Fig.1 depicts the room temperature XRD patterns of (Mg0.2Ni0.2Co0.2Cu0.2Zn0.2)O ceramics fabricated by different techniques. As can be seen, the diffraction peaks are sharpen and narrow, indicating the fabricated samples exhibit a single rock salt phase, face-centered cubic (fcc) spinel structure with () group[20]. However, although no additional peaks corresponding to impurities or secondary phase can be detected in the above-mentioned samples, implying high purity of the fabricated samples, the relative location of the Bragg (111) and (200) peaks slightly changed from one precursor to another, as shown in Fig.1b, which manifest a strong dependence on the fabrication process[21]. According to the XRD results, the interplanar distance of (200) plane for S1, S2 and S3 is 2.11 nm, 2.12 nm and 2.10 nm, respectively, the observed XRD patterns of S3 match well with the MgO phase, as listed in Table 1.
Table 1 Crystal structures, space group[22], oxidation state, co-ordination number[20] and corresponding interplanar distance of (200).
Oxide
|
Structure
|
Space Group (number)
|
Oxidation
|
Co-ordination number
|
Interplanar distance of (200) plane [nm]
|
CoO
|
Rocksalt
|
(225)
|
2+
|
VI
|
2.13
|
MgO
|
Rocksalt
|
(225)
|
2+
|
VI
|
2.10
|
NiO
|
Rocksalt
|
(225)
|
2+
|
VI
|
2.09
|
CuO
|
Tenorite
|
(15)
|
2+
|
VI
|
2.31
|
ZnO
|
Wurtzite
|
(186)
|
2+
|
IV
|
1.41
|
The micrographs reported in Fig.2 reveal the surface morphology and section view of (Mg0.2Ni0.2Co0.2Cu0.2Zn0.2)O ceramics. According to the surface morphology, as shown in Fig.2 (a-c), the surface flatness of S1 is much lower than that of S2 or S3. Combined with the sections,there are a lot of holes in S1 indicating that the density of the S1 is significantly lower than that of S2 or S3.This may due to the S1 green body is composed of hydrous chlorides, while the S2 and S3 green bodies are composed of hydroxides or oxides. Thus, the volatilization discharges of the S1 are significantly greater than that of S2 or S3 during the calcination process. Therefore, the structure of the S1 is relatively loose. At the same time, the loose structure can also affect the grain growth. So, the grain size of S1 is obviously smaller than that of S2 or S3, as shown in Fig.2 (d, e, f). In addition, compared with the d, e and f in Fig. 2, it can be seen that the S1 grains have no dominant growth orientation, while the S2 grains have the obviously dominant growth orientation. According to the XRD results in Fig. 1, the intensity of each characteristic peak in S2 is the weakest, indicating that the preferred growth orientation of S2 grains deviates from the lattice face orientations of characteristic peaks. The XRD variation trend of S1, S2 and S3 is consistent with the microstructure change trend of SEM.
The grain size of S2 is larger than that of S3, which is related to the different preparation technologies of precursors. In general, compared with the co-precipitation method (S2), it is easier to obtain nano-dispersed powders through hydrothermal method (S3)[23]. Meanwhile, the grain sizes of sintered samples are also directly affected. The grain size, characteristic peaks intensity and grain orientation degree of the different samples all fit the order (S2>S3>S1). These results indicate that there is a certain correlation between the grain size and crystal structure for high entropy ceramics.
It is worth noting that there are obvious gaps between the grains of S2, the grain bonding degree of S2 is lower than that of S3. It could be explained that the thermal stress was generated in the matrix, when the samples naturally cooled with the furnace as after sintering at high temperature. Due to the small grain size of S3, its thermal shock resistance was better than that of S2. In addition, through the section in Fig.2 (f), it is also found that there are a large number of micropores in S3, which are also very beneficial to the improvement of thermal shock resistance[24]. Therefore, the bonding tightness of S3 matrix is relatively high after cooling. The S3 was composed of nano powders which are easy to be sintered. Meanwhile, the vapor discharge leaved a lot of holes to the green body. So, the matrix contained a large number of micropores after sintering at 1200oC. While in the Fig.2(f), there are well distributed small pores located inside the grain or at the grain boundary, which will be benefit to enhancing the electromagnetic wave absorbing properties.
As shown in Fig.3, all elements are homogeneously distributed within the samples, without any tendency towards segregation formation of other phases.
It is very obviously that the grain (1) and grain (3) present a completely different microscopic morphology as seen in Fig. 4(a). It can be seen from Fig. 4(b) that the interface area includes two kinds of characteristics coming from grain (1) and grain (3) respectively. In addition, there are many defects in the bonding interfaces. According to this phenomenon, there are two reasons inferred. Firstly, there are two different matters (1 and 3), due to the uneven mixing among the components of high entropy ceramics. Secondly, 1 and 3 are the same matter, but the grain is anisotropy, the growth directions of two grains are different due to the slight differences in the crystal growth process, thus, the two grains have different morphologies observed from the same direction. So, the EDS analysis was carried out to determine composition.
Fig. 5 shows the elements distribution of (Mg0.2Ni0.2Co0.2Cu0.2Zn0.2)O c eramics (S3)with heterogeneous interfaces. The elements (Mg, Ni, Co, Cu, Zn, O) evenly distribute in entire regions. The 1 and 3 have same element composition. That is to say, the 1 and 3 may present different morphologies due to the grain anisotropy.
As seen from Fig. 6 (a, c, d), the 1 and 3 area present different electron diffraction patterns. It is noteworthy that the Fig.6 (a) and Fig.6 (c) represent diffraction patterns on different crystal faces of the same substance. The crystal plane calibration results of SADP are in good agreement with XRD (Fig. 1). Due to the difference of crystal growth direction, the angles between the electron beam incident direction and the crystal faces of different crystals are different, which leads to different diffraction patterns. It also indicates that the grain of high entropy ceramics is anisotropic and the single crystal growth is orientated. The results are also consistent with SEM (Fig. 2). Although a single grain is anisotropic, the polycrystalline high-entropy ceramics are macroscopically isotropic due to the random growth direction of each grain. It can be seen that the interface area (2) has two kinds of electron diffraction patterns. The 1 and 3 formed heterogeneous interfaces at 2 causing lattice distortions. Fig. 6(e) is HRTEM of the 2 area. It is obvious that areas near 3 are stripes, and areas near 1 are lattices. According to characteristics of cubic crystal system, the HRTEM results and SADP results support each other. The fringe spacing is about 0.266 nm, which is close to the (111) inter-planar spacing. According to SADP Fig. 6(c), the crystal plane of stripes is close to (420). The angel between (420) and (111) is more than 70o. So, the measured spacing of fringes is different from the theoretical spacing of (111). It is not perfectly regular. The solid red circles show significant distortion of the crystal lattices. In lattice area, the lattice widths are measured at two places, and the values ware 0.141 and 0.225 nm respectively. This further indicates the presence of lattice distortion at heterogeneous interfaces.
Fig.7 showed the probable EM parameters of the (Mg0.2Ni0.2Co0.2Cu0.2Zn0.2)O -paraffin composites. Fig.7 (a) shows the frequency dependencies of the real part of complex permittivity of (Mg0.2Ni0.2Co0.2Cu0.2Zn0.2)O, the ε′ value of S3 sample keeps around at 4.5, it is clear that S3 exhibits a higher real permittivity than S1 and S2. Making comparison between all these samples, it clearly shows that the three samples have similar fluctuation tendency at 2–12 GHz. As shown in Fig.7 (b), the ε′′ values of the three samples are almost the same.
The electromagnetic wave absorbing performance of (Mg0.2Ni0.2Co0.2Cu0.2Zn0.2)O is evaluated by the values of reflection loss (RL), which is based on the transmission line theory. In this work, the RL is described by the following equations:
Where the input impedance (Zin) of the composite is given by,
Where f is the frequency of electromagnetic wave, d is the thickness of the sample, c is the velocity of electromagnetic wave, is the impedance of free space, and and are the complex relative permittivity and complex permeability, respectively. Generally speaking, the RL value exceeding -10 dB demonstrates 90% absorption of incident EM wave, implying an ideal EM wave absorber. For the (Mg0.2Ni0.2Co0.2Cu0.2Zn0.2)O ceramics, as present in Fig.6, the optimum RL for S1, S2 and S3 achieved to-16.5 dB, -17.9 dB and -30.5 dB at 6.8 GHz under the thickness of 4.0 mm, respectively, and the bandwidth of S3 is 1.3 GHz (6.2 to 7.5 GHz). It can be confirmed that the enhanced microwave absorbing properties of (Mg0.2Ni0.2Co0.2Cu0.2Zn0.2)O ceramics (S3) can be obtained by the appropriate boundary defect and its micro holes. The results demonstrate that (Mg0.2Ni0.2Co0.2Cu0.2Zn0.2)O ceramics may be a promising candidate for microwave absorption materials.
After sintering at 1200 oC, the S3 samples had good microwave absorption properties. As a comment of (Mg0.2Ni0.2Co0.2Cu0.2Zn0.2)O ceramics, the melting point of CuO is 1326 oC. Therefore, it is necessary to explore the temperature limit that the microwave absorbing high entropy ceramics can withstand. At high temperature, the structure stability is the key to ensure good absorbing performance. The S3 was selected to test high temperature stability. Fig. 9 shows the side views of S3 at different temperatures. As seen from Fig. 9, the height of S3 hardly changes from 900 oC to 1200 oC. Compared with 1200 oC, the height of S3 is slightly decreased at 1300 oC. At 1400 oC, the height and width of S3 significantly decrease. These results indicate that the S3 is still stable at 1200oC.