A novel manual rotating fluid control mechanism in a microfluidic device with a finger-actuated pump for dual-mode sweat sampling

Abstract

Wearable sweat analysis using microfluidics offers a non-invasive approach for real-time health monitoring, with applications in chronic disease management, athletic performance optimization, and early-stage condition detection. However, most existing wearable sweat microfluidic devices are limited to single-mode operation either real-time or on-demand sampling and often lack precise control over sample volume, which compromises analytical accuracy and utility. To address these limitations, we present a novel wearable microfluidic device featuring a manual rotating fluid control mechanism and a finger-actuated pump for dual-mode sweat sampling. The rotational control mechanism directs sweat either into detection chambers for volume-independent sensor reactions or through the finger-actuated pump for precise volume control. The pump incorporates a dedicated collection chamber, enabling sweat accumulation and controlled delivery in a single actuation, ensuring reproducible sample volumes and facilitating on-demand analysis when required. Additionally, the device integrates two reaction chambers for simultaneous dual biomarker detection. Performance validation during a 40 minute exercise session, using a colorimetric glucose assay, demonstrated reliable sweat sampling and on-demand biochemical analysis. These results highlight the device's potential as a practical tool for personalized health monitoring and field applications.

---

1 Introduction

Wearable microfluidic devices for sweat analysis have emerged as promising platforms for non-invasive, real-time health monitoring.^1–4 Sweat, a rich source of physiological and metabolic information, offers a viable alternative to blood-based diagnostics for tracking hydration status, electrolyte balance, and biomarkers related to metabolic disorders, stress, and athletic performance.^1,5–9 Unlike traditional sampling methods, microfluidic sweat sensors enable continuous and on-demand analysis without external sample collection, making them ideal for personalized healthcare and field-based applications.^3,10 Despite these advantages, fluid management remains a significant challenge in wearable microfluidic systems, particularly in achieving precise control over sample volumes.^11–13 Real-time monitoring systems, while beneficial for continuous data acquisition, often suffer from fluid dynamics variability, leading to sample dilution or inconsistent flow that compromises assay accuracy.^12,14,15

Conversely, on-demand collection systems, which provide more controlled conditions, are often complex and difficult to integrate into compact, portable, and energy-efficient wearable devices.¹⁶ To address these challenges, recent advancements have introduced innovative fluid control mechanisms, such as finger-actuated pumps and capillary burst valves (CBVs), to improve fluid handling precision.^3,17–19 Finger-actuated pumps allow users to manually control fluid delivery, enabling sweat collection and storage for delayed analysis.^17,18,20 This feature is particularly advantageous in scenarios where real-time analysis may be affected by external contamination, sweat composition fluctuations, or motion artifacts.^9,21 By allowing sweat to be collected for delayed analysis, these devices enhance the reliability of biomarker detection, making them well-suited for applications in athletic performance monitoring and workplace health surveillance, where immediate testing may not always be feasible.^7,22,23

Finger-actuated pumps offer several advantages over traditional microfluidic systems that rely on continuous capillary flow or external pumps.^17,18,24 By giving users control over the amount of sweat collected, they ensure reproducible sample volumes, which is crucial for accurate biochemical assays particularly for low-concentration biomarkers such as glucose or stress-related hormones, which can be difficult to detect in real time due to dilution or variability.^25,26 Moreover, by storing sweat in a dedicated chamber, the pump minimizes issues such as motion artifacts, evaporation, or contamination that could otherwise compromise the analysis.^18,27,28 This delayed analysis capability is particularly important in scenarios where real-time testing is impractical or disruptive, such as during athletic activities or in extreme environments like military or firefighting.^3,29–31 In these cases, users can collect sweat throughout the activity and conduct testing at a more convenient time, enhancing both usability and accuracy. Furthermore, the need for more precise manual control of fluid flow has led to the development of manual rotation control mechanisms. These systems, which use rotational movements to regulate fluid flow, offer adjustable sample volumes and ensure more controlled delivery.³² The key advantage of this technology is its ability to complement other fluidic components, such as finger-actuated pumps, to provide a more versatile and user-friendly solution for sweat sampling.³³ The ability to manually rotate and control the fluid flow enhances the customization of fluid delivery, making it adaptable to various testing needs and more suitable for point-of-care applications. For example, Hussain et al.³⁴ (2021) demonstrated the use of rotational active valves (RAVs) for adjustable flow velocities in paper-based devices, while Mishra et al.¹⁷ (2019) integrated finger-actuated pumps with CBVs to create a reliable fluid management system for wearable sensors. These features reflect a growing trend towards manual fluid control mechanisms, offering the potential to improve precision, simplicity, and energy efficiency in wearable applications.^22,35

Nevertheless, most wearable sweat sensors function in a single operational mode either real-time continuous monitoring or on-demand sample collection.^9,36–38 While real-time sweat sensing allows for continuous monitoring of biomarkers, it often suffers from uncontrolled fluid dynamics, leading to variations in sample volume and dilution effects that can compromise assay accuracy. On the other hand, on-demand collection systems enable more controlled analysis but often require complex fluid handling mechanisms, limiting device wearability and practicality.^1,38,39 Moreover, many existing sweat sensors still lack precise sample volume control, which is crucial for ensuring reproducible biochemical assays, especially in colorimetric and electrochemical sensing applications.¹³ Thus, the absence of a versatile system that enables both continuous and controlled fluidic operation remains a fundamental challenge in wearable microfluidic sensor technology.

To overcome these limitations, we introduce a novel microfluidic device that seamlessly integrates two key fluid control mechanisms: a manual rotating fluid control system and a finger-actuated pump. This dual-functionality design enables flexible, dual-mode sweat sampling, catering to both real-time and delayed biomarker detection needs. The manual rotating control mechanism allows for the precise redirection of sweat flow within the system, offering users the ability to direct sweat either into detection chambers for real-time analysis or into the finger-actuated pump for storage and controlled analysis at a later time. When immediate analysis is required, the sweat is routed into the real-time detection chambers, where sensor reactions can occur independently of the sample volume, making this mode ideal for applications where biomarker fluctuations are minimal or the sample volume does not significantly impact assay outcomes. In contrast, when more controlled sampling is necessary, the rotational control system directs the sweat into the finger-actuated pump. This pump is designed to collect and store the sweat within a dedicated storage chamber, ensuring that the sample can be analyzed at a convenient time without interference from external environmental factors or motion artifacts. The finger-actuated pump plays a critical role in ensuring sample volume consistency. By allowing the user to control the amount of sweat collected and stored, it ensures reproducible biochemical assays, particularly for biomarkers with low concentrations, which might otherwise be difficult to detect in real time due to sample dilution or variability. The stored sweat can be delivered to the reaction chambers at a controlled rate, enabling delayed but accurate biomarker analysis. This design not only mitigates issues associated with fluctuating sample volumes but also allows for the integration of time-controlled, precision detection methods. We validated the device's performance during a 40 minute exercise session, using a colorimetric glucose assay to assess its reliability in sweat collection and biomarker detection. The results demonstrated accurate sweat sampling, effective fluid control, and reproducible biochemical analysis, confirming the device's potential for practical applications in personalized health monitoring and field-based diagnostics. This approach represents an advancement in wearable sweat sensing, bridging the gap between real-time monitoring and precise biochemical quantification.

2 Experimental

2.1 Materials and reagents

The following materials were used in the study: Bambu Lab PLA Basic, Bambu-TPU 95A (thermoplastic polyurethane), PET (polyethylene terephthalate) film, and 3M 467MP adhesive tape. Fabrication was carried out using a Bambu Lab P1P 3D printer and a Universal Laser ULTRA R9000 machine. A Harvard Apparatus syringe pump (Pump 11 Elite 11) was used for controlled fluid delivery, while a force sensor (FSR-400) and a SparkFun Thing plus ESP32 microcontroller were integrated for device functionality, with a 4.7 kΩ resistor included in the circuit. The biosensor used for device performance validation was prepared following methods reported in previous studies.⁴⁰ All other chemicals, including glucose, were purchased from Sigma-Aldrich and used as received.

2.2 Fabrication of the microfluidic system

The fabrication process began with initial sketches that focused on key components: a fluid control system, a pump, channels, and a detection mechanism. These sketches guided the engineering of a manual rotating mechanism for easy fluid flow control. CAD software, including Fusion 360, SolidWorks, and AutoCAD 2023, was used to design and create the microfluidic model. The coated layer and adhesive layer were assembled on Fusion 360 and SolidWorks ensuring that each component—microfluidic channels with capillary burst valves, the finger-actuated pump, the pump shield cover, the manual control mechanism, and the top replaceable cover and chambers—was precisely sealed. The design was validated through fluid flow simulations in COMSOL. The device was fabricated with eight layers, excluding human skin, arranged from bottom to top (Fig. 1a). A Bambu Lab P1P machine, equipped with a 0.2 mm nozzle and Bambu Studio V 01.10 software, was used for slicing and printing the microfluidic device body using a white PLA material, with a diameter of 40 mm and a thickness of 3 mm. No supports were required during printing, which took approximately 30 minutes. The same method was employed to fabricate the manual rotation fluid control, utilizing the bottom-up printing technique with a 0.2 mm nozzle, as well as the pump protection shield cover. After printing the device body and control mechanism, the printing material and nozzle were changed to a flexible TPU A92 material with a 0.4 mm nozzle size to print the finger actuator pump. A heat-resistant transparent polyester film (PET) covered with 3M 467MP adhesive tape was prepared to create the replaceable transparent cover for the entire device and chambers, as well as to form the detection paper lamination and skin adhesive layer. Following the creation of the drawing pattern in AutoCAD 2023, the exported files from SolidWorks and Fusion 360 were used in a Universal Laser System machine interfaced with VLS3.50 software to cut the top capping cover (40 mm in diameter) and the replaceable chamber covers (3.3 mm in diameter). The cutting machine settings for the PET lamination sheet were set to 35% power, 30% speed, 0.5 mm Z-axis distance, 80% vector pen up speed, 0.1 mm lens size, and 5% rotational speed. The same settings were applied to cut the medical adhesive layer. The detection paper was made from Whitman Grade 1 filter paper. Using the same laser system, the following settings were applied for cutting: red vector patterns, 7% power, 20% speed, 0.3 mm Z-axis distance, 30% vector pen up speed, 0.1 mm lens size, and 5% rotational speed to cut the 3.3 mm diameter circles. The detection paper was then carefully placed and adhered to the laminated part with the opening. After all parts were fabricated, they were cleaned and sterilized before using them.

Fig. 1 Schematic representation of the sweat microfluidic system and its operation mechanism. (a) Elements of the device (exploded view): includes a replaceable capping cover layer, main device cover layer, finger actuator pump, pump shield cover, detection paper, lamination basement, manual rotational control, reaction chambers, pump storage chamber, microfluidic device body, and skin adhesive layer. (b) Detailed device operation schematic for CBV, inlet, and outlet valves: regulates sweat flow patterns through the ducts. Detection paper: enables sweat analysis with multiple parameters in the chambers. Users can twist components to direct perspiration to the appropriate detection chamber using manual fluid control. (c) Demonstration of how the device is worn on the body (forehead and chest) during physical activity (workout). (d) Workflow for device usage: set the mode (choose between direct detection or storage sample modes), place on skin, sweat collection (sweat is absorbed during activity), data analysis (capture the detection paper image with a smartphone and perform colorimetric analysis using a computer), and a final step is to replace used components.

2.3 On-body human protocol test

Under protocol number SIAT-IRB-250115-H0954, the experimental technique of human research was examined and authorized by the Chinese Academy of Sciences' Shenzhen Institute of Advanced Technology's Institutional Review Board for Human Participants. Fig. 1a shows a perspective view of the device elements.

Fig. 1b offers a detailed view of the internal mechanisms, highlighting the outlet for collected sweat, the detection chamber, the main inlet, the sweat loss channel for measuring the sweat rate, and the CBV valves. Fig. 1c demonstrates the device's position on the forehead and chest, emphasizing the skin adhesive layer that ensures secure attachment and prevents leakage. Fig. 1d outlines the operational workflow, where users set the mode for direct analysis or sample storage, activate the pump by pushing the finger actuator, and collect sweat during exercise. After collection, users can analyze results via color analysis of the detection paper and easily replace it for multiple tests, showcasing the device's effectiveness in real-time sweat monitoring.

3 Results and discussion

3.1 Design and fabrication of the microfluidic device

Wearable sweat analysis devices face significant challenges in achieving reliable fluid control and sample handling under diverse conditions without relying on bulky, power-dependent systems.²⁶ Current devices often require automated setups or external power sources, which, while precise, compromise portability, user accessibility, and affordability – all critical factors for point-of-care testing (POCT).^1,41,42 Moreover, effectively managing small sample volumes, particularly for real-time or delayed sweat analysis, remains a persistent limitation.⁴³ Conventional systems often lack the flexibility to operate in dual modes – either for immediate or stored sample analysis – and struggle with fluid manipulation in dynamic environments.^44,43 These gaps necessitate innovative designs that integrate advanced fluid control mechanisms with intuitive user operation to enhance the versatility and reliability of sweat-based diagnostics.²⁹ To address these challenges, we developed a circular, skin-adherent microfluidic device for on-demand sweat analysis. The device integrates multiple components to enable efficient sweat collection, transport, storage, and analysis. Key innovations include a finger-actuated pump and a manual fluid control mechanism, integrated with a long, spirally curved channel and dual detection chambers for sweat rate monitoring and biomarker analysis in both real-time and on-demand scenarios.

The device was fabricated using a combination of laser cutting and 3D printing techniques, ensuring low cost, precision and reproducibility in its structural elements. Microfluidic channels were optimized to maintain consistent flow resistance while minimizing dead volume. A biocompatible adhesive layer ensures stable placement on the skin during extended sweat collection periods. As illustrated in Fig. 1a, the layered assembly includes replaceable chambers and a cover layer for repeated use and integrates detection paper within the chambers for real-time colorimetric glucose analysis. At the base of the device, a medical adhesive layer (0.2 mm thick) secures it to the user's skin. Two separate strategically positioned inlets capture sweat through direct skin contact. The main inlet, with a diameter of 2.6 mm, connects to the manual fluid control system, while the secondary inlet, slightly smaller (1.3 mm) in diameter, leads to an 800 mm-long, helix-shaped channel (0.6 mm × 1.3 mm in width and height). This channel, ending in a 0.6 mm outlet, can hold approximately 370 μL of sweat, enabling measurement of sweat loss during exercise. Central to the device's functionality is the manual rotating fluid control mechanism, located at its core. Measuring 6 mm × 5 mm, the mechanism allows users to direct sweat flow by rotating left or right at a 79.7° degree angle. This allows users to channel sweat either to the detection chambers for onsite biomarker analysis or to the finger-actuated pump chamber for storage and delayed analysis. The control mechanism includes a one-direction interior channel (0.8 mm × 1.3 mm in width and height) and is complemented by a skin adhesive layer for secure placement during use.

For sweat storage, the finger-actuated pump chamber receives sweat via a channel (0.7 mm × 1.3 mm in width and height). The pump chamber has a size of 11 mm × 2 mm in width and height, with a volume of approximately 450 μL, and is strategically positioned in the upper-left quadrant of the device. Constructed from a flexible TPU A92 material, the pump mimics a spring-like structure, facilitating efficient fluid movement while accommodating varying finger pressures. With a diameter of 13 mm and a height of 10 mm, the pump adds 2 mm to the device's height (total height: 12 mm). A protective shell encases the pump, ensuring that pressure is applied only through intentional finger contact.

The pump chamber connects to two separate channels: one from the inlet, directed by the manual rotating mechanism, and the other leading to the detection chambers for delayed biomarker analysis. To enhance fluid regulation, two series of capillary burst valves (CBVs) are integrated into the outgoing channel from the pump. These CBVs, designed with a β = 90° geometry, act as passive barriers, preventing premature fluid movement. By exploiting capillary pressure differences, the CBVs ensure that sweat is released into the detection chambers only after sufficient volume has accumulated, improving sample collection and analysis control.

The CBVs lead directly to a connection point, from which sweat is distributed to the two detection chambers via independent sub channels. Each detection chamber has a diameter of 6.6 mm and a total height of 1.3 mm, with an internal middle layer of 0.7 mm, allowing it to hold approximately 10–12 μL without the detection paper. Inside the chamber, Whitman Grade 1 filter paper (3.3 mm in diameter and 0.1 mm thick) serves as the detection substrate. The paper is supported by a laminated PET base with an adhesive layer (3.3 mm in diameter and 0.3 mm thick) with a small hole in the center (0.75 mm in diameter) for sweat entry. This design ensures that when the pump is pressed, only a small amount of liquid enters the chamber, comes into contact with the detection paper, and exits through an outlet.

The synergy between the finger-actuated pump, rotating fluid control mechanism, CBVs, and structured microfluidic channels ensures efficient sweat transport and controlled sample handling. This combination of active and passive fluid control mechanisms enhances the accuracy and reliability of sweat-based diagnostics. By supporting both on-demand and continuous sweat analysis, the device demonstrates its potential as a versatile, non-invasive platform for biomarker detection in POCT applications.

3.2 Design, fabrication and performance validation of the fluid control mechanisms

Effective fluid control is crucial in microfluidic applications to ensure precise flow regulation and prevent premature movement. In this study, three key fluid control mechanisms capillary burst valves (CBVs), a manual rotating fluid control system, and a finger-actuated pump were designed, fabricated, and validated for efficient microfluidic operation. These components were systematically evaluated through experimental and computational methods, including pressure calculations, material optimization, and force analysis, to ensure their functional reliability and performance. The integration of passive (CBVs) and active (manual control and pump) mechanisms enabled a well-regulated fluid flow system. The CBVs ensured pressure-sensitive flow control, the rotating control allowed directional adjustments, and the finger-actuated pump provided user-driven pressure modulation.

3.2.1. Design and pressure calculation of the CBVs. Capillary burst valves (CBVs) were designed to regulate passive fluid flow by preventing movement until a designated burst pressure (BP) threshold was exceeded. To achieve precise control, two CBVs (with identical burst pressures) were positioned at the output of the storage pump chamber, ensuring controlled liquid release at the junction of two channels. To calculate the burst pressure (BP) using the Young–Laplace equation in a rectangular channel, we can use the formula provided:
where σ is the liquid's surface tension, θ_A is the contact angle in the vent's channel is the min [θ_A + β; 180°], β is the diverging angle of the channel, and b and h represent the width and height of the diverging section. The BP rises as b and h decrease for hydrophobic materials at large divergence angles. The CBV has widths of 0.2 mm and divergence angles of 90°. The equation indicates that the burst pressure (BP) is influenced by the contact angle of the channel surfaces. For the typical 3D-printed PLA material using a 0.2 mm printing nozzle, the surfaces in contact with water have an average contact angle of approximately 80°.⁴⁵ The calculated burs pressure for the CBV is around 677.4 Pa based on this parameter. Primarily due to the final printing finishes, experimentally measured values are marginally lower than these estimates.

A Harvard Apparatus 11 Elite syringe pump was used to experimentally determine BP by applying incremental pressure until fluid breakthrough was observed. The measured BP closely matched theoretical values, confirming the valve's functionality (the experimental setups can be seen in Fig. S1†). The channel width (0.2 mm) and height (0.5 mm) played a crucial role in defining flow characteristics and ensuring that bursting occurred only at the designed pressure threshold. Fig. 2a illustrates the CBVs' configuration within the microfluidic system, including the diverging angle and key dimensional parameters. These results demonstrate the effectiveness of CBVs in sequential reagent delivery for microfluidic applications, preventing premature movement while ensuring precise pressure-dependent fluid flow.

Fig. 2 Illustrates the design and structure dimensions of the microfluidic device and rotating controller. (a) shows the top view of the device, highlighting the capillary burst valves (CBVs) and the interior channel design, with key measurements for each component, including various channel widths (0.2 mm, 0.8 mm, and 0.5 mm). The inset emphasizes the orientation of the CBVs. (b) features the bottom view of the manual control mechanism, detailing its dimensions (6 mm and 5.3 mm) and the rotational angle (79.2°) for fluid flow control.

While the current design of the check valve system (CBVs) demonstrates reliable performance under typical operating conditions, it faces limitations when subjected to high sweat pressures. The burst pressure (BP) of the CBVs, measured to be approximately 677.4 Pa, is significantly lower than the reported sweat pressure range of 2–3 kPa. This discrepancy indicates that the CBVs alone cannot fully endure the pressure exerted by sweat when the storage chamber is full. However, the device incorporates several design features to mitigate this issue, including a large pump chamber and optimized flow dynamics, which help distribute pressure and reduce direct stress on the valves. These features enable the CBVs to function effectively under lower-pressure conditions, ensuring reliable operation in most scenarios. Previous studies have reported BP values for comparable microfluidic systems, including BP values reported by Kang et al.,⁸ Choi, J. et al.¹⁹ and Qi et al.⁴⁶ In comparison, our device demonstrates a higher BP at some point, indicating superior fluid retention and controlled release capabilities. This enhanced performance ensures greater reliability in on-demand sweat analysis applications, where precise fluid regulation is essential for effective biomarker detection and real-time monitoring.

3.2.2. Fabrication and optimization of the manual rotating fluid control mechanism. Unlike capillary burst valves (CBVs), which passively regulate fluid movement, the manual rotating control mechanism was developed to provide active, user-driven fluid redirection. This mechanism operates through bidirectional manual rotation, allowing users to control fluid flow with precision. The design features serrated edges at the top, enhancing grip and ease of operation. With a 79.7 degree rotation angle, the controller is equipped with mechanical stops at predetermined transition points (Fig. 3) to ensure stable and accurate positioning, preventing accidental misalignment. These stops also help eliminate partial or incomplete switching, ensuring that the user cannot switch mid-way. Additionally, the ergonomic design of the rotation stops enhances usability, minimizing the risk of unintended switching. The microchannels within the rotary system are optimized to maintain smooth fluid flow, reducing the risk of turbulence and leakage while ensuring efficient operation. The fabrication process employed a Bambu-PLA material, chosen for its mechanical stability and printability. To enhance print resolution and surface smoothness, two 3D printing orientations were tested: bottom-top and top-bottom printing. Comparative analysis showed that bottom-top printing resulted in superior edge definition and minimal roughness, leading to better fluidic sealing properties (Fig. 3a and b). Additionally, nozzle size optimization was performed using 0.2 mm and 0.4 mm nozzles, with the 0.2 mm nozzle in bottom-top printing yielding the highest precision and sealing efficiency. Fig. 3c provides a comparative analysis of print resolution and surface smoothness under different fabrication conditions. These findings highlight the importance of optimized printing strategies in ensuring mechanical accuracy and functionality in microfluidic control component.

Fig. 3 Illustrates the 3D printed manual rotation fluid controller of the microfluidic device, comparing two different printing methods and nozzle sizes: 0.2 mm and 0.4 mm. (a) features the bottom view of the components. (b) highlights Top differences of the printed structure. (c) presents the front view, showing the fabrication of the main fluid path using different printing methods.

3.2.3. Design and performance validation of the finger-actuated pump. To enable a manual, power-free fluid displacement, a finger-actuated pump was designed and fabricated using Fusion 360 CAD software and printed with a flexible TPU A92 material. The pump structure featured a 0.5 mm wall thickness, optimizing flexibility while maintaining mechanical integrity (Fig. 4).

Fig. 4 Illustrates the functionality and performance of a fluid control system. (a) shows the pump's position and test fluid channel, along with a force sensor measuring the applied force during operation. (b) presents a graph comparing the applied force and estimated pressure against the fluid volume, demonstrating the relationship between these variables. (c) further explores this relationship with a fitted linear regression. Finally, (d) provides a front view and dimensions of a component, highlighting its pump design features, including three curved cones.

To prevent accidental activation of the finger-actuated pump during use, a protective shield was incorporated into the device design. The shield, fabricated from PLA Basic White—a durable and rigid material—is shaped as a shell and securely fixed onto the pump during assembly. It measures 12.3 mm (length) × 6 mm (height) × 11.4 mm (width) with an internal thickness of 1.5 mm. The shield features two internal engravings on both sides, ensuring that it remains firmly in place even during vigorous physical activities. This design not only safeguards against unintended contact but also enables controlled fluid movement by allowing users to apply deliberate pressure when needed. The integration of this protective component enhances the device's reliability and usability in real-world applications. A demonstrated video test can be seen in the ESI† (Video S1 pump protection shield). The internal shell design facilitated smooth compression and rebound, making the pump highly responsive to applied force as shown in Fig. S2.†

The finger-actuated pump was selected over a pneumatic pressure chamber for our device due to its portability, ease of use, and cost-effectiveness.²⁴ The finger pump offers a power-free method for fluid manipulation, ideal for on-the-go, point-of-care applications where external power sources are often unavailable.^46–48 It allows direct user control, making it intuitive and compact, unlike pneumatic systems that require complex infrastructure, such as compressors and valves.^24,49,50 Finger pumps are simpler to maintain and less prone to failure compared to pneumatic systems, which can face issues like air leaks and pressure instability.^18,46

To evaluate its performance, a force sensor integrated with an ESP32 microcontroller was used to measure real-time pressure variations (Fig. 4a). The force applied by the user was calculated as:

F = V_output × C

where F represents the applied force (newton), V_output is the sensor's voltage output, and C is a calibration constant.

The resulting pressure exerted by the pump was determined using:

where P is the pressure (pascal), F is the applied force, and A is the contact area.

A fluid conduit system (0.7 mm diameter) was constructed to validate performance, where controlled volumes of liquid were introduced to measure force–displacement characteristics. The data presented in Fig. 4b and c illustrate the inverse correlation between the fluid volume and the applied force required to operate the finger-actuated pump. As the fluid volume increases, the necessary actuation force decreases, which can be attributed to the underlying fluid mechanics governing the microfluidic system.

In Fig. 4b, the trend shows a clear reduction in applied force as the fluid volume increases from approximately 80 μL to 180 μL. This relationship is further validated by the estimated pressure values, which exhibit a corresponding decline. The reason for this behaviour can be explained by considering the hydrodynamic resistance within the microfluidic channel. When the fluid volume is lower, the reduced hydraulic head results in higher resistance, requiring a greater applied force to drive fluid movement. Conversely, as the fluid volume increases, the available hydrostatic pressure assists in reducing the external force required for actuation. Fig. 4c further substantiates this trend through a regression analysis, revealing a strong negative correlation (Pearson's R = −0.98087) between the applied force and fluid volume.

The near-linear relationship suggests a predictable dependence, which is essential for optimizing the pump design for practical applications. The observed correlation also indicates that minor variations in fluid volume led to proportionate changes in the force needed for operation, an important consideration for ensuring user-friendly, low-effort manual actuation in portable and point-of-care applications. These findings emphasize the efficiency of the finger-actuated pump design, where increasing the fluid volume inherently reduces the operational effort. This characteristic is particularly advantageous for wearable or disposable diagnostic systems, ensuring minimal user fatigue while maintaining effective fluid transport within the microfluidic network.

3.3 Material and printing optimization for enhanced fluid control via CBVs

The effective regulation of fluid dynamics within microfluidic systems relies heavily on the interplay between material properties and fabrication techniques. In this study, we evaluated the impact of various materials and printing parameters on the functionality of capillary burst valves (CBVs), which are critical components for achieving controlled fluid flow. The objective was to optimize these parameters to enhance device performance and reliability, particularly in diagnostic applications where precise fluid control is essential. Devices were fabricated using three materials—Bambu-TPU 95A, Bambu Resin, and Bambu-PLA—each selected for their unique mechanical and chemical properties. Two nozzle sizes, 0.4 mm and 0.2 mm, were used during the fabrication process to ensure consistency in print quality while exploring variations in structural resolution. The results, illustrated in Fig. 5, revealed distinct differences in printing quality channel shape across materials and nozzle configurations. For example, the Bambu-PLA material demonstrated superior workability, allowing the CBVs to withstand higher burst pressures without structural failure, as shown in Fig. 5c and d. In contrast, resin-based materials and Bambu-TPU-95A devices were not suitable for the development of the device, due to the limited reliability of their operation.

Fig. 5 Illustrates different microfluidic device prototypes using various materials and nozzle sizes. (a) features a device printed with a 0.4 mm nozzle and a TPU material, while panel (b) displays one made with Bambu resin. (c) shows a device printed with a 0.4 mm nozzle using another material, and panel (d) depicts a 0.2 mm nozzle design in PLA. (e) and (f) illustrate the functionality of the capillary burst valves (CBVs), demonstrating their breakage in (e) and their ability to stop fluid flow in (f).

We observed the effect of the nozzle printing size on the CBV function. Devices printed with a 0.4 mm nozzle exhibited greater variability in burst pressure thresholds compared to those fabricated with a 0.2 mm nozzle. This discrepancy can be attributed to inconsistent material deposition at critical points, which compromised the structural integrity of the larger nozzle-printed CBVs. Fig. 5e and f provide visual evidence of these differences, highlighting scenarios where CBVs failed due to breakage (“CBV-Break”) versus successfully halting fluid flow (“CBV-Stop fluid”).

We tested different materials for printing the microfluidic device, and the best choice for us was Bambu-PLA with a 0.2 mm nozzle. These results emphasize the importance of precise printing parameters in maintaining valve reliability. The mechanisms behind these observations can be linked to the inherent properties of the chosen materials. However, Bambu-TPU 95A was not a good choice for fluid dynamics, due to its limited ability to print clean, smooth channels, leading to premature failure. Conversely, the stiffness of PLA materials with a 0.2 mm nozzle mitigated stress build up at the valve edges while being beneficial for static burst pressure. Such findings are consistent with previous studies that highlight the advantages and importance of tailoring material selection to specific application requirements (Hou, X. et al.).⁵¹

These insights have broader implications for the design and development of microfluidic devices. By optimizing material properties and printing parameters, the reliability of fluid regulation systems such as CBVs can be significantly improved, thereby enhancing the performance of diagnostic devices (Racaniello, G. F. et al.).¹¹ This is particularly relevant for applications in wearable diagnostics and point-of-care testing, where precision and durability are paramount. Future research could explore hybrid materials that combine the elasticity of TPU with the structural rigidity of PLA to achieve an optimal balance of performance characteristics. By addressing key challenges in fluid regulation and valve reliability, this work contributes to the broader goal of developing robust and adaptable diagnostic tools for diverse biomedical applications.

3.4 Optimization of fluid control mechanisms for efficient sweat delivery to detection chambers

Precise fluid control is critical in microfluidic devices designed for point-of-care biomarker analysis, where sample volume and flow dynamics directly influence test accuracy and reproducibility. In this section, we present the design optimization and evaluation of the fluid control mechanism in our microfluidic device, ensuring efficient sweat delivery to the detection chambers. Given the variability in applied force from the finger-actuated pump, controlling fluid dynamics is essential to maintain consistent reagent interaction and reliable analytical outcomes. The original device design featured a finger-actuated pump for fluid manipulation, coupled with capillary burst valves (CBVs) to prevent backflow. However, uncontrolled fluid entry into the chambers led to overfilling and the washing away of the detection paper's colour.

Additionally, fluctuations in fluid volume (e.g., sweat) caused inconsistent colorimetric results, undermining the device's reliability. Fig. 6a illustrates the initial design, highlighting key features, such as the pump chamber, detection paper, and fluidic channels. To overcome these challenges, we redesigned the device chambers, increasing the diameter to 6.6 mm and the height to 1.3 mm. A middle-extruded layer (0.7 mm) was added to elevate the detection paper (3.3 mm diameter). A lamination base with a 0.75 mm central hole was also introduced to control fluid entry. This updated design ensures effective wetting of the detection paper while allowing excess fluid to exit through the outlets. Fig. 6b and c compare the fluid dynamics and performance between the initial and redesigned chambers, demonstrating notable improvements in control and functionality (ESI† Video S2 one chamber fluid test and Video S3 pump fluid test).

Fig. 6 Illustrates the design and functionality of volume control in the microfluidic device chambers and fluid test. (a) The design construction parts. (b) compares devices with and without paper inserts, showcasing the differences in fluid behaviour. (c) Chamber configurations with and without controlled lamination. (d) Simulation for the fluid behaviour within the device during different times.

To validate the device's performance, fluid dynamics were simulated using the COMSOL Multiphysics program. Simulations were performed at time intervals of 0, 240, 600, and 1200 seconds, with an inlet flow rate of 30 μL s⁻¹. The resulting color maps illustrated the progression of fluid through the microfluidic channels, showing controlled entry and exit within the chambers. In the experimental phase, the finger-actuated pump and detection chambers were assessed for precision in fluid handling. Additionally, the use of CBVs was tested to prevent backflow and ensure the integrity of the fluid flow. Initial tests were conducted using both the original and optimized designs to compare their performance. Results revealed that the optimized design effectively controlled the volume of fluid entering the detection paper chamber, preventing overflow and enhancing colorimetric accuracy.

Experimental observations confirmed that the redesigned chambers facilitated consistent fluid distribution, enabling reliable analysis. Fig. 6d presents the simulation for the fluid behaviour within the device over times, validating the improvements made through optimization. The simulation results closely aligned with experimental data, confirming the accuracy of the fluid dynamics model. Significant improvements in fluid control were observed with the optimized design. Minor discrepancies, where present, were attributed to variations in user-applied pressure during manual operation. To further assess the device's functionality, we tested it with and without the finger-actuated pump. The manual setup demonstrated the ability to handle fluid volumes ranging from 50 μL to 350 μL, with CBVs ensuring fluid integrity. Controlling the volume of sweat transported to the detection paper chamber is a critical factor in ensuring accurate and reproducible colorimetric results. To address this challenge, our device is designed with a finger-actuated pump that allows for on-demand, controlled delivery of sweat. Proper user operation is essential to minimize variability, and clear guidelines are provided to ensure that the chamber is adequately filled before applying pressure. The pump's depth prevents premature actuation by ensuring that liquid does not exit the chamber until an adequate volume is collected. Visual indicators, such as the liquid reaching and being stopped by the capillary burst valve (CBV), signal when the chamber is ready for operation, providing a reliable cue for users. The device's dual-chamber design further reduces variability: while one chamber receives the liquid first, the volume of sweat delivered to each chamber remains consistent, ensuring uniform conditions for the glucose sensing assay. Although visually confirming the pump chamber fill can be challenging in the current design, future iterations will incorporate transparent chambers, enabling real-time observation of sweat accumulation and preventing premature actuation. Additionally, visual indicators will be enhanced to clearly signal when the pump chamber is filled. In our current tests, we used high-intensity exercise to promote sweat production and visually confirmed that the chamber was sufficiently filled by observing the liquid reaching the CBV. This approach ensures that the amount of sweat interacting with the glucose sensing assay remains consistent, minimizing variability and supporting accurate and reproducible colorimetric results.

Fig. 7a and b illustrate fluid flow in the pump and detection paper directions, respectively, showcasing the system's adaptability to various volumes, a demo test can be seen in the ESI† (Video S4 rotational mechanism fluid test). CBVs played a critical role in preventing backflow and maintaining fluid control. Sweat entering the inlet chamber filled the pump chamber until the CBVs reached their burst pressure, preventing leakage. Upon activation of the finger-actuated pump, pneumatic pressure opened the CBVs, allowing precise fluid transfer to the detection chambers. Fig. 7c depicts the pump's operation and its impact on fluid flow, while Fig. 7d confirms the uniform filling of all paper chambers, an experiment test can be seen in the ESI† (Video S5 finger actuator pump pressure test).

Fig. 7 Demonstrates the operation of a microfluidic device for fluid sampling and flow control. (a) shows the sequence of fluid movement set to the direction of the pump chamber. (b) illustrates the change in the fluid direction to the detection paper and the effect of capillary burst valves (CBVs) on fluid flow. (c) and (d) Testing the finger actuator pump and activation and how the fluid is drawn into the pump chamber.

To address the challenge of flow rate control in wearable sweat analysis, our device incorporates a dual operating mode to accommodate different sampling requirements. In the passive mode, the device relies on natural sweat excretion without active flow control, similar to conventional sweat analysis systems. This mode is suitable for applications where precise flow rate regulation is not critical. In the active mode, the device uses a dual-chamber design, which includes a collection chamber and a secondary chamber for fluid transfer. The finger-actuated pump ensures that sweat is collected and delivered to the detection chambers in a consistent and reproducible volume. By manually applying pressure, the user can control the transfer of collected sweat, ensuring that sensor reactions, such as colorimetric assays, remain unaffected by fluctuations in the sample volume or flow rate. The finger-actuated microfluidic system offers significant advantages for POC diagnostics.¹⁷ Its power-free operation and user-friendly design make it particularly suitable for on-the-go applications, including sweat glucose analysis and other biochemical assays.²⁴

The improved fluid control mechanism ensures accurate and reliable results, addressing a critical need in resource-limited settings. However, variability in user-applied pressure may affect fluid flow consistency. Future efforts will focus on integrating automated actuation mechanisms and expanding the device's applicability to other biomarker analyses.

3.5 In situ application in sweat loss monitoring and on-demand sweat collection

An in situ test was conducted to evaluate the functionality of the sweat monitoring device on a 29 year-old male volunteer during a 40 minute physical exercise session using a treadmill at room temperature (∼25–29 °C). The device was placed on two body positions—the forehead and the chest to assess variations in sweat dynamics, including secretion and the flow rate, based on the anatomical location. The target areas were cleaned with alcohol and dried before placement to ensure proper adhesion and accuracy. During the exercise session, an infrared (IR) temperature sensor from Sinocare was used to monitor skin temperature in real time, capturing fluctuations in thermoregulatory responses.

The device effectively captured sweat as it began to flow from the skin's surface, using two strategically positioned inlets at the centre of its bottom side. The primary inlet (2.6 mm diameter) directed sweat into the manual fluid control system, while the secondary inlet (1.3 mm diameter) connected to a spiral-shaped channel (800 mm in length, 0.6 mm × 1.3 mm in width and height). These inlets interacted independently with distinct sweat pores, ensuring that the fluid collection from each was isolated and not influenced by the other. This design minimized potential cross-flow between the two channels by leveraging the natural distribution of sweat pores on the skin.

A picture of the device was taken at 5 minute intervals, and sweat volume was recorded based on the distance travelled, as well as the width and height of the channels. The sweat flow rate was determined by analysing the time required to cover a specific distance within the monitoring channels. This approach provided a comprehensive analysis of sweat dynamics under varying physiological conditions and device placements.

Fig. 8 presents the real-time performance of the sweat monitoring device under physiological conditions. Fig. 8b visually depicts the progressive sweat collection over time in the microfluidic chambers. The images show that sweat accumulates at a faster rate on the forehead compared to the chest, with noticeable differences emerging after 20 minutes. By 40 minutes, the forehead chamber is nearly full, while the chest chamber still contains significantly less fluid. This difference highlights variations in sweat gland density and localized heat dissipation. Fig. 8c quantifies sweat accumulation over time, demonstrating significantly higher secretion on the forehead compared to the chest. The forehead exhibited a rapid increase in sweat accumulation after approximately 15 minutes, while the chest displayed a slower and more gradual response. These results reinforce the anatomical differences in sweat secretion and flow. Fig. 8d and e establish a strong positive correlation between the sweat rate and temperature, with R² values of 0.99112 and 0.90686 for the forehead and the chest, respectively. This linear relationship underscores the device's reliability in tracking sweat dynamics in response to temperature variations, validating its potential for real-time thermoregulatory monitoring.

Fig. 8 Illustrates the on-body test for the wearable device testing real-time sweat loss during physical activity. (a) The device attached to a participant's body, positioned for fluid collection. (b) A series of images capturing the device's performance over time (0, 20, 30, and 40 minutes), highlighting its ability to collect sweat effectively. (c)–(e) depict graphs correlating sweat volume, temperature, and time, showcasing the device's responsiveness to physiological changes during exercise.

The results also showed that the device effectively controlled variability and provided accurate measurements of sweat loss over time. The variation in the sweat secretion rate between the forehead and chest was noticeable, as expected, due to anatomical differences in sweat gland distribution. The device's design allowed for efficient sweat collection from both sites, with the inlets accurately capturing sweat without significant interference from environmental factors. This confirms the device's capability to provide targeted monitoring of sweat dynamics, which is essential for accurate sweat loss measurement during physical activities. The ability to non-invasively collect sweat with high spatial and temporal resolution presents a valuable tool for personalized health monitoring and physiological studies.

3.6 Colorimetric glucose analysis as a proof of concept for applications in biomarker detection

The primary objective of this study is to establish a proof-of-concept for integrating a microfluidic device with a biochemical sensor, emphasizing the efficiency of sweat delivery via the finger-actuated pump. While the broader goal is to enable multi-biomarker analysis by equipping the detection chambers with distinct biochemical sensors, this work employs identical paper-based colorimetric glucose biosensors in both chambers to validate the system's functionality. During the in situ sweat rate monitoring test described in section 3.5, sweat entered the microfluidic device through dual inlets. One inlet directed the sweat to the sweat rate monitoring channels, while the other channelled it to the detection chambers. The manual rotatory fluid control mechanism allowed for the selection of either direct or indirect delivery to the detection chambers. In this experiment, sweat was first directed to collect at the finger-actuated pump chamber before reaching the detection zones. After completing an exercise session, fresh glucose biosensors were installed in both chambers to ensure the preservation of enzymatic activity. Initial trials using sweat directly from the pump did not yield a significant colorimetric response. This was likely due to the inherently low glucose concentration in sweat and the moderate sensitivity of the biosensors used. To address this limitation, serial dilutions of glucose (0, 100, 200, and 300 μM) were prepared using sweat collected separately by the user during exercise. This spiked sweat was used to assess the device's performance in biochemical sensor integration. Each concentration was introduced into the finger-actuated pump chamber at a controlled volume, simulating physiological sweat secretion. The pump then delivered the sweat to the detection chambers, as described earlier. Colorimetric changes were recorded by capturing images before and 3 minutes after sweat delivery. The change in red mean values (ΔR), extracted using Photoshop, represented the signal change for each glucose reaction.

Fig. 9a illustrates the progressive colorimetric changes in the detection chambers as the glucose concentration increases, confirming the sensor's ability to detect glucose variations.

Fig. 9 Device test for colorimetric glucose detection in sweat samples. (a) shows the device with different colour indicators in response to various glucose concentrations (0 μM, 100 μM, 200 μM, and 300 μM). (b)–(d) provide calibration graphs illustrating the change in chambers to glucose concentration, demonstrating a clear correlation between increasing glucose levels and the device's R response.

The quantitative results in Fig. 9b reveal that chamber 2 exhibited a consistently stronger response compared to chamber 1, despite both chambers being equipped with identical biosensors. Linear regression analysis in Fig. 3c and d further substantiates this observation, with chamber 1 yielding a response of y = 0.2074x + 3.947 (R² = 0.9257) and chamber 2 showing a higher sensitivity of y = 0.223x + 10.36 (R² = 0.953). The superior sensitivity observed in chamber 2 is attributed to its optimized microfluidic channel geometry, which facilitates more uniform and efficient sweat delivery. In contrast, chamber 1 experiences slight fluidic delay due to minor pathway resistance, resulting in a marginally weaker response. Nevertheless, both chambers exhibit excellent correlation with glucose concentration, reinforcing the robustness and reproducibility of the detection system.

An important implication of these findings is that the two chambers are not intended to serve as duplicate test sites but rather as independent units capable of simultaneous biomarker detection. This design flexibility enhances the applicability of the microfluidic platform for multi-analytical monitoring, paving the way for advanced wearable biosensors. Therefore, the integration of the microfluidic device with glucose biosensors demonstrated a strong correlation between the glucose concentration and colorimetric response, validating the system's potential for real-time biomarker analysis. The ability of the finger-actuated pump to efficiently transport sweat to the detection chambers underscores the robustness of the device, supporting its potential for future non-invasive health monitoring applications.

Conclusions

In summary, this study presents the design and development testing of a manual rotating mechanism fluid control microfluidic device with a finger-actuated pump for on-demand sweat analysis. The device is non-invasive, wearable, and reusable, and it offers flexibility for users during different activities, exercises, and other various scenarios. The device integrates manual rotating mechanism and finger-actuator pump-assisted fluid control systems with volume control design for sweat collection and colorimetric detection, enabling both immediate and delayed sample analysis. Laboratory tests demonstrated effective fluid flow, while on-body tests across various positions (forehead and chest) confirmed reliable sweat collection, with the forehead position yielding the highest and most consistent results. Performance validation under physical activity showed a positive correlation between sweat loss and body temperature while highlighting inter-individual variability.

Despite its promising performance, the device still faces some challenges, particularly with leakage in the rotating parts, which necessitates further optimization of the design and material selection. Another challenge is confirming when the pump chamber is filled in the current design. For future work, testing alternative materials, such as rubber or other options for the rotating mechanism, could enhance sealing and durability. Additionally, incorporating transparent chambers will allow for real-time observation of sweat accumulation and prevent premature actuation. Visual indicators will also be improved to clearly signal when the pump chamber is filled. Increasing the number of chambers and integrating different sensors for additional biomarkers will expand the device's functionality. We also aim to reduce the size and weight of the device for enhanced comfort and portability. Furthermore, adapting the device for broader applications, such as environmental monitoring, stress analysis, or drug testing, through customizable sensor modules, will increase its versatility. These advancements will refine the device for point-of-care (POC) monitoring, personalized healthcare, and real-time diagnostics.

Data availability

The data supporting this article, including experimental results and analysis, are included within the manuscript and its ESI.†

Author contributions

Mohamed Ishan Hassan Gama: design idea, conceptualization, visualization and fabrication, writing – original draft, methodology, investigation, formal and data analysis – review & editing, validation, and test conducting. Marwa Omer Mohammed Omer: validation, investigation, chemical preparation, formal analysis, review & editing. Saminu Abdullahi: writing – original draft, review & editing – chemical preparations, investigation, methodology, formal and data analysis. Zhu Yang: review & editing, chemical preparation. Wang Xuzhong: review & editing & test conducting. Yousuf Babiker M. Osman, Zhu Yang, Yuhang Liu, Jingzhen Li, Yingtian Li: review & editing, Zedong Nie: writing – review & editing, supervision, resources, project administration, funding acquisition.

Conflicts of interest

The authors declare that they have no conflicts of interest.

Acknowledgements

This work was supported by the National Key R&D Program of China under Grant No. 2022YFB3203702, the National Natural Science Foundation of China under Grant No. 62173318, the Science and Technology Service Network Plan of CAS-Huangpu Special Project under Grant No. STS-HP-202203, and the Key Laboratory of Biomedical Imaging Science and System, Chinese Academy of Sciences, and the State Key Laboratory of Biomedical Imaging Science and System. We would also like to express our sincere gratitude and appreciation to Dr. Xin Tong and Mr. Zhongqing Sun.

References

1. S. Apoorva, N. T. Nguyen and K. R. Sreejith, Recent developments and future perspectives of microfluidics and smart technologies in wearable devices, Lab Chip, 2024, 1833–1866.
2. L. Wei, G. Fang, Z. Kuang, L. Cheng, H. Wu and D. Guo, et al., 3D-printed low-cost fabrication and facile integration of flexible epidermal microfluidics platform, Sens. Actuators, B, 2022, 353, 131085,  DOI:10.1016/j.snb.2021.131085.
3. K. M. Clark and T. R. Ray, Recent Advances in Skin-Interfaced Wearable Sweat Sensors: Opportunities for Equitable Personalized Medicine and Global Health Diagnostics, ACS Sens., 2023, 8(10), 3606–3622.
4. H. Lin, J. Tan, J. Zhu, S. Lin, Y. Zhao and W. Yu, et al., A programmable epidermal microfluidic valving system for wearable biofluid management and contextual biomarker analysis, Nat. Commun., 2020, 11(1), 1–12,  DOI:10.1038/s41467-020-18238-6.
5. F. Gao, C. Liu, L. Zhang, T. Liu, Z. Wang and Z. Song, et al., Wearable and flexible electrochemical sensors for sweat analysis: a review, Microsyst. Nanoeng., 2023, 9(1), 1–21.
6. L. Tang, S. J. Chang, C. J. Chen and J. T. Liu, Non-invasive blood glucose monitoring technology: A review, Sensors, 2020, 20(23), 1–32.
7. F. Ju, Y. Wang, B. Yin, M. Zhao, Y. Zhang, Y. Gong and C. Jiao, Microfluidic Wearable Devices for Sports Applications, Micromachines, 2023, 14(9), 1792.
8. J. Choi, D. Kang, S. Han, S. B. Kim and J. A. Rogers, Thin, soft, skin‐mounted microfluidic networks with capillary bursting valves for chrono‐sampling of sweat, Adv. Healthcare Mater., 2017, 6(5), 1601355.
9. R. Ghaffari, D. S. Yang, J. Kim, A. Mansour, J. A. Wright and J. B. Model, et al., State of Sweat: Emerging Wearable Systems for Real-Time, Noninvasive Sweat Sensing and Analytics, ACS Sens., 2021, 6(8), 2787–2801.
10. Y. Zhang, Y. Chen, J. Huang, Y. Liu, J. Peng and S. Chen, et al., Skin-interfaced microfluidic devices with one-opening chambers and hydrophobic valves for sweat collection and analysis, Lab Chip, 2020, 20(15), 2635–2645.
11. G. F. Racaniello, T. Silvestri, M. Pistone, V. D'Amico, I. Arduino and N. Denora, et al., Innovative Pharmaceutical Techniques for Paediatric Dosage Forms: A Systematic Review on 3D Printing, Prilling/Vibration and Microfluidic Platform, J. Pharm. Sci., 2024, 113(7), 1726–1748,  DOI:10.1016/j.xphs.2024.04.001.
12. M. I. Hajam and M. M. Khan, Microfluidics: a concise review of the history, principles, design, applications, and future outlook, Biomater. Sci., 2023, 12(2), 218–251.
13. Y. Song, Y. Zhou, K. Zhang, Z. Fan, F. Zhang and M. Wei, Microfluidic programmable strategies for channels and flow, Lab Chip, 2024, 4483–4513.
14. M. Harun-Or-Rashid, M. N. Aktar, V. Preda and N. Nasiri, Advances in electrochemical sensors for real-time glucose monitoring, Sens. Diagn., 2024, 3(6), 893–913.
15. N. Brasier, J. Wang, W. Gao, J. R. Sempionatto, C. Dincer and H. C. Ates, et al., Applied body-fluid analysis by wearable devices, Nature, 2024, 636, 57–68,  DOI:10.1038/s41586-024-08249-4.
16. A. Ometov, V. Shubina, L. Klus, J. Skibińska, S. Saafi and P. Pascacio, et al., A Survey on Wearable Technology: History, State-of-the-Art and Current Challenges, Comput. Netw., 2021, 193, 108074,  DOI:10.1016/j.comnet.2021.108074.
17. N. Mishra, N. T. Garland, K. A. Hewett, M. Shamsi, M. D. Dickey and A. J. Bandodkar, A Soft Wearable Microfluidic Patch with Finger-Actuated Pumps and Valves for On-Demand, Longitudinal, and Multianalyte Sweat Sensing, ACS Sens., 2022, 7(10), 3169–3180.
18. N. N. Ahmad, N. N. Ghazali, A. T. Abdul Rani, M. H. Othman, C. C. Kee, P. K. Jiwanti and A. Rodríguez-Gómez, Finger-actuated micropump of constant flow rate without backflow, Micromachines, 2023, 14(4), 881.
19. J. Choi, Y. Xue, W. Xia, T. R. Ray, J. T. Reeder and A. J. Bandodkar, et al., Soft, skin-mounted microfluidic systems for measuring secretory fluidic pressures generated at the surface of the skin by eccrine sweat glands, Lab Chip, 2017, 17(15), 2572–2580.
20. J. Park, H. Roh and J. K. Park, Finger-actuated microfluidic concentration gradient generator compatible with a microplate, Micromachines, 2019, 10(3), 174.
21. N. Davis, J. Heikenfeld, C. Milla and A. Javey, The challenges and promise of sweat sensing, Nat. Biotechnol., 2024, 42(6), 860–871,  DOI:10.1038/s41587-023-02059-1.
22. L. Clime, J. Daoud, D. Brassard, L. Malic, M. Geissler and T. Veres, Active pumping and control of flows in centrifugal microfluidics, Microfluid. Nanofluid., 2019, 23(3), 1–22,  DOI:10.1007/s10404-019-2198-x.
23. C. Chen, Y. Fu, S. S. Sparks, Z. Lyu, A. Pradhan and S. Ding, et al., 3D-Printed Flexible Microfluidic Health Monitor for In Situ Sweat Analysis and Biomarker Detection, ACS Sens., 2024, 9(6), 3212–3223.
24. A. P. Iakovlev, A. S. Erofeev and P. V. Gorelkin, Novel pumping methods for microfluidic devices: a comprehensive review, Biosensors, 2022, 12(11), 956.
25. H. Shi, K. Nie, B. Dong, M. Long, H. Xu and Z. Liu, Recent progress of microfluidic reactors for biomedical applications, Chem. Eng. J., 2019, 361, 635–650,  DOI:10.1016/j.cej.2018.12.104.
26. A. M. V. Mohan, V. Rajendran, R. K. Mishra and M. Jayaraman, Recent advances and perspectives in sweat based wearable electrochemical sensors, TrAC, Trends Anal. Chem., 2020, 131, 116024,  DOI:10.1016/j.trac.2020.116024.
27. K. Guk, G. Han, J. Lim, K. Jeong, T. Kang and E. K. Lim, et al., Evolution of wearable devices with real-time disease monitoring for personalized healthcare, Nanomaterials, 2019, 9(6), 1–23.
28. X. Le, J. Chan, J. McMahon, J. A. Wisniewski, A. Coldham and T. Alan, et al., A Finger-Actuated Sample-Dosing Capillary-Driven Microfluidic Device for Loop-Mediated Isothermal Amplification, Biosensors, 2024, 14(9), 410.
29. N. L. Kazanskiy, S. N. Khonina and M. A. Butt, A review on flexible wearables – Recent developments in non-invasive continuous health monitoring, Sens. Actuators, A, 2024, 366, 114993,  DOI:10.1016/j.sna.2023.114993.
30. J. K. W. Lee, B. Tan, B. R. M. Kingma, F. Haman and Y. Epstein, Biomarkers for warfighter safety and performance in hot and cold environments, J. Sci. Med. Sport, 2023, 26, S71–S78,  DOI:10.1016/j.jsams.2022.12.006.
31. F. Shakeriaski and M. Ghodrat, Challenges and limitation of wearable sensors used in firefighters' protective clothing, J. Fire Sci., 2022, 40(3), 214–245.
32. G. Hussain, A. T. Jafry, S. Malik, S. F. Shah, S. Nishat and F. R. Awan, Multifunctional rotational active valve for flow control in paper-based microfluidic devices, Sens. Actuators, B, 2023, 378, 133142,  DOI:10.1016/j.snb.2022.133142.
33. S. Radhakrishnan, M. Mathew and C. S. Rout, Microfluidic sensors based on two-dimensional materials for chemical and biological assessments, Mater. Adv., 2022, 3(4), 1874–1904.
34. J. N. Hussain, N. Mantri and M. M. Cohen, Working up a good sweat - The challenges of standardising sweat collection for metabolomics analysis, Clin. Biochem. Rev., 2017, 38(1), 13–34.
35. S. Shrivastava, T. Q. Trung and N. E. Lee, Recent progress, challenges, and prospects of fully integrated mobile and wearable point-of-care testing systems for self-testing, Chem. Soc. Rev., 2020, 49(6), 1812–1866.
36. J. Heikenfeld, A. Jajack, J. Rogers, P. Gutruf, L. Tian and T. Pan, et al., Wearable sensors: Modalities, challenges, and prospects, Lab Chip, 2018, 18(2), 217–248.
37. B. Zhong, K. Jiang, L. Wang and G. Shen, Wearable Sweat Loss Measuring Devices: From the Role of Sweat Loss to Advanced Mechanisms and Designs, Adv. Sci., 2022, 9(1), 1–27.
38. M. Bariya, H. Y. Y. Nyein and A. Javey, Wearable sweat sensors, Nat. Electron., 2018, 1(3), 160–171,  DOI:10.1038/s41928-018-0043-y.
39. N. Kashaninejad, Lab Chip, 2023, 913–937.
40. W. Y. Li, S. Y. Lu, S. J. Bao, Z. Z. Shi, Z. Lu and C. M. Li, et al., Efficient in situ growth of enzyme-inorganic hybrids on paper strips for the visual detection of glucose, Biosens. Bioelectron., 2018, 99, 603–611,  DOI:10.1016/j.bios.2017.08.015.
41. R. Su, F. Wang and M. C. McAlpine, 3D printed microfluidics: advances in strategies, integration, and applications, Lab Chip, 2023, 23(5), 1279–1299.
42. H. A. Silva-Neto, I. V. Arantes, A. L. Ferreira, G. H. do Nascimento, G. N. Meloni, W. R. de Araujo, T. R. Paixão and W. K. Coltro, Recent advances on paper-based microfluidic devices for bioanalysis, TrAC, Trends Anal. Chem., 2023, 158, 116893.
43. U. A. Gurkan, D. K. Wood, D. Carranza, L. H. Herbertson, S. L. Diamond and E. Du, et al., Next generation microfluidics: fulfilling the promise of lab-on-a-chip technologies, Lab Chip, 2024, 1867–1874.
44. A. Kalkal, S. Kumar, P. Kumar, R. Pradhan, M. Willander, G. Packirisamy, S. Kumar and B. D. Malhotra, Recent advances in 3D printing technologies for wearable (bio)sensors, Addit. Manuf., 2021, 46, 102088.
45. J. Kingman and M. K. Dymond, Fused filament fabrication and water contact angle anisotropy: The effect of layer height and raster width on the wettability of 3D printed polylactic acid parts, Chem. Data Collect., 2022, 40, 100884,  DOI:10.1016/j.cdc.2022.100884.
46. W. Qi, L. Zheng, Y. Hou, H. Duan, L. Wang and S. Wang, et al., A finger-actuated microfluidic biosensor for colorimetric detection of foodborne pathogens, Food Chem., 2022, 381, 131801,  DOI:10.1016/j.foodchem.2021.131801.
47. P. Chen, C. Chen, H. Su, M. Zhou, S. Li and W. Du, et al., Integrated and finger-actuated microfluidic chip for point-of-care testing of multiple pathogens, Talanta, 2021, 224, 121844,  DOI:10.1016/j.talanta.2020.121844.
48. K. Iwai, R. D. Sochol and L. Lin, Finger-powered, pressure-driven microfluidic pump, Proceedings of the IEEE International Conference on Micro Electro Mechanical Systems (MEMS), 2011, pp. 1131–1134.
49. J. Xiang, Z. Cai, Y. Zhang and W. Wang, Mechanically programmed valving technology and the active flow switching application in centrifugal microfluidics, Sens. Actuators, B, 2018, 259, 325–331,  DOI:10.1016/j.snb.2017.10.182.
50. J. F. Hess, S. Zehnle, P. Juelg, T. Hutzenlaub, R. Zengerle and N. Paust, Review on pneumatic operations in centrifugal microfluidics, Lab Chip, 2019, 19(22), 3745–3770.
51. X. Hou, Y. S. Zhang, G. T. Santiago, M. M. Alvarez, J. Ribas, S. J. Jonas, P. S. Weiss, A. M. Andrews, J. Aizenberg and A. Khademhosseini, Interplay between materials and microfluidics, Nat. Rev. Mater., 2017, 2(5), 1–5.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5lc00171d

‡ These authors contributed equally to this work.

---