Next Article in Journal
Development of a Tool for Verifying Leakage Detection in Microfluidic Systems
Next Article in Special Issue
MXene-Enhanced Laser-Induced Graphene Flexible Sensor with Rapid Response for Monitoring Pilots’ Body Motion
Previous Article in Journal
Design Parameters Affecting the Performance of Vortex-Induced Vibration Harvesters
Previous Article in Special Issue
Advances in 2D Molybdenum Disulfide Transistors for Flexible and Wearable Electronics
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Micromachines
Volume 16
Issue 2
10.3390/mi16020123
micromachines-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

MXene–MWCNT Conductive Network for Long-Lasting Wearable Strain Sensors with Gesture Recognition Capabilities

by
Fei Wang
,
Hongchen Yu
,
Xue Lv
,
Xingyu Ma
,
Quanlin Qu
,
Hanning Wang
,
Da Chen
* and
Yijian Liu
*
Laboratory for Intelligent Flexible Electronics, College of Electronic and Information Engineering, Shandong University of Science and Technology, Qingdao 266590, China
*
Authors to whom correspondence should be addressed.
Micromachines 2025, 16(2), 123; https://doi.org/10.3390/mi16020123
Submission received: 23 December 2024 / Revised: 16 January 2025 / Accepted: 20 January 2025 / Published: 22 January 2025
(This article belongs to the Special Issue New Advances in Wearable and Flexible Sensor Devices and Their Future Prospects)
Browse Figures
Versions Notes

Abstract

:
In this work, a conductive composite film composed of multi-walled carbon nanotubes (MWCNTs) and multi-layer Ti3C2Tx MXene nanosheets is used to construct a strain sensor on sandpaper Ecoflex substrate. The composite material forms a sophisticated conductive network with exceptional electrical conductivity, resulting in sensors with broad detection ranges and high sensitivities. The findings indicate that the strain sensing range of the Ecoflex/Ti3C2Tx/MWCNT strain sensor, when the mass ratio is set to 5:2, extends to 240%, with a gauge factor (GF) of 933 within the strain interval from 180% to 240%. The strain sensor has demonstrated its robustness by enduring more than 33,000 prolonged stretch-and-release cycles at 20% cyclic tensile strain. Moreover, a fast response time of 200 ms and detection limit of 0.05% are achieved. During application, the sensor effectively enables the detection of diverse physiological signals in the human body. More importantly, its application in a data glove that is coupled with machine learning and uses the Support Vector Machine (SVM) model trained on the collected gesture data results in an impressive recognition accuracy of 93.6%.

1. Introduction

Flexible strain sensors with outstanding biocompatibility, scalability, and real-time ability have developed rapidly over the past few decades [1,2,3]. These sensors possess wide-ranging applications [4,5,6], including physiological signal monitoring [7,8,9], electronic skins [10,11,12], wearable electronic devices [13,14,15], and soft robotics [16,17,18,19]. However, there are still limitations restricting the performance of flexible strain sensors. In order to achieve an ideal flexible strain sensor, researchers have explored various strategies to enhance their performance, in which materials with excellent conductivity and compatibility are taken into account, such as carbon nanomaterials [20,21], metal nanoparticles [22,23], nanowires [24,25], reduced graphene oxide (rGO) [26,27], and MXene [28,29]. Among these materials, MXene, which exhibits excellent mechanical and electrical properties, stands out as a promising candidate for flexible wearable strain sensors [30]. Due to the layered 2D structure, MXenes, especially the most utilized Ti3C2Tx, could achieve significant resistance variations with high-level sensitivity [31], but their flake-like structures impose limitations on the strain range in flexible sensors. For example, Ti3C2Tx-based sensors achieve an impressive gauge factor (GF) of 7000, but they are limited by a sensing range of merely 5% [32]. Another strain sensor based on multi-walled carbon nanotubes (MWCNTs) could achieve an extensive sensing range of 200%, but the GF of this particular sensor is relatively low at 29.4 [33]. The distinctive spiral structures of MWCNTs allow them to stretch and recover alongside flexible substrate materials during tension and release to enhance the strain sensing range. A common challenge is that pure materials often possess inherent drawbacks that hinder the simultaneous attainment of elevated sensitivity and an expansive sensing range in strain sensors. Fortunately, the combination of MXene and MWCNTs is a promising approach that exhibits noteworthy enhancements in both sensitivity and sensing range [34]. Another key point for ideal flexible strain sensors is the choice of substrate, which plays a significant role in sensor performance. For example, a flexible strain sensor based on a cotton substrate has a strain range of only 15% due to the limitation of cotton’s own stretching [35]. In sensor design, Ecoflex, which is a silica-like polymer widely used in biomedicine, can also be used as a substrate. It is attached to the surface of the human body and does not produce adverse reactions, indicating that it has good biocompatibility [36,37]. In addition, this good biocompatibility also ensures skin safety during long-term use, wherein the good tensile property can effectively improve the performance of the sensor. Furthermore, structural modifications to the substrate could be employed to further enhance the sensor’s sensitivity, such as foam-structured substrate, multi-layered porous structured substrate, and sandpaper-like Ecoflex substrate [38,39]. Therefore, researchers could exploit new opportunities to overcome the limitations via novel material combinations, substrate selection, and structural design, which could enhance the sensitivity and expand the sensing range effectively.
Several works have focused on these key points for high-performance flexible sensors. However, most research has focused on fabricating monolayers of Ti3C2Tx-MXene or modifying MWCNTs to improve the sensor’s performance [40]. Although material modification can effectively improve its properties, complex manufacturing methods restrict its rapid development and application. Furthermore, the multi-layer MXene structure exhibits superior capability in maintaining overall structural integrity under strain due to interlayer interactions. In flexible strain sensors, when subjected to stretching, bending, or other deformations, the multi-layer architecture effectively prevents material rupture and excessive deformation. While single-layer MXene may offer enhanced electrical performance, it is comparatively inferior in terms of mechanical stability and durability [41,42]. Moreover, the fabrication method involving multi-layer clay-like Ti3C2Tx combined with pristine MWCNTs is straightforward. However, detailed analysis of the mechanism of flexible strain sensors on Ti3C2Tx/MWCNTs and sandpaper-like Ecoflex substrates is scarce. Therefore, it is urgent to systematically study whether the working mechanism of the sensor can improve its performance, and it is equally important to find its potential application fields, which is of great significance for the development of flexible strain sensors in the future.
In this study, we combine Ti3C2Tx and MWCNTs to create a multi-material system, which was coated on sandpaper-like Ecoflex substrate to fabricate flexible strain sensors. It has been observed that the composite material establishes a sophisticated conductive network, which accounts for the sensor’s broad detection range and high sensitivity. Herein, the sensor with a 5:2 weight ratio of Ti3C2Tx/MWCNTs exhibits an astonishing 240% sensing range, a high sensitivity (GF = 933) within the 180% to 240% strain range, and outstanding cyclic stability over 33,000 cycles. This flexible strain sensor allows comfortable attachment to the skin, enabling effective human body detection of joint movements, facial expressions, and subtle pulse actions. Moreover, it has been successfully applied to a data glove when coupled with machine learning, achieving 93.6% recognition accuracy for ten different gestures, indicating that it can perform classification and recognition intelligently.

2. Experiment

2.1. Material

The MWCNTs (diameter = 3–15 µm, average length = 15–30 µm, and >95% purity) were provided by Hongdakang Evolution Technology Co., Ltd., Shenzhen, China. The MXene was provided by Guangdong Foshan Xinxi Technology Co., Ltd., Guangdong, China. The Ecoflex (00–30) was purchased from Smooth-on Co., Ltd., Macungie, PA, USA.

2.2. Instruments

The deionized water machine used was UPTA−20 from Lichen Instrument Technology Co., Ltd. (Zhejiang, China), and the ultrasonic cleaner used was JP−080S from Jiemeng Technology Co., Ltd. (Chongqing, China). The 12A spin-coater used was KW−4A from the crown electronic equipment factory. The desktop, electric, constant-temperature, blast-drying oven used was DHG−9123A from Shanghai Jinghong Equipment Co., Ltd. (Shanghai, China).

2.3. Preparation of Ecoflex

Ecoflex A and Ecoflex B were combined in a 1:1 ratio in a clean, prepared beaker and stirred thoroughly to ensure a homogeneous mixture. The mixture was transferred to a vacuum pump and subjected to vacuum degassing for 10 min to remove any entrapped air bubbles. Subsequently, the uncured Ecoflex colloid was applied onto a pristine glass slide using a spin-coating technique with 50 rad/min spinning speed for 10 s. Upon completion of the spin-coating procedure, the sample was cured at room temperature for future use.

2.4. Fabrication of Ecoflex/Ti3C2Tx/MWCNT Sensors

A mixture of Ti3C2Tx and MWCNTs was applied onto the surface of Ecoflex after stirring for 24 h using a magnetic stirrer. The composite material was carefully detached from the glass slide and cut into dimensions of 4 cm × 1 cm. Finally, the Ecoflex/Ti3C2Tx/MWCNT strain sensors were assembled by affixing copper foils, each with a width of 1 cm, to both ends. The experimental groups consisted of four sensors with different conductive materials: pure MWCNT, weight ratios of Ti3C2Tx:MWCNT = 5:2, weight ratios of Ti3C2Tx:MWCNT = 10:1, and pure Ti3C2Tx. All samples were subjected to magnetic stirring at 500 rpm for 24 h to fabricate strain sensors. The entire fabrication process of the Ecoflex/Ti3C2Tx/MWCNT strain sensor is illustrated in Figure 1a, where the scanning electron microscopy (SEM) images of the strain sensor with a weight ratio of Ti3C2Tx:MWCNT = 5:2 are presented. The SEM images reveal a uniform incorporation of Ti3C2Tx material and MWCNTs, where Ti3C2Tx and MWCNTs were interconnected in an orderly manner, forming a dense sensing layer. The fabricated strain sensor could be observed macroscopically to exhibit foldability, flexibility, and stretchability, as depicted in Figure 1b.

2.5. Characterization

The scanning electron microscopy (SEM) images were measured by using the TESCAN MIRA LMS (Brno, Czech Republic). The multifunctional computer-based tensile pressure testing machine (ZHIQU ZQ-990B, Dongwan, China) was used to control the strains applied on the Ecoflex/Ti3C2Tx/MWCNT strain sensor. The resistance change in the Ecoflex/Ti3C2Tx/MWCNT strain sensor was recorded with a high-precision digital multimeter (Rigol DM3068 6.5 Digits, Suzhou, China) and a data acquisition instrument (KeySight DAQ970A, Santa Rosa, CA, USA). The calculation formulas for relative resistance change (∆R/R0) and gauge factor (GF) were as follows:
Δ R R 0 = ( R − R 0 ) R 0
G F = ( R − R 0 ) R 0 ε
where R is the resistance to which strain is applied, R0 is the initial resistance of the sensor, and ε is the deformation of the sensor.

3. Results and Discussion

Figure 2a illustrates the relationship between the relative resistance change and strain obtained through tensile testing for sensors with varying Ti3C2Tx and MWCNT weight ratios. The pure Ti3C2Tx sensor has a limited sensing range of only 50%. When the weight ratio of Ti3C2Tx:MWCNT is 10:1, the strain range of the sensor can reach 160%. When Ti3C2Tx:MWCNT = 5:2, the strain range is even more than 240%. It can be seen that the content of Ti3C2Tx significantly affects the maximum strain, which increases with decreasing content of MXene due to the weak van der Waals forces between the rigid MXene nanosheets. Meanwhile, due to the addition of MWCNTs, MXene/MWCNT interfacial interactions promote the formation of a complex conductive network favoring more conductive paths with a wider strain range and higher sensitivity [43,44]. For device sensitivity, it is found that in Figure 2a, the pure MWCNT sensor exhibits a relative resistance change of only 170.71 at the maximum strain, whereas the Ti3C2Tx:MWCNT = 5:2 sensor demonstrates a relative resistance change of 431.28. Therefore, it is crucial to control the weight ratio appropriately for the highly beneficial composite effect to ensure the device possesses both wide strain range and high sensitivity.
In order to systematically elucidate the mechanism of performance improvement in the Ecoflex/Ti3C2Tx/MWCNT sensor, the schematic diagram and SEM images are displayed in Figure 2b–h. Figure 2b is plotted to clearly illustrate the fracture mechanism in the sensing layer. It depicts a microscale schematic diagram of the sensor undergoing deformation. Initially, Ti3C2Tx and MWCNTs are evenly dispersed on the substrate surface. When the sensor undergoes stretching, the Ti3C2Tx material initiates fracture, resembling the breakup of glaciers. However, due to their smaller size compared to Ti3C2Tx, the MWCNTs are able to fill the interstitial spaces between different Ti3C2Tx sheets, effectively serving as a bridge that connects these sheets, thereby furnishing additional conductive pathways. As a result, the device maintains conductivity, but the electrical conductivity gradually decreases during this process. With continued stretching, the interconnections between the Ti3C2Tx flakes through the MWCNTs gradually decrease, resulting in a rapid decline in electrical conductivity. With further stretching, stripping occurs between Ti3C2Tx and MWCNTs, resulting in loss of sensing performance. In general, because MWCNTs are effectively connected to Ti3C2Tx as a filling material, conductive path interconnection is achieved and high performance is obtained, resulting in the strain sensor having a high relative resistance change and sensing range. In order to quantitatively illustrate the conductivity maintained across cracks at various strain levels, the top-down SEM images of Ti3C2Tx:MWCNT = 5:2 and pure Ti3C2Tx sensors under different tensile conditions are taken, as shown in Figure 2c–e and Figure 2f–h respectively. It is found that the surface of the Ti3C2Tx:MWCNT = 5:2 sensor has a dense film at the initial position, while the surface of pure Ti3C2Tx exhibits the emergence of cracks owing to its flake-like structure. At a strain of 20%, the pure Ti3C2Tx sensor shows distinct cracking, whereas the cracks on the Ti3C2Tx:MWCNT = 5:2 sensor remain inconspicuous. Combined with the conductivity in Figure 2a, these results indicate that the MWCNTs act as filler materials across infinitesimal cracks, taking pivotal effect to maintain conductivity. Upon reaching a strain of 100%, small cracks appear on the Ti3C2Tx:MWCNT = 5:2 sensor, while the cracks on the pure Ti3C2Tx sensor are further enlarged. Under this circumstance, the MWCNTs serve not only as filler material but also as a bridge that connects the individual Ti3C2Tx nanosheets. Meanwhile, the Ti3C2Tx separates entangled MWCNT mutually on account of interfacial interactions [44], which further contributes to the formation of conductive paths to maintain conductivity across cracks. The composite of MWCNTs and Ti3C2Tx indeed enhances the connectivity of conductive pathways, ensuring that the device not only possesses high sensitivity but also exhibits a broad strain range.
To further enhance the sensitivity of the strain sensor, structural modifications are implemented on the Ecoflex substrate, incorporating a sandpaper-like microstructure. Overall, through consideration of the good strain range and sensitivity, the Ti3C2Tx:MWCNT = 5:2 sensor is selected for the subsequent experiment. As shown in Figure 3a, the relative resistance changes of the sandpaper-like structure under segmented tensile testing exhibit significantly higher changes compared with the smooth structure. Figure 3b shows the mean relative resistance changes and the sensitivities of sensors with sandpaper-like substrate and smooth substrate at different strain stages. The average and standard deviation values of five sets of data of each structure are calculated to obtain corresponding bar charts and error bars. It can be seen that the five groups of data are relatively concentrated, avoiding the chance of a single group of data. In addition, it can be seen from Figure 3b that the strain sensor with sandpaper structure has a higher relative resistance change at each strain stage. In the 0–100% strain range, the GF is 54.06. In the 100–180% strain range, the GF is 260.61. In the 180–240% strain range, the GF is up to 933. In addition, the linearity of the three stages is 98%, 90%, and 99%, respectively, indicating that the sensor has good linearity over the three different strain ranges. Compared with previous reports in Figure 3c, our sensor exhibits higher GF (933) and a wider sensing range (240%), which underscores the significant potential of this sensor for applications due to the high sensitivity and the ability to withstand large strains. To elucidate the mechanism of the sandpaper-like structure, the microstructure is depicted in Figure 3d. This microstructure is achieved by introducing a crack zone mechanism to create micro-sized peaks and valleys on the surface of the Ecoflex substrate, thereby complicating the conductive network and further enhancing the device’s sensitivity. An optical microscope image of the Ecoflex surface with the sandpaper-like structure is captured (Figure 3e), which validates the presence of peaks and valleys clearly. Similarly, the SEM image of the strain sensor surface is shown in Figure 3f, where the peaks and valleys are prominently visible. These observations effectively confirm that the sandpaper-like structure effectively complicates the conductive network to enhance sensitivity.
In addition, further sensing performance measurements are performed on the 5:2 Ti3C2Tx/MWCNT sensors with sandpaper-like Ecoflex substrates. It is found that in Figure 4a, the sensor consistently exhibits excellent and stable sensing performance during each stage of strain over three cycles at different strains, wherein a wide strain range is obviously achieved. Figure 4b illustrates the relative resistance change of the strain sensor at a strain of 1% and a stretching rate of 500 mm/min. It demonstrates that the strain sensor exhibits a response time of 200 ms, indicating a commendable response rate. This fast response enables the sensor to promptly detect and monitor changes in human physiological signals in real-time. Additionally, when it comes to detecting signals with small strains, the detection limitation becomes an important evaluation criterion, which reaches an impressive value of 0.05% in Figure 4c. Moreover, it is crucial for sensors to possess excellent cyclic stability and durability as these qualities indicate the device’s lifespan. Therefore, conducting tests to assess the cyclic stability of the sensor is performed in Figure 4d. The test results show that the sensor can undergo 33,000 cycles of continuous stretching at a 20% strain, clearly indicating its reliable stability, repeatability, and durability. Furthermore, there are 40 loading–unloading cycles captured between the 16,480th and 16,520th cycles, as shown in the inset in Figure 4d. Notably, the relative resistance change remains insignificant, indicating the sensor’s effective recovery to its initial state after each cycle and its consistent maintenance of sensing performance.
Moreover, durability tests are also performed under high temperatures and in wet environments, in which 1600 cycles are applied to research the sensor performance adequately. Since both the conductive material and the base material are heat-resistant materials, the sensor still maintains good stability and durability at 100 °C under high temperature heating, as shown in Figure 4e. In addition, due to the good packaging of the sensor, it can avoid the adverse effects of a wet environment. When the sensor is subjected to a two-hour exposure in a 60% relative humidity (RH) environment and subsequently evaluated for durability, it continues to exhibit superior stability.
Based on the above analysis, it becomes evident that this Ecoflex/Ti3C2Tx/MWCNT strain sensor with sandpaper-like Ecoflex substrate possesses outstanding comprehensive sensing performance, including wide strain range, high sensitivity, fast response time, minimal detection limit, and exceptional cyclic stability. Based on the above analysis, it becomes evident that this Ecoflex/Ti3C2Tx/MWCNT strain sensor with sandpaper-like Ecoflex substrate possesses outstanding comprehensive sensing performance, including wide strain range, high sensitivity, fast response time, minimal detection limit, and exceptional cyclic stability.
Table 1 lists the key parameters of several flexible sensors (including response time, detection limits, and durability). Compared to the above references, the Ecoflex/Ti3C2Tx /MWCNT strain sensor has a faster response time, a smaller detection limit, and a sensing repeatability of 33,000 times, which are indicative of the superior performance of the sensor in our work and prove the effectiveness of the designed structure.
These qualities render this Ecoflex/Ti3C2Tx/MWCNT strain sensor highly suitable for practical applications, such as the real-time monitoring of human vital signs and the capturing of minor body movements in wearable flexible sensors. Therefore, the impressive 240% strain range makes it highly effective in detecting joint bending movements. The collected signals accurately capture human body movements and physiological signals, as depicted in Figure 5. Figure 5a–d describes the relative resistance of the index finger, wrist, arm, and leg at flexion angles of 45° and 90°, wherein the results clearly demonstrate the sensor’s ability to accurately differentiate between these different body parts’ degrees of flexion while exhibiting remarkable stability. Notably, significant and regular changes in resistance response are detected at four different joint locations. These results strongly validate the sensor’s capacity to detect major joint bending movements in the human body. Moreover, the sensor exhibits an impressive ultra-low detection limit of 0.05%, which can be effectively used for subtle change or small strain detection in the human body. For instance, Figure 5e shows that the resistance changes regularly and steadily during repeated opening and closing of the mouth, which is measured with the sensor attached to the face. When smiling, the periodic change in resistance can also be clearly detected. Meanwhile, the resistance change while smiling is obviously less than that while opening the mouth, which coincides with the actual smile and mouth opening movement amplitudes. Thus, these types of subtle changes in the human body can be distinguished easily. Additionally, another potential application in healthcare involves small strain detection, which can be placed at the pulse point to monitor pulse signals. Pulse detection is conducted in Figure 5f, where pulse signals are acquired with three waves. The P, T, and D waves are distinguished and consistent with the waveform of the output signals in the relevant study [67], proving the effectiveness of our sensors in healthcare. The detected heart rate from the subject, approximately 75 beats per minute under normal conditions, is closely aligned with the rate detected by the sensor, underscoring the sensor’s high sensitivity and real-time detection capabilities, demonstrating excellent performance even in detecting minute strain variations. Consequently, this sensor holds immense potential for providing rapid and non-invasive diagnostic methods in the healthcare industry. Furthermore, the sensor is positioned on the throat to monitor changes during various pronunciations. The resistance change curves obtained during the pronunciation of “S D U S T” (Figure 5g) clearly distinguish the pronunciation of different letters. Additionally, the resistance change curves for the two instances of pronouncing “S” are nearly identical, indicating that the sensor can not only differentiate between different letter pronunciations but also classify similar pronunciations. This remarkable speech recognition ability of the sensor opens up significant applications in speech rehabilitation training and voice control. These results show that this sensor has a remarkable capability to detect both major joint movements and subtle changes in human body.
Considering the outstanding sensing performance and stability of the sensor, it can also be applied to data gloves assisted by a Support Vector Machine (SVM) model for the classification and recognition of different hand gestures; the diagrammatic sketch is shown in Figure 6a. During the experiment, the strain sensor is first attached to the glove and connected to the KeySight DAQ970A. Subjects make gestures ranging from “0” to “9”, and the ten different gestures are shown in Figure 6b. It is obvious that in Figure 6c, each gesture provides a unique waveform. The collected data are acquired, imported, and sliced. Each signal contains five sets of 50 data points (250 data points) for a total of 1000 signal sets (100 per gesture). Subsequently, the 1000 signals are randomly divided into a training set (700 signals) and a test set (300 signals). To achieve automatic recognition of different gestures, a computationally lightweight SVM model is employed. Feature peaks are segmented using a peak detection algorithm, and machine learning is performed using the simple SVM model. Ultimately, the classification accuracy of the training set reaches 93.6%, which is sufficient for practical applications. The confusion matrix of the SVM model’s prediction results is presented in Figure 6d. The prediction accuracy for each category displays promising results, indicating the effective application potential of the sensor in data gloves, which illustrates that this sensor can be utilized with machine learning for intelligent classification and recognition.
Table 2 lists a comparison of intelligent gesture recognition rates of several flexible sensors. It can be seen that the Ecoflex/Ti3C2Tx /MWCNT strain sensor is able to show an excellent recognition rate at a high number of gestures (ten) compared to the above references. It shows that the sensor has better potential in intelligent recognition applications.

4. Conclusions

In summary, we created a strain sensor incorporating multi-layered Ti3C2Tx and MWCNTs with a sandpaper-like Ecoflex substrate that exhibited a high sensitivity and wide detection range. The MXene–MWCNT conductive network enhanced the sensors’ performance effectively, wherein 5:2 weight ratio device achieved an impressive GF of 933 in a strain range of 180% to 240% and an impressive wide strain range of 240%. A rapid response time of 200 ms and a detection limit of 0.05% could also be obtained. Furthermore, the sensor demonstrated outstanding stability and sensing performance and remained stable even after 33,000 cycles at 20% strain, with no substantial increment in relative resistance. These remarkable sensing capabilities made the sensor highly suitable for applications involving human motion, including extensive joint movements, facial expressions, and data gloves. Our results revealed the mechanism underlying the high performance in Ti3C2Tx/MWCNT strain sensors with sandpaper-like Ecoflex substrates. We suggested potential applications of this sensor in human health monitoring, human–machine interaction, speech recognition, and soft robotics, which are significant for the future development of flexible strain sensors.

Author Contributions

Conceptualization, F.W. and H.Y.; methodology, H.Y.; software, X.L. and Q.Q.; validation, X.M. and H.W.; formal analysis, H.Y.; investigation, Y.L. and H.Y.; resources, F.W. and D.C.; writing—original draft preparation, Q.Q.; writing—review and editing, F.W. and D.C.; visualization, F.W. and Y.L.; supervision, F.W. and D.C.; project administration, D.C.; and funding acquisition, D.C. All authors have read and agreed to the published version of the manuscript.

Funding

This work was supported by the national natural science foundation of China (Nos. 62371275 and 62201328), the Shandong University Youth Innovation Supporting Program (No. 2019KJN020), the Taishan Scholar Engineering Construction Fund of Shandong Province of China (No. tsqn201812066), the Natural Science Foundation of Shandong Province (Nos. ZR2023MF009, ZR2021QF121, ZR2022QF104, ZR2021QH221, and ZR2022QB193), and the Natural Science Foundation of Qingdao (No. 23-2-1-155-zyyd-jch).

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Conflicts of Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

References

  1. Gao, Y.; Wang, Y.; Dai, Y.; Wang, Q.; Xiang, P.; Li, Y.; Gao, G. Amylopectin based hydrogel strain sensor with good biocompatibility, high toughness and stable anti-swelling in multiple liquid media. Eur. Polym. J. 2022, 164, 110981. [Google Scholar] [CrossRef]
  2. Liu, L.; Xu, C.N.; Yoshida, A.; Tu, D.; Ueno, N.; Kainuma, S. Scalable elasticoluminescent strain sensor for precise dynamic stress imaging and onsite infrastructure diagnosis. Adv. Mater. Technol. 2019, 4, 1800336. [Google Scholar] [CrossRef]
  3. Abramson, A.; Chan, C.T.; Khan, Y.; Mermin-Bunnell, A.; Matsuhisa, N.; Fong, R.; Shad, R.; Hiesinger, W.; Mallick, P.; Gambhir, S.S. A flexible electronic strain sensor for the real-time monitoring of tumor regression. Sci. Adv. 2022, 8, eabn6550. [Google Scholar] [CrossRef] [PubMed]
  4. Li, Z.; Qi, X.; Xu, L.; Lu, H.; Wang, W.; Jin, X.; Md, Z.I.; Zhu, Y.; Fu, Y.; Ni, Q. Self-repairing, large linear working range shape memory carbon nanotubes/ethylene vinyl acetate fiber strain sensor for human movement monitoring. ACS Appl. Mater. Interfaces 2020, 12, 42179–42192. [Google Scholar] [CrossRef]
  5. Liu, X.; Wei, Y.; Qiu, Y. Advanced flexible skin-like pressure and strain sensors for human health monitoring. Micromachines 2021, 12, 695. [Google Scholar] [CrossRef] [PubMed]
  6. Wang, F.; Yu, H.; Ma, X.; Lv, X.; Liu, Y.; Wang, H.; Wang, Z.; Chen, D. A Highly Sensitive Strain Sensor with Self-Assembled MXene/Multi-Walled Carbon Nanotube Sliding Networks for Gesture Recognition. Micromachines 2024, 15, 1301. [Google Scholar] [CrossRef] [PubMed]
  7. Shen, Z.; Liu, F.; Huang, S.; Wang, H.; Yang, C.; Hang, T.; Tao, J.; Xia, W.; Xie, X. Progress of flexible strain sensors for physiological signal monitoring. Biosens. Bioelectron. 2022, 211, 114298. [Google Scholar] [CrossRef] [PubMed]
  8. Wang, D.; Qin, L.; Yang, W.; He, Y.; Zhang, S.; Yang, Y.; Xu, K.; Gao, P.; Yu, J.; Cai, K. A conductive hydrogel based on gain and PVA/PAA/Fe3+ for strain sensor and physiological signal detection. ACS Appl. Polym. Mater. 2021, 3, 5268–5276. [Google Scholar] [CrossRef]
  9. Yin, F.; Li, X.; Peng, H.; Li, F.; Yang, K.; Yuan, W. A highly sensitive, multifunctional, and wearable mechanical sensor based on RGO/synergetic fiber bundles for monitoring human actions and physiological signals. Sens. Actuators B Chem. 2019, 285, 179–185. [Google Scholar] [CrossRef]
  10. Chen, M.; Hu, X.; Li, K.; Sun, J.; Liu, Z.; An, B.; Zhou, X.; Liu, Z. Self-assembly of dendritic-lamellar MXene/Carbon nanotube conductive films for wearable tactile sensors and artificial skin. Carbon 2020, 164, 111–120. [Google Scholar] [CrossRef]
  11. Wu, C.; Wang, H.; Li, Y.; Kim, T.; Kwon, S.J.; Park, B.; He, Z.; Lee, S.-B.; Um, M.-K.; Byun, J.-H. Sensitivity improvement of stretchable strain sensors by the internal and external structural designs for strain redistribution. ACS Appl. Mater. Interfaces 2020, 12, 50803–50811. [Google Scholar] [CrossRef] [PubMed]
  12. Yang, X.; Liu, J.; Fan, D.; Cao, J.; Huang, X.; Zheng, Z.; Zhang, X. Scalable manufacturing of real-time self-healing strain sensors based on brominated natural rubber. Chem. Eng. J. 2020, 389, 124448. [Google Scholar] [CrossRef]
  13. Li, Y.; Zhang, K.; Geng, Q.; Nie, M.; Wang, Q.; Huang, Z.; Wu, Z.; Pi, L. Helically intersected conductive network design for wearable electronic devices: From theory to application. ACS Appl. Mater. Interfaces 2021, 13, 11480–11488. [Google Scholar] [CrossRef]
  14. Yang, C.; Wu, Y.; Nie, M.; Wang, Q.; Liu, Y. Highly stretchable and conductive carbon fiber/polyurethane conductive films featuring interlocking interfaces. ACS Appl. Mater. Interfaces 2021, 13, 38656–38665. [Google Scholar] [CrossRef]
  15. Zhao, M.; Li, D.; Huang, J.; Wang, D.; Mensah, A.; Wei, Q. A multifunctional and highly stretchable electronic device based on silver nanowire/wrap yarn composite for a wearable strain sensor and heater. J. Mater. Chem. C 2019, 7, 13468–13476. [Google Scholar] [CrossRef]
  16. Wang, Y.; Qin, W.; Hu, X.; Liu, Z.; Ren, Z.; Cao, H.; An, B.; Zhou, X.; Shafiq, M.; Yin, S. Hierarchically buckled Ti3C2Tx MXene/carbon nanotubes strain sensor with improved linearity, sensitivity, and strain range for soft robotics and epidermal monitoring. Sens. Actuators B Chem. 2022, 368, 132228. [Google Scholar] [CrossRef]
  17. Pang, Y.; Xu, X.; Chen, S.; Fang, Y.; Shi, X.; Deng, Y.; Wang, Z.-L.; Cao, C. Skin-inspired textile-based tactile sensors enable multifunctional sensing of wearables and soft robots. Nano Energy 2022, 96, 107137. [Google Scholar] [CrossRef]
  18. Thuruthel, T.G.; Shih, B.; Laschi, C.; Tolley, M.T. Soft robot perception using embedded soft sensors and recurrent neural networks. Sci. Robot. 2019, 4, eaav1488. [Google Scholar] [CrossRef]
  19. Lin, J.-C.; Liatsis, P.; Alexandridis, P. Flexible and stretchable electrically conductive polymer materials for physical sensing applications. Polym. Rev. 2023, 63, 67–126. [Google Scholar] [CrossRef]
  20. Tang, J.; Wu, Y.; Ma, S.; Yan, T.; Pan, Z. Strain-sensing composite nanofiber filament and regulation mechanism of shoulder peaks based on carbon nanomaterial dispersion. ACS Appl. Mater. Interfaces 2023, 15, 7392–7404. [Google Scholar] [CrossRef]
  21. Sun, J.; Zhao, Y.; Yang, Z.; Shen, J.; Cabrera, E.; Lertola, M.J.; Yang, W.; Zhang, D.; Benatar, A.; Castro, J.M. Highly stretchable and ultrathin nanopaper composites for epidermal strain sensors. Nanotechnology 2018, 29, 355304. [Google Scholar] [CrossRef]
  22. Cheng, H.-W.; Yan, S.; Shang, G.; Wang, S.; Zhong, C.-J. Strain sensors fabricated by surface assembly of nanoparticles. Biosens. Bioelectron. 2021, 186, 113268. [Google Scholar] [CrossRef] [PubMed]
  23. Choi, Y.K.; Park, T.; Lee, D.H.D.; Ahn, J.; Kim, Y.H.; Jeon, S.; Han, M.J.; Oh, S.J. Wearable anti-temperature interference strain sensor with metal nanoparticle thin film and hybrid ligand exchange. Nanoscale 2022, 14, 8628–8639. [Google Scholar] [CrossRef]
  24. Liu, X.; Liang, X.; Lin, Z.; Lei, Z.; Xiong, Y.; Hu, Y.; Zhu, P.; Sun, R.; Wong, C.-P. Highly sensitive and stretchable strain sensor based on a synergistic hybrid conductive network. ACS Appl. Mater. Interfaces 2020, 12, 42420–42429. [Google Scholar] [CrossRef] [PubMed]
  25. Zhang, K.; Jiang, W.; Li, X.; Gao, X. Highly stretchable and sensitive strain sensors based on modified PDMS and hybrid particles of AgNWs/graphene. Nanotechnology 2022, 34, 06LT01. [Google Scholar] [CrossRef] [PubMed]
  26. Zhao, X.; Xu, L.; Chen, Q.; Peng, Q.; Yang, M.; Zhao, W.; Lin, Z.; Xu, F.; Li, Y.; He, X. Highly conductive multifunctional rGO/CNT hybrid sponge for electromagnetic wave shielding and strain sensor. Adv. Mater. Technol. 2019, 4, 1900443. [Google Scholar] [CrossRef]
  27. You, X.; Yang, J.; Wang, M.; Hu, J.; Ding, Y.; Zhang, X.; Dong, S. Graphene-based fiber sensors with high stretchability and sensitivity by direct ink extrusion. 2D Mater. 2019, 7, 015025. [Google Scholar] [CrossRef]
  28. Luan, H.; Zhang, D.; Xu, Z.; Zhao, W.; Yang, C.; Chen, X. MXene-based composite double-network multifunctional hydrogels as highly sensitive strain sensors. J. Mater. Chem. C 2022, 10, 7604–7613. [Google Scholar] [CrossRef]
  29. Yang, Y.; Cao, Z.; He, P.; Shi, L.; Ding, G.; Wang, R.; Sun, J. Ti3C2Tx MXene-graphene composite films for wearable strain sensors featured with high sensitivity and large range of linear response. Nano Energy 2019, 66, 104134. [Google Scholar] [CrossRef]
  30. Wu, L.; Xu, C.; Fan, M.; Tang, P.; Zhang, R.; Yang, S.; Pan, L.; Bin, Y. Lotus root structure-inspired Ti3C2-MXene-Based flexible and wearable strain sensor with ultra-high sensitivity and wide sensing range. Compos. Part A Appl. Sci. Manuf. 2022, 152, 106702. [Google Scholar] [CrossRef]
  31. Pu, J.-H.; Zhao, X.; Zha, X.-J.; Bai, L.; Ke, K.; Bao, R.-Y.; Liu, Z.-Y.; Yang, M.-B.; Yang, W. Multilayer structured AgNW/WPU-MXene fiber strain sensors with ultrahigh sensitivity and a wide operating range for wearable monitoring and healthcare. J. Mater. Chem. A 2019, 7, 15913–15923. [Google Scholar] [CrossRef]
  32. Kedambaimoole, V.; Kumar, N.; Shirhatti, V.; Nuthalapati, S.; Sen, P.; Nayak, M.M.; Rajanna, K.; Kumar, S. Laser-induced direct patterning of free-standing Ti3C2–MXene films for skin conformal tattoo sensors. ACS Sens. 2020, 5, 2086–2095. [Google Scholar] [CrossRef]
  33. Nie, M.; Ren, X.; Wen, L.; Han, L.; Wang, J.; Su, S. Highly sensitive and large range strain sensor based on synergetic effects with double conductive layer structures. Sens. Actuators A Phys. 2021, 318, 112515. [Google Scholar] [CrossRef]
  34. Kong, Y.; Zhang, M.; Zhan, X.; Cao, R.; Chen, J. Retention correlated read disturb errors in 3-D charge trap NAND flash memory: Observations, analysis, and solutions. IEEE Trans. Comput.-Aided Des. Integr. Circuits Syst. 2020, 39, 4042–4051. [Google Scholar] [CrossRef]
  35. Liu, L.; Wang, L.; Liu, X.; Yuan, W.; Yuan, M.; Xia, Q.; Hu, Q.; Zhou, A. High-performance wearable strain sensor based on MXene@ cotton fabric with network structure. Nanomaterials 2021, 11, 889. [Google Scholar] [CrossRef] [PubMed]
  36. Kim, S.; Kang, J.; Lee, I.; Jang, J.; Park, C.B.; Lee, W.; Bae, B.-S. An intrinsically stretchable multi-biochemical sensor for sweat analysis using photo-patternable ecoflex. npj Flex. Electron. 2023, 7, 33. [Google Scholar] [CrossRef]
  37. Zhao, G.; Wu, T.; Wang, R.; Li, Z.; Yang, Q.; Wang, L.; Zhou, H.; Jin, B.; Liu, H.; Fang, Y. Hydrogel-assisted microfluidic spinning of stretchable fibers via fluidic and interfacial self-adaptations. Sci. Adv. 2023, 9, eadj5407. [Google Scholar] [CrossRef]
  38. Wang, H.; Zhou, R.; Li, D.; Zhang, L.; Ren, G.; Wang, L.; Liu, J.; Wang, D.; Tang, Z.; Lu, G. High-performance foam-shaped strain sensor based on carbon nanotubes and Ti3C2T x MXene for the monitoring of human activities. ACS Nano 2021, 15, 9690–9700. [Google Scholar] [CrossRef]
  39. Gao, Y.; Yu, G.; Shu, T.; Chen, Y.; Yang, W.; Liu, Y.; Long, J.; Xiong, W.; Xuan, F. 3D-printed coaxial fibers for integrated wearable sensor skin. Adv. Mater. Technol. 2019, 4, 1900504. [Google Scholar] [CrossRef]
  40. Su, Y.; Ma, K.; Mao, X.; Liu, M.; Zhang, X. Highly compressible and sensitive flexible piezoresistive pressure sensor based on MWCNTs/Ti3C2Tx MXene@ melamine foam for human gesture monitoring and recognition. Nanomaterials 2022, 12, 2225. [Google Scholar] [CrossRef] [PubMed]
  41. Huang, Y.; Zhang, J.; An, L.; Yang, Y.; Wu, Y.; Liu, R.; Jin, Y.; Liu, Q.; Shi, Q.; Liang, Y. Synergetic effect of MXene/MoS2 heterostructure and gradient multilayer for highly sensitive flexible piezoelectric sensor. Polymer 2023, 286, 126399. [Google Scholar] [CrossRef]
  42. Guo, W.-Y.; Mai, T.; Huang, L.-Z.; Zhang, W.; Qi, M.-Y.; Yao, C.; Ma, M.-G. Multifunctional MXene conductive zwitterionic hydrogel for flexible wearable sensors and arrays. ACS Appl. Mater. Interfaces 2023, 15, 24933–24947. [Google Scholar] [CrossRef]
  43. Han, Z.; Niu, Y.; Shi, X.; Pan, D.; Liu, H.; Qiu, H.; Chen, W.; Xu, B.B.; El-Bahy, Z.M.; Hou, H. MXene@ c-MWCNT Adhesive Silica Nanofiber Membranes Enhancing Electromagnetic Interference Shielding and Thermal Insulation Performance in Extreme Environments. Nano-Micro Lett. 2024, 16, 195. [Google Scholar] [CrossRef] [PubMed]
  44. Dong, M.; Tomes, O.; Soul, A.; Hu, Y.; Bilotti, E.; Zhang, H.; Papageorgiou, D.G. Hybrid Ti3C2T x MXene and Carbon Nanotube Reinforced Epoxy Nanocomposites for Self-Sensing and Structural Health Monitoring. ACS Appl. Nano Mater. 2024, 7, 3314–3325. [Google Scholar] [CrossRef]
  45. Wu, H.; Liu, Q.; Du, W.; Li, C.; Shi, G. Transparent polymeric strain sensors for monitoring vital signs and beyond. ACS Appl. Mater. Interfaces 2018, 10, 3895–3901. [Google Scholar] [CrossRef]
  46. Wang, Y.; Wang, F.; Yazigi, S.; Zhang, D.; Gui, X.; Qi, Y.; Zhong, J.; Sun, L. Nanoengineered highly sensitive and stable soft strain sensor built from cracked carbon nanotube network/composite bilayers. Carbon 2021, 173, 849–856. [Google Scholar] [CrossRef]
  47. Yang, K.; Yin, F.; Xia, D.; Peng, H.; Yang, J.; Yuan, W. A highly flexible and multifunctional strain sensor based on a network-structured MXene/polyurethane mat with ultra-high sensitivity and a broad sensing range. Nanoscale 2019, 11, 9949–9957. [Google Scholar] [CrossRef]
  48. Li, Z.-Y.; Zhai, W.; Yu, Y.-F.; Li, G.-J.; Zhan, P.-F.; Xu, J.-W.; Zheng, G.-Q.; Dai, K.; Liu, C.-T.; Shen, C.-Y. An ultrasensitive, durable and stretchable strain sensor with crack-wrinkle structure for human motion monitoring. Chin. J. Polym. Sci. 2021, 39, 316–326. [Google Scholar] [CrossRef]
  49. Kong, W.-W.; Zhou, C.-G.; Dai, K.; Jia, L.-C.; Yan, D.-X.; Li, Z.-M. Highly stretchable and durable fibrous strain sensor with growth ring-like spiral structure for wearable electronics. Compos. Part B Eng. 2021, 225, 109275. [Google Scholar] [CrossRef]
  50. Bu, Y.; Shen, T.; Yang, W.; Yang, S.; Zhao, Y.; Liu, H.; Zheng, Y.; Liu, C.; Shen, C. Ultrasensitive strain sensor based on superhydrophobic microcracked conductive Ti3C2Tx MXene/paper for human-motion monitoring and E-skin. Sci. Bull. 2021, 66, 1849–1857. [Google Scholar] [CrossRef]
  51. Fu, X.; Li, L.; Chen, S.; Xu, H.; Li, J.; Shulga, V.; Han, W. Knitted Ti3C2Tx MXene based fiber strain sensor for human–computer interaction. J. Colloid Interface Sci. 2021, 604, 643–649. [Google Scholar] [CrossRef] [PubMed]
  52. Yan, T.; Wu, Y.; Tang, J.; Pan, Z. Highly sensitive strain sensor with wide strain range fabricated using carbonized natural wrapping yarns. Mater. Res. Bull. 2021, 143, 111452. [Google Scholar] [CrossRef]
  53. He, Y.; Wu, D.; Zhou, M.; Zheng, Y.; Wang, T.; Lu, C.; Zhang, L.; Liu, H.; Liu, C. Wearable strain sensors based on a porous polydimethylsiloxane hybrid with carbon nanotubes and graphene. ACS Appl. Mater. Interfaces 2021, 13, 15572–15583. [Google Scholar] [CrossRef] [PubMed]
  54. Xu, M.; Li, X.; Jin, C.; He, Z.; Zhang, Q. High-performance epidermal strain sensor based on macro-defect graphene foams. Sens. Actuators A Phys. 2020, 303, 111721. [Google Scholar] [CrossRef]
  55. Zhang, L.; Li, H.; Lai, X.; Gao, T.; Zeng, X. Three-dimensional binary-conductive-network silver nanowires@ thiolated graphene foam-based room-temperature self-healable strain sensor for human motion detection. ACS Appl. Mater. Interfaces 2020, 12, 44360–44370. [Google Scholar] [CrossRef] [PubMed]
  56. Tian, B.; Yao, W.; Zeng, P.; Li, X.; Wang, H.; Liu, L.; Feng, Y.; Luo, C.; Wu, W. All-printed, low-cost, tunable sensing range strain sensors based on Ag nanodendrite conductive inks for wearable electronics. J. Mater. Chem. C 2019, 7, 809–818. [Google Scholar] [CrossRef]
  57. Liu, Z.; Zheng, Y.; Jin, L.; Chen, K.; Zhai, H.; Huang, Q.; Chen, Z.; Yi, Y.; Umar, M.; Xu, L. Highly breathable and stretchable strain sensors with insensitive response to pressure and bending. Adv. Funct. Mater. 2021, 31, 2007622. [Google Scholar] [CrossRef]
  58. Li, Y.; Yang, L.; He, Z.; Liu, Y.; Wang, H.; Zhang, W.; Teng, L.; Chen, D.; Song, G. Low-Cost Data Glove based on deep-learning-enhanced flexible multiwalled carbon nanotube sensors for real-time gesture recognition. Adv. Intell. Syst. 2022, 4, 2200128. [Google Scholar] [CrossRef]
  59. Zhai, W.; Zhu, J.; Wang, Z.; Zhao, Y.; Zhan, P.; Wang, S.; Zheng, G.; Shao, C.; Dai, K.; Liu, C. Stretchable, sensitive strain sensors with a wide workable range and low detection limit for wearable electronic skins. ACS Appl. Mater. Interfaces 2022, 14, 4562–4570. [Google Scholar] [CrossRef] [PubMed]
  60. Chhetry, A.; Sharifuzzaman, M.; Yoon, H.; Sharma, S.; Xuan, X.; Park, J.Y. MoS2-decorated laser-induced graphene for a highly sensitive, hysteresis-free, and reliable piezoresistive strain sensor. ACS Appl. Mater. Interfaces 2019, 11, 22531–22542. [Google Scholar] [CrossRef] [PubMed]
  61. Chen, H.; Huang, J.; Liu, J.; Gu, J.; Zhu, J.; Huang, B.; Bai, J.; Guo, J.; Yang, X.; Guan, L. High toughness multifunctional organic hydrogels for flexible strain and temperature sensor. J. Mater. Chem. A 2021, 9, 23243–23255. [Google Scholar] [CrossRef]
  62. Sun, H.; Bu, Y.; Liu, H.; Wang, J.; Yang, W.; Li, Q.; Guo, Z.; Liu, C.; Shen, C. Superhydrophobic conductive rubber band with synergistic dual conductive layer for wide-range sensitive strain sensor. Sci. Bull. 2022, 67, 1669–1678. [Google Scholar] [CrossRef] [PubMed]
  63. Cheng, R.; Zeng, J.; Wang, B.; Li, J.; Cheng, Z.; Xu, J.; Gao, W.; Chen, K. Ultralight, flexible and conductive silver nanowire/nanofibrillated cellulose aerogel for multifunctional strain sensor. Chem. Eng. J. 2021, 424, 130565. [Google Scholar] [CrossRef]
  64. Huang, K.; Wu, Y.; Liu, J.; Chang, G.; Pan, X.; Weng, X.; Wang, Y.; Lei, M. A double-layer carbon nanotubes/polyvinyl alcohol hydrogel with high stretchability and compressibility for human motion detection. Eng. Sci. 2022, 17, 319–327. [Google Scholar]
  65. Ou, K.; Wang, M.; Meng, C.; Guo, K.; Emon, N.S.; Li, J.; Qi, K.; Dai, Y.; Wang, B. Enhanced mechanical strength and stretchable ionic conductive hydrogel with double-network structure for wearable strain sensing and energy harvesting. Compos. Sci. Technol. 2024, 255, 110732. [Google Scholar] [CrossRef]
  66. Chen, J.; Wang, F.; Zhu, G.; Wang, C.; Cui, X.; Xi, M.; Chang, X.; Zhu, Y. Breathable strain/temperature sensor based on fibrous networks of ionogels capable of monitoring human motion, respiration, and proximity. ACS Appl. Mater. Interfaces 2021, 13, 51567–51577. [Google Scholar] [CrossRef]
  67. Wang, S.; Xiao, P.; Liang, Y.; Zhang, J.; Huang, Y.; Wu, S.; Kuo, S.-W.; Chen, T. Network cracks-based wearable strain sensors for subtle and large strain detection of human motions. J. Mater. Chem. C 2018, 6, 5140–5147. [Google Scholar] [CrossRef]
  68. Hu, M.; He, P.; Zhao, W.; Zeng, X.; He, J.; Chen, Y.; Xu, X.; Sun, J.; Li, Z.; Yang, J. Machine learning-enabled intelligent gesture recognition and communication system using printed strain sensors. ACS Appl. Mater. Interfaces 2023, 15, 51360–51369. [Google Scholar] [CrossRef] [PubMed]
  69. Li, W.; Wu, S.; Li, S.; Zhong, X.; Zhang, X.; Qiao, H.; Kang, M.; Chen, J.; Wang, P.; Tao, L.-Q. Gesture recognition system using reduced graphene oxide-enhanced hydrogel strain sensors for rehabilitation training. ACS Appl. Mater. Interfaces 2023, 15, 45106–45115. [Google Scholar] [CrossRef] [PubMed]
  70. Zhou, Z.; Zhang, W.; Zhang, J.; Zhang, Y.; Yin, X.; He, B. Flexible and self-adhesive strain sensor based on GNSs/MWCNTs coated stretchable fabric for gesture monitoring and recognition. Sens. Actuators A Phys. 2023, 349, 114004. [Google Scholar] [CrossRef]
  71. Tao, L.-Q.; Liu, Y.; Zou, S.; Zhang, Z.; Zhao, X. Enhancing Gesture Recognition in Low-Light Conditions Using Laser-Induced Graphene Flexible Strain Sensors. IEEE Sens. J. 2023, 24, 5256–5265. [Google Scholar] [CrossRef]
  72. Sun, Y.; Huang, J.; Cheng, Y.; Zhang, J.; Shi, Y.; Pan, L. High-accuracy dynamic gesture recognition: A universal and self-adaptive deep-learning-assisted system leveraging high-performance ionogels-based strain sensors. SmartMat 2024, 5, e1269. [Google Scholar] [CrossRef]
  73. Peng, Y.; Wang, J.; Pang, K.; Liu, W.; Meng, J.; Li, B. A physiology-based flexible strap sensor for gesture recognition by sensing tendon deformation. IEEE Sens. J. 2021, 21, 9449–9456. [Google Scholar] [CrossRef]
  74. Gao, L.; Yang, J.; Zhao, Y.; Zhao, X.; Zhou, K.; Zhai, W.; Zheng, G.; Dai, K.; Liu, C.; Shen, C. Multilayer Bionic Tunable Strain Sensor with Mutually Non-Interfering Conductive Networks for Machine Learning-Assisted Gesture Recognition. Adv. Funct. Mater. 2024, 2416911. [Google Scholar] [CrossRef]
Micromachines 16 00123 g001
Figure 1. (a) The preparation of a Ecoflex/Ti3C2Tx/MWCNT strain sensor. The insert plots in the lower left corner show the top SEM images of the Ecoflex/Ti3C2Tx/MWCNT strain sensor at magnifications of 200 μm. (b) The folding, twisting, and stretching states of the Ecoflex/Ti3C2Tx/MWCNT strain sensor.
Figure 1. (a) The preparation of a Ecoflex/Ti3C2Tx/MWCNT strain sensor. The insert plots in the lower left corner show the top SEM images of the Ecoflex/Ti3C2Tx/MWCNT strain sensor at magnifications of 200 μm. (b) The folding, twisting, and stretching states of the Ecoflex/Ti3C2Tx/MWCNT strain sensor.
Micromachines 16 00123 g001
Micromachines 16 00123 g002
Figure 2. (a) The relationship between the resistance change and deformation characteristics of strain sensors with different doping ratios. (b) Fracture mechanism of the Ecoflex/Ti3C2Tx/MWCNT strain sensor. (c–e) Top-view SEM images of the Ti3C2Tx:MWCNT = 5:2 strain sensor during stretching. (f–h) Top-view SEM images of the pure Ti3C2Tx strain sensor during stretching.
Figure 2. (a) The relationship between the resistance change and deformation characteristics of strain sensors with different doping ratios. (b) Fracture mechanism of the Ecoflex/Ti3C2Tx/MWCNT strain sensor. (c–e) Top-view SEM images of the Ti3C2Tx:MWCNT = 5:2 strain sensor during stretching. (f–h) Top-view SEM images of the pure Ti3C2Tx strain sensor during stretching.
Micromachines 16 00123 g002
Micromachines 16 00123 g003
Figure 3. (a) Real-time response curves of sandpaper-like substrate and smooth substrate sensors under 0–240% step strain. (b) Sensitivities of sandpaper-like substrate and smooth substrate sensors at different strain stages. (c) A comparison of the highest gauge factor and the maximum working range of the strain sensors with that of the previously reported strain sensors [45,46,47,48,49,50,51,52,53,54,55,56,57,58]. (d) A microscopic diagram of the sandpaper structure. (e) A light microscope image of an Ecoflex substrate with a sandpaper structure. (f) An SEM image of the surface of the Ecoflex/Ti3C2Tx/MWCNT sensor with a sandpaper structure.
Figure 3. (a) Real-time response curves of sandpaper-like substrate and smooth substrate sensors under 0–240% step strain. (b) Sensitivities of sandpaper-like substrate and smooth substrate sensors at different strain stages. (c) A comparison of the highest gauge factor and the maximum working range of the strain sensors with that of the previously reported strain sensors [45,46,47,48,49,50,51,52,53,54,55,56,57,58]. (d) A microscopic diagram of the sandpaper structure. (e) A light microscope image of an Ecoflex substrate with a sandpaper structure. (f) An SEM image of the surface of the Ecoflex/Ti3C2Tx/MWCNT sensor with a sandpaper structure.
Micromachines 16 00123 g003
Micromachines 16 00123 g004
Figure 4. (a) The cycling relative resistance variations of the strain sensors with different strains. (b) The response time under the strain of 1%. (c) The relative resistance changes as a function of time under a minimal strain of 0.05%. (d) The long-term durability test of the Ecoflex/Ti3C2Tx /MWCNT strain sensor with 33,000 stretch-and-release cycles under a 20% strain. The insert plots show the details of the 16,480–16,520 cycles. (e) Durability test of the Ecoflex/Ti3C2Tx /MWCNT strain sensor at 20% strain after heating at 100 °C 1 h. (f) Durability test of the Ecoflex/Ti3C2Tx /MWCNT strain sensor at 20% strain after 2 h at 60%RH.
Figure 4. (a) The cycling relative resistance variations of the strain sensors with different strains. (b) The response time under the strain of 1%. (c) The relative resistance changes as a function of time under a minimal strain of 0.05%. (d) The long-term durability test of the Ecoflex/Ti3C2Tx /MWCNT strain sensor with 33,000 stretch-and-release cycles under a 20% strain. The insert plots show the details of the 16,480–16,520 cycles. (e) Durability test of the Ecoflex/Ti3C2Tx /MWCNT strain sensor at 20% strain after heating at 100 °C 1 h. (f) Durability test of the Ecoflex/Ti3C2Tx /MWCNT strain sensor at 20% strain after 2 h at 60%RH.
Micromachines 16 00123 g004
Micromachines 16 00123 g005
Figure 5. Relative resistance changes in the Ecoflex/Ti3C2Tx/MWCNT strain sensor attached on the finger (a), wrist (b), arm (c), and leg (d). (e) Resistance responses of the Ecoflex/Ti3C2Tx/MWCNT strain sensor in the smiling and open mouth scenarios. (f) Pulse signal measured by the Ecoflex/Ti3C2Tx/MWCNT strain sensor. (g) The sensing performance of the Ecoflex/Ti3C2Tx/MWCNT strain sensor recorded while speaking “S D U S T”.
Figure 5. Relative resistance changes in the Ecoflex/Ti3C2Tx/MWCNT strain sensor attached on the finger (a), wrist (b), arm (c), and leg (d). (e) Resistance responses of the Ecoflex/Ti3C2Tx/MWCNT strain sensor in the smiling and open mouth scenarios. (f) Pulse signal measured by the Ecoflex/Ti3C2Tx/MWCNT strain sensor. (g) The sensing performance of the Ecoflex/Ti3C2Tx/MWCNT strain sensor recorded while speaking “S D U S T”.
Micromachines 16 00123 g005
Micromachines 16 00123 g006
Figure 6. (a) A conceptual diagram of the designed data glove. (b) An actual image of the data glove displaying “0”–“9,” representing ten different gestures. (c) The resistance change waveforms collected for the “0”–“9” gestures. (d) The confusion matrix of the prediction results using the SVM model.
Figure 6. (a) A conceptual diagram of the designed data glove. (b) An actual image of the data glove displaying “0”–“9,” representing ten different gestures. (c) The resistance change waveforms collected for the “0”–“9” gestures. (d) The confusion matrix of the prediction results using the SVM model.
Micromachines 16 00123 g006
Table 1. Compare the performance of relevant flexible strain sensors (response time, detection limit, and durability).
Table 1. Compare the performance of relevant flexible strain sensors (response time, detection limit, and durability).
CompositionResponse TimeDetection LimitDurabilityRef.
CB/PANIP/TPU 0.03%10,000[59]
MDS-LIG/PDMS250 ms 12,000[60]
PVA/DMSO/rGO/GO250 ms 4000[61]
CNTs/rGO/PDA/HF-SIO2200 ms0.1%1000[62]
AgNWs/NFC180 ms0.5%10,000[63]
CNTs/PVA508 ms 1000[64]
PAM/CA/EtOH30340 ms 200[65]
IL/TPU 0.1%1000[66]
Ecoflex/Ti3C2Tx /MWCNTs200 ms0.05%33,000This work
Table 2. A comparison of the intelligent gesture recognition rates of other flexible strain sensors.
Table 2. A comparison of the intelligent gesture recognition rates of other flexible strain sensors.
CompositionNumber of GesturesMethodRecognition RateRef.
CNT/graphene/PDMS10SVM91.6%[68]
PAM/SA/TG91D CNN83.3%[69]
GNSs/MWCNTs5LSTM95%[70]
Graphene10MobileNet CNN90%[71]
Ionogels10DCNN93.66%[72]
Graphene/PET12SVM84.7%[73]
CNTs/graphene/TPU6KNN95.7%[74]
Ecoflex/Ti3C2Tx /MWCNT10SVM93.66%This work
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Wang, F.; Yu, H.; Lv, X.; Ma, X.; Qu, Q.; Wang, H.; Chen, D.; Liu, Y. MXene–MWCNT Conductive Network for Long-Lasting Wearable Strain Sensors with Gesture Recognition Capabilities. Micromachines 2025, 16, 123. https://doi.org/10.3390/mi16020123

AMA Style

Wang F, Yu H, Lv X, Ma X, Qu Q, Wang H, Chen D, Liu Y. MXene–MWCNT Conductive Network for Long-Lasting Wearable Strain Sensors with Gesture Recognition Capabilities. Micromachines. 2025; 16(2):123. https://doi.org/10.3390/mi16020123

Chicago/Turabian Style

Wang, Fei, Hongchen Yu, Xue Lv, Xingyu Ma, Quanlin Qu, Hanning Wang, Da Chen, and Yijian Liu. 2025. "MXene–MWCNT Conductive Network for Long-Lasting Wearable Strain Sensors with Gesture Recognition Capabilities" Micromachines 16, no. 2: 123. https://doi.org/10.3390/mi16020123

APA Style

Wang, F., Yu, H., Lv, X., Ma, X., Qu, Q., Wang, H., Chen, D., & Liu, Y. (2025). MXene–MWCNT Conductive Network for Long-Lasting Wearable Strain Sensors with Gesture Recognition Capabilities. Micromachines, 16(2), 123. https://doi.org/10.3390/mi16020123

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop