Semiconductor device manufacturing: process – Forming bipolar transistor by formation or alteration of... – Including isolation structure
Reexamination Certificate
2001-11-20
2003-07-15
Nelms, David (Department: 2818)
Semiconductor device manufacturing: process
Forming bipolar transistor by formation or alteration of...
Including isolation structure
C438S360000, C438S363000
Reexamination Certificate
active
06593200
ABSTRACT:
FIELD OF THE INVENTION
This invention relates generally to semiconductor processing, and in particular, to a method of forming an integrated inductor and/or high speed interconnect in a planarized process with shallow trench isolation.
BACKGROUND OF THE INVENTION
Inductors are often employed in integrated circuits to perform specific functions according to the requirements of the circuits. Such functions include, for example, providing inductive reactance in filters and resonators, etc. An important parameter of an inductor relates to its loss characteristics, usually referred to as the quality factor (i.e. Q) of the inductor. Generally, it is desirable for an inductor to have low loss characteristics (or conversely, high Q characteristics). The higher the Q-factor of an inductor, generally the more ideal its impact is on the desired signal it is acting upon. Often inductors are combined with a capacitor to create a filter or resonator.
For example in a bandpass filter the goal is to pass signals in a selected range of frequencies, known as the “passband”, while blocking signals outside of this range. A high Q inductor in a bandpass filter allows signals in the passband to experience less loss while signals outside of this range experience more attenuation. In an oscillator, a high Q inductor is required to realize a high Q resonator. Practical benefits of a high Q resonator are better frequency stability and less unwanted spurious signals generated along with the desired signal. These unwanted spurious signals give rise to phase noise, which is an important limiting factor in digital radio performance.
Interconnects are generally used in RF integrated circuits (RFIC's) for digital radio applications and High Speed Digital (HSD) circuits for fiber-optic interface circuits. At frequencies above 1 GHz, these interconnects act more as transmission lines which guide traveling electromagnetic waves along a desired path versus the more simplistic low frequency concept of a wire carrying current from one part of a circuit to another. Similar to an inductor, the quality of an interconnect is highly influenced by its loss characteristics and noise immunity of the medium within which it resides. Lower loss interconnects reduce the gain requirements for RFICs, thereby reducing their power consumption. The lower loss interconnect in fiber-optic ICs reduces signal distortion and allows for higher frequency of operation. Increased noise immunity results in greater signal sensitivity in RFICs and reduced timing jitter in fiber-optic interface ICs.
Many factors can affect the Q of an inductor and loss characteristics of an interconnect. For instance, the conductivity of the metallization forming the inductor has an impact on the Q-factor of an inductor. Generally, the higher the conductivity of the metallization forming an inductor, the higher Q-factor of the inductor. Conversely, the lower the conductivity of an inductor, the lower the Q-factor of the inductor. Another factor that affects the Q-factor of an inductor is the proximity of another conductive layer to the inductor, since currents can be generated in this layer due to magnetic coupling to the inductor. These are known as “eddy” currents. In many semiconductors, the conductive layer is actually more like a sheet of moderately resistive material. These eddy currents dissipate power in this layer due to ohmic losses and hence remove energy from the inductor, which lowers its Q.
Similarly interconnects experience losses when layers of moderate conductivity are in close proximity, due to the same eddy current generation. This leads to increased insertion loss, which is undesirable. The presence of a moderately conductive layer underneath the high speed interconnect also provides a path way for spurious signals from adjacent interconnect to enter the main high speed signal line. This results in undesired cross-talk noise and is particularly problematic at frequencies above 1 GHz. The absence of the conductive layer underneath the high speed interconnect reduces this cross-talk effect considerably thus increasing noise immunity.
Hence, for both inductors and interconnects the elimination of moderate conductivity layers in the region below these structures improves performance. Specifically removal of this layer improves inductor Q and decreases the insertion loss of interconnects. In both cases the benefit increases with frequency, particularly above 1 GHz.
Integrated inductors are often used in radio frequency (RF) bipolar and BiCMOS technologies. However, these technologies typically incorporate a layer of low resistivity epitaxial silicon. Thus, forming an inductor proximate a low resistivity epitaxial silicon layer typically degrades the Q-factor of the inductor, which is undesirable. Similarly, high speed interconnect proximate a low resistivity epitaxial silicon layer degrades the insertion loss of the interconnect. The following explains the prior art high speed bipolar and BICMOS processes and illustrates the proximity of the epitaxial silicon layer to the inductor and the interconnect, which has undesirable consequences.
FIG. 1A
illustrates a cross-sectional view of a semiconductor device
100
at an intermediate step of a prior art method of forming a bipolar or BICMOS semiconductor. At this step, the conductor device
100
consists of a substrate
102
and a mask layer
104
including an opening
106
to define underneath a buried implant region
108
of the substrate
102
. At this step, the semiconductor device
100
is undergoing an ion implanation process to form the buried implant region
108
of the substrate
102
. The buried implant region
108
defines at least a portion of an active device being formed.
FIG. 1B
illustrates a cross-sectional view of the semiconductor device
100
at a subsequent step of the prior art method of forming a bipolar or BICMOS semiconductor. In this step, the mask layer
104
is removed. Then, a layer of low resistivity epitaxial silicon
110
is formed over the substrate
102
and the buried implant region
108
of the substrate
102
.
FIG. 1C
illustrates a cross-sectional view of the semiconductor device
100
at a subsequent step of the prior art method of forming a bipolar or BICMOS semiconductor. In this step, a shallow trench
112
is formed by removing a top portion of the low resistivity epitaxial silicon layer
110
. Also in this step, the active components can be formed.
FIG. 1D
illustrates a cross-sectional view of the semiconductor device
100
at a subsequent step of the prior art method of forming a bipolar or BICMOS semiconductor. In this step, an intermetal dielectric layer
114
is formed over the low resistivity epitaxial silicon layer
110
and within the shallow trench
112
. After forming the intermetal dielectric layer
114
, an inductor and/or an interconnect
116
is formed over the intermetal dielectric layer
114
. As
FIG. 1D
illustrates, the inductor and/or the interconnect
116
is situated over the low resistivity epitaxial silicon layer
110
. Since the layer
110
is electrically conductive, it degrades the Q-factor of the inductor and/or the interconnect
116
, which is undesirable especially for RF applications.
Thus, there is a need for a method of forming a semiconductor device (such as a bipolar or BICMOS device) including an inductor and/or an interconnect that is not situated over a conductive layer in order to prevent degradation of the Q-factor of the inductor and/or improve noise isolation properties and loss characteristics of the interconnect. Such a need and others are met with the method of forming a semiconductor device in accordance with the invention.
SUMMARY OF THE INVENTION
A first aspect of the invention relates to a method of forming a semiconductor device with inductor. The method comprises forming an epitaxial layer over the substrate, forming an opening through the epitaxial layer to expose an underlying region of the high resistivity (relative to the epi resistivity) substrate, forming a first dielectric material within the
Choutov Dmitri A.
Kalnitsky Alexander
Scheer Robert F.
Stutzin Geoffrey C.
Blakely , Sokoloff, Taylor & Zafman LLP
Le Dung Anh
Maxim Integrated Products Inc.
LandOfFree
Method of forming an integrated inductor and high speed... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method of forming an integrated inductor and high speed..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method of forming an integrated inductor and high speed... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3050588