Machine-learning-enabled starch/MXene moisture sensing bioelectronics for human monitoring and interaction

The drive toward sustainable sensors demands materials and manufacturing routes that deliver high sensing performance without generating persistent waste. We report a water-based, blade-coating process to fabricate starch/Ti 3 C 2 Tₓ MXene nanocomposite films that operate as multifunctional, moisture-responsive bioelectronics. The films form a layered conductive architecture with a Young's modulus of 6.4 GPa and tensile strength of 82.3 MPa at only 1.36 vol% MXene, exceeding the mechanical performance of previously reported biodegradable conductive films. The composite films display a low percolation threshold of 0.48 vol% and offer tuneable conductivity over 9.6 × 10 −4 to 0.48 S/m. In combination with exceptional humidity sensitivity (>700% ΔR/R 0 at 95% RH), these features enable high-performance multimodal moisture sensing, including respiration analysis, speech detection, skin hydration monitoring, and non-contact human–machine interaction (HMI). When integrated with machine-learning classification, a single transient sensor can accurately distinguish breathing patterns and recognise spoken words with 97% accuracy. In addition, the films exhibit moisture-driven actuation, excellent biocompatibility (cell viability >97%), and fully biodegrade in soil within 30 days. By combining scalable processing, intelligent sensing and transient functionality, this work positions biodegradable bioelectronics as a credible alternative to conventional wearable sensor technologies. • We demonstrate, for the first time, that humidity-induced resistance signals from natural-biopolymer based, transient films can be processed using machine learning to accurately classify breathing modes and spoken words, enabling AI-assisted physiological monitoring and HMI. • The biobased nanocomposite films exhibit exceptional humidity sensitivity (>700% ΔR/R₀ at 95% RH), enabling precise detection of moisture variations for respiration monitoring, speech recognition, skin-hydration assessment, and contactless control of a smart-car. • In addition, the films combine a low percolation threshold (0.48 vol%), tuneable conductivity, and high mechanical strength (82.3 MPa), ensuring stable electrical performance and mechanical integrity during repeated operation. • The devices fully biodegrade in soil within a month, addressing the rapidly growing challenge of electronic waste from disposable sensors and short-lifetime wearables.

Authors

Institutions

Publication Details

Journal
Chemical Engineering Journal
Published
2026-03-14
DOI
https://doi.org/10.1016/j.cej.2026.175121
Primary Topic
MXene and MAX Phase Materials
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Machine-learning-enabled starch/MXene moisture sensing bioelectronics for human monitoring and interaction

Dimitrios G. Papageorgiou, Subash Rai, Emiliano Bilotti, Yansong Li et al.
Chemical Engineering Journal
MXene and MAX Phase Materials
article

Machine-learning-enabled starch/MXene moisture sensing bioelectronics for human monitoring and interaction

Dimitrios G. Papageorgiou, Subash Rai, Emiliano Bilotti, Yansong Li, Han Zhang, Ming Dong
article en

Abstract

The drive toward sustainable sensors demands materials and manufacturing routes that deliver high sensing performance without generating persistent waste. We report a water-based, blade-coating process to fabricate starch/Ti 3 C 2 Tₓ MXene nanocomposite films that operate as multifunctional, moisture-responsive bioelectronics. The films form a layered conductive architecture with a Young's modulus of 6.4 GPa and tensile strength of 82.3 MPa at only 1.36 vol% MXene, exceeding the mechanical performance of previously reported biodegradable conductive films. The composite films display a low percolation threshold of 0.48 vol% and offer tuneable conductivity over 9.6 × 10 −4 to 0.48 S/m. In combination with exceptional humidity sensitivity (>700% ΔR/R 0 at 95% RH), these features enable high-performance multimodal moisture sensing, including respiration analysis, speech detection, skin hydration monitoring, and non-contact human–machine interaction (HMI). When integrated with machine-learning classification, a single transient sensor can accurately distinguish breathing patterns and recognise spoken words with 97% accuracy. In addition, the films exhibit moisture-driven actuation, excellent biocompatibility (cell viability >97%), and fully biodegrade in soil within 30 days. By combining scalable processing, intelligent sensing and transient functionality, this work positions biodegradable bioelectronics as a credible alternative to conventional wearable sensor technologies. • We demonstrate, for the first time, that humidity-induced resistance signals from natural-biopolymer based, transient films can be processed using machine learning to accurately classify breathing modes and spoken words, enabling AI-assisted physiological monitoring and HMI. • The biobased nanocomposite films exhibit exceptional humidity sensitivity (>700% ΔR/R₀ at 95% RH), enabling precise detection of moisture variations for respiration monitoring, speech recognition, skin-hydration assessment, and contactless control of a smart-car. • In addition, the films combine a low percolation threshold (0.48 vol%), tuneable conductivity, and high mechanical strength (82.3 MPa), ensuring stable electrical performance and mechanical integrity during repeated operation. • The devices fully biodegrade in soil within a month, addressing the rapidly growing challenge of electronic waste from disposable sensors and short-lifetime wearables.

Chemical Engineering JournalVol. 534
Vaughn College of Aeronautics and Technology (US), Queen Mary University of London (GB), University of Warwick (GB)
Openalex Percentile: Top 24%
MXene and MAX Phase Materials
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.