Developing Repeatable Measurements for Reliable Analysis of

Anal. Chem. 2009, 81, 2286–2293

Developing Repeatable Measurements for Reliable Analysis of Molecules at Surfaces Using Desorption Electrospray Ionization F. M. Green,*,† P. Stokes,‡ C. Hopley,‡ M. P. Seah,† I. S. Gilmore,† and G. O’Connor‡ National Physical Laboratory, Teddington, Middlesex TW11 0LW, U.K., and LGC, Teddington, Middlesex TW11 0LY, U.K. Desorption electrospray ionization (DESI) is a powerful ambient ionization technique that can provide highsensitivity mass spectrometry information directly from surfaces at ambient pressure. Although a growing amount of research has been devoted to exploring different applications, there are few studies investigating the basic parameters and underpinning metrology. An understanding of these is crucial to develop DESI as the robust and reliable technique required for significant uptake by industry. In this work, we begin with a systematic study of the parameters affecting the repeatability, sensitivity, and rate of consumption of material with DESI. To do this we have developed a model sample consisting of a thin uniform film of controlled thickness of Rhodamine B on glass. This model sample allowed assessment of optimal sensitivity and spot shape under different conditions. In addition, it allowed us to study the surface in more detail to understand why and how each parameter affects these. Using the model sample to optimize the instrument parameters for DESI led to an absolute intensity repeatability of better than 15%, achieved over a period of 1 day. This model sample provides valuable insight into the electrospray-sample interaction and the desorption mechanism. Confocal microscopy of areas analyzed by DESI allow droplet distribution, material utilization, and spot size to be determined. Studying surface erosion also gives the erosion rate of material, analogous to the sputtering yield in secondary ion mass spectrometry. The results of the study provide a clear description that explains the differences observed with changing electrospray parameters allowing optimization of the technique, for both spatial resolution and sensitivity. There has been an explosion in the growth of ambient surface mass spectrometries with new ionization and desorption methods developing rapidly. Surface chemical analytical techniques such as secondary ion mass spectrometry (SIMS)1 are extremely powerful, providing high-sensitivity molecular information at better * Corresponding author. Phone: +44 (0)20 8943 6153. Fax: +44 (0)208 8943 6453. E-mail: [email protected]. † National Physical Laboratory. ‡ LGC. (1) Surface Analysis: The Principal Techniques , 2nd ed.; Vickerman, J. , Gilmore, I., Eds.; Wiley: New York, 2009.

2286

Analytical Chemistry, Vol. 81, No. 6, March 15, 2009

than 200 nm spatial resolution (an ability to distinguish, through mass spectra, features at sub-200 nm). However, the requirement for SIMS analysis to be in vacuo is, for many applications, a severe drawback. This is, of course, particularly important for biological applications. In the vanguard of these ambient methods is desorption electrospray surface ionization (DESI) developed by Cooks and co-workers.2,3 DESI has already been shown to have great potential in a wide range of application areas from forensics and homeland security,4-6 through to counterfeit detection of pharmaceuticals,7 environmental analysis,8 and biological analysis.9,10 Of the wide variety of ambient desorption methods, it is clear that DESI has one of the strongest uptakes. For rapid and reliable analysis in industry, the robustness and repeatability of measurements is critical. Discussion with potential industrial users has shown that DESI uptake would be faster once repeatability, reliability, and a basic measurement infrastructure are in place. Furthermore, there are increasing industrial requirements to comply with ISO 17025 (general requirements for the importance of testing in calibration laboratories) or Food and Drug Administration regulations, which require these aspects. Over many years, the National Physical Laboratory (NPL) has developed the underpinning metrology base for surface chemical analytical techniques such as X-ray photoelectron spectroscopy (XPS), where the measurement infrastructure now allows quantification traceable to the SI (International System of Units),11 and (2) Takats, Z.; Wiseman, J.; Gologan, B.; Cooks, R. G. Science 2004, 306, 471. (3) Cooks, R. G.; Ouyang, Z.; Takats, Z.; Wiseman, J. Science 2006, 311, 1566. (4) Cotte-Rodriguez, I.; Takats, Z.; Talaty, N.; Chen, H.; Cooks, R. G. Anal. Chem. 2005, 77, 6755. (5) Justes, D. R.; Talaty, N.; Cotte-Rodriguez, I.; Cooks, R. G. Chem. Commun. 2007, 21, 2142. (6) Cotte-Rodriguez, I.; Hernandez-Soto, H.; Chen, H.; Cooks, R. G. Anal. Chem. 2008, 80, 1512. (7) Yadong, L.; Late, S.; Green, M. D.; Banga, A.; Fernandez, F. A. J. Am. Soc. Mass Spectrom. 2008, 19, 380. (8) Chen, H.; Li, M.; Zhang, Y. P.; Yang, X.; Lian, J. J.; Chen, J. M. J. Am. Soc. Mass Spectrom. 2008, 19, 450. (9) Manicke, N. E.; Wiseman, J. M.; Ifa, D. R.; Cooks, R. G. J. Am. Soc. Mass Spectrom. 2008, 19, 531. (10) Shin, Y. S.; Drolet, B.; Mayer, R.; Dolence, K.; Basile, F. Anal. Chem. 2007, 79, 3514. (11) Seah, M. P.; Spencer, S. J.; Bensebaa, F.; Vickridge, I.; Danzebrink, H.; Krumrey, M.; Gross, T.; Oesterle, W.; Wendler, E.; Rheinl¨ander, B.; Azuma, Y.; Kojima, I.; Suzuki, N.; Suzuki, M.; Tanuma, S.; Moon, D. W.; Lee, H. J.; Cho, H. M.; Chen, H. Y.; Wee, A. T. S.; Osipowicz, T.; Pan, J. S.; Jordaan, W. A.; Hauert, R.; Klotz, U.; van der Marel, C.; Verheijen, M.; Tamminga, Y.; Jeynes, C.; Bailey, P.; Biswas, S.; Falke, U.; Nguyen, N. V.; ChandlerHorowitz, D.; Ehrstein, J. R.; Muller, D.; Dura, J. A. Surf. Interface Anal. 2004, 36, 1269.

10.1021/ac802440w CCC: $40.75 Published 2009 by the American Chemical Society Published on Web 02/23/2009

secondary ion mass spectrometry (SIMS), where over 2 orders of magnitude improvement in repeatability12 has been achieved. We begin a similar approach for DESI through a systematic study of the parameters affecting repeatability, sensitivity, and rate of consumption of material. We have developed a well-controlled model sample of Rhodamine B on glass that allows the effect of key experimental parameters such as geometry and spray conditions to be studied and optimized. Although the solvent composition is a key parameter, here we keep it constant, throughout, to elucidate the effects of other parameters. Rather than complicate the present paper, in future work, the effect of the substrate, analyte, and solvent choices will be investigated. An understanding of the DESI mechanism is critical for defining the range of applicability, the limitations and advantages of DESI, and improving the repeatability, sensitivity, and spatial resolution of the technique. Research by Takats et al.13 shows that the desorption and ionization mechanisms change for different analytes, substrates, and spray solution chemistries. On the basis of research on a number of different materials, two distinct optimal geometries were found for two classes of analyte. The first group are those such as peptides and proteins, these required short spray/surface distances with near normal impact angles. Takats et al.13 suggests that this indicates that the droplet impact on the surface leads to formation of charged droplets and this results in spectra that resemble those using electrospray ionization. The second group are analytes including materials such as cholesterol and dye molecules, similar to those used here. Here, the strong dependence on spray/surface distance and temperature led Takats et al.13 to suggest that the charge transfer occurred between solvent ions and the analyte molecules, with a mechanism not necessarily involving droplets but gas phase ions. Both mechanisms of DESI are critically dependent on the spray/surface interaction; water or solvent droplets impacting with the surface of interest. Venter14 et al. used particle dynamics analysis (PDA), a Doppler method, to measure the droplet diameters and velocities both as they move from the spray tip and after impact with a surface. They found that as a droplet moves further away from the spray tip, the droplet velocity decreases. Therefore, the closer the spray tip is to the surface, the greater the velocity with which the droplets impact that surface. After droplets impact the surface, secondary droplets rebound at a range of angles, those droplets leaving the surface at smaller angles, (moving close to the surface) were found to have the highest velocities. These experimental results agree with recent modeling results by Costa15 that show the nebulizing gas from the electrospray source sweeps along close to the surface around the impact site, causing high velocities of the droplets in this region. The electrospray droplet dynamics and subsequent interaction of the droplets with the surface defines the ionization and desorption mechanisms of DESI. The electrospray process itself has been studied in detail,16-18 with work investigating droplet (12) Gilmore, I. S.; Seah, M. P.; Green, F. M. Surf. Interface Anal. 2005, 37, 651. (13) Takats, Z.; Wiseman, J.; Cooks, R. G. J. Mass Spectrom. 2005, 40, 1261. (14) Venter, A.; Sojka, P. E.; Cooks, R. G. Anal. Chem. 2006, 78, 8549. (15) Costa, A. B.; Cooks, R. G. Chem. Commun. 2007, 3915. (16) Kerbale, P.; Yeunghaw, H. Electrospray Ionization Mass Spectrometry , 1st ed.; Wiley: New York, 1997. (17) Kerbale, P.; Tang, L. Anal. Chem. 1993, 65, 972A. (18) Gomez, A.; Tang, K. Phys. Fluids 1994, 6, 404.

size, charge, and breakdown.19 This shows that the droplets typically have charges of ∼4 × 106 electrons and diameters of around 25 µm (for 10-4 M NaCl in water).19 A nebulizing gas is used in the electrospray process, and this reduces the droplet sizes further to around 4 µm and increases droplet velocity.16 The science behind water droplet impacts on surfaces has a long history and a good review is given by Yarin et al.20 This shows that a number of differing effects can be seen when a water droplet hits a hard surface, such as crown formation, where water is scattered in a circular splash about the droplet impact and this can lead to “jetting” where a corona of secondary droplets form at the top of the splash.20-22 This is one of the droplet splashing effects shown to occur in flow dynamic simulations by Costa15 of droplets hitting a wetted surface during DESI. In this study, we extend and refine the method for assessing optimal geometry as considered by Takats et al.,13 by producing a uniform model sample. This both removes the problem of sweet spots, where a concentration distribution of analyte across the surface dominates the differences in intensity that interfere with assessment of optimal sensitivity, as well as allowing us to study the surface in more detail. In addition, the model sample allows the optimal geometry to be assessed in terms of spot shape and spatial resolution. The following research also builds on the study of droplet dynamics by Venter,14 by directly visualizing the next step, the droplet interaction with the surface and the dynamics of the droplets on the surface. This ties in with the theoretical work from Costa.15 Although Costa’s work was on a small scale, by visualizing how the results of the droplet interaction with the surface could be scaled up, we are able to validate the results experimentally at the practical scale. Research by Van Berkel et al.23 investigating the spray/surface interaction, identifies three zones, an inner region where most desorption takes place, an outer elliptical zone where solvent “jetting” from the inner region impacts the surface, and a periphery of larger, slow moving droplet impacts. With the desorption region of the electrospray scanned over a thin line of analyte, the central region is shown to be where the most effective desorption and ionization takes place, although in these experiments washing effects may displace materials significantly. In this work, we study the surface crater and the surrounding disrupted material formed on a flat surface after DESI impact. This gives insights into the interaction mechanism of the electrospray jet with the surface. In addition, rates of material erosion can be calculated and correlated with the detected signal to give the efficiency and utilization of DESI for different acquisition times and conditions. This is important to allow future improvements in spatial resolution and to understand why experimental parameters affect sensitivity and repeatability. The rate at which material is removed may enable accurate quantitative measurements to be made. (19) Smith, J. N.; Flagan, R. C.; Beauchamp, J. L. J. Phys. Chem. A 2002, 106, 9957. (20) Yarin, A. L. Annu. Rev. Fluid Mech. 2006, 38, 159. (21) Levin, Z.; Hobbs, P. V. Philos. Trans. R. Soc. London, Ser. 1971, 269, 555. (22) Yarin, A. L.; Weiss, D. A. J. Fluid Mech. 1995, 283, 141. (23) Pasilis, S. P.; Kertesz, V.; Van Berkel, G. J. Anal. Chem. 2007, 79, 5956. (24) Prosolia Inc. www.prosolia.com.

Analytical Chemistry, Vol. 81, No. 6, March 15, 2009

2287

Figure 1. Schematic showing optimal DESI parameters.

EXPERIMENTAL SECTION Here, we modified a nanospray source for the QTRAP 4000 mass spectrometer (Applied Biosystems, Warrington, U.K.) to enable it to perform DESI experiments. This instrument offers good sensitivity and the capability to perform MS, MS/MS, and MS3 experiments. Furthermore, other users working with different ionization sources could easily replace the DESI head. The DESI head consists of a sample manipulator and DESI source. The sample manipulator has two parts: a manual XYZ sample stage with large (millimeter) travel in each direction to provide coarse alignment and a computer controlled XY sample stage to provide both fine adjustment of the sample position and also imaging capability with a step size of 5 µm. The sample is fixed on a PTFE tray designed to accommodate a standard glass microscope slide. The DESI source is a modified nanospray source with a manual XYZ manipulator for coarse alignment. A linear manipulator and a miniature rotational manipulator (Laser 2000, Kettering, U.K.) were added to the nanospray source to allow for fine adjustments of the electrospray tip height, d1, and the angle of incidence, R, as shown in Figure 1. We used the standard microion spray head as supplied by Applied Biosystems with stainless steel (o.d. 320 µm, i.d. 50 µm) or fused silica sprayer tips (o.d. 360 µm, i.d. 30 µm) both available from Presearch (Basingstoke, U.K.). The only other modification was the attachment of a 3 cm piece of 1.6 mm (o.d., i.d. 1.02 mm) stainless steel tubing (Anachem, Luton, U.K.) called the “sniffer”, with a 10° bend halfway along its length to the MS interface. The sniffer facilitates efficient collection of the desorbed material into the MS inlet. The original cameras and light source supplied with the nanospray source were also used to aid positioning. In this work, a consistent solvent composition consisting of 0.1% formic acid (Fisher Scientific, Loughborough, U.K.) in acetonitrile (Optigrade, LGCPromochem, Teddington, U.K.)/water (18 MΩ cm-1) (50:50) has been used throughout. Although the solvent composition may affect the sensitivity, spot size, and wetability, here, we have kept it constant in order to focus on the effects of other parameters. The standard operating conditions, unless otherwise stated, were an electrospray voltage of 5000 V, a solvent flow rate, S, of 1 µL min-1, a nebulizing gas flow rate, G, at a midpoint (6 L min-1), an angle of incidence, R, of either 0° or 40°, a tip-to-sample distance, d1, of 1.5-2 mm, a tip-to-sniffer distance, d3, of ∼5 mm, a sniffer-to-sample distance, d2, of