Arrhenius Accelerated Life Testing

How to use

1. Select Thermal Arrhenius for temperature-only acceleration or Humidity–temperature when a validated Peck humidity exponent is available.
2. Enter activation energy E_a in eV or J/mol. Enter use temperature in °C; the calculator converts every temperature to kelvin.
3. For humidity–temperature mode, enter use relative humidity and the empirical exponent n. Use values fitted to the same material, environment, and failure mechanism.
4. Enter observed life at the primary (first) stress level and the target use life in hours. Add 1–12 stress rows using the displayed temperature, humidity, and duration format.
5. Select Calculate to update acceleration, predicted life, equivalent exposure, required duration, schedule table, and log-scale sensitivity plot. Select Reset to restore the thermal example.
6. Interpret AF as use-condition life divided by stress-condition life. Review warnings whenever AF is extreme, stress is not hotter/drier as expected, or units and mechanism evidence are uncertain.
7. Do not use this screening model across phase changes, evaporation, thermal-limit crossings, oxidation or failure-mechanism shifts, or coupled humidity/voltage effects. Do not infer E_a or n from unrelated materials.

Equations used

Temperatures are converted by T [K] = temperature [°C] + 273.15. Let E_a be activation energy, T_use the use temperature, T_stress the accelerated temperature, RH the relative humidity on the same percent basis, and n the empirical humidity exponent.

C = k_B = 8.617333262145×10⁻⁵ eV/K when E_a is in eV; C = R = 8.31446261815324 J/(mol·K) when E_a is in J/mol.

Definitions, units, and limits: AF and n are dimensionless; RH is entered as percent in both numerator and denominator; all life and duration quantities use hours. Sign convention: AF exceeds 1 when the chosen stress condition accelerates degradation relative to use. The model assumes one dominant mechanism, constant conditions per schedule row, and independently established E_a and n. Multiple rows add equivalent exposure linearly (Miner-type damage assumption), which is unsuitable when sequence or recovery effects matter.