China has almost disappeared from Japan’s imports of antimony trioxide and unwrought metal and powder. Belgium and Myanmar now lead those markets. Yet replacing Chinese material has brought less growth for those origins than their commanding shares suggest: Belgian trioxide shipments fell sharply, while Myanmar’s metal volumes barely rose.

Imports are falling as manufacturers develop substitutes

Japan is buying less of both imported products. Belgium supplied 83% of Japan’s selected trioxide imports in January–July 2026, although Belgian-origin volume was 36% lower than a year earlier. Myanmar’s share of unwrought antimony and powder rose to 56%, while its volume increased by 1%.

Japanese materials companies are also competing over how much antimony products require. A 2025 trade-press report described Japanese antimony producer Nihon Seiko as developing both reduced-antimony and antimony-free composite flame retardants in response to customer concerns about supply and cost. A company that currently sells antimony compounds could continue supplying the same customer with a formulation that uses less antimony.

The companies shaping the argument

Nihon Seiko: A Japanese antimony producer whose current catalogue also lists halogen- and antimony-free composite flame retardants. It is also developing compounds that use less antimony.

Dynabook: A Japanese computer maker that discloses antimony-free multilayer circuit boards in some notebook computers. Its history pushes the commercial-use record back to 1998.

Asahi Kasei: A Japanese materials company comparing polyamide and modified-PPE choices around procurement cost, customer chemical requirements and component performance.

Unitika: A Japanese materials producer that presented an antimony-free halogenated nylon in 2025. Its release claims a lower price than conventional antimony-using material, comparable mechanical and flame performance, and improved tracking resistance.

Belgium and Myanmar take the lead

China supplied 87% of Japan’s trioxide imports in January–July 2024. Its share fell to 15% in 2025, when Belgium accounted for 49%. In the 2026 series, Belgium supplied 83%, South Korea 15% and China less than 1%.

The metal-and-powder category has a broader group of substantial origins. China’s share fell from 41% in January–July 2024 to 3% in 2025. In 2026, Japan recorded just 79kg from China. Myanmar supplied 56% of the total, with Vietnam, Thailand and Laos each accounting for roughly 11% to 16%.

How Japan’s import origins changed

Larger shares of a shrinking import market

Belgian-origin trioxide shipments fell from 754,350kg in January–July 2025 to 485,825kg a year later. Belgium’s share rose because Japan’s total imports of that product fell faster, by 62%.

Myanmar’s metal-and-powder shipments edged up from 1,122,837kg to 1,134,722kg over the same periods, while the category total fell 23%. Its increased share largely reflected the reduction in shipments from other origins.

The change creates openings for suppliers outside China, but the volume available to compete for has also fallen. Producers considering Japan need to assess the orders that remain and the uses behind them. The earlier Chinese shipments provide a poor guide to how much replacement business is available today.

Shipments from the leading replacement origins

More origins, different routes to market

GEB is seeing buyers widen their search for alternative antimony sources in its own commercial discussions. The opportunity extends beyond the countries already prominent in Japan’s import figures, although each origin offers a different product and route into manufacturing.

Tajikistan is expanding its processing base. Its government reported commissioning an antimony-metal plant under TALCO Gold in July 2026. TALCO Gold develops the country’s Kanchoch gold-antimony deposit. The new plant gives buyers a reason to investigate metal production closer to the mine; sustained output and acceptance by Japanese customers remain to be demonstrated.

Bolivia already has a record in both mining and refining. The US Geological Survey’s 2024 production table records 4,817 tonnes of mined antimony content and 2,567 tonnes of refinery output in metal and trioxide, also measured by antimony content. The refining record makes Bolivia relevant to a search for processed material, with current grades, available volumes and export terms still to establish.

Australia’s Hillgrove mine adds a more recent development. Operator Larvotto Resources announced the start of gold and antimony production on 31 August. Its September half-year report identifies an existing antimony-concentrate offtake with Wogen Resources, the contracted counterparty. Access to this output therefore depends on the offtake and downstream processing arrangements. A new mine can expand supply while its commercial route is already spoken for.

That leaves an opening for processors as well as miners. Japanese antimony processor Tohko describes refining raw metal and converting refined metal into trioxide, with grades differentiated by particle size and use. A new origin able to supply suitable feed could support an established processor’s finished product. The resulting business would connect the new producer to existing manufacturing requirements, giving customers another sourcing option while retaining their formulation.

Nihon Seiko develops compounds with less antimony

Nihon Seiko’s current product catalogue lists its STEM series of halogen- and antimony-free magnesium-hydroxide and phosphorus-based flame retardants for polypropylene and polyethylene. Separately, a 22 July 2025 trade-press report hosted by the company described development of composite flame retardants with reduced antimony and with none. The report cited customer concerns about supply and cost and described the company’s target of at least 100 tonnes of annual sales by fiscal 2027.

For Nihon Seiko, developing a different flame retardant offers a way to keep selling to customers seeking to reduce antimony use. Its knowledge of compounding and performance remains useful even when a formulation contains less of the metal. A producer selling raw antimony has a different exposure: a successful reduction in the customer’s required loading would reduce the mineral content of that order.

The fiscal-2027 target remains an ambition, with realised sales unverified. Nihon Seiko is proposing to earn revenue from products designed to reduce customers’ reliance on the material it already sells.

Price and performance support the case for substitution

In an April 2026 technical-marketing column, Asahi Kasei positioned flame-retardant polyamide and modified PPE around antimony procurement cost, PFAS and halogen-related customer requirements, and performance differences including mouldability, water absorption, dimensional stability and electrical behaviour. The column describes the company’s proposed material choices; it gives no measure of customer conversions.

Unitika made a more direct price-and-performance pitch in October 2025. The company said its antimony-free halogenated nylon was cheaper than its conventional antimony-using counterpart while retaining comparable mechanical and flame performance and improving tracking resistance. It targeted automotive and electrical applications and set a three-year annual-sales goal of JPY1 billion. The sales figure is a target. Unitika’s pitch is that a customer could lower material costs while retaining the properties needed for the component, subject to qualification in the intended use.

A customer that starts a material review because antimony is expensive or difficult to source may keep the new formulation for electrical, processing or chemical-composition reasons. If that happens, easier antimony supply alone would give the customer little reason to reverse the change.

Automation and electronics company Omron’s earlier polyester study illustrates the technical difficulty. The paper describes the high loadings and thin-wall difficulty associated with some non-halogen routes, then reports equivalent flame performance for one alternative inorganic synergist in brominated polyester under its test conditions. The study reports flame performance under those conditions; it does not establish commercial adoption or equivalent performance across other properties.

Antimony-free circuit boards have a longer history

Dynabook’s environmental sheet for notebook model W6XZHA7CAL, announced in April 2026, lists a halogen- and antimony-free circuit board. The company’s environmental history says mass production of such multilayer boards began in 1998 and continues in some notebooks.

These boards show that antimony-free designs already have a place in commercial electronics. Their history also predates the recent supply disruption by decades. Dynabook discloses neither board volumes nor avoided antimony tonnes, so its record cannot explain the latest fall in Japanese imports.

Competition extends to the customer’s next specification

An overseas antimony producer and a Japanese compound maker can view the same customer differently. The producer has an opportunity to replace a source the customer previously relied on. The compound maker can offer to reduce the customer’s exposure to antimony prices and availability by changing the formulation.

Other origins have gained share as Chinese-origin imports have receded. Nihon Seiko’s plans show why that opening gives mineral producers limited assurance about future tonnage. A Japanese customer may keep buying flame retardants from the same company while reducing the antimony it buys through them. That would give the compound maker a new product to sell while reducing demand for the mineral in those uses.

Reliable alternative supply could still help manufacturers retain formulations they already know how to make. That gives new producers a commercial opening before a redesign is qualified: supplying the antimony required by an existing process on terms the customer can sustain.

---

Evidence and limits

Japan Customs released the detailed July data on 28 August 2026. All comparisons use January–July periods; July 2026 is provisional. The customs codes are 282580021 for antimony trioxide and 811010000 for unwrought antimony and powders. Quantities are kilograms. The Herfindahl-Hirschman Index confirms the directional concentration change: trioxide moved from 0.767 in 2024 to 0.334 in 2025 and 0.705 in 2026; metal and powder moved from 0.285 to 0.294 and 0.378. Customs origin can change after transformation and therefore differs from mine provenance.

The company sources establish disclosed products, technical positioning, test results, plans and targets within their stated limits. National antimony displacement and any causal connection to the import decline remain unquantified. The analysis of competition between origins, formulations and customer relationships is GEB analysis of those records.

The additional sourcing records describe Tajik commissioning, historical Bolivian production and company-reported Australian production and offtake. Available offers and Japanese qualification remain unverified. GEB’s observation of wider buyer enquiries comes from its commercial work, as reported by Aman Singh Anand on 20 September 2026; the observation is limited to those discussions.