This week we look at a recent PTAB decision. I pulled one random case from the Board’s docket published over the last few days, where IP Copilot was not used.

---

In Ex parte Stuhrmann (Appeal 2025‑000356), the PTAB affirmed §103 rejections on claims to a photocurable ceramic formulation for additive manufacturing (see decision here). The Board leaned on three references: Cheng for particle size, Halloran for stereolithography resin principles, and Rantala for viscosity optimization. The themes were familiar—overlapping ranges and result‑effective variables—and the Board found no persuasive unexpected results.

---

What made the difference at the PTAB

Two doctrinal levers drove the affirmance.

First, the Board reiterated that the “selection of any order of performing process steps is prima facie obvious” absent contrary evidence.

Second, the record showed overlap (Cheng’s ≤10 μm average particle size sits inside the claimed 4–10 μm band) and routine optimization (viscosity), so the claimed parameters were viewed as tunable within known ranges.

Where IP Copilot could have helped — speed

Searching for invalidity searches IP Copilot takes about 20 seconds — pick the application, check the priority date, set the number of references and generate the invalidity search. Easy.

The results did end up taking ~17 minutes and we had some very good hits. Our scoring system found multiple references that are highly relevant 1 critical, 4 high and 1 medium —

While the top results we found do differ from the references presented at the PTAB, they do appear to be a relatively strong hit. Looking at the details, the results / quotes seem fairly on point.

Doing a more detailed review — IP Copilot’s Invalidity Chart can assemble, in one pass, the constituent‑level map of a photocurable ceramic formulation: acrylates (B1/B2), dispersants (C), photoinitiators (D), ceramic fillers (A), and mixing. In our 15‑minute test run, the chart surfaced mature photosensitive‑ceramic coating art (e.g., US 4,912,019; US 4,908,296; JP 4620705) and even Halloran’s stereolithography resin—handy to frame result‑effective variable arguments on viscosity and solids loading. Halloran, for instance, discusses ceramic SLA resins with “viscosity of less than 3000 mPa·s”—a concise anchor when arguing tunability.

Where the quick pass initially under‑delivered was on the two other Board‑relied references: Cheng (WO 2017/045191) and Rantala (US 2019/0061236). Cheng is particularly potent because it speaks the modern AM slurry dialect and explicitly teaches an average ceramic particle size ≤10 μm, a near‑bullseye for the claims’ D50 = 4–10 μm discussion. Rantala, meanwhile, situates viscosity squarely as a design variable for 3D‑printing materials—exactly how the Board treated it.

---

The 15‑minute IP Copilot playbook

Our practice inside IP Copilot is a two‑step sprint:

1. Breadth first — Run Invalidity Chart on the target claim set to lock down who/what/where for the constituents and any obvious knobs (particle size metrics like D50, common photoinitiators, acrylate families, dispersants). This creates a matrix of elements with citations you can carry into an Office Action response, appeal brief, or motion practice.
2. Then focus — Follow immediately with a limitation‑focused search (still inside IP Copilot) on the parameters that will be optimized and their measurement methods: here, Brookfield viscosity (spindle 64, 6 rpm, 25 °C) and the CD:S mixing ratio (75–99 : 1–25 wt %). You can also update the feature descriptions to optimize the search and focus on individual limitations we missed.

Proof that it’s practical

As we’ve written in our Weekly PTAB Reviews [1], setting up an Invalidity Chart is fast: “Setup takes seconds; this job estimated ~20 minutes.” That cadence is typical in our experience and lets teams triage quickly before investing deep attorney hours.

Takeaway

In this case, the PTAB’s logic rested on order‑of‑steps obviousness and optimization of known variables—exactly the patterns IP Copilot can surface in minutes by pulling together constituent‑level art (e.g., Halloran) and, with a targeted follow‑on, the AM‑specific range/measurement art (e.g., Cheng, Rantala). A human searcher ultimately decides what to argue—but with a 15‑minute IP Copilot pass, that person starts with the right playbook and the right exhibits in hand.