This paper was presented as part of the student paper contest associatedwith the Annual Technical Conference and Exhibition.
The Petroleum industry for decades has relied on the acquisition and correctinterpretation of reservoir logs to provide important reservoir informationused to determine the "payzone" of the reservoir. One such logging tool isbased on the principles of Nuclear Magnetic Resonance (NMR), the alignment ofhydrogen proton spins with an applied magnetic field and the subsequent protonrelaxation observed when the external magnetic force is removed. This techniquehas become a useful non destructive and seemingly matrix independent method fordetermining pore size distribution in fluid filled pores. However, the presenceof paramagnetic material such as iron or manganese within reservoir rock, particularly sandstones, increases the proton relaxation rate and as a resultthe pore size distribution appears smaller and subsequent permeabilitycalculations will be erroneous.
The significance of this paramagnetic material and its effect on surfacerelaxation is the essence of this study. There were two series of syntheticcore samples created, one with a varying paramagnetic concentration andconstant grain size and the other with a varying grain size and constantparamagnetic concentration. NMR experiments were conducted using these coresamples to study the surface relaxivity effectThe NMR analysis of thesynthetic core samples has shown a strong correlation to NMR theory.
The presence of paramagnetic material, such as iron or manganese, within thereservoir rock found in the NMR logging tool's zone of investigation, has asignificant effect on the resulting Nuclear Magnetic Resonance (NMR) signature. Pore size distributions appear smaller and subsequent permeability calculationsare inadvertently underestimated in this situation. Previous studies (Roose etal. 1995 and Kenyon et al. 1994) have shown the relationship between surfaceparamagnetic concentration and relaxation rate to be linear. This fact, whilst not disputed, was arrived at by the use of aqueous colloidal sols, whereMn[2+] was adsorbed onto either silica or calcite particles. Previous researchhas been limited by the use of this type of porous medium and as such has notbeen able to quantify, through experiment, the surface relaxation effect asseen in the natural reservoir environment. This study used homogeneous, robust, paramagnetically-doped synthetic core, as a model system for naturallyoccurring reservoir core. These core samples were doped with knownconcentrations of paramagnetic material on the grain surface.