Modeling of Particle Fates in Isothermal Plug Flow Reactor Chiara Galletti 1, Gianluca Caposciutti 1 , Giovanni Coraggio 2 , Leonardo Tognotti 1 1 Department of Civil and Industrial Engineering, University of Pisa 2 International Flame Research Foundation
29
Embed
Modeling of Particle Fates in Isothermal Plug Flow Reactor · 2021. 1. 4. · Modeling of Particle Fates in Isothermal Plug Flow Reactor Chiara Galletti1, Gianluca Caposciutti1, Giovanni
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Modeling of Particle Fates in gIsothermal Plug Flow Reactor
Chiara Galletti1, Gianluca Caposciutti1, Giovanni Coraggio2, Leonardo Tognotti1
1 Department of Civil and Industrial Engineering, University of Pisap C g g, U y2 International Flame Research Foundation
Agendag Motivations
The IFRF Isothermal Plug Flow Reactor
Procedures for heterogeneous kinetics and UQ Procedures for heterogeneous kinetics and UQ
Modeling the particle fate in IPFR
Conclusions
Motivation: predicting solid fuel combustionp g
Improving existing technologies and/or developing new ones for lid f lsolid fuels:
Oxy-coal combustion is emerging technological solution for: CCS (Carbon Capture and Sequestration); CCS (Carbon Capture and Sequestration); Abatement of NOx and SOx emissions; Retrofitting of existing coal-fired power and industrial plants.
Motivation: solid fuel combustion pyrolysis or devolatilization reactions of the solid fuel particles;
heterogeneous reactions of the residual char;
secondary gas-phase reactions of the released gases
Brown R C Iowa State Press Ames IA
4
Brown, R. C, Iowa State Press, Ames, IA,
2003.
from IFRF R&D Agenda – 2007-2013Fuel characterisationFuel characterisation
• Develop/test methodologies for fuel characterisation; establish
t l / lifi ti d (UQ)protocols/qualification procedures (UQ)
• Characterise solid (and liquid) fuels to agreed protocols– to agreed protocols
– to fill data gaps for sub-model validation & application
– includes fuels that are environmentallyincludes fuels that are environmentally and economically significant
• Biomass, Wastes, Blends with coalsI t h th t fl t• In atmospheres that reflect O2/RFG approach, temperatures and pressures of current interest to members and other sponsors
Produce and maintain DATABASES (IFRF Solid Fuel Database- http://sfdb.ifrf.net
thermal history of particles and nominal temperature are
PSD is important for evaluating the effective nominal temperature are
different the effective
trajectories and thermal histories
adhesion of particles: mass never balances
segregation of particles (depending on size)
CFD modelling as a
diagnostic tool
solid collection depends on the efficiency of separation units
measurement of gas concentration is difficult (deposition of tar)
separation units
Specific objectives: kinetics from EFRsp j EFRs often used to derive solid fuel conversions in specific
ti ditioperating conditions
Estimation of kinetics requires knowledge of particle temperature sophisticated diagnostics usually the particle heating up is neglected TP = TR
Objectives:
Analyse procedure to derive devolatilization kinetics from EFRs
I ti t ibl f t i ti Investigate possible sources of uncertainties
Can we improve the reliability of simple models by revising the d f th i d i ti ?procedure for their derivation?
CFD modelStep 1 Step 2 Step 3
POST-PROCESSING with Matlab® routines
Experiments
13
CFD settings
Simulation cases• single-phase runsthermal field• injection of inert particles thermal particle history• injection of reactive particles optimized A, E
Coal type
particle size: 65 – 90 mm VM 40.30 %; FC 47.95 %; ASH 11.75 % (dry
b i )• SebukuCoal type basis) Tin=293 K
Domain • 3D symmetric half IPFR, 1M cells
Sebuku
Temperature • 1173 K, 1373 K, 1573 K
Gas composition• At inlet: CO2 9.45 %; H2O 7.73 %; N2 79.93%; O2 2.89 % (mass
fraction)Gas composition fraction)• CO2 carrier gas for coal particles
Solvers• ANSYS Fluent, steady, second order discretization scheme• SIMPLE algorithms for the pressure-velocity couplingSolvers SIMPLE algorithms for the pressure velocity coupling• Two way coupled lagrangian tracking
Radiation • P1 model coupled with WSGGDevolatilization • SFOR, SFOR-HR, CPD 14
Single-phase thermal fieldg p
velocitytemperature
15
Injection of inert particlesj p
Coefficient of restitution
R=R⁄⁄=R
16
Injection of inert particlesj pParticle thermal histories
17
Mean particle residence time is different from expected one Ranges of residence times t and particle temperature TP exist at the same
sampling location y
Reactive simulations: devolatilization models
(Semi-empirical) models computationally cheap but poor di ti f l til i ld Constant-rate model
Standard/modified model Two-step model: Kobayashi et al (1977)
predictions of volatile yield
Two-step model: Kobayashi et al. (1977), … Multi-step kinetic model: Sommariva et al. (2009)
Phenomenological models t b t t ti ll Phenomenological models CPD FLASHCHAIN
accurate but computationally expensive
FG-DVC
Single First Order Reaction Model SimplicitySingle First Order Reaction Model • Simplicity • Easily included in CFD codes• Not generally accepted• Inaccurate for T above the Tprox
Reactive simulations: post-processingp p g
Post-processing to emulate experiments in which the position of th li b i i dthe sampling probe is varied.
19
CFD: kinetics from reactive simulationsKinetics from CFD+expNovel procedureStandard procedure
(X ) t T
EXPERIMENT EXPERIMENT
(X y) at T(X, y) at TR
(X, y) at TR
SFOR-HRhp: TP=T0+HR t
A l ti l i t ti
SFORhp: TP=TR
SFORhp: TP=TR
(X, y) at TR
V/UQ
Analytical integration over particle thermal
history
p P RArrhenius plot
p P RArrhenius plot
A E
t t, HRA, E
CFDA E A, E
Iterative procedure: 2-3 iterations needed to get fi l (A E) t
A, E
final (A, E) parameters
Kinetics from CFD+exp.
SFOR
p
SFOR-HR
Minimize the following function Minimize the following function
Arrhenius plot A,E
SFOR model: conversions
TR=1173K TR=1373K TR=1573KR R TR 1573K
Strong underestimation of conversion Experimental error bars:
horizontal: uncertainty due to sampling probe position vertical: uncertainty due to ash tracer method
i Modeling error bars: horizontal: uncertainty on residence time vertical: uncertainty on conversion
SFOR-HR model: conversions
TR=1173K TR=1373K T =1573KTR 1173K TR 1373K TR=1573K
Good predictivity
Uncertainty on kinetic parametersy p
Uncertainty on kinetic parameters
Performance of CPD model
CPD model leads to an overestimation of particle
i t l dconversions at low and medium temperature with a constant final conversion value for all temperaturesp
Model-form uncertainty: lack of accuracy in the evaluation of ultimate yield at a givenultimate yield at a given temperature
Conclusions (1)( ) Need to characterise solid fuels for practical applications:
“Fuel specific” sub models: Simple, Reduced & Reliable Kinetic Mechanisms for devolatilisation , char oxidation and gasificationgas ca o
EFRs/DTs used to derive heterogeneous kinetics in specific operating conditionsoperating conditions
Need to to qualifyqualify the EFRs/DTs procedures
Uncertainties quantification : Experimental/analysis uncertainties CFD modeling as a diagnostic tool (particle thermal histories)
Conclusions (2)( ) CFD analysis of solid fuel particles injected into a pilot-scale EFR:
particles experience different paths and thermal histories particle have a temperature lower than the reactor one at most of the
sampling positions
Iterative (2-3 cycles) joint CFD-experimental procedure to derive kinetics: Residence times from CFD calculations Particle temperature estimated with an average heating rate from
CFD model (analytical integration of the volatile release equation)
Large improvement of the agreement between experimental and predicted conversion data, even with simple SFOR model
Si ifi t d ti ti f th ki ti t ith ti l Significant underestimation of the kinetic rates with conventional assumption of constant particle temperature
Analysis of cloud of particles allows estimating uncertainty on kinetic Analysis of cloud of particles allows estimating uncertainty on kinetic parameters
References – IFRF archiveIFRF Reports at http://www.research.ifrf.net/research/new.html
IFRF Members’ Conferences and TOTeMs athttp://www ifrf net/page/conference notes/index conferenceshttp://www.ifrf.net/page/conference-notes/index-conferences
Iavarone, S., Caposciutti, G., Galletti, C., Tognotti, L., Contino, F., Parente, A.Iavarone, S., Caposciutti, G., Galletti, C., Tognotti, L., Contino, F., Parente, A.Adaptive Kinetic model for coal devolatilization in oxy-coal combustion conditions(2015) 18th IFRF Members’ Conference – Flexible and clean fuel conversion to industry Freising.
Federica Barontini, Enrico Biagini, Leonardo TognottiCharacterization of the devolatilization products of selected second generation biofuelsCharacterization of the devolatilization products of selected second generation biofuels18th IFRF Members’ Conference – Flexible and clean fuel conversion to industry, Freising, 1-3 June, 2015
Galletti, C., Tarquini, S., Bruschi, R., Giammartini, S., Coraggio, G., Tognotti, L.Ignition delay of coal particle clouds in oxy-fuel conditions(2012) 17th Members' Conference, Mafflier, France 2012
References - Journal PaperspC Galletti, G Caposciutti, L Tognotti Evaluation of scenario uncertainties in entrained flow reactor tests through CFD modelling: devolatilizationEnergy & Fuels 2016Energy & Fuels, 2016
S.Iavarone, C.Galletti, F. Contino, L.Tognotti, P.J.Smith, A.ParenteCFD-aided benchmark assessment of coal devolatilization one-step models in oxy-coal combustion conditionsFuel Processing Technology, in press 2016
Li, J., Bonvicini, G., Biagini, E., Yang, W., Tognotti, L.Characterization of high-temperature rapid char oxidation of raw and torrefied biomass fuels(2015) Fuel 143 pp 492 498(2015) Fuel, 143, pp. 492-498.
Li, J., Bonvicini, G., Tognotti, L., Yang, W., Blasiak, W.High-temperature rapid devolatilization of biomasses with varying degrees of torrefaction(2014) Fuel, 122, pp. 261-269.
Biagini, E., Tognotti, L.A generalized correlation for coal devolatilization kinetics at high temperature(2014) F l P i T h l 126 513 520(2014) Fuel Processing Technology, 126, pp. 513-520.
Galletti, C., Giacomazzi, E., Giammartini, S., Coraggio, G., Tognotti, L.Analysis of coal combustion in oxy-fuel conditions through pulsed feeding experiments in an entrained flow reactor(2013) Energy and Fuels, 27 (5), pp. 2732-2740.
Karlström, O., Brink, A., Biagini, E., Hupa, M., Tognotti, L.Comparing reaction orders of anthracite chars with bituminous coal chars at high temperature oxidation conditionsp g g p(2013) Proceedings of the Combustion Institute, 34 (2), pp. 2427-2434.
Biagini, E., Simone, M., Barontini, F., Tognotti, L.A comprehensive approach to the characterization of second generation biofuels(2013) Chemical Engineering Transactions, 32, pp. 853-858.
Karlstrom O; Brink A; Hercog J; Hupa M; Tognotti LOne-Parameter Model for the Oxidation of Pulverized Bituminous Coal Chars.One Parameter Model for the Oxidation of Pulverized Bituminous Coal Chars. ENERGY & FUELS - vol. 26 (2), 2012
Karlström, O., Brink, A., Hupa, M., Tognotti, L.Multivariable optimization of reaction order and kinetic parameters for high temperature oxidation of 10 bituminous coal chars(2011) Combustion and Flame, 158 (10), pp. 2056-2063.
Simone, M., Biagini, E., Galletti, C., Tognotti, L.Evaluation of global biomass devolatilization kinetics in a drop tube reactor with CFD aided experimentsEvaluation of global biomass devolatilization kinetics in a drop tube reactor with CFD aided experiments(2009) Fuel, 88 (10), pp. 1818-1827.