Argon Direct Analysis in Real Time Mass Spectrometry in Conjunction

Dec 20, 2012 - ... FL) was used to infuse each organic solvent at a flow rate of 30 μL/min through a fused silica capillary (Polymicro Technologies, ...
2 downloads 0 Views 1MB Size
Download PDF
Technical Note pubs.acs.org/ac

Argon Direct Analysis in Real Time Mass Spectrometry in Conjunction with Makeup Solvents: A Method for Analysis of Labile Compounds Hongmei Yang,† Debin Wan,†,§ Fengrui Song,† Zhiqiang Liu,† and Shuying Liu*,†,‡ †

Changchun Institute of Applied Chemistry, Chinese Academy of Sciences, 5625 Renmin Street, Changchun 130022, China Changchun University of Chinese Medicine, Changchun 130117, China § Graduate School of the Chinese Academy of Sciences, Beijing 100039, China ‡

S Supporting Information *

ABSTRACT: Helium direct analysis in real time (He-DART) mass spectrometry (MS) analysis of labile compounds usually tends to be challenging because of the occurrence of prominent fragmentation, which obscures the assigning of an ion to an independent species or merely a fragment in a mixture. In the present work, argon DART (ArDART) MS in conjunction with makeup solvents has been demonstrated to analyze a variety of labile compounds including nucleosides, alkaloids, glucose, and other small molecules. The results presented here confirm that Ar-DART can generate significantly less energetic ions than conventional He-DART and is able to produce the intact molecular ions with little or no fragmentation in both positive and negative ion modes. Adding a makeup solvent (absolute ethyl alcohol, methanol, fluorobenzene, or acetone) to the argon gas stream at the exit of the DART ion source can result in 1−2 orders of magnitude increase in detection signals. The sensitivity attainable by Ar-DART was found to be comparable to that by He-DART. The investigation of influence of solvents improves our understanding of the fundamental desorption and ionization processes in DART. The practical application of this rapid and high throughput method is demonstrated by the successful analysis of a natural product (Crude Kusnezoff Monkshood) extract, demonstrating the great potential in mixture research. he direct analysis in real time (DART) ion source was first developed by Cody et al.1 in 2003 and introduced as a commercial product2,3 in early 2005. It is a versatile ion source that operates in open air, allowing for the rapid, noncontact analysis of solid, liquid, and gaseous materials without any sample pretreatment. In DART, a helium (most commonly) or nitrogen plasma is created inside a ceramic flow chamber by an atmospheric pressure glow discharge initiated by applying a potential of several kilovolts. The excited beam is heated by passing through a heater chamber and flows through a grid electrode. When the DART gas comes into contact with a sample placed between the ion source outlet and the mass spectrometer interface inlet, the sample will be ionized. The DART ionization mechanisms are not yet fully understood, but the widely acceptable mechanism is Penning ionization. In the positive ion mode, the excited-state helium (ionization energy 19.8 eV) induces Penning ionization of atmospheric water molecules by collisions, generating protonated water clusters,2 followed by proton transfer to the analyte.4 Another reported ionization mechanism in the positive ion mode is direct Penning ionization of desorbed analyte whose ionization energy (IE) is below 19.8 eV,5 generating a molecular ion. In the negative ion mode, a negative ion of the analyte is formed by reaction with negative-ion clusters containing oxygen and water.2,6

T

© 2012 American Chemical Society

With the appealing features of high throughput, lack of memory effects, its relatively low tendency toward ion suppression,7,8 and simplicity, the DART ion source has been applied to a wide range of applications such as reaction monitorings,9−11 forensic science,2,12 the detection of counterfeit drugs,13−15 fruit and food analysis,16,17 metabolomics,18,19 and analysis of highly insoluble polycyclic aromatic compounds.20 Furthermore, the use of DART mass spectrometry (MS) for quantitative analysis of samples has been successfully demonstrated, such as quantitation of chemical warfare agents21 and control of strobilurin fungicides in wheat.22 Although this technique has been used with great success, one of its major drawbacks is the relatively low sensitivity. Strategies such as the use of the Vapur interface (an interface involving a novel gas ion separator)8 and adding a makeup liquid (iso-propanol) postcolumn to the HPLC effluent23 in helium DART (HeDART) MS have been employed for improving sensitivity. The comparison of the chromatograms obtained with and without iso-propanol solvent revealed the substantial enhancement in peak intensity achieved through the addition of iso-propanol as the makeup liquid.23 Solvent effects on sensitivity of He-DART Received: September 13, 2012 Accepted: December 20, 2012 Published: December 20, 2012 1305

dx.doi.org/10.1021/ac3026543 | Anal. Chem. 2013, 85, 1305−1309

Analytical Chemistry

Technical Note

MS measurements was also reported by other groups,11,17,24 indicating that high-volatility solvents were good for the sensitivity of DART. Another drawback, the fragmentation observed for some compounds in the DART source,4,25,26 will sometimes lead to a problem of how to determine whether a particular ion is an independent species or merely a fragment in a mixture. To our knowledge, until now, none of the published applications of DART MS has been concerned with the method for analysis of labile compounds. In this work, we demonstrate the application of argon DART (Ar-DART) MS in conjunction with makeup solvents for the rapid analysis of labile compounds including nucleoside, alkaloids, glucose, and other small molecules. These mass spectra were compared with those produced by HeDART MS. Ar-DART resulted in diminished fragmentation for all analytes studied versus the extensive fragmentation produced by He-DART. Organic solvents played important roles in the increase of signal intensities in Ar-DART MS. ArDART interfaced with ion trap MS in conjunction with makeup solvents has been shown to be a powerful tool for MS analyses of labile compounds with great sensitivity and speed.

tapered ceramic exit cap on the DART-SVPA source. In order to increase signal intensity, a model SP100i syringe pump (WPI, Sarasota, FL) was used to infuse each organic solvent at a flow rate of 30 μL/min through a fused silica capillary (Polymicro Technologies, Phoenix, AZ) with an inner diameter of 100 μm into the DART gas stream at the exit of the DART ion source. In our experiments, the distance between the makeup solvent line and the DART exit was about 8 mm, and the same distance between the makeup solvent line and the glass capillary tube with the sample was also employed.

■

RESULTS AND DISCUSSION Comparison of Mass Spectra Obtained by Ar-DART and He-DART. Extensive fragmentation25−27 has been found in He-DART induced by high-energy helium metastables, especially at high gas temperature and flow rates. Then, a shift to another DART gas (argon) was made to see if it could lead to a softer ionization process. As expected, Ar-DART produced intact molecular ions with little or no fragmentation. In negative ion mode, glucose was selected as an example (Figure 1). The peak at m/z 179.1 in Figure 1a represents the [M −

■

EXPERIMENTAL SECTION Chemicals and Reagents. 2′-Deoxycytidine and fluorobenzene were bought from Aladdin Chemistry Co., Ltd. (Shanghai, China). Aconitine (AC), hypaconitine (HA), and mesaconitine (MA) were purchased from National Institute for the Control of Pharmaceutical and Biological Products (Beijing, China). Glucose, diethyl ether, and dichloromethane were obtained from Beihua Fine Chemicals Co., Ltd. (Beijing, China). Acetone was acquired from Sinopharm Chemical Reagent Co., Ltd. (Beijing, China). Aspartic acid, α-cyano-4hydroxycinnamic acid (CHCA), 2,5-dihydroxybenzoic acid (DHB), 2,4,6-trihydroxy acetophenone (THAP), and ferulic acid (FA) were purchased from Sigma (St. Louis, MO). Acetonitrile, absolute ethanol, and methanol (HPLC grade) were obtained from Fisher Chemical Company. High-purity helium, nitrogen, and argon (99.999%) were supplied by Changchun Juyang Gas Co., Ltd. (Changchun, China). Ultrapure water (specific conductivity, 18.2 MΩ/cm) was produced by a Milli-Q device (Millipore, Milford, MA). DART MS Analysis. Analysis was conducted with a Finnigan LTQ ion trap mass spectrometer (San Jose, CA) equipped with a fourth generation DART source with standard voltage, pressure, and angling capability (SVPA) (IonSense, Inc., Saugus, MA). The DART ion source was operated with helium or argon for analysis and nitrogen in the standby mode. The instrument parameter settings were optimized to obtain sensitive and accurate determination of the analytes. Gas temperature was optimized to 300 °C, and gas flow rates were set to 2 L/min. Grid electrode voltages were set to 350 V (positive ion mode) and −250 V (negative ion mode), respectively. The linear ion trap mass spectrometer settings included capillary voltage 30 V (positive ion mode) or −35 V (negative ion mode); capillary temperature, 200−250 °C depending on the analyte. Automated acquisition of mass spectra was executed by data-dependent scanning using Xcalibur software (Thermo Finnigan). The analytes were introduced into the DART sample gap using the closed end of a melting point capillary tube that was directly dipped into the sample vial. For each sample, the capillary tube was held close to the DART cap for about 30 s, which was positioned 1 mm below and 1 mm in front of the

Figure 1. (a) He-DART and (b) Ar-DART mass spectra of 0.1 mg/ mL glucose in negative ion mode. The highest peak in each mass spectrum is normalized to 100, and the normalization level (NL) is indicated in the spectra.

H]− ion while the base peak at m/z 161.1 is produced by the loss of one water from the quasimolecular ion. By contrast, only the quasimolecular ion is obtained with little fragmentation in the corresponding Ar-DART mass spectrum (Figure 1b). A representative example of labile compounds in positive ion mode mass spectra is displayed in Figure 2. It is apparent in the spectrum of Figure 2a that there was a high degree of fragmentation of 2′-deoxycytidine. An abundant fragmentation 1306

dx.doi.org/10.1021/ac3026543 | Anal. Chem. 2013, 85, 1305−1309

Analytical Chemistry

Technical Note

inadequate for many challenging problems. Water whose IE is 12.6 eV did not undergo Penning ionization in Ar-DART because the metastable argon has an IE of 11.55 eV for the 3P2 state and 11.72 eV for the 3P0 state.28 In an effort to improve sensitivity, organic solvents with relatively low IE values substituted for water were examined in our work. Scheme 1 Scheme 1. Probable Reaction Occurring in Ar-DART

shows a series of reactions that describe what can happen when an organic solvent is used. They are as follows: (1) a solvent molecule (S) undergoes Penning ionization (reaction A in Scheme 1) followed by proton transfer to produce protonated Sn (n = 1 or 2, Sn means neutral solvent molecule, reaction B in Scheme 1). The analyte molecule (M) is ionized to form the protonated molecule through gas-phase ion/molecule reaction with the protonated solvent molecule (reaction C in Scheme 1). The solvent molecular ion can also react with the analyte molecule to produce the protonated analyte molecule (reaction D in Scheme 1). (2) The metastable argon which has sufficient internal energy will produce electrons. Then the kinetic energies of these electrons will thermalize within a few nanoseconds2 (reaction A in Scheme 1). It is noteworthy that the formed hot electrons can undergo resonance electron capture with analytes to produce M−• not [M − H]− in the negative ion mode (reaction E in Scheme 1). The IEs, PAs, and boiling points (Bps) of the six selected compounds are summarized in Table S-1 in the Supporting Information. With all compounds studied, 1−2 orders of magnitude increase in signal intensity was found as a result of adding a solvent to the argon gas stream at the exit of the DART ion source. The representative results are shown in Figure 3, and the observed differences among them are significant. The IE of acetonitrile is 12.2 eV, thus acetonitrile will not undergo Penning ionization in Ar-DART, resulting in almost no influence on the signal intensities as displayed in Figure 3. As for dichloromethane which has an IE of 11.33 eV, it cannot be ionized efficiently by metastable argon. Therefore,

Figure 2. (a) He-DART and (b) Ar-DART mass spectra of 50 μg/mL 2′-deoxycytidine in positive ion mode. The ions at m/z 112 and 117 shown in the inset panel are denoted as [C + H]+ and [I + H]+. The fragmentation ion at m/z 81.2 corresponds to [C5H5O]+ (II). The NL value of each spectrum is labeled on the upper right corner.

ion at m/z 112.1 corresponds to the nucleobase (cytidine) as indicated in the inset panel, interestingly, which can form adducts with the small fragment at m/z 81.2 ([C5H5O]+) in different molar ratios in the DART ion source. In addition, the prominent adduct of the nucleobase and molecular ion at m/z 339.2 was also detected. However, the number and intensity of the fragment peaks are significantly reduced, and only one minor fragmentation ion at m/z 112.1 was observed in ArDART mass spectrum of 2′-deoxycytidine (Figure 2b). In this spectrum, the dominant peaks appear at m/z 228.2 and 455.2, corresponding to the quasimolecular ion and its dimer which is a noncovalent complex, respectively, allowing us to quickly assign the parent ion. Besides, in the case of the other examined compounds including AC, HA, MA, aspartic acid, CHCA, THAP, and FA, fragmentation is not observed due to the loss of acetic acid or water in their Ar-DART mass spectra, which is abundant in the corresponding He-DART mass spectra (data not shown). The above results indicate the potential utility of Ar-DART in the analysis of mixtures and some metabolites, especially phase II conjugates. Effect of Makeup Solvents. Although this approach has been used with great success for analysis of labile compounds, it has the disadvantage of low sensitivity, which is often

Figure 3. Influence of the six organic solvents on signal intensities obtained for 2.5 mM THAP in negative ion mode and 2 mM HA in positive ion mode. Data points are mean values of three determinations. 1307

dx.doi.org/10.1021/ac3026543 | Anal. Chem. 2013, 85, 1305−1309

Analytical Chemistry

Technical Note

which is in perfect agreement with the experimental results (Figure 3). For fluorobenzene, HA was protonated through reaction D in Scheme 1 because of the PA value of fluorobenzene smaller than that of its (S − H) radical (Table S-1 in the Supporting Information). Interestingly, when analytes at concentrations of 1−2 orders of magnitude lower than those at the detection limits were analyzed by Ar-DART after injection of an effective solvent to argon gas stream, distinct quasimolecular ions appeared (data not shown). Furthermore, it was found that the higher boiling point of the solvent used was, the longer time of increased signal sustained under the same conditions. As a matter of fact, influence of organic solvents on signal intensities in He-DART was also investigated. However, there was almost no change in signal intensities probably because water in the atmosphere is enough for Penning ionization by helium in the absence of organic solvents. In short, Ar-DART can directly ionize organic solvent molecules using a stream of hot gas, but the analytes will be ionized primarily through gas-phase ion/molecule reactions with the organic solvent ions, just like atmospheric pressure chemical ionization. Our data supports the transient microenvironment mechanism proposed for addressing matrix effects for He-DART.29 The results proved that a solvent with low IE and low PA is good for the sensitivity of Ar-DART in the positive ion mode. Methanol and fluorobenzene should be the best solvents for the ionization of all the analytes in positive and negative ion modes because they have the weakest PA and lowest IE among the solvents examined, respectively. Application of Ar-DART MS in Conjunction with Makeup Solvents to the Analysis of Mixtures. The lack of fragment ions in Ar-DART should make it ideal for rapid profiling of mixtures. Methanol, fluorobenzene, absolute ethanol, and acetone are all beneficial to the sensitivity of ArDART, but absolute ethanol was selected as a makeup solvent from the point of view of its safety in human beings and environment protection. Figure 4 shows the comparison of ArDART and He-DART positive ion mode MS of alkaloid mixture extracted from Crude Kusnezoff Monkshood by diethyl ether. Figure 4a is the mass spectrum produced by Ar-DART in conjunction with absolute ethanol as a makeup solvent, which shows three major alkaloid components present in the mixture corresponding to m/z 616.4 (HA), 632.4 (MA), and 646.3 (AC). Figure 4b is the mass spectrum of the same sample produced through He-DART with our best attempt at minimizing the fragmentation. However, the fragment peaks at m/z 556.4, 572.3, and 586.4 are distinct in the He-DART mass spectrum. These three peaks are fragments corresponding to the loss of acetic acid from each of the three small quasimolecular ions at m/z 616.4, 632.4, and 646.3. The signal intensity attainable by combination of Ar-DART and absolute ethanol as the makeup solvent (NL, 5.53 × 106) was found to be comparable to that by conventional He-DART (NL, 6.44 × 106). In addition, the mixture of DHB, THAP, and CHCA was analyzed by Ar-DART and He-DART MS in negative ion mode (Figure S-3 in the Supporting Information). By contrast, the fragments observed in He-DART mass spectrum (Figure S-3b in the Supporting Information) were much more abundant than those in Ar-DART mass spectrum (Figure S-3a in the Supporting Information). The two examples illustrate the utility of Ar-DART in the analysis of mixtures, especially containing labile compounds, using both negative and positive ion modes.

in comparison, not too much signal increase was observed when dichloromethane was used as a makeup solvent. In comparison with the corresponding detection results with no makeup solvent, the signal intensities of THAP in negative ion mode were increased by 100 to over 150 times while those of HA in positive ion mode were 30−70 times when the other four organic solvents were infused (Figure 3). For the sake of clarity, absolute ethanol is taken as an example and the mass spectra of Ar-DART and those of Ar-DART in conjunction with a makeup solvent using the absolute intensity as the y-axis were shown in Figures S-1 and S-2 in the Supporting Information, respectively. By comparison, it is quite evident that makeup solvents play an important part in increasing signal intensity. Methanol, absolute ethanol, acetone, and fluorobenzene can be ionized through reaction A in Scheme 1 to generate S+• ions. The former three ones should further undergo reaction B in Scheme 1 to become [S + H]+ ions because both they and their dimers have PA values larger than their (S − H) radicals. However, in the case of fluorobenzene, the S+• ion cannot undergo reaction B in Scheme 1 to become [S + H]+ ion because the PA value of fluorobenzene is smaller than its (S − H) radical (see Table S-1 in the Supporting Information). Although the proton affinity of HA (its structure shown in Figure 4) is not reported in the literature, it will have a high PA value due to the existence of an N atom in its chemical structure. Thus, when HA was analyzed using methanol, absolute ethanol, and acetone as makeup solvents, it should be ionized through reaction C in Scheme 1, and the signal intensities should be dependent on PA(M)-PA(S or S2) values,

Figure 4. (a) Ar-DART and (b) He-DART mass spectra of Crude Kusnezoff Monkshood extract in positive ion mode. The NL value of each spectrum is labeled on the upper right corner. 1308

dx.doi.org/10.1021/ac3026543 | Anal. Chem. 2013, 85, 1305−1309

Analytical Chemistry

■

Technical Note

(11) Petucci, C.; Diffendal, J.; Kaufman, D.; Mekonnen, B.; Terefenko, G.; Musselman, B. Anal. Chem. 2007, 79, 5064−5070. (12) Steiner, R. R.; Larson, R. L. J. Forensic Sci. 2009, 54, 617−622. (13) Zhou, Z.; Zhang, J.; Zhang, W.; Bai, Y.; Liu, H. Analyst 2011, 136, 2613−2618. (14) Samms, W. C.; Jiang, Y. J.; Dixon, M. D.; Houck, S. S.; Mozayani, A. J. Forensic Sci. 2011, 56, 993−998. (15) Nyadong, L.; Harris, G. A.; Balayssac, S.; Galhena, A. S.; MaletMartino, M.; Martino, R.; Parry, R. M.; Wang, M. D.; Fernandez, F. M.; Gilard, V. Anal. Chem. 2009, 81, 4803−4812. (16) Cajka, T.; Riddellova, K.; Zomer, P.; Mol, H.; Hajslova, J. Food Addit. Contam. 2011, 28, 1372−1382. (17) Vaclavik, L.; Rosmus, J.; Popping, B.; Hajslova, J. J. Chromatogr., A 2010, 1217, 4204−4211. (18) Gu, H.; Pan, Z.; Xi, B.; Asiago, V.; Musselman, B.; Raftery, D. Anal. Chim. Acta 2011, 686, 57−63. (19) Zhou, M.; McDonald, J. F.; Fernández, F. M. J. Am. Soc. Mass Spectrom. 2010, 21, 68−75. (20) Domin, M. A.; Steinberg, B. D.; Quimby, J. M.; Smith, N. J.; Greene, A. K.; Scott, L. T. Analyst 2010, 135, 700−704. (21) Nilles, J. M.; Connell, T. R.; Durst, H. D. Anal. Chem. 2009, 81, 6744−6749. (22) Schurek, J.; Vaclavik, L.; Hooijerink, H. D.; Lacina, O.; Poustka, J.; Sharman, M.; Caldow, M.; Nielen, M. W.; Hajslova, J. Anal. Chem. 2008, 80, 9567−9575. (23) Beissmann, S.; Buchberger, W.; Hertsens, R.; Klampfl, C. W. J. Chromatogr., A 2011, 1218, 5180−5186. (24) Zhao, Y.; Lam, M.; Wu, D.; Mak, R. Rapid Commun. Mass Spectrom. 2008, 22, 3217−3224. (25) Harris, G. A.; Hostetler, D. M.; Hampton, C. Y.; Fernández, F. M. J. Am. Soc. Mass Spectrom. 2010, 21, 855−863. (26) Curtis, M.; Minier, M. A.; Chitranshi, P.; Sparkman, O. D.; Jones, P. R.; Xue, L. J. Am. Soc. Mass Spectrom. 2010, 21, 1371−1381. (27) Vaclavik, L.; Cajka, T.; Hrbek, V.; Hajslova, J. Anal. Chim. Acta 2009, 645, 56−63. (28) Dane, A. J.; Cody, R. B. Analyst 2010, 135, 696−699. (29) Song, L.; Gibson, S. C.; Bhandari, D.; Kelsey, D.; Cook, K. D.; Bartmess, J. E. Anal. Chem. 2009, 81, 10080−10088.

CONCLUSION The compounds chosen for this study are labile, which have posed significant challenges to conventional He-DART mass spectrometry due to their extensive fragmentation. The comparison between Ar-DART and conventional He-DART reveals that Ar-DART results in minimized fragmentation for both positive and negative mode mass analysis of labile compounds. The difficulties in differentiating the molecular ions from the fragment ions often encountered in a He-DART MS are eliminated. The coupling of Ar-DART with makeup solvents provides a powerful tool to obtain high sensitivity and accuracy analysis on labile analytes. Methanol, absolute ethanol, acetone, and fluorobenzene are favorable solvents for Ar-DART ionization. Ar-DART can directly ionize organic solvent molecules using a stream of hot argon, but the analytes will be ionized primarily through proton and electron transfer in positive and negative ion modes, respectively, with the organic solvent ions. The work presented here improves our understanding of the fundamental desorption and ionization processes in DART, especially when solvent is involved in the ionization. This would open the door to rapid and reliable detection of labile compounds using a DART mass spectrometer.

■

ASSOCIATED CONTENT

S Supporting Information *

Additional information as noted in text. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected], [email protected]. Fax: +86 431 8526 2886. Phone: +86 431 8526 2886. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by the National Natural Science Foundation of China (Grant No. 21175127) and the China Postdoctoral Science Foundation funded project (Grant No. 2012M511355).

■

REFERENCES

(1) Cody, R. B.; Laramée, J. A. Atmospheric pressure ion source. U.S. Patent 6,949,741, September 27, 2005. (2) Cody, R. B.; Laramée, J. A.; Durst, H. D. Anal. Chem. 2005, 77, 2297−2302. (3) Cody, R. B.; Laramée, J. A.; Nilles, J. M.; Durst, H. D. JEOL News 2005, 40, 8−12. (4) Cody, R. B. Anal. Chem. 2009, 81, 1101−1107. (5) Kpegba, K.; Spadaro, T.; Cody, R. B.; Nesnas, N.; Olson, J. A. Anal. Chem. 2007, 79, 5479−5483. (6) Zhang, J.; Huo, F.; Zhou, Z.; Bai, Y.; Liu, H. Prog. Chem. 2012, 24, 101−109. (7) Eberherr, W.; Buchberger, W.; Hertsens, R.; Klampfl, C. W. Anal. Chem. 2010, 82, 5792−5796. (8) Yu, S.; Crawford, E.; Tice, J.; Musselman, B.; Wu, J. T. Anal. Chem. 2009, 81, 193−202. (9) Sanchez, L. M.; Curtis, M. E.; Bracamonte, B. E.; Kurita, K. L.; Navarro, G.; Sparkman, O. D.; Linington, R. G. Org. Lett. 2011, 13, 3770−3773. (10) Cho, D. S.; Gibson, S. C.; Bhandari, D.; McNally, M. E.; Hoffman, R. M.; Cook, K. D.; Song, L. Rapid Commun. Mass Spectrom. 2011, 25, 3575−3580. 1309

dx.doi.org/10.1021/ac3026543 | Anal. Chem. 2013, 85, 1305−1309