On the issue of sludge and scale formation in a steam boiler

Ivan Tikhonov

Having worked for many years with water-chemical modes of steam boilers of low and medium pressure, I have noticed that even if the feed water does not meet all the requirements of the boiler manufacturers, it does not always lead to failure of the boiler (damage to the heating surfaces of the boiler by corrosion or overheating due to salt deposits). In this case the economic performance of the boiler will deteriorate, there will be an increase in purge, more frequent replacement of equalizer glasses, frequent cleaning of sensors, etc. But boilers can work at insufficient quality of feed water for at least not less than the established service life. This often happens in industries where due to the specifics of the products there is insufficient attention to the boiler and proper operation and laboratory control are not carried out. However, boilers are working during their service life in some industries and even longer, while others quickly fail. At the same time, the quality of feed water in both cases leaves much to be desired. Let’s try to figure out what cases and what parameters are critical for the operation of the steam boiler, and what parameters, even if they are exceeded, allow you to operate the equipment.

First, it should be noted that the water treatment technology for the steam boiler should ensure that all parameters of the feed water of the steam boiler are in accordance with the requirements of the manufacturer. However, I am often faced with a situation where there is some misunderstanding with the amount of residual hardness of feed water.

For low and medium pressure boilers, the feed water hardness (according to the requirements of the manufacturers) should not exceed 0.02 mg-eq/l (0.01 mmol/l). It should be taken into account that the concentration of salts in the boiler water increases due to evaporation. Accordingly, if the hardness of the feed water is 0.02 mg-eq/l, in the case where the coefficient of evaporation of boiler water is 10, the hardness of boiler water will be equal to 0.02*10=0.2 mg-eq/l. This value of water hardness is large enough for calcium to form solid particles of calcium carbonate, and magnesium to form magnesium hydroxide particles in the presence of bicarbonate in water and under the condition of distillation of carbon dioxide from water with steam.

However, when steam boilers are fed with feed water with such hardness, no scale is formed in the boilers.

Carbon dioxide is removed from the boiler water with steam and, accordingly, bicarbonate ion becomes carbonate, which in turn is hydrolyzed to form hydrate. As a result, the amount of hydrate in the boiler water increases, which can be determined by the phenolphthalein alkalinity of the boiler water. Maximum alkalinity of boiler water is allowed up to 26 mg-eq/l. As a rule, most of this alkalinity is hydrated (OH), i. e. boiler water contains a large amount of hydrate. If feed water with salts of hardness enters boiler water with a high content of hydrate, then immediately the following reactions will begin:

(1)

(2)

As we can see, in reaction (1) sodium hydrate binds hydrogen of bicarbonate into water, and bicarbonate becomes carbonate and precipitates as calcium carbonate. Remaining in the boiler water sodium bicarbonate after the removal of carbon dioxide with steam again becomes sodium hydrate.

Reaction (1) is fast enough even in water at a low temperature, not to mention instantaneous flow at the temperature of the water in the boiler.

You can do a simple experiment. Take 1 liter of tap water with hardness of 4 mg-eq/l and alkalinity of 2.5 m-eq/l, and add 0.5 grams of NaOH.

With intensive mixing of water, the formation of a suspension will begin in a few seconds in the water. After a few minutes all the carbonate hardness of the water will be in the state of suspension. At the same time, this suspension will be settling for a long time. There will be only non-carbonate hardness in water in the dissolved form. The driving force of the reaction process (1) increases as the hydrate concentration in the water increases. In this case, even the initial crystallization centers (the presence of suspended solids in water, etc.) are not required for the formation of a carbonate hardness suspension. i. e. the higher the pH value of the water, the faster the carbonate hardness salts are released from it. This process has a simple physical meaning. If the water is saturated with carbon dioxide, then such water will dissolve calcium carbonate to form calcium bicarbonate, and, accordingly, if the carbon dioxide is distilled from the water, the calcium bicarbonate will go back to the solid calcium carbonate.

It can be said that in the presence of a large amount of sodium hydrate in the boiler water, calcium and magnesium bicarbonate entering with the feed water turns into the sludge in the thickness of the boiler water, and not on the heating surfaces of the boiler.

In the absence of alkaline medium in the boiler or insufficient amount of sodium hydrate carbonate hardness salts will be deposited on the heating surfaces of the boiler, in the zone of high temperatures where there is intensive removal of carbon dioxide with evaporating water.

In my practice, there were cases of operation of fire-tube steam boilers (8 bar) with feed water hardness of more than 3.5 mg-eq/l. At the same time, the entire hardness of the source water was about 7 mg-eq/l. There was a softening of only half of the hardness of the feed water due to the failure of the softening unit. In this case, the pH of the boiler water fell from 12.3 to 11.5. That is, the hydrate concentration in the boiler water fell from 20 mmol/l to 3.2 mmol/l. At the same time, the boiler water looked like lime milk and after 2 weeks of operation in this mode, a sampling device was clogged on one of the boilers. Phenolphthalein (F) and methylorange (M) alkalinity of the boiler water is measured when operating boilers with fully softened water and with partial water softening show fundamentally different values. If during normal operation the alkalinity of phenolphthalein was F= 18-22 mmol/l, methylorange M= 1,8-2,5 mmol/l, when working with greater hardness  F=10 mmol/l, M=10 mmol/l. It is obvious that most of the boiler water hydrate obtained from sodium bicarbonate of softened feed water is spent on the process of sludge formation as shown in reactions (1, 2)

In the process of analysis on the alkalinity of methylorange a gradual decrease in the pH of samples of the boiler water is seen during dosing of hydrochloric acid  the slurry of calcium carbonate dissolves according to the formula:

(3)

This process increases the value of alkalinity for methylorange, although in boiler water the amount of dissolved sodium carbonate will be at a level of not more than 2 mmol/l. Accordingly, 8 mmol/l is the amount of carbonate that falls in the form of sludge with calcium and magnesium.

What happened to the boilers?!

Boilers worked in this mode for about 3 weeks. After the repair of the water treatment installation the internal inspection of the boiler was made. As a result of the examination it was decided not to carry out chemical washing of the boilers. The condition of the heating surfaces of the boiler was quite satisfactory. In fact, apart from the need to flush the sampling device, there were no serious problems. Almost all hardness salts were removed from the boiler in the form of sludge with continuous and periodic blowing (the number of which was significantly increased). In this case, the amount of hydrate in the boiler water was enough to prevent the formation of a dense scale on the heating surfaces of the boilers.

It should be noted that convective and screen pipes of water-tube boilers with such active sludge formation are likely to be clogged, which will lead to their overheating and destruction.

Another example from my practice. In the steam boiler room with 2 fire-tube boilers (11 bar), the water softening unit was damaged due to the ingress of steam into it. The steam got into the plastic cylinders of the water treatment installation through the heat exchanger of the heating of purified water in front of the deaerator. There was no reduction of steam and since steam entered the heat exchanger even in the absence of water flow, the heat exchanger eventually overheated and steam with a pressure of 11 bar entered in to the purified water. As a result, the softening unit is completely out of order. The boiler was fed with water with a hardness of 2.0 mg-eq/l and an alkalinity of 1.0 mmol/l (taking into account the return of condensate). Alkalinity of boiler water was: F=0,2-0,5, M= 7,5-8,5. As it can be seen actually there was no hydrate in the boiler, and a slight alkalinity of phenolphthalein was formed only due to about 15-20 mg/l sodium which was contained in the source water supply (city water supply). In such circumstances, I assumed that the boiler would not work longer than two weeks. The water of the purge line was quite clear. Almost all hardness salts were deposited on the heating surfaces under the condition of the absence of sufficient hydrate in the boiler water.

However, the boilers worked for three months. During this time, the softening unit was replaced, though not as quickly as I had said. Then the boilers were examined. The result showed almost complete clogging of the boilers with hardness salts. At the same time, there were solid formations at the bottom of the boiler, the shape of which showed how they were formed and from which part of the heat pipe they fell off during the boiling of the water. I could not say how long the boilers would have lasted, but the situation was supercritical. The boilers were chemically washed for about a week.

These two cases show fundamentally the development of the processes of solid phase formation of salts with carbonate hardness in the feed water, depending on pH. It can be concluded that the presence of hydrate in the boiler water fundamentally changes the picture of the allocation of hardness salts from water.

The use of phosphates to bind the calcium of boiler water into hydroxylapatite (Ca10(PO4)6(OH)2) is quite risky, especially in the case of insufficient control over the hardness of feed water. Hydroxylapatite is formed only in the presence of hydrate in water. In the absence of hydrate, but in the presence of hardness salts, they bind into phosphorite. Unlike hydroxyapatite, which is slurry and is excreted with the purge, phosphorite forms hard deposits. In the case of dosing phosphates into insufficiently softened water before the deaerator, phosphorite deposits can already be found in the deaerator, which can be removed only by a laborious mechanical method.

Another real case from my practice. The installation of the steam boiler water treatment consisted of reverse osmosis desalination and the second stage of softening. As a result of improper adjustment, osmosis went out of order, and the boiler room worked for about two years only at one stage of softening. The hardness of raw water was 11 mq-eq/l. The hardness of the makeup water was on average 0.15 mg-eq/l after one stage of softening. At the same time, the adjustment organization insisted on dosing phosphates into the water in the deaerator. After two years of work, I was invited to audit of this boiler room. As a result, I found out that about once in three months quarter, the boiler had to be stopped due to the fact that the water stopped coming from the deaerator. In the deaerator, phosphorite was formed, which was broken and removed manually. At the same time, the feed pumps also failed after the deaerator. Either their repair or replacement was required. As a joke it can say that this boiler used a new promising sedimentary water treatment technology, where a deaerator was used as a clarifier. If we exclude the dosage of phosphate, then with a hardness of 0.15 mg-eq/l, the fire-tube boiler could work well in this case. The fact that the coefficient of evaporation of boiler water was only 4 in this case, due to the high salt content of feed water and the lack of return of condensate. As a result, the hardness salts in the boiler water would be only 0.15*4=0.6 mg-EQ/l. This would increase the methyl orange alkalinity of the boiler water by 0.6 mg-EQ/l which would be determined only at low pH values of the sample. This sludge is removed from the boiler with purges.

Another point should be noted. It is known that only the value of the feed water hardness is set. Suppose that in a steam boiler there is a large ratio of the return of desalted condensate, about 90%. In this case, as a rule, only one stage of softening is set. As a result, the hardness after one softening stage can be obtained on average 0.1-0.07 mg-eq/l (with a reasonable consumption of salt for regeneration). After mixing the softened water and condensate, the feed water hardness will be about 0.01 mg-eq/l, which meets the requirements even more than required. But, in this case, the salt content of feed water will definitely have a rather small value and, accordingly, the coefficient of evaporation of boiler water can have very large values. In my practice I set the maximum evaporation rate at 60. The fact is that when the coefficient is about 55 the boiler water becomes significantly turbid, even if the feed water passes osmosis, and the condensate contains virtually no iron. From the point of view of the theory it is difficult for me to explain this moment, but practice shows this threshold of value. Probably substances even in small quantities in the feed water after evaporation up to 60 times cause the clouding of the water. If the feed water with hardness of 0.01 mg-eq/l evaporates by 60 times, the hardness of the boiler water will reach 0.01*60=0.6 mg-eq/l. This value is equal to the boiler water hardness value for the case of boiler feed water with hardness of 0.15 mg-eq/l and evaporation coefficient of 4. At the same time, from the point of view of the requirements, in the first case there is an obvious excess of hardness in the feed water, and in the second, the requirements of the rules are met.

Summing up an intermediate result it is necessary to tell that for boilers of low and average pressure not only the amount of hardness salts of feed water, but also the amount of alkalinity is important, to be exact their ratio in feed water is important. Alkalinity should be several times greater than the residual hardness.

As a rule, this condition is always observed, because in natural waters alkalinity takes from 50 to 80 % of the anion composition of water. But in the case of using osmosis to prepare make-up water, alkalinity as well as hardness will be removed on the membrane. In this case, the residual hardness should be replaced by sodium in the softening unit. Thus, the hardness of the water will be practically absent, and even small amount of source alkalinity is enough for the formation of a sufficient amount of hydrate in the boiler water. But there is a danger that contaminated condensate will return. Even a small amount of carbonate hardness salts in the condensate will be enough to bind hydrate into bicarbonate and precipitate. And in the case of reducing the pH of the boiler water even small amount of hardness salts will precipitate in the form of scale on the heat transfer surfaces of the boiler and substantially increase the risk of under scalecorrosion, and overuse of the fuel gas due to the increase of the thermal resistance of the heat transfer surfaces.

In conclusion, it is necessary to mention the deposits of sulfates and salts of silicic acid in the boiler. In the boiler, the formation of calcium sulfate is possible. In practice, in boilers using surface water, as a rule, sulphate deposits are not formed. Sulfate deposits will only form if there is more calcium than bicarbonate in the feed water, i. e. the feed water is practically not softened and the concentration of calcium must be greater than the concentration of bicarbonate. That for the surface sources water is extremely rare. Therefore, as a rule, sulphate deposits are formed in boilers without water treatment or boilers which are working on underground waters or if sodium sulfate is uncontrollably dosed into the water for the purposes of chemical deaeration.

The formation of salts of silicic acid in the boiler is possible in the presence of feed water hardness salts and high coefficient of evaporation of boiler water. In this case, if there is hydrate in the boiler water, it will not form a solid scale of magnesium silicate, but crumbly talc. It is believed that for the formation of talc it is enough to have hydrate in the boiler water which is one and a half times more than the salts of silicic acid. However, silicate scale is extremely difficult to remove and it is necessary to adhere to the requirements of the boiler manufacturer according to the permissible concentrations of silicon in feed and boiler water.

Conclusions:

1. The pH value of boiler water is the most important controlled parameter of the water chemical mode of boiler. The pH of the boiler water should be between 11.5 and 12.3 to prevent sedimentation for fire-tube boilers.
2. The use of different reagents for water treatment purposes should be considered only for corrective action on the boiler’s WCM and should not replace the main water treatment systems (osmosis, softening, etc.).
3. For many enterprises of non-energy profile, as a rule, the boiler room is not given enough attention, and many boilers work even without basic laboratory control. In this case, low-pressure steam boilers tend to fail well before their service life. It is necessary to measure the pH and electrical conductivity of the boiler water to carry out basic control. If there is an automatic purge on the boiler, the conductivity is measured automatically. If the pH of the boiler water is higher than 11.5 and at the same time the boiler water is transparent, this fundamentally indicates the absence of sludge formation processes in the boiler. At the same time, if the pH of the boiler water is from 8.5 to 9.5 and the water is transparent in most cases this indicates an active scale formation in the boiler, and requires emergency measures. In any case, it is necessary to visually control the boiler water for transparency. If it is not possible to ask professionals to determine the source of the ingress of hardness salts in the boiler in the case of formation of suspension in the boiler water or the drop in pH of the boiler water while maintaining the electrical conductivity values you can conduct a simple analysis by adding to the water sample about 1 grams of caustic soda (if not, then soda ash). After addition of caustic soda to the sample of water with increased hardness, a rather intensive formation of flakes will occur with stirring. After addition of caustic soda, flakes of carbonate hardness will be formed. After addition of soda ash, all hardness salts will participate in flocculation. The higher the water hardness, the faster many flakes are formed suspended in water. Make-up water and condensate must be checked. The reason for the sediment of hardness salts will be either poor operation of the softening unit or poor-quality condensate. In any case, perhaps, it will be useful for achieve operational personnel in the future to order a specialized organization for high-quality adjustment of the boiler room and the organization of automated control of the boiler.