Blockchain-Enabled Explainable Nano-Safety (BENS) Architecture for Safer Agricultural Nanotechnology: A Systematic Meta-Analysis of Nanopriming under Abiotic Stress Conditions

This study presents a nine-step Blockchain-Enabled Explainable Nano-Safety (BENS) framework designed to address the current lack of standardised, verifiable, and machine-interpretable data infrastructure that limits regulatory acceptance and safe-by-design (SbD) deployment of agricultural nanotechnology using nanomaterial-based seed, foliar, and soil interventions. The framework was validated against a curated dataset of 207 nanopriming experiments across 27 nanomaterials, 50 plant species, and 11 abiotic stress categories. The dataset was structured for blockchain readiness, cryptographically fingerprinted using SHA-256 hashing, and subjected to smart-contract-encoded SbD rules. AI-driven explainability (X-Nano / XAI) was integrated with the BENS pipeline to generate auditable Nano Safety Passports (NSPs). Each NSP is a machine-readable, blockchain-anchored record that couples a composite risk score with SHAP-derived feature-level justification for a given nanomaterial-stress combination, so that a regulator or laboratory can trace why a formulation was classified as Approved, Flagged, or Restricted and verify that the underlying data have not been altered since submission. Statistical analyses including one-way ANOVA (F = 6.445, p < 0.0001), Pearson correlation, and hierarchical clustering were employed. Silver nanoparticles (Ag, n = 35), ZnO (n = 19), and Fe NPs (n = 18) dominated the dataset. Salinity (41.1%) and drought (31.4%) were the primary target stresses. Nanoparticle sizes ranged 2–103 nm (mean: 31.1 ± 22.3 nm) with effective concentrations of 0–2000 nM (mean: 82.2 ± 188.0 nM). The BENS architecture successfully generated NSPs with blockchain-verified risk scores and OECD/ISO-compliant SbD classifications. The BENS framework provides the first comprehensive, computationally verifiable infrastructure for agricultural nanomaterial risk management framework, enabling transparent multi-laboratory data sharing and regulatory-ready explainable AI classification of nanopriming efficacy.