Data Loading...
Coupling IC to high-resolution accurate mass spectrometry Flipbook PDF
Coupling IC to high-resolution accurate mass spectrometry
138 Views
119 Downloads
FLIP PDF 3.69MB
EXECUTIVE OVERVIEW
Coupling Ion Chromatography to High-Resolution Accurate Mass Spectrometry for the Fast and Robust Analysis of Anionic Pesticides in Fruits and Vegetables by Łukasz Rajski (University of Almería, Spain) This article covers the application and evaluation of ion chromatography coupled to Orbitrap mass spectrometry for multiresidue detection of polar pesticides in fruits and vegetables. The validated method presented is fast and easy to implement for routine analysis. In all cases, it provides adequate LOQs relative to EU MRLs and the linearity, retention time variation, mass accuracy and repeatability meet the most stringent criteria of Quality Control performance for routine food control. Introduction Today we have over 1,000 pesticides, the majority of which can be analysed using very generic extraction and instrumental methods. In addition, the same LC-MS or GC-MS method can be used for many different pesticides, and multiresidue methods can be developed that allow analysis of hundreds of pesticides. However, there are still some pesticides that are difficult to address from an analytical perspective, and for which more bespoke methods are required. Some pesticides pose difficult extraction problems, others challenge the actual separation process, while still others are degraded during the analytical process posing retention problems. One group of such challenging compounds is polar pesticides. Polar pesticides actually challenge both conventional extraction processes (e.g., QuEChERS) and retention in reversed phase HPLC – the most popular separation method for pesticides. Polar pesticides cannot, therefore, be analysed using reversed phase methods which has caused considerable problems. According to the EFSA 2014 report in some EU National Control Programs, these pesticides have not even been included because of analytical problems. This is extremely alarming as some polar pesticides have very high detection rates – for example, in the case of Fosetyl-Al (controlled by fewer than five EU countries) there was a detection rate of over 30%; that is, one out of every three samples tested contained Fosetyl-Al. Extraction problems can be solved by changing the extraction solvent to a more polar one; for example, methanol, or a mixture of methanol/water. Conversely, extractions were found to be better without the addition of formic acid. To overcome retention issues, many laboratories use Thermo Scientific™ Hypercarb™ columns with variations in the mobile phase (e.g., formic acid or acetic acid), or conditioning of the column with spinach extract. However, these methods are not perfect because retention time is not always constant – in fact, subsequent injections often lead to increasingly shorter retention times. Other issues arise from the fact that some compounds in some matrices just have very low sensitivity (e.g., phosphonic acid in tomato), or very short retention times (e.g., glyphosate in strawberry).
Therefore, a different approach, to overcome these challenges is ion chromatography (IC) coupled to high-resolution mass spectrometry. In IC the stationary phase is an ion exchanger and although the technique has been used for many years, its primary application field is inorganic ions in water in conjunction with a conductivity detector. However, for analysis of pesticide residues in fruit and vegetables the conductivity detector does not have enough sensitivity or selectivity. To overcome these detector challenges IC can be coupled to high-resolution mass spectrometry. Instrumentation Figure 1 shows a modern IC-MS system, comprising a Thermo Scientific™ Dionex™ Integrion™ HPIC™ system and a Thermo Scientific™ Q Exactive™ Focus Hybrid Quadrupole-Orbitrap™ mass spectrometer.
Figure 1
An outline schematic of the experimental setup of the IC-MS setup used in these studies is shown in Figure 2. Figure 3 describes the experimental parameters. The parameters of the Q Exactive Focus MS ion source were as follows: sheath gas flow-rate: 32, auxiliary gas flow-rate: 10, sweep gas flow-rate: 0, spray voltage: 3500 V, capillary temperature: 380 °C, heater temperature: 350 °C, polarity: negative. Figure 4 shows a schematic of the Q Exactive Focus MS. For full scan, the quadrupole is wide open so the ions pass through the optics and the quadrupole, are accumulated in the C-Trap and finally are analysed in the Thermo Scientific™ Orbitrap™ mass analyser. In SIM mode, the quadrupole isolation window is very narrow, so that almost all sample matrix is Figure 2
Figure 3
removed, with only a small number of matrix ions making it to the C-Trap. In PRM mode, as with SIM mode, there is a very narrow isolation window, but before analysis in the Orbitrap mass analyser the precursor analytes are fragmented in the collision cell. Why such a complex workflow? In full scan there is normally only one ion, so another ion is required for identification – the easiest way to obtain another ion is by fragmentation. In this instance, SIM mode is required because of the low sensitivity of AMPA. Why are SIM and PRM modes more sensitive than full scan? The C-Trap is an ion trap and is responsible for gain control in the Q Exactive Focus MS system – it controls the ions that are injected into the Orbitrap mass analyser in each transient. To simplify, the C-Trap always accumulates the same number of ions. When working in full scan mode and with a particularly ‘dirty’ matrix and low concentration of the analyte, the C-Trap ion cloud will have only a small fraction of the analyte present – possibly so small that in the Orbitrap mass analyser it will be considered as noise. However, when the quadrupole is used, as in SIM and PRM modes, much of the matrix is removed meaning that a higher proportion of the ion cloud in the C-Trap is the analyte of interest. In the Orbitrap mass analyser, resolution is not constant across the whole mass range – it is inversely proportional to the square root of the massto-charge ratio. That is, for smaller masses there is more resolution. In this instance we are dealing with small ions, so the relation between m/z and resolution is very favorable. In addition, resolution depends on scanning time – more resolution requires longer scanning times, meaning fewer scans per second.
Figure 4
Eluent Generation in IC-MS A key part of the Dionex Integrion HPIC system is the eluent generation system – important for eluent gradient generation. A schematic of the portion of the system is shown in Figure 5. The electrolyte reservoir can comprise either potassium, sodium or lithium ions. The reservoir is low pressure and is separated from the high-pressure eluent generation chamber by a cation-exchange connector, which also allows cations from the reservoir to move into the chamber, but does not allow anions from the chamber to move into the reservoir. The system has two platinum electrodes which supply the necessary current. The anode is placed in the reservoir, resulting in oxidation of water to produce oxygen and hydronium ions. When one hydronium ion is released from the anode, one cation will pass through the cation-exchange connector. At the same time, on the cathode in the eluent generation chamber, there is reduction of water to hydrogen and hydroxide ions – meaning the products of this chamber are potassium, sodium or lithium hydroxide and hydrogen gas. The velocity of this reaction is controlled by the current, and by that we can control the concentration of the potassium hydroxide and create the gradient. Figure 5
Continuously Regenerated Trap Column (CRTC) The role of this device is to remove impurities from the mobile phase – the most important of which are carbonate ions. The device consists of an anion exchange bed, anion exchange membrane, anode and cathode. Figure 6 illustrates how this works. Columns Two columns were tested – the Dionex IonPac AS19 (250 mm x 2 mm x 4 µm) and the Dionex IonPac AS11-HC (250 mm x 2 mm x 4 µm). The particle diameter, pore size and crosslinking were the same for both columns, with the differences being in capacity and hydrophobicity. The Dionex IonPac AS19 column was chosen for use as the Dionex IonPac AS11-HC showed strong retention of some analytes. Suppressor The role of the suppressor is to remove potassium hydroxide from the mobile phase, and conversion of potassium salts into acids, because the effluent from the suppressor is water, potassium hydroxide and potassium Figure 6
salts. Figure 7 is a schematic of the suppressor. The suppressor has three channels with the effluent passing through the central channel, and water passing through the outer channels in countercurrent. On the anode the hydronium ions are formed and travel towards the cathode. In the central channel they neutralize the hydroxide ions, and replace the [potassium] cations in all the salts. [Potassium] cations then migrate towards the cathode to compensate the charge in the outer column – there is a reduction of water and hydroxide ions are formed. Figure 7
Figure 8
Figure 9
Conductivity Detector From an analytical point of view this detector is not very important, but using the chromatograms from this detector it is possible to determine when a lot of the matrix elutes. This provides an effective way to manipulate the gradient to separate the analytes of interest from the matrix, and control the valves to send only the analytes to the mass spectrometer. In addition, chromatograms from the conductivity detector can be used to compare different extraction methods. Figure 10
Study Results Ten analytes were studied in this work, being chlorate, perchlorate, bromide and seven organophosphorus pesticides (glyphosate, N-acetyl glyphosate, AMPA, N-acetyl AMPA, Etephon, Fosetyl-Al and phosphonic acid). All analytes showed good retention. The shortest retention time was observed for Fosetyl-Al, but it was over 5 minutes. In addition, peak shape was very good for all compounds, with the exception of AMPA, which showed some tailing. However, this could be improved by altering the gradient conditions. Figure 8 demonstrates the sensitivity improvements of SIM mode over full scan, with the example of 10 µg/kg of AMPA in onion. Figure 9 shows the sensitivity improvements of MS2 compared with full scan MS, in this case, for the analysis of 10 µg/kg of glyphosate in carrot. What was the reason for using a second pump with acetonitrile (Figure 2)? The ionisation source of the MS was electrospray and one of the key ionisation steps is desolvation. Desolvation from aqueous solution is not very efficient and the effluent from the IC is water – therefore, introducing an organic solvent (in this case acetonitrile) significantly improves the desolvation required for electrospray ionisation. Both methanol and acetonitrile were tested, but the best results were found with 0.4 mL/min of acetonitrile. Figure 10 shows extracted ion chromatograms on the left with peaks that were used for detection and quantitation, and on the right peaks that were used for identification – together with the mode the peak was extracted with. IC retention times of the analytes in various fruit and vegetable matrices were very stable. However, it was noticed that there were some variations between retention times across the different types of matrices. Another important criterion for identification is the ion ratio. Figure 11 shows three examples for glyphosate for which the permitted ion ratios are ±30% - in all three cases (solvent, carrot, onion) the ion ratio was practically the same. Working with Q Exactive high-resolution mass spectrometry, the parameter of mass accuracy was also investigated. In both MS and MS2 modes, and for different analytes across various fruit and vegetable matrices mass errors were typically below 0.1 mDa.
Figure 11
Excellent linearity from 0.01-0.5 mg/kg across both analytes and acquisition modes in an orange matrix was observed. In terms of quantitation, peak area repeatability – this was checked at two concentration levels (0.01 mg/kg and 0.05 mg/kg) across various matrices – in all cases %RSD was below 10%. In terms of quantitation results, we analysed samples from EUPT SRM-12 – the matrix was strawberry and five target polar compounds were detected (glyphosate, bromide, chlorate, phosphonic acid and N-acetyl-glyphosate). Results suggested that the quantitation was very good as all z-scores were below 2. Normally the samples were diluted five times. However in the strawberry test material very high concentrations of phosphonic acid and bromide anion (approx 20 mg/kg) were found. Subsequently, it was decided to dilute 250 times to produce concentrations similar to the concentrations from the validation study. However it was not necessary because in both cases (dilution 5 and dilution 250) the same concentration was found. Table 1 shows a series of real samples analysed by both IC-Orbitrap MS and LC-MS. The sample compounds were detected with very similar concentrations. Cations Figure 12 shows an outline of the IC-MS setup for cation analysis. The analytes tested for were trimethylsulfonium, Paraquat, Mepiquat and Diquat. Figure 13 shows a full scan MS extracted ion chromatogram for precursor ions of tomato spiked at 10 µg/kg, illustrating that retention of Table 1
Figure 12
Figure 13
About the Author Łukasz Rajski works in the research group of Dr Amadeo Fernández-Alba at the University of Almería in Spain. He studies the application of high-resolution mass spectrometry to pesticide residue detection in fruits and vegetables. His research focuses on targeted and non-targeted methods for routine pesticide analysis. He is an author of several scientific papers and book chapters about high resolution mass spectrometry and extracting pesticides from difficult matrices.
all four compounds was very good, with good peak shapes. Linearity from 0.01-0.05 mg/kg was excellent. A collaboration between:
Summary The advantages of IC-MS/MS were found to include: • Good retention of very polar compounds • Stability of retention time • Quantification by Matrix Matched Standard • Good sensitivity for all compounds • Robust chromatographic performance • Easy and fast LC-IC change