Process and structure for 50+ gigahertz transistor

Details Process and structure for 50+ gigahertz transistor Process and structure for 50+ gigahertz transistor

2000-05-30

2002-07-02

Nelms, David (Department: 2818)

Active solid-state devices (e.g., transistors, solid-state diode

Bipolar transistor structure

With enlarged emitter area

C257S587000, C438S334000

Reexamination Certificate

active

06414371

ABSTRACT:

BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention generally relates to high performance transistors operable to microwave frequencies and, more particularly, to high performance transistor designs compatible with MOS/CMOS field effect transistors in integrated circuit designs.
2. Description of the Prior Art
The increase in amount of information transferred or distributed electronically has generated a substantial demand for increased bandwidth of data processing arrangements and high frequency communication links. Computer clock frequencies in excess of 1 Gigahertz (GHz) have been projected within a relatively few years. Microwave frequencies up to 50 GHz are currently in use. High speed data links are currently operating at 40 Gigahertz and frequencies of 100 Gigahertz are foreseeable. Additionally, high frequency communication links provide some substantial advantages in that the radiation pattern from a relatively small antenna can be readily constrained to limit required transmission power and to reduce susceptibility to interception of a transmitted signal. Microwave frequencies also have other properties which are being exploited in an increasing variety of devices. Thus, the demand for extremely high performance transistors and integrated circuits containing them is increasing rapidly.
Accordingly, there have been many recent developments to increase the frequency of operation of transistors. It is known that many limitations on frequency of operation derive from parasitic capacitances and resistances which have been reduced only at substantial cost and which appear to have reached a point of diminishing returns. Relatively large performance improvements in intrinsic device (e.g. a theoretical device without parasitics) performance yields only a small increase in the performance of a corresponding realized transistor or circuit in which it is placed. Other efforts to increase operational frequency or bandwidth of transistors have largely concentrated on exotic materials and device complexity; both tending to increase transistor cost and the latter tending to reduce manufacturing yield due to the intricacy and/or criticality of transistor structures and correspondingly complex processes by which they are formed.
Further gains in operating frequency and bandwidth are complicated by a requirement for manufacturing compatibility with other types of transistors such as complementary metal-oxide-semiconductor (CMOS) field effect transistors in integrated circuits used as support signal processing circuitry at somewhat lower frequencies. Transistors capable of operating at microwave frequencies are invariably bipolar at the present time and integrated circuits including both types of transistors are referred to as BICMOS devices.
BICMOS devices require substantial process complexity for manufacture and, generally, many process steps required for bipolar devices are incompatible with CMOS and vice-versa, requiring block out masking and sequential processing steps for each type of transistor. The overall heat budgets of both types of devices must be closely coordinated and are often particularly critical for bipolar transistors. This results in extremely long and costly processes with relatively low manufacturing yield. Increased complexity of bipolar transistors to obtain higher performance can only be accommodated with extreme difficulty.
Further, compatibility with CMOS technology requires use of a lower resistivity substrate, which is a principal reason for poor performance as well as causing higher parasitic capacitances, noise, and the like. For limitation of parasitic capacitance, the substrate should ideally function only as a thermal heat sink and mechanical support. However, it is not possible to obtain semi-insulating silicon with extremely high resistivity, as is the case with gallium arsenide.
SUMMARY OF THE INVENTION
It is therefore an object of the present invention to provide a simplified bipolar transistor design of high performance which is unconditionally compatible with CMOS in integrated circuits.
It is another object of the invention to provide a family of structures and manufacturing processes which can be employed singly or in any combination to improve the high frequency performance of most practical transistor designs and other electronic elements.
In order to accomplish these and other objects of the invention, a method of forming a high frequency electronic element such as a bipolar transistor, diode, FET or the like is provided including the steps of forming active regions such as a collector, base and emitter of the electronic element in one or more layers on a first substrate, forming contacts to selected ones of the active regions including support via structures, providing a second substrate on the contacts and the support via structures, removing a portion of the first substrate, patterning the respective layers including the active elements, and completing the transistor.
In accordance with another aspect of the invention, a high frequency transistor is provided including active regions comprising separated portions of respective layers, support vias including a contact formed on an active region, a metal layer formed over the contacts and support vias, and a substrate attached to at least one of the metal layer and the support vias.
In accordance with this method and structure, substantial material which can cause parasitic capacitance can be removed and connection length which can cause parasitic resistance and inductance can be reduced. Non-standard silicon processing techniques and microwave device processing techniques can be employed to further enhance high-frequency performance.


REFERENCES:
patent: 3973271 (1976-08-01), Okumura et al.
patent: 5268324 (1993-12-01), Aitken et al.
patent: 5338698 (1994-08-01), Subbanna
patent: 5350700 (1994-09-01), Yang et al.
patent: 5406114 (1995-04-01), Brenndoerfer et al.
patent: 5455448 (1995-10-01), Benjamin
patent: 5965930 (1999-10-01), Sakamoto et al.
Ahlgren et al.; “A SiGe HBT BiCMOS Technology for Mixed Signal RF Applicatins”; IEEE BCTM 12.3; 1997, pp. 195-197, Month unknown-Dec. ?
H. Kroemer, “Heterostructure Bipolar Transistors and Integrated Circuits”; Proceedings of the IEEE, vol. 70, No. 1, Jan. 1982; pp. 13-25.
Ahlgren et al .; “Device Reliablity and Repeatability of a High Perfomrance Si/SiGe HBT BiCMOS Technology”;pp. 452-455 Year/Month Unknown.
U. Konig; “200 GHz Potential of Si-Based Devices” Solid State Phenomena vols 69-70 (1999); pp. 121-130 Month Unknown.
Subbanna et al.; “How SiGe Evolved into a Manuacturable Semiconductor Production Process”; 1999 IEEE International Solid State Circuits Conference; Digest of Technical Papers; 1999; pp. 66-68.
Lee et al.; “Deep Submicron Transferred-Substrate Heterojunction Bipolar Transistors”; Department of ECE, University of California, Santa Barbara; pp. 26-27 (Month/Year Unknown).
Groves et al.; “High Q Inductors in a SiGe BiCMOS Process Utilizing a Thick Metal Process Add-On Module”; IEEE BCTM 9.3; 1999; pp. 149-152. Sep.?
P. Russer; “Si and SiGe Millimeter-Wave Intergrated Circuits”; IEEE Transactions on Microwave Theory and Techniques, vol. 46, No. 5, May 1998, pp. 590-603.
Tayrani et al.; “X-Band SiGe Monolithic Control Circuits”; 1998 Topical Meeting on Silicon Monlithic Integrated Circuits in RF Systems; Sep. 1998; pp. 126-133.
Subbanna et al.; “Integration and Design Issues in Combining Very-High-Speed Silicon-Germanium Bipolar Transistors and ULSI CMOS for System-on-a-Chip Applications”; IEDM 99; pp. 35.1.1-35.1.4, Sep. 1999.
Volant et al.; “Fabrication of High Frequency Passives on BiCMOS Silicon Substrates”; 0-7803-5687/X/00/$10.00 Ic) 2000 IEEE (Month Unknown).
Tokumitsu et al.; “Three-Dimensional MMIC Technology: A Possible Solution to Masterslice MMIC's on GaAs and SI”; IEEE Microwave and Guided Wave Letters, vol. 5. No. 11, Nov. 1995; pp. 411-413.

Affiliated with

Also associated with

Rating

Process and structure for 50+ gigahertz transistor does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Process and structure for 50+ gigahertz transistor, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Process and structure for 50+ gigahertz transistor will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-2829559