Note: This discussion is about an older version of the COMSOL Multiphysics® software. The information provided may be out of date.
Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.
4.3a demo: Battery capacity fade. There is an unwarranted parameter.
Posted Aug 25, 2013, 3:40 a.m. EDT Battery Design Version 4.3a, Version 4.3b 6 Replies
Please login with a confirmed email address before reporting spam
An interesting thing is discovered that the newly added demo "capacity fade.mph" in 4.3a was deleted in 4.3b for unkonwn reason.
The capacity fade model was established by referreing to these two papers:
1. P. Ramadass, B. Haran, P. Gomadam, R. White, and B. Popov, “Development of
First Principles Capacity Fade Model for Li-Ion Cells,” Journal of the Electrochemical
Society, vol. 151, no. 2, pp A196–A203, 2004.
2. G. Ning, R. White, B. Popov. “A generalized cycle life model of rechargeable Li-ion
batteries”, Electrochimica Acta, vol 51, 2012-2022 (2006).
Most settings in the demo is the same as the models in the papers, except "the exchage current density of side reaction i0_side".
In the papers, i0_side was set to constant with typical value of 1.5e-6 [A/m^2].
However, in the demo of comsol 4.3a, i0_side was regarded as a varible of local current density of the main lithium intercalation (Jn). (Page 2 of models.bfc.capacity_fade.pdf)
This fundamental difference would lead to distinct simulation results.
I am very intereting if there is any argument for comsol to relate i0_side to Jn?
In fact, the simulation results appear more reasonable with Jn-related i0_side than constant i0_side. But, after refering to more than 20 papers, I have not find any phsical-based evidence to support the assumption that the i0_side is related to Jn.
The capacity fade model was established by referreing to these two papers:
1. P. Ramadass, B. Haran, P. Gomadam, R. White, and B. Popov, “Development of
First Principles Capacity Fade Model for Li-Ion Cells,” Journal of the Electrochemical
Society, vol. 151, no. 2, pp A196–A203, 2004.
2. G. Ning, R. White, B. Popov. “A generalized cycle life model of rechargeable Li-ion
batteries”, Electrochimica Acta, vol 51, 2012-2022 (2006).
Most settings in the demo is the same as the models in the papers, except "the exchage current density of side reaction i0_side".
In the papers, i0_side was set to constant with typical value of 1.5e-6 [A/m^2].
However, in the demo of comsol 4.3a, i0_side was regarded as a varible of local current density of the main lithium intercalation (Jn). (Page 2 of models.bfc.capacity_fade.pdf)
This fundamental difference would lead to distinct simulation results.
I am very intereting if there is any argument for comsol to relate i0_side to Jn?
In fact, the simulation results appear more reasonable with Jn-related i0_side than constant i0_side. But, after refering to more than 20 papers, I have not find any phsical-based evidence to support the assumption that the i0_side is related to Jn.
6 Replies Last Post Jul 2, 2015, 12:22 a.m. EDT
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
seems no reply....
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
The main idea here is that the side reaction will only occur when the graphite is expanding, since the side reaction only occurs on graphite sites that are not covered by the SEI layer.
The reference "1" states (on first page, bullet 2 under "Choice of side reaction and assumptions"), that "the solvent reactions occur only during charging".
You could of course couple the side reaction to the global state variables and set it to zero during discharge.
However, it makes more sense to couple the side reaction rate to what is going on locally in the electrode, and that is done by using the max(...,...) factor.
The reference "1" states (on first page, bullet 2 under "Choice of side reaction and assumptions"), that "the solvent reactions occur only during charging".
You could of course couple the side reaction to the global state variables and set it to zero during discharge.
However, it makes more sense to couple the side reaction rate to what is going on locally in the electrode, and that is done by using the max(...,...) factor.
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Thanks very much for your reply.
We have reached a consensus that the side reaction only occurs during charging. However, I concerned about the absolute value of i0_side rather than its sign.
If we only want to get the information of charging or discharging, it should be expressed as:
"i0_side * max(sign(liion.iloc_per1),0)" ----(1)
instead of the current expression in demo:
"i0_side * max(liion.iloc_per1/i_loc_ref),0)". ----(2)
Note that the value of Eq.(1) is zero(discharging) or CONSTANT i0_side(charnging), which is in accordance with reference [1].
However, the value of Eq.(2) is zero(discharging) or NOT CONSTANT i0_side * abs(liion.iloc_per1/i_loc_ref), which is not in accordance with reference [1].
Looking forward to your reply again。
We have reached a consensus that the side reaction only occurs during charging. However, I concerned about the absolute value of i0_side rather than its sign.
If we only want to get the information of charging or discharging, it should be expressed as:
"i0_side * max(sign(liion.iloc_per1),0)" ----(1)
instead of the current expression in demo:
"i0_side * max(liion.iloc_per1/i_loc_ref),0)". ----(2)
Note that the value of Eq.(1) is zero(discharging) or CONSTANT i0_side(charnging), which is in accordance with reference [1].
However, the value of Eq.(2) is zero(discharging) or NOT CONSTANT i0_side * abs(liion.iloc_per1/i_loc_ref), which is not in accordance with reference [1].
Looking forward to your reply again。
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Yes you are right. If you want a constant i0, only active during charging, that is the way to do it. I believe, however, that it would be more reasonable to use a higher i0 for higher charge currents, since the expansion rate will be higher for the graphite particles at higher currents.
The Comsol model is not an exact reproduction of the models of the cited papers. It should be seen more as a tutorial model (the geometrical thicknesses are different too for instance).
The Comsol model is not an exact reproduction of the models of the cited papers. It should be seen more as a tutorial model (the geometrical thicknesses are different too for instance).
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
I agree with you.
I also believe that it would be more reasonable to use a higher i0 for higher charge currents,
so I am finding more evidence or references to support this view.
Unfortunenately, I haven't found yet.
Do you have some related papers to be recommended for me?
Thank you very much!
I also believe that it would be more reasonable to use a higher i0 for higher charge currents,
so I am finding more evidence or references to support this view.
Unfortunenately, I haven't found yet.
Do you have some related papers to be recommended for me?
Thank you very much!
Please login with a confirmed email address before reporting spam
Posted:
9 years ago
I tried too. Unfortunately, there is no. The pioneering work for battery ageing was proposed in the papers of Ramadass et al. 2004 and Ning et al. 2006. A number of papers cited these two papers and attempted to enable its further applications or model reductions. However, the original two papers didn't give much information about i0.
As my understanding, the original work derived i0 from curve-fitting experiment data under specific conditions. And this model with constant i0 cannot be extended for general applications. I used this model to test battery ageing at different charging rates. It turns out battery deteriorates less at larger charging rates. This is ridiculous. So a more general and accurate model with i0 as a function of charging rates is highly needed.
As my understanding, the original work derived i0 from curve-fitting experiment data under specific conditions. And this model with constant i0 cannot be extended for general applications. I used this model to test battery ageing at different charging rates. It turns out battery deteriorates less at larger charging rates. This is ridiculous. So a more general and accurate model with i0 as a function of charging rates is highly needed.