A

sr

ds

cms©

K

1

haescaea

ecsni

0d

Materials Science and Engineering A 485 (2008) 234–242

Corrosion behaviour of Duplex stainless steels inorganic acid aqueous solutions

A.J. Invernizzi, E. Sivieri, S.P. Trasatti ∗Department of Physical Chemistry and Electrochemistry, University of Milan,

via C. Golgi 19, 20133 Milan, Italy

Received 19 March 2006; received in revised form 25 July 2007; accepted 14 August 2007

bstract

This paper reports on mass loss and electrochemical polarization behaviour of two duplex stainless steels, SAF2205 and SAF2507, in aqueousolutions of acetic acid, formic acid and their mixtures, in the presence of specific contaminant, e.g. sulphuric acid. The effect of temperature andedox couples was also investigated.

Both steels withstand corrosion in acetic acid solutions in a wide temperature and concentration range. The corrosion rate is higher in the moreissociated formic acid at the same concentration. Uniform corrosion with selective dissolution of the austenitic phase was observed in all testedolutions, being more pronounced with increasing solution aggressiveness.

The presence of sulphuric acid is severely detrimental to the corrosion resistance. The corrosion rate increases almost linearly with H2SO4

oncentration. The addition of few tens of ppm of oxidant species such as hydrogen peroxide and Fe3+ or Cu2+ improves the corrosion performancearkedly, leading to corrosion rates close to zero. In particular, the addition of H2O2 increases the corrosion potential up to the passivity of the

teels owing to the possible formation of peracids. 2007 Elsevier B.V. All rights reserved.

e diss

si<

pt

fpihddT

eywords: Duplex stainless steels; Organic acids; Polarization curves; Selectiv

. Introduction

Due to their high chromium and molybdenum contents,igher alloyed duplex grades such as SAF2205 and SAF2507re highly resistant to uniform corrosion in many industrialnvironments. Moreover, they exhibit improved pitting corro-ion resistance as a consequence of higher levels of nitrogenompared to austenitic steel grade. In the last decade, duplexpplications in chemical and petrochemical industry increasedven though not always supported by specific tests or researchctivities.

Organic acids constitute a broad family of chemicals of gen-ral formula RCOOH, where R is a hydrocarbon chain or ring,ommonly met in chemical industry. Their corrosive action is not

imple because of the high number of acids and because they areot used pure, but in mixture with other organic compounds ornorganic acids or salts. In general terms, carboxylic acid aggres-

∗ Corresponding author. Tel.: +39 02 50314207; fax: +39 02 50314224.E-mail address: stefano.trasatti@unimi.it (S.P. Trasatti).

s

swiwi

921-5093/$ – see front matter © 2007 Elsevier B.V. All rights reserved.oi:10.1016/j.msea.2007.08.036

olution

iveness increases with decreasing number of carbon atomsn the alkyl chain: C4H9COOH < C3H7COOH < C2H5COOHCH3COOH < HCOOH.The presence of contaminants makes the picture more com-

lex, since they modify the oxidizing capacity of the acid mixturehus increasing the medium aggressiveness.

Among the organic acids, acetic and formic acids are mostrequently used as chemicals or solvents in many industrialrocesses. Consequently, knowledge of their corrosion activ-ty is essential. Copper and its alloys are suitable materials toandle similar mixtures, since copper alloys do not directlyisplace hydrogen. Unfortunately, copper increasing costs andifficulty of supply make this choice sometimes unpracticable.he use of austenitic and duplex stainless steels appears to beuccessful.

The corrosion behaviour of ferritic and austenitic stainlessteels in organic acids is well documented, but relatively few

orks are dedicated to duplex stainless steels. Corrosion stud-

es on DSSs in organic acid solutions are scanty in comparisonith similar studies in mineral solutions of remarkable industrial

nterest.

e and

sga

bctc

sado3f

irstapd

acm

esim

wana

(

(

2

2

aSt

pmb

edSfa

td

2

2

lamtwiwWtT

2

lw

adegreased with anhydrous ethanol for 5 min, and dried in warm

TC

SS

A.J. Invernizzi et al. / Materials Scienc

Scribner [1] reviewed the corrosion behaviour of stainlessteels in organic acids reporting higher performance of higherrade DSSs than the homologous of the AISI 300 series in bothcetic and formic acid solutions.

Sekine et al. [2–4] worked extensively on the corrosionehaviour of stainless steels in acetic and formic acids. Theyoncluded that the corrosion rate depends markedly on concen-ration and temperature, in turn closely connected with solutiononductivity, water and oxygen content.

Chechirlian et al. [5] studied the influence of formic acid andulphites on the corrosion behaviour of AISI 300 series in aceticcid. They observed that 20 ppm of sulphites are sufficient toepassivate AISI 304 in 50% acetic acid solution. The presencef Mo increases the critical concentration up to 200 ppm for AISI16 stainless steel, improving its corrosion performance also inormic acid.

Qi et al. [6] showed that traces of sulphuric acid (≤ 1wt%)n esterification processes have detrimental effect on the cor-osion resistance of AISI 300 series in acetic acid media,o much that these steels cannot be used in such condi-ions. The same authors stated that depolarizing ions, suchs Cu2+/Cu or Fe3+/Fe2+, are beneficial to restore a stableassive state, even though this process is temperature depen-ent.

Curtis [7] investigated the corrosion behaviour of SAF2205nd SAF2507 in organic acids, concluding that in the case ofhloride contamination, acetic anhydride or formic acid, the bestaterial is SAF2507.The selective dissolution in a duplex stainless steel (DSS) is

xpected to occur because of the difference in chemical compo-ition of the constituent phases. However, the results reportedn the literature are contradictory as to which phase dissolves at

ore accelerated rate [8–12].On the basis of the above considerations, an experimental

ork was carried out on SAF2205 and SAF2507 DSSs in organiccid aqueous solutions, in the presence of specific contami-ants, such as metal redox couples (Fe3+/Fe2+ and Cu2+/Cu)nd sulphuric acid.

The following targets were pursued:

(a) to bring an additional contribution to the scarce data avail-able in the literature on this specific topic;

b) to investigate in more details the selective dissolution ofduplex;

(c) to scrutinize a few operational parameters, notably acidnature and contaminant species;

d) to assess the possibility of improving the corrosionbehaviour adding oxidant species.

a

iv

able 1hemical composition % (wt) of SAF2205 and SAF2507

C Cr Ni Mo

AF2205 0.016 22.43 5.77 3.09AF2507 0.02 24.9 6.78 3.8

a According to P.R.E. = %Cr + 3.3x%Mo + 30x%N.

Engineering A 485 (2008) 234–242 235

. Experimentals

.1. Materials

The materials used in this study were two commerciallyvailable duplex stainless steels, SAF2205 (UNS S31803) andAF2507 (UNS S32750). The chemical composition (wt%) of

he two DSSs is reported in Table 1.Coupons for mass loss tests were obtained from 2 mm thick

lates while specimens for electrochemical experiments wereachined from 12 mm diameter round bars and 10 mm square

ars for SAF2205 and SAF2507, respectively.For metallographic examination, a few specimens were

mbedded in thermoplastic resin, polished with abrasive paperown to 1000 grit, and with diamond paste down to 0.25 �m.pecimens were electrochemically etched in KOH 10 M at 1.5 Vor 5 s. Typical microstructures for axial sections of SAF2205nd SAF2507 are shown in Fig. 1.

Both materials showed an equally distributed biphase struc-ure, about 50�/50�, with no inclusions or defects of criticalimensions.

.2. Testing procedure

.2.1. Mass loss testsThe 40 mm × 20 mm × 2 mm flat coupons were used for the

ong term exposure tests. The specimens were weighted beforend after each test to determine the corrosion rate. Mass losseasurements were carried out mainly at the boiling tempera-

ure under atmospheric pressure in 1 L glass cell, where couponsere hung by means of a glass hook, with the option of full

mmersion or vapour exposure. The cell was equipped with aater refrigerator to prevent fast evaporation of the solvent.eight loss experiments lasted 6–72 h depending on test solu-

ion. Surface area/solution volume ratio was about 0.02 cm−1.emperature was controlled to ± 1 ◦C by a Vertex.

.2.2. Electrochemical testsCylindrical (SAF2205: D = 10 mm, L = 15 mm) and paral-

elepipedal (SAF2507: A = 100 mm2, H = 15 mm) electrodesere used for the electrochemical experiments.Before each experiment, specimens were ground with

brasive papers down to 600 grit, rinsed in distilled water,

ir.The electrochemical behaviour in the testing solutions was

nvestigated by means of polarization and corrosion potentialersus time curves.

N Si Creq Nieq P.R.E.a

0.17 0.55 25 10.6 370.27 0.25 28.8 16 45.5

236 A.J. Invernizzi et al. / Materials Science a

aspo

firie

2

Caic

2

sm(

oS

3

3

aTfatthaMthat corrosion proceeded selectively with respect to microstruc-ture. A typical surface of SAF2205 after exposure is reportedin Fig. 3, showing the occurrence of selective dissolution ofaustenitic phase in 6% H2SO4 solution. Some dependence of

Fig. 2. Corrosion rate of SAF 2205 (�) and SAF2507 (♦) in H2SO4 solutions.

Fig. 1. Axial microstructure: (a) SAF2205; (b) SAF2507 (400X).

The polarization curves were recorded potentiodynamicallyt 0.1666 mV s−1 using a Model 273A EG&G potenzio-tat/galvanostat. The temperature range was 20 ◦C to boilingoint, controlled to ± 1 ◦C by a Vertex thermometer. Unlesstherwise specified, all curves were recorded under nitrogen.

A polarization cell of the type described in ASTM G5 speci-cations was used with a saturated calomel electrode (SCE) as aeference and two spirals of Pt as counter electrodes The stabil-ty of the reference electrode was checked before and after eachxperiment.

.2.3. Testing solutionsSolutions were prepared from analytical grade HCOOH,

H3COOH, H2SO4 and LiClO4. Anhydrous LiClO4 was useds a supporting electrolyte to reduce ohmic effects during polar-zation experiments. Doubly distilled water was used in allases.

.2.4. X-ray photoelectron spectroscopy (XPS)

X-ray photoelectron spectroscopy was used to analyse the

urface, using an M-probe apparatus (Surface Science Instru-ents). The source was monochromatic Al K� radiation

1486.6 eV). A spot size of 200 �m × 750 �m and a pass energy F

nd Engineering A 485 (2008) 234–242

f 156.42 eV were used. The background was subtracted usinghirley’s method [13].

. Results and discussion

.1. Mass loss tests

The corrosion rates of SAF2205 and SAF2507 in sulphuriccid solution are plotted against acid concentration in Fig. 2.ests were carried out to get reference data. SAF2205 showsairly uniform corrosion resistance up to 1% H2SO4, then it devi-tes from linearity with progressive increase in corrosion rate upo 0.7 mm/y in 4% acid. SAF2507 shows quite a similar trend inhe same concentration range, exhibiting half corrosion rate atigher acid content. Optical examination of specimen surfacesfter the tests revealed a rough look typical of uniform corrosion.etallographic analysis of cross sectioned specimens showed

ig. 3. Optical microstructure of SAF2205 after exposure in 6% H2SO4 (400X).

A.J. Invernizzi et al. / Materials Science and Engineering A 485 (2008) 234–242 237

F

to

wtacophsm

aowcotc0otsab

atdowaiadLtlIw

F(

iaee

t

(at2

rda

ccAt(Htbe explained in terms of conductivity change as water contentdecreases. Corrosion morphology was always uniform in ternarymixture solutions.

ig. 4. Corrosion rate of SAF 2205 (♦) and SAF2507 (�) vs. % HCOOH.

he selective dissolution on the solution aggressiveness wasbserved.

An extensive work on the selective dissolution of DSSsas conducted by Symniotis [9,10] who described the elec-

rochemical behaviour of duplex stainless steels in sulphuriccid + hydrochloric acid environments. At low hydrochloric acidontents, she showed that preferential dissolution of ferriteccurred at low potentials, the dissolution peaks of the twohases being clearly separated on a polarization curve. At higherydrochloric acid levels, the behaviour of the 22% Cr stainlessteel may change from preferential dissolution of the ferrite toore general corrosion.Typical corrosion rates of SAF2205 and SAF2507 in formic

cid solutions are shown in Fig. 4, together with the equationbtained by fitting the data to a linear dependence y = a + bx,here “y” is the corrosion rate in mm/y, “x” is the formic acid

oncentration in percentage, and “a” and “b” are the parametersf the linear regression. The fit regression parameter (r2) is closeo unity. The rate of attack increases slightly with increasingoncentration, but more than the commonly accepted value of.1 mm/y, the corrosion rate of SAF2507 being typically onerder of magnitude lower than that of AISI 304, and about halfhat of AISI316 in the same conditions [14]. SAF2205 behavesimilarly exhibiting a corrosion rate about twice that of SAF2507t the same concentration. Corrosion morphology was uniformut a sort of selective dissolution of austenite was still evident.

SAF2205 and SAF2507 do not undergo corrosion in aceticcid in a wide range of temperature and concentrations, but inhe presence of traces of sulphuric acid the situation changesramatically. The effect of sulphuric acid on the corrosion ratef DSSs was checked in 80% acetic acid solution. This solutionas chosen since the corrosion rate becomes appreciable only

t this concentration. Fig. 5 shows corrosion data for both DSSsn boiling 80% AcOH solutions containing traces of sulphuriccid. The presence of sulphuric acid increases the corrosion rateramatically, suggesting a synergic effect between the two acids.ow levels of water and dissolved oxygen cause breakdown of

he passive film even for solutions containing only 0.1% H SO ,

2 4eading to unacceptable corrosion rates (higher than 0.1 mm/y).n 80% HCOOH solutions, sulphuric acid has the same effectith a more marked increase of corrosion rate. Since HCOOH

F%

ig. 5. Corrosion rate of both DSSs in 80% AcOH + %H2SO4: (♦) SAF2205;�) SAF2507.

s more dissociated than CH3COOH, the corrosion rate is highert the same concentration. At H2SO4 concentrations > 0.5% theffect of organic acids becomes negligible, corrosion being gov-rned by the inorganic acid (Fig. 6).

The corrosion rate of SAF2205 is about four times higherhan that of SAF2507 at the same concentration.

The corrosion behaviour of DSSs in ternary solution mixturesHCOOH/CH3COOH/H2SO4) is shown in Figs. 7 and 8. Theddition of up to 5% HCOOH at a 80% AcOH + 1% H2SO4 solu-ion causes the corrosion rate of SAF2507 to increase to about.5 mm/y. Further addition has no effect up to 15% HCOOH.

Reduction of the H2SO4 content to 0.1% causes the corrosionate to increase linearly with HCOOH content up to 12%, slightlyecreasing for further addition. However the corrosion rate islways lower than 0.6 mm/y.

When AcOH is added to 80% HCOOH + 1% H2SO4, theorrosion rate versus wt% varies linearly with a small slope:orrosion rate goes from 1.5 to 3.8 mm/y by increasing thecOH content from 0 to 16%, respectively. With SAF2205,

he corrosion behaviour in ternary solutions is quite similarFig. 8). For example, the corrosion rate in CH3COOH + 1%

2SO4 is around 5 mm/y with the HCOOH concentration inhe 0–14% range. This uncertain and unexpected behaviour can

ig. 6. Corrosion rate of SAF 2205 (♦) and SAF2507 (�) in 80% HCOOH +H2SO4.

238 A.J. Invernizzi et al. / Materials Science and Engineering A 485 (2008) 234–242

FHH

SFucaA

o

bi

area

FHH

sufficient to double the corrosion rate measured in the absenceof sulphites.

For further addition, the corrosion rate increases linearly withsulphite concentration.

ig. 7. Corrosion rate of SAF2507 in: (a) 80% AcOH + 1% H2SO4 + %COOH; (b) 80% HCOOH + 1% H2SO4 + % AcOH; (c) 80% AcOH + 0.1%

2SO4 + % HCOOH.

The influence of temperature on the corrosion rate ofAF2507 and SAF2205 in 80% AcOH + 1% H2SO4 is shown inig. 9. Temperature affects kinetics markedly, making both DSSsnusable at temperatures > 90 ◦C. An increase in temperatureauses an increase in corrosion rate up to the boiling temper-ture, with an Arrhenius-like behaviour, even for DSS in 80%cOH + 0.1% H2SO4.The corrosion rate in 80% AcOH + 0.1% H2SO4 is about one

rder of magnitude lower than in 1% H2SO4.SAF2507 in 80% HCOOH + 1% H2SO4 exhibits similar

ehaviour with corrosion rates similar or slightly lower thann AcOH at temperatures > 90 ◦C.

Besides temperature, the passive state of DSSs in organic

cid solutions is strongly influenced by species present as impu-ities or by-products [15–18]. This is the case of sulphites whoseffect on the corrosion rate of SAF2205 was checked in aceticcid solutions. As shown in Fig. 10, 20 ppm of sulphites were

Fo

ig. 8. Corrosion rate of SAF2205 in: (a) 80% AcOH + 1% H2SO4 + %COOH; (b) 80% HCOOH + 1% H2SO4 + % AcOH; (c) 80% AcOH + 0.1%

2SO4 + % HCOOH.

ig. 9. Corrosion rate of both DSSs in 80% AcOH + 1% H2SO4 as a functionf temperature: (♦) SAF2205; (�) SAF2507.

A.J. Invernizzi et al. / Materials Science and Engineering A 485 (2008) 234–242 239

Ft

scf

N

S

H

Smh

2

E

4

E

4

E

2

E

ts

ao8r

asfibTe

Table 2Corrosion rate of SAF2205 in 80% AcOH at different H2SO4 content

Testing solution

% H2SO4 With O2 Without O2

001

mAiciae

tsp

e

tt

3

stid

SFoIwas observed in the active-to-passive transition region.

The results are consistent with those of mass loss, as SAF2205in 80% CH3COOH + 1% H2SO4 corroded uniformly only closeto the boiling temperature.

Table 3Corrosion rate for SAF2205 as a function of metal ions in 80%AcOH + x%H2SO4

Testing solution Corrosion rate (mm/y)

80%AcOH + 0.1% H2SO4 + 20 ppm Cu2+(acetate) 0.00180%AcOH + 0.1% H2SO4 + 20 ppm Fe2+(acetate) 0.00380%AcOH + 0.1% H2SO4 + 20 ppm Fe3+(sulphate) 0.00280%AcOH + 1% H2SO4 + 20 ppm Cu2+(acetate) 0.0180%AcOH + 1% H2SO4 + 100 ppm Cu2+(acetate) 0.00680%AcOH + 1% H2SO4 + 100 ppm Cu2+(sulphate) 0.002

ig. 10. Relationship between corrosion rate and sulphites in 80% AcOH solu-ion (SAF2205).

The action of sulphite as oxygen scavenger in near neutralolutions is well known [19]. In acid media, the observed trendan be explained with the formation of SO2 according to theollowing chemical equilibria:

aSO3 + H2SO4 ⇔ Na2SO4 + H2O + SO2

O2•H2O ⇔ H2SO3 ⇔ HSO3

− + H+

SO3− ⇔ SO3

2− + H+

O2 and HSO3−, once formed, can easily be reduced on the

etal surface according to one of the following electrochemicalalf-reactions [20–22]:

HSO3− + 2H+ + 2e− → S2O4

2− + 2H2O◦ = 0.099 V versus NHE

H2SO3 + 4H+ + e− → S4O62− + 6H2O

◦ = 0.507 V versus NHE

HSO3− + 8H+ + 6e− → S4O6

2− + 6H2O◦ = 0.577 V versus NHE

SO2(aq.) + 2H+ + 4e− → S2O32− + H2O

◦ = 0.400 V versus NHE

hus increasing the cathodic current needed to sustain the corro-ion process.

The low stability of the SAF2205 passive film in organiccid aqueous solutions containing sulphuric acid in the absencef oxidizing species, was also confirmed by a few tests in0% AcOH + x% H2SO4 under continuous nitrogen purging toemove dissolved O2 (Table 2).

Results showed that corrosion increased markedly: thebsence of dissolved oxygen did not allow the formation of atable passive film, presumably replaced by a non protective salt

lm, as deduced from the opaque look of the coupons, coveredy green-yellow layers, likely chromium and nickel compounds.he attack exhibits uniform corrosion morphology with prefer-ntial dissolution of the austenitic phase.

8888

.1 0.01 0.3

.5 0.11 1.70.9 3.1

The need for oxidizing conditions was investigated by addingetal ions in the higher oxidation state (e.g. Fe3+ and Cu2+).few tens of ppm were sufficient to make DSS passive even

n the most aggressive environments. Table 3 summarizes theorrosion rate of SAF2205 in 80% CH3COOH + 0.1% H2SO4n the presence of diverse metal ions. At 1% H2SO4 an highermount of oxidizing species is needed to obtain the same positiveffect.

The effect of the nature of counterions, acetate or sulphate,urns out negligible with the exception of Fe2+ added as acetatealt. In that case the corrosion rate is markedly lowered if com-ared with the effect of Fe2+ added as sulphate.

This unusual behaviour was already reported in the literatureven though for different operating conditions [6].

We also observed that in order to established oxidizing condi-ions, the addition of 0.2% of hydrogen peroxide was sufficiento stop corrosion completely.

.2. Electrochemical experiments

The anodic behaviour of SAF2205 and SAF2507 is quiteimilar in all aqueous organic acid solutions, showing the exis-ence of two passive regions tending to merge as temperaturencreases. The curves show a wide passive range and the break-own potential does not change significantly with temperature.

Typical anodic potentiodynamic polarization curves forAF2205 in 80% AcOH from 20 to 100 ◦C are reported inig. 11. No distinctive active-to-passive transition region isbserved until temperatures close to the boiling point are tested.n these operating conditions, enhanced anodic current density

0%AcOH + 1% H2SO4 + 20 ppm Fe2+(acetate) 0.0130%AcOH + 1% H2SO4 + 100 ppm Fe2+(acetate) 0.0110%AcOH + 1% H2SO4 + 100 ppm Fe2+(sulphate) 3.3750%AcOH + 1% H2SO4 + 100 ppm Fe3+(sulphate) 0.007

240 A.J. Invernizzi et al. / Materials Science and Engineering A 485 (2008) 234–242

Fig. 11. SAF2205: anodic polarisation in 80% AcOH + 1% H2SO4 solution atdifferent temperatures.

Ft

im

H(

p

av1f

Table 4Atomic ratio of Cr/Fe and Si/Fe in the passivated film, determined by XPS after30 min in (a) 80%AcOH + 1%H2SO4 (SAF2205); (b) 80%HCOOH + 1%H2SO4

(SAF2507)

Cr/Fe Si/Fe

SAF2205 at E1 2.06 1.41SAF2205 at E2 1.22 1.19SAF2507 at E1 0.61 1.30S

pCLb

fifEriSri

eo

S

Dfieaio

c

ig. 12. SAF2507: anodic polarisation in 80% HCOOH + 1% H2SO4 at differentemperatures.

The anodic behaviour in 80% AcOH + 0.1% H2SO4 was sim-lar but shifted to lower current density by about an order of

agnitude.Fig. 12 shows anodic polarization curves for SAF2507 in 80%

COOH + 1% H2SO4 at temperatures in the range 70–108 ◦Cboiling point).

At temperatures lower than 90 ◦C, SAF2507 exhibits a wideassive region. Only a small active peak is visible at 90 ◦C.

In contrast, at temperatures higher than 90 ◦C, a pronounced

ctive region appeared around Ecorr + 50 mV. In particular, thealue of the critical passivation current density (icr) goes from0−6 A/cm2 to about 10−4 A/cm2 with increasing temperaturerom 70 to 108 ◦C (boiling point).

c

pfi

Fig. 13. XPS survey of SAF2507 in

AF2507 at E2 1.38 1.41

A sharp increase in current occurred at about Ecorr + 1200 mV,ossibly due to either oxidation of Cr3+ in the passive film tor6+ species, or oxygen evolution, or a combination of both.ocalized corrosion was not observed on the electrode surfacey optical examination after each polarization experiment.

To obtain data about the composition of surface oxidelms, some specimens were subjected to potentiostatic agingor 30 min at two potentials, E1 = 350 mV versus SCE and2 = 850 mV versus SCE, located in the two different passive

anges, respectively (see Figs. 11 and 12). This is to facil-tate the formation of oxide film over the surface of bothAF2205 and SAF2507. Potentiostatic experiments were car-ied out at 70 ◦C in 80%AcOH + 1%H2SO4 with SAF2205 andn 80%HCOOH + 1%H2SO4 with SAF2507.

Surface films were analysed with XPS that revealed the pres-nce of chromium oxide (Cr2O3), silicon oxide (SiO2) and ironxide (Fe2O3) as reported in recent studies [23,24].

Data are reported in Table 4, while a typical XPS survey forAF2507 in 80% HCOOH + 1% H2SO4 is shown in Fig. 13.

The findings indicate that the second passive state for bothSSs was similar, but iron enrichment occurred in the passivatedlm of SAF2507 “aged” at E1 with respect to SAF2205. How-ver, the presence of nitrogen in the passivated film of SAF2507t E1, disappearing at E2 should be noted. This observation is notn agreement with previous studies where nitrogen enrichmentf the oxide/metal interface was observed [23].

No significant variation in the Si/Fe ratio was observed, indi-ating the same level of silicon in the passivated film in everyondition.

The presence of the S2s peak in all conditions reveals thearticipation of sulphur species in the formation of the oxidelm.

80% HCOOH + 1% H2SO4.

A.J. Invernizzi et al. / Materials Science and Engineering A 485 (2008) 234–242 241

Fi

p

t

haaToiba

ctAvmiic

spc

Fa

tS

4

st

Dtib

aptti

wm

R

ig. 14. Cathodic polarization curves as a function of temperature for SAF2205n: (a) 80% AcOH + 0.1% H2SO4; (b) 80% AcOH + 1% H2SO4.

The cathodic behaviour of SAF2205, as a function of sul-huric acid content and of temperature is reported in Fig. 14.

The corrosion potential is appreciably modified by tempera-ure, shifting to more positive values as temperature increases.

Temperature causes, as expected, the cathodic current ofydrogen evolution to increase; this reaction takes place alreadyt about −200 mV versus SCE. At less negative potentialsnother process, more evident at high temperature, is observed.his process can be associated with the reduction of surfacexides, whose kinetics became higher as temperature and acidityncrease. This phenomenon is less pronounced in 0.1% H2SO4,ut at high temperatures it becomes evident again also in lowcidity media.

The effect of addition of hydrogen peroxide on the electro-hemical behaviour of SAF2205 is shown in Fig. 15, wherehe time variation of the corrosion potential of SAF2205 in 80%cOH + 1% H2SO4 is reported. The potential starts at −0.170 Versus SCE, but after about 20 min it settles at −0.275 V. Theaterial stays in this active state until the solution aggressiveness

s changed, for example, by adding H2O2. As hydrogen perox-de is added the potential shifts quickly up to about 800 mV, thatorresponds to the reversible potential of the H2O2/O2 couple.

Such a high potential is obtained both for thermal decompo-

ition and for catalytic degradation of H2O2 favoured by theresence of metal ions in solution. As the decomposition isomplete, the potential reaches a stable value of about 0.600 V

[[

ig. 15. Free corrosion potential in 80% AcOH + 1% H2SO4 at boiling temper-ture: effect of H2O2 adding for SAF2205.

hat can be assigned to the Fe3+/Fe2+surface redox reaction.AF2205 is now in the passive state.

. Conclusions

The corrosion behaviour of DSSs in organic acid aqueousolutions is strongly dependent on the acid content and nature,emperature, and contaminant type.

In aqueous solution of acetic acid SAF2205 and SAF2507SSs do not undergo corrosion in a wide range of acid concentra-

ion and temperature. SAF2507 is not subject to severe corrosionn formic acid up to 40% of concentration at the atmosphericoiling temperature.

The presence of sulphuric acid or sulphites changes the situ-tion completely with a dramatic decrease of the DSS corrosionerformance in both acids. As observed, water and oxygen, orhe presence of trace impurities, such as redox couples in solu-ion, help to maintain the protective oxide layer on DSSs, evenn the most aggressive environments.

Preferential dissolution of austenitic phase, more pronouncedith increasing solution aggressiveness, occurs in all experi-ents.

eferences

[1] L.A. Scribner, Corrosion by organic acids, Proceedings of Corrosion 2001,NACE International, TX Huston, USA, Paper No. 01343, 2001.

[2] I. Sekine, T. Kawase, M. Kobayashi, M. Yuasa, Corros. Sci. 32 (1991)815–825.

[3] I. Sekine, S. Hatekeyama, Y. Nakazawa, Electrochim. Acta 32 (1987)915–920.

[4] I. Sekine, A. Masucko, K. Senoo, Corrosion 43 (1987) 553–560.[5] S. Chechirlian, P. Eichner, H. Mazille, Influence of formic acid and sulfite on

the corrosion behaviour of austenitic stainless steels in acetic acid solutions,in: Proceedings of the 11th International Corrosion Congress, vol. 2, AIM,Firenze 2–6 aprile, 1990, p. 577.

[6] J.S. Qi, G.C. Lester, Corrosion 52 (1996) 558–565.[7] N. Curtis, Chem. Eng. 107 (2000) 94–98.[8] I.H. Lo, Y. Fu, C. Lin, W.T. Tsai, Corros. Sci. 48 (2006) 696–708.

[9] E. Symniotis, Corrosion 46 (1990) 2–12.10] E. Symniotis, Corrosion 51 (1995) 571–580.11] E. Schmidt-Rieder, X.Q. Tung, J.P.G. Farr, M. Aindow, Br. Corros. J. 31

(1996) 139–145.

2 nce a

[

[[

[[[

[[

42 A.J. Invernizzi et al. / Materials Scie

12] M. Femenia, J. Pan, C. Leygraf, P. Luukkonen, Corros. Sci. 43 (2001)1939–1951.

13] D.A. Shirley, Phys. Rev. B 5 (1972) 4709–4714.

14] Corrosion resistance of nickel-containing Alloys in Organic Acids and

Related Compounds, NiDI, publication No. 1285.15] P. Kangas, M. Newman, Anti-Corros. Methods Mater. 45 (1998) 233–238.16] Hd. Pletka, Kg. Schutze, Werkst. Korros. 43 (1992) 364–369.17] J. Duchardt, G. Grotsch, O. Schubert, Werkst. Korros. 42 (1991) 118–127.

[[[[[

nd Engineering A 485 (2008) 234–242

18] Kg. Hd Pletka, Schutze, Werkst. Korros. 45 (1994) 325–330.19] S. Muralidharan, K. Chidambaram, N. Palaniswamy, K. Balakrishnan, Bull.

Electrochem. 5 (1989) 857–860.

20] T. Hemmingsen, Electrochim. Acta 37 (1992) 2775–2784.21] T. Hemmingsen, Electrochim. Acta 37 (1992) 2785–2790.22] J.E. May, C.A.C. de Sousa, S.E. Kuri, Corros. Sci. 45 (2003) 1395–1403.23] C.A. Olsson, Corros. Sci. 37 (1995) 467–479.24] H. Weng, G. Teeter, J. Turner, J. Electrochem. Soc. 152 (2005) B99–B104.