Continuous rapid shrinking of feature size made the authorities to seek alternative patterning methods as the conventional photolithography process is reaching its intrinsic resolution limit. In this regard, some promising techniques have been proposed as the next generation lithography (NGL) that have the potentials to achieve both high volume production and very high resolution. Among them, several methods such as Extreme Ultraviolet Lithography (EUVL), Electron Beam lithography (EBL), Nanoimprint Lithography (NIL), Directed Self Assembly (DSA) and Scanning Probe Lithography (SPL) have demonstrated excellent potentials as promising candidates for future industrial nanofabrication.
However, all these technologies are in their development phases and still need further work to overcome some challenges in terms of flexibility, uniformity, high throughput, high resolution, high reliability, high- efficiency, defectivity, and cost of ownership. On the other hand, nanoelectrode nanolithography (NEL) has been developed in the laboratory and demonstrated as an efficient lithographic tool. It has been strengthened in recent years as one of the most promising methods due to its high reproducibility, low cost, and ability to manufacture nano-sized structures.
This method is based on the spatial confinement of the anodic oxidation between a conductive stamp and the sample surface. However, the non-uniformity issue severely limits the existing nanoelectrode lithography to be applied for large area nanopatterning. Besides, other issues such as stamp lifetime and low-cost stamp fabrication method need to be addressed to make this lithography technique viable for commercial applications.
A clear and explicit understanding of the mechanism at a molecular level helps to improve this technique. Therefore, this PhD thesis firstly aims to gain an in-depth understanding of nanoscale mechanisms involved in the anodic oxidation process and the parametric influence in nanoelectrode lithography through molecular dynamics (MD) simulations.
To do this, three-dimensional MD models of oxidation nanocell were developed, and a reactive force field (ReaxFF) was adopted to describe the interactions between atoms. The MD simulations were implemented in LAMMPS software and were performed by using a High-Performance Computing (HPC) service, ARCHIE-WeSt. The simulation results demonstrated two forms of adsorption of water molecules: molecular adsorption and dissociative adsorption. After breaking the adsorbed hydroxyls, the oxygen atoms insert into the substrate to form the Si−O−Si bonds so as to make the surface oxidized. A linear dependency of the electric field intensity on oxidation growth was observed. The relative humidity also showed the same linear behavior after a certain value (40%). The simulation results have been compared qualitatively with the experimental results, and they show in good agreement.
MD simulation results also showed that the crystallographic orientation of the substrate has a great impact on the oxidation process. It was revealed that the thickness of the oxide film and the initial oxygen diffusion rate follow an order of (100) > (110) > (111) at lower electric field intensities. It also confirmed that surfaces with higher surface energy are more reactive at lower electric field intensity. Crossovers occurred at a higher electric field intensity (7 V/nm) under which the thickness of the oxide film yields an order of T(110) > T(100) > T(111).
Atomic force microscope (AFM) oxidation experiments were performed to validate these results, which showed different orders for the (100) and (111) substrates, while (110) remained the largest for the oxide thickness. A good correlation has been found between the oxide growth and the orientation-dependent parameters where the oxide growth is proportional to the areal density of the surfaces.
The oxide growth also follows the relative order of the activation energies, which could be another controlling factor for the oxide growth. However, the differences between simulation and experimental results probably relate to the empirical potential as well as different time and spatial scales of the process.
Another objective of this thesis is to develop a new NEL process with a brass stamp that does not require conductive layer deposition. The brass material was chosen as it has high elastic modulus and high breaking strength, which ensures higher life expectancy. Therefore, this thesis reports the feasibility of using brass materials as the conductive stamps for NEL to shorten the process steps and reduce the production cost. The fabrication of nanostructures on the brass stamp was performed on a single point diamond turning (SPDT) machine. Some burrs were formed during the machining process, that prohibit the stamps from achieving a homogeneous contact with the substrates. Oxidation experiments were carried out with a home built NEL system. The results showed that an introduction of a thin layer of polymer (PS-OH) on the silicon substrate could improve the contact uniformity so as the oxidation.
Finally, a rolling nanoelectrode lithography process was proposed, for the first time, to scale up the nanoelectrode lithography technique for large-area nanofabrication. A test-bed was developed to realize uniform pressure distribution over the whole contact area so that the local oxidation process occurs uniformly over a large area of the samples. A brass roller wrapped with a fabricated polycarbonate strip has been used as a stamp to generate nanopatterns on a silicon surface. The experimental results indicated that a significant improvement in pattern uniformity compared to the other results was obtained with the conventional NEL process.
Moreover, the impact of pattern direction has been investigated, which shows no significant variation in the oxide pattern. Lastly, the rolling speed and the applied bias voltage were identified as the primary control parameters for the oxide growth.
|Date of Award||28 Jul 2020|
- University Of Strathclyde
|Sponsors||University of Strathclyde|
|Supervisor||Xichun Luo (Supervisor) & Yi Qin (Supervisor)|