The amount of trapped oil in hydrocarbon rich shale reservoirs recoverable through Enhanced Oil Recovery methods such as low salinity water flooding has been an ongoing investigation in the oil and gas industry. Reservoir shales typically have relatively lower amounts of swelling clays and in theory, can be exposed to a higher chemical potential difference between the native and injected fluid salinity before detrimental permeability reduction is experienced through the volumetric expansion of swelling clays. This fluid flux into the pore spaces of the rock matrix acting as a semi permeable membrane is significant enough to promote additional recovery from the extremely low permeability rock. The main goal of this paper is to determine how osmosis pressure build up within the matrix affects geomechanical behavior and hydrocarbon fluid flow. In this study we investigate Pierre shale samples with trace amount of organic content and high clay content as an initial step to fully understanding how the presence of organic content affects the membrane efficiency for EOR applications in shales using low salinity fluid injection. The same concept is also valid when slickwater is utilized as fracturing fluid as majority of the shale reservoirs contain very high salinity native reservoir fluid that will create large salinity contrast to the injected slickwater salinity.
The organic-rich reservoir shales typically have a TOC content of approximately 5 wt% or higher with TOC occupying part of the bulk matrix otherwise to be filled up by clays and other minerals. With less clay within the rock structure, the amount of associated clay swelling arising from rock fluid interaction will be limited. The overall drive of water into the matrix brings added stress to the pore fluid known as osmotic pressure acting on the matrix that also creates an imbalance in the stress state. The native formation fluid with salinity of 60,000 ppm NaCl has been used while 1,000 ppm NaCl brine is utilized to simulate the low salinity injection fluid under triaxial stress conditions in this phase of the study reported here. A strong correlation is obtained between the osmotic efficiency and effective stress exerted on the shale formation. The triaxial tests conducted in pursuit of simulating stress alteration under the osmotic pressure conditions and elevated pore pressure penetration tests indicated that the occurrence of swelling directly impact the formation permeability. These structural changes observed in our experimental results are comparable to field case studies.
The Bone Spring and Wolfcamp formations of the Delaware Basin consist of mixed sediment gravity flow and suspension sedimentation deposits. These deposits exhibit high levels of heterogeneity both at and below core and log scales. A comprehensive approach integrating core and sub-core (nanoscale) data from two key wells and well logs within central Ward County was used to characterize small scale changes in lithology, rock properties, and reservoir quality. With this approach, a total of nine facies were identified; three siliceous mudstones [1, 2, 3], three siltstones [4, 5, 6], and three carbonates [7, 8, 9]. Each is comprised of different grain size distributions, textures, mineralogies, and pore types. Facies are not unique to an individual facies associations and cannot be predicted laterally in this study. Core-based measurements of source and reservoir properties were used along with qualitative observations from thin sections and high-resolution SEM images to identify facies as primary reservoir facies, secondary reservoir facies, and non-reservoir facies. Properties concerning source, reservoir, and mechanical quality were evaluated with respect to each facies and within each stratigraphic unit; 3rd Bone Spring, Wolfcamp A, Wolfcamp B, and Wolfcamp C.
Within the study area, 210 sq. miles in central Ward County along the eastern flank of the Delaware Basin, the Bone Spring and Wolfcamp formations are in the early mature oil window (0.69% – 0.88%Ro) and consist of an intercalation of siliceous mudstones [1, 2, 3], siltstones [4, 5, 6], and carbonates [7, 8, 9]. The four reservoir facies [1, 2, 4, 5] identified are organic rich with average wt.% total organic carbon (TOC) as follows; argillaceous siliceous mudstone  (3.1 wt.%, n=21), calcareous siliceous mudstone  (3.0 wt.%, n=15), argillaceous siliceous siltstone  (2.0 wt.%, n=7), and calcareous siliceous siltstone  (2.3 wt.%, n=7). Primary reservoir facies [1, 2] are richer in type II kerogen than the mineralogically comparable but coarser-grained secondary reservoir facies [4, 5], which contain more detrital grains and type III kerogen. Lower organic content in secondary reservoir facies [4, 5] is related to the dilution of organic matter via an extrabasinal influx of detrital grains and possible consumption by benthic fauna in oxygenated conditions. Degree of anoxia, bioturbation, and silica origin all have significant implications to reservoir quality as seen in the mineralogically similar non-reservoir biogenic siliceous mudstone facies  and the primary reservoir argillaceous siliceous mudstone facies . The former contains the least amount of detrital silica and organic matter of all facies observed. Early diagenesis of radiolaria and siliceous spicules source the microcrystalline authigenic quartz that was observed to occlude pore space in this non-reservoir facies . Despite the poor source potential and reservoir quality of this facies , the high amounts of microcrystalline authigenic quartz are beneficial to reservoir geomechanics. Implications to reservoir quality identified in this work have limited utility outside of the study area away from the flank of the basin, where bioturbation, degree of anoxia, and prevalence of extrabasinal facies differ. GRI saturations, MICP measurements, NMR (T2LM) data, and core-based TOC measurements indicate siliceous calcareous siltstone  as a facies potentially making up water-bearing carrier beds. Carbonate-rich facies [6, 7, 8, 9] were sampled least from core and more work must be done to better evaluate reservoir potential of these facies.
Core-based measurements of composition and reservoir quality indicate that porosity and permeability trend positively with clay, pyrite, and TOC, and negatively with carbonate. This relationship with porosity is most evident and statistically significant in the fine-grained facies [1, 2, 3, 4, 5], where silica is always the primary constituent. Relatively high clay content, upwards of 34 wt. %, in this study is not observed to negatively impact mechanical behavior. Porosity and TOC are highest in the Wolfcamp A and lowest in the lower Wolfcamp B subdivision, a trend observed beyond core control within the two key wells and on logs throughout the study area. This is largely a function of facies distribution. Based on stratigraphic architecture, facies distribution, and lack of thick non-reservoir carbonate barriers, the Wolfcamp A and upper Wolfcamp B may be considered one flow unit. This may allow well spacing and number of wells to be strategically optimized per drilling unit. Development strategies with respect to well spacing and well planning, may be better constrained with an understanding of each facies’ source potential, reservoir and mechanical quality, and distribution within each stratigraphic interval. Findings and interpretations from this research contribute to larger scale efforts being made to: 1) understand the role of diagenesis in unconventional reservoir quality; 2) recognize implications of depositional processes in unconventional reservoirs; and 3) image unconventional facies at the nano, micro, and macro scales.
Little is known about the nature and origin of microcrystalline quartz in sandstone reservoirs or mudstone reservoirs. We have utilized advanced analytical capabilities to improve our understanding of controls on microcrystalline quartz development in several examples where porosity is preserved in deeply buried sandstone reservoirs to understand the development in siliceous mudstones.
In this study, several advanced analytical techniques were used to evaluate the crystallographic and compositional controls on the formation of microcrystalline quartz. SEM/Cathodoluminescence (CL) imaging confirms that quartz overgrowths have a complex growth history. Previous workers (Kraishan et al. 2000) suggested that CL patterns in quartz cement are largely due to trace elements rather than defects and that aluminum varies consistently between each cement phase. Electron Backscatter Diffraction (EBSD) combined with Wavelength Dispersive Spectrometry (WDS) confirms that the complex banding visible in CL is not due to changes in crystallographic orientation but more likely variations in quartz composition associated with changes in pore fluid composition and/or reservoir conditions. Secondary Ion Mass Spectrometry (SIMS) analysis provides maps of ultra-trace element distribution that confirm that trace amounts of iron, manganese, and titanium can be used as proxies for defect density and temperature. Additionally, SIMS analysis provides oxygen isotope data providing insight into the initial reservoir conditions and temperature of formation of microcrystalline quartz in several formations.
Microcrystalline quartz in the form of replacement, micropore, and overgrowth cements is present in the Wolfcamp A in the southern Delaware Basin. The amount of cementation has an effect on the reservoir quality and appears to have an impact on the petrophysical properties. The siliceous mudstones are comprised predominantly of biogenic silica (sponge spicules, radiolarians, which are the silica sources for the authigenic microcrystalline quartz), detrital grains (quartz and feldspars), pyrite framboids, and organic matter.
Integrating the results from these advanced analytical techniques has helped us develop our understanding of the processes controlling the formation of quartz cement and improved our ability to reconstruct the reservoir diagenetic history of quartz growth leading to a proposed model for predicting porosity preservation in deep, hot sandstone reservoirs and the formation of microcrystalline quartz in siliceous mudstones. This is the first research to report on spatially resolved isotopic analysis of silica cements integrated into a petrographic framework and a proposed mechanism for microcrystalline quartz growth.
Low frequency Nuclear Magnetic Resonance (NMR) spectrometers have been used extensively to study the cores from various shale plays in the US. However, in the realm of high frequency NMR measurements, these high frequency NMR studies are quite limited. Usually, the studies conducted using high frequency NMR spectrometer are performed on crushed or very small size core samples, which may not represent the real reservoir conditions. We use core samples in dimensions that are comparable in size to low field NMR and studies in addition to crushed core samples to obtain a comparative analysis in this study.
NMR measurements provide valuable insight in the characterization effort of shale plays. Since the conventional petrophysical methods of calculating the Bulk Volume Water (BVW) and Free Fluid Index (FFI) are not accurate enough, NMR data is indispensable to quantify these petrophysical parameters. NMR provides guidance in determining the organic matter hosted porosity and the bulk volume fluids held within the defined porosity, as these contribute towards the negative reservoir quality indices. Therefore, there has been extensive research using low field NMR laboratory and field measurements in shale formations. However, due to complex nanoscale structure of the tight shale formations, higher resolution is needed that can be accomplished with higher frequency measurements. Therefore, the high field high frequency NMR results of the shale samples enhances our understanding of the various pore systems in the shale formations and the bulk volume of the fluids held within these different pore systems helping us in identifying the sweet spots in the unconventional play studied.
2D NMR T1-T2 maps have been collected for preserved Eagle Ford core samples in dry and saturated with different fluids. The effect of the various saturating fluids in the experiments have been investigated on the organic and inorganic matter hosted porosity using the 2D NMR T1-T2 maps. These maps help to understand how the various saturating fluids interact with the different pore systems in shales, indicating the bulk and irreducible volume of these respective fluids held within different pore systems. It helps in delineating the way in which these pore systems contribute towards the positive and negative reservoir quality indices.
For this study, a multi-scale evaluation of the reservoir quality of the oil-prone Wolfcamp A was investigated over Reeves, Pecos and Ward counties in the Texas Delaware Basin. A highly detailed core analysis was accomplished on 1,370 ft. from 7 cores with variable stratigraphic interval coverage of the Wolfcamp A. One supplementary well outside of the focal study area was included from Culberson county to gain a further perspective of sedimentological changes in the Delaware Basin.
Facies analysis were supplemented with XRD, thin-section petrography and XRF analyses. Nine mudstone and siltstone facies and four carbonate facies have been identified over the study area. The range of TOC content for each mudstone-siltstone facies is a unique situation where eight of the mudstone and siltstone facies have both the potential to be organic-rich reservoir quality facies (≤2 wt.%) and organic-lean facies (≥2wt.%), the result of heterogeneity both vertically within a single core and laterally between cores. A reservoir quality hierarchy containing primary, secondary and tertiary reservoir quality facies that were defined on two critical parameters: average TOC content and mineralogical composition and consistency. These parameters were chosen based on the implications that these compositional controls have on the reservoir quality properties; hydrocarbons in place, mechanical properties and fracture stimulation.
The stratigraphic record of the Wolfcamp A is dominated by gravity-driven event beds that entered the basin from numerous sediment entry points. The multiple depositional processes have resulted in the complex vertical arrangement of interbedded carbonate debris flow deposits, high and low-density carbonate and siliciclastic turbidites, hybrid event beds, dilute turbulent wake and hemipelagic facies. The variability observed in core of the different facies is a result of the depositional processes.
The most significant challenge to reservoir characterization of the Wolfcamp A in the Delaware Basin, is the inability to take the sedimentological, stratigraphic and reservoir quality variabilities that are documented at the vertical scale and extrapolate into the lateral dimension. Capturing the different level of complexity of the Wolfcamp A in the Delaware Basin, and predicting lateral changes in reservoir quality requires a novel reservoir characterization approach.
In multistage hydraulic fracturing treatments, the distribution of proppant between multiple perforation clusters has a significant impact on treatment behaviors and results. Low viscosity fluids, such as slickwater fluids, are used intensively in hydraulic fracturing treatments to fracture the shale formations. Despite their low cost and their tendencies to generate more fracture networks into the formation, they lead to poor proppant suspension and, as a result, only a small portion of the fractures might be efficiently propped. This paper analyzes experimental tests conducted on proppant transportation behavior in horizontal wellbores and through the perforation clusters using high loading friction reducer (HLFR)-based fluids. It compares the results of these proppant transport experiments against other work that has been conducted in this apparatus and elsewhere.
The viscosity and elasticity of the HLFR fluids were measured under a variety of concentrations across a wide range of shear conditions. Proppant transport tests were conducted at different flow rates and proppant concentrations, utilizing a 30-foot horizontal pipe with three perforation clusters at shot densities of 4 SPF with 90-degree phasing. A range of fluid viscosities were used to transport sand particles for 20/40 and 40/70 mesh sizes.
The results show that the HLFR fluids have superior proppant suspension capabilities, when compared to other low viscosity fluids, such as basic slickwater and fresh water. The 40/70 mesh sand was observed to be transported mostly homogeneously with high loading of HLFR in the horizontal section of the apparatus, hence uniform proppant distribution was observed between the three perforation clusters. On the other hand, the 20/40 mesh sand at low flow rates and low fluid viscosities indicated that gravity was the dominant force acting upon the fluid over the momentum and the drag. In these cases, proppant particles tended to stay in the lower section of the horizontal pipe and were deposited prior to reaching the toe perforation. As a result, uneven proppant distribution was observed with higher proppant concentrations toward the first perforations. However, at high flow rates, more proppant was received at the toe cluster. This occurred because the total momentum near the first cluster prevented the proppant from turning into the perforation holes.
This paper is believed to be the first to provide a comprehensive evaluation of proppant transport and behavior in horizontal wellbores using HLFR-based fluids. An in-depth understanding of the factors that have a significant impact on proppant transport and behavior in the horizontal wellbores is critical to improving our understanding of proppant distribution and behavior during hydraulic fracturing treatments, as well as during the flowback processes.
Zhu, Ziming (Colorado School of Mines) | Fang, Chao (Virginia Polytechnic Institute and State University) | Qiao, Rui (Virginia Polytechnic Institute and State University) | Yin, Xiaolong (Colorado School of Mines) | Ozkan, Erdal (Colorado School of Mines)
In nanoporous rocks, potential size/mobility exclusion and fluid-rock interactions in nano-sized pores and pore throats may turn the rock into a semi-permeable membrane, blocking or hindering the passage of certain molecules while allowing other molecules to pass freely. In this work, we conducted several experiments to investigate whether Niobrara samples possess such sieving properties on hydrocarbon molecules. Molecular dynamics simulation of adsorption equilibrium was performed to help understand the trends observed in the experiments. The procedure of the experiments includes pumping of liquid binary hydrocarbon mixtures (C10 C17) of known compositions into Niobrara samples, collecting of the effluents from the samples, and analysis of the compositions of the effluents. A specialized experimental setup that uses an in-line filter as a mini-core holder was built for this investigation. Niobrara samples were cored and machined into 0.5-inch diameter and 0.7-inch length mini-cores. Hydrocarbon mixtures were injected into the mini-cores and effluents were collected periodically and analyzed using gas chromatography. To understand the potential effects of hydrocarbon-rock interactions on their transport, molecular dynamics simulations were performed to clarify the adsorption of C10 and C17 molecules on calcite surfaces using all-atom models. Experimental results show that the heavier component (C17) in the injected fluid was noticeably hindered. After the start of the experiment, the fraction of the lighter component (C10) in the produced fluid gradually increased and eventually reached levels that fluctuated within a range above the fraction of C10 in the original fluid; besides, the fraction of C17 increased in the fluid upstream of the sample. Both observations indicate the presence of membrane properties of the sample to this hydrocarbon mixture. Simulation results suggest that, for a calcite surface in equilibrium with a binary mixture of C10 and C17, more C17 molecules adsorb on the carbonate surface than C10 molecules, providing a mechanism that directly supports the experimental observations. Some experimental observations suggest that size/mobility exclusion should also exist. This experimental study is the first evidence that nanoporous reservoir rocks may possess membrane properties that can filter hydrocarbon molecules. Component separation due to membrane properties has not been considered in any reservoir simulation models. The consequence of this effect and its dependence on the mixture and environmental conditions (surface, pressure, temperature) are worthy of discussions and further investigations.
Production from organic-rich shale petroleum systems is extremely challenging due to the complex rock and flow characteristics. An accurate characterization of shale reservoir rock properties would positively impact hydrocarbon exploration and production planning. We integrate large-scale geologic components with small-scale petrophysical rock properties to categorize distinct rock types in low porosity and low permeability shales. We then use this workflow to distinguish three rock types in the reservoir interval of the Niobrara shale in the Denver Basin of the United States: The Upper Chalks (A, B, and C Chalk), the Marls (A, B, and C Marl), and the Lower Chalks (D Chalk and Fort Hays Limestone). In our study area, we find that the Upper Chalk has better reservoir-rock quality, moderate source-rock potential, stiffer rocks, and a higher fraction of compliant micro- and nanopores. On the other hand, the Marls have moderate reservoir-rock quality, and a higher source rock potential. Both the Upper Chalks and the Marls should have major economic potentials. The Lower Chalk has higher porosity and a higher fraction of micro-and nanopores; however, it exhibits poor source rock potential. The measured core data indicates large mineralogy, organic-richness, and porosity heterogeneities throughout the Niobrara interval at all scale.
Unconventional petroleum systems are highly complex hydrocarbon resource plays both at the reservoir scale and at the pore scale (Aplin and Macquaker, 2011; Loucks et al., 2012; Hart et al., 2013; Hackley and Cardott, 2016). These organic-rich sedimentary plays, generally described as shale reservoirs, are composed of very fine silt-and clay-sized particles with grain sizes < 62.5 μm (Loucks et al., 2009; Nichols, 2009; Passey et al., 2010; Kuila et al., 2014; Saidian et al., 2014). They undergo extensive post-depositional diagenesis that transforms rock composition and texture, hydrocarbon storage and productivity, and reservoir flow features (Rushing et al., 2008; McCarthy et al., 2011; Jarvie, 2012; Milliken et al., 2012). Although some shale rock facies can retain depositional attributes during diagenesis, many critical reservoir properties, such as, mineralogy, pore structure, organic richness and present-day organic potential, etc., are significantly perturbed (Hackley and Cardott, 2016).
Rastogi, Ayush (Colorado School of Mines) | Agarwal, Karn (Liberty Oilfield Services) | Lolon, Ely (Liberty Oilfield Services) | Mayerhofer, Mike (Liberty Oilfield Services) | Oduba, Oladapo (Liberty Oilfield Services)
Artificial Neural Network (ANN) has been used by the petroleum industry to identify key well performance drivers since the 1990's. There is usually an inverse relationship between model accuracy and model interpretability—the more interpretable a model is, the more likely it is to underperform, which is a common issue with simple linear-based regression models. Without a priori assumptions regarding the relationships between response and predictor variables, ANN has successfully been used for solving nonlinear problems.
Two field production and completion databases compiled in the DJ (Niobrara) and Permian (Wolfcamp B) basins were analyzed. The objective of this paper is to focus on the inference aspect of the ANN model, rather than just reporting the model prediction results. In order to leverage neural networks as a computation tool, a detailed pre-processing workflow is recommended which improves understanding of the impact of geology, reservoir, and completion parameters on well production. The paper emphasizes the importance of data normalization, feature selection, and outlier detection as well as their implications on prediction accuracy. Self-organizing maps are used as a multi-dimensional scaling tool for unsupervised learning to preserve topography of the dataset.
Results demonstrate that using a large Permian dataset (~3,000 wells), the ANN model explains 71% of the variance in production and does not overfit when the analysis accounts for feature selection and influential points. Using a smaller DJ dataset (<400 wells), the ANN model overfits significantly even though the model explains a higher percentage of the variation in production (78%). Using Root Mean Square Error (RMSE) as an accuracy metric for the regression model, the ANN model reduces the RMSE compared to the conventional linear model. The paper reveals the misconception of using neural networks for sparse datasets. The paper also presents key well performance indicators that can be used for a quick evaluation to determine the most economical completion methods.
Well-to-well interference is an increasingly discussed issue. Previously drilled and producing “parent” wells and recently drilled “child” wells are yielding a reduction in recovery rates in both short and long-term cases due to interference. A primary contributor to the variability in production is the presence of pressure sinks as the result of production depletion in the parent wells. Infill drilling will continue to occur in the development of unconventional plays, and it is crucial to gain an understanding of the impacts of well-to-well interference on hydraulic fracture generation.
This paper discusses a detailed approach to investigating well-to-well interference based on integrating hydraulic fracture modeling and reservoir simulation in two different formations, the Niobrara and Codell, in the Denver-Julesburg Basin. The geomechanical properties were calibrated by DFIT data and pressure matching of the parent well treatments. The resulting parent well fracture geometries were incorporated into a numerical reservoir model to determine the pressure depletion envelopes. The imported depletion model allows for the simulation of the child well treatments and associated impacts of the pressure sinks on fracture generation and the interaction between child and parent wells. The resulting depletion model provided a framework to investigate various methods to mitigate the effects of well-to-well communication in subsequent development. The developed workflow of well-to-well interference is applicable in understanding the effects of infill development in other producing basins.
The modeled child well treatments resulted in a clear indication of well-to-well communication with the parent wells that was attributable to pressure depletion. Actual field bottom-hole pressure measurements validated these results in the parent wells captured during the time of the child well treatments. Resulting proppant concentrations of the child well fractures indicated that the majority of the proppant transports towards the parent wells. Very little effective conductivity exists in the opposing direction of the depleted regions.
Slickwater treatment simulations indicate extremely asymmetric fractures that stay isolated to their respective target bench. For child wells in the same bench as the parent wells, fractures propagate directly toward the parent wells, with little to no fracture growth in the opposite direction.
Protection frac simulations indicate beneficial or detrimental results depending on the amount of repressurization that is achieved and the distance that the pressure transient extends into the reservoir. Re-pressurizing the reservoir surrounding the parent wells by 1,000 psi resulted in a reduction of well interference. A 500-psi scenario resulted in increased well interference between the parent wells. Several wells communicated with both parent wells due to the repressurization being insufficient to offset the depletion.
Natural repressurization of the reservoir to mitigate the effect of well interference was also investigated by using the reservoir model. Simulation of the parent wells being shut-in for three months prior to the child well treatments resulted in a pore pressure increase of only 280 psi. Based on the protection frac sensitivity of 500 psi, this is not a large enough repressurization to mitigate well-to-well interference successfully in the modeled scenarios.