Community Research and Development Information Service - CORDIS

Final Report Summary - NANOSIM (A Multiscale Simulation-Based Design Platform for Cost-Effective CO2 Capture Processes using Nano-Structured Materials (NanoSim))

Executive Summary:
NanoSim project created an efficient and cost effective multi-scale simulation platform based on free and open-source codes in the context of reactive fluid-particle flows. This platform connected models of various type (i.e., electronic, atomistic, mesoscopic, and continuum), and spanned an extremely wide range of scales. Specifically, the project achieved to connect atom-scale phenomena (i.e., chemical reactions, diffusion), and meso-scale phenomena in fluid-particle systems (e.g., clustering) with process-scale aspects such as economic feasibility or optimal process conditions. Therefore, the NanoSim project developed one offline linking, and one online coupling software tool.
The offline linking tool “Porto” is a free and open source software system for connecting multi-scale simulations. Porto allows models to be composed into automated workflows where the input and output from each scale is described semantically. This has made it possible to compose workflows consisting of both open source software and proprietary software tools, as the connection between different software packages used in the NanoSim project are realized independent of proprietary file formats and conventions.
For online coupling, the project created “COSI”: a co-simulation platform based on extended versions of the popular open source LIGGGHTS® and CFDEM®coupling simulation packages. These packages implement the Discrete Element Method (DEM) to solve Newton’s equation of motion, and the Finite Volume Method for classical Computational Fluid Dynamics (CFD) simulations. The combined CFD-DEM package was integrated with a newly developed simulation tool (i.e., ParScale ), as well as the postprocessor CPPPO . COSI is in active use by industry and in research, and useful for studying heat and mass transfer problems in a large variety of industrial systems, e.g., fixed and fluidized beds.
To establish these software tools, the project performed implementation tasks, as well as developed (or improved) models and materials relations (MRs) to describe relevant phenomena. MRs were mainly developed based on simulation output at a certain scale, and were then used to close the physics equations on the next coarser scale. This scientific coupling between scales is supported by sophisticated software and data management in such a way that the actual model implementation in various software packages is fully automatic.
Reaction rate constants for methane-iron oxide systems were calculated based on electronic density functional theory (DFT) and atomistic kinetic Monte Carlo (kMC) calculations. The post-processor tool “REMARC” was developed to extract and process data from DFT and kMC output, as well as to fit the resulting reaction rate constants to Arrhenius parameters. These kinetics parameters have been supplied to a variety of continuum models. These models describe species and heat transport over a wide range of length scales: (i) within porous macroscopic particles (~ 100 µm), (ii) in reactive fluid-particle suspension flows (~100 mm), as well as (iii) within a full-scale chemical reactor (~10 m).
The modeling of reactive fluid-particle suspension flows was in the focus of NanoSim, and was concerned with dense bubbling or fixed beds, as well as circulating fluidized bed systems. The modeling consisted in the integration of new, more appropriate materials relations (using the post-processing tool, CPPPO) into formalism where the discrete form of the particles is lost, i.e., a continuum model-based description of the flow. This modeling approach was selected to tackle the industrial scale issues in terms of designing and optimizing full-scale reactors. A specific example was the use of an appropriate continuum model on the device scale to quantify the process intensification advantages offered by nano-structured materials.
The NanoSim project also developed reduced (i.e., one-dimensional) continuum models for full scale fluidized bed reactors used in the chemical looping reforming (CLR) process. Such a model of the CLR reactors with the CO2 capture process was linked with power plant simulations, i.e., a system model. Once the linking approach was established, techno-economic analysis of a gas fired power plant with CO2 capture based on CLR has been carried out. The kinetic data for reforming reactions using nano-structured materials was used from DFT and kMC simulations.
Experimental work on lab-scale reactors supported these modeling activities: Supported nanostructured oxygen carrier materials were synthesized, both at gram and kilogram scale. Their performance under reforming conditions was evaluated using a lab-scale reactor for fluidized and packed bed experiments, built up within the project. The experimental results were used for validation of multi-scale modeling and demonstration of process intensification for syngas production from natural gas reforming.
NanoSim dissemination was successful to interact with projects under the same call (via the EMMC – European Materials Modeling Council). NanoSim also scientifically disseminated its results via 85 publications and conference presentations, successfully held 4 workshops in the frame of international conferences and released a number of software components as open source packages.
In summary, NanoSim’s open source software platform and the multi-scale model linking methodology was used to facilitate the rational design of second generation gas-particle CO2 capture technologies based on nano-structured materials with a particular focus on CLR. However, the final NanoSim platform is sufficiently generic for application to a wide range of gas-particle contacting processes. Project outcomes are therefore expected to have far-reaching positive impacts on the European and global process industry.

Project Context and Objectives:
WP1 - Common Environment Software Platform ("Porto")
An objective of the NanoSim project is to create an efficient and cost effective multi-scale simulation platform based on free and open-source codes. This platform will connect models spanning a wide range of scales from the atomic scale through the particle and cluster scales, the industrial equipment scale and the full system scale.
The objective of WP1 Common Environment Software Platform ("Porto") has been to provide a unified environment for information flow and data sharing. Sharing is facilitated between open and proprietary software packages, independent of proprietary file formats and conventions. This is made possible by using semantically annotated data fields.
The purpose of Porto is to build workflows and connect the different software packages used in the NanoSim project. In particular, these workflows can be used to facilitate the rational design of second generation gas-particle CO2 capture technologies based on nano-structured materials with a particular focus on Chemical Looping Reforming (CLR).
The entire software platform is made available under the Lesser General Public Licence (LGPL) which will allow the software to be used in research and commercial products, while users to contribute to the platform.

WP2 - Development of Scale-Bridging Co-Simulation Software Platform (“COSI”)
The objective of the NanoSim project is to create an efficient and cost effective multi-scale simulation platform based on free and open-source codes. The focus of WP2 was to create a core co-simulation platform called COSI which will be established based the existing CFDEM®coupling and LIGGGHTS® open source simulation engines (particle and continuum modeling). These 2 existing simulation engines are complemented by the new simulation code ParScale and a filtering tool (CPPPO). The development of COSI is the core of WP2 (also the integration of development of ParScale and CPPPO in WP4). COSI can then be used to perform detailed Euler-Lagrange simulations of gas-particle systems such as fluidized or packed beds. The data from these detailed simulations (each particle or parcels of particles resolved) can be filtered using CPPPO and be used for closures in Euler-Euler simulations. Thus, COSI provides the basis for the work in WP4 and WP5.

WP3 - Atomistic Modelling
Atomistic modeling allows for the detailed evaluation of chemical reaction mechanisms, and the corresponding thermodynamics and kinetics. This makes it possible to directly model the reactivity of given materials and chemicals without involving the particular flow conditions. Atomistic modeling thereby serves as a complement to the flow modeling performed in the other work packages of NanoSim as well as a tool to interpret and complement experiments. The particular focus of WP3 has been to determine rate constants for all relevant atomistic processes as well as the nanoparticle stability on support material. This also involves the development of a tool (REMARC) for post-processing and communication of reaction data from atomistic to particle scale and continuum scale, as well as between atomistic codes and communication of the obtained structural and reactivity data to the Porto database.

WP4 - Lagrangian Modelling
WP4 (Lagrangian modeling) connects to WP2 (development of the co-simulation software platform “COSI”, which can be seen as the framework into which tools and results of WP4 will be integrated), as well as WP5 (this work package benefits from material relations and software developed within WP4) and WP9 (i.e., exploitation via academic education and training). Also, WP4 has a connection to WP3 with respect to interoperability with the tool “REMARC” that produces reaction kinetic data.
WP 4 has two main software development objectives: i) development of a simulation tool (based on a reduced continuum model for porous particles) for predicting concentration and temperature profiles within particles, as well as ii) development of a post-processing tool for offline coupling of direct numerical (flow) simulations (DNS; based on a continuum model of fluid and particle flow) of heat and mass transfer in dense suspensions. The latter objective includes the extension of an available DNS simulator in CFDEM®. In addition, WP 4 has two main application objectives (i.e., heat transfer experiments and validation of flow predictions, reversibility checks, and the development of material relations), as well as one main documentation objective (i.e., the provision of educational resources (documentation and training material preparation to ensure future exploitation).

WP5 - Eulerian Modelling
Two approaches can be used for the numerical simulations of gas-solid flows: the Euler-Lagrange approach or the Eulerian approach. Even if the Lagrangian approach is very attractive because of the reduced number of assumptions, the colossal number of particles involved in practical application, like in chemical looping combustion, limits the Lagrangian approach to the laboratory scale. Hence, nowadays, the numerical simulation of large industrial-scale gas-particle flows can only be done by the Eulerian approach. The Eulerian modeling of dense gas-particles flows is based on the Two-Fluid Model (TFM) that is derived from the kinetic theory of granular flows. Even if the TFM approach is well established some closures laws needs to be improved for better predictability of the numerical simulations. WP5 aims to improve the TFM and especially by taking into account the effects of unresolved structures. Indeed, recently it has been highlighted that if the mesh size is too large small-scale clusters are not well predicted leading to a bad prediction of the drag force and consequently a wrong solid flow rate in a circulating fluidized bed and a wrong bed height in a dense bubbling fluidized bed. WP5 has put the focus on the reactive flows and polydisperse flows. Basically, a database of mesh-converged 2D and 3D, reactive and non-reactive periodical circulating fluidized bed has been built. A spatial filter has been applied for extracting the computational and the subgrid contribution (the missing part that requires closure) that appears when a filter is applied on the TFM. The filtered approach has been assessed by a priori and a posteriori tests. The a priori tests consist of comparing the observed values in the mesh-converged simulations to model predictions. On the other hand, a posteriori tests consist of comparing the results of coarse grid simulations using subgrid closures to resolved simulation results or to experimental data.
Although the WP5 mainly put the focus on the fluidized bed, a part has been dedicated to fixed beds. Heat and mass transfer in fixed beds have been investigated by means of Particle Resolved Direct Numerical Simulation (PR-DNS). The results of this study were subsequently used to perform simplified 1D simulations of the Packed Bed Chemical Looping Reforming (PBCLR) process. The FTFM model has also been introduced in the OpenSource platform, OpenFoam®.

WP6 - Phenomenological Modelling
1D phenomenological models are a very popular and effective framework within industry for simulation, design and optimization of reactors due to its computational efficiency and relative simplicity when compared to more complex fundamental models that are still not easily accessible by industry. The aim of the phenomenological model within NanoSim (Phenom) is to deliver results of industrial interest in a user-friendly manner and within timeframes of seconds/ minutes.
Phenom consists of an already existing transient model for fixed bed reactors developed at SINTEF that has been regularly used in industrial projects and a steady/transient model for fluidized bed reactors developed within NanoSim. The current formulation of Phenom consists of a generic 1D phenomenological model for fluidized bed reactors based on the averaging probabilistic approach proposed by Thompson et al. (1999) that combines more than one fluidization regime and allows a smoother transition between them. Phenom’s formulation includes bubbling, turbulent and fast fluidization regimes that are the most used fluidization regimes in industry and in the chemical looping reforming process. It also handles stationary and transient simulations. Other features were added to the model such as the incorporation of membranes, simulation of co-current or counter-current reactor configuration and single or a cluster of reactors.
Phenom has been applied to three different chemical looping reforming technologies: the conventional chemical looping reforming (CLR), the novel gas switching reforming (GSR) and the fuel reactor of the membrane-assisted chemical looping reforming (MA-CLR).

WP7 - Validation Experiments
There two overall objectives for NanoSim WP7 are experimental validation of:
1. Effect of nano-structured materials on the performance in chemical looping processes
2. Process intensification of chemical looping processes determined by Multiphase flow models
The effect of nano-structured materials is validated through a threefold approach:
i. Synthesis of a selected state of the art nanostructured oxygen carrier with porosity, nanoparticle distribution and fluidizable granulate size as recommended from the NanoSim platform. Inert, porous support materials of suitable shape, size and mechanical strength will be impregnated with the nanocomposite oxygen carrier material.
ii. Development of an experimental protocol based on thermogravimetric analysis data and reactor testing that will serve to validate predictions from the NanoSim model (e.g. kinetics of oxidation and reduction, mass loss and gain and thermal stability).
iii. Evaluation of the synthesis method and other relevant methods with regard to their potential for translating into industrial scalable systems.
Multiphase flow model validation is obtained through:
I. Design and construction of a lab-scale fluidized/fixed bed reactor for testing the nanostructured material under real CLR process conditions. This is applied for the demonstration of potential for process intensification through the use of nano-structured materials at high temperatures, elevated pressures and variable feed gases.
II. Data from this reactor will include instantaneous pressure, temperature and gas composition measurements over the entire reactor height. The experimental campaign will be used to verify the reactive performance and mechanical stability of the material over many cycles of CLR process operation as well as to validate the resolved reactive multiphase flow CFD models.

WP8 - Techno-Economic Assessment
The primary objective of the project NanoSim is to speed up the development of second generation CCS processes. The current work package is dedicated to techno-economic assessment in NanoSim. In line with the objectives of NanoSim, the detailed tasks of the current WP are listed below.
• Establish multiscale modeling methodology to link 1D models of reforming reactors and power plant simulations and hence reduce the time taken in design of novel CCS processes
• Process design and integration studies of pre-combustion CO2 capture methods like chemical looping reforming (CLR) in combined cycle power plants
• Identify suitable design conditions in CLR through sensitivity studies to achieve higher net electrical efficiency and CO2 avoidance rates
• Estimate the levelised cost of electricity for the combined cycle power plants integrated with CLR
• The overlying objective of this WP is also to contribute through scientific publications and presentations.

WP9 - Dissemination and Education
The objective of the NanoSim project is to create an efficient and cost effective multi-scale simulation platform based on free and open-source codes. This platform will connect models spanning a wide range of scales from the atomic scale through the particle and cluster scales, the industrial equipment scale and the full system scale. To support the information flow and data sharing between different simulation packages, the NanoSim project will develop an open and integrated framework for numerical design called Porto to be used and distributed in terms of the GNU Lesser General Public License (LGPL). A core co-simulation platform called COSI (also licensed as LGPL) will be established based on existing CFDEMcoupling (an open source particle and continuum modeling platform).
The goal WP9 is to foster dissemination for the NanoSim project, to disseminate tangible results over the lifetime of the project, how to interact with fellow projects under the same call.





Project Results:
Please see the details in the file attached.
Potential Impact:
Please see the details in the file attached.
List of Websites:
https://www.sintef.no/projectweb/nanosim/

Related information

Reported by

STIFTELSEN SINTEF
Norway
Follow us on: RSS Facebook Twitter YouTube Managed by the EU Publications Office Top