Abstract

Herein, we synthesized gold-coated ZnO/Fe3O4 nanocomposites. Initially, we prepared Fe3O4 magnetic nanoparticles based on the co-precipitation of Fe3+ and Fe2+ under aqueous ammonia as a precipitating agent. Thereafter, the ZnO/Fe3O4 composite was prepared by dispersing the synthesized magnetic nanoparticles into an alkaline zinc nitrate solution. After calcination of the precipitate, the formed ZnO/Fe3O4 composites were coated with gold nanostructures by dispersing the composites in auric acid/ethylene glycol solution in a water bath. The synthesized Au@ZnO/Fe3O4 hybrid material was able to catalyze the reduction of 4-nitrophenol (4-NP) to 4-aminophenol (4-AP). We demonstrate that this catalytic activity can be exploited for the detection of Hg2+ ions in a cosmetic product. In the presence of Hg2+ ions, the catalytic activity of Au@ZnO/Fe3O4 was greatly suppressed. This novel finding underlies a straightforward, sensitive, and highly selective detection probe for Hg2+. The material exhibited excellent analytical performance as marked by the very low limit of detection (LOD) of 2.34 nM, which was well below acceptable levels of 4.99 μM for mercury in cosmetics as set by the US Food and Drug Administration (FDA), and within the linear dynamic ranges of 0–10 nM. High recoveries ranging from 96.5 to 100.3% accompanied by excellent selectivities toward Hg2+ over potentially interfering species were obtained.

1. Introduction

The intentional adulteration of skin-lightening cosmetics with mercury is not only fraudulent but also dangerous. Long-term use of such skin-lightening cosmetics results in respiratory, cardiovascular, gastrointestinal, hepatic, renal impairment, and reproductive effects. Moreover, mercury is associated with elevated white blood cell count, changes in autoimmune response, drop in vision, and genotoxic effects [1]. The world’s leading health regulatory institutions restrict the use of mercury in cosmetics. The EU Cosmetics Regulation (EC) 1223/2009 prohibits mercury-containing cosmetic products in the EU market except for the use of thiomersal in eye products where a maximum concentration of 0.007% of mercury is permitted in eye makeup, mascara, and cleansing products [2]. The US Food and Drug Administration (FDA) under title 21 CFR 700.13 limits the amount of mercury in cosmetic products to a trace amount unavoidable under good manufacturing practice, that is, 1 ppm, except in those products intended for use around the eye, which it limits to 65 ppm [3]. Despite this, consumers continue to have access to mercury-based cosmetic products and sporadic cases as well as outbreaks of mercury poisoning still occur [4]. To ensure that these restrictions are adhered to, it is necessary to employ highly sensitive and reliable analytical methods for mercury detection.

Currently, numerous approaches based on electrochemical and optical sensors [57], organic chromophores or fluorophores [813], conjugated polymers [14], oligonucleotides [1517], DNAase [18, 19], and proteins [2022] have been established for the detection of Hg2+ ions. Unfortunately, most of these methods have the disadvantages of low selectivity, low sensitivity, and complex synthetic process [23]. Furthermore, sophisticated instrumentation including, atomic emission spectrometry (AES), atomic absorption spectrometry (AAS), and inductively coupled plasma mass spectrometry (ICP-MS) are being used for the detection of Hg2+ [24, 25]. Although these methods have the advantages of high accuracy and sensitivity, they are limited to expensive sophisticated instruments that require trained professional technicians and long operation times which limit their wide use in ordinary institutions and routine detection [26]. Therefore, there is an urgent need to establish a simple, fast, cheap, and sensitive methods for the detection of Hg2+.

As a convenient detection method for Hg2+, colorimetric strategies have received great attention due to the advantages of easy operation, rapid visual detection, and signal visibility without the aid of any advanced instrument, and they are applicable for onsite and high throughput analysis [27]. Colorimetric-sensing tools for Hg2+ based on the aggregation or antiaggregation of noble metal nanoparticles have been developed [22, 2830]. However, most of these assays are insensitive. Recently, various colorimetric methods based on enzymatic catalysis for the detection of Hg2+ ions have been proposed successfully due to their exclusive properties, such as simplicity, high specificity, ultrasensitivity, and naked eye detection [26]. Unfortunately, natural enzymes are easily affected by environmental factors; they are easily denatured, expensive, and time-consuming to prepare [31].

To avoid these disadvantages of natural enzymes, nanoenzymes; nanomaterials with intrinsic enzyme-like characteristics and with high catalytic efficiency and operational stability have been used for the detection of Hg2+ ions successfully [23, 24, 26, 3142]. Compared with natural enzymes, nanoenzymes are more stable, easy to prepare, have high yield, low cost, and tunable structure and catalytic activity [43]. Nanozymes can catalyze various types of reactions including reduction reactions, oxidation reactions, and polymerization reactions. For example, modified metal nanoparticles can catalyze the reduction of methyl orange (MO), methyl blue (MB), and 4-nitrophenol (4-NP), the oxidation of 2,2′-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) diammonium salt (ABTS) and 3,3,5,5-tetramethylbenzidine (TMB), and the polymerization of alkyl silanes [44]. It has been reported that mercury can enhance or inhibit the enzyme-like activity of noble metal nanoparticles (NPs) which enables a label-free colorimetric assay for Hg [4548]. In this regard, many materials based on the mercury-enhanced/suppressed mimetic activity of noble NPs have been prepared. Although these materials can selectively and sensitively detect Hg, they possess several drawbacks including deactivation caused by the aggregations due to high surface energy and van der Waals forces and poor recovery and reusability [27].

To overcome such challenges, hybrid nanoenzymes (those formed from the combination of two or more different materials) have been developed. The individual components bring distinct properties and the combined properties of the hybrid material may offer new or enhanced properties, which afford versatility in the design and tailoring of multifunctional devices [49]. Among the plethora of possible components of such hybrids, gold nanoparticles (AuNPs) in particular have attracted considerable interest in the detection of Hg2+ where one or more of their quantum size effects, high surface energies, and high surface-to-volume ratios can be advantageously exploited [50]. The properties of AuNPs can be readily modified because they depend on the AuNP size and shape as well as on the stabilizing ligands present on the surfaces, and all of these can be systematically modified. AuNPs possess good stability, they can be synthesized from a variety of well-established procedures, and they display a rich surface functionalization chemistry with a broad variety of organic matrices and coatings (organic compound/AuNP hybrids) and flexibility that has been exploited in the drive to afford novel hybrid multifunctional materials [51, 52]. ZnO has drawn significant research attention for decades due to its earth abundance, environmental friendliness, and excellent structural, electrical, and optical properties [53]. However, its applicability as a catalyst is limited due to the rapid recombination of electron-hole pairs caused by its wide and direct bandwidth [54]. Thus, its catalytic performance can be improved by exploiting the surface plasmon resonance (SPR) effect of AuNPs, coupled with their strong scattering and absorption of incident light [55, 56]. Au nanostructures can produce electrons induced by the surface plasmon resonance (SPR) and inject them into the conduction band of contiguous ZnO nanostructures, thus improving their catalytic activity [57]. The magnetic property of the hybrid structure afforded by the Fe3O4 allows a straightforward and effective collection of the reaction mixture with an external magnet [58]. To the best of our knowledge, there are only two reports of the fabrication of Au@ZnO/Fe3O4 hybrid material in literature [54, 59]. In one instance, Quang and Vu [59] prepared Au/ZnO/Fe3O4 composites by a mixing method using polyvinyl alcohol (PVA) as a binder and in another Cervantes-Macıas et al. [54] doped AuNPs onto ZnO–Fe3O4 NPs by adding HAuCl4 into a suspension of ZnO–Fe3O4, using sodium citrate as a reducing agent. These were applied in the photodegradation of tartrazine and oxidation of thioanisole, respectively.

Hence, in this work, we successfully employed, for the first time, polyol synthesis to coat ZnO/Fe3O4 with AuNPs. In addition, the synthesized Au@ZnO/Fe3O4 hybrid material exhibited the ability to catalyze the reduction of 4-nitrophenol (4-NP) to 4-aminophenol (4-AP). Interestingly, we found that the catalytic activity of Au@ZnO/Fe3O4 for this reaction can be significantly inhibited by Hg2+. When Hg2+ is added to the reduction reaction system, the catalytic activity of AuNPs decreases because of the inhibition of the formation of gold-hydride species, leading to a long reaction time from 4-NP to 4-AP. We take advantage of this property to develop a novel probe for the rapid, selective, and sensitive colorimetric detection trace levels of Hg2+ in a skin-lightening cosmetic product.

2. Materials and Methods

2.1. Chemicals

Iron (II) sulphate heptahydrate (FeSO4.7H2O), iron (III) chloride hexahydrate (FeCl3.6H2O), zinc nitrate (Zn(NO3)2), auric acid (HAuCl4), ethylene glycol (EG), sodium borohydride (NaBH4), mercury nitrate (Hg(NO3)2), sodium nitrate (NaNO3), copper nitrate (Cu(NO3)2), calcium nitrate (Ca(NO3)2), aluminum nitrate (Al(NO3)3), cesium nitrate (Cs(NO3)2), lead nitrate (Pb(NO3)2), and cobalt nitrate (Co(NO3)2) were purchased from Sigma Aldrich. Silver nitrate (Ag (NO3)2), sodium hydroxide (NaOH), and ammonia (NH4) were purchased from Merck. All reagents were of analytical grade and were used without additional purification. All analyte solutions were prepared in Millipore ultrapure double-deionized water.

2.2. Instrumentation

The surface morphology of the prepared nanocomposite was studied using a high-resolution Carl Zeiss Gemini SEM500 electron microscope operating at an electron accelerating voltage of 6.0 kV and working distances of 4.1 and 4.3 mm (Carl Zeiss, Germany). X-ray diffraction (XRD) patterns were obtained on a D2 Phaser XRD-300 W powder diffractometer (Bruker, AXS GmbH, Karlsruhe, Germany) for a 2θ range of 10–90 at a scan rate of 0.05 degree/s using Cu Kα radiation at 40 kV and 100 mA. Raman spectra were measured by a Horiba LabRam Raman spectrometer equipped with a 785-nm laser line as an excitation source (Jobin–Yvon Technology, France) and a CCD detector. A Thermo-Fisher UV-visible absorbance spectrophotometer was used to collect UV-vis spectra using 1 cm glass cuvettes as sample cells. An Axis Supra X-ray Photoelectron Spectrometer (XPS) (Kratos Analytical, Manchester, UK) was used to assess the surface chemical composition of the Au@ZnO/Fe3O4 and confirm the formation of an amalgam when Hg2+ ions encountered Au.

2.3. Preparation of Fe3O4 Nanospheres

The Fe3O4 magnetic nanoparticles were prepared based on the co-precipitation of Fe3+ and Fe2+ under aqueous ammonia as a precipitating agent, with some modifications [60]. In brief, FeSO4.7H2O (3.48 g) and FeCl3.6H2O (3.38 g) were each dissolved in 100 mL of deionized water and mixed under vigorous stirring. Aqueous ammonia was dropped slowly into the solution. It was observed that the solution became black. The black mixture was then aged at 85°C for 2 h. After that, the product prepared was allowed to cool overnight. Finally, the black product was washed with deionized water and dried in the oven for 6 h at 70°C.

2.4. Preparation of the ZnO/Fe3O4 Composite

In preparing the ZnO/Fe3O4 composite, we employed a previously developed method with minor modifications [61]. In brief, 0.15 g of Fe3O4 was dispersed in 200 mL of 45 mM zinc nitrate solution. The mixture was then stirred to homogenize. Under vigorous stirring, to this mixture, at 70°C, 90 mM of NaOH solution was added slowly within 1 h. The reaction mixture was kept under stirring at 70°C for 2 h. Then, it was left at room temperature overnight to complete precipitation. Thereafter, the composite was cleaned repeatedly with deionized water and then separated from the water by using a ferromagnet and placed in an oven at a temperature of 70°C for 6 h to dry. After drying, the particles were annealed at a temperature of 350°C for 1 h.

2.5. Synthesis of Gold Nanoparticles on ZnO/Fe3O4 Composite

A simple but novel thermal method was employed to deposit gold nanoparticles on ZnO/Fe3O4 composites. In brief, 0.4 g of ZnO/Fe3O4 composite was dispersed in 5 mM of 50 mL HAuCl4 solution (dissolved in ethylene glycol) and sonicated for 3 min to disperse the particles uniformly. The mixture was then placed in a water bath at 95°C for 20 min. The resulting gold-coated ZnO/Fe3O4 composites were then cleaned with deionized water and placed in the oven at 70°C for 3 h.

2.6. Catalytic Performance and Colorimetric Detection of Hg2+

Initially, the catalytic activity of the Au@ZnO/Fe3O4 composites was investigated by using the reduction of 4-NP to 4-AP as a model reaction in the presence of freshly prepared NaBH4 as the reducing agent. In a typical reaction, to a glass cuvette containing 3 mg of Au@ZnO/Fe3O4, 100 μL of deionized water and freshly prepared 0.1 mM of 1 mL 4-NP were added. To this reaction mixture, 1 mL of freshly prepared 100 mM NaBH4 was added. Thereafter, the progress of the reduction reaction was monitored by measuring the absorbance of the reaction mixture using a UV/visible spectrophotometer in the range of 250–600 nm.

The detection of Hg2+ was assessed by studying the catalytic activity of Au@ZnO/Fe3O4 composite in the presence of varying Hg2+ concentrations. Here, 3 mg of Au@ZnO/Fe3O4 was incubated in 100 mL solution with different concentrations. Then, 1 mL of 4-NP solution (0.1 mM) was added to the solution and 1 mL of freshly prepared NaBH4 solution (100 mM) was quickly added to the mixture. Then, the colorimetric reaction was allowed to proceed for 10 min. The resultant solution was scanned by a UV/Vis spectrophotometer with a wavelength range of 250–800 nm, and λmax of 400 nm was used as the detection signal.

2.7. Sensitivity and Selectivity of the Au@ZnO/Fe3O4 Composite

To estimate the sensitivity and quantitative range of the Au@ZnO/Fe3O4 for the determination of Hg2+ ions, a range of varying concentrations of Hg2+ ions (0–20 nM) were added to the sensing solution, and absorbance variations at 400 nm were recorded. The selectivity of the AuNP@ ZnO/Fe3O4 system for Hg2+ ions detection was evaluated by monitoring the UV absorption response to interference metal ions including Ag+, Na+, Cu2+, Mg2+, Ca2+, Al3+, Cs2+, Zn2+, Pb2+, Co2+, and anion NO3 at a concentration of 20 μM.

3. Results and Discussion

3.1. Fabrication of Fe3O4, ZnO/Fe3O4, and Au@ZnO/Fe3O4

The fabrication of Fe3O4, ZnO/Fe3O4, and Au@ZnO/Fe3O4 was described in the sections mentioned above. Based on equation (1), the formation of a solid-state Fe3O4, as the only solid-state product meant that the black precipitate that was described above have formed on the addition of the precipitating agent, aqueous ammonia could be attributed to the Fe3O4 solid product in equation (1).

The separation of the black precipitate, Fe3O4 solid particles, by a ferromagnet confirmed the paramagnetism of Fe3O4 as has been reported elsewhere [62]. Thereafter, when the precipitate was dispersed in an alkaline zinc nitrate solution, a grey solid powder was formed. Based on equation (2) [63], the reaction involved a double exchange that occurred between Zn(NO3)2 and NaOH, resulting in the formation Zn(OH)2, as the only solid-state product.

After annealing the grey solid powder, a brown solid powder was formed, and according to equation (3) [63], during the annealing process, zinc hydroxide had decomposed into zinc oxide.

The brown solid powder formed during annealing was attracted to a ferromagnet; this in addition confirmed that, a composite of ZnO and Fe3O4 which we named, ZnO/Fe3O4 had been formed and the paramagnetism of the Fe3O4 was retained. Finally, a brick-red composite was formed when the brown composite was reacted with auric acid in ethylene glycol at slightly high temperatures, via the polyol synthesis method. The brick-red composite which we named Au@ZnO/Fe3O4 was also attracted to a ferromagnet, confirming that the paramagnetism of the Fe3O4 was still retained. The fabrication of Fe3O4, ZnO/Fe3O4, and Au@ZnO/Fe3O4 was confirmed by characterizing for size, shape, morphology, crystallinity, and surface chemistry.

3.2. SEM Characterization

The sizes, shapes, and morphologies of the prepared black precipitate attributed to Fe3O4, brown powder attributed to ZnO/Fe3O4, and brick-red fine particles attributed to Au@ZnO/Fe3O4 composite were obtained using SEM. The SEM micrograph for Fe3O4 displayed spherical particles that were mostly agglomerated (see Figure 1(a)), with an average individual particle diameter of 26 nm measured using image J software (see Figure 1(d)). The nanosize diameter confirmed the nanoparticulate nature of the prepared Fe3O4 black precipitate. Our findings were supported by the work of Ba-Abbad et al. [64], where they reported that using ammonia as a precipitating agent resulted in spherically shaped Fe3O4 nanoparticles (NP), whereas using NaOH resulted in cubic-shaped NPs. They concluded that the formation of Fe3O4 NPs relied on the type of precipitating agent. Independently, Mazhani et al. [65] and Petcharoen and Sirivat [66] also synthesized Fe3O4 via the coprecipitation method using ammonia as a precipitating agent and similarly, their SEM imaging showed that they had produced spherical shaped Fe3O4 NPs. The referenced research works supported the nanosize and spherical shape of Fe3O4 particles that we also obtained, with ammonia as the precipitating agent.

Figure 1(b) displays the SEM micrograph of the brown ZnO/Fe3O4 composite. It displays irregular shapes of what looks like broken monoliths. Figure 1(c) is a SEM micrograph of the brick-red fine solid particles of Au@ZnO/Fe3O4 composite. According to Holade et al. [67], it displays scarcely dispersed nanoparticles of different sizes and shapes coated on top of the ZnO/Fe3O4 monoliths.

The reduction of auric acid by ethylene glycol takes place simultaneously with the oxidation of ethylene glycol, with several possible oxidation pathways and products (aldehydes, carboxylic acids, or CO2) forming, depending on the temperature, pH, and chemical environment. As already established, the shapes of the nanoparticles formed are determined by the rate at which metal atoms add to metal clusters (fluxional assemblies) to form seeds (nonfluxional), the structures of these initial seeds (twinned vs single crystal), the rate at which metal atoms add to seed faces, and the binding selectivity of capping agents. The rates at which metal atoms form and assemble are sensitive to reduction kinetics, which in turn are dependent on the reductant [68]. Therefore, we believe that the combination of moderately high reaction temperature, the solution pH, and ZnO/Fe3O4 and Au@ZnO/Fe3O4 possibly acting as catalysts in the oxidation of ethylene glycol may have resulted in the formation of different oxidation products or all the possible oxidation products as reductants of Au (III) to Au(0), hence the formation of different sizes and shapes of Au (0).

3.3. XRD Characterization

Analysis through X-ray diffraction spectrophotometry was carried out to confirm the crystalline nature of the fabricated particles. The X-ray diffraction (XRD) pattern marked Fe3O4 shown in Figure 2 exhibited 2θ peaks at 30.20°, 35.45°, 43.25°, 53.55°, 56.78°, and 62.73° marked with indices at (220), (211), (400), (422), (511), and (400) (JCPDS no. 19-0629), respectively. The peak positions and intensities matched well with those reported before for the inverse spinel structure of magnetite [69, 70] and have been reported to show the face-centered cubic structure of Fe3O4 [71]. Crystallographically, magnetite takes on a cubic inverse spinel structure due to the fact that the oxide ions form a face-centered cubic, with the iron (II) and iron (III) cations occupying one-eighth of the tetrahedral and half of the octahedral holes, respectively [72].

Furthermore, on the XRD pattern marked ZnO/Fe3O4 in Figure 2, the peaks at 2θ 31.77°, 34.42°, 36.25°, 47.54°, 56.60°, 62.86°, 66.38°, 67.96°, and 69.10° corresponding to (100), (002), (101), (102), (110), (103), (200), (112), and (201) (JCPDS no. 36-1451), respectively, were observed. The positions and relative intensities of the diffraction peaks matched well with those reported before for hexagonal zincite-type of structure [73, 74] which indicated the formation of a highly crystalline ZnO structure [58]. In this pattern, in addition to the peaks assigned to ZnO, peaks at 2θ 30.20°, 35.45°, 43.25°, 53.55°, 56.78°, and 62.73° marked with indices at (220), (211), (400), (422), (511), and (400) previously attributed to cubic spinel structured magnetite were observed in the nanocomposites, revealing that the phases of Fe3O4 nanocrystals did not change; furthermore, this confirmed the formation of ZnO/Fe3O4 composites.

Finally, on the XRD pattern marked Au@ ZnO/Fe3O4, five strong peaks that were not observed for ZnO/Fe3O4 composite appeared. The peaks were observed at diffraction angles of 38.20°, 44.40°, 64.60°,77.60°, and 81.76° with the indices (111), (200), (220), (311), and (222) (JCPDS no. 04-0784), respectively. The peaks have been attributed to the Au crystal with a cubic phase [75, 76]. The peaks attributed to the ZnO structure still appeared on this pattern. The XRD results clearly indicated the successful synthesis of Au@ZnO/Fe3O4 hybrid materials.

3.4. Raman Characterization

The formation of gold nanoparticles on the surface of ZnO/Fe3O4 was confirmed by surface-enhanced Raman spectroscopy (SERS). Gold nanoparticles can produce ultrasensitive surface-enhanced Raman scattering (SERS) signals owing to their plasmonic properties. Collective oscillation of free electrons on the surface of gold nanoparticles enhances the Raman signal of molecules which are adsorbed or are in close proximity to gold nanoparticles [77]. The mechanism relevant to SERS is the surface enhancement of the electromagnetic field at the interface between the SERS active substrate (e.g., gold) and a molecule of interest through resonant excitation of surface plasmons of the metal [78]. This mechanism was exploited to confirm the formation of AuNPs on the surface of ZnO/Fe3O4. SERS effect of the nanostructures was investigated using 4-nitrothiophenol (4-NTP) as a probe molecule because of its strong Raman activity and strong affinity to chemisorption on gold surfaces [79]. The synthesized Fe3O4, ZnO/Fe3O4, and Au@ ZnO/Fe3O4 particles were soaked in 1 mM of 4-NTP for 1 h to allow for chemisorption of the 4-NTP on the surface of the nanoparticles and then air dried before SERS analysis. The SERS spectrum in Figure 3 representing the measured conditions for Au@ ZnO/Fe3O4 showed bands at 1567 cm−1, 1336 cm−1, 1110 cm−1, 1075 cm−1, 854 cm−1, and 722 cm−1 for 4-nitrothiophenol. These could be attributed to the benzene ring C = C stretching, symmetrical NO2 stretching, C-H bending vibration, in-plane C-S stretching modes, C-H wagging vibrations, and C-S wagging vibrations, respectively [65, 80, 81]. On the other hand, Raman scattering was not observed under the measured conditions for ZnO/Fe3O4 and Fe3O4 as shown in Figure 3(ii) and (iii), respectively, indicating that the particles did not exhibit any SERS effect. This served as confirmation that gold nanoparticles were formed on the surface of ZnO/Fe3O4.

3.5. XPS Characterization

XPS analysis was carried out to assess the surface chemical composition of the Au@ZnO/Fe3O4. The XPS spectrum in Figure 4(a) shows overlapping peaks. Upon deconvolution, it resolved into four peaks at binding energies of 92 eV, 89 eV, 88 eV, and 85 eV. These peaks were attributed to Zn 3p1/2, Zn 3p3/2, Au 4f5/2, and Au 4f7/2, respectively [82]. The spectrum obtained confirmed the formation of AuNP on the surface of ZnO NP.

The XPS spectrum in Figure 4(b) shows two binding-energy peaks at 101.64 eV and 105.64 eV that were formed by the addition of Hg2+. The peaks were ascribed to Hg 4f5/2 and Hg 4f7/2, respectively [27], which revealed that Hg0 was formed on the surface of Au@ZnO/Fe3O4, confirming the formation on an Au-Hg amalgam on the surface of Au@ZnO/Fe3O4 as has been reported by Long et al. [46, 83] in the past. In the same study, Long et al. further reported that the existence of Hg0 on the surface of AuNPs had great influence on the features of the surface of AuNPs, including adsorption on the surface of AuNPs.

3.6. Construction of the Nanoprobe and Detection of Hg2+

The excellent qualities of the different components that constituted the prepared catalytic hybrid material, Au@ZnO/Fe3O4 allowed versatility in the design and tailoring of a multifunctional material with distinct properties that were brought about by the combined effect of the traits of the individual components. For instance, the prepared Fe3O4 nanoparticles provided the magnetic separation property; however, the ferromagnetic NPs aggregated too easily due to their magnetic dipole-dipole interactions and decreased the efficiency of the catalytic property of NPs [27]. Also, according to Chen et al. [84], it is notably difficult to directly attach AuNPs to the surfaces of the Fe3O4 owing to the dissimilar nature of the two species’ surfaces. To deal with this challenge, we successfully used ZnO NPs to facilitate the formation of AuNP on the nanomaterial because of the closeness of intrinsic absorption and emission bands of ZnO with the surface plasmon peak of Au at resonant coupling.

In this work, the reduction of 4-nitrophenol to 4-aminophenol in the absence and presence of the prepared Au@ZnO/Fe3O4 as a catalyst as well as in the absence and presence of Hg2+ as the analyte of interest, was monitored via UV-visible spectroscopy which showed differences in absorbance intensities at lambda max = 400 nm in the different reaction systems (see Figure 5). Figure 5(i) shows that the recorded absorbance for the reduction of 4-nitrophenol by excess NaBH4 solution in the absence of the catalyst was highest and was found to be lowest in the presence of the catalyst (see Figure 5(ii)). In the presence of the catalyst, AuNPs acted as electron relay centers and initiated the shifting of electrons from the donor, BH4 to the acceptor, 4-nitrophenol. The reacting molecules (BH4 ion and 4-nitrophenol) were simultaneously adsorbed onto the surface of the NPs. As a result, electrons were transferred from the BH4 ion to 4-nitrophenol through the NPs, causing the fast reduction of 4-nitrophenol to 4-aminophenol [85] and the lowest absorbance was recorded. Without the catalyst, the reduction of 4-nitrophenol by excess NaBH4 was very slow hence very little 4-nitrophenol was converted to 4-aminophenol and the absorbance was the highest.

However, when Hg2+ was added to the system (see Figure 5(iii)), the absorbance at 400 nm was reduced but only to a lesser extent compared to when Hg2+ was not added to the system (Figure 5(ii)), depicting a decrease in the catalytic activity of Au@ZnO/Fe3O4. This reduction in catalytic activity was because of the inhibition caused by the formation of gold-hydride species by the resulting Au-Hg amalgam formed, leading to the slow reduction of 4-nitrophenol to 4-aminophenol. The formation of Au-Hg amalgam which causes the slow reduction of 4-nitrophenol to 4-aminophenol has also been reported before by He and Zheng [86]. It should be noted that the observations in Figure 5 were accompanied by color changes. The reaction system attributed to the reduction of 4-nitrophenol by excess NaBH4 in the absence of the catalyst represented by the spectrum in Figure 5(i) was bright yellow in color and then when the catalyst was added to the system represented by the spectrum in Figure 5(ii) the solution changed to colourless and finally, when mercury was added to the system represented by the spectrum in Figure 5(iii) the solution only changed to pale yellow. The color changes, with further investigations, may be employed in the colorimetric screening of Hg2+.

3.7. Optimization of Parameters on the Detection of Hg2+

The incubation time for the colorimetric screening of Hg2+ via the catalytic reduction of 4-nitrophenol to 4-aminophenol was studied in the absence and presence of mercury in order to identify the best time at which the reaction occurs. The results are shown in Figure 6.

The spectra in Figures 6(a) and 6(b) show a decrease in the absorption signal at lambda max = 400 nm with increasing time both in the absence and presence of Hg2+, respectively, signaling the reduction of 4-nitrophenol to 4-aminophenol with time. This reaction was slower in the presence of Hg2+ than in its absence, due to an increase in the quantity of Au-Hg amalgam formed with the increase in time. The experiment was stopped at 20 mins since there was no discernible absorbance peak at lambda max = 400 nm in the absence of Hg2+. From the spectra in Figure 6(a), it was concluded that a reasonable time at which this experiment could be carried out showing discernible peaks was at any time between 8 and 12 min. Thus, for our work, 10 mins was chosen as the optimal time. It should be noted that there were corresponding color changes with time that were also observed.

The effect of the concentration of 4-nitrophenol on the absorbance signal was also studied. The results are shown in Figure 7.

As depicted by the spectra in Figures 7(a) and 7(b), the absorbance increased with increasing concentration of 4-nitrophenol at optimal 10 min incubation time, both in the absence and presence of Hg2+, respectively. In both cases, there were more 4-nitrophenol molecules in the system available to be reduced, hence the increase in absorbance with increasing concentrations of 4-nitrophenol. There were discernible peaks at all concentration levels, therefore any of the concentrations of 4-nitrophenol studied was suitable for the colorimetric screening of Hg2+ to give distinct color changes. Thus, for this work 0.05 mM of 4-nitrophenol was used.

3.8. Sensitivity and Selectivity in the Determination of Hg2+

Figure 8(a) shows the absorbance spectra of Hg2+ from 0 to 20 nM. It was found that the absorbance at 400 nm increased with increasing concentration of Hg2+ until 16 nM then began to decrease. Figure 8(b) presents the absorbance signal difference as a function of Hg2+ concentration. The absorbance signal was linearly related to Hg2+ concentration in the range of 0–10 nM. The LOD of the colorimetric sensor is calculated by using the following equation: LOD = 3σ/k, where k is the slope of the calibration curve and σ is the standard deviation of the blank solution of 4-NP absorbances recorded. It was calculated to be as low as 2.34 nM. Compared to other colorimetric methods employing Au-based nanoenzymes (Table 1), the currently proposed method fairs well. Not only does it offer high sensitivity and fast response, but it is also convenient and stable.

In addition to sensitivity, selectivity is also a critical parameter to evaluate the performance of an assay. The unique affinity of Hg2+ towards Au provided excellent selectivity towards detecting other potentially interfering metal ions. Figures 9(a) and 9(b) show that other metal ions including Ag+, Na+, Cu2+, Mg2+, Ca2+, Cs2+, Zn2+, Pb2+, Co2+, Al3+, K+, and Ni2+ do not have the ability to suppress the catalytic reduction of 4-NP by Au@ZnO/Fe3O4, instead they seem to enhance it at double the concentration of Hg2+.

3.9. Determination of Mercury-Spiked Real Samples

The performance of the proposed colorimetric method was evaluated for the determination of mercury in a cosmetic product purchased locally. The results are summarized in Table 2. As shown in Table 2, the recoveries are in the range of 96.5–100.3% and the RSD is in the range of 4.3–13.8% for n = 3. This indicates that the proposed method has a high accuracy and reliability for the analysis of Hg2+ in real samples.

4. Conclusion

In summary, we have successfully developed a high-performance colorimetric probe for the detection of Hg2+ ions based on the inhibition of the catalytic activity of Au@ZnO/Fe3O4 on the borohydride reduction of 4-NP. The probe is simple, rapid, sensitive, and selective. It can sensitively detect Hg2+ with LODs of 2.34 nM. The colorimetric change is obvious and allows one to recognize the color change at low Hg2+ concentrations. Its excellent selectivity is owed to the gold amalgam process that occurs specifically between Au and Hg. The magnetic core allows for convenient separation during analysis. Moreover, the probe has shown great potential for the accurate detection of Hg2+ in a skin-lightening cosmetic product.

Data Availability

The data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Acknowledgments

This work was supported by the Office of Research, Development, and Innovation (ORDI), Botswana International University of Science and Technology.