Effect of additive on corrosion resistance of NiFe₂O₄ ceramics as inert anodes

XI Jin-hui(席锦会)¹, XIE Ying-jie(谢英杰)², YAO Guang-chun(姚广春)², LIU Yi-han(刘宜汉)²

1. School of Materials Science and Engineering, China University of Mining and Technology,

Xuzhou 221008, China;

2. School of Materials and Metallurgy, Northeastern University, Shenyang 110004, China

Received 4 September 2007; accepted 7 January 2008

Abstract: In order to improve the corrosion resistance of NiFe₂O₄ ceramics as inert anode, additive V₂O₅ was added to raw materials NiO and Fe₂O_3. The inert anodes of nickel-ferrite ceramics were prepared by powder metallurgic method and the static corrosion rate in Na₃AlF₆-Al₂O₃ was determined by mass loss measurement. The effect of V₂O₅ on sintering property and corrosion resistance was studied. The results show that V₂O₅ can promote the grain to develop completely and improve sintering property. EDS results show the reaction product Ni₂FeVO₆ distributes along the grain boundary. The corrosion tests show that V₂O₅ is beneficial to improving corrosion resistance remarkably. The reasons that V₂O₅ can improve the corrosion resistance must be V₂O₅ promoting the gains to develop completely and Ni₂FeVO₆ distributes along the grain boundary. The stable structure can control the chemical dissolution of ceramics anode and the reinforced grain boundary can control the grain-boundary corrosion rate.

Key words: NiFe₂O₄; inert anode; additive; corrosion resistance

1 Introduction

In conventional aluminium electrolysis, the major cell reaction is

1/2Al₂O₃+3/4C=Al+3/4CO₂                     (1)

Thus, using carbon anodes has many disadvantage，such as the consumption of anode carbon and the emission of carbon dioxide and fluorocarbon. With an inert anode, the reaction[1] is

1/2Al₂O₃=Al+3/4O₂                           (2)

The introduction of inert anodes will evolve oxygen [2-3], being friendlily to environment, which makes the use of inert anode commercially attractive. Re- searches have been done in order to find the appropriate material for inert anodes in aluminum production.

However, the inert anode must meet some basic requirements[4]: to exhibit a low corrosion rate in the high-temperature melts and environment of high oxidizability; not to contaminate the produced metal Al; to be economically feasible and to be a good electric conductor. A lot of research work has been carried out,which can be divided into three classes, namely, metals [5-6], ceramic oxides[7-8], and cermets[9-10] , but no material has been found to satisfy all these requirements. One of the formidable challenges is the corrosion property.

The major corrosion mechanisms of inert anodes in electrolyte are chemical dissolution, electrolyte penetration as well as electrochemical dissolution.

As discussed by RAY[11], the cermets anode material may be corroded via several mechanisms such as oxidation, chemical dissolution (fluorination), electrochemical dissolution, reduction by dissolved metal Al, electrolyte penetration, and grain boundary attack. Under conventional electrolysis conditions, chemical dissolution and electrolyte penetration serve as two major corrosion mechanisms[12-13].

For the chemical dissolution, the reasons are the dissociation of NiFe₂O₄ ceramics and the following reactions with electrolyte[14]:

2AlF₃(s)+3FeO(s)=3FeF₂(s)+3/2O₂(g)+2Al(l)     (3)

2AlF₃(s)+Fe₂O₃(s)=2FeF₃(s)+3/2O₂(g)+2Al(l)     (4)

2AlF₃(s)+3NiO(s)=3NiF₂(s)+3/2O₂(g)+2Al(l)     (5)

As for the electrolyte penetration, LI et al[14] thought it may result from two reasons. The first one is due to pores left by the electrochemical dissolution of metal phase when anodes are polarized, or by the metal phase oxidation and the following preferential chemical dissolution. Bath penetrates into these pores by capillary effects. The second one is the selective dissolution of Fe. In the ceramic phase of the NiFe₂O₄-based anode, Fe has a fairly high solubility in cryolite, compared with Ni. So improving the density of ceramics matrix and strengthening the grain boundary are very important.

In this work, the NiFe₂O₄ ceramics with little amounts of V₂O₅ (0-2.0%, mass fraction) were prepared. By determining the density and corrosion rate of anodes and analyzing anode by EDS connected to the SEM, we expected to see how V₂O₅ affects the sintering property and corrosion resistance.

2 Experimental

2.1 Preparation of NiFe₂O₄ ceramics

The raw materials, NiO, Fe₂O₃ and V₂O₅ were all in reagent grade. NiFe₂O₄ ceramics samples with V₂O₅ were prepared by cold pressing-sintering process. NiO and Fe₂O₃ and a proper amount of V₂O₅ (0%, 0.5%, 1.0%, 1.5%, 2.0%, respectively) were mixed by ball milling and then the mixed powders were cold pressed into rectangular blocks (70 mm×18 mm×(10-11) mm) at the pressure of 100 GPa and sintered at 1 200 ℃ for 6 h.

2.2 Performance tests

The density was tested by Archimedes draining method. Samples with V₂O₅ were prepared for characterization by DSC (SDT Q600), DTA (SDT22960) and X-ray diffraction (XRD) (RigakuD/max-RB) to investigate if there are foreign phases produce.

The samples were immerged in molten cryolite, with temperature maintained at 960 ℃ for 10 h, to determine the static state corrosion rate. The electrolyte was made up of reagent grade CaF₂ and AlF₃, technical grade NaF and Al₂O₃. The CR (molar ratio of NaF to AlF₃) was kept to be 2.2, and the concentrations of CaF₂ and Al₂O₃ were both kept to be 5% (mass fraction). The eroded samples were washed in 30%AlCl₃ solution at 100 ℃ to remove the adhering melt. The corrosion rate was determined by mass loss measurement.

Some anodes were analyzed by scanning electron microscope(SEM) (SSM550) and energy dispersive spectrometer(EDS) connected to the SEM.

3 Results and analysis

3.1 Effect of V₂O₅ content on ceramics density

The changes of density and volume shrinkage with the content of V₂O₅ are shown in Fig.1. From Fig.1, it can be found that in the range of 0-2.0% V₂O₅, the higher the V₂O₅ content, the higher the density and the larger the volume shrinkage, especially when the content of V₂O₅ is more than 1%. The fact shows that V₂O₅ can promote sintering property.

Fig.1 Effect of V₂O₅ on density and volume shrinkage

The microstructures of samples without V₂O₅ and with 1.5%V₂O₅ are shown in Fig.2(a) and Fig.2(b) respectively. The grain size in the sample without V₂O₅ is small and uniform, being 2-4 μm commonly, and the grains are granulose. But when 1.5%V₂O₅ is added, the grain size accretes obviously. Furthermore, the grain size is non-uniform, being 2-8 μm, and big grain is octahedron. This shows that V₂O₅ can promote the grain to grow and develop completely.

Fig.2 SEM photographs of samples: (a) Without additive; (b) With 2.0%V₂O₅

In order to analyze the effect of V₂O₅ on sintering, the distribution of elements such as Ni, Fe, V and O on the polished surface of NiFe₂O₄ with 2.0% V₂O₅ was analyzed by scanning electron microscopy, as shown in Fig.3. It can be seen that the element V distributes along the grain boundary.

  ­­

Fig.3 SEM micrograph (a) and distribution of Ni (b), Fe (c), V (d) and O (e) on polished surface of NiFe₂O₄ with 2.0%V₂O₅

Ternary phase diagram of the system NiO-V₂O₅- Fe₂O₃[15] is shown in Fig.4. The reaction products of the compositions of NiO-Fe₂O₃-2%V₂O₅ in this work lies in the XI section. The phase compositions in this section are Ni₂FeVO₆, NiO and NiFe₂O₄.

Fig.4 Ternary phase diagram of NiO-V₂O₅-Fe₂O₃

In order to study if there are new phases produce, the DSC was carried out on the Fe₂O₃-NiO-2%V₂O₅ system, and the curve is shown in Fig.5(a). There is a small exothermic peak at 625-650 ℃. It can be determined that a chemical reaction arises at 625-650 ℃. Due to the fact that the amount of V₂O₅ is too little, the exothermic peak is not obvious. Then the DTA was carried out on the Fe₂O₃-NiO-30%V₂O₅ system, and the curve is shown in Fig.5(b). A exothermic peak exists from 625.07 ℃ to 651.64 ℃. It is same as the former. In order to study the new phase, XRD analysis was carried out. Fig.6 shows the XRD pattern of sample with 2.0%V₂O₅. It can be seen that the new phase is Ni₂FeVO₆. The peak is probably associated with the produce of Ni₂FeVO₆. Combining with Fig.3, we can know Ni₂FeVO₆ concentrates at the grain boundary. According to Ref.[15], the melting point of Ni₂FeVO₆ is low. At the sintering temperature adopted in this work, Ni₂FeVO₆ becomes liquid. The sintering of the system (Fe₂O₃-NiO-2%V₂O₅) is liquid-phase sintering, so in this process, the densification was gotten through mass transfer of liquid phase. The small grains dissolve, and the big grains grow bigger, so the grain sizes are non-uniform. And the liquid can promote the grains to grow and develop completely with the regular octahedron shape. With the amount of V₂O₅ increasing, the amount of liquid increases, so the character of liquid improving sintering is more obvious.

Fig.5 DSC curve of Fe₂O₃-NiO-2%V₂O₃ (a) and DTA curve of Fe₂O₃-NiO-30%V₂O₅ system (b)

Fig.6 XRD pattern of sample with 2.0%V₂O₅

3.2 Effect of V₂O₅ content on corrosion resistance

The effect of V₂O₅ content on corrosion rate is shown in Fig.7, from which it is obvious that adding V₂O₅ can dramatically lower the corrosion rate. Corrosion rate of sample without V₂O₅ is 0.048 2 g·cm^-2·h^-1. The corrosion rate of sample with 0.5%V₂O₅ is 0.011 g·cm^-2·h^-1, 1/4 of that of sample without V₂O₅. The corrosion rate of sample with 1.5%V₂O₅ is the lowest, being 0.001 05 g·cm^-2·h^-1, 1/46 of that of sample without additive.

Fig.7 Corrosion rate of samples with different contents of V₂O₅

After being eroded in molten salt, the two samples are all in good condition, no crack or tumescence. But the sample without V₂O₅ is not smooth, many visible pits on it. On the contrary, the sample with 1.5%V₂O₅ is smooth and no visible pits appear.

The eroded surfaces were characterized by SEM, as shown in Fig.8. From Fig.8(a), it can be found that many big holes on the grains and the grains become loosen because the grain boundary is attacked by molten cryolite. From Fig.8(b), it can be seen that the sample is eroded slightly then its surface is flat. There are only the fewer and smaller holes on the grains and grains bind closely. So the grain and grain boundary all have good corrosion resistance to melt and could effectively hold back the melt penetration.

Fig.8 SEM photographs of samples eroded by molten cryolite: (a) Without additive; (b) With 1.5%V₂O₅

The surface of sample with V₂O₅ is eroded slightly, which is coincident with the result of the samples’ corrosion rate.

The reasons that V₂O₅ can improve the corrosion resistance must be V₂O₅ promoting the gains to develop completely with the octahedron shape and Ni₂FeVO₆ concentrating at grain boundary. The stable structure could control the chemical dissolution of ceramics anode, which is beneficial to improving the corrosion resistance of sample in molten cryolite. Moreover the reaction product Ni₂FeVO₆ concentrates at grain boundary and it has good corrosion resistance, so can reinforce the corrosion resistance of the grain boundary. As the candidate materials of inert anodes, the requirement to corrosion rate is rigorous. Improving the corrosion resistance is the key problem that researchers are making great efforts to solve. In this work, adding V₂O₅ can increase the corrosion resistance of sample remarkably, so V₂O₅ is an interesting additive for NiFe₂O₄ ceramics.

4 Conclusions

1) When V₂O₅ is added to NiO, Fe₂O₃ and Ni₂FeVO₆ are produced, which has lower melting-point. During sintering, it becomes liquid and then promotes sintering property.

2) V₂O₅ can improve the corrosion resistance of samples in molten cryolite remarkably. After being eroded for 10 h, the sample with 1.5%V₂O₅ is in good condition and the corrosion rate of the sample is 1/46 of that of sample without additive.

3) The reasons that V₂O₅ can improve the corrosion resistance are that liquid promotes the grain to develop completely and Ni₂FeVO₆ concentrates at grain boundary. The stable structure could control the chemical dissolution of ceramics anode. Ni₂FeVO₆ has better corrosion resistance than NiFe₂O₄, so Ni₂FeVO₆ controls the grain-boundary corrosion rate.

References

[1] PETERSON R D, RICHARDS N E, TABEREAUX A T, KOSKI O H, MORGAN L G, STRACHAN D M. Results of 100 h electrolysis test of a cermet anode: Operational results and industry perspective [C]// CHRISTIAN M B. Light Metals. Warrendale PA: TMS, 1990: 385-393.

[2] PAWLEK R P. Inert anodes: An update [C]// SCHNEIDER W. Light Metals. Warrendale, PA: TMS, 2002: 449-456.

[3] SADOWY D R. Inert anode for the Hall-Héroult cell: The ultimate materials challenge [J]. JOM, 2001, 53(5): 34-35.

[4] OLSEN E, THONSTAD J. Nickel ferrite as inert anodes in aluminium electrolysis (Part I): Material fabrication and preliminary testing [J]. Journal of Applied Electrochemistry, 1999, 29(3): 293-299.

[5] SHI Zhong-ning, XU Jun-li, QIU Zhu-xian, GAO Bing-liang, WANG Zhao-wen. Copper-nickel superalloys as inert alloy anodes for aluminum electrolysis [J]. JOM, 2003, 55(11): 63-65.

[6] GLUCINA M, HYLAND M. Laboratory-scale performance of a binary Cu-Al alloy as an anode for aluminium electrowinning [J]. Corrosion Science, 2006, 48(9): 2457-2469.

[7] YANG Jian-hong, LIU Ye-xiang, WANG Hua-zhang. Behaviour and improvement of SnO₂-based inert anodes in aluminium electrolysis [C]// DAS S K. Light Metals. Warrendale, PA: TMS, 1993: 493-495.

[8] ZHANG Lei, JIAO Wan-li, YAO Guang-chun. Preparation of NiFe₂O₄ inert anode and its electrolysis corrosion mechanism [J]. Journal of the Chinese Ceramic Society, 2005, 33(12): 1431-1436.

[9] LAI Yan-qing, SUN Xiao-gang, LI Jie, DUAN Hua-nan, LI Xin-zheng, ZHANG Gang, TIAN Zhong-liang. Densification of Ni-NiFe₂O₄ cermets for aluminum electrolysis [J]. Trans Nonferrous Met Soc China, 2005, 15(3): 666-670.

[10] TIAN Zhong-liang, LAI Yan-qing, LI Jie, ZHANG Gang, LIU Ye-xiang. Electrical resistivity of NiFe₂O₄ ceramic and NiFe₂O₄ based cermets [J]. Trans Nonferrous Met Soc China, 2004, 14(3): 421-425.

[11] RAY S R. Inert anodes for Hall cells [C]// MILLER R E. Light Metals. Warrendale, PA: TMS, 1986: 287-298.

[12] LAI Yan-qing, TIAN Zhong-liang, LI Jie, YE Shao-long, LI Xin-zheng, LIU Ye-xiang. Results from 100 h electrolysis testing of NiFe₂O₄ based cermet as inert anode in aluminum reduction [J]. Trans Nonferrous Met Soc China, 2006, 16(4): 970-974.

[13] LUO Tao, WANG Zhao-wen, GAO Bing-liang, YU Xu-guang, SHI Zhong-ning, QIU Zhu-xian. Study on corrosion of cermet inert anode based on nickel ferrite spinel [C]// TRAVIS J G. Light Metals. Warrendale PA: TMS, 2006: 491-493.

[14] LI Jie, DUAN Hua-nan, LAI Yan-qing, TIAN Zhong-liang, LIU Ye-xiang. Effect of NiO content on corrosion behaviour of Ni-xNiO-NiFe₂O₄ cermets in Na₃AlF₆-Al₂O₃ melts [J]. Trans Nonferrous Met Soc China, 2004, 14(3): 421-425.

[15] MARIAK K, ANNA B T. Phase equilibria in the system NiO-V₂O₅-Fe₂O₃ in subsolidus area [J]. Therm Anal Calorim, 2004, 77: 65-73.

Foundation item: Project(2006A050) supported by Science and Technology Foundation of China University of Mining and Technology, China; Project (0702058C) supported by Postdoctoral Scientific Research Foundation of Jiangsu Province, China

Corresponding author: XI Jin-hui; Tel: +86-516-83591880; E-mail: xjh825@163.com

(Edited by LI Xiang-qun)