A Molecularly Imprinted Polymer-Based Porous Silicon Optical Sensor for Quercetin Detection in Wines - Supporting Information

S1 SUPPORTING INFORMATION A Molecularly Imprinted Polymer-based Porous Silicon Optical Sensor for Quercetin Detection in Wines Tiziano Di Giulioa,§, Ibrar Muhammad Asifa,§, Martina Corsib, Soumya Rajpalc, Boris Mizaikoffc, d, Nicoletta Ditarantoe, Giuseppe E. De Benedettof, Cosimino Malitestaa, Giuseppe Barillarob,*, Elisabetta Mazzottaa,* a Laboratory of Analytical Chemistry, Department of Biological and Environmental Sciences and Technologies (Di.S.Te.B.A.), University of Salento, via Monteroni, 73100 Lecce, Italy b Information Engineering Department, University of Pisa, via G. Caruso 16, 56122 Pisa, Italy c Institute of Analytical and Bioanalytical Chemistry, Ulm University, Albert-Einstein-Allee 11, 89081 Ulm, Germany d Hahn-Schickard, Sedanstrasse 14, 89077 Ulm e Chemistry Department, Aldo Moro University of Bari, Via Orabona 4, 70126 Bari, Italy f Laboratory of Analytical Mass Spectrometry, Cultural Heritage Department, University of Salento, Via Monteroni, 73100 Lecce, Italy E-mail: [email protected], [email protected] § T.D.G and M.I.A. contributed equally to this work S2 Figure S1. QU and Py molecules showing Mulliken charges (A) and molecular electrostatic potential mapping (B). The colour denotes the reactivity of the region; blue is most positive electrostatic potential, red is most electronegative electrostatic potential and green is for neutral potential. S3 Figure S2. 3D representation of an amorphous polypyrrole (PPy) matrix in CPK representation, surrounded by solvent molecules (water-ethanol, 4:1) in wireframe format, highlighting the polymersolvent system within the simulated box. S4 Figure S3. A) Reflectance spectra recorded in air on a PSiO2 scaffold during different functionalization steps: bare PSiO2 (black line), silanization with APTES (red line) and PSiO2 exposure to CDI (green line). B) Reflectance spectra recorded in air on a PSiO2 scaffold after polymer deposition using different polymerization times: 30 min (blue line), 1h (green line) and 2 hs (red line). C) Effective optical thickness changes (EOT-EOT0) achieved after polymer deposition for different time periods, namely 30 min, 1 h and 2 hs and after the washing procedures to obtain the MIPs; the EOT value recorded after quercetin anchoring on PSiO2 scaffold (EOT0) is used as reference (n=3 samples). Data are presented as mean (± s.d). S5 Figure S4. Effective optical thickness changes (EOT-EOTPSiO2) achieved for each functionalization step of A) MIPand B) NIP-sensors; the EOT value of bare PSiO2 (EOTPsiO2) scaffold is used as reference (n=3 samples). All data are presented as mean (± s.d). S6 Figure S5. Detailed C 1s signals recorded after A) PSiO2 functionalization with APTES, C) quercetin anchoring and E) PSiO2 scaffolds exposed to CDI. High-definition N 1s spectra recorded after B) PSiO2 functionalization with APTES, D) quercetin anchoring and F) PSiO2 scaffolds exposed to CDI. Spectra are fitted and charging corrected. S7 Figure S6. A) N/Si and B) C/Si atomic ratio calculated from XPS analysis of PSiO scaffold after each functionalization step up to NIP/MIP deposition. For all the samples, three different measurement positions were analyzed (n = 3). Data are presented as mean (±s.d.). S8 Figure S7. A) Calibration curves (EOT-EOT0 vs quercetin concentration) recorded on MIP-sensors prepared using different deposition times, namely 30 minutes, 1 hour and 2 hours. The sensors were used in quercetin detection tests using standard solutions prepared in water (from 2.5 to 20 M). EOT0 is measured in buffer solution without quercetin and used as reference (n=3 samples). Data are presented as mean (± s.d). B) Comparison of responses (EOT-EOT0 vs quercetin concentration) of MIP-sensors, obtained using a deposition time of 30 minutes, to quercetin solutions prepared in ultrapure water (blue column) and water/EtOH (4:1, v/v) mixture (red columns). EOT0 is measured in buffer solution without quercetin and used as reference (n=3 samples). Data are presented as mean (± s.d). S9 050 100 150 200 0 10 20 30 40 50 60 70 80 B=aCm R2=0.9816 a=4.3 m=0.53 experimental data Freundlich model EOT-EOT0 (nm) [Quercetin] (M) Figure S8. Best-fitting of the MIP sensor calibration curve (black dots) using the Freundlich isotherm model (blue trace), where B = EOT-EOT0 is the sensor output and C is quercetin concentration in solution. Fitting parameters are a=4.3 related to the median binding affinity and m=0.53, the heterogeneity index. Experimental data are presented as the mean value of n = 3 samples, with error bars representing the standard deviation.