Paper Spray Chemical Ionization: Highly Sensitive Ambient Ionization


Paper Spray Chemical Ionization: Highly Sensitive Ambient Ionization...

0 downloads 104 Views 1MB Size

Article pubs.acs.org/ac

Paper Spray Chemical Ionization: Highly Sensitive Ambient Ionization Method for Low- and Nonpolar Aromatic Compounds Donghwi Kim,† Un Hyuk Yim,‡ Byungjoo Kim,§ Sangwon Cha,*,∥ and Sunghwan Kim*,† †

Department of Chemistry, Kyungpook National University, Daegu 41566, Republic of Korea Department of Chemistry, Hankuk University of Foreign Studies, Yongin 17035, Republic of Korea ‡ Oil and POPs Research Group, Korea Institute of Ocean Science and Technology, Geoje 53201, Republic of Korea § Center for Organic Analysis, Division of Metrology for Quality of Life, Korea Research Institute of Standards and Science, Daejeon 34113, Republic of Korea ∥

S Supporting Information *

ABSTRACT: Sensitivity is an important factor determining successful mass spectrometry (MS) analysis of metabolome, protein, drugs, and environmental samples. Currently, nanoelectrospray ionization (ESI) is widely used as a sensitive ionization method. However, application of nano-ESI is limited to polar molecules and there is no atmospheric pressure ionization technique developed that can be used for MS analysis of low- and nonpolar compounds with sensitivity that can match with nano-ESI. Herein, we propose paper spray chemical ionization (PSCI) as an ionization technique that can be used to analyze low- and nonpolar aromatic compounds with high sensitivity. PSCI is based on paper spray ionization utilizing corona discharge phenomenon. PSCI can sensitively and quantitatively detect down to picogram (or femtomole) levels of lowand nonpolar aromatic compounds.

F

such as the ones acquired with ESI can be obtained with PSI. In addition, previous works have reported the paper-based corona discharge phenomenon when different experimental conditions were used.21,25−27 We think that this phenomenon has great potential for use as a new ionization method. However, limited data using this phenomenon have been reported and its potential has not yet been fully explored. PSCI utilizes this phenomenon to ionize low/nonpolar compounds. The key differences between the experimental conditions for PSI and PSCI are in the choice of solvent and applied voltage. Further details are explained in the following sections. In this study, we chose oil-contaminated soil extracts as the analytes. Many polycyclic aromatic hydrocarbons (PAHs) and heterocyclic PAHs containing N, O, and S exist in petroleum and some of them have well-documented carcinogenic or mutagenic properties.28,29 When oil spill occurs, its environmental impact can last for 20 years, and thus, long-term studies on oil spills are necessary.30−32 However, in years past, spilled oil samples in natural environment have become scarce, which makes analysis difficult. Therefore, it is critical to minimize sample consumption per analysis for long-term follow-up studies.

or successful mass spectrometry (MS) analysis, it is necessary to select a suitable ionization technique depending on the sample properties. Electrospray ionization (ESI), atmospheric pressure photoionization (APPI), and atmospheric pressure chemical ionization (APCI) are commonly used atmospheric pressure ionization (API) techniques. APPI and APCI are used to analyze low/nonpolar compounds not ionizable by ESI.1−7 High sensitivity and minimal sample consumption are other important factors when choosing an ionization technique for MS analysis. Nanoelectrospray ionization (nano-ESI) has been developed and successfully applied to improve sensitivity and lower sample consumption.8−11 Nano-ESI can successfully analyze microliter quantities of sample solutions with high sensitivity. However, no technique that can ionize low/nonpolar compounds like APPI or APCI and is comparable to nano-ESI in terms of sample consumption and sensitivity has been developed. Direct analysis in real time mass spectrometry (DART-MS) can be a good alternative to APPI and APCI when only low volumes and small amounts of sample are available.12−14 But the responses of the analytes are highly dependent on the temperature of the DART source and the matrix.14 Thus, we propose that paper spray chemical ionization (PSCI) can be used to analyze low/nonpolar aromatic molecules with sensitivity comparable to that of nano-ESI. Recently developed paper spray ionization (PSI) is a fast and convenient ionization method for the analysis of drugs and biological tissues.9,15−24 It has been well documented that spectra © XXXX American Chemical Society

Received: May 9, 2017 Accepted: August 8, 2017

A

DOI: 10.1021/acs.analchem.7b01733 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry



EXPERIMENTAL SECTION Sample Preparation. Oil-contaminated soil was collected from Sinduri beach 7 years after the Hebei spirit oil spill in Taean, Republic of Korea. The exact location of the sampling site is shown in Figure S1. The sediments were homogenized and stored at −20 °C in the freezer before analysis. Approximately 10 g of the sediment sample was extracted using a bath-type ultrasonicator (Power Sonic 410, Hwashin Technology, South Korea) with 100 mL of dichloromethane (HPLC grade, Honeywell Burdick and Jacson, Ulsan, Korea) for 6 h. The soil extract was filtered using a 0.7 μm pore size glass filter (GFF, Ø 47 mm, Whatman, U.K.) along with sodium sulfate to remove traces of water. Then, the extract was transferred to the precleaned vial and residual dichloromethane was eliminated under a gentle stream of N2. Mass Spectrometric Analysis. Analyses were performed with a Q-Exactive quadrupole orbitrap mass spectrometer (Thermo Fisher Scientific Inc., Rockford, IL). The data were collected in both positive and negative ion mode over the range of m/z 100−1500 at 1.9 scans per second, providing a mass resolution of 140000 (fwhm) at m/z 200. The automatic gain control (AGC) was set at 106 and the maximum injection time was set at 100 ms. Figure 1 shows the in-house-prepared

paper by a syringe pump (Fusion 100T, Chemyx, Stafford, TX, U.S.A.) at a flow rate of 10−40 μL/min. Three ∼7 kV voltage for ionization was applied directly to the paper and other parameters were as follows: transfer temperature at 300 °C; S-Lens level at 50 V. For ESI analysis, oil samples were dissolved in toluene/ methanol (50:50, v/v) at 1 mg/mL and injected at a flow rate of 10 μL/min. ESI parameters were usual values including spray voltage, 3.9 kV; sheath, auxiliary, and sweep gas flow rates, 7, 0, and 0 respective arbitrary units; temperature of the H-ESI II probe, 300 °C, and S-lens voltage, 50 V. For APPI and APCI analysis, oil samples were dissolved in toluene at 1 mg/mL and injected at a flow rate of 50 μL/min. The vaporizer temperature was set to 350 °C. The sheath, auxiliary, and sweep gas flow rates were 10, 5, and 0 arbitrary units, respectively. The capillary temperature was 320 °C, and the S-lens voltage was 50 V. Corona current was set at the default value for APCI MS (i.e., 5 μA for positive mode, 9.5 μA for negative mode). Data Processing. The obtained MS data were processed with an in-house-developed software (Bull. Korean Chem. Soc. 2009, 30, 2665−8). Peaks were assigned to chemical formulas within 2 ppm mass error and typical conditions for crude oil analysis were considered (CcHhNnOoSs; c unlimited, h unlimited, 0 ≤ n ≤ 5, 0 ≤ o ≤ 20, 0 ≤ s ≤ 5). Class distributions were determined by adding up abundance of peaks containing the same number of heteroatoms. The HC class designates a group of compounds composed of C and H. S1, O1, and N1 classes indicate groups of compounds with one sulfur, one oxygen, and one nitrogen, respectively.



RESULTS AND DISCUSSIONS (+)PSCI was performed with nonpolar solvents such as n-hexane or dichloromethane (DCM) and +4−7 kV ionization voltage. In contrast, (+)PSI was undertaken with 50:50 n-hexane/2-butanol and +3−4 kV voltage. All the other experimental conditions were identical for both methods. In (+)PSCI-MS analysis, the spray current was about 1.5−10 μA. The exact number varied depending on the applied solvents and voltages but the spray current of (+)PSCI is generally much higher than that of ESI (0.15 ± 0.03 μA) and lower than APCI (5.00 ± 0.05 μA). The crossover between an ESI-like and CI-like ionization process as a function of potential was investigated and it was determined to be around 4−5 kV. However, the exact crossover voltage varies depending on the experimental parameters such as shape of tips and solvent composition. (+)PSCI MS spectra acquired with hexane or DCM and (+)PSI MS spectra are provided in Figure 3. Observed total ion current (TIC) for the mass spectra are also shown in Figure S3. TIC lasted longer than 5 min for each loading. The obtained spectra were processed by the procedure described in the Experimental Section. Obtained accurate mass numbers were converted into elemental composition and classified according to their chemical classes. The obtained class distributions were plotted (Figure 2a, b, and d). In the spectra obtained with hexane and 6 kV of applied voltage, ions assigned to the S1, HC, and O1S1 classes were most abundant (Figure 2a). The HC class is abundantly observed in the spectra, as HC class compounds do not have heteroatoms and thus cannot be ionized with ESI. They were also abundant in the spectra obtained with DCM (Figure 2b). Sulfur-containing compounds in oils can be ionized well with (+)APPI MS.33−37 It was also reported that the relative abundance of sulfur-containing peaks obtained with (+)APPI MS correlated well with sulfur

Figure 1. (a) Diagram showing a procedure for paper spray chemical ionization. (b) Pictures of a sample-loaded paper tip and in-houseprepared ionization source. Q-Exactive mass spectrometer (Thermo Fisher Scientific Inc., Rockford, IL) was used.

ionization source used in this study. Extracted spilled oil samples were directly loaded onto the center of the chromatography paper tip using a disposable glass Pasteur pipet (Volac, Poulten and Graf Ltd., U.K.). It has been previously reported that electric field density of paper is higher at the tips of paper cut to smaller angles.27 In this study, the paper with dimension of 6 mm base and 14 mm height was cut to have sharp tips. Figure 1b shows the loaded paper. To estimate the amount of loaded sample on the paper, oil sample was spotted 20 times onto the rectangular paper and increased mass was measured (Figure S2). As a result, it was found that each loading delivered approximately 50 μg of sample (Figure S2). For quantitative analysis using a standard compound, 4,6-dimethyldibenzothiophene was dissolved in dichloromethane at a concentration of 5 mM. The stock solution was then diluted with dichloromethane to concentrations 5 nM−50 μM. A total of 2 μL of working solution was loaded onto a paper tip using a 10 μL syringe (Hamilton, Reno, NV, U.S.A.). Solvent was fed to the loaded B

DOI: 10.1021/acs.analchem.7b01733 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

Figure 2. Full mass spectra and distribution of major chemical classes for oil-contaminated soil extract obtained by (a) positive ion mode PSCI with hexane solvent at a flow rate of 40 μL/min and 6 kV of applied voltage, (b) positive ion mode PSCI with dichloromethane solvent (1% formic acid, v/v) at a flow rate of 35 μL/min and 6 kV of applied voltage, (c) positive ion mode PSI with hexane/2-butanol (50:50, v/v) solution at a flow rate of 10 μL/ min and 6 kV of applied voltage, (d) positive ion mode PSI with hexane/2-butanol (50:50, v/v) solution at a flow rate of 10 μL/min and 3.9 kV of applied voltage; approximately 50 nL of sample was directly loaded on a paper tip for PSCI and PSI, (e) positive ion mode APPI, and (f) positive ion mode ESI.

PSI. As discussed earlier, the corona discharge phenomenon has been observed in paper-based ionization.21,25−27 Notably, the stability of PSCI was dependent on the moisture content of the air. The PSCI signal became unstable when it dropped below 20%. The moisture content dependence of the signal stability has been reported earlier.25 It is apparent that the moisture in the air plays an important role during the ionization mechanism. To overcome this limitation, 1% formic acid (FA) was added to the solvent for (+) mode PSCI analysis and a stable PSCI signal was obtained regardless of the moisture content. We investigated the role of FA during ionization by employing deuterated hexane (D-hexane) and 1% formic acid at low humidity. The obtained spectra are shown in Figure 3. Expanded mass spectra presented in Figure 3 clearly show that mostly protonated (instead of deuterated) peaks were observed in the spectra even though D-hexane was the major solvent constituent. Therefore, it was concluded that hexane was not the major proton source for PSCI. Signal instability at low humidity and data presented in Figure 3 suggest that water in the air and added formic acid serve as proton sources during PSCI. All the PSCI spectra presented here were obtained with 1% formic acid. When DCM was used as the solvent, ice was formed at the tip of the paper (Figure S4) and that can be attributed to the low boiling point of DCM at 39.6 °C.46 Because DCM evaporates rapidly, enthalpy of vaporization lowers the temperature of the paper tip and eventually forms ice. Ice formation with a lowboiling-point nonpolar solvent was previously reported.25 It was found in this study that addition of formic acid to DCM could prevent ice formation. Suppression of apparent ice formation can be explained by freezing point depression with added formic acid. Even though similar spectra were obtained from APPI and PSCI, the critical advantage of the PSCI over APPI was the sample consumption. With PSCI, analysis could be done for up

contents of oils determined by elemental analysis quantitatively.38 On the other hand, nitrogen-containing classes assigned to N1, N1S1, and N2 were the most abundant in the spectra obtained with 50:50 n-hexane/2-butanol and 3.9 kV ionization voltage (Figure 2d). Nitrogen-containing compounds in oils are sensitively detected by use of (+)ESI MS.39−43 The same samples were also analyzed by conventional (+)APPI MS and (+)ESI MS. The resulting spectra and obtained class distributions are shown in Figure 2e, f. The ionization process of APPI is initiated by photons, but gas-phase chemical reactions also play an important role in APPI.44,45 (+)APCI MS was also attempted but the sensitivity was low and the obtained data were not considered in this study. The class distribution obtained with (+)APPI MS (Figure 2e) was very similar to ones obtained with (+)PSCI (Figure 2a, b). On the other hand, the class distribution observed with (+)ESI MS (Figure 2f) was very close to the one observed with (+) PSI (Figure 2d). When hexane and 3−4 kV ionization voltage were applied, the MS signal was unstable. Small amount of polar solvent was added to nonpolar solvent, it produced ESI-like MS spectrum even though we applied much higher voltage. The combination of 50:50 n-hexane/2-butanol and +6 kV voltage was applied and the obtained data are presented in Figure 2c. The observed class distribution was similar to that observed with PSI or ESI (Figure 2d, f) and this shows that the combination of polar solvent and high voltage does not produce PSCI. The combination of polar solvent and low voltage (e.g., hexane and 3−4 kV ionization voltage) was also applied, but unstable MS signal was observed. Therefore, it was concluded that a nonpolar solvent and higher voltage were the necessary experimental conditions for PSCI. The data presented in Figure 2 clearly shows that ionization process achieved with PSCI is quite different from the one with C

DOI: 10.1021/acs.analchem.7b01733 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

Figure 3. Full mass spectrum and expanded spectra of oil-contaminated soil extract obtained by positive PSCI with hexane-d14 (1% formic acid, v/v) and 6 kV of applied voltage.

to 5 min with approximately 50 μg of spilled oil sample. In contrast, conventional APPI requires 500 μg of spilled oil sample for a single run (500 μL of 1 mg/mL solution at a flow rate of 100 μL/min). In addition, an additional amount of sample to fill up the capillary line and syringe is required. Therefore, in practice, oil samples of the order of at least mg quantity are required for APPI analysis (at least 10 times more than that required for PSCI). To further test the sensitivity and quantitative nature of PSCI, 2 pg to 20 ng (∼10 fmol−100 pmol) of 4,6-dimethyldibenzothiophene, chosen because it is an aromatic molecule ionized not well by ESI but well with APPI, was loaded on the paper and analyzed using 50:50 v/v n-hexane/dichloromethane (1% formic acid, v/v) and +6 kV. The obtained spectra are shown in Figure 4. The raw data used to plot Figure 4 are provided in the Supporting Information (Table S1). The calibration curve was constructed by plotting the sum of intensities of both radical cations and protonated cations against the amount of loaded sample. The large plot in Figure 4 covers the entire range (2 pg∼200 ng) and the inset covers the smaller range (2 pg ∼500 pg). A linear plot was obtained, showing PSCI can be effectively used for quantitative analysis. In addition, signals can be successfully obtained by consuming femtomoles of samples, which is not possible with conventional APCI and APPI. We also analyzed other compounds with different polarities to identify the limitations of the PSCI in terms of polarity of the analytes. Anthracene, 9-phenanthrol, and 9,10-anthraquinone were analyzed by positive ion mode PSCI. The obtained spectra presented in Figure S5. Anthracene is less polar compound than 4,6-dimethyldibenzothiophene. It does not have heteroatoms thus cannot be ionized by ESI. 9-Phenanthrol and 9,10anthraquinone can be ionized with both APPI and ESI. All of the three samples exhibited a good PSCI response. Up to now, we have not observed limitations of the PSCI on application to

low polar compounds. In our experiments, however, n-alkanes were not ionized by PSCI. Negative mode analyses were performed with hexane/−6 kV (PSCI mode) or 50:50 n-hexane/2-butanol/−3.9 kV (PSI mode); the mass spectra are shown in the Supporting Information (Figure S6). As shown in Figure 5, class distributions between PSCI and PSI modes were different. O2, O2S1, and O3S1 were the most abundant classes in the spectra obtained with the PSCI mode (Figure 5a), whereas O4S1 and O3S1 were dominant classes in the PSI mode (Figure 5b). Observation of oxygenated sulfur classes from spilled oils has been well documented.47,48 The O2 containing compounds are abundant in (−)PSCI but not in (−)PSI data (refer to Figure 5). This comparison may provide information on the chemical structures of the compounds. Due to the complexity of crude oil, it is difficult to predict structures solely based on the data presented in this study. However, the difference observed in the (−)PSI and (−)PSCI is certainly an area worthy of future study. To compare with conventional techniques, samples were also analyzed with (−) mode APCI; the obtained mass spectra and class distributions are given in Figures S6 and 5, respectively. (−) mode APPI was also attempted but the signal abundance was lower than that of (−)APCI. Therefore, APCI was chosen as a conventional technique for the negative mode. Comparing ionization efficiency of APPI and APCI is not within the scope of this study and hence it should not be prematurely concluded that the sensitivity of (−)APPI is always lower than (−)APCI in the oil analysis. Figure 5a, c clearly show that the class distribution obtained from (−)APCI is like the one obtained with hexane and −6 kV ionization voltage. Based on the similarity of data shown in Figure 5, it is concluded that the corona discharge phenomenon in paper is the major ionization mechanism for the negative mode and PSCI can be operated in the negative D

DOI: 10.1021/acs.analchem.7b01733 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

Figure 4. Quantitative analysis of 4,6-dimethyldibenzothiophene (2 pg−20 ng) and a plot showing the low-concentration range (2−500 pg).

stabilizing PSCI even at low humidity with nonpolar solvent. In the quantitative analysis using 4,6-dimethyldibenzothiophene, the result showed PSCI can be effectively used for quantitation of less polar aromatic compounds down to femtomole level. The key advantage of PSCI is that low/nonpolar aromatic molecules can be ionized in a quantity that cannot be analyzed with conventional APCI and APPI. Therefore, we expect that PSCI can be effective in analyzing environmental samples often limited in availability. In that sense, PSCI can be added to the list of ambient ionization techniques51 as a method that can ionize low/nonpolar molecules in high sensitivity.

mode. The mechanism for chemical ionization has been described in the previous studies.49,50 Notably, (−)PSI signals were stable even at low moisture content of air and thus adding extra acid or base was not necessary. This indicates that ionization mechanism for (−)PSI does not involve water molecules.



ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.analchem.7b01733. Additional figures showing a sampling site, raw spectra and total ion current chromatograms and a table showing values used for quantitative analysis (PDF).



AUTHOR INFORMATION

Corresponding Authors

*Phone: +82-31-330-4377. E-mail: [email protected]. *Phone: +82-53-950-5333. E-mail: [email protected]. ORCID Figure 5. Comparison of major chemical class distribution for oilcontaminated soil extract using (a) negative-ion mode PSCI with hexane, (b) negative-ion mode PSI with hexane/2-butanol (50:50, v/v), and (c) negative-ion mode APCI.

Sunghwan Kim: 0000-0002-3364-7367



Notes

Author Contributions

The manuscript was written through contributions of all authors. The authors declare no competing financial interest.



CONCLUSION In this study, we developed the PSCI MS method for the direct analysis of low/nonpolar molecules and applied to the characterization of the oil-contaminated sediments extract. The effects of the composition of spraying solvents and the amount of the applied voltage on the stability of spectrum were examined. Adding 1% formic acid to solvent played a critical role in

ACKNOWLEDGMENTS The authors acknowledge support for this work by the grant (KCG-01-2017-06) from “Development of Advanced Oil Fingerprinting System” funded by Korea Coast Guard, by the Korea Research Institute of Standards and Science under the project “Establishing Measurement Standards for Food and E

DOI: 10.1021/acs.analchem.7b01733 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

(32) Islam, A.; Ahmed, A.; Hur, M.; Thorn, K.; Kim, S. J. Hazard. Mater. 2016, 320, 123−132. (33) Purcell, J. M.; Juyal, P.; Kim, D.-G.; Rodgers, R. P.; Hendrickson, C. L.; Marshall, A. G. Energy Fuels 2007, 21, 2869−2874. (34) Panda, S. K.; Andersson, J. T.; Schrader, W. Angew. Chem., Int. Ed. 2009, 48, 1788−1791. (35) Cho, Y.; Na, J.-G.; Nho, N.-S.; Kim, S.; Kim, S. Energy Fuels 2012, 26, 2558−2565. (36) Chiaberge, S.; Fiorani, T.; Savoini, A.; Bionda, A.; Ramello, S.; Pastori, M.; Cesti, P. Fuel Process. Technol. 2013, 106, 181−185. (37) Purcell, J. M.; Merdrignac, I.; Rodgers, R. P.; Marshall, A. G.; Gauthier, T.; Guibard, I. Energy Fuels 2010, 24, 2257−2265. (38) Kim, E.; Cho, E.; Moon, C.; Ha, J.; Cho, E.; Ha, J.-H.; Kim, S. Energy Fuels 2016, 30, 6958−6967. (39) Qian, K.; Rodgers, R. P.; Hendrickson, C. L.; Emmett, M. R.; Marshall, A. G. Energy Fuels 2001, 15, 492−498. (40) Bae, E.; Na, J.-G.; Chung, S. H.; Kim, H. S.; Kim, S. Energy Fuels 2010, 24, 2563−2569. (41) Chen, X.; Shen, B.; Sun, J.; Wang, C.; Shan, H.; Yang, C.; Li, C. Energy Fuels 2012, 26, 1707−1714. (42) Shi, Q.; Xu, C.; Zhao, S.; Chung, K. H.; Zhang, Y.; Gao, W. Energy Fuels 2010, 24, 563−569. (43) Tong, J.; Liu, J.; Han, X.; Wang, S.; Jiang, X. Fuel 2013, 104, 365− 371. (44) Ahmed, A.; Choi, C. H.; Choi, M. C.; Kim, S. Anal. Chem. 2012, 84, 1146−1151. (45) Ahmed, A.; Choi, C. H.; Kim, S. Rapid Commun. Mass Spectrom. 2015, 29, 2095−2101. (46) http://webbook.nist.gov/cgi/cbook.cgi?ID=C75092&Units= SI&Mask=4#Thermo-Phase. (47) Stanford, L. A.; Kim, S.; Klein, G. C.; Smith, D. F.; Rodgers, R. P.; Marshall, A. G. Environ. Sci. Technol. 2007, 41, 2696−2702. (48) Wang, J.; Zhang, X.; Li, G. Chemosphere 2011, 85, 609−615. (49) Carroll, D. I.; Dzidic, I.; Stillwell, R. N.; Horning, M. G.; Horning, E. C. Anal. Chem. 1974, 46, 706−710. (50) McEwen, C. N.; Larsen, B. S. J. Am. Soc. Mass Spectrom. 2009, 20, 1518−1521. (51) Cooks, R. G.; Ouyang, Z.; Takats, Z.; Wiseman, J. M. Science 2006, 311, 1566.

Clinical Nutrients”, Grant No.14011049, and by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (NRF2016R1D1A1B01006576 and 2014R1A2A1A11049946).



REFERENCES

(1) Robb, D. B.; Covey, T. R.; Bruins, A. P. Anal. Chem. 2000, 72, 3653−3659. (2) Bruins, A. P. TrAC, Trends Anal. Chem. 1994, 13, 37−43. (3) Venter, A.; Nefliu, M.; Graham Cooks, R. TrAC, Trends Anal. Chem. 2008, 27, 284−290. (4) Purcell, J. M.; Hendrickson, C. L.; Rodgers, R. P.; Marshall, A. G. Anal. Chem. 2006, 78, 5906−5912. (5) Andrade, F. J.; Shelley, J. T.; Wetzel, W. C.; Webb, M. R.; Gamez, G.; Ray, S. J.; Hieftje, G. M. Anal. Chem. 2008, 80, 2646−2653. (6) D’Andrilli, J.; Dittmar, T.; Koch, B. P.; Purcell, J. M.; Marshall, A. G.; Cooper, W. T. Rapid Commun. Mass Spectrom. 2010, 24, 643−650. (7) Purcell, J. M.; Rodgers, R. P.; Hendrickson, C. L.; Marshall, A. G. J. Am. Soc. Mass Spectrom. 2007, 18, 1265−1273. (8) Wilm, M.; Mann, M. Anal. Chem. 1996, 68, 1−8. (9) Jjunju, F. P. M.; Li, A.; Badu-Tawiah, A.; Wei, P.; Li, L.; Ouyang, Z.; Roqan, I. S.; Cooks, R. G. Analyst 2013, 138, 3740−3748. (10) El-Faramawy, A.; Siu, K. W. M.; Thomson, B. A. J. Am. Soc. Mass Spectrom. 2005, 16, 1702−1707. (11) Karas, M.; Bahr, U.; Dülcks, T. Fresenius' J. Anal. Chem. 2000, 366, 669−676. (12) Cody, R. B.; Laramée, J. A.; Nilles, J. M.; Durst, H. D. JEOL News 2005, 8−12. (13) Cody, R. B.; Dane, A. J. J. Am. Soc. Mass Spectrom. 2013, 24, 329− 334. (14) Gross, J. H. Anal. Bioanal. Chem. 2014, 406, 63−80. (15) Liu, J.; Wang, H.; Manicke, N. E.; Lin, J.-M.; Cooks, R. G.; Ouyang, Z. Anal. Chem. 2010, 82, 2463−2471. (16) Wang, H.; Liu, J.; Cooks, R. G.; Ouyang, Z. Angew. Chem., Int. Ed. 2010, 49, 877−880. (17) Manicke, N. E.; Abu-Rabie, P.; Spooner, N.; Ouyang, Z.; Cooks, R. G. J. Am. Soc. Mass Spectrom. 2011, 22, 1501−1507. (18) Wang, H.; Manicke, N. E.; Yang, Q.; Zheng, L.; Shi, R.; Cooks, R. G.; Ouyang, Z. Anal. Chem. 2011, 83, 1197−1201. (19) Zhang, Z.; Cooks, R. G.; Ouyang, Z. Analyst 2012, 137, 2556− 2558. (20) Zhang, Z.; Xu, W.; Manicke, N. E.; Cooks, R. G.; Ouyang, Z. Anal. Chem. 2012, 84, 931−938. (21) Manicke, N. E.; Yang, Q.; Wang, H.; Oradu, S.; Ouyang, Z.; Cooks, R. G. Int. J. Mass Spectrom. 2011, 300, 123−129. (22) Naccarato, A.; Moretti, S.; Sindona, G.; Tagarelli, A. Anal. Bioanal. Chem. 2013, 405, 8267−8276. (23) Qian, K.; Rodgers, R. P.; Hendrickson, C. L.; Emmett, M. R.; Marshall, A. G. Energy Fuels 2001, 15, 492−498. (24) Klampfl, C. W.; Himmelsbach, M. Anal. Chim. Acta 2015, 890, 44−59. (25) Li, A.; Wang, H.; Ouyang, Z.; Cooks, R. G. Chem. Commun. 2011, 47, 2811−2813. (26) Espy, R. D.; Muliadi, A. R.; Ouyang, Z.; Cooks, R. G. Int. J. Mass Spectrom. 2012, 325−327, 167−171. (27) Yang, Q.; Wang, H.; Maas, J. D.; Chappell, W. J.; Manicke, N. E.; Cooks, R. G.; Ouyang, Z. Int. J. Mass Spectrom. 2012, 312, 201−207. (28) Boström, C.-E.; Gerde, P.; Hanberg, A.; Jernström, B.; Johansson, C.; Kyrklund, T.; Rannug, A.; Törnqvist, M.; Victorin, K.; Westerholm, R. Environ. Health Perspect. 2002, 110, 451−488. (29) Neff, J. M.; Ostazeski, S.; Gardiner, W.; Stejskal, I. Environ. Toxicol. Chem. 2000, 19, 1809−1821. (30) Reddy, C. M.; Eglinton, T. I.; Hounshell, A.; White, H. K.; Xu, L.; Gaines, R. B.; Frysinger, G. S. Environ. Sci. Technol. 2002, 36, 4754− 4760. (31) Peterson, C. H.; Rice, S. D.; Short, J. W.; Esler, D.; Bodkin, J. L.; Ballachey, B. E.; Irons, D. B. Science 2003, 302, 2082. F

DOI: 10.1021/acs.analchem.7b01733 Anal. Chem. XXXX, XXX, XXX−XXX