| dc.description.abstract | Two main goals are pursued in this research, firstly a better correlation between two grain size measuring devices, for the mud content specifically has been established. Secondly, a grain size based work flow, developed by Daza, 2012 at TNO, to calculate permeability and capillary seal capacity of mudstones, has been calibrated, extended and improved. First of all, to obtain a better correlation, the first data set was applied, containing samples of the Rupel Clay Formation, Netherlands, and a core derived from the Belgian Boom Clay. The Belgian Boom Clay has been a target for nuclear waste disposal for decades and recently the Rupel Clay Formation has been targeted for the same reasons. These two formations are basically the same formation. Clay formations have a high permeability which is highly preferred with nuclear waste storage. The issues with the devices is that in the Netherlands, the laser diffractor is applied while in Belgium the sedigraph is used and the results are not directly comparable. It has been shown that the laser diffractor underestimates the clay fraction of the sample with respect to the sedigraph. Therefore a first step in this research is to calibrate these two grain size results. Grain size analysis of this data set has been carried out by both measuring devices and in total 16 boreholes in the Netherlands have been investigated and one core was derived from the Belgian nuclear waste disposal project. The depth of these samples ranges from 20-736m, indicating that some of the samples are of very shallow depths. In the literature, the clay fraction results should be comparable when for the laser diffractor a grain size of 8 μm is applied and for the sedigraph 2 μm. In this research a new correlation has been established at 5 μm for the laser diffractor compared to 2 μm with the sedigraph. With this improved correlation, the workflow can be continued. A next step is to integrate the improved correlation into the workflow. The workflow starts off with calculations based on formulas developed by Yang and Aplin. They established equations relating pressure coefficients to porosity and permeability, based on the clay fraction of the sample. This part of the workflow has been calibrated with a second data set, completely derived from 4 boreholes of the Boom Clay, in a synthesis research. These relationships are however for homogenous mudstone samples. Therefore to improve the workflow, heterogeneity calculations, based on Drews relationships have been implemented. Drews developed in his PhD thesis several equations relating effective stress to porosity and eventually porosity to permeability. All relationships are developed for a specific clay fraction ranging from 50% clay to 100% clay. A fourth step in this research is to select a third data set. This time a closer look was taken at the muddy cap rocks overlying the shallow gas fields, offshore northern Netherlands. The samples were derived from the A and B blocks. In total 4 cores and 49 cuttings were carefully selected in the cap rocks above the gas accumulations in order to have calibration points. The depth of these samples ranges from 312-1044m. In the sedimentological lab at the VU, these samples were analyzed by laser diffraction. These samples were implemented to verify the workflow and the calculations were carried out according to the new correlation for the clay fraction. The last step in this research is to extend the workflow, therefore 2 alternative pore throat calculation methods have been implemented. The first method is the equivalent grain size method by Nakayama and Sato. They converted pore throats into grain size by using a theoretical relation of COEF, which is the ratio of pore throat to grain size. Then this grain size is used to evaluate top seal capacity. The second alternative pore throat calculation method is the critical grain size method of d10 method, by Nordgård Bolås et al. Here the size of the critical pore throats is important to a cap rock's seal capacity. When this workflow is carried out, there will be 12 calculations of seal capacities, 6 of these seal capacities will have the heterogeneity incorporated and 6 will assume homogenous samples. In order to calibrate the computed seal capacities with the shallow gas fields, an interpretation will be made of seismic images indicating gas columns and NPHI-RHOB cross-overs, depth contour maps, and production plans. This results in the selection of three methods that render the most accurate seal capacities, two of them are based on the integration of the equivalent grain size method with and without heterogeneity calculations; the other on the critical grain size method. | |
