Semiconductor device manufacturing: process – Making field effect device having pair of active regions... – Having insulated gate
Reexamination Certificate
1998-12-02
2002-04-16
Wilczewski, Mary (Department: 2822)
Semiconductor device manufacturing: process
Making field effect device having pair of active regions...
Having insulated gate
C438S528000
Reexamination Certificate
active
06372591
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a fabrication method of a semiconductor device and more particularly, to a fabrication method of a semiconductor device using an ion-implantation technique, which is applicable to formation of shallow source/drain regions (e.g., 0.1 &mgr;m or less in depth) of a Metal-Oxide-Semiconductor Field-Effect Transistor (MOSFET) with the double drain structure.
2. Description of the Prior Art
In recent years, semiconductor devices have been becoming miniaturized more and more according to the increase in their integration scale. Under such the circumstances, a lot of electronic devices such as memory or logic devices have been integrated on a semiconductor substrate or chip. In these highly-integrated semiconductor devices, i.e., Large-scale Integrated circuits (LSIs), typically, n- and p-channel MOSFETs are used to constitute the Complementary MOS (CMOS) structure.
To cope with the progressing miniaturization and increasing integration tendency, there has been the need to solve the problems induced by the short-channel effects in MOSFETs. A known solution to the problems is to decrease the depth of source/drain regions of the MOSFETs (i.e., to use shallow source/drain regions). However, the depth decrease of the source/drain regions causes another problem that the sheet resistance of the source/drain regions is increased and simultaneously, the contact resistance of the source/drain regions with overlying wiring materials is also increased.
Another known solution to the above problems is to use the double drain structure of the source/drain regions of the MOSFET, where each of the source/drain region is formed by two parts, i.e., a shallow and high doping-concentration part located near the gate electrode and a deep and low doping-concentration part located apart from the gate electrode. The shallow and high doping-concentration part, which is located below a sidewall spacer of the gate electrode, is termed the “extension” of the source/drain region.
The shallow and high doping-concentration part contributes to the depth decrease in a source/drain region and at the same time, the deep and low doping-concentration part serves to prevent the sheet resistance and contact resistance of the souce/drain region from increasing. Accordingly, the double drain structure not only contributes to improvement in the driving capability of the MOSFET but also solves the above-described problem of the sheet resistance increase and the contact resistance increase of the source/drain region.
Recently, as a new doping method to form shallow p-n junctions or shallow source/drain regions (e.g., 0.1 &mgr;m or less in depth), solid-phase diffusion, gas-phase diffusion, plasma doping, and laser doping methods have been developed and examined. These new methods are especially effective in the doping process of boron (B) that has been usually employed as a p-type dopant in the silicon LSIs. Each of these new methods serves as an alternative of the popular ion-implantation method. This is because boron has a small mass and a high diffusion rate in silicon (Si) and as a result, shallow boron-doped regions are difficult to be formed by the popular ion-implantation method.
Thus, the popular ion-implantation method has a disadvantage that shallow doped regions or shallow p-n junctions (e.g., 0.1 &mgr;m or less in depth) are difficult to be formed. This is caused by the following main reason.
Specifically, due to the ion-implantation process of a dopant, crystal defects termed the point defects such as vacancies and interstitial atoms of Si are induced in a single-crystal Si substrate. Since the implantation-induced point defects enhances the diffusion of the implanted dopant atoms, the resultant profile and concentration of the implanted dopant atoms tend to deviate from its desired profile and concentration through a subsequent annealing process for the activation purpose.
On the other hand, the popular ion-implantation method has an advantage that the uniformity, reproducibility, controllability, and throughput are better than those of the above-identified new doping methods, i.e., the solid-phase diffusion, gas-phase diffusion, plasma doping, and laser doping methods, in the case of formation of the source/drain regions of the MOSFET. Therefore, if the generation of the implantation-induced point defects is suppressed and at the same time, the enhanced diffusion phenomenon of the dopant atoms occurring in the annealing process is well controlled, it is expected that the ion-implantation method is able to fully cope with fabrication of the future LSIs.
Additionally, the enhanced diffusion of the implanted dopant atoms, which is one of the disadvantages of the ion-implantation method, affects not only the implanted dopant atoms for the source/drain regions but also the dopant atoms existing in the channel region of the MOSFET.
For example, when the souce/drain regions with the double drain structure are formed, the deep and low doping-concentration parts of the source/drain regions, which are located apart from the gate electrode, need to have a depth enough for preventing the sheet resistance and the leakage current from increasing after (i) the silicidation process of the source/drain regions using a refractory metal such as titanium (Ti) and (ii) the formation process of the contact regions with an overlying wiring metal film. As a consequence, to form the deep and low doping-concentration parts, the same dopant as that of the shallow and high doping-concentration parts is selectively ion-implanted into the Si substrate. This ion-implantation process for the deep and low doping-concentration parts is carried out after the shallow and high doping-concentration parts (which are termed the “extensions” and located near the gate electrode) have been formed by selective ion-implantation and then, a pair of sidewall spacers have been formed at each side of the gate electrode.
In this case, the implanted dopant atoms existing in the shallow and high doping-concentration parts (i.e., the extensions) tends to be affected by the enhanced diffusion phenomenon of the implanted dopant atoms into the deep and low doping-concentration parts. This leads to the short-channel effects. Also, the distribution of the dopant atoms existing in the channel region may be affected by the same enhanced diffusion phenomenon.
Accordingly, to improve the controllability of the profile and concentration of the implanted dopant atoms in the source/drain regions with the double drain structure, it is important to control the enhanced diffusion phenomenon of the implanted dopant atoms into the deep and low doping-concentration parts. In other words, it is effective to find the ion-implantation condition minimizing or decreasing the generation of the implantation-induced crystal defects in the Si substrate.
Conventionally, as an improvement of the ion-implantation method to form the shallow junctions of the source/drain regions, the “preamorphized ion-implantation” method has been developed. In this improved method, to prevent the channeling phenomenon of the implanted boron atoms from occurring, the main surface of a single-crystal Si substrate is amorphized by an ion-implantation process of germanium (Ge) or Si to thereby form an amorphous region at the implanted main surface of the substrate prior to an ion-implantation process of boron (B). This is because the implanted boron atoms tend to be reached to a deeper level than their actual projected range in the Si substrate due to the channeling phenomenon because of its small mass. Due to the amorphous region thus formed, the channeling phenomenon of the boron atoms is effectively prevented.
With the conventional preamorphized ion-implantation method, as described above, the channeling phenomenon of the implanted boron atoms is effectively prevented. However, point defects are induced in the amorphous region by the preamorphizing ion-implantation process of Ge or Si. Accordingly, there is a disadvantage that
Mineji Akira
Saito Shuichi
Shishiguchi Seiichi
Hayes, Soloway, Hennessey Grossman & Hage, P.C.
Wilczewski Mary
LandOfFree
Fabrication method of semiconductor device using ion... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Fabrication method of semiconductor device using ion..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Fabrication method of semiconductor device using ion... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2927469