Abstract
An inherent problem with natural gas, condensate and crude oil production or transportation is the formation of gas hydrates. The usual practice for avoiding the plugging of production facilities by hydrates is to add thermodynamic inhibitors such as methanol or glycol. For a few years, a lot of works have been done for the development of a new class of low dosage additives: Kinetic Inhibitors. These compounds are designed to delay nucleation and/or prevent crystal growth. The performance level of Kinetic Inhibitors can be investigated by different methods such as hydrate formation rate, subcooling, total volume of hydrates formed, plugging of a flow loop or stirred cell… but the main technique remains the determination of the induction time. Unfortunately, in rather clean conditions (without adding any impurities), the induction time is a stochastic quantity. Therefore, to obtain a significant result, experiments must be repeated until the calculated average induction time becomes constant. This procedure is consequently time consuming and rather difficult toperformed.
We have developed a semi-batch stirred reactor in which temperature and pressure are maintained at constant values. This reactor is equipped with a turbidimetric sensor which is used to characterize in-situ the methane/water system (particle size distribution of the hydrate suspension or state of dissociation of the undersaturated solution).
Using this device, a new procedure has been found which allows us to have a constant and reproducible induction time without adding any impurities. This new and very simple method enables to determine quantitatively and qualitatively the efficiency of kinetic inhibitors both from the induction time (nucleation stage) and from the quantity of hydrates formed (growth stage) in a very short time. In this paper we will describe the apparatus and the procedure applied to obtain reproducible results. This operating procedure has been successfully applied to the characterization of kinetic inhibitors.
