A Modeling Approach for Time-Dependent Geometry Applied to Transient Heat Transfer of Aluminum Electrolysis Cells


The thermal balance of aluminum electrolysis cells (AEC) have to be rigorously controlled in order to improve the efficiency and sustainability of this industrial process. A new modeling strategy is developed to consider the displacements of solid bodies and moving boundaries in finite element models. The transient thermal-electric modeling of the AEC demonstrates the effect of an increase in operating voltage on both the anode cover and the side ledge. With higher heat generation, the anode cover deteriorates and the side ledge thickness decreases. Since the anode cover is characterized by irreversible transformations, the top heat dissipation remains higher even when the operating voltage comes back to its typical value. For the first time, the transient temperature and electric fields throughout the anode life are simulated and validated by industrial measurements. The modeling predictions have been validated from instrumented anodes and manual measurements, all performed on operating AEC.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11
Fig. 12
Fig. 13
Fig. 14
Fig. 15
Fig. 16


A :

Area (m2)

c p :

Specific heat capacity (J/kg K)

E :

Electric field (V/m)

F :

View factor

H :

Height (cm or m)

h :

Convection heat transfer coefficient (W/m2 K)

J :

Current density (A/m2)

k :

Thermal conductivity (W/m K)

L :

Length (cm or m)

q :

Heat transfer rate (W)

q″ :

Heat flux (W/m2)

\( \dot{q} \) :

Rate of energy generation per unit volume (W/m3)

t :

Time (s, h or day)

T :

Temperature (K or °C)

V :

Electric potential or voltage (V)

δ ij :

Kronecker delta

\( \nabla \) :

Gradient vector field

ε :


ρ :

Density (kg/m3)

σ :

Electrical conductivity (S/m); Stefan–Boltzmann constant (W/m2 K4)


Anode–cathode distance


Anode cover material


Aluminum electrolysis cell


Bath–metal interface


Computer-aided design


Center channel


Computational fluid dynamics


Cryolite ratio


Heat flux sensor


Mean absolute relative error


On the anode


Side channel

bold :

A vector


A matrix or a concentration of a chemical species


  1. 1.

    J. Thonstad, P. Fellner, G.M. Haarberg, J. Hives, H. Kvande, A. Sterten: Aluminium Electrolysis: Fundamentals of the Hall-Heroult Process, 3rd ed., Aluminium-Verlag, Düsseldorf, Germany, 2001, pp. 1-8.

    Google Scholar 

  2. 2.

    A.E. Gheribi, S. Poncsák, S. Guérard, J.F. Bilodeau, L. Kiss, P. Chartrand: J. Chem. Phys., 2017, vol. 146, pp. 1-10. https://doi.org/10.1063/1.4978235.

    Article  Google Scholar 

  3. 3.

    S. Poncsák, L. Kiss, A. Belley, S. Guérard, and J.F. Bilodeau: Light Met. 2015, Proc. Int. Symp., 2015, pp. 655–59.

  4. 4.

    S. Poncsák, L. Kiss, R. St-Pierre, S. Guérard, and J.F. Bilodeau: Light Met. 2014, Proc. Int. Symp., 2014, pp. 585–89.

  5. 5.

    S. Poncsák, L. Kiss, S. Guérard, J.F. Bilodeau: Metals, 2017, vol. 7, pp. 1-10. https://doi.org/10.3390/met7030097.

    Article  Google Scholar 

  6. 6.

    F. Allard, G. Soucy, L. Rivoaland, M. Désilets: J. Therm. Anal. Calorim., 2015, vol. 119, pp. 1303-1314.

    Article  Google Scholar 

  7. 7.

    A. Fallah-Mehrjardi, P.C. Hayes, E. Jak: Metall. Mater. Trans. B, 2014, vol. 45B, pp. 1232-1247.

    Article  Google Scholar 

  8. 8.

    J. Liu, A. Fallah-Mehrjardi, D. Shishin, E. Jak, M. Dorreen, M. Taylor: Metall. Mater. Trans. B, 2017, vol. 48B, pp. 3185-3195.

    Article  Google Scholar 

  9. 9.

    F. Allard, M. Désilets, A. Blais: Thermochim. Acta, 2019, vol. 671, pp. 89-102.

    Article  Google Scholar 

  10. 10.

    F. Allard, M. Désilets, M. LeBreux, A. Blais: Int. J. Heat Mass Transfer, 2019, vol. 132, pp. 1262-1276.

    Article  Google Scholar 

  11. 11.

    Q. Zhang, M.P. Taylor, J.J.J. Chen: Metall. Mater. Trans. B, 2015, vol. 46B, pp. 1520-1534.

    Article  Google Scholar 

  12. 12.

    F. Allard, M. Désilets, M. LeBreux, and A. Blais: Light Met. 2015, Proc. Int. Symp., 2015, pp. 565–70.

  13. 13.

    X. Liu, M. Taylor, and S. George: Light Met. 1992, Proc. Int. Symp., 1992, pp. 489–94.

  14. 14.

    L.N. Less: Metall. Trans. B., 1977, vol. 8B, pp. 219–225.

    Article  Google Scholar 

  15. 15.

    F. Allard, M. Désilets, M. LeBreux, and A. Blais: Light Met. 2016, Proc. Int. Symp., 2016, pp. 289–94.

  16. 16.

    H. Wijayaratne, M. Hyland, M. Taylor, A. Grama, and T. Groutso: Light Met. 2011, Proc. Int. Symp., 2011, pp. 399–404.

  17. 17.

    X. Shen: PhD thesis, University of Auckland, New Zealand, 2006.

  18. 18.

    K.A. Rye, J. Thonstad, and X. Liu: Light Met. 1995, Proc. Int. Symp., 1995, pp. 441–49.

  19. 19.

    M.A. Llavona, L.F. Verdeja, R. Zapico, F. Alvarez, and J.P. Sancho: Light Met. 1990, Proc. Int. Symp., 1990, pp. 429–37.

  20. 20.

    G. Hatem, M. Llavona, T. Log, J.P. Sancho, and T. Ostvold: Light Met. 1989, Proc. Int. Symp., 1989, pp. 365–70.

  21. 21.

    M.A. Llavona, R. Zapico, P. García, J.P. Sancho, and L.F. Verdeja: Light Met. 1988, Proc. Int. Symp., 1988, pp. 201–06.

  22. 22.

    K.E. Einarsrud, I. Eick, W. Bai, Y. Feng, J. Hua, P.J. Witt: Appl. Math. Modell., 2017, vol. 44, pp. 3–24.

    Article  Google Scholar 

  23. 23.

    B. Bardet, T. Foetisch, S. Renaudier, J. Rappaz, M. Flueck, and M. Picasso: Light Met. 2016, Proc. Int. Symp., 2016, pp. 315–19.

  24. 24.

    S. Langlois, J. Rappaz, O. Martin, Y. Caratini, M. Flueck, A. Masserey, and G. Steiner: Light Met. 2015, Proc. Int. Symp., 2015, pp. 771–75.

  25. 25.

    M. Ariana, M. Désilets, P. Proulx: Can. J. Chem. Eng., 2014, vol. 92, pp. 1951-1964.

    Article  Google Scholar 

  26. 26.

    M. Blais, M. Désilets, M. Lacroix: Appl. Therm. Eng., 2013, vol. 58, pp. 439-446.

    Article  Google Scholar 

  27. 27.

    D. Marceau, S. Pilote, M. Désilets, J.F. Bilodeau, L. Hacini, and Y. Caratini: Light Met. 2011, Proc. Int. Symp., 2011, pp. 1041–46.

  28. 28.

    Y. Safa, M. Flueck, J. Rappaz: Appl. Math. Modell., 2009, vol. 33, pp. 1479-1492.

    Article  Google Scholar 

  29. 29.

    M. Dupuis and V. Bojarevics: Light Met. 2005, Proc. Int. Symp., 2005, pp. 449–54.

  30. 30.

    T.X. Hou, Q. Jiao, E. Chin, W. Crowell, and C. Celik: Light Met. 1995, Proc. Int. Symp., 1995, pp. 755–61.

  31. 31.

    A.T. Brimmo, M.I. Hassan, Y. Shatilla: Appl. Therm. Eng., 2014, vol. 73, pp. 116-127.

    Article  Google Scholar 

  32. 32.

    M. Désilets, D. Marceau, and M. Fafard: Light Met. 2003, Proc. Int. Symp., 2003, pp. 247–54.

  33. 33.

    M. LeBreux, M. Désilets, F. Allard, and A. Blais: Numer. Heat Transf. A, 2016, vol. 69A, pp. 128-145.

    Article  Google Scholar 

  34. 34.

    Q. Wang, L. Gosselin, M. Fafard, J. Peng, B. Li: Metall. Mater. Trans. B, 2016, vol. 47B, pp. 1228-1236.

    Article  Google Scholar 

  35. 35.

    D. Picard, J. Tessier, G. Gauvin, D. Ziegler, H. Alamdari, and M. Fafard: Metals, 2017, vol. 7, pp. 1–9. https://doi.org/10.3390/met7090374.

    Article  Google Scholar 

  36. 36.

    H. Abbas, M.P. Taylor, M. Farid, and J.J. Chen: Light Met. 2009, Proc. Int. Symp., 2009, pp. 551–56.

  37. 37.

    R. Zhao, L. Gosselin, A. Ousegui, M. Fafard, D.P. Ziegler: Numer. Heat Transfer, Part A, 2013, vol. 64A, pp. 317-338.

    Article  Google Scholar 

  38. 38.

    R. Zhao, L. Gosselin, M. Fafard, J. Tessier, D.P. Ziegler: Int. J. Therm. Sci., 2017, vol. 112, pp. 395-407.

    Article  Google Scholar 

  39. 39.

    K. Stein, T.E. Tezduyar, R. Benney: Comput. Methods Appl. Mech. Eng., 2004, vol. 193, pp. 2019-2032.

    Article  Google Scholar 

  40. 40.

    A.A. Johnson and T.E. Tezduyar: Comput. Methods Appl. Mech. Eng., 1994, vol. 119, pp. 73-94.

    Article  Google Scholar 

  41. 41.

    J.N. Reddy and D.K. Gartling: The Finite Element Method in Heat Transfer and Fluid Dynamics, 3rd ed., CRC Press, Boca Raton, Florida, 2010, pp. 229-235.

    Google Scholar 

  42. 42.

    G. Vidalain, L. Gosselin, M. Lacroix: Int. J. Heat Mass Transfer, 2009, vol. 52, pp. 1753-1760.

    Article  Google Scholar 

  43. 43.

    ANSYS Inc.: ANSYS Mechanical APDL Theory Reference, Canonsburg, Pennsylvania, 2017, pp. 203–18.

  44. 44.

    M.F. Cohen and D.P. Greenberg: Comput. Graphics, 1985, vol. 19, pp. 31-40.

    Article  Google Scholar 

  45. 45.

    Y.S. Touloukian and D.P. DeWitt: Thermophysical Properties of Matter - The TPRC Data Series - Volume 8, Plenum Publishing Corporation, New York, NY, 1972, pp. 8-73.

    Google Scholar 

  46. 46.

    C.F. Windisch, B.B. Brenden, O.H. Koski, and R.E. Williford: Final report on the PNL program to develop an alumina sensor, U.S. Department of Energy, Pacific Northwest Laboratory, United States, 1992, pp. 28–29.

  47. 47.

    Hukseflux Thermal Sensors: HF01 High Temperature Heat Flux Sensor (version 1211), Delft, Netherlands, 2003, pp. 15-16.

    Google Scholar 

Download references


This study was supported by Rio Tinto Aluminium, the “Conseil de Recherches en Sciences Naturelles et en Génie du Canada” (CRSNG) and the “Fonds de Recherche du Québec - Nature et Technologies” (FRQNT). The authors wish to thank the staff at Rio Tinto Grande-Baie smelter and Arvida Research & Development Center (ARDC), especially Mr. Jean-François Bilodeau and Mr. Sébastien Guérard from ARDC, for the support provided during the realization of this work.

Conflict of interest

The authors declare that they have no competing interests.

Author information



Corresponding author

Correspondence to François Allard.

Additional information

Manuscript submitted September 9, 2018.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary material 2 (MP4 39624 kb)

Supplementary material 3 (MP4 46052 kb)

Supplementary material 1 (PDF 100 kb)


Appendix A

A modeling method for the geometry and mesh update algorithm followed by the nodal interpolation is detailed in this appendix. This method is applied to the transient thermal-electric modeling of the industrial AEC in order to consider the melting of the anode crust by a moving boundary.

A.1 Python Scripting in ANSYS Workbench

The moving geometry and mesh update algorithms are automated with a Python script interpreted by the ANSYS Workbench software. A while loop is programmed to solve each analysis step of the transient model and the results are saved automatically. The principal steps of this method are described by the scheme of Figure A1. A previously calculated or a new crust–cavity profile (moving boundary) is input in the parameter set to modify the geometry of the model. Accordingly, the geometry is updated using a Python command and a new mesh in the ANSYS Mechanical model is generated.

Fig. A1

Scheme of the modeling algorithm to solve the transient thermal-electric model with geometry and mesh updates, considering the moving boundary at the crust–cavity interface

A.2 APDL Scripting in ANSYS Mechanical

The coupled thermal-electric analysis is performed using Mechanical APDL solver in the ANSYS Mechanical software. APDL commands are integrated to read the load data stored in a CSV file, like the thermal and electric properties or the boundary conditions. The new properties or conditions are applied by APDL commands to the current analysis. The previously calculated temperatures are mapped to the new mesh with an interpolation function (*MOPER).[43] This function performs matrix operations by comparing arrays, which contain the previous nodal coordinates and temperatures, and also the new mesh information. A 3D interpolation is applied to get the nodal temperatures on the new mesh. In that simulation case, a similar interpolation method is used to apply the previous material conditions to the newly generated elements. This step allows to consider the irreversible phase transformation in the anode cover (ACM → anode crust). With the previous temperatures and material conditions applied to the new mesh, the transient model is solved for a specified analysis time, before starting the postprocessing. Subsequently, the temperatures in the anode crust are read and a new crust–cavity profile is calculated in order to reach the melting isotherm. The new profile is written in a CSV file to be read by Python commands in order to update the geometric parameter and so on. The nodal coordinates and temperatures are also exported to a CSV file to be read in the next iteration. The procedure in Figure A1 is repeated up to the final time of the analysis.

Appendix B

A moving geometry, mesh update, and nodal interpolation method is developed to reflect the anode consumption and its downward displacement in the AEC. The modeling approach is automated with Python scripting and the results are interpolated from one step to the next one with APDL commands. A moving interface for the melting of the anode crust is integrated in this modeling method. This second case additionally integrates an algorithm to consider the time-dependent geometry related to the anode consumption. The motion of the nodes that are located inside the moving parts is governed by the velocity of anode assembly, a downward movement in this case. The node coordinates are modified to follow the anode displacement. Since the geometry is moved and the mesh is automatically updated, the new mesh conforms to the quality criterium initially imposed. The previously calculated nodal temperatures are then applied as initial conditions before the new analysis starts. The complete scheme of resolution is detailed in Figure B1.

Fig. B1

Scheme of the modeling algorithm to solve the transient thermal-electric model with moving geometry, mesh update, and nodal interpolation, considering both the moving parts and the displacement of crust–cavity interface

B.1 Python Scripting in ANSYS Workbench

A new aspect of this second model is the displacement of the body parts with automatic updates of the CAD geometry by Python scripting. For each analysis time, a new position is calculated and applied to the moving parts. Since the anode is consumed by the process, its size is imposed before each resolution step. The python script also applies the new location of the moving boundary (anode crust melting) according to the postprocessing results.

B.2 APDL Scripting in ANSYS Mechanical

In the ANSYS Mechanical interface, a new mesh is generated for each resolution step. The previously moved nodal and element properties are interpolated and then applied as initial conditions for the thermal-electric model. The boundary conditions and constraints for the analysis are imposed with APDL commands. Thereafter, the model is solved and the postprocessing is performed. During postprocessing, a new crust–cavity profile is calculated only if the anode crust has melted. The nodes in the moving geometry are selected and their coordinated are displaced using matrix operations to account for the downward movement. Consequently, the nodal temperatures and element properties are saved to CSV files to be imported in the next resolution step.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Allard, F., Désilets, M. & Blais, A. A Modeling Approach for Time-Dependent Geometry Applied to Transient Heat Transfer of Aluminum Electrolysis Cells. Metall Mater Trans B 50, 958–980 (2019). https://doi.org/10.1007/s11663-019-01510-6

Download citation