Intrinsic Gettering of Copper in Silicon Wafers

Isomae, Seiichi; Ishida, Hidetsugu; Itoga, Toshihiko; Hozawa, Kazuyuki

doi:10.1149/1.1475694

Cited by 22 publications

(19 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Note that in our model the gettering efficiency has the experimentally observed 35,36 N ox r ox 2 dependency. Isomae et al 37 also reported that the gettering efficiency of copper depends on the initial copper contamination level, which is naturally included in our model.…”

Section: E Precipitation Of Nickel and Coppermentioning

confidence: 96%

Experimental and theoretical study of heterogeneous iron precipitation in silicon

Haarahiltunen

Väinölä

Anttila

et al. 2007

Journal of Applied Physics

View full text Add to dashboard Cite

Oxygen precipitation in nitrogen doped Czochralski silicon wafers. I. Formation mechanisms of near-surface and bulk defects J. Appl. Phys. 96, 3255 (2004) Heterogeneous iron precipitation in silicon was studied experimentally by measuring the gettering efficiency of oxide precipitate density of 1 ϫ 10 10 cm −3 . The wafers were contaminated with varying iron concentrations, and the gettering efficiency was studied using isothermal annealing in the temperature range from 300 to 780°C. It was found that iron precipitation obeys the so-called s-curve behavior: if iron precipitation occurs, nearly all iron is gettered. For example, after 30 min annealing at 700°C, the highest initial iron concentration of 8 ϫ 10 13 cm −3 drops to 3 ϫ 10 12 cm −3 , where as two lower initial iron concentrations of 5 ϫ 10 12 and 2 ϫ 10 13 cm −3 remain nearly constant. This means that the level of supersaturation plays a significant role in the final gettering efficiency, and a rather high level of supersaturation is required before iron precipitation occurs at all. In addition, a model is presented for the growth and dissolution of iron precipitates at oxygen-related defects in silicon during thermal processing. The heterogeneous nucleation of iron is taken into account by special growth and dissolution rates, which are inserted into the Fokker-Planck equation. Comparison of simulated results to experimental ones proves that this model can be used to estimate internal gettering efficiency of iron under a variety of processing conditions.

show abstract

Section: E Precipitation Of Nickel and Coppermentioning

confidence: 96%

Experimental and theoretical study of heterogeneous iron precipitation in silicon

Haarahiltunen

Väinölä

Anttila

et al. 2007

Journal of Applied Physics

View full text Add to dashboard Cite

show abstract

“…Since the fabrication of microelectronic devices is a very pure process and the level of possible contamination is relatively low, the results of the haze test are rather inappropriate. It has been already demonstrated that the getter eciency of oxygen precipitates for Cu strongly depends on the initial contamination level [4]. Therefore, many getter tests with low initial contamination level were developed.…”

mentioning

confidence: 99%

Development of a Storage Getter Test for Cu Contaminations in Silicon Wafers Based on ToF-SIMS Measurements

Kot

Kissinger

Sattler³

et al. 2014

Acta Phys. Pol. A

View full text Add to dashboard Cite

This work presents development and results of a storage getter test based on the measurement of the metal concentration by time-of-ight secondary ion mass spectroscopy applied for a low Cu contamination level. It was found that Cu atoms introduced into the as-grown sample by a drive-in anneal at high temperature diuse out from the bulk to the surface within 7 days of storage at room temperature. Annealing steps at low temperature should decrease the time which Cu needs for the outdiusion to the surface. However, the Cu atoms in samples subjected to outdiusion annealing at 250• C or 400• C precipitated in the bulk. The investigation of the getter eciency by the storage test in silicon samples containing oxygen precipitates has shown that in order to achieve a getter eciency above 90%, the density of bulk microdefects should be higher than 6 × 10 9 cm −3.

show abstract

“…Compared with the allowable content of copper (1-2 ppma) for solar cell [7], the obtained copper contents exhibit high values at any compositions of liquid alloys. However, copper is highly diffusive in solid silicon and can be removed by gettering treatment [20,21], hence copper introduced into solid silicon by using the Si-Al-Cu alloy was found to be insignificant.…”

Section: Solubilities Of Copper In Solid Siliconmentioning

confidence: 99%