Application of Semi-In Situ Liquid Exfoliation of Graphite To The Scalable Production of Graphene-Epoxy Nanocomposites
Application of Semi-In Situ Liquid Exfoliation of Graphite To The Scalable Production of Graphene-Epoxy Nanocomposites
Application of Semi-In Situ Liquid Exfoliation of Graphite To The Scalable Production of Graphene-Epoxy Nanocomposites
DOI: 10.1002/pc.25764
RESEARCH ARTICLE
1
School of Applied Sciences, University of
Huddersfield, Huddersfield, UK
Abstract
2
School of Chemistry and Biosciences, Recent advances in graphite wet exfoliation methods include new solvent sys-
Faculty of Life Sciences, University of tems, exfoliation-assisting agents such as pyrophosphate and use of graphite-
Bradford, Bradford, UK
intercalated compounds. These open new avenues for the scalable and versa-
Correspondence tile production of graphene/nanographite composites, with the prospect of
Daniel Belton, School of Applied Sciences, reducing barriers to the adoption of these materials for real-world applications.
University of Huddersfield, Huddersfield,
UK.
This article proposes a semi-in situ exfoliation method in which graphite is
Email: d.j.belton@hud.ac.uk exfoliated directly in the epoxy/cosolvent system via high-shear mixing. The
method is tested with three different solvent systems: epoxy/acetone; epoxy/N-
Methyl-2-pyrrolidone (NMP); and epoxy/NMP/water. Tailoring of mechanical
properties of the composites is achieved by adjusting the mixing and centrifu-
gation times and the temperature of the process. The incorporation of graphite
microparticles in the composites has a positive influence on strength and flexi-
bility at the expense of rigidity resulting in a 13.4% increase in ultimate tensile
strength (UTS) and 73% in elongation at break (εmax) at 0.04 wt% loading. The
proposed method also leads to nanocomposites with enhanced properties at
low-filler loadings. For example, the 0.01 wt% loading gave a 7.4% increase in
Young's modulus, 10.6% in UTS, and 9.1% in εmax. The scalability and versatil-
ity of this method make it a promising alternative for industrial production of
epoxy/graphene nanocomposites.
KEYWORDS
epoxy, exfoliation, graphene, graphite, nanocomposites, scalable
1 | INTRODUCTION
polymer matrices for creating graphene-polymer compos- size of its crystalline domains,[31] however, highly viscous
ites, due to their versatility and ease of processing.[5–8] solvents are believed to dampen the energy of collisions
The addition of graphene to epoxy can enhance its while promoting exfoliation.[21] As such, solvents play a
mechanical, thermal, and electrical properties, leading to significant role in HSM since they transfer the shear
new applications in industries such as aerospace, auto- forces to the graphite particles and stabilize the graphene
motive, medicine, and electronics.[9–15] and MLG platelets.[19,21,32] Polar organic solvents with
The manufacturing method of the nanocomposite surface tensions in the range 40 to 50 mJ m−2 and with
depends on the graphene material source and the Hansen solubility parameters (HSP) similar to graphene
composite's application. High-quality chemical vapor exhibit the best performance, however, exfoliation can
deposition (CVD) graphene can be grown directly or also be conducted in aqueous solutions of surfactants or
transferred to a substrate to prepare coated and lay- polymers and mixtures of isopropanol (IPA) and
ered composites.[16] Top-down graphene production via water.[19,20,29] Currently, the best HSM results are
the liquid-phase exfoliation of graphite yields a achieved with N-Methyl-2-pyrrolidone (NMP) and N-
graphene-solvent dispersion, from which graphene cyclohexyl-2-pyrrolidone,[19] however, reported yields are
must be recovered to produce the composite mate- low (<0.1%) and need to be increased. The issue of
rial.[3] Various methods for liquid-phase exfoliation of low yields can be addressed by the addition of another
graphite are available, including ball milling,[17,18] filler such as carbon nanotubes (CNTs)[22,33–35] or
high-shear mixing (HSM),[19,20] ultrasonication,[21,22] GMP/GNPs[36,37] to epoxy nanocomposites. The presence
three-roll milling,[2,23] chemical exfoliation,[24,25] and of the graphite species produced in HSM could have a
electrochemical exfoliation.[26,27] Top-down methods, in beneficial effect on the properties of graphene/epoxy
which graphene is produced from abundant graphite, composites and would further enhance the resource effi-
are lauded for their scalability and potential for indus- ciency of the process. The efficiency of the process can
trial applications, while advances in graphite exfolia- also be enhanced by using recently proposed solvents sys-
tion methods have made graphene production tems, such as NMP/water[38] and dihydrolevo-
accessible to commercial users and not-specialized lab- glucosenone.[21] While NMP demonstrates excellent
oratories. Apart from chemical exfoliation, these exfoliation efficacy, greener solvents with more desirable
methods usually yield a mixture of graphene flakes, toxicity and environmental profiles are also being actively
multilayer graphene (MLG) and graphite nano/micro- investigated.[21,39] Mixed solvent systems are particularly
platelets (GNPs and GMPs).[19,28] The graphene filler promising since the properties can be tailored by
size can be selected and controlled using centrifuga- adjusting the ratio of the solvents. First, adjusting the
tion, but the resulting supernatants require time- ratio to increase viscosity increases the exfoliation effi-
consuming analysis to determine the size and mor- ciency due to improved energy transfer.[19,20] Moreover,
phology of dispersed graphene flakes and graphite surface properties of a solvent blend expressed as surface
platelets.[19] Such analysis can be prohibitive in tensions and HSPs can be adjusted to facilitate the exfoli-
manufacturing settings due to limited access to special- ation process.[20] For example, Zhang et al reported that
ist equipment such as atomic force microscopy (AFM), the best performance of IPA/water systems is achieved
scanning electron microscopy (SEM), and transmission for around 30 vol% of IPA in which the HSP distance to
electron microscopy (TEM). graphene is the lowest.[32] The enhanced performance of
Paton et al have shown that HSM is a scalable and the aforementioned NMP/water system is due to the for-
versatile method of graphite exfoliation, whether using mation of NMP-water aggregates, which increase the vis-
industry-standard high-shear mixers (HSM)[19] or make- cosity of the solvent and the strength of the solvent-
shift equipment such as kitchen blenders.[29] In this pro- graphene interactions.[38] Viscosity and surface tension of
cess, a dispersion of graphite within a suitable solvent is a solvent can also be altered by adjusting the temperature
subjected to high-shear forces, which exfoliate graphite of the process, which further improves the versatility of
flakes into MLG and graphene. The quality and quantity solvent-assisted graphite exfoliation.[2] Epoxy resins seem
of produced graphene depend on the mixing speed and to be ideal candidates for graphite exfoliation, having
time, initial graphite concentration, graphite quality and high viscosities and surface tensions in the optimal range
on the solvent used.[19,30] For example, the amount of (40-50 mJ m−2).[40,41] Li et al successfully utilized neat
graphene platelets produced increases with increasing epoxy resin for graphene production via three roll mill-
HSM time; however, this is at the expense of lateral size, ing, which confirms the utility of epoxies in graphite
which decreases as the platelets break apart during colli- exfoliation[2,42]; however, mixed solvent systems of epoxy
sions.[19] This is unfavorable for the reinforcement of resins with a cosolvent have not previously been
composites as the graphene quality increases with the investigated.
15480569, 2020, 11, Downloaded from https://4spepublications.onlinelibrary.wiley.com/doi/10.1002/pc.25764 by CAPES, Wiley Online Library on [12/06/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
ORAWIEC ET AL. 4935
F I G U R E 1 A, Typical
approach to epoxy/graphene
composites production via liquid
exfoliation. B, Semi-in situ approach
to epoxy/graphene composites
production via liquid exfoliation.
GMPs, graphite microplatelets;
GNPs, graphite nanoplatelets; MLG,
multilayer graphene [Color figure
can be viewed at
wileyonlinelibrary.com]
Exfoliation
Names of samples Solvent Mixing time/min temperature/ C Centrifugation time/rpm
EpAct60, 90, 120 Epoxy/acetone 25/75 wt ratio 60, 90, and 120 25 30
EpNMPWc30, 45, 60 Epoxy/NMP/water 21/63/16 wt ratio 120 25 30, 45, and 60
EpNMPT15, 20, 25 Epoxy/NMP 25/75 wt ratio 120 15, 20, and 25 30
gradually increased to 9500 rpm and kept at that speed NMPW was removed in two steps, first on a rotary evapo-
for 120 minutes. The temperature of the dispersion was rator and then via vacuum distillation at 0.1 mbar and
maintained at 25 C using a water bath (15 C, 20 C, and 160 C. The removal of both solvents was confirmed by IR
25 C for epoxy/NMP systems). After the mixing, the dis- spectroscopy (Section S5). After the solvent removal, the
persion was transferred to 50 mL falcon tubes for centri- epoxy dispersions were manually mixed with the hard-
fugation, and the supernatant collected for further ener (100:30 weight ratio) and cast into silicone molds to
studies. Three solvents were studied: epoxy/acetone obtain dog-bone shapes. The curing process was under-
mixtures-25/75 wt ratio; epoxy/NMP mixtures-25/75 wt taken in a vacuum desiccator in which the shapes were
ratio; and epoxy/NMP/water mixtures (epoxy/NMPW)- degassed.
21/63/16 wt ratio. The epoxy/acetone and epoxy/NMP
dispersions were centrifuged at 2100g for 30 minutes and
the epoxy/NMPW at 2100g for 30, 45 and 60 minutes. 2.5 | Testing the properties of the
The detailed processing conditions for each sample group nanocomposites
are summarized in Table 1. All other processing condi-
tions, not included in Table 1, are as described above. An Instron universal testing machine was used to test the
mechanical properties of the nanocomoposites. The
applied measurement gap was 75 mm, and the elonga-
2.3 | Analysis of the dispersions tion speed was set to 0.3 mm min−1. The hardness of the
composites was tested using a Vickers micro indentation
The concentration of carbon material in the supernatant tester with the applied load of 300 gf for 15 seconds. The
was calculated from absorbance measurements at thermal stability of the composites was tested by ther-
660 nm. Separate calibration curves were obtained for mogravimetric analysis (TGA 1, Mettler Toledo). The
acetone and NMP dispersions. Drop casted dispersions measurements were conducted on powdered samples
were analysed by Quanta FEG 250 SEM to assess the size that were heated to 550 C at a rate of 10 K min−1.
and thickness of exfoliated particles. The VHX-2000 digi-
tal microscope equipped with VH-Z250R lens
(×250-×2500) was used complementary to SEM. Ren- 3 | RESULTS A ND DISCUSSION
ishaw inVia Raman Microscope, with a 520 nm laser,
was used to assess the quality of graphene. The size dis- 3.1 | Investigating the influence of
tributions of nano and micrographite platelets in the mixing time in epoxy/acetone systems
supernatants were measured using a Malvern Mastersizer
2000. Nicolet 380 (Thermo Electron Corporation) FT-IR Dispersions of graphite powder in epoxy/acetone mix-
spectrometer in an ATR mode was used to analyse the tures (25/75 wt ratio) were subjected to HSM for 60, 90,
chemical composition of the epoxy before and after the and 120 minutes at 25 C and centrifuged for 30 minutes
process. (samples EpAct60, 90 and 120). The increase in mixing
time from 60 to 120 minutes was applied in order to
obtain dispersions containing GMPs and GNPs of
2.4 | Preparation of graphene decreasing sizes. Based on the findings of Paton et al,[19]
nanocomposites it was also anticipated that the content of graphene and
MLG platelets in the dispersions increases with the
The nanocomposites were prepared from supernatants. increasing mixing time. Six samples per mixing time were
In the first stage, the solvent was removed from the dis- prepared, and their averaged graphite concentrations are
persion. Acetone from epoxy/acetone dispersions was presented in Table 2, along with the surface weighted
removed on a rotary evaporator. The NMP from epoxy/ mean sizes (D[3, 2]) of the dispersed graphitic species.
15480569, 2020, 11, Downloaded from https://4spepublications.onlinelibrary.wiley.com/doi/10.1002/pc.25764 by CAPES, Wiley Online Library on [12/06/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
ORAWIEC ET AL. 4937
TABLE 2 Properties of the graphitic species in the epoxy dispersions from the epoxy/acetone method
Samples Mixing time/min D (3, 2)/μm DR/μm C/mg g−1 (α660nm = 1099 ± 32 g mg−1 m−1) ID/IG
EpAct60 60 0.87 ± 0.17 1.1 ± 1.5 0.11 ± 0.04 0.32 ± 0.13
EpAct90 90 0.82 ± 0.12 1.0 ± 1.1 0.13 ± 0.09 0.32 ± 0.09
EpAct120 120 0.22 ± 0.07 0.80 ± 0.7 0.17 ± 0.06 0.35 ± 0.09
Graphite powder n/a 12 n/a n/a 0.20 ± 0.08
F I G U R E 2 A, Raman spectrum of particles from EPAct120. B, SEM image of GNP from EPAct120. GNP, graphite nanoplatelet; SEM,
scanning electron microscopy
The size of the natural graphite flakes significantly the Mastersizer's measurements of EpAct120 superna-
decreased as mixing time increased, from 0.87 μm after tants may not be accurate due to very low concentra-
60 minutes to 0.82 μm after 90 minutes, and then down tions and small platelets sizes. There are good
to 0.22 μm after 120 minutes of mixing (Table 2, correlations between Mastersizer and Raman results
Figure S3). Meanwhile, the concentration of the particles for the mean platelet's size for epoxy/NMPW method
in supernatants increased from 0.11 mg g−1 after presented in the next section, and the dispersions
60 minutes to 0.17 mg g−1 after 120 minutes (Table 2). obtained with that method are at least 10 times more
The presence of graphene species was confirmed with concentrated.
Raman spectroscopy. Figure 2A depicts a spectrum, in The mechanical properties of the nanocomposites
which the shape of the 2D peak corresponds to two layer prepared from EpAct samples are presented in Table 3
graphene[44,45]; spectra corresponding to other MLG spe- and the stress vs strain curves in Figure 3A. Figure 3B,C
cies are presented in Figure S6. A SEM picture of GNP show the SEM images of fractured surfaces of neat epoxy
from the EPAct120 sample is presented in Figure 2B. The and EpAct120 sample. The surface of the composite is
ID/IG values presented in Table 1 are the average ratios of rougher than that of the neat epoxy, crack propagation
intensities of the G and D peaks. The increase in D band lines are visible along with irregular holes and GMPs on
intensity is connected with the increased numbers of the surface. The irregularity of the propagation lines indi-
edge defects due to the fragmentation of the parti- cates that the crack deflection mechanism is responsible
cles.[19,31] Thus, the ID/IG ratio of EpAct120 samples, for the improved strength of the composite.[47] The pres-
which contain the smallest particles, is the highest. The ence of holes and GMPs on the surface indicates that the
ID/IG ratio can also be used to assess the mean platelet pullout of the filler happens during propagation,
size (DR) using Equation ((1),[19] where k has been esti- suggesting that the interface between GMPs and epoxy is
mated to be around 0.17.[46] weaker than interactions within the epoxy matrix
itself.[47] Some larger GMPs and holes are clustered, indi-
ID ID k cating that GMPs may form agglomerated regions. The
≈ + : ð1Þ
I Gplatelets I Ggraphite DR hardness of all composites was slightly enhanced in com-
parison with the neat epoxy. The thermal stability of the
As with D(3, 2), the DR values are comparable for composites was unaffected, as the decomposition temper-
EpAct60 and EpAct90 and lower for EpAct120. Likely, atures presented in Table S2 did not change significantly.
15480569, 2020, 11, Downloaded from https://4spepublications.onlinelibrary.wiley.com/doi/10.1002/pc.25764 by CAPES, Wiley Online Library on [12/06/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
4938 ORAWIEC ET AL.
T A B L E 3 Mechanical properties of
Sample Cfa/102% Eb/MPa UTSc/MPa εmaxd/% HV0.3e
the EpAc nanocomposites and neat
EpAct60 0.8 ± 0.4 2106 ± 55 45.3 ± 3.1 2.2 ± 0.5 19.2 ± 0.3 epoxy
EpAct90 1.0 ± 0.7 2166 ± 88 47.1 ± 3.0 2.4 ± 0.7 18.7 ± 0.3
EpAct120 1.3 ± 0.5 2380 ± 50 50.3 ± 4.1 2.4 ± 0.4 19.0 ± 0.3
Neat epoxy n/a 2217 ± 119 45.5 ± 3.5 2.2 ± 0.2 18.3 ± 0.2
F I G U R E 3 A, Stress vs strain curves of neat cured epoxy and a nanocomposite prepared from EpAc samples. B, Low-magnification
SEM images of fractured surfaces of neat epoxy. C, EpAct120. D, EpNMWc60. SEM, scanning electron microscopy [Color figure can be
viewed at wileyonlinelibrary.com]
The EpAct60 and EpAct90 samples exhibited an overall impede the curing process.[48] The negative effect of large
decrease in rigidity in comparison to the neat epoxy. This GMPs is at least partially diminished in the EpAct90 sam-
is probably due to the presence of large GMPs, which can ples, which have comparable properties to the neat epoxy
15480569, 2020, 11, Downloaded from https://4spepublications.onlinelibrary.wiley.com/doi/10.1002/pc.25764 by CAPES, Wiley Online Library on [12/06/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
ORAWIEC ET AL. 4939
TABLE 4 Comparison of the enhancements of mechanical properties of the epoxy nanocomposites with the literature
Abbreviations: GMPs, graphite microplatelets; GNPs, graphite nanoplatelets; MLG, multilayer graphene; NMP, N-Methyl-2-pyrrolidone;
UTS, ultimate tensile strength.
with slightly enhanced UTS and εmax. The highest H2O (21/63/16 wt ratio) were subjected to HSM for
enhancement in epoxy properties is observed for 120 minutes at 25 C and then centrifuged for 30, 45, and
EpAct120, which showed a 5% increase in Young's modu- 60 minutes (samples: EpNMPW30, 45, and 60). Addition
lus, 10.5% increase in UTS and 3.8% increase in hardness of water to NMP leads to the formation of (NMP2H2O)n
at 0.013 wt% of the filler. This nanocomposite contains aggregates formed by heteroassociation between water
the most graphene particles and the smallest GNPs and and NMP molecules, which have been shown to increase
GMPs (Table 2 and Figure S3). It can be concluded that the solvent's efficacy in graphite exfoliation.[38] The mea-
with the increase in mixing time, a higher quality filler sured surface tensions of the used epoxy and NMP are
material that enhances tensile properties more effec- close to each other and around 43 mJ m−2 (Table S1),
tively, is obtained. Although the literature reports more which is in desirable range for graphite exfoliation.[2] The
prominent enhancements of tensile properties in epoxy/NMPW mixtures have not only more suitable sur-
graphene/epoxy nanocomposites,[3,31,48] the usual filler face properties than epoxy/acetone ones but also higher
loadings are above 0.1 wt%.[3,48,49] Table 4 depicts viscosity.[38] Whereas both of these factors benefit the
improvements of the mechanical properties of epoxy exfoliation process,[19,57] the increased viscosity hinders
nanocomposites, at filler loadings even lower than those the separation process; thus, higher centrifugation times
usually reported in the literature. The improvement of were also investigated.
mechanical properties observed in EpAct120 sample It can be expected that all the dispersions presented
(7.4% improved rigidity and 10.6% improved strength) is in Table 5 contain similar amounts of graphene species
comparable to the one obtained for 0.1 wt% of graphene since they have been prepared using constant mixing
oxide[51] but at 10 times lower filler content. conditions (120 minutes). The increase in centrifugation
time results in removal of larger GMPs which is con-
firmed by the decreasing D (3, 2) and DR values, size dis-
3.2 | Investigating the influence of tribution graphs and optical microscopy (Figures S4 and
centrifugation time in epoxy/NMP/water S5). The D (3, 2) value decreases from 0.66 μm for
systems 30 minutes centrifugation to 0.46 μm for 60 minutes cen-
trifugation, which is accompanied by the decrease in the
Results presented in the previous chapter show that the concentration of the dispersions from 0.48 mg g−1 for
rigidity of epoxy nanocomposites improves with the 30 minutes centrifugation to 0.24 mg g−1 for 60 minutes
decreasing size of GMPs and with the increasing content centrifugation. Therefore, this study allows us to assess
of graphene species. In order to increase the efficiency of what kind of influence the presence of GMPs has on the
the exfoliation process, a more suitable solvent-NMP/ mechanical properties of the graphene nanocomposites.
H2O (0.2 wt fraction of H2O) in a mixture with epoxy is The concentrations of dispersions presented in Table 5
applied. Dispersions of graphite powder in epoxy/NMP/ are an order of magnitude higher than in epoxy/acetone
15480569, 2020, 11, Downloaded from https://4spepublications.onlinelibrary.wiley.com/doi/10.1002/pc.25764 by CAPES, Wiley Online Library on [12/06/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
4940 ORAWIEC ET AL.
TABLE 5 Properties of the graphitic species in the epoxy dispersions from the epoxy/NMPW method
C /mg g−1
Samples Centrifugation/min D (3, 2) /μm DR/μm (Raman) (α660nm = 740 ± 47 g mg−1 m−1) ID/IG
EpNMPWc30 30 0.66 ± 0.05 0.61 ± 0.16 0.48 ± 0.03 0.48 ± 0.08
EpNMPWc45 45 0.55 ± 0.04 0.56 ± 0.14 0.33 ± 0.02 0.50 ± 0.08
EpNMWc60 60 0.46 ± 0.04 0.51 ± 0.09 0.24 ± 0.02 0.53 ± 0.06
Graphite powder n/a 12 n/a 0.20 ± 0.08
F I G U R E 4 A, Raman spectrum of particles from EpNMPWc45. B, SEM image of MLG from EpNMPW dispersions. MLG, MLG,
multilayer graphene; NMP, N-Methyl-2-pyrrolidone; SEM, scanning electron microscopy
method even after 60 minutes centrifugation, which, (Figure 5, Table 6). The nanocomposites' hardness is
together with increased ID/IG ratios, confirms the slightly enhanced similarly to EpAc samples. The ther-
enhanced performance of epoxy/NMPW systems. More- mal stability of the EpNMPW composites has not been
over, the Mastersizer and Raman studies provide similar affected as well (Table S2). The EpNMPWc30 samples
information on the mean platelets' size, which proves the exhibit 13% improved tensile strength and 73% higher
applicability of these methods. More GMPs are likely pre- elongation at break but suffer a 20% decrease in Young's
sent in EpNMPW samples than in EPAc samples due to modulus. On the other hand, the EpNMWc60 samples,
the higher viscosity of epoxy/NMPW systems, and thus a which contain the lowest amount of GMPs, exhibit
lower value of absorption coefficient has been obtained slightly enhanced rigidity (3% increase in Young's modu-
(740 vs 1099 g mg−1 m−1). Nonetheless, Figure 4 and lus) and 4% diminished tensile strength. It seems that the
Figure S6 present Raman and SEM confirmations for the addition of GMPs to graphene nanocomposites increases
presence of graphene species. For example, the 2D peak their flexibility and strength at the cost of rigidity. In
in Figure 4A corresponds to the presence of MLG addition, composites packed with a higher number of
platelets,[44] and a thin, partially transparent MLG plate- larger platelets are more likely to have fewer cross-links
let can be seen in Figure 4B. Figure 3D shows that the due to obstructed curing process, which leads to the dete-
surface of a fracture of EpNMPWc60 is considerably rioration of the modulus.[48] Furthermore, it would seem
more rugged than of neat epoxy and EpAct120 sample. that the concentration of GMPs needs to be high enough
This indicates that the crack deflection is more promi- to improve the mechanical properties of the composite as
nent in EpNMPWc60 than in EpAct120.[46] In addition, it in EpNMPWc30; otherwise, the main effect is the inhibi-
seems that the GMPs are more homogeneously dispersed tion of the curing process, which diminishes the benefits
in EpNMPWc60 than in EpAct120 as the cracks evenly of adding graphene and MLG to the epoxy. This phenom-
cover the whole area of the sample. enon could explain the differences in properties between
With the increasing amount and size of GMPs, an EpNMPW and EpAc samples. The epoxy/acetone method
enhancement in UTS and εmax is noted; however, this is seems to provide better results than EpNMPW one with
accompanied by a significant decrease in rigidity the exception of hardness and flexibility of the
15480569, 2020, 11, Downloaded from https://4spepublications.onlinelibrary.wiley.com/doi/10.1002/pc.25764 by CAPES, Wiley Online Library on [12/06/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
ORAWIEC ET AL. 4941
Stress / MPa
30
20
EpNMPWc30
EpNMPWc45
10
EpNMPWc60
Neat epoxy
0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Strain / %
nanocomposites. Further enhancement of semi-in situ temperature). The epoxy/NMP systems were used with-
exfoliation can yield better results if nanocomposites con- out water since water losses its miscibility with
taining the smallest platelets (such as EpNMWc60 and NMP/epoxy systems at lower temperatures. Moreover,
EpAct120) with higher filler content are produced. This Table S1 shows that the surface tensions of both NMP
could be potentially achieved by tailoring the ratio of and epoxy are in a desirable region for graphite exfolia-
epoxy to the cosolvent and by changing the temperature tion, and thus the temperature change should have a
of the process as is presented in the next section. higher impact on the efficiency of the process.[2,21]
The concentrations of epoxy dispersions obtained
with this method are lower than for EpAc and EpNMPW
3.3 | Investigating the influence of the systems despite NMP being a better solvent for graphite
temperature in the epoxy/NMP systems exfoliation than acetone.[21,58] However, the DR sizes of
obtained, at every temperature, GMP and GNP platelets
Li et al have reported that the exfoliation temperate plays are smaller than in EpAct120 samples which indicates a
a significant role in in situ exfoliation of graphite in epoxy higher degree of exfoliation. Especially since higher vis-
resin via three-roll milling.[2] This phenomenon is tested cosity of NMP would impede the separation process in
in this section. Graphite dispersions in epoxy/NMP comparison with EpAc systems. Figure 6 visualizes an
(25/75 wt ratio) were subjected to HSM for 120 minutes increase in the composites' performance along with the
at 15 C, 20 C, and 25 C and centrifuged for 30 minutes decrease in the temperature of the HSM process, which
(samples: epNMPT15, 20, and 25). All of the tested sys- can be attributed to the increased exfoliation efficiency.
tems were prepared by HSM for 120 minutes, followed by Composites prepared at lower temperatures have higher
30 minutes centrifugation (after samples reached room filler concentration and contain smaller GMP and GNP
15480569, 2020, 11, Downloaded from https://4spepublications.onlinelibrary.wiley.com/doi/10.1002/pc.25764 by CAPES, Wiley Online Library on [12/06/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
4942 ORAWIEC ET AL.
F I G U R E 6 Properties of
EpNMPT composites prepared at
various temperatures. A, The filler
concentration. B, Mean platelet size
from Raman measurements. C,
Young's modulus. D, Ultimate
tensile strength. E, Elongation at
break. NMP, N-Methyl-
2-pyrrolidone; UTS, ultimate tensile
strength [Color figure can be viewed
at wileyonlinelibrary.com]
platelets. As in the previous cases, the decomposition nanocomposites were investigated: epoxy/acetone, epoxy/
temperature of the epoxy has not been affected by the NMP, and epoxy/NMP/water. The nanocomposites that
filler (Table S2). The mechanical properties of composites were produced exhibited enhanced rigidity, strength, and
prepared from dispersions exfoliated at 25 C are signifi- hardness. As such, this approach could serve as a substan-
cantly deteriorated in comparison with neat epoxy, which tial alternative for scalable industrial production of epoxy/
indicates low-filler quality and possibly curing inhabi- graphene nanocomposites, especially, given its versatility,
tancy. On the other hand, the EpNMPT15 nanocomposites scalability, and simplicity. The incorporation of both gra-
exhibit an overall enhancement of all measured mechani- phitic species and MLGs platelets was shown to compen-
cal properties (Table 4) with mean Young's Modulus of sate for the low-filler content of the composites partially.
2425 ± 73 MPa and UTS of 47.9 ± 2.3. The increase in Increase of mixing time leads to general improvement of
exfoliation's efficiency can be attributed to the change in filler's quality and thus improvement of the mechanical
surface tension of the solvent[19,20] as its temperature properties of the nanocomposites. We noted an increase in
decreases. An increase in solvent viscosity might also strength and flexibility of the material with the increasing
influence the exfoliation process, however, a change of content and size of the GMPs at the expense of decreased
just 5 to 10 is unlikely to have significant effect. Further- rigidity. As such, the properties of the nanocomposites can
more, the increase in efficiency is expected to diminish be tailored by adjusting the mixing and centrifugation
when the surface tension is no longer in the optimal range times. However, it seems that the content of GMP must
(40-50 mJ m−2).[2] meet a certain threshold to introduce the enhancement, as
with EpNMPWc30 samples; otherwise, the GMP platelets
may just impede the curing process and diminish the bene-
4 | C ON C L U S I ON S ficial effect of adding graphene and MLG sheets, as seen
with EpNMPc60. Results obtained for samples
Three cosolvent-epoxy systems for semi-in situ liquid exfo- EpNMPT15, EpNMPT20, and EpAct120 show that it is
liation of graphite and scalable production of epoxy possible to enhance both the rigidity and strength of the
15480569, 2020, 11, Downloaded from https://4spepublications.onlinelibrary.wiley.com/doi/10.1002/pc.25764 by CAPES, Wiley Online Library on [12/06/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
ORAWIEC ET AL. 4943
composites with the proposed method. Further develop- [15] A. Kumar, K. Sharma, A. R. Dixit, J. Mater. Sci. 2020, 55, 2682.
ment of this method could be achieved by exploring adjust- https://doi.org/10.1007/s10853-019-04196-y.
ments to the mixing speed, centrifugation regime, [16] V. P. Pham, H. S. Jang, D. Whang, J. Y. Choi, Chem. Soc. Rev.
2017, 46, 6276. https://doi.org/10.1039/c7cs00224f.
cosolvents, and temperature. The filler loading might be
[17] X. Fan, D. W. Chang, X. Chen, J. B. Baek, L. Dai, Curr. Opin.
further improved by increasing the cosolvent to epoxy Chem. Eng. 2016, 11, 52. https://doi.org/10.1016/j.coche.2016.
ratio. This might especially work for systems in which both 01.003.
the epoxy and the solvent have suitable surface properties [18] H. Wu, W. Zhao, H. Hu, G. Chen, J. Mater. Chem. 2011, 21,
for graphite exfoliation, such as the epoxy/NMPW system 8626. https://doi.org/10.1039/c1jm10819k.
presented in this study. Additional factors that could be [19] K. R. Paton, E. Varrla, C. Backes, R. J. Smith, U. Khan, A.
explored include the use of more industrially/environmen- O'Neill, C. Boland, M. Lotya, O. M. Istrate, P. King, T. Higgins,
tally compatible green solvents for exfoliation, and the use S. Barwich, P. May, P. Puczkarski, I. Ahmed, M. Moebius, H.
Pettersson, E. Long, J. Coelho, S. E. O'Brien, E. K. McGuire,
of graphite intercalates to facilitate exfoliation and prevent
B. M. Sanchez, G. S. Duesberg, N. McEvoy, T. J. Pennycook, C.
the inhibition of the curing process. Downing, A. Crossley, V. Nicolosi, J. N. Coleman, Nat. Mater.
2014, 13, 624. https://doi.org/10.1038/nmat3944.
FUNDING INFORMATION [20] L. Liu, Z. Shen, M. Yi, X. Zhang, S. Ma, RSC Adv. 2014, 4,
School of Applied Sciences, University of Huddersfield, 36464. https://doi.org/10.1039/C4RA05635C.
Huddersfield, UK. [21] H. J. Salavagione, J. Sherwood, M. De Bruyn, V. L. Budarin,
G. J. Ellis, J. H. Clark, P. S. Shuttleworth, Green Chem. 2017,
ORCID 19, 2550. https://doi.org/10.1039/c7gc00112f.
[22] L. Yue, G. Pircheraghi, S. A. Monemian, Carbon N Y. 2014, 78,
Marcin Orawiec https://orcid.org/0000-0001-6592-3613
268. https://doi.org/10.1016/j.carbon.2014.07.003.
[23] J. Chen, M. Duan, G. Chen, J. Mater. Chem. 2012, 22, 19625.
R EF E RE N C E S https://doi.org/10.1039/c2jm33740a.
[1] D. A. Dikin, S. Stankovich, E. J. Zimney, R. D. Piner, G. H. B. [24] D. W. Johnson, B. P. Dobson, K. S. Coleman, Curr. Opin. Col-
Dommett, G. Evmenenko, S. B. T. Nguyen, R. S. Ruoff, Nature loid Interface Sci. 2015, 20, 367. https://doi.org/10.1016/j.cocis.
2007, 448, 457. https://doi.org/10.1038/nature06016. 2015.11.004.
[2] Y. Li, H. Zhang, M. Crespo, H. Porwal, O. Picot, G. [25] N. Yousefi, X. Lin, Q. Zheng, X. Shen, J. R. Pothnis, J. Jia, E.
Santagiuliana, Z. Huang, E. Barbieri, N. M. Pugno, T. Peijs, E. Zussman, J. K. Kim, Carbon N Y. 2013, 59, 406. https://doi.
Bilotti, ACS Appl. Mater. Interfaces 2016, 8, 24112. https://doi. org/10.1016/j.carbon.2013.03.034.
org/10.1021/acsami.6b07492. [26] P. Yu, S. E. Lowe, G. P. Simon, Y. L. Zhong, Curr. Opin. Col-
[3] N. P. Singh, V. K. Gupta, A. P. Singh, Polymer 2019, 180, loid Interface Sci. 2015, 20, 329. https://doi.org/10.1016/j.cocis.
121724. https://doi.org/10.1016/j.polymer.2019.121724. 2015.10.007.
[4] X. Ji, Y. Xu, W. Zhang, L. Cui, J. Liu, Compos. Part A Appl. [27] T. C. Achee, W. Sun, J. T. Hope, S. G. Quitzau, C. B. Sweeney,
Sci. Manuf. 2016, 87, 29. https://doi.org/10.1016/j.compositesa. S. A. Shah, T. Habib, M. J. Green, Sci. Rep. 2018, 8, 1. https://
2016.04.011. doi.org/10.1038/s41598-018-32741-3.
[5] S. Zhao, H. Chang, S. Chen, J. Cui, Y. Yan, Eur. Polym. J. [28] Y. L. Zhong, Z. Tian, G. P. Simon, D. Li, Biochem. Pharmacol.
2016, 84, 300. https://doi.org/10.1016/j.eurpolymj.2016.09.036. 2015, 18, 73. https://doi.org/10.1016/j.mattod.2014.08.019.
[6] K.-S. Kim, I.-Y. Jeon, S.-N. Ahn, Y.-D. Kwon, J.-B. Baek, [29] E. Varrla, K. R. Paton, C. Backes, A. Harvey, R. J. Smith, J.
J. Mater. Chem. 2011, 21, 7337. https://doi.org/10.1039/ McCauley, J. N. Coleman, Nanoscale 2014, 6, 11810. https://
c0jm03504a. doi.org/10.1039/C4NR03560G.
[7] P. Mohan, Polym.Plast. Technol. Eng. 2013, 52, 107. https://doi. [30] H. Tao, Y. Zhang, Y. Gao, Z. Sun, C. Yan, J. Texter, Phys.
org/10.1080/03602559.2012.727057. Chem. Chem. Phys. 2017, 19, 921. https://doi.org/10.1039/
[8] F. Wang, L. T. Drzal, Y. Qin, Z. Huang, Polym. Compos. 2016, C6CP06813H.
37, 2897. https://doi.org/10.1002/pc.23487. [31] J. Kim, J. Kim, S. Song, S. Zhang, J. Cha, K. Kim, H. Yoon, Y.
[9] J. Li, H. Yan, D. Dang, W. Wei, L. Meng, J. Colloid Interface Jung, K. W. Paik, S. Jeon, Carbon N Y. 2017, 113, 379. https://
Sci. 2019, 535, 92. https://doi.org/10.1016/j.jcis.2018.09.094. doi.org/10.1016/j.carbon.2016.11.023.
[10] Y. G. Jeong, J. E. An, Polym. Int. 2014, 63, 1895. https://doi. [32] J. Zhang, L. Xu, B. Zhou, Y. Zhu, X. Jiang, J. Colloid Interface
org/10.1002/pi.4713. Sci. 2018, 513, 279. https://doi.org/10.1016/j.jcis.2017.11.038.
[11] Y. Wang, J. Yu, W. Dai, Y. Song, D. Wang, L. Zeng, N. Jiang, [33] V. C. Tung, L. M. Chen, M. J. Allen, J. K. Wassei, K. Nelson,
Polym. Compos. 2015, 36, 556. https://doi.org/10.1002/pc.22972. R. B. Kaner, Y. Yang, Nano Lett. 2009, 9, 1949. https://doi.org/
[12] J. K. Lee, S. Song, B. Kim, Polym. Compos. 2012, 33, 1263. 10.1021/nl9001525.
https://doi.org/10.1002/pc.22251. [34] M. R. Zakaria, M. H. Abdul Kudus, H. Md Akil, M. Z. M.
[13] C. Liu, C. Chen, H. Wang, M. Chen, D. Zhou, Z. Xu, W. Yu, Thirmizir, M. F. I. Abdul Malik, M. B. H. Othman, F. Ullah, F.
Polym. Compos. 2019, 40, E1294. https://doi.org/10.1002/pc. Javed, Polym. Compos. 2019, 40, E1840. https://doi.org/10.
24968. 1002/pc.25173.
[14] A. Kumar, K. Sharma, A. R. Dixit, Mol. Simul. 2020, 46, 136. [35] A. Kumar, K. Sharma, A. R. Dixit, J. Mater. Sci. 2019, 54, 5992.
https://doi.org/10.1080/08927022.2019.1680844. https://doi.org/10.1007/s10853-018-03244-3.
15480569, 2020, 11, Downloaded from https://4spepublications.onlinelibrary.wiley.com/doi/10.1002/pc.25764 by CAPES, Wiley Online Library on [12/06/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
4944 ORAWIEC ET AL.
[36] J. Li, P. S. Wong, J. K. Kim, Mater. Sci. Eng., A 2008, 483-484, [51] D. R. Bortz, E. G. Heras, I. Martin-Gullon, Macromolecules
660. https://doi.org/10.1016/j.msea.2006.08.145. 2012, 45, 238. https://doi.org/10.1021/ma201563k.
[37] S. Kumar, L. L. Sun, S. Caceres, B. Li, W. Wood, A. Perugini, [52] Y. J. Wan, L. C. Tang, L. X. Gong, D. Yan, Y. B. Li, L. B. Wu,
R. G. Maguire, W. H. Zhong, Nanotechnology 2010, 21, 105702. J. X. Jiang, G. Q. Lai, Carbon N Y. 2014, 69, 467. https://doi.
https://doi.org/10.1088/0957-4484/21/10/105702. org/10.1016/j.carbon.2013.12.050.
[38] K. Manna, H. N. Huang, W. T. Li, Y. H. Ho, W. H. Chiang, [53] I. Zaman, H. C. Kuan, J. Dai, N. Kawashima, A. Michelmore,
Chem. Mater. 2016, 28, 7586. https://doi.org/10.1021/acs. A. Sovi, S. Dong, L. Luong, J. Ma, Nanoscale 2012, 4, 4578.
chemmater.6b01203. https://doi.org/10.1039/c2nr30837a.
[39] J. Sherwood, M. De Bruyn, A. Constantinou, M. Laurianne, [54] Y. Zhang, Y. Wang, J. Yu, L. Chen, J. Zhu, Z. Hu, Polymer
C. R. McElroy, J. F. Thomas, T. Duncan, W. Raverty, J. A. 2014, 55, 4990. https://doi.org/10.1016/j.polymer.2014.07.045.
Hunt, H. J. Clark, Chem. Commun. 2014, 50, 9650. https://doi. [55] H. M. Chong, S. J. Hinder, A. C. Taylor, J. Mater. Sci. 2016, 51,
org/10.1039/c4cc04133j. 8764. https://doi.org/10.1007/s10853-016-0160-9.
[40] W. J. Feast, H. S. Munro, R. W. Richards, Polymer Surfaces [56] A. Yasmin, I. M. Daniel, Polymer 2004, 45, 8211. https://doi.
and Interfaces II, Wiley & Sons, New York 1993. org/10.1016/j.polymer.2004.09.054.
[41] S. A. Page, J. C. Berg, J. A. E. Månson, J. Adhes. Sci. Technol. [57] Y. Arao, F. Mori, M. Kubouchi, Carbon N Y. 2017, 118, 18.
2001, 15, 153. https://doi.org/10.1163/156856101743382. https://doi.org/10.1016/j.carbon.2017.03.002.
[42] Y. Li, H. Zhang, H. Porwal, Z. Huang, E. Bilotti, T. Peijs, Com- [58] J. N. Coleman, Acc. Chem. Res. 2013, 46, 14. https://doi.org/10.
pos. Part A Appl. Sci. Manuf. 2017, 95, 229. https://doi.org/10. 1021/ar300009f.
1016/j.compositesa.2017.01.007.
[43] S. Stankovich, D. A. Dikin, G. H. B. Dommett, K. M.
Kohlhaas, E. J. Zimney, E. A. Stach, R. D. Piner, S. B. T. SU PP O R TI N G I N F O RMA TI O N
Nguyen, R. S. Ruoff, Nature 2006, 442, 282. https://doi.org/10. Additional supporting information may be found online
1038/nature04969. in the Supporting Information section at the end of this
[44] A. C. Ferrari, J. C. Meyer, V. Scardaci, C. Casiraghi, M. article.
Lazzeri, F. Mauri, S. Piscanec, D. Jiang, K. S. Novoselov, S.
Roth, A. K. Geim, Phys. Rev. Lett. 2006, 97, 1. https://doi.org/
10.1103/PhysRevLett.97.187401.
[45] L. Gong, R. J. Young, I. A. Kinloch, I. Riaz, R. Jalil, K. S.
How to cite this article: Orawiec M, Belton D,
Novoselov, ACS Nano 2012, 6, 2086. https://doi.org/10.1021/ Telford R, Surtees A. Application of semi-in situ
nn203917d. liquid exfoliation of graphite to the scalable
[46] U. Khan, A. O'Neill, H. Porwal, P. May, K. Nawaz, J. N. production of graphene-epoxy nanocomposites.
Coleman, Carbon N Y. 2012, 50, 470. https://doi.org/10.1016/j. Polymer Composites. 2020;41:4933–4944. https://
carbon.2011.09.001. doi.org/10.1002/pc.25764
[47] A. Zotti, S. Zuppolini, M. Zarrelli, A. Borriello, Fracture
Toughening Mechanisms in Epoxy Adhesives. in Adhesives-
Applications and Properties, Vol. 1 (Ed: A. Rudawska), InTech,
Rijeka 2016, p. 257. https://doi.org/10.5772/65250.
[48] J. Wei, T. Vo, F. Inam, RSC Adv. 2015, 5, 7351. https://doi.org/
APPENDIX A
10.1039/c5ra13897c.
[49] S. Park, D. S. Kim, Polym. Eng. Sci. 2014, 54, 985. https://doi.
org/10.1002/pen.23368. This article is accompanied by the supporting informa-
[50] N. Domun, H. Hadavinia, T. Zhang, G. Liaghat, S. Vahid, C. tion that contains Figures from S1 to S9 and Tables S1
Spacie, K. R. Paton, T. Sainsbury, NANO 2017, 3, 85. https:// and S2.
doi.org/10.1080/20550324.2017.1365414.