Epoxidizable fatty amide-phenol conjugates

Article abstract of DOI:10.1007/s11746-011-1785-0

This paper reports the synthesis of a series of novel compounds where carboxylic acids have been linked to a phenol through amidomethyl units. For instance, one, two, or three fatty acids have been selectively appended to the phenol in yields above 75%. The fatty acid used was oleic acid, which was subsequently epoxidized. Other functional groups, such as amino acids, can be incorporated in these compounds. Examples of monomers that are suitable for polymerization were also prepared: one acrylamide, one styrene derivative, and two types of AB₂ diamino acids, all of which contain oleic units that are sites for epoxidation and crosslinking. Fatty acid aryl ethers were prepared using hydroxy fatty acids. These molecules are intended to serve as augmented analogues of epoxidized vegetable oil. Their amide groups should lead to intermolecular aggregation through hydrogen bonding, and the option to covalently include other functional groups may permit tuning of the macroscopic materials properties of films, coatings, or solids constructed from them.

Full text of DOI:10.1007/s11746-011-1785-0

J Am Oil Chem Soc (2011) 88:1229–1237
DOI 10.1007/s11746-011-1785-0
ORIGINAL PAPER
Epoxidizable Fatty Amide–Phenol Conjugates
•
Jonathan A. Zerkowski Michael J. Haas
Received: 27 September 2010 / Revised: 31 January 2011 / Accepted: 9 February 2011 / Published online: 24 March 2011
Ó AOCS (outside the USA) 2011
Abstract This paper reports the synthesis of a series of
novel compounds where carboxylic acids have been linked
to a phenol through amidomethyl units. For instance, one,
two, or three fatty acids have been selectively appended to
the phenol in yields above 75%. The fatty acid used was
oleic acid, which was subsequently epoxidized. Other
functional groups, such as amino acids, can be incorporated
in these compounds. Examples of monomers that are
suitable for polymerization were also prepared: one acryl-
amide, one styrene derivative, and two types of AB2
diamino acids, all of which contain oleic units that are sites
for epoxidation and crosslinking. Fatty acid aryl ethers
were prepared using hydroxy fatty acids. These molecules
are intended to serve as augmented analogues of epoxi-
dized vegetable oil. Their amide groups should lead to
intermolecular aggregation through hydrogen bonding, and
the option to covalently include other functional groups
may permit tuning of the macroscopic materials properties
of films, coatings, or solids constructed from them.
Introduction
Crosslinked materials produced from epoxidized vegetable
oils (EVO) have been known and studied for decades [1],
but it is only in recent years that copolymerization of EVO
with other additives, with the goal of imparting new prop-
erties, has been studied. There are two basic approaches for
crosslinking the EVO. First is self-polymerization, where a
Lewis acid catalyst is used to activate epoxide units for
reaction with themselves, forming a polyether network [2].
With this route, other epoxide monomers can be included,
such as glycidol derivatives. The second route is to incor-
porate co-monomers such as amines that are reactive with
the epoxide unit; the structure of this crosslinker can be
varied to alter the final properties of the material [3]. Some
very interesting heterogeneous blends (using natural fibers
[4], clay [5], or ceramic nanoparticles [6]) that give nano-
composites have been formed, in particular by an offshoot
of this technique: the epoxide ring is opened with a group
that is itself polymerizable, such as an acrylic unit [7].
Alternatively, the epoxides can be hydrolyzed to polyols
that react with isocyanates to give polyurethanes. Clearly, a
need has been perceived for the ability to tune the properties
of EVO, which by itself offers only a limited range of bulk
mechanical properties, even though it is an abundant and
cheap polymer building block. Since one benefit of EVO is
its green character, it would be desirable if other structural
units for incorporation into EVO could also be bio-derived.
Recent advances in the pyrolysis of lignocellulosic
material suggest one such alternative for the preparation and
modification of EVO analogues. One class of compounds
that are present in ‘‘pyrolysis oil’’ is phenols [8]. While
current pyrolysis methods do not yield high percentages or
high purities of the phenolic fraction, work is underway that
will permit greater control of the product distribution, so
Keywords Co-products (Waste) \ biobased products Á
Polymers/coatings \ biobased products Á Fats and oils Á
Oleochemistry
Mention of trade names or commercial products in this publication is
solely for the purpose of providing specific information and does not
imply recommendation or endorsement by the US Department of
Agriculture.
J. A. Zerkowski (&) Á M. J. Haas
USDA, ARS, Eastern Regional Research Center,
600 E. Mermaid Lane, Wyndmoor, PA 19038, USA
e-mail: jonathan.zerkowski@ars.usda.gov
123

1230
J Am Oil Chem Soc (2011) 88:1229–1237
that a specific class of chemical structure can be obtained
[9]. Alternatively, it has been shown that phenol can be
produced through biotech routes with microorganisms,
using glucose and the shikimic acid pathway [10].
(Alfa Aesar, Ward Hill, MA). All solvents and reagents
were used as received. Silica gel used for column chro-
˚
matography (Grade 60 A, mesh 230–400, particle size
40–63 lm), was obtained from Fisher Scientific (Fairlawn,
NJ). NMR spectra were recorded at room temperature in
CDCl3 referenced to tetramethylsilane on either a Varian
Associates (Walnut Creek, CA) Gemini 200 MHz or Inova
400 MHz instrument. LCMS data were recorded on a
Waters/Micromass (Milford, MA) ZMD instrument with
atmospheric pressure chemical ionization (APCI), using
direct injection (no column) with 9:1 CH3CN:0.5%
HCO2H in CH3CN as eluant. Masses were determined as
sodiated ionic adducts of the products, [M ? Na]?.
Our aim was to examine ways to combine this new
biobased source of phenolics with agriculture-derived fatty
acids (FA) to prepare new, and highly modifiable, versions
of EVO. The thrust of this initial work is synthetic and
exploratory, attempting to map out functional group pat-
terns that are attainable; obviously the specific lab-scale
procedures employed may require refinement before scal-
ing up. Compounds combining phenolics with unsaturated
alkyl chains are known in nature and synthetically. They
have proven to be of interest for fundamental scientific
behavior as well as for technological uses. Urushiol is the
cross-linkable component of natural lacquer [11]. Cardanol
is found in cashew nut oil, and has been shown to form
Amidomethylation of phenol (Tscherniac–Einhorn
reaction) [14] (Fig. 1)
`
fibrous aggregates [12]. Galia et al. [13] have prepared
Phenol (0.45 g, 4.8 mmol) and hydroxymethylchloroace-
tamide (1.5 g, 12 mmol) were added in one portion to
9 mL acetic acid and 1 mL sulfuric acid. The mixture was
stirred magnetically overnight, then poured onto approxi-
mately 150 mL crushed ice. The liquid was decanted from
the gummy solid, partially neutralized with approx. 2 g
NaHCO3, and extracted with ethyl acetate (5 9 100 mL)
and chloroform (2 9 100 mL). The organic fractions were
combined with the gummy solid to dissolve it, dried over
MgSO4, and evaporated. The residue could then be used as
is for further reactions, or could be purified by column
chromatography on silica gel with ethyl acetate as eluant.
Four major fractions were obtained, which were identified
by MS and NMR to be, in order of elution, 1a, 1b together
fatty acid esters with trihydroxybenzoates and used these to
crosslink silica particles. The work reported here presents a
new family of fatty acid/phenol amides that should func-
tion similarly to EVO, but with structurally distinct features
that allow a wide range of derivatization (Scheme 1).
Experimental
General
Chemicals and reagents were obtained commercially from
Sigma-Aldrich (St. Louis, MO) and Lancaster Synthesis
Scheme 1
123

J Am Oil Chem Soc (2011) 88:1229–1237
1231
6.88 (d, 6.5 Hz, 1H, 6-ArH), 6.95 (s, 1H, 3-ArH), 7.09 (d,
6.6 Hz, 1H, 5-ArH), 7.32 (br s, 1H, ortho NH), 9.50 (s, 1H,
OH). 13C NMR: 40.6, 42.4, 42.7, 43.4, 118.2, 123.9, 129.0,
129.9, 130.8, 155.4, 166.3, 168.6. 1e: 4.12 (s, 2H, para
CH2Cl), 4.15 (s, 4H, ortho CH2Cl), 4.46 (m, 6H, ArCH2),
6.82 (br s, 1H, para NH), 7.05 (s, 2H, 3- and 5-ArH), 7.36
(br s, 2H, ortho NH), 9.60 (s, 1H, OH). 13C NMR: 40.4,
42.5, 42.7, 43.2, 125.3, 129.0, 130.5, 154.0, 166.1, 167.6.
Tri-oleic and mixed oleic/amino acid adducts
(Figs. 2, 3)
Trisubstituted molecule 1e (400 mg, 0.98 mmol) was
heated to reflux in 7 mL EtOH and 3 mL conc. HCl for
2 h. The solvent was removed under a stream of nitrogen
overnight, and the solid was triturated three times with
Et2O. The residue was then dissolved in 5 mL
N,N-dimethylformamide (DMF), and 1,8-diazabicyclounde-
cene (DBU, 0.46 g, 3 mmol) was added. To this mixture
was added a solution of oleic acid (1.13 g, 4.0 mmol),
hydroxybenzotriazole hydrate (HOBt, 0.63 g, 4.1 mmol),
and diisopropyl carbodiimide (DIC, 0.40 g, 3.2 mmol) that
had been pre-mixed for 10 min in 5 mL DMF. The reaction
was stirred magnetically for 3 h, then solvent was removed
on the rotary evaporator. The residue was purified by col-
umn chromatography to give 2a (730 mg, 77%). 1H NMR:
0.88 (t, 6.7 Hz, 9H, CH3), 1.15–1.43 m, 1.51–1.71 m,
1.83–2.11 (m, 12H, =CH–CH2), 2.20 (t, 7.2 Hz, 6H,
CH2C(O)), 4.25 (d, 6.9 Hz, 2H, para ArCH2), 4.31
(d, 6.8 Hz, 4H, ortho ArCH2), 5.23–5.44 (m, 6H, olefin H),
5.96 (m, 1H, para NH), 6.58 (m, 2H, ortho NH), 7.03
(s, 2H, Ar), 10.2 (br s, 1H, OH). 13C NMR: 14.2, 22.8,
25.6, 25.8, 27.3, 29.2–29.8, 32.0, 36.6, 36.9, 40.3, 43.0,
126.0, 129.5, 129.8, 130.1, 153.9, 173.3, 174.7. APCI:
C63H110N3O4ÁNa? calc 997.8 Da, found 997.4 Da.
Fig. 1 Preparation of amidomethylated phenols
with 1c (separable using a second column with 1:1 hexane/
ethyl acetate or 5% MeOH/CHCl3), 1d, and 1e. Combined
yields of the products ranged from 85 to 94%, based on
phenol. 1a: 1H NMR: 4.13 (s, 2H, CH2Cl), 4.47 (d, 6.9 Hz,
2H, ArCH2), 6.88 (t, 6.5 Hz, 1H, Ar), 6.97 (d, 6.6 Hz, 1H,
Ar), 7.17 (d, 6.5 Hz, 1H, Ar), 7.26 (t, 6.4 Hz, 1H, Ar), 7.39
(br s, 1H, NH), 9.30 (s, 1H, OH). 13C NMR: 40.7 CH2Cl,
42.3 ArCH2, 117.9 Ar C ortho to OH, 120.5, 130.0, 130.4,
131.0, 155.4 ArC-OH, 168.7 amide C=O (since these pat-
terns recur in subsequent compounds, for brevity assign-
ments are not repeated below). 1b: 4.13 (s, 2H, CH2Cl),
4.42 (d, 6.7 Hz, 2H, ArCH2), 6.75–6.95 (br s, 1H, NH),
6.84 (d, 8.0 Hz, 2H, 2- and 6-ArH), 7.18 (d, 8.0 Hz, 2H,
3- and 5-ArH), 10.1 (s, 1H, OH). 13C NMR: 42.5, 43.6,
115.9, 129.4, 128.7, 156.1, 166.3. 1c: 4.13 (s, 4H, CH2Cl),
4.48 (d, 6.9 Hz, 4H, ArCH2), 6.85 (t, 7.2 Hz, 1H, 4-ArH),
7.20 (d, 7.3 Hz, 2H, 3- and 5-ArH), 7.46 (br s, 2H, NH),
9.42 (s, 1H, OH). 13C NMR: 40.6, 42.6, 120.2, 124.9,
130.8, 154.3, 167.5. 1d: 4.13 (s, 2H, CH2Cl), 4.15 (s, 2H,
CH2Cl), 4.45 (m, 4H, ArCH2), 6.85 (br s, 1H, para NH),
For the mixed adducts, 1e (550 mg, 1.3 mmol) was
deprotected similarly, then dissolved in 7 mL DMF with
DBU (0.61 g, 4 mmol) and treated with a premixed solu-
tion of 1:1 oleic acid (0.56 g, 2 mmol) and 3-(tert-but-
oxycarbonylamino)-propanoic acid (Boc-b-alanine, 0.38 g,
2 mmol), DIC (0.52 g, 4.1 mmol), and HOBt (0.63 g,
4.1 mmol) in 5 mL DMF. Column chromatography with
ethyl acetate provided three product groups: the trioleic
adduct 2a (19%), identical to the sample prepared above,
the mixed products 3a–3d (48%, Fig. 3), and the tris–Boc-
b-alanine adduct 3e (24%). Compounds 3a–3d could be
separated from each other by using 5% MeOH/CHCl3 as
eluant. 3a: 1H NMR: 0.90 (t, 6.6 Hz, 6H, CH3),
1.11–1.42 m, 1.45 (s, 9H, Boc tBu), 1.52–1.79 m,
1.84–2.11 (m, 8H, =CH–CH2), 2.21 (t, 7.2 Hz, 4H, oleic
C(O)CH2), 2.42 (t, 7.4 Hz, 2H, alanine C(O)CH2), 3.44 (q,
6.8 Hz, 2H, alanine CH2NH), 4.24–4.44 (m, 6H, ArCH2),
5.25–5.47 (m, 6H, Boc-NH and olefin H), 6.29 (br, 1H,
123

1232
J Am Oil Chem Soc (2011) 88:1229–1237
Fig. 2 Preparation of the tri-
oleic adduct and its epoxidation.
Reagents and conditions: a 7:3
EtOH/conc. HCl, reflux, 2 h,
then evaporation. b DMF with
DBU, then a premixed solution
of oleic acid, DIC, and HOBt
(see ‘‘Experimental’’ for
abbreviations). c mCPBA,
CH2Cl2, 16 h
Fig. 3 Preparation of the mixed
oleic/Boc-b-alanine adducts.
Reagents and conditions: a 7:3
EtOH/conc. HCl, reflux, 2 h,
then evaporation. b DMF with
DBU, then a premixed solution
of 1:1 oleic acid:Boc-b-alanine,
DIC, and HOBt. A oleic,
B Boc-b-alanine
amide), 6.58 (br, 2H, amide), 7.05 (br, 2H, Ar). 13C NMR:
14.2, 22.8, 25.6, 27.3, 28.5, 29.2–29.9, 32.0, 36.5, 40.2,
79.6, 126.0, 129.8, 130.1, 153.8, 156.4 (carbamate C=O),
174.7. 3b: 1H NMR: 0.90 (t, 6.6 Hz, 3H, CH3),
1.10–1.40 m, 1.44 (s, 18H, Boc tBu), 1.52–1.77 m,
1.91–2.09 (m, 4H, =CH–CH2), 2.20 (t, 7.3 Hz, 2H, oleic
C(O)CH2), 2.41 (t, 7.2 Hz, 4H, alanine C(O)CH2), 3.39 (q,
6.6 Hz, 4H, alanine CH2NH), 4.23 (d, 6.8 Hz, 2H, ArCH2),
4.37 (d, 6.9 Hz, 4H, ArCH2), 5.24 (br, 2H, Boc NH),
5.31–5.42 (m, 2H, olefin H), 6.04 (br, 1H, amide),
6.95–7.13 (br, 2H, amide), 7.03 (s, 2H, Ar). 13C NMR:
14.2, 22.8, 25.8, 27.3, 28.5, 29.2–29.9, 32.0, 36.3, 36.8,
40.1, 42.9, 79.7, 129.7, 129.8, 130.1, 153.7, 156.3 (carba-
mate C=O), 172.9, 173.3. 3c: 1H NMR: 0.90 (t, 6.7 Hz, 6H,
CH3), 1.06–1.38 m, 1.44 (s, 9H, Boc tBu), 1.52–1.77,
1.91–2.13 (m, 8H, =CH–CH2), 2.15-2.28 (m, 4H, oleic
C(O)CH2), 2.40 (t, 7.3 Hz, 2H, alanine C(O)CH2), 3.41 (q,
6.4 Hz, 2H, alanine CH2NH), 4.24 (d, 7.9 Hz, 2H, ArCH2),
4.29–4.41 (m, 4H, ArCH2), 5.25 (br s, 1H, Boc NH),
5.28–5.44 (m, 4H, olefin H), 5.96, 6.63, and 6.92 (each br s,
1H, amides), 7.02 (s, 2H, Ar). 13C NMR: 14.2, 22.8, 25.6,
25.8, 27.3, 28.5, 29.2–29.9, 32.0, 36.3, 36.4, 36.8, 40.2,
40.3, 43.0, 79.6, 129.6, 129.8, 130.1, 153.8, 156.2 (carba-
mate C=O), 172.5, 173.2, 175.1. 3d: 1H NMR: 0.90
(t, 6.8 Hz, 3H, CH3), 1.09–1.39 m, 1.45 (s, 18H, Boc tBu),
1.52–1.79 m, 1.92–2.09 (m, 4H, =CH–CH2), 2.20
(t, 7.2 Hz, 2H, oleic C(O)CH2), 2.28–2.49 (m, 4H, alanine
C(O)CH2), 3.23–3.50 (m, 4H, alanine CH2NH), 4.10–4.41
(m, 6H, ArCH2), 5.26–5.58 (m, 6H, Boc NH and olefin H),
6.77 (br s, 1H, NH), 6.90–7.05 (m, 3H, NH and Ar), 7.20
(br s, 1H, amide). 13C NMR: 14.2, 22.8, 25.7, 27.3, 28.5,
29.2–29.8, 32.0, 36.1, 36.4, 36.7, 40.2, 42.8, 79.5, 125.8,
126.0, 129.8, 130.1, 153.6, 156.3 and 156.4 (two distinct
carbamate C=O), 171.7, 172.7, 175.2. 3a and 3c: calc
C53H92N4O6ÁNa? 904.3 Da, found 904.1 and 904.5 Da
respectively. 3b and 3d: calc C43H73N5O8ÁNa? 811.1 Da,
found 811.1 and 811.4 Da, respectively.
Epoxidation (Fig. 2)
Tri-oleic adduct 2a (100 mg, 0.1 mmol) was dissolved in
10 mL CH2Cl2, and to it was added m-chloroperbenzoic
123

J Am Oil Chem Soc (2011) 88:1229–1237
1233
acid (77% by weight, 112 mg, 0.5 mmol). The reaction
was stirred at rt overnight, diluted with 40 mL CH2Cl2,
then washed with a dilute solution of sodium bisulfite
(2 9 50 mL) and sodium bicarbonate (2 9 50 mL). Sol-
tentatively assigned using 2D NMR spectroscopy. Briefly,
3J couplings were observed between one of the benzylic
CH2 groups and the phenol C as well as the acrylic car-
1
bonyl. H NMR: 0.91 (t, 6.7 Hz, 3H, CH3), 1.16–1.45 m,
1
vent was removed to afford 2b (96 mg, 95%). H NMR:
1.56–1.78 m, 1.91–2.11 (m, 4H, =CH–CH2), 2.17
(t, 7.3 Hz, 2H, CH2–C(O)), 4.26 (d, 6.8 Hz, ArCH2), 4.37
(d, 6.8 Hz, ArCH2), 5.26–5.43 (m, 2H, oleic olefin H),
5.66–5.84 (m, 1H, acrylic), 5.90 br s (NH), 5.97–6.18 (m,
1H, acrylic), 6.27–6.42 (m, 1H, acrylic), 6.90 (d, 7.5 Hz,
1H, Ar-6H), 6.98 (s, Ar-3H), 7.09 (d, 7.6 Hz, 1H, Ar-5H),
7.16 br s (NH). 13C NMR: 14.2, 22.8, 25.8, 27.3, 29.2-29.9,
32.0, 36.9, 40.5, 43.1, 118.2, 124.5, 128.0, 128.5, 129.5,
129.6, 129.8, 130.2, 130.7, 131.5, 155.5, 167.9, 173.3.
C29H46N2O3ÁNa? calc 493.3 Da, found 493.6 Da.
0.88 (t, 6.7 Hz, 9H, CH3), 1.15–1.48 m, 1.51–1.71 m, 2.20
(t, 7.3 Hz, 6H, CH2C(O)), 2.80–2.99 (m, 6H, HC–O
epoxide), 4.25 (d, 6.9 Hz, 2H, para ArCH2), 4.31 (d,
6.8 Hz, 4H, ortho ArCH2), 5.96 (m, 1H, para NH), 6.58 (m,
2H, ortho NH), 7.03 (s, 2H, Ar), 10.2 (br s, 1H, OH). 13C
NMR: 14.2, 22.8, 25.6, 25.7, 26.7, 27.9, 29.1–29.8, 31.9,
36.5, 36.8, 40.3, 43.0, 57.4 (epoxide C–O), 126.0, 129.5,
130.2, 153.9, 173.1, 174.6. C63H111N3O7ÁNa? 1044.8 Da,
found 1045.1 Da.
Oleic/Acryloyl Compound (Fig. 4)
Conversion to a Styrene Derivative [15] (Fig. 5)
Disubstituted molecule 1d (0.80 g, 2.6 mmol) was dis-
solved in 15 mL EtOH, and a solution of thiourea (200 mg,
2.6 mmoL) in 5 mL EtOH was added. The mixture was
heated to reflux for 2 h. Solvent was removed on the rotary
evaporator, and the residue was taken up in 5 mL DMF
with DBU (0.4 g, 2.6 mmol). A solution of oleic acid
(0.8 g, 2.8 mmol), HOBt (460 mg, 3.0 mmol), and DIC
(352 mg, 2.8 mmol) that had been premixed for 10 min in
3 mL DMF was added, and stirred for 2 h. Solvent was
removed, and the residue was purified by column chro-
matography on silica gel with ethyl acetate. Four products
were obtained: the dioleic adduct (318 mg, 18%), two
isomers of monooleic/monochloroacetic product (310 and
342 mg, respectively, 51% total), and the unchanged
starting material (dichloroacetic adduct 1d, 160 mg, 20%).
One of the mixed monosubstituted isomers (200 mg,
0.41 mmol) was selected and deprotected as above with
thiourea in EtOH. After removal of solvent, it was dis-
solved in 3 mL DMF with DBU (61 lL, 0.41 mmol), and a
premixed solution of acrylic acid (35 mg, 0.41 mmol), DIC
(63 mg, 0.41 mmol), and HOBt (76 mg, 0.41 mmol) was
added. After 2 h, solvent was removed, and the crude was
chromatographed on silica gel with ethyl acetate to afford
molecule 4 (167 mg, 87%). The isomer pattern, with the
acrylic unit ortho to the phenol OH as shown in Fig. 4, was
Molecule 5a (0.23 g, 0.58 mmol), prepared from 1b anal-
ogously to other oleic versions (see Fig. 2), was dissolved
in 5 mL pyridine and trifluoromethanesulfonic anhydride
(98 lL, 0.58 mmol) was added in an ice bath. After 2 h, tlc
showed starting material still present, so another 10 lL of
the anhydride was added. The solvent was removed, and
the product isolated by column chromatography (280 mg,
93%). This triflate 5b (0.20 g, 0.39 mmol) was then placed
in a screw-cap pressure vial, and PdCl2 (5 mg), PPh3
(10 mg), potassium vinyltrifluoroborate (KBF3(CH=CH2),
52 mg, 0.39 mmol) and Cs2CO3 (381 mg, 3 equiv.,
1.2 mmol) were added, along with 3 mL of 9:1 THF/water.
The vial was sealed and heated at 80 °C for 7 h, then
stirred at rt overnight. Removal of solvent and chroma-
tography afforded styrene derivative 5c (108 mg, 70%). 1H
NMR 5c: 0.90 (t, 6.6 Hz, 3H, CH3), 1.15–1.43 m,
1.55–1.78 m, 1.82–2.12 (m, 4H, =CH–CH2), 2.22
(t, 7.3 Hz, 2H, CH2C(O)), 4.46 (d, 6.5 Hz, ArCH2), 5.26
(d, 10.5 Hz, 1H, Ar–CH=CH(trans)), 5.32–5.45 (m, 2H,
oleic olefin), 5.70 (br s, 1H, amide), 5.75 (d, 17.6 Hz,
Ar–CH=CH(cis)), 6.72 (dd, 10.6 and 17.6 Hz, Ar–CH=),
7.25 (d, 8.0 Hz, 2H, Ar), 7.39 (d, 8.1 Hz, 2H, Ar). 13C
NMR: 14.2, 22.8, 25.9, 27.3, 29.3–29.9, 32.0, 36.9, 43.4,
114.1, 126.6, 128.1, 129.8, 130.1, 136.4, 137.0, 138.1,
173.0. C27H43NOÁNa? calc 420.3 Da, found 420.6 Da.
Fig. 4 Preparation of the oleic/acrylic adduct. Reagents and conditions: a one equiv thiourea, EtOH, reflux, then evaporation. b DMF with DBU,
then a premixed solution of oleic acid, DIC, and HOBt. c DMF with DBU, then a premixed solution of acrylic acid, DIC, and HOBt
123

1234
J Am Oil Chem Soc (2011) 88:1229–1237
Fig. 5 Conversion of a phenol
to a styrene. a (CF3SO2)2O,
pyridine. b PdCl2, Cs2CO3,
PPh3, KF3BCH=CH2, 9:1
THF/water, 80 °C
Mitsunobu (Aryl Ether) Products [16] (Fig. 6)
36.6, 37.0, 43.2, 51.6, 74.1 (C(17)-O), 116.1, 129.3, 129.9,
1
130.1, 130.3, 157.8, 172.9, 173.4. 8b: H NMR: 1.12 (d,
Molecule 1b (0.37 g, 1.9 mmol), PPh3 (0.55 g, 2.1 mmol),
and 17-hydroxy oleic acid methyl ester [17] (0.72 g,
2.3 mmol), were dissolved in 10 mL THF. To this mixture
was added diisopropyl azodicarboxylate (DIAD, 0.42 g,
2.1 mmol) in 2 mL THF. The reaction was stirred over-
night. Solvent was removed, and column chromatography
on silica gel gave 6a (805 mg, 86%). Reactions using
molecules 1c or 1d afforded similarly 6c (69%) or 6b
(73%), respectively.
6.9 Hz, 6H, CH3), 1.21–1.42 m, 1.45 (s, 18H, Boc tBu),
1.52–1.79 m, 1.94–2.09 (m, 12H, =CH–CH2), 2.14–2.24
(m, 4H, amide C(O)CH2), 2.32 (t, 7.3 Hz, 2H, ester
C(O)CH2), 3.52–3.62 (m, 2H, C(17)H–NH), 3.69 (s, 3H,
Me ester), 4.27–4.50 (m, 5H, ArCH2 and ArOC(17)H),
5.28–5.43 (m, 6H, olefin H), 5.80 (m, 3H, amide and Boc
NH), 5.98 (m, 1H, amide), 6.82 (d, 6.6 Hz, 1H, Ar-3H),
7.11–7.22 (m, 2H, Ar-5H and -6H). 13C NMR: 19.7 (C(18)
next to O–Ar), 21.3 (C(18) next to NH-Boc), 22.8, 25.0,
25.6, 25.9, 26.1, 27.3, 28.5, 29.2-29.9, 32.0, 34.3, 36.6,
37.0, 37.5, 40.7, 43.3, 47.1 (C(17)H-NHBoc), 51.5 (Me
ester), 79.0 (Boc quaternary C), 79.2 (C(17)-O), 117.9,
123.9, 129.1, 129.9, 130.1, 130.8, 155.4 (Boc carbamate
C=O), 157.7 (ArC-OH), 173.0, 173.4. 6c: C31H48Cl2N2O5Á
Na? calc 621.3 Da, found 621.6 Da. 7: C44H75NO4ÁNa?
calc 704.6 Da, found 704.4 Da. 8b: C73H128N4O9ÁNa ?
1228.8 Da, found 1228.5 Da.
Aryl ether 6a (0.18 g, 0.37 mmol) was then treated with
thiourea (50 mg, 0.66 mmol) in 10 mL EtOH for 2 h at
reflux, and solvent was removed. DMF (3 mL) and DBU
(56 lL, 0.37 mmol) were added, followed by a premixed
solution of oleic acid (113 mg, 0.4 mmol), DIC (50 mg,
0.4 mmol), and HOBt (61 mg, 0.4 mmol). After 2 h, the
solvent was removed, and column chromatography gave
molecule 7 (204 mg, 81%). Molecule 6b (350 mg,
0.6 mmol) was deprotected and coupled similarly with
2 equiv of 17-azido oleic acid (prepared as reported by us
previously [17]). The DMF solvent was removed, and the
crude taken up in THF with approximately 100 lL H2O,
then PPh3 (393 mg, 1.5 mmol) was added. The reaction
was stirred for 1 h, then Boc anhydride (436 mg, 2 mmol)
was added. The reaction was stirred overnight, solvent was
removed, and column chromatography gave molecule
Results and Discussion
Strategies for Linking Phenol to FA
We sought a synthetically flexible method for creating
FA–phenol conjugates, that is, a method that would permit
incorporation of other functional groups in a variety of
spatial arrangements. Several methods were available. The
most obvious would be just to prepare the phenyl ester, but
then there would be no options for further derivatization
using the FA carboxy unit, and further chemical modifica-
tions of the Ar ring might hydrolyze the ester. In fact this
route has been reported in a patent for preparing anti-tumor
compounds that are esters of phenols and fatty acids [18].
An alternative is to use acid-catalyzed addition of the
electron-rich aromatic ring to the olefin. While this proce-
dure is convenient on a large scale, it consumes the olefin,
so epoxidation would not be possible. Instead, we took
advantage of the electron-rich character in a different way,
using the Tscherniac–Einhorn reaction. This long-known
reaction [14] uses acid catalysis to substitute a hydroxy-
methyl amide onto the ring, affording amidomethylated
1
8b (489 mg, 65% from 6b). 6c: H NMR: 1.15–1.52 m,
1.55–1.81 m, 1.93–2.12 (m, 4H, =CH–CH2), 2.33
(t, 7.1 Hz, 2H, C(O)CH2), 3.67 (s, 3H, Me ester), 4.01–4.19
(m, 1H, ArOC(17)H), 4.09 (s, 4H, CH2Cl), 4.58 (d, 6.9 Hz,
4H, CH2NH), 5.28–5.43 (m, 2H, olefin H), 7.07–7.20 (m,
3H, 2 9 NH and Ar para H), 7.21–7.30 (m, 2H, Ar-3H and
-5 H). 13C NMR: 19.8, 25.0, 25.8, 27.3, 29.2–29.8, 34.2,
37.1, 40.0, 42.8, 51.6, 80.0 (C(17)-O), 124.6, 129.4, 129.9,
130.0, 131.3, 154.0, 165.9, 174.5. 7: 1H NMR: 0.91
(t, 6.8 Hz, 3H, CH3), 1.18–1.43 m, 1.53–1.77 m, 1.94–2.12
(m, 8H, =CH–CH2), 2.21 (t, 7.2 Hz, 2H, amide C(O)CH2),
2.32 (t, 7.3 Hz, 2H, ester C(O)CH2), 3.69 (s, 3H, Me ester),
4.26–4.37 (m, 1H, ArOC(17)H), 4.39 (d, 6.9 Hz, 2H,
ArCH2), 5.25–5.43 (m, 4H, olefin H), 5.64 (br s, NH), 6.86
(d, 8.0 Hz, 2H, Ar), 7.10 (d, 8.0 Hz, 2H, Ar). 13C NMR:
14.2, 19.8, 22.8, 25.0, 25.6, 25.9, 27.3, 29.2–29.9, 34.2,
123

J Am Oil Chem Soc (2011) 88:1229–1237
1235
Fig. 6 Synthesis of oleic aryl ethers. a 17-OH oleic acid, DIAD,
PPh3. b Thiourea, EtOH, reflux, then evaporation. c DMF with DBU,
premixed solution of oleic acid, DIC, and HOBt. d DMF with DBU,
premixed solution of 17-azido oleic acid, DIC, and HOBt. e PPh3, aq.
THF, then Boc2O
phenols. By controlling the stoichiometry of the reagents,
the product distribution can be driven toward predominant
addition of one, two, or three such units, although of course
a mixture of products is obtained. This distribution is not a
drawback in our intended use of these compounds for two
reasons. First, all the compounds can be separated from
each other via silica gel chromatography if a specific sub-
stitution pattern is desired. Alternatively, for use in mate-
rials blends with EVO, we propose that there is no need to
separate the individual compounds, they can just be
employed as a collected group. This grouping of a variety of
molecules is analogous to the heterogeneous nature of EVO
itself, since some of the fatty acid chains are saturated and
unepoxidized (stearic), some mono or di-epoxidized (oleic
or linoleic). One special benefit of the amidomethylphenol
structure that our molecules possess is the presence of the
hydrogen-bonding NH–CO units. Intermolecular hydrogen
bonding between amides is known to impart toughness to
polymers, such as nylons, that contain such groups. To the
best of our knowledge, FA/phenol conjugates linked
through an amidomethyl unit have not been previously
reported.
The Tscherniac–Einhorn reaction between two or three
equivalents of hydroxymethylchloroacetamide and phenol
proceeds smoothly at room temperature, and isolation of
the products is straightforward (Fig. 1). The chloroacet-
amide unit, which essentially serves as a protecting group,
has the benefit of being removable with different classes of
123

1236
J Am Oil Chem Soc (2011) 88:1229–1237
reagent, either acid, base, or nucleophile. We found that the
base method was cumbersome and gave unsatisfactory
results (unidentified side products). The acid route is con-
venient, but after removal of the aqueous solvent, the
chloroacetic acid is still present—if it is not removed, it
will compete with the incoming acyl unit that is intended
for ligation to the aminomethylphenol. Fortunately, the
chloroacetic acid can be fully removed by evaporation or
by trituration with diethyl ether. NMR of the material after
this treatment showed that no chloroacetic units remained.
Once the chloroacetyl group is removed, the free amine is
coupled to oleic acid or some other acid using standard
amide-forming protocols. We used carbodiimide couplings,
but other reagents could be used. In principle, acylation
could also occur at the phenol OH, but that group is a poor
nucleophile, and we have not observed any such products.
The deprotected aminomethyl phenols are obtained as their
hydrochloride salts, so during couplings we include a
stoichiometric amount of 1,8-diazabicycloundecene (DBU),
a non-nucleophilic base.
deprotection. Since its removal by a nucleophile is on a
stoichiometric basis, not catalytic, it is easier to control.
We used one equivalent of thiourea [19] to partially de-
protect the disubstituted molecule 1d, then coupled the
product to oleic acid (Fig. 4). The deprotection procedure
can be repeated with another equivalent of thiourea, then
the amine coupled to a different acyl group. To demon-
strate the appending of a group that could be used for
polymerization, the second acyl group we attached was an
acryloyl unit. We propose that the acrylic unit can be
polymerized first, then the pendant oleic chain epoxidized.
Concerning the utility of thiourea as a stepwise deprotec-
tant, results are mixed. Even though a single equivalent of
it can be added, the species in solution are not selective,
that is, once one of the chloroacetyl groups has been
removed, even though there is only one equivalent total of
thiourea, the second chloroacetyl group on the same mol-
ecule could also be removed. As a result, after the oleic
acid coupling described for 1d, we observed dioleic prod-
uct as well as unreacted 1d and the expected two isomers of
monooleic/monochloroacetyl (one with FA ortho to OH
and the other para). This product distribution of four
molecules is similar to what would be expected if we had
used HCl to remove both chloroacetyl groups in a single
step, then coupled with an equimolar mixture of acrylic
acid and oleic acid. We can only speculate, but it may be
that separation of the ortho/para isomers may be easier
when one of the groups is chloroacetyl, that is, since it is
less hydrophobic than a longer alkyl group, it is ‘‘more
different’’ from the oleic group, and polarity differences
between the isomers may aid chromatographic separation.
A second way to introduce a polymerizable unit to this
class of molecules is through the phenolic OH (Fig. 5).
Recent synthetic advances have shown how this functional
group can be readily converted to a vinyl unit [15]. These
routes rely on the easy conversion of OH to a tri-
fluoromethanesulfonate, which then reacts with a metal
catalyst and a vinylating reagent in a Suzuki–Miyaura
reaction. We found that this methodology worked well with
our oleic phenol conjugate 5a. The resulting styrene
compound 5c should be polymerizable to afford a poly-
styrene derivative with oleic amidomethyl units at the para
position, following which the olefin groups of the pendant
oleic chains would be epoxidized and crosslinked. As an
aside, it might be worth noting the structural similarity of
molecule 5a to FA derivatives of p-coumaric acid. Since
lipophilization of phenolics is an active area of research
[20], these FA–amidomethyl phenols could find a role in
such studies as well as in materials chemistry.
Synthesis of FA/amidomethylphenol conjugates
The simplest example in this new family of molecules is
the triply substituted oleyl phenol 2a (Fig. 2). After
removal of the chloroacetyl groups from 1e with HCl,
coupling with oleic acid gave 2a in good yield. Since one
ultimate use for these compounds is in crosslinking, we
also examined its epoxidation. This reaction went smoothly
to give the expected product 2b. Neither the phenol group
nor the amide interferes with the epoxidation.
A more complicated situation arises if different acylat-
ing partners are used (Fig. 3). We reacted the acid-depro-
tected tris aminomethylphenol with a mixture containing
one equivalent of oleic acid and one equivalent of Boc-b-
alanine. As expected, a mixture of products was observed:
small amounts of trioleic (2a) and tri-Boc-b-alanine
adducts, plus the mono oleic/di Boc-b-Ala and the di oleic/
mono Boc-b-Ala adducts. There is a further complication
in that the two 2:1 compounds can each be symmetric or
non-symmetric. With careful chromatography, all four
compounds can be isolated, however. Our motivation
behind performing this separation was simply to identify
the compounds, not to propose that it would be practical to
do so. Instead, we would expect that employment of these
compounds in a crosslinked matrix with EVO would sim-
ply keep all the relatives together, that is, A3, B3, and all
isomers of A2B and AB2. It is true that the B3 compound
would not take part in epoxidation and crosslinking, but it
could presumably be extracted out of the material after
these steps.
Another use for the OH group is by forming ethers with
it, using Mitsunobu protocols [16]. This route requires an
alcohol reacting partner, for which we chose a hydroxylated
fatty acid, namely 17-hydroxy oleic acid. The preparation of
One interesting manner in which the chloroacetamide
protecting group can be taken advantage of is by partial
123

J Am Oil Chem Soc (2011) 88:1229–1237
1237
3. Xu J, Liu Z, Erhan SZ, Carriere CJ (2004) Cross-linkers control
the viscoelastic properties of soybean oil-based biomaterials.
J Am Oil Chem Soc 81:813–816
4. Tran P, Graiver D, Narayan R (2006) Biocomposites synthesized
from chemically modified soy oil and biofibers. J Appl Polym Sci
102:69–75
5. Tanrattanakul V, Saithai P (2009) Mechanical properties of bio-
plastics and bioplastic–organoclay nanocomposites prepared
from epoxidized soybean oil with different epoxide contents.
J Appl Polym Sci 114:3057–3067
6. Alemdar N, Karagoz B, Erciyes AT, Bicak N (2010) Surface
modification of silica, titania, and zinc oxide micro particles with
epoxidized soybean oil for preparation of polystyrene composite
films. J Appl Polym Sci 116:165–171
7. Zhu L, Wool RP (2006) Nanoclay reinforced bio-based elasto-
mers: synthesis and characterization. Polymer 47:8106–8115
8. Mullen CA, Boateng AA (2008) Chemical composition of bio-
oils produced by fast pyrolysis of two energy crops. Energy Fuels
22:2104–2109
9. Kleinert M, Barth T (2008) Phenols from lignin. Chem Eng
Technol 31:736–745
10. Gibson JM, Thomas PS, Thomas JD, Barker JL, Chandran SS,
Harrup MK, Draths KM, Frost JW (2001) Benzene-free synthesis
of phenol. Angew Chem Int Ed 40:1945–1948
this hydroxy fatty acid from fermentation of fats or oils was
described by us previously [17]. Isolation of the arylated
hydroxy fatty acid was straightforward (Fig. 6). We also
investigated whether substitution ortho to the OH by the
chloroacetamidomethyl group would be too sterically hin-
dering to allow ether formation at the OH. It was interesting
to observe that both the 2,6-disubstituted and 2,4-disubsti-
tuted molecules 1c and 1d gave good yields of the aryl
ethers 6c and 6b. This synthetic route has the advantage of
incorporating another unit of epoxidizable fatty acid, but
one that still possesses its carboxylic group for further
derivatization. The COOH of the hydroxy fatty acid, in
other words, was not used to make a connection to the
phenol group, so it can be employed as a ‘‘carrier’’ for some
other functional group of choice, either as an ester or an
amide. In this way, the basic phenol/fatty acid structure can
be converted to another kind of AB2 monomer, as shown in
the preparation of 8b (Fig. 6). After Mitsunobu reaction
with 17-hydroxy oleic acid to give the aryl ether with 2,4-
di(chloroacetyl) amino methyl groups 6b, the chloroacetyl
groups were removed with thiourea. This is a case where
HCl deprotection of chloroacetyl could well be unaccept-
able (hydrolyzing the methyl ester or perhaps adding across
the C=C). The deprotected amine groups were then coupled
to 17-azido oleic acid, a compound we recently reported
[17]. The resulting di-azide 8a could be used in click
reactions, although, if the products were intended for later
epoxidation and crosslinking, problems might arise with the
1,2,4-triazole that the click reaction produces: it could be
oxygenated by an epoxidation agent. Instead, we used the
Staudinger reaction to reduce the azide to NH2, and pre-
pared the Boc derivative in situ. This molecule 8b is thus set
to be used as an AB2 monomer in the preparation of hy-
perbranched polymers that are epoxidizable and crosslink-
able. In conjunction with acrylamide 4 and styrene
derivative 5c, it suggests some new directions in structural
motifs for the preparation of bioderived, fatty acid-based
polymers and materials.
11. Tsujimoto T, Ikeda R, Uyama H, Kobayashi S (2001) Cros-
slinkable polyphenols from urushiol analogues. Macromol Chem
Phys 202:3420–3425
12. Jyothish K, Vemula PK, Jadhav SR, Francesconi LC, John G
(2009) Self-standing, metal nanoparticle embedded transparent
films from multi-armed cardanol conjugates through in situ
synthesis. Chem Commun 2009:5368–5370
`
´
13. Galia M, de Espinosa LM, Ronda JC, Lligadas G, Cadiz V (2009)
Vegetable oil-based thermosetting polymers. Eur J Lipid Sci
Technol 112:97–97
14. Stokker GE, Deana AA, deSolms SJ, Schultz EM, Smith RL,
Cragoe EJ Jr, Baer JE, Ludden CT, Russo HF, Scriabine A, Sweet
CS, Watson LS (1980) 2-(aminomethyl)phenols, a new class of
saluretic agents. 1. Effects of nuclear substitution. J Med Chem
23:1414–1427
15. Molander GA, Brown AR (2006) Suzuki–Miyaura cross-coupling
reactions of potassium vinyltrifluoroborate with aryl and hetero-
aryl electrophiles. J Org Chem 71:9681–9686
16. Mitsunobu O (1981) The use of diethyl azodicarboxylate and
triphenylphosphine in synthesis and transformation of natural
products. Synthesis 1981:1–28
17. Zerkowski JA, Nun˜ez A, Strahan GD, Solaiman DKY (2009)
Clickable lipids: azido and alkynyl fatty acids and triacylglyce-
rols. J Am Oil Chem Soc 86:1115–1121
18. Zaloga GP, Siddiqui R, Stillwell W (2008) Fatty acid phenolic
conjugates. US Patent Appl. 20080004342
19. Masaki M, Kitahara T, Kurita H, Ohta M (1968) A new method
for the removal of chloroacetyl groups. J Am Chem Soc
90:4508–4509
Acknowledgments We are grateful to Dr. Gary Strahan for
recording two-dimensional NMR spectra.
20. Weitkamp P, Vosmann K, Weber N (2006) Highly efficient
preparation of lipophilic hydroxycinnamates by solvent-free
References
lipase-catalyzed transesterification.
54:7062–7068
J
Agric Food Chem
1. Sperling LH, Manson JA (1983) Interpenetrating polymer net-
works from triglyceride oils containing special functional groups:
a brief review. J Am Oil Chem Soc 60:1887–1892
2. Xu J, Liu Z, Erhan SZ (2008) Viscoelastic properties of a bio-
logical hydrogel produced from soybean oil. J Am Oil Chem Soc
85:285–290
123