Rich gas injection pilot: an enhanced oil recovery approach applied to an unconventional reservoir in the Bakken petroleum system

Abstract

Economic production from shale plays focuses on hydraulic fracture techniques and well spacing optimization and yet the recovery factor ranges from 7 – 10 %. Shale plays produce large quantities of ethane and liquid rich hydrocarbon gases. Ethane, like CO2, has a low minimum miscibility pressure, has good solubility in oil, is better at mobilizing higher molecular weight hydrocarbons, and is a good viscosity reducer. The purpose of this study is to determine the technological and economic factors that could affect a rich natural gas injection pilot. The design of the study is fivefold: 1) build a dual-porosity reservoir model of the Middle Bakken (MB) formation in the West Nesson section of the Williston Basin; 2) assess the pre-injection technological factors that may contribute to difficulties or success of the pilot; 3) determine surveillance techniques that will add value and understanding to post-injection studies; 4) inject rich natural gas at different pressures and compositions into the stimulated reservoir volume of one well, to determine the incremental production of oil and/or compositional change (modified huff and puff); and 5) compare the modeling results to better understand which scenarios (injection pressure, injection composition, or both) create the best value for designing a rich natural gas injection project. This study is part of a pilot that is currently underway in the West Nesson section of the Williston Basin. The study area is a 1,280 acre drill site unit (DSU), which has 11 horizontal wells and initial production dating back to August 2013. The DSU is a low permeability, fractured reservoir containing liquid-rich hydrocarbons in the Middle Bakken (MB) and Three Forks (TF) formations. The geologic static model was built using contour maps, multiple core analysis, NMR, rock mechanics and standard well logs. I used the static geologic model to create a 3D reservoir model using petrophysical, core, PVT, and completions data. I created an equation of state (EOS) model using PVT, which was used to develop the 7-lumped component compositional model in CMG. History matching was performed in two sets, first the total fluid to obtain a reasonable material balance match, followed by matching oil, gas, and water phases. First, the study evaluated how matrix/fracture permeability, fracture porosity, and hydraulic fracture permeability impact production. The vertical to horizontal permeability ratio indicates how matrix/fracture water saturation impacts phase history matching. Fracture porosity is a significant sink for fracture to matrix flow in numerical simulation. Individual grid block size and calculated properties impacts oil recovery rate and historical saturations within the fracture system. Next, the study evaluated how injection/production cycles applied to a one well SRV affected incremental oil and cumulative production. Five cases varying pressure and composition were compared to a base case to determine the incremental oil increase or decrease. The pressure cases showed how increasing pressure increases the post injection cycle incremental oil and cumulative production. Changing the composition verified that injecting high quantities of methane decrease the incremental oil, whereas as injecting ethane and liquid rich hydrocarbons increased the incremental oil and cumulative production. Minimum miscibility pressure, hydrocarbon solubility in oil, and viscosity reduction of injectant are key properties needed to release residual oil within the matrix and increase the recovery factor. Finally, the study evaluated how low capital exposure paired with abundant quantities of liquid rich hydrocarbons can be effectively used to increase recovery factor and ultimately provide a high return on investment.