HOME
R & D
Research Demonstrations
ACADEMIA
Publications Teaching Mentoring
CONTACT
  • Conference & Seminar
    •
  • Publications
    •
    PUBLICATIONS

    Conference and Seminar Proceedings

    Johns Hopkins University Mechanical Engineering Seminar Series, October 2020

    Hybrid Physics and Machine Learning Framework: Achieving High Fidelity Modeling While Reducing the Computational Cost for First Principle Density Functional Theory Calculations, LINK

    Click on the image to see the presentation abstract.

    <b>Abstract</b><p align='left'>In this talk, I describe a novel machine learning-based technique to reduce the computational cost required to explore large design spaces of substitutional alloys. The first advancement is based on a neural network (NN) approach to predict the initial position of minority and majority ions prior to DFT relaxation. The second advancement is to allow the NN to predict the total energy for every possible minority ion position and select the most stable configuration in the absence of relaxing each trial minority configuration. The third advancement is to use the machine learning approach to make future predictions for candidate ions. A bismuth oxide materials system, (BixLayYbz)2MoO6, was used as the model system to demonstrate the developed methods and quantify the resulting computational speedup. Compared to a brute force method that requires the calculation of every permutation of minority configuration and subsequent DFT relaxation, a significant speedup was realized if the NN predicted the initial configuration of ions prior to relaxation. Implementation of the second advancement allowed the NN to predict the total energy for all possible trial configurations and down select the most stable configurations prior to relaxation. Finally, the third advancement allowed the NN to predict thermodynamic reaction barriers during nitrogen fixation to further reduce the computational cost and make future predictions for candidate ions. Validation was done by comparing the NN and DFT predictions for the position, energy, and reaction pathways.</p>

  • During the past two decades, first-principle calculations based on density-functional theory (DFT) unfolded as a successful approach to solve the electronic structure of matter. DFT is a widely used computational quantum mechanical modeling method that helps investigate the electronic structure and properties of many-body systems. The theory can reduce the many-body Schrödinger equation to an effective single-electron problem by relying on the Hohenberg-Kohn theorem and Kohn-Sham method, thus making material property predictions computationally feasible. The renowned success of DFT for describing ground-state properties for vast classes of materials such as semiconductors, insulators, half metals, semimetals, transition metals, etc., at the nanostructure scale makes it one of the most used methods for modern electronic structure analyses. Due to the extreme computational costs of most theoretical studies, limitations can and do arise when using approximation methods because accuracy is compromised in exchange for speedup time.

    • EMS TMS, June 2017

    Electronic Properties of a UiO-66 Metal-Organic Framework

  • Metal-Organic Frameworks (MOFs) have received considerable attention and fast development in the past few years. These materials have demonstrated a wide range of applications due to their porosity, tailorability of optical properties, and chemical selectivity. This report catalogs common MOF designs based on application and diversity in various fields, as well as conduct an in-depth study of inorganic substitution in a functionalized MOF.

    • Selected Publications

    Below are a list of some publications Dr. Yasin has done and collaborated on. Click on the journal cover/image to see the article abstract.

    <b>Abstract</b><p align='left'>Machine learning and artificial intelligence algorithms have expanded dramatically in use across diverse fields of research and practice. Despite the extensive benefits that these algorithms can bring to researchers, system designers, and operators alike, the adoption of these algorithms in space-related scenarios has lagged behind other fields. In order to encourage the increased adoption of artificial intelligence and machine learning techniques to space-domain-related problems, flexible modeling and simulation capabilities are needed to build stakeholder trust in these techniques. This research presents the development of a flexible Python-based modeling and simulation environment for applying Reinforcement Learning to Low Earth Orbit satellite Hyper Spectral Imaging sensor tasking. With the transition away from small numbers of highly exquisite on-orbit systems to proliferated architectures characterized by constellations of lower cost and complexity spacecraft, the methods by which payload sensors are tasked have become dynamic and complex, making the problem of determining effective sensor tasking methods an important area of research. Such a problem lends itself well to the application of Reinforcement Learning. The focus of this work is on developing the role of intelligent systems in improving the data acquisition process in a space-based hyperspectral imaging system, and showing how the developed modeling and simulation framework can be successfully employed to improve the acquisition of targets of interest. A key strength of the presented reinforcement learning application framework is its non-commercial, extensible nature, suitable for both research and educational purposes.</p>

    Reinforcement Learning Application to Satellite Constellation Sensor Tasking
    <b>Abstract</b><p align='left'>The inception of spatial transcriptomics has allowed improved comprehension of tissue architectures and the disentanglement of complex underlying biological, physiological, and pathological processes through their positional contexts. Recently, these contexts, and by extension the field, have seen much promise and elucidation with the application of graph learning approaches. In particular, neural operators have risen in regards to learning the mapping between infinite-dimensional function spaces. With basic to deep neural network architectures being data-driven, i.e. dependent on quality data for prediction, neural operators provide robustness by offering generalization among different resolutions despite low quality data. Graph neural operators are a variant that utilize graph networks to learn this mapping between function spaces. The aim of this research is to identify robust machine learning architectures that integrate spatial information to predict tissue types. Under this notion, we propose a study to validate the efficacy of applying neural operators towards classification of brain regions in mouse brain tissue samples as a proof of concept towards our purpose and compare it against various state of the art graph neural network approaches. We were able to achieve an F1 score of nearly 72% for the graph neural operator approach which outperformed all baseline and other graph network approaches within the scope of supervised learning.</p>

    Graph Neural Operators for Classification of Spatial Transcriptomics Data
    <b>Abstract</b><p align='left'>In this study, a machine learning-based technique is developed to reduce the computational cost required to explore large design spaces of substitutional alloys. The first advancement is based on a neural network (NN) approach to predict the initial position of minority and majority ions prior to DFT relaxation. The second advancement is to allow the NN to predict the total energy for every possibility minority ion position and select the most stable configuration in the absence of relaxing each trial minority configuration. A bismuth oxide materials system, (BixLayYbz)2MoO6, was used as the model system to demonstrate the developed methods and quantify the resulting computational speedup. Compared to a brute force method that requires the calculation of every permutation of minority configuration and subsequent DFT relaxation, a 1.3× speedup was realized if the NN predicted the initial configuration of ions prior to relaxation. Implementation of the second advancement allowed the NN to predict the total energy for all possible trial configurations and downselect the most stable configurations prior to relaxation, resulting in a speedup of approximately 37×. Validation was done by comparing position and energy between the NN and DFT predictions. A maximum position vector mean squared error (MSE) of 1.6 × 10⁻² and a maximum energy MSE of 2.3 × 10⁻⁷ was predicted for the worst case configuration. This method demonstrates a significant computational speedup, which has the potential for even greater computational savings for larger compositional design spaces.</p>

    A Machine Learning Approach for Increased Throughput of Density Functional Theory Substitutional Alloy Studies
    <b>Abstract</b><p align='left'>The Haber-Bosch (HB) process is the main industrial method for converting atmospheric nitrogen (N2) to ammonia (NH3). In the HB process, hydrogen is typically derived from steam reforming natural gas, which results in a large quantity of greenhouse emissions and considerable environmental concern. In providing a sustainable solution, the following research investigates a semiconductor-based photocatalytic approach that emulates the biological nitrogen fixation (BNF) pathway. By emulating the biological process, the overall reaction energy, pressure, and temperature are significantly reduced. Moreover, this allows ammonia to be produced from atmospheric nitrogen at atmospheric pressures. To aid in material selection and to provide fundamental insight into the process, the following research leverages first principle calculations to predict both the light absorption properties and the thermodynamic reaction barriers of the associative (BNF) and dissociative (HB) reaction pathways.</p> <p align='left'>A major limitation in the throughput of alloy-based Density Functional Theory (DFT) studies for large design spaces is the wall time required to find the optimal position of the minority ions. A novel method was developed that involves a machine learning approach to predict the minority ion positions prior to simulation. This lead to a significant decrease in computational wall time required to identify and solve the most stable minority configuration. By training the positions based on a select number of DFT cases, it was found that positions of other species could be predicted with mean square error (MSE) less than 1.6x10−2. Moreover, it was found that the total energy for these configurations could also be predicted with MSE less than 2.3x10−7 relative to the subsequent DFT calculations.</p> <p align='left'>In applying the developed techniques, two design spaces were investigated, (MxBiy)2MoO6 (M = Fe, La, and Yb) and sulfur desorbed MxMoyS2 (M = Co and Fe). In the case of MxMoyS2, phases with a higher concentration of Co and Fe elements resulted in a lower reaction barrier. More specifically, by incorporating Co and Fe into the structure, the affinity toward hydrogen species was exchanged with increased nitrogen affinity. In turn, the Fe containing 1T phase of Fe0.25Mo0.75S2 for the dissociative pathway was predicted to have the lowest overall reaction barrier within the MxMoyS2 design space. In the case of (MxBiy)2MoO6, higher concentration of Fe and Yb resulted in an increase of the Mo-O bond length and improved nitrogen affinity. The best composition for this design space was predicted to be the orthorhombic phase of (Fe0.25Bi0.75)2MoO6 for the associative pathway with a reaction barrier of 1.4 eV.</p>

    First Principle Study of Nitrogen Fixation on Molybdenum Containing Semiconductor Surface with a Developed Machine Learning Approach
    <b>Abstract</b><p align='left'>In this study the band gap modulation was studied in response to inorganic ion substitution within a thermally stable UiO-66 metal–organic framework (MOF). A combination of density functional theory prediction in conjunction with experimental predictions were used to map out the complete composition space for three inorganic ions (Zr, Ti, Hf) and three functional groups. The three functional groups include an amino group (NH2), a nitro group (NO2), and a hydrogenated case (H). The smallest band gap that experimentally determined was for a partially substituted UiO-66(Ti5Zr1)-NH2 resulting in 2.60 eV. Theoretical results indicated that Ti can be fully substituted within the lattice resulting in a predicted band gap as low as 1.62 eV. Modulation was a result of a mid-gap state introduced through the amino functionalization and HOMO shifting as a result of increased binding of the Ti–O–C bonds.</p>

    Study of the Inorganic Substitution in a Functionalized UiO-66 Metal-Organic Framework
    <b>Abstract</b><p align='left'>The vast majority of semiconductors photocatalysts reported for artificial nitrogen fixation have a large bandgap at around 3.0 eV, thus photocatalytic nitrogen reduction is driven mainly by ultraviolet light. In contrast, this report demonstrates for the first time that bismuth iron molybdate (Bi 3 FeMo 2 O 12 ) with a bandgap of 2.25 eV exhibits visible-light photocatalytic activity toward nitrogento-ammonia conversion. Furthermore, introduction of oxygen vacancy to this photocatalyst increases the ammonia production rate remarkably. Density functional theory (DFT) calculation reveals that the oxygen vacancies help adsorb and stabilize the N-H intermediate species, and lower the energy barrier of intermediate reactions. This work has an implication in design of semiconductor photocatalysts for sustainable ammonia synthesis under the ambient condition using solar energy.</p>

    Visible-Light Bismuth Iron Molybdate Photocatalyst for Artificial Nitrogen Fixation
    <b>Abstract</b><p align='left'>This study investigates the photocatalytic nitrogen fixation on a cation-doped surface (BixMy)2MoO6 where (M=Fe, La, Yb) in both the orthorhombic and monoclinic configurations using a density functional theory (DFT) approach with experimentally validated model inputs. The proceeding discussion focuses on the Heyrovsky-type reactions for both the associative and dissociative reaction pathway related to nitrogen reduction. Key fundamental insight in the reduction mechanism is discussed that relates the material properties of the substitutional ions to the nitrogen and hydrogen affinities. Physical insight is gathered through interpretation of bound electronic states at the surface. Compositional phases of higher Fe and Yb concentrations resulted in decreased Mosingle bondO binding and increased affinity between Mo and the N and H species on the surface. The modulation of the Mosingle bondO binding is induced by strain as Yb and Fe are implemented, this, in turn, shifts energy levels and modulates the band gap energy by approximately 0.2eV. This modification of Mosingle bondO bond as substitution occurs is a result of the orbital hybridization of Msingle bondO (M=Fe, Yb) that causes a strong orbital interaction that shifts states up toward the Fermi. The optimal composition was predicted to be an orthorhombic configuration of (Bi0.75Fe0.25)2MoO6 with a predicted maximum thermodynamic energy barrier of 1.4eV. This composition demonstrates effective nitrogen and hydrogen affinity that follows the associative or biological nitrogen fixation pathway.</p>

    Density Functional Theory Evaluation of Cation-doped Bismuth Molybdenum Oxide Photocatalysts for Nitrogen Fixation
    <b>Abstract</b><p align='left'>This study leverages density functional theory accompanied with Boltzmann transport equation approaches to investigate the electronic mobility as a function of inorganic substitution and functionalization in a thermally stable UiO-66 metal-organic framework (MOF). The MOFs investigated are based on Zr-UiO-66 MOF with three functionalization groups of benzene dicarboxylate (BDC), BDC functionalized with an amino group (BDC+NH2) and a nitro group (BDC+NO2). The design space of this study is bound by UiO-66(M)-R, [M=Zr, Ti, Hf; R=BDC, BDC+NO2, BDC+NH2]. The elastic modulus was not found to vary significantly over the structural modification of the design space for either functionalization or inorganic substitution. However, the electron–phonon scattering potential was found to be controllable by up to 30% through controlled inorganic substitution in the metal clusters of the MOF structure. The highest electron mobility was predicted for a UiO-66(Hf5Zr1) achieving a value of approximately 1.4×10−3cm2/V s. It was determined that functionalization provides a controlled method of modulating the charge density, while inorganic substitution provides a controlled method of modulating the electronic mobility. Within the proposed design space the electrical conductivity was able to be increased by approximately three times the base conductivity through a combination of inorganic substitution and functionalization.</p>

    Ab-initio Study of the Electron Mobility in a Functionalized UiO-66 Metal Organic Framework

    Load More