Safety Comparison of Li-ion Battery Technology Options for Energy Storage Systems

By Vilayanur Viswanathan, Matthew Paiss

The total heat released and rate of heat generation by Li-ion batteries during abuse spans a wide range, with forced ignition of off-gases releasing up to 20 times rated energy when subjected to external heating. This article summarizes the results of short circuit, crush, overcharge and external heating for li-ion batteries with nickel based layered oxides (NLO) and lithium iron phosphate (LFP) cathodes. The need for standardized safety testing with quantifiable metrics is highlighted in the current product safety standard UL9540. The total heat generated is proportional to ampere-hour capacity, while peak heat generation rate increases exponentially with specific energy. The 2.5X higher specific energy of NLO-based cells, relative instability of layered cathodes, and higher operating voltage make them more susceptible to thermal runaway.


Abuse Tests

Results for various abuse tests, described in UL 2580 and IEC 62660-2 standards are presented in order of increasing severity in Table 1. Abuse severity depends on nail/indenting tool dimensions, shorting resistor, crushing object shape and external heat source intensity. For external short circuit and overcharge electrical tests, cell protection devices such as positive temperature coefficient components, current interrupting devices and safety vents limit surface temperature to < 110C, while results are expected to be more severe in the absence of cell protection devices. NLO-based cells are more prone to thermal runaway, with an order of magnitude higher peak heating rate and significantly higher peak temperature. Hence early off-gas detection is of paramount importance in preventing thermal runaway.

Table 1. Summary of abuse tests and outcomes for NLO-based and LFP cells


Mitigation with cell protection device?

Test variations/description

Peak temperature NLO vs. LFP

Internal short


Nail penetration, indentation. Severity depends on tool dimensions, shape, resistance

200-800°C for NLO

110°C for LFP single cell, 

370°C for 10 LFP cells in parallel

External short


Severity depends on resistor value, which ranges from 20 to 80 mW.

105°C. Thermal runaway avoided for cells with built-in safety.



Increasing order of severity for crushing surface shape: flat plate, cylindrical bar, wedge shape.




Charge rate, maximum voltage

100°C. Thermal runaway avoided for cells with built-in safety

Adiabatic Accelerated rate calorimetry


Determines self-heating onset temperature, heating rate, total heat generated

Earlier onset of thermal runaway, higher total heat generated/generation rate for NLO.

Oven heating


·       Oven temperature ramping vs. fixed temperature

·       Various ramp rates and peak temperature

775°C for NLO

350°C for LFP.

Flame/electric heater


Severity depends on heat source kW/m2

850°C  for NLO at 35 kW/m2

580°C  for LFP at 85 kW/m2

Energy released by ignition of flammable gases 10X rated energy


Energy Storage Systems product standard UL 9540 is the required listing for stationary storage products installed in the US. The driving compliance factors for the requirement of this listing are the model Fire Codes published by ICC and the NFPA, as well as the new NFPA 855 Standard on Stationary ESS.  The 2nd edition of UL 9540-2020 requires large scale fire testing to UL 9540a. This series of tests places cells, modules, racks/units, into the worst-case thermal abuse conditions to quantify a variety of failure parameters with a test report to support the final listing.  The cell-level test identifies key worst-case conditions by inducing thermal runaway and characterizing the results.  The test is conducted in an inert atmosphere within an explosion-resistant chamber. The cells are instrumented with several thermocouples and a film heater wrapped around the cell. The temperature is ramped to identify key events such as cell venting, cell temperature, thermal runaway, gas volume and rate, and flame generation.  The gas constituents are collected in a sample container and typically sent to a specialized testing laboratory for a series of evaluations to determine total gas volume and composition. Table 2 shows a sample 9540a test report.

A key change in the UL 9540a 4th edition testing is the approval of cells that cannot be induced into thermal runaway and do not produce vent gases that present a flammability hazard, to be marked for use in ESS for use inside residential dwelling units.


Table 2. Sample UL 9540a Cell-Level Test Report

Cell Design/Chemistry

Pouch, LFP

Pouch, NMC




Nominal Voltage






Thermal Runaway Methodology

External heat 4-7°C/min

External heat 4-7°C/min

Cell Surface Temp at Gas Venting



Cell Surface Temp at Thermal Runaway



Gas Volume



Gas Composition,

CO, CO2, H2, THCs

8.4%, 25.9%, 54.1%, 11.5%

24.4%, 28.9%, 31.2%, 15.5%

Lower Flammability Limit






Su (Burning Velocity)




This data informs the safe design of modules to limit cell to cell propagation by providing feedback to module designers for fire barriers and spacing, as well as in unit or rack design to prevent heat transport to exposed components/equipment. Chemistries that offer greater resistance to thermal runaway and decreased flammable gases will experience greater acceptance in installations in multi-use occupancies.  It is recommended to include stamped fire protection engineering reports verifying the fire detection, suppression, and explosion prevention systems to be based on 9540a test reports. Additional mitigation strategies include sensors to detect flammable gases and electrolyte vapors, which precede flammable gas detection by >10 minutes at nominal charge rates.


Severity of various abuse tests depend on test conditions, which vary for the same test category. Cells with internal protection devices are effective at avoiding thermal runaway during electrical abuse regardless of chemistry. For all other tests, LFP cells are less prone to thermal runaway compared to NLO cells. While total heat generated for the most severe heating tests is an order of magnitude higher than rated energy, LFP cells exhibit lower peak temperature and temperature rise rate, which translates to precious extra minutes before thermal runaway onset. The efforts to develop cells that are resistant to thermal and mechanical abuse will be the driving force in expanding available UL 9540 listed products.  Product safety standards are setting the bar for increasing levels of safety that will help guide both chemistry selection, system design, and integration for the safe operation of ESS in a variety of installation locations and usage cases.


This article edited by Mehrdad Boloorchi

For a downloadable copy of August 2020 eNewsletter which includes this article, please visit the IEEE Smart Grid Resource Center.


Vilayanur Viswanathan
is a Senior Engineer at Pacific Northwest National Laboratory. He received his BS in Chemical Engineering from the Indian Institute of Technology, Madras, and Ph.D. from Rutgers University in Chemical and Electrochemical Engineering. His research focus areas are cost performance modeling of large-scale battery systems, battery state of health modeling, grid-scale battery testing and analysis, battery safety/reliability testing and analysis, and development of energy storage test protocols/standard for grid scale energy storage. He is the Chair for the US Technical Advisory Group to IEC TC120 to develop standards for electrical energy storage systems, and is Chair of an IEEE Flow Batteries Standards Development effort IEEE 1679.3.
Matthew Paiss serves as a Technical Advisor in the Battery Materials & Systems group. Prior to joining PNNL, he was the President of Energy Response Solutions, Inc (a Training & Consultation). He brings 28 yrs of emergency response experience retiring as a Fire Captain with the San Jose CA Fire Department. Matt has 10 years’ experience on RE Codes & Standards committees and currently serves on NFPA 855 Energy Storage Systems, UL Standards Technical Panels 9540, 1974, and IEC TC120. He served as a subject matter expert for the NFPA on energy storage and has contributed to the model Fire Code sections on PV & ESS.  He has delivered electrical safety training to over 8000 firefighters nationwide and has spoken across North America and in Europe on fire and PV/ESS safety. He has written for Fire Engineering, Home Power, SolarPro and SFPE magazines.

Past Issues

To view archived articles, and issues, which deliver rich insight into the forces shaping the future of the smart grid. Older Bulletins (formerly eNewsletter) can be found here. To download full issues, visit the publications section of the IEEE Smart Grid Resource Center.

IEEE Smart Grid Bulletin Editors

IEEE Smart Grid Bulletin Compendium

The IEEE Smart Grid Bulletin Compendium "Smart Grid: The Next Decade" is the first of its kind promotional compilation featuring 32 "best of the best" insightful articles from recent issues of the IEEE Smart Grid Bulletin and will be the go-to resource for industry professionals for years to come. Click here to read "Smart Grid: The Next Decade"