Lucent Technologies Inc.

At&t Bell Laboratories

Agere Systems Inc.

Optical Communication Products, Inc.

Leo M Chirovsky

Lucian Arthur D'Asaro

Sanghee Park Hui

Betty Jyue Tseng

Keith Wayne Goossen

John E Cunningham

Anthony L Lentine

David A Miller

George J Zydzik

Ted Kirk Woodward

William Scott Hobson

Donald W Dahringer

Philip J Anthony

Robert Allen Novotny

Shin-Shem Pei

Ronald Eugene Leibenguth

John Lopata

James Dennis Wynn

Robert A Morgan

Vincent D Mattera

A VCSEL is provided that integrates an absorbing layer sandwiched within a null of the standing wave in the emitting mirror to reduce the reflectivity and transmissivity of the emitting mirror as seen by the feedback optical wave, with minimal effect on the reflectivity of the emitting mirror as seen by the light exiting the cavity. The absorbing layer may be made of a suitable absorbing material, such as a GaAs layer in a laser emitting near 850 nm or highly doped p-layer, for instance, and may be disposed epitaxially in a semiconductor or metamorphic mirror. The absorbing layer sandwich may be incorporated into the VCSEL after the last mirror pair or at any desired position with the emitting mirror array.

Refined mirror structure for reducing the effect of feedback on a VCSEL

A lateral injection VCSEL comprises upper and lower mirrors forming a cavity resonator, an active region disposed in the resonator, high conductivity upper and lower contact layers located on opposite sides of the active region, upper and lower electrodes disposed on the upper and lower contact layers, respectively, and on laterally opposite sides of the upper mirror, and a current guide structure including an apertured high resistivity layer for constraining current to flow in a relatively narrow channel through the active region, characterized in that a portion of the lower contact layer that extends under the top electrode has relatively high resistivity. This feature of our invention serves two purposes. First, it suppresses current flow in parallel paths and, therefore, tends to make the current density distribution in the aperture more favorable for the fundamental mode. Second, it reduces parasitic capacitance.

Lateral injection vertical cavity surface-emitting laser

A technique is described for determining the performance of substrate-side emitting VCSELs formed on a wafer. The technique involves forming top-emitting VCSELs on the same wafer as bottom-emitting VCSELs and then testing the top-emitting VCSELs and using the results to determine the performance of the bottom-emitting VCSELs of the wafer.

Method of making wafer having top and bottom emitting vertical-cavity lasers

Wafer having top and bottom emitting vertical-cavity lasers

In a multi-layered dielectric mirror the higher refractive index layers comprise ZnS and the lower refractive index layers comprise a composite of approximately 95%MgF,and 5%CaF,by mole fraction. In one embodiment, the fluoride layers are e-beam deposited from an essentially eutectic melt of the two fluorides. In another embodiment, the semiconductor surface on which the mirror is formed is protected by an aluminum borosilicate glass layer. Application of the invention to the design and fabrication of VCSELs is also described.

Optical devices having ZNS/CA-MG-fluoride multi-layered mirrors

A VCSEL comprises separate current and optical guides that provide unique forms of drive current and transverse mode confinement, respectively. In one embodiment, the optical guide comprises an intracavity high refractive index mesa disposed transverse to the cavity resonator axis and a multi-layered dielectric (i.e., non-epitaxial) mirror overlaying the mesa. In another embodiment, the current guide comprises an annular first electrode which laterally surrounds the mesa but has an inside diameter which is greater than that of an ion-implantation-defined current aperture. The current guide causes current to flow laterally from the first electrode along a first path segment which is essentially perpendicular to the resonator axis, then vertically from the first segment along a second path segment essentially parallel to that axis, and finally through the current aperture and the active region to a second electrode. The dielectric mirror is deposited only after the formation of the guides in order to facilitate their fabrication.

Vertical cavity surface-emitting laser with optical guide and current aperture

A method for thermocompression bonding structures together including structures having different coefficients of thermal expansion, for example, thermocompression bonding optical diode arrays or other semiconductor structures to silicon substrates to form electronic or optoelectronic devices. The method includes aligning and contacting the structures to be interconnected, thermocompressing the structures via their contact pad elements at a bonding temperature, establishing an encapsulation temperature, applying an encapsulant material between the bonded structures, and curing the encapsulant material at the encapsulation temperature. Conventional bonding processes, which treat encapsulation as a separate step apart from bonding processes, melt the bonded assembly together and include at least one thermal cycle in which the bonded assembly is cooled to room temperature and then is re-heated to the encapsulation temperature before applying the encapsulant material. The inventive method eliminates this thermal cycle by including encapsulation as an integral step in the thermocompression bonding process and therefore reduces or eliminates the possibility of damage to bonds or their multi-layered contact elements between the time of bonding and the time of encapsulation. Resultingly, the inventive method advantageously improves the quality and reliability of the resulting bonds within the formed device compared to bonds formed in a conventional manner.

Method for thermocompression bonding structures

Solder bonding of first and second contact pad arrays is accomplished by forming contact structures, such as posts with vertical or tapered sides, on the contact pads of the first array and solder bumps on the second array. The respective contact structures should have an average cross-sectional area that is less than the average cross-sectional area of the corresponding solder bumps. The contact structures and solder bumps are then bonded by a bonding process at a temperature and pressure where the solder bumps deform and envelop at least a portion of the respective contact structures. It is possible to employ the contact structures as a compression stop during the bonding process. The temperature should be below the melting points of the contact structures. In this manner, solder bump spreading can be reduced during bonding which correspondingly reduces electrical shorting of adjacent formed interconnect bonds.

Method for solder-bonding contact pad arrays

The present invention provides a high impedance optical receiver circuit for use in integrated circuits. The receiver circuit consists of an optical detecting device connected to the gate of an FET device and further connected to a diode providing a load impedance. The FET device is connected to a biasing voltage through a biasing resistive element and to a conditioning stage output. The use of a diode to provide a load impedance allows for a smaller and easier to manufacture receiver circuit than would be possible using either a load resistor or a load FET. According to one aspect of the present invention, a digital integrated circuit employing SEED technology incorporates a plurality of diode-loaded receiver in an array of optical receiver circuits to reduce the footprint of the overall SEED circuit array.

Optical receiver with a high impedance preamplifier

An optical receiver, e.g., receiver 10 (FIG. 1), has differential optical input beams and generates an electrical output. The voltage at an electrical node between series-connected optical detector diodes is clamped within a predefined voltage range by series-connected clamping diodes, to prevent the voltage from increasing when consecutive logic one optical input beams are received. Variable bandwidth and low energy dissipation are achieved since the resistors of high input impedance and transimpedance receivers are not required. A second optical receiver, e.g., receiver 20 (FIG. 2) is a monolithic, diode-clamped S-SEED with complementary optical input beams and complementary optical output beams.

Diode-clamped optical receiver

FETs and quantum well diodes are combined on the same semi-insulating substrate, while providing the FETs with protection from spurious voltages. A deeply buried P region in the semi-insulating substrate is partitioned by a high resistivity proton implanted region, to provide both the P region of the quantum well diode and an isolating buried P layer for the FETs.

Planar fet-seed integrated circuits

Monolithic optically bistable modulator arrays, such as an M.times.N array of S-SEEDs, are electrically addressed with a matrix of electrical row and column contacts. Connected to the center node of each S-SEED is an addressing means having elements, such as diodes, transistors, or capacitors, which are electrically enabled and disabled.

Matrix addressed S-SEED optical modulator array

An optical switch in which states are defined by dynamic charge storage, rather than contention resolution, and which switches using pulsed radiation having a wavelength somewhat longer than the exciton wavelength in a SEED diode. The switch does not exhibit or need bistability but switches at a relatively low energy as compared to S-SEEDS switched at the exciton wavelength.

Dynamic optical logic using voltage pull up

A light emitting device uses a vertical cavity surface emitting laser having a voltage controlled mirror which is used, for example, to turn the laser ON and OFF by varying the reflectivity of one mirror forming the laser cavity.

Vertical cavity laser with mirror having controllable reflectivity

An optically bistable device, such as a symmetric self electro-optic effect device (S-SEED), is forced into a metastable state prior to the incidence of an optical input signal thereto, thereby increasing the sensitivity of the optically bistable device to the optical input signal, reducing both the switching time and the optical input signal energy required to switch the device. The metastable state is entered into by one of three techniques: (1) turning off the bias voltage V.sub.0 of the device with optical bias beams on then turning on the bias voltage V.sub.0 with the optical bias beams off; (2) applying a predetermined voltage to a node in the device, the predetermined voltage being substantially the metastable state voltage or V.sub.0 /2; or (3) subjecting the device to equal intensity optical bias beams having a wavelength longer than the exciton wavelength. In the second case, optical bias beams and optical input signal beams may be applied simultaneously with the application of the predetermined voltage. In the last case, optical bias beams (at the exciton wavelength) and the optical input signal beams can be applied simultaneously after removal of the long-wavelength optical bias beams, thus saving switching time. In addition, the optically bistable device may be selectively placed in the metastable state and the optical input signal have sufficient energy to switch the device only when it is in the metastable state. This allows for multiple optically bistable devices to be selectively responsive to a common optical input signal, such as in a demultiplexer.

Growing community of inventors

Bridgewater, NJ, United States of America

Leo M Chirovsky