Errors in X1t simulation

[Zhidi Wu]

Hello,

I’m setting up a reactive transport model based on my flow-through experiment in X1t and keep getting two types of errors. First error states: Porosity of node 0 is too small: -0.670590. Second error states: Mineral volume in node 0 is greater than node volume. In my experiment, fluid is injected into a core sample with a bulk volume of 57 cm3, porosity of the core sample is 0.0885. The mass of each mineral composition of the core sample is known. Injection flow rate and the pressure drop across the sample is also known. Here are the questions I have that I think might be useful to trouble shoot those two errors I mentioned above.

1.        Does X1t model consider the total injection fluid volume as an input parameter? I noticed that there is no fluid volume option, so is the fluid volume basically the pore volume when the domain is saturated, which can be calculated from domain volume and porosity?

2.        If the injection fluid volume is not directly defined, but inferred by time and pressure drop, should I set 57 cm3 as the domain volume in Domain tab (from x, y, z)? In my batch model using React, I had to scale the bulk volume and mineral mass based on injection fluid volume (1000 ml H2O).

3.        In Medium tab, is the porosity in X1t model the same as the porosity of my core sample (0.0885)?

4.        In Reactants tab, when I define the mass for minerals, are they the mass within each nodal block, or for the entire domain? I set 10 nodal blocks in my model, which means I need to divide the total mass by 10 if it’s for each nodal block.

 Please let me know if you need to know more information to troubleshoot these errors.

Thanks,

Zhidi

[Jia Wang]

Hello Zhidi,

1 and 2. Section 2.9 Boundary conditions and 2.10 Reaction intervals in the Reactive Transport Modeling guide explains how to set the fluid composition that you would inject at the boundary and the time length for how long you set the injection to last for. In section 3.2 Setting flow rate specifically informs how you can set the flow. You can set discharge directly but X1t also accepts pore_volumes flushed over the time length of your simulation. The pore space is considered as the saturated volume of the domain. In reactive transport models, the bulk volume of the domain is set by what you provide in the Domain tab. Unlike single-node apps, the bulk volume in a reactive transport model is determined by the domain size.

3. Please refer to section 2.12 Porosity evolution to see how this is handled in RTM apps.

4. Please see chapter 2.11 Kinetic reactions and gas buffers about setting reactants. If you are adding a Simple reactant, this is titrating in a set amount of mass over the course of the simulation, which is the same as React. If you are adding a kinetic mineral, then the mass or volume of the mineral set is in your initial domain. You can go about this in two ways, using either a relative unit (e.g. volume %) or absolute unit (e.g. mg) for each nodal block. Please section 2.7 Initial conditions in the same user guide to see how different types of units affect the way your problem is set up.

Before jumping into the reactive transport modeling applications with your own problem, I think it is best to familiarize yourself by examining the examples provided with the guide. A good one to start would be the example of rainwater infiltering a quartz aquifer in section 3.6 of the RTM guide. Another good example is Chapter 3.8 Example: weathering in a soil profile.

Hope this helps,
Jia Wang
Aqueous Solutions LLC