Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “POLYNOMIALS”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 973 records · Page 54

Kernel principal component analysis for stochastic input model generation

Highlights: {yields} KPCA is used to construct a reduced order stochastic model of permeability. {yields} A new approach is proposed to solve the pre-image problem in KPCA. {yields} Polynomial chaos is used to provide a parametric stochastic input model. {yields} Flow in porous media with channelized permeability is considered. - Abstract: Stochastic analysis of random heterogeneous media provides useful information only if realistic input models of the material property variations are used. These input models are often constructed from a set of experimental samples of the underlying random field. To this end, the Karhunen-Loeve (K-L) expansion, also known as principal component analysis (PCA), is the most popular model reduction method due to its uniform mean-square convergence. However, it only projects the samples onto an optimal linear subspace, which results in an unreasonable representation of the original data if they are non-linearly related to each other. In other words, it only preserves the first-order (mean) and second-order statistics (covariance) of a random field, which is insufficient for reproducing complex structures. This paper applies kernel principal component analysis (KPCA) to construct a reduced-order stochastic input model for the material property variation in heterogeneous media. KPCA can be considered as a nonlinear version of PCA. Through use of kernel functions, KPCA further enables the preservation of higher-order statistics of the random field, instead of just two-point statistics as in the standard Karhunen-Loeve (K-L) expansion. Thus, this method can model non-Gaussian, non-stationary random fields. In this work, we also propose a new approach to solve the pre-image problem involved in KPCA. In addition, polynomial chaos (PC) expansion is used to represent the random coefficients in KPCA which provides a parametric stochastic input model. Thus, realizations, which are statistically consistent with the experimental data, can be generated in an efficient way. We showcase the methodology by constructing a low-dimensional stochastic input model to represent channelized permeability in porous media.

97 MATHEMATICS AND COMPUTING↗

Uncertainties propagation in the framework of a Rod Ejection Accident modeling based on a multi-physics approach

The control of uncertainties in the field of reactor physics and their propagation in best-estimate modeling are a major issue in safety analysis. In this framework, the CEA develops a methodology to perform multi-physics simulations including uncertainties analysis. The present paper aims to present and apply this methodology for the analysis of an accidental situation such as REA (Rod Ejection Accident). This accident is characterized by a strong interaction between the different areas of the reactor physics (neutronic, fuel thermal and thermal hydraulic). The modeling is performed with CRONOS2 code. The uncertainties analysis has been conducted with the URANIE platform developed by the CEA: For each identified response from the modeling (output) and considering a set of key parameters with their uncertainties (input), a surrogate model in the form of a neural network has been produced. The set of neural networks is then used to carry out a sensitivity analysis which consists on a global variance analysis with the determination of the Sobol indices for all responses. The sensitivity indices are obtained for the input parameters by an approach based on the use of polynomial chaos. The present exercise helped to develop a methodological flow scheme, to consolidate the use of URANIE tool in the framework of parallel calculations. Finally, the use of polynomial chaos allowed computing high order sensitivity indices and thus highlighting and classifying the influence of identified uncertainties on each response of the analysis (single and interaction effects). (authors)

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Radiation Transport in Random Media With Large Fluctuations

Strongly heterogeneous media arise in several applications that include radiation shields, nuclear fuel, BWR moderators, clouds, planetary and stellar atmospheres, turbulent gases and plasmas. Neutral and charged particle transport computations in such media have relied heavily on formulating transport equations with spatially random coefficients (physical data) and developing solution methods to deal with the additional stochastic dimensions. Attempts at developing approximate closures that yield only low order statistical information (e.g., mean and variance of the flux) have proved to be highly restrictive under real physics conditions or rely on techniques that require fluctuation amplitudes to be small for robustness. Recently, stochastic spectral methods such as polynomial chaos and stochastic collocation have been developed for aleatoric uncertainty quantification and sensitivity analysis, and successfully applied in radiation transport work. Advances in UQ techniques have tended to focus on efficiently handling large numbers of uncertain variables but the rigorous stochastic basis of the approach also promotes its use in situations where stochasticity is due to spatial heterogeneity and the associated uncertainty is large. Here we apply these techniques to radiation transport in media with spatially randomly varying cross sections without restriction on fluctuation amplitudes. Specifically, we represent the cross sections as a lognormal spatial random process with specified mean, variance and covariance function and use a Karhunen-Loeve (KL) decomposition to generate cross section realizations that are strictly positive. Woodcock Monte Carlo (WMC) is then used to simulate transport using random sampling of cross sections and deterministic sampling based on a stochastic collocation technique. Numerical results for the mean and variance in the scalar flux and leakage currents are obtained for weak and strong random variations. We focus in this work on demonstrating the approach and defer application to specific problems to a future investigation. We have demonstrated that a combination of stochastic spectral representation and Woodcock Monte Carlo simulation allows efficient computation of radiation transport in strongly random media. The use of a lognormal distribution to represent the fluctuations in medium properties facilitates enforcement of strict positivity of cross section realizations, a condition that has challenged standard approaches that assume Gaussian fluctuations. The fact that the lognormal random process is a memoryless nonlinear transformation of a Gaussian process further enables efficient and accurate reconstruction of the random cross section realizations using a Karhunen-Loeve representation of the cross section. Finally, the robustness of this model enables fluctuations of arbitrarily large amplitude to be studied, using both random sampling Monte Carlo and more efficient stochastic collocation techniques. The present work is being extended to allow reconstruction of probability density functions of output variables using polynomial chaos representation, and to accommodate general cross section covariance functions. Also, techniques are being explored to more judiciously select ceiling cross sections to enhance the efficiency of Woodcock Monte Carlo for transport in random media. (authors)

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Diffusion Synthetic Acceleration for High Order S{sub N} Transport on Meshes with Curved Surfaces

Thermal radiation transport is integral to multiphysics applications such as inertial confinement fusion and astrophysics. Materials in these problems can be exceptionally optically thick and diffusive. A transport solver must be capable of accurate and efficient solutions to such problems. Previously, Woods et al. established a proof-of-concept for high order (HO) discontinuous Galerkin finite element (DGFEM) transport on meshes with curved surfaces. They numerically demonstrated the spatial convergence is order p + 1. Remarkably, these convergence rates are relatively invariant to the amount of curvature of the spatial mesh. We present here an extension of this work that demonstrates a rapidly-convergent iterative solver for the FEM transport equations on meshes with curved sides. We use the modified interior penalty (MIP) diffusion synthetic acceleration (DSA) established by Wang and Ragusa. The partially consistent MIP equations stem from the symmetric interior penalty (IP) derivation of the diffusion equation. The IP method is not stable for optically thick cells so it was combined with the diffusion conforming form (DCF), a discretized diffusion equation derived from the discretized S N equations that is not stable for intermediate optical thicknesses. The MIP equations adapt a 'switch' to the IP method in optically thinner regions and the DCF method in optically thicker regions. The literature does not reveal any Fourier analyses performed for iterative transport on meshes with curved surfaces. Wang and Ragusa performed a Fourier analysis for orthogonal meshes and our results appear to be as expected. Previous literature has reported on some of the sensitivities that affect the spectral radius of DSA schemes: polynomial order, and scattering ratios. Presently, we focus on the sensitivity of the spectral radius to polynomial order, scattering ratio, the constant C, and mesh curvature for varying cell thicknesses. We also numerically study the behavior of the spectral radius to verify our DSA implementation. A sequence of problems is designed to exercise the acceleration. As the problems become more optically thick and diffusive, the accelerated solution continues to converge toward the analytic solution. We employ the Modular Finite Element Methods (MFEM) package, developed at Lawrence Livermore National Laboratory, to generate the linear algebraic system of equations for each angle in the discrete ordinates quadrature set, and use UMFPack to solve this system of equations.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Direct S{sub N} Treatment of the Scattering Source with Phase Functions

In deterministic transport, the angular dependency of the scattering source is described using a so-called phase function. The scattering operator is commonly expressed using a spherical harmonics (P{sub N}) expansion of this phase function. Not only does this require minimal modifications if the problem dimensionality is changed, but it also contains little to no approximation whenever the scattering anisotropy degree is small. An alternative method relying on a discrete ordinates (S{sub N}) approach - essentially computing the angular integral of the phase function with a finite number of weighted directions - is theoretically possible and has shown some advantages. In particular, this method was introduced for neutron elastic scattering and was shown to be more accurate for highly anisotropic group-to-group transfer cross-sections. Benefits were further shown in the case of a collimated beam in highly forward peaked scattering media, particularly reducing the memory usage by storing an angle-to-angle collision matrix rather than a large number of angular moments. In addition, the scattering source was expressed using that same method, the phase function having been previously expanded using Legendre polynomials, in the context of achieving extreme precision while solving the 1D radiative transfer equation. Based on these previous works, we aim at exploring two questions. First, we wish to derive an expression of the phase function in reduced geometries that does not require any more angular quadrature points than for a standard S N calculation. This is particularly relevant because the reduced phase function does not solely depend on the product of the incoming and outgoing directions. Second, we wish to better understand the relationship between the S{sub N} and P{sub N} treatments. We in particular show equivalence if the phase function in the former case is expanded Legendre polynomials. This equivalence is verified numerically and the computational costs of the methods are compared on a test problem involving a Dirac delta phase function.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

On complexity of the mixed volume of parallelograms

Let K = (K{sub 1}...K{sub n}) be a n-tuple of convex compact subsets in the Euclidean space R{sup n}, and let V({center_dot}) be the Euclidean volume in R{sup n}. It is well known Herman Minkowski result (see for instance 5), that the value of the V{sub K}({gamma}{sub 1}K{sub 1} + ... {gamma}{sub n}K{sub n}) is a homogeneous polynomial of degree n, called the Minkowski polynomial, in nonnegative variables {gamma}{sub 1}...{gamma}{sub n1}, where '+' denotes Minkowski sum, and {gamma}K denotes the dilatation of K with coefficient {gamma}. The coefficient V(K{sub 1}...K{sub n}) of {gamma}{sub 1}{center_dot}{gamma}{sub 2}...{center_dot}{gamma}{sub n} is called the mixed volume of K{sub 1}...K{sub n}. Alternatively, V(K{sub 1}...K{sub n}) = ({partial_derivative}{sup n} / {partial_derivative}{gamma}{sub 1}...{partial_derivative}{gamma}{sub n})V{sub K}({gamma}{sub 1}K{sub 1}+...{gamma}{sub n}K{sub n}).

97 MATHEMATICS AND COMPUTING↗

Stochastic methods for uncertainty quantification in radiation transport

The use of generalized polynomial chaos (gPC) expansions is investigated for uncertainty quantification in radiation transport. The gPC represents second-order random processes in terms of an expansion of orthogonal polynomials of random variables and is used to represent the uncertain input(s) and unknown(s). We assume a single uncertain input-the total macroscopic cross section-although this does not represent a limitation of the approaches considered here. Two solution methods are examined: The Stochastic Finite Element Method (SFEM) and the Stochastic Collocation Method (SCM). The SFEM entails taking Galerkin projections onto the orthogonal basis, which, for fixed source problems, yields a linear system of fully -coupled equations for the PC coefficients of the unknown. For k-eigenvalue calculations, the SFEM system is non-linear and a Newton-Krylov method is employed to solve it. The SCM utilizes a suitable quadrature rule to compute the moments or PC coefficients of the unknown(s), thus the SCM solution involves a series of independent deterministic transport solutions. The accuracy and efficiency of the two methods are compared and contrasted. The PC coefficients are used to compute the moments and probability density functions of the unknown(s), which are shown to be accurate by comparing with Monte Carlo results. Our work demonstrates that stochastic spectral expansions are a viable alternative to sampling-based uncertainty quantification techniques since both provide a complete characterization of the distribution of the flux and the k-eigenvalue. Furthermore, it is demonstrated that, unlike perturbation methods, SFEM and SCM can handle large parameter uncertainty.

97 MATHEMATICS AND COMPUTING↗

A proof of the log-concavity conjecture related to the computation of the ergodic capacity of MIMO channels

An upper bound on the ergodic capacity of MIMO channels was introduced recently in [1]. This upper bound amounts to the maximization on the simplex of some multilinear polynomial p({lambda}{sub 1}, ..., {lambda}{sub n}) with non-negative coefficients. In general, such maximizations problems are NP-HARD. But if say, the functional log(p) is concave on the simplex and can be efficiently evaluated, then the maximization can also be done efficiently. Such log-concavity was conjectured in [1]. We give in this paper self-contained proof of the conjecture, based on the theory of H-Stable polynomials.

42 ENGINEERING↗

Bounds on the sample complexity for private learning and private data release

Learning is a task that generalizes many of the analyses that are applied to collections of data, and in particular, collections of sensitive individual information. Hence, it is natural to ask what can be learned while preserving individual privacy. [Kasiviswanathan, Lee, Nissim, Raskhodnikova, and Smith; FOCS 2008] initiated such a discussion. They formalized the notion of private learning, as a combination of PAC learning and differential privacy, and investigated what concept classes can be learned privately. Somewhat surprisingly, they showed that, ignoring time complexity, every PAC learning task could be performed privately with polynomially many samples, and in many natural cases this could even be done in polynomial time. While these results seem to equate non-private and private learning, there is still a significant gap: the sample complexity of (non-private) PAC learning is crisply characterized in terms of the VC-dimension of the concept class, whereas this relationship is lost in the constructions of private learners, which exhibit, generally, a higher sample complexity. Looking into this gap, we examine several private learning tasks and give tight bounds on their sample complexity. In particular, we show strong separations between sample complexities of proper and improper private learners (such separation does not exist for non-private learners), and between sample complexities of efficient and inefficient proper private learners. Our results show that VC-dimension is not the right measure for characterizing the sample complexity of proper private learning. We also examine the task of private data release (as initiated by [Blum, Ligett, and Roth; STOC 2008]), and give new lower bounds on the sample complexity. Our results show that the logarithmic dependence on size of the instance space is essential for private data release.

99 GENERAL AND MISCELLANEOUS↗

Advanced bandwidth scheduling algorithms in dedicated networks

An increasing number of high performance research network testbeds and production networks have the capability of provisioning dedicated channels for high speed data transfer in support of large scale scientific applications. Each dedicated channel in these networks typically consists of one or more physical links that are shared by multiple applications in both time and bandwidth through in advance reservation. The design of efficient bandwidth scheduling algorithms is critical to maximizing the utilization of dedicated network resources and meeting diverse end-to-end transport performance requirements. The topology of a dedicated network is represented by a graph, where each link maintains a list of residual bandwidths specified as segmented constant functions of time. Given a graph with an aggregated time-bandwidth table combining the reservation information on all links, designated source, and destination vertices, and a specified data size, we formulate and investigate five bandwidth scheduling problems to minimize the data transfer end time under different path and bandwidth constraints. (1) variable path with variable bandwidth (VPVB), which computes the widest (highest bandwidth) path in each time slot; (2) fixed path with variable bandwidth (FPVB), which computes a fixed path with varying bandwidths across different time slots; (3) variable path with fixed bandwidth (VPFB), which computes one path in each time slot with the same bandwidth; (4) fixed path with fixed bandwidth (FPFB), which computes a fixed path with a constant bandwidth; and (5) multiple fixed paths with fixed bandwidths (MFPFB), which computes multiple concurrent fixed paths with constant bandwidths. Among these problems, VPVB represents the most flexible scenario where the network resources are fully utilized and the minimal transfer end time is achieved, while FPFB imposes the most stringent transport conditions by fixing both path and bandwidth. We design an optimal algorithm for each of these scheduling problems. VPVB is solved using an extension of the classical Dijkstra's algorithm, and FPVB is solved using Maximum Permutation Algorithm, which tries all possible permutations of throughputs at different time slots to obtain the minimal transfer end time. The algorithm for VPFB, FPFB, and MFPFB are based on Adjacent Time Slots Search, which finds the minimum transfer end time by examining the possibility of data transfer in any number of contiguous time slots. All these algorithms are of polynomial or pseudo polynomial time complexity with respect to the network size and the total number of time slots in a bandwidth reservation table.

99 GENERAL AND MISCELLANEOUS↗

Unstructured discontinuous Galerkin for seismic inversion.

This abstract explores the potential advantages of discontinuous Galerkin (DG) methods for the time-domain inversion of media parameters within the earth's interior. In particular, DG methods enable local polynomial refinement to better capture localized geological features within an area of interest while also allowing the use of unstructured meshes that can accurately capture discontinuous material interfaces. This abstract describes our initial findings when using DG methods combined with Runge-Kutta time integration and adjoint-based optimization algorithms for full-waveform inversion. Our initial results suggest that DG methods allow great flexibility in matching the media characteristics (faults, ocean bottom and salt structures) while also providing higher fidelity representations in target regions. Time-domain inversion using discontinuous Galerkin on unstructured meshes and with local polynomial refinement is shown to better capture localized geological features and accurately capture discontinuous-material interfaces. These approaches provide the ability to surgically refine representations in order to improve predicted models for specific geological features. Our future work will entail automated extensions to directly incorporate local refinement and adaptive unstructured meshes within the inversion process.

58 GEOSCIENCES↗

Resonant acoustic monitoring of damage in plug‐screw feeders

Abstract Plug‐screw feeders are critical in many industrial processes for compressing slurry materials via a rotating plug‐screw feeder. Over time, increasing plug‐screw feeder wear will eventually lead to catastrophic mechanical failure. Early detection of the wear state can prevent unplanned catastrophic failures resulting in operational shut‐downs, costly repairs, and most importantly the health and safety of workers. We present a theoretical basis for a noninvasive, in operando acoustic resonant technique to monitor the wear state of plug‐screw feeders. The technique is based on tracking the resonant acoustic modes of the plug‐screw feeder, which are sensitive to the plug‐screw feeder geometry, material, and operating conditions. We implemented a multivariate polynomial model to estimate the plug‐screw feeder wear state using multiple acoustic resonances by simulating the acoustic resonant modes for two categories of wear (tip and thread damage) that are common in plug‐screw feeders. Fitting multiple resonances to the polynomial model, we demonstrate accurate estimation of the total mass loss, as well as characterization of the type of damage (i.e., tip vs. thread). Current approaches for monitoring plug‐screw feeder wear rely on shutting down the operation and visually inspecting the plug‐screw feeder. The presented acoustic technique offers a noninvasive, in operando measurement approach that mitigates unplanned catastrophic failures. The acoustic resonance technique presented in the paper has a broad range of industrial applications including the Pharmaceutical, Mining, Integrated Biorefineries (IBR), and Additive Manufacturing industries, to name a few.

36 MATERIALS SCIENCE↗

From h to p efficiently: optimal implementation strategies for explicit time-dependent problems using the spectral/ hp element method: FROM h TO p EFFICIENTLY

We investigate the relative performance of a second-order Adams–Bashforth scheme and second-order and fourth-order Runge–Kutta schemes when time stepping a 2D linear advection problem discretised using a spectral/hp element technique for a range of different mesh sizes and polynomial orders. Numerical experiments explore the effects of short (two wavelengths) and long (32 wavelengths) time integration for sets of uniform and non-uniform meshes. The choice of time-integration scheme and discretisation together fixes a CFL limit that imposes a restriction on the maximum time step, which can be taken to ensure numerical stability. The number of steps, together with the order of the scheme, affects not only the runtime but also the accuracy of the solution. Through numerical experiments, we systematically highlight the relative effects of spatial resolution and choice of time integration on performance and provide general guidelines on how best to achieve the minimal execution time in order to obtain a prescribed solution accuracy. The significant role played by higher polynomial orders in reducing CPU time while preserving accuracy becomes more evident, especially for uniform meshes, compared with what has been typically considered when studying this type of problem.

42 ENGINEERING↗

Matrix-free preconditioning for high-order H (curl) discretizations

The greater arithmetic intensity of high-order finite element discretizations makes them attractive for implementation on next-generation hardware, but assembly of high-order finite element operators as matrices is prohibitively expensive. As a result, the development of general algebraic solvers for such operators has been an open research challenge. Fast matrix-free application of high-order operators has received significant attention in the literature in the context of Poisson-type problems, but preconditioners and solvers for inverting more general operators are not very well-developed. In this paper, we consider the problem of preconditioning a definite Maxwell operator at high polynomial order without assembling a matrix. We show that given efficient preconditioners for high-order H 1 finite element problems on the same mesh, efficient H(curl) preconditioners can be constructed in an auxiliary space framework. We demonstrate the resulting preconditioners in a practical setting with tensor-product basis functions on an unstructured mesh of quadrilaterals. Overall, our approach uses a sparsified H 1 solver constructed on a low-order mesh of the nodal points of the underlying high-order space, and we show that the resulting H(curl) preconditioner is effective at very high polynomial orders for two-dimensional model problems with complicated geometry, varying piecewise constant coefficients, and curved elements. The resulting preconditioner scales with nearly optimal O(p d+1 ) floating point operation count and optimal O(p d ) memory transfer requirements, outperforming existing Maxwell preconditioners in the high-order regime.

97 MATHEMATICS AND COMPUTING↗

Overcoming element quality dependence of finite elements with adaptive extended stencil FEM (AES‐FEM)

Summary The finite element methods (FEMs) are important techniques in engineering for solving partial differential equations, but they depend heavily on element shape quality for stability and good performance. In this paper, we introduce the Adaptive Extended Stencil Finite Element Method (AES‐FEM) as a means for overcoming this dependence on element shape quality. Our method replaces the traditional basis functions with a set of generalized Lagrange polynomial basis functions , which we construct using local weighted least‐squares approximations. The method preserves the theoretical framework of FEM and allows imposing essential boundary conditions and integrating the stiffness matrix in the same way as the classical FEM. In addition, AES‐FEM can use higher‐degree polynomial basis functions than the classical FEM, while virtually preserving the sparsity pattern of the stiffness matrix. We describe the formulation and implementation of AES‐FEM and analyze its consistency and stability. We present numerical experiments in both 2D and 3D for the Poisson equation and a time‐independent convection–diffusion equation. The numerical results demonstrate that AES‐FEM is more accurate than linear FEM, is also more efficient than linear FEM in terms of error versus runtime, and enables much better stability and faster convergence of iterative solvers than linear FEM over poor‐quality meshes. Copyright © 2016 John Wiley & Sons, Ltd.

Conley, Rebecca↗

High‐order symmetric cubature rules for tetrahedra and pyramids

Abstract In this article, we present an algorithm to construct high‐order fully symmetric cubature rules for tetrahedral and pyramidal elements, with positive weights and integration points that are in the interior of the domain. Cubature rules are fully symmetric if they are invariant to affine transformations of the domain. We divide the integration points into symmetry orbits where each orbit contains all the points generated by the permutation stars. These relations are represented by equality constraints. The construction of symmetric cubature rules require the solution of nonlinear polynomial equations with both inequality and equality constraints. For higher orders, we use an algorithm that consists of five sequential phases to produce the cubature rules. In the literature, symmetric numerical integration rules are available for the tetrahedron for orders p = 1 – 10, 14 , and for the pyramid up to p = 10 . We have obtained fully symmetric cubature rules for both of these elements up to order p = 20 . Numerical tests are presented that verify the polynomial‐precision of the cubature rules. Convergence studies are performed for the integration of exponential, weakly singular, and trigonometric test functions over both elements with flat and curved faces. With increase in p , improvements in accuracy is realized, though nonmonotonic convergence is observed.

Jaśkowiec, Jan↗

Kernel Manifolds: Nonlinear‐Augmentation Dimensionality Reduction Using Reproducing Kernel Hilbert Spaces

This paper generalizes recent advances on quadratic manifold (QM) dimensionality reduction by developing kernel methods-based nonlinear-augmentation dimensionality reduction. QMs, and more generally feature map-based nonlinear corrections, augment linear dimensionality reduction with a nonlinear correction term in the reconstruction map to overcome approximation accuracy limitations of purely linear approaches. While feature map-based approaches typically learn a least squares optimal polynomial correction term, we generalize this approach by learning an optimal nonlinear correction from a user-defined reproducing kernel Hilbert space. Our approach allows one to impose arbitrary nonlinear structure on the correction term, including polynomial structure, and includes feature map and radial basis function-based corrections as special cases. Furthermore, our method has relatively low training cost and has monotonically decreasing error as the latent space dimension increases. In conclusion, we compare our approach to proper orthogonal decomposition and several recent QM approaches on data from several example problems.

kernel methods↗

Parameterized complex dielectric functions of CuIn 1−x Ga x Se 2 : applications in optical characterization of compositional non‐uniformities and depth profiles in materials and solar cells

Abstract In‐situ spectroscopic ellipsometry (SE) was employed to extract the complex dielectric functions ε = ε 1 + iε 2 over the spectral range of 0.75–6.5 eV for a set of polycrystalline CuIn 1− x Ga x Se 2 (CIGS) thin films with different alloy compositions x = [Ga]/{[In] + [Ga]}. For highest possible accuracy in ε for each CIGS thin film, specialized SE procedures were adopted including (i) deposition to a thickness of ~600 Å on smooth native oxide covered crystal silicon wafers, which minimizes the surface roughness on the film and thus the required corrections in data analysis, and (ii) measurement in‐situ , which minimizes ambient contamination and oxidation of the film surface. Assuming an analytical form for each of the ε spectra for these CIGS films, oscillator parameters were obtained in best fits, and these parameters were fit in turn to polynomials in x . With the resulting database of polynomial coefficients, the ε spectra for any composition of CIGS can be generated from the single parameter, x . In addition to enabling accurate contactless determination of bulk and surface roughness layer thicknesses of CIGS films by high speed multichannel SE, the database enables characterization of the composition and its profile with depth into these films, and even how the depth profile varies spatially within the plane of the films. In this study, depth profile parameters were found to correlate spatially with solar cell performance parameters. As a result, SE provides the capability of contactless compositional analysis of production‐scale CIGS photovoltaic modules at high speed. Copyright © 2016 John Wiley & Sons, Ltd.

Aryal, Puruswottam↗