Bio-Based Alternative to the Diglycidyl Ether of Bisphenol A with Controlled Materials Properties

Article abstract of DOI:10.1021/acs.biomac.5b00014

A series of biobased epoxy monomers were prepared from diphenolic acid (DPA) by transforming the free acid into n-alkyl esters and the phenolic hydroxyl groups into diglycidyl ethers. NMR experiments confirmed that the diglycidyl ethers of diphenolates (DGEDP) with methyl and ethyl esters have 6 and 3 mol % of glycidyl ester. Increasing the chain length of DGEDP n-alkyl esters from methyl to n-pentyl resulted in large decreases in epoxy resin viscosity (700-to-11 Pa·s). Storage modulus of DPA epoxy resins, cured with isophorone diamine, also varied with n-alkyl ester chain length (e.g., 3300 and 2100 MPa for the methyl and n-pentyl esters). The alpha transition temperature of the cured materials showed a linear decrease from 158 to 86°C as the ester length increases. The Young's modulus and tensile strengths were about 1150 and 40 MPa, respectively, for all the cured resins tested (including DGEBA) and varied little as a function of ester length. Degree of cure for the different epoxy resins, determined by FTIR and DSC, closely approached the theoretical maximum. The result of this work demonstrates that diglycidyl ethers of n-alkyl diphenolates represent a new family of biobased liquid epoxy resins that, when cured, have similar properties to those from DGEBA. (Graph Presented).

Full text of DOI:10.1021/acs.biomac.5b00014

Subscriber access provided by UNIV OF MD - ALBIN KUHN LIB
Article
A Bio-Based Alternative to the Diglycidyl Ether of
Bisphenol A with Controlled Materials Properties
Anthony Maiorana, Stephen M Spinella, and Richard A Gross
Biomacromolecules, Just Accepted Manuscript • DOI: 10.1021/acs.biomac.5b00014 • Publication Date (Web): 29 Jan 2015
Downloaded from http://pubs.acs.org on February 1, 2015
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Biomacromolecules is published by the American Chemical Society. 1155 Sixteenth
Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 36
Biomacromolecules
1
2
3
4
5
6
7
8
9
A BioꢀBased Alternative to the Diglycidyl Ether of
Bisphenol A with Controlled Materials Properties
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Anthony Maiorana1, Stephen Spinella1,2, Richard A. Gross1*
1 New York State Center for Polymer Synthesis, Department of Chemistry and Chemical
Biology, Rensselaer Polytechnic Institute, 110 8th Street, Troy, N.Y. 12180, USA.
2 NYU Polytechnic School of Engineering, Six Metrotech Center, Brooklyn, New York 11201,
USA.
*Corresponding Author
Richard A. Gross; Email: grossr@rpi.edu; Ph: 518ꢀ276ꢀ3734; Fax: 518ꢀ276ꢀ3405
KEYWORDS: Bioꢀbased epoxy resin, levulinic acid, diphenolic acid, DMA, tensile properties
ACS Paragon Plus Environment

Biomacromolecules
Page 2 of 36
1
2
3
4
5
6
7
8
9
Abstract
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
A series of bioꢀbased epoxy monomers were prepared from diphenolic acid (DPA) by
transforming the free acid into nꢀalkyl esters and the phenolic hydroxyl groups into diglycidyl
ethers. NMR experiments confirmed that the diglycidyl ethers of diphenolates (DGEDP) with
methyl and ethyl esters have 6 and 3 mol% of glycidyl ester. Increasing the chain length of
DGEDP nꢀalkyl esters from methyl to nꢀpentyl resulted in large decreases in epoxy resin
viscosity (700ꢀtoꢀ11 Pa.s). Storage modulus of DPA epoxy resins, cured with isophorone
diamine, also varied with nꢀalkyl ester chain length (e.g. 3300 and 2100 MPa for the methyl and
nꢀpentyl esters). The alpha transition temperature of the cured materials showed a linear decrease
from 158 to 86 °C as the ester length increases. The Young’s modulus and tensile strengths were
about 1,150 and 40 MPa, respectively, for all the cured resins tested (including DGEBA) and
varied little as a function of ester length. Degree of cure for the different epoxy resins,
determined by FTIR and DSC, closely approached the theoretical maximum. The result of this
work demonstrates that diglycidyl ethers of nꢀalkyl diphenolates represent a new family of bioꢀ
based liquid epoxy resins that, when cured, have similar properties to those from DGEBA.
ACS Paragon Plus Environment

Page 3 of 36
Biomacromolecules
1
2
3
4
5
6
Introduction:
7
8
9
High performance polymer composites represent an important class of high strength
materials. The low density and high strength of polymer composites allow for lightweight
alternatives to traditional structural materials. Demand for structural lightweight materials is
expected to grow due to increased demand for fuel efficient cars, aircraft, boats, and construction
of wind turbines.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The fabrication of advanced polymer composites typically involves the reinforcement of a
thermoset matrix with high strength fillers. Thermosets are often selected because they typically
have high glass transition temperatures (Tg), long service life, and are chemically resistant.1
Epoxy resins mixed with an appropriate hardener represent the majority of thermoset polymers.
The most ubiquitous of epoxy resins is the diglycidyl ether of bisphenol A (DGEBA). In addition
to composite manufacturing, epoxy resins are also used as adhesives and coatings. Bisphenol Aꢀ
based epoxy thermosets are typically considered high performance and the thermoꢀmechanical
properties are manipulated by changing the hardener. DGEBA is also a liquid epoxy resin which
can be processed at low temperatures and gives good wettability to reinforcing fillers.2
While the final properties of DGEBAꢀbased materials are excellent, there are concerns that
bisphenol A is an endocrine disruptor. Bisphenol A (BPA) has structural similarities to estrogens
and has been linked to increased risk for reduced fertility, onset of disease, and cancer.3,4,5 While
the effects of BPA on the endocrine system are still under study, the perceived risk has sparked a
demand for bisphenol A free products and research is underway to provide a replacement.6 The
US Federal Drug Administration has imposed a ban on BPAꢀbased packaging for infant
formula.7 In addition, BPA is petroleum derived.
ACS Paragon Plus Environment

Biomacromolecules
Page 4 of 36
1
2
3
4
5
6
7
8
9
There is increased interest among academic and industrial scientists to develop epoxy resins
from readily renewable carbon (i.e. bioꢀbased) that can eliminate or reduce the use of DGEBA.
Bioꢀbased feedstocks with rigid chemical structures are required to obtain high performance
alternatives to DGEBA. Rosins, produced in pines and some plants, are rigid structures that,
when epoxidized and cured, yield properties that are comparable to DGEBA.8 However,
epoxidized rosins are solids at room temperature which is undesirable for their processing into
cured materials.8 Similarly, epoxy resins prepared from esters of a polyol and 4ꢀhydroxybenzoic
acid are also solids at room temperature and high viscosity liquids at processing temperatures.
Incorporating various proportions of epoxy resins from 4ꢀhydroxybenzoates with DGEBA
resulted in cured epoxy resins with reduced modulus and Tg values relative to neat DGEBA.9
Epoxy resins from isosorbide exhibit excellent mechanical properties. However, their synthesis
is challenging due to the low reactivity of isosorbide secondary hydroxyl groups.10 Tannin
derivatives such as gallic acid were converted into epoxy resins that result in materials with
thermal and mechanical properties that surpass DGEBA. Though, the synthetic route to a fully
epoxidized gallic acid requires allylation and then epoxidization due to the high reactivity of the
carboxylic acid and the close proximity of aromatic hydroxyls.11,12 Thus, a bioꢀbased DGEBA
alternative prepared by a practical synthetic route and that results in high performance materials
with similar processing conditions to DGEBA has yet to be achieved.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The research described herein was motivated by the need to create biobased epoxy resins and
corresponding materials that overcome the above cited problems that have restricted the further
development of previous candidates. DGEBA is synthesized industrially through the reaction of
bisphenol A with a large excess of epichlorohydrin under alkaline conditions.13 A bioꢀbased
molecule that can be manufactured into an epoxy resin by a similar process to bisphenol Aꢀbased
ACS Paragon Plus Environment

Page 5 of 36
Biomacromolecules
1
2
3
4
5
6
7
8
9
resins and that exhibits similar properties when cured could be a dropꢀin replacement for
DGEBA. Furthermore, the use of lignocellulosicꢀbased chemicals is desired so that bioꢀbased
epoxy resin manufacturing is decoupled from biobased feedstocks that can also be used for food
production.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
An increasing number of bioꢀbased building blocks are now commercially produced or are at
pilotꢀplant scale. Solvay’s manufacturing of bioꢀbased ephichlorohydrin14,15 from glycerol (trade
name Epicerol) is now commercially available. While exposure to epichlorohydrin has serious
human health concerns, it is the most efficient pathway to convert hydroxyl moieties to epoxy
resins (one step).2 At the pilot plant scale, production of bioꢀlevulinic acid from cellulose by
Biofine16 and Segetis17 is being explored for commercialization. Levulinic acid is of interest as a
bioꢀbased platform chemical because it can undergo many chemical transformations.18
Furthermore, levulinic acid is produced from cellulose—the most abundant biomacromolecule
on the planet, by the acid hydrolysis of cellulose and resulting C6 sugars.19 Kinetic studies have
been conducted to understand how levulinic acid is formed from both monomeric sugars and
cellulose.20 Furthermore, at the lab scale, lignocellulose is being explored as a source of bioꢀ
based phenol. For example, fast pyrolysis is a promising approach for conversion of
lignocellulose to a mixture of phenol, guauicol, vanillin, and other phenolic compounds.21,22
Also, microwave assisted pyrolysis of lignocellulose with activated carbon yields an oil fraction
rich in phenol.23 Diphenolic acid (DPA) is synthesized by condensation of levulinic acid with
two equivalents of phenol. Structural similarities between diphenolic acid and BPA has been
recognized by others who are seeking a BPA replacement for polycarbonates.24,25 The relative
binding affinity of DPA to the estrogen receptor is an order of magnitude lower than BPA.26
ACS Paragon Plus Environment

Biomacromolecules
Page 6 of 36
1
2
3
4
5
6
Increases in yield and selectivity for the desired para, para isomer of DPA (essential for
synthesis of polycarbonates) have also been reported.25,27
7
8
9
Furthermore, DPA has been used to prepare high performance benzoxazine thermosets that
give excellent thermal and mechanical properties.28 However, benzoxazines suffer from difficult
processing conditions and high curing temperatures.29 The synthesis of DPA based
polycarbonates has also been reported and the ability to selectively deꢀprotect the carboxylic acid
to produce water soluble polycarbonates is advantageous.30 Epoxy resins based on diphenolate
esters was known to be used prior to the advent of bisphenol A and it’s use has been patented in
conjunction with vegetable oils for coatings.31 To the best of our knowledge, no peer reviewed
articles have been published describing the synthesis and structureꢀproperty relationships of DPA
based epoxies.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Due to structural similarities between DPA and BPA, we hypothesized that corresponding
epoxy resins and materials produced should have similar physical properties. However, DPA has
a free carboxylic acid which allows for a myriad of chemical modifications to tune material
properties. Herein we chose to prepare and study a series of short chain diphenolate nꢀalkyl
esters (see Scheme 1). Access to bioꢀbased short chain nꢀalkanols (e.g. ethanol, butanol) by
fermentation of sugars is already well established while work continues to increase the efficiency
of microbial producing strains to achieve higher product titers for both short and relatively
longerꢀchain alcohols for use in fuels and as chemicals.32,33 Furthermore, derivatives of DPA
may further inhibit binding to the estrogen receptor although further research is needed to assess
its toxicity.
Here, we report the first synthesis of epoxy resins from esters of DPA. Isophorone diamine
was selected as a hardening agent since it is a liquid and wellꢀknown to be effective for this
ACS Paragon Plus Environment

Page 7 of 36
Biomacromolecules
1
2
purpose.34 Bioꢀbased epoxy resins from diphenolate esters were synthesized and used without
purification to prepare cured resins that exhibit qualities of engineering materials. Furthermore,
variation of the nꢀalkyl ester chain length allowed tuning of the cured epoxy resin Tg values
(decreased with increasing ester chain length) without significant changes in the modulus.
Furthermore, direct comparisons of cured epoxy materials prepared from diphenolate esters to
those from DGEBA provide a compelling case for the potential future use of bioꢀbased DPA
esters.
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Results and Discussion:
Synthesis and Characterization of Epoxy Monomers
The epoxy monomers were prepared by a twoꢀstep process depicted in Scheme 1 (for detailed
procedures see experimental section). First, Fischer esterification with a large excess of the
desired alcohol led to the corresponding ester in yields of 85ꢀ90%. One spot was observed by
1
thin layer chromatography and peaks in the H NMR are consistent with the target esters.
Melting points for methyl and ethyl diphenolate were obtained by differential scanning
calorimetry (DSC) while only Tg values were observed for the nꢀbutyl and nꢀpentyl diphenolates.
The latter is explained by that the nꢀbutyl and nꢀpentyl diphenolates crystallize slowly and, after
vacuum distillation, cooling the flask causes vitrification of these esters.
ACS Paragon Plus Environment

Biomacromolecules
Page 8 of 36
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 1. Simple twoꢀstep synthetic pathway to synthesize nꢀalkyl diphenolate diglycidyl ethers
that differ in the chain length of the nꢀalkanol ester moiety. The diglycidyl ethers of alkyl
diphenolates (DGEDPꢀmethyl ester; DGEDPꢀethyl ester; DGEDPꢀnꢀbutyl ester; DGEDPꢀnꢀ
pentyl ester) were prepared.
Glycidation was performed by reaction of diphenolic esters with excess epichlorohydrin under
alkaline conditions in a homogeneous reaction solution. The resulting epoxy resins were
1
obtained in yields from 85ꢀ97%. Products were analyzed by HꢀNMR, 13CꢀNMR and high
resolution mass spectrometry (see Experimental) and their physical properties were further
characterized by viscosity and epoxy equivalent weight.
DGEDPꢀmethyl ester was first characterized by 1HꢀNMR (Figure 1), and the peak positions of
the higher intensity set of resonances were assigned based on predicted chemical shift positions
using ChemDraw. Further rational for peak assignments based on analysis of the corresponding
1Hꢀ1H COSY NMR spectrum (Figure SI 4) is given in the supporting information. In addition, a
set of relatively low intensity peaks observed at 2.61, 2.83, 3.17, 3.85, and 4.36 ppm were
assigned to glycidyl ester protons Hγb’, Hγa’, Hβa’, Hαb’, and Hαa’, respectively. Hence, DGEDPꢀ
methyl ester contains low levels of DGEDPꢀglycidyl ester (Figure 1). Further, a low intensity
peak was observed at 2.15 ppm corresponding to the CH2 alpha to the carbonyl labeled as H3b
that is located downfield from H3a. The origins of the glycidyl ester side reaction are likely due
ACS Paragon Plus Environment

Page 9 of 36
Biomacromolecules
1
2
3
4
5
6
7
8
9
to the proclivity of the methyl ester to undergo inꢀsitu hydrolysis by alkali forming the
corresponding carboxylate anion that reacts with excess epichlorohydrin (Scheme 2). In addition,
a group of low intensity broad multiplets were observed between 4.15 and 4.0 ppm and 3.78 and
3.69 ppm that correspond to protons of glycidyl moieties. The presence of these resonances
indicates that low contents of other side products form during glycidation with epichlorohydrin
by reactions such as hydrolysis of epoxy groups, incomplete dehydrochlorination, formation of
bound chlorides, oligomerization, and abnormal addition to the epichlorohydrin.34 This is also
consistent with findings by Nouailhas et al.35 who prepared diglycidyl ethers of resorcinol. All
1H NMR spectra are shown in the supporting information (SI 1).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
To further characterize the series of diglycidyl ether nꢀalkyl diphenolates synthesized herein,
1D 13CꢀNMR spectra were recorded. The 13CꢀNMR spectrum of DGEDPꢀmethyl ester along
with peak assignments is shown in Figure 2 while spectra for the other esters in the series are
given in the supporting information (SI 2). The spectrum in Figure 2 confirms that DGEDPꢀ
methyl ester is the major product that contains low levels of the corresponding glycidyl ester. To
validate assignments of 13CꢀNMR spectra, a Heteronuclear MultipleꢀQuantum Correlation
experiment was performed for DGEDPꢀmethyl ester. The HMQC spectrum and a discussion of
13
information on carbonꢀhydrogen connectivity that was used to assign CꢀNMR signals is given
in SI 5.
ACS Paragon Plus Environment

Biomacromolecules
Page 10 of 36
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. 1H NMR (500 MHz, CDCl3) of DGEDPꢀmethyl ester.
ProtonꢀNMR was used to determine molꢀ% values of glycidyl ester coproducts for the series of
diglycidyl ether nꢀalkyl diphenolate esters synthesized herein. This was accomplished by
1
comparing the relative HꢀNMR peak intensity of the methyl singlet (CH3) at 1.57 ppm to
integrations of protons alpha (Hαa’,Hαb’), beta (Hβa’) and gamma (Hγa’,Hγb’) to the glycidyl ester.
1
The DGEDPꢀmethyl and ethyl esters contained 6 and 3 mol% glycidyl ester, respectively. Hꢀ
NMR spectra for DGEDPꢀnꢀbutyl and DGEDPꢀnꢀpentyl esters are consistent with the proposed
structures in Scheme 1. Furthermore, signal intensities for glycidyl ester protons for DGEDPꢀnꢀ
butyl and DGEDPꢀnꢀpentyl esters is below the detection limit (< 1 mol%). The decrease in
glycidyl ester formation for longer chain nꢀalkyl esters is likely related to the lower volatility of
ACS Paragon Plus Environment

Page 11 of 36
Biomacromolecules
1
2
3
4
5
6
longer chain alcohols that would form by hydrolysis (Scheme 2) and the potential reversibility of
the hydrolysis reaction.
7
8
9
O
O
O
O
O
O
O
O
O
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
O
O-
O
O
Cl
O
-OH
O
O
O
O
O
O
DGEDP-Methyl Ester
Carboxylate Anion Intermediate
DGEDP-Glycidyl Ester
Scheme 2. Potential pathway for a side reaction with hydrolysis of the methyl ester to the
corresponding carboxylate anion under alkaline conditions to produce the glycidyl ester coꢀ
product
To further characterize the structure of the diglycidyl ether nꢀalkyl diphenolate ester
series, 13CꢀNMR spectra were recorded. The 13CꢀNMR spectrum of the methyl ester is shown in
Figure 2. 13CꢀNMR spectra of other members of the ester series are given in the supporting
information. To confirm assignments of 13C NMR spectra from carbonꢀhydrogen connectivity
data, a Heteronuclear MultipleꢀQuantum Correlation (HMQC, SI 5) spectrum was recorded. For
example, connectivity of protons at 4.36 and 3.84 ppm (Hαa’ and Hαb’) to the 13C NMR signal at
64.97 ppm allowed unambiguous assignment of this carbon to that of the glycidyl ester
methylene (6’). Connectivity between the glycidyl ester proton beta to the ester at 3.16 ppm
(Hβa’) and the 13C NMR signal at 49.28 confirmed the assignment of carbon 8’. Also, the glycidyl
ester protons gamma to the ester (Hγa’, Hγb’: 2.83, 2.61 ppm) showed connectivity to a grouping
13
of C NMR signals that are closely spaced between 44 and 45.5 ppm. Since we anticipate the
signal intensity of carbon 9’ to be of low intensity, this carbon was assigned to signal at 44.65
ACS Paragon Plus Environment

Biomacromolecules
Page 12 of 36
1
2
3
4
5
6
(see expanded region in Figure 2). Other assignments that show correlations between carbons
and protons in the HMQC are shown in Figure SI 5.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. 13CꢀNMR (500 MHz, CDCl3) for the DGEDPꢀmethyl ester showing glycidyl ester coꢀ
product peaks.
Mass spectra of the epoxy resins were obtained in electron spray ionization mode by direct
injection and ionization with ammonium due to poor solubility of the epoxy resins in typical
eluents as well as their poor ionization potential. Consequently, all m/z are reported with
+
addition of a NH4 ion [M + NH4]. The mass spectrum in Figure 3 has a high intensity signal at
444.2385 that corresponds to the DGEDPꢀethyl ester molecular ion (+NH4). The molecular ion
corresponding to the glycidyl ester coꢀproduct at 472.2328 m/z (glycidyl ester mass + NH4) is
ACS Paragon Plus Environment

Page 13 of 36
Biomacromolecules
1
2
3
4
also observed. In addition, the mass spectrum shows m/z peaks at 814 and 870 (dimer mass +
5
6
+
NH4 ) that correspond to formation of two different dimeric species. The dimer at m/z at 814 is
7
8
9
well known to occur as a coproduct that forms during epoxy resin synthesis via ringꢀopening of
the epoxide by a free phenolic group.34 The second dimer at m/z 870 was not expected and is
most likely formed by coupling of the aromatic rings under Fischer esterification conditions.
Indeed, the mass of this dimer corresponds to doubling of the molecular weight minus two
hydrogen atoms. Mass spectra of the other diglycidyl ether nꢀalkyl diphenolate esters are shown
in SI 6.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. Direct infusion ESI mass spectrum of DGEADPꢀethyl ester where all m/z peaks
+
contain NH4 .
ACS Paragon Plus Environment

Biomacromolecules
Page 14 of 36
1
2
3
4
5
6
7
8
9
The epoxy resins synthesized herein are liquids that exhibit Newtonian behavior under sheer
rates below 100 sꢀ1 (Figure 4a). Viscosity values reported in Table 1 were determined at 25 °C by
rotational rheology as an average from the Newtonian region. The diglycidyl ether of diphenolate
esters exhibit sheer thinning as the sheer rate was increased. The dependence of sheer rate on
sheer thinning for the series of DGEDPꢀnꢀalkyl ester chain lengths is as follows: methyl > ethyl
> butyl ≈ pentyl (Figure 4a). Viscosities decreased with increasing nꢀalkyl ester chain length.
This general trend is expected since longer ester groups increase epoxy resin flexibility.
However, the magnitude of viscosities decreases dramatically from the methyl to ethyl ester and
then in a linear fashion for each additional methylene unit in the nꢀalkyl ester (Figure 4b). Table
1 also shows that, although the viscosity of DGEBA is below that of DGEDPꢀnꢀpentyl ester (4
vs. 12 Pa.s), their values are similar.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
b
a
DGEBA
DGEDPꢀMethyl
DGEDPꢀEthyl
DGEDPꢀButyl
DGEDPꢀPentyl
Figure 4: a) Rotational rheology at 25 °C of synthesized epoxy resins compared to the DGEBA.
b) Viscosity as a function of ester length for the prepared resins with trend line produced from
diamond shaped symbols and exludes the square symbol.
ACS Paragon Plus Environment

Page 15 of 36
Biomacromolecules
1
2
3
4
5
6
Table 1: Comparison of viscosities, experimental and theoretical epoxide equivalent weight, and
the glass transition temperate (Tg) of diglycidyl ether diphenolate esters.
7
8
9
Viscosity (Pa.s)
Measured EEWb
175a
Theoretical EEWc
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
DGEBA
4
170
206
213
227
234
DGEDPꢀmethyl ester
DGEDPꢀethyl ester
DGEDPꢀnꢀbutyl ester
DGEDPꢀnꢀpentyl ester
792
63
18
12
215 ± 18
222 ± 8.6
243 ± 7.4
265 ± 3.7
a From certificate of analysis supplied by vendor.
bEpoxide equivalent weights (EEW) were determined by ASTM D1652. Standard deviation
values were calculated from n=3.
cDetermined from equations supplied by reference 33.
The disproportionately high viscosity for DGEDPꢀmethyl ester described above prompted us
to further analyze the above data as well as look for other explanations. To further explore the
extent that oligomerization occurs during the synthesis of diglycidyl ether nꢀalkyl diphenolate
ester resins, experimental and theoretical epoxide equivalent weights (EEWs) were determined.
If the experimental EEW is much greater than the theoretical value for DGEMDP, this would
indicate that oligomerization occurred and explain higher viscosities of the corresponding
products. However, for the methyl and ethyl esters, experimental and theoretical values of EEW
are indistinguishable given experimental variability and further indicate that oligomer formation
is minimal. Also, differences between experimental and theoretical EEW values for the nꢀbutyl
and nꢀpentyl esters are small (7 and 13%, respectively). Thus, the above analysis provides no
ACS Paragon Plus Environment

Biomacromolecules
Page 16 of 36
1
2
3
4
5
6
evidence for significant oligomerization for DGEDPꢀmethyl ester as well as its longer chain ester
analogs.
7
8
9
The discrepancy in viscosity for DGEDPꢀmethyl ester provided the motivation for the detailed
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
NMR investigations described above. Indeed, the H NMR showed that DGEDPꢀmethyl ester
differed from its longer chain ester analogs. The methyl and ethyl esters contain 6 and 3 mol% of
a glycidyl ester coꢀproduct, whereas, the corresponding nꢀbutyl and nꢀpentyl esters showed no
evidence (from 1H NMR) of glycidyl ester formation. It’s unlikely that 3 mol% glycidyl ester in
DGEDPꢀmethyl ester would result in an order of magnitude increase in viscosity relative to
DGEDPꢀethyl ester alone, but is rather a combination of intermolecular forces of the ester and
the glycidyl ester coꢀproduct. Furthermore, based on H and 13C NMR and qualitative
1
information from mass spectra, dimer formation occurs to low extents for all prepared resins and
1
is difficult to measure by H NMR. The low amount of dimer formation for all of the
synthesized epoxy resins does not explain the behavior of the DGEDPꢀmethyl ester. Dimer
formation primarily comes from aromatic coupling and ringꢀopening of an epoxide by a free
phenolic group, as discussed above, as opposed to oligomerization during glycidation. Further
work is planned to explain the disproportionately high viscosity of DGEDPꢀmethyl ester that will
include detailed studies of intermolecular interactions.
Fourier Transform Infrared Spectroscopy and Differential Scanning Calorimetry
Modulated and nonꢀmodulated differential scanning calorimetry (DSC) and Fourier transform
infrared spectroscopy (FTIR) were used to qualitatively determine the extent of curing that
resulted from placing resins under 2ꢀmetric tons of pressure for 4 h at 80 °C and then 4 h at 160
°C (see Experimental Section). FTIR of the uncured and cured resin samples displayed in Figure
SI 7 were analyzed by attenuated total reflectance. All the cured resins prepared herein showed a
ACS Paragon Plus Environment

Page 17 of 36
Biomacromolecules
1
2
dramatic decrease in the epoxy vibrational band at ~910 cmꢀ1 attributed to deformation of the
oxirane CꢀO group. For example, the intensity of the vibrational band at ~910 cmꢀ1 for the
DGEDPꢀnꢀbutyl ester decreased from 90% in the uncured state to a small peak that was difficult
to detect above signals from other neighboring absorption bands in the cured state. While the use
of FTIR as described above is not quantitative, it’s clear that the amount of oxirane moieties is
largely diminished. DGEDPꢀmethyl ester and the other resins had similar decreases in
magnitudes of the ~910 cmꢀ1 band after curing. Furthermore, after the cure cycle described
above, DSC analysis by successive heatꢀcoolꢀheat scans did not reveal an exothermic transition
above or below the glass transition. This shows that if additional cure occurred during DSC
heating it was below the instrumental level of detection. Analysis of nonꢀreversible exothermic
transitions by modulated DSC was also used since it is a more sensitive technique for analyzing
the degree of cure. Similarly, no significant exothermal transitions were found. Therefore,
consumption of epoxy moieties and concurrent events of crosslinking during the curing cycle
adopted herein results in a degree of cure that closely approaches its upper limit. All heating
curves are shown in SI 8.
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Dynamic Mechanical Analysis and Rheology
o
DMA temperature scans gave values of the storage modulus (25 C) and alpha transition
temperatures (related to glass transition) ascertained from the peak of loss modulus vs.
temperature (Table 2). The storage modulus of DGEDPꢀmethyl ester was about 25% higher than
DGEBA, whereas, longer DGEDP ester chain length analogs (ethyl, nꢀbutyl and nꢀpentyl), have
storage moduli that are about 15% below DGEBA. Thus, over the full range of DGEDPꢀesters,
storage moduli are of similar magnitude to DGEBA. Further comparison of DGEDPꢀmethyl
ester and DGEBA shows that, in addition to the higher storage modulus of the former, its Tg is
ACS Paragon Plus Environment

Biomacromolecules
Page 18 of 36
1
2
o
3
just 7 C below that of DGEBA. The relatively high storage modulus of DGEDPꢀmethyl ester
relative to DGEBA is explained by the presence of 6 mol% of the corresponding glycidyl ester
coꢀproduct that should lead to a higher crosslink density. Peaks of the loss modulus are shown in
SI 9
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Glass transition temperatures of diglycidyl ether diphenolate esters (both from DSC and peak
of loss modulus) decrease in a linear fashion with increasing nꢀalkyl ester chain length (Figure
5a). A similar shift in the tan δ peak with increasing nꢀalkyl ester chain length (Figure 5b) also
occurs. Through the analysis of the results of the DMA and DSC the nꢀalkyl chain ester moieties
are functioning as internal plasticizers facilitating movements such as those of polymer chain
segments slipping past each other at relatively lower temperatures. Indeed, the DGEDP ester side
chains provide a means by which the Tg can be tuned without changing the crosslinker or cure
cycle.
DGEBA
a
b
DGEDPꢀMethyl
DGEDPꢀEthyl
DGEDPꢀButyl
DGEDPꢀPentyl
Figure 5: a) Alpha transition temperature (peak of loss modulus) and glass transition
temperature as a function of ester length, b) Tan(δ) as a function of temperature.
ACS Paragon Plus Environment

Page 19 of 36
Biomacromolecules
1
2
3
4
5
6
7
8
In order to further examine whether increasing the nꢀalkyl chain length of DGADP disrupts
three dimensional network formation of corresponding epoxy thermosets, the average molecular
weight between crosslinks (Mx) was determined by oscillatory rheology (Table 2) by Equation 1
(derived from the rubber elasticity theory)36 where:
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(1)
ρ is the density (determined by ASTM D792ꢀ13), GN is the shear modulus, R is the universal
gas constant, and T is the temperature in kelvin. The shear modulus (GN) of cured samples was
measured by oscillatory rheology through averaging the frequency independent values of G’.
Plots are shown in SI 10. Measurements were obtained at a temperature equal to the tan δ peak +
50 °C. Mx values for DGEDPꢀesters and DGEBA are listed in Table 2.
If the ester sideꢀchains of DGEADP’s interrupt network formation, increases in Mx would be
observed and this can contribute to decreases in glass transition temperatures. DGEDPꢀmethyl
and ethyl esters have Mx values that are about 30% below that of DGEBA. This can be due in
part to the presence in these DGEDPꢀesters of 6 and 3 mol% DGEDPꢀglycidyl ester which,
presumably, increases the density of crosslinks in cured resins. DGEDPꢀnꢀbutyl and nꢀpentyl
esters have Mx values that are about 10% greater than DGEBA. This is consistent with that these
DGEDP esters do not have significantly quantities of the DGEDPꢀglycidyl ester. Storage
modulus values listed in Table 2 do not substantially change as the nꢀalkyl sideꢀchain length
increases from ethyl to nꢀbutyl and nꢀpentyl. Hence, we conclude that nꢀalkyl sideꢀchains up to
nꢀpentyl do not interrupt network formation.
Mx values determined by oscillatory rheology are higher than that expected for the high degree
of crosslinks attained (see FTIR and DSC results above) and give larger values for DGEBA than
ACS Paragon Plus Environment

Biomacromolecules
Page 20 of 36
1
2
those reported in the literature determined by different methods.37,38 There are several reasons
that explain this apparent discrepancy. It may be that some slippage of the plates occurs at 15 N
of normal force that is unavoidable due to the thermoset nature of the polymer. Increasing the
normal force that would reduce slippage can result in sample deformation. Furthermore, the
equation used to determine Mx has inherent assumption such as strain is homogenous throughout
the sample, deformation is affine, and steady state conditions at any strain.39 Thus, while Mx
values likely do not accurately describe the average molecular weight between crosslinks, they
do give valuable information on the relative crosslinking for a related series of samples.
Furthermore, determination of Mx by use of a shear parallel plate rheometer minimizes the
potential for other errors that can occur by other methods such as dynamic mechanical analysis
through single or dual cantilever where clamping effects become problematic since the whole
sample must be analyzed. An alternative method such as curing inꢀsitu (e.g. curing while
samples are within the rheometer) could be used. However, inꢀsitu curing, especially of high
modulus epoxy networks, is problematic since high torque is experienced during the cure cycle
that can damage the rheometer.
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Tensile Testing
Figure 6a shows stressꢀstrain curves recorded for the series of curedꢀepoxy materials prepared
from DGEDPꢀesters and Figure 6b shows the comparison of Young’s and Storage modulus as a
function of the different cured epoxy resins. The Young’s modulus, calculated from the linear
portion of the stressꢀstrain curve, the tensile strength at break as well as the percent strain at
break are listed in Table 2. Young’s modulus values for DGEBA, DGEDPꢀmethyl and ethyl
ester are statistically equivalent in value. A decrease in Young’s modulus by about 15% is
observed for DGEDPꢀnꢀbutyl and nꢀpentyl esters relative to DGEDPꢀmethyl and DGEDPꢀethyl
ACS Paragon Plus Environment

Page 21 of 36
Biomacromolecules
1
2
3
4
5
6
7
8
9
esters. This difference is attributed to the presence in DGEDPꢀmethyl and ꢀethyl esters of 6 and 3
mol%, respectively, of DGEDPꢀglycidyl ester. Indeed, similar to the trend in Young’s modulus
values, the average tensile strength of DGEDPꢀnꢀbutyl and nꢀpentyl esters is about 15% lower
than that of DGEDPꢀmethyl and ꢀethyl esters. However, the differences in tensile strength are not
statistically significant. The above described trends in Young’s modulus and tensile strength
agree with those obtained from the storage modulus except for DGEDPꢀethyl ester’s storage
modulus, which has statistically equivalent values to those of DGEDPꢀnꢀbutyl and nꢀpentyl
esters. Values of %ꢀstrain at break for the series of DGEDPꢀesters are statistically equivalent
except for DGEDPꢀmethyl ester, which is lower by about 50% (5 vs. about 10% strain at break).
This is explained by that DGEDPꢀmethyl ester contains 6 mol% of the corresponding glycidyl
ester that will increase the crosslink density of corresponding cured epoxy resins. The lower %ꢀ
strain at break for cured DGEDPꢀmethyl ester is consistent with its low Mx (1,600 g/mol),
although, cured DGEDPꢀethyl ester, which has a similar Mx but a lower content of glycidal ester,
has a similar %ꢀstrain at break as those of DGEDPꢀnꢀbutyl and nꢀpentyl esters. The most
prominent difference between the DGEDPꢀester family in Table 2 is Tg values that decrease from
158 to 86 °C with increasing nꢀalkyl ester chain length (see detailed discussion above). Most
satisfying is that, the epoxy materials generated by curing DGEBA and the DGEDPꢀesters under
identical conditions have similar mechanical properties.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Biomacromolecules
Page 22 of 36
1
2
3
4
5
a
b
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
DGEBA
DGEDPꢀMethyl
DGEDPꢀEthyl
DGEDPꢀButyl
DGEDPꢀPentyl
Figure 6: a) Stressꢀstrain curves of the cured DGEDPꢀester epoxy resins compared to cured
DGEBA b) Storage modulus (DMA) and Young’s modulus (tensile testing) of the cured
materials.
Table 2: Comparison of properties from dynamic mechanical analysis and tensile testing for the
series of cured resins prepared from the diglycidyl ether diphenolate ester resin series. Values of
standard deviation are given for n=3.
Strain
Storage
Modulus at
Glass
Young’s
Modulus
(MPa)c
Tensile
Strength
(MPa)c
Mx
Transition
at Break
(%)
(g/mol)d
25
°C
(°C)b
(MPa) a
DGEBA
2640 ± 80
3278 ± 63
165
158
1263 ± 49
35.4 ± 3.2
8.7 ± 1.7
2330
1600
DGEDPꢀ
Methyl
1233 ± 107
41.3 ± 12.0 4.6 ± 1.6
DGEDPꢀEthyl 2186 ± 40
DGEDPꢀButyl 2244 ± 76
136
96
1306 ± 69
1094 ± 10
41.7 ± 3.5
36 ± 1.7
9.3 ± 1.4
8.3 ± 2.3
1550
2660
DGEDPꢀ
2141 ± 40
Pentyl
86
1058 ± 18
34.8 ± 0.4
11 ± 1.5
2520
ACS Paragon Plus Environment

Page 23 of 36
Biomacromolecules
1
2
3
4
a Determined by DMA
5
6
7
8
b Determined from the peak of the loss modulus
c Determined by tensile testing from stress-strain plots
d Determined by oscillatory rheology
Thermal Stability
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Thermal stability was studied by thermal gravimetric analysis (TGA). TGA scans are shown
in Figure 7 and values of thermal degradation at 5%, 10%, and 50% weight loss (T5%, T10% and
T50%, respectively) are listed in Table S11. The ester length for cured DGEDPꢀester resins has
no observable effect on their thermal stability. The most significant difference in the thermal
stability of DGEBA and the DGEDPꢀesters is the onset of degradation at T10% that, for cured
DGEBA, is 15 oC higher than all of the cured DGEDPꢀester resins.
DGEBA
DGEDPꢀMethyl
DGEDPꢀEthyl
DGEDPꢀButyl
DGEDPꢀPentyl
Figure 7: TGA scans for cured DGEDPꢀesters and DGEBA
ACS Paragon Plus Environment

Biomacromolecules
Page 24 of 36
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Experimental
Materials: Diphenolic Acid >98% was purchased from TCI America (D1274) as a white
crystalline solid and was used as received. Diglycidyl ether bisphenol A (DGEBA, a white
crystalline solid) epichlorohydrin, sodium hydroxide, methanol, ethanol, 1ꢀbutanol, and 1ꢀ
pentanol were purchased from Sigma Aldrich and used as received. All reagents and solvents
were purchased from Sigma Aldrich and used as received.
Synthesis of Diphenolate Esters and Epoxy Resins:
Synthesis of Methyl and ethyl diphenolate esters (DPꢀesters): Diphenolic acid (50 g), methanol
or ethanol (400 mL) and concentrated sulfuric acid (1 mL) were transferred to a 1 neck 1 L round
bottom flask fitted with a reflux condenser. The reaction mixture was heated to reflux with
magnetic stirring for 16 h after which the alcohol was removed by rotary evaporation. The
contents of the reaction flask was dissolved in 250 mL of ethyl acetate, transferred to a
separatory funnel and washed with 250 mL distilled deionized water 2 times, 250 mL 10%
sodium bicarbonate 2 times, and 250 mL distilled deionized water 2 times. The organic phase
was separated, dried over anhydrous magnesium sulfate, the magnesium sulfate was removed by
filtration, ethyl acetate was removed by rotary evaporation and remaining volatiles were
removed under vacuum (1 mmHg, 3 h). Both the methyl and ethyl esters were white solids
obtained in 85% yield. Methyl diphenolate: 1H NMR (500 MHz, DMSOꢀD6) ppm δ 1.45 (s, 3H),
2.00 (t, 2H), 2.20 (t, 2H), 3.50 (s, 3H), 6.60 (d, 4H), 6.90 (d, 4H), 9.10 (s, 2H); Melting point 135
ACS Paragon Plus Environment

Page 25 of 36
Biomacromolecules
1
2
3
4
°C as the peak in the DSC from first heating scan at 10°C/min heating rate; Characteristic FTIR
5
6
bands: 1663 cmꢀ1 C=O, 1505 cmꢀ1 CꢀC aromatic, 1217 cmꢀ1 CꢀH, and 1013 CꢀO cmꢀ1. Ethyl
7
8
9
1
Diphenolate: H NMR (500 MHz, DMSOꢀD6) ppm δ 1.11 (t, 3H), 1.45 (s, 3H), 2.00 (t, 2H),
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2.20 (t, 2H), 4.00 (q, 2H), 6.60 (d, 4H), 6.90 (d, 4H), 9.10 (s, 2H); Melting point: 130 °C as the
peak in the DSC from first heating scan at 10°C/min heating rate; Characteristic FTIR bands:
1665 cmꢀ1 C=O, 1508 cmꢀ1 CꢀC aromatic, 1218 cmꢀ1 CꢀH, and 1016 CꢀO cmꢀ1
Synthesis of nꢀbutyl and nꢀpentyl diphenolate esters: 50 g of diphenolic acid, 500 mL of 1ꢀ
butanol or 1ꢀpentanol, and concentrated sulfuric acid (1 mL) were transferred to a 1 neck 1 L
round bottom flask fitted with reflux condenser. The reaction mixture was heated to 90 °C with
magnetic stirring for 16 h. The reaction mixture was then cooled and washed with 250 mL
distilled deionized water 2 times, 250 mL of 10% sodium bicarbonate 2 times, and 250 distilled
deionized water 2 times. The mixture was then dried over anhydrous magnesium sulfate. The
magnesium sulfate was filtered off and the majority of the excess alcohol was removed by rotary
evaporation. Trace amounts of alcohol were removed by vacuum distillation (1 mmHg, at 120 °C
external pot temperature) until a light orange/yellow glassy solid was obtained in yields of 85ꢀ
1
97%. Butyl Diphenolate: H NMR (500 MHz, DMSOꢀD6): ppm δ 0.84 (t, 3H), 1.28 (m, 2H),
1.38 (m, 2H), 1.45 (s, 3H), 2.00 (t, 2H), 2.20 (t, 2H), 3.94 (q, 2H), 6.60 (d, 4H), 6.90 (d, 4H),
9.10 (s, 2H); Melting point: not observed by DSC or decomposition; Characteristic FTIR bands:
1669 cmꢀ1 C=O, 1509 cmꢀ1 CꢀC aromatic, 1217 cmꢀ1 CꢀH, and 1172 CꢀO cmꢀ1. Pentyl
Diphenolate: 1H NMR (500 MHz, DMSOꢀD6): ppm δ 0.84 (t, 3H), 1.28 (m, 4H), 1.38 (m , 2H),
1.45 (s, 3H), 2.00 (t, 2H), 2.20 (t, 2H), 3.94 (q, 2H), 6.60 (d, 4H), 6.90 (d, 4H), 9.10 (s, 2H);
Melting point: not observed by DSC or decomposition; Characteristic FTIR bands: 1704 cmꢀ1
C=O, 1509 cmꢀ1 CꢀC aromatic, 1220 cmꢀ1 CꢀH, and 1178 CꢀO cmꢀ1
ACS Paragon Plus Environment

Biomacromolecules
Page 26 of 36
1
2
3
4
5
6
7
8
9
Synthesis of diglycidyl ether nꢀalkyl diphenolates (DGEDPꢀesters): The reaction was
performed in a three neck 1 L round bottom flask equipped with an overhead mechanical stirring,
a water condenser and a pressure equalizing dropping funnel. The reaction flask was charged
with an nꢀalkyl phenolate (25.00 g), epichlorohydrin (15 equiv.) and 2ꢀpropanol (115 mL). Then,
the reaction flask was heated to reflux (~115 °C) with stirring and a 20% solution of aqueous
sodium hydroxide containing 2.1 equivalents (relative to nꢀalkyl phenolate) of sodium hydroxide
was added dropwise to the reaction. After addition of the sodium hydroxide solution was
complete the reaction mixture was maintained at reflux for 30 minutes. Subsequently, the
reaction mixture was cooled to room temperature, sodium chloride was removed by filtration, the
majority of 2ꢀpropanol was removed by rotary evaporation and sodium chloride was removed by
filtration. The reaction mixture was taken up in 50 mL of dichloromethane and then transferred
to a separatory funnel, neutralized by washing with two portions of 100 mL 3 M acetic acid,
washed with 3 portions of 100 mL distilled deionized water, and dried over anhydrous
magnesium sulfate. Remaining 2ꢀpropanol or epichlorohydrin was removed by rotary
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
°
evaporation and vacuum distillation (1 mmHg, 150 C external pot temperature). The products
obtained were viscous light yellow liquids in yields of 85ꢀ97%. Diglycidyl ether diphenolate
1
methyl ester (DGEDPꢀmethyl ester) H NMR (500MHz, CDCl3): ppm δ 1.57 (s, 3H), 2.10 (t,
2H), 2.40 (t, 2H), 2.75 (dd, J = 4.75 Hz, 2Hγb), 2.90 (observed triplet, J = 4.31 Hz, 2Hγa), 3.34
(m, 2Hβa), 3.62 (s, 3H), 3.94 (dd, J = 10.91 Hz, 2Hαb), 4.19 (dd, J = 11.75 Hz, 2Hαa), 6.83 (d,
2H), 7.10 (d, 2H); 13C NMR (500MHz, CDCl3): ppm δ 174, 156.4, 141, 128, 114, 68.5, 51.6, 50,
44.71, 44.52, 36.51, 30.08, 27.60; Characteristic FTIR bands: 3461 cmꢀ1 OꢀH, 3000 cmꢀ1 CꢀH
1669 cmꢀ1 C=O, 1509 cmꢀ1 CꢀC aromatic, 1217 cmꢀ1 CꢀH, and 1172 CꢀO cmꢀ1, 914 cmꢀ1 CꢀO
1
oxirane. Diglycidyl ether diphenolate ethyl ester (DGEDPꢀethyl ester) H NMR (500MHz,
ACS Paragon Plus Environment

Page 27 of 36
Biomacromolecules
1
2
3
4
5
6
CDCl3): ppm δ 1.22 (t, 3H), 1.57 (s, 3H), 2.09 (t, 2H), 2.40 (t, 2H), 2.75 (dd, J = 4.64, 2Hγb),
2.90 (observed triplet, J = 4.40 2Hγa), 3.34 (m, 2Hβa), 3.95 (dd, J = 11.10 Hz, 2Hαb), 4.08 (q, 2H),
7
8
9
13
4.19 (dd, J = 10.90 Hz, 2Hαa), 6.83 (d, 2H), 7.10 (d, 2H); C NMR (500MHz, CDCl3): ppm δ
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
174, 156.5, 142, 128.2, 114, 69, 60.3, 50.1, 44.7, 44.5, 36.5, 30.3, 27.7, 14.1; Characteristic FTIR
bands: 3400 cmꢀ1 OꢀH, 3000 cmꢀ1 CꢀH 1665 cmꢀ1 C=O, 1508 cmꢀ1 CꢀC aromatic, 1218 cmꢀ1 Cꢀ
H, and 1016 CꢀO cmꢀ1, 912 cmꢀ1 CꢀO oxirane; Diglycidyl ether diphenolate nꢀbutyl ester
1
(DGEDPꢀnꢀbutyl ester) H NMR (500MHz, CDCl3): ppm δ 0.92 (t, 3H), 1.35 (m, 2H), 1.57
(observed multiplet, 5H), 2.09 (t, 2H), 2.40, (t, 2H), 2.75 (dd, J = 5.03 Hz 2Hγb), 2.91 (dd, J =
5.17 Hz 2Hγa), 3.35 (m, 2Hβa), 3.94 (dd, J = 10.50 Hz 2Hαb), 4.02 (t, 2H), 4.19 (dd, J = 10.69 Hz,
13
2Hαa), 6.82 (d, 2H), 7.10 (d, 2H); C NMR (500MHz, CDCl3): ppm δ 174, 156.4, 141.3, 128.2,
114, 68.5, 64.3, 50.2, 44.5, 36.6, 30.6, 30.3, 27.7, 19.1, 13.7; Characteristic FTIR bands: 3420
cmꢀ1 OꢀH, 2997 cmꢀ1 CꢀH, 1665 cmꢀ1 C=O, 1508 cmꢀ1 CꢀC aromatic, 1218 cmꢀ1 CꢀH, and 1016
CꢀO cmꢀ1, 910 cmꢀ1 CꢀO oxirane; Diglycidyl ether diphenolate pentyl ester: 1H NMR (500MHz,
CDCl3) ppm δ 0.91 (t, 3H), 1.32, (m, 4H), 1.58 (observed multiplet, 5H), 2.08 (t, 2H), 2.40 (t,
2H), 2.75 (dd, J = 4.79 Hz, 2Hγb), 2.90 (observed triplet, J = 4.60 Hz, 2Hγa), 3.34 (m, 2Hβa), 3.94
(dd, J = 11.17 Hz, 2Hαb), 4.01 (t, 2H), 4.19 (dd, J = 11.28 Hz, 2Hαa ), 6.83 (d, 2H), 7.10 (d, 2H);
13C NMR (500MHz, CDCl3): ppm δ 174, 156, 142, 128, 114, 67.1, 65.2, 50.1, 44.7, 44.5, 36.5,
30.3, 28.2, 28.0, 27.7, 22.3, 13.9; Characteristic FTIR bands: 3400 cmꢀ1 OꢀH, 3000 cmꢀ1 CꢀH,
1665 cmꢀ1 C=O, 1508 cmꢀ1 CꢀC aromatic, 1218 cmꢀ1 CꢀH, and 1016 CꢀO cmꢀ1, 912 cmꢀ1 CꢀO
oxirane; Diglycidyl ether diphenolate glycidyl ester (DGEDPꢀglycidyl ester): (only observed in
DGEDPꢀmethyl ester and DGEDPꢀethyl ester at 6 and 3 mol%, respectively: 1H NMR (500MHz,
CDCl3) ppm δ 1.57 (s, 3H), 2.15, (t, 2H), 2.20 (t, 2H), 2.61 (dd, J = 4.86 Hz, 2Hγb’), 2.83
(observed triplet, J = 4.32 Hz, 2Hγa’), 3.16 (m, 2Hβa’), 3.85 (dd, J = 12.64 Hz 2Hαb’), 4.36 (dd, J =
ACS Paragon Plus Environment

Biomacromolecules
Page 28 of 36
1
2
12.10 Hz 2Hαa’), 6.60 (d, 4H), 6.90 (d, 4H); 13C NMR (500MHz, CDCl3): ppm δ 173.57, 156.4,
141, 128, 114, 68.7, 65., 51.6, 50.1, 49.2, 44.7, 44.6, 44.5, 36.6, 36.1, 30.1, 27.7;
3
4
5
6
7
8
9
Experimental and Theoretical Epoxide Equivalent Weights (EEW): Experimental values
of EEW for the epoxide resins synthesized herein were determined by ASTM D1652. Samples
were dissolved in dichloromethane and tetraethylammonium bromide and titrated in triplicate
using a standardized perchloric acid solution in acetic acid. Crystal violet was used as an
indicator and the end point was determined when the solution turned from blue to green for
longer than 30 seconds. Theoretical values of EEW were determined by a series of equations
published in Reference 33 and that are given in the supplementary information (SI) section.
Preparation of cured epoxy materials for testing: This method takes into account that the
epoxy resins prepared herein contain small amounts of higher molecular weight fractions.
Epoxide equivalent weight (EEW) was determined by ASTM D1652 – 11e1. The epoxy resins
were mixed with nearly stoichiometric amounts of isophorone diamine (IPDA) in a disposable
aluminum pan. The weight of the resin and crosslinker were calculated by using the epoxide
equivalent weight (EEW) and amine hydrogen equivalent weight (AHEW) were calculated by
as can be seen in equations 2 and 3 to obtain the parts by weight of diamine per hundred parts
resin (phr).34
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(2)
(3)
An example would be if phr = 20 then 20 grams of diamine would be needed for 100 grams of
epoxy resin. Digylcidyl ether of Bisphenol A (DGEBA) was melted and allowed to cool to room
temperature resulting in a super cooled liquid. Epoxy resins were vigorously mixed by hand until
homogenous with the appropriate amount of IPDA and degassed by centrifugation in a syringe
(3000 rpm, 2 minutes). The resins were then transferred via syringe into stainless steel molds
ACS Paragon Plus Environment

Page 29 of 36
Biomacromolecules
1
2
3
4
5
6
7
8
9
coated with PTFE mold release agent. The three different mold geometries used were: ASTM
type V tensile bars, rectangular dual cantilever bars (35 mm x 12 mm x 3 mm), and rheology
discs (25 mm x 2 mm). The method of resin curing was adapted from a literature procedure.40 In
summary, the resins were maintained under ambient conditions in the molds overnight and then
compression molded for 4 h at 80 °C and then 4 h at 160 °C with 2 metric tons of pressure.
Dynamic Mechanical Analysis: A Thermal Analysis Q800, operating in dual cantilever mode
with an amplitude of 10 ꢀm, was used to determine the storage modulus and glass transition
temperature (determined from the peak of the loss modulus) by scanning at 2 °C/min from 0 °C
to 200 °C. All samples were run in triplicate.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Rheology: The viscosity of the different resins was determined by rotational rheology at 25 °C
using a Thermal Analysis ARꢀG2 rheometer equipped with 25 mm stainless steel parallel plate
geometry. The sheer rate was swept from 1.25 ꢀ 1250 1/s under a steady state flow with an
equilibration time of 1 min and a gap of 1000 mm. Viscosity was reported as an average from the
Newtonian region.
To determine the average molecular weight between crosslinks, a Thermal Analysis ARꢀG2
rheometer was employed that is equipped with 25 mm stainless steel parallel plate geometry and
nitrogen flow. The testing was conducted under 15 N of normal force in order to ensure good
contact between the plates and the sample and the frequency was swept from 0.1 rad/s to 100
rad/s with the shear modulus being determined from the average storage modulus (G’) that was
frequency independent. Each sample was run at a temperature 50 °C higher than the tan δ peak.
Tensile Testing: Tensile testing was conducted following ASTM D638 on compression
molded samples using an Instron 4204 equipped with a 5 kN load cell with a crosshead speed of
ACS Paragon Plus Environment