AWT-08 (Nov-08)

Association of Water Technologies, Inc. Annual Convention & Exposition

November 5 to 8, 2008 The Hilton Austin – Austin, TX

Laboratory Evaluation

of Process Variables Impacting the Performance of Silica Control Agents

in Industrial Water Treatment Programs

Zahid Amjad, Ph.D. and Robert W. Zuhl, P.E. Lubrizol Advanced Materials, Inc.

9911 Brecksville Road Cleveland, OH 44141

© 2008, The Lubrizol Corporation. All rights reserved.

Carbosperse™ K-700 Water Treatment Polymers

1/20

Abstract Preventing silica deposition and fouling of equipment surfaces in industrial water systems (e.g., cooling, desalination, geothermal) continues to be one of most difficult challenges for water technologists. The complexity of controlling silica laden feed waters stems from the limited solubility of both amorphous (polymerized) silica and metal silicates. Once formed, silica scale is extremely difficult to remove and often requires the use of mechanical and/or chemical methods; fluoride-based chemicals which present environmental and safety concerns. Therefore, industrial water systems using silica fouling prone feed waters typically employ conservative operating criteria; e.g., cooling systems limit cycles of concentration and desalination systems limit recovery. This paper and results presented herein: • Compare the performance of polymeric additives as silica polymerization inhibitors and

particulate dispersants to minimize the deposition of unwanted silica-based scales. • Show the impact of solution temperature and iron (III) on silica polymerization and

precipitation. • Compare the performance of a new silica inhibitor with several commercially available silica

inhibitors and dispersants. • Characterize the composition of silica precipitates formed during the experiments. Keywords: silica, polymerization, silicate, scaling, inhibition, water chemistry

Introduction

Silica is generally found in water supplies in three different forms:1 reactive silica, colloidal silica, and particulate silica. Reactive silica, also known as “soluble silica,” is silicon dioxide dissolved in water, creating a compound known as mono silicic acid (H4SiO4 or Si(OH)4. This is the form of silica which will react with molybdate to give the characteristic heteropoly blue color used in analytical tests. Colloidal silica is generally considered to be polymerized silica and particulate silica is larger in size and mostly comprised of sand or flocculated silica suspended in water. The two most common forms of polymeric silica encountered in water treatment systems are colloidal and gel or glassy type materials; both forms are non-crystalline or amorphous in nature.

The adverse consequences of the deposition of amorphous silica on heat exchangers and reverse osmosis (RO) membrane surfaces have been well researched.2-5 In brackish water desalination, scaling and fouling of RO membranes adversely affect the product water quality and output, energy consumption, and membrane cleaning and replacement costs. In geothermal energy utilization and cooling water systems, the fouling of equipment surfaces due to silica scaling remains one of the key problems to be solved.6, 7 Effective silica scaling control in industrial water systems necessitates an understanding of the various forms of silica present in water. The silica solubility in water systems is dependent upon several factors including pH, temperature, other ions present, and the silica form present. Figure 1 presents silica solubility as a function of solution pH. As shown, silica solubility increases with pH varying from 120 ppm at pH 6 to 140 ppm at pH 9 and increases rapidly as pH is increased from 9.5 to 10.5.8 Unlike certain silicate salts such as magnesium silicate that becomes less soluble as the temperature increases, silica solubility increases with increasing

2/20

solution temperature8 as illustrated in Figure 2. In addition, the presence of various electrolytes also influences the solubility of amorphous silica in aqueous solution.9 The potential for silica scaling increases when the dissolved silica levels in an aqueous system exceed the amorphous silica solubility limit (in the range of 120 to 150 mg/L at ambient temperature). Exceeding this saturation level in cold water (<10°C) is not as serious a problem as silica polymerization is a very slow process at lower temperatures. However, silica in excess of 180 mg/L presents a potential problem at any temperature.10 The mechanism of silica polymerization is very complex and is believed to occur via the base catalyzed reaction shown below:

Silica Polymerization Process

Si(OH)4

+ HO- ⇔ (HO)3Si -- O- + Other silicates

(HO)3Si -- O-

+

Si(OH)4

→

OH OH I I

HO – Si -- O --Si – OH I I

OH OH

→

Colloidal silica

Silica polymerization may cause to colloidal silica formation and precipitation in water systems thereby reducing heat transfer and fluid flow through equipment such as heat exchanger tubes and membranes. In addition, polyvalent metal ions (i.e., iron, aluminum, calcium, magnesium, etc.) present in make up water supplies can absorb or complex silica and catalyze the precipitation. Corroded steel pipes and heat exchangers are very prone to silica scaling. The composition and quantity of silica deposition as well as the rate at which it forms is dependent on pH, temperature, the ratio of calcium to magnesium, and the concentration of polyvalent ions in water. Furthermore, silica and silicate deposits are particularly difficult to remove once they form necessitating the use of strong chemical cleaning (based on hydrofluoric acid) or laborious mechanical removal methods. The development and application of silica control technology in industrial water systems has been well documented.11-13 Several approaches to control silica solubility have been developed to prevent silica deposition. One of the simplest methods involves keeping constituents such as silica and magnesium below the critical concentration levels necessary for precipitation. Minimizing silica deposition in industrial water systems operating in the pH 6 to 9 range requires keeping the silica concentration below 150 mg/L. In addition, if magnesium is present, the product of magnesium (as CaCO3) and silica (as SiO2) should not be allowed to exceed 20,000 mg/L.14 During the last three decades, several chemical treatment programs have been developed which allow significantly higher (than discussed earlier herein) of silica and/or magnesium levels to be maintained in re-circulating water. Most of these programs incorporate polymeric dispersants to keep silica particles dispersed and thereby avoid deposition on heat exchanger surfaces. In water treatment programs, polymers that have shown good dispersancy activity are normally anionic and have molecular weights (MW) less than 20,000 Daltons (DA). The

3/20

ideal silica deposit control agent should be able to disperse both amorphous silica and magnesium silicate as well as effectively inhibit silica polymerization. Amjad and Yorke4 in their evaluation of polymers reported that cationic-based copolymers are effective silica polymerization inhibitors. Similar conclusions were also reached by Harrar et al.5 when investigating the use of cationic polymers and surfactants in inhibiting silica polymerization under geothermal conditions. Although these cationic-based homo- and copolymers showed excellent performance in terms of silica polymerization, they offered poor silica/silica dispersancy activity. Gallup and Barcelon6 investigated the performance of a variety of organic inhibitors as alternatives to strong acids for geothermal application. Results of their study reveal that brine acidification always out-performed organic inhibitors. The performance of a formulated product containing hydroxyl phosphono acetic acid and a copolymer of acrylic acid and allyl hydroxyl propyl sulfonate ether in high hardness water containing high alkalinity and 225 mg/L silica, has been investigated.15 The inspection of heat exchangers showed essentially no deposits in the presence of formulated product compared to heavy silica and silicate deposits in the control (no treatment). Smith1 tested a number of polymers as silica inhibitors for RO application using a pilot RO unit. The results of Smith’s study reveal that a proprietary polymer exhibits excellent silica inhibitory property. Demadis and Neofotistou16 investigated the performance of several dendrimers of the polyaminoamide family as silica inhibitors for cooling water applications and observed that inhibitor performance strongly depends on the branching present in the dendrimer. As discussed above, silica scaling is a complex problem involving many processes such as polymerization, silica dispersion, precipitation of metal silicate scale, and co-precipitation of silica with commonly encountered mineral scales. Over the years, various polymers have been shown to be capable of dispersing fine particles of amorphous silica once they have formed.17-19 Polymeric dispersants are often used when the potential for particulate silica fouling exists. Although these polymeric dispersants may minimize the impact of fouling, they do not address the root problem of controlling silica polymerization. This paper and results presented herein: • Support our continued efforts to find a polymeric additive that effectively inhibits silica

polymerization as well as provides particulate dispersing properties that minimizes the deposition of unwanted silica-based scales.

• Show the impact of solution temperature and iron (III) on silica polymerization and precipitation.

• Compare the performance of a new developmental silica inhibitor with several commercially available silica inhibitors and dispersants.

• Characterize the composition of silica precipitates formed during the experiments.

Experimental Reagent grade chemicals and distilled water were used in this study in accordance with Lubrizol’s “Silica Polymerization Inhibition Test Procedure,”20 and schematically shown below and discussed in the text that follows.

4/20

Silica Polymerization Inhibition Experimental Protocol

Distilled Water

+ Sodium Silicate + Inhibitor Solution

→

Adjust to pH 7.0

→

+ Calcium and Magnesium

Solution

→

Adjust to pH 7.0

→ Sample and Filter through 0.22 micron

→ Analyze for SiO2

Silicate stock solutions were prepared from sodium metasilicate, standardized spectrophotometrically, and were stored in polyethylene bottles. Calcium chloride and magnesium chloride stock solutions were prepared from calcium chloride dihydrate and magnesium chloride hexahydrate and were standardized by EDTA titration. The results herein are reported on a 100% active inhibitor basis. Silica polymerization experiments were performed in polyethylene containers placed in a double-walled glass cell maintained at a required temperature (40, 55, or 68°C). The supersaturated solutions were prepared by adding a known volumes of sodium silicate (expressed as SiO2) solution and water in a polyethylene container. After allowing the temperature to equilibrate, the silicate solution was quickly adjusted to pH 7.0 using the dilute hydrochloric acid. The pH of solution was monitored using Brinkmann/Metrohm pH meter equipped with a combination electrode. The electrode was calibrated before each experiment with standard buffers. After pH adjustment, a known volume of calcium chloride and magnesium chloride stock solution was added to the silicate solution. The supersaturated silicate solutions were re-adjusted to pH 7.0 with dilute HCl and/or NaOH and maintained constant throughout the silica polymerization experiments. Experiments involving inhibitors were performed by adding inhibitor solutions to the silicate solutions before adding the calcium chloride and magnesium chloride solutions. Figure 3 shows the experimental set-up. The reaction container is capped and kept at constant temperature and pH during the experiment. Silicate polymerization in these supersaturated solutions was monitored by analyzing the aliquots of the filtrate from 0.22-µm filter paper for soluble silicate using the standard colorimetric method.4 The concentration of calcium and magnesium in the experimental solutions were monitored by EDTA titration method. The polymer performance for silicate polymerization inhibition was calculated according to the following equation: [SiO2] sample --- [SiO2] blank

SI = ------------------------------------------- x 100% [SiO2] initial --- [SiO2] blank Where:

SI = Silica Inhibition (%) or %SI [SiO2] sample = Silica concentration in the presence of inhibitor at 22 hr

[SiO2] blank = Silica concentration in the absence of inhibitor at 22 hr [SiO2] initial = Silica concentration at the beginning of experiment

5/20

Table 1 summarizes the description or compositions and acronyms of the polymeric and non-polymeric additives evaluated in the present study including a new development silica inhibitor (NSI).

Results and Discussion Silica Supersaturation Using the experimental set-up described above, a series of experiments were conducted at various silica supersaturation (SS) conditions. Figure 4 shows profiles of soluble silica concentration as a function of time for various silica SS solutions. It is evident from Figure 4, that silica polymerization depends on the silica concentration present in the supersaturated solutions. For highly supersaturated solutions containing 550 mg/L silica, polymerization starts immediately (as evident from the sharp decrease in soluble silica concentration with time) whereas lower silica concentration solutions polymerizes (react) at a much slower rate. At low degrees of silica supersaturation, a decrease in silica concentration is preceded by a slow polymerization reaction (induction time or β). During the induction time, the concentration of soluble silica does not change significantly. However, once polymerization begins, soluble silica begins to decrease. Figure 4 illustrates that the β values for various SS solutions are <8 hr for 450 mg/L, and >20 hr for 300 mg/L silica compared to <5 minutes for the high SS (550 mg/L silica) solutions. For the results presented in this paper, we chose 550 mg/L silica as the SS condition to evaluate the performance of various inhibitors. Inhibitor Concentration Figure 5 shows silica concentration as a function of time and P-8 (Carbosperse™ K-XP212 polymer) dosage. The data in Figure 5 indicate that dosage strongly affects the ability of K-XP212 to inhibit silica polymerization and that 50 ppm and 75 ppm dosages (as active polymer) provide similar performance.

Figure 6 examines the 22 hr data presented in Figure 5 (with the values translated from “silica conc.” to “% silica inhibition” as well as data for 175 and 350 ppm polymer dosages not shown in Figure 5). As shown in Figure 6 and summarized below, the inhibitory effect of P8 increases dramatically as dosage increases to 50 ppm and incrementally improves thereafter: P8 Dosage: 15 ppm 25 ppm 50 ppm >75 ppm Silica Inhibition: 14% 52% 84% >90% Performance: Poor Fair Excellent Excellent

Silica Inhibition Performance of Non-Polymeric and Polymeric Additives The use of non-polymeric inhibitors, such as polyphosphates and phosphonates, to control scaling (especially calcium carbonate) is well known. However, under stressed conditions (i.e., high pH, high temperature, high hardness, etc.), these phosphorus-containing inhibitors frequently react stoichiometrically with calcium ions leading to calcium-polyphosphate/phosphonate precipitation.21 The application of boric acid (BA) and/or its water soluble salts to prevent silica polymerization has been reported.22 It has been suggested that

6/20

silica inhibition by borate is perhaps due to the formation of more soluble borate-silicate complexes. To understand the importance of non-polymeric and polymeric additives as silica polymerization inhibitors, a series of experiments were conducted in the presence of 25 and 350 ppm inhibitor dosages and the results appear in Figure 7. For the non-polymeric additives (i.e., A = HEDP, B = PBTC, C= BA) tested, the data in Figure 7 indicate these materials are ineffective silica inhibitors. Although HEDP and PBTC are excellent calcium carbonate inhibitors,23 their poor performance as silica polymerization inhibitors suggests that phosphonate groups do not inhibit silica polymerization in aqueous solutions. In recent years, many studies have examined the influence of polymer architecture on the precipitation of scale-forming salts. Studies have shown that polymer performance as scale inhibitors in industrial water systems is strongly affected by polymer MW, composition (monomer types and weight ratios), and polymerization solvent. It has been reported21 that solvent polymerized polyacrylates are more tolerant to calcium ions than the water polymerized polyacrylates. For carboxylic acid containing polymers, it appears that precipitation inhibition is greatest for polymers whose MW is below 20,000 DA. However, the optimum polymer MW depends upon polymer composition and the salts being inhibited. For calcium phosphate and calcium phosphonate inhibition, acrylic acid and/or maleic acid-based co-polymers have been shown to perform better than homo-polymers of acrylic acid and maleic acid. In the case of particulate matter (i.e., clay, iron oxide) dispersion, polymers that exhibit good dispersion are typically low (<10,000 DA) MW and contain both carboxylic acid and sulfonic acid groups Figure 7 presents performance data for polymeric inhibitors containing different functional groups (e.g., carboxylic acid, sulfonic acid). It is evident that the homopolymers (P1 and P2), copolymer (P3), and terpolymers (P4, P5, P6, P7) commonly used in water treatment formulations as deposit control agents to control mineral scales and disperse suspended matter are poor (<10% inhibition) silica inhibitors. The performance data for these commonly used polymers with the exception of a copolymer blend (P8) and new developmental silica inhibitor (P9) clearly show that carboxylic acid, sulfonic acid, ester, and non-ionic groups present in the polymers exhibit poor interaction with silane groups present in silica. New Silica Polymerization Inhibitor Performance As previously discussed, a variety of additives have been developed for silica control with commercial limited success; perhaps due to limited efficiency under the field conditions or because of cost/benefit considerations. Therefore, water technologists are still seeking a cost effective silica polymerization inhibitor that minimizes the potential for silica scaling in industrial water systems. Figure 8 presents silica inhibition vs. dosage for three (3) distinctively different polymeric inhibitors. P4 is a patented terpolymer containing carboxylic acid and sulfonic acid groups designed for use in controlling calcium phosphate/phosphonate salts, metal ion stabilization, and particulate dispersion. P4 is similar to polymeric inhibitors P1 to P7 in that its composition is dominated (>50%) by carboxylic monomer groups. On the other hand, P8 and P9 are polymeric inhibitors (patented and patent pending, respectively) who compositions are distinctly different (i.e., P8 & P9 contain <50% carboxylic acid monomer groups) from the other polymeric inhibitor. The data in Figure 8 as well as those in Figure 7 indicate the following:

7/20

Additive Silica Polymerization Inhibition Performance

P4 P4 (as well as P1 to P7) provides poor (<20%) performance at all dosages tested up to 350 ppm (only up to 50 ppm dosage shown in Figure 8).

P8 P8 performance increases dramatically from poor to excellent with dosage up to 35 ppm and improves incrementally thereafter.

P9 P9 performance increases dramatically as dosage increases from 2.5 ppm (5% SI) to 10 ppm (79% SI) and improves incrementally thereafter.

At 15 ppm dosages, P8 provides >85% SI compared to <10% SI values for both P8 and P4. The absolute performance and dosage response of P8 is superior to the other additives tested except P9 which is dramatically better than all other materials tested. Figure 9 takes a closer look at a subset (7.5 to 50 ppm dosages) of the silica polymerization inhibition vs. dosage data for P8 and P9 (NSI which is a new developmental product) and reinforces earlier observations herein that P9 is a significant improvement in silica inhibitor technology.

Effect of Iron (III) The presence of iron in re-circulating waters and/or brine in reverse osmosis system, whether originating from the feed water or as a result of iron-based metal corrosion in the system or the carry-over from a clarifier using iron-based flocculating agents, can have profound effect on the performance of polymers used as scale control and dispersant. It has been previously reported that polymer performance as a calcium phosphate inhibitor and iron oxide dispersant is negatively impacted by iron (III).24 Figure 10 presents the impact 0.5 and 1.0 ppm soluble iron on the silica polymerization inhibition performance for 25 ppm dosages of P8 and P9. As shown, P9 retains >65% SI performance in the presence of 1.0 ppm. The antagonistic effect shown by Fe(III) may be attributed to the formation Fe(OH)3 – silica complex. Thus, if iron (III) is encountered in cooling waters, reverse osmosis systems, or geothermal systems, a polymer that more effectively prevents silica scaling, especially in the presence of soluble iron should be considered in developing high performance water treatment formulations.

Effect of Solution Temperature It is well documented that solubility of scale forming salts such as calcium carbonate, calcium sulfate, and calcium phosphates is inversely dependent on solution temperature.25 This solubility-temperature relationship suggests that the scaling tendency will be higher at the heat exchanger surfaces than in other parts of the re-circulating water systems. Amjad26 reported that the calcium ion tolerance of polymers decreases markedly as solution temperature increases from 25 to 70°C. However, the situation is very different for silica supersaturated system in the sense that silica solubility increases with increasing temperature thus resulting in lower supersaturation. In order to study the influence of solution temperature on polymer performance, a series of silica polymerization experiments were conducted at similar initial silica concentrations, pH 7.0, and several temperatures (i.e., 40, 55, and 68°C). Figure 11 presents polymer dosage needed to achieve >85% inhibition as a function of temperature for P8 and P9. It is evident that

8/20

solution temperature significantly influences silicate polymerization. There are two competing factors contributing to temperature influence: (a) silica supersaturation decreases with increasing temperature due to increased silica solubility and (b) rate of silica polymerization increases with increasing temperature. Figure 12 translates the inhibitor dosage vs. calculated supersaturation data for P8 and P9 based on the water chemistry used and data shown in Figure 11. Figure 12 provides a basis for estimating P8 and P9 dosages required to inhibit silica polymerization as a function of inhibitor dosage. Additional work is underway to develop predictive models for P9 dosages vs. silica supersaturation for different water chemistries. Effect of Calcium and Magnesium on Silica Polymerization It is generally known that the presence of metal ions both the rate of precipitation and affects the crystal morphology of scale forming salts. It is also reported that metal ions may form insoluble salts with silicate ions in aqueous solution. To understand the role of metal ions (i.e., Ca, Mg) in silica supersaturated solution a series of experiments were conducted wherein calcium, and magnesium, and silica concentrations were monitored as a function of time during the silica polymerization experiments. Figure 13 presents typical concentration vs. time profiles for 550 mg/L silica in the presence of 0 and 10 ppm of P9 (a new developmental silica inhibitor). As shown, the initial calcium (200 mg/L) and magnesium (120 mg/L) concentrations remain essentially constant (within the experimental error or ±3%) during the silica polymerization experiments while silica concentration decreases with time. Although not presented herein, similar results for calcium and magnesium concentrations were also observed with higher silica SS solutions and higher calcium (280 mg/L) and magnesium (168 mg/L) concentrations. The data presented in Figure 13 suggest that neither Ca nor Mg has precipitated out of the solutions in the presence and absence of P9. The silica precipitates formed during the silica polymerization experiments were studied using Electromagnetic Dispersion Spectroscopy (EDS). The EDS spectrum of the precipitates formed in the presence of a low P9 dosage is shown in Figure 14. From the spectrum, it is apparent that the precipitate consists essentially of silicon and oxygen with only trace amounts of calcium and magnesium present in the filtered solid. This observation was confirmed by analyzing calcium and magnesium ions before and after filtration wherein there was no significant concentration difference. The trace levels of Ca and Mg shown in the EDS spectrum may be due to surface adsorption of Ca and Mg on the un-washed precipitated silica.

Suspended Matter Dispersion Suspended matter, especially iron oxide and hydroxides, if present in recirculating waters are known to increase the demand for treatment chemicals and/or increase the complexity of preventing the fouling of heat transfer and other equipment surfaces. For this reason, the use of a product exhibiting excellent dispersancy activity is necessary to disperse particulate matter (e.g., iron oxide particles). In the case of silica control, an effective product must both inhibit silica polymerization and disperse particulates (that are present or form in the system). Figure 15 shows the iron oxide dispersancy activity of several polymers as determined by a Lubrizol standard test method (conditions include synthetic water, pH 7.8, 200 ppm iron oxide,

9/20

1 ppm inhibitor).21 Figure 15 indicates that P1 and P6 perform poorly as iron oxide dispersants whereas the P3 and P7 provide good and fair as iron oxide dispersion, respectively. The other inhibitors (all copolymers) inhibitors tested (i.e., P4, P5, P8, and P9) exhibit excellent (>80%) iron oxide dispersion performance. Thus, both P8 and P9 compare very favorably as particulate dispersants with the other high performance inhibitors tested. NSI Use Considerations The information available for considering NSI (P9) as a high performance silica control agent in water treatment applications include: • P9 dosage rates as a silica control agent are expected to vary based upon operating

conditions and water chemistry. Dosage requirements decrease with increasing temperature (in the range of 20 to 90ºC) due to silica solubility variability as a function of system temperature. As with any inhibitor, the presence of antagonistic materials may increase dosage requirements and/or adversely impact performance.

• Lubrizol’s laboratory testing indicates that P9 is an excellent iron oxide dispersant comparable to Carbosperse K-781 and K-798 acrylate terpolymers.

• P9 is suitable for use as an overlay to existing water treatment programs where silica control is a primary concern.

• Lubrizol has not yet completed aquatic toxicity testing for P9. However, compared to similar chemical structures, environmental experts believe that significant biodegradability over 28 days is questionable and acute toxicity should be slight to moderate for aquatic organisms. In addition, single-dose oral toxicity evaluations on a material similar to P9 were conducted with Sprague-Dawley rats and New Zealand White rabbits. Under the conditions of the testing, the acute oral LD50 of the test material was estimated to be greater than 500mg/kg in the rat and the acute dermal LD50 of the test material was estimated to be greater than 1,000 mg/kg in the rabbit.

• P9 (or its components) is (are) listed on or is (are) exempt from the U.S.A. Toxic Substance Control Act Chemical Substance Inventory.

• P9 is a partially neutralized low molecular weight water soluble copolymer expected to be supplied as a water white to amber liquid. The typical properties are not finalized but will be in the range of 35 to 45% total solids, >90% active solids, pH 5 to 8, Brookfield viscosity <1,000 cP (25°C). The anticipated packaging of P9 includes 55-gal plastic drums, intermediate bulk containers, and bulk quantities.

• Based on limited evaluations, the current guidelines for incorporating P9 as a component of water treatment formulation blends are as follows: - P9 may be blended (up to 50/50 solids weight basis) with anionic dispersants such as

Carbosperse K-700 homopolymers, acrylate copolymers, and acrylate terpolymers. - The formulating window for P9 with other water treatment additives (e.g., phosphonates,

azoles, molybdates, inorganic neutralization agents [NaOH or KOH], or strong acids [H2SO4, HCl]) is limited. Therefore, we recommend limiting formulations to less than 10% solid solids. Alternatively, if formulations exceed 10% total solids and/or are greater than pH 4.5 use coupling agents and/or organic neutralizing agents.

10/20

Future Investigations Investigations currently underway or contemplated pertaining to P9 are as follows: • Aquatic toxicity testing. • Thermal stability testing. • Magnesium silicate scale control. • Expanding formulating window knowledge base. • Field testing. • Developing predictive models for polymer dosage requirements vs. silica SS for various

water chemistries.

Summary The silica polymerization process that may lead to silica scaling in industrial water systems is complex and primarily governed by system operating conditions (e.g., pH, temperature, impurities). Unfortunately, unlike other scales (i.e., calcium carbonate, calcium phosphate) silica scaling is not easily controlled by simply adjusting pH of the re-circulating waters. Operating cooling water and RO systems under alkaline pH may lead to the formation of undesirable magnesium silicate scale. The proprietary inhibitors and/or formulated products developed and touted for silica control over the last three decades have achieved limited commercial success primarily due to either inhibitor inefficiency under the system operating conditions and/or poor cost/benefit ratios. The data presented in this paper suggest that deposit control polymers (homo-, co, and ter-polymers) commonly used as inhibitors for mineral scales (i.e., calcium carbonate, calcium sulfate, calcium phosphate, and barium sulfate) and dispersants for suspended matter perform poorly as silica polymerization inhibitors. Non-polymeric additives (i.e., HEDP, PBTC, boric acid) are also ineffective for controlling silica scaling. The performance of K-XP212 as silica polymerization inhibitor is superior to other commercial products promoted for silica control. A new developmental copolymer P9 (NSI) has shown significant improvement over existing inhibitor technology especially at lower dosages. Both P8 and P9 provide particulate matter (iron oxide) dispersion that is comparable to high performance terpolymers. Collectively, the data herein indicate that P9 is a significant improvement in silica inhibitor technology.

Acknowledgements The authors would like to thank Valerie Woodward for spectroscopic work and Steve Zamborsky for technical assistance in the experimental work. Thanks to Lubrizol for supporting the research and allowing us to present the findings at the Association of Water Technologies Annual Convention.

References 1. C. W. Smith, “Pilot Test Results Utilizing Polymeric Dispersants for Control of Silica,” Water

Soluble Polymers: Solution Properties and Application,” Z. Amjad (ed.), Plenum Press, New York, New York (1998).

2. Z. Amjad, “Latest developments in Antifouling Chemicals for RO Pretreatment Processes,”

Proceedings of Fouling Mitigation in Membrane Processes, Haifa, Israel (1999).

11/20

3. R. Sheikholeslami and S. Tan, “Effects of Water Quality on Silica Fouling of Desalination Plants,”

Desalination 126, 267-280 (1999). 4. Z. Amjad and M. Yorke, “Carboxylic Functional Polyampholytes as Silica Polymerization

Inhibitors,” U.S. Patent No. 4,510,059 (1985). 5. J. E. Harrar, L. E. Lorensen, and F. E. Locke, “Method for Inhibiting Silica Precipitation and

Scaling in Geothermal Flow Systems,” U.S. Patent No. 4,328,106 (1982). 6. D. L. Gallup and E. Barcelon, “Investigations of Organic Inhibitors for Silica Scale Control from

Geothermal Brines-II,” Geothermics 34, 756-771 (2005). 7. R. Leif, W. Bourcier, and C. Bruton, “Silica Scale Inhibition: Effects of Organic Additives on

Polymerization,” paper presented at the 2000 Geothermal Resource Council’s Annual Meeting, San Francisco, CA (2000).

8. R. K. Iler, The Chemistry of Silica, John Wiley and Sons, New York, NY, (1979). 9. S. H. Chan, “A Review on Solubility and Polymerization of Silica,” Geothermics, 18 49-56 (1989). 10. Bradley R., “Design Considerations for Reverse Osmosis Systems” in Reverse Osmosis:

Membrane Technology, Water Chemistry, and Industrial Applications, Z. Amjad (ed.), Van Nostrand Reinhold, (1993).

11. Z. Amjad and R. W. Zuhl, “An Evaluation of Silica Scale Control Additive for Industrial Water

System,” Paper No. 08368, CORROSION/2008, NACE International, Houston, TX (2008). 12. W. S. Carey, A. Solov, and L. Perez, “Method for Inhibiting Silicate Deposition,” U.S. Patent No.

5,422,010 (1995). 13. J. S. Gill, “Silica Scale Control,” Paper No. 226, CORROSION/98, NACE International, Houston,

TX (1998). 14. W. F. Hann, S. T. Robertson, and J. H. Bardsley, “Recent Experience in Controlling Silica and

Magnesium Silicate Deposits with Polymeric Dispersants,” Paper No. 93-59, International Water Conference, Pittsburgh, PA (1993).

15. L. A. Perez, J. M. Brown, and K. T. Nguyen, “Method for Controlling Silica and Water Soluble

Silicate Deposition,” U.S. Patent No. 5,256,302 (1993). 16. K. D. Demadis, A. Stathoulopoulou, and A. Ketsetzi, “Inhibition and Control of Colloidal Silica: Can

Chemical Additives Untie the Gordian Knot of Scale Inhibition?” Paper No. 07058, CORROSION/2007, NACE International, Houston, TX (2007).

17. J. S. Gill, “Inhibition of Silica - Silicate Deposit in Industrial Waters,” Colloids and Surfaces, 74,

101-106 (1993). 18. P. P. Nicholas and Z. Amjad, “Method for Inhibiting and Deposition of Silica and Silicate

Compounds in Water Systems,” U.S. Patent No. 5,658,465 (1997).

12/20

19. P. R. Young, C. M. Stuart, P. M. Eastin, and M. McCormick, “Silica Stabilization in Industrial Cooling Towers: Recent Experiences and Advances,” Technical Paper No. TP93-11, Cooling Tower Institute 1993 Annual Meeting, Houston, TX (1993).

20. “Silica Polymerization Inhibition Test Procedure,” Lubrizol technical bulletin Silica-PITP, Oct-2007. 21. Z. Amjad and R. W. Zuhl, “Factors Influencing the Precipitation of Calcium-Inhibitor Salts in

Industrial Water Systems,” Proceedings of Annual Convention, Association of Water Technologies, Phoenix, AZ (2003).

22. D. A. Meier and L. Dubin, “A Novel Approach to Silica Scale Inhibition,” Paper No. 334,

CORROSION/87, NACE International, Houston, TX (1987). 23. Z. Amjad and R. W. Zuhl, “Kinetic and Morphological Investigation on the Precipitation of Calcium

Carbonate in the Presence of Inhibitors,” Paper No. 06385, CORROSION/2006, NACE International, Houston, TX (2006).

24. Z. Amjad, J. F. Zibrida, and R. W. Zuhl, “Performance of Polymers in Industrial Water Systems:

The Influence of Process Variables”, Materials Performance, 36 (1) 32 1997). 25. J. C. Cowan and D. J. Weintritt (Eds), “Water Formed Scale Deposits,” Gulf Publishing, Houston,

TX (1976). 26. Z. Amjad, “Interactions of Hardness Ions with Polymeric Scale Inhibitors in Aqueous Systems,”

Tenside Surfactants Detergents, 42, 71-77 (2005). 27. E. Neofotisou and K. D. Demadis, “Use of Antiscalants for Mitigation of Silica Fouling and

Deposition: Fundamentals and Applications in Desalination Systems,” Desalination, 167, 257-272 (2004).

Table 1: Polymeric and Non-Polymeric Additives Evaluated Additive Composition Acronym

HEDP 1-hydroxythylidene-1,1-diphosphonic acid A PBTC 2-phosphonobutane 1,2 4-tricarboxylic acid B BA Boric acid C K-732* Poly(acrylic acid) P1 P-MA Poly(maleic acid) P2 K-775* Poly(acrylic acid: 2-acrylamido-2-methylpropane sulfonic acid) P3 K-798* Poly(acrylic acid: 2-acrylamido-2-methylpropane sulfonic acid:

sulfonated styrene) P4

Competitive-1 Poly(acrylic acid: 2-acrylamido-2-methylpropane sulfonic acid: non-ionic)

P5

Competitive-2 Poly(maleic acid:ethylacrylate:vinyl acetate) P6 Competitive-3 Proprietary acrylic copolymer P7 K-XP212* Proprietary copolymer blend P8 NSI* New proprietary copolymer P9

*Carbosperse™ K-700 polymer supplied by Lubrizol Advanced Materials, Inc.

13/20

Figure 1: Silica Solubility as a Function of pH

0100200300400500600700800900

4 5 6 7 8 9 10 11

pH

Silic

a C

onc.

(mg/

L)

Figure 2: Silica Solubilityas a Function of Temperature

0

50

100

150

200

250

300

350

0 10 20 30 40 50 60 70 80

Temperature (deg C)

Silic

a C

onc.

(mg/

L)

14/20

Figure 3: Set-Up for Silica Polymerization Inhibition Experiments

Figure 4: Silica Polymerization vs. Time as a Function of Initial Silica Concentration

0

100

200

300

400

500

600

0 5 10 15 20 25Time (hr)

Silic

a C

onc.

(mg/

L)

550 mg/L 450 mg/L 320 mg/L 150 mg/L

(Variable silica, 200 mg/L Ca, 120 mg/L Mg, pH 7.0, 40ºC)

Equilibrium SolubilityEquilibrium Solubility

ThermostattedWater In

Thermostatted

Water Out

SamplingPort0.22 μ m filter

Sample

Filtrate Retentate

Soluble SilicaAnalysis

ParticleCharacterization

pH Meter

7. 00

Magnetic Stirrer

15/20

Figure 5: Silica Polymerization vs. Time as a Function of P8 Dosage (ppm)

0

100

200

300

400

500

600

0 5 10 15 20 25Time (hr)

Silic

a C

onc.

(mg/

L)

0 ppm 15 ppm 25 ppm 50 ppm 75 ppm

(600 mg/L silica, 200 mg/L Ca, 120 mg/L Mg, pH 7.0, 40ºC)

Figure 6: Silica Polymerization Inhibition @ 22 hras a Function of P8 Dosage

0

20

40

60

80

100

15 25 50 75 175 350

P8 Dosage (ppm as active)

% S

ilica

Inhi

bitio

n

16/20

Figure 7: Silica Polymerization Inhibitionas a Function of Additive Dosage

for Non-Polymeric & Polymeric Additives

0

20

40

60

80

100

A B C P1 P2 P3 P4 P5 P6 P7 P8 P9

% S

ilica

Inhi

bitio

n

25 ppm 350 ppm

Figure 8: Silica Polymerization Inhibition for Select Copolymers vs. Dosage

0

20

40

60

80

100

0 10 20 30 40 50

Polymer Dosage (ppm)

% S

ilica

Inhi

bitio

n

P4

P8

P9

17/20

Figure 9: Silica Polymerization Inhibitionfor P8 and P9 vs. Polymer Dosage

0

20

40

60

80

100

P8 P9

% S

ilica

Inhi

bitio

n

7.5 ppm12.5 ppm25 ppm50 ppm

Figure 10: Effective of Soluble Ironon Silica Polymerization Inhibition

(25 ppm Polymer Dosages)

0

20

40

60

80

100

0 ppm Fe (III) 0.5 ppm Fe(III) 1.0 ppm Fe (III)

Soluble Iron Dosage

% S

ilica

Inhi

bitio

n

P8 P9

18/20

Figure 11: Effective of Temperature on Polymer Dosage Required to Achieve 85% Silica Polymerization Inhibition

0

10

20

30

40

50

40°C 55°C 68°CSolution Temperature

Poly

mer

Dos

age

(ppm

) P8 P9

Figure 12: Polymer Dosage Required to Achieve >85% Silica Polymerization Inhibition as a Function of SS

0

10

20

30

40

50

60

1.7 2.14 3.08

Silica Supersaturation

Poly

mer

Dos

age

(ppm

)

P8 P9

19/20

Figure 13: Silica Polymerization Impact on [Ca] and [Mg]vs. Time in the Absence & Presence of 10 ppm P9

0

100

200

300

400

500

600

0 5 10 15 20 25Time (hr)

[Sili

ca],

[Ca]

, [M

g] a

s m

g/L

Silica (mg/L) - w/o inhibitor Silica (mg/L) - w/ inhibitorCa (mg/L) Ca (mg/L)Mg (mg/L) Mg (mg/L)

Figure 14: Electromagnetic Dispersion Spectrum of Silica Precipitate

20/20

Figure 15: Iron Oxide Dispersion by Polymers(200 mg/L Iron Oxide, 1 ppm Polymer, 3 hr, pH 7.8, 100 mg/L Ca,30 mg/L Mg, 671 mg/L Sulfate, 60 mg/L Bicarbonate, 23 deg C)

0

20

40

60

80

100

P1 P3 P4 P5 P6 P7 P8 P9

% Ir

on O

xide

Dis

pers

ed

**********************************************************************************************************

Lubrizol Advanced Materials, Inc. * Cleveland, OH 44141-3247, U.S.A. Phone: 1-800-380-5397 or 216-447-5000 FAX: 216-447-6315 (USA Customer Service) 216-447-6144 (International Customer Service) 216-447-5238 (Marketing & Technical Service)

http://www.carbosperse.com

Nov-2008 The information contained herein is believed to be reliable, but no representations, guarantees or warranties of any kind are made to its accuracy, suitability for particular applications, or the results to be obtained therefrom. The information is based on laboratory work with small-scale equipment and does not necessarily indicate end product performance. Because of the variations in methods, conditions and equipment used commercially in processing these materials, no warranties or guarantees are made as to the suitability of the products for the application disclosed. Full-scale testing and field application performances are the responsibility of the user. LUBRIZOL ADVANCED MATERIALS, INC. shall not be liable for and the customer assumes all risk and liability of any use or handling or any material beyond LUBRIZOL’s direct control. The SELLER MAKES NO WARRANTIES, EXPRESS OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANT ABILITY AND FITNESS FOR A PARTICULAR PURPOSE. Nothing contained herein is to be considered as permission, recommendation, nor as an inducement to practice any patented invention without permission of the patent owner.

Lubrizol Advanced Materials, Inc. 9911 Brecksville Road, Cleveland, OH 44141, P/216-447-5000, F/216-447-5238

© 2008, The Lubrizol Corporation. All rights reserved. ™ Trademark of The Lubrizol Corporation.