Manufacturing Bits: Feb. 5

Multi-beam litho shakeout; black phosphorus; selective dep.


Multi-beam litho shakeout
The multi-beam e-beam market for lithography applications continues to undergo a shakeout amid technical roadblocks and other issues.

Last week, ASML announced that it had acquired the intellectual-property (IP) assets of Mapper Lithography, a Dutch supplier of multi-beam e-beam tools for lithography applications that fell into bankruptcy late last year.

As it turns out, ASML will not continue to develop Mapper’s multi-beam lithography technology, according to the company. Mapper’s R&D employees, who will join ASML, will work on various projects at ASML.

The move reduces the multi-beam e-beam market for lithography apps down to one company–U.S.-based Multibeam. In 2014, KLA exited from the market. Over the years, others stopped developing the technology as well.

Generally, the e-beam market can be divided into two main segments—photomask and lithography. E-beam technology is alive and well on the photomask front. For years, photomask makers have used single-beam e-beam systems to write patterns on a mask.

NuFlare is the leader in the single-beam mask writer market. Then, IMS, a subsidiary of Intel, sells a multi-beam e-beam tool for photomask writing. IMS’ technology works and is making masks today. NuFlare is working on multi-beam mask writer technology as well.

Besides photomasks, e-beam is used for lithography. This is called direct-write or maskless lithography. For this, the e-beam in a system directly patterns a wafer without using a photomask.

Originally developed by IBM in the 1980s, direct-write is attractive because it enables fine resolutions without the need of an expensive mask. But the throughputs for single-beam e-beam are too slow, making it expensive for volume IC production.

Today, Raith, Vistec and others sell single-beam direct-write tools. Direct-write e-beam is relegated to niche applications, such as select ASICs, compound semiconductors and photonics.

To solve the throughput problems, the industry has been developing direct-write technology that makes use of multiple beams. At one time, there were three players in the multi-beam lithography market—KLA, Mapper and Multibeam.

Then, at one time, TSMC was a big proponent of multi-beam lithography and invested in the technology. But the technology took longer to commercialize than expected for several reasons. First, multi-beam lithography is difficult, because it’s hard to control the individual beams. The beams tend to repeal each other in the system.

Second, the technology lacked funding. Third, besides TSMC, the industry wasn’t interested in direct-write due to technical issues. Most companies, including TSMC, have been investing and pursuing extreme ultraviolet (EUV) lithography for chip production.

Regardless, KLA built a prototype tool. In 2014, though, KLA discontinued the project when it couldn’t get the industry to fund it.

In 2016, Mapper Lithography introduced the FLX-1200, a direct-write, multi-beam e-beam system. Using a 5-kV acceleration voltage, a beam generator creates an electron beam about 3cm in diameter. Then, a blanker module splits the beam into beamlets.

For some time, Mapper and Leti were jointly developing the technology for a range of applications. The two companies were targeting security applications. The system consisted of a MEMS component. At one time, the component was manufactured in a plant in Moscow, Russia. Mapper and its Russian investor, Rusnano, were the venture partners in the plant.

Apparently, Mapper ran out of funds. There were some bidders for the company, but ASML eventually bought the assets. ASML has no plans to develop Mapper’s technology. ASML will continue to focus on optical lithography and EUV. The company sells e-beam tools for metrology apps, which is a different market.

“ASML is very interested in Mapper’s unique patents, know-how and other intellectual property and its highly skilled employees. Mapper’s IP portfolio and employees could potentially play an important role in the future development of electron-beam inspection systems,” according to officials from ASML. “Another is a more Netherlands-centric motivation. In addition, we strongly feel that it is important that Mapper’s IP and knowledge that has been accumulated over many years with the support of the Dutch economic and academic ecosystem, will now be leveraged for further innovation in the Netherlands.”

Nevertheless, Multibeam appears to be the lone vendor in multi-beam litho. David Lam, chairman and CEO of Multibeam, recently disclosed that the company has made progress. Its multi-column e-beam direct-write technology is targeted for two fronts: lithography for low-volume IC fabrication and security.

Black phosphorus
Using e-beam lithography and other techniques, A*STAR has begun to unlock the secrets of a 2D material called black phosphorus.

2D materials could enable a new class of field-effect transistors (FETs) with some intriguing electrical properties. In 2004, graphene was the first 2D material isolated. Other 2D materials include boron nitride and the transition-metal dichalcogenides (TMDs).

Black phosphorus is another 2D material, which has a high hole mobility and other properties. It has a layered honeycomb atomic structure, but the lattice is not planar, but rather it is wrinkled. One of the mysteries of the material is its heat transport properties. Heat transport is about twice as fast in the wrinkled direction, as “compared with across the wrinkles,” according to A*Star.

In the lab, researcher devised black phosphorus nanoribbons with either a zigzag or armchair orientation. “Probing the heat transport and stiffness of the nanoribbons was very challenging,” said Jing Wu, a scientist at the A*STAR Institute of Materials Research and Engineering. “We fabricated two orientations of nanoribbons by using electron-beam lithography on a thin film of black phosphorus. We then picked up the nanoribbons using nano-manipulators under a scanning electron microscope, and transferred them to our lab-built micro-electro-thermal system where they were tested using an atomic force microscope. These are techniques we have been developing and using for more than eight years.”

Researchers found a link between the thermal transport and a measure of stiffness. “The ratio of thermal conductivity between the zigzag and armchair nanoribbons is almost identical to the ratio of the corresponding Young’s modulus values, and corresponds to the relationship theorized by first principles calculations,” Wu said.

“The strong anisotropy of heat transport in black phosphorus has been theoretically attributed to the dispersion or relaxation of lattice vibrations known as phonons, but the exact origin was unclear,” said Wu. “Understanding this mechanism could help us better control heat flow in nanoelectronic devices, which would be very useful in chip design for better heat dissipation.”

Selective dep
For years, the industry has been working on an advanced patterning technology called area-selective deposition for chip production at 5nm and beyond.

AtomicLimits, a blog site, has some interesting data about the latest in selective deposition. Eindhoven University of Technology recently published a perspective article entitled: “From the bottom-up: towards area-selective atomic layer deposition with high selectivity.”

Click here for another interesting article.


Allen Rasafar says:

Thank you. I commented on this in another discussion. I was anticipating this in 2016 when ASML acquired HMI. This is the strategy I could see coming as ASML realize the limitations of the optical lithography, and EUV road map exhibit its own challenges. ASML is extending its dominant role in patterning echo system. This dominance in patterning eco system encompass Metrology (eBeam) and physical verification and validation at post print step . We hope there will be more alternative opportunities available beyond the eBeam Litho to the Semiconductor industry technologists.

Leave a Reply

(Note: This name will be displayed publicly)