References
- H. S. Lee, R. K. Panesar, L. Gonzalez-Macia, G. Barandun, F. Güder, “NFC-enabled sensing platform for the onsite determination of asparagine in food,” Materials Today Bio, vol. 36, 2025. doi:10.1016/j.mtbio.2025.102675
- D. R. Gawade, “Development of a low-power, battery-less near field communication sensor transponder for wireless sensing applications,” MRes Thesis, University College Cork, 2022. CORA UCC Repository
- K. Krorakai, S. Klangphukhiew, S. Kulchat, R. Patramanon, “Smartphone-Based NFC Potentiostat for Wireless Electrochemical Sensing,” Applied Sciences, vol. 11, no. 1, 2021. SOAR Repository
- STMicroelectronics, “AN2866 – How to design a 13.56 MHz customized antenna for ST25 NFC / RFID tags,” Application Note. ST AN2866
- I. Sommerville, Software Engineering, 10th ed. Pearson, 2015.
- ISO/IEC 15693, “Identification cards – Contactless integrated circuit cards – Vicinity cards.” ISO 15693
- Google, Material UI 3 Material UI 3
About the Project
The NFC Potentiostat is a compact, wireless electrochemical measurement device designed for point-of-care diagnostics. Leveraging Near-Field Communication technology, it enables smartphone-powered cyclic voltammetry measurements without batteries or wired connections. The system integrates custom-designed hardware, a dedicated Android application, and an AI-powered analysis agent (Gemma) to deliver real-time electrochemical sensing capabilities.
Problem & Solution
The Problem
Conventional potentiostat systems are bulky, expensive, and require wired connections and dedicated power sources. This limits their use in point-of-care diagnostics, field testing, and resource-limited settings where portable, low-cost electrochemical sensing is critically needed.
Our Solution
A fully wireless, battery-free potentiostat powered entirely via NFC energy harvesting from a smartphone. The system integrates a custom analog front-end, on-device AI for real-time analysis, and a companion Mobile app — enabling electrochemical measurements with nothing more than a phone tap.
Innovation & Value
First system to combine NFC energy harvesting, on-device AI inference (Gemma 3 1B), and voice-controlled electrochemistry in a single platform. Eliminates batteries, cables, and expensive lab equipment — reducing cost by over 90% compared to benchtop potentiostats while enabling truly portable biosensing.
Key Outputs
Custom 4-layer PCB with STM32U545 + ST25DV, 3D-printed enclosure with magnetic electrode holder, Mobile app (Sensitify Go NFC Edition) with real-time plotting, embedded firmware supporting 5 electroanalytical techniques, and on-device AI agent for automated data interpretation.
System Overview
Technical Specifications
Microcontroller
NFC Interface
ADC Resolution
DAC Output
Analog Front-End
Power Budget
Supported Techniques
Cyclic Voltammetry
Sweeps potential linearly between two vertices while measuring current. The most versatile technique for studying redox reactions, electrode kinetics, and reaction mechanisms.
Differential Pulse Voltammetry
Applies a series of fixed-magnitude pulses superimposed on a linear potential ramp. Measures the difference between two currents sampled before and after each pulse, offering superior sensitivity over CV for trace analyte detection.
Square Wave Voltammetry
Combines a staircase potential with a symmetric square wave. Forward and reverse currents are sampled and differenced, eliminating capacitive background current and achieving the highest sensitivity among pulse techniques.
Chronoamperometry
Steps the potential to a fixed value and records the resulting current vs. time. Ideal for studying diffusion-controlled processes, enzyme kinetics, and biosensor response curves following the Cottrell equation.
15 Work Areas · 4 Categories
Electronic Design
- Antenna design
- NFC potentiostat circuit design
- PCB soldering
- Embedded code (ADC & DAC)
Mechanical Design
- CNC antenna fabrication
- Design with Fusion360
- 3D printing – first model
- 3D printing – magnetic (5th attempt)
Software & AI
- Android app development
- Gemma fine-tuning (AI agent)
- Testing with the app
Testing & Validation
- Calibration
- Experimental tests
- Efficiency and cost analysis
- Ferricyanide measurements
NFC Protocol · ST25DV Fast Transfer
Packet Format
Binary protocol over ST25DV 256-byte mailbox. CRC-8 integrity check on every packet.
Command IDs
- 0x01 GET_STATUS — query device state
- 0x02 PING — echo test
- 0x14 RUN_CV — cyclic voltammetry
- 0x1E RUN_DPV — differential pulse
- 0x12 RUN_SWV — square wave
- 0x28 RUN_CA — chronoamperometry
- 0x3C RUN_LSV — linear sweep
- 0x20 SET_GAIN — configure mux channel
- 0x30 GET_RESULTS — chunked data read
- 0xFF ABORT — stop experiment
Response Codes
- 0x00 OK — success
- 0x01 BUSY — experiment running
- 0x02 ERROR — general failure
- 0x03 BAD_CMD — unknown command
- 0x04 BAD_PARAMS — invalid parameters
- 0x05 BAD_CRC — checksum mismatch
- 0x07 DATA_CHUNK — more data follows
- 0x08 DATA_LAST — final chunk
- 0x09 LOW_POWER — RF field weak
Application · NFC Edition
Real-time Plotting
WebView-based interactive charts render voltammograms and amperograms in real time as data arrives from the NFC potentiostat. Supports multi-series overlay, zoom, pan, and color customization.
WebView + JavaScript Bridge4 Experiment Modes
Dedicated fragments for Cyclic Voltammetry, Differential Pulse Voltammetry, Amperometry, and Potentiometry. Each mode has its own parameter configuration panel and visualization layout.
Fragment ArchitectureData Export
Save experiment data as CSV files for further analysis. Built-in file manager for organizing measurements. Export to spreadsheets with full metadata (timestamps, parameters, device info).
CSV & File ManagerNFC FTP Monitor
Live traffic inspector shows every mailbox read/write in real time. Debug NFC communication issues with full hex dump, auto-scrolling log, and pause/resume controls.
Protocol DebuggerOn-Device AI · Gemma 3 1B
Voice-Controlled Lab
Hands-free operation with SpeechRecognizer enables voice commands for starting experiments, switching modes, and querying results — critical in lab environments where gloves or sample handling prevent touchscreen interaction. Text-to-Speech reads back results and AI interpretations aloud.
Multiplatform Access
With grace of kotlin multiplatform, this app runs on IOS too
Future Work
Cloud Integration
Optional cloud sync for multi-device result sharing, collaborative analysis, and remote monitoring of long-running experiments. Maintaining offline-first architecture.
Multiplexed Sensing
Expand to multi-analyte detection with electrode array support. Simultaneous measurement of multiple biomarkers from a single sample.
Fine-Tuned AI Models
Train domain-specific Gemma models on electrochemistry literature and lab data to improve peak detection accuracy, automatic technique selection, and anomaly identification.
Clinical Validation
Partner with clinical labs for real-world biosensor validation. Target applications: glucose monitoring, heavy metal detection in water, and point-of-care infectious disease testing.
UML Diagrams
State Machine Diagram
Activity Diagram – Measurement Workflow
Sequence Diagram – CV Measurement
Calibration & Testing
10-Run Repeatability Test
A 10 kΩ dummy resistor was used to validate potentiostat accuracy. Ten consecutive cyclic voltammetry scans (0 – 0.5 mV, 0.1 V/s) show consistent I–V linearity with minimal inter-run drift. The gradual upward shift across runs reflects expected thermal settling of the analog front-end during initial power-up from NFC energy harvesting.
Design Evolution
Tech Stack
