In order to improve the yield enhancement speed, we adopt data mining analysis. The root cause of crack in silicon substrate induced by scrubber tool was found by data mining approach. It could not be detected by conventional knowledge-based pre-filtering approach. As a countermeasure, we switched the scrubber from high pressure clean tool to moderate pressure clean tool.

New semiconductor technologies for 300 mm wafer processing have been developed to meet the requirements for the next generation device fabrication. One of the strong candidates for supporting 300 mm semiconductor manufacturing is enlarged microwave plasma deposition technology. The polysilicon films deposited with enlarged microwave plasma technology were studied in this work. The deposited polysilicon films with the enlarged microwave plasma technology showed a variety of crystallinity as well as epitaxy on silicon substrates even at 430 degrees C substrate temperature. This low temperature deposition is very promising for application to the 300 mm semiconductor manufacture since the more crystallinity guarantees the better device performance. It was shown that the slower deposition rates promoted the higher crystallinity. It was also observed that the crystallographic orientations of the polysilicon grains changed gradually during deposition, and this was modeled and explained. It was therefore concluded that the enlarged microwave plasma deposition technology will be an important part of the future technologies for the giga dynamic random access memory era. (C) 2000 The Electrochemical Society. S0013-4651(99)11-035-8. All rights reserved.

Polycrystalline silicon is grown at a temperature of 300 °C by microwave-excited plasma enhanced chemical vapor deposition using SiH4/Xe. The grain size measured by x-ray diffraction is about 25 nm. High-density (>1012 cm−3) plasma having very low electron temperature (<1 eV) is excited by microwave irradiation using radial line slot antenna. We present the implementation of this system for the growth of poly-Si. Low-energy (3 eV), high-flux ion bombardment utilizing xenon ion on a growing film surface activates the film surface and successfully enhances surface reaction/migration of silicon, resulting in high quality film formation at low temperatures.

In this paper, a new methodology for effective process excursion monitoring using defect review/classification information is proposed. We introduce a new defect classification scheme, in which relevant defect types that are likely to be caused by the same mechanism or source are grouped into a "defect family". It is demonstrated that trending by the defect family drastically improves the detection efficiency of killer defect excursion by reducing or eliminating noise resulting from irrelevant benign defects. We compare the risks of missing critical excursions for monitoring by total defect count, killer defect count, and killer defect family, and illustrate the effectiveness of our methodology using data from actual fabline.

We have experimentally shown for the first time that threshold energies of plasma-induced deactivation for phosphorus, boron and antimony in silicon epitaxy by using a low-energy ion bombardment process [l-4]. The deactivation energy of phosphorus, boron and antimony at a growing silicon film surface is -13 eV, -5 eV and -10 eV respectively as shown in Table 1. Since the deactivation energy of boron is extremely small (< 5 eV), ion bombardment energy must be precisely controlled to be lower than 5 eV in order to make the activation ratio of dopants 100 %. The experimental results of plasma-induced deactivation energy of dopants will be crucial value for plasma processing, especially for low temperature processing using ion bombardment processes.

Due to the advances in in-line inspection technology it is now possible to obtain an early in-line prediction of yield. This paper introduces and compares two new in-line yield prediction methodologies: (1) multilayer critical area method and (2) defect-type-size kill-ratio method. These methods are more accurate than the past and other current approaches used in the semiconductor industry. The first method uses the design layout information along with the in-line defect data, whereas the second method uses the defect and yield data to empirically derive the kill-ratios. We demonstrate our methodologies using data collected in a real wafer fabrication facility at the polysilicon gate (Poly), and the first and second interconnect (Metal 1 and Metal 2) post etch inspection layers. We compare our in-line predictions with the actual yield.

For a low-energy (<30 eV) ion bombardment process, the effect of phosphorus concentration on low-temperature (350–400°C) silicon epitaxial growth is reported. The conditions of ion energy and ion flux required for realizing low-temperature epitaxial growth were precisely investigated. We found that phosphorus doping significantly enhanced silicon epitaxial growth. It is difficult to realize high-quality film growth with lightly phosphorus-doped silicon. However, large-mass, large-radius ion (xenon) bombardment is quite effective for improving the quality of silicon film with lightly phosphorus-doped silicon.

In-line monitoring of defects in VLSI manufacturing has become an indispensable tool in SPC (Statistical Process Control) and Yield Management. Wafer inspection for defects has two stages: optical wafer scanning to detect the presence of defects; and review/classification performed at the coordinates of the scanned defect to determine the defect type (for instance, shorts or opens). Typically, fabs use standard control charts, based on total defect count monitoring. However, many fabs have found this aggregate tracking of defects to be inadequate for efficient excursion or drift detection. In this paper, we demonstrate how defect type information can be utilized to optimize the excursion detection procedure. In addition, we also demonstrate how the defect type information can be useful for source identification

This paper discusses the importance of understanding and modeling the inter and intra layer defect propagation for in-line yield prediction. Some examples using real fabline data are presented to illustrate the significance of this problem.

In this paper, new methodologies for effective process excursion monitoring and defect source isolation are proposed. We introduce a new defect classification scheme, in which relevant defect types that are likely to be caused by the same mechanism or source are grouped into a "defect family." We demonstrate that trending by the defect family drastically improves the excursion detection efficiency without suffering noise from irrelevant benign defects. Furthermore we have developed a methodology for identifying the source of the excursion using defect type Pareto. This is based on the fact that the signature of defect type Pareto leads to the defect source information and thus possibly indicates the origin of the problem. Thus both process control and excursion source identification can be achieved simultaneously by effective defect classification.

We have shown for the first time that the use of large mass ions in low energy ion bombardment process is quite effective in low-temperature silicon epitaxy. By using Xe ions (mass=131) instead of Ar ions (mass=40), the minimum ion bombardment energy for 300/spl deg/C epitaxy has been drastically reduced from 20 eV to 7 eV, thus minimizing the formation of defects. It is also experimentally shown that the energy dose determined by the product of ion energy and ion flux is a key parameter for epitaxy that compensates for the reduction in the substrate temperature. Low-energy, high-flux, large-mass ion bombardment is the direction for further reducing the processing temperature while presenting high crystallinity of grown films.

A technique for formation of abrupt and arbitrary carrier depth profile in epitaxial silicon film deposited at temperatures as low as 300° C has been developed utilizing an RF-DC coupled-mode bias sputtering system. The effect of ion bombardment energy and ion flux density on the quality of silicon film has been investigated. It was found that the activation level of dopants in the grown silicon film can be widely varied by appropriately controlling the ion bombardment energy while maintaining excellent crystallinity. For controlling the carrier concentrations in the deposited silicon film with an appropriate degree of accuracy, the ion bombardment energy must be precisely controlled with a degree of accuracy higher than 0.5 V, since the properties of the deposited films are very sensitive to the ion bombardment energy. For this purpose, a technique for measurement of the plasma potential is required. While the Langmuir probe technique has been applied to the measurements in DC and also RF discharges, its use in RF discharges has been hindered by the problem of RF interference. Therefore, we have developed advanced Langmuir probes for plasma diagnoses in RF discharges at the plasma excitation frequency of 200 MHz.