The effect of MEG (mono ethyleneglycol) on the precipitation kinetics of calcium carbonate related to natural gas production from subsea wells

Abstract

Injection of inhibitors for gas hydrate formation during transportation of wet gas/condensate in pipelines strongly affects the tolerance for produced formation water. The use of mono ethylene glycol (MEG) as hydrate inhibitor can be combined with pH increase (pH stabilisation) as corrosion control for carbon steel pipelines. The pH increase promotes precipitation of iron carbonate which forms a protective layer for corrosion at the steel surface. However, the elevated alkalinity downstream of the MEG injection point, combined with calcium ions from produced formation water, might result in precipitation of calcium carbonate. The thermodynamics of the calcium carbonate- system in water/MEG mixtures is already established in literature, but to be ble to predict and understand the precipitation, and optimize separation in MEG regeneration processes, the rate towards equilibrium plays an important role. For engineering challenges as scaling and particle removal in process systems polymorphism, crystal size and habit are important factors, and are all effects of inetics. Proper data of the supersaturation are crucial for investigations of kinetics, specially to account for the effect of a co-solvent. The solubility of calcite is decreased by MEG and activity based supersaturation values were in this work calculated by MultiMEGScale7.0©. In this work the precipitation kinetics of calcium carbonate at conditions of gas processing with MEG was investigated experimentally. The kinetics will together with the hermodynamics predict precipitation and ionic consumption down the process line.

For quantitative determination of polymorphic compositions a method based on X-ray diffraction was established. Spontaneous precipitation experiments at neutral pH were performed at a wide range of supersaturations, MEG contents and temperatures in order to investigate the abundance of the three polymorphs of CaCO3 at various conditions relevant for gas processing. The polymorphic composition was found to be dependent on the conditions and switching from mixtures of all three polymorphs with a majority of aragonite at elevated temperatures and low MEG content to a higher content of vaterite at high MEG concentrations, even at comparable supersaturations. At lower temperature a mixture of calcite and vaterite was found. The induction times in the experiments were measured by conductivity and found significantly increased at higher concentrations of MEG, even if the supersaturation was increased simultaneously. The experiments were left running for 24 hours to investigate the degree of transformation into the thermodynamically stable calcite. The transformation rates were found reduced in high concentrations of MEG. Both the prolonged induction times and the decreased transformation rates were explained by reduced growth rates caused by the co-solvent.

Induction time measurements were also used for calculations of nucleation rates. The induction period is the sum of the time for nucleation growth till detectable size, and is in the literature often assumed to be inversely proportional to the nucleation rate. The nucleation rates were determined in order to develop a model for nucleation in a MEG regeneration process. The reduced growth rates caused by the MEG influenced the detection limit of the induction times to a large degree, and thus the induction times were not only representing the time for nucleation. Both a combined method of seeded and unseeded experiments; a method of counting the particles by Coulter Counter assuming all particles were formed through steady state nucleation during the induction time; and a method of including pre determined growth rate data into the induction time measurements of unseeded experiments were used to establish nucleation rate data. The latter method was assumed to be the best method for slow growing systems. The nucleation rates were found decreasing with increasing concentrations of MEG. A model for nucleation rates in the temperature range from 40-70 °C and MEG concentration from 50-70 wt% was established.

Growth rates were measured by off-line titration of calcium in seeded batch experiments, and the growth rates of all polymorphs were found reduced by MEG. Calcite was the polymorph of which growth was delayed the most and the growth rate was reduced from kr= 4.1·10-10 m/s to 0.8·10-10 m/s at 40 °C by addition of 50 wt% MEG. Aragonite was found to be the least affected polymorph at 40 °C, its kr decreased from 9.0·10-10 m/s to 5.5·10-10 m/s by the same change in solvent. The growth rate constant of vaterite was found decreasing from 1.9·10-9 m/s to 0.2·10-9 m/s when increasing from 0 to 70 wt% MEG at 40 °C, and from 2.6·10-9 m/s to 1.2·10-9 m/s when increasing from 0 to 50 wt% at 70 °C. By keeping the pH close to neutral and the calcium concentration constant a large ratio between the calcium and the carbonate contents is promoted. This ratio was resulting in diffusion controlled growth when the carbonate content was too low due to high MEG concentration, contrary to the second order surface controlled growth expected from literature. Stoichiometric experiments confirmed that the observed change in growth mechanism was an effect of the ionic ratio. Comparable seeded growth experiments were performed at high pH (~10) to examine whether the regeneration rates of CO32- from CO2 was affected by the MEG. The carbonate source was not found to have any impact on the growth rates.

A model for the growth rates as a function of temperature and MEG was established for the range of conditions investigated (40-70 °C and 0-70 wt% MEG). Growth experiments were also performed by the constant composition method to investigate growth mechanisms of vaterite, but the method was not found successful for the conditions tested (37 °C, 2.5