Why lipid nanoparticle and delivery system claims need chemistry-literate research, not just database hits.
The vocabulary problem is worse in formulation chemistry
Every field has some gap between how a patent describes an invention and how the literature describes the same underlying science. In formulation chemistry — lipid nanoparticles, controlled-release delivery systems, excipient combinations — that gap is wider than most, because the same formulation can be described accurately in terms of its components, its function, its manufacturing process, or its clinical effect, and a search built around any single one of those framings will miss prior art described in terms of the others.
A lipid nanoparticle claim might recite specific molar ratios of ionizable lipid, helper lipid, cholesterol, and PEG-lipid. The closest prior art might describe the same composition in terms of its encapsulation efficiency and particle size distribution, without ever stating the molar ratios directly — because the researchers who published it were optimizing for a different property and reported their formulation in those terms instead.
Where keyword search actually fails
Keyword and classification search is genuinely useful for building an initial candidate set in formulation chemistry — it's a reasonable way to surface documents that share obvious terminology. Where it fails is at the point where a formulation is chemically equivalent, or close enough to be relevant to novelty, but described in different terms because the disclosing document was written for a different purpose:
- Component naming inconsistency. The same lipid or excipient is frequently referred to by a trade name, a systematic chemical name, and an informal shorthand across different documents, and a search anchored to only one of these will miss documents using the others.
- Ratio versus absolute quantity. A claim stated in molar ratios and a disclosure stated in absolute quantities per batch describe the same relationship in incompatible search terms, even when the underlying formulation is identical once normalized.
- Function-first disclosures. Papers and patents from an efficacy or delivery-mechanism angle often describe what a formulation achieves — improved bioavailability, targeted release, reduced degradation — without stating the compositional details in a form that overlaps with claim language built around composition.
None of these are edge cases. They're the default way formulation chemistry gets published, which is exactly why a search strategy built primarily on keyword overlap under-performs in this field more than it does in most others.
What chemistry-literate research adds
The fix isn't a better keyword list — it's research that can normalize a formulation to its underlying chemistry regardless of how a given document chooses to describe it. That means reading disclosures for their actual compositional and functional content, not just their terminology, and being able to recognize when a molar-ratio claim and an absolute-quantity disclosure are describing the same formulation once the units are reconciled.
It also means treating structurally or functionally equivalent components as potentially interchangeable for search purposes — recognizing, for instance, that a claim reciting a specific ionizable lipid may have relevant prior art built around a structurally similar lipid with comparable ionization behavior, even if the exact compound named in the claim never appears in the disclosure.
Key takeaway
Formulation chemistry prior art search has to work at the level of the underlying chemistry, not just the vocabulary a given document happens to use. A search that stops at keyword and classification overlap will build a reasonable candidate set, but the prior art most likely to actually matter for novelty is often the disclosure that never uses the claim's own terms.