AVS1997 Session NS-ThM: Nanolithography

Thursday, October 23, 1997 8:20 AM in Room K

Thursday Morning

Time Period ThM Sessions | Abstract Timeline | Topic NS Sessions | Time Periods | Topics | AVS1997 Schedule

Start Invited? Item
8:20 AM NS-ThM-1 Moving Molecules with the STM
J.C. Dunphy (Lawrence Berkeley National Laboratory); M.K. Rose (Univ. of California, Berkeley & Lawrence Berkeley National Laboratory); S. Behler, D.F. Ogletree, M. Salmeron (Lawrence Berkeley National Laboratory)
Using a variable temperature (30-300 K) Scanning Tunneling Microscope, we have investigated the interaction between the tip and atoms or molecules adsorbed on a Pd(111) surface in ultrahigh vacuum. By sufficiently cooling the sample, thermal diffusion of the adsorbates was inhibited. STM tip cleanliness was verified by measuring I(Z) spectra over several angstroms vertical tip displacement. Under a range of tunneling conditions individual molecules of carbon monoxide, carbon, sulfur, and acetylene could be imaged without disturbing them. Modifying the tunneling conditions by either reducing the gap resistance or in some cases increasing the bias voltage caused the tip to sweep or drag the molecules over the surface. The "phase space" of tunneling conditions which induced motion was mapped with the goal of elucidating the mechanism leading to motion.
8:40 AM NS-ThM-2 "Resonant Conducting" in Nano-Pattering Hydrogen-Passivated Si(100) Surfaces by Atomic Force Microscopy
E.Z. Luo, I.H. Wilson, J.B. Xu, J.X. Ma (The Chinese University of Hong Kong)
By monitoring the electrical current image during conducting AFM nano-pattering on Si(100) surfaces, a phenomennon that we term "resonant conducting" was found. This was characterized by sharp peak at V=2.45 ±0.05 V in the averaged I-V plot. The electric current is irreversible and decays very rapidly after repeat scanning, which implies that the resonant current is associated with transient event composed of a hydrogen de-passivation and an oxygen passivation under specific electrical field. On the practical side, the "resonant conducting" provides a very sensitive method for visulization of H-passivation of silicon at nano-meter scale
9:00 AM NS-ThM-3 Electric Field and Emission Current Induced Surface Modification of Au in the Interfacial Force Microscope
T.M. Mayer, J.E. Houston, G.E. Franklin, T.A. Michalske (Sandia National Laboratories)
We discuss the role of electric field and field emission current in the modification of Au surfaces with a W probe using the interfacial force microscope in UHV. We measure both the interfacial force and the field emission current as a function of separation with a constant voltage (100 V) between tip and sample. We purposefully use large voltage and separation in order to avoid catastrophic contact. The current initially increases exponentially as the separation decreases. However, at about 46 nm separation, the current rises sharply as the surface begins to distort and rapidly close the gap. We also calculate the electric field, field emission current, and force at the sample as a function of voltage and distance, and present a model of the interaction. We show that electric fields and mechanical forces are too small for significant field evaporation, or significant elastic or plastic deformation of the surface. We propose that surface diffusion of Au atoms in a lateral electric field gradient leads to an unstable surface distortion, giving rise to formation of a mound under the tip. This will result in catastrophic mechanical contact with the tip if left unchecked. We further propose that current induced production of mobile adatoms is the source of the diffusing material. We discuss this work in relation to earlier studies using voltage pulses in the STM as a means of nanofabrication. This work was supported by US DOE contract DE-AC04-94AL85000. Sandia is a multiprogram laboratory operated by Sandia Corp., a Lockheed-Martin Company, for the US Department of Energy.
9:20 AM NS-ThM-4 Optimization of Self-Assembled Organosilane Films on Modified Si(100) Surfaces for Nanolithography Applications
F.K. Perkins, S.L. Brandow, C.S. Dulcey (Naval Research Laboratory)
Organosilane self-assembled monolayers (SAMs) may be employed as imaging layers for high resolution lithography. Patterned exposure creates a two-dimensional template of chemical reactivity which can be used for the selective attachment of various materials, including electroless Ni. Deposition of a thin layer of electroless Ni serves as an effective etch mask for pattern transfer by reactive ion etching.1 We have previously demonstrated lithographic patterning of several SAMs utilizing the low energy electrons from a vacuum STM. We have found that lithographic performance of the imaging layer is improved when the native oxide is removed prior to SAM deposition. Minimum feature sizes in SAM films (following metallization and etching) were reduced from 24 nm on native oxide covered substrates to 15 nm on treated substrates.2,3 In this work we examine the effect of surface treatments on oxide stripping and surface functionalization. Modified surfaces were used as substrates for SAM film deposition. Deposited SAM films were characterized by ellipsometry, XPS, and AFM, and compared to corresponding films grown on native oxide surfaces. Treatment conditions that result in the formation of a monolayer coverage of surface hydroxyl (-OH) groups, with minimal surface oxide, were determined. This yields optimum conditions for SAM film formation, processing repeatability, and STM lithography.


11. J.M. Calvert, in Organic Thin Films and Surfaces, Vol. 1, A. Ulman, Ed., Academic Press, Inc., New York (1994).
22. F.K. Perkins, E.A. Dobisz, S.L. Brandow, T.S. Koloski, J.M. Calvert, K.W. Rhee, J.E. Kosakowski, and C.R.K. Marrian, J. Vac. Sci. and Technol. B12, 3725 (1994).
33. F.K. Perkins, E.A. Dobisz, S.L. Brandow, J.M. Calvert, J.E. Kosakowski, and C.R.K. Marrian, Appl. Phys. Lett. 68, 550 (1996).

9:40 AM NS-ThM-5 AFM Data Storage Using a Rotating Disk: Tracking and Wear
B.D. Terris, H.J. Mamin, R.P. Ried, S.A. Rishton, D. Rugar (IBM Almaden Research Center)
One possible application of scanning probes is for high density read-only data storage. Such an application requires the means to write high resolution master disks, an inexpensive way to mass replicate the master, and a disk drive capable of reading the replica at sufficiently high data rates. Using electron beam lithography, we have written and etched data patterns in silicon oxide with feature sizes as small as 50 nm and a data density 100 times that of a CD-ROM. These features have been faithfully replicated by molding a polymer film using a photopolymerization process1. A timing-based track following servo is used to keep the tip on 100 nm data bits to within a standard deviation of 31 nm. Recently developed high frequency piezoresistive levers having in-plane tips2, under both load and tracking control, have been used to continuously read 200 nm bits for over 148 hours with only a minor decrease in tip resolution and no indication of sample wear.


1B. D. Terris, H. J. Mamin, M. E. Best, J. A. Logan, D. Rugar, and S. A. Rishton, Appl. Phys. Lett. 69, 4262 (1996).
2R. P. Ried, H. J. Mamin, B. D. Terris, L. S. Fan, and D. Rugar, Tech. Digest. Int. Conf. Solid-State Sensors and Actuators, Transducers '97 (1997).

10:00 AM NS-ThM-6 Nanolithography by Manipulation of Catalytic Metal Clusters using an Atomic Force Microscope
S.L. Brandow, W.J. Dressick, C.S. Dulcey (Naval Research Laboratory); T.S. Koloski (Integument Technologies); L.M. Shirey (Naval Research Laboratory); J. Schmidt (Centre National de la Recherche Scientifique, France); J.M. Calvert (Shipley Company)
The study of nanoparticles and small molecular clusters is currently an area of intense interest due to the unique electrical and optical properties of these materials. In this work we demonstrate the use of catalytically active nanoclusters as a novel material for atomic force microscope (AFM) nanolithography. Films were prepared from colloidal Au nanoparticles and giant Pd clusters. Lithographic patterns were generated by physically manipulating the nanoclusters into 2-dimensional patterns on silicon oxide and functionalized silicon surfaces using the AFM tip. Linewidth was found to depend on the force applied to the nanoparticles and the number of tip passes used to generate the pattern. Conditions were optimized to clear scanned areas using minimum applied force. Patterned films were used as templates for the selective deposition of electroless metal, which served as a robust plasma etch mask for pattern transfer. Patterned metallization was demonstrated for both the Au and Pd systems, giving minimum linewidths of ~ 35 nm following pattern transfer by reactive ion etching to depths of approximately 200 nm. While this work focuses on lithography applications for nanometer scale materials, the technique could be used to position active quantum sized particles in nanometer scale patterns for other applications as well.
10:20 AM NS-ThM-7 Novel Non-Contact AFM Based Approaches for the Direct Manipulation of Nanoscale 3D Objects on Surfaces
T.R. Ramachandran, A. Madhukar, B.E. Koel (University of Southern California)
We present some novel approaches towards the direct manipulation of nanoscale 3D objects on surfaces using the non-contact atomic force microscope (NC-AFM). Remarkably, a complete reversal of imaging contrast of nanoscale 3D features, from positive to negative, as a function of NC-AFM imaging conditions is experimentally observed in both in-situ ultra-high vacuum studies as well as ex-situ NC-AFM studies of nanoscale 3D features. A simple force-gradient model shows that the observed reversal of contrast is concomitant with a feedback instability, arising due to the nature of the interaction force curve between tip and sample, and leading eventually to physical tip-sample contact. In this paper, the transition from a non-contact regime to a regime of tip-sample contact is exploited to realize the twin advantages of non-contact imaging and subsequent physical manipulation of nanoscale 3D objects, for the specific case of nanoscale gold particles on a mica substrate in air at room temperature. We employ different protocols involving either an intentional, selective disabling of the NC-AFM feedback1 or a forced instability of the feedback during scanning. Our systematic studies indicate that the manipulation does not involve the dragging or pulling of the gold particles by the tip. Rather, the particles move as a result of being physically pushed by the tip due to the repulsive interactions between them and the tip. We observe that the success of the manipulation event and the distance moved by the particle is dependent on the imaging conditions used prior to the pushing of the particle. Also, the outcome of the manipulation event is seen to depend on the particular protocol used for manipulation. We propose that the observation of negative contrast NC-AFM images of nanoscale 3D objects is likely to be a universal phenomenon and that it might serve as a qualitative but useful reference point for NC-AFM based manipulation of a variety of nanoscale objects on various substrates. This work was supported by the Zohrab A. Kaprielian Technology Innovation Fund.


1C. Baur, B. C. Gazen, B. Koel, T. R. Ramachandran, A. A. G. Requicha, and L. Zini, J. Vac. Sci. Technol. B, 1997 (In Press).

10:40 AM NS-ThM-8 Ti/TiOx/Ti Single Electron Tunnel Junctions Fabricated using Si-based Inorganic Electron Beam Resist.
T. Wada, S. Gorwadkar, S. Haraichi, K. Ishii, H. Hiroshima, M. Komuro (Electrotechnical Laboratory, Japan)
We have developed an SiO2/c-Si inorganic electron beam(EB) resist process, in order to overcome the resolution limits of conventional organic resist, and successfully fabricated Ti/TiOx/Ti single electron tunnel junctions smaller than 15nm x 15nm using conventional shadow evaporation of metal. We have employed samples typically of 55nm-SiO2/118nm-c-Si/300nm-buried-SiO2/Si(100) derived from SOI substrates. C-Si becomes a support for a suspended SiO2 mask for shadow evaporation. Samples, which were delineated with a 50kV 5nmφ-EB with a dose of 3µC/cm, were developed by a solution of buffered HF. For undercut etching, we newly employed NMD-3 solution, which is commercially available as a photoresist developer from Tokyo Ohka Kogyo Co., Ltd., Japan. NMD-3 shows much less anisotropic etching nature against Si than TMAH/KOH and has very high selectivity for Si over SiO2. We could successfully fabricate the SiO2 suspended mask of minimum opening width of 12nm. The fabricated suspended mask is used for MIM junction fabrication process using double shadow deposition of Ti. The oxidation of firstly deposited Ti film was carried out at 170 degreeC with oxygen partial pressure of 100mTorr for about one hour. The surface roughness of deposited Ti film measured by AFM is about 4~6Å. The clear isolated Ti/TiOx/Ti double junctions with corresponding connecting lead pattern were obtained by employing inorganic resist lift-off process using NMD-3 solution. The fabricated Ti/TiOx/Ti junctions show clear tunneling property during preliminary measurement at RT.
11:00 AM NS-ThM-9 Nanoprecision Lithography
E.A. Dobisz, C.R.K. Marrian (Naval Research Laboratory)
This paper focuses on both the resolution and the critical dimension control in e-beam nanolithography. Several authors have reported 10-20 nm lines in polymeric resists over the past 15 years. However, these represent the best results. If nanolithography is to be practical for devices, critical dimensions must be defined with an accuracy of ± 10%, over a range of pattern densities, and a practical range of process latitude. In practice, this is difficult to produce, even at relatively large critical dimensions of 100 nm. In this work the components of e-beam exposure that limit resolution and precision are identified and guidelines are presented. The measured backscatter coefficient was 0.51, in agreement with the Monte Carlo code. The agreement of the measurements of the linespread functions (LSFs) was less ideal. The standard deviations of the forward scatter Gaussian widths were broadened by 17 nm and 25 nm, from that of the primary beam, for PMMA and SAL-601 respectively. For a 50 kV e-beam, the Monte Carlo code predicts a broadening of 5 nm. AFM measurements of latent and developed e-beam written images show that the broadening of the linespread function is due to both broader than modeled resist exposure and the development process. The AFM studies further demonstrate the resulting difficulty in defining 40 nm period gratings. A semi-empirical model is introduced in which a LSF is formed by convoluting a measured forward Gaussian with the Monte Carlo generated LSF (for secondary and backscattered electrons). The semi-empirical LSF is integrated to form the exposed patterns. The model accurately describes the small dose and process latitude for fabricating sub-100 nm features and the observed pattern formation more accurately than the Monte Carlo code. The model also shows that the reproducibility of defining fine features, even as large as 100 nm will be greatly aided by the use of an ultrafine (≤1-2 nm) e-beam probe and new ultrathin resist processes.
11:20 AM NS-ThM-10 Electron Scattering in Lithography and Transmission through Thin Films
C.R.K. Marrian, F.K. Perkins, R. Bass, M. Rebbert, I.P.I. Isaacson (Naval Research Laboratory)
Electron-solid scattering based Monte Carlo simulations are increasingly used for process optimization in electron beam lithography. Previously we reported on the importance of inelastic scattering as being key to understanding the limits of resolution and pattern distortion in electron beam nanolithography1. The form of the elastic scattering cross section in these simulations does not have a significant effect on the lithographic spread functions measured on low atomic number (silicon) substrates. This conclusion is confirmed by new studies of nanolithography on high atomic number (tungsten) substrates. However, the form of the elastic cross section does effect the profile of a beam that is transmitted through a thin film. We describe here a series of experiments where the profile of a beam of 50 keV electrons is measured after passage through thin (0.25 to 2 µm) films of low atomic number (silicon, aluminum) and high atomic number (gold, tantalum). The beam profiles were obtained by exposing a layer of electron beam sensitive resist on a substrate positioned immediately behind each film. The separation between thin film and resist varied between zero (where metal was evaporated and/or up plated onto the resist film) to ~25 µm (where the resist coated substrate was mounted directly behind a thin membrane). The width of features in the developed resist was measured as a function of incident electron dose to obtain the transmitted beam profile over three orders of magnitude in intensity. The results point to the importance of the use of the Mott cross section rather than the more usual (and computationally more tractable) screened Rutherford cross section in order to accurately describe the observed beam profiles.


1C.R.K. Marrian, F.K. Perkins, D. Park, E.A. Dobisz, M.C. Peckerar, K.-W. Rhee and R. Bass, J. Vac. Sci. Technol. B14, 3864 (1996)

11:40 AM NS-ThM-11 Arrays of Submicron Metal Rings Fabricated using NCG Replica Masks
D.H. Pearson, R.J. Tonucci, K. Bussmann (Naval Research Laboratory)
Nanochannel glass (NCG) replica masks are thin metallic films containing patterned arrays of uniform nanometer-scale voids whose sizes, positions, geometric patterns, and packing densities may be controlled to a high degree. These masks, which have been prepared from noble and refractory metals with void diameters as small as 40 nanometers and packing densities greater than 3 x 109 voids per square centimeter, are well suited for parallel patterning applications involving materials deposition and directional etching. Here we describe the fabrication of large arrays of submicron gold and nickel rings utilizing self-registered NCG replica masks in conjunction with thin film deposition, in-situ pattern-mask modification, and ion-milling. This work was supported by the Defense Advanced Research Project Agency's Microelectronics Technology Office and by the Office of Naval Research.
Time Period ThM Sessions | Abstract Timeline | Topic NS Sessions | Time Periods | Topics | AVS1997 Schedule