That looks like an interesting challenge, Glenn!
For ETO I'm only aware of gas chromatography methods. You're probably going to need to have an in-depth conversation with a couple of good labs about if/how this could be adapted for your food contact surfaces - IIRC many of the methods look for ETO and 2-chloroethanol, as the latter is formed when ETO reacts with the food matrix which would (hopefully) be absent on a clean inert food contact surface.
Depending on the required LOD, and the nature of any validation you might need to conduct, there are gas detectors that will pick up ETO. It is a relatively volatile compound so would be expected to vaporize from a surface - detection of the gaseous form immediately after application might therefore be viable with a suitably sensitive detector, but whether residual detection some time after (suspected?) application is feasible is not something I've seen any data on.
Given the recent interest in it I wouldn’t be surprised if someone out there is trying to develop a rapid detection swab system or similar, but I’ve not seen anything yet.
I'm fascinated to see if anyone has come across a method specifically for graphene detection in food - it's not a question I've seen previously. There are methods to measure e.g. total organic carbon and total inorganic carbon, which would give you a total carbon figure, but I'm not sure at present that there is a “simple” chemistry-based method that differentiates graphene from other crystal structures of carbon atoms. Arguably he defining feature of graphene is primarily the (quasi) two-dimensional hexagonal lattice structure, otherwise it’s just graphite.
It's a while since I've worked with graphene, but initial "suspected" level characterisation in those days was usually via microscopy to identify likely candidates, followed by something like atomic force microscopy, transmission electron microscopy, or Raman spectroscopy. None of these methods really lend themselves to rapid testing or to analysis of a large volume of material.
Depending on the level of (suspected) contamination, the total carbon (organic plus inorganic) might be a viable method if you’ve got reliable data for the expected result – i.e. you’d be relying on the inference that carbon content higher than your known baseline correct value was potentially graphene, or in any case is an extra unknown source of carbon worthy of further investigation. It’s complex for natural food sources though, as the baseline carbon level will presumably vary within some range, so without solid data on that I’m not sure how useful this approach would be.
Please post an update if you do come up with a solution – I’m partly curious from a purely scientific perspective, but the ETO in particular is a popular topic at present.