- Email: [email protected]
Simultaneous determination of phenylurea herbicides in yam by capillary electrophoresis with electrochemiluminescence detection
Journal of Chromatography B, 986–987 (2015) 143–148
Contents lists available at ScienceDirect
Journal of Chromatography B journal homepage: www.elsevier.com/locate/chromb
Simultaneous determination of phenylurea herbicides in yam by capillary electrophoresis with electrochemiluminescence detection Yuefang Hu ∗ College of Chemistry and Bioengineering, Hezhou University, Hezhou 542899, Guangxi, China
a r t i c l e
i n f o
Article history: Received 12 August 2014 Accepted 9 February 2015 Available online 17 February 2015 Keywords: Capillary electrophoresis Electrochemiluminescence Monuron Monolinuron Diuron Yam
a b s t r a c t A method of capillary electrophoresis (CE) coupled with electrochemiluminescence (ECL) detection has been applied to detect three major phenylurea herbicides (monuron, monolinuron and diuron) simultaneously. The effects of yam sample preparation, injection voltage and time, detection potential, detection buffer concentration and pH, Ru(bpy)3 2+ concentration, separation buffer type, separation buffer pH and concentration, separation voltage were investigated in detail. Under optimum conditions, a good baseline separation and highly sensitive detection for monuron, monolinuron and diuron were achieved. The ECL intensity (I) was in proportion to three analytes concentration () in the range of 0.1–10,000 g/L for monuron (r ≥ 0.9993), 0.1–18,000 g/L for monolinuron (r ≥ 0.9995) and 0.1–20,000 g/L for diuron (r ≥ 0.9997). The detection limits for monuron, monolinuron and diuron were 0.05, 0.04 and 0.01 g/L (S/N = 3), respectively. The developed method was successfully applied for the analysis of monuron, monolinuron and diuron residues in yam simultaneously. The average recoveries are in the ranges of 90.0–99.2% with relative standard deviations less than 3.2%. The limits of detection of the proposed method were 0.010 g/kg for monuron, 0.008 g/kg for monolinuron and diuron in yam. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Phenylurea herbicides, a kind of pesticides which includes monuron, monolinuron, diuron and metobromuron, etc. has been widely used to suppress broadleaf and grass weeds in China and other countries [1]. It was reported that phenylurea herbicides have chronic toxic, and its residues and degradation products showed certain activities in crops, soil and water, which brought potential risks to the health of human beings by environmental accumulation and food chain. For example, it has been reported that people ate the foodstuff with excessive monuron and diuron would cause carcinogenic [2], which has already caused people’s high attention. Therefore, some countries have set the maximum residue limit (MRL) of phenylurea herbicides in foodstuff. Yam is one of the Medicine-Food agricultural products with high nutritional value, with people’s awareness of health more and more strengthen, the demand for Yam are getting higher and higher. Therefore, in order to guarantee people’s health, and to provide technical support for the effective monitoring of pesticide residues in agricultural products, it is highly desirable to develop a simple
∗ Tel.: +86 13635079823. E-mail address: [email protected] http://dx.doi.org/10.1016/j.jchromb.2015.02.016 1570-0232/© 2015 Elsevier B.V. All rights reserved.
and effective method to monitor residues of phenylurea herbicides in the agricultural samples. Analytical methods for the quantification of phenylurea herbicide residues are mainly based on the chromatographic analysis, such as high performance liquid chromatography (HPLC) [3,4], high performance liquid chromatography mass spectrometry-mass spectrometry (HPLC-MS/MS) [5], liquid chromatography time-offlight mass spectrometry (LC/TOF-MS) [6], gas chromatography mass spectrometry (GC-MS) [7], thin layer chromatography (TLC) [8–10], immunoassay technique [11,12], photo-induced fluorescence (PIF) [13], graphene reinforced hollow fiber liquid phase microextraction (G-HF-LPME) [14], etc. However, these regular methods generally are high operation cost and long analytical time. Besides, the sample pretreatment procedures are rather cumbersome and expensive, which greatly diminished them application. Capillary electrophoresis (CE) has been attracted many researcher’s attention because of its advantage of high separation efficiency, short separation time, low reagent consumption, and ease of installation [15,16]. The features of the detector have great effect on the overall analytical efficiency for CE due to the small size of the separation capillaries and seriously low sample quantities injected into the capillaries. Actually, the detector should provide detection limits as low as possible without affecting the quality of separation. ECL is the procedure based on species generated at electrodes undergo high-energy electron transfer reactions
144
Y. Hu / J. Chromatogr. B 986–987 (2015) 143–148
to form excited states that emit light [17]. Nowadays, it has been widely used as a highly sensitive detection technique with its high sensitivity, wide linear range, and simple instrumentation. So, CE combining with ECL (CE-ECL) has been developed for the determination of various analytes containing tertiary amines group and their derivatives for a long time [18,19]. At present, it has become one of the high technology detection of pesticide residues, because of its advantages of high separation efficiency, good repeatability, short analysis time, high sensitivity and low solvent consumption, etc. For example, Chiu et al. [20] using 20 mM sodium phosphate (pH 9.0) and a separation voltage of 21 kV, working electrode at 1.6 V in a 300 mM sodium phosphate buffer (pH 8.0) containing 3.5 mM Ru(bpy)3 2+ , have been separated GLY and AMPA by CEECL detection, the limits of detection (LODs) for GLY and AMPA in water were 0.06 and 4.04 g mL−1 , respectively. Liu et al. [21] have used the CE-ECL to analyze mefenacet, the detection limit of 4.0 × 10−9 M, and was also applied to analyze residues of mefenacet in seedling and soil. The optimized conditions were as follows: potential applied at 1.2 V, injection time and injection voltage was 10 s and 11 kV, running buffer pH value was 7.38. Cai et al. [22] have determined pretilachlor in soil and rice using MSPD extraction by CE with field amplified sample injection and ECL detection, the limits of detection of the proposed method were 0.01 mg/kg in rice matrix and 0.008 mg/kg in soil matrix. The optimized conditions were as follows: potential applied at 1.17 V, injection time and voltage was 9 s and 8 kV, respectively, running buffer concentration and pH was 50 mmol L−1 and 7.38, respectively, buffer pH in detection cell was 8.5, separation voltage was 10 kV. However, this method seldom apply to analysis three kinds of pesticide residues simultaneously, and cannot give full play to the advantages and application potential of this high technology coupling technique. On the other hand, the influencing factors of the above application were investigated not detailed enough, and so far, the CE-ECL method has not been reported for determination of monuron, monolinuron and diuron simultaneously, and no application to yam sample was demonstrated. In this study, a simple and rapid CE-ECL method was adopted to determine phenylurea herbicides residues (monuron, monolinuron and diuron) simultaneously, and the matrix solid-phase dispersion method (MSPD) was developed to handle sample before determining [23]. Influencing factors, such as yam sample preparation, injection voltage and time, detection potential, detection buffer concentration and pH, Ru(bpy)3 2+ concentration, separation buffer type, separation buffer pH and concentration, separation voltage were investigated in detail. Residues of monuron, monolinuron and diuron in yam sample were also determined using the proposed technique. It was expected to be able to provide some theoretical basis for the quality evaluation and judgment of yam, and can provide some effective technical support for the agricultural product quality safety testing. 2. Experimental 2.1. Apparatus and reagents An MPI-A CE-ECL system was produced by Remax Electronic Science-Tech Co. Ltd. (Xi’an, China); a 42 cm length of 25 m i.d. uncoated fused-silica capillary was used (Yongnian Optical Fabric Factory, Hebei, China); ECL detection was employed using a three-electrode system: a 300 m diameter Pt disk as the working electrode, a Pt wire as the counter electrode, and a Ag/AgCl electrode (in saturated KCl solution) as the reference electrode. Tris(2,2-bipyridyl) ruthenium(II) chloride hexahydrate [Ru(bpy)3 Cl2 -6H2 O], and monuron, monolinuron and diuron standard compounds were obtained from Bailingwei Technology Company (Beijing, China). Yam was purchased from a local
pharmacy. 0.22 m membrane filters was purchased from Xinya Purification Material Factory (Shanghai, China). All water used throughout the study was double-distilled water, and reagents and chemicals used were analytical grade. 2.2. Yam sample preparation MSPD was developed to handle sample [23]: yam was weighted accurately after drying at 55 ◦ C for 4 h by oven, and grounding into powders for samples. 0.5 g sample (blank or spiked) was blended with 2.5 g dried Florisil material, and then ground together for 12 min to obtain a uniform mixture. Next, the mixture was transferred to a 10 mL syringe barrel covered in turn with a filter, anhydrous alumina at the bottom. Then, 3 cm height of anhydrous sodium sulfate and another filter were placed on the top of the mixture and then the sample mixture was compressed tightly. After successive elution with three aliquots of ethyl acetate (5 mL) for 4 min, the solutions were combined together and evaporated to dryness under a nitrogen stream. The residue was then dissolved in 100 mL 70% ethanol and soaked at room temperature for 24 h. Finally, 25 mM phosphate buffer solution (pH 8. 0) had been added and was volumed to 250 mL. Before analysis, all the yam sample were filtered through 0.22 m membrane filters. 2.3. Experimental methods Before the experiment, 350 L volume of 5 mM Ru(bpy)3 2+ and 45 mM phosphate buffer at pH 7.5 was injected into the detection reservoir, 25 mM phosphate buffer at pH 8.0 was used as separation buffer, 16 kV voltage was used as separation voltage, electrokinetic injection 10 s at 10 kV was used for sample introduction and detection potential at 1.13 V was applied at the working electrode. The biased potential of photomultiplier tube was set at 800 V. Before use, the working electrode was polished with alumina powder (0.3 and 0.05 m in turn) until a mirror-smooth surface appeared. Prior to CE injection, all the solutions were filtered through membrane filters. Then the injector end of capillary was put into phosphate buffer and electrophoresis voltage was applied, and the ECL analyzer was opened at the same time. In order to obtain accurate result, electrokinetic injection, separation and detection were began after the baseline of intensity signal to be stable (about 10 s). 3. Results and discussion 3.1. ECL behaviors of monuron, monolinuron and diuron The corresponding ECL curves of Ru(bpy)3 2+ , monuron, monolinuron and diuron were shown in Fig. 1, when a Pt electrode in 45 mM PBS (pH 7.5) with 5 mM Ru(bpy)3 2+ in the potential range from 0 to 1.3 V, only relatively weak ECL signal could be observed (Fig. 1(a0 )). With the addition of the same concentration of monuron, monolinuron and diuron, respectively. As is clearly shown in Fig. 1, the ECL intensity of three analytes increased remarkably, and the ECL intensity of diuron (Fig. 1(c)) is a little stronger than monuron (Fig. 1(a)) and monolinuron (Fig. 1(b)) at the same concentration. 3.2. Optimization of the yam sample preparation The yam samples were handled based on the MSPD method that have been reported [23]. Firstly, Florisil was chosen as the dispersant because it led to a cleaner chromatographic profile with lower baseline than C8 , C18 and silica. Various mass ratios of the dispersant to the sample matrix (0.5 g) from 7:1 to 3:1 were assayed with
Y. Hu / J. Chromatogr. B 986–987 (2015) 143–148
Fig. 1. ECL behaviors of monuron (a), monolinuron (b) and diuron (c). (a0 ) 45 mM PBS (pH 7.5) +5 mM Ru(bpy)3 2+ ; (a) a0 + 0.500 mg/L monuron; (b) a0 + 0.500 mg/L monolinuron; (c) a0 + 0.500 mg/L diuron.
10 mL ethyl acetate as the elution solvent. The mass ratio of 5:1 provided the best recoveries for three phenylurea herbicides. Secondly, four organic solvents including ethyl acetate, dichloromethane, methanol and acetone were used as the extractants to study their effect on recoveries with the mass ratios of the dispersant to the sample matrix as 5:1. It was indicated that high recoveries (>90%) were obtained with ethyl acetate and dichloromethane, and ethyl acetate was selected due to its low toxicity. The influence of ethyl acetate volume (6–18 mL) on the recoveries was also investigated, and the result shown that the recoveries reached maxima when the eluting volume was 15 mL. So, 15 mL ethyl acetate was chosen in subsequent experiments.
3.3. Optimization of injection conditions The effect of the injection voltage and time on the theoretical plates number (N) and the ECL intensity of monuron, monolinuron and diuron were investigated. The theoretical plates number was calculated by the equation: N = 5.54(tR /Wh )2 , where tR and Wh are the migration time and the width at half the maximum peak height of the analytes, respectively. The effect of injection voltage on the theoretical plates number and ECL intensity in the range of 2–18 kV as injection time set at 10 s was shown in Fig. 2, It was seen when a lower injection voltage was used, it was difficult to obtain strong intensity. However, the theoretical plates number decreased and the ECL intensity increased at high injection voltage. The probable reason is that at high injection voltage, more analytes enter into the diffusion layer and led to strong ECL intensity, but more analytes came into the detection cell resulting the analytes cannot reach the electrode surface quickly and diffuse into the solution, so the peak is retarded and broadened, and led to the theoretical plates number was decreased, so, in order to obtain high ECL intensity and theoretical plates number,10 kV was selected as the optimum injection voltage. Fixing the injection voltage at 10 kV, when the injection time changed from 2 to 10 s, the ECL intensity increased correspondingly. Over 10 s, the ECL intensity increased slowly. Furthermore, because of an excessively large injection volume and overloading occurred, the longer injection time broadened the peak shape and led to theoretical plates number decreased. As a compromise, 10 s was selected as the optimum injection time in subsequent experiments.
145
Fig. 2. The effect of injection voltage on 0.500 mg/L monuron, monolinuron and diuron theoretical plates number and ECL intensity. Conditions: separation capillary, 42 cm length (25 m i.d.); detection potential, 1.13 V; separation buffer, 25 mM phosphate buffer pH 8.0; separation voltage, 16 kV; in ECL cell: 5 mM Ru(bpy)3 2+ , phosphate buffer pH 7.5; injection time: 10 s. a, b, and c are the ECL intensity of monuron, monolinuron and diuron, respectively; a1 , b1 , and c1 are the theoretical plates number of monuron, monolinuron, and diuron, respectively.
3.4. Optimization of detection conditions 3.4.1. Effect of detection potential The optimized detection potential had a great influence on the ECL signal. As shown in Fig. 1, when the detection potential was less than 1.00 V, the ECL responses of monuron, monolinuron and diuron were relatively low and were difficult to be observed, it was indicated that Ru(bpy)3 2+ had not happened oxidation reaction on the surface of Pt electrode at these potential. When the detection potential was higher than 1.00 V, monuron, monolinuron and diuron ECL intensity increased and showed a maximum value at 1.13 V, then decreased slightly exceeded 1.13 V. Therefore, to obtain a high ECL intensity and high efficiency, 1.13 V was selected as the optimum detection potential in subsequent experiments. 3.4.2. Effect of the detection buffer concentration The effect of phosphate buffer concentration in the cell on ECL intensity of monuron, monolinuron and diuron was investigated in the concentration range from 25 to 60 mM. As can be seen from Fig. 3, at the beginning, the ECL intensity of monuron, monolinuron and diuron increased with the increase of phosphate buffer concentration, and the highest ECL intensity was obtained at 45 mM, when the concentration was exceeded 45 mM, three analytes ECL intensity decreased. In addition, when phosphate buffer concentration exceeded 45 mM, it gave higher baseline noise in experiment. Thus, a 45 mM detection phosphate buffer was chosen in further experiment. 3.4.3. Effect of the detection buffer pH It was reported that the ECL reaction of tertiary amine with Ru(bpy)3 2+ based on the buffer pH to a large extent [24], so the effect of detection buffer pH on ECL intensity was investigated within the range 6.0–10.0. As illustrated in Fig. 4, the ECL intensity of monuron, monolinuron and diuron increased quickly with increasing pH, and to a maximum at pH 7.5. However, the ECL intensity was decreased slightly when the pH value exceeded 7.5. It can be explained that the competition of the reaction of Ru(bpy)3 3+ and OH−1 ions in higher pH value [24], thus, when the pH value was higher than 8.0, the noise of baseline was very large. So, a 7.5 pH value was chosen as the optimized pH in further experiment.
146
Y. Hu / J. Chromatogr. B 986–987 (2015) 143–148
Fig. 3. The effect of detection buffer concentration on 0.500 mg/L monuron (a), monolinuron (b) and diuron (c) ECL intensity. Conditions: separation capillary, 42 cm length (25 m i.d.); detection potential, 1.13 V; separation buffer, 25 mM phosphate buffer pH 8.0; separation voltage, 16 kV; in ECL cell: 5 mM Ru(bpy)3 2+ , phosphate buffer pH 7.5; injection voltage and time: 10 kV and 10 s.
3.4.4. Effect of Ru(bpy)3 2+ concentration The ECL intensity is strongly influenced by the concentration of Ru(bpy)3 2+ that used as the ECL reagent in the system. The ECL intensity increased markedly with increasing Ru(bpy)3 2+ concentration due to the generation of more amount of electrogenerated excited species Ru(bpy)3 2+ , but the more noise was produced and the background current increased significantly when its concentration exceeded 5 mM. In addition, too high concentration was required more consumption of the expensive reagent Ru(bpy)3 2+ . So, in order to obtain higher separation efficiency, better signalto-noise ratio, stronger ECL intensity of three analytes and less consumption of reagent, 5 mM Ru(bpy)3 2+ was selected. 3.5. Optimization of separation conditions 3.5.1. Effect of separation buffer type It was known that separation buffer has great influence on separation and detection. In this study, phosphate buffer, Tris–HCl, and
Fig. 4. The effect of detection buffer pH on 0.500 mg/L monuron (a), monolinuron (b) and diuron (c) ECL intensity. Conditions: separation capillary, 42 cm length (25 m i.d.); detection potential, 1.13 V; separation buffer, 25 mM phosphate buffer pH 8.0; separation voltage, 16 kV; in ECL cell: 5 mM Ru(bpy)3 2+ , 45 mM phosphate buffer; injection voltage and time: 10 kV and 10 s.
Fig. 5. The effect of separation buffer pH on 0.500 mg/L monuron (a), monolinuron (b) and diuron (c) ECL intensity and resolutions (a0 , a/b Rs , b0 , b/c Rs ). Conditions: separation capillary, 42 cm length (25 m i.d.); detection potential, 1.13 V; separation buffer, 25 mM phosphate buffer; separation voltage, 16 kV; in ECL cell: 5 mM Ru(bpy)3 2+ , 45 mM phosphate buffer pH 7.5; injection voltage and time: 10 kV and 10 s.
borate buffer were investigated. It was indicated that phosphate buffer can obtain good resolution. In addition, the ECL intensity stability and the signal-to-noise ratio of three analytes were the best in this buffer system. Therefore, phosphate buffer was selected as separation buffer. 3.5.2. Effect of the separation buffer pH The pH of separation buffer has great influence on the electroosmotic flow (EOF) of capillary and the extent of ionization of analytes, so it affects the ECL intensity, the resolution as well as the sensitivity of the analytes. As can be seen from Fig. 5, the ECL intensity (a, b, c)and resolutions (a0 , b0 )increased as pH changed from 6.0 to 8.0, and then decreased when the pH value was higher than 8.0. The results also indicated that when pH was 8.0, the highest ECL intensity and best resolutions of monuron, monolinuron and diuron were obtained. Therefore, 8.0 of the pH value was selected. 3.5.3. Effect of the separation buffer concentration The effect of separation buffer concentration in the range 10–50 mM on monuron, monolinuron and diuron ECL intensity and resolutions was also studied. It can be seen from Fig. 6, the highest ECL intensity was produced when the separation buffer concentration was 25 mM; when the separation concentration was above 25 mM, the migration time of three analytes was prolonged gradually and the Joule heating was become stronger, which increased the baseline noises, even produced air bladder in the capillary, and led to the intensity (a, b, c) and the resolutions (a0 , b0 ) of three analytes decreased. This phenomenon was largely due to the effect of the ionic strength increase with high separation concentration, which resulted in the increase Joule heating, EOF decreased and the migration time prolonged. Therefore, in order to get the best resolutions and the highest intensity, the optimized concentration of separation buffer was selected as 25 mM. 3.5.4. Effect of the separation voltage The separation voltage has a great effects on the quality of the separation and the migration time of analytes. In order to get the optimum condition, different voltages were applied over the range of 12–20 kV. The effect of the separation voltage on the ECL intensity of monuron, monolinuron and diuron was investigated. The ECL intensity was enhanced with increasing the separation voltage in the range of 12–16 kV. However, when separation voltage was
Y. Hu / J. Chromatogr. B 986–987 (2015) 143–148
147
Table 1 Regression equations, linear range, detection limit for monuron, monolinuron and diuron (n = 6). Analyte
Regression equation
Correlation coefficient
Linear range (g/L)
Detection limit (g/L)
Monuron Monolinuron Diuron
I = 937.12+314.54 I = 935.65+365.16 I = 936.17+462.28
0.9993 0.9995 0.9997
0.1–10,000 0.1–18,000 0.1–20,000
0.05 0.04 0.01
RSD (%) Migration time
Intensity
1.6 1.9 1.8
2.3 2.6 2.2
I: ECL intensity; : analyte concentration.
Fig. 6. The effect of separation buffer concentration on 0.500 mg/L monuron (a), monolinuron (b) and diuron (c) ECL intensity and resolutions (a0 , a/b Rs , b0 , b/c Rs ). Conditions: separation capillary, 42 cm length (25 m i.d.); detection potential, 1.13 V; separation buffer, phosphate buffer pH 8.0; separation voltage, 16 kV; in ECL cell: 5 mM Ru(bpy)3 2+ , 45 mM phosphate buffer pH 7.5; injection voltage and time: 10 kV and 10 s.
more than 16 kV, the baseline noise was increased. What is more, the increasing joule heat in the capillary caused peak broadening and adversely affected the separation effectiveness. So 16 kV was chosen as the best separation voltage in this experiment. 3.6. Linearity, precision, detection limit of monuron, monolinuron and diuron Under the optimized conditions: ECL detection potential at 1.13 V, 45 mM phosphate buffer at pH 7.5, 5 mM Ru(bpy)3 2+ in the detection reservoir, and 25 mM phosphate buffer at pH 8.0 as separation buffer, separation voltage at 16 kV, electrokinetic injection 10 s at 10 kV. The typical electropherogram for 0.500 mg/L standard mixture solution was shown in Fig. 7(A). We can see three analytes were baseline separated within 8 min. To investigate the detection linearity of monuron, monolinuron and diuron by CE-ECL, a series of standard mixture solutions containing three analytes was injected consecutively six times to determine the repeatability of ECL intensity based on peak height and migration time for three analytes. The analytical results were summarized in Table 1. A comparison of the results obtained by present method with other methods is clearly shown in Table 2. The detection limit was lower than those by HPLC-MS-MS [5], GC-MS [7], immunoassay [12], PIF [13] and GHF-LPME [14], and the linear range was wider than these methods, too. The results shown the linearity and sensitive of the present method was good in detection. 3.7. Sample analysis Under the optimized conditions, the developed CE-ECL method was applied to the determination of monuron, monolinuron and diuron in yam sample that was prepared as 2.2. The spike and
Fig. 7. Electropherograms of CE-ECL. (a) monuron, (b) monolinuron, (c) diuron. (A) 0.500 mg/L standard mixture solutions; (B) yam sample; (C) yam sample spiked with 0.080 mg/kg standard mixture solutions; (D) yam sample spiked with 0.0001 mg/kg standard mixture solutions; Conditions: separation buffer concentration: 25 mM, other conditions as in Fig. 6.
recovery test for validating the accuracy of the proposed method was performed. 0.500 g yam samples separately spiked with 100 L of 5.900, 0.400, 0.080, 0.0005 mg/L standard mixture solutions (equivalent to 1.180, 0.080, 0.016 and 0.0001 mg/kg spiked levels) were pretreated and analyzed by the proposed method with six identical runs. The typical electropherograms of standard mixture solutions (A), yam sample (B) and yam sample spiked with 0.080 mg/kg (C) and 0.0001 mg/kg (D) standard mixture solution were shown in Fig. 7. As shown in Fig. 7 (C) and (D), a baseline separation for three analytes in spiked yam sample, the three analytes can be identified by the comparison of migration time of (A). From the electropherograms of (A), (C) and (D), no corresponding peak was found in yam sample (B), it indicates that the amount of monuron, monolinuron and diuron in yam is too little to be detected by this method. From Fig. 7(B)–(D), it can be observed that some small peaks of other unknown compounds, however, the migration time and the ECL intensity of them are different from three analytes. Therefore, they did not interfere with the analysis Table 2 Comparison of the results obtained by the present method with other reported methods. Method
Linear range (g/L)
Detection limit (g/L)
References
The present method HPLC HPLC-MS-MS GC-MS Immunoassay PIF G-HF-LPME
0.1–20,000 500–10,000 0.1–5 5–1000 0.01–100 – 10.0–400.0
0.01–0.05 – 0.026–0.237 5 0.1 1–28 2.0
– [3] [5] [7] [12] [13] [14]
148
Y. Hu / J. Chromatogr. B 986–987 (2015) 143–148
Table 3 Recoveries LODs and RSDs of phenylurea herbicides in yam samples (n = 6). Analyte
Found (mg/kg)
Added (mg/kg)
Total found (mg/kg)
RSD (%)
Recovery (%)
LOD (g/kg)
Monuron
ND
ND
Diuron
ND
1.166 0.0784 0.0156 0.00009 1.159 0.0785 0.0154 0.000092 1.171 0.0789 0.0156 0.00009
2.3 2.1 2.3 3.0 2.2 2.3 2.2 3.0 2.3 2.1 2.3 3.2
98.8 98.0 97.5 90.1 98.3 98.1 96.3 91.5 99.2 98.7 97.3 90.0
0.010
Monolinuron
1.180 0.080 0.016 0.0001 1.180 0.080 0.016 0.0001 1.180 0.080 0.016 0.0001
0.008
0.008
ND: not detected.
of monuron, monolinuron and diuron. The spiked recoveries were achieved using the peak height from the calibration curve under the optimized conditions, and the results are listed in Table 3. The recoveries in the range of 90.1–98.8% for monuron, 91.5–98.3% for monolinuron, and 90.0–99.2% for diuron were obtained in spiked yam sample. In addition, the RSDs of the found values were lower than 3.2% for all spiked level assays. In spiked yam sample, the detection limit (LOD) of 0.010 g/kg for monuron, the LOD of 0.008 g/kg for monolinuron and diuron. All the results shown the proposed method was suitable and accurate for phenylurea herbicides determination in actual sample detection. 4. Conclusion The method using capillary electrophoresis coupling with electrochemiluminescence was used for separation and detection of three phenylurea herbicides (monuron, monolinuron and diuron). The investigated method was found to be a simple, rapid, specific, and sensitive technique for the determination of the phenylurea herbicides (monuron, monolinuron and diuron) in yam. Moreover, because of fast separation, good selectivity, lower detection limits, powerful resolution and high sensitivity of this method, it may offer a potential method for the monitoring of environmental pollution problem or the residue content of herbicides in agricultural products. Acknowledgments This project was supported by Guangxi University of science and technology research projects No. 2013YB238.
References [1] H.-m. Ge, Z.-f. Lin, Z.-f. Yao, Y. Gao, Y.-p. Cong, H.-x. Yu, Aquat. Toxicol. 150 (2014) 165–174. [2] R. Kebeish, E. Azab, C. Peterhaensel, R. El-Basheer, Environ. Sci. Pollut. Res. 21 (2014) 8224–8232. [3] F.G. Tamayo, A. Martin-Esteban, J. Chromatogr. A 1098 (2005) 116–122. [4] M.C. Gennaro, E. Marengo, V. Gianotti, V. Maurino, J. Chromatogr. A 910 (2001) 79–86. – c, ´ ´ J. Ðordevi ´ N. Dujakovic, ´ K. Kumric, ´ T. Vasiljevic, ´ M. Lauˇsevic, ´ [5] T. Trtic-Petrovi c, Anal. Bioanal. Chem. 397 (2010) 2233–2243. [6] A. Valverde, A. Aguilera, C. Ferrer, F. Camacho, A. Cammarano, J. Agric. Food Chem. 58 (2010) 2818–2823. [7] E. Maloschik, M. Mrtl, A. Székács, Anal. Bioanal. Chem. 397 (2010) 537–548. [8] M. Murawska, I. Giza, Prog. Plant Protect. 40 (2000) 844–847. [9] T. Tuzimski, E. Soczewinski, Chromatographia 96 (2002) 219–223. [10] M. Hamad, R. Wintersteiger, J. Planar Chromatogr.—Modern TCL 15 (2002) 11–18. [11] H.Z. S¸enyuva, J. Gilbert, J. Chromatogr. B 878 (2010) 115–132. [12] P. Sharma, S. Gandhi, A. Chopra, N. Sekar, C. Raman Suri, Anal. Chim. Acta 676 (2010) 87–92. [13] P.A. Diaw, O.M.A. Mbaye, M.D. Gaye-Seye, J.-J. Aaron, A. Coly, A. Tine, N. Oturan, J. Fluoresc. 24 (2014) 1319–1330. [14] M. Sun, R.-x. Tang, Q.-h. Wu, C. Wang, Z. Wang, Food Anal. Methods 7 (2014) 1097–1102. [15] J.S. Pritchett, S.A. Shippy, J. Chromatogr. B 955–956 (2014) 81–85. ˜ ˜ [16] J.J. Berzas Nevado, G. Castaneda Penalvo, R.M. Rodríguez Dorado, Analytical Methods 921–922 (2013) 56–63. [17] G.-m. Zhu, S.-h. Long, H. Sun, W. Luo, X. Li, Z.-b. Hao, J. Chromatogr. B 941 (2013) 62–68. [18] Y. Wang, Q. Wu, M.-r. Cheng, C. Cai, J. Chromatogr. B 879 (2011) 871–877. [19] B.-y. Deng, L. Ye, H.-h. Yin, Y. Liu, S.-y. Hu, B.-f. Li, J. Chromatogr. B 879 (2011) 927–932. [20] H.Y. Chiu, Z.Y. Lin, H.L. Tu, C.W. Whang, J. Chromatogr. A 1177 (2008) 195–198. [21] S.-c. Liu, Y.-j. Liu, J. Li, M.-m. Guo, W. Pan, S.-z. Yao, Talanta 69 (2006) 154–159. [22] C. Cai, H.-y. Cheng, Y.-c. Wang, Anal. Methods 6 (2014) 2767–2773. [23] L. Xu, L.-b. Li, J.-s. Huang, T.-y. You, Talanta 118 (2014) 1–6. [24] J. Haginake, J. Sep. Sci. 32 (2009) 1548–1565.
Contents lists available at ScienceDirect
Journal of Chromatography B journal homepage: www.elsevier.com/locate/chromb
Simultaneous determination of phenylurea herbicides in yam by capillary electrophoresis with electrochemiluminescence detection Yuefang Hu ∗ College of Chemistry and Bioengineering, Hezhou University, Hezhou 542899, Guangxi, China
a r t i c l e
i n f o
Article history: Received 12 August 2014 Accepted 9 February 2015 Available online 17 February 2015 Keywords: Capillary electrophoresis Electrochemiluminescence Monuron Monolinuron Diuron Yam
a b s t r a c t A method of capillary electrophoresis (CE) coupled with electrochemiluminescence (ECL) detection has been applied to detect three major phenylurea herbicides (monuron, monolinuron and diuron) simultaneously. The effects of yam sample preparation, injection voltage and time, detection potential, detection buffer concentration and pH, Ru(bpy)3 2+ concentration, separation buffer type, separation buffer pH and concentration, separation voltage were investigated in detail. Under optimum conditions, a good baseline separation and highly sensitive detection for monuron, monolinuron and diuron were achieved. The ECL intensity (I) was in proportion to three analytes concentration () in the range of 0.1–10,000 g/L for monuron (r ≥ 0.9993), 0.1–18,000 g/L for monolinuron (r ≥ 0.9995) and 0.1–20,000 g/L for diuron (r ≥ 0.9997). The detection limits for monuron, monolinuron and diuron were 0.05, 0.04 and 0.01 g/L (S/N = 3), respectively. The developed method was successfully applied for the analysis of monuron, monolinuron and diuron residues in yam simultaneously. The average recoveries are in the ranges of 90.0–99.2% with relative standard deviations less than 3.2%. The limits of detection of the proposed method were 0.010 g/kg for monuron, 0.008 g/kg for monolinuron and diuron in yam. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Phenylurea herbicides, a kind of pesticides which includes monuron, monolinuron, diuron and metobromuron, etc. has been widely used to suppress broadleaf and grass weeds in China and other countries [1]. It was reported that phenylurea herbicides have chronic toxic, and its residues and degradation products showed certain activities in crops, soil and water, which brought potential risks to the health of human beings by environmental accumulation and food chain. For example, it has been reported that people ate the foodstuff with excessive monuron and diuron would cause carcinogenic [2], which has already caused people’s high attention. Therefore, some countries have set the maximum residue limit (MRL) of phenylurea herbicides in foodstuff. Yam is one of the Medicine-Food agricultural products with high nutritional value, with people’s awareness of health more and more strengthen, the demand for Yam are getting higher and higher. Therefore, in order to guarantee people’s health, and to provide technical support for the effective monitoring of pesticide residues in agricultural products, it is highly desirable to develop a simple
∗ Tel.: +86 13635079823. E-mail address: [email protected] http://dx.doi.org/10.1016/j.jchromb.2015.02.016 1570-0232/© 2015 Elsevier B.V. All rights reserved.
and effective method to monitor residues of phenylurea herbicides in the agricultural samples. Analytical methods for the quantification of phenylurea herbicide residues are mainly based on the chromatographic analysis, such as high performance liquid chromatography (HPLC) [3,4], high performance liquid chromatography mass spectrometry-mass spectrometry (HPLC-MS/MS) [5], liquid chromatography time-offlight mass spectrometry (LC/TOF-MS) [6], gas chromatography mass spectrometry (GC-MS) [7], thin layer chromatography (TLC) [8–10], immunoassay technique [11,12], photo-induced fluorescence (PIF) [13], graphene reinforced hollow fiber liquid phase microextraction (G-HF-LPME) [14], etc. However, these regular methods generally are high operation cost and long analytical time. Besides, the sample pretreatment procedures are rather cumbersome and expensive, which greatly diminished them application. Capillary electrophoresis (CE) has been attracted many researcher’s attention because of its advantage of high separation efficiency, short separation time, low reagent consumption, and ease of installation [15,16]. The features of the detector have great effect on the overall analytical efficiency for CE due to the small size of the separation capillaries and seriously low sample quantities injected into the capillaries. Actually, the detector should provide detection limits as low as possible without affecting the quality of separation. ECL is the procedure based on species generated at electrodes undergo high-energy electron transfer reactions
144
Y. Hu / J. Chromatogr. B 986–987 (2015) 143–148
to form excited states that emit light [17]. Nowadays, it has been widely used as a highly sensitive detection technique with its high sensitivity, wide linear range, and simple instrumentation. So, CE combining with ECL (CE-ECL) has been developed for the determination of various analytes containing tertiary amines group and their derivatives for a long time [18,19]. At present, it has become one of the high technology detection of pesticide residues, because of its advantages of high separation efficiency, good repeatability, short analysis time, high sensitivity and low solvent consumption, etc. For example, Chiu et al. [20] using 20 mM sodium phosphate (pH 9.0) and a separation voltage of 21 kV, working electrode at 1.6 V in a 300 mM sodium phosphate buffer (pH 8.0) containing 3.5 mM Ru(bpy)3 2+ , have been separated GLY and AMPA by CEECL detection, the limits of detection (LODs) for GLY and AMPA in water were 0.06 and 4.04 g mL−1 , respectively. Liu et al. [21] have used the CE-ECL to analyze mefenacet, the detection limit of 4.0 × 10−9 M, and was also applied to analyze residues of mefenacet in seedling and soil. The optimized conditions were as follows: potential applied at 1.2 V, injection time and injection voltage was 10 s and 11 kV, running buffer pH value was 7.38. Cai et al. [22] have determined pretilachlor in soil and rice using MSPD extraction by CE with field amplified sample injection and ECL detection, the limits of detection of the proposed method were 0.01 mg/kg in rice matrix and 0.008 mg/kg in soil matrix. The optimized conditions were as follows: potential applied at 1.17 V, injection time and voltage was 9 s and 8 kV, respectively, running buffer concentration and pH was 50 mmol L−1 and 7.38, respectively, buffer pH in detection cell was 8.5, separation voltage was 10 kV. However, this method seldom apply to analysis three kinds of pesticide residues simultaneously, and cannot give full play to the advantages and application potential of this high technology coupling technique. On the other hand, the influencing factors of the above application were investigated not detailed enough, and so far, the CE-ECL method has not been reported for determination of monuron, monolinuron and diuron simultaneously, and no application to yam sample was demonstrated. In this study, a simple and rapid CE-ECL method was adopted to determine phenylurea herbicides residues (monuron, monolinuron and diuron) simultaneously, and the matrix solid-phase dispersion method (MSPD) was developed to handle sample before determining [23]. Influencing factors, such as yam sample preparation, injection voltage and time, detection potential, detection buffer concentration and pH, Ru(bpy)3 2+ concentration, separation buffer type, separation buffer pH and concentration, separation voltage were investigated in detail. Residues of monuron, monolinuron and diuron in yam sample were also determined using the proposed technique. It was expected to be able to provide some theoretical basis for the quality evaluation and judgment of yam, and can provide some effective technical support for the agricultural product quality safety testing. 2. Experimental 2.1. Apparatus and reagents An MPI-A CE-ECL system was produced by Remax Electronic Science-Tech Co. Ltd. (Xi’an, China); a 42 cm length of 25 m i.d. uncoated fused-silica capillary was used (Yongnian Optical Fabric Factory, Hebei, China); ECL detection was employed using a three-electrode system: a 300 m diameter Pt disk as the working electrode, a Pt wire as the counter electrode, and a Ag/AgCl electrode (in saturated KCl solution) as the reference electrode. Tris(2,2-bipyridyl) ruthenium(II) chloride hexahydrate [Ru(bpy)3 Cl2 -6H2 O], and monuron, monolinuron and diuron standard compounds were obtained from Bailingwei Technology Company (Beijing, China). Yam was purchased from a local
pharmacy. 0.22 m membrane filters was purchased from Xinya Purification Material Factory (Shanghai, China). All water used throughout the study was double-distilled water, and reagents and chemicals used were analytical grade. 2.2. Yam sample preparation MSPD was developed to handle sample [23]: yam was weighted accurately after drying at 55 ◦ C for 4 h by oven, and grounding into powders for samples. 0.5 g sample (blank or spiked) was blended with 2.5 g dried Florisil material, and then ground together for 12 min to obtain a uniform mixture. Next, the mixture was transferred to a 10 mL syringe barrel covered in turn with a filter, anhydrous alumina at the bottom. Then, 3 cm height of anhydrous sodium sulfate and another filter were placed on the top of the mixture and then the sample mixture was compressed tightly. After successive elution with three aliquots of ethyl acetate (5 mL) for 4 min, the solutions were combined together and evaporated to dryness under a nitrogen stream. The residue was then dissolved in 100 mL 70% ethanol and soaked at room temperature for 24 h. Finally, 25 mM phosphate buffer solution (pH 8. 0) had been added and was volumed to 250 mL. Before analysis, all the yam sample were filtered through 0.22 m membrane filters. 2.3. Experimental methods Before the experiment, 350 L volume of 5 mM Ru(bpy)3 2+ and 45 mM phosphate buffer at pH 7.5 was injected into the detection reservoir, 25 mM phosphate buffer at pH 8.0 was used as separation buffer, 16 kV voltage was used as separation voltage, electrokinetic injection 10 s at 10 kV was used for sample introduction and detection potential at 1.13 V was applied at the working electrode. The biased potential of photomultiplier tube was set at 800 V. Before use, the working electrode was polished with alumina powder (0.3 and 0.05 m in turn) until a mirror-smooth surface appeared. Prior to CE injection, all the solutions were filtered through membrane filters. Then the injector end of capillary was put into phosphate buffer and electrophoresis voltage was applied, and the ECL analyzer was opened at the same time. In order to obtain accurate result, electrokinetic injection, separation and detection were began after the baseline of intensity signal to be stable (about 10 s). 3. Results and discussion 3.1. ECL behaviors of monuron, monolinuron and diuron The corresponding ECL curves of Ru(bpy)3 2+ , monuron, monolinuron and diuron were shown in Fig. 1, when a Pt electrode in 45 mM PBS (pH 7.5) with 5 mM Ru(bpy)3 2+ in the potential range from 0 to 1.3 V, only relatively weak ECL signal could be observed (Fig. 1(a0 )). With the addition of the same concentration of monuron, monolinuron and diuron, respectively. As is clearly shown in Fig. 1, the ECL intensity of three analytes increased remarkably, and the ECL intensity of diuron (Fig. 1(c)) is a little stronger than monuron (Fig. 1(a)) and monolinuron (Fig. 1(b)) at the same concentration. 3.2. Optimization of the yam sample preparation The yam samples were handled based on the MSPD method that have been reported [23]. Firstly, Florisil was chosen as the dispersant because it led to a cleaner chromatographic profile with lower baseline than C8 , C18 and silica. Various mass ratios of the dispersant to the sample matrix (0.5 g) from 7:1 to 3:1 were assayed with
Y. Hu / J. Chromatogr. B 986–987 (2015) 143–148
Fig. 1. ECL behaviors of monuron (a), monolinuron (b) and diuron (c). (a0 ) 45 mM PBS (pH 7.5) +5 mM Ru(bpy)3 2+ ; (a) a0 + 0.500 mg/L monuron; (b) a0 + 0.500 mg/L monolinuron; (c) a0 + 0.500 mg/L diuron.
10 mL ethyl acetate as the elution solvent. The mass ratio of 5:1 provided the best recoveries for three phenylurea herbicides. Secondly, four organic solvents including ethyl acetate, dichloromethane, methanol and acetone were used as the extractants to study their effect on recoveries with the mass ratios of the dispersant to the sample matrix as 5:1. It was indicated that high recoveries (>90%) were obtained with ethyl acetate and dichloromethane, and ethyl acetate was selected due to its low toxicity. The influence of ethyl acetate volume (6–18 mL) on the recoveries was also investigated, and the result shown that the recoveries reached maxima when the eluting volume was 15 mL. So, 15 mL ethyl acetate was chosen in subsequent experiments.
3.3. Optimization of injection conditions The effect of the injection voltage and time on the theoretical plates number (N) and the ECL intensity of monuron, monolinuron and diuron were investigated. The theoretical plates number was calculated by the equation: N = 5.54(tR /Wh )2 , where tR and Wh are the migration time and the width at half the maximum peak height of the analytes, respectively. The effect of injection voltage on the theoretical plates number and ECL intensity in the range of 2–18 kV as injection time set at 10 s was shown in Fig. 2, It was seen when a lower injection voltage was used, it was difficult to obtain strong intensity. However, the theoretical plates number decreased and the ECL intensity increased at high injection voltage. The probable reason is that at high injection voltage, more analytes enter into the diffusion layer and led to strong ECL intensity, but more analytes came into the detection cell resulting the analytes cannot reach the electrode surface quickly and diffuse into the solution, so the peak is retarded and broadened, and led to the theoretical plates number was decreased, so, in order to obtain high ECL intensity and theoretical plates number,10 kV was selected as the optimum injection voltage. Fixing the injection voltage at 10 kV, when the injection time changed from 2 to 10 s, the ECL intensity increased correspondingly. Over 10 s, the ECL intensity increased slowly. Furthermore, because of an excessively large injection volume and overloading occurred, the longer injection time broadened the peak shape and led to theoretical plates number decreased. As a compromise, 10 s was selected as the optimum injection time in subsequent experiments.
145
Fig. 2. The effect of injection voltage on 0.500 mg/L monuron, monolinuron and diuron theoretical plates number and ECL intensity. Conditions: separation capillary, 42 cm length (25 m i.d.); detection potential, 1.13 V; separation buffer, 25 mM phosphate buffer pH 8.0; separation voltage, 16 kV; in ECL cell: 5 mM Ru(bpy)3 2+ , phosphate buffer pH 7.5; injection time: 10 s. a, b, and c are the ECL intensity of monuron, monolinuron and diuron, respectively; a1 , b1 , and c1 are the theoretical plates number of monuron, monolinuron, and diuron, respectively.
3.4. Optimization of detection conditions 3.4.1. Effect of detection potential The optimized detection potential had a great influence on the ECL signal. As shown in Fig. 1, when the detection potential was less than 1.00 V, the ECL responses of monuron, monolinuron and diuron were relatively low and were difficult to be observed, it was indicated that Ru(bpy)3 2+ had not happened oxidation reaction on the surface of Pt electrode at these potential. When the detection potential was higher than 1.00 V, monuron, monolinuron and diuron ECL intensity increased and showed a maximum value at 1.13 V, then decreased slightly exceeded 1.13 V. Therefore, to obtain a high ECL intensity and high efficiency, 1.13 V was selected as the optimum detection potential in subsequent experiments. 3.4.2. Effect of the detection buffer concentration The effect of phosphate buffer concentration in the cell on ECL intensity of monuron, monolinuron and diuron was investigated in the concentration range from 25 to 60 mM. As can be seen from Fig. 3, at the beginning, the ECL intensity of monuron, monolinuron and diuron increased with the increase of phosphate buffer concentration, and the highest ECL intensity was obtained at 45 mM, when the concentration was exceeded 45 mM, three analytes ECL intensity decreased. In addition, when phosphate buffer concentration exceeded 45 mM, it gave higher baseline noise in experiment. Thus, a 45 mM detection phosphate buffer was chosen in further experiment. 3.4.3. Effect of the detection buffer pH It was reported that the ECL reaction of tertiary amine with Ru(bpy)3 2+ based on the buffer pH to a large extent [24], so the effect of detection buffer pH on ECL intensity was investigated within the range 6.0–10.0. As illustrated in Fig. 4, the ECL intensity of monuron, monolinuron and diuron increased quickly with increasing pH, and to a maximum at pH 7.5. However, the ECL intensity was decreased slightly when the pH value exceeded 7.5. It can be explained that the competition of the reaction of Ru(bpy)3 3+ and OH−1 ions in higher pH value [24], thus, when the pH value was higher than 8.0, the noise of baseline was very large. So, a 7.5 pH value was chosen as the optimized pH in further experiment.
146
Y. Hu / J. Chromatogr. B 986–987 (2015) 143–148
Fig. 3. The effect of detection buffer concentration on 0.500 mg/L monuron (a), monolinuron (b) and diuron (c) ECL intensity. Conditions: separation capillary, 42 cm length (25 m i.d.); detection potential, 1.13 V; separation buffer, 25 mM phosphate buffer pH 8.0; separation voltage, 16 kV; in ECL cell: 5 mM Ru(bpy)3 2+ , phosphate buffer pH 7.5; injection voltage and time: 10 kV and 10 s.
3.4.4. Effect of Ru(bpy)3 2+ concentration The ECL intensity is strongly influenced by the concentration of Ru(bpy)3 2+ that used as the ECL reagent in the system. The ECL intensity increased markedly with increasing Ru(bpy)3 2+ concentration due to the generation of more amount of electrogenerated excited species Ru(bpy)3 2+ , but the more noise was produced and the background current increased significantly when its concentration exceeded 5 mM. In addition, too high concentration was required more consumption of the expensive reagent Ru(bpy)3 2+ . So, in order to obtain higher separation efficiency, better signalto-noise ratio, stronger ECL intensity of three analytes and less consumption of reagent, 5 mM Ru(bpy)3 2+ was selected. 3.5. Optimization of separation conditions 3.5.1. Effect of separation buffer type It was known that separation buffer has great influence on separation and detection. In this study, phosphate buffer, Tris–HCl, and
Fig. 4. The effect of detection buffer pH on 0.500 mg/L monuron (a), monolinuron (b) and diuron (c) ECL intensity. Conditions: separation capillary, 42 cm length (25 m i.d.); detection potential, 1.13 V; separation buffer, 25 mM phosphate buffer pH 8.0; separation voltage, 16 kV; in ECL cell: 5 mM Ru(bpy)3 2+ , 45 mM phosphate buffer; injection voltage and time: 10 kV and 10 s.
Fig. 5. The effect of separation buffer pH on 0.500 mg/L monuron (a), monolinuron (b) and diuron (c) ECL intensity and resolutions (a0 , a/b Rs , b0 , b/c Rs ). Conditions: separation capillary, 42 cm length (25 m i.d.); detection potential, 1.13 V; separation buffer, 25 mM phosphate buffer; separation voltage, 16 kV; in ECL cell: 5 mM Ru(bpy)3 2+ , 45 mM phosphate buffer pH 7.5; injection voltage and time: 10 kV and 10 s.
borate buffer were investigated. It was indicated that phosphate buffer can obtain good resolution. In addition, the ECL intensity stability and the signal-to-noise ratio of three analytes were the best in this buffer system. Therefore, phosphate buffer was selected as separation buffer. 3.5.2. Effect of the separation buffer pH The pH of separation buffer has great influence on the electroosmotic flow (EOF) of capillary and the extent of ionization of analytes, so it affects the ECL intensity, the resolution as well as the sensitivity of the analytes. As can be seen from Fig. 5, the ECL intensity (a, b, c)and resolutions (a0 , b0 )increased as pH changed from 6.0 to 8.0, and then decreased when the pH value was higher than 8.0. The results also indicated that when pH was 8.0, the highest ECL intensity and best resolutions of monuron, monolinuron and diuron were obtained. Therefore, 8.0 of the pH value was selected. 3.5.3. Effect of the separation buffer concentration The effect of separation buffer concentration in the range 10–50 mM on monuron, monolinuron and diuron ECL intensity and resolutions was also studied. It can be seen from Fig. 6, the highest ECL intensity was produced when the separation buffer concentration was 25 mM; when the separation concentration was above 25 mM, the migration time of three analytes was prolonged gradually and the Joule heating was become stronger, which increased the baseline noises, even produced air bladder in the capillary, and led to the intensity (a, b, c) and the resolutions (a0 , b0 ) of three analytes decreased. This phenomenon was largely due to the effect of the ionic strength increase with high separation concentration, which resulted in the increase Joule heating, EOF decreased and the migration time prolonged. Therefore, in order to get the best resolutions and the highest intensity, the optimized concentration of separation buffer was selected as 25 mM. 3.5.4. Effect of the separation voltage The separation voltage has a great effects on the quality of the separation and the migration time of analytes. In order to get the optimum condition, different voltages were applied over the range of 12–20 kV. The effect of the separation voltage on the ECL intensity of monuron, monolinuron and diuron was investigated. The ECL intensity was enhanced with increasing the separation voltage in the range of 12–16 kV. However, when separation voltage was
Y. Hu / J. Chromatogr. B 986–987 (2015) 143–148
147
Table 1 Regression equations, linear range, detection limit for monuron, monolinuron and diuron (n = 6). Analyte
Regression equation
Correlation coefficient
Linear range (g/L)
Detection limit (g/L)
Monuron Monolinuron Diuron
I = 937.12+314.54 I = 935.65+365.16 I = 936.17+462.28
0.9993 0.9995 0.9997
0.1–10,000 0.1–18,000 0.1–20,000
0.05 0.04 0.01
RSD (%) Migration time
Intensity
1.6 1.9 1.8
2.3 2.6 2.2
I: ECL intensity; : analyte concentration.
Fig. 6. The effect of separation buffer concentration on 0.500 mg/L monuron (a), monolinuron (b) and diuron (c) ECL intensity and resolutions (a0 , a/b Rs , b0 , b/c Rs ). Conditions: separation capillary, 42 cm length (25 m i.d.); detection potential, 1.13 V; separation buffer, phosphate buffer pH 8.0; separation voltage, 16 kV; in ECL cell: 5 mM Ru(bpy)3 2+ , 45 mM phosphate buffer pH 7.5; injection voltage and time: 10 kV and 10 s.
more than 16 kV, the baseline noise was increased. What is more, the increasing joule heat in the capillary caused peak broadening and adversely affected the separation effectiveness. So 16 kV was chosen as the best separation voltage in this experiment. 3.6. Linearity, precision, detection limit of monuron, monolinuron and diuron Under the optimized conditions: ECL detection potential at 1.13 V, 45 mM phosphate buffer at pH 7.5, 5 mM Ru(bpy)3 2+ in the detection reservoir, and 25 mM phosphate buffer at pH 8.0 as separation buffer, separation voltage at 16 kV, electrokinetic injection 10 s at 10 kV. The typical electropherogram for 0.500 mg/L standard mixture solution was shown in Fig. 7(A). We can see three analytes were baseline separated within 8 min. To investigate the detection linearity of monuron, monolinuron and diuron by CE-ECL, a series of standard mixture solutions containing three analytes was injected consecutively six times to determine the repeatability of ECL intensity based on peak height and migration time for three analytes. The analytical results were summarized in Table 1. A comparison of the results obtained by present method with other methods is clearly shown in Table 2. The detection limit was lower than those by HPLC-MS-MS [5], GC-MS [7], immunoassay [12], PIF [13] and GHF-LPME [14], and the linear range was wider than these methods, too. The results shown the linearity and sensitive of the present method was good in detection. 3.7. Sample analysis Under the optimized conditions, the developed CE-ECL method was applied to the determination of monuron, monolinuron and diuron in yam sample that was prepared as 2.2. The spike and
Fig. 7. Electropherograms of CE-ECL. (a) monuron, (b) monolinuron, (c) diuron. (A) 0.500 mg/L standard mixture solutions; (B) yam sample; (C) yam sample spiked with 0.080 mg/kg standard mixture solutions; (D) yam sample spiked with 0.0001 mg/kg standard mixture solutions; Conditions: separation buffer concentration: 25 mM, other conditions as in Fig. 6.
recovery test for validating the accuracy of the proposed method was performed. 0.500 g yam samples separately spiked with 100 L of 5.900, 0.400, 0.080, 0.0005 mg/L standard mixture solutions (equivalent to 1.180, 0.080, 0.016 and 0.0001 mg/kg spiked levels) were pretreated and analyzed by the proposed method with six identical runs. The typical electropherograms of standard mixture solutions (A), yam sample (B) and yam sample spiked with 0.080 mg/kg (C) and 0.0001 mg/kg (D) standard mixture solution were shown in Fig. 7. As shown in Fig. 7 (C) and (D), a baseline separation for three analytes in spiked yam sample, the three analytes can be identified by the comparison of migration time of (A). From the electropherograms of (A), (C) and (D), no corresponding peak was found in yam sample (B), it indicates that the amount of monuron, monolinuron and diuron in yam is too little to be detected by this method. From Fig. 7(B)–(D), it can be observed that some small peaks of other unknown compounds, however, the migration time and the ECL intensity of them are different from three analytes. Therefore, they did not interfere with the analysis Table 2 Comparison of the results obtained by the present method with other reported methods. Method
Linear range (g/L)
Detection limit (g/L)
References
The present method HPLC HPLC-MS-MS GC-MS Immunoassay PIF G-HF-LPME
0.1–20,000 500–10,000 0.1–5 5–1000 0.01–100 – 10.0–400.0
0.01–0.05 – 0.026–0.237 5 0.1 1–28 2.0
– [3] [5] [7] [12] [13] [14]
148
Y. Hu / J. Chromatogr. B 986–987 (2015) 143–148
Table 3 Recoveries LODs and RSDs of phenylurea herbicides in yam samples (n = 6). Analyte
Found (mg/kg)
Added (mg/kg)
Total found (mg/kg)
RSD (%)
Recovery (%)
LOD (g/kg)
Monuron
ND
ND
Diuron
ND
1.166 0.0784 0.0156 0.00009 1.159 0.0785 0.0154 0.000092 1.171 0.0789 0.0156 0.00009
2.3 2.1 2.3 3.0 2.2 2.3 2.2 3.0 2.3 2.1 2.3 3.2
98.8 98.0 97.5 90.1 98.3 98.1 96.3 91.5 99.2 98.7 97.3 90.0
0.010
Monolinuron
1.180 0.080 0.016 0.0001 1.180 0.080 0.016 0.0001 1.180 0.080 0.016 0.0001
0.008
0.008
ND: not detected.
of monuron, monolinuron and diuron. The spiked recoveries were achieved using the peak height from the calibration curve under the optimized conditions, and the results are listed in Table 3. The recoveries in the range of 90.1–98.8% for monuron, 91.5–98.3% for monolinuron, and 90.0–99.2% for diuron were obtained in spiked yam sample. In addition, the RSDs of the found values were lower than 3.2% for all spiked level assays. In spiked yam sample, the detection limit (LOD) of 0.010 g/kg for monuron, the LOD of 0.008 g/kg for monolinuron and diuron. All the results shown the proposed method was suitable and accurate for phenylurea herbicides determination in actual sample detection. 4. Conclusion The method using capillary electrophoresis coupling with electrochemiluminescence was used for separation and detection of three phenylurea herbicides (monuron, monolinuron and diuron). The investigated method was found to be a simple, rapid, specific, and sensitive technique for the determination of the phenylurea herbicides (monuron, monolinuron and diuron) in yam. Moreover, because of fast separation, good selectivity, lower detection limits, powerful resolution and high sensitivity of this method, it may offer a potential method for the monitoring of environmental pollution problem or the residue content of herbicides in agricultural products. Acknowledgments This project was supported by Guangxi University of science and technology research projects No. 2013YB238.
References [1] H.-m. Ge, Z.-f. Lin, Z.-f. Yao, Y. Gao, Y.-p. Cong, H.-x. Yu, Aquat. Toxicol. 150 (2014) 165–174. [2] R. Kebeish, E. Azab, C. Peterhaensel, R. El-Basheer, Environ. Sci. Pollut. Res. 21 (2014) 8224–8232. [3] F.G. Tamayo, A. Martin-Esteban, J. Chromatogr. A 1098 (2005) 116–122. [4] M.C. Gennaro, E. Marengo, V. Gianotti, V. Maurino, J. Chromatogr. A 910 (2001) 79–86. – c, ´ ´ J. Ðordevi ´ N. Dujakovic, ´ K. Kumric, ´ T. Vasiljevic, ´ M. Lauˇsevic, ´ [5] T. Trtic-Petrovi c, Anal. Bioanal. Chem. 397 (2010) 2233–2243. [6] A. Valverde, A. Aguilera, C. Ferrer, F. Camacho, A. Cammarano, J. Agric. Food Chem. 58 (2010) 2818–2823. [7] E. Maloschik, M. Mrtl, A. Székács, Anal. Bioanal. Chem. 397 (2010) 537–548. [8] M. Murawska, I. Giza, Prog. Plant Protect. 40 (2000) 844–847. [9] T. Tuzimski, E. Soczewinski, Chromatographia 96 (2002) 219–223. [10] M. Hamad, R. Wintersteiger, J. Planar Chromatogr.—Modern TCL 15 (2002) 11–18. [11] H.Z. S¸enyuva, J. Gilbert, J. Chromatogr. B 878 (2010) 115–132. [12] P. Sharma, S. Gandhi, A. Chopra, N. Sekar, C. Raman Suri, Anal. Chim. Acta 676 (2010) 87–92. [13] P.A. Diaw, O.M.A. Mbaye, M.D. Gaye-Seye, J.-J. Aaron, A. Coly, A. Tine, N. Oturan, J. Fluoresc. 24 (2014) 1319–1330. [14] M. Sun, R.-x. Tang, Q.-h. Wu, C. Wang, Z. Wang, Food Anal. Methods 7 (2014) 1097–1102. [15] J.S. Pritchett, S.A. Shippy, J. Chromatogr. B 955–956 (2014) 81–85. ˜ ˜ [16] J.J. Berzas Nevado, G. Castaneda Penalvo, R.M. Rodríguez Dorado, Analytical Methods 921–922 (2013) 56–63. [17] G.-m. Zhu, S.-h. Long, H. Sun, W. Luo, X. Li, Z.-b. Hao, J. Chromatogr. B 941 (2013) 62–68. [18] Y. Wang, Q. Wu, M.-r. Cheng, C. Cai, J. Chromatogr. B 879 (2011) 871–877. [19] B.-y. Deng, L. Ye, H.-h. Yin, Y. Liu, S.-y. Hu, B.-f. Li, J. Chromatogr. B 879 (2011) 927–932. [20] H.Y. Chiu, Z.Y. Lin, H.L. Tu, C.W. Whang, J. Chromatogr. A 1177 (2008) 195–198. [21] S.-c. Liu, Y.-j. Liu, J. Li, M.-m. Guo, W. Pan, S.-z. Yao, Talanta 69 (2006) 154–159. [22] C. Cai, H.-y. Cheng, Y.-c. Wang, Anal. Methods 6 (2014) 2767–2773. [23] L. Xu, L.-b. Li, J.-s. Huang, T.-y. You, Talanta 118 (2014) 1–6. [24] J. Haginake, J. Sep. Sci. 32 (2009) 1548–1565.