This repository contains research work on BLAST (Biocompatible biLayer graphene-based Artificial Synaptic Transistor) devices - a breakthrough in neuromorphic computing that mimics brain synapses using graphene and Nafion materials. The project focuses on electrical characterization and stress testing of these novel neuromorphic devices using Agilent Source Measure Units (SMUs).
- What are BLAST Devices?
BLAST devices represent a significant advancement in neuromorphic computing, bridging the gap between traditional CMOS-based computers and brain-like computational systems. These devices exhibit:
Long-term Potentiation (LTP): Similar to biological synaptic connections
Metaplasticity: Ability to modify plasticity based on synaptic activity history
Ultra-low switching energy: Significantly lower than previous neuromorphic devices
Biocompatibility: Safe for biological system integration
- Device Architecture
The BLAST device consists of:
Graphene layers: Providing excellent electrical conductivity
Nafion electrolyte: Ion-conducting polymer whose properties change with proton concentration
Source-Drain configuration: Standard transistor layout for electrical measurements
- Research Objectives
Electrical Characterization: Comprehensive testing of BLAST device electrical properties
Stress Testing: Evaluating device performance under various electrical stress conditions
Synaptic Behavior Analysis: Investigating neuromorphic properties like LTP and metaplasticity
Parameter Optimization: Finding optimal operating conditions for different applications
- Equipment Used
Agilent/Keysight Source Measure Units (SMUs): Precision electrical measurement and sourcing
Keysight B2900 Quick IV Measurement Software: Data acquisition and analysis
Custom test fixtures: For device mounting and electrical connections
- Test Protocols
- Ramp and Level Testing
Systematic voltage/current ramping
Static level measurements
I-V characterization curves
- Pulse Testing
Pulse width modulation studies
Pulse amplitude variations
Response time measurements
- Stress Testing
Extended operation under various conditions
Temperature cycling (planned)
Voltage stress applications
- Synaptic Behavior Demonstration
Successfully demonstrated synaptic-like current responses
Observed characteristic triangular wave patterns indicating proper device operation
Measured current ranges: ~10⁻⁶ to 10⁻⁴ Amperes
Pulse Width Dependency
Device response varies predictably with pulse width
Decreasing pulse width shows progressive current modulation
Consistent with biological synaptic timing dependencies
Device Reliability
Stable operation across multiple test cycles
Reproducible electrical characteristics
Low noise floor in measurements
- Data Analysis
Key Metrics Analyzed
Current-Voltage (I-V) characteristics
Pulse response dynamics
Long-term stability
Switching energy calculations
Conductance modulation range
- Analysis Tools
Keysight Quick IV
Python-based data processing scripts
MATLAB integration for advanced signal analysis
Statistical analysis of device-to-device variations
This research was conducted as part of undergraduate research at UMass Amherst BME Labs. For questions about methodologies or data, please reach out through the contact information below.