Alkene isomerisation catalysed by a ruthenium PNN pincer complex

Article abstract of DOI:10.1002/chem.201403236

The [Ru(CO)H(PNN)] pincer complex based on a dearomatised PNN ligand (PNN: 2-di-tert-butylphosphinomethyl-6-diethylaminomethylpyridine) was examined for its ability to isomerise alkenes. The isomerisation reaction proceeded under mild conditions after activation of the complex with alcohols. Variable-temperature (VT) NMR experiments to investigate the role of the alcohol in the mechanism lend credence to the hypothesis that the first step involves the formation of a rearomatised alkoxide complex. In this complex, the hemilabile diethylamino side-arm can dissociate, allowing alkene binding cis to the hydride, enabling insertion of the alkene into the metal-hydride bond, whereas in the parent complex only trans binding is possible. During this study, a new uncommon Ru⁰ coordination complex was also characterised. The scope of the alkene isomerisation reaction was examined. The catalyst tested positive! A dearomatised ruthenium PNN (2-di-tert-butylphosphinomethyl-6-diethylaminomethylpyridine) pincer complex, [Ru(CO)H(PNN)], was evaluated as an alkene isomerisation catalyst. The isomerisation reaction was greatly accelerated by the addition of alcohols, in particular isopropanol. Isomerisation of terminal to internal alkenes took place at room temperature. A mechanism was proposed based on variable-temperature NMR spectroscopy.

Full text of DOI:10.1002/chem.201403236

DOI: 10.1002/chem.201403236
Full Paper
&
Homogenous Catalysis
Alkene Isomerisation Catalysed by a Ruthenium PNN Pincer
Complex**
Sꢀbastien Perdriau,[a] Mu-Chieh Chang,[a] Edwin Otten,*[a] Hero J. Heeres,*[b] and
Johannes G. de Vries*[a, c]
Abstract: The [Ru(CO)H(PNN)] pincer complex based on
a dearomatised PNN ligand (PNN: 2-di-tert-butylphosphino-
methyl-6-diethylaminomethylpyridine) was examined for its
ability to isomerise alkenes. The isomerisation reaction pro-
ceeded under mild conditions after activation of the com-
plex with alcohols. Variable-temperature (VT) NMR experi-
ments to investigate the role of the alcohol in the mecha-
nism lend credence to the hypothesis that the first step in-
volves the formation of a rearomatised alkoxide complex. In
this complex, the hemilabile diethylamino side-arm can dis-
sociate, allowing alkene binding cis to the hydride, enabling
insertion of the alkene into the metal–hydride bond, where-
as in the parent complex only trans binding is possible.
During this study, a new uncommon Ru0 coordination com-
plex was also characterised. The scope of the alkene isomeri-
sation reaction was examined.
Introduction
tion catalysts have been developed based on various metals
such as Pd,[15–17] Ru,[18–22] Ti,[23–25] V,[26] Fe,[27–29] Rh,[30–32] Mo,[33]
Ni,[34–36] etc., but also metal-free systems such as frustrated
Lewis pairs.[37] One of the most active catalysts is the rutheni-
um-based “alkene zipper” developed by Grotjahn and co-work-
ers.[38,39]
Although isomerisation is often observed as an unwanted side
reaction,[1–3] it is also put to good use, in particular in tandem
reactions, such as isomerising hydroformylation, carbonylation
or hydrocyanation. Examples include the Evonik/Oxeno isomer-
ising hydroformylation of a mixture of butenes to n-pentanal,[4]
and the Dupont process for the production of adiponitrile
through isomerising hydrocyanation of 3-pentenenitrile.[5] High
yields and selectivities have been obtained for catalysis involv-
ing alkene isomerisation in tandem with hydroformylation/re-
duction to alcohols,[6] methoxycarbonylation,[7] metathesis,[8]
conjugate addition,[9] hydroboration,[10] cyclisation[11] and trial-
kylsilylation.[12] Other examples are the isomerisation of allyl
benzenes such as estragol or safrole, which are of great inter-
est for the fragrance industry.[13,14] A wide range of isomerisa-
Recently, researchers have focused on the selectivity of the
alkene isomerisation. Beller and Grotjahn have developed cata-
lysts which allow a selective one-position shift of the double
bond with high selectivity for E isomers,[27,40] whereas other
groups focused on the synthesis of the less thermodynamically
favoured Z isomers.[41]
Two different mechanisms are generally believed to be oper-
ative for alkene isomerisation. The alkyl mechanism starts with
an organometallic hydride species. Coordination of the alkene
to the metal centre followed by insertion into the MÀH bond
gives two possible metal alkyl intermediates. b-H elimination
will then generate either the same alkene or its double-bond
shifted isomer. The p-allyl mechanism initiates with coordina-
tion of the double bond to a low-valent metal centre. A p-allyl
complex is generated together with an MÀH bond by oxidative
addition of the CÀH bond in the allylic position. A reductive
elimination can take place in two directions and will provide
either the olefin isomer or the starting olefin.[42]
[a] S. Perdriau, M.-C. Chang, Dr. E. Otten, Prof. Dr. J. G. de Vries
Stratingh Institute for Chemistry
University of Groningen, Nijenborgh 4
9747 AG Groningen (The Netherlands)
Fax: (+31)50-3634296
E-mail: edwin.otten@rug.nl
[b] Prof. Dr. H. J. Heeres
Department of Chemical Engineering
University of Groningen, Nijenborgh 4
9747 AG Groningen (The Netherlands)
E-mail: h.j.heeres@rug.nl
Among the catalysts developed for this reaction, only a few
belong to the family of the pincer complexes[43–47] and, to our
surprise, the reactivity of PNN (PNN: 2-di-tert-butylphosphino-
methyl-6-diethylaminomethylpyridine) ruthenium complexes
with alkenes has not been explored. In particular, the pyridine-
based PNN ruthenium pincer complexes are of great interest,
due to the unique participation of the ligand in the reaction
(non-innocent ligand).[48] This involves dearomatisation/aroma-
tisation of the pyridine ring of the ligand without a change in
[c] Prof. Dr. J. G. de Vries
Leibniz Institut fꢀr Katalyse e.V.
An der Universitꢁt Rostock
Albert-Einstein-Strasse 29a
18059 Rostock (Germany)
E-mail: johannes.devries@Catalysis.de
[**] PNN=2-di-tert-butylphosphinomethyl-6-diethylaminomethylpyridine
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201403236.
Chem. Eur. J. 2014, 20, 15434 – 15442
15434
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
perature (entry 7). Substantial progress was achieved with the
addition of an equivalent of isopropanol to the substrate. An
overnight reaction in [D8]THF led to almost full conversion of
both 1-octene and estragol even at room temperature (en-
tries 6 and 8).
A GC/MS-FID (flame ionisation detector) analysis of the mix-
ture from the reaction with 1-octene after 4.5 days in [D8]THF
indicates a good selectivity for 2-octene (77%), present as
a cis/trans mixture (Figure 1). Other octene isomers were ob-
Scheme 1. Activation of the PNN ruthenium catalyst.
the metal oxidation state. This mechanism permits the facile
activation of a variety of XÀH bonds (Scheme 1).
With this unique feature, a wide range of catalytic reactions
has been performed, such as dehydrogenative coupling of al-
cohols into esters or of amines and alcohols into amides, ester
hydrogenation, etc.[49–52] The presence of a RuÀH bond in the
PNN Ru complex 1 and the robustness of this class of com-
pounds prompted us to explore its use as an olefin isomerisa-
tion catalyst, the results of which are presented here.
Results and Discussion
Preliminary results
Figure 1. Isomeric composition of 1-octene isomerisation (Table 1, entry 6).
Initial alkene isomerisation studies using complex 1 were per-
formed with 1-octene in [D6]benzene plus catalyst (2 mol%).
Low activity was observed at room temperature (3% conver-
sion after 16 h, entry 1, Table 1). Higher reaction temperatures
(658C) improved the conversion somewhat to 22% after 8 h
(entry 2). A higher catalyst loading did not improve the rate of
the reaction and only 42% conversion was reached after 2.5
days. Also, the use of a more polar solvent ([D8]THF) did not
bring significant improvement either at room temperature or
higher (entries 4 and 5). Similar results were obtained with es-
tragol as substrate, even though a slightly better conversion
(52%) was observed after an overnight reaction at room tem-
served as well in smaller quantities (11% total for 3- and 4-
octene). Octane (11%) was present in the mixture as the only
side-product. Since this type of catalyst is known to dehydro-
genate alcohols, we suggest that transfer hydrogenation from
isopropanol is taking place as a side reaction during the pro-
cess to yield the saturated product. We were intrigued by the
dramatic influence of isopropanol on the isomerisation rate,
and decided to study the interaction of 1 with isopropanol in
more detail.
The role of isopropanol: a VT-NMR study
Table 1. Isomerisation reactions.
The role of isopropanol was studied by performing variable-
temperature NMR (VT-NMR) on complex 1 and isopropanol. A
first series of NMR spectra of 1 were recorded in [D8]toluene
solution in the absence of isopropanol between 258C and
À508C. In this temperature range, no significant change of the
characteristic NMR resonances of 1 were observed, except for
a broadening of all signals and a slight up-field shift of the sig-
nals corresponding to the CH2 protons of the NEt2 group.
When one equivalent of isopropanol was added to this solu-
Entry Substrate Catalyst Solvent
Temp. Time Conversion
loading
[mol%]
[8C]
[h]
[%]
1
2
1-octene
“
2
”
C6D6
“
RT
65
16
4
3
17
8
22
3
4
“
“
10
”
”
RT
RT
40
2
traces of
isomers
42
traces of
isomers
11
60
3
[D8]THF
“
1
tion and an H NMR spectrum was recorded at room tempera-
ture,
a
mixture of the starting material
1
(RuÀH, d=
21
4
5
6
“
“
”
”
15
À26.7 ppm) and free isopropanol was observed together with
a small amount of the known dihydride 2 (Scheme 2, d=
À4.4 ppm) which stems from isopropanol dehydrogenation[49]
(confirmed by the presence of acetone). By cooling down the
21
2
27
48
[D8]THF+1equiv RT
iPrOH[a]
19
96
3
93
>99
15
7
8
estragol
“
“
”
[D8]THF
RT
reaction mixture,
a
new hydride appears (RuÀH, d=
21
2
52
61
À15.9 ppm), which becomes the major complex below À508C
(3, Figure 2A). NOESY NMR measurements at À708C (Fig-
ure 2B) show cross-peaks diagnostic for exchange between
free isopropanol and the new ruthenium complex 3, indicating
[D8]THF+1equiv RT
iPrOH[a]
19
86
[a] iPrOH/substrate, 1/1 molar ratio.
Chem. Eur. J. 2014, 20, 15434 – 15442
www.chemeurj.org
15435
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Scheme 2. Reactivity of the PNN ruthenium complex 1 with isopropanol.
that this species contains an isopropoxide group (CH, d=
4.70 ppm; diastereotopic Me, d=1.95/1.75 ppm), which has
also been confirmed by COSY NMR measurement at À708C
(Figure 2C). Thus, the NMR data indicate the new species 3 to
be the complex [Ru(CO)(H)(OiPr)(PNN)], in which the ligand is
rearomatised by proton transfer from isopropanol (Scheme 2).
Compound 1 and related Ru complexes have been shown to
react with XÀH bonds (X=H, OH, NHR)[49,52] through metal–
ligand cooperation. Reaction of 1 with isopropanol to form 3,
in which the NEt2 group is presumably hemilabile, has been
suggested by Milstein and co-workers to be a key step in the
(catalytic) reactivity of 1.[49] This has recently been confirmed
by detailed DFT calculations, which shows that a complex simi-
lar to 3 but with a 2-methoxyethoxide rather than an isoprop-
oxide ligand is only 2.3 kcalmolÀ1 uphill from 1 plus 2-methox-
yethanol due to an unfavourable entropic contribution.[53]
Compound 3 should be favoured at low temperature accord-
ing to these calculations, in agreement with the experimental
observations. Our data show that formation of 3 is reversible
and warming the NMR tube to room temperature results in
the quantitative regeneration of 1. The limited temperature
range in which 3 can be studied prevents a detailed characteri-
sation of the fluxional behaviour of the NEt2 arm. However, as
indicated by 1H NMR at À308C (Figure 2A), 1 and 3 are present
in a 1 to 1 ratio. A more detailed analysis of the NOESY NMR
spectra at À308C shows cross-peaks diagnostic for exchange
between the two diastereotopic N(CH2CH3)2 signals of 1 (d=
0.65 and 0.80 ppm), while a single broad signal is observed in
the case of 3 (d=0.96 ppm). This suggests that indeed this
arm is less strongly bound (hemilabile) in 3 (see the Support-
ing Information).
Figure 2. Reaction of PNN complex with iPrOH: A) VT-NMR spectrum of the
~
^
hydride region; B) NOESY spectrum at À708C, : iPrOH, : Ru-OiPr; C) COSY
spectrum at À708C.
Equilibria similar to that between 1/isopropanol and 3 have
been demonstrated for NÀH activation of amines by [RuH(CO)-
(PNP)] (PNP=2,6-bis-(di-tert-butylphosphinomethyl)pyridine).[54]
In this case, the observation at room temperature by NMR
spectroscopy of the starting material with free amine, the coor-
dination complex or the activation of the amine NÀH bond
was highly dependent on the electronic properties of the sub-
strate. We thus decided to react compound 1 with one equiva-
lent of the more acidic o-allyl phenol. Instant formation of
a new species (7) with a hydride resonance at d=À15.7 ppm
was observed similar to the isopropoxide complex 3
(Scheme 3). This product possesses an aromatic pyridine ring
as well as diastereotopic ÀP(tBu)2 and ÀN(CH2CH3)2 groups, in-
dicative of a C1 symmetric compound. It also shows a single
peak in the 31P{1H} NMR spectrum (d=114 ppm). X-ray charac-
terisation of 7 indicated the presence of a phenoxide ligand
trans to the hydride (Figure 3).
Coordination of the ÀNEt2 group rather than the olefin was
observed in both solution and solid state. A comparison with
Chem. Eur. J. 2014, 20, 15434 – 15442
www.chemeurj.org
15436
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
and 44 ppm), proving that all hydrogens of the coordinated
ethylene are inequivalent (Supporting Information). Based on
this, we formulate 6 as the 18e organometallic compound [Ru-
(C2H4)2(CO)(PNN)], in which the ruthenium metal centre is in
the zero oxidation state, two ethylene molecules are coordinat-
ed trans to each other, and the diethylamino arm is off the
metal centre.
Scheme 3. Reactivity of the PNN ruthenium complex 1 with o-allylphenol.
Although not very common, ruthenium(0) PNP pincer com-
plexes have already been synthesised[55] and were suggested
to be intermediates in a water splitting reaction with the cata-
lyst used in the present study.[50]
The stability of complex 6 was tested by warming the NMR
tube to room temperature. This resulted in the regeneration of
the starting material 1 even though the process seems slow.
The position of the equilibrium reaction (Scheme 4) is highly
dependent on the ethylene pressure. Whereas no quantitative
reaction was obtained with only 1 bar of ethylene, removal of
pressure leads to a full reversible formation of the starting
catalyst 1.
When the reaction of the PNN ruthenium complex 1 with
ethylene was performed in the absence of isopropanol, the
catalyst remained mostly unchanged. Only traces of 6 were ob-
served, which underlines the important role of isopropanol for
the reactivity with alkenes. The formation pathway of 6 with-
out isopropanol is still unclear. It is possible that ethylene plays
the role of a hydride transfer agent. Another explanation
Figure 3. X-ray structure of complex 7. Hydrogen atoms, ÀNEt2 and ÀPtBu2
groups have been omitted for clarity. Selected bond distances [ꢂ]: Ru1ÀP3,
2.261; Ru1ÀN1, 2.108; Ru1ÀN2, 2.256; Ru1ÀH300, 1.523; Ru1ÀO2, 2.200;
Ru1ÀC28, 1.821; C28ÀO1, 1.168.
the known hydroxide complex[50] shows only two main differ-
ences: a shorter Ru1ÀO2 bond length and an elongated Ru1À would be a simple rearrangement of 1, as has recently been
N2 in the case of the phenoxide.
observed with bispyridine PNN ruthenium complexes.[56]
Based on literature precedents and the NMR studies report-
ed here, we propose a mechanism for 1-octene isomerisation
with compound 1, as shown in Scheme 5. It involves the initial
activation of isopropanol, allowing the formation of a vacant
coordination site cis to the hydride. After coordination of the
alkene, the isomerisation proceeds via the formation of a ruthe-
nium-alkyl complex.
A second NMR experiment with the PNN ruthenium com-
plex 1 was performed in the presence of ethylene in an at-
tempt to characterise alkene coordination complexes such as
the putative intermediate 5. For this purpose, equimolar
amounts of isopropanol and 1 were mixed in [D8]toluene to-
gether with ethylene (4 bar). However, NMR spectroscopy
showed no evidence for the formation of the expected RuII
complex with a coordinated olefin group. Instead, leaving the
reaction mixture overnight at À208C led to the quantitative
formation of a new Ru species 6 (Scheme 4). The 31P{1H} NMR
spectrum shows a single peak at d=75.8 ppm. The ligand
backbone was confirmed to be aromatic with equivalent alkyl
groups for both the ÀN(CH2CH3)2 and ÀP(tBu)2 side arms,
which suggests that 6 possesses Cs symmetry. Surprisingly, no
Isomerisation of 1-octene followed over time
The composition of the 1-octene isomerisation reaction mix-
ture was followed over time to gain insight in the relative rates
of the reactions (Scheme 6). The reaction was performed in iso-
propanol at room temperature with a catalyst loading of
0.5 mol%. Samples taken at different time intervals were ana-
lysed by GC/MS-FID after calibration of the different octene
isomers and using cyclooctane as external standard. The re-
sults are presented in Figure 4.
1
hydride resonance could be located, and the H NMR spectrum
shows new signals attributed to coordinated ethylene, the in-
tegration of which corresponds to two ethylene molecules per
pincer ligand. These resonances are shifted upfield from free
ethylene (d=3.10, 2.87 and 1.50 ppm) and appear in a 1:2:1
ratio. A HSQC NMR experiment indicated the signal at d=
2.87 ppm to correlate with two different carbon atoms (d=41
We observe that 98% of 1-octene is already converted after
5 h. At this point in time 2-octene is the main product of the
reaction, reaching almost 88%. 3-Octene is slowly emerging,
whereas only traces of 4-octene are detected. Samples taken
after one, two and 21 days of reaction show that the composi-
tion is slowly approaching the calculated equilibrium
composition.
A comparison between the calculated[30] and the experimen-
tal isomeric octene distribution reveals that after 21 days the
thermodynamic equilibrium is indeed approached, although
not fully reached (Table 2). This might be due to catalyst deac-
Scheme 4. Proposed CH2=CH2 adduct complex of 1.
Chem. Eur. J. 2014, 20, 15434 – 15442
www.chemeurj.org
15437
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Catalyst loading
A study of the influence of the
catalyst loading on the selectivi-
ty of the reaction was performed
using estragol. The latter was
chosen instead of 1-octene as it
simplifies the analyses considera-
bly, because only two isomerisa-
tion products are formed (cis-
and trans-anethole), along with
the saturated side product 10, 1-
methoxy-4-propylbenzene (see
Scheme 7 and Figure 5).
Comparing the results after
17 h, we find that although the
amount of product increases at
higher catalyst loading, the same
is true for the formation of the
Scheme 5. Proposed mechanism for octene isomerisation catalysed by 1 in the presence of isopropanol.
Table 2. Calculated versus experimental product distribution.[a]
Octene isomers DrH8 DDrH8 Calc. [%] Exptl. [%] (excl. octane)
45 h
504 h
1-octene
À125
10
0.5
24.6
5.1
21.8
9.7
30.1
8.3
–
0.8 (0.8)
38.1 (40.6)
9.4 (10.1)
25.0 (26.7)
4.5 (4.8)
13.4 (14.3)
2.6 (2.8)
6.3
0.6 (0.6)
28.0 (31.8)
7.1 (8.0)
25.8 (29.3)
4.5 (5.1)
18.5 (21.0)
3.6 (4.0)
11.9
trans-2-octene
cis-2-octene
trans-3-octene
cis-3-octene
trans-4-octene
cis-4-octene
octane
À115.5 0.5
À119.4 4.4
À115.8 0.8
À117.8 2.8
Scheme 6. Isomerisation of 1-octene catalysed by 1 in the presence of iso-
propanol.
À115
0
À118.2 3.2
tivation but possibly also to kinetic effects due to the large re-
activity differences between the terminal and internal alkenes.
The isomerisation reaction from a terminal alkene to a 2-alkene
is in general much faster than the isomerisation of an internal
alkene which is in accordance with the results obtained. It is
also not very surprising that the internal alkenes are mostly
present in their trans form, this being the thermodynamically
more stable isomer. Finally, formation of octane was also ob-
served under these conditions. However, the formation of this
side product stayed below 10%, even after two days.
–
–
[a] DrH8=hydrogenation enthalpies [kJmolÀ1]; DDrH8=relative formation
enthalpies [kJmolÀ1]; Calc.
RT))]*100.
[%]=exp(ÀDDrH8/RT)/[S(exp(ÀDDrH8/
Scheme 7. Estragol isomerisation.
saturated side product 10. Thus, a maximum yield of
anethole is reached with 5 mol% catalyst loading.
Alcohol screening
Although the addition of isopropanol clearly increas-
es the rate of the reaction, it also leads to the forma-
tion of saturated product through transfer hydroge-
nation (Figure 1). For this reason we tested the use
of other alcohols (ethanol, t-amylalcohol) for 1-
octene isomerisation. With ethanol, the reaction per-
forms well with a 1-octene conversion of 83%
(Figure 6) after 15 h, which is about 15% lower than
in isopropanol. Ethanol is also capable of transfer hy-
Figure 4. Isomerisation of 1-octene catalysed by 1 followed over time.
Chem. Eur. J. 2014, 20, 15434 – 15442
www.chemeurj.org
15438
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
formed in isopropanol as this resulted in the highest rates
(vide supra). Product compositions were determined by GC
and, in addition, isolated yields are reported as well. Reactions
with vinyl cyclohexane (entry 1a–d) showed that the isomerisa-
tion of a terminal double bond to an internal tri-substituted
double bond is possible with 1 even at room temperature
with reasonable conversion. C=C double-bond isomerisation
into the ring was not observed. Saturated side product ac-
counted for 16% in the product composition. Although in-
creasing the temperature is beneficial for the conversion, it
also decreases the selectivity. Performing the reaction neat
with only 1.2 mol/mol of isopropanol led to a lower conversion
with an increased amount of saturated ethyl cyclohexane. It is
likely that all the alcohol has been consumed in the transfer
hydrogenation reaction.
^
Figure 5. Estragol isomerisation as a function of catalyst loading ( estragol,
~
&
anethole, 1-methoxy-4-propylbenzene) after 17 h.
4-Methyl-1-pentene was converted selectively into 4-methyl-
2-pentene (Table 3, entry 2). None of the other isomers or the
saturated product were detected.
No isomerisation of citronellene (Table 3, entries 3a,b) oc-
curred at room temperature. Although reactivity was observed
at 508C, large amounts of saturated product were also formed.
This issue could be solved by using tert-amyl alcohol. Full iso-
merisation of the terminal double bond was observed whereas
the internal double bond was not reactive under the condi-
tions employed. The small percentage of saturated compound
found can be explained by secondary alcohol impurities pres-
ent in the tert-amyl alcohol.
Figure 6. Alcohol effect on octene isomerisation.
Surprisingly high temperatures were required to convert 1,7-
octadiene (Table 3, entries 4a,b). The main products were 1,6
and 2,6-octadiene with very little octene and no saturated
octane detected.
drogenation and octane is observed in amounts up to 6%. The
isomeric composition of the reaction mixtures in EtOH and
iPrOH are comparable after an overnight reaction, though less
4-octene is obtained with ethanol. This suggests that not only
the 1-octene to 2-octene isomerisation but also the subse-
quent 2-octene to 3- and 4-octene in ethanol are slower than
in isopropanol.
Isomerisation of 1,5-cyclooctadiene (COD) (Table 3, entrie-
s 5a,b), proceeded sluggishly, yielding mainly 1,4-COD at room
temperature, whereas cyclooctene was the major product
when the reaction was performed at 708C.
It is clear that the undesired transfer hydrogenation reaction
to saturated alkanes is directly linked to the presence of a hy-
drogen atom on the carbon in the alpha position of the alco-
hol. Thus, the use of a tertiary alcohol should solve this issue.
With this in mind, a reaction was performed using tert-amyl al-
cohol as solvent. A reaction at room temperature for a 16 h re-
action time resulted in very low conversion (23%) with essen-
tially only 2-octene as product. At higher temperatures (708C),
84% conversion was obtained. In addition, no octane was de-
tected, which confirmed our hypothesis. Furthermore, the se-
lectivity for 2-octene was high, which indicates that the C=C
internal bonds are not prone to isomerisation in the presence
of a tertiary alcohol. The higher temperature needed with tert-
amyl alcohol may be caused by steric factors that influence
alkene coordination, particularly when the double bond is in-
ternal. This is supported by the very small amount of 3 and 4-
octene isomers found, even at 708C.
Allyl benzene as well as alkyloxy substituted estragol and sa-
frole (Table 3, entries 6–8b) were easily isomerised at room
temperature with good selectivities. The thermodynamically
most stable trans isomer was always obtained as the main
product.
Whereas diethylallylamine was not reactive at room temper-
ature in isopropanol, or neat at 708C, full conversion was ob-
served at 708C in the presence of a few equivalents of isopro-
panol (Table 3, entry 9). This again illustrates the important role
of isopropanol.
The isomerisation of unsaturated linear alcohols using 1 was
also tested. Interestingly, the reaction of 5-hexen-1-ol resulted
in a highly selective isomerisation to 4-hexene-1-ol (94%)
(Table 3, entry 10). Knowing that the alcohol function in these
substrates can also be activated by the catalyst, both coordina-
tion of the -OH group and the double bond may take place. A
similar type of coordination complex has already been ob-
served by Bꢃckvall and co-workers.[57] The stability of this possi-
ble intermediate may be influenced by the chain length, thus
restricting the extent of the isomerisation. The fact that iso-
merisation of 9-decen-1-ol leads to a mixture of isomers seems
to prove this hypothesis (Table 3, entry 11).
Substrate scope and limitations
The scope of the isomerisation reaction was determined by
testing a number of substrates (Table 3). All reactions were per-
Chem. Eur. J. 2014, 20, 15434 – 15442
www.chemeurj.org
15439
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
When unsaturated ketones
were subjected to the isomerisa-
tion reaction, full conversion of
the starting material was ob-
tained (Table 3, entries 12–13).
However, the product was not
the expected C=C double-bond
isomer but rather the unsaturat-
ed alcohol, resulting from trans-
fer hydrogenation of the ketone
function. Moreover, and to our
surprise, no isomerisation of the
double bond was detected.
When tert-amyl alcohol was
used in an attempt to avoid
transfer hydrogenation, no con-
version at all was obtained with
full recovery of the starting ma-
terial. The chain length between
the alcohol function and the
double bond is the same as in
the product of the 5-hexen-1-ol
isomerisation, which strength-
ens the hypothesis of the forma-
tion of the stable intermediate
where both the alcohol and
alkene may be coordinated to
the metal centre.
Table 3. Substrate scope.
Entry Substrate
Conditions
Solvent
Product composition[a]
Conv. Isolated
[%]
yield[b] [%]
1a
b
RT
508C
iPrOH
iPrOH
neat
84% (16%)
70% (30%)
77% (23%)
61
87
22
–
68
–
c
708C, 6 h
1.2 equiv iPrOH/
cat.
2
RT
[D8]-iPrOD
>99%
100
91
3a
b
c
RT
508C
708C
iPrOH
iPrOH
t-AmOH
–
0
–
45% (55%)
95% (5%)
49
100
94
–
4a
b
RT
708C
iPrOH
iPrOH
isomers
isomers 94%+octene 6%
7
–
77
63
5a
b
RT
iPrOH
iPrOH
17
79
–
29%
33%
71%
2% (65%)[c]
708C
71
98% (E/Z:90/10)
(2%)
6
RT
iPrOH
80
77
70
92% (E/Z:87/13)
(8%)
7a
b
RT
iPrOH
85
91
708C
t-AmOH
100% (E/Z:98/2) 100
8a
b
RT
708C
iPrOH
t-AmOH
88% (2%)
99% (1%)
57
100
76
87
Finally, room temperature iso-
merisation of allyl phenols
(Table 3, entries 14a–15b) was
performed and shown to be un-
successful. As in the case of o-
allyl phenol (vide supra),
a stable alkoxide complex may
be formed. However, isomerisa-
tion activity was observed when
performing a stoichiometric re-
action of the catalyst with these
phenolic substrates. After one
day, 50% conversion into the
conjugated isomer was ob-
served and this was close to full
conversion after two days. After
the addition of a second equiva-
lent of phenol, the isomerisation
process proceeds and is com-
plete after 3.5 days at room
temperature. Thus, although iso-
merisation seems possible in the
presence of phenols, the reac-
tion rates are very low. This was
further supported by the iso-
merisation of o-allyl phenol at
1308C in toluene. After 16 h, full
conversion to 2-(prop-1-en-1-yl)-
phenol was observed.
9a
b
RT
iPrOH
–
0
100
–
91
708C, 2.5 equiv neat
iPrOH/cat.
708C
>99%[d]
c
neat
–
0
–
10
RT
iPrOH
97
83
94%
6%
11
RT
iPrOH
100
79
12a
b
RT
708C
iPrOH
t-AmOH
100%
–
88
0
34
–
13a
b
RT
708C
iPrOH
t-AmOH
100%
–
>99 89
0
–
14a
b
RT
1308C
iPrOH
[D8]toluene
–
0
100
–
75
100%
15a
b
RT
508C
iPrOH
iPrOH
–
–
0
0
–
–
[a] Number in parentheses indicates the saturated product from hydrogenation; [b] Isolated product consists
of a mixture of the different products and starting material; [c] Cyclooctene; [d] Traces of saturated product are
observed in GC-MS/FID.
Chem. Eur. J. 2014, 20, 15434 – 15442
www.chemeurj.org
15440
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
VT-NMR experiments
Conclusion
In an oven-dried Young’s NMR tube, a solution of 1 (0.022 mmol)
was dissolved in [D8]toluene and isopropanol (1 equiv) was added.
The reaction mixture was degassed using three freeze–pump–thaw
cycles. The required amount of ethylene to reach 1 or 4 bars ethyl-
ene in the NMR Young’s tube was collected in a calibrated gas
bulb. The tube was plunged into liquid nitrogen and the ethylene
was further condensed into it. The tube was then carefully sealed
and slowly warmed up to room temperature. The desired tempera-
ture was reached stepwise. The tube was arbitrarily left for 10–
15 min. for establishment of equilibrium before NMR experiments
were recorded.
The [Ru(CO)H(PNN)] catalyst developed by Milstein was exam-
ined for its ability to isomerise alkenes. Isomerisation was
found to be slow but could be accelerated by alcohols, in par-
ticular, using isopropanol as an additive. Transfer hydrogena-
tion also occurs to some extent upon the use of this alcohol.
This side reaction could be suppressed by the use of tertiary
alcohols such as tert-amylalcohol, but at the cost of catalyst ac-
tivity. VT NMR experiments and a model reaction with ethene
lend credence to the hypothesis that reaction with isopropanol
leads to formation of a [Ru(CO)H(OR)(PNN)] complex by metal/
ligand cooperation. In this complex the diethylamino sidearm
can dissociate, allowing binding of the alkene in the cis posi-
tion relative to the hydride, followed by insertion of the
alkene. In the parent complex, the alkene can only bind trans
to the hydride, which explains why the isomerisation reaction
is slow in the absence of alcohols. A new bis ethylene adduct
Ru0 complex was discovered during the VT-NMR study. A range
of substrates was examined, allowing determination of the
scope and limitations of the synthetic methodology. Whereas
a single shift of a terminal to a linear 2-alkene proceeds
smoothly at room temperature, isomerisation of a terminal
alkene to a tri-substituted internal alkene requires a higher
temperature, leading to lower selectivity. Isomerisation of inter-
nal alkenes is much slower and tri-substituted internal alkenes
do not react at all. Functional groups like alcohols, ethers,
phenyl groups or di-alkyl amines do not influence the isomeri-
sation, whereas with phenols stable complex formation re-
duces the catalyst activity.
CCDC-999441 (7) contains the supplementary crystallographic data
for this paper. This data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif.
Acknowledgements
The authors thank the Smart Mix Program of the Dutch Minis-
try of Economic Affairs and the Dutch Ministry of Education,
Culture and Science, Royal DSM NV, Merck & Co and Avantium
for funding of this work within the framework of the CatchBio
Program. Paul Alsters, Lavinia Panella and Bernard Kaptein (all
DSM) are thanked for stimulating discussions. We thank Pieter
van der Meulen for helping to set-up the VT-NMR experiments.
Keywords: alkene · isomerisation · alcohols · pincer complex ·
ruthenium
[1] S. H. Hong, D. P. Sanders, C. W. Lee, R. H. Grubbs, J. Am. Chem. Soc.
2005, 127, 17160–17161.
[2] C. S. Higman, L. Plais, D. E. Fogg, ChemCatChem 2013, 5, 3548–3551.
[3] K. E. Allen, D. M. Heinekey, A. S. Goldman, K. I. Goldberg, Organometal-
lics 2013, 32, 1579–1582.
[4] J. G. de Vries, in Science of Synthesis Reference Library: C-1 Building
Blocks in Organic Synthesis 1 (Ed.: P. W. N. M. van Leeuwen), Thieme,
Stuttgart, 2014, pp. 193–227.
[5] M. Kimura, Y. Tamaru in Modern Organonickel Chemistry, Wiley-VCH,
Weinheim, 2005, pp. 137–170.
When comparing this catalyst system with other published
catalysts we see that most other catalysts operate at higher
temperatures, usually around 708C. We have achieved alkene
isomerisation at room temperature, which is rare. Also, the cat-
alyst is selective enough to allow the isomerisation reaction to
stop at the first internal alkene with good selectivity, which
most catalysts cannot do at higher temperatures.
[6] L. Wu, I. Fleischer, R. Jackstell, I. Profir, R. Franke, M. Beller, J. Am. Chem.
Soc. 2013, 135, 14306–14312.
[7] C. Jimꢀnez-Rodriguez, G. R. Eastham, D. J. Cole-Hamilton, Inorg. Chem.
Commun. 2005, 8, 878–881.
[8] D. M. Ohlmann, N. Tschauder, J.-P. Stockis, K. Gooßen, M. Dierker, L. J.
Gooßen, J. Am. Chem. Soc. 2012, 134, 13716–13729.
[9] D. M. Ohlmann, L. J. Goossen, M. Dierker, Chem. Eur. J. 2011, 17, 9508–
9519.
[10] J. V. Obligacion, P. J. Chirik, J. Am. Chem. Soc. 2013, 135, 19107–19110.
[11] L. J. Gooßen, D. M. Ohlmann, M. Dierker, Green Chem. 2010, 12, 197–
200.
Experimental Section
Typical isomerisation reaction
All reactions were carried out under a protective nitrogen atmos-
phere using a glove box and Schlenk techniques.
[12] T. Huber, D. Firlbeck, H. M. Riepl, J. Organomet. Chem. 2013, 744, 144–
148.
A
vial (4 mL) was loaded with a solution of the catalyst
(0.044 mmol) in isopropanol (2 mL) and cyclooctane (20 mL) was
added as internal standard followed by the substrate (1.8 mmol).
The vial was sealed with a screw cap and septum after addition of
a stir bar and the reaction mixture was stirred at room temperature
for 16 h. Conversion was determined using GC-MS/FID. The reac-
tion was quenched by exposure to air. The product was isolated
after evaporation of isopropanol and passing the residue over
a short silica column with gradient elution from pentane to 1:5
AcOEt/pentane. The product obtained was further analysed by
NMR techniques.
[13] B. Lastra-Barreira, P. Crochet, Green Chem. 2010, 12, 1311–1314.
[14] T. Karlen, A. Ludi, Helv. Chim. Acta 1992, 75, 1604–1606.
[15] T. Kochi, T. Hamasaki, Y. Aoyama, J. Kawasaki, F. Kakiuchi, J. Am. Chem.
Soc. 2012, 134, 16544–16547.
[16] D. Gauthier, A. T. Lindhardt, E. P. K. Olsen, J. Overgaard, T. Skrydstrup, J.
Am. Chem. Soc. 2010, 132, 7998–8009.
[17] P. Mamone, M. F. Grꢄnberg, A. Fromm, B. A. Khan, L. J. Gooßen, Org.
Lett. 2012, 14, 3716–3719.
[18] C. J. Yue, Y. Liu, R. He, J. Mol. Catal. A 2006, 259, 17–23.
[19] S. K. Sharma, V. K. Srivastava, R. V. Jasra, J. Mol. Catal. A 2006, 245, 200–
209.
Chem. Eur. J. 2014, 20, 15434 – 15442
www.chemeurj.org
15441
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[20] T. J. Donohoe, T. J. C. O’Riordan, C. P. Rosa, Angew. Chem. 2009, 121,
1032–1035; Angew. Chem. Int. Ed. 2009, 48, 1014–1017.
[21] F. Ding, Y. Sun, F. Verpoort, Eur. J. Inorg. Chem. 2010, 1536–1543.
[22] F. Ding, S. Van Doorslaer, P. Cool, F. Verpoort, Appl. Organomet. Chem.
2011, 25, 601–607.
[23] E. Shaviv, M. Botoshansky, M. S. Eisen, J. Organomet. Chem. 2003, 683,
165–180.
[24] S. Achyutha Rao, M. Periasamy, J. Organomet. Chem. 1988, 342, 15–20.
[25] M. Akita, H. Yasuda, K. Nagasuna, A. Nakamura, Bull. Chem. Soc. Jpn.
1983, 56, 554–558.
[26] J. Nieman, J. W. Pattiasina, J. H. Teuben, J. Organomet. Chem. 1984, 262,
157–169.
[27] R. Jennerjahn, R. Jackstell, I. Piras, R. Franke, H. Jiao, M. Bauer, M. Beller,
ChemSusChem 2012, 5, 734–739.
[28] M. Mayer, A. Welther, A. Jacobi von Wangelin, ChemCatChem 2011, 3,
1567–1571.
[41] C. Chen, T. R. Dugan, W. W. Brennessel, D. J. Weix, P. L. Holland, J. Am.
Chem. Soc. 2014, 136, 945–955.
[42] R. Crabtree, The Organometallic Chemistry of the Transition Metals,
Wiley-VCH, Weinheim, 2009.
[43] S. Biswas, Z. Huang, Y. Choliy, D. Y. Wang, M. Brookhart, K. Krogh-Jes-
persen, A. S. Goldman, J. Am. Chem. Soc. 2012, 134, 13276–13295.
[44] A. R. Chianese, S. E. Shaner, J. A. Tendler, D. M. Pudalov, D. Y. Shopov, D.
Kim, S. L. Rogers, A. Mo, Organometallics 2012, 31, 7359–7367.
[45] S. M. M. Knapp, S. E. Shaner, D. Kim, D. Y. Shopov, J. A. Tendler, D. M. Pu-
dalov, A. R. Chianese, Organometallics 2014, 33, 473–484.
[46] K. McGowan, A. Veige, WO Pat. 2,012,047,517, 2012, 1.
[47] K. McGowan, K. Abboud, A. Veige, Organometallics 2011, 30, 4949–
4957.
[48] C. Gunanathan, D. Milstein, Acc. Chem. Res. 2011, 44, 588–602.
[49] J. Zhang, G. Leitus, Y. Ben-David, D. Milstein, J. Am. Chem. Soc. 2005,
127, 10840–10841.
[29] K. R. Sawyer, E. A. Glascoe, J. F. Cahoon, J. P. Schlegel, C. B. Harris, Orga-
nometallics 2008, 27, 4370–4379.
[30] T. C. Morrill, C. A. D’Souza, Organometallics 2003, 22, 1626–1629.
[31] A. Seayad, M. Ahmed, H. Klein, R. Jackstell, T. Gross, M. Beller, Science
2002, 297, 1676–1678.
[32] C. Cai, S. Yu, G. Liu, Adv. Synth. Catal. 2011, 353, 2665–2670.
[33] S. C. Marinescu, A. J. King, R. R. Schrock, R. Singh, P. Mꢄller, M. K. Takase,
Organometallics 2010, 29, 6816–6828.
[34] E. G. Kuntz, G. Godard, J. M. Basset, O. M. Vittori, Oil Gas Sci. Technol.
2007, 62, 781–785.
[35] X. Wang, L. K. Woo, J. Mol. Catal. A 1998, 130, 171–176.
[36] H. J. Lim, C. R. Smith, T. V. RajanBabu, J. Org. Chem. 2009, 74, 4565–
4572.
[37] M. Pꢀrez, L. J. Hounjet, C. B. Caputo, R. Dobrovetsky, D. W. Stephan, J.
Am. Chem. Soc. 2013, 135, 18308–18310.
[50] S. W. Kohl, L. Weiner, L. Schwartsburd, L. Konstantinovski, L. J. W.
Shimon, Y. Ben-David, M. A. Iron, D. Milstein, Science 2009, 324, 74–77.
[51] E. Balaraman, B. Gnanaprakasam, L. J. W. Shimon, D. Milstein, J. Am.
Chem. Soc. 2010, 132, 16756–16758.
[52] C. Gunanathan, Y. Ben-David, D. Milstein, Science 2007, 317, 790–792.
[53] H. Li, X. Wang, F. Huang, G. Lu, J. Jiang, Z.-X. Wang, Organometallics
2011, 30, 5233–5247.
[54] E. Khaskin, M. A. Iron, L. J. W. Shimon, J. Zhang, D. Milstein, J. Am. Chem.
Soc. 2010, 132, 8542–8543.
[55] E. Fogler, M. A. Iron, J. Zhang, Y. Ben-David, Y. Diskin-Posner, G. Leitus,
L. J. W. Shimon, D. Milstein, Inorg. Chem. 2013, 52, 11469–11479.
[56] R. Barrios-Francisco, E. Balaraman, Y. Diskin-Posner, G. Leitus, L. J. W.
Shimon, D. Milstein, Organometallics 2013, 32, 2973–2982.
[57] J. B. ꢂberg, M. C. Warner, J.-E. Bꢃckvall, J. Am. Chem. Soc. 2009, 131,
13622–13624.
[38] D. B. Grotjahn, C. R. Larsen, J. L. Gustafson, R. Nair, A. Sharma, J. Am.
Chem. Soc. 2007, 129, 9592–9593.
[39] D. B. Grotjahn, WO Pat. 2,009,015,360, 2009, DOI US8501032B2.
[40] C. R. Larsen, G. Erdogan, D. B. Grotjahn, J. Am. Chem. Soc. 2014, 136,
1226–1229.
Received: April 24, 2014
Published online on September 26, 2014
Chem. Eur. J. 2014, 20, 15434 – 15442
www.chemeurj.org
15442
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim