Non-enzymatic detection of glucose-ketone ratio via formation of different floral honeys-capped silver nanoparticles and iron-acetate complexes on 3D-connector microfluidic paper-based analytical devices

• Integrated detection system aids ketogenic diet management via glucose-ketone ratio (GKR). • 3D-µPADs enhance the accuracy and precision of GKR detection. • Glucose detection based on the formation of silver nanoparticles (AgNPs) stabilized by various floral honeys, while ketone detection integrates Schiff-base and iron-acetate complex formation. • Method validation, which includes precision, accuracy, LOD, and LOQ, ensures reliable GKR detection. Obesity is a significant risk factor for diabetes mellitus, a metabolic disorder often managed through ketogenic diets that enhance ketogenesis. However, excess ketone accumulation can lead to ketoacidosis, a serious complication in diabetes. Accurate monitoring of the glucose-ketone ratio (GKR) is crucial for evaluating ketogenic diets. This study introduces the 3D-Connector microfluidic paper-based analytical devices (3D-µPADs) for GKR measurement in artificial urine, providing enhanced accuracy and precision. The 3D-Connector innovation reduces capillary diffusion, minimizing cross-contamination between samples and reagents while optimizing analyte distribution. The device supports non-enzymatic glucose and ketone (acetoacetate) detection; glucose is assessed through silver nanoparticles (AgNPs) synthesis stabilized by various floral honey types, and acetoacetate is measured via iron-acetate complex formation. This study further investigates the efficiency of different floral honey as capping agents for AgNPs formation, aiming to improve the accuracy and reproducibility of GKR detection. AgNPs are characterized using ATR-FTIR and UV-Vis spectrophotometry, while ImageJ is utilized to quantify colorimetric reaction distances on µPADs. Optimized parameters identified rambutan honey as the most effective for GKR detection on 3D-µPADs, with optimum conditions of 10% (v/v) honey, 60 mM silver nitrate, 1 M glycine in pH 9.4 phosphate buffer, and a 15-minute reaction time. Validation of this method confirmed excellent linearity (R² = 0.9971), high precision (RSD = 2.85%) and accuracy (93.40 - 99.83%), with limits of detection and quantification of 5.16 and 15.63 mM/dM, respectively.