Summary
Severe injectivity decline during seawater injection and produced-water
reinjection is a serious problem in offshore waterflood projects. The
permeability impairment occurs because of the capture of particles from
injected water by the rock, both internally in the pores and externally in a
filter cake. The reliable modeling-based prediction of injectivity decline is
important for injected-water-treatment design and management (injection of
seawater or produced water, water filtering, etc.).
The classical deep-bed filtration model includes a single overall
description of particle capture. During laboratory or field data interpretation
using this model, it is usually assumed that several simultaneously occurring
capture mechanisms are represented in the model by a single overall mechanism.
The filtration coefficient, obtained by fitting the model to the laboratory or
field data, represents the total kinetics of the particle capture. The purpose
of this study is to justify this approach of using an aggregated single
filtration coefficient.
A multiple-retention deep-bed filtration model needs to describe several
simultaneous capture mechanisms. The kinetics of the different capture
mechanisms can differ from one another by several orders of magnitude. This
greatly affects the particle propagation in natural reservoirs and the
resulting formation damage. In this study, a model for deep-bed filtration
taking into account multiple particle-retention mechanisms is discussed. It is
proven that the multicapture model can be reduced to a
single-capture-mechanism deep-bed filtration model. The method for
determination of the capture kinetics for all individual capture processes from
the breakthrough curve is discussed. As an example, the complete
characterization of filtration with monolayer and multilayer deposition of iron
oxide colloids is performed using particle-breakthrough curves from
coreflooding.
© 2009. Society of Petroleum Engineers
View full textPDF
(
473 KB
)
History
- Original manuscript received:
24 October 2005
- Meeting paper published:
15 February 2006
- Revised manuscript received:
1 March 2009
- Manuscript approved:
21 March 2009
- Published online:
16 July 2009
- Version of record:
28 September 2009