Skip to main content

Lead, Arsenic, and Cadmium Are Carcinogens. Mercury Is Something Else.

Lead, arsenic, cadmium, and mercury are the four heavy metals that regulators worry about most — in food, in water, in cosmetics, and in industrial chemicals. They get grouped together in headlines as "toxic heavy metals," as if they were interchangeable. In the regulatory data, they are not. Each is dangerous in its own specific way, and the way a chemical database classifies them shows the difference clearly.

We queried the K-REACH database, which holds 47,517 chemical substances registered in Korea, each carrying nine regulatory flags. Searching for the compounds of these four metals returns hundreds of entries, and the pattern of flags tells you something the "toxic heavy metal" label hides: three of the four are dominated by cancer risk, and one is not.

The four metals by the numbers

Counting the registered compounds of each metal and their flags:

Metal Compounds Toxic CMR CMR share
Lead18015911363%
Cadmium1071056561%
Arsenic22221777%
Mercury464212%

The toxic flag is almost universal — nearly every compound of all four metals carries it. The CMR flag, which marks a substance as carcinogenic, mutagenic, or reprotoxic, is where they split. For lead, cadmium, and arsenic, the majority of registered compounds are flagged CMR. For mercury, almost none are.

Three carcinogens and one that isn't

Arsenic has the highest share, with 77% of its compounds flagged CMR. That fits the wider scientific picture: arsenic and its inorganic compounds are classified as Group 1 human carcinogens by the International Agency for Research on Cancer — the highest certainty category. Cadmium is also a Group 1 carcinogen, and 61% of its compounds carry the flag here. Lead compounds sit at 63%, reflecting lead's classification as a probable carcinogen alongside its other well-documented harms.

Mercury is the outlier. Only one mercury compound in the database carries a CMR flag. This is not a gap in the data — it reflects how mercury actually harms. Mercury's primary danger is neurotoxicity, not carcinogenicity. It damages the nervous system, and in the case of methylmercury, crosses the placenta to affect fetal brain development. It is one of the most serious developmental toxins known, but it is not primarily a cancer-causing agent, and the flag pattern captures that distinction exactly.

Instead, mercury picks up a different flag. The elemental form is classified as persistent — a persistent environmental pollutant — which lead, cadmium, and arsenic in their elemental entries are not. Mercury cycles through air, water, and the food chain, accumulating in fish as methylmercury, which is why fish-consumption advisories exist. The database records mercury as a persistent, neurotoxic hazard rather than a carcinogenic one. Different metal, different danger, different flag.

Where this meets cosmetics

Heavy metals matter to cosmetics too, but they enter the picture differently. In cosmetic regulation, lead, arsenic, cadmium, and mercury are almost never intentional ingredients — they are impurities, controlled through maximum trace limits in finished products rather than through ingredient bans. That is why a search of cosmetic ingredient databases returns colorant compounds and a handful of mercury salts rather than the metals themselves.

Mercury is the exception that shows up directly. Our cosmetic data includes entries for mercury compounds such as phenylmercuric salts, with the note that mercury compounds are readily absorbed through the skin. The K-REACH data explains why those cosmetic restrictions exist: every mercury compound with GHS data in the chemical database carries a "Danger" signal word. The cosmetic rule and the chemical classification are two views of the same hazard — one governing trace limits in a face cream, the other governing the compound as an industrial substance.

Why the distinction matters

Treating the four metals as one undifferentiated hazard leads to the wrong risk questions. For arsenic and cadmium, the dominant long-term concern is cancer, which shapes how exposure limits are set and how contaminated sites are prioritized. For mercury, the governing concern is neurological and developmental harm, which is why the sharpest advice targets pregnant women and young children, and why mercury policy focuses on environmental accumulation and the food chain.

A compliance team, a toxicologist, or a public-health program that needs to know "what kind of harm" — not just "how toxic" — reads it directly from the flags. The count of CMR-flagged compounds is a fast proxy for whether a metal's regulatory profile is built around cancer risk or something else. Three of these four are; one is built around persistence and neurotoxicity instead.

What the data does not show

The counts here are of registered compounds, not of exposure or contamination. A metal with more registered compounds in the database is not necessarily more common in products or the environment — it may simply have more distinct chemical forms on the register. The compound totals also depend on our search by name, which captures entries containing the metal's name; a differently-worded entry could fall outside the count.

Flag classifications reflect the regulatory notices recorded in the database, and classifications evolve as new toxicological data emerges. The figures reflect the database at the time of writing and should be checked against the current official source before any registration, labeling, or safety decision.

Methodology and Sources

Chemical and regulatory data: K-REACH Chemical Substance database, 47,517 registered substances, nine regulatory flags per substance. We searched for compounds by metal name (lead, arsenic, cadmium, mercury) and counted the toxic and CMR flags for each group. Cosmetic context is from the K-Beauty Cosmetic Ingredients database.

External verification: the carcinogen classifications (arsenic and cadmium as IARC Group 1; lead as a probable carcinogen; mercury as a developmental neurotoxin rather than a primary carcinogen) were cross-checked against IARC monographs and published toxicology reviews. The flag counts are taken from the database as shown.

For related chemical analysis, see 47,000 Chemicals in Korea's Database, Only 97 Banned and the K-REACH API guide. For the underlying data, see K-REACH Chemical Substance on RapidAPI.

Important Notice: This article is for informational purposes only. It is not legal, regulatory, medical, or safety advice. Chemical classifications change and must be verified against the current official source before use in any registration, SDS, labeling, or compliance decision. For full terms, see our Disclaimer.

Decoded Korea publishes data-driven analysis of Korean cosmetic ingredients, chemical regulations, and safety data.

Labels:

Comments

Post a Comment

Popular posts from this blog

Your 20% Vitamin C Serum Has No Concentration Limit. In Any Country.

Vitamin C serums are sold at 10%, 15%, 20%, and sometimes 30% concentration. These are high numbers for an active ingredient. For comparison, the EU recently capped retinol at 0.3% in most cosmetics. So how is 20% vitamin C legal when 0.3% retinol is the EU limit? Because no country regulates L-Ascorbic Acid concentration in cosmetics. None of the 10 markets in our database — EU, Korea, Japan, China, Taiwan, ASEAN, Brazil, Argentina, the US, or Canada — has a limit. 76 forms of vitamin C Our database contains 76 ingredients with "Ascorbic" or "Ascorbyl" in their INCI names. Vitamin C in skincare is not a single ingredient — it is a family of compounds, each with different stability, potency, and skin penetration characteristics. The ones consumers encounter most often: Form INCI name Characteristics L-Ascorbic Acid Ascorbic Acid Most potent, least stable. Requires low pH. The form used in 10–20% serums MAP Magnesium Ascorbyl Phosphate Stable,...
Post a Comment
Read more

Retinol Had No Limits in Most Countries Until the EU Set Them in 2024

Retinol is probably the most recommended ingredient in skincare. Dermatologists recommend it. Beauty editors rank it above everything else. K-Beauty brands put it in serums, creams, and sleeping masks. It has decades of published research behind it. For most of that history, most countries did not regulate how much retinol a cosmetic product could contain. Canada had a limit (1.0% Retinol Equivalent), but Korea, Japan, China, the US, and the EU had none. That changed in April 2024, when the EU adopted Regulation 2024/996 and set concentration limits well below Canada's. We checked our database to see how this compares across 10 countries. What retinol is Retinol is a form of vitamin A. In skincare, it promotes cell turnover — old skin cells shed faster, new ones replace them. This is why retinol is used for fine lines, uneven skin tone, acne, and texture. It is one of the few cosmetic ingredients with clinical evidence supporting its anti-aging claims. Retinol is not a single...
Post a Comment
Read more

Banned in Europe, Legal in Korea: 5 Ingredients That Split Global Regulators

The same moisturizer that sits on a shelf in Seoul could be pulled from a store in Paris. Not because of a labeling error or a packaging defect — but because an ingredient inside is classified differently by two governments looking at the same scientific data. This happens more often than most consumers realize. The EU and South Korea — both major cosmetics markets with sophisticated regulatory systems — frequently disagree on where to draw the line. The EU's cosmetics regulation (EC 1223/2009) maintains one of the strictest banned substance lists in the world, with over 1,600 entries in Annex II. Korea's MFDS takes a different approach, often allowing the same ingredients under specific concentration limits or without restriction. Neither system is wrong. They operate under different regulatory philosophies: the EU leans toward the precautionary principle — restrict first, revisit later. Korea tends toward risk management — allow under controlled conditions, monitor outcomes...
Post a Comment
Read more