PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
February 6, 2026Sensors2 citationsOpen Access

Mechanical Stability of Amorphous Silicon Thin-Film Devices on Polyimide for Flexible Sensor Platforms

View Full Paper
GPGiulia PetrucciFCFabio CappelliMBM. Baldini

Key Points

  • The aim is to assess the mechanical stability of hydrogenated amorphous silicon sensors on polymer substrates under bending stress.
  • Deposited n- and p-doped amorphous silicon layers on 50 µm thick Kapton®.
  • Structured layers into two-terminal thin-film resistors to measure resistivity under bending.
  • Conducted electrical measurements with current paths oriented parallel and perpendicular to the bending axis.
  • Fabricated a linear array of p-i-n sensors on Kapton® with different thicknesses and characterized under bending.
  • n-doped layers showed limited variations, while p-doped layers were more sensitive to mechanical stress.
  • Critical-radius behavior was observed in p-type layers under transverse bending.
  • p-i-n diodes maintained rectifying behavior during bending at various radii.
  • Only minor differences were noted in reverse current related to substrate thickness.

Abstract

Hydrogenated amorphous silicon (a-Si:H) is a mature thin-film technology for large-area devices and thin-film sensors, and its low-temperature growth via Plasma-Enhanced Chemical Vapor Deposition (PECVD) makes it particularly suitable for biomedical flexible and wearable platforms. However, the reliable integration of a-Si:H sensors on polymer substrates requires a quantitative assessment of their electrical stability under mechanical stress, since bending-induced variations may affect sensor accuracy. In this work, we provide a quantitative, direction-dependent evaluation of the static-bending robustness of both single-doped a-Si:H layers and complete p-i-n junction stacks on polyimide (Kapton®), thereby linking material-level strain sensitivity to device-level functionality. First, n- and p-doped a-Si:H layers were deposited on 50 µm thick Kapton® and then structured as two-terminal thin-film resistors to enable resistivity extraction under bending conditions. Electrical measurements were performed on multiple samples, with the current path oriented either parallel (longitudinal) or perpendicular (transverse) to the bending axis, and resistance profiles were determined as a function of bending radius. While n-type layers exhibited limited and mostly gradual variations, p-type layers showed a stronger sensitivity to mechanical stress, with a critical-radius behavior under transverse bending and a more progressive evolution in the longitudinal one. This directional response identifies a practical bending condition under which doped layers, particularly p-type films, are more susceptible to strain-induced degradation. Subsequently, a linear array of a-Si:H p-i-n sensors was fabricated on Kapton® substrates with two different thicknesses (25 and 50 µm thick) and characterized under identical bending conditions. Despite the increased strain sensitivity observed in the single-layers, the p-i-n diodes preserved their rectifying behavior down to the smallest radius tested. Indeed, across the investigated radii, the reverse current at −0.5 V remained consistent, confirming stable junction operation under bending. Only minor differences, related to substrate thickness, were observed in the reverse current and in the high-injection regime. Overall, these results demonstrate the mechanical robustness of stacked a-Si:H junctions on polyimide and support their use as sensors for wearable biosensing architectures. By establishing a quantitative, orientation-aware stability benchmark under static bending, this study supports the design of reliable a-Si:H flexible sensor platforms for curved and wearable surfaces.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Petrucci et al. (2026) studied this question.

synapsesocial.com/papers/698585ea8f7c464f23009a69https://doi.org/10.3390/s26031026
Ask AI
Helpful
Bookmark
Share
View Full Paper