Indian Journal of Engineering & Materials Sciences Vol. 8, June 2001, pp. IS3-160

Corrosion behaviour of sintered ferrous alloys and Ferro-TiC in H2S04 and N aCI solutions

M Debata & G S Upadhyaya

Department of Materials and Metallurgical Engineering, Indian Institute of Technology, Kanpur 208 016, India

Received 5 September 2000; accepted 4 April 2001

The room temperature corrosion behaviour of plain carbon steel Fe-O.SC, low alloy steel Fe-SCu-O.SC, tool s'teels ASP 23 and ~SP 30, a.nd ferro-TiC cermet was investigated in H2S04 (O. I-S N) and 0.6 N NaCI solutions by potentiodynamic polarIZation techntque. The study also includes X-ray diffraction, optical microscopy and scanning electron microscopy of the as-received as well as corroded surfaces. Porosities present in Fe-O.SC and Fe-SCu-O.SC steels were found to be responsible for their poor corrosion resistance. ASP 30 steels were found to exhibit good corrosion resistance as compared to ASP 23 in both the media.

The addition of both copper and carbon to sintered iron offers greater strength and hardness than the addition of either alone. A detailed sintering study of Fe-8%Cu, C (>0.8%) and Fe-Cu-l.O%C (Cu>6%) premixes has been reported by Majima and Mitani I. Fe-Cu or Fe-Cu-C steels exhibit transient liquid phase sintering.

The most widely used production practices for PIM tool steels use gas atomization and hot isostatic pressing (HIP). Such steels exhibit super solidus liquid phase sintering. Cobalt in such steels does not act as carbide former, but enters into the matrix. Cobalt increases the solidus temperature, thus permitting the use of high austenitizing temperatures to achieve greater solutioning of alloying elements. Furthermore, cobalt enhances hot hardness and temper resistance.

Steel-bonded titanium carbides are the type of cermets , which can be machined and are corrosion, oxidation and wear resistant. One of the attractive properties of steel bonded TiC is that it is heat treatable. Among the steel-bonded titanium carbides, ferro-TiC grades are very common, for which various grades have been developed2

.

The present investigations deal with the corrosion behaviour of plain carbon steel Fe-O.SC, low alloy steel Fe-SCu-O.SC, tool steels ASP 23 and ASP 30, and ferro-TiC in H2S04 media of different concentrations and 0.6 N NaCI.

Experimental Procedure The chemical compositions of the alloys and other

relevant data are given in Table 1. The steel powder

premix compacts were sintered in hydrogen atmosphere at 1 120°C for one hour. Full details of the process are already reported3

. Fully dense ASP 23 and ASP 30 PIM high-speed steels were supplied by MIS Associated Swedish Steels AB, Stockholm, Sweden. The anti segregation process (ASP) for production of such steels is also known4

• The ferro-TiC cermets are generally liquid phase sintered in vacuum in the temperature range of 13S0-1S00°C depending on the steel binder compositions. The material in the form of short bars was supplied by Alloy Technology International, Inc., West Nyack, NY.

The cylindrical specimens of diameter ranging from 10 to 16 mm were first ground fl at on a gri nding belt and then polished using a series of emery papers up to 1000 grit size. The final wet polishing was subsequently carried out on a polishing wheel using 0.3 ~m size alumina powder suspension in distilled water. Optical metallography was carried out on these alloys. In case of ferro-TiC, there was no necessity of etching, as the polished surface revealed the microstructure. The etchant used for Fe-O.SC and Fe-SCu-O.SC was FeCh solution and for ASP 23 and ASP 30, it was Nita!' After metallography, the samples were mounted in epoxy with electrical connections through the rear of the mounts.

Potentiodynamic scans for the alloys were taken in different concentrations of H2S0 4 (0.1-5 N) and 0.6 N NaCl solutions. The polarization measurements were carried out using EG & G Parc Model Versastat-II combined with M3S2 programmer, the scanning rate was 1 m V Is. For testing, the samples were subjected as the working electrodes and connected in an electrical

IS4 INDIAN 1. ENG. MATER. SCI., JUNE 200 1

circuit which allowed impression of an external voltage and measurement of corrosion current against saturated calomel electrode (SCE). Graphite was used as counter electrode. The corrosion testing was carried out at room temperature.

The X-ray diffraction study of as-received and corroded surfaces was carried out with the help of IsoDebyefIex-2002 X-ray diffractometer using CUka radiation. The scanning micrographs of the corroded surfaces were taken using JEOL, 840 A scanning electron microscope.

Results Anodic polarization studies - Results of

potentiodynamic polarization studies, carried out on all the aIIoys in H2S04 media, are presented in Figs I-S and Table 2. There is an increase in Ecorr value towards posi tive side with concentration of H2S04 in Fe-O.SC, Fe-SCu-O.SC, and ASP 30 steels. With increase in concentration of H2S04, increase in lcorr

value is obtained for ASP 23 and ASP 30. AIl other aIIoys did not show any particular trend .

Table I -Chemical composition of sintered alloys with density and porosity

Alloy Density (gcm-3)

Nominal comJ2osition

Fe-0.5C 6.56 C Cu Fe

0.5 0 Bal (wt. %)

Fe-5Cu-0.5C 6.50 C Cu Fe

0.5 5.0 Bal (wt. %)

ASP23 8.54 C Si Mn Cr Mo W Co V Fe

1.27 0.3 0.3 4.2 5.0 6.4 3. 1 bal (wt. %)

ASP 30 8.63 C Si Mn Cr Mo W Co V Fe

1.27 0.3 0.3 4.2 5.0 6.4 8.5 3.1 bal (wt. %)

ferro-TiC 6.34 TiC Tool steel binder

C Cr Mo Fe

45 (vol. %) 0.8 10.0 3.0 Bal (wt. %)

Table 2 - Anodic polarization characteristics of alloys in H2S04 medi um

Alloy H2SO4 Ecorr Icorr (J..lAlcm2) Corrosion rate

Concentration (N) (mV) mpy mmpy

Fe-0.5C 0.1 -511.6 3629 1993 50.63 1.0 - 477.5 9907 544 1 92.21 2.0 - 467.6 2935 1581 40.17 3.0 -440.2 7325 4023 102.2 5.0 -437.2 3985 2 189 55 .59

Fe-5Cu-0.5C 0.1 -492.9 6658 3621 9 1.96 1 -456.6 15670 8522 216.5 2 - 448.4 5220 2839 72.10 3 - 439.6 6852 3726 94.64 5 -425 .1 4602 2503 63.57

ASP23 0. 1 - 515.3 39.89 15.14 0.382 I -469.4 1007 382.0 9.704 2 -462.7 1502 570.0 14.48 3 - 437.9 1677 634.4 16.17 5 -44 1.9 2076 788.1 20.02

ASP 30 0.1 -456.7 70.01 23.86 0.606 I -426.7 23 1.6 78.94 2.005 2 -420.9 311.5 106.2 2.697 3 - 414.7 382.6 123.2 3.129 5 - 409.9 410.7 140.0 3.556

Ferro-TiC 0. 1 -514.0 103.8 43.92 1.1 16 I - 443.9 15.65 8.735 0.221 2 -434.8 19420 8390 213.3 3 -439.9 8379 3622 92.0 5 - 429.1 4381 1889 47.97

DEBATA & UPADHYA YA: CORROSION BEHAVIOUR OF SINTERED FERROUS ALLOYS AND FERRO-TiC ISS

900

700

500

;;-300 -5.

III U '/}

100

~ - 100 III

~ - 300

J: -5 00 !--~. -=-::--=.- .:..:- -:..c" -:;;;;-~='"-=-"=

- 700

I "'~':::::::: . .......... ,

I

O.1N \ ) p

CurreOi Density Alem'

Fig. I - Typical anodic polarization curves of Fe-O.SC steel in H2S04 of various concentrations

;;­E III U Vl

200

~ -zoo Ul

" jj · 400 o "-

·600

) O'N 'N )

5N

\ " - BOO 6~L..L.J...LlJJ._';-5 -L..LU.JJ_il!4--'---W.LIJJ.J:l...-L...LLil.L_ul

z--'--'--'-!..U.J.ILL..L.J...LLJ.J.J

-I

~ Curren! Density Alem'

Fig . 2 - Typical anodic polarization curves of Fe-SCu-O.SC steel in H2S04 of various concentrations

1400r -rn-m'"I·''''"TTIT1]''I,..,..,mnnr-r-rTTTTmJ''I---rTTTm"ry--,r-rrmnr-rrrn""" ".

Current Density Aicm2 IOn

Fig. 3 - Typical anodic polarization curves of ASP 23 tool steel in H2S04 of various concentrations

Fig. 1 shows typical anodic polarization curves for Fe-O.SC steel, where the plots for 0.1 Nand S N show active and passive regions. Fig. 2 shows the anodic polarization curves for Fe-SCu-O.SC steel, where no clear passive regions are obtained. For ASP 30, there is a distinct passive region for 0.1 N (Fig. 4) and for ASP 23 a passive region (Fig . 3) in 1 N solution. The anodic polarization curves for ferro-TiC (Fig. S) show

;;­E

1000

;;:;- 600 U Vl

~ 200 III .. . ~ -200 <; p..

· 600

-I OO~ 6L..-.L.1...LlJJ.~~l..J.liuL.J-)...LU~-'-.LJ..J..u._~2---'--'--'-J.lill_I!-L..L.J~

Current Dens ity Alem'

Fig. 4 - Typical anodic polarization curves of ASP 30 tool steel in H2S04 of various concentrations

1400

,~ 1000 > E t':l 600 Vl

vi >

Ul 200

~

~ -2 00 o p..

-6 00

-I 00~9~ .Lll~...L.I..JWJ_11lL7 ....u..w_lI1Jl6-LJLWll_ 51L.L.l.llU_uL

4 -LLUJ_lI1Jl

3--L..1.llll.!_ZL.i...lillJl_!L

, ..J....UllWi

Current Densi ty A/em!

Fig. S - Typical anodic polarization curves of ferro-TiC in H2S04

of various concentrations

no clear passive region; the curves corresponding to 0.1 Nand 1 N solutions are similar. For the sake of convenience, corrosion rates of all the alloys with respect to H2S04 concentration are plotted in Fig. 6 and Table 2 also provides data on anodic polarization characteristics of alloys in H2S04 medium. A comparison of both Fe-O.SC and Fe-SCu-O.SC, and also ASP 23 and ASP 30 shows similar behaviour. The trend is similar to the variation of l eorr value with the H2S04 concentration.

Results of potentiodynamic anodic polarization studies, carried out on these alloys in 0.6 N NaCl solution are presented in Figs 7-8 and Table 3. The curves show active and passive regions. The trend in corrosion rate variation is:

Fe-SCu-O.SC > Fe-O.SC > ferro-TiC> ASP 23 > ASP 30

X-ray diffraction studies - Due to the formation of relatively very thin oxide film of the surface of the alloys, no characteristic peaks for corresponding oxides were obtained. X-ray study was able to detect only the base elements. The lattice parameters for the

156 INDIAN J. ENG. MATER. SCI., JUNE 200l

9000.------------ - --, o Ferro TiC o Fe - O.se

7500 '" Fe- 5Cu -0.5C

<lI~

0 (; a:

1000 c o ASP 23 0 o ASP 30 'iii 0 ... 800 ... 0 u

600

2 3 4 5 H2S04 Concentration, N

Fig. 6 - Corrosion rate variation of different alloys with respect to H2S04 concentration

500

:> 300

E :;;- 100 u V>

~ -100 Ul

] -300

c ~ -sao 0..

- 700

Current Densiry A/cm2 n

10

Fig. 7 - Anodic polari zation curves of Fe-0.5C and Fe-5Cu-0.5C steel s in 0.6 N NaCI

1400

:> ~ 1000

tJ en 600 ~

Ul 200

'" § <5 - 200

0..

- 600

I ! j

()~ / ) " / . )"1

Ferro - TiC cermr:;.-_/;;;-;U J/~~

-", __ /~30 ~ Tool s teel

...... ---- .. :.:;:.:;;:.:.:::.~.

Current Density A/em:!

Fig. 8 - Anodic polarization curves of ASP 23 and ASP 30 tool steels along with ferro-TiC in 0.6 N NaCi

TiC matrix for ferro-TiC and base element iron for all the alloys in as-received and cQrroded conditions in H2S04 & NaCI are given in Table 4.

Optical and SEM studies - The optical micrographs of the as-polished surfaces for ferro-TiC and as-etched surfaces for all other alloys are shown in Fig. 9. Both ASP 23 and ASP 30 show fine grain structures. The SEM micrographs of the corroded surfaces of all the alloys are shown in Fig. 10. The surfaces of Fe-5Cu-0.5C and ASP 23 are attacked to a greater extent in comparison with Fe-O.SC and ASP 30.

The SEM micrographs of the corroded surfaces of these alloys in 0.6 N NaCI at different magnifications are shown in Fig. 11. The micrographs for Fe-5Cu-0.5C show greater corrosion attack as compared to Fe-0.5C. The surfaces of ASP 23 and ASP 30 steels are not much affected.

Discussion

Corrosion in H2S04 media

The anodic reaction for all the alloys 111 H2S04

solution is:

.. . (1)

Table 3 - Anodic polari zation characteristics of alloys in 0.6 N NaCI solution

Alloy Ecoer Icon (~A/cm2) Corrosion rate (mY) mpy mmpy

Fe-0.5C -494.0 43.05 23.65 0.600

Fe-5Cu-0.5C -541.5 86.28 46.92 1.192

ASP23 -574.3 7.595 2.883 0.073

ASP 30 -552.1 5.890 2.008 0.051

Ferro-TiC -512.9 24.76 10.70 0.271

.('

DEBATA & UPADHYA YA: CORROSION BEHAVIOUR OF SINTERED FERROUS ALLOYS AND FERRO-TiC 157

,,-

r

~ ."- .,\ i ' . :: . ~

. ' '. I

"-

Fig. 9 - Optical micrographs of the as-received alloys: (a) Fe-O.SC; (b) Fe-SCu-O.SC; (c) ASP 23; (d) ASP 30; and (e) ferro-TiC

Table 4 - Lattice parameter (nm) variation of the matrix element of as-received and corroded alloys in different media

Alloy

Fe-O.SC Fe-SCu-O.SC

ASP 23 ASP 30 ferro-TiC

As-received condition

0.2870 0.2876

0.2870 0.2883

0.2887 (Ferrous matrix) 0.432S (TiC)

Corroded condition S N H2S04 0.6 N NaCI

0.2866 0.2867 0.2873 0.2872

0.2866 0.2796

0.4307 (FeO) 0.4310 (TiO)

0.2864 0.2899

0.4321 (FeO) 0.4306 (TiO)

158 INDIAN J. ENG. MATER. SCI., JUNE 200 1

Fig. 10- SEM micrographs of the corroded surfaces of various alloys tested in 5 N H2S04 (a) Fe-0.5C; (b) Fe-5Cu-0.5C; (c)-(d) ASP 23 at different magnificati ons; and (e)-(f) ASP 30 at different magn ifications

In most of the cases, iron goes into solution as divalent iron which is oxidized to the trivalent state by atmospheric oxygen. Hydrogen evolution is the cathodic reaction .

. .. (2)

As the hydrogen ions are active, sulphate ions do not participate in the electrochemical reaction. A linear behaviour in corrosion rate vs H2S04

concentration of iron is reported5 within the concentration range of 25% (0-9.5 N). Similar results were obtained for ASP 30 and ASP 23 steels. The curves for the rest of the alloys are found to show ilTegular trends. The effect on cOlTosion rate with

respect to H2S0 4 concentration does not follow a _uniform pattern - first because of ionization effects in the aqueous solution and second, because of changes that occur in the characteristics of any film of the corrosion products that may be present on the alloy surfaces.

The beneficial effect of eu in reducing the corrosion rates of steels in H2S0 4 is also reported6

.7

.

However, the present investigation shows that Fe-Seu-o.se steel is less cOlTosion resistant as compared to Fe-O.Se. This may be due to the more heterogeneous microstructure for the former (Fig. 9). In addition, the lattice distortion with di ssolution of eu in the felTous matrix also is responsible for more

DEBATA & UPADHYAYA: CORROSION BEHAVIOUR OF SINTERED FERROUS ALLOYS AND FERRO-TiC 159

Fig. Il - SEM micrographs of the corroded surfaces of various alloys tested in 0.6 N NaCl: (a)-(b) Fe-O.SC at different magnifications; (c)-(d) Fe-SCu-O.SC at different magnifications; (e)-( f) ASP 23 at different magnifications; (g)-(h) ASP 30 at different magnifications; and (i) ferro-TiC

160 INDIAN J. ENG. MATER. SCI., JUNE 2001

corrosion. As far as porosity in these steels is concerned, the values are more or less equal for both the steels (Table 1). This confirms that the present change in corrosion behaviour is not dictated by the variation in porosities. 8.5 wt. % of Co in ASP 30 steel appears to be responsible for more corrosion resistance compared to ASP 23 .

In case of ferro-TiC cermet, there is much ilTegularity in the corrosion rate variation. The abnormal value at 2 N is not clear. However, the higher corrosion rates obtained may be attributed to a greater attack of the solution on the binder phase, i.e. steel.

Corrosion in O.6N NaCI solution

Relatively very high values of corrosion rates for Fe-O.SC and Fe-SCu- O.SC steels are attributed to the porosity level present in them. In the presence of aggressive cf ions, there is a breakdown of the passive film that is formed on the surface. This is evident from Figs 7-8 . ASP 23 and ASP 30 steels owe their corrosion resistance to the maintenance of the passive film . The higher corrosion rate of ferro-TiC is obviously due to the greater attack of the binder phase by NaCI solution .

Conclusions Porosity plays a vital role in determining the

corrosion behaviour of sintered alloys. Both H2S04

and NaCI media more rapidly attack Fe-SCu-O.SC sintered steel than Fe-O.SC steel. This is due to the more heterogeneous microstructure in the former. ASP 30 tool steel is more corrosion resistant as compared to ASP 23 tool steel in both H2S04 and NaCI media. The sudden increase in corrosion rate of ferro-TiC after certain H2S04 concentration may be attributed to the binder phase. In the case of 0.6 N NaCl solution, ferro-TiC is in between low alloy steel and tool steel in its corrosion behaviour.

References 1 Majima K & Mitani H, Transactions JIM, IS (1977) 663.

2 Upadhyaya G S, Silltered Metallic and Ceramic materials­Preparations, properties and applications (John Wiley and Sons Ltd. , Chichester, U.K.), 1999.

3 Ranjan. M. Tech. Thesis, Indian Institu te of Technology. Kanpur, May 2000.

4 Hellman P, Larker H, Pfeffer J B & Stromblad I, in Modem developments in powder metallurgy, Vol. 4 (Plenum Press, New York), 1970,573.

5 Ward R, in Metals halldbook, Vol. 10, Sth Edition. (ASM International, Metals Park, Ohio), 1975, 16S.

6 Endo H & Morioka S, Dissolution Phenomenon of Copper containing Steels in AqueoLis Sulfuric Acid Solutions of VarioLis Concentrations, Paper presented at the Third Symposium, Japanese Metal Association. April 1935.

7 Will iams E & Komp M E, Corrosion. 21 (1)(1965) 9.