HEA Alloy Design ML Patent Landscape 2026
The machine-learning-driven high-entropy alloy design space is in a clear growth phase, with filings expanding roughly 90% over the recent window and Chinese academic institutions holding the dominant positions. Activity is concentrated in China but fragmented across many applicants, leaving meaningful white space in processing and application branches adjacent to the core computational chemistry cluster.
Chinese universities lead a still-fragmented field
Shanghai University holds the top position among ranked applicants, followed by Zhengzhou University and Zhejiang Weixiang Materials Technology Co., Ltd., each with 5 patent families, and a cluster of institutions at 4 patent families including the University of Science and Technology Beijing, Central South University, Harbin University of Science and Technology, Seoul National University R&DB Foundation, and Lehigh University.
The top five filers account for only 16% of the combined total of the hundred largest filers, indicating a highly fragmented competitive structure with no single organization holding a commanding share. The tier gap between the leader and the second tier is narrow — one or two families — so rankings are fluid and could shift quickly with a single filing burst.
| # | Applicant | Patent families | Share |
|---|---|---|---|
| 1 | Shanghai University | 7 | |
| 2 | Zhengzhou University | 5 | |
| 3 | Zhejiang Weixiang Materials Technology Co., Ltd. | 5 | |
| 4 | UNIV OF SCI & TECH BEIJING | 4 | |
| 5 | Central South University | 4 | |
| 6 | HARBIN UNIV OF SCI & TECH | 4 | |
| 7 | SEOUL NATIONAL UNIVERSITY R&DB FOUNDATION | 4 | |
| 8 | Lehigh University | 4 | |
| 9 | Hamilton Sundstrand Corporation | 3 | |
| 10 | Hunan University | 3 |
| # | Applicant | Patent families | Share |
|---|---|---|---|
| 11 | Beijing Institute of Technology | 3 | |
| 12 | Northwestern Polytechnical University | 3 | |
| 13 | City University of Hong Kong | 3 | |
| 14 | Guizhou University | 3 | |
| 15 | POSTECH Academy-Industry Foundation | 3 | |
| 16 | Dalian University of Technology | 3 | |
| 17 | Northeastern University (China) | 2 | |
| 18 | Soochow University | 2 | |
| 19 | Indian Institute of Technology Kharagpur | 2 | |
| 20 | Xi’an Jiaotong University | 2 |
The dominance of universities across the top tier signals that IP in this field is still largely being generated by academic research groups rather than industrial players, with Hamilton Sundstrand Corp being a notable exception as the highest-ranked industrial assignee. This creates an acquisition and licensing opportunity for manufacturers seeking foundational ML-for-HEA patents.
The most recent 18–24 months of data are understated due to patent publication lag; 2025 and 2026 figures should be read as floor estimates, not final counts. Longer-window growth, applicant concentration, and technology-route coverage are therefore more reliable signals than the latest-year bar alone.
Filings are accelerating and the technology mix spans alloy composition to AI infrastructure
The annual trend chart illustrates the field’s growth trajectory from 2017 onward, while the technology composition chart reveals how activity is distributed across alloy, computational, and processing branches.
Annual filing trend
Activity was minimal through 2018, then stepped up from 2019 onward, with a pronounced acceleration from 2022 to 2024. The 2025 and 2026 bars are understated due to publication lag and should not be interpreted as a plateau or decline — they represent incomplete counts for those years.
↗ Hover for values · click a bar to ask EurekaTechnology composition
C22C (Alloys) is the dominant branch, reflecting the core alloy-composition claims, followed by G16C (Computational chemistry) and the paired G06F/G06N digital-processing and AI-models branches. B22F (Powder metallurgy) and C23C (Coating and surface deposition) represent the largest secondary clusters, confirming that ML is being applied not just to composition prediction but also to processing routes. Downstream application branches such as B33Y (Additive manufacturing), B22D (Metal casting), and A61F (Implants and prostheses) each carry very low counts, signaling underserved territory.
↗ Hover for values · click a bar to ask EurekaHighly cited patent families surfaced by the query
Citation-heavy patent families returned by the query. Use this section as citation context, not as a curated list of the most topic-specific patents.
Efficient High-Entropy Alloys Design Method Includ…
Embodiments relate to a system for predicting thermodynamic phase of a material. The system includes a phase diagram image scanning processing module configured to scan a binary phase diagram for each material to be used as a component of a high-entropy alloy (HEA). The system includes a feature computation processing module configured to generate a primary… (excerpt from the patent abstract)
| # | Patent | Citations |
|---|---|---|
| 1 | 一种基于机器学习的高熵合金硬度预测方法 | 50 |
| 2 | High-entropy multielement alloy matrixcoating comp… | 41 |
| 3 | 一种CrFeCoNiNbx高熵合金及其制备方法 | 39 |
| 4 | High-entropy alloy foam and manufacturing method f… | 37 |
| 5 | High entropy alloy doped with boron and method for… | 24 |
| 6 | Efficient High-Entropy Alloys Design Method Includ… | 22 |
| 7 | 基于扩散多元节技术的沉淀强化高熵合金成分设计和制备方法 | 21 |
| 8 | 基于扩散多元节技术的沉淀强化高熵合金成分设计和制备方法 | 20 |
Ranked by total forward citations. Citation counts favour older and broadly cited patent families, and broad or adjacent patents may appear when they match the search scope. Treat this section as citation context, not as a curated list of the most topic-specific patents. Some patent titles may be shown in their original, non-English language where an accurate translation could not be guaranteed.
What the competitive structure means for R&D investment
Four structural factors shape the risk and opportunity profile for teams entering or expanding in ML-driven HEA design: the field’s lifecycle stage, its fragmentation, the collaboration patterns visible in the data, and the geographic concentration of filings.
Growth stage — annual filings still rising
The lifecycle evidence classifies this field as Growth, with annual filings still rising and the most recent years understated by publication lag. A 90% growth rate over the recent window confirms that foundational IP is still being staked. Teams entering now can still establish meaningful positions, but the window for low-competition claim space is narrowing as academic output accelerates.
Lifecycle: GrowthFragmented top tier — no entrenched incumbent
With the top five filers holding only 16% of the combined total of the hundred largest filers, no single entity has locked up the space. The leading applicant, Shanghai University, holds 7 patent families — a margin easily bridged by a focused filing campaign. Industrial entrants with engineering resources can realistically compete with the current academic leaders in a short time horizon.
HHI: LowUniversity–industry and cross-institutional pairs are emerging
The most active co-filing pair is Shanghai University and Zhejiang Weixiang Materials Technology Co., Ltd., with 2 joint filings. Central South University has co-filed with AECC Beijing Institute of Aeronautical Materials, signaling aerospace-sector engagement. In the Korean ecosystem, Seoul National University R&DB Foundation has co-filed with both the Korea Institute of Science and Technology and the University of Ulsan Industry-Academic Cooperation Foundation, suggesting a coordinated national research effort. These collaborations highlight access points for licensing or consortium entry.
Ecosystem: FormingChina-dominant filing base; US and EPO underweighted
China accounts for the overwhelming majority of filing activity, with the United States and Europe (EPO) each carrying substantially fewer records. India also appears with a notable presence. The thin US and EPO coverage relative to China means that many Chinese-origin inventions may not yet be protected in key commercial markets, creating freedom-to-operate opportunities for Western manufacturers — but also risk if Chinese filers begin pursuing international prosecution more aggressively.
Lead office: ChinaGo beyond the landscape: Eureka’s TRIZ Solution agent breaks down an R&D problem and returns patented concept solutions, each with a technical approach and cited patent & literature evidence.
| Applicant | Collaborator | Co-filings |
|---|---|---|
| Shanghai University | Zhejiang Weixiang Materials Technology Co., Ltd. | 2 |
| Central South University | AECC Beijing Institute of Aeronautical Materials | 1 |
| Seoul National University R&DB Foundation | Korea Institute of Science and Technology | 1 |
| Seoul National University R&DB Foundation | University of Ulsan Industry-Academic Cooperation Foundation | 1 |
Co-filing pairs, ranked by the number of jointly-filed patent families.
Shanghai University leads on computational chemistry; Lehigh University anchors Western alloy-composition work
The top tier is dominated by Chinese academic institutions with a strong computational focus, while Lehigh University stands out as the leading Western research university and Hamilton Sundstrand Corp is the only industrial player in the upper ranks.
Shanghai University
Shanghai University leads with 7 patent families and is classified as a new entrant in momentum terms, with 4 recent families — indicating that its position is built on recent activity rather than a long legacy. Its technology emphasis falls squarely on G16C Computational chemistry and C22C Alloys, reflecting a strategy centered on ML-driven property prediction and composition optimization. Its active co-filing relationship with Zhejiang Weixiang Materials Technology Co., Ltd. suggests a commercialization pathway beyond pure academic output.
families: 7Lehigh University
Lehigh University holds 4 patent families and is also classified as a new entrant in momentum terms, with 4 recent families, meaning its entire current portfolio has been filed in the recent window. Its technology focus spans C22C Alloys and B22F Powder metallurgy, distinguishing it from the computationally oriented Chinese leaders and positioning it at the synthesis and processing end of the ML-for-HEA pipeline — a differentiated niche with relatively few competing filers.
families: 4| Applicant | Recent (3 yrs) | Trend |
|---|---|---|
| Shanghai University | 4 | ▲ new entrant |
| Northwestern Polytechnical University | 1 | ▲ new entrant |
| Zhejiang Weixiang Materials Technology Co., Ltd. | 4 | ▲ new entrant |
| Lehigh University | 4 | ▲ new entrant |
| Harbin University of Science and Technology | 3 | ▲ new entrant |
| University of Science and Technology Beijing | 4 | ▲ new entrant |
| Central South University | 3 | ▲ new entrant |
Under-served branches where ML-for-HEA coverage is thin relative to technical potential
Several processing and application branches sit adjacent to the dominant alloy-composition and computational-chemistry core but carry disproportionately low patent counts. These observations of relative sparsity are worth evaluating against technical feasibility and commercial entry paths.
B33Y · Additive manufacturing (3D printing)
B33Y holds only 9 records in a corpus of 135 patent families, representing roughly 3% of the technology composition. ML-guided process-parameter optimization for HEA additive manufacturing — where the compositional and thermal space is vast and experimental iteration is expensive — is a plausible technical application of the same predictive models used for composition design. The low count suggests that the intersection of ML, HEA, and AM process control is not yet well staked. Teams with existing AM expertise and ML tooling could extend their HEA composition models into process-parameter claims with incremental investment.
Search this in Eureka →C23C · Coating and surface deposition
C23C carries 15 records, representing roughly 5% of composition. HEA coatings are an established application area for wear, corrosion, and high-temperature resistance, yet the integration of ML for coating composition and deposition-parameter optimization remains sparsely protected. Given the clear industrial demand from aerospace and tooling sectors — and the presence of Hamilton Sundstrand Corp in the broader corpus — this branch has a plausible commercial pull. Entry would involve coupling existing ML property-prediction frameworks to coating-specific performance targets such as hardness gradients and oxidation resistance.
Search this in Eureka →How leading applicants differ by technology route
Route coverage across the main technology branches in the current evidence set.
| Player | C22C 30 · Alloys | C22C 1 · Alloys | G16C 20 · Computational chemistry | G16C 60 · Computational chemistry | G06N 3 · Computing based on AI models |
|---|---|---|---|---|---|
| Northwestern Polytechnical University | Strong · 3 | Strong · 3 | Absent | Strong · 2 | Strong · 2 |
| Zhejiang Weixiang Materials Technology Co., Ltd. | Absent | Absent | Strong · 3 | Strong · 4 | Strong · 3 |
| Shanghai University | Absent | Absent | Strong · 4 | Strong · 3 | Moderate · 2 |
| Harbin University of Science and Technology | Absent | Absent | Moderate · 2 | Strong · 4 | Strong · 3 |
| Lehigh University | Strong · 4 | Strong · 4 | Absent | Absent | Absent |
| Zhengzhou University | Strong · 4 | Strong · 4 | Absent | Absent | Absent |
| Seoul National University R&DB Foundation | Strong · 4 | Strong · 3 | Absent | Absent | Absent |
Frequently asked questions
The corpus currently covers 135 patent families in scope. Activity has grown roughly 90% over the recent measurement window, and the most recent filing years are understated due to publication lag, so the true current size is likely higher.
Shanghai University leads with 7 patent families, followed by Zhengzhou University and Zhejiang Weixiang Materials Technology Co., Ltd. at 5 patent families each. The top five filers collectively account for 16% of the combined total of the hundred largest filers, reflecting a fragmented competitive landscape.
China is the dominant filing jurisdiction by a wide margin. The United States and Europe (EPO) each account for far fewer records, and India also appears with a notable presence. The thin non-Chinese coverage may represent freedom-to-operate opportunities in Western markets.
The evidence classifies the field as Growth, with annual filings still rising and the 90% recent growth rate confirming active IP staking. The most recent filing years are understated by publication lag and should be read as floor estimates.
Yes. The most active co-filing pair identified is Shanghai University with Zhejiang Weixiang Materials Technology Co., Ltd., with 2 joint filings. Central South University has co-filed with AECC Beijing Institute of Aeronautical Materials, and in Korea, Seoul National University R&DB Foundation has co-filed with both the Korea Institute of Science and Technology and the University of Ulsan Industry-Academic Cooperation Foundation.
B33Y (Additive manufacturing) and C23C (Coating and surface deposition) carry the lowest counts relative to their technical adjacency to the dominant ML and alloy-composition core. B22D (Metal casting) is also sparsely covered. These are observations of relative sparsity; commercial viability depends on specific technical and market conditions that should be validated independently.
Ready to map your own HEA alloy ML landscape?
Join 18,000+ innovators using PatSnap Eureka to map any technology landscape: search 2B+ patents and papers, surface key assignees, and generate a report like this in minutes.
Disclaimer. This page is generated from PatSnap Eureka data drawn from a limited snapshot of global patent and scientific-literature records, and is provided for general information and reference only.
Patent data carries inherent limitations: recent filings (typically the most recent 18–24 months) are under-counted due to standard publication lag; counts may be reported at either a patent-family or a patent-record basis and are not always directly comparable; classification, applicant-name, and citation data may contain errors, duplicates, or omissions; and the underlying search query defines and constrains the scope shown. As a result, the analysis may be incomplete or inaccurate and may not reflect the full technology landscape.
Nothing on this page constitutes an exhaustive prior-art, novelty, freedom-to-operate, or validity search, nor does it constitute legal, financial, investment, or professional advice, and it should not be relied upon as such. Any patent, commercial, or strategic decision should be verified independently and reviewed with qualified patent, legal, and domain professionals. PatSnap makes no warranties, express or implied, as to the accuracy, completeness, or fitness for any particular purpose of the information presented.