A mineral solver is an algorithm that tries to find the proportions of already assumed minerals present in the rock. Several mathematical methods have been tried historically, from linear system of equations, to artificially trained neural networks, and even empirical heuristic approaches. Linear system solvers require a set of good quality log curves, and not too many minerals to solve, otherwise the problem would be mathematically undetermined. Mineral solvers offer only a coarse guide to understand the rock composition. Nothing replaces the experience of qualified geologists and petrophysicists.
The GeolOil Petrophysical Functions module has mineral solvers for carbonate systems with limestones, dolomites, anhydrites, and halite salts (NaCl). GeolOil also solves clean eolian systems with sandstones and dolomites.
The figure below shows the GeolOil mineral solvers panel to compute the proportion of dolomites
No mineral solver is perfect. Curves may be noisy, unreliable, and even if the log curves are perfect, many other not considered minerals might be present. In the well log plot shown below from a Montana carbonate reservoir, GeolOil achieved very good results computing the mineral proportions of limestone, dolomite, anhydrite, and salt.
The rightmost thin bar, with colors grey (anhydrite), light green (limestone, and salt), and dark green (shaly dolomite), is the interpretation manually made by an experienced geologist already familiarized with the depositional environment. The Minerals track on the right, shows the result of the mineral proportions solved by GeolOil from the PEFZ photoelectric factor, and RHOZ bulk density curves. The pink are filling represents the proportion of dolomite, indigoblue represents the proportion of limestone, orange the proportion of anhydrite, green the proportion of salt, and blank white represents the zones for which GeolOil could not find a stable or reliable solution. Notice that there are blends limestones and dolomites, and the overall match is good.
The figure below shows a well log plot result of a mineral solver for limestone, dolomite, anhydrite, and salt
GeolOil provides also a QuartzDolomite solver for eolian environments. It is particularly useful to estimate the matrix density for partial blends of quartz and dolomites:
The figure below shows the mineral solver panel for a sandstonedolomite eolian environment to compute matrix density
The figure below shows a mineral solved log plot for a sandstonedolomite eolian environment
GeolOil Petrophysical Functions module also ships with a clays solver for clastic environments, aimed to solve approximate proportions of quartz, silt, illite, smectite, kaolinite, and chlorite. It offers mostly semiquantitative analyses of the clay minerals present. Even "hard lab" techniques like XRD X Ray Diffraction are also semiquantitative.
The figure below shows the mineral solver panel for clays.
✔ NOTE: The GLOG file workflow for this log is available for download with the set of optional interpretation examples.
The figure below shows a mineral solved log plot for a clay minerals in a clastic environment
Another mineral solver shipped with the GeolOil Petrophysical Functions module is a clayscarbonate mineral solver, aimed to solve approximate proportions of clays, silt, quartz, calcite, and dolomite. It is usually more accurate than the former clays solver, (clays mineral composition is always very difficult to separate, but GeolOil does a useful qualitative work on this) so the proportions found can be more similar than those found via XRD diffraction.
The figure below shows the mineral solver panel for clayscarbonates.
✔ NOTE: The GLOG file workflow for this log is available for download with the set of optional interpretation examples.
The figure below shows a mineral solved log plot for clays and carbonates. Notice the reasonable match with XRD lab points
Related article:
