Working with water insoluble organic molecules in aqueous media: fluorene derivative-containing polymers as sensory materials for the colorimetric sensing of cyanide in water

Abstract

This paper describes a strategy followed to achieve a sensing phenomenon in aqueous media using water-insoluble organic molecules. We have prepared a methacrylamide and a methacrylate with pendant cyanide chemosensors based on a fluorene-derivative motif, and we have fabricated highly hydrophilic membranes by means of copolymerising these hydrophobic monomers with others. Therefore, upon absorption of water in the membranes, solvated ions enter the membrane by a simple diffusion mechanism, reaching the hydrophobic chemosensor motifs and giving rise to a macroscopic sensing phenomenon. In this way, we have prepared solid materials (dense membranes or films) capable of selectively detecting cyanide, with an extremely low detection threshold, in aqueous solution by means of colour changes (naked-eye sensing) (13 ppb). Nevertheless, the key point of this research is the description of the possibilities of anchoring organic insoluble molecules (i.e., drugs, fungicides, bactericides, sensing probes, etc.) to solubilise them in water, or to prepare hydrogels, permitting the use of these molecules in aqueous media or in biological media for medical, biological or biochemical purposes.

---

Introduction

Molecular recognition is a widespread, naturally occurring key event in nature. The recognition between two or more molecules occurs when they are chemically and geometrically, or structurally, complementary. Thus, the “fit together” takes place spatially, and the binding to each other arises from intermolecular feeble interactions, including hydrogen bonds, electrostatic interactions, hydrophobic interactions and weak metal coordination.^1,2 In nature, the processes take place in aqueous environments and are usually implicated in natural macromolecules or even in complex and organised systems, such as cells. Examples of the recognition process include the binding of an enzyme to a substrate,³ a drug to a biological target,^4,5antigen/antibody recognition in the immune system,^6,7 the formation of messenger RNA from DNA templates,⁸etc. In this context, the development of chemosensors for the detection of target chemicals is a field of current interest.⁹

Naturally, recognition in biological systems takes place in aqueous environments. The active sites, for example in an enzyme, have the shape needed to fit the guest molecule, or a part of the guest molecule, and the interaction relies on feeble bonds that take place due to the hydrophobic microdomain in a hydrophilic general domain provided by the chemical and by the quaternary structure of a protein. In a very simplified manner, the competitiveness of the water molecules to establish these weak integrations is locally diminished or altered by the active site microsurroundings.

Mimicking nature, water insoluble receptors may be chemically anchored to a linear or crosslinked hydrophilic polymer, giving rise to water-soluble sensing polymers or to water-swelled sensing networks. The latter approach permits easy control of swelling by means of increasing the crosslinking ratio. That is, the hydrophilic character of a polymer, derived from the chemical composition of the monomers, may be controlled not only by the monomer nature, but also by increasing the crosslinking monomer content of the network; thus, mechanically impairing the water uptake. Thus, an induced partially hydrophobic character may be addressed with an overall hydrophilic polymer associated with the mechanical stretching of the network upon swelling with water, as we have previously described.^10,11 Moreover, the polymers usually have good thermal and mechanical properties, can be easily transformed into end materials, e.g., into films or coatings obtained by casting, to produce cheap sensing devices (e.g., user-friendly “naked-eye” film sensors). These polymers may also be incorporated as coating at the end of an optic fibre connected to a portable UV/Vis diode-array or fluorescence detector.^12–14

For example, cyanide is a highly toxic, chemical target currently used in many industrial processes.^15–17 Due to its toxicity, the United States Environmental Protection Agency (EPA) has set the maximum contaminant level (MCL) for cyanide in drinkingwater at 0.2 ppm, rendering its detection important in terms of both environmental and industrial control.

The detection ofcyanide through various supramolecular approaches¹⁶ (the binding site-signalling subunit,¹⁸ the chemodosimeter^19,20 or the displacement^21–24) has been extensively reported. Nevertheless, few examples of sensing of the anion in aqueous solutions,^20,22,24,25 some of them showing detection limits lower than the MCL,^16,22 have been described. Most cyanide sensing molecules are insoluble in water because of their organic nature, preventing their use in aqueous media. We felt that detection ofcyanide by means of changes in optical signals upon interaction of a target guest with a sensor molecule could give rise to the development of inexpensive, portable devices for environmental testing inside or outside the laboratory.^9,15 Prior to our investigation, few examples existed describing materials exhibiting colorimetric, or fluorescence, recognition with potential use as solid systems or practical kits for “dip-in” naked-eye cyanide detection (e.g., films or membranes,^10,25,26quantum dots²⁷ or supported materials²⁸). We hypothesised that the chemical anchoring of a sensing motif to a hydrophilic polymer chain could provide a water-rich environment to the sensing moiety upon polymer swelling, thus permitting the solvated cyanide ions to reach the guest or the transducer cores, giving rise to the sensing phenomenon, and resulting in cyanide detection.

Experimental

All materials and solvents were commercially available and used as received, unless otherwise indicated. They included the following: sodium cyanide (Panreac, 98%), 2-isocyanato-9H-fluorene (Aldrich, 98%), ethanolamine (Merck, 98%), cesium carbonate (Aldrich, 99%), malononitrile (Aldrich, 99%), 2-nitro-9H-fluorene (Aldrich, 98%), Pd/C (Aldrich, 10%), methacryloyl chloride (Fluka, 97%), 2-(2-aminoethoxy ethanol) (Aldrich, 98%), fluorenone (Aldrich, 98%), ethylene glycol dimethacrylate (Aldrich, 98%), DMSO (Merck, 99%), acetone (Aldrich, 99%), ethanol (Aldrich, 99%), dichloromethane (Aldrich, 99%), DMF (Aldrich, 99%), hexanes (Aldrich, 99%), and ethyl acetate (Aldrich, 99%). Pyridine was dried under reflux over sodium hydroxide for 24 h and was distilled over 4 Å molecular sieves. Azo-bis-isobutyronitrile (AIBN, Aldrich, 99%) was recrystallised twice from methanol.

Intermediates and monomers

The two methacrylamide and methacrylate monomers were synthesised following the reaction conditions described in Scheme 1.

Scheme 1 Monomer synthesis and labels.

2-Ethoxyethyl methacrylate (1). This commercially available monomer (Aldrich, 99%) was purified by distillation under high vacuum.

2-(2-Aminoethoxyethanol) methacrylamide (2). To 20 mL of an ice-cold, methanolic solution of 2-(2-aminoethoxy ethanol) (2.4 mL, 24 mmol) was slowly added 2.4 mL (25 mmol) of methacryloyl chloride diluted in 20 mL of THF under a nitrogen blanket. Potassium hydroxide (1 M, aqueous) was added to maintain a pH of 8–9 throughout the reaction. The mixture was warmed to rt over a 4 h period. Afterwards, it was quenched by addition of hydrochloric acid to obtain a final pH of 5, and the product was concentrated. Finally, sodium chloride was added to the crude residue, and it was extracted with dichloromethane. Yield: 97%. ¹H NMR (300 MHz, CDCl₃-d₆): δ (ppm), 6.98 (s, 1H); 5.66 (m, 1H); 5.26 (m, 1H); 4.48 (s, 1H); 3.64 (m, 2H); 3.51 (m, 4H); 3.43 (m, 2H); 1.87 (m, 3H). ¹³C NMR (75.4 MHz, CDCl₃-d₆): δ (ppm), 169.34; 139.72; 120.37; 72.43; 69.82; 61.63; 39.81; 18.86. EI-LRMS m/z: 174.0 (M⁺, 10), 113.0 (12), 112.0 (26), 111.0 (21), 98.0 (25), 69.0 (100), 45.0 (30), 44.0 (17), 41.0 (54). FTIR [wavenumbers (cm⁻¹)]: ν_OH,NH: 3503–2489, ν_C=O: 1704; δ_N–H: 1606.

Synthesis of 1-(9H-fluoren-7-yl)-3-(2-hydroxyethyl)urea (3). To a 500 mL flask fitted with a mechanical stirrer were added 25 mmol of 2-isocyanato-9H-fluorene and 50 mL of dichloromethane. Subsequently, 25 mmol of ethanolamine was added dropwise, and the solution was stirred at 0 °C for 30 min and then at rt for 6 h. Finally, the product 1-(9H-fluoren-7-yl)-3-(2-hydroxyethyl)urea was filtered off and washed with dichloromethane. Mp: 212 ± 1 °C. Yield: 96%. ¹H NMR, δ (400 MHz, DMSO-d₆): 8.71 (s, 1H, NH); 7.77 (m, 3H, ArH); 7.54 (m, 1H, ArH); 7.36 (m, 2H, ArH); 7.25 (dt, 1H, J = 1.1, 7.4 Hz); 6.30 (t, 1H, J = 5.6 Hz, NH), 4.84 (t, 1H, J = 4.8 Hz, OH), 3.89 (s, 2H); 3.26 (c, 4H, J = 5.3 Hz). ¹³C NMR, δ (100.6 MHz, DMSO-d₆): 156.34, 144.84, 143.39, 142.30, 140.75, 135.31, 127.59, 126.51, 125.86, 121.86, 121.07, 119.97, 117.39, 115.33, 61.42, 42.84, 37.44. EI-LRMS m/z: 269.1 (M⁺, 7), 181.1 (100), 268.1 (39), 207.1 (22), 165.1 (19), 78.0 (14). FTIR [wavenumbers (cm⁻¹)]: ν_OH: 3355; ν_NH: 3209; ν_c=o: 1656.

Synthesis of 1-(2-hydroxyethyl)-3-(9-oxo-9H-fluoren-7-yl)urea (4). To a 125 mL flask equipped with a reflux condenser was added 13 mmol of 1-(9H-fluoren-7-yl)-3-(2-hydroxyethyl)urea dissolved in 50 mL of DMSO. To the reaction mixture was added 67 mmol of cesium carbonate, and the solution was stirred vigorously at rt for 5 d. Finally, the solution was precipitated in water to give the orange product 1-(2-hydroxyethyl)-3-(9-oxo-9H-fluoren-7-yl)urea. Mp: 222 ± 1 °C. Yield: 98%. ¹H NMR, δ (400 MHz, DMSO-d₆): 8.93 (s, 1H, NH); 7.81 (d, 1H, J = 2.1 Hz, ArH); 7.64 (t, 2H, J = 7.7 Hz, ArH); 7.57 (m, 2H, ArH); 7.47 (dd, 1H, J = 2.1, 8.1 Hz, ArH); 7.29 (dt, 1H, J = 1.0, 7.2 Hz, ArH); 6.33 (t, 1H, J = 5.6 Hz, NH); 4.82 (t, 1H, J = 5.1 Hz, OH); 3.50 (c, 2H, J = 5.3 Hz); 3.22 (c, 2H, J = 5.6 Hz), ¹³C NMR, δ (100.6 MHz, DMSO-d₆): 194.01, 155.70, 145.16, 142.60, 136.97, 136.05, 134.82, 134.00, 128.78, 124.48, 123.55, 122.22, 120.90, 113.84, 60.94, 42.43. EI-LRMS m/z: 283.1 (M⁺, 3); 78.0 (17); 282.1 (20); 221.0 (36); 195.1 (100). FTIR [wavenumbers (cm⁻¹)]: ν_OH: 3384; ν_NH: 3347; ν_C=O: 1701.

Synthesis of 1-(9-(dicyanomethylene)-9H-fluoren-7-yl)-3-(2-hydroxyethyl)urea (5). To a round-bottomed flask fitted with a mechanical stirrer and a reflux condenser was added 16 mmol of 1-(2-hydroxyethyl)-3-(9-oxo-9H-fluoren-7-yl)urea in 50 mL of dimethylformamide (DMF). Subsequently, 30 mmol of malononitrile was added, and the solution was stirred at 100 °C for 5 d, yielding a precipitate. The precipitate was filtered off and washed with acetone at reflux for 5 min. The product was dried thoroughly in a vacuum oven at 60 °C. Mp: 235 ± 2 °C. Yield: 40%. ¹H NMR, δ (400 MHz, DMSO-d₆): 8.92 (s, 1H, NH); 8.10 (d, 1H, J = 1.8 Hz, ArH); 7.98 (d, 1H, J = 7.8 Hz, ArH); 7.60 (dd, 1H, J = 1.8, 8.3 Hz, ArH); 7.46 (m, 3H, ArH); 7.22 (dt, 1H, J = 1.5, 7.5 Hz, ArH); 6.30 (t, 1H, J = 5.5 Hz, NH); 4.82 (t, 1H, J = 5.1 Hz, OH); 3.52 (c, 2H, J = 5.3 Hz); 3.23 (c, 2H, J = 5.3 Hz). ¹³C NMR, δ (100.6 MHz, DMSO-d₆): 161.16, 155.63, 143.19, 142.34, 135.71, 134.90, 134.66, 133.86, 128.46, 126.36, 123.48, 122.25, 120.90, 115.92, 114.14, 113.71, 75.62, 61.08, 42.43. EI-LRMS m/z: 313.1 (2), 312.1 (7), 311.1 (4), 269.1 (33), 270.1 (5), 243.1 (100), 201.1 (2), 163.1 (2), 78.0 (4). FTIR [wavenumbers (cm⁻¹)]: ν_NH,OH: 3355; ν_CN: 2213; ν_C=O: 1671.

Synthesis of 2-(3-(9-(dicyanomethylene)-9H-fluoren-7-yl)ureido)ethyl methacrylate (6). 1.3 mmol of 1-(9-(dicyanomethylene)-9H-fluoren-7-yl)-3-(2-hydroxyethyl)urea was dissolved in 15 mL of pyridine in a round-bottom flask. Afterward, 1.7 mmol of methacryloyl chloride was added dropwise, and the solution was stirred at room temperature for 3 h. Subsequently, the solution was precipitated in slightly acidified water at 0 °C. The product 2-(3-(9-(dicyanomethylene)-9H-fluoren-7-yl)ureido)ethyl methacrylate was extracted from the crude residue using a Soxhlet apparatus with acetone. The acetone was then removed by distillation, and the product was purified by column chromatography using a mixture of hexane : ethyl acetate (3 : 1). Mp: 212 ± 2 °C. Yield: 83%. ¹H NMR, δ (400 MHz, DMSO-d₆): 8.99 (s, 1H, NH); 8.30 (m, 1H, ArH); 8.12 (d, 1H, J = 7.8 Hz, ArH); 7.64 (m, 3H, ArH); 7.54 (t, 1H, J = 7.5 Hz, ArH); 7.32 (t, 1H, J = 7.7 Hz, ArH); 6.41 (t, 1H, J = 5.9 Hz, NH); 6.12 (s, 1H, CH₂); 5.74 (s, 1H, CH₂); 4.18 (t, 1H, J = 5.4 Hz, CH₂); 1.93 (s, 3H, CH₃). ¹³C NMR, δ (100.6 MHz, DMSO-d₆): 167.17, 161.26, 155.71, 143.23, 142.26, 136.46, 135.80, 135.25, 134.86, 134.06, 128.69, 126.56, 123.90, 122.43, 121.15, 116.24, 114.27, 113.87, 81.67, 76.15, 64.61, 18.57. EI-LRMS m/z: 399.1 (M⁺, 2), 398.1 (8), 313.1 (1), 270.1 (7), 243.1 (100), 226.1 (1), 163.0 (2). FTIR [wavenumbers (cm⁻¹)]: ν_NH: 3355; ν_CN: 2220; ν_C=O: 1708.

Synthesis of 9H-fluoren-2-amine (7). To a hydrogenation flask was added 48 mmol of 2-nitro-9H-fluorene mixed with 20 mL of ethanol and 5 g of Pd/C (10% Pd), and the system was purged and pressured with hydrogen to 75 psi. The mixture was stirred at 60 °C for 3 h, and the hydrogen consumed was replaced every 15 min. The solid was filtered off and dried. Mp: 132 ± 2 °C. Yield: 62%. ¹H NMR, δ (400 MHz, DMSO-d₆): 7.64 (d, 1H, J = 7.2 Hz, ArH); 7.55 (dd, 1H, J = 2.0, 8.1 Hz, ArH); 7.47 (d, 1H, J = 7 Hz, ArH); 7.29 (t, 1H, J = 7.2 Hz, ArH); 7.14 (m, 1H, ArH); 6.82 (s, 1H, ArH); 6.64 (m, 1H, ArH); 5.26 (s, 2H, NH₂); 3.77 (s, 2H, CH₂). ¹³C NMR, δ (100.6 MHz, DMSO-d₆): 149.36, 145.52, 142.60, 130.63, 127.40, 125.56, 121.41, 118.95, 113.73, 111.27, 57.08, 37.16, 19.48. EI-LRMS m/z: 181.1 (M⁺, 100), 180.1 (70), 165.1 (6), 90.0 (10). FTIR [wavenumbers (cm⁻¹)]: ν_NH: 3443, 3355.

Synthesis of 2-amino-9H-fluoren-9-one (8). To a flask equipped with a reflux condenser was added 30 mmol of 9H-fluoren-2-amine dissolved in 50 mL of DMSO. Subsequently, 89 mmol of cesium carbonate was added, and the solution was vigorously stirred for 5 d at rt. Finally, the solution was precipitated in water, and the garnet product 2-amino-9H-fluoren-9-one was thoroughly washed with water. Mp: 157 ± 2 °C. Yield: 93%. ¹H NMR, δ (400 MHz, DMSO-d₆): 7.51–7.39 (m, 4H, ArH); 7.20–7.14 (m, 1H, ArH); 6.85–6.83 (m, 1H, ArH), 6.74–6.69 (m, 1H, ArH); 5.70 (2H, s, NH₂). ¹³C NMR, δ (100.6 MHz, DMSO-d₆): 195.11, 151.32, 146.71, 136.27, 135.86, 134.02, 131.76, 127.66, 124.50, 122.96, 120.09, 119.27, 110.27. EI-LRMS m/z: 195.2 (M⁺, 100), 167.1 (5), 78.0 (17). FTIR [wavenumbers (cm⁻¹)]: ν_NH: 3450, 3355; ν_C=O: 1701.

Synthesis of 2-(2-amino-9H-fluoren-9-ylidene)malononitrile (9). To a flask equipped with a reflux condenser was added 8 mmol of 2-amino-9H-fluoren-9-one dissolved in 20 mL of DMF. Afterward, 10 mmol of malononitrile was added portion-wise, and the solution was stirred at 100 °C for 5 d. The product 2-(2-amino-9H-fluoren-9-ylidene)malononitrile was filtered off and purified by column chromatography with a mixture of hexanes : ethyl acetate (1 : 1). Mp: 262 ± 1 °C.Yield before purification: 80%. Yield after purification: 25%. ¹H NMR, δ (400 MHz, DMSO-d₆): 8.07 (d, 1H, J = 7.8 Hz, ArH); 7.51–7.40 (m, 4H, ArH); 7.19 (dt, 1H, J = 1.9, 7.3 Hz, ArH); 6.71 (dd, 1H, J = 2.0, 8.1 Hz, ArH); 5.88 (s, 2H, NH₂). ¹³C NMR, δ (100.6 MHz, DMSO-d₆): 162.31, 151.04, 144.62, 135.84, 133.70, 129.86, 127.27, 126.49, 123.00, 119.95, 119.39, 114.55, 114.21, 112.18. EI-LRMS m/z: 242.1 (M⁺, 3), 243.1 (100), 217.1 (1), 215.1 (15), 189.1 (7), 78.0 (28). FTIR [wavenumbers (cm⁻¹)]: ν_NH: 3480, 3370; ν_CN: 2220; ν_C=O: 1613.

Synthesis of N-(9-(dicyanomethylene)-9H-fluoren-7-yl)methacrylamide (10). 2.1 mmol of 2-(2-amino-9H-fluoren-9-ylidene)malononitrile was dissolved in 8 mL of pyridine in a round-bottom flask. Afterward, 2.7 mmol of methacryloyl chloride was added dropwise to the solution, and it was stirred at rt for 3 h. The product N-(9-(dicyanomethylene)-9H-fluoren-7-yl)methacrylamide was purified by column chromatography using a mixture of hexane : ethyl acetate (2 : 1) and was washed in a Soxhlet assembly with ethanol. Mp: 224 ± 1 °C. Yield before purification: 87%. Yield after purification: 20%. ¹H NMR, δ (400 MHz, DMSO-d₆): 10.21 (s, 1H, NH); 8.78 (m, 1H, ArH); 8.19 (d, 1H, J = 7.8 Hz, ArH); 7.84–7.73 (m, 3H, ArH); 7.60 (t, 1H, J = 7.5 Hz, ArH); 7.40 (t, 1H, J = 7.7 Hz, ArH); 5.87 (s, 1H, CH₂); 5.61 (s, 1H, CH₂); 2.00 (s, 3H, CH₃). ¹³C NMR, δ (100.6 MHz, DMSO-d₆): 167.81, 160.92, 142.72, 140.75, 137.39, 135.84, 134.61, 134.27, 129.20, 126.80, 126.66, 122.22, 121.61, 121.15, 118.66, 114.36, 113.74, 76.56, 19.40. EI-LRMS m/z: 310.1 (M⁺, 1), 311.1 (73), 69.0 (100), 188.0 (9), 78.0 (5). FTIR [wavenumbers (cm⁻¹)]: ν_NH: 3377; ν_CN: 2220; ν_C=O: 1679.

The model compounds used to clarify the sensing mechanism were synthesised following the reaction conditions described in Scheme 4.

Synthesis of 2-(9H-fluoren-9-ylidene)malononitrile (11). 11 mmol of 9H-fluoren-9-one was dissolved in 25 mL of DMF. Afterward, 16.5 mmol of malononitrile was added, and the solution was stirred at reflux for 5 d. The solution was precipitated in water, and the solid was recrystallised from acetone. Mp: 240 ± 1 °C. Yield: 80%. ¹H NMR, δ (400 MHz, DMSO-d₆): 8.25 (d, 2H, J = 7.8 Hz, ArH); 7.91 (d, 2H, J = 7.6 Hz, ArH); 7.66 (t, 2H, J = 7.5 Hz, ArH); 7.49 (dt, 2H, J = 7.7 Hz, ArH); ¹³C NMR, δ (100.6 MHz, DMSO-d₆): 161.02, 142.53, 135.83, 134.26, 130.13, 126.72, 122.30, 114.20, 76.67. EI-LRMS m/z: 228.1 (M, 100), 76.0 (1), 63.0 (2). FTIR [wavenumbers (cm⁻¹)]: ν_CN: 2213.

Synthesis of 2-(9-cyano-9H-fluoren-9-yl)malononitrile (12). 8.8 mmol of 2-(9H-fluoren-9-ylidene)malononitrile was dissolved in 15 mL of DMSO. Afterward, 17.6 mmol of sodium cyanide was added, and the solution was stirred at rt for 2 h. The solution was precipitated in water, and the solid was filtered off and washed with water. The product was dried in a vacuum oven at 60 °C. Mp: 159 ± 1 °C. Yield: 76%. ¹H NMR, δ (400 MHz, DMSO-d₆): 8.14 (d, 2H, J = 7.5 Hz, ArH); 8.07 (d, 2H, J = 7.7 Hz, ArH); 7.74 (t, 2H, J = 7.6 Hz, ArH); 7.65 (t, 2H, J = 7.6 Hz, ArH); 6.81 (s, 1H). ¹³C NMR, δ (100.6 MHz, DMSO-d₆): 140.92, 137.74, 132.57, 130.11, 125.10, 122.49, 117.10, 111.20, 47.78, 32.56. EI-LRMS m/z: 255.1 (M⁺, 12), 190.0 (100), 163.0 (6), 63.0 (2). FTIR [wavenumbers (cm⁻¹)]: ν_CH,sp³: 2886; ν_CN: 2242.

Membrane preparation

The membranes were prepared by radical polymerisation of mixtures of the 4 different co-monomers (1, 2, 6, 10) using ethylene glycol dimethacrylate as a cross-linking agent (1%) and AIBN (1 wt%) as a thermal radical initiator. The monomers (6) and (10) were added in DMF. Eight different polymer films were prepared in this fashion by varying the molar concentration of monomers. The reactions were performed in silanised glass moulds of 100 μm thickness in an oxygen-free atmosphere at 65 °C for 5 h. Upon demoulding, the film was air dried for two days and overnight under vacuum at rt.

Measurements

¹H and ¹³C NMR spectra were recorded with a Varian Inova 400 spectrometer operating at 399.92 and 100.57 MHz, respectively, with deuterated dimethyl sulfoxide (DMSO-d₆) as solvent. Infrared spectra (FT-IR) were recorded with a Nicolet Impact spectrometer or with a JASCO FT/IT-4100 fitted with a PIKE TECH “Miracle” ATR. Thermogravimetric analysis (TGA) data were recorded on a 5 mg sample under a nitrogen or oxygen atmosphere on a TA Instrumenta Q50 TGA analyser at a scan rate of 10 °C min⁻¹. UV-Vis spectra were recorded using a Varian Cary3-Bio UV-Visspectrophotometer. Sensing measurements performed in water were performed at physiological pH (7.5, buffered with Tris). Apparent response time was recorded using a digital stopwatch. The different polymeric films were dipped in a water solution of sodium cyanide (3 M), and the time was taken when the colour of the film turned colourless to the naked eye. Each experiment was repeated three times, and the results were averaged. The water-swelling percentage (WSP), the weight percentage of water uptake by the films upon soaking until equilibrium, in pure water, was obtained from the weights of a dry sample film (w_d) and a water-swelled (the membranes were immersed in pure water at 20 °C until the swelled equilibrium was achieved) sample film (w_s) as follows: 100 × [(w_s − w_d)/w_d].

Results and discussion

The main aim of this research was to extend some of the properties or characteristics of organic molecules, insoluble in water, to aqueous media by means of chemically incorporating them into a hydrophilic polymer network. To investigate our hypothesis, we prepared two methacrylic and methacrylamide monomers containing a fluorene derivative as a colorimetric cyanide sensing motif [(6) and (10)] and copolymerised them with either a hydrophobic or a hydrophilic monomer [(1) and (2), respectively]. The cross-linking agent 1,2-ethanedioldimethacrylate was used to obtain different, dense polymer films or membranes (Scheme 2 and Table 1), which were optically transparent and had good mechanical properties, even after water swelling, from which cyanide solid sensing strips could be easily obtained with a hand blade from the films.

Scheme 2 Material preparation. Synthesis of the dense membrane or copolymer network, or hydrogel.

Table 1 Composition and properties of polymeric films prepared

Membrane or film	1 (x)^a	2 (y)^a	6 (z)^a	10 (z)^a	WSP ^b	Raw film^c	Film + CN−^c	Response time/s^d
a Molar composition ratio of the monomers used for the preparation of the films. b Water-swelling percentage. c Digital picture of the films taken over a written paper upon soaking in pure water and in an aqueous solution of cyanide (10 equiv. per equiv. of the sensing fluorene derivative motif), respectively; the pictures have been taken on written paper to show the transparency and the different shades of the films. d Naked-eye response time (—: without response, n.a.: not applicable).
(a)	100	—	2	—	3			—
(b)	80	20	2	—	20			—
(c)	60	40	2	—	80			72
(d)	50	50	2	—	110			33
(e)	—	100	2	—	275			20
(f)	—	100	—	2	250			35
(g)	50	50	0.2	—	185			n.a.
(h)	50	50	—	0.2	160			n.a.

---

Monomer synthesis and characterisation

The reaction steps performed to obtain the new methacrylate (6) and methacrylamide (10) monomers incorporating the cyanide sensing motifs were straightforward and well documented from an organic chemistry viewpoint and produced the monomers from widely available chemicals in good yield and purity. These monomers were incorporated as comonomers in the dense crosslinked membranes at a molar percentage lower than 2%, meaning that their presence in the membrane network had a negligible impact in the mechanical properties, or handling, of the materials due to their low molar and weight contents. Purification by column chromatography yielded monomers of high purity, as demonstrated by NMR spectra. The intermediates and monomers were characterised by IR, ¹H and ¹³C NMR spectroscopy, thereby fully confirming the chemical structure of all the products. As an illustrative example, Fig. 1 depicts the ¹H NMR spectra of the monomers.

Fig. 1 ¹H NMR of the monomers containing the sensing motifs (DMSO-d₆, * = solvent signals, multiplets corresponding to protons 7 of monomer (6) were observed at 3.4 ppm, with water signal).

The hydrophobic and hydrophilic monomers (1) and (2) represented the 98.0% or the 99.8% molar content of the membranes, excluding the crosslinker. Thus, monomer (1) was chosen because it is a commercial chemical and (2) was chosen for reasons including the following: (a) their easy and cheap preparation and purification; (b) their hydrophilicity; (c) their full compatibility with (1), thus permitting the design of a membrane with a la carte hydrophilicity (i.e., the water uptake or gel behaviour); and (d) the good handling of the crosslinked membranes obtained with it, both as dry or water-swelled materials.

Membrane preparation and characterisation

Structural monomers (1) and (2), as well as the crosslinker, were fully miscible and could be mixed in any molar ratio. On the other hand, the monomers containing the sensing motifs (6) and (10) were highly insoluble and could not be solved in common organic solvents other than DMF and DMSO. Thus, preparation of the dense membrane started with a solution of the radical initiator and the mixture of monomers (1) and (2) (600 μL), to which the functional monomers (6) or (10) were added in small quantities of DMF (500 μL). Lowering the amount of DMF gave rise to precipitation or insolubilisation of (6) or (10), and an increase in the DMF content led to excessive plasticisation of the membrane as well as to handling problems immediately after demoulding. The homogeneous polymerisation mixture was then nitrogen bubbled, to eliminate the solved oxygen, and immediately injected in silanised glass moulds, and upon heating, the “almost” bulk polymerisation was conducted smoothly. The silanisation of the glass mould facilitated demoulding of the plasticised material, in which DMF was first air dried and then dried under vacuum (always at rt).

The structure of the copolymer network of the membrane, the sensing material, is depicted in Scheme 2 along with a digital picture of one of the membranes, and the composition of each of the 8 membranes and their properties are shown in Table 1. The good optical properties of the membranes can be appreciated in the picture inset of Scheme 2.

The membranes were characterised in the dry solid state by ATR-FTIR showing the absorption bands corresponding to the stretching and bending of the groups presented in the chemical structure, as depicted in Fig. 2 for membrane (d).

Fig. 2 ATR-FTIR of membrane (d) before (black line) and after (red line) hindering in a water cyanide solution.

Membrane hydrophilicity. The membrane hydrophilic or hydrophobic character was related to the water-swelling percentage (WSP); that is, it was related to the weight percentage of water uptake by the films upon soaking until equilibrium in pure water at rt.²⁹Fig. 3 shows a clear relationship between the WSP and the molar composition ratio (MCR) of the hydrophilic monomer. Thus, excluding membrane (a), which had no content of hydrophilic monomer, the relationship of WSP and MCR was linear, indicating that the WSP was intimately related with the material hydrophilicity and can be easily tuned.

Fig. 3 Water-swelling percentage (WSP) of membranes (a) to (e) vs. the molar composition ratio (MCR) of the hydrophilic monomer 2 (the code of each membrane, as shown in Table 1, is depicted close to each symbol). The straight line is a linear fit of the data corresponding to membranes with a molar content of 1 lower than 100%.

Thermal properties. The membranes showed reasonably good thermal resistance in a nitrogen atmosphere, as seen by the thermogravimetric technique. A first degradation onset was observed at approximately 180 °C, a second at approximately 300 °C, and a 10% weight loss was observed around 330 °C. The weight loss pattern showed an initial loss probably corresponding to water molecules obtained upon thermal condensation of the lateral hydroxyl motifs, followed by the loss of the pendant of the methacrylic group followed by the concomitant loss of the methacrylamide side chains. The char yield at 800 °C was near 0% due to the aliphatic nature of the main chain (a sample TGA curve can be found in the ESI†).

Mechanical properties. All of the membranes were transparent, and their mechanical properties were good, both as dry and as water-swelled materials. Membranes of 80 × 50 × 0.1 mm (length, height, width) were easy to handle and could be transformed into appropriate sensing strips of approximately 20 × 5 × 0.1 mm using a home cutter or scissors. These strips showed Young's modulus of approximately 100 MPa and tensile strength of approximately 7 MPa (films (d) and (g)), as determined from their strain–stress patterns.

Cyanide sensing

Depending on the molar composition ratio (MCR) of the monomers used for preparation of the films, their observed colours were brownish or reddish [films (a) to (f)] or light violet [films (g) and (h)], both when dry and when water-swelled. Upon soaking in aqueous solutions of cyanide, the films turned colourless in less than 72 seconds, which represents a low apparent response time,²⁸ with the lower values corresponding to the higher hydrophilic character of the membrane, depicted in Table 1 and Fig. 2 by the water-swelling percentage (WSP). Fig. 4 depicts the evolution of the colour of a membrane strip upon soaking in an aqueous solution of cyanide.

Fig. 4 Colour disappearance of a film strip of the membrane (d) upon soaking in an aqueous solution of cyanide.

The sensing mechanism. The mechanism of the colorimetric response of the membranes to the cyanide ions is depicted in Scheme 3.³⁰

Scheme 3 Schematic representation of the proposed sensing mechanism (background: digital picture of film (d)).

Membranes prepared with monomers containing the urea subgroup [monomer (6), Scheme 1] and without it [monomer (10), Scheme 1] showed a similar response toward cyanide, indicating that this group plays a negligible role in the chemodosimeter response of the fluorene derivative motif. Nevertheless, the fluorene derivative with and without the urea unit influenced the UV/Vis spectra of the corresponding membranes, as shown below. The binding characteristics of the ureagroups were probably not observed due to the aqueous media, the hydrogen bonds and water solvation and the neutral conditions.

The UV/Vis calculated spectra, by means of molecular modelling of the reaction of model compounds of copolymers, supported the proposed mechanism. The model compounds used to model the behaviour of the sensing motifs within the membrane had a similar structure to (6) and (10), replacing the methacrylate and the methacrylamidegroup by a 2-methylpropanoate or 2-methylpropanamidegroups; thus, mimicking the polymer network backbone in the membrane. Modelling of (6) in aqueous media gave rise to absorption in the visible region that disappeared upon reaction with cyanide, as depicted in Fig. 5. The experimental spectra of membrane (d) showed comparable behaviour.

Fig. 5 Theoretical calculation of UV/Vis spectra of a model molecule of monomer 6 before and after reaction with CN⁻ (UV-Vis DFT calculations at B3LYP/6-31G*, in water), along with the spectra corresponding with membrane (d) before and after immersing it in water solution of CN⁻ (10⁻¹ M), solid lines.

This mechanism was fully confirmed by means of the preparation and characterisation of 2-(9H-fluoren-9-ylidene)malononitrile and its reaction with cyanide to give 2-(9-cyano-9H-fluoren-9-yl)malononitrile, as depicted in Scheme 4. The former reacted with cyanide in a selective and quantitative manner though a Michael reaction.

Scheme 4 Synthesis of the model 2-(9-cyano-9H-fluoren-9-yl)malononitrile.

The sensing sensitivity and selectivity of the membranes. To test the colorimetric anion sensing ability of the membranes (a) to (f), different anions were added to a quartzUV/Viscell containing the films, resulting in a change in the absorption spectra with a gradual diminishment of the absorbance of the 445 nm band with the cyanide concentration. In contrast, the film remained completely silent in the presence of a broad set of other anions (benzoate, acetate, chloroacetate, methanesulfonate, p-toluenesulfonate, nitrate, carbonate, chloride, fluoride). On the other hand, the monomers and the membranes showed a colorimetric response with basic anions, such as OH⁻ and PO₄³⁻, and this response was cancelled buffering the system at neutral pH, which is especially adequate for environmental and biological applications (Fig. 6). The development of colours of urea-containing receptors upon interaction with anions was thoroughly described by Esteban-Gómez et al.³¹ The response tocyanide can be seen in Fig. 7, which shows titration of the films with aqueous CN⁻ at a physiological pH, and the inset shows the corresponding binding isotherm. The best responses in terms of response time and colour variation were obtained with film (d), probably due to its moderate hydrophilicity, giving rise to a detection limit lower than 260 ppb, which is slightly above the MCL. Membranes containing monomer (6) did not show a maximum because an intense band centred at 352 nm saturated the UV/Vis spectrometer. A real-time movie showing the colorimetric response of a membrane strip [film (d)] upon soaking it in a cyanide aqueous solution can be found in the ESI†.

Fig. 6 Solution of monomer (6) in DMSO/water (90 : 10, v/v), 2 × 10⁻³ M, before and after addition of hydroxide or phosphate anions 2 × 10⁻² M).

Fig. 7 Selected UV/Vistitration curve of film (d) with cyanide in buffered water (pH = 7.5) (inset: cyanide concentration vs. I/I₀ at 445 nm).

Upon lowering the MCR of the copolymer from 2% (films (a)–(f)) to 0.2% [films (g) and (h)], the absorption bands of the UV region were observed {a maximum with shoulders at 351 nm [film (g)], and two maxima were observed at 290 and 354 nm [film (h)]}. The progressive addition of cyanide to the cell containing the copolymer films gave rise to subsequent diminishment of the absorbance intensity of these bands, thus permitting titration and giving rise to a detection limit lower than 13 ppb for film (h), which is much lower than the MCL (Fig. 8), and 260 ppb for film (g) (Fig. 9).

Fig. 8 Selected UV/Vistitration curve of film (h) with cyanide in buffered water (pH = 7.5) (inset: absorbance intensity ratio at 290 and 354 nm (I₂₉₀/I₃₅₄) vs.cyanide concentration).

Fig. 9 Selected UV/Vistitration curve of film (g) with cyanide in buffered water (pH = 7.5) (inset: cyanide concentration vs. I/I₀ at 351 nm).

Conclusion

In summary, water-insoluble organic molecules (6) and (10), with a chemodosimeter behaviour toward the cyanide anion, were chemically incorporated in the lateral chain of a methacrylic copolymer to give different, dense membranes. The design of the membrane composition allowed for the preparation of a solid material with ‘a la carte’ hydrophilic character. Upon swelling with water, the membrane permitted the colorimetric detection ofcyanide with an extremely low detection limit. Thus, through diffusion into the swelled membrane, the solvated ions reached the highly hydrophobic sensing organic motifs, giving rise to a recognition phenomenon. Following this procedure, this supramolecular sensing phenomenon can be easily performed by means of anchoring organic sensing molecules to appropriate polymeric nets. Using this approach, we are now successfully developing membranes with the ability of detecting biologically and medically important molecules. Moreover, we are extending our studies to the preparation of linear copolymers for the water solubilisation of all types of chemicals, with an eye on the development of new drugs.

Acknowledgements

The authors gratefully acknowledge financial support provided by the Spanish Ministerio de Educación y Ciencia—Feder (MAT2008-00946/MAT) and by the Junta de Castilla y León (BU001A10-2).

References

1. N. M. Bergmann and N. Peppas, Prog. Polym. Sci., 2008, 33, 271.
2. B. N. Chen, S. Piletsky and A. P. F. Turner, Comb. Chem. High Throughput Screening, 2002, 5, 409.
3. A. Tulinsky, Semin. Thromb. Hemostasis, 1996, 22, 117.
4. M. Britschgi, S. von Greyerz, C. Burkhart and W. J. Pichler, Curr. Drug Targets, 2003, 4, 1.
5. P. Cudic, D. C. Behenna, J. K. Kranz, R. G. Kruger, A. J. Wand, Y. I. Veklich, J. W. Weisel and D. G. McCafferty, Chem. Biol., 2002, 9, 897.
6. E. J. Sundberg and R. A. Mariuzza, J. Protein Chem., 2002, 61, 119.
7. R. Jimenez, G. Salazar, K. K. Baldridge and F. E. Romesberg, Proc. Natl. Acad. Sci. U. S. A., 2003, 100, 92.
8. S. A. Hofstadler and R. H. Griffery, Chem. Rev., 2001, 101, 377.
9. R. Martínez-Máñez and F. Sancenón, Chem. Rev., 2003, 103, 4419.
10. F. García, J. M. García, B. García-Acosta, R. Martínez-Mañez, F. Sancenón, N. San-José and J. Soto, Chem. Commun., 2005, 2790.
11. B. García-Acosta, F. García, J. M. García, R. Martínez-Mañez, F. Sancenón, N. San-José and J. Soto, Org. Lett., 2007, 9, 2429.
12. J. M. García, F. C. García, F. Sena and J. L. de la Peña, Prog. Polym. Sci., 2010, 35, 623.
13. K. C. Persaud, Mater. Today, 2005, 8, 38.
14. H. N. Kim, Z. Guo, W. Zhu, J. Yoon and H. Tian, Chem. Soc. Rev., 2011, 40, 79.
15. K. W. Kulig, in Cyanide Toxicity, US Department of Health and Human Services, Atlanta, GA, 1991; Guidelines for Drinking-Water Quality, World Health Organization, Geneva, 1996.
16. For a recent review see: Z. Xu, X. Chen and J. Yoon, Chem. Soc. Rev., 2010, 39, 127.
17. S. Vallejos, P. Estévez, F. C. García, F. Serna, J. L. de la Peña and J. M. Garacía, Chem. Commun., 2010, 46, 7951.
18. T. Agou, M. Sekine, J. Kobayashi and T. Kawashima, J. Organomet. Chem., 2009, 694, 3833; V. Kumar, M. P. Kaushika, A. K. Srivastava, A. Pratap, V. Thiruvenkatam and T. N. Row, Anal. Chim. Acta, 2010, 663, 77; J. Wang and C.-S. Ha, Tetrahedron, 2010, 66, 1846.
19. See, for instance: T. Ábalos, S. Royo, R. Martínez-Máñez, F. Sancenón, J. Soto, A. M. Costero, S. Gil and M. Parra, New J. Chem., 2009, 33, 1641; H.-T. Niu, D. Su, X. Jiang, W. Yang, Z. Yin, J. He and J.-P. Cheng, Org. Biomol. Chem., 2008, 6, 3038; Y. Sun, G. Wang and W. Guo, Tetrahedron, 2009, 65, 3480; S. K. Kwon, S. Kou, H. N. Kim, X. Chen, H. Hwang, S.-W. Nam, S. H. Kim, K. M. K. Swamy, S. Park and J. Yoon, Tetrahedron Lett., 2008, 49, 4102; D.-G. Cho, J. H. Kim and J. L. Sessler, J. Am. Chem. Soc., 2008, 130, 12163; J. L. Sessler and D.-G. Cho, Org. Lett., 2008, 10, 73; J. Ren, W. Zhu and H. Tian, Talanta, 2008, 75, 760; S.-J. Hong, J. Yoo, S.-H. Kim, J. S. Kim, J. Yoon and C.-H. Lee, Chem. Commun., 2009, 189; Y.-K. Yang and J. Tae, Org. Lett., 2006, 8, 5721; G. Qian, X. Li and Z. Y. Wang, J. Mater. Chem., 2009, 19, 522; J. O. Huh, Y. Do and M. H. Lee, Organometallics, 2008, 27, 1022; T. Agou, M. Sekine, J. Kobayashi and T. Kawashima, Chem.–Eur. J., 2009, 15, 5056; M. Jamkratoke, V. Ruangpornvisuti, G. Tumcharen, T. Tuntulani and B. Tomapatanaget, J. Org. Chem., 2009, 74, 3919.
20. L. Peng, M. Wang, G. Zhang, D. Zhang and D. Zhu, Org. Lett., 2009, 11, 1943; Y. Sun, Y. Liu and W. Guo, Sens. Actuators, B, 2009, 143, 171; J. Jo and D. Lee, J. Am. Chem. Soc., 2009, 131, 16283; Y. Sun, Y. Liu, M. Chen and W. Guo, Talanta, 2009, 80, 996; H.-T. Niu, X. Jiang, J. He and J.-P. Cheng, Tetrahedron Lett., 2009, 50, 6668; K.-S. Lee, H.-J. Kim, G.-H. Kim, I. Shin and J.-I. Hong, Org. Lett., 2008, 10, 49.
21. J. H. Lee, A. R. Jeong, I.-S. Shin, H.-J. Kim and J.-I. Hong, Org. Lett., 2010, 12, 764; R. Guliyev, O. Buyukcakir, F. Sozmen and O. A. Bozdemir, Tetrahedron Lett., 2009, 50, 5139; L. Shang, L. Zhang and S. Dong, Analyst, 2009, 134, 107; F. H. Zelder, Inorg. Chem., 2008, 47, 1264; C. Männel-Croisé and F. Zelder, Inorg. Chem., 2009, 48, 1272.
22. X. Lou, L. Zhang, J. Qin and Z. Li, Chem. Commun., 2008, 5848.
23. X. Chen, S. W. Nam, G. H. Kim, N. Song, Y. Jeong, I. Shin, S. K. Kim, J. Kim, S. Park and J. Yoon, Chem. Commun., 2010, 46, 8953.
24. Z. Xu, J. Pan, D. R. Spring, J. Cui and J. Yoon, Tetrahedron, 2010, 66, 1678; L. Shang, L. Jin and S. Dong, Chem. Commun., 2009, 3077; P. Kaur, S. Kaur and K. Singh, Inorg. Chem. Commun., 2009, 12, 978; Y. Liu, K. Ai, X. Cheng, L. Huo and L. Lu, Adv. Funct. Mater., 2010, 20, 951; X. Lou, J. Qin and Z. Li, Analyst, 2009, 134, 2071; C. Männel-Croisé, B. Probst and F. Zelder, Anal. Chem., 2009, 81, 9493; S.-Y. Chung, S.-W. Nam, J. Lim, S. Park and J. Yoon, Chem. Commun., 2009, 2866.
25. N. Gimeno, X. Li, J. R. Durrant and R. Vilar, Chem.–Eur. J., 2008, 14, 3006.
26. Q. Zeng, P. Cai, Z. Li, J. Qina and B. Z. Tang, Chem. Commun., 2008, 1094; Z. Li, X. Lou, H. Yu, Z. Li and J. Qin, Macromolecules, 2008, 41, 7433; Z. Ekmekci, M. D. Yilmaz and E. U. Akkaya, Org. Lett., 2008, 10, 461.
27. A. Touceda-Varela, E. I. Stevenson, J. A. Galve-Gasión, D. T. F. Dryden and J. C. Mareque-Rivas, Chem. Commun., 2008, 1998.
28. P. Kaur, D. Sareen, S. Kaur and K. Singh, Inorg. Chem. Commun., 2009, 12, 272.
29. V. Compañ, P. Tiemblo, F. García, J. M. García, J. Guzmán and E. Riande, Biomaterials, 2005, 26, 3783.
30. H. D. Hartzler, J. Org. Chem., 1966, 31, 2654.
31. D. Esteban-Gómez, L. Fabbrizzi and M. Licchelli, J. Org. Chem., 2005, 70, 5717.

---

Footnote

† Electronic supplementary information (ESI) available: Intermediates and monomer NMR and IR-FT spectra, a TGA curve, as well as a movie showing the colorimetric response of a film (d) toward cyanide. See DOI: 10.1039/c1py00013f

---