Date of Award

May 2026

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Petroleum Engineering

First Advisor

Andreas Michael

Abstract

Reactive-transport phenomena during CO2 injection in carbonate reservoir were examined in detail. Three core samples from the primary producing units of North Dakota’s Upper Red River Formation (U2-H055, U4-H069, and U6-H009) were used to experimentally investigate geochemical reactions and petrophysical alterations, impacting carbon dioxide enhanced-oil-recovery (CO2-EOR) and geologic-carbon-storage (GCS) projects. During corefloods using a CO2/brine mixture, two samples demonstrated early permeability increases followed by declines, whereas the third sample demonstrated an early decrease; all samples gradually recovered at a later stage. Nuclear magnetic resonance (NMR) measurements revealed slight porosity increases due to dissolution-induced wormholes, while the pore-size distribution (PSD) shifted toward smaller pores, because of mineral precipitations and particle-induced pore blockages. X-ray diffraction and fluorescence (XRD/XRF) analyses confirmed dolomite and calcite dissolutions and precipitations following CO₂/brine flooding. Dissolutions predominantly occurred during the early injection stages, as indicated by increased calcium (Ca²⁺) and magnesium (Mg²⁺) concentrations in the produced water, measured by inductively-coupled-plasma mass spectrometry (ICP-MS). These effects were concentrated near the core inlets, resulting in wormholes creation and particle detachments, observable at the pore scale by scanning-electron-microscopy (SEM) images and energy-dispersive X-ray spectroscopy (EDS). Conversely, mineral precipitations at the core outlets were reflected by decreases in the Ca²⁺ and Mg²⁺ concentrations in the produced water, also confirmed through SEM and EDS. Furthermore, core-scale compositional models coupled with a reactive-transport models were developed for each of the three core samples to investigate the impact of geochemical reactions and petrophysical alterations on CO₂-EOR performance. The developed models were history matched against analytical oil recoveries predicted by the Buckley-Leverett theory before being used to forecast outcomes of CO₂-EOR application. Model validation was achieved by matching ionic concentration variations in the produced water, mineralogical changes, and pressure responses observed during coreflood experiments. CO₂-induced carbonate dissolution and precipitation exert primary controls on permeability and, consequently, overall CO₂-EOR performance. Reactive-transport simulations, compared to corresponding base-case (constant-permeability) scenarios, illustrated that mineral dissolutions in sample U2-H055 enhanced sweep efficiency, increasing oil recovery to ~68% versus ~63% if these dissolutions are neglected. In contrast, samples U4-H069 and U6-H009 exhibited ~3% lower oil recoveries due to precipitation-induced permeability reductions (via pore blockages) and earlier CO₂/brine breakthroughs. At field scale, the history-matched compositional model was used to forecast incremental CO₂-EOR production in a sector of the reservoir that includes four existing wells, incorporating laboratory-derived dynamic-permeability variations. Results differed from corresponding base-case (constant-permeability) simulations, with predicted differences of 0.586 MMSTB in oil and 344 MMscf in gas, representing incremental recovery changes of 1.25% and 6.14%, respectively, over the final three years of the five-year simulated injection period. Reduced permeabilities led to lower oil recoveries, highlighting the strong influence of dynamic-permeability variations on the overall CO₂-EOR performance and the importance of its inclusion in field-scale development assessments.

Share

COinS