Microsoft Word - 235 completed

IIUM Engineering Journal, Vol. 12, No. 5, 2011: Special Issue -1 on Science and Ethics in Engineering

Kaderi et al.

OBSERVATION ON VOID FORMED IN OXIDE SCALE OF

Fe-Cr-Ni ALLOY AT 1073K IN DRY AND HUMID

ENVIRONMENTS

AKBAR KADERI
1
, AHMAD ZAKI MOHD ZAINAL

1
, HANAFI ANI

AND RAIHAN OTHMAN
2

1
Department of Manufacturing and Materials Engineering,

2
Department of Science in Engineering,

Kulliyyah of Engineering, International Islamic University Malaysia,

Jalan Gombak, 53100, Kuala Lumpur, Malaysia.

mhanafi@iium.edu.my

ABSTRACT: Void formation in oxide scale during high temperature oxidation is a

common phenomenon. Over a long period of time voids will affect the mechanical

property of scales by influencing the cracking and spalling. Voids formed in dry

environment are different than that of formed in humid environment. With the presence

of water vapor in humid environment the formation of void will increase, thus greater

number of void compared to that in dry environment. Fe-Cr-Ni alloy samples were

exposed isothermally at 1073 K in air (��� = 0.21atm = 2.1 � 10
Pa) and humid (air +

steam) environments. XRD analysis done to all samples confirms that Fe2O3, Fe3O4,

NiCr2O4, FeCr2O4, Cr2O3 and NiO phases exist in the scale. EDX analysis done shows

varying compositions of Fe,Cr,Ni and O in outer and inner oxide scale, oxide scale/metal

interface and metal. Field emission scanning electron microscope (FE-SEM) was used to

investigate voids formed in the cross sections of the oxidized samples. Volume fraction

of voids in the oxide scale was calculated in accordance to the cross sectional area

fraction of voids in the scale. It shows that Fe-Cr-Ni alloy samples exposed in humid

environment has as high as 71% more voids than that exposed in dry environment. It is

concluded that the humid environment increased the number of void formed in the oxide

scale, thus facilitates the exfoliation of protective scale during the high temperature

oxidation.

ABSTRAK: Pembentukan gelembung udara di dalam lapisan oksida ketika proses

pengoksidaan di suhu tinggi merupakan satu fenomena biasa. Pada satu jangka masa

yang panjang gelembung-gelembung ini akan memberi kesan kepada sifat mekanikal

oksida dengan mempengaruhi pembentukan keretakan dan pengelupasan oksida.

Gelembung udara yang terbentuk di dalam persekitaran kering berbeza daripada yang

terbentuk di dalam persekitaran lembap. Dengan adanya wap air, pembentukan

gelembung akan bertambah berbanding yang terbentuk di dalam persekitaran kering.

Sampel aloi Fe-Cr-Ni telah dioksidakan secara isoterma pada suhu 1073 K di dalam

udara (��� = 0.21atm = 2.1 � 10
Pa) dan lembap (udara + wap air). Analisis

Pembelauan Sinar – X (XRD) kepada semua sampel menunjukkan oksida yang terbentuk

ialah Fe2O3, Fe3O4, NiCr2O4, FeCr2O4, Cr2O3 dan NiO. Analisis Penyebaran Tenaga

Sinar – X (EDX) menunjukkan komposisi Fe, Cr, Ni dan O yang berubah - ubah di

lapisan oksida luar dan dalam, oksida/ antara muka logam dan logam. Mikroskop

Imbasan Elektron-Pancaran Medan (FE-SEM) digunakan untuk meneliti gelembung di

dalam oksida pada keratan rentas sampel. Pecahan isi padu gelembung yang terbentuk

pada oksida dikira dengan merujuk kepada pecahan luas keratan rentas pada oksida

tersebut. Sampel aloi Fe-Cr-Ni yang dioksidakan di dalam persekitaran lembap

IIUM Engineering Journal, Vol. 12, No. 5, 2011: Special Issue -1 on Science and Ethics in Engineering

Kaderi et al.

mempunyai kandungan gelembung udara 71% lebih banyak berbanding dengan yang

dioksidakan di dalam persekitaran kering. Kesimpulannya persekitaran lembap

meningkatkan bilangan gelembung yang terbentuk di dalam lapisan oksida, sekaligus

memudahkan pengelupasan oksida semasa pengoksidaan suhu tinggi.

KEYWORDS: high temperature oxidatio;, Fe-Cr-Ni alloy; void formation; quantitative

analysis of voi;, dry environment; humid environment

1. INTRODUCTION

Formation of voids in oxide scale especially in the proximity of metal/oxide interface

or outer oxide scale/inner oxide scale interface is common in metals during high

temperature oxidation. The formation will affect the oxidation mechanism and mechanical

properties of the scale and also the metal itself. The tendency of voids to concentrate at

outer oxide scale/inner oxide scale interface will result in spalling of the outer oxide scale

[1].

Voids within iron oxide scales and the iron-rich oxides formed on dilute Fe-Cr and

Fe-Al alloys developed far more faster in the presence of water vapor. Rahmel et al. [2, 3]

have reported that presence of H2O(g) caused the scale to develop porosity. Ehlers et. al.

[4] observed that 9 mass% Cr ferritic steels form rapidly growing scales of porous

magnetite plus spinel during oxidation at 923 K in gases containing water vapor.

Contemporary literatures [2-6] on effect of water vapor on Fe-Cr or Fe-Al alloys

mostly describe formation of voids in qualitative manner but few are attempting to study it

quantitatively. Understanding on the formation of voids quantitatively is very important

because voids formation affect development of scales microsructure. Maruyama et al. [7]

proposed that void formation in a growing oxide scale during high temperature oxidation

of a metal closely related with the divergence of ionic fluxes. Further improvement of the

treatment by Ueda et. al. [8], by including all ionic fluxes into the calculation of the

chemical potential distribution and quantitatively explained the position and the volume of

voids formed in the magnetite scale. Until recently, Maruyama et al. [9] has successfully

applied the quantitative estimation on the void formed in a single phase magnetite scale

grown on iron substrate at 823 K in oxdizing environment containing oxygen partial

pressure, ��� = 4.2 � 10
�
� Pa. However, most qualitative and quantitative studies on

void formation cited earlier done in a strictly controlled oxidizing environment especially

by lowering the ��� , so that only specific oxide scale grows on the metal substrate at

specific temperature.

Therefore in this study, quantitative observation was focusing on void formed in a

multiple oxide scale grown on Fe-Cr-Ni alloy at 1073 K in normal air (��� = 0.21atm =

2.1 � 10 Pa) and humid (air + steam) environment.

2. EXPERIMENTAL METHODS

2.1 Sample Preparation of Fe-Cr-Ni Alloys

Fe-Cr-Ni alloys were prepared from high-purity Fe (99.99%), Cr (99.99%), Ni

(99.99%) plates by arc melting in Ar gas. The alloys were further annealed for 86.4 ks at

1373 K in vacuum. After annealing the alloys were sliced into coupons of 1mm thick and

the surfaces of the coupons were ground with emery paper up to #2000. Then the samples

were polished with 3µ m alumina paste to a mirror-like finish. The samples were cleaned

IIUM Engineering Journal, Vol. 12, No. 5, 2011: Special Issue -1 on Science and Ethics in Engineering

Kaderi et al.

with acetone via ultrasonic agitation. The chemical composition of the alloys were

analyzed through Energy Dispersive X-Ray (EDX). The chemical composition of Fe-Cr-

Ni alloys are shown in Table 1.

Table 1: Chemical composition of Fe-10 mass % Cr-10 mass % Ni alloy.

Element Fe Cr Ni

Mass % 79.61 10.78 9.61

2.2 Oxidation Experiment of Fe-Cr-Ni Alloys

Figure 1 shows the experimental setup. The sample was placed at the isothermal zone

of the furnace. An R-type thermocouple was placed at the isothermal zone beside the

sample to monitor the temperature. The distance between the sample and the thermocouple

is 5 mm. The isothermal zone is 30 mm in range.

Fig. 1: Experimental setup for oxidation in humid condition at 1073 K.

For the oxidation in dry environment, the sample was heated to 1073 K and held for

86.4 ks and 172.8 ks respectively in air. For the oxidation in humid environment, a boiler

is placed directly below the furnace. The water vapor was channeled into the furnace once

the heating process started. The continous supply of water to the boiler will ensure that the

water vapor will be produced continously throughout the oxidation process. The oxidation

process in humid environment was also hold for 86.4 ks and 172.8 ks respectivey. After

oxidation, the samples are cooled to room temperature in the furnace.

2.3 Phase Identification

Oxide phases formed were analyzed by X-ray diffraction (XRD) using the divergence

slit of 1 degree. The diffraction angle is from 20° to 80° and the scanning speed was

3.0000 °/min.

The XRD patterns were compared with α-Fe (JCPDS 00-006-0696), FeO (JCPDS 00-

006-0615), Fe2O3 (JCPDS 01-085-0599), Fe3O4 (JCPDS 00-019-0629), NiFe2O4 (JCPDS

IIUM Engineering Journal, Vol. 12, No. 5, 2011: Special Issue

01-074-1913), NiCr2O4 (00-023

1479), and NiO (JCPDS 00-047

2.4 Chemical Composition of Each o

Scale, Inner Oxide Scale, Inner Oxide Scale/

Alloy

The concentration of main elements of each samples after being oxidized were

quantitatively determined by EDX. The analysis was done on 4 spots; Spot 1: Outer oxide

scale, Spot 2: Inner oxide scale, Spot 3: Metal/Oxide interface and Spot 4: Metal. Figure 2

shows an example of an area of a sample’s 4 spots being analyzed to determine the main

elements concentration.

Fig. 2: An FE-SEM micrograph shows the spots for EDX analysis.

2.5 Volume Fraction of Voids

Oxidized samples were mounted individually in resin. Prior to mounting, 2 mm of

each of sample was cross –

samples after being cross-sectioned. The exposed cross

ground up to #2000. Then the samples were polished with 0.05 µ m alumina to a mirror

like finish.

Fig. 3: Schematics of samples after being cross

Surface morphology was observed by field emission scanning electron microscopy (FE

SEM). Grids of uniformed size were traced on each images and area fraction of voids in

the scale was measured. It was ass

to the cross-sectional area fraction of voids in the scale

applied by Akiba et al. [11]

substrate at 1373 K.

g Journal, Vol. 12, No. 5, 2011: Special Issue -1 on Science and Ethics

023-0432), FeCr2O4 (JCPDS 00-034-0140), Cr

047-1049).

omposition of Each of Main Elements after Oxidation:

Inner Oxide Scale/Fe-Cr-Ni Alloy Interface a

The concentration of main elements of each samples after being oxidized were

by EDX. The analysis was done on 4 spots; Spot 1: Outer oxide

scale, Spot 2: Inner oxide scale, Spot 3: Metal/Oxide interface and Spot 4: Metal. Figure 2

shows an example of an area of a sample’s 4 spots being analyzed to determine the main

SEM micrograph shows the spots for EDX analysis.

f Voids in the Oxide Scale

Oxidized samples were mounted individually in resin. Prior to mounting, 2 mm of

sectioned by diamond saw. Figure 3 is a schematic of the

sectioned. The exposed cross – section part of the samples was

ground up to #2000. Then the samples were polished with 0.05 µ m alumina to a mirror

Schematics of samples after being cross-sectioned.

Surface morphology was observed by field emission scanning electron microscopy (FE

SEM). Grids of uniformed size were traced on each images and area fraction of voids in

the scale was measured. It was assumed that the volume fraction of void in the scale equal

sectional area fraction of voids in the scale [10]. The same assumption was

[11], in order to calculate voids formed in NiO grown of Ni

1 on Science and Ethics in Engineering

Kaderi et al.

0140), Cr2O3 (00-038-

Outer Oxide

and Fe-Cr-Ni

The concentration of main elements of each samples after being oxidized were

by EDX. The analysis was done on 4 spots; Spot 1: Outer oxide

scale, Spot 2: Inner oxide scale, Spot 3: Metal/Oxide interface and Spot 4: Metal. Figure 2

shows an example of an area of a sample’s 4 spots being analyzed to determine the main

SEM micrograph shows the spots for EDX analysis.

Oxidized samples were mounted individually in resin. Prior to mounting, 2 mm of

diamond saw. Figure 3 is a schematic of the

section part of the samples was

ground up to #2000. Then the samples were polished with 0.05 µ m alumina to a mirror-

Surface morphology was observed by field emission scanning electron microscopy (FE-

SEM). Grids of uniformed size were traced on each images and area fraction of voids in

umed that the volume fraction of void in the scale equal

. The same assumption was

, in order to calculate voids formed in NiO grown of Ni

IIUM Engineering Journal, Vol. 12, No. 5, 2011: Special Issue -1 on Science and Ethics in Engineering

Kaderi et al.

3. RESULTS AND DISCUSSION

3.1 Phase Identification by XRD and EDX Analysis

Figure 4 shows the XRD patterns of the sample surfaces.It was shown there is

formation of Fe2O3 in all samples. Less intense and small peaks of Cr2O3 and NiO phase

only detected for samples oxidized in dry environment. The formation of FeCr2O4 and

NiCr2O4 spinels were observed in all samples. Fe3O4 were also detected only in samples

oxidized both in humid and dry environment.

Fig. 4: XRD patterns of samples in dry and humid environment at 1073 K after

86.4 ks and 172.8 ks.

The EDX Profile was examined as shown in Fig. 5a, b, c, and d. The metal/oxide

scale interface for all samples were set at x = 0 µ m for the sake of easy observation. The

elemental composition of Fe in the outer scale is within a range of 67% - 72%. The

composition of O in the outer oxide scale is also substantial for all samples which are

showing a range of 30 – 35%. Correspond with the XRD pattern, the predomination of Fe

and O concentration in the outer scale can be explained by the presence of Fe2O3 and

Fe3O4. The inner scale shows the element of Fe as well as O has decreased. There are

traces of Ni and Cr concentration in this inner scale compared to the outer scale. The inner

scale is corresponding to the XRD pattern of NiCr2O4 and FeCr2O4. In the internal oxide

zone (IOZ), it was observed that the concentration of Fe increased while the concentration

of O is decreasing. The concentration of Cr and Ni also shows increment compared to that

of in the inner scale zone. These suggested that spinel oxides of NiCr2O4 and FeCr2O4

formed in IOZ region. Finally, in the Fe-Cr-Ni alloy region the concentrations of Fe, Cr,

and Ni are at their highest but the concentration of O is zero. The decreasing oxygen

composition closer to the alloy and decreasing metal composition further away from the

alloy was caused by the diffusion of metal ion and oxide ion under the influence of

chemical potential gradients.

IIUM Engineering Journal, Vol. 12, No. 5, 2011: Special Issue -1 on Science and Ethics in Engineering

Kaderi et al.

Fig. 5: EDX profile of samples in dry and humid environment at 1073 K after

86.4 ks (a, b) and 172.8 ks (c, d).

3.2 Volume Fraction of Voids in the Oxide Scale

Figure 6 shows the FE-SEM micrographs of cross section of samples oxidized in dry

and humid environment. Voids are observed to be different from the inner oxide scale than

that of outer oxide scale for all samples.

Larger size of voids are observed in the vicinity of outer oxide scale /inner oxide

scale. Smaller size of voids are observed to be congregated in the metal reservoir of inner

oxide scale. For samples oxidized in dry environment, voids are spherical in shape. While

samples oxidized in wet environment, voids are columnar in shape.

IIUM Engineering Journal, Vol. 12, No. 5, 2011: Special Issue -1 on Science and Ethics in Engineering

Kaderi et al.

Fig. 6: FE-SEM micrographs of cross section of samples oxidized in dry (a, b) and humid

(c, d) environment at 1073 K after 86.4 ks (a, c) and 172.8 ks (b, d).

Figure 7 shows the volume fraction of voids in the inner oxide scale. The volume fraction

of voids is in good agreement with the governing equation (1) proposed by Maruyama [9]:

The volume fraction of voids in scales, �� :

�� �
�����

��

������
�
(1)

�� is volume fraction of voids in scale, �����
�

is void formation in the scale during

oxidation, ! is inner scale and " is outer scale. Multiplication of the length,(L - l) with 1 ×
1 is to project the void formed into cubic dimension represented in a real oxide scale

formed on the Fe-Cr-Ni alloy. Equation (1) shows the formation of voids is independent

of time.

IIUM Engineering Journal, Vol. 12, No. 5, 2011: Special Issue -1 on Science and Ethics in Engineering

Kaderi et al.

Fig. 7: The area fraction of voids in the inner oxide scale measured in the cross section of

samples oxidized in dry and humid environment at 1073 K after 86.4 ks and 172.8 ks.

Figure 8 shows the volume fraction of voids in the outer oxide scale. However an anomaly

occurred for volume fraction of voids in outer oxide scale of samples oxidized in dry

environment. It was shown that the volume fraction of voids in outer scale increased with

time.

Fig. 8: The area fraction of voids in the outer oxide scale measured in the cross section of

the sample oxidized in dry and humid environment at 1073 K after 86.4 ks and 172.8 ks.

IIUM Engineering Journal, Vol. 12, No. 5, 2011: Special Issue -1 on Science and Ethics in Engineering

Kaderi et al.

Nevertheless, the increases in volume fraction of voids for samples oxidized in humid

environment shown in Fig. 7 and 8 are phenomenal than that of samples oxidized in dry

environment. The increase is ranging from 59% to 86%.

It is interesting to note that the volume fraction of voids in outer oxide scale in this

study are in a range of 0.06 to 0.29. Studies on void formation in magnetite scale by

Maruyama [9] shows very small volume fraction of voids in a range of 0.05 to 0.06.

Meanwhile, NiO scale on Ni by Akiba [11] shows volume fraction of voids less than 0.01.

Both studies were conducted in a controlled ��� environment. Therefore, the fact that the

oxidation for the current study was done in normal air with higher ��� has contributed to

the formation of greater magnitude of void in the outer scale.

4. CONCLUSION

The void formation in oxide scales of Fe-Cr-Ni alloy samples exposed isothermally

at 1073 K in air (��� =0.21atm = 2.1 � 10
Pa) and humid (air + steam) were investigated.

It has been shown that the protective Cr2O3 phases and NiO phases occurred in samples

oxidized in dry environment while non-protective Fe3O4 phase occurred in all samples

oxidized in dry and humid environment. The volume fraction of voids formed in inner

scale increased by 84% for 86.4 ks and 86% for 172.8 ks when oxidized in humid

environment. On the other hand, volume fraction of voids formed in outer scales increased

by 76% for 86.4 ks and 59% for 172.8 ks when oxidized in humid environment. The

oxygen partial pressure, ��� greatly influenced the volume fraction of voids formed in the

oxide scale of Fe-Cr-Ni alloy. Hence the oxidation in air contributes greater formation of

void than in controlled environment.

ACKNOWLEDGEMENT

The International Islamic University Malaysia funded this project through the Research

Endowment Grant B (EDWB 0906-333) and the Ministry of Higher Education Malaysia

funded this project through Fundamental Research Grant Scheme (FRGS 0510-127). The

authors gratefully acknowledge the financial support.

REFERENCES

[1] R.J. Ehlers, P.J. Ennis, L. Singheiser, W.J. Quadakkers, and T. Link. “Significance of scale
spalling for the lifetime of ferritic 9–10% Cr steels during oxidation in water vapour at

temperatures between 550 and 650 °C.” Proceedings of the European Federation of

Corrosion Workshop. 2001.

[2] A. Rahmel, and J. Tobolski. “Einfluss von Wasserdampf und Kohlendioxid auf die
Oxidation von Eisen in sauerstoff bei hohen Temperaturen.” Corrosion Science 5.5 (1965):

333-340.

[3] A. Rahmel, and J. Tobolski. “Einfluß von Wasserdampf und Kohlendioxyd auf die
Oxydation von Eisen-Silicium-Legierungen in Sauerstoff bei Temperaturen von 750 bis

1050°C.” Werkstoffe und Korrosion 16.8 (1965): 662-676

[4] J. Ehlers, D.J. Young, E.J. Smaardijk , A.K. Tyagi , H.J. Penkalla , L. Singheiser , W.J.
Quadakkers. “Enhanced oxidation of the 9%Cr steel P91 in water vapour containing

environments.” Corrosion Science 48 (2006): 3428-3454

[5] J. śurek, E. Wessel, L. Niewolak, F. Schmitz, T. -U. Kern, L. Singheiser, and W. J.
Quadakkers. “Anomalous temperature dependence of oxidation kinetics during steam

oxidation of ferritic steels in the temperature range 550–650 °C.” Corrosion Science 46.9

(2004): 2301-2317

IIUM Engineering Journal, Vol. 12, No. 5, 2011: Special Issue -1 on Science and Ethics in Engineering

Kaderi et al.

[6] L. Mikkelsen, and S. Linderoth. “High temperature oxidation of Fe-Cr alloy in O2-H2-H2O
atmospheres; microstructure and kinetics.” Materials Science & Engineering A, 361 (2003):

198-212

[7] T. Maruyama, N. Fukagai, M. Ueda, and K. Kawamura. “Chemical Potential Distribution
and Void Formation in Magnetite Scale Formed in Oxidation of Iron at 823 K.” Materials

Science Forum 461-464 (2004): 807-814

[8] M. Ueda, K. Kawamura and T. Maruyama. “Void formation in Magnetite Scale Formed on
Iron at 823 K -Elucidation by Chemical Potential Distribution-.” Materials Science Forum

522-523 (2006) 37-44

[9] T. Maruyama, and M. Ueda. “Void Formation Induced by the Divergence of the Diffusive
Ionic Fluxes in Metal Oxides Under Chemical Potential Gradients.” Journal of the Korean

Ceramic Society 47.1 (2010): 8-18

[10] W. D. Kingery, H. K. Bowen, and D. R. Uhlmann. Introduction to Ceramics, Second
Edition, pp. 526-530, Wiley - Interscience Publication, New York, USA 1976

[11] K. Akiba, M. Ueda, K. Kawamura, and T. Maruyama. “Quantitative Prediction of Voids
Formation in a Growing Nickel Oxide Scale at 1373 K.” Materials Transactions 48.10

(2007): 2753-2761