Gold Nanoparticle-Coated ZrO2-Nanofiber Surface as a SERS ... - MDPI

nanomaterials Article

Gold Nanoparticle-Coated ZrO2-Nanofiber Surface as a SERS-Active Substrate for Trace Detection of Pesticide Residue Han Lee 1 , Jiunn-Der Liao 1,2, *, Kundan Sivashanmugan 1 , Bernard Haochih Liu 1 Chih-Chien Chen 1 , Guo Dung Chen 3 and Yung-Der Juang 4 1

2 3 4

*

ID

, Wei-en Fu 3 ,

Department of Materials Science and Engineering, National Cheng Kung University, 1 University Road, Tainan 701, Taiwan; [email protected] (H.L.); [email protected] (K.S.); [email protected] (B.H.L.); [email protected] (C.-C.C.) Medical Device Innovation Center, National Cheng Kung University, 1 University Road, Tainan 701, Taiwan Center for Measurement Standards, Industrial Technology Research Institute, No. 321, Kuang Fu Road, Sec. 2, Hsinchu 300, Taiwan; [email protected] (W.-e.F.); [email protected] (G.D.C.) Department of Materials Science, National University of Tainan, Tainan 700, Taiwan; [email protected] Correspondence: [email protected]; Tel.: +886-6-2757575 (ext. 62971); Fax: +886-6-2346290

Received: 4 May 2018; Accepted: 1 June 2018; Published: 3 June 2018

Abstract: Trace detection of common pesticide residue is necessary to assure safety of fruit and vegetables, given that the potential health risk to consumers is attributed to the contamination of the sources. A simple, rapid and effective means of finding the residue is however required for household purposes. In recent years, the technique in association with surface-enhanced Raman scattering (SERS) has been well developed in particular for trace detection of target molecules. Herein, gold nanoparticles (Au NPs) were integrated with sol-gel spin-coated Zirconia nanofibers (ZrO2 NFs) as a chemically stable substrate and used for SERS application. The morphologies of Au NPs/ZrO2 NFs were adjusted by the precursor concentrations (_X, X = 0.05–0.5 M) and the effect of SERS on Au NPs/ZrO2 NFs_X was evaluated by different Raman laser wavelengths using rhodamine 6G as the probe molecule at low concentrations. The target pesticides, phosmet (P1), carbaryl (C1), permethrin (P2) and cypermethrin (C2) were thereafter tested and analyzed. Au NPs/ZrO2 NFs_0.3 exhibited an enhancement factor of 2.1 × 107 , which could detect P1, C1, P2 and C2 at the concentrations down to 10−8 , 10−7 , 10−7 and 10−6 M, respectively. High selectivity to the organophosphates was also found. As the pesticides were dip-coated on an apple and then measured on the diluted juice containing sliced apple peels, the characteristic peaks of each pesticide could be clearly identified. It is thus promising to use NPs/ZrO2 NFs_0.3 as a novel SERS-active substrate for trace detection of pesticide residue upon, for example, fruits or vegetables. Keywords: surface-enhanced Raman scattering; pesticide residue; gold nanoparticles; Zirconia nanofibers

1. Introduction Agricultural production and safety, food industries, as well as the consumers are of great concern to governments [1]. The use of pesticides in agricultural products aims to increase the yield, as well as improve the quality of crops. Pesticide residue in food and in the environment provokes great public concern since it could pose potential health risks to not only the consumers but also the earth. Various analytical methods, including gas chromatography [2] and high-performance liquid chromatography [3], have been applied for trace detection of pesticides [4,5]. However,

Nanomaterials 2018, 8, 402; doi:10.3390/nano8060402

www.mdpi.com/journal/nanomaterials

Nanomaterials 2018, 8, 402

2 of 11

these methods are time consuming, labor intensive, and usually require complicated procedures for sample preparation. Thus, research has been conducted to develop advanced detection techniques instead of traditional methods to provide rapid, nondestructive food quality and safety evaluation and analysis for the industry [1]. As an emerging technology, surface-enhanced Raman scattering (SERS) techniques are becoming increasingly widespread and accessible for accurate and specific identification of chemical or microbiological contaminants in foodstuffs [6]. Over the past decades, the technique of SERS has been well developed as a sensitive and selective method, using, for example, nanostructured surfaces [7]. When target species with Raman-active modes are adsorbed on a nanostructured surface of noble metal(s), particular peak enhancement can be found in the Raman spectrum. Target species could be identified by verifying the specific fingerprints, which comprise information about vibrational modes in compounds. Advances in nanofabrication techniques have assisted the as-prepared substrates capable of lowering the detection limit of sensing target molecules or biomolecules. Particular enhancement of Raman-active modes is mainly attributed to two mechanisms: a chemical [8] and an electromagnetic (EM) effect [9]. For the former, as the target molecules are chemically adsorbed upon a SERS-active substrate, Raman-active modes in molecules can be enhanced in different degrees, depending on the distances of the modes with respect to the substrate surface [10]. For the latter, a strong EM may be generated from the morphologies of nanostructures. As the target species is trapped/adhered to the proximity of a metal nanostructure (usually Au, Ag or Cu), Raman intensity is enhanced due to the amplification of the EM field resonance, not only by the size and shape of the nanostructure, but also the interactions among Raman laser wavelength, the target species, and the substrate surface in the microenvironment [11]. Researchers have investigated the formation of hot spots, which are small regions of a highly enhanced EM field that leads to high enhancement factor (EF). These regions can be observed or even predicted using computer simulation and modeling to describe the quantum effect on subnanometer gaps [12]. Usually, a noble metal surface with the structure of high roughness [7], edges [13], hollow cavities [14], and so on, is advantageous to produce EM as well as surface plasmon [15]. Strategies for fabricating nanostructures can be categorized as, for example, top-down [4], bottom-up [16], combination [17], and template-assisted techniques [18]. In addition, for example, Zirconia (ZrO2 ), especially the thin-film type, is one of the promising common materials used for a base SERS-active substrate. [19] By manipulating the structure of ZrO2 into, for example, a nanofiber, a large surface area is gained. As the ZrO2 nanofibers’ (ZrO2 /NFs’) structure is combined with gold nanoparticles (Au NPs) to gain the EM effect between NPs or intra-Au NPs [13], the hybrid nanostructure may increase EM at the metal–semiconductor junction and generate a strong local EM field at the interface. [20] In addition, Au NPs embedded in ZrO2 NFs are competent to prevent the electron-hole pairs from recombination; thus a significant enhancement of SERS is anticipated. In this work, ZrO2 NFs are therefore made by a spin-coated sol-gel method and formed as templates; subsequently, Au NPs are deposited upon ZrO2 NFs. The as-designed Au NPs embedded upon ZrO2 NFs as the substrate (denoted as Au NPs/ZrO2 NFs) is proposed. The structure of NFs is designed to increase the surface areas for detecting the target species, while the embedded Au NPs upon the as-formed ZrO2 NFs are anticipated to gain the additional effect of SERS. Thereafter, different pesticides at low concentrations are tested and measured to interpret their competences for trace detection of pesticide residue. 2. Experimental Section 2.1. Fabrication of ZrO2 NFs and NPs/ZrO2 NFs Zirconium tetrachloride (ZrCl4 , 98%, Acros Organics, Geel, Belgium) was used as the precursor for the synthesis of ZrO2 . The precursor solutions with various concentrations were prepared by dissolving appropriate amounts of ZrCl4 in 10 mL isopropanol (99.8%, Panreac AppliChem Barcelona, Spain).


3 of 11

Nanomaterials 2018, 8, x FOR PEER REVIEW

3 of 11

After vigorous stirring, precursor solutions were stored in sealed glassware aged at room Barcelona, Spain). After the vigorous stirring, the precursor solutions were stored inand sealed glassware temperature for 24 h. 100 µL of the precursor solution was spin-coated onto a 2 cm × 2 cm silicon and aged at room temperature for 24 h. 100 μL of the precursor solution was spin-coated onto awafer 2 cm ◦ C and 70%, respectively. at22500 rpm forwafer 30 s at ambient temperature and relative humidityand of 25relative × cm silicon atan 2500 rpm for 30 s at an ambient temperature humidity of 25 °C Prior70%, to coating, siliconPrior wafers were pre-cleaned withwere hydrochloric acid (37%, Panreac AppliChem, and respectively. to coating, silicon wafers pre-cleaned with hydrochloric acid (37%, Barcelona, Spain) and then ethanol (99.9%, Merck KGaA, Darmstadt, Germany) to remove the organic Panreac AppliChem, Barcelona, Spain) and then ethanol (99.9%, Merck KGaA, Darmstadt, Germany) ◦ contaminants the surface. The as-prepared samples were heated atsamples 100 C for 10 min to evaporate to remove the on organic contaminants on the surface. The as-prepared were heated at 100 °C ◦ C for 3 h in the air for densification. The as-formed ZrO NFs the solvent and then calcined at 500 2 The for 10 min to evaporate the solvent and then calcined at 500 °C for 3 h in the air for densification. are distinguished as are ZrOdistinguished X is 2the concentration precursor in M. The fabrication 2 NFs_X, where as-formed ZrO2 NFs as ZrO NFs_X, where X isofthe concentration of precursor in procedures of ZrO NFs are simply illustrated in Figure 1a. 2 M. The fabrication procedures of ZrO2 NFs are simply illustrated in Figure 1a.

Figure 1. (a) Steps for preparing Au NPs/ZrO2 NFs: 1. mixing isopropanol with ZrCl4; 2. forming Figure 1. (a) Steps for preparing Au NPs/ZrO2 NFs: 1. mixing isopropanol with ZrCl4 ; 2. forming ZrO2 thin film by spin-coating method; 3. forming ZrO2 NFs by removing solvents; (b) steps for ZrO2 thin film by spin-coating method; 3. forming ZrO2 NFs by removing solvents; (b) steps for depositing Au NPs upon ZrO2 NFs by e-beam evaporator; (c) SERS mechanism based on Au NPs depositing Au NPs upon ZrO2 NFs by e-beam evaporator; (c) SERS mechanism based on Au NPs deposited uponrandom randomZrO ZrONFs; 2 NFs; (d) SERS signals from Au NPs deposited upon ZrO2 NFs, no deposited upon (d) SERS signals from Au NPs deposited upon ZrO2 NFs, no Raman 2 Raman signal from the surface of Au NPs Si (100) ZrO2 NFs without the integration of Au signal from the surface of Au NPs upon Si upon (100) and ZrOand 2 NFs without the integration of Au NPs. NPs.

Au NPs NPs were were then onto ZrO NFs by by using using an an electron electron beam beam evaporator evaporator (VT1-10CE, (VT1-10CE, Au then deposited deposited onto ZrO22 NFs ULVAC Inc., Chigasaki, Japan) with a thickness of about 1.5 nm and at a rate of 0.1 Å/s. The ULVAC Inc., Chigasaki, Japan) with a thickness of about 1.5 nm and at a rate of 0.1deposition Å /s. The −6 torr. The SEM images was operated under ultra-high vacuum conditions maintained below 7 × 10 −6 deposition was operated under ultra-high vacuum conditions maintained below 7 × 10 torr. The after deposition shown inare Supporting 1 (a) with a magnification of 5 × 105 , and SEM images afterare deposition shown inData Supporting Data 1 (a) with a magnification of 5Supporting × 105, and 6 . The results show that the nano-sized Au NP gold Data 1 (b) with a higher magnification of 10 6 Supporting Data 1 (b) with a higher magnification of 10 . The results show that the nano-sized Au particles arranged the nanofiber In addition, theaddition, Energy Dispersive Spectroscopy NP gold closely particles closely on arranged on thesurface. nanofiber surface. In the Energy Dispersive (EDS) mapping image of the cross-sectioned sample AuNPs/ZrO NFs_0.3 is shown in the Supporting 2 AuNPs/ZrO2NFs_0.3 is Spectroscopy (EDS) mapping image of the cross-sectioned sample shown in Data 2. The particle-size distribution of the nano-Au NPs with the range from 30 to 45 nm is also the Supporting Data 2. The particle-size distribution of the nano-Au NPs with the range fromshown 30 to in the Data in which a narrow represents a uniform size of Au NPs. 45 nmSupporting is also shown in3,the Supporting Datadistribution 3, in whichcurve a narrow distribution curve represents a The as-formed as Auwere NPs/ZrO where X is the concentration of 2 NFs_X, as uniform size ofsamples Au NPs.were The distinguished as-formed samples distinguished Au NPs/ZrO 2 NFs_X, where precursor in M. The fabrication of AuinNPs/ZrO simply illustrated in Figure 1b. As illustrated 2 NFs_X is of X is the concentration of precursor M. The fabrication Au NPs/ZrO 2 NFs_X is simply illustrated in Figure 1c with Au NPs deposited upon ZrO NFs, intra-Au NP interactions “1” in NP the in Figure 1b. As illustrated in Figure 1c with2 Au NPs deposited upon ZrO2 (marked NFs, intra-Au figure) tend to be much stronger than electron transfer at the junction. In case of Au NPs distributed interactions (marked “1” in the figure) tend to be much stronger than electron transfer at the junction. over ZrO NFs (marked “2” inover the figure), there exists a“2” recombination electron-hole pairs, leading In case of2Au NPs distributed ZrO2 NFs (marked in the figure),ofthere exists a recombination to a significant improvement of SERS properties. As appropriate laser wavelength is applied, surface of electron-hole pairs, leading to a significant improvement of SERS properties. As appropriate laser wavelength is applied, surface plasmon resonance of Au NPs and hot spots between Au NPs may presumably occur and contribute to the effect of SERS.


4 of 11

plasmon resonance of Au NPs and hot spots between Au NPs may presumably occur and contribute to the effect of SERS. 2.2. Structural and Morphological Characterization The compositions of ZrO2 NFs and NPs/ZrO2 NFs were analyzed by an X-ray diffractometer (XRD, MiniFlex II, Rigaku, Japan) using CuKα radiation with scanning angles ranging from 20◦ to 61.5◦ . The obtained XRD patterns were compared with JCPDS card No. 89-6976 [21] and 65-1022 [22]. The photo-images of the surfaces of ZrO2 NFs and NPs/ZrO2 NFs were taken by a high-resolution thermal field emission scanning electron microscope (FE-SEM, JSM-7000, JEOL, Tokyo, Japan), which was operated at an accelerating voltage of 10 kV. All the samples subsequent to FE-SEM were platinum-coated in advance. The dimensions of ZrO2 NFs and Au NPs were thereafter presented and determined by FE-SEM photo-images and software Image J (National Institutes of Health, USA). 2.3. Enhancement Evaluation for the Effect of SERS The test probe molecule, rhodamine 6G (R6G), was used as the referenced target species. R6G was diluted in aqueous solution to a concentration of 10−3 M as the standard solution. A quantity of 5 µL R6G standard solution was then placed on each substrate and dried at room temperature for subsequent analysis. A Raman spectrum was obtained by using Raman spectrometer with a confocal microscope (Renishaw, United Kingdom). He-Ne and diode lasers with excitation wavelengths of 633 and 785 nm were respectively applied. An air-cooled CCD was used as the detector and the incident power was about 3 mW. The samples were scanned with an exposure time of 10 s over an area of 1 µm × 1 µm (the size of the laser spot was 1 µm, as shown in Figure 1d), using a 50× objective. The SERS spectra were averaged from 10 consecutive measurements on different samples. All Raman spectra were normalized by using the peak fit software. 2.4. Trace Detection of Pesticide Residue and Those on Apples The optimal SERS-active substrates were thereafter examined using four types of pesticides, namely phosmet (P1), carbaryl (C1), permethrin (P2) and cypermethrin (C2). An aqueous solution of each pesticide was diluted to concentrations ranging from 10−2 to 10−10 M. To verify the effect of SERS with respect to organophosphates, mixed pesticides were prepared to the concentration of 10−3 M. A quantity of 5 µL from single and diluted pesticide and mixed pesticides were respectively placed on the substrates and dried at room temperature for subsequent analyses. The test apples were purchased from a local market and then immersed in a solution of each pesticide in 10−2 M. The apple containing pesticide on the surface was thereafter cut into pieces. Each pesticide on the outer surface of the sliced apple was extracted by soaking a 1 cm × 1 cm apple peel in 500 µL ethanol. The as-formed mixture was then vigorously shaken for 10 min and subjected to 4500 rpm of centrifugation for 5 min. A final product with a quantity of 5 µL was placed on the respective substrates and dried at room temperature for subsequent analyses. 3. Results and Discussion 3.1. The Quality of ZrO2 NFs and Au NPs/ZrO2 NFs In Figure 2, XRD patterns of ZrO2 NFs obtained from the various precursor concentrations are shown. Characteristic peaks of tetragonal zirconia, namely (011), (002), (110), (112), (020), (013) and (121) resulted. Additional peaks from monoclinic zirconia, including (−111), (111) and (311), were also observed. The result indicates that ZrO2 is successfully prepared by a spin-coated sol-gel method. Nevertheless, the intensity of the characteristic peaks is lowered with the decreased concentration of precursor solution as the coverage of the generated ZrO2 is still insufficient.

Nanomaterials 2018, 8, 402 Nanomaterials 2018, 8, x FOR PEER REVIEW

5 of 11 5 of 11

Figure 2. 2. XRD XRD patterns patterns of of ZrO ZrO2 NFs with different ZrCl4 concentrations. Figure 2 NFs with different ZrCl4 concentrations.

In Figure 3, FE–SEM photo-images of Au NPs/ZrO2 NFs were obtained from various precursor In Figure 3, FE–SEM photo-images of Au NPs/ZrO NFs were obtained from various precursor concentrations, corresponding to their XRD patterns2 (ZrO2 NFs) shown in Figure 2. At the concentrations, corresponding to their XRD patterns (ZrO2 NFs) shown in Figure 2. At the beginning, beginning, the base ZrO2 NFs were significantly generated, as the precursor concentrations were the base ZrO2 NFs were significantly generated, as the precursor concentrations were higher than higher than 0.2 M (i.e., ZrO2 NFs_0.2, _0.3, _0.4, and _0.5); the diameter of ZrO 2 NFs was measured 0.2 M (i.e., ZrO2 NFs_0.2, _0.3, _0.4, and _0.5); the diameter of ZrO2 NFs was measured around around 35 to 50 nm. With the decreased precursor concentrations, that is, ZrO2 NFs_0.05 and _0.1, 35 to 50 nm. With the decreased precursor concentrations, that is, ZrO2 NFs_0.05 and _0.1, the surface the surface of ZrO2 NFs was broad and flat, as shown in Figure 3a,b. From Figure 3c–f, as the of ZrO2 NFs was broad and flat, as shown in Figure 3a,b. From Figure 3c–f, as the precursor precursor concentration was increased, the hydrolytic rate was decreased; the morphology of the concentration was increased, the hydrolytic rate was decreased; the morphology of the base ZrO base ZrO2 NFs became agglomerated and connected with thinner nanofibers. As the increase of2 NFs became agglomerated and connected with thinner nanofibers. As the increase of surface areas surface areas for the subsequent Au NP deposition is of significance, ZrO2 NFs with distinct size for the subsequent Au NP deposition is of significance, ZrO NFs with distinct size and dimension and dimension of nanofibers are preferable in this study. 2 Thereafter, the sample Au NPs/ZrO2 of nanofibers are preferable in this study. Thereafter, the sample Au NPs/ZrO NFs_0.3 (Figure 3d) NFs_0.3 (Figure 3d) suitably resulted. As determined by software Image J from2200 random Au NPs, suitably resulted. As determined by software Image J from 200 random Au NPs, the diameter of Au the diameter of Au NPs in the range from 30 to 45 nm was averaged, which was corresponding to NPs in the range from 30 to 45 nm was averaged, which was corresponding to the rough diameter of the rough diameter of ZrO2 NFs. ZrO2 NFs. 3.2. 3.2. The The Effect Effect of of SERS SERS for for the the Samples Samples of of Au Au NPs/ZrO NPs/ZrO22 NFs NFs In Figure 4, 4, SERS SERS spectra spectra were were taken taken from from the the samples samples of of Au Au NPs/ZrO NPs/ZrO2 NFs with the test In Figure 2 NFs with the test −3 M attached to them. The Raman laser with the wavelengths of 633 (Figure 4a) molecule R6G of 10 molecule R6G of 10−3 M attached to them. The Raman laser with the wavelengths of 633 (Figure 4a) and nm (Figure (Figure 4c) 4c) were were respectively respectively used. used. The The characteristic characteristic peaks peaks of ofR6G R6Gupon uponAu AuNPs/ZrO NPs/ZrO 2 and 785 785 nm 2 NFs were significantly detected and enhanced, as compared to the flat substrate. The most intense NFs were significantly detected and enhanced, as compared to the flat substrate. The most intense peak at at1361 1361cm cm whichis is assigned to the stretching of in C–C in aromatics, is usually used to −1−1, ,which peak assigned to the stretching of C–C aromatics, is usually used to calculate −1 calculate EF. By taking peak cm at 1361 as the reference, theenhancement peak enhancement 633laser nm −1 ascm the EF. Bythe taking the peakthe at 1361 the reference, the peak by 633bynm laser wavelength was higher than that by 785 nm. In Figure 4b,d, Au NPs/ZrO 2 NFs_0.3 showed the wavelength was higher than that by 785 nm. In Figure 4b,d, Au NPs/ZrO2 NFs_0.3 showed the most 7 under the excitation of 633 nm laser wavelength, which is most EF ×of102.1 × 10the 7 under intenseintense EF of 2.1 excitation of 633 nm laser wavelength, which is corresponding corresponding to the sample with distinct size and dimension of nanofibers, shown in Figure 3d. to the sample with distinct size and dimension of nanofibers, shown in Figure 3d. Particularly, Particularly, the gaps between the Auabundant NPs generate abundant “hot-spot” structures for the gaps between the aggregated Auaggregated NPs generate “hot-spot” structures for SERS, which are SERS, which are homogeneously distributed on the substrate, making a robust and reproducible homogeneously distributed on the substrate, making a robust and reproducible enhancement of the enhancement of the Raman signal feasible. Raman signal feasible.


6 of 11 6 of 11

Figure 3. SEM micrographs of ZrO2 NFs with different ZrCl4 concentrations. Morphologies from the Figure 3. SEM micrographs of ZrO2 NFs with different ZrCl4 concentrations. Morphologies from the surfaces of NPs/ZrO2 NFs_X with (a) X = 0.05 (b) X = 0.1, (c) X = 0.2, (d) X = 0.3, (e) X = 0.4, and (f) X = surfaces of NPs/ZrO2 NFs_X with (a) X = 0.05 (b) X = 0.1, (c) X = 0.2, (d) X = 0.3, (e) X = 0.4, and 0.5. (f) X = 0.5.


7 of 11 7 of 11

The effect effect of SERS on Au NPs deposited Figure 4. The deposited upon random random ZrO ZrO22 NFs with with different different ZrCl ZrCl44 concentrations were using the molecular probe R6G andR6G different laserRaman wavelengths. concentrations wereevaluated evaluated using the molecular probe andRaman different laser (a) and (b) are(a) the intensity factor of 633 nm laser; similarly to (c)similarly and (d) to with wavelengths. and (b) areand the enhancement intensity and enhancement factor of 633 nm laser; (c) 785 nm and (d) laser. with 785 nm laser.

Pesticides Using Using the the Optimal Optimal Au Au NPs/ZrO NPs/ZrO22 NFs_0.3 3.3. Trace Detection of Pesticides Using Au 22 NFs_0.3 the substrate, in in Figure 5a,5a, thethe characteristic SERS peaks for P1, Au NPs/ZrO NPs/ZrO NFs_0.3asas the substrate, Figure characteristic SERS peaks for 3 M were C1, C1, P2 and C2 in M−were examined and taken as theas references. The major characteristic peaks P1, P2 and C210in−3 10 examined and taken the references. The major characteristic −1 ), −1), C1 (713, of P1 (606, 713, 1192, and 1776 cm1776 1379, 1441 and1441 1582and cm−11582 ), P2cm (1002, peaks of P1654, (606, 654, 713,1379, 1192,1407 1379, 1407 and cm−1 ), C1 (713, 1379, − 1 −1 −1 1017, 1162,1017, 1209 1162, and 1582 1017, 1209, 1582 and1209, 2130 cm were2130 identified. P2 (1002, 1209cm and) and 1582C2 cm(1002, ) and C21162, (1002, 1017, 1162, 1582) and cm−1 ) − 1 − 1 −1, δ(C=O)), C1 (1379 cm−1, symmetric ring vibration), P2 (1002 The most intenseThe peaks of P1 (606 cm were identified. most intense peaks of P1 (606 cm , δ(C=O)), C1 (1379 cm , symmetric ring − 1 −1 , benzene −1 cm , benzene ring cm breathing vibration) and C2 vibration) (1002 cm−1and , benzene ring vibration) were vibration), P2 (1002 , benzene ring breathing C2 (1002 cmbreathing ring breathing respectively determined at low concentrations pesticides. The of detection were vibration) were respectively determined at low of concentrations of limits pesticides. The limits of estimated detection −8 M for as 10−6 M for cypermethrin 10−7 M for carbaryl and permethrin and 10(P2), were estimated as 10−6 M for (C2), cypermethrin (C2), 10−7 M(C1) for carbaryl (C1) and (P2), permethrin and − 8 phosmet which(P1), in practice, the meet tracethe concentration requirement of, for of, example, food 10 M for(P1), phosmet which in meet practice, trace concentration requirement for example, security [23]. There are five per test Each Each sample is performed from from 20 spots. The food security [23]. There aresamples five samples pergroup. test group. sample is performed 20 spots. results of using these concentrations can be found in the Supporting Data 4–8. The results of using these concentrations can be found in the Supporting Data 4–8. The application of SERS to to analyze analyze multiple multiple pesticides pesticides (in (in 10 10−−33 M) M) was was carried carried out using Au NPs/ZrO2 2NFs_0.3 NPs/ZrO NFs_0.3as asthe thesubstrate. substrate.The Thecharacteristic characteristicpeaks peaksof ofeach each pesticide, pesticide, shown shown in in Figure Figure 5a, 5a, usedas as references the presence of the pesticides. Figure 5b, qualitative were used the the references for thefor presence of the pesticides. In Figure 5b,Inqualitative measurements measurements from seven identical with multiple pesticides showed that2 Au NPs/ZrO from seven identical solutions with solutions multiple pesticides showed that Au NPs/ZrO NFs_0.3 was2 NFs_0.3 was competent the to presence distinguish the pesticides. presence ofRelatively four pesticides. peak competent to distinguish of four low peakRelatively intensitieslow naturally intensitiesfrom naturally resulted from of a low of each due pesticide. However, due the resulted a low concentration each concentration pesticide. However, to the overlapping of to peaks, −1 forfound −1 C2, overlapping of peaks, increases of relative peak intensities at C2, 1002and cm−1 P2cm and increases of relative peak intensities were found at 1002 cmwere P2 and at for 1379 for −1 for P1 and C1. and at 1379 cm P1 and C1.


8 of 11 8 of 11

Figure 5. 5. (a) (a) The atat the concentration of Figure The characteristic characteristic SERS SERSpeaks peaksfor forthe thepesticides pesticidesP1, P1,C1, C1,P2P2and andC2C2 the concentration −3 M; − 3 10 (b) SERS signals from 7 samples (i to vii) for the subsequent mixture of P1, C1, P2 and C2. of 10 M; (b) SERS signals from 7 samples (i to vii) for the subsequent mixture of P1, C1, P2 and C2. Their characteristic characteristic peaks peaks were were identified. identified. Their

ZrO2 shows a strong affinity to phosphoryl group [24], which could make organophosphates ZrO2 shows a strong affinity to phosphoryl group [24], which could make organophosphates more more competitive to adsorb on the ZrO2 surface. To verify the effect of ZrO2 templates on competitive to adsorb on the ZrO2 surface. To verify the effect of ZrO2 templates on organophosphates, organophosphates, the reduction of Raman intensity for each pesticide in various mixtures the reduction of Raman intensity for each pesticide in various mixtures compared to that in compared to that in standard solutions is shown in Table 1. The reduction of Raman intensity of standard solutions is shown in Table 1. The reduction of Raman intensity of phosmet [25], which is phosmet [25], which is an organophosphate, was less than the other pesticides in the mixtures. The an organophosphate, was less than the other pesticides in the mixtures. The results indicate that Au results indicate that Au NPs/ZrO2 NFs_0.3 exhibited higher selectivity to organophosphates due to NPs/ZrO2 NFs_0.3 exhibited higher selectivity to organophosphates due to the property of ZrO2 the property of ZrO2 templates. The utilization of ZrO2 NFs can be exploited as a substrate for templates. The utilization of ZrO2 NFs can be exploited as a substrate for enhancing Raman intensity enhancing Raman intensity on the adsorbed molecules. ZrO2 showed a strong affinity toward the on the adsorbed molecules. ZrO2 showed a strong affinity toward the phosphate group on parathion phosphate group on parathion molecules, which provides sensitivity and selectivity of the sensing molecules, which provides sensitivity and selectivity of the sensing film [26], as nitroaromatic NPs film [26], as nitroaromatic NPs strongly bind to the ZrO2 nanofiber surface [27]. The deposition of strongly bind to the ZrO2 nanofiber surface [27]. The deposition of gold particles on the fibers further gold particles on the fibers further amplifies Raman signals due to SERS. This study suggests that amplifies Raman signals due to SERS. This study suggests that Raman signals can be finely tuned in Raman signals can be finely tuned in intensity and effectively enhanced in nanofiber mats and intensity and effectively enhanced in nanofiber mats and arrays by properly tailoring the architecture, arrays by properly tailoring the architecture, composition and light-scattering properties of the composition and light-scattering properties of the complex networks of filaments. In addition, the large complex networks of filaments. In addition, the large area of inter-Au NP surface plasmon area of inter-Au NP surface plasmon resonance on ZrO2 NFs as a SERS-active substrate was applied to resonance on ZrO2 NFs as a SERS-active substrate was applied to distinguish multiple pesticides distinguish multiple pesticides due to their formation of hot spots (Figure 1). due to their formation of hot spots (Figure 1). Table 1. Raman intensity reduction of each pesticide in various mixtures compared to that in Table 1. Raman intensity reduction of each pesticide in various mixtures compared to that in standard solutions. standard solutions. Raman Intensity Reduction(%) (%) bb Raman Intensity Reduction P1 C1 P2 P2 C2 P1 C1 C2 i 04.0 68.1 i 04.0 68.1 ii -ii 02.5 02.5 64.7 64.7 iii -- 39.4 39.4 iii 18.9 18.9 iv 27.2 83.9 60.9 iv 27.2 83.9 60.9 v 20.1 82.1 49.7 v 33.6 20.1 82.1 - 94.3 49.7 vi 60.2 vi 25.5 33.6 94.3 80.3 60.2 vii 80.4 70.5 a The concentration of each pesticide b the Raman vii in various25.5 80.410−3 M;80.3 70.5 mixtures was intensity reduction was the ratio Mixture a

Mixture a

of Raman intensity of each in mixtures that in standard solutions. a The concentration ofpesticide each pesticide intovarious mixtures was 10 −3 M;

the Raman intensity reduction was the ratio of Raman intensity of each pesticide in mixtures to that in standard 3.4. Detection of Simulated Pesticides on Apples solutions. b

According to world food ethics (which are based on USA food rules) [28], the agricultural crops 3.4. Detection ofwith Simulated Pesticides on Apples contaminated residual pesticides (C1, P1, C2, P2) are certainly harmful to human health [29]. To simulate the detection pesticides on fruits, spiked with 10−2 [28], M standard solution According to world of food ethics (which areapples basedwere on USA food rules) the agricultural crops contaminated with residual pesticides (C1, P1, C2, P2) are certainly harmful to human health

Nanomaterials 2018,2018, 8, 402 Nanomaterials 8, x FOR PEER REVIEW

9 of 119 of 11

[29]. To simulate the detection of pesticides on fruits, apples were spiked with 10 −2 M standard

containing C1,containing P1, C2 and P2.P1, Subsequently, the apple peels soaked ethanol to in extract pesticides. solution C1, C2 and P2. Subsequently, thewere apple peels in were soaked ethanol to extract pesticides. The concentration each pesticide in extracts 10−4 M.containing SERS The concentration of each pesticide in of extracts was calculated to was 10−4calculated M. SERStospectra spectra containing pesticides extracted from apple peels are in Figure 6a. peaks The characteristic pesticides extracted from apple peels are shown in Figure 6a. shown The characteristic in the obtained peaks in the obtained SERS spectra matched those in the SERS spectra from the standard SERS spectra matched those in the SERS spectra from the standard solutions, as shownsolutions, in Figure 6b. as shown in Figure 6b. The enhancement was attributed to mainly high EMinduced effect, which The enhancement was attributed to high EM effect, which was by Auwas NPs.mainly The SERS induced by Au NPs. The SERS effect occurred at the metal–semiconductor junction and generated effect occurred at the metal–semiconductor junction and generated strong local electromagnetic (EM) strong local electromagnetic (EM) field at the interface in the hybrid nanosystem [30]. Notably, the field at the interface in the hybrid nanosystem [30]. Notably, the physical methods prevent samples physical methods prevent samples from chemical or residual contaminations and provide a large from density chemical or residual contaminations and provide a large2 density of Raman hot-spot of Raman hot-spot areas. Our optimized Au NPs/ZrO NFs substrate was applied forareas. Our optimized Au NPs/ZrO NFs substrate was applied for pesticide detection. 2 pesticide detection.

Figure 6. SERS spectra (a) from the pesticide-containing apple peels in comparison with (b) a

Figure 6. SERS spectra (a) from the pesticide-containing apple peels in comparison with (b) standard solution. a standard solution. 4. Conclusions

4. Conclusions

To have a simple and fast detection of pesticide residue on fruits or vegetables, a novel

To have with a simple and fast detection of pesticide residue on fruitsstable or vegetables, a novel material material a cost-effective solution has been proposed. Chemically ZrO 2 NFs are prepared a nonthermal spin-coated sol-gel followed by depositing AuZrO NPs2 and as an by with by a cost-effective solution has beenmethod, proposed. Chemically stable NFsforming are prepared integratedspin-coated Au NPs/ZrO 2 NFsmethod, substrate. The optimized sampleAu AuNPs NPs/ZrO 2 NFs_0.3 been a nonthermal sol-gel followed by depositing and forming as has an integrated 7. Au NPs/ZrO2 NFs_0.3 is competent to distinguish proved to be SERS-active with an EF of 2.1 × 10 Au NPs/ZrO2 NFs substrate. The optimized sample Au NPs/ZrO2 NFs_0.3 has been proved the characteristic Raman peaks from four 7kinds of pesticide residue; their detection limits can be to be SERS-active−6 with an EF of 2.1 × 10 . Au NPs/ZrO2 NFs_0.3 is competent to distinguish lowered to 10 –10−8 M. Besides, Au NPs/ZrO2 NFs show high selectivity to organophosphates when the characteristic Raman peaks from four kinds of pesticide residue; their detection limits can be multiple pesticides are present. For a practical application on the diluted juice containing sliced lowered to 10−6 –10−8 M. Besides, show highpesticide selectivity to organophosphates 2 NFs pesticide-containing apple peels, Au the NPs/ZrO characteristic peaks of each could also be clearly whenidentified. multipleMoreover, pesticides are present. For a practical application on the diluted juice containing Au NPs/ZrO2 NFs can be made on a large surface area and are thus promising slicedfor pesticide-containing apple peels, the characteristic peaks of each pesticide could also be clearly flexible and extensible applications. identified. Moreover, Au NPs/ZrO2 NFs can be made on a large surface area and are thus promising for flexible and extensible applications. Supplementary Materials: The following are available online at http://www.mdpi.com/2079-4991/8/6/402/s1, Figure S1: Supporting Data 1: (a) Low magnification and (b) high magnification of FE-SEM images for the sample AuNPs/ZrO2 NFs_0.3. A uniform Au NPs are deposited onto ZrO2 NFs. Figure S2: Supporting Data 2:


10 of 11

The EDS-mapping image of the cross-sectioned sample AuNPs/ZrO2 NFs_0.3. The green spot is Au element, while the red spot is Zr one. Figure S3: Supporting Data 3: The size distribution histogram of Au NPs on the sample AuNPs/ZrO2 NFs_0.3. To calculate the sizes, random 200 particles on the Supporting Data 1(a) are chosen. Figure 4S: Supporting Data 5: (a) SERS spectra of phosmet standard solutions of various concentrations, and (b) the molecular structure of phosmet. Figure 5S: Supporting Data 6 (a) SERS spectra of carbaryl standard solutions of various concentrations, and (b) the molecular structure of carbaryl. Figure 6S: Supporting Data 7: (a) SERS spectra of permethrin standard solutions of various concentrations, and (b) the molecular structure of permethrin. Figure 7S: Supporting Data 8 (a) SERS spectra of cypermethrin standard solutions of various concentrations, and (b) the molecular structure of cypermethrin. Table S1: Supporting Data 4: Raman spectra of the analytes. Author Contributions: Han Lee designed the study, analyzed the data, generated the figures and wrote the manuscript. Chih-Chien Chen performed functional experiments and data analysis and generated the figures. Kundan Sivashanmugan, Bernard Haochih Liu, Wei-en Fu and Yung-Der Juang helped in the design of the study. Jiunn-Der Liao designed the study, interpreted results, modified the manuscript. Guo Dung Chen edited the manuscript. Acknowledgments: This research was supported in part by (received funding from) the Headquarters of University Advancement at the National Cheng Kung University, which is sponsored by the Ministry of Education, Taiwan, under grant number D107-F2301 and MOST-103-2221-E-006-067-MY3. Conflicts of Interest: The authors declare no conflicts of interest.

References 1. 2.

3.

4.

5.

6. 7.

8.

9. 10.

11. 12.

Pu, H.; Xiao, W.; Sun, D.W. SERS-microfluidic systems: A potential platform for rapid analysis of food contaminants. Trends Food Sci. Technol. 2017, 70, 114–126. [CrossRef] Koesukwiwat, U.; Lehotay, S.J.; Miao, S.; Leepipatpiboon, N. High throughput analysis of 150 pesticides in fruits and vegetables using QuEChERS and low-pressure gas chromatography–time-of-flight mass spectrometry. J. Chromatogr. A 2010, 1217, 6692–6703. [CrossRef] [PubMed] Wu, Q.; Chang, Q.; Wu, C.; Rao, H.; Zeng, X.; Wang, C.; Wang, Z. Ultrasound-assisted surfactant-enhanced emulsification microextraction for the determination of carbamate pesticides in water samples by high performance liquid chromatography. J. Chromatogr. A 2010, 1217, 1773–1778. [CrossRef] [PubMed] Cheng, C.; Yan, B.; Wong, S.M.; Li, X.; Zhou, W.; Yu, T.; Fan, H.J. Fabrication and SERS performance of silver-nanoparticle-decorated Si/ZnO nanotrees in ordered arrays. ACS Appl. Mater. Interfaces 2010, 2, 1824–1828. [CrossRef] [PubMed] Chen, G.; Cao, P.; Liu, R. A multi-residue method for fast determination of pesticides in tea by ultra performance liquid chromatography-electrospray tandem mass spectrometry combined with modified QuEChERS sample preparation procedure. Food Chem. 2011, 125, 1406–1411. [CrossRef] Yaseen, T.; Sun, D.W.; Cheng, J.H. Raman imaging for food quality and safety evaluation: Fundamentals and applications. Trends Food Sci. Technol. 2017, 62, 177–189. [CrossRef] Sivashanmugan, K.; Liao, J.D.; Liu, B.H.; Yu, L.C. AuGa2 on focused Ga ion beam-fabricated Au nanorod array for trace detection of melamine cyanurate in milk solution. Appl. Phys. Express 2014, 8, 017001. [CrossRef] Sivashanmugan, K.; Liao, J.D.; Liu, B.H.; Yao, C.K. Focused-ion-beam-fabricated Au nanorods coupled with Ag nanoparticles used as surface-enhanced Raman scattering-active substrate for analyzing trace melamine constituents in solution. Anal. Chim. Acta 2013, 800, 56–64. [CrossRef] [PubMed] Halvorson, R.A.; Vikesland, P.J. Surface-enhanced Raman spectroscopy (SERS) for environmental analyses. Environ. Sci. Technol 2010, 44, 7749–7755. [CrossRef] [PubMed] Liu, B.; Han, G.; Zhang, Z.; Liu, R.; Jiang, C.; Wang, S.; Han, M.Y. Shell thickness-dependent Raman enhancement for rapid identification and detection of pesticide residues at fruit peels. Anal. Chem. 2011, 84, 255–261. [CrossRef] [PubMed] Guerrini, L.; Graham, D. Molecularly-mediated assemblies of plasmonic nanoparticles for Surface-Enhanced Raman Spectroscopy applications. Chem. Soc. Rev. 2012, 41, 7085–7107. [CrossRef] [PubMed] Lee, A.; Andrade, G.F.; Ahmed, A.; Souza, M.L.; Coombs, N.; Tumarkin, E.; Kumacheva, E. Probing dynamic generation of hot-spots in self-assembled chains of gold nanorods by surface-enhanced Raman scattering. J. Am. Chem. Soc. 2011, 133, 7563–7570. [CrossRef] [PubMed]


13.

14. 15.

16. 17. 18.

19.

20. 21.

22. 23.

24.

25. 26. 27. 28.

29.

30.

11 of 11

Sivashanmugan, K.; Liao, J.D.; Liu, B.H.; Yao, C.K.; Luo, S.C. Ag nanoclusters on ZnO nanodome array as hybrid SERS-active substrate for trace detection of malachite green. Sens. Actuator B-Chem. 2015, 207, 430–436. [CrossRef] Xie, H.N.; Larmour, I.A.; Smith, W.E.; Faulds, K.; Graham, D. Surface-enhanced Raman scattering investigation of hollow gold nanospheres. J. Phys. Chem. C. 2012, 116, 8338–8342. [CrossRef] Wang, X.; Wang, C.; Cheng, L.; Lee, S.T.; Liu, Z. Noble metal coated single-walled carbon nanotubes for applications in surface enhanced Raman scattering imaging and photothermal therapy. J. Am. Chem. Soc. 2012, 134, 7414–7422. [CrossRef] [PubMed] Strelau, K.K.; Schüler, T.; Möller, R.; Fritzsche, W.; Popp, J. Novel Bottom-Up SERS Substrates for Quantitative and Parallelized Analytics. ChemPhysChem. 2010, 11, 394–398. [CrossRef] [PubMed] Innocenzi, P.; Malfatti, L.; Falcaro, P. Hard X-rays meet soft matter: When bottom-up and top-down get along well. Soft Matter. 2012, 8, 3722–3729. [CrossRef] Huang, Z.; Meng, G.; Huang, Q.; Chen, B.; Zhu, C.; Zhang, Z. Large-area Ag nanorod array substrates for SERS: AAO template-assisted fabrication, functionalization, and application in detection PCBs. J. Raman Spectrosc. 2013, 44, 240–246. [CrossRef] Wolosiuk, A.; Tognalli, N.G.; Martínez, E.D.; Granada, M.; Fuertes, M.C.; Troiani, H.; Soler-Illia, G.J. Silver nanoparticle-mesoporous oxide nanocomposite thin films: A platform for spatially homogeneous SERS-active substrates with enhanced stability. ACS Appl. Mater. Interfaces 2014, 6, 5263–5272. [CrossRef] [PubMed] Jiang, R.; Li, B.; Fang, C.; Wang, J. Metal/Semiconductor Hybrid Nanostructures for Plasmon-Enhanced Applications. Adv. Mater. 2014, 26, 5274–5309. [CrossRef] [PubMed] Lee, H.; Yao, C.K.; Liao, J.D.; Shao, P.L.; Thi, M.H.N.; Lin, Y.H.; Juang, Y.D. Annealed thin-film zirconia coating adhered on 316 L stainless steel as a bio-inert indwelling needle. Mater. Des. 2015, 88, 651–658. [CrossRef] Lee, H.; Liao, J.D.; Sivashanmugan, K.; Liu, B.H.; Weng, S.L.; Juang, Y.D.; Yao, C.K. Dual properties of zirconia coated porous titanium for a stiffness enhanced bio-scaffold. Mater. Des. 2017, 132, 13–21. [CrossRef] Tsaboula, A.; Papadakis, E.N.; Vryzas, Z.; Kotopoulou, A.; Kintzikoglou, K.; Papadopoulou-Mourkidou, E. Environmental and human risk hierarchy of pesticides: A prioritization method, based on monitoring, hazard assessment and environmental fate. Environ. Int. 2016, 91, 78–93. [CrossRef] [PubMed] Zhou, H.; Xu, S.; Ye, M.; Feng, S.; Pan, C.; Jiang, X.; Zou, H. Zirconium phosphonate-modified porous silicon for highly specific capture of phosphopeptides and MALDI-TOF MS analysis. J. Proteome Res. 2006, 5, 2431–2437. [CrossRef] [PubMed] Fan, Y.; Lai, K.; Rasco, B.A.; Huang, Y. Analyses of phosmet residues in apples with surface-enhanced Raman spectroscopy. Food Control 2014, 37, 153–157. [CrossRef] Liu, G.; Lin, Y. Electrochemical Sensor for Organophosphate Pesticides and Nerve Agents Using Zirconia Nanoparticles as Selective Sorbents. Anal. Chem. 2005, 77, 5894–5901. [CrossRef] [PubMed] Wang, M.; Li, Z. Nano-Composite ZrO2 /Au Film Electrode for Voltammetric Detection of Parathion. Sens. Actuator B-Chem. 2008, 133, 607–612. [CrossRef] Abdalla, A.A.; Afify, A.S.; Hasaan, I.E.; Mohamed, A. Studying the Effect of Household-Type Treatment and Processing on the Residues of Ethion and Profenofos Pesticides and on the Contents of Capsaicinoids in Green Chili Pepper Using GC-MS/MS and HPLC. Food Anal. Meth. 2017, 11, 1–12. [CrossRef] Sivashanmugan, K.; Lee, H.; Syu, C.H.; Liu, B.H.C.; Liao, J.D. Nanoplasmonic Au/Ag/Au nanorod arrays as SERS-active substrate for the detection of pesticide residue. J. Taiwan Inst. Chem. Eng. 2017, 75, 287–291. [CrossRef] Siddhanta, S.; Thakur, V.; Narayana, C.; Shivaprasad, S.M. Universal metal-semiconductor hybrid nanostructured SERS substrate for biosensing. ACS Appl. Mater. Inter. 2012, 4, 5807–5812. [CrossRef] [PubMed] © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).