Search

The photovoltaics (PV) technology landscape is evolving rapidly. To predict the potential and scalability of emerging PV technologies, a global understanding of these systems' performance is essential. Traditionally, experimental and computational studies at large national research facilities have focused on PV performance in specific regional climates. However, synthesizing these regional studies to understand the worldwide performance potential has proven difficult. Given the expense of obtaining experimental data, the challenge of coordinating experiments at national labs across a politically-divided world, and the data-privacy concerns of large commercial operators, however, a fundamentally different, data-efficient approach is desired. Here, we present a physics-guided machine learning (PGML) scheme to demonstrate that: (a) The world can be divided into a few PV-specific climate zones, called PVZones, illustrating that the relevant meteorological conditions are shared across continents; (b) by exploiting the climatic similarities, high-quality monthly energy yield data from as few as five locations can accurately predict yearly energy yield potential with high spatial resolution and a root mean square error of less than 8 kWhm$^{2}$, and (c) even with noisy, heterogeneous public PV performance data, the global energy yield can be predicted with less than 6% relative error compared to physics-based simulations provided that the dataset is representative. This PGML scheme is agnostic to PV technology and farm topology, making it adaptable to new PV technologies or farm configurations. The results encourage physics-guided, data-driven collaboration among national policymakers and research organizations to build efficient decision support systems for accelerated PV qualification and deployment across the world.

A fundamental challenge in physics-informed machine learning (PIML) is the design of robust PIML methods for out-of-distribution (OOD) forecasting tasks. These OOD tasks require learning-to-learn from observations of the same (ODE) dynamical system with different unknown ODE parameters, and demand accurate forecasts even under out-of-support initial conditions and out-of-support ODE parameters. In this work we propose a solution for such tasks, which we define as a meta-learning procedure for causal structure discovery (including invariant risk minimization). Using three different OOD tasks, we empirically observe that the proposed approach significantly outperforms existing state-of-the-art PIML and deep learning methods.

Agrivoltaics (AV) is a dual land-use approach to collocate solar energy generation with agriculture for preserving the terrestrial ecosystem and enabling food-energy-water synergies. Here, we present a systematic approach to model the economic performance of AV relative to standalone ground-mounted PV (GMPV) and explore how the module design configuration can affect the dual food-energy economic performance. A remarkably simple criterion for economic feasibility is quantified that relates the land preservation cost to dual food-energy profit. We explore case studies including both high and low value crops under fixed tilt bifacial modules oriented either along the conventional North/South (N/S) facings or vertical East/West (E/W) facings. For each module configuration, the array density is varied to explore an economically feasible design space relative to GMPV for a range of module to land cost ratio (M_L) - a location-specific indicator relating the module technology (hardware and installation) costs to the soft (land acquisition, tax, overheads, etc.) costs. To offset a typically higher AV module cost needed to preserve the cropland, both E/W and N/S orientated modules favor high value crops, reduced (<60%) module density, and higher M_L (>25). In contrast, higher module density and an increased feed-in-tariff (FIT) relative to GMPV are desirable at lower M_L. The economic trends vary sharply for M_L< 10 but tend to saturate for M_L> 20. For low value crops, ~15% additional FIT can enable economic equivalence to GMPV at standard module density. The proposed modeling framework can provide a valuable tool for AV stakeholders to assess, predict, and optimize the techno-economic design for AV, Comment: Submitted to IEEE Journal of Photovoltaics

Agrivoltaics (AV) is an emerging technology having symbiotic benefits for food-energy-water needs of the growing world population and an inherent resilience against climate vulnerabilities. An agrivoltaic system must optimize sunlight-sharing between the solar panels and crops to maximize the food-energy yields, subject to appropriate constraints. Given the emerging diversity of monofacial and bifacial farms, the lack of a standardized crop-specific metric (to evaluate the efficacy of the irradiance sharing) has made it difficult to optimize and assess the performance of agrivoltaic systems. Here we introduce a new metric, light productivity factor (LPF), that evaluates the effectiveness of irradiance sharing for a given crop type and PV array design. The metric allows us to identify optimal design parameters including the spatial PV array density, panel orientation, and single axis tracking schemes specific to the PAR needs of the crop. By definition, LPF equals 1 for PV-only or crop-only systems. The AV systems enhances LPF between 1 and 2 depending on the shade sensitivity of the crop, PV array configuration, and the season. While traditional fixed-tilt systems increase LPF significantly above 1, we find LPF is maximized at 2 for shade-tolerant crops with a solar farm based on single axis sun tracking scheme. Among the fixed tilt systems, East-West faced bifacial vertical solar farms is particularly promising because it produces smallest variability in the seasonal yield for shade sensitive crops, while providing LPF comparable to the standard North-South faced solar farms. Additional benefits include reduced soiling and ease of movement of large-scale combine-harvester and other farming equipment., Comment: 9 pages

The commercialization of non-volatile memories based on ferroelectric transistors (FeFETs) has remained elusive due to scaling, retention, and endurance issues. Thus, it is important to develop accurate characterization tools to quantify the scaling and reliability limits of FeFETs. In this work, we propose to exploit an analytical expression for the Memory Window (MW, i.e., the difference between the threshold voltages due to polarization switching) as a tool to: i) identify a universal scaling behavior of MW regardless of the ferroelectric material; ii) predict endurance and explain its weak dependence on writing conditions; iii) give an alternative explanation for MW being lower than theoretical limits; and, based on this, iv) devise strategies to maximize MW for a given ferroelectric thickness. According to these findings, the characterization and analysis of MW would enable the systematic comparison of next-generation FeFET based on emerging ferroelectric materials., Comment: 9 pages, 9 figures

Cyberphysical systems (CPS) are ubiquitous in our personal and professional lives, and they promise to dramatically improve micro-communities (e.g., urban farms, hospitals), macro-communities (e.g., cities and metropolises), urban structures (e.g., smart homes and cars), and living structures (e.g., human bodies, synthetic genomes). The question that we address in this article pertains to designing these CPS systems to be resilient-from-the-ground-up, and through progressive learning, resilient-by-reaction. An optimally designed system is resilient to both unique attacks and recurrent attacks, the latter with a lower overhead. Overall, the notion of resilience can be thought of in the light of three main sources of lack of resilience, as follows: exogenous factors, such as natural variations and attack scenarios; mismatch between engineered designs and exogenous factors ranging from DDoS (distributed denial-of-service) attacks or other cybersecurity nightmares, so called "black swan" events, disabling critical services of the municipal electrical grids and other connected infrastructures, data breaches, and network failures; and the fragility of engineered designs themselves encompassing bugs, human-computer interactions (HCI), and the overall complexity of real-world systems. In the paper, our focus is on design and deployment innovations that are broadly applicable across a range of CPS application areas., Comment: 36 pages, 2 figures, NSF-supported workshop on Grand Challenges in Resilience, held at Purdue, March 20-21, 2019

Agrivoltaics is an innovative approach in which solar photovoltaic (PV) energy generation is collocated with agricultural production to enable food-energy-water synergies and landscape ecological conservation. This dual-use requirement leads to unique co-optimization challenges (e.g. shading, soiling, spacing) that make module technology and farm topology choices distinctly different from traditional solar farms. Here we compare the performance of the traditional optimally-titled North/South (N/S) faced monofacial farms with a potential alternative based on vertical East/West (E/W)-faced bifacial farms. Remarkably, the vertical farm produces essentially the same energy output and photosynthetically active radiation (PAR) compared to traditional farms as long as the PV array density is reduced to half or lower relative to that for the standard ground-mounted PV farms. Our results explain the relative merits of the traditional mono facial vs. vertical bifacial farms as a function of array density, acceptable PAR-deficit, and energy production. The combined PAR/Energy yields for the vertical bifacial farm may not always be superior, it could still be an attractive choice for agrivoltaics due to its distinct advantages such as minimum land coverage, least hindrance to the farm machinery and rainfall, inherent resilience to PV soiling, easier cleaning and cost advantages due to potentially reduced elevation., Comment: IEEE Journal of Photovoltaics

Molding compounds (MCs) have been used extensively as an encapsulation material for integrated circuits, however, MCs are susceptible to moisture and charge spreading over time. The increase in dissipation factor due to the increase of parasitic electrical conductivity ({\sigma}) and the decrease in dielectric strength (E_MC^Crit) restrict their applications. Thus, a fundamental understanding of moisture transport will suggest strategies to suppress moisture diffusion and broaden their applications. In this paper, we 1) propose a generalized effective medium and solubility (GEMS) Langmuir model to quantify water uptake as a function of filler configuration and relative humidity; 2) investigate the dominant impact of reacted-water on {\sigma} through numerical simulations, mass-uptake, and DC conductivity measurements; 3) investigate electric field distribution to explain how moisture ingress reduces E_MC^Crit; and finally 4) optimize the filler configuration to lower the dissipation factor, and enhance E_MC^Crit. The GEMS-Langmuir model can be used for any application (e.g., photovoltaics, biosensors) where moisture diffusion leads to reliability challenges., Comment: 22 pages, 6 figures, Elsevier journal format

Moisture- and temperature-activated corrosion of metal fingers, mechanical stress induced delamination, and failure of solder bonds rank among the leading failure mechanisms of solar modules. The physics of moisture ingress, diffusion and reaction have been explored in detail, but the electrical implications of corrosion and delamination on specific front-surface grid geometry is not fully understood. In this paper, we show that the module efficiency loss due to corrosion, delamination, and solder bond failure (CDS) involves a complex interplay of voltage/current redistribution, reflected as a loss in photocurrent as well as decrease/increase in shunt/series resistances. Our work will redefine the interpretation of experimental J-V characteristics features due to degradation mechanisms, integrate a variety of scattered and counter-intuitive experimental results within a common theoretical framework, and inform CDS-resistant grid design for solar modules., Comment: 8 pages, 11 figures

We demonstrate memory devices based on proton doping and re-distribution in perovskite nickelates (RNiO3, {R=Sm,Nd}) that undergo filling-controlled Mott transition. Switching speeds as high as 30 ns in two-terminal devices patterned by electron-beam lithography is observed. The state switching speed reported here are 300X greater than what has been noted with proton-driven resistance switching to date. The ionic-electronic correlated oxide memory devices also exhibit multi-state non-volatile switching. The results are of relevance to use of quantum materials in emerging memory and neuromorphic computing., Comment: 4 figures

The uncertainties associated with technology- and geography-specific degradation rates make it difficult to calculate the levelized cost of energy (LCOE), and thus the economic viability of solar energy. In this regard, millions of fielded photovoltaic (PV) modules may serve as a global testbed, where we can interpret the routinely collected maximum power point (MPP) time-series data to assess the time-dependent "health" thereof. The existing characterization methods, however, cannot effectively mine/decode these datasets to identify various degradation pathways of the corresponding solar modules. In this paper, we propose a new methodology, i.e., the Suns-Vmp method, which offers a simple and powerful approach to monitoring and diagnosing time-dependent degradation of solar modules by physically mining the MPP data. The algorithm reconstructs "IV" curves by using the natural illumination- and temperature-dependent daily MPP characteristics as constraints to fit the physics-based compact model. These synthetic IV characteristics are then used to determine the time-dependent evolution of circuit parameters (e.g., series resistance) which in-turn allows one to deduce the dominant degradation mode (e.g., corrosion) for the modules. The proposed method has been applied to analyze the MPP data from a test facility at the National Renewable Energy Laboratory (NREL). Our analysis indicates that the solar modules degraded at a rate of 0.7 %/year due to discoloration and weakened solder bonds. These conclusions are independently validated by outdoor IV measurement and on-site imaging characterization. Integrated with physics-based degradation models or machine learning algorithms, the method can also serve to predict the lifetime of PV systems.

With the rapidly growing interest in bifacial photovoltaics (PV), a worldwide map of their potential performance can help assess and accelerate the global deployment of this emerging technology. However, the existing literature only highlights optimized bifacial PV for a few geographic locations or develops worldwide performance maps for very specific configurations, such as the vertical installation. It is still difficult to translate these location- and configuration-specific conclusions to a general optimized performance of this technology. In this paper, we present a global study and optimization of bifacial solar modules using a rigorous and comprehensive modeling framework. Our results demonstrate that with a low albedo of 0.25, the bifacial gain of ground-mounted bifacial modules is less than 10% worldwide. However, increasing the albedo to 0.5 and elevating modules 1 m above the ground can boost the bifacial gain to 30%. Moreover, we derive a set of empirical design rules, which optimize bifacial solar modules across the world, that provide the groundwork for rapid assessment of the location-specific performance. We find that ground-mounted, vertical, east-west-facing bifacial modules will outperform their south-north-facing, optimally tilted counterparts by up to 15% below the latitude of 30 degrees, for an albedo of 0.5. The relative energy output is the reverse of this in latitudes above 30 degrees. A detailed and systematic comparison with experimental data from Asia, Europe, and North America validates the model presented in this paper. An online simulation tool (https://nanohub.org/tools/pub) based on the model developed in this paper is also available for a user to predict and optimize bifacial modules in any arbitrary location across the globe.

Process-related and stress-induced changes in threshold voltage are major variability concerns in ultra-scaled CMOS transistors. The device designers consider this variability as an irreducible part of the design problem and use different circuit level optimization schemes to handle these variations. In this paper, we demonstrate how an increase in the negative steepness of the universal mobility relationship improves both the process-related (e.g., oxide thickness fluctuation, gate work-function fluctuation), as well as stress-induced or reliability-related (e.g., Bias Temperature Instability or BTI) parametric variation in CMOS technology. Therefore, we calibrate the universal mobility parameters to reflect the measured variation of negative steepness in uniaxially strained CMOS transistor. This allows us to study the extent of (process-related and stress-induced parametric) variation resilience in uniaxial strain technology by increasing the negative steepness of the mobility characteristics. Thus, we show that variability analysis in strained CMOS technology must consider the presence of self-compensation between mobility variation and threshold voltage variation, which leads to considerable amount of variation resilience. Finally, we use detailed circuit simulation to stress the importance of accurate mobility variation modeling in SPICE analysis and explain why the variability concerns in strained technology may be less severe than those in unstrained technology., Comment: 25 pages, 7 figures

For commercial one-sun solar modules, up to 80% of the incoming sunlight may be dissipated as heat, potentially raising the temperature 20 C - 30 C higher than the ambient. In the long term, extreme self-heating erodes efficiency and shortens lifetime, thereby dramatically reducing the total energy output. Therefore, it is critically important to develop effective and practical (and preferably passive) cooling methods to reduce operating temperature of PV modules. In this paper, we explore two fundamental (but often overlooked) origins of PV self-heating, namely, sub-bandgap absorption and imperfect thermal radiation. The analysis suggests that we redesign the optical properties of the solar module to eliminate parasitic absorption (selective-spectral cooling) and enhance thermal emission (radiative cooling). Our Comprehensive opto-electro-thermal simulation shows that the proposed techniques would cool the one-sun and low-concentrated terrestrial solar modules up to 10 C and 20 C, respectively. This self-cooling would substantially extend the lifetime for solar modules, with The corresponding increase in energy yields and reduced LCOE., Comment: Discussion of low concentration PV is added

In this paper, we present a physics-based analytical model for CIGS solar cells that describes the illumination- and temperature-dependent current-voltage (I-V) characteristics and accounts for the statistical shunt variation of each cell. The model is derived by solving the drift-diffusion transport equation so that its parameters are physical, and, therefore, can be obtained from independent characterization experiments. The model is validated against CIGS I-V characteristics as a function of temperature and illumination intensity. This physics-based model can be integrated into a large-scale simulation framework to optimize the performance of solar modules as well as predict the long-term output yields of photovoltaic farms under different environmental conditions.

Self-heating in surrounding gate transistors can degrade its on-current performance and reduce lifetime. If a transistor heats/cools with time-constants less than the inverse of the operating frequency, a predictable, frequency-independent performance is expected; if not, the signal pattern must be optimized for highest performance. Typically, time-constants are measured by expensive, ultra-fast instruments with high temporal resolution. Instead, here we demonstrate an alternate, inexpensive, cyclostationary measurement technique to characterize self-heating (and cooling) with sub-microsecond resolution. The results are independently confirmed by direct imaging of the transient heating/cooling of the channel temperature by the thermoreflectance (TR) method. A routine use of the proposed technique will help improve the surrounding gate transistor design and shorten the design cycle., Comment: 12 pages, 2 figures, Journal

Sub-threshold swing (S) defines the sharpness of ON-OFF switching of a Field Effect Transistor (FET) with S=0 corresponding to abrupt switching characteristics. While thermodynamics dictates S to be greater than or equal to 60mV/dec for classical FETs, "Landau switches" use inherently unstable gate insulators to achieve abrupt switching. Unfortunately, S=0 switching is always achieved at the expense of an intrinsic hysteresis, making these switches unsuitable for low-power applications. The fundamental question therefore remains: Under what condition, hysteresis-free abrupt switching can be achieved in a Landau switch? In this paper, we first provide an intuitive classification of all charge based switches in terms of their energy landscapes and identify two-well energy landscape as the characteristic feature of Landau switches. We then use nanoelectromechanical field effect transistor (NEMFET) as an illustrative example of a Landau switch and conclude that a flat energy landscape is essential for hysteresis-free abrupt switching. In contrast, a hysteresis-free smooth sub-60mV/dec switching is obtained by stabilizing the unstable gate insulator in its unstable regime. Our conclusions have broad implications and may considerably simplify the design of next charge based logic switch., Comment: 7 pages, 6 figures

In this review article, we focus on emerging nanocantilever based biological sensors and discuss the response of nanocantilevers towards bio-molecules capture. The article guides the reader through various modes of operation (e.g., static or dynamic) to detect the change in characteristics (e.g., mass, stiffness, and/or surface stress) of cantilever due to adsorption of bio-molecules on cantilever surface. First, we explain the classical linear resonant mode mass sensors and static stress based sensors. The effect of operating the cantilever in nonlinear regime is then illustrated through examples of bifurcation based mass sensors and electromechanical coupling based Flexure-FET biosensors. We believe that a new class of nonlinear sensors, with their extraordinary sensitivity towards bio-molecules capture, could be the potential candidate for low cost point-of-care applications., Comment: 17 pages and 7 figures

Continued scaling of nanoscale transistors leads to broad device-to-device fluctuation of parameters due to random dopant effects, channel length variation, interface trap generation, etc. In this paper, we obtain the statistics of negative bias temperature instability (NBTI)-induced interface defect generation in ultra-scaled MOSFET by Markov Chain Monte-Carlo (MCMC) solution of Reaction-Diffusion (R-D) model. Our results show that the interface defect generation at a particular stress time, i.e., NIT}@tSTS in small transistors should follow a skew-normal distribution and that the generation and annealing of interface defects are strongly correlated. Next, we use a random percolative network to demonstrate (which is also consistent with previously published results in literature based on separate techniques) that the distribution of threshold voltage shift for single interface defect, i.e., {\Delta}VT@NIT is exponential, with finite number of transistors having zero {\Delta}VT. Finally, we show that the statistics of {\Delta}VT@tSTS - based on the convolution of NIT@tSTS and {\Delta}VT@NIT - is broadly consistent with the available experimental data in literature., Comment: This paper has been withdrawn by the author. A similar paper is already published in Journal of Computational Electronics with the following link: http://www.springerlink.com/content/y0p362uhh3gm0u12/fulltext.pdf