Multiscale Pore Research Articles

Clarifying the contribution of multiscale pores to physical properties of Chang 7 tight sandstones: insight from full-scale pore structure and fractal characteristics

The pore structure and its heterogeneity of tight reservoirs are key factors affecting the storage and percolation of crude oil. The pore system of Chang 7 tight sandstone has multi-scale and multi-type characteristics. However, the contribution of different pore types and pore sizes to the physical properties of Chang 7 tight sandstone is still unclear. In this paper, we collected a suite of Chang 7 tight sandstones to investigate the full-scale pore structure and fractal characteristics by casting thin section, field emission scanning electron microscope, two-dimensional multi-scale backscattered scanning electron microscopy, N2 adsorption (NA) and NMR. The pore diameters of Chang 7 tight sandstone are usually distributed between 0.001 and 20 μm. Intercrystalline pores are mainly distuibuted <500 nm. Dissolution pores vary from 100 nm to 100 μm. Residual intergranular pores range from 1 μm to 40 μm. Based on the fractal characteristics, pore system is divided into macropores (mainly >300 nm), mesopores (mainly 7–300 nm), and micropores (mainly <7 nm). Micropores are adsorb-fluid pores that do not contribute to the storage and percolation but contribute significantly to contrasting specific surface area. Mesopores represent bound-fluid pores and only contribute to total porosity but not to permeability. Macropores represent movable-fluid pores, contributing to both porosity and permeability. The content and heterogeneity of macropores control the quality of Chang 7 tight sandstone. When macropore volume is >12×10−3 mL/g, the continuous percolation network consists entirely of macropores, resulting in higher porosity and permeability of the reservoir. Moreover, reservoir physical properties show excellent correlation with macropore heterogeneity. These results demonstrate that the content and heterogeneity of macropores are key indicators indicating the quality of the Chang 7 tight sandstones.

A multi-mechanism numerical simulation model for CO2-EOR and storage in fractured shale oil reservoirs

Under the policy background and advocacy of carbon capture, utilization, and storage (CCUS), CO2-EOR has become a promising direction in the shale oil reservoir industry. The multi-scale pore structure distribution and fracture structure lead to complex multiphase flow, comprehensively considering multiple mechanisms is crucial for development and CO2 storage in fractured shale reservoirs. In this paper, a multi-mechanism coupled model is developed by MATLAB. Compared to the traditional Eclipse 300 and MATLAB Reservoir Simulation Toolbox (MRST), this model considers the impact of pore structure on fluid phase behavior by the modified Peng–Robinson equation of state (PR-EOS), and the effect simultaneously radiate to Maxwell–Stefan (M–S) diffusion, stress sensitivity, the nano-confinement (N-C) effect. Moreover, a modified embedded discrete fracture model (EDFM) is used to model the complex fractures, which optimizes connection types and half-transmissibility calculation approaches between non-neighboring connections (NNCs). The full implicit equation adopts the finite volume method (FVM) and Newton–Raphson iteration for discretization and solution. The model verification with the Eclipse 300 and MRST is satisfactory. The results show that the interaction between the mechanisms significantly affects the production performance and storage characteristics. The effect of molecular diffusion may be overestimated in oil-dominated (liquid-dominated) shale reservoirs. The well spacing and injection gas rate are the most crucial factors affecting the production by sensitivity analysis. Moreover, the potential gas invasion risk is mentioned. This model provides a reliable theoretical basis for CO2-EOR and sequestration in shale oil reservoirs.

Thermal Stability and Weather Resistance of a Bionic Lotus Multiscale Micro-Nanostructure TiC/TiN-Ni/Mo Spectral Selective Absorber Based on Laser Cladding-Induced Melt Foaming.

As the core of a solar collector, a solar selective absorbing coating has become the key material for efficient utilization and development of solar energy. In order to overcome the core scientific problem of poor high-temperature stability of optical properties caused by atomic diffusion between layers in the high-temperature environment of traditional multilayer solar absorption coatings, the multiscale lotus bionic porous structure was constructed by using a TiC/TiN-Ni/Mo material system with excellent intrinsic absorption performance. The melt foam method and laser cladding technology were combined to deposit the multiscale bionic porous structure solar selective absorption coating in situ by a laser-induced melt foaming strategy. It was found that the bubbles in the molten pool were affected by Marangoni force, gravity, buoyancy, and surface tension. The unescaped bubbles formed pores near the surface of the coating and showed a bimodal distribution. For the multiscale addition of 1.8 and 0.4 μm pore-forming agents with a mass fraction of 15 wt %, the multiscale pores enhance the scattering and secondary absorption of light and reduce the amplitude of electromagnetic wave to electron vibration; at the same time, the small-size conductor effect formed by the hole increases the surface electron concentration, strengthens the absorption of light, and enhances the magnetic field strength at the hole, and the structural absorbing effect is significant. The coating absorptivity α reaches 85%, and the temperature stability is excellent. Compared with the substrate, the coating has a smaller corrosion current density, larger polarization resistance, and strong corrosion resistance. The research shows that the laser-induced multiscale bionic porous structure coating obtains the intrinsic absorption-structural absorption composite absorption mechanism and realizes the integration of flexible design and efficient manufacturing of high-temperature solar absorption coatings.

Hierarchical cluster and principal component analyses of multi-scale pore structure and shale components in the Upper Triassic Chang 7 Member in the Ordos Basin of China

An important issue concerning the sustainable production of shale oil is the characterization of pore structure at various scales, but there is an insufficient understanding of the coupled relationship between pore structure and shale components. Integrated analyses using field emission-scanning electron microscopy, mercury intrusion porosimetry, nitrogen physisorption, and small angle neutron scattering were conducted on nine shale samples from the Upper Triassic Chang 7 Member in Ordos Basin to investigate the relationship between multi-scale pore structure and shale minerals and organic matter (OM). The results show that the clays minerals (mainly illite-smectite) are usually combined with OM to form complex OM-hosted pores, and inorganic pores consist of both inter- and intra-particle pores (mostly related to clay platelet clusters). Three groups of shales classified by hierarchical cluster analyses (HCA) vary widely in pore structure, with the highest pore volume and largest pore diameter observed in a decreasing order of Group (1), Group (2), and Group 3. Principal component analysis (PCA) shows that brittle minerals (mainly quartz) have preserved interP pores that contribute to the development of high pore volumes and large pore-throat diameters, and ductile clays minerals and OM tend to fill pores and reduce pore diameter and volume. The complex nature of shale porosity is influenced by multiple factors, and hence accurate and comprehensive characterization requires the use of diverse methods including both experimental and theoretical analyses. An integrated approach to elucidating the factors controlling pore structure in medium-to-low maturity shale will help with efficient shale oil production.

Thermochemical energy storage performances of Co3O4-based honeycombs with multi-scale composite pores

Metal oxide redox system characterized open-loop operation, high energy density, and high reversibility, which is one of the most promising thermochemical energy storage technologies for the next-generation concentrated solar power plants. Most of the previous studies focused on the material properties, while the energy storage performance of oxide monolithic bodies, such as energy storage density and heat storage/release rate, has rarely been reported. In this paper, a novel and effective method for the improvement of the energy storage performance of oxide honeycombs with multi-scale composite pores is proposed and presented. Methyl cellulose and four common biomass materials, including bagasse, loofah, juniper leaf, and pine needle, were used as pore-forming agents for oxide honeycombs. The pyrolysis of methyl cellulose forms small-scale pores in the honeycomb bodies and contributes the major pore volume. The biomass templates further optimize the pore characteristics, all biomass-templated Co3O4-based honeycomb bodies have higher total pore volume and permeability than that of biomass-free honeycomb body. The honeycomb with 2.5 wt% pine needle achieves the highest energy storage density, with an average of 694.62 kJ/kg during the second to fifteenth cycles. In addition, the honeycomb containing 7.5 wt% pine needle has the highest energy store/release rate. Biomass-templated Co3O4-based honeycomb shows a high crushing strength, and there are no obvious cracks and structural deformations on its surface after 15 thermal cycles. The findings demonstrate that the methyl cellulose and biomass template method can effectively optimize the pore structure and energy storage properties of the oxide honeycombs, thereby expanding the application prospects of Co3O4-based honeycombs in thermochemical energy storage.

Insights into mechanism and modelling of dynamic moisture sorption in dual-porous insulation materials with micro- and nano-scale pores

HypothesisUnderstanding moisture sorption in porous insulation materials is challenging due to the influence of multiscale pore structures on phase behavior and transport properties. Dynamic moisture sorption in dual-porous materials is likely co-determined by interior micro- and nano-scale pores, and an accurate physical model for predicting moisture evolution can be developed by clarifying the sorption mechanisms. ExperimentsMoisture behavior during the dynamic sorption of dual-porous insulation material is measured by low-field nuclear magnetic resonance (NMR) experiments. The contributions of micro- and nano-scale pores to the adsorbed moisture are differentiated using NMR relaxometry, and the evolution of moisture morphology is quantitatively analyzed. FindingsAnalysis of T2 evolution reveals that the moisture in nano-scale pores alters from adsorption layers to liquid with increasing relative humidity (RH), while minimal sorption occurs in micro-scale pores. Moisture is mainly transferred as vapor molecules at low RH levels, with the dynamic sorption enhanced by molecular diffusion in micro-scale pores. Capillary flow in nano-scale pores dominates moisture transport when RH rises above a threshold, leading to a significant increase in apparent moisture diffusivity. According to the elucidated mechanism, a physical model is further developed to predict moisture sorption inside dual-porous insulation materials, and it may serve as a basis for evaluating and optimizing the performance of dual-porous systems in different environments.

Efficient electro-demulsification of O/W emulsions and simultaneous oil removal enabled by a multiscale porous biocarbon electrode

Emulsion wastewater contain substantial amounts of oil and various additives, which pose threats to the environment and human health. Demulsification is a crucial pretreatment stage for wastewater. This study aims to identify a novel electro-demulsification method with high oil removal efficiency and low energy consumption. Modified carbonized birch wood with a unique isotropic multiscale pore structure is used as a self-standing electrode to treat a toluene oil-in-water (O/W) emulsion. The electrode must have a highly porous structure to facilitate efficient water diffusion and oil adsorption. It must also have high electronic conductivity to expedite polarized molecular electrophoresis to realize penetration into the pores and, subsequently, demulsification. Guided by an applied electric field force, polarized O/W droplets are drawn toward the electrode, revealing electrical characteristics distinct from those of polarized organic molecules. This electric field force augments the capture and adhesion of droplets by the electric double layer at the electrode interface. Consequently, adsorbed droplets in close proximity to the electrode rupture due to the combined influence of the electric field force and the electrostatic effects stemming from the electrode's multiscale porous structure. This synergistic action enables demulsification to occur efficiently at low energy consumption levels. This study has revealed that electro-demulsification can effectively treat toluene emulsions stabilized by various surfactants and microemulsion containing toluene. Therefore, this electro-demulsification technology can be further developed for various types of water pollution.

The multi-scale pore structure of superfine pulverized coal. Part 2. Mesopore and closed pore morphology

Superfine pulverized coal exhibits extraordinary properties in thermal conversion processes like pyrolysis and combustion. The exploration of closed pores is of great significance for studying of the macromolecular structure, and developing advanced clean combustion technologies. In this work, the influences of coal rank and particle size on the mesopore characteristics (i.e., pore size, volume, and specific surface area), especially closed pores, were obtained by combining nitrogen adsorption and synchrotron-based small-angle X-ray radiation (SAXS). In addition, combined with the closed pore volume measured by He density and SAXS, research focuses on the formation and opening mechanisms of closed pores in different rank coals at a molecular level. The results indicate that both the bituminous (NMG) and anthracite coal (HN) have a few large open pores (>100 nm), with large amount of small mesopores and micropores (<2 nm). The HN coal has more open mesopores which have a smaller pore size than NMG coal. The closed pore proportion of HN coal in the studied pore size range is between 2 and 52 %, while that of NMG coal is between 47 and 67 %. In addition, both coals have a certain amount of closed pores formed by shear force and ash blockage, while the matrix compression also contribute to certain types of closed pores in NMG. These findings contribute to illustrating the closed pore structure in coal at the molecular level, which sheds light on constructing comprehensive macromolecular pore models of coal and thus promoting the development of clean coal utilization technologies.

A FRACTAL-BASED OIL TRANSPORT MODEL WITH UNCERTAINTY REDUCTION FOR A MULTI-SCALE SHALE PORE SYSTEM

The challenges of modeling shale oil transport are numerous and include strong solid-fluid interactions, fluid rheology, the multi-scale nature of the pore structure problem, and the different pore types involved. Until now, theoretical studies have not fully considered shale oil transport mechanisms and multi-scale pore structure properties. In this study, we propose a fractal-based oil transport model with uncertainty reduction for a multi-scale shale pore system. The fractal properties of the shale pore system are obtained using high-resolution scanning electron microscope (SEM) imaging combined with laboratory core sample gas permeability measurements to reduce the model uncertainty. This fractal-based oil transport model accounts for boundary slippage, fluid rheology, the adsorption layer, and different pore types. We further pinpoint the effects of the fractal properties (pore fractal dimension, tortuosity fractal dimension), the shale pore properties (pore type, pore size, total organic carbon in volume), and the fluid properties (yield stress, liquid slippage, adsorption layer) on the shale oil permeability and mobile oil saturation using the proposed model. The results reveal that the size of the inorganic pores has the largest influence on the shale oil transport properties, followed by the yield stress, tortuosity fractal dimension, and the fractal dimension of the inorganic pores.

Phase Behaviors of Gas Condensate at Pore Scale: Direct Visualization via Microfluidics and In-Situ CT Scanning

Summary Gas condensate is stored in multiscale pores, fractures, and vugs within geological formations. Confinement within these structures significantly influences the phase behavior of gas condensate, rendering it challenging to characterize through conventional bulk pressure/volume/temperature (PVT) measurements. In this study, we used microfluidics and in-situ computed tomography (CT) scanning to directly measure the upper dewpoint of gas condensate and the gas/oil ratio in porous media during depressurization. We used two microfluidic chips with different pore sizes to investigate the confinement effects on gas condensate phase behavior at various scales, including pores as small as 50 nm. Our results revealed a significant increase in the upper dewpoint within the pores compared to bulk PVT measurements, with a more pronounced deviation at smaller pore sizes. Additionally, the proportion of condensate oil in porous media exceeded that observed in bulk PVT measurements at the same pressure. To validate our microfluidic findings, we conducted in-situ CT scanning experiments using a porous media model created by packing quartz particles. CT scans revealed pores ranging from a few micrometers to over 100 micrometers. Consistently, we observed an increase in the upper dewpoint and liquid ratio within these pores. Our study provides crucial experimental evidence indicating that the phase behavior of gas condensate in porous media deviates from bulk PVT measurements. The observed increase in the upper dewpoint, even within micrometer-sized pores, has important implications for phase equilibrium calculations.

Thermo-responsive polymer-based Janus biogenic-nanosilica composite, part B: Experimental study as a multi-functional synergistic shale stabilizer for water-based drilling fluids

Wellbore instability caused by shale swelling and dispersion has always been a thorny issue for drilling complex formations. Nanotechnology has offered a revolutionary solution for shale stabilization in water-based drilling fluids (WBDFs). Recent technological advancement in nanomaterials for optimal performance using economically and environmentally sustainable pathway is an utmost desire in oil and gas industries. In this research, Janus nanocomposite (TRJN) with rice husk (RH)-extracted biogenic nanosilica core (RH-NSP) having the shell integrating synergistic benefits of thermo-responsive and amino-based materials, which was reported in our previous work was assessed as a novel shale stabilizer in drilling fluids. The product is anticipated to simultaneously improve the chemical inhibition and nanoplugging performances in shaly formations. TRJN was assessed as an additive in WBDF in comparison with symmetrical thermo-responsive polymer-based nanocomposite (TRN) counterpart and parent RH-NSP. Results from the rheological test revealed that, albeit both nanocomposites (i.e., TRN and TRJN) could provide a certain rheological control above the LCST compared to RH-NSP, but TRJN-blended fluid was less sensitive to increasing temperatures. The inhibitory effects were investigated based on cuttings hot rolling recovery rate, linear swelling rate, pressure penetration rate, tensile strength of shale reduction rate, and high temperatures and high pressures filtration loss. TRJN demonstrated higher chemical inhibition compared to RH-NSP and TRN, driving a better capacity to control pressure penetration. The higher mean free energy of TRJN/Na-Mt (sodium montmorillonite) system than TRN/Na-Mt system was an indication of its stronger interactions with Na-Mt particle, which triggered changes in microstructures, considering the results of zeta potential, FT-IR, surface wettability, X-ray diffraction, and SEM analyses. This, therefore contributing to firmly trap the TRJN into the multi-scale pore structure of the shale and beneficially promoted the construction of a tight hydrophobic plugging layer over the shale surface from the outward face. Lastly, the synergistic enhancement mechanisms of shale inhibition with TRJN was further proposed.

Research on petrophysical properties and porosity evolution of fractured coal mass under cyclic impact for coalbed methane exploitation

In the process of coalbed methane (CBM) extraction, coal seam penetration modification is frequently subjected to several cycle impact due to drilling-blasting method and deflagration fracturing method. Therefore, the split Hopkinson pressure bar (SHPB) was utilized to investigate the impact cycle effect and confining pressure effect on dynamic behavior of coal. Furthermore, the low-field nuclear magnetic resonance (NMR) was utilized to evaluate the modification of multiscale pore before and after 5 cycles impacts. Finally, the 3D profile scanner was utilized to quantify fracture surfaces and assess fracture roughness variation. The results showed that there existed the 6 MPa critical confining pressure that altered the dynamic mechanical properties of coal. Due to the combined effect of the confining pressure and cycle impact, the damage variable based on the energy method showed a log-normal distribution. With increasing strain rate, the micropores evolved into mesopores and macropores. There was a critical strain rate that caused the ratio of effective porosity to total porosity to shift from increasing to decreasing. Furthermore, the fracture roughness was shown to be positively correlated with the ratio and negatively correlated with seepage fractal dimension. The research findings can provide theoretical guidance for the safer and more efficient CBM exploitation.

Multiscale Modeling of the Membrane Electrode Assembly in Polymer Electrolyte Fuel Cells: Interaction of Pore Size and Wettability of Porous Layers

Design of the membrane electrode assembly (MEA) plays a crucial role in polymer electrolyte membrane fuel cells (PEMFCs). However, predictive modeling of mass, charge and heat transport considering primary microstructural information rather than previously measured effective transport properties is challenged by the multiscale pore structure found in an MEA, which spans around 4-5 orders of magnitude (from tens of nanometers in the catalyst layer to tens of micrometers in the gas diffusion layer). It is not clear how to incorporate microstructural information to guide design, while preserving moderate computational cost (see log-scale pore size distribution in an MEA in Figure a). In this work, a 3D two-phase, non-isothermal model is presented, which combines continuum and pore network (PN) formulations in a single CFD framework. Transport at fine pore scales of the catalyst layer and the microporous layer (below 1 um) is modeled by means of a continuum bundle-of-capillary-tubes (BCT) submodel, and transport at coarse pore scales of the gas diffusion layer and manufacturing defects above 1 um (cracks, gaps, etc.) by means of a continuum-based PN submodel. The BCT submodel accounts for microscopic (<20 nm), mesoscopic (20-300 nm) and macroscopic (300 nm-1 um) pore scales considering a subdivision of pore size distributions in the integral volume-averaged formulation. The interplay between pore size distribution and wettability of the various porous layers is analyzed, with a focus on the role of microporous layer wettability and defects, depending on operating relative humidity and temperature (see water distribution in Figure b from an invasion-percolation simulation). Best design practices of porous assemblies are extracted for maximum performance from a large computational campaign, including stochastic effects from different sample realizations. Analysis of degradation phenomena will be the focus of subsequent work to determine optimal microstructures arising from a trade-off between performance and durability. Figure 1

Synergistic impacts of fracture–matrix system on recovery rate during spontaneous imbibition in tight sandstones

Hydraulic fracturing is considered a promising technology for enhancing the recovery rate from tight oil/gas reservoirs. However, the hydraulic fracturing process inevitably leaves a portion of the fracturing fluid in the reservoir. This causes a noticeable decrease in the recovery rate of oil/gas. Thus, the flow mechanism of fracturing fluids in fracture–matrix system remains unclear. In this study, spontaneous imbibition (SI) experiments were conducted on sandstone samples with different fracture widths. Subsequently, magnetic resonance imaging (MRI) was used to monitor the displacement of the imbibition leading edge at the fracture edges. Finally, the recovery rate was quantified using the T2 spectra of multi-scale pores during the SI process. The experimental results revealed the following: (1) A significant contribution from the pre-rapid growth stage was observed during the SI process. (2) The heterogeneity of sandstone sample resulted in a rough imbibition leading edge and a decreasing gradient in fluid saturation over time. (3) The increase of distilled water in multi-scale pores during the SI process varied, specifically a preferential increase in micropores, followed by mesopores and macropores. (4) When the sandstone sample had a wider fracture, the synergistic effect between the fracture-matrix system was significantly enhanced, and a higher recovery rate was observed.

Multiscale fusion of tight sandstone digital rocks using attention-guided generative adversarial network

Tight sandstone exhibits heterogeneity due to its multiple mineral compositions and multiscale pore structures, presenting significant challenges for digital rock modeling. To capture the multiscale information inherent in heterogeneous rocks, we utilized an attention-guided generative adversarial network (AttentionGAN) to combine X-ray micro-computed tomography (micro-CT) and scanning electron microscope (SEM) images of tight sandstone, thereby producing large-scale, high-precision rock images. The reconstructed digital rock images comprehensively showcase the multiscale pore structures, including intergranular and clay micropores, as well as the distribution of minerals. Mineral content, two-point correlation functions, representative volume elements, and pore radius distributions were calculated for both the real and reconstructed images to confirm the efficacy of the proposed AttentionGAN. The results demonstrate that AttentionGAN is capable of generating high-precision rock images with a wide field of view. This offers a beneficial methodology for characterizing heterogeneous rocks and numerically simulating rock physics processes.

A novel lattice model to predict chloride diffusion coefficient of unsaturated cementitious materials based on multi-typed pore structure characteristics

This paper develops a novel lattice diffusive model to quantitatively study the chloride diffusion coefficient in unsaturated cementitious materials, in which the pore voxels are redistributed to make a better representation of a real microstructure of hardened cement paste. Considering the hierarchical microstructure and different drying-wetting cycles, water distributions in multiscale pore structures are modelled and the structure characteristics of water-filled pores, including water connectivity, water tortuosity and effective porosity, are computationally extracted based on that. A lattice diffusion network is established to predict relative chloride diffusion coefficient by combining the effect of both water saturation degree and pore structure characteristics. The predicted results are validated against experimental data, and a concise analytical equation is proposed to predict the relative chloride diffusion coefficient. The equation indicated that the relative chloride diffusion coefficient is proportional to water connectivity but inversely proportional to the square of water tortuosity. Besides, the lattice model's quantitative results reveal that the water connectivity and water tortuosity are highly related to pre-water loading processes, and influenced by the gel pore fraction, which in turn will affect the relative chloride diffusion coefficient. Compared with existing equations and non-redistributed models, the present model could improve the prediction accuracy significantly.

Phase behavior of gas condensate in fractured-vuggy porous media based on microfluidic technology and real-time computed tomography scanning

With the development of gas-condensate reservoirs, the phase behavior of gas condensate undergoes continuous changes. The phase behavior of gas condensate in PVT cylinders is different from that in fractured-vuggy gas-condensate reservoirs because multiscale pore structures can affect the phase behavior. By vuggy, we mean the formation of a reservoir contains cavity structures. To elucidate the influence of fractured-vuggy porous media on phase behavior of gas condensate, a microfluidic chip and a fractured-vuggy carbonate sample were prepared for pressure depletion experiments. The phase behavior of gas condensate in the chip during depressurization was observed via brightfield optical microscopy. The real-time computed tomography scanning was used to analyze pore structure of the sample and to obtain the phase behavior of gas condensate under different pressures. The results showed that the dew point pressure of the gas condensate was increased from 12.0 MPa in PVT cylinders to 14.8 MPa in nanochannels of the chip. The condensate liquid appeared in micropores at 14.4 MPa. The change of the average radius, coordination number, and tortuosity of condensate liquid ganglia were calculated and analyzed using pore network models. The shape factor and the Euler number were used to classify the condensate liquid into four types including clustered, branched, membranous, and droplets ganglia. The condensate liquid was generally distributed as clustered ganglia. This study provided evidence of the effect of fractured-vuggy porous media on phase behavior of gas condensate and clarified that the phase behavior of gas condensate varies in multi-scale pores.

Prediction of production indicators of fractured-vuggy reservoirs based on improved Graph Attention Network

Various storage and seepage spaces exist in fractured-vuggy carbonate reservoirs composed of multi-scale dissolution pores and fractures. The frequent regulation of the working system causes nonlinear and unstable production data along with complex water breakthrough characteristics, which lead to difficulty in real-time prediction. Traditional methods based on the water drive curve and autoregressive machine learning ignore the spatial correlation among production wells and local geological characteristics. To address these problems, we propose a model based on improved attention, the spatiotemporal multi-graph convolutional network (STMCN), for production prediction. The unit production wells are abstracted as directed graph network nodes to establish adjacency, connectivity and correlation graphs to extract the spatial and semantic features from different perspectives through graph convolutional networks. To depict the law of fluid movement, we realize the fusion of spatial and semantic information through dynamic routing. Aiming at the autocorrelation characteristics of production sequences, the model uses the self-attention mechanism to capture the dependencies in production sequences. The gating mechanism is designed to achieve dynamic production prediction by adaptively aggregating the spatiotemporal characteristics. This paper evaluates the predictive performance of our model by two real-world datasets of the Tahe Oilfield. The results show that the mean absolute error (MAE), root mean squared error (RMSE), and mean absolute percentage error (MAPE) of our proposed model for the prediction of well A2 were 3.19, 4.66 and 0.07, respectively, which were better than the relatively new baseline model.

3D Printed Cellulose Nanofiber Aerogel Scaffold with Hierarchical Porous Structures for Fast Solar-Driven Atmospheric Water Harvesting.

Hygroscopic salt-based composite sorbents are considered ideal candidates for solar-driven atmospheric water harvesting. The primary challenge for the sorbents lies in exposing more hygroscopically active sites to the surrounding air while preventing salt leakage. Herein, a hierarchically structured scaffold is constructed by integrating cellulose nanofiber and lithium chloride (LiCl) as building blocks through 3D printing combined with freeze-drying. The milli/micrometer multiscale pores can effectively confine LiCl and simultaneously provide a more exposed active area for water sorption and release, accelerating both water sorption and evaporation kinetics of the 3D printed structure. Compared to a conventional freeze-dried aerogel, the 3D printed scaffold exhibits a water sorption rate that is increased 1.6-fold, along with a more than 2.4-fold greater water release rate. An array of bilayer scaffolds is demonstrated, which can produce 0.63g g-1 day-1 of water outdoors under natural sunlight. This article provides a sustainable strategy for collecting freshwater from the atmosphere.

A pore network-based multiscale coupled model for rapid permeability prediction of tight sandstone gas

Tight sandstone has multiscale pore structures. Gas transport in tight sandstones involves several length scales and the complicated physics. Gas transport characteristics in tight sandstone can be represented by apparent gas permeability. Microstructure-based accurate and efficient calculation of gas permeability is challenging. By combining CT and SEM images along with statistical analysis, we present a novel pore network-based multiscale coupled model (MCPNM) to rapidly predict the apparent gas permeability. CT images are adopted to extract the large-scale pore network (LPNM) and the clay component, and SEM image is used to get the properties of small-scale pores. Upscaled model (UM) of small-scale pores is built via statistical analysis and then assigned to the clay domains. The LPNM and UM are coupled as MCPNM by the cross-scale connection structure with variable diameter. The pore spaces at several length scales and the flow characteristics in them are included in the MCPNM. We validate the MCPNM by comparing the calculated apparent gas permeability to the results of available multiscale pore network models and the experimental data. Compared with the available multiscale pore network models, MCPNM solves the accuracy/efficiency trade-off of tight gas permeability prediction. The effects of pressure, temperature, and gas type on gas permeability are studied. The MCPNM simplifies the permeability calculation process and can accurately and rapidly predict tight sandstone gas permeability.