XRF · LEACH · FLUORESCENCE · A PRIMER

XRF says lead. Does that mean lead in your food?

Most public lead reporting (XRF readings, ppm headlines, viral panic posts) measures the wrong thing for the wrong question. Here is what XRF actually measures, what leach tests measure, and where FluoroSpec fits between them.

The two questions, and why they get mixed up

Every lead test you'll ever read about is answering one of two questions:

  • Is there lead present? (an elemental question)
  • Will lead get into a person? (a bioavailability question)

These are not the same question. Most public lead reporting treats them as if they were, which is how a 12,000 ppm XRF reading on a fired piece of stoneware ends up next to a 12,000 ppm reading on a flaking 1920s windowsill, with the same headline. One of those is a real exposure risk. The other usually is not. The XRF can't tell the difference. That's not a flaw of XRF, it's just outside what XRF was built to do.

What XRF actually measures

X-ray fluorescence shoots a beam at a surface and reads the elemental signature that bounces back. It is excellent at one specific job: detecting how much lead (or arsenic, or cadmium) is present in the top few microns of whatever it's pointed at, in elemental form, regardless of what that lead is chemically bound to. It is the standard tool for housing assessments, regulatory compliance work, and pigment surveys.

What XRF cannot tell you:

  • Whether the lead is locked into a vitreous matrix (fired glaze, leaded glass), or sitting on the surface as pigment, paint, or decal.
  • Whether contact with food, saliva, or moisture will free that lead.
  • Whether a child mouthing the object will absorb anything.

That last point is the one that matters. Lead presence is a precondition for exposure. It is not the same as exposure.

What leach testing measures

Leach testing puts an object in contact with a food simulant (typically 4% acetic acid, sometimes water, sometimes saliva analogue) for hours to days, then measures how much lead came off. This is the closest thing the regulatory world has to a real exposure proxy. FDA's Compliance Policy Guide for ceramic foodware uses leach data, not XRF, to set its action levels.

Independent leach testing on fired ceramics with XRF readings up to ~3,000 ppm has consistently found no detectable lead migration into food simulant. Lead in fired glaze is chemically bound in the vitreous matrix and doesn't readily dissolve. That is the missing context behind most XRF panic.

Leach testing has its own problems: it's slow, it's expensive, it requires a lab, and it destroys the object. You can't leach-test the dish on your shelf this afternoon. Which is exactly the gap FluoroSpec was built for.

Where the mismatch matters most

XRF says yes, leach says no

Fired glaze, sealed leaded glass, fully cured ceramic

Lead is locked into the vitreous matrix. XRF reads high. Real-world lead migration into food is typically below detection limits unless the glaze is cracked, worn, or heavily acid-etched.

XRF says yes, leach says yes

Surface paint, decals, decorations on top of glaze, worn rims

Lead is in pigment form, sitting on top of the matrix. Mouthing, abrasion, or acidic food (citrus, tomato sauce, vinegar) pulls it off. This is the meaningful exposure risk most people are actually worried about.

Both items can show 50,000+ ppm on XRF. They are not the same item. Treating them the same is what produces the lead-panic news cycle. It also produces hundreds of dollars in unnecessary thrift-store dish discards every week.

Where FluoroSpec fits

FluoroSpec is methylammonium bromide dissolved in isopropanol. When that solution contacts surface-reactive lead pigment, the bromide and the lead form methylammonium lead bromide perovskite quantum dots, which fluoresce bright green under 365 nm UV light.

The chemistry is selective in a way XRF is not. The reaction needs lead in an accessible, surface-reactive form. Lead locked in a vitreous matrix doesn't react. Lead sitting on the surface as pigment does. So a green glow on a painted decoration is the same kind of signal as an acetic acid leach: it tells you the lead can be freed.

  • If FluoroSpec glows on an item: retire it from food contact. The lead is in a form that can move.
  • If it doesn't glow: the lead, if any, isn't in surface-reactive form for that condition. (Caveat: badly worn or acid-etched glazes can shift between these states.)

FluoroSpec is not a replacement for leach testing in a regulatory context. It is a 30-second proxy for the question most consumers actually care about: is this dish safe to keep using. For that question, surface reactivity is a far better answer than elemental presence.

How to read the lead database

The DetectLead lead database publishes XRF data for thousands of consumer items, sourced from EverythingLead, Lead Safe Mama, and reader submissions. Each entry shows the XRF reading and a verdict. When you read it, keep this in mind:

  • An XRF "Not Lead Safe" verdict is a flag, not a diagnosis. It tells you lead is present at a level above the verdict threshold. It does not tell you whether that lead is in a form that can reach a person.
  • For fired ceramic, leaded crystal, and sealed glass, treat XRF readings as a screening signal: worth a second look, not necessarily a discard.
  • For painted surfaces, decals, plastics, and worn pieces, treat XRF readings closer to face value: surface lead at any meaningful ppm is a real exposure pathway.
  • If you want the actionable answer for a specific item in your house, FluoroSpec on the painted areas is what we'd use ourselves.

Test your own dishes

30-second at-home test. No lab. No mailing samples. Tells you which items in your kitchen have surface-reactive lead, and which ones are fine to keep using.

Get the kit · $75
Does this mean XRF is wrong?

No. XRF is correct about what it measures: elemental presence. The mistake is reading XRF results as if they answered "will this poison me," when they answer "is lead in here." Both answers are useful. Confusing them is what produces the panic.

What about cumulative exposure from many "low" items?

Bioavailable lead from many small sources adds up the same way bioavailable lead from one big source does. That's why the FluoroSpec answer (does any of it actually move) is the one we keep coming back to. Two dozen fired ceramic plates with sealed glaze contribute roughly nothing. Two flaking painted mugs can contribute a measurable share of a child's daily intake.

Can FluoroSpec miss lead?

Yes, in two situations. First, lead deeply embedded in an intact vitreous matrix (which is also the lead least likely to ever reach a person). Second, lead in pigment forms that don't react with bromide chemistry (rare, mostly historical). FluoroSpec is a strong screen for the bioavailable-lead question, not a perfect one. Pair it with caution on flaking, worn, or acid-etched items.

What does FDA actually allow?

FDA's Compliance Policy Guide 7117.07 sets leach action levels for ceramic foodware (3.0 ppm Pb for flatware, 2.0 ppm for small hollowware, 1.0 ppm for large hollowware, 0.5 ppm for cups and mugs, in 4% acetic acid extraction). Notice none of those numbers are XRF readings. They're leach numbers. That's the regulatory standard.

Why do XRF reports include arsenic and cadmium?

XRF reads several heavy metals at once. Arsenic and cadmium have their own toxicity profiles and their own bioavailability stories, similar to lead. Cadmium in particular shows up in red, orange, and yellow ceramic glazes. The same "is it bound or surface-reactive" question applies, and FluoroSpec doesn't directly answer it for cadmium or arsenic. For those, the heuristic is: same surface conditions that release lead release everything else.

Sources: FDA CPG 7117.07 (Ceramic foodware lead action levels), 21 CFR §165 (food packaging migration), Wallace et al. ceramic leach studies, EverythingLead XRF dataset (CC-BY-SA 4.0). Last updated 2026-05-08.