Anal. Chem. 1994,66, 3733-3736
Use of Condensation Figures To Image Low-Energy Ion Beam Damage of Monolayer Films Thomas E. Kane, Vincent J. Angeiico, and Vicki H. Wysocki’ Department of Chemistry, 100 1 West Main Street, Box 84-2006, Richmond, Virginia 23284-2006
Monolayer films are bombarded by low-energy (10-100 eV) ion beams, and condensation figures (CFs), or breath figures, are used to image the damage. The monolayer films are prepared on gold using long-chain alkanethiols (R(CHz)SH), where R is the polar w-terminal group, OH (n = 11) or [N(CH&+Br] (n = 20). After bombardment, the films are removed from vacuum, and clean water vapor is allowed to condense onto the monolayer films, forming droplet patterns. Differences in water droplet size, shape, and populationdensity per unit area are observed between the damaged and undamaged monolayer regions, due to differences in surface polarity between the undamaged polar thiolates and those that have lost the o-terminal group via ion-surface collisions. The final droplet pattern illustrates the dimensionsand shape of the lowenergy ion beam. The ion beam damage site observed is a well-defined, cross-shape 8 mm X 6 mm, which is smaller than the dimensions of the surface. In order to improve the focusing of the ion beam, a leaky dielectric ELFS tube lens is added to the lens assembly prior to the surface, at the entrance of the first quadrupole (a second quadrupoleis positioned after the surface to collect and analyze the scattered ion flux). With the ELFS lens in place, the resulting cross shape is thinner, and the horizontal portion of the cross is greatly diminished, as expected. Magnification of the damaged apolar spot region in both cases yields sphericalwater droplets, smaller in diameter than those observed for the undamaged polar regions. Investigations employing an array of small droplets (micrometer diameter) have been applied in the imaging of large (1.5 cm X 1.5 cm) area surfaces.l-’O The resulting image patterns have been hi~torically’-~ called breath figures, as they were first obtained by simply cooling the back of a surface with ice and fogging the front with one’s breath. Condensation figures (CFs) employ the same technique, but the surface temperature is regulated and an inert gas is bubbled through clean water (or some other liquid) and passed over the surface at a regulated flow rate. For C F experiments employing polyethylene surfaces, mean droplet diameters ranging from 4 to 200 pm have been cited,6and depend on the time allotted (1) Rayleigh Nature 1911, 86, 416. (2) Aitken, J. Nuture 1912, 90, 516. (3) Rayleigh Nufure 1912, 90, 436. (4) Beysens, D.; Knobler, C. M. Pfiys. Reo. Lei. 1986, 57, 1433. (5) Mcakin, P.; Family, F. J. Pfiys. A 1989, 22, L225. ( 6 ) Briscoc, B. J.; Galvin, K. P. J. Pfiys. D 1990, 23, 422. (7) Fritter, D.; Knobler, C. M.; Beysens, D. A. Pfiys. Rev. A 1991,43, 2858. (8) Steyer, A.; Guenoun, P.; Beysens, D.; Knoblcr, C. M. Pfiys. Reo. A 199144, 827 1. (9) L6pe2, G.P.; Biebuyck, H. A.; Frisbie, C. D.; Whitesides, G.M. Science 1993, 260, 647. (10) Kumar, A.; Whitesides, G . M. Science 1994, 263, 60.
0003-2700/94/0366-3733$04.50/0 0 1994 American Chemical Society
for vapor condensation and the degree of nearest-neighbor droplet coalescence. Wetting experiments involving a single large droplet, 2-5 mm in diameter, have been performed by Whitesided l-14 in an attempt to employ a macroscopic system, Le., the measurement of the droplet contact angle, to describe the microscopic intermolecular relationships between self-assembled monolayer (SAM) films and the monolayer-liquid and liquid-air interfaces. Self-assembled monolayers are formed by the spontaneous chemical adsorption of organic molecules, such as alkanethiols, onto a solid substrate in a highly ordered fashion. One of several general trends observed for water droplets on monolayer films possessing polar o-terminal groups is an increase in the surface area of the droplet adsorbed to the surface (wetting). For a water droplet of a given volume, the result is an increase in the droplet diameter (increased wetting) with a larger contact angle (>9Oo).1l-14 In the case of condensation figures on a homogenous SAM film chemisorbed on gold (film prepared from a single type of alkanethiol, AuS(CH2),R), droplets of roughly the same volume and diameter are evenly distributed on the surface,gJo and the size and number of droplets per unit area depend on the polarity of the monolayer film and the length of time the vapor is allowed tocondense. Whitesides has recently showngJ0 how CFs may be employed to image films composed of more than one thiol type. A solution of one thiol having a polar (or nonpolar) w-terminal head group is drawn or printed onto a gold surface, and a thiol of opposite polarity is used to “fill in” the remaining surface space. The resulting monolayer surface is a mosaic of localized patches or discrete lines of different monolayer material, imaged with CFs by droplets of different size and population density. The damage of a homogenous monolayer film by collisions with low-energy (10-100 eV) ions has been noted by Wysocki and c o - w o r k e r ~who , ~ ~ used the monolayer film as a target for surface-induced dissociation (SID) experiments. In SID experiments, the impinging molecular ion fragments following collision with the surface, and structural information on the ion is obtained. l 6 Earlier investigations determined how the damage sustained by the monolayer during the ion-surface (11) Bain, C. D.; Whitesides, G. M. J. Am. Cfiem. SOC.1988, 110, 3665. (12) Bain,C. D.;Troughton, E. B.;Tao,Y.-T.;Evall, J.; Whitesides,G. M.;Nuzzo, R. G.J. Am. Cfiem. SOC.1989, 111, 321. (13) Whitesides, G.M.; Laibinis, P. E. Lnngmuir 1990, 6, 87. (14) Laibinis, P.E.; Whitesides, G. M.; Allara, D. L.; Tao, Y.-T.; Parikh, A. N.; Nuzzo, R. G.J. Am. Cfiem. SOC.1991, 113. 7152. (15) Kane, T. E.; Somogyi, A.; Wysocki, V. H. Org. Muss Specrrom. 1993, 28,
1665.
(16) Cooks, R. G.;Ast, T.; Mabud, Md. A. Inr. J. Muss Specrrom. Ion Processes 1990, 100, 209.
Analytical Chemistty, Vol. 66, No. 21, November 1, 1994 3733
13.0mn
6
yL X
collisions affected the mass spectral data and how long the SID behavior of a given monolayer surface remains consistent." The degree of surface damage has b u n probed in these earlier experiments with a three-party method: a molecular ion probe is used to monitor the hydrocarbon present on the monolayer,and the results are then related to the relative degree of disorder (inherent or i n d u d ) on the monolayer film. Thequestionofexactly whereon thesurface thedamage takes place has not b u n answered. It is not known whether the ion beam produces a localized or diffuse region of damage or migrates across the surface due to changes in the surface electric field. A description of the instrument used in these experiments is given in ref 17. Two quadrupoles (01,42) are arranged inaWOfashion,with thesurfacepositionedat theintersection of the quadrupole axes. The surface is positioned 4S0 relative to the ion optical path of both quadrupoles. A complete method for monolayer preparation is given in ref 15. The surface dimensions are rectangular, 13 mm X 17 mm in size. They are held in the instrument with a ceramicsupport and a stainless steel frame (Kimball Physics, Wilton, NH) as illustrated in Figure 1. A +20 V potential is applied to the surface, and the ion-surface collision energy is achieved by floating the source to +W V. The Are+beam is produced at 8 X le7 Ton by bombardment with 70 eV electrons. The instrument base pressure is 2 X le7Ton. For this investigation, the surfaces are prepared using a single o-functionalized alkanethiol, resulting in a largely homogeneous and uniform monolayer film. The films examined in this investigation are produced from HSB ~ ] thiolates . (CH2)l lCOH or [ H S ( C H ~ ) X N ( C H ~ ) ~ +Both (17) W m . V. H.; Ding. J-M.; Jwr J. L;Gllahan. J. SOC.Ma~sSprctrom.1992 3.27.
3734 Am-/-,
H.;King. F.L.1.Am.
Vd. 66,No. 21, Mwwntnw 1, 1994
contain an apolar alkyl chain and a polar terminal group. X-ray photoelectron spectroscopy and droplet contact angle data suggest that for self-assembled monolayen of thiolates on gold. the polar head group extends out from the substrate in an ordered fashion.!* The surfaces are bombarded using a 70 eV Are+ beam. The ensuing damage disrupts the monolayer, presumably exposing the underlying nonpolar moieties,and the result isa regionon thesurface with exposcd nonpolar alkyl groups (see data below). Any change in the surface polarity where the ion beam strikes is observed as a change in the water droplet pattern. Depositionof clean water droplets on monolayer films is believed to be nondestructive, whichallows theuseofCFs toimage theSAM surfacesbefore and after any number of ion-surface collision experiments. Figure 2a represents the C F image observed for a spot produced when Are+ (m/z 40) is mass selected and allowed to collide into the AuS(CH2)''OH monolayer at 70 eV for 4 h. After removal of thesurface from vacuum, thecondmtion of water droplets onto the OH-terminated surface produces a pattern where the beam presumably has sputtered and disrupted the monolayer film. It is important to note that after +36 h of ion beam damage, no evidence of damage to the monolayer film is visibleprior tocondensing water droplets (not shown). From the image of Figure 2, it is immediately clear that the Are+ beam, to a large extent, is localized in a region which is smaller than the total available surface area. The degree to which light is reflected and scattered by the water droplets depends on the droplet size, curvature, and population density per unit area. The corona-like effect, or outline of light around the damaged region, is due to the light source of the microscope being positioned at an angle so the intermediate-sized droplets (thost adsorbed where both polar and apolar monolayer molecules exist) predominantly reflect the light to the camera. Although not visible in the figure, the droplets also act as prisms of different sizes, and the result isa seriesofdiscrete,coMxntric ringsof color observed around thedamaged area. Figure 2b isa SOX magnificationof Figure 2a at the edge of the damage region. The droplets appear spherical and smaller in the apolar damaged region (upper portion). The smaller size inherently allows for a greater number of droplets per unit area. Control surfaces which have been subjected to time in the instrument vacuum (2 days, not shown) but not used in any SID experiments show a uniform droplet pattern appropriate to their hydrophobicity/ hydrophilicity but do not show the kind of droplet pattern (Figure 2) associated with the monolayendisrupted by an ion beam. Mechanical damage associated with the Kimball Physics frame along the perimeter edge of the control surface (see Figure 2a) isobserved, however. In thecaseof monolayer surfaces prepared from thiols with no polar head group (e.g., HS( C H 2) I 7c H 3 or HS( C H 2) 2( CF2)fCF3) and sput tercd with a 70 eV argon beam for 6 h, the observed C F droplet pattern is uniform across the entire surface, with no visual indication of ion beam damage. This is the expected result since no change in the hydrophobicity is expected for these surfaces. The ion beam image in Figure 2a is clover-like, or crossshaped, with dimensions of 8 mm (xdirection) and 6 mm @direction). This is in accordance with the predicted shape
Flgure 2. CF images. (a, top) AuS(CH2),,0H monolayer film after damage by a 70 eV Are+beam for 4 h. (b, middle) Close-up of an edge of the damage region above, Illustrating the different droplet patterns. (c, bottom) [ A U S ( C H ~ ) ~ ~ N ( C H ~monolayer ) ~ + B ~ ] after being damaged for 4 h by a 70 eV Ar'+ beam with a leaky dielectric lens positioned at the entrance to Q1.
of an ion beam leaving a quadrupole assembly.18-20This shape may also be compared to those experimentally observed by Cooks and co-workers22and Todd and co-workers20,21with an ion beam imaging assembly which included postacceleration (la) Campana, J. E. Int. J. Moss Spectrom. Ion Phys. 1980, 33, 101. (19) Friedman, M. H.; Ycrgcy, A. L.; Campana, J. E. J. Phys. E 1982, 15, 53. (20) Short,R. T.; Grimm, C. C.; Todd, P. J. J. Am. Soc. Mass Spectrom. 1991, 2, 226. (21) Grimm, C. C.; Short, R. T.; Todd, P. J. J. Am SOC.Mass Spectrom. 1991, 2. 362.
lenses and a phosphor screen detector to plot the low-energy ion beam shape in a tandem instrument. The cross shape is produced by the ions entering the quadrupole off-axis, over a range of different forward angles. As the ions approach one of the four quad rods, they are deflected and head toward the opposite rod that has the same dc voltage polarity. A gross oscillatory motion ensues. When the ion beam image is projected onto a flat surface, the effect is observed as a band whose position indicates the direction of the ion beam oscillation motion. A cross is observed on the surface due the overlapping oscillating trajectories induced by the four rods (two pairs) in the quadrupole. Figure 2a is oriented as the surface would appear viewed from the source, through Q1. It is important to note that in the instrument setup the surface is positioned at 4 5 O relative to Q1 but was flat against a microscope stage when photographed. The location and minor distortion of the ion beam damage are attributed to the positioning of the surface in the instrument and the tuning of the lens assemblies both before and after the surface. In an attempt to maximize collection of the scattered ion signal from a 70 eV collision, the voltage on the first lens positioned immediately after the surface is biased 30 V below the surface voltage. In addition to increased collection of the surface scattered ions, the incoming ion beam appears to be "pulled" closer to 4 2 . Differences in the shape of the left vs right sides of the vertical portion of the cross may also be explained in this way. Note also that the cross-shaped ion beam was projected on a surface at 4 5 O , which may result in a "smearing" of the right side of the ion beam. The wide black regions (Figure 2a) with the bright circular shapes appearing along the extreme left and right sides of the surface are the result of mechanical damage to the monolayer incurred in the region with the Kimball Physics stainless steel frame, which is used to hold the surface in position (see Figure 1). A reasonable estimate of the primary beam damage area is obtained by using the hole imprints made by the frame as a gauge (hole diameter, 1.6 mm; hole center-to-center distance, 3.8 mm). Note that for collisions of Are+ at 70 eV for equivalent time duration, the [AuS( CH&N( CH3)3+Br-] monolayer surface produces a damage image comparable in size and shape (not shown) to that produced from the A u S ( C H ~ ) I ~ Osurface H (Figure 2a). The droplet pattern observed on the undamaged region of the [ A U S ( C H ~ ) ~ O N ( C H ~monolayer ) ~ + B ~ ] suggests a more polar surface than the AuS(CH2)110H surface, as the droplets appear virtually flat and are irregularly shaped and sized. The droplet pattern in the damage region is comparable to that observed for the [OH] damage region. Figure 2c represents the damage image obtained with the same experimental conditions (Ar'+, 70 eV, 4 h), for a [AuS(CH2)2oN(CH3)3+Br] monolayer surface when a leaky dielectric ELFS tube lens23is added to the lens stackassembly in the ion source, at the entrance to Q1. A major purpose of the ELFS lens in a quadrupole instrument is to reduce the energy distribution of the incoming ion beam and to minimize the fringing fields associated with theQ1 DCvoltage. Ideally, both features result in a narrow, cylindrical ion beam. The C F damage image observed with the leaky dielectric lens in (22) Bier, M. E.; Amy, J. W.; Cooks, R. G.; Syka, J. E. P.; Ccja, P.; Stafford, G. Int. J. Mass Spectrom. Ion Processes 1987, 77, 31. (23) Ketkar, S. N.; Fitc, W. L. Rev. Sci. Instrum. 1988, 59, 987.
Analytical Chemistry, Vol. 66, No. 21, November 1, 1994
3735
place is similar in shape to that obtained in Figure 2a; however, the cross shape is smaller and thinner, and the horizontal portion is greatly diminished. This would suggest that the incoming ion beam is of lower overall current, with a decreased number of off-axis ions entering the quadrupole. The presence of the vertical-shaped spot on the monolayer surface suggests that the effectiveness of this ELFS lens in this particular assembly is only partially achieved and that the ion beam packet is still entering the quadrupole off axis or off center. The ions are still directed toward one of the rod pairs of the quadrupole, and the resulting gross oscillatory motion appears as a vertical line on the surface. Condensation figure imaging of a low-energy ion beam can be compared with conventional imaging techniques, such as the phosphor screen assembly employed by Cooks et a1.22 and Todd et a1.20*2'While the phosphor screen has advantages such as real-time measurements without venting the instrument in order to acquire the signal, the CF image produced on the monolayer is a direct representation of the low-energy ion beam behavior within the instrumental system. Phosphor screens require an acceleration of low translational energy ions to keV energies in order to induce light emission. To achieve this, additional equipment must be inserted into the ion optical path, which to some extent may compromise the observed ion beam image. The C F image displayed on the monolayer is an uncompromised image of the low-energy beam within the normal electrical assembly. It is important to note that the beam employed in this experiment was specifically chosen as a destructive agent. Atomic argon, having no vibrational or rotational modes of freedom, transfers a greater portion of the translational energy into the monolayer molecules than does a large molecular ion. The ions commonly employed in SID investigations do not induce much sputtering, and in some cases (e.g., protonated peptides) no sputtering peaks are observed in the collisional spectra. A logical next step in this investigation is to employ
3730 Analytical Chemistry, Vol. 66, No. 21, November 1, 1994
C F images to determine relative degrees of damage to monolayer films associated with ion-surface collisions by molecular ions of different size or odd- vs even-electron configurations, etc. Other surface imaging techniques, specifically scanning electron microscopy, will be employed to map bromine concentrations on the monolayer in order to compare the results to the C F data. In summary, condensation figures have been shown to be useful in imaging the shape and position of low-energy ion beams upon collision with monolayer films. The polar head groups of the thiolate monolayer are sputtered with a 70 eV Ar'+ ion beam, disrupting the surface and exposing the underlying nonpolar alkyl chain. The patterns observed are a result of differences in droplet formation on surface regions of different polarity. From the droplet pattern it is determined that the ion beam is smaller than the area of the surface and predominantly localized in a clover-like shape 8 mm X 6 mm in size. Increased time of the ion beam bombardment results in a spot of increased size and decreased shape definition. This suggests that the position of the ion beam does not migrate to a great extent on the surface but the general area of the spot increases. C F images on monolayer surfaces represent a simple, inexpensivemethod to identify the shape and location of a low-energy ion beam and may be adapted to a variety of instruments without msjor changes to the electric lens assemblies.
ACKNOWLEDGMENT This work was supported by the National Science Foundation (Grant No. CHE-9224719). We thank Naotoshi Nakashima (Department of Applied Chemistry, Nagasaki University) for providing the HS(CH2)20N(CH3)3+Br- and HS(CH2)llOH compounds. Received for review June 29, 1994. Accepted July 7, 1994." Abstract published in Advance ACS Abstracts. September 1, 1994.