A worked sequence for the sixty minutes after activity ends up somewhere it should not be: what happens in the first two minutes, why cleaning and surveying run in opposite directions, and what has to be written down before the shift ends.
A vial slips out of its shield onto the bench, or a syringe is dropped on the way to the injection room, or a patient vomits twenty minutes after a therapy dose. What happens in the next hour decides whether it becomes a cleaned surface and a signed record, or a week of finding activity on door handles.
The full procedure belongs in the department's local rules. This is the shape of it, and the parts that people get wrong under pressure.
Nothing in the first two minutes involves cleaning. It is all containment and people.
All of that depends on a spill kit that is in the hot lab rather than in an office upstairs. It needs absorbent pads, spare gloves, overshoes, waste bags with ties, long forceps, tape, a marker, a warning sign, a spare coat and a laminated copy of the procedure. Check it on a fixed date each quarter, because a kit that has been raided for gloves is not a kit.
This is the part that is genuinely counter-intuitive, and getting it the wrong way round is how contamination spreads during the clean-up rather than before it.
Clean from the outside in. Start at the outer edge of the contaminated area and work towards the centre, changing pads frequently and bagging each one as it comes off, so that the active area gets smaller with every pass rather than being smeared wider.
Survey from the outside in as well, but for a different reason: begin in a clean area to establish a background you can trust, then work inwards until the readings rise. If you start the survey standing over the spill, everything reads high, including the probe itself, and the boundary of the contaminated zone is never established. Record the background and the highest reading before you begin cleaning, so there is a before and an after.
Between passes, resurvey. When the readings approach background, take a wipe and count it, because a survey meter cannot distinguish fixed contamination from removable contamination and only the removable fraction is what spreads. If a residual spot turns out to be fixed and will not come off, the answer is to cover it, label it and let it decay rather than to keep attacking it with solvent. With technetium that is a matter of days. Everything then goes into the shielded waste bin, segregated by half-life, on the register, exactly as described in radioactive waste in a nuclear medicine department.
Two areas are forgotten with dull consistency: the underside of the bench overhang, where splashes land and nobody looks, and the floor between the spill and the sink, which is exactly the route everybody took while dealing with it.
The record is written the same day, while people remember, and it holds what happened, when, which radionuclide and roughly what activity, who was present, the dose rate and contamination readings before and after, what was done, the final survey result, whether anyone was personally contaminated and what dose was assigned to them, who was informed and when, and what has been changed so it does not recur. Whether the incident is reportable to the Bangladesh Atomic Energy Regulatory Authority is a threshold question, so establish the reporting criteria while writing your local rules rather than on the afternoon it happens.
The final item is the one that pays for itself. Run the procedure once a year as a drill, using water with a dye and no activity, with the people who actually work the bench rather than the people who wrote the document. It takes twenty minutes, it always reveals that the pads are in the wrong cupboard and the survey meter battery is flat, and it means the first time the procedure is read is not the first time it is needed.