Skip to main content

Battery Lifespan

Graph of relative capacity (ranging from .75 to 1) of battery in percent over time (ranging from 0 to 15 years) for three different climates (represented by Minneapolis, Houston and Phoenix) compared to a range of temperatures in 5-degree Celsius increments over 15 years. Trend lines from upper left to lower right reflect diminished capacity over time and shorter lifespan in Phoenix.

Impact of climate on battery calendar lifetime with no thermal management. Simulation using NREL Battery Life Model.

Graph of relative capacity (ranging from .75 to 1) of battery in percent over time (ranging from 0 to 10 years) for four different cooling types (liquid/fluid, passive, air, and air cooling) for a low-resistance cell over 15 years. Trend lines from upper left to lower right reflect diminished capacity over time and shorter lifespan for passively cooled batteries.

Impact of battery thermal management on calendar lifetime in Phoenix, Arizona hot climate. Simulation using NREL Battery Life Model.

3-D visualization with a pyramid shape (representing a macro-scale mechanical/ electrochemical model) with a callout detailing one section of the rendering.

Multi-physics model of electrochemical/mechanical/thermal degradation processes. Figure courtesy of partner institution University of Colorado-Boulder.

To compete with conventional vehicles, electric-drive vehicles (EDVs) and their batteries must perform reliably for 10 to 15 years in a variety of climates and duty cycles. NREL researchers use the battery life-predictive model, together with systems-level vehicle thermal design models, to assess lithium-ion (Li-ion) battery:

  • Chemical and mechanical degradation caused by environment and cycling
  • Performance, lifespan, and cost tradeoffs
  • Excess power, energy, and thermal management system requirements
  • Warranty, second use, and other business decision factors.

Degradation Mechanisms

Predictive models of Li-ion battery reliability must consider the multitude of electrochemical, thermal, and mechanical degradation modes experienced by batteries in application environments. Complicating matters, storage and cycling patterns can trigger varied degradation pathways for Li-ion batteries. Rates of degradation are controlled by factors such as temperatures, electrochemical operating windows, and charge/discharge rates.

Equations relating relative resistance and relative capacity to calendar fade and cycling fade. Text is presented in five sections: 1) "Calendar fade  SEI growth, loss of cyclable lithium, coupled with cycling, a1 b1 = f"; 2) "Cyclic fade  Active material structure degradation and mechanical fracture, a2 c2 = f; 3) Relative Resistance R = a1 tz + a2N; 4) "Relative Capacity Q = min(QLi, Qsites)"; and 5)"QLi = b0 + b1 tz + ..." and "Qsites = C0 = C2N + ...". The "Calendar fade" text is connected to the first part of the following equations. The "Cycling fade" text is connected to the remaining part of the following equations.

Example of reduced-order life model capturing degradation in battery resistance and capacity performance with dependence on time, number of cycles, environment, and electrochemical state.

Lifetime Prediction Modeling

The NREL lifetime model is the only predictive model of its kind, extending expensive laboratory battery-aging datasets to untested real-world scenarios. The model captures degradation effects due to both calendar time and cycle aging, including constant discharge/charge cycling, as well as more complex cycling profiles, such as those found in vehicles and grid storage applications.

The model's trial functions are statistically regressed to Li-ion cell life datasets where cells are aged under various conditions. Using this statistical framework, degradation mechanisms, along with temperature and cycling factors, are regressed to produce a lifetime prognostic model.

NREL's lifetime prediction model has been licensed to multiple companies and is integrated into other NREL analysis tools to improve the fidelity of vehicle systems, fleet evaluation and testing, ancillary load, and battery ownership modeling analyses. The model is also applied in real-time control algorithms.

Physics of Degradation

Predictive models are also used to provide feedback during the cell design process. Compared to experimentation, physical models of stress and degradation allow engineers to better understand the impacts of design concepts on lifetime and accelerate the design process. For example, predictive models can enable a designer to reduce stresses in an electrode stack to avert a shortened lifespan.

Multi-physics models of battery degradation include:

  • Non-uniform degradation due to temperature and potential imbalance in large cells
  • Solid/electrolyte interphase layer formation and growth
  • Micro-scale and macro-scale mechanical stress and degradation, coupled with electrochemistry and temperature


NREL's Publications Database offers a wide variety of documents related to the development of batteries and energy storage systems for EDVs. The following publications document project activities in modeling of battery lifetime:


Kandler Smith

Email | 303-275-4423