Predicting the Watertightness of a Membrane
Generation of a three-dimensional model of a membrane and analysis of morphological properties
The requirements for waterproof materials for everyday products such as waterproof sportswear and building materials or in medical care areas such as protective clothing and work equipment are becoming increasingly complex. However, application conditions and performance requirements vary considerably. The highest water pressure before penetration of such materials is usually measured in elaborate tests. Simulations on material structures at the micro and meso scale in GeoDict can significantly reduce this experimental effort.
Various conditions and environmental influences can be represented in GeoDict. For example, capillary pressures can be calculated using various pore morphology methods. For this purpose, in addition to the normal pore morphology method, the dynamic pore morphology method can be used to evaluate the distribution of wetting and non-wetting phases for a given (quasi-stationary) capillary pressure and thus determine saturation. The capillary pressure curve is obtained by repeating this calculation for a variety of capillary pressures.
The application workflow in overview:
- Generate grains based on a FIB-SEM image or a µCT scan.
- Modeling the membrane by adding fiber structures and binders.
- Calculate the water saturation in the membrane with increasing water pressure.
What was the result?
- Using different structure generators, a waterproof membrane could be simulated reliably in GeoDict.
- The simulation results contain detailed information about the water permeability of the membrane depending on the water pressure.
- The saturation simulations enabled to identify the limits of the membrane technology.
The modeling of the desired membrane is shown step by step with different modules of GeoDict. For this purpose, first the grains were created and then the fibers and the binder were added to the structure. The exact materials as well as parameters, such as the fiber orientations, can be defined for each component.
- FiberGeo contains a selection of more than 25 different fiber types, which can be customized.
- Properties such as the fiber diameter or the exact orientation and distribution of the fibers can be precisely defined.
- Material properties can be assigned to each component individually. For example, multiple fiber types are also possible in one model.
- The generated structure can be represented true to the original with the help of binders and in line with industrial production.
Generate grains with GrainGeo
Many different structure types can be generated in GeoDict. In this example, grains are generated with GrainGeo at the beginning, which serve as nodes of the added fibers in the further course. A variety of properties of the grains can be defined. In addition to the exact shape, type and orientation of the grains, the distribution and the extent of the overlap of the grains can also be defined.
Add fiber structures and binders with FiberGeo
In the next step, fiber structures are added to the already created grains in order to represent the membrane faithfully.
Subsequently, the various components must be bonded together. This is achieved by adding binder. Also here, the ratio, the contact angle and the distribution of the binder can be selected individually. Once all the components have been joined together accordingly, a waterproof membrane is obtained, which can now be tested and improved through various simulations in GeoDict.
After the membrane has been reconstructed using different structure generators, various properties of the membrane can then be investigated. Since in this case the membrane is waterproof and consists of a porous material, the water permeability was investigated in relation to the water pressure.
With SatuDict, different methods can be chosen to simulate the pore morphology in order to determine the distribution of the two phases within the porous medium. The dynamic pore morphology method used here allows a dynamic simulation of the drainage and imbibition processes, even for non-monotonic capillary pressure profiles. Thus, the capillary pressure can drop during a drainage simulation when the non-wetting phase passes a pore constriction (bottleneck pore).
In this simulation, non-monotonic capillary pressure gradients are used, resulting in several intermediate steps being calculated for the same pressure value to avoid immediate filling of the entire structure.
At the beginning of the drainage process (of air), when the initial pressure is low, the porous structure is by 100% saturated with air. As the pressure increases, the membrane initially withstands the water and even at a water pressure of over 2 bar, the membrane remains watertight. If the pressure continues to increase and reaches a value of 3.6 bar, the water begins to penetrate the structure, displacing the air from the structure. At the end of the drainage process, the structure is by 71% filled with water and the remaining 29% of the structure is still saturated with air.