Buffers/pH Control

Although corrosion inhibitors are frequently separated from the buffering/pH control agents, in most cases they are intertwined. Inhibitors generally function only in certain pH ranges. For example, nitrite is typically used under alkaline conditions. Others, such as some of the dibasic acid mixtures, show best performance at neutral pH. Regardless of the inhibitor type, acid pH conditions are never acceptable.

Given the variety of metals and elastomers in closed loops, the pH range is a compromise. The primary metallurgies used in North America are mild steel and copper alloys. These metals require a pH range of 8.5 to 10.5. Aluminum, which is uncommon in North America but is seen in Europe, prefers a neutral pH range for optimum protection. At one time, caustic (sodium hydroxide) was commonly used to raise pH. Since caustic is not a buffer, controlling pH is difficult, and excursions are not unusual. Except for cost, there is little to recommend caustic, and the risk related to its use is high. Also rarely used, carbonate buffers are sometimes encountered where other more suitable buffering systems cannot be used. Carbonate and hydroxide may have application in nuclear plants where borate cannot be accepted. Otherwise, they have little to recommend them.

The most widely used pH control agents are all buffers:

• Ortho phosphate,
• Borate salts, and
• Organic amines.

Ortho phosphate, typically dipotassium phosphate (or adjusted mixtures of di- and tripotassium phosphate) can be used to control the pH between 9 to 10.5. At the high levels used for corrosion control, ortho phosphate provides excellent buffering capacity. This buffering function is particularly useful in glycol containing loops, where the ortho phosphates can neutralize glycol breakdown products for extended periods of time.

Another inorganic buffering system is borate (or metaborate). Borates provide good buffering capacity and allow a slightly higher pH than phosphates. This means that less can be used to obtain the desired pH. Unlike phosphate, borate functions in the sole role of a buffer. It has effectively no impact on corrosion rates beyond that seen from pH.

Organic amines provide a number of benefits compared to inorganic buffers. Depending on the pH range desired, one can select from the commonly available amines to tailor the pH of the fluid to the exact value required. In addition, morpholine has been shown to be a passivity agent in its own right. Another benefit of the amines is that the amount needed in the fluid is low (200 to 300 mg/L). The low concentration coupled with their low intrinsic conductivity means amines contribute less to conductivity than inorganic buffers. The lower conductivity means amines are useful in conductivity sensitive systems like cooling water loops for welders.

Film Former Cause Corrode

Among the four filmers used, ortho-phosphate is the most common. Glycol manufacturers widely use ortho-phosphate in the dual role of corrosion inhibitor and pH buffer in their formulations. At normal use concentrations (1000 to 5000 mg/L as PO4), phosphate protects against corrosion on ferrous and non-ferrous alloys. The primary mode of action is via precipitation at the anode to form insoluble metal phosphates. This low solubility of phosphate salts is why manufacturers recommend using good quality (i.e., soft or distilled/deionized) water12 for diluting phosphate containing products like glycols.

The ability of phosphate to form a protective film by directly precipitating is both its strength and weakness. While it will film the metal surfaces, it will just as readily precipitate with metal ions or hardness salts in the bulk water system. This competition between useful and non-productive reactions is the major liability associated with phosphate. Since ortho phosphate is an anodic inhibitor, if the concentration falls below the critical level (200 to 300 mg/L), rapid corrosion attack will occur, this mean that on the surface have oxidized. Phosphonates are related to inorganic phosphates, which include ortho and polyphosphate. HPA (hydroxy-phosphonoacetic acid) is the best example of a phosphonate. These chemicals are effective cathodic inhibitors. HPA can be employed where a product with a low environmental impact is preferred. Acceptable corrosion rates can be obtained at levels of 50 to 200 mg/L (as HPA). The phosphorous contribution of 25 to 150 mg/L (as PO4) is well below what would be necessary with inorganic phosphates.

A third class of film formers is the various dibasic acids. They can be used for pH buffering and corrosion inhibition, much like phosphate. Dibasic acids work because of their limited solubility with transition metals (iron and copper) and alkaline earth cautions (hardness). As the corrosion process takes place at the anode, iron ions go into solution. The dibasic anion reacts with the iron ions and precipitates at the corrosion site, stopping corrosion.

Unfortunately, dibasic acids are biodegradable. In chilled water loops this can pose a serious limitation. In particular, where there is already a supply of biologically available nitrogen (i.e., nitrite), rapid biological growth can quickly consume these nutrients in a few days.

Also, as they rely on iron ions to form the inhibitor film, losses can occur if a considerable amount of corrosion product is present in a system before adding the dibasic acid. In systems (new or old) with significant amounts of corrosion products present, loss of this type of inhibitor can be dramatic.

Substituted triazines are the final type of film former. They react selectively with iron to form an inhibitory film on the metal. Although they are widely used in oil-field applications as down hole corrosion inhibitors, they have not seen extensive use in closed loops. The hexanoic acid triazine compound is an example of this chemistry.

Oxidizing Agents for Waste

In contrast to reducing agents, oxidizers either react directly with the metal surface (chromate and nitrite) or work in conjunction with oxygen (molybdate) to achieve a passive film on the metal. The standard inhibitor in this category, nitrite, has been used to inhibit corrosion of mild steel for many years1, 2 in neutral or alkaline aqueous solutions. Nitrite is the only remaining anodic oxidizing inhibitor that can still be used. Unlike molybdate, another anodic inhibitor, nitrite does not need oxygen.3 For this reason, it is very effective in closed systems. It has been proposed that nitrite protects ferrous metal by an oxidation-reduction process where ferrous hydroxide forms a passive magnetite layer.

The overall reaction is:

9 Fe (OH)2 + NO2- ---------> 3 Fe3O4 + NH4+ + 2OH- + 6 H2O

Whether nitrite is used alone or in conjunction with pH buffering agents, relatively high concentrations are needed to establish a protective film, usually on the order of 700 to 1200 mg/L to completely inhibit pitting corrosion. Once the protective film has been established, the nitrite concentration can be lowered slightly to 700 to 1000 mg/L. Some sources have stated that the required nitrite level is influenced by the amount of chloride and sulfate present in the water5, 6 because they can affect the stability of the magnetite layer. As with all anodic inhibitors, severe pitting can occur at low concentrations (<500 mg/L as NaNO2). In other words, too little nitrite is actually worse than none at all because low levels of nitrite will speed up the corrosion process. Loss of nitrite can occur via electrochemical and biological processes. In the former case, if corrosion continues, nitrite can be reduced at the cathode to form ammonia7 according to the equation:

NO2– + 5H+ + 6e- -------> NH3 + 2 OH-

In chilled water loops (or hot water systems that are not in operation) exposure to bacteria has the potential to oxidize nitrite to nitrate or reduce it to ammonia or nitrogen. Controlling biological activity is difficult because oxidizing biocides (like chlorine) will oxidize the nitrite to nitrate, and the efficacy of non-oxidizing biocides tends to be less certain. Difficulty in preventing biological degradation of nitrite has always been a serious limitation.

The use of molybdate for corrosion protection in cooling water, either open recalculating or closed loop. While molybdate is not as strong an oxidizing agent as chromate, it can function in this role in the presence of oxygen. In the presence of oxygen, molybdate will convert hematite (Fe2O3 or red rust) to magnetite (Fe3O4 or magnetic black rust). This process is quite visible as boilers (either hot water or steam) change from a reddish color to black when treated with molybdate.

This mechanism predominates at higher concentrations (>50 mg/L as Mo). By contrast, molybdate’s efficacy as an anodic (or pitting) inhibitor is related to its ability to accumulate within the acidic part of a pit and block the corrosion process. Use of molybdate alone at <20>50 mg/L as Mo), molybdate (in the presence of oxygen) is capable of passivating boiler metal. The only concern regarding molybdate use is related to its accumulation in sludge from waste treatment plants if the sludge is spread on agricultural land. However, based on its low human and aquatic toxicity, molybdate is not severely restricted in most areas of North America. The last oxidizer is hexavalent chromium, or chromate, which, while a very effective corrosion inhibitor is seldom used due to concerns of health/environmental effects. Being a strong oxidizing agent, chromate is capable of converting hematite to magnetite. The reduced chromium becomes incorporated into the resultant oxide layer.

Water Treatment In Closed Systems

Closed systems are commonly classed by their function — either heating (hot water) or cooling (chilled water). They typically rely on water or water-based solutions as their heat transfer fluid.

While various ways exist to classify water treatment product technologies, one of the simplest is whether the primary corrosion inhibitors are reducing agents, oxidizing agents, or film formers. In addition to the primary corrosion inhibitors, factors such as fluid pH and copper ions also affect corrosion. Supplemental agents are used to buffer/control pH and minimize yellow metal corrosion.

The primary corrosion inhibitors, which focus on ferrous alloys, and the supplemental inhibitors comprise the total treatment package.

Reducing Agents

Reducing agents are not commonly used due to inherent limitations with available chemicals. Reducing agents work by removing oxygen from solution so it is not available to corrode metals. Although rarely seen nowadays, tannins have been used both to remove oxygen and form an iron-tannin film on steel surfaces. Tannins are low cost, easy to formulate, testable, and readily available.

The drawbacks for tannins include their tendency to form organic deposits on heat exchange surfaces. These deposits may eventually require chemical cleaning to remove. More significantly, while they do scavenge oxygen, the rate of reaction is not rapid. It is common to have oxygen corrosion in spite of maintaining reasonable tannin residual. Sulfite is another type of reducing agent treatment that still shows up from time to time. The reaction of sulfite with oxygen has been well studied. It is generally recognized that the reaction proceeds via a free radical mechanism. The overall equation is:

½ O₂ + Na₂SO₃ --------> Na₂SO₄

With this technology, sufficient sulfite must be present at all times. Otherwise free oxygen will exist and the boiler metal will corrode. The normal approach involves maintaining an excess of sulfite (30 – 50 mg/L Na2SO3) in the water.

The sulfite residual acts as a “sponge” to react with oxygen that enters the system. Without regular testing, the risk of corrosion is quite high, since even temporary losses of the sulfite residual can lead to corrosion.

A secondary drawback is that as sulfite is fed to the system (to cope with the ongoing ingress of oxygen), the concentration of sulfate builds. Increasing sulfate increases the conductivity of the water and its corrosively as well as the potential for SRB (sulfate reducing bacteria) growth. Organic reducing agents such as hydrazine (N2H4) and DEHA (diethyl hydroxylamine) have been used.

However, decomposition (catalyzed by copper) and health concerns for hydrazine have largely eliminated their use. In the case of hydrazine, its breakdown to ammonia has resulted in failures related to intensive copper corrosion by the following mechanism.

3 N₂H₄ + 6 OH^¯ ---------> 2 NH₃ + 2 N₂ + 6 H₂O

NH₃ + Cu+² ---------> Cu (NH₃)²⁺

Both hot and cold loops have all the conditions for these reactions to take place, including the oxygen needed to oxidize the copper metal and ammonia that dissolves copper oxide. The dissolved copper is free to plate on to steel surfaces in the system where it can cause galvanic corrosion. When the copper plates out, it releases the ammonia, which is free again to repeat the cycle. As a result, ammonia will rapidly corrode copper (and its alloys).