SPE Production & Operations
Volume 25,
Number 1,
February 2010,
pp. 19-24
Summary
This case history tracks the continual improvement cycle for the
fluid-separation process of a heavy-oil/oil-sands production facility in
northern Alberta over a period of 3 years. The major challenge posed by the
operator of this 13 to 16°API crude oil was to move away from injection of two
separate demulsifier formulations to injection of a single product. This was
not an easy task because of the very different conditions that existed at the
two injection locations. The first location was at a series of injection points
upstream of the gathering stations before separation where temperatures could
reach subzero conditions, and the second was at the battery receiving facility
where heating increased temperatures to 100°C. Water cut and shear were also
very different, and the operator required a very strict 0.2% basic sediments
and water (BS&W) on the crude exiting any of the four treater tanks. To
complicate issues further, crude-oil viscosity ranged from 500 to 5,000 cp.
A unique bottle testing method was developed and used to simulate the field
conditions as accurately as possible. Details are given on the chemistry of the
individual components of the demulsifier determined to be so crucial to
adequate performance and how this was optimized in the field after being
identified from the bottle tests.
Results show how careful consideration was given to the concentration of the
demulsifier bases in the blends, and show the curious observation that dilution
of the final product made a big difference to the final performance in the
field. Elaboration is given on potential mechanisms explaining the dilution
effect, and this paper will conclude with observations on how careful design of
field testing followed by field implementation can indeed solve complex
separation issues and address individual-well, battery, and field
requirements.
© 2009. Society of Petroleum Engineers
View full textPDF
(
369 KB
)
History
- Original manuscript received:
31 July 2008
- Meeting paper published:
20 October 2008
- Revised manuscript received:
1 December 2008
- Manuscript approved:
20 December 2008
- Published online:
10 September 2009
- Version of record:
1 March 2010