# IP Strategy for Deep Tech Startups Intellectual property in deep tech materials science is a moat or it is nothing. The filing strategy, the choice between patents and trade secrets, and the management of regulatory data exclusivity together determine whether a company controls its market or merely participates in it. ## The Natural Compound Problem Synthetic molecules are patentable as compositions of matter. Natural compounds — plant extracts, bioactive polyphenols, mineral particles that exist in geological deposits — are not patentable in their naturally occurring form. This is the central IP challenge for natural antimicrobials companies. The solution is to patent everything adjacent to the natural compound: - **Purified or isolated forms.** A specific purity grade, stereoisomer, or concentrated fraction of a naturally occurring compound can be novel and non-obvious relative to the crude extract. - **Formulations.** The combination of an active compound with specific carriers, stabilisers, emulsifiers, or encapsulants is patentable if the combination produces unexpected performance. - **Delivery mechanisms.** How the active is embedded in a polymer matrix, how it releases in response to humidity or temperature, how it disperses through a coating — these are patentable method claims. - **Use patents.** A new application for a known compound — specifically, preventing *Botrytis cinerea* on fresh strawberries at a defined concentration range — is patentable as a use claim even if the compound itself is not. - **Process patents.** The extraction, purification, or synthesis process that produces the active ingredient at commercial purity may be novel and patentable even if the molecule is not. ## The Patent Hierarchy Not all patents are equal. Rank them by defensibility: **1. Composition of matter claims** (broadest, most valuable). Cover the compound or material as a thing — not how it is made or what it is used for. If a competitor must use this material to compete, a composition patent is a hard block. **2. Method of manufacture claims.** Cover how the material is produced. Harder to enforce because manufacturing methods are typically trade secrets — competitors can claim a different process even if the output is identical. **3. Method of use claims.** Cover specific applications. Easier to design around; competitors can attack different applications or dosing ranges. **4. Formulation patents.** Cover specific compositions. Valuable if the formulation delivers a genuinely non-obvious performance benefit, but narrow in scope. A defensible portfolio holds claims at multiple levels. Composition claims block direct imitation. Use claims cover specific applications where competitors might try to enter with different chemical entities. Formulation patents protect the deployable product. ## Freedom to Operate The natural products space is crowded. University research programmes around the world file patents on plant extracts, polyphenols, essential oil components, and antimicrobial mechanisms. A natural antimicrobials company must conduct a Freedom to Operate (FTO) analysis before product launch in each major market. FTO asks: can we make, use, and sell this product in this jurisdiction without infringing a third party's in-force patents? FTO failure can stop a commercialisation programme entirely. A competitor's blocking patent can be designed around (reformulate to avoid infringing claims), licensed (negotiate an in-licence, typically a royalty stream), or challenged (inter partes review in the US, opposition proceedings in Europe). Each path takes time and capital. Conducting FTO early — before regulatory submissions are filed — preserves options. ## PCT Filing Strategy The Patent Cooperation Treaty (PCT) allows a single international filing that preserves priority across 150+ member states for 18 months before national phase entry. The priority year (12 months from first filing) is the working period: file the provisional, generate additional supporting data, broaden or sharpen claims before international publication. Do not waste the priority year. National phase entry is expensive. Select markets strategically based on where commercial value will be created and where manufacturing or licensing partners operate. For a natural antimicrobials company with a Singapore base and European ambitions: US, EU (via EPO), Singapore, Australia, and one or two ASEAN markets with meaningful food industries (Indonesia, Thailand) are a defensible national phase shortlist. ## Trade Secrets: The Durable Layer Patents expire in 20 years from filing. Trade secrets — properly maintained — never expire. The application know-how required to embed an antimicrobial compound into a specific polymer at a specific loading, through a specific extrusion process, with a specific compatibiliser, to achieve a specific release profile — this is knowhow that competitors cannot extract from the patent specification alone. It represents years of failed experiments that the patent does not describe. Protect application know-how as trade secret: - Document it internally with version control and access logs - Restrict sharing to employees and contractors under NDA with non-disclosure obligations that survive employment termination - Never include it in patent specifications (patent specifications are public) A layered IP architecture combines patents (public, time-limited) with trade secrets (private, potentially unlimited). When the patent expires, the trade secret keeps competitors from replicating performance even with freedom to use the composition. ## Regulatory Data as IP-Adjacent Moat A regulatory approval dossier — the GRAS file, the EFSA safety opinion, the BPR active substance dossier — is not intellectual property in the legal sense. But it functions as a moat that is often harder to replicate than a patent. **Why:** regulatory submissions require years of safety and efficacy data, substantial capital (a BPR active substance approval costs €1–3M in studies and fees), and domain expertise in regulatory affairs. A competitor cannot simply read the public dossier and replicate it — they must generate their own data through their own studies. Data exclusivity protections formalise this in some jurisdictions: - EU BPR grants 10 years of data exclusivity for new active substance approvals (Article 59, Regulation 528/2012). A second applicant cannot reference the first applicant's data during this period. - US FDA grants 5 years of new chemical entity exclusivity; up to 12 years for biological entities. The company that achieves first regulatory approval sets the bar and runs with a data exclusivity clock that forces competitors to repeat the full study burden independently. ## IP Chain-of-Title In due diligence, IP chain-of-title is the first thing investors and enterprise customers examine. Chain-of-title questions: - Does the company actually own the IP, or does a founder's previous employer or a university have a claim? - Were invention assignments executed correctly when founders joined? - If research was conducted under an A*STAR collaboration or NRF grant, what are the IP co-ownership terms? - Are there pre-existing licences from universities or research institutes that limit field-of-use? Resolve chain-of-title before any funding round. Ambiguous ownership blocks investment and creates legal exposure that is far more expensive to resolve post-funding than pre-funding. --- *Part of [[Natural Antimicrobials & Sustainable Materials MOC]]*