Physics-based Analysis Research Articles

Introduction to this special section: Data analytics and machine learning

The explosion of data types and volumes we are experiencing in the oil field, particularly in unconventional reservoir development, is stretching the capabilities of traditional manual workflows. The complex interactions between dynamic reservoir properties and the many-faceted well completions process are governed by complex physics that may be only partially understood. Furthermore, our best attempts to perform controlled experiments can often produce highly varied results, suggesting the stochastic nature of production from unconventionals and that we may benefit from adding new tools to our toolkit of deterministic, physics-based analysis.

Large-signal variability of microwave power amplifiers through efficient device sensitivity-based physical modeling

In the present article they present a comprehensive variability analysis of microwave power amplifiers in various operating classes by means of a recently developed numerically efficient technique for the physics-based variability analysis of devices operated in nonlinear conditions. Both a class A and a deep class AB power amplifier are considered, with the active device undergoing variations of main physical parameters such as doping, gate work function, and gate length. The results presented show for the first time that the physics-based variability analysis of nonlinear circuits is feasible and allows for a direct link between the PA performance uncertainty and the active device technological parameter variations. Both deterministic and statistical variations are taken into account, showing that the approach of Donati Guerrieri et al. (2016) allows for an aggressive reduction in computation time with respect to the variability analysis by means of conventional approaches, while retaining very good accuracy for parameter variations up to 10% of their nominal value. The highest sensitivity of the class A stage is found for the drain efficiency, while the class AB stage has a more involved behavior, with the highest sensitivity shifting from the drain efficiency to the output power as the input power increases.

Enhanced Surface Integrity From Cryogenic Machining of AZ31B Mg Alloy: A Physics-Based Analysis With Microstructure Prediction

Enhanced Surface Integrity From Cryogenic Machining of AZ31B Mg Alloy: A Physics-Based Analysis With Microstructure Prediction

An Improved Fault-to-Site Analysis Tool Towards Fully HPC-Enhanced Physics-Based Urban Area Response Estimation

This paper describes the improvement to a fault-to-site analysis tool for use in a fully HPC-enhanced physics-based urban earthquake response analysis. We developed a three-step procedure, involving coarse and refined partitioning for the preprocessing stage, that achieves good balance in computational load and FE nodes used in communications. An end-to-end approach to computing in K computer was also implemented. The results of strong- and weak-scaling tests validate the capability of the tool for application to dynamic analysis involving billions of unknowns. In application to reproduction of observed ground motions, improvements in solution in comparison with previously conducted validations were obtained. The tool was also used to generate synthetic ground motions for analyzing the response of thousands of buildings in an urban area due to a scenario earthquake.

Effect of explosive cratering on embankment dams

The efficient management and protection of critical infrastructure such as dams is relevant to safety of the population of low-lying downstream communities. Ramifications of dam breach on these communities would be severe and include loss of electricity, flooding, injury and fatality. The economic losses can be enormous depending on the size of the dam and level of emergency preparedness of the community. As acts of terror continue to increase globally, dams appear to be attractive targets due to the capacity to adversely impact downstream communities and to attract media attention. Embankment dams are more susceptible to cratering and, in turn, overtopping and erosion of the dam crest. This paper investigates the effects of explosions on embankment dams by using a high-fidelity physics-based numerical analysis software program – LS-DYNA. The impacts of explosive size, reservoir level and soil density were examined while measuring the dimensions of the resulting crater. The investigation showed that an increase in explosive size led to an increase in crater diameters as well as the depth. In considering the reservoir level, the reservoir was considered at maximum capacity, at mid-capacity and embankment without a reservoir. The results showed that the crater dimensions were larger at maximum capacity and lowest when there was no water in reservoir. At a given soil density, when the dam fill materials are saturated, the crater dimensions are larger in terms of diameters and depths as compared to when fill materials are not saturated.

Optimized Standard Cells for All-Spin Logic

All-Spin Logic (ASL) devices provide a promising spintronics-based alternative for Boolean logic implementations in the post-Complementary Metal-Oxide Semiconductor (CMOS) era. In principle, any logic functionality can be implemented in ASL. In practice, the performance of an ASL gate is significantly affected by layout choices, but such implications have not been adequately explored in the past. This article proposes a systematic approach for building standard cells in ASL, which are a basic building block in an overall design methodology for implementing large ASL-based circuits. We first propose a new technique to reduce the magnet count for an ASL majority gate but still ensure correct functioning through layout optimization methods. Building on physics-based analysis, we then build a standard cell library with diverse functionality and characterize the library for delay, energy, and area. We perform delay-optimized technology mapping on ISCAS85 benchmark circuits using our library. Our approach results in circuits that are 12.90% faster, consume 26.16% less energy, and are 33.56% more area efficient compared to a standard cell library that does not incorporate layout-based optimization techniques of our work.

Phases and Interfaces from Real Space Atomically Resolved Data: Physics-Based Deep Data Image Analysis.

Advances in electron and scanning probe microscopies have led to a wealth of atomically resolved structural and electronic data, often with ∼1-10 pm precision. However, knowledge generation from such data requires the development of a physics-based robust framework to link the observed structures to macroscopic chemical and physical descriptors, including single phase regions, order parameter fields, interfaces, and structural and topological defects. Here, we develop an approach based on a synergy of sliding window Fourier transform to capture the local analog of traditional structure factors combined with blind linear unmixing of the resultant 4D data set. This deep data analysis is ideally matched to the underlying physics of the problem and allows reconstruction of the a priori unknown structure factors of individual components and their spatial localization. We demonstrate the principles of this approach using a synthetic data set and further apply it for extracting chemical and physically relevant information from electron and scanning tunneling microscopy data. This method promises to dramatically speed up crystallographic analysis in atomically resolved data, paving the road toward automatic local structure-property determinations in crystalline and quasi-ordered systems, as well as systems with competing structural and electronic order parameters.

Reducing Detail Design and Construction Work Content by Cost-Effective Decisions in Early-Stage Naval Ship Design

The Office of the Under Secretary of Defense, Acquisition, Technology, and Logistics (AT&L) recently presented analyses of cost and schedule growth on Major Defense Acquisition Programs (MDAPs) over the last 20 years (2013, 2014). For naval ships, AT&L (2013) concluded that contract work content growth (not capability growth) dominates total cost growth statistically. In addition, costs-over-target are significant and reflect poor cost estimation or faulty framing assumptions. AT&L (2014) also concluded prices on fixed-price contracts are only “fixed” if the contractual work content remains fixed, but this is often not the case. The authors show that under-sizing the ship during concept design studies increases ship outfit density and adds complexities to the design. These early-stage design decisions on sizing the ship are a major contributor to unnecessary work content growth later in Detail Design and Construction (DD&C) that cannot be eliminated no matter how productive the shipbuilder. However, new ship design methods are being developed and integrated with legacy physics-based design and analysis tools into a Rapid Ship Design Environment (RSDE) that will enable a more rational process for initially sizing ships. The authors also identify the need for early-stage design measures of complexity and ship costing tools that are more sensitive to these measures, and propose solutions that will aid decision-makers in reducing DD&C work content by making cost-effective design decisions in early-stage naval ship design.

Reducing Detail Design and Construction Work Content by Cost-Effective Decisions in Early-Stage Naval Ship Design

The Office of the Under Secretary of Defense, Acquisition, Technology, and Logistics (AT&L) recently presented analyses of cost and schedule growth on Major Defense Acquisition Programs (MDAPs) over the last 20 years (2013, 2014). For naval ships, AT&L (2013) concluded that contract work content growth (not capability growth) dominates total cost growth statistically. In addition, costs-over-target are significant and reflect poor cost estimation or faulty framing assumptions. AT&L (2014) also concluded prices on fixed-price contracts are only "fixed" if the contractual work content remains fixed, but this is often not the case. We show that under-sizing the ship during concept design studies increases ship outfit density and adds complexities to the design. These early-stage design decisions on sizing the ship are a major contributor to unnecessary work content growth later in Detail Design and Construction (DD&C) that cannot be eliminated no matter how productive the shipbuilder. However, new ship design methods are being developed and integrated with legacy physicsbased design and analysis tools into a Rapid Ship Design Environment (RSDE) that will enable a more rational process for initially sizing ships. We also identify the need for early-stage design measures of complexity and ship costing tools that are more sensitive to these measures, and propose solutions that will aid decision-makers in reducing DD&C work content by making cost-effective design decisions in early-stage naval ship design.

Microstructural characterization and recrystallization kinetics modeling of annealing cold-rolled vanadium microalloyed HSLA steels

In this study, we investigated the microstructural evolution and recrystallization kinetics during the annealing of cold-rolled Vanadium microalloyed HSLA (denoted as HSLA-V) steels using both experimental characterization techniques and theoretical model analysis method. Our experimental results indicate that the recrystallization kinetics can be quantified with the evolution of volume fraction of recrystallization as a function of annealing time, and is strongly dependent on the processing parameters (i.e., cold-rolling reduction and annealing temperature) of the HSLA-V steels. Validated by experimental data, our physics-based model analysis revealed that the recrystallization process in the HSLA-V steels consisted of an early stage with both nucleation and grain growth processes and a later stage with grain growth only. Moreover, the growth of the recrystallized grains was found to follow one dimensional preferential growth mechanism, owing to the dispersed V(C,N) precipitates, in the HSLA-V steels. Consequently, our new analysis model is demonstrated to be useful to gain insights into the recrystallization kinetics in the annealed cold-rolled HSLA-V and even other microalloyed steels.

A Universal Hardware-Driven PVT and Layout-Aware Predictive Failure Analytics for SRAM

The impact of device variability, temperature, and technology CAD-based layout parasitics on low-voltage static random access memory (SRAM) yield is explored using a novel variability-aware statistical methodology. Threshold voltage, $V_{t}$ , mismatches for planar 22- and 14-nm FinFET SRAM transistors are characterized based on unique array-like structures for capturing process voltage and temperature (PVT) impact on variability. In general, the mismatches are shown to be a consistent and unique function of $V_{\rm dd}$ , doping, and temperature across the two technologies. Stronger $V_{t}$ mismatch impact is observed as a function of $V_{\rm dd}$ and doping in the 22-nm technology, with higher mismatch recorded at lower temperatures. In the 14-nm technology, doping is found to have the strongest impact on $V_{t}$ mismatch, and the mismatch increases with $V_{\rm dd}$ despite the reduced drain-induced barrier lowering effects. Similar to the 22-nm technology, the mismatch increases at lower temperatures. Front-end-of-the-line capacitance effects are found to be more significant than back-end-of-the-line effects in 14-nm technologies, as opposed to planar technologies. Accurate parasitic capacitance modeling along with PVT-aware variability process variations for different 22-/14-nm cell arrangements are incorporated into a physics-based statistical analysis methodology for accurate $V_{\rm min}$ analysis. The yield analysis results are corroborated with hardware yield using 4–16-Mb inline SRAM macro monitors. The methodology is unique in the industry, gives insight into the technology-circuit interactions, and is able to effectively predict the SRAM yield bounds.

A Unified Approach to the Sensitivity and Variability Physics-Based Modeling of Semiconductor Devices Operated in Dynamic Conditions—Part I: Large-Signal Sensitivity

We present here and in the companion paper (Part II) a general framework for the modeling of semiconductor device variability through the physics-based analysis of the small-change sensitivity. We consider a very general class of dynamic device operation, i.e., the periodic or quasi-periodic large signal (LS) time-varying regime, and we evaluate the sensitivity of both dc and harmonic components of the device dynamic working point with respect to process or physical device parameters. The proposed technique is based on the linearization of the physics-based device model around a nominal parameter, and extends to the dynamic case the already established Green’s function approach to the numerically efficient dc sensitivity analysis. As an example of application, we consider a class A GaAs MESFET microwave power amplifier; the sensitivity of the LS working point with respect to doping and gate work function variations is evaluated through the proposed approach and compared with the result of repeated LS amplifier analyses, showing that the numerically efficient small-change sensitivity approach provides reliable predictions for parameter variations up to 10% of the nominal value.

A Unified Approach to the Sensitivity and Variability Physics-Based Modeling of Semiconductor Devices Operated in Dynamic Conditions. Part II—Small-Signal and Conversion Matrix Sensitivity

We present here and in Part I a general framework for the modeling of semiconductor device variability through the physics-based analysis of the small-change sensitivity. While Part I focuses on the sensitivity of the device terminal currents in periodic large-signal (LS) operation, we extend here the analysis to the linearized device representations, i.e., to the sensitivity of the small-signal (SS) admittance matrix (SS sensitivity) and the conversion admittance matrix (SS-LS sensitivity). The proposed technique is based on the linearization of a physical device model around a nominal process parameter, and on the evaluation of relevant Green’s functions linking the parameter variations to the terminal performance. This provides, for the first time, a unified and computationally efficient simulation framework for the device physics-based sensitivity in dc, SS, LS, and SS-LS conditions. To highlight the accuracy of the approach when compared with the incremental evaluation, we discuss two case studies, concerning the SS-LS sensitivity of a class A GaAs MESFET-based amplifier and the SS sensitivity of an AlGaN/GaN microwave HEMT.

Design of CdZnTe and Crystalline Silicon Tandem Junction Solar Cells

A tandem photovoltaic system consisting of cadmium zinc telluride/crystalline silicon (CZT/c-Si) combines the successful technologies of cadmium telluride and silicon into a single platform and offers the potential efficiencies up to 46% in theory. However, the highest efficiency fabricated CZT/c-Si tandem cell is only 16.8% today. In this paper, we develop a detailed model for single-junction CZT and tandem CZT/c-Si PV cells, which is verified with experimental data. Based on this model, we propose three hypotheses to explain the anomalously low $V_{\rm oc}$ observed in tandem cells: a low-quality tunnel junction, a Schottky barrier, and through-thickness shunting path. We then suggest a simple experiment to distinguish these hypotheses. After that, we provide a physics-based analysis of the magnitude of all the loss mechanisms present in the cell and an experimental strategy to mitigate each one. Ultimately, we predict that the ideal efficiency of CZT/c-Si tandem cell could reach 34.1%, if all these loss mechanisms were mitigated, and the CZT bandgap were adjusted to 1.8 eV, without requiring any improvement in bulk or surface recombination rates.

Numerical and experimental study of circular disc electromechanical impedance spectroscopy signature changes due to structural damage and sensor degradation

The article describes a study on the use of the electromechanical impedance spectroscopy method for crack-damage detection with piezoelectric wafer active sensors bonded to thin-walled structures. The study combines analytical, finite element method, and experimental methods. The specimens under consideration consisted of circular plates with circular piezoelectric wafer active sensors bonded at the center. Some of the plates were pristine, and others were damaged. The damage consists of simulated cracks (thin laser-machined slits) placed at various locations away from the center. For pristine plates, the electromechanical impedance spectroscopy spectrum was predicted with two numerical methods: (a) analytically using closed-form solutions and (b) Comsol 4.3 Multi-physics software using two-dimensional axisymmetric elements. The results were compared with experimental measurements performed with an HP 4194 impedance analyzer. This comparison allowed us to tune the model parameters. For plates with simulated cracks, the electromechanical impedance spectroscopy spectrum was predicted with three-dimensional finite element method models and then compared with experiments. Overall, the effect of crack damage on electromechanical impedance spectroscopy signature was found to consist of frequency shifts, peak splitting, and appearance of new peaks. These effects were predicted by the numerical models and confirmed by experiments. Modeshapes were also measured using a scanning laser Doppler vibrometer and compared with finite element method predictions. Interesting new modeshape features (local high-frequency modes) were observed near the crack. These features tend to explain some of the new peaks appearing in the electromechanical impedance spectroscopy spectrum. Further refining of the finite element method model was performed to capture variability effects associated with plate thickness, piezoelectric wafer active sensors thickness, adhesive layer thickness, and imperfect adhesive layer. The effect on the piezoelectric wafer active sensors asymmetrically placed electrodes and soldering points was also studied. It was found that these variations modify the electromechanical impedance spectroscopy signatures, but much less than the changes induced by the presence of the crack. Hence, it was concluded that the capability of the electromechanical impedance spectroscopy method to detect actual damage was not affected by these variations. A proper physics-based analysis and interpretation of the electromechanical impedance spectroscopy spectral changes are performed.

A study of keyhole geometry in laser welding of zinc-coated and uncoated steels using a coaxial observation method

Experiments were conducted over a large process parameter space using a 2kW multi-mode fiber laser, and the effect of zinc-coating on the keyhole geometry was investigated using a high-speed coaxial video camera. From the top and bottom coaxial surface images of erratic keyhole motions that were obtained from separately conducted experiments, time-averaged keyhole structures were calculated using a physics-based semi-statistical frame analysis. For uncoated steel, the keyhole bottom aperture is located mostly within the laser beam area. In this case, because there is no zinc evaporation, the bottom aperture tends to be closed to absorb enough laser energy for sustaining a keyhole. For zinc-coated steel, the keyhole is mostly open at the bottom, and the front keyhole wall is tilted so that the laser beam is located on the front keyhole wall. The keyhole tilting angle is more important for zinc-coated steel and the beam interaction area is more relevant to uncoated steel.

Physics-based stability analysis of MOS transistors

In this work, a physics-based model is derived based on a linearization procedure for investigating the electrical, thermal and electro-thermal instability of power metal–oxide–semiconductor (MOS) transistors. The proposed model can be easily interfaced with a circuit or device simulator to perform a failure analysis, making it particularly useful for power transistors. Furthermore, it allows mapping the failure points on a three-dimensional (3D) space defined by the gate-width normalized drain current, drain voltage and junction temperature. This leads to the definition of the Safe Operating Volume (SOV), a powerful frame work for making failure predictions and determining the main root of instability (electrical, thermal or electro-thermal) in different bias and operating conditions. A comparison between the modeled and the measured SOV of silicon-on-insulator (SOI) LDMOS transistors is reported to support the validity of the proposed stability analysis.

(Invited) Thermal Limitations in Wide Bandgap (WBG) Semiconductor Power Switching Devices

Power switching devices made using wide bandgap (WBG) semiconductors such as Silicon Carbide (SiC) and Gallium Nitride (GaN) have the potential to make transformative impact on electrical energy generation, transmission, distribution and conditioning [1]. Although SiC power diodes and power MOSFETs are commercially available for voltage ratings up to 1,700 volts, a careful review of the published data suggests that reliability of WBG power devices may be severely compromised compared to identically rated silicon power devices. For example, dv/dt, avalanche, and safe-operating area (SOA) of SiC power diodes and MOSFETs are inferior to silicon power devices with identical ratings. This paper will present a simple physics-based thermal analysis of the power handling capability of WBG power devices and explain the possible causes of lower reliability ratings. [1] K. Shenai et al, “Current status and emerging trends in wide bandgap (WBG) semiconductor power devices,” ECS J. Solid State Sci. and Tech. 2(8), N3055-N3063, Jul. 2013.

(Invited) Thermal Limitations in Wide Bandgap (WBG) Semiconductor Power Switching Devices

Power switching devices made using wide bandgap (WBG) semiconductors such as Silicon Carbide (SiC) and Gallium Nitride (GaN) have the potential to make transformative impact on electrical energy generation, transmission, distribution and conditioning. Although SiC power diodes and power MOSFETs are commercially available for voltage ratings up to 1,700 volts, a careful review of the published data suggests that reliability of WBG power devices may be severely compromised compared to identically rated silicon power devices. For example, dv/dt, avalanche, and safe-operating area (SOA) of SiC power diodes and MOSFETs are inferior to silicon power devices with identical ratings. This paper will present a simple physics-based thermal analysis of the power handling capability of WBG power devices and explain the possible causes of lower reliability ratings.

Microstructure Effects on Cutting Forces and Flow Stress in Ultra-Precision Machining of Polycrystalline Brittle Materials

This paper presents a physics-based analysis to quantitatively describe the effects of grain size, grain boundaries, and crystallographic orientation on the flow stress of the polycrystalline material and thereby on the cutting and thrust forces. The model has been experimentally validated, in terms of the force intensities and sensitivities to microstructure attributes such as the grain size and the misorientation by comparing the forces to measured data in micromachining of polycrystalline silicon carbide (p-SiC). Molecular dynamics (MD) simulations are performed to explore the effects of grain boundaries and misorientation and to validate the modeling analysis in the context of resulting force ratios.