Repository logo
 

General Device Integration Strategies for Two-Dimensional Materials


Type

Thesis

Change log

Authors

Wang, Ruizhi 

Abstract

Despite the extraordinary wealth of unique properties of 2D layered materials (2DLM), no large scale commercial application has been achieved so far. The central challenge hereby is the scalable manufacture of these materials. While mechanical exfoliation, also famously known as the scotch-tape method, leads to materials of extremely high quality, the method itself is non-scalable due to the highly stochastic deposition yield. Major advancement has been achieved in recent years regarding the scalable production of 2DLMs, especially using chemical vapour deposition (CVD). It is now common to produce large areas of graphene (Gr) or hexagonal boron nitride (h-BN), which are two members of the family of 2DLMs, in the laboratory using CVD. Often the only limit to the size of the sample is dictated by the dimensions of available equipment. Most studies have targeted the improvement of the quality of the grown material. The focus hereby has been the growth of ever-larger single-crystalline regions by lowering the nucleation density or by merging aligned domains.

Most studies fail to acknowledge the actual key challenge. Nearly all emerging applica- tions require the integration of 2DLMs into stacks of so called van der Waals heterostructures and their deposition onto insulating substrates. Since direct deposition of such structures on dielectrics has been proven to be an elusive goal, the most promising approach so far is the growth of 2DLMs on a catalyst with a subsequent transfer to the target substrate. The bottleneck of this approach has been the lack of sufficient transfer methods. A number of these have been proposed for CVD Gr and h-BN. Still the introduction of contamination and damage remain major constraints, which is exceptionally severe in case of heterostructures that rely on atomically clean interfaces. 2DLMs will only be a true candidate for commercial applications if sufficiently clean transfer methods are found that will enable large scale fabrication.

The work presented in this thesis addresses this challenge in two ways. The first is to develop new and improved transfer methods for existing combinations of 2DLM and catalyst. Thereby the aim is to base the method on a detailed understanding of their interaction and thus to devise a general rationale for transfer. The proposed method, which is referred to as Lift-Off Transfer (LOT), makes use of the weak interaction between 2DLMs and certain types of catalysts. It is shown how intercalation processes result in the local oxidation of the substrate followed by selective oxide dissolution, which releases the 2DLM film. Not only is the method highly versatile, but it also yields Gr and h-BN films of high quality compared to traditional transfer methods without requiring additional post-transfer annealing. While LOT is a significant improvement over existing transfer method, it still requires bringing the 2DLM into contact with a solution, which is a potential source of contamination.

It has been demonstrated, that CVD Gr, when processed using optimized methods, will show similar performance as mechanically exfoliated Gr. While these results are a promising first step towards more scalable processes, it still relies on mechanically exfoliated h-BN, which acts as a stamp that is used to delaminate the Gr from the growth catalyst. Thus, the focus is shifted on how to process and transfer CVD h-BN, which can then be used as the initial capping layer for the transfer of further layers of 2DLMs. To that end, an improved deposition process of h-BN has been developed that allows the growth of h-BN with individual domain exceeding 0.5 mm. More importantly, these h-BN films can be easily transferred using an entirely delamination based approach that makes use of the weak interaction between the specifically chosen catalyst and the h-BN. This enables the sequential pick up additional layers to create multilayer h-BN with atomic precision, and also direct fabrication h-BN/Gr heterostructures.

Based on a thorough understanding of the interaction between 2DLMs and their substrate, this thesis presents new strategies for device integration. Hereby not only a method is proposed that is an incremental improvement over existing ones, but an entirely new approach is presented that enables the clean and scalable device integration of 2DLMs. This work paves the path for future large scale applications of 2DLMs.

Description

Date

2018-09-19

Advisors

Hofmann, Stephan

Keywords

2D material, Graphene, h-BN, hexagonal boron nitride

Qualification

Doctor of Philosophy (PhD)

Awarding Institution

University of Cambridge
Sponsorship
EPSRC