Integration of Experimental Methods and Molecular Dynamics Simulations for a Comprehensive Understanding of Enhancement Mechanisms in Graphene Oxide (GO)/Rubber Composites

doi:10.21203/rs.3.rs-2888129/v1

Download PDF

Research Article

Integration of Experimental Methods and Molecular Dynamics Simulations for a Comprehensive Understanding of Enhancement Mechanisms in Graphene Oxide (GO)/Rubber Composites

https://doi.org/10.21203/rs.3.rs-2888129/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 24 Jun, 2023

Read the published version in Journal of Polymer Research →

You are reading this latest preprint version

Graphene oxide (GO) exhibits great application in rubber industry due to its unique two-dimensional nanosheet structure, significant specific surface area, good barrier property, and high reactivity. However, different rubbers, such as carboxylated nitrile butadiene rubber (XNBR), natural rubber (NR), and styrene butadiene rubber (SBR), have different interactions with GO, which has great influence on the reinforcement effect of GO to the rubber matrix. In this work, the enhancement mechanism of GO on NR, SBR, and XNBR was studied by combining experiments with molecular dynamics (MD) simulation. The results show that GO/XNBR nanocomposites had the smallest potential energy difference (ΔW_a), mean square displacement (MSD), and free volume fraction (FFV), resulting in excellent solvent resistance, and dynamic and mechanical properties. This study provides a new way to explore the macroscopic properties of rubber nanocomposites through molecular-level simulation.

graphene oxide

molecular dynamics simulation

rubber composites

interaction

Graphene oxide (GO) is a derivative of graphene, possessing a distinctive two-dimensional nanosheet structure that endows it with a notable specific surface area, and high reactivity^[1]. The abundance of active groups, such as hydroxyl, epoxy, carboxyl, and carbonyl groups, presenting on the surface of GO enables effective interaction with the rubber matrix^[2]. As a result, GO has emerged as a popular choice for augmenting the mechanical properties, thermal stability, and barrier properties of rubber. For example, we reported that the tensile strength and thermal conductivity of GO/carboxylated nitrile rubber (XNBR) nanocomposites were significantly increased by 370% and 40%, respectively, by adding 3 phr GO to XNBR^[3]. Mao et al.^[4] improved the gas barrier performance of the system by filling the butadiene-styrene-vinyl-pyridine rubber/styrene butadiene rubber (VPR/SBR) composite with GO, and the gas transmission rate decreased by 53.8% at 0.4% filler volume fraction.

Molecular dynamics (MD) simulation is now recognized as a promising technique for acquiring comprehensive insights into interfacial interactions between filler and matrix, which has the potential to facilitate the development of high-performance polymer nanocomposites at the atomic level through novel design strategies. For instance, Cui et al.^[5] used MD simulation to establish a sandwich molecular model to study the enhancement effect of GO on the mechanical properties and friction properties of nitrile rubber. Simulation results show that stronger interfacial interaction and better dispersion of GO in rubber matrix can be obtained by introducing functionalized GO. Verma et al.^[6] studied the fracture properties of graphene nanosheet (GNS) reinforced polymer composites by MD simulation. The results show that the addition of OH-functionalized GNS can obtain higher interfacial interaction energy and better fracture strength.

In polymer nanocomposites, the surface properties of both the nanofiller and polymer play a crucial role in determining the dispersion of the filler and interfacial interactions^[7]. In the case of GO, the abundant reactive groups on its surface exhibit unique interactions with diverse rubber substrates, thereby imparting a multitude of possibilities for enhancing various rubber properties. Natural rubber (NR) is a non-polar crystalline rubber with cis-1,4-polyisoprene units forming its main chain which has broad application in the rubber industry^[8]. XNBR is a polar rubber composed of units of acrylonitrile, 1,4-butadiene, and acrylic acid^[9], which has already been used as an alternative of traditional NBR in harsher application conditions. Styrene butadiene rubber (SBR) is a nonpolar elastomer with benzene rings forming its molecular chains that are used in automobile tires, sporting goods, and other fields^[10]. Therefore, exploring the interaction between GO and the above three typical rubbers is of great significance for optimizing the performance of GO/rubber nanocomposites.

Despite of the extensive use of MD simulations to investigate the characteristics of nanofiller reinforced polymer composites, to date, no comprehensive study has explored quantitatively the differences in the interaction between GO and various rubbers, along with their underlying molecular mechanisms.

In light of the aforementioned research gap, our study aimed to investigate the molecular-level characteristics of three distinct nanocomposites, namely GO/NR, GO/SBR, and GO/XNBR. By employing MD simulations, we computed various features such as binding energy (E_binding), solubility parameters (δ), fractional free volume (FFV), and mean square displacement (MSD). These findings, combined with experimental data such as surface energy, mechanical properties, dynamic mechanical properties, and solvent resistance, were utilized to study the mechanism underlying the improved performance of GO-reinforced XNBR, SBR, and NR. Our research provides crucial insights for developing nanocomposites with superior properties.

2.1 Materials

NR was provided by Lishenhang International Trade Co., Ltd. (Shanghai, China). SBR was purchased from Nanjing Yangzi Petrochemical Jinpu Rubber Co., Ltd (Nanjing, China). XNBR was purchased from Nandi Chemical (Zhenjiang, China) Co., Ltd. Graphite powder, sulfuric acid (H₂SO₄,98%), sodium nitrate (NaNO₃), potassium permanganate (KMnO₄), hydrogen peroxide (H₂O₂,30%), hydrochloric acid (HCl, 37%), sodium chloride (NaCl), analytical purity, chemical reagent Co., Ltd. The N-isopropyl-N’-phenyl-p-phenylenediamine (4010NA), stearic acid (SA), zinc oxide (ZnO), N-cyclohexyl-2-benzothiazole sulfinamide (CZ) and sulfur (S), industrial purity, Nanjing Jinsanli Rubber and Plastic Co., Ltd. (Nanjing, China).

2.2 Preparation of GO/rubber nanocomposites

GO was prepared by the Hummers method^[11]. The GO dispersion was added to rubber latex, stirred for 15 min, and then flocculated with NaCl solution. The resulting solids were repeatedly washed and filtered with deionized water and dried in a vacuum oven. At room temperature, rubber (100 phr), filler (GO, 3 phr), and compounding ingredients (4010NA, 2 phr; ZnO, 2 phr; SA, 2.4 phr; CZ, 2.2 phr; S, 1.5 phr) were added to the open mill to obtain rubber compounds. GO/NR, GO/SBR, and GO/XNBR nanocomposites were then prepared on a flat vulcanizing machine at 160°C and 15 MPa by compression molding. The properties of the vulcanized rubber samples were tested after being placed at room temperature for one day.

2.3 Characterization and tests

Elemental analysis of GO was performed by X-ray photoelectron spectroscope (XPS-PHI QUANTERA II, Ulvac-Phi, Japan). The chemical structures of GO/rubber nanocomposites were recorded by Fourier transform infrared spectra were recorded within wavenumbers ranging from 400 to 4000 cm^− 1(FTIR-8400 S-type spectrometer, Shimadzu, Japan). Raman spectra were recorded with an inVia-H31894 Raman spectrometer (Renishaw Corporation, Britain) by an argon-ion laser at an excitation wavelength of 514.5 nm and a resolution of 1 cm^− 1. The dispersion morphology of the nanocomposites was observed by scanning electron microscope (SEM-JSM-6380 LV, FEIGECH, Netherlands). The sample was brittlely broken in liquid nitrogen and then sprayed with gold for testing. The static mechanical performances of nanocomposites were characterized by an electronic universal testing machine (CMT 4254, Shenzhen New Sansi Material Testing Co., Ltd.) according to ASTM D-412. The tensile test spline was dumbbell-shaped, and the test rate was 500mm/min. The dynamic mechanical properties of nanocomposites were investigated using a dynamic mechanical thermal analyzer (DMA-Q800, American TA company). The sample specification was (20.0 × 4.0 × 0.5) mm, heating rate 5°C min^− 1, frequency 1 Hz, amplitude 10 µm, temperature − 80 ~ 50°C.

The swelling behavior of the sample was tested in toluene at room temperature according to ASTM D-471. The composite material was made into small discs, and the weighing mass was m₁. After immersion in toluene, the soaked sample at swollen equilibrium with stable weight was wiped and weighed to obtain (m₂). The solvent adsorption constant (Q_e) of the composite could be calculated through Eq. (1)^[12], and the test results took the average value of five tests.

$${Q_e}(mol\% )=100 \times \frac{{({m_2} - {m_1})/{M_s}}}{{{m_1}{m_m}/({m_n}+{m_m})}}$$

Where m_n is the mass of filler in the composite, m_m is the mass of rubber in the composite, and M_S is the molar mass of toluene (92.14 g mol^− 1).

The contact angle of rubber with water and diiodomethane was tested three times by contact angle meter, and the average value was taken. The surface energy is calculated by the Fowkes equation^[13].

$${\gamma _l}(1+\cos \theta )=2(\sqrt {\gamma _{s}^{d}\gamma _{l}^{d}} +\sqrt {\gamma _{s}^{p}\gamma _{l}^{p}} )$$

$${\gamma _l}=\gamma _{l}^{p}+\gamma _{l}^{d}$$

Among them, ${\gamma }_{l}^{p}$ and ${\gamma }_{l}^{d}$ are the polar and dispersive components of liquid surface energy, respectively, and ${\gamma }_{s}^{p}$ and ${\gamma }_{s}^{d}$ are the polar and dispersive components of solid surface energy, respectively. ${\gamma }_{l}$ is the total surface energy of the liquid, which is the sum of the dispersion component and the polar component. The ${\gamma }_{l}^{p}$ and ${\gamma }_{l}^{d}$ of deionized water in this study were 51.0 mJ m^− 2 and 21.8 mJ m^− 2, respectively. The ${\gamma }_{l}^{p}$ and ${\gamma }_{l}^{d}$ of diiodomethane are 2.3 mJ m^− 2 and 48.5 mJ m^− 2, respectively^[14].

The mixed gel was placed for one week, weighed 0.5g (ω₁), wrapped with filter paper (ω₂), soaked in 100 ml toluene at room temperature, changed the solvent every two days, soaked for 7 days, vacuum dried to constant weight. The calculation formula of bound rubber content (BdR%) is as follows^[15]:

$$BdR\left( \% \right)=\frac{{{w_2} - {w_1}{m_f}/\left( {{m_f}+{m_r}} \right)}}{{{w_1}{m_r}/\left( {{m_f}+{m_r}} \right)}} \times 100$$

where m_f is the mass of the filler in the composite, and m_r is the mass of the rubber in the composite.

By comparing the XPS result (Fig. S1), GO in which the percentage of carbon atoms and oxygen atoms is 66.84% and 33.16% was chosen as the model(C₃₇₅O₇₈(OH)₇₈(COOH)₂₀, (Fig. 1)). GO sheet model was then constructed and hydrogen atoms were added to terminate the unsaturated edge configuration. All the initial configurations used for MD simulations were built by Materials Studio 7.0 software.

A stable conformational model of GO was obtained through an energy minimization and annealing procedure. The smart method implemented in Materials Studio was used for energy minimization. During annealing, the system was gradually heated to a starting temperature of 300 K, followed by a mid-cycle temperature of 500 K, and then cooled back to 300 K. Subsequently, we constructed amorphous cells for GO/rubber composites and MD simulation using the Condensed-phase Optimized Molecular Potential for Atomistic Simulation Study (COMPASS) force field after energy minimization and annealing.

To further ensure that the system has reached equilibrium, 1000 ps MD simulations with NVT (constant number of particles, volume, and temperature) ensemble were followed by another 1000 ps NPT (constant number of particles, pressure, and temperature) ensemble. The construction of the composite models is illustrated in Fig. 1. In all the simulations, 3D periodic boundary conditions were employed. The box size of the MD simulation is 40 Å ⋅ 40 Å ⋅ 40 Å. The temperature was controlled by the Berendsen thermostat, and the pressure was controlled by the Andersen barostat^[16]. The van der Waals interactions were calculated by the Lennard-Jones function with a cutoff distance of 12.5 Å and the electrostatic interactions were calculated by the Ewald summation method. Newton’s equation of motion was solved by the Verlet velocity integration method with a time step of 1 fs.

4.1 Characterization of GO/rubber composites

FTIR of GO and GO/rubber nanocomposites are illustrated in Fig. 2(a). In the FTIR spectrum of GO, the broad peak at 3417 cm^− 1 is ascribed to the vibration of -OH, and sharp peaks at 1714, 1628, 1393, 1225, and 1055 cm^− 1 are attributed to the stretching vibration of C = O, stretching vibration of C = C, O-H deformation vibration from C-OH, C-O stretching vibration from the C-OH, and stretching vibration of C-O-C^[17], respectively. The FTIR spectrum of pure XNBR (Fig. S2) displays peaks at 3440 cm^− 1 (-OH), 2926, 2851 cm^− 1 (-CH₂-),1742(C = O) and 1702 cm^− 1 (C = O related to hydrogen bonding)^[18]. Upon the incorporation of GO into XNBR, the stretching vibration peak of the -OH group (attributed to XNBR) shifted from 3440 cm^− 1 to 3434 cm^− 1. This shift suggests the occurrence of hydrogen bonding between the carboxyl group of XNBR and the hydroxyl group on GO, which primarily resulted from hydrogen bonding-averaging electron cloud density. This interaction led to a decrease in the characteristic peak's wavenumber^[19]. The FTIR spectra of both GO/SBR and GO/NR nanocomposites exhibit a combination of the FTIR spectra of GO and the corresponding rubber matrix (SBR and NR, Fig. S2), suggesting the absence of hydrogen bonding between GO and SBR or NR.^{[20, 21]} Fig. 2(b) illustrates the number of total hydrogen bonds and the number of interfacial hydrogen bonds in the MD simulation of the GO/rubber nanocomposite system. It is evident that the GO/XNBR system exhibits the highest number of both total hydrogen bonds and interfacial hydrogen bonds, indicating the strongest interaction between the GO and XNBR interface. In contrast, there are no interfacial hydrogen bonds in GO/SBR and GO/NR systems, and the difference in total hydrogen bond number is negligible.

Raman spectroscopy was further employed to analyze the interfacial interactions between GO with different rubber matrix. As shown in Fig. 2(c), the GO delivers two Raman active modes including 1595 cm^− 1 (G band) and 1357 cm^− 1 (D band), the D band is activated by the breathing vibrations of carbon atoms in-plane terminations of defective graphite, and the G band assigned to the in-plane C-C stretching vibration of sp² carbon system^[22]. Compared to GO, the G band of GO/SBR nanocomposites exhibits a higher frequency shift (from 1595 to 1601 cm^− 1), while the D band remains almost unchanged. This suggests the occurrence of interfacial π-π interactions between GO and the benzene of SBR, leading to noticeable charge-transfer effects that are typically accompanied by frequency shifts of the G band^[23]. However, both the D band and G band of GO/XNBR and GO/NR nanocomposites are similar to those of GO, suggesting that there no π-π interactions between GO and XNBR or NR. The mechanism of the interactions between GO and the different rubber matrices (NR, SBR, and XNBR) are shown in Fig. 2(d). π-π interactions were observed between GO and SBR, and hydrogen bonding interactions were observed between GO and XNBR. In contrast, GO interacted with NR mainly through Van der Waals force interactions.

4.2 Thermodynamic compatibility and dispersion behavior

The morphology of the fracture surface of the composites was observed by SEM to characterize the dispersion of GO in the rubber matrix, as shown in Fig. 3. The fracture surface of GO/NR nanocomposite is rough (Fig. 3(a)) with the obvious GO extraction from NR matrix (white arrows), indicating that the poor dispersion of GO in NR matrix. In the case of GO/SBR nanocomposites (Fig. 3(b)), it was observed that the GO sheets underwent slight stacking and agglomeration within the rubber matrix. However, in the GO/XNBR system (Fig. 3(c)), no significant interface was observed between GO and XNBR, indicating that GO exhibited the best dispersion effect within the XNBR matrix.

The solubility parameter is an important physicochemical parameter used to predict the compatibility of components^[24]. The solubility parameter (δ) can be calculated from the square root of cohesive energy density (CED) through Eq. (5), where CED is the cohesive energy per unit volume. And δ (the square root of cohesive energy density) can also be calculated by two-component solubility parameters δ_vdW and δ_elec (the dispersive and electrostatic components of solubility parameters). Besides, R (compatibility parameter) is defined by the closeness between GO and rubber matrix^[9]. These parameters can be calculated as follows:

$$\delta =\sqrt {CED}$$

$${\delta ^2}=\delta _{{vdw}}^{2}+\delta _{{elec}}^{2}$$

$$R=\sqrt {{{({\delta _{vdw,m}} - {\delta _{vdw,m}})}^2}+{{({\delta _{ele,m}} - {\delta _{ele,f}})}^2}}$$

The CED and δ of GO, XNBR, SBR, and NR were calculated by MD simulation and the result was listed in Table 1. It can be seen that the δ of GO and XNBR are the closest, indicating that GO and XNBR have the best compatibility, which is consistent with the aforementioned analysis of SEM images. Furthermore, it is a widely accepted fact that R signifies the separation between the solubility parameter of the rubber and the solubility parameter of the filler in spherical coordinates^[25]. A lower R is indicative of better compatibility and dispersion between the rubber and filler. Notably, as shown in Fig. 4(a), the GO/XNBR nanocomposite exhibited the smallest R, indicating optimal compatibility between GO/XNBR.

Table 1

Simulated values of solubility parameters and *CED* of GO, XNBR, SBR, and NR
Samples	δ(MPa^1/2)	δ_vdw(MPa^1/2)	δ_elec(MPa^1/2)	CED(J cm^− 3)
GO	21.7	16.2	14.5	470.9
XNBR	19.2	18.9	3.45	368.6
SBR	17.1	16.8	3.02	292.4
NR	16.8	16.5	2.9	282.2

The varied dispersion of GO in different rubber matrices can be elucidated by interfacial thermodynamics^[7], wherein the filler dispersion in the matrix is influenced by the surface energies of both GO and rubber. The surface energies were evaluated by measuring the contact angles between deionized water and diiodomethane on the surfaces of GO, NR, SBR, and XNBR(Fig. 4b and 4c). According to Equations (2) and (3), the surface energies of GO, NR, SBR, and XNBR were calculated, and the results are listed in Table 2. It should be noted that GO and XNBR, which have carboxyl groups on the chain of XNBR and oxygen-containing groups on the surface of GO, have higher γ^d (dispersive components of surface energy), γ^p (polar components of surface energy), and γ (surface energy) than those of SBR and NR. Due to their similar polarities, XNBR and GO exhibit higher compatibility than NR and SBR. The intermolecular forces are directly correlated to the surface energy, so the highest γ of GO and XNBR means the strongest interaction between them.

Table 2

Surface energy of GO, XNBR, SBR, and NR
Samples	γ^d (mJ m^− 2)	γ^p (mJ m^− 2)	γ (mJ m^− 2)
GO	30.4	13.2	43.6
XNBR	29.9	11.0	40.9
SBR	27.9	9.6	37.5
NR	20.3	8.6	28.9

Adhesion energy (W_rf) describes the bonding strength between rubber and filler, which is the work required to separate two adjacent phases^[26]. The larger adhesion energy indicates that the rubber is more tightly bound to the filler and the interfacial interaction is stronger. Potential energy difference (ΔW_a) represents the potential energy difference between rubber and filler^[27]. A smaller ΔW_a indicates the better dispersion of filler in rubber. W_rf and ΔW_a can be calculated through Eq. (8) and Eq. (9)^[28]:

$${W_{rf}}=2\sqrt {\gamma _{r}^{d}\gamma _{f}^{d}} +2\sqrt {\gamma _{r}^{p}\gamma _{f}^{p}}$$

$$\vartriangle {W_a}=2{(\sqrt {\gamma _{f}^{d}} - \sqrt {\gamma _{r}^{d}} )^2}+2{(\sqrt {\gamma _{f}^{p}} - \sqrt {\gamma _{r}^{p}} )^2}$$

Here, subscripts r and f denote rubber and filler, respectively. Therefore, γ_r^p and γ_r^d are the polar and dispersive components of rubber surface energy. γ_f^p and γ_f^d are the polar and dispersive components of the surface energy of GO. From a thermodynamic standpoint, the aggregation of fillers is propelled by ΔW_a. When ΔW_a approaches 0, the surface characteristics of the rubber and filler become comparable, making it challenging for the filler to agglomerate. Conversely, as the surface properties between the filler and rubber diverge, the filler is more likely to aggregate^[29].

As shown in Fig. 5(a) and (b), the W_rf of GO/XNBR nanocomposites is maximum due to the strong interaction force caused by hydrogen bonding. The similarity in polarity between GO and XNBR, coupled with their closely aligned surface energies, results in a lower ΔW_a compared to that of GO/NR and GO/SBR nanocomposites. As previously noted, the interaction between GO and non-polar rubbers such as SBR and NR results in a large ΔW_a, hindering the dispersion of GO in these materials.

The interaction strength between the filler and the rubber matrix could be characterized by the binding energy (E_binding), which is defined as the negative value of the interaction energy (E_inter) of the two components. In MD simulation, E_binding was calculated through Eq. (10)^[5]:

$${E_{binding}}= - {E_{\operatorname{int} er}}= - \left( {{E_{totol}} - {E_{filler}} - {E_{matrix}}} \right)$$

Here, E_total was the total energy of the system, E_filler and E_matrix were the energies of separate filler and rubber matrix, respectively. According to the definition, a positive high E_binding suggests good compatibility between filler and rubber matrix, and two-phase separation morphology may not appear in the composites^[22]. E_binding of GO/XNBR, GO/SBR, and GO/NR nanocomposites were calculated using MD simulation and the results were shown in Table 3. The E_binding of GO/XNBR nanocomposites was observed to be higher than that of GO/SBR and GO/NR nanocomposites, indicating the superior compatibility and interaction between GO and the XNBR matrix. E_elec and E_vdw denote the electrostatic and van der Waals energy, respectively. Notably, the magnitude of electrostatic interactions is strongly dependent on the electronegativity of the functional groups. In contrast to the GO/SBR and GO/NR nanocomposites, the GO/XNBR nanocomposite exhibits the highest E_elec, attributable to the carboxyl group’s heightened electronegativity within XNBR.

Table 3

Energy of MD simulation of GO/NR, GO/SBR, and GO/XNBR nanocomposites
Samples	E_total (kcal mol^− 1)	E_filler (kcal mol^− 1)	E_matrix (kcal mol^− 1)	E_binding (kcal mol^− 1)	E_vdw (kcal mol^− 1)	E_elec (kcal mol^− 1)
GO/XNBR	12132	4356	6794	982	92	253
GO/NR	8390	4356	4469	435	135	163
GO/SBR	5672	4361	780	531	163	177

4.3 Mechanical properties and thermodynamic parameters

The stress-strain curve and tensile strength of NR, XNBR, and SBR before and after GO addition were shown in Fig. 6. (a-d). The tensile properties of GO/NR nanocomposites exhibited no significant improvement compared to those of pure NR. This lack of improvement may be attributed to the poor compatibility between NR and GO, resulting in extensive agglomeration of GO within the NR matrix and consequent disruption of the strain crystallization of NR. GO had a significant enhancing effect on SBR, the tensile strength, 100% tensile modulus (M100), and 300% tensile modulus (M300) increased by 222%, 40%, and 231%, respectively (Table. S1.), which is due to the π-π interaction between GO and SBR. Additionally, the tensile strength of GO/XNBR nanocomposites reached 26.72 MPa, which was 155% higher than that of pure XNBR. The superior mechanical properties of GO/XNBR nanocomposites may be attributed to the hydrogen bond interfacial interaction between XNBR and GO, facilitating stress transfer and thereby improving mechanical properties.

The significant CED of the polymer is representative of strong intermolecular contact between large molecule chains, which is defined as the energy required to transform the condensed phase polymer into a collection of chains^[30]. The CED of the pure rubber and GO/rubber nanocomposites were shown in Fig. 6(e). Compared with NR and SBR, the CED of XNBR is the highest, which is due to the molecular chains of XNBR containing a high proportion of polar groups, resulting in strong polar forces in addition to the dispersion force. Despite that CED of NR is only slightly less than that of SBR, self-reinforced nature of NR and ease of crystallization under stress contribute to its superior tensile strength compared to that of SBR. On the other hand, compared with GO/NR and GO/SBR nanocomposites, the increase of CED from XNBR to GO/XNBR nanocomposites is the highest, which verified the strongest interfacial interaction between GO and XNBR, accord with the results of tensile tests.

In order to further confirm the interaction between GO and the three different types of rubber, the bound rubber content (BdR%) of the GO/NR, GO/SBR, and GO/XNBR nanocomposites were determined. BdR% refers to a type of rubber that cannot be extracted with a suitable solvent, as it is the result of a strong interaction between rubber and filler^[31]. As shown in Fig. 6(f), the BdR% of GO/XNBR is the highest, while that of GO/NR is the lowest, indicating that GO and XNBR have the strongest interfacial connection, consistent with the mechanical properties tests.

4.4 Dynamic mechanical properties and chain motion

The storage modulus (E′) of GO/XNBR, GO/SBR, and GO/NR nanocomposites and corresponding pure rubber were displayed in Fig.S3(a-c). In the glassy region (below T_g), the storage modulus of GO/NR, GO/SBR, and GO/XNBR nanocomposites increased in the temperature range from − 70 to 40°C after adding GO. The increased storage modulus of GO incorporated nanocomposites are caused by the reinforcing effect of nanofillers which restrict the mobility of polymer chain, and provides better load bearing ability to composite materials.

The loss factor (tan δ) as a function of temperature was studied by DMA. As shown in Fig. 7 (a-c), the addition of GO increased the glass transition temperature (T_g) of XNBR by 2.2°C (-0.9 ~ 1.3°C), the T_g of SBR by 1.8°C (-39.3 ~ -37.5°C), and the T_g of NR by 0.9°C (-40.7 ~ -39.8°C), respectively. Owing to the stronger interaction between GO and XNBR, the extent of T_g enhancement in XNBR after the addition of GO is greater than those observed in SBR and NR. In addition, the high degree of GO dispersion in the XNBR matrix enlarges the region in which rubber molecules come into contact with nanofillers, thereby limiting the mobility of most rubber molecules^[32]. However, the addition of GO does not significantly impact the T_g of NR. It is due to the significant agglomeration (as observed in SEM) of GO in NR matrix, which can not restrict the mobility of NR rubber molecules.

For an equilibrium system, the movement ability of particles in equilibrium is often characterized by mean square displacement (MSD). The MSD of particles was calculated through Eq. (11)^[33]:

$$MSD=\frac{1}{N}\sum\limits_{{i=1}}^{N} {\left\langle {{{\left| {{r_i}(t) - {r_i}(0)} \right|}^2}} \right\rangle }$$

where N is the total number of particles in the system, r_i (0) and r_i (t) are the position vectors of particle t at the initial time and t time respectively, and < > represents the ensemble average.

The MSD of GO/XNBR, GO/SBR, and GO/NR are shown in Fig. 7 (d-f). After adding GO, the MSD of GO/XNBR, GO/SBR, and GO/NR nanocomposites at 500 ps decreased by 28.25%, 22.85%, and 12.26%, respectively, compared with that of the corresponding pure rubber. A lower MSD indicates a stronger interaction between the rubber and filler, which limits the molecular chain's ability to move. In GO/XNBR nanocomposites, the hydrogen bonding between GO and XNBR resulted in a significant reduction in MSD, indicating a denser structure that restricts the movement of rubber chains. Similarly, the MSD of GO/SBR nanocomposites was much lower than that of pure rubber owing to the π-π interaction between GO and SBR. However, the MSD of GO/NR nanocomposites did not decrease significantly due to the weak van der Waals connection and partial physical entanglement between GO and NR, which limited the molecular chain's restriction.

4.5 Solvent resistance and molecular free volume

Figure 8(a) presents the Q_e of pure rubber and nanocomposites. The molecular chain of XNBR has minimal free volume due to the influence of internal hydrogen bonds^[34], leading to the lowest degree of swelling. Incorporating GO into XNBR created strong chemical interfaces between GO and XNBR and established a filling network due to the well-dispersed GO in XNBR, which contributed to high solvent resistance, significantly reducing the free volumes between XNBR molecules and increasing the routes and diffusion times of solvent molecules in XNBR matrices.

According to the free volume theory proposed by Fowkes and Flory, fractional free volume (FFV) is the ratio of free volume (FV(r)) to total volume (V). It can be used to characterize the stacking capacity of molecular chains and calculated through Eq. (12)^[35]:

$$FFV(r)=\frac{{FV(r)}}{V} \times 100\%$$

As shown in Fig. 8(b), due to the hydrogen bond between GO and XNBR rubber chains and the π-π interaction between GO and SBR, the XNBR chains and SBR chains in GO/XNBR and GO/SBR nanocomposites are stacked more compactly, resulting in FFV of the two nanocomposites decreased by 57.37% and 33.3%, respectively. The migration time and path of solvent molecules in the system are prolonged, and the solvent resistance is improved.

The molecular-level insights and macroscopic features of three systems, namely GO/NR, GO/SBR, and GO/XNBR were investigated by combining MD simulations and experiments. The simulation outcomes revealed that GO exhibited the best dispersibility in XNBR, as confirmed by the highest E_binding and R for GO and XNBR. The MSD and FFV of both XNBR and SBR showed a significant decrease with the incorporation of GO due to the hydrogen bonding and π-π interactions between GO and XNBR and SBR, respectively. The experimental results demonstrated that GO/XNBR and GO/SBR nanocomposites outperformed pure rubber with regard to their mechanical characteristics, dynamic mechanical properties, and solvent resistance. However, the performance of NR was not significantly improved by the addition of GO, owing to the poor dispersion and interaction of GO in NR. Our work reveals the relationship between the interaction between GO and rubber as well as the reinforcing effect from the molecular level, and provides a new way to design high performance rubber nanocomposites.

Acknowledgements

We acknowledge financial support from Xi’an Jiaotong-Liverpool University Research Development Fund (PGRS2112001), Shanghai Aerospace Science and Technology Innovation Fund (SAST2022-097).

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

X. Duan, S. Cheng, R. Tao, Z. Zhang, G. Zhao (2022) Synergistically enhanced thermal control ability and mechanical properties of natural rubber for tires through a graphene/silica with a dot-face structure. Adv Compos Hybrid Mater 5:1145.
H. Bian, J. Xue, G. Hao, Y. Hao, M. Xie, C. Wang, Z. Wang, L. Zhu, Y. Xiao (2022) High thermal conductivity graphene oxide/carbon nanotubes/butyl rubber composites prepared by a dry ice expansion pre-dispersion flocculation method. J Appl Polym Sci 139:51897.
J. Wang, H. Jia, Y. Tang, D. Ji, Y. Sun, X. Gong, L. Ding (2013) Enhancements of the mechanical properties and thermal conductivity of carboxylated acrylonitrile butadiene rubber with the addition of graphene oxide. J Mater Sci 48:1571.
Y. Mao, S. Wen, Y. Chen, F. Zhang, P. Panine, T. W. Chan, L. Zhang, Y. Liang, L. Liu (2013) High Performance Graphene Oxide Based Rubber Composites. Sci Rep 3:2508.
J. Cui, J. Zhao, S. Wang, Y. Li (2021) A comparative study on enhancement of mechanical and tribological properties of nitrile rubber composites reinforced by different functionalized graphene sheets: Molecular dynamics simulations. Polym Compos 42:205.
A. Verma, A. Parashar (2018) Molecular dynamics based simulations to study the fracture strength of monolayer graphene oxide. Nanotechnology 29:115706.
H. Yang, F. Cai, Y. Luo, X. Ye, C. Zhang, S. Wu (2020) The interphase and thermal conductivity of graphene oxide/butadiene-styrene-vinyl pyridine rubber composites: A combined molecular simulation and experimental study. Compos Sci Technol 188:107971.
N. Candau, E. Vives, A. Ines Fernandez, M. Lluisa Maspoch (2021) Elastocaloric effect in vulcanized natural rubber and natural/wastes rubber blends. Polymer 236:124309.
F. Cai, Y. Luo, W. Yang, X. Ye, H. Zhang, J. Zhu, S. Wu (2021) Study on the thermal and dielectric properties of covalently modified GO/XNBR composites. Mater Des 198:109335.
X. Zhang, X. Xue, Q. Yin, H. Jia, J. Wang, Q. Ji, Z. Xu (2017) Enhanced compatibility and mechanical properties of carboxylated acrylonitrile butadiene rubber/styrene butadiene rubber by using graphene oxide as reinforcing filler. Composites Part B 111:243.
W. S. Hummers, R. E. Offeman (1958) Preparation of Graphitic Oxide. J Am Chem Soc 208:1334.
Z. Zhang, P. Chen, W. Nie, Y. Xu, Y. Zhou (2020) Enhanced mechanical, thermal and solvent resistance of silicone rubber reinforced by organosilica nanoparticles modified graphene oxide. Polymer 203:122772.
C. Yin, X. Zhao, J. Zhu, H. Hu, M. Song, S. Wu (2019) Experimental and molecular dynamics simulation study on the damping mechanism of C5 petroleum resin/chlorinated butyl rubber composites. J Mater Sci 54:3960.
J. Hou, S. Huang, N. Zhang, D. Liu, Z. Zhang, H. Yang, L. Zong, Y. Duan, J. Zhang (2021) Reinforcing natural rubber by amphiphilic graphene oxide for high-performance catheters. Polymer 232:124142.
B. Mensah, J. K. Efavi, D. S. Konadu, G. P. Manu (2022) Graphene-maleic anhydride-grafted-carboxylated acrylonitrile butadiene-rubber nanocomposites. Heliyon 8:11974.
H. Zhang, F. Cai, Y. Luo, X. Yu, S. Wu (2022) Grafting silica onto reduced graphene oxide via hydrosilylation for comprehensive rubber applications: Molecular simulation and experimental study. Polym Compos 43:5332.
S. Wan, X. Lu, H. Zhao, S. Chen, S. Cai, X. He, R. Zhang (2022) Effect of Graphene Oxide Modified with Organic Amine on the Aging Resistance, Rolling Loss and Wet-Skid Resistance of Solution Polymerized Styrene-Butadiene Rubber. Materials 13:1025.
S. M. Krzeminska, A. A. Smejda-Krzewicka, A. Leniart, L. Lipinska, M. Woluntarski (2020) Effects of curing agents and modified graphene oxide on the properties of XNBR composites. Polym Test 83:106368
H. Kang, K. Zuo, Z. Wang, L. Zhang, L. Liu, B. Guo (2014) Using a green method to develop graphene oxide/elastomers nanocomposites with combination of high barrier and mechanical performance. Compos Sci Technol 92:1.
S. Inphonlek, N. Bureewong, K. Jarukumjorn, P. Chumsamrong, C. Ruksakulpiwat, Y. Ruksakulpiwat (2022) Preparation of Poly(acrylic acid-co-acrylamide)-Grafted Deproteinized Natural Rubber and Its Effect on the Properties of Natural Rubber/Silica Composites. Polymers 14:4602.
A. Khan, L. K. Kian, M. Jawaid, A. A. P. Khan, M. M. Alotaibi, A. M. Asiri, H. M. Marwani (2022) Preparation of Styrene-Butadiene Rubber (SBR) Composite Incorporated with Collagen-Functionalized Graphene Oxide for Green Tire Application. Gels 8:161.
F. Cai, G. You, K. Luo, H. Zhang, X. Zhao, S. Wu (2020) Click chemistry modified graphene oxide/styrene-butadiene rubber composites and molecular simulation study. Compos Sci Technol 190:108061.
D. R. Dreyer, S. Park, C. W. Bielawski, R. S. Ruoff (2010) The chemistry of graphene oxide. Chem Soc Rev 39:228.
Q. Wei, Y. Zhang, Y. Wang, W. Chai, M. Yang (2016) Measurement and modeling of the effect of composition ratios on the properties of poly(vinyl alcohol)/poly(vinyl pyrrolidone) membranes. Mater Des 103:249.
Y. Luo, X. Chen, H. Liu, H. Zhang, M. Song, J. Liu, Z. Luo (2021) Precisely tailoring the thermodynamic compatibility between single-walled carbon nanotubes and styrene butadiene rubber via fully atomistic molecular dynamics simulation and theoretical approach. Comput Mater Sci 186:109995.
Y. Han, K. C. Lai, A. Lii-Rosales, M. C. Tringides, J. W. Evans, P. A. Thiela (2019) Surface energies, adhesion energies, and exfoliation energies relevant to copper-graphene and copper-graphite systems. Surf Sci 685:48.
J. Zhang, J. Ma, L. Zhang, C. Zong, A. Xu, Y. Zhang, B. Geng, S. Zhang (2020) Enhanced breakdown strength and suppressed dielectric loss of polymer nanocomposites with BaTiO₃ fillers modified by fluoropolymer. Rsc Adv 10:7065.
Y. Luo, Y. Wu, K. Luo, F. Cai, T. Zhai, S. Wu (2018) Structures and properties of alkanethiol-modified graphene oxide/solution-polymerized styrene butadiene rubber composites: Click chemistry and molecular dynamics simulation. Compos Sci Technol 161:32.
L. Qu, L. Wang, X. Xie, G. Yu, S. Bu (2014) Contribution of silica-rubber interactions on the viscoelastic behaviors of modified solution polymerized styrene butadiene rubbers (M-S-SBRs) filled with silica. Rsc Adv 4:64354.
C. Li, A. Strachan (2018) Cohesive energy density and solubility parameter evolution during the curing of thermoset. Polymer 135:162.
L. Huang, F. Yu, Y. Liu, A. Lu, Z. Song, W. Liu, Y. Xiong, H. He, S. Li, X. Zhao, S. Cui, C. Zhu (2023) Understanding the Reinforcement Effect of Fumed Silica on Silicone Rubber: Bound Rubber and Its Entanglement Network. Macromolecules 56:1
X. Dong, Q. Liu, L. Cui, Y. Yu, M. Zhang (2014) Molecular simulation and experimental study on propylene dehumidification through a PVA-PAA blend membrane. J Mater Chem A 2:16687.
J. Tan, C. Chen, Y. Liu, J. Wu, D. Wu, X. Zhang, X. He, Z. She, R. He, H. Zhang (2020) Molecular simulations of gas transport in hydrogenated nitrile butadiene rubber and ethylene-propylene-diene rubber. Rsc Adv 10:12475.
F. Pan, F. Peng, Z. Jiang (2007) Diffusion behavior of benzene/cyclohexane molecules in poly(vinyl alcohol)-graphite hybrid membranes by molecular dynamics simulation. Chem Eng Sci 62:703.
X. Liu, X. Zhou, F. Kuang, H. Zuo, J. Huang (2021) Mechanical and Tribological Properties of Nitrile Rubber Reinforced by Nano-SiO₂: Molecular Dynamics Simulation. Tribol Lett 69:54.

Download PDF

Journal Publication

published 24 Jun, 2023

Read the published version in Journal of Polymer Research →

Reviewers agreed at journal
15 May, 2023
Reviewers invited by journal
15 May, 2023
Editor invited by journal
11 May, 2023
Editor assigned by journal
03 May, 2023
First submitted to journal
02 May, 2023

You are reading this latest preprint version

Integration of Experimental Methods and Molecular Dynamics Simulations for a Comprehensive Understanding of Enhancement Mechanisms in Graphene Oxide (GO)/Rubber Composites

Status:

Journal Publication

Version 1

Abstract

Figures

1 Introduction

2 Experimental Details

2.1 Materials

2.2 Preparation of GO/rubber nanocomposites

2.3 Characterization and tests

3 Molecular Simulation Details

4 Results and Discussion

4.1 Characterization of GO/rubber composites

4.2 Thermodynamic compatibility and dispersion behavior

4.3 Mechanical properties and thermodynamic parameters

4.4 Dynamic mechanical properties and chain motion

4.5 Solvent resistance and molecular free volume

5 Conclusion

Declarations

References

Status:

Journal Publication

Version 1