J. Phys.

IV France 122 (2004) 69-73

” EDP Sciences, Les Ulis
DOI: 10.1051/jp4:2004122010

Outbursts formation on low carbon and trip steel

grades during hot-dip galvanisation

E.J. Petit1, L. Lamm1 and M. Gilles2

1
LETAM (UMR 7078 CNRS-Université de Metz), Ile du Saulcy, 57045 Metz Cedex 1, France
2
UMICORE, Kasteelstraat 7, 2250 Olen, Belgium

Abstract. Low carbon and TRIP grade steels have been hot dip galvanised in order to study outbursts
formation. Microstructure and texture of intermetallic phases have been observed after selective
electrochemical etching by scanning electron microscopy. Potential versus time (chronopotentiometric)
characteristics were recorded in order to monitor surface modifications. This combination of techniques
enable to quantify and observe intermetallic phase one by one. The overall thickness of coating on both
substrates are similar. However, microstructures of Fe-Zn intermetallic phases are very different on both
grades. In particular, the 9 phase is dense on standard steel but develops a highly branched filament
structure on TRIP steel. The transformation of 9 phase to G and *1 are limited on TRIP steel. Differences of
texture provide clues for understanding mechanisms of formation of outbursts. They can account for the
differences of mechanical properties and corrosion resistance. Silicon from the substrate influences the
reactivity of TRIP steels due to capping and local reactions.

Résumé. La formation des outbursts a été étudiée sur un acier bas carbone et sur un acier TRIP galvanisés.
Les épaisseurs des revêtements sont similaires. Néanmoins, les observations microscopiques et les
érosions électrochimiques montrent que la répartition des phases intermétalliques et leurs microstructures
diffèrent sensiblement en fonction de la nature du substrat. Ces différences expliquent les propriétés
mécaniques et anticorrosions. L’encapsulation de la surface par les oxydes de silicium freine la
transformation de la phase dzêta en delta et gamma sur l’acier TRIP.

1 INTRODUCTION

Outbursts are microstructures made of Fe-Zn phases developing within anticorrosion coatings during
hot-dip galvanisation. They evidence violent local reaction between the galvanisation alloy and steel.
They impair mechanical and anticorrosion properties of the coating not only because intermetallic
phases are brittle and offer a poorer resistance to corrosion, but also because their dispersion produces
grain boundaries which are paths for corroding molecules. Finally outbursts deteriorate the visual
aspect of the coating due to pitting when they reach the surface. This paper compares outbursts grown
on standard low carbon steel and TRIP (transformation induced plasticity) grade steel aiming at
understanding mechanisms of triggering and growth.

Reactivity of galvanisation alloy with low carbon steels is usually regulated by aluminium addition
above 0.15 wt. % in the alloy. Aluminium reacts rapidly with iron and produces a thin Fe2Al5 layer
inhibiting momentarily the formation of Fe-Zn phases during hot dipping. Silicon from the substrate

Article published by EDP Sciences and available at http://www.edpsciences.org/jp4 or http://dx.doi.org/10.1051/jp4:2004122010

 70 JOURNAL DE PHYSIQUE IV

also increases reactivity when concentration in steel ranges of some tenths of wt. %. This is known as
the Sandelin effect. In high resistance steels like TRIP’s, silicon concentration reaches about ~1 wt. %
(~2 at. %) and causes contradictory effects on reactivity due to the formation of a SiO2 protective
layer. This layer results from segregation to the surface and preferential oxidation during hot rolling
and surface preparation. It prevents contact between the galvanising bath and steel. Vigorous reactions
forming outbursts start at cracks in the oxide layer. For these reasons high silicon grade steels are
difficult to galvanise and studying the formation of outbursts deserves special attention.

2 EXPERIMENTAL

The 2 mm sheets of low carbon (hot rolled and cold rolled) HE400M and (hot rolled) TRIP steels were
similarly prepared by alkaline degreasing and acid etching. A 500 g/l zinc and amonium chloride
solution was used for fluxing at 70 °C. After drying in air at 95°C, hot-dip galvanising was performed
at 450°C (without preheating) in a zinc alloy containing 1.1 wt.% tin, 0.1 wt.% bismuth and
0.07 wt.% aluminium. Galvanisation lasted 4 minutes in order to break the silicon oxide layer on TRIP
steel. Zinc in excess was drained by gravity and cooling proceeded naturally in air.

Electrochemical etching were performed in (NaCl, ZnSO4 : 3.4 M, 0.32 M) with acidity fixed at pH =
4 by addition of sulphuric acid. A platinum counter electrode has been used. Anodic dissolution were
performed at constant current density, at room temperature with no steering. Potentials measured
versus time during galvanic dissolution (chronopotentiometric data) are referred to a saturated calomel
electrode (SCE potentials). Standard potentials can be obtained by adding 241 mV.

Chronopotentiometric data have been quantitatively analysed by integrating current over time in order
to correlate charge transfer to the amount of matter etched from the surface. We assumed a double
charge both on iron and zinc ions. Values were converted in thickness without hypothesis on the
nature and composition of each phase, considering average mass and density for zinc, iron and
intermetallic compounds. Averaging causes a + 10 % error because densities of iron, zinc and
compounds are very close.

3 RESULTS

Cross section and chemical etching with nital (2%) reveal intermetallic phases (figure 1). Overall
thickness’ of coatings on both steels are close. A, B, C labels on figure 1 refer respectively to K
(pure zinc), 9 and G phases [1]. Proportion and tridimensional distribution into the coating depends on
the substrate. Quantitative measurements are collected in table 2. Peak to peak variations are
reported too.

Figure 2 shows variations of sample potential during etching. Corrosion potential A, B and C are
fingerprints of K, 9 and G phases. Potential of iron is the highest. (E). Potentials compare with similar
data from the literature [3]. Quantitative analysis of chronopotentiometric data are reported in table 1.
In order to get quantitative agreement between thickness’ measured by microscopy and dissolution,
the galvanic yield is assumed to be between 0.66 and 0.75 when erosions are performed at ~15
mA/cm2. Both kind of data are complementary since microscopy measures mean thickness ; and
electrochemical etching, weight composition. Finally, since electrochemical potentials are
sufficiently different, selective etching of zinc and iron-zinc compounds is possible. Micrographs

Article published by EDP Sciences and available at http://www.edpsciences.org/jp4 or http://dx.doi.org/10.1051/jp4:2004122010

 XXX JEEP 71

obtained by electron scanning microscopy and taken at time pointed by the arrows on figure 1 are
presented on figure 3 to 6.

Figure 1. Cross cuts of coatings on standard (a.) and TRIP (b.) steels.

Table 1. Thicknesses (in microns) of (sub)layers evaluated from micrographs and dissolution.

TRIP steel Standard low carbon steel

Cross section Dissolution Cross section Dissolution
Total 95 + 10 90 85 + 7 84
K 40 + 25 30 33 + 11 18
9 50 + 20 50 40 + 9 43
G ? 11 17 + 6 24

-300
2
15 mA/cm E
-400
Low carbon steel 3
-500 D
SCE potential (mV)

Trip steel
Steel
-600
2 C
-700

1 B
-800

-900 A

-1000
Coating
-1100
-160 -120 -80 -40 2
0 40
Transfered charge (Cb/cm )

Figure 2. Chronopotentiometric data recorded during dissolution at constant current = 15 mA/cm2.

Coating on TRIP steel exhibit extended surface defects (figure 3). Pits entering deeply into the bulk
degrade corrosion resistance. They are caused by emergence of dense underlying intermetallic
structure and local dewetting of zinc. Smooth surface are obtained on standard steel.

Article published by EDP Sciences and available at http://www.edpsciences.org/jp4 or http://dx.doi.org/10.1051/jp4:2004122010

 72 JOURNAL DE PHYSIQUE IV

Figure 3. Surfaces of galvanised TRIP (a.) and standard (b.) steels.

Figure 4 shows that phase 9forms bunches of filaments on TRIP steel which are not observed on
standard steel. Spatial dispersion and porosity of 9 on TRIP probably favours retention of Kphase
(pure zinc) by capillarity during drain off. Dewetting when dense entanglements of 9 filaments reach
the surface corroborates his hypothesis.

Figure 4. Micrographs of the 9 phase on TRIP (a.) and standard (b.) steels.

Figure 5 shows that phase G takes a typical microstructure made of submicron sheets [3]. Sheets are
differently oriented on both substrates : - hexagonal in shape and laid on the surface on TRIP steel, -
perpendicular to the interface on standard steel (dovetailing hinders lateral extension). Coverage
seems more continuous on standard steel. On TRIP steel, piles of sheets are separated by large holes.
Orientations are probably inherited from the parent 9 phase.

Figure 5. Micrographs of the G phase on TRIP (a.) and standard (b.) steels.

Article published by EDP Sciences and available at http://www.edpsciences.org/jp4 or http://dx.doi.org/10.1051/jp4:2004122010

 XXX JEEP 73

Figure 6 shows that phase *1 appears as a cracked smooth film on TRIP steel. XRD signals from the
substrate reveal that the thickness of this layer is ~1 µm. Shiny spots on the picture could be insulating
materials like SiO2. The micrograph of the standard steel discloses nodules. XRD indicates that the
thickness of this layer is larger than 5 µm. Nodules are then probably extremities of a columnar
structure not distinguished on figure 1.

Figure 6. Micrographs of the *1 phase on TRIP (a.) and standard low carbon (b.) steels.

4 DISCUSSION AND CONCLUSION

Selective electrochemical etching of galvanised standard low carbon and TRIP grade steels has been
exploited in order to get a tridimensionnal and quantitative information about the microstructure of
outbursts. These complex microstructures contrast with the layered one observed on galvannealed
coatings. Differences of microstructures of intermetallic phases observed account for the variation of
mechanical and anticorrosion properties of coatings on both steels.
Variations of proportions, morphology and texture of intermetallic phases indicate that composition of
steel and interface structure plays an important role on reaction, nucleation, growth and
transformation of Fe-Zn phases. As a matter of fact, the parent phase 9is directly produced from the
liquidus. Reactions are local on TRIP due to surface screening by the SiO2 layer. Filaments probably
evidences nucleation and fast growth of the 9 phase from the surface as soon as iron saturates the
alloy. Limited transformation of 9 phase to secondary G and * phases on TRIP evidences that the SiO2
layer hinders later iron diffusion.
Additional studies on samples galvanised with increasing duration will provide information about
kinetics of growth and phase transformation, as well as on the nature of the reaction starting points.

References

[1] A.R. Marder, Prog. in Mat. Sci. 45 (2000) 191-271.

[2] X.G. Zhang and I.C. Bravo, Corrosion 50 (1994) 308-317.
[3] A. Besseyrias, F. Dalard, J.J. Rameau and H. Baudin, Corrosion Sci. 39 (1997) 1883-1896.

Article published by EDP Sciences and available at http://www.edpsciences.org/jp4 or http://dx.doi.org/10.1051/jp4:2004122010