This is from the SP4 paper https://pubs.acs.org/doi/10.1021/acs.analchem.1c04200 , which is really good btw . When ever I see optimization steps like this, it just screams to me that this is very hard to reproduce across labs/people. I generally think it's a good trade to trade sensitivity for reproducibility in most cases which I why I like using the King Fisher and Magnetic beads over doing it by hand