| Summary: | This dissertation examines various aspects of optical disk data storage systems from the point of view of the irradiance and phase distributions at the exit pupil of the objective lens. The research topics were chosen in order to address some of the problems facing future generations of optical disk systems. Future optical disks will have a much greater areal data density, and will undoubtedly use a shorter wavelength in order to decrease the size of the optical stylus. The research in this dissertation examines some of the problems inherent in the move to shorter wavelengths. For example, at short wavelengths, the tolerance on acceptable disk tilt becomes tighter; disks must either be manufactured with a tighter flatness tolerance, or a system must be devised such that the problems caused by disk tilt are corrected inside the optical disk drive. (Such a system is described in Chapter 6.) The other topics in this dissertation address similar problems, all of which are essential for a more complete understanding of optical disk technology. After a brief introduction to the irradiance distribution at the exit pupil of the objective lens (typically called the baseball pattern), we describe a novel focusing/tracking technique. Many optical disk drives use the astigmatic technique, in which the baseball pattern is projected onto an astigmatic lens that focuses the beam onto a quadrant detector placed between the two line foci of the lens. We experimentally demonstrate that by projecting the baseball pattern onto a ring lens (i.e. a lens that focuses light to a ring rather than a single spot), we are able to produce steeper focus-error signals that are more resistant to feed-through (induced on the focus-error signal by track crossings during the seek operation) than the astigmatic technique. We then examine the effects of substrate birefringence and tilt on the irradiance and phase distributions at the exit pupil of the objective lens. The irradiance and phase patterns are calculated and experimentally verified for the cases of no substrate birefringence, birefringence aligned with the incident polarization, and birefringence aligned at 45° to the incident polarization. The irradiance at the exit pupil is also calculated and experimentally verified for a grooved substrate for various amounts of substrate tilt. We then examine two distinct effects that are dependent on the incident polarization direction. The first of these is the excitation of surface plasmons at the interface between the dielectric substrate (or air, if the optical disk's storage layer is air-incident) and the metallic thin films in the disk. These plasmons are responsible for dips in the zeroth order diffraction efficiency curves of a metal grating at certain angles of incidence. The dips appear as dark bands in the baseball pattern and are seen only when there is a component of incident polarization that lies perpendicular to the tracks. The location of these bands is derived from theoretical considerations and is shown to depend on the track pitch and the materials involved, but not on the groove depth or width. The band locations are confirmed by zeroth order diffraction efficiency measurements as a function of incident angle. A possible negative effect of these bands is the introduction of additional fluctuations and noise into the focusing and push-pull tracking signals. The second of the polarization-dependent effects concerns the differences in tracking performance with respect to the direction of the incident-light polarization. In optical disk storage systems, the signal that provides tracking information is dependent on the groove shape, the optical constants of the materials involved, and the polarization state of the incident light. We show that the tracking signals can be described by two measurable quantities, both of which are largely independent of aberrations in the optical system. Using these two quantities, we match the tracking performance of a given optical disk to an equivalent disk having rectangular grooves - the adjustable parameters being the rectangular groove depth and the duty cycle. By assumption, the rectangular grooves modulate only the phase of the incident beam and disregard its state of polarization. The effective groove depth and the duty cycle thus become dependent on the polarization state of the incident beam. We examine the dependencies for various disks having different groove geometries and different combinations of materials. Next, we use the baseball pattern as a diagnostic tool to develop and demonstrate the concepts of a servo system for the correction of disk tilt. Since disk tilt produces primarily coma in the beam focused onto the disk, the system uses a "variable coma generator" to produce and equal and opposite amount of coma as that caused by the tilted disk. The magnitude and direction of disk tilt are detected using the light reflected from the front facet of the disk substrate. Finally, we address a major obstacle in the construction of future generation optical disk testers - the use of shorter wavelengths and thinner substrates. A typical aspheric singlet used as the objective lens in optical disk data storage systems will not work at different wavelengths or with different substrate thicknesses, due to spherochromatism. Using two microscope objectives with adjustable collars and a pair of relay lenses, we have constructed a system in which a diffraction-limited spot of any wavelength in the range of 0.4 μm - 0.7 μm can be moved by as much as ± 1 00 μm in both the focusing and tracking directions. This is accomplished by simply moving an aspheric singlet mounted in an off-the-shelf optical head. The system uses the adjustable collars of the microscope objectives to correct for the spherochromatism of the singlet, and to accommodate the various thicknesses of the substrates.
|
|---|
