Thermodynamics of Inclusion Formation and Its Influence

on the Corrosion Behavior of Cu Bearing Duplex Stainless Steels

Soon-Hyeok Jeon1, Soon-Tae Kim1, In-Sung Lee1, Joo-Hyun Park2,Kwang-Tae Kim3, Ji-Soo Kim3 and Yong-Soo Park1;*

1Department of Material Science and Engineering, Yonsei University, 134 Shinchon-dong, Seodaemun-gu, Seoul 120-749, Korea2School of Material Science and Engineering, University of Ulsan, Moogeo-dong, Nam-gu, Ulsan 680-749, Korea3Stainless Steel Research Group, POSCO Technical Research Laboratories,Goedong-dong, Nam-Gu, Pohang, Gyeongbuk 790-785, Korea

To elucidate the thermodynamics of inclusion formation and its influence on the corrosion behavior of Cu bearing duplex stainless steels,potentiodynamic and potentiostatic polarization tests, a SEM-EDS analysis of inclusions, and thermodynamic calculations of the formation ofinclusions were carried out. While the resistance to general corrosion of the noble copper contained alloy-1.5Cu in a deaerated 2M H2SO4 washigher than that of the alloy-BASE, the resistance to pitting corrosion of copper contained alloy-1.5Cu in a deaerated 0.5N HCl + 1N NaCl and30mass% NaCl was lower than that of the alloy-BASE due to an increase of interface areas between inclusions and matrix acting as preferentialpit initiation sites. The thermodynamic calculation for the formation of Cr-containing oxide inclusions was in good agreement with theexperimental results. [doi:10.2320/matertrans.M2010355]

(Received October 12, 2010; Accepted December 20, 2010; Published February 9, 2011)

Keywords: copper, duplex stainless steel, potentiostatic, scanning electron microscope (SEM), inclusion, pitting corrosion

1. Introduction

Duplex stainless steel (DSS) is the stainless steel that hasmicrostructure where both ferrite and austenite phases arepresent in approximately equal volume fraction. Duplexstainless steel has a high resistance to pitting and crevicecorrosion, high mechanical properties, and a better costperformance than austenitic stainless steel because of itslower Ni content. Highly alloyed DSS with Cr, Mo and N hasbeen increasingly used for various applications such as powerplants, desalination facilities and chemical plants due to theexcellent resistance to localized corrosion and stress corro-sion cracking (SCC).1–3)

It has been well known that the addition of copper (Cu) toferritic, austenitic or duplex steels improves the resistance touniform corrosion in sulfuric media.4–8) It was reported in theprevious studies that the mechanism on the beneficial effectof Cu addition on the steels is based on the suppressionof anodic dissolution by elemental Cu deposition on thecorroded surface of stainless steels.9)

In acid chloride media, the studies of redeposited metalliccopper have been conducted due to the presence of Cl� ions.Hermas et al.10) assumed that Cu additions diminish thecorrosion rate and the stainless steel dissolution in dilutedacid chloride solutions as a result of the accumulation ofmetallic Cu on the surface and the later formation of CuCl,which protects the oxide film. In a comparison with the roleof copper in stainless steels in acidic solutions, Ujiro et al.11)

showed that the protective effect of the redeposited metalliccopper at the active alloy surface can also occur in acidicchloride media, but it is less effective since the presenceof Cl� ions diminishes the stability of the deposits, whichdissolve as CuCl2 in solutions of low pH and high chlorideconcentration.

In general, pit initiation occurs as a result of creviceformation at between the inclusion and the metallic matrix.12)

The interface area between the inclusion and the metallicmatrix appears to affect the resistance to pitting corrosion bysupplying the preferential area as a role of pit initiation site.

The initiation mechanism of pitting corrosion due tosulfide inclusions has mainly been issued during severaldecades. Recently, Williams et al.13) found that a thin FeSlayer would be formed around the MnS inclusions, withinwhich a pit can be triggered. Zheng et al.14) showed that anumber of nano-sized MnCr2O4 inclusion particles wereembedded in the MnS medium, which generating localMnCr2O4/MnS nano-galvanic cells and this acts as thereactive site and catalyses the dissolution of MnS.

Meanwhile, metastable pit growth is a well-documentedfeature of the pitting corrosion of stainless steel in chloridesolutions.15–17) Metastable pits of stainless steels often formin the entire range of passivating potential.18,19) The numberof metastable pitting events has been shown to be a functionof the potential,20–22) the potential scan rate,23,24) the chlorideconcentration in the solution,25,26) the oxide thickness and thealloy composition.23) Burstein assumes that there are twodistinct processes before stable pit formation occurs: pitnucleation and growth of the metastable pit.25,27)

In this study, to elucidate the thermodynamics of inclusionformation and its influence on the corrosion behavior of Cubearing duplex stainless steels, potentiodynamic and poten-tiostatic polarization tests, a scanning electron microscope-energy dispersive spectroscope (SEM-EDS) of inclusionsand thermodynamic calculation for the formation of inclu-sions were conducted.

2. Experimental Procedures

2.1 Material and heat treatmentThe experimental alloys were manufactured using a high*Corresponding author, E-mail: yongsoop@yonsei.ac.kr

Materials Transactions, Vol. 52, No. 3 (2011) pp. 416 to 422#2011 The Japan Institute of Metals

frequency vacuum induction furnace and then hot rolled toplates of 6mm thickness. The experimental alloys were cutand solution heat-treated for 5min per 1mm thickness at1090�C and then quenched in water. Chemical compositionsof the experimental alloys are presented in Table 1.

2.2 Computation of inclusion formation behaviorTo elucidate the effects of copper addition on the

formation of inclusions and the resistance to pitting corrosionof alloys, the commercial thermochemical computing soft-ware, FactSage� (ver. 6.1, Thermfact & GTT-Technology)was used with the Fact53 compound database, FToxid oxidedatabase, and FSstel steel alloy database.28) The FToxiddatabases contain data for pure oxides and oxide solutions of20 elements. First, for major oxide components (MOCs) suchas Al2O3, CaO, FeO, Fe2O3, MgO, and SiO2, all MOCs havebeen fully optimized and evaluated together at all composi-tions. All available data for binary, ternary and quaternarysub-systems have been fully optimized. Second, for thesystems containing MnO with the MOCs, most binary andternary sub-systems between MnO and the MOCs have beenevaluated and optimized. Third, when the CrO and Cr2O3 arepresent but not containing SiO2, the systems Al2O3-CaO-CrOx-FeOx-MgO-NiO has been extensively optimized overmost composition regions where the data are available.Finally, for the systems containing CrO and Cr2O3 with theMOCs, the Al2O3-CaO-CrOx-SiO2 system has been fullyoptimized. The FSstel database is based on relevant steelsub-systems from the old SGTE SGSL solution databasebut the model and data for different liquid MeO and Me2Ooxide associates have been incorporated in the liquid phase.These are used in conjunction with the Fact53 database forcalculations relating to equilibrium concentrations of oxygenresulting from particular processes such as deoxidation ofsteel during melting and casting processes. The liquid steelphase is also compatible with the data for oxide solutionsand stoichiometric oxides in the FToxid database.28) Thisdatabase has successfully been used in the evaluation andmodeling of various oxide systems.29–33) The more details forthe documentation on regard of FactSage� program and itsmodeling developments are available in the literatures.34)

2.3 Corrosion testsTo analyze the effect of copper addition on the resistance

to pitting corrosion of the experimental alloys, a potentiody-namic anodic polarization test was conducted. The pittingpotential (Ep) and the passive current density were conductedin a deaerated 25mass% NaCl solution at 70�C and adeaerated 30mass% NaCl solution at 75�C and a deaerated0.5N HCl + 1N NaCl solution at 40�C and a deaerated 2MH2SO4 solution at 60�C according to the ASTM G 5.35) Test

specimens were joined with copper wire through soldering(95mass% Sn-5mass% Sb), and then mounted with an epoxyresin. One side of the sample was ground to 600 grit usingSiC abrasion paper. After defining the exposed area of thetest specimen as 1 cm2, the remainder was painted with atransparent lacquer. The test was conducted at a potentialrange of �0:65VSCE � þ1:1VSCE and a scanning rate of0.06V/min, using a saturated calomel electrode.

The current transients through the potentiostatic test weremeasured in a deaerated 25mass% NaCl solution at 70�Cwith an applied potential of 0VSCE in the passive regionof the potentiodynamic anodic polarization curves at whichmetastable pitting can occur.27,36) The current transients wererecorded for a duration of 3600 s.

A critical crevice temperature (CCT) test was conducted in6mass% FeCl3 + 1mass% HCl of pH + 0 per ASTM G 48-Method F.37) The specimens were ground to 100 grit usingSiC abrasion paper. The initial temperature of 6mass%FeCl3 + 1mass% HCl for the critical crevice temperaturetest was 25�C. The solution temperature was increased by5�C per 24 h from the initial temperature of 25�C accordingto ASTM G 48-Method F. After the test was completed,corrosion products were removed in acetone. Pitting corro-sion is considered to be present if the local attack is 0.025mmor greater in depth.

2.4 Micro-structural characterizationTo observe the optical microstructures of the alloys, they

were ground to 2000 grit using SiC abrasive papers, polishedwith diamond paste, and then electrolytically etched using10mass% KOH.

The chemical compositions of various inclusions wereanalyzed using a SEM and an EDS attached to a SEM.

3. Results and Discussion

3.1 Effect of Cu addition on the distribution of inclu-sions

Figure 1 shows back-scattered electron (BSE) images ofthe inclusions in the experimental alloys. The chemicalcompositions and the morphologies of the inclusions of theblack spots were analyzed using a SEM-EDS. As presented inFig. 1, the inclusions in the alloy-BASE were composed ofthe main type of (Mn, Cr, Al) oxides and (Mn, Cr, Fe, Al)oxides. The inclusion in the Cu added alloy-1.5Cu werecomposed of the main type of (Mn, Cr, Al) oxides and (Mn,Cr, Fe, Al) oxides. The chemical compositions of inclusionsin the Cu added alloy-1.5Cu are similar to those in the alloy-BASE.

Figure 2(a) presents the effects of the copper addition onthe distribution of inclusions per frame area in the exper-

Table 1 Chemical compositions of the experimental alloys (mass%).

Alloy

DesignationC Cr Ni Mo W Si Mn Cu S N Fe PREN�

BASE 0.020 27.29 7.06 2.58 3.39 0.22 1.46 — 0.0037 0.33 Bal. 51.3

1.5Cu 0.031 27.35 6.60 2.58 3.42 0.30 1.43 1.32 0.0040 0.33 Bal. 51.4

�PREN (Pitting Resistance Equivalent Number) = mass% Cr + 3.3(mass% Mo + 0:5xmass%W) + 30xmass% N

Thermodynamics of Inclusion Formation and Its Influence on the Corrosion Behavior of Cu Bearing Duplex Stainless Steels 417

imental alloys. Notably, the alloy-1.5Cu had inclusions ofvarious sizes ranging from fine inclusions to coarse inclu-sions as large as 13 mm, and had a large number of inclusions,compared with those of the alloy-BASE.

As copper content increased, the number and the areafraction of inclusions per frame area increased. The numberof inclusions per frame area in the alloy-1.5Cu was increasedby about 1.5 times, compared with those of the alloy-BASE(Fig. 2(b)). The area fraction of inclusions per frame area in

the alloy-1.5Cu was increased by about 2.5 times, comparedwith those of the alloy-BASE (Fig. 2(b)). Figure 2(c) showsthat the effects of the copper addition on the length ofinclusions per frame area in the experimental alloys. Themean length (2.18 mm) of the inclusions in the alloy-1.5Cuwas much bigger than that (1.52 mm) of the alloy-BASE.The maximum length (13.43 mm) of the inclusions in thealloy-1.5Cu was bigger than that (11.19 mm) of the alloy-BASE.

(a)

Mn : 26.55 %

Cr : 32.63 %

Fe : 2.61 %

Al : 7.53 %

O : 30.67 %

150µm

1.5µm

µm1.5

Mn : 25.18 %

Cr : 35.54 %

Al : 8.90 %

O : 30.37 %

(Mn,Cr,Fe,Al)O

(Mn,Cr,Al)O

(b)

Mn : 25.87 %

Cr : 39.00 %

Al : 5.17 %

O : 29.96 %

(Mn,Cr,Al)O

1.5

1.5

Mn : 19.25 %

Cr : 36.25 %

Fe : 9.62 %

Al : 5.54 %

O : 29.35 %

(Mn,Cr,Fe,Al)O

150

µm

µm

µm

Fig. 1 Back-scattered electron (BSE) images and SEM-EDS of the inclusions in the experimental alloys: (a) the alloy-BASE and (b) the

alloy-1.5Cu.

BASE0

100

200

300

400

500

600

700

800

900

1000

0

100

200

300

400

500

600

700

800

900

1000

Th

e n

um

ber

of

Incl

usi

on

s

/Fra

me

Are

a (5

.93x

105

µm2 ) T

he area o

f Inclu

sion

s ( µm

2)

/Fram

e Area (5.93x10 5µ

m2)

The area of inclusionsThe number of inclusions

Experimental Alloys0

0

10

20

30

40

50

60 BASE 1.5Cu

Size of inclusions (µm)

Th

e n

um

ber

of

Incl

usi

on

s

/Fra

me

Are

a (5

.93x

105

µ m2 )

(a)

Mean length0123456789

1011121314

Th

e le

ng

th o

f in

clu

sio

ns

(µm

)

/Fra

me

Are

a (5

.93x

105

µm2 )

BASE 1.5Cu

1413121110987654321 1.5Cu

(b)

(c)

Min. lengthMax. length

Fig. 2 Effect of copper addition on the distribution, the number, the area and the length of inclusions per frame area of the experimental

alloys: (a) the distribution of inclusions, (b) the number and the area of inclusions and (c) the length of inclusions.

418 S.-H. Jeon et al.

Figure 3 shows the average composition of the inclusionsin the alloys investigated in the present study. It seems thatthe compositions of the inclusions are quite similar betweenthe alloy-BASE and alloy-1.5Cu.

3.2 Effects of Cu addition on the formation behavior ofinclusions

The effect of copper on the activity of chromium can beanalyzed using the interaction parameter concept. In a multi-component system, the effect of various elements on theactivity coefficient of component i ( fi) is expressed byformula (1).38)

log fi ¼ log f ii þ log fji þ log f ki þ . . . ð1Þ

where fji is the effect of component j on the activity

coefficient of component i. Since log fi is some function ofthe weight percents of components i; j; k; . . ., the Taylor-series expansion yields,

log fi ¼ ðmass%iÞ@ log fi

@ðmass%iÞþ ðmass%jÞ

@ log fi

@ðmass% jÞ

þ ðmass%kÞ@ log fi

@ðmass%kÞþ . . . ð2Þ

Here, the interaction parameter eji is defined as

eji ¼

@ log fi

@ðmass% jÞð3Þ

The effect of copper on the activity coefficient of chromiumis known as formula (4) and thus be calculated as a functionof copper content as shown in Fig. 4.39)

eCuCr ¼@ log fCr

@ðmass%CuÞ¼ 0:016 ð4Þ

As shown in Fig. 4, the activity coefficient of chromiumlog f CuCr linearly increases with an increase of copper content.This means that the addition of copper enhanced the drivingforce of the formation reaction of Cr-containing oxide phasesduring melting and solidification processes.

More quantitative analysis for the formation behavior ofoxide and sulfide inclusions in alloys BASE and 1.5Cu iscarried out with an aid of a commercial thermochemicalcomputing software, FactSage�6.1.18,34) The calculatedamounts of precipitated oxide and sulfide inclusions areplotted in Fig. 5 as a function of copper content at 1090�C,

which is relatively close to the solution heat-treatmenttemperature. The precipitated content of MnAl2O4 (spinel)inclusion decreases at copper content greater than about0.5mass% and disappears at ½Cu� > 1mass%. The MnAl2O4

phase is non-stoichiometric oxide compound, which meansthat there is a solubility of Al2O3 and MnO in this phase.Actually, we found that the content of alumina of theinclusions in BASE alloy is larger than that found in 1.5Cualloy. Consequently, even though we can not exactlyconfirm the stoichiometry of the spinel-type inclusionsat this time, it is believed that the MnO-Al2O3(-Cr2O3)inclusions would be formed in BASE alloy. However, thecontent of Cr-containing oxide inclusions such as corundum(87mass%Al2O3-13mass%Cr2O3) and tetragonal spinel(Cr[Mn,Cr]2O4) phases increases with increasing coppercontent greater than 0.5mass%, followed by saturation withthe content of 170 ppm and 75 ppm, respectively. The contentof MnS does not change through the content of copperinvestigated in the present study. This computing resultindicates that the addition of copper thermodynamicallyenhances the formation of Cr-containing oxide inclusions,which is in good agreement to the present findings shown inFig. 1 and Fig. 2.

3.3 Effects of Cu addition on the corrosion behaviorFigure 6 shows the effect of copper addition on the critical

crevice temperature (CCT) of the experimental alloys in

0

10

20

30

40

50

60

SiO2

Al2O

3MnO Cr

2O

3

Th

e C

on

ten

t o

f in

clu

sio

ns(

%)

Type of Inclusions

BASE 1.5Cu

FeO

Fig. 3 The contents on type of inclusions of experimental alloys.

0.00.000

0.005

0.010

0.015

0.020

0.025

0.030

log

f CrC

u

Content of Cu, wt%2.01.51.00.5

Fig. 4 Effect of copper on the activity coefficient of chromium in molten

alloy-BASE.

0.00

50

100

150

200

250

300

MnAl2O

4 spinel

Cr(Mn,Cr)2O

4 tetragonal spinel

MnS

87%Al2O

3-13%-Cr

2O

3 (corrundum)

Con

tent

of p

reci

pita

tes,

ppm

Content of Cu, wt%

2.01.51.00.5

Fig. 5 Content of precipitated oxide and sulfide inclusions as a function of

copper content at 1090�C.

Thermodynamics of Inclusion Formation and Its Influence on the Corrosion Behavior of Cu Bearing Duplex Stainless Steels 419

6mass% FeCl3 + 1mass% HCl per ASTM G 48-method F.The resistance to pitting corrosion of copper contained alloy-1.5Cu was lower than that of the alloy-BASE due to adecrease in the CCT.

Figure 7 shows the effect of copper addition on thepotentiodynamic polarization behavior of the experimentalalloys in a deaerated 2M H2SO4 solution at 60�C accordingto ASTM G 5. Based upon a decrease of the critical passivecurrent density and primary passivation potential in activeregion, the resistance to general corrosion of the noble coppercontained alloy-1.5Cu in a deaerated 2M H2SO4 solutionwas higher than that of the alloy-BASE.

Figure 8 shows the effect of copper addition on thepotentiodynamic polarization behaviors of the experimentalalloys in a deaerated 0.5N HCl + 1N NaCl solution at 40�Cper ASTM G 5. Based upon an increase of the passive currentdensity in the passive region and the critical passive currentdensity in active region, the resistance to pitting corrosionof copper contained alloy-1.5Cu was lower than that of thealloy-BASE.

Figure 9(a) shows the effect of copper addition on thepotentiodynamic polarization behaviors of the experimentalalloys in a deaerated 25mass% NaCl solution at 70�Caccording to ASTM G 5. Based upon an increase of currentdensity above oxygen evolution potential (Fig. 9(a)), theresistance to pitting corrosion of the Cu added alloy-1.5Cuare similar to that of the alloy-BASE. However, thepassivation behavior of the Cu added alloy-1.5Cu appearedunstable because of the fluctuation of current density in thepassive region, compared with that of the alloy-BASE.Figure 9(b) shows the effect of copper addition on thepotentiodynamic polarization behavior of the experimentalalloys in a deaerated 30mass% NaCl solution at 70�C. Ingeneral, the pitting potential (Ep) is defined as the breakdownpotential destroying a passive film. As the Ep of an alloyincreases, the resistance to pitting corrosion of the alloyincreases. As copper content increased, the resistance topitting corrosion decreased due to a decrease in the Ep

(Fig. 9(b)). Hence, the resistance to pitting corrosion of thealloy-1.5Cu containing Cu was inferior to that of the alloy-

BASE0

10

20

30

40

50

60

70

80C

riti

cal P

itti

ng

Tem

per

atu

re (

°C)

Experimental Alloys

1.5Cu

Fig. 6 Effect of copper addition on critical crevice temperature of the

experimental alloys in 6mass% FeCl3 + 1mass% HCl per ASTM G 48-

method F.

10-6 10-5 10-4 10-3-400

-200

0

200

400

600

800

1000

1200

Po

ten

tial

(m

V v

s. S

CE

)

BASE 1.5Cu

Current Density(A /cm2 )

10-6 10-5 10-4-300

-250

-200

-150

-100

-50

0

50

100

150

200

Po

ten

tial

(m

V v

s. S

CE

)

BASE 1.5Cu

Current Density(A /cm2 )

Fig. 7 Effect of copper addition on potentiodynamic polarization behavior of the experimental alloys in deaerated 2M H2SO4 solution at

60�C according to ASTM G 5.

10-5 10-4 10-3 10-2-600

-400

-200

0

200

400

600

800

1000

1200

Po

ten

tial

(m

V v

s. S

CE

)

BASE 1.5Cu

Current Density(A /cm2)

Fig. 8 Effect of copper addition on potentiodynamic polarization behav-

iors of the experimental alloys in deaerated 0.5N HCl + 1N NaCl

solution at 40�C according to ASTM G 5.

420 S.-H. Jeon et al.

BASE because pitting potential (Ep: +0.64VSCE) of thealloy-BASE is much higher than that (Ep: +0.01VSCE) ofthe alloy-1.5Cu.

Figure 10(a) presents the potentiostatic polarization be-haviors (the current transient behaviors) for the experimentalalloys in a deaerated 25mass% NaCl solution at 70�C withan applied potential of 0VSCE in the passive region. Thepotentiostatic test was conducted to observe the initiation ofpitting corrosion and the repassivation of meta-stable pits of

the experimental alloys. Figure 10(b) shows that the numberof the current spikes corresponding to the initiation of pittingcorrosion and the repassivation of the meta-stable pits. Thenumber of the current spikes of the alloy-1.5Cu was 49whereas that of the alloy-BASE was 14. It was assumed thatthere are two distinct processes before the stable formationsof pits occur: pit nucleation and growth of the meta-stablepit.25) As presented in Fig. 10(b), the degree of the growth ofmeta-stable pits in the alloys-1.5Cu seemed to increase

10-7 10-6 10-5 10-4 10-3 10-2

-1000

-800

-600

-400

-200

0

200

400

600

800

1000

1200 BASE 1.5Cu

Po

ten

tial

(m

V v

s. S

CE

)

Current Density(A /cm2 )10-8 10-7 10-6 10-5 10-4 10-3 10-2

-1200

-1000

-800

-600

-400

-200

0

200

400

600

800

1000

1200

Po

ten

tial

(m

V v

s. S

CE

)

BASE 1.5Cu

Current Density(A /cm2 )

(a) (b)

Fig. 9 Effect of copper addition on potentiodynamic and potentiostatic polarization behaviors of the experimental alloys in deaerated

NaCl solution according to ASTMG 5: (a) deaerated 25mass%NaCl solution at 70�C and (b) deaerated 30mass%NaCl solution at 75�C.

00

10

20

30

40

50

60

70

Cu

rren

t D

ensi

ty (

µA/c

m2)

Time (s)

BASE 1.5Cu

00

10

20

30

40

50

60

70

Time (s)

Cu

rren

t D

ensi

ty (

µ A/c

m2 )

(a)

BASE0

5

10

15

20

25

30

35

40

45

50

55

Th

e n

um

ber

of

met

asta

ble

pit

s

Experimental Alloys

(b)

1.5Cu

3600300024001800120060036003000240018001200600

Fig. 10 Effect of copper addition on potentiostatic polarization behaviors at an applied potential of +0.64VSCE. of the experimental

alloys in deaerated 25mass% NaCl solution at 70�C according to ASTM G 5: (a) potentiostatic polarization behavior and (b) the number

of metastable pits.

Thermodynamics of Inclusion Formation and Its Influence on the Corrosion Behavior of Cu Bearing Duplex Stainless Steels 421

compared to those in the alloy-BASE. Therefore, it wasclarified that the resistance to the initiation of pittingcorrosion and the repassivation of meta-stable pits of thealloy-1.5Cu was inferior to that of the alloy-BASE.

The reasons that the resistance to pitting corrosion with theaddition of copper decreased are as follows: the deteriorationin the resistance to pitting corrosion as a result of the copperaddition seems to be associated with the inclusions in thealloy. In particular, the pitting potential of the Cu added alloydecreased greatly with an increase in the area fraction ofinclusions per frame area. That is, as the interface areasbetween the inclusion and the matrix in the copper addedalloy-1.5Cu increased compared to those in the alloy-BASE,the preferential sites for the initiation of pitting corrosion inthe alloy-1.5Cu increased compared to those in the alloy-BASE. Hence, it was clarified that the resistance to pittingcorrosion of the copper added alloy-1.5Cu was decreased dueto an increase of interface areas between inclusions andmatrix acting as preferential pit initiation sites.

4. Conclusions

To elucidate the thermodynamics of inclusion formationand its influence on the corrosion behavior of Cu bearingduplex stainless steels, potentiodynamic and potentiostaticpolarization tests, a SEM-EDS (scanning electron micro-scope-energy dispersive spectroscope) of inclusions andthermodynamic calculation for the formation of inclusionswere conducted.

(1) The number and the area fraction of inclusions perframe area in the alloy-1.5Cu were bigger than those of thealloy-BASE due to an increase of activity of chromium. Thechemical compositions of inclusions in the Cu added alloy-1.5Cu are similar to those in the alloy-BASE.

(2) The thermodynamic calculation for the formation ofCr-containing oxide inclusions was in good agreement withthe experimental results.

(3) Based upon a decrease of the critical passive currentdensity and primary passivation potential in active region, theresistance to general corrosion of the noble copper containedalloy-1.5Cu in a deaerated 2M H2SO4 solution was higherthan that of the alloy-BASE.

(4) Based upon an increase of the passive current densityin the passive region and the critical passive current densityin active region and a decrease of pitting potential, theresistance to pitting corrosion of copper contained alloy-1.5Cu in a deaerated 0.5 N HCl + 1N NaCl and 30mass%NaCl solutions was lower than that of the alloy-BASE due toan increase of interface areas between inclusions and matrixacting as preferential pit initiation sites.

(5) The Cu added alloy improved the resistance to generalcorrosion due to the noble Cu enriched on surface in sulfuricacid solution whereas it deteriorated the resistance to pittingcorrosion due to an increase of interface areas betweeninclusions and matrix acting as the preferential sites of pitinitiation in chloride (Cl�) solution. The addition of Cu to thealloy had a positive or negative influence in the differentenvironments.

REFERENCES

1) J. Olson and S. Nordin: Proc. Duplex Stainless Steels Symp., The

Hague, (1986) pp. 219–225.

2) E. Perteneder, J. Tosch, P. Reiterer and G. Rabensteiner: Proc. Duplex

Stainless Steels Symp., The Hague, (1986) pp. 48–56.

3) J.-O. Nilsson: Mater. Sci. Technol. 8 (1992) 685–700.

4) M. Seo, G. Hultquist, C. Leygraf and N. Sato: Corros. Sci. 26 (1986)

949–960.

5) Y. Jiangnan, W. Lichang and S. Wenhao: Corros. Sci. 33 (1992) 851–

859.

6) G. Hultquist, M. Seo, T. Leitner, C. Leygraf and N. Sato: Corros. Sci.

27 (1987) 937–946.

7) B. Postrach, I. Garz and H. H. Stehblow: Werkstoffe und Korrosion 45

(1994) 508–516.

8) L. Fedrizzi, A. Molinari, F. Deflorian and A. Tiziani: British Corros. J.

26 (1991) 46–50.

9) A. Yamamoto, T. Ashiura and E. Kamisaka: Boshoku Gijutsu 35

(1986) 448–454.

10) A. A. Hermas, K. Ogura, S. Yakagi and T. Adachi: Corrosion 51 (1995)

3–10.

11) T. Ujiro, S. Satoh, R. W. Staehle and W. H. Smyrl: Corros. Sci. 43

(2001) 2185–2200.

12) J. Stewart and D. E. Williams: Corros. Sci. 33 (1992) 457–474.

13) D. E. Williams, M. R. Kilburn, J. Cliff and G. I. N. Waterhouse: Corros.

Sci. 52 (2010) 3702–3716.

14) S. J. Zheng, Y. J. Wang, B. Zhang, Y. L. Zhu, C. Liu, P. Hu and X. L.

Ma: Acta 58 (2010) 5070–5085.

15) D. E. Williams, C. Westcott and M. Fleischmann: J. Electrochem. Soc.

132 (1985) 1804–1811.

16) H. Bohni and L. Stockert: Werkst. Korros. 40 (1989) 63–70.

17) W. M. Carroll and M. B. Howley: Corros. Sci. 30 (1990) 643–655.

18) P. C. Pistorius and G. T. Burstein: Phil. Trans. R. Soc. Lond. A 341

(1992) 531–559.

19) U. Reichau, H. D. Pletka and K. G. Schutze: Werkst. Korros. 43 (1992)

520–526.

20) G. T. Burstein and R. M. Souto: Electrochim. Acta 40 (1995) 1881–

1888.

21) G. S. Frankel, L. Stockert, F. Hunkeler and H. Boehni: Corrosion 43

(1987) 429–436.

22) G. T. Burstein and S. P. Mattin: Phil. Mag. Lett. 66 (1992) 127–131.

23) D. E. Williams, J. Stewart and P. H. Balkwill: Corros. Sci. 36 (1994)

1213–1235.

24) M. Keddam, M. Krarti and C. Pallotta: Corrosion 43 (1987) 454–464.

25) P. C. Pistorius and G. T. Burstein: Corros. Sci. 36 (1994) 525–538.

26) S. T. Pride, J. R. Scully and J. L. Hudson: J. Electrochem. Soc. 141

(1994) 3028–3040.

27) P. C. Pistorius and G. T. Burstein: Corros. Sci. 33 (1992) 1885–1897.

28) www.factsage.com

29) J. H. Park, J. G. Park, D. J. Min, Y. E. Lee and Y. B. Kang: J. Eur.

Ceram. Soc. 30 (2010) 3181–3186.

30) J. H. Park and H. Todoroki: ISIJ Int. 50 (2010) 1333–1346.

31) J. H. Park, I. H. Jung and S. B. Lee: Met. Mater. Int. 15 (2009) 677–681.

32) J. H. Park, S. B. Lee, D. S. Kim and J. J. Pak: ISIJ Int. 49 (2009) 337–

342.

33) J. H. Park, S. B. Lee and H. R. Gaye: Metal. Mater. Trans. B 39B

(2008) 853–861.

34) C. W. Bale, E. Belisle, P. Chartrand, S. A. Decterov, G. Eriksson, K.

Hack, I.-H. Jung, Y.-B. Kang, J. Melancon, A. D. Pelton, C. Robelin

and S. Petersen: Calphad 33 (2009) 295–311.

35) ANNUAL BOOK of ASTM STANDARDS, ASTMG 5 (1995) pp. 48–58.

36) G. T. Burstein, P. C. Pstorius and S. P. Mattin: Corros. Sci. 35 (1993)

57–62.

37) ANNUAL BOOK of ASTM STANDARDS, ASTM G 48-method A

(1995) pp. 174–179.

38) H. G. Lee: Chemical Thermodynamics for Metals and Materials,

(Imperial College Press, London, 1999) pp. 97–130.

39) Steelmaking Data SourceBook, revised edition, Japan Society for the

Promotion of Science (JSPS), 19th Committee on Steelmaking,

(Gordon and Breach Science Publishers, New York, NY, 1988)

pp. 273–325.

422 S.-H. Jeon et al.