Highly Deformable Porous Electromagnetic Wave Absorber Based on Ethylene–Propylene–Diene Monomer/Multiwall Carbon Nanotube Nanocomposites

The need for electromagnetic interference (EMI) shields has risen over the years as the result of our digitally and highly connected lifestyle. This work reports on the development of one such shield based on vulcanized rubber foams. Nanocomposites of ethylene–propylene–diene monomer (EPDM) rubber and multiwall carbon nanotubes (MWCNTs) were prepared via hot compression molding using a chemical blowing agent as foaming agent. MWCNTs accelerated the cure and led to high shear-thinning behavior, indicative of the formation of a 3D interconnected physical network. Foamed nanocomposites exhibited lower electrical percolation threshold than their solid counterparts. Above percolation, foamed nanocomposites displayed EMI absorption values of 28–45 dB in the frequency range of the X-band. The total EMI shielding efficiency of the foams was insignificantly affected by repeated bending with high recovery behavior. Our results highlight the potential of cross-linked EPDM/MWCNT foams as a lightweight EM wave absorber with high flexibility and deformability.


CURING BEHAVIOR AND CROSSLINK DENSITY (CLD)
Curing characteristics of the prepared compounds were evaluated at 160 °C according to the ASTM 5289 using a rubber curing rheometer (Rubber Process Analyzer, Alpha Technologies, Akron, USA). Scorch time (t2) and optimum cure time (t90) were measured based on the obtained rheographs.
Degree of crosslinking at t90 for each individual compound was evaluated by means of equilibrium swelling method using Flory-Rehner equation (Eq. S1) [1]. For this purpose, five circular cured test pieces with a diameter of 1 cm were cut and immersed in toluene (99.5%) at room temperature for 72 h [2].
where ν is the CLD (mol/ cm 3 ), VR is the volume fraction of EPDM rubber after immersion in toluene, V is the molecular volume of toluene (106.8 cm 3 /mol), and μ is the interaction parameter between EPDM and toluene (0.5). Figure S1 shows the curing curves of all fabricated EPDM/MWCNTs nanocomposite compounds containing various levels of MWCNTs, with and without foaming agent obtained at 160°C. The values of minimum torque (ML), maximum torque (MH), scorch time (t2), and t90 (optimum cure time, the time for the torque that is equal to the 90% of ΔM=MH-ML), are also given in Table S1. The results reveal that ΔM values increase with increasing MWCNTs content, which is indicative of the reinforcing effect of the MWCNTs [3][4][5]. Additionally, the scorch time reduces, from about 5 min to around 1 min, indicating an acceleration of the start of the vulcanization reaction. This result could be attributed to the higher thermal conductivity of both solid and foamed EPDM/MWCNTs nanocomposites (thermal conductivities are reported below). In hot curing of rubber compounds, the speed of temperature rise in various points of the compound is governed by thermal diffusivity factor ( p .C κ α = ρ ; where κ, ρ, and Cp denote thermal conductivity, mass density, and specific heat capacity, respectively). Hence, as κ increases, the time required for the compound to reach the mold temperature becomes shorter [3]. Finally, foamed samples also exhibit lower values of scorch time and t90 compared to the solid samples with the same MWCNTs contents, indicating an acceleration effect of the blowing agent. The decomposition of sulphohydrazide and its byproducts activate the sulfur vulcanization of EPDM. Similar observation has also been reported by other researchers [3][4][5][6].  Torque (dNm)

Time (min)
The CLD value would influence the mechanical properties such as deformation and elasticity as well as ease of segmental polarization when subjected to high-frequency electrical waves. Therefore, CLD of both solid and foamed samples at their corresponding optimum cure (t90) was measured by the swelling method using the Flory-Rehner equation (Eq. S1). Table S1 shows the CLD of both solid and foamed samples increases with MWCNTs loading. This would be attributed to the activation of crosslinking reaction by MWCNTs particles, and also physical crosslinks formed by the MWCNTs within the microstructure of the samples [7][8]. It is worth to note that introducing cellular structure does not have any obvious effect on CLD. Moreover, the obtained results are in consistency with the results discussed earlier.

DYNAMIC MELT RHEOLOGICAL BEHAVIOR
The flowability of the compounds plays a crucial role in the foaming evolution and morphology of the cellular structure. Hence, we analyze the flow behavior and rheological properties as well as dispersion of MWCNTs, using dynamic oscillatory rheometry in the molten state by a reo-mechanical spectrometer (RMS, Paar Physica UDS 200, Graz, Austria). Oscillatory shear melt rheological measurements were conducted using a parallel-plate geometry with a diameter of 25 mm and a gap of 1 mm under controlled shear deformation mode. The linear viscoelastic characteristics of the samples without curing and foaming ingredients were recorded at 160 °C within the angular frequency range of 0.01-1000 (rad/s) with a small strain amplitude of 1% (in the linear viscoelastic region, which was obtained from the strain sweep test). Figure S2.a and S2.b show the variation of the storage modulus (G՛ ) and complex viscosity (η*) versus frequency, respectively. The sample without MWCNTs exhibits Rouse-like viscoelastic behavior i.e. terminal behavior of G՛ at low angular frequency range; whereas all compounds containing MWCNTs show pseudo solid-like or nonterminal behavior, indicating the presence of physical networks of MWCNTs within the EPDM matrix. This is consistent with the increase in melt elasticity (G') and η* as well as shearthinning characteristics with MWCNTs loading fraction [9][10].
To get an insight into the micromorphology, extent of polymer-filler interaction, and dispersion state of MWCNTs, the linear viscoelastic behavior of the prepared nanocomposites was analyzed via calculation of relaxation time spectrum for storage and viscous modulus data by using Paar Physica UDS 200 software. Figure S2.c displays the weighted relaxation spectra, λ.H(λ), versus λ, where H(λ) and λ denote the relaxation time distribution function and relaxation time, respectively. Interestingly, the neat EPDM shows three characteristics peaks located at 0.0126 s, 0.13 s, and 1.34 s. These could be associated with the presence of micro-phases including, ethylene, propylene segments in the backbone of EPDM chain, and attached ethylene long branches with different ease of relaxation. Meanwhile, the inclusion of MWCNTs leads to a change in the weighted relaxation spectra. The low concentration sample (2 phr MWCNTs) presents only one broad peak, i.e. the chain motion is slowed down by the presence of the MWCNTs and their possible interaction with the micro-phases. Higher concentrations (>2 phr) exhibit no characteristic relaxation peak, typical of solid-like features, i.e. large relaxation times, and, hence, suggest the presence of 3D physical networks by the MWCNTs within the EPDM matrix. This is consistent with the dynamic melt rheological results presented in Figure S2.a and b, as MWCNTs particles percolate with each other and restrict the molecular motion leading to no or infinite relaxation times. Similar observation has been reported for the nanocomposites based on Poly(lactic acid) and cellulose nanocrystals [11].

THERMOGRAVIMETRIC ANALYSIS (TGA)
For better understanding, the extent of MWCNTs dispersion and interfacial adhesion between the EPDM matrix and MWCNTs, thermogravimetric analysis (TGA) was conducted on the prepared samples using a thermal analyzer, TA Instruments Series Q500 (New Castle, USA). Samples with the weight of 20 mg were heated from 40 to 800 °C at a heating rate of 10 °C/min under nitrogen atmosphere (with a flow rate of 90 mL /min).
As a high-performance polymer nanocomposite, thermal stability is considered as a key parameter for high-tech applications. The TGA thermographs of solid and foamed EPDM/MWCNTs nanocomposites are displayed in Figure S6.a and b, respectively. As can be seen, compared to the neat and unfilled EPDM samples all EPDM/MWCNTs nanocomposites exhibit higher temperatures for the onset of thermal degradation. This indicates that EPDM segments are thermally shielded by MWCNTs aggregates. However, the foamed samples display lower T90 (the temperature that the sample loses only 10% of its weight) which is attributed to the presence of air molecules inside the cells, activating the onset of thermal degradation of EPDM segments.