Electromagnetic Wave Shielding Effectiveness Based on Carbon Microcoil-Polyurethane Composites

Carbon microcoils (CMCs) were deposited onto Al 2 O 3 substrates using C 2 H 2 /H 2 as source gases and SF 6 as an incorporated additive gas in a thermal chemical vapor deposition system. CMC-polyurethane (PU) composites were obtained by dispersing the CMCs in the PU with a dimethylformamide additive. The electromagnetic wave shielding properties of the CMC-PU composites were examined in the frequency range of 0.25–1.5 GHz. The shielding effectiveness (SE) of the CMCs-PU systematically increases with increasing the content of CMCs and/or the layer thickness. Based on these results, the main SE mechanism for this work was suggested and discussed.


Introduction
Recently, the importance of preventing electromagnetic interference (EMI) on diverse electronic equipment has increased with the rapid development of radiation sources and the high reliability requirements for electronic devices [1][2][3].
Absorption, reflection, and multiple reflections of EM radiation by the electronic components are regarded as the main shielding mechanisms for EMI [3][4][5][6]. For materials having high electric constants or magnetic permeability, absorption is known to be the major EMI shielding mechanism [3]. The absorption loss is a function known as ( is the electrical conductivity relative to copper; is the relative permeability). For metals, the EMI reflection from the metal's free electrons is regarded as the major shielding mechanism [3,4], usually requiring an interaction between the mobile charge carriers and the electromagnetic fields in the radiation. Consequently, the shield tends to be electrically conducting, although it does not need a complete connection in the conduction path [5]. In comparison, the reflection loss is a function of / . In general, the reflection loss decreases with increasing frequency, whereas the absorption loss increases with increasing frequency [6]. Additionally, the reflection loss is independent of the shield thickness, while the absorption loss is proportional to the thickness of the shield [6]. Meanwhile, the multiple reflections caused by interfaces and/or the various surfaces of the shielding materials are regarded as another primary EMI shielding mechanism. This mechanism requires a large surface or interface area, such as a composite material containing fillers. Fu and Chung reported that small fillers with high surface areas in a composite gave rise to enhanced shielding performance owing to the skin effect, namely, the interaction of highfrequency radiation with only the surface of the materials [7].
Up to the present, metals have been commonly used as EMI shielding materials in the 0.5-2.0 GHz range for mobile phones. Owing to the increasing demand for lightweight and moldable materials, polymer-matrix composites for use as EMI shielding materials are strongly desired for portable electronic devices, avionic electronics, and so forth, [7,8]. In this respect, carbon materials (including carbon fibers, carbon nanotubes, carbon blacks, carbon coils, graphites, etc.) have been seen as promising candidates for EMI shielding materials [9][10][11][12][13]. In particular, carbon microcoils (CMCs), which have DNA-like double helix geometry, are appealing as electromagnetic wave absorbers because they appear to 2 Journal of Nanomaterials  induce an electrical current, consequently generating a magnetic field [12,13]. Indeed, the coil geometry was understood to be an effective form for inducing current through an inductive electromotive force, unlike that of the straight-or powder-like forms. Unfortunately, the geometries of the carbon coils were diverse in the as-grown state. Additionally, their diameter could vary from the nanometer to the micrometer scale. Indeed, the electrical properties of the helically coiled CMCs can have considerable geometry-dependent variation, similar to that of straight carbon nanotubes [14]. Therefore, it would be indispensable to be able to control the geometry (diameter, pitch, length, and turning direction) of the carbon coils for their application as electromagnetic wave absorbers.
Previously, we reported that the injection of SF 6 gas during the initial reaction stage was effective on controlling the geometry of the formed CMCs [15][16][17]. In this work, we obtained the geometrically controlled CMCs using the SF 6 additive and then formed CMC-polyurethane (PU) composites by dispersing the CMCs in the PU with a dimethylformamide (DMF) additive. Polymethyl methacrylate [12] and paraffin wax [13] have been reported for the polymer matrix of carbon coils. In this work, however, we chose PU for the matrix because it is more readily available. The EM wave shielding properties of the CMC-PU composites were examined according to the weight percent of the CMCs in the PU and the composite layer thickness across the frequency range of 0.25-1.5 GHz. Based on these results, we determined the main shielding mechanism for the CMC-PU composites.

Experimental
For the carbon coil deposition, a thermal chemical vapor deposition (TCVD) system was employed. C 2 H 2 and H 2 were used as the source gases, and SF 6 , an incorporated additive gas, was injected into the reactor during the reaction. The flow rate for C 2 H 2 , H 2 , and SF 6 was fixed at 15, 35, and 35 standard cm 3 per minute (sccm), respectively. Table 1 shows the detailed reaction conditions for the CMC deposition. The detailed morphologies of the as-grown carbon coils were investigated using field emission scanning electron microscopy (FESEM).
For the CMC-PU composites, the CMCs were dispersed in the PU with the addition of DMF using an ultrasonic system. The range of PU molecular weight in this work was 60,000∼70,000. After 120 min of on/off ultrasonic treatment at 500 W and 20 kHz, a paste-type CMC-PU-DMF mixture was obtained. We prepared three kinds of samples (samples A, B, and C) having different CMC composition ratios in the paste-type CMC-PU-DMF mixtures as shown in Table 2. After manufacturing the three kinds of paste-type samples, each sample was coated onto a circular-shaped glass plate 133 mm in diameter. For the coating, about 20 mL of the paste-type sample was poured onto the glass plate, and then the coated samples were dried naturally in a fume hood for about 24 hours. Finally, the weight of the coated samples was measured using an electronic balance (BJ210s, Sartorius).
For the electrical resistivity measurements, we fabricated square CMC-PU sheets having dimensions of 40 (length) × 35 (width) mm. For the coating, about 2 mL of the pastetype samples was poured onto the sheet, and then the coated sheets were naturally dried in the fume hood for 24 hours. The volume resistivity (Ω cm) of the sheets was measured by fourpoint probe (labsysstc-400, Nextron) using Ohm's law and a correction factor at room temperature [18].
The SE of the CMC-PU composites was analyzed using a network analyzer (SynthNV2 3b, Windfreak Tech.) in accordance with ASTM D4935-99. The setup consisted of a sample holder with its outside connected to the network analyzer (see Figure 1). The coaxial sample holder (Electro Matrix EM-2107A) and the coaxial transmission test specimen were set according to ASTM D4935-99 as shown in Figure 2. The performance measurement range of the SE for the CMC-PU composites was from 250 MHz to 1.5 GHz. Figure 3 shows the FESEM images of the substrate surface morphologies for samples produced by the continuous C 2 H 2 + H 2 flow process with 5 min of SF 6 gas flow addition during the initial reaction. Figures 3(b) and 3(c) show the magnified Journal of Nanomaterials images of Figures 3(a) and 3(b), respectively. As shown in these figures, the well-structured CMCs were prevalent on the surface on the sample. These results reveal that the continuous C 2 H 2 + H 2 flow process with the 5 min of SF 6 gas flow addition during the initial reaction can give rise to the dominant form of the CMCs. As shown in Figure 3(c), the first and second rings of the CMCs are clearly observed, indicating that the formation of the CMCs in this work follows the typical double-helix geometry for the structure. For the shape of the rings constituting the coils, circular-type morphology was observed. Figure 4(a) shows a representative photograph of the coated paste-type CMC-PU-DMF mixture on the glass plate. Figure 4(b) shows a representative cross-sectional FESEM image of the coated samples after the natural drying process. As shown in Figure 4(b), the existence of the CMCs in the coated layers could be clearly observed. The weight variation for the coated layers of the different samples was measured as a function of the number of coatings. The time intervals between the coatings were approximately 24 hours. As shown in Figure 5, the weight of the coated layers gradually increased with the number of coatings. Although we poured almost  the same amount (∼20 mL) of the paste-type sample for every coating, the increased weight of the coated layers of the samples was not directly proportional to the number of coatings. This was due to the volatile DMF, which evaporated during the natural drying process. As expected, sample C, having the highest composition ratio of CMCs in the CMC-PU-DMF mixture, gave rise to the highest weight of the coated layers among the samples, irrespective of the number of coatings. For the thickness of the coated layers, we estimated that the proportional ratio for the layers' weight (g), to the layer thickness (mm), was approximately 10, although the thickness of the coated layers might be uneven at different locations on the sample. This estimation was carried out by the comparison between the cross-sectional FESEM image of a certain point of the sample and the weight of the sample.

Results and Discussion
After measuring the weights of the coated layers on the samples, we examined the volume resistivity dependence of the CMC-PU sheets relative to the number of coatings as shown in Figure 6. The sheet corresponding to sample C has the lowest volume resistivity among the sheets under similar thicknesses (see Figure 6 and the inset in Figure 6).  As expected, the higher amount of CMCs in the PU-CMC-DMF mixture, the higher the electrical conductivity of final CMC-PU sheet. Considering only the reflection and the absorption effects as the main shielding mechanisms for the EM interference of this work, the SE of the EM interference for the electrically conductive polymer composites can be estimated by the empirical equation of Simon [19]: where SE is in dB, is the volume resistivity (Ω cm) at room temperature, is the thickness of the sample (cm), and is the measurement frequency, respectively. Indeed, the multiple reflections effect, likely significant in the nanoscale filler system, was thought to be minimized in this work because we used only micron-sized carbon coils. In the previous equation, the combined first and second terms, namely, Journal of Nanomaterials 50 + 10 log 10 ( ) −1 , show only the reflection mechanism SE. The third term, namely, 1.7 ( / ) 1/2 , indicates only the absorption mechanism SE. As indicated by this equation, the SE from the reflection decreases with increasing measurement frequency, while the SE from the absorption increases with increasing measurement frequency. For the composite layer thickness dependence, only the SE from the absorption mechanism increases with increasing thickness ( ).
For the SE measurement, we investigated the SE variation compared to the different CMC composition ratios in the CMC-PU-DMF mixture (see Figure 7) and the different thicknesses of the composite (see Figure 8). As shown in Figure 7, we compared the SE of the CMC-PU composites for samples A-C under similar composite layer thicknesses. Sample C, with the highest electrical conductivity of the CMC-PU composites in this work, also has the highest SE among the samples. This indicates that the reflection effect may work as the dominant SE mechanism for the EMI in these composites. However, the combined result of Figures  6 and 7 shows that the increase in the SE is not directly proportional to the volume resistivity increase. Accordingly, this result strongly informs us that the other mechanisms besides reflection, namely, the absorption effect, will play a critical role as the SE mechanism for the EMI of these composites. Measuring the frequency dependence of the SE for these composites, we could not confirm any variation of the SE with the frequency. However, we are convinced that the SE did not decrease with increasing measurement frequency in the range of 1.0 GHz to 1.5 GHz. This result also reveals that the main EMI mechanism in these composites is not the reflection effect. Furthermore, Figure 8 clearly reveals that the SE increases with an increase of the coated layer thickness, clearly indicating that the absorption mechanism is the composite's main EMI mechanism in this work.
As shown in Figure 8, for Sample B, layers with thicknesses greater than 2.0 mm show a SE above 20 dB, demonstrating that more than 99% of the injected electromagnetic waves have been shielded. Even from the industrial point of view, this value is considered sufficient for use across numerous application fields. As the previous reports, CMCs,  associated with the absorption mechanism of the SE, would be useful mainly in the relatively high frequency (above several GHz) region [12,13]. However, this work confirms that the CMC-PU composite will be applicable even in the mobile communication region (around 1∼2 GHz).

Conclusions
The highest electrical conductivity of the PU-CMC composites in this work corresponds to the highest SE among samples under similar coated layer thicknesses. However, the increase of the SE is not directly proportional to the increase of the volume resistivity. Additionally, the SE did not decrease with increasing measurement frequency in the range of 1.0 GHz to 1.5 GHz. Furthermore, the SE increases with an increase of the composite layer thickness. Based on these results, we confirm that the main EMI mechanism for the composites in this work is the absorption mechanism.