In recent years, wearable flexible pressure sensors have garnered significant attention and research due to their ultra-thin structure, lightweight nature, high flexibility, and conformability, making them highly promising for applications in human physiological signal monitoring, biomedical devices, and human-machine interfaces^[1−5]. To meet the requirements of practical applications, wearable flexible pressure sensors need to exhibit high sensitivity, particularly under low-pressure conditions, to detect subtle pressure changes caused by small activities such as heartbeats, pulses, and breathing^[6−10]. Furthermore, to mimic the sensing capabilities of human skin, a wide detection range is necessary^[11]. Based on their working principles, pressure sensors can be categorized into piezoresistive^[12−14], capacitive^{[15, 16]}, piezoelectric^{[17, 18]}, and triboelectric types^[19−22]. Among these, flexible piezoresistive sensors have received widespread attention due to their wide detection range, simple structure, and straightforward fabrication process. Meanwhile, electrospinning, characterized by its simplicity, efficiency, and cost-effectiveness, is well-suited for large-scale practical applications. The nanofiber membranes prepared by electrospinning exhibit high porosity, high aspect ratio, large specific surface area, low density, and low compressive modulus, making them ideal materials for the sensing layers of flexible piezoresistive sensors^[23].

MXene is a new type of 2D layered transition metal carbide, nitride, or carbonitride, represented by the general chemical formula M_n+1X_nTx (n = 1−3), where M is an early transition metal, X is carbon and/or nitrogen, and T represents functional groups (O, F, or OH)^[24−26]. MXene materials exhibit exceptional properties, including high electrical conductivity, large specific surface area, tunable interlayer spacing, and excellent mechanical performance. With adjustable electrical, optical, and electrochemical properties, MXene holds significant potential for applications in various fields, such as energy storage, catalysis, and sensing^[27−30]. In the domain of flexible pressure sensors, Ti₃C₂Tx is the most extensively studied MXene material. Its surface is rich in functional groups, enabling dispersion into other materials to form multifunctional composites, making it a promising candidate for high-performance sensor applications.

In this work, we developed a composite nanofiber membrane (CNM) based on MXene (Ti₃C₂Tx)/polyethylene oxide (PEO) using a simple and efficient electrospinning process. PEO possesses good water solubility, biocompatibility, and non-toxic properties, and can be processed into high-porosity, fully absorbable functional dressings, which have wide applications in biomedical fields. The flexible pressure sensor based on the MXene/PEO CNM demonstrated excellent sensitivity (44.34 kPa⁻¹ at 0−50 kPa), relatively short response/recovery times (150/45 ms), a low-pressure detection limit (6 Pa), and outstanding mechanical stability after more than 10 000 operational cycles. Due to its remarkable sensing performance, this flexible pressure sensor shows great potential for applications in health monitoring, motion detection, and human-machine interfaces.

To prepare the MXene/PEO solution, MXene was first dispersed in deionized water and ultrasonicated for 1 h to ensure uniform dispersion. Separately, poly(ethylene oxide) (PEO) was dissolved in deionized water with constant stirring at room temperature until fully dissolved, forming a 5 wt% PEO solution. The MXene dispersion was then gradually added to the PEO solution while stirring, resulting in a mixed solution containing 0.4 wt% MXene. The mixture was continuously stirred for 24 h until a homogeneous MXene/PEO solution was obtained. This solution was subsequently used in the electrospinning process to fabricate the composite nanofiber membrane (CNM).

The MXene/PEO CNM was fabricated using an electrospinning process. A high voltage of 15 kV was applied to the tip of the syringe needle, with the syringe flow rate set at 0.4 mL∙h⁻¹. The distance between the aluminum foil collector and the needle tip was maintained at 15 cm, and the ambient humidity was kept below 20%. The MXene/PEO mixed solution was then loaded into the syringe and electrospun to form the nanofiber membrane.

The prepared MXene/PEO CNM was utilized as the sensing layer for a piezoresistive pressure sensor, with polished Cu thin films serving as the electrodes. The sensor was then encapsulated using polyimide (PI) and polydimethylsiloxane (PDMS) on the top and bottom, respectively, forming a flexible pressure sensor. In the pulse wave monitoring system, the MXene/PEO-based flexible sensor was connected to a serpentine wire, signal processing circuit, and wireless transmission circuit. The signal processing circuit converted the change in resistance of the sensor under applied pressure into a corresponding voltage signal. This analog signal was then digitized by an analog-to-digital converter (ADC) and stored in a microcontroller unit (MCU). The wireless transmission circuit transmitted the data to a smart terminal for real-time monitoring and analysis.

The flexible pulse measurement system is primarily designed to achieve pulse wave signal acquisition, filtering, signal processing, and wireless transmission through the pressure sensor. At the core of the circuit control, a low-power ARM Cortex-M3 72 MHz microcontroller (MCU) manufactured by STMicroelectronics, model STM32F103T8U6TR, is employed. Its rich peripheral functionalities and high operating frequency fully meet the system’s circuit control requirements. For pressure sensing signal acquisition, a voltage divider method is used to convert the resistance changes of the sensor into voltage signals. These voltage signals are then digitized using the MCU’s built-in 12-bit analog-to-digital converter (ADC). A digital bandpass filtering algorithm is integrated within the MCU to obtain high signal-to-noise ratio (SNR) pulse wave signals. The MCU transmits the processed data via a universal synchronous/asynchronous receiver transmitter (USART) to a wireless Bluetooth transmission module (BT-11), which then forwards the data to a mobile device for processing and display. The system is powered by a 3.7 V lithium battery, with a voltage conversion chip (662 K) used to stabilize the output to 3.3 V, providing power to the system components.

Wearable flexible sensors can detect electrophysiological signals, physical signals, biochemical signals, and optoelectronic signals, and possess excellent conformability and high biocompatibility, making them ideal for monitoring key physiological signals of the human body^[31]. Pulse waves are a crucial component of cardiovascular information^{[32, 33]}. Studies have shown that quantitative analysis of pulse waves can provide cardiovascular health-related information, such as arterial blood pressure, heart rate, and vascular stiffness, offering a rapid and non-invasive method for diagnosing cardiovascular diseases^{[33, 34]}. Pressure sensing technology can effectively reconstruct the original waveform of pulse waves, while offering advantages such as high sensitivity and simple structure and fabrication processes. By integrating flexible pressure sensors with wireless transmitters, real-time monitoring and analysis of physiological signals can be achieved on smart devices, enabling precise monitoring and personalized services (Fig. 1(a)).

The wireless wearable pressure sensor based on MXene/PEO adopts a sandwich device structure, with a CNM prepared via electrospinning as the pressure-sensitive layer (Fig. 1(b)). Copper electrodes were used for the upper and lower layers, and the device was encapsulated using PI and PDMS substrates, respectively (for detailed parameters, refer to the experimental section). Scanning electron microscope (SEM) of the MXene/PEO CNM were shown in the enlarged image on the right side of Fig. 1(b) and Fig. S1. Fig. S2 showed the X-ray diffraction (XRD) plot of the MXene/PEO CNM. In healthcare applications such as physiological signal monitoring, the required pressure detection range spans several orders of magnitude. Traditional pressure sensors often struggle to combine a wide pressure detection range with high sensitivity, limiting their application across different scenarios. The CNM prepared via electrospinning exhibits high porosity, high aspect ratio, and low compressive modulus. When pressure is applied, the gap between the electrodes and the material is compressed, reducing contact resistance. Simultaneously, the MXene nanosheets on the nanofibers make contact under compression, decreasing the pathway resistance. The flexible pressure sensor based on MXene/PEO CNM demonstrates ultra-high sensitivity over a wide detection range, enhancing the reliability of physiological signal monitoring. Compared to existing pressure sensors, the MXene/PEO CNM-based flexible pressure sensor offers significant advantages in terms of detection range and sensitivity (Fig. 1(c))^{[11, 35−42]}.

A piezoresistive pressure sensor consists of an active layer made of pressure-sensitive material and two electrodes, which convert external pressure variations into changes in the device's resistance signal^[43]. The total resistance of the sensor is composed of the electrode resistance (R₁) and the resistance of the active layer (R₂)^[44]. For piezoresistive pressure sensors, the change in the sensor's resistance signal primarily depends on the variation in the resistance of the active layer. The resistance (R) of the piezoresistive pressure sensor can be calculated using the following formula:

$R=\rho L/A\mathrm{.}$

Here ρ is the resistivity, L is the length, and A is the cross-sectional area. In general, for flexible piezoresistive pressure sensors based on the piezoresistive effect, changes in resistance are mainly due to changes in resistivity. A schematic of the sensor structure is shown in Fig. 2(a). When the sensor is subjected to a small applied force, the MXene/PEO sensitive layer begins to deform. However, the contact resistance R₁ and the resistance R₂ of the MXene/PEO remain relatively high, resulting in a relatively low current in the circuit. When the sensor is subjected to a larger force, the electrodes make full contact with the MXene/PEO, significantly reducing the contact resistance. Simultaneously, the contact area within the MXene/PEO material increases under pressure, reducing the sensitive layer's resistance. The applied pressure also shortens the resistance path length, further decreasing the overall resistance of the sensitive layer and increasing the current. Fig. 2(b) shows that the MXene/PEO-based sensor can detect a minimum pressure of 6 Pa, capable of responding to subtle external deformations, which places it ahead of other sensors^{[21, 25]}. Fig. 2(c) displays the response test of the sensor under pressure levels ranging from 0.10 to 500 kPa. The application of repeated pressure at different levels (five times each) shows clear differentiation in pressure response, with stable results and no significant fluctuations.

Moreover, response time is another important parameter for sensors, as rapid, lag-free response ensures timely detection under external pressure. The response time of the MXene/PEO-based flexible pressure sensor under an applied pressure of 20 kPa is 150 ms, with a recovery time of 45 ms upon pressure release (Fig. 2(d)), which is attributed to the elastic deformation of the substrate, causing a slight delay in the response. Fig. 2(e) and 2(f) present the current−voltage (I−V) curves of the sensor under different pressures, demonstrating a wide and distinguishable pressure detection range for the MXene/PEO-based flexible pressure sensor. Compared to the initial no-pressure state, the sensor’s current shows a significant change upon the application of even slight pressure, indicating excellent sensitivity to low-pressure monitoring. Fig. 2(g) displays the sensitivity of the MXene/PEO-based flexible pressure sensor, defined as:

$S=\delta (\Delta I/I_0)/\delta p\mathrm{.}$

Here $\Delta I$ = I − I₀, I is the current under an external load, I₀ is the current without an external load, and P is the applied pressure^[5]. As pressure increases from 0.1 to 500 kPa, the current response increases in a segmented linear manner. The MXene/PEO pressure sensor exhibited sensitivities of 44.34 kPa⁻¹ (R² = 0.97) and 12.99 kPa⁻¹ (R² = 0.99) in the pressure ranges of 0.1−50 kPa and 50–500 kPa, respectively, as determined by the slopes in Fig. 2(g). This demonstrates high sensitivity across a wide detection range, establishing a foundation for its applications in human physiological signal detection and human−machine interface systems.

The fabrication process of the flexible pressure sensor based on MXene/PEO CNM proposed in this work is illustrated in Fig. 3(a). First, a mixed solution of MXene and PEO with the appropriate mass fraction is prepared and thoroughly stirred. Next, the MXene/PEO CNM is fabricated through electrospinning. Finally, copper is used as the electrode, and the device is encapsulated with PDMS and PI to obtain the complete pressure sensor. Detailed experimental steps can be found in the "Experimental Section". Fig. 3(b) and 3(c) are SEM images of the MXene/PEO CNM. The difference is that the mass fraction of PEO in Fig. 3(b) is 4% and the mass fraction of PEO in Fig. 3(c) is 5%. We tested the dynamic response of the flexible pressure sensor based on MXene (0.4 wt%)/PEO (4 wt%) CNM (Fig. S3). Due to the relatively increased mass fraction of MXene, the response current level of the sensor is significantly improved, but the $\Delta I/I_0$ of the device is significantly reduced. At 4 wt% PEO, the reduction in polymer matrix may lead to closer contact between MXene nanosheets, forming a more effective conductive network. This results in a higher absolute current response. However, excessive proximity or aggregation of MXene may lead to a saturation effect, that is, the response to the stimulus becomes less obvious, thereby reducing the relative rate of change. Multiple layers of MXene nanosheets are attached to the nanofibers, which form the backbone of the CNM, with fiber diameters ranging from 130 nm to 1.26 μm. The nanofibrous network structure of MXene/PEO, prepared by electrospinning, serves as the sensing layer of the sensor. When the sensor is subjected to external pressure, the nanofiber structure of MXene/PEO is compressed and deformed by force, which increases the contact area inside the material and reduces the resistivity, thereby reducing the contact resistance. On the other hand, the resistance length of the material is reduced, which reduces the resistance. MXene/PEO CNMs prepared by electrospinning enable high sensitivity over a wide pressure detection range.

In addition, we evaluated the sensing performance of the MXene/PEO-based flexible pressure sensor under various bending conditions to simulate the sensor's sensitivity and adaptability in detecting human motion. Four bending angles of 15°, 30°, 45°, and 60° were applied, resulting in distinguishable current signals for each angle. This result indicates that the sensor can be utilized for human motion detection and angle recognition in human-machine interfaces. To test the sensor's cycling stability, we conducted more than 10 000 cycles of loading and unloading (Fig. 3(e)). The enlarged inset in Fig. 3(e) clearly shows that there were no significant changes in current amplitude at the start, middle, and end points, demonstrating the excellent stability of the sensor and its promising potential for practical applications.

The MXene/PEO-based flexible pressure sensor demonstrates high sensitivity, a wide response range, and excellent stability, making it highly promising for real-time monitoring of human activities and physiological signals. Pulse waves are a critical source of information for monitoring cardiovascular health, and numerous studies have shown that quantitative analysis of pulse waves can provide valuable cardiovascular system information such as arterial blood pressure, heart rate, vascular stiffness, and blood flow velocity^[27]. We designed a flexible system for wireless monitoring of human pulse waves using the MXene/PEO-based pressure sensor. This system can be worn directly on the wrist for pulse detection and wirelessly transmit the signals to a smart device, offering high portability and flexibility (Fig. 4(a)). The pulse monitoring system consists of three main components: a flexible sensing electrode, a flexible signal processing system, and a smart terminal (Fig. S4). The flexible sensing and signal processing system has excellent bendability, allowing it to conform closely to the skin, enhancing both comfort and wearability.

The characteristic peak positions in pulse waveforms, such as the "P" (percussion), "T" (tidal), and "D" (dicrotic) waves, correspond to different phases of the cardiac cycle. The "P" and "T" peaks represent early and late systolic peaks, respectively, while the "D" wave appears during diastole^{[45, 46]}. During human pulse monitoring, the flexible sensing and signal processing system is attached to the wrist. Fig. 4(b) and 4(c) present the real-time wrist pulse signals recorded before and after exercise, with clearly identifiable characteristic peaks (i.e., percussion wave, tidal wave, and diastolic wave). Since the user's movements and differences in skin surface characteristics will affect the changes in pressure applied to the sensor, there will be certain baseline fluctuations in the tested pulse wave curve. At rest, the frequency of the pulse waveform is 61 Hz (Fig. 4(b)). After 10 min of aerobic exercise, the frequency of the pulse wave increased significantly, becoming 91 Hz (Fig. 4(c)). The sensor is capable of accurately capturing real-time pulse fluctuations and displaying distinct waveforms, making it suitable for disease diagnosis and medical monitoring applications.

In conclusion, a high-performance flexible piezoresistive pressure sensor based on MXene/PEO CNM was developed. The MXene/PEO CNM was fabricated via electrospinning, allowing for uniform attachment of MXene onto the nanofibers to form the pressure-sensitive layer. Owing to the high porosity and low compressive modulus of the sensitive layer, the sensor exhibited excellent sensing characteristics, including high sensitivity (44.34 kPa⁻¹ at 0−50 kPa, 12.99 kPa⁻¹ at 50–500 kPa), a wide pressure detection range (0–500 kPa), an ultra-low detection limit (~6 Pa), and outstanding stability (over 10 000 cycles). Furthermore, due to its high sensitivity over a broad pressure range, we designed a pulse monitoring system based on the MXene/PEO flexible pressure sensor, enabling real-time monitoring and recording of pulse waves. This demonstrates its immense potential in wearable medical devices and health monitoring applications.

Supplementary materials to this article can be found online at https://doi.org/10.1088/1674-4926/24110017.