Getting The Electron Mass Right In Practice
The electron mass is a foundational constant in physics, but it's not something most people actually need to look up as a standalone value. It's 9.1093837015 × 10^-31 kilograms, or about 0.51099895000 MeV/c² in energy units. That's the CODATA 2018 recommended value, which is what you'll find in pretty much any handbook or reference table. The uncertainty is in the last two digits, so for anything beyond ultra-high-precision work you can safely use 9.11 × 10^-31 kg and not introduce any meaningful error. I've seen people trip over this in computational physics classes where they're setting up a simulation and they type in the mass wrong by a factor of ten. I did this myself early on when I was running a Monte Carlo code for electron scattering. I put in 9.11 instead of 9.11e-31 because I'd gotten confused about whether the value needed to be in kg or g. The results came back looking like the electrons were barely moving through the target material at all. Took me an hour to trace it back. Now I always sanity-check my constants against known limiting behaviors before running anything expensive. The real subtlety here isn't memorizing the number. It's understanding when the rest mass is sufficient and when you actually need the effective mass or the energy-dependent relativistic mass. In condensed matter physics, for example, electrons in a semiconductor have an effective mass that can be significantly different from the free electron rest mass. In silicon, the longitudinal effective mass is about 0.98 m_e and the transverse is about 0.19 m_e. If you're doing anything with band structure calculations and you use the bare electron mass, your results will be wrong. Not slightly wrong. Fundamentally wrong in the way that makes the whole simulation useless.
Another thing people miss: the electron mass in atomic units is just 1. So if you're working in Hartree atomic units, which half the quantum chemistry codes use internally, you don't enter the mass at all. You just let the code assume m_e = 1. I've wasted more time than I want to admit debugging unit conversion errors between SI and atomic units. The lesson is to figure out what unit system your tool expects before you start entering values. For high-energy physics applications, you'll usually see the mass expressed in MeV/c². The value 0.51099895000 MeV/c² is precise enough for collider simulations, detector reconstructions, and anything involving electromagnetic showers in calorimeters. The conversion back to kilograms is straightforward but introduces its own rounding errors if you're not careful. I typically keep the value in MeV/c² throughout the calculation and only convert to SI at the very end for reporting purposes. The downside of relying on CODATA recommended values is that they don't always match the precision you need for your specific application. If you're doing Penning trap measurements or hydrogen spectroscopy at the part-per-trillion level, the CODATA uncertainty might be too large. In those cases you need to look up the specific experimental measurement that's relevant to your work and use its reported value and uncertainty directly.
There's also the question of whether to use the classical electron radius or the Compton wavelength when setting up cross-section calculations. These aren't masses themselves but they're derived from the electron mass and show up constantly in formulas. The Compton wavelength is about 2.426 × 10^-12 m and it's extremely common in scattering problems. Mixing up which derived quantity your formula needs is another classic source of error. If you're building a reference sheet or a constants module for a codebase, I'd recommend storing the value with one extra digit of precision beyond what you think you need, and keeping the uncertainty separate. That way when someone asks why your simulation disagrees with the literature by a fraction of a percent, you can actually tell them whether it's a physics issue or just a rounding issue.
Get the Full Details
