Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add model equations for simple Dead-End Filtration Model #5

Open
daklauss opened this issue Jul 2, 2024 · 8 comments · Fixed by #9
Open

Add model equations for simple Dead-End Filtration Model #5

daklauss opened this issue Jul 2, 2024 · 8 comments · Fixed by #9

Comments

@daklauss
Copy link
Member

daklauss commented Jul 2, 2024

In GitLab by @schmoelder on Jul 2, 2024, 15:22

In previous works, Tarek had written down some equations for a simple 0D-filtration model. Some of these equations were in a jupyter notebook, others already implemented in Python. Because I find it hard to keep track, I would like to ask you @d.klauss and @AntoniaBerger to consolidate them and write them down here in this thread (including all the assumptions and extensions). Then, we will use this to implement the model in the CADET-Python-Simulator.

@daklauss
Copy link
Member Author

daklauss commented Jul 5, 2024

In GitLab by @AntoniaBerger on Jul 5, 2024, 09:00

Simple model for Dead-End

A simple model for Dead-end filtration yielding the filtrate/permeate volume using Darcy's law can be:

$$\frac{dV_p(t)}{dt} = JA = A \cdot\frac{ \Delta p}{\mu \cdot R(t)} = A \cdot \frac{\Delta p}{\mu \cdot (R_{m}+R_{c}(t))}$$
With

  • $V$ volume of the filtrated collection (permeate) -> Answer to Question
  • $J$ Filtration flux (Flow rate per unit area)
  • $A$ Filtration area
  • $R_m$ resistance of the medium
  • $R_c$ resistance of the cake
  • $\Delta p$ pressure difference
  • $\mu$ dynamic Viscosity

Where the cake's resistance is:
$$R_{c}(t) = \alpha \cdot h_{cake}(t) = \alpha \cdot \gamma \cdot \frac{V_p(t)}{A}$$
With

  • $h_{cake}$ Height of the cake
  • $\gamma$ concentration of the suspended particles in the feed ( Should be revised ; [c] = mol / $m^3$; [ $\gamma_c$] = -)
  • $\alpha$ Specific cake resistance

$\color{red}{\text{Assumptions}}$:

  1. $\gamma$, $\Delta p$ constant
  2. $\alpha$ is constant $\Rightarrow$ incompressible cake.
  3. Constant $R_{m}$ $\Rightarrow$ no membrane fouling.
  4. Membrane has $100%$ efficiency $\Rightarrow$ all particles above certain size excluded.
  5. Single component
  6. No cake volume $V_{c} = 0$ $\Rightarrow$ $V_{f} = V_{p}$
  7. Perfect separation $V_c = V_s$; $V_p = V_l$
  8. $\gamma$ $\approx$ $c_{f}$. where $c_{f}$ is the feed's concentration.

@daklauss
Copy link
Member Author

daklauss commented Jul 5, 2024

In GitLab by @AntoniaBerger on Jul 5, 2024, 09:04

Model Extension with new Assumptions:

1. Cake has non volume zero and is time dependent:

  • $V_f = V_c + V_l$
  • $V_c(t) = h_{c}(t) \cdot A$
  • Has a individual volume flow
    $$\dot{V_c} = J A $$
  • With Carman-Kozeny we can add porosity of the cake:
    $$\dot{V_c} = A \cdot \beta \cdot \frac{\Delta P \cdot D^2}{\mu \cdot h_c \cdot K}$$
    With
  • Porosity factor $\beta = \frac{\varepsilon^3}{(1-\varepsilon)^2}$
  • Height of the cake $h_c = \frac{V_c}{A}$
  • $D$ the particle size

2. $\Delta p$ is time dependent

3. Multi component

4. Non perfect membrane

@daklauss
Copy link
Member Author

Wrote our simple model with constant flow $\dot{V}$ and every other constant = 1 as residualf and calculatet with ida. Because $R_c$ growth with $V$ linear, $\Delta p$ does too:
$$0 = \dot{V} -1$$
and
$$0 = -\Delta p \cdot A + \mu \left( R_m + \alpha\cdot \gamma \frac{V}{A} \right)\dot{V} $$

FigureSimpleDef

@schmoelder
Copy link
Contributor

Hey @daklauss,

how did you calculate this? Did you already integrate this into CADET-Python-Simulator or was this calculated "outside"?

@daklauss
Copy link
Member Author

Calculated outside, wanted to test sundials ida a bit to get confortable with residuals, and the solver itself. The simple model confuses me a bit. Because how it is written down above it doesn't say anything about the Seperated Volume $V_s$. Assumption 6. and 7. is a bit confusing because 6. states $V_c = 0$ and 7. states $V_s = V_c$. So $V_s = 0$. Does it mean this simple model is there to calculate preasure drop?

Hey @daklauss,

how did you calculate this? Did you already integrate this into CADET-Python-Simulator or was this calculated "outside"?

@daklauss
Copy link
Member Author

daklauss commented Jul 23, 2024

Simple Model for Dead End V2

Darcy's law provides a simple model for calculating the preasure Drop at a membrane:

$$\dot{V}_P = \Delta p \cdot \frac{A}{\mu} \cdot \frac{1}{R}$$

  • $\dot{V}_P$ flow of the Permeate through the membrane
  • $\Delta p$ pressure drow of the unit operation, so combined drop of cake and membrane?
  • $A$ membrane area
  • $\mu$ viscosity of our liquid
  • $R$ hydraulic Resistance of our Unit Operation

With $R$ as the combined Resistance of the Cake and Membrane:

$$R = R_m + R_c $$

  • $R_m$ Membrane resistance
  • $R_c$ Cake resistance

With $h_c = V_c / A$ we can describe the Cake Resistance as:

$$R_c (t) = \alpha \cdot h_{cake} (t) = \alpha \cdot \frac{V_c (t)}{A}$$

  • $h_c$ Cake height
  • $\alpha$ Specific cake resistance
  • $V_c$ Cake volume

We can split the Volume of the feed into two seperate parts:

$$ V_f = V_c + V_p \Rightarrow \dot{V}_f = \dot{V}_c +\dot{V}_p $$

  • $V_f$ Volume of the feed

Assuming our filtration efficency is $100$% $= 1$, our Cakevolume is:

$$ \dot{V}_c = \gamma_s \cdot \dot{V}_f $$

  • $\gamma$ Proportion of filtered material

Proportion of the filtered material is:

$\gamma_s = c^n _s \cdot \frac{M}{\rho ^m} = \frac{c^n _s}{\rho^n}$

or if Volume is zero:

$\gamma_s = V^n \cdot c^n _s $

  • $M$ Molar Mass of suspended material
  • $\rho^n$ Molar density of suspended material
  • $V^n$ Molar Volume of suspended material
  • $C^n _s$ molar concentration of suspended material
  • $\rho^m$ mass density of suspended material

Assumptions

  1. our filtration efficency is constant equal to $1$
  2. Our Filter does filter as soon as the stream enters the cake
  3. Single Component
  4. Constant $R_m$

Open Questions

  1. What about the viscosity? it should change inbetween $V_f$ and $V_P$ with Filtered particles

@schmoelder
Copy link
Contributor

  1. What about the viscosity? it should change inbetween $V_f$ and $V_P$ with Filtered particles

viscosity is a property of the fluid. As long as we do not discretize the cake, let's (for now) assume that vicosity is simply given by the feed stream.

$K$ weird konstant?

Where does the $K$ come from?

@daklauss
Copy link
Member Author

As far as if looked for reference $K$ is described as "an empirical constant which depends on bed tortuosity."

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging a pull request may close this issue.

2 participants