Fluids

Why is the mass transfer rate zero, and is there a diffusion process?

• h395523899
Subscriber

Why is the mass transfer rate zero, and is there a diffusion process?

Is this diffusion caused by convection?

What principle?

• raul.raghav
Subscriber

Convection and diffusion are different mass transport phenomena. Convection describes the bulk downstream transport of the fluid with a certain velocity u; Diffusion is a mass transfer phenomenon which occurs due to the random molecular motion (Brownian motion) in order to achieve an equilibrium state of uniform spatial distribution. The general transport equation is given by

So make sure you frame your question correctly and carefully. And moreover, it would very difficult to comment on your issue without knowing what you're trying to do or what physics you are trying to model and how you're modeling them. Could you provide us more information about that?

• h395523899
Subscriber

First of all, thank you for answering my question.

I want to simulate the diffusion of ions in the presence of low concentrations of salt water in place of high concentrations of salt water.

I use an unresponsive species transport model, laminar flow, unsteady state calculations

When I set the mass transfer coefficient of the ionic material to zero, what happens between the two liquids is shown in the figure.

I think the software also simulates convection.So I want to understand whether the principle of software simulation convection is in line with the situation I want to simulate.

This picture shows the ion mass distribution cloud

I think that if the mass transfer rate is zero, there should be no intermediate transition area

• raul.raghav
Subscriber

Can you attach a schematic of your geometry with the boundary conditions you specified? Species transport model with laminar flow sounds like the best approach for the problem. How did you define the salt water in Ansys Fluent? And how did you define the different concentrations at different regions of your geometry?

• h395523899
Subscriber

The 2d model is 10m long and 20cm wide

Entrance speed 0.1m/s

Pressure outlet 0

Do not consider the temperature

What I am simulating is potassium chloride solution

The concentration is converted to a mass percentage by the degree of mineralization. For example, 1000 mg/L can be considered as a 1000 mg/1000 g potassium sulfide solution.

I set the concentration of the incoming liquid, and then set the concentration of potassium chloride solution in the container

However, there is no potassium sulfide solution in the material library, so I replaced it with hydrogen ions first.

• h395523899
Subscriber

• h395523899
Subscriber

I seem to have found my problem. I have set the speed too high. The actual fluid is already turbulent, so it causes diffusion, although the mass transfer coefficient is zero.

I reduced the speed, controlled the Reynolds number within the laminar flow range, and the mass transfer rate was 0, and there was no diffusion between the liquids.

One problem here is that if the boundary conditions I set are already turbulent, then laminar flow is selected for simulation. What will happen? Does the program simulate the turbulence? Or does it simulate a bunch of wrong results according to laminar flow?

• raul.raghav
Subscriber

I thought I understood what you were trying to model but I'm again confused reading your last post. Would you able to answer the following questions?

1. You mentioned that you were trying to model "the diffusion of ions in the presence of low concentrations of salt water in place of high concentrations of salt water". The transfer of the species due to a concentration gradient is a diffusion process. In order to simulate this, you would have to make sure that diffusion dominates. And you can achieve this by setting the inlet velocity very low.

2. When you say you set the mass transfer coefficient zero, what do you mean by it? Is it because you set the mass diffusivity zero?

3. If the velocity is high, convection dominates and the effect of diffusion would small. And if your aim is to model diffusion, you would have to reduce the velocity (depends on your problem) such that the convective effects are small. And by setting the appropriate diffusion coefficient at low enough velocities, you can model diffusion. And again, for diffusion, you would need to have regions of different species concentration.

4. Laminar model numerically solves the Navier Stokes (NS) equation while the turbulence models solve the Reynolds Averages Navier Stokes (RANS) equation. I'm just talking about RANS turbulence models here. There are other turbulence models like LES, DES which I'm not mentioning here just to keep things simple. So if you model a turbulent flow using the laminar model, it would not be able to capture/estimate the fluctuations in the flow which is caused due to turbulence. A turbulent flow has eddies of various length scales and timescales which needs to be resolved or modeled by using turbulence models.

And if you want to learn more about turbulence, I can suggest you some resources which would be of help.

• h395523899
Subscriber

1. Yes, I just want to simulate this. Let proliferation dominate and convection can be ignored.

2. I set the mass transfer coefficient to zero to see if the convection has any effect on the simulation. After I set it to zero, I discovered that a concentration gradient band was still produced. Because the mass transfer coefficient has been set to zero, I think it is due to convection. So I want to know what principle or equations the software uses to simulate convection. I change my boundary conditions through principles or equations to reduce the impact of convection.

3. Later, I found that the speed I set was 0.1m/s would cause turbulence, so I changed it to 0.001m/s. It was found that the impact of convection basically disappeared. I used the patch function on the initial interface to set the solution ion concentration in the vessel to a low concentration of 0.005, and the inlet fluid ion mass coefficient was set to 0.02. I do not know if this is right.

4. I mentioned turbulence and laminar flow. I wanted to know what the software would do if it used a laminar flow model to model the turbulent flow. By your answer I know that it will simulate a bunch of wrong results.

5. In the previous question and answer, I mentioned a lot of the word “proliferation.” But some “proliferation” was just to describe the phenomenon. My narrative was not very accurate and it caused you inconvenience. I am very sorry, so in (2) I describe this phenomenon as a "concentration gradient band," which may be caused by diffusion or by convection.