Solid-State Electrolyte Patent Data Processing
Structure the formulations, process steps, characterization data, and performance metrics in solid-state electrolyte patents into a comparable database. The trouble with this kind of data isn't extraction—it's comparability: the same ionic conductivity means something entirely different depending on the test temperature and the fabrication route.
Where This Kind of Task Gets Hard
The comparability problem with formulation data1. Formulation bases aren't consistent
Within the same batch of patents, component contents may be expressed as wt%, mol%, at%, or even volume ratios; minor components are often given as "balance" or a range value. These notations are not sloppy—they are the applicant's definition of the formulation boundary. Forcing them into a unified number actually throws that layer of information away.
2. Process and performance are tightly coupled
Sintering temperature, pressing pressure, annealing time, atmosphere conditions—change any one of them and the final performance changes. If you extract the performance numbers without the corresponding process, the data can't answer "why is this formulation better".
3. Performance data often lacks conditions
Ionic conductivity is only meaningful with a test temperature—values at room temperature and at high temperature can differ by several orders of magnitude. The same applies to activation energy and the electrochemical window. When conditions are missing, the value is still "true", but it isn't comparable.
4. Characterization bases are inconsistent
Whether particle size is given as D50 or D90, whether relative density is referenced to the theoretical or the measured value, and how XRD peaks are labeled—each patent writes it differently. A data point with an unclear basis becomes noise, not a sample, once it enters the database.
What We Cover
Formulation · Process · Characterization · Performance- Basic information: material system, chemical formula, applicant, publication number, and filing date
- Components and contents: original wt% / mol% / at% bases preserved, including range values and balance
- Process steps and parameters: temperature, time, pressure, atmosphere, heating/cooling rates
- Step order numbered as in the original; not merged and not rewritten
- Characterization data: XRD, SEM, particle size distribution, relative density
- Performance metrics: ionic conductivity, activation energy, electrochemical window, cycling data
- Test conditions: temperature, pressure, cell configuration, and test method
- Source location: publication number, paragraph number, table number, and figure number
Formulation, process, and performance are split into three tables that can be linked to each other, joined through the publication number and example number, so you can either query one type of data on its own or reconstruct the complete record of "a formulation + a process + a set of performance data".
Delivery Pipeline
Four steps in one line · fully loggedDeliverables
What you get is not just an annotation result- Structured dataset: JSON / CSV / Excel, with formulation, process, and performance split into linked tables; fields corresponding one-to-one to the Schema
- Annotation guidelines document: field definitions, edge cases, acceptance criteria
- QC report: sampling ratio, error categories, rework records
- Provenance log: each data point maps to its publication number and paragraph location, annotator, and modification time
Related Cases
Real delivery scenariosStructuring of solid-state electrolyte basic information, process steps, characterization data, and performance metrics, covering patent data in the new materials domain.
Solid-state electrolytesPatent processingPerformance metricsSimilar materials patent processing also covers annotation of material entities, elements/components, weight percentages, performance attributes, and fabrication processes. The two workflows share the same Schema design approach; the difference lies in the field bases.
Frequently Asked Questions
Questions we're often asked before working togetherWill wt% and mol% be unified into a single basis?
No. The original bases are preserved as-is and labeled separately, without forced conversion. Converting between the two requires the molecular weight of each component, which patents don't necessarily provide—forcing a conversion would introduce untraceable error, so keeping the original basis is actually more rigorous.
How are "balance" and range values handled?
Recorded as a range or balance exactly as in the original, with no conversion to point values. This kind of notation itself embodies the applicant's definition of the formulation boundary; turning "balance" into a specific number would lose that layer of information.
What if performance data doesn't state its test conditions?
Recorded faithfully as missing conditions, with no imputation. Ionic conductivity can differ by several orders of magnitude at different temperatures; without the test temperature the value isn't comparable—flagging that is more valuable than filling in a plausible value.
How do you guarantee the order of the process steps?
Steps are numbered in the original order and the original wording is retained; similar steps are not merged. The order itself is often one of the patent's points of protection, and merging would destroy that information.
What formats do you output?
Delivered as JSON / CSV / Excel, with formulation, process, and performance split into linked tables joined through the publication number and example number, and fields corresponding one-to-one to the co-designed Schema.
Continue Reading
Related pagesStructure formulas and formulations in materials patents often appear on the same page, so the two workflows can be run together.
Browse all the solutions we currently offer by industry domain and specific task.
Send us a representative patent family, and we can run a small pilot annotation batch so you can see directly how formulation bases and test conditions are handled.