Somnath Ghosh - US grants

This Research Initiation Award is made to develop stable, accurate and efficient computational models for large deformations of metals and composites. The finite element method with arbitrary Lagrangian-Eulerian kinematic description will be implemented for industrial forming processes.

The prime objective of the research is to develop a system of computational models for response analysis of actual heterogeneous materials. In these materials, microscopic particulates or fibers (in reinforced composites) or precipitates and pores, (in alloy systems) of arbitrary shapes and sizes, are randomly dispersed in the matrix. The intended research is based upon a vigorous collaboration between university and industry. The research will blend voronoi cell elements, homogenization method and extend them, to innovate a novel and sophisticated system of computational modules, that will accurately depict the complex deformation mechanisms and material evolution in actual multi phase materials.

9457603 Ghosh This NYI award is to provide linkage between fundamentals of mechanics, materials science, design and manufacturing. Novel methods of computational mechanics using multiple scale analyses for concurrent design and processing of advanced materials will be employed. Industrial interaction will be used to validate the procedures. ***

9810096 Ghosh A conference is conducted focusing on six major areas in Mechanics and Materials: (a) crystal imperfections, (b) polycrystals and properties, (c) constitutive modeling, (d) plasticity and failure, (e) metal forming and casting, and (f) process and product design. The conference provides directions for industrial relevance of these topics. Current obstacles and constraints to tailoring material properties and to optimizing process and product designs are identified. ***

This Grant Opportunity for Academic Liaison with Industry (GOALI) project will involve the principal investigator and researchers at the ALCOA Technical Center (ATC). This award is to institute an integrated experimental-computational research program targeted at improving productivity and product quality. This will be accomplished by developing advanced multiple scale computational models and software for simulating failure in materials processing and subsequent process design. It will augment two newly initiated thrust areas at ATC on: (1) understanding the effect alloy microstructure and process parameters on edge cracking in cold rolling processes; and (2) improving cut-edge quality and minimize burr and debris in material shearing processes such as slitting. Furthermore, collaborations with Scientific Forming Technologies Corporation (SFTC) and Ohio Supercomputer center will elevate the simulation software to the advanced level of high performance computing needed for solving complex industrial problems.

Experimentally motivated-validated, multi-scale computer models will introduce adaptive multi-level hierarchy and represent details of material microstructure from quantitative metallography, to simulate ductile fracture. The adaptive multi-level system of computational modules will concurrently predict variables at the scales of the work-piece and model localization and ductile fracture as a phenomenon of incidence, interaction and propagation of damage across the scales. Adaptive hierarchy of computational sub-domains with varying resolutions will differentiate between non-critical and evolving critical regions and will zoom in at 'damaging hotspots' for pure microscopic simulations. Microscopic stress and damage analysis of real microstructures will be done by the specially developed Voronoi cell finite element models. The forming simulations will be coupled with design methodologies and software to predict 'optimal' microstructural constitution and process parameters. Experiments on localization and failure in deformation processing will be performed, and will include mechanical testing of notched specimens for load-displacement and fracture toughness data, and real process like cold rolling and sheet metal shearing for

Integrated computational- experimental program for ductility and failure in cast aluminum alloys
Somnath Ghosh, Bhaskar Majumdar and Steve Harris

Today's The automotive industry is faced with major challenge of improvedmustis faced with major challenges to improve performance to and reduce weight ratio in order to meet lofty fuel efficiency and emissions standards at low cost. Improved mAs designs become more complex and the power output requirements increase, the practical limits of their ductility and ultimate strength are being reached.,. which Scrapped parts and downtime can cost a manufacturer millions of dollars in terms of scrapped parts and downtime. Compounding The proposed Industry-University Collaborative GOALI research is aimed at addressing thisese issues for meeting goals of materials with high ductility, ultimate strength and strength at low cost. It will build a collaborative relation between the Ohio State University (PI), New Mexico Institute of Technology and Ford Research Laboratory (industry partner) to launch an integrated experimental-computational research program. The program will augment a major thrust area at FRLord called Virtual Aluminum Casting or VAC that is targeted to (a) reduce product development time (b), improve quality and performance, and reduce scrap and (c) improve performance and lower weight, and (d) reduce costs.
and cycle time.
The proposed program will develop a system of experimentally validated adaptive multiple scale computational models for predicting localization and ductile fracture of cast Al-Si components from microstructural information and process conditions. The models will simulate the evolution of microstructural features such as voids and secondary phases into incipient cracks and determine how the loss of ductilityductile failure depends on alloy properties and on the , distributions and interactions of different phases in the microstructure. The mechanics of particle fracture, interface decohesion, matrix rupture and damage percolation through the dendritic network will be studied. The role of porosity size and distribution on failure will also be investigated. Various dDevelopmental modules will include: (i) Quantitative metallography using SEM, and orientation imaging microscopy (OIM), and microstructural characterization to identify and characterize critical microstructure features that control important material response; (ii) Mechanical tests accompanied bywith in-situ SEM and fracture surface observations, computer imaging and microstructural characterization observation to generate strain fields and to provide understanding of critical mechanisms in the failure process; (iii) Neutron diffraction measurements and Raman microprobe techniques for microstress evolution and probabilistic strength estimation of particles; (iv) Development of an adaptive multi-level model for multiple scale analysis to predict the failure process as a phenomenon of multi-scale incidence and propagation of cracks; (v) Development of image-based microstructural Voronoi Cell finite element model for efficient and accurate analysis of plastic deformation, strain localization and damage evolution in nonuniform heterogeneous microstructures; and (vi) Incorporation of a probabilistic analysis framework to account for the effect of input variabilities on ductility and failure. The major intellectual merit of the proposed research is in its innovative blend of state state-of of-the the-art computational tools andwith experimental methods to advance provide a comprehensive analysis tool and design methodology for advanced metallic materialscast metals to increase their effective utilization. The uniqueness of this approach is in the broad attack on the problem: (a) iIntroduction of adaptive hierarchical and multi-scale computational models, incorporating image-based microstructural models to depict the percolation of damage at different length scales;. T (b) he iIncorporation of detailed microstructures at the critical regions of evolving damage and localization is possible through the efficient and accurate Voronoi Cell finite elementFE model, being developed by the PI.; and (c) Robust validation of the models through rigorous feedback from multi-scale experiments and material characterization by using in-situ SEM, orientation imaging microscopy (OIM), in-situ neutron diffraction and Raman microprobe. To the best knowledge of the investigators, there is a lack of such a necessary comprehensive approach to the understanding of response and failure characteristics of complex cast microstructures. The program, upon completion, is expected towill provide a good understanding of stress and strain evolution ofin the the complex phases in cast Al, their strength levels, and damage initiation and percolation through the network of brittle, ductile and porous phases.

The broader impact of the program will occur on two fronts. front, the It will reach beyond the automotive industry to aid the entire casting industry, where significant gains in alloying and solidification technology is are often stymied by unknowns regardingnot knowing how variability in material and process parameters affect damage tolerance and ductility. The methodology will allow industry be able to leapfrog thepresent technology and use these lightweight allows into in new safety -critical applications, armed with the knowledge that ductility and fracture can now be predicted with a reasonable degree of confidence. The second front will be oGraduate students will intern at FRL every summer and NMT students will have access to equipment at the national labs. As a consequence of the university-industry collaboration collaboration, students in this program will have a strong interaction with and mentorship from industrial researchers . There will also be student interaction with researchers at Sandia National Laboratory. researchers. In addition, the national laboratories will be involved in the experimental component of the work.

This grant provides support to enable students to participate in the 2004 NUMIFORM conference, which will be held at the Ohio State University, June 13-17. The conference, which rotates between the U.S., Europe and Japan, and which is held every three years, brings together researchers in the field of industrial forming processes to discuss recent advances and future directions in the modeling of materials and manufacturing processes. The conference emphasizes traditional processing of metals and polymers, and will include advanced materials and emerging manufacturing approaches at different scales. Topics will include bulk and sheet forming, superplastic forming, casting, molding, quenching, materials joining and powder forming, machining, thermal mechanical processing, chemical processing, polymer processing, surface treatment, food processing, composite manufacturing and reactive liquid molding.

This grant will enable the participation of about 20 students, and will provide them with an enhanced educational experience, and give them a chance to meet the top researchers in their chosen field. The topical areas of this conference are vital to the economic health of the nation, and include areas where it is important to maintain a skilled workforce.

The proposed collaborative research will build on computational models developed by S. Ghosh (principal investigator, OSU) and R. Ghanem (co-principal investigator, USC) to develop an integrated computational system for probability based multi-scale model (PMM) of ductile fracture in lightweight metals and alloys with precipitates and inclusions. It will incorporate (a) a probabilistic multi-scale pre-processor consisting of a multi-scale image builder and morphology based domain partitioning; (b) image based microstructural Voronoi Cell finite element model for accurate, high resolution analysis and probabilistic characterization of plastic deformation, strain localization and ductile fracture; (c) stochastic homogenization based continuum plasticity-damage model for macroscopic analysis in the multi-scale material model. The PMM simulator will help in the development of light alloy components with optimal material microstructures for superior failure properties such as ductility.

Major materials, automotive and aerospace industries will benefit significantly from the comprehension and flexibility of this computational system in handling a variety of problems for different materials, loading and geometric constructs. Specifically the research will aid the materials industry, to leapfrog present technology and use these lightweight alloys in new safety-critical applications. Graduate students at two institutions will work together with researchers at Ford Research Laboratory to develop significant interdisciplinary expertise and industrial exposure. This interaction will provide a well-rounded education as students train to enter the technical workforce. New courses in Stochastic Material Modeling and Multiscale Analysis will be developed. The PI?s will organize workshops at respective universities to introduce students and industrial partners to this novel area of research.

In this collaborative effort, the PI?s will develop a system of 3D image-based computational models for finite deformation crystal plasticity modeling that incorporates morphological features and crystallographic orientations of real material microstructures. An innovative grain-based adaptive finite element model will be created for accurate and efficient analyses, overcoming limitations of conventional FE methods. Each grain in a polycrystalline aggregate will be represented by a single super-element incorporating a special hybrid assumed stress-plastic strain formulation. The element will feature adaptive augmentation of resolution to represent evolving localization zones and nascent cracks in the microstructure. Crystallographic dislocation densities will also be incorporated as spatial field variables, which can convect and localize to affect plastic hardening. Effective multi-scaling schemes for coupling the dislocation density model with efficient coarse-grained crystal plasticity models will be developed. The GAFEM developments will also be accompanied by a multiple time scale model for simulating large number of cycles required to capture initiation and evolution in fatigue crack simulations. New strategies for the measurement of microstructural response to mechanical loading will be incorporated for calibrating constitutive models. Novel, complementary experiments at different scales, will use a combination of focused ion beam (FIB) micromachining of test articles coupled with testing in a modified nanoindentor and other testing methods, enabling extraction of high quality uniaxial and simple bending constitutive data, and state-of-the-art SEM and TEM observations. Interfaces will be characterized with pillar/cantilever testing of bicrystals and with a combination of orientation microscopy (OM) analysis, surface strain mapping and optical interferometry. These procedures will form a basis for the development of property databases necessary for verification and validation of microstructure-sensitive deformation and damage modeling.

The program, upon completion, will provide an unprecedented detailed and integrated understanding of the role of microstructure on deformation and failure characteristics in Ti alloys. This is critical to reliable materials design, especially with respect to creep and fatigue characteristics. Success of this new paradigm will be of great interest to industries such as GE Aviation since the time and cost saving for inserting advanced alloys in safety-critical applications will be tremendous and will allow industry to leapfrog present technologies.

Intellectual merit: The US National Congress of Computational Mechanics has become a premier conference in the field of Computational Science and Engineering, not only in the USA but also internationally. It encompasses a broad range of disciplines and has evolved into a confluence of multidisciplinary ideas. For students and post-doctoral researchers, this is an ideal platform for exposure to the state of the art developments in a broad range of fields. The conference provides an excellent platform for younger researchers to showcase their ideas and research progress, to experts in various fields and gain instant recognition.

Broader impact: Students and post-doctoral researchers will participate through presentations in technical sessions and poster sessions. Furthermore, panel discussions are planned with experts from industry, academia and government (Air Force, Army, Navy and DOE labs) to provide future directions. The organizing committee will try to encourage underrepresented minority and women students to participate in this congress by awarding travel scholarships.

The objective of this Grant Opportunity for Academic Liaison with Industry (GOALI) project, in collaboration with General Motors (GM) R&D Center, will develop a novel system of experimentally validated, physics-based multi-level computational models for simulating multi-scale deformation and predicting ductile failure in Magnesium alloys. Simulations will focus on high strain rate impact loading and forming processes. The multi-scale model will develop modules at three relevant scales, viz. (i) macroscopic scale of structural components, (ii) microscopic scale of individual grains and polycrystalline aggregates, and (iii) atomic scales of crystal lattices for postulating crack evolution. Both homogenization and localization strategies will be incorporated in a hierarchical-concurrent framework. Specific developmental modules include: (i) image-based three-dimensional microstructure generation of polycrystalline aggregates, (ii) image-based crystal-plasticity FEM modeling of microstructural deformation and failure, (iii) homogenized continuum plasticity-damage laws for incorporation in macroscopic models, and (iv) macroscopic simulations. GM will be responsible for experiments at different scales to validate and calibrate the computational models. The program will provide an unprecedented understanding of the role of microstructure on deformation and failure characteristics in Magnesium alloys.

Magnesium alloys have a huge potential for enhancing fuel efficiency and energy economization in the automotive industry, which will be facilitated. The program will aid the materials industry, where materials development is stymied by not knowing how variabilities at different scales affect performance and processing. Graduate students will have a strong interaction with eminent industrial researchers. The program will educate next generation of engineers on the challenges of emerging technologies, with a unique academia-industry perspective.

This grant provides partial funding to support workshops on Multiscale Modeling for Multifunctional Analysis and Design (MMMAD) and Challenges in Computational Multiscale Materials Modeling (CCMMM) to be held in Arlington, Virginia. These workshops are part of a series of three interdisciplinary workshops that will focus on experimental and computational methods in multiscale analysis and design. The workshops will be held from 2 thru 5 May 2011 and are entitled:
(i) Experimental Multiscale Methods (EMM),
(ii) Multiscale Modeling for Multifunctional Analysis and Design (MMMAD), and
(iii) Challenges in Computational Multiscale Materials Modeling (CCMMM).
The workshop series will be sponsored by multiple government agencies, viz. ARO, AFOSR and NSF. The EMM workshop will focus on measurement of deformation and experimental methods across scales. It will also address validation of multiscale models by means of multiscale experiments. The MMMAD workshop will discuss specific challenges pertaining to the development of robust multiscale computational analysis-design platforms for multifunctional materials and structures. The CCMMM workshop will address critical issues that confront the computational multiscale materials modeling research discipline for comprehensive analysis of complex material structure interactions at a broad range of length and time scales. The workshops will address critical challenges and needs in establishing robust spatio-temporal multiscale modeling of material systems.

The potential of the multiscale modeling paradigm is unfathomable at the current stages, given the tremendous growth in various aspects of this discipline. The theme is consistent with current trends in Integrated Computational Materials Science and Engineering (ICMSE) and Simulation based Engineering & Sciences (SBES) to guide the design of novel systems that meet specifications of performance and reliability. The proposed workshop series will bring together researchers from academia, industry and national laboratories to identify major challenges and research objectives for developing robust multiscale modeling and design platforms. A comprehensive report will result from this workshop which will be an important roadmap for future strategic developments in this field.

The research objective of this grant is the development of physics-based crack evolution models in polycrystalline metallic materials for use in crystal plasticity finite element models. The unified deformation-fracture models will be developed by upscaling variables from atomic-scale, molecular dynamics simulations with explicit depiction of the atomic structure and deformation mechanisms. Upscaling will provide a framework for cohesive laws and associated evolution laws for deformation variables along specific slip systems. A systematic approach will be undertaken, encompassing the following tasks: (i) molecular dynamics (MD) simulations of 3D crystalline lattice systems to simulate the evolution of ductile and brittle crack evolution under a variety of force fields, (ii) development of accelerated MD methods to simulate longer time scales consistent with experiments, (iii) characterization and identification of crack tip deformation mechanisms, followed by quantification of the local atomic structure and evolution, and (iv) development of energy-based laws governing crack growth in the process zone by partitioning the total energy corresponding to different physical mechanisms.

Benefits to the scientific community will be accomplished through the establishment of novel crack evolution models of critical importance in analysis and design of materials for better fracture toughness and ductility. Advancement in accelerated molecular dynamics modeling will allow modeling at strain rates of realistic experiments. Collaboration with Sandia and Brown Deer Technology will enable technology transfer of deliverables for broader utilization. Students will have strong interactions with eminent researchers at Sandia National Laboratories and a small-scale development company to see the direct impact of their work.

This grant provides funding to support the International Union of Theoretical and Applied Mechanics (IUTAM) Symposium on "Integrated Computational Structure-Material Modeling of Deformation and Failure under Extreme Conditions," to be held in Baltimore, MD on June 10-22, 2016. The symposium will focus on different material classes and cover a range of spatial and temporal scales needed for physics-based modeling of deformation and failure under extreme operational parameters. The symposium will bring together top researchers from academia, government laboratories and industry. A number of junior faculty members and young researchers from academia and government laboratories will be invited to participate in the IUTAM symposium. Funding through this proposal would increase the number of junior faculty and researchers participants to prepare the next generation of scientific leaders in this field.

High demands on future systems are challenging the operational limits of materials like metals, ceramics and composites. The design of enhanced systems to meet performance and reliability challenges requires advanced methods of analysis and simulation to provide a comprehensive understanding of the structure-material behavior. Distinguished speakers from academia, industry and government laboratories have been invited to address the state of the art in the thematic areas of the symposium and to make approximately 35 presentations. There will be three panel discussion sessions focused on the prediction of material behavior during fatigue, impact and blast loading, and radiation and high temperature environments. Additionally, a poster session is planned to showcase research progress by junior researchers.

Additive manufacturing (AM) processes for metallic materials, namely powder-bed fusion, and various powder-feed and wire-feed systems, are bringing dramatic changes to the manufacturing industry. The unprecedented agility achieved through layer-by-layer material addition is enabling near net-shape production of complex components. Despite this promise and progress, the qualification and acceptance of AM parts have been compromised by the lack of consistency in material behavior and life-limiting properties, e.g., undesirable ductility and fatigue behavior. These inconsistencies are often attributed to subtle, yet characteristic variations in the microstructure (morphology and defect structure), possibly a consequence of small perturbations in the AM process parameters. By implementing a multi-pronged approach incorporating innovative methods of integrated computational materials engineering (ICME), such as physics-based multi-scale modeling, multi-objective design, and materials characterization and testing, this research will address this shortcoming by developing a robust framework for location specific material properties and behavior in structures fabricated by AM processes. Ultimately, it will provide guidance on the links between process parameters and product performance and life, paving the acceptance of AM parts in critical applications. The research will promote the sciences associated with ICME; advance the national health, prosperity, and welfare by establishing the provenance of AM manufacturing; and secure the national defense by enabling near on-demand, net-shape production of critical parts in theaters. Collaborative partnerships with Sandia National Laboratories and JHU/Applied Physics Laboratory will be a strength of this research. Graduate students on this project will undergo multi-disciplinary training in state-of-the-art techniques in multi-scale modeling, design optimization, materials characterization and modern manufacturing processes. The collaboration with Sandia and JHU/APL will also provide them exposure to the most advance technological advancements that will help their professional career development.

Two specific developmental modules will be realized for AM processed metals and alloys with polycrystalline-polyphase microstructures. They are: (i) development and implementation of parametrically homogenized constitutive/damage models (PHCMs) with explicit representation of microstructural descriptors in the form of representative aggregate microstructural parameters or (RAMPs), and (2) development of PHCM-based robust design methods for location-specific material design in structures designed, e.g., by topology optimization. Furthermore, PHCM-based simulations readily estimate the effect of variations in the microstructure on structure-scale variables like stresses, strains, strength, and even ductility or fatigue life. This will facilitate sensitivity analysis in location-specific material design in AM processed structures that have been conceptualized by design methods like topology optimization. A multi-objective function framework will be incorporated for identifying a Pareto front, defining the achievable cross-property space. The outcome of the detailed material design undertaking will be a structural-scale layout of the microstructure (in terms of RAMPs) for a robust, site-dependent distribution of macro-scale properties. This will guide the selection of AM processes parameters and routes to improve structure-material performance and life.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

This award will support a Grant Proposal Writing Workshop to be held in conjunction with the 2022 Engineering Mechanics Institute Conference in Baltimore, Maryland, 31 May to 3 June 2022. Early career researchers typically have limited exposure and experience with grant proposal writing and the proposal review process. This workshop aims to provide early career researchers with opportunities to learn about the National Science Foundation (NSF) and its merit review process and engage in a series of interactive sessions with recent NSF awardees discussing best practices and strategies in proposal writing. An emphasis will be placed on recruiting women and underrepresented minorities in engineering to participate in the workshop. By providing clarity, guidance, and feedback, the workshop will support the next generation of engineering researchers in improving the quality of their future proposals and, ultimately, the quality of research and broader impacts guided by clearly defined objectives, plans, and mechanisms to assess success.

A series of interactive seminars, panels, and other activities will be held in this workshop with the goal of improving the quality of future proposals from participating early career researchers. Participants will include senior doctoral students, postdoctoral researchers, and early career faculty. Speakers and panelists will feature recent NSF awardees, including awardees from NSF’s Faculty Early Career Development Program. Session topics will include the following: (i) an overview of the NSF, the Civil, Mechanical, and Manufacturing Innovation Division, and the Merit Review Process; (ii) best practices and strategies in grant proposal writing; (iii) shared experiences from recent grant awardees; and (iv) small breakout sessions providing participants the opportunity to ask detailed questions and receive feedback on project summaries that may serve as the basis of a future proposal. Additionally, participants will be encouraged to present a poster of their choice for display during the workshop. A key feature of the workshop is that it will be linked with the 2022 Engineering Mechanics Institute Conference, which will provide opportunities for the early career researchers to network and discuss their research with researchers from academia, national laboratories, and industry, while also ensuring that workshop discussions and feedback will be directly relevant and accessible to participants from the community of engineering mechanics and civil infrastructure.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.