A self-powered human gait monitoring sensor for osteoarthritis prevention Open Access

With the increasing aging population, medical and health diagnosis has posed new and severe challenges to current medical devices and healthcare personnel.^1–3 The flexible wearable sensing technology based on the Internet of things (IoTs) provides a new technological foundation for medical diagnosis and research, promoting the process of medical intelligence.^4,5 Wearable sensors can recognize various postures of human motion and provide feedback on these posture features through electrical signals.⁶ Thus, wearable sensors for monitoring human motion posture are increasingly receiving attention from the medical industry. Currently, human pose recognition mainly relies on video motion recognition technology, and sensor-based motion recognition technology can overcome fuzzy points in video recognition and reduce the cost of later data processing.^7,8 Wearable posture sensors can have potential applications in medical diagnosis and analysis, such as gait analysis of patients with osteoarthritis.^9,10 Early diagnosis and intervention in the early stage of joint cartilage regeneration can maximize the delay of progression and prevent the occurrence of irreversible lesions. These sensors can provide real-time and accurate monitoring information but still rely on conventional battery power and face challenges with human-friendly materials. In addition, abandoned chemical batteries pollute the environment, and their limited lifespan requires cyclic charging, making it hard to meet the needs of sensors and other devices that require long-term operation.^11–13 Hitherto, breakthroughs have been made in energy harvesting technologies, including piezoelectric,¹⁴ optoelectronic,¹⁵ thermoelectric,¹⁶ electromagnetic,¹⁷ and triboelectric.¹⁸ Among them, triboelectric nanogenerator (TENG) can harvest various low-frequency mechanical energies and has received widespread attention in recent years.^19–31 TENG devices can convert mechanical energy (walking, running, and touching) into continuous electrical energy, and the energy obtained from human motion is green, safe, and renewable.^32,33 Due to the human compatibility, high performance, and a wide range of preparation materials, TENG devices have promising applications in flexible wearable sensors, energy harvesters, and medical-assisted diagnosis.³⁴ 

The flexible TENG devices also have applications in human motion behavior sensing, such as human gait, finger joint, and elbow joint movements.³⁵ Due to the wide selection of preparation materials, TENG devices can be prepared from various flexible commercial materials, which provides an effective way for their large-scale production.³⁶ Compared to other sensors (piezoresistive sensor³⁷ and capacitive sensor³⁸), TENG sensors have substantial environmental adaptability and a fast response rate, and can be applied to monitor various complex human movements. Most importantly, they do not require an additional power supply. TENG sensors can monitor the gait of patients with osteoarthritis and recognize abnormal human behaviors and changes. Due to the fact that the contact electrification mechanism can occur between any two materials, the exploration of new triboelectric materials is an important direction and hot topic in TENG device research. In addition, surface roughness and triboelectric material thickness are two factors that influence the TENG device performance.³⁹ Moreover, the film preparation technology requires professional synthesis technology and special equipment, which limits the application scenarios of TENG devices. A promising approach is to explore commonly used commercial materials in daily life, use them for TENG device preparation, and explore the application fields of TENG to promote the development of TENG devices. In recent years, significant progress has been made in related research. In 2019, Xia et al. first reported a commercial cosmetic fixing powder as a triboelectric material for fabricating a flexible TENG device, bringing a new idea to the triboelectric material selection.⁴⁰ In 2021, Xia and Xu further proposed a novel TENG based on facial mask as a mechanical energy harvester and flexible touch sensor, resulting in a new perspective on TENG research based on cosmetics.⁴¹ Hence, exploring the application of cosmetics in triboelectric materials is of great significance, and there is still a research gap to be filled. Meanwhile, expanding the application of flexible TENG devices in the medical monitoring field is also a potential pathway to promote the development of TENG devices .

Hence, we used eyebrow powder to fabricate a triboelectric nanogenerator (E-TENG) for bio-mechanical energy harvesting and gait monitoring of patients with osteoarthritis. In detail, the eyebrow power layer following the sample single process is the triboelectric part with a polytetrafluoroethylene (PTFE) film. Under a maximum separation distance (5 mm) and a maximum motion frequency (6 Hz), the E-TENG device can attain a open-circuit voltage (V_oc) of 169 V and a short-circuit current (I_sc) of 5.5 µA. Meanwhile, the transfer charge (Q_sc) of E-TENG can reach 20.8 nC, and the maximum output power of E-TENG can arrive at 175 µW (load resistance: 20 MΩ), which indicates the potential application of powering low-power electronic devices. From the results, after 15 000 consecutive operations, the E-TENG can still maintain a stable output state, which also provides a feasibility for long-term work of the E-TENG in human gait monitoring. The E-TENG can detect various human gait patterns (walking, running, and jumping), finger motion, and elbow joint movements. Further research has shown that the E-TENG can be used for gait recognition and monitoring in patients with osteoarthritis, providing reference data for osteoarthritis prevention and treatment.

Significantly, the function of eyebrow powder is to increase the weight of the eyebrows, which can create a hazy eyebrow shape that is natural and suitable for daily makeup use. Simultaneously, the contact electrification mechanism has been proven by relevant research to occur between any two materials,³⁹ which brings new ideas to utilize eyebrow powder to prepare triboelectric sensors. Accordingly, we propose using eyebrow powder as a novel triboelectric material for human gait monitoring and analysis, and this sensing application can serve as a tool for preventing osteoarthritis [Fig. 1(a)]. The E-TENG consists of three parts: triboelectric functional layers (PTFE layer@eyebrow powder layer), a conductive electrode (copper tape), and a supporting structure (Kapton film). The simple and efficient preparation process provides a convenient path for the E-TENG application, as demonstrated in Fig. 1(b). Concretely, two Kapton films were cut as the substrate and support components of the E-TENG device [Fig. 1(b-1)]. Next, the conductive copper tape was applied to the surfaces of the two Kapton films, with one piece of copper tape facing upward [Figs. 1(b-2) and 1(b-3)]. Then, a layer of eyebrow powder was applied on the surface of the conductive copper foil with the adhesive facing upward [Fig. 1(b-3)]. Then, a piece of PTFE tape was stuck on the surface of another piece of copper foil [Fig. 1(b-5)].

There are three different colors of eyebrow powder, and here, we will code these three colors as Nos. (1), (2), and (3), as illustrated in Fig. 1(c). The brush can be an effective tool for preparing the eyebrow powder layer, which provides fast single-step preparation technology. Figure 1(d) displays three pieces of eyebrow powder layer with different colors, and a coin is used to reflect the sample size. Figure 1(e) shows the picture of the E-TENG device (size: 2 × 2 cm²). To demonstrate the distribution characteristics of eyebrow powder on the adhesive surface, we use an energy dispersive spectrometer (EDS) to analyze the elemental distribution in the No. (1) eyebrow powder layer [Figs. 1(f-1)–1(f-4)]. The test results indicate that the eyebrow powders evenly distributed on the adhesive surface, resulting in a high efficiency triboelectric contact between the eyebrow powder layer and PTFE layer.

The elemental spectrogram of No. (1) eyebrow powder layer indicates the element type of No. (1) eyebrow powder (Fig. S1, supplementary material). According to the results (Table S1, supplementary material), the content of C, O, and Si in No. (1) eyebrow powder is relatively high. Compared to No. (1) eyebrow powder, the Na element has been added to No. (2) eyebrow powder, while the K element is missing (Fig. S2, supplementary material). In addition, the content of the Fe element increased from 2.69 to 8.67% (Table S2, supplementary material). As for No. (3) eyebrow powder (Fig. S3, supplementary material), the Ti element has been added. Meanwhile, the content of the Fe element increased to 18.84%, which is the main reason for the change in the color of the eyebrow powder (Table S3, supplementary material). Moreover, a mechanical exciter is used to produce mechanical excitation with various amplitudes and frequencies. An electrometer (Keithley 6517) is used to measure the output voltage, output current, and transfer charge of the E-TENG.

Many polymer materials have been designed to form the triboelectric part of TENG devices, including nylon, PET films, silk fibers, and various composite materials. They have excellent triboelectric properties and flexibility but face complex processes and professional technical challenges in the surface texture treatment and preparation of triboelectric layers. Hence, we proposed a single painting process for preparing the eyebrow powder layer by using a brush here, and it displays tremendous potential for use in the medical diagnosis industry. The single-painting process represents a high efficiency and low cost, and the triboelectric material of eyebrow powder also reveals skin compatibility and high adherence. A multiple painting technology specifically demonstrates the preparation process of the eyebrow powder layer, as shown in Figs. 2(a-1)–2(a-3). As the number of paintings increases, the distribution density of eyebrow powder on the adhesive surface will gradually increase. Finally, it will be densely distributed on the glue surface [Fig. 2(b)]. Meanwhile, strengthening methods, such as pressing, are used to increase the firmness of the eyebrow powder, and a hair dryer can be used to remove the excess eyebrow powder to obtain a uniform and stable eyebrow powder layer. The influences of painting cycles on the quantity of eyebrow powder and the electrical output (V_oc, I_sc, and Q_sc) of the prepared E-TENG were determined, as demonstrated in Figs. 2(c) and 2(d). The sample size in the comparative experiment remains constant, and the average value is taken through multiple experiments for comparison. According to the results, when the painting cycles improved from 1 to 8, the eyebrow powder quantity on the adhesive surface improved from 4.5 to 11.9 mg. The output voltage varies across different painting cycles, suggesting the presence of optimal painting cycles that result in the highest electrical output. From the results, when painting cycles reach 6, the output voltage reaches its maximum. When the painting cycles exceed 6, the output voltage decreases. The reason behind this trend is that when the number of painting cycles is less than 6, with an increase in painting cycles, the density of eyebrow powder distributed on the adhesive surface gradually rises. Consequently, the effective triboelectric area also expands, resulting in an enhancement of the output performance. However, as the painting cycles increase, the thickness of the eyebrow powder layer increases, which hinders the improvement of the output performance.

Considering the influence of eyebrow powders with different colors on the output performance, we determined the surface micro-textures of three eyebrow powder layers. Based on the scanning electron microscopy (SEM) images shown in Figs. 2(e-1)–2(e-3), it is evident that all three eyebrow powders possess a sheet-like microstructure, with No. (3) eyebrow powder demonstrating a higher level of delicacy. Further testing has shown that the E-TENG prepared from No. (3) eyebrow powder can achieve the highest electrical output, as shown in Figs. 2(f)–2(h), since a more delicate surface texture can bring more sufficient triboelectric area. Moreover, the V_oc of the E-TENG based on No. (1), No. (2), and No. (3) eyebrow powders can reach 36.49, 39.38, and 44.71 V, respectively. The Q_sc of the E-TENG based on No. (1), No. (2), and No. (3) eyebrow powders can arrive at 19.35, 20.5, and 23.7 nC, respectively. The I_sc of the E-TENG based on No. (1), No. (2), and No. (3) eyebrow powders can arrive at 1.61, 1.9, and 2.19 µA, respectively. According to elemental analysis, the deeper the color of the eyebrow powder, the higher the Fe element in the eyebrow powder. Obviously, the content of the Fe element is also the main factor causing the difference in triboelectric charges. A higher content of the Fe element will lead to significant differences in the morphological characteristics of the eyebrow powder, leading to a higher accumulation effect of frictional charges. Thus, this research used No. (3) eyebrow powder as the research object.

Considering the impact of different polymer films on the output performance of the E-TENG, we tested the output performance of the E-TENG when the eyebrow powder layer and different polymer films form the triboelectric pair (Fig. S4, supplementary material). The experimental results indicate that the friction electric pair composed of the PTFE film and the eyebrow powder layer produces the highest output performance for E-TENG. Figure 3(a) illustrates the working mechanism of the E-TENG. In the initial state [Fig. 3(a-1)], the eyebrow powder layer and PTFE layer are in a separate state, and the two surfaces do not carry any charges. Under external forces, the eyebrow powder layer and PTFE layer are pressed and electrons are transferred from the eyebrow powder layer to the PTFE layer during the triboelectric contact process [Fig. 3(a-2)]. When the external force acting on the E-TENG device is withdrawn, the PTFE layer and eyebrow powder layer separate, and electrons flow in the driving circuit and generate current [Fig. 3(a-3)]. Until the separation distance between the PTFE layer and the eyebrow powder layer reaches its maximum, the electrons in the circuit are not flowing [Fig. 3(a-4)]. When the surfaces of the PTFE layer and eyebrow powder layer approach again, electrons in the circuit will flow in the opposite direction and produce a reverse current [Fig. 3(a-5)]. When the E-TENG device is installed in shoes, a continuous reciprocating driving force is produced during human movements, causing the E-TENG device to generate alternating current signals for gait monitoring in patients with osteoarthritis.

Furthermore, to examine the effect of motion frequency on the electrical output, we set the maximum motion distance to 5 mm and measured indicators such as V_oc, I_sc, and Q_sc at different motion frequencies to evaluate the output performance. Figure 3(b) illustrates the V_oc at various motion frequencies (2–6 Hz), and the stable trend of the open circuit voltage can ignore the motion frequency. However, the increasing frequency of motion will drive electrons in the circuit to transfer at a higher speed, bringing an increase in the I_sc of the E-TENG device [Fig. 3(c)]. Similar to the trend of V_oc variation, the increasing frequency cannot influence the stable tendency of Q_sc of the E-TENG [Fig. 3(d)]. Furthermore, when the frequency arrives at 2 Hz, the E-TENG exhibits an upward trend in the output performance with an increase in the separation distance (1–5 mm), as depicted in Figs. 3(e)–3(g). A higher separation distance causes a higher potential difference between the triboelectric layers, resulting in a higher output performance.

In addition, we determined the output performance under a maximum separation distance (5 mm) and a maximum motion frequency (6 Hz). The E-TENG can achieve a V_oc of 169 V and a I_sc of 5.5 µA, which is comparable to the output performance of previous TENG devices based on cosmetics^40,41 [Figs. 4(a) and 4(b)]. Moreover, the Q_sc of the E-TENG is 20.8 nC, as shown in Fig. 4(c). Since the E-TENG needs to be installed on the human body for medical diagnosis, a sustainable output from E-TENG devices is crucial. According to reliability testing [Fig. 4(d)], after 15 000 consecutive operations, the output performance shows a stable trend, which also provides a feasibility for the long-term work of E-TENG in human gait monitoring. The electrical outputs (voltage, current, and power) of the E-TENG under different loads have also been investigated [Fig. 4(e)]. According to the results, the maximum output power of the E-TENG can arrive at 175 µW (load resistance: 20 MΩ), which indicates the potential application of powering low-power electronic devices. In addition, we conducted research on the self-charging system based on the E-TENG device [Fig. 4(f)]. According to the charging curve of different capacitors in Fig. 4(g), capacitors with lower capacitance values can reach high voltage more quickly, indicating that the E-TENG has unique advantages in micro-energy collection and storage. Furthermore, the energy stored in capacitors can provide power to low-power electronic devices, such as electronic calculators, demonstrating potential applications, according to the charging/discharging curve in Fig. 4(h).

Due to the correlation between TENG's electrical signal and mechanical motion, TENG devices can play a role in the self-powered sensors, especially in human gait monitoring. Gait is relevant to physiological factors, such as human physiological structure, knee joint health status, and psychological factors. As the fingerprints, it contains individual-specific information, as shown in Fig. 5(a). Human gait information monitoring for special populations is significant, especially for osteoarthritis prevention and diagnosis. To demonstrate the application of E-TENG in joint monitoring, we installed the E-TENG on fingers and elbows to measure the electrical signal from the E-TENG during joint movement. According to the results [Fig. 5(b)], when the finger bends at different angles, the E-TENG device installed on the finger can produce various output voltage peaks. Specifically, when the finger bending angle grows from 30° to 90°, the output voltage of the E-TENG device improves from 9.4 to 19 V, which clearly reflects the monitoring function of the bE-TENG on joints. Similarly, the E-TENG device can achieve the monitoring of elbow movement. Under different degrees of bending, it can be judged by the electrical signal from the E-TENG device [Fig. 5(c)]. Furthermore, the E-TENG device can indirectly monitor human gait information by sensing different gait types. For example, the E-TENG installed inside the shoes can detect signals when the human body falls [Fig. 5(d-1)]. The pulse signal is generated by the rapid separation of the two triboelectric materials of the E-TENG after the human body falls, resulting in a strong pulse signal, as illustrated in Figs. 5(d-2) and 5(d-3). For the walking movement [Fig. 5(e)], the electrical signal of the E-TENG device can accurately evaluate information such as the number of steps, walking speed, and stride characteristics of walking. Compared to walking movement, there are significant differences in the output signals of the E-TENG between jumping and running, including peak values and signal frequency, as illustrated in Figs. 5(f) and 5(g). In addition, these signal differences can distinguish between different types of human gait, which is of great help for gait analysis. Ulteriorly, we simulated the walking state of normal individuals and patients with knee osteoarthritis and measured the output signals of E-TENG under two different gaits, as shown in Figs. 5(h) and 5(i). From the results, compared to the output signal of E-TENG under normal walking, the output signal of E-TENG under walking in patients with osteoarthritis shows significant differences, especially the signal peak. The reason for this result is that the knee of osteoarthritis patients is difficult to bend, and when walking, the stride height is low and flat, which leads to a lower maximum separation distance between the two triboelectric layers of the E-TENG device, resulting in a lower peak electrical signal. The early pathological changes of osteoarthritis are mainly joint cartilage damage. Early monitoring and intervention with E-TENG can delay progression, prevent irreversible lesions, and improve the patient’s treatment cycle since the joint cartilage still has regenerative ability in the early stage. Thus, the application of E-TENG in monitoring osteoarthritis in this research is of great significance.

In conclusion, the E-TENG based on eyebrow powder was fabricated for bio-mechanical energy harvesting and gait monitoring of patients with osteoarthritis. Under a maximum separation distance (5 mm) and a maximum motion frequency (6 Hz), the E-TENG device can attain a V_oc of 169 V and a I_sc of 5.5 µA. Meanwhile, the Q_sc of E-TENG is 20.8 nC, and the maximum output power of E-TENG can arrive at 175 µW (load resistance: 20 MΩ). From the reliability testing results, after 15 000 consecutive operations, the E-TENG can still maintain a stable output state, which also provides a feasibility for the long-term work of E-TENG in human gait monitoring. The E-TENG can detect various human gait patterns (walking, running, and jumping), finger motion, and elbow joint movements. Further research has shown that the E-TENG can be used for gait recognition and monitoring in patients with osteoarthritis, providing reference data for osteoarthritis prevention and treatment.

Figure S1 SEM image of (1) eyebrow powder layer and analysis of the corresponding spectral elements.

Figure S2 SEM image of (2) eyebrow powder layer and analysis of the corresponding spectral elements.

Figure S3 SEM image of (3) eyebrow powder layer and analysis of the corresponding spectral elements.

Figure S4 The output voltage of five TENG based eyebrow powders and different negative triboelectric materials.

Table S1 Elemental composition and proportion of (1) eyebrow powder.

Table S2 Elemental composition and proportion of (2) eyebrow powder.

Table S3 Elemental composition and proportion of (3) eyebrow powder.

This work was supported by the National Key Research and Development Program of China (Grant No. 2018YFA0703000).

The authors have no conflicts to disclose.

Y.D. and Y.L. contributed equally to this work.

Yunyi Ding: Conceptualization (lead); Data curation (lead); Formal analysis (lead); Funding acquisition (lead); Investigation (lead); Methodology (lead); Project administration (lead); Resources (lead); Software (lead); Supervision (lead); Validation (lead); Visualization (lead); Writing – original draft (lead); Writing – review & editing (lead). Yichen Luo: Conceptualization (lead); Data curation (lead); Formal analysis (lead); Funding acquisition (lead); Investigation (lead); Methodology (lead); Project administration (lead); Resources (lead); Software (lead); Supervision (lead); Validation (lead); Visualization (lead); Writing – original draft (lead); Writing – review & editing (lead). Xue Zhou: Conceptualization (equal); Data curation (equal); Formal analysis (equal); Funding acquisition (equal); Investigation (equal); Methodology (equal); Project administration (equal); Resources (equal); Software (equal); Supervision (equal); Validation (equal); Visualization (equal); Writing – original draft (equal); Writing – review & editing (equal). Shaojie Zhang: Conceptualization (equal); Data curation (equal); Formal analysis (equal); Funding acquisition (equal); Investigation (equal); Methodology (equal); Project administration (equal); Resources (equal); Software (equal); Supervision (equal); Validation (equal); Visualization (equal); Writing – original draft (equal); Writing – review & editing (equal). Yayu Li: Conceptualization (supporting); Data curation (supporting); Formal analysis (supporting); Funding acquisition (supporting); Investigation (supporting); Methodology (supporting); Project administration (supporting); Resources (supporting); Software (supporting); Supervision (supporting); Validation (supporting); Visualization (supporting); Writing – original draft (supporting); Writing – review & editing (supporting). Bin Zhang: Conceptualization (supporting); Data curation (supporting); Formal analysis (supporting); Funding acquisition (supporting); Investigation (supporting); Methodology (supporting); Project administration (supporting); Resources (supporting); Software (supporting); Supervision (supporting); Validation (supporting); Visualization (supporting); Writing – original draft (supporting); Writing – review & editing (supporting).

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