Strategies for synthesis of epoxy resins from oleic acid derived from food wastes

Article abstract of DOI:10.1002/pola.28203

The use of biomass-sourced chemical feedstocks creates a conflict over land use between food and fuel/chemical production. Such conflict could be reduced by making use of the annual 1.3 Pg food waste resource. Oleic acid is available from seed oils such as pumpkin, grape, avocado and mango. Its esterification with diols 1,3-propanediol, resorcinol and orcinol was used to form diesters and the naturally occurring norspermidine was used to prepare a diamide, all under ambient conditions. These compounds were then epoxidized and polymerized. When esterification was followed by epoxidation and subsequent curing at elevated temperature with p-phenylenediamine or diethylenetriamine, hard insoluble resins were formed. When the sequence was changed such that the epoxidized oleic acid was first reacted with cis-1,2-cyclohexanedicarboxylic anhydride and then esterified with orcinol and resorcinol, insoluble crosslinked polymers were also obtained.

Full text of DOI:10.1002/pola.28203

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
Strategies for Synthesis of Epoxy Resins from Oleic
Acid Derived from Food Wastes
Theodore Hayes, Yingxue Hu, Sandra A. Sanchez-Vazquez,
Helen C. Hailes, Abil E. Aliev, Julian R. G. Evans
Department of Chemistry, University College London, 20 Gordon Street, London WC1H 0AJ, United Kingdom
Correspondence to: J. R. G. Evans; E-mail: j.r.g.evans@ucl.ac.uk
Received 22 April 2016; accepted 7 June 2016; published online 00 Month 2016published online 00 Month 2016
DOI: 10.1002/pola.28203
ABSTRACT: The use of biomass-sourced chemical feedstocks
creates a conflict over land use between food and fuel/chemi-
cal production. Such conflict could be reduced by making use
of the annual 1.3 Pg food waste resource. Oleic acid is avail-
able from seed oils such as pumpkin, grape, avocado and
mango. Its esterification with diols 1,3-propanediol, resorcinol
and orcinol was used to form diesters and the naturally occur-
ring norspermidine was used to prepare a diamide, all under
ambient conditions. These compounds were then epoxidized
and polymerized. When esterification was followed by epoxida-
tion and subsequent curing at elevated temperature with
p-phenylenediamine or diethylenetriamine, hard insoluble res-
ins were formed. When the sequence was changed such that
the epoxidized oleic acid was first reacted with cis-1,2-cyclo-
hexanedicarboxylic anhydride and then esterified with orcinol
and resorcinol, insoluble crosslinked polymers were also
obtained. C
V
2016 Wiley Periodicals, Inc. J. Polym. Sci., Part A:
Polym. Chem. 2016, 00, 000–000
KEYWORDS: biomass; epoxy resins; esterification; nanocompo-
sites; oleic acid
INTRODUCTION Petroleum is the main feedstock for mono-
The objective here was to explore the synthesis of novel
epoxy resins using a food waste, oleic acid, as the feedstock.
World annual production of mango is 36 Tg and mango seed
oil contains 40% oleic acid, grape seed oil 15–20%, pumpkin
seed oil 32% and avocado seed oil 17%, so industrial and
agricultural food waste can be a significant source of oleic
acid: consumer food waste is generally too contaminated to
be a suitable resource.
mer and polymer production and nearly 7% of world oil and
1
gas is used for their manufacture. More challenging extrac-
tion methods are likely to increase the background petro-
leum price in the long term so that biomass becomes
2
–4
increasingly viable as a chemical feedstock.
The literature
on biomass-derived polymers has been dormant while petro-
chemicals dominated the market but is now expanding to
5
–7
reflect interest in ‘sustainable’ technologies
and plant oils
Unsaturated fatty acids are obtained by treating triglycerides
with sodium hydroxide followed by separation. They have at
least one alkene functional group which may be epoxidized.
These can undergo polymerization reactions based on a suc-
cession of intermolecular cross-linkages via epoxide ring
openings. As the extent of cross-linking increases, the resin
8
,9
are a focus for this attention. The conversion of land-use
in poorer communities from food to biofuel production has
caused conflict which would recur, mutatis mutandi, for
1
0,11
12
chemical feedstock production.
It has thus been argued
that the industrial fraction of the 1.3 Pg per annum food
waste resource can be exploited for materials production
thus lessening land-use conflict. Food waste scales with pop-
ulation as does the demand for materials. Vegetable oils are
triglyceride-based, generally liquids at room temperature
16
becomes progressively more solid and less soluble.
Epoxy resins can be cured by nucleophile initiated direct
reaction between two resin molecules (homopolymeriza-
1
3
17
with 94–96 wt % fatty acids which can be differentiated
by carbon chain length and the positions of double bonds,
typically 14–22 carbon atoms and 0–3 alkenyl degrees of
tion) and ring opening by photoinitiators (cationic photo-
1
8
polymerization)
but the most common and important
methods are those which use a curing agent or hardener
(copolymerization). These can be polyfunctional amines,
1
4
unsaturation. The configuration and distribution of fatty
acids influences the physical and chemical properties of
1
9
cyclic anhydrides, or polyphenols. The many types of epoxy
resins, curing processes, and additives (such as plasticizers
1
5
materials produced from them.
Additional Supporting Information may be found in the online version of this article.
2016 Wiley Periodicals, Inc.
V
C
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 00, 000–000
1

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
and reactive diluents) give rise to materials with a wide
range of properties in applications, such as paints, adhesives,
and composites
The two main classes of crosslinking agents applied to epox-
ides for polymerization are amines and anhydrides. Such
1
6,20
37,38
reactions have been widely studied.
The amine curing
3
7
process takes place at the stoichiometric ratio and room
temperature as amines exhibit high nucleophilicity. Anhy-
dride curing has been studied for the last 50 years. Usually
Natural unsaturated fatty acids tend to occur as cis-iso-
mers. Oil is extracted from seed waste by pressing and
oleic acid can be isolated by low temperature crystalliza-
tion. Epoxidation of oleoyl unsaturations can be achieved
by the action of peroxides or peracids. Two common
reagents used for this purpose are meta-chloroperbenzoic
39
an initiator is used as the epoxy/anhydride does not react
directly and therefore ring opening of the anhydride curing
4
0,41
reagent is required.
In this case, a tertiary amine was
used as initiator (triethylamine).
2
1
15
acid (mCPBA) and hydrogen peroxide (H O ). Previous
2
2
investigations have found that epoxidation using H2O2
works best in the presence of an organic acid such as ace-
tic or formic acid, which forms a peracid with the perox-
The use of biomass for preparing structural resins is not
new. In the interwar years, Henry Ford invested heavily in
4
2
the use of soybean oil in vehicle manufacture. The use of
smectite clays for reinforcing polyamides also originated in
1
5,22
ide.
The carboxylic group in oleic acid could
4
3
the motor industry in 1985 and such polymer–clay nan-
potentially be used to replace added carboxylic acids for in
4
4
2
3
composites have expanded to provide materials with supe-
rior elastic modulus, strength, heat distortion, and barrier
properties. Three types of composite are seen: (a) a conven-
tional composite in which the polymer does not enter the
galleries, (b) an intercalated nanocomposite in which the
polymer distributes itself throughout the galleries but the
clay layers remain stacked, and (c) an exfoliated nanocompo-
site in which clay layers are dispersed and separated. The
last is preferred as it provides superior properties but
actually most nanocomposites contain both exfoliated and
intercalated fractions. Such nanocomposites can be made by
melt intercalation, solution intercalation and in situ intercala-
tive polymerization in which the unreacted polymer precur-
sors are hosted by the galleries before reacting. Such
composites, prepared from food waste derived polymers
offer the prospect of a composite material with extremely
low embodied fossil carbon.
situ peracid generation as, according to Fong et al., pera-
cid could be generated when the ester group and hydrogen
peroxide are combined. However, this reaction can only
happen either at a pH of around 10 or in the presence of
2
3,24
catalyst.
Epoxy resins with two epoxide groups per molecule can be
cured using polyamines without needing a catalyst or sol-
1
9,22
vent.
Possible polyamine hardeners include triethylene-
tetramine (TETA) and p-phenylenediamine (PPD). Diepoxides
that may be appropriate for this procedure are epoxidized
dioleates (diesters of oleic acid) or dioleamides (diamides of
oleic acid). Dioleates may be prepared at room temperature
via a Steglich esterification between oleic acid and a diol
such as orcinol or resorcinol, using 4- (dimethylamino) pyri-
2
5
dine (DMAP) and N,N’-dicyclohexylcarbodiimide (DCC).
2
6
Reaction of the extract from aloes, or the plant resin asa-
2
7
foetida with potassium hydroxide can be used to produce
orcinol or resorcinol respectively. Dioleamides may be pre-
pared at room temperature by reaction of oleoyl chloride
In this work, two approaches to form oleic acid-based poly-
mers have been used. First, oleic acid was coupled to diols
and a diamine to give two alkenyl groups per molecule then
epoxidized and polymerized. In an alternative approach oleic
acid was first epoxidized then reacted with anhydride to
crosslink and then esterified.
2
8–30
with a polyamine.
Norspermidine possesses two primary amines and one steri-
cally hindered secondary amine group. It can therefore be
used to prepare a dioleamide. Although toxic, norspermidine
occurs naturally in a number of plants, algae, and bacteria,
EXPERIMENTAL
3
1
32
33
such as Lucerne, red algae, and V. parahaemolyticus.
Materials and Measurements
The major component of mango butter is oleic acid (40–
One of the most straightforward methods to produce oleic
acid-based polymers is via homopolymerization: epoxidation
of the acid followed by crosslinking directly between the
epoxide groups. Previous investigations have shown that
these processes often demand high curing temperatures of
4
6%) which can be separated from other components
mainly stearic acid at 40–45% with palmitic, linoleic and
arachidic acids together ꢀ10%) by crystallization of satu-
(
45,46
rated acids from aqueous acetone solution at 225 8C
to
give the oleic acid at >90% purity. In this work, commer-
cially available oleic acid at 90% purity (Sigma Aldrich
Chemical, Gillingham, UK) was used in the synthesis of
monomers. It was envisaged that impurities will be compara-
ble to those in technical grade oleic acid and will not influ-
ence the epoxidation reaction.
100–200 8C applied for periods of 1–4 h for effective cross-
linking to be achieved, but that a solvent is not usually
3
4
needed. Homopolymerization also requires a Lewis base
initiator to proceed such as piperidine, p-phenylenediamine
1
7,34
(
PPD) or triethylamine (NEt
3
).
One promising route to
thermosetting materials from biomass is the crosslinking of
3
5
epoxidized sucrose esters of fatty acids for uses in coat-
ings, adhesive, and composites and tensile strengths of 18
Orcinol, resorcinol, oleic acid (90%), 4-(dimethylamino) pyri-
0
dine, N,N -dicyclohexylcarbodiimide, hexane, ethyl acetate,
3
6
MPa have been obtained.
norspermidine, pyridine, benzene, oleoyl chloride, acetone,
2
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 00, 000–000

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
meta-chloroperbenzoic acid, diethylenetriamine, titanium
IV) n-butoxide (99%) and p-phenylenediamine were pur-
chased from Sigma Aldrich Chemical, Gillingham, UK. Trie-
(dimethylamino)pyridine (DMAP) (0.03 equiv.) and then
N,N -dicyclohexylcarbodiimide (DCC) (2.4 equiv.). The reac-
0
(
tion was stirred for 24 h at room temperature, the solid
formed was removed by filtration and the solvent removed
in vacuo. Either crude material was directly carried through
to the next epoxidation step or was purified by column chro-
matography as detailed below.
V
R
thylamine, Geduran silica gel 60 and 30% hydrogen
peroxide were from Merck KGaA, Darmstadt Germany. For-
mic acid (98%) was from VWR Lutterworth, UK. Toluene
and dichloromethane were purchased from Fisher Scientific,
Loughborough, UK. 1,3-Propanediol and cis-1,2-cyclohexane-
dicarboxylic anhydride were purchased from Alfa Aesar. Deu-
terated chloroform CDCl3 (99.8%) was purchased from
Apollo scientific, Heysham, UK. The montmorillonites used
were Nanofil 116VR and CloisiteVR 10A (Rockwood Company,
4
7
Oleoate propyl diester diol
The reaction was carried out using oleic acid (69 g, 0.22
mol; 90% purity), 1,3-propanediol (7.6 g, 0.10 mol), DMAP
(
(
3.6 g, 0.03 mol), and DCC (49.4 g, 0.24 mmol) in CH Cl2
200 mL). The urea precipitate formed was removed by fil-
2
Southern Clay Products, TX) and in some cases were modi-
fied as described below. Other reagents were used as sup-
plied without modification or purification.
tration and product isolated from the filtrate (32 g, 53%).
NMR spectroscopic analysis indicated that the product was
isolated in >90% purity so it was taken directly through to
1
IR spectra were taken on a Bruker-FTIR-ALPHA Platinum
ATR with a single reflection diamond ATR module with a
scan rate of 16 and a wavelength range from 4000 to 600
the next epoxidation step. H-NMR (300 MHz; CDCl
3
) d
5.29–5.35 (4H, m, 2 3 CH@CH), 4.14 (4H, t, J 6.0 Hz, 2 3
CH OCO), 2.29 (4H, t, J 7.5 Hz, 2 3 CH CO), 1.93–2.03 (10H,
2
2
2
1
m, 4 3 5CHCH , CH CH O), 1.58–1.63 (4H, m, 2 3 CH ),
cm . A Siemens D500 diffractometer with Cu-Ka radiation
2
2
2
2
1
1
.19–1.34 (40H, m, 20 3 CH ), 0.87 (6H, t, J 6.0 Hz, 2 3
(k 5 1.54056 Å) was used for X-ray diffraction measure-
2
CH CH ).
ments with a scanning step of 0.058 and a scan time of 4 s
per step. Mass spectra were recorded on a Waters Acquity
uPLC SQD using HPLC grade water and acetonitrile (both
with 0.1% formic acid) as the solvents. Solution NMR spectra
2
3
Orcyl dioleate
The reaction was carried out using oleic acid (17.28 g, 55.1
mmol; 90% purity), orcinol (3.10 g, 25.0 mmol), DMAP
were recorded on Bruker spectrometers AMX300, Avance
(
0.092 g, 0.75 mmol) and DCC (12.4 g, 60.0 mmol) in CH Cl2
1
2
5
3
00 and Avance III 600, with H-NMR Larmor frequencies of
00, 500, and 600 MHz, respectively. Chemical shifts (in
(
100 mL). The product was purified by column chromatogra-
phy (hexane/ethyl acetate 30:1) to give orcyl dioleate as a
pale yellow oil (7.59 g, 47%). H-NMR (500 MHz; CDCl ) d
ppm) relative to tetramethylsilane (TMS) were calibrated
1
1
3
using the residual solvent peak ( H 7.26 ppm in CDCl ). Cou-
3
6.77 (2H, s, ArH), 6.68 (1H, s, ArH), 5.30–5.41 (4H, m, 2 3
pling constants (J) were measured in Hertz (Hz) and multi-
CH@CH), 2.52 (4H, t, J 7.5 Hz, 2 3 CH CO), 2.35 (3H, s,
1
2
plicities for H-NMR coupling are expressed as s (singlet), d
ArCH ), 1.91–2.09 (8H, m, 4 3 5CHCH ), 1.68–1.78 (4H,
3
2
(
doublet), t (triplet), quint (quintet), and m (multiplet). A
quint, J 7.5 Hz, 2 3 CH CH CO), 1.19–1.44 (40H, m, 20 3
2
2
Bruker Avance 300 NMR spectrometer with a 7.05 T wide-
bore magnet and a 4 mm magic-angle spinning (MAS) probe
was used to record solid-state NMR spectra. High-resolution
CH ), 0.84–0.93 (6H, t, J 7.0 Hz, 2 3 CH CH ); m/z (CI1)
2
2
3
1
1
653 (M , 67%), 265 (CH (CH ) HC@CH(CH ) CO , 100%).
3 2 7 2 7
1
3
solid-state C-NMR spectra at 75.5 MHz were recorded at
ambient probe temperature using cross-polarization (CP),
MAS (8 kHz), high-power proton decoupling (HPPD), and
total suppression of spinning sidebands (TOSS). Typical
Resorcyl dioleate
The reaction was carried out using oleic acid (17.3 g, 55.1
mmol; 90% purity), resorcinol (2.77 g, 25.2 mmol), DMAP
(0.092 g, 0.75 mmol) and DCC (12.4 g, 60.0 mmol) in CH Cl2
(100 mL). The product was purified by column chromatogra-
phy (hexane/ethyl acetate 30:1) to give resorcyl dioleate as
a pale yellow oil (13.2 g, 82%). H-NMR (500 MHz; CDCl ) d
7.34 (1H, t, J 8.0 Hz, ArH), 6.95 (2H, d, J 8.0 Hz, ArH), 6.89
(1H, s, ArH), 5.30–5.41 (4H, m, 2 3 CH@CH), 2.53 (4H, t, J
6.8 Hz, CH CO), 1.93–2.10 (8H, m, 4 3 @CHCH ), 1.73 (4H,
2
1
acquisition conditions were: H 908 pulse duration 2.45 ls;
1
3
contact time 2 ms; recycle delay 5 s. C CPMAS TOSS spectra
with nonquaternary suppression (NQS) were also acquired
with a 50 ls dephasing delay to suppress signals from
1
3
motionally rigid CH and CH carbons. For direct detection of
2
1
3
C nuclei with HPPD, the ARINGDEC pulse sequence was
2
2
used with suppression of probe ringdown effects, which also
suppresses the background signal. Typical acquisition condi-
quint, J 6.8 Hz, 2 3 CH CH CO), 1.19–1.46 (40H, m, 20 3
2 2
CH ), 0.87 (6H, t, J 7.0 Hz, 2 3 CH CH ); m/z (CI1) 639
2
2
3
1
3
1
1
tions were: C 908 pulse duration 5 3.0 ls; recycle delay 5
(M , 71%), 265 (CH (CH ) (CH) (CH ) CO , 100%).
3 2 7 2 2 7
1
3
1
5 s. The solid-state C-NMR chemical shifts are expressed
4
8
Norspermidine dioleamide
To norspermidine (1.28 g, 9.75 mmol) and pyridine (1.65 g,
0.9 mmol) in benzene (50 mL), oleoyl chloride (6.03 g, 20.0
relative to TMS using the resonance line at 176.46 ppm for
the a polymorph of glycine as an external standard.
2
Monomer Synthesis and Purification
Diesters
To oleic acid (2.2 equiv.), and the diols 1,3-propanediol (1.0
equiv.), orcinol (1.0 equiv.), or resorcinol (1.0 equiv.) in
mmol) was added dropwise at ambient temperature. The
solution was stirred for 15 min and the resulting solid sepa-
rated by filtration under reduced pressure and then recrys-
talized from acetone (twice) to give norspermidine
dioleamide as a yellow solid (2.63 g, 41%). Mp 248–256 8C
dichloromethane (CH Cl ) at
0
8C was added 4-
2
2
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 00, 000–000
3

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
1
(
acetone); H-NMR (500 MHz; CDCl ) d 6.78–6.90 (2H, br s,
m, 2 3 CH CHOCHCH ), 1.18–1.47 (40H, m, 20 3 CH ), 0.90
2 2 2
(6H, t, J 7.0 Hz, 2 3 CH CH ); m/z (ES1) 685 (M , 100%).
2 3
3
1
RCONH), 5.30–5.40 (4H, m, 2 3 CH@CH), 3.40 (4H, t, J 7.0
Hz, 2 3 CONHCH ), 2.96 (4H, t, J 7.0 Hz, 2 3 CH NH), 2.25
4H, t, J 6.5 Hz, 2 3 CH CO), 2.13 (4H, quint, J 7.0 Hz, 2 3
2
2
Resorcyl-di-9,10-epoxystearate
(
2
To resorcyl dioleate (4.78 g, 7.48 mmol) in chloroform (40
mL) mCPBA (9.18 g, 41.0 mmol; 77% purity) was added.
The mixture was stirred at room temperature for 1.5 h. The
resulting precipitate of meta-chlorobenzoic acid by-product
was removed by filtration, and the filtrate was washed with
NHCH CH ), 1.92–2.09 (8H, m, 4 3 @CHCH ), 1.71–1.80 (1H,
2
2
2
br s, CH NHCH ), 1.66 (4H, quint, J 6.5 Hz, 2 3 CH CH CO),
2
2
2
2
1
.18–1.40 (40H, m, 20 3 CH ), 0.86 (6H, t, J 7.0 Hz, 2 3
2
1
CH CH ); m/z (ES1) 660 (M , 100%), 396 (RCONH(CH )
2 3
NH(CH ) NH , 44%).
2
3
1
2
3
5
% sodium hydrogen carbonate, dried (MgSO ) and the
4
Epoxidation Reactions
Two epoxidation methods were used, with either hydrogen
solvent removed in vacuo to yield resorcyl-di-9,10-
epoxystearate (3.86 g, 77%) as a pale yellow wax. Mp 27–33
1
5,49
1
peroxide
or meta-chloroperbenzoic acid: the highest
8C; H-NMR (500 MHz; CDCl ) d 7.37 (1H, t, J 8.5 Hz, ArH),
3
yielding preparations are described below:
6.97 (2H, d, J 8.5 Hz, ArH), 6.92 (1H, s, ArH), 2.88–2.96 (4H,
m, 2 3 CHOCH), 2.52 (4H, t, J 6.5 Hz, 2 3 CH CO ), 1.67–
2
2
9
,10-Epoxystearic acid
1
.78 (4H, m, 2 3 CH CH CO ), 1.48–1.58 (8H, m, 2 3
2 2 2
To a stirred solution of oleic acid (90%; 3.57 g, 11.3 mmol)
in toluene (30 mL) was added formic acid (95%; 1.9 mL,
2 2 2
CH CHOCHCH ) 1.18–1.48 (40H, m, 20 3 CH ), 0.84–0.95
1
(
6H, m, 2 3 CH CH ); m/z (CI1) 671 (M , 59%), 281
2 3
47.8 mmol) and the reaction was heated to 80 8C. Hydrogen
1
(CH (CH ) CHOCH(CH ) CO , 100%).
3
2 7
2 7
peroxide solution was then added dropwise (54.8 mL, 546
mmol; 30% solution). The reaction was then cooled to room
temperature, separated, and the organic phase washed with
Norspermidyl di-9,10-epoxystearamide
To a stirred solution of norspermidine dioleamide (0.970 g,
1.47 mmol) in toluene (20 mL) was added formic acid
(95%; 3.3 mL, 83.1 mmol), and the reaction was heated to
80 8C. Hydrogen peroxide solution was added dropwise
(5.95 g, 52.5 mmol; 30% solution). The reaction was then
cooled to ambient temperature, separated and the organic
water (3 3 20 mL), dried (MgSO
4
), and evaporated to give
9
,10-epoxystearic acid (1.38 g, 41%) as a pale yellow waxy
5
0
1
solid. Mp 56–59 8C (lit., Mp 59.5 8C ); H-NMR (500 MHz;
CDCl ) d 9.27–11.18 (1H, br s, COOH), 2.91 (2H, m, CHOCH),
3
2
.35 (2H, t, J 6.5 Hz, CH CO H), 1.64 (2H, quint, J 6.5 Hz,
2 2
CH CH CO H), 1.42–1.56 (4H, m, CH CHOCHCH ), 1.16–1.45
phase washed with water (3 3 30 mL), dried (MgSO ), and
2
2
2
2
2
4
(
20H, m, 10 3 CH ), 0.87 (3H, t, J 7.0 Hz, CH CH ); m/z
2 2 3
evaporated to give norspermidyl di-9,10-epoxystearamide
2
1
(ES2) 297 ([M–H] , 100%).
(
0.61 g, 60%) as a yellow paste. H-NMR (500 MHz; CDCl )
3
d 6.58–7.08 (2H, br s, RCONH), 3.46 (4H, t, J 6.0 Hz, 2 3
CONHCH ), 2.94–3.00 (4H, m, CH NHCH ), 2.87–2.97 (4H, m,
9
,10-Epoxystearate propyl diester diol
To a stirred solution of oleate propyl diester diol (6.08 g,
0.0 mmol) in toluene (5 mL) was added formic acid (0.74
2
2
2
CHOCH), 2.29 (4H, t, J 6.5 Hz, 2 3 CH
2
CONH), 2.11 (4H,
1
quint, J 6.5 Hz, 2 3 NHCH CH ), 1.74–1.85 (1H, br s,
2
2
mL, 20 mmol). Hydrogen peroxide solution was then added
dropwise (2 mL, 20 mmol; 30% solution), followed by rais-
ing the temperature slowly to 80 8C. The reaction was cooled
to room temperature, separated and the organic phase
CH NHCH ), 1.60–1.69 (4H, quint,
J
6.5 Hz,
2
3
2
2
CH CH CONH), 1.32–1.44 (8H, m, CH CHOCHCH ), 1.18–1.32
2
2
2
2
(
40H, m, 20 3 CH
2
), 0.86 (6H, t, J 7.0 Hz, CH
2 3
CH ); m/z
1
(ES1) 692 (M , 25%).
washed with water (2 3 50 mL), dried (MgSO ), and evapo-
4
rated to give 9,10-epoxystearate propyl diester diol (3.18 g,
Amine-Epoxy Reactions
5
0%; ꢀ50% purity with oleate propyl diester diol remain-
For the homopolymerization of 9,10-epoxystearic acid, 1 eq.
and PPD/NEt3 initiator (0.05 eq.) were mixed, placed in a
glass sample tube, and heated in an oven at 100 8C/150 8C/
200 8C for 1 h or 5 h.
1
ing) as a pale yellow wax/syrup. H-NMR (500 MHz; CDCl )
3
d 4.14 (4H, t, J 6.5 Hz, 2 3 CH OCO), 2.91 (4H, m, 2 3
2
CHOCH), 2.33 (4H, m,
2 3 CH CO ), 1.96 (2H, m,
2 2
CH CH OCO), 1.55–1.70 (8H, m, 4 3 CH ), 1.16–1.54 (40H,
2
2
2
m, 20 3 CH ), 0.91 (6H, t, J 6.7 Hz, 2 3 CH CH ).
For the epoxy dioleates, two amine curing agents were used,
2
2
3
(
(
i) diethylenetriamine (DET) and (ii) p-phenylenediamine
PPD) using the following procedures. The diepoxy com-
Orcyl-di-9,10-epoxystearate
To orcyl dioleate (1.26 g, 1.93 mmol) in chloroform (20 mL)
mCPBA (2.21 g, 9.86 mmol; 77% purity) was added. The
mixture was stirred at room temperature for 1.5 h. The
resulting precipitate of meta-chlorobenzoic acid by-product
was removed by filtration, and the filtrate was washed with
pounds and diamine curing agents were mixed in 1:1 equiva-
lents at ambient temperature until a homogeneous mixture
was obtained. In some cases, it was left to react for up to
7
1
days (0.6 Ms) and in others it was heated in an oven at
50 8C for 16 h.
5% sodium hydrogen carbonate, dried (MgSO ) and the sol-
4
vent removed in vacuo to yield orcyl-di-9,10-epoxystearate
Anhydride Curing Reactions
Initially, the anhydride curing step using cis-1,2-cyclohexane-
1
(
1.15 g, 87%) as a yellow oil. H-NMR (500 MHz; CDCl3)
d 6.79 (2H, s, ArH), 6.71 (1H, s, ArH), 2.92 (4H, m, 2 3
CHOCH), 2.53 (4H, t, J 7.5 Hz, 2 3 CH CO ), 2.36 (3H, s,
3
dicarboxylic anhydride (CH) with triethylamine (NEt ) as ini-
tiator was applied after esterification and epoxidation. The
epoxidized monomer, CH and NEt3 were mixed in molar
2
2
ArCH ), 1.74 (4H, quint, J 7.5 Hz, 2 3 CH CH CO ), 1.51 (8H,
3
2
2
2
4
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 00, 000–000

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
TABLE 1 Homopolymerization of Epoxystearic Acid
Initiator
Time/h
Temp/ 8C
% Ring Opening
Color
Form
PPD
PPD
PPD
PPD
PPD
PPD
1
1
1
5
5
5
1
1
1
5
5
5
100
150
200
100
150
200
100
150
200
100
150
200
52
Dark brown
Dark brown
Dark brown
Dark brown
Dark brown
Dark brown
Pale yellow
Brown-orange
Brown-orange
Pale yellow
Brown-orange
Brown-orange
Soft wax
92
Viscous liquid
Viscous liquid
Viscous liquid
Viscous liquid
Brittle solid
99
70
97
N/A
49
NEt
NEt
NEt
NEt
NEt
NEt
3
3
3
3
3
3
Soft wax
90
Viscous liquid
Viscous liquid
Viscous liquid
Viscous liquid
Rubbery solid
98
76
97
N/A
ratios of 0.5/0.5/0.0085, respectively. The curing time and
modified). Clays (iii), (iv) and (v) were prepared in this labo-
ratory by Mr Illya Litvinov following the procedure recom-
mended by Akelah et al. A Siemens D500 diffractometer
1
5
temperature were prompted by the work of Nicolau et al.
5
1
53
which used 165 8C for 3 h and by Rosch et al., who used
00 8C for 1 h. Products were stored in a vacuum desiccator
2
with Bragg-Brentano geometry and silicon (111) presample
at room temperature.
monochromator was used with Cu Ka radiation with wave-
1
length 1.54056 Å.
In subsequent experiments, the epoxidized but unesterfied
oleic acid (9,10-epoxystearic acid) was reacted with anhy-
dride curing agent. The reaction product was subsequently
mixed with one of the diols (1,3-propanediol, orcinol, or res-
orcinol) in a round bottom flask in a molar ratio of 2:1. The
mixture was raised to 160 8C by use of a silicone oil bath
and 0.01 mL of titanium (IV) n-butoxide was added to the
mixture. Reaction took place for 1 h after which the tem-
perature was decreased to 120 8C for 30 min in order to
remove all water produced from the esterification reaction.
Solubility tests were performed on the polymers thus
obtained.
RESULTS AND DISCUSSION
Products of Homopolymerization of Epoxystearic Acid
The epoxidation of oleic acid was carried out as described in
the experimental and below. Homopolymerization was car-
ried out using PPD and NEt . The extent of epoxide ring
opening was estimated using H-NMR spectroscopy by divid-
ing the integral of the signal from remaining epoxy protons
over the integral from the CH COOH signal (equivalent to
two protons) and then subtracting that value from one.
Results for homopolymerization of epoxystearic acid as a
function of initiator, curing temperature and curing time
4
6
3
1
5
2
2
Interaction with Smectite Clays
(
Table 1) showed that samples heated at 200 8C for 5 h
XRD was used to assess swelling in order to explore the via-
bility of forming polymer–clay nanocomposites by in situ
polymerization. Samples of (i) clay, (ii) clay 1 methanol, and
formed nonsticky solids. They were insoluble in all solvents
tested and could be heated above 200 8C without melting,
indicative of substantial curing. Materials made with PPD
(
iii) clay 1 monomer 1 methanol were made using 1 g of
were noticeably harder and more brittle than the NEt equiv-
3
monomer, 1 g of clay and 10 mL of methanol. All the sam-
ples were exposed for 10 min to the ultrasonicator at a duty
cycle 0.5 with constant amplitude of 60% with the exception
of organoclay which was hand mixed. A control with solvent
and clay alone were used to establish the effect of solvent on
the clay. All samples were dried at 60 8C overnight to
remove solvent.
alent, which was more rubbery.
Increasing the temperature from 100 to 150 8C and extending
the reaction time from 1 to 5 h at 100 8C produced a substan-
tial rise in epoxy ring opening. This time–temperature
dependence of ring opening agrees with literature reports.
As with polymers produced from diepoxides (vide infra), sam-
3
4
ples that were cured to a greater extent had a darker color a
trend that was much less pronounced with NEt initiation
3
which gave transparent, light orange-brown colored materials.
V
R
Five different clays were used: (i) Nanofil 116 , an unmodi-
fied montmorillonite (Rockwood Company, Southern Clay
V
R
Products, TX) (ii) Cloisite 10A which is modified with a
benzyl (hydrogenated tallow alkyl)dimethyl, salt (Rockwood
Company, Southern Clay Products, TX), (iii) Nanofil 116
treated with aminolauric acid (ALA modified), (iv) Nanofil
Samples obtained as waxes had similar texture and softness
to epoxystearic acid, reaction conditions of 100 8C for 1 h
being insufficient to completely melt the resin thus restrict-
ing mobility and hence the extent of ring opening. Samples
obtained as viscous liquids appeared to become more
116 treated with diethylenetriamine (DET modified) and (v)
Nanofil 116 treated with p-phenylenediamine (PPD
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 00, 000–000
5

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
1
3 3
FIGURE 1 The H-NMR spectrum (in CDCl ) of the product of polymerization of epoxystearic acid in the presence of NEt at 150 8C
for 5 h. A schematic presentation of the corresponding polymerization process is also shown.
1
viscous and less soluble as percentage ring opening
increased, fully cured resin being completely insoluble.
H-NMR spectroscopy characterization data of the product
1
9
47
(Fig. 2) was identical to that previously reported. Orcyl
and resorcyl dioleate were also generated using DCC/DMAP
Thus while homopolymerization of epoxystearic acid
requires fewer supplementary reagents and one less reaction
step, high temperature heating periods are a disadvantage.
1
couplings and were characterized using H-NMR and mass
48
spectrometry (MS). Norspermidine dioleamide was pre-
pared using standard procedures [e.g., Ref. 29] from oleoyl
chloride, norspermidine and pyridine and was characterized
5
4
In reports, initiation, by UV irradiation shows promise.
Here, subsequent work focused on epoxidized dimers which
offer a family of promising resins with considerable scope
for development.
1
by H-NMR and MS. Use of oleoyl chloride and direct cou-
pling with polyamines is a more atom efficient method for
amide synthesis than via Steglich couplings which produces
urea waste. Future work will also focus on ester synthesis
avoiding Steglich couplings. It is possible that due to the
higher nucleophilicity of the secondary amine, secondary
amide formation could also occur. However, the primary
1
Figure 1 shows the H-NMR spectrum of the product of poly-
merization in the presence of NEt at 150 8C for 5 h.
3
The product of polymerization shows two distinct signals at
4.82 and 3.57 ppm. These are characteristic for
>
CH AOACOA and >CH AOA protons, respectively. Based
X
Y
on these observations, the following structure is suggested
for the product of polymerization:
A schematic presentation of the corresponding polymeriza-
tion process is presented in the inset of Figure 1.
Monomer Synthesis: Diesters and Diamide
The diesterification reaction between 1,3-propanediol and
oleic acid using DCC/DMAP proceeded in good yield and the
FIGURE 2 Diesters and diamide synthesized.
6
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 00, 000–000

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
1
however no H-NMR signals corresponding to the diols were
observed. Formation of the epoxides was further confirmed
by MS. Epoxidation of norspermidine–dioleamide to give E in
good purity using H O was achieved in 60% yield. mCPBA
2
2
was also employed as an alternative epoxidising agent to
H2O2 for the orcyl and resorcyl dioleates and the reaction
can be performed at ambient temperature. Epoxidation by
mCPBA of the orcyl and resorcyl dioleates gave C and D in
87 and 77% yields, respectively, 12 and 8% higher yields
than when using hydrogen peroxide.
Anhydride Curing of Epoxystearic Acid Followed by
Esterification
The products of epoxidized oleates with a cyclic anhydride
1
5
as curing agent are neither crosslinked, rubber-like, or
5
1
highly flexible according to the literature. This was con-
firmed with epoxystearic acid A before using the diesters for
the reaction. Curing reactions were carried out at 165 8C for
FIGURE 3 Epoxides synthesized.
1
5
3
h, as in Nicolau’s work, and 200 8C for 1 h as in Rosch’s
5
1
work. It should be noted that the carboxylic acid group
can also react with the anhydride leading to ring opening
and subsequent polymerization.
amines are less sterically hindered. Previous work has indi-
cated that triamide formation requires higher reaction tem-
peratures and that the diamide is the kinetically favoured
4
9
product. The dioleamide product (Fig. 2) was recrystallized
twice: MS data of this purified material showed a peak at m/
z 924, corresponding to the triamide with a relative abun-
dance of <2% which was almost negligible compared to the
dioleamide molecular ion peak at 660. The diamide has pre-
Infrared spectra of reaction products of epoxystearic acid
and cis-1,2-cyclohexanedicarboxylic anhydride under these
conditions (165 8C for 3 h/200 8C for 1 h) were similar (Fig.
21
4
). Peaks in the region of 1735–1729 cm are attributed to
4
8
viously been investigated as a soil hydrophobic agent.
Some samples of orcyl and resorcyl dioleate were purified
using column chromatography while for norspermidine-
0
dioleamide, recrystallization was used. Residual N,N -dicyclo-
hexylurea (a side-product of the DCC coupling reaction) and
oleic acid could reduce the extent of polymerization: oleic
acid has only one olefin moiety per molecule and may thus
induce termination. Removal of oleic acid from crude orcyl
and resorcyl dioleate may be important due to the potential
plasticizing effect of oleic acid.
Monomer Synthesis: Epoxidation of Oleic Acid, the
Diesters, and Diamide
Initially hydrogen peroxide was used at high temperature to
5
1
carry out the epoxidation as previously reported.
The
epoxy group is not strongly characterized in the infrared
5
5
spectrum so conversion of oleic acid and its diester to
1
epoxides was monitored using H-NMR spectroscopy: epoxi-
dation was indicated by disappearance of the alkene CHs at
ꢀ
5.3 ppm and appearance of a peak at ꢀ2.9 ppm which cor-
responded to the CH protons of the epoxy group. Oleic acid
was epoxidized to give epoxide A (Fig. 3) in high purity and
5
0
4
1% yield. However, using hydrogen peroxide and a simi-
lar procedure, the epoxidized diester B was formed in 50%
yield, and 55% purity (oleate propyl diester diol remained
1
from analysis of the H-NMR spectrum). Using analogous
FIGURE 4 FT-IR of the product of the reaction of epoxystearic
acid and cis-1,2-cyclohexanedicarboxylic anhydride from (a)
procedures orcyl dioleate and resorcyl dioleate were fully
epoxidized to give epoxides C and D in 75 and 69% yields,
respectively. During epoxidation in the presence of formic
acid, the hydrogen peroxide may also cleave the ester bond,
1
165 8C for 3 h and (b) 200 8C for 1 h (above) and the H-NMR
spectrum (in CDCl ) of the product heated at 165 8C for 3 h
3
(below).
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 00, 000–000
7

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
1
according to H-NMR and FT-IR), were then esterified at 160
5
2
8
C with titanium (VI) n-butoxide using three different diols:
resorcinol, orcinol, and 1,3-propanediol, essentially making
esterification the last, rather than the first step. When 1,3-
propanediol was used, there was no change in the waxy
appearance of the product of epoxystearic acid and anhy-
dride. When orcinol was employed, upon cooling the product
solidified after 1 h at ambient temperature. During the ester-
ification reaction with resorcinol, the product solidified after
7
min producing a dark-brown polymer.
After orcinol esterification, no solubility was observed in tolu-
ene but the material was partially soluble in CDCl and DMSO.
3
For the reaction product from resorcinol, there was no detecta-
ble solubility in these solvents or in toluene after 1 week. The
absence of solubility suggested that crosslinking of the poly-
mer occurred when esterified. Thus the pathway of using a
diol to crosslink an epoxystearic acid previously reacted with
anhydride emerges as a route to a family of thermosetting
polymers much deserving of further development.
FIGURE 5 The scheme of the reaction of epoxystearic acid
with anhydride (n 5 degree of polymerization). [Color figure
can be viewed in the online issue, which is available at
wileyonlinelibrary.com.]
13
Figure 6 shows the solid-state C CPMAS TOSS NMR spec-
trum for resorcinol esterified resin which was consistent
with the conjugation of resorcinol to the polymer. In the area
of carboxylic carbons between 166 and 183 ppm, the rela-
tively strong signal at 174 ppm can be assigned to carboxylic
carbons attached to the cyclohexane ring (e.g., 173.9 ppm in
the stretching vibrations of the C=O of the carboxylic acid
2
1
group while those at 1250 and 1170 cm
correspond to
the CAOAC bonds of the ester linkages. Peaks around 3000
2
1
56
cm represent the CH asymmetric and symmetric stretch-
dimethyl cyclohexane-1,2-dicarboxylate, while the signal at
2
ing vibrations.
167 ppm is characteristic for carboxylic carbons attached to
the benzene ring. The signal at 183 ppm is due to the acid
5
7
1
In the H-NMR spectrum of anhydride-cured products
CO H carbon attached to the alkyl chain (e.g., 180.3 ppm in
2
obtained at 165 8C from epoxystearic acid (Fig. 4), signals at
58
octanoic acid ), which is attributed to unesterified CO H
2
ꢀ
5 ppm were attributed to the H protons of the ester link-
A
groups in the resorcinol esterified resin. The signal due to
the resorcyl quaternary CAO carbons is observed at 152–
age formed by the incorporation of the anhydride. Three
broad signals can be identified at 5.10, 4.99, and 4.83 ppm
with the integral intensity ratio of approximately 1:12:2. The
signals at 4.83 and 3.57 ppm are due to the homopolymer of
the epoxystearic acid (Fig. 1). The smallest signal at 5.10
160 ppm, with the signal maximum at 158 ppm. As con-
firmed by the CPMAS TOSS NQS experiments, other signals
due to the protonated aromatic resorcyl carbons are at
ꢀ
129 ppm (meta to aryl CAO) and ꢀ108 ppm (ortho/para
1
13
ppm may be due to the end-group >CH N or >CH O pro-
tons, although further studies are needed in order to verify
the origin of this signal. Broad overlapping signals between
A
A
to aryl CAO). Aliphatic >CHAO C-NMR signals can be
2.5 and 3 ppm are assigned to the CH protons of the 1,2-
substituted cyclohexane fragment in the anhydride-cured
polymer, although a fraction of these is due to unopened
oxirane protons at 2.90 ppm. The other signals are consist-
1
5
ent with the material reported by Nicolau.
1
5
The previously suggested scheme of anhydride curing reac-
tion from oleic acid is shown in Figure 5 and is in agreement
with our results described above, although the presence of
1
the NEt3 end group may need further verification as its
replacement by the hydroxyl group cannot be ruled out. The
products were soluble in chloroform, DMSO, acetone, tolu-
ene, and DMF, indicating they were not crosslinked.
It was decided to investigate whether the acid groups could
be esterified after the epoxidation step, effectively giving a
diol the role of curing agent. The products of epoxystearic
acid and anhydride obtained at 165 8C (being uncrosslinked
1
3
FIGURE 6 The solid state CPMAS TOSS C-NMR spectrum for
epoxystearic acid reacted with cis-1,2-cyclohexanedicarboxylic
anhydride and then esterified with resorcinol.
8
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 00, 000–000

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
are at 102–136 ppm and the aliphatic >CHAO carbons are
at 68–86 ppm. The remaining aliphatic methylene carbons
are at 21–44 ppm and the methyl carbons are at 15 ppm.
The orcinol–diepoxide (epoxide C) which had not been sub-
jected to column purification was cured with PPD at 150 8C
for 16 h and produced a hard solid that was insoluble in
chloroform after 2 weeks indicating that the purification
step is not essential for creating a strong solid.
FIGURE 7 Amine curing agents.
observed at 74.3 ppm. The relatively small signals at 64.9
and 19.9 ppm are assigned to the ACH OA and MeACH A
2
2
5
9
carbons of titanium (IV) n-butoxide, respectively.
The
cyclohexyl CH signals are observed at 45 ppm and remaining
Similarly when column-purified epoxides B–E were subjected
aliphatic carbons are at ꢀ14–35 ppm.
to high temperature curing with PPD by heating for 16 h at
150 8C, hard solids were formed. These conditions were
Amine Curing of Epoxy Distearates
Initially, attempts were made to polymerize diepoxystearates
effective in curing these epoxy resins whereas lower temper-
atures or heating times yielded samples that were incom-
pletely solidified. All PPD-cured diepoxide resins produced
insoluble black solids which were highly resistant to scratch-
ing or indentation. After initial curing, the samples were
tested at 200 8C where no significant signs of melting or
alteration of their characteristic features were observed sug-
gesting they had formed thermosets. These observations
imply that these samples had all formed extensively cured
epoxy resins. Once the resins had cured they appeared
strong and highly resistant to removal from the glass sample
tube. These properties suggest that the polymers have the
potential for applications as adhesives with high temperature
resistance. Initially, the epoxy resins were highly transparent,
colorless or pale yellow oils. On addition of PPD the mix-
tures turned orange-brown color and became slightly cloudy
becoming increasingly dark in color as the reactions pro-
gressed. The final products were black and opaque.
(
Fig. 3: epoxides B, C, and D) at room temperature using
diethylenetriamine (DET) and p-phenylenediamine (PPD)
curing agents (Fig. 7) with a 1:1 ratio of the amine and
epoxy group.
Secondary and primary amines are capable of forming a
linkage to one and two epoxy groups respectively. Hence,
one DET monomer can in principle link to five epoxy groups
and one PPD monomer can link four epoxy groups. PPD as
an aromatic amine, typically produces cured resins with
high chemical and thermal resistance properties but tends
to require long cure cycles at high temperatures. The FTIR
spectra of the products were similar. Absorptions in the
21
region of 3350–3400 cm
can be attributed to the NAH
bond vibration of the amine group from the curing agents,
indicating that there is residual unreacted amine. The
broadening of the signal in the region of 3200–3570 cm
21
PPD appears to be a highly effective hardener though
unfortunately it does not occur naturally. Preliminary tests
were carried out into the hardening efficacy of norspermi-
dine as curing agent however these yielded viscous liquids
as products. The molecular structure of DET is much shorter
overall than norspermidine and PPD, which also cures well,
and is more rigid than DET given its central aromatic ring.
Possibly, the terminal amine can more readily cross-link two
could also be ascribed to the stretching vibrations of AOH
groups from the opening of the oxirane ring in the curing
reaction.
1
In the room temperature cured materials the H-NMR spec-
tra showed weak signals from residual epoxy group protons
even when all curing agent appeared to have been con-
sumed; products were all in the form of waxes and were
soluble in chloroform and acetone.
Epoxidized resorcinol (epoxide D) and epoxide B cured with
PPD cured with PPD and DET which had not been subjected
to column purification were then heated at 150 8C for 16 h.
They formed hard solids which were insoluble in chloroform.
After 2 weeks with intermittent shaking a slight yellow dis-
coloration was seen suggestive of leaching of minor amounts
of low molecular weight material but the bulk remained as a
robust solid.
1
3
The solid-state C-NMR spectrum (Fig. 8) shows only a
small signal at 58 ppm, which can be attributed to either the
1
5,60
residual oxirane carbon (expected at 54–57 ppm
) or to
the end group >CHAN carbon. This confirms that the major-
ity (if not all) of epoxy rings have been opened. As con-
1
3
firmed by C CPMAS NQS TOSS spectra, aromatic CAN and
CAO peaks are at 144–162 ppm and the ester remains intact
with a carbonyl signal at 173 ppm. The aromatic CH carbons
1
3
FIGURE 8 The solid state CPMAS TOSS C-NMR spectrum of
epoxide D cured with PPD.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 00, 000–000
9

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
FIGURE 11 X-ray diffraction patterns of organoclay with oleic
acid using methanol as solvent (traces have been shifted on
the ordinate for clarity). [Color figure can be viewed in the
online issue, which is available at wileyonlinelibrary.com.]
FIGURE 9 X-ray diffraction patterns of clay as-received and clay
dispersed in methanol and then dried (traces have been shifted
on the ordinate for clarity). [Color figure can be viewed in the
online issue, which is available at wileyonlinelibrary.com.]
not provide a pathway to nanocomposite formation with
untreated montmorillonite.
epoxy-monomers when the reacting amines are not so
widely spaced apart.
The XRD traces for the clays treated with epoxystearic acid
showed that for PPD modified clay, no intercalation occurred
and d001 increased by only 0.2 Å. The d001 for ALA modified
clay actually decreased by 1.5 Å, suggesting that methanol
extraction of the aminolauric acid took place. The DET modi-
fied clay also did not accept the epoxystearic acid.
Interactions with Smectite Clay
The swelling of clays by polymer precursors indicates the
possibility of preparing nanocomposites by intercalation fol-
lowed by in situ curing and was assessed by change of the
basal plane spacing (d001) of pristine clay and four modified
clays using XRD. Methanol was used as solvent to introduce
organic species so a control trace is shown in Figure 9 for
the effect of methanol alone. The (001) peak at 7.18 (12.5 Å)
for the untreated clay is shifted slightly to 7.28 (12.2 Å) sug-
gesting methanol removes some gallery water. The metha-
nol–MMT trace therefore offers a control for all subsequent
experiments: methanol being used as solvent in each case.
Figure 11 shows XRD patterns of the commercially available
V
R
Cloisite 10A with oleic acid. This organophilic clay which is
treated with a benzyl(hydrogenated tallow alkyl)dimethyl
salt giving it an initially higher basal spacing is generally
44
regarded as a receptive host for polymers.
Oleic acid
increased d001 from 19.2 to 23.2 Å, an expansion of 21%
and a new peak appeared at 15.8 Å.
The use of untreated clay is preferred for an industrial pro-
cedure since no additional step is needed and the cost of
feedstock is lower but these advantages must be set against
the mechanical and barrier properties of the resulting com-
posite. This work indicates that the polymer precursors do
not intercalate from solution in untreated clay and even in
clay that has been pretreated to make it compatible with
epoxy resins. The commercial clay Cloisite 10A with an ini-
tially much expanded basal plane does host oleic acid in the
galleries and provides a direction to epoxy-clay nanocompo-
sites with low petroleum input. In support of this conjecture,
an epoxidized soybean oil has been shown to produce inter-
calated nanocomposites with 5 wt % of a comparable orga-
noclay that offers 3.5 times modulus and twice the tensile
Reagent oleic acid dissolved in methanol had no significant
effect on d001 after 86 ks (1 day), 2.6 Ms (1 month), or 7.8
Ms (3 months). Similarly, the epoxystearic acid produced a
single peak at 12.5 Å after 2.6 Ms (1 month) immersion (Fig.
10). It appears that intercalation of these precursors does
6
1
strength without loss of elongation at break.
CONCLUSIONS
FIGURE 10 X-ray diffraction patterns of clay treated with oleic
acid (1 and 3 months) and epoxystearic acid (1 month) using
methanol as solvent (traces have been shifted on the ordinate
for clarity). [Color figure can be viewed in the online issue,
which is available at wileyonlinelibrary.com.]
This work sets out several alternatives to the homopolymeriza-
tion of epoxystearic acid so that oleic acid from seed waste can
be incorporated into materials manufacture. Oleic acid was
esterified with propanediol, orcinol and resorcinol to provide a
precursor with two double bonds which were then epoxidized
10
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 00, 000–000

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
is using hydrogen peroxide or meta-chloroperbenzoic acid.
Similarly a dioleamide was prepared with norspermidine and
epxoidized. These precursors were cured at 150 8C using dieth-
ylenetriamine or p-phenylenediamine to produce hard insolu-
ble resins. Furthermore, the products of epoxystearic acid and
anhydride obtained at 165 8C which were not crosslinked
could then be esterified using resorcinol or orcinol so
that esterification became the last step. While neither oleic
acid nor epoxystearic acid intercalated into unmodified mont-
morillonite, an organo-clay was found that acted as a host for
oleic acid and this opens the possibility for in situ intercalation
so that a new family of epoxy nanocomposites could be pre-
pared with considerably reduced petroleum input.
19 RAPRA Technology Ltd., Recent Developments in Epoxy
Resins.: RAPRA Review Reports, Report 91, RAPRA, Shrews-
bury, UK, 1996; pp. 25–28.
2
0 P. Mohan, Polym.-Plast. Technol. Eng. 2013, 52, 107–125.
1 P. Roumanet, F. Lafleche, N. Jarroux, Y. Raoul, S. Claude, P.
2
Guegan, Euro. Polym. J. 2013, 49, 813–822.
2 V. V. Goud, S. Dinda, A. V. Patwardhan, N. C. Pradhan,
Asia-Pacific J. Chem. Eng. 2010, 5, 346–354.
3 R. A. Fong, S. N. Lewis, R. J. Wiersema, A. G. Zielske,
2
2
Glycolate ester peracid precursors, U.S. Patent No. 4,778,618.
18 Oct., 1988.
24 G. Ch aꢀ vez, R. Hatti-Kaul, R. A. Sheldon, G. Mamo, J. Mol.
Catal B: Enzym. 2012, 89, 67–72.
25 J. Zuo, S. Li, L. Bouzidi, S. S. Narine, Polymer 2011, 52,
4
503–4516.
ACKNOWLEDGMENTS
26 World of Chemicals, Orcinol [online], WOC Labs Pte. Lim-
ited, Singapore, 2013; http://www.worldofchemicals.com/chem-
icals/chemical-properties/orcinol.html [Accessed 20 Mar 2014].
The authors are grateful to John Hill for the mass spec-
trometry and to Illya Litvinov for synthesis of treated
clays. We are grateful to the National Council of Science
and Technology of Mexico (CONACYT) for providing a sti-
pend for S. A. S.-V.
2
7 Net Industries Online Encyclopedia, Resorcin [online], 2014;
http://encyclopedia.jrank.org/RAY_RHU/RESORCIN_meta_dioxy-
benzene_C5H4.html [Accessed 20 Mar 2014].
28 R. R. Mod, F. C. Magne, G. Sumrell, J. Am. Oil Chem. Soc.
1
971, 48, 254–256.
2
9 eCompound, Common reaction procedures [online], 2007;
http://www.ecompound.com/Reaction%20reference/Lab%20rat%
REFERENCES AND NOTES
2
0procedures.htm [Accessed 03 Dec 2013].
1
2
C. K. Williams, M. A. Hillmyer, Polym. Rev. 2008, 48, 1–10.
3
0 G. Astarita, B. Di Giacomo, S. Gaetani, F. Oveisi, T. R.
IBIB.2012-13, International Business Directory for Innovative
Compton, S. Rivara, G. Tarzia, M. Mor, D. Piomelli, J. Pharma-
col. Exp. Ther. 2006, 318, 563–570.
Bio-Based Plastics and Composites Nova-Institute GmbH
Huerth, Ger. and Bioplastics Magazine; 2012.
3
1 B. Rodriguez-Garay, G. C. Philips, G. D. Kuehn, Plant Phys-
3
H. J. Endres, A. Siebert-Raths, Engineering Biopolymers:
iol. 1989, 89, 525–529.
Market, Manufacturing, Properties and Applications; Hanser
3
3
2 K. Hamana, S. Matsuzaki, J. Biochem. 1982, 91, 1321–1328.
3 S. Yamamoto, S. Shinoda, M. Makita, Biochem. Biophys.
Publications, Cincinnati, OH, USA, 2011.
4
C. Michael, T. Michael, International Business Directory for
Res. Commun. 1979, 87, 1102–1108.
Innovative Bio-Based Plastics and Composites, Nova-Institute
3
4 H. Dannenberg, W. R. Harp, Anal. Chem. 1956, 28, 86–90.
GmbH, Bioplastics Magazine; 2012. ISBN : 978-3-9812027-5-5.
3
2
5 X. Pan, P. Sengupta, D. C. Webster, Biomacromolecules
011, 12, 2416–2428.
5
J. -M. Raquez, M. Del eꢀ glise, M. -F. Lacrampe, P. Krawczak,
Prog. Polym. Sci. 2010, 35, 487–509.
A. Rouilly, L. Rigal, J. Macromol. Sci. C: Polym. Rev. 2002,
C42, 441–479.
3
6 X. Pan, D. C. Webster, Macromol. Rapid Commun. 2011, 32,
6
1
324–1330.
3
3
7 G. Tillet, B. Boutevin, B. Ameduri, Prog. Polym. Sci. 2011,
6, 191–217.
7
8
X. Tong, Y. Ma, Y. Li, Appl. Catal. A: Gen. 2010, 385, 1–13.
C. K. Williams, M. A. Hillmyer, Polym. Renew. Resour. Polym.
3
8 T. M. Goulding, Epoxy Resin Adhesives, Handbook of Adhe-
Rev. 2008, 48, 1–10
sive Technology, A.Pizzi, K.L.Mittal (Eds.) Marcel Dekker, NY,
9
R. T. Mathers, J. Polym. Sci. A: Polym. Chem. 2012, 50, 1–15.
0 J. B. S. F. Filho, M. Horridge, Land Use Policy 2014, 36, 595–604.
2
3
4
003, pp. 809–824.
1
9 J. Luston, F. Vass, Adv. Polym. Sci. 1984, 56, 91–133.
0 H. Lee, K. Neville, Handbook of Epoxy Resins; McGraw-Hill
1
1 B. Eickhout, H. V. Meijl, T. V. Rheenen, Land Use Policy
007, 24, 562–575.
2
Book Company: New York, 1967; pp 5.2–5.13.
1 K. J. Saunders, Organic Polymer Chemistry; Chapman and
Hall: London, 1988; pp 424–427.
2 R. M. Wik, Henry Ford and Grass-Roots America; Univ.
12 S. A. Sanchez-Vazquez, H. C. Hailes, J. R. G. Evans, Polym.
4
Rev. 2013, 53, 627–694.
3 M. A. R. Meier, J. O. Metzger, U. S. Schubert, Chem. Sci.
Rev. 2007, 36, 1788–1802.
1
4
Michigan Press: Ann Arbor, MI, 1992; pp 148–152.
14 F. S. Guner, Y. Yaget, A. T. Erciyes, Prog. Polym. Sci. 2006,
43 A. Okada, A. Usuki, Macromol. Mater. Eng. 2006, 291, 1449–
31, 633–670.
1476.
1
5 A. Nicolau, R. M. Mariath, E. A. Martini, D. S. Martini, D.
44 B. Chen, J. R. G. Evans, H. C. Greenwell, P. Boulet, P. V.
Coveney, A. A. Bowden, A. Whiting, Chem. Soc. Rev. 2007, 37,
568–594.
Samios, Mater. Sci. Eng. C 2010, 30, 951–962.
6 C. A. May, Epoxy Resins: Chemistry and Technology, 2nd
ed.; Marcel Dekker, Inc.: New York, 1988; 10016, pp 1–3.
1
45 E. Mularczyk, J. Drzymala, Sep. Sci. Technol. 1989, 24,
17 N. Galego, A. Vazquez, R. J. J. Williams, Polymer 1994, 35,
151–155.
857–861.
46 G. Haraldsson, J. Am. Oil Chem. Soc. 1984, 61, 219–222.
1
8 W. F. Schroeder, S. V. Asmussen, M. Sangermano, C. I.
47 L. Maisonneuve, T. Lebarbe, T. H. N. Nguyen, E. Cloutet, B.
Gadenne, C. Alfos, H. Cramail, Polym. Chem. 2012, 3, 2583–2595.
Vallo, Polym. Int. 2013, 62, 1368–1376.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 00, 000–000
11

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
4
8 R. G. Bistline, W. M. Linfield, W. B. Wise, P. E. Pfeffer, P. E.
55 J. Coates, Interpretation of Infrared Spectra; A Practical
Approach, Encyclopedia of Analytical Chemistry; 2000, pp
Sonnet, E. G. Piotrowski, J. Am. Oil Chem. Soc. 1987, 64, 744–748.
1
0815–10837.
49 B. R. Moser, S. Z. Erhan, Eur. J. Lipid Sci. Technol. 2007,
109, 206–213.
56 D. E. James, J. K. Stille, J. Org. Chem. 1976, 41, 1504–1511.
5
1
0 D. Swern, G. N. Billen, J. T. Scanlan, J. Am. Chem. Soc.
948, 70, 1226–1228.
57 F. Busque, P. de March, M. Figueredo, J. Font, Tetrahedron
1995, 51, 1503–1508.
5
1 J. Rosch, R. Mulhaupt, Polym. Bull. 1993, 31, 679–685.
58 J. V. Paukstelis, E. F. Byrne, J. Org. Chem. 1977, 42, 3941–3944.
59 B. S. Buyuktas, Transit. Metal Chem. 2006, 31, 786–791.
5
2 J. Ma, Y. Pang, M. Wang, J. Xu, H. Ma, X. Nie. J. Mater.
Chem. 2012, 22, 3457–3461.
6
0 N. D. Sachinvala, D. L. Winsor, R. K. Menescal, I. Ganjian,
5
1
3 A. Akelah, P. Kelly, S. Qutubuddin, A. Moet. Clay Minerals.
994, 29, 169–178.
W. P. Niemczura, M. H. Litt, J. Polym. Sci. A: Polym. Chem.
1998, 36, 2397–2413.
54 M. Sangermano, G. Malucelli, G. Delleani, A. Priola, Polym.
61 V. Tanrattanakul, P. Saithai, J. Appl. Polym. Sci. 2009, 114,
Int. 2007, 56, 1224–1229.
3057–3067.
12
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 00, 000–000