Morphological & Electrical Characterisation of Nanosheet Networks

Abstract

Since the discovery of the stability and excellent electrical properties of the graphene monolayer, there has been a great deal of excitement about the superlative electrical, mechanical and thermal properties of 2D materials. While individual nanosheets show excellent electrical properties, this promise has not translated to networks composed of nanosheets, which regularly show properties that are considerably inferior to their constituent nanosheets. A number of reasons have been proposed to explain these inferior electrical properties, including the network morphology and the resistance of the junctions between the nanosheets. The aim of this thesis is to develop methods to characterise the morphology of nanosheet networks and the electrical properties of the junctions between the nanosheets so that the properties of nanosheet networks can be better understood and optimised. The morphology of a nanosheet network affects its performance in a range of applications, from energy storage to sensing to printed electronics. While it is known that the network morphology is important, techniques to investigate it have remained limited. Focused ion beam-scanning electron microscope nanotomography (FIB-SEM nt) is presented as a technique which can be used determine an array of network morphological parameters. Using size-selected graphene networks as a case study, the ability of this technique to characterise network morphology was investigated. A variety of properties, including network porosity, nanosheet alignment, nanosheet aggregation, and nanosheet and pore tortuosity factors were calculated. It was found that each of these parameters varies with the size of the nanosheets in the network. This technique was then extended to investigate the morphology of size-selected networks of tungsten disulphide (WS2) and silver nanosheets (AgNSs). The morphology of the networks is found to vary with the constituent nanosheets, as well as the nanosheet size. The technique was then extended to heterostack devices. The interface between 2 size-selected graphene networks and a silver nanoparticle (AgNP) network, which is printed on top of the graphene networks, was investigated. It was found that the penetration of the AgNPs into the network depends on the size of the graphene nanosheets in the network. The technique was also applied to an out-of-plane WS2 device with an indium tin oxide (ITO) bottom electrode and a gold top electrode, finding that shorting between the top and bottom electrode was caused by the high surface roughness of the WS2 network. The electrical properties of nanosheet networks are limited by the high resistance of the junctions between the nanosheets, as well as the morphology of the network. A model which relates network conductivity to the resistance of these junctions, the network morphology and the electrical properties of the constituent nanosheets was derived. This model suggests that the network conductivity depends on the size of the nanosheets in the network, with ii the nature of this dependence being controlled by the nanosheet carrier concentration. To investigate this size-selected graphene and WS2 networks were produced. The graphene network conductivity increases with decreasing nanosheet length, in line with the predictions of the model. Using the model the nanosheet conductivity was calculated as 32 ± 2 kS.m−1 and a value of 1.4 ± 0.1 kΩ was calculated for the junction resistance. The size-selected WS2 network conductivity exhibited the opposite behaviour with nanosheet length, peaking at larger nanosheet lengths and sharply decreasing as the length decreased. This behaviour is also in line with the predictions of the model. Using the model the WS2 nanosheet conductivity was calculated as 0.6 ± 0.1 S.m−1 and the junction resistance was calculated as 17 ± 3 GΩ. These results demonstrate the importance of optimising nanosheet length for nanosheet network devices. While solution-processed nanosheet networks show promise for electronic devices, producing these devices is energy and time consuming. Recently, a technique which allows for the mass-production of nanosheets by mechanical exfoliation has been demonstrated. This technique was used to produce mechanically exfoliated nanosheets, yielding large area, thick, high aspect ratio nanosheets. These nanosheets were then stacked to form a network. To understand the morphology of these networks FIB-SEM nt was used to image these networks, finding that the networks are composed of well-aligned nanosheets, with a low porosity and surface roughness. These morphological properties suggest that the network should show excellent electrical properties. The electrical properties of the networks were investigated by ionic liquid gated transistor measurements. It was found that the mobility of the networks depended on the starting material, with nanosheets exfoliated from crystal MoS2 showing mobility of 2.2 ± 0.3 cm2.V−1.s−1, while networks of nanosheets exfoliated from MoS2 powder had a mobility of 0.025 ± 0.002 cm2.V−1.s−1. These mobilities are below what has been obtained from electrochemically exfoliated MoS2 nanosheet networks. To understand the low mobilities of the networks exfoliated from crystal, the nanosheet and junction resistance were calculated. This was achieved by combining the model derived for network conductivity with electrical impedance spectroscopy (EIS). The junction resistance was calculated as 890 ± 130 kΩ and the nanosheet conductivity was calculated as 110 ± 20 S.m−1, suggesting that the low network mobility was caused by the low mobility of the MoS2 nanosheets. To further understand charge transport in these networks temperature dependent EIS was carried out, revealing that the nanosheet conductivity is limited by phonon scattering of charge carriers while the junctions transition from Miller- Abrahams hopping (MAH) at room temperature to 3D-variable range hoping (3D-VRH) at low temperatures. The MAH has an associated activation energy of 64 meV, while from the 3D-VRH regime a localisation length of 0.4 nm was calculated for the charge carriers in the MoS2 nanosheet.