Imaging and Patterning of Monomolecular Resists by Zone-Plate

J. Phys. Chem. B 2003, 107, 13133-13142

13133

Imaging and Patterning of Monomolecular Resists by Zone-Plate-Focused X-ray Microprobe Ruth Klauser,* I.-H. Hong, and S.-C. Wang National Synchrotron Radiation Research Center, 101 Hsin-Ann Road, Hsinchu 300, Taiwan

M. Zharnikov,* A. Paul, and A. Go1 lzha1 user Angewandte Physikalische Chemie, UniVersita¨t Heidelberg, 69120 Heidelberg, Germany

A. Terfort Anorganische und Angewandte Chemie, UniVersita¨t Hamburg, 20146 Hamburg, Germany

T. J. Chuang Center for Condensed Matter Sciences, National Taiwan UniVersity, Taipei 106, Taiwan ReceiVed: June 11, 2003; In Final Form: September 8, 2003

Soft X-ray scanning photoelectron microscopy (SPEM) was applied to image and characterize molecular patterns produced by electron irradiation of various aliphatic and aromatic thiol-derived self-assembled monolayers (SAMs). The observed chemical contrasts allowed us to monitor complex phenomena which occurred as a result of electron-beam patterning, the exposure of the patterned films to ambient, and the irradiation of the films by the X-ray microprobe during image acquisition. The latter effect has been analyzed in detail and utilized for direct lithographic writing in the SAM resists by the zone-plate-focused X-ray beam. The results demonstrate the capabilities of the SPEM technique both for chemical imaging and as a fabrication tool for micro- and nanolithography.

1. Introduction The development of novel approaches for the fabrication of micro- and nanostructures and, in particular, chemical and biological patterns is an important technological and scientific issue. One of the perspective methods is electron-beam patterning of chemisorbed monomolecular films, self-assembled monolayers (SAMs), which are well-ordered 2D-assembles of long-chain molecules attached to a suitable substrate.1,2 A flexible molecular architecture of the SAM constituents allows to use a wide range of substrates and tailor adsorption and wetting phenomena at the SAM-ambient interface, whereas the molecular size of these constituents makes SAMs an ideal platform for the fabrication of micro- and nanostructures. For the latter purpose, different approaches have been developed, including microcontact printing,3,4 scanning microprobe printing,5,6 and electron-beam patterning.7-16 Typical sizes of SAMpatterns produced by microcontact printing,3,4 micromachining,17 and microwriting18 are 0.1-100 µm. At the same time, the patterning with electron beam,16 atomic force microscope (AFM), and scanning tunneling microscope (STM) tips5,6 was successfully performed on nanometer length scale. Among these approaches, electron-beam lithography has attracted special attention over the past decade because of the flexibility of this technique and the improved understanding of the phenomena occurring at irradiation of ultrathin organic films with ionizing radiation. The interaction of electron19-30 and X-ray31-34 beams with different SAMs has been reported in numerous papers and * Corresponding authors. R. Klauser e-mail: [email protected]; M. Zharnikov e-mail: [email protected].

reviewed recently by Zharnikov and Grunze.35 The response of SAMs to low-energy electrons (10-300 eV) and soft X-rays (100-1500 eV) is similar, because the irradiation-induced modification in the latter case is predominately caused by photoand secondary electrons.35 In general, two competitive processes take place, namely, irradiation-induced decomposition and quasipolymerization. The balance of these processes is determined by the composition of the SAM constituent and the molecular packing in the monolayer. In particular, it has been shown that the aliphatic thiol-derived SAMs respond differently toward lowenergy electron exposure than the respective aromatic films.28,35 The former systems become heavily damaged under the electron beam with loss of the orientational and conformational order, partial dehydrogenation with CdC double bond formation, desorption of films fragments, and the transformation of the pristine thiolate species to dialkylsulfides. At the same time, the electron-induced damage is greatly reduced in aromatic films, where the irradiation induces cross-linking between the neighboring aromatic moieties, which prevents both an extensive damage of the headgroup-substrate interface and the complete loss of the orientational order. The structural integrity of the irradiated aromatic SAMs opens the possibility for an exclusive modification of the terminal functional groups. As a first example of this approach, the electron-induced transformation of the functional nitro groups in nitro-biphenylthiolate SAMs to amino moieties has been realized.29 The respective chemical patterning was performed on both micrometer and nanometer length scales, and the fabricated patterns have been successively

10.1021/jp0307396 CCC: $25.00 © 2003 American Chemical Society Published on Web 10/30/2003

13134 J. Phys. Chem. B, Vol. 107, No. 47, 2003 used for laterally selective attachment of organic compounds and biological macromolecules.29,36 At present, the imaging of the SAM-based lithographic patterns has only been carried out by scanning electron microscopy (SEM) and AFM. Considering that these methods do not provide any chemical information, it is desirable to implement an imaging technique based on the chemical identity of the SAM constituents and to perform a direct spectromicroscopic characterization of the patterned SAMs. This can be provided by soft X-ray scanning photoelectron microscopy (SPEM),37,38 as will be demonstrated in this paper. In the first part of our study, we applied SPEM to image and characterize molecular patterns produced by electron irradiation of aliphatic and aromatic thiol-derived SAMs through masks (proximity printing). In the second part, we explored the modification of SAMs by the zone-plate-focused X-ray beam and utilized it for direct lithographic writing to give a first example of SAM-based soft X-ray lithography. Although the diameter of the zone-platefocused X-ray beam (∼100 nm) in our current experimental setup is larger than the diameter of a well-focused electron beam (less than 10 nm), a reduced scattering of X-rays in the sample may result in a better or comparable spatial resolution. A successful example of a similar approach has been recently shown by the ELETTRA-SPEM group, who created submicron fluorescent patterns of color centers in LiF films with a microprobe.39 Naturally, the spatial resolution of SPEM can be substantially improved in the future when ultrafine zone plates become available. In the present study, we performed in situ patterning of three different monomolecular resists by the microbeam with subsequent imaging of the fabricated patterns. Proximity printing in polymers using “conventional” X-ray lithography has already demonstrated features of less than 100 nm.40,41 The setup could be therefore applied for writing into the SAM resists as well. Our approach, however, aims in particular in the combination of direct writing and chemical imaging to provide a basic understanding of the patterning process. 2. Experimental Section The substrates were prepared by deposition of gold films with 100-300 nm thickness on titanium-primed polished singlecrystal Si(100) wafers. These films predominately exhibit a (111) orientation, as implied by the presence of the corresponding forward-focusing peaks in the polar-angle intensity distribution of the Au 4f electrons42 and by the observation of the characteristic binding energy (BE) shift of the Au 4f surface component in high-resolution X-ray photoemission spectra.43 The SAMs were formed by immersion of the substrates in 1 mM solutions of the respective substances in ethanol or dichloromethane for 24 h with subsequent rinsing and drying. Three different kinds of SAMs were used in this investigation, namely, the SAMs formed from alkanethiols (AT), semifluorinated alkanethiols (SFAT), and aromatic thiols (ART). The AT SAM of this study was hexadecanethiolate or C16 (CH3(CH2)15SH) while SFAT and ART SAMs were CF3(CF2)9(CH2)11SH (F10H11) and nitrobiphenylthiolate (NBPT: NO2(C6H4)2SH), respectively. C16 was purchased from Fluka; details of the synthesis of F10H11 and NBPT can be found elsewhere.29,44,45 The patterning of the SAMs was performed by electrons with energy of 300 eV through masks with mesh sizes of #1500 and #2000 (proximity printing). The doses were 10 000 µC/cm2 and 30 000 µC/cm2 for the C16/F10H11 and NBPT films, respectively. The patterning was carried out in ultrahigh vacuum (UHV). The resulting patterns were exposed to ambient before

Klauser et al. the SPEM characterization, which was performed in a separate UHV chamber. Some of the NBPT patterns were acylated by immersion in a solution of trifluoroacetic acid anhydride (TFAA). This modification results in the selective attachment of trifluoroacyl groups to the electron-beam-converted amino groups.29 The imaging and characterization of the SAM-patterns were performed with a UHV SPEM system at the microscopy branch of the U5 undulator beamline of NSRRC in Hsinchu, Taiwan (see refs 38, 46, 47 for details). The samples were mounted on a piezo-driven flexure stage of the manipulator. The focusing of the photon beam to submicron spot size was achieved by soft X-ray optics, on the basis of Fresnel amplitude zone plate with a diffraction-limited resolution of 92 nm and an efficiency maximum for a photon energy of about 350 eV. The entire system has a spatial resolution of ∼200 nm in the relevant photon energy range. The photoelectrons emitted from the illuminated spot were collected by a hemispherical electron energy analyzer with a 16-channel detection scheme. Photoelectron images of the patterned surfaces were obtained by scanning the sample with respect to the photon beam and recording a 16-channel spectrum at every sample position. This resulted in the simultaneous acquisition of 16 SPEM images within the energy range covered by the channeltrons (∼12 eV for standard settings of the analyzer), which also allows to derive a 16-point spectrum for every image pixel. In addition to these “pixel spectra”, conventional micro-XPS spectra were taken from selected areas of the samples with both focused and unfocused (without zone plate and order-sorting aperture) beams. The respective photon flux density was calculated from the beam size and electron current from a phosphor screen illuminated by the photon beam. C 1s, Au 4f, N 1s, and F 1s SPEM images and microspectra of patterned SAMs were recorded with primary photon energies of 385-390 eV (C 1s and Au 4f), 630 eV (N 1s), and 870 eV (F 1s). Along with the spectromicroscopy and microspectroscopy experiments, a patterning of nonstructured samples by the focused X-ray beam has been performed. The same experimental settings were applied for both imaging and patterning, with the only differences in pixel size and dwell time for the pixel acquisition. 3. Results and Discussion 3.1. Electron-Beam Patterned SAMs. Alkanethiolates on gold substrate are the most studied SAMs and often used as a test system. Figure 1 shows Au 4f7/2 and C 1s SPEM images of an e-beam patterned C16/Au sample. The Au 4f7/2 image is presented as measured while the C 1s image is the result of the subtraction of two images, namely, the image at the kinetic energy of the C 1s emission and the image at a higher kinetic energy corresponding to inelastic background. Below the images in Figure 1, the corresponding intensity profiles taken across the solid lines are given. The electron-induced pattern with a mesh square opening of 7.5 µm and a mesh stripe width of 5.5 µm can be clearly distinguished in both Au 4f7/2 and C 1s images, which exhibit inverse contrasts. The Au photoemission signal from nonirradiated areas has higher intensity than that for the irradiated areas, whereas the inverse intensity relation is observed for the C 1s signal. For both Au 4f7/2 and C 1s images, the intensity profiles across the border between the irradiated and nonirradiated area are not very sharp and correspond to a lateral resolution of ∼1.5 µm. Far from the border, the intensity difference between the nonirradiated and irradiated areas is ∼27% and ∼24% for the

Imaging and Patterning of Monomolecular Resists

J. Phys. Chem. B, Vol. 107, No. 47, 2003 13135

Figure 1. Au 4f7/2 and C 1s SPEM images of electron-beam patterned C16/Au taken at a photon energy (PE) of 388 eV. The image size is 24 × 24 µm2 (120 pixel × 120 pixel). The dwell time for the pixel acquisition was 60 ms. Bottom panels: intensity profiles taken along the solid lines as indicated in the images.

Figure 2. A sketch to illustrate the procedure of electron-beam patterning and contrasts observed in the SPEM images before and after air exposure of the sample.

Au 4f7/2 and C 1s image, respectively. The 24% contrast in the latter image was obtained after the subtraction of a background, as mentioned above. The nonprocessed C 1s image has a much weaker contrast because of the contribution from an intense inelastic background, which predominately originates from the Au 4f electrons. The contrast in the “background” image mimics the contrast of the Au 4f image and is opposite to the C 1s contrast, as shown by us in a preliminary study of electronbeam-patterned C12/Au and C18/Au samples.38 An alternative to the subtraction of a background image is the acquisition of the carbon KLL Auger image,38 which is, however, difficult

because of a short focal length of the zone plate at the relevant photon energies (around 300 eV). The contrast observed in the Au 4f7/2 and C 1s images is in a contradiction with the results of in situ spectroscopic measurements, which revealed a reduction of SAM thickness because of irradiation-induced desorption of the SAM constituents and their fragments.25,35 The respective thickness reduction depends on the length of the alkyl chain and the electron dose and is about 30% (with respect to the pristine layer) for a dose of 10 000 µC/cm2 in the case of C16/Au.25,27 In accordance with these results, the electron-exposed areas (squares) of the

13136 J. Phys. Chem. B, Vol. 107, No. 47, 2003

Figure 3. C 1s HRXPS spectra of F10H11/Au taken at a PE of 386 eV for different X-ray exposure. The exposure was set by selection of the exit slit of the beamline (10 µm, 20 µm, and 100 µm, respectively) while the dwell time was kept fixed.

patterned SAMs should reveal a larger Au 4f and a smaller C 1s signals, as illustrated schematically in Figure 2. On the contrary, the opposite intensity relations are observed in Figure 1. This “disagreement” can be explained by the adsorption of airborne carbon-containing molecules on the irradiated areas during the air exposure, which presumably occurs parallel to the adsorption of oxygen-containing species.35 An enhanced absorption affinity of the irradiated areas is not surprising considering the fact that electron irradiation creates chemically active sites at the SAM-ambient interface and noticeably increases its roughness.22,35 A direct proof that the observed intensity contrast is caused by the adsorption of airborne molecules can be obtained by the investigation of the SFAT SAMs. The molecular constituents of these films are composed of fluorocarbon and hydrocarbon parts with the former part building the SAM-ambient interface.48,49 Figure 3 displays the C 1s photoemission spectra of a nonpatterned SFAT SAM, F10H11/Au taken at a photon energy of 386 eV with an unfocused X-ray beam. The bottom spectrum collected at a low photon flux exhibits three emissions at 284.1, 290.6, and 292.8 eV corresponding to the CH2, CF2, and CF3 moieties, respectively. The emission from the hydrocarbon chain is much smaller as compared to that for the fluorocarbon part because of a strong suppression of the respective signal by the fluorocarbon overlayer; the mean free path for the C 1s photoelectrons at a given kinetic energy (∼100 eV) is only about 4 Å. The spectrum drastically changed when the layer is exposed to a higher photon flux, as, for example, provided by opening the exit slit of the beamline from 10 to 20 µm (middle spectrum in Figure 3): The intensities of the emissions related to the fluorocarbon part dramatically decrease, whereas a new emission at 287.3 eV appears, the signal at 284.1 eV increases, and all peaks become broader. Whereas the broadening is associated with film disordering, the intensity changes originate from an extensive desorption of fluorocarbon species and fluorine, which results in a partial transformation of residual CF2 and CF3 entities to C-F moieties (a new peak at 287.3 eV) and to C-C/ CdC species (an additional signal at the position of the CH2 emission).30 For an even higher X-ray exposure at 100 µm slit

Klauser et al. opening (top spectrum in Figure 3), almost all fluorinecontaining species are desorbed, and the spectrum is dominated by the broad peak from disordered saturated and nonsaturated hydrocarbons. Electron irradiation of SFAT SAMs results in similar changes of the XPS spectra,30 so that the spectra given in Figure 3 can be considered as representative for the pristine and irradiated areas of patterned SFATs. The respective Au 4f and C 1s SPEM images are shown in Figure 4. The contrast and intensity profile of the Au 4f image are similar to those of the patterned C16/ Au (see Figure 1): The nonirradiated areas exhibit a higher (by 28%) Au 4f signal than the areas exposed to electrons. The respective C 1s images acquired at the positions of the CH2 and CF2 emissions (see Figure 3) reveal opposite contrasts. Whereas the hydrocarbon-related image has the same contrast as the C 1s image for C16/Au (a stronger emission from irradiated parts), the fluorocarbon-related image exhibits the inverse contrast in agreement with the reduction of the CF2related emissions after electron irradiation (see Figure 3). As there are no fluorocarbon species in the ambient, no respective absorption occurred upon the air exposure, and hence there is no contrast inversion in the “CF2” image. At the same time, the adsorption of airborne hydrocarbon species results in a further increase in intensity of the already intense CH2/C-C/ CdC emission, so that the contrast in the respective image (middle panel in Figure 4) is even stronger than that of C16/ Au (the C 1s image in Figure 1) even without any background subtraction. Along with the possibility to modify the entire SAM, there is an option to modify only the functional groups, keeping the structural integrity of the SAM within a framework of chemical lithography. In NBPT/Au used for this study, an irradiationinduced transformation of the nitro functional groups to amino entities takes place, as shown by previous XPS measurements, where the simultaneous reduction in the intensity of N 1s emission at 400.0 eV (NO2) and increase in the intensity of N 1s emission at 406.0 eV (NH2) was observed.29 At the same time, the respective C 1s and S 2p spectra exhibited only slight changes, because of the cross-linking of the phenyl groups, which stabilizes the film and reduces the extent of irradiationinduced desorption to 10-15%.35 It was therefore expected that the C 1s and Au 4f SPEM images of patterned NBPT should show a weak contrast, whereas there should be a pronounced contrast in the N 1s images corresponding to the emissions from NO2 and NH2 groups. However, the Au 4f7/2 and C 1s images of electron-beam-patterned NBPT/Au in Figure 5 exhibit similar contrast patterns as aliphatic SAMs (see Figure 1). The Au signal shows a comparable decrease in intensity (28%) at going from nonirradiated to irradiated areas, and only the contrast in the C 1s image is noticeably weaker than that for the aliphatic SAMs, even though the background subtraction procedure was applied. Apart from the weaker contrast, it is obvious that despite the reduced damage in NBPT/Au, the modification induced by the electron beam is already sufficient to cause the adsorption of carbon-containing species from ambient onto the irradiated areas (even if to a smaller extent as compared to the aliphatic SAMs). Presumably, the same factors, namely, an enhanced surface roughness and the appearance of chemically active sites are responsible for this phenomenon. Whereas the electron-beam-induced patterns could be clearly recognized in the Au 4f and C 1s images of NBPT/Au, no contrast due to the nitro-amino transformation could be identified in the N 1s imagessthe only observed contrast was provided by the inelastic background dominated by the gold



Figure 4. Au 4f7/2, C 1s (CH2), and C 1s (CF2) SPEM images of electron-beam patterned F10H11/Au taken at a PE of 387 eV. The image size is 40 × 40 µm2. The C 1s (CH2) image also contains contributions from the irradiation-induced C-C and CdC moieties, which have a similar binding energy as CH2 groups. Bottom panels: intensity profiles taken along the lines as indicated in the images.

Figure 5. Au 4f7/2 and C 1s SPEM images of electron-beam-patterned NBPT/Au taken at a PE of 387 eV (image size 24 × 24 µm2). Bottom panels: intensity profiles taken along the lines as indicated in the images.

signal as previously shown by us.50 The reason for the failure of the chemical imaging in this particular case is the low intensity of the N 1s emissions compared to the inelastic

background, as displayed in the left panel of Figure 6. One has also to consider that the dwell time for the pixel acquisition (imaging) is noticeably smaller than the dwell time of the spectra


Klauser et al.

Figure 6. Left: Survey photoemission spectra of an electron-beam-patterned NBPT/ Au sample taken at a PE of 630 eV with focused and unfocused (without zone-plate and order-sorting aperture) X-ray beam. Right: Survey photoemission spectra from patterned and TFAA-treated NBPT/Au taken at a PE of 870 eV with focused and unfocused X-ray beam.

Figure 7. A series of N 1s HRXPS spectra of electron-beam-patterned NBPT/Au taken sequentially at a PE of 630 eV with a defocused X-ray beam. The beam spot illuminated both irradiated and nonirradiated (by electrons) areas. The duration of X-ray exposure is marked at the respective spectra.

in Figure 6, which suggests that the N 1s emissions, small but clearly perceptible in the spectra, provide at the imaging a contribution which cannot be distinguished from the noise level.

Similar results have been obtained for the TFAA-modified NBPT-patterns. Whereas the Au 4f and C 1s images (not shown) exhibited contrasts similar to those of the C16 and NBPT patterns, perceptible contrast could be observed neither in the N 1s nor F 1s spectral ranges representing the nitro-amino transformation and the TFAA attachment, respectively. The reasons for the lack of contrast are once more the small relative intensities of the characteristic N 1s and F 1s emissions, as shown in the wide-scan spectra in the right panel of Figure 6. At first sight, the problem of low-intensity emissions can be easily overcome by improving the signal-to-noise ratio in the relevant spectral ranges, which might lead to a perceptible contrast after the subtraction of a background image, just in the same way as it was done for the C 1s images of C16/Au and NBPT/Au. However, the signal-to-noise ratio can only be improved by an increase of the dwell time for the pixel acquisition, which is limited by an X-ray induced damage of the SAM patterns by the imaging microprobe. Whereas the extent of such damage was rather small for the experimental settings used to acquire the images in Figures 1, 4, and 5, it will become larger as soon as the dwell time increases. An example is provided by the wide scan microspectra in Figure 6, which were acquired under a prolonged (as compared to the SPEM imaging) exposure of the same area. The intensities of the characteristic N 1s and F 1s emissions in the spectra collected with an unfocused X-ray beam (without ZP/OSA) are noticeably larger than those in the spectra acquired with a focused beam (with ZP/OSA), which suggests a modification of the nitro and TFAA moieties by the X-ray microbeam during the spectra acquisition. 3.2. X-ray Beam Patterned SAMs. Whereas the effect of the X-ray microprobe may be disturbing in SPEM-based microspectroscopy and spectromicroscopy, it can be utilized for a tailored modification of a SAM resist. While it is well known that X-rays induce radiation damage to organic films upon prolonged exposure time, this effect should be even more



Figure 8. Au 4f7/2 and C 1s SPEM images (image size 70 × 70 µm2) of soft X-ray microbeam patterned C16/Au taken at a PE of 387 eV. See text for further details. Bottom panels: intensity profiles taken along the lines as indicated in the images. A stripe of damaged area at the right end of the images stems from a longer (some milliseconds) stay of the focused beam at the end of each line scan.

dramatic for a focused beam.49 Calibration measurements for our present SPEM setup gave X-ray induced currents of 0.082 nA and 2.6 nA for the focused and unfocused beams, respectively. The corresponding areas of the beam spot at the sample position were about 0.03 µm2 and 2 × 105 µm2, respectively. This results in a 105 larger photon flux density for the focused beam, which is crucial for taking micro-photoemission spectra, which requires an X-ray exposure of 1-2 min. At the same time, X-ray induced damage is presumably of minor importance for the imaging mode because of a small dwell time for the pixel acquisition, which was generally kept at 40-100 ms. To monitor the effect of X-ray irradiation on SAMs and, particularly, on the terminal groups of the NBPT/Au samples, we have recorded sequential photoemission spectra with the unfocused beam in the course of X-ray exposure. Figure 7 depicts a series of N 1s spectra of an electron-beam-patterned NBPT/Au; the beam spot illuminated both electron-beammodified and pristine areas, so that the signals from both amino and nitro species could be expected. In fact, the first spectrum, which took 90 s to be completed, clearly shows two peaks at 406.0 and 399.8 eV assigned to the nitro and partially protonated amino species. Whereas the first scan spectrum exhibits comparable intensities for the nitro- and amino-related emissions, the intensity of the former peak drastically decreases already in the second scan spectrum. After 21 min of X-ray exposure, the nitro peak was not no longer perceptible in the spectrum and the emission related to the amino functional group became the only spectral feature. This peak shifted slightly to 399.4 eV (as compared to 399.8 eV in the first scan spectrum), which indicates a larger amount of fully reduced species, but does not exhibit any further change in position or intensity up to an exposure time of 2 h.

A similar modification of the terminal group after X-ray exposure was observed for a TFAA-treated NBPT/Au sample. The spectra presented in Figure 7 once more demonstrate that SAMs can be modified even with an unfocused X-ray beam. This effect should be noticeably larger with a zone-plate-focused microbeam, which opens the possibility for in situ patterning of SAMs by SPEM. As in imaging, first experiments were performed with a “reference” C16/Au sample. Figure 8 displays an example of microbeam writing (letter “T”) and the successive imaging. The patterning was performed with a pixel width of 500 nm and a dwell time of 90 ms, while the imaging of the fabricated letter was carried out with a pixel width of 700 nm and a dwell time of 60 ms. The fabricated X-ray lithographic pattern shows a good contrast and better lateral resolution than the ex situ e-beam patterns in Figures 1, 4, and 5 fabricated in the proximity printing geometry. In agreement with the assumption on the impact of airborne molecules for the ex situ patterns, the in situ patterns, as in Figure 8, exhibit the expected contrasts, which are now only representative for the irradiationinduced desorption. An especially high-contrast patterning could be achieved in SFAT SAMs, for example, F10H11/Au, as displayed in Figure 9. The experimental settings for the patterning and imaging were the same as for the C16/Au sample. The strong C 1s signal related to the CH2/C-C/CdC entities in the X-ray-irradiated areas is clearly distinguished from the nonirradiated parts, which exhibit much smaller intensity of this signal, as illustrated by the HRXPS spectra in Figure 3. The impact of the X-ray microbeam should be the same as the one provided by electron irradiation, so that the top and bottom spectra in Figure 3 can be considered to be representative for the X-ray irradiated and pristine areas, respectively. The lateral resolution of the X-ray


Figure 9. Two-dimensional C 1s (CH2) SPEM images of soft X-ray microbeam patterned F10H11/Au taken at a PE of 387 eV (image size 70 × 70 µm2). Bottom panel: intensity profile taken along the line as indicated in the image. The “CH2” image also contains contributions from the irradiation induced C-C and CdC moieties, which have a similar C 1s binding energy as CH2 groups.

beam patterning taken from the edge of the intensity profile is 0.8 µm. The observed contrast is not a topographic but a chemical one. Even though the intensity contrasts of the C 1s (CH2) images of ex situ and in situ patterned F10H11/Au in Figures 4 and 9, respectively, are the same (a higher intensity from the irradiated areas), the topographic contrasts might be just the opposite. Whereas the thickness of the irradiated areas in the in situ patterns (X-rays) is reduced because of the desorption of C-F species, the respective thickness reduction in the ex situ patterns (electrons) is presumably overcompensated by the additional adsorption of airborne molecules during air exposure. This is confirmed by the contrast observed in the Au 4f images. Whereas the Au 4f image of the e-beam pattern in Figure 4 has the opposite contrast with respect to the C 1s (CH2/ C-C/CdC) image, the Au 4f image (not shown) of the X-ray pattern in Figure 9 exhibits the same contrast as the C 1s (CH2/ C-C/CdC) image. The patterns which can be produced by the focused X-ray beam of our current SPEM setup are limited to square or rectangular shapes. Also, the scanning electronics does presently not allow the writing of single pixel lines, whose width is determined just by the beam spot size. The realization of such

Klauser et al. a writing mode would certainly improve the performance and the spatial resolution of soft X-ray lithography with a SAM resist. Another example of soft X-ray lithography is given in Figure 10, where the Au 4f images of several different patterns written in NBPT/Au are shown. The dwell times for the writing and imaging of the “R” and “S” letters (left panels in Figure 10) were again 90 ms per pixel (pixel width 500 nm) and 60 ms per pixel (pixel width 700 nm), respectively. It is clearly seen that the contrast is not as strong as in the C16 and F10H11 resists (the respective C 1s contrast is even lower). This can be explained by a lower impact of the focused X-ray beam on the aromatic films as compared to the aliphatic ones, similar to the electron-beam irradiation. In particular, the extent of irradiationinduced desorption from the latter systems (∼30%)27 is noticeably larger than from the former ones (∼10%),35 which is of importance for the contrast in the SPEM images, as far as the SAM patterns were not exposed to ambient. In addition, multiexposure lithographic patterns were fabricated as, for example, displayed in the right part of Figure 10 by the example of NBPT/Au. First, a small 10 × 10 µm2 square was written with a pixel width of 200 nm and a dwell time of 160 ms (4 × 40 ms), followed by the writing of a larger 50 × 50 µm2 square with a pixel size of 500 nm and a dwell time of 60 ms, and the imaging of the fabricated pattern with a pixel width of 700 nm and a dwell time of 60 ms. The intensity profile shows two steps of surface modification. As expected, the 10µm inner square with the prolonged exposure time exhibits a stronger Au signal than other areas. The smaller pixel size and the longer dwell time result in a total X-ray exposure time which is about 9 times larger than that for the larger square. The respective increase in intensity is, however, only 22% while the intensity difference between the large square and the outer areas of the image is about 15%. The observed intensity relations correlate with a general evolution of the electron/X-ray induced modification of SAMs, which progresses rather fast at initial stages of irradiation and slows down at a prolonged exposure. 4. Summary and Conclusions SAM-based lithography is a promising approach to fabricate micro- and nanostructures, including chemical and biological ones. Taking advantage of the conscious design of a suitable resist, a practical implementation of this technique relies on a detailed knowledge of the patterning process and the direct control of the fabricated patterns. As demonstrated in the present paper, both these goals can be addressed by soft X-ray scanning photoelectron spectromicroscopy, which not only allows a chemical imaging of SAM-based lithographic patterns, but gains new insights into their reactivity and the course of the patterning process. Molecular patterns produced by electron irradiation of three different kinds of SAMs, namely, aliphatic, semifluorinated, and aromatic thiolates on Au substrates were imaged and characterized. The aliphatic SAMs become heavily damaged upon the electron-beam exposure, including a noticeable thickness reduction. However, because of an enhanced surface roughness and chemical activity of the irradiated areas, selective adsorption of the airborne molecules onto these areas occurs upon air exposure, so that their thickness becomes larger than that in the nonirradiated areas. This effect is clearly seen by the inverse contrasts of the Au 4f and C 1s SPEM images compared to the expectation on the basis of in situ thickness reduction. The effect of airborne adsorbates was important for the aromatic SAMs as well. The Au 4f and C 1s SPEM images of



Figure 10. Au 4f7/2 SPEM images (image size 70 × 70 µm2) of soft X-ray microbeam patterned NBPT/Au taken at a PE of 387 eV and intensity profile taken along the lines as indicated in the image of a multiexposure lithographic pattern (right upper panel).

the respective lithographic patterns exhibited similar contrasts as those of the aliphatic films. Even though the extent of the irradiation-induced desorption is much smaller for the aromatic films, the reactivity of the SAM-ambient interface was obviously high enough to initiate the adsorption of carbon-containing molecules from ambient. A better understanding of the properties of the patterned films was provided by the example of semifluorinated SAMs. For the respective lithographic patterns, a “true” chemical imaging of the different carbon species participating in C-H and C-F bonds was performed. The observed contrasts could be clearly correlated with the chemical and physical changes induced by the electron beam and subsequent exposure to ambient. An unequivocal distinction between the effects of air exposure and those intrinsic to the patterning was achieved. Whereas SPEM imaging was rather straightforward for intense emissions, a strong inelastic background and a low signal-to-noise ratio hindered the observation of a chemical contrast for low-intense photoemission peaks. Even though the signal-to-noise ratio can be partly improved by increase of the dwell time for the imaging, there are limitations to this increase because of X-ray induced damage by the SPEM microprobe. The latter effect is certainly a disturbing factor. On the other hand, a focused X-ray beam can be utilized for direct lithographic writing, as shown in this study by the example of in situ X-ray microbeam patterning of aliphatic, semifluorinated, and aromatic SAMs with subsequent imaging of the fabricated patterns. Those images generally showed a better lateral resolution and a sharper contrast pattern as compared to the ex situ e-beam patterns fabricated in the proximity printing geometry. This is presumably due to the limitations of proximity printing and “clean” in situ fabrication without the adsorption

from ambient, which can additionally blur the pattern. A particularly good contrast and lateral resolution have been achieved for the C 1s (C-H) image of F10H11/Au, where an intense desorption and decomposition of the fluorocarbon part occurred upon exposure to the microbeam, which causes a high C-H/C-C/CdC intensity from the irradiated areas. In the NBPT/Au, the contrast of microbeam patterns is noticeably weaker as compared to C16/Au and F10H11/Au. This originates in the fact that the only observable contrast is caused by the X-ray induced damage in irradiated areas. The degree of damage, however, is largely reduced in aromatic films, as discussed above. At present, lithographic writing in the SAM resists by the zone-plate-focused soft X-ray beam is still inferior to the focused electron beam. It can be used, however, as a suitable tool to study basic aspects of SAM patterning and to optimize the patterning process. Acknowledgment. The authors are very grateful to M. Grunze for his support. Also, the authors like to thank G.-C. Yin for providing the SPEM image analysis program and M.L. Huang for technical help. We gratefully acknowledge the support of the Ministry of Education of R.O.C. under grant number 89-N-FA01-2-4-5 and the BMBF of Germany under grant GRE1HD. References and Notes (1) Ulman, A. Chem. ReV. 1996, 96, 1533. (2) Thin films: self-assembled monolayers of thiols; Ulman, A., Ed.; Academic Press: San Diego, CA, 1998. (3) Kumar, A.; Whitesides, G. M. Science 1994, 263, 60. (4) Loo, Y.-L.; Willett, R. L.; Baldwin, K. W.; Rogers, J. A. J. Am. Chem. Soc. 2002, 124, 7654.

13142 J. Phys. Chem. B, Vol. 107, No. 47, 2003 (5) Xu, S.; Liu, G. Langmuir 1997, 13, 127. (6) Fuierer, R. R.; Carroll, R. L.; Feldheim, D. L.; Gorman, C. B. AdV. Mater. 2002, 14, 154. (7) Tiberio, R. C.; Craighead, H. G.; Lercel, M. J.; Lau, T.; Sheen, C. W.; Allara, D. L. Appl. Phys. Lett. 1993, 62, 476. (8) Lercel, M. J.; Tiberio, R. C.; Chapman, P. F.; Craighead, H. G.; Sheen, C. W.; Parikh, A. N.; Allara, D. L. J. Vac. Sci. Technol., B 1993, 11, 2823. (9) Lercel, M. J.; Redinbo, G. F.; Pardo, F. D.; Rooks, M.; Tiberio, R. C.; Simpson, P.; Craighead, H. G.; Sheen, C. W.; Parikh, A. N.; Allara, D. L. J. Vac. Sci. Technol., B 1994, 12, 3663. (10) Lercel, M. J.; Redinbo, G. F.; Rooks, M.; Tiberio, R. C.; Craighead, H. G.; Sheen, C. W.; Allara, D. L. Mircroelectron. Eng. 1995, 27, 43. (11) Lercel, M. J.; Rooks, M.; Tiberio, R. C.; Craighead, H. G.; Sheen, C. W.; Parikh, A. N.; Allara, D. L. J. Vac. Sci. Technol., B 1995, 13, 1139. (12) Mu¨ller, H. U.; David, C.; Vo¨lkel, B.; Grunze, M. J. Vac. Sci. Technol., B 1995, 13, 2846. (13) David, C.; Mu¨ller, H. U.; Vo¨lkel, B.; Grunze, M. Microelectron. Eng. 1996, 30, 57. (14) Lercel, M. J.; Craighead, H. G.; Parikh, A. N.; Seshadri, K.; Allara, D. L. Appl. Phys. Lett. 1996, 68, 1504. (15) Hild, R.; David, C.; Mu¨ller, H. U.; Vo¨lkel, B.; Kayser, D. R.; Grunze, M. Langmuir 1998, 14, 342. (16) Go¨lzhaüser, A.; Geyer, W.; Stadler, V.; Eck, W.; Grunze, M.; Edinger, K.; Weimann, Th.; Hinze, P. J. Vac. Sci. Technol., B 2000, 18, 3414. (17) Abbott, N. L.; Kumar, A.; Whitesides, G. M. Chem. Mater. 1994, 6, 596. (18) Kumar, A.; Abbott, N. L.; Kim, E.; Biebuyck, H.-A.; Whitesides, G. M. Acc. Chem. Res. 1995, 28, 219. (19) Baer, D. R.; Engelhard, M. H.; Schulte, D. W.; Guenther, D. E.; Wang, L.-Q.; Rieke, P. C. J. Vac. Sci. 1994, A12, 2478. (20) Rowntree, P.; Dugal, P. C.; Hunting, D.; Sanche, L. J. Phys. Chem. 1996, 100, 4546. (21) Seshadri, K.; Froyd, K.; Parikh, A. N.; Allara, D. L.; Lercel M. J.; Craighead, H. G. J. Phys. Chem. 1996, 100, 15900. (22) Mu¨ller, H. U.; Zharnikov, M.; Vo¨lkel, B.; Schertel, A.; Harder, P.; Grunze, M. J. Phys. Chem. B 1998, 102, 7949. (23) Olsen, C.; Rowntree, P. A. J. Chem. Phys. 1998, 108, 3750. (24) Zerulla, D.; Chasse, T. Langmuir 1999, 15, 5285. (25) Zharnikov, M.; Geyer, W.; Go¨lzhaüser, A.; Frey, S.; Grunze, M. Phys. Chem. Chem. Phys. 1999, 1, 3163. (26) Heister, K.; Frey, S.; Go¨lzhaüser, A.; Ulman, A.; Zharnikov, M. J. Phys. Chem. B 1999, 103, 11098. (27) Zharnikov, M.; Frey, S.; Heister, K.; Grunze, M. Langmuir 2000, 16, 2697. (28) Geyer, W.; Stadler, V.; Eck, W.; Zharnikov, M.; Go¨lzhaüser, A.; Grunze, M. Appl. Phys. Lett. 1999, 75, 2401.

Klauser et al. (29) Eck, W.; Stadler, V.; Geyer, W.; Zharnikov, M.; Go¨lzhaüser, A.; Grunze, M. AdV. Mater. 2000, 12, 805. (30) Frey, S.; Heister, K.; Zharnikov, M.; Grunze, M. Phys. Chem. Chem. Phys. 2000, 2, 1979; 2000, 2, 3721. (31) Laibinis, P. E.; Graham, R. L.; Biebuyck, H. A.; Whitesides, G. M. Science 1991, 254, 981. (32) Ja¨ger, B.; Schu¨rmann, H.; Mu¨ller, H. U.; Himmel, H. J.; Neumann, M.; Grunze, M.; Wo¨ll, Ch. Z. Phys. Chem. 1997, 202, 263. (33) Wirde, M.; Gelius, U.; Dunbar, T.; Allara, D. L. Nucl. Instrum. Methods Phys. Res., Sect. B 1997, 131, 245. (34) Heister, K.; Zharnikov, M.; Grunze, M.; Johannson, L. S. O.; Ulman, A. Langmuir 2001, 17, 8. (35) Zharnikov, M.; Grunze, M. J. Vac. Sci. Technol., B 2002, 20, 1793 and references therein. (36) Go¨lzhaüser, A.; Eck, W.; Geyer, W.; Stadler, V.; Weimann, T.; Hinze, P.; Grunze, M. AdV. Mater. 2001, 13, 806. (37) Kiskinova, M. Surf. Interface Anal. 2000, 30, 464. (38) Klauser, R.; Hong, I.-H.; Lee, T.-H.; Yin, G.-C.; Wei, D.-H.; Tsang K.-L.; Chuang, T. J.; Wang, S.-C.; Gwo, S.; Zharnikov, M.; Liao, J.-D. Surf. ReV. Lett. 2002, 9, 213. (39) Larciprete, R.; Gregoratti, L.; Danailov, M.; Montereali, R. M.; Bonfigli, I.; Kiskinova M. Appl. Phys. Lett. 2002, 80, 3862. (40) Forber, R. A.; Chen, Z. W.; Menon. R.; Grygier, R.; Mrowka, S., Turcu, I. C. E.; Gaeta. J.; Cassidy, K.; Smith, H. I. J. Vac. Sci. Technol. 2002, 20, 2984. (41) Toyota, E.; Hori, T.; Khan, M.; Cerrina, F. J. Vac. Sci. Technol. 2001, 19, 2428. (42) Ko¨hn, F. Diploma Thesis, Universita¨t Heidelberg, Heidelberg, Germany, 1998. (43) Heister, K.; Zharnikov, M.; Grunze, M.; Johansson, L. S. O. J. Phys. Chem. B 2001, 105, 4058. (44) Strong, L.; Whitesides, G. M. Langmuir 1988, 4, 546. (45) Graupe, M.; Koini, T.; Wang, V.; Nassif, G. M.; Colorado, R., Jr.; Villazana, R. J.; Dong, H.; Miura, Y. F.; Shmakova, O. E.; Lee, T. R. J. Fluorine Chem. 1998, 93, 107. (46) Hong, I.-H.; Lee, T.-H.; Yin, G.-C.; Wei, D.-H.; Juang, J.-M.; Dann, T.-E.; Klauser, R.; Chuang, T. J.; Chen C.-T.; Tsang K.-L. Nucl. Instrum. Methods, A 2001, 467/468, 905. (47) Ko, C.-H.; Klauser, R.; Wei, D.-H.; Chan H.-H.; Chuang, T. J. J. Synchrotron Rad. 1998, 5, 299. (48) Frey, S.; Heister, K.; Zharnikov, M.; Grunze, M.; Tamada, K.; Colorado, R., Jr.; Graupe, M.; Shmakova, O. E.; Lee, T. R. Isr. J. Chem. 2001, 40, 81. (49) Zharnikov, M.; Grunze, M. J. Phys. Condens. Matter 2001, 13, 11333. (50) Klauser, R.; Zharnikov, M.; Hong, I.-H.; Wang, S.-C.; Go¨lzhaüser, A.; Chuang, T. J. J. Phys. IV 2003, 104, 459.

Imaging and Patterning of Monomolecular Resists by Zone-Plate

Recommend Documents