UMD Theses and Dissertations

Permanent URI for this collectionhttp://hdl.handle.net/1903/3

New submissions to the thesis/dissertation collections are added automatically as they are received from the Graduate School. Currently, the Graduate School deposits all theses and dissertations from a given semester after the official graduation date. This means that there may be up to a 4 month delay in the appearance of a given thesis/dissertation in DRUM.

More information is available at Theses and Dissertations at University of Maryland Libraries.

Browse

Filters

2015 - 20162

Settings

Subject: search.filters.subject.Reaction factorization

Search Results

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    REACTION NETWORK ANALYSIS FOR THIN FILM DEPOSITION PROCESSES
    (2016) Ramakrishnasubramanian, Krishnaprasath; Adomaitis, Raymond; Chemical Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
    Understanding the growth of thin films produced by Atomic Layer Deposition (ALD) and Chemical Vapor Deposition (CVD) has been one of the most important challenge for surface chemists over the last two to three decades. There has been a lack of complete understanding of the surface chemistry behind these systems due to the dearth of experimental reaction kinetics data available. The data that do exist are generally derived through quantum computations. Thus, it becomes ever so important to develop a deposition model which not only predicts the bulk film chemistry but also explains its self-limiting nature and growth surface stability without the use of reaction rate data. The reaction network analysis tools developed in this thesis are based on a reaction factorization approach that aims to decouple the reaction rates by accounting for the chemical species surface balance dynamic equations. This process eliminates the redundant dynamic modes and identifies conserved modes as reaction invariants. The analysis of these invariants is carried out using a Species-Reaction (S-R) graph approach which also serves to simplify the representation of the complex reaction network. The S-R graph is self explanatory and consistent for all systems. The invariants can be easily extracted from the S-R graph by following a set of straightforward rules and this is demonstrated for the CVD of gallium nitride and the ALD of gallium arsenide. We propose that understanding invariants through these S-R graphs not only provides us with the physical significance of conserved modes but also give us a better insight into the deposition mechanism.
  • Thumbnail Image
    Item
    REACTION FACTORIZATION FOR DISTRIBUTED DEPOSITION SYSTEMS: APPLICATION TO COPPER FILM GROWTH
    (2015) Arana-Chavez, David; Adomaitis, Raymond A; Chemical Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
    A model reduction methodology based on a Gauss-Jordan reaction factoriza- tion for thin-film deposition reaction systems is developed in this thesis. The fac- torization generates a transformation matrix that is used to create a new coordinate system that guides the separation of the deposition process time scales by decoupling the net-forward reaction rates to the greatest extent possible. The new coordinate space enables recasting the original model as a singular perturbation problem and consequently as a semi-explicit system of differential-algebraic equations (DAE) for the dominant dynamics in the pseudo-equilibrium limit. Additionally, the factor- ization reveals conserved quantities in the new reaction coordinate system as well as potential structural problems with the deposition reaction network. The reaction factorization methodology is formulated to be suitable for appli- cation to dynamic, spatially distributed reaction systems. The factorization provides a rigorous pathway to decouple the time evolution and the spatial distributions of deposition systems when the dynamics of reactor-scale gas-phase transport are fastrelative to the deposition process. Moreover, the factorization approach provides a solution to the problem of formulating Danckwerts-type boundary conditions where gas-phase equilibrium reactions are important. The reaction factorization is used to study the chemical vapor deposition of copper on a tubular hot-wall reactor using copper iodide as the Cu precursor. A film-growth mechanism is proposed from experimental observations that the copper films deposited on quartz substrates suggest a Volmer-Weber growth mode. A model based on this mechanism is used to track spatial distribution of the average Cu island size in the reactor. The rate expressions used in the Cu deposition model are determined using absolute rate theory. To carry out these calculations in an organized manner, a library of object-oriented classes are created in the Python programming language.