400338 Flowloop Measurements of Hydrate Blockage Risk in 
Under-Inhibited Gas-Dominant Systems

Monday, April 27, 2015
Exhibit Hall 5 (Austin Convention Center)
Mauricio Di Lorenzo1,2, Zachary M. Aman2, Karen Kozielski1, Bruce W.E. Norris2, Michael L. Johns2 and Eric F. May2, (1)Petroleum Production, CSIRO, Clayton, Australia, (2)School of Mechanical and Chemical Engineering, University of Western Australia, Crawley, Australia

In conventional flowlines, gas hydrate formation is mitigated through the injection of thermodynamic hydrate inhibitors (THIs) that shift the hydrate stability criteria toward high pressure and lower temperature. Complete hydrate avoidance with THIs substantially increases infrastructure and operating costs, which may affect the economic viability of deepwater assets. Reducing the rate of THI injection below the complete inhibition threshold may translate to significant cost savings, but adoption of this practice requires a rigorous experimental confirmation that such under-inhibition will not increase the risk of hydrate blockage formation. In this study, we have deployed a high-pressure sapphire autoclave and high-pressure single-pass flowloop to measure hydrate formation and resistance-to-flow (motor torque or pressure drop) in a matrix of under-inhibited conditions. 20 autoclave studies were performed with deionised water containing 0-20 wt% monoethylene glycol (MEG). The rate of hydrate growth and severity of blockage formation decreased monotonically with the initial MEG fraction; visual observation of the sapphire cell suggested that MEG reduced hydrate film growth and particle deposition on the wall. The hydrate growth rate was accurately predicted within 2 vol% by a kinetic growth model. 20 gas-dominant flowloop experiments were performed with 10 vol% water flowing under annular conditions. Increasing the amount of MEG in water monotonically decreased (i) the hydrate growth rate, (ii) the rate of pressure drop increase, and (iii) the frequency of sloughing events. At subcoolings below 5 K, the pressure drop increase due to wall deposition and the global hydrate growth rate were compared. The results suggest that approximately half the hydrate was deposited on the pipeline wall; further studies are underway to expand this understanding to systems at higher subcooling. Together, these experimental studies demonstrate that a reduction in MEG fraction of up to 10 wt% from the complete inhibition requirement did not result in an imminent increase to hydrate blockage risk. The hydrate growth rate may be readily estimated through a kinetic model, where MEG fractions above 10 wt% were consistently observed to improve hydrate transportability.

Extended Abstract: File Not Uploaded