Semiconductor device manufacturing: process – Making field effect device having pair of active regions... – On insulating substrate or layer
Reexamination Certificate
2002-12-06
2003-12-23
Nhu, David (Department: 2818)
Semiconductor device manufacturing: process
Making field effect device having pair of active regions...
On insulating substrate or layer
C438S770000, C438S198000
Reexamination Certificate
active
06667197
ABSTRACT:
FIELD OF THE INVENTION
This invention relates generally to the manufacture of high speed semiconductor microprocessors, application specific integrated circuits (ASICs), and other high speed integrated circuit devices. More particularly, this invention relates to a method of forming a uniform thermally grown oxide layer over surfaces with high doping level and different doping type.
BACKGROUND OF THE INVENTION
Since its invention in the 1960's, the thermal oxidation of silicon is considered to be the enabling process for modern integrated circuits. After more than four decades of extensive investigation, there is a vast art on various aspects of the thermal oxidation and its applications to the fabrication of integrated circuits and other microstructures. Several key properties of thermal oxidation distinguish it from other methods of forming dielectric on semiconductor. These include: (1) a nearly electrically perfect interface between silicon and silicon oxide; (2) high dielectric strength of the oxide; and (3) excellent control of the uniformity of the oxide film.
These properties have enabled the use of thermal oxide as, for example, the gate dielectric, the isolator in various LOCOS (local isolation of silicon) structures, the isolator for various IT (isolation trench) structures, the gate sidewall isolator/spacer, and the screen oxide for ion implantation.
The kinetics of thermal oxidation are well understood. The Deal-Grove (linear-parabolic) model provides an insight into physical and chemical processes occurring during the thermal oxidation of silicon. The model was put forward in the 1960's as a physical interpretation of experimental data, and since then its validity has been repeatedly confirmed. According to this model, the oxide growth rate is limited by the speed of the interface reaction in the thin film regime. (The oxide film is less than about 500 angstrom.) Consequently, the oxidation rate is greatly affected by doping level and type. Indeed, the presence of a large amount of dopant alters the free energy of the silicon interface, changing the rate of interfacial chemical reactions. In addition, dopant type (n or p) affects the free energy in different ways. The dependence of the oxide growth rate on doping level and type is often referred to in the art as “the dopant effect.” For instance, the dependence of the oxidation rate on phosphorus doping level is illustrated in “
Silicon Processing for the VLSI Era, Volume
1
: Process Technology
,” by S. Wolf and R. N. Tauber, 2nd edition, Lattice Press, Sunset Beach, Calif., 2000, pp. 198-207, 213-215. The data show a rapid increase of the oxidation rate in the thin film regime while there is a negligible dopant effect in the parabolic, thick film regime. As the concentration of phosphorus increases from 1×10
19
to 1×10
21
cm
−3
, the growth rate increases in excess of 1000 percent. There is a negligible dopant effect for a phosphorus concentration below 1×10
19
cm
−3
. The oxidation art teaches that the dopant effect is fundamental in nature, and therefore cannot be eliminated by simply changing either the type of oxidation tool (e.g., horizontal furnace, vertical furnace, rapid thermal processor) or the oxidation ambient (e.g., dry, wet, high pressure, low pressure).
The dopant effect can pose a serious obstacle to the fabrication of a uniform thermal oxide film over a silicon surface with different doping level and type. Therefore, there is a need in the art for a versatile oxidation method with reduced dopant effect.
U.S. Pat. No. 5,412,246 to Dubuzinsky et al. describes plasma assisted oxidation of silicon and silicon nitride performed at a low plasma power. Dubuzinsky et al. teach that a high power plasma oxidation may cause damage to the grown oxide film. Therefore, a low plasma power process was selected to produce high-quality oxide films. Dubuzinsky et al. also disclose useful embodiments such as a low-temperature method of forming an oxide spacer on a doped gate. However, Dubuzinsky et al. do not disclose anything about the uniformity of the oxide spacer as a function of doping concentration in the gate, nor do they teach anything about the uniformity of the oxide spacer as a function of gate doping type.
U.S. Pat. No. 5,443,863 to Neely et al. describes a low temperature plasma assisted oxidation process. The plasma is created up stream of the processing zone with a microwave plasma electrical discharge. Neely et al. teach that such oxidation process can be conducted at a temperature below 300° C. Neely et al. describe a useful embodiment where a silicon carbide film is oxidized at a low temperature. However, Neely et al. do not discuss the dopant effect of the disclosed process, nor do they describe application of their process to structures with varying high dopant level or type.
U.S. Pat. No. 5,946,588 to Ahmad et al. describes a method of forming gate oxide. Ahmad et al. teach a thermal oxidation process where a silicon surface is oxidized in a low-temperature sub-atmospheric ozone ambient. Inherently, gate oxides are grown on lightly doped substrates and are not subjected to the oxidation doping effect. Therefore, Ahmad et al. specify a preferred range of doping level of 3×10
16
to 5×10
17
cm
−3
, far lower than the range investigated in the present application (1×10
19
to 1×10
22
cm
−3
). Moreover, Ahmad et al. do not discuss the uniformity of the oxide film as a function of doping concentration in the substrate, nor do they discuss any other aspects of the doping effect.
U.S. Pat. No. 5,738,909 to Thakur et al. describes a method of forming thin oxides on a semiconductor substrate. Thakur et al. teach a method where a portion of the oxidation process is conducted in an ozone ambient in order to increase the oxide growth rate. In addition, Thakur et al. teach that ultraviolet radiation can speed up the oxidation process even further. Thakur et al. do not teach anything about the uniformity of the oxide film as a function of doping concentration in the substrate, nor do they discuss any other aspects of the doping effect.
U.S. Pat. No. 5,700,699 to Han et at describes a method of forming gate oxide for thin film transistor (TFT). The gate oxide is formed with plasma assisted oxidation. The plasma is created with the aid of electron cyclotron resonance electrical discharge (ECE). Inherently, gate oxides are grown on a lightly doped semiconductor and are not subjected to the oxidation doping effect. Even though Han et al do not specify a preferred range of doping level, the doping level of the transistor channel is known to be typically less than 5×10
18
cm
−3
, a range far lower than the one investigated in the present application (1×10
19
to 1×10
22
cm
−3
). Consequently, Han et al. do not teach anything about the uniformity of the oxide film as a function of doping concentration in the substrate, nor do they discuss any other aspects of the doping effect.
U.S. Pat. No. 5,238,849 to Sato describes a method of fabricating a bipolar transistor. Sato teaches a method of forming an oxide layer between the crystalline base and polycrystalline emitter. The layer is formed with oxygen ions resulting in a substoichiometric silicon oxide. Sato does not teach anything about the uniformity of the oxide film as a function of doping concentration in the crystalline base, nor does he discuss any other aspects of the doping effect.
U.S. Pat. No. 6,358,867 to Tews et al., the disclosure of which is incorporated herein by reference, describes an orientation independent oxidation of silicon. Tews et al. teach a thermal oxidation method where a substantially uniform silicon oxide film is grown on various crystallographic planes of silicon. The main advantage of the method is achieved by employing atomic oxygen as the main oxidizing agent. Tews et al. also teach that the orientation effect is mainly due to the different surface density of silicon atoms on different crystallographic planes. As a resul
Dokumaci Omer H.
Doris Bruce B.
Gluschenkov Oleg
LandOfFree
Method for differential oxidation rate reduction for n-type... 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 for differential oxidation rate reduction for n-type..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for differential oxidation rate reduction for n-type... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3125829