Chapter 10 Simulation of Oxygen Precipitation

Schrems, M.

doi:10.1016/s0080-8784(08)60253-7

Cited by 11 publications

(12 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…where u p is given by the Gibbs-Thomson equation 21 as [11] The precipitate radius, R, depends on the interstitial oxygen flux into precipitates, so the time dependence of precipitate size can be written as 12 [12] As a boundary condition, we assume the point defect concentrations are in equilibrium at the wafer surface…”

Section: Discussionmentioning

confidence: 99%

Oxide Precipitate‐Induced Dislocation Generation in Heavily Boron‐Doped Czochralski Silicon

Ono

Romanowski

Asayama

et al. 1999

J. Electrochem. Soc.

View full text Add to dashboard Cite

Czochralski (CZ) silicon wafers are widely used in the semiconductor industry for integrated circuit fabrication. CZ silicon wafers contain interstitial oxygen atoms which are incorporated during crystal growth and are supersaturated during subsequent heat-treatments. Oxygen precipitation reactions readily occur and the precipitate growth at high temperature (>800ЊC) is normally accompanied by generation of extended defects, such as stacking faults and prismatic punched-out dislocation (POD) loops. 1,2 These oxide-related defects can be useful when they act as metallic impurity gettering sites during device processing; however, if they are located in the near-surface region, these defects may negatively impact device performance and yield.The nature and geometry of the prismatic POD loops have been reported. 2 To lie on {110} planes of the silicon crystal where they form by cross glide of a shear loop initially introduced on {111} planes. Taylor et al. 3 investigated the critical size of oxide precipitates for loop generation in lightly doped silicon. In Ref. 3, the strain around the precipitates was estimated by considering the supersaturation of silicon interstitials. However, the behavior of POD loop generation by oxide precipitates in heavily boron-doped wafers has not yet been fully examined. Since oxygen precipitation is enhanced in heavily boron-doped silicon, it is expected that the supersaturation level for silicon interstitials should also be larger in heavily borondoped silicon. Furthermore, Hahn et al. 19 reported that the occurrence of complex dislocation loops, even on the reduced-stress polyhedral precipitates, is related to a lower barrier for dislocation nucleation and higher dislocation mobility in heavily boron-doped silicon. On the other hand, Fukuda and Ohsawa 6 reported that as the boron concentration increased above 10 18 cm Ϫ3 , the mechanical strength of the crystal improved due to dislocation pinning by boron atoms. In this work, we have investigated the POD loop generation by oxide precipitates in heavily boron-doped wafers using detailed transmission electron microscopy (TEM) observations to understand the mechanism of POD loop generation. ExperimentalBoron-doped 200 mm diam, p-type, (100) CZ silicon wafers with resisitivities of 40, 18, and 9 m⍀ cm were used in this study. The initial oxygen concentrations of these wafers were 11.0-12.1 ϫ 10 17 atoms/cm 3 , as determined by the old ASTM standard. These wafers were subjected to a prolonged annealing in a N 2 ambient at 800, 900, and 1000ЊC for 4-64 h. After annealing, TEM samples were prepared by mechanical polishing followed by argon ion milling at an acceleration voltage of 4 kV. TEM observations were carried out in a 200 kV Topcon EM-002B system along the [001] direction of the silicon matrix. Oxide precipitate size, density, and the extent of PODs were measured. An elemental analysis was performed by energy-dispersive X-ray spectroscopy (EDS) using a Voyager analyzer system from Noran Instruments mounted on a JEOL JEM-2010. Figure ...

show abstract

Section: Discussionmentioning

confidence: 99%

Oxide Precipitate‐Induced Dislocation Generation in Heavily Boron‐Doped Czochralski Silicon

Ono

Romanowski

Asayama

et al. 1999

J. Electrochem. Soc.

View full text Add to dashboard Cite

show abstract

“…Different models of the oxygen precipitation in silicon can be found in the literature. Schrems [4] has described a model based on the diffusion of oxygen monomers with the aid of rate equations and the Fokker-Planck equation. Li Long [5] improved this model by the assumption of a diffusing dimer species in addition to the monomer species.…”

Section: Modeling Of Oxygen Precipitationmentioning

confidence: 99%

Optimized Parameters for Modeling Oxygen Nucleation in Silicon

Zschorsch¹,

Hölzl

Rüfer³

et al. 2003

SSP

View full text Add to dashboard Cite

A parameter study for simulation oxygen precipitate densities is performed by the method of Taguchi for designing experiments. For each model a set of parameters set is concluded which describes experimental data optimal in the form of bulk micro defect (BMD) densities. A temperature dependent surface energy for the oxygen precipitates is fitted to obtain minimal deviations between experimental results and modeling results.

show abstract

“…Various studies have been conducted to overcome the complexity and make the prediction of the amount of precipitation possible. For example, C V was measured by the combined technique of Pt-diffusion at 730˚C and deep level transient spectroscopy (DLTS) (11)(12)(13), while the time evolution of oxygen precipitation could be simulated by a model that consisted of the combination of chemical-reaction equation and Fokker-Planck equation (14). However, most of the available techniques and metrologies still have difficulty to be applied practically such as quality control in the silicon wafer manufacturing or the characterization of a device failure mechanism in terms of extend defect formation.…”

Section: Introductionmentioning

confidence: 99%

Practical Evaluation Method of Oxygen Precipitation in the Czochralski Silicon

Lee¹,

Hong²,

Kim³

et al. 2016

ECS Trans.

View full text Add to dashboard Cite

A simple, practical way to evaluate process-induced oxygen precipitates was investigated. Effective values of fastest nucleation point (TC , J max) were derived as parameters of precipitate densities from the as-grown oxygen precipitate nuclei distributions by fitting them into the cumulative function F(T) of the nucleation rate J = J(Oi, CV, T). Precipitate densities were clearly explained by TC and J max in the entire Oi and point defect region. From the relations of Oi and Cv on the fastest nucleation point in the nucleation rate, TC is closely related to the CV and expected as a potential indicator of evaluating the practical CZ-Si manufacturing.

show abstract

Chapter 10 Simulation of Oxygen Precipitation

Cited by 11 publications

References 46 publications

Oxide Precipitate‐Induced Dislocation Generation in Heavily Boron‐Doped Czochralski Silicon

Oxide Precipitate‐Induced Dislocation Generation in Heavily Boron‐Doped Czochralski Silicon

Optimized Parameters for Modeling Oxygen Nucleation in Silicon

Practical Evaluation Method of Oxygen Precipitation in the Czochralski Silicon

Contact Info

Product

Resources

About