Ratiometric sensing of fluoride ions using Raman spectroscopy

Centre for Molecular Nanometrology, WestCHEM, Department of Pure and Applied Chemistry, Technology and Innovation Centre, University of Strathclyde, 99 George Street, Glasgow G1 1RD, UK. WestCHEM, Department of Pure and Applied Chemistry, University of Strathclyde, 295 Cathedral Street, Glasgow G1 1XL, UK. †Present address: Technische Universtät Braunschweig, Institut für Technische Chemie, Franz-Liszt Str. 35a, 38106 Braunschweig, Germany. Email: nicholas.tomkinson@strath.ac.uk; karen.faulds@strath.ac.uk; duncan.graham@strath.ac.uk

Raman maps on paper test strips: All Raman maps were processed in WiRE 4.4 TM software enabling cosmic ray removal, noise filtering and baseline subtraction. A custom MATLAB ® script was then used to perform ratiometric analysis on the Raman spectral map (20 μm × 20 μm; 400 spectra). False-colour images for the test paper strips were created based on the peak intensity ratio: 2105 cm -1 / 2160 cm -1 and 2231 cm -1 / 2237 cm -1 . The images were scaled between 0-0.2 (alkyne) or 0-1.0 (nitrile) and are presented in the Parula LUT available in MATLAB ® .

General Procedures
All reagents were obtained from commercial sources, including Sigma-Aldrich, Alfa Aesar and Fluorochem and used without purification unless otherwise stated. The abbreviations Et2O and NEt3 refer to diethyl ether and triethylamine, respectively. The term "in vacuo" refers to evaporation under reduced pressure using a rotary evaporator connected to a diaphragm pump, followed by the removal of trace volatiles using a high vacuum (oil) pump. The term "purged" refers to atmospheric exchange via 3 evacuation/refill cycles using a Schenck line fitted to a cylinder of inert gas and a high vacuum (oil) pump. Flash chromatography was carried out using Fischer Scientific chromatography grade silica 60 Å particle size 35-70 micron. Analytical thin layer chromatography was carried out using aluminium-backed plates coated with Machery-Nagel pre-coated TLC sheets, coated in 0.20 mm silica gel 60 with UV254 fluorescent indicator. Sheets were visualized under UV light (at 254 nm) or stained using p-anisaldehyde. Nuclear magnetic resonance (NMR) spectra were recorded on a Bruker Avance 500 spectrometer, operating at 500 MHz ( 1 H) and 125 MHz ( 13 C). Chemical shifts were reported in parts per million (ppm) in the scale relative to CDCl3, 7.26 ppm for 1 H NMR and 77.16 for 13 C NMR. Multiplicities are abbreviated as: s, singlet; d, doublet. Coupling constants are measured in Hertz (Hz). Melting points were obtained on a Stuart SMP11 device. Infrared spectra were recorded in the range 4000-600 cm -1 on a Shimadzu IRAffinity-1 equipped with an ATR accessory.

Synthesis of Sensor 1, 4-((trimethylsilyl)ethynyl)benzonitrile
A flame dried 20 mL reaction vial was charged with 4-bromobenzonitrile (273 mg, 1.50 mmol, 1.0 eq), bis(triphenylphosphine)palladium (II) dichloride (10.5 mg, 0.015 mmol, 1 mol%) and copper (I) iodide (3.0 mg, 0.015 mmol, 1 mol%), then sealed purged with nitrogen. NEt3 (degassed by 3 freeze-pump-thaw cycles, 6 mL) was added to the reaction vial via syringe, followed by (trimethylsilyl)acetylene (230 μL, 1.65 mmol, 1.1 eq) and the reaction was heated to 80 °C with stirring for 4 h. After cooling to ambient temperature, the reaction mixture was diluted with Et2O (20 mL), filtered through a pad of celite® and evaporated in vacuo. Purified by silica flash chromatography (3% Et2O/petroleum ether 40-60) to yield the title compound 1 as a white solid (285 mg, 1.43 mmol, 95%).  Figure S4 Control reactions for the desilylation reaction of sensor 1. A A mixture of sensor 1 (5 mM in THF:water 1:1 v/v) was analysed using Raman spectroscopy at t = 0 min and t = 30 min. The acquisitions show no peak at 2109 cm -1 (i.e. no desilylated product 2 was observed). Raman spectra were acquired using 785 nm excitation and a 20× objective lens (~180 mW) for 10 s. B Analysis of sensor 1 with different counter anions. Sensor 1 (5 mM in THF) was treated with either THF (Blank) or TBAX (X= F, Cl, Br; 50 mM prepared from a 100 mM stock in water). Raman spectra were acquired after 30 min at 20 °C using 785 nm for 10s using a 20× objective lens (~180 mW). Data represents the mean peak area ratio at 2109/2162 cm -1 from three replicates with error bars ± S.D. C and D Analysis of sensor 1 with different counter anions. Sensor 1 (5 mM in THF) was treated with NaX (X= fluoride, acetate, ascorbate, hydrogen carbonate, citrate, nitrate and phosphate; 50 mM prepared from a 100 mM stock in water). Raman spectra were acquired after 30 min at 20 °C using 785 nm for 10 s using a 20× objective lens (~180 mW). Data represents the mean peak area ratio at 2109/2162 cm -1 from three replicates with error bars ± S.D. E As per C and D, but using NaI and at high pH using NaX (X = carbonate and hydroxide) showing decomposition of the sensor occurs.  1 v/v)). Raman spectra were acquired after 15 min at RT using λex = 785 nm using a 5× lens (~180 mW) for 10 s. Spectra representative of 3 repeats. This reaction shows a decrease in absorbance at ~280 nm which is in good agreement with a similar TMS-protected alkynesee Supplementary Information File Ref. [2] . . Raman spectra were acquired using 785 nm for 0.5 s using a 5× objective lens (~180 mW) and normalised to the intensity of the THF solvent peak at 1450 cm -1 (CH2 def). The reaction monitoring started ~5 s after the addition of TBAF. D Reaction profiles for the desilylation reaction of 1 (5 mM) with TBAF (7.5 mM) in (i) THF, (ii) THF:water (90:10 v/v), (iii) THF:water (80:20 v/v), (iv) THF:water (75:25 v/v) and (v) is a control reaction where [TBAF] = 0 mM. The peak areas (Ap) at 2162 cm -1 (C≡C-TMS, 1) and 2109 cm -1 (desilylated alkyne, 2) are plotted as a function of time. Raman spectra were acquired using 785 nm for 1 s using a 5× objective lens (~180 mW) and normalised to the intensity of the THF solvent peak at 1450 cm -1 (CH2 def).

Figure S8
Paper-based detection of fluoride using Raman sensor 1 using the ratio of the nitrile band. A These maps accompany those acquired in Figure 4. Filter paper was pre-treated with sensor 1 (100 mM in THF, 10 µL) before air-drying and subsequent treatment with TBAF in THF at the indicated concentrations. Raman maps were acquired across 20 μm × 20 μm (1 μm pixel size; 400 spectra) using 785 nm laser excitation with a 20× objective lens (~180 mW) for 0.5 s. The maps represent the ratio of the nitrile band at 2231 cm -1 /2237 cm -1 (for desilylated product/sensor 1). B Expanded view of the Raman spectra presented in Figure  4B, indicating signal at 2109 cm -1 when 125 μM TBAF is added.  Figure S9 Repeat analysis of the paper-based detection of fluoride using Raman sensor 1.
Filter paper was pre-treated with sensor 1 (100 mM in THF, 10 µL) before air-drying and subsequent treatment with TBAF in THF at the indicated concentrations. Raman maps were acquired across 20 μm × 20 μm (1 μm pixel size; 400 spectra) using 785 nm laser excitation with a 20× objective lens (~180 mW) for 0.5s. The maps represent the following ratios: A 2105 cm -1 /2160 cm -1 ; B 2231 cm -1 /2237 cm -1 (for desilylated product/sensor 1) and C the average Raman spectra from the maps presented in A and B.  Figure S10 Paper-based detection of fluoride using a Raman microscope and a handheld spectrometer. Filter paper was pre-treated with sensor 1 (100 mM in THF, 10 µL) before airdrying and subsequent treatment with TBAF in THF (either 125 μM or 0 μM). Point spectra were acquired using either (A) -(B) a Raman microscope (785 nm laser excitation with a 20× objective lens (~180 mW) for 10s) or (C) a handheld spectrometer (785 nm laser excitation for 10s (~55 mW). Three repeat spectra from the same paper test strip are provided in each case.