| Time: |
February 21, 2017 |
| Lecturer: |
Kata Kurgyis,
Department Petroleum Engineering, Montanuniversität Leoben, Austria |
| Venue: |
Pfaffenwaldring 61, Raum U1.003 (MML), Universität Stuttgart
|
| Download as iCal: |
|
Many studies indicated that recovery of crude oil by water flooding (in secondary or tertiary recovery mode) can be improved by simple and cost effective means through modifying the ionic composition of the injection water. Low-salinity water flooding and alkaline flooding belong into this family of cost effective IOR/EOR methods. Both methods act differently - while low-saline water modifies essentially the wettability of the rock, the main EOR mechanism of alkaline flooding is to create and mobilize surfactants in situ in order to lower the interfacial tension between the injection water and the oil.The mechanism governing Low Salinity Effects (LSE) during water flooding is believed to be a combination of several physicochemical processes like multi-component ion exchange (MIE, Lager, 2006) and double layer expansion (DLE, Ligthelm et al., 2008). These fundamental LSE are introduced in 4 mechanistic and phenomenological models proposed in the work of Kuznetsov et al. (2005).The objective of the here presented research is to develop the numerical capability for an explicit description of water-based IOR/EOR processes on the continuum-scale. The aim is to use the model for the interpretation of core flood experiments and to serve as basis for upscaling. In this context, an explicit description means to implement the change of wetting properties due to the composition of the injected water by a set of chemical processes linked to concentrations and to measured and interpolated flow properties (relative permeability and capillary pressure).The relevant processes captured in the mechanistic and phenomenological models are implemented in the open-source flow and transport simulator, DuMux (DUNE for Multi-Phase, Component, Scale, Physics...). In the first simplified versions of the phenomenological model, the changing wetting conditions are linked to a single dissolved salinity tracer in order to relate the alteration of fluid-rock interaction properties to a varying salinity, then to a single adsorbed tracer for the purpose to introduce adsorption.The first mechanistic model suggests 2 reactions, one showing monovalent and one displaying divalent ion adsorption to rock surface. In the second mechanistic model the reaction set is extended with an additional reaction in order to capture the relative contribution of divalent ionic species to MIE and DLE.
