How to split nanowires by Atomic Layer Etching

The master thesis by Sabbir A. Khan based on work completed at Lund Nano Lab, Lund University Sweden, has finally been released from secrecy [Patent application WO2017157902A1]. Please find below the abstract and link to the full version. Sabbir is now a Ph.D. candidate in Quantum Materials at Microsoft Quantum Materials Lab (MQML) and Center for Quantum Devices, Niels Bohr Institute, University of Copenhagen, Denmark. There his research focuses on as-grown epitaxial nanowire based networks for scalable quantum computation. The main goal is to develop innovative fabrication techniques that can realize new device concepts based on as-grown nanowire networks.

 

Evaluation of Atomic Layer Etching Possibility at Lund Nano Lab [LINK PDF]

Next you will have a chance to meet Sabbir at the PSE 2018 Tutorial, Plasma assisted atomic level processing – PEALD & ALE Sunday, September 16, 2018 in Garmisch-Partenkirchen, Germany.

Abstract

In modern electronics, device downscaling demands atomic precision control and Atomic Layer Etching (ALE) can provide this prime capability with minute device damage. ALE, also known as layer-by-layer etching, is a technique of removing atomically thin layers from the surface of materials in a controlled way. This technique is now very crucial for nanofabrication and semiconductor industry in order to achieve atomic scale resolution. This is why the overall goal of this diploma project was to investigate the possibility for ALE at Lund Nano Lab and to reveal different limitations with our current equipment. In order to achieve this goal we have done experiments with conventional system used for reactive ion etching. In addition, the ALE has been done on GaP nanowires and on Si surface patterned with high-resolution Electron Beam Lithography (EBL). The results of these experiments indicate that the process can be used to make stamps for nanoimprint lithography in a highly controlled way and that the low ion energy etching process can be used for direct nanowire splitting. We show different limitations for ALE with our current equipment and provide recommendations for new equipment dedicated for this process. In this way, the work presented here opens up the possibility for further studies of ALE with conventional equipment, shows some aspects of it’s importance for nanofabrication and suggests new applications for the ALE processes.

Popular Abstract

Nowadays electronic devices are getting smaller and much more efficient. However, it’s getting much harder to fabricate such small devices. Specifically, device fabrication with feature sizes below 10 nm (a human hair is 100,00 nm wide) is a big challenge. For this, atomic level control is needed. Atomic layer etching (ALE) is one of the key technologies that can provide atomic controlled etching of different materials by a cyclic etching, where an (sub) atomically thick layer is etched in every cycle. There are also other technologies, which have a potential for the sub 10 nm fabrication. For instance nanoimprint lithography, which is alike book printing but for extremely small features in nanometer range, and epitaxial semiconductor nanowires grown from seed catalytic particles by different epitaxial techniques have a very big potential for extremely fine fabrication of nanostructures. These techniques are also very active research areas in Lund and have already enabled many important applications. This is why the combination of ALE with these techniques may open up many new interesting opportunities. For example, the nanoimprint lithography can be only as good as the stamps, which are used for the nanoimprint, and ALE may provide a very good mean of the stamp fabrication with subatomic precision.

Sabbir A. Khan splitting Nanowires at Lund Nanolab using the Oford Instruments Plasmalab 100 System.

Different III-V semiconductor nanowires are very important for their electrical and optical properties. Diameter of the usual nanowires lies in the 40-100 nm range and with ALE it might be potentially possible to shrink the nanowire diameter to the sub 10 nm range. In this research work we used a system for reactive ion etching, similar to the systems, which are widely used in semiconductor industry for semiconductor device fabrication, for testing ALE possibility in Lund Nano Lab. We demonstrated that with this equipment it is possible to perform ALE and used this process for etching semiconductor horizontal nanowires and to make stamps for nanoimprint lithography. Surprisingly, we found that, due to some specific properties of the ALE process and a hexagonal cross section of nanowires, which we used in our experiments, after ALE each nanowire is split in to two very thin nanowires. We believe that here the inclined nanowire surfaces act as a mask for the etch process and that potentially this technique can enable fabrication of ever smaller semiconductor devices in a controllable and industrially relevant way. For instance, a core part of every transistor is a transistor channel and we can foresee that this technique may enable splitting of the transistor channel into two channels without additional expensive and challenging lithographic steps. In this way we may enable further downscaling of transistors in a very economical and practical way by this helping further downscaling of electronic devices