Sustainable and efficient methodology for CLA synthesis and identification

Article abstract of DOI:10.1039/c2gc35792e

Microwave-assisted organic synthesis and continuous-flow techniques have been successfully employed for the preparation of conjugated linoleic acids (CLA), compounds with high health beneficial effects. A good production rate of CLA was obtained. A sustainable methodology for the differentiation of both positional and geometrical CLA isomers (diene), based on the analysis by NMR spectroscopy of the resulting Diels-Alder cycloadducts with an appropriate dienophile, was developed.

Full text of DOI:10.1039/c2gc35792e

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Green Chemistry
Cite this: Green Chem., 2012, 14, 2584
www.rsc.org/greenchem
PAPER
Sustainable and efficient methodology for CLA synthesis and identification†
Maria Moreno,a M. Victoria Gomez,*b Cristina Cebrian,a Pilar Prieto,a Antonio de la Hoza and
Andres Moreno*a
Received 22nd May 2012, Accepted 9th July 2012
DOI: 10.1039/c2gc35792e
Microwave-assisted organic synthesis and continuous-flow techniques have been successfully employed
for the preparation of conjugated linoleic acids (CLA), compounds with high health beneficial effects.
A good production rate of CLAwas obtained. A sustainable methodology for the differentiation of both
positional and geometrical CLA isomers (diene), based on the analysis by NMR spectroscopy of the
resulting Diels–Alder cycloadducts with an appropriate dienophile, was developed.
13C-NMR spectroscopy was reported as a tool to identify CLAs;
Introduction
however, only predicted data were given for some of the
isomers.6 On the other hand, the positions of double bonds in
Due to their unique biological properties, conjugated linoleic
acids (CLAs) have received a great deal of attention in the last
few decades. These particular fatty acids show conjugated diene
systems and are found in tissues of ruminant animals and, conse-
quently, in meat and dairy products. They are formed as inter-
mediate compounds in the biohydrogenation of linoleic acid
(LA) carried out by microorganisms in the rumen. However,
several positional and geometrical isomers (6,8- to 12,14-18:2)
are formed. Among these, 9-cis,11-trans-octadecadienoic acid
(9c,11t-CLA) is the main isomer and constitutes about 1% of the
fatty acids of milk fat. Moreover, this isomer exhibits highly
interesting specific physiological activities such as anti-obesity,
anti-carcinogenic, anti-atherogenic, anti-diabetic, antioxidant,
apoptotic, immunomodulatory and osteosynthetic effects.1
Several strategies can be followed for synthesizing CLAs. One
of the employed methods is the alkali isomerization of linoleic
acid, which leads to mixtures of geometric isomers of CLA.2
Another approach is based on the dehydration of ricinoleic acid
methyl ester (RAME) in the presence of KHSO4 or other expens-
ive dehydration reagents, such as DBU (1,8-diazabicyclo-
(5.4.0)-undec-7-ene).3 It is also possible to obtain 9c,11t-CLA
from linoleic acid by means of microbial synthesis or chemo-
enzymatic conversions.4 However, the first method is the most
widely used because of economic reasons.
conjugated dienes can often be verified by mass spectrometry
(MS) of dimethyloxazoline (DMOX) derivatives.7 In addition,
conjugated double bonds of CLAs can react specifically with
certain dienophiles in Diels–Alder cycloaddition reactions. Thus,
4-methyl-1,2,4-triazoline-3,5-dione (MTAD) readily forms stable
adducts with such conjugated dienes and these adducts are suffi-
ciently volatile to be analyzed by GC.8 A related reagent to
MTAD has been used for sensitive fluorescence detection of
CLAs in human serum by HPLC.9 However, all these methods
only allowed the determination of positional isomers, not giving
information on the double bond geometry.
Microwave irradiation has been successfully applied in
organic chemistry.10,11 Spectacular accelerations in comparison
to conventional heating are attained, resulting in shorter reaction
times. In this sense, Diels–Alder reactions, which are often
limited by the reversibility of the reaction, have been accom-
plished with great success due to the reduction of reaction
times.12 It is worth noting that, under microwave heating,
the selectivity (chemo-, regio- and stereoselectivity) can be
modified in relation to that obtained with conventional heating.13
Further improvements can be achieved by using neutral and
recyclable ionic liquids (ILs), or under solvent-free reaction con-
ditions, which give rise to environmentally friendlier
protocols.14
We herein report the development of a new methodology to
identify conjugated linoleic acids in a complex matrix. The
study concerns the search and implementation of the most
appropriate dienophile for identifying CLA isomers (diene) by
NMR analysis of the corresponding formed Diels–Alder
cycloadducts. The use of maleic anhydride in microwave-
assisted solvent-free cycloaddition reactions of highly reactive
CLAs permits the differentiation of both positional and geo-
metrical CLA isomers by 13C-NMR spectroscopy. Furthermore,
two new sustainable approaches for the synthesis of CLAs are
reported.
As for the analytical tools, gas chromatography (GC) and
high-performance liquid chromatography (HPLC) have been
extensively used to identify this kind of fatty acid.5 Moreover,
aDepartamento de Química Orgánica, Facultad de Química,
Universidad de Castilla-La Mancha, 13071 Ciudad Real, Spain.
E-mail: Andres.Moreno@uclm.es; Fax: +34 926 295 318;
Tel: +34 926 295 300
bInstituto Regional de Investigación Científica Aplicada (IRICA),
Universidad de Castilla-La Mancha, 13071 Ciudad Real, Spain
†Electronic supplementary information (ESI) available. See DOI:
10.1039/c2gc35792e
2584 | Green Chem., 2012, 14, 2584–2594
This journal is © The Royal Society of Chemistry 2012

View Article Online
Scheme 2 Diels–Alder reactions of sorbic acid (1).
Table 1 Diels–Alder reaction conditions of sorbic acid (1) with
different dienophiles
Entry
Dienophile
T (°C)
t (min)
P (W)
Yield (%)
1
2
3
4
TCE (3)
PBQ (4)
NMM (5)
MA (6)
80
150
100
150
15
15
15
15
30
120
30
—
—
72 (7) + 18 (8)
93 (9)
100
with differences in their 13C-NMR chemical shifts, and provided
that NMR spectroscopy allows the elucidation of the CLA geo-
metric (cis/trans) isomerism in contrast to GC and HPLC
analytical techniques.
Thus, some compounds similar in structure to CLA were pur-
chased to be used as dienes in order to understand the reactivity
of CLAs in Diels–Alder cycloadditions for certain sustainable
reaction conditions. Then, sorbic acid (1) and 2,4-trans,trans-
hexadienyl acetate (2) were chosen. On the other hand, focused
on finding the best dienophile, tetracyanoethylene (TCE) (3),
p-benzoquinone (PBQ) (4), N-methylmaleimide (NMM) (5) and
maleic anhydride (MA) (6) were selected, because they are
known to be highly reactive in this type of reaction, stable and
their products easily characterized.15
Fig. 1 Olefinic 1H- and 13C-NMR data of 9c,11t CLA (■) and 10t,12c
CLA (●) isomers in a cheese organic extract sample (a), synthetic
mixture isomers (b) and the corresponding formed Diels–Alder cyclo-
adducts (c).
Reactions were optimized and carried out under solvent-free
conditions in a monomode microwave reactor Discover CEMTM
.
Different power and temperatures were used depending on the
dienophile and in all reactions microwave irradiation was
required for 15 min (Tables 1 and 2).
In the case of sorbic acid (1), excellent yields were obtained
from N-methylmaleimide (5) and maleic anhydride (6)
(Scheme 2). However, no products were obtained from tetracya-
noethylene (3) because of an uncontrolled absorption of the
microwave irradiation, resulting in a very quick increase of the
reaction temperature and an undesirable reaction mixture. p-Ben-
zoquinone (4) did not turn out to be a good dienophile since
changes in the reaction parameters (time, temperature and
power) did not result in products. Yields of the reaction are
Scheme 1 General scheme of the methodology. Step 1: synthesis of
CLA. Step 2: Diels–Alder cycloaddition reaction.
Results and discussion
Diels–Alder reactions of conjugated linoleic acids
The necessity of the development of a new methodology for dif-
ferentiating the positional and geometrical isomers of CLA mix-
tures is illustrated by three different key facts. (i) The separation
of the different CLA isomers is a rather difficult task by means
of the commonly used purification techniques; (ii) their identifi-
cation in a food complex matrix is also difficult because of their
low natural abundance and the presence of other compounds
(Fig. 1a); (iii) the different CLA isomers show up similar data in
1H and 13C-NMR spectra (Fig. 1b), even more, the chemical
shifts in the 13C-NMR spectrum depend on the presence of other
isomers in the food matrix. Therefore, we focused our attention
on the development of a methodology for such purpose implying
the NMR analysis of the stereochemistry of the corresponding
formed cycloadducts as a result of an appropriate Diels–Alder
reaction (Scheme 1, step 2), presumably with simpler spectra,
1
shown in Table 1. The pure products were identified by H and
13C-NMR spectroscopy as well as mass spectrometry.
The formation of stereoisomeric products should only be poss-
ible as a consequence of isomerization of the double bonds,
which was not observed. However, this is not the case of reaction
of sorbic acid (1) with NMM (5) (Table 1, entry 3) which pro-
duced diastereoisomers 7 and 8 in a 4 : 1 ratio; both result from
the endo addition. The stereochemistry of these products was
inferred by NOE difference experiments.
On the other hand, the previous set of cycloaddition reactions
were carried out again but employing 2,4-t,t-hexadienyl acetate
(2) as a diene (Scheme 3, Table 2). These reactions led to good
results for all dienophiles except for p-benzoquinone (Table 2,
This journal is © The Royal Society of Chemistry 2012
Green Chem., 2012, 14, 2584–2594 | 2585

View Article Online
Scheme 3 Diels–Alder reactions of 2,4-t,t-hexadienyl acetate (2).
Scheme 4 Diels–Alder reactions of conjugated linoleic acid (14–15).
Table 2 Diels–Alder reaction conditions of 2,4-t,t-hexadienyl acetate
(2) with different dienophiles
results were achieved using microwave irradiation for 15 min at
150 °C in a 1 : 2 diene–dienophile molar ratio. It should be
noted that all reactions occurred in the absence of solvent.
Entry
Dienophile
T (°C)
t (min)
P (W)
Yield (%)
1
The pure products were identified by H and 13C-NMR spec-
1
2
3
4
TCE (3)
PBQ (4)
NMM (5)
MA (6)
80
150
100
150
15
15
15
15
30
120
30
88 (10)
—
93 (11)
95 (12)
troscopy (Fig. 1c) as well as mass spectrometry. A thorough
NMR study was performed in order to elucidate the stereo-
chemistry of the formed cycloadducts consisting of the following
NMR experiments: g-COSY, g-HMQC, g-HMBC, TOCSY,
g-HSQC-TOCSY and HMBC-TOCSY. Particularly, 1D-NOESY
and 2D-NOESY experiments made it possible to unequivocally
identify the different cycloaddition isomers. In addition, it
allows us to conclude that the cycloaddition of maleic anhydride
(6) with conjugated linoleic acids is also 100% stereoselective
because the endo isomer is formed exclusively without any of
the exo isomer.
Thus, cycloadducts 18 and 19 were obtained from the CLA
isomers synthesized by the alkali isomerization procedure, with
100% conversion of reagents in the corresponding products and
86% of total reaction yield (Scheme 4). It is important to note
that complete conversion is needed for the applicability of this
methodology, in order to avoid the presence of NMR signals,
corresponding to the starting material, in the region of study. At
this point, it was possible to identify the different CLA geometri-
cal isomers used as the starting material, after having elucidated
the stereochemistry of the isolated cycloadducts. Accordingly,
from linoleic acid isomerization 9c,11t-CLA (14) and 10t,12c-
CLA (15) (1 : 1 ratio) were achieved, illustrating that microwave
irradiation did not modify the selectivity because no differences
were found with respect to the conventional reported method.2
Note that cycloadduct 18 is derived from CLA 14 and cyclo-
adduct 19 from CLA 15. More importantly, the 13C-NMR spec-
trum of the corresponding cycloadducts is simpler than that
observed for the CLA isomers (Fig. 1c vs. 1b) due to the mag-
netic and chemical equivalency present in the reaction product.
Moreover, the cycloadducts chemical shift values are indepen-
dent of the presence of other isomers in contrast to that observed
for the corresponding CLA isomers. These facts clearly illustrate
the validity of the here reported method.
100
entry 2), where changes in reaction parameters (time, tempera-
ture and power) did not result in the desired products. No iso-
merization of double bonds was observed. The pure products
were identified by 1H and 13C-NMR spectroscopy as well as
mass spectrometry. All reactions are 100% stereoselective and
produce the endo stereoisomer according to the general require-
ments of the Diels–Alder reaction. The stereochemistry of these
products was inferred by NOE difference experiments.
Based on such results, we can conclude that the best dieno-
phile is the maleic anhydride (6) due to its high stereoselectivity
and high reaction yields. Accordingly, the study of the reactivity
of conjugated linoleic acids was carried out using maleic an-
hydride (6) as a dienophile.
Two sustainable approaches, microwave-assisted organic syn-
thesis as well as continuous flow techniques, were successfully
employed (vide infra) for the synthesis of CLA, slightly modify-
ing the reported procedure2 of the alkali isomerization of linoleic
acid (13) (Scheme 1, step 1). They resulted in a mixture of two
positional and geometrical isomers of CLAs which showed very
similar NMR data (Fig. 1b). In fact, no differences are found in
1
the H-NMR spectra, whilst slight differences are observed in
the 13C-NMR spectra (Fig. 1b). Due to the fact that these two
methods can lead to the modification of the selectivity of the
reaction,13 the assumption of having obtained the same geo-
metric isomerism as described2 can not be made. Different
methodologies such as column chromatography and HPLC were
employed for the separation of the CLA isomers with unsuccess-
ful results. Nevertheless, the purpose of this work was the devel-
opment of a methodology for the identification of CLA isomers
in a complex matrix, the accomplishment of the aforementioned
separation being unnecessary.
On the other hand, two other CLA isomers originated from
isomerization of 9c,11t-CLA (14) and 10t,12c-CLA (15) cata-
lyzed by iodine, expecting to observe the trans/trans geometry
as reported.16 The Diels–Alder cycloaddition of such dienes
with maleic anhydride afforded the cycloadducts 20 and 21,
After the selection of maleic anhydride (6) as a dienophile,
Diels–Alder reactions were performed employing the initially
obtained CLA (diene) mixtures (Scheme 1, step 2). The best
2586 | Green Chem., 2012, 14, 2584–2594
This journal is © The Royal Society of Chemistry 2012

View Article Online
Scheme 5 Diels–Alder reactions of conjugated linoleic acid (16–17).
with 100% conversion of the diene and 80% of total reaction
yield (Scheme 5). After elucidation of the stereochemistry of the
isolated cycloadducts, the geometric isomerism of the starting
material was deduced to be trans/trans, particularly 9t,11t-CLA
(16) and 10t,12t-CLA (17) (1 : 1 ratio). Cycloadduct 20 is
derived from CLA 16 and cycloadduct 21 from CLA 17.
In short, 9c,11t-CLA (14), 10t,12c-CLA (15), 9t,11t-CLA
(16) and 10t,12t-CLA (17) are the CLA isomers obtained. Note
that they represent the most interesting CLA isomers from a bio-
logical point of view, where 9c,11t-CLA (14) is significantly the
most important.1
Fig. 2 Fully optimized B3LYP/6-31G* structures of the different
Diels–Alder cycloadducts of CLA and maleic anhydride.
Table 3 Comparison of NOE experiment data (%) (after irradiating the
protons of interest) with the distances (H–H) theoretically calculated (Å)
To verify the aforementioned conclusions, we performed the
Diels–Alder reactions with some pure commercially available
CLA isomers (9c,11t-CLA (14), 10t,12c-CLA (15) and 9t,11t-
CLA (16)). For all cases, quantitative yields in the formation of
the cycloadduct were obtained for maleic anhydride (6) as the
dienophile in our reaction conditions. The pure products were
H–H
Cyc
NOE
dH–H
Cyc
NOE
dH–H
H
H
H
H
H
H
H
H
3′a–H4′
18
1
3
3
1
4
2
2
4
3.03
2.29
2.47
3.38
2.29
3.02
3.43
2.48
20
5
5
2
2
5
5
2
2
2.38
2.40
3.34
3.34
2.38
2.38
3.34
3.34
7′a–H7′
3′a–H8
7′a–H1′′
3′a–H4′
7′a–H7′
3′a–H9
7′a–H1′′
1
identified by H and 13C NMR spectroscopy as well as mass
19
21
spectrometry, confirming that the former relation found for each
pair CLA–cycloadduct (i.e. cycloadduct 18 derived from 9c,11t-
CLA (14)) was correct. The comparison of the NMR spectra of
our resulting reaction mixture, in which some CLA isomers are
present, with a purchased CLA isomer would be possible, but
only in the case in which a limited number of isomers are
present as illustrated in Fig. 1b (only two isomers and with slight
considering chloroform as the solvent. The most stable structures
for each type of cycloadduct are depicted in Fig. 2. Both
cycloadducts 18 and 19 adopted a conformation of half-chair
(twisted form). In contrast, cycloadducts 20 and 21 exhibited a
preference for a boat conformation. Nonetheless, in all cases the
alkyl substituents on the central cyclohexene ring occupied equa-
torial positions in comparison with the less sterically demanding
hydrogen atoms, which remained in the axial positions.
differences in 13C-NMR chemical shift values (Δδ13C-NMR
=
0.1–0.025 ppm)). However, this comparison can not be done in
a complex matrix (Fig. 1a) since the presence of other com-
pounds not only makes the identification a difficult task but also
alters the NMR chemical shifts of the CLA isomers. These facts
illustrate the utility of this reported methodology.
Moreover, it is worth noting that the structural differences
among conformers corresponding to the same cycloadduct were
related to the disposition of the alkyl chains, and not to the
central bicyclic unit.
In order to verify the optimized structures, some relevant cal-
culated proton distances and the corresponding NOE values after
irradiating the protons of interest are collected in Table 3. As is
shown for each compound, shorter distances are associated with
higher NOE values and vice versa. In this sense, proton distances
of more than 3.00 Å gave NOE values of 1–2%, meanwhile
proton distances of ca. 2.40 Å led to NOE values of 3–5%.
Therefore, this excellent agreement of theoretical distances and
NMR data for each optimized structure of the cycloadducts
In the attempt to further confirm the stereochemistry of the
different cycloadducts, theoretical calculations were performed.
One of the most important uses of theoretical distance geometry
is for deriving conformations that are consistent with the exper-
imental distance information, especially distances obtained from
NMR experiments as in this case.17 Therefore, the corresponding
structures at their ground state were optimized. At a first stage,
Monte Carlo Multiple Minimum simulations were carried out to
preliminarily evaluate the most stable conformers for each cyclo-
adduct (see ESI† for the structure of all the conformers). The
OPLS_2005 force field as implemented in MacroModel v.9 was
used for all molecular modelling. As a second step, the selected
conformers were optimized at the B3LYP(PCM)/6-31G* level,
This journal is © The Royal Society of Chemistry 2012
Green Chem., 2012, 14, 2584–2594 | 2587

View Article Online
Fig. 3 Experimental setup of the flow system.
and quenching the reagents continuously through the microreac-
tor. This approach permits rapid experimentation and scale-up,
thus shortening the time from research to development and pro-
duction. Furthermore, from an environmental point of view, pro-
duction of hazardous waste is also reduced.20
In this sense, in order to improve the synthesis of CLA by the
alkali isomerization of linoleic acid (13) and to produce rela-
tively high amounts of CLA isomers, a very valuable com-
pound,21 we envisaged the application of flow chemistry to this
reaction. Thus, one of the lines was fed with linoleic acid in
ethylene glycol and the other one with the catalyst (KOH in
ethylene glycol). Both streams were mixed in a T-mixer and the
reaction was carried out in a high temperature module. Then, the
output was collected (Fig. 3).
Reaction conditions were optimized by modulating the temp-
erature, time and equivalents of KOH. The best conditions were
found at 180 °C and a residence time of 30 min, achieving a
100% conversion of LA to CLA. It should be noticed that the
reaction was completed in 30 min instead of the 4 h reported
with conventional heating.2 The results were similar to the
microwave approach, resulting in isomers 9c,11t-CLA (14) and
10t,12c-CLA (15) (1 : 1) (Scheme 6). The scale-up afforded 1 g
of CLA isomers per hour. Note that the scale-up process is
limited by the solubility of KOH in ethylene glycol. This good
production rate makes it interesting to carry out epidemiological
studies1 usually limited by the low availability of this
component.
Scheme 6 Isomerization of linoleic acid (13) to CLA isomers.
confirms the obtained structures as the most stable ones.
Additionally, it allows us to establish the value of the employed
theoretical model for describing the geometrical parameters of
the here investigated cycloadducts.
Synthesis of conjugated linoleic acids
Microwave-assisted organic synthesis and continuous flow tech-
niques were employed for the synthesis of CLAs. Initially, basic
isomerization of linoleic acid (13) was performed following a
slightly modified reported method (Scheme 6).2 Microwave
irradiation was successfully applied. In comparison to conven-
tional heating, reaction time was drastically reduced from 4 h to
15 min when employing microwave heating. Tedious and large
steps in the preparation of the reaction were avoided because the
addition of reagents was carried out in a one pot procedure. In
this way, 9c,11t and 10t,12c conjugated linoleic acids (14 and
15) were obtained with a 1 : 1 ratio and high yields (90%). No
changes in isomerization products ratio were observed by using
conventional heating or microwave irradiation.
On the other hand, isomers 9t,11t-CLA (16) and 10t,12t-CLA
(17) were obtained by isomerization from the mixture of pro-
ducts 14 and 15 (1 : 1 ratio) previously prepared. The reaction
was carried out in the presence of iodine at room temperature
during 5 h (Scheme 6).16 An equimolecular mixture of these two
isomers was provided with a 75% reaction yield. No isomeriza-
tion towards cis,cis double bonds was observed.
Therefore, isomerization of linoleic acid (13) to CLA was suc-
cessfully transferred to flow systems, further proving to be a
valuable alternative to batch procedures and producing more
product in a given time than an analogous batch reactor.18
Focused on the development of
a more sustainable
methodology for the synthesis of CLAs, which avoids the use of
ethylene glycol, neutral and recyclable ionic liquids were
effectively used as a green reagent and solvent in the
dehydration of ricinoleic acid (RA) (22).3 Currently, ionic
liquids are widely used in synthesis due to their interesting pro-
perties such as low vapour pressure, high thermal stability and
easy recyclability.22
First of all, ricinoleic acid (22) was heated with 1-butyl-3-
methylimidazolium hexafluorophosphate (23) under microwave
irradiation for 30 min at 140 °C and the corresponding CLA
isomers were obtained in 32% conversion along with some side
products (Scheme 7). We carried out various modifications on
the reaction conditions, such as prolonged reaction times and
increased reaction temperatures, to further increase the reaction
performance, but to no avail. Consequently, a range of ionic
liquids (Table 4) was investigated for their effects on the
dehydration of RA (22). All the ionic liquids were
As is shown in Scheme 6, pairs of isomers of CLA are
obtained which could not be separated either by column chrom-
atography or by HPLC to afford the necessary quantities to
execute the Diels–Alder cycloadditions. Therefore, we have
worked with pairs of isomers, as shown in the foregoing discus-
sion for the CLA identification.
The same procedure was carried out in a flow system using a
minifluidic reactor,18 R2+R4 Vapourtec. Facile automation,
secured reproducibility, improved safety and process reliability
are several advantages when working on the continuous flow
regime. Over the last 15 years, micro- and minifluidic reactors
have entered the field of flow chemistry.19 Minifluidic reactors
present some advantages compared to microfluidic reactors such
as improved flow capacities, lower pressure drop and no block-
ing of channels among others.18 Optimization of reaction con-
ditions is performed by control of residence time, and scalability
of this kind of reaction is simply a matter of pumping, mixing
2588 | Green Chem., 2012, 14, 2584–2594
This journal is © The Royal Society of Chemistry 2012

View Article Online
hexafluorophosphate (23) in order to increase the conversion to
CLAs. In this manner, the formation of the lactone (29) was
almost suppressed and high conversions to CLA isomers were
obtained (Table 4, entry 7). The small quantity of the lactone
(29) detected in the mixture could be due to a possible
transesterification reaction.
Furthermore, in order to check the recyclability of the ionic
liquid, the products were extracted after completion of the reac-
tion with ethyl ether, and the remaining ionic liquid was used
again in a new reaction. There were no significant changes in the
yields of reactions after three cycles.
In addition, water has been used as a solvent to conduct these
reactions as an alternative to ionic liquids under the same exper-
imental conditions. However, no reaction product was obtained.
In summary, three new sustainable approaches for the syn-
thesis of CLAs are reported; microwave-assisted organic syn-
thesis as well as continuous flow techniques have been
successfully employed. Following our goal, it was noticed that it
was much better to use the conjugated linoleic acids obtained by
alkali isomerization of linoleic acid (13) than those obtained by
dehydration of ricinoleic acid (22), because the latter provides a
higher number of isomers. However, this is a mild, efficient and
environmentally friendly alternative for the dehydration of
ricinoleic acid.
Scheme 7 Dehydration of ricinoleic acid (22) to CLA isomers.
Table 4 Effect of different ionic liquids on the dehydration of
ricinoleic acid (22) under microwave irradiation
CLA (%) LA (%) RAL (%) RA (%)
Entry Ionic liquid
(14–17)
(13)
(29)
(22)
1
2
3
4
[bmim]PF6 (23)
32
34
32
11
13
7
22
21
12
3
46
45
9
5
6
7
<1
—
—
47
81
76
85
—
[hmim]PF6 (24)
[bmim]Br (25)
[hmim]Br (26)
[bmim]Cl (27)
[hmim]Cl (28)
Experimental
General
5
5
6
1
Microwave irradiations were performed in a CEM Discover
microwave reactor. Temperature was monitored during the reac-
tion by the standard IR pyrometer included in the microwave
reactor. Analysis by layer chromatography was performed on
aluminium oxide 60 F254 silica gel plates and visualised using
short-wave ultra-violet light or with a KMnO4 solution. All sep-
arations by column chromatography were carried out on silica
gel Merck type 60 (0.040–0.063 mm).
7a
[bmim]PF6 (23) 57b
42c
IL, ionic liquid; CLA, conjugated linoleic acid; LA, linoleic acid; RAL,
ricinoleic acid lactone; RA, ricinoleic acid (unreacted). a Ricinoleic acid
methyl ester. b CLA methyl esters. c LA methyl ester.
commercially available, except 1-hexyl-3-methylimidazolium
hexafluorophosphate (24) and 1-hexyl-3-methylimidazolium
bromide (26) which were synthesized as described in the Exper-
imental section. Again, the procedure gave rise to mixtures of
dienoic acids, such as LA (13) as well as positional and geo-
metrical isomers of CLA (Scheme 7), which were identified and
quantified by NMR. In addition, there was another secondary
product, whose 1H and 13C-NMR spectra revealed that it was the
lactone 29 from ricinoleic acid due to the intramolecular esterifi-
cation between carboxyl and hydroxyl groups.
According to the results shown in Table 4, the nature of the
counteranion in the ionic liquid seems to play a crucial role on
the effect of the dehydration of ricinoleic acid (22). The best
results were achieved with 1-butyl-3-methylimidazolium hexa-
fluorophosphate (23) (Table 4, entry 1) and 1-hexyl-3-methyl-
imidazolium hexafluorophosphate (24) (Table 4, entry 2),
because they enhanced the concentrations of CLAs in compari-
son with other ionic liquids, and there was no unreacted RA.
Nevertheless there was a significant increase in the formation of
the lactone (29). We became interested in exploring ways to
avoid the formation of compound 29, so we have used ricinoleic
Melting points are uncorrected. All NMR spectra were
recorded on a Varian Inova 500 MHz spectrometer in CDCl3 at
1
278 K and at 499.769 MHz and 125.678 MHz for H and 13C
NMR, respectively. Chemical shifts were referenced to internal
standard TMS. Assignment of spectra was carried out using
NOESY-1D, NOESY-2D, g-COSY, g-HMQC, g-HMBC,
TOCSY, g-HSQC-TOCSY and HMBC-TOCSY experiments.
The NOESY-1D spectra were recorded with the following acqui-
sition parameters: mixing time 800 ms and number of scans 256.
Two-dimensional NMR spectra were acquired using 96 scans
and 128 increments. The g-HSQC-TOCSY and HMBC-TOCSY
experiments were recorded using 256 scans and different mixing
times: 30, 80, 150 and 300 ms. The pulse programs were taken
from the standard Varian pulse sequence library. All spectra were
Fourier transformed with MestReNova 6.2.0 software.
High resolution mass spectrometry (HRMS) was carried out
using the electronic impact technique at 70 eV on a VG Auto-
Spec spectrometer. In some cases a QStar pulsar i spectrometer
with positive electrospray ionization was used.
Commercially available starting materials and solvents were
used without previous purification.
acid
methyl
ester
and
1-butyl-3-methylimidazolium
This journal is © The Royal Society of Chemistry 2012
Green Chem., 2012, 14, 2584–2594 | 2589

View Article Online
Diels–Alder reactions of sorbic acid (1)
column chromatography of silica gel using hexane–ethyl acetate
(5 : 1) as the eluent. Fractions containing product 10 were col-
lected separately and the solvent was removed in vacuo.
A mixture of sorbic acid (1) (5 mmol, 0.56 g) and N-methyl-
maleimide (5) (5 mmol, 0.60 g) in solvent free conditions was
heated under microwave irradiation for 15 min at 100 °C. Purifi-
cation of product 7 was achieved by column chromatography of
silica gel using hexane–ethyl acetate (5 : 1) as the eluent; chan-
ging the solvent to ethyl acetate product 8 was obtained. Frac-
tions containing products 7 and 8 were removed separately
in vacuo and pure products were achieved.
[(1S,4R)-5,5,6,6-Tetracyano-4-methylcyclohex-2-en-1-yl]acetate
1
(10). (1.17 g, 88%). Yellow oil. H-NMR (CDCl3, ppm) δ: 5.96
(1H, ddd, J = 10.8, 3.5, 2.8 Hz, 3-H), 5.74 (1H, dt, J = 10.8, 2.5
Hz, 2-H), 4.58 (1H, dd, J = 12.1, 5.3 Hz, CH2(Ha)), 4.40 (1H,
dd, J = 12.1, 8.4 Hz, CH2(Hb)), 3.50–3.45 (1H, m, 1-H),
3.30–3.24 (1H, m, 4-H), 2.18 (3H, s, CH3CO), 1,66 (3H, d, J =
7.5 Hz, CH3). 13C-NMR (CDCl3, ppm) δ: 170.0 (CH3CO),
128.9 (C-3), 121.0 (C-2), 111.2, 111.1, 109.8, 109.3 (4CN), 62.5
(CH2), 43.0 (C-6), 41.1 (C-1), 39.2 (C-5), 37.5 (C-4), 20.5
(CH3CO), 17.4 (CH3). HR-MS/EI C14H12N4O2: calcd 268.0960;
found 268.0966.
A mixture of 2,4-trans,trans-hexadienyl acetate (2) (5 mmol,
0.72 g) and N-methylmaleimide (5) (5 mmol, 0.60 g) in solvent
free conditions was heated under microwave irradiation for
15 min at 100 °C. The further work-up was carried out as
described above for product 10.
(3aR,4S,7R,7aR)-2,7-Dimethyl-1,3-dioxo-2,3,3a,4,7,7a-hexa-
hydro-1H-isoindole-4-carboxylic acid (7).15 (0.80 g, 72%).
1
Colourless solid. M.p. 176–177 °C. H-NMR (CDCl3, ppm) δ:
9.43 (s, COOH), 5.95 (1H, dd, J = 9.5, 6.6 Hz, 5-H), 5.90 (1H,
dd, J = 9.5, 3.9 Hz, 6-H), 3.83 (1H, dd, J = 6.6, 2.6 Hz, 4-H),
3.66 (1H, dd, J = 9, 2.6 Hz, 3a-H), 3.19 (1H, dd, J = 9, 7.6 Hz,
7a-H), 2.97 (3H, s, NCH3), 2.79–2.76 (1H, m, 7-H), 1.25 (3H,
d, J = 7.3 Hz, CH3). 13C-NMR (CDCl3, ppm) δ: 178.8 and
177.6 (C-1 and C-3), 176.4 (COOH), 136.7 (C-6), 123.8 (C-5),
43.1 (C-7a), 41.8 (C-3a), 40.3 (C-4), 28.9 (C-7), 24.8 (NCH3),
16.6 (CH3). HR-MS/EI C11H13NO4: calcd 223.0845; found
223.0854.
[(3aS,4S,7R,7aR)-2,7-Dimethyl-1,3-dioxo-2,3,3a,4,7,7a-hexa-
hydro-1H-isoindol-4-yl]acetate (11). (1.18 g, 92%). Colourless
1
solid. M.p. 77–78 °C. H-NMR (CDCl3, ppm) δ: 5.75 (2H, m,
(3aR,4S,7S,7aR)-2,7-Dimethyl-1,3-dioxo-2,3,3a,4,7,7a-hexa-
hydro-1H-isoindole-4-carboxylic acid (8). (0.21 g, 18%). Col-
ourless solid. M.p. 175–176 °C. H-NMR (CDCl3, ppm) δ: 9.45
5-H and 6-H), 4.68 (1H, dd, J = 11.2, 7.2 Hz, CH2(Hb)), 4.55
(1H, dd, J = 11.2, 8.4 Hz, CH2(Ha)), 3.24 (1H, dd, J = 9.1, 6.5
Hz, 3a-H), 3.05 (1H, dd, J = 9.1, 7 Hz, 7a-H), 2.89 (3H, s,
NCH3), 2.61–2.59 (1H, m, 7-H), 2.45–2.42 (1H, m, 4-H), 2.09
(3H, s, CH3CO), 1.45 (3H, d, J = 7.3 Hz, CH3). 13C-NMR
(CDCl3, ppm) δ: 177.1 and 177.0 (C-1 and C-3), 170.8
(CH3CO), 135.5 (C-5), 129.0 (C-6), 64.4 (CH2), 44.8 (C-7a),
42.4 (C-3a), 35.8 (C-4), 31.1 (C-7), 24.5 (NCH3), 20.9
(CH3CO), 16.6 (CH3). HR-MS/EI C13H15NO4: calcd 251.1158;
found 251.1143.
A mixture of 2,4-trans,trans-hexadienyl acetate (2) (5 mmol,
0.72 g) and maleic anhydride (6) (5 mmol, 0.50 g) in solvent
free conditions was heated under microwave irradiation for
15 min at 150 °C. The further work-up was carried out as
described above for product 10.
1
(s, COOH), 6.32 (1H, dt, J = 9.5, 3.5 Hz, 5-H), 5.75 (1H, dt, J =
9.5, 3.4 Hz, 6-H), 3.76 (1H, dd, J = 8.8, 5.7 Hz, 3a-H),
3.19–3.16 (1H, m, 4-H), 3.15 (1H, t, J = 8 Hz, 7a-H), 2.91 (s,
3H, NCH3), 2.47–2.45 (1H, m, 7-H), 1.44 (3H, d, J = 7.4 Hz,
CH3). 13C-NMR (CDCl3, ppm) δ: 177.2 and 176.6 (C-1 and
C-3), 175.1 (COOH), 134.8 (C-6), 125.8 (C-5), 43.8 (C-7a),
43.7 (C-3a), 40.1 (C-4), 31.5 (C-7), 24.8 (NCH3), 16.6 (CH3).
HR-MS/EI C11H13NO4: calcd 223.0845; found 223.0854.
A mixture of sorbic acid (1) (5 mmol, 0.56 g) and maleic
anhydride (6) (5 mmol, 0.50 g) in solvent free conditions was
heated under microwave irradiation for 15 min at 150 °C. Purifi-
cation of product 9 was achieved by column chromatography of
silica gel using hexane–ethyl acetate (1 : 1) as the eluent. Frac-
tions containing product 9 were collected separately and the
solvent was removed in vacuo.
[(3aS,4S,7R,7aR)-7-Methyl-1,3-dioxo-1,3,3a,4,7,7a-hexahydro-
isobenzofuran-4-yl]acetate (12). (1.20 g, 97%). Colourless solid.
1
M.p. 111–112 °C. H-NMR (CDCl3, ppm) δ: 5.88–5.86 (2H, m,
(3aS,4S,7R,7aR)-7-Methyl-1,3-dioxo-1,3,3a,4,7,7a-hexahydro-
isobenzofuran-4-carboxylic acid (9).15 (0.98 g, 93%). Colourless
solid. M.p. 174–175 °C. H-NMR (CDCl3, ppm) δ: 5.86 (1H,
dt, J = 10, 2.9 Hz, 5-H), 5.56 (1H, dt, J = 10, 2.9 Hz, 6-H), 3.67
(1H, dc, J = 10.6, 2.9 Hz, 4-H), 3.50 (1H, dd, J = 10.6, 6 Hz,
3a-H), 3.07 (1H, t, J = 6 Hz, 7a-H), 2.56–2.51 (1H, m, 7-H),
1.15 (3H, d, J = 7.5 Hz, CH3). 13C-NMR (CDCl3, ppm) δ: 172.6
and 172.3 (C-1 and C-3), 170.8 (COOH), 131.8 (C-6), 120.1
(C-5), 43.0 (C-7a), 41.1 (C-3a), 39.7 (C-4), 30.6 (C-7), 18.2
(CH3). HR-MS/ESI C10H10O5: calcd 210.0528; found 210.0546.
5-H and 6-H), 4.57 (1H, dd, J = 11.3, 7.8 Hz, CH2(Hb)), 4.49
(1H, dd, J = 11.3, 7.7 Hz, CH2(Ha)), 3.60 (1H, dd, J = 9.4, 6
Hz, 3a-H), 3.38 (1H, dd, J = 9.4, 7.3 Hz, 7a-H), 2.69–2.63 (1H,
m, 4-H), 2.53–2.47 (1H, m, 7-H), 2.09 (3H, s, CH3CO), 1.44
(3H, d, J = 7.4 Hz, CH3). 13C-NMR (CDCl3, ppm) δ: 171.4
(CH3CO), 170.9 and 170.8 (C-1 and C-3), 135.5 (C-5), 129.2
(C-6), 63.5 (CH2), 45.5 (C-7a), 43.1 (C-3a), 35.0 (C-4), 30.4
(C-7), 20.7 (CH3CO), 16.2 (CH3). HR-MS/ESI C12H14O5: calcd
238.0841; found 238.0836.
1
Diels–Alder reactions of conjugated linoleic acid (14–17)
Diels–Alder reactions of 2,4-trans,trans-hexadienyl acetate (2)
General procedure. The corresponding conjugated linoleic
acids (14–17) (0.75 mmol, 0.210 g) and maleic anhydride (6)
(1.5 mmol, 0.155 g) in solvent free conditions were heated under
microwave irradiation for 15 min at 150 °C. After the mixture
had cooled to room temperature diethyl ether (50 mL) was added
A mixture of 2,4-trans,trans-hexadienyl acetate (2) (5 mmol,
0.72 g) and tetracyanoethylene (3) (5 mmol, 0.65 g) in solvent
free conditions was heated under microwave irradiation for
15 min at 80 °C. Purification of product 10 was achieved by
2590 | Green Chem., 2012, 14, 2584–2594
This journal is © The Royal Society of Chemistry 2012

View Article Online
and the organic layer was washed several times with water, in
order to eliminate the excess of maleic anhydride. The organic
layer was separated and dried with anhydrous sodium sulphate.
After filtration the solvent was removed in vacuo. Purification of
cycloadducts 18 to 21 was achieved by column chromatography
of silica gel using hexane–ethyl acetate (5 : 1) as the eluent. Frac-
tions containing pure products were collected separately and the
solvent was removed in vacuo.
trans-octadecadienoic acid (17). (0.224 g, 40%). Yellow oil.
1H-NMR (CDCl3, ppm) δ: 5.82 (2H, s, 5′-H and 6′-H), 3.35
(2H, dd, J = 4.5, 2 Hz, 3′a-H and 7′a-H), 2.35 (2H, t, J = 7.5 Hz,
2-H), 2.28–2.20 (2H, m, 4′-H and 7′-H), 1.91–1.82 (2H, m, 1′′-
H), 1.81–1.74 (2H, m, 9-H), 1.68–1.60 (2H, m, 3-H), 1.44–1.40
(4H, m, 8-CH2 and 2′′-CH2), 1.39–1.22 (12H, m, 4–7 and 3′′-4′′-
CH2), 0.91 (3H, t, J = 7 Hz, 5′′-H). 13C-NMR (CDCl3, ppm) δ:
178.7 (COOH), 171.5 and 171.4 (C-1′ and C-3′), 133.9 and
133.8 (C-5′ and C-6′), 44.9 (C-3′a), 44.8 (C-7′a), 36.4 (C-4′),
36.3 (C-7′), 33.7 (C-2), 31.7 (C-3′′), 30.9 (C-9), 30.6 (C-1′′),
29.4 (C-7), 29.1 (C-6), 29.1 (C-5), 29.0 (C-4), 27.9 (C-7), 27.7
(C-2′′), 24.6 (C-3), 22.6 (C-4′′), 14.1 (C-5′′). HR-MS/EI
C22H34O5: calcd 378.2416; found 378.2445.
8-[(3aS,4R,7R,7aR)-7-Hexyl-1,3-dioxo-1,3,3a,4,7,7a-hexahydro-
isobenzofuran-4-yl]octanoic acid (18). From 9-cis,11-trans-octa-
decadienoic acid (14). (0.231 g, 43%). Yellow oil. 1H-NMR
(CDCl3, ppm) δ: 5.93–5.85 (2H, m, 5′-H and 6′-H), 3.34 (1H,
dd, J = 9.5, 7 Hz, 7′a-H), 3.13 (1H, dd, J = 9.5, 3.5 Hz, 3′a-H),
2.74–2.67 (1H, m, 4′-H), 2.62–2.56 (1H, m, 7′-H), 2.36 (2H, t,
J = 7.5 Hz, 2-H), 1.70–1.61 (4H, m, 3-CH2 and 1′′-CH2 (NOE
assigned)), 1.60–1.52 (2H, m, 8-H), 1.50–1.27 (16H, m, 4–7 and
2′′-5′′-CH2), 0.88 (3H, t, J = 7 Hz, 6′′-H). 13C-NMR (CDCl3,
ppm) δ: 179.6 (COOH), 174.4 and 171.8 (C-1′ and C-3′), 131.5
(C-5′), 131.2 (C-6′), 45.3 (C-3′a), 44.1 (C-7′a), 35.0 (C-4′), 34.6
(C-8), 33.9 (C-2), 33.6 (C-7′), 31.6 (C-4′′), 29.2 (C-1′′), 29.1
(C-3′′), 29.0 (C-6), 29.0 (C-5), 28.8 (C-4), 27.8 (C-7), 27.2
(C-2′′), 24.5 (C-3), 22.5 (C-5′′), 14.0 (C-6′′). HR-MS/EI
C22H34O5: calcd 378.2416; found 378.2446.
Linoleic acid (13) isomerization
Linoleic acid (13) was isomerized to 9-cis,11-trans and 10-
trans,12-cis conjugated linoleic acids (14 and 15) slightly modi-
fying the previous work of Chin et al.2 Linoleic acid (13)
(0.87 mmol, 0.25 g) was mixed with KOH (12 mmol, 0.78 g) in
2.5 mL of ethylene glycol and refluxed at 160 °C for 15 min
under microwave irradiation and an argon atmosphere. To the
reaction mixture, 5 mL of methanol were added followed by
acidification with 20 mL of HCl 3 N. The reaction mixture was
extracted with hexane (3 × 5 mL), and then the organic layer
was washed thrice with 30% methanol in water and three times
with distilled water. Subsequently, it was dried over anhydrous
sodium sulphate and hexane was removed in a rotary evaporator
under vacuum. A colourless liquid was obtained corresponding
with the mixture of 9c,11t-CLA (14) and 10t,12c-CLA (15) in
equal proportions (0.23 g, 90%).
9-[(3aS,4S,7S,7aR)-1,3-Dioxo-7-pentyl-1,3,3a,4,7,7a-hexahydro-
isobenzofuran-4-yl]nonanoic acid (19). From 10-trans,12-cis-
1
octadecadienoic acid (15). (0.229 g, 43%). Yellow oil. H-NMR
(CDCl3, ppm) δ: 5.92–5.85 (2H, m, 5′-H and 6′-H), 3.35 (1H,
dd, J = 9.5, 7 Hz, 3′a-H), 3.14 (1H, dd, J = 9.5, 3.5 Hz, 7′a-H),
2.75–2.68 (1H, m, 7′-H), 2.61–2.57 (1H, m, 4′-H), 2.35 (2H, t,
J = 7.5 Hz, 2-H), 1.68–1.59 (4H, m, 3-CH2 and 9-CH2 (NOE
assigned)), 1.57–1.52 (2H, m, 1′′-H), 1.45–1.28 (16H, m, 4–8
and 2′′-4′′-CH2), 0.89 (3H, t, J = 7 Hz, 5′′-H). 13C-NMR
(CDCl3, ppm) δ: 179.8 (COOH), 174.4 and 171.8 (C-1′ and
C-3′), 131.4 (C-5′ and C-6′), 45.3 (C-7′a), 44.1 (C-3′a), 35.1
(C-7′), 34.6 (C-1′′), 33.9 (C-2), 33.6 (C-4′), 31.6 (C-3′′), 31.7
(C-9), 29.4 (C-7), 29.1 (C-6), 29.1 (C-5), 29.0 (C-4), 28.9 (C-8),
27.7 (C-2′′), 24.6 (C-3), 22.5 (C-4′′), 14.0 (C-5′′). HR-MS/EI
C22H34O5: calcd 378.2416; found 378.2442.
9-cis,11-trans Octadecadienoic acid (14). Colourless oil.
1H-RMN (CDCl3, ppm) δ: 6.32–6.26 (1H, m, 11-H), 5.94 (1H,
t, J = 11 Hz, 10-H), 5.66 (1H, dt, J = 14.5, 7 Hz, 12-H), 5.29
(1H, dt, J = 11, 7.5 Hz, 9-H), 2.35 (2H, t, J = 7.5 Hz, 2-H), 2.15
(2H, ddd, J = 14.5, 7.5, 1.5 Hz, 8-H), 2.09 (2H, c, J = 7 Hz, 13-
H), 1.63 (2H, q, J = 7.5 Hz, 3-H), 1.42–1.23 (16H, m, 4–7-CH2
and 14–17-CH2), 0.88 (3H, t, J = 7 Hz, 18-H). 13C-RMN
(CDCl3, ppm) δ: 179.7 (C-1), 134.8 (C-12), 129.9 (C-9), 128.7
(C-10), 125.5 (C-11), 34.0 (C-2), 32.9 (C-13), 31.7 (C-16), 29.6
(C-7), 29.4 (C-14), 29.1 (C-15), 29.0 (C-5), 29.0 (C-6), 28.9
(C-4), 27.6 (C-8), 24.6 (C-3), 22.6 (C-17), 14.1 (C-18).
8-[(3aS,4S,7R,7aR)-7-Hexyl-1,3-dioxo-1,3,3a,4,7,7a-hexahydro-
isobenzofuran-4-yl]octanoic acid (20). From 9-trans,11-trans-
1
octadecadienoic acid (16). (0.222 g, 40%). Yellow oil. H-NMR
(CDCl3, ppm) δ: 5.82 (2H, s, 5′-H and 6′-H), 3.37 (2H, dd, J =
4.5, 2 Hz, 3′a-H- and 7′a-H), 2.36 (2H, t, J = 7.5 Hz, 2-H),
2.27–2.20 (2H, m, 4′-H and 7′-H), 1.90–1.81 (2H, m, 1′′-H),
1.80–1.72 (2H, m, 8-H), 1.64 (2H, q, J = 7 Hz, 3-H), 1.47–1.40
(4H, m, 7-CH2 and 2′′-CH2), 1.39–1.32 (8H, m, 4–6 and 3′′-
CH2), 1.31–1.29 (4H, m, 4′′ and 5′′-CH2), 0.89 (3H, t, J = 7 Hz,
6′′-H). 13C-NMR (CDCl3, ppm) δ: 179.8 (COOH), 171.6 and
171.5 (C-1′ and C-3′), 133.9 and 133.7 (C-5′ and C-6′), 44.9
(C-3′a), 44.8 (C-7′a), 36.3 (C-4′), 36.2 (C-7′), 33.9 (C-2), 31.7
(C-4′′), 30.6 (C-8), 30.5 (C-1′′), 29.2 (C-3′′), 29.1 and 29.0 (C-5
and C-6), 28.9 (C-4), 27.9 (C-7), 27.8 (C-2′′), 24.5 (C-3), 22.6
(C-5′′), 14.0 (C-6′′). HR-MS/EI C22H34O5: calcd 378.2416;
found 378.2444.
10-trans,12-cis
Octadecadienoic
acid
(15). Colourless
oil.1H-RMN (CDCl3, ppm) δ: 6.32–6.26 (1H, m, 11-H), 5.94
(1H, t, J = 11 Hz, 12-H), 5.65 (1H, dt, J = 14.5, 7 Hz, 10-H),
5.30 (1H, dt, J = 11, 7.5 Hz, 13-H), 2.35 (2H, t, J = 7.5 Hz,
2-H), 2.15 (2H, ddd, J = 9, 7.5, 1.5 Hz, 14-H), 2.09 (2H, c, J =
7 Hz, 9-H), 1.63 (2H, q, J = 7.5 Hz, 3-H), 1.41–1.25 (16H, m,
4–8-CH2 and 15–17-CH2), 0.89 (3H, t, J = 7 Hz, 18-H).
13C-RMN (CDCl3, ppm) δ: 179.5 (C-1), 134.6 (C-10), 130.1
(C-13), 128.6 (C-12), 125.6 (C-11), 33.9 (C-2), 32.8 (C-9), 31.5
(C-16), 29.4 (C-8), 29.3 (C-15), 29.2 (C-7), 29.1 (C-5), 29.1
(C-6), 29.0 (C-4), 27.7 (C-14), 24.6 (C-3), 22.5 (C-17), 14.0
(C-18).
Isomers 9-trans,11-trans and 10-trans,12-trans conjugated
linoleic acids (16 and 17) were achieved using a common
method of cis/trans isomerization catalyzed with iodine.16 To a
9-[(3aS,4S,7R,7aR)-1,3-Dioxo-7-pentyl-1,3, 3a,4,7,7a-hexa-
hydroisobenzofuran-4-yl]nonanoic acid (21). From 10-trans,12-
This journal is © The Royal Society of Chemistry 2012
Green Chem., 2012, 14, 2584–2594 | 2591

View Article Online
solution of the previously obtained mixture of isomers 9-cis,11-
trans and 10-trans,12-cis conjugated linoleic acids (14 and 15)
(0.35 g, 1.25 mmol) in 80 mL of hexane, a small amount of
iodine crystals was added. The mixture was stirred for 5 h at
room temperature and sunlight. The reaction mixture was
washed with Na2SO4 0.1 M (3 × 30 mL) to remove iodine, then
the organic layer was washed thrice with distilled water and
dried over anhydrous sodium sulphate and hexane was removed
in vacuo. A white solid was obtained corresponding with the
mixture of 9t,11t-CLA (16) and 10t,12t-CLA (17) in equal pro-
portions (0.26 g, 75%).
1-butyl-3-methylimidazolium chloride (27) and 1-hexyl-3-
methylimidazolium chloride (28), was heated with microwaves
for 30 min at 140 °C under an argon atmosphere. After com-
pletion of the reaction, the mixture was cooled at room tempera-
ture and then extracted with ethyl ether or ethyl acetate (3 ×
10 mL), and the ionic liquid was recovered in order to be used in
the next cycle. The organic layer was washed with water, dried
over anhydrous sodium sulphate and solvent was removed
in vacuo. The same procedure was used for ricinoleic acid
methyl ester. In all cases, yields were determined by 1H-NMR.
(R,Z)-13-Hexyloxacyclotridec-10-en-2-one
(29). 1H-RMN
9-trans,11-trans Octadecadienoic acid (16). White solid.
1H-RMN (CDCl3, ppm) δ: 6.07–5.95 (2H, m,10-H and 11-H),
5.64–5.50 (2H, m, 9-H and 12-H), 2.35 (2H, t, J = 7.5 Hz, 2-H),
2.05 (4H, c, J = 7.2 Hz, 8-CH2 and 13-CH2), 1.64 (2H, q, J =
7.5 Hz, 3-H), 1.44–1.20 (16H, m, 4–7-CH2 and 14–17-CH2),
0.89 (3H, t, J = 7 Hz, 18-H). 13C-RMN (CDCl3, ppm) δ: 179.5
(C-1), 132.5 (C-12), 132.2 (C-9), 130.4 (C-10), 130.3 (C-11),
33.9 (C-2), 32.6 (C-13), 32.5 (C-8), 31.7 (C-16), 29.4 (C-7),
29.3 (C-14), 29.1 (C-15), 29.0 (C-5), 29.0 (C-6), 28.9 (C-4),
24.6 (C-3), 22.6 (C-17), 14.1 (C-18).
(CDCl3, ppm) δ: 5.48–5.41 (1H, m, 9-H), 5.37–5.28 (1H, m,
10-H), 4.88 (1H, q, J = 6.2 Hz, 12-H), 2.32 (2H, t, J = 7.5 Hz,
2-H), 2.27 (2H, t, J = 7.5 Hz, 11-H), 2.07 (2H, c, J = 6.8 Hz,
8-H), 1.59 (2H, q, J = 7.5 Hz, 3-H), 1.57–1.49 (2H, m, 13-H),
1.42–1.21 (18H, m, 4–7-CH2 and 14–17-CH2), 0.88 (3H, t, J =
6.1 Hz, 18-H). 13C-RMN (CDCl3, ppm) δ: 173.6 (C-1), 132.3
(C-9), 124.1 (C-10), 73.6 (C-12), 34.5 (C-11), 33.9 (C-2), 33.4
(C-13), 31.6 (C-16), 29.6 (C-7), 29.4 (C-15), 29.0 (C-5), 29.0
(C-6), 28.9 (C-4), 27.2 (C-8), 25.2 (C-14), 24.9 (C-3), 22.4
(C-17), 13.9 (C-18).
1-Hexyl-3-methylimidazolium hexafluorophosphate (24) was
synthesized according to the method of Leadbeater et al.23 On
the other hand, 1-hexyl-3-methylimidazolium bromide (26) was
obtained using the procedure of Varma and Namboodiri24
slightly modified. A mixture of 1-methylimidazole (0.165 g,
2 mmol) and 1-bromohexane (0.383 g, 2.2 mmol) was heated
under microwave irradiation at 80 °C for 15 min. After com-
pletion of the reaction, the mixture was washed with ethyl
acetate and dried under vacuum.
10-trans,12-trans Octadecadienoic acid (17). White solid.
1H-RMN (CDCl3, ppm) δ: 6.06–5.96 (2H, m, 11-H and 12-H),
5.61–5.52 (2H, m, 10-H and 13-H), 2.35 (2H, t, J = 7.5 Hz, 2-
H), 2.04 (4H, c, J = 7.2 Hz, 9-CH2 and 14-CH2), 1.64 (2H, q, J
= 7.5 Hz, 3-H), 1.45–1.22 (16H, m, 4–8-CH2 and 15–17-CH2),
0.88 (3H, t, J = 7 Hz, 18-H). 13C-RMN (CDCl3, ppm) δ: 179.9
(C-1), 132.5 (C-13), 132.3 (C-10), 130.5 (C-11), 130.4 (C-12),
33.9 (C-2), 32.6 (C-14), 32.5 (C-9), 31.4 (C-16), 29.4 (C-8),
29.2 (C-15), 29.1 (C-7), 29.0 (C-5), 29.0 (C-6), 28.9 (C-4), 24.6
(C-3), 22.5 (C-17), 14.0 (C-18).
1-Hexyl-3-methylimidazolium bromide (26) (0.416 g,
85%). Colourless viscous liquid. 1H-RMN (CDCl3, ppm) δ:
9.98 (1H, s, 2-H), 7.70 (1H, d, J = 1.8 Hz, 4-H), 7.54 (1H, d, J
= 1.8 Hz, 5-H), 4.33 (2H, t, J = 7.5 Hz, 6-H), 4.11 (3H, s, 12-
H), 1.91 (2H, q, J = 7 Hz, 7-H), 1.38–1.26 (6H, m, 8,9 and 10-
CH2), 0.87 (3H, t, J = 7 Hz, 11-H).13C-RMN (CDCl3, ppm) δ:
136.6 (C-2), 123.6 (C-4), 121.8 (C-5), 49.7 (C-6), 36.4 (C-12),
30.7 (C-9), 29.9 (C-7), 25.5 (C-8), 22.0 (C-10), 13.6 (C-11).
General procedure for the linoleic acid isomerization in
flow. The manifold system (pumps, valves, PFA tubing and
reactor) of a Vapourtec R2+R4 unit was dried with isopropyl
alcohol (2 mL min−1, 15 min) and ethylene glycol (0.5 mL
min−1, 15 min). For the previous optimization of the reaction
conditions, a solution of the linoleic acid (13) (0.16 M) in ethy-
lene glycol was loaded into a sample loop (2 mL) and a solution
of KOH (0.44 M) in ethylene glycol was loaded into a second
sample loop (2 mL). The two sample loops were switched in-
line into streams of ethylene glycol, each flowing at 0.167 mL
min−1 and mixed in the T-mixer. The mixture was then heated in
the reactor (10 mL) at 180 °C and 30 min of residence time, and
the output was collected. For the scale-up of the process, the
concentration was increased 10 times and a longer residence
time was needed for complete conversion (45 min). Only
23.5 ml of ethylene glycol was needed for the production of 1 g
of CLAs. The further work-up was carried out as described
above for the alkali isomerization.
Computational studies
All calculations included in this paper were carried out using the
GAUSSIAN 09 series programs,25 with the standard 6-31G*
basis set.26 In order to include electron correlation at a reason-
able computational cost, density functional theory (DFT)27 was
used. In this study, calculations were carried out by means of the
three-parameter functional developed by Becke et al., which is
usually denoted as B3LYP.28 Zero-vibrational energies (ZPVEs)
were computed at the B3LYP/6-31G* level and were not scaled.
Solvent effects were estimated by means of polarization conti-
nuum models (PCM),29 using chloroform as the solvent. Macro-
Model v.930 was used in its MCMM (Macromodel Monte Carlo
Multiple Minimum) mode of conformational search in which a
Monte Carlo method31 is used to modify the input structures by
random changes in the torsion angles and/or molecular positions.
The OPLS_2005 force field32 as implemented in MacroModel
was used for energy calculation. MacroModel v.9 was run in
Ricinoleic acid (22) dehydration
A mixture of ricinoleic acid (22) (0.33 mmol, 100 mg) and the
corresponding ionic liquid (1 mmol), such as 1-butyl-3-methyl-
imidazolium hexafluorophosphate (23), 1-hexyl-3-methyl-imida-
zolium hexafluorophosphate (24), 1-butyl-3-methyl-imidazolium
bromide (25), 1-hexyl-3-methylimidazolium bromide (26),
2592 | Green Chem., 2012, 14, 2584–2594
This journal is © The Royal Society of Chemistry 2012

View Article Online
4 A. S. Demir and F. N. Talpur, J. Agric. Food Chem., 2010, 58, 1646–
1652.
“perform automatic setup” mode, and the number of steps was
set in 1000.
5 Advances in Conjugated Linoleic Acid Research, ed. J. L. Sebedio,
W. W. Christie, R. Adlof, AOCS Press, Champaign, Illinois, 2003, vol. 2;
A. Ruiz-Rodriguez, G. Reglero and E. Ibañez, J. Pharm. Biomed. Anal.,
2010, 51, 305–326; O. Berdeaux, S. Fontagné, E. Sémon, J. Velasco,
J. L. Sébédio and C. Dobarganes, Chem. Phys. Lipids, 2012, 165, 338–
347.
Conclusions
We herein report a sustainable and efficient methodology for the
identification of different positional and geometrical CLA
isomers in a mixture, by NMR analysis of Diels–Alder cyclo-
adducts. Maleic anhydride was found to be the most appropriate
dienophile to react with CLAs (diene) in solvent-free micro-
wave-assisted Diels–Alder cycloadditions due to the high stereo-
selectivity and high reaction yields obtained. Note that it is not
possible to differentiate among CLA isomers in a food matrix by
NMR because of their low abundance and presence of other
compounds which alter the CLA isomers chemical shift and
overlap signals, however, the Diels–Alder cycloadducts show up
differences in the 13C NMR chemical shifts. Importantly, NMR
allows the identification of both, positional and geometrical iso-
merism, in contrast to the reported analytical tools. Previous
reports containing a complete NMR characterization, including
all the methylene groups, of the CLA isomers could not be
found, reporting here their first 1H- and 13C-NMR data. To
develop this methodology, the synthesis of CLA is required.
Thus, microwave-assisted organic synthesis and continuous-flow
techniques are employed, enabling a reduction of the reported
reaction times. The use of ionic liquids, a reduced amount of sol-
vents and the simplicity of the isolation procedure make this
method a green alternative for the preparation of this kind of
compound. The continuous-flow process provides a good pro-
duction rate for CLA isomers, valuable and important com-
pounds from the biological point of view. Work is underway to
broaden the applicability of this methodology for the identifi-
cation of CLA isomers on a complex matrix (food, drugs).
6 M. S. F. Lie Ken Jie, M. K. Pasha and M. S. Alam, Lipids, 1997, 32,
1041–1044; A. L. Davis, G. P. McNeill and D. C. Caswell, Chem. Phys.
Lipids, 1999, 97, 155–165; M. S. F. Lie Ken Jie, Eur. J. Lipid Sci.
Technol., 2001, 103, 594–632.
7 V. Spitzer, F. Marx and K. Pfeilsticker, J. Am. Oil Chem. Soc., 1994, 71,
873–876.
8 G. Dobson, J. Am. Oil Chem. Soc., 1998, 75, 137–142; M. J. T. Reaney,
Y. D. Liu and W. G. Taylor, J. Am. Oil Chem. Soc., 2001, 78, 1083–1087.
9 G. Read, N. R. Richardson and D. G. Wickens, Analyst, 1994, 119, 393–
396.
10 V. Polshettiwar, M. N. Nadagouda and R. S. Varma, Aust. J. Chem.,
2009, 62(1), 16–26; V. Polshettiwar and R. S. Varma, Acc. Chem. Res.,
2008, 41(5), 629–639; A. K. Bose, M. J. Manhas, B. K. Banik and
E. W. Robb, Res. Chem. Intermed., 1994, 20, 1; S. Caddick, Tetrahedron,
1995, 52, 10403; S. A. Galema, Chem. Soc. Rev., 1997, 25, 233;
P. Lidström, J. Tierney, B. Wathey and J. Westman, Tetrahedron, 2001,
57, 9225; C. O. Kappe, Angew. Chem., Int. Ed., 2004, 43, 6250; B.
L. Hayes, Aldrichim. Acta, 2004, 37, 66; M. V. Gomez, R. Caballero,
E. Vazquez, A. Moreno, A. de la Hoz and A. Diaz, Green Chem., 2007,
9, 331–336; R. B. N. Baig and R. S. Varma, Chem. Soc. Rev., 2012, 41,
1559–1584.
11 B. L. Hayes, Microwave Synthesis: Chemistry at the Speed of Light,
CEM, Matthews, NC, 2002; Microwave-Assisted Organic Synthesis, ed.
P. Lidström, J. P. Tierney, Blackwell Scientific, Oxford, 2005;
C. O. Kappe and A. Stadler, in Microwaves in Organicand Medicinal
Chemistry, ed. R. Mannhold, H. Kubinyi, G. Folkers, Wiley, Weinheim,
2005, vol. 25; Microwaves in Organic Synthesis, ed. A. Loupy, Wiley-
VCH, Weinheim, 2nd edn, 2006.
12 F. Langa, P. de la Cruz, A. de la Hoz, A. Díaz-Ortiz and E. Díez-Barra,
Contemp. Org. Synth., 1997, 4, 373–386; A. de la Hoz, A. Díaz-Ortiz,
A. Moreno and F. Langa, Eur. J. Org. Chem., 2000, 3659; A. Díaz-Ortiz,
J. M. Fraile, A. de la Hoz, M. V. Gómez, J. A. Mayoral, A. Moreno,
P. Prieto, L. Salvatella, A. Saiz and E. Vázquez, Eur. J. Org. Chem.,
2001, 2891–2899; A. de la Hoz, A. Díaz-Ortiz, A. Moreno and F. Langa,
in Microwaves in Cycloadditions in Microwaves in Organic Synthesis, ed.
A. Loupy, Wiley-VCH, Weinheim, 2002, ch. 9; A. Díaz-Ortiz, A. de la
Hoz and A. Moreno, Curr. Org. Chem., 2004, 8, 903–918; K. Bougrin,
M. Soufiaoui and G. Bashiardes, in Microwaves in Cycloadditions in
Microwaves in Organic Synthesis, ed. A. Loupy, Wiley-VCH, Weinheim,
2nd edn, 2006, ch. 11.
Acknowledgements
The authors acknowledge financial support from the Junta de
Comunidades de Castilla-La Mancha (project PBI08-0259-
7224), and from the Spanish Ministry of Science and Innovation
(project CTQ2011-22410). M. M. acknowledges Junta de
Comunidades de Castilla-La Mancha for her F.P.I
fellowship. M. V. G. acknowledges Marie Curie Reintegration
Grants, and Albacete Science and Technology Park for the
INCRECYT research contract. Technical support from High
Performance Computing Service of University of Castilla-La
Mancha is gratefully acknowledged. Bio serae Laboratories
S. A. are also acknowledged for providing CLA.
13 A. de la Hoz, A. Díaz-Ortiz, J. M. Fraile, M. V. Gómez, J. A. Mayoral,
A. Moreno, A. Sainz and E. Vázquez, Synlett, 2001, 753–757;
A. Moreno, M. V. Gómez, E. Vázquez, A. de la Hoz, A. Díaz-Ortiz,
P. Prieto, J. A. Mayoral and E. Pyres, Synlett, 2004, 1259–1263.
14 P. T. Anastas and J. C. Warner, Green Chemistry: Theory and Practice,
Oxford University Press, New York, 1988; P. T. Anastas and
T. C. Williamson, Green Chemistry: Frontiers in Benign Chemical Syn-
thesis and Process, Oxford University Press, New York, 1988;
D. Dallinger and C. O. Kappe, Chem. Rev., 2007, 107, 2563–2591;
T. Razzarq and C. O. Kappe, ChemSusChem, 2008, 1, 123–132;
R. S. Varma, Green Chem., 2008, 10, 1129–1130; M. J. Hernáiz,
A. R. Alcántara, J. I. García and J. V. Sinisterra, Chem.–Eur. J., 2010, 16,
9422–9437.
15 G. D. Khandelwal and B. L. Wedzicha, Food Chem., 1997, 60, 237–244;
U. Biermann, W. Butte, T. Eren, D. Haase and J. O. Metzger, Eur. J. Org.
Chem., 2007, 3859–3862.
16 K. Eulitz, M. P. Yurawecs, N. Sehat, J. Fritsche, J. A. G. Roach,
M. M. Mossoba, J. K. G. Kramer, R. O. Adolf and Y. Ku, Lipids, 1999,
34, 873–877.
17 R. Leach, Molecular Modelling: Principles and Applications, Prentice
Hall, 2nd edn, 2001.
18 J. Wegner, S. Ceylan and A. Kirschning, Chem. Commun., 2011, 47,
4583–4592.
19 M. V. Gomez, H. H. J. Verputten, A. Diaz-Ortiz, A. Moreno, A. de la
Hoz and A. H. Velders, Chem. Commun., 2010, 46, 4514–4516.
20 Microreactors: New Technology for Modern Chemistry, ed. W. Ehrfield,
V. Hessel and H. Löwe, Wiley-VCH, Weinheim, 2000; B. P. Mason,
Notes and references
1 Y. L. Ha, J. Storkson and M. W. Pariza, Cancer Res., 1990, 50, 1097–
1101; C. Ip, S. F. Chin, J. A. Scimeca and M. W. Pariza, Cancer Res.,
1991, 51, 6118–6124; M. W. Pariza, Am. J. Clin. Nutr., 2004, 79(suppl),
1132S–1136S; R. Kumar, A. Bhatia and D. Arora, J. Clin. Diagn. Res.,
2009, 3, 1953–1967; Y. Park, J. Food Comp. Anal., 2009, 22S, S4–S12.
2 S. F. Chin, W. Liu, J. M. Storkson, Y. L. Ha and M. W. Pariza, J. Food
Comp. Anal., 1992, 5, 185–197.
3 L. Yang, Y. Huang, H. Q. Wang and Z. Y. Chen, Chem. Phys. Lipids,
2002, 119, 23–31.
This journal is © The Royal Society of Chemistry 2012
Green Chem., 2012, 14, 2584–2594 | 2593

View Article Online
K. E. Price, J. L. Steinbacher, A. R. Bogdan and D. T. McQuade, Chem.
Rev., 2007, 107, 2300–2318; H. R. Sahoo, J. G. Kralj and K. F. Jensen,
Angew. Chem., 2007, 119, 5806–5810; H. R. Sahoo, J. G. Kralj and
K. F. Jensen, Angew. Chem., Int. Ed., 2007, 46, 5704–5708; C. Wiles and
P. Watts, Eur. J. Org. Chem., 2008, 1655–1671; Microreactors in Organic
Synthesis and Catalysis, ed. T. Wirth, Wiley-VCH, Weinheim, 2008;
T. Illg, P. Lög and V. Hessel, Bioorg. Med. Chem., 2010, 18, 3707–3719;
R. L. Hartman, J. P. McMullen and K. F. Jensen, Angew. Chem., Int. Ed.,
2011, 50, 7502–7519; J. M. de Muñoz, J. Alcazar, A. de la Hoz and
A. Diaz-Ortiz, Tetrahedron Lett., 2011, 52, 6058–6060; J. M. de Muñoz,
J. Alcazar, A. de la Hoz and A. Diaz-Ortiz, Eur. J. Org. Chem., 2012,
260–263.
S. S. Iyengar, S. J. Tomasi, M. Cossi, N. Rega, N. J. Millam, M. Klene,
J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts,
R. E. Stratmann, O. Yazyev, O. A. J. Austin, R. Cammi, C. Pomelli,
J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski,
G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels,
O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski and D. J. Fox, Gaus-
sian 09, (Revision A.1), Gaussian, Inc., Wallingford, CT, 2009.
26 R. Ditchfield, W. J. Hehre and J. A. Pople, J. Chem. Phys., 1971, 54,
724–728; W. J. Hehre, R. Ditchfield and J. A. Pople, J. Chem. Phys.,
1972, 56, 2257–2261; P. C. Hariharan and J. A. Pople, Theor. Chim.
Acta, 1973, 28, 213–222; P. C. Hariharan and J. A. Pople, Mol. Phys.,
1974, 27, 209–215; M. S. Gordon, Chem. Phys. Lett., 1980, 76, 163–
168.
21 T. Tanaka, M. Hosokawa, Y. Yasui, R. Ishigamori and K. Miyashita,
Int. J. Mol. Sci., 2011, 12, 7495–7509.
22 R. Kumar, A. Sharma, N. Sharma, V. Kumar and A. K. Sinha,
Eur. J. Org. Chem., 2008, 5577–5582.
23 N. E. Leadbeater, H. M. Torenius and H. Tye, Tetrahedron, 2003, 59,
2253–2258.
27 R. G. Parr and W. Yang, Density-Functional Theory of Atoms and Mol-
ecules, Oxford University Press, New York, 1994.
28 W. Kohn, A. D. Becke and R. G. J. Parr, Phys. Chem., 1996, 100,
12974–12980; A. D. Becke, J. Chem. Soc., 1993, 98, 5648–5652;
A. D. Becke, Phys. Rev. A, 1988, 38, 3098–3100.
24 R. S. Varma and V. V. Namboodiri, Chem. Commun., 2001, 643–644.
25 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb,
J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson,
H. Nakatsuji, M. Caricato, X. Li, X. H. P. Hratchian, A. F. Izmaylov,
J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota,
R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao,
H. Nakai, T. Vreven, Jr. J. A. Montgomery, J. E. Peralta, F. Ogliaro,
M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov,
R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, A. J. C. Burant,
29 S. Miertus, E. Scrocco and J. Tomasi, J. Chem. Phys., 1981, 55, 117–
129; B. Mennucci and J. Tomasi, J. Chem. Phys., 1997, 106, 5151–5158;
R. Cammi, B. Mennucci and J. Tomasi, J. Phys. Chem. A, 2000, 104,
5631–5637.
30 MacroModel, version 9, Schrödinger, LLC, New York, NY, 2007.
31 N. Metropolis, A. W. Rosenbluth, M. N. Rosenbluth, A. H. Teller and
E. Teller, J. Chem. Phys., 1953, 21, 1087–1092.
32 W. L. Jorpensen, D. S. Maxwell and J. Tirado-Rives, J. Am. Chem. Soc.,
1996, 118, 11225–11235.
2594 | Green Chem., 2012, 14, 2584–2594
This journal is © The Royal Society of Chemistry 2012