Three Isomeric Zn(II)-Sn(IV)-Zn(II) Porphyrin-Triad-Based Supramolecular Nanoarchitectures for the Morphology-Dependent Photocatalytic Degradation of Methyl Orange.

Three isomeric Zn(II)-Sn(IV)-Zn(II) porphyrin-based triads (T2, T3, and T4) were synthesized by the reaction of common Zn(II) porphyrins (ZnL) with different Sn(IV) porphyrins (SnP n ). The Sn(IV) porphyrin precursors differ with respect to the position of the pyridyl-N atoms. All compounds were characterized by 1H NMR, UV-vis, fluorescence spectroscopy, electrospray ionization-mass spectrometry, and field-emission scanning electron microscopy measurements. In these structures, the intramolecular cooperative metal-ligand coordination of the 3-pyridyl nitrogen in SnP 3 with axial ZnL and the π-π interactions between the adjacent porphyrin triad are the determining factors affecting the nanostructures of T3. Owing to the geometrical constraints of the SnP 2 center, this type of interaction is not possible for T2. Therefore, only the π-π interactions affect the self-assembly process. In the case of SnP 4 , intermolecular coordinative interactions and then π-π interactions are responsible for the nanostructure of T4. The morphology-dependent photocatalytic degradation of methyl orange (MO) dye in aqueous solution under visible light irradiation was observed for these photocatalysts, and the degradation ratio of MO varied from 76 to 94% within 100 min. Nanorod-shaped T3 exhibited higher performance compared to nanosphere T2 and nanoflake T4.

Introduction

Highly ordered and well-defined nanomaterials have drawn tremendous attention in the fields of applied and materials chemistry owing to their varied applications, such as the conversion and storage of solar energy,[1] catalysis,[2] sensing,[3] molecular recognition,[4] and biomedicine.[5] In addition, these materials have high surface areas,[6] excellent corrosion resistance,[7] high electrical conductivity,[8] outstanding optoelectronic properties,[9] and high thermal stability[10] compared to their parent constituents. Thus, a large number of building blocks have been used to construct nano- and microstructured nanomaterials having well-defined shapes and dimensions.[11−14] Porphyrinoids (free porphyrin or metalloporphyrin compounds) are emerging building blocks for the fabrication of self-assembled functional nanomaterials having light-harvesting properties or charge-transfer functionality.[15−18] In general, free porphyrin or metalloporphyrin compounds self-assemble and form large aggregates in solution. Several intermolecular noncovalent interactions (for example, hydrogen bonding, van der Waals, π–π stacking, ligand coordination, electrostatic, hydrophobic, and hydrophilic interactions) are responsible for the self-assembly of porphyrinoid compounds.[19−23] The morphology of these nanostructures is not always regular, making them unsuitable for practical applications. Therefore, the design and construction of novel nanoarchitectures having definite sizes, shapes, and dimensions is challenging. Sn(IV) porphyrin compounds have been widely used as building blocks in the construction of nanomaterials, including nanofibers,[24] nanotubes,[25] nanosheets,[26] nanorods,[27] and nanocomposites,[28] and these nanoaggregates have applications for the production of hydrogen gas under visible-light irradiation[29] and the photocatalytic degradation of organic dyes in aqueous solution.[30,31] Because of the oxophilic nature of Sn(IV) centers, they form stable six-coordinated complexes with two trans-oriented oxyanion ligands (alkoxides or carboxylates) and thus are ideal scaffolds for the construction of supramolecular architectures.[32−38] These compounds have interesting photophysical properties and are diamagnetic. Therefore, structural information can be obtained from the Sn nuclei, which are NMR-active. To date, a large number of methods, including ionic self-assembly,[39] reprecipitation,[40] sonication,[41] metal–ligand coordination,[42] and surfactant-assisted methods, have been used for the construction of well-defined, discrete self-assembled porphyrin nanoaggregates.[43]

Recently, we have reported several porphyrin-based nanostructures obtained from the reaction between trans-dihydroxo-[5,15-bis(3-pyridyl)-10,20-bis(phenyl)porphyrinato] tin(IV) and different Zn(II) porphyrins.[30,31] Intramolecular coordination between the pyridyl nitrogen atom of the Sn(IV) porphyrin and Zn(II) porphyrin, followed by the self-assembly of porphyrin triads, is the main driving force for the formation of these nanostructures. However, studies have been limited to the use of the 3-pyridyl position. Therefore, in this study, we examined the coordination of other pyridyl positions in the Sn(IV) porphyrin to Zn(II) porphyrin. Here, we report three new triad compounds synthesized from the reaction of a Zn(II) porphyrin, ZnL, with three different tin(IV) porphyrin precursors. These Sn(IV) porphyrin precursors differ in the trans-pyridyl positions (Chart ). The aim of the present study is to determine the correlation between the coordination mode (pyridyl-N of the Sn(IV) porphyrin with the Zn(II) porphyrin) and its effect on the nanomorphology. Our results suggest that the coordination structure affects the morphology of these nanostructures.

Chart 1

Chemical Structures of Three Isomeric Triads Used in the Present Study

In addition, porphyrin-based nanomaterials have tremendous applications in the removal of pollutant dyes from aqueous solutions.[30,31] Water pollution caused by the discharge of highly toxic and hazardous dyes poses a serious threat to the environment and has negative effects on aquatic flora and fauna.[44] Methyl orange (MO) is one of the 20 dyes detected most frequently in wastewater, and it can cause respiratory, skin, and eye irritation. MO is a water-soluble aromatic azo dye (commonly used as a pH indicator) and is extensively used in several industries, including the textile, paper, printing, and food industries. Every year, the textile industry discharges large quantities of MO into the aqueous ecosystem, resulting in nonpotable water.[45,46] Thus, we also studied the morphology-dependent photocatalytic degradation of MO in aqueous solution by these photocatalysts.

Experimental Section

Materials and Methods

All chemicals were purchased from Sigma-Aldrich and used without further purification unless otherwise stated. The trans-dihydroxo-[5,15-bis(2-pyridyl)-10,20-bis(phenyl)porphyrinato] tin(IV) (SnP), trans-dihydroxo-[5,15-bis(3-pyridyl)-10,20-bis(phenyl)porphyrinato] tin(IV) (SnP), and trans-dihydroxo-[5,15-bis(4-pyridyl)-10,20- bis(phenyl)porphyrinato] tin(IV) (SnP) complexes were synthesized according to a previously reported procedure.[47,30,37] All experiments were carried out under a dry argon atmosphere using standard Schlenk techniques. Toluene was purified by distillation from a sodium/benzophenone ketyl solution, whereas pyrrole was obtained from a solution of calcium hydride. Steady-state UV–vis spectra were recorded on a Shimadzu UV-3600 spectrophotometer. The fluorescence spectra were recorded using a Shimadzu RF-5301PC fluorescence spectrophotometer. The 1H NMR spectra were obtained on a Bruker BIOSPIN/AVANCE III 400 spectrometer at 293 K, and the electrospray ionization mass spectra (ESI-MS) were recorded on a Thermo Finnigan linear ion trap quadrupole mass spectrometer. Elemental analysis was performed using an EA 1110 Fisons analyzer. Field emission scanning electron microscopy (FE-SEM) images were obtained using a MAIA III SEM device. For sample preparation, each compound was suspended in toluene at a fixed concentration (c = 0.5 mM) and centrifuged at 13 500 rpm for 20 s. After that, this solution was drop-cast on the surface of Cu tapes and then dried in air. After deposition and before the FE-SEM measurements, a Pt coating was applied. The Brunauer–Emmett–Teller (BET) surface area using N2 adsorption isotherms at 77 K was determined with an analyzer (BELSORP-mini volumetric adsorption equipment). The total organic carbon (TOC) values of the degradation of MO were detected with a JEOL JEM-3010 total organic carbon analyzer to determine the extent of mineralization.

Synthesis of 4-Octoxybenzaldehyde

For the synthesis of 4-octoxybenzaldehyde, we used a previously reported procedure with some modifications.[48] Briefly, a mixture of 4-hydroxybenzaldehyde (8.84 g, 72.4 mmol), K2CO3 (14 g, 109 mmol), and KI (1.08 g, 7.24 mmol) was dissolved in dry dimethylformamide (DMF, 50 mL) and stirred at room temperature under an argon atmosphere. Then, 1-bromooctane (14 mL, 72.4 mmol) was added to the above reaction mixture and held at 100 °C for 24 h. The reaction mixture was then filtered, and the filtrate was removed under vacuum. The residue was dissolved in ethyl acetate and washed with water. The organic layer was then dried over anhydrous MgSO4, and the solvent was evaporated under reduced pressure. Pure 4-octoxybenzaldehyde was separated from the crude product by column chromatography (ethyl acetate/n-hexane = 1/10) and dried under vacuum. Yield: 15.2 g (90%). 1H NMR (400 MHz, CDCl3, ppm): δ 0.84 (t, 3H, CH3), 1.15–1.30 (m, 10H, alkyl chain), 1.74 (m, 2H, O–C–CH2), 3.97 (t, 2H, O–CH2), 6.94 (d, J = 8.7 Hz, 2H, H3,5-PhCHO), 7.39 (d, J = 8.7 Hz, 2H, H2,6-PhCHO), 9.81 (s, H, CHO).

Synthesis of meso-5-(4-Hydroxyphenyl)-10,15,20-tris(4-octoxyphenyl)porphyrin, H

The free base porphyrin was synthesized by mixed aldehyde condensation in propionic acid using 1.0 equiv of 4-hydroxybenzaldehyde, 3.0 equiv of 4-substituted benzaldehyde, and 4.0 equiv of pyrrole. In a typical reaction, pyrrole (0.77 mL, 11.1 mmol) was added dropwise to a solution of 4-hydroxybenzaldehyde (0.34 g, 2.8 mmol) and 4-octoxybenzaldehyde (1.97 g, 8.4 mmol) in propionic acid (200 mL) under reflux. The mixture was stirred for 3 h, and then the solvent was evaporated under reduced pressure. The crude product was purified by column chromatography (SiO2, n-hexane/CH2Cl2 eluent) to afford HL and recrystallized from a CH2Cl2/acetonitrile mixture to give a violet–purple powder. Yield: 0.243 g (8%). Anal. Calcd for C73H90N4O4: C, 80.62; H, 8.34; N, 5.15; R, 5.88. Found: C, 81.21; H, 8.62; N, 4.90; R, 5.27. 1H NMR (400 MHz, CDCl3, ppm): δ −2.79 (s, 2H, NH), 0.92 (t, 9H, CH3), 1.34–1.48 (m, 24H, alkyl chain), 1.63 (m, 6H, O–C–C–CH2), 1.96 (m, 6H, O–C–CH2), 4.22 (t, 6H, O–CH2), 7.16 (d, J = 8.4 Hz, 2H, H2,6-PhOH), 7.30 (d, J = 8.4 Hz, 6H, H2,6-Ar), 8.04–8.08 (m, 8H, meso-O-Ar + meso-O-PhOH), 8.85 (m, 8H, β-pyrrole). UV–visible (CHCl3): λnm (log ε), 423 (5.31), 522 (4.21), 559 (3.92), 597 (3.87), 649 (3.83). Emission (CHCl3, λnm): 652, 712.

Synthesis of meso-5-(4-Hydroxyphenyl)-10,15,20-tris(4-octoxyphenyl)porphyrinato)zinc(II), ZnL

Zn(OAc)2·2H2O (0.053 g, 0.29 mmol) was added to a solution of HL (0.122 g, 0.12 mmol) in a mixture of tetrahydrofuran (THF, 15 mL) and methanol (15 mL). The mixture was refluxed for 3 h. The solvent was then removed, and the solid product was purified by column chromatography (SiO2, n-hexane/CH2Cl2 eluent). The crude product was recrystallized from CH2Cl2/acetonitrile to yield a brownish-red color. Yield: 0.113 g (90%). Anal. Calcd for C73H88N4O4Zn: C, 76.18; H, 7.71; N, 4.87; R, 11.24. Found: C, 76.02; H, 7.93; N, 4.73; R, 11.32. 1H NMR (400 MHz, deuterated dimethyl sulfoxide (DMSO-d6), ppm): δ 0.92 (t, 9H, CH3), 1.30–1.50 (m, 24H, alkyl chain), 1.56 (m, 6H, O–C–C–CH2), 1.90 (m, 6H, O–C–CH2), 4.24 (t, 6H, O–CH2), 7.17 (d, J = 8.4 Hz, 2H, H2,6-PhOH), 7.32 (d, J = 8.4 Hz, 6H, H2,6-Ar), 7.96 (d, J = 8.0 Hz, 2H, meso-O-PhOH), 8.05 (d, J = 8.0 Hz, 6H, meso-O-Ar), 8.81 (m, 8H, β-pyrrole), 9.90 (s, 1H, OH). UV–visible (CHCl3): λnm (log ε), 425 (5.40), 560 (4.03), and 601 (3.78). Emission (CHCl3, λnm): 610, 657.

General Procedure for the Synthesis of the Triads

A mixture of ZnL (0.092 g, 0.08 mmol) and SnP (0.031 g, 0.04 mmol) was added to anhydrous toluene (10 mL) and refluxed for 48 h under an Ar atmosphere. Then, the solvent was removed (for T2 and T4), and the solid compound was purified by neutral silica chromatography (n-hexane/CH2Cl2 eluent). For T3, the reaction mixture was cooled to room temperature, and the colored precipitate was filtered out, washed with toluene, and dried under vacuum.

Synthesis of Triad T2

Brick-red compound. Yield: 0.092 g, 80%. Anal. Calcd for C178H176N14O8SnZn2: C, 74.00; H, 6.14; N, 6.76; R, 13.10. Found: C, 73.68; H, 6.27; N, 6.54; R, 13.51. 1H NMR (400 MHz, DMSO-d6, ppm): δ 0.92 (t, 18H, CH3), 1.24–1.48 (m, 48H, alkyl chain), 1.58 (m, 12H, O–C–C–CH2), 1.88 (m, 12H, O–C–CH2), 2.65 (d, J = 8.6 Hz, 4H, α-bridging phenyl), 4.26 (t, 12H, O–CH2), 6.70 (d, J = 8.8 Hz, 4H, β-bridging phenyl), 7.26 (d, J = 8.4 Hz, 12H, m-phenyl-axial), 7.48 (m, 8H, m,p-phenyl-central + H4-Py-central), 7.70 (m, 14H, meso-O-phenyl-axial + H5-Py-central), 8.09 (m, 6H, H3-Py-central + meso-O-phenyl-central), 8.38 (d, J = 7.8 Hz, 4H, β-pyrrole-axial), 8.55 (d, J = 7.8 Hz, 4H, β-pyrrole-axial), 8.70 (m, 8H, β-pyrrole-axial), 9.08 (d, J = 8.2 Hz, 4H, β-pyrrole-central), 9.18 (d, J = 8.2 Hz, 4H, β-pyrrole-central), 9.32 (d, J = 5.4 Hz, 2H, H6-Py). UV–vis (CHCl3): λnm (log ε), 422 (5.60), 560 (4.60), and 602 (4.35). Emission (CHCl3, λnm): 598, 644.

Synthesis of Triad T3

Bluish-green compound. Yield: 0.105 g, 91%. Anal. Calcd for C178H176N14O8SnZn2: C, 74.00; H, 6.14; N, 6.76; R, 13.10. Found: C, 73.82; H, 6.47; N, 6.66; R, 13.05. 1H NMR (400 MHz, DMSO-d6, ppm): δ 0.91 (t, 18H, CH3), 1.20–1.48 (m, 48H, alkyl chain), 1.57 (m, 12H, O–C–C–CH2), 1.90 (m, 12H, O–C–CH2), 2.67 (d, J = 8.6 Hz, 4H, α-bridging phenyl), 4.26 (t, 12H, O–CH2), 6.70 (d, J = 8.8 Hz, 4H, β-bridging phenyl), 7.27 (m, 18H, {m,p-phenyl}-axial), 7.60–7.76 (m, 20H, m,p-phenyl-central + H5-Py-central + meso-O-phenyl-axial), 8.12 (m, 6H, H4-Py-central + meso-O-phenyl-central), 8.42 (d, J = 7.8 Hz, 4H, β-pyrrole-axial), 8.57 (d, J = 7.8 Hz, 4H, β-pyrrole-axial), 8.72 (m, 8H, β-pyrrole-axial), 9.08 (d, J = 8.2 Hz, 4H, β-pyrrole-central), 9.18 (d, J = 8.2 Hz, 4H, β-pyrrole-central), 9.30 (d, J = 5.2 Hz, 2H, H6-Py), 9.55 (s, 2H, H2-Py). UV–visible (CHCl3): λnm (log ε), 428 (5.49), 453 (5.35), 565 (4.81), 613 (4.78). Emission (CHCl3, λnm): 598, 645.

Synthesis of Triad T4

Dark-red compound. Yield: 0.097 g, 84%. Anal. Calcd for C178H176N14O8SnZn2: C, 74.00; H, 6.14; N, 6.76; R, 13.10. Found: C, 73.49; H, 6.32; N, 6.56; R, 13.65. 1H NMR (400 MHz, DMSO-d6, ppm):): δ 0.91 (t, 18H, CH3), 1.22–1.48 (m, 48H, alkyl chain), 1.58 (m, 12H, O–C–C–CH2), 1.89 (m, 12H, O–C–CH2), 2.66 (d, J = 8.6 Hz, 4H, α-bridging phenyl), 4.26 (t, 12H, O–CH2), 6.70 (d, J = 8.8 Hz, 4H, β-bridging phenyl), 7.26 (d, J = 8.4 Hz, 12H, m-phenyl-axial), 7.56 (m, 6H, m,p-phenyl-central), 7.72 (d, J = 8.0 Hz, 12H, meso-O-phenyl-axial), 8.14 (m, 8H, H3-Py-central + meso-O-phenyl-central), 8.42 (d, J = 7.8 Hz, 4H, β-pyrrole-axial), 8.58 (d, J = 7.8 Hz, 4H, β-pyrrole-axial), 8.75 (m, 8H, β-pyrrole-axial), 9.09 (m, 8H, β-pyrrole-central), 9.26 (d, J = 5.4 Hz, 4H, H6-Py). UV–visible (CHCl3): λnm (log ε), 426 (5.53), 448 (5.01), 563 (4.57), 605 (4.43). Emission (CHCl3, λnm): 598, 646.

Photocatalytic Degradation of Methyl Orange

The photocatalytic activity of the nanostructures was studied by monitoring the degradation of methyl orange (MO) dye in an aqueous medium. In a typical process, 10 mg of the photocatalyst (T2, T3, or T4) was added to 150 mL of an aqueous MO solution (20 mg L–1, pH 7) with stirring. Before visible-light irradiation, the mixture was kept in the dark for 15 min to reach an adsorption–desorption equilibrium. Subsequently, the reaction mixture was exposed to visible light irradiation from a 150 W xenon arc lamp (ABET Technologies, USA) with a UV cutoff filter at room temperature. All sample solutions were collected by centrifugation (to remove the photocatalyst) at given time intervals to determine the MO concentration by UV–vis spectroscopy at 461 nm.

Results and Discussion

Synthesis and Characterization

For the synthesis of the three isomeric triads in Chart , we used the “axial bonding” approach.[30,31,49−52] All of these triads were synthesized by refluxing meso-5-(4-hydroxyphenyl)-10,15,20-tris(4-octoxyphenyl)porphyrinato)zinc(II) (ZnL) with the corresponding trans-dihydroxo Sn(IV) porphyrins SnP in anhydrous toluene for 48 h under an argon atmosphere (Scheme ). Complete details of the experiments are given in the Experimental Section. The average yield was found to be more than 70% in all cases. The strong affinity of the aryloxides toward the Sn(IV) porphyrin center was the driving force for the formation of these triads. All of the synthesized compounds were fully characterized by elemental analysis and spectroscopic techniques including 1H NMR spectroscopy, ESI-MS, UV–vis spectroscopy, and fluorescence spectroscopy.

Scheme 1

Synthesis Scheme Used to Prepare the Triads

The 1H NMR spectra of all three isomeric triads and the individual unconnected monomers (HL and ZnL) are given in the Supporting Information (Figures S1–S5). All of the peaks were assigned, and the peak integrals are given in the Experimental Section. From the 1H NMR spectra of these three triads, it is clear that the resonance positions of the aromatic protons as well as the β-pyrrolic protons of the central Sn(IV) porphyrin are not significantly different from those of the starting Sn(IV) porphyrins (SnP).

In the 1H NMR spectra, the pyridyl protons of the triads appear as doublets (9.32 ppm for T2, 9.30 and 9.55 ppm for T3, and 9.26 ppm for T4). In contrast, the β-pyrrolic protons of triads T2 and T3 appear from 9.08 to 9.18 ppm (doublet), and those for T3 appear as a doublet at 9.09 ppm. The remaining aromatic protons of the central SnP unit in the triads appear from 7.5 to 8.15 ppm. On the other hand, the protons of the axial Zn(II) porphyrins in the triads are different from those of the monomeric ZnL in terms of the splitting patterns and resonance positions. This is due to the strong ring-current effects of the central Sn(IV) porphyrins on the axial Zn(II) porphyrins. In particular, the aryloxy protons appeared as a doublet at 7.17 and 7.96 ppm in the NMR spectrum of monomeric ZnL. In the triads, these two protons experienced strong shielding effects from the ring-current effects of the central Sn(IV) porphyrins, appearing as a doublet at 2.65 and 6.70 ppm. The Δδ values [that is, δ(ZnL) – δ(triad)] for these protons are approximately 4.52 and 1.26 ppm, respectively. In addition, the β-pyrrole protons of the axial Zn(II) porphyrin in the triads are exposed to ring-current effects similar to that of monomeric ZnL. Specifically, the β-pyrrole protons resonated at 8.81 ppm in monomeric ZnL. However, these protons in the triads were shifted upfield and split into three different positions (approximately 8.40, 8.55, and 8.70 ppm). The chemical shifts of the remaining protons associated with the axial Zn(II) porphyrins showed no further changes compared to those of monomeric ZnL. 1H NMR measurements have been extensively used to explore “axial-bonding”-type architectures based on the interactions between the protons of axial Zn(II) porphyrins and those of the central Sn(IV) porphyrin involving ring-current-induced shifts and resonance couplings.[30,31,49,50]

The results of ESI-MS measurements of all three triads are shown in the Supporting Information (Figures S6–S8). From Figure S6, it is clear that compound T2 was fragmented during the MS experiments.[30,31] However, the molecular ion peak [T2 + H]+ at a mass-to-charge (m/z) ratio of 2889.89 appeared with low intensity. On the other hand, a base peak at 769.13, corresponding to [SnP + H]+, was observed. Similar MS patterns were observed for the other two triads.

The UV–visible spectra of all three triads in CHCl3 solution, along with that of ZnL, are shown in Figure . The maximum absorbance (λmax) and molar extinction coefficient (ε) of all of the compounds are reported in the Experimental Section. The monomeric Zn(II) porphyrin (ZnL) had a Soret band (at approximately 425 nm) and Q-bands (at approximately 560 and 601 nm). Insignificant changes in the spectrum of T2 were observed (the Soret band at 422 nm is quite broad) compared to that of ZnL. Furthermore, the molar extinction coefficient of T2 is very close to the sum of those of the monomeric Sn(IV) porphyrin and ZnL. In contrast, the spectral features of T3 and T4 were different from those of T2. The spectrum of T3 is broad in the Soret region with a small red shift (approximately 428 nm) along with the Q-bands (approximately 565 and 613 nm) compared to those of ZnL. Additionally, a peak at approximately 453 nm was also found for T3. In the case of T4, the Soret band is present at approximately 426 nm, and the Q-bands are present at approximately 563 and 605 nm, along with a small peak at 448 nm.

Figure 1

UV–visible spectra of the three triads and ZnL in CHCl3.

The ground-state data indicate that there are minimal perturbations to the electronic structures of the individual macrocyclic π-systems in these triads. In addition, axial–axial, basal–basal, or axial–basal couplings between the axial Zn(II) porphyrin and basal Sn(IV) porphyrin rings were not observed.[30,31,49,50] Therefore, the red shift in the peak position (λmax) and the peak broadening of the Soret bands indicate that a J-type or step-like self-assembled arrangement is possible in solution.[30,31,53]

Steady-state fluorescence spectra of all three triads, along with those of ZnL, were recorded in CHCl3 and are shown in Figure . ZnL yielded two-banded fluorescence spectra having emission maxima at 610 and 657 nm. In contrast, all of the triads showed similar two-banded spectra having emission maxima at 598 and 645 nm. The peak-to-peak intensity ratio also decreased from ZnL to these triads. As shown in Figure , the emissions of all these triads were quenched, and the degree of quenching was significantly affected by their aggregation in solution.[30,31]

Figure 2

Fluorescence spectra of all three triads and ZnL in CHCl3 (λext = 550 nm). The optical density (OD) of each sample solution was fixed at 0.25.

Supramolecular Self-Aggregation

The morphology and self-assembly behavior of all of the compounds were analyzed using FE-SEM. The FE-SEM images are shown in Figure . Figure a shows that ZnL has a nanostructured morphology, but the particle shape and size are very irregular. In contrast, nanospheres of different sizes were observed for T2 (Figure b), and the average diameter of the smaller nanospheres ranged from 60 to 80 to nm, whereas the average diameter of the larger nanospheres varied from to 220 to 240 nm. Well-defined nanorods were observed for T3 (Figure c), the average length of the nanorods varied from 160 to 900 nm, the average width of the nanorods ranged from 40 to 90 nm, and the height of the nanorods was approximately 20 nm. On the other hand, variably shaped nanoaggregates were observed for T4 (Figure d) and comprised small particles, along with some long nanoflakes with interconnected smaller particles. The average lengths of the nanoflakes varied from 800 to 1200 nm. The average width of the nanoflakes varied from 200 to 260 nm.

Figure 3

FE-SEM images showing the morphologies of (a) ZnL, (b) T2, (c) T3, and (d) T4.

From the FE-SEM images in Figure , it is clear that all of the compounds formed nanoscale aggregates. Typically, porphyrin (free-base porphyrin or metalated porphyrin) molecules are aggregated into nanostructures via self-assembly. Here, monomeric Zn(II) porphyrins are self-assembled via π–π stacking interactions between adjacent porphyrin molecules and form nanostructures. Similarly, T2 aggregated to form spherical particles as well. In contrast, the morphology of T3 was quite different: nanorods with well-defined shapes and sizes. The pyridyl groups in T3 coordinate intramolecularly to the Zn ion in the axial Zn(II) porphyrin, thus locking the axial porphyrin units into position, which facilitates self-assembly via π–π stacking interactions through face-to-face interactions.[30,31] In contrast, for T4, the coordination of the pyridyl-N to the axial Zn(II) porphyrins occurs in an intermolecular manner. Therefore, intermolecular coordination and π–π stacking interactions drive the aggregation of T4. A similar coordination mode is not possible in T2 because of the geometrical constraints within the molecule. It should be noted that the other starting porphyrin monomers, SnP, did not show any aggregation under our experimental conditions.[30,31]

Photocatalytic Degradation of Methyl Orange (MO)

The photocatalytic degradation of MO in aqueous solution was used to evaluate the photocatalytic activity of the three triads under visible light irradiation, and the results are shown in Figure . For comparison, the photocatalytic performance of monomeric ZnL was also measured. The degradation of MO dye in the presence of monomeric SnP was not measured. A low degree of dye degradation was observed in the absence of both visible light and a photocatalyst (Figure ). Therefore, both visible light irradiation and a photocatalyst are required to degrade MO. Time-dependent measurements of the absorption spectra of MO in aqueous solution in the presence of the T3 nanorods under visible light irradiation are shown in Figure S9. The degradation of MO dye was monitored by measuring the decay of the absorbance peak at 461 nm with respect to irradiation time. The photocatalytic efficiency of the different nanomaterials for the degradation of MO is expressed using the degradation efficiency, (C0–C)/C0, where C0 is the initial concentration of MO and C is the concentration at time t. As shown in Figure , of the photocatalysts, T3 exhibited the best efficiency for the degradation of MO within 100 min of visible-light irradiation, and the observed degradation ratios reached 94, 86, 76, and 50% for T3, T4, T2, and ZnL, respectively. These results suggest that the efficiency of the photocatalytic degradation of MO is affected by the photocatalyst morphology. In general, different morphologies can affect the catalytic activity because they provide different sizes, surface areas, numbers of active sites, and numbers of defects. Especially for T3, the one-dimensional rodlike shape probably promotes the electron-transfer process, providing more efficient photocatalytic performance. Similarly, the nanofibers of ZnTPyP (meso-tetra(4-pyridyl)porphyrinato)Zn(II) exhibited distinct photocatalytic performance for the photodegradation of rhodamine B (RhB) molecules compared to the nanosphere, and it was revealed that the formation of J-aggregates in the nanofibers facilitates the electron-transfer process, resulting in distinct photocatalytic performance.[54] In addition, the ZnTPyP nanowire showed enhanced photocatalytic performance for the MO degradation and H2 production compared to other morphologies such as nanospheres and nanooctahedra.[55] On the other hand, the surface area of the catalyst had little influence on the performance. Each catalyst exhibited a low surface area, for example, the BET surface area of the best-performing T3 was measured to be 30 m2/g by using N2 adsorption isotherms at 77 K (Figure S10). For comparison, commercial TiO2 (anatase) and ZnO nanoparticles have surface areas of approximately 186 and 88 m2/g, respectively.

Figure 4

Photocatalytic degradation of MO in aqueous solution (pH 7, 298 K) under visible-light irradiation.

The reaction kinetics for the decomposition of MO were also investigated. For this, we used the pseudo-first-order model as expressed by ln(C0/C) = kt, which is commonly used for photocatalytic degradation experiments if the initial concentration of the pollutant or dye is low. In this equation, k is the pseudo-first-order rate constant. Using the data in Figure , we plotted ln(C0/C) versus t and obtained the rates of MO decomposition by these photocatalysts (Figure S11). The first-order rate constant for the degradation of MO by T3 (0.0251 min–1) is higher than those of T4 (0.018 min–1), T2 (0.015 min–1), and ZnL (0.006 min–1). Notably, these rates are comparable to or even better than those previously reported for inorganic photocatalysts such as Ag2O (0.007 min–1),[56] ZnO (0.005 min–1),[57] and TiO2 (0.015 min–1).[58]

Consecutive degradation reactions were conducted to verify the stability of the photocatalysts in aqueous solution. The catalysts were separated by centrifugation after each reaction, washed twice with distilled water, and dried under vacuum (323 K). Subsequently, the catalysts were used for repeated dye-degradation cycles. Importantly, the photocatalyst efficiency remained intact even after 10 consecutive cycles. As depicted in Figure , 94% MO was degraded within 100 min for the first run, and the photocatalytic performance decreased slightly, marked by 90% MO degradation, after 10 consecutive cycles. The high stability of T3 was confirmed by FE-SEM. No significant change was observed in the morphology of T3 after 10 cycles as shown in Figure .

Figure 5

Consecutive cycling photodegradation curves of MO with T3 in aqueous solution.

Figure 6

FE-SEM images of T3 before and after 10 consecutive cycles.

Next, we optimized the photocatalyst/MO ratio during the degradation experiments. For this purpose, different concentrations of MO (5, 10, 15, and 20 mg L–1) with a constant amount of photocatalyst (10 mg) were used at 298 K and pH 7. It was found that 20 mg L–1 of MO dye solution (150 mL) with 10 mg of photocatalyst yielded the best performance (Figure S12). We further examined the influence of pH on the photocatalytic degradation of MO dye. In Figure S13, it can be seen that the degradation performance is comparable to over 80% in the range of pH 5–9. However, the performance declined significantly under highly acidic or basic conditions. Inhibited coordination bonding under these harsh conditions can cause the nanostructures to collapse, thus degrading the performance.

Several possible mechanisms for the photodegradation of pollutant dyes in aqueous solutions using similar photocatalysts have been reported.[30,31] When a photocatalyst is subjected to visible-light irradiation in the presence of MO, the nanomaterial absorbs light, and the valence band (VB) electrons are promoted to the conduction band (CB) across the band gap. This causes the formation of electron–hole pairs (e–/h+) pairs at the surface of the photocatalyst. Strong intermolecular π–π interactions intensify the electron delocalization over the nanoaggregates and minimize the recombination energy of the excited electrons. Then, the photogenerated holes (h+) react with H2O to produce hydroxyl radicals (•OH), and the electrons react with the dissolved O2 to produce superoxide radical anions (O2–•). These highly reactive photogenerated superoxide radical anions and hydroxyl radicals react with MO, resulting in its degradation.

The mechanism consists of several steps for the porphyrin triad (T), as shown in eqs –5.

The reactive species involved in the photocatalytic reactions in the presence of T3 were revealed by radical trapping analysis.[59] As shown in Figure S14, para-benzoquinone (BQ) and tert-butyl alcohol (tBuOH) were used to capture superoxide radical anions (O2–•) and hydroxyl radicals (•OH), respectively, to investigate the effect of different reactive species on the photocatalytic degradation of MO by T3. The results show that the MO degradation rate in the presence of T3 is affected by both tBuOH and BQ, suggesting that hydroxyl and superoxide radicals are the major reactive species contributing to the catalytic oxidation of MO. For further details on the detection of all intermediate species (singlet or triplet oxygen species, hydroxyl radicals, or superoxide radical anions), readers are directed to our previous reports.[30,31]

To obtain details of the degradation products of MO after photocatalysis, we analyzed the reaction mixture obtained after 1 h by ESI-MS (Figure S15). Compared to the MS pattern of MO, that of the degraded mixture suggests that MO was degraded by visible light photocatalysis.[60] As suggested by the data in Figure S15, three different degradation mechanisms can be proposed. First, the successive demethylation and addition of hydroxyl groups to MO leads to the formation of two intermediates with m/z values of 290 and 320. Second, the cleavage of the azo double bond leads to the generation of intermediates having m/z values of 156 and 172. Finally, the breakdown of the sulfonate group from the ring leads to the formation of intermediates with an m/z value of 79. Other possible intermediates having m/z values of 100, 108, 136, 240, 260, 276, 306, and 322 are also shown in Figure . The formed intermediates were further fragmented into small molecules and finally mineralized into CO2 and H2O. Additionally, the total organic carbon (TOC) value was measured to evaluate the removal of MO dye by photocatalysts.[61] The TOC removal percentage achieved using T3 was only 77%.

Figure 7

Possible intermediates of the reaction between MO and T3 under visible-light irradiation in aqueous solution.

Conclusions

Three isomeric Zn(II)–Sn(IV)–Zn(II) porphyrin triads were synthesized from the reaction of ZnL with different pyridyl-N-containing SnP complexes. The intramolecular cooperative metal–ligand coordination of the 3-pyridyl nitrogen in SnP with axial ZnL, followed by the π–π interactions between adjacent porphyrin triads, is the determining factor affecting the nanomorphology of T3. Owing to geometrical constraints, this type of interaction is not possible for T2, for which only π–π interactions affect the self-assembly process. In contrast, in the case of T4, intermolecular interactions and π–π interactions are responsible for the nanostructure. Moreover, the morphology-dependent photocatalytic degradation of MO dye under visible light irradiation was observed for these photocatalysts, and nanorod-shaped T3 performed better than nanosphere T2 and nanoflake T4. We consequently demonstrated that tuning the coordination mode enabled us to develop nanostructured photocatalysts with high efficiency for the degradation of MO. Several metal oxides (Ag2O,[56] ZnO,[57] and TiO2[58]), a bismuth-based photocatalyst,[62] and a graphitic carbon nitride (g-C3N4)-based photocatalyst[63] have been used in advanced oxidation processes (AOPs) for the degradation of pollutant dyes. In particular, TiO2 has been used as a common photocatalyst because of its low cost, high stability, low toxicity, and high catalytic efficiency toward the degradation of organic pollutants in water, but its large band gap (∼3.2 eV) limits the utilization of sunlight owing to the low photocatalytic efficiency in the visible region. Our strategy is much more expandable in the development of visible-light-active photocatalysts because the morphology-controlled fabrication can be readily achieved by molecular engineering. Therefore, precisely engineered porphyrin-based materials could be an alternative to the photocatalysts for the treatment of dye-containing wastewater in the near future.