Magnetic nanostructures by atom optics

Abstract

Direct-write atom lithography is a technique in which highly periodic nanostructures are directly deposited on a substrate by patterning an atomic beam with a standing light wave, tuned near a resonance frequency of the corresponding atom. In this thesis an extension of this technique for the production of ferromagnetic nano- structures is investigated. Of all ferromagnetic elements Fe is the most suitable to per- form atom lithography. Light with a wavelength of 372 nm is needed to produce Fe nanostructures by atom lithography. Also, the atomic beam should be highly collimated, for which laser cooling is commonly used. For this purpose, a frequency doubled Ti:S laser tuned to 744 nm is constructed, capable of producing more than 100 mW of light at a wavelength of 372 nm. Laser cooling requires a laser with a frequency stability better than the natural linewidth of the atom. To meet this requirement, the laser is locked on the atomic transition by means of polarization spectroscopy on a hollow cathode discharge cell. The discharge cell produces a sufficient density of ground state Fe atoms at a temperature which is more than a factor of two lower than the temperature necessary to produce that density in a thermal cell. With this technique the laser is locked stable within 0.2 MHz for hours. A laser cooling simulation based on the semiclassical approach is presented. Although the intention was to simulate laser cooling of Fe with this code, for the commonly used pxpy polarization gradient configuration an analysis of the simulation results shows that the semiclassical approach is only valid for transitions with recoil parameters ?r on the order of 10i4 or less. For the standard laser cooling transitions only the transitions in Rb and Cs satisfy this condition. A drastic reduction in calculation time compared to quantum Monte-Carlo calculations is to be expected for these transitions by implementing an analytical approach to the long-term contribution of the diffusion coefficient. For Fe, full quantum Monte-Carlo calculations are necessary. Laser cooling of Fe on the 372 nm transition has been studied experimentally. This transition, although the only one realistically available, is far from perfect for laser cooling, as it is not a fully closed transition: it has a leak ratio of approximately 1/240 to other states. Nevertheless, a collimation of aRMS = 0.17 mrad could be achieved by laser cooling. However, contrary to the normally obtained central beam flux increase, the ground state atom flux decreased to approximately 70 % of the non-cooled value. Because of both the complexity of the laser cooling setup and the restrictions in perfor- mance, we have also attempted to produce Fe nanostructures without using laser cooling techniques, but simply relying on geometrical collimation of the atomic beam. In this way, Fe nanolines are deposited with a pitch of 186 nm, a FWHM width of 50 nm, and a height of up to 6 nm. This opens the way for applying direct-write atom lithography to a wide variety of elements. In a MOKE (Magneto-Optical Kerr effect)-setup we are able to verify that the deposited Fe-layer is ferromagnetic. However, the magnetic anisotropy due to the presence of the nanolines can not be detected with a MOKE-setup or a MFM (Magnetic Force Microscope). In contrast, the magnetization of background-free nano- lines, deposited through a mechanical mask with a pitch of 744 nm and a FWHM of