Azuddin, Farhana Jaafar (Group Research & Technology, PETRONAS Institute of Petroleum Engineering, Heriot-Watt University) | Davis, Ivan (Institute of Petroleum Engineering, Heriot-Watt University) | Singleton, Mike (Institute of Petroleum Engineering, Heriot-Watt University) | Geiger, Sebastian (Institute of Petroleum Engineering, Heriot-Watt University) | Mackay, Eric (Institute of Petroleum Engineering, Heriot-Watt University) | Silva, Duarte (Institute of Petroleum Engineering, Heriot-Watt University)
When CO2 is injected into an aquifer, the injected CO2 is generally colder than the reservoir rock; this results in thermal gradients along the flow path. The temperature variation has an impact on CO2 solubility and the kinetics of any mineral reactions. Core flood experiments and associated reactive transport simulations were conducted to analyse thermal effects during CO2 injection in a dolomitic limestone aquifer and to quantify how CO2 solubility and mineral reactivity are affected.
The experiments were conducted by injecting acidified brine into an Edwards Limestone core sample. A back pressure of 400 psi and injection rates of 30 mL/hr and 300 mL/hr were used. A range of temperatures from 21 °C to 70 °C were examined. Changes in the outlet fluid composition and changes in porosity and permeability were analysed. A compositional simulation model was used to further analyse the experiments. The simulations were history-matched to the experimental data by changing the reactive surface area and the kinetic rate parameter. The calibrated model was then used to test the sensitivity to CO2 injection rate and temperature.
The impact of temperature on CO2-induced mineral reactions was observed from changes in mineral volume, porosity and permeability. The reaction rate constants estimated from the outlet solution concentrations are much lower than existing data for individual minerals. The estimated specific surface areas for carbonate minerals are in reasonable agreement with published values. The numerical investigations showed that at the lower temperatures, despite the reaction rates being slower, the solubility of the minerals was higher, and so as a result of these competing effects, moderately elevated calcium and magnesium concentrations were observed in the effluent. At higher temperatures, the solubilities of the minerals were lower, but now the reactions rates were higher, so similar effluent concentrations could be achieved. However, at higher flow rates, characterized by a lower Damköhler number, the residence times were shorter, and so lower effluent concentrations were observed. Additionally, the solubilities of calcite and dolomite varied to different extents with temperature, and so the calcium to magnesium molar ratio in the effluent brine increased with increasing temperature.
The change in mineral composition during CO2 injection varies between the near well zone and the deeper reservoir. Near the well where the temperatures will be lower, solubilities are elevated, but the kinetic reaction rates and residence times will be lower, somewhat limiting dissolution. Deeper in the aquifer the solubilities will be reduced and residence times will be longer, enabling an equilibrium to be established. Modelling is thus required to connect these flow regimes.
Modelling multiscale-multiphysics geology at field scales is non-trivial due to computational resources and data availability. At such scales it is common to use implicit modelling approaches as they remain a practical method of understanding the first order processes of complex systems. In this work we introduce a numerical framework for the simulation of geomechanical dual-continuum materials. Our framework is written as part of the open source MATLAB Reservoir Simulation Toolbox (MRST). We discretise the flow and mechanics problems using the finite volume method (FVM) and virtual element method (VEM) respectively. The result is a framework that ensures local mass conservation with respect to flow and is robust with respect to gridding. Solution of the coupled linear system can be achieved with either fully coupled or fixed-stress split solution strategies. We demonstrate our framework on an analytical comparison case and on a 3D geological grid case. In the former we observe a good match between analytical and numerical results, for both fully coupled and fixed-stress split strategies. In the latter, the geological model is gridded using a corner point grid that contains degenerate cells as well as hanging nodes. For the geological case, we observe physically plausible and intuitive results given the boundary conditions of the problem. Our initial testing with the framework suggests that the FEM-VEM discretisation has potential for conducting practical geomechanical studies of multiscale systems.
Azevedo, Leonardo (Cerena/Decivil, Instituto Superior Técnico) | Demyanov, Vasily (Institute of Petroleum Engineering, Heriot-Watt University) | Lopes, Diogo (Cerena/Decivil, Instituto Superior Técnico) | Soares, Amílcar (Cerena/Decivil, Instituto Superior Técnico) | Guerreiro, Luis (Partex Oil & Gas)
Geostatistical seismic inversion uses stochastic sequential simulation and co-simulation as the perturbation techniques to generate and perturb elastic models. These inversion methods allow retrieve high-resolution inverse models and assess the spatial uncertainty of the inverted properties. However, they assume a given number of a priori parametrization often considered known and certain, which is exactly reproduce in the final inverted models. This is the case of the top and base of main seismic units to which regional variogram models and histrograms are assigned. Nevertheless, the amount of existing well-log data (i.e., direct measurements) of the property to be inverted if often not enough to model variograms and its histograms are biased towards the more sand-prone facies. This work shows a consistent stochastic framework that allows to quantify uncertainties on these parameters which are associated with large-scale geological features. We couple stochastic adaptive sampling (i.e., particle swarm optimization) with global stochastic inversion to infer three-dimensional acoustic impedance from existing seismic reflection data. Key uncertain geological parameters are first identified, and reliable a priori distributions inferred from geological knowledge are assigned to each parameter. The type and shape of each distribution reflects the level of knowledge about this parameter. Then, particle swarm optimization is integrated as part of an iterative geostatistical seismic inversion methodology and these parameters are optimized along with the spatial distribution of acoustic impedance. At the end of the iterative procedure, we retrieve the best-fit inverse model of acoustic impedance along with the most probable value for the location of top and base of each seismic unit, the most likely histogram and variogram model per zone. We couple stochastic adaptive sampling (i.e., particle swarm optimization) with global stochastic inversion to infer three-dimensional acoustic impedance from existing seismic reflection data. Key uncertain geological parameters are first identified, and reliable a priori distributions of potential values are assigned to each parameter. The type and shape of each distribution reflects the level of knowledge about this parameter. Then, particle swarm optimization is integrated as part of an iterative geostatistical seismic inversion methodology and these parameters are optimized along with the spatial distribution of acoustic impedance. At the end of the iterative procedure we retrieve the best-fit inverse model of acoustic impedance along with the most probable value for the location of top and base of each seismic unit, the most likely histogram and variogram model per zone.
Hu, Yisheng (State Key Laboratory of Oil & Gas Reservoir Geology & Exploitation, Southwest Petroleum University) | Mackay, Eric (Institute of Petroleum Engineering, Heriot-Watt University)
Produced water chemical compositional data are collected from a carbonate reservoir which had been flooded by North Seawater for more than 20 years, so there is an opportunity to analyse the large amount of produced water data collected, understand the brine/brine and brine/rock interactions and explore the impact factors behind them. In some publications, core flood experimental tests were performed with chalk cores or carbonate columns in order to make an understanding of possible chemical reactions occurring triggered by injected water with different composition (Seawater, low salinity water or any other brine). However, most of the time the laboratory conditions where core flooding experiments are implemented cannot fully simulate the real reservoir conditions. Therefore, in this study, with the help of the valuable produced water dataset and some basic reservoir properties, a one-dimensional reactive transport model is developed to identify what in situ reactions were taking place in the carbonate reservoir triggered by seawater injection.
From the perspective of reservoir mineralogy, calcite, as the dominant mineral in the carbonate reservoir, is relatively more chemically reactive than quartz and feldspar which are usually found in sandstone. Whether calcite is initially and dominantly present in the carbonate reservoir rock is dissolved under seawater flooding or not is the first key issue we focused on. The effects of calcite dissolution on the sulphate scaling reactions due to incompatible brine mixing and the potential occurrence of carbonate mineral precipitation induced by calcite dissolution are investigated and discussed in detail. The comparison of simulation results from the isothermal model and the non-isothermal model show the important role of temperature during geochemical processes. The partitioning of CO2 from the hydrocarbon phase into injected brine was considered through calculation of the composition of reacted seawater equilibrated with the CO2 gas phase with fixed partial pressure (equivalent with CO2 content), then subsequently the impact of CO2 interactions on the calcite, dolomite and huntite mineral reactions are studied and explained. We also use calculation results from the model to match the observed field data to demonstrate the possibility of ion exchange occurring in the chalk reservoir.
Jarrahian, Kh. (Institute of Petroleum Engineering, Heriot-Watt University) | Sorbie, K. S. (Institute of Petroleum Engineering, Heriot-Watt University) | Singleton, M. A. (Institute of Petroleum Engineering, Heriot-Watt University) | Boak, L. S. (Institute of Petroleum Engineering, Heriot-Watt University) | Graham, A. J. (Institute of Petroleum Engineering, Heriot-Watt University)
The bulk "apparent adsorption" behavior (Γapp, vs. Cf) of 2 polymeric scale inhibitors (SI), PPCA and PFC, onto carbonate mineral substrates has been studied for initial solution pH values of pH 2, 4 and 6. The 2 carbonate minerals used, calcite and dolomite, are much more chemically reactive than sandstone minerals (e.g. quartz, feldspars, clays etc.) which have already been studied extensively. In nearly all cases, precipitates formed at higher SI concentrations were due to the formation of sparingly soluble SI/Ca complexes. A systematic study has been carried out on the SI/Ca precipitates formed, by applying both ESEM/EDX and particle size analysis (PSA), and this identifies the morphology and the approximate composition of the precipitates.
For PPCA, at all initial solution pH values, regions of pure adsorption (Γ) ([PPCA] <100ppm) and coupled adsorption/ precipitation (Γ/Π) are clearly observed for both calcite and dolomite. PFC at pH = 4 and 6 also showed very similar behavior with a region of pure adsorption (Γ) for [PFC] < 500ppm and a region of coupled adsorption/precipitation (Γ/Π) above this level. However, the PFC/calcite case at pH 2 showed only pure adsorption, while the PFC/dolomite case at pH 2 again showed coupled adsorption/ precipitation at higher PFC concentrations. For both SIs on both carbonate substrates, precipitation is the more dominant mechanism for SI retention than adsorption above a minimum concentration of ~100 – 500 ppm SI. The actual amount of precipitate formed varies from case to case, depending on the specific SI, substrate (calcite/dolomite) and initial pH (pH 2, 4 and 6).
Although the qualitative behavior of both PPCA and PFC was similar on both carbonate substrates, the apparent adsorption of PPCA was higher on calcite than on dolomite; PFC apparent adsorption was higher on dolomite than on calcite. It is discussed in the paper how these observations are related to the reactivity of the different carbonate minerals, the resulting final pH (which affects the dissociation of the SI), Ca-SI binding and the solubility of the resulting complex.
Lopes, Diogo (CERENA/Instituto Superior Técnico) | Azevedo, Leonardo (CERENA/Instituto Superior Técnico) | Demyanov, Vasily (Institute of Petroleum Engineering, Heriot-Watt University) | Guerreiro, Luís (Partex Oil & Gas)
Petro-elastic models retrieved from seismic inversion allow inferring the spatial distribution of the subsurface properties of interest, for example, acoustic impedance and porosity, to allow better reservoir characterization and field management (
There is a potential in simultaneously assessing uncertainty and integrating data with different resolution (i.e., well-log and seismic reflection data) through iterative geostatistical seismic inversion methodologies (e.g.
This work introduces a statistical framework that couples stochastic adaptive sampling and Bayesian inference to assess uncertainty associated with the large scale geological parameters such as: regional variogram models ranges, regional probability distribution for the elastic property of interest and the structural and stratigraphic interpretation. The proposed methodology was tested and implemented in a real case study from an onshore Middle East field. The results show how the synthetic seismic reflection data matches the recorded one with respect to uncertainty in the large-scale geological parameters.
Presentation Date: Tuesday, September 26, 2017
Start Time: 10:10 AM
Presentation Type: ORAL
Joonaki, Edris (Institute of Petroleum Engineering, Heriot-Watt University) | Burgass, Rod (Institute of Petroleum Engineering, Heriot-Watt University) | Tohidi, Bahman (Institute of Petroleum Engineering, Heriot-Watt University)
In this communication, a novel approach for combined Enhanced Oil Recovery (EOR) and asphaltene inhibition is proposed employing a New Class of Asphaltene Inhibitor (NCAI).
Following primary and secondary oil production, appropriate EOR techniques can increase oil recovery. Gas injection is commonly used, however there are risks associated with asphaltene precipitation and deposition which could result in flow assurance concerns in addition to changes in wettability.
This work investigates the effect of a newly developed AI on the wettability of synthesized substrates. An Environmental Scanning Electron Microscopy (ESEM) technique is utilized for monitoring the wettability alterations of the substrates caused by NCAI. This research also presents adsorption equilibrium and kinetics data for the NCAI on sandstone rocks. Batch experiments are conducted at various temperatures (from 25 °C to 75 °C) to illustrate the adsorption behaviour of this new chemical compound on reservoir rock surfaces. Conductivity measurements are utilized to determine the amount of the chemical compound adsorbed on crushed sandstone rock surfaces. The adsorption results are examined by employing various adsorption isotherms of Langmuir, Freundlich, Temkin, and Linear. The obtained experimental adsorption kinetic results are described and evaluated by pseudo-first-order, pseudo-second-order and intra-particle diffusion models. It can be inferred from the results that higher NCAI concentrations results in higher adsorption of NCAI onto rock surfaces. These experimental and modelling investigations provide a new tool for understanding the usefulness of this NCAI as a surface active agent in chemical based EOR techniques, especially for integrated processes with asphaltene precipitation and deposition inhibition.
Graham, A. J. (Institute of Petroleum Engineering, Heriot-Watt University) | Singleton, M. A. (Institute of Petroleum Engineering, Heriot-Watt University) | Sorbie, K. S. (Institute of Petroleum Engineering, Heriot-Watt University) | Collins, Ian R (Upstream Technology Chemical Sciences Centre of Expertise)
Work was undertaken to systematically investigate the factors that affect the formation of zinc sulfide in an aqueous system. Experiments were performed at a series of temperatures from room temperature up to 90 °C, at a range of initial pH values and in two brines systems. The effect of pH was further examined by changing the salt from Na2S.9H2O to NaSH.xH2O, therefore changing the initial sulfide source from S2- to HS-, as part of an ongoing method development strategy.
A variation of the standard barium sulfate static bottle test was used, in which the two bottles to be mixed contained aqueous "H2S" and zinc, respectively. Having pH adjusted the zinc brine to values calculated by the FAST sulfide model, the brines were pre-heated to the required temperature and mixed. Aliquots were removed at 2, 4 and 24 hours to perform elemental analysis by ICP and pH measurements were performed on all samples once they had returned to room temperature. In addition, particle size analysis and ESEM examination of the resulting precipitate were also performed for a subset of the samples prepared.
The reaction between zinc and aqueous "H2S" was quantitative at all temperatures up to 90 °C and in both brines. The final pH values of the supernatant were independent of the zinc brine pH and instead were dependent on the molar ratio of zinc and sulfide ions. A high pH, sulfide dominated, and a low pH, zinc dominated, plateau region were seen with a sharp inflection between the two. As a consequence, reaching field representative pH values was seen to be extremely difficult while retaining the ability to alter the relative concentrations of the reacting ions. Altering the sulfide source yielded the same trend, albeit with different absolute values. These observations have been rationalised with reference to the thermodynamic constants governing the reaction through scale prediction modelling.
The work presented here provides a greater understanding of the factors governing the formation of zinc sulfide scale and the considerations required for more industrially relevant formation and inhibition experiments in the future.
The increased reliance on CO2 sequestration and Enhanced Oil Recovery techniques such as tertiary gas injection or WAG for oil field development has necessitated an in depth analysis of three-phase flow in porous media for reservoir modelling and production enhancement techniques. Prediction of three-phase flow phenomena requires a sound understanding of the fundamental flow physics in water-wet, intermediate and oil-wet rocks in order to derive physically robust flow functions, namely relative permeability and capillary pressure curves.
A new pore-network model for rocks with arbitrary wettability employing the thermodynamic oil formation and collapse criteria was developed and successfully validated by Al Dhahli et al. in January 2013. This model has been used in the current study to assess the influence of network extraction methodologies, wettability alteration after ageing and capillary effects on multiphase flow characteristics at pore level. Also, the study aims to understand the effects of capillary height functions on flow phenomena. These analogies are successfully illustrated for the Berea sandstone network models.
The flow functions for pore-network analyses were subsequently imported to a reservoir model so as to carry out pore to reservoir scale integration. Flow function normalization and de-normalization were performed with respect to the flowing fluid saturations. This aided in better curve interpolations and fewer curve smoothening. Consequently, curve interpolations based on flowing saturations are suggested for three-phase pore- to reservoir scale integrations in future analyses.
Pore network extraction and flow modelling has contributed to a paradigm shift in today's reservoir modelling approach. However, these applications are presently being limited to research level only and pore scale modelling is still in a relatively nascent stage. Near reservoir conditions such as intermediate wettability, variable flow functions and reservoir heterogeneity form a part of future work in this subject. Further study in this field can delineate dynamic properties of the reservoir, bringing the industry one step closer to effectively recovering immobile/trapped fluid from brown field reservoirs.
Sand production is considered one of the major problems that would affect the economical value of any EOR project. The produced sand will start to accumulate in the wellbore till it kills the Well. This is done by increasing the pressure drop (both frictional and hydrostatic) in the borehole to a point where no more fluid can be produced. It can also exhibit a mechanical damage to the pipes, chokes, surface facility lines and would require continues cleaning.
There are several sand control equipment that can be installed whether in the wellbore or in the formation itself. However, these sand control equipment not only add a cost element to the whole project, but also most of them reduce the Productivity Index of the well. Consequently more wells need to be drilled in order to compensate the production loss due to the installation of sand control equipment.
This paper examines the effect of using Sucker Rod Pump (SRP) and All Metal Progressive Cavity Pump (AMPCP) on the sand production. Both pumps can handle solids to some degree, however, the mechanism of each pump will affect the building of sand arch and thus will impact the amount of sand volume produced.
By comparing the sand production profiles of two vertical wells under Cyclic Steam Stimulation (CSS) completed in the same zone, it was found that the use of AMPCP can help reduce sand production significantly over the SRP. This is attributed to the difference in working mechanisms of both pumps. In AMPCP, the fluid is being pumped by a smooth movement, while in SRP, the upstroke and downstroke movement of the pump have an effect similar to the plunger, which adversely impact the sand arch building. As a result, SRP has a higher tendency to produce more sand compared to that of AMPCP.
Hundreds of wells have been drilled vertically in order to achieve the target production of the Shaly-sand reservoir. Due to the shallowness of the reservoir at depth of around 600’ and the unconsolidated nature of sandstone, sand production soon showed its hindrance potential in the development of the field. Several attempts were made to control sand production, such as using different perforation charges with different phasing and sand screens. The application of sand screens was in efficient as they were plugged within 2-3 weeks of production. They would require a cleaning operation to resume the production and such operation proved to be very costly.
Since many wells were drilled for the development of the Shaly-sand oil field, simply buying sand control equipment for this large number of wells will result in the project being un-economic. So another way of controlling the sand production had to be evaluated to implement a reasonable sand production technique with better economics.
Progressive Cavity Pumps (PCP) were used in the cold pilots that were conducted in the Shaly-sand oil field here in Kuwait, but due to the nature of CSS operation, AMPCP pumps were needed to handle the heat of the injected steam and the heat of the produced fluid. The difference between PCP and the AMPCP pumps is that the stator is made of rubber in the PCP pump while it is made of metal in AMPCP, which tolerates high temperatures of the CSS operation. Beyond this point, they share the same components and they work in the same mechanism. Another advantage of the AMPCP is that steam injection can be done directly while the pump is in the well, by simply unscrewing the rotor out of the stator. Once the steam injection is finished, the rotor is screwed back in the stator and the pump can be started.