Patents and Additive Manufacturing: What insights can mining PBF-EB data reveal about the industry and the technology?

Intellectual Property (IP) protection is an important competitive tool in the arsenal of companies wishing to maximise the value of the product development in which they have invested much effort and many resources. Patents are a central mechanism to protect the value of any technology development. Many Additive Manufacturing companies, including startups short on resources, actively seek to patent their key inventions. However, companies often make a mistake in over-patenting, putting a strain on budgets without any apparent return on the investment. After all, patent protection requires resources.

Because the very notion of patents involves revealing an innovative idea, companies often submit patent applications at an early stage and leave the decision whether to go through the often-slow process of getting them granted – let alone enforcing them – to a later stage. It is well known that one needs deep pockets to enforce patents. That does not mean, of course, that one should avoid applying for them at all; indeed, investors often expect broad-scale patent protection to mitigate the risks of investment. Instead, a sensible and well-thought-out strategy should be considered the first step in patent protection.

While not all technologies warrant patent protection, those that do must undergo global patenting processes, to which companies contribute a significant and detailed amount of data about their technologies. Analysis of the data can reveal trends and trace technological relationships that extend well beyond the dry legal and procedural mechanics of the patenting process. The mountain of data contains aggregated, global information on what companies and researchers are doing, and the ability to mine this data allows companies to understand the nuances of their colleagues’ work and any market trends.

Why patents?

So why do companies apply for patents, and what do they hope to get out of them? Applying for patent protection offers several strategic advantages, including:

Exclusivity:

• A patent gives the inventor exclusive rights to use, produce, and sell the invention and prevents others from copying or using the invention without a licence.

Competitive advantage:

• Patent protection allows a company to establish a monopoly on its technology, thus giving it an edge over competitors.

Revenue

• Patents can be licenced to others, creating a potential revenue stream and potentially increasing the value of a business for investors and other stakeholders.

Credibility

• Patents can cement a company’s reputation and demonstrate its commitment to innovation.

Attracting investment and raising valuation

• Patents can help attract funding from investors seeking a unique product or business idea to fund. Patents can also raise a company’s valuation.

Strategic relationships

• Patents often enable companies to form partnerships, joint ventures, or collaborations by ensuring protection of assets.

History of patent protection in AM

Patents have always been an important factor in the development of the Additive Manufacturing industry, with the first key patent awarded in 1986 for the Vat Photopolymerisation process of stereolithography (Fig. 2). The value of patents in safeguarding a nascent technology was illustrated by the Material Extrusion (MEX) process, of which Fused Deposition Modelling (FDM) was the first commercial offering when introduced by Stratasys in 1992. Twenty years after those MEX patents began to expire, the Additive Manufacturing industry experienced its first hyper-growth period, based on a flood of companies entering the market using the MEX process.

The number of patents applied for and awarded in the AM area also started showing high rates of growth in 2012. One factor was the ending of the monopoly on Stratasys’ basic patent; new entrants began developing ideas that followed from that original technology, sparking greater interest in wider Additive Manufacturing and leading many newcomers into the industry. The rate of patent applications and awards is a useful indicator of trends in an industry. Because patenting often occurs well before products are launched, patents can also be an indicator of the industry’s future health and prospects.

The Wohlers Report includes the number of US patent applications and awards. In 2012, fewer than 300 patents were awarded, and about 150 applications were filed. Five years later, in 2017, the number of AM-related patents awarded reached almost one thousand. That same year, the number of US patent applications outstripped the number of awarded patents by 100%. By 2020, the number of AM patents granted in the US peaked at more than five thousand, a compounded annual growth rate of almost 43% over eight years. In 2021, the number of awarded patents plateaued and has declined slightly each year since then.

Disputes and infringements

With patents so integral to the industry, there have, of course, been several disputes punctuating the history of the AM industry. Most recently, Stratasys announced that it was suing Bambu Labs – an up-and-upcoming desktop polymer MEX AM machine manufacturer – for infringement of certain patents. Recently, Continuous Composites filed a claim agaist Markforged. Markforged eventually settled the dispute in September 2024 and agreed to pay $25 million in licencing fees in exchange for a cross licence, allowing it to continue sales of its products.

In the metals area, a 2018 jury case found that Markforged had not infringed patents awarded to Desktop Metal; that case has been well documented and analysed in the literature [1], so will not be covered here.
In earlier times, 3D Systems was known to avidly protect its intellectual property, which effectively maintained a competitive edge in stereolithography. Notably, Japanese companies that had begun to develop machines based on this technology in the 1980s and 1990s were blocked from selling their systems outside of Japan.

The patent data system

The international patent processing system is a framework that takes copious technical data and squeezes it into an established format. While each country has its own process, under international agreements such as the Patent Cooperation Treaty, an application in one country is recognised as valid and shares a priority date with patent systems in all member countries. To gain actual protection in a particular country, however, the applicant has to go through a ‘national phase’ in the jurisdictions of interest where it makes sense to invest in patent protection. Each jurisdiction has its own rules, filings and fee structure, which is what makes international patent protection so expensive and favours inventors from companies with deep pockets.

The filing contains a number of different kinds of data:

• Bibliographical information, including application and patent numbers, inventor information, filing dates, priority dates, and where the patent is filed
• Technical data such as the title of the invention, abstract, description, claims and drawings
• Legal status data such as filing status, date of the grant, and expiry date.

The content of the patent describes the innovation for which protection is sought. Patent offices have highly developed patent classification systems; International Patent Classification (IPC) is a common hierarchical one with different levels of detail, while the Cooperative Patent Classification (CPC) system is more detailed. The CPC was developed jointly by the European Patent Office and the United States Patent and Trademark Office to provide better accuracy and detail.

The established format for data classification and publication is what allows modern data processing methods to extract insights from data published by patent systems across the world. Most of this information is available digitally after publication which, depending on the jurisdiction, is typically eighteen months after filing. Several patent data processing and analysis products are available to ease the task of sifting through vast amounts of descriptive text data and aggregate them to provide a picture of who is doing what in a particular area.

The patent system is designed to protect innovation, and so explaining how the claimed innovation differs from previously filed or granted patents is a key element of how patents are written. These new patents will, therefore, reference a previous patent for the purposes of differentiation or will be cited by patent examiners during an office action on new patents as relevant for determining the nature of the innovation. Similarly to academic papers, important prior patents will be cited more frequently than less important patents. As shall be illustrated, this serves as an indicator of the technological value and importance of the patent being cited. Big data processing methods are able to display a picture that is much more nuanced and insightful than simply listing patents having a particular CPC code.

Throughout this article, the authors have used the Lexis Nexis Patent Sight data processing platform to conduct patent analysis.

Case study: Patent analysis of Electron Beam Powder Bed Fusion (PBF-EB)

To demonstrate how complex patent data from the global patenting system can be mined for insight and prediction, we will look at a relatively compact subsection of metal AM: PBF-EB. This technology accounts for between 4-8% of sales of metal AM machines in the industry today. For comparison, the leading AM technology – Laser Beam Powder Bed Fusion (PBF-LB) – accounts for 80-90%. PBF-EB is the ISO-ASTM designation of the process, but, in the industry, and also in patent filings, the technology may be referred to as electron beam melting, EBM, Ebeam, Electron Beam Additive Manufacturing (EBAM), and Electron Beam Selective Melting (EBSM).

Laser and electron beam PBF share some common elements: the selective fusing (or melting) of metal powder in a powder bed, spreading (or recoating) a new layer of metal powder onto a previously melted layer (after the fused layer has been lowered by one layer), and then selectively melting the next layer.
While the basic technology of the electron beam was not invented for AM, the combination of melting powder by an electron beam source to selectively build a design layer by layer is the innovation of PBF-EB. Some companies active in electron beam technology for welding and electron microscopy have, in recent years, expanded into PBF-EB as a new business opportunity. This fact will require some understanding when searching the patent data bases, because simply searching the term ‘electron beam melting’ will include areas not related to AM.

Selecting a portfolio

The way we began analysing the patent landscape in PBF-EB is to begin with the companies that are currently active in this area. The use of electron beam technology for AM was pioneered by Sweden’s Arcam, starting in 1997. In 2016, Arcam was acquired by GE Aerospace, which established a business unit called GE Additive to manage its AM activities. In 2024, GE Additive was renamed Collibrium Additive. The patent landscape analysis also reveals M&A activities and change of ownership status. Table. 1 shows the companies from around the world commercially active in PBF-EB and which were analysed for this case study.

Patents filed

A search was conducted to show the patent activity of the portfolio of companies active in the space. To filter out patent filings from large companies just as GE Aerospace and JEOL, the search was limited to Additive Manufacturing CPC B33Y (Fig. 3). To allow for filings connected to other Additive Manufacturing technologies, the search was limited to filings containing variations of the text ‘electron beam’ or ‘particle beam.’ A total of 272 applications were identified. The filings, as expected, were dominated by GE Aerospace or GE Vernova, the successor corporate entities that acquired Arcam and its patent portfolio. The next biggest patent filers were Chinese companies QBeam and Sailong, which began their patenting activity in around 2015. Starting in 2018, we find patent activity starting to appear from PBF-EB start-up company Freemelt. In 2020, newcomer Wayland Additive began to appear in the filings (Fig. 4).

A deeper understanding of the filings is revealed when we examine what is known as ‘External Technology Relevance.’ This measure is calculated to answer the following question: How frequently are patents of a particular company cited in subsequent patent applications by all other applicants? A surprising result in technology relevance is Wayland Additive. Although Wayland has to date filed only five electron beam patents (they may have filed other applications relating to mechanisms and devices in their machine, such as powder handling), the external technology relevance is already relatively high, suggesting that their innovation has generated quite a bit of attention since its filing at the end of 2020 in the UK in filing 2020407.9 (GB) (Fig. 5).

Based on what the company has announced and what has been published in past issues of Metal AM magazine, it developed a method of neutralising charged particles in the powder bed, a well-known issue in PBF-EB technology. The commercial expression of this innovation is called NeuBeam. It turns out that in the patenting world, this patent application garnered a great deal of attention.

Relevance over time

One would expect that a patent or patent application that has been around for a long time would be cited more frequently than a new patent application. Therefore, a patent garnering more attention closer to its filing date should be considered a significant patent.

Fig. 6 reveals that the youngest patent application plotted against external technology relevance is a filing by Chinese company Sailong. Even though it was filed only two years ago, it has attracted a relatively significant amount of attention. The patent is entitled ‘A method for improving surface quality of an Additive Manufacturing part and an Additive Manufacturing device.’ The industry has evidently taken note.

Where did the applicants seek protection?

As expected, the US dominated when searching where the PBF-EB portfolio was active. China showed very active patenting applications, more than in either Europe or Japan (Fig. 7).

Patents in PBF-EB: taking a broader view

Who else is patenting more broadly in this field, even though there is probably no intention to commercialise the invention as a standalone product? Processing from the patent data set indicates that the academic segment is quite active in this space.

This analysis uses a CPC classification to limit the search to patents only relating to the making of articles using metal powder. This classification (B22F) includes patents for Powder Metallurgy, such as Metal Injection Moulding. To filter out the general Powder Metallurgy or metal Binder Jetting patents, the search was limited to those patents referencing electron beam technology.

It turns out that 135 academic institutions have filed for patents touching upon PBF-EB in some way. Of the eleven institutions in this area with a portfolio size of more than ten patents, ten are from China. Fraunhofer is the only institution from western countries that makes the listing of top patenting institutions in PBF-EB. Among smaller portfolio sizes (less than ten patents per institution), we start to see additional western institutions.

However, when one ranks filings by academic institutions according to relevance, a different picture emerges: only three Chinese institutions make the top fifteen.

This finding supports the relatively well-known fact that Chinese institutions and the Chinese government encourage researchers and companies to file patents. Since the patent process is expensive, without budgetary support, many of these patents would probably not be filed. Time will tell how many of these applications are worth anything. One thing is clear, though: once an idea has been included in a patent application anywhere, it effec­tively becomes common knowledge. If not claimed, the patent owner does not have a monopoly. The ideas revealed in patent filings can, therefore, be used by anyone, meaning that competitiveness relies on who actually makes a good product that uses the innovation and who can produce it more cost-effectively.

Invented in China

Since China seems to be an up-and-coming market, where patents are being filed both by Chinese organisations and international companies seeking patent coverage in China, we can zoom in to get greater clarity on the China patent landscape in recent times.

The data reveals that forty-one organisations have filed at least three patent applications in the electron beam area in the last three years. Of them, twenty-three were universities or research institutes, and sixteen were companies; the remainder was the government. Among the companies, Sailong was by far the most active in patenting technologies, with thirty-three patent applications since January 2021. The other commercially active company offering a PBF-EB machine, Qbeam, has not appeared to have filed applications since 2021.

The second most active company was the China Aviation Industry Corporation. Since there is no evidence that this company plans to offer a PBF-EB machine, we can assume that their patenting interest relates to applications of PBF-EB for the aerospace industry.

Active in PBF-EB

Worldwide, several companies are active in PBF-EB patenting, even though they are not directly involved in the development of machines and, instead, are probably mainly interested in applications of the technology.

Table. 2 lists companies with at least eight patents in PBF-EB AM. The usual suspects, such as GE, JEOL and Sailong, appear here. A second group of companies, such as Siemens Energy, RTX, Stryker and Boeing, use the technology for their own businesses.

Another group of companies consists of AM suppliers of PBF-LB machines, including EOS, Trumpf, Sodick and Nikon. Could these companies be planning in the future to expand to PBF-EB? Could they be considering acquisitions? Could they merely be covering their bases by adding electron beam to their core PBF-LB businesses? The patent data can tip us off to further examine what companies’ plans may be in the space.

Proterial is a diversified Japanese industrial manufacturing concern that has roots in Hitachi Metals, with annual sales of $7.9 billion in 2023. Could it be interested in expanding its position in AM, possibly via acquisition in the footsteps of Nikon? Could IHI Corp, a $9 billion Japanese company also be interested in AM? INEOS is a Japanese company active in the oil and gas segment with annual sales of more than $100 billion. What is their interest in AM?

Another company to note is Matsuura, primarily a manufacturer of traditional machine tools, although one that has also dabbled in AM in the past several years with laser-based hybrid machines. Could it be planning to develop an offering in PBF-EB machines?

Divergent 3D is well known in AM circles for its automotive manufacturing technology, where PBF-LB is an important component. Could it also be considering an expansion into Electron Beam? It would certainly be interesting to review the ten patents it has in its portfolio to try and find out.
Several names not usually associated with AM also pop up on the list, and it is intriguing to understand what connection, present or future, they have to the PBF-EB segment.

In conclusion

Smart mining of patent data can be insightful for AM industry competitors, analysts and investors. It can show us the technology leaders and highlight the most important or relevant patents. It can reveal territorial strengths and weaknesses. New companies can be revealed, or existing companies’ plans to enter the market might be divulged. Technology mavericks owning the most cited patents can be identified. Potential collaborative efforts or business combinations can be planned.

The patent data system is a rich ecosystem containing a large amount of information. In some cases, we may not be able to learn all the answers from the data on a company’s plans and business strategy, but it can certainly guide us in asking the right questions.

Authors

Joseph Kowen

Joseph is an industry analyst and consultant who has been involved in Additive Manufacturing since 1999. He is an Associate Consultant at Wohlers Associates, part of ASTM International’s AM Center of Excellence.

www.linkedin.com/in/joseph-kowen-a5129b3/

Gil Perlberg

Gil is an intellectual property consultant and advisor with over thirty years of experience across diverse technology sectors, helping businesses chart sensible, business-driven strategies for intellectual property management. He is recognised by Intellectual Asset Management as a leading global IP strategist.

www.linkedin.com/in/gilperlberg/
www.perl-ip.com

Sources

Processing of patent data for this article was performed using LexisNexis PatentSight+

References

[1] Frontiers in Research Metrics and Analytics, Volume 7, 02 September 2022