Journal Information
Download PDF
More article options
Original Article
DOI: 10.1016/j.jmrt.2018.05.029
Open Access
Electrochemical investigation on effect of sodium thiosulfate (Na2S2O3) and ammonium chloride (NH4Cl) on carbon steel corrosion
Prince Kumar Baranwal, Prasanna Venkatesh Rajaraman
Corresponding author

Corresponding author.
Department of Chemical Engineering, Indian Institute of Technology Guwahati, Guwahati 781039, India
This item has received

Under a Creative Commons license
Received 13 January 2018, Accepted 31 May 2018
Article information
Full Text
Download PDF
Figures (13)
Show moreShow less
Tables (2)
Table 1. Ecorr and icorr values attained from potentiodynamic polarization curves.
Table 2. Best fit EEC parameters obtained for carbon steel dissolution in different corrosive media; the values within bracket indicates the error %.
Show moreShow less

The effect of sodium thiosulfate (Na2S2O3) concentrations (0.01, 0.1 and 1M) on carbon steel corrosion behavior in 3.75M ammonium chloride (NH4Cl) solution was investigated. The potentiodynamic polarization results show that the icorr increases from 8.7×10−6Acm−2 to 7.2×10−4Acm−2 when thiosulfate (S2O32−) concentration increases from 0 to 1M. Electrochemical impedance spectroscopy measurements also mimic the same trend. However, it is also observed that the icorr value starts decreasing when S2O32− concentration equals or exceeds the NH4Cl concentration (Cl:S2O32−1:1) in the electrolyte solution. The polarization behavior and field-emission scanning electron microscopy images reveal that the carbon steel undergoes general corrosion at lower concentrations (0.01M S2O32− and 0.1M S2O32−) and, both pitting and general corrosion at higher concentration (1M S2O32−). Based on the experimental results, the mechanistic reaction pathway, which involves two dissolution paths for carbon steel dissolution is suggested.

Carbon steel
Ammonium chloride
Sodium thiosulfate
Electrochemical techniques
Corrosion mechanism
Full Text

In oil refinery industries, ammonium chloride (NH4Cl) corrosion has been reported to be one of the main causes of equipment and piping failures [1–3]. Severe corrosion failures result from aggressive species such as chlorides, sulfur, amines, salts, carbon dioxide and acids present in crude oil [1–9]. Among them, polythionates, especially thiosulfates (S2O32−) has been reported to be a well identified problem in oil refinery units, particularly in the sulfurous processing streams [4–6], in addition to pulp and paper industry [10,11] and nuclear power industry [12]. Of all the sulfur species, S2O32− is found to be one of the most intriguing, metastable polysulfur oxyanions, which can undergo both oxidation and reduction or disproportionation and decomposition [5,13,14] depending upon the corresponding electrochemical/chemical environment. Marcus et al. [15] developed potential-pH diagram for Fe in water containing various sulfur species including S2O32− at 25°C and 300°C.

Besides, S2O32− renders a thermodynamic basis for use as a substitute of noxious hydrogen sulfide (H2S) in laboratory scale corrosion studies of sour gas conditions [5,6]. This was basically proposed based on the following reactions, where the elemental sulfur generated via both electrochemical (Reaction 1) and chemical (Reaction 2) reactions, may further undergo disproportionation (Reaction 3)/reduction (Reaction 4) to produce H2S:

Thus, investigating the combined effect of S2O32− and chloride (Cl) ions on corrosion of various metal alloys has been of great interest. The corrosive effect of S2O32−–Cl ions [10–12,16–23] on various metal alloys have been reported. Wensley et al. [10] reported that polysulfides and S2O32− present in white liquor caused a detrimental effect on mild steel by degrading the passivating ability of the steel. It was observed that the sulfur adsorbed on the surface prevented the repassivation process in Cl solutions. X-ray photoelectron spectroscopy (XPS) studies [22] showed that sulfate ions are reduced on the metal electrode (M) surface as per the following reactions:

It was reported that Cl:S2O32− ratio is one of the most significant factors in enhancing pitting corrosion of steels in Cl media. Also, it was found that S2O32− ions are not corrosive enough when present alone, as it does not initiates the breakdown of the passive film formed on the steel surface [16,17,19,24]. However, these studies were conducted for other steel alloys and hence, understanding the corrosion mechanism of carbon steel in Cl–S2O32− system will be of interest in the present work.

Similarly, in the studies reported in the literature, NaCl solutions were mainly considered and to the best of author's knowledge, the effect of S2O32− ions in the presence of NH4Cl solution for carbon steel was not investigated yet. Although, cations will not participate in the electrochemical reactions directly, it may alter the anionic activity and, thereby affect the corrosion rate and mechanism. Many parameters such as pH of the electrolyte solution, hydrolysis of cations, hardness of cations, corrosion inhibitory effect of cations, formation and solubility of corrosion products in the presence of given cation will influence the corrosion rate as well as the mechanism [25–28]. Hence, the investigation of effect of sodium thiosulfate (Na2S2O3) on the corrosion behavior of carbon steel in NH4Cl solution is of particular interest in the present work.

The electrochemical techniques such as linear polarization, cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS) were performed to analyze the corrosion mechanism of carbon steel in NH4Cl solution containing S2O32− ions. In addition, other techniques such as X-ray diffraction (XRD), field emission scanning electron microscopy (FESEM) and energy dispersive X-ray spectroscopy (EDS) were employed to support our research findings.


All the measurements were carried out with carbon steel of following composition (wt.%): C (0.22–0.26) %; Si (0.11–0.14) %; Mn (1.03–1.06) %; P (0.04% maximum); S (0.03% maximum); Cr (0.03% maximum); and rest Fe. The electrolytes were prepared utilizing chemicals of analytical reagent grade (NH4Cl from Himedia, India; Na2S2O3 from Loba Chemie, India) and distilled water. The experiments were performed at natural pH of the solution: 4.22 for pure 3.75M (20wt.%) NH4Cl solution; 4.54, 5.32 and 6.16 respectively for 3.75M NH4Cl solution containing 0.01M, 0.1M and 1M Na2S2O3 respectively. The pH values of pure 0.01M, 0.1M and 1M Na2S2O3 solutions are 6.5, 7.57 and 8.99 respectively. The microscopy image of carbon steel sample was obtained using an optical microscope (Leica DM2500 M, Germany), which reveals the presence of ferrite, bainite and pearlite phases (Figure S1).

2.2Electrochemical measurements

The electrochemical measurements were performed using a potentiostat (PGSTAT 204, Switzerland), and a conventional three electrode system. Ag/AgCl (3M KCl) was used as the reference electrode and a Pt wire as the counter electrode. The working electrode was cylindrical carbon steel of 8mm diameter with exposed surface area of 0.5cm2 to the electrolyte solution. The solution temperature was controlled at (25.5±1)°C. During the experiments, the corrosion cell was housed inside a faraday cage to minimize the external noises. Prior to each experiment, the surface of the working electrode was pretreated by grounding sequentially with 220, 320, 600 and 1000 grades of silicon carbide (SiC) paper, followed by polishing with 1.0μm and 0.3μm alumina powder. Then, the electrode was rinsed with distilled water, and ultrasonicated to remove any particle adhered on the specimen surface. The electrode was finally rinsed and dried using tissue paper. The sample was pretreated to ensure that the results obtained are repeatable and reproducible. All the experiments were performed at least thrice to check the repeatability and reproducibility of the results. On the whole, all the experiments were carried out in naturally aerated stagnant electrolyte solutions.

2.2.1Polarization measurements

The polarization measurements were performed once the open circuit potential (OCP) reached a stable value after 1h (refer Figure S2). Potentiodynamic polarization was employed by sweeping the potential from −250mV (vs. Ag/AgCl) to +500mV (vs. Ag/AgCl) w.r.t OCP, at a scan rate of 1mVs−1. Potentiostatic polarization measurements were also performed at various overpotentials in order to understand the pitting behavior of carbon steel.

2.2.2Cyclic voltammetry

The cyclic voltammogram of carbon steel in different concentrations of corrosive media were obtained by sweeping the potential linearly from starting potential, −1.20V (vs. Ag/AgCl) to −0.20V (vs. Ag/AgCl) and then reversed back to the starting potential at a potential scan rate of 50mVs−1. Scanning was also performed from −0.20V (vs. Ag/AgCl) to −1.20V (vs. Ag/AgCl) and vice versa, and no significant change in peak position and intensity was observed.

2.2.3Electrochemical impedance spectroscopy

Before each experiment, the working electrode (specimen) surface was exposed to open circuit for a period of 1h. The impedance responses were obtained by employing an amplitude signal of 10mV (rms) at the corresponding OCP value, within the frequency range of 10kHz to 10mHz, with 7 frequencies per decade. The impedance results were further analyzed with electrical equivalent circuit (EEC) model using Nova software (Metrohm Autolab, Switzerland). The EIS data obtained was validated with Kramers Kronig transform in order to ensure the stability, linearity and causality of the system.

2.3Surface analysis using immersion tests

The surface morphology of carbon steel was inspected via field emission scanning electron microscope (FESEM) [Zeiss, Sigma, Germany], in order to confirm the type of corrosion attack. The steel specimen was immersed in nitric acid (70%, as received), and rinsed with distilled water. Further, the specimen was polished with SiC paper and abrasive alumina powder. Subsequently, the sample was rinsed with distilled water, ultrasonicated, dried and then introduced in the respective corrosive media for 12h. At the end of 12h, the specimen was taken out of the system, washed with distilled water and dipped in nitric acid to remove the corrosion layer formed on the surface. The steel specimen was then rinsed with distilled water, air dried and finally, the micrograph images were taken.

Energy dispersive X-ray spectroscopy (EDS) was performed to elucidate the elemental composition of the corrosion products using FESEM [Zeiss, Sigma, Germany]. The steel specimen was polished in a similar fashion as above, and then submerged in the system of interest for 72h.

2.4X-ray diffraction measurement

X-ray diffraction (XRD) was used to characterize the corrosion product formed on the carbon steel surface submerged in solutions containing NH4Cl and Na2S2O3. XRD patterns were obtained using TTRAX-III (Rigaku) diffractometer (Japan), with Cu Kα (λ=1.5406Å) radiation (50kV, 100mA). The steel specimen was exposed to the respective solution of interest for a period of 72h, so that sufficient corrosion product is obtained for XRD detection. The corrosion product formed were scrubbed out of the specimen, dried, and placed on a quartz plate to provide a planar surface in the diffractometer. The XRD pattern was acquired over a range of 10°–70°, with a scan rate of 4° per minute.

3Results and discussion3.1Potentiodynamic polarization

In the current investigation, the NH4Cl concentration was fixed at 3.75M and the concentration of Na2S2O3 was varied from 0.01M to 1M, to understand the combined effect on carbon steel corrosion. The concentrations used in the study are usually encountered in the refineries, as reported in the literatures [1–3,16,17,19,22,23]. Fig. 1 shows the potentiodynamic polarization behavior of carbon steel in NH4Cl solution containing various concentrations of S2O32− ions. It is to be noted here that the molar ratio of Na2S2O3 to NH4Cl is less than one for all the cases. Table 1 shows Ecorr and icorr values, which were obtained from these polarization curves by means of corrosion fit analysis, using Nova 1.10.4 software. In some cases, where the polarization curves do not possess linear range long enough for extrapolation, estimating Tafel slopes were difficult and the reported values might not be accurate. However, the overall trend will not be changed. The polarization curves clearly depict a considerable effect of S2O32− on the electrochemical behavior of carbon steel in NH4Cl system. One of the main observations from these polarization curves is that Ecorr value shifts toward negative potential with the addition of 0.01M S2O32− to the NH4Cl solution. However, the value of Ecorr keeps increasing to a more positive potential with further increase in S2O32− concentration. Similar kind of behavior is also reported for carbon steel in caustic soda–thiosulfate system [10], and other systems as well [29,30]. On the other hand, the icorr value is increasing from 8.7×10−6Acm−2 to 7.2×10−4Acm−2 when the S2O32− concentration increases from 0M to 1M. It implies that the addition of S2O32− ions to the NH4Cl solution accelerates the corrosion rate of carbon steel.

Fig. 1.

Potentiodynamic polarization curves of carbon steel corroded in 3.75M NH4Cl solution containing various concentrations of S2O32−. Scan rate=1mVs−1.

Table 1.

Ecorr and icorr values attained from potentiodynamic polarization curves.

Electrolyte concentration  βa (mVdec−1βc (mVdec−1Ecorr (V)  icorr (Acm−2
Pure 3.75M NH4Cl  137.9  62.9  −0.683  8.7×10−6 
3.75M NH4Cl+0.01M S2O32−  142.7  362.9  −0.743  1.3×10−4 
3.75M NH4Cl+0.1M S2O32−  148.1  375.2  −0.719  3.8×10−4 
3.75M NH4Cl+1M S2O32−  160.4  326.4  −0.655  7.2×10−4 
Pure 0.01M S2O32−  190.6  159.8  −0.665  2.4×10−5 
Pure 0.1M S2O32−  146.5  315.6  −0.743  4.1×10−5 
Pure 1M S2O32−  132.2  328.3  −0.789  7.1×10−5 

The substantial increase in the cathodic corrosion current density could also be observed because of the presence of S2O32− ions in the NH4Cl solution. It would be attributed to the role of these species in either of the reduction reaction (1) or (4) as mentioned in the introductory section. The solution pH [31,32], electrochemical potential [14] and nature of the surface being corroded (passivated/non-passivated) [12,23,33,34] determines the typical cathodic reactions occurring on metal surface. Besides, the elemental sulfur produced may further react with Fe2+ to form FeS film, thereby accelerating the corrosion process [23]. Similar effect is also reported for NaCl solution [23], i.e. the addition of Na2S2O3 to NaCl solution influences both the anodic and cathodic reactions.

It is also interesting to see that the polarization curves (Fig. 1) show non-linear behavior, which is more evident especially at 0.1M and 1M Na2S2O3 concentrations. In particular, 1M Na2S2O3 solution shows constant current regime, for a potential range from −0.58V to −0.45V (vs. Ag/AgCl), which indicates the formation of pseudo-passive layer (the current remains more or less constant at 1.1mAcm−2) on the carbon steel surface. This layer then finally breaks down at a breakdown potential (Eb) of −0.449V (vs. Ag/AgCl), as evident by the increase of current. The pseudo-passive layer comprises of a mixture of sulfides and oxides. However, the major constituent could be sulfides as the formation of the same is thermodynamically more feasible than the formation of oxides [34].

The reduction of S2O32− to sulfur is more likely to happen on non-passivated surfaces [12,22,34]. In NH4Cl solutions, the surface is not completely covered with an oxide film, which triggers the formation of sulfur and thereby FeS layer on the carbon steel surface. The formation rate of FeS will probably be increasing with S2O32− concentration, which results in the formation of pseudo-passive layer as evident by the constant current regime in polarization curve. At higher anodic potential, the pseudo-passive layer breaks down indicating the susceptibility of carbon steel to localized corrosion. In general, the sulfide film is more porous and non-protective comparing to that of oxide film [34]. Thus, this pseudo-passive layer could not completely protect the underlying metal, and as a result, general corrosion also occurs simultaneously at higher Na2S2O3 concentration. Simultaneous occurrence of general and pitting corrosion in acidic and neutral solution is also reported for other metals [35,36]. However, at lower Na2S2O3 concentrations, the corrosion is of only general type as observed in the case of pure NH4Cl solutions [37].

As the Na2S2O3 concentration varies from 0.01M to 1M, the pH value increases from 4.54 to 6.16. According to Pourbaix diagram, the passivation due to Fe1+XS/FeS is thermodynamically feasible in this pH range of 4–8 [15,38,39]. However, the pseudo-passive layer is formed only at higher concentration of Na2S2O3. Thus, these results clearly depict that concentration of Na2S2O3 plays a significant role in determining the nature of the corrosion process.

In order to understand further, whether the observed behavior at higher Na2S2O3 concentrations is due to the presence of S2O32− ions alone or because of the combined effect of S2O32− and Cl ions, the polarization experiments were carried out in pure Na2S2O3 solutions (0.01M–1M) and the results are shown in Fig. 2. For an increase in Na2S2O3 concentration, the curves indicate a marked increase in the current density from 2.4×10−5Acm−2 to 7.1×10−5Acm−2 with decrease in Ecorr value from −0.665V to −0.789V (vs. Ag/AgCl). The higher current density in pure S2O32− solutions (2.4×10−5Acm−2 at 0.01M Na2S2O3 vs. 8.7×10−6Acm−2 at 3.75M NH4Cl) indicates that S2O32− ions are more aggressive compared to that of Cl ions in corroding carbon steel surface. Besides, it does not show constant current regime for long potential range as observed in case of 1M S2O32− in 3.75M NH4Cl solution (Fig. 1). However, a non-linear behavior is observed in the curve at 1M S2O32− concentration, which may be due to the formation of a porous and non-protective sulfide layer on the surface. This implies that susceptibility to pitting enhances only in presence of both Cl and S2O32− species. Thus, when the solution contains only NH4Cl or Na2S2O3, carbon steel undergoes general corrosion, whereas when the solution contains both the species, then carbon steel undergoes either general or pitting corrosion depending on the concentration of both the species.

Fig. 2.

Potentiodynamic polarization curves of carbon steel corroded in various concentrations of pure S2O32− solutions. Scan rate=1mVs−1.

The polarization experiments were further carried out at different molar ratios of Na2S2O3 to NH4Cl: <1, 1 and >1 and the results are shown in Fig. 3. The pH of the corresponding solutions varied from 6.17 to 7.13. It is to be noted that the concentration of NH4Cl is fixed at 0.93M (5wt.%) instead of 3.75M, and the concentration of Na2S2O3 is varied to get these desired ratios. At 3.75M NH4Cl, achieving ratio 1 and >1 is difficult as the solution is saturated. It appears from these results that initially the corrosion rate of the carbon steel in NH4Cl solution increases markedly to a maximum on addition of Na2S2O3 (as observed in 3.75M NH4Cl solution, Fig. 1) and then starts decreasing when the ratio of Cl and S2O32− species becomes 1:1 or higher than 1 as shown in Fig. 4.

Fig. 3.

Potentiodynamic polarization curves of carbon steel corrosion in various Cl to S2O32−concentration. Scan rate=1mVs−1.

Fig. 4.

Ecorr and icorr values acquired for various ratio of Cl:S2O32−.

The reason for icorr showing a maxima at particular S2O32− concentration could be explained as follows: In pure NH4Cl solution, the surface coverage of oxide layer is very low [37]. Na2S2O3, being an oxidizer, increases the oxidation state of the carbon steel when added to NH4Cl solution and thereby increases the dissolution rate at lower concentrations. However, higher concentrations of S2O32− leads to the formation of thick layer of FeS on the carbon steel surface, which restricts the Cl ions or O2 molecules to diffuse through it, and thereby finally lowers the dissolution rate. Thus, both Cl and S2O32− has a pronounced effect on the corrosion mechanism.

3.2Potentiostatic experiments

The potentiostatic experiments were also carried out for carbon steel in the electrolyte solution containing 3.75M NH4Cl and 1M Na2S2O3 in order to understand the pitting behavior at two different overpotentials: (a) 50mV below Eb, and (b) 50mV above Eb. The results obtained are shown in Fig. 5. When the overpotential is below the pitting potential, the current is constant (1.2×10−4Acm−2) with time, which implies that there is no formation of passive layer on the electrode surface. However, when the overpotential is above the pitting potential, the current decreases and then increases sharply to 9×10−3Acm−2 indicating the growth of pit with time. After 1600s, the current value does not increase significantly (saturates at 9.3×10−3Acm−2), which implies that the growth of pit is impeded by the deposition of corrosion products inside the pit.

Fig. 5.

Potentiostatic measurements of carbon steel in 3.75M NH4Cl solution containing 1M S2O32− at overpotentials: (a) 50mV below pitting potential (Eb): −0.499V (vs. Ag/AgCl), and (b) 50mV above pitting potential (Eb): −0.399V (vs. Ag/AgCl).

3.3Cyclic voltammetry

Fig. 6(a)–(d) shows the cyclic voltammograms obtained for carbon steel in NH4Cl media containing S2O32− of varied concentrations. The voltammograms are interpreted in terms of Fe as the content of other elements in carbon steel is very low. In pure 3.75M NH4Cl solution, two anodic peaks A1 and A2 appeared at −0.581V and −0.393V (vs. Ag/AgCl) respectively. The peak A1 corresponds to oxidation of Fe to Fe2+ and the peak A2 corresponds to oxidation of Fe2+ to Fe3+. Besides, the high anodic current (A4) is observed at higher anodic potentials for all the cases, which is due to the combined effect of transpassive dissolution and oxygen evolution. In the cathodic side, one peak C1 appears at −0.991V (vs. Ag/AgCl) which is more likely to be the conjugate peak of A1. The peak intensity of C1 for all the cases comes out to be slightly more than its counter peak A1, which suggests that most of the species being oxidized are completely reduced.

Fig. 6.

Cyclic voltammograms for carbon steel immersed in: (a) pure 3.75M NH4Cl, (b) 3.75M NH4Cl+0.01M S2O32− solution, (c) 3.75M NH4Cl+0.1M S2O32− solution, and (d) 3.75M NH4Cl+1M S2O32−. Scan rate=50mVs−1.

On addition of 0.01M Na2S2O3 to the NH4Cl solution, the peak A1 shifts toward more negative potential to −0.644V (vs. Ag/AgCl) and one more peak A3 appears close to A1 peak at −0.586V (vs. Ag/AgCl), which may be due to competitive diffusion/adsorption of both Cl and S2O32− ions on the surface. The peak A3 is more likely to be the oxidation of Fe to form FeS on the electrode surface along with Fe oxide/chloride layer. Thus, both reactions occur simultaneously on the electrode surface. In the cathodic region, in addition to C1, one more peak C2 appears at −0.759V (vs. Ag/AgCl), which is more likely to be the conjugate peak of A3, i.e. the reduction of FeS. Thus, the electrode surface might be covered with mixture of oxide and sulfide layers. When the concentration of S2O32− in NH4Cl solution is increased to 0.1M, the two peaks A1 and A3 merge together, and form as one broad peak A1 that appears at −0.653V (vs. Ag/AgCl). It is indicating the increase in aggressiveness of the system due to the S2O32− ions. The peak A2 is not observed for these two cases, which may be due to the reason that, the generation of Fe3+ is overshadowed by transpassive and oxygen evolution reactions.

However, in NH4Cl solution containing 1M S2O32−, there appears a region of lower and constant current density from −0.83V to −0.65V (vs. Ag/AgCl), with the disappearance of A1 peak. The peak A2, which is arising at a comparatively higher overpotential, −0.466V (vs. Ag/AgCl) corresponds to oxidation of both oxides and sulfides. This suggests that a pseudo-passive layer is formed on the entire surface. This also corroborates with the results obtained from the polarization measurements (Fig. 1) where formation (insignificant increase in current density) and breakdown of passive layer (significant increase in current density) is observed on the steel surface. The occurrence of positive hysteresis in the CV also indicates the possibility of pitting corrosion on carbon steel surface.

3.4Electrochemical impedance spectroscopy

The impedance spectra obtained for various concentration of S2O32− in NH4Cl solution are shown in Fig. 7(a)–(c). Although two time constants (capacitance loops) were observed for all the cases, the two time constants were not clearly distinguishable in some cases, as it is merged together. The electrical equivalent circuit (EEC) showed in Fig. 7(d) is employed to analyze the impedance data quantitatively. Here, Rs represents the solution resistance, Q1 and R1 corresponds to constant phase element (CPE) associated with electrical double layer and charge transfer resistance respectively. The other parameters Q2 and R2 corresponds to capacitance and resistance associated with reduction reactions respectively, while Q3 and R3 is associated with diffusion phenomenon and chemical reactions occurring in bulk solution respectively. The EEC parameters obtained are shown in Table 2. It is clearly evident that Q1 is increasing and R1 is decreasing with increase in the S2O32− concentration. Thus, the total impedance associated with Faradaic reaction decreases as the Na2S2O3 concentration increases. Instead of capacitor, CPE is employed in EEC model to quantify the non-idealities observed in the capacitor behavior [40–44]. The electrical double layer capacitance is calculated from the CPE element using the following Brug's equation and the values are shown in the same Table 2.

where Cdl is the double layer capacitance, Y0 and n are the CPE parameters, Rs and Rct are the solution resistance and charge transfer resistance respectively. The values of Cdl obtained for the present system are high. However, similar values were reported for carbon steel in H2S environment [44,45] and for other systems as well [46–48]. The other capacitive element Q2 is increasing, while the resistance R2 is decreasing with increase in S2O32− concentration. It indicates the fact that the cathodic reaction rate increases with S2O32− concentration. The capacitance values are higher compared to Cdl. It might be attributed to the adsorption of various corrosion products on the carbon steel surface [44]. The exponent value of Q3 is nearly 0.5 in all these cases, which signifies the diffusion effect. The total impedance arising from these parameters is decreasing with S2O32− concentration. This indicates that the corrosion rate increases with S2O32− concentration as observed in the polarization measurements.

Fig. 7.

EIS data obtained from carbon steel corrosion in 3.75M NH4Cl solution for varied concentrations of S2O32− (a) Nyquist plot, (b) Bode modulus plot and (c) Bode phase plot; continuous line indicates the EEC fitting of corresponding impedance data. (d) Equivalent circuit used to fit the EIS data.

Table 2.

Best fit EEC parameters obtained for carbon steel dissolution in different corrosive media; the values within bracket indicates the error %.

EEC parameters  3.75M NH4Cl  3.75M NH4Cl+0.01M S2O32−  3.75M NH4Cl+0.1M S2O32−  3.75M NH4Cl+1M S2O32− 
Rscm20.8 (1%)  0.7 (1.5%)  0.8 (1.1%)  0.9 (1%) 
Q1Y10−1cm−2sn4.7×10−4 (3.6%)  1.1×10−2 (2.2%)  1.2×10−2 (2.3%)  1.4×10−2 (3.3%) 
Q1n1  0.8 (0.5%)  0.85 (0.6%)  0.85 (0.6%)  0.86 (0.9%) 
Cdl (Fcm−26.4×10−5  5×10−3  5.1×10−3  6.6×10−3 
R1cm210 (1.1%)  5.5 (1.9%)  4.3 (1.3%)  3.7 (1.7%) 
Q2Y02−1cm−2sn7.5×10−2 (1.3%)  7.7×10−2  1.4×10−1 (1.3%)  1.6×10−1 (2.4%) 
Q2n2  0.6 (0.6%)  0.6 (1.5%)  0.68 (0.8%)  0.65 (1.5%) 
R2cm2778.7 (3.8%)  321.5 (1.8%)  221 (2.2%)  153.2 (1%) 
Q3Y03−1cm−2sn2.4×10−2 (2.9%)  5.1×10−2 (4%)  5.4×10−2 (3%)  5.5×10−2 (2.1%) 
Q3n3  0.51 (1%)  0.52 (1.6%)  0.51 (0.9%)  0.54 (0.7%) 
R3cm218.6 (1.1%)  14.8 (1.5%)  7.5 (1.2%)  7 (1%) 
3.5FESEM and EDS

The surface morphology of carbon steel after immersion in NH4Cl solution containing S2O32− of various concentrations are shown in Fig. 8(a)–(c). As expected from the polarization behavior, uniform corrosion is observed at lower S2O32− concentrations (0.01M and 0.1M), as observed in pure NH4Cl solution [37]. While pits are clearly observed at higher S2O32− concentration (1M). It supports our observation from polarization measurements that a pseudo-passive layer is formed when the concentration of Cl and S2O32− ions in solution are higher, which is attacked further by Cl ions resulting in localized corrosion.

Fig. 8.

FESEM micrographs of carbon steel after immersion in: (a) 3.75M NH4Cl+0.01M S2O32− solution, (b) 3.75M NH4Cl+0.1M S2O32− solution, and (c) 3.75M NH4Cl+1M S2O32− solution, for a period of 12h; Square indicates the pit formed on the surface.

Fig. 9(a)–(d) shows the results obtained from EDS analysis, carried out on corrosion products formed on carbon steel surface, at different concentrations of S2O32− in 3.75M NH4Cl solution. In pure NH4Cl solutions, the corrosion film is mainly oxides and chlorides of Fe. When Na2S2O3 is added to the NH4Cl solution, the sulfur peak appears indicating the formation of sulfide layer on the surface. Thus, the corrosion products are oxides and chlorides of Fe in Cl only solution and a mixture of oxides, chlorides and sulfides in Cl–S2O32− solutions. EDS analysis were also carried out on the corrosion product formed inside the pit. The result confirms the presence of both sulfur and chlorine in the pit (Figure S3). EDS mapping of carbon steel after reaction with 3.75M NH4Cl–1M Na2S2O3 is also presented in Fig. 10(a)–(d). The result shows that the species S, Cl and O are present on the carbon steel surface as well as on the pit. These results imply that both sulfur and chlorine species took part in the formation of stable pits as reported by other researchers [12,34]. Also, the species S and O are more concentrated inside the pit. It indicates that the pits are covered with corrosion products as observed in the potentiostatic measurements.

Fig. 9.

EDS analysis of corrosion products formed on the surface of carbon steel after immersion in: (a) pure 3.75M NH4Cl, (b) 3.75M NH4Cl+0.01M S2O32− solution, (c) 3.75M NH4Cl+0.1M S2O32− solution, and (d) 3.75M NH4Cl+1M S2O32− solution, for a period of 72h.

Fig. 10.

EDS mapping of carbon steel exposed to 3.75M NH4Cl+1M S2O32− solutions for a period of 72h.


The crystalline phases of corrosion film formed on carbon steel surface in NH4Cl–Na2S2O3 solutions were identified by XRD spectra and the results are presented in Fig. 11. The peaks were identified using JCPDS software. XRD spectra exhibits various crystalline phases including α-FeOOH, FeCl3, FeS, Fe2O3 depending upon the concentration of S2O32− in NH4Cl solution. The peaks corresponding to FeS did not appear in pure NH4Cl solution as expected and the corrosion film comprises of only FeOOH, FeCl3 and Fe2O3. Although, the dissolution of carbon steel in NH4Cl solution occurs via Fe2+, the oxidation of these products to Fe3+ is more common in non-deaerated solution containing dissolved oxygen or during its exposure to atmospheric air. The appearance of FeS peaks in the recorded XRD spectra for NH4Cl–S2O32− solutions clearly depicts the formation of FeS layer on the carbon steel surface. This is also corroborating our results obtained from other techniques employed in the present investigation.

Fig. 11.

XRD pattern of corrosion products obtained after carbon steel corroded in 3.75M NH4Cl solution containing S2O32− of varied concentrations. Peak label indicates: +, (FeOOH); −, (FeCl3); *, (FeS); &, (Fe2O3).

3.7Dissolution mechanism of carbon steel in S2O32−–Cl solutions

The pitting corrosion is reported for various metal alloys in Cl only solutions [12,49–51] and Cl–S2O32− solutions [12,16,52]. Point defect model (PDM) is widely employed to explain the pitting corrosion of metal alloys [53,54]. As PDM could be applied only when strong passive film is present on the carbon steel surface, the following mechanistic reaction pathway is suggested to explain the features observed in electrochemical and non-electrochemical experiments.

The following anodic reaction occurs in NH4Cl solution (corresponds to peak A1 in CV).

Besides, the following chemical reactions also occur in NH4Cl solution.
The dissolution–precipitation reactions in presence of S2O32− ions (corresponds to peak A3 in CV) are given by the following anodic reaction.
In the above dissolution reaction, H2S is regenerated and further reacts with Fe2+ to form FeS.

The possible cathodic reactions occurring in the system are,

The peaks corresponding to the above cathodic reactions were not observed in CV. It is probably because these reactions were buried under the other major reduction process of iron oxides and sulfides. The overall reaction network is presented in Fig. 12.

Fig. 12.

The reaction network of carbon steel dissolution in NH4Cl–Na2S2O3 solutions.

In case of carbon steel in NH4Cl solution, the corrosion layer does not completely block the carbon steel surface as observed from polarization measurement. Consequently, general corrosion occurs as reported in our earlier studies [37]. Thus, based upon these results, it could be envisaged that in Cl only solutions, an oxide-chloride layer could be formed on the carbon steel surface but not as a continuous layer as shown in Fig. 13a.

Fig. 13.

Schematic representation of dissolution mechanism of carbon steel in: (a) pure NH4Cl solution, (b) NH4Cl solution containing lower concentration of S2O32−, and (c) NH4Cl solution containing 1M S2O32−; *, ^, #, † indicates species O2, Cl, H2O and S2O32−, respectively.

With the addition of lower concentrations of S2O32−, FeS could be precipitated on the carbon steel surface without blocking the surface layer completely as shown in Fig. 13b. As the S2O32−addition enhances the dissolution rate also, the formation of FeS would enhance the anodic reaction rate by acting as a reduction site [23,34]. Further, the pseudo-passive layer forms on the carbon steel surface only in the presence of both S2O32− and Cl ions, especially at higher concentrations. This implies that the pseudo-passive layer formed on the carbon steel surface is a mixture of oxides, chlorides and sulfides as evident from EDS and XRD results. Also, CV results confirm that the redox reactions due to both S2O32− and Cl ions are occurring on the carbon steel surface, unlike in pure NH4Cl solutions.

The FeS/FeO/FeCl2 could also undergo partial oxidation (corresponds to peak A2 in CV) as per the following equation [32,34].

Further, these iron oxides dissolved in the solution via transpassive reaction along with oxygen evolution (corresponds to peak A4 in CV).

Thus, the dissolution of Fe in Cl–S2O32− solution occurs via two paths as shown in Fig. 12. As observed from polarization measurements, the corrosion rate in pure Na2S2O3 is more comparing to the corrosion rate in pure NH4Cl solutions. Thus, the dissolution rate via path B would be higher. Besides, the concentration of these individual ions would also play a critical role in determining the nature and the rate of the process occurring at metal-electrolyte solution interface. For example, when the concentration of the S2O32− is increased to 1M (the ratio of S2O32− to Cl is still less than one), the rate of dissolution via reaction path B would enhance further and form a complete layer consisting of both oxides and sulfides on the surface as shown in Fig. 13c. Subsequently, Cl ions penetrates into this layer and causes pitting corrosion as explained before. Due to the pseudo-passivity of film formed on the carbon steel surface, general corrosion also occurs together. When S2O32− concentration equals or exceeds the Cl ion concentration, the decrease in current density is observed from polarization measurements. It is probably due to the formation of thicker sulfide layer formed on the electrode surface, which limits the diffusion of active species.


The following conclusions are drawn from this present study:

  • 1.

    The corrosion rate increases first with the increase in S2O32− concentration in NH4Cl solution. However, further increase in S2O32− concentration decreases the corrosion rate. It would be attributed to formation of thicker film on carbon steel surface when the Na2S2O3 concentration is higher.

  • 2.

    The nature of corrosion (uniform or pitting corrosion) is determined by the concentration of S2O32− ions in Cl solution. At lower concentrations, general corrosion is observed, while at higher concentrations, simultaneous occurrence of pitting and general corrosion is observed.

  • 3.

    EDS and XRD results reveals that the corrosion products are mainly FeCl3, Fe2O3 and FeOOH in NH4Cl solution, while FeS was formed in addition to oxides and chlorides of Fe with the addition of Na2S2O3.

  • 4.

    The dissolution of Fe occurs via two different paths, and both involves electrochemical and chemical steps. The concentration of both NH4Cl and Na2S2O3 determines the rate of these steps.

Conflicts of interest

The authors declare no conflicts of interest.


This work was supported by DST–SERB, India (SERB/F/1365/2014-15). We acknowledge DST–FIST, Government of India (SR/FST/PSII-020/2009) for XRD analysis and Central Instruments Facility of Indian Institute of Technology, Guwahati, India for providing FESEM Spectroscopy facility.

Appendix A
Supplementary data

The following are the supplementary data to this article:

O. Forsen, J. Aromaa, M. Tavi
Corrosion resistance of different materials in dilute ammonium chloride – bearing environments
Corros Sci, 35 (1993), pp. 297-301
P.P. Alvisi, V.D.F.C. Lins
Acid salt corrosion in a hydro-treatment plant of a petroleum refinery
Eng Fail Anal, 15 (2008), pp. 1035-1041
K. Toba, M. Ueyama, K. Kawano, J. Sakai
Corrosion of carbon steel and alloys in concentrated ammonium chloride solutions
Corrosion, 68 (2012), pp. 1049-1056
H.H. Horowitz
Chemical studies of polythionic acid stress corrosion cracking
Corros Sci, 23 (1983), pp. 353-362
S. Tsujikawa, A. Miyasaka, M. Ueda, S. Ando, T. Shibata, T. Haruna
Alternative for evaluating sour gas resistance of low-alloy steels and corrosion-resistant alloys
Corrosion, 49 (1993), pp. 409-419
M. Kappes, G.S. Frankel, N. Sridhar, R.M. Carranza
Reaction paths of thiosulfate during corrosion of carbon steel in acidified brines
J Electrochem Soc, 159 (2012), pp. C195-C204
F.F. Eliyan, F. Mohammadi, A. Alfantazi
An electrochemical investigation on the effect of the chloride content on CO2 corrosion of API-X100 steel
Corros Sci, 64 (2012), pp. 37-43
S. Freitas, M.M. Malacarne, W. Romão, G.P. Dalmaschio, E.V.R. Castro, V.G. Celante
Analysis of the heavy oil distillation cuts corrosion by electrospray ionization FT-ICR mass spectrometry, electrochemical impedance spectroscopy, and scanning electron microscopy
Fuel, 104 (2013), pp. 656-663
F.F. Eliyan, A. Alfantazi
Mechanisms of corrosion and electrochemical significance of metallurgy and environment with corrosion of iron and steel in bicarbonate and carbonate solutions – a review
Corrosion, 70 (2014), pp. 880-898
D.A. Wensley, R.S. Charlton
Corrosion studies in kraft white liquor: potentiostatic polarization of mild steel in caustic solutions containing sulfur species
Corrosion, 36 (1980), pp. 385-389
T. Laitinen
Localized corrosion of stainless steel in chloride, sulfate and thiosulfate containing environments
Corros Sci, 42 (2000), pp. 421-441
D. Xia, S. Song, R. Zhu, Y. Behnamian, C. Shen, J. Wang
A mechanistic study on thiosulfate-enhanced passivity degradation of Alloy 800 in chloride solutions
Electrochim Acta, 111 (2013), pp. 510-525
R.J. Tykodi
In praise of thiosulfate
J Chem Educ, 67 (1990), pp. 146-149
T. Hemmingsen
The electrochemical reaction of sulphur-oxygen compounds – Part I. A review of literature on the electrochemical properties of sulphur/sulphur-oxygen compounds
Electrochim Acta, 37 (1992), pp. 2775-2784
P. Marcus, E. Protopopoff
Potential-pH diagrams for adsorbed species: application to sulfur adsorbed on iron in water at 25°C and 300°C
J Electrochem Soc, 137 (1990), pp. 2709-2712
H.S. Kuo, H. Chang, W.T. Tsai
The corrosion behaviour of AISI 209 stainless steel in thiosulfate ion containing saturated ammonium chloride solution
Corros Sci, 41 (1999), pp. 669-684
R.C. Newman, H.S. Isaacs, B. Alman
Effects of sulfur compounds on the pitting behaviour of type 304 stainless steel in near-neutral chloride solutions
Corrosion, 38 (1982), pp. 261-265
R.C. Newman, W.P. Wong, H. Ezuber, A. Garner
Pitting of stainless steels by thiosulfate ions
Corrosion, 45 (1989), pp. 282-287
N.J. Laycock
Effects of temperature and thiosulfate on chloride pitting of austenitic stainless steels
Corrosion, 55 (1999), pp. 590-595
D. Tromans, L. Frederick
Effect of thiosulfate on crevice corrosion of stainless steels
Corrosion, 40 (1984), pp. 633-639
R. Roberge
Effect of the nickel content in the pitting of stainless steels in low chloride and thiosulfate solutions
Corrosion, 44 (1988), pp. 274-280
C.D. Thual, D. Costa, W.P. Yang, P. Marcus
The role of thiosulfates in the pitting corrosion of Fe-17Cr alloys in neutral chloride solution: electrochemical and XPS study
Corros Sci, 39 (1997), pp. 913-933
H.M. Ezuber
Influence of temperature and thiosulfate on the corrosion behaviour of steel in chloride solutions saturated in CO2
Mater Des, 30 (2009), pp. 3420-3427
P.R. Roberge, S. Wang, R. Roberge
Stainless steel pitting in thiosulfate solutions with electrochemical noise
Corrosion, 52 (1996), pp. 733-737
G.I. Ogundele
The effect of cation on the corrosion of carbon and stainless steels in differing chloride environments
Corrosion, 45 (1989), pp. 981-983
K. Otani, M. Sakairi
Effects of metal cations on corrosion of mild steel in model fresh water
Corros Sci, 111 (2016), pp. 302-312
S. Das, P.K. Baranwal, P.V. Rajaraman
Effect of soft cations on carbon steel corrosion in chloride media
Corros Rev, (2018), pp. 1-9
F.F. Eliyan, A. Alfantazi
On the theory of CO2 corrosion reactions – investigating their interrelation with the corrosion products and API-X100 steel microstructure
Corros Sci, 85 (2014), pp. 380-393
E.G. Webb, R.C. Alkire
Pit initiation at single sulfide inclusions in stainless steel I. Electrochemical microcell measurements
J Electrochem Soc, 149 (2002), pp. B272-B279
F.F. Eliyan, E.S. Mahdi, A. Alfantazi
Electrochemical evaluation of the corrosion behaviour of API-X100 pipeline steel in aerated bicarbonate solutions
Corros Sci, 58 (2012), pp. 181-191
R.J. Biernat, R.G. Robins
High temperature potential/pH diagrams for the sulfur-water system
Electrochim Acta, 14 (1969), pp. 809-820
R.J. Biernat, R.G. Robins
High-temperature potential/pH diagrams for the iron-water and iron-water-sulfur systems
Electrochim Acta, 17 (1972), pp. 1261-1283
M. Kappes, G.S. Frankel, N. Sridhar, R.M. Carranza
Corrosion behaviour of carbon steel in acidified, thiosulfate containing brines
Corrosion, 68 (2012), pp. 872-884
L. Choudhary, D.D. Macdonald, A. Alfantazi
Role of thiosulfate in the corrosion of steels: a review
Corrosion, (2015), pp. 1-65
B. Zaid, D. Saidi, A. Ben Zaid, S. Hadji
Effects of pH and chloride concentration on pitting corrosion of AA6061 aluminium alloy
Corros Sci, 50 (2008), pp. 1841-1847
M.S. Darmawan
Pitting corrosion model for reinforced concrete structures in a chloride environment
Mag Concrete Res, 62 (2010), pp. 91-101
P.K. Baranwal, R.P. Venkatesh
Investigation of carbon steel anodic dissolution in ammonium chloride solutions using electrochemical impedance spectroscopy
J Solid State Electrochem, 21 (2017), pp. 1373-1384
A.H. Espejel, M.P. Pardave, R.C. Sierra, M.R. Romo, M.T.R. Silva, E.M.A. Estrada
Kinetics and mechanism of the electrochemical formation of iron oxidation products on steel immersed in sour acid media
J Phys Chem B, 115 (2011), pp. 1833-1841
D.D.J. Antia
Desalination of water using ZVI (Fe0)
Water, 7 (2015), pp. 3671-3831
A. Sadkowski
On the ideal polarizability of electrodes displaying cpe-type capacitance dispersion
J Electroanal Chem, 481 (2000), pp. 222-226
F.F. Eliyan, A. Alfantazi
Corrosion of the heat-affected zones (HAZs) of API-X100 pipeline steel in dilute bicarbonate solutions at 90°C – an electrochemical evaluation
Corros Sci, 74 (2013), pp. 297-307
R.P. Venkatesh, S. Ramanathan
Electrochemical characterization of Cu dissolution and chemical mechanical polishing in ammonium hydroxide-hydrogen peroxide based slurries
J Appl Electrochem, 40 (2010), pp. 767-776
B. Hirschorn, M.E. Orazem, B. Tribollet, V. Vivier, I. Frateur, M. Musiani
Determination of effective capacitance and film thickness from constant-phase-element parameters
Electrochim Acta, 55 (2010), pp. 6218-6227
J. Tang, Y. Shao, J. Guo, T. Zhang, G. Meng, F. Wang
The effect of H2S concentration on the corrosion behaviour of carbon steel at 90°C
Corros Sci, 52 (2010), pp. 2050-2058
R. Feng, J. Beck, M.Z. Moroz, S.N. Lvov
Electrochemical corrosion of ultra-high strength carbon steel in alkaline brines containing hydrogen sulfide
Electrochim Acta, 212 (2016), pp. 998-1009
A.V. Benedetti, P.T.A. Sumodjo, K. Nobe, P.L. Cabot, W.G. Proud
Electrochemical studies of copper, copper–aluminium and copper–aluminium–silver alloys: Impedance results in 0.5M NaCl
Electrochim Acta, 40 (1995), pp. 2657-2668
M.A. Migahed, A.A. Attia, R.E. Habib
Study on the efficiency of some amine derivatives as corrosion and scale inhibitors in cooling water systems
RSC Adv, 5 (2015), pp. 57254-57262
J.C. Liu, S.W. Park, S. Nagao, M. Nogi, H. Koga, J.S. Ma
The role of Zn precipitates and Cl anions in pitting corrosion of Sn–Zn solder alloys
Corros Sci, 92 (2015), pp. 263-271
L. Cáceres, T. Vargas, L. Herrera
Influence of pitting and iron oxide formation during corrosion of carbon steel in unbuffered NaCl solutions
Corros Sci, 51 (2009), pp. 971-978
B. Lin, R. Hu, C. Ye, Y. Li, C. Lin
A study on the initiation of pitting corrosion in carbon steel in chloride-containing media using scanning electrochemical probes
Electrochim Acta, 55 (2010), pp. 6542-6545
N. Aouinaa, F.B. Célérier, F. Huet, S. Joiret, H. Perrot, F. Rouillard
Single pit initiation on 316L austenitic stainless steel using scanning electrochemical microscopy
Electrochim Acta, 56 (2011), pp. 8589-8596
J.O. Park, M. Verhoff, R. Alkire
Microscopic behaviour of single corrosion pits: the effect of thiosulfate on corrosion of stainless steel in NaCl
Electrochim Acta, 42 (1997), pp. 3281-3291
D.D. Macdonald
The point defect model for the passive state
J Electrochem Soc, 139 (1992), pp. 3434-3449
Y. Zhang, M. Urquidi-Macdonald, G.R. Engelhardt, D.D. Macdonald
Development of localized corrosion damage on low pressure turbine disks and baldes: I. Passivity
Electrochim Acta, 69 (2012), pp. 1-11
Copyright © 2018. Brazilian Metallurgical, Materials and Mining Association
Journal of Materials Research and Technology

Subscribe to our Newsletter

Article options
Supplemental materials
Cookies policy
To improve our services and products, we use cookies (own or third parties authorized) to show advertising related to client preferences through the analyses of navigation customer behavior. Continuing navigation will be considered as acceptance of this use. You can change the settings or obtain more information by clicking here.