Functionalized monolayers on mesoporous silica and on titania nanoparticles for mercuric sensing

Abstract

Heterogeneous “naked-eye” colorimetric and spectrophotometric cation sensors were prepared by immobilization of an azobenzene-coupled receptor onto mesoporous silica (AR-SiO₂) or titania nanoparticles (AR-TiO₂) via sol–gel or hydrolysis reactions. The optical sensing ability of AR-SiO₂ was studied by addition of metal ions such as K⁺, Ca²⁺, Sr²⁺, Co²⁺, Cd²⁺, Pb²⁺, Zn²⁺, Fe³⁺, Cu²⁺ and Hg²⁺ ions (all as chlorides) in aqueous solution. Upon the addition of Hg²⁺ ion in suspension, the AR-SiO₂ resulted in a color change from yellow to deep red. No significant color changes were observed in the parallel experiments with K⁺, Ca²⁺, Sr²⁺, Co²⁺, Cd²⁺, Pb²⁺, Zn²⁺, Fe³⁺ or Cu²⁺ ion. These findings confirm that the AR-SiO₂ can be useful as chemosensors for selective detection of Hg²⁺ ion over a range of metal ions in aqueous solution. Also, the color change of AR-SiO₂ was independent of the presence of anions NO₃⁻, ClO₄⁻, Br⁻ and I⁻. We also prepared a portable chemosensor kit by coating a 4 μm thick film of AR-TiO₂ onto a glass substrate. We found that this AR-TiO₂ film detects Hg²⁺ ion at pH 7.4 with a sensitivity of 28 nM. Finally, we tested the effect of pH on AR-TiO₂ with Hg²⁺ ion between pH 1.0 to 11.0. The absorbance and color changes of AR-TiO₂ were almost constant between pH 4 and 11. The results imply that the AR-TiO₂ film is applicable as a portable colorimetric sensor for the detection of Hg²⁺ ion in the environmental field.

---

Introduction

Chemical sensors are molecular receptors that transform into analytically useful signals upon binding to specific guests. Such sensors have attracted interest in fields such as waste management, environmental chemistry, clinical toxicology, and bioremediation of radionuclides due to their potential for easy detection, and quantification of pollutant species.^1–9 Among these, the selective detection of metal ions such as mercury and lead is critical for environmental monitoring, as these are highly toxic and common pollutants.

Various approaches for mercury screening, including atomic absorption/emission spectroscopy and inductively coupled plasma mass spectrometry, have been investigated.^10–12 Although these instrumental analyses are currently used in applications relevant to the detection of metal toxins, there is still a need to develop inexpensive and simple methods for the detection of metal toxins. Among the various sensor approaches, optical sensors that allow on-site, real-time, qualitative, or semiquantitative detection without the use of bulky or complex spectroscopic instrumentation have received a great deal of attention as promising methods for the determination of pollutant species in environmental analysis.^13–16 However, most of the sensors developed so far are kinetically slow with a limited sensitivity for detection below the permissible level of metal toxins, indicating a lack of control of the remote sensing of pollutant species.

Very recently, organic-inorganic hybrid materials have been investigated as new materials for ion recognition and sensing. Receptors immobilized on inorganic nanomaterials such as SiO₂, Al₂O₃ and TiO₂ constitute solid chemosensors and, as such, have important advantages when used in the heterogeneous solid-liquid phase.^17–28 Firstly, receptors immobilized on an inorganic support can remove the guest molecules (toxic metal ions or anions) from the pollutant solution. Secondly, the inorganic oxides can be fabricated as functionalized porous nanomaterials. In particular, nanocrystalline TiO₂ films are potentially useful materials for optical sensors due to their high surface area and excellent optical transparency in the visible region of the spectrum (λ > 400 nm).

Mesoporous silica and titania nanoparticles (SBA-15 and MCM-41) are promising as inorganic support materials due to their homogeneous porosity and large surface area. Only limited examples of such heterogeneous sensors have been reported,^17–28 however. Such sensors take advantage of the independent solubility of the receptor in water and in organic solvents. Very recently, for example, Martínez-Máñez et al. reported on a fabrication of the squaraine-based receptor immobilized on a mesoporous 3D hybrid material for Hg²⁺ ion detection and adsorption.²⁹ Addition of the colorless hybrid material to a solution containing Hg²⁺ ion resulted in a rapid and dramatic change to deep blue. The solution also showed a remarkable fluorescence. However, the hybrid material could not be renewed after adsorption of Hg²⁺ ion because it acts as the chemodosimeter for these ions. In addition, this hybrid material was selectively detectable for Hg²⁺ in a mixed solvent of acetonitrile and water due to the solubility of the squaraine receptor at high Hg²⁺ concentration.

Furthermore, Willner et al. reported that oligonucleotides immobilized on gold nanoparticles enabled the colorimetric detection of Hg²⁺.³⁰ The gold nanoparticles were highly aggregated in the presence of Hg²⁺ ion, which changed from red to blue. Also, the gold nanoparticles were not renewable after adsorption of Hg²⁺ ion.

Based on this idea, in this work we explore a new approach to the development of nanomaterial chemosensors. We synthesized the azobenzene-coupled acyclic receptor 6 containing a triethoxylsilane moiety (Scheme 1) and immobilized 6 onto the surface of mesoporous silica by sol–gel reaction. We also attached 4 containing a carboxylic acid to titania nanoparticles of 50 nm diameter in order to prepare a portable chromogenic sensor. The receptors 4 and 6 were immobilized by different methods on the two different supports. To the surface of mesoporous silica, the receptors were attached by covalent bonds via a sol–gel grafting method. To the TiO₂ nanoparticles, the receptors were attached via hydrogen bonds. After immobilizing the receptors, we transformed the ester of receptor 4 or 6 into an acid-type receptor attached onto mesoporous silica and titania by hydrolysis. We herein describe the fabrication of an acyclic acid-type receptor immobilized on mesoporous silica (AR-SiO₂) and the same receptor on titanium oxide nanoparticles (AR-TiO₂), both of which selectively change color upon the addition of Hg²⁺.

Scheme 1 Synthetic method. Reaction conditions: (i) BrCH₂CO₂Et, 60 °C; (ii) diazonium salt, DMF, 0 °C; (iii) TiO₂ nanoparticles, reflux; DMF = N,N′-dimethylformamide; (iv) HCl, methanol; (v) 3-Aminopropyltriethoxysilane, DCC, DMAP, EA, RT; (vi) mesoporous silica, toluene, reflux; (vii) HCl, methanol; (viii) NaOH.

Experimental

Characterization

¹H and ¹³C NMR spectra were recorded on a Bruker 300 apparatus. Infrared spectra of silica powder pellets were obtained using a Shimadzu FT-IR 8400S and the mass spectrum was obtained with a JEOL JMS-700 mass spectrometer. Powder X-ray diffraction patterns were recorded on a Siemens D8 Advance diffractometer using Cu-Kα radiation and a secondary monochromator. Transmission electron microscopy (TEM) images were captured with a JEOL JEM – 2100 F microscope. Nitrogen-adsorption isotherms were measured at 78 K on a Micromeritics ASAP 2010 analyzer. Thermal gravimetry analysis was conducted on a TA SDT Q600 with a heating rate of 10 °C min⁻¹ using a Pt pan in air. The temperature was scanned from 25 °C to 900 °C.

Preparation of mesoporous silica

Mesoporous silica was synthesized starting from the preparation of a hydrochloric acid solution of P-123 [(poly(ethylene oxide)-poly(propylene oxide)-poly(ethylene oxide) triblock copolymer)]. Tetraethyl orthosilicate (TEOS) was then added and the mixture was stirred at 40 °C for 20 h. The molar composition was 1 : 5.9 : 193 : 0.017 TEOS : HCl : H₂O : P-123. The solid was aged at 65 °C for 1 day and then was filtered, washed and dried at 90 °C. To cleave the template to generate mesopores, 1.0 g of as-synthesized SBA-15 was mixed with 100 mL of 60 wt % H₂SO₄ solution and refluxed at 95 °C for 1 day. The product was recovered by washing with water and dried at 90 °C. To generate mesopores, the acid-treated sample was heated to 200 °C in air. To remove cationic surfactants from the resulting dried fiber-like flocculates and particles, the sample was calcined in a box furnace in air at 500 °C for 5 h, with a ramp rate of 1 °C min⁻¹.

Immobilization of receptor 6 onto mesoporous silica

Compound 6 (100 mg) was dissolved in toluene (10 mL). The mesoporous silica (100 mg) was added as a solid. The suspension of silica was stirred in under reflux conditions for 24 h in toluene. Then, the collected solid was washed copiously with toluene (50 mL) to rinse away any surplus 5 and dried under vacuum.

Hydrolysis of mesoporous silica immobilized with 6

To obtain the acid from the receptor (7) attached to mesoporous silica, the mesoporous silica immobilized with receptor 6 was first dispersed in methanol (2 mL). The suspension of silica was maintained for 10 h at room temperature. After filtration, the solid product was washed with Na₂CO₃ solution and water and dried.

Preparation of TiO₂ nanoparticles

The TiO₂ nanoparticles were obtained by a method reported previously.²⁵

Preparation of receptor 4 immobilized TiO₂ films

TiO₂ paste was prepared from a sol–gel colloidal suspension containing 12.5 wt% TiO₂ and 6.2 wt% Carbowax 20000. The nanocrystalline TiO₂ particles were synthesized by the following procedure: 20 mL of titanium iso-propoxide (Aldrich Co.) were injected into 5.5 g of glacial acetic acid under argon atmosphere and stirred for 10 min. The mixture was transferred into a flask containing 120 mL of 0.1 M nitric acid at room temperature and stirred vigorously. The flask was left uncovered and heated at 80 °C for 8 h. After cooling, the solution was filtered using a 0.45 mm syringe filter, diluted to 5 wt% TiO₂ by adding H₂O and then autoclaved at 220 °C for 12 h. The colloids were re-dispersed with a 60 s cycle burst from an LDU Soniprobe. Finally, the solution was concentrated to 12.5 wt% TiO₂ on a rotary evaporator using a membrane vacuum pump at a temperature of 40 °C. Carbowax 20.000 was added and the resulting paste was stirred slowly overnight to ensure homogeneity. For the film TiO₂ preparation, 3 mL of the solution detailed above were spread on glass slides with a glass rod and using adhesive tape as spacers. After the films were dried in air, they were sintered at 450 °C for 20 min. All samples used in this work were films of 4 μm thickness. Functionalization of the TiO₂ films was achieved by immersing the films in a 1 mM solution of receptor 4 in 1 : 1 acetonitrile/tert-butyl alcohol mixture overnight, followed by rinsing in ethanol to remove unadsorbed receptor 4.

Coloration and extraction of metal ions by AR-SiO₂ particles and AR-TiO₂ films

The AR-SiO₂ (10 mg) was added to metal ion solutions (5.0 equiv with respect to 1 anchored to the AR-SiO₂). The mixture was stirred for 10 min. After filtration, the concentration of metal ions was analyzed by ICP (DX-500, DIONEX). The UV-vis spectra of metal-loaded AR-SiO₂ were measured by a Shimadzu UV-2401 PC UV-Vis spectrophotometer. In addition, the reversibility of metal adsorption was measured. EDTA (10.0 μM, 2 mL) was added to the AR-SiO₂ suspension (at pH 7.4) to remove Hg²⁺ ion bound to AR-SiO₂. Then, the UV-vis absorption change of the H₂O suspension of AR-SiO₂ was measured. In addition, the AR-TiO₂ films were dipped in cation aqueous solutions of interest in buffered condition, and then dried in air, with changes in coloration noted by visual inspection as well as by measurement of UV-vis absorption.

Compound 3

To a 50 mL round bottom flask, o-anisidine (2) was added to ethylbromoacetate in potassium dihydrogen phosphate in CH₃CN (10 mL). The reaction mixture was maintained for 10 h at 60 °C. After cooling to room temperature, the reaction mixture was reduced, then was dissolved into CHCl₃ (50 mL) and washed with water (2 × 30 mL). The organic phase was dried over MgSO₄ and reduced to oil. The crude product was purified on a silica column using 70 : 30 Hex : EtOAc to produce 3 as a white solid. Properties of 3 are as follows: m.p. 78–80 °C; ¹H NMR (300 MHz, CDCl₃, TMS) δ = 6.87 (d, ²J(H,H) = 12.3 Hz, 1H; Ar–H), 4.20 (d, ²J (H,H) = 7.2 Hz, 1H; Ar–H), 3.80 (s, 3H; CH₃) 1.26 (t, ³J(H,H) = 7.2 Hz, 6H; CH₃); ¹³C NMR (300 MHz, CDCl₃, TMS) δ = 171, 151, 122, 120, 119, 111, 77, 77, 77, 76, 60, 55, 53, 30, 14; IR (KBr): ν = 3393, 3090, 2931, 2853, 1740, 1653, 1373, 1032, 751 cm⁻¹; High resolution mass spectrum (HRMS) (FAB⁺) m/z 295.14 [(M + H)⁺ calcd for C₁₅H₂₁NO₅:295.16]. Anal. Cald for C₁₅H₂₂NO₅: C, 61.00; H, 7.17; N, 4.74. found: C, 60.68; H, 7.23; N, 4.69.

Compound 4

To a 50 mL round bottom flask, o-amino benzoic acid (1.047 g, 7.5 mmol) was added which contained a solution of 1 : 1, THF : H₂O (20 mL) and NaNO₂ (0.5261 g, 7.6 mmol). While this mixture was being stirred at 0 °C, HCl (1.0 mL, 12 M) was slowly added. The resulting solution was added dropwise to a 100 mL single neck round bottom flask containing the appropriate ester (3) (6.8 mmol) in a solution of 1 : 1 THF : H₂O (40 mL). This solution was held at 0 °C for approximately 2 h and then was left stirring overnight at room temperature. The resulting dark red solution was reduced down, dissolved into CHCl₃ (60 mL) and washed with water (2 × 30 mL). The organic phase was dried over MgSO₄ and reduced to oil. The crude product was purified on a silica column using 70 : 30 Hex : EtOAc, to produce 4 as a bright red solid. Properties of 4 are as follows: m.p. 101–103 °C; ¹H NMR (300 MHz, CDCl₃, TMS) δ = 8.10 (d, ²J(H,H) = 7.75 Hz, 1H; Ar–H), 7.59 (d, ²J(H,H) = 7.2 Hz, 1H; Ar–H), 7.48 (t, ³J(H,H) = 7.8 Hz, 1H; Ar–H), 6.9 (d, ²J(H,H) = 4.5 Hz, 1H; Ar–H), 4.18 (m, 8H), 3.8 (s, 3H; CH₃), 1.26 (t, ³J(H,H) = 7.2 Hz, 6H; CH₃); ¹³C NMR (300 MHz, CDCl₃, TMS) δ = 187, 182, 154, 144, 135, 134, 134, 133, 133, 127, 127, 125, 120, 116, 58, 43, 42, 23, 18,; IR (KBr): ν = 3414, 3270, 2940, 2856, 1870, 1735, 1637, 968, 804 cm⁻¹; HRMS (FAB⁺) m/z 443.17 [(M + H)⁺ calcd for C₂₂H₂₅N₃O₇: 443.20]. Anal. Cald for C₂₂H₂₅N₃O₇: C, 59.59; H, 5.68; N, 9.48. found: C, 60.11; H, 5.80; N, 9.35.

Compound 6

To a 20 mL round bottom flask, 4 (250 mg, 0.42 mmol) was added to aminopropyltriethoxysilane (1.0 g, 5.58 mmol) in ethyl acetate (10 mL). The mixture was stirred under nitrogen for two minutes and then cooled to 0 °C. 1-Hydroxybenzotriazole (HOBt, 68 mg, 0.5 mmol) and dicyclohexyl carbodiimide (DCC, 1.870 g, 9.07 mmol) were added simultaneously as a mixture of solids. The reaction mixture was allowed to warm to room temperature and stirred for 24 h. The precipitate was removed by filtration. The filtrate was then washed with aq. NaHCO₃ (satd.) and aq. NaHSO₄ (satd.). The solution was dried with MgSO₄ and the solvent was removed by rotary evaporation to produce a reddish crude. This crude product was purified by column chromatography (silica-gel, n-hexane: ethyl acetate, 1 : 1, R_f 0.45) to give the product (1.24 g, 65%). Properties of 6 are as follows: m.p. 115–117 °C; ¹H NMR (300 MHz, DMSO) δ = 7.5, 7.5, 7.4 (m, 3H) 6.9, 6.8, 6.8, 6.7 (m, 4H), 4.17 (t, ³J(H,H) = 6.9 Hz, 1H; Ar–H), 3.79 (s, 3H), 1.73 (t, ³J(H,H) = 2.4 Hz, 1H; Ar–H) 1.24 (m, 15H); IR (KBr): ν = 3414, 3270, 2940, 2856, 1870, 1735, 1637, 1121, 968, 804 cm⁻¹; HRMS (FAB⁺) m/z 646.30 [(M + H)⁺ calcd for C₃₁H₄₆N₄O₉Si: 646.34]. Anal. Cald for C₃₁H₄₆N₄O₉Si: C, 57.56; H, 7.17; N, 8.66. found: C, 57.71; H, 7.02; N, 8.33.

Compounds 5 and 7

The synthetic methods will be described in detail later.

Results and discussion

Synthesis of acyclic azobenzene-based receptor

The synthesis of the target compound 1 is difficult due to the polymerization of the triethoxyl group during hydrolysis, which removes the ethyl groups of 4 by pathway A in Scheme 1. Thus, we tried to obtain 1 by hydrolysis of 4 in acidic conditions after immobilization of receptor 6 onto the surface of mesoporous silica by pathway B. To prepare a selective chromogenic receptor for Hg²⁺, we began with 2, then formed o-anisidine (3) by alkylation using two equivalents of ethylbromoacetate in the presence of potassium dihydrogen phosphate and potassium iodide in CH₃CN. The azobenzene ester 4 was made by the reaction of the diazonium salt of o-aminobenzoic acid with 3. Subsequently, treatment with 4 and aminopropyltriethoxysilane as precursor for the sol–gel reaction afforded compound 6 as a yellow powder. Then, immobilization of the chromogenic receptor 6 to mesoporous silica was conducted under reflux for 24 h in toluene. In this process, the triethoxylsilyl group of 6 attached onto mesoporous silica undergoes hydrolysis and attaches covalently to the surface of the mesoporous silica.¹⁵ After cooling to room temperature, the yellow solid product was filtered, washed with THF, and dried. Finally, a chemosensing material was obtained as a light yellow solid by hydrolysis of 7 attached onto mesoporous silica under acidic conditions. The acyclic receptor 1 immobilized mesoporous silica (AR-SiO₂) was characterized by transmission electron microscopy (TEM), UV-vis spectroscopy, thermogravimetric analysis (TGA) and FT-IR.

Preparation of AR-SiO₂

In Fig. 1 and Fig. S1 (ESI),† TEM images of AR-SiO₂ and the mesoporous silica clearly show the formation of a well-ordered hexagonal arrangement of mesoporous channels before and after attaching the receptor 1. To investigate the porosity changes of the mesoporous silica induced by the introduction of the azobenzene-coupled derivative 1, we measured the surface area, pore volumes and pore diameters of both mesoporous silica and AR-SiO₂ using nitrogen adsorption-desorption isotherms (Fig. 1B). The mesoporous silica has a BET (Brunauer-Emmett-Teller) surface area of 1119.72 m² g⁻¹ and a pore volume of 0.49 cm³ g⁻¹. On the other hand, we observed that the AR-SiO₂ has a BET surface area of 673.21 m² g⁻¹ and a pore volume of 0.26 cm³ g⁻¹. The mesoporous silica and the AR-SiO₂ have BJH (Barrett-Joyner-Halenda) pore diameters of 2.24 and 2.15 nm, respectively (Fig. S2, ESI).† The decreased surface area and pore diameter in AR-SiO₂ are attributable to the attachment of the acyclic azo-coupled derivative to the mesoporous silica. Furthermore, from the results of the TGA measurement (Fig. S3, ESI),† we determined that the AR-SiO₂ consists of only 18.0 wt% of 1.

Fig. 1 (A) TEM image of AR-SiO₂. (B) The nitrogen adsorption-desorption isotherms of (a) mesoporous silica and (b) AR-SiO₂.

For further structural proof of the AR-SiO₂, we carried out IR spectroscopy of both mesoporous silica and AR-SiO₂. For the mesoporous silica, IR peaks appear at 3450, 1658 and 1084 cm⁻¹. For AR-SiO₂ (Fig. S4, ESI),† peaks appear at 3382, 2976, 2933, 2884, 1626, 1615, 1570, 1471, 1446, 1428 and 1382 cm⁻¹. These new peaks originate from the acyclic azo-coupled receptor 1, providing solid evidence that 1 is indeed attached to the surface of the mesoporous silica.

Preparation of the azobenzene-based receptor 4 immobilized TiO₂ film (AR-TiO₂)

In consideration of extending its usefulness, immobilization of the chromogenic receptor 4 into the TiO₂ nanoparticles was conducted under reflux conditions for 24 h in toluene. In this process, the receptor 4 undergoes hydrolysis and attaches covalently to the surface of TiO₂.^9,13 After cooling to room temperature, the red solid product was filtered, washed with ethanol, and then dried. For the structural proof of the TiO₂, we took IR spectroscopy of both TiO₂ and AR-TiO₂. For TiO₂, IR peaks appear at 3391, 1639 and 655 cm⁻¹ whereas for AR-TiO₂ appeared at 3421, 2977, 2926, 2874, 1728, 1633, 1594, 1513, 1439, 1371, 1284, 1131, 1069 and 655 cm⁻¹ (Fig. S5, ESI)† where new peaks are originated from azobenzene-coupled receptor 4, giving us a solid evidence that 4 is certainly attached onto the surface of the TiO₂ nanoparticles. In addition, TEM images of TiO₂ nanoparticles and AR-TiO₂ were observed (Fig. 2). Both TiO₂ nanoparticles before and after immobilization of the receptor 4 showed 30∼60 nm of diameters without any morphological change.

Fig. 2 TEM image of (A) TiO₂ and (B) AR-TiO₂ nanoparticles.

To prepare AR-TiO₂ films onto glass slides, 3 mL of the solution detailed above were spread on the glass slides with a glass rod and using adhesive tape as spacers. After the films were dried in air, they were sintered at 450 °C for 20 min. Sensitization of the TiO₂ films was achieved by immersing the films in a 1.0 mM solution of receptor 4 in 1 : 1 acetonitrile/tert-butyl alcohol overnight, followed by rinsing in ethanol to remove unadsorbed receptor 4. Finally, the slight yellow color of the AR-TiO₂ film was obtained.

Coloration of AR-SiO₂ with metal ions

We investigated the spectroscopic properties of the AR-SiO₂ towards the metal ions K⁺, Ca²⁺, Sr²⁺, Co²⁺, Cd²⁺, Pb²⁺, Zn²⁺, Fe³⁺, Cu²⁺ and Hg²⁺ in aqueous solution (pH = 7.4, 5.0 equiv with respect to 1 anchored to the mesoporous silica). Interestingly, upon the addition of Hg²⁺ in H₂O suspension, AR-SiO₂ resulted in a color change from light yellow to red within 10 s (Fig. 3A(b) and 3B(b)). In this category of chromophore, photoexitation causes a net electronic charge transfer from the donor end (bridgehead nitrogen) to the acceptor end within the chromophore. Thus, the observed color change is ascribed to the formation of a complex with strong coordination bonding between the bridgehead nitrogen atom of aniline group of 1 and Hg²⁺.^7,23 Except for Hg²⁺, no significant color changes were observed in the experiments using other metal ions (Fig. 3A(c–i) and Fig. S6, ESI).† These findings confirm that AR-SiO₂ can be useful as a colorimetric sensing material for the selective detection of Hg²⁺ in the presence of other metal ions. This is a rare example of chromogenic sensing for a specific metal ion by functional inorganic nanomaterials. The association constant (K) for the Hg²⁺ coordination to AR-SiO₂ was calculated to be 3.05 × 10⁵ M⁻¹ (log K = 5.48).²⁸

Fig. 3 The colorimetric response of (A) dried and (B) H₂O suspension samples of AR-SiO₂ (5.0 mg) in the (a) absence and the presence of (b) HgCl₂ (5.0 equiv), (c) CoCl₂ (5.0 equiv), (d) CdCl₂ (5.0 equiv), (e) PbCl₂ (5.0 equiv), (f) ZnCl₂ (5.0 equiv), (g) FeCl₃ (5.0 equiv) and (h) CuCl₂ (5.0 equiv) at pH = 7.4.

The UV-vis spectrum of Hg²⁺-loaded AR-SiO₂ is unchanged in the presence of an excess amount of K⁺, Ca²⁺, Sr²⁺, Co²⁺, Cd²⁺, Pb²⁺, Fe³⁺, Cu²⁺ and Zn²⁺ (Fig. S7, ESI),† indicating that AR-SiO₂ shows great promise as a selective chemosensor for the detection of Hg²⁺.

To develop the AR-SiO₂ as a general mercury cation sensor which is independent of the anion present, we investigated the anion effect by addition of other anion such as NO₃⁻, ClO₄⁻ and Br⁻ (Fig. S8, ESI).† As observed for the HgCl₂ solution, the color of the AR-SiO₂ also changed from yellow to red. This finding indicates that the AR-SiO₂ can be employed for the detection of Hg²⁺ independently of the other anion(s) present.

The Hg²⁺-loaded AR-SiO₂ was isolated to confirm the binding efficiency of Hg²⁺ by AR-SiO₂. Then, the solid UV-vis spectrum for the Hg²⁺-loaded AR-SiO₂ was compared with that for AR-SiO₂ alone (Fig. 4A). The Hg²⁺-loaded AR-SiO₂ spectrum exhibits an absorption maximum at 495 nm, whereas the absorption peak appears at 310 nm for the Hg²⁺-free one, indicating that the Hg²⁺ is efficiently bound to 1 attached in the AR-SiO₂ by covalent bonds.

Fig. 4 (A) Solid UV-vis spectra of AR-SiO₂ (5.0 mg) in the (a) absence and (b) the presence of HgCl₂ (5.0 equiv) and (c) after addition of EDTA (10.0 μM, 2 mL). (B) Picture and (C) proposed structure of AR-SiO₂·Hg²⁺ before and after treatment of EDTA (10.0 μM).

In addition, we confirmed the reversibility of the color change of AR-SiO₂ by removing the Hg²⁺ ion bound to AR-SiO₂ by treatment with EDTA. As expected, the red color of AR-SiO₂ in the presence of Hg²⁺ ion was changed into light yellow upon the EDTA (10.0 μM) treatment (Fig. 4B). Once again, the color change was fully reversible with the addition of EDTA (10.0 μM). Because the Hg²⁺ ion bound to AR-SiO₂ is dissociated by EDTA, the AR-SiO₂ can be repeatedly used by renewing with EDTA (Fig. 4C). Clearly, AR-SiO₂ exhibits excellent reusability, as almost no loss in AR-SiO₂ sensitivity was observed after eight repeated dipping/rinsing cycles.

In order to understand the coordination behavior between the receptor 1 attached to AR-SiO₂ and the Hg²⁺ ion, we made repeated attempts to obtain the crystal structure of the complex 1 with the Hg²⁺ ion, but were not successful. As an alternative, we measured the UV-vis spectra of AR-SiO₂ with the addition of Hg²⁺ ion to confirm the stoichiometry between 1 attached onto AR-SiO₂ and the Hg²⁺ ion. A spectral variation of AR-SiO₂ in H₂O was observed upon the gradual addition of HgCl₂. As a function of the Hg²⁺ concentration, a new absorption band centered at 495 nm leading to an obvious color change from light yellow to red was observed. The red-shift from 310 to 495 nm of the absorption of AR-SiO₂ is attributed to a strong binding affinity between the nitrogen atom of 1 attached to AR-SiO₂ and the Hg²⁺ ion (Fig. S9, ESI†).⁷ The stoichiometry for the 1 complex with Hg²⁺ was examined by a Job plot. As shown in Fig. S8,† it was found that the 1-Hg²⁺ complex concentration approaches a maximum when the molar faction of [1]/[1] + [Hg²⁺] is about 0.5, indicating that it forms a 1 : 1 complex of 1 attached to AR-SiO₂ with Hg²⁺ ion, as show in Fig. S10, ESI.†

Fig. S11 (ESI)† shows the standard calibration data (Abs. vs. [Hg²⁺]) for AR-SiO₂. A linear response is observed (between 1.0 μM and 10 μM) with a sensitivity of ∼1.0 μM. This sensitivity is equivalent to those previously reported for spectrophotometric sensors anchored to mesoporous aluminosilicates.²⁴

The above results encouraged us to test the separation of Hg²⁺ from the waste solution. The test sample was prepared by adding AR-SiO₂ (10 mg) to 1 mL of waste containing 2.0 μM Hg²⁺. The Hg²⁺-loaded AR-SiO₂ was then removed by filtration from the waste. To determine the amount of Hg²⁺ separated by the AR-SiO₂, the amount of Hg²⁺ left in the waste solution was determined by inductively coupled plasma mass spectrometry (ICP-MS). The ICP-MS measurements indicated that only 7% of the original Hg²⁺ remained in the waste solution, suggesting that the AR-SiO₂ removed 93% of the Hg²⁺. We also analyzed the Hg²⁺-loaded AR-SiO₂ sample by UV-vis spectroscopy (Fig. S12, ESI).† The absorbance of the Hg²⁺-loaded AR-SiO₂ was 0.26, coinciding with the calibration curve (Fig. S11, ESI).†

Furthermore, we examined the pH effect on the Hg(II) uptake by measuring the absorbance of AR-SiO₂ treated with Hg(II) solutions (pH 1.0–12.0). As shown in Fig. S13 (ESI),† no significant absorbance or color changes were observed between pH 2–9, suggesting the proposed AR-SiO₂ sensor can be used in pH 2–9.

The extraction ability of the AR-SiO₂ was also estimated by measuring the amount of Hg²⁺ ion adsorbed on the AR-SiO₂ by ICP, resulting in 90% of Hg²⁺ ion being extracted by AR-SiO₂. This result suggests that the AR-SiO₂ is potentially useful as a stationary phase for the separation of Hg²⁺ ion in liquid chromatography. Also, the adsorption capacity of the AR-SiO₂ was measured through the solid extraction using solutions of binary metal ions (Hg²⁺/Fe³⁺, Hg²⁺/Co²⁺, Hg²⁺/Cd²⁺, Hg²⁺/Pb²⁺ and Hg²⁺/Zn²⁺) resulting in 85.7–90.5% of Hg²⁺ ion being adsorbed by AR-SiO₂ (Fig. 5). In contrast, other metal ions such as Zn²⁺, Cd²⁺, Pb²⁺ and Zn²⁺ were extracted into the solid phase at percentages of only 2.5–5.2% in the competition system. These results suggest that AR-SiO₂ is a useful adsorbent for the selective separation of Hg²⁺ over a range of transition and heavy metal ions.

Fig. 5 Adsorption capacities of AR-SiO₂ for (a) single and (b) binary metal ions in H₂O.

Coloration of TiO₂ films with metal ions

We now address the metal cation sensing ability of AR-TiO₂ films. For the cation sensing experiments, the AR-TiO₂ films were dipped into cation solutions of interest, then dried in air, with any color change determined by “naked-eye” inspection and by measurement of solid UV-vis absorption spectra. In all cases, changes in film color were found to be complete within 1–2 s of immersion, with the resulting coloration being stable for in excess of one day, both for films immersed continuously in the cation solution and for films dried in air. Typical data are shown in Fig. 6, showing a photographic image of the films after immersion in 1.0 mM solutions of chloride salts of the seven cations (Co²⁺, Cd²⁺, Pb²⁺, Zn²⁺, Fe³⁺, Cu²⁺ and Hg²⁺). The yellow color of the AR-TiO₂ film changed to a deep red color when the film was dipped in Hg²⁺ solution. On the other hand, no significant color changes were observed with the other metal ions. These results indicate that the AR-TiO₂ film selectively detected only the Hg²⁺ ion. Since the preparation of the AR-TiO₂ films prepared in this work is much simpler than the sol–gel grafting method, we propose that this AR-TiO₂ film is highly useful as a chemosensor to detect Hg²⁺ ions.

Fig. 6 Color changes observed after the AR-TiO₂ films were dipped in 1.0 mM aqueous solutions of the analytes: (a) the absence and the presence of (b) HgCl₂, (c) CoCl₂, (d) CdCl₂, (e) PbCl₂, (f) ZnCl₂, (g) FeCl₃ and (h) CuCl₂.

We also observed the solid UV-vis spectra of the AR-TiO₂ films after dipping into the various cation solutions. Fig. 7 shows the corresponding absorption spectra for selected films. It is apparent that the film coloration is only sensitive to the Hg²⁺ ion. Hg²⁺ exposure resulted in a deep red coloration of the film caused by a shift of the absorption maximum from 310 to 492 nm. This bathochromic shift is similar to that observed for AR-SiO₂. On the other hand, no significant changes in the UV-vis spectra were observed for any other cations tested (including Co²⁺, Cd²⁺, Pb²⁺, Zn²⁺, Fe³⁺ and Cu²⁺) for concentrations up to 1.0 mM. Furthermore, the bathochromic shift observed with the Hg²⁺ ion was insensitive to the presence of other cations in the solution, indicating that other metal cations do not interfere with the ability of AR-TiO₂ to detect Hg²⁺.

Fig. 7 (A) Solid UV-vis spectra of AR-TiO₂ films in the (a) absence and (b) the presence of HgCl₂ (1.5 μM) aqueous solution.

The color change observed following the dipping of AR-TiO₂ into Hg²⁺ solution (1.0 μM) was found to be fully reversible when AR-TiO₂ was rinsed thoroughly with EDTA (10.0 μM). Reusability was evaluated by repeated dipping/rinsing cycles, with the AR-TiO₂ absorption spectrum being recorded after each step. Typical data are shown in Fig. 8. The data show that AR-TiO₂ exhibits excellent reusability, as almost no loss in AR-TiO₂ sensitivity was observed after eight repeated dipping/rinsing cycles. In addition, the AR-TiO₂ film before and after dipping of Hg²⁺ ion solution showed transparency with a deep red color (Fig. S14, ESI).†

Fig. 8 Plot for the absorbance of AR-SiO₂ by alternated dipping in 1.0 μM aqueous solution of Hg²⁺ ion (“ON”) and 10.0 μM EDTA (“OFF”). The cyclic index is the number of alternating dipping/rinsing cycles, with the vertical axis showing the absorbance for the AR-SiO₂ at 492 nm.

To develop the AR-TiO₂ as a general Hg²⁺ ion sensor which is independent of any anions present, we also added the anions as I⁻, Br⁻, NO₃⁻ and ClO₄⁻. With the anion addition, the color of the AR-TiO₂ in HgCl₂ solution still changed from yellow to deep red. This finding indicates that AR-TiO₂ can be employed for the detection of Hg²⁺ independently of other anion(s) present.

Fig. S15 (ESI)† shows standard calibration data (Abs. vs. [Hg²⁺]) for AR-TiO₂. A linear response is observed (between 28 nM and 1.0 μM) with a sensitivity of ∼28 nM. This sensitivity is much higher than those previously reported for spectrophotometric sensors anchored to mesoporous aluminosilicates.²⁴ Furthermore, the sensitivity of the AR-TiO₂ film for Hg²⁺ was much higher than that of AR-SiO₂, because AR-TiO₂ films are transparent and/or have higher immobilization of receptor onto TiO₂ nanoparticles. These findings demonstrate that AR-TiO₂ has strong potential as a portable chemosensor for Hg²⁺ ions.

AR-TiO₂ film function as a pH buffer

We consider next the buffer action of the AR-TiO₂ film. The AR-TiO₂ film was immersed in aqueous solutions with pH values ranging between 1.0 and 11.0. Control experiments were conducted using 1.0 mM of Hg²⁺ solution controlling the pH range. The resulting film coloration is shown in Fig. 9. The absorbances of AR-TiO₂ in the presence of Hg²⁺ ion were almost constant, with pH values ranging between 4 and 11. The absorbances of AR-TiO₂ are independent between pH 4 and 11. We conclude that AR-TiO₂ can reliably detect the Hg²⁺ ion over the wide pH range of pH 4–11.

Fig. 9 Color changes observed after AR-TiO₂ films were dipped in Hg²⁺ aqueous solution (1.0 mM) at different pH values.

Conclusions

We have fabricated AR-SiO₂ and AR-TiO₂ using a functional azobenzene-coupled acyclic receptor. The AR-SiO₂ recognized and separated Hg²⁺ with a high degree of selectivity among heavy metal ions in aqueous solution. Beyond its immediate applications in environmental cleanup, AR-SiO₂ provides a unique opportunity to introduce molecular binding sites and to rationally design the surface properties of inorganic nanomaterials. Furthermore, AR-TiO₂ films with 4 μm thickness on glass substrates were prepared as portable chemosensors for Hg²⁺ ion in aqueous solution. The AR-TiO₂ films show excellent sensitivity and selectivity for Hg²⁺ ion. The AR-TiO₂ film detected Hg²⁺ in pH = 7.4 with a sensitivity of 28 nM. The Hg²⁺ detection ability was unaffected by the presence of other cations or anions. The results imply that the AR-TiO₂ film has strong potential as a portable colorimetric sensor for detection of Hg²⁺ in the environmental field. We believe the combination of well-defined inorganic nanomaterials and functionalized organic receptors can play a pivotal role in the development of a new generation of hierarchical structures and functionalized composites.

Acknowledgements

This work was supported by Korea Science and Engineering Foundation Grant (2009-0074100, 2008-2001014), and World Class University (WCU) Program supported from Ministry of Education, Science and Technology (R32-2008-000-20003-0), and Environmental-Fusion Project (191-091-004) supported from Ministry of Environment.

References

1. G. K. Walkup and B. Imperiali, J. Am. Chem. Soc., 1996, 118, 3053.
2. A. Miyawaki, J. Llopis, R. Helm, J. M. McCaffery, J. A. Adams, M. Ikura and R. Y. Tsien, Nature, 1997, 388, 882.
3. M. M. Henary and C. J. Fahrni, J. Phys. Chem. A, 2002, 106, 5210.
4. J. Homola, S. S. Yee and G. Gauglitz, Sens. Actuators, B, 1999, 54, 3.
5. F. Turner, Science, 2000, 290, 1315.
6. P. Chen and C. He, J. Am. Chem. Soc., 2004, 126, 728.
7. S. J. Lee, J.–E. Lee, J. Seo, I. Y. Jeong, S. S. Lee and J. H. Jung, Adv. Funct. Mater., 2007, 17, 3441.
8. S. J. Lee, S. S. Lee, J. Y. Lee and J. H. Jung, Chem. Mater., 2006, 18, 4713.
9. E. Palomares, R. Vilar, A. Green and J. R. Durrant, Adv. Funct. Mater., 2004, 14, 111.
10. E. C. Lima, J. L. Brasil and A. H. D. P. Santos, Anal. Chim. Acta, 2003, 484, 233.
11. E. C. Lima, J. L. Brasil and J. C. P. dos Vaghetti, Talanta, 2003, 60, 103.
12. N. G. Becka, R. P. Franks and K. W. Bruland, Anal. Chim. Acta, 2002, 455, 11.
13. E. Palomares, R. Vilar and J. R. Durrant, Chem. Commun., 2004, 362.
14. J. S. Wu, I.-C. Hwang, K. S. Kim and J. S. Kim, Org. Lett., 2007, 9, 907.
15. M. H. Lee, J. S. Wu, J. W. Lee and J. S. Kim, Org. Lett., 2007, 9, 2501.
16. E. M. Nolan and S. J. Lippard, Chem. Rev., 2008, 108, 3443.
17. S. O. Obare, R. E. Hollowell and C. J. Murphy, Langmuir, 2002, 18, 10407.
18. T. Balaji, S. A. El-Safty, H. Matsunaga, T. Hanaoka and F. Mizukami, Angew. Chem., Int. Ed., 2006, 45, 7202.
19. E. Palomares, M. V. Martínez-Díaz, T. Torres and E. Coronado, Adv. Funct. Mater., 2005, 15, 803.
20. R. Metivier, I. Leray, B. Lebeau and B. Valeur, J. Mater. Chem., 2005, 15, 2965.
21. M. Boiocchi, M. Bonizzoni, L. Fabbrizzi, G. Piovani and A. Taglietti, Angew. Chem., Int. Ed., 2004, 43, 3847.
22. M. H. Lee, S. J. Lee, J. H. Jung and J. S. Kim, Tetrahedron, 2007, 63, 12087.
23. S. J. Lee, J. H. Jung, J. Seo, I. Yoon, K.-M. Park, L. F. Lindoy and S. S. Lee, Org. Lett., 2006, 8, 1641.
24. A. B. Descalzo, D. Jimenz, J. E. Haskouri, D. Beltran, P. Amoros, M. D. Marcos, R. Martinez-Manez and J. Soto, Chem. Commun., 2002, 562.
25. E. Palomares, R. Vilar, A. Green and J. R. Durrant, Adv. Funct. Mater., 2004, 14, 111.
26. W. S. Han, H. Y. Lee, S. H. Jung, S. J. Lee and J. H. Jung, Chem. Soc. Rev., 2009, 38, 1904.
27. H. J. Kim, S. J. Lee, S. Y. Park, J. H. Jung and J. S. Kim, Adv. Mater., 2008, 20, 3229.
28. H. Y. Lee, D. R. Bae, J. C. Park, H. Song, W. S. Han and J. H. Jung, Angew. Chem., Int. Ed., 2009, 48, 1239.
29. J. V. Ros-Lis, R. Casasús, M. Comes, C. Coll, M. D. Marcos, R. Martínez-Máñez, F. Sancenón, J. Soto, P. Amorós, J. E. Haskouri, N. Garró and K. Rurack, Chem.–Eur. J., 2008, 14, 8267.
30. D. Li, A. Wieckowska and I. Willner, Angew. Chem., Int. Ed., 2008, 47, 1.

---

Footnote

† Electronic supplementary information (ESI) available: Measurement data. See DOI: 10.1039/b915975d

---