High-Purity Aluminum: Two Thresholds

High-purity aluminum is aluminum refined beyond ordinary smelter metal so that specific impurities fall below a controlled limit. Primary aluminum metal from the Hall-Héroult process runs at roughly 99.7% to 99.9% purity, with iron, silicon, zinc and gallium as the main impurities, according to a 2021 review of purification methods in Metals [Curtolo et al.].

Everything above that range depends on which threshold the seller means. That is the first thing to establish before comparing two numbers.

High-purity aluminum carries two thresholds

A Chinese producer and a specialist supplier can both write “high purity aluminum”, or “high purity aluminium”, while describing material that is a grade apart.

Xinjiang Joinworld’s 2025 annual report defines its high-purity aluminum as primary aluminum refined to an aluminum content above 99.9%, sold as ingots for electronic aluminum foil, electronic lead wires and high-purity alumina powder. That is 3N.

The two usages are a table apart:

The specialist literature and the established producers use a higher floor. The same 2021 review notes that aluminum purity has no officially standardized terminology and that countries classify it by their own criteria. Its application survey places 3N8 to 4N8 material in capacitor foils and reserves 5N and above for semiconductors and superconductors. Sumitomo Chemical lists 4N, 5N and 6N grades at 99.99%, 99.999% and 99.9999%, with individual element limits down to fractions of a part per million. Hydro quotes a range of 99.998% to 99.9999%, or 4N8 to 5N5.

Both usages are current. Neither is an error. They are different floors attached to one English phrase, which means a Chinese capacity or output figure and a 4N to 6N specialty figure cannot be added, averaged or set side by side without saying which threshold each uses.

The nines are an aggregate, not a specification

A purity percentage sums whatever the analyst chose to measure. Three applications of the metal care about three different things.

Three applications pull the same metal in different directions:

Capacitor foil accounts for the bulk of the 3N8 to 4N8 range: more than 75% of aluminum metal at that grade goes into electrolytic capacitor foils, and the review gives a purity floor of 99.95% for anode foil and 99.998% for cathode foil. A manufacturer’s technical description sets out what the foil has to do: it is etched to multiply its effective surface area, and then anodized so that an aluminium oxide layer forms as the dielectric. My reading is that the purity floor exists to keep that etching and anodizing behaviour uniform across a very thin sheet, which is why the anode and cathode floors differ rather than sharing one number.

Sputtering targets sit at the other end. An aluminum sputtering target belongs to the highest-purity demand block: the review estimates that about 96% of 5N and higher aluminum metal goes to the semiconductor industry, and it identifies a constraint that a total purity figure would hide: thorium and uranium are alpha emitters, and above a few parts per billion their radiation can cause malfunction in integrated circuits. A target can therefore pass a headline purity test and still be unusable.

Cryogenic hardware shows the same logic in a physical measurement rather than a specification. Sumitomo reports a residual resistivity ratio of 330 for 4N aluminum metal, 3,600 for 5N and 16,000 for 6N, against 9 for a standard 1070 alloy. At cryogenic temperature the 6N material reaches a thermal conductivity of about 40,000 W/(m·K), which is why it appears in conduction cooling for superconducting magnets. At that temperature the number a buyer needs is the resistivity ratio, and a purity figure does not supply it.

Aluminum refining takes more than one step

Aluminum refining beyond primary metal uses two industrial routes, and they are usually combined rather than chosen.

Three-layer electrolysis, introduced by Hoopes in 1901, floats three liquids of different density in one cell. A dense aluminum-copper alloy at the bottom acts as the anode, an electrolyte sits in the middle, and purified aluminum collects on top. The review reports roughly 13,000 kWh of direct current per tonne, current efficiency of 95% to 98%, and an output ceiling around 4N8, with the anode alloy replaced every two to three years as iron accumulates.

Segregation, also called fractional crystallization, exploits the fact that impurities dissolve differently in liquid and solid aluminum. Crystals grown slowly from a melt reject much of the impurity into the remaining liquid, which is then drained away. Repeated passes raise the grade.

Because neither route alone goes far beyond 4N8, producers chain them. Sumitomo lists both three-layer electrolysis and segregation among its production methods, and Hydro describes its route as three-layer electrolysis combined with segregation refining. The review records that three-layer electrolysis followed by segregation and a final zone-refining pass has reached 7N, with a residual resistivity ratio around 90,000.

The practical consequence for a buyer is that a supplier’s route tells you which impurities it is good at removing. It does not tell you the final grade, and the final grade does not tell you the route.

One Chinese chain shows why tonnage misleads

Joinworld’s filing describes an integrated chain running from energy and alumina through primary aluminum to high-purity aluminum, alloy products, electronic aluminum foil and electrode foil. The company states that it operates both the three-layer electrolysis and the segregation processes, and that it has achieved volume production of ultra-high-purity aluminum target blanks. A 5N5-grade target blank appears elsewhere in the filing, in its industry discussion rather than in that production claim.

The capacity and volume figures are where a reader can go wrong. The filing gives 78,000 tonnes of high-purity aluminum liquid capacity and 30,000 tonnes per year of electronic aluminum foil capacity. For 2025 it reports 92,266 tonnes of high-purity aluminum produced but only 55,346 tonnes sold externally.

That gap is internal consumption, not unsold inventory. The filing’s own footnote states that sales volumes cover external sales only and exclude internal transfers, so the usual inventory identity does not hold. An outside analyst who reads the sales line as output will understate the plant, and one who reads output as merchant supply will overstate the market.

Grade is not margin

The price ladder is steep, and the rungs are not the same kind of quote:

| Material | Price | Basis | | ------------ | ----------------- | -------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | 2N7 aluminum | about USD 2.45/kg | London Metal Exchange index, June 2021 | | 4N pellets | about USD 275/kg | commercial price list, same period | | 6N pellets | about USD 900/kg | commercial price list, same period | The review cites a June 2021 London Metal Exchange figure of about USD 2.45 per kilogram for 2N7 aluminum, against roughly USD 275 per kilogram for 4N pellets and USD 900 per kilogram for 6N pellets from a commercial price list for the same period. The LME number is an index price for primary metal; the two higher numbers are pellet prices. None is a current quotation. |

Producer margins in 2025 tell a less tidy story. Joinworld’s high-purity aluminum segment earned CNY 1.41 billion of revenue, up 39.15%, at an 11.91% gross margin. Electronic aluminum foil earned 19.18%. Electrode foil, one step further downstream, earned 1.42% while its revenue fell 18.49%, and group net profit attributable to shareholders fell 44.81%.

The company attributes the pressure to capacity additions across the aluminum electronic materials industry, lower processing fees, and higher wastewater treatment standards at its electrode foil park. I take those as the company’s account of its own year, not as an industry-wide measurement.

My inference from the two datasets, kept separate because they describe different markets and different years, is that the price ladder measures what a grade can be worth, while the margin line measures what a producer can keep. Between the two sits qualification: the customer’s acceptance of a specific material for a specific process. A 4N or 5N label does not carry that acceptance with it.

What to verify before comparing suppliers

Ask for the threshold first. A statement that a supplier makes high-purity aluminum is incomplete until it names the percentage and the elements counted.

Ask which impurity matters for the intended step. Iron and silicon dominate capacitor foil discussions, and targets are the source material for films deposited on substrates such as silicon wafers. Thorium and uranium dominate target discussions. Residual resistivity ratio dominates cryogenic applications. A supplier strong in one is not automatically strong in another.

Ask for the route, then for evidence at the point of use. Route indicates which impurities the process removes well. Qualification evidence indicates whether a customer has accepted the material in production.

Ask whether a tonnage figure is output, capacity or external sales. In an integrated plant those differ by design, as Joinworld’s own footnote shows.

Ask for the date and market of any price. Grade premiums move, and a 2021 Western pellet price is not a 2026 contract.

The useful conclusion is narrow. High-purity aluminum is a family of thresholds and process routes, and the term alone cannot rank two suppliers. The grade, the controlled impurity, the downstream step and the customer’s acceptance have to be named together before a comparison means anything.

Methodology

This guide uses a peer-reviewed review of aluminum purification, three manufacturer technical documents, and one listed company’s audited annual report. The review’s price table is dated June 2021 and describes Western market conditions; its application shares carry no date or geography in the review. The annual report covers 2025 and one Chinese producer. These are reported separately and are not combined into a single premium or margin calculation. No global market size, share or growth rate is inferred, and no current price is stated.

Sources

1. Curtolo, Xiong and Friedrich, “High- and Ultra-High-Purity Aluminum, a Review on Technical Production Methodologies”, Metals 11(9):1407, 6 September 2021. https://www.mdpi.com/2075-4701/11/9/1407
2. Sumitomo Chemical, “High purity aluminium” product brochure (manufacturer disclosure). https://www.sumitomo-chem.co.jp/english/products/files/docs/en_a07004.pdf
3. Hydro, “High purity aluminum” product page (manufacturer disclosure). https://www.hydro.com/us/global/aluminum/products/casthouse-products/high-purity-aluminum/
4. Xinjiang Joinworld Co., Ltd., 2025 Annual Report, filed 31 March 2026. https://money.finance.sina.com.cn/corp/view/vCB_AllBulletinDetail.php?stockid=600888&id=12038883
5. Nichicon, “General Description of Aluminum Electrolytic Capacitors”, technical notes hosted by Massachusetts General Hospital. https://www.nmr.mgh.harvard.edu/~reese/electrolytics/tec1.pdf