Mechanics-Consistent Adhesion in Soft Wearables
Status: Active Research (Aims 1 & 2) | Output: Presented at ACS Fall 2025
1. The Situation: The “False Positive” Trap
The Gap: Current industry standards (like ASTM D1876) assume that higher average peel force equals better adhesion.
The Problem: I identified that this assumption fails for soft wearables. In stretchable textiles, up to 90% of the “measured force” is actually just the fabric stretching (dissipation), not the glue holding. This creates a “False Positive Trap”: selecting adhesives that look strong in the lab but fail functionally on the human body due to “stick-slip” instability.
2. The Action: Rational Engineering
Instead of relying on misleading averages, I applied First Principles Thinking to re-engineer how we define failure.
- Logic over Data: I deconstructed the force profile to distinguish between True Adhesion (the peaks, representing crack resistance) and Artifacts (the troughs, representing slip).
- The Metric: I introduced the Peel Stability Index (PSI). Unlike raw force, this metric quantifies consistency. A high PSI predicts that a sensor will perform reliably without electrical noise, preventing costly device failures downstream.
3. The Execution: Automated Python Pipeline
To remove human bias and speed up analysis, I built a custom Python pipeline that operationalizes this physics-based logic.
(Noisy & Unreliable)"/]:::input subgraph "The Python Pipeline (Automated Logic)" Clean["Step 1: Signal Cleaning
(Remove Vibration Artifacts)"]:::logic Detect{"Step 2: Physics Check
Is it Adhesion or Stretching?"}:::logic Filter["Step 3: Feature Extraction
Isolate True Peak Load"]:::logic end Output[/"Final Output:
Reliability Score (PSI)"/]:::result %% Connections Raw --> Clean Clean --> Detect Detect -- "Stretching (Dissipation)" --> Clean Detect -- "True Adhesion" --> Filter Filter --> Output
