Surface oxidation and decarburization

Other serious problems encountered during hot stamping are surface oxidation and decarburization. Decarburization becomes serious when the metal is heated to temperatures of 700°C or above when carbon in the metal reacts with gases containing oxygen or hydrogen [14]. Imai et al [15] investigated galvannealed steel sheets heated to 900 °C for 5 minutes and found that the coating completely transformed into a zinc oxide layer and there’s iron zinc solid solution containing 20–30 wt.% zinc left after hot stamping while iron oxide was not observed on the surface. Therefore, most sheet metal blanks are pre-coated with a protective layer to solve these problems. Al–Si coating layer is commonly used for preventing scale formation on the steel during the direct hot stamping operation [16-20]. Borsetto et al. [20] studied the influence of thermal process parameters on the chemical behavior of the coating of the Al–Si layer, and the results showed that this protective layer prevents the formation of scales in the direct hot stamping process. Lee also reported that thin Al-10 pct Si alloy (~25) coating help to prevent the steel from decarburization and surface oxidation during the high-temperature thermal cycle in hot stamping process. However, due to the lower forming limits of the Al–Si layer compared to the base material in the initial state at room temperature, the hot-dip aluminized sheets cannot be used for the indirect process and they are not suitable for cold forming. Moreover, Al-Si coating does not provide cathodic protection, like zinc or cadmium, but possesses a high barrier protection. From another aspect, according to Mori and Ito [21], two kinds of different oils were applied for the prevention of oxidation, which were conducted and evaluated in a cooling experiment without forming and in a hot bending experiment. The examination of the sheet surface showed that the number of lubrications (up to 4 times) reduced the

either adding suitable elements that will form compact scales (such as alumina and silica) above in the coating or covering a protective barrier layer (ex. Painting or Oils) will help to reduce the level of oxidation.

2.1.4 Different types of protective coatings for hot stamping process

De Cooman et al. [78] reviewed different coatings for hot stamping use, including Aluminized coating, Galvanized coating (GI), Galvannealed coating (GA), Zn-Ni coating and Hybrid coating (see Fig. 2.1.3). It can be noticed that the research of Zn-Ni as a protective coating for hot stamping published are comparatively fewer than other coatings. The aluminized coating forms two different structures through different heat treatment methods, type 1 exhibits layered structure containing Fe₂AlSi₂ phase and Fe2Al5 phase while type 2 contains ductile Fe3Al phase and FeAl phase. In addition, the Fe-Zn intermetallic compounds and α-Fe(Zn) phase can be obtained by galvanized and galvannealed coating through the holding time during heat treatment. The hot stamping related characteristics of protective coatings are listed in Table 2.1.2 (The hybrid coating is not discussed in this study). Both two types of aluminized coatings exhibit excellent oxidation resistance by forming compact and thin Al2O3 layer on the surface. GI and GA are also reported to have enough oxidation resistance due to the trace of Al in the coating diffuses to the surface and form Al2O3. Therefore, the oxides on the GI and GA surface after hot stamping contain Al₂O₃ and ZnO. On the other hand, the Zn-Ni coating exhibits the poorest oxidation resistance among the protective coatings. No Al in the Zn-Ni coating results in large amounts of ZnO on the surface after hot stamping. Liquid metal induced embrittlement (LMIE) is commonly observed in two metals with low mutual solubility. The solubility of Zn in α-Fe is high while it is relatively low in γ-Fe.

When the temperature is above peritectic temperature, there will be liquid zinc present.

If liquid zinc contact withγ-Fe, it will lead to the embrittlement of austenitic steel by facilitating the decohesion of grain boundaries [79]. There’s no LMIE found in aluminized coatings for the rapid formation of high melting point Fe-Al intermetallic compounds. The Zn-Ni coating can avoid LMIE by increasing nickel content in the coating. LMIE phenomenon is commonly seen in GI and GA coatings where liquid zinc might present during hot stamping. The use of indirect hot stamping process or increasing the holding time at high temperatures are developed to prevent LMIE on GI and GA specimens [8, 22-23,78, 80]. Aluminized coatings provide barrier protection by forming stable corrosion products while GI, GA and Zn-Ni coatings provide cathodic protection due to its more negative electrochemical potential than steel. Since the boiling point of Zn (907°C) is relatively low, the coating evaporation will become serious when heating above the boiling temperature of Zn. Therefore, the evaporation problem occurs in GI and GA coatings while in Zn-Ni coating is suppressed by stronger Zn-Ni bonding. This problem is not found in aluminized coatings due to the higher boiling point of Al (2519°C). The layered structure of type 1 aluminized coating is good for resistance spot welding in Drillet’s study [81], which is reported to have a wide welding current range of 1.4 kA. Faderl [80] studied the RSW properties of GI specimen after hot stamping. It was found that the surface oxides on the surface will increase the contact resistance of RSW. This result was also found in Genderen’s study on GA stamped specimens [82]. Therefore, the surface oxides must be eliminated from the coating prior to welding. The Zn-Ni also need to clean the surface oxides to reduce the contact resistance, but it was reported that Zn-Ni coating has lower electrode wear and wider welding current range than that of GI and GA coatings. In the aspect of paintability, it is necessary for GI and GA to be phosphated before painting [80] while aluminized coating can be painted directly without phosphate layer [83]. On the other

hand, the formability of type 2 aluminized is better than type 1 for the presence of ductile Fe3Al and FeAl phases. Both Fe-Zn and Zn-Ni intermetallic phases are brittle at room temperature. The α-Fe(Zn) phase is ductile under tensile stress at high temperatures, but the study of Fe-Zn-Ni γ phase are not reported in papers.

Fig. 2.1.3 Comparison of the different coating systems for hot stamping application [78]

2.2 Anomalous co-deposition theory

According to Brenner’s definition proposed in 1963 [24], since the less noble metal zinc deposits preferentially on the cathode and its percentage in deposit is higher than that in the electrolyte, which typically occurs in zinc-iron group metal co-deposition.

Based on electrochemical theory, the nickel deposition should prior to zinc in the co-deposition of zinc–nickel alloy since the equilibrium potential of nickel is far more positive than zinc; however, many research found that zinc deposits preferentially in most practices [25-28]. Although there are many attempts made to explain the anomalous co-deposition, still no universal theory accepted due to the complicated kinetic process.

2.2.1 Hydroxide suppression theory

One model proposed that anomalous co-deposition was attributed to the increase of Ph value at the cathode surface being able to induce zinc hydroxide precipitation, which inhibits nickel discharge [29-32]. In Dahms’ research [31], the experiment was conducting with dropping mercury as cathode. The result showed that the anomalous co-deposition was observed at high overpotential deposition region, which might due to the hydrogen evolution of proton causing the increase of Ph value, thus zinc hydroxide formed on the cathode. In order to prove this hypothesis, Dahms also did the experiment of single iron group metal deposition, and the results showed that deposition rate was not hindered by its hydroxide. As the critical Ph for precipitation of iron-group metal hydroxides is significantly higher than for precipitation of zinc hydroxide, the former may not form so that M-deposition requires direct discharge of M²⁺ ions through the zinc hydroxide film. Fukushima’s group [30, 33-35] also studied the mechanism of

anomalous co-deposition, and they found two strong evidences to support hydroxide suppression model. First, the Ph value on the cathode increase drastically when the current density achieves the critical value, which causes the anomalous co-deposition.

Second, there’s a sharp increase of the impedance between cathode and reference electrode. However, there are still some experimental results contrary to this theory. In Fabri Miranda’s study, the zinc content in coatings decreased with the rise of Ph value of zinc-nickel alloy electrolyte. In addition, this theory does not explain the strong inhibition of nickel reduction observed in the normal deposition region, the high current efficiency during anomalous codeposition and the increase in nickel content in the alloy with increasing pH value.

2.2.2 Underpotential deposition (UPD) theory

Another theory assumes that underpotential deposition (UPD) of Zn provides an alloy surface that is different from the parent metal for the continuous codeposition, which means thepreferential deposition of zinc is due to the underpotential deposition of zinc on the surface of iron group metals since zinc deposition potential on surfaces of iron group metals is much higher than that on surfaces of the other metal, thus making the deposition of the less noble component preferable [36-37]. However, if the model was correct, once a monolayer is deposited, the UPD should be terminated and the ions in solution should “sense” only the last layer deposited on the surface. Hence, such a theory is valid only if an alternating multilayer coating is formed, but it cannot explain how zinc deposits continuously when the surface of zinc–nickel coatings is entirely occupied by zinc atoms. Moreover, this theory also cannot describe the preferential deposition of nickel in some specific conditions, such as low current density and low deposition potential [29, 38-40].

2.2.3 Exchange current density theory

According to a third theory, the great difference between the exchange current densities of Zn and the iron-group metal results is a significant difference between the thermodynamic and the practical nobility. Therefore, the magnitude of the exchange current density is generally much greater for Zn compared to Ni, Co and Fe [36, 41-43].

In Hegde group’s study [44], they found this model more appropriate to explain the deposition behavior in the Zn–Ni-Co system. It should be noted that such a model may be proper for electroplating under galvanostatic conditions in their study, where a high current consumption by one element must be at the expense of another element, but their results may not be applicable to electroplating under potentiostatic conditions .

2.3 Factors of Zn-Ni co-deposition process

Generally speaking, the alloy co-deposition process is affected by many factors, including bath composition, temperature, additives, current density, deposition potential and bath Ph [38, 45–50]. Mathias et al. [51] studied the effect of bath composition, temperature, electrolyte velocity, current density on the alloy composition and current density distribution within the cell. The experimental results showed that composition uniformity could be achieved when the mass transfer was fast relative to the electrode kinetics, so that the surface concentrations of the reacting species remained essentially at their bulk values. Therefore, the temperature and bath composition could affect the average alloy composition through their influence on the electrode kinetics. In conclusion, an acceptable theory would be developed if the zinc–nickel co-deposition mechanism under various deposition conditions is made certain.

2.3.1 Deposition temperature

The effect of temperature on the electrodeposition of zinc–nickel alloy from alkaline bath and acid bath had been studied [39, 52-53]. The structure, composition, mechanical, optical and thermal property of electroplating alloys has strong relationships to deposition temperature. In most cases, the deposition rate is also subject to deposition temperature since the diffusion of metal ion from bulk to cathode will be accelerated with the rise of deposition temperature. With Lee’s group studied Zn-Ni co-deposition in alkaline bath [39], Abou-Krisha’s group in sulphate bath [52] and Qiao’s group [53] in chloride bath, the results of them showed that the nickel content in zinc–nickel alloy coating was dependent on the deposition temperature when other plating parameters were fixed. The nickel content in the coating of three different

electrolytes went up with the increase of temperature, which might due to the result of intrinsically slow nickel kinetics. According to the literatures [38-40], it is suggested that the preferential deposition of nickel is attribute to a mixed intermediate ZnNi⁺_ads which catalyzes the reduction of nickel ion. The preferential deposition of zinc is attributed to the intermediate Zn⁺_ads which play a catalytic role on the deposition of zinc rich coatings. Qiao’s research group pointed out that the formation of Zn⁺ads on the surface of cathode accelerating the reduction of zinc and blocking the nickel to deposit at low temperature. On the other hand, once the temperature increases, the formation of mix intermediate ZnNi⁺ads dominates the reduction process, which catalyzes nickel deposition and hydrogen evolution on the surface of cathode.

2.3.2 Bath content ratio

Rehim et al. [32] observed that the Ni content in the deposits decreases with increasing Zn content or with decreasing Ni content in the bath when other parameters are fixed, which might due to the fact that an increase in metal content in the bath tends to oppose the depletion of that metal in the cathodic diffusion layer. Roventi et al. [40]

studied the relationship between the deposition potential and bath content ratio. The results showed that at a potential of -700 Mv the alloy composition is almost constant and does not seem to depend on the bath composition; while deposition potential decreases, the nickel contents in the deposits rises with the increase of Ni²⁺ ion content percentage in the bath. Byk et al. [56] investigated the influence of different bath content ratio on compositions in the deposits, deposition rate and current efficiency in chloride bath and ammonical diphosphate bath. The deposition rate and current efficiency reduces with Ni²⁺ ion content percentage increase in the bath, while the

ammonical diphosphate bath, with the change is sharper in ammonical diphosphate bath than in chloride bath.

2.3.3 Additives

Albalat et al. [45] have examined the relationship between additives and deposit properties in Zn-Ni coatings. In their work, the electroplating of Zn-Ni alloys from a chloride bath containing two brighteners (phenolic derivative and an unsaturated aromatic compound) and a leveling agent (an aromatic carboxylate) has been studied under different plating conditions. According to the results, the composition and morphology of the alloys deposited depended on the concentration of all the additives and also on the temperature. On the other hand, from their results it could be concluded that the corrosion resistance of the coating was related more to their morphology than to their composition. The best behavior could be obtained with alloys that had nodular grains of measurable size, while those that had elongated or non-measurable grains always gave lower corrosion resistance.

2.3.4 Electrolyte type

Many studies of Zn-Ni co-deposition system were conducting in chloride bath [26, 29, 40, 51, 53-54, 56] and in sulfate bath [30, 34, 37, 55]. One reaction model has been proposed by Chassaing and Wiart [54] in chloride bath. A mixed compound (ZnNi⁺ads) dominates the deposition of zinc-rich alloys. This compound plays a role as a catalyst for nickel deposition, and is incorporated in the alloy co-deposition with increasing polarization, thus allowing zinc to deposit preferentially. At low cathodic polarization, which normal co-deposition usually occurs, the deposition of nickel-rich alloys was

attributed to a mixed intermediate (ZnNi⁺ads), which catalyses the reduction of Ni²⁺ ion.

On the contrary, at high cathodic polarizations, which anomalous co-deposition often occurs, zinc preferential discharge is attributed to the intermediate Zn⁺_ads, catalyst for the deposition of zinc rich deposits. Metal cations can easily form chloro-complexes in chloride electrolyte [51]; however, it should be considered the reactive species in sulfate electrolyte are Ni²⁺ and Zn²⁺ ions. There are four factors found in Fabri Miranda’s research in sulfate bath [55], (1) ZnNi⁺ads acts as a catalyst in nickel deposition (2) The intermediate H_ad involved in hydrogen evolution at low polarizations (3) The intermediate adion Zn⁺ads acts as a catalyst in zinc deposition at high polarizations (4) The anionic Zn/NiSO₄^2-_ads inhibits zinc deposition at intermediate polarizations through the incorporation of S deposition.

Chassaing’s reaction model in chloride bath [54]

Low polarizations and high temperature: normal deposition and hydrogen evolution

2

High polarizations and low temperature: anomalous deposition

*

Fabri Miranda’s reaction model in sulfate bath [55]

Low polarizations and high temperature: normal deposition and hydrogen evolution

(7.2)

High polarizations and low temperature: anomalous deposition

(9)

Fig. 2.3.1 Scheme of the reduction reactions that predominate in different potential domains [55].

2.4 High temperature oxidation theory

From simple oxidation reaction:

)

The solid scale MO will separate the two reactants as shown below in Fig. 2.4.1

Fig. 2.4.1 Diagram of separated two reactants (metal and gas) [57].

According to the textbook [57], for the oxidation to move on, at least one reactant need to penetrate the scale, either metal must be transported through the oxide to the oxide–gas interface and react there, or oxygen must be transported to the oxide–metal interface and react there.

There are two types of the scale are shown in Fig.2.4.2, which also presents two oxidation mechanisms. The growth of the scale of these two types is different. Cation migration from substrate leads to scale formation at the scale–gas interface whereas anion migration from atmosphere leads to scale formation at the metal–scale interface.

In most cases, to describe simultaneous migration of ions and electrons it is necessary to assume that the oxides are non-stoichiometric compounds. Non-stoichiometric ionic compounds are classified as semiconductors with the chemical formula M1+δO and M_1−δO (the value of δ varies widely from 0.05 to 0.001). NiO, FeO &Cr₂O₃ are p-type cation-deficit semiconductors; therefore, the cations will migrate with electrons from the scale–metal interface to the scale–gas interface during oxidation. On the other hand,

ZnO is an n-type cation-excess semiconductor, having interstitial Zn ions and equivalent electrons within the conduction band. The oxidation occurs when anions migrate from the scale–gas interface to the scale–metal interface. The scheme of ZnO and NiO are shown in Fig. 2.4.3.

Fig. 2.4.2 Interfacial reactions and transport processes for high-temperature

oxidation mechanisms (a) p-type(cation mobile) (b) n-type(anion mobile) [57].

(a) (b)

Fig. 2.4.3 (a) Interstitial cations and excess electrons in ZnO – an n-type metal-excess semiconductor. (b) Typical p-type metal-deficit semiconductor NiO with cation vacancies and positive holes. [57].

When two reactants are separated by the scale produced, it is necessary to postulate that ionic and electronic transport processes through the oxide are accompanied by ionizing phase-boundary reactions and formation of new oxide at a site whose position depends on whether cations or anions are transported through the oxide layer.

Take a simplified treatment of diffusion-controlled oxidation in textbook [57].

Assume all thermodynamic equilibriums are established at each interface, and the cationic transport through the scale governs the rate of oxidation, the process are listed below. The outward cation flux, j_M²⁺, is equal and opposite to the inward flux of cation defects (here taken to be vacancies). This model is shown in Figure 2.4.4.

Thus, j_M²⁺ can be expressed as in Equation: [2.3.1]:

where x is the oxide thickness, D_VM is the diffusion coefficient for cation vacancies, and C’VM and C’’VM are the vacancy concentrations at the scale–metal and scale–gas

where x is the oxide thickness, D_VM is the diffusion coefficient for cation vacancies, and C’VM and C’’VM are the vacancy concentrations at the scale–metal and scale–gas