Atom-Thin Sensors Turn Breath and Sweat Into Real-Time Health Data

A new generation of wearable chemical sensors is quietly rewriting what a face mask, a patch of fabric, or a strip of paper can do. In a mini review published in Frontiers in Sensors, researchers at Gauhati University in India survey the rapid progress in flexible sensors built from two-dimensional (2D) materials—graphene derivatives, MXenes, and transition metal dichalcogenides (TMDs)—that detect volatile organic compounds (VOCs) and humidity in real time. These atomically thin materials combine enormous surface area, mechanical flexibility, tunable surface chemistry, and the ability to operate at room temperature, making them unusually well suited for devices that must bend, stretch, and sit against human skin for hours. The review, led by Anurag Kashyap, Rajnandan Lahkar, and Hemen Kalita, is notable for tackling VOC and humidity sensing together, because in the real world the two signals are inseparable: water vapor is always present in exhaled breath and ambient air, and it can swamp or distort the very chemical signals wearables are meant to capture.

The stakes are more than academic. VOCs such as ammonia, acetone, ethanol, and nitrogen dioxide are not only industrial pollutants linked to respiratory disorders, neurological damage, and cancer; several are also biomarkers that leak out of the body in exhaled breath. Ammonia in breath, for example, rises in chronic kidney disease, while acetone tracks lipid metabolism, and volatile sulfur compounds signal oral disease. Humidity, meanwhile, shapes health in its own right—high levels foster microbial growth and respiratory difficulty, low levels dry out airways, and both extremes affect indoor air quality, pharmaceutical stability, and the reliability of electronic products. A wearable that can read these variables continuously, comfortably, and cheaply would open the door to non-invasive diagnostics that today require clinic visits, blood draws, or bulky analytical instruments.

The physical platform matters as much as the chemistry. The review organizes wearable sensors by substrate: polymers such as polydimethylsiloxane (PDMS), polyethylene terephthalate (PET), polyimide (PI), and thermoplastic polyurethane (TPU), which are elastic, transparent, lightweight, and conform comfortably to skin thanks to their low glass transition temperatures; textiles, whose intrinsic roughness and porosity boost analyte adsorption and which can be functionalized by dip-coating, spray deposition, or screen printing; and paper, a biodegradable, porous, low-cost medium whose fast vapor diffusion makes it ideal for disposable point-of-care devices. Each choice involves trade-offs—polymers raise disposal concerns, textiles remain underexplored for VOC and humidity detection despite their comfort, and paper suffers from limited mechanical durability. Strikingly, most 2D-material sensors demonstrated so far still live on polymer or paper rather than fabric, a gap the authors flag as an emerging research frontier.

Underneath the variety of devices lies a common transduction principle. In chemiresistive sensors, the dominant architecture for wearables because of its simplicity and microwatt-to-milliwatt power draw, adsorbed VOC or water molecules donate or accept electrons depending on their molecular orbital energies relative to the Fermi level of the 2D material. Ammonia donates electrons to p-type graphene or WS2, reducing hole concentration and raising resistance; nitrogen dioxide withdraws electrons and lowers it. Surface chemistry fine-tunes this interaction: hydroxyl, carboxyl, and epoxy groups on graphene oxide act as preferential hydrogen-bonding sites for polar VOCs and water, while engineered defects and edge sites serve as high-energy binding locations that amplify sensitivity. In polyaniline (PANI) hybrids, a reversible protonation and deprotonation mechanism dominates—the insulating emeraldine base converts to conductive emeraldine salt—with the 2D material providing parallel conduction pathways that sharpen the response.

Heterostructures push performance further still. Stacking two 2D materials, such as WS2 on MoS2 or MXene on reduced graphene oxide, creates a built-in electric field at the junction; gas adsorption then modulates the width of the depletion region, magnifying the resistance change beyond what either material could deliver alone. Impedance-based readout offers an alternative to DC resistance, using AC complex impedance to decouple bulk and interfacial contributions and thereby resist humidity artifacts. These engineering levers—heterostructure formation, conductive polymer incorporation, defect engineering, and integration with metal-organic frameworks (MOFs)—form the review’s core toolkit for turning exotic nanomaterials into practical sensors.

The concrete devices are where the field’s momentum becomes vivid. For breath-based ammonia sensing, one team coated a graphene/PANI hybrid onto disposable polypropylene masks, achieving a 250 percent response at 50 ppm ammonia with a 100 parts-per-billion detection limit and—critically—a stable response across 20 to 100 percent relative humidity. Another group built an organic field-effect transistor sensor from oleylamine-functionalized graphene oxide blended with a DPP polymer and SEBS elastomer, reaching 67.01 percent per ppm sensitivity and a 9.19 ppb limit of detection while remaining stable up to 95 percent RH, because the hydrophobic oleylamine chains suppress water adsorption. A dual-signal sensor combining resistive and visual readouts tolerated 99 percent RH, and an rGO/polypyrrole composite fabric doubles as both an ammonia detector and a wearable Joule heater for responsive therapy.

MXenes and TMDs bring their own strengths. A termination-modified Ti3C2Tx MXene sensor on polyimide uses hydrocarbon surface groups to repel water while amplifying ethanol sensitivity for real-time breath alcohol monitoring, responding in 14 seconds and recovering in 37.5. A directionally freeze-cast, vertically aligned MXene/PEDOT:PSS aerogel provides a mechanically robust chemiresistive platform for wide-range breath analysis. A WS2/MoS2 heterostructure on conductive foam detected nitrogen dioxide at 0.01 ppm—51.5 percent response—while simultaneously monitoring occlusal force and maxillofacial movement for oral disease diagnosis. On the humidity side, an rGO/MoS2 composite achieved a 0.65-second response across 11 to 95 percent RH, the fastest among all sensors reviewed, and MXene/SWCNT and MXene/silk devices on laser-induced graphene delivered sub-two-second and even 0.1-second responses for respiration and touch-free sensing.

MOF composites add molecular selectivity to the mix. Although MOFs are not themselves 2D materials, pairing their porous, size-selective cages with graphene’s conductivity yields striking results: a nickel-MOF-on-graphene ethanol sensor detected 1 ppm with 1.03-second response times and survived 100,000 bending cycles, while a ZIF-8/laser-scribed graphene sensor exploited a molecular sieving effect plus machine learning to discriminate lung cancer biomarker VOCs with 96.5 percent accuracy at room temperature. Yet the review is candid about the field’s soft spots. Many VOC sensors were never tested under humid conditions or degrade badly when they are; response and recovery times are often unreported; MXenes drift more than graphene in humid air and risk oxidizing into potentially toxic byproducts; and explicit wireless transmission via Bluetooth or NFC remains experimentally rare, even as neural networks and machine learning accelerate data processing.

The authors’ roadmap for the next decade is correspondingly concrete. Selectivity, the stubborn weakness of broadly responsive 2D materials, should be attacked with selective surface receptors and electronic-nose sensor arrays coupled to pattern recognition. Energy harvesting via triboelectric or piezoelectric nanogenerators could yield self-powered wearables, while biocompatible encapsulation and dermal irritation studies must precede any real skin deployment. Scalable fabrication—inkjet printing, screen printing, roll-to-roll processing, and laser-induced graphene—will determine whether laboratory heroes reach the market, and the authors estimate current technology readiness at only TRL 3 to 4. Standardized testing protocols with defined humidity levels, analyte concentrations, flow rates, and deformation conditions, plus rigorous clinical validation in human cohorts, are urgently needed. If those pieces fall into place, the destination is clear: biodegradable substrates, self-healing electronics, AI-assisted data fusion, and multifunctional patches that turn the moisture and molecules we exhale every second into a continuous, personalized stream of medical and environmental intelligence.

Subject of Research: Wearable VOC and humidity sensors based on two-dimensional materials such as graphene, MXenes, and TMDs

Article Title: Recent advances in 2D-material-based wearable VOC and humidity sensors

Article References: Kashyap, A., Lahkar, R., & Kalita, H. (2026). Recent advances in 2D-material-based wearable VOC and humidity sensors. Frontiers in Sensors, 7, Article 1883762. https://doi.org/10.3389/fsens.2026.1883762

Image Credits: AI Generated

DOI: 10.3389/fsens.2026.1883762

Keywords: wearable sensors, 2D materials, graphene, MXene, transition metal dichalcogenides, volatile organic compounds, humidity sensing, breath analysis, flexible electronics, chemiresistive sensors, metal-organic frameworks, health monitoring