Abstract
A modified QuEChERS (quick, easy, cheap, effective, rugged, and safe) method and ultrahigh-performance liquid chromatography coupled to tandem mass spectrometry (UHPLC–MS/MS) were optimized and validated for 16 antibiotics belonging to three families (macrolides, quinolones, and sulfonamides) that were found in preserved eggs. Samples were extracted in 4 mL water and 10 mL acetonitrile with 1% acetic acid and subjected to a cleanup procedure using dispersive solid-phase extraction with C18 and primary secondary amine sorbents, prior to detection by UHPLC–MS/MS. Matrix-matched calibration was used for quantification to reduce the matrix effect with limits of quantification in the range of 0.3–3.0 μg/kg. Validation of the method was conducted by recovery and precision experiments. Recoveries of the spiked samples ranged from 73.8% to 127.4%, and the intra- and inter-day relative standard deviations were lower than 21.2% and 22.3%, respectively. This method was successfully applied to the analysis of antibiotics in preserved egg samples.
Introduction
Preserved eggs, also known as “Pidan” or “century eggs,” are a traditional and popular food in China. They can be made from chicken, duck, or quail eggs in a mixture of clay, ash, salt, quicklime, and rice hulls over a period of several weeks to months, depending on the production method [1, 2] (Figure 1).
Antibiotics, mainly macrolides, quinolones, and sulfonamides, have been used on a large scale in poultry farms in China to control disease and as growth promoters [3, 4]. Consequently, these antibiotics may accumulate in eggs, potentially posing risks to the health of consumers [5–8]. To minimize the risks and ensure food safety, several international organizations and countries have taken action and established maximum residue limits (MRLs) with regards to the levels of antibiotics in eggs. In the European Union, erythromycin has an MRL of 150 μg/kg, while the use of quinolones (enrofloxacin and ciprofloxacin) and sulfonamides is not permitted in laying hens [9]. The same legislation and MRLs have been established by the Chinese government (announcement nos. 176 and 1519 from the Ministry of Agriculture of the People’s Republic of China) [4]. However, due to the popular consumption of preserved eggs, sensitive and effective methods for the simultaneous determination of antibiotic levels in such food items are necessary.
In the last decade, liquid chromatography (LC) has been used for the detection of multiresidues [7, 10, 11] and single families of antibiotics, including macrolides [12, 13], quinolones [14, 15], and sulfonamides [16, 17]. LC coupled to an ultraviolet (UV) source or diode array [16, 18, 19], fluorescence [20], a time-of-flight (TOF) analyzer [21], or biosensor [22] has been used to detect antibiotics. Meanwhile, LC coupled to tandem mass spectrometry (LC–MS/MS) has become a promising technique for the analysis of antibiotics due to its selectivity and sensitivity at low concentrations. The selectivity can be further increased when ultra-high performance LC (UHPLC) is used with MS/MS [7].
However, the different physicochemical properties of antibiotics make it difficult to extract and clean these analytes simultaneously. Current available methods include solid-phase extraction (SPE) [21] and pressurized liquid extraction (PLE) [11], but these techniques are tedious and time-consuming. In the last few years, a modified extraction procedure known as QuEChERS (quick, easy, cheap, effective, rugged, and safe) has been used for the extraction and purification of a variety of chemicals, including antibiotics and veterinary drugs in many matrices [4], such as bees [23], fish, and other edible tissues [8]; however, few studies have applied it to the assessment of eggs, including preserved eggs.
In this study, a modified QuEChERS sample extraction and dispersive SPE cleanup method was developed with UHPLC–MS/MS analysis for the simultaneous determination of 16 antibiotics in preserved eggs. The method was found to be rapid, safe, sensitive, and cheap and was applied to determine the antibiotic residues in preserved egg samples.
Materials and Methods
Chemicals and Reagents
Antibiotic standards (roxithromycin, azithromycin, clarithromycin, erythromycin, sparfloxacin, lomefloxacin, sulfachloropyridazine, enrofloxacin, sulfamethoxydiazine, sulfameter, sulfamonomethoxine, sulfadimidine, sulfamethazine, sufisomezole, and sulfadiazine) were purchased from Dr. Ehrenstorfer (Augsburg, Germany), while gatifloxacin was supplied by Toronto Research Chemicals Inc. (Canada).
HPLC-grade acetonitrile (ACN) and ammonium acetate were supplied by Tedia (Ohio State, USA), formic acid (purity, >99%) was obtained from Roe Scientific Inc. (USA), and ultrapure water was prepared by a Milli-Q gradient water system (Millipore, Bedford, MA, USA). Bondesil-C18 (40 μm, 100 g), primary secondary amine (PSA; 40 μm, 100 g), and graphitized carbon black (GCB) were obtained from Agilent (CA, USA). Acetic acid (purity, >99.8%) was supplied by CNW Technologies GmbH (Germany). Anhydrous magnesium sulfate (MgSO4) and sodium chloride (NaCl) were supplied by Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). Nylon filters (0.22 μm) were purchased from Tianjin Jinteng Experiment Equipment Co., Ltd. (Tianjin, China).
Stock standard solutions of the individual compounds with concentrations of 1000 mg/L were prepared by accurately weighing the powders followed by dissolution in ACN and storage at −20 °C in the dark. A multicompound working standard solution with a concentration of 10 mg/L was prepared for each compound using appropriate dilutions with ACN, with prepared solutions stored at −20 °C in the dark.
Apparatus and Software
LC was performed using an Agilent 1200 SL Series Rapid Resolution LC System (Santa Clara, CA, USA), which consisted of a binary pump, autosampler, vacuum degasser, and a column oven, with an Extend-C18 column (100 mm × 2.1 mm, 1.7 μm particle size). The LC system was coupled to an Agilent 6460 Triple Quadrupole LC/MS (QqQ), which was operated using an electrospray ionization (ESI) interface in positive ion mode. Data acquisition was performed using the Mass Hunter Workstation software (B.01.03). The nitrogen evaporator was obtained from Plus Century (Beijing, China). The centrifuge was obtained from HERMLE Labortechnik GmbH (Stuttgart, Germany).
Sample Preparation
Twenty preserved egg samples were purchased from retail markets in WuHan (China). Preliminary analyses were performed using our method to determine whether they could be used as blank samples. These were used for validation experiments.
Homogenized preserved eggs (2 ± 0.05 g; both the yolk and egg white) were weighed into 50 mL polypropylene centrifuge tubes, followed by 4 mL water. The samples were then vortexed for 1 min prior to the addition of 10 mL ACN with 1% acetic acid. After that, the tubes were mixed on a rotary agitator for 20 min and then 4 g MgSO4 and 1 g NaCl were added. Finally, the sample tubes were capped tightly and placed on a vortex mixer for 1 min, prior to centrifugation at 5000 rpm for 5 min at 4 °C.
An aliquot (1.5 mL) of the upper ACN layer from each tube was transferred into a 2 mL tube that contained 50 mg PSA, 50 mg C18, and 100 mg MgSO4. After that, tubes were capped tightly, vortexed for 1 min, and then centrifuged at 5000 rpm for 5 min at 4 °C. A 1 mL aliquot of supernatant was then transferred into a 2 mL tube and evaporated to dryness under a gentle stream of nitrogen at 40 °C for each sample. The dry residue was then reconstituted with 1 mL of a solution containing ACN and 5 mmol/L ammonium acetate with 0.1% formic acid in water (50:50, v/v). Filtration through a 0.22 μm nylon filter provided the sample for UHPLC–MS/MS analysis.
UHPLC–MS/MS Analysis
Chromatographic samples were separated with a mobile phase consisting of ACN (eluent A) and 5 mmol/L ammonium acetate with 0.1% (v/v) formic acid in water (eluent B). The flow rate was 0.25 mL/min. The elution started with 10% of eluent A, followed by an equilibration time of 3 min; the concentration of A was then linearly increased up to 90% over 3 min, followed by an equilibration time of 1 min, and then returned to the initial conditions over 3 min. Re-equilibration followed for a time of 5 min, so that the total run time was 15 min. The column temperature was maintained at 40 °C, and the injection volume was 10 μL.
MS/MS Conditions
MS analysis was operated using ESI in the positive mode. Multiple reaction monitoring (MRM) with two mass transitions that acquired two specific precursor–product ion transitions per target compound was selected, of which the most abundant transition was used for quantification and the second most abundant transition was used for confirmation. The MS was connected to the UHPLC system through an ESI interface. Nitrogen was used as the drying and sheath gas. The source parameters were as follows: gas temperature, 325 °C; gas flow, 6 L/min; nebulizer, 40 psi; sheath gas temperature, 325 °C; sheath gas, 11 L/min; capillary voltage, 3500 V; nozzle voltage, 500 V; and dwell time, 20 ms. The parameters of the transitions, the applied cone voltages, and the collision energies are shown in Table 1.
Retention time windows (RTWs) and MS/MS parameters of the selected antibiotics
Compound | RTW (min) | Ionization mode | Cone voltage (V) | Quantitation transitiona | Confirmation transitiona |
---|---|---|---|---|---|
Roxithromycin | 7.96 | + | 200 | 838.1 > 680.4(21) | 838.1 > 158.1(33) |
Azithromycin | 7.95 | + | 141 | 750 > 158.1(41) | 750 > 116.1(49) |
Clarithromycin | 7.44 | + | 165 | 749 > 158.1(29) | 749 > 116.1(45) |
Erythromycin | 7.74 | + | 160 | 734.2 > 158.1(29) | 734.2 > 116.1(53) |
Sparfloxacin | 7.42 | + | 147 | 393.4 > 375.1(17) | 393.4 > 264(37) |
Gatifloxacin | 7.35 | + | 167 | 376.4 > 358.2(17) | 376.4 > 261.1(29) |
Enrofloxacin | 7.27 | + | 140 | 360 > 342.2(18) | 360 > 316.2(14) |
Lomefloxacin | 7.12 | + | 108 | 352.3 > 308.1(13) | 352.3 > 265.1(21) |
Sulfachloropyridazine | 7.65 | + | 98 | 285.2 > 108(25) | 285.2 > 92.1(29) |
Sulfamethoxydiazine | 7.31 | + | 113 | 281.2 > 108(26) | 281.2 > 92(30) |
Sulfameter | 7.15 | + | 118 | 281 > 108(26) | 281 > 92(30) |
Sulfamonomethoxine | 7.59 | + | 118 | 281 > 108(26) | 281 > 92(34) |
Sulfadimidine | 7.15 | + | 100 | 279 > 186(14) | 279 > 124(22) |
Sulfamerazine | 4.80 | + | 120 | 265 > 156(14) | 265 > 108(26) |
Sufisomezole | 7.76 | + | 100 | 254.3 > 156(13) | 254.3 > 108.1(25) |
Sulfadiazine | 3.08 | + | 108 | 251.3 > 156(9) | 251.3 > 108.1(21) |
Collision energy (in eV) is given in brackets.
Validation Study
The matrix effect, linearity, accuracy, precision, limits of detection (LODs), and limits of quantitation (LOQs) were established by a validation procedure with spiked preserved egg samples.
The matrix effects were evaluated by the slopes obtained in the calibration with matrix-matched calibration (concentration levels: 5, 10, 20, 50, and 100 μg/L) and those obtained with solvent standards. Linearity was evaluated using matrix-matched calibration, spiking blank extracts at five concentration levels (5, 10, 50, 100, and 200 μg/kg). Accuracy and intra-day precision were validated by spiking blank preserved egg samples at three fortification levels (10 μg/kg, 50 μg/kg, and 100 μg/kg) with five replicates for each concentration. Inter-day precision was evaluated by analyzing five spiked samples at one fortification level on three separate days. LODs and LOQs were determined as the amounts of analyte for which the signal-to-noise ratios (S/N) were equal to 3 and 10, respectively.
Results and Discussion
Optimization of UHPLC–MS/MS Conditions
Recently, LC and UHPLC coupled to MS/MS have emerged as suitable techniques for the simultaneous detection of multiresidue antibiotics due to the robustness of the resulting analyses at low concentration levels [24–26]. In order to obtain appropriate peak shapes and resolution, the C18 column has been most widely used [10, 27, 28] and was adopted in this study with a mobile phase consisting of ACN and water with 5 mmol/L ammonium acetate and 0.1% (v/v) formic acid so as to obtain good sensitivity and resolution. Although sulfameter, sulfamethoxydiazine, and sulfamonomethoxine have the same MRM parameters, they also have relative separations in this mobile phase. Thus, a total run time of 10 min, followed by a reequilibration time of 5 min, provided the best chromatographic results with minimum analysis time.
Identification of the precursor and product ions, the fragments, and the collision energies was achieved with a Masshunter Optimizer (B.03.01) in ESI positive mode with injection of 1 μg/mL matrix-matched standard solution. Under these conditions, protonated [M+H]+ molecules were detected as the precursor ions for all analytes. The MS/MS parameters for each compound are shown in Table 1.
Optimization of the Sample Pretreatment
Selection of Extraction Solvent
An appropriate sample treatment is essential for reliable results in multiclass antibiotics analysis. In some reported antibiotics detection methods [11], ACN has been considered the best extraction solvent due to its ability to precipitate proteins and lipids, which could degrade drug residues during sample treatment. To increase throughput, a QuEChERS-based procedure that could reduce sample handling time was chosen in this study prior to UHPLC–MS/MS analysis.
Bearing in mind the characteristics of the preserved egg matrix, 4 mL water was added to samples, followed by ACN with 1% acetic acid as the extraction solvent. Previous reports have suggested that suitable recoveries may be obtained when acetic acid is used; however, it resulted in a more obvious matrix enhancement effect (Figure 2). Notably, the recoveries of quinolone and sulfonamide antibiotics increased in the presence of acetic acid, while the recovery of macrolide antibiotics decreased.
Optimization of the Cleanup Procedure
Coextracted compounds can interfere with the analysis of antibiotics by UHPLC–MS/MS and reduce the lifetime of the column, making a purification procedure indispensable. However, the use of conventional SPE cartridges, such as OASIS HLB [7] or StrataX [21], is time-consuming and expensive. Here, dispersive SPE was utilized with sorbents to purify the samples in a more efficient and cost-effective process.
Previous reports have suggested that PSA, C18, and GCB sorbents can provide adequate results with UHPLC–MS/MS [23, 30, 31]. As a consequence, to achieve the highest recoveries, we assessed different combinations of dispersive tube sorbents, including: C18 + MgSO4 (50 mg + 150 mg), C18 + PSA + MgSO4 (50 mg + 150 mg + 150 mg), and C18 + PSA + GCB + MgSO4 (50 mg + 50 mg + 50 mg + 150 mg). A spiked concentration of 20 μg/kg was used during the procedure, and the impact on the recovery is shown in Figure 3. The recoveries of all antibiotics decreased with the addition of GCB, while a conjugation of PSA and C18 resulted better recovery than C18 alone.
Validation
Matrix Effect
Signal suppression or enhancement may occur due to the influence of matrix components on the release of ions from the electrospray droplets to the gas phase or due to competition between the analytes and co-eluents for the available charge. To evaluate the matrix effect, the slope ratios obtained in the matrix-matched calibration were compared to those obtained in pure solvent, and signal suppression/enhancement (SSE) values were calculated for the analytes [32, 33]. Usually, an SSE value between 0.8 and 1.2 demonstrates that signal suppression or an enhancement effect is negligible, while values outside this range indicate a strong matrix effect [24]. The SSE values of target analytes are shown in Figure 4. A matrix effect was noted for several analytes, indicating that matrix-matched calibration curves should be used to avoid matrix interference.
Linearity
To avoid matrix interference, matrix-matched calibration curves were used in this study. Different concentration levels of standard series (5, 10, 50, 100, and 200 μg/kg) were added to the blank samples and subjected to sample pretreament, prior to detection by UHPLC–MS/MS. At last, matrix-matched calibration curves for the detection of antibiotics were obtained using the integrated peak areas of the target compounds against their concentrations. These analyses showed good linearity for the selected antibiotics with correlation coefficients (R2 > 0.995) (Table 2).
Linear equations and correlation coefficients of the assessed antibiotics
Compound | Linear equation | R 2 |
---|---|---|
Roxithromycin | y = 44.46x + 59.17 | 0.998 |
Azithromycin | y = 21.61x + 5.855 | 0.996 |
Clarithromycin | y = 20.55x − 94.0 | 0.994 |
Erythromycin | y = 26.69x + 35.16 | 0.997 |
Sparfloxacin | y = 29.62x + 304.0 | 0.995 |
Gatifloxacin | y = 149.0x − 120.3 | 0.991 |
Enrofloxacin | y = 527.9x + 3185 | 0.997 |
Lomefloxacin | y = 80.72x + 161.9 | 0.995 |
Sulfachloropyridazine | y = 24.56x + 197.8 | 0.996 |
Sulfamethoxydiazine | y = 217.9x + 447.4 | 0.998 |
Sulfameter | y = 179.6x + 275.2 | 0.999 |
Sulfamonomethoxine | y = 17.52x + 61.42 | 0.993 |
Sulfadimidine | y = 264.0x − 1.038 | 0.999 |
Sulfamerazine | y = 145.2x + 4.673 | 0.999 |
Sufisomezole | y = 84.87x + 486.4 | 0.994 |
Sulfadiazine | y = 127.6x + 103.1 | 0.999 |
Accuracy and Precision
Recoveries and intra-day precision were evaluated by adding standard to analyte-free egg samples at low, intermediate, and high concentration levels. Inter-day precision was assessed using the intermediate concentration level. The mean recoveries and coefficients of variation of the target compounds are shown in Table 3. The relative standard deviations (RSDs) of the intra-day precision ranged from 6.4% to 21.2% and the RSDs of the inter-day precision ranged from 7.4% to 22.3% in samples spiked at a concentration of 50 μg/kg. Figure 5 shows the MRM chromatograms of blank samples that were fortified with target analytes at a concentration of 50 μg/kg. All these data demonstrated that the established method has good accuracy and precision.
Validation parameters of the developed method
Compound | Recovery (%)a | LOD (μg/kg) | LOQ (μg/kg) | Inter-day precision | ||
---|---|---|---|---|---|---|
10 μg/kg | 50 μg/kg | 100 μg/kg | ||||
Roxithromycin | 74.3 (16.0) | 127.4 (13.9) | 110.1 (12.2) | 0.6 | 2.0 | 17.2 |
Azithromycin | 99.8 (20.9) | 98.8 (18.7) | 117.8 (8.0) | 0.4 | 1.3 | 21.8 |
Clarithromycin | 118.2 (11.9) | 99.7 (14.0) | 99.8 (8.1) | 0.1 | 0.3 | 16.4 |
Erythromycin | 78.4 (20.0) | 113.4 (21.2) | 110.3 (9.9) | 0.9 | 3.0 | 22.3 |
Sparfloxacin | 81.3 (11.9) | 79.1 (10.5) | 79.5 (9.4) | 0.1 | 0.3 | 15.2 |
Gatifloxacin | 104.7 (10.4) | 82.9 (9.5) | 83.0 (6.1) | 0.1 | 0.3 | 11.7 |
Enrofloxacin | 98.6 (8.9) | 88.78 (11.8) | 95.4 (5.5) | 0.3 | 1.0 | 12.5 |
Lomefloxacin | 122.8 (10.1) | 98.3 (15.6) | 97.8 (7.6) | 0.4 | 1.3 | 16.4 |
Sulfachloropyridazine | 85.3 (18.2) | 88.1 (14.2) | 91.5 (12.6) | 0.4 | 1.3 | 18.8 |
Sulfamethoxydiazine | 79.2 (20.4) | 93.6 (13.5) | 80.49 (7.4) | 0.2 | 0.7 | 21.7 |
Sulfameter | 81.2 (11.6) | 76.3 (8.2) | 82.6 (5.6) | 0.3 | 1.0 | 12.5 |
Sulfamonomethoxine | 119.2 (15.1) | 105.8 (14.5) | 100.5 (2.7) | 0.4 | 1.3 | 18.3 |
Sulfadimidine | 88.3 (9.3) | 83.2 (8.4) | 85.8 (7.1) | 0.5 | 1.6 | 7.9 |
Sulfamerazine | 99.7 (13.0) | 76.1 (6.4) | 83.8 (8.0) | 0.4 | 1.3 | 9.6 |
Sufisomezole | 83.5 (30.8) | 90.8 (10.6) | 90.2 (10.0) | 0.3 | 1.0 | 20.6 |
Sulfadiazine | 86.9 (13.0) | 73.8 (6.8) | 81.8 (6.3) | 0.8 | 2.6 | 7.4 |
Inter-day precision is given in brackets (n = 5).
LOD and LOQ
The LODs and LOQs were evaluated based on the peak area of the lowest concentration matrix-matched calibration solution in the linear range, and the results are described in Table 3. The LOD ranged from 0.1 to 0.9 μg/kg, while the LOQs were <3.0 μg/kg.
Application to Real Samples
The developed method was applied to 20 preserved egg samples collected from markets in WuHan. The results indicated that the method was suitable for the detection of analytes in real samples, and all the samples analyzed in this study presented levels of antibiotics lower than the LOQ.
Conclusions
A new method based on a QuEChERS extraction procedure and UHPLC–MS/MS detection was developed and validated for the simultaneous determination of 16 antibiotics. Samples were extracted with 4 mL water and 10 mL acetonitrile with 1% acetic acid, purified by PSA and C18 sorbent in the presence of MgSO4, and then detected by UHPLC–MS/MS. The results demonstrated good linearity, accuracy, precision, LOD, and LOQ, which indicated that the proposed method was highly sensitive and could efficiently determine trace amounts of these 16 antibiotics in preserved egg samples.
Acknowledgments
This research was supported by Hubei Provincial Centre for Disease Control and Prevention funded project (Y2013W01, Y2013W07). Meanwhile, we thank Elixigen Corporation (Huntington Beach, California, USA) for helping in proofreading and editing the English of final article.
References
- 6.
Chico J. ; Rúbies A.; Centrich F.; Companyó R.; Prat M. D.; Granados M. J. Chromatogr. A 2008, 1213, 2.
- 7.↑
Frenich A. G. ; Aguilera-Luiz M. D. M.; Vidal J. L. M.; Romero-González R. Anal. Chim. Acta. 2010, 661, 2.
- 8.↑
Lopes R. P. ; Reyes R. C.; Romero-González R.; Vidal J. L. M.; Frenich A. G. J. Chromatogr. B 2012, 895-896, 1.
- 9.↑
Commission regulation (EU) No. 37/2010 of 22 December 2009 on pharmacologically active substances and their classification regarding maximum residue limits in foodstuffs of animal origin.
- 10.↑
Capriotti A. L. ; Cavaliere C.; Piovesana S.; Samperi R.; Laganà A. J. Chromatogr. A 2012, 1395, 23.
- 12.↑
Bogialli S. ; Ciampanella C.; Curini R.; Corcia A. D.; Laganà A. J. Chromatogr. A 2009, 1216, 40.
- 13.↑
Spisso B. F. ; Ferreira R. G.; Pereira M. U.; Monteiro M. A.; Cruz T. Á.; Costa R. P. D.; Lima A. M. B.; Nóbrega A. W. D. Anal. Chim. Acta. 2010, 682, 1–2.
- 18.↑
Gigosos P. G. ; Revesado P. R.; Cadahía O.; Fente C. A.; Vazquez B. I.; Franco C. M.; Cepeda A. J. Chromatogr. A 2000, 871, 1–2.
- 19.↑
Heller D. N. ; Ngoh M. A.; Donoghue D.; Podhorniak L.; Righter H.; Thomas M. H. J. Chromatogr. A 2002, 774, 1.
- 21.↑
Peters R. J. B. ; Bolck Y. J. C.; Rutgers P.; Stolker A. A. M.; Nielen M. W. F. J. Chromatogr. A 2009, 1216, 46.
- 22.↑
Huet A. C. ; Charlier C.; Singh G.; Godefroy S. B.; Leivo J.; Vehniäinen M.; Nielen M. W. F.; Weigel S.; Delahaut P. Anal. Chim. Acta. 2008, 623, 2.
- 23.↑
Lombardo-Agüí M. ; García-Campana A. M.; Gámiz-Gracia L.; Cruces-Blanco C. Talanta 2012, 93, 15.
- 24.↑
Lopes R. P. ; Reyes R. C.; Romero-González R.; Frenich A. G.; Vidal J. L. M. Talanta 2012, 89, 30.
- 25.
Chiaochan C. ; Koesukwiwat U.; Yudthavorasit S.; Leepipatpiboon N. Anal. Chim. Acta. 2010, 682, 1–2.
- 27.↑
Romero-González R. ; Frenich A. G.; Vidal J. L. M.; Prestes O. D.; Grio S. L. J. Chromatogr. A 2011, 1218, 11.
- 28.↑
Aguilera-Luiz M. M. ; Martínez Vidal J. L.; Romero-González R.; Frenich A. G. Food Chem 2012, 132, 4.
- 29.
Tang C. ; Huang Q.; Yi-Yi U.; Peng X. Chinese J. Anal. Chem. 2009, 37, 8.