Colorimetric detection and determination of Fe(III), Co(II), Cu(II) and Sn(II) in aqueous media by acrylic polymers with pendant terpyridine motifs

Acceptedmanuscrip.Publishedmanuscriplink: http://www.sciencedirect.com/science/article/pii/S0925400515306948 doi:10.1016/j.snb.2015.11.116  Colorimetric detection and determination of Fe(III), Co(II), Cu(II) and Sn(II) in aqueous media by acrylic polymers with pendant terpyridine motifs Miriam Trigo-López, Asunción Muñoz, Saturnino Ibeas, Felipe Serna, Félix Clemente García and José Miguel García* Departamento de Química, Facultad de Ciencias, Universidad de Burgos, Plaza de Misael Bañuelos s/n, 09001 Burgos, Spain. Fax: (+) 34 947 258 831, Tel: (+) 34 947 258 085. E-mail: [email protected] Graphical abstract ABSTRACT Colorimetric cation responsive water soluble polymers and manageable films or membranes have been designed. The sensory materials respond with a colour change to the presence in water of Fe(III), Co(II), Cu(II), and Sn(II). The colour change is specific of each metal cation, and enables its identification (purple for iron, orange for cobalt, green for copper, and yellow for tin). The design of the materials relies on an addition monomer having a terpyridine moiety, which behaves as a dye in presence of transition metal cations due to its proven chelating capability toward these species and the colour development that always accompany the metallic complex formation. Water solutions of the sensory linear polymers allow for the UV/vis titration of Fe(III), Co(II), Cu(II), 1 Acceptedmanuscrip.Publishedmanuscriplink: http://www.sciencedirect.com/science/article/pii/S0925400515306948 doi:10.1016/j.snb.2015.11.116 2 and Sn(II) with a limit of detection of 1.3x10-7, 6.4x10-8, 1.3x10-5 and 1.4x10-5 M, respectively. On the other hand, sensory kits, cut from sensory membranes, permitted the visual quantification of the cations in a dynamic range of five decades (1x10-7 - 5x10-3 M) for Fe(III) and Co(II) and of two decades (9x10-5 - 9x10-3 M) for Cu(II) and Sn(II). Titration curves can also be drawn from a picture taken to the sensory kits with a smartphone, by using the digital colour definition of the materials as analytical signal. Also, after entering into contact with hands, shapes of metallic objects (iron and cobalt containing tools) can be colour revealed by pressing the hands on paper or cotton fabrics wetted with water solutions of the linear sensory polymer. Keywords Sensory polymers, terpyridine, cation detection, forensic applications 1. INTRODUCTION The costless, in-situ, and fast detection and quantification of transition metal cations in pure water are of the utmost environmental, industrial, and health importance. Traditional techniques, such as atomic absorption spectroscopy (AAS) or inductively coupled plasma mass spectrometry (ICP-MS), enable the selective and precise detection and quantification of the mentioned chemical species. However, they are heavy, bulky, extremely expensive techniques and require trained personnel. On the other hand, chemical sensors have become a simple species detection method for non-trained personnel, especially if the transduction is chromogenic (i.e., by a colour change) and the detection can be carried out visually. Chemical sensors are an emerging technology with expanding applicability to a number of fields, such as civil security, environmental control and remediation, medicine, and industrial control. Moreover, sensory polymers, which are Acceptedmanuscrip.Publishedmanuscriplink: http://www.sciencedirect.com/science/article/pii/S0925400515306948 doi:10.1016/j.snb.2015.11.116 3 macromolecules that have receptor motifs (or binding sites) in their structure, represent a step further and show significant advances over discrete (or low molecular mass) chemosensors. Thus, polymers can be prepared or transformed into films, coatings or finished sensory materials with different shapes. Distinct polymer geometries are achievable (linear, spherical, and tridimensional crosslinked network). They can be easily designed to work in hydrophobic or hydrophilic environments and can be used to sense both vapours and liquids. Also, they exhibit collective properties sensitive to minor perturbations. Finally, their sensory moieties cannot migrate with the concomitant increase in the performance stability along time, improve in the thermal and chemical resistance, and can be easily reused. Chemosensory polymers following a research methodology based on a guaranteed of success strategy have been prepared. That is to say, once chemical species to be detected are selected, the so-called targets, fully confident receptors for such targets are looked in scientific literature to find, usually, discrete organic molecules that are insoluble in water; then, their chemical structure is slightly modified by including a polymerizable group; and finally it is copolymerized with commercial hydrophilic and hydrophobic monomers to have water soluble linear polymers and crosslinked membranes with gel structures that allow for the detection in 100% water. The target species for this work are transition metal cations (Fe(III), Co(II), Cu(II) and Sn(II)), and the receptor core is based on a terpyridine (2,2’:6’,2’’-terpyridine, tpy) motif. Morgan and Burstal isolated tpy and described its purple complex with iron(II) in the 1930s [1]. Since then tpy has become one of the most used ligands with multiple applications in different research and technological areas, such as coordination polymers [2-4]; sensors for anion [5-8], cations [9-17], both[18,19], and biomolecules Acceptedmanuscrip.Publishedmanuscriplink: http://www.sciencedirect.com/science/article/pii/S0925400515306948 doi:10.1016/j.snb.2015.11.116 4 [20,21]; gelation and solvochromic sensors [22]; luminescent converters [23-25]; photon harvesting [26] and catalysis [27]. Terpyridine derivatives are multivalent pyridine ligands that exhibit strong binding affinity toward a broad set of transition metal cations. This extremely strong interactions come both from the d-d* back bonding of the cations to the N-heterocycle rings and from the chelate effect [28,29]. Here, the tpy structure has been modified with a polymerizable methacrylamide group and two kind of chromogenic sensory materials have been prepared: linear polymers and solid film-shaped dense membranes comprised of crosslinked polymer networks. The tpy-monomer (~1% mol) was copolymerized with a balance of hydrophilic and hydrophobic commercial co-monomers (~99% mol) to give linear copolymers and networks (membranes). The linear polymers are water soluble. The membranes are solid, exhibit gel behaviour, and can be used to prepare manageable solid sensory kits. Both type of materials respond with development of different colour depending on the presence of Fe(III), Co(II), Cu(II) and Sn(II) in 100% water under controlled acidic conditions. The detection in 100% water is relevant for real-life environmental applications, whereas the acidic conditions avoid the analytical complexity of dealing with presence of different species of each cation (for instance, for iron at pH 2 only Fe3+ species are present, while at pH > 2 there are different equilibria of Fe3+, Fe(OH)2+, Fe(OH)2+, and Fe(OH)3). Also, the usual presence in most research studies of organic solvents, due to the insolubility in water of conventional probes, influence the selectivity of the tpy core [30]. 2. EXPERIMENTAL PART 2.1. Materials Acceptedmanuscrip.Publishedmanuscriplink: http://www.sciencedirect.com/science/article/pii/S0925400515306948 doi:10.1016/j.snb.2015.11.116 5 All materials and solvents were commercially available and used as received, unless otherwise indicated: 2-acetylpyridine (98%, Alfa Aesar), iodine (≥ 99.8%, SigmaAldrich), 4-nitrobenzaldehyde (99% Alfa Aesar), ammonium acetate (97%, Alfa Aesar), tin chloride anhydrous (98% , Alfa Aesar), methacryloyl chloride (97%, Alfa Aesar), triethylamine (TEA) (≥ 99%, Aldrich), pyridine (≥ 99%, Probus), sodium hydroxide (99.9%, VWR-Prolabo), hydrochloric acid (37%, VWR-Prolabo), N-methyl-2pyrrolidone (NMP) (99%, Aldrich), ethanol (99.97%, VWR-Prolabo), diethyl ether (≥ 99,5%, Aldrich), SnCl2 anhydrous (98%, Alfa Aesar), Fe(NO3)3·9H2O (VWR-Prolabo), Cu(NO3)2·3H2O (98%, Sigma-Aldrich), Co(NO3)2·6H2O (≥ 99%, Labkem), NaCl (≥ 99%, Sigma-Aldrich), KCl (99.5%, Scharlau), Al(NO3)2·9H2O (≥ 989%, SigmaAldrich), Pb(NO3)2 (≥ 99%, Fluka), LiCl (≥ 99%, Sigma-Aldrich), Zn(NO3)2·6H2O (98%, Aldrich), Mg(NO3)2·6H2O (≥ 99%, Labkem), Cd(NO3)2 (98.5%, Alfa Aesar), Ni(NO3)2·6H2O (98.5%, Sigma-Aldrich), methyl methacrylate (MMA) (99%, Aldrich), 1-vinyl-2-pyrrolidone (VP) (≥ 99%, Sigma-Aldrich), ethylene glycol dimethacrylate (EGDMMA) (98%, Sigma-Aldrich), Azo-bis-isobutyronitrile (AIBN, ≥ 98%, Aldrich) was recrystallised twice from methanol. 2.2. Instrumentation and 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 chloroform (CDCl3) as the solvent. UV/vis spectra were recorded using a Hitachi U-3900 UV/vis spectrophotometer. Acceptedmanuscrip.Publishedmanuscriplink: http://www.sciencedirect.com/science/article/pii/S0925400515306948 doi:10.1016/j.snb.2015.11.116 6 Infrared spectra (FT-IR) were recorded with a FT/IR-4200 FT-IR Jasco spectrometer with an ATR-PRO410-S single reflection accessory. Low-resolution electron impact mass spectra (EI-LRMS) were obtained at 70 eV on an Agilent 6890N mass spectrometer. Thermogravimetric analysis (TGA) data were recorded for a 5-mg sample under a nitrogen or oxygen atmosphere on a TA Instrument Q50 TGA analyser at a scan rate of 10ºC min-1. The limiting oxygen index (LOI) was estimated using the following experimental Van Krevelen equation: LOI = 17.5 + 0.4 CR, where CR is the char yield weight percentage at 800ºC, which was obtained from the TGA measurements under a nitrogen atmosphere. The water-swelling percentage (WSP) of the membrane was obtained from the weights of a dry sample membrane (ωd) and a water-swelled sample membrane (ωs) as follows: 100 x [(ωs-ωd)/ωd] (the membrane was immersed in pure water at 20ºC until the swelled equilibrium was achieved). To determine the tensile properties of the polymer films (membranes), strips (5 mm in width and 30 mm in length) were cut from polymer films of 112 and 115µm thickness for Mem1 and Mem2, respectively, on a SHIMADZU EZ Test Compact Table-Top Universal Tester at 20ºC. Mechanical clamps were used and an extension rate of 5 mm min-1 was applied using a gauge length of 9.44 mm. At least 6 samples were tested for each polymer, and the data was then averaged. The limit of detection (LOD) and limit of quantification were estimated by the following equations: LOD = 3.3 x SD/s and LOQ = 10 x SD/s, where SD is the standard deviation of a blank sample and s is the slope of the calibration curve in a region of low concentration of target species. Acceptedmanuscrip.Publishedmanuscriplink: http://www.sciencedirect.com/science/article/pii/S0925400515306948 doi:10.1016/j.snb.2015.11.116 7 The qualitative and quantitative chromogenic responses of sensory squares (~5x5 mm) cut from membranes (Mem1 and Mem2) toward Fe(III), Co(II), Cu(II) and Sn(II) in water solution were studied by immersing the squares in a number of sealed vials with 1 mL of buffered water, containing each vial a known concentration of one of the target cations (pH = 2, buffer: KCl-HCl). The resident time was 24 hours and the temperature 25 °C. The qualitative evaluation of the sensing performance of the materials was carried out visually. On the other hand, the quantitative study was performed using a digital picture of the sensory squares taken with a smartphone by treatment of the colour definition data of each disc (RGB parameters, R = purple, G = green, B = blue). These parameters were obtained for each square directly after taking the photograph of the set squares through using the app called ColourMeter of a conventional Android smartphone (for each square 121 (11 × 11) pixels were averaged). The three RGB parameters were reduced to one variable (PC1, principal component 1), using principal component analysis (PCA), which provided an account of >78% of the information on the three RGB parameters, thus allowing for the elaboration of simple 2D titration curves ([cation] vs. PC1) with concomitant noise reduction, without a significant loss of information, and with independence of the type of camera, lighting, quality of the image and so on [31,32]. 2.3. Synthesis of sensory monomer The sensory monomer containing the tpy-motif (5) was prepared according with the procedure schematically shown in Scheme 1. Scheme 1. Acceptedmanuscrip.Publishedmanuscriplink: http://www.sciencedirect.com/science/article/pii/S0925400515306948 doi:10.1016/j.snb.2015.11.116 8 Synthesis of 3-(4-nitrophenyl)-1-(pyridine-2-yl)prop-2-en-1-one (1). 2.5 mL of an aqueous solution of 10% NaOH was added to a suspension of 6.25 g (41.4 mmol) of 4nitrobenzaldehyde in 50 mL of ethanol. To the resulting mixture cooled at 0 ºC, 5.0 g (41.2 mmol) of 2-acetylpyridine was added dropwise for 3 h. The solution was stirred at 0 ºC for 2 h. The precipitate formed was collected by filtration and washed with ethanol. Yield 6.91 g (66%). 1H NMR H (400 MHz, CDCl3, Me4Si): 8.75 (1H, d, J 4.7 Hz, pyridyl-H); 8.42 (1H, d, J 16.1 Hz, CH=CH); 8.26 (2H, d, J 8.8 Hz, Ph); 8.19 (1H, d, J 7.9 Hz, pyridyl-H); 7.91 (1H, d, J 16.1 Hz, CH=CH); 7.90 (1H, td, J 7.7 Hz, 1.7 Hz, pyridyl-H); 7.85 (2H, d, J 8.9 Hz, Ph); 7.52 (1H,ddd, J 7.5 Hz, 4.7 Hz, 1.2 Hz, pyridylH). 13C NMR, C (100.6 MHz, CDCl3, Me4Si): 189.06, 153.73, 149.11, 148.67, 141.44, 141.36, 137.32, 129.36, 127.46, 124.94, 124.25, 123.20. EI-LRMS (m/z (%)): 255.07 (18), 254.07 (M+, 100), 226.07 (23), 225.06 (68), 180.07 (18), 179.07 (31), 130.04 (16), 102.04 (29), 79.03 (19), 78.03 (20). FTIR [Wavenumbers (cm-1)]: ar C-H: 3075; C=O: 1671; C=N: 1578; as NO2: 1511; s NO2: 1334. Synthesis of 1-pyridylacylpyridinium iodide (2). To a solution of 2 g (15.6 mmol) of 2acetylpyridine in 20 mL of pyridine 4.60 g (17.6 mmol) of I2 was added and heated at 100 ºC under N2 atmosphere for 3 h. The mixture was then cooled at room temperature and filtered off and washed with ether. The dry solid was then washed with ethanol. A black solid was obtained. It was immediately used in the next synthetic step because it is sensitive to ambient conditions. Yield 4.03 g (75%). 1H NMR H (300 MHz, CDCl3, Me4Si): 8.75 (1H, d, J 4.7 Hz, pyridyl-H); 8.42 (1H, d, J 16.1 Hz, CH=CH); 8.26 (2H, d, J 8.8 Hz, Ph); 8.19 (1H, d, J 7.9 Hz, pyridyl-H); 7.91 (1H, d, J 16.1 Hz, CH=CH); 7.90 (1H, td, J 7.7 Hz, 1.7 Hz, pyridyl-H); 7.85 (2H, d, J 8.9 Hz, Ph); 7.52 (1H, ddd, J 7.5 Hz, 4.7 Hz, 1.2 Hz). Acceptedmanuscrip.Publishedmanuscriplink: http://www.sciencedirect.com/science/article/pii/S0925400515306948 doi:10.1016/j.snb.2015.11.116 9 Synthesis of 2-(4-(4-nitrophenyl)-6-(pyridin-2-yl)pyridin-2-yl)pyridine (3). To a solution of 10% dry ammonium acetate in 20 mL of ethanol, 1 g (3 mmol) of (2) and 0.78 g (3 mmol) (1) were added. The mixture was refluxed for two days, the solvent removed and the product used without further purification. Yield 1.00 g (93%). 1H NMR H (400 MHz, CDCl3, Me4Si): 8.75 (2H, pyridyl-H); 8.73 (2H, d, J 4.3 Hz, pyridyl-H); 8.68 (2H, d, J 8.0 Hz, pyridyl-H); 8.36 (2H, d, J 8.8 Hz, Ph); 8.04 (2H, d, J 8.8 Hz, Ph); 7.90 (2H, td, J 7.8 Hz, 1.6 Hz, pyridyl-H); 7.38 (2H, pyridyl-H). 13C NMR, C (100.6 MHz, CDCl3, Me4Si):156.53, 155.81, 149.35, 148.33, 148.02, 145.12, 137.17, 128.43, 124.34, 124.34, 121.56, 119.07. EI-LRMS (m/z (%)):355.11 (25), 354.11 (M+, 100), 309.12 (20), 308.12 (78), 306.09 (9), 229.07 (57), 203.06 (4), 177.05 (4), 153.54 (5), 78.02 (5). FTIR [Wavenumbers (cm-1)]: C=N: 1585; as NO2: 1514; s NO2: 1351. Synthesis of 4-(2,6-di(pyridin-2-yl)pyridin-4-yl)benzenamine (4). A mixture of 3.66 g (10.3 mmol) of (3) and 12.24 g (64.5 mmol) of anhydrous tin(II) chloride in concentrated hydrochloric acid (100 mL) was heated at 70 °C for 6 h. The solid was filtered off and stirred in a 10% aqueous solution of sodium hydroxide for 1 hour. Then it was filtered off and washed with water. Yield 3.07 g (92%). 1H NMR H (400 MHz, CDCl3, Me4Si): 8.73 (2H, ddd, J 4.8 Hz, 1.8 Hz, 0.9 Hz, pyridyl-H); 8.69 (2H, s, pyridyl-H); 8.67 (2H, dt, J 8.0 Hz, 1.1 Hz, pyridyl-H); 7.87 (2H, td, J 7.7 Hz, 1.8 Hz pyridyl-H); 7.78 (2H, d, J 8.5 Hz, Ph); 7.35 (2H, ddd, J 7.5 Hz, 4.8 Hz, 1.2 Hz, pyridylH); 3.87 (2H, s, NH2). 13C NMR, C (100.6 MHz, CDCl3, Me4Si):156.70, 155.88, 150.14, 149.22, 147.67, 136.95, 128.53, 128.38, 123.81, 121.49, 117.93, 115.37. EILRMS (m/z (%)):325.13 (24), 324.13 (M+, 100), 323.12 (19), 296.11 (6), 246.10 (15), 219.09 (6), 162.09 (6), 78 (4). FTIR [Wavenumbers (cm-1)]: NH2: 3481, 3387; C=N: 1599; NH2: 1338; C-N: 1185. Acceptedmanuscrip.Publishedmanuscriplink: http://www.sciencedirect.com/science/article/pii/S0925400515306948 doi:10.1016/j.snb.2015.11.116  16  6.4x10-8 and 1.9x10-7 M for Co(II)). On the other hand, the response to Cu(II) and Sn(II) was rather complex with LOD and LOQ in the sub milimolar level (1.3x10-5 and 3.8x10-5 M for Cu(II) and 1.4x10-5 and 4.2x10-5 M for Sn(II)). Only the low concentration region is shown in Figure 4. Full titration curves and UV/vis titration spectra are shown in Figures S6 to S9, ESI. Figure 2. Figure 3. Figure 4. Solid sensory materials give similar colours upon entering into contact with the cations (Fe(III), Co(II), Cu(II) and Sn(II)), and the differences between solution an solid state for each cation, shown in Figure 2, correspond to concentration differences and colour perception and digitalization. Thus, these materials permitted the preparation of solid titration kits for the visual detection and quantification of cations. Accordingly, squares cut from Mem1 allowed for the naked eye study of Fe(III) and Co(II) content by analysing the purple and orange colour development, respectively (Figure 5a). The lower sensitivity of the sensory polymers toward Cu(II) and Sn(II), compared with the sensitivity toward Fe(III) and Co(II), was resolved by increasing the tpy sensory motifs within the membranes. Therefore, Mem2, which have a molar tpy content four times higher than Mem1, permitted the visual titration of Cu(II) and Sn(II) by following the green and yellow colour development, respectively. The digital colour of pictures of sensory squares (sensory kits) shown in Figures 5a and 5b were used to build titration Acceptedmanuscrip.Publishedmanuscriplink: http://www.sciencedirect.com/science/article/pii/S0925400515306948 doi:10.1016/j.snb.2015.11.116  17  curves with quantification purposes, as previously described. The titration curve for Fe(III) is shown in Figure 5c (the data and titration curves for Co(II), Cu(II) and Sn(II) are depicted in the ESI, Tables S1-S12 and Figures S10-S13) [31,32]. Figure 5. 3.6. Response time There are key parameters for sensor performance for real live measurements: reliability, accurateness, environmental inertness, and short response time. The response time was calculated as follows: for LCp solutions by UV/vis spectroscopy as the time needed to achieve 99% of the absorbance variation (ESI, Figure S14). This time was 10, 6, 2, and 12 min for Fe(III), Co(II), Cu(II), Sn(II) respectively (cations concentration of 5x10-6, 5x10-5, 1x10-4, 7.5x10-4 M, respectively). The apparent response time of the sensory films (membranes) was slower because of the diffusion of the species into the membrane, and the sensory squares were left immersed overnight. 3.7. Interference study The selectivity of the sensory materials as colorimetric transition metal cation sensors was tested in the presence of a broad set of cations (NaCl, KCl, Cu(NO3)2·3H2O, Co(NO3)2·6H2O, Al(NO3)2·9H2O, Pb(NO3)2, LiCl, Zn(NO3)2·6H2O, Mg(NO3)2·6H2O, Cd(NO3)2, SnCl2 and Ni(NO3)2·6H2O, SnCl2, Fe(NO3)3·9H2O,Cu(NO3)2·3H2O, Co(NO3)2·6H2O). Thus, a solution of LCp in water (pH = 2, buffer: KCl-HCl, concentration of sensory motifs in water was 1x10-3 M -equivalents of pendant sensory tpy motifs per litre-) with a cocktail of these cations (concentration of each cation = 9.09x10-4 M) was used in this study. The UV/vis spectra show that the set of cations are true interferents (ESI, Figure S15). This is a cause of the broad chelating effect of tpy Acceptedmanuscrip.Publishedmanuscriplink: http://www.sciencedirect.com/science/article/pii/S0925400515306948 doi:10.1016/j.snb.2015.11.116  18  motifs. This means that a full range of complexes are formed, though only a few are coloured. Accordingly, real sample can contain these ions at different concentrations and this fact may be a drawback for the identification and quantification of the target cations. However, the target cations can also be evaluated in presence of these interferents by the standard addition procedure. For instance, by adding increasing and known quantities of Fe(III) to the sample containing the cocktail of 13 cations (concentration of each cation = 1x10-6 M), the Fe(III) concentration calculated was 1.3x10-6 M, similar to the real content, 1.0x10-6 M (ESI, Figure S16). Furthermore, the accuracy of the Fe(III) concentration determination was verified by measuring the cation concentration in a sample prepared with tap water from our laboratory spiked with Fe(III), thus emulating a real sample. The titration curve was prepared using tap water (the iron concentration of this tap water, determined by inductively coupled plasma mass spectrometry -ICP-, was about 19.4 ppb). The Fe(III) concentration added to the tap water was 1.5x10−6 M, and the calculated concentration was 1.4x10−6 M, in good accordance with the added Fe(III) concentration. In parallel, the standard addition method permitted the calculation of the iron concentration of tap water, giving a result of 20.3 ppb, also in agreement with the concentration calculated by ICP. 3.8. Forensic applications and metal recognition Iron containing tools leave a small amount of metal upon entering into contact with other surfaces. This is relevant in forensic applications where it is important to detect the imprint of certain objects in hands, like knives. In this sense, Figure 6 and a video shows the shape of a lag screw in a finger after holding it with two fingers. After pressing lag screw with the fingers, the index finger is pressed on a filter paper impregnated with an aqueous solution of LCp. The iron particles are oxidized by the Acceptedmanuscrip.Publishedmanuscriplink: http://www.sciencedirect.com/science/article/pii/S0925400515306948 doi:10.1016/j.snb.2015.11.116  19  oxygen of the air, the sweat and water and recognized by the tpy motifs giving rise to the complex formation with the concomitant purple colour development (see video, ESI). Moreover, it is possible to know if a tool has iron in its composition by dipping in a water solution of LCp. Immediately after immersion, the solution turn purple (see Figure 6 and video, ESI). Cobalt, a critical raw material [40], can also be analysed in a similar way. Figure 6 and the video (ESI) also shows how a widia drill bit, which has an iron-based shaft with a tungsten carbide (with 6-10% of cobalt) blade at the tip off the drill. After being touched with a finger, the widia blade leave on the filter paper, previously wet with the water solution of LCp, its orange imprint surrounded by the purple imprint of the iron holding the widia tip. Figure 6. 4. CONCLUSIONS Terpyridine, as a proven chelating agent for monoatomic metallic cations, has been used to prepare colorimetric sensory solid polymers both as water soluble linear polymers and as solid films (membranes). Water solution of the linear polymer turned purple, orange, green or yellow upon being in contact with Fe(III), Co(II), Cu(II) and Sn(II), respectively. The colour development permitted the visual differentiation of the cations and the titration with the UV/vis technique. The limit of detection was sub-micromolar for Fe(III) and Co(II), and micromolar for Cu(II) and Sn(II). In a similar fashion, the solid films have gel behaviour and turn coloured upon immersing in water containing the aforementioned cations. Squares cut from the membranes (5x5 mm) behaved as Acceptedmanuscrip.Publishedmanuscriplink: http://www.sciencedirect.com/science/article/pii/S0925400515306948 doi:10.1016/j.snb.2015.11.116  20  solid sensory kits from which the cations could be differentiated and their concentration estimated by the naked eye. Pictures taken to the kits permitted the titration using the colour definition of the sensory squares as analytical input. The dynamic range was of five decades (lower concentration = 1x10-7 M) for Fe(III) and Co(II) and of two decades (lower concentration = 9x10-5 M) for Cu(II) and Sn(II). The chosen of proven organic chelating agents and their chemical anchorage to polymer backbones have proven to be successful strategies to prepare sensory materials for water environments. The colour output of the sensory systems and the manageable solid kits allowed for the visually use of these materials by non-specialized personnel and, concomitantly, portable devices such as smartphones permit the fast and costless quantification of target species. Moreover, forensic applications are envisaged, e.g., the shape of metallic objects after entering into contact with hands can be colour revealed by pressing the hands on paper or cotton fabrics wetted with water solutions of the linear sensory polymer. Acknowledgments The financial support provided by the Spanish Ministerio de Economía y Competitividad-Feder (MAT2014-54137-R) and by the Consejería de Educación – Junta de Castilla y León (BU232U13) is gratefully acknowledged. Apendix A. Supplementary data A file and a video. The file containing experimental part (intermediates and monomer characterization); study of interaction of LCp, Mem1 and Mem2 with Fe(III), Co(II), Cu(II), and Sn(II); interference study; and response time. The video showing the forensic applications and metal recognition of metallic goods with solutions of LCp. Acceptedmanuscrip.Publishedmanuscriplink: http://www.sciencedirect.com/science/article/pii/S0925400515306948 doi:10.1016/j.snb.2015.11.116  21  References  [1] G. T. Morgan, F. H. Burstall, Dehydrogenation of pyridine by anhydrous ferric chloride, J. Chem. Soc. (1932) 20-30. [2] C. E. Housecroft, 4,2′:6′,4′′-Terpyridines: diverging and diverse building blocks in coordination polymers and metallomacrocycles, Dalton Trans. 43 (2014) 6594-6604. [3]. E. C. Constable, C. E. Housecroft, M. Neuburger, S. Schaffner, L. J. Scherer, Preparation and structural characterisation of terpy-cored dendrimers and dendriplexes, Dalton Trans. (2004) 26352642. [4] P. Wang, T. Okamura, H.-P. Zhou, W.-Y. Sun, Y.-P. Tian, Metal complex with terpyrindine derivative ligand as highly selective colorimetric sensor for iron(III), Chin. Chem. Lett. 24 (2013) 20-22. [5] B. Zhang, Y. Li, W. Sun, Anion-sensitive 2,4-dinitrophenylhydrazone-containing terpyridine derivative and its platinum chloride complex, Eur. J. Inorg. Chem. (2011) 4964-4969. [6] C. Bhaumik, D. Maity, S. Das, S. Baitalik, Anion sensing studies of luminescent bis-tridentate ruthenium(II) and osmium(II) complexes based on terpyridyl-imidazole ligand through different channels, Polyhedron 52 (2013) 890-899. [7] Q.-Y. Cao, M. Li, L. Zhoua, Z.-W. Wang, New 2,2′:6′,2′′-terpyridines as colorimetric and fluorescent sensors for fluoride ions, RSC Adv. 4 (2014) 4041–4046. [8] S. Bhowmik, B. N. Ghosh, V. Marjomäki, K. Rissanen, Nanomolar pyrophosphate detection in water and in a self-assembled hydrogel of a simple terpyridine-Zn2+ complex, J. Am. Chem. Soc. 136 (2014) 5543-5546. [9] A. Fermi, G. Bergamini, M. Roy, M. Gingras, P. Ceroni, Turn-on phosphorescence by metal coordination to a multivalent terpyridine ligand: a new paradigm for luminescent sensors, J. Am. Chem. Soc. 136 (2014) 6395-6400. [10] C. Goze, G. Ulrich, L. Charbonnière, M. Cesario, T. Prangé, R. Ziessel, Cation sensors based on terpyridine-functionalized boradiazaindacene, Chem. Eur. J. 9 (2003) 3748-3755. [11] Y. Zhang, C. B. Murphy, W. E. Jones Jr, Poly[p-(phenyleneethynylene)-alt-(thienyleneethynylene)] polymers with oligopyridine pendant groups: highly sensitive chemosensors for transition metal ions, Macromolecules 35 (2002) 630-636. [12] M. A. R. Meier, U. S. Schubert, Fluorescent sensing of transition metal ions based on the encapsulation of dithranol in a polymeric core shell architecture, Chem. Commun. (2005) 46104612. [13] C. Fan, X. Wang, P. Ding, J, Wang, Z, Liang, X, Tao, Synthesis, photophysical and iron-sensing properties of terpyridyl-based triphenylamine derivatives, Dyes Pigm. 95(2012) 757-767. [14] H. Li, S.-J. Zhang, C.-L. Gong, Y.-F. Li, Y. Liang, Z.-G. Qi, S. Chen, Highly sensitive and selective fluorescent chemosensor for Ni2+ based on a new poly(arylene ether) with terpyridine substituent groups, Analyst 138 (2013) 7090-7093. [15] Y. Hong, S. Chen, C. W. T. Leung, J. W. Y. Lam, J. Liu, N.-W. Tseng, R. T. K. Kwok, Y. Yu, Z. Wang, B. Z. Tang, Fluorogenic Zn(II) and chromogenic Fe(II) sensors based on terpyridinesubstituted tetraphenylethenes with aggregation-induced emission characteristics, ACS Appl. Mater. Interfaces 3 (2011) 3411-3418. [16] M. Zheng, H. Tan, Z. Xie, L. Zhang, X. Jing, Z. Sun, Fast response and high sensitivity europium metal organic framework fluorescent probe with chelating terpyridine sites for Fe3+, ACS Appl. Mater. Interfaces 5 (2013) 1078-1083. [17] M. E. Padilla-Tosta, J. M. Lloris, R. Martínez-Máñez, A. Benito, J. Soto, T. Pardo, M. A. Miranda, M. D. Marcos, Bis(terpyridyl)-Ruthenium(II) units attached to polyazacycloalkanes as sensing fluorescent receptors for transition metal ions, Eur. J. Inorg. Chem. (2000) 741-748. [18] Z.-B. Zheng, Z.-M. Duan, J.-X. Zhang, K.-Z. Wang, Chromogenic and fluorogenic sensing properties toward cations and anions by a terpyridine/phenylimidazo [4,5-f]phenanthroline hybrid, Sens. Actuators B 169(2012) 312-319. [19] C. Bhaumik, S. Das, D. Maity, S. Baitalik, A terpyridyl-imidazole (tpy-HImzPh3) based bifunctional receptor for multichannel detection of Fe2+ and F− ions, Dalton Trans. 40 (2011) 11795-11808. [20] C. Y.-S. Chung, V. W.-W. Yam, Induced self-assembly and Förster resonance energy transfer studies of alkynylplatinum(II) terpyridine complex through interaction with water-soluble poly(phenylene ethynylene sulfonate) and the proof-of-principle demonstration of this two-  Acceptedmanuscrip.Publishedmanuscriplink: http://www.sciencedirect.com/science/article/pii/S0925400515306948 doi:10.1016/j.snb.2015.11.116  22   component ensemble for selective label-free detection of human serum albumin, J. Am. Chem. Soc. 133 (2011) 18775-18784. [21] H.-L. Kwong, W.-L. Wonga, C.-S. Lee, C.-T. Yeung, P.-F. Teng, Zinc(II) complex of terpyridinecrown macrocycle: A new motif in fluorescence sensing of zwitterionic amino acids, Inorg. Chem. Commun. 12 (2009) 815-818. [22] G. N. Tew, K. A. Aamer, R. Shunmugam, Incorporation of terpyridine into the side chain of copolymers to create multi-functional materials, Polymer 46 (2005) 8440-8447. [23] R. Shunmugam, G. N. Tew, Polym, Dialing in color with rare earth metals: facile photoluminescent production of true white light, Adv. Technol. 18 (2007) 940-950. [24] Etienne Baranoff, Jean-Paul Collin, Lucia Flamigni, Jean-Pierre Sauvage, From ruthenium(II) to iridium(III): 15 years of triads based on bis-terpyridine complexes, Chem. Soc. Rev. 33 (2004) 147155. [25] E. G. Moore, M. Benaglia, G. Bergamini, P. Ceroni, Synthesis, Stability and Sensitised Lanthanide Luminescence of Heterobimetallic d/f Terpyridine Complexes, Eur. J. Inorg. Chem. (2015) 414-420. [26] S. Caramori, J. Husson, M. Beley, C. A. Bignozzi, R. Argazzi, P. C. Gros, Combination of cobalt and iron polypyridine complexes for improving the charge separation and collection in ru(terpyridine)2-sensitised solar cells, Chem. Eur. J. 16 (2010) 2611-2618. [27] A. Winter, G. R. Newkome, U. S. Schubert, Catalytic applications of terpyridines and their transition metal complexes, ChemCatChem 3 (2011) 1384-1406. [28] H. Hofmeier, U. S. Schubert, Recent developments in the supramolecular chemistry of terpyridine– metal complexes, Chem. Soc. Rev. 33 (2004) 373-399. [29] P. R. Andres, U. S. Schubert, New functional polymers and materials based on 2,2′:6′,2″-terpyridine metal complexes, Adv. Mater. 16 (2004) 1043-1068. [30] R. shanmugam, G. J. Gabriel, C. E. Smith, D. A. Aamer, G. N. Tew, Highly selective colorimetric aqueous sensor for mercury, Chem. Eur. J. 14 (2008) 3904-3907. [31] H. El Kaoutit, P. Estevez, F. C. Garcia, F. Serna, J. M. Garcia, Sub-ppm quantification of Hg(II) in aqueous media using both the naked eye and digital information from pictures of a colorimetric sensory polymer membrane taken with the digital camera of a conventional mobile phone, Anal. Methods 5 (2013) 54-58. [32] S. Vallejos, A. Muñoz, S. Ibeas, F. Serna, F. C. Garcia, J. M. Garcia, Solid sensory polymer substrates for the quantification of iron in blood, wine and water by a scalable RGB technique, J. Mater. Chem. A 1 (2013) 15435-15441. [33] P. Lainé, F. Bedioui, P. Ochsenbein, V. Marvaud, M. Bonin, E. Amouyal, A new class of functionalized terpyridyl ligands as building blocks for photosensitized supramolecular architectures. synthesis, structural, and electronic characterizations, J. Am. Chem. Soc., 124 (2002) 1364-1377. [34] F. Tessore, D. Roberto, R. Ugo, M. Pizzotti, Terpyridine Zn(II), Ru(III), and Ir(III) complexes: the relevant role of the nature of the metal ion and of the ancillary ligands on the second-order nonlinear response of terpyridines carrying electron donor or electron acceptor groups, Inorg. Chem.44 (2005) 8967-8978. [35] E. Zysman-Colman, J. D. Slinker, J. B. Parker, G. G. Malliaras, S. Bernhard, Improved turn-on times of light-emitting electrochemical cells, Chem. Mater. 20 (2008) 388-396. [36] G. A. Koohmareh, M. Sharifi, Synthesis, characterization, and coordination behavior of copoly(styrene-maleimide) functionalized with terpyridine, J. Appl. Polym. Sci. 116 (2010) 179183. [37] W. Y. Ng, X. Gong, W. K. Chan, Electronic and light-emitting properties of some polyimides based on bis(2,2‘:6‘,2‘ ‘-terpyridine) ruthenium(II) complex, Chem. Mater. 11 (1999) 1165-1170. [38] A. Earnshaw, N. Greenwood, Chemistry of the Elements, 2nd Ed., Elsevier Butterworth-Heinemann, Burlington, 2005, ch 19. [39] S. Vallejos, P. Estévez, S. Ibeas, F. C. García, F. Serna, J. M. García, An organic/inorganic hybrid membrane as a solid “turn-on” fluorescent chemosensor for coenzyme A (CoA), cysteine (Cys), and glutathione (GSH) in aqueous media, Sensors 12 (2012) 2969-2982. [40] Report on critical raw materials for the EU, Report of the Ad hoc Working Group on defining critical raw materials, European Commission, May 2014. Downloaded: July 14, 2015 (http://ec.europa.eu/enterprise/policies/raw-materials/files/docs/crm-report-on-critical-rawmaterials_en.pdf). 1  Tables Table 1. Thermal (TGA) and mechanical properties of materials. The thermal properties were evaluated in inert (N2) and oxidizing (synthetic air) atmospheres. Polymer Thermal properties Mechanical properties Atmosphere LOIc) Tensile stren g th, MPa Young’s Modulus, MPa Elongation, % N2 Air T5,a ) °C T10,a ) °C CR, b) % T5,a ) °C T10,a ) °C Mem1 242 344 4 224 318 19.1 33 660 9 Mem2 241 342 4 245 333 19.1 39 730 8 LCp 389 408 1 353 399 17.9 n.a. d) n.a. d) n.a. d) a ) 5% weight loss (T5), 10% weight loss (T10); b) CR: char yield at 800 ºC; c ) limiting oxygen index, calculated from the TGA data (LOI = 17.5 + 0.4 CR, CR at 800 ºC unde r nitrogen).d) n.a.: not applicable. 1  Captions Graphical abstract Films and water soluble polymers as sensory materials for the detection and quantification of iron, cobalt, copper and tin salts in water, and for forensic applications. Schemes Scheme 1. Synthesis of acylic monomer (5). Scheme 2. Synthesis and chemical structure of sensory linear polymer (LCp). Scheme 3. Synthesis and chemical structure of sensory membranes Mem1 and Mem2. The picture shows the membranes after removal from the mould. Figures Figure 1. UV/vis study of the interaction of Fe(III) (M) with tpy-motif (L) of LCp in water (pH = 2, buffer = HCl-KCl): a) absorbance variation spectra. Blue and red spectra correspond to formation of 1:1 (ML) and 1:2 (ML2) complexes tpy motif:Fe(III), respectively (inset: separation of processes with indication of the isosbestic points); b) absorbance at 567 nm vs Fe(III) concentration. The continuous line is the fitting of the data according with Eq. (1) (inset: Job´s plot (tpy is molar fraction of tpy sensory motifs, absorbance at 567 nm); and c) the species distribution. The concentration of sensory motifs in water was 1x10-4 M (equivalents of pendant sensory tpy motifs per litre). Figure 2. Colour development sensory materials upon entering into contact withFe(III), Co(II), Cu(II) and Sn(II): a) LCp in water solution (pH = 2, buffer = KCl-HCl, concentration of cations = 5x10-4 M). The concentration of sensory motifs in water was 0.01 M (equivalents of pendant sensory tpy motifs per litre); and b) Mem1 after immersion in water (pH = 2, buffer = KCl-HCl, concentration of cations = 5x10-4 M, immersion time = 24 h). Figure 3. Titration curves of Fe(III) and Co(II) with aqueous solutions of LCp (pH = 2, buffer = KCl-HCl, concentration of cations ranging from 6.75x10-8 to 6.57x10-4 M for Fe(III) and 7.99x10-8 to 7.54x10-4 M for Co(II). Insets: expansion of the low concentration region. The concentration of sensory motifs in water was 1x10-4 M for the Fe(III) titration and 1x10-3 M for the Co(II) titration (equivalents of pendant sensory tpy motifs per litre). Figure 4. Titration curves of: a) Cu(II); and b) Sn(II) with aqueous solutions of LCp (pH = 2, buffer = KCl-HCl, concentration of cations ranging from 1.64x10-5 to 1.46x10- 2  3 M for Cu(II) and 3.44x10-5 to 1.50x10-3 M for Sn(II) M. The concentration of sensory motifs in water was 1x10-3 M (equivalents of pendant sensory tpy motifs per litre). Figure 5. Visual titration of Fe(III), Co(II), Cu(II) and Sn(II) in water (pH = 2, buffer = KCl-HCl) with squares cut from Mem1 or Mem2. Each square was dipped in the water solution containing each cation for 24 h. The sensory membrane and the cation concentration of the water solution was: a) membranes = Mem1, Fe(III) and Co(II) concentration, from left to right: 1x10-7, 5x10-7, 1x10-6, 5x10-6, 1x10-5, 5x10-5, 1x10-4, 5x10-4, 1x10-3, 5x10-3 M; b) membranes = Mem2, Cu(II) and Sn(II) concentration, from left to right: 9x10-5, 3x10-4, 6x10-4, 9x10-4, 3x10-3, 6x10-3, 9x10-3 M; and c) Fe(III) titration curve obtained from the picture taken to the squares immersed in water with Fe(III) (PC1: principal component 1 that encompass three digital colour parameters (RGB) of each sensory square). Figure 6. Detecting iron and cobalt with a water solution of LCp (pH = 2, buffer: KClHCl, 1.14 mg/mL (concentration of sensory motifs was 1.75x10-3 M -equivalents of pendant sensory tpy motifs per litre-)): a) shape of a lag screw in a finger; b) coloration of the LCp solution caused by the iron of the lag screw; and c) shape of a widia blade (tungsten carbide + 6-10% of cobalt) of a drill bit. 9  0.0 1.5 3.0 4.5 6.0 7.5 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.00 3.50x10 -5 7.00x10 -5 1.05x10 -4 1.40x10 -4 0.0 0.1 0.2 0.3 0.4 0.5 0.6  Absorbance 567 nm [Fe(III)], 10 3 M Absorbance 567 nm [Fe(III)], M  0.0 1.5 3.0 4.5 6.0 7.5 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 0.0 3.0x10 -5 6.0x10 -5 9.0x10 -5 1.2x10 -4 1.5x10 -4 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 [Co(II)], 10 3 M  Absorbance 518 nm [Co(II)], M  Absorbance 518 nm Figure 3. Titration curves of Fe(III) and Co(II) with aqueous solutions of LCp (pH = 2, buffer = KCl-HCl, concentration of cations ranging from 6.75x10-8 to 6.57x10-4 M for Fe(III) and 7.99x10-8 to 7.54x10-4 M for Co(II). Insets: expansion of the low concentration region. The concentration of sensory motifs in water was 1x10-4 M for the Fe(III) titration and 1x10-3 M for the Co(II) titration (equivalents of pendant sensory tpy motifs per litre). 10  0.0 0.5 1.0 1.5 2.0 2.5 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09  Absorbance 450 nm [Cu (II)], 10 4 M a) 0.0 0.2 0.4 0.6 0.8 1.0 0.00 0.05 0.10 0.15 [Sn (II)], 10 3 M  Absorbance 420 nm b) Figure 4. Titration curves of: a) Cu(II); and b) Sn(II) with aqueous solutions of LCp (pH = 2, buffer = KCl-HCl, concentration of cations ranging from 1.64x10-5 to 1.46x10-3 M for Cu(II) and 3.44x10-5 to 1.50x10-3 M for Sn(II) M. The concentration of sensory motifs in water was 1x10-3 M (equivalents of pendant sensory tpy motifs per litre).  a) b) c) Fi gu KC l sol u co n co n 5x1 left titr a Fe( I (R G Fe(III) Co(II) Cu(II) Sn(II) ure 5. Vis u l -HCl) wit h u tion cont a n centration n centration, 0 -4 , 1x10 -3 , to right: 9 a tion curve I II) (PC1: G B) of each 10-7 -2 -1 0 1 2 PC1 u al titration h squares c u a ining eac h of the wat e from left t o 5x10 -3 M; 9 x10 -5 , 3x1 0 obtained f r principal c sensory sq u 10-6 1 0 [ of Fe(III), C u t from Me m h cation fo r e r solution o right: 1x 1 b ) membra n 0 -4 , 6x10 -4 , r om the pi c o mponent u are). 0 -5 10-4 Fe ( III ) ], M 11 C o(II), Cu( I m 1 or Me m r 24 h. T h was: a) m 1 0 -7 , 5x10 - 7 n es = Me m 9x10 -4 , 3 x c ture taken 1 that enc o 10-3    I I) and Sn( I m 2. Each s q h e sensor y embranes = 7 , 1x10 -6 , 5 x m 2, Cu(II) a n x 10 -3 , 6x10 to the squ a o mpass thr e 10-2 I I) in water u are was d i membran e = Mem1, F x 10 -6 , 1x10 n d Sn(II) c o - 3 , 9x10 -3 M a res immer s e e digital c (pH = 2, b u i pped in th e e and the F e(III) and -5 , 5x10 -5 , 1 o ncentratio n M ; and c) s ed in wat e c olour par a u ffer = e water cation Co(II) 1 x10 -4 , n , from Fe(III) e r with meters  Fi gu HCl, pen d of t h (tun g u re 6. Dete c 1.14 mg/ m d ant sensor y h e LCp sol u g sten carbi d c ting iron a n m L (conce n y tpy motifs u tion cause d d e + 6-10% n d cobalt w i n tration of s per litre-)) d b y the iro of cobalt) o 12 i th a water s ensory m o : a) shape o o n of the la g o f a drill bit s olution of o tifs was 1. o f a lag scr e g screw; an d . LCp (pH = 75x10 -3 M w in a fing e d c) shape o = 2, buffer: K -equivale n e r; b) color a o f a widia b K Cln ts of a tion b lade 1  SUPPORTING INFORMATION Colorimetric detection and determination of Fe(III), Co(II), Cu(II) and Sn(II) in aqueous media by acrylic polymers with pendant terpyridine motifs Miriam Trigo-López, Asunción Muñoz, Saturnino Ibeas, Felipe Serna, Félix Clemente García and José Miguel García* Departamento de Química, Facultad de Ciencias, Universidad de Burgos, Plaza de Misael Bañuelos s/n, 09001 Burgos, Spain. Fax: (+) 34 947 258 831, Tel: (+) 34 947 258 085. E-mail: jm[email protected] Table of contents S1. Experimental part. Intermediate and monomer characterisation..…………..... 2 S2. Titration of Fe(III) with LCp..……………………………............................... 6 S3. Titration of Co(II) with LCp.………...……………………………………….. 6 S4. Titration of Cu(II) with LCp.............................................................................. 7 S5. Titration of Sn(II) with LCp………………………………………………….. 7 S6. Titration of cations with pictures taken to Mem1 and Mem2………………... 8 S6.1. Titration of Fe(III) with a picture taken to Mem1……………………. 8 S6.2. Titration of Co(II) with a picture taken to Mem1.……………………. 9 S6.3. Titration of Cu(II) with a picture taken to Mem2…………………….. 10 S6.4. Titration of Sn(II) with a picture taken to Mem2…………………….. 11 S7. Response time……………………………………………………………........ 12 S8. Interference study...………………………………………………………….... 13 2  S1. Experimental part. Intermediate and monomer characterisation a) NO NO2 b) 4000 3500 3000 2500 2000 1500 1000 500 0 20 40 60 80 100 Transmittance (%) Wavenumber, cm -1  c)  8.8 8.6 8.4 8.2 8.0 7.8 7.6  ppm  d) 190 180 170 160 150 140 130 120  ppm  Figure S1. Characterisation of (3-(4-nitrophenyl)-1-(pyridine-2-yl)prop-2-en-1-one (1): a) chemical structure; b) FT-IR; c) 1H NMR; d) 13C NMR (NMR solvent: DMSO-d6). a)  b) 9.5 9.0 8.5 8.0 7.5 7.0  ppm Figure S2. Characterisation of 1-pyridylacylpyridinium iodide (2): a) chemical structure; b) 1H NMR (NMR solvent: DMSO-d6). 3  a)   b) 4000 3500 3000 2500 2000 1500 1000 500 20 30 40 50 60 70 80 90 100 Wavenumber, cm -1 Transmittance (%)  c)  8.8 8.6 8.4 8.2 8.0 7.8 7.6 7.4  ppm  d) 160 150 140 130 120  ppm  Figure S3. Characterisation of 2-(4-(4-nitrophenyl)-6-(pyridin-2-yl)pyridin-2yl)pyridine (3): a) chemical structure; b) FT-IR; c) 1H NMR; d) 13C NMR (NMR solvent: DMSO-d6). 4  a)  b) 4000 3500 3000 2500 2000 1500 1000 500 0 20 40 60 80 100 Wavenumber, cm-1 Transmittance (%)  c) 8.74 8.36 7.98 7.60 7.22 6.84 4.0 3.8  ppm  d) 160 150 140 130 120  ppm  Figure S4. Characterisation of 4-(2,6-di(pyridin-2-yl)pyridin-4-yl)benzenamine (4): a) chemical structure; b) FT-IR; c) 1H NMR; d) 13C NMR (NMR solvent: DMSO-d6).  5  a)  a) 4000 3500 3000 2500 2000 1500 1000 500 20 30 40 50 60 70 80 90 100 Wavenumber, cm -1 Transmittance (%)  b) 9.0 8.5 8.0 7.5 7.0 6.5 6.0 5.5 2.1 1.8  ppm  c) 170 160 150 140 130 120 20  ppm  Figure S5. Characterisation of N-(4-(2,4-di(pyridin-2-y)pyridine-4yl)phenyl)methacrylamide (5): a) chemical structure; b) FT-IR; c) 1H NMR; d) 13C NMR (NMR solvent: DMSO-d6). 6  S2. Titration of Fe(III) with LCp 400 450 500 550 600 650 700 750 800 0.0 0.1 0.2 0.3 0.4 0.5 0.6 [Fe(III)]  Absorbance  nm Blank [Fe(III)] = 6.57x10 - 4 M Figure S6. UV/Vis titration spectra of Fe (III) with a water solution of LCp (pH = 2, buffer = KCl-HCl, concentration of cations ranging from 6.75x10-8 to 6.57x10-4 M). The concentration of sensory motifs in water was 1x10-4 M (equivalents of pendant sensory tpy motifs per litre). S3. Titration of Co(II) with LCp 450 500 550 600 650 0.0 0.5 1.0 1.5 2.0  nm  Absorbance [Co(II)] Blank [Co(II)] = 7.54x10 - 4 M Figure S7. UV/Vis titration spectra of Co (II) with a water solution of LCp (pH = 2, buffer = KCl-HCl, concentration of cations ranging from 7x99·10-8 to 7x54·10-4 M). The concentration of sensory motifs in water was 1x10-3 M (equivalents of pendant sensory tpy motifs per litre). 13  S8. Interference study a) 400 450 500 550 600 650 700 0.0 0.5 1.0 1.5 2.0 2.5 3.0  Absorbance , nm Fe (III) + Co (II) + Sn (II) + Cu (II) Sn (II) Cu (II) Co (II) Fe (III) b) 400 450 500 550 600 650 700 750 800 0.0 0.5 1.0 1.5 2.0 2.5  Absorbance  , nm Fe(III) Na + , K + , Li + , Cu(II), Co(II), Al(III), Pb(II), Zn(II), Mg(II), Cd(II), Sn(II), Ni(II), Fe(III) Figure S15. Interference study. Water solutions of LCp (concentration of sensory motifs was 1x10-3 M (equivalents of pendant sensory tpy motifs per litre)) upon adding a: a) cocktail of cations that forms coloured complexes with tpy motifs (Fe(III), Co(II), Cu(II), Sn(II). Concentration of each cation = 1.3x10-3 M); b) cocktail of 13 cations (Na+, K+, Li+, Cu(II), Co(II), Al(III), Pb(II), Zn(II), Mg(II), Cd(II), Sn(II) and Ni(II), Fe(III),Cu(II), Co(II)). Concentration of each cation = 9.09x10-4 M). a) 504567630693 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08  , nm  Absorbance b) 0.00 2.10x10 -6 4.20x10 -6 6.30x10 -6 0.01 0.02 0.03 0.04 0.05 0.06 0.07  Absorbance [Fe (III)], M Equation y = a + b*x Adj. R-Sq 0.9828 Value Standard E B Interce 0.0107 0.00188 B Slope 7964.3 429.03007 Figure S16. Standard addition of an aqueous solution of Fe (III) to a water solution of LCp (concentration of sensory motifs was 2x10-3 M (equivalents of pendant sensory tpy motifs per litre)) containing a cocktail of 13 cations (Na+, K+, Li+, Cu(II), Co(II), Al(III), Pb(II), Zn(II), Mg(II), Cd(II), Sn(II) and Ni(II), Fe(III),Cu(II), Co (II)). The concentration of each cation was 1x10-6 M; b) least square fitting of the absorbance at 567 nm vs molar concentration of Fe(III). (Calculated concentration of Fe(III) = 1.3x10-6. Real concentration of Fe(III) and each cation = 1x10-6 M).