Boiler Water Treating Compositions

One of the biggest challenges in boiler water treatment lies in the development of simple, easily monitored, and easily controlled programs. Ideally would be one product which can prevent scale, provide heat transfer surface protection, and protect condensate systems. However, the state-of-the-art practices have not been able to meet this challenge. Chelant programs, for example, are capable of eliminating hardness deposits. They are also known, however, to cause corrosion under certain conditions. While the chelants are capable of solubilizing hardness metal ions, their strong affinity toward iron ions may actually be the corrosion mechanism. Excessive residual chelants may not only prevent the formation of magnetite but also strip the boiler of its protective magnetite films.

The purpose of this invention is to:

(1) develop programs which provide exceptional scale prevention without the corrosion potential;

(2) develop programs which provide similar scale prevention capabilities as chelants without the corrosion potential associated with chelant application.

These purposes are accomplished utilizing certain water-soluble anionic vinyl polymers in a particular dosage range either alone or in conjunction with certain low molecular weight water-soluble polymeric dispersants including sulphonate-containing, hardness-dispersing polymers.

The invention provides a method of treating hardness present in boiler waters and scale formed on heat transfer surfaces in contact with such waters to prevent and remove scale caused by such hardness which comprises treating such waters with a water-soluble anionic vinyl polymer containing at least 30% and, preferably, 70%-100% by weight of carboxylate functionality with said polymer having a molecular weight.sup.1 within the range of 500-50,000 and with the amount of such polymer being sufficient to provide between 1-30 ppm per ppm of hardness present in the boiler waters. By the term, "hardness," we mean to include soluble and insoluble compounds of calcium, magnesium, iron, copper, aluminum, and the like.

In addition to the above characteristics, the water-soluble anionic vinyl polymer must also interact with hardness ions to sequester them. The sequestration must be of such magnitude as to yield a chelation value of at least 200 as measured by specific ion electrodes.

In a preferred embodiment of the invention, the molecular weight range of the carboxylate polymers is within the range of 1,000-30,000.

In another preferred embodiment, there is utilized in combination with the anionic water-soluble vinyl polymers another water-soluble polymer having dispersant properties such as a sulphonate-containing polymer which is capable of acting as a dispersant for any excess hardness not acted upon by the sequestrant anionic water-soluble vinyl polymer.

These polymers, as indicated, have molecular weights ranging between 500-50,000, with a preferred molecular weight range being within the range of 1,000-30,000.

The polymer may be homopolymers or copolymers of vinyl carboxylate-containing monomers. "Carboxylate-containing monomers" means that the carboxylic acid groups are either in the form of the free acid or of a water-soluble salt thereof such as alkali metal, ammonia or amine. In the case of acrylic acid polymers, it would include the amide.

Thus, the homopolymers of acrylic acid, methacrylic acid, maleic acid, fumaric acid, itaconic acid, and the like may be used. Polyacrylamide, when added to the boiler water, undergoes hydrolysis to convert portions or all of the amide groups to carboxylate groups, and as such, is also included.

In addition to using these homopolymers, water-soluble copolymeric forms may also be employed. When the copolymers are used, the amount of carboxylate should be at least 30% by monomer weight ratio of the copolymers.

A preferred group of carboxylate polymers are those derived by the hydrolysis of the corresponding polyacrylamides. These materials, after either caustic or acid hydrolysis, will contain between about 10-30% by weight of amide groups. A most preferred group of carboxylate polymers are those obtained by polymerizing acrylic acid with acrylamide at a 3:1 monomer weight ratio.

As indicated, the amount of polymers used to treat the hardness contained in the boiler waters should be between 1-30 ppm per ppm of hardness.

It has been found that the preferred water-soluble anionic vinyl polymers must exhibit a chelation value in excess of 200, and preferably in excess of 300. When applied to the invention, chelation value means the average chelation value from both calcium and magnesium determinations.

Chelation value is defined as the milligrams of calcium or magnesium expressed in terms of calcium carbonate complexed by one gram of active sequestrant. In this work, it is measured by specific ion electrode techniques. A known increment of sequestrant is added to a system containing a known amount of free (uncomplexed) calcium or magnesium. The decrease in calcium/magnesium activity (concentration) is then a direct measure of complexed species. This amount is then converted (ratioed) to yield the chelation value.

The effective mole ratio can also be computed using this information. By dividing the chelation value into 100,000 an equivalent weight for the sequestrant is determined. If the molecular weight is known, then the mole ratio is found by dividing the molecular weight by the equivalent weight. For EDTA and NTA, the value should approximate unity. For polymers, this number varies with the molecular weight and is generally greater than unity.

In a preferred mode of the invention, the carboxylate polymers described above are used in conjunction with a water-soluble polymer which is capable of dispersing hardness.

The polymeric dispersants used in this preferred mode of the invention are anionic water-soluble vinyl polymers. To be operative, they must be capable of dispersing suspended matter that normally occurs in boiler waters. They may be further characterized as containing either carboxylate functionality or sulphonate functionaltiy. Additionally, they may be characterized as having a molecular weight of at least 500 to about 50,000.

The water-soluble dispersing polymers useful in this invention may be chosen from the carboxylate containing water-soluble vinyl polymers such as, vinyl sulphonate-acrylic acid copolymers, vinyl sulfonate-methacrylic acid copolymers, sulfonated styrene-maleic anhydride copolymers, and acrylamide/acrylate copolymers.

The preferred water soluble dispersing polymer is a vinyl sulphonate copolymer synthesized from vinyl sulphonate and acrylic acid. This dispersant molecule generally contains from 5-25 mole percent of the vinyl sulphonate or its alkali metal (preferably Na) salts and from 95-75 mole percent of acrylic acid and its water-soluble alkali metal or ammonium salts. Preferably, the acrylic acid-vinyl sulphonate copolymers contain 10-20 mole percent of the vinyl sulphonate and from 90-80 mole percent of acrylic acid. The molecular weights of these preferred dispersant polymers range from as low as 500 to as high as 50,000. Molecular weight ranges of from 750-50,000 are preferred with a molecular weight range of approximately 900-15,000 being especially preferred. Ideally, the molecular weight will range from 1,000-6,000. It is surprising to find that these dispersant molecules may or may not be chelant or sequestrant molecules if treatment levels are drastically increased.

Another class of polymeric dispersants are the low molecular weight polyacrylic acids and their water-soluble salts. These materials have a molecular weight range of 1,000-5,000. The ratios in which these materials are used with the carboxylate polymers are the same as described above for the acrylic acid-vinyl sulphonate copolymers.

Yet, another class of polymeric dispersants are the low molecular weight sulphonated copolymers of styrene and maleic anhydride. These materials are preferably present as the sodium salt of sulphonated copolymers of styrene and maleic anhydride and are typically known and commercially sold as Versa TL.RTM.-3.sup.2 products. Other sulphonated copolymers of styrene and maleic anhydride are also found useful in this application when combined with the above-described carboxylate polymers having sequestrant properties.

To summarize with respect to the most preferred operational method, it may be stated that the preferred method of treating hardness present in boiler waters which are in contact with heat transfer surfaces to prevent and remove scale caused by such hardness comprises treating such waters with:

(a) a water-soluble anionic vinyl polymer containing at least 30% by weight of a carboxylate functionality, said polymer having a molecular weight within the range of 500-50,000, and

(b) a second anionic water-soluble vinyl polymer dispersant.

To summarize with respect to the most preferred polymeric dispersants, it may be stated that the boiler waters are preferably simultaneously treated with both the sequestrant (chelant) water-soluble anionic vinyl polymer mentioned above and a second anionic water-soluble vinyl polymer dispersant chosen from the group consisting of carboxylate-containing, water-soluble vinyl polymers, vinyl sulphonate-acrylic acid copolymers, vinyl sulphonate-methacrylic acid copolymers, sulphonated styrene-maleic anhydride copolymers, and acrylamide-acrylic acid copolymers.

It is particularly interesting to note that most of the chelant or sequestrant anionic vinyl polymers show an ability to disperse solids in boiler water systems if treatment levels are below those required to chelate hardness ions. As a result of this observation, the chelant or sequestrant polymer may be used in quantities above those quantities necessary to chelate all of the hardness initially found in the boiler system. When this occurs, it has been observed that any hardness ions or scale existing in the boiler system may be removed from the system and the additional chelant polymer may act as a dispersant, as well as a sequestrant for hardness contamination.

Thus, the expression and related terms relating to the addition of a second anionic water-soluble vinyl polymer dispersant in combination with the sequestrant polymer is meant to include the phenomenon described above, i.e., use of excess sequestrant polymer to provide both sequestration and dispersancy.

The ratio of carboxylate polymer to acrylic acid-vinyl sulphonate copolymer, when they are used, is within the range of 30:1 to 1:30 with 20:1 to 30:1 being a preferred range, with 20:1 being most preferred.

In general, the ratio of sequestrant carboxylate polymer to dispersant polymer is also within the range of 30:1 to 1:30. A preferred sequestrant polymer to dispersant polymer ratio is between 30:1 and 10:1 with a most preferred ratio of sequestrant polymer to dispersant polymer being 20:1. In all cases, this ratio is on a weight:weight basis.

To illustrate the many advantages of the invention, the following is presented by way of example.

For purposes of understanding the following tests, a series of various polymers and known boiler water treatment chemicals were evaluated. These evaluations were preformed in two series of testing programs. The first testing program involved the measurement of the chelation values of various polymers so as to determine by initial screening the potential of each polymer to function adequately as a boiler transport material.

The experimental design used to test the chelation value of a series of polymers was as follows:

Solutions of calcium or magnesium ions were titrated with solutions of various polymeric and other sequestering agents. The residual unsequestered metal ion concentration (or, more correctly, activity) was measured by means of a Specific Ion Electrode (henceforth S.I.E.). This data was ultimately converted to graphical representations of sequestrant performance. Sequestrant performance for Ca ion was measured by a calcium specific electrode, manufactured by Orion Research, Model 93-20. Sequestrant performance for Mg ion was measured by a Divalent cation electrode, Model 93-32, again manufactured by Orion Research. The electrode response is measured, as the sequestrant solution is added incrementally to the hardness solution. The desired solution pH is automatically maintained by feeding potassium hydroxide solution from a Mettler DV10 which is controlled by a Mettler DK10/11 system.

A short period of time is allowed after each sequestrant addition before taking a reading so that the electrode can come to equilibrium with the solution. Noise levels are typically .+-.0.2 mV using mechanical stirring (higher using a magnetic stirrer).

Prior to each titration, the S.I.E. was calibrated with standard solutions containing 1,000, 100, 10, and 1 ppm calcium or magnesium.

The S.I.E. responds to activity rather than concentration. For calcium measurements, a high, constant, ionic strength is maintained by addition of 6 g/l potassium chloride to all solutions (i.e., standards, sequestrant, and calcium sample). This maintains a constant activity coefficient for the calcium ion. The divalent sensing electrode used for magnesium measurements is subject to interference from both sodium and potassium ions at fairly low concentrations so no ionic strength buffer can be used in this case.

Typical operating conditions were 2 or 3 g/l active polymeric sequestrant titrated against (in all cases) 100 ml of 100 ppm metal ion. Under these conditions, most titrations were essentially complete after the addition of 40-50 ml of sequestrant. Sequestrant solutions were usually added in 2 to 3 ml increments. Sequestrant was added slowly so as to avoid the formation of bubbles which could be trapped at the base of the electrodes and result in incorrect readings.

All measurements were made at room temperature (measurements above about 40.degree. C. will result in rapid electrode deterioration).

The data from these experiments were graphically displayed or preferably converted to a usable form by a computer program which was written specifically for these experiments. The computer program obtains the best straight line fit through the origin using a reiterative, least-squares approach, allowing calculation of chelation values for each polymeric species.

For calcium measurements, a pH of 10 was used in most cases. Initial studies gave results indicating chelation greater than theoretical. At a pH of 9, the results were in good agreement with theory. The discrepancy at pH 10 may be due to a competing reaction (e.g., magnesium hydroxide formation). All magnesium measurements were made at a pH of 9 after this.

An attempt was made to correct magnesium results for the effects of any sodium present. However, when corrections obtained from sodium chloride solutions were applied to an NTA.Na.sub.3.H.sub.2 O titration, the "corrected" results were very unreasonable (much greater than theoretical chelation). No further attempts to correct for sodium were made.

Chelation values were determined from the initial slope of the titration curve which plots percent metal ion sequestered versus grams of active polymer added. This calculation gives practical chelation values in that no consideration has been given to which complexes are formed, the effects of competing equilibria, or various stability constants.

In the case of the polymers studied, a comparison of chelation values may be more valid than they might be in the case of strong complexing agents, such as EDTA or NTA, where a simple comparison of chelation values is not necessarily a good guide to chelation performance.

As later results will show, to be effective as a transport agent in boilers, the chelation value for the polymeric sequestrant must be above 200 and must give clear solutions for both calcium and magnesium ion test solutions. The preferred average chelation value is above 300. Most of the polymers tested appeared to sequester magnesium ion better than they were able to sequester calcium ion. However, to be successful, a polymer must have, as stated above, a chelation value above 200, preferably above 300, and be able to sequester both calcium ions and magnesium ions to give clear solutions.

The data in Table I compares different sequestrants and the chelation values obtained using the above described test. As can be seen from this Table, the sequestrants tested include not only well-known complexing agents such as EDTA and NTA but also polymeric sequestrants, as well as other sequestrants. As will be shown later, only those sequestrants which have chelation values above 200 can be shown to function as effective transport agents in a boiler system.

Other sequestrants on the list are not satisfactory as transport agents because of known thermal degradation in a boiler system. Such agents are the phosphate containing compounds listed in Table I.

Table II identifies each of the polymeric species tested

Table III lists results for the polymeric sequestrants of this invention, as well as other more common sequestrants versus magnesium ion. Again, as can be seen, those polymeric sequestrant agents which have chelation values above 200 and are thermally stable give excellent results in boiler transport. Of particular note in Table III is the result for citric acid. Although, a very large chelation value is obtained, this material does not effectively transport magnesium or calcium hardness when tested in the boiler. It is expected that these results are due to the fact that citric acid thermally decomposes when exposed to boiler operating conditions. This very well may be the benefit for the low molecular weight polymeric carboxylate polymers of this invention, that is, that thermal stability is obtained while maintaining sequestrant activity for hardness ions at proper dosages.

Of particular note in Table III is the fact that Polymer C, though giving a chelation value in excess of 200, yields a somewhat cloudy solution with magnesium at pH 9 and would not be expected to perform as well as the other carboxylate containing polymeric sequestrants of this invention. This problem might be solved by increasing the concentration of this polymer or by combining this polymer with other materials giving improved results. The polymeric materials that did not perform well with calcium were not tested for magnesium since both ions must be complexed before adequate boiler transport systems can be achieved.

Some of the more promising carboxylate-containing polymers having chelation values above 200 were tried in experimental boiler water systems. The experimental boiler is described in the paper, "The Investigation of Scaling and Corrosion Mechanisms Using Process Simulation," by J. A. Kelly, P. T. Columbo, and G. W. Flasch, paper No. IWC-80-10, given at the 41st Annual Meeting, International Water Conference, Pittsburgh, Pa., Oct. 20-22, 1980.

Table IV indicates the formulations and sequestrant polymers chosen to be tested in the experimental boiler program. Included in these tests were a phosphorous-containing sequestrant, as well as a water-soluble polymer which does not contain measurable amounts of carboxylate functionality.

Most of the experiments were conducted at 1,000 psig, 110,000 Btu/ft.sup.2 -hr heat flux, and 10 concentration cycles. The Composition I polymer was tested more extensively at 250, 600, and 1,500 psig. This laboratory boiler is of the type described in U.S. Pat. No. 3,296,027, which is incorporated herein by reference.

Feedwater was typically deionized water containing 1 ppm Ca, 0.5 ppm Mg.sup.4, and 0.5 ppm SiO.sub.2. Sulfite residual was maintained at 25.+-.5 ppm at 600 psig and 10.+-.5 ppm at 1,000 psig. Boiler water `O` alkalinity was maintained at 160-180 ppm. The pH of the polymers was adjusted to 9.

Dosage Profiles of a number of polymers were obtained under three conditions. It is apparent in FIG. 1 that:

1. the recommended dosage of Comp. I combination polymer for hardness control is about 5.3 ppm active polymers/ppm total hardness at 1,000 psig, and

2. at dosages below the recommended, the combination polymer preferentially transports Ca rather than Mg ions.

FIGS. 1, 2, and 3 indicate that Comp. I as well as Comp. III and Comp. IV have threshold inhibition capability at low dosages for the Ca ions, and chelate.sup.3 hardness ions at high treatment levels.

In general, all the tested acrylate-, acrylamide-, and vinyl sulphonate-based polymers give excellent hardness control at high dosages, as long as they contain sufficient carboxylate functionality. Among these, Comp. II and IV are the most effective. Results are listed below.

Boiler Pressure=1,000 psi, Heat Flux=110,000 Btu/ft.sup.2 -hr, Ca=1, Mg=0.5, SiO.sub.2 =0.5 ppm in the feedwater.

Boiler Pressure=1,000 psi, Heat flux=250,000 Btu/ft.sup.2 -hr Feedwater contained 1 ppm Ca, 0.5 ppm Mg, and 0.5 ppm SiO.sub.2.

Boiler Pressure=600 psi, Heat Flux=110,000 Btu/hr-ft.sup.2, Ca=1, Mg=0.5, SiO.sub.2 =0.5 ppm in the feedwater

Higher recommended dosage is required for hardness control at higher boiler pressure. the increase in treatment level is probably due to the decomposition of polymer.

As the data below indicate, the dosage required for a complete hardness recovery increased at 1,500 psig. Corrosion rate, as measured by the iron content in the blowdown, did not increase.

Ca=1, Mg=0.5, SiO.sub.2 =0.5 ppm in the feedwater. Pressure=1,500 psig, heat Flux=110,000 Btu/ft.sup.2 -hr

The combination polymer had no problem controlling hardness at low pressure (250 psi) boiler applications. Heat flux was 110,000 Btu/ft.sup.2 -hr.

Feedwater contained Ca=3, Mg=1.5, Na.sub.2 SO.sub.4 =42.6, NaCl=10, SiO.sub.2 =5, Fe=1 ppm, and enough NaHCO.sub.3 to give an M alkalinity 40 in the feedwater or 400 in boiler water, SO.sub.3.sup.= =30 ppm in the blowdown.

Feedwater Contained 3 ppm Mg and all the other components listed in 1.

Comp. I treatment can recover from moderate hardness, silica, and treatment upsets.

Pressure of the boiler=1,000 psig, Heat flux=110,000 Btu/hr-ft.sup.2. Comp. I treatment ratio 7.88 polymer/ppm hardness when initial hardness was 1 ppm Ca and 0.5 ppm Mg. Total polymer held constant throughout the test as hardness was varied.

Ca/Mg/SiO.sub.2 Swing Effect, 1,000 psi and 250,000 Btu/hr-ft.sup.2.

Feedwater contained 1 ppm Ca, 0.5 ppm Mg, and 2.5 ppm SiO.sub.2, 1,000 psi and 110,000 Btu/hr-ft.sup.2.

Scale removal using Comp. I appears feasible if hardness and silica can be discharged by the blowdown. Adequate Comp. I treatment can transport boiler deposits in addition to the hardness in the feedwater. It enhanced passivation of the boiler heat transfer surface and formed a black, magnetite film.

Feedwater contained Ca=1, Mg=0.5, and SiO.sub.2 =0.5 ppm Pressure=1,000 psig, Heat flux=110,000 Btu/hr-ft.sup.2

Feedwater contained no hardness and no silica, a badly fouled boiler. Pressure=1,000 psig, Heat flux-110,000 Btu/hr-ft.sup.2.

Feedwater contained no hardness and no silica and the boiler was relatively clean.

Pressure=1,000 psig, Heat flux=110,000 Btu/hr-ft.sup.2. Treatment=23.63 ppm Comp. I per ppm total hardness, assuming total hardness=1 ppm.

Heat flux in the range of 110,000 to 250,000 Btu/hr-ft.sup.2 had little influence on hardness recovery. At heat fluxes greater than 300,000 Btu/hr-ft.sup.2 there was a thin film of deposition on the heat transfer surface.

Test Condition: 1,000 psi, Feedwater contained 1 ppm Ca, 0.5 ppm Mg, and 0.5 ppm SiO.sub.2.

Comp. III/Comp. IV combination polymer, although less effective than Comp. I, gave reasonable hardness control at 1,000 psig. This combination polymer could be used in NH.sub.3 sensitive applications.

Ca=1, Mg=0.5, SiO.sub.2 =0.5 ppm in the feedwater. Pressure=1,000 psig, Heat Flux=110,000 Btu/ft.sup.2

The combination polymer Comp. I was initially added to the boiler at a dosage of 7.9 ppm active per ppm total hardness. The dosage was maintained for eight days. The average calcium and magnesium recoveries were 118% and 101%, respectively. Initial hydrogen level was 11 ppb but dropped to 1.5 ppb the same day. It leveled off to 0.4 ppb. Hydrogen values of 1.0-1.2 ppb are equivalent to background levels. The high initial rise of hydrogen frequently occurs in a boiler just brought on line. In addition, sulfite residuals for the first few days were lower than desired and contributed to hydrogen generation. Iron in the blowdown started off high at 2-3 ppm and declined to 1.1 ppm after eight days. The condensate frequently had a pH greater than 9, and contained small amounts of ammonia.

At this point the polymer dosage was decreased to 3.9 and the test was continued at this condition for six days. This treatment level is less than 2/3 the recommended. The average calcium and magnesium recoveries for this period were 96% and 81%, respectively. It was anticipated from the experimental scale boiler results that hardness recoveries would decline and that magnesium would be more affected than calcium. Hydrogen dropped to 0.3 ppb and iron decreased to 0.3 ppm. The temperature of the high heat flux area remained constant, indicating no scaling.

At this point, the polymer dosage was further reduced to 2.6 and held at this condition for three days. Calcium and magnesium recoveries declined to 89% and 78%, respectively, while hydrogen and iron levels were fractionally lower. During low level treatment, the temperature in the horizontal test section increased 30.degree. F. which indicated that scale was being deposited.

The polymer dosage was then restored to the original level of 7.9 for fifteen days. Within the first day, the temperature of the horizontal test section dropped 30.degree. F. As anticipated, calcium and magnesium recoveries increased dramatically and averaged 122% and 111%, respectfully. These high recovery values suggest that the polymer treatment program is removing deposits previously laid down when under-treating. Similarly, it is postulated that at the start of this test, the 118% calcium recovery was due to removal of boiler deposits that remained in the system from the previous test. Hydrogen and iron remained at relatively low levels.

FIGS. 4, 5, and 6 depict percent hardness recovery, H.sub.2 level, and iron level in the blowdown as a function of days of test. Polymer treatment was then stopped for 7 days. Only trace amounts of calcium and magnesium were recovered during this period, while hydrogen and iron remained virtually unchanged.

During the remaining period of the test, polymer dosage at 7.9 was alternated with no treatment. In the presence of polymer, recoveries of calcium and magnesium averaged 115% and 106%, respectfully.