Stanford students, postdocs, faculty, and staff interested in energy are invited to our 11th annual summer series. Learn about cutting-edge science and clean-energy breakthroughs from the students doing the research.
MONDAYS, July 12 – August 23, 4:00 – 5:15 pm, via Zoom.
Please register with your Stanford email address here. Zoom details will be sent to you shortly.
Speaker Schedule, Abstracts and Bios
Weier Wan (left) and Haitong Li (right): Novel Computing System Enables Energy-Efficient AI Processing
Abstract: Performing computationally demanding artificial intelligence (AI) tasks consumes a significant amount of energy using today’s computer hardware, leading to both enormous data-center energy usage and limited edge device stand-by time. Majority of the energy in today’s AI hardware is spent at moving massive data between separate compute and memory units that locate on different chips. Our study shows that by combining a novel system integration technology that integrates multiple layers of compute and memory devices vertically on the same chip, and energy-efficient AI accelerator architectures that substantially reduce data movement activities, we can obtain orders-of-magnitude improvement in AI processing energy-efficiency to power the ever-demanding AI workloads of the future.
Bio: Weier Wan is pursuing a Ph.D. degree at Stanford University, Department of Electrical Engineering, supervisedby Prof. H.-S. Philip Wong. His research centers around building energy-efficient Compute-In-Memory hardware for AI acceleration. His research efforts span the full-stack of AI system including machine learning, chip design and testing. Previously he received his Master degree from Stanford University in 2017, and Bachelor degree from University of California, Berkeley.
Haitong Li is an EE PhD candidate at Stanford University, supervised by Prof. H.-S. Philip Wong. He received M.S. in electrical engineering from Stanford University in 2017, and B.S. in microelectronics from Peking University, China, in 2015. His research focuses on energy-efficient machine learning systems enabled by emerging nanotechnologies. He is a recipient of 2019 IEEE EDS PhD Student Fellowship and 2016 IEEE EDS Masters Student Fellowship.
Gustavo Cezar: Powernet: A Cloud-Based Behind-the-Meter Resource Management Platform
Abstract: Coordinating behind-the-meter (BTM) distributed energy resources is critical to ensure efficiency and reliability for consumers facing an increasingly variable grid supply. Such coordination of heterogeneous resources at scale, outside of very controlled environments, has remained a challenge due to limitations in sensing, communications and modeling in field settings. We develop Powernet, an end-to-end, cloud-based system for cost effective, scalable, and secure coordination of BTM resources within and across multiple consumer sites. We demonstrated the system in our lab and in a dairy farm where cost savings ranged from 40.75% to 92.20% and decreases reliance on grid by up to 92.68%.
Bio: Gustavo Cezar received his B.S. in Controls and Automation Engineering and M.Sc. in Mechanical Engineering at PUC-Rio, in Brazil where he developed the mechanical, electrical and control system for internal combustion engines testbeds. He is currently a PhD candidate at CEE working with Prof. Ram Rajagopal developing hardware, software, and algorithms to enable high penetration of distributed energy resources into the grid, and taking academic research to real-world applications.
Peter Csernica: Battery Degradation Through Cycling‑Induced Oxygen Release
Abstract: While Li‑rich layered oxides offer higher reversible capacities than traditional Li‑ion positive electrodes, they suffer from significant voltage degradation over cycling. Here, we use X‑ray spectromicroscopy to show that oxygen originating in the material bulk is slowly released over hundreds of cycles. Surprisingly, the released oxygen manifests as persistent oxygen vacancies, which contribute to voltage degradation due to both structural and electronic considerations. Additionally, we show that, even within the same electrode, particle morphology has a profound impact on the oxygen release process. Our work informs promising chemical, structural, and morphological approaches to achieving stability in Li‑rich electrodes.
Bio: Peter Csernica received his A.B. in Chemistry from Cornell University in 2016. He is currently a 5th year PhD student in Materials Science working with Professor William Chueh. His work focuses on the understanding of the atomic and electronic structure of positive electrode materials for Li‑ion batteries.
David Boyle: Corrosion of Lithium Metal Anodes and its Microscopic Origins
Abstract: Rechargeable lithium (Li) metal batteries must have long cycle life and calendar life. Emphasis has been placed on prolonging the cycle life of Li metal anodes, but calendar ageing is studied less often. Here, we show the effects of calendar ageing on Li metal anodes in several electrolytes. We use cryogenic transmission electron microscopy to show that chemical corrosion of Li and the continuous growth of the solid electrolyte interphase—a passivation film on Li—cause losses of capacity during storage. Surprisingly, state-of-the-art electrolytes with long cycle life are also susceptible to chemical corrosion. The results establish new design criterion for Li metal batteries and show that the rate of solid electrolyte interphase growth and the surface area of electrodeposited Li metal must simultaneously be minimized.
Bio: David is a Stanford University PhD student of Chemistry in Professor Yi Cui’s lab. Prior tojoining Stanford, he graduated with a Bachelor's degree in chemistry from James Madison University. There, he worked with Professor Ashleigh Baber on understanding the surface chemistry of Au-based heterogeneous catalysts. Now at Stanford, he studies the impact of electrolyte composition on several kinetic processes associated with lithium metal anodes. These processes include growth of the solid electrolyte interphase, nucleation and growth of lithium, and the interfacial electron transfer kinetics of lithium deposition.
Shang Zhai: Carbon-free Hydrogen from Natural Gas Pyrolysis
Abstract: One of the biggest challenges for this century is to meet people’s growing need for affordable energy, while decreasing greenhouse gas emissions at Gigaton (Gt) scale. Thermochemical processes are the only type of Gt-scale processes developed by human beings today. Meanwhile, cheap and carbon-free hydrogen (H2) is a critical missing piece to realize carbon neutrality. Natural gas pyrolysis leverages this abundant resource to produce carbon-free H2 and solid carbon that is easy to sequester. The long lasting challenge in commercialization of natural gas pyrolysis is carbon deposition on catalyst. Our work creates a special “non-stick” oxide layer as catalyst support.
Bio: Shang Zhai is a postdoctoral fellow in Pr. Arun Majumdar’s lab in Mechanical Engineering. He obtained his PhD from Stanford in 2020. Shang is passionate about applying thermal sciences and materials chemistry to revolutionary technologies for a sustainable future. His research has focused on designing and understanding mixed metal oxides for thermochemical energy cycles, including water splitting, CO2 splitting and methane conversion.
Justin Luke: Joint Optimization of Electric Vehicle Fleet Operations and Charging Station Siting
Abstract: Charging infrastructure is the coupling link between power and transportation networks, thus determining station siting is necessary for planning of power and transportation systems. This presentation introduces a linear program that optimizes for station siting and macroscopic fleet operations in a joint fashion. From a case-study of an electric vehicle fleet operating in San Francisco, our results show that small EVs with low procurement costs are the most cost-effective in terms of total ownership costs. Furthermore, the optimal siting of charging stations is more spatially distributed than the current placements of stations and reduces total costs by up to 10%.
Bio: Justin Luke is a PhD candidate in Civil and Environmental Engineering and is co-advised by Professor Marco Pavone (Autonomous Systems Lab) and Professor Ram Rajagopal (Sustainable Systems Lab). His research focuses on grid integration of autonomous electric vehicle fleets, in particular identifying synergies with renewable energy resources integration. Justin has obtained a MS in Electrical Engineering at Stanford University in 2020 and a BS in Energy Engineering at the University of California, Berkeley in 2018.
Ejeong Baik: What is Different About Net Zero Energy Systems
Abstract: Previous literature has highlighted the value of clean firm resources in decarbonizing electricity systems cost-effectively. This study utilizes three independent capacity expansion and dispatch models to assess three different types of clean firm resources and their roles in decarbonizing California’s electricity grid. The analysis finds that while each clean firm resource can provide similar cost savings value to the grid, the resources also complement each other to achieve the most savings when operating in a grid together.
Bio: EJ Baik received her BSE in Civil and Environmental Engineering from Princeton University in 2016. She is currently a 5th year PhD student in Energy Resources Engineering. EJ works with Professor Sally Benson on decarbonizing large-scale energy systems.
Pouya Rezazadeh: Design and Optimization of Integrated Infrastructure for Maximum Sustainability
Abstract: While covering only 3% of the earth’s land, cities accommodate 55% of world population and are responsible for more than two-thirds of global energy consumption and greenhouse gas emissions. My research tries to improve the livability and reduce the energy use and emissions of urban districts by integrating the design of energy system, building mix, wastewater treatment, and EV charging infrastructure while concurrently optimizing them for sustainability targets. Results show that urban districts designed with this integrated approach are more socially, economically, and environmentally sustainable than districts designed with the traditional, segregated approach.
Bio: Pouya did his undergrad in Civil & Environmental Engineering at Sharif University of Technology, and is now a 4th year PhD candidate in Civil & Environmental Engineering, minoring in Computer Science, at Stanford. Under the supervision of Professor Michael Lepech, Pouya works on frameworks for optimally integrating the design of infrastructure to achieve higher sustainability in urban areas.
Emily Lacroix: Managing Anoxic Microsites to Increase Carbon Storage in Agricultural Soils
Abstract: Enhancing soil carbon storage in managed lands is a promising biological negative emissions strategy for carbon dioxide removal from the atmosphere. Anoxic microsites, small zones of oxygen depletion in otherwise aerated soils, are an unquantified determinant of soil carbon storage. Unlike other controls on soil carbon storage, anoxic microsites can be actively managed. In this talk, we present two methods for quantifying the soil carbon preservation imparted by anoxic microsites in upland soils. Our results show that anoxic microsites enhance carbon storage in soils that were previously thought to have low carbon storage potential.
Bio: Emily Lacroix received her B.A. in Chemistry and Environmental Studies from Dartmouth College. She is currently a 4th year Ph.D. candidate in Earth System Science, advised by Professor Scott Fendorf. Emily’s research explores how anoxic microsites contribute to soil carbon storage across soils of different properties and management practices.
Julian Vigil: Understanding and Controlling Defects in Halide Perovskite Semiconductors
Abstract: Crystalline semiconductors serve as the basis for many current and next-generation energy technologies, including photovoltaics and solid-state lighting. Stable and efficient optoelectronic devices based on these materials are engineered, in many cases, to include or exclude sparse defects owing to their outsized influence on the properties of the bulk material. We highlight the prevalence of point defect reactions occurring in the halide perovskites, an emergent class of optoelectronic materials, and present our efforts to quantify, characterize, and mitigate point defect formation in halide double perovskites.
Bio: Julian Vigil received his B.S. in Chemical Engineering from the University of New Mexico in 2017 and an M.Phil. in Chemistry from the University of Cambridge in 2018. He is currently a third-year Ph.D. candidate in Chemical Engineering. Julian works with Prof. Hemamala Karunadasa and Prof. Michael Toney on understanding defects in halide perovskites and X-ray scattering and spectroscopy measurements.
Yuhao Nie: Training Machine Vision Systems for PV Power Output Prediction
Abstract: Solar photovoltaic (PV) is growing rapidly in the last decade. Fluctuations in PV power output due to short-term cloud events can have large impacts in areas with high solar PV penetration. For this reason, significant effort has been made to predict PV output using a variety of methods. Images of the sky contain a wealth of information, but this information is challenging to extract and use for reliable predictions. More recently, efforts have shifted to using machine vision systems to “read” the sky and make predictions of PV panel output. In this lecture, I will introduce some of our work on such systems.
Bio: Yuhao Nie is currently a 3rd PhD student in Energy Resources Engineering working with Professor Adam Brandt. His current research focuses on developing machine vision systems for short-term solar forecasting. His past work includes quantifying the environmental and economic impact of different energy systems, such as biofuels from forest residues and liquefied natural gas, as well as building optimization models to assist decisions in energy systems.
Lily Buechler: Accelerating Power Flow Simulations with Machine Learning
Abstract: Power flow simulations are used extensively by electric utilities and researchers for grid operation and planning. These simulations involve solving the nonlinear power flow equations using numerical methods, which can be computationally expensive especially for large power systems. We develop a machine learning-based framework for accelerating power flow simulations by adaptively training a data-driven model in-the-loop with a Newton-Raphson solver. We implement this framework in the GridLAB-D power flow simulation engine and validate performance on a variety of distribution systems, resulting in significant computational improvements.
Bio: Lily Buechler is a 4th year PhD student in Mechanical Engineering working with Professor Ram Rajagopal. She also works with the Grid Integration, Systems, and Mobility (GISMo) group at SLAC National Laboratory. Her research focuses on data-driven control, optimization, and simulation in power system applications. She received her B.S. from Tufts University in 2017 and M.S. from Stanford in 2019, both in Mechanical Engineering.
Anela Arifi: Exploratory Conception of Sustainable Bioenergy Supply Chains
Abstract: While decarbonizing light-duty vehicles is relatively straightforward through electrification, decarbonizing heavy-duty vehicles mandates energy sources with higher energy density and is a fundamental challenge that needs to be tackled to mitigate climate change. With its high energy density, ethanol produced from agricultural residues is a suitable biofuel and is expected to have the lowest cost of production for the indefinite future. However, removing agricultural residues from the soil to produce ethanol is widely viewed as competitive with soil fertility. Our modelling evidence shows that returning processed agricultural residues resulting from ethanol production to the soil has a greater potential for increasing soil fertility than leaving raw agricultural residues.
Bio: Anela is an E-IPER Ph.D. student and a Knight-Hennessy scholar co-advised by Chris Field and Sally Benson. Her research focuses on the role of bioenergy in decarbonization scenarios and the nexus between socio-economic and environmental factors affecting bioenergy systems. While in her home country, Bosnia and Herzegovina, Anela developed renewable energy systems for rural communities using waste chicken feathers and municipal waste. To raise awareness about energy poverty, she spoke at TED and the UN.
Austin Flick: Scalable Processes for Manufacturable Perovskite Solar Modules
Abstract: Perovskite solar cells have demonstrated a considerable rise in power conversion efficiency (>25%) in recent years. However, significant barriers towards commercialization remain—these performances have yet to be achieved on large areas (~1 m2) as the record-achieving fabrication methods are inherently unscalable. Additional insights from cost-modeling of perovskite fabrication reveal high-impact areas for the development of alternative, high-throughput fabrication techniques. The co-optimization of performance, throughput, and cost enables the efforts demonstrated in this work towards roll-to-roll and sheet-to-sheet manufacturing of perovskite solar modules. These efforts are further validated by preliminary LCOE calculations that demonstrate a pathway towards commercialization.
Bio: Austin Flick received both his Bachelor's and Master's degrees, with distinction, from Stanford University's Department of Materials Science and Engineering in 2019. He is currently a 2nd year PhD student in Materials Science working with Professor Reinhold Dauskardt towards the development of scalable, high-throughput methods to fabricate perovskite solar modules.