Modeling the parameters influencing the hardness of the molten zone in submerged arc welding: Focus on manganese-adsorbed boehmite nanoparticles Open Access

Welding is a fundamental technique in engineering and manufacturing, and it enables the creation of continuous, durable structures by joining materials. This process is integral to industries such as construction, automotive manufacturing, and aerospace engineering, where the formation of strong and permanent joints is essential.^1–3 Among the numerous welding methods, submerged arc welding (SAW) distinguishes itself by its high efficiency and adaptability. This technique involves an arc generated between a consumable electrode wire and a workpiece, shielded by a layer of powdered flux.^4,5 This protective environment stabilizes the arc and improves deposition rates, rendering SAW particularly suitable for welding both ferrous and non-ferrous materials of varying thicknesses. The method is widely employed in fabricating critical structures such as pressure vessels and pipelines, where precision and long-term reliability are paramount.^6–8 

Recent advances in welding research have increasingly focused on integrating innovative technologies to optimize efficiency and improve joint quality. Among these, nanotechnology has emerged as a transformative field with the potential to revolutionize welding processes. By making use of the unique physical, chemical, and mechanical properties of materials at the nanoscale, nanotechnology can provide novel opportunities for improving welding performance.^9,10 A large number of studies have demonstrated the significant impact of nanoparticles on welding outcomes. For instance, Choudhary et al. examined the role of flux compositions containing compounds such as CaO, Al₂O₃, TiO₂, and MgO in improving the mechanical properties of low-carbon steel welds.¹¹ Their research revealed that these compositions led to notable improvements in weld hardness and impact strength. Similarly, the incorporation of nanoparticles such as tungsten carbide (WC) and titanium nitride (TiN) has been shown to lead to remarkable improvements in weld microstructure and mechanical attributes, with WC nanoparticles notably increasing impact strength.¹² Other investigations have highlighted the effects of titanium dioxide (TiO₂) nanoparticles on weld geometry and chromium oxide (Cr₂O₃) nanoparticles on hardness in low-carbon steel welds.^13,14 These findings underscore the potential of nanoparticles to refine weld geometry and improve mechanical properties. Titanium dioxide nanoparticles have also been observed to facilitate the formation of acicular ferrite needles, which are associated with improved weld toughness and ductility.¹⁵ Further advancements have been achieved with nano-Al₂O₃-treated wires in SAW, which have been found to significantly enhance weld-metal properties.

These studies have demonstrated improvements in hardness, elongation, and bending performance, demonstrating the versatility of nanoparticle-enhanced welding.¹⁶ Collectively, these results highlight the potentially profound influence of nanoparticles on the microstructural and mechanical characteristics of welds; however, it is essential to recognize that optimal weld quality is determined not only by the type of nanoparticles used but also by the careful control of process parameters. Critical factors such as arc voltage, current, welding speed, and electrode stick-out significantly can influence weld properties by affecting arc stability, heat input, penetration depth, and cooling rates.^17–23 For example, the arc voltage governs the heat distribution, while the current impacts the penetration and weld composition. Similarly, the welding speed controls the cooling rate, which is crucial for controlling microstructural transformations.^24–26 

Despite the significant progress that has been made in understanding the individual effects of nanoparticles and specific welding parameters, the combined influence of these factors remains underexplored. This gap in knowledge presents a unique opportunity for research, particularly in the development of innovative nanoparticle compositions. The current study addresses this gap by introducing manganese-adsorbed boehmite nanoparticles into the SAW process. Boehmite, an aluminum oxide hydroxide, possesses distinct properties that make it an excellent candidate for nanoparticle-based welding enhancement. Adsorbing manganese onto the boehmite surface further enhances its functional capabilities, potentially improving the weld quality. This approach is complemented by a systematic investigation of key process parameters, including the arc voltage, current, electrode stick-out, and welding speed. The primary objective of this research is to evaluate the combined effects of the use of manganese-adsorbed boehmite nanoparticles and the welding parameters on the hardness of the molten zone in SAW. By examining the interplay of these factors, this study seeks to provide valuable insights into the optimization of nanoparticle-enhanced welding techniques.

Nanoparticles are materials that are characterized by their extremely small size, typically ranging between 1 and 100 nm, which gives them unique properties. Their small size results in them having different physical, chemical, and mechanical behaviors from those of bulk materials. These characteristics—such as increased surface area, reactivity, and mechanical strength—make nanoparticles highly valuable in diverse fields such as nanomedicine, energy storage, electronics, and material science. The ability to control and modify these properties at the nanoscale has led to significant advances in various technological applications.

Among the different types of nanoparticles, aluminum oxide (Al₂O₃), known as alumina, stands out due to its remarkable stability and versatile applications. Alumina is widely recognized for its insulating, thermal, and chemical properties, which make it ideal for use in a broad range of industries, from ceramics to electronics. Additionally, alumina can exist in different phases, one of which is boehmite [γ-AlO(OH)]. Boehmite is a hydrated form of aluminum oxide that features hydroxyl groups on its surface. These groups play an important role in facilitating interactions with other materials, allowing boehmite nanoparticles to adsorb various elements.²⁷ The utility of boehmite nanoparticles extends beyond their basic material properties. When subjected to high temperatures, boehmite transforms into α-alumina, a stable form that has excellent thermal and mechanical properties. This phase transition, along with the nanoparticles’ inherent catalytic properties, means that boehmite is particularly useful in high-temperature applications, including welding and automotive manufacturing. Moreover, the synthesis of boehmite nanoparticles is relatively inexpensive, and this has contributed to their widespread adoption in several industrial sectors, such as the oil and automotive industries, where they are used for adsorption purposes and are incorporated into composite materials.^28–30 

In welding, nanoparticles can have a particular role in improving weld quality. The addition of nanoparticles to the weld pool can improve the mechanical properties of the joint by refining the microstructure and improving hardness, toughness, and impact resistance. A novel approach in this field is the adsorption of manganese cations onto the surface of boehmite nanoparticles. Under welding conditions, these modified nanoparticles transform into alumina and manganese oxide, both of which play notable roles in improving the overall performance of the weld. This method offers a low-cost solution to enhancing weld properties without significantly increasing production costs.^31,32

The preparation of manganese-adsorbed boehmite nanoparticles involves a multi-step process. Initially, two distinct solutions are prepared: one containing sodium hydroxide (NaOH) dissolved in distilled water, and the other containing aluminum nitrate [Al(NO₃)₃ · 9H₂O] in a separate quantity of distilled water. These solutions are carefully mixed, and the resulting mixture is then subjected to ultrasonic treatment at 25 °C for several hours, allowing the formation of boehmite nanoparticles. Following this, the mixture is filtered and dried at 220 °C to obtain solid boehmite particles. Next, to introduce manganese onto the boehmite nanoparticles, a potassium permanganate (KMnO₄) solution is prepared. The boehmite nanoparticles are then added to this solution, which is placed in an ultrasonic bath for 15 min to facilitate an even distribution of particles. The solution is heated and stirred for several hours to ensure the successful adsorption of manganese ions onto the boehmite surface. This process, as depicted in Fig. 1, effectively creates manganese-adsorbed boehmite nanoparticles.

The successful adsorption of manganese onto boehmite nanoparticles can be confirmed through microscopic and spectroscopic analysis. An electron microscope image of the nanoparticles reveals their morphology, while energy-dispersive spectroscopy (EDS) analysis can be used to identify the presence of both aluminum and manganese, verifying the adsorption process (Fig. 2). These results demonstrate the effectiveness of this synthesis method for producing nanoparticles with functionalities suitable for welding applications. Incorporating manganese-adsorbed boehmite nanoparticles into the welding process offers a promising approach to significantly improving the mechanical properties of welds, including improved hardness and resistance to deformation. This innovative method enables the production of higher-quality welds without incurring substantial additional costs, thus making it an appealing solution for industries that require very high welding performance, such as the aerospace, automotive, and construction sectors. The introduction of these nanoparticles not only provides a cost-effective enhancement to the welding process but also provides the potential to overcome challenges associated with traditional welding methods.

The present research contributes to the advancement of welding technology by highlighting the transformative potential of nanoparticle-based interventions. Specifically, a combination of boehmite nanoparticles and manganese cations provides a unique mechanism for improving weld quality, thereby offering an effective means of optimizing welding outcomes.

This study employed SAW using a semi-automatic robot, the PARS CAT P2310, in combination with the PARC ARC 1203T power supply. The base material for the welding trials was St37 steel, prepared according to the AWS D1.1 standard. Steel samples with dimensions of 15 × 50 × 150 mm³ were carefully cut using a milling machine to ensure uniformity and consistency across all specimens. Table I presents the chemical composition of St37 steel, detailing the primary elements and their concentrations. This is essential for understanding the behavior of the material during the welding process.

The selection of the filler metal and the flux materials, which play a significant role in the quality of the weld, is also a key consideration. Table II shows the chemical composition of the filler metal, while Table III provides the chemical composition of the flux used during the experiments. These materials are critical in determining the structural integrity and performance of the weld, and knowledge of their detailed compositions can be used to help to explain their interactions with both the base material and the nanoparticles during welding.

The primary goal of this research was to investigate the impacts of various welding parameters on the hardness of the molten zone—an area that plays an important role in determining the mechanical properties of the final weld. Key welding parameters—including the arc voltage, current, electrode stick-out, welding speed, and thickness of the nanoparticle coating applied to the steel samples—were systematically varied throughout the experiments. The choice of nanoparticle coating thicknesses (0.25, 0.50, 0.75, and 1 mm) was informed by a combination of an extensive literature review, preliminary trials, and the novel nature of this study. Given that this research represents the first attempt to explore the incorporation of nanoparticles in submerged arc welding to enhance the mechanical properties and microstructure of welded joints, there were no predefined standards for nanoparticle coating thicknesses. As a result, a range of thicknesses was examined to systematically evaluate their effects on the weld’s performance and identify any threshold values that might optimize the weld’s characteristics. A design of experiments (DOE) methodology was adopted to ensure a robust and statistically sound investigation. This approach facilitated a controlled examination of the influence of nanoparticle coating thicknesses while accounting for other welding parameters. Various combinations of process variables were tested using the DOE methodology, enabling the extraction of reliable data to provide meaningful insights into the optimal nanoparticle thickness for improving the overall weld quality and mechanical performance.

After applying the nanoparticle coatings, welding was performed according to pre-established parameters. The hardness of the molten zone was then measured to evaluate the influence of the welding parameters and the nanoparticle coating thickness on the final weld properties. Table IV presents the design matrix that was used in the experiments, detailing the specific combinations of welding parameters and nanoparticle thicknesses tested. Table V lists the measured hardness values for each experimental condition, providing a clear indication of how the welding parameters and nanoparticle coating thicknesses influenced the mechanical properties of the weld.

Finally, Fig. 3 illustrates the weld-bead geometry produced with a sample having a 1-mm-thick nanoparticle coating. The geometry of the weld is influenced by multiple factors, including the welding parameters and the presence of nanoparticles. The uniformity of the weld bead is a crucial determinant of weld consistency and structural integrity. Additionally, the hardness distribution across the molten zone plays a key role in assessing the weld’s overall performance. The cooling rate and solidification process, which are both influenced by the presence of nanoparticles, contribute to variations in hardness, with different regions of the weld exhibiting distinct hardness levels due to localized thermal effects.

Hardness is a key mechanical property that determines a material’s ability to resist localized plastic deformation, and it plays an important role in evaluating the performance of materials, particularly under mechanical stress. It is commonly used in industry as a non-destructive testing measure because hardness values are often correlated with other mechanical properties, such as tensile strength, impact toughness, and fatigue resistance. In the context of welding, hardness measurements are particularly important for evaluating the quality and performance of the weld metal, as they provide insights into the structural integrity and durability of the joint. Given its importance, in the present study, the hardness of welds produced under varying welding parameters was investigated with a specific focus on the effects of nanoparticle coatings on the weld metal.

The Vickers hardness test was chosen for this investigation due to its precision and widespread use in the evaluation of materials with fine microstructures, such as welds. A force of 62.5 kg was applied with a pyramidal diamond indenter for 10 s to measure the hardness of the weld metal. The central region of the weld, where the fusion zone is most concentrated, was selected as the target area for these hardness measurements. This central region is critical because it influences the overall strength and performance of the welded structure. By evaluating the hardness in this region, we can infer the effects of various welding parameters on the properties of the weld. A quadratic regression model was used to explore the relationship between the process parameters and the resulting hardness in the weld zone.

The p value for this model was found to be less than 0.0001, which indicates that the regression model is highly statistically significant. This means that the relationships between the input parameters and the hardness are not due to random variation. Furthermore, the sum of the square model was considerably larger than the residual value, reinforcing the adequacy of the model in explaining the variability observed in the data. The residual analysis revealed a lack-of-fit value of 0.0828, which is not significant, suggesting that the model adequately fits the observed data.

These results indicate that the model provides a reliable and accurate representation of the relationship between welding parameters and weld hardness. The model also demonstrated low variability, with an average hardness value of 148.37 and a standard deviation of 1.46. This indicates that the experimental data are consistent and that the model is robust in predicting the hardness of the weld. The coefficient of variation for the model was calculated to be 0.9845, reflecting a high level of precision in the measurements. Additionally, several statistical measures—including the R², adjusted R², and predicted R² values—were used to assess the model’s explanatory power and predictive capability. The R² value of 0.9776 indicates that the model explains nearly 98% of the variability in the weld hardness, while the adjusted and predicted R² values of 0.9652 and 0.9386, respectively, further support the model’s ability to predict new data accurately. The adequate precision, which measures the signal-to-noise ratio (SNR), was found to be 36.3452, suggesting that the model has a high SNR and is therefore well suited to making reliable predictions. This reinforces the validity of the regression model as a tool for optimizing the welding process and predicting the hardness of welds under different conditions.

To further evaluate the model’s reliability and the normality of the residuals, several diagnostic plots were generated. A standard probability plot of the studentized residuals was used to check for normality, while a plot of studentized residuals versus predicted values was used to assess constant error. These plots, shown in Figs. 4 and 5, respectively, indicate that the residuals follow a normal distribution, supporting the assumption of normality and confirming the adequacy of the model. Additionally, the expected versus actual plot and the histogram of experimental data, shown in Figs. 6 and 7, respectively, were used to compare the predicted values with the observed measurements. These diagnostic tools confirmed that the model accurately predicted the weld hardness with minimal deviations from the experimental data.

In summary, this study presents a detailed analysis of the relationship between welding parameters and the hardness of welds, using a quadratic regression model to establish predictive relationships. The statistical analysis confirms that the model is both valid and reliable, and it can provide a dependable framework for optimizing welding processes. The diagnostic analyses further support the model’s suitability for predicting the mechanical properties of welds under varying conditions.

Figure 8 presents plots detailing the influence of key welding parameters on the hardness of the fusion zone in the welded metal. The analysis shows that variations in the arc voltage and current significantly affect the fusion-zone hardness. Specifically, increasing the arc voltage leads to a decrease in hardness, a trend that is also observed with higher current values. This is because higher voltage and current increase the heat input into the weld pool, promoting grain growth and ultimately reducing the hardness of the weld metal.

In contrast, increasing the welding speed and increasing the electrode stick-out result in higher hardness in the welded region. These two factors reduce the heat input to the weld pool, which helps to refine the grain structure and improve the mechanical properties of the weld. The relationship between the welding parameters and the heat input is key to understanding these trends. Higher arc voltage and current mean that more heat is provided to the workpiece, leading to larger grain sizes in the molten zone. Larger grains typically decrease the hardness of the material, as they offer less resistance to deformation. Conversely, increasing the welding speed reduces the amount of heat transferred to the weld pool, allowing for the formation of finer grains and increasing hardness. This observation is supported by previous reports, which found that finer microstructures are associated with improved hardness and strength of welded joints.^33–35 

The specific influence of the welding speed on the hardness is particularly notable. As the welding speed is increased from 300 to 500 mm/min, the hardness of the fusion zone increases from ∼140 to around 155 HV (Vickers hardness). This change is attributed to the reduced heat input at higher speeds, which promotes rapid cooling of the weld pool and leads to finer grain structures. The higher hardness values observed at increased welding speeds further emphasize the importance of controlling the heat input to achieve the desired mechanical properties in the final weld.

Figure 9 presents a collection of three-dimensional response surfaces that illustrate the complex interactions between the process input parameters—arc voltage, current, electrode stick-out, welding speed, and nanoparticle layer thickness—and the resulting hardness of the fusion zone. Each plot in this collection explores the combined effects of these input parameters and visually represents their influence on the hardness of the fusion zone.

Figure 9(a) specifically demonstrates the relationship between the current, arc voltage, and fusion-zone hardness. The plotted surface reveals a nonlinear interaction between these parameters, with the hardness varying from 133.24 to 163.78 HV. The contour lines on the base plane further highlight the combined effects of the current and arc voltage on the hardness. The curvature of the surface suggests that specific combinations of these two parameters lead to optimized hardness, and this can serve as an indicator for process optimization to achieve desirable mechanical properties in weld zones.

Figure 9(b) examines the relationship between the nanolayer thickness, the arc voltage, and the hardness of the fusion zone. The hardness here ranges from 133.24 to 163.78 HV, similar to the range seen in Fig. 9(a). The contour lines on the base plane reveal the interaction trends between the nanolayer thickness and the arc voltage. The curvature of the surface indicates a nonlinear correlation, where certain combinations of these parameters result in maximized hardness values. This representation demonstrates the complexity of the interactions and the significant role that both the nanolayer thickness and the arc voltage play in determining the mechanical properties of the fusion zone.

Figure 9(c) provides a detailed illustration of the interaction between the welding speed, electrode stick-out, and the resulting hardness of the fusion zone. The hardness values here range from 133.24 to 163.78 HV, with the highest hardness observed at elevated welding speeds and minimal electrode stick-out, represented by the red and orange regions. In contrast, the lowest hardness values are found at lower welding speeds and higher electrode stick-out values, depicted by blue and green areas. The contour lines offer a two-dimensional perspective of these transitions, showing clear gradients between regions with varying hardness. The experimental red data points on the surface align well with the modeled trends, validating the predictions and illustrating the distribution of hardness across different parameter combinations.

Figure 9(d) illustrates the interaction between the electrode stick-out, the current, and the resulting hardness of the fusion zone. The hardness values, ranging from 133.24 to 163.78 HV, are highest at minimal electrode stick-out and lower current, as indicated by the red and orange gradients. As the electrode stick-out and current increase, the hardness progressively decreases, as indicated by the transition to blue and green regions. The contour lines on the base plane depict these gradient transitions, detailing the influence of each parameter on the hardness distribution. The red data points, derived from experimental results, match the modeled surface, confirming the accuracy of the predicted trends and again validating the model’s reliability.

Finally, Fig. 9(e) explores the relationship between the welding speed, the current, and the resulting hardness of the fusion zone. The hardness values range from 133.24 to 163.78 HV, with higher hardness observed at higher welding speeds and lower current values, as represented by the red and orange regions on the surface plot. Conversely, lower hardness values are associated with slower welding speeds and higher current values, as depicted by the blue and green areas. The contour lines delineate these transitions, highlighting the sensitivity of hardness to changes in welding parameters. The red experimental data points again align with the modeled surface, further validating the accuracy of the model across various parameter combinations.

Figure 10 presents a scanning electron microscope (SEM) image coupled with an EDS spectrum, illustrating the microstructural changes in the weld metal influenced by the incorporation of manganese-adsorbing boehmite nanoparticles. The particles labeled “A” and “B” in the image correspond to the manganese oxide formed during the welding process as a result of the thermal reaction with boehmite nanoparticles. These oxides act as nucleating points, facilitating the refinement of the grain structure and consequently increasing the hardness and toughness of the weld metal. The accompanying EDS spectrum further corroborates the presence of key elements such as aluminum (Al) and manganese (Mn), which are integral to the nanoparticle-induced modifications in the molten zone.

The observed microstructure, which is characterized by fine, uniform grains and minimal cracking, suggests that the nanoparticles significantly improve the solidification process, preventing excessive grain growth and contributing to enhanced mechanical properties in the weld zone. Figure 11 presents optical microscopy images of the microstructure of the weld metal under two distinct conditions: with and without the addition of a 1-mm-thick layer of boehmite nanoparticles. Sample 13, which lacks nanoparticles, exhibits a coarse ferrite structure, and this is attributed to the high heat input during welding and the absence of growth-limiting agents. This coarse structure results in lower hardness and reduced resistance to deformation. In contrast, sample 19, which incorporates a layer of boehmite nanoparticles, displays a refined microstructure with acicular ferrite and smaller grain sizes. The transformation of the boehmite nanoparticles into aluminum and manganese oxides during the welding process accelerates nucleation, preventing excessive grain coarsening and leading to a finer, more homogeneous microstructure. This structural refinement results in significantly improved hardness and toughness, improving the overall mechanical performance of the weld. These changes have the potential to be instrumental in improving weld quality in high-demand industrial applications, such as in the aerospace and energy sectors, where superior weld strength and durability are essential.

SAW has long been recognized as a highly effective process for joining thick materials, offering remarkable efficiency and consistency. This research sought to enhance the performance of SAW by examining the combined influence of the process parameters and the addition of manganese-adsorbed boehmite nanoparticles on the final mechanical properties of the weld metal. The results shed light on how the interplay between advanced materials and tailored welding conditions can increase weld quality. In particular, it was found that the welding speed and electrode stick-out have pivotal roles in achieving improved hardness, with higher values leading to refined grain structures and superior mechanical strength. Conversely, excessive heat input, caused by increased arc voltage and current, was observed to reduce hardness by encouraging grain coarsening.

A key contribution of this study lies in the application of manganese-adsorbed boehmite nanoparticles, which undergo decomposition under the thermal influence of the welding arc, yielding aluminum oxide and manganese oxide. This process was found to lead to not only a significant increase in hardness—achieving a peak value of 152.15 HV in samples incorporating a 1-mm-thick nanoparticle layer—but also a notable refinement of the grain structure. Detailed microstructural analysis confirmed these outcomes, revealing a transition from coarse ferrite to fine acicular ferrite. This structural evolution, which was accompanied by a 30% reduction in average grain size, demonstrates the ability of nanoparticles to enhance both the toughness and overall mechanical performance of the weld. The use of functionalized nanoparticles in the SAW framework represents a transformative step in welding science. By combining precise parameter control with the unique attributes of manganese-adsorbed boehmite nanoparticles, this study presents a robust approach to improving weld quality. The findings enrich the understanding of nanoparticle-enhanced welding processes and set the stage for future explorations into the long-term reliability and scalability of this innovative technique, opening new frontiers for industrial applications.

The authors would like to thank Razi University for their assistance throughout the research. The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.

The authors have no conflicts to disclose.

All authors contributed to the study’s conception and design. Manuscript preparation, data collection, and data analysis were performed by F.R. A.S. performed the experiments and collected the data. M.A. designed and supervised the whole project and revised and analyzed data, and F.K. analyzed and revised the manuscript.

Farhad Rahmati: Formal analysis (equal); Methodology (equal); Writing – original draft (equal); Writing – review & editing (equal). Ali Shafipour: Data curation (equal); Investigation (equal). Masood Aghakhani: Conceptualization (equal); Data curation (equal); Formal analysis (equal); Project administration (equal); Validation (equal). Farhad Kolahan: Project administration (equal); Supervision (equal).

The data that support the findings of this study are available within the article.