BASIC WATER CHEMISTRY
This article describes water treatment. We will discuss several terms including pH, dissolved oxygen, conductivity, and chlorides. We will review the principles of ion exchange. And in the final section, we will review boiler feedwater and boiler water treatment methods and problems, and blowdown.
The importance of good water chemistry cannot be overemphasized. Your plant represents a significant investment on the part of your parent company. Maintaining it in good working order is not only a matter of valve, electrical, and pump maintenance, but also maintaining chemistry controls within the applicable specifications. Controlling the chemistry of the plant will extend the life of the plant, increase reliability, and reduce maintenance costs.
Water treatment is necessary to remove the impurities that are contained in water as found in nature. Control or elimination of these impurities is necessary to combat corrosion, scale formation, and fouling of heat transfer surfaces throughout the facility and support systems.
The following are three reasons for using very pure water at the plant site:
Corrosion products and other impurities may deposit on core surfaces and other heat transfer regions, which results in decreased heat transfer capabilities by fouling surfaces or blockage of critical flow channels. Areas of high concentrations of these impurities and corrosion products may also lead to extreme conditions of the various corrosion processes with resultant failure of components or systems.
There are several processes used to purify the water in the systems and water used as makeup. Deaeration is used to strip dissolved gases, filtration is effective in the removal of insoluble solid impurities, and ion exchange removes undesirable ions and replaces them with acceptable ions.
This section introduces some basic chemistry terms and reasons why each must be attended in the plant.
The reason for controlling pH of the steam plant water is to minimize and control corrosion. The presence of excess H ions in solution results in an acidic condition. Acidic conditions are detrimental to the materials of construction in a number of ways.
An acidic condition in the steam plant water results in processes that are potentially harmful to the system as follows. First, a low pH promotes rapid corrosion by deteriorating, or "stripping off," the protective corrosion film. Second, corrosion products, such as ferrous oxide (Fe2O3), which is predominant in the corrosion film, are highly soluble in an acidic solution. Figure 1 depicts how the corrosion rate increases as the pH decreases. Thus, for facilities not using aluminum components, a neutral or highly basic pH is less corrosive.
Figure 1: Corrosion Rate Increases as pH Decreases
pH is normally maintained at a value between 8 and 9.5, which means that the facility is operating the feed and condensate systems in a “basic condition” versus an “acidic condition.” The boiler is operated at a much higher pH, usually around 10.5.
Acidic pH values will cause attack on the boiler metal with resultant solution and corrosion of the metal. Corrosion is general in nature over the entire surface, with some localized attack. The obvious control measure to prevent such corrosion is neutralization of acidic characteristics by the use of alkali. In the treatment of boiler water, soda ash, and caustic soda commonly are employed for this purpose.
As previously stated, except in the special cases of corrosive attack involving high-pressure boiler operation, experience has shown that it is desirable to maintain boiler water pH of approximately 10.5. This pH value is sufficiently high to avoid acidic attack on boiler metal and also provides a minimal alkaline (basic or nonacidic environment in the boiler water for the precipitation of scale-forming salts by the internal treatment employed).
Control of corrosion in the condensate system can be made by selection of either neutralizing amine, such as morpholine and ammonia, or filming amines. The final selection can and should be made with the assistance of the water consultant.
Control of the dissolved oxygen content at your facility is of paramount importance because of its contribution to increased corrosion. The base reactions of concern regarding high concentrations of dissolved oxygen are the following:
These reactions are dependent on both the concentration of oxygen and temperature.
Reaction 1 is predominant at high temperatures (>400°F) in the presence of lower oxygen concentrations. This corrosion film, ferrous oxide, is also known as magnetite and is a black, generally tightly adherent film that provides a protective function to surfaces within the facility.
Reaction 2 occurs at temperatures below 400°F in the presence of higher oxygen concentrations. Ferric oxide (FeO) is more commonly known as rust and is generally a reddish color. This corrosion product adheres loosely to surfaces and is, therefore, easily removed and transported throughout the system for subsequent deposition and possible irradiation.
In either of the reactions, the corrosion rate is accelerated by increased concentrations of dissolved oxygen and can be aggravated further by the presence of other substances that may be present in the system. In addition to the direct contribution to corrosion, oxygen reacts with nitrogen to lower the pH of the water, which also results in an increased rate of corrosion. Oxygen and nitrogen react to form nitric acid. In all the reactions presented, it can be seen that oxygen concentrations promote corrosion. It follows then that if corrosion is to be minimized, oxygen concentrations must be maintained as low as possible. Concentration may be monitored on a continuous basis by using an inline analyzing system or periodically by withdrawing a sample volume and analyzing that sample. Monitoring oxygen levels is done not only to ensure that no oxygen is available for corrosion, but also to indicate the introduction of air into the system.
Free oxygen (dissolved) can and will corrode feedlines, economizers, steam drums, and downcomers. The corrosion can be general but is more likely evident as pitting of internal surfaces. Idle boilers not properly stored will corrode until either the oxygen has been used or the boiler returned to service. In any case, the boiler life will be shortened.
Two methods of oxygen control are:
Oxygen content is specified to be less than 7.0 ppb by weight at the economizer or feedwater heater inlets. This level of oxygen typically is attained in a correctly operated deaerator if all sprays, trays, and nozzles are in place and the steam supply is maintained. In addition to the water being deaerated to this oxygen level, further oxygen level control is achieved through the use of oxygen scavengers such as sulfite or hydrazine. Sodium sulfite or hydrazine can be used at all pressures up to 1,000 psig. Both chemicals are best used after the deaerator. In addition to hydrazine and sodium sulfite, complex organic oxygen scavengers are available. These scavengers may decompose at higher boiler operating pressures, contributing carbon dioxide or other multiple short-chain organics to the water. In high purity water systems, these compounds will affect the pH and conductivity.
Conductivity of facility water is measured to provide an indication of dissolved ionic substances in the coolant. Conductivity measurements provide quantitative rather than qualitative information, because it is possible to determine the total conductivity of the ions present but not the specific types of ions present. Because many ions, such as iron (Fe ), chromium (Cr ), copper (Cu ), and aluminum (Al ), are susceptible to forming oxides and plating out as scale on heat transfer surfaces, steam plant conductivity is normally controlled at a level as low as practicable and consistent with pH. By monitoring conductivity levels in the systems, the operator is able to cross-check the chemistry of these systems, thereby achieving a higher confidence level in the parameters measured.
Regardless of the operating limits specified for a given facility, operating relationships can be established between pH and conductivity levels of the coolant. Excessively high conductivity levels are an indication of the presence of undesired ions. This condition warrants further investigation to locate the source of the impurity because, in addition to other chemistry problems, it contributes to general corrosion by increasing the reaction rates of the electrochemical cells. The purity of the makeup water and any pH control agents added should be verified to determine the cause. pH should also be checked because of the relationship of these parameters. Other chemistry parameters should also be checked, such as Cl and F. After the cause of high conductivity has been determined, appropriate steps should be taken to return conductivity to its normal value. One method that is often used is a feed-and-bleed procedure, whereby water is added to and drained from the facility at the same time. If this method is used, verification of makeup water purity must be ensured to prevent compounding the problem. Low conductivity is also an indicator of a potential problem because, in high-purity basic systems, the only possible cause of low conductivity is a low pH. For example, in a system using high-pH ammonium hydroxide control, the introduction of air into the facility could result in the formation of nitric acid (HNO) with a reduction in pH.
Another parameter that is carefully monitored and controlled in most facilities is chlorides (Cl). The reason for maintaining the chloride ion concentration at the minimum level practicable is that several forms of corrosion are affected by the chloride ion, and the type of greatest concern is chloride stress corrosion. When high levels of Cl are suspected or detected, immediate steps must be taken to eliminate the source and remove Cl from the system because of the potential consequences. If Cl is present in the system, one method of removing it is to initiate a feed-and-bleed operation after determining that makeup water supplies are not the source of contamination. Because of the large volume of water normally contained in the system, cleanup by this method involves considerable amounts of pure water and a significant amount of time. An additional problem associated with feed-and-bleed operations is the resulting change (usually lowering) in pH.
This lowering of pH has the potential to further aggravate the occurrence of chloride stress corrosion. During conditions that require the use of feed-and-bleed to correct a chemistry anomaly of any type, increased attention to all parameters becomes increasingly important.
Ion exchange is a process used extensively at generating stations to control the purity and pH of water by removing undesirable ions and replacing them with acceptable ones. Specifically, it is the exchange of ions between a solid substance (called a resin) and an aqueous solution. Depending on the identity of the ions that a resin releases to the water, the process may result in purification of water or in control of the concentration of a particular ion in a solution.
An ion exchange is the reversible exchange of ions between a liquid and a solid. This process generally is used to remove undesirable ions from a liquid and substitute acceptable ions from the solid (resin). The devices in which ion exchange occurs commonly are called demineralizers. This name is derived from the term demineralize, which means the process where impurities present in the incoming fluid (water) are removed by exchanging impure ions with H and OH ions, resulting in the formation of pure water. H and OH are present on the sites of resin beads contained in the demineralizer tank or column.
There are two general types of ion exchange resins: those that exchange positive ions, called cation resins, and those that exchange negative ions, called anion resins.
Chemically, both types are similar and belong to a group of compounds called polymers, which are extremely large molecules that are formed by the combination of many molecules of one or two compounds in a repeating structure that produces long chains. A mixed-bed demineralizer is a vessel, usually with a volume of several cubic feet, that contains the resin. Physically, ion exchange resins are formed in the shape of very small beads, called resin beads, with an average diameter of about 0.005 millimeters. Wet resin has the appearance of damp, transparent amber sand and is insoluble in water, acids, and bases. Retention elements or other suitable devices in the top and bottom have openings smaller than the diameter of the resin beads. The resin itself is a uniform mixture of cation and anion resins. The ratio is normally two parts cation resin to three parts anion resin. In some cases, there may be chemical bonds formed between individual chain molecules at various points along the chain. Such polymers are said to be cross-linked. This type of polymer constitutes the basic structure of ion exchange resins. In particular, cross-linked polystyrene is the polymer commonly used in ion exchange resins. However, chemical treatment of polystyrene is required to give it ion exchange capability, and this treatment varies depending on whether the final product is to be an anion resin or a cation resin.
The ions (H and Cl) are replaceable by other ions. That is, H will exchange with other cations in a solution, and Cl will exchange with other anions. In its final form, an ion exchange resin contains a huge, but finite, number of sites occupied by an exchangeable ion. All of the resin, except the exchangeable ion, is inert in the exchange process. Thus, it is customary to use a notation such as R-Cl or H-R for ion exchange resins. R indicates the inert polymeric base structure and the part of the substituted radical that does not participate in exchange reactions. The term R is inexact because it is used to represent the inert portion of both cation and anion resins, which are slightly different. Also, the structure represented by R contains many sites of exchange, although only one is shown by the notation, such as R-Cl. Despite these drawbacks, the term R is used for simplicity.
A particular resin may be prepared in different forms according to the identity of the exchangeable ion attached. It is usually named according to the ion present on the active sites. For example, the resin represented by R-Cl is said to be the chloride form of the anion resin, or simply the chloride form resin. Other common forms are the ammonium form (NH -R), hydroxyl 4 form (R-OH), lithium form (Li-R), and hydrogen form (H-R).
The mechanics of the ion exchange process are somewhat complicated, but the essential features can be understood on the basis of equilibrium concepts and recognition that the strength of the ionic bond between the resin and an ion varies with the particular ion. That is, for a particular resin, different ions experience different attractions to the resin. The term affinity is often used to describe the attraction between a resin and a given ion. This affinity can be described quantitatively by experimental determination of a parameter called the relative affinity coefficient.
Exchange capacity is the amount of impurity that a given amount of resin is capable of removing.
The term normally applied to ion exchanger effectiveness is decontamination factor (DF), which is defined as “a ratio of the concentration (or activity) of the fluid at the inlet compared to the concentration (or activity) at the effluent,” which expresses the effectiveness of an ion exchange process.
The preboiler equipment (feedwater heater, pumps, lines, etc.) is constructed of a variety of materials often including copper alloys, carbon steel, stainless steel, and phosphor bronze. The reduction or prevention of corrosion depends upon the optimum pH level that usually ranges from 8.0 to 9.5.
Generally, boiler feedwater is filtered and deionized prior to the introduction to the steam cycle as make-up. Make-up water is required to offset the losses that occur due to valve packing leaks, turbine gland leaks, boiler feed pump gland leakage, blowdown, or other vents in the steam cycle. These losses must be replenished over time.
Condensate boiler feedwater is one of the best sources of makeup water but sometimes contains contaminants, such as acidic gases, that make it corrosive to metal surfaces. Other contaminants from industrial plants can range from hydrocarbons to oil. This valuable recyclable water should be purified and reused it at all possible.
If magnetic oxides are present in the feedwater, they can be removed by filtering on resin beds or by electromagnet filters. Magnetic oxides typically are iron compounds such as iron oxide or iron silicates. Deposits of iron oxide are typically found in boilers operating with very pure feedwater.
The usual source of these deposits lies in corrosion external to the boiler. Corrosion of iron or steel can result in the solution of iron by the condensate or feedwater, with its subsequent precipitation under the higher temperature and alkalinity conditions of the boiler water. The usual causes of such corrosive action are dissolved oxygen and carbon dioxide. The prevention of the iron oxide deposit in the boilers requires elimination of the corrosive effect of these gases. The source of dissolved oxygen should be determined through the use of mechanical and/or chemical deaeration.
While internal corrective measures such as the use of organic sludge conditioning agents may be applied to minimize iron oxide deposits, the basic solution to the problem lies in correcting the corrosive conditions that led to the solution of these metals by the condensate or feedwater. In some cases, the source of iron oxides and iron silicate boiler corrosion is not the result of corrosion external to the boiler. Corrosive attack on the boiler metal by high caustic concentrations or dissolved oxygen may be the cause of this problem.
Oil typically is not found in boiler feedwater. Oil may be introduced into feedwater through pump or steam turbine lubrication system leaks. Oil may also be introduced into steam cycle equipment during maintenance activities. Oil has a tendency to float on water. Consequently, only a small amount of oil present in boiler drum water is removed by blowdown. It is practically impossible to obtain a representative sample of boiler water for oil content testing. Samples used for oil content testing may be taken from condensate or boiler feedwater. The determination of oil can rarely be employed as a control test since no accurate rapid method of determination has been developed. Some of the most serious boiler operation problems, such as tube warpage, rupture, localized overheat, foaming, priming, and scale formation, can be traced to oil contamination of boiler feedwater. Overheating of boiler heating surfaces by oil occurs because the oil forms a thin film on the surface. The film acts as an insulator that retards the heat transfer process. This prevention of rapid heat transfer results in an increase in tube metal temperature.
Compounded oils may cause foaming of the boiler water. The impure steam resulting from such boiler water may foul superheater tubes, turbine blades, clog lines, and traps; it could also foul other heat transfer equipment.
Oil in boiler water should be kept at an absolute minimum. No method of internal conditioning can successfully cope with this problem, and oil should be removed externally before it enters the boiler. Various types of equipment are available for removing oil, including mechanical and chemical methods.
Mechanical methods can be applied to steam and condensate, whereas chemical methods are restricted to condensate only.
Silica can be present in water in two distinct forms. Silica is expressed in water as percent silicon dioxide (SiO2). Silica is difficult to remove from water and will result in scaling of heating surfaces or steam turbine components.
Dissolved solids are in true solution and cannot be removed by filtration. Suspended solids are not in true solution and can be removed by filtration.
Total solids represent the sum of the suspended and dissolved solids. Solids in boiler water may be magnetic or nonmagnetic.
The origin of the dissolved solids present in boiler water lies in the solvent action of water in contact with the minerals of the earth or boiler components. Suspended solids are small particles of insoluble matter, mechanically introduced by the turbulent action of the water on the solid or boiler parts.
Suspended solids are objectionable in boiler water. These solids may be corrosive in nature or form scale on boiler heating surfaces. For these reasons, suspended solids must be eliminated from boiler water.
Dissolved solids are typically sulfates, bicarbonates, and chlorides of calcium, magnesium, and sodium. Each of these ions may produce a specific effect in boiler feedwater. However, the additive effect of various boiler water constituents may produce a tendency for carryover.
Calcium and magnesium salts are the most common source of boiler scale. Internal chemical treatment is used to prevent deposit and scale formation from the residual hardness concentrations remaining in the feedwater and also to maintain clean boiler heating surfaces.
The most common source of scale is the breakdown of calcium bicarbonate to form calcium carbonate under the influence of heat. The precipitation of calcium carbonate to form boiler scale readily takes place where the boiler feedwater contains any appreciable quantity of calcium bicarbonate.
The action illustrated above may also be the cause for the formation of feedline and economizer deposits and deposits on feedwater heater surfaces.
Hardness in water is the sum of the calcium and magnesium present. Hardness is also reported as calcium hardness and magnesium hardness. In order to subdivide the hardness in this manner, it is necessary to determine the calcium as well as total hardness. The difference is assumed to be the magnesium hardness. Hardness in water is objectionable for its soap’s destroying and scale-forming properties. In boiler feedwater conditioning, the hardness of water can cause the formation of scale on evaporative surfaces, as well as excessive sludge or "mud," unless properly treated. Hardness may also cause scale and deposits in feedwater heaters, feedlines, and economizers.
The formation of scale and sludge deposits on boiler heating surfaces is the most serious problem encountered in steam generation. The object of the majority of the external treatment processes is to remove from the boiler feedwater those substances that will contribute to scale or deposit formations in the boiler.
The primary cause of scale formation is decreasing solubility levels with increasing temperature. Consequently, the higher the boiler operating temperature, the more insoluble become the encrusting salts. No method of external chemical treatment operates at a temperature as high as that of the boiler water. Therefore, as the feedwater temperature is elevated to the operating boiler water temperature and concentrations, the solubility of the scale forming salts is exceeded and they crystallize from solution as scale on the boiler heating surfaces. Boiler scale creates a problem in boiler operation because the different types of scale formed all posses a low degree of heat conductivity. The presence of scale is, therefore, equivalent to spreading a thin film of insulation across the path of heat transfer from the high temperature gasses to the boiler water. The presence of the heat insulating material will retard heat transfer and cause a loss in boiler efficiency. Stack gas temperature may increase as the boiler absorbs less heat.
After initial treatment of the water that becomes feedwater, the boiler water needs internal treatment. There are five readily accepted methods of internal treatment used for natural circulation drum-type boilers. These methods are:
The phosphate-hydroxide boiler water treatment prevents hardness compounds from baking on, or scaling, boiler heat transfer surfaces. The action of this method occurs because an excess of hydroxide alkalinity is maintained. The phosphate internal boiler water treatment process relies on sodium phosphate to convert hardness in boiler water to virtually insoluble calcium phosphate. Trisodium phosphate, disodium hydrogen phosphate, and sodium metaphosphate are the chemical compounds typically used for this method.
Calcium phosphate solids are readily removed by the continuous blowdown from the drum or the bottom blowdown (referred to as the intermittent blowdown). This method is characterized by high levels of total suspended solids in the continuous blowdown from the drum. In this treatment method, phosphates (PO4) passivate internal tube heating surfaces.
High-operating pressure heating surfaces cannot tolerate intentional formation of solids. To ensure solids loadings are not exceeded, the maximum operating pressure for this treatment method is 1,500 psig.
The coordinated phosphate treatment is an extension of the phosphate-hydroxide treatment and provides a better balance between acid and alkaline constituents. In this method, the action is accomplished by having no excess of "tree" hydroxide available. The control Na/PO4 ratio for this treatment is at or slightly below 3. This treatment method minimizes the potential for caustic corrosion and ensures that the deposits formed are easily removed, producing lower solids levels and high steam purity. Additionally, acids formed are neutralized and surface passivation is achieved by phosphate (PO4) in the water.
The congruent phosphate internal water treatment is based on maintaining a Na/PO4 ratio of 2.3 to 2.6. This method relies on a family of curves and is aimed at protecting the boiler heating surface tubes from either caustic gouging which can occur above the 2.6 control range or acid phosphate tube failures, which can occur below the 2.3 control range.
The chelant treatment uses a complex metal compound to dissolve the hardness and allow removal from the boiler water by continuous blowdown. Chelant is not used above 1,500 psig, and some sources recommend limiting its use to 1,000 psig. The pH control of the feedwater and boiler water is important to protect the iron from chelant attack. One (1) ppm excess chelant in the boiler water is usually satisfactory.
The all-volatile treatment employs only volatile chemicals for condensate return pH control, feedwater, and boiler water. The commonly used additives are ammonium hydroxide, cyclohexylamine, and morpholine. The dissolved oxygen scavenger typically used with AVT treatment is hydrazine. Higher concentrations of this additive generally are used during startup and shutdown operations due to less efficient deaeration during these periods.
Phosphate treatment is used in most drum-type boilers for pH control and protection against hardness. It is, however, difficult to control in transient situation. Changes caused by hideout, and variations in water volume, affect the phosphate concentration. Not all boilers experience phosphate hideout. Whether or not it will occur will depend on the relationship between phosphate concentrations, metal temperatures, and the availability of concentrating sites and deposits. Minimizing deposits in boiler steam generating tubes reduces phosphate hideout.
Excessive hideout normally is corrected by chemically cleaning followed by conditioning of the cleaned surfaces with phosphate until a stable film of magnetite is reformed on the tube surfaces. Even though alkaline treatment may be difficult to control in some instances, its effectiveness in controlling contaminants is not impaired and can be used effectively in emergencies.
Steam issuing from a boiler can be carrying droplets of boiler water (mechanical carryover) or vaporization of boiler water salts. The sum of the mechanical and vaporous carryover is the total carryover. Carryover is the term applied to the continuous entrainment of a relatively small quantity of boiler water solids with the steam.
As boiler water is carried over in the steam, deposits will form in superheaters, nonreturn valves, piping, or steam turbines. These deposits act as insulators in superheater heating surfaces, allowing the tube metal temperatures to rise. Serious loss to turbine efficiency may result or encrustation of governor valves may permit overspeeding of and damage to the machine.
The most common causes of carryover attributable to incorrect operation are those associated with foaming or priming in the boiler and too high a water level in the drum. In all these cases, significant amounts of water enter the steam separator and reduce its efficiency. Priming and foaming are always undesirable and may be dangerous, and their cause should always be investigated and corrective action taken. If carryover is occurring due to incomplete steam separation or desuperheating with impure water, the superheater will certainly be fouled or damaged. An increase in pressure drop across the superheater or loss of superheater tubes are sure warning signals of possible carryover.
Foaming is the formation of a large amount of foam or bubbles in the boiler due to failure of steam bubbles to coalesce and break. The foam produced may entirely fill the boiler drum steam space or it may be of relatively minor depth. In either case, this foaming condition causes appreciable boiler water entrainment or steam moisture content. Increased steam moisture content increases the solids loading to the superheaters. Excessive dissolved and suspended boiler water solids cause foaming. Sudden or dramatic load swings may also cause foaming.
Priming is characterized by large amounts of water passing out of the boiler with the steam, usually in intermittent slugs. Priming is more violent than foaming. This action is similar to the "bumping" experienced when water is boiled in an open beaker. It may occur simultaneously with foaming. High water levels in steam drums promotes priming.
HRSG evaporator circuits are designed to operate with continuous and intermittent blowdowns. The HRSG is designed to normally operate with a 1% continuous blowdown rate and no intermittent blowdown. The continuous blowdown rate can be adjusted to meet steam purity requirements depending on drum water quality.
During normal operation, feedwater constantly enters the drum as steam leaves. The impurities in the feedwater and those separated out of the steam will remain in the boiler drum water. The continuous removal of impurities by a blowdown line is referred to as continuous blowdown (CBD). The CBD line is routed to either the boiler drain tank or the blowdown flash tank. This routing can be achieved through proper valving of the lines. A needle type globe valve is used to set the CBD flow. If the impurities or solids are not removed by a CBD, the solids will become more and more concentrated and eventually deposit on internal tube surfaces as scale or result in carryover to the superheater and steam turbine. The tube surface scale formation reduces heat transfer and can lead to overheating and eventual tube failures.
The intermittent removal of sludge or deposits through a bottom blowdown line from feed header is referred to as intermittent blowdown (IBD). The IBD is designed to remove any sludge formed in the boiler water and maintain boiler water chemistry within the design limits. Unlike the CBD, the IBD is operated manually for short durations (several seconds) to remove suspended solids that may have settled in lower feeder headers. Feeder headers are the lowest water space in the HRSG evaporator circuit.
During cold startup and load swings, the IBD system is available to remove excessive amounts of feedwater to control boiler drum level and water solids for steam purity considerations.
Intermittent blowdown is not normally needed after the unit is stabilized at a normal operating load, provided the continuous blowdown system can maintain drum water solids level. Increases in solids concentration may be attributed to upsets in water treatment or changes in water chemistry.