Three‐dimensional (3D)‐printed MXene high‐voltage aqueous micro‐supercapacitors with ultrahigh areal energy density and low‐temperature tolerance

Abstract

The rapid advancement in the miniaturization, integration, and intelligence of electronic devices has escalated the demand for customizable micro‐supercapacitors (MSCs) with high energy density. However, efficient microfabrication of safe and high‐energy MXene MSCs for integrating microelectronics remains a significant challenge due to the low voltage window in aqueous electrolytes (typically ≤0.6 V) and limited areal mass loading of MXene microelectrodes. Here, we tackle these challenges by developing a high‐concentration (18 mol kg⁻¹) “water‐in‐LiBr” (WiB) gel electrolyte for MXene symmetric MSCs (M‐SMSCs), demonstrating a record high voltage window of 1.8 V. Subsequently, additive‐free aqueous MXene ink with excellent rheological behavior is developed for three‐dimensional (3D) printing customizable all‐MXene microelectrodes on various substrates. Leveraging the synergy of a high‐voltage WiB gel electrolyte and 3D‐printed microelectrodes, quasi‐solid‐state M‐SMSCs operating stably at 1.8 V are constructed, and achieve an ultrahigh areal energy density of 1772 μWh cm⁻² and excellent low‐temperature tolerance, with a long‐term operation at −40°C. Finally, by extending the 3D printing protocol, M‐SMSCs are integrated with humidity sensors on a single planar substrate, demonstrating their reliability in miniaturized integrated microsystems.

INTRODUCTION

Microscale energy storage units are vital for the future development of multifunctional microsystems, integrating portable electronics and smart devices.^1–3 To harmonize with these devices, it is necessary to apply advanced microfabrication techniques, optimizing space usage and fabricating devices on the same substrates to create high-performance integrated microsystems. Advanced microfabrication techniques hold immense potential for building such systems, despite practical implementation challenges. The need to overcome these hurdles has sparked significant interest in designing and microfabricating “on-chip” energy storage units. Notably, on-chip planar micro-supercapacitors (MSCs), as high-power devices, have been proven to be suitable for integration with miniaturized electronics. However, the persisting issue of low energy density in MSCs remains to be addressed. The energy density of MSCs is largely influenced by factors such as the manufacturing process, microelectrode, electrolyte, and configuration of different components.^4–7

Recently, MSCs with three-dimensional (3D) architectures have been studied for their potential to improve areal energy storage performance, due to their ability to load high-mass active materials per unit area. Advanced 3D printing technology can efficiently regulate the loading of electrode material by controlling the number of printing layers with 3D architectures, surpassing the thickness limit of electrodes prepared using the conventional doctor blade coating technique, and result in high areal capacitance per unit.⁸ Furthermore, in practical applications, it can cater to the needs of custom energy storage devices on a large scale. However, the microfabrication of 3D MSCs exploiting extrusion-based 3D printing technologies hinges on the availability of printable inks with high conductivity and electrochemical activity.^9,10

Despite significant progress in 3D printing electrodes and electrochemical devices, most printable inks require a variety of additives like auxiliary solvents or surfactants to control their rheological properties. For post-printing, these additives significantly affect the electrochemical properties of the electrode and usually necessitate removal.^11,12 Hence, development of electrochemically active and highly conductive additive-free inks for upscaling microfabrication through 3D printing is crucial. The recently developed 2D MXene (Ti₃C₂T_x) obtained by etching Ti₃AlC₂^13–15 possesses high electrical conductivity, good rheological properties, excellent charge storage capability, and hydrophilicity,^16–18 making it suitable for preparing inks compatible with 3D printing technology to create high-quality MSC electrodes.^11,19,20 The 3D architecture offers ample porosity to provide transport channels for ions/electrons, enhancing the electrochemical performance. In this context, 3D printing technologies have been explored to construct microelectrode architectures, such as 3D MXene aerogel-based MSC (79 mF cm⁻² and 3.9 µWh cm⁻²),²¹ solid-state MXene-MSC (2.1 F cm⁻² and 24.4 μWh cm⁻²),²² and flexible all-MXene MSC (1035 mF cm⁻² and 51.7 μWh cm⁻²).²³ However, these solutions’ energy density remains unsatisfactory due to the low operating voltage of only 0.6 V, leaving much room for improvement.

The electrolyte is a key component of MSCs and significantly influences their operating voltage window, which in turn affects the device's energy density.^24,25 Therefore, it is paramount to regulate and design the electrolyte accurately. Currently, the stable working voltage of aqueous electrolytes is generally limited by the thermodynamic decomposition voltage of water (1.23 V).^26,27 Organic electrolytes (≥2 V) and ionic liquids (≥3 V) can achieve a higher operating voltage window.²⁸ Although they have broadened the operating voltage window, the high flammability and volatility of organic electrolytes cause serious safety issues, while ionic liquids are expensive and suffer from low ionic conductivity, hindering their large-scale applications. Therefore, there is ongoing research aiming to explore new electrolytes that outperform traditional ones.^29,30 In this regard, a new type of high-concentration “water in salt” (WiS) electrolyte is capable of breaking the thermodynamic shackles of water, significantly widening the electrochemical stability window (ESW) of aqueous electrolytes due to the significantly inhibited activity of water molecules and the production of aqueous solid–electrolyte interphases on the electrode.^26,31–33 Hence, the application of a WiS electrolyte is expected to considerably increase the voltage of MSCs, thereby promoting further advancements in the energy density of 3D-printed all-MXene symmetric MSCs (M-SMSCs).

Herein, we prepared a highly concentrated additive-free MXene ink that facilitates rapid 3D printing of customizable and thick MXene microelectrodes on various substrates such as flexible polymer films, rubber, and cloth with excellent adhesion. Further focusing on high energy density, we developed a green, economical, high-concentration WiB gel electrolyte matched with MXene microelectrodes. With increased concentration, the interaction between ions and water was significantly strengthened, reducing the activity and inhibiting the decomposition of water at higher voltages. This aqueous electrolyte was used in 3D-printed additive-free M-SMSCs, significantly widening the operating voltage window to 1.8 V. Combined with the easily available thick electrodes from 3D printing, the M-SMSCs achieved an ultrahigh areal energy density of 1772 μWh cm⁻², at least one order of magnitude higher than previously reported MSCs. Additionally, due to the ultrahigh ionic conductivity (97 mS cm⁻¹) and ultra-low freezing point (−77°C) of the LiBr-gel electrolyte, even at −40°C, 3D-printed M-SMSCs could still function over an extended period, proving their practicability in extreme conditions. To demonstrate practical applications, we constructed an all-MXene-printed integrated microsystem capable of energy storage and intelligent moisture sensing.

EXPERIMENTAL SECTION

Preparation of MXene ink

Monolayer Ti₃C₂T_x MXene colloidal solutions were prepared using an in situ hydrofluoric acid-based etching strategy. The high-concentration Ti₃C₂T_x MXene aqueous solution was achieved through a three-step process of etching, stripping, and concentrating. Initially, 6.4 g of lithium fluoride (LiF, 99%; Aladdin Reagent Co., Ltd.) was added to 80 mL of 9 M hydrochloric acid (HCl; Shanghai Runjie Chemical Reagent Co., Ltd.) to serve as an etchant. The etchant was placed in a 200-mL Teflon reactor and thoroughly stirred at 500 rpm to dissolve the LiF. Subsequently, 4.0 g of Ti₃AlC₂ powders (400 mesh; Shandong Xiyan New Material Technology Co., Ltd.) was slowly added to the above solution. With the addition of Ti₃AlC₂, the etching reaction occurs immediately. Therefore, the Ti₃AlC₂ material should be added slowly to avoid the initial violent exotherm leading to solution boiling and material oxidation. The etching reaction was carried out in a water bath environment of 30°C for 24 h to ensure complete etching of Ti₃AlC₂. After washing with dilute hydrochloric acid and aqueous solution sequentially, excess LiF and acid were removed. The exfoliation process involved 2 h of shaking treatment at 2000 rpm (Multi Reax oscillator, Heidolph) combined with 0.5 h of sonication. Subsequently, the unprecipitated black colloidal solution was collected after centrifugation at 3500 rpm for 30 min (H1850; Xiangyi Centrifuge Instrument Co., Ltd.), resulting in the monolayer Ti₃C₂T_x MXene solution. To increase the concentration of MXene material in the aqueous solution, the solution was centrifuged at 13,000 rpm for 20 min. After removing the upper transparent solution, the solid slurry of high-concentration monolayer MXene at the bottom was collected (solid content: ~13%).

Preparation of WiB gel electrolyte

Specifically, aqueous LiBr electrolytes (1, 2, 5, 10, and 18 mol kg⁻¹ (m)) were obtained by dissolving LiBr salts (Aladdina) in deionized water at room temperature. The “water-in-LiBr” gel electrolytes were prepared by mixing approximately 0.007 µm-sized SiO₂ powder (10 wt% of the solvent; Sigma-Aldrich) with the aforementioned LiBr electrolytes.

3D printing fabrication of M-SMSCs

The 3D-printed M-SMSCs were fabricated using a 3-axis micro-positioning stage (ZZ-221; Zhongzhi Automation Co., Ltd.) equipped with an air-powered fluid dispenser (JND983A) following a preprogrammed patterning procedure. The typical printing speed for MXene inks, housed in separate syringes attached to a 320 μm nozzle, is ~4.5 mm s⁻¹ at an air pressure of 25 psi. The thickness of the MXene microelectrode finger (with a length of 3 mm, a width of 340 μm, and interspace of 460 μm) could be facilely controlled by adjusting the number of printed layers, labeled as M-SMSCs-xL (where x represents the printed layers). After printing, the printed microelectrodes were freeze-dried for 24 h to remove excess water. Finally, M-MSCs with a two-electrode configuration were attained after drop-casting the “water-in-LiBr” gel electrolyte. The conductive copper tape is used as the wire to connect the microelectrode with the external circuit for testing.

Fabrication of a planar integrated microsystem

First, the microelectrodes of interdigitated M-SMSCs, the current collectors of sensors, and the corresponding interconnects were printed on PET substrates using MXene inks under the same 3D printing conditions as before. Then, a humidity-sensitive SENS-H200 ink (Shanghai MiFang Electronic Technology Co., Ltd.), primarily composed of polyvinyl alcohol and water, was blade-coated onto the MXene current collector, and the prepared humidity sensor was dried at 100°C for 20 min. Finally, a layer of 18 m LiBr-gel electrolyte was coated on the interdigitated microelectrodes of M-SMSC. The planar integrated microsystem of M-SMSC and sensor was obtained.

RESULTS AND DISCUSSION

MXene inks for 3D printing M-SMSCs

By selectively etching the Al layer of bulk Ti₃AlC₂ (Figure S1) and stripping it in distilled water, a hydrophilic dispersion of MXene (Ti₃C₂T_x) nanosheets was obtained (Figures S2 and S3). The MXene nanosheets, displaying a thickness of approximately 1 nm (Figure S4), appear nearly transparent under a high-resolution transmission electron microscope (HRTEM, Figure S5A). The corresponding selected-area electron diffraction (SAED) pattern reveals a well-defined hexagonal crystal symmetry (Figure S5B). Highly concentrated pure MXene ink (Figure S6) suitable for 3D printing was obtained by ultrahigh-speed centrifugation of the diluted dispersion of MXene nanosheets. The resulting MXene ink shows high viscosity and excellent rheology (Figure S7), which are favorable for precise layer-by-layer patterning in the vertical direction, particularly for extrusion-based 3D printing manufacturing technology. The microfabrication process of M-MSCs based on highly concentrated aqueous MXene ink is illustrated in Figure 1. MXene microelectrodes with different thicknesses were fabricated by 3D printing MXene ink (Figure 1A). Interestingly, MXene ink could be easily printed on a variety of substrates such as glass, polyethylene terephthalate (PET), rubber, and cloth (Figure 1B–H). Even under various states (e.g., distortion and inversion), there was no structure delamination or degradation of the printed patterns, confirming good adhesion between the microelectrode and the substrate. Figure 1B presents a photograph of 5 × 3 interdigital microelectrode arrays printed on a glass substrate (10 cm × 10 cm), demonstrating the potential for large-scale rapid manufacturing. Furthermore, the profile display of multilayer MXene microelectrodes printed on the PET substrate in different states (Figure 1D,E) remained well maintained even when inverted, without deformation or collapse. More personalized geometries can be precisely 3D-printed into as-designed patterns on the PET substrate for improving the esthetic appeal of MSCs (Figure 1F,G). Figure 1H shows custom-shaped 3D M-MSCs on a soft cloth substrate, showcasing high versatility and flexibility. The morphologies of the side and cross-section of the microelectrode fingers are shown in Figure 1I–L. The favorable contact between adjacent layers ensures the structural robustness of the ultrathick microelectrodes (Figure 1I,J). This is also corroborated by the cross-section view of the microelectrodes, where 3D-printed interdigital fingers were effectively fused, sharing an internal network of electron conduction paths formed by MXene nanosheets and interconnected ion transporting pores constructed by freeze-drying to remove H₂O (Figure 1K,L). The 3D-printed MXene microelectrode possesses a specific surface area of 29.4 m² g⁻¹ with abundant pores ranging from macropore to mesopore (Figure S8). Such dense and varied porous structures can promote the penetration of electrolytes and provide many open pathways for fast ionic diffusion, contributing to the high electrochemical performance of M-SMSCs.^21,22

[IMAGE OMITTED. SEE PDF]

Physicochemical properties of water-in-LiBr electrolyte

To shed light on the structural features of electrolytes, the molecular structure evolution between low- and high-concentration LiBr electrolytes was first analyzed by molecular dynamics (MD) simulations (Figures 2A,B and S9). With increasing concentration, the number of free water molecules around salt ions decreases. Notably, in the 1 m (mol kg⁻¹) LiBr electrolyte (Figure 2A), solvated Li⁺ ions with four (45%) or five (43%) water molecules (Table S1) were formed via oxygen atoms and well separated from Br⁻ (Figure 2A1,A2). In sharp contrast to 18 m LiBr electrolyte (Figure 2B), the majority of the water molecules were linked to Li⁺ ions via oxygen atoms and one/two Br⁻ ions (Figure 2B1,B2 and Table S2). Owing to the limited H₂O, the Br⁻ ions partially coordinated with Li⁺ ions, inducing a strong cation-anion interaction. Based on equilibrium MD simulations, a concentration-dependent coordination between Li⁺, Br⁻, and H₂O occurred (Figures S10 and S11). From 1 to 18 m, the average coordination number of water molecules around Li⁺ decreased from 4.40 to 2.37, while the average coordination number of Br⁻ increased from 0.08 to 1.66 (Figure S10), and matched well with the typical molecular conformations (Figure 2A,B). It is suggested that the anion Br⁻ began to participate in the solvation the shell structure of Li⁺ at 18 m and substituted fractional water molecules, reducing the coordination between Li⁺ and H₂O (Figure 2B), which consequently resulted in weakened Br⁻/O···H and intensified Br⁻/O···Li⁺ interactions (Figure S11), and decreased the coordination number of Br-H, O-H, and O-Li (Figure S10). From 1 to 18 m, the original hydrogen-bonding network was destroyed and the number of free H₂O decreased in this process. The effective electrostatic interactions between H₂O and electrolyte cations reduced the activity of water and inhibited its decomposition, thus widening the larger operating voltage window at 18 m.

[IMAGE OMITTED. SEE PDF]

The results of the theoretical calculation results were consistent with the experimental observations in Raman (Figure 2C) and Fourier transform infrared spectra (FTIR, Figure 2D). A broad Raman/FTIR band was observed with O–H stretching vibration in pure water. From 1 to 18 m, the broadband of O–H stretching vibration gradually disappeared, and until at 18 m, only a sharp peak at 3459 cm⁻¹ existed in Raman spectra (Figure 2C). Similarly, as the wave number of the tensile vibration increased, the wave number of the bending vibration decreased, and the peak value became more acute in the FTIR spectra (Figure 2D), attributed to the perturbation with H₂O–Li⁺–Br⁻ interactions. There was a significant reduction in the abundance of water clusters at a high concentration of 18 m. The strong hydrogen bonds of free H₂O were significantly broken and the solvent framework of the solution was reconstructed. The electronic environment of electrolyte ions in 1–18 m LiBr aqueous electrolyte was characterized by nuclear magnetic resonance (NMR, Figure S12). Signals of ¹H and ⁷Li were blue-shifted with increasing salt concentration, and the electron density around H atoms in water molecules and Li atoms decreased and the shielding weakened. This is caused by the decrease of H bonds and the enhancement of the O–Li interaction,³⁴ which is consistent with the theoretical results.

The ionic conductivity, viscosity, price, and safety of electrolytes are critical parameters for the overall performance and practicability of MSCs. The physicochemical properties of LiBr at various concentrations are shown in Figure 2E and Table S3. Despite the molality being as high as 18 m, its viscosity is only 8.09 mPa s at 25°C (Figure S13), significantly lower than the viscosity of the previously reported high-concentration electrolytes (20 m LiCl, 21 m LiTFSI, 30.8 m KFSI, and 42 m LiTFSI + 21 m Me₃EtN·TFSI electrolyte: 13.3, 36.2, 14.5, and 407 mPa s, respectively, Table S3).^26,30,35,36 Due to their low viscosity, the ionic conductivity of 18 m LiBr electrolyte is as high as 105 mS cm⁻¹. Even after the gel electrolyte was formed, the ionic conductivity is still as high as 97 mS cm⁻¹, which is higher than those of the 20 m LiCl (72.7 mS cm⁻¹), 21 m LiTFSI (8.21 mS cm⁻¹), 30.8 m KFSI (43 mS cm⁻¹), and 42 m LiTFSI + 21 m Me₃EtN·TFSI (0.91 mS cm⁻¹) electrolytes. Thermogravimetry measurements (Figure 2F) indicated that the weight loss of 18 m LiBr was just 2.53 wt% on increasing the temperature to 100°C, lower than those obtained for 1 m LiBr (26.21 wt%) and 5 m LiBr (12.86 wt%). Water evaporation was inhibited in high-concentration electrolytes, showing a better water retention function. Owing to the reconstructed hydration environments, the thermal stability of water molecules increased with salt molality, gradually enhancing heat tolerance. The superior thermal stability of electrolytes significantly ameliorates safety property and long-term usability compared with the dilute electrolytes.^37,38

To understand the effect of LiBr concentration on the electrochemical performance of MXene, we first studied the cyclic voltammetry (CV) curves of M-SMSCs at a scan rate of 5 mV s⁻¹ (Figure 2G). As observed, the voltage window of M-SMSCs can be expanded to 1.8 V without obvious water splitting in 18 m LiBr aqueous gel electrolyte. However, in 1, 5, and 10 m LiBr aqueous gel electrolyte, CV curves polarized to varying degrees when the voltage was higher than 1.2 V, indicative of water decomposition. Obviously, with increasing salt concentration, the hydrolysis was significantly inhibited, which was beneficial for the widening of the voltage window. This phenomenon can be explained by the significant decrease in free H₂O at high concentrations, the transformation of the solvation structure of Li⁺ ions, and the involvement of Br⁻ ions, which strengthen the interaction between anions and cations while reducing the activity of water.

Electrochemical characteristics of M-SMSCs

The operating voltage windows of M-SMSCs in the 18 m WiB gel electrolyte (Figure S14) were identified by CV measurements (Figure 3A). When the voltage exceeded 1.6 V, a pair of redox peaks emerged, further increasing the working voltage to 1.8 V. It is noted that the appearance of a redox peak is attributed to the redox reaction of a large amount of Br⁻ in the high-concentration electrolyte at this potential,³⁹ which further inhibits the decomposition of water. The inherent redox of Br⁻ renders it an advantageous component in enhancing charge storage capacity. However, when the operating potential increases beyond 1.8 V, the CV curve begins to polarize and the water in the electrolyte is decomposed. Therefore, the printed M-SMSCs reach a high operation voltage of 1.8 V, which is the highest value among recently reported aqueous MXene-based SMSCs (Figure 3B). Most reported aqueous MXene-based SMSCs deliver a low output voltage (<1.0 V, Table S4). Given that an ESW as high as 1.8 V can be achieved with the cost-effective LiBr electrolyte, this strategy is highly promising. Notably, the WiB electrolyte, characterized by its high concentration and redox properties, serves multiple roles: (i) facilitating ion conduction, (ii) limiting water decomposition, and (iii) acting as electrochemically active species. This multi-role functionality ensures smooth ion transport, broadens the operating voltage window, and endows M-SMSCs with additional capacitance.

[IMAGE OMITTED. SEE PDF]

Furthermore, we constructed M-SMSCs by replicate printing with 1, 2, 4, and 8 layers (Figure S15, denoted as M-SMSCs-1L, M-SMSCs-2L, M-SMSCs-4L, and M-SMSCs-8L, respectively) and investigated the dependence of the areal capacitances and electrical conductivity of microelectrodes with thickness. The shapes of CV curves (Figure 3C) and galvanostatic charge-discharge (GCD) profiles (Figure 3E,F) are well maintained at 200 mV s⁻¹ and 100 mA cm⁻², which confirms that M-SMSCs-1L shows outstanding rate performance. The CV and GCD results of M-SMSCs-2L, M-SMSCs-4L, and M-SMSCs-8L devices are shown in Figures S16–S18, confirming capacitive behavior.

At 5 mV s⁻¹, M-SMSCs-1L, M-SMSCs-2L, M-SMSCs-4L, and M-SMSCs-8L show high areal capacitances of 473, 878, 1779, and 3901 mF cm⁻², respectively (Figure 3D). The capacitance of M-SMSCs scales in an approximately linear fashion with an increased number of printed layers. At 100 mA cm⁻² (Figure S19), M-SMSCs-1L, M-SMSCs-2L, M-SMSCs-4L, and M-SMSCs-8L devices still maintain high areal capacitances of 285, 490, 1162, and 2669 mF cm⁻², respectively, showing excellent rate capability even in the case of thick layers. The current-voltage curves of a 3D-printed MXene microelectrode finger (Figure S20A) show that the resistance decreases with increasing number of layers. This was further corroborated using electrochemical impedance spectroscopy (EIS), and the ion and charge transport kinetics of the 3D-printed M-SMSCs (Figure S20B) was investigated. The equivalent series resistance (ESR) of M-SMSCs decreases with the incremental layers. Specifically, M-SMSCs-8L shows the lowest ESR (15.7 Ω) and a nearly vertical line at the low-frequency region, indicative of rapid ion diffusion and good capacitive behavior (Figure S20A). Thus, M-SMSCs retain good rate performance even at thick electrodes with a greater number of layers.

For all devices, increasing scan rate results in decreased areal capacitance, which is also influenced by the number of printed layers. At 2 mA cm⁻², M-SMSCs-1L showed high energy and power densities of 209 µWh cm⁻² and 1.8 mW cm⁻², respectively (Figure 3E). On 50-fold increasing to 100 mA cm⁻² (Figure 3F), 61% of the energy density is retained (128 µWh cm⁻²), while the power density increases to 90 mW cm⁻². The highest areal energy and power densities measured for M-SMSCs-8L are 1772 μWh cm⁻² and 180 mW cm⁻², respectively. These values are much higher than those of previous MXene-based MSCs. Figure 3G compares Ragone plots of various MXene-based MSC devices and further showcases the potential of high energy/power density of 3D-printed MXene interdigitated SMSCs. The calculated areal energy density of our M-SMSC increases up to 1772 µWh cm⁻², being several orders of magnitude higher than those of SMSCs using different printing technologies (Table S5), including screen printing, laser-scribing, inkjet printing, and 3D printing. Electrochemical characterization further confirmed that satisfactory cycling performance (93% after 10,000 cycles) was achieved in the M-SMSC-1L module at 50 mA cm⁻² and 1.8 V (Figure 3H). M-SMSCs in customized shapes (letters M, S, and C) connected in series can easily light up 42 light-emitting diodes (LEDs) in both flat (Figure 3I) and bent (Figure 3J) states, demonstrating excellent flexibility. This goes to show that by manipulating the configuration (series/parallel), 3D-printed M-SMSCs can be personalized for use as flexible microscale power units. Such modules can meet different power/energy requirements and have the potential to power the microgrid and miniaturized integrated systems.

To implement high-voltage and high-capacitance integrated circuits, M-SMSCs arrays in series (Figure S21a) and parallel (Figure S21b) configurations were assembled directly via 3D printing. Five in-series SMSCs connections were designed to improve output voltage. It is observed that the GCD profiles of M-SMSCs connected in series showed the operating voltage with a linear increase from a single 1.8 to 5.4 V for three units and 9 V for five units (Figure 4A). Furthermore, the operating voltage of M-SMSCs linked in parallel from one to five units remained unchanged, but the output capacitance was effectively enhanced (Figure 4B). Interestingly, concentric M-SMSC showed only minor fluctuations in capacitance retention from 0° to 180° bending angles was effectively enhanced (Figures 4C and S22), signifying the excellent flexibility of the shape-customized devices.

[IMAGE OMITTED. SEE PDF]

Electrochemical performance at low temperatures is critical for several practical applications of microscale energy storage devices. Usually, at low temperatures, aqueous electrolytes get easily solidified, leading to sluggish ion diffusion, weakened ionic conductivity, and deterioration of the electrolyte-electrode interface, thus reducing the performance of MSCs. Therefore, reducing the freezing point of electrolyte is regarded as an effective way to realize high-performance low-temperature aqueous MSCs. From differential scanning calorimetry (DSC) curves (Figure 4D), it is revealed that the high content of LiBr plays a key role in hindering electrolyte freeze. Owing to the breakdown of the hydrogen-bond network between free water in a high-concentration LiBr solution, the freezing point of the electrolyte decreases from −24°C (1 m) to −77°C (18 m). Then, we systematically evaluate the electrochemical properties of M-SMSCs-1L in 18 m LiBr-gel electrolyte at various temperatures ranging from 20 to −40°C, as shown in the temperature-dependent electrochemical characteristics presented in Figure 4E–H. From 20 to −40°C, the discharge time of M-SMSCs-1L slowly decreases (Figure 4E). It still showed a high capacitance of 353 mF cm⁻² at 2 mA cm⁻² obtained by the GCD profile under −40°C, corresponding to high capacitance retention of 76% at 20°C (Figure 4F). EIS at different temperatures (Figure 4G) showed that the resistance change is relatively small from 20 to −40°C, and the low ESR value only increases from 20 Ω at 20°C to 52 Ω at −40°C (24 Ω at 0°C, 33 Ω at −20°C). Such low-temperature tolerance of high-concentration LiBr and the high compatibility with microelectrodes ensure that MSCs with excellent electrochemical performance retained high capacitance at low temperatures. Meanwhile, nearly 96% of the capacitance is retained after 10,000 cycles at 10 mA cm⁻² (Figure 4H), showing excellent durability and low-temperature performance. These results demonstrate that M-SMSCs have broad application prospects in cold conditions.

3D printing an M-SMSC-sensor integrated microsystem

With the advent of the intelligence era, there has been strong interest in the miniaturization of intelligent products and their integration onto chips or flexible substrates. As power-consuming smart devices cannot function independently, they need an external power supply. Typically, the electrical device and the power supply are connected through external wires, coupled with the bulky volume of the traditional power supply, which is unfavorable for the development of flexible wearable and portable smart devices. Therefore, dual-functional monolithic devices integrating miniature energy storage and conversion on the same flexible substrate have emerged. To verify the feasibility of 3D printing high-performance planar M-SMSCs for practical applications in microsystem integration, we continuously printed M-SMSCs and humidity sensors on flexible PET substrates using MXene inks (Figures 5A and S23). The same conductive MXene ink was used to print the interconnects between MSC and the sensor directly, eliminating the use of metal wires and metal current collectors, thus realizing their planar integration. The integrated microsystem was minutely evaluated considering various aspects. The self-discharge performance of M-SMSC (Figure S24) shows that it can maintain a voltage of more than 0.9 V within 10,000 s after being fully charged, indicating that M-SMSC can provide a continuous power supply for the sensor. Figure 5B shows the change in the current response of a humidity sensor powered by M-SMSC in response to continuous blowing by a person. The M-SMSC-driven sensor responded quickly and stably when a person blew air on it. The sensor was blown several times, and its response value fluctuated between 10.7% and 8.1% (Figure 5C). The M-SMSC-driven sensor microsystem could operate normally and showed good sensing performance, which can be used for humidity measurement in various daily life situations. Figure 5D–F shows that the M-SMSC-driven humidity sensor is more sensitive to a wet wipe. When the sensor touched the wet wipe, a rapid response occurred, and the response value could reach 65%–75%, which could be applicable for unique wearable devices, such as detecting the degree of wetness of a baby's diaper so as to notify parents of the need to change the diaper. The above results demonstrate the feasibility, uniqueness, and application potential of our printed integrated microsystems.

[IMAGE OMITTED. SEE PDF]

CONCLUSION

In conclusion, our prepared aqueous MXene ink with excellent rheology without additional additives enables 3D printing of customizable M-SMSCs on various substrates. We coupled these microelectrodes with our newly developed green, economical, and high-concentration WiB gel electrolyte, which has multifunctional attributes. The resulting M-SMSCs show a considerably enhanced voltage window of 1.8 V and an ultrahigh areal capacitance close to 4 F cm⁻², resulting in an excellent areal energy density of 1772 μWh cm⁻², far surpassing previously reported MXene-based SMSCs. Furthermore, the 3D-printed all-M-SMSCs maintain their normal operation at −40°C, demonstrating their applicability and practicality in extreme environments. As verification of the practical application, we further printed the all-in-one integration microsystem of M-SMSCs and humidity sensor on the same substrate, which showed a quick and reliable response to changes in moisture, indicating their potential for integrated microsystem applications. This feasible approach holds great potential for fabricating high-performance energy storage modules, either individually or in cooperation with other electronic devices. This work will inspire more researchers to further explore high-voltage MXene-based aqueous microscale energy storage devices along with printable integrated microsystems and accelerate the advancement of 3D-printed high-performance integrated electronics.

ACKNOWLEDGMENTS

This work was financially supported by the National Natural Science Foundation of China (Grant Nos. 22005297, 22125903, 51872283, 22209175, 22209176), the National Key R&D Program of China (Grant No. 2022YFA1504100), the Support Program for Excellent Young Talents in Universities of Anhui Province (Grant No. 2022AH030134), the Anhui Province Higher Education Innovation Team: Key Technologies and Equipment Innovation Team for Clean Energy (Grant No. 2023AH010055), the Strategic Priority Research Program of the Chinese Academy of Sciences (Grant No. XDB36030200), the Dalian Innovation Support Plan for High Level Talents (2019RT09), the Dalian National Laboratory for Clean Energy (DNL), CAS, DNL Cooperation Fund, CAS (DNL202016, DNL202019, DNL202003), DICP (DICP I2020032), the Doctor Research Startup Foundation of Suzhou University (2023BSK015), the China Postdoctoral Science Foundation (Grant Nos. 2020M680995, 2021M693127), and the International Postdoctoral Exchange Fellowship Program (Talent-Introduction Program, YJ20210311).

CONFLICT OF INTEREST STATEMENT

The authors declare that there are no conflicts of interests.

References

Zhang P, Wang F, Yu M, Zhuang X, Feng X. Two‐dimensional materials for miniaturized energy storage devices: from individual devices to smart integrated systems. Chem Soc Rev. 2018;47(19):7426‐7451.

Simon P, Gogotsi Y. Perspectives for electrochemical capacitors and related devices. Nat Mater. 2020;19(11):1151‐1163.

Zhu Q, Li J, Simon P, Xu B. Two‐dimensional MXenes for electrochemical capacitor applications: progress, challenges and perspectives. Energy Storage Mater. 2021;35:630‐660.

Kim BC, Hong J‐Y, Wallace GG, Park HS. Recent progress in flexible electrochemical capacitors: electrode materials, device configuration, and functions. Adv Energy Mater. 2015;5(22): [eLocator: 1500959].

Wang R, Yao M, Niu Z. Smart supercapacitors from materials to devices. InfoMat. 2020;2(1):113‐125.

Zhang P, Li J, Yang D, Soomro RA, Xu B. Flexible carbon dots‐intercalated MXene film electrode with outstanding volumetric performance for supercapacitors. Adv Funct Mater. 2023;33(1): [eLocator: 2209918].

Zhang P, Soomro RA, Guan Z, Sun N, Xu B. 3D carbon‐coated MXene architectures with high and ultrafast lithium/sodium‐ion storage. Energy Storage Mater. 2020;29:163‐171.

Wang S, Ma J, Shi X, Zhu Y, Wu Z‐S. Recent status and future perspectives of ultracompact and customizable micro‐supercapacitors. Nano Res Energy. 2022;1: [eLocator: e9120018].

Zhu C, Liu T, Qian F, et al. 3D printed functional nanomaterials for electrochemical energy storage. Nano Today. 2017;15:107‐120.

Li H, Liang J. Recent development of printed micro‐supercapacitors: printable materials, printing technologies, and perspectives. Adv Mater. 2020;32(3): [eLocator: 1805864].

Zhu Y, Wang S, Ma J, Das P, Zheng S, Wu ZS. Recent status and future perspectives of 2D MXene for micro‐supercapacitors and micro‐batteries. Energy Storage Mater. 2022;51:500‐526.

Zhang Y, Zhu Y, Zheng S, et al. Ink formulation, scalable applications and challenging perspectives of screen printing for emerging printed microelectronics. J Energy Chem. 2021;63:498‐513.

Ghidiu M, Lukatskaya MR, Zhao M‐Q, Gogotsi Y, Barsoum MW. Conductive two‐dimensional titanium carbide ‘clay’ with high volumetric capacitance. Nature. 2014;516(7529):78‐81.

Zhou A, Liu Y, Li S, et al. From structural ceramics to 2D materials with multi‐applications: a review on the development from MAX phases to MXenes. J Adv Ceram. 2021;10(6):1194‐1242.

Liu P, Liu W, Liu K. Rational modulation of emerging MXene materials for zinc‐ion storage. Carbon Energy. 2022;4(1):60‐76.

Truby RL, Lewis JA. Printing soft matter in three dimensions. Nature. 2016;540(7633):371‐378.

Anasori B, Lukatskaya MR, Gogotsi Y. 2D metal carbides and nitrides (MXenes) for energy storage. Nat Rev Mater. 2017;2(2): [eLocator: 16098].

Zhang Y‐Z, Wang Y, Jiang Q, El‐Demellawi JK, Kim H, Alshareef HN. MXene printing and patterned coating for device applications. Adv Mater. 2020;32(21): [eLocator: 1908486].

Zhu Y, Ma J, Das P, Wang S, Wu Z‐S. High‐voltage MXene‐based supercapacitors: present status and future perspectives. Small Methods. 2023;7(8): [eLocator: 2201609].

Jia J, Zhu Y, Das P, et al. Advancing MXene‐based integrated microsystems with micro-supercapacitors and/or sensors: rational design, key progress, and challenging perspectives. J Materiomics. 2023;9(6):1242‐1262.

Tetik H, Orangi J, Yang G, et al. 3D printed MXene aerogels with truly 3D macrostructure and highly engineered microstructure for enhanced electrical and electrochemical performance. Adv Mater. 2022;34(2): [eLocator: 2104980].

Yang W, Yang J, Byun JJ, et al. 3D printing of freestanding MXene architectures for current‐collector‐free supercapacitors. Adv Mater. 2019;31(37): [eLocator: 1902725].

Orangi J, Hamade F, Davis VA, Beidaghi M. 3D printing of additive‐free 2D Ti₃C₂T_x (MXene) ink for fabrication of micro‐supercapacitors with ultra‐high energy densities. ACS Nano. 2020;14(1):640‐650.

Gou Q, Zhao S, Wang J, Li M, Xue J. Recent advances on boosting the cell voltage of aqueous supercapacitors. Nano‐Micro Lett. 2020;12(1):98.

Tian X, Zhu Q, Xu B. “Water‐in‐salt” electrolytes for supercapacitors: a review. ChemSusChem. 2021;14(12):2501‐2515.

Suo L, Borodin O, Gao T, et al. “Water‐in‐salt” electrolyte enables high‐voltage aqueous lithium‐ion chemistries. Science. 2015;350(6263):938‐943.

Zhu Y, Zheng S, Qin J, et al. 2.4 V ultrahigh‐voltage aqueous MXene‐based asymmetric micro‐supercapacitors with high volumetric energy density toward a self‐sufficient integrated microsystem. Fundam Res. 2024;4(2):307‐314. [DOI: https://dx.doi.org/10.1016/j.fmre.2022.03.021]

Xu K. Nonaqueous liquid electrolytes for lithium‐based rechargeable batteries. Chem Rev. 2004;104(10):4303‐4418.

Li Z, Gadipelli S, Li H, et al. Tuning the interlayer spacing of graphene laminate films for efficient pore utilization towards compact capacitive energy storage. Nat Energy. 2020;5(2):160‐168.

Zhu Y, Zheng S, Lu P, et al. Kinetic regulation of MXene with water‐in‐LiCl electrolyte for high‐voltage micro‐supercapacitors. Natl Sci Rev. 2022;9(7):nwac024.

Liang T, Hou R, Dou Q, Zhang H, Yan X. The applications of water‐in‐salt electrolytes in electrochemical energy storage devices. Adv Funct Mater. 2021;31(3): [eLocator: 2006749].

Dou Q, Wang Y, Wang A, et al. “Water in salt/ionic liquid” electrolyte for 2.8 V aqueous lithium‐ion capacitor. Sci Bull. 2020;65(21):1812‐1822.

Hu M, Zhang H, Hu T, Fan B, Wang X, Li Z. Emerging 2D MXenes for supercapacitors: status, challenges and prospects. Chem Soc Rev. 2020;49(18):6666‐6693.

Zheng J, Tan G, Shan P, et al. Understanding thermodynamic and kinetic contributions in expanding the stability window of aqueous electrolytes. Chem. 2018;4(12):2872‐2882.

Ko S, Yamada Y, Yamada A. A 62 m K‐ion aqueous electrolyte. Electrochem Commun. 2020;116: [eLocator: 106764].

Chen L, Zhang J, Li Q, et al. A 63 m superconcentrated aqueous electrolyte for high‐energy Li‐ion batteries. ACS Energy Lett. 2020;5(3):968‐974.

Jiang Q, Lei Y, Liang H, Xi K, Xia C, Alshareef HN. Review of MXene electrochemical microsupercapacitors. Energy Storage Mater. 2020;27:78‐95.

Zhang J, Cui C, Wang P‐F, et al. “Water‐in‐salt” polymer electrolyte for Li‐ion batteries. Energy Environ Sci. 2020;13(9):2878‐2887.

Li M, Li X, Qin G, et al. Halogenated Ti₃C₂ MXenes with electrochemically active terminals for high‐performance zinc ion batteries. ACS Nano. 2021;15(1):1077‐1085.

Zhang C, McKeon L, Kremer MP, et al. Additive‐free MXene inks and direct printing of micro‐supercapacitors. Nat Commun. 2019;10:1795.

Peng Y‐Y, Akuzum B, Kurra N, et al. All‐MXene (2D titanium carbide) solid‐state microsupercapacitors for on‐chip energy storage. Energy Environ Sci. 2016;9(9):2847‐2854.

Xu S, Dall'Agnese Y, Wei G, Zhang C, Gogotsi Y, Han W. Screen‐printable microscale hybrid device based on MXene and layered double hydroxide electrodes for powering force sensors. Nano Energy. 2018;50:479‐488.

Yue Y, Liu N, Ma Y, et al. Highly self‐healable 3D microsupercapacitor with MXene‐graphene composite aerogel. ACS Nano. 2018;12(5):4224‐4232.

Li H, Li X, Liang J, Chen Y. Hydrous RuO₂‐decorated MXene coordinating with silver nanowire inks enabling fully printed micro‐supercapacitors with extraordinary volumetric performance. Adv Energy Mater. 2019;9(15): [eLocator: 1803987].

Abdolhosseinzadeh S, Schneider R, Verma A, Heier J, Nüesch F, Zhang C. Turning trash into treasure: additive free MXene sediment inks for screen‐printed micro‐supercapacitors. Adv Mater. 2020;32(17): [eLocator: 2000716].

Li H, Hou Y, Wang F, et al. Flexible all‐solid‐state supercapacitors with high volumetric capacitances boosted by solution processable MXene and electrochemically exfoliated graphene. Adv Energy Mater. 2017;7(4): [eLocator: 1601847].

Qin L, Tao Q, Liu X, et al. Polymer‐MXene composite films formed by MXene‐facilitated electrochemical polymerization for flexible solid‐state microsupercapacitors. Nano Energy. 2019;60:734‐742.

Yang J, Pan Z, Zhong J, Li S, Wang J, Chen PY. Electrostatic self‐assembly of heterostructured black phosphorus–MXene nanocomposites for flexible microsupercapacitors with high rate performance. Energy Storage Mater. 2021;36:257‐264.

Yu C, Gong Y, Chen R, et al. A solid‐state fibriform supercapacitor boosted by host‐guest hybridization between the carbon nanotube scaffold and MXene nanosheets. Small. 2018;14(29): [eLocator: 1801203].

Shen B‐S, Wang H, Wu L‐J, Guo RS, Huang Q, Yan XB. All‐solid‐state flexible microsupercapacitor based on two‐dimensional titanium carbide. Chin Chem Lett. 2016;27(10):1586‐1591.

Wu C‐W, Unnikrishnan B, Chen IWP, Harroun SG, Chang HT, Huang CC. Excellent oxidation resistive MXene aqueous ink for micro‐supercapacitor application. Energy Storage Mater. 2020;25:563‐571.

Huang H, Chu X, Xie Y, et al. Ti₃C₂T_x MXene‐based micro‐supercapacitors with ultrahigh volumetric energy density for all‐in‐one Si‐electronics. ACS Nano. 2022;16(3):3776‐3784.

Li X, Li H, Fan X, Shi X, Liang J. 3D‐printed stretchable micro‐supercapacitor with remarkable areal performance. Adv Energy Mater. 2020;10(14): [eLocator: 1903794].

Zhou G, Li MC, Liu C, Wu Q, Mei C. 3D printed Ti₃C₂T_x MXene/cellulose nanofiber architectures for solid‐state supercapacitors: ink rheology, 3D printability, and electrochemical performance. Adv Funct Mater. 2021;32(14): [eLocator: 2109593].

Ma J, Zheng S, Cao Y, et al. Aqueous MXene/PH1000 hybrid inks for inkjet‐printing micro‐supercapacitors with unprecedented volumetric capacitance and modular self‐powered microelectronics. Adv Energy Mater. 2021;11(23): [eLocator: 2100746].

Zhang Y, Wang L, Zhao L, et al. Flexible self‐powered integrated sensing system with 3D periodic ordered black phosphorus@MXene thin‐films. Adv Mater. 2021;33(22): [eLocator: 2007890].