Synergetic effect of ruthenium and basicity sites in the Ru-MgAl catalyst for hydrogen-free production of conjugated linoleic acids

Article abstract of DOI:10.1039/c5ra00417a

A series of Ru-MgAl composite oxide catalysts prepared by calcining the ruthenium grafted hydrotalcite-like precursor at various temperatures were used in the hydrogen-free production of conjugated linoleic acid. The effect of calcination temperature on the textural, base and catalytic properties of the materials was investigated. Results indicated that the Ru-MgAl composite oxides calcined at 450 °C showed high activity, namely, CLA productivity, CLA production rate and TOF up to 1.52 g CLA g (CLA) L^-1 (solvent) min^-1, 284 g (CLA) g^-1 (Ru) h^-1 and 102.6 mol (LA converted) mol^-1 (Ru) h^-1. Moreover, the biologically active CLA isomers, cis-9, trans-11, trans-10, cis-12 and trans-9, trans-11-CLA, were the main products, while almost no hydrogenated products were formed. Meanwhile, the role of ruthenium and basicity sites in the catalytic reaction has been studied. It was found that the basicity sites of the Ru-MgAl catalyst and the ruthenium activity sites seem to have a synergic effect on the catalytic reaction. The possible reaction mechanism for the isomerization was also proposed. This journal is

Full text of DOI:10.1039/c5ra00417a

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Synergetic effect of ruthenium and basicity sites in
the Ru–MgAl catalyst for hydrogen-free
production of conjugated linoleic acids†
Cite this: RSC Adv., 2015, 5, 20248
ab
b
ac
b
Jiebo Chen, Xinxiang Chen, Ying Zheng* and Qinglu Li
A series of Ru–MgAl composite oxide catalysts prepared by calcining the ruthenium grafted hydrotalcite-
like precursor at various temperatures were used in the hydrogen-free production of conjugated linoleic
acid. The effect of calcination temperature on the textural, base and catalytic properties of the materials
ꢀ
was investigated. Results indicated that the Ru–MgAl composite oxides calcined at 450 C showed high
ꢁ
1
activity, namely, CLA productivity, CLA production rate and TOF up to 1.52 g CLA g (CLA) L (solvent)
ꢁ
1
ꢁ1
ꢁ1
ꢁ1
ꢁ1
min , 284 g (CLA) g (Ru) h and 102.6 mol (LA converted) mol (Ru) h . Moreover, the biologically
active CLA isomers, cis-9, trans-11, trans-10, cis-12 and trans-9, trans-11-CLA, were the main products,
while almost no hydrogenated products were formed. Meanwhile, the role of ruthenium and basicity
sites in the catalytic reaction has been studied. It was found that the basicity sites of the Ru–MgAl
catalyst and the ruthenium activity sites seem to have a synergic effect on the catalytic reaction. The
possible reaction mechanism for the isomerization was also proposed.
Received 8th January 2015
Accepted 12th February 2015
DOI: 10.1039/c5ra00417a
www.rsc.org/advances
Conventionally, CLAs can be prepared by microbial biosyn-
theses and isomerization of linoleic acid with alkaline as cata-
1
. Introduction
Conjugated linoleic acid (CLA) is a collective term used to lysts. Some microorganisms contain specic isomerase enzymes
describe the mixture of positional and geometric isomers of that are able to transform linoleic acid into CLAs. Unfortunately,
linoleic acid with conjugated double bonds. A vast amount of these bacterial species can transform the CLAs into saturated
literature describing the abundant health benets of CLAs is products such as stearic acid. Moreover, bacteria do not
1
available today. They decrease the body fat quantity and synthesize the isomer cis-9, trans-11-CLA until the linoleic acid
2
3
increase muscle, possess anti-inammatory, cancer-preventive concentration is high, producing the inhibition of bio-
4
5
10
effects, exert benecial effects on the skeletal system, act as hydrogenation. Henceforth, these bacterial species cannot be
immunostimulants and decrease the probability of asthma utilized for the industrial production of CLAs. For basic catalysis
occurrence. However, several studies have shown that only a method, the amount of the two formed isomers (cis-9, trans-11
6
7
small number of CLA isomers, namely, cis-9, trans-11-CLA and and trans-10, cis-12-CLA) is almost equal (selectivity z 50%) and
7
trans-10, cis-12-CLA present interesting biomedical and nutri- the yield is quite high (over 80%). However, the use of the strong
tional properties, thus demonstrating their potential application basic potassium hydroxide or sodium methoxide is not envi-
as functional foods. Specically, the cis-9, trans-11-CLA isomer ronmentally friendly. Furthermore, CLA in its chemical form of a
(
c9, t11-CLA) has been proposed as potential anti-tumour agent free fatty acid during alkaline isomerization is easily oxidized in
11
whereas the trans-10, cis-12-CLA isomer (t10, c12-CLA) affects the air; other homogeneous catalysts, such as RhCl(pph
3 3
) ,
8
regulation of lipids (cholesterol) and body mass. Meanwhile, [RhCl(C
H ) ]
8 14 2 2
have been tested for the isomerization of linoleic
recent research indicates that the trans-9, trans-11 (t9, t11-CLA) acid, it does not cause hydrolysis of the nal product and it
9
isomer also promotes good health. Hence these three fatty acids yields highly conjugated products. But the process is not envi-
12
are of great interest to food and health research.
ronmentally friendly and the catalyst is difficult to separate.
Instead, a heterogeneous catalyst would be easy to lter and
reuse, and environmental problem and separation difficulties
a
College of Chemistry and Chemical Engineering, Fujian Normal University, Fuzhou,
13–16
Fujian 350007, China. E-mail: zyingth@sina.com; Fax: +86 59183750182; Tel: +86 can be avoided. Bernas et al.
did pioneering work on
5
9183750182
producing conjugated linoleic acid using heterogeneous cata-
lysts, it was showed that a great variety of supported-metal
catalysts were tested for isomerization of linoleic acid. Kreich
and Claus also reported the synthesis of CLAs over Ag/SiO2
catalyst. The catalytic results at 438 K yielded selectivity values
of 35% to c9, t11-CLA and 26% to t10, c12-CLA for a 69% of
b
College of Life Sciences, Fujian Agriculture and Forestry University, Fuzhou 350002,
China
c
College of Materials Science and Engineering, Fujian Normal University, Fuzhou
12
3
†
1
50007, China
Electronic supplementary information (ESI) available. See DOI:
0.1039/c5ra00417a
20248 | RSC Adv., 2015, 5, 20248–20255
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
conversion with a 12% of selectivity to hydrogenation products. under autogenous water vapor pressure. Aer 16 h, the
17
Cardo et al. used mordenite and ZSM-5 to yield CLA isomers by precipitate formed was ltered and washed thoroughly with
ꢁ
carbenium ions. However, it can be found that the hydrogen is hot distilled water until the ltrate was free from Cl ions as
needed for the majority of the heterogeneous catalytic process test by silver nitrate solution. The obtained lter cake was
ꢀ
to form the half-hydrogenated intermediates, which will nally dried in an oven at 80 C for 14 h. The solid material named as
transform into CLAs. That is to say, the elevated level of Ru–HT. For the comparison, the sample labeled as HT was
hydrogen will lead to formation of unwanted hydrogenated by- also synthesized as per the above-mentioned procedure
10
products. Though the production of CLAs by heterogeneous without adding RuCl
3 2 2 2
$xH O or MgCl $6H O. These samples
catalysts without hydrogen is a difficult and complicated were formed by calcination in air at different temperature for
ꢀ
ꢁ1
process, it can reduce the production of unwanted by-products. 4 h with a heating rate of 10 C min . The XRF analysis
18
Bernas et al. used Ru/C catalyst for the isomerization of lino- results showed that the Ru-content of the catalyst was
leic acid without the use of H . Philippaerts et al. described the 0.8 wt%, which was very similar to the ratio in the coprecipi-
19
2
2
H -free production of CLA using Ru loaded zeolite catalysts with tation solution. Consequently, the catalysts were named as
high CLA production rate. The main advantage of this catalyst is Ru–MgAlt, MgAlt or Ru–Alt, where t stands for calcination
that no hydrogen pretreatment or addition of hydrogen donors temperature.
is required and almost no hydrogenated products are formed.
However, apart from the above references, there are few publi-
cations concerning the use of heterogeneous catalysts for
2.3 Characterization of catalysts
production of CLAs without hydrogen. Hence, the design of a Powder X-ray diffraction (XRD) patterns of synthesized catalysts
reusable heterogeneous catalyst with high activity towards the were recorded on a Philips X'Pert MPD system equipped with
formation of the physiologically important c9, t11-, t10, c12- and XRK 900 reaction chamber, using Ni-ltered Cu Ka radiation
˚
ꢀ
t9, t11-CLA isomers without using hydrogen in mild atmo- (l ¼ 1.54056 A) over a 2q range of 5–80 . Operating voltage and
sphere, remains a challenge. In view of this, this paper presents current were kept at 40 kV and 40 mA, respectively. The iden-
a study of the use of Ru–MgAl composite oxides for production tication of crystalline phases was made by matching the JCPDS
of conjugated linoleic acids in the absence of hydrogen. While les. Fourier transform infra-red (FT-IR) spectra of synthesized
almost no hydrogenated by-products are formed, high produc- catalysts were recorded with a PerkinElmer Spectrum GX FT-IR
ꢁ1
tivity of and selectivity for benecial CLAs is achieved. The role spectrometer in the region of 400–4000 cm using KBr pellets.
of catalyst basicities and ruthenium for the isomerization was X-ray uorescence analyses (XRF) were carried out with the use
also discussed. As know as we could, there is not dates' were of an Axios mAX 4 kW, PANalytical. N adsorption–desorption
2
ꢀ
found for the utilization of Ru–MgAl composite oxides catalyst isotherm at ꢁ196 C was obtained using an TriStar 3000 model
for obtaining benecial CLAs.
of gas adsorption analyzer from Micromeritics, Inc. Prior to N
adsorption, the samples were previously out gassed at 300 C for
2
ꢀ
8
h. BET specic areas were calculated from these isotherms
2
. Experimental
using the BET method. Evolved gases analysis of the precursor
decomposition was performed by mass spectrometry
2.1 Materials
21
(EGA-MS). Samples (20 mg) were placed in a U-shaped quartz
Linoleic acid and methyl linoleate of 95% purity was purchased
from Aladdin Reagent Co. (China). Three standard CLA methyl
ester (CLAME) isomers [c9, t11; t10, c12 and t9, t11] and methyl
heptadecanoate were purchased from Nu-ChekPrep. INC and
reactor incorporated within a ow system, aer pretreating at
1
was heated to 800 C at 10 C min , under He ow of
50 mL min . Exhaust gases were inducted into the mass
spectrometer (Hiden HPR20), the mass fragments of OH
ꢀ
ꢁ1
00 C for 1 h under a He ow of 50 mL min , then the sample
ꢀ
ꢀ
ꢁ1
ꢁ
1
ꢀ
kept at ꢁ20 C. Various chemicals including MgCl $6H O,
2
2
AlCl $6H O, NaOH, Na CO , BF $2CH OH, RuCl , undecane
3
2
2
3
3
3
3
(
m/z ¼ 17) and CO
2
(m/z ¼ 44) were recorded as a function of the
were all purchased from Aladdin Reagent Co. (China). All
chemicals were used as purchased without further purication.
pyrolysis temperature. CO
2
-TPD studies of the samples were
performed using a TPR/TPD Micromeritics ASAP 2920 instru-
ment equipped with a thermal conductivity detector (TCD).
Samples (100 mg) were pretreated under a helium stream at 450
2.2 Catalyst synthesis
ꢀ
ꢀ
ꢁ1
ꢁ1
Ruthenium graed hydrotalcites were synthesized by the
C for 4 h (10 C min , 50 mL min ). Then, temperature was
2
0
ꢀ
ꢁ1
hydrothermal method. In a typical synthesis procedure, an decreased until 70 C, and a ow of pure CO
aqueous solution (A) containing MgCl $6H O (90 mmol), subsequently introduced into the reactor during 1 h. The CO
AlCl $6H O (20 mmol) and RuCl $xH O (0.83 mmol) in TPD was carried out between 70–750 C under a helium ow
00 mL double distilled deionized water was prepared. The (10 C min, 30 mL min ). H
2
(50 mL min ) was
2
2
2
-
ꢀ
3
2
3
2
ꢀ
ꢁ1
1
2
-TPR studies of the samples were
solution A was added dropwise into a second solution (B) performed using a TPR/TPD Micromeritics ASAP 2920 instru-
containing Na CO3 (130 mmol) in 100 mL double distilled ment equipped with a thermal conductivity detector (TCD).
2
deionized water, in around 45 min under room temperature. Before reduction, the sample (around 50 mg) was dried under
ꢀ
The mixture was maintained by adding NaOH solution to keep helium ow at 120 C for 4 h. TPR was carried out between room
ꢀ
ꢀ
ꢁ1
a constant pH of 10. Content was transferred into the Teon temperature and 750 C at a heating rate of 10 C min under
ꢀ
ꢁ1
coated stainless steel autoclave and aged at 80 C for 16 h 5% H
2
ow in argon (20 mL min total ow).
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 20248–20255 | 20249

View Article Online
RSC Advances
.4 Catalysis measurements
Paper
2
Isomerization of linoleic acid (>98% purity) was conducted in a
stirred batch reactor (25 mL) which was provided with a reux
condenser and a heating jacket. In a typical experiment, 10 mL
ꢁ
1
1
mol L linoleic acid solvent and 0.4 g catalyst were charged
all at once. A nitrogen ow was fed through the reactor before
and during the isomerization to have an inert atmosphere and
the reactor outlet was locked by a uid to prevent back-diffusion
of oxygen into the reactor. The system was stirred at 400 rpm in
the experiments to keep the catalyst uniformly dispersed in the
reaction medium, and also to eliminate external mass-transfer
ꢀ
problems. Catalytic tests were performed at 180 C for 2 h.
For reusability of the catalysts, spent catalyst was washed with n-
hexane to remove adsorbed reactant/products from the surface
ꢀ
of catalysts. Aer that the catalyst was dried for 10 h at 80 C and
reused for isomerization reaction.
Fig. 1 The XRD patterns of Ru–HT and HT.
2.5 Products analysis
The sample from the reactor was methyl esterication by using
BF –methanol. Aer methyl esterication, 20 mL of the sample
3
was dissolved in 1 mL isooctane, then it was analyzed by a gas
chromatograph (GC, Agilent 7890A) equipped a 100 m HP-88
column (inner diameter: 0.25 mm and lm thickness:
2
0
like layers. All above features indicate that ruthenium was
incorporated in the hydrotalcite layers.
22
The XRD patterns (Fig. 2) of sample thermally at various
temperature revealed that the layered structure of hydrotalcite
destroyed in the temperature range of 350–400 C. For those
calcined at 450 C and higher temperature, the diffraction peaks
corresponding to MgO and MgAl O spinel can be observed,
which indicated that the hydrotalcite completely transformed
into composite oxides at the temperature above 450 C. More-
over, when calcination temperature increased, the peak posi-
tions of MgO and spinel did not change, which indicated the
good stability of composite oxide derived from the Ru–HT
ꢀ
0
.25 mm), ame ionization detector (FID) unit, and an auto-
ꢀ
ꢀ
sample injector. Injector and detector temperature were 240 C.
The column temperature was maintained at 120 C for 1 min,
and then increased at 10 C min to 175 C and held there for
ꢀ
2 4
ꢀ
ꢁ1
ꢀ
ꢀ
ꢀ
ꢁ1
ꢀ
15 min and then increased at 5 C min to 210 C and held
ꢀ
ꢁ1
ꢀ
there for 5 min and then increased at 5 C min to 230 C and
held there for 5 min. Heptadecane was used as the internal
standard. Most CLA isomers were identied based on retention
times, using standard references. Other CLA isomers were
2
4
hydrotalcite.
The evolution of the gaseous products formed during the
thermal decomposition of the interlayer anions incorporated in
the studied Ru–HT is shown in Fig. 3. The two peaks 208 C and
254 C corresponded to the elimination of physically adsorbed
water and interlayer water; the third peak of 385 C can be
ascribed to the removal of OH from the brucite-like layer as
water molecules. Simultaneously, it is clear that CO
are transformed into CO , which leaves the material starting
23
identied based on literature data.
ꢀ
ꢀ
3. Results and discussion
ꢀ
The XRD patterns of the as-synthesized HT and Ru–HT are
shown in Fig. 1. For the Ru–HT, the diffraction peaks are typical
characteristics of the hydrotalcite (JCPDS le no. 14-0191),
indicating that Ru–HT was successfully prepared. These
reections at respective 2q are no difference and revealed a good
dispersion of ruthenium. The intensities of (003) and (006)
planes, which are directly related to the crystallinity were
observed to decrease to 85% for Ru–HT samples as compared to
pristine HT (Table 1). Decrease in the crystallinity on intro-
ducing the ruthenium cations in hydrotalcite structure could be
attributed to the increase in the number of cations of higher
ionic radii in brucite sheet. Increase in the value of a was also
ꢁ
25
2ꢁ
3
anions
2
ꢀ
from about 250 C, the maximum of CO evolution rate is at
2
ꢀ
about 435 C.
In order to better understand the change on the functional
group of Ru–HTs during calcination, FTIR (Fig. 4) measure-
ments have been carried out. The FTIR spectrum of the uncal-
cined sample is typical of HTLCs compounds containing mainly
carbonate anions. Three general types of IR-active vibrations of
hydrotalcites can be distinguished: molecular vibrations of the
˚
observed for Ru–HT sample (3.066 A) as compared to HT sample
˚
˚
(
3.062 A) due to larger ionic radii of ruthenium (0.68 A) as
Table 1 Physical characterization of HT and Ru–HT samples
˚
compared to aluminum (0.53 A). Decrease in the value of unit
˚
Unit cell
parameter (a), A
Unit cell
parameter (c), A
cell parameter c was observed for Ru–HT sample (23.47 A) as
˚
˚
˚
Samples
Crystallinity, %
compared to the value for HT (23.50 A). This results into
decrease in charge density on layers due to weaker interaction
HT
Ru–HT
3.062
3.066
23.50
23.47
100
85
(or decrease in Coulombic attractive force) between the nega-
tively charged interlayer anions and positively charged brucite
20250 | RSC Adv., 2015, 5, 20248–20255
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
Fig. 2 The XRD patterns of Ru–MgAl350 (a), Ru–MgAl400 (b), Ru–
MgAl450 (c), Ru–MgAl550 (d), Ru–MgAl700 (e).
Fig. 4 The FTIR spectra of Ru–HT (a), Ru–MgAl350 (b), Ru–MgAl400
(c), Ru–MgAl450 (d), Ru–MgAl700 (e).
hydroxyl groups, lattice vibrations of the octahedral layers and
vibrations of the interlayer species. The band observed at
the presence of water in the interlayer space. No signicant
changes within these ranges were observed in the spectra of
calcined HTLCs (Fig. 4b–d). It showed that calcination did not
reduce the number of hydroxyl groups. On the other hand, the
ꢁ1
1
360 cm is attributed to the n asymmetric stretching of the
3
carbonate anions. Compared with the FTIR spectrum of the
uncalcined sample, remarkable changes are observed in the IR
ꢁ
1
bands in the region below 1000 cm , which can be assigned to
ꢀ
ꢀ
26
spectra of HTLCs calcined at 350 C and 400 C. Some authors
the M–O (M ¼ Zn, Ni and Al) vibrations, is fully agrees with the
20
have observed this rearrangement of carbonate anions during
thermal annealing, and, in general, it happens along with the
anionic graing. The graing of the anions with the layer leads
to the formation of a new layered phase with an interlayer
distance smaller than that observed in the parent sample. This
assumption is supported by XRD results and discussion. When
description in the literature. The FTIR study entirely consis-
tent with the results obtained by means of the EGA-MS and
XRD.
The composite oxides were used to isomerize linoleic acid to
conjugated linoleic acid. Fig. 5 showed the CLA conversion as a
function of catalysts calcination temperature, suggesting that the
calcination temperature has a marked effect on the catalyst
ꢀ
the calcination temperature increased to 450 C, remarkable
ꢁ
1
ꢀ
ꢀ
changes are observed. The 1360 cm
band disappear
activity. In the range of 350 C to 450 C, the activity of the
catalyst increased with the increasing calcination temperature.
completely. It can be indicated that the carbonate anions were
ꢀ
ꢀ
removed when being calcined above 450 C. The bending mode
The catalyst calcined at 450 C exhibited the best excellent
ꢁ
1
H–O–H from H O was observed at 1640 cm thus conrming
2
Fig. 5 Effect of calcination temperature on CLA conversion. Reaction
ꢀ
Fig. 3 EGA-MS profiles of evolution of the gaseous products of the temperature 180 C, catalyst weight 0.4 g, reaction time 2 h, 10 mL 1 M
thermal decomposition of interlayer anions built in Ru–HT.
LA solution.
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 20248–20255 | 20251

View Article Online
RSC Advances
Paper
Table 2 Specific surface area of catalysts at different calcination temperature
Sample
Specic surface area (m g
Ru–HT
Ru–MgAl350
Ru–MgAl400
Ru–MgAl450
Ru–MgAl550
Ru–MgAl700
2
ꢁ1
)
79
112
177
155
138
116
catalytic activity. However, the catalyst activity sharply dropped oxidized ruthenium species are well-stabilized in the catalyst
ꢀ
when the temperature increased to 550 C and higher. For matrix. Moreover, this reduction temperature is much higher
2
Ru–MgAl450, the 65% conversion was observed while only 7% than that found for bulk RuO , and apparently indicates that a
conversion was sustained for Ru–MgAl700. Generally speaking, strong interaction of the ruthenium species exists with the
high surface area of catalyst would provide more surface area for support. Obviously, one could be found that in the series of
ꢀ
active sites and higher chances for the catalysts to be exposed samples calcined from 350 to 450 C, the stronger interaction of
to reactant molecule. However, it's worth mentioning that ruthenium species with the support, the better the catalytic
Ru–MgAl350 and Ru–MgAl400 activities were much lower than performances was. As increasing the calcination temperature
ꢀ
ꢀ
Ru–MgAl450 though their surface area (Table 2) were higher than from 550 C to 700 C, a new reduction peak appears in the low
it, which indicated that the surface area was not the main factor temperature, which may be attributed to the reduction of bulk
2
responsible for the catalytic activity. Base on the above XRD and RuO , thus suggesting a larger metallic particle size. This
FT-IR characterization, we think that at low calcination temper- behavior may be explained ruthenium atoms segregate to minor
29
ature, the transition to Ru–MgAls from Ru–HT has not fully quantities of RuO
2
. Clearly, that ruthenium atoms segregate to
formed; the linoleic acid molecules could not be fully close to the RuO is unfavourable for the catalysis activity; the catalysis
2
active sites due to the limitation of interlayer groups. As the activity of Ru–MgAl700 is much lower than that of Ru–MgAl450
calcination temperature increased, the carbonate anions were and even lower than that of Ru–MgAl350. Thus the ruthenium
completely removed, which resulted in the exposed active sites has a profound effect on the catalytic performance of the series
on the surface of the catalysts. Besides, the crystal gradually of Ru–MgAl catalysts.
tended to reach perfection, thus the catalytic activity was
It was reported that the basicity of catalysts played a decisive
increased. However, when the calcination temperature was role in the double bond isomerization, which has been sug-
increased higher, the catalytic activities were decreased, espe- gested to proceed via a carbanion intermediate on the basic
30
cially for the Ru–MgAl700 sample, the LA conversion was only sites and then transferred to the terminal production. The
%. Obviously, others factor may affect their catalytic activity. basicity of the series of Ru–MgAl oxides has been determined
Fig. 6. shows H -TPR proles of the Ru–HT samples calcined using CO -TPD technique (Fig. 7). This technique affords
7
2
2
at different temperature. For Ru–MgAl350, Ru–MgAl400 and information about the strength and amount of basic sites from
Ru–MgAl450, there is only a broad reduction peak with desorption temperature and the peak area, respectively. In all
ꢀ
maximum at around 450 C, respectively. The peak can be cases, the graphs exhibit two desorption peaks, between 50 and
2
7,28
ꢀ
ꢀ
assigned to the reduction of RuO
2
.
Thus it could be stated 150 C and from 370 until 680 C. The differences of the rst
ꢀ
that when the Ru–HT samples is calcined below 450 C, a few desorption temperature for all the samples were small, but the
2
amount of ruthenium is segregated as RuO , most of the second desorption peak shi toward high temperature as the
2 2
Fig. 6 The H -TPR profiles of Ru–MgAl350 (a), Ru–MgAl400 (b), Ru– Fig. 7 The CO -TPD profiles of Ru–MgAl350 (a), Ru–MgAl400 (b),
MgAl450 (c), Ru–MgAl550 (d), Ru–MgAl700 (e).
Ru–MgAl450 (c), Ru–MgAl550 (d), Ru–MgAl700 (e), MgAl450 (f).
20252 | RSC Adv., 2015, 5, 20248–20255
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
Table 3 Comparison of different catalysts described in the literature for the production of CLA in different reaction atmosphere (RAs)
ꢀ
a
b
c
d
Entry
Substrate
Catalyst
Solvent
T [ C]
RA
Y
CLA
%
P
R
TOF
Ref.
1
2
3
4
5
6
7
8
9
Methyl linoleate
Methyl linoleate
Linoleic acid
Linoleic acid
Linoleic acid
Linoleic acid
Linoleic acid
Methyl linoleate
Linoleic acid
Linoleic acid
Methyl linoleate
Linoleic acid
KOH
Ru/C
Ru/C
Ru/AlB2BOB3B
Ru/H–Y
Ethylene glycol
n-Decane
n-Decane
n-Decane
n-Decane
n-Decane
n-Decane
n-Decane
n-Decane
n-Decane
n-Decane
n-Decane
180
250
165
165
120
165
150
165
180
180
180
180
NB2B
NB2B
NB2B
NB2B
NB2B
HB2B
HB2B
NB2B
NB2B
NB2B
NB2B
NB2B
97
31
58
30
24
67
4
75
65
5
0.63
0.019
0.90
0.0048
0.002
0.021
0.0005
0.391
1.4
—
13.6
23
—
5.3
17.2
0.85
0.4
0.7
0.1
99.8
18
18
18
18
18
18
18
18
0.83
0.8
1.5
0.2
234
438
33.7
—
Ag/SiOB2B
Au/C
Ru/Cs–USY e
Ru–MgAl450
Ru–Al450
Ru–MgAl450
MgAl450
102.6
7.9
—
This work
This work
This work
This work
10
11
12
0.108
—
—
z0
z0
—
—
a
b
ꢁ1
ꢁ1
c
ꢁ1
ꢁ1
d
Yield of CLA. CLA productivity [g (CLA) L (solvent) min ]. CLA production rate [g (CLA) g (Ru) h ]. Turnover frequency [mol (LA
ꢁ1
ꢁ1
e
converted) mol (Ru) h ]. The Ru-content of the catalyst was detected by XRF ¼ 0.8%.
ꢀ
ꢀ
calcination temperature increased from 350 C to 550 C, which basicity amounts as catalyst (Table 3, entries 10), the CLA yield
indicated that the strength of basicity of Ru–MgAl increased was very low. In addition, when methyl linoleate was used as the
with the rise of calcination temperature, and the Ru–MgAl550 reactant (Table 3, entries 11), the CLA yield was almost
sample exhibits the highest basicity among this family of cata- approaching zero. The above two tests revealed that the basic
lysts. Interestingly, the same trend has taken place in the above sites of the catalysts were very important for the isomerization,
linoleic acid isomerization reaction (Fig. 5), namely, the catal- but it is not the only decisive role. We therefore suppose that the
ysis activity increased with the strength of basicity of catalysts; it basic site maybe plays an absorption role. To identify the active
seems that the catalytic activity is related to the basicity strength centers responsible for the isomerization reaction, another test
of the catalyst. However, some interesting results were found. was performed by using MgAl450 devoid of ruthenium, the CLA
When we used Ru–Al450 which was reported to have a few yield was also approaching zero, (Table 3, entries 12). It is worth
Fig. 8 Possible reaction mechanism for the isomerization of linoleic acid to conjugated linoleic acid using Ru–MgAl catalyst.
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 20248–20255 | 20253

View Article Online
RSC Advances
Paper
mentioning that the basic strength and the basic amount
(Fig. 7c and f) for the Ru–MgAl450 and MgAl450 were almost
equal. Hence, those above facts let us to think that the basicity
sites of catalyst and ruthenium could have a synergic effect on
the catalytic activity, namely, the Ru–MgAl was a bifunctional
catalyst where the catalyst not only provided the basic sites, but
also provided the noble metal activity sites necessary in the
isomerization reaction. What's more, the basic site plays an
absorption role and the ruthenium activity site responsible for
the double bond migration reaction. According to the above
facts, the possible reaction mechanism for the isomerization of
linoleic acid to conjugated linoleic acid was proposed as follows
(
Fig. 8): initially, the linoleic acid was absorbed on basic sites of
the solid base surface (step I). The next step II is the interaction
electron of 1 to form a p-complex 2 with RuO . The equilibrium
2
step III gives p-allyl metal complex 3 through allyl-H-migration
from g-carbon atom to metal. The hydride shi from p-allyl
metal-hydride complex 3 to a-carbon atom gives complex 4 via
next equilibrium step IV. The dissociation of active catalyst
species from the complex 5 results the formation of conjugated
linoleic acid.
Fig. 10 Reusability of catalysts for isomerization of linoleic acid (gray
¼
Ru–MgAl450, black ¼ impregnated Ru-catalyst).
are 25.680, 25.994, 26.681 min, respectively. Other CLA
isomers were also formed, but the amounts were small. The
above results indicate that the catalyst shows high selectivity
for desirable c9, t11-, t10, c12- and t9, t11-CLA isomers.
For a catalyst, the reusability in a sense is more important than
activity. The optimal Ru–MgAl450 catalyst was tested in a recycling
experiment. Conversion for isomerization of linoleic acid was
observed to remain unchanged even aer the h run (Fig. 10),
conrming that the catalyst was reusable for the isomerization
reaction without loss in its activity. In the rst run, the catalytic
activity of sample synthesized by impregnation of ruthenium on
hydrotalcite was similar to that of Ru–MgAl450 sample. However,
the conversion rate of linoleic acid decreases sharply aer the
second run which might be due to the leaching of ruthenium from
support. This suggests that the substitution of ruthenium for Mg
or Al cations increases the stability of catalysis, which makes Ru–
MgAl an active and reusable catalyst.
Comparison of the production of CLA for different catalysts
in different reaction atmospheres described in the literature
was showed in Table 3. Most of the catalysts except Ru/Cs–USY
showed low productivity. However, the productivity and TOF of
ꢁ
1
ꢁ1
Ru–MgAl450 reached 1.52 g (CLA) L (solvent) min and
1
ꢁ
1
ꢁ1
02.6 mol (LA converted) mol (Ru) h . What's more, the
productivity of Ru–MgAl450 could even be comparable to that
of the homogeneous process used industrially today. It is
important for a heterogeneous catalyst with high selectivity to
form physiologically CLA isomers. Fig. 9 shows the CLA
isomers determined by GC analysis. Peak identications for
three CLAs were conrmed by spiking with standard CLAs. The
result indicates that three major CLA isomers were formed
during the conjugation reaction, namely, c9, t11-CLA (14%),
t10, c12-CLA (16%) and t9, t11-CLA (35%). Their retention time
4. Conclusions
The Ru–MgAl composite oxides catalysts prepared by thermal
decomposition of ruthenium graed hydrotalcite were used to
isomerization of linoleic acid to conjugated linoleic acids
without hydrogen. Results indicated that the catalyst calcined at
ꢀ
4
50 C showed the highest catalytic activity, compared with
other heterogeneous processes reported in the literature, higher
productivities of and selectivities for CLA can be obtained. The
Ru–MgAl was a bifunctional catalyst and the basicity sites and
ruthenium had a synergic effect on the catalytic activity; mean-
while, the basic site plays an absorption role in the isomerization
reaction, and the ruthenium activity site responsible for the
double bond migration reaction. Because it had high produc-
tivity as well as durability and it also produced benecial CLA
isomers products, the catalyst of Ru–MgAl450 is suitable for the
isomerization of linoleic acid to conjugated linoleic acid; this
research may also accelerate the development of CLA-enriched
functional foods.
Fig. 9 Gas chromatogram of the isomerization produces from linoleic
acid.
20254 | RSC Adv., 2015, 5, 20248–20255
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
1
3 A. Bernas, N. Kumar, P. Laukkanen, J. Vayrynen, T. Salmi
and D. Y. Murzin, Appl. Catal., A, 2004, 267, 121.
Acknowledgements
The authors are grateful for the nancial support from the 14 A. Bernas, P. M. Arvela, N. Kumar, B. Holmbom, T. Salmi and
Natural Science Foundation of Fujian Province of China
D. Y. Murzin, Ind. Eng. Chem. Res., 2003, 42, 718.
no. 2014J01032), the Undergraduate Innovation and Entrepre- 15 N. Chorfa, S. Hamoudi and K. Belkacemi, Appl. Catal., A,
(
neurship training program innovative training project of Fujian
2010, 387, 75.
Province and Youth Foundation of Fujian Agriculture and 16 A. Bernas, N. Kumar, P. M. Arvela, N. V. Kul'kova,
Forestry University (2011xjj08 and 2012xjj29).
B. Holmbom, T. Salmi and D. Y. Murzin, Appl. Catal., A,
003, 245, 257.
2
1
1
7 X. Cardo, O. Bergada, Y. Cesteros and P. Salagre, Chem. Eng.
J., 2012, 183, 450.
8 A. Bernas, N. Kumar, P. Maki-Arvela, B. Holmbom, T. Salmi
and D. Y. Murzin, Org. Process Res. Dev., 2004, 8, 341.
References
1
A. Bernas, P. Laukkanen, N. Kumar, P. M. Arvela, J. Vayrynen,
E. Laine, B. Holmbom, T. Salmi and D. Y. Murzin, J. Catal., 19 A. Philippaerts, S. Goossens, W. Vermandel and M. Tromp,
002, 210, 354.
ChemSusChem, 2011, 4, 757.
I. Sluijs, Y. Plantinga, B. Roos, L. I. Mennen and M. L. Bots, 20 S. K. Sharma, P. A. Parikh and R. V. Jasra, J. Mol. Catal. A:
Am. J. Clin. Nutr., 2010, 91, 175.
Chem., 2010, 317, 27.
M. Coakley, M. Johnson, E. McGrath, S. Rahman, R. Ross, 21 I. M. Cabrera, M. L. Grannados and J. L. G. Fierro, Phys.
2
2
3
G. Fitzgerald, R. Devery and C. Stanton, Nutr. Cancer, 2006,
6(1), 95.
W. Shen, C. C. Chuang, K. Martinez, T. Reid, J. M. Brown,
Chem. Chem. Phys., 2002, 4, 3122.
22 F. Basile, G. Fornasari, M. Gazzano and A. Vaccari, J. Mater.
Chem., 2002, 12, 3296.
5
4
5
L. Xi, L. Hixson and R. Hopkins, J. Lipid Res., 2013, 54(4), 909. 23 M. P. Yurawecz, J. K. G. Kramer, O. Gudmundsen,
S. McClelland, C. Cox, R. O. Connor, M. Gaetano,
C. McCarthy, L. Cryan, D. Fitzgerald and O. Belton,
Atherosclerosis, 2010, 211, 96.
L. Gorissen, K. Raes, S. Weckx, D. Dannenberger, F. Leroy,
L. D. Vuyst and S. D. Smet, Appl. Microbiol. Biotechnol., 25 P. Kustrowski, D. Sulkowska, L. Chmielarz, A. R. Lasocha,
M. W. Pariza and S. Banni, Advances in Conjugated Linoleic
Acid Research, AOCS Press, Champaign, 2006, vol. 3.
24 L. Ai, C. Zhang and L. Meng, J. Chem. Eng. Data, 2011, 56,
4217.
6
7
2010, 87, 2257.
B. Dudek and R. Dziembaj, Microporous Mesoporous Mater.,
2005, 78, 11.
B. Yang, H. Chen, Z. Gu, F. Tian, R. P. Ross, C. Stanton,
Y. Q. Chen, W. Chen and H. Zhang, J. Appl. Microbiol., 26 C. H. Zhou, J. N. Beltramini, C. X. Lin, Z. P. Xu, G. Q. Lu and
014, 117, 430.
A. Tanksale, Catal. Sci. Technol., 2011, 1, 111–122.
L. D. Whigham, M. E. Cook and R. L. Atkinson, Pharmacol. 27 A. Ballarini, P. Benito, G. Fornasari, O. Scelza and A. Vaccari,
Res., 2000, 42, 503.
Int. J. Hydrogen Energy, 2013, 38, 15128.
B. Jia, G. Wu, X. Fu, X. Mo, M. Du, Y. Hou and S. Zhu, Mol. 28 Z. W. Yang, W. M. Guo, J. D. Lin and D. W. Liao, Chin.
Reprod. Dev., 2014, 81, 20.
J. Catal., 2006, 27, 378.
0 R. V. Salamon, E. V. Visi, C. D. Andras, Z. C. Kiss and 29 F. Basile, G. Fornasari, M. Gazzano and A. Vaccari, J. Mater.
2
8
9
1
J. Csapo, Acta Universitatis Sapientiae – Alimentaria, 2012, 5,
2.
1 P. Pakdeechanuan, K. O. Intarapichet, L. N. Fernando and
I. U. Grun, J. Agric. Food Chem., 2005, 53, 923.
Chem., 2002, 12, 3296.
3
30 J. Puriwat, W. Chaitree, K. Suriye, S. Dokjampa,
P. Praserthdam and J. Panpranot, Catal. Commun., 2010,
12, 80.
1
12 M. Kreich and P. Claus, Angew. Chem., 2005, 117, 7979.
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 20248–20255 | 20255