Fig. 2: Design and characterization of the wearable device.

a Schematic illustrating the mechanism of glucose measurement by CA and Na⁺/pH detection by OCP. CA, chronoamperometry; OCP, open-circuit potential. b, c Layer assembly (b) and schematic design (c) of the wearable device. d Optical images of the fabricated device on the forearm skin. Scale bars, 1 cm. e, f Stepwise testing of the glucose sensor from (e), and i-t curves with the corresponding correlation curve (inset) from 0 to 3 mM at an initial potential of −0.1 V versus RE (f) (n = 3 technical replicates). g Selectivity performance of the glucose sensor at low concentrations. h Storage stability of the glucose sensor over 14 days (n = 3 independent sensors). i Sensitivity of the glucose sensor under varying pH conditions, performed in Tris-HCl buffer solutions with pH values of 5, 6, 7, 8, 9, and 10 (n = 5 technical replicates). pH 5: P = 0.0087 (vs. pH 6), P = 0.0061 (vs. pH 7), P < 0.0001 (vs. pH 9 or 10); pH 6: P = 0.0013 (vs. pH 8), P < 0.0001 (vs. pH 9 or 10); pH 7: P = 0.0009 (vs. pH 8), P < 0.0001 (vs. pH 9 or 10); pH 8: P < 0.0001 (vs. pH 9 or 10). Statistical analysis was performed using one-way ANOVA with Tukey’s multiple comparison. **P < 0.01, ***P < 0.001, and ****P < 0.0001. j Calibration curve of glucose sensitivity as a function of pH. Glucose sensitivity was normalized at pH 7, revealing a quadratic relationship between glucose sensitivity and pH (n = 5 technical replicates). S, sensitivity. k, l Glucose detection error with and without pH correction at pH 5 (k) and 10 (l). m, q Reversibility of the Na⁺ sensor (m) and pH sensor (q). n, r Potentiometric responses and corresponding correlation curve (inset) of the Na⁺ sensor (n) and pH sensor (r) (n = 3 technical replicates). o, s Selectivity of the Na⁺ sensor (o) and pH sensor (s) under various interferences. p, t Storage stability of the Na⁺ sensor (p) and pH sensor (t) over 14 days (n = 3 independent sensors). Data are presented as mean values  ±  SD.