Modeling, approximation, and time optimal temperature control for binder removal from ceramics

Carmen Chicone; Stephen J. Lombardo; David G. Retzloff

doi:10.3934/dcdsb.2021034

Article Contents

2022, Volume 27, Issue 1: 103-140. Doi: 10.3934/dcdsb.2021034

This issue Previous Article Approximate dynamics of a class of stochastic wave equations with white noise Next Article Monotonic and nonmonotonic immune responses in viral infection systems

Modeling, approximation, and time optimal temperature control for binder removal from ceramics

1.
Department of Mathematics, University of Missouri, Columbia, MO 65211, USA
2.
Department of Mechanical and Aerospace Engineering, and Department of Chemical Engineering, Columbia, MO 65211, USA
3.
Department of Chemical Engineering, University of Missouri, Columbia, MO 65211, USA

^* Corresponding author: Stephen J. Lombardo
^* Corresponding author: Stephen J. Lombardo

Received: September 2020

Revised: November 2020

Early access: February 2021

Published: January 2022

Abstract / Introduction Full Text(HTML) Figure(7) / Table(1) Related Papers Cited by

Abstract

The process of binder removal from green ceramic components-a reaction-gas transport problem in porous media-has been analyzed with a number of mathematical techniques: 1) non-dimensionalization of the governing decomposition-reaction ordinary differential equation (ODE) and of the reaction gas-permeability partial differential equation (PDE); 2) development of a pseudo steady state approximation (PSSA) for the PDE, including error analysis via $ L^2 $ norm and singular perturbation methods; 3) derivation and analysis of a discrete model approximation; and 4) development of a time optimal control strategy to minimize processing time with temperature and pressure constraints. Theoretical analyses indicate the conditions under which the PSSA and discrete models are viable approximations. Numerical results indicate that under a range of conditions corresponding to practical binder burnout conditions, utilization of the optimal temperature protocol leads to shorter cycle times as compared to typical industrial practice.

Keywords:

Mathematics Subject Classification: Primary: 35Q92, 35Q93, 93C95; Secondary: 34E15.

Citation:

Full Text(HTML)

Figure 1. Top panel: Centerline pressure versus temperature for linear temperature rise test case $ \beta = 5.6 \times 10^{-4} $ K/s (that is $ T(t) = \beta t+300 $) and green body length $ L = 0.01 $ m using numerical experiments with PDE model (5), PSSA (46), and three-station model (64). Bottom three panels: Relative centerline PSSA error for green body lengths (from top to bottom) $ 0.001 $ m, $ 0.01 $ m and $ 0.1 $ m

Download: Full-size image PowerPoint slide

Figure 2. Numerically approximated graphs of $ L^2 $ norms of difference between $ \epsilon \rho $ and corresponding PSSA versus temperature in physical units ([$ \rho \sqrt{L/2} $]) = mol/m$ ^{5/2} $ for the linear temperature rise (57) test case where $ \beta = 5.6\times 10^{-4} $ and green body lengths $ L = 0.001, 0.01, 0.1 $ m (depicted in order from bottom to top). The $ L^2 $ error is obtained using finite-difference approximation (MacCormack's method) of the solution of the model equations (5) and numerical quadrature for the $ L^2 $ norm, and the theoretical bound is computed likewise from the solution of the ODE (55)

Download: Full-size image PowerPoint slide

Figure 3. Numerically approximated graphs of the absolute value of the quality function and the absolute difference between $ u: = \epsilon \rho $ and the corresponding PSSA versus temperature (both computed at the centerline) for the linear temperature rise (57) test case where $ \beta = 5.6\times 10^{-4} $ and three green body lengths $ L = 0.001, 0.01, 0.1 $ m (depicted in order from bottom to top). The solution of the model equations (5) is approximated using MacCormack's method, and the quality function is computed using (44). Note: To benchmark the values, $ \rho_0 = 40 $ mol/m$ ^3 $

Download: Full-size image PowerPoint slide

Figure 4. Graphs of numerical approximations of pressure versus temperature for the linear temperature rise test case $ \beta = 5.6 \times 10^{-4} $ K/s (that is $ T(t) = \beta t+300 $) are depicted for a solid cylinder of height $ 0.03 $ cm and radius $ 0.005 $ cm using the PDE model (5) and corresponding PSSA at the geometric center computed using one (PSSA$ _1 $), two (PSSA$ _2 $), and ten (PSSA$ _{10} $) of its (transport scaled) power series representation (29) with post process conversion to corresponding physical units

Download: Full-size image PowerPoint slide

Figure 5. Top panel depicts optimal temperature protocol versus time for green body length $ 0.01 $ m, maximum temperature rise rate of 10 K per minute and maximum gas pressure $ 200\, 000 $ Pa for PDE model, PSSA, and three-station model. Bottom panel shows pressure versus temperature during the optimal temperature protocol for the three models

Download: Full-size image PowerPoint slide

Figure 6. Top panel depicts for the three-station model an approximation of the time-optimal temperature protocol for a green body of length $ 0.01 $ m for maximum temperature rise rates $ 0.015 $, $ 0.2 $, $ 0.5 $, $ 10.0 $ K per minute and maximum allowed gas pressure $ 200\,000 $ Pa. Bottom panel depicts corresponding pressure versus temperature

Download: Full-size image PowerPoint slide

Figure 7. Top panel depicts a comparison of the time optimal temperature protocols for the solid cylinder of height 3 cm and radius 0.5 cm as computed using the PDE model and ten terms of the series expansion representation (29) of its corresponding PSSA for the case of maximum heating rate 10 K per minute and maximum allowed gas pressure $ 150\,000 $ Pa, where the maximum is computed at the geometric center of the cylinder. Bottom panel depicts corresponding geometric center gas pressure versus temperature

Download: Full-size image PowerPoint slide

Table 1. Model Parameters

Name	Value	Description
$ T_0 $	300.0 K	initial temperature
$ P_0 $	$ 1.0\times 10^5 $ Pa	ambient gas pressure
$ R $	8.314 J/mol-K	ideal gas constant
$ M $	$ 4.4\times 10^{-2} $ kg/mol	molecular weight of gas
$ E $	$ 2.22\times 10^5 $ J/mol	activation energy
$ A $	$ 1.67 \times 10^{16} $ /s	pre-exponential factor
$ \mu $	$ 2.5\times 10^{-5} $ Pa-s	viscosity of gas
$ S $	$ 6\times 10^6 $/m	specific surface
$ k $	$ 5 $	tortuosity
$ \epsilon_{20} $	0.4	initial binder volume fraction
$ \epsilon_3 $	0.5	ceramic volume fraction
$ \rho_2 $	$ 1.0\times 10^3 $ kg/m$ ^3 $	polymer density
$ L $	0.001–0.1 m	green body length

| Show Table

DownLoad: CSV

Related Papers

Cited by

References

[1]	M. Bisi, F. Conforto and L. Desvillettes, Quasi-steady-state approximation for reaction-diffusion equations, Bull. Inst. Math., 2 (2007), 823-850.
[2]	J. Elderling, Normally hyperbolic invariant manifolds. The noncompact case, Atlantis Studies in Dynamical Systems, 2 (2013). doi: 10.2991/978-94-6239-003-4.
[3]	J. R. G. Evans, M. J. Edirisinghe, J. K. Wright and J. Crank, On the removal of organic vehicle from moulded ceramic bodies, Proc. R. Soc. London A, 432 (1991), 321-340.
[4]	K. Feng and S. J. Lombardo, Modeling of the pressure distribution in three-dimensional porous green bodies during binder removal, J. Am. Ceram. Soc., 86 (2003), 234-240. doi: 10.1111/j.1151-2916.2003.tb00005.x.
[5]	N. Fenichel, Persistence and smoothness of invariant manifolds for flows, Indiana Univ. Math. J., 21 (1972), 193-226. doi: 10.1512/iumj.1972.21.21017.
[6]	N. Fenichel, Geometric singular perturbation theory for ordinary differential equations, J. Diff. Eqs., 31 (1979), 53-98. doi: 10.1016/0022-0396(79)90152-9.
[7]	S. J. Fraser, The steady state and equilibrium approximations: A geometrical picture, J. Chem. Phys., 88 (1988), 4732-4738. doi: 10.1063/1.454686.
[8]	R. M. German, Theory of thermal debinding, Int. J. Powder Metall., 23 (1987), 237-245.
[9]	K. Kim and Y. Yao, The Patlak–Keller–Segel model and its variations: Properties of solutions via maximum principle, SIAM J. Math. Anal., 44 (2012), 568-602. doi: 10.1137/110823584.
[10]	J. A. Lewis, Binder removal from ceramics, Annual Rev. Mater. Sci., 27 (1997), 147-173. doi: 10.1146/annurev.matsci.27.1.147.
[11]	L. C-K. Liau and C-C. Chiu, Optimal heating strategies of polymer binder burnout process using dynamic optimization scheme, Ind. Che. Res., 44 (2005), 4586-4593. doi: 10.1021/ie049143a.
[12]	L. C.-K. Liau, B. Peters, D. S. Krueger, A. Gordon, D. S. Viswanath and S. J. Lombardo, Role of length scale on pressure increase and yield of poly(vinyl butyral)-barium titanate-platinum multilayer ceramic capacitors during binder burnout, J. Am. Ceram. Soc., 83 (2000), 2645-2653. doi: 10.1111/j.1151-2916.2000.tb01609.x.
[13]	K. C. Liddell, Shrinking core models in hydrometallurgy: What students are not being told about the pseudo-steady approximation, Hydromet., 79 (2005), 62-68. doi: 10.1016/j.hydromet.2003.07.011.
[14]	J. D. Logan, An Introduction to Nonlinear Partial Differential Equations, Wiley, Hoboken, 2008. doi: 10.1002/9780470287095.
[15]	S. J. Lombardo, Minimum time heating cycles for diffusion-controlled binder removal from ceramic green bodies, J. Amer. Ceram. Soc., 98 (2015), 57-65.
[16]	S. J. Lombardo and Z. C. Feng, Pressure distribution during binder burnout in three-dimensional porous ceramic bodies with anisotropic permeability, J. Mat. Res., 17 (2002), 1434-1440. doi: 10.1557/JMR.2002.0213.
[17]	S. J. Lombardo and Z. C. Feng, Analytic method for the minimum time for binder removal from three-dimensional porous green bodies, J. Mat. Res., 18 (2003), 2717-2723. doi: 10.1557/JMR.2003.0378.
[18]	S. J. Lombardo and D. G. Retzloff, A process control algorithm for reaction-diffusion minimum time heating cycles for binder removal from green bodies, J. Amer. Ceram. Soc., 102 (2019), 1030-1040. doi: 10.1111/jace.15964.
[19]	S. J. Lombardo and D. G. Retzloff, Reaction-permeability optimum time heating policy via process control for debinding green ceramic components, Adv. Appl. Ceram., 119 (2020), 150-157. doi: 10.1080/17436753.2019.1707393.
[20]	L. A. Segel and M. Slemrod, The quasi-steady-state assumption in biochemistry: A case study in perturbation, SIAM Rev., 31 (1989), 446-477. doi: 10.1137/1031091.
[21]	I. Stakgold, K. B. Bischoff and V. V. Gokhale, Validity of the pseudo-steady-state approximation, Int. J. Engng. Sci., 21 (1983), 537-542. doi: 10.1016/0020-7225(83)90101-5.
[22]	G. Y. Stangle and I. A. Aksay, Simultaneous momentum, heat and mass transfer with chemical reaction in a disordered porous medium: Application to binder removal from a ceramic green body, Chem. Eng. Sci., 45 (1990), 1719-1731. doi: 10.1016/0009-2509(90)87050-3.
[23]	T. A. Turányi, S. Tomlin and M. J. Pilling, On the error of the quasi-steady-state approximation, J. Phys. Chem., 97 (1993), 63-172.
[24]	D-S. Tsai, Pressure buildup and internal stresses during binder burnout: numerical analysis, AIChE J., 37 (1991), 547-554. doi: 10.1002/aic.690370408.
[25]	J. L. Vázquez, The Porous Medium Equation: Mathematical Theory, Clarendon Press, Oxford, 2007.