Corrosion Inhibitor

Control of copper corrosion is critical in any closed loop. While copper and its alloys are quite corrosion resistant, the impact of even low corrosion rates can be dramatic. When copper corrodes, soluble copper ions plate out onto mild steel components. When this happens, the more inert copper metal becomes a “permanent” cathode on the metal surface. At this point, the corrosion process, which had been spread over the entire steel surface, now becomes localized and continues at an accelerated rate. As this proceeds, instead of having a low general corrosion rate, high local corrosion rates will be seen. Azoles are used to prevent the initial corrosion of copper alloys, as well as to inhibit copper deposits on mild steel surfaces.

MBT (mercaptobenzothiazole), a low cost, effective inhibitor, has been used for many years with good results. More and more commonly, TT (tolyltriazole) has become the inhibitor of choice due to cost considerations and its superior resistance to the corrosive effect of chloride ions. In contrast to precipitating agents, the nitrogen atoms in the azoles bond to the copper metal via copper oxide molecules on the surface. The protective layer that is formed enhances the natural corrosion resistance of copper and copper alloys.

Monitoring
Monitoring of closed loops entails verifying that the treatment program is meeting the goals for corrosion, deposits, microbiological activity, and product/treatment level. The corrosion monitoring should also have some provision to show that localized or pitting type attack is not taking place.

Corrosion
Monitoring relies on using a model for what is happening to a system. It is meant to prevent surprises such as in Figure 3. Instead of having to cut out pipes to see if corrosion is taking place, a well-designed monitoring program will provide the same information about the efficacy of the water treatment program with considerably less effort. While a physical inspection of boilers and chillers (or other related components) is the most effective way to determine overall system performance, other approaches can provide the same information on the cleanliness of the equipment.

Corrosion coupons are the most common type of monitoring, since they provide information on overall corrosion rates as well as the type of corrosion that is taking place. A coupon will give the following data:

• General/overall corrosion rate,
• Pitting corrosion rate,
• Indicators of biological attack, and
• Evidence of galvanic attack.

In a well-maintained system, the corrosion rate should be virtually nil, and there should be no sign of localized attack. In chilled water loops, it is also important to assess whether microbiologically induced corrosion (MIC) is occurring. Coupons are one of the easiest ways to look at this aspect. Most closed loops contain a variety of alloys, and yellow metals are quite common. If they are not adequately protected, the dissolved copper ions (from the corrosion process) can deposit onto steel components and cause galvanic attack. If this problem is suspected, the copper content on the surface of a steel coupon will confirm if it is an issue or not. To be effective, corrosion coupons need to meet a number of criteria to mimic a closed system:

1. The coupons should be of similar metallurgy to the components in the system. Mild steel (typically 1010) and copper (although brass and other more specialized alloys may be appropriate) coupons should be used.
2. Exposure periods should be varied, with some as short as 30 days and some being allowed to remain in for up to a year.
3. The flow rate through the coupon rack should be close to what various sections of the loop experience. This could range from the normal flow velocities of 1.2 to 1.5 m/s (4 to 5 fps) to as low as 0.03 m/s (0.1 fps).
4. Since temperature has a large impact on corrosion rates, the coupons should see the same temperature (or as close as possible to it) as the hottest section of the system. A good location is on the supply header, shortly after the boiler. With chilled water, the return header is an ideal location.

If coupons are left in for short periods (less than 30 days), the corrosion rate will be artificially inflated. The real value of the corrosion rate is partially the level itself, but the major value is the trend over several months or years. Consistently low corrosion rates, with no localized areas of metal loss, is the goal.

If localized attack occurs, it is normally a good idea (assuming that obvious causes such as low chemical residuals are not the cause) to have the coupon checked for copper plating and MIC. Most water treatment companies provide this sort of service as part of their service program. In some cases, consultants also can arrange to have this type of testing done, though they typically use outside laboratories, which might be less experienced in what to check for. Installing coupons to copy conditions in zones where the water velocity is low is important. It is rare to find a closed loop circuit that does not have low flow sections or is periodically stagnant. Although the high inhibitor levels used in closed systems should reduce the effect of stagnant or low flow conditions, it is necessary to ensure that the program being used meets this critical performance criteria.

Figure 4 shows the effect that reducing water velocity has on corrosion rates for a conventional borate-nitrite program. Although the corrosion rates should only be taken as relative indicator of what might happen in a system, the important point is that going from a water velocity of 1.5 m/s (5 fps) to 0.3 m/s (1 fps) allowed corrosion rates to more than triple. Higher inhibitor levels did minimize this effect, yet the general trend is still apparent. Not taking into account the effect that water velocity has on the inhibitor program has caught more than one installation by surprise.

Low-cost electrochemical measuring devices (to determine instantaneous corrosion rates) are becoming more common and accessible. Electrochemical monitoring provides rapid results (probes can generate accurate rates after only a few days) and makes it possible to trend corrosion rates not only using monthly averages but also on a day by day basis. Being able to get reliable data on corrosion rates over such short periods allows one to determine if system operation (e.g., periodic shutdowns, etc.) is affecting the corrosion protection being provided. Although not as high tech as electrochemical monitoring methods, a simple and effective technique to follow corrosion trends is to measure iron and copper concentrations in the closed loop fluid. As system metal corrodes, it goes into the fluid. Although it is present in a precipitated form, the metal concentration in the water does serve as a good trend indicator of what is happening. While it does not tell what is causing corrosion, it does provide a quick indication of when things begin to go wrong or confirmation that the program is continuing to meet agreed upon standards.

Deposits
Deposits are rarely a concern in a closed system. The best way of monitoring deposits is to record makeup rates. If the system is operating within normal limits (makeup 10% of system volume per year) the risk of scaling is minimal. It should be noted that some treatment programs, specifically phosphate inhibitors used in some glycols, are quite sensitive to hardness in the water used. If good quality water is not used, the phosphates can react with the hardness present and form deposits.