Technique Overcomes Scarcity of Solubility Data in Scale Prediction for HP/HT Conditions
The complete paper describes the generation of additional fundamental solubility data under HP/HT conditions and comparison of the obtained values with several existing models.
Accurate scale-prediction modeling is only possible when reliable mineral solubility data are available under the required conditions. It is recognized that the relative paucity of high-pressure/high-temperature (HP/HT) solubility data can result in inaccurate predictions because current models extrapolate from data obtained under more conventional conditions. The complete paper describes the generation of additional fundamental solubility data under HP/HT conditions and comparison of the obtained values with several existing models.
The limited availability of HP/HT mineral solubility data in simple brine conditions and the paucity of solubility data in complex brine systems means that existing scale-prediction models, including those used in conventional software in common industry use, are forced to extrapolate, resulting in inaccurate and potentially misleading predictions.
Efforts have been made in recent years to address this issue by generating additional solubility data under HP/HT conditions to allow improved models to be developed. It is apparent from the reported work, and comparisons with previous literature, that these measurements are not trivial and require detailed understanding, development, and validation to ensure that reliable measurements are obtained. However, full details, comparison, and verification of the various experimental methods used are rarely presented and discussed, particularly in older works. The experimental method used must demonstrate that equilibrium solubility has been reached, preferably in a reasonable laboratory time period. The complete paper discusses the experimental considerations that must be taken into account when generating HP/HT solubilities, and presents methods to confirm that true equilibrium conditions have been reached.
Solubility measurements were made using a purpose-built laboratory test rig capable of attaining 250°C and 30,000 psi, though data described here were obtained between 50 and 200°C and between 10,000 and 19,000 psi. The design and validation of this rig against literature data has been previously described in detail in paper SPE 184568, and is only presented in summary in the complete paper. A schematic of the apparatus is shown in Fig. 1, while Fig. 2 shows the pressure vessel used in the system.
There are two primary experimental approaches that can be applied, and both have been investigated in the course of this work, using the same combined flow-through and static method. The first is to approach equilibrium from below by dissolving the test mineral into the brine composition of interest. In the second, the precipitation approach, a mixture of two fluids, one containing the relevant cation and one the corresponding anion, are mixed to create a solution oversaturated slightly in the test mineral before being applied to the sample. This seeding is intended to result in crystallization of the excess solute, relieving the supersaturation and resulting in a solution containing equilibrium concentration of the mineral under the chosen conditions. However, the rate of the dissolution and precipitation processes can vary considerably for a given mineral and, if this is not recognized, inaccurate solubility measurements will result.
Solubility measurements under HP/HT conditions using the dissolution approach revealed that for barite at 200°C, the concentration of barium measured in the solution that was flowing over the mineral pack with a residence time of 18 minutes was similar to that in the eluted portion that had been in contact with the mineral for 18 hours. In contrast, the results shown for anhydrite dissolution at 150°C indicate that the calcium concentration from the shut-in period is considerably higher than that from the continuous-flow period. In this case, it would also be necessary to perform measurements with a longer shut-in time to determine whether 18 hours is sufficient for the system to reach equilibrium.
A rigorous approach to determine whether true equilibrium has been reached involves performing both dissolution and precipitation experiments for the same mineral and test conditions to verify that these converge to the same value. Of course, the actual equilibrium value will not be known before starting the measurements, particularly in cases in which it is known that existing models are inaccurate, so it may not be possible to determine initially whether a particular test using the precipitation approach will actually result in precipitation or dissolution. Screening a range of injection fluid concentrations is, then, one method that can be used to gain confidence in the equilibrium solubility measured. The authors’ approach is to conduct the dissolution method initially at reasonably long residence times to estimate the solubility, use that as a starting point, and then perform precipitation tests with initial concentrations of the mineral ions increased stepwise.
A complicating factor when processing data generated using the precipitation method is that the output cation-toanion ratio is well known but not necessarily 1:1. To compare the results from different conditions, calculation of the ion product of the output concentrations is necessary.
If equilibrium is reached under the test conditions, the same value for the output ion product should be reached regardless of the input brine concentrations and variation in pump flow rates, provided these are not so far from 1:1 that the overall composition of the brine is changed markedly, and along with it, the activity coefficients. Therefore, plotting the input ion concentration against the output ion concentration after the 18-hour shut-in and during the flowing periods (with 18-minute residence time) should provide an indication of the behavior of the system and whether equilibrium is being achieved.
Another factor highlighted by the authors’ experimental results is that in precipitation experiments, the form of the mineral depositing must be taken into account. It is important to demonstrate that the system has not simply reached a metastability by saturating a less thermodynamically stable polymorph that has more facile precipitation kinetics, rather than representing the true thermodynamic equilibrium solubility of the most stable form.
The complete paper discusses considerations required for the experimental design of HP/HT mineral solubility measurements and, in particular, steps that can be taken to ensure that equilibrium is reached. The authors find that particular care must be taken to achieve this for each mineral and condition to be tested because results differ and different approaches will be required. Some minerals, such as barite, reach equilibrium rapidly by a dissolution approach under HP/HT conditions, whereas others, such as anhydrite, require prolonged reaction times, in which case a precipitation-based approach can be more suitable, allowing solubility data under HP/HT conditions to be derived under reasonable laboratory time periods. The authors describe an approach combining both methods to allow verification that a true equilibrium value is reached and not a transient, steady-state value or one characteristic of an intermediate metastable state.
The presence of calcium cations was found to increase the solubility of barium sulfate under the conditions tested (between 100–200°C and 10,000–19,000 psi), with the solubility increasing with increasing calcium ion concentration, at least up to 1 mol/kg. The experimental values were compared with those generated from several oilfield-scale-prediction modeling packages and found to differ significantly, with some models predicting barium sulfate solubility up to 2.5 times the experimentally determined value for some conditions. This finding demonstrates the limitations of scale-prediction-modeling software, in which reliable predictions are predicated on having reliable calibration data, something that remains scarce for HP/HT conditions.
This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper SPE 193564, “Scale Prediction and Mineral Solubility Under HP/HT Conditions,” by David Nichols, Neil Goodwin, and Gordon Graham, SPE, Scaled Solutions, et al., prepared for the 2019 SPE International Conference on Oilfield Chemistry, Galveston, Texas, 8–9 April. The paper has not been peer reviewed.