Facile Construction of Self-Powered Electronic Textiles for Comprehensive Respiration Analysis

Respiration is one of the most important physiological processes of the human body. Continuous monitoring of respiration with wearable sensors can provide abundant information related to the health status of the respiratory system. However, the vast majority of the developed respiration sensors are constructed on nontextile substrates, resulting in dissatisfactory convenience, comfortness, and aesthetics. Additionally, special micro‐/nanomaterials are inevitably involved to construct the respiration sensors, which pose potential hazards to the human body due to the exfoliation and inhalation of these micro‐/nanomaterials. Here, humidity‐ and temperature‐sensitive electronic textiles (e‐Textiles) are presented, which are fabricated based on easily accessible and biofriendly materials (e.g., sodium chloride, glycerin, aluminum fiber, carbon fiber, and polyvinyl alcohol). Moreover, the proposed e‐Textiles use a potentiometric sensing mechanism, featuring self‐powered signal outputs and ultralow power consumption. The humidity and temperature‐sensing functionalities can also be in situ integrated into commercial masks for constructing full‐textile and all‐in‐one e‐Masks for comprehensive respiration monitoring. As demonstrations, both single‐point respiration monitoring (e.g., intensity, frequency, etc.) and 2D respiratory analysis (i.e., airflow distribution) can be realized with the e‐Masks. This work provides a facile and scalable approach to manufacture self‐powered and fully fabric respiration sensors for comprehensive respiratory analysis.


Introduction
Respiration is a vital physiological process closely related to the health conditions of the human body.[3] Among the family of wearable respiration monitoring devices, smart electronic masks (e-Masks) play an important role due to their unique capabilities for both disease prevention and respiration monitoring.[7] Meanwhile, e-Masks with sensing capability can be used to acquire valuable information related to the status of respiratory process and the symptoms of related diseases. [8,9]Therefore, developing electronic masks with both disease protective ability and respiratory monitoring functionality is of great significance to reduce medical costs and improve the population's health status.
In recent years, different types of electronic sensors have been integrated into commercial masks to construct wearable respiratory monitoring systems.In general, the flexible sensors used for respiratory monitoring could be divided into two categories based on their working principles or sensing mechanisms.One class of respiration sensors is fabricated based on active sensing mechanisms, such as resistive sensors, [10][11][12][13][14][15][16][17][18] capacitive sensors, [19][20][21][22] and so on. [23]These sensors convert external stimuli generated from respiration (e.g., pressure, temperature, humidity, etc.) into the changes of electrical resistance or capacitance variations.For example, Ma et al. prepared a fibrous capacitive humidity sensor by winding, coating, and sputtering technologies. [20]This fibrous humidity sensor can be integrated into a fabric for breath monitoring.Li et al. fabricated flexible respiration sensors via printing technique based on porous graphdiyne (GDY) nanomaterial. [15]The abundant active sites of GDY can adsorb water molecules, thus exhibiting variation in electrical signals under humidity change.Integrating the printed GDY respiration sensors with a mask allows to achieve respiratory monitoring functionality.However, for the resistive and capacitive sensors as mentioned above, a constant bias voltage is prerequisite to operate the sensors and the power consumption could be as high as tens of mW per single-sensing device, [24] which limits their large-scale deployment and longterm operation.
As an alternative, electronic sensors based on passive sensing mechanisms are proposed to mitigate the issues related to power consumption.Typical passive sensors include electret sensors, [25,26] triboelectric sensors, [27][28][29][30][31] electrostatic sensors, [32] moisture-electric sensors, [33][34][35] electrochemical sensors, [36][37][38][39][40] and so on. [41]This category of sensors does not necessitate an external power supply to operate the devices and can generate electrical signal outputs by themselves under external stimulations (e.g., pressure, humidity, temperature, etc.).For instance, Wang et al. demonstrated an air-flow-driven triboelectric nanogenerator with a flexible nanostructured polytetrafluoroethylene (n-PTFE) thin film vibrating in an acrylic tube under airflow. [30]hong et al. reported a smart mask integrated with an electrostatic pressure sensor, which can generate open-circuit voltage outputs under pressure variation. [32]He et al. developed a moisture power generator, which can produce electrical current or voltage outputs under humidity stimulations and can be used to construct a self-powered smart mask for respiration monitoring. [34]Additionally, Li et al. reported a self-powered electrochemical humidity sensor based on the metal-air redox reaction. [36]espiration sensors based on aforementioned passive working principles exhibit a self-powered operation manner and ultralow power consumption, revealing promising application in manufacturing respiratory monitoring systems.
Despite the remarkable progress achieved as mentioned above, for the vast majority of the currently developed respiration sensors, there are two major issues that hinder their large-area practical application.On the one hand, the fabrication of respiration sensors usually involves a variety of special micro-/ nanomaterials to construct the active sensing layer or the conductive electrodes.30] These nontextile-based respiration sensors are attached or immobilized onto commercial masks to monitor the respiration process. As a result, theomfortness, convenience, and aesthetics of the respiration monitoring system could be significantly compromised.Recently, fibrous respiration sensors or textile-based respiration sensors have been designed for respiration monitoring.[20,34,45] However, these fibrous or textile-based respiration sensors are directly attached to commercial masks, not in situ and seamlessly integrated into the masks to form an all-in-one respiration monitoring system.Moreover, the fibrous or textile-based respiration sensors are fabricated based on active sensing mechanisms and necessitate continuous power supply.[20,45] Hence, developing fully textile-based, in situ integrated, and self-powered electronic masks with easily accessible and biofriendly materials as well as facile fabrication process is high necessary, but has rarely been explored so far.
Here, we propose and demonstrate the facile construction of self-powered electronic textiles (e-Textiles, Figure 1) using easily accessible and biofriendly raw materials.Such e-Textiles feature both humidity-and temperature-sensing capability via a selfpowered manner, which makes the e-Textiles ideal candidates for continuous respiration monitoring.A potentiometric sensing mechanism based on the metallic corrosion effect was employed to fabricate the proposed e-Textiles.Specifically, fibrous aluminum (Al) electrodes and fibrous carbon electrodes were used as the two electrodes and integrated into a polyvinyl alcohol/ sodium chloride/glycerin (PVA/NaCl/Gly)@textile electrolyte.With this material system, a self-generated potential difference output can be developed between the Al and carbon electrodes due to metallic corrosion reactions happening on the Al electrode. [46,47]Importantly, the potential difference output capability relies heavily on the properties (e.g., ionic activity, electrical impedance, etc.) of the PVA/NaCl/Gly@textile electrolyte.Regulating the properties of the PVA/NaCl/Gly@textile electrolyte via humidity or temperature stimulations can significantly modulate the potential difference outputs between the Al and carbon electrodes, enabling to realize potentiometric humidity or temperature-sensing functionalities.Specifically, the humidity and temperature-sensing functionalities are positively correlated with each other, that is, both higher humidity and higher temperature give rise to higher sensor signal outputs, as discussed below.Therefore, this kind of e-Textile can be a good candidate for comprehensive respiration monitoring.Based on this new sensing principle, fully fabric and self-powered smart e-Masks can be facilely fabricated.The as-prepared e-Masks exhibit good wearing comfortness, superior cost-efficiency, and good capability to monitor the respiration process with ultralow power consumption.The comprehensive comparisons between this e-Mask with other reported e-Masks are summarized in Table S1, Supporting Information.As promising applications, both singlepoint respiration monitoring as well as 2D respiratory mode analysis were successfully demonstrated with this e-Mask.The design philosophy and fabrication methodology demonstrated here open up new opportunities for the manufacturing of novel respiration monitoring systems in future.

Fabrication and Characterization of the e-Textile
Figure 2a illustrates the fabrication process of the e-Textile.An aqueous solution mixture composed of PVA, NaCl, and Gly was prepared, followed by coating a specified amount of PVA/NaCl/ Gly mixture onto a piece of cleaned commercial textile.After drying, a layer of PVA/NaCl/Gly was coated onto the surface of the textile substrate, forming a PVA/NaCl/Gly@textile composite electrolyte.For the composite electrolyte, PVA was used as a hydrophilic polymer matrix to enhance the adhesion between the PVA/NaCl/Gly electrolyte and the textile substrate.NaCl was introduced as the ion source and provides abundant mobile ions and ionic conductance for the electrolyte.Additionally, Gly, which can bind with water molecular tightly and act as a humectant, was used to modulate the water content and thus regulate the ionic activity of the electrolyte. [48,49]Finally, Al fibers and carbon fibers, which act as two electrochemical electrodes, were sewn into the PVA/NaCl/Gly@textile electrolyte with controlled layout, forming a fully fabric, air-permeable, and comfortable e-Textile.
Figure 2b-e shows typical scanning electron microscope (SEM) images of the textile before and after being coated with PVA/NaCl/Gly electrolyte.After coating PVA/NaCl/Gly onto the textile, the woven structure and the porous morphology were not significantly affected (Figure S1, Supporting Information), while the textile surface becomes rough and rugged, revealing the successful deposition of PVA/NaCl/Gly electrolyte onto the textile.Besides, Na and Cl elements can be found on the surface of PVA/NaCl/Gly@textile (Figure 2f-g) due to the deposition of PVA/NaCl/Gly on the textile.In addition, Fouriertransform infrared (FTIR) analysis (Figure 2h) shows that the coating of PVA/NaCl/Gly onto the textile gives rise to a new broad absorption peak around 3300 cm À1 , which can be assigned to the O-H stretching vibration of coated PVA and Gly.
Notably, the deposition process of PVA/NaCl/Gly onto the textile substrate can be well controlled by changing the weight ratio of M c to M i (M c /M i ), where M i refers to the weight of the initial textile before coating and M c refers to the weight of coated PVA/NaCl/Gly aqueous mixture solution.As shown in Figure 2i, the weight ratio of the final dried PVA/NaCl/ Gly@textile (M f ) to the initial textile (M i ) (M f /M i ) increased with M c /M i , indicating that the coating process can be well regulated.Moreover, the properties (e.g., electrical impedance) of PVA/NaCl/Gly@textile can also be modulated easily, as shown in Figure 2j, enabling us to adjust and regulate of the performance of the e-Textile.

Sensing Mechanism of the e-Textile
The basic working principle of the e-Textile is illustrated in Figure 2k.With Al and carbon fibrous electrodes integrated in the PVA/NaCl/Gly@textile electrolyte, an electrical potential difference can be developed and recorded between the two fibrous electrodes.This is due to the fact that the Al fiber acts as an anode and the superficial Al atoms can be dissociated into metal ions via losing electrons, resulting in a negatively charged surface.Accordingly, the carbon fiber acts as a cathode with possible oxygen reduction and/or hydrogen evolution on this electrode, which, however, still remains vague and has not been clearly verified. [47,50,51]he developed potential difference output between the two fibrous electrodes relies heavily on the properties (e.g., impedance) of the PVA/NaCl/Gly@textile electrolyte, as shown in Figure 2l.Importantly, the electrical impedance of the PVA/NaCl/Gly@textile electrolyte is highly sensitive to external humidity or temperature stimulations.Specifically, when the PVA/NaCl/Gly@textile electrolyte is exposed to different humid environments, the water adsorption content of the electrolyte is different.Higher environmental humidity gives rise to higher water content and higher ion activity of the PVA/NaCl/Gly electrolyte.As a result, the ionic conductivity of the PVA/NaCl/Gly electrolyte could be improved and the output capability of the potential difference developed between the two electrodes can be enhanced when increasing environmental humidity.In addition, external temperature variation can modulate the ion mobility, leading to the change of the ionic conductivity of the electrolyte.Increasing the temperature will increase the migration capability of the ions and the conductivity of the electrolyte.As a result, the output capability of the potential difference between the two electrodes can be improved when increasing the environment temperature.Therefore, the potential difference outputs between the two fibrous electrodes can be regulated by external humidity or temperature stimulations.
It is worth pointing out that the e-Textiles do not necessitate an external power supply and have self-generated signal outputs, featuring ultralow power consumption when compared with other traditional humidity or temperature-sensing devices.In detail, the generated potential difference outputs arise from the electrochemical reactions of the two fibrous electrodes.To verify this assumption, the polarization curve of the Al fiber electrode in NaCl solution is recorded and a corrosion potential of 1.17 V can be obtained (Figure 2m).This confirms that metallic dissociation occurs on the fibrous Al electrode, leaving rich electrons on the Al electrode.Accordingly, on the fibrous carbon electrode, there could be two possible electrochemical reactions: oxygen reduction and/or hydrogen evolution.To further verify the self-powered property of the e-Textiles, we set up an Al/NaCl/carbon system, where saturated NaCl solution acts as the electrolyte, and Al and carbon act as the two electrodes.By connecting eight Al/NaCl/carbon systems in series, a total voltage output about 2 V can be recorded.This voltage output generated by the Al/NaCl/carbon system is high enough to light up a light-emitting diode (LEDs), as shown in Figure 2n, confirming the power-supplying capability of such Al/NaCl/carbon systems.For the e-Textiles, the potential difference outputs are also self-generated, featuring self-powered sensing characteristics.

Sensing Performance of the e-Textile
During respiration, the humidity and temperature near the mouth and nostril positions vary simultaneously.Monitoring the humidity and temperature variations allows to analyze the respiration process in real time, which can be realized with our proposed e-Textiles.On the one hand, the potential difference outputs of the e-Textile can be regulated by humidity change, as the water content and ionic conductivity of the hydrophilic PVA/NaCl/Gly@textile electrolyte can be improved by increasing the environment humidity, as illustrated in Figure 3a.In order to evaluate the humidity sensing performance, a measurement platform was set up using specific saturated salt solutions to create different relative humidity (RH) conditions at room temperature (≈25 °C) (Figure S2a, Supporting Information).The electrical impedance of the e-Textile can be influenced by the change of RH (Figure S3a, Supporting Information).Figure 3b exhibits the potential difference outputs measured between the Al and carbon electrodes in different RHs.The potential difference output increases significantly with the increase of the RH in the range of 11-97%.Figure 3c shows the humidity sensitivity of the e-Textile, which can be defined as follows.
where ΔV and ΔRH correspond to the potential difference variation and the RH change, respectively.With the increase of RH, the sensitivity gradually increases and reaches the maximum of 5.23 mV %RH À1 at 97% RH.The humidity sensitivity of the e-Textile is comparable with other reported humidity sensors (Table S2, Supporting Information).The e-Textile also exhibits desirable signal stability over a long period time (≈200 s) under a specific humidity in the range of 11-97% (Figure 3d).In addition, the reliability of the e-Textile in a cyclic humidificationdehumidification process (from 23% RH to 57% RH) was tested.
As shown in Figure 3e, repeatable signal variations of the e-Textile can be detected over ten cycles, illustrating the repeatability of the e-Textile for humidity sensing.Furthermore, the long-term stability of the e-Textile was also investigated.The signal outputs of the e-Textile remain relatively stable at an ambient condition (i.e., 25 °C and 84% RH), demonstrating good stability and durability of the e-Textile (Figure 3f ).The above results demonstrate that the e-Textile have desirable humidity-sensing performance.
On the other hand, the developed e-Textile is also sensitive to temperature variation (Figure 3g).This is due to the fact that temperature change would affect the ion mobility and conductivity of the PVA/NaCl/Gly@textile electrolyte (Figure S3b, Supporting Information).A measurement platform was set up to investigate the temperature-sensing performance of the e-Textile (Figure S2b, Supporting Information).Figure 3h depicts the variation in potential difference outputs of the e-Textile when gradually increasing the temperature from 20 to 45 °C step by step at a gradient of 5 °C.The potential difference outputs of the e-Textile increase with the elevation in temperature and are relatively stable when maintaining the temperature.Figure 3i illustrates the potential difference outputs of the e-Textile under different temperature conditions in the temperature range of 20-45 °C.The temperature sensitivity of the e-Textile can be calculated, as shown in the inset of Figure 3i, which can be defined as follows.
where ΔV and ΔT are correspond to the potential difference variation and the temperature change, respectively.With the increase of temperature, the sensitivity gradually increases and reaches the maximum of 4.42 mV °CÀ1 at 45 °C.
The stability of the temperature-sensing performance of the e-Textile was also investigated (Figure 3j).When changing the temperature from 25 to 35 °C repeatedly for multiple cycles, the response curve shows desirable repeatability and reliability.The typical response and recovery behaviors of the e-Textile were investigated as well.When changing the temperature from 25 to 40 °C, or from 25 to 5 °C, the response curve of the e-Textile is shown in Figure 3k.It can be observed that the e-Textile exhibits good capability for continuous temperature monitoring.

Application of e-Mask for Single-Point Respiratory Monitoring
Monitoring the humidity and temperature variations near the mouth and nostril positions allows to analyze the respiration process in real time.Notably, the humidity and temperature-sensing functionalities of the developed e-Textile are positively correlated with each other (Figure S4, Supporting Information), which is highly desired for respiratory monitoring.In addition, the signal outputs of this e-Textile would not change significantly under moderate mechanical stimulations (e.g., bending, twisting, etc., Figure S5, Supporting Information), exhibiting desirable mechanical stability for respiration monitoring.
As a demonstration, medical masks were used as the textile substrate to fabricate the e-Masks.Medical masks generally consist of three layers, including the outer protective layer, the middle filtering layer, and the inner layer, as depicted in Figure S6, Supporting Information.PVA/NaCl/Gly electrolyte could be coated onto the selected area of the filtering interlayer of the medical mask, followed by sewing Al fiber and carbon fiber electrodes into the mask.A digital photograph of the fabricated e-Mask is shown in Figure S7, Supporting Information.Under different RH environments and temperature environments, e-Mask still exhibits good sensing performance, as shown in Figure S8, Supporting Information.Moreover, in a certain period of time (e.g., several days), the response behaviors of the e-Mask are relatively stable (Figure S9, Supporting Information).Although the baseline decreases slightly over time, the overall signal responses do not exhibit significant change and the e-Mask still has good capability to detect the respiration process.
Coating PVA/NaCl/Gly electrolyte and sewing Al and carbon electrodes onto the filtering interlayer of the mask does not remarkably affect the filtration efficiency, filter resistance, and moisture permeability (Figure S10 and S11, Supporting Information).A self-designed and custom-made circuit board was used to record and transmit the data of the e-Mask to a personal computer (Figure 4a).With a subject wearing the e-Mask, the exhaled airflow passes through the e-Mask.As the exhaled airflow from the body generally has relatively higher humidity and higher temperature compared with the ambient air, the potential difference outputs of the e-Mask increase accordingly, as shown in Figure 4b.When the subject conducts inhalation, the inhaled ambient air has relatively lower humidity and lower temperature than the exhaled airflow, and the potential difference outputs of the e-Mask decrease, as shown in Figure 4b.During continuous respiration with both exhalation and inhalation processes, the signal outputs of the e-Mask show periodic variations.
The frequency and amplitude of the periodic signal variations allow us to resolve the respiratory frequency and the depth of breathing, as illustrated in Figure 4b.Based on this principle, the exhalation and inhalation processes can be reflected in the response behaviors of the e-Mask, as shown in Figure 4c.
The signal outputs of the e-Mask can be continuously recorded by a custom-made circuit board.The design and operation principles of the circuit board are depicted in Figure 4d.The potential difference signal is first sent to an operational amplifier with fixed amplification (1), followed by a voltage regulator (2) to ensure the amplifier's stable operation.Then, the signal is transferred to the STM 32 microprocessor (4) and concentrated for final processing, with the reference voltage (3) powering the STM 32 microprocessor's analog-to-digital converter.The recorded signals are finally displayed on a software for visualization and postprocessing, as depicted in Figure 4e.At current stage, only wired signal transmission method is employed.Wireless signal transition has not been achieved, which would be explored in future.
Figure 4f shows the typical response curve of the e-Mask when a subject conducted 3 min's continuous normal respiration.Based on this response curve, the respiratory rate is calculated to be around 16 bpm (beat per minute), which agrees well with the practical situation.More importantly, different respiration conditions, including normal breathing, deep breathing, rapid breathing, and stopped breathing, can be detected and differentiated by wearing the e-Mask (see Video S1, Supporting Information).Figure 4g depicts the response curves of the e-Mask when the subject conducted different respiration conditions.The respiratory rate and the breathing intensity differ significantly from different respiratory states.For example, the signal intensity recorded from deep breathing is approximately three times that of normal breathing, while the respiratory rate is approximately half of the normal breathing.Furthermore, during stopped breathing state, the potential difference output does not vary periodically anymore.When the subject returned to a normal breathing state, a periodic response signal variation can be detected again.
Moreover, the respiration status before and after exercise can be distinguished with the e-Mask.Figure 4h shows the response curve of the e-Mask before and after the subject taking exercise.It can be observed that the breathing depth and breathing frequency of the subject increase significantly after exercise.The respiration frequency was calculated to be 19 bpm before exercise and 35 bpm after exercise, as shown in Figure 4i, which is in good accordance with the actual situation.
In addition, an intelligent and real-time respiration analysis system was designed and constructed.As shown in Figure S12 and Video S2, Supporting Information, the intelligent respiration analysis system is capable of analyzing as well as displaying the important respiration parameters, such as respiratory rate, respiratory intensity, and respiratory state.
Moreover, e-Mask can distinguish the status of coughing and sneezing, as shown in Figure S13, Supporting Information.Besides, different types of textile substrates could be employed to construct e-Textiles and e-Masks for respiration monitoring (Figure S14, Supporting Information).In addition, the response performance of our proposed e-Textiles and other types of conventional sensors (including resistive sensors and capacitive sensors) were compared and are presented in Figure S15, Supporting Information.

Application of e-Mask for 2D Respiratory Analysis
In addition to monitoring of respiration depth and respiration rate, analyzing the respiratory modes (i.e., how the respiratory airflow passes through the mouth and nasal cavities) can reflect more valuable information related to the respiratory habit and status of a person.For example, when breathing via the nostril, the nasal hairs can block the dust in the inhaled air. [52]Also, the nostril cavity can humidify and warm the inhaled air, which is of great benefit for the respiratory tract.In contrast, some breathing modes can be harmful to one's health.For instance, excessive breathing via the mouth cavity can cause dryness and discomfort to the throat area.As the mouth is unable to filter the dust in the air, excessive mouth breathing can lead to diseases such as stomatitis and pharyngitis due to the inhaled dust and germs.Besides, excessive mouth breathing can affect the alignment of the teeth and the shape of the face. [53]Moreover, when a person has illnesses such as colds, one or even two of the nasal cavities can be congested, which also affects the respiratory modes.Therefore, monitoring and analysis of the respiratory modes can provide rich information about the health status of a person.
To analyze the respiratory modes, an e-Mask with 3 Â 3 sensing unit array was designed based on the airflow distribution (Figure 5a).In general, the human body breathes through the two nostrils and the mouth.To distinguish different breathing patterns, we first set three sensing points (①, ②, and ⑤ in Figure 5a) near the two nostrils and the mouth.These three sensing points are directly subjected to the breathing airflow.On the other hand, since the mask and the human face will form a relatively airtight space, the airflow will spread around the bridge of the nose, cheeks, chin, and other parts in this space.Therefore, 3 Â 3 sensing points were adopted and in situ fabricated on the outer protective layer of a medical surgical mask.The specific locations of each sensing unit are numbered and shown in Figure 5a.A digital photograph of the e-Mask with 3 Â 3 sensing units is shown in Figure 5b, and a digital photograph of volunteer wearing e-Mask is shown in Figure 5c.
Here, four different respiratory modes, including nasal breathing, right nostril breathing, left nostril breathing, and oral-nasal breathing, were used to evaluate the performance of the e-Mask.Figure 5d illustrates the airflows under different respiratory modes.A custom-made circuit board was used to collect the signals from the sensing units under a constant respiration frequency, and the response curves of the sensing units are shown in Figure 5e.By averaging the signal outputs from the sensing units separately, 2D contour plots that reflect the airflow distributions can be acquired (Figure 5f ).For example, when breathing via the nasal cavities, the average signal outputs near the two nasal cavities are higher than that of other positions, because the air flows downward from the nostrils.For the left nostril breathing mode, the average signal outputs located on the left side are higher because the airflow is exhaled downward from the left nostril cavity.The signal output distribution for the right nostril breathing mode is opposite when compared with that of left nostril breathing mode.When breathing via both mouth and nasal cavities, the signal outputs of the central area are higher than that of other positions.These signal output contour plots of the sensing units agree well with the airflow distributions in practical situations.
In order to further verify the reliability of the above results, an infrared camera was used to take infrared images of the subject when conducting different respiratory modes with a mask worn on the face, as shown in Figure 5g.The distribution of the temperature on the mask can be clearly seen from the infrared images, which also illustrates the distribution of respiratory airflow.The contour maps in Figure 5f are in good accordance with the infrared imaging maps in Figure 5g, verifying the reliability of this study.In conclusion, different respiratory modes can be resolved and analyzed with the e-Mask, which is a useful tool to study the respiratory habit and conditions.

Conclusion
In summary, a new class of self-powered e-Textiles with both humidity and temperature-sensing functionalities are successfully constructed based on easily accessible and biofriendly materials.The potential difference outputs measured between the fibrous Al electrode and carbon electrodes of the e-Textiles can be well regulated by external humidity and temperature stimulations.The resultant e-Textiles exhibit good humidity and temperature-sensing capability as well as reliable stability and repeatability.Based on this humidity and temperaturesensing strategy, smart e-Masks with desirable wearing comfortness, air permeability, and good capability to monitor the respiration process via a self-powered manner can be constructed.As proof-of-concept demonstration, diverse respiratory conditions (e.g., normal breathing, deep breathing, rapid breathing, and stopped breathing) and different respiratory modes (e.g., nasal breathing, right nostril breathing, left nostril breathing, and oral-nasal breathing) can be well resolved with the e-Masks.This work provides a scalable and low-cost strategy for constructing humidity-and temperature-sensitive e-Textiles and e-Masks with self-generated signals, exhibiting promising applications in personalized health analysis and early disease diagnosis in future.

Experimental Section
Materials: Commercial modal fabric composed of 95%wt cotton and 5%wt spandex, commercial silk fabric, and nonwoven fabric were selected as three representative textile substrates to fabricate the e-Textiles.Commercial surgical masks with three fabric layers (including the inner layer, the filtering interlayer, and the outer protective layer) were used to fabricate the e-Mask.Polyvinyl alcohol (PVA 1799), sodium chloride (NaCl), and glycerin (Gly) were purchased from Kelong Co., Ltd.The aluminum wire and carbon fibers were purchased from Aoshuo Co., Ltd. and Dongbang Co., Ltd., respectively.
Preparation of PVA/NaCl/Gly Solution Mixture: PVA solution (5 wt%) was prepared by dissolving PVA (5 g) powder in deionized water (95 g) and stirred at 70 °C for 3 h to obtain a homogeneous and transparent PVA solution.Then, NaCl particles were added into the PVA solution and stirred until completely dissolved.Finally, Gly was added into the above mixture and mixed uniformly, as shown in Figure S16a, Supporting Information.The mass ratio of NaCl to PVA and the mass ratio of Gly to PVA in the final solution mixture were 2.24% and 64%, respectively.
Preparation of the e-Textiles: The purchased textiles were cut into the designed shape and washed by deionized water three times in an ultrasonic cleaner.Then, the textiles were dried in an air-ventilated oven at 70 °C for 20 min, as shown in Figure S16b, Supporting Information.To prepare the PVA/NaCl/Gly@textile electrolytes, the cleaned textiles were dipped into the PVA/NaCl/Gly mixture solution for 15 min.Finally, the textiles coated with PVA/NaCl/Gly solution were set in a plastic dish and dried in an oven at 70°Cfor 20 min.After that, the fibrous aluminum and carbon, which acted as two electrodes, were sewed into the as-prepared PVA/NaCl/Gly@textile electrolyte with a space of ≈10 mm.
Fabrication of the Smart e-Mask: A common surgical mask with three layers (including the inner layer, the filtering interlayer, and the outer protective layer) was used to fabricate the smart e-Mask.Specifically, the filtering interlayer of the surgical mask was employed as the substrate, onto which PVA/NaCl/Gly was uniformly coated.It is worth noting that a diluted PVA/NaCl/Gly solution with a concentration of 3%wt PVA was used to coat the designated area of the filtering interlayer of the surgical mask uniformly, so that the air permeability of mask would not be affected significantly without excessive PVA/NaCl/Gly coated on the filtering layer.Then, the mask coated with PVA/NaCl/Gly was dried for 20 min at 70 °C.Finally, an aluminum wire and a carbon fiber were sewn onto the coated area with a space of ≈1 cm.For the smart e-Mask with the 3 Â 3 sensing unit array to analyze the airflow distribution of the respiration, the outer protective layer of the mask was selected as the substrate to fabricate the electrolyte, with 3 Â 3 aluminum wires and 3 Â 3 carbon fibers sewn into nine sensing unit areas (Figure 5b).The sensor stitching length was set to be ≈3 cm, the lateral spacing was set to be ≈3 cm, and the vertical spacing was set to be ≈1 and ≈0.5 cm.
Characterization and Measurement: The surface morphologies and the energy dispersive spectroscopy (EDS) mapping images of the textiles were obtained with a field-emission SEM (Zeiss Sigma 300, Germany).The chemical structures of the samples were analyzed by a FTIR spectrometer in the wavelength range of 400-4000 cm À1 (ThermoScientific Nicolet iS50, USA).The polarization curve of metal aluminum in NaCl solution was measured with a electrochemical workstation (AUTOLAB PGSTAT 302F, Switzerland).The electrical impedance of the electronic textiles was measured by a digital inductance, capacitance, resistance meter (Tong Hui TH2817B, China).The potential difference signal outputs of the e-Textile were recorded by a custom-designed circuit board (Figure 4d).The humidity sensing behaviors of the e-Textile were characterized in specific humidity environments, which were created with saturated salt solutions.Specifically, saturated salt solutions were prepared by dissolving LiCl (RH 11%), C 2 H 3 KO 2 (RH 23%), K 2 CO 3 (RH 43%), NaBr (RH 57%), KI (RH 69%), NaCl (RH 75%), KCl (RH 84%), and K 2 SO 4 (RH 97%) into deionized water in a closed test bottle at 25 °C.
For single-point respiratory monitoring, an e-Mask integrated with one sensing unit was worn by the subject on the face.The subject sat on a chair and in a resting state initially.Then, the subject exchanged the respiratory states intentionally (e.g., switching from normal breathing to deep breathing, followed by a pause in breathing after returning to normal breathing, and maintaining it for a period of time).During this process, the signal outputs of sensor were recorded by the custom-made circuit board.In addition, to evaluate the breathing conditions before and after exercise, the normal respiratory state before exercise was first measured with the e-Mask.Then, the respiratory state immediately after running for 10 min was evaluated with the e-Mask.
For 2D respiratory analysis, an e-Mask integrated with nine sensing units was worn by the subject on the face.The subject sat on a chair in a resting state.Then, the subject conducted a specific respiratory mode (e.g., right nostril breathing) and kept the respiratory mode as stable as possible.During this process, the signal outputs of each sensing unit were recorded individually with the custom-made circuit board.As the respiratory mode was kept relatively stable, the recorded signal variation of the sensing unit at different periods of time was stable as well.As shown in Figure S17, Supporting Information, the signal outputs of the sensing unit were nearly constant over 30 min, demonstrating the reliability of this measurement method.All human subjects involved in the test on human bodies agreed to all tests and the picture in the manuscript with informed consent, and all tests were approved by the Scientific Ethical Committee of Sichuan University.

Figure 1 .
Figure 1.Design concept of the e-Textiles and e-Masks.a) Component and working mechanism of the e-Textiles.b) Illustrative response behaviors of the e-Textiles.c) Application of the e-Masks for respiration monitoring and analysis.

Figure 2 .
Figure 2. Preparation, characterization, and working mechanism of the e-Textiles.a) Schematic illustrating the preparation process of the e-Textile.b,c) SEM images showing the surface morphology of the original textile before coating PVA/NaCl/Gly electrolyte.d,e) SEM images giving the surface morphology of the textile after coating PVA/NaCl/Gly electrolyte.Scale bars: (b,d) 400 μm; (c,e) 8 μm.f-g) EDS mapping of Na and Cl element on the PVA/NaCl/Gly@textile. h) FTIR spectra of the textiles before and after coating of PVA/NaCl/Gly electrolyte.i) Relative weight change of PVA/NaCl/ Gly@textile electrolytes prepared with different weight ratios of 1, 1.5, 2, 2.5, and 3. j) The relationship between impedance and different weight ratios of 1, 1.5, 2, 2.5, and 3. k) Illustration showing the self-powered sensing mechanisms of the e-Textile.l) The relationship between potential difference outputs and the electrical impedance measured between the two electrodes.m) The recorded polarization curve of Al in saturated NaCl solution.n) Schematic diagram and picture showing that an LED can be lit up by eight Al/NaCl/carbon systems connected in series.

Figure 3 .
Figure 3. Sensing performance of the e-Textile.a) Illustration showing the humidity sensing mechanism of the e-Textile.b) The average output values of the e-Textile under different RH environments.c) The humidity sensitivity of the e-Textile.d) Continuously monitored signal outputs values of the e-Textile under different RH environments.e) Cyclic test of the e-Textile in the humidity range of 23-57% for ten cycles.f ) Long-term stability test of the e-Textile under 84% RH for different days (1 day, 7 days, 15 days, and 30 days).g) Schematic showing the temperature-sensing mechanism of the e-Textile.h) The continuous response curve of the e-Textile when increasing the temperature from 20 to 45 °C in a stepwise manner.i) The signal outputs of the e-Textile under different temperatures.The inset shows the temperature sensitivity of the e-Textile.Response behaviors the e-Textile when the temperature varies between 25 and 35 °C for multiple cycles at a fixed RH of 45%.k) Typical response/recovery curves of the e-Textile when the temperature increases from 25 to 40 °C and decreases from 25 to 5 °C.

Figure 4 .
Figure 4. Application of e-Mask for single-point monitoring of human respiration.a) Schematic diagram showing the real-time monitoring of human respiration with the e-Mask.b) Diagram depicting the response behavior of the e-Mask during respiration.c) Typical response signal of the e-Mask in three cycles' normal breathing.d) Schematic diagram giving the circuit board design and operating principle.e) The signal visualization software used for the circuit board.f ) The response signal of the e-Mask during 3 min continuous normal respiration.g) Response signal curves of the e-Mask when the subject conducted different respiratory states.h) Response signals of the e-Mask before and after exercise.i) Histogram comparing respiratory rate before and after exercise.

Figure 5 .
Figure 5. Application of the e-Mask for 2D respiration pattern analysis.a) Schematic illustrating 3 Â 3 sensing unit array integrated on the e-Mask.b) Digital photograph of the e-Mask.c) Digital photograph of volunteer wearing e-Mask.d) Schematic diagrams showing the airflow distributions for four different respiration modes, including nasal breathing, right nostril breathing, left nostril breathing, and oral-nasal breathing.e) Recorded potential difference curves of each sensing unit under different respiration modes.f ) Distribution in potential difference outputs of the 3 Â 3 sensing unit under different respiration modes.g) Infrared imaging maps showing the temperature distributions (i.e., respiratory airflow distributions) under different respiration modes.