Chemical and Materials IP Checklist: Genus and Species Scope, Data Sufficiency, Formulation Secrecy, Regulatory Confidentiality, and Freedom to Operate

By Casey Scott McKay · 2024-10-27

This checklist runs a chemical or materials IP position from the analytical bench outwards, because the single most consequential decision in the field — patent or keep secret — is answered by a laboratory rather than by doctrine. Phase one establishes what an independent analyst can recover from the product. Phase two audits the specification against the claimed genus, since the distribution of working examples determines how much scope survives. The middle phases cover the comparative data programme, range criticality, the layered claim set, and the regulatory and commercial channels through which composition leaks without anyone intending it. The closing phases handle collaboration paper, trade secret controls for process know-how, freedom to operate by structure search, post-grant readiness, and the annual review that keeps secrecy decisions aligned with improving instrumentation.

IP and Technology > Patent Counseling Transactions | Checklist | Published 27 October 2024 - Updated 9 May 2025 | Casey Scott McKay - marksy.us

Summary. This checklist runs a chemical or materials IP position from the analytical bench outwards, because the single most consequential decision in the field — patent or keep secret — is answered by a laboratory rather than by doctrine. Phase one establishes what an independent analyst can recover from the product. Phase two audits the specification against the claimed genus, since the distribution of working examples determines how much scope survives. The middle phases cover the comparative data programme, range criticality, the layered claim set, and the regulatory and commercial channels through which composition leaks without anyone intending it. The closing phases handle collaboration paper, trade secret controls for process know-how, freedom to operate by structure search, post-grant readiness, and the annual review that keeps secrecy decisions aligned with improving instrumentation.

Keywords: chemical IP checklist · analytical recoverability test · specification ladder · example distribution · Markush scope audit · unexpected results data · criticality boundary data · polymorph screening record · product-by-process fallback · impurity fingerprint · TSCA confidentiality reassertion · safety data sheet withholding · disclosure register · collaboration assignment · material transfer reach-through · process compartmentalisation · purchase and analyse monitoring · post-grant readiness · term cliff mapping · portfolio pruning

How to use this checklist

| Field | Detail | |---|---| | Who runs it | Patent counsel with the project chemist, regulatory affairs, and analytical services | | When | On invention disclosure; before each filing; annually across the portfolio | | Time required | Four to six weeks including outside analytical work | | Gates | Analytical recoverability determined; example distribution fixed before filing; confidentiality claims substantiated | | Output | A protection decision memo, a specification ladder, a comparative data plan, and a disclosure register | | Companion documents | Protecting a Chemical or Materials Invention and Claiming a Compound |

The matter. A specialty chemicals business has developed a new additive that improves the durability of an industrial coating. The lead compound is one of a family the chemists believe is broadly active. Eleven analogues have been made, all from the same synthetic route and all clustered around the lead. The team presented preliminary results at a conference five months ago. A premanufacture notice will be required before commercial supply. Two large customers have asked for full composition disclosure as a condition of qualification. The synthesis is run by a toll manufacturer under a two-page confidentiality agreement, and the process parameters were emailed to them. Nobody has asked what a competitor's laboratory could determine from a sample of the finished coating.

Phase 1. Establish analytical recoverability

[ ] Send the product to an independent laboratory with an open brief. "Tell us what is in this and how it was made." Why. The patent-or-secret decision turns on this answer and on nothing else. Trap. Asking your own analytical group, which knows what to look for and will therefore find it too easily.
[ ] Record separately what is recoverable about composition and what is recoverable about process. Why. They almost always differ, and the answer is usually to patent the first and keep the second. Trap. A single yes-or-no conclusion that conceals the split.
[ ] Accept that reverse engineering is lawful. Why. Bonito Boats v. Thunder Craft Boats treats it as legitimate discovery and 18 U.S.C. § 1839 excludes it from misappropriation. Trap. A secrecy strategy premised on competitors behaving honourably about analysis.
[ ] Confirm both regimes are available. Why. Kewanee Oil v. Bicron confirms trade secret law is not preempted by patent law. Trap. Treating the choice as exclusive across the whole invention rather than element by element.
[ ] [Gate] Record the decision in a one-page memo stating what was recoverable, what is patented, what is kept secret, and why. Why. It supports reasonable measures later and explains the strategy to a successor. Trap. A decision made in a meeting and never written down.

Phase 2. Audit the specification against the claim

[ ] Count the claimed space and the examples. Why. A genus with millions of members and eleven clustered examples is the standard failure pattern, and it is visible in five minutes. Trap. Assessing scope by reading the claim rather than by enumerating it.
[ ] Plot where the examples sit within the claimed space. Why. Examples at the edges of each substituent position support the genus; examples around the lead support the lead. Trap. Counting examples without mapping them.
[ ] Test against Amgen Inc. v. Sanofi. Would a skilled artisan have to screen to reach the members at the edge? Why. A roadmap for trial and error is not enablement. Trap. Confidence that the art is predictable enough, in a field courts treat as unpredictable.
[ ] Test against Ariad Pharmaceuticals v. Eli Lilly. Does the specification identify a structural feature common to the genus and correlated with the property? Why. Possession may be shown by structural features or by a representative number of species; a specification that offers neither relies on example count alone. Trap. A specification that describes what the compounds do and never says why.
[ ] Apply the In re Wands factors explicitly and write down the answers. Why. They are the framework the examiner and the court will use. Trap. Treating them as a litigation topic rather than a drafting tool.
[ ] Strip functional limitations out of structural claims. Why. "Wherein R is any group that improves solubility" imports the entire Amgen problem into a claim that did not have it. Trap. Functional language introduced during prosecution to escape a prior art reference.
[ ] Write intermediate subgenera into the specification explicitly. Why. European practice frequently refuses intermediate generalisations from lists as added matter. Trap. Relying on fallback positions that exist only in the drafter's intention.
[ ] [Gate] Do not file until the example distribution matches the intended claim. Why. Additional syntheses cost weeks; lost scope is permanent. Trap. Filing to meet a conference deadline and planning to add examples in the continuation, which does not work.

Phase 3. Build the comparative data

[ ] Identify the closest prior art before designing the experiments. Why. In re Dillon makes structural similarity a prima facie case, and only comparison against the closest art rebuts it. Trap. Comparing against a distant analogue, which signals avoidance.
[ ] Make the data commensurate with the claim. Why. Results for one species narrow the claim to that species. Trap. A single spectacular result presented as support for a genus.
[ ] Frame the result as unexpected, with a reason. Why. In re Papesch allows properties to rebut structure, but "better" is not "unexpected". Trap. A declaration asserting surprise without stating what the art predicted.
[ ] Design once for both frameworks. Why. United States practice asks about unexpected results against the closest art; European practice asks for a technical effect within a problem-and-solution structure. Trap. Two badly designed studies that satisfy neither.
[ ] Preserve failed experiments and inactive analogues. Why. They are the best evidence of unpredictability against an obvious-to-try attack under KSR International v. Teleflex. Trap. Laboratory clear-outs that discard exactly the records that would have won.
[ ] Prepare 37 C.F.R. § 1.132 declarations properly. Personal knowledge, attached data, stated conditions. Why. Conclusory inventor declarations are discounted. Trap. A declaration drafted by counsel and signed without the underlying data attached.

Phase 4. Ranges, criticality, and solid forms

[ ] Generate data at and just outside every claimed boundary. Why. Overlapping ranges are prima facie obvious and routine optimisation is not inventive — In re Aller. Trap. Midpoint data offered as proof of criticality.
[ ] Look for positive teaching away, not silence. Why. Art that discourages the claimed range is valuable; art that omits it is not. Trap. Characterising an absence of disclosure as a teaching away.
[ ] Record the polymorph screen. Conditions tried, forms found, forms that would not form. Why. It answers the obvious-to-try argument that screening was routine, illustrated by Pfizer v. Apotex. Trap. A screen run and not documented.
[ ] Test whether the prior art process inherently produces the claimed form. Why. Atlas Powder v. Ireco and Schering v. Geneva Pharmaceuticals anticipate what the art necessarily produced, whether or not it was recognised. Trap. Assuming novelty because nobody named the form.
[ ] State the analytical method for any property or impurity limit. Why. Different methods give different numbers, which is a definiteness problem under Nautilus v. Biosig Instruments. Trap. A numerical limit with no method, in a field where method choice moves the number.

Phase 5. Build the layered claim set

[ ] Claim 1: the broad composition or genus. Widest scope the specification honestly supports. Why. It is the claim you defend and may lose. Trap. Drafting it to the scope you want rather than the scope the data reaches.
[ ] Intermediate independent or dependent claims: the subgenera. Structural, not property-based. Why. These survive when claim 1 does not. Trap. Subgenera defined by the property they were selected for, which is functional claiming again.
[ ] A separate formulation claim. Active, excipients, ratios, physical form. Why. It is usually the most durable claim because it maps onto the actual product. Trap. Leaving formulation to a dependent claim on a genus that falls.
[ ] A process claim. Why. 35 U.S.C. § 271(g) reaches imported product made abroad by a patented process. Trap. Omitting it because the process is also a trade secret, which is a false choice where offshore manufacture is foreseeable.
[ ] Method-of-use claims drafted with indirect infringement in mind. Why. 35 U.S.C. § 271(b) and 35 U.S.C. § 271(c) are the practical route against a supplier. Trap. A method claim performed only by end users with no supplier step.
[ ] Product-by-process only as a fallback. Why. Abbott Laboratories v. Sandoz makes the process limitations binding on infringement while validity is assessed on the product. Trap. Relying on it against a competitor using a different route.
[ ] Consider an impurity-limited claim. Why. A limit a competitor's route cannot meet is a route-specific claim in composition form. Trap. A limit the prior art already met inherently.

Phase 6. Close the disclosure channels

[ ] Substantiate every 15 U.S.C. § 2613 confidentiality claim at the time it is made. Why. An unsubstantiated claim fails and publishes the identity. Trap. A generic name that is not as specific as practicable.
[ ] Diarise confidentiality reassertion dates. Why. Lapsed claims publish composition and nobody notices until a competitor cites it. Trap. Reassertion treated as a regulatory affairs task with no legal diary entry.
[ ] Review safety data sheets for unnecessary disclosure. Why. 29 C.F.R. § 1910.1200 permits withholding specific identity as a trade secret. Trap. A sheet naming everything because it was simpler.
[ ] Model foreign registration disclosure before entering a market. Why. Public inventories and full-composition registration defeat secrecy in the markets where you sell. Trap. Entering a market and discovering the disclosure afterwards.
[ ] Control customer specification disclosure. Why. Qualification pulls composition detail out of sales teams, and it then lives in the customer's supply chain. Trap. Disclosure under a framework NDA signed years earlier by procurement.
[ ] Gate conference abstracts, posters, and theses. Why. 35 U.S.C. § 102(b) forgives the applicant's own disclosure for twelve months domestically and nowhere else. Trap. A student poster that ends the European position.
[ ] [Gate] Maintain a disclosure register recording every channel, what was disclosed, the confidentiality basis, and the reassertion date. Why. It is the single most useful document a chemical IP function keeps. Trap. Three partial lists in three departments.

Phase 7. Collaboration and supply paper

[ ] Identify who conceived what before assuming ownership. Why. Inventorship follows conception, not funding or effort; a contract research organisation synthesising to specification usually conceives nothing. Trap. Naming everyone who worked on the project, which is its own invalidity risk.
[ ] Avoid joint ownership. Why. 35 U.S.C. § 262 lets each co-owner practise and license without accounting, all co-owners are necessary parties to enforcement, and 35 U.S.C. § 116 makes a contributor to one claim an inventor of the whole patent. Trap. Accepting joint ownership as a fair compromise, which it is not.
[ ] Use present-tense assignment in every research agreement. Why. "Agrees to assign" leaves title with the assignor. Trap. A template inherited from a services agreement.
[ ] Separate background from foreground explicitly. Libraries, assays, analytical methods, and process know-how are background; compounds and data arising are foreground. Why. Field-of-use allocation is how parties share foreground without deadlock. Trap. A definition of background that swallows the foreground.
[ ] Read material transfer agreements for reach-through. Why. Rights claimed in inventions made using transferred material can encumber a programme years later. Trap. MTAs signed by scientists without review.
[ ] Negotiate publication timing, not publication rights. Why. An academic partner's right to publish is not negotiable in substance. Trap. A review window shorter than the time it takes to prepare a chemical filing.
[ ] Review toll manufacturing terms. Confidentiality reaching the toller's suppliers, prohibition on producing for others using the process, ownership of improvements, destruction at termination. Why. The toller receives recipe, parameters, and methods. Trap. A two-page confidentiality agreement covering a full process transfer.
[ ] Check distribution agreements for local mark registration. Why. A distributor registering your mark in its territory converts a supply dispute into a hostage negotiation. Trap. Brand terms omitted because the agreement is about chemistry.

Phase 8. Protect the process

[ ] Compartmentalise the process. Order of addition, catalyst treatment, thermal profile, quench timing held separately. Why. A departure that carries one compartment is survivable; one that carries all of them is not. Trap. A single master batch record with universal access.
[ ] Test the programme against 18 U.S.C. § 1839. Why. Reasonable measures are an element of the claim under the Defend Trade Secrets Act, not an aspiration. Trap. A policy with no access controls behind it.
[ ] Include the 18 U.S.C. § 1833 whistleblower notice. Why. Omission forfeits exemplary damages and fees against that person. Trap. Notice in the handbook rather than the agreement.
[ ] Run exit forensics for anyone with parameter access. Why. Process know-how travels in a spreadsheet. Trap. Exit procedures triggered by seniority rather than by access.
[ ] Consider what is observable from outside. Why. E.I. duPont deNemours v. Christopher found improper means without trespass; the same reasoning means plant construction, deliveries, and effluent are all information. Trap. Assuming the fence is the perimeter.

Phase 9. Freedom to operate

[ ] Search by structure, not by text. Why. The question is whether a compound is a member of a claimed set, which keyword search cannot answer. Trap. A literature-style search report presented as clearance.
[ ] Follow membership with a validity assessment. Why. In chemistry the broad claim is often invalid, and an opinion that stops at membership has done half the work. Trap. Advising a design-around that was never necessary.
[ ] Read the prosecution history for surrendered territory. Why. Festo v. Shoketsu Kinzoku Kogyo creates a presumption of surrender from narrowing amendments, and Warner-Jenkinson v. Hilton Davis Chemical requires element-by-element analysis. Trap. Assessing the claim as granted without the file.
[ ] Search process claims separately. Why. A compound clear of composition claims may still be caught on import under 35 U.S.C. § 271(g). Trap. Composition-only clearance for an imported product.
[ ] Design around with chemists and counsel in the same room. Why. Moving outside a substituent list is a chemical decision with a legal test. Trap. A legal memo describing the boundary and a laboratory that never reads it.
[ ] Obtain a written opinion where exposure is material. Why. Opinions remain relevant to wilfulness and enhanced damages under Halo Electronics v. Pulse Electronics. Trap. An oral clearance recorded in nobody's file.

Phase 10. Term, post-grant, and enforcement readiness

[ ] File provisionals that actually enable and describe. Why. Priority extends only to what was enabled and described; later-added genus members take the later date. Trap. A thin provisional filed to beat a disclosure, followed by a broad non-provisional.
[ ] Protect term adjustment. Why. Extensions of time and unnecessary continuations cost days of exclusivity worth more than the fees saved. Trap. Routine one-month extensions taken for convenience.
[ ] Plan the 35 U.S.C. § 156 extension choice where regulatory review applies. Why. One patent per approved product, chosen by the patentee. Trap. Defaulting to the composition patent when the formulation patent faces the real threat.
[ ] Screen for obviousness-type double patenting across the family. Why. Long families and continuations generate it, and a terminal disclaimer ties term and enforceability. Trap. Discovering it in litigation.
[ ] Assemble the 35 U.S.C. § 112 support file before grant. Example distribution rationale, structural-feature explanation, unpredictability evidence. Why. Post-grant review reaches section 112 grounds within nine months of issue. Trap. Assembling it after a petition is filed.
[ ] Contest the definition of the person skilled in the art early. Why. It determines whether screening was routine, and petitioners draft it to suit the outcome. Trap. Accepting the petitioner's definition and litigating everything downstream of it.
[ ] Run purchase-and-analyse monitoring continuously. Documented provenance, validated method, sealed retention. Why. Detection in chemicals is analytical and the chain of custody is contested in nearly every case. Trap. Monitoring triggered by suspicion, which is always late.
[ ] Build the impurity reference data set on your own material now. Why. Where the product is identical whichever route made it, the by-product profile is the only route evidence available. Trap. Trying to establish a baseline after the dispute starts.

Phase 11. Annual review

[ ] Re-test secrecy decisions against current instrumentation. Why. A composition unrecoverable a decade ago may be trivially resolvable now. Trap. A secrecy decision treated as permanent.
[ ] Map claims to current products. Why. Portfolios drift toward protecting what the company used to sell. Trap. Counting filings instead of coverage.
[ ] Diary the term cliffs within five years. Why. The commercial response — formulation filings, process improvements, brand, cost position — needs years of lead time. Trap. Learning the expiry date from a competitor's launch.
[ ] Audit the marketing library against the patent positions. Why. Bulletins describing ranges as "typical" and grades as "essentially the same" will be produced against you. Trap. An annual review that covers filings but not literature.
[ ] Review confidentiality reassertion deadlines and any lapsed claims. Why. This is the most avoidable disclosure in the field. Trap. Assuming regulatory affairs is tracking it.
[ ] Prune the portfolio. Why. Maintenance on discontinued grades is pure cost and obscures what matters. Trap. Renewing by default because abandonment requires a decision.

Phase 12. Working the example matter

[ ] Deal with the conference presentation first. Why. Preliminary results presented five months ago are a disclosure. The domestic grace period under 35 U.S.C. § 102(b) leaves seven months; absolute-novelty jurisdictions are already lost unless what was presented was genuinely less than the claimed invention. Obtain the slides and the abstract, and assess precisely what they enabled. Trap. Assuming a poster was too preliminary to count, without reading it.
[ ] Stop and count before filing. Why. Eleven analogues from one synthetic route, all clustered around the lead, will support a claim to that cluster and not to the family the chemists believe is active. Trap. Filing on the eleven and planning to prove the genus later.
[ ] Commission analogues at the edges. Why. A dozen additional syntheses spanning the substituent positions is a few weeks of laboratory time and is the difference between a family claim and a lead-compound claim. Trap. Treating the additional chemistry as a research cost rather than a legal one.
[ ] Send the finished coating to an outside laboratory. Why. Nobody has asked what a competitor could recover from the product, which is the question that decides whether the additive should be patented, kept secret, or both. Trap. Assuming an additive at low loading in a cured coating is undetectable, which modern instrumentation frequently disproves.
[ ] Re-paper the toll manufacturer before the next campaign. Why. A two-page confidentiality agreement covering a full process transfer, with parameters sent by email, is neither a reasonable measure under 18 U.S.C. § 1839 nor a workable contract. It needs improvement ownership, a prohibition on producing for others by the process, subcontractor obligations, and destruction at termination. Trap. Leaving it because production depends on the relationship.
[ ] Prepare the premanufacture notice confidentiality claim now, with substantiation. Why. 15 U.S.C. § 2613 requires substantiation at the time of the claim and a generic name as specific as practicable. Trap. Filing the notice and adding the confidentiality substantiation afterwards.
[ ] Answer the two customers with a controlled disclosure. Why. Full composition disclosure into two customers' specification systems is a permanent publication into their supply chains. Offer performance data, a specific-purpose confidentiality undertaking naming the document, or disclosure of a masked composition sufficient for qualification. Trap. Sales agreeing to full disclosure to win the qualification.
[ ] [Gate] File nothing until the example distribution, the analytical answer, and the disclosure position are all recorded. Why. Each one changes what the application should say. Trap. A filing date driven by a board meeting.

Phase 13. Materials-specific questions

[ ] Treat composition ranges as the Markush analogue. Why. The overlapping-range obviousness analysis applies directly, and criticality at the boundary is the only reliable escape. Trap. Assuming alloy and polymer claims escape the chemistry case law.
[ ] Decide how microstructure will be claimed and characterised. Why. Identical composition processed differently gives different properties, and microstructure is hard to define with the precision a claim requires. Trap. Drifting into product-by-process claiming by default and inheriting the Abbott Laboratories v. Sandoz limitation.
[ ] Keep property limitations out of independent claims. Why. A claim to a material "having a tensile strength greater than X" is a functional claim, with the full Amgen Inc. v. Sanofi enablement burden. Trap. Property claims drafted because the composition is hard to define.
[ ] Check published industry specifications as prior art. Why. Where a material is sold against a specification, the specification defines the composition envelope and may anticipate. Trap. Treating a specification as a commercial document rather than a publication.
[ ] Claim coatings and surface treatments separately from the substrate. Why. The coated article, the coating composition, and the application process reach three different infringers. Trap. A single claim to the finished article that misses the coating supplier.
[ ] Address recyclability and end-of-life claims deliberately. Why. They are becoming regulatory obligations and marketing claims at once, and a specification asserting environmental performance will be read against the advertising. Trap. Environmental language written into a patent by a marketing reviewer.

Phase 14. Documents this checklist should produce

[ ] A protection decision memo. What an outside laboratory recovered about composition and about process, what was patented, what was kept secret, and why. Why. It supports reasonable measures and survives staff turnover. Trap. A decision recorded only in a meeting minute.
[ ] A specification ladder plan. Genus, subgenera, preferred lists, and the map of where examples sit within the claimed space. Why. It is the document that makes the filing decision reviewable. Trap. A claim set drafted without a scope map behind it.
[ ] A comparative data plan and file. Closest prior art identified, experiments designed for both frameworks, boundary data for every range, failures preserved. Why. It is the evidence the whole non-obviousness case rests on. Trap. Data generated reactively against each rejection.
[ ] A disclosure register. Every regulatory filing, safety data sheet, customer specification, bulletin, and presentation, with the confidentiality basis and reassertion date. Why. No single function sees all these channels. Trap. Three partial registers.
[ ] A monitoring file. Sample purchases with provenance, validated analytical methods, impurity reference data, sealed retentions. Why. It converts enforcement from a project into a decision. Trap. Building it after the first suspected infringement.

Phase 15. International filing decisions

[ ] Write the combinations you may want as explicit embodiments. Why. European practice treats combining a preferred value from one list with a preferred value from another as an intermediate generalisation and often refuses it as added matter. Trap. Relying on United States amendment practice in a European file.
[ ] Identify the objective technical problem before drafting. Why. European obviousness runs on problem-and-solution: closest prior art, difference, objective problem, obviousness of the solution. Data designed only for an unexpected-results argument frequently fails to identify the effect. Trap. Assuming one comparative study serves both offices without being designed to.
[ ] Make the technical effect plausible on the application as filed. Why. Sufficiency and plausibility have no exact domestic analogue, and post-filing data cannot rescue an effect the application never made plausible across the claimed scope. Trap. A specification that asserts breadth and demonstrates a point.
[ ] Audit every self-disclosure against the grace period map. Why. 35 U.S.C. § 102(b) forgives twelve months domestically; most jurisdictions forgive nothing. Trap. A thesis, poster, or customer bulletin that quietly ends the foreign position.
[ ] Budget for unity of invention objections. Why. A Markush claim that proceeds as one invention domestically may attract multiple search fees or restriction elsewhere. Trap. Discovering the cost at national phase entry.
[ ] Consider utility model routes for formulations and articles where available. Why. Fast unexamined rights suit short product cycles. Trap. Assuming they are available for compounds, which in several jurisdictions they are not.
[ ] Control translation cost by structuring fallbacks. Why. Fifteen pages of undifferentiated preferred-value lists are expensive to translate and add little; a small number of explicit embodiments cost less and work better. Trap. Length treated as a proxy for support.

Phase 16. Inheriting an existing portfolio

[ ] Search the company's own publication history first. Conference programmes, theses, customer bulletins, trade press, regulatory dockets. Why. Self-disclosure is the most common invalidity ground and the easiest to find. Trap. Searching competitors and never searching yourself.
[ ] Map every granted claim to a current product. Why. Unmapped portfolios usually protect discontinued grades and leave the current range uncovered. Trap. Reporting filing counts as coverage.
[ ] Check the assignment chain on the top ten patents. Inventor assignments, corporate transfers, recordation. Why. Gaps are common, cheap to fix while people are still employed, and fatal to enforcement if a co-inventor is missing. Trap. Assuming recordation happened because the deal closed.
[ ] Re-test the secrecy assumptions with an outside laboratory. Why. Instrumentation improves; assumptions do not. Trap. Inheriting a secrecy decision without inheriting its evidence.
[ ] Pull the regulatory confidentiality file and diarise every reassertion. Why. Lapsed claims publish composition silently. Trap. Assuming the regulatory function tracks legal deadlines.
[ ] Read the top ten supply agreements for improvement, residuals, benchmarking, and local registration clauses. Why. This is where value leaves quietly. Trap. Reviewing only the agreements legal negotiated.
[ ] Present findings as a remediation plan, not an audit. Why. Inherited portfolios reflect earlier constraints, and cooperation is needed from the people who made those decisions. Trap. A report that reads as criticism and produces defensiveness instead of assignments.

Phase 17. Substitution and green chemistry programmes

[ ] Expect a structural obviousness problem. Why. Regulatory pressure to replace a restricted substance is itself evidence of a known problem and a known direction of solution, which supports an obvious-to-try argument under KSR International v. Teleflex. Trap. Treating the substitution itself as the invention.
[ ] Locate the invention in unexpected performance, not in avoidance. Why. A replacement that matches or exceeds the restricted substance where the art taught that the restricted moiety was responsible for the performance is genuinely non-obvious. Trap. A specification whose only asserted advantage is regulatory compliance.
[ ] Capture the technical prejudice contemporaneously. Literature and internal assessments saying the restricted component was essential. Why. It is the best available non-obviousness evidence and it exists because people wrote it while opposing the regulation. Trap. Discovering it after the documents were archived or deleted.
[ ] Coordinate regulatory affairs and patent filing. Why. Consultation responses and restriction dockets are publications, and they routinely disclose the substitution being developed. Trap. Two functions that do not speak until national phase entry.
[ ] Review environmental marketing language against the specification. Why. A benefit asserted in the patent will be quoted in a false advertising complaint if the marketing overstates it. Trap. Two teams writing about the same product to different standards of proof.

Phase 18. Training the laboratory

[ ] Teach three habits, not a syllabus. Do not publish before filing; keep the failed experiments; write down why the result surprised you. Why. Those three produce most of the evidence every later dispute needs. Trap. An annual compliance session covering everything and changing nothing.
[ ] Give the chemists a one-page disclosure form that asks what the art taught, what was tried and failed, and what surprised the team. Why. It captures the non-obviousness narrative while it is still remembered. Trap. A form that only asks what the compound is.
[ ] Put one reviewer between the laboratory and every external document. Why. Abstracts, posters, theses, bulletins, and specification sheets all publish, and all are issued without review by default. Trap. A review process so slow that people route around it.

Outcome. A protection decision memo grounded in what an outside laboratory could actually recover; a specification and claim ladder whose examples match its scope; a comparative data file built before the first office action; a disclosure register with live reassertion dates; and a monitoring programme that can produce an analysed, provenance-documented sample on the day a decision to enforce is taken.

The five things people get wrong

One. Filing before the examples exist. The distribution of working examples across the claimed space is the single largest determinant of surviving scope, and it cannot be fixed later. A continuation does not add examples; it inherits the disclosure. Every week spent making analogues at the edges of the intended genus buys more scope than any amount of prosecution argument.

Two. Comparing against the wrong prior art. Comparative data against a distant analogue is worse than none: it tells the examiner and the court that the applicant identified the real comparison and declined to make it. The closest prior art should be identified before the experiments are designed, not after the rejection arrives.

Three. Treating patent and secrecy as a single choice. The composition and the process almost always have different analytical recoverability, and the right answer is usually to patent one and keep the other. A blanket decision either way gives away something for nothing.

Four. Letting the composition out through the side doors. Regulatory filings, safety data sheets, customer qualification packs, conference posters, and technical bulletins each publish composition, and none of them is reviewed by the people who chose secrecy. The disclosure register exists because no single function sees all these channels.

Five. Assuming the range is critical because the claim says so. Criticality is data at the boundary. Without it, an overlapping prior art range makes the claim prima facie obvious under In re Aller, and the internal technical bulletin describing the same range as "typical" will be waiting in the exhibit list.

Key Authorities at a Glance

| Authority | Proposition | |---|---| | 35 U.S.C. § 101 | Patentable subject matter | | 35 U.S.C. § 102 | Anticipation; grace period; inherency | | 35 U.S.C. § 103 | Obviousness | | 35 U.S.C. § 112 | Enablement, written description, definiteness | | 35 U.S.C. § 116 | Joint inventors | | 35 U.S.C. § 154 | Term and adjustment | | 35 U.S.C. § 156 | Patent term extension | | 35 U.S.C. § 262 | Joint owners; no accounting | | 35 U.S.C. § 271 | Infringement; imported products of a patented process | | 35 U.S.C. § 282 | Presumption of validity | | 15 U.S.C. § 2604 | TSCA premanufacture notice | | 15 U.S.C. § 2613 | TSCA confidentiality claims | | 18 U.S.C. § 1833 | Whistleblower immunity notice | | 18 U.S.C. § 1836 | DTSA civil action | | 18 U.S.C. § 1839 | Trade secret definition | | 29 C.F.R. § 1910.1200 | Hazard communication trade secret | | 40 C.F.R. § 720 | TSCA new chemicals | | 37 C.F.R. § 1.75 | Claim form | | 37 C.F.R. § 1.132 | Evidence declarations | | Amgen Inc. v. Sanofi | Full-scope enablement | | Nautilus v. Biosig Instruments | Definiteness | | KSR International v. Teleflex | Obvious to try | | Kewanee Oil v. Bicron | Secrecy not preempted | | Bonito Boats v. Thunder Craft Boats | Reverse engineering legitimate | | Warner-Jenkinson v. Hilton Davis Chemical | Equivalents element by element | | Festo v. Shoketsu Kinzoku Kogyo | Prosecution history estoppel | | Halo Electronics v. Pulse Electronics | Enhanced damages discretion | | Ariad Pharmaceuticals v. Eli Lilly | Written description separate | | In re Wands | Undue experimentation | | In re Dillon | Structural obviousness | | In re Papesch | Properties rebut structure | | In re Petering | Small genus describes species | | In re Aller | Routine optimisation | | Pfizer v. Apotex | Salt screening routine | | Schering v. Geneva Pharmaceuticals | Inherent anticipation | | Atlas Powder v. Ireco | Inherency without appreciation | | Abbott Laboratories v. Sandoz | Product-by-process limits bind | | Idenix Pharmaceuticals v. Gilead Sciences | Genus enablement failure | | Juno Therapeutics v. Kite Pharma | Written description for genus | | E.I. duPont deNemours v. Christopher | Improper means | | TSCA CBI substantiation | Confidentiality claims | | Unexpected results practice | Rebutting obviousness |