| OCR Text |
Show Surfactant effects on doping of GaAs grown by organometallic vapor phase epitaxy J. K. Shurtleff, S. W. Jun, and G. B. Stringfellowa) Department of Materials Science and Engineering, University of Utah, Salt Lake City, Utah 84112 ✂ Received 21 December 2000; accepted for publication 19 March 2001 ✄ Recently, the addition of the isoelectronic surfactant Sb during organometallic vapor phase epitaxy ✂ OMVPE ✄ of GaInP was shown to eliminate ordering, resulting in a significant change in the band gap energy. These results suggest that surfactants added during growth could have profound affects on other important properties of semiconductors, such as doping. This letter presents the results of a recent study on the effects of the isoelectronic surfactant Sb on doping in GaAs. The addition of a small amount of triethylantimony during OMVPE of GaAs is found, using secondary ion mass spectroscopy analysis, to increase the Zn doping concentration from 6 ✁ 1018 atoms/cm3 to 9 ✁ 1018 atoms/cm3, a factor of 1.6. The amount of antimony introduced into the solid is only 2 - 3 ✁ 1017 atoms/cm3. The addition of Sb also increases the impurity concentration of In in GaAs, but does not affect the concentration of Te or P. © 2001 American Institute of Physics. ☎ DOI: 10.1063/1.1371790 ✆ To control the properties of advanced semiconductor de- vices and structures, the surface properties must be con- trolled during vapor phase growth processes, in particular for organometallic vapor phase epitaxy ✂ OMVPE ✄ and molecu- lar beam epitaxy. The use of surfactants to control the sur- face morphology, growth mode, and surface reconstruction during growth of elemental1 and III/V2-6 semiconductors has been demonstrated. Recently, the use of surfactants to change a major semiconductor property, the band gap en- ergy, was reported.7-9 In vapor phase growth, surfactants typically refer to sub- stances that accumulate at the surface during growth and alter the surface properties. Generally, surfactants are sub- stances with a low solubility ✂ i.e., they are rejected from the solid ✄ and a low vapor pressure ✂ i.e., they do not rapidly evaporate ✄ . This results in a high surface concentration of the surfactant that in turn can profoundly affect the surface en- ergy, surface diffusion, surface reconstruction, adatom at- tachment, and step structure.4 This letter presents the results of a recent study on the effects of the isoelectronic surfactant Sb on doping in GaAs. Secondary ion mass spectroscopy ✂ SIMS ✄ depth profile mea- surements of doped GaAs samples grown by OMVPE with and without the addition of a small amount of triethylanti- mony ✂ TESb ✄ clearly indicate a change in the doping con- centration. The addition of Sb increases the Zn doping con- centration by a factor of 1.6. The addition of Sb also increases the impurity concentration of In in GaAs, but does not affect the concentration of Te or P. All of the GaAs layers discussed in this letter were grown in a horizontal, infrared-heated, atmospheric pressure, OMVPE reactor. Semi-insulating GaAs substrates with sin- gular ✂ 001 ✄ orientation were cleaned by standard degreasing followed by a 1 min etch in a 2 NH4 OH:12H2O:1H2O2 so- lution. The substrates were then rinsed in de-ionized water for 5 min and blown dry with N2 before being loaded into the reactor. Trimethylgallium ✂ TMGa ✄ at 7 °C and tertiarybuty- larsine ✂ TBA ✄ at ✝ 7 °C were used as the organometallic pre- cursors. TESb at ✝ 7 °C was used as the surfactant precursor. The TESb to group III ratio in the vapor was 0.012 for all of the doping experiments. Dimethylzinc with a Zn/IIIV ratio of 0.08 and diethyltellerium with a Te/IIIV ratio of 0.000 38 were used as dopants in the GaAs layers. The indium and phosphorus contamination seen in the GaAs layers is due to memory effects from the previous growth of GaInP. The carrier gas was Pd-diffused H2. All of the layers were grown at a temperature of 620 °C with a V/III ratio of 40, a total flow rate of 5200 ml/min, a growth rate of 1.3 ✞ m/h, and a TMGa partial pressure in the vapor of 2.0 ✁ 10✟ 2 Torr. Dur- ing growth of the GaAs layers, the dopant was added during the entire 36 min deposition. TESb was added after 12 min of deposition and removed after 12 additional min of growth. All of the GaAs layers were smooth and mirror-like when examined using Nomarski phase contrast optical microscopy. SIMS depth profiles of the doped GaAs layers were per- formed by Applied Microanalysis Laboratory using a Cam- eca ims-3f system. A Cs✠ primary beam was used to analyze the GaAs layers for Sb✟, Te✟, and P✟, and an O2✠ primary beam was used to analyze for Zn✠ and In✠. Figure 1 shows the SIMS depth profile of a Zn doped GaAs epilayer. As expected, at a constant DMZn concentra- tion in the vapor ✂ Zn/IIIV✡ 0.08 ✄ , the Zn concentration in the layer slowly builds up to concentration of 5.8 ✁ 1018 atoms/ cm3. However, when a small amount of TESb is added to the vapor ✂ Sb/III ✡ 0.012 ✄ the Zn concentration in the layer in- creases sharply to 8.5 ✁ 1018 atoms/cm3, a 60% increase. As can be seen, the Sb concentration in the layers is very small ✂ 2-3 ✁ 1017 atoms/cm3 ✄ . Note that after the TESb is removed from the vapor, as indicated by a decrease in the Sb concen- tration in the epilayer, the Zn concentration decreases as well. The correlation between the change in the Zn and Sb concentrations in the layer clearly indicates that Sb affects a the incorporation of Zn in GaAs. ☛Electronic mail: stringfellow@coe.utah.edu APPLIED PHYSICS LETTERS VOLUME 78, NUMBER 20 14 MAY 2001 0003-6951/2001/78(20)/3038/3/$18.00 3038 © 2001 American Institute of Physics Downloaded 10 Oct 2007 to 155.97.12.90. Redistribution subject to AIP license or copyright, see http://apl.aip.org/apl/copyright.jsp It has been shown previously that Sb acts as a surfactant in GaInP. For example, surface photoabsorption measure- ments have shown that a small amount of TESb significantly changes the surface reconstruction of GaInP.8 It is expected that Sb also acts as a surfactant in GaAs. Therefore, it is, perhaps, not surprising that a small amount of Sb can affect the incorporation of dopants in GaAs. The SIMS data shown in Fig. 1 are direct evidence that the surfactant Sb affects dopant incorporation. The affect of the surfactant Sb on other dopants was also investigated. Figure 2 shows the SIMS depth profile of a GaAs epilayer that was inadvertently doped with In. The In contamination is due to memory effects from the previous growth of GaInP layers. The correlation between the change in the In and Sb concentrations strongly suggests that the surfactant Sb also affects the incorporation of In in GaAs. Figure 3 shows the SIMS depth profile of a Te ✂ donor ✄ doped GaAs epilayer. The DETe was held constant at a Te/IIIV ratio of 0.000 38. In this case, the addition of TESb ✂ Sb/IIIV✝ 0.012 ✄ during growth did not change the Te con- centration. The Sb concentration in the layer is similar to that in the Zn and In doping experiments which indicates that Sb did collect on the surface during growth of the Te doped GaAs. This suggests that the mechanism for Te incorporation is different than for Zn and In. Note that Zn and In both reside on group III sites; whereas, Te, an n-type dopant, is incorporated on group V sites. As shown in Figs. 1, 2, and 3, the concentration of P inadvertently present in the GaAs epilayers was also mea- sured. Apparently, Sb has little affect on the concentration of P, which is incorporated on group V sites. As was the case for In, the P contamination most likely came from memory effects. The P concentration is the highest (8.2 1018 atoms/cm3) in the Te doping experiment ✂ Fig. 3 ✄ which was the first GaAs epilayer to be grown after the growth of GaInP. The P concentration decreased which each consecu- tive GaAs deposition. The lowest P concentration (2.0 1018 atoms/cm3) was seen in the last GaAs deposition, shown in Fig. 1. The effect of the surfactant Sb on the In distribution coefficient is the most easily analyzed. Clearly, at the growth temperature used in these experiments, 620 °C, some of the In reaching the surface is able to evaporate before being incorporated; otherwise, there could be no clear explanation for the increase of In incorporation due to Sb on the surface. This suggests that the Sb acts to inhibit In evaporation. How- ever, this is not likely due to stronger bonding of In to the Sb-covered surface, since In-Sb bonds are known to be much weaker than In-As bonds. The classical theory of el- emental incorporation into the solid10 suggests that In evapo- ration occurs as it diffuses on the ✂ 001 ✄ terrace, before being incorporated into the crystal lattice by attachment at a step edge. In this case, the ability of In to be incorporated is related to the probability of In reaching a step edge and stick- ing there. Thus, either an increase in group III adatom sur- face diffusion coefficient or an increase in the group III stick- ing coefficient at the step edge would cause an increase in In incorporation into the GaAs. Studies of the effect of the sur- factant Sb on lateral compositional modulation in GaInP sug- gests that Sb does, indeed, increase the surface diffusion co- efficients of Ga and In on the ✂ 001 ✄ surface.11 It is also possible that Sb will increase the sticking coefficient of group III adatoms at ☎✁ 110 ✆ step edges, due to occupation of the ‘‘dangling'' group V sites at the step edge by Sb, since Sb has a much lower volatility than As.12 In fact, Sb is ob- served to increase the step velocity somewhat. FIG. 1. SIMS depth profile of Zn doped GaAs epilayer grown with a Zn/III ratio in the vapor of 0.08. After 12 min of growth, a small amount of TESb ✞ Sb/IIIV✟ 0.012 ✠ was added to the system. 12 min later, the Sb was removed and an additional 12 min of Zn doped GaAs was grown. FIG. 2. SIMS depth profile of In doped GaAs epilayer. In ✞ and P ✠ contami- nation are from a system memory effect. An Sb/III ratio in the vapor of 0.012 was used. The deposition cycle followed the same three step process described for Fig. 1. FIG. 3. SIMS depth profile of Te doped GaAs epilayer grown with a Te/III ratio in the vapor of 0.000 38 and a Sb/III ratio in of vapor of 0.012. The deposition cycle followed the same three step process described for Fig. 1. Appl. Phys. Lett., Vol. 78, No. 20, 14 May 2001 Shurtleff, Jun, and Stringfellow 3039 Downloaded 10 Oct 2007 to 155.97.12.90. Redistribution subject to AIP license or copyright, see http://apl.aip.org/apl/copyright.jsp This mechanism also explains the increase in Zn incor- poration, since Zn also incorporates on the group III sublat- tice. In addition, the lack of a change in incorporation of elements occupying the group V sublattice, ✂ i.e., Te and P ✄ is explained, since neither effect described above changes group V incorporation into the solid. This work indicates that an isoelectronic surfactant such as Sb can affect dopant incorporation in GaAs. The addition of small amounts of TESb during growth of GaAs epilayers by OMVPE significantly affects the incorporation of Zn and In, but not Te and P. The effect of surfactants on doping provides insights into the mechanism for atomic incorpora- tion during vapor phase growth of semiconductors. This re- confirms the importance of surface processes during OMVPE. The ability to control major semiconductor proper- ties, such as conductivity type and concentration, by simply adding a small amount of surfactant during OMVPE growth may profoundly affect the manufacturing of many important semiconductor devices. The authors gratefully acknowledge the financial support of the Department of Energy without which this important research could not have been performed. 1M. Copel, M. C. Reuter, E. Kaxiras, and R. M. Tromp, Phys. Rev. Lett. 63, 632 1989 ✁ . 2 J. Massies, N. Grandjean, and V. H. Etgens, Appl. Phys. Lett. 61, 99 1992 ✁ . 3 R. Yakimova, T. Paskova, and I. Ivanov, J. Cryst. Growth 129, 143 1993 ✁ . 4 E. Tournie and K. H. Ploog, Thin Solid Films 231, 43 1993 ✁ . 5 J. E. Cunningham, K. W. Goossen, W. Jan, and M. D. Williams, J. Vac. Sci. Technol. B 13, 646 1994 ✁ . 6 R. Kaspi, D. C. Reynolds, K. R. Evans, and E. N. Taylor, Inst. Phys. Conf. Ser. 141, 57 1995 ✁ ; B. R. A. Neves, M. S. Andrade, W. N. Rodrigues, G. A. M. Safar, M. V. B. Moreira, and A. G. de Oliveira, Appl. Phys. Lett. 72, 1712 1998 ✁ ; D. Reinking, M. Kammier, M. Horn-von Hoegen, and K. R. Hofmann, ibid. 71, 924 1997 ✁ . 7 J. K. Shurtleff, R. T. Lee, C. M. Fetzer, and G. B. Stringfellow, Appl. Phys. Lett. 75, 1914 1999 ✁ . 8 R. T. Lee, C. M. Fetzer, J. K. Shurtleff, and G. B. Stringfellow, J. Appl. Phys. 87, 3730 2000 ✁ . 9 C. M. Fetzer, R. T. Lee, J. K. Shurtleff, G. B. Stringfellow, S. M. Lee, and T. Y. Seong, Appl. Phys. Lett. 76, 1440 2000 ✁ . 10 A. A. Chernov, in Handbook of Crystal Growth, edited by D. T. J. Hurle Elsevir Science, Amsterdam, 1994 ✁ , Vol. 3. 11 C. M. Fetzer, R. T. Lee, and G. B. Stringfellow, Appl. Phys. Lett. to be published ✁ . 12 H. Asai, J. Cryst. Growth 80, 425 1987 ✁ . 3040 Appl. Phys. Lett., Vol. 78, No. 20, 14 May 2001 Shurtleff, Jun, and Stringfellow Downloaded 10 Oct 2007 to 155.97.12.90. Redistribution subject to AIP license or copyright, see http://apl.aip.org/apl/copyright.jsp |
