Wireless and Flexible Optoelectronic System for In Situ Monitoring of Vaginal pH Using a Bioresorbable Fluorescence Sensor - Supporting Information

Supporting Information for Adv. Mater. Technol., DOI: 10.1002/admt.202201600 Wireless and Flexible Optoelectronic System for In Situ Monitoring of Vaginal pH Using a Bioresorbable Fluorescence Sensor Alessandro Paghi, Martina Corsi, Antonino A. La Mattina, Gabriella Egri, Lars Dähne, and Giuseppe Barillaro* Supporting Information Wireless and Flexible Optoelectronic System for In-Situ Monitoring of Vaginal pH Using a Bioresorbable Fluorescence Sensor Alessandro Paghi1+, Martina Corsi1+, Antonino A. La Mattina1, Gabriella Egri2, Lars Dähne2, and Giuseppe Barillaro1* 1Dipartimento di Ingegneria dell’Informazione, Università di Pisa, via G. Caruso 16, 56122 Pisa, Italy. 2Surflay Nanotec GmbH, Max-Planck-Straße 3, 12489 Berlin, Germany. +These authors contributed equally to this work. *Corresponding author: [email protected] Experimental Section Materials and chemicals Single side polished silicon boron-doped wafers (p++ type) with resistivity of 0.8 - 1.2 mΩ×cm, orientation <100>, thickness of 500 - 550 µm, were purchased from Siltronix Silicon Technologies (France). Oxidized single side polished p+-type silicon wafers with resistivity of 100 mΩ×cm, orientation <100>, thickness of 625 µm, and a 110-nm-thick silicon dioxide layer on top, were provided by STMicroelectronics (Italy). Aqueous hydrofluoric acid (HF, 48%), sodium hydroxide (NaOH, 98%), aqueous hydrochloric acid (HCl, 37%), sodium acetate (CH3COONa, 99%), sodium phosphate monobasic monohydrate (PBS, 98%), tris(hydroxymethyl)aminomethane (TRIS, 99%), sodium chloride (NaCl, 99%), glacial acetic acid (AcOH, 99.7%), D-(+)-Glucose (C6H12O6 ≥ 99.5%), and bovine serum albumin from bovine serum (BSA, ≥ 98%), were purchased from Sigma Aldrich (Germany). Absolute ethanol (EtOH, 99.9%), isopropyl alcohol (IPA, 99.5%), diethyl ether (Et2O, > 99%), glycerol (99.5%), and potassium hydroxide (KOH, 99 %), were purchased from Carlo Erba Reagents (Italy). Calcium chloride (CaCl2, 95 %) was purchased from Fisher Scientific. Modified-rhodamine labelled poly-allylamine-hydrochloride (PAH:Fr (1:135), Mw = 90,000) and modified-rhodamine labelled poly-methacrylic-acid (PMAA:Fr (1:219), Mw = 100,000) were provided by Surflay Nanotec (Germany). Polydimethylsiloxane (PDMS, Sylgard 184) base and thermal curing agent were purchased from Cecchi S.r.l. (Italy). Polyethylene terephthalate (PET, MELINEX 506) was provided by DuPont Teijin Films (UK). 2 N skin toughness synthetic human tissues were purchased from SynDaver (USA). Aqueous solutions were prepared using deionized water (DIW, 18.2 MΩ×cm) filtered by Elix® (Merck Millipore, Germany) and Milli-Q® Academic (Merck Millipore, Germany). All buffers were prepared in deionized water (DIW) and pH-adjusted with NaOH (5 M) and HCl (1 M) aqueous solution. Preparation of porous silica (PSiO2) scaffold on bulk silicon Porous silicon (PSi) layers were prepared by room temperature (RT) anodic etching of p++-type silicon samples (1.5 cm × 1.5 cm) using 2 mL of HF:EtOH (3:1 v/v) solution under ambient light. A custom-made Teflon cell with platinum wire cathode and aluminium flat anode was used for the electrochemical anodization of the silicon sample over a circular area of 0.567 cm2. A source measurement unit (SMU 2602A, Keithley, USA) was used to set the etching current and measure the voltage between anode and cathode. A first sacrificial PSi layer was prepared at 300 mA cm-2 for 20 s and, after rinsing of the sample with EtOH for 120 s to remove residual HF, fully dissolved in a solution of NaOH(1M):EtOH (9:1 v/v) for 120 s [1]. The sample was rinsed with DIW and EtOH, then a second PSi layer (i.e., the porous scaffold) with thickness of ~4.5 µm and porosity of ~76.5 % was etched at constant current density of 300 mA cm-2 for 45 s. Eventually, the sample was rinsed with EtOH for 120 s and Et2O for 60 s to obtain a crack-free PSi layer. The as-prepared PSi layer was thermally oxidized to porous silica (PSiO2) in a muffle furnace (ZB/1, ASAL, Italy) at 1000 °C for 5 min in ambient air. Preparation of PDMS slabs PDMS slabs were prepared by mixing base and curing agents (10:1 by weight). A mass of 12 g of the mix was poured in a polystyrene Petri dish (9 cm in diameter), vacuumed for 30 min using a rotation pump to remove air bubbles, then thermally cured in the ventilated oven (G-Therm 035, Fratelli Galli, Italy) at 90°C for 1 hour. After curing, square PDMS slabs (1.5 cm × 2.5 cm, thickness 2 mm) were cut with a razor blade. Before use, the slabs were sonicated in EtOH for 5 min to remove excess of curing agent from the PDMS surface. Preparation of PSiO2 scaffold on PDMS slab PSi layers were prepared using the same setup detailed in the section “Preparation of porous silica (PSiO2) scaffold on bulk silicon”. A first sacrificial PSi layer was prepared at 300 mA cm-2 for 20 s and, after rinsing of the sample with EtOH for 120 s to remove residual HF, fully dissolved in a solution of NaOH(1M):EtOH (9:1 v/v) for 120 s [1]. The sample was rinsed with DIW and EtOH, then a second PSi layer with thickness of ~100 nm and porosity of ~61.5 % was etched at constant current density of 5 mA cm-2 for 40 s to obtain a barrier layer inhibiting diffusion of PDMS oligomers inside the porous scaffold after the transfer-printing procedure, as previously reported [2]. A third PSi layer (i.e., the porous scaffold) with thickness of ~4.5 µm and porosity of ~76.5 % was etched underneath the barrier layer at constant current density of 300 mA cm-2 for 45 s. Eventually, the sample was rinsed with EtOH for 120 s and Et2O for 60 s to obtain a crack-free PSi stack. The PSi stack was transfer-printed on a flexible PDMS slab following the method reported by Wan et al. [3]. First, the as-etched sample was extracted from the Teflon cell and the edge of the PSi stack scratched with a diamond tip to separate it from lateral bulk silicon; second, the soscratched sample was inserted again in the Teflon cell and the PSi stack was electrochemically detached from the native substrate by electropolishing of the bulk silicon underneath the PSi stack, upon application of a constant current density of 800 mA cm-2 for 0.1 s in 1 mL of a HF:EtOH (1:1 v/v) solution. The HF concentration of the solution was reduced with respect to that used for the etching of the PSi stack to lower the electropolishing current density [4]. Once detached, the freestanding PSi membrane was gently rinsed in EtOH and in Et2O, then glided (barrier layer on top) onto the non-polished side of a silicon chip and topped with a second non-polished silicon chip (in contact with the barrier layer). Eventually, the PSi membrane sandwiched between the two silicon samples was oxidized to PSiO2 in a muffle furnace (ZB/1, ASAL, Italy) at 1000 °C for 5 min in ambient air. After oxidation, the top silicon sample was removed, and the membrane transfer printed on a PDMS slab by gently pressing the slab on the barrier layer on top of the PSiO2 membrane. Eventually, the PDMS-PSiO2 assembly was lifted off the silicon sample at the bottom. Layer-by-Layer (LbL) coating of PSiO2 scaffolds with Rhodamine-labelled polyelectrolytes PSiO2 scaffolds on bulk silicon and PDMS slab were conformably coated with a multilayer stack of two rhodamine (Fr)-labelled polyelectrolytes with opposite charge leveraging electrostatic layer-bylayer (LbL) assembly [5][6], namely, poly-allylamine-hydrochloride (PAH:Fr) and polymethacrylic-acid (PMAA:Fr) that are positivelyand negatively-charged, respectively. Specifically, 1) a PAH:Fr solution (1 mg mL-1 in 50 mM TRIS buffer with 200 mM NaCl at pH 8) was drop cast (50 μL) on top of the PSiO2 scaffold and incubated for 1 hour at RT; 2) then, a PMAA:Fr solution (1 mg mL-1 in 50 mM TRIS buffer with 200 mM NaCl at pH 8) was drop cast (50 μL) on top of the PAH:Fr-coated PSiO2 scaffold and incubated for 1 hour at RT. After each coating step, the PSiO2 scaffold on PDMS slab was rinsed with EtOH for 5 s, DIW for 1 min, and eventually gently dried at 30 °C for 5 min in a ventilated oven (G-Therm 035, Fratelli Galli, Italy), whereas the PSiO2 scaffold on bulk silicon was rinsed with DIW and dried under a N2 flow. Coating steps 1) and 2) were repeated as needed to increase the number of polyelectrolytes in the assembled stack, namely, from (PAH:Fr)1 to (PAH:Fr/PMAA:Fr)3. Morphological characterization of PSi scaffold Top view and cross-section morphological characterization of PSi layer was carried out using a scanning electron microscope (FEG-SEM, Zeiss SUPRA) with a 10 kV acceleration voltage at different magnifications. Pore diameter distribution was obtained from the analysis of 250000× topview SEM image using the “equivalent radius” function of Gwyddion software. Optical microscope (Leica DM2500 M) was equipped with a blue laser diode (CP450, λex = 450 nm, 4.5 mW, 3.2 × 1.0 mm2 elliptical beam shape, Thorlabs, USA) focused on the LbL-coated PSiO2 layer on bulk silicon orthogonally to the surface, and used to corroborate the cross-sectional LbL coating of PSiO2 with both PAH:Fr and PMAA:Fr. Optical characterization of PSi and PSiO2 scaffolds by reflectance measurements Reflectance spectra of bare and LbL-coated, PSi and PSiO2 scaffolds on bulk silicon were measured in the wavelength range 400 - 1000 nm at normal incidence in air. Light exiting from the source (Halogen light source HL-2000, OceanOptics, USA) was fed through one arm of a bifurcated fibreoptic probe (QR200-7-SR, OceanOptics, USA) orthogonally onto the sample surface, and the reflected light was collected through the other arm into a UV-VIS spectrometer (USB2000+UVVIS, Ocean Optics, USA). Acquisition parameters, namely, integration time, average scan number, and boxcar width, were: 2 ms, 2, and 5 with the spectrometer working normalized reflectance mode. Porosity and thickness of the PSi layer were evaluated by best-fitting the reflectance spectra with homemade Matlab software (MathWorks, USA). Fast Fourier Transform Reflectance Spectroscopy (FFT-RS, Hamming window) of the reflectance spectra of PSi and PSiO2 scaffolds was used to calculate the Effective Optical Thickness (EOT) = 2neff L value, with L thickness and neff effective refractive index of the scaffolds. Photoluminescence measurements of PSiO2 scaffolds coated with fluorescent polyelectrolytes Photoluminescence spectra of PSiO2 scaffolds on bulk silicon and PDMS slab and control SiO2 flat substrates coated with polyelectrolytes were acquired in the wavelength range 400 - 1000 nm in air for dry samples or through a sapphire window for wet samples secured within a flow cell. The excitation source was a blue laser diode (CP450, λex = 450 nm, 4.5 mW, 3.2 × 1.0 mm2 elliptical beam shape, Thorlabs, USA) focused orthogonally on the sample by means of the bifurcated fiberoptic probe (QR200-7-SR, OceanOptics, USA). The photoluminescence was collected through the other arm of the bifurcated fibre-optic probe and redirected to the UV-VIS spectrometer (USB2000+UV-VIS, Ocean Optics, USA) using an optical fiber (M35L01, Thorlabs, USA) with an in-line long-pass filter with cut-off wavelength of 515 nm (GL-OG515-3-12, Optoprim, Italy) secured in its in-line filter holder (FH-INL, Optoprim, Italy). Acquisition parameters, namely, integration time, average scan number, and boxcar width, were: 10s, 1, and 5 for measures carried out in air; 6s, 1, and 5 for measures carried out through a sapphire window on samples within the flow cell, with the spectrometer working in scope mode. Thickness evaluation of PAH:Fr/PMAA:Fr polyelectrolyte stacks Ellipsometry measurements were performed on flat SiO2(110 nm-thick)/Si samples (2 × 1.5 cm2) coated with PAH:Fr and PMAA:Fr polyelectrolytes using an ellipsometer (RudolphResearch AutoEL-II Automatic Ellipsometer) with an excitation wavelength of 633 nm at 70° incidence angle. Coating of the samples with the polyelectrolyte stack was carried out according to the following procedure: 1) the SiO2/Si sample was dipped in 4 mL of PAH:Fr (1 mg mL-1 in 50 mM TRIS buffer with 200 mM NaCl at pH 8) at RT for 120 s, washed in 40 mL of TRIS buffer pH 8 for 30 s to remove the excess of polymer, and dried with N2; 2) the PAH:Fr-coated sample was dipped in 4 mL of a PMAA:Fr solution (1 mg mL-1 in 50 mM TRIS buffer with 200 mM NaCl at pH 8) at RT for 120 s, washed in 40 mL of TRIS buffer pH 8 for 30 s, and eventually dried with N2. The steps 1) and 2) were repeated as needed to increase the number of polyelectrolytes in the assembled stack, namely, from (PAH:Fr)1 to (PAH:Fr/PMAA:Fr)3. Thickness evaluation of the deposited polyelectrolyte stack was performed at each step by bestfitting of experimental raw values (Δ and Ψ) using a customized homemade Matlab software (MathWorks, USA). Evaluation of intrinsic sensitivity of PAH:Fr and PMAA:Fr to pH PAH:Fr and PMAA:Fr solutions (0.05 mg mL-1 in 10 mM PBS buffer with 100 mM NaCl) with different pH values, namely 3, 4, 5, 6, 7, and 7.5, were placed into a quartz µ-fluorescence cuvette (CV10Q700, 700 μL volume, Thorlabs, USA) to measure photoluminescence spectra. The excitation source was a blue laser diode (CP450, λex = 450 nm, 4.5 mW, 3.2 × 1.0 mm2 elliptical beam shape, Thorlabs, USA) focused on the cuvette, and the 90° photoluminescence was collected through an optical fibre (M35L01, OceanOptics, USA) into a UV-VIS spectrometer (USB2000+UV-VIS, Ocean Optics, USA). Acquisition parameters were: integration time 100 ms, average scan number 1, boxcar width 5, with the spectrometer working in scope mode. Preparation of synthetic Vaginal Fluid (VF) solution VF solutions were prepared by adding AcOH (16.7 mM) and glycerol (1.73 mM) to DIW, then dissolving NaCl (100mM), KOH (25 mM), CaCl2 (2 mM), glucose (27.8 mM), and BSA (0.27 mM) [7]. pH Sensing with polyelectrolyte-coated PSiO2 scaffold using UV-VIS spectrometer pH sensing with polyelectrolyte-coated PSiO2 scaffolds, both on bulk silicon and PDMS slab, was carried out either in batch using a custom PDMS O-ring or securing the sensor within a custom-made flow-cell, at a constant temperature of 37 °C in a ventilated oven (G-Therm 035, Fratelli Galli, Italy). The photoluminescence spectrum was acquired as detailed in section Photoluminescence measurements of PSiO2 scaffolds coated with fluorescent polyelectrolytes, after immersion of the sensor in a PBS solution (10mM PBS with 100mM NaCl) or VF solution with different pH values in the range 3 – 7.5. As to batch experiments, a PDMS O-ring (base:curing agent 10:1 w/w, curing temperature of 90 °C, curing time of 1 h) with diameter of 1.2 cm and thickness of 3 mm was prepared and glued on the bulk silicon or PDMS slab carriers of the polyelectrolyte-coated PSiO2 scaffolds using a small quantity of PDMS prepolymer, then cured at RT for one day, except that for sensors used to carry out bending and twisting experiments, namely, 1, 10, and 100 bending cycles at different curvature radii, namely, 1, 2 and 2.5 cm and 100 twisting cycles, for which the PDMS prepolymer was cured in oven at 40 °C for 5 min. Next, a 500 µL of PBS solution (10 mM PBS with 100mM NaCl) or VF solution with different pH values, namely, 3, 4, 5, 6, 7, and 7.5 was dropped inside the chamber defined by the PDMS O-ring so that the sensor was fully immersed in the solution. Photoluminescence spectra of the sensor at a given pH value were acquired through the solution after 20 min from immersion. The solution was then sucked out and the sensor rinsed with DIW for three times to remove excess of the solution, before immersing the sample in a new solution with different pH value. Figure S4. Ellipsometry measurements of the thickness of PAH:Rh and PMAA:Rh multilayer stacks assembled on a 110-nm-thick SiO2 layer on Si. Data are reported as the average value measured over three replicates, with error bars representing the standard deviation (n=3). Figure S5. Bright field and fluorescence optical images of a PSiO2 scaffold coated with a multilayer stack of PAH:Fr and PMAA:Fr polyelectrolytes. a,b) Bright-field and c,d) fluorescence optical microscope images of the cross section of a PSiO2 scaffold coated with a fluorescent polymer stack (PAH:Fr/PMAA:Fr)2+(PAH:Fr)1. Scale bar in a,c) is 1 mm. Scale bar in b,d) is 20 µm. e,f) Intensity profile of the fluorescence emission calculated from c,d) using ImageJ. Figure S6. Sensing performance of pH sensors on native silicon. a) Photoluminescence spectra in PBS solution of a pH sensor on native silicon for different pH values in the range 3 – 7.5 at 37 °C. b) Calibration curve (photoluminescence intensity at 530 nm vs. pH value) of pH sensors on native silicon measured in PBS solution and VF simulant for different pH values in the range 3 – 7.5 at 37°C. Data are reported as the average value measured over 3 samples and 3 sensing cycles per each sample, with error bars representing the standard deviation (n=3). Figure S7. Intrinsic sensitivity of PAH:Fr and PMAA:Fr polyelectrolytes to pH. PL intensity change vs. pH value achieved for the pH sensor as well as for PAH:Fr, and PMAA:Fr solutions in PBS in the pH range 3 – 7.5. The PL intensity is normalized to the PL value achieved for each system at pH=7.5 for comparison purposes. Figure S8. Long-term performance of the pH sensor. Sensitivity values of the pH sensor measured over the pH range 3 – 7.5 in PBS at 37 °C after 0, 50, and 100 hours of continuous operation, as extrapolated from Figure 3g. Figure S9. Performance of pH sensor after bending and twisting experiments. a) Various pictures of a pH sensor on PDMS slab, also bent around a cylinder with radius of 1 cm and twisted. b, c) Sensitivity values of the pH sensor measured over the pH range 3 – 7.5 in PBS at 37 °C after b) 1 and c) 10 bending cycles, at different curvature radii. Figure S10. pH sensing with a light-emitting diode (LED) and a photodiode (PD) pair. a) Sketch of the setup used to demonstrate pH sensing with a LED-PD pair, instead of a spectrometer. b) Calibration curve (output voltage of the PD readout circuit vs. pH value) of the pH sensor on PDMS in the range of pH 3 – 7.5 measured with the setup in (a) in PBS solution and VF simulant. c) Sensitivity values extrapolated from (b). Data are reported as the average value measured over at least 3 samples and 3 sensing cycles per each sample, with error bars representing the standard deviation (n>=3). Figure S11. Optical transmittance of PDMS slabs. Transmittance spectrum of PDMS slabs with different thicknesses. Figure S12. Schematic of the driving/readout optoelectronic circuit implemented on the sensing vaginal ring. Top, from left to right: a LED (emission peak at 455 nm) is powered to stimulate the pH sensor. The fluorescence emission of the pH sensor (emission peak at 530 nm) is collected by the PD (maximum sensitivity at 560 nm), which is equipped with an optical long-pass filter paper (cutoff at 520 nm) on top to remove contribution of the LED light to the PD photocurrent. The PD photocurrent is converted by means of a transimpedance amplifier (TIA, resistive gain factor = 100 kΩ) obtaining an output voltage VOUT proportional to the PL intensity of the pH sensor. The analog VOUT signal is then converted into a digital signal using the analog-to-digital converter (ADC) integrated in the microcontroller ATmega328P and wireless transmitted via the Bluetooth module of the controller to a smartphone. Bottom, from left to right: photodiode bias voltage VPD is obtained from the supply voltage VSS (+/-3V) using a resistive voltage divider and a unit-gain buffer circuit. References [1] M. J. Sailor, Porous Silicon in Practice. Weinheim, Germany: Wiley, 2011 [2] S. Mariani et al., “Moldless Printing of Silicone Lenses with Embedded Nanostructured Optical Filters,” Adv. Funct. Mater., vol. 30, no. 4, p. 1906836, Jan. 2020. [3] Wan Y, Krueger NA, Ocier CR, Su P, Braun P V., Cunningham BT. Resonant Mode Engineering of Photonic Crystal Sensors Clad with Ultralow Refractive Index Porous Silicon Dioxide. Adv Opt Mater. 2017;5(21):1-7. doi:10.1002/adom.201700605 [4] V. Lehmann, Electrochemistry of Silicon, vol. 3. 2002. [5] S. Mariani et al., “Layer-by-layer biofunctionalization of nanostructured porous silicon for high-sensitivity and high-selectivity label-free affinity biosensing,” Nat. Commun., vol. 9, no. 1, p. 5256, Dec. 2018. [6] S. Mariani, A. Paghi, A. A. La Mattina, A. Debrassi, L. Dähne, and G. Barillaro, “Decoration of Porous Silicon with Gold Nanoparticles via Layer-by-Layer Nanoassembly for Interferometric and Hybrid Photonic/Plasmonic (Bio)sensing,” ACS Appl. Mater. Interfaces, vol. 11, no. 46, pp. 43731–43740, Nov. 2019. [7] D. H. Owen and D. F. Katz, “A vaginal fluid simulant,” Contraception, vol. 59, no. 2, pp. 91– 95, Feb. 1999.