Evidence for a primarily nonradiative Si,O defect in GaP

Bachrach, R. Z.; Lorimor, O. G.; Dawson, L. R.; Wolfstirn, K. B.

doi:10.1063/1.1661078

Cited by 20 publications

(13 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Deliberate doping with 10 to 30 ppm oxygen has been observed to reduce both 9 and the electroluminescent efficiency of greenemitting LED's (11). In other work, a complex involving Si and oxygen has been observed to be an important nonradiative center when large concentrations of 02 were added to Si-doped melts (5). Consequently, if the reduction in O2 concentration of the melt results in the longer minority carrier lifetimes, one should be able to add to the melt an impurity which has a strong affinity for oxygen but is not incorporated into the GaP in an appreciable concentration as a means of increasing the minority carrier lifetime.…”

Section: Fig 2 Scanning Electron Microscope Photograph Using Secondmentioning

confidence: 94%

See 1 more Smart Citation

Observations on Si Contamination in GaP LPE

Lorimor¹,

Haszko²,

Dapkus³

1975

J. Electrochem. Soc.

Self Cite

View full text Add to dashboard Cite

Silicon has long been known to be an unintentional impurity often present in GaP; however, the concentration of the Si and its effects have not been known. It has been observed that the net donor concentration in undoped LPE layers fluctuates from ,~1 X 10i6 cm-3 < Nd --Na ~ 2-4 • 10 i7 cm-S for different growth runs-when growth is in ,--1 atm of H2. This background Nd --Na is attributed to unintentionally present Si donors. The Si concentrations in these as well as S-and Zn-doped layers as determined by ion microanalyzer measurements range from ,--6 • 10 i6 cm -3 ~< XSsi ~< 2 • 10 is cm -a with the typical value being ,~5~ • 10 i7 cm -3. In the undoped layers XSsi often exceeds Nd --Na. The variations in both background Nd --Na and XSsi are attributed to variations in the oxygen level of the gas ambient which is inversely related to the XSsi. An important reaction resulting in Si contamination is H2 reduction of the quartz used in the construction of the LPE growth system. With a growth sequence designed to grow both a Zn-and oxygendoped p-layer onto an unctoped n-layer in the same growth run, SiO2 precipitates have been observed in the p-layer near the p-n junction. The identification of these precipitates as SiO2 is further proof of the high levels of Si contamination that can occur during LPE growth. Minority carrier lifetimes for Si-doped LPE layers with Nd --Na > 5 • 1017 cm -3 are considerably longer than have been obtained by doping with S or Te. The increased lifetime is attributed to either a reaction involving Si which results in the reduction in the concentration of an important nonradiative center or the fact that the usual dopants S and Te themselves are involved in a nonradiative center.For several years it has been known that Si is often found as an unintentional impurity in GaP (1); however, thi.s Si contamination has received little attention, especially in the LPE layers used in LED's. In this paper we review both previous work and present additional data which together provide a better understanding of Si contamination resulting from LPE growth.Early work by Trumbore et aI. (2) studied Si incorporation into intentionally Si-doped samples. One observation was that emission spectroscopic measurements always yielded a higher value for the Si concentration than did a spectrophotometric (a chemical separation followed by a spectroscopic measurement) determination. They felt that the spectrophotometric analysis was more accurate and so weighed those data points more heavily in their analysis. However, the same data were later reinterpreted (3) by suggesting that the spectr0photometric method yields the isolated Si (or Si-Si pairs) in solid solution while the emission spectroscopic (as well as electron microprobe) method yields the total Si content including Si in the form of some second phase (e.g., SiO2) or complex which is insoluble in the reagents used to dissolve the GaP for the spectrophotometric analysis. Thus, there is early evidence that the incorporation of Si into GaP may be complicated in...

show abstract

Section: Fig 2 Scanning Electron Microscope Photograph Using Secondmentioning

confidence: 94%

“…In one experiment an LPE layer was grown from 925~ from a melt containing 0.01 atom per cent (a/o) radioactive Si (5). The net donor concentration, Nd --Na, was 1.6 • l0 is cm -3 while the total Si concentration, XSsl, was 1.7 • 1019 cm-a.…”

mentioning

confidence: 99%

Observations on Si Contamination in GaP LPE

Lorimor¹,

Haszko²,

Dapkus³

1975

J. Electrochem. Soc.

Self Cite

View full text Add to dashboard Cite

show abstract

“…5 indicate that at 800 ~ and ll00~ Eq. [10] overestimates the concentration of Si(x~,idsi) at the eutectic valley by about a factor of 2.7 and 1.7, respectively. However, as there exist no ternary liquidus data for xl~s > 0.…”

Section: Discussionmentioning

confidence: 94%

“…[7], for an ideal liquid solution the equation of the eutectic valley is given by (~,lsi == 1, aSsi ~_ 1) xl'idsi = exp (~OSsi --~Olsi ) [10] The results of sample calculations based on Eq. [10] in comparison with the data from Fig. 5 indicate that at 800 ~ and ll00~ Eq.…”

Section: Discussionmentioning

confidence: 99%

Calculation of the Liquidus Isotherms and Component Activities in the Ga-As-Si and Ga-P-Si Ternary Systems

Jordan¹,

Weiner²

1974

J. Electrochem. Soc.

View full text Add to dashboard Cite

Liquidus isotherms and thermodynamic activities have been calculated for the Ga-As-Si and Ga-P-Si ternary systems. The calculations are based on the assumption that the activities of the components in the liquid phase can be represented with sufficient accuracy by the ternary regular solution model. A least square analysis of the liquidus data for Si-doped GaAs yields the three ternary interchange energies (i.c.e.'s). These are then used to construct the liquidus isotherms for Si-doped GaAs, Ga-and As-doped Si, and the ternary eutectic valley, all of which are in excellent agreement with the experimental results over a wide temperature and composition range. The ternary Ga-As and Ga-Si i.c.e.'s are in very good accord with the i.c.e.'s derived from the binary data but a similar test for the Si-As i.c.e, shows a less satisfactory agreement. As the Ga-P-Si system has not yet been investigated, the calculations presented here are based solely on binary data and they need experimental confirmation. The implications of these results to crystal growth and the limitations of the regular solution model in the vicinity of the compound composition SiAs are also discussed.

show abstract

“…These include recombination via deep states at Si-0 complexes (Bachrach et al 1972b) and Cu impurities (Wessels 1975). In p-type pulled crystals the role of gallium vacancies has been indirectly investigated in some depth and a convincing amount of circumstantial evidence has been produced, which has associated with the thermodynamically derived gallium vacancy concentration (Jordan et a1 1971).…”

Section: Deep Level Recombinationmentioning

confidence: 99%

Sonication Enables Effective Iron Leaching from Green Tuff at Low Temperature

Nakamura¹,

Okawa²,

Kawamura³

et al. 2011

Jpn. J. Appl. Phys.

View full text Add to dashboard Cite

The present understanding of the operation of green-emitting GaP LED'S is reviewed. All existing visible LED devices which are made in III-V compound semiconductors are inefficient, and it is known that this is due to the low visible luminescence efficiency of the epitaxial material incorporated in the devices. In green-luminescent GaP this inefficiency is becoming understood, and quantitative analyses of the important mechanisms are described. The radiative processes include free-exciton and boundexciton recombination which are important in p-type and n-type material with and without nitrogen doping, but in all materials the recombination is dominated by nonradiative processes which have proved to be elusive and difficult to eliminate. Particular emphasis is therefore placed on recent advances in the quantitative classification of these dominant non-radiative processes in n-type material. These are : (1) recombination at deep defect levels positioned 0.75 eV from the valence band, (2) diffusion-limited recombination at dislocations, and (3) surface and interface recombination.

show abstract

Evidence for a primarily nonradiative Si,O defect in GaP

Cited by 20 publications

References 13 publications

Observations on Si Contamination in GaP LPE

Observations on Si Contamination in GaP LPE

Calculation of the Liquidus Isotherms and Component Activities in the Ga-As-Si and Ga-P-Si Ternary Systems

Sonication Enables Effective Iron Leaching from Green Tuff at Low Temperature

Contact Info

Product

Resources

About