Friction Stir Welding: A Large-Scale Scientometric Analysis of Research Impact and Dynamics Over Three Decades

Friction stir welding (FSW), since its 1991 invention by TWI, has become widely used for lightweight alloys, along with techniques like friction stir processing (FSP), friction stir spot welding (FSSW), and friction stir additive manufacturing (FSAM). This work conducts a Scientometric analysis of three decades of FSW research to describe publication growth, scientific structure, productivity, and collaboration trends. Analysis of 19,476 FSW-related Scopus records from 1995 to 2025 was performed using Bibliometrix/Biblioshiny (R 4.3.1) for citation metrics, coauthorship networks, keyword co-occurrence, and thematic mapping. The database includes 1,184 sources, 22,951 authors, and 62,339 references, with a mean of 4.12 authors/article; 3.4% are single-authored, with international coauthorship of 18.15%. FSW publications increased from 2 papers in 1995 to over 1,400 per year in 2020, peaking at 1,564 in 2019. This shows an annual growth rate of 24.03% and 23.38 citations per document. The logistic model matched the trend (R2 of 0.982, RMSE of 72.91), suggesting 30,072 potential publications; 95% focus on microstructure, mechanical properties, and aluminum/magnesium alloys, while expanding into dissimilar joining and FSAM. The 2005 publication “Friction Stir Welding and Processing” remains fundamental, with more than 6,700 citations. China and India lead in output with 3,989 and 3,798 papers, respectively, but have lower MCP ratios than the UK (47.6%) and Saudi Arabia (70.3%). Harbin Institute of Technology leads institutions with 585 articles. The thematic analysis revealed that microstructure–mechanical property relationships and aluminum alloy joining represent the most common and well-established research motor topics, while dissimilar material joining, friction stir additive manufacturing (FSAM), and friction stir processing (FSP) for surface composite fabrication are developing as high-growth niche areas. Keyword trend analysis further highlighted machine learning-assisted process optimization, high-entropy alloys, and robotic FSW systems as rapidly growing themes over the most recent decade, from 2015 to 2025, reflecting the transition of the FSW-based techniques field toward digitalization and advanced material systems.