Advances in the Additive Manufacturing of Superalloys

Abstract

This study presents a bibliometric analysis of the evolution and research trends in the additive manufacturing (AM) of superalloys over the last decade (2015–2025). The review follows a structured methodology based on the PRISMA 2020 protocol, utilizing data from the Scopus and Web of Science (WoS) databases. Particular attention is devoted to the intricate process–structure–property relationships and the specific behavioral trends associated with different superalloy families, namely Ni-based, Co-based, and Fe–Ni-based systems. The findings reveal a substantial growth in scientific output, with the United States and China leading contributions and an increasing trend in international collaboration. Key research areas include process optimization, microstructural evolution and control, mechanical property assessment, and defect minimization. The study highlights the pivotal role of technologies such as laser powder bed fusion, electron beam melting, and directed energy deposition in the fabrication of high-performance components. Additionally, emerging trends point to the integration of machine learning and artificial intelligence for real-time quality monitoring and manufacturing parameter optimization. Despite these advancements, challenges such as anisotropic properties, porosity issues, and process sustainability remain critical for both industrial applications and future academic research in superalloys.

1. Introduction

Superalloys are distinguished by their exceptional mechanical strength, thermal stability, and corrosion resistance, making them indispensable in gas turbine engines for aerospace and energy applications [1]. Their ability to work under temperatures close to 85% of their melting point makes them indispensable in critical components such as turbine blades and combustion chambers in industries such as aeronautics and energy [2,3]. On the other hand, the development of additive manufacturing (AM) has revolutionized their production, making it possible to manufacture complex structures and integrated cooling systems, previously unfeasible with traditional methods. This optimizes weight, thermal efficiency, and durability in different sectors [4,5]. As a result, interest in the AM of superalloys has grown significantly in recent years, both in industry and academia [6].

Superalloys are traditionally classified into three main categories according to their base element, each with distinctive characteristics that determine their specific application: (i) nickel-based, (ii) cobalt-based, and (iii) iron–nickel-based. Nickel-based superalloys, such as the Inconel, Waspaloy, and René families, constitute the most widely used group due to their exceptional combination of high-temperature strength, microstructural stability, and corrosion resistance in oxidizing environments. These alloys typically contain significant proportions of chromium, cobalt, molybdenum, tungsten, aluminum, and titanium elements that contribute to their specific hardening mechanisms and microstructural stability under extreme conditions [7,8,9]. Cobalt-based superalloys, including the Haynes series and HS alloys, offer outstanding resistance to wear and erosion at elevated temperatures and are particularly valuable in applications where abrasion and hot corrosion are primary concerns [10,11]. Finally, iron–nickel-based superalloys, which include variants such as Incoloy, A286, and Nimonic, provide an economic balance between cost and performance, maintaining adequate mechanical properties at moderately high temperatures while offering greater processability than their nickel- or cobalt-based counterparts [12,13].

The outstanding performance of superalloys at elevated temperatures is primarily due to three hardening mechanisms that operate synergistically: (i) solid solution hardening, (ii) carbide dispersion, and (iii) precipitation hardening. Solid solution hardening is achieved by incorporating elements such as chromium, molybdenum, tungsten, and small additions of tantalum, zirconium, niobium, and boron, which distort the crystal lattice, impeding the movement of dislocations and providing stable strength even at elevated temperatures [14,15]. Carbide dispersion hardening represents another critical mechanism, where extremely hard particles such as TiC, BC, ZrC, TaC, Cr₇C₃, Cr₂₃C₆, Mo₆C, and W₆C are strategically distributed in the microstructure, acting as effective barriers to the movement of dislocations and reinforcing especially the grain boundaries, contributing significantly to creep resistance [16,17]. Nevertheless, the most distinctive and crucial mechanism for nickel-based superalloys is precipitation hardening, principally through the prime gamma phase (γ′), with a Ni₃Al or Ni₃Ti composition, which coherently precipitates with the gamma austenitic matrix (γ). This γ′ phase provides a unique combination of toughness and ductility, acting as an obstacle to the movement of dislocations without compromising the overall toughness of the alloy, which is critical for fatigue and creep resistance at elevated temperatures [18]. The controlled interaction between the γ′ matrix and the γ′ precipitates forms a microstructural basis that allows these alloys to maintain their mechanical properties under conditions where most metallic materials would degrade rapidly [19].

The convergence between AM technologies and superalloys represents a particularly promising technical frontier for expanding the application capabilities of these advanced materials. AM processes for superalloys typically use metal powders with specifically optimized compositions and characteristics, including controlled particle size distributions, spherical morphologies to maximize flowability, and precise chemical compositions that ensure the formation of the desired microstructural phases after thermal processing. Predominant technologies include selective laser melting or laser powder bed fusion (SLM/LPBF) [20,21], electron beam melting (EBM) [22,23], direct energy deposition (DED) [24,25], and wire arc additive manufacturing (WAAM) [26,27], each with specific advantages depending on the component geometry, dimensional accuracy requirements, and target mechanical properties. The ability to process superalloys through AM has opened new possibilities for applications previously limited by manufacturing constraints, such as lightweight structures with internal geometries optimized for heat dissipation, monolithic components that previously required assembly of multiple parts, and biomimetically inspired geometries that maximize strength-to-weight ratios [28,29]. However, this field faces significant challenges related to property anisotropy, porosity, residual stresses, and precise microstructural control, all of which require a thorough understanding of the interactions between process parameters, powder composition, and thermo-mechanical transformations during fabrication and post-treatment. Figure 1 presents a schematic representation of most important advanced superalloys, associated hardening mechanisms, and AM techniques.

The rapid evolution and diversification of research in the field of AM superalloys has generated a growing mass of scientific literature spanning multiple disciplines. For this reason, it is necessary to apply a systematic bibliometric analysis to identify emerging trends, quantify the evolution of different lines of research, map international collaborative networks, and assess the relative impact of various technological approaches in this dynamic field. Moreover, particular attention is devoted to the intricate process–structure–property relationships and the specific behavioral trends associated with different superalloy families, namely Ni-based, Co-based, and Fe–Ni-based systems. Thereby, it is possible to obtain a comprehensive view of the current research scene, identifying consolidated areas of knowledge, recognizing gaps that require further attention, and visualizing the geographical and institutional distribution of expertise in this technological domain.

2. Materials and Methods

To conduct a comprehensive bibliometric analysis in superalloys within the context of AM, a structured methodology was employed, consisting of five distinct stages (Figure 2): (i) identification and diagnosis of the relevant scientific literature, (ii) collection and organization of bibliometric data, (iii) data analysis and processing, (iv) visualization and presentation of findings, and (v) synthesis of results and formulation of conclusions. This structured methodology ensured a systematic approach, enabling the identification of key research trends, emerging topics, and potential gaps in the literature.

During stage I, it is necessary to conduct a diagnosis of the scientific publications in the state of the art. The search strategy employed in this study leveraged two of the most authoritative scientific databases: Scopus and Web of Science (WoS). These databases were selected due to their extensive coverage of the peer-reviewed literature, ensuring the retrieval of high-impact and high-quality publications [30]. The search string, indicated in Figure 3, was carefully designed to include the term “superalloy” within the title, abstract, and keywords, ensuring the selection of studies specifically focused on this category of advanced metallic materials. To capture the broad spectrum of research on AM, additional keywords were incorporated, including “additive manufacturing”, “3D printing”, and the abbreviation “AM”. The use of Boolean operators and asterisk symbol increased the search sensitivity, reducing the risk of missing relevant studies as term variations were considered. In this regard, the bibliometric data were collected from 2015 to March 2025, that is, the last ten years.

In stage II, the scientific publications were systematically collected and organized. The study adhered to the PRISMA 2020 guidelines to ensure methodological rigor and transparency in the selection process, following the identification, screening, eligibility, and inclusion stages [31,32], as shown in Figure 4. During the identification phase, a total of 3931 records were retrieved from Scopus and WoS, with Scopus contributing 1945 records and WoS 1986. In this step, data refinement involved removing duplicate entries between both databases, which accounted for 1055 records. Moreover, 6 records were flagged as ineligible by automated filtering tools, reducing the dataset to 2870 documents for further screening. No reports were excluded or removed at this stage, ensuring that all potential sources were carefully evaluated. The eligibility assessment phase involved a detailed review of each study based on predefined inclusion and exclusion criteria. Conference papers (396), proceedings (214), non-English studies (169), book chapters (42), and conference reviews (41) were excluded to maintain a focus on peer-reviewed journal articles. Furthermore, 40 studies that did not meet the inclusion criteria were removed for reasons such as retraction, editor letters, and other exclusions. As a result, 1770 studies were deemed relevant and included in the final bibliometric dataset. This rigorous selection process ensured that only high-quality, peer-reviewed research contributions were considered, reinforcing the reliability and validity of the bibliometric results.

In stage III, bibliometric data, including sources, authors, documents, clustering, and citations, were systematically extracted and structured using Bibliometrix and Biblioshiny from the R-4.4.2 software [33,34,35]. Bibliometrix was instrumental in cleaning and organizing the mentioned data. Meanwhile, Biblioshiny facilitated interactive visualizations, generating network diagrams, tables, maps, and trend analyses to track the field’s evolution.

In stage IV, the processed data were presented through a combination of visual representations by using Origin 2021 and Datawrapper. These tools enabled the generation of network diagrams, bar charts, heat maps, and trend analyses, facilitating a comprehensive exploration of the relationships between key research components. The use of temporal trend graphs further provided insights into the evolution of research interests, showcasing how specific topics have gained prominence over time [36].

In stage V, the findings were synthesized into a coherent narrative and structured format, offering a comprehensive overview of the research in superalloys in the era of AM. This stage highlighted major advancements, knowledge gaps, and emerging directions, facilitating an informed discussion on the field’s development. The bibliometric analysis’s findings form the basis for further investigation, pointing researchers in the direction of unexplored areas and possible multidisciplinary partnerships.

3. Results

3.1. General Document Results

The bibliometric analysis conducted in this study covers the timespan from 2015 to March 2025, as mentioned in the previous section. As shown in Figure 5, the dataset comprises 1770 documents sourced from 329 scientific journals, with an annual growth rate of 22.27%. The records of all these documents are presented in Table S1. The outcomes show an ongoing increase in scientific output, which reflects the growing interest in this area. The average document age of 3.33 years suggests that research in this area is relatively recent and rapidly evolving, with an average citation rate of 27.38 citations per document, indicating a high level of academic impact. In terms of document content, a total of 3852 keywords plus and 3107 author keywords were identified, demonstrating the thematic diversity of the studies analyzed. The author network comprises 4520 researchers, with a significant degree of collaboration—each document has an average of 5.84 co-authors, and 18.19% of collaborations are international, highlighting the global nature of research in this domain. Regarding document types, the vast majority are scientific articles (1731) due to the inclusion/exclusion criteria used during the PRISMA 2020 protocol, with a small number of data papers (3) and early-access articles (36).

Figure 6 illustrates the evolution of research output and citation dynamics over the study period. The production of scientific papers has followed an upward trajectory, starting with 11 documents in 2015 and peaking at 343 publications in 2024. As of March 2025, the number of publications stands at 108, reflecting a consistent trend in research output for the year to date. This significant growth highlights the increasing attention paid to the field of study, which is aligned with other studies related to AM [37,38]. Despite the growing number of publications, the mean total citations per article and per year (MeanTCperArt and MeanTCperYear) exhibit a declining trend. While early publications received high citation rates due to their pioneering nature, the recent surge in articles may have led to a more distributed citation impact. This tendency implies that, as the field develops, the overall citation density is stabilizing, even though the research output is still growing.

3.2. Scientific Journal Results

Table 1. Most influential scientific journals according to Bradford’s law.

Scientific Journal	Ranking	Frequency	Cumulative Frequency	Zone
Materials Science and Engineering A	1	131	131	Zone 1
Additive Manufacturing	2	123	254	Zone 1
Journal of Alloys and Compounds	3	76	330	Zone 1
Materials	4	66	396	Zone 1
Materials & Design	5	57	453	Zone 1
Metals	6	53	506	Zone 1
International Journal of Advanced Manufacturing Technology	7	44	550	Zone 1
Journal of Materials Engineering and Performance	8	44	594	Zone 2
Materials Characterization	9	43	637	Zone 2
Acta Materialia	10	34	671	Zone 2

presents an evaluation of the most influential scientific journals in the field, classified according to Bradford’s Law, which helps to identify core sources of knowledge production. The ranking is based on the frequency of publications per journal, with a cumulative frequency analysis that distinguishes between Zone 1, where the highest concentration of relevant research is published, and Zone 2, which includes publications that help spread knowledge more widely. On the one hand, Materials Science and Engineering A emerges as the most influential journal, accumulating 131 publications. Following closely, Additive Manufacturing ranks second with 123 publications, highlighting its importance in the field of advanced materials and manufacturing processes. The presence of journals such as Journal of Alloys and Compounds, Materials & Design, and Metals in Zone 1 indicates a strong research focus on metallic materials, composites, and alloy engineering. On the other hand, Zone 2 includes journals with a moderate but significant research impact, such as Materials Characterization and Acta Materialia, which specialize in microstructural analysis and material behavior. Their inclusion suggests that structural and mechanical characterization remains a crucial aspect of materials science research.

Figure 7 depicts the cumulative growth of publications in the top five journals from Table 1 over the last decade, illustrating a marked increase in scholarly output. The most noticeable increase in publication output occurred after 2015, reflecting both a larger network of scholars in this field and more general technological advancements. In this regard, Materials Science and Engineering A and Additive Manufacturing have experienced a more rapid growth trajectory from 2019 to the present compared to Journal of Alloys and Compounds, Materials, and Materials & Design, which have followed a more similar upward. The sustained rise across all five journals attests to the multidisciplinary nature of superalloy research, encompassing materials science, mechanical engineering, and computational modeling.

Table 2. Highly impactful scientific journals ranked by number of publications, total citations, h-index, g-index, and m-index.

Scientific Journal	Number of Publications	Total Citations	h-Index	g-Index	m-Index	Publication Year Start
Additive Manufacturing	123	6420	46	77	4.18	2015
Materials Science and Engineering A	131	5177	35	70	3.18	2015
Materials & Design	57	3291	29	57	2.90	2016
Journal of Alloys and Compounds	76	2165	27	44	2.70	2016
Acta Materialia	34	3801	23	34	2.09	2015
International Journal of Advanced Manufacturing Technology	44	1019	20	31	2.00	2016
Materials	66	1070	20	30	2.22	2017
Journal of Materials Science & Technology	25	1121	17	25	1.54	2015
International Journal of Fatigue	26	889	16	26	1.60	2016
Materials Characterization	43	962	16	30	1.60	2016

provides a comparative overview of highly impactful scientific journals by considering the number of publications, total citations, and key citation metrics (h-index, g-index, and m-index). In this regard, the h-index measures productivity and impact by counting publications with at least h citations. The g-index assigns more weight to highly cited papers, ensuring the top g articles accumulate at least g² citations. The m-index normalizes the h-index by the years since the first publication, allowing for fairer comparisons over time. This ranking reveals the prominence of Additive Manufacturing at the top, a testament to the surging research interest in novel fabrication routes and their implications for superalloy performance. Materials Science and Engineering A and the Journal of Alloys and Compounds maintain high impact scores, reflecting their established reputation for disseminating advanced research on alloy design, characterization, and property optimization. Meanwhile, journals such as Materials and Materials & Design show consistently strong metrics, underscoring the role of integrated design strategies and computational modeling in guiding modern superalloy development.

3.3. Author Results

Figure 8 features the most prolific authors in the field by both the absolute number and percentage of scientific articles published. J. Li stands out as the most active contributor, with 65 publications, representing 9.27% of the total output. Y. Li and Y. Zhang follow, each with 43 publications, while other notable contributors, such as Y. Wang and Y. Zhou, maintain strong publication records. The relatively even distribution among the top-ranked authors suggests a collaborative research environment where multiple researchers are significantly shaping advancements in superalloy AM studies.

Table 3. Highly impactful authors ranked by number of publications, total citations, h-index, g-index, and m-index.

Author	Number of Publications	Total Citations	h-Index	g-Index	m-Index	Publication Year Start
M. Kirka	28	1959	19	28	1.72	2015
J. Li	65	989	18	29	2	2017
Y. Li	43	1308	17	36	1.7	2016
Y. Wang	39	925	17	30	1.7	2016
X. Lin	29	963	16	29	1.6	2016
X. Wang	34	1256	16	34	1.6	2016
Y. Zhang	43	781	16	27	1.77	2017
R. Dehoff	16	1809	15	16	1.36	2015
W. Huang	21	1079	15	21	1.5	2016
S. Babu	15	1592	14	15	1.27	2015

ranks highly impactful authors based on their number of publications, total citations, and key citation metrics (h-index, g-index, and m-index). M. Kirka leads in overall influence, with the highest citation count (1959) and an h-index of 19, indicating both productivity and research impact. While J. Li holds the highest number of publications (65), his citation metrics are slightly lower, suggesting a high output but with a relatively lower citation-per-paper ratio. Authors like Y. Li, Y. Wang, and X. Wang exhibit balanced profiles, combining strong publication records with robust citation indices. The m-index values indicate sustained contributions over time, with authors like Y. Zhang (1.77) and M. Kirka (1.72) showing a consistent impact since their first publication.

3.4. Country and Affiliation Results

Figure 9 provides a detailed breakdown of scientific output by corresponding authors’ countries, illustrating both total publication counts and the proportion of single- and multiple-country collaborations. China stands out as the leading contributor, with 521 articles, representing 31.5% of the total publications. The USA follows with 340 articles (20.6%), while India, Germany, and the United Kingdom contribute 98, 82, and 67 publications, respectively. The distribution of single-country and multiple-country publications reveals distinct research collaboration patterns. China and the USA exhibit a strong inclination towards domestic research, with 441 and 304 single-country publications, respectively. Conversely, countries like Canada and France show a higher engagement in international collaborations, as reflected in their relatively larger proportions of multiple-country publications. This pattern suggests that, while leading nations have well-established research infrastructures supporting extensive local studies, other countries increasingly rely on cross-border partnerships to advance their contributions in the area.

Figure 10 highlights the global distribution of scientific research in the field. China (1379) and the USA (964) lead significantly, followed by Germany (229), the UK (224), and India (215), reinforcing their strong research presence. France (114), Sweden (109), Italy (100), and Canada (110) also contribute notably, while Japan (90), South Korea (82), and Singapore (63) reflect Asia’s growing role. Emerging players like Iran (58), Russia (56), and Brazil (20) indicate a widening research landscape. The data underscore the dominance of North America, Europe, and Asia, while regions with lower outputs present opportunities for increased collaboration.

Figure 11 highlights the most cited countries, revealing significant disparities in total citations and average article impact. The USA leads with 14,830 total citations and an average of 42.3 citations per article, indicating both high research output and strong academic influence. China follows with 12,621 citations, but with a lower average of 21.1 citations per article, suggesting a high volume of publications with varying impact. European countries such as Germany (3433 citations, 40.9 average), the UK (2791 citations, 41 average), and France (1732 citations, 35.3 average) demonstrate a strong academic impact, although with fewer total publications. Canada (2415 citations, 36.6 average) and Australia (1232 citations, 33.3 average) also show notable contributions, while Singapore and Sweden stand out for their high average citations despite lower total counts. These trends indicate that North America and Europe continue to dominate in terms of influence, while China’s increasing presence suggests a growing impact in the field.

Figure 12 depicts the cumulative growth of scientific publications by research institutions, illustrating the contributions of key affiliations over time. Oak Ridge National Laboratory (163 documents) and the U.S. Department of Energy (116 documents) lead, reinforcing the USA’s strong institutional support for research in superalloy AM. The Chinese Academy of Sciences (90) and Northwestern Polytechnical University (80) show China’s expanding research presence, driven by state-sponsored initiatives and academic–industry collaborations. The Indian Institute of Technology System (59) reflects India’s growing investment in advanced materials research. The consistent rise in publications from these institutes indicates ongoing support, global cooperation, and the emphasis on advances in AM. The dominance of U.S. institutions, combined with China’s rapid growth and India’s emerging contributions, suggests an evolving global research landscape in this field.

4. Discussion

4.1. High-Impact Scientific Publications Evaluation

Table 4. The top 10 cited scientific articles from 2015 to 2025.

Ref	Authors	Scientific Journal	Total Citations	Total Citations Per Year	Normalized Total Citations
[39]	J. Martin et al.	Nature	2077	230.78	17.71
[40]	N. J. Harrison et al.	Acta Materialia	521	47.36	3.62
[41]	C. Todaro et al.	Nature Communications	513	85.50	11.99
[42]	W. Tucho et al.	Materials Science and Engineering A	474	52.67	4.04
[43]	B. Cheng et al.	Additive Manufacturing	466	46.60	4.22
[44]	R. Dehoff et al.	Materials Science and Technology	425	38.64	2.95
[45]	T. Trosch et al.	Materials Letters	407	40.70	3.68
[46]	N. Raghavan et al.	Acta Materialia	406	40.60	3.67
[22]	E. Chauvet et al.	Acta Materialia	397	49.63	5.92
[20]	D. Li et al.	Journal of Materials Science and Technology	385	35.00	2.67

presents the top 10 most impactful articles in the field, providing statistics on total citations, annual citation rates, and normalized citation impact. By analyzing these articles, we identified five dominant research directions that have significantly influenced the development of AM of superalloys over the past decade: (i) computational alloy design and microstructure prediction; (ii) process parameter optimization; (iii) in situ monitoring and defect mitigation strategies; and (iv) post-processing and mechanical performance enhancement.

The first identified trend pertains to computational alloy design and microstructure prediction. Here, the computational approaches have progressively evolved from simple thermal models to sophisticated multi-physics simulations incorporating fluid dynamics and phase transformation kinetics, enabling researchers to predict complex phenomena like Marangoni convection effects on keyhole stability in superalloys with high vapor pressure elements, such as aluminum and titanium. The groundbreaking work by J. Martin et al. in Nature (2077 citations, 230.78 citations/year) established the foundation for physics-based modeling in superalloy development for AM processes [39]. Their models incorporated thermal gradient calculations across varying scan strategies to predict stress distribution and microstructural evolution with unprecedented accuracy. Building upon this foundation, E. Chauvet et al. in Acta Materialia (397 citations, 49.63 citations/year) advanced the field through crystallographic texture prediction models that optimize γ′ phase distribution in nickel-based systems [22]. Their work demonstrated that scan strategy rotation angles between 45° and 67° significantly altered dendrite growth direction in René N5 superalloy, producing unique spiral microstructural patterns. Customized mechanical properties for high-temperature applications in aerospace components have been made possible by the focused manipulation of crystallographic texture using computational models.

The second identified trend focuses on process parameter optimization. AM process optimization is increasingly complex, with numerous controllable parameters beyond standard ones like laser power and scan speed. Recent advances use normalized diagrams and enthalpy-based models to improve printability assessment and transferability across alloys and AM systems. The comprehensive work by B. Cheng et al. in Additive Manufacturing (466 citations, 46.60 citations/year) established systematic process mapping methodologies for laser powder-bed fusion of Inconel 7185 [43]. Their research quantified the complex interrelationships between volumetric energy density parameters and defect formation mechanisms. As evidenced in the search results, the conventional processing map for defect-free print represented by laser power and scan velocity as the two main axes consists of four distinct regions: keyholing, balling, lack of fusion, and the printable region. Similarly, R. Dehoff et al. in Materials Science and Technology (425 citations, 38.64 citations/year) developed process–structure–property relationships for electron beam melting of nickel-based superalloys [44]. Their work pioneered the concept of site-specific microstructural control through beam modulation strategies, where beam focus and scan speed variations within a single layer could produce alternating columnar and equiaxed grain structures. Moreover, N.J. Harrison et al. in Acta Materialia (521 citations, 47.36 citations/year) revealed anisotropic creep behavior in AM-processed René N5 through advanced electron backscatter diffraction analysis [40]. Their characterization of crystallographic texture in relation to scanning strategies demonstrated that creep deformation mechanisms differ significantly between horizontally and vertically built specimens. Finally, T. Trosch et al. in Materials Letters (407 citations, 40.70 citations/year) introduced sustainable cobalt-free superalloys specifically designed for AM processes [45]. Their work optimized processing atmospheres to control oxidation kinetics of reactive elements like aluminum and titanium in these novel alloy compositions.

The third identified direction focuses on in situ monitoring and defect mitigation strategies. According to the search results, although keyhole mode in power bed fusion is often avoided because of porosity, some studies contend that a stable keyhole mode can improve penetration and layer adherence, thus boosting superalloy performance. The highly cited work by C. Todaro et al. in Nature Communications (513 citations, 85.50 citations/year) introduced revolutionary synchrotron X-ray imaging techniques for the real-time observation of melt pool dynamics in superalloys [41]. Their high temporal resolution imaging revealed previously unseen mechanisms of keyhole formation and collapse that lead to lack-of-fusion defects in nickel-based superalloys. Complementing this work, N. Raghavan et al. in Acta Materialia (406 citations, 40.60 citations/year) developed machine learning models that correlate thermal signatures with microstructural defect formation [46]. Their neural network approach achieved 93% accuracy in predicting keyhole porosity from thermal imaging data, establishing a foundation for closed-loop control systems. As indicated in the search results, porosity prediction has been a major focus of machine learning applications in AM, with studies by using Monte Carlo simulation for optimal process map.

The last identified focus area is the post-processing and optimization of mechanical performance. This trend indicates how post-processing methods, like surface changes and heat treatments, can improve a material’s mechanical qualities. Advanced methodologies aim to refine microstructures, reduce residual stress, and improve performance for high-demand applications. W. Tucho et al. in Materials Science and Engineering A (474 citations, 52.67 citations/year) demonstrated how hybrid hot isostatic pressing combined with customized heat treatment cycles can improve the ultimate tensile strength of AM-processed Hastelloy X by 28% compared to as-built conditions [42]. Their systematic evaluation of post-processing parameters optimized solution treatment temperatures and holding times to achieve ideal carbide precipitation while minimizing grain growth. In this sense, D. Li et al.’s work in the Journal of Materials Science and Technology (385 citations, 35.00 citations/year) showed that laser shock peening can enhance fatigue life through engineered compressive residual stresses in Inconel 625 [20].

Despite the wealth of microstructural data reported across AM techniques, a critical analysis reveals that comparisons are often hindered by inconsistencies in reporting grain size, dendritic arm spacing, or secondary phase formation. For instance, while SLM is known for producing fine cellular–dendritic grains, the influence of scan strategy and alloy composition on grain refinement is not consistently quantified. Similarly, although EBM builds benefit from in situ annealing due to high preheat, the degree of recrystallization and its effect on phase stability are still poorly correlated across studies. These gaps point to a broader need for harmonized microstructural characterization protocols that allow meaningful comparison between techniques and materials.

4.2. Keyword Occurence Evaluation

Figure 13 displays a wide array of keywords associated with research on superalloys in AM. To better understand the focus areas and emerging trends, these keywords can be grouped to understand the primary concerns and research directions in the area.

A dominant theme in the AM of superalloys is the intricate relationship between microstructure (584 occurrences) and mechanical properties (453 occurrences), which collectively govern how components behave under service conditions [47,48,49]. The rapid heating and cooling cycles characteristic of AM can yield unique morphology (55 occurrences)—including columnar grains, cellular dendrites, and finely dispersed precipitates—that significantly influence strength (75 occurrences) and deformation (82 occurrences) behavior [50,51]. These features often evolve dynamically during the build and post-processing stages, making microstructural evolution (67 occurrences) a key area of study for researchers aiming to optimize performance. Moreover, phenomena such as grain growth (76 occurrences) and phase formation can be beneficial or detrimental depending on whether they enhance load-bearing capacity or promote crack initiation sites [52]. As a result, attempts to balance high strength, ductility, and stability throughout a variety of temperatures are made by adjusting laser parameters, scanning techniques, and thermal treatments. To guarantee that superalloy parts fulfill strict specifications for fatigue life [53], creep resistance [54], and overall mechanical robustness [55], it is essential to comprehend and regulate these microstructural characteristics.

Research on superalloy (304 occurrences) in the field of high-temperature materials clearly identifies nickel alloys (213 occurrences) as a top option, with nickel-based (90 occurrences) versions serving as the foundation for demanding industries such as power generation and aerospace [56,57,58]. Within this category, nickel-based superalloys (80 occurrences) families—often broadly referred to as nickel-based superalloys (62 occurrences)—stand out for their ability to maintain mechanical integrity at elevated temperatures while resisting oxidation and corrosion [59]. One of the most extensively studied examples is Inconel 718 (148 occurrences), prized for its robust weldability and relatively straightforward heat-treatment (216 occurrences) protocols, which facilitate widespread adoption in AM [60,61,62]. Nevertheless, numerous other alloys (55 occurrences), including Inconel 625, Hastelloy X, and various cobalt-augmented grades, are also under investigation to address specialized performance needs [63,64,65]. Interestingly, stainless steel (56 occurrences) occasionally appears in comparative studies or in multi-material builds, serving either as a reference baseline or as a support structure material [66,67]. This continuous investigation of both new and pre-existing superalloys highlights a larger movement toward tailoring chemical compositions to AM’s particular heat cycles, which will ultimately increase the variety of industrial uses.

Optimizing AM for superalloys necessitates a deep understanding of how process parameters and thermal histories domine key metallurgical events such as solidification (125 occurrences), precipitation (88 occurrences), and phase transformations. During the build, local temperature (135 occurrences) gradients and cooling rates profoundly influence the solidification path, shaping both the grain structure and potential segregation of alloying elements [68,69]. Post-heat-treatment (216 occurrences)—often involving solution annealing, aging, or even specialized protocols to target specific strengthening phases—further refines microstructural features, including the delta phase (55 occurrences) in Inconel 718 (148 occurrences) [70,71]. Although this δ phase can aid in grain boundary control, excessive amounts can reduce ductility. Here, researchers pay close attention to microstructural evolution (67 occurrences) throughout the entire process, seeking to balance fine-scale precipitation (88 occurrences) of γ′/γ″ phases with the mitigation of defects or brittle constituents [72,73]. Predicting component performance under service conditions requires an understanding of these transitions since considerable grains or less-than-ideal phase distributions can compromise mechanical integrity and reduce part longevity.

A variety of AM methods are employed to fabricate superalloys, with laser (95 occurrences)-based powder bed fusion (often referred to as laser melting (66 occurrences)) standing out as a leading technology for producing complex, near-net-shape parts. In this approach, a high-intensity laser selectively fuses layers of powder (70 occurrences) in a controlled powder bed (69 occurrences) environment, facilitated by advanced 3D printers (92 occurrences) equipped with sophisticated optics and inert gas handling [74,75]. Alternatively, deposition (129) techniques like directed energy deposition allow for the creation or repair of large-scale components by melting laser powders (64 occurrences) as they are fed into the melt pool [24,76]. To reduce defects as porosity or lack of fusion, each of these procedures necessitates the careful calibration of the build parameters—laser power, scanning speed, and layer thickness [21,77,78]. Powder characteristics, including particle shape, size distribution, and chemical purity, are equally critical, as they influence flowability and final part quality [25]. The larger objective of producing strong, high-performance superalloy parts that are financially feasible for large-scale production is supported by this unrelenting quest for parameter optimization.

In general, one of the persistent challenges in AM is the development of residual stress (57 occurrences), thus also affecting superalloys. This is due to the steep thermal gradients and rapid cooling rates that occur during layer-by-layer manufacturing. These stresses might cause warping, cracking, or early failure if they are not adequately controlled, which would compromise the high mechanical qualities for which superalloys are renowned [79]. To address this, researchers employ multiple strategies: preheating the build plate to reduce temperature gradients, adjusting scan patterns to distribute heat more evenly, and applying post-build stress relief or hot isostatic pressing [80,81,82]. The goal is to produce fully functional components (68 occurrences)—from turbine blades to rocket engine parts—capable of withstanding prolonged exposure to high temperatures, mechanical loads and corrosive environments. For this reason, the AM community is increasingly focusing on the integration of material science knowledge with advanced process control, with the aim of producing geometrically complex and reliable superalloy parts under real-world operating conditions.

Beyond listing frequent terms, the keyword analysis reveals a research concentration on certain materials (e.g., Inconel 718) and processes (e.g., SLM), while others like Co-based alloys or WAAM are underrepresented. This imbalance may reflect technological inertia or limited accessibility, but it also points to research gaps in systems with high industrial potential. Addressing this asymmetry could broaden the knowledge base and diversify future innovation paths.

4.3. Keywords Co-Occurrence Analysis

The co-occurrence diagram stated in Figure 14 provides a visual representation of how frequently specific keywords appear together across the literature, effectively mapping the conceptual superalloys AM. Here, each node corresponds to a keyword, and its size often reflects the overall frequency of that keyword: the links or edges between nodes indicate how frequently two keywords co-occur in the same publication, with thicker or denser connections typically signifying stronger thematic relationships.

The red cluster centers on microstructure, mechanical properties, and behavior, highlighting the intricate interplay between metallurgical phenomena and performance outcomes. As previously mentioned, keywords like precipitation, heat treatment, and solidification frequently co-occur with microstructure, reflecting a deep focus on how phase transformations and grain evolution impact creep resistance, fatigue life, and tensile strength. The fact that behavior is so closely linked to mechanical properties underscores how the microstructural state directly influences the alloy’s overall response under stress. Consequently, researchers in this cluster advocate a multi-pronged approach, akin to the tetrahedron of materials engineering, which integrates microstructure, properties, performance, and processing into a single, iterative design cycle. This grouping underlines the need for a multi-pronged approach to AM, such as the tetrahedron of materials engineering that links microstructure, properties, performance, and processing [83].

Meanwhile, the blue cluster turns around both materials and AM techniques, highlighting the dual focus of researchers in this domain. On the one hand, keywords such as Ni-based superalloys, nickel alloys, laser powders, and Inconel 718 indicate a strong emphasis on material selection and composition, which are crucial for optimizing mechanical performance and manufacturability. On the other hand, the presence of terms like 3D printing, powder bed fusion, directed energy deposition, and laser parameters in proximity suggests a parallel effort to refine process conditions, ensuring that feedstock characteristics, scanning strategies, and thermal cycles align to minimize defects and enhance structural integrity. The close interconnection between these themes underscores that alloy selection, process optimization, and performance evaluation are not isolated aspects but rather part of an integrated continuum.

While strong clusters are observed around microstructure and mechanical properties, key concepts like anisotropy, texture, or heat treatment remain weakly connected. This suggests that these aspects are often studied in isolation, rather than in integrated process–structure–property frameworks. Bridging these thematic gaps would foster more holistic approaches to AM superalloy development.

4.4. Thematic Evaluation

Figure 15 illustrates a thematic map that categorizes research topics on two key dimensions: development degree (density) on the vertical axis and relevance degree (centrality) on the horizontal axis.

Motor themes are central and well developed, which means they are well-integrated, fundamental subjects that propel advancement in discipline. In this study, the red cluster in the upper-right quadrant includes Ni-based superalloys, electron beam melting, nickel superalloy, and creep. The prominence of Ni-based superalloys reflects their dominance in AM applications due to their exceptional mechanical strength, corrosion resistance, and thermal stability [84]. The existence of electron beam melting indicates that this AM process is commonly linked to superalloys, most likely because it can process materials at high temperatures with less residual stress [23,85]. Furthermore, as superalloys are frequently used in high-temperature settings where creep resistance is a crucial design consideration, the inclusion of creep highlights the field’s emphasis on long-term mechanical performance.

Basic themes are central but moderately developed, indicating that they form the foundation of research but still have significant expansion. The blue and green clusters in this quadrant encompass terms such as additive manufacturing, laser powder bed fusion, Inconel 718, selective laser melting, microstructure, mechanical properties, laser additive manufacturing, and Inconel 625. These terms collectively represent the core of AM research for superalloys, where a significant emphasis is placed on material–process–property relationships. The extensive use of laser-based AM techniques for creating superalloy components is demonstrated by the existence of laser powder bed fusion and selective laser melting. The inclusion of Inconel 718 and Inconel 625, two of the most studied nickel-based superalloys in AM, further reinforces the central role of material-specific investigations in this domain.

Niche themes are highly developed but less relevant, indicating that they reflect niche or new study fields that might not be extensively incorporated into the larger field just yet. The purple cluster, featuring machine learning, falls into this category. The presence of machine learning as a niche theme suggests that while the application of artificial intelligence and data-driven approaches in AM superalloys is a well-explored topic within certain research communities, it has not yet become a core aspect of mainstream studies. However, its high density implies that significant advancements are being made in this area, focusing on predictive modeling of microstructural evolution, process optimization, and defect detection [29,83,86]. This pattern is consistent with the growing use of Industry 4.0 approaches in AM, where AI-powered analytics are being used to improve quality assurance and manufacturing efficiency.

Emerging or declining themes include topics with low development density and low relevance, meaning they are either emerging fields that require further exploration or declining areas that are losing research interest. The blue cluster in this quadrant includes 3D printing, solidification, and wire arc additive manufacturing. The positioning of 3D printing suggests that, while it is still a common term in AM research, it is becoming less frequently emphasized in specialized studies, possibly due to a shift toward more precise terminology such as laser-based techniques or directed energy deposition. Moreover, wire arc additive manufacturing appears in this quadrant, suggesting that, while it has potential applications for superalloy processing, it remains a relatively underexplored technique due to the lower mechanical properties and control in comparison with laser techniques [87,88,89].

Thematic mapping confirms the maturity of certain topics but also reveals limited development in areas such as residual stress, HIP, and WAAM. Their peripheral position signals that critical challenges in AM metallurgy remain insufficiently explored. Future research could benefit from aligning more closely with these underdeveloped yet high-impact directions.

4.5. Comparative Evaluation of Microstructure, Texture, and Post-Treatment Effects Across AM Techniques

The four common AM processes—selective laser melting or laser powder bed fusion (SLM/LPBF), electron beam melting (EBM), laser/plasma directed energy deposition (DED), and wire arc AM (WAAM)—produce markedly different microstructures and properties in superalloys, owing to their distinct thermal histories (energy input, cooling rates, preheat) and melt pool dynamics. In general, SLM/LPBF yields very fine, rapid-solidification microstructures (columnar/cellular grains ~1–10 µm), and EBM (with high build preheat) produces coarser, more homogeneous grains (often equiaxed or large columnar ~10–100 µm), while DED (laser/plasma) and especially WAAM (arc-based) produce much larger, often heterogeneous grains (columnar dendritic grains > 100–200 µm) with extensive reheated zones. These differences strongly influence porosity, precipitate formation, crystallographic texture, and ultimately mechanical performance, as stated in previous sections.

Table 5. Key microstructural, mechanical, and thermal-treatment characteristics of Ni-based, Co-based, and Fe–Ni superalloys fabricated by SLM, EBM, DED, and WAAM.

Process	Ni-Based Superalloys	Co-Based Superalloys	Fe-Ni-Based Superalloys
SLM/ LPBF	Microstructure: Fine columnar/cellular grains (10–50 µm) along BD; strong dendritic segregation. Properties: Very high as-built UTS (often >800 MPa), low elongation, high hardness, but significant H/V anisotropy. Texture: Strong <001> fiber texture in the build direction (epitaxial growth). Post-HT: Solutionizing + aging dissolves Laves/δ and precipitates γ′/γ″; increases strength, partially homogenizes anisotropy.	Microstructure: Ultrafine (<1 µm cell) hierarchical grains with columnar texture; minor as-built γ′. Properties: Exceptional nano-hardness and tensile strength (>1.0 GPa) with good ductility in as-built. High crack susceptibility unless alloy is specially designed. Texture: Strong <001> fiber texture.Post-HT: Aging precipitates Co3(Al,W) γ′; further strengthens.	Microstructure: Similar to Ni alloys (fine γ dendrites), since Ni is dominant; no δ/γ″ phases present (e.g., in IN718). Properties: As-built strength intermediate; anisotropic. Texture: Columnar FCC grains along BD. Post-HT: Conventional 2-step aging to form Ni3Nb and Ni3Al γ″/γ′. Often requires higher solvus/aging due to Fe.
EBM	Microstructure: Coarser columnar grains (~20–200 µm) with reduced segregation and porosity; banded features by build height. Properties: High ductility; strength slightly lower than SLM; anisotropy typically arises from aligned pores, not texture. Texture: Weaker due to layer annealing; grains may recrystallize. Post-HT: HIP and solutionizing above γ′ solvus produce porosity-free, coarse equiaxed grains, subsequent aging yields uniform γ′.	Microstructure: Coarse, recrystallized grains (equiaxed or large columns) due to high preheat. Little in situ γ′; traces of planar segregates at layer boundaries. Properties: Good ductility, high residual strength after HIP. Texture: Weak, as large grains recrystallize. Post-HT: HIP + aging produces uniform γ′ (Co3Al) precipitates and eliminates defects.	Microstructure: Coarser γ grains than SLM, similar to Ni-EBM. Some planar segregation. Properties: Similar to EBM Ni alloys. Texture: Moderate; large FCC grains. Post-HT: Similar to Ni-based EBM; HIP homogenizes, then ages for γ″/γ′.
DED	Microstructure: Large (~50–200 µm) columnar grains with extensive reheating bands. Dendrites are coarser than SLM, but finer than WAAM. Significant microsegregation (Nb, Mo) persists. Properties: As-built UTS < SLM but > WAAM; ductility moderate. Porosity is generally lower than SLM. Texture: Strong columnar (build direction) with many annealing twins after solution heat. Post-HT: Often only solution + aging needed (similar to welded 718); HIP may be used to heal pores. Microstructure after HT still retains larger grains	Microstructure: Large columnar grains, similar to Ni-DED. Possibly some planar Co–Cr precipitation patterns. Properties: Good strength/ductility balance; anisotropic (build vs. transverse). Texture: Columnar FCC (Co/Ni) grains with <001> tendency. Post-HT: Solution + aging gives γ′ precipitation. HIP can reduce heterogeneity.	Microstructure: Similar to Ni-DED: large γ grains (often >200 µm). Dendritic segregation of Cr/Nb. Properties: As-built moderate strength. Texture: Columnar (FCC). Post-HT: Solution + aging, may require HIP if defects.
WAAM	Microstructure: Coarsest structure: very large (>100 µm) columnar/polygonal γ grains across walls. Pronounced interdendritic segregation (Nb, Mo, Nb) and often oxide inclusions. Properties: Lower strength (<SLM/DED) but high ductility. Very anisotropic—properties vary strongly between build and transverse directions. Fatigue life is generally the lowest. Texture: Strong epitaxial columns along deposition; little recrystallization. Post-HT: Even after solution + aging, grains remain large (no recrystallization after typical HT). HIP is often needed to close remaining pores; mechanical properties may still lack wrought strength.	Microstructure: Similar coarse columnar grains, though Co alloys may form more carbides (W/Ta carbides) and interdendritic phases. Properties: Moderately high ductility; strength limited by grain size. Texture: Strong fiber texture along weld path. Post-HT: Same as above; HIP + aging to induce γ′; grains remain coarse.	Microstructure: As with Ni alloys, extremely coarse grains and segregation. Properties: Lower strength; highest ductility. Texture: Strong columnar. Post-HT: Retains coarse grains; heavy forging/HIP may be needed to approach wrought properties.

summarizes key trends in as-built microstructure, texture, mechanical behavior, and heat-treatment response for Ni-based, Co-based, and Fe–Ni (Ni-Fe) superalloys fabricated by each AM method.

4.5.1. Ni-Based Superalloys

In SLM/LPBF, Ni superalloys (e.g., Inconel 718, Haynes 282, IN625) solidify extremely rapidly, forming elongated columnar grains along the build direction with a fine cellular–dendritic substructure [50,90,91]. The interdendritic regions are enriched in Nb, Mo, etc., and often contain Laves or δ phases in the as-built state [92]. The microstructure is heavily work-hardened by residual stresses, resulting in high strength (often exceeding wrought values) but relatively low ductility and strong anisotropy (horizontal vs. vertical build orientations). For example, as-built SLM IN718 exhibits very fine dendritic cells with interdendritic Laves phases and shows higher UTS but lower elongation than wrought, with tensile anisotropy between build directions [93]. In contrast, EBM of Ni alloys (e.g., IN718) uses a 800–1000 °C bed preheat, which causes in situ annealing: the as-built microstructure is coarser and more heterogeneous by build height, with fewer super-saturated phases and lower residual stress [94]. Porosity in EBM tends to be aligned along the build direction (due to layer-by-layer solidification and keyholing) and texture can be weaker because high thermal gradients are alleviated by preheating. DED (laser/plasma) IN718 or 625 produces intermediate features: large columnar grains that span multiple tracks, significant reheated bands, and partially coarsened precipitates. In one study of L-PBF versus L-DED IN718, L-PBF (SLM) specimens had nearly all grains < 100 µm, whereas L-DED and WAAM specimens had the majority of grains > 200 µm [95]. <The WAAM (arc) of Ni superalloys (e.g., IN625, IN718) yields the coarsest structures, with very large (>50–100 µm) polygonal or columnar grains and pronounced elemental segregation. For example, WAAM–625 produced large polygonal γ grains (~100 µm) with heavy Nb-and-Mo-rich segregation to interdendritic zones. WAAM often produces banded or columnar dendritic structures spanning layers, which can degrade ductility perpendicular to bands [96]. In summary, for Ni-based alloys, SLM provides fine grains and cells (high strength, anisotropic); EBM yields coarser, less stressed grains (good ductility, low residual stress); DED produces intermediate coarseness; and WAAM yields very coarse, columnar grains with extensive segregation.

Mechanically, as-built SLM Ni superalloys typically show the highest yield and tensile strengths but limited elongation, whereas EBM parts are somewhat less strong but often tougher due to reduced defect content. In a recent comparative study, as-printed Inconel 718 achieved only ~50–60% of wrought UTS, whereas solid-solution alloys (IN625) reached ~64–72% of wrought strength [97]; this reflects the fact that precipitation-strengthened alloys like 718 require careful aging to achieve full strength. DED-fabricated IN718 and WAAM generally show lower strength and higher ductility than SLM counterparts, due to larger grains and partial in situ softening. Fatigue and creep performance in Ni alloys follow similar trends: fine-grained SLM parts can have high fatigue limits if defects are minimized, but columnar/segregated WAAM material tends to have lower fatigue endurance unless homogenized by HIP.

Texture studies indicate that SLM Ni alloys often develop strong <001> fiber textures along the build direction, since each layer re-melts a narrow molten pool that solidifies epitaxially [98]. EBM parts (with high preheating) can show weaker, more random textures because partial recrystallization can occur during build [99]. DED and WAAM also produce columnar grains, but the multiple reheats and wide melt pools often produce a mixture of orientations and twin formation, e.g., after solution annealing, L-PBF, and L-DED IN718 showed annealing twins, while WAAM IN718 did not. In all cases, the characteristic epitaxial growth during AM leads to anisotropic properties (e.g., higher strength in the build-plane than along build-height) unless counteracted by heat treatment.

Post-build heat treatment is critical for all Ni-based AM alloys. SLM parts are typically solutionized and aged to dissolve Laves and precipitate γ′/γ″; this increases strength and reduces anisotropy but does not eliminate the fine grain structure. HIP above the γ′ solvus (and subsequent aging) is often used for EBM/DED/WAAM parts to eliminate porosity and homogenize microstructure. For example, HIPping a Co–Ni superalloy EBM build at 1245 °C (above the γ′ solvus) produced a porosity-free, fully recrystallized coarse-grained microstructure, and follow-on aging restored a uniform γ′ dispersion [100]. In Ni alloys, rapid-cool SLM microstructures are highly supersaturated; so, standard IN718-like heat treatments (solution + aging) are effective at precipitating γ′/γ″ and MC carbides [92]. In contrast, the slower-cooling DED/WAAM structures may precipitate some phases in situ; they often require only solution treatment and aging (and possibly forging/HIP) to reach wrought-like performance, since excessive precipitation during build can limit age response.

4.5.2. Co-Based Superalloys

Co–Cr–W/Co–Ni–Cr-type superalloys (e.g., Haynes 25/Haynes 188/CoCr alloy 702, and newer Co–Ni–Al–W alloys) show analogous trends, but with some differences. The SLM of Co-based superalloys yields very fine cellular structures and often very strong <001> textures. For instance, SLM-processed GH5188 (Co–Ni–W superalloy) formed a hierarchical microstructure of elongated grains with each grain containing ultrafine (<1 µm) cells and exhibited a strong <001> fiber texture along the build direction [98]. These fine microstructures show exceptionally high levels of hardness and tensile strength (e.g., >1.0 GPa), even in the as-printed condition. In situ phase formation in Co alloys is limited at high cooling rates; so, as-built Co superalloys generally lack equilibrium γ′–Co₃ (Al,W) precipitates; a post-build aging produces a uniform γ′ dispersion much like in Ni alloys [101].

In the EBM of Co/Ni superalloys (often with high Co content), high preheat similarly produces large, low-stress grains. Recent alloy designs (e.g., an equal-Co–Ni–Cr–Al–Ta–W alloy) have achieved crack-free EBM builds by reducing solidification cracking tendency. Such EBM Co–Ni alloys, after HIP + aging, transform into coarse equiaxed grains with well-formed γ′ particles [102]. In general, EBM Co alloys tend to have lower residual stress and porosity than SLM, but may exhibit less kinetic stabilization of cells or fine substructure during processing. For DED and WAAM with cobalt alloys, the trends are similar: coarser columnar dendrites and grains, with considerable macrosegregation (e.g., of carbide-forming elements) across multi-layer builds. One report of WAAM-fabricated Co–Ni–Cr “Elgiloy” noted large columnar grains and a high density of carbides in the weld metal, reflecting slow cooling [103].

Mechanically, SLM Co-based parts often show excellent combinations of strength and ductility—comparable to or exceeding wrought—thanks to their ultrafine microstructures. For example, as-printed GH5188 outperformed its cast/wrought counterpart in nano-hardness and strength [104,105]. EBM Co/Ni prints (especially after proper heat treatment) can also achieve > 1 GPa UTS with >10% elongation [99]. Fatigue data are sparser, but the same general rule holds: finer SLM microstructures are favorable for fatigue life, whereas large-grained WAAM is tougher but weaker in fatigue. Crystallographic texture in Co-superalloy AM parts is usually strong (SLM) to moderate (EBM), as Co–Ni FCC tends to form coherent <001> solidification textures. Anisotropy in Co alloys is thus significant in as-built parts, but like Ni alloys can be mitigated by recrystallizing heat treatments [106].

The post-treatment of Co alloys follows the same principles as Ni alloys. SLM Co alloys (which are essentially Co–Ni FCC matrices) are solutionized/aged to precipitate Co₃ (Al,W) γ′. EBM/DED/WAAM parts often also benefit from HIP + aging: high-temperature HIP eliminates pores/cracks and produces uniform coarse grains, followed by aging to re-precipitate γ′ [107,108].

4.5.3. Fe–Ni (Fe-Ni) Superalloys

This class includes Ni–Fe–Cr alloys (e.g., Fe–Ni Inconel-family alloys or specialized Fe–Ni–Co rocket alloys) in which Ni (often 40–50%) and Fe are both major elements. Their behavior in AM lies between Ni-based and ferrous alloys. For example, additive Inconel 718 (~50%Ni, ~18%Fe) exhibits a dendritic γ structure like other Ni superalloys. SLM 718 shows fine cellular grains with Laves/δ phases and strong anisotropy [92], whereas EBM 718 (high preheat) shows coarser columns and fewer low-melting eutectics [109,110]. DED/WAAM 718 produce very large columnar grains and often twin-laden γ, as noted above. The mechanical performance of Fe–Ni alloys is similarly anisotropic: untreated AM IN718 often falls well short of wrought strength (~50–60% of wrought UTS in as-built) [97], requiring the classic two-stage aging to precipitate Ni₃Nb (γ″/γ′) and reach design properties. Texture in Fe–Ni alloys follows the same epitaxial columnar pattern as Ni-based FCC. Post-build, Fe–Ni superalloys (like IN718) are typically heat-treated with solutionizing, and then aging to precipitate the γ″/γ′ phases. Welded 718 (and thus WAAM/DED 718) often retains δ (Ni₃Nb) if not fully solutioned [111]; therefore, higher solutionizing temperatures or HIP may be needed to fully eliminate columnar topology and isotropize the microstructure.

In summary, the interplay between processing parameters and material response is central to tailoring superalloy performance through AM. Key variables such as laser power, scanning speed, energy density, layer thickness, and preheat temperature directly influence solidification rates, thermal gradients, and melt pool dynamics. These in turn control microstructural features like grain size and morphology, dendritic arm spacing, porosity distribution, and residual stress. For example, high scanning speeds and low energy density can result in a lack of fusion defects, while excessive power may induce keyholing and the vaporization of volatile elements.

This process–structure relationship ultimately governs mechanical behavior. Fine columnar grains produced by SLM can yield high tensile strength and hardness but also lead to strong anisotropy and limited ductility—traits acceptable for aerospace components where weight and high-temperature strength are prioritized. In contrast, DED or WAAM processes produce coarser grains and more ductile behavior, making them more suitable for large, structural parts in energy or marine applications where dimensional stability and damage tolerance are critical.

Establishing this process–structure–property–application link is essential for intelligent AM process selection. It also guides the design of tailored post-processing treatments (e.g., HIP, solutionizing, aging) to mitigate defects or anisotropic behavior introduced during fabrication. Here, the implementation of design for AM principles has played a pivotal role in advancing the capabilities of superalloy components by enabling innovative geometrical designs—such as lattice structures, conformal cooling channels, and topology-optimized parts—that were previously unattainable through traditional methods. These design strategies are critical not only for performance enhancement, but also for minimizing thermal gradients and residual stress accumulation during the build process.

4.6. Sector-Specific Applications and Industrial Relevance of Am Superalloys

The industrial implementation of superalloys for AM shows clear sectorial patterns that reflect both technological maturity and functional requirements. To map this distribution, specific terms related to major industrial sectors were introduced in the bibliometric query, as summarized in

Table 6. Supplementary industrial sector keywords used in combination with core AM superalloy search filter.

Sector	Query Modification	Percentage
Aerospace	AND (“aerospace*” OR “jet engine” OR “turbine blade” OR “aircraft” OR “space” OR “propulsion”)	17.5
Power Generation	AND (“power plant” OR “gas turbine” OR “thermal” OR “energy” OR “combustor” OR “heat exchanger”)	61.7
Automotive	AND (“automot*” OR “vehicle” OR “car”)	12.1
Biomedical	AND (“biomedical” OR “implant” OR “prosthesis” OR “orthopedic” OR “bone” OR “dental”)	4.3
Marine	AND (“marine” OR “naval” OR “ship” OR “seawater” OR “propeller” OR “submarine”)	4.4

* Term variations and enhancement of the search output.

.

Power generation emerges as the leading domain, accounting for 61.7% of sector-linked publications. This prominence is driven by the widespread use of Ni-based superalloys in turbine components, combustor linings, and thermal shielding structures, where resistance to high-temperature deformation and oxidation is essential. Aerospace follows with 17.5% of the output, leveraging AM to produce weight-optimized components such as turbine blades and nozzles with integrated cooling channels. The automotive sector contributes an estimated 12.1%, focusing on thermally loaded components like turbochargers and exhaust manifolds, where superalloys enable higher efficiency and durability. Meanwhile, the biomedical (4.3%) and marine (4.4%) sectors represent emerging application areas. In biomedical engineering, Co-Cr superalloys processed by AM are increasingly used in orthopedic implants and dental prosthetics due to their superior wear resistance and biocompatibility. In the marine sector, research emphasizes corrosion-resistant components exposed to saline environments, such as propellers and valves. This industrial segmentation not only highlights current adoption trends but also reveals growth opportunities in underrepresented sectors, particularly as AM processes become more cost-efficient and qualification standards evolve.

5. Future Directions

Future advancements in the AM of superalloys must focus on several directions. First, it is important to refine microstructural control to improve mechanical performance. Research should prioritize precise control over phase evolution, grain refinement, and porosity mitigation to optimize properties such as anisotropy, creep resistance, fatigue life, and tensile strength, which are limitations on the actual industrialization. For this purpose, the investigation and development of fitted heat treatments, in situ monitoring techniques, and new alloy compositions will be crucial to balancing performance and manufacturability.

Although AM techniques such as SLM, EBM, DED, and WAAM have enabled new geometries and performance in superalloys, each presents critical limitations. SLM and EBM suffer from residual stresses and anisotropic mechanical behavior due to rapid solidification and directional grain growth. DED often results in lower resolution and higher surface roughness, while WAAM is hindered by large grain sizes and severe segregation, leading to inferior mechanical properties unless extensive post-processing is applied. These constraints must be carefully considered when selecting an AM route for specific superalloy applications and emphasize the need for further innovation in process control, alloy design, and post-treatment strategies.

The integration of artificial intelligence, specifically, machine learning, will play a transformative role in AM superalloys. AI-driven models can predict defects, optimize processing parameters, and improve quality control through real-time monitoring. In this review, there is a lack of AI in this field in comparison to other materials or technologies is testified. Additionally, innovations including computational materials design and digital twins will speed up alloy creation and adaptive process control, allowing for increased consistency and efficiency.

Lastly, future research needs to prioritize sustainability. The environmental impact of AM operations will be reduced with the use of strategies to increase energy efficiency, powder reuse, and material circularity. Investigating different alloy compositions with more recyclable or sustainable components could improve resource efficiency even more without sacrificing performance.

A multidisciplinary approach that integrates materials science, manufacturing engineering, and AI-driven process control will be essential for advancing the search field, safeguarding their continued development for aerospace, energy, and other high-performance applications. The integration of these advancements can be a competitive advantage.

6. Conclusions

This bibliometric analysis on superalloys in AM from 2015 to 2025, based on Web of Science and SCOPUS data and structured according to the PRISMA 2020 protocol, highlights significant research trends and developments. The study confirms the growing scientific output in this domain, with North America and Europe maintaining a dominant influence, while China demonstrates an increasing impact in the field.

The evaluation of high-impact scientific publications identifies four major research directions: (i) computational alloy design and microstructure prediction, (ii) process parameter optimization, (iii) in situ monitoring and defect mitigation strategies, and (iv) post-processing and mechanical performance enhancement. These topics have determined advancements in AM superalloys, focusing on material–process–property relationships and manufacturing reliability. Notably, the limited standardization of microstructural characterization and the inconsistent reporting of texture, porosity, or grain size trends across studies represent a barrier to knowledge consolidation, which future efforts must address.

The optimization of AM processes remains a central challenge in superalloy manufacturing, especially in key technologies such as laser powder bed fusion, electron beam melting, and directed energy deposition. The relationship between processing parameters and final material properties is a critical aspect, as the formation of porosity, residual stress, and mechanical anisotropy directly affect the quality and reliability of manufactured components. The integration of advanced in situ monitoring and thermal control techniques has been an area of particular interest to mitigate these effects and improve repeatability in production. However, limitations specific to each AM method—such as residual stress in SLM, low resolution in DED, and coarse grain size in WAAM—must be better understood to guide process selection and alloy adaptation strategies.

Another relevant finding of this review is the increasing application of artificial intelligence and machine learning in the development of strategies for the control and optimization of superalloy AM processes. However, compared to other technological fields, the use of machine learning remains limited. The implementation of data-driven predictive models will enable the real-time optimization of manufacturing parameters, defect prediction, and increased efficiency in the validation of parts produced using AM. To achieve this potential, future studies should prioritize the creation of open-access datasets and standardized benchmark scenarios for model training and validation.

Future research should prioritize microstructural control to improve mechanical properties, address process optimization challenges in key AM techniques, and integrate artificial intelligence to improve defect prediction and quality control. In addition, the sustainability of AM superalloys must be enhanced through energy-efficient processes, powder reuse strategies, and environmentally friendly material selection. A multidisciplinary approach combining materials science, manufacturing engineering, and AI-driven process control will be essential to the industrial adoption of AM superalloys for high-performance applications. In conclusion, unlocking the full potential of AM superalloys will require coordinated efforts across academia and industry to bridge current knowledge gaps, standardize best practices, and translate laboratory findings into robust, scalable manufacturing solutions.