Elsevier

Corrosion Science

Volume 45, Issue 6, June 2003, Pages 1329-1341
Corrosion Science

Characterisation of the coke formed during metal dusting of iron in CO–H2–H2O gas mixtures

https://doi.org/10.1016/S0010-938X(02)00251-2Get rights and content

Abstract

Carbon deposits formed on the surface of iron samples during carburisation at 700 °C in a gas mixture of 75%CO–24.81%H2–0.19%H2O were characterised by using scanning electron microscopy (SEM), X-ray diffraction (XRD), Mössbauer spectroscopy and transmission electron microscopy (TEM). Cross-section observation of the iron sample by light optical microscopy revealed the formation of cementite after only 10 min reaction, together with a thin layer of graphite. After 4 h reaction, a thick coke layer was formed on top of the cementite surface. SEM surface observation indicated the formation of filamentous carbon in the coke layer. Further analysis of the coke by XRD and Mössbauer showed the presence of mainly Fe3C and small amount of Fe2C but no metallic iron in the carbon deposit. TEM analysis of the coke detected very convoluted filaments with iron-containing particles at the tip or along their length. These particles were identified to be cementite by selected area diffraction. Carbon deposits produced at the same temperature but with other gas compositions were also analysed by using XRD. It was found that with a low content of CO, e.g. 5%, both α-Fe and Fe3C were detected in the coke. Increasing CO content to more than 30%, iron carbide was the only iron-containing phase.

Introduction

Metal dusting is a corrosion phenomenon deteriorating iron, low and high alloy steels and Ni- or Co-based alloys in strongly carburising gas atmospheres (carbon activity ac>1) at elevated temperatures (400–800 °C). The mechanism [1], [2], [3] proposed for iron-based alloys involves a super-saturation of iron with carbon and subsequent growth of a cementite layer at the surface which acts as a barrier for further carbon transfer. Consequently, graphite is deposited on the cementite surface, lowering the carbon activity to ac=1 in the Fe3C/graphite interface and thereby initiating cementite decomposition. It is assumed that the iron atoms formed from Fe3C decomposition diffuse within graphite to the outside and form small iron particles. These particles then strongly catalyse further carbon deposition causing a vast growth of reaction product ‘coke’.

Coke is a mixture of metal-containing particles and carbon deposits, which is often as filaments with very fine metal-containing particles located at the tip or in the filament. According to the above mentioned mechanism these particles should be in the form of iron. In fact fine iron particles were detected in the dust of pure iron samples by Chun et al. [3] and Pippel et al. [4] using transmission electron microscopy (TEM). However, cementite was also reported to be in the dust as an important part and sometimes the only iron-containing phase [5], [6], [7], [8], [9]. Hochman [5] identified the particles in the filaments to be iron and in some cases carbide, but he did not give details of this result. Wei et al. [6] reported that iron particles were present in the coke if low carbon activity gas (e.g. ac=3.3) was used, while with extremely high carbon activities (e.g. ac=4580 at 500 °C) only carbide was detected [6], [9]. Recently, Toh et al. [8] used XRD and TEM to analyse a coke of Fe–25Cr–(0–25)Ni carburised by 68%CO–26%H2–6%H2O at 680 °C (ac=2.9). They found that for the steel with low contents of Ni (0–5 wt.%) only cementite was detected in the coke and at the tip of filaments. For alloys with high Ni contents (10–25 wt.%), in addition to carbide, some austenite was detected. Zeng et al. [7] reported that cementite is the only detectable iron-containing phase in the coke of iron carburised at 593 °C in the gas of 72.4%H2–8.1%CO2–17.2%CO–2.3%H2O (ac=27.3). These reports imply that the type of metal-containing phase in the coke, iron or carbide, depends on the sample and on the reaction conditions. The mechanism of forming these different phases is not very clear and needs further research.

The aim of this work is to characterise the coke formed by carburising a pure iron sample at 700 °C in a 24.81%H2–75%CO–0.19%H2O gas mixture by scanning electron microscopy (SEM), X-ray diffraction (XRD), Mössbauer spectroscopy and TEM. The coke prepared with other gas compositions was analysed by XRD only.

Section snippets

Experimental

Pure iron samples (discs of 20 mm in diameter and 1.1 mm in thickness) were used for the experiments. The samples were first annealed at 850 °C for 1 h in a pure H2 gas atmosphere and then ground on SiC paper to grade 1000. Afterwards, the sample was hung on a microbalance (Sartorius 7287 with an accuracy of 1 μg) by silica hooks. Helium gas was introduced into the chamber of the microbalance to protect the balance. The composition of reaction gas was 24.81–94.81 vol% H2, 5–75 vol% CO and 0.19

Iron carburisation and coke formation

Fig. 1 shows the TGA curve of an iron sample carburised at 700 °C in a 75%CO–24.81%H2–0.19%H2O gas mixture. The iron carburisation is fast at the early stage of the reaction. Then this rate decreases slightly and finally increases drastically after 2 h reaction.

Metallographic cross-sections of specimens after 10 min and 4 h carburisation are shown in Fig. 2. It is clear that even after only 10 min reaction cementite is formed together with a graphite layer on the surface (Fig. 2(a)). A very

Discussion

Following the reasoning of Grabke et al. [1], [2] and some other researchers [3], [4] the iron-containing particles should be α-Fe, since they are in direct contact with graphite. Recent investigations showed already that most of the iron-containing particles are in fact cementite [7], [8], [9]. The same result was also obtained in a different field of research, i.e. catalytic formation of graphite filaments from fine iron powder [14], [15], [16], [17], [18], [19]. Walker et al. [14] found by

Conclusions

A pure iron sample was carburised at 700 °C in a 75%CO–24.81%H2–0.19%H2O gas mixture. Thermogravimetric analysis showed an initial quick carburisation followed by a slowing down of the mass gain and finally, a marked increase of the rate of mass gain. Cross-section observation of the iron samples showed that cementite was formed already after 10 min reaction, together with a thin layer of graphite. After 4 h reaction, a thick coke layer was formed on the surface. Coke formed during metal

Acknowledgements

The authors would like to thank Prof. J. Hesse, Institut für Metallphysik und Nukleare Festkörperphysik, TU Braunschweig, for Mössbauer analysis. The authors would also like to thank Mrs. E. Bartsch for TEM analysis and H. Falkenberg and M. Nellessen for preparing the metallographic cross-sections and the SEM analysis. Support of this study by the Deutsche Forschungsgemeinschaft is greatly acknowledged.

References (19)

  • J Zhang et al.

    Corros. Sci.

    (2003)
  • P Coquay et al.

    Acta Mater.

    (2000)
  • S.J Campbell et al.

    Mater. Sci. Eng. A

    (1997)
  • A.J.H.M Kock et al.

    J. Catal.

    (1985)
  • L.D Mansker et al.

    Appl. Catal.

    (1999)
  • H.J Grabke et al.

    ISIJ Int.

    (2001)
  • H.J Grabke

    Mater. Corros.

    (1998)
  • C.M Chun et al.

    Mater. Corros.

    (1999)
  • E Pippel et al.

    Mater. Corros.

    (1998)
There are more references available in the full text version of this article.

Cited by (0)

1

Tel.: +49-211-679-2481; fax: +49-211-679-2537.

2

Tel.: +49-211-679-2246; fax: +49-211-679-2537.

View full text