Abstract: A method of measuring the thickness of ultra-thin (e.g., <200 Å) oxide formed on a semiconductor device uses a reference sample (i.e., another silicon wafer) which has been pre-processed to include a relatively thick oxide surface layer. The thickness of the reference oxide (t) is measured using any conventional technique (such as an ellipsometer). An ultra-thin oxide is then simultaneously formed on both the reference sample and semiconductor device. The total oxide thickness (T) of the dual-layer structure on the reference sample is then measured (again, using any conventional technique), and the difference between the two measured values (T−t=&dgr;) is defined as the thickness of the ultra-thin oxide layer.

Abstract: A beam expander for providing coupling between a semiconductor optical device and an optical fiber comprises a double layer structure that may be integrated with the optical device. The first, underlying layer of the expander comprises a relatively high refractive index material (e.g., 3.34), thus providing improved coupling efficiency between the optical device and the fiber. The second, covering layer of the expander comprises a relatively low refractive index material (e.g., 3.28), for providing the large mode size desired at the fiber input. The parameters of each layer can be adjusted independently, allowing for the two criteria (coupling efficiency and mode size) to be separately optimized.

Abstract: A method for bonding an oxide-containing member to an aluminum surface, includes the steps of: providing a substrate, wherein a portion of a surface of the substrate has aluminum thereon; positioning an oxide-containing member on the aluminum surface of the substrate; and bonding the oxide-containing member to the aluminum surface by pressing the oxide-containing member against the aluminum surface while simultaneously heating the interface between the oxide-containing member and the aluminum surface. The interface between the oxide-containing member and the aluminum surface is heated with a laser.

Abstract: A fiber array connector is passively self-aligned to provide optical coupling between a first N×M array of optical fibers and a second N×M array of optical fibers. Each fiber array is inserted in a separate half of the connector, where each individual fiber is then supported through an aperture in a fiber array connector piece part. The connector piece part comprises a stack of substrate members, processed to include apertures for supporting the fibers in an array formation. The top substrate member of each stack is further processed to include alignment apertures so that as the two fiber array connector halves are mated the alignment fiducials on the connector piece parts will mate and self-align the fiber arrays.

Abstract: A beam expander for providing coupling between a semiconductor optical device and an optical fiber comprises a double layer structure that may be integrated with the optical device. The first, underlying layer of the expander comprises a relatively high refractive index material (e.g., 3.34), thus providing improved coupling efficiency between the optical device and the fiber. The second, covering layer of the expander comprises a relatively low refractive index material (e.g., 3.28), for providing the large mode size desired at the fiber input. The parameters of each layer can be adjusted independently, allowing for the two criteria (coupling efficiency and mode size) to be separately optimized.

Abstract: A fiber array connector is passively self-aligned to provide optical coupling between a first N×M array of optical fibers and a second N×M array of optical fibers. Each fiber array is inserted in a separate half of the connector, where each individual fiber is then supported through an aperture in a fiber array connector piece part. The connector piece part comprises a stack of substrate members, processed to include apertures for supporting the fibers in an array formation. The top substrate member of each stack is further processed to include alignment apertures so that as the two fiber array connector halves are mated the alignment fiducials on the connector piece parts will mate and self-align the fiber arrays.

Abstract: A fiber array connector is passively self-aligned to provide optical coupling between a first N×M array of optical fibers and a second N×M array of optical fibers. Each fiber array is inserted in a separate half of the connector, where each individual fiber is then supported through an aperture in a fiber array connector piece part. The connector piece part comprises a stack of substrate members, processed to include apertures for supporting the fibers in an array formation. The top substrate member of each stack is further processed to include alignment apertures so that as the two fiber array connector halves are mated the alignment fiducials on the connector piece parts will mate and self-align the fiber arrays.

Abstract: A method of measuring the thickness of ultra-thin (e.g., <200 Å) oxide formed on a semiconductor device uses a reference sample (i.e., another silicon wafer) which has been pre-processed to include a relatively thick oxide surface layer. The thickness of the reference oxide (t) is measured using any conventional technique (such as an ellipsometer). An ultra-thin oxide is then simultaneously formed on both the reference sample and semiconductor device. The total oxide thickness (T) of the dual-layer structure on the reference sample is then measured (again, using any conventional technique), and the difference between the two measured values (T−t=&dgr;) is defined as the thickness of the ultra-thin oxide layer.

Abstract: A fiber array subassembly for use with a two-dimension MEMS mirror array comprises a plurality of separate substrate elements (for example, silicon substrates) stacked one upon the other and formed to include via holes that, upon stacking, form the fiber array apertures. By controlling the via location on each substrate element, precise registration between the location of each optical device in the array (such as mirror elements in a MEMS arrangement) and the communication fibers can be achieved. The stack of substrates may also be formed to including a “mechanical stop” for each fiber, ensuring a precise separation between each fiber endface and its associated lensing elements.

Abstract: A variable fiber optic attenuator is formed of a first, movable section of optical fiber and a second, fixed section of optical fiber. The fibers are disposed end-to-end on a substrate, where the substrate includes a cavity formed underneath an end portion of the movable fiber. As supported, the fiber endfaces are aligned such that complete optical coupling is achieved between the fibers. Upon application of an external force to the first, movable fiber, the end portion of the fiber is forced downward into the cavity (such as, for example, by application of an external magnetic field and the inclusion of a permalloy sleeve around the end portion). As the fiber moves downward into the cavity, the endface will become misaligned with the endface of the fixed fiber, leading to attenuation of the signal passing through the fixed fiber. By controlling the force applied to the movable fiber, the degree of attenuation can be well-controlled.

Abstract: A self-aligned 1×N fiber optic switch uses silicon optical bench technology to provide the alignment between the single fiber and the “selected” one-out-of-N fibers. A first substrate stack is used to support the single fiber, using a via etched through the thickness of the stack, and includes alignment fiducials formed on the top surface thereof. A second substrate stack is used to similarly support an array of N fibers, the array disposed in a circular arrangement with respect to the top surface of the second substrate stack. A set of alignment fiducials is also formed on the top surface of the second substrate stack. One alignment feature includes a “centering” fiducial to allow for the two stacks to be properly aligned. Another alignment fiducial includes a circular channel, having a diameter greater than that of a fiber array, and allowing for alignment sphere to “rest” in the channel when the substrate stacks are mated.

Abstract: A cleaving apparatus, system and method which inhibit damage to cleave edges of laser bars are described. The cleaving apparatus includes a cleaving device having a first and a second slope meeting at a solitary cleave line. The slopes are angled to provide a solitary cleave point at the cleave line. Preferably, the first slope is between zero and four degrees and the second slope is about twenty degrees above a support structure upon which the cleaving device is mounted. A source of air may optionally provide air directed at the cleave line to assist in cleaving the laser bars.

Abstract: A method of depositing material on a substrate using a shadow mask. The mask includes a plurality of etched features which correspond to a plurality of features in the substrate. Spheres are provided in the features of the mask or substrate so that the mask and substrate are aligned when each of the spheres occupy both of the corresponding features in the mask and substrate.

Abstract: A method and apparatus are provided for freeing laser bars stuck in a facet coating fixture. A roller device is provided that rolls into a facet coating fixture channel and applies a downward force to a stuck laser bar. The roller device provides an uncomplicated and reliable system for handling the bars.

Abstract: A holding apparatus including a fixture is described. The fixture of the holding apparatus, which receives a semiconductor laser bar for optical coating, includes a base and a pair of walls. The walls and the base have inner surfaces, and the walls have a top surface upon which the laser bar rests. An extending portion of the semiconductor laser bar fits in a cavity formed by the base and the walls. The facet edges of the laser bar do not extend horizontally beyond the walls of the fixture. A coating apparatus applies a coating to the facet edges of the laser bar without contacting any coating material on the extending portion.

Abstract: An arrangement for providing attachment and alignment between an array of surface emitting lasers and an array of optical fibers utilizes a first substrate for supporting the laser array and a second substrate for supporting the fibers. The second substrate is formed to including a plurality of apertures that are disposed to self-align with the laser array upon attachment. Alignment fiducials may be formed on both substrates to facilitate the alignment.

Abstract: A silicon optical bench arrangement is used to provide for proper angular orientation between various components of an optical subassembly. A set of "optical stops" comprising alignment spheres held in pyramidal detents are formed in predetermined locations across the surface of an optical substrate. The separate optical components are positioned adjacent to the optical stops and subsequently held in place by bonding or other attachment means. By properly positioning the optical stops with respect to one another, any desired orientation between the optical components can be achieved.

Abstract: The invention is a method for protecting an aluminum layer from corrosion when it is formed by means of a photoresist mask on the same substrate as a multilayer such as titanium, platinum, and gold. A protection layer such as titanium is formed over the aluminum layer prior to stripping the photoresist mask, and is removed subsequent to stripping and prior to bonding components to the aluminum layer.

Abstract: A system is provided for freeing laser bars stuck in a facet coating fixture. A roller device is provided that rolls into a facet coating fixture channel and applies a downward force to a stuck laser bar. The roller device and coating fixture combination provide an uncomplicated and reliable system for handling the bars.

Abstract: A tin-rich gold-tin solder is disclosed which is particularly advantageous for self-aligning applications. When utilized with gold-plated bond locations, the out-diffusion of tin from the solder during heating functions to shift the composition of the remaining solder closer to the eutectic value, thus preserving the liquid state of the solder and improving its reflow quality with respect to conventional eutectic solders.