Resist based neutral atom lithography with metastable 23S1 helium(He. ) has been used to produce small structures in both goldand palladium. A beam of He. from a reverse flow, DC dischargesource is collimated by the bichromatic optical force followed bythree optical molasses velocity compression stages. The He. beamthen crosses a region where a mechanical or optical mask createsthe desired pattern.In the first experiments, a self assembled monolayer (SAM) ofnonanethiol is grown on a gold coated silicon wafer and then exposedto the bright, collimated beam of He. after a Nickel micromesh mask. The mesh protects parts of the SAM from the 20 eVof internal energy deposited by the impact of a He. atom onto thesample surface. The pattern of the mesh is transferred into the~200 Å gold layer with a standard selective etch that removes theGold from under the damaged SAM molecules. The samples wereanalyzed with an Atomic Force Microscope and Scanning ElectronMicroscope.The second set of experiments were performed using an opticalmask to pattern a gold coated silicon wafer. In the optical mask,the He. atoms experience the dipole force while traversing a standinglight wave of &lambda1<\sub> = 1083 nm light tuned 490 MHz above the 23S1<\sub>→ 23P2 transition or, in separate experiments, &lambda2<\sub> = 389 nmlight tuned 80 MHz below the 23S1<\sub>→ 33P2 transition. Using theoptical masks, He. can be focused or channeled into parallel linesseparated by &lambda/2 by varying the intensity of the standing lightwave.The small structures created by the 389 nm optical mask beganapproaching the gold surface granularity. A ~200 Å layer of palladium on silicon was chosen as the palladium granularity is muchsmaller and the SAM of nonanethiol will still form on the surface.Experiments using a nickel micro mesh as a mechanical mask havedemonstrated similar features to those in gold.
Resist based neutral atom lithography with metastable 23S1 helium(He. ) has been used to produce small structures in both goldand palladium. A beam of He. from a reverse flow, DC dischargesource is collimated by the bichromatic optical force followed bythree optical molasses velocity compression stages. The He. beamthen crosses a region where a mechanical or optical mask createsthe desired pattern.In the first experiments, a self assembled monolayer (SAM) ofnonanethiol is grown on a gold coated silicon wafer and then exposedto the bright, collimated beam of He. after a Nickel micromesh mask. The mesh protects parts of the SAM from the 20 eVof internal energy deposited by the impact of a He. atom onto thesample surface. The pattern of the mesh is transferred into the~200 Å gold layer with a standard selective etch that removes theGold from under the damaged SAM molecules. The samples wereanalyzed with an Atomic Force Microscope and Scanning ElectronMicroscope.The second set of experiments were performed using an opticalmask to pattern a gold coated silicon wafer. In the optical mask,the He. atoms experience the dipole force while traversing a standinglight wave of &lambda1<\sub> = 1083 nm light tuned 490 MHz above the 23S1<\sub>→ 23P2 transition or, in separate experiments, &lambda2<\sub> = 389 nmlight tuned 80 MHz below the 23S1<\sub>→ 33P2 transition. Using theoptical masks, He. can be focused or channeled into parallel linesseparated by &lambda/2 by varying the intensity of the standing lightwave.The small structures created by the 389 nm optical mask beganapproaching the gold surface granularity. A ~200 Å layer of palladium on silicon was chosen as the palladium granularity is muchsmaller and the SAM of nonanethiol will still form on the surface.Experiments using a nickel micro mesh as a mechanical mask havedemonstrated similar features to those in gold.
A bright metastable helium (He∗) beam is collimated sequentially with the bichromatic force and three optical molasses velocity compression stages. Each He∗ atom in the beam has 20 eV of internal energy that can destroy a molecular resist assembled on a gold coated silicon wafer. Patterns in the resist are imprinted onto the gold layer with a standard selective etch. Patterning of the wafer with the He∗ was demonstrated with two methods. First, a mesh was used to protect parts of the wafer making an array of grid lines. Second, a standing wave of λ=1083 nm light was used to channel and focus the He∗ atoms into lines separated by λ/2. The patterns were measured with an atomic force microscope establishing an edge resolution of 80 nm. Our results are reliable and repeatable.
Optical forces using lasers allow precise control over the motion of atoms. The bichromatic optical force (BF) is unique in its large magnitude and velocity range, arising from the absorption and stimulated emission processes. These properties were used to transversely collimate a beam of metastable helium (He∗ ) using the 2^{3}S − 2^{3}P transition. The collimation created a very bright beam of He∗ , allowing experiments of neutral atom lithography. The He∗ beam was used to pattern material surfaces using a resist-based lithography technique, where the pattern was determined by either mechanical or optical masks. The optical masks produced features with a separation of half the wavelength of the light used. Patterning was successfully demonstrated with both IR and UV optical masks. The etched pattern resolution was ∼ 100 nm and limited by the material surface. Further experiments were performed studying the ability of the bichromatic force to cool. The finite velocity range of the BF allows estimation of a characteristic cooling time which is independent of the excited state lifetime, only depending on the atomic mass and optical transition energy. Past experiments, including the helium collimation used for neutral atom lithography, have demonstrated that the BF can collimate and longitudinally slow atomic beams, but required long interaction times that included many spontaneous emission (SE) events. The effect of SE can be mitigated, and is predicted to not be necessary for BF cooling. Since the BF cooling time is not related to the excited state lifetime, a transition can be chosen such that the cooling time is shorter than the SE cycle time, allowing direct laser cooling on atoms and molecules that do not have cycling transitions. Experiments using the helium 2^{3}S− 3^{3}P transition were chosen because the BF cooling time (285 ns) is on the order of the average SE cycle time (260 ns). Numerical simulations of the experimental system were run predicting compression of the atomic velocity distribution. Our experimental results demonstrate the stimulated nature of the force through many atomic recoils from the light in less than one SE cycle time. A large range of the atomic velocity distribution is accelerated and accumulates at the velocity limit of the force. This accumulation results in an increase in the velocity space density, demonstrating cooling. | 112 pages
A high flux metastable Neon atomic beam was designed and characterised. Atom optical enhancement of the beam using a two-dimensional optical collimator, novel two beam Zeeman slower and hexapole magnetic guide was performed to produce a UV-free metastable flux of (4.35 x 109 ± 4 x 107) atoms s−1. Investigations of several resists for neutral atom lithography were undertaken. A quantitative investigation of the wetting properties of ethanethiol (ET) and dodecanethiol (DDT) self-assembled monolayers (SAM's) exposed to various metastable dosages was carried out. A mechanism for the poor lithographic patterning using ET was proposed and the negative contrast patterning observed for this SAM was similar to those observed for bare gold substrates and were attributed to mechanical pump oil (MPO) contamination resists. Negative patterned resists were used to produce 7.5 μm iron dots on a silicon substrate via neutral atom lithography. This scheme was found to be very robust and free from the laser cooling issues of alternative direct depositional schemes. Numerical simulations have shown that two dimensional arrays of magnetic nanodots may be produced this way, paving the way for a magnetic storage option with a density of " 0.1 Gbit mm−2.
In this thesis experiments are described, where a beam of metastable helium is used for atom optic interferometry experiments as well as for lithography. The broad potential of atom optics for fundamental as well as applied research could thus be used. In the first experiment described in this work, dealing with atom lithography, a lithographic resist technique was implemented, basing on the structural changes of single molecules of a self-assembling monolayer of dodecanethiol due to Penning processes induced by metastable helium atoms. This molecular layer could be laterally structured with respect to its hydrophobicity properties. In a chemical etching step, the structure in the local wetability could be transferred into an underlying gold layer, producing sharp structures there. It could be shown experimentally, that this change of the wetability is essential for the process. In the second experiment, the Talbot effect was investigated with de-Broglie waves, which is the phenomenon of interferometric self imaging of a transmission grating at the characteristic distance of the Talbot length from the grating. Special interest was taken in the demonstration of the fractional Talbot effect, that is the self imaging of the transmission grating with reduced period, because this effect is in principle useful for the production of very small structures in the atomic density distribution of an atom beam. Fractional self imaging up to seventh order with periods of one micron could be observed in the experiment. The third experiment is on diffraction of de-Broglie waves by a wire and on refraction of atomic de-Broglie waves by a rotationally symmetric potential. The Poisson interference spot in the center of the shadow of the wire was experimentally demonstrated as well as Fresnel diffraction fringes at the edges of the wire. Well defined voltages could be applied to two plate electodes opposite the wire, producing a radial electric field in the vicini
Atom lithography is a technique to structure layers of atoms during deposition, using interactions of near-resonant light fields with neutral atoms. The basic scheme uses a standing wave light field, aligned just above a substrate, while a beam of atoms impinges perpendicularly on the standing wave light field. The interaction of the light field and the atoms focuses the atoms towards the crests or troughs of the standing wave intensity, dependent on the tuning of the light field. The focused atoms are subsequently deposited on the substrate, resulting in an array of lines, whose periodicity is determined by the wavelength of the light field. The standing wave light field can therefore be seen as an array of lenses, focusing the atom beam into an array of lines. The first experiments on atom lithography were performed in the 1990s with sodium and chromium. In this thesis experiments with iron are described, which is a magnetic element and thus allows for the creation of magnetic nanostructures. To focus atom beams into structures using atom lithography, well-colimated atom beams are required. The most common technique to create collimated atom beams is transverse laser cooling, but iron has no closed optical transition suitable for laser cooling. Fortunately Smeets and Te Sligte found that even then atom lithography of iron is possible. If the standing wave light field is considered to be an array of lenses, the focusing of the atom beam on a single lens is determined by the local collimation of the atom beam at that position. This local collimation is determined by the geometry of the beam source and the distance of the beam source to the standing wave lens. Using a small beam source (e.g. 1 mm diameter) positioned far from the standing wave (e.g. 1 m), the geometrically collimated atom beam is suitable for atom lithography. Chapter 3 shows what the implications of geometric collimation are on the structure formation in atom lithography. The geometric effects that locally collimate the atom beam on the sample also introduce local offset angles of the atom beam with respect to the focusing light field. Structure formation has been observed with lines of up to 7 nm high on a background layer of 15 nm average height, over areas of up to 400 µm × 6 mm. This indicates that the offset angles can vary over an 8 mrad range, which is more than an order of magnitude larger than the local angle of collimation. The offset angle influence the local geometry of the deposited structures: an increasing offset angle decreases the local line height, increases the width and creates a skewness in the local line shape. The local geometry of the lines has been modelled with a Monte Carlo particle tracking model and the results were compared to experiments. It was found that the highest and narrowest lines suffer from significant broadening due to surface diffusion in experiments, limiting the achievable structure width of iron on SiOx samples to about 80 nm. We have also found that in atom lithography without laser cooling, each standing wave lens is effectively imaging the source geometry onto the substrate. We therefore propose using structured atom beam sources to image more complex patterns on sub-wavelength scales in a parallel way. The reason to use iron in atom lithography, is the option to create magnetic nanostructures. However, in previous experiments no clear magnetic signature of the atom lithographic structuring of iron was found. Therefore in Chapter 4 firstly the magnetic properties of unstructured layers of iron (Fe), nickel (Ni) and FexNi1-x are investigated. These layers are all ferromagnetic, but with a reduced magnetic moment compared to bulk values due to contaminations and crystal growth effects. As a reference for periodic atom lithography structures, periodic line structures are created with interference lithography in a polymer layer and onto these structures thin layers of FexNi1-x are shadow deposited, thereby creating a layer periodically modulated in both height and elemental composition. This periodic modulation in composition has been observed with SEM-EDX on sub-micron scales. The structures showed a clear magnetic anisotropy with an easy axis along the direction of the lines, in accordance with expectations. In Chapter 5, the magnetic properties of atom lithographic iron line structures are presented. Using MOKE microscopy, direct comparison of structured and unstructured parts of a sample is possible, which allows for the observation of clear magnetic signatures due to the atom lithographic structuring. In layers of an average thickness of about 15 nm, no anisotropy is induced by the line structures, while a magnetic easy axis along the direction of the lines is expected. The line structures introduce an isotropic increase in the coercivity of the layers, indicating that the line structures can be considered as isotropic corrugations instead of directional line structures. For layers of 30 nm average thickness and thus higher line structures, a clear magnetic anisotropy is observed for the highest line structures. This anisotropy can be seen as a magnetic easy axis along the direction of the lines, as expected for line structures. However, we also observe a sharp increase in coercivity for applied fields perpendicular to the lines. This phenomenon is intriguing and may be explained by a pinning of head-to-head domain walls along the direction of the lines. In Chapter 5, we also report on the first results of co-deposition of FexNi1-x, where the Fe is structured into lines using atom lithography, while the Ni is deposited uniformly, thereby creating an alloy of modulating composition. Magnetic analysis of these structures indicates that these structures are anisotropic, both the coercivity and the shape of domain walls is dependent on the angle of the applied field relative to the line structures. Finally in Chapter 6, we investigate the possibilities to focus a thermal beam of atoms into a single 100 nm spot using light fields, thereby creating a nanopencil suitable for deposition of nanostructures. A Monte Carlo particle tracking program was developed to model the all optical focusing or funneling of 10 µm sized atom beams to a 100 nm spot. This model included effects of initial beam divergence, magnetic substructure, laser cooling, and spontaneous as well as stimulated diffusion and it was applied to a number of promising light field configurations. The best results obtained for experimentally realistic settings are a beam focused to a full width at half maximum of FWHM = 0.55 µm in a blue detuned hollow beam where atoms are focused towards the dark center, or a FWHM = 0.31 µm in a red detuned axicon light field, where atoms are focused towards the maximum intensity at the heart line. However, in the latter case the focused beam has a flux only 25 times larger than the background flux, which is unpractical for applications. The limiting factors of all schemes are heating due to stimulated diffusion and the limited interaction time available for a thermal beam of atoms. A nanopencil could be possible if it were based on the focussing of a slow and / or monochromatic atom beam, but this would seriously complicate the practical application of such a device.
Atom lithography is a technique in which a light field is used to pattern an atomic beam. This patterned flux is then deposited onto a substrate, resulting in a nanostructured thin film. The smallest structures that have been made thus far using this technique are around 30 nm wide. This thesis investigates the technique, expanding its possibilities. The work-horse for the development of atom lithography has been Cr, as this transition metal atom has a closed transition in a wavelength range that is accessible to dye lasers. We extend the technique to Fe, the first ferromagnetic element to be used for atom lithography. The setup that was used to do this experiment is described, along with its critical design parameters. We present nanostructures that are typically 50 nm wide, and up to 4 nm high. The spacing between the nanolines is 186.05 nm. The nanostructure profile is compared to that of a simulated deposition process, and found to match. To the authors’ knowledge, this is the first demonstration of direct write atom lithography without laser cooling. A preliminary incursion into the magnetic properties of the nanostructures deposited is presented. In addition to giving an overview of the general ferromagnetic properties that might be expected, a deeper investigation of the magnetic anisotropy of the nanostructures deposited in this experiment is given. Novel resonant light masks are used in an experiment performed at the University of Konstanz (Germany). These light masks, using exactly instead of nearly resonant light, reveal some intriguing quantum mechanical effects. The most important of these features is the possibility to place structures closer together – at quarter wavelength spacings rather than half-wavelength intervals. Finally, the influence of surface diffusion on the structures obtained in an atom lithography experiment is investigated using kinetic Monte Carlo simulations. Several diffusion limiting effects are investigated; the influence of small amounts of residual reactive background gas is found to describe the experimental observations.