Cooling Towers: Corrosion Mechanisms to Critical Failures — Part 3

The Most Common Corrosion Mechanisms

Figure 1 illustrates a classic example, albeit in an artificial environment, of general corrosion in which anodes and cathodes constantly shift around the metal surface. General corrosion is the baseline to which water-system engineers design equipment for a lifespan of about 30 years. Unfortunately, many corrosion mechanisms are localized, where in some cases, through-wall penetration of tubes, pipes, tank walls and other structures has occurred within months and sometimes even weeks.

Proper selection of chemical treatment programs is necessary to minimize localized corrosion, but so is materials selection. For example, a common myth is that stainless steel is a cure-all for corrosion difficulties, when in fact some stainless steels will rapidly corrode in certain environments.

The ferrous hydroxide (Fe(OH)₂) reaction product will further react with oxygen to produce solid compounds, basically rust (hydrated ferric oxides, Fe₂O₃∙xH₂O) that settle on the metal surfaces.  It is not uncommon for these corrosion products to accumulate on top of active corrosion sites to form tubercules. Other dissolved elements/compounds such as chloride and sulfate may then accumulate underneath the tubercules to increase the corrosion potential. 

Given the ready supply of oxygen via air drawn through a cooling tower and the dissolved oxygen (DO) that enters with the makeup water, a primary goal of treatment programs is protecting carbon steel from DO attack (we will examine treatment details in part 6 of this series).

Pitting

The oxygen attack outlined above is localized and can be thought of as a form of pitting. Other impurities can cause pitting.  

A classic example is chloride pitting of the 304 and 316 austenitic stainless steels. These materials are very popular for heat exchanger tubes and plate-and-frame HX components. Stainless steels owe their corrosion resistance to the oxide layer that forms on the surface per the addition of 12% or more chromium to the alloy. Stainless steels need a constant supply of oxygen to maintain this protective layer. However, chloride ions can penetrate the oxide film to establish localized corrosion cells. Consider the following personal case history. 

Case Study: Stainless Steel Selection Pitfalls in Power Plant Design

For several years, one of my primary tasks was to review the water treatment portion of engineer, procure, construct, or EPC, specifications for combined cycle power plants. These specs came from outside design firms. Almost without exception, the design engineers would select 304L or 316L stainless steel as the tube material for the steam surface condensers.

But in numerous instances it was clearly obvious that the design team had not considered the potential for chloride pitting. 

Corrosion knowledge has advanced greatly over the years, and it is now known that even seemingly low chloride concentrations (approximately 200 mg/l for 304L and 500 mg/l for 316L) can induce pitting corrosion, especially underneath deposits.² 

Because most modern cooling systems have a cooling tower at the heart, all impurities, including chloride, cycle up in concentration due to evaporation from the tower. While the makeup water chlorides may be below corrosive concentrations, the recirculating water may be well above.

Chloride pitting resistance is frequently the primary factor in heat exchanger materials selection for water-cooled exchangers. A well-known guide is the pitting resistance equivalent number (PREN) chart as shown in Figure 4.

Higher-alloy materials beyond the 300 austenitic stainless series are recommended for cooling waters with appreciable chlorides. For highly brackish water and seawater, the ferritic and super-ferritic alloys are often recommended. Consider another important and insidious pitting example. As we will further examine in the microbiologically influenced corrosion (MIC) discussion below, sulfides can be extremely corrosive to many metals. Figure 5 shows sulfide pitting in an extracted and longitudinally cut section of a 90-10 copper-nickel condenser tube.

The tubes had been installed as a replacement for aging Admiralty brass material. Within 18 months, the new tubes failed from numerous through-wall penetrations. Investigation revealed that the tube manufacturer had used a lubricant containing sulfides during the fabrication process. They did not remove lubricant residuals following tube fabrication. Once the condenser was placed in service, the sulfides rapidly attacked the metal. A second retubing became necessary.

Microbiologically Influenced Corrosion (MIC)

Even garden-variety deposits can be problematic because they establish oxygen differential cells, where oxygen deficiency underneath a deposit causes that location to become anodic to the remaining metal. Beyond that issue, though, problems may arise from microbial metabolic processes. Sessile microbial organisms and the slime layer they produce typically harbor many organisms, including anaerobic bacteria whose metabolic processes generate corrosive compounds. A leading culprit is sulfate-reducing bacteria, such as desulfovibrio desulfuricans, which extracts oxygen from sulfate to produce hydrogen sulfide.

I directly witnessed the aftermath of a situation in which a steam condenser with over 15,000, 316L SS tubes was left open to the atmosphere, with cooling water remaining in the tubes, during a month-long scheduled maintenance outage. Upon unit startup, the plant chemists discovered severe impurity ingress to the condensate. Visual inspection revealed thousands of pinhole leaks in the tubes, forcing a unit shutdown and complete retubing of the condenser.

Other factors influence corrosion potential, including fluid velocity, surface imperfections, disruptions of metal structure and chemistry at weld locations, the list goes on. Additional details are available in the references below and from the Cooling Technology Institute. Part 4 of this series will examine issues related to mineral deposition and scale formation in cooling systems.