BZPEERReading, trust and collaboration for research
Menu

Discover

GraphFollow connections around a paper, topic or saved view.

Workspaces

HomePick up where your research flow left off.InboxHandle requests, replies and notifications from one place.CollectionsCurate reading lists, evidence sets and shareable dossiers.ResearchSee your private provenance graph, trail and imports.CollaborationMove from discovery into focused discussion rooms.PublishDraft, preview and publish full-text research articles.

Network

ExpertsFind specialists worth following or asking for help.CommunitiesJoin research circles organized around shared work.PartnersSee external organizations active around the same topics.

Opportunities

ListingsBrowse roles, calls and collaborations with clearer fit signals.

Account

Log inReturn to your reading and collaboration workspace.Create accountStart saving work, following people and joining teams.
Log inCreate account

Researcher profile

Tonio Buonassisi

Tonio Buonassisi contributes to research discovery and scholarly infrastructure.

ResearcherAffiliation not importedOpen to collaborate
cond-mat.mtrl-sciMachine Learningeess.SPphysics.app-phphysics.comp-phcond-mat.mes-hallphysics.flu-dynphysics.soc-phRobotics

Trust snapshot

Quick read

Trust 21 - EmergingVerification L1Unclaimed author
9works
0followers
9topics
4close collaborators

Actions

Decide how to stay connected

Follow researcher0

Identity and collaboration

How to connect with this researcher

Claiming links this public author record to a researcher profile and unlocks direct collaboration workflows.

Log in to claim

Direct collaboration

Open a focused conversation when the fit is right

Claim this author entity first to unlock direct invitations.

Research graph

See the researcher in context

Open full explorer

Inspect adjacent work, topics, institutions and collaborators without jumping out to a separate graph page.

Building this graph slice

BZPEER is loading the nearby papers, people, topics and institutions for this page.

Published work

9 published item(s)

preprint2022arXiv

A Machine Learning and Computer Vision Approach to Rapidly Optimize Multiscale Droplet Generation

Generating droplets from a continuous stream of fluid requires precise tuning of a device to find optimized control parameter conditions. It is analytically intractable to compute the necessary control parameter values of a droplet-generating device that produces optimized droplets. Furthermore, as the length scale of the fluid flow changes, the formation physics and optimized conditions that induce flow decomposition into droplets also change. Hence, a single proportional integral derivative controller is too inflexible to optimize devices of different length scales or different control parameters, while classification machine learning techniques take days to train and require millions of droplet images. Therefore, the question is posed, can a single method be created that universally optimizes multiple length-scale droplets using only a few data points and is faster than previous approaches? In this paper, a Bayesian optimization and computer vision feedback loop is designed to quickly and reliably discover the control parameter values that generate optimized droplets within different length-scale devices. This method is demonstrated to converge on optimum parameter values using 60 images in only 2.3 hours, 30x faster than previous approaches. Model implementation is demonstrated for two different length-scale devices: a milliscale inkjet device and a microfluidics device.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Introducing flexible perovskites to the IoT world using photovoltaic-powered wireless tags

Billions of everyday objects could become part of the Internet of Things (IoT) by augmentation with low-cost, long-range, maintenance-free wireless sensors. Radio Frequency Identification (RFID) is a low-cost wireless technology that could enable this vision, but it is constrained by short communication range and lack of sufficient energy available to power auxiliary electronics and sensors. Here, we explore the use of flexible perovskite photovoltaic cells to provide external power to semi-passive RFID tags to increase range and energy availability for external electronics such as microcontrollers and digital sensors. Perovskites are intriguing materials that hold the possibility to develop high-performance, low-cost, optically tunable (to absorb different light spectra), and flexible light energy harvesters. Our prototype perovskite photovoltaic cells on plastic substrates have an efficiency of 13% and a voltage of 0.88 V at maximum power under standard testing conditions. We built prototypes of RFID sensors powered with these flexible photovoltaic cells to demonstrate real-world applications. Our evaluation of the prototypes suggests that: i) flexible PV cells are durable up to a bending radius of 5 mm with only a 20 % drop in relative efficiency; ii) RFID communication range increased by 5x, and meets the energy needs (10-350 microwatt) to enable self-powered wireless sensors; iii) perovskite powered wireless sensors enable many battery-less sensing applications (e.g., perishable good monitoring, warehouse automation)

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Machine Learning with Knowledge Constraints for Process Optimization of Open-Air Perovskite Solar Cell Manufacturing

Perovskite photovoltaics (PV) have achieved rapid development in the past decade in terms of power conversion efficiency of small-area lab-scale devices; however, successful commercialization still requires further development of low-cost, scalable, and high-throughput manufacturing techniques. One of the critical challenges of developing a new fabrication technique is the high-dimensional parameter space for optimization, but machine learning (ML) can readily be used to accelerate perovskite PV scaling. Herein, we present an ML-guided framework of sequential learning for manufacturing process optimization. We apply our methodology to the Rapid Spray Plasma Processing (RSPP) technique for perovskite thin films in ambient conditions. With a limited experimental budget of screening 100 process conditions, we demonstrated an efficiency improvement to 18.5% as the best-in-our-lab device fabricated by RSPP, and we also experimentally found 10 unique process conditions to produce the top-performing devices of more than 17% efficiency, which is 5 times higher rate of success than the control experiments with pseudo-random Latin hypercube sampling. Our model is enabled by three innovations: (a) flexible knowledge transfer between experimental processes by incorporating data from prior experimental data as a probabilistic constraint; (b) incorporation of both subjective human observations and ML insights when selecting next experiments; (c) adaptive strategy of locating the region of interest using Bayesian optimization first, and then conducting local exploration for high-efficiency devices. Furthermore, in virtual benchmarking, our framework achieves faster improvements with limited experimental budgets than traditional design-of-experiments methods (e.g., one-variable-at-a-time sampling).

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Processing Induced Distinct Charge Carrier Dynamics of Bulky Organic Halide Treated Perovskites

State-of-the-art metal halide perovskite-based photovoltaics often employ organic ammonium salts, AX, as a surface passivator, where A is a large organic cation and X is a halide. These surface treatments passivate the perovskite by forming layered perovskites (e.g., A2PbX4) or by AX itself serving as a surface passivation agent on the perovskite photoactive film. It remains unclear whether layered perovskites or AX is the ideal passivator due to an incomplete understanding of the interfacial impact and resulting photoexcited carrier dynamics of AX treatment. In the present study, we use TRPL measurements to selectively probe the different interfaces of glass/perovskite/AX to demonstrate the vastly distinct interfacial photoexcited state dynamics with the presence of A2PbX4 or AX. Coupling the TRPL results with X-ray diffraction and nanoscale microscopy measurements, we find that the presence of AX not only passivates the traps at the surface and the grain boundaries, but also induces an α/δ-FAPbI3 phase mixing that alters the carrier dynamics near the glass/perovskite interface and enhances the photoluminescence quantum yield. In contrast, the passivation with A2PbI4 is mostly localized to the surface and grain boundaries near the top surface where the availability of PbI2 directly determines the formation of A2PbI4. Such distinct mechanisms significantly impact the corresponding solar cell performance, and we find AX passivation that has not been converted to a layered perovskite allows for a much larger processing window (e.g., larger allowed variance of AX concentration which is critical for improving the eventual manufacturing yield) and more reproducible condition to realize device performance improvements, while A2PbI4 as a passivator yields a much narrower processing window. We expect these results to enable a more rational route for developing AX for perovskite.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

What Information is Necessary and Sufficient to Predict Materials Properties using Machine Learning?

Conventional wisdom of materials modelling stipulates that both chemical composition and crystal structure are integral in the prediction of physical properties. However, recent developments challenge this by reporting accurate property-prediction machine learning (ML) frameworks using composition alone without knowledge of the local atomic environments or long-range order. To probe this behavior, we conduct a systematic comparison of supervised ML models built on composition only vs. composition plus structure features. Similar performance for property prediction is found using both models for compounds close to the thermodynamic convex hull. We hypothesize that composition embeds structural information of ground-state structures in support of composition-centric models for property prediction and inverse design of stable compounds.

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

An invertible crystallographic representation for general inverse design of inorganic crystals with targeted properties

Realizing general inverse design could greatly accelerate the discovery of new materials with user-defined properties. However, state-of-the-art generative models tend to be limited to a specific composition or crystal structure. Herein, we present a framework capable of general inverse design (not limited to a given set of elements or crystal structures), featuring a generalized invertible representation that encodes crystals in both real and reciprocal space, and a property-structured latent space from a variational autoencoder (VAE). In three design cases, the framework generates 142 new crystals with user-defined formation energies, bandgap, thermoelectric (TE) power factor, and combinations thereof. These generated crystals, absent in the training database, are validated by first-principles calculations. The success rates (number of first-principles-validated target-satisfying crystals/number of designed crystals) ranges between 7.1% and 38.9%. These results represent a significant step toward property-driven general inverse design using generative models, although practical challenges remain when coupled with experimental synthesis.

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

Interpretable and Explainable Machine Learning for Materials Science and Chemistry

While the uptake of data-driven approaches for materials science and chemistry is at an exciting, early stage, to realise the true potential of machine learning models for successful scientific discovery, they must have qualities beyond purely predictive power. The predictions and inner workings of models should provide a certain degree of explainability by human experts, permitting the identification of potential model issues or limitations, building trust on model predictions and unveiling unexpected correlations that may lead to scientific insights. In this work, we summarize applications of interpretability and explainability techniques for materials science and chemistry and discuss how these techniques can improve the outcome of scientific studies. We discuss various challenges for interpretable machine learning in materials science and, more broadly, in scientific settings. In particular, we emphasize the risks of inferring causation or reaching generalization by purely interpreting machine learning models and the need of uncertainty estimates for model explanations. Finally, we showcase a number of exciting developments in other fields that could benefit interpretability in material science and chemistry problems.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Bridging the gap between photovoltaics R&D and manufacturing with data-driven optimization

Novel photovoltaics, such as perovskites and perovskite-inspired materials, have shown great promise due to high efficiency and potentially low manufacturing cost. So far, solar cell R&D has mostly focused on achieving record efficiencies, a process that often results in small batches, large variance, and limited understanding of the physical causes of underperformance. This approach is intensive in time and resources, and ignores many relevant factors for industrial production, particularly the need for high reproducibility and high manufacturing yield, and the accompanying need of physical insights. The record-efficiency paradigm is effective in early-stage R&D, but becomes unsuitable for industrial translation, requiring a repetition of the optimization procedure in the industrial setting. This mismatch between optimization objectives, combined with the complexity of physical root-cause analysis, contributes to decade-long timelines to transfer new technologies into the market. Based on recent machine learning and technoeconomic advances, our perspective articulates a data-driven optimization framework to bridge R&D and manufacturing optimization approaches. We extend the maximum-efficiency optimization paradigm by considering two additional dimensions: a technoeconomic figure of merit and scalable physical inference. Our framework naturally aligns different stages of technology development with shared optimization objectives, and accelerates the optimization process by providing physical insights.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Techno-economic model of a second-life energy storage system for utility-scale solar power considering li-ion calendar and cycle aging

While the use of energy storage combined with grid-scale photovoltaic power plants continues to grow, given current lithium-ion battery prices, there remains uncertainty about the profitability of these solar-plus-storage projects. At the same time, the rapid proliferation of electric vehicles is creating a fleet of millions of lithium-ion batteries that will be deemed unsuitable for the transportation industry once they reach 80 percent of their original capacity. The repurposing and deployment of these batteries as stationary energy storage provides an opportunity to reduce the cost of solar-plus-storage systems, if the economics can be proven. We present a techno-economic model of a solar-plus-second-life energy storage project in California, including a data-based model of lithium nickel manganese cobalt oxide battery degradation, to predict its capacity fade over time, and compare it to a project that uses a new lithium-ion battery. By setting certain control policy limits, to minimize cycle aging, we show that a system with SOC limits in a 65 to 15 percent range, extends the project life to over 16 years, assuming a battery reaches its end-of-life at 60 percent of its original capacity. Under these conditions, a second-life project is more economically favorable than a project that uses a new battery and 85 to 20 percent SOC limits, for second-life battery costs that are less than 80 percent of the new battery. The same system reaches break-even and profitability for second-life battery costs that are less than 60 percent of the new battery. Our model shows that using current benchmarked data for the capital and O&M costs of solar-plus-storage systems, and a semi-empirical data-based degradation model, it is possible for EV manufacturers to sell second-life batteries for less than 60 percent of their original price to developers of profitable solar-plus-storage projects.

0 citations0 reviewsNo code linkedOpen access