Fluorescence Optosensing of Inorganic Ions in Aqueous Media

Mar 3, 2011 - Instituto de Biología Molecular y Celular, University of Miguel Hernández de Elche, Avenida Universidad s/n E-03202 Alicante, Spain. §...
4 downloads 0 Views 2MB Size
Download PDF
ARTICLE pubs.acs.org/ac

Conjugated Polymer Microspheres for “Turn-Off”/“Turn-On” Fluorescence Optosensing of Inorganic Ions in Aqueous Media  lvarez-Diaz,† Alfonso Salinas-Castillo,*,‡ María Camprubí-Robles,§ Jose M. Costa-Fernandez,† Adrian A Rosario Pereiro,† Ricardo Mallavia,‡ and Alfredo Sanz-Medel† †

Department of Physical and Analytical Chemistry, University of Oviedo, Avenida Julian Claveria 8, E-33006, Oviedo, Spain Instituto de Biología Molecular y Celular, University of Miguel Hernandez de Elche, Avenida Universidad s/n E-03202 Alicante, Spain § Department of Physiology and Biomedical Physics, Medical University of Innsbruck, Fritz-Pregl Strasse 3, A-6020 Innsbruck, Austria ‡

bS Supporting Information ABSTRACT: The synthesis and characterization of a novel water-compatible microsized material, based on fluorescent conjugated polymers (CPs), and its applicability for optical sensing of inorganic ions of environment interest (copper and cyanide) in water media is here described. Polyfluorene-based fluorescent CPs were synthesized and functionalized with imidazole moieties (selective recognition element) and a terminal double bond (covalently linked to an organic matrix) through a postfunctionalization strategy. Further, microspheres of the novel imidazole-functionalized fluorescent CPs, able to work in water media, were synthesized via a microemulsion and polymerization procedure. The synthesized imidazole-functionalized CP microspheres were then evaluated as fluorescence “turn-Off” sensing materials for Cu2þ detection in aqueous media. Analyte detection was based on the quenching effect of the Cu2þ, selectively recognized by the imidazole group, on the polymer fluorescence emission. The developed optosensor exhibits a detection limit of 1 μg/L for the determination of Cu2þ in water with a reproducibility of 4%. The synthesized microsized material was also evaluated for the “turn-on” optosensing of cyanide in water, measuring the recovery of the emission signal from the CP that has been previously deactivated by the presence of quencher species. The “turn-On” optosensor allows the selective determination of free cyanide in aqueous solution with high sensitivity (detection limit of 18 μg/L), obtaining a reproducibility of 2.9%. A high sample throughput (between 7 and 12 samples per hour) was achieved in both cases. Analytical applicability of the fluorescent CP microsphere materials has been successfully demonstrated by tap and mineral water analysis.

T

hese days, there is an increasing demand for chemical sensors in many areas and disciplines. High sensitivity and ease of operation are the two main issues for sensor development. Fluorescence techniques can fulfill these requirements and, therefore, are highly valuable for chemical sensing. However, implementation of fluorescent probes in functional devices maintaining the sensitivity is still very challenging. Thus, extensive efforts are currently being done in the search of improved materials for fluorescent sensing. Polymers are still the most common support for luminescent chemical sensors. Their success relies on their simplicity to be processed to small particles and thin films that can be easily deposited onto optical fibers for sensor fabrication.1 During the last decades, chemical indicators have been immobilized in polymeric matrixes mainly by simple impregnation, doping, or covalent attachment.1,2 In addition, fluorescent polymers have been also synthesized by copolymerization with fluorescent polymerizable monomers.3-5 Alternatively, fluorescent conjugated polymers (CPs) have emerged as improved selective recognition and transduction materials for chemical sensing purposes.6-8 CPs are polyunsaturated compounds with alternating single and double bonds along the polymer chain. This conjugation between each repeated unit creates a semiconductive “molecular wire”, providing the polymers with very useful optical and optoelectronic properties. CPs, which can exhibit a strong luminescence, are extremely r 2011 American Chemical Society

sensitive to minor external structural perturbations or to electron density changes within the polymer. CPs have the ability to selfamplify their fluorescence quenching response9 due to perturbation of the electronic network upon binding of analytes (analyte binding produces a trapping site whereby the excitation is effectively deactivated by electron transfer quenching). As compared with conventional fluorophores, in the CP approach one single interaction can quench a large number of fluorophores, and the complete fluorescence quenching would be observed upon interaction with even a single analyte molecule. As a result of all these favorable properties, the development of fluorescent sensing materials based on CPs is captivating the attention of the scientists. Fluorescent CPs have been employed in different sensing applications, and particularly as sensitive probes for the detection of metal ions.10 Typically, CP chemosensors contain in their structure some acceptor groups to trap the metals ions, such as bipyridyl, terpyridyne, quinoline segments, or imidazole.11,12 Nevertheless, most reported CPs exhibit rather poor solubility in water media (a serious drawback for practical applications), probably due to their low charge densities which compete with the aromatic π-π stacking of the hydrophobic backbones. The Received: December 15, 2010 Accepted: February 7, 2011 Published: March 03, 2011 2712

dx.doi.org/10.1021/ac103268r | Anal. Chem. 2011, 83, 2712–2718

Analytical Chemistry

ARTICLE

Scheme 1. Synthesis and Representative Structure of Copolymer P1 by Postfunctionalization

incorporation of CPs into an appropriate solid support (e.g., in microbeads, layer-by-layer coated with silica or surfacegrafted)13-15 has been described as an efficient strategy to fine-tune the optical properties of the fluorophore as well as to improve the stability of the polymer.16-18 However, the physical entrapment of the CPs into a solid matrix typically suffers from serious problems of heterogeneity of the CPs in the solid material and lack of stability due to their leaching. Such described problems constitute serious handicaps for further sensing applications (poor stability and reproducibility are observed). Alternatively, covalent immobilization of the CPs into a polymeric matrix would increase its operational stability and its own life (leaching or aggregation problems would be reduced). Following this latter approach, recently we have reported a novel fluorescent CP synthesized including imidazole moieties, that was able to selectively recognize Cu2þ, and containing vinyl groups bonded to the polymer side chain, allowing further cross-linking with organic matrix supports via radical polymerization.19 The promising results obtained in such early work encouraged us to further develop new fluorescent CP microsphere materials for their implementation in an optosensing system for monitoring of trace levels of inorganic ions in natural water samples. Development of novel optosensors for monitoring of trace levels of inorganic species (e.g., Cu2þ and cyanide) in water samples is a demand of environmental interest.20-27 Actually, the World Health Organization (WHO) recommends that the concentration of Cu2þ in drinking water must not exceed 2 mg/L (31 μM) and establishes a reference value of 70 μg/L (2.7 μM) for the maximum permissible concentration of cyanide in tap water. The synthesized imidazole-functionalized conjugated polymers microspheres were here evaluated as fluorescent materials in the development of a “turn-Off” optosensor for Cu2þ and of a “turn-On” optosensor of free cyanide in water samples.

’ EXPERIMENTAL SECTION Chemicals and Materials. All chemical reagents were of analytical grade and used as received without further purification. Copper(II) sulfate pentahydrate (99-100.5%) and acetic acid (min 99.8%) were purchased from Merck (Darmstadt, Germany). Sodium cyanide g97%, ethylenediaminetetraacetic acid sodium salt dehydrated g99% (EDTA), and sodium sulfate anhydrous were purchased from Fluka (Steinheim, Germany). Potassium thiocyanate g98%, sodium phosphate tribasic dodecahydrate g98%, polyvinyl alcohol, methyl methacrylate (MMA), ethylene glycol dimethacrylate, 2,20 -azobisisobutyronitrile, and tetraethyl orthosilicate (TEOS) were purchased from Sigma-Aldrich (Milwaukee, WI, U.S.A.). Salts of the different cations studied (sodium hydroxide g99%, calcium chloride

dihydrate g99%, sodium chloride 99-102%, potassium chloride g99%, magnesium sulfate heptahydrate g99.5%) and ICP standard solutions (1000 mg/L) of iron, cobalt, zinc, manganese, copper, mercury, and lead were obtained from Merck. Sodium nitrate g99% was obtained from Panreac (Barcelona, Spain). Sodium fluoride g99%, acetylacetone g99.5%, boric acid (99.5-100.5%), and sodium acetate trihydrate (99.5-100.5%) were purchased from Merck. Freshly prepared ultrapure deionized water (Milli-Q3 RO/Milli-Q2 system, Millipore, U.K.) was used in all experiments. Preparation of Solutions. A standard solution of 2  10-3 M copper was prepared by dissolving the appropriate amount of CuSO4 3 5H2O in deionized Milli-Q water [the exact copper concentration was obtained by inductively coupled plasma mass spectrometry (ICPMS) measurements]. Then, concentrated HNO3 was added to have a final concentration of 0.2%, and the stock solution was stored at 4 °C during a maximum period of 2 weeks. Working solutions were prepared daily by dilution with Milli-Q water of an intermediate 10-5 M Cu2þ solution, and the pH was adjusted by adding 10% of 1 M sodium acetate/acetic acid buffer solution at pH 6.5. A 10-4 M EDTA solution (regenerating solution) was prepared daily by diluting a 10-2 M stock EDTA solution in deionized Milli-Q water. The pH of the working solution was adjusted to 6.5 (using a sodium acetate/acetic acid buffer), and the solution was stored at 4 °C. The 10-3 M standard solutions of sodium cyanide were prepared in a 0.1 M sodium borate/boric acid buffer solution (pH 10.5) to ensure a proper basic medium for safety reasons. Then, they were stored at 4 °C during a maximum period of 1 week. Cyanide working solutions were prepared by diluting the standard solution with the sodium borate/boric acid buffer. Synthesis and Characterization of Fluorescent CPs. The synthesis and characterization of the conjugated polymers here used were already described in an earlier communication.19 After synthesis, the structure of the CP (see Scheme 1) was confirmed by NMR and FT-IR spectroscopy. The detailed synthetic procedure and the characterization data are reported in the Supporting Information. CP Microspheres Preparation. The sensing material, able to work in aqueous media, was developed by incorporating the fluorescent CPs into an appropriate solid support by means of a covalent attachment. Thus, microspheres of the imidazole-functionalized fluorescent CPs were developed via a microemulsion and polymerization procedure. For such purpose, methacrylic acid containing polar groups was included in this structure, to favor the diffusion of water into the microspheres. Then, the copolymerization of MMA (polar monomer) with the imidazole-modified CPs was performed using ethylene glycol 2713

dx.doi.org/10.1021/ac103268r |Anal. Chem. 2011, 83, 2712–2718

Analytical Chemistry dimethacrylate as cross-linker and 2,20 -azobisisobutyronitrile as a radical initiator. The detailed synthetic procedure of the microspheres is reported in the Supporting Information. Synthesis of Sol-Gel Particles. For preparation of the sensing material, the CP microspheres were mixed with appropriate amounts of sol-gel inorganic particles. The detailed synthetic procedure of the sol-gel particles is reported in the Supporting Information. Sensing Materials. Here, we used 125-160 μm TEOS solgel particles mixed at different concentrations with CP microspheres. We performed a mixture because the microspheres have very low size and exhibit a high fluorescence. Therefore, we mixed fluorescent CP microspheres with a higher-size particulate inert material just before packing in the flow cell. With this mixture, that allowed us to avoid their compaction of the material inside the flow cell and easily prevented the detector saturation. General Procedure. The detailed description of the developed optosensing manifold optosensor is also reported in the Figure S1 (Supporting Information). After the injection of the samples or standards in the flow injection analysis (FIA) system, they passed through the fluorescence flow cell packed with the sensing material. Fluorescence emissions from the packed sensing material (λex = 378 nm, λem = 418 nm) were continuously monitored at room temperature (20 ( 3 °C). Reagent blanks were prepared and measured following the same procedure (in this case, Milli-Q water was injected instead of the inorganic ion standards). After analyte measurement, the sensing phase was regenerated by the sequential injection of 2 mL of a 10-4 M EDTA solution (for Cu2þ optosensing) or 10-5 M Cu2þ solution (in the case of the CN- optosensor) by using the second FI valve. Analysis of Real Samples. For Cu2þ determination, before injection in the optosensing system, the collected water samples were filtered through a 0.2 μm nylon cartridge, spiked with 1 mg/ L NaF and 500 μg/L acetylacetone (in order to mask the major interferences expected to be present, iron and zinc, respectively) and diluted to a final volume of 50 mL with the 0.1 M acetic acid/ sodium acetate buffer to ensure a pH 6.5. For cyanide optosensing, the real samples were first spiked with different analyte concentrations. Then, they were filtered through a 0.2 μm nylon cartridge and diluted 1:2 with the 0.1 M boric acid/sodium borate buffer to ensure a pH 10.5. Safety Cautions. Special caution was taken in the preparation and handling of all cyanide solutions and during the preparation process of the conjugated polymers. Nitrile globes, protective clothing, and safety glasses were worn all the time. Preparation of cyanide stock solutions and synthesis of the conjugated polymers were carried out under an appropriate extractor hood. Also, special care was taken to avoid the release of chemicals to the environment. Thus, hazardous wastes were collected, stored in appropriate glass containers, labeled, and sent to Wastes Management Company (COGERSA, Asturias, Spain).

’ RESULTS AND DISCUSSION Characterization of the Sensing Material. The morphology of the synthesized microspheres was assessed by scanning electron microscopy and fluorescence microscopy. As can be seen in Figure 1, the resulting microspheres have a regular spherical shape with diameters between 1 and 35 μm. On the other hand, fluorescence microscopy images demonstrated that

ARTICLE

Figure 1. (a) SEM micrographs of conjugated polymers microspheres. (b) Confocal fluorescence microscopy imaging of a single CP microsphere in water media.

the CPs were uniformly distributed inside the microspheres and they had kept their fluorescent properties after immobilization. The photostability of the sensing material was studied by continuously measuring the fluorescence emission of the CP microspheres when irradiated with a xenon discharge lamp (peak power equivalent to 75 kW) at 378 nm, both in dry media and when flowing in an aqueous carrier solution (0.1 M acetate/ acetic acid buffer at pH 6.5). There was no sign of degradation (photobleaching) when illuminating continuously the CP microspheres during more than 9 h (emission signal of the CPs was stable during that time). The long-term stability of the synthesized sensing materials was also evaluated. It has been found that the optical features and the analyte response characteristics of the polymers remained unaltered at least 1 year after synthesis, when stored at room temperature in the absence of light. Development of a “Turn-Off” Optosensor for Cu2þ Monitoring. The synthesized imidazole-functionalized CP microspheres were evaluated as fluorescence “turn-off” sensing materials for detection of trace levels of Cu2þ dissolved in aqueous media. Analyte detection was based on the quenching effect of the Cu2þ, selectively recognized by the imidazole group, on the CPs’ fluorescence emission. Confocal Microscopy Characterization. Figure 2 shows the confocal microscopy images of a fluorescent CP microsphere taken at different levels in the z-axis, before and after interaction with Cu2þ ions. As can be seen, the fluorescence intensity of the microsphere is homogeneously distributed throughout the whole material (also confirmed by the 3D image composition of the microsphere using the same z-stack images). Additionally, we 2714

dx.doi.org/10.1021/ac103268r |Anal. Chem. 2011, 83, 2712–2718

Analytical Chemistry

ARTICLE

Figure 2. Sequence of confocal microscopy images of the CP taken in the z-axis (z-stack) in presence of Cu2þ. (a) The sequence of images represents a total of seven slices with a thickness of 1 μm (right panel) taken in a small microsphere (smaller than 10 μm). (b) In presence of copper (Cu2þ) there is a significant decrease in the fluorescence signal of the microsphere. The left panel represents a 3D image of a small microsphere generated after the composition of the images taken at different levels in the polymer (z-stack, panel A).

Figure 3. Excitation and emission spectra of different mixtures of polymer microspheres with a sol-gel matrix. Fluorescence excitation (λem = 415 nm) and emission (λex = 375 nm) spectra of the mixtures of polymer microspheres with a TEOS sol-gel: (a) 3.1/30.0, (b) 1.3/30.0, and (c) 0.5/30.0 mg/mg.

found that the CP attached to the microspheres is stable in aqueous media. These features are highly valuable for their use in optosensing applications. As shown in Figure 2, the intensity of fluorescence of the CPs was greatly decreased in presence of water-dissolved Cu2þ. These results confirm that the CP maintained its capability to recognize Cu2þ after its immobilization into the microspheres. On the other hand, although the Cu2þ-evoked responses in the polymer are observed throughout the whole polymer, a stronger decrease in the signal is produced at the cortical level where this divalent ion first contacts to the microsphere. Thus, the decrease on the fluorescence intensity observed after Cu2þ interaction in all the slices acquired by confocal fluorescence microscopy measurements confirms that the copper-induced fluorescence decrease affects not only the surface of the microsphere but also the whole CP even in the more internal part of the microsphere. Sensing Material Preparation and Regeneration. Considering the low size of the microspheres (1-30 μm) and that these materials exhibit a very intense fluorescence emission, a mixture of the fluorescent CP microspheres with a higher-size particulate inert material was needed before packing in the flow cell, in order to avoid their compaction of the material inside the flow cell (that would generate overpressure in the system) and also to prevent the saturation of the detector. In our experiments, TEOS sol-gel particles (an inert material without significant luminescent emission or absorption at the wavelengths of the CPs) with diameters between 125 and 160 μm were mixed with the microspheres. Three mixtures, containing 0.5, 1.3, and 3.1 mg of fluorescent microspheres per 30.0 mg of sol-gel, were thoroughly evaluated for Cu2þ optosensing (in terms of response intensity and response times) following the general procedure. Spectra of the fluorescence of different mixtures are

Figure 4. Response profile of the optosensing system to three injections of 50 μg/L Cu2þ.

collected in Figure 3. The fluorescence emission profile from the microspheres (excitation and emission wavelengths and spectral width) was not affected by the presence of the inert sol-gel matrix. On the other hand, the mixture containing the smaller amount of polymer microspheres gave rise to lower intense and less reproducible signals. Conversely, the higher sensitivity and better reproducibility was obtained when using the material containing 3.1 mg of microspheres per 30.0 mg of sol-gel. Therefore, this sensing material was chosen for further experiments. The removal of retained Cu2þ was achieved by simply passing 2 mL of a 10-4 M solution of EDTA through the microspheres between each sample injection, proving the sensing material can be easily regenerated in about 1 min. EDTA was found to be able to form strong complexes with Cu2þ, releasing the analyte from the imidazole group of the CP, with the subsequent recovery of the fluorescence emission. As an example, Figure 4 shows the 2715

dx.doi.org/10.1021/ac103268r |Anal. Chem. 2011, 83, 2712–2718

Analytical Chemistry

ARTICLE

Figure 5. Interference assay for copper(II): 25 μg/L Cu(II) in all cases, either alone or together with 10 mg/L Ca(II), Mg(II), or Na(I), 4 mg/L K(I), and the rest of cations in the same concentration as that of Cu(II); pH 6.5 with a 0.1 M acetic acid/acetate buffer. Three replicates per point.

response profile of the optosensing system to sequential injections of samples containing 0.475 μg/L of Cu2þ. Optosensor Optimization: Effect of pH and Ionic Strength. The effect of several important parameters (typically affecting the fluorescence emission and the complexation equilibrium) on the response of the sensing material to Cu2þ was studied. First of all, the effect of pH in a range between 4.5 and 8.0 was studied in order to select the optimum conditions for the determination of copper. Different water solutions were prepared, containing 25 μg/L Cu2þ and varying the pH either by addition of a 0.1 M acetic acid/acetate buffer or a 0.1 M PBS buffer. The samples were analyzed following the general procedure. Results showed (see the Supporting Information, Figure S2) that changes in the pH affect significantly the obtained analytical signals, 6.5 being the optimal pH value. At low pHs the Cu2þ signal was greatly reduced because imidazole is occupied by protons rather than Cu2þ. Conversely, when the pH is higher than 7 the formation of copper hydroxides takes place. Natural waters may contain high concentrations of several ionic species. Therefore, the effect of the ionic strength on the response of the sensing material to Cu2þ was studied by measuring the signals obtained after injection in the optosensor system of water solutions containing a fixed 25 μg/L Cu2þ concentration and increasing NaCl concentrations up to 5000 mg/L (higher than values commonly found in tap waters). The results obtained from this study showed that variations in the ionic strength in the assayed range do not affect the response of fluorescent CP microspheres to Cu2þ (see Figure S3 in the Supporting Information). Analytical Characterization of the “Turn-Off” Cu2þ Optosensors. Calibration graphs were obtained from the fluorescent signals of triplicate 5 mL injections of aqueous standards of increasing Cu2þ concentrations following the general procedure. The quenching efficiency of the fluorescence by Cu2þ nearly fitted to the Stern-Volmer equation (I0/I = KSV[Q] þ 1), which relates the CP fluorescence intensity, I, at different concentrations of the analyte quencher, [Q] (I0 being the fluorescence intensity at [Q] = 0 and KSV being the Stern-Volmer constant). The apparent Stern-Volmer constant, determined from the Cu2þ calibration graphs, was Ksv ≈ 1.4  105 M-1. Linear dynamic range of the optosensor extends up to, at least, 75 μg/L (maximum concentration tested). The reproducibility of the proposed method, evaluated as the standard deviation of five replicates of a sample containing 25

Table 1. Copper(II) Concentrations Found in Different Mineral and Tap Water Samples with the Proposed Optosensing System, by Resorting to an External Calibration, Standard Addition Method, and by ICPMS Measurements external

standard addition

ICPMS

calibration (μg/L)

method (μg/L)

(μg/L) 166 ( 3

M1

159 ( 7

164 ( 8

M2

82 ( 4

85 ( 4

85 ( 6

M3

33 ( 3

35 ( 3

31 ( 1

μg/L of Cu2þ, was (4%. The detection limit, calculated as the concentration of Cu2þ which produced an analytical signal that is 3 times the standard deviation of the blank signal (IUPAC criterion), was 1 μg/L of copper. This detection limit was low enough for Cu2þ control in tap water samples following the criteria established by the WHO. However, for applications requiring higher sensitivity, such detection limits could be improved if the amount of sample injected in the optosensing system is increased (conversely, an increase in the volume of sample injected will result in higher analysis times). It must be indicated that the total response time of the optosensor required to perform a single measurement (time elapsed from a sample injection until a next injection could be done) is lower than 5 min. Moreover, the sensing material could be easily regenerated and used in at least up to 60 cycles of quenching and regeneration, without noticeable changes in the sensing response characteristics. Selectivity. Following the general procedure, the effect of foreign cations on the fluorescence signal was studied. The response of the CP microspheres to solutions containing 25 μg/L Cu2þ in the presence of relevant ionic species such Ca2þ, Mg2þ, Naþ, Kþ, Mn2þ, Fe2þ, Co2þ, Zn2þ, Pb2þ, and Hg2þ (at concentration levels similar to those usually found in mineral waters) was investigated. The results are collected in Figure 5. As can be seen, most of the studied species do not affect the determination of Cu2þ. Only the presence of iron, lead, and zinc could result in an error on the Cu2þ determination. Significant amounts of iron and zinc would be expected to be found together with copper in tap water samples, mainly due to the corrosion of pipes, although no interfering concentrations of lead should be found. Thus, to minimize the possible effect of such species in the determination of Cu2þ with the proposed system, the addition to 2716

dx.doi.org/10.1021/ac103268r |Anal. Chem. 2011, 83, 2712–2718

Analytical Chemistry

ARTICLE

Figure 6. Interference assay for cyanide: 0.6 mg/L CN- in all cases, either alone or together with the potential interferences, all of them in a 100 mg/L concentration; pH 10.5 with a 0.1 M boric acid/borate buffer. Three replicates per point.

the sample of sodium fluoride demonstrated to be effective for selective removal of iron, whereas the addition of acetylacetone to the sample was found to be effective to mask the interfering Zn2þ (see Figure S4 in the Supporting Information). Application to Real Sample Analysis. The proposed optosensing system was applied to the determination of Cu2þ in mineral and tap water samples. Samples were analyzed following the general procedure, and Cu2þ concentrations were obtained from a conventional calibration curve and following a standard addition methodology. Table 1 collects the results obtained. As can be observed, there is a good agreement between the results obtained from the optosensor and the results obtained when the same samples were analyzed by an alternative method (ICPMS), demonstrating the high potential of the novel CPs optosensor for the analysis of Cu2þ in real samples. Characterization of the CP Microspheres for “Turn-On” Optosensing of Cyanide. The synthesized microsized material was also evaluated for development of a “turn-on” optosensor of trace levels of free cyanide in water. Cyanide optosensing is based on the measurement of the recovery of the emission signal from the CPs that has been previously deactivated by the presence of a controlled concentration of a metal quencher.28-31 Cyanide Recognition. In this optosensor scheme, the CP microspheres are first exposed to a metal quencher, and consequently the luminescence was quenched. Then, the addition of cyanide, which forms highly stable complexes with the quencher, extracts the metal from the imidazole moieties of the CPs and their luminescence is recovered (see eqs E1 and E2 in the Supporting Information). Cyanide optosensing procedure consists on an initial injection of 10 mL of a 10-3 M Cu2þ standard solution in the carrier (Milli-Q water with 0.1 M acetic acid/ sodium acetate and pH 6.5). After interaction with the CP microspheres, their fluorescence emission is quenched, and then, water solutions (blanks or cyanide standards) were injected in the optosensing system for analysis. After measuring of the fluorescence recovery (analyte signal) a regeneration of the microsized sensing material was achieved by the injection of 2 mL of 10-5 M Cu2þ standard solution. The intensity of fluorescence of the CPs was gradually recovered after injection in the optosensor system of water samples with increased concentration of dissolved CN-. These results confirm that the proposed scheme allows the determination of trace levels of CN- in water samples.

The effect of the pH on the response of the fluorescencequenched CP microspheres to CN- was studied. Different water solutions containing a 1.3 mg/L CN- concentration and varying the pH in the range from 9 to 13 by the addition of a boric acid/ sodium borate buffer to a final concentration of 0.1 M were prepared and analyzed with the optosensor. Samples with pH values below 9.0 were not tested for safety reasons. We found that the optimal pH value was of about 10.5, ensuring the best limit of detection. Analytical Characterization of the Turn-On Optosensor for CN- Determination. The observed effect of cyanide on the luminescence emission of the quenched CP microspheres could be used to develop a method for this anion determination. A good linear relationship (r > 0.9985) was observed up to cyanide concentrations of 5.19 mg/L (maximum concentration tested). The limit of detection, calculated following the 3σ IUPAC criteria, was 18 μg/L of cyanide, with a reproducibility (evaluated from the relative standard deviation of 10 replicates of a 0.6 mg/L CN- standard solution) of 2.9%. It should be indicated that the total response time of the optosensor required to perform a single measurement (time elapsed from a sample injection until a next injection could be done) is about 7 min. Moreover, the sensing material could be easily regenerated and used in at least up to 60 cycles of quenching and regeneration, without noticeable changes in the sensing response characteristics. Selectivity. An interference assay was also carried out to observe the possible effect on the response of the optosensor to cyanide when other anions different from the analyte were present. The response of the CP microspheres to solutions containing 0.6 mg/L CN- in the presence of relevant ionic species such SO42-, NO3-, PO43-, SCN-, CH3COO-, and Cl- (at a concentration level over 10 times higher to those of the analyte) was investigated. The results are collected in Figure 6. As can be seen, no significant interference effect was found. Application to Real Samples Analysis. The proposed optosensing system was applied to the determination of free cyanide in several mineral and tap water samples. Since no cyanide was detected in those water samples, they were spiked with different concentrations of the anion prior to their injection in the optosensor system. The samples were analyzed following the general procedure, and obtained results are shown in Table 2. As can be observed, good recoveries were achieved in all the cases, 2717

dx.doi.org/10.1021/ac103268r |Anal. Chem. 2011, 83, 2712–2718

Analytical Chemistry

ARTICLE

Table 2. Recoveries for Several Cyanide-Spiked Water Samples spiked CN (mg/L)

recovery (%)

M1 (Valtorre)

2.07

106 ( 6

M2 (tap) M3 (Valtorre)

2.07 1.55-

103 ( 3 99 ( 5 104 ( 6

M4 (Fontvella)

1.55

M5 (tap)

1.55

96 ( 7

Milli-Q

1.55

95 ( 5

thus demonstrating the high potential of the novel CPs optosensor for the analysis of trace levels of cyanide in real samples.

’ CONCLUSIONS Microspheres of novel imidazole-functionalized CPs were synthesized and analytically evaluated as fluorescent materials in the development of a “turn-off” optosensor for Cu(II) and of a “turn-on” optosensor of free cyanide in water samples. Besides the high photostability and long-term stability exhibited by the CPs, the novel microsized sensing materials are able to work in water media, are fully regenerable, and allowed a rapid determination of trace levels of cyanide and copper ions in natural water samples with high sensitivity. Application of the sensing materials for the detection of the target ions in natural waters was successfully demonstrated. This research represents an example of the potential, but still unexplored, applicability of water-compatible CPs to the detection of nonluminescent analytes (e.g., anions and cations). ’ ASSOCIATED CONTENT

bS

Supporting Information. Details about the synthesis and characterization of the fluorescent conjugated polymers, the synthesis of the microspheres, the detailed procedure for synthesis of the sol-gel particles, the description of the optosensing manifold, and the standard addition protocol for Cu2þ optosensing in real samples. This material is available free of charge via the Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*Phone: þ34-965222475. Fax: þ34-966658758. E-mail: [email protected].

’ ACKNOWLEDGMENT Financial support from the Spanish Ministry of Education and Science (research projects MAT-2008-05670, CTQ2006-02309, and CTQ2010-16636) and FEDER program cofinancing funds and Fundacion Caja Murcia is gratefully acknowledged. Adrian  lvarez-Diaz acknowledges the Ph.D. scholarship from the A Asturias Government (ref BP07-061).

(3) Suri, J. T.; Cordes, D. B.; Cappuccio, F. E.; Wessling, R. A.; Singaram, B. Angew. Chem., Int. Ed. 2003, 42, 5857–5859. (4) Brasola, E.; Mancin, F.; Rampazzo, E.; Tecilla, P.; Tonellato, U. Chem. Commun. 2003, 3026–3027. (5) Meallet-Reanult, R.; Pansu, R.; Amigoni-Gerbier, S.; Larpent, C. Chem. Commun. 2004, 2344–2345. (6) Swager, T. M. Acc. Chem. Res. 1998, 31, 201–207. (7) Liu, Y.; Schanze, S. Anal. Chem. 2008, 80, 8605–8612. (8) Wang, Y.; Zhang, Y.; Liu, B. Anal. Chem. 2010, 82, 8604–8610. (9) Thomas, S. W.; Joly, G. D.; Swager, T. M. Chem. Rev. 2007, 107, 1339–1386. (10) Fan, L. J.; Zhang, Y.; Murphy, C. B.; Angell, S. E.; Parker, M. F. L.; Flynn, B. R.; Jones, W. E., Jr. Coord. Chem. Rev. 2009, 253, 410–422. (11) McQuade, D. T.; Pullen, A. E.; Swager, T. M. Chem. Rev. 2000, 100, 2537–2574. (12) Fabre, B.; Simonet, J. Coord. Chem. Rev. 1998, 178, 1211–1250. (13) Ravindranath, R.; Ajikumar, P. K.; Hanafiah, N. M.; Knoll, W.; Valiyaveettil, S. Chem. Mater. 2006, 18, 1213–1218. (14) Liao, J. H.; Swager, T. M. Lagmuir 2007, 23, 112–115. (15) Ogawa, K.; Chemburu, S.; Lopez, G. P.; Whitten, D. G.; Schanze, K. S. Langmuir 2007, 23, 4541–4548. (16) Montamedi, F.; Ihn, K. J.; Ni, Z.; Srdanov, G.; Wudl, F.; Smith, P. Polymer 1992, 33, 1102–1104. (17) Huynh, W. U.; Dittmer, J. J.; Alivisatos, A. P. Science 2002, 295, 2425–2427. (18) Skaff, H.; Sill, K.; Emrick, T. J. J. Am. Chem. Soc. 2004, 126, 11322–11325. (19) Salinas-Castillo, A.; Camprubi-Robles, M.; Mallavia, R. Chem. Commun. 2010, 46, 1263–1265. (20) Dash, R. R.; Gaur, A.; Balomajumder, C. J. Hazard. Mater. 2009, 163, 1–11. (21) Wang, J.; Ha, C. S. Tetrahedron 2010, 66, 1846–1851. (22) Zheng, J.; Xiao, C.; Fei, Q.; Li, M.; Wang, B.; Feng, G.; Yu, H.; Huan, Y.; Song, Z. Nanotechnology 2010, 21, 45501–45505. (23) Guliyev, R.; Buyukcakin, O.; Sozmen, F.; Bozdemir, O. A. Tetrahedron Lett. 2009, 50, 5139–5141. (24) Jin, W. J.; Fernandez-Arguelles, M. T.; Costa-Fernandez, J. M.; Pereiro, R.; Sanz-Medel, A. Chem. Commun. 2005, 883–885. (25) Fernandez-Arguelles, M. T.; Jin, W. J.; Costa-Fernandez, J. M.; Pereiro, R.; Sanz-Medel, A. Anal. Chim. Acta 2005, 549, 20–25. (26) Lin, W.; Long, L.; Chen, B.; Tan, W.; Gao, W. Chem. Commun. 2010, 46, 1311–1313. (27) Kumar, R.; Bhalla, V.; Kumar, M. Tetrahedron 2008, 35, 8095–8098. (28) Song, T.; Xu, J.; Gheng, G. Inorganic Chemistry; Higher Education Press: Beijing, China, 2004. (29) Li, Z.; Lou, X.; Yu, H.; Li, Z.; Qin, J. Macromolecules 2008, 41, 7433–7439. (30) Zeng, Q.; Cai, P.; Li, Z.; Qin, J.; Tang, B. Z. Chem. Commun. 2008, 1094–1096. (31) Yang, R.; Wu, W.; Wang, W.; Li, Z.; Qin., J. Macromol. Chem. Phys. 2010, 211, 18–26.

’ REFERENCES (1) Basabe-Desmont, L.; Reinhoudt, D. N.; Crego-Calama, M. Chem. Soc. Rev. 2007, 36, 993–1017. (2) Bosch, P.; Catalina, F.; Corrales, T.; Peinado, C. Chem.—Eur. J. 2005, 11, 4314–4325. 2718

dx.doi.org/10.1021/ac103268r |Anal. Chem. 2011, 83, 2712–2718