Anal. Chem. 2008, 80, 1208-1214
Unexpected Analyte Oxidation during Desorption Electrospray Ionization-Mass Spectrometry Sofie P. Pasilis, Vilmos Kertesz, and Gary J. Van Berkel*
Organic and Biological Mass Spectrometry Group, Chemical Sciences Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831-6131
During the analysis of surface-spotted analytes using desorption electrospray ionization-mass spectrometry (DESI-MS), abundant ions are sometimes observed that appear to be the result of oxygen addition reactions. In this investigation, the effect of sample aging, the ambient lab environment, spray voltage, analyte surface concentration, and surface type on this oxidative modification of spotted analytes, exemplified by tamoxifen and reserpine, during analysis by DESI-MS was studied. Simple exposure of the samples to air and to ambient lighting increased the extent of oxidation. Increased spray voltage also led to increased analyte oxidation, possibly as a result of oxidative species formed electrochemically at the emitter electrode or in the gas phase by discharge processes. These oxidative species are carried by the spray and impinge on and react with the sampled analyte during desorption/ionization. The relative abundance of oxidized species was more significant for the analysis of deposited analyte having a relatively low surface concentration. Increasing the spray solvent flow rate and the addition of hydroquinone as a redox buffer to the spray solvent were found to decrease, but not entirely eliminate, analyte oxidation during analysis. The major parameters that both minimize and maximize analyte oxidation were identified, and DESI-MS operational recommendations to avoid these unwanted reactions are suggested. Desorption electrospray ionization-mass spectrometry (DESIMS) is increasingly being used as a tool for general surface sampling, high throughput analysis, and chemical imaging because of its capability for in situ analyte detection from unmodified surfaces under ambient sampling conditions.1-3 In DESI, a nebulizing gas/solvent mixture is used to create a high-velocity spray consisting of charged solvent microdroplets in a gas stream.4 This flow of solvent microdroplets impacts the surface, creating a surface solvent flow that mixes with and dissolves at least a portion of the analyte with which it comes in contact. The continual * Corresponding author. Phone: 865-574-1922. Fax: 865-576-8559. E-mail:
[email protected]. (1) Taka´ts, Z.; Wiseman, J. M.; Gologan, B.; Cooks, R. G. Science 2004, 306, 471-473. (2) Taka´ts, Z.; Wiseman, J. M.; Cooks, R. G. J. Mass Spectrom. 2005, 40, 12611275. (3) Cooks, R. G.; Ouyang, Z.; Taka´ts, Z.; Wiseman, J. M. Science 2006, 311, 1566-1570. (4) Venter, A.; Sojka, P. E.; Cooks, R. G. Anal. Chem. 2006, 78, 8549-8555.
1208 Analytical Chemistry, Vol. 80, No. 4, February 15, 2008
impact of the spray on the surface results in the formation of analyte-containing secondary droplets; these secondary droplets, as well as gas-phase ions liberated from the droplets, are then carried into the atmospheric sampling capillary of the mass spectrometer. Localized analyte dissolution, mixing, and creation of secondary droplets through splashing are used to advantage in the technique known as “reactive” DESI-MS. Here, chemical reagents are deliberately added to the spray solvent to facilitate the detection of the target analyte by increasing sensitivity and specificity or to study chemical reactions. Reactive DESI can be a valuable method for sample analysis, as demonstrated by its use for the detection of explosives,5,6 for the oxidation of copper(II) dibutyl dithiocarbamate to the charged Cu(III) complex,7 for the modification and detection of cis-diols8 and cyclic acetals,9 and as a screening assay for artesunate antimalarials.10,11 The formation of noncovalent enzyme-substrate complexes and metal complexes has also been observed.3 The studies noted above illustrate positive examples of the use of reaction chemistry to advantage in DESI. However, reactive DESI may also inherently occur as a result of adventitious reactive species contained in the impacting DESI gas/solvent plume. These reactive species might include oxygen present in the spray solvent or reactive oxygen species (ROS) formed during the electrostatic spray process. The presence of oxygen or ROS in the impacting spray has been postulated as being the cause of one-electron oxidation of a copper carbamate complex during DESI analysis.7 In the course of experiments carried out in our laboratory, we have experienced this unexpected form of reactive DESI, which has led to undesired oxidation, mainly in the form of oxygen addition, for a variety of compounds under investigation (verapamil, reserpine, tamoxifen, 4-hydroxy tamoxifen, clomiphene, toremiphene). This adventitious or inadvertent type of reactive (5) Cotte-Rodrı´guez, I.; Taka´ts, Z.; Talaty, N.; Chen, H.; Cooks, R. G. Anal. Chem. 2005, 77, 6755-6764. (6) Cotte-Rodrı´guez, I.; Chen, H.; Cooks, R. G. Chem. Commun. 2006, 953955. (7) Nefliu, M.; Cooks, R. G.; Moore, C. J. Am. Soc. Mass Spectrom. 2006, 17, 1091-1095. (8) Chen, H.; Cotte-Rodrı´guez, I.; Cooks, R. G. Chem. Commun. 2006, 597599. (9) Sparrapan, R.; Eberlin, L. S.; Haddad, R. S.; Cooks, R. G; Eberlin, M. N.; Augusti, R. J. Mass Spectrom. 2006, 41, 1242-1246. (10) Ricci, C.; Nyadong, L.; Ferna´ndez, F. M.; Newton, P. N.; Kazarian, S. G. Anal. Bioanal. Chem. 2007, 387, 551-559. (11) Nyadong, L.; Green, M. D.; De Jesus, V. R.; Newton, P. N.; Ferna´ndez, F. M. Anal. Chem. 2007, 79, 2150-2157. 10.1021/ac701791w CCC: $40.75
© 2008 American Chemical Society Published on Web 01/10/2008
EXPERIMENTAL SECTION Materials and Reagents. LC/MS grade methanol was obtained from Mallinckrodt Baker, Inc. (Phillipsburg, NJ). Isopropyl alcohol was obtained from EMD Chemicals (Gibbstown, NJ). Water was purified with a Milli-Q system (Millipore Corp., Billerica, MA). Ammonium acetate was purchased from Aldrich Chemical Co. (St. Louis, MO). Hydroquinone, tamoxifen, and reserpine were purchased from Sigma-Aldrich (St. Louis, MO). All compounds were used as received. A standard stock solution of tamoxifen (1 mM) was prepared by dissolving the compound in isopropyl alcohol. The reserpine stock solutions used ranged from 0.77 to 0.79 mM in concentration and were prepared in 77% isopropyl alcohol/23% water. Working standards were prepared by dilution of the stock solutions with isopropyl alcohol as needed. PTFE-printed slides (Electron Microscopy Sciences, Fort Washington, PA), glass slides (Gold Seal Products, Portsmouth, NH), frosted glass slides (Fisher Scientific, Pittsburgh, PA), Whatman No. 42 filter paper (Whatman Inc., Florham Park, NJ), and aluminum-backed TLC sheets (5 × 20 cm) precoated with silica gel 60 F254 (EM Reagents, Darmstadt, Germany) were used as surfaces for these experiments. EM Quant peroxide test strips (EMD Chemicals, Gibbstown, NJ) were used to test the DESI spray for the presence of peroxides.
Sample Preparation. Surfaces were spotted using either 0.25 µL Drummond micropipets or a 2 µL Eppendorf pipet set to deposit 0.25 µL. Sample spots delivered with the Drummond micropipets were ∼2 mm in diameter on the PTFE-printed slides. Sample spots delivered with the Eppendorf pipet were ∼1.5 mm in diameter on the PTFE-printed slides, ∼7 mm in diameter on glass, and ∼3 mm in diameter on Whatman No. 42 filter paper. DESI-MS. The DESI-MS setup used for these experiments has been described in detail elsewhere.13 In brief, the mass spectrometer was a ThermoFinnigan LCQ Deca ion trap (ThermoFinnigan, San Jose, CA) operated using Xcalibur, version 1.3, software. A 7.1 cm long tube extension (508 µm i.d., 803 µm o.d.) to reach out over the sample surfaces was press fit into the standard heated atmospheric pressure sampling capillary. The DESI emitter, as described previously,13 was a 5.2 cm long, taper tip, fused silica capillary (50 µm i.d., 360 µm o.d., New Objective, Woburn, MA). The inner diameter of the nebulizing gas tube was 500 µm, giving a nebulizing gas (nitrogen) jet annulus area of ∼9.5 × 10-8 m2. The ES emitter was mounted 2-3 mm from the surface to be analyzed and was arranged at a ∼55° angle to the surface. The DESI spray plume impacted the surface to be analyzed about ∼2 mm back from the sampling end of the atmospheric pressure sampling capillary extension. Nebulizer gas (nitrogen) flow rates were measured by placing a digital flow meter (GFM 37, 0-50 L/min range, Aalborg Instruments, Orangeburg, NY) between the instrument gas source and the DESI emitter. The nebulizer flow gas was set via the instrument control software to 2.5 L/min (440 m/s at the aperture of the nebulizer). The ES high voltage was applied to the stainless-steel body of the spray head. The ES solvent was delivered to the emitter by a syringe pump using a 1.0 mL glass syringe. A grounded union was placed in the transfer line (75 cm long, 127 µm i.d., 1/16 in. o.d., PEEK tubing) between the syringe pump and the DESI emitter. The background loop current was measured by grounding this union through a Keithley (Cleveland, OH) 610C electrometer. In positive ion mode, the anodic emitter electrode current can be calculated by summing the cathodic background loop current and the anodic electrospray current shown in the XCalibur software’s Tune Plus window. Methanol was used as the spray solvent at a flow rate of 5 µL/min. The surface to be analyzed was placed upon an insulating block secured to an MS 2000 x/y/z robotic platform (Applied Scientific Instrumentation, Inc., Eugene, OR). This platform and associated manual positioning controls and computer control software have been previously described.13 The surface was positioned such that the bottom of the atmospheric pressure sampling capillary extension was just touching the surface. A charge-coupled device (CCD) camera (Costar Imaging, Anaheim, CA) and a monitor were used to aid the positioning of the DESI emitter and sampling capillary. A second CCD camera with an Optem 70XL zoom lens (Qioptiq, Fairport, NY) was mounted directly over the surface to be analyzed and centered on the DESI plume impact region. The image was output to a second monitor for a magnified view of the impact plume and sample spot and used to aid in x-y positioning. The surfaces were scanned from left to right relative to the sampling capillary of the mass spectrometer at a scan rate of 100 µm/s.
(12) Hovorka, S. W.; Scho ¨neich, C. J. Pharm. Sci. 2001, 90, 253-269.
(13) Van Berkel, G. J.; Kertesz, V. Anal. Chem. 2006, 78, 4938-4944.
Figure 1. Structure and mass-to-charge ratio for the compounds investigated.
DESI may be problematic, especially for the analysis of analytes having oxidative degradation pathways, such as pharmaceutical compounds.12 Many drugs are easily oxidized and may have oxidation products of the same mass-to-charge values or be structurally equivalent to the drug metabolites. In general, unwanted analyte oxidation hinders the analysis of unknowns and, in targeted analysis, diminishes the signal intensity of the precursor ion in the spectrum, a significant problem for quantitative analysis, especially when working close to the limit of detection for a compound. In the present report, we investigate the effect of ambient conditions, spray voltage, analyte surface concentration, and surface type on oxidation of the model compounds tamoxifen and reserpine (Figure 1) during DESI-MS. The parameters that both minimize and maximize oxidation of these analytes are delineated and, at least in part, explained. From this, recommendations are made for DESI-MS operation to avoid these unwanted reactions.
Analytical Chemistry, Vol. 80, No. 4, February 15, 2008
1209
Safety Considerations. The DESI emitter floats at high ES voltage, and appropriate shields and interlocks should be used to avoid accidental contact with this component. RESULTS AND DISCUSSION One anticipated widespread use of DESI-MS is for the analysis of sample spot arrays.14 Because analytes are deposited on these surfaces prior to analysis and because of the ambient nature of the typical DESI-MS experiment, the analyte spots can be exposed to air and light for significant periods of time (from minutes to hours or more). As such, the opportunity exists for simple ambient oxidation processes to alter the analytes. For example, this “autoxidation” may occur upon diffusion of molecular oxygen into the lattice structure of the crystallized analyte, a process that may be enhanced by the presence of a moisture layer resulting from humid conditions and that can be photolytically initiated.12 As such, the extent of autoxidation that any given analyte can undergo in a specific time period has the potential to vary from day to day with changes in the laboratory environment. The extent of autoxidation was monitored in our experiments using desorption sonic spray ionization (DeSSI).15 That is, equivalent DESI conditions were used, but the high voltage on the sprayer was set to 0 kV. Turning off the high voltage eliminated the possibility of oxidation caused by reactive species created electrochemically in solution16 or in the gas phase by, in this case, unintended discharge processes.17,18 This option was feasible in this case because the analytes under investigation could be sampled and ionized using both techniques, although signal levels were ∼5-6 times lower when using DeSSI than when using DESI at 4 kV. This will not be true for all analytes. Given the conditions in our laboratory and the analytes chosen for investigation, autoxidation could generally be kept to a minimum if the time between sample deposition and analysis was kept under 10 min and overhead fluorescent lighting fixtures were turned off. Keeping the deposited samples in an inert atmosphere (N2) until just prior to the analysis was found to allow a longer time window between sample deposition and analysis without oxidation. When operating under these ambient conditions to minimize autoxidation, we consistently noted a significantly greater abundance of analyte oxidation products when using DESI-MS rather than DeSSI-MS. We also noted that the oxidation products were not observed in ESI mass spectra acquired either in ESI mode or by spraying a micromolar solution of the compound of interest at the surface on the same day and with the same emitter in the DESI geometry. Furthermore, a test strip for peroxides tested positive after it had been exposed to the DESI spray (2 mM ammonium acetate in methanol) for 10 min at 5 kV, while no color change was seen at 0 kV (Figure S1, Supporting Information). According to the instruction sheet provided with the test strip, a peroxidase is used to transfer oxygen from peroxide to an organic redox indicator, which forms a blue oxidation product upon (14) Pasilis, S. P.; Kertesz, V.; Van Berkel, G. J. Anal. Chem. 2007, 79, 59565962. (15) Haddad, R.; Sparrapan, R.; Eberlin, M. N. Rapid Commun. Mass Spectrom. 2006, 20, 2901-2905. (16) Van Berkel, G. J.; Kertesz, V. Anal. Chem. 2007, 79, 5510-5520. (17) Mahle, N. H.; Cooks, R. G.; Korzeniowski, R. W. Anal. Chem. 1983, 55, 2272-2275. (18) Maleknia, S. D.; Chance, M. R.; Downard, K. M. Rapid Commun. Mass Spectrom. 1999, 13, 2352-2358.
1210 Analytical Chemistry, Vol. 80, No. 4, February 15, 2008
Figure 2. Mass spectra showing the effect of spray voltage on tamoxifen oxidation during DESI analysis from a PTFE-printed slide at a spray voltage (background loop current, spray current) of (a) 0 (0.1 and 0.1 µA), (b) 2 (6.4 and 0.5 µA), (c) 4 (13.1 and 1.1 µA), and (d) 6 kV (19.9 and 1.9 µA) for a 50 pmol spot of tamoxifen on a PTFEprinted slide. (e) Oxidation efficiency as a function of background loop current. Each point represents 5-6 replicate analyses. Oxidized species appear at m/z 388 and 404. The spray solvent used was 2 mM ammonium acetate in methanol at a solvent flow rate of 5 µL/ min and a gas flow rate of 440 m/s.
exposure to H2O. This result was consistent with the delivery of a peroxide or other ROS species (which may possibly react with the redox indicator independently of the peroxidase) from the DESI emitter to the surface. To investigate this DESI-mediated oxidation process further, we examined the effects of spray voltage (and associated emitter electrode current), analyte surface concentration, and surface type on the extent of analyte oxidation. Increases in spray voltage (and the associated emitter electrode current) led to increased abundance of analyte oxidation products in the DESI mass spectra. However, analyte oxidation was not seen during ESI in DESI mode (analyte-containing spray impacting the surface and sampled by the sampling capillary) under conditions equivalent in terms of spray voltage, gas flow rate, and solvent flow rate to those used during DESI analysis. Nor was oxidation seen during ESI mode, in which the analyte-containing spray was sampled directly such that no surface contact was made. Figure 2 shows the effect of increasing spray voltage (emitter electrode current) on tamoxifen
Table 1. Major Electrochemical Oxidation Reactions Involving Water Anticipated in Typical ES Solvent Systems oxidation (positive-ion mode)
E0/V vs SHE32
2H2O ) O2 + 4H+ + 4e2H2O ) H2O2 + 2H+ + 2eH2O + O2 ) O3 + 2H+ + 2eH2O ) •OH + H+ + e-
1.229 1.763 2.075 2.38
oxidation during DESI analysis from a PTFE-printed slide. The spray solvent used was 2 mM ammonium acetate in methanol. The product at m/z 388, which increases from a relative abundance of 4% at a spray voltage of 0 kV (0.1 µA) to 38% at 6 kV (19.9 µA) (Figure 2a-d), has the same mass-to-charge value as several important tamoxifen metabolites, including tamoxifen N-oxide, R-hydroxytamoxifen, and 4-hydroxytamoxifen.19 Also seen at 6 kV was an oxidation product at m/z 404 resulting from a second hydroxylation; 4-hydroxytamoxifen N-oxide, yet another tamoxifen metabolite, has an m/z value of 404.19 Products of electrochemical oxidation of tamoxifen include N-desmethyltamoxifen at m/z 358, also an important metabolite, and one or more of the species listed above, as well as a product at m/z 404.20 Figure 2e shows the overall oxidation efficiency as measured by the total oxidized product abundance. Increases in spray voltage also led to increased abundance of analyte oxidation products of the tamoxifen-related compounds 4-hydroxytamoxifen, clomiphene, and toremifene. Increases in spray voltage also led to increases in analyte oxidation product ion abundances for verapamil. Figure S2a-d (Supporting Information) shows spectra of tamoxifen, 4-hydroxytamoxifen, toremifene, and clomiphene acquired at spray voltages of 0 and 5 kV. Figure S3 shows spectra of verapamil acquired at spray voltages of 0 and 5 kV. Note that these oxidation products were also observed when using both unmodified methanol and acetonitrile. However, by adding 2 mM ammonium acetate to the methanol spray solvent the extent of oxidation was increased. Adding this electrolyte increased solution conductivity, and thus the current at the emitter electrode was significantly increased. These conditions generally increase the solution concentration of electrochemically produced products, which would include potential reactive species.16 Thus, the data in Figure 2 are consistent with, but not conclusive for, the electrochemical formation of one or more long-lived oxidative species at the DESI emitter electrode that can be transmitted to the surface and oxidize the analyte. ROS species that can form during electrolysis of H2O include, for example, the hydroxyl radical (•OH), singlet oxygen (1O2), and peroxo compounds (i.e., •OOH and H2O2), as well as ozone (O3).21 E0 values for these reactions are given in Table 1. The hydroxyl radical has a lifetime on the order of 10-10 s in CH3OH22 and is thus unlikely to survive for the time required for travel from the point of generation in the emitter to the spray tip (∼1.3 s at a flow rate of 5 µL/min) and from the spray tip to the surface (∼4.5-7 µs at a gas flow (19) Lu, W.; Poon, G. K.; Carmichael, P. L.; Cole, R. B. Anal. Chem. 1996, 68, 668-674. (20) Van Berkel, G. J.; Deng, H., unpublished results. (21) Wabner, D.; Grambow, C. J. Electroanal. Chem. 1985, 195, 95-108. (22) Buxton, G. V.; Greenstock, C. L.; Helman, W. P.; Ross, A. B. J. Phys. Chem. Ref. Data. 1988, 17, 513-886.
rate of 440 m/s). However, singlet oxygen can survive for 7 µs in CH3OH,23 and H2O2 is quite stable. Because we desired to examine oxidation under typical ambient DESI conditions, we made no attempt to eliminate oxygen or water in these experiments. Thus, molecular oxygen and trace quantities of water are presumably present in the spray solvent and, of course, are present in the lab air (20% O2, ca. 50-60% relative humidity at 22 °C). Oxygen and ROS species can potentially be generated in the emitter through the oxidation of water. Deliberate addition of up to 1% v/v water to the spray solvent did not increase the amount of oxidation observed. Degassing the spray solvent to remove oxygen did not decrease the extent of analyte oxidation. A significant quantity of molecular oxygen can also rapidly diffuse through the surface of the Taylor cone and be transported inside the ES capillary where, if the capillary is made of a conductive material, it may become involved in electrochemical processes.24 In the case of the experiments described here, however, a nonconductive silica capillary was used; thus, it seems improbable that any electrochemical reactions involving oxygen could take place at the capillary, although oxygen absorbed in this way could certainly be transmitted to the surface via the DESI spray. Because little or no oxidation was seen under DeSSI (0 kV) conditions, oxidation by the molecular oxygen present in the spray solvent before electrolysis appears unlikely to be a significant factor in the oxidation of the analytes examined here. However, molecular oxygen could feasibly play a role in oxidation if the reactant species are formed by a different mechanism at the DESI emitter spray tip. One means to prevent the electrochemical formation of reactive species is to add a redox buffer, such as hydroquinone (E0 ) 0.699 V vs SHE), to the spray solvent.25,26 This maintains the interfacial potential at the DESI emitter electrode below that required for water oxidation (Table 1) and is therefore expected to mitigate the formation of ROS from water electrolysis. The effect of adding hydroquinone to the spray solvent during DESI analysis of 198 pmol spots of reserpine from a PTFE-printed slide is shown in Figure 3a-d. Prior to the addition of hydroquinone to the spray solvent, there was significant formation of the major oxidation products 3,4-dehydroreserpine (M + H+, m/z 607), 1-hydroxy3,4-dehydroreserpine (M + H+, m/z 623), and 1-hydroxyreserpine (M + H+, m/z 625) (Figure 3a). Ions at m/z 641 and 657 are presumably the results of further hydroxylation reactions. As the concentration of hydroquinone in the spray solvent was increased, the abundances of the oxidized products decreased slightly. At a hydroquinone concentration of 10 mM, the relative abundance of 1-hydroxyreserpine (m/z 625) decreased to ∼20% from its initial relative abundance of ∼32% (Figure 3b). Sodiated ions (M + Na+) of reserpine, 1-hydroxyreserpine, and the two products of hydroxylation appear at m/z 631, 647, and 663 and 679, respectively. At a hydroquinone concentration of 30 mM, the relative abundance of 1-hydroxyreserpine (m/z 625) decreased to ∼14% (Figure 3c). However, increasing the hydroquinone concentration to 60 mM had no further effect (Figure 3d). In fact, the relative abundance of 1-hydroxyreserpine (m/z 625) actually increases to ∼20%. (23) Merkel, P. B.; Kearns, D. R. J. Am. Chem. Soc. 1972, 94, 7244-7253. (24) Pozniak, B. P.; Cole, R. B. Anal. Chem. 2007, 79, 3383-3391. (25) Moini, M.; Cao, P.; Bard, A. J. Anal. Chem. 1999, 71, 1658-1661. (26) Smith, A. D.; Moini, M. Anal. Chem. 2001, 73, 240-246.
Analytical Chemistry, Vol. 80, No. 4, February 15, 2008
1211
Figure 3. Mass spectra showing the effect of hydroquinone in spray solvent on reserpine oxidation during DESI analysis of a 198 pmol spot from a PTFE-printed slide for (a) no hydroquinone, (b) 10 mM hydroquinone, (c) 30 mM hydroquinone, and (d) 60 mM hydroquinone in 2 mM ammonium acetate in methanol at a spray voltage of 5 kV. Oxidized species are marked with an asterisk. (e) Oxidation as a function of hydroquinone concentration. Each point represents 5-6 replicate analyses. The solvent flow rate was 3 µL/min, and the gas flow rate was 440 m/s.
Figure 3e shows the effect of the addition of hydroquinone to the spray solvent on the total oxidized product abundance. In the case of tamoxifen, the addition of hydroquinone to the spray solvent likewise decreased, but did not eliminate, oxidation (data not shown). We also observed that increasing the solvent flow rate was effective in mitigating the extent of analyte oxidation (Figure 4a-c). By increasing the solvent flow rate from 3 to 5 to 10 µL/ min the relative abundance of 1-hydroxyreserpine decreased from ∼32 to 13 to 6%, respectively. Figure 4d shows the effect of flow rate on the total oxidized product abundance. The increased solvent flow rate would decrease the concentration of electrolysis products because the emitter electrode current is relatively independent of flow rate.27 The fact that the hydroquinone redox buffer did reduce the extent of reserpine oxidation and the extent of oxidation decreased with increased solvent flow rate is consistent with the electrochemical origin of at least some of the species leading to oxidation of this analyte. (27) Kertesz, V.; Van Berkel, G. J. J. Mass Spectrom. 2001, 36, 204-210.
1212 Analytical Chemistry, Vol. 80, No. 4, February 15, 2008
Figure 4. Mass spectra showing the effect of solvent flow rate on reserpine oxidation during DESI analysis of a 198 pmol spot from a PTFE-printed slide for (a) 3 µL/min, (b) 5 µL/min, and (c) 10 µL/min. Oxidized species are marked with an asterisk. (d) Oxidation as a function of solvent flow rate. Each point represents 5-6 replicate analyses. The spray solvent was 2 mM ammonium acetate in methanol. The spray voltage was 5 kV, and the gas flow rate was 440 m/s.
An alternative explanation for the electrochemical origin of ROS, ozone, or other species leading to analyte oxidation is the creation of these species through a corona discharge.17,18,28 Mahle et al.17 noted the hydroxylation of aromatic hydrocarbons during atmospheric pressure chemical ionization conditions, which they ascribed to gas-phase reactions with species formed from molecular oxygen in the discharge. Downard and colleagues18,29,30 used the discharge phenomena that can occur in pneumatically assisted electrospray to tag proteins with oxygen functionalities. To enhance that process they used pure O2 as a nebulizing gas and a high emitter electrode voltage (8 kV). In our DESI experiments we did not specifically examine the effect of a possible corona discharge on analyte oxidation. However, the use of air as a nebulizing gas (ca. 20% O2) did not increase the amount of oxidation detected over that observed when using N2, nor was the characteristic blue glow of the discharge observed. That the amount of oxidation was not increased may not be surprising, given that the DESI experiment, while using N2 as a nebulizing (28) Thomas, M. C.; Mitchell, T. W.; Blanksby, S. J. J. Am. Chem. Soc. 2006, 128, 58-59. (29) Wong, J. W. H.; Maleknia, S. D.; Downard, K. M. Anal. Chem. 2003, 75, 1557-1563. (30) Wong, J. W. H.; Maleknia, S. D.; Downard, K. M. J. Am. Soc. Mass Spectrom. 2005, 16, 225-233.
Figure 5. Extracted ion chromatograms showing the “coffee ring” effect on tamoxifen oxidation during DESI analysis of a 5 pmol spot of tamoxifen at a spray voltage of 4 kV. The spray solvent used was 2 mM ammonium acetate in methanol at a solvent flow rate of 5 µL/min and a gas flow rate of 440 m/s. The scan rate was 100 µm/s. Figure 7. Mass spectra showing oxidation of reserpine during DESI analysis of 193 pmol spots from (a) a PTFE-printed slide, (b) Whatman No. 42 filter paper, and (c) a glass slide. Oxidized species are marked with an asterisk. The spray solvent used was 2 mM ammonium acetate in methanol at a spray voltage of 4 kV. The solvent flow rate was 5 µL/min, and the gas flow rate was 440 m/s.
Figure 6. Mass spectra showing the effect of surface concentration on tamoxifen oxidation during DESI analysis from a PTFE-printed slide for an (a) 250, (b) 50, and (c) 5 pmol for a tamoxifen spot at a spray voltage of 4 kV. Oxidized species appear at m/z 388 and 404. (d) Oxidation as a function of surface concentration. Each point represents 3-4 replicate analyses. The spray solvent used was 2 mM ammonium acetate in methanol at a solvent flow rate of 5 µL/min and a gas flow rate of 440 m/s.
gas, was conducted in ambient air. Pure O2 was not tested because of safety restrictions in our laboratory. However, ionic species consistent with oxygen additions to the analyte were more abundant in our spectra as the high voltage was increased. In fact, the DESI spray current measured at 5 kV was 1.9 µA, which is significantly higher than the current levels normally associated with a pure electrospray process (typically
