ChemInform Abstract: OXYNITRIDE DEPOSITION KINETICS IN A SILANE‐CARBON DIOXIDE‐AMMONIA‐HYDROGEN SYSTEM

Gaind, A. K.; Ackermann, Gerhard; Lucarini, V. J.; Bratter, R. L.

doi:10.1002/chin.197730031

Cited by 2 publications

(8 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The dependence of refractive index of SilOml~p films on NH3:CO2 ratio and the relationship of infrared absorption maxima with the refractive index have been previously established (9). Throughout this paper we have reported the physicochemical proper~ies in terms of the composition of SilOmNp films rather than their refractive indexes.…”

Section: Resultsmentioning

confidence: 95%

“…The novelty of these procedures lies in the fact that they are both written with the underlying assumption that SilOmNp films are homogenous mixtures of Si3N4 and SiO2. Silicon oxynitride films are clearly not mixtures for two reasons: (i) the infrared spectra do not show the separation of SiO2 and SigN4 absorption peaks, and (it) the kinetic investigation by Gaind et al (9) clearly shows that the deposition rate constants of Si1OmNp cannot be determined by assuming codeposition of SiO2 and SigN4. Thus neither the Lorentz-Lorenz theory nor the procedure of molar averaging based solely on the properties of SiO2 and Si3N4 (i.e., refractive index and density) should have yielded any meaningful results.…”

Section: Discussionmentioning

confidence: 97%

“…The details of the reactor used (10), the sample preparation, thickness measurement procedures, and typical deposition conditions have been described elsewhere (9). During this study, 1000-1500A films were used for the determination of most properties, except for stress measurements for which film thicknesses ranged between ~600-8000/k.…”

Section: Methodsmentioning

confidence: 99%

“…Since molar refractivity and molecular weight are a constant for a given material (11) the density is a function of the re/ractive index only (9). Therefore the difference in silicon oxynitride etch rates at two deposition temperatures (i.e., 900 ~ and 1000~C) for films with identical refractive indexes and composition (i.e., infrared spectra) cannot be due to the difference in density.…”

Section: P -[7]mentioning

confidence: 97%

“…It was shown earlier by Gaind et al (9) that the SiH4-CO2-NHs-H2 chemical system results in better process control, especially in forming Si3N4-rich oxynitride films; e.g., SilOmNp films with refractive index ~1.73 can be formed with NH3:CO2 ratio of ~6 at 900~C, whereas for other chemical systems, the NH3:oxidant ratio must be at least 50, usually ~5000 depending upon the oxidant (1)(2)(3)(4)(5).…”

mentioning

confidence: 95%

See 4 more Smart Citations

Physicochemical Properties of Chemical Vapor‐Deposited Silicon Oxynitride from a SiH4 ‐ CO 2 ‐ NH 3 ‐ H 2 System

Gaind

Hearn

1978

J. Electrochem. Soc.

Self Cite

View full text Add to dashboard Cite

The physicochemical properties of amorphous silicon oxynitride (SilOmND) films deposited from a SiH4-CO.2-NH~-H2 system have been investigated. The properties of CVD Si3N4 and SiO2 deposited from SiH4-NH3-H2 and SiH4-CO2-H2, respectively, were also investigated, wherever appropriate, to determine the end points of the SilO~No series. The film stress is a continuous, approximately parabolic function of silicon oxynitride composition. The SilOmN D composition also determines its resistance to oxidation in steam (950~ It is shown that a ,-~500k thick film with the composition of Si2ON2 is an adequate barrier to steam oxidation for 6 hr at 1000~ The etch rate of SilOmN, films is a function of composition in 7:1 BHF. In refluxing I4_~PO4, a peak in etch rate exists at a film composition of ~64% equivalent Si3N4. SilOmNo films deposited between 900 ~ and 1000~C densify 2-3% after "heat-treatment" at 1000~ for 2 hr and their etch rates decrease by ~50%; whereas the stress increases (more tensile) as a result of heat-treatment.Silicon oxynitride (SilOmNp) films have been chemical vapor deposited from various chemical systems (1-6), and the resulting properties have been reported(1-8). It was shown earlier by Gaind et al. (9) that the SiH4-CO2-NHs-H2 chemical system results in better process control, especially in forming Si3N4-rich oxynitride films; e.g., SilOmNp films with refractive index ~1.73 can be formed with NH3:CO2 ratio of ~6 at 900~C, whereas for other chemical systems, the NH3:oxidant ratio must be at least 50, usually ~5000 depending upon the oxidant (1-5).It is our purpose to show the similarity between the silicon oxynitrides produced by the above system and those reported in the literature. This is done by comparing physicochemical properties wherever possible with the existing data in the literature, we will at the same time be filling in gaps in the knowledge of physicochemical properties of silicon oxynitride, especially those which are important in the processing of devices (bipolar, I~'ET, etc.) which require wet etching, e.g., the oxidation barrier of oxynitrides to steam and the etch rate in refluxing phosphoric acid. There are no available data in the literature regarding these physicochemical properties.From a device-processing point of view, the importance of the effect of postdeposition "heat-treatment" (which simulates processes like diffusion or oxidation) on physicochemical properties of these dielectric films cannot be overemphasized. Thus we have investigated the effect of heat-treatment on SilOmNp film stress, etch rate, and changes in density (thickness). We have also investigated the effect of the deposition temperature on SitOmNp physicochemical properties by investigating these at two deposition temperatures, 900 ~ and 1000~ Experimental

show abstract

Section: Resultsmentioning

confidence: 95%

Section: Discussionmentioning

confidence: 97%

Section: Methodsmentioning

confidence: 99%

Section: P -[7]mentioning

confidence: 97%

mentioning

confidence: 95%

See 3 more Smart Citations

Physicochemical Properties of Chemical Vapor‐Deposited Silicon Oxynitride from a SiH4 ‐ CO 2 ‐ NH 3 ‐ H 2 System

Gaind

Hearn

1978

J. Electrochem. Soc.

Self Cite

View full text Add to dashboard Cite

show abstract

Oxidation Studies of Crystalline CVD Silicon Nitride

Tressler

Spear

et al. 1989

J. Electrochem. Soc.

205

View full text Add to dashboard Cite

15271. The diffusion model successfully predicted minimum A1 concentrations for protective A1203 scale formation during cyclic oxidation of Ni-Cr-AI(Zr) alloys containing more than 20% Cr.2. The success of the diffusion model in predicting N~ is strongly dependent on the ability to account for partial oxide spallation. The use of the oxide spalling model (16) provides this ability.3. It is not necessary to account for surface recession or the effect of the Cr concentration gradient on the diffusion of A1 when predicting minimum A1 concentrations for protective A1203 scale formation on Ni-Cr-AI(Zr) alloys.4. The decrease and subsequent recovery in the value of the solute concentration at the alloy surface is dependent on the amount of oxide which spalls each cycle and increases with an increasing number of thermal cycles. The increase in the amount of recovery with an increasing number of cycles is without a satisfactory explanation. ABSTRACTOxidation studies of CVD a-Si3N4 were performed in dry oxygen, oxygen-argon, and oxygen-nitrogen-argon gas mixtures of various oxygen and nitrogen partial pressures at a total pressure of I atm at 1100~176Parallel oxidation studies of single-crystal silicon were also conducted for direct comparison. It was observed that the oxidation of both Si3N4 and Si followed parabolic growth kinetics with activation energies of about 115 kcal/mol and about 30 kcal/mol, respectively. The formation of a single layer of SiO2 and evolution of N2 could not account for the much lower parabolic rate constants and much higher activation energy for Si3N4 than for Si during the oxidation. Detailed characterization of the oxidation scales using ellipsometry, step-by-step etching, SIMS, and XPS techniques indicated that a duplex oxidation scale consisting of SiO2 and SieN20 was formed when Si3N4 was oxidized. The intermediate Si2N20 scale was identified as a singlephase material, not a physical mixture of Si3N4 and SIO2. The low oxidation rate and high activation energy for SigN4 during the oxidation were attributed to the formation of Si2N20 and low oxygen diffusivity in this structurally dense phase. ) unless CC License in place (see abstract). ecsdl.org/site/terms_use address. Redistribution subject to ECS terms of use (see 129.81.226.78 Downloaded on 2014-09-04 to IP ) unless CC License in place (see abstract). ecsdl.org/site/terms_use address. Redistribution subject to ECS terms of use (see 129.81.226.78 Downloaded on 2014-09-04 to IP ABSTRACTThe available thermodynamic and phase equilibria data for the binary systems Cd-Te, Cd-Se, and Cd-S have been analyzed using an associated solution model for the liquid phases and considering the CdTe, CdSe, and CdS phases as line compounds. The phase diagrams and the thermodynamic properties calculated from the model parameters agree well with the experimental data.) unless CC License in place (see abstract). ecsdl.org/site/terms_use address. Redistribution subject to ECS terms of use (see 129.81.226.78 Downloaded on 2014-09-04 to IP

show abstract

ChemInform Abstract: OXYNITRIDE DEPOSITION KINETICS IN A SILANE‐CARBON DIOXIDE‐AMMONIA‐HYDROGEN SYSTEM

Cited by 2 publications

References 1 publication

Physicochemical Properties of Chemical Vapor‐Deposited Silicon Oxynitride from a SiH4 ‐ CO 2 ‐ NH 3 ‐ H 2 System

Physicochemical Properties of Chemical Vapor‐Deposited Silicon Oxynitride from a SiH4 ‐ CO 2 ‐ NH 3 ‐ H 2 System

Oxidation Studies of Crystalline CVD Silicon Nitride

Contact Info

Product

Resources

About