Skip to main content

Diagnosing degradation and thermal risk in commercial-format batteries

Battery Cells, Diagnostics & Safety

Controlled experiments, multimodal measurements, and predictive models connect commercial-cell behavior to degradation mechanisms and thermal risk.

Discuss an Experiment Explore Our Research

Man working in a lab with green light in the background

Realistic cell formats and operating conditions

Experiments are designed around the formats and duty profiles used in demanding energy-storage applications—not idealized benchtop conditions alone.

  • 2170 and 4680 cylindrical cells, pouch cells, and large-format cells
  • Controlled temperature and humidity exposure
  • Charge and discharge rates from steady cycling to pulse and dynamic operating profiles
  • Synchronized electrical, thermal, impedance, and environmental measurements

These controlled campaigns establish how operating conditions shape performance, degradation, and safety in commercial-format batteries.

Multi-physics modeling

Low-temperature charging and lithium plating

Cold charging can produce lithium-plating behavior that is difficult to observe directly but leaves coupled electrical, impedance, and thermal signatures.

Ongoing work: Commercial cylindrical cells are tested across charging rates and controlled low-temperature conditions while measurements are compared with physics-based models.

Possible studies: Voltage relaxation, EIS, surface-temperature evolution, protocol optimization, and early indicators of plating-related degradation can be combined in project-specific campaigns.

Experimental measurements

Thermal-event forecasting and safety

Surface-temperature fields, electrical response, and sensor data are combined to identify the onset and spatial development of thermal risk.

Ongoing work: The DoD DEPSCoR project connects controlled cycling and thermography with electrochemical–thermal models and machine-learning forecasts.

Possible studies: Thermal nonuniformity, self-heating, hotspot onset, sensor placement and response, early-warning signatures, controlled abuse, and mitigation strategies can be evaluated within a defined research protocol.

Battery module with thermal sensors, management electronics, and feedback-control signals

Commercial-cell performance and aging

Multi-cycle testing resolves how capacity, efficiency, impedance, and thermal response evolve under realistic service conditions.

Ongoing work: Custom duty profiles and environmental exposure are used to connect measurable cell response with degradation across repeated cycles.

Possible studies: Constant-current cycling, pulse protocols, dynamic loads, EIS tracking, temperature and humidity effects, and EV or electric-aircraft duty profiles can be configured for cylindrical, pouch, and large-format cells.

Current federal project · Battery diagnostics & safety

Machine Learning-Assisted Forecasting of Thermal Events in Rechargeable Batteries

U.S. Department of Defense · DEPSCoR
FA9550-24-1-0164 · June 2024–May 2027

PI: Vitaliy Yurkiv · Co-PI: Todd A. Kingston, Iowa State University

Experimental, multiphysics, and machine-learning methods are used together to sense, forecast, and mitigate thermal events in rechargeable batteries.

Read the University of Arizona project story

Experimental capabilities

The laboratory integrates synchronized measurements across three complementary capability areas.

Electrochemical testing

Cycling, pulse protocols, dynamic loads, sequential EIS, and custom duty profiles for commercial-format cells.

Thermal and environmental measurements

Infrared thermography, temperature and humidity control, hotspot mapping, and coupled electro-thermal datasets.

Safety and reliability

Explosion-resistant testing, controlled abuse, vibration, mechanical loading, and safety-focused research protocols.

From experiment to prediction

Design →

Define the mechanism, conditions, observables, and safe operating envelope.

Measure →

Synchronize electrical, thermal, impedance, environmental, and sensor data.

Diagnose →

Resolve signatures, transitions, and likely internal processes.

Calibrate →

Use the dataset to constrain physics-based and data-driven models.

Predict

Forecast behavior at the next condition and test the prediction.

Every dataset is planned to answer a scientific question, test a hypothesis, and improve decisions about performance, aging, and safety.

Facilities, equipment, and shared resources

Facilities support the experimental questions above; they are the means, not the center of the research.

ESC Lab-managed experimental systems

  • ARBIN LBT21044 and LBT21084UC battery cyclers with Gamry 5000E EIS integration
  • FLIR A70 infrared camera for noncontact temperature mapping
  • AES BHD-508 temperature/humidity chamber with integrated safety modifications
  • NEWARE MFB-220-2K containment chamber for controlled battery-safety studies
  • Instron 4204 mechanical testing and Qualmark T2.0 X-LF vibration/thermal-stress capability

Computation and project-dependent access

Local GPU workstations support code development and analysis, including an NVIDIA A6000 system with 48 GB GPU memory and 256 GB RAM. Large simulations use UArizona high-performance computing resources; shared GPU resources are available for project-dependent work.

Projects may also draw on AME machine-shop and composites support, FDM/SLA prototyping, wind-tunnel and flow-diagnostics facilities, and UArizona core microscopy, materials-characterization, and nondestructive-evaluation resources. Access is scheduled and governed by each host facility.


Research-use boundary

Capabilities support academic research, method development, and certification-informed pre-screening. They are not presented as formal product certification or formal HIRF, lightning, EMC, or DO-160 qualification. Availability and safe operating envelopes are confirmed for each project.

Discuss an Experiment Explore Our Research