In Situ Synthesis of CuO/Cu₂O Nanoparticle-Coating Nanoporous Alumina Membranes with Photocatalytic Activity under Visible Light Radiation

Abstract

We report the synthesis and characterization of CuO/Cu₂O film supported on nanoporous alumina membranes (NAMs) and the photocatalytic properties in the removal of the organic pollutant methyl orange (MO). For this purpose, transparent nanostructured membranes were fabricated and sequentially modified with APTS ((3-aminopropyl)-trimethoxysilane) and EDTAD (ethylenediaminetetraacetic dianhydride) to form a highly functionalized surface with high density of carboxyl groups, which easily complex with copper cations. The Cu²⁺-modified membranes were annealed in a chemical vapor deposition (CVD) furnace to form a well-ordered nanostructured coating of CuO/Cu₂O with photocatalytic properties. These modifications were followed by characterization with FT-IR and UV–visible spectra, X-ray photoelectron spectroscopy (XPS), field emission scanning electron microscopy (FE-SEM), and energy dispersive X-ray spectroscopy (EDS). Finally, the photocatalytic performance of the NAM-CuO/Cu₂O nanostructured membranes was tested in the aqueous removal of MO dye as a model reaction system. Our results revealed 50% photocatalytic removal of MO under continuous light irradiation for 2 h. The procedure presented in this work provides an adequate approach for the fabrication of nanostructured devices with photocatalytic properties for the degradation of organic compounds.

1. Introduction

Nanostructured alumina membranes with ordered honeycomb-like pore arrangement exhibit outstanding physical, chemical, mechanical, optical, and electrical properties. These promising materials have unique qualities, such as long-range ordered nanostructures, large surface area, high porosity, controllable pore size, high chemical stability, and relatively low cost of fabrication. Importantly, the highly porous surface of nanoporous alumina membranes (NAMs) can be chemically activated, modified, or functionalized with diverse types of molecules (e.g., organosilanes, amino acids, and organic ligands) [1,2,3,4]. Advantageously, this strategy could allow us to tune the physicochemical and wettability properties of the NAM surface, which is highly desirable for diverse technological uses, such as templates for biosensors, filters, magnetic recording media, photonic crystals, plasmonic devices, catalysis, and biocatalysis [5,6,7,8,9,10].

Thermal treatment is a viable strategy that is useful for the surface activation of NAM surfaces. This results in a notable increment of the density of hydroxyl groups, which can act as active sites to bind different molecules. The surface functionalization of NAM with selective, specific, and sensitive chelating molecules results in a versatile substrate or template with uses such as retention and chemical reduction of metal ions to form stabilized metal nanoparticles. Considering this, ethylenediaminetetraacetic acid (EDTA), a hexadentate ligand containing both carboxylate and amine groups, is a well-known metal chelating agent that has received much attention due to its versatility in the surface modification of various types of materials. EDTA is a highly useful reagent that is widely used in industrial cleaning, the nuclear industry, pharmaceuticals, and the manufacturing of textile leather, rubber, and paper [11,12,13,14,15].

Among numerous cations that are suitable to be adsorbed by EDTA, copper derivates have attracted much attention because of their numerous applications and advantages. Specifically, copper oxides (CuO and Cu₂O) are p-type semiconducting compound members of the family of copper compounds employed in diverse applications, such as batteries, biosensors, solar energy conversion, and photocatalysis. Interestingly, CuO and Cu₂O nanoparticles exhibit a relatively narrow band gap (~1.2 and 2.0–2.2 eV, respectively), resulting in exploitable visible-light-active behavior that is very useful for energy conversion, photoconductive functionalities, and photocatalytic properties [16,17,18,19].

Based on this, the combination of a supporting high-porosity matrix, a chelating agent, and semiconducting nanoparticles should lead to relevant physicochemical properties and catalytic activity. This would constitute an attractive approach for obtaining highly functional multiphasic nanomaterial for applications in photocatalysis. We seek to contribute to the studies and applications of inorganic materials (e.g., copper oxide, alumina nanoparticles) and organic materials (e.g., ethylenediaminetetraacetic dianhydride (EDTAD)) with high availability, stability in various media, and reasonable cost for applications of technological importance (e.g., dye pollutant degradation).

Accordingly, we report the fabrication, physicochemical characterization, and photocatalytic behavior of CuO/Cu₂O nanoparticles supported on NAMs. For this purpose, NAM was chemically treated with aminopropylsilane and then with EDTAD agent (denoted as NAM-EDTA). Later, copper oxide nanoparticles were obtained and immobilized on the NAM-EDTA support by immersion in an aqueous solution of a copper salt precursor for a determined time and subsequent thermal annealing treatment. Notably, the obtained support of combined NAM and CuO/Cu₂O, referred to as NAM-CuO/Cu₂O, showed reasonable photocatalytic activity for the degradation of methyl orange (MO) dye, which was recognized as an acute toxin.

2. Experimental

2.1. Chemicals

Pure aluminum foil (99.997%) was obtained from Alfa Aesar^®. Experiments were performed using 3-aminopropyl-trimethoxysilane (APTS, 97.0%; Sigma-Aldrich, St. Louis, MO, USA), EDTAD (97.5%; Sigma-Aldrich, St. Louis, MO, USA), copper(II) acetate (Cu(CH₃COO)₂, 99.9%; Sigma-Aldrich, St. Louis, United, methyl orange (85.0%; Sigma-Aldrich, St. Louis, MO, USA), dimethylsulfoxide (for Analysis, PanReac, Chicago, United States) 30% hydrogen peroxide (H₂O₂, Emsure^®, Merck, Darmstadt, Germany), ethanol (p.a., Merck, Darmstadt, Germany), perchloric acid (HClO₄, 67%, Merck, Darmstadt, Germany), and Milli-Q water.

2.2. Membrane Fabrication

NAMs were fabricated by a two-step anodization process [20]. Briefly, high-purity aluminum (99.999%) was first electropolished in a solution of HClO₄/EtOH at 1:4 for 5 min at 20 V. The first anodization was performed for 780 min at 40 V using a 0.3 M oxalic acid solution at 5 °C. Afterwards, removal of alumina was performed in a solution with 6.0 wt.% phosphoric acid and 1.8 wt.% chromic acid, which was kept at 60 °C for 3 h. The second step of anodization was performed for 643 min using 0.3 M oxalic acid solution at 5 °C. Removal of aluminum was performed in 20% HCl w/w + 0.1 M CuCl₂ solution for 20 min at room temperature. Finally, etching was carried out for 50 min using 5 wt.% phosphoric acid solution at 25 °C.

2.3. Membrane Modification

Previously fabricated anodic alumina membranes were used to obtain photocatalytic supports. Firstly, the membranes were treated with 30% H₂O₂ at 90 °C for 1 h and rinsed with Milli-Q water. Later, the samples were dried in an oven at 120 °C for 3 h and then stored in a desiccator for further use. This procedure was performed in order to increase hydroxyl groups on the surface of pores [21,22]. Next, the membranes were treated with 2% APTS ethanol solution and 160 μL of 1M acetic acid aqueous solution for 1 h at 50 °C. Subsequently, the modified membrane was rinsed with abundant ethanol. After this treatment, the samples were dried under N₂ stream and kept in a desiccator overnight [23]. Later, the membranes were treated with a solution of 0.1 M EDTAD in DMSO for 4 h at 75 °C and rinsed 3 times with abundant DMSO. Hydrolysis reaction was carried out by immersing the membranes into a 1 M acetic acid solution overnight. In this stage, NAM modified with EDTAD (NAM-EDTAD) was produced.

To obtain immobilized copper (II) and copper (I) oxide nanoparticles, NAM-EDTAD membranes were immersed in 1M Cu (CH₃COO)₂ water solution at 70 °C for 1 h. After this, the resultant samples were thermally annealed by conventional chemical vapor deposition under a controlled flow rate (200 sccm) of air at 400 °C for 30 min to produce NAM-CuO/Cu₂O membranes. By weight difference between the NAM-EDTAD and NAM-CuO/Cu₂O membranes, it was determined that the percentage of copper oxides in the membrane was approximately 2.7%. Note that this content was estimated in triplicate to ensure reproducibility.

2.4. Membrane Characterization

FT-IR and UV spectra of the membranes were recorded on Thermo Nicolet 6700 (Thermo Fischer Scientific, Waltham, MA, USA) and a HR4000CG-UV-NIR spectrophotometer (Ocean Optics, Orlando, FL, USA), respectively. Morphological structures of NAM-CuO/Cu₂O membranes were analyzed using an FEI QUANTA FEG 250 field emission scanning electron microscope (FE-SEM, FEI, Hillsboro, OR, USA) equipped with an Oxford X-MAX50 energy dispersive X-ray spectroscopy (EDS) analyzer (Oxford Instruments plc, Oxford, UK). To obtain better resolution in SEM micrographs, samples were sputtered with 4 nm of gold. Electron micrographs were recorded in high-vacuum mode under an acceleration voltage of 20 kV. Finally, X-ray photoelectron spectroscopy (XPS) analysis was performed in a monochromator SPECS XPS (Specs Group, Berlin, Germany) with a PHOBIOS detector using an Al kA (1486.71 eV, Specs Group, Berlin, Germany) anode as an X-ray source.

2.5. Photocatalytic Activity Tests

The photocatalytic activity of obtained membranes was studied under both dark conditions and visible light irradiation. For photocatalytic activity tests, MO was selected as a target pollutant. The detailed procedure is as follows. MO solution (20 mg/L) was used for the photodegradation test. Initially, 2 mL of MO solution was transferred to a quartz spectrophotometer cell and a piece of NAM-CuO/Cu₂O membrane (0.5 × 0.5 cm) was added, with constant air bubbling at ambient temperature. Afterwards, the MO concentration was monitored by a UV–vis spectrophotometer (Ocean Optic HR4000CG-UV-NIR) by simply removing the membrane of the solution.

A solar simulator with a 1500-W Xenon lamp (Newport Corp., Irvine, CA, USA) was utilized as a visible light source. The lamp was located 13 cm above the glass beaker. Similar experiments were carried out for MO degradation under dark conditions. The MO degradation efficiency (N) was calculated as follows:

N = (A₀ − A)/A₀

where A₀ is the original absorbance of MO solution at 463 nm and A is the absorbance of MO at a different degradation time.

Additionally, the possible recyclability of NAM-CuO/Cu₂O system was assessed using these materials in three consecutive runs of the MO photodegradation. After each cycle, the membrane systems were washed with Milli-Q water before the next use. Note that all experiments were performed at least in triplicate.

3. Results and Discussion

Optical images of the initial nanoporous alumina membrane and the NAM-EDTAD and NAM-CuO/Cu₂O supports are shown in Figure 1. Figure 2 shows a schematic representation of the chemical modification of anodic alumina membranes to obtain NAM-EDTAD systems and the subsequent process for the incorporation of CuO/Cu₂O nanoparticles (i.e., the generation of NAM-CuO/Cu₂O). Notably, the previous chemical activation (i.e., the formation of hydroxyl groups) in the membrane surface would play a pivotal role in the proposed modification of porous alumina [23].

Hydroxylated nanoporous alumina membranes were treated with APTS to generate a propylamine surface, which can be modified by amine substitution using EDTAD to form a corresponding amide bond [11,12,13]. Thus, by employing this strategy, a versatile and relatively robust material exhibiting adequate photocatalytic performance could be obtained. Later, membranes were annealed under air atmosphere to yield CuO/Cu₂O nanoparticles [24,25].

Figure 3 shows the normalized FT-IR spectra of the surfaces of the NAM-APTS, NAM-APTS-EDTAD, and NAM-CuO/Cu₂O systems. The previous treatment of NAM with H₂O₂ allowed us to generate an enriched surface of activated –OH groups. Thus, a covalent bond between hydroxyl groups and silane derivatives is conveniently formed on the surface of porous alumina [23]. The presence of Si—O stretching was observed at approximately 1125 cm⁻¹ [26]. Additionally, the signals at approximately 1300 and 1200 cm⁻¹ could be ascribed to C—N stretching and C—C bending vibrations, respectively (Figure 3a).

The NAM-EDTAD spectrum shows a sharper band at 1200 cm⁻¹ compared with that of silanol stretching, which indicates the successful incorporation of EDTAD (Figure 3b). Interestingly, a broad band centered at ~900 cm⁻¹ was also detected and corresponded to O—H bending vibrations (carboxylic acid group) [27]. In the case of the NAM-CuO/Cu₂O system spectrum, most of the signals previously observed in the NAM-EDTAD spectrum were clearly suppressed (Figure 3c). These spectral differences provide pertinent evidence of the thermal degradation of organic compounds and the formation of copper oxide nanoparticles on the surface of the NAM support. The initial coordination of copper ions by EDTAD would be a crucial step for the generation of a CuO/Cu₂O nanoparticle coating.

Figure 4 shows the UV–vis spectra of unmodified anodic alumina membranes along with the respective chemical modifications. For NAM, the typical absorption band was detected at 250 nm and ascribed to the electron photoexcitation from the valence band to the conduction band of the oxidized aluminum surface. In the case of the membrane modified with EDTA, the respective spectra showed slight differences compared with pristine NAM. This behavior can be explained by the absence of absorption in the visible range of the chemical groups present in the modified membranes. Subsequently, membranes modified with EDTA were treated with copper acetate, so the complexation of copper ions by the EDTA-enriched surface should be expected. Indeed, a characteristic absorption maximum corresponding to the complex of copper ions and EDTA was observed at approximately 300 nm [28]. The membrane surface also changed from transparent to pale blue, which helps to confirm the formation of the Cu-EDTA complex.

After annealing treatment, the surface changed from pale blue to slightly yellow-brownish, which is evidenced by the UV–visible spectrum. The absorption band located below 400 nm is broader compared with that obtained with EDTAD modification prior to the annealing process. These changes in both the color and the UV spectra can be attributed to the formation of copper oxide nanoparticles (NAM-CuO/Cu₂O) [29,30].

The optical band gap of the NAM-CuO/Cu₂O support was estimated using the Tauc equation as follows:

$${\left(\mathsf{\alpha }\mathrm{hv}\right)}^{\mathrm{n}}=\mathrm{A}\left(\mathrm{h}\mathsf{\nu }-{\mathrm{E}}_{\mathrm{g}}\right)$$

where $\mathsf{\alpha }$ is the absorption coefficient, hν is the photon energy, E_g is the optical band gap of the analyzed sample, A is a constant, and n = 2 for a direct allowed band gap transition. The optical band gap value of the films was estimated by extrapolation of the straight line of the plot of (αhν)² versus photon energy. The obtained Tauc plot (Figure 4b) shows a band gap of 2.7 eV for the modified membrane. This band gap is slightly higher than previous reports [31], which can be attributed to the combined formation of CuO and Cu₂O, as well as the possible influence of the membrane material Al₂O₃ [32].

Figure 5 shows SEM micrographs for the neat Al₂O₃ membrane and the NAM-CuO/Cu₂O samples. The image of the neat alumina membrane helps to confirm the high porosity ascribed to H₂O₂ treatment. The mean pore diameter was estimated to be 65.2 ± 2.2 nm, which is a typical value for this type of material. Interestingly, the 45°-tilted SEM image of the neat membrane showed unpolluted inner channels with highly regular and straight pores.

The EDX analysis (Figure 5c) displayed the chemical composition of this membrane and denoted the expected presence of aluminum and oxygen. In the case of NAM-CuO/Cu₂O support (Figure 5a), the pore diameter decreased to approximately 43.5 ± 2.3 nm, exhibiting a highly homogeneous and regular surface with clearly distinguishable open pores. The SEM micrograph at the angle of 45° (Figure 5b) denotes a change in the roughness, morphology, and regularity of inner walls throughout channels compared with the neat membrane. Likewise, the modified membrane exhibits open pores, confirming a significantly regular distribution of the CuO/Cu₂O film on the membrane surface.

Notably, the EDX analysis of treated membranes showed the appearance of copper signals. These results help corroborate the successful modification of the membrane surface with copper oxide nanoparticles. The modification of the membrane surface with ATPS as a chemically active molecule, and then with EDTA as a chelating agent, is a relatively low-cost, innovative, and viable strategy to capture and retain copper ions in order to obtain copper oxide nanoparticles after annealing treatment.

XPS analysis was performed to determine the chemical states of Cu species in the nanoporous alumina membranes. High-resolution spectra in the Cu 2p region of membranes from before and after annealing are shown in Figure 6. In both samples, it is possible to decompose the obtained spectrum in two doublet contributions, Cu²⁺ 2p (932.2 and 952.0 eV) and Cu⁺ 2p (934.2 and 954 eV), which are in good agreement with the literature [33,34]. Before annealing, it is possible to observe a higher intensity of the doublet related to Cu⁺ contribution. This could be due to the partial reduction of Cu²⁺ species, which is promoted by the presence of EDTAD on the NAM surface [33,35]. As expected, after the annealing process, the contribution related to Cu²⁺ increased in intensity and there was also a strong increment in the satellite signals to about 942 and 962 eV, which are also related to the increment of Cu²⁺ ions [36]. The presence of different oxidation states after annealing suggests the formation of mixed CuO/Cu₂O oxides forming a type 2 semiconductor heterojunction on the surface of the membrane [35,37].

To confirm the photocatalytic properties of the NAM-Cu/Cu₂O systems, degradation of MO was performed in the presence of NAM-CuO/Cu₂O in aqueous medium for 2 h (Figure 7a,b). Approximately 25% and 50% MO was removed in the dark and under visible light irradiation, respectively. This reflects the predominant effect of the light irradiation on the process of degradation of the organic molecule of MO. Considering this, the proposed mechanism would address the adsorption of the MO onto the membrane surface and subsequent degradation.

Under light irradiation conditions, the reaction rate (Figure 7c) is significantly faster than in the dark conditions, which is attributed to the adequate photoresponse ability of CuO/Cu₂O nanoparticles [38]. In dark conditions, the removal of MO can be explained by the generation of Cu⁺, which is fundamental in the MO removal process in dark conditions [39].

Importantly, control experiments in the absence of NAM-CuO/Cu₂O support, using neat Al₂O₃ membranes, were conducted in order to confirm the photocatalytic activity of the modified membrane. For all the mentioned cases, no reaction was identified as no significant change at 460 nm was detected for MO in the UV–vis absorption spectrum. Therefore, the copper (I and II) oxide nanoparticles on the Al₂O₃ membranes play a pivotal role in the photocatalytic process of MO degradation.

Another important aspect to consider is the recyclability of the NAM-CuO/Cu₂O systems, which was assessed by applying the same membrane in three consecutive catalytic cycles. After each cycle, the membrane was washed and rinsed with a large amount of Milli-Q water. The photoactivity of the NAM-CuO/Cu₂O systems was estimated as follows:

$$Photoactivity \left(\%\right)=\frac{A_x}{A_0};x=1, 2 and 3$$

where A_x is the absorbance after each cycle and A₀ is the absorbance before the first cycle (note that these absorbances were monitored at 370 nm, which corresponds to CuO/Cu₂O absorption). The photoactivity decreased slightly after each catalytic cycle (in the third cycle it decreased to approximately 95%). Despite the moderate photoactivity of NAM-CuO/Cu₂O against MO degradation, this material showed that it is potentially recyclable for more than one use cycle.

4. Conclusions

The procedure presented allowed the in situ generation of CuO/Cu₂O nanoparticles on the surface and in the channels of NAMs. Chemically activated anodic alumina membranes were able to be modified with EDTAD as a chelating agent to promote the complexation of copper ions in aqueous medium. After this process, photoactive copper oxide nanoparticles were produced by annealing treatment. The presence of Cu²⁺ and Cu⁺ species (i.e., the combined presence of CuO and Cu₂O nanoparticles) was adequately elucidated by XPS technique. The in situ formation of copper oxide nanoparticles in the anodic alumina membranes generated significant changes in the chemical, optical, morphological, and photocatalytic properties in MO photodegradation.