Carbonate reservoir rocks of the Najmah formation in Kuwait, with low porosity and low permeability, have been characterized using integrated digital and physical rock analyses methods. High-resolution imaging and analyses determined the microstructural characters of mineral matrix, organic matter (OM) distribution, organic and inorganic pore types, size distribution, and permeability variation within this kerogen-rich Late Jurassic stratigraphic unit.
Considerable heterogeneity of porosity and permeability was observed in the 100-ft studied interval of the Najmah Formation. Two-dimensional scanning electron microscopy (2D-SEM) imaging and three-dimensional focused ion beam SEM (3D-FIB-SEM) imaging highlighted the different types of porosities present within the formation rock. At each depth, several 2D-SEM images were used for characterization and selection of representative locations for extracting 3D FIB-SEM volumes. The 3D volumes were digitally analyzed and volumetric percentages of OM and total porosity were determined. The porosity was further analyzed and quantified as connected, nonconnected, and associated with organic matter. Connected porosity was used to compute absolute permeability in the horizontal and vertical directions in the area of interest.
Porosity associated with OM is an indicator of OM maturity and flow potential. It has been categorized as pendular type, spongy large grain, spongy small grain, fracture porosity within the OM, grain boundary fractures and intergranular porosity covering the entire OM. Permeability is not only influenced by porosity within OM or even apparent transformation ratio (ATR), it is also dependent on pore connectivity, pore sizes, and heterogeneity (e.g., high-permeability streaks). For high porosity samples, almost all pores are connected and contributing to permeability. For low porosity samples with high permeability, the flow is mainly through microfractures. It is possible that intergranular clay pores in highly thermally mature rocks were originally filled with OM and that, during progressive thermal maturation, transformation of OM to hydrocarbon(s) removed much of the pore filling OM.
It has also been observed that, although the total organic carbon (TOC) content of the rocks is significant (up to 18 wt%), and good maturity index (VR0>1), only few examined samples show good connected porosity within the OM. It is essential to evaluate the porosity within the OM thorough high-resolution measurements for pinpointing the prospective layers for future stimulated horizontal wells in this organic-rich source unit. These intervals can be considered as the potential sweet spots after integration with detailed petrophysics and geomechanical parameters for optimized well planning and completion design.
Reservoir pressure and stress behaviors are simulated during elapsed time between sequential fractures in multistage and multipad hydraulic fracturing scenarios to analyze interference between fractures and to manage potential fracture given interactions. Commonly current implemented completion techniques such as conventional sequential fracturing (CSF) and modified zipper fracturing (MZF) are investigated. Finite element analysis is implemented to simulate time dependent post-fracture behavior of formation pore pressure distribution and stress magnitude in multistage and multipad fracturing jobs in horizontal wellbores. A coupled stress-displacement to hydraulic diffusivity modeling technique was implemented to estimate formation pore pressure distribution and to account for the effective in-situ stresses in the near fracture region. Governing equations, methodology, and boundary conditions are provided in detail to allow readers to implement of this technique.
Completion techniques rely on the principle of elapsed time between consecutive fractures as a mean to minimize interference between fractures optimizing resources at the same time. Post-fracture reservoir pressure and stress time-dependent behaviors within the stress shadow area are discussed. These parameters are computed at several observation points inside the reservoir such as at the tip of the fracture, away of the fracture, and at the parent horizontal wellbore. This analysis provides unique insight of scenarios occurring during sequential fracturing jobs. Permeability of rock matrix is a controlling factor of the diffusion processes occurring post-fracture. It was found that unconventional shales exhibiting extremely low permeability values show low diffusion of hydraulic fracturing fluids from the induced fracture to the formation. For this reason, disappearance of stress shadow effects in the near induced fractures region are not being allowed during elapsed times between sequential stages executed in the named completion techniques. These findings are relevant because inadequate management of elapsed time does not favor the subsequent induced fractures to grow parallel to the previous one due to changes of direction of stresses. When interference becomes critical, the phenomenon identified as "Frac-Hits" can also be predicted; therefore, managed using this tool.
Lou, Xuanqing (Pennsylvania State University) | Chakraborty, Nirjhor (Pennsylvania State University) | Karpyn, Zuleima (Pennsylvania State University) | Ayala, Luis (Pennsylvania State University) | Nagarajan, Narayana (Hess Corp.) | Wijaya, Zein (Hess Corp.)
The design of oil recovery processes by gas injection or vapor solvent relies on knowledge of diffusion coefficients to enable meaningful production predictions. However, lab measurements of diffusion coefficients are often performed on bulk fluids, without accountability for the hindrance caused by the pore network structure and tortuosity of porous media. As such, our ability to predict effective diffusion coefficients in porous rocks is inadequate and, additional laboratory work is needed to investigate the impact of the medium itself on transport by diffusion. In addition, experimental data on multi-phase diffusion coefficients are particularly scarce for tight rocks. This study therefore proposes an experimental methodology, based on a pressure-decay technique, to measure diffusion of injected gas in oil saturated porous rocks. A diffusion experiment of gas into bulk oil (without porous medium) provides an upper limit estimation of this gas-liquid diffusion coefficient. Diffusion experiments using limestone and Bakken shale provide insight into different degrees of restriction in high permeability versus low permeability media. Two analytical models and one numerical model were implemented and compared to determine the diffusion coefficients from the time-dependent experimental pressure-decay data. These diffusion coefficients were found in agreement with literature on corresponding data, demonstrating the validity of the modeling approaches used. Results indicate considerable hindrance to diffusion in porous media relative to bulk oil and relates to the tortuosity and constrictivity of the rock matrix. The diffusion coefficient of methane in bulk oil is 3.8 × 10−9 m2/s. In our limestone sample, this diffusion coefficient drops by one order of magnitude, ranging between 1.5 to 6.5 × 10−10 m2/s and, it drops by another order of magnitude in the Bakken shale sample to 2.0 × 10−11 m2/s.
The efficiency of miscible gas flooding relies on the mass transfer rate of species from one phase to another. Determining the diffusion coefficient of a gas-oil system is a critical factor to characterize this transfer phenomenon between gas and oil. Many gas-oil systems are being investigated to measure the diffusion coefficients, losing sight of the hindrance effect under reservoir conditions due to the pore network structure of porous media. Therefore, effective diffusion coefficients in porous media, which have more practical guiding significance, need further investigation. In this study, an experimental methodology based on the pressure-decay technique is proposed for capturing the gas-oil diffusion signals in porous rocks. A numerical model developed by our research group is implemented for estimating all diffusion parameters under different conditions. Results demonstrate the feasibility of this numerical simulation model. Experimental work and numerical simulation were performed on three different scenarios: gas-liquid diffusion in bulk Bakken oil, oil-saturated limestone and Bakken shale samples. The results present quantitative insight into the sensitivity of the diffusion coefficients to the degree of porous media restriction. Moreover, effective diffusion coefficients respond to pore network tortuosity, where more tortuous pore networks provide stronger hindrance effect on the molecular diffusion process. According to the results, this study can provide a technical support for correlating the effective diffusion coefficient to the relevant properties of rocks, which is helpful for establishing more accurate simulation models when the effective diffusion coefficients are missing.
The main goal for an operator developing an unconventional reservoir project is to maximize NPV per acre by optimizing its completion strategy. This can be achieved by applying a comprehensive approach that accounts for key well treatment controlling parameters, their impact on the future production performance, and economic uncertainty. In this work, we developed and applied a workflow to explore the impact of various completion parameters and determine the completion strategy with the maximum economic gain.
The workflow integrates petrophysical well log and core data, along with PVT lab experiments with normalized permeabilities calculated from microseismic attributes to initialize the reservoir model. The reservoir model is then calibrated using actual field data to generate a history matched model. Since this model is developed based on microseismic data and represents a realistic network of fractures created during stimulation, it can be further used to analyze the impact of main completion parameters, well spacing and configuration, on the production performance of the wells.
The workflow is applied to three wells drilled in a gas reservoir in the Marcellus Shale. Because abundant field data were available, we can be certain that the calibrated reservoir model accurately matches the reservoir behavior. Detailed analysis of the reservoir model shows the presence of undepleted zones which indicates the current well spacing is too wide. However, the frac hits recorded through microseismic monitoring and pressure interference with nearby wells suggests a tighter well spacing will result in energy loss and over-stimulation. Therefore, an economic analysis is used to evaluate the various well spacing and configuration scenarios and their implications in terms of cost-benefits.
Various well spacing scenarios are created for the original and the proposed chevron pattern well configurations. For each scenario, the EUR, NPV per well, and NPV per acre are calculated to represent maximum gas production, the overall profitability of the pad, and the economic success of the project, respectively. Three gas price scenarios are used for calculation of the NPV's to analyze the impact of the market condition on the economics of the project. The analysis demonstrates that tighter well spacing, independent of gas price, leads to the improved NPV per acre, reduction of EUR, and an increase in well communication as shown by the newly developed well communication index. The models reveal that a monotonic relation between well spacing and NPV per acre does not exist due to the complex nature of the created fracture network and competition between two opposite factors: frac hits that arises at tighter well spacing and unstimulated zones that diminish.
We showed that obtaining optimized well spacing and configuration could only be achieved through applying a comprehensive workflow that not only accounts for the impact of various well design and configuration parameters on production but also their economic implications defined in terms of NPV per acre. It is important to note that the integration of microseismic data was essential for the success of the workflow since it provides a realistic picture of the pathways connecting the adjacent wells which facilitate well communication.
Fan, Dian (University College London) | Wang, Wendong (China University of Petroleum, East China) | Ettehadtavakkol, Amin (Texas Tech University) | Su, Yuliang (China University of Petroleum, East China)
Molecular Dynamics (MD) simulation has been helpful to study liquid transport through simple pore structures, e.g., single straight pores. However, to study the overall flowing capability through a porous medium, e.g., shale rocks, heterogeneities should be considered at the Representative Elementary Volume (REV) scale. We propose an analytical permeability model for liquid through a nanoporous REV by accounting for the heterogeneity, tortuosity, and wettability features.
We model oil slippage and adsorption in hydrophobic pores and hydrophilic pores to investigate the apparent slippage phenomenon in the mix-wet porous media. The fractal theory is applied to characterize REV-scale heterogeneities including pore size distribution, pore-throat tortuosity, and pore surface roughness. We particularly modify the classical slippage factor by a fractal tortuosity to study liquid slippage through tortuous pores by using MD data for straight nanotubes.
The proposed model gives an insight to liquid transport mechanisms in nanoporous and heterogeneous porous media and contributes to understanding hydrocarbon production in tight reservoirs at field scales. The results show two competing impacts of pore confinement. 1) The apparent oleic slippage is the result of liquid-solid interactions, and more importantly, the pore confinement's effect. Oil can slip in hydrophobic organic pores evidently, quantitatively comparable to oil slippage in hydrophilic inorganic pores, due to a higher pore-throat tortuosity and a smaller pore size. 2) The apparent oil permeability is the result of intrinsic permeability and apparent slippage. Despite a comparable slippage factor in both organic and inorganic pores, the apparent permeability in organic matter is restricted by stronger pore confinement.
In the past decade, researchers have been actively investigating flow behaviors of fluids in carbon nanotubes (CNTs) (Striolo, 2006; Ho, 2017). Flow enhancement and liquid (e.g., water and oil) slippage through CNTs are commonly observed under experimental conditions, where the apparent flow velocity is several orders of magnitude higher than that predicted by the classical Hagen-Poiseuille equation (de Gennes, 2002; Whitby and Quirke, 2007; Joseph and Aluru, 2008; Myers, 2011; Podolska and Zhmakin, 2013).
Crandall, Dustin (National Energy Technology Laboratory) | Gill, Magdalena (National Energy Technology Laboratory, LRST) | Moore, Johnathan (National Energy Technology Laboratory, LRST) | Brown, Sarah (West Virginia Geological and Economic Survey) | Mackey, Paige (National Energy Technology Laboratory, ORISE)
The behavior of fractured low-permeability rock in many subsurface formations is critical for unconventional resource extraction. Understanding how flow through individual fractures changes during shearing, and what influence heterogeneity of the rock has on shearing behavior, was the focus of our laboratory study. Computed tomography (CT) scanning of fractured rocks undergoing shear was coupled with numerical simulations of fluid flow through these fractures. We sheared multiple cores from the Marcellus and Eau Claire shales in a closed system with confining pressures of greater than 1000 psi. Samples were manually sheared in a step wise fashion. After each shearing event we assessed the bulk hydrodynamic response by measuring permeability through the core and performed a high-resolution CT scan to understand how the principal and secondary fractures were changing in the core volume. The mineralogy of each sample was examined via x-ray fluorescence.
A range of interdependent characteristics influence fracture network evolution and sample cohesion: mineralogy, lithological heterogeneity, principal fracture morphology, fracture asperities, and shearing direction in relation to bedding. We found that samples sheared parallel to bedding were less likely to develop extensive networks of secondary fractures, with secondary fracture growth contingent on the presence of large asperities. Fracture permeability tended to increase with continued shear and secondary fracture development, but a high variance existed between samples. In some instances, permeabilities decreased in response to shear-initiated aperture reduction due to fracture mating. Gouge formation is another factor contributing to the transmissivity decreases, particularly in shale-dominated fracture regions. The ability to study this complex behavior in a controlled fashion using CT scanning enables a view into processes that impact production in many unconventional formations. Findings show that small scale features and details can play a significant role in fracture behavior and should be accounted for.
Shale properties vary significantly and understanding how fractures evolve due to geomechanical stressing can improve our understanding of how to effectively stimulate a variety of formations.While hydraulic fracturing is a large-scale activity, the microfabric and heterogeneity of shale can control fracture evolution and flow properties. Upscaling the impact of microfabric and heterogeneity is poorly captured in most modeling and planning efforts; this disconnect between small scale features and large-scale operations is understandable. It is difficult to measure changes in fractures directly, difficult to implement upscaled equations of value, and difficult to know if studied laboratory/outcrop samples are representative of activities in the subsurface. This study describes the observed behavior of two distinctly different shales under controlled geomechanical stressing to examine what impact small features have on fracture evolution. By examining two shales with distinctly different structure and composition our goal is to understand when inclusions of micro-features in upscaling is critical to understanding system dynamics.
McClure, Mark (ResFrac Corporation) | Bammidi, Vidya (Keane Group) | Cipolla, Craig (Hess Corporation) | Cramer, Dave (ConocoPhillips Company) | Martin, Lucas (Formerly with Apache Corporation, now with Marathon Oil Company) | Savitski, Alexei (Shell International Exploration and Production Inc.) | Sobernheim, Dave (Keane Group) | Voller, Kate (Range Resources Corporation)
This paper summarizes findings from a one-year study sponsored by seven operators and service companies to investigate interpretation of diagnostic fracture injection tests (DFIT’s). The study combined computational modeling, a diverse collection of field data, and operator experience. DFIT simulations were performed with a three-dimensional hydraulic fracturing, wellbore, and reservoir simulator that describes fracture propagation, contacting of the fracture walls, and multiphase flow. Interpretation procedures were applied to estimate stress, permeability, and pressure from the synthetic data. The interpretations were compared to the simulation input parameters to evaluate accuracy. Based on the results, new techniques were developed, existing techniques were refined, and an overall interpretation protocol was developed. The techniques were applied to interpret over thirty field DFIT’s drawn from shale plays across the US and Canada, and the methods were evaluated in the context of operator experience. The results are applicable to fracturing tests in formations with permeability ranging from nanodarcies to 10s of microdarcies. The minimum principal stress is estimated by identifying the ‘contact pressure’ when the fracture walls come into contact, causing fracture compliance and system storage coefficient to decrease. After the walls come into contact, the pressure transient is controlled by the interplay of changing fracture compliance, deviation from Carter leakoff, and multiphase flow. The contact pressure is slightly greater than the minimum principal stress. It can be identified from either a plot of dP/dG or a relative stiffness plot. Permeability is estimated using the G-function method, a newly developed h-function method that accounts for deviation from Carter leakoff, and impulse linear flow. These three methods, which are based on linear flow geometry, require an estimate of fracture area. We derive equations for estimating area using mass balance equations, accounting for wellbore storage and fluid leakoff. The results from field data show that impulse linear permeability estimates are usually 2-5 times lower than estimates derived from the G-function and h-function methods, apparently indicating a difference between effective permeability during leakoff and permeability during flow of reservoir fluid through the formation. Impulse radial flow regime may be used for estimating permeability, but should be used with caution. Simulation results indicate that a variety of processes can cause an apparent radial trend that is not actually radial flow. Simulations and field data indicate that ‘false radial’ is very common in gas reservoirs and, if applied, leads to a large overestimate of permeability. Production history matching using overestimated permeability will underestimate fracture length, potentially resulting in suboptimal choices for well and cluster spacing.
Water saturation and permeability are crucial petrophysical properties to evaluate unconventional reservoirs. However, there is no agreement on accurately estimating these properties from logs. Thus, there is a need to develop scale dependent petrophysical correlations to improve the estimation of these properties. As a result, this work aims to use digital rock properties from high-resolution images of unconventional carbonate mudrock samples to develop petrophysical correlations to improve water saturation and permeability estimates. Focused ion beam scanning electron microscopy (FIB-SEM) images were obtained from four carbonate mudrock samples from the Middle East and were segmented into the individual components: calcite, organic matter, pore space, and pyrite. Each image was subdivided into eight sub-sections to study scale dependence. Image analysis provided component details such as porosity, pore size distribution, connectivity, and geometric tortuosity. The impact of varying fluid saturation was investigated by introducing two fluid phases in the segmented pore space. Digital rock (DR) simulations were performed to estimate absolute permeability and electrical resistivity. The results from the DR methodology are discussed with reference to porosity and permeability data from the Gas Research Institute (GRI) method and a relative comparison to the log data is included. The DR results were used to develop petrophysical correlations to predict water saturation and permeability from electrical resistivity. The results show that a limited amount of pyrite and organic matter within a non-conductive calcite framework can change the electrical resistivity by several orders of magnitude. When combined with low porosity, high salinity water, and changing saturation, the results show that some variations in log responses may be attributed to the changing rock matrix and fabric as opposed to saturation. The permeability results also demonstrate that the low porosity, limited connectivity, and resulting tortuosity have a significant impact. These DR-guided correlations may improve the estimates of water saturation and permeability using resistivity logs.
Inyang, Ubong (Halliburton) | Cortez-Montalvo, Janette (Halliburton) | Dusterhoft, Ron (Halliburton) | Apostolopoulou, Maria (University College London) | Striolo, Alberto (University College London) | Stamatakis, Michail (University College London)
Estimating the effective permeability and microfracture (MF) conductivity for unconventional reservoirs can be challenging; however, a new method for estimating using a stochastic approach is discussed. This new analysis method estimates matrix permeability and the unpropped and propped MF conductivities during laboratory testing where MFs were propped with ultrafine particles (UFPs).
Kinetic Monte Carlo (KMC) simulations form the basis of the method used to estimate effective permeability of the core sample. First, the stochastic model was implemented to calculate effective matrix permeability of a small core taken from unfractured Eagle Ford and Marcellus formation samples using scanning electron microscopy (SEM) images and adsorption data to obtain the pore-size distribution (PSD) within the sample. The KMC approach then evaluated the effect of various parameters influencing the conductivity of laboratory-created MFs. Case studies considered for this work investigate the conductivity improvement of a manmade MF as a function of the UFPs used as proppants that maintain width under high stress, the UFP (proppant) concentration, and the UFP flow perpendicular into a secondary or adjacent MF zone (2ndMF) penetrating the face of an opened MF during flow testing under stress. The leakoff area widths considered were 1, 2, and 3 mm and can be propped or unpropped.
Results obtained for the unfractured Eagle Ford and Marcellus samples closely correlate with other computational and experimental data available. For the laboratory-prepared nonpropped and propped MF samples, the effective propped width was determined to have the greatest effect on the MF conductivity, which increased by two orders of magnitude in the presence of the UFPs. The remaining two factors—proppant concentration and length of 2ndMFs—helped improve the effective MF conductivity in a linear manner; the highest proppant concentration and the 2ndMF zone resulted in the highest fracture conductivity achieved. Insight obtained from this study can be used to optimize fracturing designs by including UFPs and to create strategies for maximizing hydrocarbon recovery during development of unconventional resources where MFs are opened during stimulation treatments.