Transformation of Methyl Linoleate to its Conjugated Derivatives with Simple Pd(OAc)2/Lewis Acid Catalyst

Article abstract of DOI:10.1007/s11746-017-3052-5

With the rapid depletion of fossil resources, the exploitation of biomass to partly replace fossil resources as the source of carbon in the chemical industry constitutes a promising alternative for the near future. This work introduces catalytic transformation of vegetable oil, i.e., methyl linoleate, to its conjugated esters by a simple Pd(OAc)₂/Sc(OTf)₃ catalyst, which has extensive applications in industry. It was found that adding non-redox metal ions like Sc(III) to a simple Pd(OAc)₂ catalyst can effectively improve its isomerization activity in toluene/t-BuOH solvent, whereas Pd(OAc)₂ alone is inactive. Preliminary mechanistic investigations together with previous studies suggested that the in situ-generated heterobimetallic Pd(II)/Sc(III) dimer serves as the key species for methyl linoleate isomerization, and the reaction proceeds by [1,3]-hydrogen shift mechanism involving a formal Pd(II)/Pd(IV) cycle.

Full text of DOI:10.1007/s11746-017-3052-5

J Am Oil Chem Soc
DOI 10.1007/s11746-017-3052-5
ORIGINAL PAPER
Transformation of Methyl Linoleate to its Conjugated Derivatives
with Simple Pd(OAc) /Lewis Acid Catalyst
2
1
1
1
1
1
Ahmed M. Senan · Sicheng Zhang · Shuhao Qin · Zhuqi Chen · Guochuan Yin
Received: 31 March 2017 / Revised: 19 September 2017 / Accepted: 2 October 2017
©
AOCS 2017
Abstract With the rapid depletion of fossil resources, the
exploitation of biomass to partly replace fossil resources
as the source of carbon in the chemical industry con-
stitutes a promising alternative for the near future. This
work introduces catalytic transformation of vegetable oil,
i.e., methyl linoleate, to its conjugated esters by a simple
Pd(OAc) /Sc(OTf) catalyst, which has extensive applica-
Introduction
Recently, versatile vegetable oils have attracted attention for
industrial applications because of their renewable availabil-
ity with expanded growth as needed [1–7]. Beside their wide
applications as biofuel after transesteriꢀcation, transforma-
tion of vegetable oils to conjugated derivatives is also very
attractive [8–10]. For example, conjugated methyl linoleate
(CML) has large markets in the preparation of biopolymers
and drying oils for paints and coatings, and they are also
used in versatile nutritional, therapeutic and pharmacologic
applications because of their beneꢀcial physiological eꢁects
[11–16]. Currently, conjugated vegetable oils are commer-
cially produced by an alkali isomerization process in which
vegetable oils are treated with a strong base like NaOH or
KOH at high temperature (i.e. 200–250 °C), followed by
neutralization with acid [17, 18]. Apparently, the on-going
alkali isomerization process is not environmentally benign
in view of its harsh conditions. Consequently, transition
metal ions-based catalytic isomerizations of vegetable oils,
for example, methyl linoleate (ML), to the corresponding
conjugated oils has recently attracted attention. In the lit-
erature, Wilkinson-type catalysts like Ru and Rh complexes
2
3
tions in industry. It was found that adding non-redox metal
ions like Sc(III) to a simple Pd(OAc) catalyst can eꢁec-
2
tively improve its isomerization activity in toluene/t-BuOH
solvent, whereas Pd(OAc) alone is inactive. Preliminary
2
mechanistic investigations together with previous studies
suggested that the in situ-generated heterobimetallic Pd(II)/
Sc(III) dimer serves as the key species for methyl linoleate
isomerization, and the reaction proceeds by [1,3]-hydrogen
shift mechanism involving a formal Pd(II)/Pd(IV) cycle.
Keywords Biomass transformation · Methyl linoleate
isomerization · Pd(II) catalyst · Lewis acid · Promotional
eꢁect
having the PPh ligand have demonstrated encouraging
3
Electronic supplementary material The online version of this
article (doi:10.1007/s11746-017-3052-5) contains supplementary
material, which is available to authorized users.
catalytic eꢂciency under low temperature [19–26]. Larock
reported that a [RhCl(C H )] catalyst with SnCl and
8
12
2
2
(
p-CH C H ) P (TPP) can isomerize soybean oil with 90%
3 6 4 3
*
Guochuan Yin
yield in absolute ethanol at 60 °C under Ar. In this work, the
gyin@hust.edu.cn
role of SnCl was assigned as a Lewis acid to abstract chlo-
2
1
ride anions from an in situ-generated RhCl(TPP) complex
School of Chemistry and Chemical Engineering,
Key Laboratory of Material Chemistry for Energy
Conversion and Storage (Huazhong University of Science
and Technology), Ministry of Education, Hubei Key
Laboratory of Material Chemistry and Service Failure,
Huazhong University of Science and Technology,
Wuhan 430074, People’s Republic of China
2
+
to form the active [Rh(TPP) ] species [21].
2
In exploring non-redox metal ions serving as Lewis acids
to improve the catalytic eꢂciency of redox metal complexes
[
27–32], we unexpectedly found that, in Wacker-type oxi-
dations, the Lewis properties of copper(II) play signiꢀcant
1
3
Vol.:(0123456789)

J Am Oil Chem Soc
roles in oleꢀn oxidations in addition to its redox proper-
ties [33]. Inspired by these ꢀndings, we explored non-redox
metal ions as Lewis acid-promoted oleꢀn isomerization
of toluene/t-BuOH (2.75:0.25, v/v) in a glass tube; methyl
linoleate (53.0 mg, 0.18 mmol) was added to the above solu-
tion and the reaction mixture was magnetically stirred at
110 °C in an oil bath for 9 h. Conjugated diene mixtures as
the total products were determined by GC analysis with an
internal standard method. Next, the solvent was removed
under reduced pressure, and the remaining mixtures were
dissolved in CH Cl . Then, it was puriꢀed by ꢃash chroma-
and its oxidative coupling with indoles by a Pd(OAc) cata-
2
lyst. In those reactions, the role of Lewis acid is assigned
to improve the C–H bond activation ability of the Pd(II)
cation through the formation of heterometallic Pd(II)/Lewis
acid (LA) species, and related DFT calculations have sup-
ported this assignment [34, 35]. Here, we report an example
2
2
tography with a mixture of petroleum ether and ethyl acetate
(10:1, v/v) as eluent solvent, aꢁording the CML products in
86.8 ± 5.7% yield. In this procedure, the Pd(OAc) /Sc(OTf)
of such a simple Pd(OAc) /LA system in biomass utilization,
2
that is, catalyzing isomerization of methyl linoleate to its
2
3
corresponding conjugated derivatives, whereas Pd(OAc)
catalysts were stuck to the silica gel medium. Control experi-
ments using Pd(OAc) or Sc(OTf) as the catalyst were car-
2
alone is inactive.
2
3
1
ried out in parallel. H NMR (400 MHz, CDCl ) characteri-
3
zation of isolated conjugated methyl linoleate products gave:
Experimental
δ 5.99 (d, J = 13.8 Hz, 1H), 5.74–5.46 (S, 1H), 5.46–4.85
(
m, 2H), 3.66 (s, 3H), 2.30 (t, J = 7.5 Hz, 2H), 2.07–1.92
The chemicals used in this work were all commercially
(m, 4H), 1.61 (dd, J = 15.7, 8.8 Hz, 2H), 1.42–1.20 (m,
16H, −(CH ) ), 0.89 (dd, J = 13.4, 7.1 Hz, 3H,CH CH ).
available. Pd(OAc) was purchased from Strem Chemi-
2
2 8
3
2
−
−
cals, Al(OTf) , Sc(OTf) , Y(OTf) (OTf =CF SO ) came
3
3
3
3
3
from Shanghai Dibai Chemical, and Mg(OTf) , Ca(OTf) ,
General Procedures for Catalytic Kinetics of Methyl
Linoleate Isomerization to its Conjugated Derivatives
by Pd(OAc) /Sc(OTf) Catalyst
2
2
Zn(OTf) , NaOTf from Shanghai Shaoyuan. Methyl
2
linoleate came from Aladdin, and octadecane from Tianjin
Heons BioChem. The solvents including t-BuOH, methanol,
ethanol, isopropyl alcohol, ethylbenzene and toluene were
purchased from the local Sino Pharm Chemical Reagent.
Gas chromatography (GC) analysis was performed using
n-octadecane as the internal standard on a 9750 GC (Fuli,
Zhejiang, China) using the following column and tempera-
ture program: AE.XE-60: 30 m × 0.32 mm, i.d., 0.5 μm
2
3
In a typical procedure, Pd(OAc) (2 mg, 0.009 mmol),
2
Sc(OTf) (8.9 mg, 0.018 mmol) and methyl linoleate
3
(53.0 mg, 0.18 mmol) were dissolved in 3 mL of toluene/t-
BuOH (2.75:0.25, v/v) in a glass tube, and n-octadecane was
added as internal standard to the above reaction solution.
The resulting reaction mixtures were magnetically stirred
at 110 °C in an oil bath. The conjugated diene products
were determined by GC analysis using an internal standard
method at diꢁerent intervals. A control experiment using
(
Lanzhou Atech Technologies, Gansu, China); the constant
temperature of 180 °C was held for 10 min for analysis. The
conversion was determined as the consumption of ML as
determined by GC relative to the initial ML added. The total
yield was determined as the formation of all the CML prod-
ucts as determined by GC relative to the initial ML added.
All the data for the conversions and yields are the average
of three runs of the reaction, and the selectivity of the total
CML was obtained by dividing the average total yield of
CML by the average ML conversion.
Pd(OAc) alone as catalyst was conducted in parallel.
2
Results and Discussion
Using methyl linoleate (ML) as a vegetable oil model, a
series of non-redox metal ions were tested as Lewis acids
One GC trace graph is provided in the electronic supple-
mentary material as Fig. S6. NMR analysis was performed
on a Bruker AV400, GC–MS analysis was performed on
an Agilent-B-WAX spectrometer, and UV–Vis spectra
were recorded on an Analytikjena specord 205 UV–Vis
spectrophotometer.
to promote simple Pd(OAc) catalyzed oleꢀn isomerization
2
in toluene/t-BuOH at 110 °C, and the results are summa-
rized in Table 1. In the literature, the reported catalysts for
eꢂcient vegetable oil isomerization include Ru- and Rh-
based metal complexes with phosphine ligands like PPh ,
3
while a Pd-based catalyst was not reported [19–26]; even
in common oleꢀn isomerization, the reported Pd catalysts
were less active than those that were Ru or Rh based [36].
General Procedures for Catalytic Isomerization
of Methyl Linoleate to its Conjugated Derivatives
by Pd(OAc) /Sc(OTf) Catalyst
As shown in Table 1, Pd(OAc) alone as a catalyst is also
2
inactive for the transformation of ML to its conjugated
2
3
methyl linoleates (CML). Adding NaOTf to Pd(OAc) did
2
In a typical procedure, Pd(OAc) (2 mg, 0.009 mmol) and
not exhibit any improvement, and the promotional eꢁects of
2
Sc(OTf) (8.9 mg, 0.018 mmol) were dissolved in 3 mL
bivalent metal salts like Ca(OTf) , Mg(OTf) , and Zn(OTf)
3
2
2
2
1
3

J Am Oil Chem Soc
Table 1 Non-redox metal
Entrya
Conv. (%)b
Total yield of CML (%)c
Selectivity (%)d
Catalyst
Lewis acid
ions promoted methyl linoleate
isomerization to its conjugated
1
2
3
4
5
6
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
–
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
–
3.1 ± 2.0
Trace
Trace
N.D.
Trace
Trace
Trace
Trace
Trace
–
–
–
–
–
–
–
64.0
96.0
96.7
67.1
82.0
derivatives by Pd(OAc)
2
NaOTf
catalyst in toluene/t-Butanol
Zn(OTf)2
Mg(OTf)2
Ca(OTf)2
InCl3
Sc(OTf)3
Cu(OTf)2
Al(OTf)3
Sc(OTf)3
Y(OTf)3
ZrCl4
4.9 ± 2.3
4.2 ± 1.2
2.7 ± 1.6
8.6 ± 3.6
87.0 ± 3.6
76.4 ± 3.8
53.0 ± 2.8
99.6 ± 0.4
18.9 ± 3.7
35.8 ± 2.7
e
7
8
9
48.9 ± 2.4
52.0 ± 1.8
96.3 ± 2.4(86.8 ± 5.7)
12.7 ± 1.9
29.4 ± 3.0
1
1
1
0f
1
2
ND Not detected
a
Conditions: 3 mL toluene/t-BuOH (2.75:0.25, v/v); Pd(OAc) (3 mM), Lewis acid (6 mM), methyl
2
linoleate (60 mM), 110 °C, 9 h, under air
b
Conversion was determined as the consumption of ML as determined by GC relative to the initial ML
added
c
The total yield was determined as the formation of all the CML products as determined by GC relative to
the initial ML added
d
The selectivity of the total CML was obtained by dividing the average total yield of CML by the average
ML conversion
e
f
1
3
8.6 ± 3.6% yield of isolated products were obtained as saturated ester through H NMR analysis
The value presented in parentheses was isolated yield
are also minimal, except that adding Cu(OTf) to Pd(OAc)
2
2
Conversion %
Total Yield%
Pd(II) alone
oꢁered 76.4 ± 3.8% conversion of ML with 48.9 ± 2.4%
1
00
yield of CML. Notably, trivalent metal ions are more eꢁec-
2
+
tive than bivalent metal ions (except Cu ), in which adding
Al(OTf) or Sc(OTf) to Pd(OAc) provided 53.0 ± 2.8%
80
Pd(II)/Sc(III)
3
3
2
6
4
2
0
0
0
0
or 99.6 ± 0.4% conversion of ML with 52.0 ± 1.8% or
9
8
6.3 ± 2.4% yield of CML, respectively. In particular,
6.8 ± 5.7% of the isolated yield can be achieved with
Pd(OAc) alone
Sc(OTf) as Lewis acid. It is worth mentioning that the CML
2
3
products are a collection of products not just the 9,11-isomer
(
via infra). To address whether carboxylic acid was gener-
ated through hydrolysis of the ester, one isomerization reac-
tion was speciꢀcally conducted under the optimal conditions
0
2
4
6
8
10
Time (h)
(
Table 1, entry 10) by extending the reaction time to 24 h,
Fig. 1 The time course of Pd(II)-catalyzed transformation of methyl
and the mixture was dissolved in CDCl after removal of the
3
1
3
linoleate to its conjugated derivatives in the presence or absence of
toluene/t-BuOH solvent for C NMR analysis. As shown
in Fig. S7, the chemical shift at 174.28 ppm, corresponding
to the carbonyl carbon, is diꢁerent from the chemical shifts
of the carboxyl carbon in lionleic acid and its conjugated
products (around 180 ppm) [37, 38], but can be assigned to
the carboxyl carbon in esters [39, 40]. Clearly, the hydrolysis
does not happen under current isomerization conditions. In
Sc(OTf) . Conditions: 3 mL of Toluene/t-BuOH (2.75:0.25, v/v),
3
Pd(OAc) (3 mM), Sc(OTf) (6 mM), methyl linoleate (60 mM),
2
3
110 °C, under air. The conversion and yield were determined as the
consumption of ML and the formation of all the CML products,
respectively, as determined by GC relative to the initial ML added
Sc(OTf) on the kinetics of ML isomerization by Pd(OAc)
3
2
3
the control experiment, using Sc(OTf) as catalyst, oꢁered
catalyst, and it clearly demonstrates that adding Sc(OTf)
3
8
7.0 ± 3.6% conversion of ML. However, no CML prod-
signiꢀcantly acclerates Pd(II)-catalyzed ML isomerization.
Similar to isomerization of allylbenzene in previous stud-
ies [34], a stronger Lewis acid also oꢁered a better promo-
tional eꢁect for ML isomerization in the present studies. In
ucts were detected in GC analysis, and the isolated product
1
(
38.6 ± 3.6% yield) was identiꢀed as a saturated ester by H
NMR analysis (vide infra). Figure 1 displays the inꢃuence of
1
3

J Am Oil Chem Soc
addition to the super-promotional eꢁect of Sc(OTf) , adding
is no chemical shifts observed at 5.2–6.4 ppm corresponding
to the vinylic hydrogens of the conjugated ester, thus exclud-
ing the formation of the CML products (Fig. 2c). Notably,
the disappearance of non-conjugated vinylic hydrogens at
5.2–5.4 ppm suggested the addition reaction occurring on
3
Cu(OTf) also disclosed a remarkable promotional eꢁect in
2
ML isomerization, oꢁering 48.9 ± 2.4% yield of the CML
products, while neither Pd(OAc) nor Cu(OTf) alone is
2
2
inactive for isomerization. However, Cu(OTf) alone oꢁered
2
3
3.8 ± 4.1% isolated yield of the saturated ester as well as
this product. Clearly, the roles of Sc(OTf) in ML trans-
3
does Sc(OTf) . It is worth emphasizing that, unlike Wacker-
formation are distinctly diꢁerent between the presence and
3
type oxidations and related C–H bond functionalizations by
a Pd(II)/Cu(II) catalyst, in which the Cu(II) cation is long
believed to serve as an oxidant to the reduced Pd(0), here,
the redox properties of the Cu(II) are not essential for Pd(II)-
catalyzed ML isomerization. Therefore, this work further
exhibited the signiꢀcant role of Lewis acid properties, rather
than the redox properties, of the Cu(II) cation in promot-
ing Pd(II)-catalyzed oleꢀn isomerization. Remarkably, the
isomerizations were conducted under air in the present stud-
ies, whereas previously reported oleꢀn isomerizations were
absence of the Pd(OAc) catalyst. One is to promote Pd(II)-
2
catalyzed isomerization of ML to the CML products, the
other is to transform ML to the saturated ester product by
addition reaction.
In the literature, it has been reported that isomerization
of ML can generate a list of conjugated isomerization prod-
ucts including (9Z,11E)-CLA, (10E,12Z)-CLA, (9Z,11Z)-
CLA and (9E,11E)-CLA, etc., in which (9Z,11E)-CLA
and (10E,12Z)-CLA showed signiꢀcant biological activi-
ties (CLA conjugated linoleate acid) [23]. Here, through
GC–MS analysis of the isolated CML products, it was
found that, in the major peaks, the ratio of (9E,11E)-CML,
(10E,12E)-CML, (9Z,11E)-CML and (10E,12Z)-CML is
33.3:33.5:6.6:26.6% (Scheme 1), while a few minor peaks
have not yet been identiꢀed by the GC–MS database. In
addition, one minor peak was identified as 9-octadece-
noic acid (Z)-methyl ester (methyl oleate), which conꢀrms
that the addition reaction also happens in the presence of
generally conducted under N or Ar [21–23], which high-
2
lights the merit of the catalyst system demonstrated here.
1
In H NMR characterizations of ML substrate and CML
products, the chemical shift observed at 2.7–2.8 ppm is
known as the methylene protons positioned between the
two C=C bonds in ML as shown in Fig. 2a. Disappearance
of this chemical shift with the occurrence of new chemical
shifts at 5.2–6.4 ppm, corresponding to the vinylic hydro-
gens of the conjugated ester, indicated the isomerization
of ML to the corresponding CML. Meanwhile, the chemi-
cal shift for non-conjugated vinylic hydrogens in ML at
Sc(OTf) , even though it is not the major reaction in the
3
presence of the Pd(OAc) catalyst.
2
Two distinct mechanisms have been proposed in the lit-
erature for Pd(II)-catalyzed oleꢀn isomerization [41–44].
One proceeds by [1,2]-hydrogen shift mechanism in which
the formation of the Pd(II)-hydride moiety is crucial to ini-
tiate a [1,2]-hydrogen shift. In this mechanism, a proton
source is generally necessary, at least, as co-solvent. The
5
.2–5.4 ppm also disappeared simultaneously (Fig. 2b). In
the case of using Sc(OTf) alone as a catalyst which pro-
3
1
vided 87.0 ± 3.6% conversion of ML, although the H NMR
spectrum of the isolated products indicated the disappear-
ance of the methylene protons at 2.7–2.8 ppm for ML, there
Fig. 2 1H NMR spectra of
methyl linoleate substrate with
its conjugated and saturated
(a) Methyl linoleate
products in CDCl . a ML
3
substrate, b CML products, and
c the isolated product using
Sc(OTf) alone as catalyst
3
(
b) Mixture of conjugated methyl
linoleate (CML) products
(
c) Methyl stearate (saturated ester)
8
.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0
f1 (
ppm
)
1
3

J Am Oil Chem Soc
Scheme 1 The major products
identiꢀed by GC–MS analysis
from methyl linoleate isomeri-
zation with the Pd(OAc) /
2
Sc(OTf) catalyst
3
Table 2 The inꢃuence of solvent on the Pd(II)-catalyzed methyl
oꢁered 55.4 ± 1.4% conversion of ML with 35.5 ± 2.2%
yield of ester exchange product, that is, ethyl linoleate, with
only 5.2 ± 1.2% of CML products. Even in the case of using
methanol as the solvent which can avoid ester exchange
occurring, it provided only 33.6 ± 2.2% conversion of ML
with 23.6 ± 1.8% yield of CML, similar to using ethylben-
zene as solvent (25.4 ± 2.2% yield), but still less than that
using dry toluene as solvent (38.0 ± 3.6% yield). In particu-
lar, using t-BuOH alone as solvent also oꢁered 57.8 ± 1.8%
conversion of ML; however, no CML products were gen-
erated. In the case of using toluene/ethanol as solvent, it
oꢁered 54.4 ± 2.2% conversion with only 12.4 ± 1.4% yield
of CML, and 33.9 ± 3.1% yield of ester exchange product
was obtained, that is, ethyl linoleate. Remarkably, the com-
bination of toluene with t-BuOH (2.75:0.25, v/v) as solvent
provided complete conversion of ML with 96.3 ± 2.4% yield
of the CML products. It is worth mentioning that, although
the Pd(OAc) /Sc(OTf) catalyst exhibited roughly similar
linoleate transformation to its conjugated derivatives
Entrya Solvent
Conv. (%)b Total yield
Selectivity (%)d
of CML
c
(
%)
1
2
3
4
5
6
7
8
Ethylbenzene
29.5 ± 2.8 25.4 ± 3.2 86.1
46.4 ± 1.1 38.0 ± 3.6 81.9
Toluene
Ethanol
Methanol
t-BuOH
55.4 ± 1.4 5.2 ± 1.2
33.6 ± 2.2 23.6 ± 1.8 70.2
57.8 ± 1.8 Trace
9.3
–
Toluene/ethanol 54.4 ± 2.2 12.4 ± 1.4 22.8
Toluene/t-BuOH 99.6 ± 0.4 96.3 ± 2.4 96.7
Toluene/metha- 41.0 ± 3.8 31.4 ± 1.3 76.4
nol
9
1
Toluene/isopro- 20.6 ± 4.6 16.3 ± 2.1 78.9
panol
0
Ethylbenzene/t- 38.2 ± 1.2 33.2 ± 1.3 87.0
BuOH
2
3
a
selectivity in transforming ML to CML in several solvents
(Table 2, entries 1, 2 and 7–10), the toluene/t-BuOH solvent
oꢁers the highest catalytic eꢂciency for this transformation.
The inꢃuence of solvent composition between toluene
and t-BuOH was next investigated in detail and the results
are summarized in Table 3. As stated above, using t-BuOH
alone as solvent oꢁered 57.8 ± 1.8% conversion of ML but
no CML formation was observed. In a total volume of 3 mL
of solvent, adding toluene to t-BuOH as solvent mixture con-
tinuously improved the yield of CML up to 96.3 ± 2.4% at
the ratio of 2.75:0.25 (v/v), while using toluene alone as sol-
vent caused a sharp decrease of catalytic eꢂciency, oꢁering
only 38.0 ± 3.6% yield of CLM with only 81.9% selectivity.
Clearly, the presence of the large amount of t-BuOH did
not improve the catalytic eꢂciency and selectivity, while
using t-BuOH alone as solvent did not oꢁer the expected
CML products. On the other hand, using dry toluene alone
as solvent does provide CML as major products, even though
the catalytic eꢂciency is relatively poor. In experimental
Conditions: solvent (3 mL), Pd(OAc) (3 mM), Sc(OTf) (6 mM),
2
3
methyl linoleate (60 mM), 110 °C except entries 3–5 reꢃuxing at
9
2
b
0 °C, 9 h, under air. Entries 6–10, the solvent mixture with ratio of
.75:0.25 (v/v) in which alcohol is minor
Conversion was determined as the consumption of ML as deter-
mined by GC relative to the initial ML added
c
The total yield was determined as the formation of all the CML
products as determined by GC relative to the initial ML added
d
The selectivity of the total CML was obtained by dividing the aver-
age total yield of CML by the average ML conversion
other proceeds by [1,3]-hydrogen shift mechanism through
the formal Pd(II)/Pd(IV) cycle, in which the formation of the
allyl-Pd(II) intermediate is the key step for isomerization. In
the present studies, the inꢃuence of solvent composition on
catalytic eꢂciency was investigated to address the role of
alcohol in the solvent mixture. As shown in Table 2, using
either dry ethylbenzene or toluene as solvent, the Pd(OAc) /
2
Sc(OTf) catalyst provided 29.5 ± 2.8 or 46.4 ± 1.1% con-
3
version of ML with 25.4 ± 3.2 or 38.0 ± 3.6% yield of CML
products, respectively. Although toluene serves as a more
eꢂcient solvent than ethylbenzene for isomerization, their
selectivity to CML is similar, that is, 86.1% for ethylbenzene
and 81.9% for toluene, respectively. Using ethanol as solvent
preparation, it was found that Pd(OAc) does not dissolve
2
in pure t-BuOH at room temperature but dissolves in tolu-
ene, while Sc(OTf) does not dissolve in toluene but dis-
3
solves in t-BuOH. Taking all this information together, it
can be concluded that t-BuOH does not serve as the proton
1
3

J Am Oil Chem Soc
Table 3 The inꢃuence of solvent composition on isomerization of
methyl linoleate by Pd(OAc) /Sc(OTf) catalyst
2
3
6
4
0
0
Entrya Toluene/t-BuOH Conv. (%)b Total yield Selectivity (%)d
of CML
c
(
%)
Con%
Total Yield%
1
2
3
4
5
6
7
8
0:3
0.5:2.5
1:2
1.5:1.5
2:1
2.5:0.5
2.75:0.25
3:0
57.8 ± 1.8 trace
–
59.4 ± 3.0 32.6 ± 2.2
61.2 ± 2.9 42.0 ± 1.6
74.1 ± 2.7 67.2 ± 2.0
84.3 ± 2.3 75.4 ± 1.9
87.2 ± 3.0 77.4 ± 1.4
99.6 ± 0.4 96.3 ± 2.4
46.4 ± 1.1 38.0 ± 3.6
54.8
68.5
90.8
90.9
88.9
96.7
82.0
20
0
0.0
0.5
1.0
1.5
2.0
The ratio of Sc(lll)/Pd(ll)
a
Fig. 3 The inꢃuence of the Sc(III)/Pd(II) ratio on the Pd(II) cata-
lyzed methyl linoleate isomerization. Conditions: 3 mL of toluene/t-
BuOH (2.75:0.25, v/v), Pd(OAc)2 (3 mM), methyl linoleate (90 mM),
Conditions: solvent (3 mL), Pd(OAc) (3 mM), Sc(OTf) (6 mM),
2
3
methyl linoleate (60 mM), 110 °C, 9 h, under air
b
Conversion was determined as the consumption of ML as deter-
1
10 °C, 9 h, under air. The conversion and yield were determined as
mined by GC relative to the initial ML added
the consumption of ML and the formation of all the CML products,
respectively, as determined by GC relative to the initial ML added
c
The total yield was determined as the formation of all the CML
products as determined by GC relative to the initial ML added
d
The selectivity of the total CML was obtained by dividing the aver-
age total yield of CML by the average ML conversion
step in the catalytic cycle, thus extra amounts of Sc(OTf)
3
are not essential for the Pd(OAc) catalyst.
2
In previous studies, it has been observed by UV–Vis
source to participate in the isomerization reaction through a
spectra that adding non-redox metal salts like Sc(OTf) or
3
[
1,2]-hydrogen shift mechanism, in which the formation of
Al(OTf) to Pd(OAc) in acetonitrile would generate the new
3
2
the Pd(II)-hydride moiety is crucial for isomerization [37,
9]. Here, the toluene/t-BuOH solvent mixture facilitates
the complete dissolution of both Pd(OAc) and Sc(OTf)
Pd(II) species, and it was further evidenced by NMR stud-
ies [33–35]. Taken together with the facts that heterobime-
tallic Pd(II) dimers having an acetate bridge with bivalent
3
2
3
2
+
2+
2+
2+
in one system. It is worth mentioning that, in either tolu-
ene or t-BuOH solvent, much palladium black formation
was observed after the reaction, while in the solvent ratio
of 2.75:0.25 (v/v), the reaction medium remains clean after
the reaction, which also highlights the merit of the mixture
solvent.
or trivalent metal ions, including Ba , Sr , Ca , Mn ,
2
+
2+
2+
2+
3+
3+
3+
4+
Co , Ni , Cd , Zn Ce , Nd , Eu and Ce , have
been widely reported [45–48], we proposed the formation
of the heterobimetallic Pd(II)/Sc(III) or Pd(II)/Al(III) dimer
having an acetate bridge as the key species for Wacker-type
oxidation, oleꢀn isomerization and oxidative coupling reac-
tions [33–35]. In the present studies, the UV–Vis kinetics
of Pd(OAc) /Sc(OTf) in toluene/t-BuOH solvent also indi-
The influence of the Sc(III)/Pd(II) ratio on methyl
linoleate isomerization is displayed in Fig. 3. It was found
that, with the ratio increased from 0 to 1, the catalytic eꢂ-
ciency increased quickly. However, when the ratio continu-
ously increased from 1 to 2, the eꢂciency improvement is
relatively minor. This phenomenon is similar to those in
allylbenzene isomerization, but distinctly diꢁerent from
2
3
cated the formation of the new Pd(II) species. As shown in
Fig. 4, adding Sc(OTf) to the toluene/t-BuOH solution of
3
Pd(OAc) at room temperature caused a gradual change of
2
the characteristic absorbance of Pd(OAc) around 410 nm.
2
Since there is no absorbance change beyond 600 nm with no
that in Wacker-type oxidations catalyzed by Pd(OAc) /Lewis
palladium black observed during this period, the absorbance
2
acid [33–35]. In Wacker-type oxidation, the Sc(III)/Pd(II)
ratio of 4 is much more eꢂcient than the ratio of 1. In that
reaction, the roles of the Sc(III) cation were assigned to two
aspects: one is to generate the heterobimetallic Pd(II)/Sc(III)
dimer with an acetate bridge, which is the key active species
for oxidation, while the other is to generate a Sc(III)···H-
Pd(II) intermediate which facilitates dioxygen insertion
to form the HOO-Pd(II) intermediate, thus preventing the
reductive elimination of the Pd(II)-hydride moiety to gener-
ate the inactive palladium black. Here, in ML isomerization,
dioxygen insertion of the H-Pd(II) bond is not an elementary
change below 600 nm by adding Sc(OTf) also implicated
3
the formation of new Pd(II) species as well as those in ace-
tonitrile, although their UV–Vis spectra are distinctly dif-
ferent, possibly because of diꢁerent solvent media. Accord-
ingly, similar heterobimetallic Pd(II)/Sc(III) dimera having
an acetate bridge can be proposed in this Pd(OAc) /Sc(OTf)
2
3
system in the toluene/t-BuOH solvent mixture, which is
responsible for ML isomerization such as that for allylben-
zene isomerization in acetonitrile [34].
Based on previous studies together with the
UV–Vis spectra changes demonstrated here, a simpliꢀed
1
3

J Am Oil Chem Soc
Then, reductive elimination of this intermediate releases the
9,11 or 10,12-conjugated methyl linoleate products, includ-
ing four major products as (9E,11E)-CML, (10E,12E)-
CML, (9Z,11E)-CML and (10E,12Z)-CML as disclosed
by GC–MS analysis. As shown in Table 1, in the absence
of Sc(OTf) , Pd(OAc) alone as catalyst is inactive for
0
0
0
0
0
0
.5
.4
.3
.2
.1
.0
Pd+Sc
Pd+Sc 5 min
Pd+Sc 10 min
Pd+Sc 15 min
Pd+Sc 20 min
Pd+Sc 25 min
Pd+Sc 30 min
Pd+Sc 40 min
Pd+Sc 50 min
Pd+Sc 60 min
Pd+Sc 90 min
Pd+Sc 180 min
3
2
ML isomerization, indicating that the Pd(II) species from
Pd(OAc) in toluene/t-BuOH solvent is incapable of acti-
2
vating the C–H bond of the methylene group positioned
between two C=C bonds in the ML substrate. While adding
Sc(OTf) to Pd(OAc) generates a new Pd(II) species, that
3
2
is, the proposed heterobimetallic Pd(II)/Sc(III) dimer which
is more powerful in C–H bond activation, leading to the
eꢂcient isomerization of ML to the CML products. Similar
improved isomerization capability of the Pd(II) species was
4
00
600
800
1000
Wavelength (nm)
also observed in allylbenzene isomerization by Pd(OAc) /
2
Al(OTf) catalyst, and supported by related DFT calcula-
3
Fig. 4 The UV–Vis kinetics of Pd(OAc) /Sc(OTf) in 3 mL of
2
3
tions [34, 35].
toluene/t-BuOH (2.75:0.25, v/v) at room temperature
ML isomerization mechanism was proposed as shown in
Scheme 2. Regarding the facts that (1) t-BuOH alone as
solvent does not oꢁer the CML products, (2) dry toluene
alone as solvent oꢁers CML as the major products, even
though the catalytic eꢂciency is relatively low, and (3) the
lower content of t-BuOH in toluene/t-BuOH solvent mixture,
that is, the ratio of 2.75:0.25 (v/v), oꢁers the better catalytic
eꢂciency for isomerization, it can be rationally proposed
that ML isomerization proceeds by [1,3]-hydrogen shift
mechanism in which a formal Pd(II)/Pd(IV) catalytic cycle
proceeds [37–40]. As shown in Scheme 2, after coordina-
tion of the C=C bond either from the 9-position (left side in
Scheme 2) or the 12-position (right side in Scheme 2), the
Pd(II)/Sc(III) dimeric species next activates the methylene
protons which have the chemical shift at 2.7–2.8 ppm, to
generate a formal allyl-Pd(IV)-hydride/Sc(III) intermediate.
Conclusion
This work disclosed that adding non-redox metal salts like
Sc(OTf) to a simple Pd(OAc) catalyst can effectively
3
2
improve its catalytic eꢂciency for methyl linoleate isomer-
ization to conjugated methyl linoleate in toluene/t-BuOH
solvent, while Pd(OAc) alone is inactive. The formation of
2
the heterobimetallic Pd(II)/Sc(III) dimer having an acetate
bridge was proposed as the key active species for isomeriza-
tion. The reaction is proposed to proceed by a [1,3]-hydro-
gen shift mechanism in which the formal Pd(II)/Pd(IV) cycle
was involved in the catalytic cycle. Notably, this non-redox
metal ion-promoted methyl linoleate isomerization was
conducted under air, which highlights its merits when com-
pared with those Rh- and Ru-based isomerizations under
N or Ar. Thus, this Pd(II)/Lewis acid catalyst may oꢁer a
2
Scheme 2 Proposed mechanism for transformation of methyl linoleate to its conjugate isomers by the Pd(OAc) /Sc(OTf) catalyst
2
3
1
3

J Am Oil Chem Soc
new opportunity for transforming vegetable oils to the cor-
responding conjugated isomers in their industrial utilizations
through catalytic isomerization.
17. Westfechtel A, Albiez W, Zander L, Horlacher P (2006) Method
for the production of conjugated linoleic acid. US20060106238
A1
1
8. Saebo A, Skarie C, Jerome D, Haroldsson G (2002) Conju-
gated linoleic acid compositions and methods of making same.
US6410761 B1
Acknowledgement This work was supported by the National Natu-
ral Science Foundation of China (no. 21273086 and 21573082). The
GC–MS analyses were performed in the Analytical and Testing Center
of Huazhong University of Science and Technology.
19. Philippaerts A, Goossens S, Jacobs PA, Sels BF (2011) Catalytic
production of conjugated fatty acids and oils. Chem SusChem
4:684–702
2
0. Krompiec S, Jerzy J, Majewski J, Grobelny J (1998) Isomeriza-
tion of vegetable oils catalyzed by Ruthenium complexes. Pol J
Appl Chem 42:43–48
2
1. Larock RC, Dong X, Chung S, Reddy CK, Ehlers LE (2001)
Preparation of conjugated soybean oil and other natural oils and
fatty acids by homogeneous transition metal catalysis. J Am Oil
Chem Soc 78:447–453
References
1
.
Raꢀq M, Lv YZ, Zhou Y, Ma KB, Wang W, Li CR, Wang Q
(
2015) Use of vegetable oils as transformeroils–a review. Renew
22. Behr A, Witte H, Bayrak Z (2013) Homogeneous metal com-
plex catalyzed conjugation of methyl linoleate. Eur J Lipid Sci
Technol 115:721–728
Sust Energ Rev 52:308–324
2
.
Melo-Espinosa EA, Piloto-Rodríguez R, Goyos-Pérez L, Sierens
R, Verhelst S (2015) Emulsiꢀcation of animal fats and vegetable
oils for their use as a diesel engine fuel, an overview. Renew Sust
Energ Rev 47:623–633
23. Andjelkovic DD, Min B, Ahn D, Larock RC (2006) Elucida-
tion of structural isomers from the homogeneous Rhodium-
catalyzed isomerization of vegetable oils. J Agric Food Chem
54:9535–9543
3
4
.
.
Pꢀster DP, Xia Y, Larock RC (2011) Recent advances in vegetable
oil-based polyurethanes. Chemsuschem 4:703–717
Xia Y, Larock RC (2010) Vegetable oil-based polymeric mate-
rials synthesis, properties, and applications. Green Chem
24. Quirino RL, Larock RC (2012) Rh-based biphasic isomerization
of carbon–carbon double bonds in natural oils. J Am Oil Chem
Soc 89:1113–1124
1
2:1893–1909
25. Consorti CS, Aydos GL, Ebeling G, Dupont J (2009) Multiphase
catalytic isomerisation of linoleic acid by transition metal com-
plexes in ionic liquids. Appl Catal A 371:114–120
5
6
7
.
.
.
Sivasamy A, Cheah KY, Fornasiero P, Kemausuor F, Zinoviev
S, Miertus S (2009) Catalytic applications in the production of
biodiesel from vegetable oils. Chemsuschem 2:278–300
Lu Y, Larock RC (2009) Novel polymeric materials from vegeta-
ble oils and vinyl monomers: preparation, properties, applications.
Chemsuschem 2:136–147
26. Pertici P, Ballantini V, Catalano S, Giuntoli A, Malanga C, Vitulli
G (1999) (6-Naphthalene)(4-cycloocta-1, 5-diene) ruthenium (0)
as eꢂcient catalytic precursor for the isomerization of methyl
linoleate under mild conditions. J Mol Catal A 144:7–13
27. Dong L, Wang Y, Lv Y, Chen Z, Mei F, Xiong H, Yin G (2013)
Lewis-acid-promoted stoichiometric and catalytic oxidations by
manganese complexes having cross-bridged cyclam ligand: a
comprehensive study. Inorg Chem 52:5418–5427
Lima GES, Nunes EV, Dantas RC, Meneghetti MR, Meneghetti
SMP (2015) Systematic investigation of the oxidative polym-
erization of linseed oil catalyzed by Co(II), Mn(II), and Fe(II)
complexes with chelating nitrogen ligands. J Eur Lipid Sci Tech
1
17:229–234
28. Zhang Z, Coats KL, Chen Z, Hubin TJ, Yin G (2014) Inꢃuence of
calcium (II) and chloride on the oxidative reactivity of a manga-
nese (II) complex of a cross-bridged cyclen ligand. Inorg Chem
53:11937–11947
8
9
.
.
Philippaerts A, Joost VA, Bert S (2013) Conjugated linoleic acids
and conjugated vegetable oils: from nutraceutical to bio-polymer.
J Eur Lipid Sci Tech 115:717–720
Yavari A, Hamedi M, Heshmati A, Haghbin S (2010) Retrac-
tion Are (CLA) isomers good or bad trans fats. Lipid Technol
29. Chen Z, Yang L, Choe C, Lv Z, Yin G (2015) Non-redox metal
ion promoted oxygen transfer by a non-heme manganese catalyst.
Chem Commun 51:1874–1877
2
2:227–229
1
1
1
0. Lu Y, Larock RC (2010) Novel biobased plastics, rubbers, com-
posites, coatings and adhesives from agricultural oils and by-
products. ACS Sym Ser 1043:87–102
30. Zhang J, Wang Y, Luo N, Chen Z, Wu K, Yin G (2015) Redox
inactive metal ion triggered N-dealkylation by an iron catalyst
with dioxygen activation: a lesson from lipoxygenases. Dalton
Trans 44:9847–9859
1. Meier MA, Metzger JO, Schubert US (2007) Plant oil renewable
resources as green alternatives in polymer science. Chem Soc Rev
31. Lv Z, Zheng W, Chen Z, Tang Z, Mo W, Yin G (2016) Synergistic
oxygen atom transfer by the Ruthenium complexes with non-redox
metal ions. Dalton Trans 45:11369–11383
3
6:1788–1802
2. Wang CS, Yang LT, Ni BL, Shi G (2009) Polyurethane networks
from diꢁerent soy-based polyols by the ring opening of epoxidized
soybean oil with methanol, glycol, and 1, 2-propanediol. J Appl
Polym Sci 114:125–131
32. Guo H, Chen Z, Mei F, Zhu D, Xiong H, Yin G (2013) Redox
inactive metal ion promoted C–H activation of benzene to phenol
with Pd(bpym): Demonstrating new strategies in catalyst designs.
Chemistry 8:888–891
1
1
3. Andjelkovic DD, Valverde M, Henna P, Li F, Larock RC (2005)
Novel thermosets prepared by cationic copolymerization of vari-
ous vegetable oils—synthesis and their structure–property rela-
tionships. Polymer 46:9674–9685
33. Qin S, Dong L, Chen Z, Zhang S, Yin G (2015) Non-redox
metal ions can promote Wacker-type oxidations even better than
copper (II), a new opportunity in catalyst design. Dalton Trans
44:17508–17515
4. Ip C, Chin SF, Scimeca JA, Pariza MW (1991) Mammary cancer
prevention by conjugated dienoic derivative of linoleic acid. Can-
cer Res 51:6118–6124
34. Senan AM, Qin S, Zhang S, Lou C, Chen Z, Liao RZ, Yin G
(2016) Non-redox metal-ion-accelerated oleꢀn isomerization by
palladium(II) catalysts: density functional theory (DFT) calcula-
tions supporting the experimental data. ACS Catal 6:4144–4148
35. Zhang SC, Qin S, Lou C, Senan AM, Liao RZ, Yin G (2016)
Non-redox metal ion promoted oxidative coupling of indoles with
oleꢀns by the palladium (II) acetate catalyst through dioxygen
1
1
5. Whigham LD, Cook ME, Atkinson RL (2000) Conjugated linoleic
acid, implications for human health. Pharm Res 42:503–510
6. Rickert R, Sébédio JL, Steinhart H (2001) Conjugated linoleic
acid (CLA), physiological eꢁects and production. J Eur lipid Sci
Rech 103:56–61
1
3

J Am Oil Chem Soc
3
activation: experimental results with DFT calculations. Org Bio-
mol Chem 14:4146–4157
pincer catalysts. Experimental evidence for an η -allyl pathway
and an unconventional mechanism predicted by DFT calculations.
J Am Chem Soc 134:13276–13295
3
3
3
3
4
4
4
4
6. Sharma SK, Srivastava VK, Pandya PH, Jasra RV (2005) Solvent-
free isomerization of methyl chavicol to trans-anethole using tran-
sition metal complexes as catalysts. Catal Commun 6:205–209
7. Marwah RG, Fatope MO, Deadman ML, Al-Maqbali YM, Hus-
band J (2007) Musanahol: a new aureonitol-related metabolite
from a Chaetomium sp. Tetrahedron 63:8174–8180
44. Knapp SMM, Shaner SE, Kim D, Shopov DY, Tendler JA,
Pudalov DM, Chianese AR (2014) Mechanistic studies of alkene
isomerization catalyzed by CCC-pincer complexes of iridium.
Organometallics 33:473–484
45. Kozitsyna NY, Nefedov SE, Dolgushin FM, Cherkashina NV,
Vargaftik MN, Moiseev II (2006) Heterodimetallic PdII- based
carboxylate-bridged complexes: synthesis and structure of single-
crystalline Pd II–M (M=Mn II, Co II, Ni II, Cu II, Zn II, Nd III,
Eu III, Ce IV) acetates. Inorg Chim Acta 359:2072–2086
46. Akhmadullina NS, Cherkashina NV, Kozitsyna NY, Gekhman AE,
Vargaftik MN (2009) Unexpected participation of nucleophiles
in the reaction of palladium (II) acetate with divalent 3d metals.
Kinet Catal 50:396–400
8. Duꢁy PE, Quinn SM, Roche HM, Evans P (2006) Synthesis of
trans-vaccenic acid and cis-9-trans-11-conjugated linoleic acid.
Tetrahedron 62:4838–4843
9. Biswas A, Sharma BK, Willett JL, Erhan SZ, Cheng HN (2008)
Room-temperature self-curing ene reactions involving soybean
oil. Green Chem 10:290–295
1
3
0. Bus J, Sies I, Lie Ken Jie MSF (1976) C-NMR of methyl,
methylene and carbonyl carbon atoms of methyl alkenoates and
alkynoates. Chem Phys Lipids 17:501–518
47. Pandey RN, Henry PM (1974) Interaction of palladium (II) ace-
tate with sodium and lithium acetate in acetic acid. Can J Chem
52:1241–1247
1. Sen A, Lai TW (1981) Catalytic isomerization of alkenes by pal-
ladium (II) compounds. An alternative mechanistic view. Inorg
Chem 20:4036–4038
48. Nefedov SE, Perova EV, Yakushev IA, Kozitsyna NY, Vargaftik
MN, Moiseev II (2009) Competition between 3d metals (II) and
palladium (II) in the reaction of heterobimetallic complexes Pd
(μ-OOCMe)4M (OH2)(M=Ni Co, Mn) with azobenzene. Inorg
Chem Commun 12:454–456
2. Cramer R (1966) Oleꢀn coordination compounds of rhodium.
III. The mechanism of oleꢀn isomerization. J Am Chem Soc
8
8:2272–2282
3. Biswas S, Huang Z, Choliy Y, Wang DY, Brookhart M, Krogh-
Jespersen K, Goldman AS (2012) Oleꢀn isomerization by iridium
1
3