Ai-designed phase-change and endothermic cooling materials for diaper comfort
AI-native materials informatics program that designs cooling materials for the diaper microclimate, targeting 1–3 °C skin-interface temperature reduction while preserving softness, rewet, and breathability. The platform ranks candidate formulations using phase-change, thermally conductive, moisture-responsive, and endothermic mechanisms, and validates prototypes through thermal, rewet, breathability, and safety screening.
Overview
This AI-native materials-informatics program targets a well-known hygiene-product challenge: cooling the diaper microclimate without degrading softness, rewet performance, breathability, cost, or manufacturability. The approach integrates computational materials design with experimental prototyping to deliver ranked, testable thermal-management options.
The program considers four cooling mechanisms: phase-change materials (latent-heat absorption), thermally conductive fillers (heat spreading), moisture-responsive polymers (behavior change in wet conditions), and endothermic formulations (heat absorption through dissolution). Because each mechanism affects multiple comfort and manufacturing properties, the platform treats the problem as a multi-objective optimization and ranks potential candidates according to cooling magnitude, latent-heat buffering, safety, processability, cost, and comfort trade-offs. The output is a down-selected manufacturable solution plus a reusable prediction model.
Technical specifications
The program uses the Polymerize One AI-native R&D stack, which combines a unified data management (“Cortex”), domain-specific materials models, and partner-laboratory prototyping. It follows a design–make–test–learn loop:
- In silico design: Polymer models incorporate phase-change, thermally conductive, moisture-responsive, and endothermic components into candidate formulations and rank them against target cooling (approximately 1–3 ×C), heat flux, latent-heat behavior, rewet, breathability, cost, and safety guardrail.
- Prototyping and characterization: top-ranked candidates are turned into functional layers or diaper mock-ups; tests include simulated wet or humid insult with thermal skin-interface measurements, rewet evaluation, and breathability measurement.
- Down-selection and de-risking: the lead formulation is refined, early safety and migration checks are run, and converting/cost constraints are addressed to arrive at a manufacturable candidate.
The use of multi-opposed models allows the platform to navigate softness/cost/breathability trade-offs that typically make cooling additions difficult, and can generate explainable recommendations for informed R&D decisions.
Technology readiness level
The platform itself has already delivered notable computational and laboratory validation: one end-to-end phase-change material development was completed in a 65% faster timeline, and an analogous multi-property formulation refinement (paint-cohesion) reduced prediction error from 16% to 8% across 50+ ingredients and 15+ properties. This indicates that the model-design methodology is demonstrated for adjacent materials challenges.
The specific diaper-microclimate thermal comfort technology is at early TRL 3–4 stage. It is being developed for a wet/humid insult protocol, with prototype-side mock-ups to be measured for cooling, thermal flux, rewet, and breathability before manufacturing scale-up. The program is not yet at production-level TRL; it requires a validation collaboration to complete down-selection, safety screens, and scale confirmation.
About Polymerize
Polymerize is an AI-native platform specifically engineered to transform materials research and development. It operates as a system of intelligence that integrates a unified data foundation, domain-specific AI models, and explainable insights to accelerate the discovery, development, and scaling of new materials. By digitizing experimental workflows, the platform replaces manual, spreadsheet-based processes with a centralized, data-driven backbone. The solution features over 35 domain-guided models trained on polymers, chemicals, and advanced materials, which allow R&D teams to predict material properties, optimize formulations, and gain scientific explanations for AI-driven recommendations.
The platform serves global R&D teams across industries such as automotive, specialty chemicals, coatings, and packaging, helping them reduce failed experiments and shorten go-to-market timelines. By capturing and leveraging institutional knowledge, Polymerize enables organizations to turn years of trial-and-error into targeted innovation, ensuring that data from past experiments informs future research. Customers use the platform to achieve significant ROI through increased discovery speed and streamlined operational forecasting, with security compliance standards like SOC 2 and ISO 27001 integrated to protect sensitive intellectual property. Founded in 2020 and headquartered in Singapore, the company maintains a global presence with offices and strategic collaborations spanning Asia, Europe, and the United States.
delivered to your inbox.