Silane Consumption Research Articles

Study of high-tech process furnace using inherently safer design strategies (II). Deposited film thickness model

In the low-pressure chemical vapor deposition (LPCVD) process, there are many factors that could affect the thin film thickness and homogeneity in the thin film deposition process. Besides temperature, the volume flow rate of the reactant gas, the Mole fraction, and pressure are all affecting factors. On the basis of the forecast mode result of the thermal model of wafer temperature distribution in the previously published article, this study used the optimal method to calculate the conversion rate of the surface of each wafer under simulated conditions of various furnace temperatures and Mole fractions, as well as silane flow rates and pressures, and finally calculated the thin film growth rate and homogeneity of the surface of each wafer.This study validated four types of process conditions, and found that the results were very close to the experimental values in the previous studies, when conditions such as temperature, Mole fraction of silane, flow rate and pressure in the process were changed. This study thus concluded that the proposed process film thickness mode could meet the requirement of the process capacity. This study found that adjusting the heating zone temperature to regulate the film thickness growth rate was able to control uniformity of the film thickness. These feasible adjustment measures could be very effective on the industry commercial processes.Furthermore, the film thickness mode established in this study was used to discuss the application level of the inherently safer design strategies. This study mode could be used to adjust the temperature, mole fraction (mf), flow rate (F) and pressure (P) in the process for a single machine. Under the condition of meeting the film thickness process recipes required by the process, the silane flow rate was lowered through the adjustment of the control system. The study took the silane supply system as the basis of the case study. Based on a daily required silane consumption of 64.4 L, the daily required amount was 1.5 cylinders in the case of using 40 L high-pressure cylinders. The total daily required amount would be three 40 L high-pressure cylinders, including one more reserve amount on top of the daily required amount, which could match the inherently safer design philosophy and the principles of intensification, attenuation and limitation of effects.In terms of the factory evaluation, among all of the 12-inch furnace machines in Taiwanese wafer factories, nine out of 47 sets use silane, and the total daily consumption reaches 579.6 L. The total required amount, including the reserve, is 27 cylinders for 40 L high-pressure cylinders. As for the setting of the gas cabinet, the required amount could be reduced, since the gas supply pipeline could be designed in a multiple circuit. The minimum reactant gas input amount could be obtained in combination with the proposed film thickness mode, and the safety could be improved greatly, in accordance with the inherently safer design philosophy and the principles of intensification, attenuation and limitation of effects. In addition to the process capacity, this study also took the elevation of the inherently safer level into consideration, which could be provided to the industry for more comprehensive design consideration.

Improve silane utilization for silicon thin film deposition at high rate

In order to investigate the silane utilization during silicon thin film deposition, one quadrupole mass spectrometer was used to monitor the partial pressure of silane during high pressure and high power deposition process. Relationship of silane consumption with silane concentration, excitation power and reaction pressure was investigated. The results show that increasing silane concentration leads to lower silane consumption. Increasing excitation power can improve the silane utilization, which is proved by higher deposition rate. However, when the power is higher than a certain value, much more power will consume by the transmission line and matching network. The power coupled into the plasma saturates to a certain value and then no more silane decomposes any more. In addition, the silane consumption decreased and then increased with increasing reaction pressure, which is related with the voltage of electrodes. These results could help to improve silane utilization for low cost production and better quality of film.

Mechanistic study of nickel-catalyzed ynal reductive cyclizations through kinetic analysis.

The mechanism of nickel-catalyzed, silane-mediated reductive cyclization of ynals has been evaluated. The cyclizations are first-order in [Ni] and [ynal] and zeroth-order in [silane]. These results, in combination with the lack of rapid silane consumption upon reaction initiation, are inconsistent with mechanisms involving reaction initiation by oxidative addition of Ni(0) to the silane. Silane consumption occurs only when both the alkyne and aldehyde are present. Mechanisms involving rate-determining oxidative cyclization to a metallacycle followed by rapid reaction with the silane are consistent with the data obtained.

Total SiH4/H2 pressure effect on microcrystalline silicon thin films growth and structure

Abstract The effect of the total SiH4/H2 gas pressure (1–10 Torr) on the growth rate, the film crystallinity and the nature of hydrogen bonding of microcrystalline silicon thin films deposited by 13.56 MHz plasma-enhanced chemical vapour deposition (PECVD) was investigated under well-controlled discharge conditions. The deposition rate presents an optimum for 2.5 Torr, which does not follow the trend of silane consumption that increases with pressure and is attributed to an increase in plasma density. The film crystallinity increases with pressure from 1–2.5 Torr and then remains almost the same, whereas the films deposited at 1 Torr are highly stressed. On the other hand, hydrogen bonding is also drastically affected.

Temperature Dependence and Silane Consumption During Particle Formation in Ar–Silane RF Capacitively Coupled Plasma

We propose a generalized model, based on a simple balance equation, explaining the temperature dependence of the agglomeration time of nanoclusters in Ar-SiH/sub 4/ plasmas. This model allows easy incorporation of specific mechanisms and verification of their effect on particle growth. We consider extra silane consumption due to back diffusion, an important issue for the laboratory plasma experiments and modeling results interpretation.

Gas phase and surface kinetics in plasma enhanced chemical vapor deposition of microcrystalline silicon: The combined effect of rf power and hydrogen dilution

A gas phase and surface simulator of highly diluted silane in hydrogen rf discharges used for the deposition of microcrystalline silicon has been developed. The model uses the spatial density distribution of SiH (X 2Π) radicals measured using laser induced fluorescence and the total silane consumption for estimating the primary electron induced silane dissociation, thus avoiding fluid or statistical approaches commonly used for the prediction of electron impact rate coefficients. A critical analysis is made for the relative importance of all the parameters involved either in the gas phase chemistry or in the surface processes. The model results are compared to experimental data concerning disilane production and film growth rate over a wide range of rf power densities in 2% and 6% SiH4 in H2 discharges. The good agreement between experimental and model results allows for the extension of the discussion to the composition of the radical flux reaching the substrate, the relative contribution of each of the radicals to the film growth, and the most probable mechanism of microcrystalls formation under typical conditions of low and high microcrystalline silicon deposition rate.

Electron-impact silane dissociation and deposition rate relationship in the PECVD of microcrystalline silicon thin films

An investigation of the relation between silane electron impact dissociation and deposition rates of microcrystalline silicon thin films, has been performed in highly diluted SiH 4 in H 2 discharges, by applying a combination of experimental measurements and modeling of the process. A wide range of frequencies (13.56 MHz - 50 MHz), power densities (11 mW/cm 2 -162 mW/cm 2 ) and silane partial pressures (2% - 6%) at two total SiH 4 /H 2 pressures of 0.5 Torr and 1 Torr has been studied. In the lower pressure, independent of all other discharge parameters, SiH 4 primary dissociation has been found to be responsible for about 70% of the total silane consumption in the discharge, while a fraction of 12% of the initially produced silicon hydrides are incorporated into the growing film. The increase of pressure leads to a drop of the contribution of the SiH4 primary dissociation to the total silane consumption and to an increase of the deposition efficiency of the initially produced radicals to a value of 26%. This result is attributed to the production of additional, two silicon atom precursors via secondary gas-phase reactions.

Silane consumption and conversion analysis in amorphous silicon and silicon nitride plasma deposition using in situ mass spectroscopy

In situ mass spectroscopy is used to sense plasma deposition of silicon and silicon nitride, to analyze gas phase reactant depletion, and the efficiency of silane conversion into the thin film. A double-differentially pumped quadrupole mass spectrometer was used to monitor the SiH4, Si2H6, and H2 effluent from 100% SiH4 and 2% SiH4/He silicon deposition, and SiH4/He/N2 silicon nitride deposition processes. No significant changes in gas phase nitrogen related species were observed during nitride deposition. However, the Si species show significant process dependence allowing reaction analysis. Disilane species were produced at low powers in the SiH4/He/N2 process, but no amino–silane species were observed. The silane consumption and silicon incorporation efficiency are shown to depend on gas dilution, residence time, and reactor geometry.

Studying of silane thermal decomposition mechanism

The mechanism of silane thermal decomposition is investigated in a flow reactor. The time dependencies of silane consumption and disilane formation were compared with those parameters of solid product (aerosol particles) such as concentration, total hydrogen content in solid product, and fraction of hydrogen contained in solid product as polyhydride groups (SiH2)n. Silane loss and gaseous product formation were analyzed using a mass spectrometer. The hydrogen content in solid product was analyzed by the methods of IR-spectroscopy and hydrogen evolution. Based on a simple kinetic scheme we qualitatively explained the experimental dependencies of silane conversion and disilane formation, the effective activation energy of the decomposition process, and the amount of polyhydride groups in the solid product on reaction time and initial silane concentration. © 1998 John Wiley & Sons, Inc. Int J Chem Kinet 30: 99–110, 1998.

Studying of silane thermal decomposition mechanism

The mechanism of silane thermal decomposition is investigated in a flow reactor. The time dependencies of silane consumption and disilane formation were compared with those parameters of solid product (aerosol particles) such as concentration, total hydrogen content in solid product, and fraction of hydrogen contained in solid product as polyhydride groups (SiH2)n. Silane loss and gaseous product formation were analyzed using a mass spectrometer. The hydrogen content in solid product was analyzed by the methods of IR-spectroscopy and hydrogen evolution. Based on a simple kinetic scheme we qualitatively explained the experimental dependencies of silane conversion and disilane formation, the effective activation energy of the decomposition process, and the amount of polyhydride groups in the solid product on reaction time and initial silane concentration. © 1998 John Wiley & Sons, Inc. Int J Chem Kinet 30: 99–110, 1998.

Arrhenius-like behavior in plasma reactions

An Arrhenius-like behavior of the silane dissociation rate constant as a function of the power actually fed in a rf discharge system is reported. The rate constant is calculated using mass-spectrometric measurements of silane consumption for 50 and 75 mTorr discharges, while the power consumed in the discharge is determined by using detailed voltage and current measurements. The slope of the natural logarithm of the dissociation rate constant as a function of the inverse of the discharge power increases with pressure. This is attributed to a change in the shape and/or the mean energy of the electron energy distribution function, while the linearity indicates an analogous increase of the active electron concentration with increasing discharge power, excluding a significant change in the shape of the electron energy distribution function.

Kinetics of Power Deposition and Silane Dissociation in Radio‐Frequency Discharges

Measurements of the actual power and silane consumption as a function of excitation voltage and pressure in a completely characterized radio‐frequency, glow‐discharge chamber used for the deposition of a‐Si:H, are reported. The influence of these parameters on the discharge power and impedance is presented and discussed with consideration of the basic electron heating and power dissipation mechanisms. The discharge is more efficient, in terms of the amount of energy spent per dissociated silane molecule, at higher pressures and lower voltages. This amount tends toward the dissociation threshold of silane for lower power densities. The rate constant of electron impact dissociation of silane, calculated from silane consumption, increases with power with a different trend for each pressure, and is higher for lower pressures ranging from 0.08 to .

Real-time process sensing and metrology in amorphous and selective area silicon plasma enhanced chemical vapor deposition using in situ mass spectrometry

We have used mass spectroscopy to observe and analyze, in real-time, gas phase reactants and product species in plasma enhanced chemical vapor deposition (PECVD) of silicon. We describe a doubly differentially pumped mass spectrometry system to sample the exhaust stream of a large area plasma CVD reactor operating at 0.4–1.5 Torr. We show real-time quantitative analysis of silane consumption and hydrogen production for deposition of hydrogenated amorphous silicon and for pulsed-gas selective area silicon deposition. The ability of mass spectrometry to observe process faults in real time is also demonstrated. Mass spectroscopy is a useful nonintrusive process-state sensor for real-time metrology of plasma deposition, for example, to quantify gas phase species, and to characterize reactions occurring on the substrate surface. Based on our results, we discuss potential advanced manufacturing applications of real-time mass spectrometry in amorphous silicon and selective area silicon plasma deposition, including indirect wafer-state sensing, fault analysis and classification, and run-to-run and real-time process control.

On chemical kinetics of silicon deposition from silane (I). The first order chemical reaction

AbstractTheoretical expressions for silicon layer deposition in consequence of silane decomposition within an open isothermal reaction tube has been derived for the case of homogeneous as well as heterogeneous gas reaction. Layer growth distribution along reaction tube axis has been taken into consideration as well as average layer growth rate related to silane consumption during its passage through the tube on the one hand and to growth rate distribution along the tube on the other. Comparing the theoretical results with experimentally based data homogeneous rather than heterogeneous reaction mode might be preferred. In consequence, however, layer growth rate should be linearly effected by the ratio of reactor gas volume to total substrate surface area. In a practical sense average growth rate, and so axial growth homogeneity, should be expected the higher the lower silane consumption efficiency would be adjusted.

On chemical kinetics of silicon deposition from silane (II). Comparing 1st- and 0.5th-reaction order

AbstractTheoretical expressions for silicon layer deposition from silane pyrolysis within an open isothermal reactor tube has been derived for 0.5th order of chemical reaction with restricting the discussion to homogeneous gas reaction, and compared to the analogous expressions of chemical reaction characterized by 1st order relationship. Both kinds of theoretical equations give results free from contradictions when applied to experimentally based data of silane consumption on the one hand and silicon deposition distribution on the other, at least, with respect to those data published by Puga. First order reaction is characterized by an activation energy of ΔEaSi = 34 – 39 kcal/mole, and 0.5th order reaction by ΔEaSi = 23 – 26 kcal/mole. Chemical reaction of 0.5th order is to be preferred rather than 1st order.