CORROSION PREVENTION WITH SODIUM SILICATE1

Nausha Asrar, Anees U. Malik, Shahreer Ahmed Research And Development Center Saline Water Conversion Corporation P. O. Box # 8328, Al-Jubail 31951, Kingdom of Saudi Arabia

SUMMARY Corrosion of steel structures is a common problem in everyday life. Use of carbon steel or mild steel as a construction material is taken as an onus despite its poor corrosion resistance properties at ambient temperature. This is because of its abundance, easy availability, low cost, good formability and workability and remarkable strength. The corrosion in steel structures is generally controlled by 3 methods, viz., cathodic protection, coating or use of an inhibitor. For pipe lines, storage tanks or similar structures, silicate or silicate based materials have been used as inhibitor for more than 70 years. Silicate in water provides protection to iron and steel by prohibiting its anodic dissolution. In this project, tests were carried out to study the corrosion prevention properties of Sodium silicate supplied by Adwan Chemical Industries. Immersion and electrochemical tests of steels in tap water with and without addition of Sodium Silicate have been carried out under static (stagnant) and dynamic (flowing) conditions. In order to simulate the in situ condition, tests were also carried out in a corrosion test loop. These tests were carried out in the corrosion laboratory using polished carbon steel coupons. Corrosion rates were determined by weight loss method. Polished coupons were immersed in static and dynamic tap water containing different amounts of sodium silicate ranging between 3 - 200 ppm. Results obtained show positive effect of sodium silicate in controlling the corrosion of carbon steel immersed in tap water. It has also been observed that sodium silicate is more effective in controlling the corrosion of carbon steel in dynamic water than in
1

Issued as Technical Report N0.TR3804/EVP95013 in July 1998.

1856

static water. However, in dynamic tap water, saturated with dissolved oxygen, optimum concentration required to provide maximum corrosion inhibition, appears to be quite high (> 100 ppm). As the inhibitive effect of Sodium Silicate depends upon dissolved oxygen, its dosing rate should always be optimized in the system to be protected.

1.

INTRODUCTION

1.1

Background

In July 1995, ADWAN Chemical Industries, Riyadh approached Department of Research and Development for testing their product, Sodium Silicate, in prevention of corrosion in steel pipe lines. The R&D Center Al-Jubail was asked to asses the feasibility of such testing and if it is in affirmative then to submit an Evaluation Project Proposal to the Department of R&D, Riyadh.

Subsequently a project proposal was submitted to Riyadh and an agreement was signed between SWCC and ADWAN Chemical Industries Co. Ltd. on March 26, 1996 (7.11.1416H) to carry out the testing of Sodium Silicate at R&D Center, AL-Jubail.

1.2

Sodium Silicate as a Corrosion Inhibitor

Sodium silicates are the most commonly used commercially available alkali silicates. These materials are usually manufactured as glasses and if dissolved in water, form viscous alkaline solutions. The soluble silicates, called water glass, have been used in cements, coatings, water treatment and ore benefication. These applications are based on the ability of silicates to form gels or to react with multivalent metal ions in solutions or oxide surface [Encyclopedia, 1988]. Sodium silicate has been used for the inhibition of corrosion of steel for more than 70 years. The general principle lies in feeding dissolved silicate to the metal water pipe where a protective film is formed on the inner surface of the pipe. Sodium silicate has been used for prevention of corrosion in water heating pipelines, centralized heat supply

1857

system, steam boilers, industrial plant circulated municipal water, water transmission steel pipelines, etc., [Akolzin, A. P. et al 1971; Varfolomeev, M. Yu et al 1972; Akolzin, A. P. et al 1974, 1981]. According to Robinson et al [1987] the basic mechanism of rust control by silicate is not explainable by silica iron complex as envisaged by earlier investigators [Dart, J. et al 1970, 1972]. A colloidal solution formation is more consistent with the facts. The influence of pH on the corrosion inhibition of iron and mild steel by sodium silicate is studied and it has been found that dissolution of sodium silicate in water generally increases the pH and the latter appears to play a pivotal role in the inhibition of corrosion [Lahodny, S. et al 1981]. A positive effect due to soluble silicate species resulting from pH change could be distinguished. Armstrong [1994] reported that in unbuffered solutions, sodium silicate is effective as a corrosion inhibitor for iron by inhibiting the anodic dissolution of the metal. However, the reduction of the corrosion rate in the pH range 9.6-11.6 can be accounted for, entirely by the change in pH caused by the addition of sodium silicate. Shams El-Din et al [1991] in a recent study showed that the corrosion of rusted steel in conditioned potable water was decreased by a maximum of 47% in the presence of 16 ppm Na2SiO3. However, once protection has been achieved, 4 ppm was the minimum amount to sustain a state of permanent inhibition.

Keeping in view, the great versatility of sodium silicate in preventing corrosion of iron and steels and its promising application in carbon steel and galvanized steel pipelines, it was proposed to carry out testing of this chemical in corroded or rusted steel water pipelines. In this study corrosion prevention properties of sodium silicate, supplied by Adwan Chemicals Industries, was tested. Immersion and electrochemical studies of steels in tap water, with and without sodium silicate, were carried out under static (stagnant) and dynamic (flow) conditions.

1.3

What are Soluble Silicates?

The most common method of preparing soluble sodium or potassium silicates involves the fusion of silica sand and the appropriate carbonate at high temperatures in a furnace

1858

to produce a water soluble glass. The reaction that takes place in the furnace is as follows, where Me is either sodium or potassium and x is the weight ratio: Me2CO3 + x SiO2 Me2O . xSiO2 + CO2 (1)

Silicate products with weight ratios of silica (SiO2) to alkali (Na2O or K2O) of up to 4.0 or even higher can be produced. The highest practical ratio is about 3.8. The fused glasses are dissolved in water to produce syrupy solutions, with solid contents from about 30 to 50%. The most common commercial liquid sodium silicate is a product having a weight ratio of silica to alkali (as Na2O) of 3.22, and with 37 to 38% solids. The ratio can be reduced by adding sodium hydroxide, to give a series of liquid products, or, in the low ratio range, a number of solid hydrous or anhydrous materials. A wide range of liquid forms are available, with SiO2:Na2O ratios from 3.8 to 1.6, and solid contents from about 25 to 55%. Because of the variation in properties over the range of products, it is important to specify the silicate ratio and concentration when discussing an application. The terms sodium silicate, soluble silicate, or water glass are not explicit enough. Furthermore, in water treatment applications, it should be clear whether dosages are expressed in terms of silica (SiO2), silicate solids, or sodium silicate solution as received. The liquids are similar in that they are all complex solutions of silicate anions, together with hydroxyl and sodium ions, but the properties differ depending on the ratio and concentration. The distribution of siliceous species shifts toward higher order polymers as the ratio of SiO2 to Na2O increases. This is accompanied by a rapid decrease in free alkalinity and an increase in the colloidal nature of the product. The stability or solubility domains for various species in equilibrium with amorphous silica are shown in a plot by Stumm and Morgan (Figure 1) [1970]. The areas of importance in this concentration-pH diagram are: the monomeric region (Region I), where the mononuclear species, such as Si(OH)4, Si(OH)3-, and SiO2(OH)22predominate; the multimeric region (Region II), where silicate polymers are stable; and the insolubility region (Region III), where amorphous silica precipitates. The

1859

commercial silicate products are found in Region II. Silica species seen in Region I are normally present in natural water. Several chemical reactions are characteristic of commercial silicate solutions. The concentrated solutions are strong alkalis, with pH values from about 11.0 to 13.0. The pH values of diluted silicate solutions, as a function of alkali contained by the silicate, are shown in Figure 2. For the lower loadings, the pH values range between about 9 and 10. However, concentrations typically used for potable water treatment which are in the ppm range, the pH is not affected greatly and, in fact, has a value nearly equal to that of untreated water. When the alkalinity is decreased or removed by neutralization, ion exchange or dialysis, the siliceous species will polymerize at a rate dependent on the silica concentration and pH value. At SiO2 concentrations of 0.5% or greater, the solution will pass through a colloidal solution state and ultimately form a solid hydrous gel (Region III, Figure 1). In the low dosages often used in water applications, the opposite reaction takes place over a wide pH range, i.e., more polynuclear forms found in Region II and III depolymerize the mononuclear form. At low to medium pH values, depolymerization occurs over a period of hours to days, depending on silica concentration [Lange, K. R. et al 1968].

Katsanis, Krumrine, and Falcone [1983] showed the time dependence of initial precipitation on the silica concentration and pH value (Figure 3). The pH range of minimum stability for the silicate solutions is 6 to 8. With aging, precipitates form over wider and wider ranges of silica concentrations and pH values, but after 24 h there is little additional precipitation in this pH region. At lower pH values, however, larger changes continue until the equilibrium conditions shown in Figure 1 are satisfied. Another reaction of interest to water chemists is the combination of the silica anions with multivalent cations to form insoluble hydrous metal silicates. This very general reaction, occurring with alkaline earth and other di- and trivalent cations, is used in the manufacture of pigments and fillers, soil grouting, waste treatment, metal casting, and many other industrial processes.

1860

1.4

The Nature of Corrosion Inhibition by Silicate

Silicates inhibit corrosion by forming a thin silicate film on the surfaces of metal water pipes as the silicate is fed through them. The films are almost invisible and do not increase in thickness with time. Once the film is formed, the protection is maintained as long as silicate treatment is continued; if stopped, the protection is gradually lost; if damaged, the film is self-healing, as long as the silicate feed is continued. The preferred product for this use is the 3:22 SiO2:Na2O weight ratio. A dosage of 24 to 25 ppm as silica (60 to 63 gal silicate solution per million gal of water) is recommended for the first month or two, to form film as quickly as possible. Feed rate can then be reduced to a maintenance dosage of 8 to 10 ppm SiO2 (20 to 25 gal/MG); maintenance dosages as low as 4 to 5 ppm SiO2 have been used successfully. For lower pH waters, a more alkaline silicate should be used, such as the 2.00 weight ratio, to reduce acidity and thereby enhance the formation of silicate films. At the levels of SiO2 treatment recommended, figures 1 and 3 show the precipitation of silica is not likely throughout the entire pH scale. However, those treatments have been shown to be quite effective, it is reasonable to assume that the thin silicate film on the metal pipe is formed by an adsorption mechanism. In old water lines, where corrosion and scale are present, silicate treatment may displace the accumulated deposits and release the rust [Stericker, W. et al 1938]. There may be temporary red water problem, but subsequently the pipes will be as cleaner as they originally were. It is advantageous to start the silicate treatment immediately following mechanical cleaning and flushing of the pipes if possible.

1.5

Theory of Silicate Inhibition

The silicates are considered anodic inhibitors, initially forming a film on anodic areas. Lehraman and Schuldener [1951] investigated the mechanism of the process and concluded that the deposition of the film depends on the presence of small amounts of corrosion products on the metal surfaces.

1861

At the anodic areas, the oxidation of a metal occurs: metal metaln+ + ne(2)

For ferrous metal pipes: Feo + 2H2O Fe(OH)2 + 2H+ + 2eIn the presence of oxygen, the ferrous ions will be oxidized to ferric: 2Fe++ + O2 + 2H+ 2Fe+++ + H2O (4)

(3)

The negatively charged siliceous species can react with the metallic cations to form the protective film. However, with continuous feeding of the silicates, the films extend to cathodic areas [Lehrman 1964]. This is important, because as Scheider and Stumm [1964] point out, the reduction in the free surface for anodic action, without a corresponding reduction in the surface available for cathodic action, will lead to an increase in current density at any unprotected anodic sites and result in accelerated corrosion and pitting at these sites. For this reason, it is important to use a maintenance dosage that is high enough to preserve the film throughout the entire system. Maintenance dosages as low as 4 ppm and lower dosages should be tried gradually. Microscopic and X-ray examinations of the films show two layers, with most of the silica in the surface layer adjacent to the water. Most of the silica is amorphous. When the hydrous metal oxide or metal silicate has been covered with a silica layer, further growth is halted, which accounts for the observation that continued silicate feed does not cause a build-up of the films.

2.

OBJECTIVES To study the effect of sodium silicate on corroded rusted steel pipe lines. To measure the corrosion rate of carbon steel in tap water in absence and presence of sodium silicate.

1862

 To determine the optimum concentration of Sodium Silicate dosing for corrosion inhibition of the rusted carbon steel. To determine the minimum concentration of Sodium Silicate dosing for maintaining the state of permanent inhibition.

3. 3.1

EXPERIMENTAL Immersion Test

The corrosion tests on carbon steel has been carried out in tap water under static and dynamic conditions. The composition of the tap water used during the experiments is given in Table-1. 3.1.1 Coupon Preparation

Commercial grade carbon steel containing 0.3% carbon was used for the tests. Carbon steel coupons of 100 x 40 x 4 mm size with a hole of 6 mm near the edge were cut from a sheet. Coupons were machined and abraded sequentially with silicon carbide papers of grades 180, 320 and 600. Polished coupons were washed, degreased with ethyl alcohol and dried up. After taking the initial weight and dimension, coupons were hanged in the silicate solutions of different concentrations. In order to avoid galvanic and crevice corrosion, coupons were loosely tightened in the holes using Teflon nuts and bolts. 3.1.2 Sodium Silicate Solution Preparation

Sodium silicate solutions of required concentrations were prepared as follows: Chemical Composition: Na2O SiO2 Ratio Sp. Gr. (at 20 oC) : : : : 9.8 % 31.2 % 3.18 1.44

1863

Normality =

Sp.Gr. x % Purity x 1000 Mol. Wt. x 100 1.44 x 31.2 x 1000 60.0855 x 100 = 7.47 N = 448.838 g. = 448838 ppm

Amount of silicate present in 1 litre = 7.47 x 60.0855 1000 ppm Stock. Solution Volume of silicate in one litre = 1000 x 1000 448838

= 2.22 ml

or

= 2.22 x 1.44 = 3.208 gm.

Stock solution was further diluted to prepare solutions of required concentrations. 3.1.3 Calculation of Corrosion Rate and % Inhibition

Corrosion rates were calculated according to the following equation: 534 W Corrosion Rate (mpy) = -----------------D A t Where; W = Weight loss (mg) D = Density of coupon material (g/ cm3) A = Area (inch2) t = Time of exposure (Hr.)

In order to evaluate the effect of inhibitor, percentage inhibition was calculated according to the following equation: (C.R.)TR - (C.R.)UTR % Inhibition = -------------------------- 100 (C.R.)UTR Where;

1864

(C.R.)UTR (C.R.)TR

= Corrosion rate in untreated tap water = Corrosion rate in treated water

Percentage of corrosion inhibition by silicate treatment of tap water in static, dynamic and loop tests is given in Table - 2. 3.1.4 Immersion in Static Water

Polished coupons were immersed in tap water containing 0, 5, 10, 15, 25, 35, 45, 100, 150 and 200 ppm of silicate. Glass beakers of 1000 cc were used to immerse the coupons for the period of 14, 28 and 42 days. Silicate solutions were daily replenished by fresh ones. In order to avoid experimental errors, two coupons were immersed in each solution. After immersion coupons were taken out, their condition was observed visually and photographed. After visual observations coupons were cleaned in a ultrasonic bath. Coupons were dried and their weights were taken to determine the weight loss and corrosion rate (Table 2). 3.1.5 Immersion in Dynamic Water Coupons were immersed in silicate solutions of above mentioned concentrations stirred with magnetic stirrer with 1300 rpm (velocity = 2 m/ sec). Duration of the test and weight loss measurement procedures were the same as in case of static water. Results are given in Table 2. While studying the effect of 0, 5, 10, and 15 ppm of silicate, coupons were exposed for 14, 28 and 42 days. During these tests it was observed that the corrosion behavior of carbon steel showed similar trend during all the three durations. It does not show any remarkable difference in the corrosion behavior during longer duration (42 days). Similar observations were made during the use of 25, 35 and 45 ppm of silicate when the tests were carried out for 14 and 28 days. In view of this, other experiments were carried out only for 14 days.

3.2

Corrosion Loop Test

1865

In order to study the inhibitive effect of silicate in flowing tap water, tests were carried out in a dynamic corrosion test loop. The test loop assembly is comprised of a plastic tank of 200 liter capacity, stainless steel pump, temperature, pressure, flow rate and dissolved oxygen measuring system (Figure 4). All the pipes and fittings of the loop were of PVC. At one end of the pipeline of the loop, two polished and initially weighed carbon steel coupons of known dimension (100 40

4 mm) were fixed on retractable

coupon holder. In order to measure the dissolved oxygen and temperature of the recirculating water, the probe of Dissolved Oxygen Monitor (Yellow Springs Instrument Co., Model 56) was connected with the water flowing in the pipeline and on-line monitoring of dissolved oxygen was done. Coupons were allowed to oxidize for 24 hours in the loop. After 24 hours, the required amount of silicate was added to the 200 liters of the tap water recirculating in the loop. Tests were carried out at 35 C in 0, 3, 5, 10 and 15 ppm of silicate solutions for 160 hours. The flow rate was maintained at 80 GPM (2.5 m/ Sec). After 160 hours of exposure, coupons were taken out, cleaned in ultrasonic bath, dried and finally weighed to know the weight loss/ area. The corrosion rate is calculated, from the obtained weight loss, time of exposure and original exposed surface area of the material, by the formula mentioned in section 2.1.4.

3.3

Open Circuit Potential (OCP) Measurement

While carrying out the loop test, OCP measurement was also carried out under the same condition and concentration of silicate. OCP measurement was carried out to study the change in potential of the carbon steel immersed in flowing water with time and concentration of silicate. The OCP curve pattern gives an idea about the protective behavior of the oxide scales formed on the metal surface.

For this study, probe electrodes (for Petrolite Corrosion Meter) of 1018 MS material and 44 6 mm dimension were abraded with 600 SiC paper and finally mirror polished with emery paper. Polished electrode was washed with ethyl alcohol and dried. Polished electrodes were inserted into the pipe line of the loop with the help of a tightly fitted rubber cork and connected to the plotter for on-line measurement of change in potential against calomel electrode. OCP of the carbon steel electrodes were plotted for 160 hours at a chart speed of 2 cm/ hr. All the electrodes were preoxidized for 24 hours in tap water. Silicate was added after 24 hours to make the solution of required concentration.

1866

To maintain a situation of the constant dosing, water was replenished after every 24 hours. 4. 4.1 RESULTS AND DISCUSSION Immersion Test

The corrosion data of the steel in water under static and dynamic conditions (Table 2) indicate that the trend of the corrosion inhibition by different concentrations of silicate did not change with time. However, the magnitude of corrosion rate has increased with the treatment period. This behavior indicates that the rate of formation and thickening of the passive film decrease with time, and in the present condition the above mentioned concentration of silicate is unable to passivate the steel coupon. Figure 6 shows typical plots of corrosion rate vs silicate concentration in tap water for a 14 day test.

On the basis of the study of the corrosion rate of carbon steel immersed in the tap water containing 5 - 200 ppm of sodium silicate, following generalizations can be made: In static tap water containing up to 100 ppm silicate, corrosion of carbon steel was always less than that in the dynamic water. Under dynamic conditions, the tap water containing more than 100 ppm silicate shows protection.

Sodium silicate has been found very effective in controlling corrosion of carbon steel by many investigators [Akolzin, A. P. et al 1971; Varfolomeev, M. Y. et al 1972; Akolzin, A. P. et al 1974, 1981; Robinson, R. B. et al 1987; Dart, J. et al 1970, 1972; Lahodny, O. et al 1981] and the optimum quantity of sodium silicate required to suppress the corrosion has been cited as less than 10 ppm. The present results show that under static immersion conditions, silicate fails to suppress the corrosion rates even at concentrations as high as 200 ppm. Under dynamic immersion conditions, corrosion rates appear to decrease at silicate dosing higher than 100 ppm. There is lowering down of corrosion rates by 17.8 and 71.0 % at silicate concentrations of 150 and 200 ppm, respectively.

1867

It is important to consider that above immersion tests were carried out in beaker, where the water gets saturated with oxygen (6 ppm dissolved oxygen in static condition and 8.5 ppm dissolved oxygen in dynamic condition). Oxygen concentration may also affect the performance of silicate. Also, it is evident from the above results that silicate is more effective in flowing water than stagnant one. It is a pointer towards the fact that corrosion prevention mechanism of silicate does depend on the flow rate of the water. It is also to be emphasized that the dynamic tests carried out in beaker are far away from the real situation of water flowing in a pipe line. In view of this, it was thought worthwhile to carry out a close circuit loop test where the flow rate was 80 GPM (flow velocity 2.5 m/ sec.) and the dissolved oxygen will be comparatively less.

4.2

Loop Test

In the loop test corrosion rate was determined at a flow rate of 2.5 m/ sec. which is not much different from the flow rate maintained in the beaker during dynamic immersion test. However, the dissolved oxygen concentration in the loop test was much lower (5.5 ppm) than that found during the dynamic immersion test (8.5 ppm).

Change in pH during the loop test was also monitored. The pH of the fresh tap water was found in the range of 7.2 - 7.5, but after exposing the carbon steel coupons for five days, the pH of the water was shifting towards more basic range, showing a value of 8.2. The alkalinity of the tap water could be due to the presence of rust (FeOOH) in the water. The pH pattern of silicate treated tap water at different silicate concentrations was as follows:

3 ppm Silicate Solution 5 ppm Silicate Solution 10 ppm Silicate Solution 15 ppm Silicate Solution

pH = 8.0 pH = 8.2 pH = 8.4 pH = 8.5

The alkalinity does not show any considerable rise with time.

1868

The results of corrosion test loop showed in general, relatively high corrosion as compared to that observed during dynamic and static immersion tests. But inspite of very high corrosion rates, during loop tests, silicate was found very effective in inhibiting the corrosion of carbon steel. Corrosion rates observed after treating the tap water with silicate are given in table 2. The plots of corrosion rates against the silicate concentration, show that the addition of silicate in the range of 3 - 15 ppm inhibits the corrosion by >45% (Figure 7). However, out of the above mentioned concentrations, 5 ppm appeared to be the most effective amount which can inhibit the corrosion by more than 62% (Figure 8). It is also evident from Figure 7 that an increase in the concentration of silicate above 5 ppm does not increase the corrosion protection.

Comparing the corrosion behavior of carbon steels in sodium silicate treated tap water tested under three different conditions e.g., static immersion, dynamic immersion and dynamic loop, the results of loop test appear to be of far reaching importance. In loop tests, corrosion rates are although more due to high flow velocity, silicate is far more effective in inhibiting the corrosion than other two types of tests.

In the classical approach, the corrosion of iron is represented by the anodic reaction: Fe Fe 2+ + 2e-

------(i)

And by cathodic reaction: 2 H+ + 2 e-

H2

------(ii) 2OH- ------(iii)

In the presence of oxygen depolarization takes place: O2 + H2O + 2e-

The effect of velocity of water on steel is largely determined by oxygen diffusion and film formation. The rate of corrosion is generally controlled by oxygen depolarization. The rate of diffusion of oxygen to the metallic surface depends on the rate of diffusion through the corrosion product films which act as a barrier to the passage of oxygen. For instance, one of such well known films is of Fe(OH)3 formed by the reaction:

1869

2 Fe + 2H2O + O2

2 Fe(OH)2

----(iv)

2 Fe(OH)2 + H2O + O2

2 Fe(OH)3 ----(v)

In addition to the hydrated oxides, a layer of magnetite (Fe3O4) or FeO.Fe2O3 often forms between iron oxides (FeO) and hematite (Fe2O3). Actually, the various oxides and hydroxides of iron form a rather complicated system of compounds. The compound FeOOH has been found to exist in three different crystal forms plus an amorphous form [Metal Hand Book, 1987]. The occurrence of the various oxide species is dependent on pH, oxygen availability and the composition of the steel. Rust formation in open atmosphere: 4 Fe + H2O + 3 O2

2 Fe2O3 . H2O

Rust formation in water:

It generally forms two layers: FeOOH + FeOOH FeOOH + Fe3O4 (Outer, loose, crystalline) (Inner, dense, crystalline)

The actual nature of the rust film is important because FeOOH seems to be more adherent than hydrated Fe3O4 and Fe2O3 [Metal Hand Book, 1987], and therefore, more likely to show the corrosive attack, but the higher oxides and oxy-hydroxides are more prone to spallation.

The rust layer, therefore, provides resistance to the diffusion of oxygen. The first resistance is offered to the passage of oxygen from the fluid to the porous corrosion product film. The second resistance is applied by the porous solid layer filled with fluid.

1870

Cathodic reaction proceeds very rapidly at the metal surface because of its low activation energy. The concentration of oxygen at the metal surface is therefore, very small as compared to the bulk concentration. The main resistance to the mass transfer is offered by Fe(OH)3. At higher velocities, the resistance to the mass transfer is decreased and the rate of corrosion is increased. However, this has also been reported that the rate of corrosion is not only related to the velocity of the water but it also depends on the type of corrosion product and temperature. For instance, Butler and Stroud [1961] has reported a maxima in the corrosion rate at a velocity of 1 m/ sec. as compared to that at 2 m/ sec. while temperature is 25 C (Figure 8). But at 55 C, corrosion rate increases with velocity when measured at the above said velocities. At further higher temperature of 76 C, the corrosion rate again decreases with rise in velocity. This is attributed to the more compactness of the corrosion product at higher temperatures which makes it more resistant at higher velocities [Whiteman, G. E. et al 1924]. Thus, oxygen diffusivity, temperature and corrosion product play an important role in the corrosion of carbon steel and the rate of corrosion depends on the dominance of one or more factors. Although increasing oxygen concentration at first accentuates corrosion of iron in water, as one expects from equation (iii), it is found that beyond a critical concentration the corrosion rate may drop again to low values [Groesbeck, E. et al 1931].

Regardless of the actual pH of water between pH 4 and 9.5, the surface of iron is always in contact with an alkaline saturated solution of hydrous ferrous oxide. Since the corrosion product film next to the iron is essentially unchanged by external conditions within the above range of pH, the corrosion rate is altered only by change in dissolved oxygen and velocity of water. Also, so far the effect of pH is concerned, it is well known that the corrosion rate of iron in aerated solution is independent of pH ranging between 4 and 10, at a pH > 10 the corrosion rate decreases rapidly with increasing pH [Whiteman, G. E. et al 1924]. Dissolution of Na2SiO3 in water generally increases the pH since silicic acid H2SiO3 is a weak acid with pK1 = 9.5 and pK2 = 12.7 and the following reaction occurs: SiO32- + H2O H.SiO3- + OH

1871

As after addingNa2SiO3 in the range of 3 - 15 ppm the pH of the solution does not exceed pH 9, while the corrosion rate is considerably suppressed, it appears that Na2SiO3 is controlling the corrosion of steel not by changing the pH but by some other mechanism.

High concentration of dissolved oxygen (8.5 ppm) and velocity of water ( 2 m/ Sec.) appear to be the cause of higher corrosion rate observed during the dynamic immersion test carried out at 25 C. The analysis of the corrosion product shows that the rust formed during the static and dynamic immersion tests carried out in the beakers was hydrated loose -Fe2O3 and Fe3O4 which give comparatively less protection to the steel from corrosion [Metal Hand Book, 1987]. No remarkable effect of the inhibitor addition during static and dynamic immersion tests can be attributed to the formation of the above mentioned loose hydrated oxides which may be due to result of the comparatively high concentration of dissolved oxygen.

The remarkable suppression in the corrosion rate of carbon steel in 3 - 15 ppm of Na2SiO3 solution observed during loop test is an interesting and important aspect of this study. It has been discussed by many authors that the nature of the oxide scales formed on the steel surface plays an important role in determining the inhibitive effect of Na2SiO3 in water [24 - 32]. Chen et al [1991], postulated that a layered structure of Fe2O3/FeO/Fe is responsible for the lowering of the corrosion rate of iron in contact with both aerated and deaerated Silicate solutions. Presence of FeO in the rust formed on the steel coupons exposed in the loop was confirmed by X-ray diffraction analysis. Formation of FeO along with -Fe2O3 is expected due to low dissolved oxygen (5.5 ppm) as compared to the dynamic test (8.5 ppm) or also due to higher flow rate.

It is also very important to determine the optimum concentration of the Na2SiO3 which should be fed continuously. This is because of the fact that after covering the anodic site, the film of the inhibitor may extend to the cathodic areas. As Schneider and Stumm reported [1964], any reduction in free surface for anodic action without a corresponding reduction in the surface available for cathodic action, will lead to an increase in current density at any unprotected anodic sites and result in accelerated corrosion and pitting at these sites.

1872

These results indicate that it is very important to use a maintained dosage of Silicate which could be most suitable to preserve the protective film throughout the system. Also, excess aeration of the flowing water should be avoided to enhance the inhibitive effect of Na2SiO3. This findings disagree with the contention of Katasanis et al [1986] that Silicate treatment would be more effective with high oxygen content.

4.3

Open Circuit Potential (OCP) Test

The open circuit potentials of clean and rusted C-steel coupons were followed as a function of time in the untreated and silicate treated tap water until a constant value was attained. At the moment of immersion in water, the clean coupons registered potentials ranging between -450 to -480 mV vs SCE. These potentials slowly drifted towards more negative values and attained constancy in the range of -570 mV to -632 mV, after around 15 - 20 hours of immersion. The negative shift in potential is indicative of the destruction of the air-formed pre-immersion oxide film, and the attack on the metal. Shifting of the potential towards the original OCP is an indication of passivation of the metal surface. It is evident from Figure 9 that an addition of 3 - 5 ppm of Sodium Silicate to tap water provides a maximum passivation to the C-steel coupons as the OCP shifts from ~ -630 mV to ~ -560 mV, while a higher concentration of 10 and 15 ppm of silicate could not shift the OCP below -609 mV. These results are in consistence with the findings of the Loop Test and confirms the remarkable corrosion inhibition when 3 5 ppm of silicate is added to the flowing tap water. 5. CONCLUSIONS

On the basis of Static and Dynamic Immersion, Loop and Electrochemical Tests, the following conclusions can be made: 1. Optimum concentration of silicate depends upon the dissolved oxygen and the rate of flow of the water. 2. Although the corrosion rates of the steel in dynamic condition are invariably higher than those under static condition but silicate acts

1873

more effectively as an inhibitor when the water is in dynamic condition. 3. From the loop test it is evident that at ambient temperature, 5 ppm concentration of silicate is sufficient to suppress the corrosion of carbon steel by 62% in tap water having a dissolved oxygen level of 5.5 ppm and a flow rate of 2.5 m/Sec.

4. Electrochemical tests indicate that silicate concentration in the range of 3-5 ppm provides remarkable inhibition to the flowing tap water.

6.

RECOMMENDATIONS 1. In order to protect carbon steel pipes, it is recommended to maintain a continuous dosing of 5 ppm of silicate in tap water because it is effective in controlling the corrosion of C-steels as shown by the results of loop tests. 2. Due to the clear effect of oxygen content in normal tap water, the role of other oxidizing agent contents, such as chlorine, is to be investigated in a separate study. Such an overall oxidizing agents effect could lead to a better understanding about silicate dose optimization.

REFERENCES 1. Akol'zin, A. P. et al (1971) 'Use of sodium silicate for preserving the equipment of water heating pipe lines'. Energetik, 7, 8-9. 2. Akol'zin, A. P. et al (1974) 'Use of sodium silicate to prevent corrosion of boiler steel'. Energetik, 6, 12-13. 3. Akol'zin, A. P. et al (1974) 'Use of sodium silicate for protecting heating system from corrosion. Zh. Telplo energy, 35-490.

1874

4. Akol'zin, A. P. et al (1981) 'Protection of boilers from downtime corrosion by sodium silicate solution prepared by using hard water'. Prom. Energy, 7, 41-43. 5. Armstrong, R. D. and Zhou, S. (1988) Corrosion Sci., 28, 1177. 6. Armstrong, R. D., Peggs, L. and Walsh, A. (1994) 'Behavior of sodium silicate and sodium phosphate (tribasic) as corrosion inhibitor for iron'. J. Applied Electrochemistry, 24, 1244. 7. Butler, B. and Strand, G. (1961) British Corrosion Journal. 1, 110. 8. Chen J. R. and Chao, H. Y. (1991) Surface Sci., 247, 352. 9. Dart, J. and Foley, P. D. (1970) 'Preventing iron deposition with sodium silicate'. Journal AWWA, 62, 663. 10. Dart, J. and Foley, P. D. (1972) 'Silicate as Fe, Mn deposition preventive in distribution systems'. Journal AWWA, 64, 244. 11. Duffeck, E. F. and Mckinney, B. S. (1956), ibid, 103, 645. 12. Encyclopedia of Polymer Science and Engineering, 'Silicates' (1988), 2nd Ed., 15,178. Fujita, N., Matsuura, C. and Ishigure, K. (1989) Corrosion. 45 (11), 901. 13. Groesbeck, E. and Waldron, L. (1931), Proc. Amer. Soc., Testing Materials, 31, partII, 279-291. 14. Kalsanis, E. P., Kumrine, P. H. and Falcone, J. S. Jr. (1983) Chemistry of precipitation and scale formation in geological systems. Presented at International Symposium on Oil-field and Geothermal Chemistry, Denver, CO, June, 1- 3. 15. Katsanis, E. P., Esmonde, W. B. and Spencer, R. W. (1986) Matter. Perform.. 25, 19. 16. Lahodny, O. S. and Kasleau, L. (1981) 'The influence of pH on the inhibition of corrosion of iron and mild steel by sodium silicate'. Corrosion Science, 21, 265. 17. Lange, K. R. and Spencer, R. W. (1968) Environmental Science and Technology, 2(3), 212. 18. Lehrman, L. and Shuldener, H. L. (1951) Journal of the American Water Works Association. 43, 175. 19. Lehraman, L. (1964) Journal of the American Water Works Association. 56(8), 1009. 20. Metal Hand Book (1987),American Society of Metals, 13,511. 21. Pryor, M. J. and Cohen, M. (1953) J. Electrochem. Soc., 100, 203.

1875

22. Robinson, R. B., Minear, R. A. andHolden, J. M. (1987) 'Effects of several ions on iron treatment by sodium silicate and hypochlorite'. Journal AWWA, 79, 116. 23. Schneider, C. R. and Stumm, W. (1964) Journal of the American Water Works Association. 56(5), 621. 24. Shams El-Din, A. M., Saber, T. M. and Salem, A. H. (1991), Proc. Vth Middle East Corrosion Conference (NACE), Bahrain, October 28-30, 331. 25. Shuldner, H. L. and Sussman, S. S. (1960), ibid, 16, 126. 26. Stericker, W. (1938) Industrial and Engineering Chemistry. 30 (3), 348. 27. Stumm, W. and Morgan, J. J. (1970) Aquatic chemistry, an introduction emphasizing chemical equilibria in Natural Waters. Wiley-Interscience, New York, 395. 28. Varfolomeev, M. Yu. and Glyakhov, O. (1972) 'Use of sodium silicate for the protection of centralized heat supply system from internal corrosion'. Elek. Sta. 10, 3134. 29. Whiteman, G. E., Russell, R. P. and Alliery, V. J. (1924) Ind. Eng. Chem.. 16, 665. 30. Wood, J. W., Beecher, J. S. and Laurence, P. S. (1957), Corrosion NACE, 13, 41.

1876

1877

Table 2. Percentare Corrosion Inhibition by Addition of Sodium Silicate in Tap Water During Different Tests S. No. Type of Corr. Test Corrosion Rate in Untreated Water (mpy) 1.07 Silicate Concentration (ppm) 5 10 15 25 35 45 100 150 200 5 10 15 25 35 45 100 150 200 3 5 10 15 Corrosion Rate in Silicate Treated Tap Water (mpy) 1.07 1.04 1.13 1.55 1.53 1.60 1.76 1.56 1.13 12.58 10.96 10.96 13.94 17.15 16.30 12.08 9.42 3.30 107.20 76.15 109.06 91.65 % Inhibition

Static Immersion Test

0.0 -2.8 5.6 44.8 43.0 49.5 64.5 45.8 5.6 9.5 -4.5 -4.5 21.4 49.4 42.0 5.2 -17.8 -71.0 -46.9 -62.2 -45.9 -54.5

Dynamic Immersion Test

11.48

Loop Test

201.5

1878

EFFECT OF Ph ON THE SOLUBILITY OF AMORPHOUS SILICA

1. 2. 3. 4.

SiO2 (amorphous) + 2H2O Si(OH)4 SiO(OH)3 Si(OH)4

= = = =

log k (250C) Si(OH)4 -2.7 + Si(OH)3 + H -9.46 -2 + SiO2(OH)2 = H -12.56 -2 + Si4O6(OH)6 + 2H + 4H2O -12.57

Figure 1. Species in equilibrium with amorphous silica. Diagram computed from equilibrium constants (250C). The line surrounding the shaded area gives the maximum soluble silica. The mononuclear wall represents the lower concentration limit, below which multinuclear silica species area not stable. In natural waters the dissolved silica is present as monomeric silicic acid.

1879

Figure 2.

PH values of sodium silicate solutions at 200C. Numbers on lines are SiO2/Na2O ratio by weight.

Figure 3.

Solubility diagrams for 3.22 weight ratio Na-silicate as a function of pH for aging times up to 1 month at 250C.

1880

CORROSION TEST LOOP

OM V EC FM TI V RESERVIOR

Drain RC EC FM OM P = = = = Electrochemical Cell Flow Meter Oxygen Monitor Pump FM PG PG RC TI V = = = = V Pressure Gauge Retractable Coupon Holder Temperature Indicator Valve P V

Figure 4. Flow Diagram of Corrosion Test Loop.

1881

Figure 5. Corrosion test loop used for the study of corrosion inhibition by sodium silicate

1882

14 12
Dynamic Static

CORROSION RATE (mpy)

10 8 6 4 2 0 0 5 10 15

SILICATE CONCENTRATION (PPM) Figure 6. Corrosion of carbon steel in sodium silicate containing tap water (A 14 Day Test Result)
250

200

CORROSION RATE (mpy)

150

100

50

0 0 3 5 10 15

SODIUM SILICATE CONC. (ppm) Figure 7. Corrosion behavior of carbon steel in silicate treated tap water during loop test.

1883

70

62.2 54.5

60 46.9 45.9 50

% Inhibition

40

30

20

10

0 3 5 10 15

Conc. of Silicate (ppm)

Figure 8. Corrosion inhibition of carbon steel by sodium silicate treatment in loop test.

1884

-675

-650

POTENTIAL (-mV)

-625

-600

-575 Tap Water -550 Tap Water + 3 ppm Tap Water + 5ppm -525 Tap Water + 10 ppm Tap Water +15 ppm -500 0 20 40 60 80 100 120 140 160

TEST DURATION (Hr. ) Figure 9. Change in open circuit potential of carbon steel in sodium silicate treated tap water during loop test.

1885