Increased selectivity in the formation of the phenoxy ether of methyl lesquerolate over chloroalkyl-modified SBA-15-SO3H catalysts

Article abstract of DOI:10.1016/j.apcata.2009.11.002

A novel synthetic catalysis was developed for selective modification of useful bioproducts from the oil of Lesquerella fendlerii, an agricultural commodity of increasing importance. Modification of the surface of the solid acid catalyst SBA-15-SO₃H with the chloroalkyl groups CH₂CH₂Cl, has improved the selectivity in the synthesis of the phenoxy ether of methyl 14-hydroxy-11-eicosenoate (methyl lesquerolate). A series of catalysts, SBA-15(EtCl)_xSO₃H where x = 0, 2, 4, 8, have been prepared and characterized by ¹³C CPNMR, surface properties, and elemental analyses. Selectivity increased with increased loading of the chloroalkyl group. This selectivity can be attributed to the decrease in the rate of conversion of the phenoxy ether to the Friedel-Crafts adducts. This study demonstrates synthetic approaches can develop novel catalysts to increase yields and selectivity for value-added bioproducts from new crops.

Full text of DOI:10.1016/j.apcata.2009.11.002

Applied Catalysis A: General 373 (2010) 90–97
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Increased selectivity in the formation of the phenoxy ether of methyl lesquerolate
§
over chloroalkyl-modified SBA-15-SO H catalysts
3
a,
b
Michael A. Jackson *, Michael Appell
a
New Crops and Processing Unit, National Center for Agricultural Utilization Research, Agricultural Research Service, United States Department of Agriculture,
815 N. University St. Peoria, IL 61604, United States
1
b
Mycotoxin Research Unit, National Center for Agricultural Utilization Research, Agricultural Research Service, United States Department of Agriculture,
815 N. University St. Peoria, IL 61604, United States
1
A R T I C L E I N F O
A B S T R A C T
Article history:
A novel synthetic catalysis was developed for selective modification of useful bioproducts from the oil of
Received 6 July 2009
Lesquerella fendlerii, an agricultural commodity of increasing importance. Modification of the surface of
Received in revised form 23 October 2009
Accepted 2 November 2009
Available online 6 November 2009
the solid acid catalyst SBA-15-SO
3 2 2
H with the chloroalkyl groups CH CH Cl, has improved the selectivity
in the synthesis of the phenoxy ether of methyl 14-hydroxy-11-eicosenoate (methyl lesquerolate). A
13
series of catalysts, SBA-15(EtCl)
x
SO
3
H where x = 0, 2, 4, 8, have been prepared and characterized by
C
CPNMR, surface properties, and elemental analyses. Selectivity increased with increased loading of the
chloroalkyl group. This selectivity can be attributed to the decrease in the rate of conversion of the
phenoxy ether to the Friedel–Crafts adducts. This study demonstrates synthetic approaches can develop
novel catalysts to increase yields and selectivity for value-added bioproducts from new crops.
Published by Elsevier B.V.
Keywords:
Lesquerella
Phenoxy ether
SBA-15-SO
3
H
GIAO/B3LYP
1
. Introduction
Lesquerella fendlerii is a winter annual native to the Southwest-
There is a long history of this chemistry starting shortly after the
original work of Friedel and Crafts in 1877 [2]. Nicolet and DeMilt
[3] synthesized phenylstearate by addition of benzene to oleic acid
using AlCl as catalyst. It would be 7 years before it was suggested
ern United States. This member of the mustard family produces a
small seed which contains about 30% oil which is typically 58%
hydroxyfatty acids with the predominant fatty acid being 11-cis-
3
that the products formed were actually a mixture [4] and another
30 years before gas chromatographic methods showed that there
were 12 (soon to be 15) positional isomers that formed [5]. Studies
of this reaction would continue and include sulfuric acid [6],
methanesulfonic acid [7,8], p-toluenesulfonic acid, clays [9], and
solid acids such as Amberlyst 15 [10] as catalysts. The facile nature
of the alkylation of aromatics with oleic acid is indicated by the
success of all of these catalysts with almost any aromatic
derivative of benzene from toluene, phenol, anisole, cresols,
xylenes, and naphthalenes. Although all acid catalysts and all
aromatics yield the same mix of 12–15 positional isomers, this has
not eliminated interest in this class of compounds as lubricant
additives [11]. We had the expectation that the alcohol group in
methyl lesquerolate could at least serve to limit the number of
positional isomers which could form. This turns out to be partially
true, but when phenol is used, one of the products formed is the
phenoxy ether.
1
4-hydroxyeicosenoic acid [1]. In order for lesquerella to become a
viable new crop, uses for the novel oil, or novel products from this
oil, will need to be developed. The development of new products
will require efficient catalytic conversions with high selectivity in
order to minimize waste from what is already a limited resource.
The purpose of this investigation is to develop novel catalysts that
improve the selectivity in the synthesis of value-added products
from the oils of Lesquerella fendlerii.
Methyl 11-cis-14-hydroxyeicosenoate, hereafter referred to as
methyl lesquerolate, contains both a double bond and alcohol
group as potential reactive sites and we initially focused on adding
an aromatic ring to the double bond by Friedel–Crafts alkylation.
§
Mention of trade names or commercial products in this publication is solely for
There is extensive literature on the synthesis of ethers for use as
gasoline oxygenates [12–14] and longer chain ethers are known to
increase the cetane number in diesel fuel [15–17]. Furthermore,
long chain ethers offer greater lubricity for diesel fuel which will be
the purpose of providing specific information and does not imply recommendation
or endorsement by the U.S. Department of Agriculture.
*
Corresponding author. Tel.: +1 309 681 6255; fax: +1 309 681 6524.
E-mail address: Michael.jackson@ars.usda.gov (M.A. Jackson).
0926-860X/$ – see front matter. Published by Elsevier B.V.
doi:10.1016/j.apcata.2009.11.002

M.A. Jackson, M. Appell / Applied Catalysis A: General 373 (2010) 90–97
91
Table 1
Molar ratios of the reagents used in the catalyst syntheses.
Catalyst
TEOS
MPTMS
VTMS
2 2
H O
SBA-15-SO
3
H
100
100
100
100
100
15.1
5.1
3.0
1.6
3.2
–
212
212
212
212
212
SBA-15(EtCl)
SBA-15(EtCl)
SBA-15(EtCl)
SBA-15(EtCl)
2
4
8
8
SO
SO
SO
3
3
3
H
H
H
10
Scheme 1. Catalyst synthesis.
12.1
13.5
13.5
3 2
(SO H)
in 2N HCl and by grafting the functional groups onto preformed
SBA-15. In a typical cocondensation preparation, 56.8 mmol TEOS,
needed with the continued reduction of sulfur levels in these fuels
18]. Small ethers such as methyl and ethyl tert-butyl ether are
8.6 mmol MPTMS, and 140 mmol H
solution (molar concentrations: TEOS, 0.227; MPTMS, 0.034; H O ,
2 2
O were added to 250 ml P123
[
2 2
prepared by addition of the appropriate alcohol to isobutylene over
solid acid catalysts [14,19] whereas others are prepared by the
dehydration of alcohols to make the symmetrical ethers [15], also
over solid acids.
Tight supplies of biomass will demand the biorefinery be very
efficient. High efficiency can only be accomplished with good
catalyst selectivity especially in the presence of the multiple
functionalities of biomass. Herein, we report on the increased
selectivity toward ether synthesis from methyl lesquerolate and
phenol directed by the surface properties of functionalized SBA-15
catalysts.
0.56) and stirred at 300 rpm and 50 8C for about 18 h and then held
static at 100 8C for 24 h. The solid product was collected by
filtration and washed with warm (45 8C) water. This was repeated
several times by resuspending the solid in warm water followed by
collection on a frit. Finally, the solid was suspended in ethanol and
collected. The product was then dried under vacuum at 60 8C.
Table 1 contains the ratios of reactants used in the syntheses of the
different catalysts. SBA-15-vinyl was prepared by reacting
8.6 mmol VTMS with 57 mmol TEOS in 2N HCl/P123 in the
2 2
absence of H O . In the reactions containing the VTMS, the vinyl
group reacts with HCl in the presence of hydrogen peroxide to give
the chloroethyl group.
2
. Experimental
Grafted catalysts were prepared in 2N HCl or toluene.
Functionalized silanes were added to SBA-15 in quantities to
match those prepared by cocondensation. The SBA-15 used for the
grafted catalyst was prepared at 50 8C, aged at 100 8C, and calcined
at 650 8C (22–650 8C at 5 8C/min hold for 2 h) to remove residual
surfactant. Functionalized products were collected by filtration
and washed with water and ethanol.
2.1. Materials
Reagents were from commercial sources. Tetraethylorthosilica
(TEOS), mercaptopropyltrimethylsilane (MPTMS), vinyltrimethox-
1
ysilane (VTMS), Amberlyst 15, and the surfactant Pluronic P123
were purchased from Sigma–Aldrich (Milwaukee, WI, U.S.A.).
Phenol was from Fisher Scientific (Pittsburgh, PA, U.S.A.).
2.2. Analytical methods
2
.1.1. Preparation of methyl lesquerolate
Textural properties of the catalysts were calculated from
nitrogen isotherms collected at 77 K using a Quantachrome
Autosorb-1. Surface areas were calculated using the multipoint
BET method and pore sizes calculated using the BJH method. For
the chloroalkyl-modified catalysts, the desorption branch of the
isotherms was used for the BJH calculation, whereas for the SBA-
Cold-pressed, refined, bleached, and deodorized lesquerella oil
was prepared in-house and converted to methyl esters by reaction
with sodium methoxide. This mixture of FAMES was distilled to
yield a fraction enriched to 95% in methyl lesquerolate as
previously described [20].
3
15-SO H, the adsorption branch was used [21]. Temperature
2
.1.2. Preparation of catalysts
programmed desorption measurements were also made using the
Quantachrome Autosorb-1 with a thermal conductivity detector.
The following program was used. Samples (ꢀ225 mg) were dried
The propylsulfonic acid-functionalized SBA-15 was prepared
both by cocondensation of TEOS and MPTMS in a 3% P123 solution
Fig. 1. Nitrogen adsorption–desorption profiles and pore size distribution of the functionalized SBA-15 catalysts. ^, SBA-15-SO
5(EtCl) SO H; *, SBA-15(EtCl) SO H; *, SBA-15(EtCl) (SO H)
3 2 3
H; ~, SBA-15(EtCl) SO H; &, SBA-
1
4
3
8
3
8
3
2
.

9
2
M.A. Jackson, M. Appell / Applied Catalysis A: General 373 (2010) 90–97
under flowing helium at 110 8C for 40 min. This was allowed to
cool to 60 8C and ammonia was passed over the samples for
1
0 min. This was followed by a 25 min. Helium purge during which
time the TCD was turned on. The temperature was then ramped
from 60 8C to 600 8C at 10 8C/min. Quantitative analyses of acid
sites on each catalyst were performed in aqueous solution by first
equilibrating the solids in 1 N NaCl followed by titration using a
7
51 GPD Titrino from Metrohm. The titrant was 0.086 M KOH.
Gas chromatographic separation of the reaction mixture was
accomplished on a Hewlett-Packard 6890 using a Supelco SPB1
column (30 m  0.25 mm, 0.25
m
m film thickness) running a
À1
heating profile of 125 8C for 2 min then ramping at 308 min to
3
3
20 8C and holding for 5 min. The detector was a FID operating at
00 8C.
Mass spectra were collected on a Hewlett-Packard 5973N mass
detector operating at 70 eV. Separations were performed as above.
Methyl 11-cis-14-phenoxyeicosenoate mass spectrum m/z (rela-
+
tive intensity): EI 416 (8 M ), 323 (26), 291 (65), 191 (71), 97 (76),
9
4 (80), 55 (100). Friedel–Crafts adducts were characterized as a
+
Fig. 2. NH
SBA-15; (B) SBA-15 pretreated with NH
pretreated with NH (E) SBA-15(EtCl)
15(EtCl) SO H pretreated with NH .
3
temperature programmed desorption profiles from three catalysts. (A)
; (C) SBA-15-SO H; (D) SBA-15-SO
SO pretreated with NH (F) SBA-
group by m/z EI 416 (ꢀ7 M ), 133 (35–75), 107 (100). Mass 107
HOC CH) was used to distinguish Friedel–Crafts adducts from
3
3
3
H
(
6 5
H
3
;
4
3
H
3
;
the phenoxy ether. Side products tentatively identified as the cyclic
8
3
3
ethers 2-methyl decanoyl-5-hexyltetrahydrofuran m/z EI 255 (42),
+
1
2
23 (38), 155 (96), 137 (40), 95 (100), 55 (98). CI (CH
4
) (341 m ).
1
-Methyl decanoyl-3-hexyloxetane m/z EI 255 (14), 223 (14), 155
+
1
(38), 113 (77), 95 (100), 55 (58). CI (CH
4
) (341 m ).
1
13
H and C NMR spectra were obtained on a Bruker ARX-400
1
with a 5 mm dual proton/carbon probe (400 MHz H/100.61 MHz
13
C) using CDCl
4-phenoxyeicosenoate: H NMR:
.45 (s, 1H), 1.6 (m, 4H), 2.0 (m, 2.1H), 2.35 (t, 2.4H), 2.4 (m, 1.2H),
3
as the solvent in all measurements. Methyl 11-cis-
1
1
1
3
1
1
d 0.9 (t, 3.1H), 1.3 (m, 19.3H),
13
.65 (s, 3H), 4.2 (m, 0.8H), 5.45 (m, 1.9H), 7.1 (dd, 4.4H). C NMR:
d
4.1, 22.6, 25.0, 27.5, 29, 31.7, 32.7, 33.7, 34.1, 51.5, 77.9, 116.1,
20.5, 124.6, 125.3, 129.4, 132.5, 133.7, 158.5, and 174.3. The cyclic
13
ethers as a mixture C NMR:
d 78.6, 79.3, 51.4, 36.1, 34.1, 32.1,
3
1.8, 31.0, 29.7, 29.5, 29.4, 29.2, 29.1, 26.2, 25.0, 22.6, and 14.0.
Solid phase CPMAS C NMR spectra were recorded on a Bruker
13
Avance 300 with a rotor speed of 3200 rpm.
Sulfur and chlorine analyses were performed by Galbraith
Laboratories, Knoxville, TN.
2.3. Catalytic conversion of methyl lesquerolate
Reactions were carried out in 2-port 10 ml round-bottomed
flasks fitted with vigreux condensers which were stoppered. The
flasks were charged with catalyst, internal standard (hexadecane),
and 2.0–4.0 ml melted phenol. These were immersed in an oil bath
set at the reaction temperature and 100–500
ml methyl lesquer-
olate were injected at time zero. Fifty microlitres of samples were
withdrawn to follow reaction progress.
2.4. GIAO/B3LYP determination of chemical shifts
Density functional theory calculations were carried out using
Parallel Quantum Solutions (Fayetteville, AR, U.S.A.) hardware and
Table 2
Textural properties of the SBA-15 derived catalysts as prepared by cocondensation.
+
Catalyst
S
BET
2
V
BJH
MPD
H
S
Cl:S
(
m /g) (ml/g) (nm) (meq/g) (meq/g)
SBA-15
860
407
664
448
666
536
1.39
0.89
1.42
1.21
1.35
1.31
8.7
3.7
8.3
8.7
8.3
8.7
NA
NA
NA
SBA-15-SO
3
H
0.96
0.34
0.21
0.08
0.17
1.02
0.43
0.24
0.14
0.26
0.015
1.4
SBA-15(EtCl)
SBA-15(EtCl)
SBA-15(EtCl)
SBA-15(EtCl)
2
4
8
8
SO
SO
SO
3
3
3
H
H
H
3.3
1
3
Fig. 3. CPMAS C NMR spectra showing incorporation of the chloroalkyl group. (A)
6.1
SBA-15-vinyl; (B) SBA-15-EtCl; (C), SBA-15-SO
3 2 3
H; (D) SBA-15(EtCl) SO H. *Marks
(SO
3
H)
2
2.7
residual P123, SSB marks spinning side bands.

M.A. Jackson, M. Appell / Applied Catalysis A: General 373 (2010) 90–97
93
Table 3
software v3.2 [22]. Structures were initially modeled with the
HyperChem 7.52 program (Gainesville, FL, U.S.A.) and PM3 semi-
empirical method [23]. Geometry optimization was performed in
the gas phase using B3LYP/6-31G(d) and B3LYP/6-31G(d,p) levels
of theory and performed on delocalized internal coordinates using
the Eigenvector following Algorithm with the convergence criteria
1
3
Convergence of B3LYP-GIAO C chemical shifts relative to tetramethylsilane for
observed and unobserved substituents at the 6-31G** level of theory.
1
3
d
C
Group
Si–CH
Calculated
Found
2
–CH
2
–Cl
–Cl
46.3
20.5
47
16
À6
À4
set at 1 Â 10 Hartree and a gradient of less than 3 Â 10 a.u.
24]. Gauge including atomic orbital (GIAO) methods were applied
for shielding tensor calculations on geometry optimized structures
25,26]. Results are reported as chemical shifts relative to the
reference standard, tetramethylsilane.
Si–CH
2
–CH
2
[
Si–CH
2
–CH
2
–OH
–OH
58.2
18.2
–
–
Si–CH
2
–CH
2
[
Si–CH5CH
Si–CH5CH
2
120.9
137.3
130
138
2
Si–CH
Si–CH
Si–CH
2
2
2
–CH
–CH
–CH
2
–CH
2
–SO
–SO
3
H
H
59.3
20.6
12.6
54
18
16
3
. Results and discussion
2
–CH
2
3
2 2 3
–CH –SO H
3.1. Characterization of catalysts
Table 4
Scheme 1 shows the chemical conversion which occurs
Selectivity as measured by initial rates of reaction and phenoxyether turnover
frequencies for the catalysts studied.
following cocondensation of the TEOS with MPTMS and VTMS.
In the absence of H the vinyl group does not undergo attack by
2 2
O
Catalyst
Selectivity,
PE TOF
mol/min/H eq)
E
app
the chloride and a vinyl substituted silica results [27]. Fig. 1 shows
the nitrogen adsorption–desorption isotherms and the pore size
distribution plots obtained for the catalysts. These data were
collected from catalysts prepared by cocondensation of the
functionalized silicas with TEOS and after template removal. The
isotherms can be classified as Type IV with hysteresis loops typical
of mesoporous materials. Further refinement of the classification
PE FC
+
S = k /k
(
m
(kJ/mol)
Amberlyst
SBA-15-SO
1.2
2.8
3.2
3.7
5.1
16
27
10
8
74
74
65
55
46
3
H
SBA-15(EtCl)
SBA-15(EtCl)
SBA-15(EtCl)
2
4
8
SO
SO
SO
3
H
3
H
3
H
2
3
places SBA-15-SO H in group IVb and all of the doubly modified
derivatives in group IVa [28]. This indicates that the presence of the
vinyl group, or the chloroalkyl group, affects the assembly of the
SBA-15. Table 2 contains the textural properties which are typical
of SBA-15 derivatives previously published [27,29]. The column
showing the Cl:S ratio is a measure of the chloroalkyl to
the vinyl group to the chloroalkyl group, is incomplete. However,
we will continue to refer to the catalysts by the stoichiometric
ratios used for the syntheses. Fig. 2 shows the results of
temperature programmed desorption experiments. The catalysts
used for these measurements were prepared by grafting in order to
escape interference caused by pyrolysis of the P123 template. Our
propylsulfonic acid ratio. The very low ratio for SBA-15-SO
3
H
points out that there is little residual chloride remaining from the
3
results for SBA-15-SO H are similar to, but notably different from
2
2
N HCl solution used in the syntheses. All of the others are about
0–30% lower than calculated from the synthesis mixture,
those previously published [30]. Specifically, we used samples free
of template and we started our scans at a lower temperature. The
template-free samples result in sharper peak shape from the
indicating that incorporation of the vinyl group, or conversion of
Scheme 2. Synthesis of phenoxy ether and byproducts.

9
4
M.A. Jackson, M. Appell / Applied Catalysis A: General 373 (2010) 90–97
catalysts not pretreated with NH
3
and the lower starting
temperature reveals a weak acid site common to all the SBA-15
samples. We do not see a broad desorption of NH from SBA-15 at
high temperatures. In addition to the weak acid site, the sulfonic
acid catalysts show a stronger acid site indicated by NH
3
3
desorption starting at about 220 8C. Comparing this peak to the
weak acid site, it can be seen that the number of strong acid sites is
highest for SBA-15-SO
SO H to SBA-15(EtCl)
3
H and decreases through SBA-15(EtCl)4-
SO H. This is expected from both the
3
8
3
stoichiometries used in the syntheses and the aqueous titrations of
the catalysts. It can also be noted that the strength of the strong
acid site, attributed to the sulfonic acid group, is unaffected by the
presence of the chloroalkyl group.
13
Fig. 3 shows the C NMR spectra obtained for the catalysts and
the chloroalkyl precursor, SBA-15-vinyl. This was collected by
cocondensation of TEOS and VTMS in 2N HCl but in the absence of
hydrogen peroxide. The resonances at 125 ppm and 132 ppm are in
agreement with those published previously [27]. The spectrum of
SBA-15-SO
groups nearest the silicon at about 17 ppm and a resonance at
4 ppm corresponding to methylene carbons adjacent to the
sulfonic acid group. The spectrum of SBA-15-(EtCl) SO H shows
the same peaks as SBA-15-SO H but also has a resonance at 47 ppm
attributed to the carbon bonded to chlorine. Table 3 compares the
3
H shows resonances attributed to the two methylene
5
2
3
3
13
C chemical shifts obtained by experiment and calculated using
density functional theory GIAO methods. DFT-GIAO methods at the
-31G(d,p) level have demonstrated to be suitable for conforma-
6
tional analysis and chemical shift predictions of substituted silica
compounds, similar to those investigated in this study [31]. The
strong convergence between calculated and found values is
meaningful in that we had considered that the vinyl group would
be converted to an alcohol through an epoxide intermediate. The
values in the table show that this is not the case and that the
chloroalkyl group results from addition of HCl to the double bond.
3.2. Catalytic conversion of methyl lesquerolate
Scheme 2 shows the reactions under study. The addition of
phenol to methyl lesquerolate results in the formation of the
phenoxy ether (PE), a pair of cyclic ethers, and several isomers of
Friedel–Crafts (FC) adducts. These last two are dead ends but the
phenoxy ether can be converted to the Friedel–Crafts adducts. The
adducts appear to form by this route as well as directly from the
methyl lesquerolate by a simultaneous alkylation and dehydration
resulting in several isomers containing the unsaturated fatty acid
ester. Fig. 4 shows the reaction course resulting from the addition
of phenol to methyl lesquerolate over Amberlyst 15, SBA-15-SO
3
H,
and SBA-15(EtCl) SO H at 120 8C. All the reactions were performed
8
3
with comparable levels of acid equivalents of catalyst. Also, since
these reactions were run at temperatures above 100 8C and none of
the reaction profiles indicates inhibition by water formed as a side
product, we did not examine the role of water in the reaction nor
the relative hydrophilicity of the catalysts as others have done [32].
The data points across the bottom of the three plots are the yields
of the methyl dienoates and the cyclic ethers. The former, mostly
methyl 11,13-eicosadienoate but including positional isomers, is
present in the starting material and changes only slightly over the
course of the reaction, indicating that dehydration of the methyl
lesquerolate is not a significant reaction under these conditions.
The cyclic ethers result from intramolecular attack of the alcohol
group on the double bond. These two isomers are formed with a
total yield of about 6% with all the catalysts studied. The impact of
each catalyst is seen in the yields of PE and the FC adducts. With
Amberlyst 15, the two products are produced at about the same
initial rate with the former ultimately being converted to the latter
to the point where FC adducts are the only isolated products. With
Fig. 4. Reaction profile for the reaction of methyl lesquerolate and phenol over (A)
Amberlyst 15; (B) SBA-15-SO H; and (C) SBA-15(EtCl) SO H. *, methyl
3
8
3
lesquerolate; *, methyl eicosenoates; &, cyclic ethers; &, phenoxy ether; ~,
Friedel–Crafts adducts.
3
the SBA-15-SO H catalyst, PE is formed at a greater rate resulting in
a higher yield than the FC adducts, but again, the PE is lost to FC
adducts at a rapid rate. The doubly functionalized catalyst, SBA-
8 3
15(EtCl) SO H, yields a much greater initial rate of formation of PE
compared to FC adducts and the conversion of PE to FC adducts
occurs at a very low rate giving the highest yield of PE.
These differences were further examined by comparing the
relative rates of formation of PE and FC adducts in the early stages
of conversion. For these reactions the substrate to catalyst ratio
was increased so that data could be collected while the reaction
was still under kinetic control. Conversion of up to 20% of the

M.A. Jackson, M. Appell / Applied Catalysis A: General 373 (2010) 90–97
95
3 4 3
Fig. 5. Relative rate of formation of the phenoxy ether (solid circles) and Freidel–Crafts adducts (open circles) over (A) Amberlyst; (B) SBA-15-SO H; (C) SBA-15(EtCl) SO H;
8 3
(D) SBA-15(EtCl) SO H.
methyl lesquerolate gave linear production rates. Fig. 5 shows the
results from these experiments. An additional catalyst, SBA-
rapidly converted to the Friedel–Crafts adducts. The porous SBA-15
catalysts show an increase in PE yield with increasing loadings of
the chloroalkyl group. It can further be seen that the increased
selectivity for the chloroalkyl-modified catalysts is manifested late
in the reaction indicating that it is the result of the relative rate of
conversion of PE to FC products. We, therefore, examined the
conversion of the ether to Friedel–Crafts adducts using a partially
purified sample of PE (approximately 75% PE with the remainder
dienoic acids and cyclic ethers). The turnover frequencies for this
reaction are included in Table 3. The catalysts can be divided into
groups with lower and higher turnover frequencies. The catalysts
1
4 3
5(EtCl) SO H is included here to show that the chloroalkyl
functionality contributes directly to the increased selectivity. In
PE FC
order to compare the catalyst selectivity, we use S = k /k where
PE
FC
k
and k are the rate constants for the formation of the two
products. Amberlyst 15 shows almost no selectivity with S = 1.2
and the SBA-15 catalysts show an increase in selectivity with an
increased loading of the chloroalkyl group. Table 4 contains these
selectivity data.
Fig. 6 shows the selectivity of each catalyst versus conversion.
Here selectivity is defined as the percentage of converted methyl
lesquerolate resulting in PE. Again, the lack of selectivity from
Amberlyst 15 is clear with a maximal selectivity of 50%. This
macroreticular solid acid leads to low levels of the ether that are
in the latter group, Amberlyst 15 and SBA-15-SO
selectivity toward PE formation.
3
H, have lower
3.3. Eapp, pre-exponential terms, and the compensation effect
Fig. 7 shows the Arrhenius’ plots for the catalyzed reaction of
methyl lesquerolate and phenol. Initial reaction rates were
collected at 10 8C intervals from 100 8C to 140 8C. The apparent
activation energies are shown in Table 4. The apparent activation
energies fall to the lower end of the broad range of published
values for ethers formed by the reaction of alcohols with
isobutylenes [12] and are much lower than the 150 kJ/mol
published for the formation of dioctyl ether from octanol over
zeolite BEA [15] and the 122 kJ/mol found for the formation of
dipentyl ether from pentanol over Nafion [33]. It is interesting to
see that as the measured activation energies fall, selectivity
improves. This is contrary to the expectation that lower activation
energies will lead to more side reactions [34].
Fig. 8 is the Constable plot for the SBA-15 catalysts which shows
the linear relationship between Eapp and the log of the pre-
exponential factor, termed the compensation effect [35]. The effect
has been classically attributed to the reduction in entropy caused
by the adsorption of reactants onto heterogeneous catalysts [36]. It
has recently been observed in the hydrogenation of tetralin on
platinum [37] and in the Knoevenagel condensation over a
Fig. 6. Performance curves for several catalysts in the conversion of methyl
lesquerolate to the phenoxy ether. ^, SBA-15-SO
SBA-15(EtCl) SO H; *, SBA-15(EtCl) SO H; *, SBA-15(EtCl)
Amberlyst 15.
3
H; ~, SBA-15(EtCl)
2
SO
3
H; &,
2
; &,
4
3
8
3
8
(SO
3
H)

9
6
M.A. Jackson, M. Appell / Applied Catalysis A: General 373 (2010) 90–97
catalyst, compensation results from the alignment of the reactants
on the catalyst surface in proximity to the basic active site, thereby
serving to reduce transition state entropy by organizing the
substrates. It seems reasonable then, in light of these arguments,
that the chloroalkyl functionalization serves to enhance reactivity
of the methyl lesquerolate by drawing it into the catalyst pores and
into the vicinity of the acid groups, thereby reducing entropy.
The intriguing mechanism of selectivity by the catalysts
deserves further discussion. It should be noted that hydrogen
bonds have been observed between chloro substituents and
hydrogen bond donors [39]. Likewise, hydrogen bridge bonds have
been observed between chloro substituents and the alcohol group
of an alkynol moiety, with calculations at the B3LYP/6-31G* level
indicating
a
favorable interaction of ꢀ16 kJ/mol [40]. We
calculated a similar interaction between the alkenol fragment of
methyl lesquerolate and the chlorine of the chloroethyl moiety and
found this to be 20.2 kJ/mol, suggesting that a hydrogen bond from
the methyl lesquerolate to the chlorine is the source of the
compensation effect. However, similar calculations on the
interaction of phenol with the chloroalkyl group gave a favorable
interaction of 19.4 kJ/mol. Considering this similar interaction with
the catalyst surface and the large excess of phenol in our reaction
conditions, it would seem the mechanism is more complex than
selective interaction through molecular recognition of the alkenol
Fig. 7. Arrhenius plots for four SBA-15 derived catalysts. ^, SBA-15-SO
3
H; ~, SBA-
1
5(EtCl) SO H; &, SBA-15(EtCl) SO H; *, SBA-15(EtCl) SO H.
2
3
4
3
8
3
(
Fig. 9). Furthermore, the most selective catalyst was developed
through an iterative synthetic approach with the population of
surface sites governed by the equilibrium of synthesis. Thus, the
mechanisms for which the catalyst prefers interactions with the
reactants over products are complex, and correspondingly likely
involves preferred surface and solvophobic interactions of the
reactants and products [41,42].
3.4. Highest yields of the phenoxy ether
A final catalyst was prepared which had both the higher acid
loading and selectivity resulting from high loading of chloroalkyl
groups. SBA-15(EtCl) (SO H) was prepared with 14.4% loading of
functionalized silica differentiating it from the others which were
3% functionalized. This yielded a catalyst with higher turnover
8
3
2
Fig. 8. Constable plot generated from the Arrhenius plot data in Fig. 7.
1
frequency and with high selectivity. As can be seen in the
performance curve in Fig. 6, the selectivity from this catalyst was
the highest and the reaction rate, in per gram catalyst, was high
enough to allow preparation of the phenoxy ether in a reasonable
amount of time.
4. Conclusion and summary
The current study demonstrates the successful development of
a novel catalyst capable of selective modification of a major
bioproduct of Lesquerella fendlerii, promising agricultural
commodity. We have shown that modifying the surface of SBA-
5-SO H with chloroalkyl groups affects the selectivity of the
a
1
3
catalysts in the reaction of methyl lesquerolate with phenol. This
increase in selectivity toward the ether product, methyl 11-cis-14-
phenoxyeicosenoate, results from a decrease in its conversion to
the Friedel–Crafts adducts. B3LYP density functional calculations
were useful to characterize the catalyst and probe the interactions
of reactants. Increase in yield will now allow us to prepare enough
of this new product for characterization of its bulk properties and
assessment as a new biobased product.
Fig. 9. Representation of methyl lesquerolate interacting with the sulfonic acid
residue of the catalyst.
bifunctional SBA-15 catalyst [38]. In the hydrogenation of tetralin,
the authors propose the compensation effect results from the
increased strength of adsorption of the substrate onto metal
clusters, leading to lower entropy. In the Knoevenagel condensa-
tion of benzaldehyde with ethyl cyanoacetate over the bifunctional
References
[
[
[
1] S.C. Cermak, K.B. Brandon, T.A. Isbell, Ind. Crops Res. 23 (2006) 54–64.
2] G.A. Olah, Friedel–Crafts Chemistry, vol. 1, Wiley Interscience, New York, 1973.
3] B.H. Nicolet, C.M. DeMilt, J. Am. Chem. Soc. 49 (1927) 1103–1106.
[4] J. Harmon, C.S. Marvel, J. Am. Chem. Soc. 54 (1932) 2515–2527.

M.A. Jackson, M. Appell / Applied Catalysis A: General 373 (2010) 90–97
97
[
[
[
[
[
5] F.D. Smith, H.E. Kenney, A.J. Stirton, J. Org. Chem. 30 (1965) 885–888.
6] E.T. Roe, W.E. Parker, D. Swern, J. Am. Oil Chem. Soc. 36 (1959) 656–659.
7] A. Eisner, T. Perlstein, W.C. Ault, J. Am. Oil Chem. Soc. 41 (1962) 290–292.
8] Y. Nakano, T.A. Foglia, J. Am. Oil Chem. Soc. 84 (1984) 569–573.
[26] K. Wolinski, J.F. Hinton, P. Pulay, J. Am. Chem. Soc. 112 (1990) 8251–8260.
[27] H.M. Kao, P.C. Chang, J.D. Wu, A.S.T. Chiang, C.H. Lee, Micropor. Mesopor. Mater.
97 (2006) 9–20.
[28] F. Rouquerol, J. Rouquerol, K. Sing, Adsorption by Powders and Porous Solids, vol.
439, Academic Press, San Diego, 1999.
9] F.O. Barret, C.G. Goebel, US Patent 3,074,983, (1 January 1963), Ohio, USA.
[
[
[
[
10] H. Kohashi, T.A. Foglia, J. Am. Oil Chem. Soc. 61 (1984) 1048–1052.
11] K.D. Black, F.D. Gunstone, Chem. Phys. Lipids 79 (1996) 87–94.
12] N. Boz, T. Dogu, K. Murtezaoglu, G. Dogu, Appl. Catal. A: Gen. 268 (2004) 175–182.
13] R.G. Herman, F.H. Khouri, K. Klier, J.B. Higgins, M.R. Galler, C. Terenna, J. Catal. 228
[29] I.K. Mbaraka, B.H. Shanks, J. Catal. 229 (2005) 365–373.
[30] S. Sreevardhan Reddy, B. vid Raju, V. Siva Kumar, A.H. Padmasri, S. Narayanan, K.S.
Rama Rao, Catal. Commun. 8 (2007) 261–266.
[31] L. Fernandez, P. Viruela-Martin, J. Latorre, C. Guillem, A. Beltr a´ n, P. Amoros,
Theochem. J. Mol. Struct. 850 (2008) 94–104.
[32] R. van Grieken, J.A. Melero, G. Morales, J. Mol. Catal. A: Chem. 256 (2006) 29–36.
[33] R. Bringuq, J. Tejero, M. Iborra, J.F. Izquierdo, C. Fitq, F. Cunill, Chem. Eng. J. 145
(2008) 135–141.
[34] K. Yoshizawa, Y. Shiota, T. Yamabe, J. Am. Chem. Soc. 120 (1998) 564–572.
[35] G.C. Bond, A.D. Hooper, J.C. Slaa, A.O. Taylor, J. Catal. 163 (1996) 319–327.
[36] W. Conner, J. Catal. 78 (1982) 238–246.
(
2004) 347–361.
14] T.V. Kovalchuk, J. Kochkin, H. Sfihi, V.N. Zaitsev, J. Fraissard, J. Catal. 263 (2009)
47–257.
15] I. Hoek, T.A. Nijhuis, A.I. Stankiewicz, J.A. Moulijn, Appl. Catal. A: Gen. 266 (2004)
09–116.
[
[
2
1
[
[
16] G.A. Olah, T. Shamma, G.K.S. Prakash, Catal. Lett. 46 (1997) 1–4.
17] J. Tejero, C. Fite, M. Iborra, J.F. Izquierdo, R. Bringue, F. Cunill, Appl. Catal. A: Gen.
3
08 (2006) 223–230.
[37] M.F. Williams, B. Fonf e´ , C. Woltz, A. Jentys, J.A.R. van Veen, J.A. Lercher, J. Catal. 251
(2007) 497–506.
[38] S.L. Hruby, B.H. Shanks, J. Catal. 263 (2009) 181–188.
[39] J.A. Christopher, F.L. Atkinson, B.D. Bax, M.J.B. Brown, A.C. Champigny, T.T.
Chuang, E.J. Jones, J.E. Mosley, J.R. Musgrave, Bioorg. Med. Chem. Lett. 19
(2009) 2230–2234.
[
[
18] J.W. Goodrum, D.P. Geller, Bioresour. Technol. 96 (2005) 851–855.
19] M.A. Salomon, J. Coronas, M. Men e´ ndez, J. Santamar ı´ a, Appl. Catal. A: Gen. 200
(
2000) 201–210.
20] T.A. Isbell, B.A. Lowery, S.S. DeKeyser, M.L. Winchell, S.C. Cermak, Ind. Crop Prod.
3 (2006) 256–263.
[
2
[
[
[
[
[
21] M. Thommes, Introduction to Zeolite Science and Practice, Elsevier, 2007, p. 495.
22] Parallel Quantum Solutions, Fayetteville, AR, 2005.
23] Hypercube, Inc., Gainesville, FL, 2006.
24] J. Baker, J. Comput. Chem. 7 (1986) 385–395.
25] R. Ditchfield, J. Chem. Phys. (1976) 3123–3133.
[40] R. Banerjee, G.R. Desiraju, R. Mondal, J.A.K. Howard, Chem. Eur. J. 10 (2004) 3373–
3383.
[41] C.A. Hunter, Angew. Chem. Int. Ed. 43 (2004) 5310–5324.
[42] D. Weber, J. Mitchell, J. McGregor, L.F. Gladden, J. Phys. Chem. C 113 (2009) 6610–
6615.