Assessment of the Electrolyte Composition in the Degree of Sensitization in AISI 347H Stainless Steel

Abstract

This study evaluated the suitability of the depassivator in revealing low degrees of sensitization (DOS) in samples of wrought AISI 347H austenitic stainless steel in the as-received condition and solution heat treated (SHT) at 1050℃. Assessment of the electrolyte composition was performed by the double loop electrochemical potentiokinetic reactivation (DL-EPR) test at room temperature. The electrolyte 1.0M H₂SO₄ + 0.50M HCl was found to be adequate for detecting low DOS. Microstructural characterization of the as-received material revealed the presence of Cr-rich carbides. These carbides were responsible of the susceptibility to IGC.

1. Introduction

Austenitic stainless steels (ASS) are the most commonly alloys in industries where aggressive environments prevail¹⁾. The AISI 347 is an ASS constituted by a matrix of austenite with the presence of ferrite in small percentages and it is stabilized with Nb. Typical applications refer to industries as oil extraction, refineries, thermo-electric and nuclear power plants due to its high corrosion resistance at high temperature²⁾. The principal characteristics of this alloy are the combination of good mechanical property and corrosion resistance in comparison to conventional ASS due to the presence of Nb which reduces the sensitization phenomenon and increases creep resistance as a result of the formation of NbC³⁾. Thermodynamically, the formation of niobium carbides are more favored and takes place in a broader range of temperatures⁴⁾ as compared to chromium carbides (450–850℃)^5,6). The precipitation of NbC occurs in grain boundaries, non-coherent twin boundaries, dislocations and stacking faults^5,7,8). A requisite for nucleation and growth of niobium carbides is met when there is at least a Nb:C mass ratio of 10:1³⁾. In this alloy, the highest affinity of C for Nb than Cr enhances the resistance to intergranular corrosion (IGC) by reducing the susceptibility to localized attack in virtue of the minimization of chromium depleted zones⁹⁾. Nevertheless, when there is C available in solid solution, it is possible the precipitation and growth of chromium-rich carbides⁸⁾ and the IGC resistance of this alloy may be reduced when it is subjected to thermomechanical processes as hot deformation¹⁰⁾. In this regard, Murr and Advani¹¹⁾ suggested a SHT in the range of temperatures between 1000–1100℃ to dissolve these chromium-rich carbides after hot deformation and improving the resistance to IGC.

The double loop electrochemical potentiokinetic reactivation (DL-EPR) test has been used as an effective technique for estimating the degree of sensitization (DOS) in ASS. Modifications to the conventional DL-EPR test have been made in the electrolyte depending on the type of alloy, but the most common depassivator or activator in the dissolution of metals is potassium thiocyanate. This compound is used due to its catalytic effect in diluted solutions containing sulfuric acid, because potassium thiocyanate promotes the dissolution of the passive layer in chromium depleted zones, increasing the current density during the anodic dissolution in the activation process and subsequently along the grain boundaries during the reactivation loop¹²⁾. It is established that a disproportionate reaction of the decomposition of potassium thiocyanate takes place producing the adsorption of S in the surface of the sample, which is stable in the potential range of anodic dissolution, where S compounds are reduced according to the reaction   

\[{\rm SCN^-} \to {\rm CN^-} + {\rm S_{ads}}\](1)

Based on this reaction, it was observed that an increase above 0.02 M in the concentration of potassium thiocyanate led to general corrosion and reduced the possibility of the passivation of the material even at high potentials¹³⁾. In highly corrosion resistant Ni-based alloys the substantial increase of potassium thiocyanate can act as inhibitor due to the high concentration of S adsorbed on the surface of the material. To overcome this inconvenient, hydrochloric acid was used as an activator in a greater concentration than sulfuric acid along with potassium thiocyanate in concentration up to 0.001 M¹⁴⁾, with good reproducibility¹⁵⁾.

The use of Nb, Ti, Zr and V as stabilizing elements delays the onset of sensitization due to the formation of carbides with these elements at higher temperature than chromium carbides with much less C available in solid solution. The outcome of the precipitation of carbides other than chromium carbides is an increase in IGC resistance^6,16,17). In spite of the use of stabilizing elements, Hong et al.¹⁸⁾ found the precipitation of chromium carbides in 347 ASS after hot deformation. Precipitation of chromium rich carbides, M₂₃C₆, is feasible considering the kinetics of this process. Dissolution of niobium carbides and Nb free regions enable carbon diffusion into Cr-rich zones and thereby precipitation of chromium carbides making the steel susceptible to IGC^3,19). In this context, the thermo-metallurgical history of these alloys is crucial in their performance. A SHT has proved to be advisable for restoring the resistance to IGC as reported by Kina et al.²⁰⁾ in samples with different metallurgical conditions. Assessment of the susceptibility to localized attack by these authors was performed in a 0.5M sulfuric acid + 0.01M potassium thiocyanate solution.

This study was undertaken in order to assess the capability of an electrolyte to disclose light DOS using the DL-EPR test at room temperature in AISI 347 ASS by modifying the activator agent with additions of Cl⁻ ions with and without KSCN.

2. Experimental Details

Plates, 7 mm thick, of wrought AISI-347H ASS with the chemical composition given in Table 1 were used. Samples of 10 × 20 × 7 mm were taken from the as-received ASS sheet and subjected to a SHT at 1050℃ and held at temperature for 30, 60 and 90 minutes followed by water quenching. Dissolution of chromium carbides occurs at this temperature along with grain growth of the austenitic matrix^11,20). For microstructural characterization, the samples were mirror like polished following standard metallographic preparation and etched by immersion-stirring in a solution containing 8.43 mL hydrochloric acid + 2.80 mL nitric acid + 3.75 mL ethylic alcohol, rinsed with a stream of water and dried. Grain size was measured, with software facilities, in samples with and without SHT from digital images captured in the optical microscope (OM) and scanning electron microscope (SEM) equipped with an energy dispersive X-ray detector (EDX).

Table 1 Chemical composition of the AISI 347H ASS (mass%).

C	Mn	S	P	Si	Cr	Ni	Nb	Co	Cu	Mo	N	Ti	Al
0.04	1.5	0.001	0.03	0.36	17.3	9.3	0.64	0.28	0.45	0.41	0.04	0.005	0.004

In order to evaluate the DOS by the DL-EPR test, samples were embedded in epoxy resin and a copper wire was attached in the rear for electric connection and used as working electrode in a conventional three electrodes electrochemical cell using a graphite bar as auxiliary electrode and a saturated calomel electrode (SCE) as reference electrode. Samples were previously ground with emery paper (1200 grit), rinsed with distilled water and degreased with ethylic alcohol. Before the DL-EPR tests, samples were stabilized at open circuit potential (E_oc) during 20 minutes. Cyclic potentiokinetic polarization was conducted from E_oc and open to air at a scanning rate of 1 mV/s to an overpotential of 200 mV vs. SCE at ambient temperature (25℃) using a potentiostat Solartron 1280B.

The composition of the electrolytes used is shown in Table 2. Tests were made three times for reproducibility using samples subjected to identical surface preparation and fresh electrolyte in every electrochemical test. The DOS was determined with the Ir/Ia ratio, where Ir and Ia are the maximum peaks of the current density during anodic reactivation and activation, respectively. After the DL-EPR test, samples were taken into the SEM to observe the degree of damage and morphology of the corrosion features generated with every electrolyte. Correlation between DL-EPR curves and visual observations of the samples in the SEM pointed to the solution with better reproducibility for detecting light DOS due to the presence of M₂₃C₆ or Cr-rich phases in terms of quantitative and qualitative results.

Table 2 Composition of the electrolytes essayed.

E1	0.5M H2SO4 + 0.01M KSCN
E2	1.0M H2SO4 + 0.01M KSCN
E3	1.0M H2SO4 + 0.01M KSCN + 0.5M NaCl
E4	1.0M H2SO4 + 0.50M HCl

3. Results

3.1 Microstructural characterization

Figure 1 shows the typical microstructure of the AISI 347H ASS in the as-received condition. The microstructure corresponds to an austenitic matrix with twins and niobium carbides allocated in the grain boundaries and within the grains. A mean grain size of 13 µm was measured for the ASS in the as-received condition. The microstructures of the SHT samples are shown in Fig. 2. It can be seen grain growth with time when holding at temperature, namely; 16.5, 26.5 and 60.6 µm for holding times of 30, 60 and 90 minutes, respectively. In the microstructure of the heat-treated samples the presence of twins is also observed. SHT of the AISI-347H ASS also resulted in the modification of the niobium carbides. Typically, solution annealing of ASS`s is performed in the temperature range of 1020 to 1150℃⁵⁾. The SHT of 1050℃ is close to the temperature of dissolution of niobium carbides. Thus, solution annealing of the samples at this temperature gave rise to a slow dissolution-reprecipitation process. For heat-treating times of 30 and 60 minutes, slight coarsening of the niobium carbides is appreciated. Particle size measurements from SEM images of the NbC particles reveals a very small increase from 1.01 ± 0.75 µm for the as-received stainless steel to 1.103 ± 0.59 and 1.205 ± 0.68 µm for holding times of 30 and 60 minutes, respectively, and a decrease to 0.7 ± 0.36 µm for holding after 90 minutes. This behavior is explained by the dissolution-reprecipitation mechanism activated at this temperature where for the longest holding time, besides the dissolution of chromium carbides (this made more C available for precipitation of NbC), a very large fraction of the initial NbC particles dissolved and reprecipitated into very fine NbC particles. In addition, the sample held at temperature for 30 minutes exhibits an even distribution of NbC with little clustering while the sample held for 60 minutes presents a number of austenitic grains with fewer NbC and other grains with clustering of niobium carbides as pointed by the arrows in Fig. 2(b). Holding for 90 minutes at 1050℃, Fig. 2(c), resulted in finer NbC evenly distributed in the austenitic matrix and the presence of colonies of large niobium carbides as indicated by the circle in Fig. 2(c).

Fig. 1

Characteristic microstructure of the as-received AISI-347H ASS as observed in the SEM.

Fig. 2

Microstructures of the AISI-347H ASS as observed in the SEM after SHT at 1050℃ for (a) 30, (b) 60 and (c) 90 minutes.

3.2 Assessment of the electrolyte composition

Figures 3(a) and (b) show the DL-EPR curves of the as-received AISI-347H ASS essayed with the distinct electrolytes and the Ir/Ia ratios, respectively. It can be seen from Fig. 3(a) that increasing the concentration from 0.5 (E1) to 1 M (E2) of sulfuric acid only produces a little increase in the activation current (approximately 0.01 A/cm²). For these electrolytes according to Fig. 3(b) the Ir/Ia ratios indicate that the as-received alloy is not susceptible to IGC as the DOS is negligible. The curve of the composition E3 shows a very high increase in Ia of approximately 0.035 A/cm² with respect to composition E1. However, the reactivation current, Ir, peaks also at a high value. This suggests that for the composition of electrolyte E3 generalized corrosion instead of localized corrosion is taking place. Regarding solution E4, the plot in Fig. 3(b) shows that little DOS may be revealed with this composition. In addition, a significant reduction in Ia is observed (approximately 10 times with respect to solution E1), this effect is ascribed to the elimination of the excess of S that can be adsorbed on the surface of the sample by substituting potassium thiocyanate by hydrochloric acid.

Fig. 3

Evaluation of the effect of the electrolyte composition on the IGC resistance of as-received AISI-347H ASS; (a) DL-EPR curves and (b) Ir/Ia ratios.

The above observation can be related to the fact that with the concentration of potassium thiocyanate in solution E1 and E2 it is not possible to detect little DOS induced by the thermomechanical process experienced by the as-received material. However, substitution of the depassivator with another containing Cl⁻ ions (sodium chloride or hydrochloric acid) promotes the activation and the dissolution of chromium depleted zones restricting thus adsorption of SO₄²⁻ from the reaction   

\[{\rm H_2}{\rm SO_4} \to {\rm {SO_{4}}^{2-}} + {\rm H_2}\](2)

This mechanism enables reaction of the Cl⁻ ions with the metallic matrix without increasing the surface area of elemental S adsorbed. Nevertheless, in presence of potassium thiocyanate there is an increase in the adsorption of S promoting generalized corrosion due to the significant concentration of SO₄²⁻ + Cl⁻¹ and the formation of S₂O₃²⁻ ions resulting too much aggressive for the AISI-347H alloy. Examination of the surface of the samples in the SEM after DL-EPR tests, as shown in Fig. 4(a), confirms generalized corrosion. Conversely, when hydrochloric acid substitutes potassium thiocyanate, adsorption of elemental S on the metallic surface is prevented during anodic dissolution. This effect leads to a significant reduction in the Ia and Ir peak values because only SO₄²⁻ + Cl⁻¹ ions are involved in the oxidation of the metal matrix during the DL-EPR test. Figure 4(b) shows the features of the surface after exposure to electrolyte E4 during the DL-EPR test. In this instance, it is only observed localized corrosion in grain boundaries as well as in the vicinities of niobium carbides.

Fig. 4

Surface damage of the as-received samples as observed in the SEM after DL-EPR test in electrolyte; (a) E3 and (b) E4.

Figure 5 shows complementary characterization, in detail, by SEM-EDX of the as-received sample exposed to electrolyte 4. Figure 5(a) reveals the presence of precipitates in the grain boundaries. The EDX spectrum of the angular precipitate pointed by the arrow in the micrograph of Fig. 5(a) along with its elemental quantification given in Table 3 indicate that these precipitates correspond to Cr-rich carbides, likely of the M₂₃C₆ type. A number of precipitates with different morphologies are observed in the micrograph of Fig. 5(c). In this instance, the experimental evidence gathered by EDX elemental microanalysis showed that these phases correspond to niobium carbides. Thus, according to these results it is feasible to use the solution E4 at ambient temperature for detecting small DOS when niobium and chromium carbides coexist.

Fig. 5

Details of the microstructure of the as-received samples as observed in the SEM after DL-EPR test in electrolyte E4.

Table 3 Elemental quantification of the EDX spectra shown in Fig. 5.

		C	Si	Cr	Fe	Mo	Mn	Nb	Ni
Spectra b	mass%	5.97	----	5.26	17.05	----	0.57	67.92	3.14
	at.%	29.27	----	5.95	17.95	----	0.61	42.99	3.20
Spectra d	mass%	2.06	0.45	25.84	70.58	1.04	----	----
	at.%	8.76	0.83	25.35	64.48	0.55	----	----

3.3 Effect of the SHT on IGC resistance

Once that the use of hydrochloric acid as depassivator was found to be more adequate as compared to potassium thiocyanate or sodium chloride, DL-EPR testing of the SHT samples was carried out in order to evaluate the effect of the heat treatment in terms of susceptibility of the samples to IGC. Figure 6 shows characteristic DL-EPR curves and the Ir/Ia ratios as a function of the metallurgical condition of the AISI 347H stainless steel. From the plot shown in Fig. 6(b), it is evident that the Ir/Ia ratio of the SHT samples significantly decreased with respect to the value of the as-received material. As a matter of fact, this behaviour was expected. Microstructural characterization of the as-received AISI 347H stainless steel disclosed the presence of both NbC and Cr-rich carbides. Thus, heating of the samples up to 1050℃ not only coarsened the grain structure and changed the features of the niobium carbides, it also dissolved the chromium carbides formed during hot rolling of the stainless steel. Rapid cooling of the samples by water quenching prevented any re-precipitation of chromium carbides with any C available after holding at temperature for some time. The Ir/Ia ratio of the SHT samples drew an approximate value of 0.01 meaning that, virtually, AISI 347H stainless steel in these metallurgical conditions is not susceptible to experience IGC. This assumption is further supported by the features observed in the SEM of these samples after the DL-EPR test as shown in Fig. 7. The images, captured in secondary electron mode, do not exhibit localized corrosion of the surfaces. Only very few isolated pits were observed.

Fig. 6

Evaluation of the effect of SHT on the IGC resistance. (a) DL-EPR curves and (b) DOS of the SHT samples as given by the Ir/Ia ratio.

Fig. 7

Surface characteristics of the SHT samples as observed in the SEM after DL-EPR test in electrolyte E4.

The findings of this study indicate that care must be taken in the use of Nb stabilized stainless steel, because thermomechanical processing may induce susceptibility to IGC by the precipitation of Cr-rich carbides. This metallurgical condition is likely to be worsened during subsequent fusion welding of components. Thermal cycles experienced in the heat affected zone of the base material induce dissolution of NbC and because of the rapid cooling complete reprecipitation of NbC will not occur, leaving C available for nucleation of new and further growth of pre-existant chromium carbides.

4. Conclusions

• (1)    The presence of niobium carbides, as it is well known, in the microstructure of the stabilized austenitic stainless steel exhibited a positive effect in reducing the DOS. However, analysis of the as-received plate revealed some susceptibility to IGC as a result of its thermomechanical history. The engineers must be aware that this problem may be worsened by a subsequent fusion welding step of the AISI-347H alloy.
• (2)    It is possible to detect small DOS in the AISI-347 stainless steel with the use of hydrochloric acid as depassivator in substitution of potassium thiocyanate in DL-EPR test.
• (3)    The as-received AISI-347H stainless steel presented the highest DOS due to the precipitation of chromium carbides during its thermomechanical processing.
• (4)    A SHT at 1050℃ for 30 minutes is effective in dissolving chromium carbides so that IGC resistance of wrought AISI-347H stainless steel is restored.

Acknowledgement

VLCH thanks CONACyT for the scholarship provided.

REFERENCES

• 1)  R.L. Plaut, C. Herrera, D.M. Escriba, P. Rangel-Rios and A.F. Padilha: Mater. Res. 10 (2007) 453–460. 10.1590/S1516-14392007000400021
• 2)  I. Hajiannia, M. Shamanian and M. Kasiri: Mater. Des. 50 (2013) 566–573. 10.1016/j.matdes.2013.03.029
• 3)  T. Fukunaga, K. Kaneko, R. Kawano, K. Ueda, K. Yamada, N. Nakada, M. Kikuchi, J.S. Barnard and P.A. Midgley: ISIJ Int. 54 (2014) 148–152. 10.2355/isijinternational.54.148
• 4)  P. Lacombe, B. Baroux, G. Béranger, L. Colombier, J. Hochmann: Les Editions de Physique (1993).
• 5)  A.F. Padilha and P. Rios: ISIJ Int. 42 (2002) 325–337. 10.2355/isijinternational.42.325
• 6)  H. Uno, A. Kimura and T. Misawa: Corrosion 48 (1992) 467–474. 10.5006/1.3315962
• 7)  J. S. Silcock, A. W. Denham: institute of metals monograph, Institute of Metals 33 (1969).
• 8)  B. Sasmal: J. Mater. Sci. 32 (1997) 5439–5444. 10.1023/A:1018643617491
• 9)  K.H. Lo, C.H. Shek and J.K.L. Lai: Mater. Sci. Eng. 65 (2009) 39–104. 10.1016/j.mser.2009.03.001
• 10)  S. Rahimi, D. Engelberg and J. Marrow: Corros. Sci. 53 (2011) 4213–4222. 10.1016/j.corsci.2011.08.033
• 11)  L. Murr and Advani: Mater. Charact. 39 (1997) 575–598. 10.1016/S1044-5803(97)00145-9
• 12)  A.P. Majidi and M.A. Streicher: Corrosion 40 (1984) 584–593. 10.5006/1.3581921
• 13)  M. Momeni, M.H. Moayed and A. Davoodi: Corros. Sci. 52 (2010) 2653–2660. 10.1016/j.corsci.2010.04.015
• 14)  T.F. Wu and W.T. Tsai: Corros. Sci. 45 (2003) 267–280. 10.1016/S0010-938X(02)00100-2
• 15)  M. Prohaska, H. Kanduth, G. Mori, R. Grill and G. Tischler: Corros. Sci. 52 (2010) 1582–1592. 10.1016/j.corsci.2010.01.017
• 16)  M. Terada, D. Escriba, I. Costa, E. Materna-Morris and A. Padilha: Mater. Charact. 59 (2008) 663–668. 10.1016/j.matchar.2007.05.017
• 17)  A.S. Lima, A.M. Nascimento, H.F.G. Abreu and P.d. Lima-Neto: J. Mater. Sci. 40 (2005) 139–144. 10.1007/s10853-005-5699-9
• 18)  S.M. Hong, M.Y. Kim, D.J. Min, K. Lee, J.H. Shim, D.I. Kim, J.Y. Suh, W.S. Jung and I.S. Choi: Mater. Charact. 94 (2014) 7–13. 10.1016/j.matchar.2014.04.002
• 19)  R. Ayer, C. F. Klein and C. N. Marzinsky: Metallurgical Transactions A 23 (1992) 2455–2467. 10.1007/BF02658049
• 20)  Y. Kina, V.M. Souza, S.S.M. Tavares, J.A. Souza and H.F.G.d. Abreu: J. Mater. Process. Technol. 199 (2008) 391–395. 10.1016/j.jmatprotec.2007.08.011