close
close

EUV light source could meet chipmakers’ future lithography needs | Research & Technology | Jul 2024

BERLIN, July 2, 2024 — Scientists from Helmholtz-Zentrum Berlin (HZB), Tsinghua University, and the German National Metrology Institute Physikalisch-Technische Bundesanstalt (PTB) are laying the foundation for a future source of coherent UV radiation, known as a steady-state microarray (SSMB).

According to the researchers, the SSMB could provide a means of generating coherent synchrotron radiation in an electron storage ring to deliver kilowatt (kW)-level average power radiation in the extreme UV (EUV) regime.

Semiconductor manufacturers are demanding shorter wavelengths to etch smaller structures. The SSMB could meet the power requirements of lithography applications that cannot be met by established accelerator technologies. It could also provide applications in various scientific and industrial fields with very high-brightness X-rays at high repetition rates.

A pulsed laser propagates with the electron beam through the MLS U125 undulator and imposes an energy modulation. The same undulator serves as a radiator during subsequent passes of the electron beam. The radiation from the undulator is detected by a fast photodiode, while the laser pulse is blocked from the detection path using an electro-optical switch. Courtesy of HZB/Communications Physics.


A pulsed laser propagates with the electron beam through the MLS U125 undulator and imposes an energy modulation. The same undulator serves as a radiator during subsequent passes of the electron beam. The undulator radiation is detected by a fast photodiode, while the laser pulse is blocked from the detection path using an electro-optical switch. Courtesy of HZB/Communications Physics.


When ultrafast electrons are deflected, they emit synchrotron radiation that can be used in storage rings where the particles are magnetically forced onto a closed path. This longitudinally incoherent light consists of a broad spectrum of wavelengths and its high degree of brightness makes it an excellent tool for materials research. Monochromators can be used to select individual wavelengths in the spectrum, but this reduces the radiant power by several orders of magnitude.

In 2010, researcher Alexander Chao, a member of the current research team, showed that if the electron bunches orbiting in a storage ring become shorter than the wavelength of the light they emit, the radiation emitted becomes coherent and more powerful.

Researcher Xiujie Deng, also a member of the current team, defined the SSMB parameters for a specific type of circular accelerator with low-alpha rings. These rings create short particle clusters that are only a μm long after interacting with a laser.

“You have to realize that the electrons in a storage ring are not distributed homogeneously,” says researcher Arnold Kruschinski. “They move in groups that are typically about a centimeter long and about 60 centimeters apart.” That’s six orders of magnitude larger than the microgroups proposed in the 2010 research, Kruschinski said.

In 2021, the researchers validated the parameters created by Deng, using what they believe to be the first storage ring designed for low-alpha operation. Through extensive experiments, the team has now fully validated Deng’s theory for microbunch generation. “For us, this is an important step on the path to a new type of SSMB radiation source,” Kruschinski said.

Videology Industrial Grade Cameras - NEW 2MP 2024 MR Camera

Project manager Jörg Feikes (left) and researcher Arnold Kruschinski (right) in the control room of BESSY II and MLS. Courtesy of Ina Helms/HZB.


Project Manager Jörg Feikes (LEFT) and researcher Arnold Kruschinski (RIGHT) in the control room of BESSY II and MLS. Courtesy of Ina Helms/HZB.


The team’s systematic studies are performed in an ongoing proof-of-principle experiment, where microbunching is generated from an energy modulation imposed by a 1064 nm laser. The results obtained so far confirm the dependence of the microbunching process on the laser modulation amplitude and the accuracy of the theoretical description of the microbunching mechanism.

The results further show that the influence of transverse-longitudinal coupling dynamics matches the team’s theoretical expectations and can be manipulated with a high degree of precision. The SSMB scheme with the most potential for achieving high-power EUV radiation is based on the use of transverse-longitudinal coupling dynamics to generate efficient micro-bunching.

Starting with an electron storage ring with a high repetition rate in the megahertz (MHz) range, SSMB could potentially use an optical laser modulator to replace the radiofrequency cavity as the main longitudinal focusing element, creating persistent microbunchlets in the circular accelerator. High average power coherent radiation could be produced in this way. With an appropriate higher harmonic generation scheme, the wavelengths of the generated radiation could reach the EUV regime, producing an EUV radiation source with a power level suitable for lithography. SSMB could also serve as a high-brightness, narrow-bandwidth UV radiation source for angle-resolved photoemission spectroscopy.

Although efforts have been made to improve the capability of free electron lasers (FELs) to generate ultrashort, high peak power radiation pulses up to the hard X-ray regime, FELs are not yet capable of delivering high average power radiation at the kW level at short wavelengths. SSMB has the potential to fill this gap.

Confirmation of key elements of the SSMB theory provides, the team says, a solid basis for continuing proof-of-principle efforts to build a prototype SSMB synchrotron radiation source. Preparations for the next phase of the SSMB experiment are underway.

Jörg Feikes, project manager at HZB, believes that it will take some time before the SSMB is introduced as an EUV radiation source. He sees parallels between the development of the SSMB and that of the FELs. “After initial experiments and decades of development work, this idea has evolved into a kilometer-long superconducting accelerator,” he said. “This kind of development takes a very long time. It starts with an idea, then a theory, then there are experimenters who gradually realize it, and I think the SSMB will evolve in the same way.”

The research was published in Physics of natural communications (www.doi.org/10.1038/s42005-024-01657-y).