Microstructural investigation of low-activity and high-activity aluminide coatings fabricated by vapor phase aluminizing on IN792 superalloy

Abstract

To improve the oxidation and hot corrosion resistance of components, parts for gas turbine engines made of nickel-based superalloys are often coated with aluminide coatings. Various methods apply these coatings on superalloys. This research examined the vapor phase aluminizing method. The samples were aluminized through a single-step vapor phase at 1050 °C for a duration of 4 h. This method separates the vapor phase formation chamber from the coating chamber, produces Al_xCl_y vapor in the coating powder input, and transports it to the coating chamber via argon gas to form the aluminide coating. The powder composition was 18 wt% Al, 3 wt% NH₄Cl, and 79 wt% Al₂O₃, and the coating powder inputs considered were 20, 30, 35, and 100 g. After coating, the microstructure and formation of the coating were evaluated using scanning electron microscopy (SEM), energy dispersive spectroscopy (EDS), X-ray diffraction (XRD), and Energy-Dispersive X-ray Spectroscopy (EDS) mapping. The results showed that the thickness of coated samples increased with using more coating powder input from 21.3 μm for 20 g to 68.7 μm for 100 g. The results revealed that powder weights of 20, 30, and 35 g formed low-activity high-temperature (LAHT) aluminide coatings with a double-layer microstructure, with the upper layer free of precipitates, indicating outward diffusion of Ni. The formation of a triple-layer coating for a powder weight of 100 g revealed the presence of precipitates in the upper layer, signifying the inward diffusion of Al. In all coating layers, the matrix phase was β-NiAl, where the outer layer was Al-rich and the interdiffusion zone (IDZ) was Ni-rich.

Introduction

Stainless steels introduced in the early twentieth century provided an initial foundation for high-temperature engineering applications. However, subsequent investigations revealed the insufficiency of these materials under demanding thermal conditions. In response to the growing industrial requirements, the metallurgical field developed superalloys as an alternative to stainless steel^1,2. Superalloys are a class of alloys engineered to preserve mechanical properties at elevated temperatures. Nevertheless, these materials exhibit limited resistance to hot corrosion and oxidation when subjected to the severe operating conditions encountered in turbine environments, characterized by extremely high-temperatures and the presence of salt deposits. Protective coatings typically protect superalloys to mitigate these challenges^3–6.

Among various protective coatings, including MCrAlY overlays, and diffusion coatings, aluminide coatings have been widely adopted. Their extensive use is due to their ability to form a continuous and adherent alumina (α-Al₂O₃) scale upon exposure to elevated temperatures, thereby providing effective oxidation and corrosion resistance in high-temperature environments^7,8. The application of aluminide coatings to the surfaces of turbine components composed of nickel-based superalloys can be achieved by diffusion and overlay techniques^9,10. Various processes can be employed to apply aluminide coatings on nickel-based superalloys, including slurry^11,12, pack cementation¹³, out-of-pack cementation¹⁴, and chemical vapor aluminizing (CVA) techniques^15–17. Furthermore, the optimization of reaction chamber and coating composition can be achieved through the utilization of modeling methodologies^18–21. To comprehend the phase formation within aluminide coatings, the Al-Ni phase diagram serves as a fundamental tool. The formation of Ni_xAl_y intermetallic compounds is primarily governed by two parameters: the relative diffusion rates of Al and base alloy elements and the high-temperature behavior of aluminum halides in the vapor phase reactions that happen during pack cementation, out-of-pack, and chemical vapor aluminizing processes^22–24.

Different aluminizing techniques exhibit varying strengths and limitations in terms of cost, efficiency, and coating performance⁷. Pack cementation, a widely used method, involves placing the sample in direct contact with the coating powder within a chamber, followed by high-temperature treatment in a furnace. This technique is favored for its simplicity and cost-effectiveness; however, it is limited in application, as only surfaces directly exposed to the powder receive coating. Consequently, complex cooling channels and internal surfaces in turbine blades often remain uncoated^25,26. Out-of-pack aluminizing, by contrast, separates the sample from the coating powder. The sample is positioned above the powder chamber and subjected to high-temperature processing under an inert argon atmosphere. This approach offers several advantages, including high coating purity, prevention of powder entrapment, and the ability to coat internal cooling channels. Additionally, it facilitates aluminide layer modification with other elements, enhancing its functional properties^27–30. Chemical vapor deposition (CVD) employs halide-activated Al vapors generated externally by passing HCl gas over Al powder at approximately 300 °C. These vapors react with nickel in the superalloy substrate to form β-NiAl coatings. While this method enables uniform coating on intricate internal surfaces, its complexity and high cost make it less economically viable for certain applications^16,26,31,32. The vapor phase aluminizing method presents a cost-effective alternative to CVD. By separately controlling the temperature of the coating powder (750 °C) and the sample (1050 °C), it minimizes contamination risks associated with halide gas formation. This technique retains the advantages of vapor phase reactions while significantly reducing operational expenses, offering a controlled and optimized deposition process suitable for high-performance applications.

Goward and Boone²³ were the first to classify aluminizing processes into two distinct groups: high-activity low-temperature (HALT) and low-activity high-temperature (LAHT), based on the process temperature and Al activity levels. The LAHT method makes the formation of aluminide coatings with a double-layer microstructure in a single step (LAHT) at approximately 1000 °C, a process driven by the outward diffusion of Ni. The high-activity method deposits coatings with a triple-layer structure, either in a single step of high-activity high temperature (HAHT) at around 1000 °C or in two steps (HALT): the first step involves coating at around 750 °C, and the second step involves heat treatment at 1000 °C, resulting from the inward diffusion of Al at low temperatures and outward diffusion of Ni at higher temperatures^3,24,33.

Aluminide coatings made with LAHT methods have coarser grain morphology and fewer grain boundary diffusion paths than HAHT coatings. Additionally, the outer layer of LAHT coatings suppresses the formation of alloying element precipitates such as chromium, titanium, tungsten, molybdenum, rhenium, and hafnium. These microstructural differences contribute to the coating-enhanced oxidation and hot corrosion resistance³⁴.

The precipitates of the mentioned alloying elements typically develop phases characterized by a topologically close-packed (TCP) structure, resulting in the establishment of the Secondary Reaction Zone (SRZ) beneath the (IDZ). These phases can severely degrade the mechanical properties of the substrate. The addition of Ru to superalloys can prevent the formation of these harmful phases³⁵. TCP phases are typically characterized by ‘basket weave’ sheets that align with the octahedral planes in the FCC matrix^36,37.

At high temperatures, the aluminide coating on superalloys protects with the formation of a uniform, dense, and continuous layer of α-Al₂O₃ on the surface of the component³⁸. It prevents the diffusion of harmful elements such as sulfur and oxygen, thereby preventing corrosion. Coated superalloys are sensitive to sulfur, and its presence causes cracking and spalling in corrosive environments at high-temperatures³⁹. The simple aluminide coating in this case uses Pt⁴⁰, Cr⁴¹, Si^42–44, Co⁴⁵, Zr⁴⁶, Hf⁴⁷, Ce⁴⁸, Y⁴⁹, and other elements to improve resistance to hot corrosion and oxidation⁵⁰.

Traditional CVD, known for its ability to produce high-quality coatings through external chloride formation and impurity removal, comes with significant costs due to the need for specialized equipment. The proposed method is by generating AlCl vapor within a separate chamber and transporting it using argon to the coating chamber in a distinct zone of the same furnace. This approach significantly reduces costs compared to conventional CVD systems. This study aims to produce LAHT and HAHT aluminide coatings using vapor phase method in a single step, and also the coating formation mechanisms have been explained through detailed experimental analysis.

Experimental

Materials and sample preparation

Table 1 displays the chemical composition of the IN792 alloy used as a substrate in this research. The Superalloy was cut into samples with the dimensions of 10 × 10 × 3 mm³ and was ground using SiC papers up to the mesh number of 1200, then cleaned with acetone in an ultrasonic bath and dried.

Table 1.

Chemical composition of the substrate alloy IN792 (wt. %).

Cr	Co	Al	Ti	Ta	W	Mo	Si	Zr	Fe	Ni
12.53	8.92	3.82	3.86	3.70	4.67	2.01	0.2	0.3	0.50	Bal.

The powder mixture included 18 wt% Al powder (commercial grade, 20 μm) as the donor, 3 wt% NH₄Cl powder (laboratory grade) as an activator, and 79 wt% Al₂O₃ powder (commercial grade, 50 μm) as an inert filler. The experiments were conducted using four different coating powder weights: 20, 30, 35, and 100 g. The experimental samples were designated A20, A30, A35, and A100, respectively. The experiments were conducted using four different coating powder weights: 20, 30, 35, and 100 g. These weights were selected systematically to define the process window, optimize the LAHT coating (30 and 35 g), and investigate the transition to a HAHT coating at high aluminum activity (100 g). The experimental samples were designated A20, A30, A35, and A100, respectively. The numerical suffixes indicate the corresponding amounts of coating powder applied. Figure 1 schematically depicts the coating procedure utilized in the aluminizing process. The temperature of the coating chamber was adjusted to 1050 °C, while that of the chamber containing the coating powder was set to 750 °C. The sample was kept at this temperature for four hours to finish the coating process. To maintain a neutral atmosphere and transport the generated Al_xCl_y vapor from the coating powder input to the coating chamber, argon gas was flowed throughout the chamber from the start of the heating operation until the end of the coating period. After finishing the coating process, the sample was cooled under flowing Ar.

Fig. 1.

Schematics of the coating equipment used for vapor phase.

Characterization

Thermodynamic calculations were performed by the HSC Chemistry software. The cross-sectional morphology of the aluminide coating was examined using a scanning electron microscope (SEM, TESCAN VEGA/XMU, Czech Republic). The elemental composition at the cross-section of the aluminide coating was investigated using an EDS. In order to investigate the accumulation of various components in the coating, an EDS mapping was employed. The surface phase compositions of the aluminide coating were detected using an X-ray diffraction spectrometer (Bourevestnik DRON-8, Russia) with Cu-Kα radiation. Scans were performed over a 2θ range of 10° to 80° at a scan rate of 4°/min. XRD results were evaluated using the X’Pert phase identification software (Version 3.0) with the PDF2-2013 and COD-2016 databases.

Result and discussion

Characterization of coatings

Figure 2 shows the cross-sections of coated samples deposited using different coating powder inputs. Table 2 shows the chemical composition of the various sections of the coatings marked in Fig. 2 of the coatings.

Fig. 2.

Cross-section of the coating formed for different amounts of weight in the aluminizing processes: (a) A20, (b) A30, (c) A35, (d) A100.

Table 2.

Chemical composition (wt.%) of different zones formed in the coatings

Weight of Powder mixture (g)	Zone	Al	Ni	Cr	Co	Ti	W	Mo	C
20	A	48.74	42.33	3.56	4.56	0.22	0.27	0.59	0
	B	29.49	32.37	21.69	3.92	2.73	4.20	4.00
	S	10.76	53.46	16.91	6.78	2.82	5.78	3.49	0
30	A	53.33	36.49	5.50	3.23	0.43	0.31	0.41	0
	B	16.21	46.03	18.21	7.29	1.76	5.24	5.23	0
	C	1.32	2.86	3.76	0.29	23.55	20.42	18.77	29.003
	S	10.97	51.39	18.35	7.03	3.69	4.89	3.68	0
35	A	38.68	53.82	3.41	2.72	0.74	0.28	0.35	0
	B	18.56	48.68	17.62	5.87	0.61	4.35	4.31	0
	S	11.02	51.08	19.08	6.93	3.01	5.58	3.30	0
100	A	32.66	46.82	8.16	5.85	0.56	3.25	2.70	0
	B	29.36	57.82	4.05	3.43	2.01	1.93	1.40	0
	C	13.21	39.63	25.08	7.79	3.79	6.48	4.02	0
	D	6.27	50.98	19.36	7.01	4.46	7.99	3.93	0
	S	8.64	59.20	16.78	7.57	2.69	3.17	1.95	0

Figure 2(a) and Table 2 demonstrate the microstructure and chemical composition formed using 20 g of coating powder, respectively. These results revealed that the aluminide coatings consists of two layers. The outer layer has the β-NiAl phase, which is rich in Al, while the IDZ also contains the β-NiAl phase that is rich in Ni. Figure 3 shows that the phases on the coated surface were β-NiAl and AlO₃, as shown by XRD analysis. Figure 4 shows that the coating had a thickness of 15.4 μm for the outer layer, 5.9 μm for the IDZ, and a total thickness of 21.3 μm. This thickness is relatively low for aluminide coatings due to an insufficient Al reservoir for the formation of a protective oxide scale against oxidation and hot corrosion, resulting in a reduced service life of the coating. Furthermore, the outer layer phase contains oxide scale, giving the coating an uneven surface. As a result, this amount of coating powder is insufficient. The prepared powder mixture—containing the designated proportions and weights—was subjected to a stirring process in a Jar mill machine, which operates without balls and is specifically designed for mixing and homogenizing powders.

Fig. 3.

X-ray diffraction pattern of aluminide coatings deposited by vapor phase.

Fig. 4.

Thickness of coating layers at different weights of coating powder.

The microstructure and chemical composition of the surface layer produced using 30 g of coating powder are illustrated in Fig. 2(b) and Table 2, respectively. The coating contains two layers, and the β-NiAl phase forms similarly to the 20 g sample. Figure 3 shows that the phase on the coated surface was β-NiAl, as shown by XRD analyses. Figure 4 shows that the coating has the outer layer of 12.8 μm, the IDZ of 9.3 μm, and a total thickness of 22.1 μm. Although the overall thickness of the coating has not changed significantly, it is now more uniform and does not contain any Al₂O₃ on the surface. Furthermore, the IDZ has achieved a relatively ideal thickness. To enhance the coating thickness and remove Al₂O₃, the amount of coating powder has increased. The anomalously high carbon content measured in Zone C (29.03 wt%) is attributed to the presence of a large, pre-existing carbide precipitate from the IN792 substrate. During the LAHT process, as Ni diffuses outward, less mobile carbide-forming elements (such as Ti, Ta, W) become concentrated in the interdiffusion zone, a phenomenon consistent with the literature.

Figure 2(c) and Table 2 demonstrate the microstructure and chemical composition of the surface layer formed using 35 g of coating powder, respectively. As shown by the XRD analysis in Fig. 3, the surface of the A30 and A35 samples consists of the β-NiAl phase, which corresponds to the dual-layer microstructure observed in Fig. 2(b) and 2(c).

Figure 4 shows that the coating had the outer layer thickness of 20.1 μm, the IDZ thickness of 11.6 μm, and a total thickness of 31.7 μm. The coating had a total thickness of over 30 μm and was homogeneous, with no AlO₃ on the surface. Figure 3 shows that the phase on the coated surface was β-NiAl, as shown by XRD analyses.

The microstructure and chemical composition of the surface layer formed with 100 g of coating powder are shown in Fig. 2(d) and in Table 2, respectively. Figure 3 shows that the primary phases on the coated surface were β-NiAl, and Ni₃Cr₂, XRD confirmed this finding. The coating is composed of three layers. The outer and middle layers consist of the β-NiAl phase rich in Al, while the IDZ contains the β-NiAl phase rich in Ni. Figure 4 shows that the thickness of the outer, middle, and the IDZ of the coating were 41.4, 10.7, and 16.6 μm, respectively with a total thickness of 68.7 μm.

Figure 4 illustrates the direct correlation between increased coating powder weight and overall coating and the IDZ thickness. By increasing the powder weight from 20 to 100 g, the overall thickness of the coating has nearly tripled, owing primarily to the expansion of the middle and the outer layers. This result demonstrates that the content of Al in the coating chamber has a direct effect on the coating thickness. However, the behavior of the outer layer is more complex, and in some circumstances, such as the thickness reduction in the coated sample with 30 g powder compared to 20 g, the overall trend is not observed. This phenomenon is the result of a change in the growth mechanism of the outer layer.

Figure 5 displays the distribution of Ni and Al elements in coated samples. According to the figure, the upper portion of the coating exhibits a higher content of Al compared to Ni. Conversely, the lower portion demonstrates an increase in Ni content and a corresponding decrease in Al content. This phenomenon is attributed to diffusion, whereby Al, present in greater quantities at the coating surface, diffuses into the substrate while Ni, concentrated within the substrate, diffuses towards the coating surface. The IDZ is characterized by the presence of additional elements, such as chromium and cobalt, which can influence the diffusion behavior of Al and Ni. These interactions may cause the formation of intermetallic compounds or solid solutions, leading to lower overall content of both elements in the IDZ compared to the upper and lower layers. Figure 5 illustrates a direct correlation between the quantity of coating powder and the diffusion depth of Al within the coating and substrate. As the coating powder content increases, Al exhibits deeper diffusion into the substrate. This observation is further supported by the EDS mapping, which align with the results obtained from EDS. As a result, an increase in coating powder leads to deeper Al diffusion, causing the IDZ to form deeper within the substrate.

Fig. 5.

EDS mapping of the aluminide coating after aluminizing (a) Ni, (b) Al, (c) combination of Ni and Al.

Figure 6 illustrates the elemental composition of the aluminide coating in the IDZ and the outer layers. The composition of the IDZ in the LAHT aluminide coating exhibits a direct relationship between the weight of the coating powder and the Al content. Specifically, an increase in the mass of the coating powder correspondingly elevates the Al content within the IDZ. As the weight of the coating powder elevates, the content of Al increases, whereas the levels of other elements, including chromium and cobalt, diminish. Despite having a higher overall Al content, the IDZ of the HAHT aluminide coating exhibits a lower Al content within the layer compared to the LAHT the IDZ.

Fig. 6.

Chemical composition of the aluminide coating: (a) IDZ layer, (b) outer layer.

Figure 2 illustrates that the IDZ in HAHT aluminide coatings is thicker than that found in LAHT coatings, although Fig. 6 shows that the Al content in the IDZ of the HAHT aluminide coating is lower than in the LAHT coatings. This observation indicates that Al has diffused more extensively into the substrate, resulting in a reduced Al percentage within the IDZ of the HAHT aluminide coating. In both HAHT and LAHT aluminide coating processes, an increase in the weight of coating powder results in a higher Al content at the outer layer of the coating. In the case of HAHT coatings, the elevated Al content is extensively distributed, leading to the formation of two distinct layers: a middle layer and a top layer.

Formation mechanism

The following equations describe the reactions involved with aluminide formation on the surface. As the temperature increases throughout the powder-containing chamber, a variety of reactions occur. Thermodynamic calculations were performed by the HSC Chemistry software^13,23. To determine the thermodynamic feasibility of Al deposition reactions, the Gibbs free energy is calculated from 0 to 1200 °C. These processes are studied thermodynamically without considering the kinetics of reactions.

The first reaction occurs when NH₄Cl decomposes (Reaction 1). As shown in Fig. 7, the Gibbs free energy of the reaction becomes negative at 370 °C, implying that this is the minimum temperature required for the reaction to start.

1

Fig. 7.

Temperature-dependent Gibbs free energy change for Al deposition on a substrate.

After the formation of HCl gas, other reactions take place to produce Al_xCl_y, notably AlCl (Reaction 2), AlCl₂ (Reaction 3), and AlCl₃ (Reaction 4).

As depicted in Fig. 7, AlCl₃ is identified as the most thermodynamically stable allotrope of Al_xCl_y. At the specified operating temperature of 1050 °C, with the simultaneous presence of Cl ions and Al, AlCl₃ is initially the dominant species. However, in environments with a high concentration of Al in the reaction chamber, AlCl becomes the predominant form of Al_xCl_y. Specifically, once AlCl₃ is formed, it transforms into AlCl upon interaction with excess Al, facilitated by the thermal stability of Al_xCl_y at 1050 °C. Upon entering AlCl in the coating chamber, a reaction occurs in which Al interacts at an atomic level with the substrate surface, producing AlCl₃ gas species under conditions devoid of excess Al. As AlCl₃ is produced, it is removed from the coating chamber by the flowing Ar gas. This mechanism is illustrated in Fig. 8. AlCl is responsible for the majority of Al deposition on the substrate surface²².

Fig. 8.

The mechanism of AlCl formation and precipitation of Al on a substrate.

Figure 7 shows that the Gibbs free energy for all of these processes becomes negative at temperatures over 370 °C.

2

3

4

The formation of the aluminide coating on the substrate proceeds as described in Reaction 9. To elucidate the underlying mechanism, Reactions 5 to 8 are proposed as intermediate steps. The standard Gibbs free energy changes (∆G°) for these reactions were individually evaluated, and the cumulative ∆G° was found to match the value calculated for Reaction 9, thereby confirming the thermodynamic plausibility of the proposed reaction pathway. The gaseous products emerging from the reaction zone include hydrogen, which is subsequently atomized within the coating chamber (Reaction 5). The decomposition of AlCl leads to the generation of Cl₂ and Al (Reaction 6). This Cl₂ subsequently reacts with hydrogen to produce HCl gas (Reaction 7). When HCl gas reacts with AlCl, the resulting products are AlCl₃ and hydrogen gas (Reaction 8). Throughout this series of reactions, hydrogen gas, produced from the decomposition of the NH4​Cl activator, appears to play a crucial role. It is consumed in Reaction (7) to convert Cl₂ into the active HCl gas and is also a product of Reaction (8), suggesting its participation is key to the overall deposition pathway.

5

6

7

8

9

The term ΔG_Al(Ni)​ represents the partial molar Gibbs energy of mixing for Al in the Ni-based substrate at the process temperature of 1050 °C. Figure 7 shows that all reactions are thermodynamically feasible within the temperature range of 370 °C to 1200 °C. In a closed system, such as pack cementation, AlCl reacts with the substrate to produce AlCl₃, which subsequently recirculates back into the reaction chamber, thus maintaining the cycle. Conversely, this study used an open system, such as CVD, where the AlCl₃ gas is expelled from the system.

The mechanism of aluminide coating production is determined by the chemical activity of the Al source in the coating chamber^23,51. As shown in Fig. 9(a), in the samples with 20, 30, and 35 g of coating powder, under LAHT conditions in the coating chamber, the diffusion coefficient of Al is smaller than that of Ni, resulting in outward diffusion of Ni. The coating process began with the outward diffusion of Ni from the substrate and its interaction with Al. The outer layer is unlikely to contain substrate carbides; however, some elements from the substrate may diffuse outward alongside Ni. The area below the outer layer became an interdiffusion zone as a result of Ni loss from the substrate and the intermixing of Al from the outer layer into the substrate. As Ni content decreased and Al content increased in the IDZ, the superalloy matrix phase (γ₊γ′) changed into the β-NiAl phase. According to the limitations of the solubility of the β-NiAl phase compared to the substrate matrix phase, precipitates rich in Co, Mo, Cr, carbides, and sigma phase form, which are usually shaped like columns^23,52.

Fig. 9.

Schematics showing the mechanism of coating formation during the aluminizing process of (a) low-activity type, (b) high-activity type.

As shown in Fig. 9(b), with 100 g of coating powder, the activity of Al in the coating chamber increased, resulting in a higher diffusion coefficient for Al than Ni. This results in Al diffusion to the substrate surface, producing the δ-Ni₂Al₃ or β-NiAl phase in the matrix. In this situation, due to the high-temperature and Al activity, the formation of the coating and the subsequent heat treatment occur simultaneously. The matrix’s solubility decreased as the amount of Al in the substrate increased, causing the dissolved alloying elements in the γ phase (nickel solid solution) to precipitate as small particles. After this step was done, Ni diffused from the substrate and combined with Al from the outer layers to form the middle layer, which had fewer precipitates. As Ni diffused from the substrate and formed the β phase in the middle layer, an interdiffusion zone formed in the nickel-depleted region⁵³.

Conclusions

Aluminide coatings were successfully formed on the IN792 nickel-based superalloy using a single-step vapor phase aluminizing method that is a cost-effective alternative to conventional CVD. The microstructure and formation mechanism were influenced by the weight of the coating powder, transitioning from a double-layer low-activity (LAHT) coating at lower powder weights to a triple-layer (HAHT) coating at 100 g of powder. While this study clarifies the formation mechanisms, it is important to note its limitations. The work is focused on microstructural analysis, and further research is required to evaluate the long-term oxidation and hot corrosion performance of these coatings. Furthermore, the process was demonstrated on a laboratory scale using small samples. Scaling this vapor phase technique for large, complex industrial parts would require significant engineering to ensure uniform temperature and vapor concentration, representing a key challenge for future industrial application.

The main conclusions are summarized as follows:

1. The weight of the coating powder directly influences coating thickness and activity. Increasing powder weight resulted in thicker coatings, with total thickness growing from 21.3 μm (at 20 g) to 68.7 μm (at 100 g). However, at low aluminum activity (20 g powder), the coating quality was limited by the formation of a non-uniform surface and an AlO₃ scale.

2. Increasing aluminum activity in the chamber promotes the inward diffusion of Al into the substrate, leading to a transition from an outward Ni-diffusion dominated (LAHT) mechanism to an inward Al-diffusion dominated (HAHT) mechanism.

3. Thermodynamic analysis shows that the process is initiated by the decomposition of the NH₄Cl activator at temperatures above 370 °C, which generates the HCl gas required for the subsequent aluminum-transport reactions. All key reactions are thermodynamically feasible at the process temperature of 1050 °C.

4. The hydrogen gas generated from the NH₄​Cl activator is integral to the vapor-phase reaction mechanism. Thermodynamic analysis shows it is involved in the regeneration of the HCl activator and the overall reduction of aluminum-chloride species at the substrate surface, facilitating the deposition of Al.

Author contributions

Ali Azari Beni: Validation, Investigation, Visualization, Writing – original draft, Writing – review & editing. Saeed Rastegari: Conceptualization, Methodology, Resources, Supervision, Writing – review & editing.

Data availability

The datasets used and/or analyzed during the current study available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.