Loading today’s stories
Product1 distinct publisher3 min readPublished
Conventional X-ray cannot always resolve what sits beneath a multilayer device, so Rigaku and Tohoku have opened an institute in Sendai where the first three years go to soft X-ray methods and to the people who read the data.
The Product Desk · Product desk
Compiled by The Product DeskSomething wrong?How this is made
The measurement a process engineer wants is dull to state and hard to get: whether a feature buried under half a dozen other layers still matches the shape the design called for, and whether that can be learned from a wafer still intended for shipment. In the account published by Interesting Engineering, that is the pressure behind the launch: devices now carry more layers and a wider mix of materials, features keep shrinking, and inspection has to happen without cutting the sample apart or damaging it [3]. Conventional X-ray does not always get there, either failing to resolve the smallest structures or failing to capture changes accurately through the stack [4].
Soft X-rays are the first bet, lower in energy and longer in wavelength than the conventional kind [6]. Tohoku's side of that bet is NanoTerasu, its next-generation synchrotron radiation facility, which researchers can aim at structures that defeat conventional methods [7]. Rigaku brings the instrument and metrology side, and the two say they will combine those into new measurement methods and improved data interpretation [9].
The thing being pitched and the thing being funded are not quite the same. The pitch is seeing under the layers, but the work booked for the next three years is soft X-ray microstructure measurement for semiconductors plus information science techniques for X-ray analysis [10], and by the partners' own description, technology development is only half the mission, with training the other half [12]. Tohoku is supplying the information science because the measurements generate datasets complex enough to require serious analysis [8], and as architectures get denser, interpreting the data is as demanding as acquiring it [5]. The wall here is part signal and part knowing what the signal means.
The timeline is worth the arithmetic. Kazuhiko Omote, general manager of Rigaku's X-ray Research Laboratory, notes the field has more than 130 years behind it and says he is convinced X-rays still hold possibilities yet to be revealed [15]. The institute's initial commitment is three years, extendable on research results [2], which works out to roughly 2 percent of that span [18]. Three years is a review point rather than a ship date, and the same review governs whether the work reaches electronic and functional materials [17].
The nearest-term beneficiaries are named in the plan itself: graduate students supported by the Rigaku Scholarship System, and whoever takes the credit-bearing courses the institute wants to introduce [13][14]. Daichi Chiba, who directs Tohoku's International Center for Synchrotron Radiation Innovation Smart, describes the aim as a meeting point across generations and disciplines rather than a bolt-on of Rigaku technology to NanoTerasu [16].
If you have your own buried measurement, sort it on two axes instead of on vendor language. First, can you get the number today without destroying the part. Second, once you have the number, does anyone trust the reading without a specialist. Non-destructive and readable falls under a process problem you already own, while reachable but unreadable calls for an analysis hire or a software purchase. Unreachable but well understood once measured is an instrument line item with a date attached. Unreachable and unreadable is the quadrant this institute occupies, and the planning consequence is that your controls have to hold for at least three more years without that number in hand.
Ranked by verification strength, evidence, and original report placement.
The Rigaku-Tohoku University Co-Creation Research Institute for X-ray Metrology started on August 1 at Tohoku University in Sendai, Japan, aimed at expanding what X-ray metrology can measure inside advanced semiconductor structures.
The partners will work together for an initial three years, with an extension possible based on research results.
Devices now contain smaller features, more layers and increasingly diverse materials, and engineers need to inspect those structures without cutting them apart or damaging them.
Conventional X-ray techniques cannot always resolve the smallest structures or accurately capture changes through multilayer devices.
As chip architectures become denser, measurements must distinguish structures buried beneath multiple material layers, which makes both the X-ray measurement and the interpretation of its data increasingly demanding.
One major research direction involves soft X-rays, which have lower energy and longer wavelengths than conventional X-rays.
Distinct publishers with included, body-backed reporting in this cluster.
1 article · August 28, 2026
Follow any of these and your For You feed starts watching them — no settings page required.
science
Cerium oxide's oxygen vacancies pull graphene growth down to 300C1 distinct publisher
Evidence-backed comparisons of source perspectives and observed adoption signals. Read the methodology
Which Builder, Operator, and Investor concerns the observed source mix emphasized—not a truth score.
Evidence, demonstrated adoption, hype gap, incentives, and confidence are assessed independently, each on its own current evidence. How these are measured.
One outlet, announcement-shaped
Every fact in this story — the August 1 start, the three-year term, the division of labour, both quotations — reaches us through Interesting Engineering alone, and reads as material the two partners supplied. Nothing here is implausible; nothing here has been checked by anyone outside the partnership, and the piece carries no measurement, no figure and no document you could go and read.
A door has opened; nothing has shipped
What has demonstrably happened is an opening: a named institute, a start date, a host campus with a synchrotron already standing next to it, and a scholarship line for graduate students. No fab has run a soft X-ray recipe, no instrument has been productised, no measurement result has been published. That is close to the floor, and the reporting does not pretend otherwise.
Headline outruns the programme
Interesting Engineering's prose is properly hedged — capabilities 'could' be extended, other materials 'could eventually' follow — but the headline says semiconductor X-ray metrology exposes buried microscopic flaws in multilayer chips, in the present tense, as though the capability is in hand. Three years were commissioned precisely because it is not. The overstatement sits in the packaging, not the body.
Both speakers sell something
Rigaku sells X-ray analytical instruments, and it is Rigaku's laboratory chief who tells us the field still hides numerous possibilities. Tohoku hosts NanoTerasu, which gains from named industrial partners and paying users. The scholarship money runs one way — from the instrument vendor to the students most likely to specify instruments later. That alignment does not make the account false, but there is no voice in it with an interest in saying the idea might not work.
Dates firm, outcomes unwritten
Take the dull facts as settled — start dates, titles, a three-year term are the things institutional announcements rarely get wrong. Everything that would matter to a chipmaker or a shareholder is unsettled: whether soft X-rays actually resolve features under the stack, whether the information-science half produces usable interpretation, whether the extension is ever exercised. One account, work barely begun, so hold the direction loosely and the specifics not at all.