Design and Performance of an Instrument for Soft Landing of

Aug 2, 2007 - Omar Hadjar, Peng Wang, Jean H. Futrell, Yury Dessiaterik, Zihua Zhu, ... (10) Alvarez, J.; Cooks, R. G.; Barlow, S. E.; Gaspar, D. J.; ...
0 downloads 0 Views 536KB Size
Download PDF
Anal. Chem. 2007, 79, 6566-6574

Design and Performance of an Instrument for Soft Landing of Biomolecular Ions on Surfaces Omar Hadjar, Peng Wang, Jean H. Futrell, Yury Dessiaterik, Zihua Zhu, James P. Cowin, Martin J. Iedema, and Julia Laskin*

Pacific Northwest National Laboratory, Fundamental Science Directorate and Environmental Molecular Sciences Laboratory, Richland, Washington 99352

A new ion deposition apparatus was designed and constructed in our laboratory. Our research objectives were to investigate interactions of biomolecules with hydrophilic and hydrophobic surfaces and to carry out exploratory experiments aimed at highly selective deposition of spatially defined and uniquely selected biological molecules on surfaces. The apparatus includes a hightransmission electrospray ion source, a quadrupole mass filter, a bending quadrupole that deflects the ion beam and prevents neutral molecules originating in the ion source from impacting the surface, an ultrahigh vacuum (UHV) chamber for ion deposition by soft landing, and a vacuum lock system for introducing surfaces into the UHV chamber without breaking vacuum. Ex situ analysis of surfaces following soft landing of mass-selected peptide ions was performed using 15 keV Ga+ time-of-flight secondary ion mass spectrometry and grazing incidence infrared reflection-absorption spectroscopy. It is shown that these two techniques are highly complementary methods for characterization of surfaces prepared with a range of doses of mass-selected biomolecular ions. We also demonstrated that soft landing of peptide ions on surfaces can be utilized for controlled preparation of peptide films of known coverage for fundamental studies of matrix effects in SIMS. Soft landing of mass-selected ions onto surfaces is relevant to a broad range of scientific disciplines, including surface science, materials science, mass spectrometry, interfacial surface imaging, and spectroscopy. This technique was first introduced by Cooks and co-workers in 1977 as a means of surface modification.1 Ionsurface collisions at hyperthermal (5 mm) spot size were conducted by placing the target surface after the resolving quadrupole. In these experiments, the spot size was defined using a 5-10-mm restricting aperture plate mounted 4 mm in front of the surface. With this configuration, we achieved 300-500-pA current of mass-selected peptide ions on the target and reduced the deposition time by 1 order of magnitude. However, it should be noted that because the surface is located on axis with the electrospray ion source and maintained at a relatively high pressure of of 4 × 10-5 Torr, it is most likely exposed to molecules originating in the atmospheric pressure interface during the time of the deposition. Ion-Transfer Optics. As briefly mentioned above, m/zselected ions are injected through a 2-mm conductance limit following the gate valve and are turned 90° by an electrostatic quadrupole held at 10-7 Torr to eliminate neutral contamination in the UHV region. Einzel lenses precede and follow the turning quadrupole to guarantee proper focusing of the ion beam entering and leaving this turning element. The ions are injected into the UHV deposition chamber held at 2 × 10-9 Torr through a 2-mm conductance limit and focused with another Einzel lens. The ions are finally decelerated immediately in front of the soft-landing surface; length of the deceleration region is limited to 2 mm by an electrically isolated high-transmission mesh mounted right after the Einzel lens. This combination allows us to control soft-landing energies and spot sizes from around 5 mm down to 1-mm diameter in the UHV chamber. (27) Brubaker, W. M. Adv. Mass Spectrom. 1968, 4, 293.

Analytical Chemistry, Vol. 79, No. 17, September 1, 2007

6567

Figure 1. (a) Schematic view of the ion soft-landing instrument: I, electrospray source (760 Torr). II, high-transmission ion funnel (2 × 10-1 Torr). III, ion thermalization and focusing stage (10-2 Torr). IV, m/z ion selection stage (4 × 10-5 Torr). V, 90° ion bending stage (10-7 Torr). VI, UHV chamber for ion soft landing (2 × 10-9 Torr). VII, surface introduction stage (from 760 to 2 × 10-8 Torr). (1) Syringe pump, (2) HV needle, (3) heated capillary, (4) electrodynamic ion funnel, (5) collision quadrupole (CQ), (6) 1-mm conductance limit (CL), (7) prefilter, (8) resolving quadrupole (RQ), (9) postfilter, (10) Einzel lenses, (11) gate valve, (12) 2-mm CL, (13) electrostatic quadrupole (bender), (14) deceleration area, (15) surface and phosphorus screen detector, (16) CCD camera, and (17) magnetic translator. (b) Schematic view of the sample-transfer system for the transfer and positioning of the surface holder inside the UHV chamber (see text for more details).

Focusing of the ion beam prior to ion deposition experiments is facilitated by a phosphorus screen detector mounted just below the surface holder. When this detector is lifted up into the ion beam path, a CCD camera displays a zoomed image of the spot on a monitor screen. A mesh in front of the MCP detector establishes the same electric field region that is used for the ion deposition. This allows us to monitor the spot size and beam quality for different collision energies chosen for our soft-landing experiments. Optimization of potentials throughout this complex ion optical system is facilitated by modeling ion trajectories using SIMION7. An illustration of calculated trajectories is presented in Figure 2 for ions exiting the RQ and transferred to the surface. Figure 2a 6568 Analytical Chemistry, Vol. 79, No. 17, September 1, 2007

is the 3D work bench view of ion trajectories that illustrates our electrostatic optics configuration. The ion beam cross section broadening and angular dispersion introduced by the RQ are approximated by using different starting points and initial angles of the ions. Figure 2b is a potential energy surface view of the system that provides additional insight on the effect of different elements of the ion optics on ion trajectories. We note that optimum focus potentials deduced from the simulation are very close to the optimum focus potentials found experimentally. Ion Deposition. The first step in ion deposition is tuning the ion beam using the ion visualizer. Once the desired optimum focus is achieved, the visualizer is lowered down out of the ion path and replaced by the surface holder. The surface holder used in

Figure 2. SIMION simulation of ion trajectories for a 4-mm beam of doubly protonated GS exiting the RQ (a) 3D plot of the ion optics from the resolving quadrupole to the surface. (b) Potential surface plot showing the transport of the ion beam and the sudden deceleration stage right before the surface.

our soft-landing apparatus allows us to mount two target surfaces that are alternately used for soft-landing experiments, for example, two SAM substrates with different terminal groups. Ion beam currents on the target are measured by an analog electrometer and range from a few picoamperes to ∼60 pA. For doubly protonated gramicidin S as an example and using its measured28 geometrical cross section of 263 Å2 with 60 pA uniformly distributed over a spot size of 3-mm diameter deposits 7 × 1011 ions in 60 min and creates one-fourth of a monolayer (assuming 100% landing efficiency and close packing of soft landed ions). It should be noted that the actual efficiency of the soft landing was not examined in this study. Quantification of the amount of softlanded molecules or ions is difficult because the retention of different species on substrates depends on the binding strength to the surface, the physical and chemical properties of the surface, the structure of the ion and its kinetic energy, the pressure in the deposition region, and other factors. It follows that the values of surface coverage used in this work present only an upper estimate of the actual coverage achieved in our experiments. Precise control of the size and position of the ion beam allows us to deposit up to four well-resolved 1-3-mm spots on each of the two 10 × 10 mm2 surfaces. This capability is important for quantitative investigation of the effect of different factors (e.g., ion dose, initial charge state, the kinetic energy of projectile ions, etc.) on the soft-landing efficiency. UHV Chamber. Our soft-landing experiments were carried out in an UHV chamber maintained at a base pressure of 2 × 10-9 Torr using a 1000 L/s Thermionics ion pump (model COV1000). The sample is mounted on a Thermionics model FM104/ 1.88-2-2-4/B8T/LT xyz manipulator mounted on a differentially pumped rotary platform. Though not used in this work, this manipulator includes heating and cooling capabilities with a thermocouple connection for measuring and controlling the sample temperature. Vacuum Lock Sample Changer. A vacuum lock system was designed for sample introduction and removal that reduces the (28) Ruotolo, B. T.; Tate, C. C.; Russell, D. H. J. Am. Soc. Mass Spectrom. 2004, 15, 870.

pressure from ambient to 4 × 10-8 Torr in 30 min. A magnetic translator is used to move the surface holder in and out of the deposition chamber. A Thermionics model STLC sample-transfer system illustrated in Figure 1b is composed of the sample platen that holds the surface,29 the transfer fork, which is twisted to secure the platen and move it between the two vacuum chambers, and the sample dock that receives the platen from the fork. A simple twisting motion frees the platen from the fork while the dock both holds the platen within the deposition chamber and provides electrical connections to the platen. The special turn-tolock design ensures smooth and confident transfer of the sample into the ion deposition region. Materials and Chemicals. Gold substrates that are the base for forming alkyl thiol self-assembled monolayers were purchased from SPI Supplies (100-nm gold vacuum sputtered onto 500 µm thick 10 × 10 mm2 silicon wafers with 10-nm chromium adhesion layer). SAMs of 10-carboxy-1-decanethiol (COOH-SAM) and 1-dodecanethiol (HSAM) were prepared following literature procedures.30,31 After thorough cleaning, the substrates were immersed in the container with 1 mM solution of the corresponding thiol in ethanol for 12 h. The SAM surfaces were then removed from the thiol solutions, ultrasonically washed in ethanol (acetic acid/ethanol in the case of COOH-SAM) for 5 min, quickly dried with dry nitrogen gas, and placed in the UHV chamber to minimize sulfur oxidation.32 For similar preparation of alkanethiols on Au(111), scanning tunneling microscopy33 demonstrated that a hexagonal close-packed crystal lattice with a (x3×x3)R30° arrangement is formed. The spacing of the alkane chains is 4.97 Å as determined by low-energy electron diffraction.34 Note that, unlike the HSAM, the quality of the COOH-SAM is harder to control; however, formation of good-quality SAMs is achieved when acetic acid is added to ethanol in the wash solution, leading to efficient removal of multilayers by disruption of interplane hydrogen bonds between carboxyl groups.31 Substance P (SP), Gramicidin S (GS), and bradykinin (BK) were purchased from Sigma-Aldrich (St. Louis, MO) and used as received. Samples are dissolved in methanol at a concentration of 1 mg/mL. The stock solution is further diluted to 0.1 mg/mL in a specific mixture solution of methanol/ water/acetic acid for electrospraying. The acidity of the solution is adjusted to optimize the ion signal; decreasing the pH enhances higher charge-state protonation of the peptide in the electrospray process. TOF-SIMS Experiments. Samples prepared with the new apparatus were characterized ex situ using a PHI TRIFT II TOFSIMS instrument (Physical Electronics, Eden Prairie, MN) in the Environmental Molecular Sciences Laboratory (EMSL). Briefly, in TOF-SIMS, desorption of material is induced by 15-keV Ga+ bombardment of the surface. Secondary particles ejected from the target surface include atomic, fragment, and molecular ions and clusters thereof. The secondary ion yield is a complex and sample(29) Thevuthasan, S.; Baer, D. R.; Englehard, M. H.; Liang, Y.; Worthington, J. N.; Howard, T. R.; Munn, J. R.; Rounds, K. S. J. Vac. Sci. Technol., B 1995, 13 (4), 1900-1905. (30) Chidsey, C.; Liu, G.; Rowntree, P.; Scoles, G. J. Chem. Phys. 1989, 91, 4421. (31) Wang, H.; Chen, S.; Li, L.; Jiang, S. Langmuir 2005, 21, 2633. (32) Willey, T. M.; Vance, A. L.; van Buuren, T.; Bostedt, C.; Terminello, L. J.; Fadley, C. S. Surf. Sci. 2005, 576 (1-3), 188-196. (33) Widrig, C. A.; Alves, C. A.; Porter, M. D. J. Am. Chem. Soc. 1991, 113 (8), 2805-2810. (34) Strong, L.; Whitesides, G. M. Langmuir 1988, 4, 546.

Analytical Chemistry, Vol. 79, No. 17, September 1, 2007

6569

Figure 3. Spot size characterization. (a) Photograph of the softlanding instrument phosphorus screen detector showing the cross section of ∼3-mm diameter of a beam of doubly protonated GS. (b) TOF-SIMS line profiles of the deposited spot plotted as the integral peak signal of GS on the COOH-SAM surface along both the X and Y axes (solid circles). The solid line is the Gaussian fit showing a spot size of ∼3 mm at the base and a fwhm of 0.8 mm.

specific function of energy deposition in the surface layers of the material and is determined by the projectile parameters and physical processes of energy dissipation on the surface.35 This TOF analyzer incorporates three electrostatic analyzers that fully compensate for energy dispersion of the secondary ions resulting in mass resolution of ∼4000 at nominal mass 1000. ToF-SIMS spectra were acquired for 5-10 min using a 15-keV pulsed gallium ion beam (500 pA, 5-ns pulse width, 10-kHz repetition rate) focused onto the 100 × 100 µm2 area. The primary ion dose of 1011 ions/ cm2 used in these experiments is well below the static limit (10121013 ions/cm2) at which detectable surface ablation occurs.36 It should be noted that it takes ∼10 min to remove the sample from the soft-landing apparatus UHV and introduce it into the TOFSIMS vacuum system. For SP soft landed on the FSAM surface, we previously demonstrated that such exposure to laboratory air reduces the total peptide signal in SIMS by 30-40%.11 IRRAS Experiments. IRRAS measurements were performed on a Bruker IFS 66V/S spectrometer operating at ∼6 mbar. A mercury-cadmium-telluride detector was used to collect spectra with a resolution of 4 cm-1. The angle of incidence was 80° from the surface normal. Before every soft landing, the unmodified surface was used as background reference. For both sample and reference, 512 scans were collected. RESULTS AND DISCUSSION Spot Characterization. Shown in Figure 3a is a photograph of the phosphorus screen detector display illustrating the optical image of an ion beam of ∼3-mm cross section. This image is compared with the spot profile (Figure 3b) obtained from a horizontal (X) and vertical (Y) TOF-SIMS scan through the peptide spot produced by soft landing of the ion beam shown in Figure 3a. Figure 3b shows the integral values of the singly protonated peptide peak as a function of the position on the peptide spot. We refer to this type of TOF-SIMS scan as a “line profile” of a soft-landed spot. For this example, we deduce from Figure 3b that the peptide is located within a symmetrical spot that is wellfitted by a Gaussian curve with fwhm of 0.8 mm and diameter of 3 mm at the baseline. This result is typical of repeated ion deposition experiments showing a close match between the actual (35) Demirev, P. A. Mass Spectrom. Rev. 1995, 14, 279. (36) Benninghoven, A. Z. Phys. B 1970, 230, 403-417.

6570 Analytical Chemistry, Vol. 79, No. 17, September 1, 2007

Figure 4. (a) Section of a blank mass spectrum of the COOH-SAM and the HSAM surfaces showing similar fragment distributions. (b) A zoom in the 43-45 amu area showing a clear difference between both COOH-SAM (red) and HSAM (blue).

size of the deposited spot and the cross section and shape of the optical image used for tuning the ion beam. Selective Deposition of Peptides on SAM Surfaces. Selective deposition of peptides on the hydrophobic HSAM and hydrophilic COOH-SAM surfaces in the UHV chamber was performed using a mixture of three model peptides: SP, GS, and BK. Mass-selected beams of doubly protonated peptide ions of 3-pA current each generated by electrospray ionization of a mixture of the three peptides were soft landed in three different spots on each of the two target surfaces. Figure 4a shows TOFSIMS spectra of the HSAM and COOH-SAM surfaces prior to ion deposition. Both spectra are very similar but exhibit characteristic peaks at low mass that are emphasized in Figure 4b where the mass spectrum from m/z 43 to 45 is displayed for the two SAM surfaces. One can think of the structure of alkanethiol-based SAMs as consisting of three partssnamely, the thiol group (SH), which covalently binds to the two-dimensional Au crystal by loss of a hydrogen atom, a spacer group (CH2)n defining the length of the molecule, and finally the terminal group responsible for modified surface reactivity toward the soft-landed ionic peptide. This simple picture rationalizes the low-mass peaks in Figure 4b as consisting of alkyl ion peaks common to both SAMs and several oxygencontaining fragment peaks that differentiate the carboxyl end group from the methyl group. Specifically the three common peaks are observed at 43.05 (C3H7+), 44.06 (C3H8+), and 44.98 amu (CHS+), and the remainder are oxygen-containing groups at 43.02 (C2H3O+), 43.96 (CO2+), 44.02 (C2H4O+), 44.99 (COOH+), and 45.03 amu (C2H5O+) that can only originate from the COOHterminated SAM. TOF-SIMS spectra obtained following peptide deposition (nominal collision energy of 40 eV) on COOH-SAM (left side spectra) and HSAM (on the right) surfaces are shown in Figure 5. The exposure time was 20 min for each of the six spots, resulting in deposition of 0.02 ng of mass-selected peptide ions onto a 2.5-mm-diameter spot, corresponding to an average coverage of 0.6% of a monolayer for this experiment. At this extremely low surface coverage, peptide-peptide interactions are virtually nonexistent and we may assume that the TOF-SIMS signal

Figure 5. TOF-SIMS spectra obtained following selective deposition (7 amu mass window) of three doubly protonated peptides from a mixture solution in three well-separated spots on the COOH-SAM (left panels) and HSAM (right panels) surfaces.

response is linear with ion dose.10,37-40 The spectra correspond to the maximum peptide signal found at the center of each softlanded spot. In addition to common SAM substrate peaks, characteristic peptide peaks dominated by singly protonated, [M + H]+, peptide ions, are observed in all spectra. The spectra demonstrate that there is no cross-contamination between the spatially separated regions of ion deposition. For all three peptides, deposition on the COOH-SAM surface results in higher TOF-SIMS signals. However, it should be noted that the total signal in the background spectrum of the COOH-SAM surface is ∼1.7 times higher than the SIMS signal obtained from the HSAM surface. It follows that signal increase on the COOH-SAM surface can be in part attributed to higher yield of secondary ion formation on this surface. The reviewer suggested verifying this finding by examining spectra of authentic peptides deposited from solution in similar amounts. It should be noted that it is extremely difficult, if at all possible, to prepare a uniform film of peptides on the HSAM surface using this approach. While the solvent evaporates from the surface, the molecules clump up, resulting in formation of poorly characterized peptide films. For this reason, our attempts to perform the comparison between TOF-SIMS spectra obtained following soft landing and TOF-SIMS spectra of the samples deposited from solutions were not successful. Soft landing presents a new approach for preparation of well-defined uniform films on a variety of substrates. Note that almost exclusively singly protonated parent ions are observed following soft landing of doubly protonated peptide ions on SAM surfaces. Our recent study showed that only a small amount of doubly protonated ions, [M + 2H]2+, is observed using ex situ analysis of surfaces following soft landing of peptide ions.41 Specifically, we found that the [M + 2H]2+/[M + H]+ ratio is ∼0.01 for the COOH-SAM surface and ∼5 × 10-4 for the HSAM (37) Muddiman, D. C.; Gusev, A. I.; Hercules, D. M. Mass Spectrom. Rev. 1995, 14, 383. (38) Chatterjee, R.; Riederer, D. E.; Postawa, Z.; Winograd, N. J. Phys. Chem. B 1998, 102 (21), 4176-4182. (39) Schnieders, A.; Mollers, R.; Benninghoven, A. Surf. Sci. 2001, 471 (1-3), 170-184. (40) Ruschenschmidt, K.; Schnieders, A.; Benninghoven, A.; Arlinghaus, H. F. Surf. Sci. 2003, 526 (3), 351-355.

surface. Because of insufficient signal intensity, no [M + 2H]2+ ion was observed in TOF-SIMS spectra shown in Figure 5 except for GS on the COOH-SAM surface, for which the ratio of the doubly protonated to the singly protonated peptide signal is 0.01. We also demonstrated that only a small fraction of ions survive exposure of the HSAM surface to the laboratory air, while complete neutralization occurs following SL on the COOH-SAM surface.41 Other peptide-related peaks observed in TOF-SIMS spectra include peptide cationized on gold, [M + Au]+, loss of ammonia from the protonated peptide, [M - NH3]+, and several characteristic peptide fragments. Fragmentation observed in TOF-SIMS analysis of soft-landed peptides is typically dominated by the formation of low-mass immonium ions. Abundant immonium ions of proline (P, m/z 70.067) and phenylalanine (F, m/z 120.078) were observed for all three peptides. Monitoring of Peptide Deposition by TOF-SIMS and IRRAS. The two ex situ methods used for characterization of surfaces following peptide deposition experiments are strongly complementary and will be presented together. TOF-SIMS is much more sensitive overall and, as already demonstrated, is capable of mapping both spatial location and chemical composition of deposited peptides. However, the secondary ion signal is a complex function of projectile and target characteristics; it also exhibits very strong deviations from linearity with surface coverage that will be described in more detail below. In contrast, the much less sensitive IRRAS technique is relatively free from surface coverage artifacts. Figures 6 and 7 illustrate results from our study of the dose dependence of the signal for soft-landed doubly protonated SL ([SP + 2H]2+) on the two SAMs at impact energy of 40 eV. As discussed in the Experimental Section, these experiments were performed by placing the surface right after the resolving quadrupole at a pressure of 4 × 10-5 Torr. (Note that the bending quadrupole was not used in these experiments.) Line profiles of the soft-landed spots corresponding to deposition of 0.1, 0.3, and 0.6 ng of peptide are shown in Figure 6a. The absolute values of (41) Laskin, J.; Wang, P.; Hadjar, O.; Futrell, J. H.; Alvarez, J.; Cooks, R. G. Int. J. Mass Spectrom. 2007, 265, 237-243.

Analytical Chemistry, Vol. 79, No. 17, September 1, 2007

6571

Figure 6. TOF-SIMS signal response to different doses of 40-eV (SP + 2H)2+ soft landed on the COOH-SAM and HSAM surfaces: (a) line profiles of three spots corresponding to deposition of 0.1, 0.3, and 0.6 ng of mass-selected SP ion on each surface; (b) total peptide signal as a function of the dose (see text).

TOF-SIMS signals were normalized to the background signal intensity to minimize run-to-run variability. Figure 6b shows the dependence of the TOF-SIMS signal integrated over the spot profile on the ion dose. As noted previously, signals from peptides soft landed on the COOH-SAM are somewhat higher than for the methyl-SAM surface. The upper scale of Figure 6b shows the surface coverage realized in these experiments, assuming uniform coverage of spots 2 mm in diameter. For this low-dose deposition regime (below 25% surface coverage), TOF-SIMS peptide signal obtained for both surfaces is a linear function of the ion dose. This result is in agreement with previous TOF-SIMS investigations that demonstrated linear increase in secondary ion yield at surface coverage below 25%37 and our recent data showing linear increase in SIMS peptide signal following soft landing on the FSAM surface at similar coverage.10 Figure 7 illustrates normalized results from our experiments at significantly higher doses (0.9, 4.5, and 17 ng) for the same peptide deposited on both COOH-SAM and HSAM surfaces. Figure 7a shows the line profiles obtained as a function of dose. The qualitative result is similar for both surfaces. The spot corresponding to the lowest dose of 0.9 ng of peptide exhibits the nice Gaussian shape described earlier and shown in Figure 3. At the higher dose of 4.5 ng, corresponding to ∼75% of a monolayer surface coverage, a pronounced dip in the signal is observed in the line profile. This effect is even more obvious at the highest dose investigated. Here the signal drops essentially 6572 Analytical Chemistry, Vol. 79, No. 17, September 1, 2007

to zero at the center of the deposited spot. These effects, similar for both SAM surfaces, are especially pronounced for the COOHSAM, presumably reflecting higher efficiency of capture of the peptide by the highly polar terminal group. We note that the strong suppression of the TOF-SIMS signal is observed not only for the peptide signal but for all sputtered ions, including those characteristic of the SAM substrate. Figure 7b shows our IRRAS results for the two higher dose experiments just described. At the lower dose of 0.9 ng of peptide, the IR signal was too small for quantitative interpretation. For the two higher doses, amide I and amide II bands of the peptides are clearly observed on both surfaces. The highest IRRAS signal is observed for 17 ng of the peptide soft landed on the COOH-SAM, corresponding to an average surface coverage of ∼2 monolayers. At this level of coverage, the TOF-SIMS spectrum is essentially zero throughout the entire mass range, including substrate ions as well as peptide ions. This is evidently characteristic of formation of a relatively thick layer of peptide molecules that effectively suppresses sputtering and secondary ion formation. In this high coverage region, the IRRAS technique provides a true measure of the amount of peptide deposition achieved by soft landing. Figure 7b shows a quantitative increase of the amide band intensities, confirming the linear dependence of accumulated peptide with dose. However, while the IRRAS signal is proportional to the concentration of deposited species, it cannot be used for quantitation of the amount of soft-landed molecules because the signal intensity is determined by surface selection rulessonly those vibrational modes that give rise to an oscillating dipole perpendicular to the surface contribute to the absorption. It follows that the IRRAS signal depends on the orientation of molecules on the surface, which makes quantification very difficult. Figure 8 shows the full TOF-SIMS spectra for three different areas on the SP peptide spot. The first area (area I) represents the spectrum taken outside the peptide spot associated with the background. The second area (area II) represents the spectrum with maximum peptide signal taken a couple millimeters inside the peptide spot. The third area (area III) represents the spectrum taken at the spot center associated with the dent shown in Figure 7. Moving from area II to area III suppresses the peptide signal by a factor of 360 while the most dominant SP fragments F (m/z ) 120) and P (m/z ) 70) are only suppressed by a factor of 7 and 5, respectively. The estimated coverage around the center of the peptide spot is ∼4-6 monolayers. Our IRRAS data demonstrate that the accumulation of peptides on the HSAM and COOH-SAM surfaces is linear with dose, and unusual results obtained in our TOF-SIMS experiments reflect the complex and not fully understood matrix effects on the formation of secondary peptide ions during bombardment of surfaces with fast projectiles. Several studies36-40,42-44 have demonstrated that SIMS spectral intensities depend on the substrate (metal conductor, semiconductor, insulator), deposited neutral species, film thickness, nature of the bombarding ion, beam energy, and flux. Ions and neutrals are ejected from the surface and fluxes of both neutrals and ions are dependent on the (42) Delcorte, A.; Medard, N.; Bertrand, P. Anal. Chem. 2002, 74 (19), 49554968. (43) Delcorte, A.; Garrison, B. J. J. Phys. Chem. B 2003, 107 (10), 2297-2310. (44) Delcorte, A.; Garrison, B. J. J. Phys. Chem. B 2004, 108 (40), 15652-15661.

Figure 7. High-dose. 40-eV deposition of doubly protonated substance P, (SP + 2H)2+, on the COOH-SAM and HSAM surfaces. (a) TOFSIMS line profiles showing three areas: I (background) II (maximum peptide signal), and III (peptide spot center). Shown also is a strong signal suppression at the center of the spot with increase in the dose for both surfaces. (b) IRRAS spectra of the same surfaces showing and increase of the peptide signal (amides I and II) with the increase in the dose.

Figure 8. (a) TOF-SIMS spectra probing different areas of the COOH-SAM surface characterized in Figure 7a. Area I is the background spectrum taken few millimeters away from the peptide spot. Area II is spectrum showing the maximum peptide signal. Area III is the spectrum taken at the center of the spot showing strong signal suppression for surface and peptide-related peaks. Insets are high-mass region multiplied by a factor of 8. (b) Zoom around the two major peptide fragments through the three regions (see text for detailed discussion).

parameters noted. Detailed studies focused on understanding of matrix effects in SIMS have been reported for a number of small molecules adsorbed on different substrates.38-40 It has been demonstrated that suppression of secondary ion signal can be attributed to the reduced sputtering efficiency (sputter-induced matrix effect) and to the decrease in the ionization probability with increase in surface coverage (ionization matrix effect).39 However, we are not aware of similar reports utilizing uniform peptide films of varying coverage on surfaces. Consequently, it is not currently possible to present a clear interpretation of our TOF-

SIMS results. The situation is especially complicated by secondary charge-transfer reactions that may occur in the plume of species ejected by ion impact where chemical properties of species in the plume strongly affect the ion distribution. Similarly, the constituents of the plume are influenced by the composition of both the substrate and the surface layer. Fortunately the low coverage results are unambiguous as are the high coverage IRRAS data; many studies have shown linear dependence of the SIMS ion signal on surface coverage for coverage below 0.5 monolayer. For example, Muddiman et al.37 Analytical Chemistry, Vol. 79, No. 17, September 1, 2007

6573

presented a careful study of the effect of the film thickness of polystyrene on metal surfaces on the SIMS response and defined three different regionssbelow 0.25 ML, the signal was linear, then showed saturation effect up to a coverage of 0.5 ML, and finally decreased almost to zero at coverage above 0.8 ML. The strong increase of our secondary ion signal as one moves from the unmodified surface (area I in Figure 8) up the edge of the deposited spot corresponds to the linear region, while the decline corresponds to region 3 in that study. Related observations have been reported by Benninghoven et al.36,39 and Winograd et al.,38 who found linear and nonlinear yield regions for both sputtered ions and postionized neutrals. The decline of the secondary ion signal of [M + H]+ ions to zero observed in this study is clearly unexpected and likely represents a change in both the effective substrate and secondary reactions as ions are ejected from the surface. Clearly, peptide layer is a poor matrix for the formation of secondary ions. It is noteworthy that low-mass peptide fragments are suppressed by a factor of only 5-7 while the [M + H]+ signal is reduced to near zero. Two effects could contribute to the lower suppression of the fragment ion signal. Bombardment of a relatively thick peptide layer with 15-keV Ga+ beam may contribute to efficient fragmentation of neutral peptide molecules upon direct collision followed by ionization of the resulting fragments. Alternatively, secondary charge-transfer reactions could be affected by the surface coverage. In this case, the increase in the relative abundance of fragment ions at high coverage could be attributed to more exothermic charge-transfer reactions resulting in ionization of sputtered peptide molecules followed by fast fragmentation. CONCLUSION An ion deposition apparatus for selective soft landing of biomolecular ions onto well-characterized SAM surfaces has been described, and its capabilities have been illustrated. Two modes of operation of the instrument were evaluated. In the first mode, the surface is positioned in the UHV chamber maintained at 2 × 10-9 Torr. Good mass resolution and spatial resolution for relatively high beam currents (up to 60 pA) were achieved simultaneously. Visualization of the ion beam prior to the deposition enables precise control of the size and position of the ion beam, which allows us to deposit up to four well-resolved 1-3mm spots on each of the two different surfaces. This capability is important for quantitative investigation of the effect different

6574

Analytical Chemistry, Vol. 79, No. 17, September 1, 2007

factors on the soft landing efficiency The kinetic energy of projectile ions is controlled by the potential difference between the dc offset of the collisional quadrupole and the surface and can be varied in the range from 2 to 200 V per charge. In the second mode, which enables fast deposition of mass-selected ions over a larger area (5-8 mm spot), the surface is positioned right after the mass-resolving quadrupole maintained at 4 × 10-5 Torr. High ion currents (up to 500 pA) and large spot sizes obtained in this mode were essential for relatively fast preparation of surfaces for characterization using IRRAS. Ex situ characterization of surfaces was performed using TOFSIMS and IRRAS. In this study, we compared soft landing of several model peptides on the hydrophobic (HSAM) and hydrophilic (COOH-SAM) surfaces at different deposition doses. Although TOF-SIMS is a very sensitive technique that provides important information on the spatial distribution of deposited ions and is capable of detecting very low doses of soft-landed ions, severe matrix effects prohibit its utilization for characterization of surfaces with high peptide coverage. Soft landing of peptide ions on surfaces can be utilized for controlled preparation of peptide films of known coverage for fundamental studies of matrix effects in SIMS of large nonvolatile molecules. We demonstrated that IRRAS is an invaluable technique for chemical characterization of surfaces following high-dose deposition experiments. Future experiments will utilize both techniques for detailed studies of physical and chemical processes associated with deposition of large complex ions on semiconductive surfaces. ACKNOWLEDGMENT This work was performed at the W. R. Wiley Environmental Molecular Sciences Laboratory (EMSL), a national scientific user facility sponsored by the U.S. Department of Energy’s Office of Biological and Environmental Research and located at Pacific Northwest National Laboratory (PNNL). PNNL is operated by Battelle for the U.S. Department of Energy. Research at EMSL was supported by the Laboratory Directed Research and Development Program at PNNL and the grant from the Chemical Sciences Division, Office of Basic Energy Sciences of the U.S. Department of Energy.

Received for review March 26, 2007. Accepted June 25, 2007. AC070600H