A Direct Measurement of the Activation Potential of Stainless Steels in Nitric Acid

The rate of anodic dissolution and the associated activation potential that characterizes the passive-active transition of stainless steels have been measured directly for the ﬁrst time in nitric acid. The anodic dissolution current under cathodic polarization in pure nitric acid,inabsenceofchlorides,ismaskedbyintensecathodichydrogenreduction.Inthiswork,atomicemissionspectroelectrochemistry(AESEC)wasusedtorecordsimultaneouslythedissolutionrateoftheindividualalloyingelementsofstainlesssteelsaswellastheoverallcathodiccurrent.Thismethodologyhasbeenusedtoquantifytheinﬂuenceofseveralparametersontheactivationpotential:nitricacidconcentration,temperature,andtheadditionofsiliconinthesteelcomposition.©TheAuthor(s)2017.PublishedbyECS.ThisisanopenaccessarticledistributedunderthetermsoftheCreativeCommonsAttribution4.0License(CCBY,http://creativecommons.org/licenses/by/4.0/),whichpermitsunrestrictedreuseoftheworkinanymedium,providedtheoriginalworkisproperlycited.[DOI:10.1149/2.0081709jes]Allrightsreserved.

Nitric acid, HNO 3 , is a widely-used electrolyte in nuclear reprocessing plants for spent nuclear fuel. 1,2 In addition to its acidic properties, HNO 3 is a strong oxidizing agent and therefore material choice for the industrial devices must follow strict specifications. Some austenitic stainless steels (SS) such as the 18Cr-10Ni type SS are frequently chosen because of their high corrosion resistance in concentrated nitric acid. 1 The cathodic and anodic reactions of stainless steel in concentrated nitric acid have been the object of numerous investigations. [3][4][5][6][7] Cathodic processes involved in austenitic SS corrosion in concentrated nitric acid have been investigated since the beginning of the 20 th century. 6,[8][9][10] However, in the very low range of potentials of interest for the present work, the proton reduction reaction is expected to prevail. 6 The anodic reactions of stainless steel in the active state have proven difficult to investigate due to the fact that stainless steel is spontaneously passive in concentrated nitric acid, and when polarized to the active potential domain, the high cathodic current completely masks the anodic current. Under certain conditions, the potential of nitric acid can find itself closer to the active domain. Some other acidic electrolytes such as H 2 SO 4 or HCl have more clearly shown a spontaneous activation of SS in spite of high initial open circuit potentials 11,12 or when polarized for a long time close to the active domain. 13 This supports an interest into exploring the passive layer stability on the edge of the active dissolution. When the electrode potential becomes increasingly cathodic and approaches the active state, the oxides making up the passive film become thinner, less protective, and the dissolution rate of the steel increases. In general, the thickness of the passive film is determined by a steady state between film growth and film dissolution. To a first approximation, the rate of film growth will decrease with decreasing potential, while film dissolution is less dependent on potential and more a function of electrolyte pH. Although the oxidation rate of the stainless steel should, in theory, decrease with decreasing potential, the rate of elemental dissolution will increase due to the decreased thickness of the passive film. The potential below which the dissolution rate becomes measurable is referred to as the "activation potential", E a .
Many studies of the "activation potential" have been made on various materials including pure iron, 11,12 iron-chromium alloys 12,14 and more complex alloys involving nickel and other elements. 16,17 They lead to an establishment of a linear dependence between the activation potential and the logarithm of the proton activity. These measurements have mostly been performed in sulfuric acid using linear * Electrochemical Society Member. z E-mail: kevin.ogle@chimie-paristech.fr sweep voltammetry. 12,14,18 This is possible because there is a sufficiently wide electrochemical window to measure active dissolution at low potentials without interference from cathodic reactions. Frankenthal et al. 18 showed that an accurate measurement of E a was possible to within a few millivolts. However, in nitric acid, no cathodic loop behavior (total current that becomes positive over a short range of potentials in the active domain of the steel) can be observed because overall current is strongly negative. Such a method is not appropriate for concentrated nitric acid due to the high cathodic current in the active potential domain. Measurements of E a can also be based on the spontaneous activation behavior of stainless steels in sulfuric acid. 15,17 After a passivation polarization of the sample in sulfuric acid, the potential at which passivity decays spontaneously to an active potential on shutting off this anodic polarization has been also called the "Flade potential". 16,19 Once again, this method cannot be applied to nitric acid because the sample is spontaneously passive. It must also be noticed that using a voltammetric linear scan may have an impact on the mechanism of passivity breakdown in the case of chromium rich passive layers. 18 King and Uhlig 12 highlighted that linear scan measurements doubled the slope E a = f(pH) and that this difference could be related to differing chemical equilibria accompanying the breakdown of passivity.
In several references, samples were passivated in nitric acid and their activation potential was then studied in sulfuric acid. 11,20 According to these authors, stainless steel passivated in HNO 3 and in H 2 SO 4 show a very similar behavior. However, they were not able to perform the in situ measurement since below the corrosion potential, the contribution of the anodic dissolution (active and passive) of the sample to the overall electrochemical current is completely masked by the cathodic reduction of nitrate. The use of chlorides 23 provided a way to increase anodic current density and thus a chance to observe an anodic contribution greater than the reduction current at low potentials, which was a proof of the active dissolution, but did not permit easily the determination of the activation potential. A few attempts have been made at measuring the activation potential by gravimetry in a discontinuous way to measure this anodic dissolution rate below the corrosion potential. The material was polarized at low potentials where the reduction reaction is intense, and the weight loss recorded after several hours of polarization. 3,21 Such a method raises the issue of the solution chemical equilibria over the measurement time. It requires a massive electrode that releases a high quantity of metallic ions, while at the same time the counter electrode, in the same reactor, can highly disturb gas equilibria in the environment. The analysis of the increasing cathodic current during a long-term polarization of samples in HCl enabled Moshaweh and Burstein 13 to identify the activation of SS under high cathodic currents. Activation was deduced qualitatively and indirectly by observing the enhancement of the cathodic hydrogen reduction due to the exposed metallic surface.
In this work, a novel method of directly measuring the activation potential of stainless steel in HNO 3 will be presented using atomic emission spectroelectrochemisty (AESEC). In this way, the anodic dissolution of the stainless steel was measured directly despite a cathodic current which is orders of magnitude larger. The variation of the activation potential was measured as a function of nitric acid concentration, temperature and silicon concentration of the steel and interpreted in terms of oxide stability. Previous work concerning the application of AESEC to stainless steel was essentially limited to the case of sulfuric acid, H 2 SO 4 . Briefly, the electrochemical kinetics of dissolution in the active state were investigated in the earliest AESEC publication. 22

Experimental
Materials.-18Cr-15Ni-3.5Si SS (designated in the present work as Si-rich SS) and 18Cr-10Ni SS (designated in the present work as 304L SS) were used in this work for AESEC measurements. The specimens were cut into squares of 20 mm × 20 mm × 1 mm. The elemental analysis of the steel was performed by glow discharge optical emission spectroscopy (GD-OES) using a GD-Profiler from Horiba-Jobin Yvon. From the signals given by the GD-OES, the bulk of the sample was clearly reached from 10 to 40 μm depth and the composition of the steel was averaged over two craters on different samples. This analysis is given in Table I. Si-rich and 304L SS are very similar stainless steels in composition, except for silicon and nickel. Relative ratios between major elements Fe, Cr and Ni are comparable. Prior to any use, the specimens were cleaned with ethanol and acetone in an ultrasonic bath, then polished to a mirror finish with 0.03 μm diamond paste. Polishing ensures good sealing in the AESEC flow cell and allows possible ex situ surface analysis of the sample if required. All samples were then left for passivation at open circuit potential for 24 h in nitric acid at the desired concentration and at room temperature in separate beakers. The passive film was then analyzed by X-ray photoelectron spectroscopy prior to any AESEC measurement.
Electrolytes.-Deionized water (18.2 M cm) was prepared with a Millipore system and used for all electrolytes. Nitric acid 68% (Sigma Aldrich) was used to prepare the solutions. pH of each solution was verified using a Mettler-Toledo DL55 titrator and 1 mol dm −3 (reagent grade, Sigma Aldrich). All glassware was protected with a paraffin film to avoid any hazardous contamination.
The experiments were performed in a temperature range of 28 to 80 • C (301 to 353 K) using a recirculating water system connected to a thermocryostat (LAUDA) constant temperature bath. Water from the bath was circulated through a hollow copper block connected to the rear of the working electrode so that the electrode was heated directly. The electrolyte reservoir was also heated in the constant temperature bath. Electrical isolation between the block and the sample was designed to prevent both current leakage and ensure heat transfer.
Electrochemical measurements.-The AESEC flow cell has been described in detail in previous publications. 22,27 It consists of a threeelectrode cell with the stainless steel specimen as the working electrode (Si-rich SS or 304L SS), a small platinum grid as a counter electrode, a mercury-mercurous sulfate reference electrode (MSE, E = +0.65 V vs. SHE in saturated K 2 SO 4 ). The flow cell consists of a small volume working electrode compartment (approximately 0.2 cm 3 ) with electrolyte input at the bottom and output at the top. The flow rate was measured accurately (1% precision) for each experiment and nominally 3 cm 3 min −1 . The reference and counter electrodes were in a separate compartment separated from the working electrode by a porous membrane to allow passage of electrical current but avoid bulk mixing of the two solutions. A Gamry Reference 600 potentiostat functioning in the potentiodynamic linear polarization mode was used to measure electrochemical current density j e with a 0.2 mV s −1 scan rate. It should be mentioned that using a dynamic measurement could impact the values obtained, 12,13 however, the slow scan rate used here enables the measurement of electrical current at a quasi-stationary state of the electrochemical reactions within a reasonable total time of experiment. The analog outputs of the potentiostat were fed into the A/D converter and data acquisition software of the ICP-OES spectrometer so that the ICP-OES intensity data and the electrochemical data were on exactly the same time scale.
Atomic emission spectroelectrochemistry.-The experimental set-up including data acquisition has been described in detail in Ogle et al. 22 Briefly, the working electrode releases ions into the electrolyte in the flow cell. The electrolyte is then continuously fed into the plasma of the ICP-OES where the emission intensities of the different ions are measured simultaneously. These emission intensities were converted into concentration using standard ICP-OES calibration techniques. Electrolyte transport was implemented via a peristaltic pump. Electrolyte transfer into the plasma was realized via a concentric glass nebulizer and a cyclonic spray chamber. The ICP-OES used in this work was an Ultima 2C from Horiba Jobin Yvon consisting of a 40.68 MHz inductively coupled Ar plasma, operating at 1 kW and interfaced to independent polychromator and monochromator optical modules. A 50 cm focal length Paschen-Runge polychromator was used equipped with an array of photomultiplier tube detectors at given wavelengths allowing the measurement of up to 50 preselected elements simultaneously. Emission wavelengths were chosen for maximum sensitivity and low interferences. The monochromator (1 m focal length) with a Czerny-Turner configuration is dedicated for high spectral resolution of a single element. In the present work, the monochromator was used to monitor the Cr signal. Wavelengths used for each element and corresponding detection limits are given in Table II.  where f is the flow rate of the electrolyte, A the surface area, F the Faraday constant, z M the oxidation state of the element M, and M M its molar weight. Total dissolution current j can be defined as the sum of major elements dissolution currents: and j will be compared to electrical current j e * which is obtained by convoluting the electrical current measured by the potentiostat, j e , using an experimental transfer function h(t), where h(t) represents the distribution of residence times in the flow cell. 27 Complex physical processes contribute to the broadening of h(t). These processes include diffusion from the surface into the flowing electrolyte stream, mixing in the channel flow cell, spreading out during the laminar flow in the capillaries between the cell and the spectrometer, and the complicated nebulization system itself. Despite this complexity, an empirical function was simulated in the form of a log-normal distribution after an experimental measurement. 24,27 This convolution treatment is necessary to compare electrochemical current with elementary equivalent currents that are estimated from the elementary concentration transients in solution. The convolution integral is: 22 The detection limits 3σ are calculated as following: where σ blank is standard deviation of the background and α the sensitivity factor calculated from the calibration curves of each element at their specific wavelength.
X-ray photoelectron spectroscopy (XPS).-XPS analyses were carried out with a Thermofisher Escalab 250 XI spectrometer using a monochromatic X-ray Al Kα source. The instrument was calibrated in energy with the silver Fermi level (0 eV) and the 3d 5/2 core level of metallic silver (368.3 eV). The C-1s signal was used to correct a possible charge effect: the CC/CH contribution of C-1s spectra was fixed at 285.0 eV. The analysis zone consisted of a 900 μm diameter spot.
No etching of the surface was done before the experiment. The data processing was performed using the commercially available Avantage software. The main parameters used to decompose XPS spectra into the various contributions of major elements of the alloy are presented in Table III.

Results and Discussion
Anodic dissolution below the corrosion potential.-The AESEC method permits a direct measurement of the anodic dissolution of metals even when they cannot be detected in the electrical current. A typical example of this for the Si-rich stainless steel in HNO 3 is given in Fig. 1. A cathodic linear polarization sweep was performed beginning at the open circuit potential of 0.19 ± 0.01 V vs. MSE where the steel is passive. As no significant dissolution rate is measured between the open circuit potential and −0.70 V vs. MSE, only a −0.70 V to −0.90 V vs. MSE range of potential is presented in Fig. 1. Shown as a function of potential are the electrical current, j * e , and the elemental currents, j M , and the sum of the elemental currents J . Fig.  2 shows j * e , j Mn , and j on a semilogarithmic scale. Ammonia, NH 3 , can be expected to be produced below 0.05 V vs. NHE (−0.60 V vs. MSE). 7 However, in the same region the proton reduction reaction should also take place and dominate. 7 As the potential decreases in the cathodic direction, j e * increases systematically approaching −31 mA cm −2 at −0.9 V vs. MSE reflecting the reduction of H + to H 2 . Supporting the idea that the proton reduction is the major reaction, two Tafel slopes of −40 mV and −120 mV per decade can be read on |j e * | in Fig. 2 that are usually assimilated, respectively, to Volmer-Herovsky and Volmer-Tafel proton reduction mechanisms on metals in acidic environment. 28 Fig. 1 gives as a function of potential, j e * and the elemental dissolution currents of the alloying elements (j M , where M = Fe, Cr, Ni, Si, Mn). Like the cathodic current, the elemental dissolution currents, j M , also increase monotonically as the potential decreases below −0.7 V vs. MSE. Expressed as equivalent faradaic currents, their sum j is also shown by way of comparison to j e * (Fig. 2). Their systematic increase clearly demonstrates the loss of passivity as the potential becomes increasingly cathodic. Note however that at the final point, −0.9 V vs. MSE, j is only 2.4 mA cm −2 which is quite negligible as compared to the −31 mA cm −2 of j e * . This demonstrates the capacity of the AESEC technique to quantitatively detect very low anodic dissolution rates under circumstances when the electrochemical interface is dominated by the cathodic reaction. The activation transient of the stainless steel is not visible in the conventional polarization curve of j e * vs. E. The operative definition of the activation potential E a used in this work is shown in Fig. 2. The logarithm of |j | and |j Mn | are plotted as a function of potential. As Mn is the alloying element with the lowest detection limit and is completely soluble in the HNO 3 electrolytes used here, it is a logical candidate for the determination of E a . The detection limit, expressed in A cm −2 , is 1.5 × 10 −7 A cm −2 and is shown in Fig. 2 (lower dashed line). Passive dissolution of the sample was not detectable by AESEC, being below the detection limit. Therefore, E a is defined as the first potential where j Mn rises to five times the detection limit (upper dashed line), in this case 7.5 × 10 −7 A cm −2 . This measurement is actually coherent with any similar measurement on all of the elemental signals and enables determination of E a with a precision of ±10 mV. Reproducibility of this measurement is also consistent with an uncertainty of 10 mV over 3 experiments in spite of an intense cathodic reaction that may lead to scatter in the dissolution rates. 13 Anodic dissolution was demonstrated to be congruent by the results of Fig. 3. The dissolution rates of Cr, Ni, Si, and Mn in μg s −1 cm −2 divided by the mass ratio of the chemical composition of the steel given in Table I, are plotted as a function of the Fe dissolution rate. The good superposition of all the dissolution rates as a function of ν Fe reveals  the non-selective nature of the active dissolution. It can be noticed that Mn presents a rigorously non-selective behavior with respect to Fe, which supports the choice of Mn as the reference element for E a determination. The non-selective behavior for all M supports the conclusion that the active dissolution domain has been reached. One could expect an excess of Cr dissolution if Cr oxide were dissolving at the surface of the sample. 33 A zoom of the low current values (inset to Fig. 3) does not confirm this expectation. However, Cr deviation could be too small to be seen in Fig. 3, and therefore transient dissolution rate analysis should be more appropriate to measure any Cr enrichment. Such experiments were performed by Ogle et al. 25 in sulfuric acid using active-passive cycles.
Activation potential depending on pH.-Previous measurements of the activation potential demonstrated a proportional relationship between E a and the pH in sulfuric acid for several materials from pure iron to various types of iron-based alloys [12][13][14][15][16]19,29 AESEC measurements of E a for the Si-rich SS in different concentrations of nitric acid enabled assessment of this relationship. Measurements of the activation potential of the Si-rich SS are shown in Fig. 4 for 2, 4 and 6 mol dm −3 HNO 3 .
Due to their method of measurement, most of studies cited before were not able to measure E a for high proton activities. As AESEC is able to work with concentrated electrolytes, measurement of E a of the Si-rich SS was possible with the identical precision even for concentrations up to 6 mol dm −3 .
Each nitric acid concentration is associated with a certain proton activity which was calculated after Fallet's 30 work upon the stoichiometric activity coefficient of the proton in 28 • C binary HNO 3 -H 2 O solutions. This calculation takes into account the incomplete dissociation of the nitric acid. The resulting activities are given in Table IV. The total cathodic current j e * increased with increasing proton activity. This is not surprising since the proton reduction reaction is expected to dominate below −0.65 V vs. MSE. If the cathodic reaction is enhanced by the activity of the proton, it can also be accelerated by the passivity breakdown of the working electrode. The activation potential increases with the activity of the proton a H + as shown in Fig. 4 with a slope of 0.12 V ± 0.03 V (Fig. 5). According to Rocha et al.'s measurements 14 on various iron-chromium alloys at room temperature, in the case of linear sweep voltammetry measurement, the activation potential of such alloys in sulfuric acid followed the Equation 5: with the n value of 0.5 as described in Rocha et al., 14 which is an experimental value determined for an alloy that contains more than 15 wt% in chromium, the activation potential curve should display a slope of 0.116 (n = 0.5). Uhlig and King 11 showed that in the case of pure iron, n = 1, leading to a Nernstian slope of E a vs. pH for Fe dissolution in acidic electrolytes. They suggested that for Fe-Cr alloys, the passivity breakdown mechanism would have an impact on the n-value. When spontaneously activated in sulfuric acid, n = 1, but if activation was performed through linear polarization, they found a value of n = 0.5. The results presented in Fig. 5 seem to be very similar to what was found in other electrolytes. 11,16,18 Therefore, the E a of stainless steel in very concentrated nitric acid is comparable to sulfuric acid in terms of proton activity dependence.
The free enthalpy of formation may be accessed from the yintercept of Fig. 5. Rocha et al. 14 defined this y-intercept as the standard activation potential E • a and measured −0.26 V vs. SCE which is higher than −0.87 vs. MSE found in the present work (by about 200 mV). Uhlig et al. 12,29 showed that a standard free enthalpy of formation, r G • , of the oxide can be obtained from the value of E • a and that it helps to assess the alloy's affinity to oxygen in the electrolyte. The calculation is based on the equation of oxide formation as follow: where M is the metallic element considered. Such affinity might also depend strongly on the chemical composition and microstructure of the passive layer.  These considerations lead to the idea that the oxide formation kinetics could be related to a specific oxide layer. Therefore, the value of E • a could be explained by a different passive layer that is preferentially formed in the three different concentrations. XPS quantifications were performed for passive layers formed at Si-rich SS surface in the three nitric acid concentrations at 28 • C and in sulfuric acid 2 mol dm −3 at 28 • C. The results ascertained in Table V present relative concentrations in iron, chromium and silicon oxides regarding levels Fe-2p 3/2 , Cr-2p 3/2 and Si-2p.
The differences between nitric and sulfuric electrolytes are not considered as significant and the results also demonstrate a reproducible passive layer composition in all nitric electrolytes. The oxide film's nature and thickness (relatively evaluated by the ratio of oxide and metallic peaks' areas) do not seem to be sensitive to the concentration or nature of the electrolyte. It is very likely that the solubilities of all species in the oxide layer increase with the proton activity. The concentration of nitrates could also play a role but was not investigated in the present work.
Activation potential depending on temperature.-The proportional relationship between E a and pH has been assimilated to a Nernst type equation by several authors. 12,20 It can then be expected that E a also depends linearly on temperature. AESEC measurements of the E a of Si-rich SS were performed in 4 mol dm −3 HNO 3 at different temperatures. Fig. 6 displays the total dissolution rates obtained for  the temperatures of 28 • C, 40 • C, 60 • C and 80 • C (respectively 301 K, 313 K, 333 K and 353 K). E a is between −0.75 and −0.60 V vs. MSE in these conditions. The higher the temperature, the higher the dissolution rate measured at a given potential, resulting in a shift of E a to higher potential. With increasing temperature, j e * also increases. This may be considered as a consequence of temperature elevation under the hypothesis of hydrogen reduction whose reaction rate follows the Butler-Volmer equation.
When the E a values extracted from Fig. 6 are plotted as a function of temperature, a proportionality factor of 0.0027 V/K ± 0.0004 is observed (Fig. 7). For a better legibility of the calculation, temperatures have been converted from degrees Celsius to Kelvin. Under the hypothesis of a Nernst type evolution of E a vs. T, the slope of the curve of Fig. 7 will depend on several parameters such as the activities of the elements in the alloy and the number of electrons exchanged. These parameters cannot be determined easily. Nevertheless, the linear dependence between E a and temperature is demonstrated through the present work.
Activation potential depending on the silicon content in the alloy.-The silicon enrichment of 18Cr-10Ni type SS has been shown to provide a homogeneous corrosion morphology in very oxidizing electrolytes. 10,31 In the present work, XPS measurements of the Sirich SS confirmed a significant presence of silicon in the passive layer which is also higher than in its bulk composition (Table VI). As compared to the 304L stainless steel, which is a 18Cr-10Ni type SS with 0.34 wt% of silicon in its composition (Table I), Si-rich SS presents a chemically different oxide layer. The chromium rich 304L SS's passive layer, Cr 2 O 3 , was quantified by XPS measurements displayed in Table VI.
The oxide peak energies measured for silicon suggest that the alloy is oxidized into some mixed iron-chromium silicates whose chemistry is not easy to determine. Similar conclusions can be found in Robin et al. 32 The activation potentials of such different passive layers may  provide information about their respective properties in nitric acid, for example their affinity to oxygen.
The activation potentials of the 304L SS were measured by AESEC and compared in the same conditions as the Si-rich SS. Fig. 8 shows the total dissolution currents measured for the 304L SS during the activation of the sample as compared to the Si-rich SS's in a 4 mol dm −3 HNO 3 at 28 • C. The reduction reaction increases shortly after the break of passivity occurs, as observed in Fig. 6. Considering that the interface reactivity is deeply modified by the break of passivity, one can expect an increase of the reduction kinetics regardless of the reduced species.
A gap of about 200 mV appears between E a of the 304L (−0.95 V vs. MSE) and Si rich stainless steel (−0.78 V vs. MSE). The equilibrium potential of the couple Cr(III)/Cr(II) is −0.41 V vs. NHE 35,36 corresponding to −1.06 V vs. MSE and is consistent with E a of 304L SS. However, if the active dissolution was only the consequence of Cr(III) reduction, such a large difference in E a between these two very similar alloys would not be expected. Given the XPS results, it is very likely that Si lowers the free enthalpy of formation of the oxides, r G • . Such a result is not trivial, since Vetter 34 listed some standard potentials of oxide electrodes calculated from r G • and the Si oxide electrode (−1.51 V vs. MSE) has a lower potential than Cr (−1.25 V vs. MSE).

Conclusions
The AESEC technique was used to measure the activation potential, E a , corresponding to the passive to active transition of stainless steel under conditions of net cathodic reaction rate by analyzing the anodic dissolution rate directly and independently from the cathodic reaction. The value of E a in a given environment gives information on the alloy's affinity to oxygen and thus the stability of the oxides formed. The relationship between E a and temperature was demonstrated from 28 • C to 80 • C and hydrogen ion activity from 2 to 12.8.
It was confirmed by XPS measurements that the initial passive layer is similar for all conditions in terms of thickness, structure and chemistry, enhancing the hypothesis of both an increase of the oxide solubility with the proton activity and an impact of the nitrates on the oxygen interaction with the alloy.
A linear dependence between the activation potential and pH was found, similar to what has been found in the literature in other electrolytes. The slope of this linear dependence was found to be very similar to that in sulfuric acid, although the standard activation potential was shifted to lower values. This shift might be explained by several parameters including the nature of the electrolyte (impact of NO 3 − ).
A linear dependence was found between the activation potential and the temperature (28 • C to 80 • C). The general tendency of this evolution would be in good agreement with a Nernst-type equation. However, determination of all parameters would require further investigations of activities of the alloyed elements.
It was shown that silicon in the stainless steel increases the value of the activation potential. The passive layer displays a higher activation potential as it contains less chromium and more silicon. It is proposed that Si lowers the affinity of the alloy to oxygen.