Jul-2009

Multiple corrosion mechanisms in crude distillation overhead system

Extensive troubleshooting efforts determine distinct corrosion mechanisms simultaneously attacking multiple areas of an atmospheric tower overhead system

George Duggan and Randy Rechtien, Baker Hughes
Lionel Roberts, Irving Oil

Article Summary

With declining crude quality and the high profit potential of opportunity crudes, refiners continue to face a difficult balancing act when controlling corrosion: determining the optimum combination of crude blends, unit operations, corrosion-control programs and unit maintenance in order to achieve the greatest return on investment for the refinery. Understanding the source of corrosion is critical. It is only after the root cause of the corrosion is properly identified that viable mitigation solutions can be selected based on the refiner’s unique circumstances and short- and long-term goals.

The Baker Petrolite TopGuard overhead corrosion-control program from Baker Hughes Inc is a comprehensive, engineering-based program designed to meet refiner’s profit objectives in the most cost-effective way possible. Working closely with each refiner, Baker Hughes provides the knowledge required to effectively manage the corrosive impact of specific crude blends and operating conditions. The following case study provides a detailed summary of the troubleshooting efforts and the methods implemented to successfully reduce the threat of corrosion-related failures at a refinery in Canada.

History of corrosion incidents
More than four years ago, problematic episodes of corrosion occurred in the overhead condensing system of the #3 crude unit atmospheric tower at Irving Oil in Saint John, New Brunswick, Canada. Corrosion occurred in three separate locations in the overhead, with each location experiencing a different mechanism of attack. Although uncommon, there are industry examples of the simultaneous occurrence of different corrosion mechanisms in a single overhead system.¹

For this particular system, the primary sources of corrosion were strongly related to unit operating conditions, contaminant levels in the crude and, ultimately, contaminant levels in the tower overhead itself. The refinery processed blends of either sweet or sour crudes in blocked operation. These alternating crude slates, combined with seasonal variations in tower operations, produced a wide range of corrosive environments in the overhead. In particular, spikes in overhead hydrochloric acid (HCl) concentration increased the formation potential of ammonium chloride (NH4Cl) salt and made pH control of overhead drum water more difficult. During sour crude processing, increased levels of hydrogen sulphide (H2S) in the overhead produced preferential attack on copper-based equipment. Higher system temperatures and higher flow rates during some operating modes created localised zones in which velocity-accelerated corrosion was prevalent.

The variations in system conditions required more diligence on the part of operators, inspectors and corrosion-control engineers to address the problems. Multiple analytical and monitoring techniques were required to identify the cause of the corrosion mechanism and to develop appropriate mitigation options. To this end, efforts were conducted to correlate operational changes with periods of increased corrosion activity. The Baker Petrolite Ionic Model was employed to calculate amine-hydrochloride salt formation temperatures and to define “safe” operating envelopes. Detailed compositional analyses of scale deposits and metallurgical analyses of weight loss coupons were also performed. Traditional methods for measuring metal loss rates provided insight into the magnitude of corrosion activity as well.

System overview
The #3 crude unit atmospheric distillation tower overhead (Figure 1) comprises a set of two parallel shell-and-tube exchangers (E-22001 A/B) that are vertically oriented. In these exchangers, process vapours exchange heat with cold crude oil. The exchanger outlet streams are combined and then fed to a set of eight air coolers (E-22027 A-H). The vapour/liquid mixture from the air coolers is separated in the D-22001 drum. Naphtha from the drum is divided between both the tower reflux and the product. A portion of sour water from the drum is continuously recycled to the overhead vapour line for use as wash water.

The E-22001 A/B exchanger tube bundles are constructed of 70/30 copper/nickel alloy (UNS C71500). The remaining overhead equipment is constructed of carbon steel. The existing corrosion-control program includes an oil-soluble inhibitor injected into the overhead vapour line via a reflux carrier. Neutralisation is provided by an aqueous ammonia solution (approximately 20% concentration), which is injected into the overhead line via the water wash. Currently, the ammonia injection rate is adjusted to maintain a nominal target pH range of 6–6.5 in the overhead drum. During the periods of corrosion activity discussed herein, the drum was typically operated at a pH of 7.0 or higher.

Corrosion rate monitoring was measured via a combination of electrical resistance (ER) probes and weight loss coupons at several locations in the overhead. Specifically, a total of seven monitoring devices were installed as follows: the main overhead vapour line (x 1); the E-22001 A/B inlets (x 4); and the E-22001 A/B outlets (x 2). In addition, frequent UT measurements were taken by the refinery’s inspection department. In general, the rates measured at these locations were within acceptable ranges. However, there were occasional periods of unacceptably high corrosion rates (>0.25 mm/year [>10 mpy]) at these locations. The periods of high corrosion rates usually correlated to variations in operating modes and/or switches in crude blends.

The corrosion activity in the atmospheric tower overhead was most severe from late 2004 through late 2006. During this period, the refinery processed several different crudes of varying sulphur content. Typically, the unit was operated in blocked operation: several days of sweet crude and then several days of sour crude, and so on. Crude blends with sulphur contents below 1.0% were considered sweet. In addition to variations in crude sulphur, the tower was operated on seasonal cycles (summer vs winter operation). These seasonal modes covered a relatively wide range of tower operations, particularly in terms of tower top temperature and overhead flow rate. Table 1 summarises the typical conditions under each of the four primary operating modes. There were three distinct areas of the overhead that experienced periods of excessive corrosion activity:
•  Pressure relief valves at the tower top
•  E-22001 A/B tube bundles
•  E-22001 A/B outlet elbows.