Experimental Use Exception

The experimental use exception permits inventors to test and refine inventions publicly without triggering the § 102 prior art bar: STATUTORY CONTEXT: 35 U.S.C. § 102(b)(1) (AIA) bars patents when the invention was publicly disclosed (including on-sale or public use) more than one year before the effective filing date; the exception is judge-made; it is not explicitly stated in the statute; FOUNDATIONAL CASE: City of Elizabeth v. American Nicholson Pavement Co. (S.Ct. 1877): the inventor installed an experimental road pavement section on a public street and observed it for several years; the Supreme Court held this was not a 'public use' that triggered the bar because the purpose was experimental — to test the invention in its intended use environment; PRINCIPLE: 'An inventor may continue experiments in perfecting his invention, making them as public as the nature of the invention requires, without losing his right to a patent. The use of an invention by the inventor himself, or by another person under his direction, by way of experiment, and in order to bring the invention to perfection, has never been regarded as [a public use]'; EXPERIMENTAL vs. COMMERCIAL USE: the exception requires that the primary purpose of the public use or sale was experimentation, not commercial exploitation; if the primary purpose was to make money, test market demand, or establish commercial relationships, the exception does not apply; APPLIES TO BOTH PUBLIC USE AND ON-SALE BARS: the exception can negate both the § 102 public use bar (using the invention publicly) and the § 102 on-sale bar (selling or offering for sale the invention), though courts look at different factors for each.

Courts evaluate a multi-factor test to determine whether use was experimental: ALLEN ENGINEERING CORP. v. BARTELL INDUSTRIES (Fed. Cir. 2002) and similar cases identify these factors: FACTORS SUPPORTING EXPERIMENTAL USE: (1) the necessity of public testing — the invention cannot be tested except through actual use in its intended environment; (2) the inventor monitored and controlled the tests; (3) the inventor received test reports and feedback; (4) the test units were labeled or marked as experimental; (5) confidentiality agreements were in place with test users; (6) payment was not received (or only cost recovery) for the test units; (7) the inventor continued to refine the invention based on test results; (8) there were features of the invention that could not be determined without real-world testing; FACTORS UNDERMINING EXPERIMENTAL USE: (1) the invention was offered for sale commercially (including price quotes, purchase orders, or commercial agreements); (2) the test was primarily to test market demand or commercial acceptability, not technical performance; (3) the inventor had no feedback mechanism or control over the testing; (4) the invention was fully reduced to practice and ready for commercial sale before testing; (5) commercial sales occurred simultaneously or shortly after testing; (6) the testing occurred in the normal course of commerce without any experimental notation; PFAFF INTERACTION: the Pfaff v. Wells Electronics (S.Ct. 1998) on-sale bar test requires that the invention be 'ready for patenting' (reduced to practice or depicted in drawings/written description) — if the invention is not yet ready for patenting, experimental testing may continue without triggering the bar even without the experimental use exception.

The Pfaff decision redefined the on-sale bar and affects how the experimental use exception operates: PFAFF v. WELLS ELECTRONICS (S.Ct. 1998): the on-sale bar under § 102(b) (pre-AIA) / § 102(a)(1) (AIA) is triggered when, before the critical date: (a) the claimed invention was the subject of a commercial offer for sale; AND (b) the invention was 'ready for patenting' (reduced to practice OR described sufficiently to enable a POSITA to practice it); PRE-PFAFF EXPERIMENTAL USE: before Pfaff, experimental use was a key exception to the on-sale bar even if there was a commercial offer for sale; POST-PFAFF INTERACTION: if the invention is NOT yet ready for patenting, there is no triggering on-sale bar even without the experimental use exception; if the invention IS ready for patenting AND there was a commercial offer for sale, then experimental use analysis matters: the commercial nature of the offer is significant; if the offer was truly commercial (not a test arrangement), experimental use does not negate the bar; HELSINN HEALTHCARE v. TEVA PHARMACEUTICALS (S.Ct. 2019): a CONFIDENTIAL commercial sale agreement triggered the on-sale bar even though the sale was not publicly known; the AIA's 'available to the public' language did not change the secret sale rule for the on-sale bar; PRACTICAL TENSION: a company that sells products under a strict confidentiality agreement for clinical testing purposes — is this experimental use or a commercial sale? Post-Helsinn, a commercial sale is on-sale bar even if confidential; experimental use analysis still applies if the testing was the primary purpose.

There is a separate, narrow common law experimental use exception to patent infringement — distinct from the prosecution-stage experimental use exception: EXPERIMENTAL USE AS AN INFRINGEMENT EXCEPTION: Whittemore v. Cutter (Story, C.C.D. Mass. 1813) suggested that making a patented invention solely for experimental purposes was not infringement; this is the common law 'research exemption' from infringement (not the § 102 exception); CURRENT STATUS — MADEY v. DUKE UNIVERSITY (Fed. Cir. 2002): the Federal Circuit dramatically narrowed the common law research exemption; held that even a non-profit research university conducting fundamental scientific research with a patented laser does not qualify for the experimental use exemption from infringement; the test: 'any use which is in keeping with the legitimate business of the infringer' is NOT experimental; a university's legitimate business includes teaching and research — even non-commercial fundamental research; PRACTICAL RESULT: the common law research exemption from infringement is essentially non-existent post-Madey; virtually any use of a patented invention, even in academic research, is infringement; STATUTORY RESEARCH EXEMPTION: § 271(e)(1) — the Hatch-Waxman research exemption: making, using, or selling a patented invention 'solely for uses reasonably related to the development and submission of information' to the FDA is not infringement; this is the 'safe harbor' for generic drug testing; Merck KGaA v. Integra Lifesciences (S.Ct. 2005): the § 271(e)(1) exemption is broader than FDA submission activities — covers preclinical research that might lead to FDA submissions; DISTINCTION: the § 102 experimental use exception (prosecution bar) and the common law research exemption (infringement defense) are SEPARATE doctrines with separate requirements.

To rely on the experimental use exception during prosecution, inventors should document the experimental nature of pre-filing testing from the outset: DOCUMENT EXPERIMENTAL PURPOSE: maintain written records showing the purpose of each test was to perfect the invention, not sell it commercially; keep test logs, engineering notebooks, data sheets, and design iteration records; CONTROL AND MONITOR TESTS: assign each test unit a tracking number; require testers to return reports, provide feedback, and return units; the more control the inventor maintains, the stronger the experimental use argument; USE CONFIDENTIALITY AND EXPERIMENTAL AGREEMENTS: have testers sign confidentiality agreements identifying the product as a test prototype; include language stating that the testing is to perfect the invention, not for commercial use; agreements should require feedback and may require return of the test unit; AVOID COMMERCIAL INDICIA: do not set a commercial price, issue invoices (or issue only cost-recovery invoices with experimental labels), take purchase orders, or enter supply agreements during the experimental period; if money changes hands, document it as cost recovery for testing, not a sale; KEEP TESTING PERIOD REASONABLE: courts look at whether the length of testing was consistent with legitimate experimentation; a 10-year 'test period' with no technical modifications raises doubt; periodic design iterations and documented improvements support the exception; PROSECUTE WITHIN ONE YEAR OF REDUCTION TO PRACTICE: when the invention is fully reduced to practice (it works for its intended purpose), the experimental phase has typically ended; file the patent application promptly — the one-year clock starts when experimental use ends; KEEP RECORDS TO SHOW IN PROSECUTION: if an examiner or later challenger cites experimental activities as prior art, you need to produce contemporaneous records proving the experimental nature; documents created after the fact are less persuasive.