Metathesis of cardanol over ammonium tagged Hoveyda-Grubbs type catalyst supported on SBA-15

Article abstract of DOI:10.1016/j.cattod.2017.09.049

Cardanol (a mixture of phenols with C₁₅ aliphatic unsaturated chain in meta position, manufactured from cashew nut shell liquids) is an abundant and till now practically waste natural material. An important way leading to its utilization as a r

Full text of DOI:10.1016/j.cattod.2017.09.049

Accepted Manuscript
Title: Metathesis of cardanol over ammonium tagged
Hoveyda-Grubbs type catalyst supported on SBA-15
ˇ
ˇˇ
´
˚
Authors: Hynek Balcar, Nadezda Zilkova, Martin Kubu,
´ˇ ˇ´
´
Miroslav Polasek, Jirı Zednık
PII:
S0920-5861(17)30661-2
DOI:
Reference:
https://doi.org/10.1016/j.cattod.2017.09.049
CATTOD 11054
To appear in:
Catalysis Today
Received date:
Revised date:
Accepted date:
19-5-2017
4-9-2017
24-9-2017
ˇ
ˇˇ
´
˚
Please cite this article as: Hynek Balcar, Nadezda Zilkova, Martin Kubu,
´ˇ
ˇ´
´
Miroslav Polasek, Jirı Zednık, Metathesis of cardanol over ammonium
tagged Hoveyda-Grubbs type catalyst supported on SBA-15, Catalysis
Today https://doi.org/10.1016/j.cattod.2017.09.049
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

METATHESIS OF CARDANOL OVER AMMONIUM TAGGED HOVEYDA-
GRUBBS TYPE CATALYST SUPPORTED ON SBA-15
Hynek Balcar,a, Naděžda Žilková,a Martin Kubů,a Miroslav Polášek,a Jiří Zedníkb
a J. Heyrovský Institute of Physical Chemistry, v.v.i., Academy of Sciences of the Czech
Republic, Dolejškova 3, Prague 8, CZ-182 23, Czech Republic, b Department of Physical and
Macromolecular Chemistry, Faculty of Science, Charles University in Prague, Hlavova 8
2030, CZ-128 43, Prague 2, Czech Republic.
Corresponding author: Hynek Balcar, e-mail: hynek.balcar@jh-inst.cas.cz
Graphical abstract
Highlights
Ammonium tagged Hoveyda-Grubbs catalyst on SBA-15 was effective in cardanol
metathesis.
The catalyst was also used in cross-metathesis with alkenes and unsaturated esters.
The Ru leaching was low and the catalyst was used in continuous flow reactor.

Abstract
Cardanol (a mixture of phenols with C15 aliphatic unsaturated chain in meta position,
manufactured from cashew nut shell liquids) is an abundant and till now practically waste
natural material. An important way leading to its utilization as a renewable source of valuable
phenolic compounds is represented by olefin metathesis.
Quaternary ammonium tagged Ru complex of Hoveyda-Grubbs type immobilized on SBA-15
mesoporous molecular sieve by non-covalent interactions was proved as a very efficient
catalyst for cardanol metathesis under mild conditions (35°C, toluene, 0.1 mol% of catalyst).
The catalyst was used also in cross-metathesis of cardanol with alkenes (1-hexene, 1-heptene,
3-hexene, 4-octene) and unsaturated esters (methyl acrylate, 1,4-diacetoxybutene-2) under
high yields of cross-metathesis products. The catalysts exhibited low leaching (from 0.2 ppm
to 2.9 ppm Ru in products) and can be used in a continuous flow reactor (cumulative TON =
2500).
Keywords: cardanol, olefin metathesis, Hoveyda-Grubbs type catalyst, immobilized catalyst,
SBA-15, flow chemistry.
1. Introduction
Cardanol is an important natural product received from cashew nut shell liquid (CNSL)[1,2].
Cashew nut tree (Anacardium occidentale L.) is growing in tropical regions (Brazil, Vietnam,
Nigeria, India, Cote d’Ivoire) and the world cashew nut crop is more than 3 millions tons per
year[1]. Cashew nut shell (about 25% of weight) is normally discarded as a waste. CNSL is
produced from nut shells either by solvent extraction or roasting process [3]. Solvent
extracted CNLS contains mainly anacardic acid, CNSL from roasting process is rich in
cardanol (at high temperature anacardic acid decarboxylates). Cardanol then is obtained by
vacuum distillation as a pale yellow viscous liquid and contained a small amount of cardol
(Fig. 1). Cardanol itself is a mixture of four m-substituted phenols with C15 aliphatic chains
of different degree of unsaturation (3,2,1,0 isolated C=C bonds in chain). Similar mixture of
resorcin analogues is called cardol. The composition of cardanol varies in some extent and
depends on cashew nut origin and method of CNSL extraction.

OH
OH
OH
COOH
R
R
HO
R
cardanol
anacardic acid
cardol
R =
Fig. 1. Main constituents of CNSL.
Till recently cardanol has been used for production of resins and polymer materials. In last
decade, its importance as a renewable source of phenolic compounds was recognized and
new chemical transformations leading to valuable products have been studied [1,2]. In
addition to cardanol hydrogenation, epoxidation, and carbonylation, olefin metathesis has
been investigated especially for synthesis of new fine chemicals [4-10]. Among other
important products, 3-(non-8-enyl)phenol prepared by cardanol ethenolysis [9] has attracted
attention as a precursor molecule for environmental friendly detergents [2].
Metathesis of cardanol proceeds smoothly with Grubbs-, and Hoveyda-Grubbs type Ru
catalysts under mild conditions. Protection of phenol OH group is not necessary. However,
relatively high amounts of Ru catalysts were used (usually 1 mol % , in some case even 5
mol% [6], in the case of cardanol ethenolysis [9] exceptionally 0.05 mol % of Ru catalyst was
used). It is a result of low catalyst stability under conditions applied (catalyst decomposition
by reaction with cardanol and impurities in it, inherent catalyst instability). Another problem
is connected with homogeneous character of the reaction, i.e. separation of catalyst residues
from reaction mixture. This may be solved by immobilization of Ru catalysts on solid
supports. Mesoporous molecular sieves (e.g. SBA-15) and two-dimensional zeolites have
been proved as advantageous support for immobilization of originally homogeneous catalysts
[11,12]. Grubbs second generation catalyst and Hoveyda-Grubbs type Zhan 1B catalyst
immobilized on SBA-15 were successfully tested in cardanol metathesis and cardanol cross-
metathesis with ethylene and diacetoxybutene [13]. Cardanol was used without special
purification and high conversions were achieved under mild conditions (batch reactor,
toluene, 35-60°C, 1-2 mol% of catalyst, 2h of reaction time). Nevertheless, with regards to
cardanol properties, a low possibility of its further purification, and the desirability of scaling
up this process, searching for more efficient and stable catalysts is still necessary.
Recently, Hoveyda-Grubbs type catalysts with ammonium tags were developed and
immobilized on SBA-15 by non-covalent interactions [14,15]. These catalysts showed

excellent effectivity in metathesis reaction of various substrates, and especially the catalyst
with diisopropylphenyl substituents on NHC ligand 1 (Fig.2) exhibited high robustness
allowing flow reactor application.
The goal of this contribution is the investigation of the Hoveyda-Grubbs type catalysts with
ammonium tags immobilized on the SBA-15 in cardanol metathesis and cross-metathesis
with various co-substrates. We followed activity and selectivity of these catalysts under both
batch and flow conditions. According to our best knowledge, cardanol metathesis in flow
conditions has not been reported till now.
DIPP = 2,6-diisopropylphenyl
DIPP
DIPP
1
Fig. 2. Catalyst 1 (AquaMet SIPr).
2. Materials and methods
2.1. Materials
Purely siliceous SBA-15 mesoporous molecular sieve (particle size about 1μm) was
synthesized as described elsewhere [16] using a tetraethoxysilane (TEOS; 98%, Aldrich) as a
silica source and triblock copolymer Pluronic P123 (EO20PO70EO20; BASF) as a structure-
directing agent. Catalyst AquaMet SIPr (1) was purchased from Apeiron Synthesis S.A.
(Poland). Toluene (for HPLC, HiPerSolv Chromanorm, VWR Prolabo Chemicals) and
dichloromethane (for HPLC, LachNer, Czech Rep.) were purified using a Pure Solv PS-MD-
7 Solvent Purification System (Innovative Technology, Amesbury, MA). Toluene was stored
over molecular sieves 4A under Ar. Dichloromethane was degassed and stored in vacuo in

dark. 1-Hexene (Fluka, purum), 1-heptene (Fluka, purum), cis-3-hexene (Alfa Aesar), trans-
4-octene (Aldrich) were purified by metallic Na, cis-1,4-diacetoxy-2-butene (96%, Alfa
Aesar), methyl acrylate (99%, Aldrich) and nonane (99%, Sigma-Algrich) were dried by
molecular sieve (Fluka). Cardanol was purchased from Shanghai Boyle Chemical Co., Ltd.
China, stored in cold and dark and used without further purification.
2.2. Catalyst preparation
About 1 g of SBA-15 was dried in a Schlenk tube in vacuo at 300°C for 3 h. Then 10-20 mL
dichloromethane and calculated amount of 1 were added under Ar. After 30 min stirring the
solid phase turned to slightly green and the supernatant became colorless. The supernatant
was removed by filtration and the solid phase was washed 3 times by 10 mL
dichloromethane. Finally, the solid catalyst was dried in vacuo at room temperature.
Catalysts containing 0.09, 0.5, and 1.0 wt% of Ru [1/SBA-15 (0.09), 1/SBA-15 (0.5), and
1/SBA-15 (1.0)] have been prepared.
N2 adsorption isotherms (77K, ASAP 2020, Micromeritics) reveal that mesoporous character
and narrow pore size distribution of SBA-15 support were preserved in the catalyst (see
Figure S1 and S2 in Supplementary material). The comparison of textural parameters of
SBA-15 and 1/SBA-15 (1.0) is given in Table 1. Due to the low Ru concentration (0.07
Ru/nm2) the decrease in void volume is small and in average pore diameter practically
negligible. It is in accord with previous findings [14].
Table 1. Textural parameters of parent support and 1/SBA-15 (1.0) catalyst.
SBET (m2/g)
864
V (cm3/g)
1.10
D (nm)
5.6
SBA-15
1/SBA-15 (1.0)
699
0.99
5.5
SBET = BET surface area, V = void volume, D = average pore diameter.
2.3.Metathesis experiments and product analysis
The metathesis in batch reactors were carried out under Ar using standard Schlenk technique.
Cardanol (together with a co-substrate in case of cross-metathesis) was added to the catalyst
suspension in toluene under stirring (800 rpm). The samples of reaction mixture were
collected at given reaction times, the catalyst was removed by centrifugation and samples
were analyzed by GC, GC-MS and HPLC-MS.
The flow experiments were carried out on a bench-top continuous-flow reactor X-cube
(Thales Nano Inc., Budapest, Hungary). Catalyst (about 50 mg) was placed into the cartridge
- reactor in an atmosbag under Ar. The toluene solution of cardanol, co-substrate (if any) and
nonane (GC internal standard) was pumped from the stock Schlenk tube to the reactor heated
to the required temperature. Flow rate was 0.1 mL/min. The liquid on the outlet was collected
and analyzed by GC.

A high resolution gas chromatograph Agilent 6890 with a DB-5 column (length: 50 m, inner
diameter: 320 μm, stationary phase thickness: 1 μm), equipped with a 7683 Automatic Liquid
Sampler and a FID detector was used for product analysis. Temperature parameters: inlet
temperature 250°C, detector temperature 250°C, temperature program 100°C (hold 5 min),
10°C/min to 150°C, 20°C/min to 200°C (hold 3 min), 30°C/min to 290°C (hold 10 min),
30°C/min to 325°C (hold 10 min), 20°C/min to 350°C. Flow gas: helium, initial pressure:
150kPa, constant flow 2.8 ml/min. Sample volume was 1 μl, split ratio 25:1. n-Nonane was
used as an internal standard. Individual products were identified by GC–MS
(ThermoFinnigan, FOCUS DSQ II single Quadrupole) and HPLC–MS.
HPLC-MS analyses were performed using a Waters 1525μ binary HPLC pump connected to
a Waters 2487 dual λ absorbance UV detector and a Quattro Premiere XE tandem quadrupole
mass spectrometer (Waters). An AcclaimTM 120 C18 column (130 × 3 mm, 3 μm particle
size) was used. HPLC separation of products was achieved utilizing a binary gradient elution
with acetonitrile with 0.05% of formic acid (component A) and water with 0.05% of formic
acid (component B) controlled by the following program: 0 min. – 80% A; 20 min. – 100%
A; 28 min. – 100% A. A flow rate of 0.5 mL/min was used. The samples were injected using
a Rheodyne valve furnished with 5 μL sample loop.
During the HPLC analyses, full mass spectra were recorded in the range of m/z 200 - 800
with scan duration of 2 s. The electrospray ion source was operated in negative ion mode
(ESI−), and thus (M−1)− ions, i.e. substituted phenolates were produced, mass-analyzed and
detected. The ESI− settings were as follows: capillary voltage −2.7 kV, cone voltage 50 V,
source temperature 100°C, desolvation temperature 200°C, desolvation gas (N2) flow 300
L/h. For quantitative HPLC analysis, the standard addition method was used in connection
with UV detector set at 280 nm. A mixture of cis- and trans-1,16-bis(3-
hydroxyphenyl)hexadecene was used as an internal standard (for preparation and
characterization of these compounds see chapter “Preparation and characterization of product
10” in Supplementary material).
Cardanol conversions were determined from GC (using an internal standard). Conversion (C)
was calculated according to Eq. 1 and yield (Yi) of product i according to Eq.2.
C(%) = 100(1- ccard /c0
)
card
(1)
(2)
Yi (%) = 100(ci/ c0
)
card
The experimental error in determination of cardanol conversion was ±2% (absolute) for all
metathesis and cross-metathesis reactions.
Preparative thin-layer chromatography (TLC) was performed on Aldrich PLC Silica Gel 60
F254 1 mm preparative plates using a mixture of hexanes : dichloromethane : diethylether / 6 :
3 : 1 as a mobile phase.

NMR spectra were recorded on a Varian UNITYINOVA 400 instrument (proton frequency 400
MHz) or Varian 300 NMR System in CD2Cl2 or CDCl3. Chemical shifts ( scale, ppm) were
obtained by first order analysis. Spectra were referenced to the solvent signal  (CD2Cl2) =
5.32 ppm,  (CDCl3) = 7.25 ppm for 1H NMR and solvent signal (CD2Cl2) = 54.00 ppm, 
(CDCl3) = 77.00 ppm for 13C NMR.
The Ru content in catalysts and Ru leaching into the reaction mixtures were determined by
ICP-MS at the Institute of Analytical Chemistry, Institute of Chemical Technology, Prague.
Solid samples containing silica were decomposed in a microwave decomposition unit with
the mixture of HF and HNO3. For details see ref. [17].
3. Results and discussion
3.1. Metathesis of cardanol in batch reactor
HPLC-MS analysis revealed that in cardanol obtained from the supplier and used in all
following experiments cardanol components (see Fig.1) were present in the molar ratio 3-
(pentadeca-8,11,14-trienyl)phenol (2) : 3-(pentadeca-8,11-dienyl)phenol (3) : 3-(pentadec-8-
enyl)phenol (4) : 3-(pentadecyl)phenol = 42.8 : 18.2 : 36.4 : 2.6. In addition to this the
cardanol sample contained a certain amount of cardols impurities (8.6%). With regard to the
practical applications we worked with cardanol as obtained without any purification.
Moreover, the cardol impurities and their metathesis products were not taken into accounts.
Fig.3 shows conversion curves for cardanol metathesis over 1/SBA-15 (0.09), and 1/SBA-15
(1.0) (50 mg and 5 mg, respectively) at cardanol/Ru molar ratio 1000. Rapid onset of the
reaction and high conversions are visible in both cases. The conversions achieved with
1/SBA-15 (0.09) and 1/SBA-15 (1.0) are practically the same. It is to expect because of the
same substrate and Ru concentrations in the reaction mixture. The difference in the weight
amounts of catalysts used and different concentrations of Ru centers on the catalyst surface
have no visible effect under conditions applied.
The experiments with the same weight amounts (10 mg) of catalysts 1/SBA-15 (0.09),
1/SBA-15 (0.5), and 1/SBA-15 (1.0), and 0.5 mmol of cardanol in the same solvent volume
(Fig. S3 in Supplementary material) showed only slightly lower conversions for 1/SBA-15
(0.5) in comparison with 1/SBA-15 (1.0) at the early stages in accord with the lower Ru
concentration in the system. The conversions after prolonged reaction time were the same.
On the other hand very low conversions were achieved with 1/SBA-15 (0.09) indicating
very rapid deactivation of catalyst by impurities and/or by substrate in high excess. The
cardanol/Ru ratio seems to be the main factor influencing the results of cardanol metathesis
(see below).

100
80
60
40
20
0
0
50
100
150
300
350
time, min
Fig. 3. Cardanol conversion vs. time for 1/SBA-15 (0.09) - () and 1/SBA-15 (1.0) - ().
Toluene, 35°C, cardanol/Ru = 1000, initial cardanol concentration c0card = 0.13 mol/L.
Fig. 4 shows the decrease in HPLC peak areas corresponding to individual cardanol
components during metathesis reaction. Assuming the same extinction coefficient (UV
detection at 208 nm) we can compare the consumption rates of individual components. The
initial rate (i.e. the reduction of peak area in the first 3 min) decreased in the order 2 > 3 > 4.
After 300 min of the reaction, the components 2 and 3 were consumed nearly completely,
however, about 40% of the original amount of component 4 persisted in the reaction mixture.
This difference in consumption rates cannot be explained by different initial concentrations of
cardanol components (initial concentrations of 2 and 4 do not differ too much, initial
concentration of 4 is higher than that of 3) but it reflects the participation of individual
components in different metathesis reactions (see below).
100000
80000
60000
40000
20000
0
0
50
100
150
200
250
300
reaction time, min

Fig. 4. Changes of HPLC peak areas of cardanol components 2 ( ), 3 (), 4 () during
metathesis reaction. 1/SBA-15 (1.0), toluene, 35°C, cardanol/Ru molar ratio = 1000, c0
=
card
0.13 mol/L.
Theoretically, individual cardanol components can react in many self-metathesis and cross-
metathesis reactions, giving rise numerous different alkenes and phenolic compounds.
However, only some of them were found in considerable amounts. The reaction products
contained (see a typical HPLC chromatogram in Fig.S4 and GC chromatogram in Fig.S5 in
Supplementary material): 1,4-cyclohexadiene (5), 4-undecene (6), 7-tetradecene (7), 3-(non-
8-enyl)phenol (8), 3-(dodec-8-enyl)phenol (9), and 1,16-bis(3-hydroxyphenyl)hexadecene
(10). The yields of selected products (calculated from GC chromatograms) in the course of
metathesis reaction over 1/SBA-15 (0.09) and 1/SBA-15 (1.0) are displayed in Fig.5. For
1/SBA-15 (0.5) a similar product development was observed (Fig.S6 in Supplementary
material). The common features of these developments are (i) rapid formation of 5 in yields
corresponding to the consumption of 2; (ii) rapid formation of 8, however in considerably
lower yields than the yields of 5 , (iii) formation of 9 (yield about 7%), (iv) very low yields
of alkenes 6 and 7 even at high cardanol conversion (3%, 2%, respectively). The low yield of
7 is rather surprising, because the yield of 10 (calculated from HPLC-MS), was about 15%.
Assuming 10 to be a product of self-metathesis of 4 [5], 7 and 10 should be formed in
approximately same amounts.
40
35
30
25
20
15
10
5
0
0
50
100
150
200
250
300
time, min
Fig. 5. Time development of selected products during cardanol metathesis over 1/SBA-15
(0.09) - () and 1/SBA-15 (1.0) - (). Toluene, 35°C, cardanol/Ru = 1000, initial cardanol

concentration c0
= 0.13 mol/L. Products: 5 (black), 6 (green), 7 (red), 8 (blue), 9
card
(magenta).
To explain the product formation in a real cardanol mixture under conditions applied we set
up a tentative scheme of the main reactions proceeding in this system (Scheme 1). A rapid
consumption of 2 and formation of 5 in the short reaction times indicates the monomolecular
self-metathesis of 2 (reaction (a) in Scheme 1). The second product of this self-metathesis
reaction 8 was partially consumed in a consecutive reaction. As bisphenol compound 10 was
found in a considerable amount even at the early stages of the reaction (when 7 was not
observed), we suggest 8 reacted with 4 and 3 in cross-metathesis reactions (b) in Scheme 1.
Hydrocarbons C8H16 and C8H14 were found in GC-MS, however, their quantification was not
possible. Bimolecular self-metathesis of 4 - reaction (c) in Scheme 1 - seems to be strongly
limited (maybe for steric reasons in coordination to the catalyst): corresponding tetradecene 7
was formed in very low amounts. Similarly, undecene 6 was formed in a low yield in cross-
metathesis reaction (d) in Scheme 1. Product 9 - 3-(dodec-8-enyl)phenol - was formed in
cross-metathesis of 3 and 4 - reaction (e) of Scheme 1.
It may be noticed, that the position of double bond in compounds in Scheme 1 is assumed in
accord with metathesis mechanism. The double bond shift isomerization during metathesis in
some extent cannot be excluded. Cis-trans isomerization during metathesis reaction was
observed: starting cardanol is practically all cis-, during reaction trans-isomers were
identified among cardanol components as well as among the reaction products.

Scheme 1. The formation of the main metathesis products in cardanol system.
To asses the possibility of higher TONs achievements, the reaction was performed using
different substrate/Ru molar ratios (Fig.6). With increasing cardanol/Ru ratio TON increased
till cardanol/Ru ratio 2500, when conversion 64% and the highest TON = 1600 were
achieved. Further increase in the cardanol/Ru ratio led to the decrease in both final

conversion and TON. Catalyst deactivation either by reaction with cardanol impurities or
with cardanol itself occurred. In accord with the proposed reaction mechanism (Scheme 1)
the yield of product 8 increased with increasing cardanol/Ru ratio up to cardanol/Ru = 2500
from 13% to 20% and the yield of 5 was kept on 39% (practically maximum possible yield).
TON = 1600 is several times higher than that achieved in cardanol metathesis over Ru
catalysts previously [13].
TON=405
80
60
40
20
0
TON=1000
TON=1600
TON=1000
0
50
100
150
200
250
time, min
Fig. 6. Cardanol conversion curves at cardanol/Ru molar ratios = 500 (), 1250 (), 2500
() and 4000 (). 1/SBA-15 (0.09), toluene 1.2 mL, 35°C, c0card = 0.21mol/L.
3.2.Metathesis of cardanol in continuous flow reactor
Temperature 60 °C and cardanol concentration in the feed c0card = 0.15 mol/L were chosen as
the optimal reaction conditions for cardanol metathesis in a flow system. For 1/SBA-15
(0.09) it was found (see Table 2, No 4,5 and Fig. S7 in Supplementary material) that 50°C led
to low initial conversion, on the other hand the temperature increase to 75°C caused very fast
catalyst deactivation similarly as the doubled cardanol concentration in the feed. The effect
of Ru content in catalyst is presented in Fig. 7 and Table 2, No 1-3. Cardanol conversion vs.
time plots for 1/SBA-15 (0.09), 1/SBA-15 (0.5), and 1/SBA-15 (1.0) clearly show that the
catalyst stability is limited: after a certain period the cardanol conversion fell down. This
period increased with Ru content in catalyst in the order 1/SBA-15 (0.09) < 1/SBA-15 (0.5)
< 1/SBA-15 (1.0). The average cardanol conversion (calculated from the cardanol
conversion vs. time curve as cumulative amount of cardanol consumed in the experiment and
the total amount of cardanol entered the reactor) increased in the same order, cumulative
TON (calculated as the cumulative amount of cardanol consumed in the experiment divided
by the amount of Ru in catalyst in the cartridge – both quantities in mol) in reverse order.
1/SBA-15 (1.0) delivered the highest average cardanol conversion (59%), 1/SBA-15(0.09)
the highest cumulative TON (2500 and/or 2800). These TONs are higher than the highest
TON achieved in batch reactor (1600).

80
60
40
20
0
0
50
100
150
200
250
300
350
400
time, min
Fig. 7. Cardanol conversion vs. time for 1/SBA-15 (0.09) (), 1/SBA-15 (0.5) (), and
1/SBA-15 (1.0) (). Flow reactor, toluene, 0.1mL/min, 60°C, c0card = 0.15 mol/L.
Table 2. Average cardanol conversions, cumulative TONs, and Ru leaching in flow system.
No catalyst
T ,
°C
c0
mol/L
,
Average
conversion,
CA, %
Cumulative Ru leaching a, b
TON
card
1
2
3
4
5
1/SBA-15 (0.09) 60
0.15
0.15
0.15
0.30
0.15
30
41
59
22
31
2500
960
1.3%, 0.2 ppm
1.0%, 1.3 ppm
1.5%, 2.9 ppm
_
1/SBA-15 (0.5)
1/SBA-15 (1.0)
60
60
800
1/SBA-15 (0.09) 60
1/SBA-15 (0.09) 75
2800
200
7.0%, 1.7 ppm
Toluene, 0.1 mL/min, CA=100 ∫(ccard-c0card).dt/c0card.t (t = experiment time), a in % of
original Ru content in catalyst, b in ppm related to the original cardanol amount.
Catalyst leaching of 1/SBA-15 (0.09), 1/SBA-15 (0.5), and 1/SBA-15 (1.0) in these
experiments (determined from the collected liquid phase) increased with the increasing
amounts of Ru in the catalysts from 0.2 ppm to 2.9 ppm with respect to the starting amount of
cardanol (Table 2). Corresponding losses of Ru in catalysts were from 1.0 to 1.5% at 60°C
and 7% at 75°C. Therefore, the reduction of Ru content in catalysts cannot be responsible for

the continuous loss of their activity. On the other hand, the negligible concentration of Ru in
a liquid phase ensured the acceptable level of product contamination by catalyst residues.
3.3. Cardanol cross-metathesis
Cross-metathesis of cardanol with linear alkenes and with unsaturated esters were studied
both in batch and flow reactors. The results achieved in batch reactor are collected in Table
3. Cross-metathesis with symmetrical alkenes can be used for changing the lengths of
aliphatic chain in cardanol molecule [10]. Using these reactions, phenol with 11 carbon atom
chain (product 12, eq. 3) can be prepared from cardanol and 3-hexene, and similarly phenol
with 12 carbon atom chain (product 9, eq. 4) from cardanol and 4-octene. Product 9 may
arise also from cardanol metathesis according Scheme 1(e) (see above). If alkene is used in
excess, phenols 9 and 12 are produced in high yields (see Table 3). On the other hand,
complementary hydrocarbons C10 and C11 (see Scheme S1 and S2 and Figs S8 and S9 in
Supplementary material) are formed in significantly lower amounts than 9 and 12. Negligible
amounts of phenol 8 indicates its possible reactions with 3-hexene and 4-octene under
formation of 9 and 12.
(3)
(4)
1-hexene and 1-heptene gave rise to phenols with aliphatic chains having 13 and 14 carbon
atoms, respectively (eqs 5 and 6). However, their yields were only 25 and 27%, respectively,
which is connected with a consumption of considerable amounts of both 1-alkene in self-
metathesis reactions leading to 5-decene and 6-dodecene, respectively (see Schemes S3 and
S4 and Figs S10 and S11 in Supplementary material). There is an additional difference in
behavior of 1-akenes in cross-metathesis with cardanol: product 8 was found in high amounts
(yields of 46 and 52%, respectively). These amounts are higher than that of 2 in original
cardanol, which suggests that 8 must be formed also in a reaction of 1-alkenes with 3 or 4.
Thus cross-metathesis with 1-alkenes may represent an alternative way to product 8 in
addition to cardanol ethenolysis [9,13].

(5)
(6)
Despite its electron-deficient double bond methyl acrylate reacted smoothly under mild
reaction conditions and delivered expected phenol derivative 15 (eq.7) in 45% yield (see
Table 3, No 5). Long chain ester 16 was formed in a significantly lower amount as well as
product 8 (see Scheme S5 and Fig.S12 in Supplementary material). Product 15 was isolated
by TLC and characterized by NMR and MS (see Fig.S13, Supplementary material).
Changing 1/SBA-15 (0.09) for 1/SBA-15 (0.5) (Table 3, No 6) higher cardanol conversion
and higher yield of product 15 were achieved in a shorter reaction time (1h) as expected.
TON however decreased more than 4 times.
Cis-1,4-diacetoxybut-2-ene (DAB) contains two ester groups in the molecule. It might be the
reason for a rapid catalyst deactivation and rather low yield of product 17 under conditions
applied (eq. 8, Table 3, No 7). Aliphatic acetate 18 was formed even in a lower amout (see
Scheme S6 and Fig. S14 in Supplementary material). Product 17 was isolated and
characterized similarly as product 15 (see Fig.S15, Supplementary material).
(7)
(8)

Table 3. Cardanol cross-metathesis in batch reactor
Cardanol
conversion,
C, %
Selected
products
Yield of
products, Y i , %
selected
Co-substrate
Reaction
time, h
3-hexene
5
93
72
94
90
86
95
42
Y12 = 68
Y8 = 1
Y9= 43
Y8 = 3
Y13= 25
Y8 = 46
Y14 = 27
Y8 = 52
Y15 = 45
Y8 = 5
1
2
3
4
5
6
7
12
8
9
8
13
8
14
8
15
8
15
8
17
8
4-octene
5
1-hexene
5
1-heptene
3.5
5
methyl acrylate
methyl acrylate*
cis-1,4-diacetoxybut-2-ene
1
Y15= 69
Y8= 5
Y17 = 8
Y8 = 14
5
1/SBA-15 (0.09), toluene, 35°C, co-substrate/cardanol/Ru molar ratio = 2000/500/1, c0
=
card
0.21 mol/L. * 1/SBA-15 (0.5) , co-substrate/cardanol/Ru molar ratio = 400/100/1.
Cross-metathesis of cardanol with selected co-substrates was also carried out in a flow
system. 1/SBA-15 (0.5) and temperature 60°C were used. 1/SBA-15 (0.5) was chosen
because fast catalyst deactivation was observed for 1/SBA-15 (0.09) in some cases. For
example in the system cardanol + 3-hexene (molar ratio1:4) cardanol conversion only 11%
was achieved. Table 4 summarizes the main results. Average cardanol conversion and
cumulative TON were calculated as in the previous section. From the conversion vs. time plot
(Fig. S16 in Supplementary material) it is seen that cross-metathesis with symmetrical
alkenes proceeded in a similar way as metathesis of cardanol. Especially for 3-hexene, high
TON (1550) and high yield of product 12 (49%) were achieved.
For methyl acrylate and DAB, however, the catalyst deactivation was faster than for cross-
metathesis with alkenes and therefore the TONs and yields fell down. For cross-metathesis
with methyl acrylate (Table 4, No 3) Ru leaching was established as 4.5% with respect to the
Ru catalyst content and 4.9 ppm with respect to cardanol. This leaching is significantly higher
than that found for cardanol metathesis under same conditions (Table 2), indicating a possible
effect of substrate polarity on Ru leaching. Nevertheless, this leaching can be still considered
as small, confirming strong bonding interaction between 1 and SBA-15 support.

Table 4. Cardanol cross-metathesis in flow system
Average cardanol Cumulative
Cumulative
yield of the
key product, %
No co-substrate
Time,
h
conversion,
TON
CA, %
3-hexene
4.5
5.7
4.25
5.7
72
50
47
28
1550
Y12 = 49
1
2
3
4
4-octene
1250
905
Y9 = 29
Methyl acrylate
cis-1,4-diacetoxybut-2-ene
Y15= 11
Y17 = 23
540
1/SBA-15 (0.5), toluene, 60°C, c0
= 0.15 mol/L, co-substrate/cardanol molar ratio = 4,
card
flow rate 0.1 mL/min, CA=100 ∫(ccard-c0card).dt/c0card.t (t = experiment time), cumulative TON
with regard to cardanol.
4. Conclusions
Quaternary ammonium tagged Ru complex of Hoveyda-Grubbs type (AquaMet SIPr)
immobilized on SBA-15 mesoporous molecular sieve by non-covalent interactions was
proved as efficient catalyst for cardanol metathesis under mild conditions (35°C, toluene, 0.1
mol% of catalyst). Cardanol was used without any purification as a mixture of components.
Metathesis reaction proceeded rapidly and high cardanol conversion (more than 80%) was
achieved.
Time development of concentrations of individual cardanol components as well as of selected
products during metathesis allowed us to set up a tentative scheme of reactions proceeding in
this complex system under applied reaction conditions. Triply unsaturated cardanol
component reacted very rapidly by self-metathesis under formation of 1,4-cyclohexadiene
and 3-(non-8-enyl)phenol. Then 3-(non-8-enyl)phenol reacted by cross-metathesis with di-
and mono-unsaturated cardanol components under formation of biphenolic compounds.
The catalyst was successfully tested in a flow system (60°C, toluene, 0.1 mL/min). The
cumulative TON achieved was 2500, which surpassed the maximum TON in batch reactor
(1600). The Ru leaching was low: from 1.0 to 1.5% of the original Ru content in catalyst,
which corresponded to the possible product contamination with Ru in the range 0.2 – 2.9
ppm. However, catalyst stability was limited: after certain period cardanol conversion fell
down. Catalyst deactivation probably by impurities or reaction with substrate itself occurred.

The catalyst was tested also in cross-metathesis of cardanol with alkenes (1-hexene, 1-
heptene, 3-hexene, 4-octene) and unsaturated esters (methyl acrylate, 1,4-diacetoxybutene-2).
The cross-metathesis with alkenes effectively changed the length of cardanol aliphatic chain.
The cross-metathesis with esters allowed the introduction of ester group on the end of
cardanol aliphatic chain. Especially in the case of cross-metathesis with methyl acrylate, high
cardanol conversion and high yield of cross-metathesis product were achieved.
Acknowledgements
The authors thank M. Horáček, Z. Flenerová and V. Varga (all from J.Heyrovský Institute,
Prague) for GC-MS analysis and for preparation of compounds cis-/trans-10.
Financial
support from the Czech Science Foundation (project No. 17-01440S) is gratefully
acknowledged.
Appendix: Supplementary material.
References
[1] E. B. Mubofu, Pure Appl. Chem. 88 (2016) 17–27.
[2] V. S. Balachandran, S. R. Jadhav, P. K. Vemula, G. John, Chem.Soc.Rev. 42 (2013) 427-
438.
[3] P. P.Kumar, R. Paramashivappa, P. J. Vithayathil, P. V. S. Rao, A. S. Rao, J. Agric. Food
Chem. 50 (2002) 4705-4708.
[4] Y.-C. Guo, G. Mele, F. Martina, E. Margapoti, G. Vasapollo, W.-J. Xiao, J. Organomet.
Chem. 691 (2006) 5383-5390.
[5] G. Mele, J. Li, E. Margapoti, F. Martina, G. Vasapollo, Catal. Today 140 (2009) 37-43.
[6] G. Vasapollo, G. Mele, R. Del Sole, Molecules 16 (2011) 6871-6882.
[7] J. A. Mmongoyo, Q. A. Mgani, S. J. M. Mdachi, P. J. Pogorzelec, D. J. Cole-Hamilton,
Eur. J. Lipid Sci. Technol. 114 (2012) 1183-1192.
[8] S. Baader, P. E. Podsiadly, D. J. Cole-Hamilton, L. J. Goossen, Green Chem., 16 (2014)
4885-4890.
[9] J. Julis, S. A. Bartlett, S. Baader, N. Beresford, E.J. Routledge, C. S. J. Cazina, D. J. Cole-
Hamilton, Green Chem., 16 (2014) 2846-2856.
[10] J. E. Mgaya, S.A. Bartlett, E. B. Mubofu, Q. A. Mgani, A. M. Z. Slawin, P. J.
Pogorzelec, D. J. Cole-Hamilton, ChemCatChem 8 (2016) 751-757.
[11] H.Balcar, J. Čejka, Coord.Chem.Rev. 257 (2013) 3107-3124.
[12] M. V. Opanasenko, W. J. Roth, J.Čejka, Catal. Sci. Technol., 6 (2016) 2467-2484.
[13] T. Shinde, V. Varga, M. Polášek, M. Horáček, N. Žilková, H. Balcar, Appl. Catal. A:
Gen. 478 (2014) 138-145.
[14] J. Pastva, K. Skowerski, S. J. Czarnocki, N. Žilková, J. Čejka, Z. Bastl, H. Balcar, ACS
Catal. 2014, 4, 3227-3236.
[15] K. Skowerski, J. Pastva, S. J. Czarnocki, J. Janoscova, Org. Process Res. Dev. 19 (2015)
872-877.
[16] P. Topka, H. Balcar, J. Rathouský, N. Žilková, F. Verpoort, J. Čejka, Microporous
Mesoporous Mater. 96 (2006) 44–54.
[17] J. Pastva, J. Čejka, N. Žilková, O. Mestek, M. Rangus, H. Balcar, J. Mol. Catal. A:
Chem. 378 (2013) 184–192.