Synthesis of novel decorated one-dimensional gold nanoparticle and its application in ultrasensitive detection of insecticide

Abstract

Contamination of soil, groundwater and food with methyl parathion, which is one of the most hazardous insecticides, is causing wide concern. Motivated by the urgent demand for trace analysis of insecticide from environmental samples, searching for a novel highly sensitive or selective analytical approach has been a longstanding interest. The as-prepared one dimensional gold nanoparticles, with different aspect ratios, have been decorated with mono-6-thio-β-cyclodextrin so as to efficiently capture and detect methyl parathion insecticide via a surface enhanced Raman scattering (SERS) technique for the first time. A detailed comparison among different aspect ratios of the one dimensional gold nanoparticles suggests that the one dimensional gold nanoparticles with aspect ratio 2 can be provided as an excellent SERS active substrate for detecting the insecticide in the present study. Due to the efficient formation of a host–guest complex between the hybridized cavity and methyl parathion, identification of the insecticide can be observed at picomolar level according to the fingerprint Raman peaks. Excellent sensitivity and selectivity for detection of the insecticide using the hybridized one dimensional gold nanoparticles without a Raman label indicate that it is a simple and ultrasensitive approach for detecting insecticides in contrast with some other traditional detection approaches.

---

Introduction

Surface enhanced Raman scattering (SERS) holds great potential as a powerful analytical technique for rapid and sensitive chemical or biological analysis.^1–11 Two mechanisms contribute to the SERS effect, one electromagnetic (EM) enhancement and one chemical (charge transfer or CT) enhancement in nature; moreover, electromagnetic enhancement stems from the enhancement of the local electromagnetic field incident on absorbed molecules at the noble metallic nanoparticles with the size scale 10–200 nm.¹² On the other hand, chemical enhancement is highly dependent on the electronic resonance-charge transfer between absorbed molecules and metal surfaces at atomic scale roughness features, which increases the Raman scattering cross section due to an increase of the polarizability of the molecule.¹³ Although coinage metal nanoparticles have been widely used as SERS substrates,^14–25 investigations on different shaped noble metallic nanoparticles are very active.^26–35 Previous investigations on SERS properties of the one dimensional gold nanoparticles suggested that the SERS intensity could be obviously strengthened when molecules was absorbed on a surface covered with aggregated nanorods thanks to the enhancement of the electric field among the particles in the aggregates.³⁶ In comparison with a number of previous investigations focused on the evaluation of SERS properties of different shaped metallic nanoparticles, applications in chemosensing or biosensing of controllable shaped metallic nanoparticles via the SERS technique are very limited. To our knowledge, detailed detection of insecticides on the basis of one dimensional gold nanoparticles with different aspect ratios through SERS has not been reported, which implies that this approach may be a potential candidate for the novel trace analysis of insecticides.

As one of the most hazardous insecticides which is widely used to control sucking and chewing insects,³⁷ the detection limit of methyl parathion insecticide in vegetables is in the range 0.1–1.0 ppm according to recommendations from the Collaborative International Insecticide Analytical Council (CIPAC).³⁸ As far as the present detection approaches for insecticide, e.g., gas chromatography mass spectroscopy (GC-MS),³⁹ electrochemical,⁴⁰ high-performance liquid chromatography (HPLC)^41–43 and electrophoresis⁴⁴ are concerned, some disadvantages, viz., poor detection limit, lengthy measurement time, and indirect measurement, limit their applications. In contrast, surface enhanced Raman scattering (SERS) provides great advantages for detecting chemical and structural information of compounds on the microscopic scale.

As far as prinstine gold nanoparticles are concerned, it is difficult to directly bind to target molecules for detection, which indicates that it is necessary to decorate the gold particles via a chemical approach, e.g., cystamine, DPAA, MTA, etc.^45,46 Therefore, in order to efficiently detect insecticide molecules using one dimensional gold nanoparticles, chemical decoration, e.g., cyclodextrin, has to be considered. Cyclodextrins are cyclic oligosaccharides, which are composed of six, seven, or eight glucopyranose units in α, β, and γ forms, respectively. Based on a special molecular structure—hydrophobic internal cativity and hydrophilic external surface^47,48—it can be expected that an inclusion complex with guest molecules can be efficiently formed between aromatic insecticide molecules and cyclodextrin decorated one dimensional gold nanoparticles. Therefore, trace SERS analysis for the detection of methyl parathion insecticide has been firstly performed by hybridized mono-6-thio-β-cyclodextrin with one dimensional gold nanoparticles with different aspect ratios. The excellent SERS properties of the one dimensional gold nanoparticles with a specific aspect ratio can remarkably enhance the SERS signal of methyl parathion insecticide.

Experimental section

Chemical materials and instrumentation

HAuCl₄·4H₂O (99.9%), NaBH₄ (99%), Vitamin C (99.9%), CTAB (99%), BDAC (99%) and AgNO₃ (99.9%), tri-sodium citrate (99.9%), p-toluenesulfonyl chloride (analytical purity), thiourea (analytical purity) were obtained from Aldrich. β-CD was purchased from Sinopharm Chemical Reagent Co., Ltd., China. The double distilled water that was used throughout the experiments was purified using a Milli-Q system.

Surface enhanced Raman spectra were collected using 25 mW of 632.8 nm radiation from a helium neon laser. Integration times were provided as 12 s. Transmission electron microscopy (TEM) images were acquired with a Hitachi 800 microscope operating at 200 kV. The ultraviolet-visible (UV-Vis) spectra were used to characterize the optical properties of one dimensional gold nanoparticles with different aspect ratios and measured in quartz cuvettes with a Shimadzu UV2550 spectrophotometer. FTIR spectra were recorded on a Nexus-870 spectrophotometer (KBr pellets). NMR spectra were acquired at 25 °C using a Bruker Avance spectrometer.

Synthesis of one dimensional gold nanoparticles with different aspect ratios

One dimensional gold nanoparticles were synthesized by a seed-mediated approach with modifications from the previous reports.^49–51 Preparation of the one dimensional gold nanoparticles with different low aspect ratios could be divided into three steps. The preparation of the seed solution could be described as follows in detail. 10 mL, 0.5 mM HAuCl₄ was mixed with 10 mL, 0.2 M CTAB. Subsequently, 600 μL, 0.02 M ice-cold NaBH₄ was added to the mixed solution. Vigorous stirring of the seed solution could be kept for 120 s; furthermore, the seed solution was settled at room temperature for 1 h. As the second step, the growth solution of the one dimensional gold nanoparticles was prepared. 10 mL, 0.2 M CTAB was mixed with 1 mL, 4 mM AgNO₃ solution at ambient temperature. Then, 10 mL, 1 mM HAuCl₄ was added to the growth solution followed by addition of 140 μL, 0.08 M vitamin C. Finally, the one dimensional gold nanoparticles with low aspect ratio 2 : 1 were prepared by addition of 24 μL seed solution to the growth solution. Compared to synthesis of the one dimensional gold nanoparticles with low aspect ratio, a binary surfactant mixture of CTAB and BDAC has been employed in order to synthesize the one dimensional gold nanoparticles with a relatively high aspect ratio 5 : 1. Firstly, 10 mL, 0.15 M BDAC was added to 0.546 g of CTAB and dissolved by sonication in order to prepare the surfactant mixtures. Then, 400 μL of 0.004 M AgNO₃, 10 mL, 0.001 M HAuCl₄ and 140 μL, 0.08 M vitamin C were added to the mixed solution, respectively. Finally, 24 μL seed solution was added to growth solution and settled overnight.

One dimensional gold nanoparticles with high aspect ratios can be synthesized via a three-step seed mediated method, which was described by Murphy;⁵¹ however, extra purification of large aspect ratio one dimensional gold nanoparticles from a large quantity of gold nanoparticles has to be performed as in a previous report.⁵¹ Initially, the mixture including 0.5 mL, 0.1 M HAuCl₄ and 0.01 mL, 0.1 M tri-sodium citrate was dissolved in 19 mL H₂O. Then, 600 μL, 0.1 M ice cold NaBH₄ was added to the mixture accompanied with vigorous stirring for 2 min. The as-prepared seed solution has to be used within 10 min. 64.06 g CTAB and 181.3 mg HAuCl₄·4H₂O were dissolved in the 880 mL H₂O. The as-prepared solution in amounts of 45 mL, 140 mL and 1.575 L was transferred to flask A, B and C, respectively. Subsequently, 250 μL, 770 μL and 8.75 mL of 0.1 M vitamin C solution have been added into flasks A, B and C, respectively. 4 mL of seed solution was added to flask A and mixed thoroughly; furthermore, 12.4 mL solution was transferred from the flask A to flask B immediately. Finally, all of the solution in flask B was added to flask C. The color of solution in the flask C changed to dark-red after 3 min and flask C was settled overnight. The mixtures of the one dimensional gold nanoparticles with very high aspect ratio and large quantity of nanoplates or nanospheres have been formed; moreover, the long one dimensional gold nanoparticles could be precipitated from the solution to the bottom of flask C. In order to purify the long one dimensional gold nanoparticles from nanoplates or nanospheres, an oxidizing Au(III)/CTAB complex, viz., dissolving 0.3645 g CTAB and 1 mL 0.0005 M HAuCl₄ in 9 mL H₂O, have been prepared. 1 mL of the solution was added to a suspension of the one dimensional gold nanoparticles/nanoplates and settled for 14 h. Then, the supernatant containing nanodisks was removed and the 1 mL solution at the bottom of the flask was redispersed in 10 mL 0.1 M CTAB solution followed by addition of 1 mL Au(III)/CTAB solution. The purification process was repeated four times so that large amount of nanoplates or nanospheres could be separated from the one dimensional gold nanoparticles with high aspect ratios.

Synthesis of mono-6-deoxy-6-(p-toylsulfonyl)-β-cyclodextrin (β-CDtos)

70 g β-cyclodextrin was suspended in 550 mL H₂O, and 20 mL aqueous solution including 6.6 g NaOH was added dropwise to the β-cyclodextrin suspension so that the suspension gradually became transparent. 10.1 g p-toluenesulfonyl chloride in 30 mL acetonitrile was added dropwise to the transparent solution so as to form a white precipitate. After stirring for 2 h at ambient temperature, the precipitate could be removed by suction filtration and the filtrate was refrigerated for a couple of days at 4 °C. The white precipitate was recovered via suction filtration and recrystallized twice to obtain a pure white solid. ¹H NMR (400 MHz, (CD₃)₂SO, 25 °C, TMS, δ): d = 2.43 (s, 3 H), 3.20–3.40 (m, overlaps with HOD), 3.44–3.66 (m, 28 H), 4.17–4.52 (m, 6 H), 4.77 (br s, 3 H), 4.84 (br s, 4 H), 5.64–5.84 (m, 14 H), 7.43 (d, J = 8.07 Hz, 2 H), 7.75 (d, J = 8.11 Hz, 2 H) ppm. Anal. Calcd for C₄₉H₇₆O₃₇S·3H₂O: C, 43.81; H, 6.15; S, 2.39. Found: C, 43.58; H, 6.21; S, 2.28.

Synthesis of mono-6-thio-β-cyclodextrin

A mixture of 2 g β-CDtos and 2 g thiourea in 100 mL 80% methanol–water (v/v) was heated at reflux for 2 days and the solvent of the reaction mixture was removed in vacuum. Subsequently, the white solid was added to 30 mL methanol and stirred for 1 h. After filtration, the residue was dissolved with a 10 wt% 69 mL NaOH solution and stirred at 50 °C for 5 h. When the pH value of the reaction mixture was adjusted to 2 with 10 wt% HCl, the pale yellow solution was formed. Then, 5 mL trichloroethylene was added to the solution and stirred overnight. The resulting white precipitate was recovered by suction filtration and washed with water. To confirm the presence of thio groups in the product, NaNO₂ and HCl were added into the as-prepared solution. The color of the solution changed to red, which indicated the existence of thio group in the product. After recrystallization, the white solid was obtained. ¹H NMR (400 MHz, (CD₃)₂SO, 25 °C, TMS, δ): d = 2.03 (m, SH), 2.50–3.20 (m, 2 H), 3.26–3.47 (m, overlapping with HDO), 3.56–3.76 (m, 28 H), 4.38–4.52 (m, 6 H), 4.83, 4.91 (br d, 7 H), 5.59–5.84 (m, 14 H) ppm. Anal. Calcd. for C₄₂H₇₀O₃₄S·7H₂O: C, 39.50; H, 6.63; S, 2.51. Found: C, 39.23; H, 7.02; S, 2.38.

Preparation of mono-6-thio-β-cyclodextrin–one dimensional gold nanoparticles–methyl parathion insecticide and SERS experimental details

After synthesis of CTAB-stabilized one dimensional gold nanoparticles, excess CTAB molecules on the one dimensional gold nanoparticles (5 mL solution) with different aspect ratios were removed by centrifuging once at 10 000 rpm, discarding the supernatant and redispersing the particles in pure water. Subsequently, a 500 μL solution of one dimensional gold nanoparticles was combined with varying amounts of mono-6-thio-β-cyclodextrin in DMF to give final concentrations of 1.00, 5.00, 25.0, 50.0, 100.0, 200.0 and 500.0 × 10⁻⁵ M. These mixtures were kept at 40 °C under constant sonication for 24 h. This procedure was then followed by centrifuging at 10 000 rpm for 20 min to remove excess CTAB and mono-6-thio-β-cyclodextrin. Finally, the mixtures were dried in vacuum at 60 °C overnight to remove water and DMF. The complexes involving different amounts of mono-6-thio-β-cyclodextrin were conjugated with 1 mL 1 × 10⁻⁶ M insecticide solution at ambient temperature for overnight. Peaks heights of the Raman mode at 1393 cm⁻¹ were determined and plotted against different concentrations of mono-6-thio-β-cyclodextrin. On the basis of the observations, accompanied with the increase of concentration of mono-6-thio-β-cyclodextrin, the intensity of the Raman mode at 1393 cm⁻¹ can be increased simultaneously; however, when the concentration of mono-6-thio-β-cyclodextrin is increased to 10⁻³ M, the intensity of the Raman mode at 1393 cm⁻¹ can almost be kept identical, which suggests that the absorption of cyclodextrin on the one dimensional gold nanoparticles has been saturated. After purification of hybridized mono-6-thio-β-cyclodextrin–one dimensional gold nanoparticles with different aspect ratios, 100 μL of a hybridized 10⁻³ M hybridized one dimensional gold nanoparticle solution was added to a 1 mL 1 × 10⁻⁸ M insecticide solution. Peaks heights of the Raman mode at 1393 cm⁻¹ were determined and plotted against aspect ratio with hybridized mono-6-thio-β-cyclodextrin-one dimensional gold nanoparticles with different aspect ratios. SERS results reveal that hybridized one dimensional gold nanoparticles with an aspect ratio ca. 2 can give higher enhancement for detecting methyl parathion insecticide compared to hybridized one dimensional gold nanoparticles with aspect ratios ca. 5 and 15. Hence, the one dimensional gold nanoparticle solution with an aspect ratio ca. 2 has been selected and equilibrated with varied amounts of methyl parathion insecticide in the range 10⁻⁶ to 10⁻¹² M overnight. Subsequently, the aqueous solution was centrifuged, removing possible excess insecticide, and the precipitation was dissolved in water and equilibrated under ambient conditions for 10 min prior to spectral analysis.

Results and discussion

Characterization and analysis of unhybridized and hybridized one dimensional gold nanoparticles

The one dimensional gold nanoparticles with different aspect ratios can be yielded by seed-mediated and surfactant directed synthesis, which is modified as compared to previous reports.^49–51 In contrast with strong reducing agent NaBH₄, the mild reducing agent vitamin C is selected in the process of growth for one dimensional gold nanoparticles. As far as the reducing ability of vitamin C is concerned, it can not directly reduce the metal salt to the elemental metal. It is necessary to add seed particles into the growth solution because the further redox reaction could occur on the seed surface and the autocatalytic process contributes to yield larger nanoparticles. Therefore, based on the previous investigations,⁵¹ Au⁺ can be firstly formed via a redox reaction between Au³⁺ and vitamin C. Subsequently, by aid of seed particles as nucleation centres, the Au⁺ can be reduced to Au⁰ through vitamin C on their surface. According to reports from Jana et. al., the one dimensional gold nanoparticles can be easily transformed to the gold nanospheres in the absence of silver ions.⁵² In order to inhibit the growth of gold nanospheres, it is necessary to add a certain amount of AgNO₃ in the reaction system, which can lead to formation of the one dimensional gold nanoparticles. Therefore, optimizing the amount of AgNO₃ in the growth solution contributes to obtaining high-yields of the one dimensional gold nanoparticles.

As shown in Fig. 1A, the one dimensional gold nanoparticles with low aspect ratio (ca. 2 : 1) can be formed if 24 μL seed solution is added to growth solution containing 1 mL AgNO₃ and the one dimensional gold nanoparticles. If 24 μL gold seed is added to the growth solution containing CTAB/BDAC cosurfactant, the one dimensional gold nanoparticles with relatively high aspect ratio (ca. 5 : 1) should be yielded, which is reflected in the TEM micrograph of Fig. 1B. Interestingly, the one dimensional gold nanoparticles with different aspect ratios can also be observed via a colorimetric approach, that is, the color of the one dimensional gold nanoparticles with aspect ratio 5 : 1 could be observed as a wine-red color as compared to the blue color of the aspect ratio 2 one dimensional gold nanoparticle solution. As for the one dimensional gold nanoparticles with aspect ratio ca. 15, it can also be well distributed, which is shown in Fig. 1C.

Fig. 1 TEM micrographs of the one dimensional gold nanoparticles synthesized by seed-mediated methods. (A) The one dimensional gold nanoparticles with the lowest averaged aspect ratio ca. 2 : 1. (B) The one dimensional gold nanoparticles with averaged aspect ratio ca. 5 : 1. (C) The one dimensional gold nanoparticles with very large averaged aspect ratio ca. 15 : 1. The scale bars in each micrograph measure 100 nm.

Compared to the single plasmon absorption band (ca. 520 nm) of the gold nanospheres, two plasmon absorption bands of the one dimensional gold nanoparticles, viz., transverse band and longitudinal bands, can be observed; furthermore, the longitudinal plasmon absorption band of the one dimensional gold nanoparticles could be redshifted and broadened with the increase of aspect ratio. In the present experiment, the one dimensional gold nanoparticles with aspect ratio 2 : 1 have a transverse band at 520 nm and a longitudinal plasmon band at 621 nm (shown in Fig. 2A). However, in the case of the one dimensional gold nanoparticles with aspect ratio 5 : 1, its longitudinal plasmon band is redshifted to 765 nm (refer to Fig. 2B). If the aspect ratio of one dimensional gold nanoparticles is increased to 15 : 1, a very strong longitudinal band could be located at 1583 nm. Therefore, according to the combined results of TEM and UV-Vis experiments, it can be concluded that the one dimensional gold nanoparticles with different aspect ratios have been synthesized through controlling surfactants, AgNO₃ and seed solution.

Fig. 2 Optical spectra of the one dimensional gold nanoparticles (A) the one dimensional gold nanoparticles with the lowest averaged aspect ratio ca. 2 : 1. (B) The one dimensional gold nanoparticles with averaged aspect ratio ca. 5 : 1. (C) The one dimensional gold nanoparticles with very large averaged aspect ratio ca. 15 : 1.

Before applying one dimensional gold nanoparticles with various aspect ratios to detect insecticide, cyclodextrin decoration of these one dimensional gold nanoparticles is necessary so as to efficiently bind with the target molecules. It is well known that most of the macrocyclic host molecules are decorated on the surface of nanoparticles via a ligand exchange approach. Motivated by this idea, a novel thiolated β-cyclodextrin derivative without an alkyl chain is designed and synthesized and compared to previous reports^53,54 so as to effectively associate with these one dimensional gold nanoparticles and decrease the distance between insecticide molecules and the surface of the one dimensional gold nanoparticles. Furthermore, the products of mono-6-deoxy-6-(p-toylsulfonyl)-β-cyclodextrin (β-CDtos) and mono-6-thio-β-cyclodextrin have been characterized by FTIR, and are shown in Fig. 3. In comparison with β-CDtos and β-CD, SH stretching vibrational bands can be observed at 2602 cm⁻¹ (shown in Fig. 3C), clearly suggesting that β-cyclodextrin has been thiolated. CTAB stabilized one dimensional gold nanoparticles and cyclodextrin decorated one dimensional gold nanoparticles are shown in Fig. 4A and 4B, respectively. The FTIR spectrum of CTAB on the one dimensional gold nanoparticles exhibits two strong vibrational bands of C–CH₂ in methylene chains, viz., asymmetric stretching band at 2919 cm⁻¹ and symmetric stretching band at 2850 cm⁻¹. After the one dimensional gold nanoparticles are decorated by mono-6-thio-β-cyclodextrin, the strong vibrational bands at 2919 cm⁻¹ and 2850 cm⁻¹ become quite weak, which indicate CTAB molecules capping around the one dimensional gold nanoparticles have been substituted by mono-6-thio-β-cyclodextrin. Moreover, the appearance of the intense absorption bands in hybridized cyclodextrin–one dimensional gold nanoparticles, i.e., 3418 cm⁻¹, 1655 cm⁻¹, 1038 cm⁻¹, which are identical with those in mono-6-thio-β-cyclodextrin, suggest that the surface of the one dimensional gold nanoparticles is efficiently modified by mono-6-thio-β-cyclodextrin. The CTAB capping on the one dimensional gold nanoparticles can be easily substituted by mono-6-thio-β-cyclodextrin due to the strong interaction between Au and SH.

Fig. 3 (A) FT-IR spectrum of β-cyclodextrin, (B) FT-IR spectrum of mono-6-deoxy-6-(p-toylsulfonyl)-β-cyclodextrin (β-CDtos), (C) FT-IR spectrum of mono-6-thio-β-cyclodextrin.

Fig. 4 (A) FT-IR spectrum of hybridized CTAB-one dimensional gold nanoparticles with aspect ratio 2, (B) FT-IR spectrum of mono-6-thio-β-cyclodextrin-one dimensional gold nanoparticles with aspect ratio 2.

In order to further confirm the replacement on the surface of one dimensional gold nanoparticles, the gel electrophoresis and preliminary ζ-potential measurement experiments have been carried out and these results suggest that CTAB-coated one dimensional gold nanoparticles are positively charged. However, as observed from gel electrophoresis (Fig. S1, ESI†), mono-6-thio-β-cyclodextrin–one dimensional gold nanoparticles (lane B) could run in the positive direction in contrast with CTAB coated nanoparticles (lane A), indicating that one dimensional gold nanoparticles are negatively charged if mono-6-thio-β-cyclodextrin is decorated on their surfaces. The analogous observations could also be found in the recent report on the decoration of mercaptocarboxylic acid (MCA), SH-PEG on the one dimensional gold nanoparticles.⁵⁵

Additionally, observations of decoration of one dimensional gold nanoparticles can be further confirmed by TEM images. As shown in Fig. 5, although the size of hybridized cyclodextrin–one dimensional gold nanoparticles can be increased compared to undecorated one dimensional gold nanoparticles with identical aspect ratios ca. 2 (33 × 16 nm), aspect ratios of the decorated one dimensional gold nanoparticles could not be obviously changed. In order to confirm TEM observations, comparison of UV-Vis spectra between decorated and undecorated one dimensional gold nanoparticles has been performed. As shown from their UV-Vis spectra (Fig. S2, ESI†), the longitude plasmon band can only be slightly redshifted accompanied with the decoration of mono-6-thio-β-cyclodextrin, indicating that the aspect ratio (ca. 2) should not be significantly changed. More importantly, the longitude plasmon band of the decorated one dimensional gold nanoparticles could not be broadened compared to the undecorated one dimensional gold nanoparticles. It is indicated that no aggregations take place in the process of decoration, which is consistent with the observations in TEM experiments. It should be pointed out that the redshift phenomenon on the longitude plasmon band of thiol-derivatives decorated one dimensional gold nanoparticles can also observed in the previous experiments, e.g., MCA decorated one dimensional gold nanoparticles,⁵⁵ which are in good agreement with the present observations.

Fig. 5 TEM micrographs of one dimensional gold nanoparticles and hybridized one dimensional gold nanoparticles. (A) The one dimensional gold nanoparticles with aspect ratio 2. (B) Hybridized mono-6-thio-β-cyclodextrin–one dimensional gold nanoparticles with aspect ratio 2. The scale bars in each micrograph measure 100 nm.

Hence, combined results of TEM, UV-Vis, gel electrophoresis and preliminary ζ-potential measurements, suggest that the one dimensional gold nanoparticles have been efficiently decorated by mono-6-thio-β-cyclodextrin.

SERS analysis of methyl parathion insecticide trapped by mono-6-thio-β-cyclodextrin–one dimensional gold nanoparticles

Considering the fact that the inclusion of complexes between cyclodextrin–one dimensional gold nanoparticles and methyl parathion insecticide can be effectively formed via a host–guest interaction (shown in Fig. 6), SERS detection can be performed on the basis of remarkable SERS properties of the one dimensional gold nanoparticles. SERS detection of insecticide using hybridized cyclodextrin–one dimensional gold nanoparticles with different aspect ratios are performed in aqueous solution in order to obtain an accurate evaluation of the enhancement effect for the insecticide molecule trapped by cyclodextrin associated with one dimensional gold nanoparticles with different aspect ratios because plasmon coupling from the aggregation of one dimensional gold nanoparticles can be efficiently inhibited. As mentioned above, one dimensional gold nanoparticles or hybridized one dimensional gold nanoparticles can be well dispersed (see Fig. 5) and no obvious aggregation phenomena can be observed, which indicates that the present measurements guarantee that there are no hybridized one dimensional gold nanoparticles aggregating which is detected in these SERS experiments; moreover, SERS spectra could not be convoluted with aggregation and plasmon coupling effects. In comparison with the Raman spectra of methyl parathion powder and SERS spectra in polydimethylsiloxane microfluidic channel,⁵⁶ the characteristic peaks in the present SERS spectra, i.e., 1598 cm⁻¹, 1393 cm⁻¹, 1246 cm⁻¹, 1132 cm⁻¹, 1003 cm⁻¹ and 851 cm⁻¹, could be suggested as the fingerprint vibrational bands of the methyl parathion trapped by SH-β-cyclodextrin–one dimensional gold nanoparticles. The vibrational mode at 1598 cm⁻¹ can be assigned as the phenyl ring stretch in methyl parathion, which could be reproducible from the theoretical predictions using B3LYP/6-31G* in Gaussian 03 software,⁵⁷viz., the phenyl ring stretching vibrational bands could be located at 1603 cm⁻¹ using scaled factor 0.9613.⁵⁸ As illustrated in Table 1, the strong vibrational bands at 1393 cm⁻¹ are assigned as N–O stretches, which is good agreement with the theoretical vibrational frequencies 1390 cm⁻¹. As far as C–O stretching vibration between carbon atom from phenyl ring and oxygen is concerned, the experimental vibrational band could be observed at 1246 cm⁻¹, which can be reproduced by theoretical vibrational band at 1259 cm⁻¹ with scaling factor 0.9940.⁵⁹ On the other hand, according to experimental and theoretical observations, stretching vibrations between oxygen atom and carbon atoms in methyl groups can be located at 1003 cm⁻¹. As for C–N stretching vibrational bands, the experimental and theoretical vibrational bands can be located at 1132 cm⁻¹ and 1147 cm⁻¹ (Table 1), respectively. Compared to the C–O stretch, P–O stretching vibrations in methyl parathion could be shifted towards low wavenumbers 851 cm⁻¹, which is exactly the same as the theoretical vibrational frequency 851 cm⁻¹ (Table 1). It should be mentioned that the SERS spectrum of mono-6-thio-β-cyclodextrin-one dimensional gold nanoparticles without insecticide molecules have been given in Fig. S3, ESI†, suggesting that no obvious signal can be observed at the region of fingerprint Raman active vibrational bands of the methyl parathion and disturbance from substrates of decorated one-dimensional gold nanoparticles could be ignored. On the basis of combinations of the experimental and theoretical Raman spectra, the methyl parathion insecticide is indeed captured by the hybridized mono-SH-β-cyclodextrin–one dimensional gold nanoparticles system.

Fig. 6 Schematic representation of inclusion complex of methyl parathion insecticide–mono-6-thio-β-cyclodextrin–one dimensional gold nanoparticles for SERS experiments.

Table 1 Observed and calculated fingerprint Raman wavenumbers (cm⁻¹) of methyl parathion insecticide

Vibrational description	Observed		Calculated at B3LYP/631G(d) level^a
	Powder	SERS
a Calculated vibrational wavenumbers scaled by 0.9613 (higher wavenumber) and 0.9940 (lower wavenumber).
Phenyl stretch	1596	1598	1603
N–O stretch	1373	1393	1390
C–O stretch (Phenyl-O)	1216	1246	1259
C–N stretch	1107	1132	1147
C–O stretch (CH3–O)	1039	1003	1003
P–O stretch	857	851	851

---

SERS behaviors of hybridized one dimensional gold nanoparticles with different aspect ratios for detecting methyl parathion insecticide

Adsorption could be believed to occur via the formation of a covalent bond between Au and SH of SH-β-cyclodextrin. It can be assumed that the benzene ring of methyl parathion insecticide in the cavity extends out of the one dimensional gold nanoparticles surface and its molecular axis should be aligned with the local field and perpendicular to the one dimensional gold nanoparticles. Therefore, Raman modes with a large component along the axis can be favored in comparison with the other modes, viz., the vibrational mode at 1393 cm⁻¹ are strongly uniaxial along the molecular axis as compared to the other vibrational modes. Hence, the vibrational mode is obviously strengthened in SERS spectrum and can act as sensitive fingerprint Raman peaks. The intensity of the vibrational mode at 1393 cm⁻¹ and signal-to-noise are qualitatively better for methyl parathion trapped by mono-SH-β-cyclodextrin–one dimensional gold nanoparticles with different aspect ratios 2 than that trapped by mono-SH-β-cyclodextrin–one dimensional gold nanoparticles with different aspect ratios 5 or 15, suggesting that aspect ratio ca. 2 one dimensional gold nanoparticles can provide better SERS enhancements than the other two one dimensional gold nanoparticles with aspect ratios 5 or 15. The analytical sensitivity (AS) in the SERS can be estimated by using a reference sample without SERS contribution and it can be described as AS = I_SERS/I_RS × C_RS/C_SERS, where I_SERS and I_RS denote the Raman intensities of the target molecule in the SERS and the non-SERS conditions, respectively. Considering that identical experimental conditions, e.g., laser wavelength, laser power, the same preparation conditions etc., can be rigorously controlled, the estimations of analytical sensitivity are quite intuitive and reproducible. The SERS spectra of methyl parathion at 10⁻⁸ M captured by decorated one dimensional gold nanoparticles with different aspect ratios, i.e., 2, 5 and 15, have been given in Fig. S4, ESI†. Through comparing the peak intensity of the characteristic band at 1393 cm⁻¹ with a normal Raman peak of a reference methyl parathion at 0.1 M, the analytical sensitivity in the SERS of the hybridized cyclodextrin–one dimensional gold nanoparticles with aspect ratio 2 can be estimated to be 1.11 × 10⁹ at 10⁻⁸ M; in contrast, the analytical sensitivity in the SERS from the hybridized cyclodextrin–one dimensional gold nanoparticles with aspect ratio 5 and aspect ratio 15 can be estimated to be 6.27 × 10⁸ and 2.54 × 10⁸, respectively, suggesting that the analytical sensitivity in the SERS of decorated one dimensional gold nanoparticles with aspect ratio 2 are a factor of 10 greater than for the other decorated one dimensional gold nanoparticles as SERS substrates. This remarkably high SERS enhancement with respect to the other one dimensional gold nanoparticles substrates can be ascribable to the greater electromagnetic enhancement for the one dimensional gold nanoparticles with aspect ratio 2, which possess plasmon absorption overlap with the excitation source. Although investigations on SERS using substrates with tunable plasmon resonance are limited, observed differences in enhancement factors for the one dimensional gold nanoparticles can be compared to previous theoretical or experimental studies of wavelength-dependent SERS on spherical metallic substrates. According to the previous experimental results,⁶⁰ the magnitude of electromagnetic enhancement M^EM can be estimated by

M^EM = [E^L (ω_I)/E^I (ω_I)]⁴ (1)

Where E^L and E^I mean that the total and incident electric fields, respectively, which are dependent on the incident light frequency. The total incident electric field is the sum of the incident and the induced field, E^ind, which are from the electrodynamic environment of the nanoparticle. The largest wavelength dependent contributions to M^EM are from the E^ind term,

E^L (ω_I) = E^I (ω_I) + E^ind (ω_I) (2)

E^ind ∝ (ω/c)² (ε₂ − ε₁) (3)

where the magnitude of E^ind is dependent on the dielectric constant of the metal nanoparticle, ε₁ and that of the surrounding medium, ε₂. According to the model for analyte molecules absorbed on the surface of gold nanoparticles, simulated M^EM values can reach a maximum of ca. 10³ for excitation energies at the plasmon resonant wavelength (ca. 520 nm), but lower than 10 at 400 nm and ca. 100 for incident wavelength of 700–1200 nm. According to the theoretical simulations, the M^EM value for gold SERS substrates with plasmon bands in resonance with the incident radiation could be 10–100 times greater than those without overlapping bands, which are good agreement with analytical sensitivity for different hybridized cyclodextrin–one dimensional gold nanoparticles with different aspect ratios in detecting methyl parathion insecticide.

Additionally, it should be mentioned that the present analytical sensitivity in the SERS is quite high, which leads to ultrasensitive detection of insecticide compared to the most recent report using simply direct absorption of herbicide at 3 × 10⁻⁷ M level on one dimensional gold nanoparticles.⁶¹ It can be partly correlated with the condensation effect of the insecticide in the cyclodextrin cage, i.e., the target insecticide molecules are located in the cyclrodextrin with much higher density than in solution, obviously larger signals can be detected in addition to the SERS effect for each molecule.

Sensitivity for detection of methyl parathion insecticide using hybridized one dimensional gold nanoparticles with aspect ratio 2

In order to evaluate the sensitivity of the SERS detection, different concentrations of methyl parathion insecticide ranging from 10⁻⁶ to 10⁻¹² M can be analyzed using hybridized one dimensional gold nanoparticles with aspect ratio 2. As shown in Fig. 7, the SERS intensity of vibrational bands at 1393 cm⁻¹ (N–O stretching vibration), which is quite high sensitive to the concentration of methyl parathion insecticide, suggests that the present detection limit can be as low as ppt level. Hence, the detection limit in our SERS probe is obviously lower than the previous reported results, which are correlated with three factors. Firstly, as mentioned above, the primary enhancement in SERS can be attributed to electromagnetic effect and electromagnetic field decrease in strength with distance from the point source. Therefore, analyte molecules can benefit from the enhancement of SERS-active substrate even if the analyte are some distance away from the enhancing surface. According to previous reports, the enhancement decrease ten fold with a distance of 2–3 nm.^62–64 As far as mono-6-thio-β-cyclodextrin decoration on the one dimensional gold nanoparticles are concerned, the distance between the insecticide molecule and the surface of one dimensional gold nanoparticles can be efficiently shortened without long tethered alkyl chain, which contributes to maximizing SERS effect. Secondly, based on the fact that the enhancement has a strong dependence on the coupling degree between the longitudinal surface plasmon resonance and the incident line, the present overlapping can be tuned through changing the aspect ratio of the one dimensional gold nanoparticles, viz., the aspect ratio 2 of decorated one dimensional gold nanoparticles with SPR at 654 nm and the excitation wavelength at 632.8 nm, which can maximize the enhancement. Finally, the cavity diameter of β-cyclodextrin is ca. 0.62 nm, which can be well matched with the size (0.61 nm) of methyl parathion insecticide and efficiently capture the analyte. Moreover, the complex stability constant log K between β-cyclodextrin and methyl parathion insecticide is measured as 3.75 ± 0.23 via a microcalormetric titration experiment, indicating that the interaction between the insecticide and cavity of β-cyclodextrin is enough to efficiently capture methyl parathion.

Fig. 7 (A) SERS spectra of methyl parathion insecticide at different concentration from 10⁻¹² M to 10⁻⁶ M. (B) Plot demonstration how Raman intensity at 1393 cm⁻¹ changes upon addition of different concentration of methyl parathion insecticide.

Selectivity for detection of methyl parathion insecticide using hybridized one dimensional gold nanoparticles with aspect ratio 2

In order to investigate the selectivity of a hybridized system for detecting methyl parathion insecticide by SERS, some other persistent organic pollutants, e.g., mirex, phthalocyanine, hydroquinol, resorcinol and 1,3-phenylenediamine etc. have been selected. The present results reveal that no Raman signal can be observed in the presence of these pollutants, which implies that good selectivity for detecting methyl parathion using hybridized cyclodextrin–one dimensional gold nanoparticles in the SERS experiments (shown in Fig. 8). Because host–guest interactions through the cavity lead to the formation of inclusion complexes within the cavity with guest molecules smaller than the cavity size, some pollutants, e.g., phthalocyanine, whose size is larger than the cavity, can not be efficiently captured by the present hybridized system so that obvious SERS signal can not be observed. On the other hand, due to driving forces for the formation of the cyclodextrin inclusion complex arising from apolar–apolar interactions between guest molecules and host cavities, some polar pollutants, e.g., 1,3-phenylenediamine can not be effectively trapped by the cavity, which lead to unobservable SERS signals. Furthermore, different representative vibrational modes could be observed in the different pollutant molecules and it can be expected that our SERS probes can be applied for the multiplex detection of insecticide from environment sample.

Fig. 8 Selectivity of the present SERS detection on the other pollutants (10⁻⁶ M) compared to methyl parathion insecticide (10⁻⁸ M).

Conclusions

In conclusion, a simple, cheap and ultrasensitve detection of the most common insecticide methyl parathion via SERS approach based on hybridized mono-6-thio-β-cyclodextrin-one dimensional gold nanoparticles without Raman label, have been demonstrated in the present work for the first time. Different mono-6-thio-β-cyclodextrin decorated the one dimensional gold nanoparticles with different aspect ratios 2, 5 and 15 have been selected for detecting insecticide using SERS approaches, suggesting that enhancement from one dimensional gold nanoparticles with aspect ratio 2 has excellent detection capability compared to the one dimensional gold nanoparticles with aspect ratio 5 or 15. Furthermore, the one dimensional gold nanoparticles with aspect ratio 2 have been selected as SERS substrate for detecting the insecticide and the detection limit could be qualified as picomolar level. Compared to some other methods for the trace analysis of insecticide methyl parathion, the present detection capability has been increased several orders of magnitude. It can be expected that the new simple, cheap and reliable SERS analysis for detection of insecticides from environmental sample, which is based on fingerprint Raman peaks, could be widely applied in the future.

Acknowledgements

This work is supported by National Basic Research Program of China (2007CB936603), Innovation project of Chinese Academy of Sciences (KSCX2-YW-G-058) and Innovation fund for talented personnel of Anhui Province (2009Z035).

References

1. X. X. Han, Y. Kitahama, T. Itoh, C. X. Wang, B. Zhao and Y. Ozaki, Anal. Chem., 2009, 81, 3350.
2. H. Ko and V. V. Tsukruk, Small, 2008, 4, 1980.
3. R. A. Tripp, R. A. Dluhy and Y. P. Zhao, Nano Today, 2008, 3, 31.
4. W. Wang, C. M. Ruan and B. H. Gu, Anal. Chim. Acta, 2006, 567, 121.
5. R. A. Álvarez-Puebla, R. Contreras-Cáceres, I. Pastoriza-Santos, J. Pérez-Juste and L. M. Liz-Marzán, Angew. Chem., Int. Ed., 2009, 48, 138.
6. R. M. Jarvis, A. Brooker and R. Goodacre, Faraday Discuss., 2006, 132, 281.
7. H. Ko, S. Chang and V. V. Tsukruk, ACS Nano, 2009, 3, 181.
8. W. Xie, L. Wang, Y. Y. Zhang, L. Su, A. Shen, J. Q. Tan and J. M. Hu, Bioconjugate Chem., 2009, 20, 768.
9. L. L. Sun, Y. H. Song, L. Wang, C. L. Guo, Y. J. Sun, Z. L. Liu and Z. Li, J. Phys. Chem. C, 2008, 112, 1415.
10. R. A. Alvarez-Puebla, D. S. Dos Santos Jr. and R. F. Aroca, Analyst, 2007, 132, 1210.
11. M. Y. Sha, H. X. Xu, S. G. Penn and R. Cromer, Nanomedicine, 2007, 2, 725.
12. A. D. McFarland, M. A. Young, J. A. Dieringer and R. P. Van Duyne, J. Phys. Chem. B, 2005, 109, 11279.
13. A. Otto, I. Mrozek and C. Pettenkofer, Surf. Sci., 1990, 238, 192.
14. Y. L. Wang, W. Ren, S. J. Dong and E. K. Wang, J. Raman Spectrosc., 2009, 40, 571.
15. R. Sato-Berrú, R. Redón, A. Vázquez-Olmos and J. M. Saniger, J. Raman Spectrosc., 2009, 40, 376.
16. A. Musumeci, D. Gosztola, T. Schiller, N. M. Dimitrijevic, V. Mujica, D. Matrin and T. Rajh, J. Am. Chem. Soc., 2009, 131, 6040.
17. M. Sardo, C. Ruano, J. L. Castro, I. López-Tocón, P. J. A. Ribeiro-Claro and J. C. Otero, Chem. Phys. Lett., 2008, 467, 101.
18. L. Y. Cao, P. Diao, L. M. Tong, T. Zhu and Z. F. Liu, ChemPhysChem, 2005, 6, 913.
19. J. A. Dieringer, K. L. Wustholz, D. J. Masiello, J. P. Camden, S. L. Kleinman, G. C. Schatz and R. P. Van Duyen, J. Am. Chem. Soc., 2009, 131, 849.
20. M. Z. Si, Y. P. Kang and Z. G. Zhang, J. Raman Spectrosc., 2009, 40, 1319.
21. P. C. Wu, C. G. Khoury, T. H. Kim, Y. Yang, M. Losurdo, G. V. Bianco, T. Vo-Dinh, A. S. Brown and H. O. Everitt, J. Am. Chem. Soc., 2009, 131, 12032.
22. J. Zhao, L. Jensen, J. H. Sung, S. L. Zou, G. C. Schatz and R. P. Van Duyne, J. Am. Chem. Soc., 2007, 129, 7647.
23. P. Zhang and Y. Y. Guo, J. Am. Chem. Soc., 2009, 131, 3808.
24. R. A. Alvarez-Puebla and R. F. Aroca, Anal. Chem., 2009, 81, 2280.
25. L. O. Brown and S. K. Doorn, Langmuir, 2008, 24, 2277.
26. J. Kneipp, X. T. Li, M. Sherwood, U. Panne, H. Kneipp, M. I. Stockman and K. Kennipp, Anal. Chem., 2008, 80, 4247.
27. H. J. Chen, Y. L. Wang and S. J. Dong, J. Raman Spectrosc., 2009, 40, 1188.
28. E. N. Esenturk and A. R. H. Walker, J. Raman Spectrosc., 2009, 40, 86.
29. R. Sanci and M. Volkan, Sens. Actuators, B, 2009, 139, 150.
30. S. H. Ciou, Y. W. Cao, H. C. Huang, D. Y. Su and C. L. Huang, J. Phys. Chem. C, 2009, 113, 9520.
31. B. K. Jena and C. R. Raj, Chem. Mater., 2008, 20, 3546.
32. L. Cui, A. Wang, D. Y. Wu, B. Ren and Z. Q. Tian, J. Phys. Chem. C, 2008, 112, 17618.
33. P. N. Sisco and C. J. Murphy, J. Phys. Chem. A, 2009, 113, 3973.
34. G. V. Maltzahn, A. Centrone, J. H. Park, R. Ramanathan, M. J. Sailor, T. A. Hatton and S. N. Bhatia, Adv. Mater., 2009, 21, 3175.
35. H. I. Peng, C. M. Strohsahl, K. E. Leach, T. D. Krauss and B. L. Miller, ACS Nano, 2009, 3, 2265.
36. B. Nikoobakht and M. A. El-Sayed, J. Phys. Chem. A, 2003, 107, 3372.
37. M. C. Pablos Espana, F. J. Arrebola-Liebnas, A. Garrido-Frenich and J. L. Martinez-Vidal, Int. J. Environ. Anal. Chem., 1999, 75, 165.
38. Collaborative International Insecticide Analytical Council (CIPAC) Handbook, CIPAC Ltd., 1994, vol. 1B, p. 1849.
39. C. Goncalves and M. F. Alpendurada, J. Chromatogr., A, 2004, 1026, 239.
40. A. L. Hart, W. A. Collier and D. Janssen, Biosens. Bioelectron., 1997, 12, 645.
41. A. Di Muccio, I. Camoni, M. Ventriglia, D. A. Barbini, M. Mauro, P. Pelosi, T. Generali, A. Ausili and S. Girolimetti, J. Chromatogr., A, 1995, 697, 145.
42. E. Paradopoulou-Mourkidou and J. J. Patsias, J. Chromatogr., A, 1996, 726, 99.
43. A. I. Valenzuela, R. Lorenzini, M. J. Redondo and G. Font, J. Chromatogr., A, 1999, 839, 101.
44. A. M. Stockton, T. N. Chiesl, J. R. Scherer and R. A. Mathies, Anal. Chem., 2009, 81, 790.
45. A. Neely, C. Perry, B. Varisli, A. K. Singh, T. Arbneshi, D. Senapati, J. R. Kalluri and P. C. Ray, ACS Nano, 2009, 3, 2834.
46. W. L. Daniel, M. S. Han, J. S. Lee and C. A. Mirkin, J. Am. Chem. Soc., 2009, 131, 6362.
47. J. Szejtli, Chem. Rev., 1998, 98, 1743.
48. K. A. Connors, Chem. Rev., 1997, 97, 1325.
49. A. P. Alivisatos and C. Sonnichsen, Nano Lett., 2005, 5, 301.
50. C. J. Murphy, T. K. Sau, A. M. Gole, C. J. Orendroff, J. Gao, L. Gou, S. E. Hunyadi and T. Li, J. Phys. Chem. B, 2005, 109, 13857.
51. A. Gole and C. J. Murphy, Chem. Mater., 2004, 16, 3633.
52. N. R. Jana, L. Gearheart and C. J. Murphy, Adv. Mater., 2001, 13, 1389.
53. D. Vizitiu, C. S. Walkinshaw, B. I. Gorin and G. R. J. Thatcher, J. Org. Chem., 1997, 62, 8760.
54. M. T. Rojas, R. Königer, J. F. Fraser Stoddart and A. E. Kaifer, J. Am. Chem. Soc., 1995, 117, 336.
55. A. Wijaya and Hamad-Schifferi, Langmuir, 2008, 24, 9966.
56. D. Lee, S. Lee, G. H. Seong, J. Choo, E. K. Lee, D. G. Gweon and S. Lee, Appl. Spectrosc., 2006, 60, 373.
57. M. J. Frisch et. al. Gaussian 03, Revision E.01, Gaussian, Inc.: Wallingford, CT, USA 2004.
58. (a) M. W. Wong, Chem. Phys. Lett., 1996, 256, 391; (b) J. A. Pople, H. B. Schlegel, R. Krishnan, D. J. Dfrees, J. S. Binkley, M. J. Frish, R. A. Whiteside, R. H. Hout and W. J. Hehre, Int. J. Quant. Chem. Symp., 1981, 15, 269; (c) H. A. Jiménez-Vázquez and J. Tamariz, J. Phys. Chem. A, 2001, 105, 1315; (d) S. K. Ignatov, P. G. Sennikov, A. G. Razuvaev, L. A. Chuprov, O. Schrems and B. S. Ault, J. Phys. Chem. A, 2003, 107, 8705; (e) P. Blowers, K. F. Tetrault and Y. Trujillo-Morehead, Theor. Chem. Acc., 2008, 119, 369.
59. D. J. Clouthier and D. L. Joo, J. Chem. Phys., 1997, 106, 7479.
60. H. X. Xu, J. Aizpurua, M. Kall and P. Apell, Phys. Rev. E: Stat. Phys., Plasmas, Fluids, Relat. Interdiscip. Top., 2000, 62, 4318.
61. J. C. S. Costa, R. A. Ando, A. C. Santana, L. M. Rossi, P. S. Santos, M. L. A. Temperini and P. Corio, Phys. Chem. Chem. Phys., 2009, 11, 7491.
62. A. Campion and P. Kambhampati, Chem. Soc. Rev., 1998, 27, 241.
63. K. Kneipp, H. Kneipp, I. Itzkan, R. R. Dasari and M. S. Feld, Chem. Rev., 1999, 99, 2957.
64. M. J. Weaver, S. Zou and H. Y. H. Chan, Anal. Chem., 2000, 72, 38 A.

---

Footnotes

† Electronic supplementary information (ESI) available: Gel electrophoresis, UV-vis and SERS spectra. See DOI: 10.1039/c0jm00040j

‡ These authors contributed equally to this work.

---