18. - 20. 5. 2011, Brno, Czech Republic, EU

THE QUALITY OF CAST IRON SURFACE CREATED AS THE RESULT OF HOT - DIP ZINC GALVANIZING PRECEDED BY HIGH TEMPERATURE OXIDATION

Dariusz JEDRZEJCZYK, Maciej HAJDUGA

University of Bielsko-Biala, 43-309 Bielsko-Biala, Willowa 2, Poland, E-mail address: djedrzejczyk@ath.bielsko.pl

Abstract

The paper describes the high temperature oxidation as the method that allows avoiding some problems arising during hot-dip zinc galvanizing of cast iron. Authors studied the influence of the high-temperature oxidation, as the preliminary stage previous to coating with zinc on the change of surface layer structure as well as subsurface layer of cast iron with flake, vermicular and nodular graphite. Received effects were compared to these obtained during cast iron coating without preliminary thermal processing. To observation both optical and scanning microscope was applied. As the consequence of conducted high-temperature oxidation in subsurface layer of cast iron pores have been created, that in result of coating in liquid zinc were filled with new phase and in this way the new zone with different properties was obtained. To obtain such composite subsurface layer the preliminary chemical treatment is necessary. It was additionally stated that the cast iron layer enriched in zinc is considerably thicker than layers got with application of other methods. Thickness of sub-surface layer where “after-graphite“ pores are filled with zinc depends directly on kind of graphite. In cast iron with flake and vermicular/compacted graphite depth of penetration reaches 120 μm, whereas in nodular cast iron it reaches only 15μm, although sometimes single filled with zinc voids are observed at 75μm depth. Moreover, roughness of cast iron being previously oxidized measured after zinc coating is sometimes even lower than roughness of zinc coating cast iron after turning.

Keywords: High temperature oxidation, Hot dip zinc galvanizing, Cast iron.

1. INTRODUCTION

There are several factors influencing thickness and properties of Zn coating created at the Fe-C alloys surface. Generally, analysis of Fe-Zn equilibrium diagrams shows that expected coating growth sequence should be as follows: Γ1, δ, ζ, η [1-3]. Coating thickness can be controlled by: introducing Al addition to liquid Zn that creates layer blocking coating growth; growth mechanisms can be change by introducing to liquid zinc such addition, like: Ni, Ti, V, Mn [4]; introducing Pb, Bi and Sn can change bath viscosity and surface tension [5]; greater addition of Pb, Bi and Sn results in Zn separation from alloyed coating layer. Aluminum decreases growth of alloyed layers in consequence of inter-metallic phase creation – Al2Fe5 (Fig. 1a) [6]. Nickel concentrates in Fe-Zn phases and causes higher structure compactness. Also temperature level exerts essential influence on coating quality. In temperature range up to 480oC and higher than 530oC Zn coating growth is according parabolic low. Between 490 and 520oC, alloyed layer crystallizes according linear dependence causing rapid and uncontrolled coating growth. Decreasing temperature to 430 oC eliminates Sandeline effect [8]. That is why during hot dip Zn coating of steel elements usually temperatures below 480oC are used. From the other side high temperature Zn coating is used during cast iron treatment. The created at cast iron Zn coating should consist of: Γ1, δ and η phases. So, it follows from the above data analysis that scientific research concentrates mainly on growth model [9,10] and proper choice of bath chemical composition, temperature, fluxing and etching than suitable material surface preparation. It regards especially cast iron zinc coating that was poorly spread until recently because of problems associated with: the large content of Si and P in base material, etching and processing of the casting surface that demonstrates the high content of SiO2 and graphite precipitations.

18. - 20. 5. 2011, Brno, Czech Republic, EU

a b

Fig. 1 Graphite and grain boundaries influence on zinc coating structure: a – scheme of Fe-Al layer breakdown mechanism at steel surface [5]; b – graphite precipitation inside Zn coating (own investigations).

Generally the opinion was formulated that coating of cast iron creates greater problems than steel with high E equivalent value, and proper coatings (without defects) can be created only at decarburized surface of malleable cast iron (decarburization allows avoiding the adverse influence of coating pollutions by graphite precipitations (Fig 1b)) [6]. Therefore using protective zinc covers at cast-iron castings was only occasionally, mainly at the production of elements of industrial fittings made from malleable cast iron, which in the last years (from technical and economic considerations) is replacing with the ductile cast iron. Because it follows from the literature data that creation of Zn layer at crude nodular cast iron casting not requires any special treatment during zinc bath preparation [6], coatings created at crude surface of castings both in high and low temperature processing are continuous and steady. In industrial condition very difficult is to change treatment parameters (bath temperature and chemical composition) only for single elements. This is the reason why surface state stays significant parameter influencing the thickness and quality of coated layer (Fig. 2).

Fig. 2. Influence of the bath temperature and surface quality on the kinetic of zinc – coating growth at the cast iron: 1 - layer created at the crude casting surface; 2 - layer created at the grinding surface; λC - total

alloyed and zinc layer thickness [6].

The assumed investigation targets was: formation of composite subsurface layer, increasing of Zn diffusion depth, formation of outside Zn layer without graphite precipitations and impurities, good quality of coating expressed by relatively low surface roughness, high corrosion resistance of created layer.

18. - 20. 5. 2011, Brno, Czech Republic, EU

2. METHOD OF INVESTIGATION

Experiment was composed of the following stages: proper sample preparation (casting, oxidation, zinc coating), samples surface analyse and corrosion tests. Sample preparation - investigations refers to cast iron with flake, vermicular and nodular graphite melted in induction furnace of medium frequency. Liquid metal was cast to sand mould with dimensions φ30x300mm (cast iron with flake and vermicular graphite) and Y2 ingots (cast iron with nodular graphite).The chemical composition of applied cast iron is presented in Table 1.

Table 1 Chemical composition of cast iron applied in experiment

From such cast ingots samples φ=11mm and length l=110mm were turned. Then samples were put to silite furnace PKS 600/25 to oxidize. The experiment was led in five different temperatures - 850, 900, 950, 1000 and 1050oC. Samples have been taken out from the furnace separately after: 2, 4, 6, 8, 10 and 12 hours. Before oxidation and after cooling dawn samples were exactly measured, weighed, and metallographic specimens were prepared to observe surface perpendicular to sample’s axis. The scale layer was removed in two stage procedure: mechanically by sandblasting, as well as chemically, by dipping in solution of oxalic acid. After scale layer removal the hot dip zinc coating in industrial conditions has carried out. Process consisted of the following action: degreasing – dirt and oil removal; pickling – rust, scale and carbon deposit removal; rinsing – hydrochloric acid removal; fluxing – increasing of zinc adhesion to alloy; galvanizing in temperature 460 oC, within t=1min in bath enriched with Ni, Bi and Al; cooling – decreasing of sample temperature. Surface analyse - to optical observation microscope "NEOPHOT 2” as well as “Axiovert A -100” were applied, whereas further examination was made with application of scanning microscope “Jeol J7” and X-ray analyzer “JCXA – JEOL”. Sample’s surface quality was described additionally by roughness measurements made after scale layer removal - „MAHR” profile measurement gauge with ”Perthometer Concept” software. Corrosion tests - corrosion resistance will be evaluated using samples φ=14,7 mm and thickness 7mm, by application of potentiostat SI 1286, that enables registration of polarization curves in three electrodes system. Before measure start, samples will be placed in corrosion solution – 0,5M Na2SO4, in temperature 25oC, within 24h. Samples will be subjected to polarization in the same solution, from potential 1000mVNEK in anode direction, with rate 1mV/s. Results will be compared to these measured for steel tested in similar conditions.

3. RESULTS OF INVESTIGATIONS

Cast iron oxidation leads to several effects: scale layer creation, decarburization, after-graphite voids creation (with dependence on graphite kind). The scale layer covering cast iron samples can be divided on two zones - first situated outside the initial surface of sample where porosity is directed parallel to the sample radius as well as internal layer where the direction of porosity is diverse. It can be stated that cast iron oxidation at high temperature leads to similar effect that is observed in case of pure iron - low carbon steel – at the surface the scale layer forms that is composed of three layers: wustite (FeO), magnetite (Fe3O4) and hematite (Fe2O3), additionally because of higher silicon content the internal scale layer in composed of wustite and fayalite mixture. Moreover in subsurface layer of cast iron the area free from graphite which thickness achieves even 1,6mm is observed. So, generally the observed structure of cast iron can be divided on three layers. The conducted process of high-temperature oxidation leads to creation not only the scale

Graphite shape

Chemical composition, wg. %

C Si Mn P S Mg Ce

Flake 3,32 1,80 0,55 0,065 0,035 0 0

Nodular 3,63 2,55 0,10 0,025 0,007 0,045 0

Vemicular 3,65 2,58 0,08 0,023 0,008 0,025 0,015

18. - 20. 5. 2011, Brno, Czech Republic, EU

layer and subsurface porous layer but also causes the change of metal matrix character from pearlitic to ferritic. Both measured parameters: the diameter of metal core as well as the samples mass enlarges more intensely in initial stage of oxidation. This is probably caused by growth of scale layer, which influence can be compared to the coat inhibiting the course of oxidation process. Additionally, the silicon content interaction can’t be neglected; especially in cast iron where its value is much higher than in steel. Silicon favours the decrease the speed of oxidation, both during external, as well as internal corrosion. It follows from obtained dependence that porous layer thickness can be controlled by suitable selection of temperature and the time of oxidation as well as it depends on graphite shape existing in cast iron. Porous layer thickness decreases when graphite shape is changing from flake to vermicular and nodular. Moreover, whereas in cast iron with flake and vermicular after-graphite voids are filled with new product, in nodular cast iron chemical composition of the filler is similar to metallic matrix (Fig. 3). Chemical composition of phase filling the after–graphite (flake and vermicular) pores is similar to internal scale layer covering sample surface. With reference to above creation of composite layer in nodular cast iron is practically impossible, whereas in cast iron with flake and vermicular graphite necessary is not only removal of outside scale layer but also after graphite pores filler. The assumed effect can be realized only by proper chemical treatment. When the chemical processing was used – dipping in solution of oxalic acid, the penetration depth considerable increase and achieves 120µm in cast iron with vermicular and flake graphite.

a b

c d

Fig. 3. Structure and linear changes of measured elements concentrations after oxidation: a, b – cast iron

with vermicular graphite; c, d – cast iron with nodular graphite.

Differentiation of the oxidation effect results from diverse graphite structure: flake and vermicular graphite precipitations are connected to each other, whereas nodular graphite precipitations are separated. Thickness of zinc layer that was created at the surface of cast iron samples is stable and reaches 70-100µm. Essential in this case is fact that graphite precipitation frequently penetrate in coated zinc layer and can in this way reduce it tightness (Fig.1b, 4a).

18. - 20. 5. 2011, Brno, Czech Republic, EU

a b

Fig 4 Microstructure of cast iron with flake graphite after hot dip zinc galvanizing; a – zinc coating without oxidation, b – zinc coating with oxidation, 950 oC, 8h, scale removed mechanically.

a b

Fig. 5 The change of Zn concentration at the cross section of created Zn layer; a – points of analysis, b – change tendency.

Such situation is not observed in case of oxidized cast iron. Cast iron, where the scale layer was removed only mechanically – using sandblasting, does not reveal deep zinc penetration inside the “after-graphite” pores – it reach about 15μm (Fig. 4b). It is evident that oxidation process results in higher cast iron surface roughness. Surface state is changing more rapidly in cast iron with flake graphite. In this case roughness express by Ra value reach even more than 9µm, after 12 hour of oxidation at 950 oC. When graphite shape change from flake to vermicular and nodular, roughness changes stay smaller. Measured roughness in vermicular and nodular cast iron after 12 hour oxidation at 950 oC reach correspondingly 8,2 and 6,8µm. Although zinc coating makes surface more smooth the final quality of obtained surface changes in dependence on parameters of high-temperature oxidation. The greatest surface smoothness - Ra<6,3μm was got for samples oxidized at parameters not higher than T≤1050o C and t≤8h. For example, roughness measured at sample with flake graphite, that was oxidized before zinc coating (T=950 oC, t = 8h) gained value Ra = 3,62μm. It means decreasing in comparison to oxidized surface after sand blasting about 3,26 µm. Although the measured values Ra not differs essentially, it is visible that roughness obtaining at the surface galvanized after turning (Ra = 4,06μm) is higher that this one measured after galvanizing treatment combined with oxidation (compare surface at Fig 4a and 4b). The higher Fe (lower Zn) content was measured in internal coating sub-layer, which decreases with increasing of distance from cast iron surface. Such results confirm the growth model presented in literature. It follows from measured Fe, Zn and C concentration that ζ and η phases thickness reaches about 60 µm whereas the rest of layer is occupied by δ phase (Fig 5).

18. - 20. 5. 2011, Brno, Czech Republic, EU

Realization of assumed results:

1. Formation of the composite subsurface layer, √ 2. Increasing of Zn diffusion depth, √ 3. Formation of outside Zn layer without graphite √ precipitations and impurities, 4. Good quality of coating expressed by relatively √ low surface roughness, 5. Higher corrosion resistance of created coating. ?

Fig. 6. Scheme presenting assumed targets realization

4. CONCLUSIONS

1. During the experiment, as the result of two stage hot-dip zinc galvanizing of cast iron, the following targets were realized (Fig. 6): formation of the composite Zn-metallic matrix subsurface layer; increased Zn diffusion depth; formation of outside Zn layer without graphite precipitations and impurities; good quality of coating expressed by relatively low surface roughness. 2. Conducted investigation allows to suppose that corrosion resistance of such created coating should be better than resistance of the layer obtained without high temperature oxidation. The formulated assumption will be verified by corrosion tests in 0,5M Na2SO4 and results will be compared to these measured for steel tested in comparable conditions 3. Creation of composite layer in nodular cast iron is practically impossible, whereas in cast iron with flake and vermicular graphite necessary is not only removal of outside scale layer but also after graphite pores filler. The assumed effect can be realized only by proper chemical treatment, without necessity of pressure processing. The use of chemical processing causes, that depth of zinc penetration gets higher - to 120μm level. 4. Roughness measured after zinc coating of cast iron being previously oxidized is sometimes even lower than roughness of zinc coating of cast iron after turning. It follows first from negative influence of graphite precipitations uncovered during turning that can change locally the thickness and structure of coated layer and decrease in this way its tightness.

LITERATURE

[1] JORDAN C.E., MARDER A.R. Fe-Zn phase formation in interstitial – free steels hot-dip galvanized 450oC. Journal of Material Science, 32, 1997, 5593-5602. [2] MITA K., IKEDA T., MAEDA M. Phase diagram study of Fe-Zn intermetallics. Journal of Phase Equilibria, 23, 2000, 1808-1815. [3] SU X., TANG N.Y., TOGURI J.M. Thermodynamic evaluation of the Fe-Zn system. Journal of Alloys and Compounds, 325, 2001, 129-136. [4] REUMOUNT G., FOCT J., PERROT P. New possibilities for the galvanizing process: the addition of manganese and titanium to the zinc bath. From Conference Proceedings. Intergalva 2000, Berlin 2000, 66-71. [5] FRATESI R., RUFFINI N. Use of Zn-Bi-Ni alloy to improve zinc coating appearance and decrease zinc consumption in hot dip galvanizing. From Conference Proceedings. Intergalva 2000, Berlin 2000, 98-104. [6] MURDER A.R. The metallurgy of zinc coated steel. Progress in Materials Science, 45, 2000, 191-201. [7] KOPYCINSKI D. Monograph, AGH Krakow, 2006. [8] KOPYCINSKI D., GUZIK E., WOLCZYNSKI W, Coating Zn formation during hot dip galvanizing, Materials Engineering, 4, 2008, 289-292. [9] KOPYCINSKI D., GUZIK E., Zinc layer at the nodular cast iron surface. Materials Engineering, 6, 2008, 780-783.