Paving the Way: An Accelerator on a Microchip

Estimated reading time: 3 minutes

Electrical Engineers at the Department of Accelerator Physics of the TU Darmstadt are developing a concept of a laser-driven electron accelerator that is so small that it can be produced on a silicon chip and is cost-effective and versatile.  

Particle accelerators are usually large and expensive. That should change now. The "Accelerator on a Chip International Program" (AChIP) , sponsored by the American Gordon and Betty Moore Foundation, has set itself the goal of realizing an electron accelerator on a silicon chip. The basic idea here is to replace metal accelerator structures with glass or silicon and use a laser instead of a microwave generator as an energy source. The higher electric field load capacity of glass can increase the rate of acceleration and thereby transfer the same energy to the particles in a shorter distance, making the accelerator approximately 10 times shorter than conventional same-energy accelerators. 

A challenge here is that the vacuum channel for the electrons on a chip can only be very small, which requires an extremely strong focusing of the electron beam. The magnetic focusing channels used in conventional accelerators are far too weak for this purpose. This means that a completely new focusing concept has to be developed for an accelerator on a chip.

A Decisive Solution

As part of the TU profile area "Particle Beams and Matter" , the AChIP Group at the Department of Accelerator Physics (Department of Electrical Engineering and Information Technology of the TU Darmstadt) has given the junior scientist Dr. Ing. Uwe Niedermayerrecently presented a crucial solution. To focus the electrons in only 420 nanometer wide channel, the laser fields themselves are to be used. The concept is based on changing the relative phase of the electrons to the laser abruptly, which leads to the fact that one gets alternating focusing and defocusing in the two directions of the plane of the chip surface. This gives stability in both directions. The concept is similar to a ball on a saddle. The ball will fall down regardless of the direction of the saddle. However, if you turn the saddle continuously, the ball stays stable on the saddle. The same thing is done by the electrons in the channel on the chip. 

Perpendicular to the chip surface, only a weaker focus is necessary, and a single quadrupole magnet can be used that encloses the entire chip. This concept is similar to that of a conventional linear accelerator. However, for the on-chip accelerator, the electron dynamics have been changed to achieve a two-dimensional design that can be realized with lithographic techniques from the semiconductor industry. 

Niedermayer is currently a visiting scholar at the American Stanford University, which is the AChIP programtogether with the University of Erlangen. He works there with the AChiP colleagues on the realization of the accelerator on the chip in an experiment chamber the size of a shoebox. The laser source is a commercially available system, which is adapted by a complicated non-linear optics. The aim of the AChIP program, which runs until 2020, is to obtain electrons with an energy of one megaelectron volt from the chip. This corresponds to the electrical voltage of about one million batteries. Furthermore, ultrashort (<10 ^ -15 seconds) electron pulses are to be realized, as they are necessary for a scalable accelerator on the chip according to the concept from Darmstadt. 

Applications in Industry And Medicine

The applications of such an accelerator are in the industry and in medicine. An important long-term goal is to realize a compact, coherent X-ray source for the characterization of materials. A medical application would be, for example, an accelerator endoscope, with which one could irradiate tumors from the inside of the body with electrons. A particular advantage of this new accelerator technology is that the chips can be produced cost-effectively in large quantities, making the accelerator for everyone or the accelerator lab possible for every university. Furthermore, there are opportunities to use cost-effective coherent X-ray sources in the semiconductor industry in processes of photolithography,