Subscriber access provided by University of Sussex Library
Article
Direct coupling of solid-phase microextraction (SPME) with mass spectrometry: sub-pg/g sensitivity achieved using a dielectric barrier discharge ionization (DBDI) source Mario Francesco Mirabelli, Jan-Christoph Wolf, and Renato Zenobi Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.6b01507 • Publication Date (Web): 22 Jun 2016 Downloaded from http://pubs.acs.org on June 28, 2016
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 14
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
Direct coupling of solid-phase microextraction (SPME) with mass spectrometry: sub-pg/g sensitivity achieved using a dielectric barrier discharge ionization (DBDI) source Mario F. Mirabelli, Jan-Christoph Wolf and Renato Zenobi* ETH Zurich, Department of Chemistry and Applied Biosciences, 8093 Zurich, Switzerland. ABSTRACT: We report a new strategy for the direct coupling of SPME with mass spectrometry, based on thermal desorption of analytes extracted on the fibers, followed by ionization by a dielectric barrier discharge ionization (DBDI) source. Limits of detection as low as 0.3 pg/mL and a linear dynamic range of ≥ 3 orders of magnitude were achieved, with a very simple and reproducible approach. Different from DART, DESI or LTP, the desorption of the analytes from the SPME devices in our setup is completely separated from the ionization event. This enhances the reproducibility of the method and minimizes ion suppression phenomena. The analytes were quantitatively transferred from the SPME to the DBDI source, and the use of an active capillary ionization embodiment of the DBDI source greatly enhanced the ion transmission to the MS. This, together with the extraordinary sensitivity of DBDI, allowed sub-pg/mL sensitivities to be reached, and to skip conventional and time-consuming chromatographic separation.
INTRODUCTION Since its introduction in 1990 by Pawliszyn and Arthur [1], SPME has become one of the most frequently used sampling techniques, and is nowadays very widespread for fast and solvent-free sample enrichment and clean-up [2]. A large number of methods for analysis of different compound classes have been reported in the literature, also thanks to the large selection of different commercially available fiber coatings. Before the introduction of ambient ionization techniques like desorption electrospray ionization (DESI) in 2004 [3] and direct analysis in real time (DART) in 2005, [4] SPME was mainly used in combination with gas chromatography (GC) and more marginally to liquid chromatography (LC). In the case of GC, the SPME fiber is introduced in the injection port, and the analytes are thermally desorbed and focused at the head of the column, before being chromatographically separated. Depending on the volatility of the compounds, a cryofocusing system might be required to allow a good focusing at the column’s head. Direct interfacing to MS are less widespread compared to chromatographic approaches, although different direct couplings have been realized over the last years [5]. A variety of geometrical configurations of SPME were used for diverse applications, including environmental [6], bioanalytical [7] and food analysis [8]. One example of direct coupling is based on the direct introduction of the fiber inside the high vacuum region of the ionization source (e.g., electron impact, EI) before mass spectrometric detection. The analytes are thermally desorbed, and the very low pressure inside the ionization source also facilitates their volatilization. When elemental information about metals is wanted, inductively coupled plasma (ICP) can be used as ionization source, in combination with thermal or solvent-assisted desorption [9,10]. Laser desorption/ionization (LDI) can also be used to couple SPME to mass spectrometry, with both vacuum and atmospheric pressure (AP) ionization. The vacuum LDI approach
was first reported by Chen and Sun [11]. It is easy to realize but quite laborious, since the fibers have to be manually positioned on the metal target plate, limiting the throughput. APMALDI analysis of SPME was initially reported by Pawliszyn’s group [12], which used a silanized optical fiber for extraction, with the matrix αHCCA being applied onto the fiber before laser desorption. This approach may be problematic for low molecular weight molecules, due to the presence of matrix peaks in the low mass range. To overcome this limitation, surface-enhanced LDI (SELDI) can be used [13]. In atmospheric pressure interfaces to mass spectrometry (APIMS), SPME materials are usually desorbed with a solvent that has a high affinity for the compounds of interest, and then ionized by electrospray ionization (ESI) [14] or atmospheric pressure chemical ionization (APCI) [15], depending on the nature of the analytes. In these cases, the need to desorb the analytes with an organic solvent, usually with a volume of ≥ 70 µL, limits the enrichment potential of SPME. The coupling of SPME to nano-ESI has also been explored [16], and small desorption volumes (i.e., less than 10 µL) have been used in combination to nano-ESI to increase the enrichment factor. The “ambient revolution” in mass spectrometry opened new possibilities in the use of SPME, and in addition to the standard fiber geometry, other arrangements have been tested for specific applications. Direct interfacing between SPME devices and ambient MS is developing rapidly these years [5], the driving force behind this interest being the possibility to perform a fast sample enrichment and clean-up in one single step, skipping chromatographic separation, and greatly simplifying and shortening the whole analytical procedure. Intrinsic characteristics of the direct analysis of SPME devices by ambient mass spectrometry ionization sources are ease of use, high sensitivity, and high throughput, since analytes are ionized and detected simultaneously. As recently reported by Pawliszyn’s group, SPME can be efficiently coupled with MS using DART [17], which achieves
ACS Paragon Plus Environment
Analytical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
remarkable signal linearity and very low limits of detection (LODs) for diazepam, cocaine and other prohibited substances. Online coupling of in-tube SPME to DART-MS was also reported [18]. Nano-ESI was recently used in combination with SPME for analysis of small organisms [19], using surface-coated probe nanoelectrospray ionization mass spectrometry (SCP-nanoESI-MS). Specific perfluorinated compounds could be quantified by modifying a C18 anion exchange adsorbent on the surface of a metal probe. Biocompatible SPME fibers have also been employed in combination with nano-ESI [20] for bioanalytical applications, having the advantages of being compatible with in-vivo sampling and to minimizing protein adsorption to the extraction phase when complex biological matrices are extracted [7]. This Bio-SPME-nano-ESI approach showed good linearity and accuracy, but the requirement of disposable nano-ESI emitters and need for careful handling and control during desorption and ionization could represent a limitation in the ease of use and operational costs. The same authors reported the use of coated-blade SPME devices, from which the analytes were electrosprayed [21]. Other approaches are based on SPME-DESI-MS [22] and solvent-assisted spray-based SPME-MS [23]. Recently, another SPME-LTP-MS coupling was reported [24]. The authors used an ionization source based on active capillary ionization for the analysis of chemical warfare simulants, with custom-made SPME materials employed for both extraction and as ionization electrode in the DBDI source. Reported LODs in water were above 24 ng/g. However, since the SPME material is constantly surrounded by reactive plasma species, the resulting ionization is harder and significant fragmentation/degradation products could be expected for labile compounds. At the moment, with the exceptions of solvent desorption, all direct couplings with ambient mass spectrometry ionization sources reported in the literature are based on the simultaneous desorption/ionization of analytes from the SPME devices. This is the case for DART, DESI, blade spray and for LTP, where the desorption step is coupled to the ionization step. Although this does not necessarily represent a drawback, it leads to less reproducible results because the entire process happens in an open environment; in such a situation, the geometric parameters (distances and angles between SPME device, ionization source and MS) play a crucial role and have to be very carefully optimized and controlled. Here we report a direct interfacing of SPME with a flowthrough DBDI source, where desorption from the SMPE device and ionization are spatially and temporally separated. DBDI sources have become increasingly widespread during the last few years [25], and are mostly being used for ambient mass spectrometry. The one used in this study was developed by our research group [26]. It was employed in several applications, including the detection of chemical warfare agents [27, 28] and for breath analysis [29]. It was also recently coupled with liquid chromatography for ultra-trace pesticide analysis in food matrices [30], with LODs at the low pg/mL level. We show that direct SPME-DBDI-MS coupling can effectively compete with chromatography for special applications with reduced matrix loads. The sensitivity achieved allowed the detection of pesticides and drugs at the sub-pg/mL concentration level.
Page 2 of 14
EXPERIMENTAL SECTION Materials Cocaine, cocaine-D3, diazepam, diazepam-D5, imipramine, atrazine, ametryn, parathion, triethyl thiophosphate, metolachlor and chlorpyrifos methyl were obtained from Sigma-Aldrich (Buchs, Switzerland). In this study commercial 100 µm PDMS and 65 µm PDMS/DVB SPME fibers (Sulpelco, Bellefonte, PA) were used. Mass spectrometry Initial optimization of the desorption conditions was performed on a LCQ Deca XP (Thermo Scientific, San José, CA, USA). Full optimization and final data were acquired using a Thermo LTQ Orbitrap, in full scan mode with a resolution of 30,000 (FWHM at m/z 400). The LTQ interface parameters were as follows: capillary voltage, 4V; tube lens voltage, 65 V; capillary temperature, 275°C. The acquisition was performed with a mass window of 50 to 400 m/z, with 1 micro scan, and with a maximum injection time of 250 ms. Automatic gain control was used. Desorption system In order to achieve a high sensitivity, the compounds extracted by SPME fibers have to be quantitatively desorbed in a short amount of time: the shorter this time, the higher the gas-phase concentration of analyte and, therefore, the sensitivity. As commonly observed in thermal desorption SPME, also in our setup, the desorption temperature was found to play an important role in the desorption profile. The vaporization of analytes from the SPME devices was achieved in a lab-built stainless steel desorption chamber (Figure 1), which was connected to the DBDI source (and therefore to the MS). A constant nitrogen gas flow humidified to 90% (R.H. at 25ºC) was allowed to pass through it, due to the underpressure inside the MS. The use of a split gas inlet system allowed to maintain an overflow of heated gas behind the position of the SPME fiber. This way, both the desorption chamber and the source were maintained at ambient pressure, an important requirement for a correct operation of the DBDI source employed. In addition, since the fiber is positioned after the split system during desorption, everything that is desorbed enters the DBDI source. Due to its unique construction design, our DBDI source can be imagined as an elongation of the MS ion transfer capillary. This allowed continuous sampling/ionization and a virtually 100% transmission for analytes (in neutral form) from the desorption chamber to the DBDI. In order to optimize the desorption process, desorption chambers with different inner diameters (i.e., 1.6, 3.4 and 5.0 mm) were tested. While liners of 0.75 mm i.d. are generally employed for SPME-GC, our system requires a much higher gas flow rate than the few mL/min used in GC systems. The gas flow rate through the ionization source depends on the type of instrument and ion transfer line used; it was 0.7L/min for the LTQ Orbitrap and 1.5L/min for the LCQ Deca XP. This flow was incompatible with a 0.75mm i.d. chamber with an inserted SPME fiber. A dramatic difference in desorption kinetics was found for different chambers. A temperature of ≥ 250°C and a desorption chamber of 1.6 mm i.d. were optimal, resulting in sharp desorption profiles for both fiber tested, with no carryover effect observed after 5 min of desorption.
ACS Paragon Plus Environment
Page 3 of 14
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
Figure 1. (left) schematic and photograph of the SPME-DBDI setup and its interfacing with a Thermo LCQ; (right) TIC and EICs of a mix of 6 analytes at 1 ng/mL concentration in water with a 65 µm PDMS/DVB fiber. Extraction was performed for 30 minutes, at room temperature and with magnetic stirring of 700 rpm, desorption temperature was 250°C, desorption chamber i.d. was 1.6 mm.
Ionization source Our DBDI source can be easily interfaced to any MS system having an atmospheric pressure interface. It consists of two concentric electrodes, onto which an alternating current is applied. The operating frequency was 5.75 kHz, at a voltage of 1.6 kVp-p. The ionization is soft due to the characteristics of the resulting low temperature plasma. It is generated between two ring-shaped electrodes. Regular air or nitrogen can be used as reagent gas, with very similar performance. Nitrogen was used here, to avoid the interference of room air contaminants (e.g., plasticizers) and to avoid oxidation of the SPME fibers during the desorption. The nitrogen was humidified to 90% (R.H. at 25ºC) to maximize the ionization efficiency [30]. The residence time of the gas-phase analytes in the plasma itself is short, resulting in a very low energy transfer from the plasma to the molecules. Therefore, they are ionized but not fragmented significantly, yielding mostly [M+H]+ ions. Because of the fast desorption and ionization processes, the extent of fragmentation did not depend much on the vaporization temperature, but rather on the transit of the molecules through the reactive plasma. The sampling of the ions from the source is a crucial aspect, as such ambient sources are known to be strongly affected by geometric parameters. Our setup employs a highly efficient ion sampling technique, the so-called “active capillary sampling”: the neutral analytes are drawn into an extended inlet capillary of the MS, and ionized during transfer into the vacuum. This differs from other (ambient) ionization sources, as the ionization happens in a confined volume inside the source itself, and not in an open environment like in ESI, APCI or conventional LTP sources. Since the source is constructed as an extension of the MS-inlet itself, the robustness and ion transmission into the MS is greatly increased. SPME extraction Optimization of extraction conditions is crucial in SPME. After evaluating the extraction time profiles for diazepam, cocaine, atrazine, ametryn, parathion and triethyl thiophosphate using the 65 µm PDMS/DVB and the 100 µm PDMS fibers, a pre-equilibrium extraction time of 30 min was chosen. The extraction was performed at room temperature (25°C) using a small magnet to stir the liquid with a magnetic hotplate (700 rpm). The extraction volume was 1.4 mL.
Quantification Quantification was performed using 65 µm PDMS/DVB fibers, by analyzing water samples spiked at concentrations between 10 pg/mL and 300 ng/mL. Cocaine-D3 (2 ng/mL), diazepam-D5 (3 ng/mL), metolachlor (2 ng/mL) and chlorpyrifos methyl (2 ng/mL) were used as internal standards. Samples volume was 1.4 mL. For each analyte, a mass window of 2 ppm was used for creation of extracted ion chromatograms (EICs) and integration. The signal was integrated for 0.5 minutes starting from the moment the fiber was entered the desorption chamber. This time interval ensured that most of the ion signal was integrated. In all cases, a 1/x weighting was used for the calibration. The LODs were evaluated experimentally, by analyzing samples with decreasing concentrations of analytes, until no more signal was observed. Because of the high resolution of the LTQ Orbitrap, no background noise was observed in the extracted ion chromatograms within a 2 ppm mass tolerance window around the exact mass of the compounds considered. Therefore, the classical LOD determination based on the evaluation of the signal-to-noise ratio could not be applied.
RESULTS AND DISCUSSION Unlike other approaches, the SPME device has not to be positioned very precisely in front of the mass spectrometer in our approach, and a good reproducibility can be attained even if the geometric parameter (angles, distances) between the ion source, the SPME and the MS are not precisely controlled. The fibers can be manually introduced into the desorption chamber, because the only relevant geometric parameter to be controlled to obtain high reproducibility of the results are independent of the positioning of the SPME, namely the DBDI parameters (inter-electrode distance, applied voltage, operating frequency) and the desorption parameters (temperature, chamber inner diameter). Optimal ionization conditions (i.e., highest ion abundance) were achieved with optimized electrode positions [30]. The ion abundance is maximized for a specific electrode position, which was found to be virtually independent of the analyte type. The plasma characteristics (e.g., density of reactive species, and plasma length) also depend on the applied voltages and on the operating frequency.
ACS Paragon Plus Environment
Analytical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
A good and rapid desorption of analytes from the SPME fibers was achieved by optimizing the desorption temperature, and by using a small diameter chamber. These parameters have to be optimized during the method development, and cannot be accidentally modified during measurement. In fact, in set-ups where tight control of angles and distances is needed, the use of high precision translation/rotation stages is required, while we were able to operate our system manually, with minimum attention required by the operator during measurements.
Figure 2. Desorption profiles for diazepam (1 ng/mL) using the same desorption chamber (1.6 mm i.d.) but different desorption temperatures for a 65 µm PDMS/DVB fiber (n=3).
Page 4 of 14
Since the flow rate in our desorption chamber was much higher than that in a conventional GC injector, the desorption profiles of the used SPME fibers was evaluated (Figure 2). These profiles not only depend on the temperature, but also on the carrier gas flow. In our system, the best desorption profiles were obtained with a 1.6 mm desorption chamber, which led to a higher gas flow velocity around the fiber. With the optimized parameters reported in the experimental section peak widths (FWHM) were less than six seconds for the analytes considered. For cocaine and diazepam, 95% and 94% of ion signal was within the first 0.5 minutes of desorption, respectively (Figure 2). Triazine and organophosphorus pesticides showed the sharpest desorption profiles, with 100% desorption within 0.5 minutes. Diazepam was most strongly retained by the fiber during desorption, but it was also completely desorbed within 4 minutes. With the reported performance, no cryo-focusing system is needed to further reduce the ion peaks width. Quality control procedures were carried on routinely, with blank experiments, using the same analytical workflow, but without extraction and/or with extraction from pure water. No carryover effect was observed after 5 min of desorption. This was verified by observing the EICs, showing no residual signal after that time, and by desorbing the fiber a second time after the first desorption. The absence of carryover was observed over the entire concentration range, from 10 pg/mL to 300 ng/mL. This represents a clear advantage over other techniques in the literature, where carryover was observed for concentrated samples, even after rinsing of the SPME devices with solvent [17,21]. In these cases, however, a direct comparison is not possible since a higher amounts of extraction phase than in the case of the fibers was used, and different desorption strategies were employed.
Figure 3. Quantification of selected drugs and pesticides in water. a) Calibration curve for cocaine (10 pg/mL to 300 ng/mL), with its isotopologue, cocaine-D3, as internal standard (2 ng/mL). b) Calibration curve for diazepam (10 pg/mL to 300ng/mL), with its isotopologue, diazepam-D5, as internal standard (3 ng/mL). c) Calibration curve for ametryn (10 pg/mL to 30 ng/mL), with metolachlor (2 ng/mL) as internal standard. d) Calibration curve for parathion (10 pg/mL to 30 ng/mL), with chlorpyrifos methyl (2 ng/mL) as internal standard. Analyses were performed in duplicate for each concentration.
ACS Paragon Plus Environment
Page 5 of 14
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
Having optimized the desorption and obtained very fast desorption profiles, we faced some technical challenges that limit the range of instrumentation that can be employed in practice. Since no chromatography is performed, all compounds are desorbed from the fiber and ionized at the same time. The desorption process usually liberates most absorbates in less than 10 seconds, in which all ions also have to be analysed. For targeted analysis of few compounds this is not a challenge, since a sufficient number of MS/MS transitions can be monitored during this time interval, e.g., with an ion trap or a triple quadrupole mass spectrometers. However, when the number of compounds to be detected is too high or especially when part of them are unknown, the use of a high resolution (preferably higher than 30,000) mass analyzers is required, since they can deliver full scan spectra from which the elemental composition of the ions can be determined. The LTQ Orbitrap used in this study was only capable of 1 full scan acquisition every 0.8 seconds at a resolution of 30,000 (FWHM at m/z 400). The use of newer Orbitrap instruments, capable of several scan/s with the same resolution, would allow a better peak shape to be obtained, which should improve the overall quantitation performance. Compared to the Orbitrap instrument used, a triple quadrupole would easily allow to increase sensitivity by at least one order of magnitude. Calibration curves for cocaine, diazepam, ametryn and parathion are reported in Figure 3. In addition to the very high sensitivity, another advantage of our technique is its high reproducibility (Figure 4). Based on isotopically labeled internal standards, we found intra-day relative standard deviation (RSD) values of 2.1% for diazepam and 3.7% for cocaine, from water solutions at a concentration of 1 ng/mL. Inter-day (5 days) RSD values for the same analytes were 2.9% and 2.1%, respectively. The use of internal standards is preferred especially when the number of analytes in the sample is high or when matrix effects are not negligible. Another advantage is that they are also desorbed from the fiber with the same kinetics of their isotopologues, allowing to compensate for time-dependent suppression effects, if present. In our experiments and for the reported concentration range, such effects were not observed, and the quantification can also be performed with non-isotopologue internal standards or even without any internal standard. Because of the high sensitivity achieved, shorter extraction times can be used, provided a precise control of the time, without significantly affecting the sensitivity. As an example, for the analytes tested, an extraction time of 5 min delivered 20 to 27% of the total ion signal compared to that of a 30 min extraction, which was still sufficient to obtain LODs in the low pg/mL level. They were 0.3 pg/mL for cocaine and diazepam, 1 pg/mL for ametryn and 3 pg/mL for parathion. These values are very satisfactory, and among the lowest reported to date. As a comparison, an analogous SPME-MS direct coupling application, based on coated blade spray, allowed to obtain LODs of 0.1 pg/mL for cocaine in PBS [21]. This value is lower than the one reported here, but it was obtained using a triple quadrupole instrument in tandem MS mode, and using an extraction device significantly larger than a SPME fiber. A direct comparison is therefore difficult. In addition, since MS/MS transition were used, a non-targeted screening was not possible, while in our case full scan spectra are acquired continuously. LOD values for cocaine in urine obtained for SPME analyzed in transmission mode by DARTMS/MS were 2 pg/mL [17] and 100 pg/mL [31]. When a similar low temperature plasma ionization source (based on DBDI)
is used for direct coupling [24], the achieved LODs for organophosphate compounds in water were 24, 31 and 68 ng/g for DMMP, DEEP and PinMPA, respectively, more than four orders of magnitude higher than our results for parathion. Intra-day and inter-day repeatability are very good when an isotopically labelled internal standard is used, i.e., for cocaine and diazepam, and acceptable when no standards are used. For water samples, when no internal standard is used, RSD% remained below 10% for intra-day and below 20% for inter-day, except for atrazine (Figure 4). This most likely results from changes in the ionization efficiency, which is common when using ambient ionization techniques. The inter-day RSD was calculated based on the signal response during different days, and it can be easily reduced with an internal standard (not necessarily an isotopologue), as shown for cocaine and diazepam. Short (e.g., 1 min or less) extractions are preferred for analyte concentrations in the ng/mL to µg/mL range, and should be performed with an autosampler to avoid high RSD% due to the difficulty of finely controlling the extraction time. Because of the simplicity of our setup, the entire extraction and thermal desorption process can be easily automated with commercial autosamplers, enabling high throughput (up to 1 sample per minute), with the minimum analysis time mostly dependent on the extraction time used.
Figure 4. Intra-day and inter-day relative standard deviation evaluated at 1 ng/mL (n=5) for selected pesticides and drugs in pure water, tap water and surface water.
Real samples: matrix effect When dealing with real samples, matrix effects are often substantial. To investigate the robustness of our method, we performed a series of tests on the compatibility of our DBDI source with the presence of co-extracted interfering molecules. Two surface water samples were collected in the proximity of our campus, and analyzed first to exclude the presence of de-
ACS Paragon Plus Environment
Analytical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
tectable levels of the analytes under investigation. One of the samples was particularly turbid. No filtering was performed before extraction. Interestingly, a low pesticide contamination could be detected in these surface waters (Table 1). compound 1-naphtyl acetamide a
b
m/z theoretical 186.0913
m/z measured 186.0912
∆m (ppm) -0.14
C (pg/mL) 10-20
metalaxyl
280.1543
280.1537
-0.64
