Has anyone integrated handheld XRF at the shaker for on-the-fly chemostrat? We trialed a Vanta for 120 s per run on 45 Wolfcamp B samples in Reeves County last week; Ca/Si and Sr tracked our carbonate benches, but Ba from barite and oil-base stain skewed things — curious what calibration routines or cleanup steps have held up in the field.
We kept the Vanta but went 3×40 s on sieved 1–2 mm chips after a hexane→IPA→DI rinse and switched to Compton‑normalized Geochem mode; we also auto-flag any pass with Ba >2 wt% as a mud-hit and toss it. On Wolfcamp B in Reeves, Sr/Ca and Ca/Si settled down once we hand‑picked obvious barite grains and fit a simple Ba-correction from barite‑spiked blanks. “Ba from barite and oil-base stain skewed things” is the pain () — are you doing on‑rig drift checks with an SRM like NIST 2711a every few hours?
The ‘Ba from barite’ ghosting drives me nuts too — at 120 s per run you’re fine on precision. What’s helped us most is a mud-only blank every tour: quick hotplate dry, shoot it, and apply that Ba offset to each cutting run so Sr and Ca/Si stay interpretable. @cgreen23’s Compton norm is solid; if you don’t like offsets, try reporting to Ti as an immobile normalizer — it’s been stable for us in Reeves.