Corrosion behaviour of mild steel in chloride solutions with Na 2 HPO 4 additive and the development of antifreeze

Abstract

Chloride salts can prevent roads from freezing in cold regions, together with severe corrosion on steel constructions. To develop an ecological and low-corrosive antifreeze, di-sodium hydrogenphosphate (Na₂HPO₄) was chosen as the additive into chloride salts. The addition of Na₂HPO₄ into either of the antifreezes of natural salt (N/S), NaCl, CaCl₂ or MgCl₂ can suppress the corrosion reaction of mild steel in the 3.0% antifreeze solution. Moreover, the addition of CaCl₂ or MgCl₂ into the antifreeze containing NaCl and 2.0% Na₂HPO₄ significantly decreased the corrosion rate. According to polarisation and XPS analyses, it is evident that the obtained low corrosion rates on mild steel are related to the suppressed cathodic and anodic reactions and the formation of a protective film containing Fe, P, O, Ca or Mg.

This paper is part of a supplementary issue from the 17th Asia-Pacific Corrosion Control Conference (APCCC-17).

Introduction

To protect roads from freezing of water in cold winter regions, antifreezes, such as NaCl and CaCl₂, are usually sprayed on roads to depress the freezing point of water or to melt ice and snow [1–5]. However, the chloride ions dissolved in water generally cause corrosion of metallic structures on/in roads as well as running automobiles. It is known as the salt damage of corrosion [4–9]. Various anticorrosive components, such as phosphates, chromates, nitrites, magnesium salts, aluminium salts, acetates, methylglucoside, triethanolamine, organozinc compounds, have been utilised as corrosion inhibitors into the traditional antifreezes of NaCl or CaCl₂ [1,2]. The use of non-toxic phosphates, particularly the safe hydrogenphosphate in the food industry had attracted the interest of researchers in suppression of the corrosion of steels [1,3,10]. Till date, it had been confirmed from a continuous immersion corrosion test that the addition of Na₂HPO₄ to NaCl really promoted rather than suppressed the corrosion of steel [1]. On the other hand, the corrosion inhibition effect was confirmed when adding the Na₂HPO₄ together with CaCl₂ or MgCl₂ to the NaCl solution.

However, the authors considered that it should be more prudent to assess the result of the addition of Na₂HPO₄ to NaCl before drawing final conclusion. Furthermore, the wet and dry cyclic corrosion test, which is closer to the real corrosion environment, should be carried out, and the polarisation measurement as well as the surface composition analysis should be utilised to clarify the mechanism. Although the addition of K₂HPO₄ in antifreeze had appeared in opened patents [1], Na₂HPO₄, which is safely used as food additive and has higher antifreeze ability, has not been tried. With the aim of developing low cost, ecological, low-corrosive and mass-produced antifreezes, it is worthy to know the effect of adding Na₂HPO₄ to NaCl, MgCl₂ or CaCl₂ on the corrosion behaviour and understand the mechanism. Based on the above considerations, the authors had developed an ecological and low-corrosive antifreeze easily mass-produced antifreeze by adding di-sodium hydrogen phosphate (Na₂HPO₄) into a natural salt (N/S) [3]. In the N/S, mainly the NaCl and a little of impurities of CaCl₂ and MgCl₂ are contained. However, even if the low corrosiveness of the antifreeze on mild steel had been confirmed in the wet and dry cyclic test, the influence of each component in the N/S and the amount of Na₂HPO₄ on the corrosion behaviour of the steel should be assessed.

Therefore, here in this work, several types antifreezes containing NaCl, MgCl₂, CaCl₂ and Na₂HPO₄, were produced and their aqueous solutions were used in the wet and dry cyclic corrosion test to get the corrosion rate of mild steel. Furthermore, based on the polarisation and X-ray photoelectron spectroscopy (XPS) analysis, the mechanism of the corrosion behaviour of mild steel in the aqueous solution of the antifreezes is discussed.

Experimental

Preparation of solution with dissolved antifreeze and the mild steel specimen

Sodium chloride (NaCl), calcium chloride (CaCl₂) and magnesium chloride (MgCl₂) were considered as antifreezes and the latter two (CaCl₂, MgCl₂) were also treated as additives into NaCl. In addition, the natural salt (N/S) containing about 96.0% NaCl and 0.23% CaCl₂ as well as 0.19% MgCl₂ was also used as one of the antifreezes. On the other hand, di-sodium hydrogen phosphate (Na₂HPO₄) was used as the corrosion inhibitor to be added into the above antifreezes. It is the reagent of MgCl₂·6H₂O but not that of MgCl₂ was used in this work. Therefore, the ratio of MgCl₂·6H₂O rather than that of MgCl₂ was described in the antifreeze. The 3.0 wt-% antifreeze solution containing several types of the above salts (N/S, NaCl, CaCl₂, MgCl₂·6H₂O or Na₂HPO₄) was used in the wet and dry cyclic corrosion test and the polarisation measurement. In the following description, the ratios of each type of salt (N/S, NaCl, CaCl₂, MgCl₂·6H₂O or Na₂HPO₄) in the antifreeze are noted rather than their real concentrations in the solution.

A mild steel plate (JIS SPCC of composition (C < 0.15%, Mn < 0.60%, P < 0.100%, S < 0.035%)) with size of 67 mm × 50 mm × 1 mm was used as the specimen to evaluate the corrosiveness of the antifreezes. The specimens were ultrasonically cleaned in water and acetone and the weight change was measured using a precise balance.

Wet and dry cyclic corrosion test

The specimens were kept in 3.0 wt-% solution of antifreeze in a 300-mL glass beaker. The solution was kept in a chamber with constant temperature (296 ± 2 K) and humidity (50 ± 2%RH) for 7 days, with the cyclically wet immersion for 1 day and then dried for another 1 day in the same chamber. Air was continuously pumped into the chamber during the test. After that, the rust formed on the specimen was carefully removed by a soft brush without damaging the uncorded steel in a solution of 1.0% propargyl alcohol (C₃H₄O) and 5.0% hydrogen chloride (HCl), following which they were cleaned in water and acetone. From the weight change before and after corrosion test, the corrosion rate of steel was calculated in the unit of mdd (mg/(dm²·day)).

Polarisation measurement

The specimen surface was masked with silicone sealant, leaving only a central test zone of 10 mm × 10 mm as the polarisation area. The polarisation curve was measured with the three-electrode configuration in the above solutions at 303 ± 2 K with a platinum sheet counter electrode and Ag/AgCl (S.S.E.) reference electrode using an HAB151 potentiostat of HOKUTO DENKO. In the polarisation measurement, potential was swept with a rate of 0.33 mV s⁻¹ from the cathodic potential of –600 mV to the anodic side to clarify the current variation on both the cathodic and the anodic reactions. The test solution was bubbled with air for 1.8 ks prior to polarisation.

XPS analysis

After the wet and dry cyclic test, several selected specimen surfaces were analysed by XPS (Shimadzu Co.: AXIS ULTRA). The X-ray was the Kα spectrum from Mg target, generated at the voltage of 15 kV and the anodic current of 10 mA. The air pressure in the vacuum chamber was lower than 5 × 10⁻⁷ Pa and the analysis area was 2 mm × 1 mm. The analysed binding energy ranges from 1100 to 0 eV with a scanning step of 0.2 eV per 300 ms without pre-etching the surface. From the obtained spectra, the components on the surface were determined and discussed in relation to the corrosion behaviour of the steel.

Ice melting capacity

A glass dish with 6.70 g distilled water was first frozen in a cold chamber with a constant temperature of 268 K. Next, 0.35 g salt particles (99.9% NaCl) or 0.35 g of the developed antifreeze (97.3% N/S, 2.7% Na₂HPO₄) was sprayed on the ice. The dish was then sealed with a glass cover and kept in the above cold chamber for 3.6 or 21.6 ks. The dissolved water on the ice was removed by a tissue paper and the weight change of the tissue paper was measured to obtain the amount of dissolved ice.

Results and discussion

Wet and dry cyclic corrosion test

Solutions containing NaCl, Na₂HPO₄, CaCl₂ and MgCl₂

Figure 1(a,b) shows the appearances of the solutions of 3.0 wt-% N/S, NaCl, CaCl₂ or MgCl₂·6H₂O before and after the wet and dry cyclic corrosion test. Figure 1(c) shows the mild steel surface after the test without removing the rust. No appreciable difference can be found from both the solutions and the specimen surfaces. According to the following description (Figures 2, 8, and 10), it is evident that the corrosion rates in the CaCl₂ and MgCl₂ solution are a slightly greater than those in the N/S and NaCl solutions.

Figure 1

Appearance of solutions of 3.0% N/S, NaCl, CaCl₂ and MgCl₂·6H₂O before (a) and after (b) wet and dry cyclic corrosion and (c) specimens after wet and dry cyclic corrosion without the removal of rust.

Figure 2 shows the corrosion rates of the mild steel obtained from the wet and dry cyclic corrosion test in 3.0% (NaCl + x% Na₂HPO₄) or 3.0% (N/S + x% Na₂HPO₄) solutions, where the x% represents the ratio of Na₂HPO₄ in the antifreeze of (NaCl + Na₂HPO₄) or (N/S + Na₂HPO₄). The corrosion rates of steel in 3.0% NaCl or 3.0% N/S are almost the same as 20.6 mdd. The value decreased to 16.5 mdd in the NaCl solution when 0.5% Na₂HPO₄ was added into antifreeze and remained same even when increasing the ratio of Na₂HPO₄ in the antifreeze to 20.0%. The value decreased to 10.3 mdd when the Na₂HPO₄ ratio of 30.0% and a value of 0.4 mdd was obtained at the Na₂HPO₄ ratio of 100% (solution concentration: 3.0% Na₂HPO₄). On the other hand, comparing to the NaCl solution, the decrease in corrosion rate in the N/S solution is much larger when adding Na₂HPO₄. The value decreased to 6.2 mdd at the ratio of 0.5% Na₂HPO₄, 0.59 mdd at the ratio of 2.0% Na₂HPO₄. The value remained almost the same low level when the Na₂HPO₄ ratio is 2.0% and above. This means that the Na₂HPO₄ really has an ability to inhibit the corrosion progress in both NaCl and N/S solution, and the suppression effect is much larger in the N/S solution than in the NaCl solution. The result in the NaCl solution is different from that reported literature [1], where a continuous immersion test was carried out. Since the antifreeze has 96.0% NaCl and a little of CaCl₂ as well as MgCl₂ in the N/S, the Ca²⁺ and Mg²⁺ should have played important roles with the to inhibit the corrosion reaction. Perhaps, an insoluble film had been formed on the steel surface to decelerate the corrosion reaction. Since a much small corrosion rate was obtained in the 3.0% (N/S + 2.0% Na₂HPO₄) solution, it should be important to note the role of CaCl₂ and MgCl₂ in (NaCl + 2.0% Na₂HPO₄) antifreeze in the following experiments.

Figure 2

Corrosion rates obtained in 3.0% (NaCl + x% Na₂HPO4) or in 3.0% (N/S + x% Na₂HPO₄) solutions.

Figure 3 shows the corroded specimen surfaces with rust. Brighter surfaces can be seen in case of lower corrosion rates. Slight corrosion occurred in the N/S solution when the Na₂HPO₄ ratio is over 2.0%, leaving a bright non-corroded surface where a protective film might have been formed. On the other hand, the occurrence of the bright surface in NaCl solution is delayed only when the Na₂HPO₄ ratio is over 30.0%.

Figure 3

Specimen surfaces after wet and dry cyclic corrosion in 3.0% (NaCl + x% Na₂HPO₄) or in 3.0% (N/S + x% Na₂HPO₄) solution, without the removal of rust.

Figure 4 shows the corrosion rates of the steel obtained in 3.0 wt-% (NaCl + 2.0% Na₂HPO₄ + x% CaCl₂) solutions, where the x% means the ratio of CaCl₂ in the antifreeze of (NaCl + Na₂HPO₄ + CaCl₂). From this figure, it can be seen that the corrosion rate largely decreased from 20.4 to 3.5 mdd when adding 0.5% CaCl₂ into the antifreeze of (NaCl + 2.0% Na₂HPO₄). The rate then gradually increased to 7.2 mdd with the increase in addition of CaCl₂ ratio to 3.0% and kept almost the value in the case of 10.0% CaCl₂. Figure 5 shows the solution appearance before and after the test as well as the corroded specimen surfaces with rust. The appearance of the solutions before the test is also shown as inset in Figure 4. No large difference could be found on the specimen surface. On the other hand, it can be seen from the solution pictures that the white precipitation appeared just when the ratio of Na₂HPO₄ is more than 1.0%, which is well corresponding to the variation of the corrosion rate. The precipitated matter is considered as CaHPO₄, which has only a solubility of 0.02 g in 100 mL water at 298 K. Too many Ca²⁺ ions promoted the formation of slightly soluble CaHPO₄ at much early stage before the corrosion test, which decreased the effective inhibitor species of in the solution. On the other hand, it also indicates that the condensation and protective film containing Ca and P should have also slowly formed during the corrosion reaction, which directly inhibited the corrosion reaction. However, the much more quickly formed CaHPO₄ before corrosion test is porous and would adhere on the surface, which might act as obstacle for formation of protective film.

Figure 4

Corrosion rates obtained in solution with the concentration of 3.0% (NaCl + 2.0% Na₂HP0₄ + x% CaCl₂).

Figure 5

Appearance of solutions of 3.0% (NaCl + 2.0% Na₂HPO₄ + x% CaCl₂) before (a) and after (b) wet and dry cyclic corrosion and (c) specimens after wet and dry cyclic corrosion without removing rust.

Figure 6 shows the corrosion rates of the steel obtained in 3.0% (NaCl + 2.0% Na₂HPO₄ + x% MgCl₂·6H₂O) solutions, where the x% means the ratio of MgCl₂·6H₂O in the antifreeze of (NaCl + Na₂HPO₄ + MgCl₂·6H₂O). Different from the results in Figure 4, the corrosion rate gradually decreased with the increase in the ratio of MgCl₂·6H₂O in the antifreeze and the smallest value was obtained at the MgCl₂·6H₂O ratio of 10.0%.

Figure 6

Corrosion rates obtained in solutions with the concentration of 3.0% (NaCl + 2.0% Na₂HPO₄ + x% MgCl₂·6H₂O).

Figure 7 shows the solution appearance before and after the test as well as the corroded specimen surfaces with rust. The appearance of the solution before the immersion of the steel is also shown as inset in Figure 6. According to the observation of the specimen, lesser corrosion had occurred when the ratio of MgCl₂·6H₂O is 3.0% and above. According to these observations, no precipitation can be found from all the conditions with various MgCl₂·6H₂O ratios (0.1-10.0%). It suggests that the solubility of MgHPO₄ in water is large enough to avoid precipitation in this experiment.

Figure 7

Appearance of solutions of 3.0% NaCl + 2.0%Na₂HPO₄ + x%MgCl₂·6H₂O) before (a) and after (b) wet and dry cyclic corrosion and specimens after wet and dry cyclic corrosion without removing rust.

Solutions containing CaCl₂, MgCl₂ and Na₂HPO₄

Figure 8 shows the corrosion rates obtained in 3.0 wt-% (CaCl₂ + x% Na₂HPO₄) solutions, where x% represents the ratio of Na₂HPO₄ in the antifreeze of (CaCl₂ + Na₂HPO₄). When the ratio of Na₂HPO₄ increased from 0.1 to 1.0%, the corrosion rate gradually decreased. However, the value increased again with the increase in the Na₂HPO₄ from 4.0 to 30.0%. The smallest corrosion rate was about 5.5 mdd, which was obtained at the Na₂HPO₄ ratio of 1.0-4.0%. Figure 9 shows the solution appearance before and after the test as well as the corroded specimen surfaces with rust. From the specimen surface observation and the solutions after corrosion test, the corrosion in the solution with Na₂HPO₄ ratio more than 1.0% is slighter. According to the solution observation (Figure 8: inset; Figure 9), it is found that precipitation occurred when the Na₂HPO₄ ratio is over 1.0%. This tendency is very near to that described in Figure 4, where CaCl₂ was added into the solution with NaCl and Na₂HPO₄. Therefore, the white precipitation should be also CaHPO₄, which decreased the effective component of Ca²⁺ ions in the solution in this case. Of course, the dissolved ions with the Ca²⁺ ions in the solution should have helped to form a protective film when it was formed during corrosion.

Figure 8

Corrosion rates obtained in solutions with the concentration of 3.0% (CaCl₂ + x% Na₂HPO₄) solution.

Figure 9

Appearance of solutions of 3.0% (CaCl₂ + x%Na₂HPO₄) before (a) and after (b) wet and dry cyclic corrosion and (c) specimens after wet and dry cyclic corrosion without removing rust.

Figure 10 shows the corrosion rates obtained in 3.0 wt-% (MgCl₂·6H₂O + x% Na₂HPO₄) solutions, where x% represents the ratio of Na₂HPO₄ in the antifreeze of (MgCl₂·6H₂O + Na₂HPO₄). It is known that with the increase of Na₂HPO₄ addition the corrosion rate gradually decreased until to the ratio of Na₂HPO₄ of 3.0%. After that, the corrosion rate almost did not vary even when increasing the ratio to 30.0%.

Figure 10

Corrosion rates obtained in solutions with the concentration of 3.0% (MgCl₂·6H₂O + x% Na₂HPO₄) solution.

Figure 11 shows the solution appearance before and after the test as well as the corroded specimen surfaces with rust. From the solution observations, it is found that the solution kept transparent until the Na₂HPO₄ ratio of 3.0%. This means that the amount of MgHPO₄ in the solution had been over its solubility when the Na₂HPO₄ ratio is more than 3.0%. However, though different from that obtained in Figures 4 and 8, the precipitation did not largely weaken the corrosion reaction.

Figure 11

Appearance of solutions of 3.0% (MgCl₂·6H₂O + x%Na₂HPO₄) before (a) and after (b) wet and dry cyclic corrosion, and specimens after wet and dry cyclic corrosion without removing rust.

According to the above results, high corrosion suppression effect was obtained in the 3.0% (NaCl + 2.0% Na₂HPO₄ + 0.5% CaCl₂) and 3.0% (CaCl₂ + 1.0% Na₂HPO₄). The corresponding solution concentrations of the possible CaHPO₄ in the solutions are respectively about 0.018 and 0.028%, which are very near the solubility of CaHPO₄ (0.02 g per 100 mL water at 298 K). Therefore, it is reasonable to consider that the best inhibition effect can be obtained at the saturated concentration of CaHPO₄ and much more Ca²⁺ or will cause the precipitation of CaHPO₄ at the early stage. Such early precipitated CaHPO₄ generally adheres on the steel surface in a porous state, which obstructs the formation of condensation and protective film. On the other hand, although the MgHPO₄ was also as insoluble with almost the same solubility (0.025 g per 100 mL water at 298 K), the increased MgCl₂ in solution might have helped to form protective film during the corrosion reaction.

Polarisation curves

Figure 12 shows several typical polarisation curves obtained in various solutions. The polarisation curves of the mild steel in NaCl and N/S are almost the same. In Figure 12(a), the corrosion potential became nobler when 2.0% Na₂HPO₄ is added into NaCl, and the corrosion current decreased due to the suppression of the anodic rather than the cathodic reaction. This result is different from that reported literature [1]. It indicates the adsorption of ions or the formation of compound containing iron and phosphorous on the anodic dissolution sites. The corrosion potential decreased further a little, when the ratio of Na₂HPO₄ increased, with slight increased suppression of cathodic reaction. In Figure 12(b), comparing to those in CaCl₂ or MgCl₂ solution, both corrosion potential and corrosion current largely decreased when adding Na₂HPO₄ into CaCl₂ or MgCl₂, with large suppression of cathodic reaction. These results indicate the formation of protective film containing phosphorous and calcium or magnesium to suppress cathodic reaction. According to Figure 12(c), the corrosion potential became nobler when adding MgCl₂·6H₂O to the (NaCl + 2.0% Na₂HPO₄) antifreeze, and the corrosion current decreased with the increase of MgCl₂·6H₂O due to both the suppression of cathodic and anodic reactions. When 0.5% CaCl₂ was added to the antifreeze of (NaCl + 2.0% Na₂HPO₄), both the corrosion potential and the corrosion current became much lower, which is much similar with that obtained at (N/S + 2.0% Na₂HPO₄) in Figure 12(d). In the both cases of (N/S + 2.0% Na₂HPO₄) and (NaCl + 2.0% Na₂HPO₄ + 0.5% CaCl₂), both their cathodic and anodic reactions, especially the cathodic reactions were suppressed. It indicates that the corrosion inhibition mechanism of the antifreeze of (N/S + 2.0% Na₂HPO₄) is due to the formation of a protective film containing P, Fe, Ca or Mg. The above polarisation curves have good correlation with the wet and dry cyclic test.

Figure 12

Polarisation curves of steel in different solutions with the total salt concentration of 3.0%.

XPS analysis

Figure 13 shows the XPS spectra of mild steel after the wet and dry test for 7 days in different solutions with the total salt concentration of 3.0%. The elements of Fe, Ca, P and O were detected from each of the surfaces. Comparing to other three, more Ca was detected on the surface tested in the solution containing (N/S + 2.0% Na₂HPO₄). The peak of Mg 2p and Fe 3p are overlapped on the surface corroded in the solution containing (NaCl + 2.0% Na₂HPO₄ + 2.0% MgCl₂·6H₂O). The main peak of P 2p lies at 131.8 eV, indicating the existence of on the surface. Therefore, it is evident that the elements of Fe, Ca, Mg, P and O have promoted the formation of protective film during corrosion.

Figure 13

XPS spectra of mild steel after the wet and dry test for 7 days in different solutions with the total salt concentration of 3.0%.

Considering that FeHPO₄, CaHPO₄ and MgHPO₄ have poor solubility in water, the isolate or the complex ones might have been formed as part of the protective film on the steel surface. Since direct and certain evidence of such compounds on steel surface had not been found, it is reasonable to consider that the suppression of corrosion should be due to the formation of a protective film containing Fe, O, P, Ca or Mg. Perhaps, the Fe, Ca or Mg atoms combined on steel can attract large ions to the surface, which can reduce the adhesion of chloride ions.

Capacity of dissolving ice

The capacities of dissolved ice of a mine salt (99.9% NaCl) and the developed antifreeze of (N/S + 2.7% Na₂HPO₄) are shown in Table 1. The mine salt is currently used as antifreeze. According to this table, the dissolved ice amount by the developed antifreeze is the same as that by the mine salt after 3.6 ks. After 21.6 ks, the value by the developed antifreeze was kept as 95% of that of mine salt. Therefore, there is no problem to use the new antifreeze with low corrosiveness instead of the mine salt. Of course, this antifreeze can also be used as snow or ice melting agent. Perhaps, the Na₂HPO₄ together with Ca²⁺ or Mg²⁺ can be used as a potential inhibitor in some other ways.

Table 1

Amount of dissolved ice by mine salt and the developed antifreeze with natural salt of NaCl and 2.7% Na₂HPO₄ at 268 K.

Salts (antifreezes)	Dissolved ice (g)
After 3.6 ks	After 21.6 ks
Mine salt (99.9%NaCl)	2.30	4.36
Developed antifreeze (N/S + 2.7%Na2HPO4)	2.30	4.16
Rate	100%	95%

Conclusions

To develop ecological and low-corrosive antifreezes, Na₂HPO₄ with different weight ratios was added to antifreezes of natural salt (N/S), NaCl, CaCl₂ and MgCl₂. The corrosion rate of mild steel was measured in the aqueous solution containing the antifreeze by a wet and dry cyclic test. The polarisation and XPS analyses were used to discuss the inhibition suppression mechanism. The following results are obtained:

The addition of Na₂HPO₄ into either of the antifreezes containing natural salt (N/S), NaCl, CaCl₂ or MgCl₂ can suppress the corrosion of mild steel (SPCC) in the 3.0 wt-% antifreeze aqueous solutions. The addition of CaCl₂ or MgCl₂ to the antifreeze of (NaCl + 2.0% Na₂HPO₄) significantly decreased the corrosion rate of steel.

The polarisation shows that the with Ca²⁺ or Mg²⁺ can suppress both the anodic and cathodic reactions but with larger effect on the cathodic reaction of the mild steel. The observation of solution and specimen as well as the XPS analysis indicates the formation of a protective film containing iron, phosphorous, oxygen and calcium or magnesium.