Nanospray Mass Spectrometry with Indirect Conductive Graphite

Stable electrospray was achieved in the investigated flow range (150−900 nL min-1), and salbutamol, diphenhydramine, and nortriptyline (Mw: 239−26...
4 downloads 0 Views 396KB Size
Anal. Chem. 2004, 76, 4241-4244

Nanospray Mass Spectrometry with Indirect Conductive Graphite Coating Peter Viberg,*,†,‡ Staffan Nilsson,‡ and Kerstin Skog†

Division of Applied Nutrition and Food Chemistry, Department of Food Technology, Engineering and Nutrition, and Department of Technical Analytical Chemistry, Center for Chemistry and Chemical Engineering, Lund University, P.O. Box 124, SE-22100 Lund, Sweden

An easy and cost-effective method to manufacture a robust conductive graphite coating for nanospray mass spectrometry (nESI-MS) and capillary electrophoresis (CE)nESI-MS is described. The method involves graphite coating of a tube sleeve, into which the nESI emitter is inserted and connected to a transfer capillary, instead of coating the actual emitter. The coating, made of graphite from a pencil and epoxy glue, was stable over long periods of use (>80 h) and showed excellent resistance toward various solvents. Stable electrospray was achieved in the investigated flow range (150-900 nL min-1), and salbutamol, diphenhydramine, and nortriptyline (Mw: 239263 g mol-1) were detected in the nanomole per liter range during continuous pumping. CE-nESI-MS analysis gave excellent signal-to-noise ratios for 100-fmol injections. The technique allows simple exchange of the nESI emitter to suit a specific flow rate, and it minimizes risk of corona discharge. Over the past 10 years, sheathless nanospray mass spectrometry (nESI-MS) has developed into a widely used ion source for MS. Benefits of sheathless nESI include minimized sample dilution, low flow rate operation, and high ionization efficiency.1 Sheathless operation requires that the spray voltage is applied directly to the electrosprayed solution, unlike in sheath flow operation where the electrical contact occurs via the sheath flow. Commonly the outside of the nanospray emitter is coated in conductive materials, such as gold or silver through sputtering, plating or vapor deposition,2-5 gold powder,6 graphite,7,8 silver epoxy,9 or conductive polymer.10,11 However, one of the disadvan* Corresponding author: (e-mail) [email protected] or peter.viberg@ teknlk.lth.se; (phone) +46 46 2228314. † Division of Applied Nutrition and Food Chemistry, Department of Food Technology, Engineering and Nutrition. ‡ Department of Technical Analytical Chemistry, Center for Chemistry and Chemical Engineering. (1) Wilm, M.; Mann, M. Anal. Chem. 1996, 68, 1-8. (2) Kriger, M. S.; Cook, K. D.; Ramsey, R. S. Anal. Chem. 1995, 67, 385-389. (3) Wu, J.-T.; Qian, M. G.; Li, M. K.; Liu, L.; Lubman, D. M. Anal. Chem. 1996, 68, 3388-3396. (4) Bateman, K. P.; White, R. L.; Yaguchi, M.; Thibault, P. J. Chromatogr., A 1998, 794, 327-344. (5) McComb, M. E.; Krutchinsky, A. N.; Ens, W.; Standing, K. G.; Perrealt, H. J. J. Chromatogr., A 1998, 800, 1-11. (6) Barnidge, D. R.; Nilsson, S.; Markides, K. E. Anal. Chem. 1999, 71, 41154118. (7) Chang, Y. Z.; Her, G. R. Anal. Chem. 2000, 72, 626-630. (8) Smith, D. R.; Wood, T. D. Anal. Chem. 2003, 75, 7015-7019. 10.1021/ac049728e CCC: $27.50 Published on Web 05/19/2004

© 2004 American Chemical Society

tages with conductive coatings occurs during malfunctioning of the emitter, e.g., due to clogging or breakage, in which case the coated emitter needs to be replaced, which can be time-consuming and costly. Furthermore, corona discharges are likely to occur, which may destroy the coating and add noise to the mass spectra. The spray voltage has also been applied to the electrosprayed solution via precolumn metal union,12 a T-junction with electrode contact,13 a wire electrode inserted through the capillary tip,14,15 and through a porous junction created by hydrofluoric acid etching on a fused-silica capillary.16 This report demonstrates an effective, robust, cost-effective, and straightforward method to establish electrical contact in ESI, without the disadvantages discussed above. The method involves a conductive graphite coating of a tube sleeve that connects a nESI emitter to a transfer capillary instead of coating of the emitter itself. EXPERIMENTAL SECTION Chemicals and Reagents. Nortriptyline, diphenhydramine, and salbutamol were obtained from Sigma (St. Louis, MO). Methanol (analytical grade) was purchased from J. T. Baker (Phillispburg, NJ), and hydrofluoric acid (48%) from Merck (Darmstadt, Germany). Formic acid was obtained from Riedel-de haen (Seelze, Switzerland), and water was purified by a MilliQ purification system (Millipore, Bedford, MA). Capillaries (380-µm o.d.; 25- and 75-µm i.d.) made of polyimide-coated fused silica were purchased from Polymicro Technologies (Phoenix, AZ). Sample and Solvent Preparation. Electrolyte consisting of methanol and water (1:1 v/v) with the addition of formic acid (0.1% v/v) was prepared for CE-nESI-MS and nESI-MS experiments. A stock solution of salbutamol (Mw ) 239 g mol-1), diphenhydramine (Mw ) 255 g mol-1), and nortriptyline (Mw ) 263 g mol-1) (1 mmol L-1 each) was prepared in methanol. Standard solutions, (9) Wahl, J. H.; Gale, D. C.; Smith, R. D. J. Chromatogr., A 1994, 659, 217222. (10) Wetterhall, M.; Nilsson, S.; Markides, K. E.; Bergquist, J. Anal. Chem. 2002, 74, 239-245. (11) Maziarz, E. P.; Lorenz, S. A.; White, T. P.; Wood, T. D. J. Am. Soc. Mass Spectrom. 2000, 11, 659-663. (12) Haskins, W. E.; Wang, Z.; Watson, C. J.; Rostand, R. R.; Witowski, S. R.; Powell, D. H.; Kennedy, R. T. Anal. Chem. 2001, 73, 5005-5014. (13) Waterval, J. C.; Bestebreurtje, P.; Lingeman, H.; Versluis, C.; Heck, A. J.; Bult, A.; Underberg, W. J. Electrophoresis 2001, 22, 2701-2708. (14) Fong, K. W.; Chan, T. W. J. Am. Soc. Mass Spectrom. 1999, 10, 72-75. (15) Hannis, J. C.; Muddiman, D. C. Rapid Commun. Mass Spectrom. 1998, 12, 443-448. (16) Janini, G. M.; Conrads, T. P.; Wilkens, K. L.; Issaq, H. J.; Veenstra, T. D. Anal. Chem. 2003, 75, 1615-1619.

Analytical Chemistry, Vol. 76, No. 14, July 15, 2004 4241

Figure 1. Illustrations of a capillary connector used for connecting a nESI emitter to a CE/transfer capillary and to supply ESI voltage. (A) Photograph of the nESI emitter inserted into the graphite-coated tube sleeve and secured to the connector, ready for nESI-MS analysis. (B) Schematic cross section view of the capillary connector, CE/transfer capillary, and nESI emitter. High voltage, which is applied to the graphite coating of the tube sleeve via an alligator clip, is transferred along the graphite coating to the liquid junction between the CE/transfer capillary and nESI emitter, where it enters the solution to create electrospray from the nESI emitter outlet (positioned ∼1 mm from the inlet to the mass spectrometer).

for analysis, were prepared by dilution of the stock solution in electrolyte 50, 500, 5000, and 50 000 times (20 000, 2000, 200, and 20 nmol L-1). Further information about the analytes, including molecular structures and CE and CEC optimization, is found elsewhere.17 nESI-MS Interface Design and Procedure for Graphite Coating. A plastic zero dead-volume connector (Upchurch Scientific, Oak Harbor, WA) was used to connect a nESI emitter to a transfer/CE capillary and to provide electrical contact for the nESI voltage. The nESI voltage was applied via an alligator clip to the graphite-coated tube sleeve and transferred through the graphite to the electrosprayed solution via a liquid junction (Figure 1). The graphite was applied to the tube sleeve as follows: A piece of tube sleeve, 380-µm i.d. and 600-µm o.d. (Upchurch Scientific) was first coated in a layer of two-component epoxy glue (Danalim, Denmark), and then wiped against a paper tissue to ensure a thin and even layer of glue. The tube sleeve was then put into a 1.5mL Eppendorf tube containing ∼0.25 mL of graphite powder, obtained by rotating a pencil lead (HB, Staedler Mars micro carbon, Staedler, Germany) against a sanding paper (300 grains/ in.). The Eppendorf tube was shaken to spread graphite powder over the tube sleeve. The graphite-coated tube sleeve was then rolled between two pieces of paper to create an even layer of graphite and to force the graphite powder into the glue. A piece of capillary was pushed through the tube sleeve to clear it from graphite powder, and the tube sleeve was left to cure at room temperature for minimum of 1 h. Finally, the nESI emitter was inserted into the tube sleeve (the emitter extending 1-2 cm from the tube sleeve) and connected to a CE/transfer capillary according to Figure 1. (17) Viberg, P.; Karlsson-Jornten, M.; Petersson, P.; Spe´gel, P.; Nilsson, S. Anal. Chem. 2002, 74, 4595-4601.

4242

Analytical Chemistry, Vol. 76, No. 14, July 15, 2004

nESI-MS during Continuous Pumping. Nanospray emitters were prepared by etching 25-µm-i.d. fused-silica capillaries in hydrofluoric acid (48%, Merck), as described elsewhere.18 (N.B.: The hydrofluoric acid should be used with extreme care and neutralized with CaCl2 solution after use.) The nanospray emitter was cut to a length of 5 cm. The nESI emitter performance was validated by electrospraying standard solutions of salbutamol, diphenhydramine, and nortriptyline at different flow-rates (100-900 nL min-1) and at different sample concentrations (20-20 000 nmol L-1). The standard solutions were pumped hydrodynamically utilizing a pressurized vessel.19 The flow rate was estimated from the time required to pump a plug of sample solution through the electrolyte-filled capillary at a given pressure and the volume of the capillary and nanospray emitter. CE-nESI-MS. In another set of experiments, an in-houseconstructed CE system19 was connected to the nESI-MS setup. The inlet end of the capillary (100 cm in length with 75-µm i.d.) was pressurized by nitrogen gas to make hydrodynamic injections, and to modify the flow during separation. To allow enough time for separation, no pressure was used in the beginning of the separation, but the pressure was switched on prior to elution of the sample molecules. A standard solution of salbutamol, diphenhydramine, and nortriptyline was injected during 3 s at 1 bar. Pressure-aided CE separation was performed at 18.5 kV or 185 V cm-1 (20 kV to the injection end, and 1.5 kV to the graphite coating). Mass Spectrometric Detection. Detection was performed on a ThermoFinnigan LCQ Deca ion trap mass spectrometer (18) Santesson, S.; Ramı´rez, I. B.-R.; Viberg, P.; Jergil, B.; Nilsson, S. Anal. Chem. 2004, 76, 303-308. (19) Viberg, P.; Nilsson, S.; Skog, K. Anal. Bioanal. Chem. 2004, 378, 17291734.

Figure 3. Total ion chromatograms (left) and their corresponding average mass spectra (right). A 20 µmol L-1 solution of salbutamol, diphenhydramine, and nortriptyline (m/z 240.1, 256.0, and 264.1, respectively) was pumped at different flow rates during scanning.

Figure 2. Light microscopy photographs of a graphite-coated tube sleeve. Inset is a magnification.

(ThermoFinnigan, San Jose, CA), operated in positive full-scan mode. Xcalibur software was used for data processing. The mass spectrometer was tuned on nortriptyline, and the heated capillary was maintained at 115 °C. Sheathless nESI was performed at 1.5 kV spray voltage with the nanospray emitter positioned ∼1 mm from the heated capillary inlet on the mass spectrometer using an x,y,z stage. RESULTS AND DISCUSSION nESI-MS during Continuous Pumping. The graphite coating proved to be very stable over long periods of use and provided stable and sensitive nESI. The graphite coating had, prior to the experiments presented in this report, been used in a preliminary study including more than 30 h of electrospraying without any decrease in the MS signal. The graphite coating proved to be very resistant toward different solvents, which was verified by continuous pumping of methanol, water, and electrolyte through the capillaries for several hours at flows of >1 µL min-1, without reduction in the MS signal. In other subsequent experiments (data not shown), the same graphite coating was used for another 50 h of electrospraying (including both continuous pumping and CE separations), still without decrease in the MS signal or any noticeable changes to the graphite coating. Figure 2 shows a light microscopy photograph of a graphite-coated tube sleeve. Figure 3 shows total ion chromatograms and mass spectra (average of 230 scans), from electrospraying of standard solution (20 µmol L-1) at different flow rates (100-900 nL min-1). The total ion current (TIC) at 100 nL min-1 showed instability with variations around 20-25%, but flow rates at or above 150 nL min-1 stabilized the TIC. The peaks immediately to the right of each

Figure 4. Sum of signal intensities of salbutamol, diphenhydramine, and nortriptyline (2 µmol L-1 each), versus different flow rates.

sample peak are isotopic peaks from the sample molecules and not contaminants. Figure 4 shows the sum of the signal intensities from salbutamol, nortriptyline, and diphenhydramine (2 µmol L-1, respectively, injected as a mixture) versus the flow rate. It is clear from Figure 4 that the detection was mass sensitive up to ∼250 nL min-1 and concentration sensitive above that flow. Consequently, a flow rate above 250 nL min-1 was chosen for the CE-nESI-MS experiment. Figure 5 shows mass spectra (average of 150 scans) at sample concentrations ranging from 20 to 20 000 nmol L-1. The sample peaks are easily detected down to 200 nmol L-1, but at 20 nmol L-1, the noise intensity is approaching that of the sample molecules. Figure 6 displays a wide mass-range spectrum, without any large peaks corresponding to contaminants, obtained from electrospraying the standard solution (20 µmol L-1) at 500 nL min-1 (average of 100 scans, between m/z 50 and 2000). Since the graphite coating is applied to the outside of the tube sleeve and Analytical Chemistry, Vol. 76, No. 14, July 15, 2004

4243

Figure 5. Mass spectra from salbutamol, diphenhydramine, and nortriptyline [m/z 240.1 (A), 256.0 (B), and 264.1 (C)] at different concentrations at 500 nL min-1 flow rate. Figure 7. Electropherogram from a 100-fmol injection of (a) nortriptyline, (b) diphenhydramine, and (c) salbutamol (5 nL, 20 µmol L-1).

analytes from a 100-fmol injection (20 µmol L-1). The signal-tonoise ratios calculated by the software were 1081, 986, and 530 for diphenhydramine, nortriptyline, and salbutamol, respectively, which indicate that the limit of detection was considerably better than the injected amount in this experiment.

Figure 6. Avarage of 100 scans from a 20 µmol L-1 solution of salbutamol, diphenhydramine, and nortriptyline (m/z 240.1, 256.0, and 264.1, respectively) pumped at 500 nL min-1.

thus not exposed to the flow of electrolyte, the risk of contamination from the epoxy glue or from the graphite is minimized. Furthermore, the risk of corona discharge is reduced with this graphite coating, since it is not applied to the outlet end of the nESI emitter. Although the emitter was positioned only ∼1 mm from the heated capillary of the mass spectrometer, the spray voltage could be set up to 6 kV without discharging. CE-nESI-MS Experiments. Pressure-aided CE was used in order to produce a high enough flow to produce a stable electrospray. This is needed since the low pH of the electrolyte produces a too low electroosmotic flow (EOF). Salbutamol, nortriptyline, and diphenhydramine were eluted at 400 nL min-1. Figure 7 shows the reconstructed electropherograms of the three 4244 Analytical Chemistry, Vol. 76, No. 14, July 15, 2004

CONCLUSIONS The conductive-coating method presented in this report is quick, robust, and easy to use. The coating only involves graphite (from a pencil), epoxy glue, and a low dead-volume capillary connector. The coating proved to be stable over long periods of use and showed resistance toward organic and aqueous solvents. This method has been developed and used for the following major reasons: (i) The nESI emitter can freely be exchanged for one with different dimensions to suit a specific flow rate. (ii) In case of clogging, the emitter can easily be replaced. (iii) If the emitter needs replacement, the old coated tube sleeve can still be used with the new emitter. (iv) Uncoated emitters are used (less expensive). (v) The lack of coating on the nESI emitter minimizes risk of corona discharging. ACKNOWLEDGMENT This work has been carried out with financial support from the Swedish Council for Forestry and Agricultural Research and from The Swedish Cancer Society. Received for review February 17, 2004. Accepted April 16, 2004. AC049728E