Shale Nano-Pore Structures And Confined Fluid Behavior

Shale Nano-Pore Structuresand Confined Fluid BehaviorProject Number: FWP FE 406/408/409 2Hongwu Xu, Qinjun Kang, Rex HjelmRajesh Pawar & George GuthrieLos Alamos National LaboratoryU.S. Department of EnergyNational Energy Technology LaboratoryMastering the Subsurface Through Technology Innovation, Partnerships and Collaboration:Carbon Storage and Oil and Natural Gas Technologies Review MeetingAugust 1-3, 2017

Presentation Outline Technical Status– Experimental Studies (SANS)– Computer Simulations (LBM) Accomplishments to Date Synergy Opportunities Project Summary2

Why Shale Matrix?Problem Although shale oil/gas production in the U.S.has increased exponentially, the currentenergy recovery rates are extremely low: 10% for oil and 35-40% for gas. The production rate for a given well typicallydeclines rapidly within one year or so.Cause Small pore sizes (a few to a few hundred nm)and low permeability (10-16-10-20 m2) of shalematrices.Technical ChallengeMarcellus Characterize shale nanopore structures andunderstand the confined fluid behavior.Matrix-fracture fluid transfer3

Shale Nano-Pore Structure (Open vs. Closed) Open vs. closed nanopores - the proportion and distribution of which are tied tothe permeability of a shale matrix. Estimate original oil/gas in place (OOIP/OGIP) more reliably. Intrinsically heterogeneous; in organic (kerogen) and inorganic (clay)components. Unconventional formations - a mix of different lithologies and their hydrocarbonproductions vary (i.e., one shale can produce much better than the other).WolfcampilliteWolfcampOM-rich; nocarbonateOM-poor,carbonate-richMarcellus4

Nano-Pore Fluid Confinement (Pressure Mgmt.) Changes in fluid flow behavior: the traditional, Darcy’s-law-based approaches areinadequate. Shifts in liquid/vapor phase boundaries/diagrams (bubble/dew points).Fluid in shaleDarcy’s lawKnudsen number: Kn λ/r, r - average poresize; λ - mean free path of gas moleculesFailure to take into account the nano-confinement effectscan underestimate the ultimate oil/gas recoveries by ashigh as 50% (Sapmanee 2011; Nojabaei et al. 2012).This understanding is critical for developing optimum field production strategies.5

Basic Shale Matrix Characterization Compositions (mineralogy & chemistry) X-ray diffraction (XRD) – Mineral compositions X-ray fluorescence (XRF) – Chemical compositions Differential scanning calorimetry (DSC) / Thermogravimetry (TG) – TOC/water contents;kerogen thermal maturity Microstructure Scanning electron microscopy (SEM) / Focused ion beam (FIB); X-ray/neutron tomographytomographyWolfcamp Dark Layerillite dehydrationReflectingthermalmaturitykerogen pyrolysis6

Shale Nano-Pore CharacterizationConventional Techniques Gas adsorption and mercury intrusion/immersion porosimetryCumulative Pore Volume (cc/g) Transmission electron microscopy (TEM)N2 Adsorption0.05TEMMarcellus (High TOC)MarcellusKerogen Nano-Pores0.04High TOC0.030.02Low TOC0.010.000102030Pore Diameter (nm) Larger pores dominate the overall pore volumes and pore surface areas. The high TOC shale is more porous than the low TOC shale: Kerogen is more porous.7

Shale Nano-Pore CharacterizationConventional Techniques Gas adsorption / mercury porosimetry: limited to measuring open pores TEM: requires thin specimens and measures a small area.Neutron Scattering Neutrons are highly penetrating (e.g. compared with X-rays) Probing samples at depths Ease of combination with sample environments (e.g. a pressure cell) Neutrons are sensitive to hydrogen (rich in hydrocarbons and water) & its isotopes8

Small-Angle Neutron Scattering Small- and ultra-small-angle neutron scattering (SANS/USANS) characterizepores ranging from 1 nm to 10 µm (SANS: 1-100 nm; USANS: 100 nm-10 µm). Combine with controlled environment cells to probe the properties of fluids(hydrocarbon/water) in nanopores.2dsinθ λ Q 4πsinθ/λI N(ΔρV)2P(Q)S(Q)MarcellusHydrostatic cell (3 kbar) bedding planeEffect of texture of nanoporesGas-mixing systemOedeometer for simulating pore pressure(500 bar) overburden stress (100 bar).LANL-Developed Pressure Systems9

Develop an Approach to DistinguishOpen vs. Closed Shale Nano-Pores SANS/USANS signals reflect the difference in scattering length densitybetween the rock matrix and the pore space of a rock. Sensitive to isotopes, especially H & D (opposite signs of neutron scattering). Use a H/D mixture (e.g. H2O/D2O & CH4/CD4) to match the scattering of therock matrix to reveal closed versus open pores Contrast Matching.105dry104wet70% D2O 30% H2OMarcellus Shale-1I(Q) (cm )100010010neutron scatteringhydrogendeuteriumThe slope difference reflectsthe amount of closed pores.10.10.001628 nm0.010.163 nm6 nm-1Q(A )d10

Water Imbibition – Water Stays in the Matrixpsipsipsipsi Higher pressures ( 4K psi) had little effect. Water entered into larger pores (tens of nm). On decreasing P, water remained in the pores.11

LBM Modeling of Gas Flow in Shale MatrixPressuremedia: 70% D2O 30%GasH2Oflow– to throughmatch thematrix Dusty gas model(a superimpositionmodel):nanoporesconsists of contributions from viscous flow, Knudsen diffusion & surface diffusion.Viscous flowKnudsen diffusionCorrection FactorViscous flowKnudsen diffusionsurface diffusionSurface diffusionfcf c ka / kd 1 Dk,eff µpkd For conventional porous media,apparent permeability ka intrinsicpermeability kd, correction factor 1. For tight formations, ka kd.Cunningham & Williams (1980) The traditional, Darcy’s lawunderestimates the gas transport12rate in shale matrix.

Develop a Predictive Method forModeling Gas Flow in Shale MatrixPressure media: 70% D2O 30% H2O – to match the matrixMarkov Chain MonteCarlo (MCMC) method(2.8 nm/pixel)SEM image of a shaleReconstructed 3D pore structure The correction factors can be up to 100,depending on the pressure. Decreasing pressure can increase the matrixflow ability by 100 times Wellbore pressurecycling to increase the production? First numerical study – providing a predictivecapability.13

Effect of Surface Diffusion on Gas FlowPressure media: 70% D2O 30% H2O – to match the matrix Surface diffusion of the adsorbed gas in kerogen nanopores can enhanceor reduce the apparent permeability. Most obviously, it enhances thepermeability for smaller pores and higher diffusivities.14

Effect of Mixed Wettability on theRelative Permeability of Oil-Water flowPressure media: 70% D2O 30% H2O – to match the matrixOil-Wet SolidO/W 4/1O/W 3/2O/W 2/3O/W 1/4Water-Wet SolidPorous media with different wettability properties. Grey color denotesoil wet and black color water wet. The fractions of water wet solids are0.0, 0.2, 0.4, 0.6, 0.8 and 1.0, respectively, from top left to bottom right.15

Flow Blockage of Two-Phase Fluids(Oil/Gas Water)Pressure media: 70% D2O 30% H2O – to match the matrixOil-Wet SolidWater-Wet SolidOil-Wet Solid :Water-Wet Solid 2:3Oil/Water 1/1When the oil saturation is moderate (0.3-0.7), the total relative permeability ofthe mixed wet porous media is smaller than that of the purely oil-wet or waterwet porous media, indicating higher resistance to oil-water flow.16

Accomplishments to Date Developed a capability/approach to measure open vs. closed shalenanopores at reservoir conditions important for betterestimating original gas/oil in place. Examined the water imbibition phenomenon in shale matrix important for addressing the question of ‘where does the water goduring fracking?’. Discovered the enhanced gas flow ability in shale matrix by 100times via decreasing pressure wellbore pressure cycling toincrease gas production? Predicted the reduced total relative permeability of mixed wetporous media compared to that of a purely oil-wet or water-wetmedium higher resistance to two-phase oil/gas-water flow.17

Synergy Opportunities Multi-Lab Synergies and Collaborations on UnconventionalGas/Oil Research Common field site: Marcellus and MSEEL Sample sharing: Avoid redundant sample characterizationand provide/share complementary information obtainedwith different techniques Geochemistry/mineralogy collaboration between LANL,SLAC, Sandia, LBL and NETL. Synergies with CO2 Sequestration (caprock properties)18

Project Summary Key Findings Determination of open vs. closed shale nanopores is important forbetter estimating original gas/oil in place and for predicting wellproduction performance. While increasing pressure generally opens fractures to facilitate gasflow, decreasing pressure can also enhance gas flow in shale matrix.This finding suggests the production can potentially be increased viawellbore pressure cycling. Next Steps Characterize open/closed nanopores for a set of representative shalelithologies and link the characteristics with the production data. Predict the gas flow in shale matrix based on the determined nanoporestructures and incorporate the results into DFN modeling to simulatethe production curve.19

Questions?20

Appendix: Benefit to the Program Program goals being addressed:The magnitude of the natural gas resource recoverable from domestic fracturedshales has only been recognized within the past decade as a combination of drillingand well completion technology advancements, which have made it possible toproduce gas from shales at economic rates. NETL research efforts focus on furtherrefining these technologies, characterizing the geology of emerging shale plays, andaccelerating the development of technologies that can reduce the environmentalimpacts of shale play development. Project benefits statement:This research project is developing an approach for characterizing shale nanoporestructures and their confined fluid behavior with high fidelity. The obtained newknowledge will reveal the key factors controlling the production tail and thus will helpdevelop optimum long-term field production strategies to enhance hydrocarbonrecovery.21

Appendix: Project OverviewGoals and Objectives Develop a fundamental understanding of what controlshydrocarbon transport at different scales, using an integration ofexperimental and modeling methods.– Experimental studies and pore-scale modeling of shale matrixnanopore structures and their fluid behavior What are the characteristics of shale nanopores? How to better estimate the original gas/oil in place? How do fluids move within the matrix and to fractures?22

Appendix: Organization ChartGeorge Guthrie(Project Lead)Task 1: Reservoir-scaleModeling(Satish Karra PI)Task 2: Core-scaleexperiments(Bill Carey PI)Jeffrey HymanNataliia MakedonskaHari ViswanathanRichard MiddletonMark PorterJoaquín JiménezMartínezLuke FrashTask 3: Micro-scale(Hongwu Xu PI)Li ChenMei DingRex HjelmQinjun KangRajesh Pawar23

Appendix: Gantt ChartFY16Understanding Basic Mechanisms in Natural GasProduction using Reservoir-Scale ModelingConcludedExperimental Study of In Situ Fracture Generation and FluidMigration in Shale.ConcludedProbing Hydrocarbon Fluid Behavior in NanoporousFormations to Maximize Unconventional Oil/Gas Recovery ConcludedAssessment of current approaches to understandingHydrocarbon productionLarge-scale fracture controls on hydrocarbon production inthe Marcellus shaleTributary zone fractures (small-scale) contributions tohydrocarbon production in the Marcellus shaleFundamental Matrix Properties in Relation to PredictingHydrocarbon Migration into Fractured Marcellus ShaleIntegration of Large-Scale Fractures, Tributary Fracturesand the MatrixFY17FY18ConcludedOn trackOn trackOn track24