Aluminum deposition is an industrially important process that is often not optimized in an electrochemical sense. One method of growing thin aluminum films is electrodeposition; this process has been gaining traction in ionic liquids over the last two decades, although details of the process are still under debate. These inconsistencies are largely attributed to widely variable results based on electrolyte purity and a strong dependence on a multitude of parameters including composition, temperature and even experimental set up. This work is targeted to elucidate some of these properties with electrochemical and rheological measurements, alongside NMR and microscopy techniques.

                Some of the most promising aluminum deposition is done in chloroaluminate ionic liquids. These solutions have high aluminum concentration and low vapor pressure but typically higher viscosity then molecular solvents or aqueous chemistries. 1-ethyl-3-methyl imidazolium chloride (EMIC) and AlCl₃ in acidic ratios will readily deposit aluminum. This electrolyte deposits aluminum by the reduction of Al₂Cl₇^- to aluminum and produces AlCl₄^- anions in a relatively complex 3 electron transfer mechanism. The composition of the electrolyte is determined by the ratio of AlCl₃ to EMIC and is confirmed by NMR. The varying compositions of the ionic liquid influence both the physical and electrochemical properties. The Walden rule applies well to most ionic liquids and states that the conductivity and viscosity are related, and as a corollary the mobility of ions is dependent as well. We systematically investigate the diffusion parameters that lead to insight of reaction kinetics of the chloroaluminate system.

                There are methods that have been developed for this type of investigation that do not apply to this system because of a few complications. For example, the Cottrell model indicates a diffusion controlled system, while attempting Koutecky-Levich does not apply because the system cannot reach a diffusion limited regime in the window of the electrolyte. We assert that in a quiescent system the ionic interactions of the electrolyte become quite strong and resistant to cleavage, leading to restricted Al₂Cl₇^- diffusion. The structuring and interaction of the ionic liquid may be shown in the rheological data, indicating that the solutions are thixotropic. Since the conductivity and diffusivity are dependent on the viscosity, we assert that as the solution is allowed to rest and coordinate in a preferred structure the diffusivity would decrease significantly. This implies that the measured diffusion rates in a quiescent experiment are not relevant in a system that implements convection, which is almost any real system.

We apply data derived from the quiescent system experiments to dynamic systems by understanding how the macroscale physical properties change as the system goes from static to dynamic. As mentioned above, the Cottrell equation can measure the diffusion rates of an electrochemically active species, which allows measurement of transport for Al₂Cl₇^-. Proton pulse field gradient spin echo nuclear magnetic resonance spectroscopy (PGSE NMR) can measure the diffusivity of the cationic species in the system (EMIm⁺), and the bulk conductivity was used to determine the combined mobility of ions in solution. The anions and cations contribute to the bulk conductivity separately, which necessitates the multiple methods of investigation, as well as the redundancy of using solutions of varying compositions. All of these experiments give results based on a static solution.

We apply the viscosity dependence to the data gathered in static solutions to predict properties in dynamic conditions. A close approximation of zero shear viscosity was determined through falling ball experiments. The dynamic viscosity was measured by a rheometer at varying shear rates. The diffusion properties and conductivity are related to viscosity. Composition and temperature variation in the viscosity studies gave insight to the mobility of both the anions and cations under all static or dynamic conditions, and point to complex interaction at the electrode interface for deposition from the Al₂Cl₇^- in this ionic liquid.

Using the diffusivity and composition data gathered from this set of experiments allows for investigation of the reaction kinetics, and optimization of the kinetics to diffusion balance. Increasing the plating rate reduces fabrication times but infringing on the precursor diffusion limits leads to dendritic and poor quality deposits. Adjusting the parameters to give similar kinetic and diffusion properties of the Al₂Cl₇^- in this electrolyte will give the best deposition properties based on both rate and dense, compact and level morphologies.

Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-AC04-94AL85000.