Summary
With more and more salt cavern gas storage under construction, the corrosion
problems of leaching string, production string, and production casing require
more attention by experts on gas storage. This paper introduces several
corrosion phenomena that are encountered during the operation of leaching salt
caverns and during natural-gas injection and withdrawal. The main corrosion
factors are discussed such as brine, air, microbes, components of natural gas,
gas injection velocity, operating status, and others. The corrosion mechanisms
are also analyzed. To prevent the corrosion effects, a series of measures are
successfully applied during different construction phases of the gas storage
facility in the Yangtze River Delta, such as oxygen scavenger and biocide,
coating protection, annulus protection liquid, and cathodic protection.
Introduction
The two principal types of underground storage sites used in China today are
depleted reservoirs in oil and/or gas fields and salt formations. Each type has
its own physical characteristics (porosity, permeability, retention capability)
and economics (site preparation costs, deliverability rates, cycling
capability), which govern its suitability to particular applications.
Salt-formation storage facilities provide very high withdrawal and injection
rates compared with their working gas capacity. Base gas requirements are
relatively low. To date, a large amount of natural-gas storages in the bedded
salt deposits are being constructed by PetroChina as a component of the
infrastructure necessary to accommodate the gas flow from the west-to-east
pipeline. It is critical to ensure safe gas consumption in the Yangtze River
Delta, which is the main gas-consumption area.
Solution mining is the process by which a void or cavity is created in an
underground salt formation for storage purposes (RP 1114 1994). This cavity is
created by the dissolution of salt when fresh water is injected into the
formation under controlled conditions. The resulting brine is displaced out of
the developing caverns for further processing or disposal. Solution mining of
the cavity is usually achieved by controlled circulation of water through two
concentric leaching strings down the wellbore and up again to the surface. The
leaching strings include an intermedium string and a central string, and it can
be seen in Fig. 1. Solution mining of the caverns represents about 25 to 35% of
the investment in gas storage (Favret 2004). Taking to several years to
complete, it is a long process that requires large water resources and that
produces just as much brine with a salt concentration of 200 to 310 kg/m3,
which is used by the chlorine and sodium chemical industries, or reinjected
into the subsoil, or even pumped into the sea.
During the process of solution mining, the leaching strings and wellhead are
often corroded and eroded by the brine, which contains plentiful salt,
dissolved oxygen, and particulates. Therefore, the circulating pump pressure is
increased, and the service life of leaching strings is reduced. Furthermore,
the sealing property of the wellhead is also affected. These can all be seen in
Fig. 2. According to construction unit surveys, the corrosion of leaching
string and wellhead contributes to a solution mining cost increase of at least
15%, and at worst it can even cause a salt cavern to be discarded as
useless.
Once the cavern is completed, the well is recompleted for dewatering and
first gas filling. Gas is then injected through the annulus between the
debrining string and the production string, and brine is produced through the
debrining tubing. This can be seen in Fig. 3. Natural-gas injection and
debrining are a very important part of the construction process. This process
generally takes 6 to 8 months, and then, after the debrining string is pulled
out, the gas-storage facility will be used for daily peaking, seasonal gas
deliveries, and emergency gas supply.
© 2008. Society of Petroleum Engineers
View full textPDF
(
440 KB
)
History
- Original manuscript received:
14 June 2006
- Meeting paper published:
30 May 2006
- Revised manuscript received:
27 March 2007
- Manuscript approved:
18 April 2007
- Version of record:
20 February 2008
View Cart
