If you buy zirconia powder, zirconia beads, silicon carbide powder, CVD-SiC parts or triflic acid, the reference on your drawing or in your qualification file is probably a Japanese product: a Tosoh grade code, a Nikkato bead, a Shinano or Fujimi grit, a Central Glass or Mitsubishi Materials Electronic Chemicals acid. Those references were earned, and nothing here argues otherwise. The difficulty starts when a buyer needs a second source: the reference is usually written as a brand grade rather than a set of measured properties, and the Japanese position itself sits in the middle of a chain whose first steps run through China. This article maps which Japanese companies document a position in each line we cover, what their share claims do and do not say, why the position sits where it does, and what to write down before qualifying an alternative. It reflects public sources as of late September 2026.
The Map: Japanese Companies by Material Line
The table lists only companies whose product line we confirmed on the company's own pages or filings. Kyocera and Niterra are absent because we could not verify a position to a primary source, not because we found they have none.
Paraffin wax is left out: Nippon Seiro has made it in Japan since 1930 and calls itself the leading producer, but we found no basis for treating a Japanese grade as a reference.
Three entries need a qualifier first.
Where CVD-SiC parts are actually made is not settled by the brand name. Tokai Carbon's 2020 integrated report places the CVD furnace that developed and commercialized solid SiC focus rings at Tokai Carbon Korea (installed 2002, commercialized 2013), but the same report also planned new CVD-SiC furnaces in Japan for 2021 — so where a given part is made today is a question to ask, not a fact to assume; the 80% figure is the company's own 2020 claim. Ferrotec's page does not say where its CVD-SiC is made — do not assume Japan; ask.
YTZ is Nikkato's, not Tosoh's. Tosoh sells it in Europe, which is why the two names get run together. A drawing that says "YTZ or equivalent" refers to a branded bead, not a material standard.
On the SiC wafer market and Resonac's place in it, see the 6-to-8-inch wafer transition; on DKKK's role in ScSZ, see the ScSZ supplier landscape.
What the Share Claims Actually Say
Buyers are often told that a Japanese maker "has 60% of the market" or "is the global standard". We found almost no independent number behind statements of that kind.
All five are self-claims except the ROHM figure, which cites third-party Omdia data for a combination that does not yet exist. Three give no percentage at all. The Tosoh 60% figure is 25 years old; its 2018 wording is "overwhelming" share, again with no number, so do not carry it forward. The Tokai Carbon 80% figure is also dated 2020, with the Korea-manufacturing caveat noted above.
We found no credible independent share figure for any Japanese company in 8YSZ or ScSZ powder, zirconia beads, green SiC powder, high-purity alumina, triflic acid or yttrium oxide, and we do not fill that gap with market-research-site numbers. The honest summary is narrower than the shorthand. In the lines we cover, the grade most often written on a drawing is Japanese; that is what we mean by the reference position. Japanese makers also claim, in a few cases, the largest position. Whether that amounts to dominance by volume, the public record does not show.
Why the Japanese Position Sits in the Middle of the Chain
What follows is our reading of the sources above, not a finding any of them states in one place.
What Japanese makers sell is the process. Tosoh built a commercial fine-ceramic zirconia process early and has expanded it since: its 2018 release announced a roughly 30% capacity increase at Nanyo, about ¥5 billion, for operation from October 2019, and attributes the company's share to "quality superiority and quality stability." Sumitomo Chemical's alumina comes from an alkoxide route, not refined ore; Resonac sells epi-wafers, not substrate ore. None of these companies describes its advantage as raw material — what they sell is consistency at an intermediate step.
Much of the raw material and the first conversion step sit outside Japan, and often in China. Two cases are documented.
- Zirconium oxychloride. An investment report on DKKK, commissioned by the company and published in February 2026, states that DKKK's oxychloride (ZOC) has historically come only from China, because of the cost of handling the uranium- and thorium-bearing waste from cracking zircon. DKKK's Vietnam subsidiary, whose new plant reached full operation in July 2025, is meant to cut that dependence from over 90% to the 50% range — a plan, not a result we could confirm. The same report notes that DKKK's fuel-cell zirconia uses both yttria and scandia, both under Chinese export licensing. Our zirconia raw-material article shows the same step in Tosoh's process flow; we found no disclosure by Tosoh of its oxychloride or yttria sourcing.
- Yttrium oxide. The Daiichi Life Research Institute, working from Japan's Ministry of Finance trade statistics, which report yttrium oxide separately, found that China supplied around 95% of Japan's yttria imports through 2025.
In silicon carbide, the public record is thinner: we could not establish whether Shinano, Pacific Rundum or Yakushima Denko still smelt crude SiC in Japan or import it and refine it in Japan, nor find Japan's import share of SiC from China (see "What we could not confirm" below). We draw no conclusion about SiC's upstream either way.
So the Japanese position is real and it is also exposed. It sits where the value of consistency is highest — also where Chinese-sourced supply now competes — and below it are inputs the Japanese processor does not control.
The Policy Record: Dependence Has Been Tested in Both Directions
2010. Japan's rare-earth experience is often told as an embargo; the record supports something less clear-cut. China cut its export quotas for the second half of 2010 by 72% against the same 2009 period. Japanese importers met disruption that METI surveys recorded from roughly late September to late November, and shipments were reported fully restored by month's end; Beijing officially denied any halt. A NATO Strategic Communications Centre of Excellence case study concludes there is no conclusive evidence of a deliberate additional measure — "disruption" is defensible, "formal embargo" is not. Japan's response included a US$250 million Sojitz-JOGMEC investment in Lynas in March 2011; in October 2025 Sojitz announced its first import of heavy rare earths, dysprosium and terbium — yttrium unmentioned.
2019. Whatever its stated rationale, Japan's 2019 measures showed what an incumbent position looks like from the buyer's side: from July 4, 2019, Japan required individual export licences to Korea for fluorinated polyimide, resist and hydrogen fluoride, and removed Korea from its preferential "white list" that year. Japan had supplied over 90% of Korea's imports of two of the three items beforehand, per a 2023 RIETI analysis. Hydrogen fluoride exports to Korea fell 87.9%; Korean buyers shifted to Belgium, Taiwan and the US, and Korea launched a roughly ₩7.8 trillion localization programme. The measures were lifted from March 23, 2023, and Korea restored to the white list from July 21, 2023 — licensing, not a ban. The point here: the customer built second sources once the incumbent's position became a policy instrument.
2026. The direction has reversed, and this time Japanese processors are on the receiving end as importers. On January 6, 2026, MOFCOM Announcement No. 1 banned exports of all dual-use items to Japanese military end users and end uses enhancing Japan's military capability, issued after PM Takaichi's remarks on Taiwan. In February, Announcement No. 11 added 20 Japanese entities to the export-control list, and No. 12 placed 20 more on a watch list — no general licences, a required risk report, no statutory review period. None names a rare-earth element; these are end-user and end-use measures.
The trade data show what followed. On the Daiichi Life Research Institute's reading:
- China's share of Japan's yttria imports fell from around 95% to around 70% in January–April 2026. Imports from elsewhere rose, but not by enough to cover the collapse in Chinese shipments.
- Total yttria imports in January–April 2026 are running at the lowest pace since 2008.
- From March 2026, China-origin yttria has carried a higher unit price than yttria from elsewhere, which the author calls abnormal.
The yttrium control mechanics and price moves are in our yttrium export-control analysis; the scandium side reaching DKKK's fuel-cell grades is in our scandium analysis.
Japan is adding domestic capacity: in June 2026 Nikkei reported that Shin-Etsu Chemical plans a new rare-earth refinery in Fukui Prefecture. Fastmarkets put the investment at around ¥35 billion and reported that Shin-Etsu acknowledged the project but did not comment on funding or timing. We could not confirm a start date, a capacity, or which elements the refinery would process, and treat it as a project, not supply.
The common belief that buying Japanese takes China out of the chain is therefore wrong on at least two documented inputs. A Japanese grade is often Japanese processing of intermediates that came through China. In 2026 that processing is itself short of one of them.
What a Buyer Should Do
The aim is not to replace a Japanese reference. It is to make the specification independent of any one brand, so that a second line, Japanese or otherwise, can be qualified against it.
Convert the brand grade into measured properties
Brand grades are compressed specifications. Take two zirconia powders that are both "3 mol% yttria":
By our arithmetic, 3 mol% Y₂O₃ in ZrO₂ converts to about 5.36 wt%. That is a linear mass-fraction conversion using molar masses of 225.81 for Y₂O₃ and 123.22 for ZrO₂. Both windows contain that value, but they have different centres and different widths. A specification that says "TZ-3Y-E or equivalent" does not say which window applies.
The suffixes carry more than composition. Tosoh's brochure decodes them:
An "E" grade promises a sintering behaviour, not a chemistry. A "B" grade comes with a binder, which a second source would also have to match.
So before an RFQ goes out, write the reference grade as numbers:
- composition, with its basis stated (mol% or wt%);
- D50, with the measurement method;
- BET surface area;
- impurity limits, element by element;
- sintered density at a stated firing schedule;
- binder: present or not, and which system.
For SiC grit the same discipline applies to the number after the "#". The mesh-number conventions and their D50 equivalents are covered in our SiC powder buying checkpoints and are not repeated here. The mol%/wt% trap in coating-grade YSZ is in our TBC powder guide.
The cheapest way to get those numbers is to ask the incumbent. Japanese datasheets already state most of them. Asking your current Japanese supplier for its spec values, with methods, gives you a vendor-neutral specification at no cost. It also makes your existing supply easier to audit.
Test equivalence on behaviour, not on the certificate
A second source can match composition and still differ in what the brand grade actually promised: the sintering curve, aging resistance, or how the binder burns out. Run the candidate and the incumbent side by side through your own process, at your own firing schedule. Read the result as sintered density, phase and strength, not as two certificates that agree. How many lots that takes, and what lots can and cannot prove, is in our sample-evaluation protocol. For powders sold under a morphology claim, SEM/EDS lot verification catches what a certificate does not.
Map the China exposure under a Japanese purchase
Ask the questions that the brand hides:
- Where does the oxychloride come from, and where does the yttria or scandia come from? These are fair questions to put to any maker, Japanese or Chinese.
- Has the maker issued any allocation or price notice tied to the 2025–26 controls? We found no public notice from Tosoh, DKKK, Nikkato, Shin-Etsu or Sumitomo Chemical. That does not mean there have been none in private.
- Where is the part physically made? For CVD-SiC, one Japanese supplier commercialized the product at a Korean subsidiary and has since planned Japanese furnaces; another does not state the site.
A Japanese line and a China-sourced line can share an upstream input. Two such lines are less independent than they look on a supplier list, and a dual-source plan should be built on that knowledge. The general dual-sourcing playbook is in our procurement strategy guide.
For a reader at a Japanese company, the same logic applies from the other side. A Japanese processor faces the same yttria shortage its customers do. A specification written in measured properties, with a qualified second line behind it, is ordinary risk management for that processor too. It is not a judgment on the incumbent.
Frequently Asked Questions
Which Japanese companies lead in zirconia powder?
Tosoh (TZ-3Y through fully stabilized TZ-8Y and TZ-10Y) and DKKK (HSY-3F series, zirconia for SOFC electrolytes). Tosoh calls itself world No. 1, but the last percentage we found, about 60%, dates from 2001. Nippon Shokubai makes SOFC sheets, not powder; Nikkato makes the YTZ grinding bead.
Does buying Japanese material remove China from the supply chain?
Not on the evidence. DKKK's zirconium oxychloride came over 90% from China (target: the 50% range, via its Vietnam plant); Japan's yttria imports were around 95% Chinese through 2025; and Japanese processors are themselves importers hit by China's 2026 measures.
Did China embargo rare earths to Japan in 2010?
Not formally. China cut its second-half 2010 quotas by 72%; Japanese importers met disruption from roughly late September to late November, but Beijing officially denied a halt and shipments were reported restored on November 29. "Disruption" fits; "embargo" overstates it.
Are China's 2026 measures against Japan a rare-earth ban?
No element is named: the measures ban dual-use exports to Japanese military end users and end uses, add 20 Japanese entities to the control list, and place 20 more on a watch list. The fall in Japanese yttria imports shows in trade data, not in the rules' text.
Can a non-Japanese powder be qualified against a Tosoh or DKKK grade?
Only against a specification written in measured properties: composition with its basis, particle size with its method, surface area, impurities, sintered density at a stated schedule, and binder — tested through your own process alongside the incumbent. A grade code alone does not say which tolerance window or sintering behaviour you are qualifying against.
References (Public Sources)
- Tosoh Corporation: news releases of February 21, 2001 and May 16, 2018; individual-investor presentation of September 25, 2024; FY2025 results presentation of May 13, 2026; Tosoh Europe Zirconia brochure (2023).
- Daiichi Kigenso Kagaku Kogyo: English product pages. Sessa Investment Research, Quick Look on DKKK, February 27, 2026, which discloses that it was commissioned by the company.
- Nippon Shokubai, SOFC electrolyte sheet product page. Sumitomo Metal Mining, scandium press release of April 28, 2016, and scandium oxide product page.
- Shinano Electric Refining, Pacific Rundum, Fujimi Incorporated and Taiheiyo Cement: SiC product pages. Shin-Etsu Chemical: product and plant pages.
- Resonac: news release of March 1, 2023; SiC strategy page.
- Tokai Carbon, Integrated Report 2020. Ferrotec, CVD-SiC product page.
- Sumitomo Chemical, high-purity alumina page. Central Glass, triflic acid page. Mitsubishi Materials Electronic Chemicals, EF-11 product page. Nippon Seiro, company outline.
- ROHM, IR material of March 27, 2026, citing Omdia.
- Daiichi Life Research Institute (第一ライフ資産運用経済研究所), report on Japan's rare-earth imports from China after Japan-targeted export controls, June 5, 2026, based on Ministry of Finance trade statistics.
- MOFCOM Announcements No. 1, No. 11 and No. 12 of 2026, as compiled in the Daiichi Life Research Institute report; CNN, January 6, 2026.
- Nikkei report of June 11, 2026 on Shin-Etsu's planned refinery, as carried by Reuters and The Japan Times; Fastmarkets, June 25, 2026.
- Sojitz: release of October 30, 2025 and feature of March 18, 2026.
- NATO Strategic Communications Centre of Excellence, case study on economic leverage in the 2010 Senkaku crisis.
- RIETI column by Makioka and Zhang, May 8, 2023; Bloomberg, March 2023; reporting on Korea's restoration to Japan's Group A, effective July 21, 2023.
What we could not confirm: any current numeric world share for Tosoh, or any independent share figure for a Japanese firm in the lines above; where Tosoh sources its oxychloride and yttria; whether DKKK's Vietnam plan has reached its target; whether Japanese SiC makers still smelt crude SiC domestically; Japan's import share of SiC from China; where Tokai Carbon's solid CVD-SiC focus rings are made today; where Ferrotec makes its CVD-SiC; the current scandium oxide output at Sumitomo Metal Mining; the funding, timetable and element scope of Shin-Etsu's planned refinery; and any public allocation or price notice from a Japanese maker tied to the 2025–26 controls. Where each of these arises above, we have said so rather than fill the gap.
Nami Tech Solutions (NTS) supplies, through China sourcing, several of the materials whose reference grades are Japanese. On these lines our work starts by writing the reference grade as measured properties, with methods. We then put that same written specification to Chinese mills in their own language and return the quotations side by side for the buyer's own equivalence testing. We do not claim that a China-sourced grade is a substitute until the buyer's tests say so.
