Xiao-min Xu

Direct determination of melamine in dairy products by gas chromatography/mass spectrometry with coupled column separation

A coupled capillary column system was developed for the qualitative and quantitative determination of melamine with isotope internal standard in dairy products by gas chromatography/mass spectrometry (GC/MS) without derivatization. A 30 m of DB-5 ms ((5%-phenyl)-methylpolysiloxane, 0.25 mm i.d., 0.25 microm df) coupled with a 1.5 m of Innowax (polyethylene glycol, 0.32 mm i.d., 0.25 microm df) by a quartz capillary column connector was introduced as separation column. Three advantages were discussed for the coupled system. The sample was fortified with a ring-labeled (13)C(3)(15)N(3)-melamine as an isotope internal standard and extracted by 1% of trichloroacetic acid aqueous solution. 2.2% of lead acetate solution was then added to deposit protein in the sample matrix. After purification by cation exchange cartridge, the sample solution was directly injected and detected by GC/MS. A six-point calibration curve ranging from 0.05 to 2 mg kg(-1) of melamine in sample was used to establish instrument response. The recovery was 93.9-102% with relative standard deviation from 3.1 to 8.7% when isotope internal standard used. The calculated method detection limit was 0.01 mg kg(-1).

Multi‐residue analysis of pesticides in tea by online SEC‐GC/MS

A multi-residue method for the analysis of pesticides in tea was developed by online size exclusion chromatography (SEC)-GC/MS with full scan mode. The sample was fortified with chlorpyrifos-d(10) isotope internal standard and extracted by acetonitrile. After purification by primary secondary amine sorbent and solvent exchange by SEC mobile phase, the sample was detected by online SEC-GC/MS. The purification result of the online system was evaluated by comparing the correlation between Chinese cabbage and tea matrix. The factors for method optimization included sample preparation, matrix effects and the instrument parameters of each online component. Scatter plot was introduced in this study to directly illustrate the results of the condition optimization and matrix effects in the online system. For most of the pesticides, the average recoveries ranged from 70 to 130% and the RSD were below 15%. The feasibility of the application of full scan mode in multi-residue determination of trace amounts of pesticides (LODs below 0.01 mg/kg) in a complex matrix was discussed.

Simultaneous determination of 3-monochloropropane-1,2-diol and acrylamide in food by gas chromatography-triple quadrupole mass spectrometry with coupled column separation.

Both 3-monochloropropane-1,2-diol (3-MCPD) and acrylamide are contaminants found in heat-processed foods and their related products. A quantitative method was developed for the simultaneous determination of both contaminants in food by gas chromatography-triple quadrupole mass spectrometry (GC-MS/MS). The analytes were purified and extracted by the matrix solid-phase dispersion extraction (MSPDE) technique with Extrelut NT. A coupled column (a 3 m Innowax combined with a 30 m DB-5 ms) was developed to separate both compounds efficiently without derivatization. Triple quadrupole mass spectrometry in multiple reaction monitoring mode (MRM) was applied to suppress matrix interference and obtain good sensitivity in the determination of both analytes. The limit of detection (LOD) in the sample matrix was 5 μg kg(-1) for 3-MCPD or acrylamide. The average recoveries for 3-MCPD and acrylamide in different food matrices were 90.5-107% and 81.9-95.7%, respectively, with the intraday relative standard deviations (RSDs) of 5.6-13.5% and 5.3-13.4%, respectively. The interday RSDs were 6.1-12.6% for 3-MCPD and were 5.0-12.8% for acrylamide. Both contaminants were found in samples of bread, fried chips, fried instant noodles, soy sauce, and instant noodle flavoring. Neither 3-MCPD nor acrylamide was detected in the samples of dairy products (solid or liquid samples) and non-fried instant noodles.

The bioaccessibility of polychlorinated biphenyls (PCBs) and polychlorinated dibenzo-p-dioxins/furans (PCDD/Fs) in cooked plant and animal origin foods

In this study, we compared the effect of boiling and frying food preparation methods in determining the bioaccessibility of polychlorinated biphenyls (PCBs) and polychlorinated dibenzo-p-dioxins/furans (PCDD/Fs) in rice, cabbage, milk powder, eggs, beef, and fresh water fish. We then used these data to calculate a toxic equivalent (TEQ) for risk assessment and compared it to published values that did not account for bioaccessibility. When the foods were prepared by boiling, the mean bioaccessibility (%) in rice (PCBs: 16.5±1.0, PCDD/Fs: 4.9±0.3) and cabbage (PCBs: 4.2±0.9, PCDD/Fs: 1.9±0.7) were lower than in animal origin foods (beef, PCBs: 49.0±3.3, PCDD/Fs: 7.8±0.9; egg, PCBs: 29.7±3.1, PCDD/Fs: 8.6±1.3; fish, PCBs: 26.9±2.5, PCDD/Fs: 7.9±1.3; milk powder, PCBs: 72.3±1.6, PCDD/Fs: 28.4±1.2). When fried in cooking oil, the bioaccessibilities of all analytes in all foods increased, but the increase in plant based foods (rice, PCBs: 3.4×, PCDD/Fs: 3.6×; cabbage, PCBs: 10.3×, PCDD/Fs: 7.9×) was greater than that of animal origin foods (beef, PCBs: 1.6×, PCDD/Fs: 3.4×; egg, PCBs: 2.1×, PCDD/Fs: 1.8×; fish, PCBs: 2.8, PCDD/Fs: 3.2×). Comparison of PCBs/PCDD/Fs bioaccessibility in rice and cabbage showed that bioaccessibility was greater in the low fat, high carbohydrate/protein content food (rice) than in the low carbohydrate/protein, low fat content food (cabbage), regardless of the method used to prepare the food. Adjusting for bioaccessibility reduced the gross estimated daily intake (EDI) of 112pgWHO-TEQ/day, by 88% and 63% respectively for foods prepared by boiling and frying. Our results indicate that: 1) The method used for cooking is an important determinant of PCBs/PCDD/Fs bioaccessibility, especially for plant origin foods, 2) there might be a joint fat, carbohydrate and protein effect that influences the bioaccessibilities of PCBs/PCDD/Fs in foods, and 3) use of bioaccessibility estimates would reduce the uncertainty in TEQ calculations.

Study of chloropropanols in soy sauce by gas chromatography-triple quadrupole mass spectrometry with coupled column separation without derivatisation.

A qualitative and quantitative determination method for chloropropanols in soy sauce was developed by GC-MS/MS with coupled column separation without derivatisation. The target compounds were 1,3-dichloropropan-2-ol (1,3-DCP), 2,3-dichloropropan-1-ol (2,3-DCP), 3-monochloropropane-1,2-diol (3-MCPD) and 2-monochloropropane-1,3-diol (2-MCPD). 3-MCPD-d5 was used as an isotope internal standard for MCPDs and 1,3-DCP-d5 for DCPs. After spiking with internal standards and mixed with 1 g of Extrelut™ NT, about 1 g of the sample was washed by 4 ml of hexane and the analytes were eluted with 15 ml of 95% (v/v) ethyl ether/hexane mixture. The concentrated extract was directly injected without derivatisation, separated by a coupled column – a 3 m Innowax (polyethylene glycol) combined with a 30 m DB-5 ms ((5%-phenyl)-methylpolysiloxane) by a quartz capillary column connector – and detected by GC-MS/MS. The limits of detection (LODs) in the sample matrix were 1.0, 2.5, 5.0 and 5.0 µg kg−1 for the above chloropropanols, respectively. The relative proportions of 2-MCPD/3-MCPD ranged from 0.19 to 3.74 in soy sauce samples. 2,3-DCP and 2-MCPD were more stable than 1,3-DCP and 3-MCPD under alkaline condition. The levels of chloropropanols can be decreased by an alkaline treatment process.

Determination of dioxin-like polychlorinated biphenyls in 1 mL whole blood using programmable temperature vaporization large volume injection coupled to gas chromatogram and high-resolution mass spectrometry.

A sensitive method based on programmable temperature vaporization large volume injection coupled to gas chromatogram and high-resolution mass spectrometry (PTV-GC-HRMS) has been developed for the determination of ultra trace levels of dioxin-like polychlorinated biphenyls (DL PCBs) in small amounts of human blood. Blood samples (1mL) were first extracted by column extraction and then purified with column chromatorgraphies. Final extracts (20μL) were introduced to the PTV injector under the solvent vent mode and detected by GC-HRMS (SIM mode). PTV parameters were observed by changing one factor at a time (practical conditions: vent flow: 50mLmin(-1), vent pressure: 0kPa and vent time: 0.1min), recoveries of most PCB congeners ranged from 55.1% to 108%, and method detection limits were in the range of 0.11-1.63pgg(-1).

Analysis of monofluoroacetic acid in urine by liquid chromatography-triple quadrupole mass spectrometry and preparation of the positive sample by the bioconversion from monofluoroacetamide to monofluoroacetic acid in vitro.

Whether as a rodenticide or as a natural product, monofluoroacetic acid (FAcOH) may cause poisoning to humans or animals for its high acute toxicity. Urine is one of the most typical specimens for forensic diagnosis when poisoning case about FAcOH happens. The positive sample containing FAcOH plays a key role for the development of an accurate and reliable analytical method. The bioconversion from monofluoroacetamide (FAcNH2) to FAcOH in urine in vitro was studied for the preparation of positive urine sample containing FAcOH without standard spiking or animal experiment. The average bioconversion rates were 0%, 18.6% and 41.3% when incubated the FAcNH2 spiked urine in vitro for 21days at -20°C, room temperature (RT) and 37°C, respectively. Afterwards, a fast and sensitive analytical method was developed for determination of FAcOH in urine. Samples were diluted with water containing formic acid and cleaned with polymeric anion exchange (PAX) cartridge. The acid eluate was neutralized with ammonium hydroxide and directly measured by hydrophilic interaction liquid chromatography-triple quadrupole mass spectrometry (LC-MS/MS) using basic mobile phase condition. The limit of detection and limit of quantification of FAcOH in urine were 2 and 5ngmL(-1), respectively. The linear range was 5-1000ngmL(-1) with a correlation coefficient of r=0.9993 in urine calibrated with internal standard. The recoveries at four spiking levels (5, 10, 50 and 500ngmL(-1) in urine) were 87.2%-107% with relative standard deviations ranged between 4.3%-8.8%.

Quantitative analysis of vitamin K1 in fruits and vegetables by isotope dilution LC-MS/MS

A method has been developed for quantitative analysis of vitamin K1 in vegetables and fruits by LC-MS/MS. The method employed a neutral aluminum oxide column for sample clean-up. Considering matrix effects, an isotope standard (vitamin K1-D7) was added to correct matrix interferences, as well as pretreatment losses and instrument variability. Spiked recoveries in model matrices were 88.3–90.1% for spinach, 84–110% for lotus roots, and 108–115.6% for grapes. The relative standard deviations (RSDs) were 1.0–7.1% for intra-day and 2.8–8.8% for inter-day. The limit of detection (LOD) and limit of quantification (LOQ) were 0.001 mg kg−1 and 0.004 mg kg−1, respectively. The method is sensitive and reliable, which is successfully applied to different vegetable and fruit matrices. Vitamin K1 in fruit samples ranged from 0.008 mg kg−1 to 0.401 mg kg−1, and in vegetable samples from 0.0032 mg kg−1 to 3.973 mg kg−1.