PolySi films deposited with and without oxygen doping using rapid thermal chemical vapor deposition (RTCVD) have been investigated. Experimental results show that RTCVD systems can be used to provide high deposition rates (900-1000 Å/min at 700 °C) for both oxygen-doped and non-oxygen-doped polySi films. The surface roughness of the RTCVD polySi film is about half that of conventional LPCVD polySi films. The surface roughness and grain size of the RTCVD polySi film can be further reduced using oxygen doping. The catastrophic breakdown strength for capacitors using oxygen-doped polySi electrodes are improved compared with the breakdown strength for capacitors using non-oxygen-doped polySi electrodes. Electrical resistivities of B, P and As doped samples of polySi films with oxygen doping are found to be larger than those of polySi films without oxygen doping. Resistivities of silicides formed on the oxygen-doped polySi samples are approximately the same for those of silicides formed on non-oxygen-doped polySi samples.

We report a systematic study of dopant diffusion behavior for thin gate oxides and polysilicon implanted gate structures. Boron behavior is emphasized and its behavior is compared to that of As+ and BF2+. Dopant activation is achieved by rapid thermal annealing. Test structures with 100 Å, 60 Å and 30 A gate oxides and ion implanted polysilicon gate electrodes were fabricated and characterized after annealing by SIMS, SEM, TEM, and C-V measurements. For arsenic implanted structures, no dopant diffusion through a gate oxide of 30 Å thickness and an annealing condition as high as 1100°C/10s was observed. For boron implanted structures, as indicated by SIMS depth profiling, structures annealed at 1000°C/10s exhibit a so-called critical condition for boron diffusion through a 30 Å gate oxide. Boron dopant penetration is clearly observed for 60 Å gate oxides at an annealing condition of 1050°C/10s. The flatband voltage shift can be as high as 0.56 volts as indicated by C-V measurements for boron penetrated gate oxides. However, 100 Å gate oxides are good diffusion barriers for boron at an annealing condition of 1100°C/10s. For BF2 implanted structures, the diffusion behavior is consistent with behavior reported in the literature.

In this study, we present characterization of Metal-Oxide-Semiconductor (MOS) capacitors fabricated on carbon (C14) implanted silicon substrates. Carbon was implanted at an energy of 50 keV with doses ranging from 1 × 1012 cm−2 to 4.1×1015 cm−2. Metal-Oxide-Silicon (MOS) capacitors were fabricated and used to determine the MOS capacitance-voltage (C-V) and capacitance-time (C-t) behavior. These measurements revealed a strong correlation between carrier lifetime and the carbon dose. Degradation in lifetime was observed for carbon dose levels as low as 4 × l012 cm−2. At carbon doses equal to and above 6.4 × l013 cm−2, extremely low generation lifetimes were obtained (∼ 10−7 sec). On the other hand, degradation in C-V characteristics was observed only for carbon doses above 2.7 × l014 cm−2. Below this dose, both flatband voltage and interface trap density of the carbon implanted samples were comparable to those of the monitors. Analysis of the samples by cross sectional transmission electron microscopy revealed the absence of extended defects even in samples with high carbon dose levels.

In this paper, we explore selective Si epitaxy by UltraHigh Vacuum Rapid Thermal Chemical Vapor Deposition (UHV-RTCVD) using Si2H6, H2, and Cl2 with particular emphasis on selectivity robustness. Two key parameters considered in this study were partial pressures of Si2H6 and H2. It was found that excessive increases in either partial pressure could lead to selectivity degradation. The two mechanisms by which the observed selectivity degradation can be explained are as follows: A higher Si2H6 partial pressure provides a larger flux of Si atoms which directly influences the probability of reaching the critical nuclei size for stable nuclei formation while an increase in H2 partial pressure reduces the desorption rate of Si adatoms from the insulator surface by reducing the available Cl in the gas phase for SiCl2 formation. The impact of process parameters on epitaxial defect density was also evaluated using darkfield imaging. The results clearly indicate increasing defect density upon increases in both the chlorine flow rate and the level of contamination introduced through the silicon source gas.

Silicon nucleation on silicon dioxide and selective silicon epitaxial growth (SEG) were studied in an ultrahigh vacuum rapid thermal chemical vapor deposition (UHV‐RTCVD) reactor using 10% diluted in . Silicon was deposited on patterned Si (100) substrates over a pressure range of 10–100 mTorr and a temperature range of 650 and 850°C. Under these conditions, the growth rate ranged from 50 to 330 nm/minute, demonstrating compatibility with single wafer manufacturing throughput requirements. A pressure dependence in the activation energy in the surface reaction limited regime was observed and attributed to a variation in the steady‐state hydrogen coverage on the growing surface. The incubation time for loss of selectivity via Si nucleation on was found to increase at lower pressure and remained constant over the experimental temperature range. However, the incubation thickness defined as the film thickness that can be deposited before loss of selectivity occurs was found to increase both at low pressures and high temperatures. We show that a 100 nm thick epitaxial film can be grown selectively with no Cl addition at 750°C/10 mTorr.

The research focus of this paper is the result of collaborations between NC State University and Four Dimensions, Inc. As a result of this partnership, the challenge of probing a patterned wafer with a Hg-probe system that was engineered by Four Dimensions to probe unpatterned wafers emerged. To address this challenge, a novel mask was designed, fabricated, and used to obtain patterned arrays of field-oxide isolated gate oxides as well as so-called "L" strips of gate oxide. Mercury-gate capacitance-voltage measurements from three sites (i.e., unpatterned gate oxides and field oxides plus gate oxides as well as patterned field oxides/gate oxides) were used to validate a relatively simple algorithm for the extraction of oxide thickness. In addition, optical oxide thickness was extracted from the "L" strips, and excellent agreement with the electrical value is demonstrated. This "single wafer" methodology for extracting electrical and optical metrics will have a significant impact on process development of advanced gate dielectrics within the SRC/SEMATECH Center for Front End Processes.

Transmission electron microscopy (TEM) was used to characterize image contrast obtained from doping-dependent etching of p-n junctions in silicon. The local variations in crystal thickness give rise to the appearance of thickness fringes which may be interpreted as two-dimensional iso-concentration contours that map the dopant distribution. The samples used for the study consisted of solid source diffusions of boron into substrates of varying resistivities of both n- and p-type. The factors which affect the interpretation of dopant profiles obtained from selective chemical etching of cross section TEM samples is addressed. One-dimensional chemical dopant concentration data were derived from secondary ion mass spectroscopy and one-dimensional carrier concentration data were derived from spreading resistance profiling.

In May of 1996, the Engineering Research Center for Advanced Electronic Materials Processing (ERC for AEMP) at North Carolina State University (NCSU) received funding from the Semiconductor Research Corporation (SRC) to pursue what is called the SRC Summer '96 Continuous Quality Improvement Initiative. As a result of this situation, three graduate students were selected to participate in the project and a number of researchers from the ERC-AEMP were brought on board to support the effort. The objective of the project was to continuously improve the quality of an in situ gate-stack process for the across-stage fabrication of MOS capacitors. The research pathways to achieve this objective were process characterization as well as definition, materials analysis and statistical analysis. This paper will discuss the key issues in this effort.

We have previously reported a process for low temperature selective silicon epitaxy using Si2H6, H-2, and Cl-2 in an ultrahigh vacuum rapid thermal chemical vapor deposition reactor.(1) Selective deposition implies that growth occurs on the Si surface but not on any of the surrounding insulator surfaces. Using this method and process chemistry, the level of Cl species required to maintain adequate selectivity has been greatly reduced in comparison to SiH2Cl2-based, conventional CVD approaches.(2,3) In this report, we have extended upon the previous work and provide information regarding the selectivity of the silicon deposition process to variations in the growth conditions. We have investigated the selectivity of the process to variations in disilane flow/partial pressure, growth temperature, and system contamination. We demonstrate that increases in either the Si2H6 partial pressure or flow rate, the process temperature, pr the source contamination levels can lead to selectivity degradation. In regard to the structural quality of the selective epitaxial layers, we have observed epitaxial defects that have appeared to be a strong function of two basic conditions: the contamination level of the process and the chlorine flow rate or chlorine partial pressure. Overall, the results in this study indicate several process conditions that can inhibit the quality of a selective silicon deposition process developed for single-wafer manufacturing.

Doping-dependent contrast in secondary electron images of p/n junctions in silicon obtained in a field-emission scanning electron microscope was observed and characterized. The optimum experimental conditions for observing this ‘‘electronic’’ contrast were established by investigating the effect of microscope and material parameters on the magnitude of the contrast. The contrast between the bright p-type areas and the darker n-type areas was maximized at an accelerating voltage of ∼1 kV, and when a through-the-lens detector configuration was employed. Secondary electron contrast profiles of boron doped p+/n junctions in silicon showed a good correlation with secondary ion mass spectroscopy depth profiles of the atomic concentration down to the 1017 cm−3 level. However, similar results were not obtainable for n+/p junctions. It is demonstrated that this contrast effect may be exploited for obtaining two-dimensional dopant profiles directly from secondary electron images of p+/n junctions provided that the technique is empirically calibrated against a one-dimensional dopant profiling method.

The kinetics of the nucleation and growth of Si films on amorphous SiO2-covered Si using rapid thermal chemical vapor deposition from SiH4 and Si2H6 (5% in He) were compared at temperatures between 600 and 800 °C and reactant gas pressures between 1 and 25 mTorr. Quantitative assessment of the nucleation parameters and the structures of the deposited Si films have been determined using in situ real time single wavelength and spectroscopic ellipsometry. In addition to ellipsometry, atomic force microscopy, scanning electron microscopy, and cross-sectional transmission electron microscopy were used ex situ to observe the nucleation stage and the microstructures of the films. This study compares the initial growth parameters for SiH4: nuclei density (6×108 cm−2), nuclei size (94 nm), incubation time (4.2 min), and degree of selectivity (42 nm) with those for Si2H6: 1.3×1010 cm−2, 31 nm, 0.4 min, and 10 nm, respectively. The incubation times for SiH4 and Si2H6 are different, as is the degree of selectivity, but they show similar activation energies of about 1 eV in the 600–800 °C range. The Si film quality in terms of surface roughness and grain structure was better for the Si film derived from Si2H6 than from SiH4.

This investigation is concerned with the influence of a vacuum prebake on oxygen and carbon levels at epitaxial silicon/silicon (100) interfaces. The epitaxial layers are deposited in an ultrahigh vacuum, rapid thermal reactor using chemical vapor deposition techniques. Secondary ion mass spectroscopy (SIMS) is used to evaluate carbon and oxygen levels at the epitaxy/substrate interface. We show that a vacuum prebake can be effectively used following a standardex situ clean that consists of an RCA clean, a dilute (5%) dip, and a rinse in deionized water. The results show that if epitaxial deposition is initiated by introducing the reactive gases into the chamber at the prebake temperature, oxygen and carbon levels below the sensitivity limits of secondary ion mass spectroscopy are obtained at the epitaxy/substrate interface. This result can be reproducibly achieved with a low thermal budget prebake of even after a relatively long rinse (∼300 s) in deionized water. We propose that the mechanism responsible for cleaning is thermal desorption of oxygen and hydrocarbons from the (100) surface of silicon. We show that the surface obtained with this ex situ clean is very stable and, hence, the wafer can be left in a clean ultrahigh vacuum environment for many hours without detectable changes in the oxygen and carbon levels. On the other hand, results indicate that when the prebake is terminated by cooling the wafer to the ambient temperature of the reactor, carbon is readsorbed on the silicon surface at a peak concentration of . We also show that when a small amount of hydrogen is introduced into the reactor during the prebake, a higher thermal budget is required to remove oxygen from the surface. This observation is attributed to a higher background associated with the presence of hydrogen. It is concluded that vacuum prebake is an attractive surface preparation technique which effectively reduces oxygen and carbon levels on a silicon (100) surface below the SIMS sensitivity limits.

In this paper, a novel raised p+−n junction formation technique is presented. The technique makes use ofin- situ doped, selectively deposited Si0.7Ge0.3 as a solid diffusion source. In this study, the films were deposited in a tungsten halogen lamp heated cold-walled rapid thermal processor using SiCl2H2, GeH4, and B2H6. The microstructure of the Si0.7Ge0.3 layer resembles that of a heavily defected epitaxial layer with a high density of misfit dislocations, micro-twins, and stacking faults. Conventional furnace annealing or rapid thermal annealing were used to drive the boron from thein- situ doped Si0.7Ge0.3 source into silicon to form ultra-shallow p+−n junctions. Segregation at the Si0.7Ge0.3/Si interface was observed resulting in an approximately 3:1 boron concentration discontinuity at the interface. Junction profiles as shallow as a few hundred angstroms were formed at a background concentration of 1017 cm−3.

We present the use of the Si2H6/H2/CL2 chemistry for selective silicon epitaxy by rapid thermal chemical vapor deposition (RTCVD). The experiments were carried out in an ultrahigh vacuum rapid thermal chemical vapor deposition reactor. Epitaxial layers were grown selectively with growth rates above 150 nm/min at 800 °C and 24 mTorr using 10% Si2H6 and H2 and Cl2 with a minimum Si:Cl ratio of 1. Excellent selectivity with respect to SiO2 and Si3N4 was obtained indicating that very low Cl2 partial pressures are sufficient to preserve selectivity. In situ doping results with B2H6 show that sharp doping transitions and a wide range of B concentrations can be obtained with a slight B incorporation rate reduction with Cl2 addition. Our results indicate that UHV-RTCVD with the Si2H6/H2/Cl2 chemistry yields highly selective Si epitaxy with growth rates well within the practical throughput limits of single wafer manufacturing and with a potential to reduce the Cl content below the levels used in conventional SiH2Cl2 based selective epitaxy processes.

In this Letter, we report our results on surface preparation, involving in situ cleaning and passivation for low-temperature Si epitaxy in a multichamber cluster tool. The experiments were carried out in a three-chamber reactor which mimics a cluster tool. The results indicate that residual O on the dilute HF-treated Si surface (ex situ cleaned) can be reduced below the detection limit of secondary ion mass spectroscopy (SIMS) by in situ baking at 750°C for 15 s in an ultra-high vacuum environment or in H2 (pressure = 240 mTorr). We show that the extremely reactive Si surface can be passivated against recontamination by exposing it to a low-pressure Si2H6 environment at the ambient temperature immediately following the in situ clean. When the unpassivated samples are exposed to an air pressure of 10−6 Torr in the load-lock, O adsorbs on the surface up to 50% of a monolayer within 10 min. Under the same conditions, with passivation, the oxygen levels remain below the detection level of SIMS. Surface passivation will be extremely useful in applications that require wafer transfer between chambers such as in multichamber cluster tools.

A study of Si nucleation and deposition on SiO2 was performed using disilane and hydrogen in an ultra high vacuum rapid thermal chemical vapor deposition reactor in pressure and temperature ranges of 0.1 – 1.5 Torr and 625 – 750°C. The film analysis was carried out using scanning electron microscopy, transmission electron microscopy and atomic force microscopy. At lower pressures, an incubation time exists which leads to a retardation in film nucleation. At 750°C, the incubation time is 10s at 0.1 Torr and decreases to less than Is at 1.5 Torr. The nuclei grow and form three dimensional islands on S1O2, and as they coalesce, result in a rough surface morphology. At higher pressures, the inherent selectivity is lost resulting in a higher nucleation density and smoother surface morphology. For ˜ 2000 Å thick films, the root-mean-square surface roughness at 750ÅC ranges from 110Å at 0.1 Torr to 40Å at 1.5 Torr. Temperature also strongly influences the film structure through surface mobility and grain growth. At 1 Torr, the roughness ranges from 3Å at 625°C to 60Å at 750°C. The grain structure at 625°C/1Torr appears to be amorphous, whereas at 750°C the structure is columnar. The growth rate at 625°C/1.5 Torr is 1200 Å/min provides a surface roughness on the order of atomic dimensions which is comparable to or better than amorphous silicon deposited in LPCVD furnaces.