Bioinspired microtexturing for enhanced sweat adhesion in ion-selective membranes

With the advancement of wearable technology, real-time monitoring of physiological parameters such as hydration status and electrolyte balance has become increasingly important, especially during physical exercise. Dehydration and electrolyte imbalances (such as hyponatremia) can lead to serious health problems, yet current health monitoring devices mostly rely on conventional sweat sensors, which tend to fail under dynamic motion and typically depend on skin-adhesive interfaces that can cause allergies or irritation. Although existing ion-selective membrane (ISM) sensors can effectively detect electrolytes such as sodium ions in sweat, their stability under dynamic conditions is poor. In addition, the intrinsic hydrophobicity of the membrane limits efficient contact with sweat, thereby impairing sensor performance. Thus, improving the stability and reliability of sensors during movement, while avoiding the discomfort caused by adhesives, has become an urgent challenge. “To address these issues, we propose a bioinspired microstructuring design that mimics the surface microtexture of rose petals to enhance the wettability and self-cleaning properties of the ion-selective membrane, thereby improving sensor performance in dynamic environments, particularly for non-contact sweat monitoring.” said the author Marc Josep Montagut Marques, a researcher at Waseda University, “This innovation not only overcomes the limitations of traditional sensors but also enhances device comfort and wearability, promoting the practical deployment of health-monitoring technologies.”

The fabrication of the bioinspired rose-petal microstructured ion-selective membrane (ISM) sweat sensor can be summarized in two core steps. First, electrodes and bioinspired molds are prepared: multi-walled carbon nanotubes are dispersed in isopropyl alcohol, ultrasonicated, and vacuum-filtered to form a carbon nanotube forest (CNTF) sponge film, which is then dried at elevated temperature, cut into strips, and coated at one end with Ag/AgCl conductive ink using a blade-coating method to define the reference/contact electrode region. In parallel, the inner and outer surfaces of rose petals are used as templates; petals are pressed onto glass slides, PDMS is cast and cured on top, and the resulting PDMS replicas with negative rose-petal microstructures are peeled off to serve as elastic molds. Subsequently, the ion-selective membrane is formulated, replicated, and integrated with the electrode. A membrane cocktail containing sodium ionophore, a hydrophobic counter anion, PVC matrix, and the plasticizer DOS is prepared in THF and homogenized. This ISM precursor solution is spin-coated onto the PDMS molds bearing the rose-petal microtexture to form an approximately 100 μm thick sodium-selective membrane that replicates the surface microstructures. After solvent evaporation, the free-standing membrane is carefully detached, cut into small pieces, and bonded to the CNTF electrode tips using a small amount of precursor solution as an adhesive layer, yielding sodium ion-selective sweat-sensing electrodes with a rose-petal–inspired microtextured surface.

The characterization and experimental results of materials and devices can be divided into the following three aspects. First, in terms of surface wettability and adhesion behavior, contact angle and hysteresis measurements demonstrate that the bioinspired microtextures markedly improve the wettability and sweat adhesion of the ion-selective membrane. The unmodified flat ISM exhibits a contact angle of about 90°, whereas the bioinspired type A texture (derived from the wrinkles on the outer side of the rose petal) reduces the contact angle to approximately 76–77°, indicating enhanced wettability. The type B texture (derived from the “island–spike” structures on the inner side of the petal) slightly increases the contact angle but more closely mimics the complex wetting behavior of the natural petal. Compared with the baseline sensor C, surfaces A and B show significantly higher maximum in-plane and lateral water load, as well as greater self-adhesion capacity; notably, the self-adhesion of surface A increases by about 200%. Static and dynamic cycling tests under planar and lateral motion further show that surface A can retain droplets for many more motion cycles than the control, indicating that the wrinkle-inspired structure offers superior droplet retention and resistance to detachment under dynamic conditions. Second, regarding electrochemical performance and non-contact sensing capability, all three ISMs (A, B, and C) display near-Nernstian potential responses in 0–10⁻¹ M NaCl, with the bioinspired type B surface exhibiting the highest sensitivity at roughly 82% of the theoretical value, followed by type A at about 76% and the baseline C at about 74%. Geometrical analysis indicates that the bioinspired textures increase the effective interfacial area by approximately 16–22%, and this increased area is consistent with the observed trend in sensitivity enhancement. Using 3D-printed microchannels with different air gaps (0.5–2 mm), the authors further evaluate signal response and “self-cleaning” behavior under flow. The results show that the type A surface can still reliably maintain a liquid interface and avoid readout failure even at a 2 mm gap, and exhibits the smallest signal drift over multiple NaCl/deionized water flushing cycles, indicating excellent liquid retention and ion clearance under non-contact conditions. Finally, in wearable and on-body experiments, the bioinspired type A ISM is integrated into a rigid wrist-worn device with a 2 mm channel gap and tested in a ~20 min treadmill running session. As the channel transitions from air-filled to sweat-filled, the sensor achieves stable real-time Na⁺ potential monitoring and shows rapid, reversible responses when low-concentration NaCl solutions are externally applied to simulate “rewetting/recirculation,” while bubble-induced disturbances and motion artifacts remain limited. In a battery-powered, wireless version, additional tests involving horizontal and vertical arm swings are conducted; fast Fourier transform analysis reveals no evident noise peaks corresponding to the arm swing frequency, and the slopes from linear fitting are comparable to those in static experiments. These results confirm that the bioinspired structure provides robust signal stability and strong motion-artifact resistance under realistic movement conditions.

This article proposes and fabricates a bioinspired microtextured sodium ion-selective membrane (ISM) sweat sensor based on the surface microstructures of rose petals. Using a rose petal–PDMS–PVC replication process, the authors construct ISMs with wrinkle and island–spike microstructures and integrate them onto carbon nanotube forest (CNTF) electrodes, achieving non-contact sweat sodium sensing with air gaps up to 2 mm. Compared with conventional flat membranes, the bioinspired ISM exhibits enhanced wettability, liquid adhesion, self-cleaning behavior, and resistance to motion artifacts, and its practicality is validated in wrist-worn wearable tests. However, the replication fidelity of submicron structures is still constrained by the material properties of PDMS and PVC, leading to discrepancies from the native petal morphology; the mechanical durability of CNTF electrodes, ion accumulation, bubble interference, and signal drift under large channel gaps also remain partially unresolved, and the microtexture itself cannot fundamentally improve ion selectivity. “In future, we will focus on high-fidelity microstructure replication and advanced high-resolution polymer systems, extending the approach to multiple ion-selective membranes, developing more durable carbon-based composite electrodes with optimized microfluidic channel designs, and integrating these sensors with low-power electronics and machine learning algorithms to build intelligent, multi-parameter sweat health monitoring platforms for applications such as prosthetics, exoskeletons, and high-intensity work environments.” said Marc Josep Montagut Marques.

Authors of the paper include Marc Josep Montagut Marques, Takayuki Masuji, Mohamed Adel, Ahmed M. R. Fath El-Bab, Kayo Hirose, Kanji Uchida, Hisashi Sugime, and Shinjiro Umezu.

This work was supported by Japan Society for Promotion of Science under its Grants-in-Aid for Scientific Research KAKENHI grant numbers 24K21600, 23K26069, and 23K26077.

The paper, “Bioinspired Microtexturing for Enhanced Sweat Adhesion in Ion-Selective Membranes” was published in the journal Cyborg and Bionic Systems on Aug 5, 2025, at DOI: 10.34133/cbsystems.0337.