Ultrasensitive biosensing meta-garment via wetting gradient effect for heat-exhaustion warning

Introduction

Firefighting, an inherently dangerous occupation, requires professionals to go through complex and changeable extreme environments while engaging in fire suppression and life rescue operations^1,2. Unfortunately, the intense physical labor performed under high ambient temperatures frequently leads to heat-related illnesses, which are a significant cause of mortality among firefighters^3,4. Heat exhaustion, a common heat-related illness in this profession, is characterized by dehydration, tachycardia, and electrolyte imbalances⁵. Accompanied by multiple organ failure, they will further progress to thermoplegia (mortality rate up to 70%) without timely detection and treatment^6,7. Sodium, potassium, and glucose are crucial electrolytes that maintain body health and electrolyte metabolism, and their concentration levels can serve as the early diagnostic criteria for heat exhaustion^8,9,10,11,12. Furthermore, heart rate as a vital biomarker reflecting the basic life state and neural activity of human beings, provides insights into the severity of heat exhaustion^13,14,15. Therefore, the development of an integrated real-time health monitoring system has been regarded as a crucial and effective solution to reduce the incidence of heat exhaustion and safeguard the well-being of firefighters.

In recent years, significant advancements have made in biomarkers monitoring systems, with applications extending to the analysis of metabolism gases^16,17,18,19, excrement^20,21, and biofluid^22,23,24,25. However, traditional monitoring equipment falls short of the needs of firefighters’ rescue operations, as it lacks the capabilities for rapid movement, load-bearing and real-time alert. In this context, fiber-based sweat monitoring systems have shown immense potential for wearable personal health monitoring^26,27,28,29. These systems can noninvasively monitor molecular-level changes and assess individuals’ physiological states, thanks to their unique permeability^30,31, biocompatibility^32,33, and weavability^34,35. However, the undifferentiated absorption effect between conventional fiber-based sensors and integrated fabrics often leads to inefficient sweat collection, causing sensors to be triggered only after significant perspiration has occurred. This results in high detection volumes and slow response times, which can delay the early warning of abnormal signals. Additionally, a limitation of current fiber-based systems is their reliance on single biosensor units without integrated heart rate detection, which is crucial for identifying heat exhaustion. To address these issues, optimizing the fiber’s structural design to reduce detection volume and response time, and integrating sweat biosensors with heart rate detection devices, are seen as effective strategies for developing an early warning system for heat exhaustion.

Herein, a fully integrated multi-biomarker monitoring garment is proposed for heat-exhaustion warning. This innovative garment employs core-sheath structured sensing fibers for active sweat capture and fabric-based electrodes for electrocardiograph (ECG) detection, enabling the simultaneous detection of five biomarkers: heart rate, glucose, pH value, sodium (Na⁺), and potassium (K⁺). Thanks to the wetting gradient effect of biosensing fibers, the system achieves a low detection volume of 0.1 μL and a rapid response time of 1.4 s. Furthermore, 12-lead ECG signals and heart rate are accurately captured through fabric-integrated electrodes, which are securely sewn into the garment. Last, but not least, integrated garment was employed to monitor the biomarkers of a volunteer in real-time, with the capability to wirelessly transmit the data to a cell phone. Consequently, this work provides a promising approach for the early and personalized health management of firefighters, leveraging cutting-edge wearable technology for enhanced safety and efficiency.

Results

Illustration and fabrication of the multi-biomarker monitoring system

The wetting gradient effect assisted firefighter protective garment, designed for multi-biomarker biosensing (pH, glucose, Na⁺, and K⁺) and 12-lead ECG monitoring, was developed based on the super-absorbent core-sheath sensing fibers (CS-SF) and fabric-based electrodes (Fig. 1a). Specifically, CS-SF was prepared by depositing various functional materials onto the stainless-steel/cotton blended fiber (S/CF), which was then intertwined with the viscose fiber sheath. Benefited from the hydrophilic sheath layer with hierarchical spiral, a wetting gradient effect is formed between biosensing fibers and integrated textile substrate. Thus, sweat could rapid diffuse and permeate into the CS-SF, absorbed by sensing core effectively (Supplementary Fig. 1). Regarding fabric-integrated electrodes, polyester fabric was first coated with silver layer through chemical plating process and then filled with sponge to ensure the tight contact between the conductive fabric and skin (Supplementary Fig. 2). In order to meet clinical requirements, ten fabric-based electrodes with high conductivity (246 mΩ m⁻²) were firmly attached to the garment for 12-lead ECG monitoring, covering both limb leads and chest leads (Supplementary Fig. 3). By integrating the fabric electrodes and four CS-SFs, and further incorporating chips, the protective meta-garment can monitor health data in real-time and wirelessly transmit it to a mobile device.

Ultrasensitive biosensing meta-garment via wetting gradient effect for heat-exhaustion warning — **Fig. 1: Schematic illustration and application of the firefighter protective meta-garment.**

The working principle of the biosensors was illustrated in Fig. 1b. Polyaniline (PANI), glucose oxidase and selective ionophore were continuously deposited onto S/CF as the transducer, active layer and ion-specific adsorptive layer, respectively (Supplementary Fig. 4). These components were capable of rapidly transducing the concentrations of corresponding analytes into electrical signals. Scanning electron microscope (SEM) images revealed that the pristine S/CF was smooth with tangled yarn hairiness. After the deposition of active materials, PANI nanosheets, glucose oxidase layer, ionophore membrane and Ag/AgCl layer were evenly and densely spread over the surface of the yarn (Supplementary Fig. 5).

Owing to the simple and convenient processing method, sensing fiber can be continuously prepared and seamlessly integrated into commercial fabric, demonstrated their promising potential for industrial-scale production (Fig. 1c, d). Ultimately, the incorporation of these fibers and electrodes into firefighter protective gear enabled the real-time tracking of target biomarkers, effectively ensuring the safety of firefighters (Fig. 1e).

Sweat capture ability and electronic property of the CS-SF

The hierarchical spiral structure of the as-prepared CS-SF was clearly revealed in Supplementary Fig. 6. Viscose fiber sheath was tightly and regularly wound on the surface of the S/CF, establishing a core-sheath configuration. As a polymer substance composed of long chains, viscose fiber was enriched with hydroxyl (-OH) functional groups. These groups were predisposed to forming hydrogen bonds with water molecules, thereby endowing the CS-SF with a super-absorbent property suitable for sweat management. Moreover, the microporous structure of viscose enhanced its capability for water absorption and retention. Under the action of capillaries, water will rapidly penetrate into the fiber when a liquid contact with the surface of the viscose fiber (Fig. 2a and Supplementary Fig. 7). Consequently, viscose fiber exhibited a remarkable sweat absorption capacity of approximately 800%, a stark contrast to the absorbency observed in other common fiber substrates—194%, 88%, and 71% for S/CF, nylon, and polyethylene, respectively. The contact angle of viscose fiber was 18 degrees, endowing its efficiency in swiftly attracting and dispersing sweat away from the skin (Supplementary Figs. 8 and 9). Upon contact, the viscose fiber’s inherent hydrophilicity triggered a capillary effect that actively channels sweat, facilitating its diffusion and absorption. This dynamic interaction led to a notable increase in the diameter of the viscose fiber—approximately 50% after imbibing liquid droplets tinted with red dye (Fig. 2b). Such enriched sweat envelopment around the core sensing fiber considerably enhanced the biomolecular reaction efficiency with the active layer, simultaneously ensuring a stable circuitry connection.

The sweat capture ability of the CS-SF was demonstrated by injecting liquid droplets onto the integrated fabrics fabricated by embroidering CS-SF into common fabric (Fig. 2c and Supplementary Fig. 10). Since the significant wetting gradient effect, liquid droplets could only be accurately captured and collected by the CS-SF. Conversely, liquid was indiscriminately captured and randomly collected in the common sensing fiber (SF), leading to a limited amount of sweat in the sensing area, which affected the stable connection of the system (Supplementary Fig. 11). These findings highlighted the substantial advantages of CS-SF in capturing sweat and constructing a high-sensitivity system. To further clarify the gradient wettability mechanism, we delved into the wettability behavior of a liquid on the solid surface. According to the previous research, wettability behavior is governed by the solid-liquid-gas three-phase boundary, which can be described by Young’s model, Wenzel model and Cassie–Baxter model³⁶ (Fig. 2d). For an ideally smooth solid surface, Young’s equation can be obtained by balancing the horizontal surface tension in the system³⁷ (Eq. (1)). However, solid surface exhibit varying degrees of roughness in reality. In such scenarios, the contact angles predicted by the Wenzel model and the Cassie–Baxter model are expressed by Eqs. (2) and (3)³⁷, respectively.

$$cos {theta }_{Y}=frac{{gamma }_{{sg}}-{gamma }_{{sl}}}{{gamma }_{mathrm{lg}}}$$

(1)

$$cos {theta }_{w}=rfrac{{gamma }_{{sg}}-{gamma }_{{sl}}}{{gamma }_{{gl}}}$$

(2)

$$cos {theta }_{{CB}}=fleft(cos {theta }_{Y}+1right)-1$$

(3)

where θ_Y, θ_w and θ_CB are the equilibrium contact angle of the solid surface of Young’s, Wenzel and Cassie–Baxter, respectively, γ_sg, γ_sl and γ_lg are the surface tension of solid-gas, solid-liquid and liquid-gas, respectively, r is the roughness coefficient, and f is the ratio of the contact area of the solid-liquid interface to the actual surface area within a unit area.

Based on Eqs. (1)–(3), it is evident that roughness significantly impacts the hydrophilic properties of a surface. Therefore, gradient wettability surfaces can be formed by employing asymmetric contact angles to induce a micro-roughness gradient³⁸. Driven by the surface energy difference, droplets are propelled along the wettability gradient, facilitating their coalescence and subsequent merging into larger droplets³⁹. This approach allows for precise point-to-point fluid control and regulation (Fig. 2e). The interfacial imbalance force, commonly referred to as the wetting gradient driving force, can be mathematically expressed as Eq. (4)⁴⁰

$$F={2R}_{gamma mathrm{lg}}{int }_{0}^{pi /2}left({cos theta }_{A}-{cos theta }_{B}right)cos phi dvarphi$$

(4)

where R is the base radius of the contact between the droplet and the interface, θ_A and θ_B respectively represent the contact angle of the droplet before and after the wetting gradient, and ϕ represents the polar angle.

The fundamental electronic characteristics of the CS-SF were subsequently explored. A tiny amount of artificial sweat was continuously dripped onto the fiber and the circuit could be activated with merely 0.1 μL (Fig. 2f), suggesting the excellent sweat diffusion ability. Notably, the glucose-sensing fiber demonstrated a swift response time of 1.4 seconds upon exposure to high concentrations of sweat (Fig. 2g). Compared with the previously reported fiber-based electrochemical sensors^{41,42,43,44,45,46,47,48,49,50,51}, the CS-SF exhibited the lowest detection volume and the most rapid response time to date (Fig. 2h).

Electrochemical sensing performance of the CS-SFs

The electrochemical sensing performances of the four CS-SFs were individually tested in various analyte solutions. For glucose-sensing, glucose molecules are catalyzed by glucose oxidase to form gluconic acid while oxygen is consumed to produce hydrogen peroxide. The consumption of oxygen led to the change of electrical signal, reflecting the concentration of glucose in the solution (Fig. 3a). The glucose sensing CS-SF exhibited distinct current responses and performed a superior linear relationship with a sensitivity of 101.6 nA μM⁻¹ as the glucose concentration ranged from 0 to 250 μM (Fig. 3d). In the case of pH sensing, the conductive polyaniline undergoes deprotonation in the presence of H⁺, causing the fiber sensor to exhibit a potential change with varying pH values (Fig. 3b). The pH sensing fiber demonstrated a sensitivity of 51.25 mV pH⁻¹ and displayed a strong linear correlation within the pH range of 3.0–7.0 (Fig. 3e), encompassing the physiological pH range (4.5–6.5). Ion-sensing fibers were fabricated by depositing an ion-selective membrane and a PEDOT: PSS layer on the S/CF, serving as an ion-specific adsorptive layer and ion-to-electron transducer, respectively. These layers created ion channels in the sensing fibers capable of selectively binding to Na⁺ and K⁺ ions in sweat, resulting in a change in surface potential. The potential difference correlated with the ion concentration, providing an accurate indicator (Fig. 3c). A good linear response of the open circuit potentials to Na⁺ and K⁺ was achieved with sensitivities of 55.33 and 50.83 mV dec⁻¹, respectively (concentrations of 10–160 mM and 2–32 mM) (Fig. 3f, g). The sensing fibers were tested on at least 5 samples with excellent reproducibility (Supplementary Fig. 12).

**Fig. 3: Electrochemical sensing performance of the CS-SF.**

The selectivity of the CS-SF is crucial due to the presence of numerous metabolites in sweat. Therefore, interfering substances were intentionally added to the relevant solutions, and their impact on the electrical signals was assessed. It could be observed that these interfering substances had minimal effect on the target sensing system, indicating the excellent selectivity of the CS-SF (Fig. 3h–k). Furthermore, the electrochemical performances of the CS-SFs remained stable during testing for over 2000 seconds, indicating their robust long-term operational stability (Supplementary Fig. 13).

Wearable performance and robust monitoring performance

Wearable performance, which includes air breathability, moisture permeability, flexibility, and washing performance, plays a crucial role in the integrated garment system. Water vapor from commercial humidifiers could easily pass through the integrated garment system without restriction, suggesting the superior air breathability (Fig. 4a). Moreover, the integrated CS-SF could effectively remove heat from the skin through water evaporation after absorbing sweat. Infrared thermal images indicated that the skin area covered by integrated CS-SF was cooler compared to areas covered by common polyethylene film at the same ambient temperatures (Fig. 4b). Additionally, CS-SF exhibited excellent flexibility as it could be cut, twisted, knotted, woven, bent and coiled, showcasing its potential in smart electronic textiles application (Supplementary Fig. 14).

The electrochemical sensing performance of CS-SFs under different conditions was further investigated at the same concentration (glucose: 50 μM, pH: 6, Na⁺: 20 mM, and K⁺: 4 mM). During the dynamic deformation process, the sensing detection results remained close to 100% even after being bent for 50 cycles, as depicted in Fig. 4c, d. The stability of the performance was crucial for future practical applications. Furthermore, washability was tested by washing sensing fibers for 50 cycles in a washing machine. Each sensing fiber maintained over 90% electrochemical sensing performance, confirming the excellent washing durability of the CS-SF (Supplementary Fig. 15). To adapt to environmental diversity, the electrochemical sensing performance of CS-SFs was tested under temperature variations from 5 to 50 °C. The normalized intensity of the four sensing fibers all remained nearly 100%, demonstrating superior stability to temperature changes (Supplementary Fig. 16). It is worth noting that the performance of the glucose-sensing fiber showed gradual improvement and subsequent degradation with increasing temperature, indicating that the enzyme activity would initially increase within a certain temperature range. However, beyond a certain temperature threshold, the enzyme activity would decrease or even cease. Additionally, the long-term stability of the CS-SF was evaluated over 60 days (Fig. 4e). The current or potential responses of the same CS-SFs remained at least 90% of the initial value, highlighting the high long-term stability.

Ten fabric-based electrodes were scattered in the corresponding position to achieve 12 lead-ECG test and display via LEPU Dynamic ECG detector (TH12) (Fig. 4f and Supplementary Fig. 17). To evaluate the practicality of the device, ECG signals were collected by the electrodes both before and after exercise (Fig. 4g). It could be observed that ECG signals were accurately recorded with a significantly higher frequency after physical activity, showcasing the high reliability of the fabric-based electrodes in personal health monitoring applications.

Integrated ultrasensitive biosensing meta-garment for on-body analysis

Motivated by the sensitive sensing performance and comfortable wearable performance, an ultrasensitive biosensing meta-garment was developed to monitor both sweat sampling and heart rate among firefighters during their exercise routines or rescue operations. This garment was designed to promptly detect any abnormal changes in targeted biomarkers when a firefighter was in a potentially unhealthy physiological state. Upon detecting such changes, the garment not only alerted the wearer but also automatically sent a notification to the rescue center (Fig. 5a). Health-related data gathered by the garment could be wirelessly uploaded to a smartphone or computer via Bluetooth, and from there, it could be synchronized with Cloud storage. This feature aimed to enable the remote, real-time monitoring of multiple firefighters through a wireless data terminal. Furthermore, to enhance their understanding of personal health status, firefighters had the option to download and analyze their health data directly from the Cloud (Fig. 5b). This refined integration of technology into protective garment underscored our commitment to advancing firefighter safety and health monitoring. Through this innovative approach, we aspired to establish a more secure and informed environment for firefighters during their critical missions, ensuring that they are promptly alerted to potential health risks and provided with immediate access to their health metrics.

Leveraging the mature and cost-effective textile manufacturing techniques, a firefighter protective meta-garment was developed based on the fabric-based electrodes and CS-SFs, performing great potential in large-scale production (Supplementary Fig. 18). It was capable of detecting multiple biomarkers, collecting signals via an integrated chip, and then wirelessly transmitting the data to a Bluetooth-enabled smartphone equipped with a custom-developed application for continuous monitoring, data storage, and transmission (Fig. 5c, d and Supplementary Fig. 19). In a practical demonstration, a volunteer was dressed in the firefighter protective meta-garment for real-time on-body monitoring. During a running test, the volunteer’s heart rate quickly increased and stabilized at approximately 130 beats per minute (Fig. 5d, e). As the CS-SF absorbed enough sweat, the pH value and concentrations of glucose, Na⁺ and K⁺ were consistently transmitted through the integrated monitoring system in time. This successful application proved the promise of the CS-SF-based intelligent monitoring system for health monitoring and early warning in firefighting scenarios.

Discussion

In conclusion, this study introduces a biosensing meta-garment incorporating super-absorbent fiber-integrated biosensors and fabric-integrated electrodes designed for monitoring sweat composition and heart rate in firefighters. By taking advantage of the wetting gradient effect, the biosensing fibers presented exceptional stability and sensitivity, which required only 0.1 μL of sweat and 1.4 s to detect target molecules. Moreover, their performance remained consistent through various conditions, including washing, bending and exposure to a wide range of temperature fluctuations. And the sensing performance could still keep over 60 days. Additionally, the fabric-based electrodes were used to collect heart rate data through a 12-lead ECG system. This work provides an effective strategy for developing protective garment for firefighter, significantly advancing the implementation of an early health warning system for firefighters.

Methods

Materials

Stainless-steel/cotton blended fiber, modified cellulose fiber, polyester fabric, sponge, polyamide fiber, and polyethylene fiber were provided by Qingdao Qichu Intelligent Technology Co., Ltd. Glucose oxidase from aspergillus niger (GOx, 30 U mg⁻¹) was acquired from Sigma-Aldrich. Chitosan, potassium chloride (KCl), sodium chloride (NaCl), calcium chloride (CaCl₂), ammonium chloride (NH₄Cl), magnesium chloride (MgCl) and silver nitrate (AgNO₃) were purchased from Sinopharm Chemical Reagent Co., Ltd. 0.01 M PBS phosphate buffer dry powder (pH 7.2–7.4) was provided by Beijing Solarbio Science& Technology Co., Ltd. Graphene suspension was obtained from Ningbo Morsh Technology. Co., Ltd. Aniline, sulfuric acid(H₂SO₄), tetrahydrofuran, cyclohexanone, acetic acid and ammonia were purchased from Tianjin Chemical Reagent Co., Ltd. Calomel counter and platinum wire were purchased from Shanghai Yueci Electronic Technology Co., Ltd. PEDOT/PSS aqueous dispersion was acquired from Guangzhou Aibang Synthetic Materials Co., Ltd. Glucose anhydrous, sodium tetraphenyl boron (NaTFPB), sodium tetraphenylborate (NaTPB), high-molecular-weight poly(vinyl chloride) (PVC), bis(2-ethylhexyl) sebacate (DOS), valinomycin and sodium ionophore X were provided by Shanghai Macklin Biochemical Co. Ltd. Ag/AgCl ink was Shangdong AEX Chemical Technology Co. Limited.

Preparation of the CS-SF

For the preparation of core-sheath sensing fiber, sensing fibers were first continuous prepared by the home-made coating and electrochemical deposition system, and then wrapped with modified cellulose fibers by a rotating motor. The motor was operated at a speed of 60 rad min⁻¹.

Glucose Sensing Fiber: To prepare high performance biological sensors, the fiber electrodes should be modified with biocompatible polymers and enzyme. For the preparation of glucose sensor, S/CF was firstly immersed into the glucose oxidase solution (40 mg/ml in PBS) for 30 min and then dried at room temperature. Subsequently, GOx/S/CF electrodes were guided through the roller into the chitosan solution (5 mg/mL in 50 mM acetic acid) to from a fixed film on the surface to immobilize GOx molecules. Finally, the modified fiber electrode was wrapped on the collector roller. The speed of roller was 20 mm/s.

pH Sensing Fiber: The pH-sensing fiber was obtained by coating graphene as a mediator to improve conductivity in an aqueous suspension containing 60 wt% graphene nanosheets. The effective mass loading of graphene was 1.37–1.39 mg cm⁻¹. And then, PANI was electrodeposited at 0.7 V in an aqueous solution consisting of 0.1 M aniline and 0.5 M H₂SO₄, as calomel and platinum wire were served as the reference and counter electrodes, respectively. During the polymerization, G/S/CF was guided through the motor and wrapped to the electric holder in the beaker. The speed of roller was 20 mm/s.

Ion-Sensing Fiber: PEDOT: PSS layer was first coated onto S/CF through the routing of roller. Electrode was then composited with the Na⁺ selective membrane or K⁺ selective membrane. The Na⁺ selective membrane was prepared by dissolving NaTFPB (0.55 mg), PVC (33 mg), DOS (65.45 mg) and sodium ionophore X (1 mg) in 660 μL tetrahydrofuran, stirring for 2 h. Similarly, the K⁺ selective membrane was prepared by dissolving NaTPB (0.5 mg), PVC (32.75 mg), DOS (64.75 mg) and valinomycin (2 mg) in 350 μL cyclohexanone. The speed of roller was 20 mm/s.

Preparation of the fabric-based electrode

Firstly, ammonia was slowly added into AgNO₃ solution (0.2 mg/mL) until the solution became clear. Glucose (0.4 mg/mL) was then added to the solution under stirring for 15 min and polyester fabric was immersed into it. Finally, sliver-plated polyester was prepared by placing in the solution for 24 h. To obtain the fabric-based electrode, sliver-plated polyester and sponge were fixed on the cloths layer by layer.

Fabrication of the firefighter protective garment

The Ag/AgCl reference electrode was prepared by coating a layer of Ag/AgCl ink on the surface of S/CF and dry under the heater. And four core–sheath sensing fibers were embroidered into the fabric substrate with reference electrode to fabricate multi-biomarker monitoring textiles. Furthermore, ten fabric-based electrodes were integrated into clothes according to the distribution of 12-lead electrodes. Based on the biosensors and electrodes for ECG, firefighter protective garment was fabricated.

Characterizations

Scanning electron microscope (SEM, ZEISSEVO18) was used to examine the morphologies of the CS-SF and fabric-based electrode. Optical Contact Angle and 3D Shape Measurement system (DSA100) was used to investigate the wettability of the fibers. Resistivity of the fiber are characterized by a System Source Meter (KEITHLEY). An IR thermal camera (FLIR, A300 series) was used to capture the temperature profile of common film and the CS-SF.

Electrochemical test

Three-electrode system was adopted as the CS-SF, Ag/AgCl and platinum wire were served as the working, reference and counter electrodes, respectively. The sensing time of the CS-SF was tested through amperometric I–t curve. Similarly, I–t curve was also used for testing glucose-sensing fiber at 0.6 V (vs Ag/AgCl). Real-time open-circuit potential measurement was used to test the pH, Na⁺, and K⁺-sensing fibers in analyte solutions. To increase the concentration of analyte solution, 0.1 microliter high concentration of corresponding solution was quickly added to the low one. In addition, selectivity of the CS-SF was tested through adding interfering ions to the target analyte solution and the change of signals was observed. All the electrochemical performances were performed using a CHI660e electrochemical workstation (Shanghai Chenhua Instrument Co., Ltd.) and repeated at least five times.

On-body sweat analysis

A 24-year-old healthy male was recruited from Qingdao University and gave informed consent before participation in the study. During exercise, firefighter protective garment was dressed on the subject while electrolyte concentration could be monitored through biosensors and heart rate could be detected through electrodes integrated LEPU Dynamic ECG detector (TH12). The detected signal could be collected via the chip in time and transmit wirelessly to the custom-developed application of smartphone.