Nickel Release Rules for Beauty Devices: REACH Limits, EN 1811 Testing and What Buyers Must Ask

2026-11-10 · By the Aolemon export team

A compliance folder for an IPL handset usually contains a CE declaration, an LVD and EMC report, a RoHS report and, if there is a battery, a UN 38.3 test summary. None of those documents addresses nickel. Nickel is not one of the ten substances restricted by RoHS, so a clean RoHS report says nothing about it. What the folder typically says instead is a single line in the specification: housing in ABS, contact surface in stainless steel, with no alloy grade, no surface finish and no migration data attached. That line is where importers get caught, because the European restriction on nickel does not regulate what the metal is made of. It regulates how much nickel leaves the part per square centimetre per week, measured on the finished article after it has been worn.

The scope question is narrower and more awkward than most buyers expect. The restriction applies to parts intended to come into direct and prolonged contact with the skin, and ECHA guidance defines prolonged contact as more than ten minutes on three or more occasions within two weeks, or thirty minutes on one or more occasions within two weeks. A handset pressed to the jaw for a ten-minute session three times a week is inside that definition. So is a massage head run over a shoulder for fifteen minutes, a metal epilator head, a stainless frame around a cooling window, and the metal face of a facial roller. A test scope therefore has to be written as a list of parts chosen by how the product is used, not as one line called the product. Screws, battery contacts, USB shells and buttons under a plastic membrane are normally out; anything the user holds or presses against skin for minutes at a time is in.

The limit for skin-contact parts is 0.5 micrograms per square centimetre per week. The stricter 0.2 figure applies only to post assemblies inserted through pierced skin, so it is not the one that governs a beauty appliance, although suppliers sometimes quote it to sound thorough. The reference method is EN 1811:2011+A1:2015: the part is exposed to artificial sweat at 30 degrees Celsius for one week and the released nickel is divided by the exposed area. For a coated part the sequence is different and this is where most reports cheat. EN 12472:2020 first simulates two years of normal wear and corrosion, and only then does EN 1811 run on the aged sample. A pass on a pristine, freshly polished sample is not evidence of compliance with a rule whose whole point is that the coating must hold for at least two years of normal use. Ask for the ageing step in the method line, not just the result.

In practice the failures cluster in a few places. The classic one is a nickel underlayer. A zinc die-cast handle finished with a copper-nickel-chrome decorative stack has a bright nickel layer doing the levelling, and wherever the top chrome is thin, at edges, threads, parting lines or a spot that was over-polished, the nickel beneath is what the sweat reaches. Gold flash under a tenth of a micron over nickel on brass behaves the same way. Electroless nickel on aluminium is another repeat offender. Among stainless parts, a bead-blasted or brushed finish presents far more real surface area than a mirror polish and can push a marginal alloy over the line. Two shortcuts are worth naming because they come up in every negotiation: a magnet tells you whether a part is ferritic 430 or austenitic 304 and 316, and it tells you nothing about release; and the claim that a part contains no nickel is simply false for 304, which is roughly 8 to 10.5 percent nickel, and for 316L, which is 10 to 14 percent plus molybdenum. 316L usually releases less not because it holds less nickel but because its passive film is more stable.

The engineering answer is to stop treating this as a test and start treating it as a specification. On the drawing, name the alloy grade, the temper, the surface finish in Ra, whether the part is passivated, and if anything is plated, state that no nickel underlayer is permitted. If a zinc casting has to be plated for cost reasons, ask the plater to declare the full stack, because nickel-free stacks exist and differ in colour and price. Then close the loop on change control: purchasing substituting 304 for 316L, or a new plating lot from a different line, is a material change for this restriction even though the drawing number stays the same. Incoming inspection cannot substitute for the migration test. Plating thickness and salt spray hours are useful, but neither predicts nickel release, because release depends on porosity and passivation rather than thickness alone. Plenty of parts pass both and still fail EN 1811.

A report that survives a challenge has a recognisable shape. It comes from an ISO/IEC 17025 laboratory, which in practice often means a CNAS-accredited lab operating under the ILAC mutual recognition arrangement. It names the method and the version, shows photographs of the samples, lists each tested part and maps it to the bill of materials, states the exposed area in square centimetres, gives a result per part against the limit, and lists the models or SKUs the conclusion covers. A report on a stainless steel sheet is not a report on an article. Alongside it, keep a REACH declaration referencing the current SVHC candidate list, a SCIP notification where a candidate-list substance is present above 0.1 percent by weight, and, since the General Product Safety Regulation has applied from 13 December 2024, a technical file whose risk assessment actually names nickel as a hazard considered. For the United States there is no federal nickel release limit, but California lists soluble nickel compounds under Proposition 65 for reproductive toxicity, effective 26 October 2018, with no safe-harbour level, so the defence there is an exposure assessment or a warning.

Three sentences belong in the purchase order. First, the parts in direct and prolonged skin contact are listed in an annex, and nickel release from each must not exceed 0.5 micrograms per square centimetre per week by EN 1811, with EN 12472 ageing applied to any coated part. Second, a report covering the production configuration is due before the first shipment and again after any change of alloy, plating stack or surface finish, at the supplier's cost. Third, parts that fail are rejected and reworked or replaced, not shipped with a waiver. The commercial logic is simple: upgrading a small metal contact part from plated zinc to 316L usually costs cents per unit, while a retail withdrawal, a re-test campaign and a scrapped container cost the year's margin on that SKU. And the answer to the first email tells you most of what you need. A factory that replies with a part list, a finish specification and a test plan has been through this before. A factory that replies that its material is stainless steel 304 and therefore contains no nickel has not.

← Back to all articles