Added bipolar membrane unit model and costing #1500
Conversation
| Figure 1. Schematic representation of a bipolar electrodialysis unit | ||
|
|
||
|
|
||
| One bipolar membrane along with the **Acidate** and **Basate** channels can thus be treated as a modelling unit that can |
There was a problem hiding this comment.
Are acidate and basate typical terminology for BPMD?
There was a problem hiding this comment.
They are not. I needed a name for the acid & base channel.
I went off diluted/ concentrated channel which are called diluate/concentrate (in ED documentation). Extrapolating from this I chose acidate/basate.
If anybody has better suggestion I am happy to use it.
There was a problem hiding this comment.
In theory they are an acid concentrate and base concentrate (increasing in concentration), where the brine is a brine diluate (deceasing in concentration) based on the donation of ions toward the objective of the system, correct?
There was a problem hiding this comment.
Yes, they do increase concentration. But I think there is a subtle difference. BPMED increase concentration via production of new acid/base (through water splitting) and not by just transporting ions across the membranes. I feel like that deserved a distinction in nomenclature. Thoughts?
There was a problem hiding this comment.
I'd go with either acid and base or acidic and basic.
There was a problem hiding this comment.
@adam-a-a That's reasonable. I have changed the names to Acidic & Basic
| @@ -6,6 +6,7 @@ Unit Models | |||
|
|
|||
| anaerobic_digester | |||
| aeration_tank | |||
| Biploar_electrodialysis_0D | |||
There was a problem hiding this comment.
| Biploar_electrodialysis_0D | |
| bipolar_electrodialysis_0D |
| @@ -0,0 +1,340 @@ | |||
| Bipolar Electrodialysis (0D) | |||
There was a problem hiding this comment.
This is great! While we're here, you should also add an rst file to describe costing, e.g.,: https://watertap.readthedocs.io/en/latest/technical_reference/costing/detailed_unit_model_costing.html
There was a problem hiding this comment.
That's a good idea. However, I made some changes to the unit model costing and would like some feedback (before committing it to the docs).
While doing analysis I realised that a big component (negative cost/value addition) of the bipolar costing might be best done on the flowsheet level. So, I have removed it (in the most recent version).
As it stands the BPMED costing is very similar to the ED costing. The difference being the variable name for number of membranes. (cell_pair_num vs cell_num). Is there a way to inherit the ED costing?
Also, just for future reference I plan to add the bpmed + ED unit model (a configuration that is commonly used and something that I have been using for analysis). This would require minor alteration to the costing. It would be nice to know the capabilities/limits of the inheritance. Since that would inform whether I need a costing just for bipolar membrane.
I am happy to elaborate more on this issue.
| @@ -0,0 +1,340 @@ | |||
| Bipolar Electrodialysis (0D) | |||
| ============================ | |||
There was a problem hiding this comment.
Looks like you updated the terminology to acidic and basic channels. One other place where this should be updated is in the png diagram.
There was a problem hiding this comment.
Thanks for pointing that out. Somehow I missed that.
| ion and water transport across the membrane along with water splitting. Modelled transfer mechanisms include | ||
| electrical migration, diffusion of ions, osmosis, electroosmosis, and water splitting. The following are the key | ||
| assumptions made: |
There was a problem hiding this comment.
This sounds great. A question that comes to mind is -- what notable mechanisms does the model not account for at this time (but technically could be based on latest literature)? I suggest creating an issue to track any such mechanisms so that we chart a path for future enhancement, whether that be by you or another developer later on.
There was a problem hiding this comment.
@adam-a-a This is a good question. I actually faced the other problem. It was challenging finding the appropriate steady state equations for bipolar membrane (there are many equations/features for dynamic bipolar modelling but that's a can of worms for another day).
Even now there are some features (like the boundary layer details) that have not been added to the unit model because I can't find appropriate equations.
|
|
||
| * The **acidic** and **basic** side channels have identical geometry. | ||
| * For each channel, component fluxes are uniform in the bulk solutions (the 0-dimensional assumption) and are set as the average of inlet and outlet of each channel. | ||
| * Steady state: all variables are independent on time. |
There was a problem hiding this comment.
| * Steady state: all variables are independent on time. | |
| * Steady state: all variables are independent of time. |
| * **acidic** channel | ||
| * **basic** side channel |
There was a problem hiding this comment.
Is one called acidic and the other called basic or basic_side? I'd remove "side" if not used in the attribute name to avoid confusion.
|
|
||
| "Time", ":math:`t`", "[t] ([0])\ :sup:`1`" | ||
| "Phase", ":math:`p`", "['Liq']" | ||
| "Component", ":math:`j`", "['H\ :sub:`2` \O', 'Na\ :sup:`+`', 'Cl\ :sup:`-`', 'H\ :sup:`+`', 'OH\ :sup:`-`']" |
There was a problem hiding this comment.
It is not yet clear whether this unit works with MCAS, the NaCl prop package, both, and/or other property models. Can you elaborate?
There was a problem hiding this comment.
How do I check this? I took what the ED folks have been using and ran with it. So, theoretically whatever is applicable for the prop packages in the ED unit model should be applicable here (with the caveat that users should also supply H_+ and OH_- ).
| to fully solve the model. The exact degrees of freedom depend on the mode of operation. For the simplest case where no water | ||
| splitting occurs and the bipolar membrane acts like a simple electrodialysis memrbane these are: |
There was a problem hiding this comment.
| to fully solve the model. The exact degrees of freedom depend on the mode of operation. For the simplest case where no water | |
| splitting occurs and the bipolar membrane acts like a simple electrodialysis memrbane these are: | |
| to fully solve the model. The exact degrees of freedom depend on the mode of operation. For the simplest case, where no water | |
| splitting occurs and the bipolar membrane acts like a simple electrodialysis membrane, these are: |
| ion_dict = { | ||
| "solute_list": ["Na_+", "Cl_-", "H_+", "OH_-"], | ||
| "mw_data": { | ||
| "H2O": 18e-3, |
There was a problem hiding this comment.
| "H2O": 18e-3, | |
| "H2O": 18e-3, |
Note, you no longer need to supply MW for water anymore.
| }, | ||
| } | ||
|
|
||
| This model, by default, uses H\ :sub:`2`\ O as the solvent of the feed solution. |
There was a problem hiding this comment.
I probably should wait until I get to the code, but what happens when the user omits H_+ and OH_- from the solute_list? Would the unit model raise the appropriate exception? If so, and I may have missed it, you should add a note here stating that H2 and OH must be provided.
There was a problem hiding this comment.
Yes, the user needs to supply H_+ and OH_- . I have now added a note saying that it must be added.
…rrect error in water flux. Updated unit model and test (changes are minor)
…rrect error in water flux. Updated unit model and test (changes are minor) v1.1
…rrect error in water flux. Updated unit model and test (changes are minor) v1.2
…rrect error in water flux. Updated unit model and test (changes are minor) v1.2
|
|
||
| "Component mass balance", ":math:`N_{j, in}^{acidic \: or\: basic}-N_{j, out}^{acidic\: or\: basic}+J_j^{acidic\: or\: basic} bl=0`", ":math:`j \in \left['H_2 O', '{Na^+} ', '{Cl^-} '\right]`" | ||
| "mass transfer flux, basic, solute", ":math:`J_j^{C} = -t_j^{bpem}\frac{\xi i_{lim}}{ z_j F}`", ":math:`j \in \left['{Na_+} ', '{Cl^-} '\right]`" | ||
| "mass transfer flux, acidic, Water ions", ":math:`J_j^{C} = \frac{i - i_{lim}}{ z_j F}`", ":math:`j \in \left['{H^+} '\right]`" |
There was a problem hiding this comment.
I would annotate "Water ions" or replace them with "proton" and "hydroxide".
| :math:`mol\, m^{-3}` and are taken to be constants. :math:`f(E)` is the second Wien effect driven enhanacidicent of the | ||
| dissociation rate under applied electric field. It requires as input temperature and relative permittivity (:math:`\epsilon_r`). | ||
| To close the model :math:`\lambda = E_{crit} \epsilon_0 \epsilon_r / (F \sigma)` | ||
|
|
There was a problem hiding this comment.
I recommend tabulating these equations along others for easier following. It could be even better to tabulate them side by side with the catalysis-modeled scenario. I also recommend referencing them more precisely, i.e., annotating which eq is from which literature.
There was a problem hiding this comment.
@lbibl This is a very good suggestion. I have rewritten the doc file to tabulate the equations (and correct some errors).
|
|
||
| Bipolar electrodialysis, an electrochemical separation technology, has primarily been used to generate acids and bases | ||
| from waste salts produced in water purification. By providing in-situ access to valuable raw materials bipolar membranes can | ||
| significantly reduce the total cost of the operation. This cell stack is shown in Figure 1 with **basic** and **acidic** channels, that produce base and acid |
There was a problem hiding this comment.
This sentence somewhat puzzles me without specifying what is the "raw material" and what operation.
| .. csv-table:: **Table 2.** List of Degree of Freedom (DOF) | ||
| :header: "Description", "Symbol", "Variable Name", "Index", "Units", "DOF Number \ :sup:`1`" | ||
|
|
||
| "Temperature, inlet_acidic", ":math:`T^acidic`", "temperature", "None", ":math:`K`", 1 |
There was a problem hiding this comment.
Please check the rst formatting in the printout throughout. For instance, it should probably be T^{acidic}, etc.
|
|
||
| "mass transfer flux, acidic/basic, Water ions", ":math:`J_j^{C} = J_{diss}`", ":math:`j \in \left['{H^+, OH^-} '\right]`" | ||
| "mass transfer flux, acidic H\ :sub:`2`\ O", ":math:`J_j^{C} = t_w^{bpem} \left(\frac{i}{F}\right)+\left(L^{bpem} \right)\left(p_{osm}^CEM-p_{osm}^AEM \right)\left(\frac{\rho_w}{M_w}\right) - J_{diss}`", ":math:`j \in \left['H_2 O'\right]`" | ||
| "mass transfer flux, basic, H\ :sub:`2`\ O", ":math:`J_j^{C} = -t_w^{bpem} \left(\frac{i}{F}\right)-\left(L^{bpem} \right)\left(p_{osm}^CEM-p_{osm}^AEM \right)\left(\frac{\rho_w}{M_w}\right) - J_{diss}`", ":math:`j \in \left['H_2 O'\right]`" |
There was a problem hiding this comment.
Check formatting throughout.
…anup to remove unused performance metrics
…anup to remove unused performance metrics v1.1
Fixes/Resolves:
(replace this with the issue # fixed or resolved, if no issue exists then a brief statement of what this PR does)
Summary/Motivation:
Adding the bipolar membrane unit model capability
Changes proposed in this PR:
Legal Acknowledgement
By contributing to this software project, I agree to the following terms and conditions for my contribution: