Active solid-state devices (e.g. – transistors – solid-state diode – Incoherent light emitter structure
Reexamination Certificate
2001-02-20
2002-12-03
Lee, Eddie (Department: 2815)
Active solid-state devices (e.g., transistors, solid-state diode
Incoherent light emitter structure
C257S082000, C257S083000, C257S084000, C257S102000, C257S103000, C257S184000, C257S225000, C257S233000
Reexamination Certificate
active
06489635
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
Briefly described, this invention relates to an improved method for fabricating a p-i-n semiconductor photodiode that results in improved uniformity and device scalability over large substrate sizes.
2. Description of Related Art
Linear and 2-D arrays of InGaAs p-i-n photodiodes are commercially important for a wide range of applications. Linear arrays are used as wavelength monitors for wavelength division multiplexing (WDM) communication systems (C. Koeppen, J. L. Wagener, T. A. Strasser, and J. DeMarco, “High resolution fiber grating optical network monitor,” Proc. National Fiber Optic Engineers Conference, 1998.), spectroscopy (G. H. Olsen, “InGaAs Fills the Near-IR Array Vacuum,” Laser Focus World 27, p. 48 (1991)), and machine vision applications. Two-dimensional arrays are used in many applications ranging from surveillance to industrial process control. Linear photodiode arrays are available with up to 512 elements, while 2-D arrays are available with up to 320×240 elements. These devices are large in comparison to most semiconductor devices. A 512 linear array is 2 cm in length and a 2-D array is also large, with a 320×240 array being 1.3 cm×1 cm. This large device size drives the need for a device structure and process which is uniform and scalable with substrate size. Conventional methods of device fabrication used a vapor-phase Zn diffusion to define the device active region.
Increasing substrate sizes characterizes the history of semiconductor fabrication. This increase in size is driven by the cost effectiveness of scaling the process to larger and larger wafer sizes. The trend is particularly true of more complex devices containing large numbers of devices, such as transistors. The compound semiconductor industry is driven by a similar trend as the devices fabricated from these materials contain a larger number and size of elements such as transistors and diodes. The GaAs industry has scaled wafer sizes from 2 inch diameters (50 mm) up to 6 inches (150 mm) in recent years as the chips fabricated from this material have achieved increasing levels of device integration. The InP-based materials systems are experiencing a similar pressure to increase the substrate size to take advantage of the cost advantages of larger wafer sizes.
As the wafer size is increased, the processing of the devices becomes more challenging. To take advantage of the larger wafer size, the process must scale with the area of the wafer. The invention discussed herein addresses the problem of scaling the diffusion of impurities to large wafer sizes.
The following patents may be generally relevant to the state of the art and the present invention.
U.S. Patent entitled “Photodetector” (U.S. Pat. No. 4,597,004) describes a photodetector in which the preferred embodiment is structured with a non-patterned p-contact layer, and in which a buffer layer is added to prevent excess diffusion from the p-doped layer.
U.S. Patent entitled “Photovoltaic Diode with First Impurity of Cu and Second of Cd, Zn, or Hg” (U.S Pat. No. 3,836,399) describes a device in which the diffusion is over a broad area and for which no specific method or patterned structure is indicated.
U.S. Patent entitled “Enhancement of Photoconductivity in Pyrolytically Prepared Semiconductors” (U.S. Pat. No. 4,616,246) describes a device prepared using n-type layers grown, but not subsequently patterned or diffused, to fabricate the device.
U.S. Patent entitled “Buried Heterostructure Devices with Unique Contact-Facilitating Layers” (U.S. Pat. No. 4,661,961) describes a fabricated device in which the contact layers are patterned by growth of additional material to define the device active region.
U.S. Patent entitled “Passivation of InP by Plasma-Deposited Phosphorus” (U.S. Pat. No. 4,696,828) shows a method for passivating surface states using a phosphorus layer.
U.S. Patent entitled “Lateral P-I-N Photodetector” (U.S. Pat. No. 4,746,620) describes a device in which the p-type and n-type regions of the photodiode are defined by alloying doped metallic contacts onto the semiconductor substrate.
Finally, U.S. Patent entitled “PIN Junction Photovoltaic Element with p or n-type Semiconductor Layer Comprising Non-Single crystal Material Containing Zn, Se, H in an Amount of 1 to 4 Atomic % and Dopant and I-Type Semiconductor Layer Comprising On-Single Crystal Si(H,F) Material” (U.S. Pat. No. 4,926,229) describes a device fabrication process which uses non-single crystal and in which H is incorporated into the material.
The existing methods of scaling the diffusion of impurities into the semiconductor in order to define the device include sealed ampoule diffusion, open tube diffusion using a gaseous diffusion source, and the use of surface-deposited films of source dopant.
Sealed ampoule diffusion (Y. Yamamato, H. Kanabe, “Zn Diffusion in In
x
Ga
1−x
As with ZnAs
2
Source,” Japn. J. of Appl. Phys., vol. 19, No. 1, Jan 1980, pp. 121-128.) is characterized by the difficulty of handling large diameter quartz ampoules. In this method, the semiconductor wafer is sealed in a quartz ampoule containing a solid source of diffusion such as ZnAs
2
. The ampoule is placed on a high vacuum pumping system and evacuated to 10
−6
torr. The quartz ampoule is then sealed manually using an oxygen/hydrogen torch. The sealed ampoule is then placed in a furnace at the appropriate diffusion temperature. The deficiencies of this method include the manual handling problems involved with large diameter quartz, the large thermal mass requiring longer diffusion times, and the time consuming vacuum processing and sample preparation.
Open tube diffusion may be implemented in a number of different embodiments. Common to all of these embodiments (M. Wada, M. Seko, K. Sakakibara, and Y. Sekiguchi, “Zn diffusion into InP using dimethylzinc as a Zn source,” Japn. J. of Appl. Phys., vol. 28, no. 10, October 1989, pp. L1700-L 1703; C. Blaauw, et al, “Secondary ion mass spectrometry and electrical characterization of Zn diffusion in n-type-InP,” J. Appl. Phys. 66 (2), 1989, pp 605-610; N. Arnold, et al, “Diffusion in III-V semiconductors from spin-on film sources,” J. Phys. D: Appl. Phys., 17, 1984, pp. 443-474.) is the requirement to maintain a sufficient overpressure of column V elements to prevent the compound semiconductor from decomposing at the temperatures required for impurity diffusion. This overpressure is usually achieved with the use of a gaseous source of phosphine (PH
3
) or arsine (AsH
3
). Both of these gases are extremely hazardous and require special handling, effluent control, licensing, and safety precautions. The cost of such gas handling is usually prohibitive and becomes more complex for larger and larger wafer sizes.
Spin-on diffusion sources are becoming more widely used (N. Arnold, R. Schmitt, K. Heime, “Diffusion in III-V semiconductors from spin-on film sources,” J. Phys. D: Appl. Phys., vol. 17, 1984, pp. 443-474.). They are scalable with device size and do not suffer the same requirements for column V overpressure as open tube or sealed-ampoule diffusion. This process does, however, suffer from the standard process variability common to all thin film deposition techniques. The source quality (freshness) and application to the substrate require precise control.
Diffuision is the preferred method of introducing impurities for p-i-n photodiode fabrication. The principal reason is that the material damage introduced by the ion implantation process causes defects that increase the dark current of the device (I. M. Tiginyanu, I. V. Kravetsky, V. V. Ursaki, G. Marowsky, H. L. Hartnagel, “Crystal order restoration and Zn-impurity activation in InP by As++ plus—coimplantation and annealing,” Physica Status Solidi (A) Applied Research v 162 n 2 August 1997. p R9-R10). Large dark currents are detrimental to photodiode performance for most applications.
In general, the prior art for diffusion of impurities are limited in process control and scalability. This inventio
Lee Eddie
Lee Eugene
Sensors Unlimited
Woodbridge & Associates PC
Woodbridge Esq. Richard C.
LandOfFree
Epitaxially grown p-type diffusion source for photodiode... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Epitaxially grown p-type diffusion source for photodiode..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Epitaxially grown p-type diffusion source for photodiode... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2961478