Jarrahian, Khosro (Heriot-Watt University) | Sorbie, Kenneth (Heriot-Watt University) | Singleton, Michael (Heriot-Watt University) | Boak, Lorraine (Heriot-Watt University) | Graham, Alexander (Heriot-Watt University)
Scale inhibitor (SI) squeeze treatments in carbonate reservoirs are often affected by the chemical reactivity between the SI and the carbonate mineral substrate. This chemical interaction may lead to a controlled precipitation of the SI through the formation of a sparingly soluble Ca/SI complex which can lead to an extended squeeze lifetime. However, the same interaction may in some cases lead to uncontrolled SI precipitation causing near-well formation damage in the treated zone. This paper presents a detailed study of the various retention mechanisms of SI in carbonate formations, considering system variables such as the (carbonate) formation mineralogy, the type of SI and the system conditions. Apparent adsorption (Γapp) experiments, described previously (
For all SIs, both adsorption (Γ) and precipitation (�?) retention mechanisms were observed, with the dominant mechanism depending on SI chemistry, temperature and mineralogy. Differences were observed between the "apparent adsorption" (Γapp) levels of polymeric, phosphonate and phosphate ester scale inhibitors, as follows: For the polymeric SIs (PPCA, PFC and VS-Co), the highest retention levels were observed at low pH for all carbonate substrates, due to the increase in divalent cations (Ca2+ and Mg2+) available from rock dissolution for SI-M2+ precipitation. For phosphonate (DETPMP) and phosphate ester (PAPE) SIs, the retention level was greatest at higher pH values, as the SI functional groups were more dissociated and hence available for complexation with M2+ ions. The polymeric VS-Co showed the lowest amount of precipitation (Γapp ~ 1.2 mg/g) in contact with dolomite substrate due to the presence of sulphonate groups (low pKa); indeed this showed low Γapp which was predominantly pure adsorption. However, a small amount of precipitate was observed by ESEM/EDX and PSA. For polymeric inhibitors, the retention level (Γapp) was highest on calcite (highest relative calcium content), followed by limestone and then dolomite. Phosphonate and phosphate ester SIs showed the highest retention levels on dolomite (higher final solution pH and more SI dissociated), followed by limestone and calcite. For all SI species, higher retention (more precipitation, �?) was observed at elevated temperature. At lower temperatures, a more extended region of pure adsorption was observed for all SIs.
For the polymeric SIs (PPCA, PFC and VS-Co), the highest retention levels were observed at low pH for all carbonate substrates, due to the increase in divalent cations (Ca2+ and Mg2+) available from rock dissolution for SI-M2+ precipitation. For phosphonate (DETPMP) and phosphate ester (PAPE) SIs, the retention level was greatest at higher pH values, as the SI functional groups were more dissociated and hence available for complexation with M2+ ions.
The polymeric VS-Co showed the lowest amount of precipitation (Γapp ~ 1.2 mg/g) in contact with dolomite substrate due to the presence of sulphonate groups (low pKa); indeed this showed low Γapp which was predominantly pure adsorption. However, a small amount of precipitate was observed by ESEM/EDX and PSA.
For polymeric inhibitors, the retention level (Γapp) was highest on calcite (highest relative calcium content), followed by limestone and then dolomite. Phosphonate and phosphate ester SIs showed the highest retention levels on dolomite (higher final solution pH and more SI dissociated), followed by limestone and calcite.
For all SI species, higher retention (more precipitation, �?) was observed at elevated temperature. At lower temperatures, a more extended region of pure adsorption was observed for all SIs.
The information presented in this study will help us in SI product selection for application of squeeze treatments with longer squeeze lifetimes in carbonate reservoir based on mineralogy and reservoir conditions. In addition, this study provides valuable data for validating models of the SI/Carbonate/Ca/Mg system which can be incorporated in squeeze design simulations.
Baghban Salehi, Mahsa (Chemistry & Chemical Engineering Research Center of Iran) | Mousavi Moghadam, Asefe (Chemistry & Chemical Engineering Research Center of Iran) | Jarrahian, Khosro (Heriot-Watt University)
Preformed Particle Gel (PPG) is an appropriate solution for conformance control and managing water production in low permeable reservoirs. Rheological behavior evaluation of these deformable particles is a key factor in designing composition to achieve the best conformance control treatment due to the viscoelastic behavior of these particles along with their swelling. The purpose of this paper is to evaluate the network parameters of PPGs through swelling tests, rheology and determining its role in maintaining their structural strength. Several PPG hydrogels were prepared by varying the concentrations of polyacrylamide and Cr(OAc)3 as copolymer and crosslinker, respectively. The characterization of these hydrogels was performed using Scanning Electron Micrographs (SEM), Electron Dispersion X-ray analysis (EDX), Environmental Scanning Electron Microscopy (ESEM), ThermoGravimetric Analysis (TGA), and Differential ThermoGravimetry (DTG). The correlation between reaction conditions and network parameters of polymer networks such as, molecular weight of the polymer chain between two neighboring crosslinks, crosslink density, and size fraction have been determined. The swelling of the hydrogels was found through the Fickian diffusion mechanism. In this case, the diffusion rate of water in the 3D structure of the hydrogel is less than the relaxation of the polymeric chain, resulting in a significant increase in the PPG particles volume. As PPG was invaded such as in the reservoir by formation water or oil, repeatedly, the sensitivity factor was measured to ensure the swelling in the electrolyte solution. Based on rheological tests, the dynamic modulus of the swelled PPG was strongly dependent on the concentration and consequently network parameters. Also, through the optimization of the network parameters, the appropriate composition from the point of view of strength (complex modulus of 4×104 Pa) and salt sensitivity of 0.5 was presented. In addition, the results of the TGA/DTG test demonstrated the thermal stability of the sample was in temperature range 245 to 340°C. The determination and analysis of the network parameter is the novel technique for predicting the hydrogel performance in porous media and investigating its strength under harsh reservoir conditions. In other words, determination of the network parameter can be a shortcut to ensure the success of the gel performance in porous media.
Produced water composition analysis provides evidence of what geochemical reactions are taking place in the reservoir. This information can be useful for predicting and managing oilfield mineral scale resulting from brine supersaturation.
This paper presents results of a study of the produced brine compositions from three wells in a field operated in the North Sea, with geochemical modelling complementing the analysis. The findings presented in this work provide evidence of magnesium depletion and sulphate retardation in a sandstone reservoir at 130° C.
This adjusted formation water composition was then used for calculations of the injection water fraction in each of the produced water samples. The Reacting Ions Toolkit was used to plot data in a variety of formats, including ion concentration vs. ion concentration, ion concentration vs. injection water fraction and ion concentration vs. time to identify trends and to examine the extent of involvement of the various ions in geochemical reactions.
The breakthrough of sulphate, a component primarily introduced during seawater flooding, was retarded during injection water breakthrough. Observed sulphate concentrations were lower than predicted for the case of brine/brine interactions only. The implication of this sulphate reduction was lower minimum inhibitor concentration required to control scale formation and longer squeeze treatment lifetimes for the operator.
A brine/rock interaction mechanism was proposed that involves magnesium depletion and is reproduced in the reactive transport model. 1D reactive transport modelling was performed to match possible
CO2 Water-Alternating-Gas injection (CO2 WAG), which involves complex phase and flow behaviour, is still a challenging task to simulate and predict accurately. In this paper, we focus specifically on the regime of viscous fingering flow in CO2 WAG in heterogeneous systems because of its importance. We investigated two key physical processes that occur during near-Miscible WAG (nMWAG) processes, namely oil stripping (Mechanism 1, M1) and low-interfacial-tension (IFT) film flow effects (Mechanism 2, M2). The low IFT effects in M2 manifest themselves in an increased mobility of oil phase due to film flow process (discussed below). The importance of properly simulating the interaction of viscous, compositional (M1), and low-interfacial-tension effects (M2) is clearly demonstrated in this study. Our specific aim is to improve the modelling of CO2 displacement in the transition from immiscible to miscible flows in CO2 WAG processes.
We simulated both immiscible and near-miscible CO2 WAG and also continuous CO2 displacements with unfavourable mobility ratios for 1D and 2D systems. 2D heterogeneous permeability fields were generated with certain Dykstra-Parsons coefficients and dimensionless correlation ranges. IFT (σgo) was calculated by the simulator as part of the compositional simulation using the McLeod-Sugden equation. The consequent IFT effects on relative permeability was imposed using two commonly used models, i.e.
We tested various combinations of oil-stripping effects (M1) and IFT effects (M2) to evaluate the potential impact of each mechanism on the flow behaviour such as the local displacement efficiency, the tracking of tracer flow and the ultimate oil recovery. Oil bypassed by viscous fingering/local heterogeneity, can be efficiently recovered by WAG in the cases where both M1 and M2 are taken into account (as opposed to either mechanism being considered alone). Through tracer analysis, we found that a major recovery mechanism in near-miscible displacement was
WAG (Water-Alternating-Gas) schemes have been applied in Brazilian carbonate reservoirs aiming to minimize residual oil saturation and gas flaring by recycling CO2 naturally being produced alongside hydrocarbon gas. However, applying WAG injection in highly reactive and heterogeneous carbonate rocks can potentially create severe scaling problems. This work develops a reactive transport simulation-based workflow to evaluate the impact of key WAG design parameters on oil recovery, scale deposition risk and CO2 storage to support multi-objective decision-making.
Compositional simulations of WAG scenarios were performed as part of a sensitivity study followed by statistical analysis in order to quantify to what extent the outcomes of interest are sensitive to variations on four WAG design parameters: WAG ratio, CO2 concentration in the injection gas stream, injection rate and solvent slug-size. We established an Equation-of-State (EoS) using PVT data, a representative geochemical model and well constrains designed to control production of injected fluids. Scale risk was assessed by calcite changes around the wells, precipitation in well tubing and surface facilities, and water breakthrough.
Results of this study showed that values of calcite rate constant (
Ultimately, we demonstrate the importance of integrating multiphase miscible displacement with geochemical reactions while modeling complex CO2-EOR in carbonate reservoirs and address how key design parameters impact our desired outcomes, knowledge that promotes a more robust decision-making framework.
Al Kalbani, M. M. (Heriot-Watt University) | Jordan, M. M. (Nalco Champion) | Mackay, E. J. (Heriot-Watt University) | Sorbie, K. S. (Heriot-Watt University) | Nghiem, L. (Computer Modelling Group Ltd.)
Barium Sulphate (BaSO4) scale is a serious problem that is encountered during oilfield production and has been studied mainly for fields undergoing water flooding. Chemical Enhanced Oil Recovery (cEOR) processes involve interactions between the injected brine and the formation brine, rock and oil. Very little work has appeared in the literature on how cEOR processes can influence the severity of the mineral scaling problem that occurs in the field and how this can be managed. This study investigates barium and sulphate co-production behaviour, the deposition of BaSO4 in the formation and in the producer wellbore, and its inhibition during polymer (P), surfactant (S) and surfactant-polymer (SP) flooding cEOR processes.
Reservoir simulation has been used in this study, employing homogenous and heterogeneous 2D areal and vertical models. Data from the literature are used to define the parameters controlling the physical and chemical functionality of surfactant and polymer (e.g. oil-water interfacial tension, IFT, polymer viscosity and surfactant and polymer adsorption). Assessment is made of the minimum inhibitor concentration (MIC) required to control scale that is predicted to occur due to changes in brine composition induced by the water and chemical flooding processes. The expected retention and release of a phosphonate scale inhibitor during squeeze treatments in the production wells is modelled.
The high viscosity and more stable polymer slug reduces the mixing between the injected and the formation brines, reducing BaSO4 scale precipitation in the formation and delaying the potential scale risk in the producer wellbore compared to normal water flooding. Polymer adsorption causes retardation of the polymer front compared to the sulphate front, accelerating the scale risk in the wellbore. Polymer with low salinity make-up brine and low sulphate concentration not only improves polymer viscosity and enhances recovery, it also delays and reduces the scale risk in the formation and the producer. During surfactant flooding, from an oil recovery perspective, the optimal phase type and salinity can be any of the three microemulsion phase types, depending on the system multiphase parameters. However, the scaling risk can be different to that in the water flooding case, depending on the IFT, ME phase type, the injected salinity and sulphate concentration. In SP flooding, low salinity make-up brine is preferred to enhance oil recovery, and it also delays and reduces scale risk. The impact of the changing brine composition due to ion reactions affected the required MIC values over time. The impact of the MIC and salinity changes on inhibitor retention and release then influences the treatment volumes required to control scale over field life.
The study shows that barium and sulphate co-production and the evolving scale risk depend on the mobility ratio (which is determined by the injected brine and oil viscosities), on the oil-water IFT and on the level of chemical adsorption. The severity of the scale risk is also impacted by the flood techniques utilised, with the extent of reservoir reactions have an effect on the MIC required to control scale and the squeeze treatment volumes required to maintain production after breakthrough.
In this work, we investigate different approaches for history matching of imperfect reservoir models while accounting for model error. The first approach (base case scenario) relies on direct Bayesian inversion using iterative ensemble smoothing with annealing schedules without accounting for model error. In the second approach the residual, obtained after calibration, is used to iteratively update the covariance matrix of the total error, that is a combination of model error and data error. In the third approach, PCA-based error model is used to represent the model discrepancy during history matching. However, the prior for the PCA weights is quite subjective and is generally hard to define. Here the prior statistics of model error parameters are estimated using pairs of accurate and inaccurate models. The fourth approach, inspired from
The effects of H2S on system integrity, sulphide scaling potential and health and safety in oil and gas production is well recognized and understood. However, as part of a wider study on pH dependent scale predictions, the authors have identified an additional challenge associated with the presence and/or development of H2S in reservoirs containing carbonates: higher H2S concentration reflects in higher calcium carbonate scaling potential. The intention of this work is to demonstrate the impact of H2S using a real field case scenario and investigate how the variability in water cut, aqueous phase composition, CO2 and H2S concentration can impact the well carbonate scaling potential and ultimately its productivity.
To model pH dependent scales correctly, it is necessary to integrate PVT calculations with the aqueous phase thermodynamic mineral scaling calculations. This has been extensively discussed in previous publications by the authors. For this work, a commercial integrated PVT and scale prediction software package was used to determine the scale prediction profile from reservoir to the first stage of topside separation. In addition, to investigate the impact of PVT on the final results, a second PVT software employing a different equation of state (EOS) is used and the results obtained from this calculations are coupled with the same aqueous phase model using the Heriot-Watt scale prediction workflow.
The well selected for this study shows productivity issues as well as signs of presence of calcium carbonate scale. However, scale prediction calculations carried out in the past did not show any potential for carbonate scale formation at the given conditions. After rigorously accounting for variations in water cut over time, as well as for increased H2S due to reservoir souring, our work clearly shows a correlation between a gradual loss of well productivity and carbonate scaling potential.
This work clearly demonstrates the impact of H2S on calcium carbonate scaling potential and highlights the importance of correctly modelling CO2 and H2S partitioning in gas/oil/water at the different stages of production, from reservoir to topside separation. Following this study, it has also been possible to offer specific well treatment and testing recommendations to verify the results and try to obtain improvements in production efficiency.
Moreover, the application of our approach to a real field scenario shows how some field findings associated with carbonate scale problems can be explained only by correctly modelling the full three phase system (oil, gas and water). Some aspects of this approach are frequently overlooked and not linked correctly to carbonate scale formation.
Zaidin, Mohd Fakrumie (Petronas Research Sdn Bhd) | Kantaatmadja, Budi Priyatna (Petronas Research Sdn Bhd) | Chapoy, Antonin (Heriot-Watt University) | Ahmadi, Pehzman (Heriot-Watt University) | Burgass, Rod (Heriot-Watt University)
The X field is one of PETRONAS's Research & Development (R&D) initiative plans involving separation of high CO2-Hydrocarbon gas and CO2 storage in offshore Malaysia. The X field is a high pressure high temperature (HPHT) carbonate reservoir with a temperature of 423 K and pressure of 36.0 MPa with about 500 m of gas column. It was chosen as a candidate due to its well and reservoir data completeness. The plan is technically challenging as it involves re-injecting produced supercritical CO2 back into an aquifer reservoir for permanent storage. Recently acquired X field DST data indicates the presence of CO2 in the aquifer, up to a level nearing saturation. Information of the initial CO2 concentration level in the aquifer reservoir is critical to ensure the success of the CO2 injection. Predictions on this initial CO2 solubility have been made using available well data, however the reliability of the results has to be validated by an experimental study. Therefore, an extensive experimental approach to measure initial CO2 solubility in the X field aquifer reservoir has been conducted. As pressure, temperature and salinity are the important key parameters that influence CO2 solubility, detailed information about X field gas and aquifer brine compositions are well determined prior to the solubility measurement. Utilizing lab facilities at Heriot-Watt University (HWU), measurements were conducted at T=423.15 K and pressure at 36.0 MPa to mimic the X field aquifer conditions. The experimental results obtained are compared against available literature data, Duan Model and sCPA-EoS model and reasonable agreements were observed. Experimental results indicated that the X field aquifer is not fully saturated with CO2 and it could accommodate an additional 6 mol% of CO2 dissolved in the brine. In addition, approximately 6 mol% of hydrocarbon will be recovered from the same aquifer system as a result of CO2 injection due to the CO2-Hydrocarbon displacement. This paper details lab measurements of initial CO2 solubility in the X field aquifer, including preparation, experimental procedure, results and discussion as well as suggested future works. Reservoir simulation incorporating the experimental data obtained from this study is necessary and recommended, for getting a full picture of the CO2 injection program for the current Carbon Capture Utilization & Storage (CCUS) project.
This paper is based on the analysis of the ultrasonic/sonic data of the 9 5/8-in. casing logging of the 14 wells of the Varg field within the Norwegian Continental Shelf. While writing this papper Varg field was undergoing a plug and abandonment (P&A) phase after 19 years of production. High-quality bonding is observed behind the 9 5/8-in. casing far above expected theoretical top of cement within single casing areas. This bonding is attributed to the formation influence. Formation is used as an alternative to traditional cement barriers during P&A, and its use is regulated by the legislation.
The paper aims to develop better understanding of the mechanisms responsible for formation bonding development. The percentage of observed bonding at "high" and "high and moderate-to-high" quality is calculated within each well and is related to the various factors that could influence formation bonding development. Factors such as duration of time lapsed from well completion to well logging, type of well (producer versus injector), geological formation, type of drilling mud used, duration of production periods, volumes of production, and well deviation and azimuth were looked at to determine observable trends and relationships.
The results of the study allowed us to conclude which factors are critical or influence formation bonding. Based on the observations, recommendations can be made for the selection of the first well to be logged on the planned P&A campaigns. Correct selection of the first well saves time and resources on the formation testing for the qualification of the formation as a barrier.