1 Introduction

Advances in electromagnetic interference (EMI) shielding materials have sparked considerable attention in almost every electronics industry for attenuating electromagnetic radiation of complex electronic systems [1, 1i).

Fig. 1
figure 1

Scalable manufacturing of C-MXene@PI composite foams: schematic of preparation process of a MXene flakes and b C-MXene@PI composite foams, c TEM (inset shows electron diffraction image) and d AFM (inset shows the height profile of a monolayer) images of the MXene flakes. e, f Photographs of freestanding and large-area (~ 60 × 60 cm2) C-MXene@PI composite foams before and after rolling, showing the mechanical robustness and flexibility. SEM images of the g micrometer-sized pores and h cell walls, and i Ti-, C-, O-, and N- element map**s of the C-MXene@PI composite foams

Full size image

The interfacial interactions between MXene and PI as well as the flexible and robust PI scaffolds endowed the C-MXene@PI composite foams with excellent mechanical flexibility including bendability, rollability, twistability, and even foldability (Fig. 2a). Even though soaked in the liquid nitrogen at an extreme temperature of -196 ℃, the ultra-flexibility of the C-MXene@PI composite foam was still maintained (Fig. 2b, Video S1). In contrast, the commercial PU foams upon bending broke easily in this low-temperature condition (Fig. S3, Video S2), proving the significance of PI scaffolds on our flexible and reliable composite foams especially for extreme conditions. Furthermore, the chemical crosslinking of MXene flakes increased the water contact angle from 0° to 118° (Figs. 2c and S4a, Video S3), which was mainly ascribed to the introduction of the hydrophobic backbone in the PMDI [57]. As displayed in FTIR curves (Figs. 2d and S5), the characteristic peaks of polyimide (1720, 1780 cm−1 for C=O and 1380 cm−1 for C-N) presented in both pure polyimide and MXene-coated polyimide foams (Fig. 2d). After incubated with chemical crosslinker PMDI, the emblematic bands (around 1410 and 1504 cm−1) of benzene ring in PMDI can be observed on the spectra of C-MXene@PI (Fig. S5a). The new appearance of characteristic benzene ring vibrations and a CO–NH mode (1700 cm−1 for C=O in the urethane bonding, Fig. S5b) in C-MXene@PI showed the successful chemical coating of PMDI on the MXene. Consequently, such hydrophobic coating led to excellent stability and water resistance of the MXenes composites. After ultrasonic treatment for 20 min of the MXene@PI and C-MXene@PI composite foams were immersed in water, the former deteriorated completely, while the latter stayed as one piece, and there was no MXene detachment (Figs. 2e and S4b).

Fig. 2
figure 2

Photographs of a the C-MXene@PI composite foams showing ultra-flexible performance including bendability, rollability, twistability, foldability at room temperature and b the composite soaked in liquid nitrogen (LN, -196 °C) with maintained mechanical flexibility. c Contact angles of the MXene@PI and C-MXene@PI composite foams. d FTIR curves of PI, MXene@PI, and C-MXene@PI foams. e Photograph of C-MXene@PI foams with several drops of water on the surface, as well as the MXene@PI and C-MXene@PI foams after ultrasonic treatment in water for 30 min, showing that the C-MXene@PI foams are waterproof. f XRD patterns of the PI and C-MXene@PI foams. g X-band EMI SE of the PI, MXene@PI, and C-MXene@PI foams. X-band EMI SE of the h MXene@PI and i C-MXene@PI composite foams after stored in a 95% RH environment and a temperature of 60 °C for different days and j the corresponding change of SE as a function of time. In the first day (d = 0), the foams were in a dry state

Full size image

3.2 EMI Shielding Performance of the Composite Foam

In addition to the significant color change of the PI scaffolds upon the coating of MXene nanoflakes (Fig. S6), the 2-theta angle of ~ 7.22° in XRD pattern corresponding to the interlayer gaps of 1.2 nm between the MXene nanoflakes further shows a well-preserved structure of MXene flakes (Fig. 2f), which indicates a high utilization efficiency of MXenes’ conductivity and EMI shielding properties [22]. The coated MXene flakes on PI skeleton led to a remarkable increase in X-band EMI SE from 1.4 to 60 dB for the MXene@PI composite foams at a thickness of 1.5 mm. It is worth noting that EMI SE describes the attenuation capability of samples to the incident EM waves (Table S2), and an EMI SE value of 20 dB corresponding to a 99% attenuation of the waves is generally required for commercial applications [13, 19, 22]. Apart from the satisfactory EMI SE, our composite foams showed larger SE values than other PI-based composites (Table 1) and typical conductive nanomaterial embedded porous composites ever reported at similar thicknesses [58], e.g., 5-mm-thick AgNW/PI [59], 2-mm-thick CNT/PI [60], 2.3-mm-thick CNT/PU [16], 2.3-mm-thick graphene/PEI [61] composite foams reached SE values up to 12.8, 41.1, 50.5, and 12.8 dB, respectively. Chemical crosslinking of MXene flakes shown ignorable influence on the electrical conductivity and EMI SE properties (Figs. 2g and S7a), demonstrating the potentials of our C-MXene@PI composite foams as high-performance EMI shielding materials. More intriguingly, the chemical crosslink efficiently improved the oxidation stability of the C-MXene@PI composites in H2O/O2 environment (Fig. 2h–i), which is crucial for long durability of the composites in practical applications. After being stored in a 95% RH environment and a temperature of 60 ℃ for 2 days, EMI SE of the MXene@PI composite foams remarkably decreased and the shielding effect almost disappeared after 6 days in the same condition. In contrast, the C-MXene@PI composite foams maintained a high EMI SE of 44.4 dB after being stored in the same condition for 6 days (Fig. 2j). The behavior for resistance change of the composite foam stored in such condition is consistent with that of EMI SE (Fig. S7b). Therefore, stable, durable C-MXene@PI composite foams with a remarkable EMI SE were achieved successfully.

In the facile and scalable “layer-by-layer” dip-coating approach, EMI SE of C-MXene@PI composite foams can be widely controlled by adjusting the coating layers/times with MXene suspensions (Fig. 3a). The EMI SE of the 1.5-mm-thick porous composites increased with increasing coating layers, e.g., EMI SE reached a commercial SE value after a 4 times coating (namely 4 Layers MXene, 4L), while it increased to 41 and 60 dB after 10 and 16 times coatings, respectively. Nevertheless, when the composites were coated by 18 (18L) and more times, the increase of EMI SE reached a plateau. To better realize the behavior, we concluded the density and electrical conductivity of the C-MXene@PI composite foams as a function of coating layer (Fig. 3b). The density increased with the increasing of the coating layers, which correlates to the increased MXene loading and related thickness on PI scaffolds. According to the MXene-PDMI layer thickness (around 0.8 μm) identified from the SEM image of a 14L C-MXene/PI composite foam (Fig. S2b), we can easily calculate the MXene-PDMI thickness on the PI skeletons based on the measured density of the composite foams. Furthermore, a clear observation in Fig. 3a, 1L C-MXene@PI composite foams can already form sufficient conductive paths due to the efficient adhesion of MXene on the interconnected PI cell walls. This led to the rapid transformation from insulation to conduction of the porous scaffolds. With increasing of coating layers, more MXene nanoflakes were stacked on the cell walls, forming improved conductive networks, which eventually affected the electrical conductivity and EMI SE of the composites. However, the changes of electrical conductivity, as well as EMI SE, were not remarkably, especially above 10L coatings, which indicates good compatibility between MXene and PI fibers, and abundant intact conductive paths have been well established during the initial several coating rounds. Additionally, the EMI SE was controlled by adjusting the thickness of the C-MXene@PI composite foams, e.g., 16L composites reached an EMI SE value of 43.7 to 80.8 dB at a thickness of 0.5 to 3.0 mm (Fig. 3c). Moreover, the porous composites showed excellent resistance and EMI SE stability upon mechanical deformation, the resistance and EMI SE remained almost constant after the sample was bent for 1000 cycles (Figs. 3d and S7c). Briefly, the controllable and stable EMI shielding performance further demonstrates the great promises of our C-MXene@PI foams for practical applications.

Fig. 3
figure 3

source reference used in this figure are summarized in Table S1)

a X-band EMI SE of C-MXene@PI composite foams with various dip-coating layers. b Density and electrical conductivity of the C-MXene@PI composite foams as a function of the dip-coating layers. c The 16L C-MXene@PI composite foams at various thicknesses, and d the 10L and 16L C-MXene@PI composite foams before and after 1000 times bending treatment, showing the good shielding stability against mechanical deformations. e EMI shielding performance (SET, SEA, and SER) and SSE of C-MXene@PI composite foams as function of sample density. f Theoretically calculated EMI shielding performance (Th-SET, Th-SEA, and Th-SER) of a bulk based on the electrical conductivity of corresponding C-MXene@PI foams and the comparison to the experimentally tested EMI shielding performance (SET, SEA, and SER) of the C-MXene@PI composite foams. g Schematic showing the proposed EMI shielding mechanism of the C-MXene@PI composite foams for ultrahigh EMI shielding performance. h EMI shielding performance comparison of the C-MXene@PI composite foams with other typical porous architectures: SSE/d values of typical EMI shielding macrostructures with corresponding SE values (the experimental data and

Full size image

In order to realize the EMI shielding mechanism of the C-MXene@PI composite foams, we measured the shielding by absorption (SEA) and reflection (SER) (Fig. 3e). Generally, EMI shielding performance of conductive porous composites is influenced by the reflection, absorption, and multi-reflections, corresponding to the mobile charge carriers, electric dipoles, and interior interfaces/surfaces, respectively [19, 56, 62]. The micrometer-sized pores induced more reflections or scatterings of incident EM waves, which had more interactions with the pore walls in the C-MXene@PI composite foams, efficiently increasing SEA [16,17,18, 63]. The large mismatch of conductivity in the interfaces between the MXene and PI also led to high interfacial polarization [8], which combined with the abundant charge carriers from MXenes [23, 63, 64], resulting in increased SEA of the pore walls. In addition, the MXene terminal functional groups were considered to give rise to electric dipoles under the electric field of the EM wave [17, 22, 26], improving the SEA of the MXene-based composites. Thus, the SEA dominated the total SE (SET), which sum up both SER and SEA. With increasing density of the C-MXene@PI composite foams derived from increased MXene loadings, SET and SEA of the composites thus increased significantly and achieved maximum values of 62.5 and 54.9 dB, respectively, at a density of 48.7 mg cm−3. Furthermore, we had theoretically calculated the EMI shielding performance (Th-SET, Th-SEA, and Th-SER) (the details of the theoretical calculation method are shown in our previous work [13, 16]) of a homogenous shield based on the conductivities derived from our C-MXene@PI composite foams (Fig. 3f). Obviously, apart from the similar SER values, Th-SEA was obviously lower than the experimentally tested SEA. This was attributed to that the introduced multi-reflections caused by the porous structure gave rise to more interactions of the incident waves with the MXene/PI composite cell walls, which effectively absorbed the EM waves derived from the conduction and polarization loss capability. As a consequent, the experimentally tested SET was much higher than the Th-SET. In short, we could efficiently demonstrate that the high EMI shielding performance of our C-MXene@PI composite foams is attributed to the synergistic interactions among MXene, PI skeleton and the porous structure (Fig. 3g).

To better realize the lightweight EMI shielding architectures, we calculated the SSE of the C-MXene@PI composite foams with various densities (Fig. 3e). Interestingly, the SSE initially had a significant increase with increasing density, and it reached the extremum of 1,442 dB cm3 g−1 at a density of 41 mg cm−3, and a further increase in density led to the drop of SSE. This illustrated that suitable coating times of the MXene nanoflakes on the PI skeleton were vital for better utilizing the MXene for the EMI shielding porous architectures. Notably, the EMI SE reached a value of 59.2 dB at such a large SSE for the C-MXene@PI composite foams, which showed opposite behavior with other composites having a higher SSE value with decreasing density or SE values [5, 16, 18]. Here, the efficient design of our C-MXene@PI composite foams contributed both higher SE and SSE values, which significantly outperformed that of other typical nanofiller embedded polymeric porous composites at similar thickness (Table 1), e.g., EMI SE was around 22, 41, and 23 dB for the graphene/PI [65], CNT/PI [58], and CNT/PU [16] porous composites at SSE values of 78.6, 1280, and 1184 dB cm3 g−1, respectively. Certainly, a higher SE associated with a minimum material consumption is crucial for achieving high-performance EMI shielding architectures, and thus, SSE/d was proposed in our previous work [13, 16, 53] to efficiently evaluate the lightweight EMI shields. Herein, the SSE/d of C-MXene@PI composite foams could reach 9,612 and 21,317 dB cm2 g−1 at a SE of 59.2 and 43.7 dB, respectively. This performance was superior to that of other porous EMI materials including commercial carbon foams, carbon nanotube sponges, graphene-based foams, metal-based foams, and other MXene-based porous composites (Fig. 3h, Table S1). Particularly, the EMI shielding performance among the PI-based architectures is compared in Table 1, effectively proving the superiority of EMI performance of C-MXene@PI composite foams. Combined with the service stability, as well as the energy-efficient, facile, and scalable preparation method, the lightweight, ultra-flexible, and robust C-MXene@PI composite foams with outstanding EMI shielding performance show huge potentials in the applications of aerospace, portable electronics and smart wearable devices.

3.3 Electrothermal Performance of the Composite Foam

Besides the EMI application, multifunctionality is always expected for such lightweight conductors. First extension was on electrothermal materials, which convert electric power to thermal energy have attracted more and more attention due to the rapid development of electronic engineering [47, 48, 51, 52]. Traditional electrothermal materials, including Fe–Cr–Al or Ni–Cr-based alloys and electrothermal ceramics, suffer from complicated manufacturing process, heavy weight, inflexible shape and low heating efficiency. Here, thermogravimetric analysis (TGA) curves of pure PI and C-MXene@PI foams showed the good thermal stability of the PI-based architectures (Fig. 4a), in contrast to most polymers. Combined with the efficient conductive networks, the ultra-flexible C-MXene@PI composite foams showed great potentials for lightweight and stable electrothermal heaters. Therefore, various DC voltages were applied to the C-MXene@PI composite foams (4 × 4 × 0.15 cm3), and the currents flowing through the foams were observed. Figure 4b showed the effect of these low DC voltages on the temperature of the 10L C-MXene@PI composite foams. The composites had evident, stable, and reversible electrothermal effect at these voltages. For example, the composite foams with a large size of 40 × 40 × 1.5 mm3 were able to reach up to 34, 53, 84, and 114 °C in tens of seconds at 4, 6, 8, and 10 V, respectively. At the same time, the life-time tests for the composite heaters had been carried out. As shown in Fig. 4c, a DC voltage of 6 V was applied to the composite heaters for 24 h, and the temperature was measured every half hour. It was found that the steady-state heating of the samples, which further demonstrated a stable electrothermal effect of the composite. A stable and remarkable electrothermal performance of the C-MXene@PI composite foams was achieved at low DC voltages, demonstrating the possibility and prospects in various application areas.

Fig. 4
figure 4

Electrothermal performance of the C-MXene@PI composite foams. a TGA curves of the C-MXene@PI composite and pure PI foams in air, b the electrothermal curves of the C-MXene@PI composite foams (4 × 4 × 0.15 cm3) at various voltages, and c life-time tests of the C-MXene@PI composite foams for 24 h. d The steady-state temperature versus input power density curves of the C-MXene@PI composite foams and commercial Fe–Cr–Al and Ni–Cr alloy plates. Thermal image map of the e C-MXene@PI composite foams and f commercial metal-based heaters, showing the uniform temperature of the C-MXene@PI composite foams. Electrothermal deicing performance. Demonstration for a deicing application of the C-MXene@PI composites: g the composite without applied voltage and h the composite at an applied voltage of 8 V

Full size image

The equilibrium temperature of the composites showed a linear relation with the input power densities (Fig. 4d). The slope of the steady-state temperature versus input power density revealed the heat performance (Hp = dT/dP) of the composite heater [53, 56], as the greater the Hp value, the higher the steady-state temperature of the heater, at certain initial temperature and input electric power. The Hp value of the C-MXene@PI composite samples was 253.1 °C cm2 W−1, higher than that of other systems including CNTs, rGOs and metal thin films, as shown in Table 2. In addition, the composite foam heaters reached higher temperature than the typical commercial metal electrothermal heater Fe–Cr–Al and Ni–Cr alloy plates at the same input power density (Fig. 4e, f). In other words, the input power required for the composite foams was less than that of the commercial alloy-based heaters in order to reach the same temperature with the same heater area. More intriguingly, arising from the uniform distribution of MXene on PI skeleton, the lightweight C-MXene@PI composite foam heaters had an even temperature distribution, which could not be achieved by the heavy commercial heaters composed of metal alloy strips which easily had local overheating (Fig. 4e, f). Based on the high electrothermal performance, a demonstration for a high-performance deicing or anti-icing application at low voltages is well shown (Fig. 3g, h). Combining the excellent flexibility, high heat performance, low working voltage and uniform temperature distribution under 12 V in the range of automobile power supply, and the robust behavior demonstrated by the life-time test, the C-MXene@PI composite foam heaters show extended prospects in practical applications and can be produced on large scale for health-care-related products. The combination of the excellent MI shielding and electrothermal performance further efficiently demonstrates the great application potentials of the C-MXene@PI composites as multifunctional devices.

Table 2 Comparison of Hp of typical heaters
Full size table

3.4 Electromechanical Sensing Performance of the Composite Foam

Since our C-MXene@PI composite foams are highly flexible and robust, as the second extension, they can be effortlessly attached to human body as wearable sensors for detecting the human motions [43,44,45]. As the three-dimensional (3D) composite samples were bent, stretching and compressing were caused in different sides (Fig. 5a). The stretched and compressed side of the 3D composite sensors corresponds to the formation of less and more conductive paths in the C-MXene@PI composites, respectively, which correspond to the increased and decreased electrical resistance, respectively (Fig. 5b–d). The gauge factor corresponding to the sensitivity of the foam sensors is also calculated based on the bending angle (Fig. S8), and we can conclude that the bending-induced stretching leads to a high sensitivity. When the composites bent to an angel and then kept steady, the resistance of the sensor initially changed due to the deformation and then kept stable. Moreover, a larger relative resistance change is observed for a larger bending angle, showing the usefulness of our sensors in the detection of various human motions, from small strain to large strain activities. Additionally, even the bending speed of the sample was very fast, the change of the resistance could be well detected and distinguished. Therefore, the detected relatively resistance change’s signal of the sensor attached on a finger bending circularly can be displayed. When the finger bent to a fixed angel, then returned periodically, the resistance of the C-MXene@PI composite foam sensor increased and recovered accordingly and periodically (Fig. 5e, f), ascribing to the stretching caused by the bend of the composite foams. In contrast, when the composite foam was attached on the wrist bending circularly, the resistance decreased and recovered periodically due to the compression caused by the bend of the composite foams (Fig. 5g, h). According to the measured electrical signals from the composite sensor, we can easily deduce that in the testing period, the finger or wrist repeatedly bends and moves back quickly with similar amplitude, and number and of bends in each interval.

Fig. 5
figure 5

Detections of human motions for the flexible C-MXene@PI composite foams as electromechanical strain sensors. a Bending-induced stretch and compression modes of the C-MXene@PI composite foams. b, c The resistance changes (ΔR/R = (R1-R)/R, where R1 and R are the resistance and initial resistance of the sample, resistance) of the C-MXene@PI composite foam upon various bending angle induced stretch. d The resistance changes of the C-MXene@PI composite foam upon various bending angle induced compress. e, f Detections of finger bending activities with cyclic bending which induces resistance change of the C-MXene@PI composite foam. g, h Detections of wrist bending activities with cyclic bending which induces resistance change of the C-MXene@PI composite foam

Full size image

4 Conclusion

Lightweight, ultra-flexible, and large-area C-MXene@PI composite foams were prepared in a facile and scalable dip-coating followed by a chemical crosslinking approach. The coactions of the highly porous yet robust PI scaffolds, coated MXene nanoflakes, and the effective chemical crosslink treatment rendered the C-MXene@PI foams with hydrophobicity, waterproof capability, anti-oxidation and extreme-temperature stability. The combination also benefits high utilization of the MXene electrical conductivity, and an interfacial polarization between the MXene and PI. As-prepared C-MXene@PI foams show an ultrahigh EMI shielding performance accompanied by excellent durability and reliability. The 1.5-mm-thick C-MXene@PI composite foams exhibit a satisfactory X-band EMI SE of 22.5 to 62.5 dB at a density of 28.7 to 48.7 mg cm−3. Excellent SSE and SSE/d values of 1,971 dB cm3 g−1 and 21,317 dB cm2 g−1 are achieved, respectively, for the C-MXene@PI foams, significantly surpassing other typical porous architectures. Furthermore, two possible extensions of the C-MXene@PI foams were exanimated. A rapid reproducible, and stable electrothermal effect of C-MXene@PI foams at low voltages is demonstrated, which shows ultrahigh heat performance and uniform heat distribution. Secondly, the excellent performance of the C-MXene@PI foams as flexible wearable sensors is well demonstrated with sensitive and reliable detection capability of human motions. The integration of ultrahigh EMI shielding, electrothermal, and electromechanical sensing performances of the C-MXene@PI foams as well as their facile and scalable production approach suggest the promising application potentials for next-generation flexible and multifunctional electronic devices and aerospace.