The behaviour of solid powder aluminized steel in molten SAC205 solder alloy

Introduction

Although the new lead-free, tin-based solders comply with EU regulations and contain no lead, they cause severe degradation of soldering tools, greatly reducing their lifespan. Equilibrium intermetallic compounds (FeSn and FeSn₂) form at the soldering temperature (around 593 K), where Fe and Sn are in contact. Even after multiple soldering cycles, an intermetallic compound layer of FeSn or FeSn₂ develops at the interface of the Fe/solder alloy, continuously thickening during soldering as Fe atoms dissolve from the layer directly into the compound. In manual soldering, the iron can even be pierced, exposing the inner Cu, which further reacts with tin. For selective wave soldering nozzles, the uneven surface leads to bouncing of the solder wave, preventing the formation of proper joints. When using lead-containing solders (before 2006), companies were less concerned about the slow deterioration of the solder, as the tool problems did not significantly affect quality or cost. However, with the adoption of new lead-free solders, tool degradation has accelerated 4 to 5 times compared to lead-based solders.^1–3 The increasing concern about preventing soldering iron degradation has inspired research, with several studies exploring the causes of tool failure. Japanese researchers T. Takemoto ⁴ and H. Nishikawa ⁵ were among the first to address this issue. T. Takemoto et al. studied the resistance of the iron coating on manual soldering tools to various tin-based lead-free solders. They conducted immersion experiments on pure iron and alloy steel samples, as well as on actual soldering iron tips. The experiments involved immersion times of 2–8 h at temperatures between 350–450 °C. They found that the dissolution of the iron layer occurs 3–3.5 times faster in lead-free solders than in lead-based Sn-37Pb solders. Higher temperatures and longer immersion times further accelerate this degradation. ⁶ Another study examined the dissolution of stainless steel immersed in lead-free solder, yielding similar results. Since stainless steel is resistant to aggressive tin-based lead-free solders, this suggests it is not a suitable solution for protecting the copper core of the tools. ⁷ Nishikawa et al. reduced the formation of the FeSn₂ phase by adding Co to the solder. Although this process is effective, it negatively impacts the quality and reliability of solder joints. ⁸ J. Watanabe et al. studied the resistance of an Fe-multi-walled carbon nanotube (MWCNT) composite in lead-free solder. They found that the resistance of the composite increases, but the wetting decreases, which has a negative impact for the stability of the solder wave. ⁹ Additionally, producing composites is much more costly, making economical device production an important consideration. M. Benke et al. explored the possibility of making steels of different compositions (DC04, C45, CK60, C105U) resistant to the harmful effects of aggressive lead-free solder through nitriding. Besides preventing the formation of the FeSn/FeSn₂ intermetallic phase, the nitride layer was well wetted by the solder melt. ¹⁰ In a study by Zsolt Sályi et al., the resistance of two-phase iron boride (FeB/Fe₂B) coatings on various steel substrates (DC04, C45, CK60, and C105U) was tested against lead-free liquid solder alloy SAC309 (Sn-3Ag-0.9Cu). Results showed no reaction between the borided steels and the solder melt after 40 days of continuous testing. Moreover, a borided nozzle was successfully tested under industrial conditions. ¹¹ M. Benke et al. also examined the interaction between TiB₂-coated steel samples and Sn-Ag-Cu SAC309 by static immersion at 593 K for 40 days. The study compared TiB₂-coated steels with native Ti and B oxides on the surface and oxide-free TiB₂-coated steels. Results indicated that none of the TiB₂ coatings—native oxide-covered and oxide-free— reacted with the solder and so both maintained proper wetting. ¹²

Considering these limitations, aluminized coatings may be regarded as one of the most promising potential protection solutions. Aluminized steels are well-established in high-temperature and corrosive industrial environments—such as boiler tubes, petrochemical pipelines, hot-stamping automotive components, and erosion-resistant energy-sector parts—owing to the formation of a dense and chemically stable Al₂O₃ scale.^13–16 This oxide layer provides exceptional oxidation and corrosion resistance even under long-term thermal exposure,^14,17 suggesting that Fe–Al intermetallic coatings may also offer reliable protection in aggressive molten Sn-based lead-free solder environments. Traditional solid powder aluminization is usually carried out at 900–1000 °C, which leads to high energy consumption, coarse microstructural changes and reduced efficiency. Further studies have used a pack aluminization method based on Al, Al₂O₃ and NH₄Cl powders on steel and stainless steels, and it was found that the resulting aluminide coatings have a high level of corrosion resistance in aggressive environments. ¹³ The optimal coating quality was observed in the range of 700–900 °C, and aluminization carried out at 800 °C for 3 h resulted in a dense and 180 μm thick layer, providing the highest hardness and oxidation resistance. ¹⁴ Aluminized steels have a wide range of applications: in the automotive industry they provide high mechanical strength and weldability,^15,16 reduce corrosion in heat-resistant boiler parts, ¹⁷ in oil and gas pipelines at high temperatures the Fe₂Al₅ phase forms and acts as a protective layer,^18,19 and modifications based on Al-Si alloys increase the durability of automotive safety parts.^20–22 In the energy sector, aluminide coatings have been shown to be effective in enhancing erosion resistance in 9Cr-1Mo type steels. ²³ However, at high temperatures (≥900 °C), a decrease in protective capacity, layer erosion and spallation have been noted. ²⁴ It is shown that the efficiency of the aluminization process is directly dependent on the applied temperature, time and powder composition. The intermetallic phases formed (FeAl, Fe₃Al, Fe₂Al₅, etc.) directly affect the main properties of the layer, such as hardness, oxidation and corrosion resistance. In particular, the dense Al₂O₃ protective layer formed at high temperatures ensures the long-term stability of the material. Therefore, aluminization is still considered one of the most promising methods for creating a reliable protective coating in the automotive industry, energy systems, and metal structures operating in high-temperature environments.

Despite the extensive literature on aluminized steels in high-temperature and corrosive environments, no studies have examined the metallurgical interaction between aluminized steel and modern SAC-type lead-free solder melts. The dissolution behavior, long-term stability, and interfacial reaction mechanisms at the aluminized steel/Sn-based solder interface remain unreported. Therefore, the aim of this work is to investigate the resistance of solid-powder aluminized CK60 steel against prolonged exposure (40 days) to a stationary SAC205 lead-free solder alloy at 593 K, with emphasis on interfacial reactions, oxide stability, and possible formation of Fe–Sn or Al–Sn intermetallic compounds.

Materials and methods

Materials

For the immersion test, CK60 steel was chosen as the main substrate material due to its relatively low cost, widespread industrial use, and easy availability. The chemical composition of CK60 steel is shown in Table 1. The substrates prepared for the experimental tests were brick-shaped, measuring approximately 4 mm × 12 mm × 16 mm. The lead-free SAC205 alloy was selected as the solder material for the soldering tests. The chemical composition of the SAC205 alloy is provided in Table 2.

OPEN IN VIEWER

Table 1.

Chemical composition of the examined CK60 steel (nominal values, wt.% and at.%). ²⁵

Elements	C	Mn	Si	Ni	Cr	Mo	P	S
wt%	0.57 to 0.65	0.6 to 0.9	max 0.4	max 0.4	max 0.4	max 0.1	max 0.03	max 0.035
at%	2.77	0.74	0.39	0.19	0.21	0.03	0.05	0.06

OPEN IN VIEWER

Table 2.

Chemical composition of the SAC205 solder alloy determined by ICP (wt.% and at.%). ²⁶

Elements	Sn	Ag	Cu	Ni	Fe	Bi	Pb	Sb
wt%	Ball.	1.95	0.57	0.0064	0.038	0.004	0.0162	0.0086
at%	Ball.	2.13	1.06	0.013	0.08	0.023	0.0092	0.0083

Solid powder diffusion aluminization

Before starting the first powder aluminization experiment, the surface of the steel substrates was carefully cleaned with alcohol to remove oil and dust residues. The materials used in the study were: silicon carbide (SiC, 177–250 µm, 25 kg/bag, Műhelynet, Hungary), aluminum oxide (Al₂O₃, EKF 90, 150–180 µm, 25 kg/bag, WFDental, Hungary), ferroaluminum (Fe–Al, <45 µm, Stanford Advanced Materials, USA) and ammonium chloride (NH₄Cl, ≥99.5%, PG Chemicals, Slovakia). All materials were used as is, without further processing. FeAl powder was selected in a ratio of 99.5 m/m% and NH₄Cl 0.5 m/m%. NH₄Cl was mixed with FeAl powder as an activator. The addition of NH₄Cl is used to chemically remove oxide layers formed on the surface of FeAl particles and to activate the surface for metallurgical reactions. At high temperatures, NH₄Cl decomposes at 612 K (while the melting point would be 793 K ²⁷ ) by releasing HCl gases, which dissolve the surface oxides to chloride form, thereby cleaning the surface. As a result, FeAl powder adheres better to metal surfaces. 356.644 g of FeAl powder was accurately measured for the experiment. This amount constitutes 99.5 m/m% of the mixture. Accordingly, the required amount of NH₄Cl was calculated to be 1.786 g. The total mass of the mixture was 358.43 g. Next, the FeAl powder and NH₄Cl powder were thoroughly mixed to form a homogeneous mixture. After a compression process, the mixture with the steel specimens was closed with a plate cover. To prevent oxidation, thick layers of Al₂O₃ (aluminum oxide) and SiC (silicon carbide) powders were applied to the top of the plate cover. These layers serve as a passive barrier against the oxidizing environment and were mechanically compacted before sintering. The top of the aluminizing container was finally closed with a steel lid. Figure 1 illustrates the structure of the steel container. The closed container was placed in a furnace heated by three-phase resistance at 1323.15 K (1050 °C) for 6 h. After thermochemical heat treatment, the samples were removed and cleaned for further testing and microstructural analysis.

OPEN IN VIEWER

Figure 1.

Schematic cross-section of the aluminizing container and the different layers.

Dissolution test

Before the dissolution test, some of the aluminized samples were cut in half to allow direct contact of the SAC205 solder with the aluminized region of the specimen (shown in Figure 2). The immersion tests were performed in a specially designed dissolution test simulator (illustrated in Figure 3). In this test, the samples were immersed into SAC205 solder melt for 40 days. The test temperature was maintained at 593 K (320 °C). Two immersion test configurations were examined, each using aluminized CK60 steel specimens. The experiments consisted of (i) an intact aluminized sample and (ii) a half-cut aluminized sample in which the aluminized layer was intentionally exposed through sectioning. Each configuration was tested with its own dedicated specimen (two specimens in total), and both tests showed consistent interfacial behavior and microstructural characteristics, confirming the reproducibility of the results. The specimen was intentionally half-cut to expose a bare steel region adjacent to the intact aluminized coating. This design allowed direct comparison of the interfacial reactions occurring on coated and uncoated surfaces under identical dissolution conditions. The exposed steel area enabled clear observation of Fe dissolution and Fe–Sn intermetallic layer formation, phenomena that cannot be effectively evaluated when the entire surface is aluminized. Additionally, the bare steel region provided enhanced wetting of the molten Sn, functioning as a reliable reference surface for distinguishing the dissolution kinetics, intermetallic growth behavior, and wetting characteristics between aluminized and non-aluminized steel.

OPEN IN VIEWER

Figure 2.

A schematic drawing of the (a) aluminized specimen, (b) half-cut specimen used for immersion tests, showing the reacting interfaces.

OPEN IN VIEWER

Figure 3.

The schematic structure of the dissolution simulator.

Results

Except for one case, all the states of the samples were examined on multiple specimens. The as-aluminized, the intact samples subjected to dissolution tests and the half-cut samples subjected to dissolution tests were examined on three specimens and the same results were obtained. Only in the case of the cracked sample subjected to dissolution test was one specimen used.

Microstructure of the aluminized samples prior to immersion tests

The optical microscope mosaic image presented in Figure 4(a) shows the formation of a uniformly distributed, continuous, and compact aluminized coating on the surface of the CK60 steel sample. This structure indicates that an effective diffusion process has occurred between the FeAl powder and the steel substrate, resulting in significant penetration or deposition of the Al on the substrate surface. Such diffusion activity confirms the high degree of bonding and interfacial interaction. The image presented in Figure 4(b) visually confirms that this coating was formed to a thickness of approximately 400 μm and that it had a consistent, uniform structure. Note that in the centre, the pearlite + cementite structure formed from austenite during cooling from the aluminizing temperature, whereas in the aluminized region, a single phase was formed.

OPEN IN VIEWER

Figure 4.

Optical micrographs of an aluminized CK60 steel specimen: a) mosaic view; b) higher magnification.

Figure 5 presents the XRD patterns obtained from the aluminized steel specimens. Since the penetration depth of the incident X-rays is far smaller than the overall coating thickness (∼390 µm), the recorded diffraction response originates almost exclusively from the aluminized layer. The diffractogram contains peaks associated with the FeAl intermetallic compound (B2-type, ordered BCC) identified on the basis of the PDF 03-065-0985 reference file. The measured lattice parameter of 2.91553 Å, which is in accordance with the Fe–Al phase diagram at elevated temperatures, where the BCC solid-solution region spans a wide range of compositions. ²⁸

OPEN IN VIEWER

Figure 5.

XRD spectrum and formed phases of the aluminized steel samples.

Microstructure of the aluminized samples after immersion tests

Native oxide-covered Fe-Al/solder interface

Figure 6 shows the cross-sectional microstructure and the corresponding EDS line scan analysis recorded from the aluminized steel and solder melt after 40 days of immersion test. In order to accurately correlate the microstructure with the chemical distribution, the coating thickness determined from the EDS profile was marked on the SEM image. The EDS line scan results (Figure 6(b)) show a gradual decrease of Al concentration from approximately 40 at% from the outer surface of the specimen to almost 0 at% at the region of untreated steel. The total thickness of aluminized layer was based on this region, being ∼180 µm. The absence of sharp structural discontinuities across the coating region indicates that the process is diffusion-controlled, i.e., the formation of a continuous diffusion layer rather than sharply delimited intermetallic sub-layers. Such a compositional gradient is typically characteristic of the pack-aluminization process, resulting in a continuously varying Al profile as a result of the inward diffusion of Al into Fe.

OPEN IN VIEWER

Figure 6.

(a) SEM image of the intact aluminized steel/solder interface after 40 days of immersion test; b) EDS line-scan across the aluminized layer into the solder (∼180 µm).

Sn was not detected in the Aluminized region of the specimen, yielding that there was no reaction between the solder melt and the aluminized specimen. The scan of O shows a sharp peak at the specimen/solder interface, which proves that there is an oxide layer on the surface of the specimen. The oxide is most probably Al oxide since the Al concentration is relatively high (∼40 at%) at the specimen's surface and it is known that a native Al-oxide forms rapidly when subjected to air. At this point, it is assumed that this native oxide layer acts as a barrier between the specimen and the solder melt and prevented any reaction between them.

To examine the native-oxide-free specimen/solder interface, some specimens were cut in half after aluminization, but prior to the dissolution tests. Thus, non-aluminized steel/solder and aluminized steel/solder interfaces could be examined after the dissolution tests. Figure 7 shows SEM images of the steel/solder and aluminized steel/solder interfaces after the 40 days immersion test and the corresponding EDS line analysis results. In the uncoated steel region (Figures 7(a) and (b)), the EDS profile shows the formation of a clear reaction zone between the steel and the liquid Sn. A clear intermetallic layer is observed, and the line scan shows a sharp transition from the Fe-rich steel to Fe-Sn intermetallic region, then the Sn-rich solder region. The observed intermetallic phase with medium Fe (∼20–30 at%) and high Sn (∼30–50 at%) values ​​in the range of about 80–150 µm is fully consistent with the stoichiometry of FeSn₂ and confirms the formation of a continuous and relatively thick layer of this phase at the steel/Sn interface. In sharp contrast, the aluminized steel/Sn interface (Figure 7(c)) exhibits a completely different behavior. Here, no Fe–Sn intermetallic layer is detected. EDS line analysis shows that the aluminized (Al-containing) coating remains chemically stable. At its surface, the Fe and Al signals decrease rapidly to zero, while the Sn concentration increase sharply as we move to the Sn-rich solder zone. This proves that the diffusion of Fe from the aluminized steel into the liquid solder is completely suppressed. The Al signal also disappears sharply when moving to the solder region, indicating that no Al–Sn intermetallic products are formed under the given conditions. The O scan does not show a sharp peak with ∼40 at% concentration at the specimen/solder interface. However, the native Al-oxide layer is most likely present at the specimen interface, since there was sufficient time gap (hours) between the sample cutting and the dissolution test. Thus, on the fresh surface of the non-aluminized region Fe-oxides, while Al-oxides on the fresh surface of the aluminized region probably formed. This also explains the different behaviors of the two regions. Iron oxides present on the unprotected steel surface lose their stability and decompose at the immersion test temperature. As a result, the exposed Fe atoms can directly react with the liquid solder. On the contrary, the Al–oxide film formed on the surface of the aluminized layer remains stable throughout the experiment and effectively protects the steel from direct contact with the solder. It is well known in the literature that Al–oxide or other protective barrier films have very low wettability with metal solutions, including liquid Sn.^29,30 Therefore, cracks are present at the solder/natural oxide layer interface.

OPEN IN VIEWER

Figure 7.

SEM images and EDS line scans of the half-cut specimens at the non-aluminized steel/SAC205 and aluminized steel/SAC205 interfaces after 40 days immersion test. (a–b) Non-aluminized steel shows a continuous FeSn₂ intermetallic layer (≈80–150 µm). (c) Aluminized steel exhibits no Fe–Sn reaction, with the stable Al–oxide layer preventing Fe diffusion.

Native oxide-free Fe-Al/solder interface

Figure 8 shows an SEM micrograph after dissolution test of a specimen in which cracks were introduced intentionally during cutting, prior to the dissolution test. The role of the cracks was to reveal if any reaction occurs between aluminized steel and solder melt if the native Al-oxide layer is broken. Note that multiple cracks are present, with different orientations. The different phases were identified with EDS analysis. It can be seen that the molten solder penetrated along the crack. As a result, Fe atoms from the exposed steel substrate reacted with the solder, leading to the formation of multiple FeSn₂ intermetallic regions adjacent to the crack. EDS analysis confirms that regions of the FeAl matrix became surrounded by FeSn₂, indicating that the crack acted as a preferential penetration and reaction pathway, effectively bypassing the otherwise protective aluminized layer. This behavior stands in clear contrast to the fully intact aluminized interface shown in Figure 7(c), where no Fe–Sn intermetallics were detected and the coating remained chemically stable. The comparison demonstrates that although the aluminized layer efficiently inhibits FeSn₂ formation in undamaged regions, any mechanical breach can locally expose the steel substrate and promote rapid Fe–Sn intermetallic growth.

OPEN IN VIEWER

Figure 8.

SEM image of a specimen cracked prior to the dissolution test.

Figure 9 shows the EDS elemental distribution maps for the cracked area shown earlier in Figure 8. The Al map confirms that the crack is entirely within the aluminized region, while the Sn maps show that solder infiltrated this defect and reached Fe-rich areas exposed by the local fracture of the coating. These chemical maps thus verify the mechanism shown in Figure 8, indicating that the damage of the native oxide coating creates a direct pathway for solder penetration and subsequent FeSn₂ formation.

OPEN IN VIEWER

Figure 9.

Element maps of a specimen cracked prior to the dissolution test: a) SEM image; b) Al; c) Fe; d) Sn.

Figure 10 presents a higher-magnification view of the cracked region together with an EDS line scan taken across the defect. The compositional variations along the scan reveal several distinct zones. In the initial segment (0–10 μm), corresponding to the solder, Sn constitutes the dominant element, whereas Fe and Al remain at low levels. Although their contents are small, they are not zero, indicating that both Fe and Al locally dissolved into the molten solder through the crack. This is followed by two intervals (approximately 12–32 μm and 42–57 μm) where Fe, Sn, Al, and O vary simultaneously. In these regions, the Fe and Sn signals increase and decrease in parallel, with an Fe:Sn ratio close to 1:2, confirming the presence of FeSn₂ formed due to solder infiltration into the crack. These findings demonstrate that once the native Al-oxide barrier is locally fractured, the aluminized steel becomes reactive to molten SAC205, enabling FeSn₂ formation. Importantly, such reaction products were not observed either in the intact specimen or in the aluminized region of the half-cut specimen. This further verifies that, in those cases, the continuous Al-oxide layer remained intact and prevented any metallurgical reaction between the coating and the solder. The accompanying rise in O within the reaction zones indicates oxidation of Sn and/or dissolved Al. Between these reaction regions, two segments (approximately 32–42 μm and 57–70 μm) exhibit nearly constant Fe and Al contents in an approximately 1:1 ratio. This composition is characteristic of FeAl, confirming that these parts of the scan correspond to intact regions of the aluminized coating rather than the underlying steel substrate.

OPEN IN VIEWER

Figure 10.

(a) SEM image of the cracked specimen and (b) EDS line scan of Al, Fe, Sn, and O (at.%). The scan identifies the Sn-rich solder (0–10 µm), FeSn₂ in crack-infiltrated regions (∼12–32 µm and 42–57 µm), and intact FeAl coating (∼32–42 µm and 57–70 µm), with O indicating localized Al oxidation, highlighting the compositional variation across the defect.

Thermodynamic analysis

A simplified thermodynamic analysis was performed to interpret the experimental results. This analysis considered the effect of Sn, the main component of the SAC205 alloy, as it has the highest concentration (almost unity activity) and is more reactive than the other two components of the solder alloy, Ag or Cu. ⁴ In addition, the formation of Fe–Al intermetallic phases was also studied.

Discussion

It was found that the bare steel reacted with the SAC205 solution, resulting in the formation of the FeSn₂ phase, which is consistent with the results of other authors. ⁵ This observation is also consistent with the Fe–Sn phase diagram, which shows the presence of FeSn and FeSn₂ phases at the experimental temperatures. ¹ It is worth noting that among the alloying elements in the steel in this study, only Fe forms intermetallic compounds with Sn. ¹ On the other hand, when the Al-coated steel samples were immersed in liquid SAC205, no reaction phases were detected; moreover, no diffusion of Sn, Fe, or Al elements was observed. These results are only valid in the presence of a naturally formed Al₂O₃ oxide layer on the Al-coated steel's surface. These observations indicate that Al oxides do not react with Sn-based solder solutions. Although this study applied a single aluminizing parameter (1050 °C for 6 h), previous works have clearly shown that aluminizing temperature, holding time and aluminium activity together control the thickness, phase composition and cracking behavior of Fe–Al coatings. Thermodynamic and diffusion studies on the Fe–Al system indicate that higher treatment temperatures shift the coating towards high-temperature phases such as FeAl or Fe₃Al, whereas moderate temperatures and higher Al activity favour the formation of Fe₂Al₅ and FeAl₃ near the steel surface.^36,37 In hot-dip aluminizing, where aluminium activity is high, relatively short treatments in the 700–900 °C range typically lead to rapid growth of a brittle Fe₂Al₅ layer. ³⁸ By contrast, powder-pack aluminizing, characterized by lower effective Al activity and slower diffusion, promotes the formation of FeAl when the treatment temperature exceeds ∼1000 °C and the holding time is sufficient to reach the corresponding equilibrium field.^39–41 Thermal stability investigations further demonstrate that FeAl, owing to its cubic B2 lattice, exhibits better mechanical integrity and reduced crack susceptibility compared with Fe₂Al₅, which tends to form tongue-like morphologies and readily cracks under thermal or mechanical loading. ⁴² These parameter-dependent effects are consistent with the FeAl-rich, crack-free coating obtained in the present work at 1050 °C for 6 h and help to rationalize its excellent resistance to molten SAC205 solder under prolonged exposure.

Conclusions

Based on the performed experiments and thermodynamic analysis, the following conclusions can be drawn. Native oxides formed on the surface of aluminized steel act as a barrier and prevent any atomic dissolution reaction between the aluminized steel and Sn-based solder melts. The adhesion is poor between the native oxide-covered aluminized steel and the Sn-based solder melt. These results suggest that native oxide-covered aluminized steels can be suitable candidates for non-wettable soldering components. When native oxide-free aluminized steel is in contact with Sn-based solder melts, equilibrium FeSn_₂ forms_, which involves the dissolution of Fe atoms from the aluminized steel into the forming FeSn₂ phase. Thus, cracking of the native oxide layer on the surface of aluminized steels leads to degradation. No reaction was observed between native oxide-covered or native oxide-free aluminized steel and the alloying elements (namely, Ag and Cu) of Sn-based solder melt.