Ultrafast universal fabrication of configurable porous silicone-based elastomers by Joule heating chemistry

Significance

Silicone-based elastomers (SEs) carry wide applications owing to their stable Si−O backbone with configurable functional groups. However, the traditional manufactures of SEs are either not very effective or excessively environmentally harmful. In this work, a universal and green Joule heating strategy is developed to rapidly yield SEs at programmable structures. Intriguingly, the cross-linking of SEs can be easily customized, which represents a key advance in the field of rapid polymerization of polymers. Consequently, an array of nanocomposites containing inorganic nonmetals, metals, metal compounds, and organics are in situ intercalated within the SEs matrix to produce diverse functional SEs, demonstrating their appealing utility in polymer-based material synthesis and wider applications.

Abstract

Silicone-based elastomers (SEs) have been extensively applied in numerous cutting-edge areas, including flexible electronics, biomedicine, 5G smart devices, mechanics, optics, soft robotics, etc. However, traditional strategies for the synthesis of polymer elastomers, such as bulk polymerization, suspension polymerization, solution polymerization, and emulsion polymerization, are inevitably restricted by long-time usage, organic solvent additives, high energy consumption, and environmental pollution. Here, we propose a Joule heating chemistry method for ultrafast universal fabrication of SEs with configurable porous structures and tunable components (e.g., graphene, Ag, graphene oxide, TiO₂, ZnO, Fe₃O₄, V₂O₅, MoS₂, BN, g-C₃N₄, BaCO₃, CuI, BaTiO₃, polyvinylidene fluoride, cellulose, styrene-butadiene rubber, montmorillonite, and EuDySrAlSiO_x) within seconds by only employing H₂O as the solvent. The intrinsic dynamics of the in situ polymerization and porosity creation of these SEs have been widely investigated. Notably, a flexible capacitive sensor made from as-fabricated silicone-based elastomers exhibits a wide pressure range, fast responses, long-term durability, extreme operating temperatures, and outstanding applicability in various media, and a wireless human–machine interaction system used for rescue activities in extreme conditions is established, which paves the way for more polymer-based material synthesis and wider applications.

Elastomers, with their unique flexibility and versatility, have emerged as modern materials with broad applications in cutting-edge technologies, including mechanics (1, 2), optics (3, 4), stretchable electronics (5, 6), and soft robotics (7). Among them, silicone-based elastomers (SEs), consisting of a stable Si−O−Si backbone with programmable side groups and terminal groups (8, 9), have attracted considerable interest in wearable electronics (10–12), health care (13), shock-absorbing elasticity (14, 15), and thermal insulation (16) due to their desirable characteristics, such as thermal stability, electrical insulation, biocompatibility, elasticity, and machinability (17–19). Despite the advantages of SEs, their vast potential for flexible applications and inorganic functionality is constrained by their compact structure and limited inherent organic polymer characteristics (20–22). Porous SEs (PSEs) have attracted considerable interest because of their abundant pores and functional components, which endow SEs with light weight, breathability, sufficient contact sites, excellent deformability, and multifunctionality (23–25). However, manufacturing specific structures to extend more advantages and compounding functional materials with pure SEs to enable new functionalities present significant challenges (26–28).

The main approaches previously reported for synthesizing PSEs are focused on conventional, slow polymerization methods, such as oven or stage heating with pore templates (29–31). These methods can produce hierarchical porous products but suffer from uneven pore distribution (32), time- and energy-consuming strategies due to the long reaction time required at high temperatures, and the reaction environment absorbing much power which results in energy loss (33, 34). Although new synthetic strategies, such as microwave (MW) irradiation (35), UV curing (36), and supercritical foaming (37), have been gradually developed and have been demonstrated as efficient preparation methods of PSEs, most of these routes still involve intricate steps and high costs. Regarding the structural engineering of PSEs, template-assisted methods and gas-generated methods are currently popular approaches for achieving this goal (33, 38). Nevertheless, the postprocessing required for template removal introduces hazardous by-products and can affect the intrinsic properties of SEs (39–41). Furthermore, the in situ decomposed gas templates, typically referred to as NH₃ or CO₂, are considered to generate environmental pollution and accelerate climate warming (42, 43). To date, MW technologies with small amounts of organic solvents or H₂O mixed with prepolymers for pore formation have been proven effective and environmentally friendly in constructing hierarchical PSEs (44, 45). However, an investigation into the mechanism of programmable preparation and the manufacture of configurable structures has not yet been conducted. To address this issue, more in situ reaction factors should be thoroughly evaluated to study the reaction mechanism and achieve effective fabrication. Thus, developing a scalable, one-step, efficient, and clean templating strategy that involves no harm to the environment for preparing high-quality PSEs with tunable composition is highly desirable.

To this end, this work employs a Joule heating chemistry strategy to fabricate a series of PSEs with variable components [graphene, Ag, graphene oxide (GO), TiO₂, ZnO, Fe₃O₄, V₂O₅, MoS₂, BN, g-C₃N₄, BaCO₃, CuI, BaTiO₃, polyvinylidene fluoride (PVDF), cellulose, styrene-butadiene rubber (SBR), montmorillonite, and EuDySrAlSiO_x] within seconds. Intense MW irradiation enables ultrafast thermal dynamics via bond vibration with an alternating magnetic field (46, 47), which leads to the catalytic curing of prepolymers into robust molecular networks while simultaneously synthesizing PSEs with ultrasmall holes by limiting H₂O movement during the rapid solidification process. The proposed Joule heating engineering was energy efficient, time saving, and environmentally friendly for the preparation of PSEs with ultraporous construction filled with 5-μm pores. In addition, owing to the superior capacitive pressure sensing performance and scenario-applicable characteristics, the PSEs can be made into capacitive mechanical sensors utilized for human motion monitoring and human−machine interactions for emergency rescue in extreme scenarios.

Results and Discussion

Ultrafast Universal Fabrication of Configurable PSEs.

A synthetic strategy for ultrafast universal fabrication of programmable and commercial PSEs is proposed, and its schematic is depicted in Fig. 1. Generally, the variable components were dispersed homogenously and loaded into uncured prepolymers consisting of vinyl-terminated siloxane and Si−H moieties to form a uniform hybrid gel. Typically, H₂O is the only eco-friendly spongy template used in this method instead of adding an environmentally harmful organic blowing agent, which was stirred with the hybrid gel for the formation of a highly phase-separated gel system. Subsequently, the phase-separated gel system was processed with MW irradiation, leading to the Joule heating chemistry that acted as a field-assisted heating or sintering induced by the coordinated vibration or movement of polar molecules or charged particles under high-frequency alternating electric fields (48, 49). The excitation of H₂O and the loading components thus synergistically produced intense heat (50). During the thermodynamics of Joule heating chemistry, the C=C and Si−H in the substrate and the cross-linker are facilitated to assemble into Si−C−C bonds within seconds to build a robust polymer network, which is successfully compounded with embedded variable components for reliable polymer–component bonds. Moreover, the extra Si−H bonds transform into Si−OH bonds, which undergo intramolecular dehydration to create Si−O−Si in the polymer networks, rendering the formation of PSEs with variable components. Thus, the established polymer networks result in limited movement and gathering of H₂O, yielding cured PSEs with a controllable pore size. Finally, the elastomer was dried for 24 h to fully eliminate the H₂O for practical applications.

Fig. 1.

Schematics illustrating the preparation and synthetic mechanism of PSEs via the ultrafast Joule heating chemistry.

By manipulating the mass ratio of H₂O/SE at 0%, 50%, 100%, 150%, and 200% (defined as PSE-0, PSE-50, PSE-100, PSE-150, and PSE-200), the morphology and structure of the obtained PSEs were identified through high-resolution field-emission scanning electron microscopy (FESEM). Pore structures of PSEs were confirmed in manageable evolution (SI Appendix, Fig. S1), suggesting the feasibility of the proposed route to fabricate structurally controllable PSEs. Significantly, PSE-200 exhibits a high level of porosity due to the presence of numerous small spherical pores of approximately 5 μm in diameter. These pores are considerably smaller than those described in previous studies (33, 38, 44). Furthermore, the utilization of the Joule heating chemistry approach can be extended to incorporate additive manufacturing technology. The obtained hybrid gel, which was in a 200% mass ratio of H₂O/SE, was three dimensionally (3D) printed with “NPU” models through direct ink writing and then microwaved for 120 s. The infrared radiation (IR) photographs demonstrate the intense synthetic process of the printed gel (SI Appendix, Fig. S2A). The detected temperatures reached approximately 100 °C in 120 s, ensuring the rapid thermodynamic curing of the gel. The optical diagram implies a successful customized design of MW-cured PSEs in which the shape was maintained compared with the initial state (SI Appendix, Fig. S2B). Impressively, compared with traditional pore-forming templates and preparation methods for PSEs, such as using PS balls, sugar, salt, NaHCO₃, etc., and curing at high temperature for hours or dozens of hours (SI Appendix, Table S1), Joule heating chemistry for preparing PSEs was demonstrated to be time saving, efficient, and environmentally friendly. SI Appendix, Fig. S2 C and D, shows the internal structure of the 3D-printed PSE-200, which was also confirmed to be ultraporous with ultrasmall pores of 5 μm. In addition, PSE foams with designable shapes and sizes are capable of being obtained in various models (SI Appendix, Fig. S2E), ensuring Joule heating technologies as a scalable and manageable approach to PSE preparation. To evaluate the mechanical properties of as-synthesized PSE-200, the elastomers were pressed (~80% compressive strain, SI Appendix, Fig. S2F and Movie S1) and stretched (~167% tension strain, SI Appendix, Fig. S2G and Movie S2), which can perfectly recover back to their initial state, exhibiting extraordinary flexibility and recoverability of the PSEs. As such, it was demonstrated that fast, efficient, and environmentally friendly Joule heating chemistry could be applicable to produce PSEs with a variety of structural specializations.

To investigate the Joule heating chemistry mechanism, a series of studies were conducted to evaluate the influencing factors. First, various heat treatment routes for curing PSE with H₂O in a 200% H₂O/SE mass ratio were compared, including traditional direct heating at different temperatures (5, 25, 60, 80, 100, and 120 °C). The cured PSEs treated by conventional heating were all in porous inner structures due to the introduction of H₂O for pore construction (SI Appendix, Fig. S3). SI Appendix, Fig. S4, depicts the pore size of the as-obtained PSEs, which decreased significantly from 40 to 10 μm between 5 and 80 °C and then remained stable at approximately 6 μm from 80 to 120 °C. This result indicated that high-temperature curing can effectively limit water movement by promoting the curing of PSE and reducing the accumulation of H₂O, thus resulting in a smaller pore size. Joule heating chemistry technology was demonstrated to enable the reaction temperature to reach 98 °C in just 2 min (SI Appendix, Fig. S5), suggesting extraordinary contributions for ultrasmall pore sizes. In addition, the special heat transport mode of MW irradiation was also revealed. SI Appendix, Fig. S6, establishes an overall uniform pore distribution by Joule heating, while a hierarchical structure (SI Appendix, Fig. S7A) with ball-like pores inside (SI Appendix, Fig. S7 B and C) and irregular wedge-shaped channels outside (SI Appendix, Fig. S7 D and E) was obtained via the 100 °C oven heating method, suggesting a uniform heat transport mode of Joule heating inside and nonuniform conventional heating from outside to inside. Therefore, it has been established that the application of Joule heating chemistry is viable for the effective synthesis of PSEs, resulting in a more homogeneous and smaller pore structure in contrast to conventional preparation techniques.

Furthermore, we evaluated the impact of H₂O content on the controllable pore structures during Joule heating chemistry by varying the mass ratios of H₂O/SE. In Fig. 2A, the pore-forming process is depicted, which encompasses two distinct stages: The first stage involves a continuous increase in pore size when the ratio is less than 100%, and the second stage involves an increase in pore number when the ratio is 100%. FESEM results essentially confirmed the efficacy of this strategy. The detailed average diameters and numbers of pores suggest that the pore size increased from 0 to 5.6 μm from 0 to 150% and decreased to 4.5 μm at 200%, while the numbers of pores in the unit area maintained significant increases (SI Appendix, Fig. S8 A–F). The reaction times for PSEs in 0%, 50%, 100%, 150%, and 200% mass ratios of H₂O/SE were measured to be 20, 8, 4, 3, and 2 min, respectively (Fig. 2B), implying an increasing curing speed. At a slow curing rate, PSE-50 exhibits macropores over 500 μm in the upper part and extremely small micropores below 2 μm in the middle part of the sample (SI Appendix, Fig. S9A), resulting from the relatively fast movement and aggregation of H₂O on the top (SI Appendix, Fig. S9B). In contrast, when the H₂O content was increased to 200%, the curing speed was much faster, and the H₂O movement was more difficult, restricting the H₂O movement to form pores with a larger diameter and a smaller number. Thereby, the pore formation was dependent on the curing speed relative to the H₂O moving speed, causing limited H₂O movements during the liquid−solid conversion process. SI Appendix, Fig. S10, shows the viscosities of the uncured gels obtained after adding different masses of H₂O. This illustrates the increased viscosity and decreased fluidity of gels with the addition of H₂O, which are caused by the heavier phase separation between the H₂O and prepolymers, making it more difficult for H₂O to move. It is worth mentioning that the presence of H₂O was shown to have a substantial impact on increasing reaction temperatures, as the temperatures reached 53.6, 57.4, 66.8, 88.8, and 98.4 °C within 2 min for 0%, 50%, 100%, 150%, and 200% mass ratios, respectively. In contrast to the temperature of an empty container, which generated little heat from 21.2 to 33.0 °C in 8 min (SI Appendix, Fig. S11), the prepolymer chains of PSE vibrated with the alternating electric field and generated heat by friction, demonstrating a distinctive MW response. Furthermore, we investigated the influence of MWs on the synthesis of PSE by Fourier transform infrared (FTIR) analysis. For the SE synthesized by MW in the absence of H₂O, there are two distinct peaks assigned to Si−O bonds located at 1,062 and 1,051 cm⁻¹ (SI Appendix, Fig. S12), proving the intrinsic self-assembly reaction in the MW environment. However, for the PSE prepared by Joule heating technology with MWs and H₂O, only a peak assigned to Si−O bonds is located at 1,055 cm⁻¹ (51), verifying the unique formation of Si−O bonds induced by Joule heating chemistry. This suggests that H₂O not only plays the role of a pore template but also, perhaps more importantly, serves as a MW absorbing seed, rotating vigorously with the high-frequency alternating electric field and colliding with prepolymers of SE which have weak MW responses. As shown in the IR images of the uncured prepolymers in 200% H₂O at different reaction times (SI Appendix, Fig. S5B), a stable synthetic process occurred at approximately 98 °C from the second minute to the end without excessive temperature (52, 53). We treated the precursor for 1 min and repeated the process after it had cooled for 6 min, during which the recorded temperature sharply increased to 90 °C and then decreased to 40 °C (SI Appendix, Fig. S13). Therefore, H₂O is confirmed to be capable of quickly providing heat and maintaining temperature, revealing a smart, self-adjusted synthesis strategy for controllable preparations.

Fig. 2.

(A) The schematic (Upper) and the corresponding FESEM images (Bottom) of the structural control with different H₂O/SE mass ratios. (B) Reaction temperatures of PSEs prepared with different H₂O/SE mass ratios. (C) GPC analysis of reaction system in different reaction times. (D) Normalized FTIR curves of the reaction system during different reaction times. Detailed normalized FTIR curves of (E) Si−C bonds and (F) Si−O bonds. (G) Simulated temperature distribution of (I) conventional heating methods and (II) Joule heating chemistry of the reaction system at a 200% mass ratio of H₂O/SE.

Subsequently, a further study of the reaction mechanism during MW irradiation was carried out, in which the uncured gels with an H₂O/SE mass ratio of 200% were treated by varying the MW times (t = 0, 30, 60, 90, and 120 s), and denoted as MW0, MW30, MW60, MW90, and MW120, respectively. During the reaction, the viscosity of the gels at different times (t = 0, 30, and 60 s) was measured using a viscometer. The viscosity of MW60 increased with increasing shear rates, suggesting the production of macromolecules within 60 s due to significant cross-linking reactions (SI Appendix, Fig. S14). Gel permeation chromatography (GPC) was carried out to measure the relative molecular mass of the synthesized reactants at different times, with shorter retention times indicating larger molecular weights. Significant molecular weight increases were observed in MW60 at 5.2 min, with a relative molecular mass of 1.48 × 10¹⁰ g mol⁻¹, which was significantly larger than the 1.0 × 10⁷ and 2.9 × 10³ g mol⁻¹ found at MW0 (27.2 and 32.4 min) ascribed to the base material and curing agent (Fig. 2C). The results indicate that the flash polymerization process began within 30 to 60 s. Moreover, the GPC analysis showed that the relative molecular mass of MW90 was beyond the detection range, and species with low relative molecular masses were not detected, indicating that the molecules had completed polymerization before 90 s. Therefore, the entire curing process was verified to occur within 30 to 90 s.

FTIR analysis was used to understand the synthetic process of PSEs under MW radiation. Characteristic peaks of cured PSEs were detected compared with the base material and the curing agent of the prepolymers. There is a disappearance of Si−H bonds of the curing agent at 2,160 cm⁻¹ (SI Appendix, Fig. S15) and an appearance of C−C bonds of cured PSE at 544 cm⁻¹ (SI Appendix, Fig. S16A), and a slight shift from 791 to 788 cm⁻¹ for Si−C from the base material to PSE (SI Appendix, Fig. S16B) occurs, implying that Si−H bonds reacted with C=C bonds to form Si−C−C bonds. Meanwhile, an obvious appearance of Si−OH bonds at 909 cm⁻¹ (SI Appendix, Fig. S15) and a shift of Si−O bonds from 1,012 to 1,009 cm⁻¹ (SI Appendix, Fig. S16C) were detected, which originated from the oxidation process of Si−H to Si−OH and the dehydration reaction to form new Si−O−Si. Furthermore, the uncured gels exposed to different MW times were characterized to assess the instant progress of curing PSE, and the absorption spectrum was normalized with respect to the Si−CH₃ bonds at 2,960 cm⁻¹ (Fig. 2D). The normalized absorbance peaks of Si−O and Si−C bonds, which are located at 1,060 to 1,012 cm⁻¹ and 791 cm⁻¹, respectively, were detected and varied with reaction time. Fig. 2E and SI Appendix, Fig. S17A, display detailed Si−C bond peaks, where slight intensity changes and peak displacements were observed from MW0 to MW60, indicating that the initial 60 s was a heat production process without an obvious Si−C reaction. However, for MW90, a significant intensity increase and peak shift were detected compared to MW60, illustrating the formation of new Si−C bonds derived from the reaction between C=C bonds and Si−H bonds after 60 s. MW90 to MW120 did not show a noticeable difference, indicating that the polymerization of Si−C was completed before 90 s, which is consistent with the previous analysis. Therefore, the curing reaction of Si−C was confirmed to occur during 60 to 90 s. The detailed curing process of Si−O is exhibited in Fig. 2F and SI Appendix, Fig. S17B. The intensity of Si−O bonds increased significantly over time due to the continuous oxidation process and dehydration reaction to form Si−O−Si bonds. The generation of Si−O bonds was mainly observed during 60−90 s, while the slightly increased Si−O intensity at MW120 resulted from the internal reaction of Si−OH in polymer networks to form Si−O−Si as GPC analysis confirmed that excess cross-linking agent reacted before 90 s.

Theoretical studies were conducted to simulate and investigate the Joule heating chemistry. First, we explored the heat transfer mechanisms of conventional heating methods and Joule heating chemistry. Compared with the traditional external hot–cold (bottom–up) heating method, Joule heating chemistry was confirmed to be a special heating method for internal uniform heat generation (Fig. 2G). The calculated temperatures of the reaction systems are depicted in SI Appendix, Fig. S18A, and it took 60 s and 2,400 s for Joule heating chemistry and conventional heating, respectively, to reach above 90 °C, indicating a faster heating effect of Joule heating chemistry. In detail, the conventional heating method produced 35 °C in 60 s, while Joule heating chemistry achieved 92 °C at the same time (SI Appendix, Fig. S18B), proving that the thermal effect of Joule heating chemistry was greater than that of conventional heating. The influencing factors of Joule heating chemistry were also investigated. As the H₂O content increased, the temperature of the reaction systems increased significantly after the same reaction time (SI Appendix, Fig. S19). The calculated temperatures of the reaction system at the mass ratio of H₂O/SE in 50%, 100%, 150%, and 200% were verified to reach 47, 73, 85, and 92 °C, respectively, at 60 s (SI Appendix, Fig. S20). To ensure the absorbing factor of the reaction system, the electric field distribution was studied. As shown in SI Appendix, Fig. S21A, H₂O was confirmed to be the main factor that could effectively absorb waves in the alternating electric field and produce excellent thermal effects, which was associated with a high electromagnetic power loss density. Impressively, obvious MW thermal responses were also found at 0.1 s and originated from evenly dispersed H₂O (SI Appendix, Fig. S21B), verifying the uniform spatial thermal response and heat transfer of the heating mode. In summary, the MW-facilitated cross-linking of Joule heating chemistry engineering was demonstrated to be rapid and effective compared to the conventional curing method (SI Appendix, Fig. S22), which took over 60, 40, and 30 min in 60, 80, and 100 °C ovens, respectively.

Customized Fabrication of Porous Silicone-Based Elastomer Composites.

Various porous SEs coordinated with different components (PSECs), including graphene, Ag, GO, TiO₂, ZnO, Fe₃O₄, V₂O₅, MoS₂, BN, g-C₃N₄, BaCO₃, CuI, BaTiO₃, PVDF, cellulose, SBR, montmorillonite, and EuDySrAlSiO_x (designated PSE–graphene, PSE–Ag, PSE–GO, PSE–TiO₂, PSE–ZnO, PSE–Fe₃O₄, PSE–V₂O₅, PSE–MoS₂, PSE–BN, PSE–g-C₃N₄, PSE–BaCO₃, PSE–CuI, PSE–BaTiO₃, PSE–PVDF, PSE–cellulose, PSE–SBR, PSE–montmorillonite, and PSE–EuDySrAlSiO_x, respectively) on behalf of pure elements, oxides, sulfides, nitrides, inorganic compounds, organic compounds, and other functional materials, were universally fabricated within one step via the MW synthesis strategy in seconds, as depicted in Fig. 3 A1–R1. Characterizations of the as-prepared PSECs were carried out to verify their morphological structures and chemical compositions. FESEM was employed to characterize the inner structure of the PSECs. As shown in Fig. 3 A2–R2, numerous microscopic spherical cavities were formed inside the cured colloid due to the phase separation between the large amount of mixed H₂O and the prepolymers during cross-linking. Furthermore, the pore sizes of PSECs (SI Appendix, Fig. S23 A–R) were determined to be around 5 to 200 μm due to the various mass ratios of H₂O and SE, indicating a universal approach to producing PSECs. The strategy for preparing diverse PSECs is remarkably simple, rapid, universal, and environmentally friendly, making it a prospective manufacturing route for customized functional designs.

Fig. 3.

Optical pictures and FESEM images of as-fabricated PSECs: (A1 and A2) PSE–graphene, (B1 and B2) PSE–Ag, (C1 and C2) PSE–GO, (D1 and D2) PSE–TiO₂, (E1 and E2) PSE–ZnO, (F1 and F2) PSE–Fe₃O₄, (G1 and G2) PSE–V₂O₅, (H1 and H2) PSE–MoS₂, (I1 and I2) PSE–BN, (J1 and J2) PSE–g-C₃N₄, (K1 and K2) PSE–BaCO₃, (L1 and L2) PSE–CuI, (M1 and M2) PSE–BaTiO₃, (N1 and N2) PSE–PVDF, (O1 and O2) PSE–cellulose, (P1 and P2) PSE–SBR, (Q1 and Q2) PSE–montmorillonite, and (R1 and R2) PSE–EuDySrAlSiO_x.

X-ray diffraction (XRD) patterns of pure PSE and PSECs were evaluated (SI Appendix, Fig. S24). Pure PSE displays a single broad peak centered at 22°, while PSECs exhibit distinct characteristic peaks (SI Appendix, Fig. S24), respectively. This indicated that multiple materials have been successfully incorporated into the curing process of prepolymers. The FTIR spectra of the as-synthesized PSECs are illustrated in SI Appendix, Fig. S25A. The characteristic absorption bands of pure PSE at approximately 2,961, 1,078, and 1,016 cm⁻¹ are attributed to the symmetrical deformation vibration of the Si−CH₃ bonds and the stretching vibration of the Si−O and Si−C bonds, respectively. Furthermore, the PSECs displayed similar FTIR peaks to those of pure PSE, suggesting that the introduction of the variable components will not generate significant restrictions on prepolymers curing since the components were thoroughly dispersed in the prepolymers during the cross-linking process. SI Appendix, Fig. S25B, demonstrates similar one-peak Si−O bonds of PSECs, implying that the MW curing effect on Si−O bonds is universally identified compared to that depicted in SI Appendix, Fig. S12. Meanwhile, the interaction between embedded functional components and prepolymers, such as MoS₂ and V₂O₅, was also established. The FTIR spectrum of PSE–MoS₂ (SI Appendix, Fig. S26) displays the appearance of the S−O bond at 1,040 cm⁻¹ and the disappearance of the Si−OH bond at 910 cm⁻¹, demonstrating the formation of the S−O−Si bond via S bonding with Si−OH, thus achieving a polymer–component compounding PSE–MoS₂. Similarly, the FTIR spectrum of PSE–V₂O₅ shows the newly generated Si−O stretching vibration at 853 cm⁻¹ and a significant decrease in Si−CH₂ at 755 cm⁻¹, implying the generation of Si−O−V bonds between V₂O₅ and prepolymer chains and the restriction of ethynyl reacting with Si−H to form Si−CH₂−C. Significantly, the FTIR spectra of PSE–MoS₂ and PSE–V₂O₅ reveal a noticeable decrease in the intensity of the −OH peak. This indicates a weaker hydrophilicity of the materials (54), thus leading to a reduction in the maximum mass ratio of H₂O/SE to 100%, which is lower than that of pure PSE of 200%. Therefore, it can be determined from Fig. 3 G2 and H2 that the pore diameters of these two PSECs are larger.

The thermal properties of PSECs were investigated by thermogravimetric analysis (TGA). Almost no mass loss is observed below 120 °C because the H₂O has already been removed during the drying process (SI Appendix, Fig. S27A), and significant mass loss is observed for pure PSE from 400 to 800 °C due to the pyrolysis of organic functional groups of PSEs (55) (SI Appendix, Fig. S27B). Notably, the thermogravimetric behaviors of PSECs with a small amount of compounding materials (mass proportion of 4.7%) during the pyrolysis process were significantly distinct from those of pure PSE, revealing the strong molecular interactions of polymer to components between siloxane chains and variable components. Stress−strain curves were employed to assess the mechanical characteristics of the prepared PSECs (SI Appendix, Fig. S28). The PSECs all exhibited excellent elasticity up to 80% compression and higher mechanical strength than pure PSE at the same strain, implying the significant role of polymer–component bonds in mechanical properties. Therefore, it has been established that compounding materials effectively participate in the formation of PSECs by forming polymer–component bonds, enabling the programmable design of molecular structures and the regulation of properties.

Sensing Performance of the Flexible Capacitive Mechanical Sensors.

Owing to their soft, lightweight, and human-friendly characteristics, the variable MW-prepared PSEs were employed as the flexible sensing layer of capacitive mechanical sensors, and the sensing behavior was evaluated. As illustrated in Fig. 4A, when subjected to external pressure, the inner structure of the sensing layer undergoes extrusion deformation, and the thickness of the sensing layer decreases, inducing a sharp increase in capacitance according to Eq. 1 (56):

$$C=\epsilon S/4\Pi \mathit{kd}\mathit{,}$$ [1]

Fig. 4.

(A) Schematic illustration of capacitive sensing mechanism and assembled capacitive mechanical sensors. (B) Capacitive sensing performance of PSEs obtained in different H₂O/SE mass ratios. (C and D) Capacitive responses of PSEs under different strains. (E) Capacitive responses under different frequencies. (F) Stability of capacitive responses. (G and H) Response times under different mass loadings. (I) Capacitive sensing performance under different temperatures. (J) Cycling performance of the capacitive mechanical sensor.

where ε refers to the dielectric constant, S represents the electrode area, k is the electrostatic constant, and d is the dielectric thickness. The sensor was assembled with two flexible conductive copper electrodes on top and bottom to ensure effective electron transmission, with the PSEs placed between the electrodes as the dielectric layer. Therefore, the mechanical sensor can be fabricated in a small size (1 cm × 1 cm × 1 mm) with extraordinary flexibility, as pressed and folded in the optical images (SI Appendix, Fig. S29). Classical tensile measurements were adopted to investigate the mechanical properties of the PSEs at different H₂O/SE mass ratios (SI Appendix, Fig. S30A). Compressive stress−strain curves showed that the mechanical strength decreases with increasing H₂O/SE mass ratio, as measured at 1,511, 988, 987, 652, and 285 kPa at 70% strain for the samples with mass ratios of 0%, 50%, 100%, 150%, and 200%, respectively. Specifically, PSE-50 and PSE-100 exhibited similar mechanical properties due to their similar pore structures, which matched well with the SEM images (SI Appendix, Fig. S1). In addition, PSE-200 has the lowest mechanical strength due to the largest porosity, achieving the best flexibility and compressibility. Furthermore, the capacitive strain sensing performances of PSEs with different H₂O/SE mass ratios are displayed in SI Appendix, Fig. S30B. The calculated ΔC/C₀ values are 32%, 62%, 72%, 111%, and 184% at 70% strain for the samples with mass ratios of 0%, 50%, 100%, 150%, and 200%, respectively, indicating that the sensing capability is positively correlated with the H₂O/SE mass ratio. The H₂O/SE mass ratio effectively enhanced porosity of 0%, 2%, 6%, 59%, and 66% (SI Appendix, Fig. S31), thus increasing deformation sites during compression. PSE-200 had the highest sensing performance for the same mechanical behavior. As a key parameter for pressure sensing, the sensitivity of capacitive mechanical sensors is defined as Eq. 2 (57):

$$S=(\mathrm{\Delta }C/C_0)/\mathrm{\Delta }P\mathit{,}$$ [2]

where S is represented by the capacitive sensitivity, C₀ refers to the initial capacitance before loading, ΔC is the change in capacitance with changeable pressure, and ΔP represents the change in pressure. From above, the pressure–capacitance curves of PSEs prepared in different H₂O mass ratios are provided in Fig. 4B, and the sensitivities are fitted in SI Appendix, Fig. S32. The sensitivities of PSE-200 were calculated to have significantly highest capacitive sensitivities of 0.08865 kPa⁻¹ and 0.04091 kPa⁻¹ for strain ranges from 0 to 20% and 20 to 80%, compared with PSEs fabricated with low H₂O mass ratios. Meanwhile, the pressure sensitivity decreases with increasing pressure, consistent with the reduction in pore size during the compression process (58). Compared with the sensing performance and preparation schemes of other PSEs and polymer-based porous elastomers (SI Appendix, Table S1), the PSE-200 obtained via Joule heating chemistry was confirmed to have great potential for pressure sensing and highly efficient preparation.

PSE-200 was employed as an excellent dielectric layer for capacitive mechanical sensors. Fig. 4 C and D presents six stretching–releasing cycles of the sensor at various strains (10%, 25%, 40%, 60%, and 70%) at a frequency of 0.5 Hz, showing fast real-time dynamic responses with excellent repeatability. Furthermore, the capacitive responses at 10% strain with different frequencies (0.1, 0.5, 1, and 2 Hz) are demonstrated in Fig. 4E, implying the competence of the sensors for different application scenarios. To investigate the response time and signal stability of the device, weights with different masses (2, 10, and 20 g) were instantly placed on the top of the sensor and then removed after retaining 10 s. The simultaneous capacitive signal outputs showed that the capacitance increased rapidly and maintained a stable level during compression and then recovered to the initial signal value after release (Fig. 4F), indicating a rapid response and steady signal output under different pressures. Furthermore, the response time and recovery time of the signals were measured as 230 ms for both the ascent and descent processes of the weights of 2, 10, and 20 g, as illustrated in Fig. 4 G and H, indicating the extremely fast and highly reversible capacitive response of the sensor. The thermal tolerance of the sensor shows that the static capacitance maintains stability over a wide operating temperature range (20 to 230 °C) with only a slight decline (SI Appendix, Fig. S33), attributed to the dielectric constant of SE decreasing insignificantly with temperature (59). In addition, the pressure-induced capacitive responses were investigated at different temperatures. Under various temperatures (15, 50, 150, and 230 °C), the device was employed to detect weights with different masses (2, 10, 20, 50, and 100 g), and the curves delivered similar pressure–capacitance values without apparent deviation, meaning excellent thermal tolerance with a wide applicable temperature range (Fig. 4I). To investigate the durability of the sensor, the signal output of the sensor was measured under over 15,000 stretching–releasing cycles (20% strain, 1 Hz), suggesting outstanding stability and durability without evident decay (Fig. 4J and SI Appendix, Fig. S34 A–C, for details).

Motion Monitoring and Human−Machine Interaction System for Diverse Application Scenarios.

Owing to the exceptional sensitivity and cycling stability, the capacitive mechanical sensor fabricated based on the flexible PSEs can be applied to human signal monitoring systems. Fig. 5A demonstrates that by directly attaching the sensor to the surface of the skin using commercial medical tape, the movement signals generated by the different parts of the body were detected via capacitive responses. Fig. 5B and SI Appendix, Figs. S35 and S36, exhibit the repeatable movements of the ankle, throat, finger, elbow, wrist, and foot, respectively, showing the delicate joint movements by real-time monitoring capacitive signals. Moreover, the frequency, amplitude, and waveform of ΔC/C₀−time curves were successfully applied to identify the evident characteristic response signals. Simultaneously, the device can also distinguish different extents of movement by monitoring finger bending at different angles. As a result, ΔC/C₀ presented a marked increase from 15 to 58% as the bending angle of the fingers increased from 30 to 90°. Similarly, the capacitive response to high pressure was evaluated by monitoring the forefoot pressure when sitting and standing by raising and lowering the forefoot. These results confirmed that the mechanical sensor can detect various kinematic performances, implying great potential for motion monitoring. Additionally, the thermal tolerance characteristic of the sensor makes it suitable for applications in extreme environments. Fig. 5C shows that the sensor can work effectively at an ambient temperature of 80 °C (SI Appendix, Fig. S37A), with a stable and prominent signal output. Similarly, the device is also able to detect signals at a low ambient temperature of −10 °C (SI Appendix, Fig. S37B). Additionally, owing to the chemical stability of the materials, the fabricated device can be utilized to sense stress signals in water (SI Appendix, Fig. S37C) and oil (SI Appendix, Fig. S37D). The characteristic signals of finger movements matched well with those collected in the air, indicating excellent compatibility with various applicable media.

Fig. 5.

(A) Schematic illustration of human motion signal monitoring. (B) Capacitive signals via monitoring ankle flexing, throat swallowing, wrist bending, and finger bending. (C) Capacitive signals of sensors in different application scenarios as in high temperature, in low temperature, in water, and in oil. (D) Schematic illustration of human−machine interactions and applications in extreme conditions as simulated fire and ice scenarios.

The remarkable performance of the PSE with satisfactory capacitive sensitivity makes it ideal for the application of human−machine interaction. Based on the capacitive mechanical sensors, a wireless human−machine interaction system containing a human signal monitoring system and a wireless manipulator system was designed (Fig. 5D and SI Appendix, Fig. S38), which enables the real-time monitoring of capacitive signals from sensors attached to fingers. The signals collected from fingers are processed through a capacitance-to-digital converter (PCAP01) and a microcontroller board (STM32) with a resolution of 0.001 pF and a set sampling interval of 300 ms, converted into signals recognizable by the manipulator, and further wirelessly interconnected with the manipulator through Bluetooth. The system can instantly identify and capture the signals of the sensors attached to fingers to accurately distinguish different hand gestures and then drive the manipulator to present the same gestures from “zero” to “five” (SI Appendix, Fig. S39). Simultaneously, the wireless human−machine interaction system with continuous, accurate, and fast responses is shown in Movie S3. Moreover, the human−machine interaction system was employed in simulated extreme environments with fire and ice, as exhibited in Fig. 5D and Movies S4 and S5, where the ambient temperature reached over 188 °C and below −5 °C, respectively. The application demonstrated in such extreme conditions verifies the strong environmental applicability of the sensors for human−machine interactions used in extreme rescue activities.

Conclusions

In summary, a method was proposed and widely investigated to produce programmable PSECs using a Joule heating chemistry strategy. The proposed universal approach is capable of fabricating structurally controllable PSECs with variable components (graphene, Ag, GO, TiO₂, ZnO, Fe₃O₄, V₂O₅, MoS₂, BN, g-C₃N₄, BaCO₃, CuI, BaTiO₃, PVDF, cellulose, SBR, montmorillonite, and EuDySrAlSiO_x) within seconds, significantly reducing the fabrication time consumed by traditional preparation methods and optimizing the manufacturing process in an environmentally friendly way. Owing to the abundant deformation sites and exceptional stability of the PSEs, capacitive mechanical sensors based on the elastomer with 5 μm diameter were fabricated with a wide pressure range (0 to 500 kPa), a fast response time (230 ms), long-term durability (over 15,000 cycles), and extreme operating temperatures (−10 to 230 °C). Furthermore, the sensors were used in human motion monitoring and are compatible with various media, including air, water, and oil. Finally, a wireless human−machine interaction system capable of operating in extreme environments was successfully constructed, showing great application potential in extreme rescue activities.

Materials and Methods

Chemicals.

All chemical reagents employed in our studies are available commercially and used directly without further purification. The base material and curing agent of SE (Sylgard 184) were provided by Dow Corning Corporation. Graphene was supplied by Tanfeng Technology Co., Ltd. GO (99%), Ag (99%), TiO₂ (99%), ZnO (99%), Fe₃O₄ (99%), MoS₂ (99.5%), BN (99.9%), g-C₃N₄ (99.2%), BaCO₃ (99.95%), and BaTiO₃ (99.9%) were purchased from Shanghai Aladdin Biochemical Technology Co., Ltd. V₂O₅ (99.2%) was provided by Alfa Aesar. CuI (99.5%), cellulose, SBR, and montmorillonite were obtained from Beijing Innochem Technology Co., Ltd. PVDF was obtained from MTI Corporation. EuDySrAlSiO_x was purchased from Guangzhou Qihong Technology Materials Co., Ltd. Ultrapure water was used during all experiments.

Preparation of PSEs.

The base and cross-linking reagents of SE were initially mixed at a mass ratio of 10:1. Subsequently, abundant ultrapure water was added to the uncured mixture in mass ratios of 50%, 100%, 150%, and 200% respectively, forming a gel-like precursor. Five grams of precursor in different mass ratios (50%, 100%, 150%, and 200%) was then placed into a commercial MW oven for the curing reaction under a working power of 800 W for 8, 4, 3, and 2 min, respectively. Finally, the acquired PSEs were dried in an oven (80 °C, 24 h) to eliminate H₂O.

Preparation of PSECs.

The base and curing agent of SEs were initially mixed at a mass ratio of 10:1. Subsequently, variable components (graphene, Ag, GO, TiO₂, ZnO, Fe₃O₄, V₂O₅, MoS₂, BN, g-C₃N₄, BaCO₃, CuI, BaTiO₃, PVDF, cellulose, SBR, montmorillonite, and EuDySrAlSiO_x) were added into the uncured precursor in a mass ratio of 1:20 for the fabrication of PSECs systems. The resulting mixtures (5 g) were thoroughly stirred with ultrapure water in mass ratios of 200% for graphene, Ag, GO, TiO₂, ZnO, Fe₃O₄, BN, BaCO₃, CuI, BaTiO₃, PVDF, cellulose, SBR, montmorillonite, EuDySrAlSiO_x and 100% for V₂O₅, MoS₂, and g-C₃N₄, respectively. The gel-type precursors were then placed into the MW oven for 120 s reaction (800 W) for thorough curing and dried in the oven (80 °C, 24 h) for full H₂O removal.

Fabrication of Capacitive Mechanical Sensors.

The PSEs were cut into 10 mm × 10 mm × 1 mm pieces as the dielectric layer. Cu films were attached to the upper and lower surfaces as conducting electrodes, and copper wires served as conductors. Polydimethylsiloxane films were used to encapsulate the device.

Characterization.

The morphology and structure of the samples were investigated using FESEM (FEI, Verios G4). Crystalline structures of the PSEs and PSECs were determined by XRD (Bruker, D8 Discover, Cu K_α, 2θ scan range of 5 to 80°). The chemical composition was identified by FTIR (Bruker, Tensor II). TGA (NETZSCH, 209 F3) analysis was employed on the SEs detecting from 30 to 800 °C at a heating rate of 10 °C min⁻¹ in N₂. The absolute viscosities of the samples were characterized at room temperature with a viscometer (Brookfield). The weight-average molecular weights (Mw) and number-average molecular weights (Mn) of the polymers under different reaction times were recorded by GPC (ALLIANCE E2695). The IR images were examined by an IR camera (FLIR E85). Mechanical tests of SEs were characterized by an electronic universal testing machine (INSTRON, 3344), and all samples were cut to a size of 10 × 10 × 5 mm³ with the same procedure for compressive tests. The capacitive response signals of the assembled sensors were obtained via an LCR meter (TH2838) with a resolution of 0.000001 pF in 10 Hz. The Joule heating reaction was conducted in the MW oven (Midea, M1−L201B). COMSOL Multiphysics 6.1 software (COMSOL Inc. Boston, MA) was employed to deal with the numerical problem with temperature-dependent properties. The permittivity, permeability, and thermal conductivity were input to the simulation software. The conventional heating mode and the Joule heating chemistry mode were constructed in the actual sizes (2 × 2 × 2 cm³). The heating temperature of conventional heating was set as 100 °C, and the frequency of Joule heating chemistry was set as 2.45 GHz. Since the reaction system was fabricated with H₂O in SE, the exact size of the shape (H₂O in around 5 μm ball diameter and SE in 100 μm × 100 μm square) was used in the numerical model construction. The heat transfer simulation was performed with a heat transfer module in the software. The electric field distribution was simulated by a radio frequency module (electromagnetic waves, frequency domain). The temperature distribution was carried out via multi-physics module (electromagnetic heat).

Supplementary Material

Appendix 01 (PDF)

Movie S1.

Video of pressing processes for PSE-200.

Movie S2.

Video of the stretching process for PSE-200.

Movie S3.

Video of wireless human−machine interactions by distinguishing the gestures from “zero” to “five”.

Movie S4.

Video of the operating process of wireless human−machine interaction system in a simulated fire scenario.

Movie S5.

Video of a wireless human−machine interaction system acting in a simulated ice environment.

Data, Materials, and Software Availability

All study data are included in the article and/or supporting information.