Introduction

Electromagnetic interference (EMI), which could cause detrimental effects on the performance of electronic devices, has drawn growing attentions with the repaid development of highly sensitive circuits1,2. Ceramic-based EMI shielding materials have been identified as novel electromagnetic shielding options in the past few years owning to their promising applications as light-weight structural EMI shielding components in areas of aircraft and aerospace3,4,5,6. Compared to traditional metal- and carbon-based EMI shielding materials, ceramic composites exhibit the great advantage of being mechanically stiff and thermo-stable, which could guarantee their functionality at harsh environments6. Besides, the relatively high permittivity and dielectric loss of ceramics are favourable for the attenuation of EMI waves4,5,7,8,9. Nevertheless, the commonly inferior electrical conductivity becomes a major obstacle to obtain high EMI shielding effectiveness (SE) in ceramics. Although the electrical conductivity and EMI SE of ceramics could be both significantly enhanced by incorporating highly conductive carbon materials such as carbon nanotubes (CNTs)4,10, carbon fibers (Cf)6,11, graphene nanoplatelets (GNPs)5,7,8,12, and pyrolytic carbon13,14, the incorporation of those carbon nanostructures in ceramic composites is hindered by the difficulty of homogenous dispersion and their poor oxidation resistance at high temperatures. Accordingly, it is of great interest to explore intrinsically conductive ceramics as effective high-temperature EMI shielding materials.

MAX phase ceramics which represent a class of layered ternary transition-metal carbides and nitrides with a general formula of Mn+1AXn (wherein M is an early transition metal, A is an A-group element, X is either C or N, and n varies from 1 to 3)15 are ideal candidates of high-temperature structural EMI shielding materials owning to their satisfactory mechanical properties, intrinsically superior electrical conductivity and the nano-layered structure15,16. Ti3AlC2 and Ti3SiC2 are two most studied MAX phases and both have been reported to exhibit high EMI SE17,18. Preliminary investigation demonstrates that the nano-layered structure of MAX phases made a considerable contribution to the total shielding effectiveness17.

As was well known, the physical properties of composites usually show remarkable variations upon the change of their microstructures such as porosity, average grain size and interfaces2,19,20,21,22. The physical properties of ceramics could be possibly enhanced by a large scale through the optimization of microstructures. Especially for highly conductive nano-layered structures, the grain size, aspect ratios and the alignment are supposed to have a decisive influence on the EMI shielding properties20. For example, significantly enhanced shielding efficiency was recently reported in highly aligned graphene/polymer nanocomposites20. However, no studies on the dependences of microstructures on EMI shielding properties of ceramics have been reported to our best knowledge. Accordingly, it is of great interest to study the effects of microstructure on the EMI shielding properties not only to obtain excellent high-temperature EMI shielding properties in MAX phase ceramics but also to unfold the origin of high EMI SE in ceramics with unique microstructure. Compared to Ti3SiC2, Ti3AlC2 exhibits much better oxidation resistance at high temperatures23 and is therefore more suitable for high-temperature applications. Accordingly, Ti3AlC2 was chosen as study object, and four typical types of nano-layered Ti3AlC2 ceramics with different microstructures were fabricated by pressureless and hot-press sintering as well as their combination. As a comparison, the EMI shielding properties of highly conductive TiC ceramics without nano-layered structure were also examined. The influences of microstructure on the high-temperature EMI shielding properties were investigated systematically, and the mechanisms for high EMI SE in Ti3AlC2 ceramics were presented.

Results and Discussion

Figure 1 exhibits the SEM images of polished and etched surfaces of the various Ti3AlC2 ceramics prepared by different methods (given in Table 1). Obviously, all Ti3AlC2 ceramics exhibit distinct microstructures. The pressureless-sintered sample shows certain amounts of pores, which is consistent with its relatively lower relative density around 91% as indicated in Table 1. Besides, the aspect ratio of pressureless-sintered Ti3AlC2 grains is relatively low due to the absence of pressure during sintering. All hot-pressed Ti3AlC2 ceramics exhibit dense microstructures and layered grains with higher aspect ratios. The distributions of grain level dimension for different Ti3AlC2 ceramics were statistically calculated and exhibited in Fig. S1. The majority of grains for pressureless-sintered Ti3AlC2 ceramic (TAC-A) show level dimensions in the range of 8–14 µm. The average level dimension of hot-pressed samples increases significantly with increasing sintering temperature and duration, which is consistent with previous study24. Both the Ti3AlC2 ceramics sintered at 1350 °C for 15 min (TAC-B) and 1400 °C for 2 h (TAC-C) show broad level dimension distributions. The TAC-B shows a level dimension peak around 10 µm while the distribution peak of TAC-C locates at a higher size range around 20 µm. A followed thermal process at 1500 °C for 10 h further promotes the grain growth with a dominant percentage of grains with level dimension larger than 25 µm and a considerable percentage of grains with level dimension larger than 40 µm. All the Ti3AlC2 ceramics exhibit the same single-phase crystal structure without detectable second phases, which can be evidenced by the XRD patterns shown in Fig. S2 in the supporting information.

Figure 1
figure 1

SEM images of polished surfaces of various Ti3AlC2 ceramics with distinct microstructures. (a) TAC-A; (b) TAC-B; (c) TAC-C; (d) TAC-D. All Ti3AlC2 ceramics show typical layered grains. The pressureless sintered sample shows certain degree of pores. With the increase of hot-pressing temperature, the microstructure becomes denser and shows significant grain growth.

Full size image
Table 1 Comparison of relative density and electrical conductivity of various Ti3AlC2 and TiC ceramics.
Full size table

The total SE (SET) of EMI shielding materials can be characterized as follows25:

$${{\rm{S}}{\rm{E}}}_{{\rm{T}}}=10\,{\rm{l}}{\rm{o}}{\rm{g}}({{\rm{P}}}_{{\rm{i}}}/{{\rm{P}}}_{{\rm{t}}})={{\rm{S}}{\rm{E}}}_{{\rm{R}}}+{{\rm{S}}{\rm{E}}}_{{\rm{A}}}+{{\rm{S}}{\rm{E}}}_{{\rm{M}}}$$
(1)

in which SET, SEA SER and SEM denote the total SE, SE due to reflection, absorption loss and multiple reflections, respectively. SET, SER and SEA were calculated from the S-parameters as follows26,

$${{\rm{S}}{\rm{E}}}_{{\rm{T}}}=-\,10\,{\rm{l}}{\rm{o}}{\rm{g}}\,{|{{\rm{S}}}_{12}|}^{2}$$
(2)
$${{\rm{S}}{\rm{E}}}_{{\rm{R}}}=-\,10\,{\rm{l}}{\rm{o}}{\rm{g}}(1-{|{{\rm{S}}}_{11}|}^{2})$$
(3)
$${{\rm{S}}{\rm{E}}}_{{\rm{A}}}=-\,10\,{\rm{l}}{\rm{o}}{\rm{g}}(\frac{{|{{\rm{S}}}_{21}|}^{2}}{1-{|{{\rm{S}}}_{11}|}^{2}})$$
(4)

Figure 2a–c compare the frequency dependences of room-temperature SET, SER and SEA respectively of Ti3AlC2 ceramics with different microstructures. The SET exhibits notable differences among various Ti3AlC2 ceramics. The pressureless sintered Ti3AlC2 ceramic exhibits the lowest SET around 30 dB, which is comparable to those of most ceramic-based composites5,11,27. The hot-pressed Ti3AlC2 ceramics exhibit remarkably enhanced SET compared to pressureless sintered sample. Further, the room-temperature SET of hot-pressed samples increases monotonically with the increase of grain size. TAC-D ceramic with the largest grain size shows the highest SET over 40 dB at the whole Ku-band, which means over 99.99% of the incident radiation could be effectively blocked with only 0.01% transmission. The room-temperature EMI SET of TAC-D ceramic is higher than most ceramic-based EMI shielding composites containing conductive particles or carbon nano-fillers3,4,5,6,8,27, while the preparation of Ti3AlC2 ceramics is much less complicated. Moreover, the high SET of TAC-D ceramic is almost independent of the frequency. In Equation (1), the SEA can be regarded as the energy dissipation of the electromagnetic microwave in the absorber and therefore the multiple reflection SEM in single phase Ti3AlC2 ceramics actually is included in absorption because the re-reflected waves could get absorbed or dissipated within the material1. Accordingly, the SET could be expressed as,

$${{\rm{S}}{\rm{E}}}_{{\rm{T}}}={{\rm{S}}{\rm{E}}}_{{\rm{R}}}+{{\rm{S}}{\rm{E}}}_{{\rm{A}}}$$
(5)
Figure 2
figure 2

The frequency dependences of (a) the total shielding effectiveness SET, (b) reflection loss SER and (c) absorption loss SEA of various Ti3AlC2 ceramics with distinct microstructures measured at room-temperature. Ti3AlC2 ceramics with various microstructures exhibit different EMI shielding capability and those differences mainly arose from the distinct absorption loss.

Full size image

In order to have a deeper understanding on the origin of microstructure-dependent EMI shielding properties in Ti3AlC2 ceramics, the dependences of microstructure on the reflection loss SER and absorption loss SEA of Ti3AlC2 ceramics were investigated and the results were exhibited in Fig. 3b and c respectively. It can be seen that the SER of Ti3AlC2 ceramics is almost independent of their microstructure. All Ti3AlC2 ceramics with various microstructures show similar SER values between 10 and 14 dB, especially, the SER of all the hot-pressed ceramics exhibit almost the same values at the whole Ku-band frequency range. The reflection loss of EMI shielding materials arises mainly from the impedance mismatch between the sample and the free space25. The extremely high electrical conductivity (~106 S/m) of Ti3AlC2 ceramics could lead to a huge impedance mismatch and therefore give rise to a high reflection loss. The pressureless sintered Ti3AlC2 ceramics exhibit inferior electrical conductivity due to high porosity (Table 1), accordingly, its SER is relatively lower. While for dense Ti3AlC2 ceramics, the influence of grain size on the conductivity is negligible and therefore the SER becomes almost independent of the microstructure. All Ti3AlC2 ceramics exhibit high absorption loss SEA (Fig. 2c) which is much larger than SER, indicating that Ti3AlC2 ceramics are typically absorption-dominant EMI shielding materials. Specially, high SEA of 30 dB which is almost 3 times larger than SER was obtained for TAC-D ceramic at 12.4 GHz. Being different from SER, the absorption loss shows obvious grain size dependence and becomes the major contribution to the variation of SET among Ti3AlC2 ceramics with different microstructures, as demonstrated in Fig. 2c. Hot-pressed Ti3AlC2 ceramics exhibit higher SEA than pressureless sintered sample. For hot-pressed Ti3AlC2 ceramics, the SEA increases gradually with increasing grain size. Therefore, it can be concluded that dense and coarse-grained microstructure with higher aspect ratios is more favorable for high EMI attenuation and the total EMI shielding in nano-layered Ti3AlC2 ceramics.

Figure 3
figure 3

Comparison of the shielding effectiveness of TiC ceramics with different grain size. The shielding effectiveness was measured at room temperature. The shielding capability of TiC ceramics shows much weaker grain size dependence compared to Ti3AlC2 ceramics.

Full size image

In order to further validate the significant contribution of the nano-layered structure to the absorption loss, the EMI shielding properties of highly conductive TiC ceramics without nano-layered structures were characterized and compared with those of Ti3AlC2 ceramics. Figure S3 shows the fracture surfaces of TiC ceramics sintered at different temperatures. Both TiC ceramics exhibit different microstructures from Ti3AlC2 ceramics without layered structures. It is apparent that the grain size of TiC-B is much larger than that of TiC-A. The frequency dependences of the EMI SE for different TiC ceramics were shown in Fig. 3. It is interesting to notice that, being different to that of Ti3AlC2 ceramics, the change of grain size did not bring significant variation of SET for TiC ceramics. Especially, the SEA for fine- and coarse-grained TiC ceramics are almost the same. Compared to Ti3AlC2 ceramics, the TiC ceramics shows much lower SEA while higher SER. The microstructure-independent SEA of TiC ceramics varies in the range of 18–25 dB at the whole frequency range. The SER has almost equal contribution to the total shielding for both TiC ceramics. The different shielding properties between Ti3AlC2 and TiC further confirms that the nano-layered structure is responsible for the superior and microstructure-dependent SEA in Ti3AlC2 ceramics. The reflection loss of Ti3AlC2 ceramics is considerably lower than that of TiC ceramics although both ceramics exhibit similar electrical conductivity as demonstrated in Table 1.

Similar to most EMI shielding materials with two-dimensional nano-layered structures, the high SEA of nano-layered Ti3AlC2 ceramics arises mainly from two major attenuation mechanisms: the high electrical/dielectric loss and layered architecture1,5,19,20,22. Figure 4 presents the real (\({\varepsilon }^{{\rm{^{\prime} }}}\)) and imaginary (ε″) parts of permittivity as well as the dielectric loss tan defined as9

$$tan\delta ={\varepsilon }^{{\rm{^{\prime} }}{\rm{^{\prime} }}}/{\varepsilon }^{{\rm{^{\prime} }}}$$
(6)

of TAC-D ceramic measured at room-temperature. It can be seen that Ti3AlC2 ceramics exhibit moderate real permittivity around 15 while significantly higher imaginary part of ~40. Consequently, the dielectric loss of Ti3AlC2 ceramics reaches a considerably high value around 2.5. The high electrical/dielectric loss of Ti3AlC2 ceramics arises from the abundant free electrons and a large quantity of unpaired defects16. The conduction current owing to the skin effect as well as the eddy current will contribute to the ohmic loss, and the unpaired defects will contribute to the dielectric loss. Compared to the electrical/dielectric loss, the multiple reflections also played a significant role in achieving high SEA in nano-layered structures such as graphene and MAX phases, and the multiple reflections are more likely to display microstructure dependences in ceramics. Ti3AlC2 ceramics exhibit layered grains and therefore abundant flat grain boundaries. When the incident EMI radiations encounter those flat grain boundaries, a portion of the EMI radiations gets immediately reflected due to the impedance mismatch of adjacent layers with the others penetrating through these interfaces. The penetrated EMI radiations can be reflected back and forth between the adjacent grain boundaries. These multiple reflections could increase the length of EMI transmission pathway by a large scale and enhance the attenuation of EMI accordingly1,7,8,20. In coarse grains with high aspect ratios, a higher degree of multiple reflections can be expected due to the large surface area of a single grain boundary and thereby a higher absorption loss. Consequently, the origin of microstructure dependent EMI shielding effectiveness can be ascribed to the enhanced degree of multiple reflections with increasing level dimensions of layered Ti3AlC2 grains.

Figure 4
figure 4

The frequency dependent permittivity (real part ε′ and imaginary part ε″) and loss tangent (tanδ = ε″/ε′) of TAC-D ceramic measured at room-temperature. The Ti3AlC2 ceramics show high imaginary permittivity ε″ and high dielectric loss at the whole Ku band.

Full size image

Figure 5 shows the frequency dependences of SET at different temperatures up to 600 °C for different Ti3AlC2 ceramics. It can be seen that the SET of all Ti3AlC2 ceramics only shows a slight decrease with increasing temperature. The variations of SET from room-temperature to 600 °C for all Ti3AlC2 ceramics is less than 10%. A high SET around 38 dB was still maintained at 600 °C for TAC-D ceramic, demonstrating that the EMI shielding properties of Ti3AlC2 ceramics have excellent thermo-stability and is well suitable for high-temperature EMI shielding applications. Figure S4 shows the frequency dependences of SER and SEA of different Ti3AlC2 ceramics measured at 600 °C. By comparing Fig. 2 and Fig. S4, it can be easily observed that for all Ti3AlC2 ceramics the SER kept almost the same with increasing temperature to 600 °C. Although the electrical conductivity of Ti3AlC2 ceramics tend to decrease with increasing temperature, the giant impedance mismatch between Ti3AlC2 ceramics and the air made the SEA almost unaffected by the slight decrease of electrical conductivity. The decrease of SET solely arises from the decrease of SEA with increasing temperature. The SEA decrease equally for all Ti3AlC2 ceramics and TAC-D ceramic still exhibits the highest SEA around 26 dB in the Ku-band frequency range at 600 °C. The slight decrease of absorption loss with increasing temperature can be ascribed to the less internal friction of dipole reorientation at high temperatures.

Figure 5
figure 5

The frequency dependent SET of various Ti3AlC2 ceramics measured at different temperatures. (a). TAC-A; (b). TAC-B; (c). TAC-C; (d). TAC-D. The shielding effectiveness of all Ti3AlC2 ceramics exhibit satisfactory temperature stability and high EMI SE still persist at high temperatures.

Full size image

Conclusions

Various Ti3AlC2 ceramics with distinct microstructures were prepared by pressureless sintering and hot-pressing. Their EMI shielding properties were characterized at Ku-band frequency range from room temperature to 600 °C. It was found that the EMI shielding properties of Ti3AlC2 ceramics display remarkable microstructure dependences. High EMI shielding effectiveness over 40 dB was obtained in coarse-grained Ti3AlC2 ceramic and the excellent shielding properties were well maintained at high temperatures up to 600 °C. The variations of EMI SET with changing microstructure dominantly arises from the variation of absorption loss SEA which is related to the high electric/dielectric loss and more importantly the nano-layered structure of Ti3AlC2. Coarse Ti3AlC2 grains with high aspect ratio are more favourable for high multiple reflections of EMI waves and consequently more absorption loss. Highly conductive TiC ceramics without nano-layered structure exhibit almost microstructure-independent shielding properties and much lower absorption loss than that of Ti3AlC2 ceramics. These results indicate that Ti3AlC2 ceramic are promising high-temperature EMI shielding materials.

Methods

Ti3AlC2 powder (purity ≥ 98%, ∼300 mesh) was purchased from Forsman (Beijing) Scientific Co., Ltd. Four types of Ti3AlC2 ceramics referred as TAC-A, TAC-B, TAC-C and TAC-D respectively were prepared by different sintering techniques at different temperatures. In the case of TAC-A, the Ti3AlC2 powder was first cold-pressed into pellets of 50 mm in diameter and 2 mm in thickness under a uniaxial pressure of 50 MPa, and then pressureless-sintered in Ar atmosphere at 1450 °C for 120 min; In the case of TAC-B, TAC-C and TAC-D, the Ti3AlC2 powder was placed in a 50 mm diameter graphite die and hot-pressed in Ar atmosphere under different conditions: TAC-B was hot-pressed at 1350 °C for 15 min; TAC-C was hot-pressed at 1400 °C for 120 min; TAC-D was hot-pressed at 1400 °C for 120 min followed by a pressureless sintering in Ar atmosphere at 1450 °C for 500 min. During the hot-press sintering a uniaxial pressure of 30 MPa was applied. Two distinct types of TiC ceramics with 50 mm in diameter and 2 mm in thickness were prepared by hot-pressing at 1800 °C for 15 min (TiC-A) and 1950 °C for 2 h (TiC-B) respectively using commercial TiC powders (∼100 nm, Forsman (Beijing) Scientific Co., Ltd.). A uniaxial pressure of 60 MPa was applied during sintering. The bulk density of the prepared ceramics was measured using the Archimedes’ method. Crystalline phases were characterized by X-ray diffraction (XRD). For microstructure characterization, the Ti3AlC2 samples were polished using SiC papers and diamond suspensions down to 0.5 µm. The well-polished surfaces were further etched in an acid solution with a volume ratio of HF: HNO3:H2O = 1:1:3 for 45 s. The morphology of different specimens was observed by field emission scanning electron microscopy (FE-SEM). The grain size distribution of Ti3AlC2 ceramics was analysed on SEM images (with a magnification of 1000×) of polished and etched surfaces using ImageJ software. In order to ensure the accuracy of measurement, at least 1000 grains from the polished and etched surfaces were considered to statistically obtain the grain level dimension distributions.

For EMI SE characterization, specimens with dimensions of 22.86 mm × 10.16 mm × 1.00 mm were cut and polished. The magnitudes of complex scattering parameters (S-parameters) that correspond to reflection (S11 or S22) and transmission (S21 or S12) in the frequency range of 12.4~18 GHz (Ku band) were determined through wave-guide method using a vector network analyzer (Agilent N5230A). For accuracy of measurement, the device is carefully calibrated with Through-Reflect-Line (TRL) approach. The high-temperature measurement of SE was performed in a waveguide heated by an inner heater at a rate of 10 °C/min. The temperature range is 25–600 °C and each temperature spot was stabilized for 10 min in order to ensure the accuracy of measurement.