Cp′ReO₃ complexes (1b, 2b, and 3b) afforded significantly less 2-octene (presumably from isomerization of the primary 1-octene product) in comparison to those containing per-alkylated Cp-moieties (4b and 6b).

Full text of DOI:10.1021/acs.organomet.6b00120

Article
pubs.acs.org/Organometallics
Synthesis of Cyclopentadienyl-Based Trioxo-rhenium Complexes and
Their Use as Deoxydehydration Catalysts
Suresh Raju,† Christian A. M. R. van Slagmaat,† Jing Li,† Martin Lutz,‡ Johann T. B. H. Jastrzebski,†
Marc-Etienne Moret,† and Robertus J. M. Klein Gebbink*,†
†Organic Chemistry and Catalysis, Debye Institute for Nanomaterials Science, Faculty of Science, Utrecht University,
Universiteitsweg 99, Utrecht 3584 CG, The Netherlands
‡Crystal and Structural Chemistry, Bijvoet Center for Biomolecular Research, Faculty of Science, Utrecht University, Padualaan 8,
Utrecht 3584 CH, The Netherlands
S
* Supporting Information
ABSTRACT: The cyclopentadienyl-based trioxo-rhenium complexes CptttReO3
(Cpttt = 1,2,4-tritert-butylcyclopentadienyl) 1b and Cp*ReO3 (Cp* = pentam-
ethylcyclopentadienyl) 4b) are known to be active catalysts for the deoxydehydration
(DODH) of vicinal diols to olefins. Here, we report on the preparation of a series of
complexes of the general formula Cp′ReO3 (Cp′ = 1,3-di-tert-butylcyclopentadienyl
(Cptt, 2b, 18%), 1,2,3-triisopropylcyclopentadienyl (3b, 4%), 1,2,3-trimethyl-4,5,6,7-
tetrahydroindenyl (6b, 36%), and tetramethylcyclopentadienyl (7b, 33%)) in which
the electronic and steric properties of the Cp′ ligand are varied. These complexes
were synthesized via oxidative decarbonylation from the corresponding Cp′Re(CO)3
complexes with either H2O2 or tBuOOH. An in situ NMR investigation revealed that
complexes 2b, 3b, and 6b are unstable under the oxidizing reaction conditions. This information was used to determine the
optimal reaction time to isolate the complexes 2b, 3b, and 6b. These “piano-stool” configurated Cp′ReO3 complexes were
characterized spectroscopically and by single crystal X-ray diffraction. The new complexes 2b, 3b, 6b, and 7b were found to be
generally less thermally stable than CptttReO3 (1b) or Cp*ReO3 (4b). It appeared that 2b and 6b were better catalysts for the
DODH of 1,2-octanediol to octene with PPh3 as an oxygen-atom acceptor. Remarkably, the less substituted Cp′ReO3 complexes
(1b, 2b, and 3b) afforded significantly less 2-octene (presumably from isomerization of the primary 1-octene product) in
comparison to those containing per-alkylated Cp-moieties (4b and 6b).
expensive stoichiometric reductant (PPh3), and rapid catalyst
decomposition to multinuclear clusteric oxo-rhenium spe-
cies.15,16 These issues were partly overcome by introducing a
bulky 1,2,4-tritert-butylcyclopentadienyl (Cpttt) ligand:
CptttReO3 displays better TONs (up to 1400) and is able to
perform DODH with other reductants such as sec. alcohols and
hydrogen.17a,b Hence, systematic variations of multiply
substituted Cp-based trioxo-rhenium catalysts are desirable
for further optimization.
INTRODUCTION
■
Sustainable industrial production of chemicals will benefit from
new synthetic routes starting from renewable feedstocks such as
biomass-derived polysaccharides and sugar and polyol deriva-
tives. Unlike the currently used fossil hydrocarbons, these new
starting materials are overfunctionalized with oxygen atoms.1−4
Hence, there is a renewed interest in catalytic deoxygenation
methods, and especially in deoxydehydration (DODH)
reactions that convert vicinal diols to olefins.5−7 The best
catalysts for these reactions are currently rhenium-based; in
particular, methyltrioxorhenium (MTO) has been shown to be
a versatile DODH catalyst in combination with a range of
external reductants.8−11 The simple methyl substitution that
keeps the high Lewis acidity of the rhenium center is thought to
facilitate the coordination of substrate molecules. However,
derivatives of MTO containing longer chain alkyl groups are
unstable even at low temperatures, preventing systematic
variations for further catalyst development.12−14
At the beginning of this work, only a limited number of other
Cp-based trioxo-rhenium complexes were known in the
literature: CpReO3,18−20 (CpMe)ReO3,19 and (CpEtMe4)-
ReO3.21 This is likely due to the synthetic difficulties
encountered when preparing these high-valent Re(VII) trioxo
complexes. Two methods were reported for the preparation of
Cp′ReO3 complexes depending on the redox potential of the
Cp′ ligands (Cp′ = substituted cyclopentadienyl; substituted
indenyls are also abbreviated as Cp′ for simplicity).22 For
electron-poor ligands, the method of choice is the trans-
However, the cyclopentadienyl-based trioxo-rhenium catalyst
Cp*ReO3 (Cp* = 1,2,3,4,5-pentamethylcyclopentadienyl) is
long known to catalyze DODH reactions of vicinal diols
including bioderived polyols.15 However, this system suffers
from some major drawbacks such as low (<55) catalytic
turnover number (TON), the requirement of a strong and
metalation reaction of XReO3 (X− = ReO4 , Cl−, and
−
CF3COO−) with Cp′2Zn or Bu3Cp′Sn to obtain Cp′ReO3
Received: February 12, 2016
© XXXX American Chemical Society
A
DOI: 10.1021/acs.organomet.6b00120
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
(Cp′ = Cp18−20 or CpMe19). In contrast, highly substituted,
electron-rich Cp′ReO3 (Cp′ = Cp*,23−25 CpEtMe4,21 or
Cpttt17) complexes can be obtained by oxidative decarbon-
ylation of Cp′Re(CO)3 with H2O2 as oxidant.26 In addition,
oxidative decarbonylation for the formation of Cp*ReO3 was
demonstrated with other oxidants such as O2,27,28 O3,29
Mn2O7,30 tBuOOH,31 and dimethyldioxirane.32 These elec-
tron-rich Cp′ReO3 complexes display improved thermal
stability compared to that of electron-poor Cp′ReO3
complexes. The decomposition temperature was determined
by thermogravimetric analysis: Cp, 134.2 °C; CpMe, 144.7 °C;
Cp*, 207.8 °C; and CpEtMe4, 166.0 °C.13,22,31,33
Scheme 1. Synthesis of Alkyl-Substituted Cp′ReO3
Complexes by Oxidative Decarbonylation of Cp′Re(CO)3
Complexes by Using H2O2 Solution
Here, we report the preparation of a series of alkyl-
substituted Cp′ReO3 complexes (1b−7b, Chart 1) by oxidative
Chart 1. Cp′ReO3 Complexes 1b−7b (Including Isolated
Yields)
evaporation in vacuo afforded bright yellow crystals of 1b, 4b,
and 5b.
In contrast, when 2a, 6a, (3a + 3a′), and 7a were subjected
to these reaction conditions, a yellow-brown oily mixture was
obtained. According to thin layer chromatography (TLC) and
NMR, it was found not to contain the starting Cp′Re(CO)3
complexes (2a, 6a, (3a + 3a′), and 7a) or the expected
Cp′ReO3 complexes (1b, 6b, (3b + 3b′), and 7b). The
synthesis attempts for the phenyl-substituted complexes
Cp′ReO3 (Cp′ = CpPh5, CpPh4H, IndPh3) using H2O2
biphasic conditions were also not successful. The starting
complex 7a was further tested under varying biphasic reaction
conditions. Excluding catalytic H2SO4 under prolonged reflux
reactions also resulted in an intractable mixture of products.
However, the appearance of the yellow color typical of Cp-
ligated trioxo-rhenium complexes at early reaction times
prompted us to stop the reaction after a short reaction time
(1 h), which allowed for the isolation of 7b as a yellow solid in
33% (Scheme 1).
Attempts to synthesize 2b under similar biphasic conditions
with H2O2 as the oxidant were unsuccessful. Thus, we turned to
the methodology developed by Kochi for Cp*ReO3 (31−36%
maximum yield), which involves homogeneous oxidation of
Cp*Re(CO)3 by tBuOOH (purified by distillation) over short
reaction times (5−10 min) at 80 °C.31 In our case, a
commercially available tBuOOH solution (6 M in decane)
was used instead. After the addition of tBuOOH, a solution of
2a in benzene remained colorless for 45 min at room
temperature. It was then placed in an oil bath at 85 °C,
resulting in a color change to yellow over 10−15 min followed
by dark-brown after 20 min, at which point the reaction was
stopped by cooling the mixture in an ice bath. The dark-brown
oily mixture obtained after solvent evaporation did not contain
the desired trioxo-rhenium species. However, the initial
appearance of a light yellow color suggested that the desired
product might be formed transiently.
decarbonylation of the corresponding tricarbonyl complexes
(1a−7a). tBuOOH is identified as the oxidant of choice for
sensitive complexes that decompose under oxidative conditions,
and we report an NMR method to determine the optimal
conditions and reaction time in these cases. Under the same
conditions, the desired Cp′ReO3 complexes were not observed
for aryl-substituted Cp′H ligands CpPh5H (1,2,3,4,5-pentaphe-
nylcyclopentadiene), CpPh4H2 (1,2,3,4-tetraphenylcyclopenta-
diene), or IndPh3H (1,2,3-triphenylindene). The catalytic
activity of complexes 1b−7b in the DODH of 1,2-octanediol
to 1-octene with PPh3 as reductant is compared to investigate
structure−activity relationships.
RESULTS AND DISCUSSION
■
Synthesis of Trioxo-rhenium Complexes. The alkyl-
substituted Cp′Re(CO)3 complexes 1a−7a (3a and 3a′ is 1,2,3-
and 1,2,4-Cp-substituted (CpiPr3H2)Re(CO)3, respectively)
can be prepared under neat reaction conditions by reacting
the corresponding Cp′H ligands with Re2(CO)10, while aryl-
substituted complexes are obtained by the reaction of LiCp′
(Cp′ = CpPh5, CpPh4H, IndPh3) with BrRe(CO)5.17c The
known rhenium trioxo complexes 1b, 4b, and 5b were
synthesized by decarbonylative oxidation of the corresponding
rhenium tricarbonyl complexes (1a, 4a, and 5a) according to
literature procedures (Scheme 1).17,23 A biphasic oxidation
reaction was performed by using an excess (20−30 equiv) of
H2O2 (35% in water) with a catalytic amount of conc. H2SO4 in
benzene at 80 °C for 15 h. After the reaction, the mixture was
extracted with benzene, and the organic phase was washed with
water, 5% aqueous NaHCO3, and brine. Then, solvent
The homogeneous character of the latter reactions rendered
them amenable to in situ monitoring by NMR, which allowed
one (1) to determine whether the desired products are formed
at all and (2) to determine optimal conditions (reaction time
and temperature) for a series of substituted Cp-ligands. Thus,
complexes 1a−7a were subjected to the following experiment:
To a solution of the Cp′Re(CO)3 complex in C6D6 (ca. 40
mM) a stoichiometric amount of tBuOOH (6 equiv) was
added. There was no visible reaction at RT, and the reaction
mixtures were heated gradually from RT inside the NMR
B
DOI: 10.1021/acs.organomet.6b00120
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
spectrometer until the conversion of tricarbonyl rhenium
complex started (Scheme 2). The reaction was monitored at
that temperature until no more conversion of the starting
Cp′Re(CO)3 complex was observed.
Scheme 2. Oxidative Decarbonylation of Cp′Re(CO)3
Complexes to Cp′ReO3 Complexes Using tBuOOH
Figure 2. Formation of Cp′ReO3 under oxidation conditions at 50 °C
in C6D6. CptttReO3 (1b) formation was followed at 70 °C, and it
reached a constant value (ca. 28%) in 4 h.
In most cases, the formation of Cp′ReO3 started at 50 °C, as
evidenced by the appearance of new Cp−H signals at lower
field than those of the corresponding Cp′Re(CO)3 complex
with some exceptions: the formation of 1b occurred only at 70
°C, and the 1,2,4-tri-isopropylcyclopentadienyl trioxo-rhenium
complex (vide infra) was not detected at all. Similar
experiments were also conducted with aryl-substituted 1,2,3,4-
tetraphenylcyclopentadienyl tricarbonylrhenium, (CpPh4H)Re-
(CO)3, in benzene at 50−80 °C or with excess of tBuOOH (12
equiv), but no conversion was observed. Increasing the reaction
temperature up to 100 °C in toluene-d8 also did not afford the
desired (CpPh4H)ReO3, albeit a small amount of (CpPh4H)-
Re(CO)3 was converted (ca. 20−30%, not quantified exactly).
The obtained reaction profiles of the formation of Cp′ReO3
complexes (1b−2b and 4b−7b) from the corresponding
Cp′Re(CO)3 complexes are depicted in Figures 1−3. There
Figure 3. Sum of unidentified rhenium components (100 −
(Cp′Re(CO)3 + Cp′ReO3) %) under in situ 1H NMR oxidation
reaction conditions at 50 °C in C6D6. The reaction starting from
CptttRe(CO)3 (1a) was followed at 70 °C.
10 ppm are found in the spectra, and these were tentatively
assigned to oxidation products originating from the Cp′-ligand.
Remarkably, the previously reported trioxo-rhenium com-
plexes 1b, 4b, and 5b were observed to be stable under these
oxidation reaction conditions: their concentration reaches a
plateau and does not decay under prolonged heating. Moderate
product yields of 52% and 50% were obtained (by NMR
experiments) for 4b and 5b corresponding to the conversion of
4a and 5a in 74% and 80%, respectively (Figures 1 and 2). The
formation of 1b was achieved in a maximum of 28% yield out of
57% conversion at 70 °C in 4 h. In these three cases, the
missing rhenium components were close to only 30%.
In contrast, the yield of complexes 2b, 6b, and 7b reached a
maximum yield of 28, 30, and 53%, respectively, followed by a
relatively slow decay under prolonged heating (Figure 2). In
the meantime, the conversion of Cp′Re(CO)3 complexes
reached its final value after 3 h. For these complexes, the
summed missing unidentified rhenium components account for
>60% (Figure 3).
Figure 1. Depletion of Cp′Re(CO)3 under oxidation conditions at 50
°C in C6D6. The CptttRe(CO)3 (1a) reaction was followed at 70 °C,
and its conversion reached a constant value (ca. 57%) in 4 h.
The inseparable mixture of isomers of (CpiPr3H2)Re(CO)3
(3a′/3a ratio = 73:27%) was oxidized under the same reaction
conditions. The reaction profile over time showed that
conversion of both isomers occurred simultaneously, but only
one of the corresponding trioxo-rhenium complexes (3b, 10%)
was observed. After 85 min, only 3a′ remained in the mixture
(3a′/3a = 30:0%, based on the initial amount), and a trace
amount of 3b (1.5%) was also observed (Figure 4). To confirm
the origin of 3b, an enriched sample (3a′/3a = 95:5%) was
obtained by column chromatography after heating (40 min) the
mixture of 3a′/3a = 73:27% under the oxidation conditions.
was an induction period for all the complexes, except for the
formation of 6b. After this time, a sudden increase of the
reaction rate was observed to reach nearly the optimal yield for
Cp′ReO3 in a very short time (see Figure 2). This varying
induction period (10−45 min) under exactly the same reaction
conditions is somewhat surprising and is probably indicative of
a radical chain reaction starting from tBuO· radicals. In all cases,
there is a substantial amount of unidentified missing rhenium
components, suggesting parallel decomposition pathways.
1
Many different additional H NMR signals ranging from 1 to
C
DOI: 10.1021/acs.organomet.6b00120
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
preheated oil bath at 50 °C, and the optimal reaction time as
determined by NMR was applied (see Experimental Section).
After the reaction, the mixture was cooled in an ice bath
followed by washings with water. This allows the removal of the
main decomposition product HReO4, evidenced by an acidic
(pH ∼ 1) aqueous phase, and any water-soluble oxidized
organics. Furthermore, either repeated crystallization by adding
hexane or pentane, or column chromatography followed by
crystallization afforded pure Cp′ReO3 (2b, 3b, and 6b) as
yellow crystals. The overall isolated yields are 2b (18%), 3b
(4%), and 6b (36%) (see Experimental Section for more
details).
Figure 4. Oxidation of the isomer mixture of (CpiPr3H2)Re(CO)3
(3a′/3a ratio = 73:27%) at 50 °C in C6D6. The filled data point
indicates the reaction corresponding to the 3a′/3a ratio = 95:5%. Only
the conversion of 3a occurred, and no oxo-rhenium product was
observed.
All newly synthesized Cp′ReO3 complexes (2b, 3b, 6b, and
1
7b) were characterized by NMR spectroscopy. All H NMR
signals, particularly the distinct Cp−H resonances, are shifted
to a low field region compared to the signals in the
corresponding Cp′Re(CO)3 complexes. Furthermore, ESI-MS
analysis showed protonated molecular species either with or
without the coordination of additives like pyridine.
Subjecting the new sample (3a′/3a = 95:5%) to the same
oxidation conditions showed a large consumption of 3a′ in 90
min (3a′/3a = 28:0%) but no appreciable formation of any
trioxo-rhenium complex. This indicates that the trioxo complex
3b originates from the minor isomer 3a. As 3b has been
identified crystallographically as containing the 1,2,3-tri-
isopropylcyclopentadienyl ligand (vide infra), we infer that
the minor isomer 3a contains the same ligand.
The oxidation reaction profiles indicate that per-alkylated
(4b and 5b) or very bulky (1b) Cp′ReO3 complexes are less
susceptible to decomposition under oxidative conditions, in
accord with their convenient synthesis in a biphasic medium in
good yields. Conversely, the less-substituted Cp-analogues (1b
and 7b) were prone to decomposition and pass a maximum
yield upon prolonged heating. Although the tetrahydroindenyl
complex 6b contains a per-alkylated Cp moiety, it still
decomposed under oxidative conditions, which may be due
to weaker C−H bonds in the fused cyclohexyl ring.
The reaction mechanism of the oxidative decarbonylation is
not clearly understood. Kochi and co-workers proposed that an
initial oxidative elimination of CO might lead to an unsaturated
Cp*Re(CO)2 species, followed by quick CO exchange with oxo
ligands.31,32 However, Herrmann and co-workers isolated some
bimetallic species containing oxo-bridges and carbonyl ligands
from oxidation reactions of Cp*Re(CO)3,23,34 which indicates a
more complex reactivity behavior.
The solid-state FT-IR spectra of the Cp′ReO3 complexes
display the typical Re−O bands in the range of 800−950 cm−1
and no ReC−O (1880−2010 cm−1) bands. The antisym-
metrical and symmetrical Re−O stretching vibrations of the
(ReO3) moiety in 1b−7b are typically observed in the range of
869−884 and 907−917 cm−1, respectively.39 The (Re−O)asym
band of the less-substituted 2b (882 cm−1) and 3b (884 cm−1)
are found at slightly higher energy than that of 1b (877 cm−1);
this band for these three complexes appear at higher energy
than that for the per-alkylated complexes 4b−6b (869−875
cm−1) (Table 1). This suggests that the ReO multiple bond
is weakened by strongly donating Cp-ligands. The complex 7b
lies in between these two types: an identical (Re−O)asym
vibration at 877 cm−1 for both 1b and 7b indicates that they
are comparable in terms of electronics. However, the lower
symmetry experienced by the bulkier 1b is reflected by the
higher-energy (Re−O)sym vibrations at 917 cm−1 compared to
that of the highly symmetrical 7b ((Re−O)sym = 910 cm−1).
The spectral differences between the 5b21 and 4b35 have been
interpreted in a similar way by lowering symmetry.
The X-ray crystal structures of 2b, 3b, 6b, and 7b all display a
three-legged “piano-stool” geometry similar to that of the
literature reported complexes 1b, 4b, and 5b (Figure 5). All of
the Re−O bond distances are in the range of 1.7209(15)−
1.735(4) Å for 1b−7b, except for the slight deviations
previously reported for 5b (1.664(4)−1.716(4) Å). Also, the
bond angles in 4b (O−Re−O = 103.7(6)−110(1)°) are larger
On the basis of the above in situ results, the alkyl-substituted
Cp′ReO3 1b, 3b, and 6b were isolated on a preparative scale.
These reactions were performed under N2 atmosphere using a
Table 1. Selected Analytical Data of Cp′Re(CO)3 Complexes of 1a−7a
X-ray crystal structure (Å/deg)
a
e
f
Cp′ReO3
IR (cm−1
)
Re−O
Cp−Re
2.0830(8)
Δ
Re−O
O−Re−O
b
1b
877, 917
882, 916
884, 919
869, 907
875, 915
875, 909
877, 910
0.032
0.040
0.059
1.7239(15)−1.7267(14)
1.725(5)−1.735(4)
1.7209(15)−1.7281(14)
1.73(1)−1.744(1)
1.664(4)−1.716(4)
1.724(4)−1.733(2)
1.729(3)−1.733(2)
103.86(8)−105.39(8)
103.82(19)−105.1(3)
104.41(7)−105.82(9)
103.7(6)−110(1)
104.7(2)−106.7(2)
105.35(17)−105.89(15)
105.01(16)−105.74(11)
2b
3b
2.068(3)
2.0637(8)
c
4b
d
2.023(4) (2.032(4))
2.076(4)
0.028 (0.046)
0.051
5b
6b
7b
2.062(4)
0.032
2.0628(15)
0.054
a
b
c
IR data recorded by directly measuring the solid samples (no KBr or solvent). Ref 17. Ref 35. The earlier reported highly symmetrical complex 4b
(Cp*ReO3) exhibited both crystal twinning and disorder,35 precluding accurate metric determination Ref 21. Distance between the ring centroid
and the metal. Ring slippage Δ is defined as the distance between the ring centroid and the perpendicular projection of the metal on the least-
squares plane of the ring.
d
e
f
D
DOI: 10.1021/acs.organomet.6b00120
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
The chemical nature of the decomposition products is not
generally known. Some insight was gained from the
decomposition of 2b: a concentrated solution of 2b (20 mg,
48 mM) was monitored in CDCl3 at RT under air. The color
changed from yellow to dark-brown to purple-red and a
precipitate appeared over the course of 1 day. The insoluble
green solid was removed from the purple-red solution by
filtration. After solvent evaporation in vacuo, a purple-red solid
(2c) was obtained (12.2 mg), which was crystallized in CHCl3/
ether (2:1). Spectroscopic data for 2c are consistent with the
tetranuclear structure depicted in Figure 6, which is analogous
to the previously reported, brown Cp*2Re2O3(ReO4)2.38
Figure 5. Molecular structures of 2b (top left), 3b (top right), 6b
(bottom left), and 7b (bottom right) in the crystal, drawn at the 50%
probability level. Hydrogen atoms are omitted for clarity. Complexes
2b and 7b are located on a crystallographic mirror plane.
than that in the other Cp′ReO3 complexes (O−Re−O =
103.8(2)−105.0(2)°.
Figure 6. Proposed structure of 2c formed by the decomposition of
2b..
The Cp-centroid distance to rhenium (Cp−Re) in all of the
Cp′ReO3 complexes (2b−7b) was found in the range of
2.062(4)−2.0830(8) Å, which is significantly longer than the
corresponding Cp′Re(CO)3 analogues (1.9538(6)−1.959(8)
Å). This is due to the replacement of π-electron accepting CO
ligands by π-electron donating oxo ligands.
1
The H NMR spectrum of 2c in CDCl3 displays two sets of
signals belonging to two distinct Cptt-ligands in a 1:1 ratio.
Further, the Re−O vibrations appear at 702, 731, 811, 874, 901,
915, and 931 cm−1 in the IR spectrum, similar to the vibrations
shown by Cp*2Re2O3(ReO4)2 at 688, 700, 729, 827, 855, 899,
934, and 973 cm−1. An ESI-MS analysis of 2c showed signals
corresponding to a cationic species formed by loss of a neutral
ReO3 fragment from 2c (M − ReO3)+ at m/z = 1025.1582
(calcd. m/z = 1025.1577). However, the elemental analysis data
of a sample of 2c did not match with the proposed structure,
which is likely due to organic impurities resulting in higher C
and H percentages. The proposed structure contains two
formally reduced Re(V) centers, suggesting that its formation
would require a reductant; interestingly, the Cp* analogue
Cp*2Re2O3(ReO4)2 was indeed formed in the presence of
PPh3. The fact that 2c is formed from CpttReO3 (2b) in the
absence of an external reductant suggests that it may originate
from a disproportionation reaction in which the Cptt ligand is
oxidized and released, in agreement with the presence of
“naked” trioxo-rhenium units in the structure. Moreover,
tetranuclear 2c can also be used as a catalyst in DODH
reactions. At 2 mol % loading, 2c selectively forms 1-octene
(89%) out of 90% conversion of 1,2-octanediol at 180 °C for
15 h. This observation could make higher nuclear Re complexes
of interest in further DODH catalyst development and in
mechanistic studies.
In all cases, the η5-Cp bonded Cp-ring is perfectly planar
with a very small ring-slippage of Δ = 0.032−0.059 Å (Δ is the
distance between the ring centroid and the perpendicular
projection of the metal on the least-squares plane of the ring).36
This commonly occurring slight ring-slippage in Cp′ReO3 has
been attributed to the trans-influence of the π-electron donating
oxo ligands.26,37 The molecules of 2b and 7b are located on
exact, crystallographic mirror planes, which pass through a C−
Re−O plane. In 3b, the Re−C(CpiPr) bond lengths
(2.4039(17)−2.4244(17) Å) of the substituted Cp-carbons
are significantly longer than those of the unsubstituted Cp-
carbons (Re−C(CpH): 2.3592(17)−2.3679(18) Å). This
might be due to the highly asymmetric steric bulk. In the
case of 6b, the typical puckered conformation of the fused
cyclohexyl group is observed alongside torsion angles of
−16.7(5) and −13.6(7) Å for the flipped methylene units
(Figure 5).
In general, the spectroscopic and structural properties of the
half-sandwiched “piano-stool” complexes are comparable to
previously known Cp′ReO3 complexes (1b and 4b−5b). In
contrast, the stability of Cp′ReO3 complexes appears to be
sensitive to the subtle changes in the Cp-substituents, which is
investigated further in the next section.
Stability of Trioxo-rhenium Species. A striking property
of the less substituted Cp trioxo-rhenium complexes (2b, 3b,
6b, and 7b) is their relative instability in comparison to the Cp*
or Cpttt analogues under oxidation reaction conditions. To
assess whether this arises from intrinsic thermal lability, the
thermal stability of trioxo-rhenium compounds was examined
in some detail. First, all pure Cp′ReO3 complexes (2b, 3b, 6b,
and 7b) could be stored as yellow solids below 0 °C without
decomposition for months even under air atmosphere.
However, at room temperature (both under air and N2
atmosphere), these solids gradually turned first to dark-brown
then to dark-green oily mixtures in 1−2 days.
The thermal stability of complexes 2b, 3b, and 6b was also
studied in solution. Solutions of each Cp′ReO3 complex in
C6D6 (ca. 40 mM) were heated to 50 °C, and the
1
decomposition reactions were monitored by H NMR. The
complexes 2b and 3b started to decompose within 15−30 min,
but per-alkylated 6b was stable in C6D6 even at 70 °C (30 min).
Hence, complexes 2b and 3b are intrinsically unstable under
the conditions of their preparation, which can be tentatively
ascribed to the reduced steric protection of the Re center in
comparison with the Cpttt (1b) and Cp* (4b) analogues. In
contrast, per-alkylated 6b is thermally stable under the same
conditions, suggesting that its decomposition arises from
E
DOI: 10.1021/acs.organomet.6b00120
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
reaction with the oxidant. Indeed, rapid decomposition of 6b
was observed at 50 °C in the presence of tBuOOH. This is
likely due to the presence of weaker (secondary) C−H bonds
in 6b than in the Cp* analogue 4b, which would be more
susceptible to H atom abstraction by radicals derived from
tBuOOH during reaction.
Reaction profiles were recorded at 135 °C for the best
performing catalysts 2b and 6b and for the benchmark catalysts
1b and 4b (Figure 7). In general, an apparent zero-order
Catalytic Studies. All Cp′ReO3 complexes (1b−7b) were
found to be catalytically active in the DODH of 1,2-octanediol
to 1-octene. PPh3 (1.1 equiv) was used as the stoichiometric
reductant in chlorobenzene at 180 °C or 135 °C to completely
convert 1,2-octanediol in 12−15 h (Scheme 3).17 Complexes
Scheme 3. Cp′ReO3-Catalyzed DODH of 1,2,-Octanediol to
1-Octene in Chlorobenzene
Figure 7. DODH reaction profile for 1,2-octanediol conversion to 1-
octene at 135 °C. In the case of CptttReO3, 23% of 1,2-octanediol
remained after 10 h. For 4b, 15% trans-2-octene was formed as a side-
product after 10 h, which is denoted as 4b (t2O) in the figure.
1b−3b showed high selectivity to 1-octene (94−96%) at 135
°C, whereas a small amount of the isomerization product 2-
octene (5−6%, cis-2-octene/trans-2-octene ≈ 1:1) was
observed at 180 °C (Table 2). Notably, the per-alkylated
reaction was observed consistent with the rate-limiting
extrusion of olefin from the Re-diolate.40 An induction period
was observed for 1b and 4b, suggesting that the trioxo-rhenium
complex might act as a precatalyst and not part of the real
DODH cycle. Abu-Omar and co-workers recently reported on
similar observations, i.e., a zero-order rate and an induction
period in MTO-catalyzed DODH reactions.10 It may thus be
that related species are involved in these reactions.
Table 2. Cp′ReO3-Catalyzed DODH of 1,2,-Octanediol in
a
Chlorobenzene
b
b
yield (%) at 180 °C
yield (%) at 135 °C
1-
C-2-
oct
T-2-
oct
1-
C-2-
oct
T-2-
oct
catalyst oct
t (h) oct
t (h)
Except for 4b (and 5b), which produced a significant amount
of 2-octene product, the other catalysts were highly selective to
yield 1-octene at 135 °C. The formation of trans-2-octene for
4b mainly took place at the start of the reaction, which is
coupled with an apparent induction period. This indicates an
initially formed rhenium species would facilitate this reaction
pathway. The amount of trans-2-octene remained the same
after some time, while 1-octene formation gradually increased
between 2−4 h (Figure 7). 1b only gave 1-octene in 74% after
10 h with traces of 2-octene isomers out of 77% diol
conversion. Interestingly, the catalysts 2b and 6b were found
to act as much faster DODH catalysts to obtain >96% of 1-
octene within 6 and 8 h, respectively. No observable 2-octene
isomers were formed in these reactions (Table 2).
c
1b
2b
3b
4b
5b
6b
7b
89
91
95
66
61
66
70
3
3
12
15
15
15
12
15
12
94
96
94
84
72
96
96
1
0
0
1
2
0
0
1
15
15
15
10
15
15
15
2
3
0
2
2
0
14
9
15
10
12
4
14
19
0
10
2
0
a
Reaction conditions: 1,2-octanediol (0.5 mmol), Cp′ReO3 catalyst
(0.01 mmol, 2 mol % loading), mesitylene (0.5 mmol, internal
standard), PPh3 (0.55 mmol), PhCl (5 mL, 100 mM based on diol),
and N2 atmosphere. Determined by GC. Complete conversion to 1,2-
octanediol. 1-oct/C-2-oct/T-2-oct = 1-octene/cis-2-octene/trans-2-
b
c
octene. Literature reported data in ref 20.
The rather similar reaction rates (after the induction period)
could be interpreted in terms of a common catalyst being
involved for the cases shown in Figure 7. The instability of the
isolated high-valent Re(VII) Cp′ReO3 complexes (2b, 3b, 6b,
and 7b) at high temperatures may indeed point to the
formation of such a common catalyst. However, the differences
in product selectivity between the different catalysts strongly
suggests that the Cp-ligands are still bound in the active species,
i.e., the Cp′ReO3 species do not simply decompose to a
common catalyst. This apparent discrepancy can be resolved by
considering the fact that both the reduction of the Cp′ReO3
species to Cp′ReO2 and the subsequent condensation reaction
with the diol substrate are rapid at room temperature
already.17b Therefore, Cp′ReO3 cannot be the resting state in
the DODH reaction, which takes several hours at 135 °C.
Under the catalytic reaction conditions with excess reductant, it
therefore is most plausible that the starting Cp′ReO3 species
are transformed into either a mononuclear, Cp′-ligated resting
state species, with the corresponding mono-oxo Re(V)diolate
being a likely candidate, or into (Cp′-ligated) reduced or mixed
complexes 4b−6b afforded a significant amount of 2-octene
products (19−29%, cis-2-octene/trans-2-octene ≈ 1) at 180 °C
and at 61−66% 1-octene formation. At 135 °C, 6b and 7b very
selectively gave 1-octene (96%), but 4b and 5b produced
substantial amounts of trans-2-octene (14−19%) together with
1-octene (72−84%).
Therefore, a high reaction temperature promotes the
formation of 2-octene in all cases as reported previously for
1b, but the reason why per-alkylated (i.e., highly electron
donating) Cp-substituted catalysts (4b−6b) significantly
facilitate the formation of the formal isomerization product is
unclear. Control experiments carried out at 180 °C for 15 h
starting from 1-octene in chlorobenzene in the presence of 4b
(2 mol %) did not show any isomerization product. However,
under catalytic conditions, i.e., including 1.1 equiv of PPh3, 22%
of 1-octene was isomerized to a mixture of cis-2-octene (12%)
and trans-2-octene (10%) at 180 °C in 15 h. Therefore, a
(reduced) rhenium species formed during the course of the
DODH reaction is likely to be responsible for the resulting 2-
octene products.
F
DOI: 10.1021/acs.organomet.6b00120
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
and a PerkinElmer Clarus SQ 8 T Mass Spectrometer. GC
measurements were performed by using a PerkinElmer Autosystem
XL gas chromatograph equipped with a PerkinElmer Elite-17 column
(length = 30 m, internal diameter = 0.32 mm, and film thickness =
0.50 mm), and with a flame ionization detector. GC method: 40 °C, 5
min; 3 °C min−1 to 55 °C; 20 °C min−1 to 250 °C; and 250 °C, 10
min.
valent multinuclear Re-oxo or Re-diolate species. Further
experimental inquiries are required to delineate these
speculative DODH reaction intermediates.
CONCLUSIONS
■
Multiply substituted Cp-based trioxo-rhenium complexes
(Cp′ReO3) were synthesized by oxidative decarbonylation of
the corresponding Cp-based tricarbonyl rhenium complexes,
either by using aqueous H2O2 or tBuOOH as the oxidant. In
situ NMR experiments for the reaction with tBuOOH were
carried out at 50 °C to observe a maximum build up (up to
55%) followed by decomposition of the Cp′ReO3 complexes at
an extended reaction time. The per-alkylated Cp*ReO3 and
(CpEtMe4)ReO3 complexes were exceptionally stable under
these conditions showing no decomposition. Using the
optimized oxidation reaction conditions (ca. 30 min), the
synthesis of (CptBu2H3)ReO3 (2b, 18%), (CpiPr3H2)ReO3
(3b, 4%), (1,2,3-Me3(tetrahydroindenyl))ReO3 (6b, 36%),
and (CpMe4H)ReO3 (7b, 33%) was achieved. Spectroscopic
characterization, together with X-ray crystal structure determi-
nation, showed the typical “piano-stool” half-sandwich structure
of these complexes. Preliminary studies showed that the pure
Cp′ReO3 complexes 2b, 3b, 6b, and 7b are thermally labile and
are prone to decomposition to multinuclear oxo-rhenium
clusters.
Synthesis of CpttReO3 2b. CpttRe(CO)3 (1.556 g, 3.476 mmol) was
placed in a dried Schlenk flask and degassed in vacuo for 30 min. Then
it was dissolved in benzene (65 mL), followed by the addition of
tBuOOH (6.0 M in decane, 2.9 mL, 6 equiv) under a nitrogen
atmosphere, and vigorous stirring. The reaction mixture was immersed
into a preheated oil bath at 50 °C and allowed to react for 25 min,
during which a color change from colorless to yellow-orange and the
formation of gas (carbon monoxide) were observed. Subsequently, the
reaction mixture was immediately cooled in an ice bath. A sample of
the crude reaction mixture was analyzed by TLC and 1H NMR, which
showed the formation of the desired product and approximately a 1:1
ratio of the carbonyl- and oxo-complexes. In open air, the reaction
mixture was successively washed with water (2 × 30 mL) and brine
solution (1 × 30 mL), dried over MgSO4, and concentrated in vacuo
(10 mL). Upon addition of cold hexane (40 mL) the product was
precipitated and isolated by centrifugation and decantation of the
hexane. To avoid a highly concentrated solution of 2b, it was always
left over with ca. 10 mL of solution to which cold hexane added to
precipitate the product, and this procedure was repeated 2 more times
to collect most of the product. A bright yellow crystalline solid was
obtained by combination of the product fractions (262 mg, 0.637
mmol, 18%). Crystals suitable for X-ray diffraction were grown from a
CH2Cl2/hexane 2:1 mixture at −30 °C; the unreacted CpttRe(CO)3
(272 mg, 0.608 mmol, 18%) and the brown oily mixture (662 mg,
unidentified mixtures). In addition, the water used for the workup was
shown to be strongly acidic (pH ≤ 1), which indicates the presence of
perrhenic acid.
All complexes were found active in catalytic deoxydehydra-
tion (DODH) of 1,2-octanediol to provide very good to
excellent octene yields. Surprisingly, a slight modification (6b
and 7b) in the electron donating property of Cp* or the steric
bulkiness (2b and 3b) of Cpttt (1,2,4-tritert-butylcyclopenta-
dienyl) ligands decreased thermal stability and resulted in
improved catalytic properties in comparison with those of the
previously reported Cp*ReO3 complex, leading to catalytic
properties similar to those of the CptttReO3 complex. Per-
alkylated Cp-ligated catalysts (4b and 6b) afford significantly
more 2-octene isomers than their less substituted counterparts
at a higher reaction temperature. At a lower temperature, 1b−
3b, as well as 6b, provided excellent 1-octene yields without the
formation of 2-octene. Following the catalytic reactions in time
has provided further insight, including the presence of
induction times for some and zero-order kinetics for other
catalysts. These experiments also suggest the involvement of
Cp′-ligated Re-species as the resting state and active species
during catalysis. Accordingly, ongoing investigations in our
laboratory target the development of new Cp-based trioxo-
rhenium catalysts, the application of these Cp′ReO3 catalysts
for the conversion of biomass-derived polyols, as well as a
detailed understanding of the mechanism by which these
catalysts operate in DODH reactions.
1H NMR (400 MHz), 25 °C, CDCl3 (7.26 ppm): 1.43 (s, 18H,
tBu), 6.48 (t, 1H, Cp−H, 4JH,H = 2.4 Hz), 6.54 (d, 2H, Cp−H, 4JH,H
=
2.4 Hz) ppm. 1H NMR (400 MHz), 25 °C, C6D6 (7.16 ppm): 1.16 (s,
4
18H, tBu), 5.89 (d, 1H, Cp−H, JH,H = 2.8 Hz), 6.18 (t, 2H, Cp−H,
4JH,H = 2.8 Hz) ppm. 13C NMR (400 MHz), 25 °C, CDCl3 (77.16
ppm): 30.05, 33.53, 106.71, 107.72, 140.65 ppm. 13C NMR (400
MHz), 25 °C, C6D6 (127.06 ppm): 29.71, 33.12, 106.76, 107.04,
139.76 ppm. FT-IR: 661, 674, 851, 882, 898, 916, 1030, 1063, 1174,
1206, 1252, 1368, 1397, 1455, 1471, 1489, 2872, 2908, 2965, 3084,
3127 cm−1. ESI-MS (in CH2Cl2 with pyridine): m/z = 492.1554
{[CpttReO3 + py + H]+, calcd. 492.1549}. Elemental analysis calcd
(%) for C13H21ReO3 (412.10): C 37.94, H 5.14; found, C 38.42, H
5.18.
Synthesis of (CpiPr3H2)ReO3 3b. The liquid isomer mixture of
(CpiPr3H2)Re(CO)3 (1.682 g, 3.644 mmol) was introduced into a
dried Schlenk flask and degassed in vacuo for 30 min. Then, it was
dissolved in benzene (80 mL), followed by the addition of 6 equiv of
tBuOOH (6.0 M in decane, 3.8 mL, 22.8 mmol, 6.2 equiv) under a
nitrogen atmosphere and vigorous stirring. The reaction mixture was
immersed into a preheated oil bath at 50 °C and allowed to react for
15 min, during which a color change from very light brown to bright
yellow was observed. Subsequently, the reaction mixture was
immediately cooled down in an ice bath. A sample of the crude
reaction mixture was used for TLC, which showed the formation of a
newly formed species, and the expected incomplete conversion of
(CpiPr3H2)Re(CO)3. In open air, the reaction mixture was washed
with water (2 × 80 mL) and brine (1 × 80 mL). The separated organic
layer was then dried over MgSO4 and concentrated in vacuo. From the
resulting liquid mixture (0.7241 g), the desired product was isolated
over a silica column. The unreacted tricarbonyl rhenium complexes
and organic side products were eluted first with petroleum ether as
eluent. After that, the yellow product fraction was eluted with
dichloromethane and petroleum ether (1:1). A yellow viscous liquid
was obtained (0.5618 g), which contained only one trioxo-rhenium
complex mixed with the high boiling organics. The product could not
be precipitated at room temperature, but yellow, needle-shaped
EXPERIMENTAL SECTION
■
General Considerations. All chemicals including rhenium
complexes were degassed either by exposure in vacuo or freeze−
pump−thaw cycles. Benzene and THF were distilled from Na/
benzophenone. Toluene, THF, and acetonitrile solvents were obtained
from a MBraun MB SPS-800 and stored over 4 Å molecular sieves.
Unless otherwise stated, all other commercial chemicals were used
without further purification. NMR spectra were recorded on a Varian
VNMRS400 (400 MHz) at 298 K. Infra-red spectra were recorded
using a PerkinElmer Spectrum One FT-IR spectrometer in the range
of 650−4000 cm−1. ESI-MS spectra were recorded using a Waters
LCT Premier XE instrument. GC-MS spectra were recorded on a
PerkinElmer Clarus 680 Gas chromatograph, equipped with a
PerkinElmer Elite 5MS column (15m × 0.25 mm ID × 0.25 mm),
G
DOI: 10.1021/acs.organomet.6b00120
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
crystals were grown (65.3 mg, 4.2%) by adding hexane (5 mL) and
storing the resulting solution at −30 °C for 3 days. The obtained
crystals were suitable for X-ray crystal structure determination.
Furthermore, 10% of the unreacted (CpiPr3H2)Re(CO)3 was collected
from the column separation (159.4 mg, 0.345 mmol). In addition, the
water used for the workup was strongly acidic (pH ≤ 1), which
indicates the presence of formed perrhenic acid.
impurities were seen at the sides of the flask. An analytically pure
sample was obtained by recrystallization in a 2:1 mixture of
dichloromethane/hexane (1/0.5 mL) at −30 °C. 1H NMR (400
MHz), CDCl3 (7.26 ppm): 2.08 (s, 6H, Me), 2.32 (s, 6H, Me), 5.95
(s, 1H, CpH). 13C NMR (400 MHz), CDCl3 (77.16 ppm): 10.52,
12.05, 104.64, 122.52, 124.88 ppm. ESI-MS (in CH3CN): calcd. for
[C9H13ReO3 + H]+: 357.0501; found: 357.0434. Elemental analysis
calcd. (%) for C9H13ReO3 (355.45): C 30.42, H 3.69; found, C 29.86,
H 3.93.
1H NMR (400 MHz), 25 °C, CDCl3 (7.26 ppm): δ = 1.29 (s, 6H,
Me, J =), 1.40 (s, 6H, Me), 1.46 (s, 6H, Me), 3.06 (two overlapping
septets, 3H, CHMe2), 6.39 (s, 2H, Cp−H) ppm. 1H NMR (400
MHz), 25 °C, C6D6 (7.16 ppm): 0.81, (s, 6H, Me), 1.19 (s, 6H, Me),
1.21 (s, 6H,Me), 2.73 (sept, 1H, CHMe2), 2.79 (sept, 2H, CHMe2),
5.83 (s, 2H, Cp−H) ppm. 13C NMR (400 MHz), 25 °C, CDCl3
(77.16 ppm): 20.82, 22.62, 24.35, 26.73, 26.83, 103.15, 123.37, 141.15.
FT-IR: 478, 548, 607, 684, 714, 841, 861, 884, 919, 1017, 1056, 1099,
1206, 1284, 1306, 1368, 1384, 1457, 2969, 3114. ESI-MS (in CH3CN
with pyridine and HCOOH as additives): m/z = 506.1701 {[M + py +
H]+, calcd. 506.1706}. Elemental analysis calcd (%) for C14H23ReO3
(426.12): C 39.52, H 5.45; found, C 39.34, H 5.33.
General Procedure for Tricarbonyl Rhenium Oxidation
Profile Determination. Unless otherwise described, all reaction
mixtures were prepared in a Teflon-capped Young-type NMR tube,
which was placed in a Schlenk flask under a nitrogen atmosphere. The
actual reactions were performed inside a Varian 400 MHz
spectrometer with variable-temperature programming. The intended
Cp′Re(CO)3 (10 mg) and 1,4-di-tert-butylbenzene internal standard
(1 equiv) were degassed in vacuo for 30 min, and then dissolved in
degassed C6D6 (0.5 mL). tBuOOH (6.0 M in decane, 6 equiv) was
added by using a calibrated microliter syringe. Immediately, the NMR
tube was closed inside the Schlenk flask under a counterflow of
nitrogen, and the resulting mixture was shaken thoroughly. The initial
Synthesis of (1,2,3-Me3(tetrahydroindenyl))ReO3 6b. (1,2,3-
Me3(tetrahydroindenyl))Re(CO)3 6a (125.0 mg, 0.304 mmol) was
placed in a dried Schlenk flask and degassed in vacuo for 30 min. Then,
it was dissolved in benzene (6.25 mL), followed by the addition of
tBuOOH (6.0 M in decane, 0.3 mL, 6 equiv. based on rhenium) under
a nitrogen atmosphere and vigorous stirring. The reaction mixture was
immersed into a preheated oil bath at 50 °C and allowed to react for
20 min, during which a color change from colorless to bright yellow
was observed. Subsequently, the reaction mixture was immediately
cooled in an ice bath. A small sample of the crude reaction mixture was
analyzed by TLC, which showed the formation of a newly formed
species, and the expected incomplete conversion of 6a. In open air, the
reaction mixture was washed with demineralized water (2 × 7 mL) and
brine solution (1 × 7 mL). The isolated organic layer was then dried
over MgSO4 and concentrated in vacuo. A first product fraction was
isolated by precipitating (1,2,3-Me3(tetrahydroindenyl)ReO3 in
hexane at −30 °C (yield = 21.1 mg), followed by centrifugation,
and decantation. The remaining liquid, which still contained more
product, was separated over a silica column. The unreacted tricarbonyl
rhenium complexes and organic side products were eluted first with
petroleum ether as eluent. After that, the yellow product fraction was
eluted with dichloromethane and petroleum ether (1:1) as eluent
(yield = 23.4 mg). Combination of the two product fractions yielded a
total of yield 36% (44.5 mg, 0.113 mmol). Crystals suitable for X-ray
diffraction were grown from a CH2Cl2/hexane (2:1) mixture at −30
°C. The mass balance of the reaction was verified by collecting the
unreacted 6a (70.8 mg, 0.172 mmol, 57%). In addition, the water used
for the workup was shown to be strongly acidic (pH ≤ 1), which
indicates the presence of formed perrhenic acid.
1
quantities of the reaction mixture were determined by recording a H
NMR spectrum at 25 °C. For a second initial quantity check, whenever
necessary, the temperature was increased to 40 °C for approximately
15 min, in order to shift the broad peroxide peak sufficiently below 7
ppm, so that overlap with the peak of 1,4-di-tert-butylbenzene was
avoided. After that, the temperature was increased to 50 °C, except for
CptttRe(CO)3 that required 70 °C for the reaction to occur. A
1
preacquisition delay H NMR recording with intervals of 5 min was
programmed to follow the change in quantities of starting material and
products during the reaction. Immediately after the reaction, a sample
of the crude reaction mixture was dissolved in hexane and analyzed by
GC-MS. The concentrations were calculated based on the integrals of
Cp−H signals and in some cases methyl signals in the 1H NMR
spectra.
Procedure for the Decomposition Study of 2b to 2c. In a
Schlenk flask, CpttReO3 (20 mg, 0.0486 mmol) was dissolved in
CDCl3 (1.0 mL) under a nitrogen atmosphere and left at room
temperature. Within 1 day, the mixture had turned completely dark,
bearing a red-brown solution and the insoluble dark green solid.
Intermediate 1H NMR analyses showed the formation of many
decomposition products in the solid, and a new complex species in the
liquid. The red-brown solution was filtered, and evaporated in vacuo to
yield a dark red solid (12.2 mg). This solid was then recrystallized
from a mixture of chloroform and diethyl ether (2:1) at −30 °C, and
dark red block-shaped crystals of 2c were obtained.
1H NMR (400 MHz), 25 °C, CDCl3 (7.26 ppm): 1.19 (s, 18H, tBu,
4
4JH,H = 2.4 Hz), 1.49 (s, 18H, tBu), 5.67 (t, 1H, Cp−H, JH,H = 2.4
1H NMR (400 MHz), 25 °C, CDCl3 (7.26 ppm): 1.75 (m, br, 2H,
CH2 exo), 1.83 (m, br, 2H, CH2 endo), 2.10 (s, 6H, Me), 2.21 (s, 3H,
Hz), 6.05 (d, 2H, Cp−H, 4JH,H = 2.4 Hz), 6.13 (d, 2H, Cp−H, 4JH,H
=
2.4 Hz), 6.52 (t, 1H, Cp−H, JH,H = 2.4 Hz) ppm. 13C NMR (400
MHz), 25 °C, CDCl3 (77.16 ppm): 30.30, 30.78, 33.95, 34.79, 77.37,
96.46, 96.92, 107.87, 108.25, 134.18, 137.13 ppm. FT-IR: 702, 731,
811, 874, 901, 915, 931, 1070, 1167, 1251, 1363, 1464, 1702, 2961,
3096 cm−1. ESI-MS (in CH2Cl2): m/z = 1025.1582 {[(CpttReO)2O-
(ReO4)]+, calcd. 1025.1577}; 1421.2621 {[(CpttReO2)3ReO3]+, calcd.
1421.26} Elemental analysis calcd (%) for C26H42Re4O11 (1278.10): C
24.48, H 3.32; found, C 27.11, H 3.76.
4
Me), 2.61 (m, br, 2H, Cp-CH2 exo), 2.73 (m, br, 2H, Cp-CH2 endo
)
ppm. 1H NMR (400 MHz), 25 °C, C6D6 (7.16 ppm): 1.22 (m, br, 2H,
CH2 exo), 1.56 (m, br, 2H, CH2 endo), 1.57 (s, 6H, Me), 1.71 (s, 3H,
Me), 1.90 (m, br, 2H, Cp-CH2 exo), 2.46 (m, br, 2H, Cp-CH2 endo
)
ppm. 13C- NMR (400 MHz), 25 °C, C6D6 (117.06 ppm): 9.60, 9.63,
21.51, 21.55, 110.40, 116.37, 121.38, 122.08 ppm. IR: 790, 813, 875,
909, 1028, 1143, 1192, 1242, 1332, 1366, 1378, 1422, 1714, 2865,
2936, 3401 cm−1. ESI-MS (in CH3CN with pyridine): m/z = 460.0936
{[M + Na + CH3CN]+, calcd. 460.0899};51 476.1229 {[M + pyridine
+ H]+, calcd. 476.1236}.
General Procedure for Catalytic Deoxydehydration. Unless
otherwise described, all reaction mixtures were prepared inside the
glovebox under nitrogen atmosphere. 1,2-Octanediol (286 mg, 1.96
mmol), PPh3 (580 mg, 2.21 mmol), and mesitylene (Internal standard,
225 mg, 1.87 mmol) were dissolved in chlorobenzene (20 mL).
Aliquots (5 mL) of this stock solution were added to each Cp′ReO3
catalyst (0.01 mmol) in a 15 mL thick-walled glass pressure tube (Ace)
fitted with a Teflon screw-cap. Then, the closed reaction flask was
immersed into a preheated oil bath at 135 °C. At regular intervals, to
plot the reaction profile, the reaction tube was cooled down in an ice
bath and then warmed to room RT before taking aliquots of the
reaction mixture inside the glovebox. However, the reaction was
stopped only once after 15 h when the final catalytic data were needed.
Synthesis of (Me4CpH)ReO3 7b. (Me4CpH)Re(CO)3 7a (0.7 g, 1.8
mmol) was degassed in a dried Schlenk flask by stirring for 30 min in
vacuo, followed by the addition of benzene (35 mL) and H2O2 (35% in
water, 3.5 mL, 20 equiv). The resulting mixture was refluxed (75 °C)
for 1 h, after which the yellow organic layer was separated from the
water layer, which was extracted with benzene (2 × 5 mL) in open air.
Then, the organic layer was washed with water (2 × 50 mL), 5%
NaHCO3 solution (1 × 50 mL), and brine (2 × 50 mL), before it was
dried over MgSO4. After the removal of benzene in vacuo, 7b was
obtained as a yellow solid in 33% yield (0.209 g), and traces of oily
H
DOI: 10.1021/acs.organomet.6b00120
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
The samples were diluted (ca. 10 times) with acetone (for olefins) or
ethyl acetate (for diols) in open air for GC analysis. All olefinic
products were known compounds and were calibrated against
mesitylene for quantification. Alternatively, all the reagents (0.5
mmol substrate, 0.01 mmol catalyst, 0.55 mmol PPh3, and 0.5 mmol
mesitylene) were weighed in a scintillation glass vial for the
experiments carried out at 135 or 180 °C for 15 h, which also applies
to the control experiments starting from 1-octene. Then, 5 mL of PhCl
was used to dissolve before transferring into the pressure tubes.
X-ray Crystal Structure Determinations. Compound 2b.
C13H21O3Re, Fw = 411.50, yellow plate, 0.45 × 0.28 × 0.07 mm3,
monoclinic, P21/m (no. 11), a = 6.3051(3), b = 19.3260(14), c =
6.3244(5) Å, β = 118.867(3)°, V = 674.88(8) Å3, Z = 2, Dx = 2.025 g/
cm3, and μ = 9.00 mm−1. 10858 Reflections were measured on a
Bruker Kappa ApexII diffractometer with sealed tube and Triumph
monochromator (λ = 0.71073 Å) at a temperature of 150(2) K up to a
resolution of (sin θ/λ)max = 0.66 Å−1. The crystal appeared to be
nonmerohedrally twinned with a 2-fold rotation about hkl = (0,0,1) as
twin operation. Consequently, two orientation matrices were used for
the integration of the intensity data with the Eval15 software.41 An
analytical absorption correction was performed with PLATON36
(correction range 0.08−0.69) resulting in a HKLF5 file.42 1992
reflections were unique (Rint = 0.054), of which 1983 were observed [I
> 2σ(I)]. The structure was solved with automated Patterson methods
using the program DIRDIF-08.43 Least-squares refinement was
performed with SHELXL-201344 against the F2 of all reflections. N
on-hydrogen atoms were refined freely with anisotropic displacement
parameters. All hydrogen atoms were included in calculated positions
and refined with a riding model. 86 parameters were refined with no
restraints. R1/wR2 [I > 2σ(I)]: 0.0260/0.0648. R1/wR2 [all refl.]:
0.0261/0.0649. S = 1.179. Batch scale factor for the second twin
component BASF = 0.2995(16). Residual electron density between
−1.49 and 2.73 e/Å3. Geometry calculations and checking for higher
symmetry were performed with the PLATON program.36
Compound 3b. C14H23O3Re, Fw = 425.52, yellow block, 0.26 ×
0.11 × 0.08 mm3, monoclinic, P21/n (no. 14), a = 8.5282(2), b =
12.0181(3), c = 14.2586(3) Å, β = 93.993(1)°, V = 1457.85(6) Å3, Z =
4, Dx = 1.939 g/cm3, and μ = 8.33 mm−1. 35418 Reflections were
measured on a Bruker Kappa ApexII diffractometer with sealed tube
and Triumph monochromator (λ = 0.71073 Å) at a temperature of
150(2) K up to a resolution of (sin θ/λ)max = 0.65 Å−1. Intensity data
were integrated with the Eval15 software.41 Analytical absorption
correction and scaling was performed with SADABS45 (correction
range 0.35−0.64). 3343 Reflections were unique (Rint = 0.015), of
which 3257 were observed [I > 2σ(I)]. The structure was solved with
SHELXT.46 Least-squares refinement was performed with SHELXL-
201444 against the F2 of all reflections. Non-hydrogen atoms were
refined freely with anisotropic displacement parameters. All hydrogen
atoms were located in difference Fourier maps. Hydrogen atoms H4
and H5 of the Cp ring were refined freely with isotropic displacement
parameters. All other hydrogens were refined with a riding model. 178
parameters were refined with no restraints. R1/wR2 [I > 2σ(I)]:
0.0110/0.0263. R1/wR2 [all refl.]: 0.0115/0.0264. S = 1.090. Extinction
parameter EXTI = 0.00048(4). Residual electron density between
−0.90 and 1.39 e/Å3. Geometry calculations and checking for higher
symmetry were performed with the PLATON program.36
Non-hydrogen atoms were refined freely with anisotropic displace-
ment parameters. All hydrogen atoms were located in difference
Fourier maps and refined with a riding model. 178 parameters were
refined with 1 restraint (floating origin). R1/wR2 [I > 2σ(I)]: 0.0098/
0.0227. R1/wR2 [all refl.]: 0.0112/0.0230. S = 1.065. Flack parameter48
x = 0.291(11). Residual electron density was between −0.49 and 0.25
e/Å3. Geometry calculations and checking for higher symmetry were
performed with the PLATON program.36
Compound 7b. C9H13O3Re, Fw = 355.39, yellow plate, 0.08 × 0.05
× 0.03 mm3, orthorhombic, Pnma (no. 62), a = 5.9964(3), b =
13.1856(7), c = 12.0268(7) Å, V = 950.91(9) Å3, Z = 4, Dx = 2.482 g/
cm3, and μ = 12.75 mm−1. 13014 reflections were measured on a
Bruker Kappa ApexII diffractometer with sealed tube and Triumph
monochromator (λ = 0.71073 Å) at a temperature of 150(2) K up to a
resolution of (sin θ/λ)max = 0.70 Å−1. Intensity data were integrated
with the Saint software.49 Multiscan absorption correction and scaling
were performed with SADABS45 (correction range 0.32−0.43). 1452
reflections were unique (Rint = 0.033), of which 1280 were observed [I
> 2σ(I)]. The structure was solved with direct methods using
SHELXS-97.50 Least-squares refinement was performed with
SHELXL-9750 against the F2 of all reflections. Non-hydrogen atoms
were refined freely with anisotropic displacement parameters. All
hydrogen atoms were located in difference Fourier maps. Hydrogen
atom H1 of the Cp ring was refined freely with an isotropic
displacement parameter. All other hydrogens were refined with a
riding model. 69 parameters were refined with no restraints. R1/wR2 [I
> 2σ(I)]: 0.0189/0.0416. R1/wR2 [all refl.]: 0.0251/0.0434. S = 1.067.
Residual electron density was between −1.07 and 2.13 e/Å3. Geometry
calculations and checking for higher symmetry were performed with
the PLATON program.36
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.organo-
met.6b00120.
The NMR spectra of all the new complexes (PDF)
X-ray data (CIF)
AUTHOR INFORMATION
Corresponding Author
*E-mail: R.J.M.KleinGebbink@uu.nl.
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This research has been performed within the framework of the
CatchBio program. We gratefully acknowledge the support of
the Smart Mix Program of The Netherlands Ministry of
Economic Affairs and The Netherlands Ministry of Education,
Culture, and Science. The X-ray diffractometer has been
financed by The Netherlands Organization for Scientific
Research (NWO). M.E.M. acknowledges funding from the
European Union Seventh Framework Programme (FP7/2007-
2013) under grant agreement PIIF-GA-2012-327306 (IIF-
Marie Curie grant) and the Sectorplan Natuur-en Scheikunde
(Tenure-track grant at Utrecht University).
Compound 6b. C12H17O3Re, Fw = 395.45, yellow-green block, 0.23
× 0.13 × 0.11 mm3, orthorhombic, Pna21 (no. 33), a = 10.9327(4), b
= 8.1327(2), c = 13.2961(4) Å, V = 1182.18(6) Å3, Z = 4, Dx = 2.222
g/cm3, and μ = 10.27 mm−1. 13606 reflections were measured on a
Bruker Kappa ApexII diffractometer with sealed tube and Triumph
monochromator (λ = 0.71073 Å) at a temperature of 150(2) K up to a
resolution of (sin θ/λ)max = 0.65 Å−1. The crystal was cracked into
several crystal fragments. Only the intensity data of the major fragment
were integrated with the Eval15 software.41 Analytical absorption
correction and scaling were performed with SADABS45 (correction
range 0.21−0.57). 2709 reflections were unique (Rint = 0.016), of
which 2544 were observed [I > 2σ(I)]. The structure was solved with
direct methods using SIR-97.47 Least-squares refinement was
performed with SHELXL-201444 against the F2 of all reflections.
REFERENCES
■
(1) Adduci, L. L.; McLaughlin, M. P.; Bender, T. A.; Becker, J. J.;
Gagne, M. R. Angew. Chem., Int. Ed. 2014, 53, 1646−1649.
(2) ten Dam, J.; Hanefeld, U. ChemSusChem 2011, 4, 1017−1034.
(3) Schlaf, M. Dalton Trans. 2006, 4645−4653.
(4) Werpy, T.; Petersen, G.; Aden, A.; Bozell, J.; Holladay, J.; White,
J.; Manheim, A.; Eliot, D.; Lasure, L.; Jones, S. Top Value Added
I
DOI: 10.1021/acs.organomet.6b00120
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Chemicals from Biomass: Volume 1 - Results of Screening for Potential
Candidates from Sugars and Synthesis Gas, 2004.
(5) Raju, S.; Moret, M.-E.; Klein Gebbink, R. J. M. ACS Catal. 2015,
5, 281−300.
(6) Dethlefsen, J. R.; Fristrup, P. ChemSusChem 2015, 8, 767−775.
(7) Boucher-Jacobs, C.; Nicholas, K. M. Top. Curr. Chem. 2014, 353,
163−184.
(8) Shiramizu, M.; Toste, F. D. Angew. Chem., Int. Ed. 2012, 51,
8082−8086.
(9) Shiramizu, M.; Toste, F. D. Angew. Chem., Int. Ed. 2013, 52,
12905−12909.
(10) Liu, S.; Senocak, A.; Smeltz, J. L.; Yang, L.; Wegenhart, B.; Yi, J.;
Kenttamaa, H. I.; Ison, E. A.; Abu-Omar, M. M. Organometallics 2013,
32, 3210−3219.
(11) Yi, J.; Liu, S.; Abu-Omar, M. M. ChemSusChem 2012, 5, 1401−
1404.
(36) Spek, A. L. Acta Crystallogr., Sect. D: Biol. Crystallogr. 2009, 65,
148−155.
(37) Herrmann, W. A.; Herdtweck, E.; Floel, M.; Kulpe, J.;
̈
Kusthardt, U.; Okuda, J. Polyhedron 1987, 6, 1165−1182.
̈
(38) Herrmann, W. A.; Serrano, R.; Kusthardt, U.; Guggolz, E.;
̈
Nuber, B.; Ziegler, M. L. J. Organomet. Chem. 1985, 287, 329−344.
(39) Bencze, E; Mink, J.; Nem
V.; Kuhn, F. E. J. Organomet. Chem. 2002, 642, 246−258.
(40) Gable, K. P.; Phan, T. N. J. Am. Chem. Soc. 1994, 116, 833−836.
(41) Schreurs, A. M. M.; Xian, X.; Kroon-Batenburg, L. M. J. J. Appl.
Crystallogr. 2010, 43, 70−82.
(42) Herbst-Irmer, R.; Sheldrick, G. M. Acta Crystallogr., Sect. B:
Struct. Sci. 1998, 54, 443−449.
́
́
eth, C.; Herrmann, W. A.; Lokshin, B.
̈
(43) Beurskens, P. T.; Beurskens, G.; de Gelder, R.; Garcia-Granda,
S.; Gould, R. O.; Smits, J. M. M. (2008) The DIRDIF2008 Program
System; Crystallography Laboratory, University of Nijmegen: The
Netherlands.
(44) Sheldrick, G. M. Acta Crystallogr. 2015, C71, 3−8.
(45) Sheldrick, G. M. (1999) SADABS: Area-Detector Absorption
(12) Herrmann, W. A.; Romao, C. C.; Fischer, R. W.; Kiprof, P.; de
Mer
́
ic de Bellefon, C. Angew. Chem., Int. Ed. Engl. 1991, 30, 185−186.
(13) Herrmann, W. A.; Kuhn, F. E. Acc. Chem. Res. 1997, 30, 169−
180.
(14) Herrmann, W. A.; Kuhn, F. E.; Romao, C. C.; Huy, H. T.;
̈
Correction, v2.10; Universitat Gottingen: Gottingen, Germany.
̈
̈
̈
(46) Sheldrick, G. M. Acta Crystallogr., Sect. A: Found. Adv. 2015, 71,
̈
̃
3−8.
Wang, M.; Fischer, R. W.; Kiprof, P.; Scherer, W. Chem. Ber. 1993,
126, 45−50.
(47) Altomare, A.; Burla, M. C.; Camalli, M.; Cascarano, G. L.;
Giacovazzo, C.; Guagliardi, A.; Moliterni, A. G. G.; Polidori, G.;
Spagna, R. J. Appl. Crystallogr. 1999, 32, 115−119.
(48) Flack, H. D. Acta Crystallogr., Sect. A: Found. Crystallogr. 1983,
39, 876−881.
(15) Cook, G. K.; Andrews, M. A. J. Am. Chem. Soc. 1996, 118,
9448−9449.
(16) Gable, K. P.; Zhuravlev, F. A.; Yokochi, A. F. T. Chem. Commun.
1998, 799−800.
(49) Bruker (2001) SAINT-Plus; Bruker AXS Inc.: Madison, WI.
(50) Sheldrick, G. M. Acta Crystallogr., Sect. A: Found. Crystallogr.
2008, 64, 112−122.
(51) The CH3CN adduct can be explained by the Lewis acidity of the
high-valent trioxo-rhenium species, the use of CH3CN for the sample
preparation, and the presence of residual Na+ sources in the ESI-MS
setup.
(17) (a) Raju, S.; Jastrzebski, J. T. B. H.; Lutz, M.; Klein Gebbink, R.
J. M. ChemSusChem 2013, 6, 1673−1680. (b) Raju, S.; Jastrzebski, J. T.
B. H.; Lutz, M.; Witteman, L.; Dethlefsen, J. L.; Fristrup, P.; Moret,
M.-E.; Klein Gebbink, R. J. M. Inorg. Chem. 2015, 54, 11031−11036.
(c) Raju, S.; van Slagmaat, C. A. M. R.; Lutz, M.; Kleijn, H.;
Jastrzebski, J. T. B. H.; Moret, M.-E.; Klein Gebbink, R. J. M.
Manuscript in preparation.
(18) Kuhn, F. E.; Herrmann, W. A.; Hahn, R.; Elison, M.; Blumel, J.;
̈
̈
Herdtweck, E. Organometallics 1994, 13, 1601−1606.
́
(19) Herrmann, W. A.; Taillefer, M.; de Meric de Bellefon, C.; Behm,
J. Inorg. Chem. 1991, 30, 3247−3248.
(20) Hogerl, M.; Kuhn, F. E. Z. Anorg. Allg. Chem. 2008, 634, 1444−
̈
̈
1447.
(21) Okuda, J.; Herdtweck, E.; Herrmann, W. A. Inorg. Chem. 1988,
27, 1254−1257.
(22) Kuhn, F. E.; Herrmann, W. A.; Hahn, R.; Elison, M.; Blumel, J.;
̈
̈
Herdtweck, E. Organometallics 1994, 13, 1601−1606.
(23) Herrmann, W. A.; Voss, E.; Floel, M. J. Organomet. Chem. 1985,
̈
297, C5−C7.
(24) Herrmann, W. A.; Correia, J. D. G.; Kuhn, F. E.; Artus, G. R. J.;
̈
Romao, C. C. Chem.−Eur. J. 1996, 2, 168−173.
̃
(25) Thiel, W. R.; Fischer, R. W.; Herrmann, W. A. J. Organomet.
Chem. 1993, 459, C9−C11.
(26) Herrmann, W. A.; Okuda, J. J. Mol. Catal. 1987, 41, 109−122.
(27) Herrmann, W. A.; Serrano, R.; Kusthardt, U.; Ziegler, M. L.;
̈
Guggolz, E.; Zahn, T. Angew. Chem., Int. Ed. Engl. 1984, 23, 515−517.
(28) Klahn-Oliva, A. H.; Sutton, D. Organometallics 1984, 3, 1313−
1314.
(29) Gable, K. P.; Phan, T. N. J. Organomet. Chem. 1994, 466, C5−
C6.
(30) Herrmann, W. A.; Kiprof, P.; Rypdal, K.; Tremmel, J.; Blom, R.;
Alberto, R.; Behm, J.; Albach, R. W.; Bock, H. J. Am. Chem. Soc. 1991,
113, 6527−6537.
(31) Wallis, J. M.; Kochi, J. K. Inorg. Chim. Acta 1989, 160, 217−221.
(32) Wolowiec, S.; Kochi, J. K. Inorg. Chem. 1991, 30, 1215−1221.
(33) Szyperski, T.; Schwerdtfeger, P. Angew. Chem., Int. Ed. Engl.
1989, 28, 1228−1231.
(34) Herrmann, W. A.; Serrano, R.; Schafer, A.; Kusthardt, U.;
̈
̈
Ziegler, M. L.; Guggolz, E. J. Organomet. Chem. 1984, 272, 55−71.
(35) Burrell, A. K.; Cotton, F. A.; Daniels, L. M.; Petricek, V. Inorg.
Chem. 1995, 34, 4253−4255.
J
DOI: 10.1021/acs.organomet.6b00120
Organometallics XXXX, XXX, XXX−XXX