Lead and cadmium in dark chocolate, and what the testing shows

1. 01

   The headline percentages: 43 percent lead, 35 percent cadmium, zero arsenic

   A multi-year analysis of 72 consumer cocoa-containing products purchased in four cohorts across 2014, 2016, 2019 and 2022 found that 43 percent exceeded the California Prop 65 maximum allowable dose level for lead, 35 percent exceeded it for cadmium, and none exceeded it for arsenic (Hands et al. 2024). The thresholds applied were 0.5 micrograms a day for lead, 4.1 micrograms a day for cadmium, and 10 micrograms a day for arsenic. Arsenic, despite appearing in headlines, was not a problem in a single product tested.

2. 02

   The second ruler: 70 of 72 cleared the FDA lead limit

   Against the separate US FDA interim reference level for lead, 97.2 percent of the products tested, 70 out of 72, fell below the limit (Hands et al. 2024). The apparent contradiction with the 43 percent figure is a difference in what each threshold is for. Prop 65 maximum allowable dose levels are conservative regulatory triggers for warning labels set well below levels of expected harm; FDA interim reference levels are derived to hold exposure within a target blood lead concentration. Reporting one without the other overstates or understates the risk.

3. 03

   The distribution: medians below threshold, outliers doing the work

   Median concentrations of each metal tested were lower than even the conservative Prop 65 thresholds, which the authors read as indicating a potentially large effect of product outliers (Hands et al. 2024). That is the single most useful structural fact here. The typical dark chocolate product is not the problem; a minority of high-contamination products pull the exceedance rate upward. It moves the practical question from whether to eat dark chocolate to which specific product, which is a solvable question with per-lot testing data.

4. 04

   Where the real exposure risk sits: dose and stacking

   The authors concluded that contamination in more than half the products tested may not pose appreciable risk for the average person at a single serving, but that consuming some of the products tested, or more than one serving a day in combination with non-cocoa-derived sources, may add up to exposure exceeding the Prop 65 thresholds (Hands et al. 2024). Two habits carry most of the risk: eating well beyond one serving daily, and stacking cocoa powder, cacao nibs, chocolate protein powders and cocoa supplements that nobody counts together.

5. 05

   The counterintuitive result: organic products tested worse

   Products labelled organic were significantly more likely to demonstrate higher levels of both cadmium and lead (Hands et al. 2024). The likely mechanism is agronomic rather than procedural: cadmium is taken up from soil and varies with regional geology, while lead contamination in cocoa arises largely post-harvest during drying and transport. Organic certification regulates synthetic inputs and addresses neither pathway. Paying an organic premium as a contamination hedge is buying the opposite of the intended outcome.

6. 06

   Why cadmium is assessed differently from lead

   The two metals arrive by different routes and behave differently in the body. Cadmium enters the cocoa bean through root uptake from soil, making it a function of growing region, while lead contamination is largely acquired after harvest during drying and handling, making it a processing and supply-chain variable. Cadmium accumulates in the kidney over years, which is why a daily habit is evaluated differently from occasional consumption, and why the single-serving framing in the exceedance statistics understates the position of a heavy daily eater.

7. 07

   What a proportionate response looks like

   Nothing in the data supports abandoning dark chocolate, which remains among the densest sources of several nutrients: roughly 500 to 940 micrograms of copper per ounce in the copper ranking on this site, about 65 mg of magnesium per ounce in the magnesium foods ranking, and a leading polyphenol contributor by weight. The proportionate response is to hold the serving near an ounce rather than a bar, avoid stacking multiple cocoa products in a day, and prefer brands publishing per-lot heavy metal results.

What people ask us next.

Q01

How much lead and cadmium is in dark chocolate?

It varies enormously by product, which is the central finding. Across 72 cocoa-containing products tested over 2014 to 2022, 43 percent exceeded the California Prop 65 maximum allowable dose level for lead of 0.5 micrograms a day and 35 percent exceeded the cadmium level of 4.1 micrograms a day, while none exceeded the arsenic level. Median concentrations of every metal were below even those conservative thresholds, indicating a large effect from product outliers (Hands et al. 2024).

Q02

Is dark chocolate safe to eat every day?

At about one serving, for most products, the tested data suggest yes. The authors concluded that contamination in more than half the products tested may not pose appreciable risk for the average person when consumed as a single serving, but that consuming some of the products tested, or more than one serving per day in combination with non-cocoa sources of heavy metals, may add up to exposure exceeding the Prop 65 thresholds (Hands et al. 2024). Daily quantity and product choice are the two levers.

Q03

Does organic dark chocolate have fewer heavy metals?

No, the opposite was observed. Products labelled organic were significantly more likely to demonstrate higher levels of both cadmium and lead (Hands et al. 2024). The explanation is agronomic: cadmium is taken up from soil and varies with regional geology, and lead contamination largely occurs post-harvest during drying and transport. Organic certification regulates synthetic inputs and addresses neither route, so it provides no protection against either metal.

Q04

Why do the reported percentages disagree so much between sources?

Because different thresholds are being applied to the same products. Against California Prop 65 maximum allowable dose levels, 43 percent of tested products exceeded the lead limit; against the US FDA interim reference level for lead, 97.2 percent of the same products, 70 of 72, fell below the limit (Hands et al. 2024). Prop 65 levels are conservative warning-label triggers set well below expected-harm levels; FDA interim reference levels target a blood lead concentration. Both figures are accurate.

Q05

Should you worry about arsenic in chocolate?

The tested data say no. Zero percent of the 72 cocoa-containing products analysed exceeded the Prop 65 maximum allowable dose level for arsenic of 10 micrograms a day (Hands et al. 2024). Arsenic frequently appears in headlines about chocolate because it is measured alongside lead and cadmium, but in this multi-year dataset it was not an exceedance in a single product. For dietary arsenic, the analysis of arsenic in rice on this site is the relevant one.

Q06

How do you reduce heavy metal exposure from cocoa without giving it up?

Three levers, in order of effect. Hold the daily amount near a single serving of about an ounce rather than a bar, since the exceedance statistics are calculated per serving. Avoid stacking cocoa products, because cocoa powder, cacao nibs, chocolate protein powders and cocoa supplements draw on the same input and nobody totals them. And prefer brands that publish per-lot heavy metal testing, since the risk is concentrated in outlier products rather than spread across the category (Hands et al. 2024).