A question of tritium

[dt-woods]

A question was raised on the amount of tritium in our models.

https://github.com/USEPA/ElectricityLCI/issues/255

In the ELCI.csv flow mapper, the conversion between 1 kg of tritium (hydrogen-3) to 1 kbq is given as: 359712230215827.3 (3.59x10^14).

A quick back-of-envelope calculation:

Activity (Bq) = λ × N

where λ is the decay constant and N is the number of nuclei.

1. Decay constant λ: λ (s⁻¹) = ln(2) / T½ = 0.693147 / 12.32 years ≈ 5.62 × 10⁻⁵ s⁻¹

2. Number of nuclei N in 1000 g (or 1 kg, since molar mass is approximately 3 g/mol):

Number of moles (n) = mass / molar mass = 1000 g / 3 g/mol ≈ 333.33 mol

Number of nuclei (N) = n × N_A = 333.33 mol × 6.022 x 10²³ mol⁻¹ ≈ 2.01 x 10²⁶ nuclei

3. Activity: Activity (Bq) = λ × N = 5.62 × 10⁻⁵ s⁻¹ × 2.01 x 10²⁶ nuclei ≈ 1.13 x 10¹⁴ Bq (113 × 10⁹ kBq)

This is a bit different.

[bl-young]

I am not opposed to changing the conversion factor. Do you anticipate that the next version of eLCI will need other changes to flow mapping, are all these flows still relevant? If i recall, many of the flows that required mapping were from upstream models, which may have changed at this point?

If you want to change the conversion factor you are welcome to do a PR.

Per https://github.com/USEPA/ElectricityLCI/issues/255, the source of the high tritium emissions was apparently USLCI, which I believe should already have been using fedef flows and so no mapping would have been necessary.