GROWTH OF SEMICONDUCTOR MATERIALS BY HYDRIDE VAPOR PHASE EPITAXY USING AN EXTERNAL ALUMINUM CHLORIDE GENERATOR

Abstract

Disclosed herein is the controlled epitaxy of Al.sub.xGa.sub.1-xAs, Al.sub.xIn.sub.1-xP, and Al.sub.xGa.sub.yIn.sub.1-x-yP by hydride vapor phase epitaxy (HVPE) through use of an external AlCl.sub.3 generator.

Claims

1. A method for the deposition of Al-containing III-V materials by hydride vapor phase epitaxy (HVPE) through use of an external AlCl.sub.3 generator.

2. The method of claim 1 wherein the source temperature of the external AlCl.sub.3 generator is about 400 degrees Celsius.

3. The method of claim 1 wherein the AlCl.sub.3 molecules do not decompose during the deposition process.

4. The method of claim 1 wherein a deposition temperature ranges from 620 to 700 degrees Celsius.

5. The method of claim 1 wherein the Al-containing III-V materials comprise Al.sub.xGa.sub.1-xAs where x is from 0 to 1.

6. The method of claim 5 wherein the V/III ratio of Al-containing III-V materials is from 10 to 300.

7. The method of claim 1 wherein the group V species is selected from the group consisting of nitrogen, phosphorus, arsenic, and antimony.

8. The method of claim 1 further comprising the use of AsH.sub.3.

9. A method for the HVPE deposition of lattice-matched Al.sub.xIn.sub.1-xP and Al.sub.xGa.sub.yIn.sub.1-x-yP wherein x varies from 0 to 1 comprising the use of an external AlCl.sub.3 generator.

10. The method of claim 9 wherein the source temperature of the external AlCl.sub.3 generator is about 400 degrees Celsius.

11. The method of claim 9 wherein the AlCl.sub.3 molecules do not decompose during the deposition process.

12. The method of claim 9 wherein a deposition temperature ranges from 620 to 700 degrees Celsius.

13. The method of claim 9 wherein the V/III ratio of Al-containing III-V materials is from 10 to 300.

14. The method of claim 9 wherein the group V species is selected from the group consisting of nitrogen, phosphorus, arsenic, and antimony.

15. The method of claim 9 further comprising the use of AsH.sub.3.

16. An optoelectronic device made by using a method for the deposition of Al-containing III-V materials by HVPE through use of an external AlCl.sub.3 generator.

17. The optoelectronic device of claim 16 wherein the source temperature of the external AlCl.sub.3 generator is about 400 degrees Celsius.

18. The optoelectronic device of claim 16 wherein the AlCl.sub.3 molecules do not decompose during the deposition process.

19. The optoelectronic device of claim 16 wherein a deposition temperature ranges from 620 to 700 degrees Celsius.

20. The optoelectronic device of claim 16 wherein the Al-containing III-V materials comprise lattice-matched Al.sub.xIn.sub.1-xP and Al.sub.xGa.sub.yIn.sub.1-x-yP.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0022] FIG. 1 depicts a plot of the equilibrium constant for deposition of solid AlAs and GaAs using various group III precursors and As-vapor (As.sub.4) or As-hydride (AsH.sub.3).

[0023] FIG. 2 depicts a diagram of an embodiment of an HVPE reactor used in methods disclosed herein, including an external AlCl.sub.3 generator (not to scale).

[0024] FIG. 3 depicts x.sub.Al (left) and growth rate (right) of Al.sub.xGa.sub.1-xAs epilayers as a function of H.sub.2 carrier flow rate to the external Al boat. The HCl flow rate to the boat, as well as all other growth conditions, was held constant.

[0025] FIG. 4 depicts x.sub.Al (left) and growth rate (right) of Al.sub.xGa.sub.1-xAs epilayers grown with varying (T.sub.S) at constant source temperature (T.sub.D). All other growth parameters were held constant.

[0026] FIG. 5 depicts x.sub.Al (left) and growth rate (right) of Al.sub.xGa.sub.1-xAs epilayers grown with varying deposition temperature (T.sub.D) at constant source temperature (T.sub.S).

[0027] FIG. 6 depicts x.sub.Al (left) and Al.sub.xGa.sub.1-xAs growth rate (right) as a function of AsH.sub.3 flow rate in epilayers grown with T.sub.D=650° C. and all other parameters constant.

[0028] FIG. 7 depicts x.sub.Al (left) and Al.sub.xGa.sub.1-xAs growth rate (right) as a function of AsH.sub.3 carrier flow rate in epilayers grown with T.sub.D=650° C. and all other parameters constant.

[0029] FIG. 8 depicts (004) omega-2theta HRXRD scans of Al.sub.xGa.sub.1-xAs epilayers nearly spanning the entire compositional space.

[0030] FIGS. 9a and 9b depict transmission spectra of in FIG. 9a, Al.sub.xIn.sub.1-xP and FIG. 9b, Al.sub.xGa.sub.yIn.sub.1-x-yP epilayers bonded to glass.

[0031] FIG. 10 depicts x.sub.Al in Al.sub.xIn.sub.1-xP for epilayers grown nearly lattice-matched to GaAs as a function of InCl/AlCl.sub.3 ratio, assuming complete conversion of HCl to MCl.sub.x.

[0032] FIG. 11 depicts the X-ray (004) scan and Nomarski image of lattice matched AlInP grown by HVPE using methods and reactors as disclosed herein. FIG. 11a depicts a (004) omega-2theta XRD scan of Al.sub.0.53In.sub.0.47P epilayer. FIG. 11b depicts a Nomarski microscope image of Al.sub.0.53In.sub.0.47P with different Al content (20×).

[0033] FIG. 12 depicts the internal quantum efficiency at various wavelengths of light for an AlGaAs passivated single junction GaAs solar cell. FIG. 12 depicts two upright GaAs devices passivated with Ga.sub.0.5In.sub.0.5P and Al.sub.0.4Ga.sub.0.6As that achieved similar quantum efficiency indicating the AlGaAs grown at low temperature, 650° C., (growth temperature also conducive to the growth of GaInP) provides effective passivation.

DETAILED DESCRIPTION

[0034] Prior to using the methods and devices disclosed herein, it was thought that the quaternary alloy AlGaInP material cannot be grown by conventional high-volume growth processes such as liquid-phase epitaxy (LPE) and hydride vapor-phase epitaxy (VPE). The difference in thermodynamic stability of aluminum phosphide (AlP) and indium phosphide (InP) makes compositional control extremely difficult by LPE. Additionally, the problem of forming a stable aluminum chloride (AlCl) compound during hydride or chloride vapor-phase epitaxy has prevented the successful growth of Al-containing phosphides by VPE.

[0035] Disclosed herein are methods and devices for the controlled epitaxy of Al.sub.xGa.sub.1-xAs, Al.sub.xIn.sub.1-xP, and Al.sub.xGa.sub.yIn.sub.1-x-yP by HVPE through use of an external AlCl.sub.3 generator. By limiting the Al-source temperature to 400° C. the formation of AlCl.sub.3 was promoted instead of AlCl, a precursor that otherwise prevents controlled deposition of multinary Al-containing compounds and is reactive with quartz reactors. It was shown that conversion of HCl to AlCl.sub.3 in the source zone reaches a maximum at this temperature. The effects of deposition temperature, V/III ratio, and group V precursor species on the Al.sub.xGa.sub.1-xAs solid composition and growth rate were determined. It was discovered that the presence of AsH.sub.3 at the growth front was effective at kinetically promoting the incorporation of Al into the growing film. The controlled deposition of Al.sub.xGa.sub.1-xAs was demonstrated, and for the first time, it was demonstrated that Al.sub.xIn.sub.1-xP, and Al.sub.xGa.sub.yIn.sub.1-x-yP growth is possible by HVPE. Using methods and devices disclosed herein, the deposition of new heterobarrier optoelectronic devices with Al-containing layers by HVPE has been demonstrated, results that were previously unattainable.

[0036] Disclosed herein are methods and devices for the deposition of Al-containing III-V compounds by HVPE using an external AlCl.sub.x generator. AlCl.sub.3 generation was selected for through use of a 400° C. source temperature, enabling controlled deposition of the entire compositional range of Al.sub.xGa.sub.1-xAs from x.sub.Al=0-1. It was verified that the AlCl.sub.3 molecule is insensitive to decomposition at a typical range of temperatures employed in our reactor. The effect of growth conditions such as growth temperature, V/III ratio, and group V species on x.sub.Al and Al.sub.xGa.sub.1-xAs growth rate were evaluated. Conditions that select for AsH.sub.3 over As.sub.2/As.sub.4 strongly promote Al-incorporation. The growth of lattice-matched Al.sub.xIn.sub.1-xP and Al.sub.xGa.sub.yIn.sub.1-x-yP was demonstrated for the first time by HVPE, overcoming previous difficulties with the AlCl precursor that prevented their growth. These results demonstrate that the controlled deposition of Al-containing arsenides and phosphides is possible for HVPE, contrary to what was previously taught in the art.

[0037] First, a series of Al.sub.xGa.sub.1-xAs samples was grown at T.sub.D=650° C. to test the HCl->AlCl.sub.x conversion efficiency in the Al-boat at 400° C. by varying the H.sub.2 carrier flow rate, H.sub.2(Al), to the Al boat with a constant HCl(Al) flow rate and all other growth conditions constant. This experiment tests the conversion efficiency by varying the residence time of the HCl in the boat. At 400° C., thermodynamic calculations indicate that nearly 100% of the input HCl should be converted to AlCl.sub.x if the system is permitted to reach equilibrium. However, kinetic limitations prevent the system from reaching equilibrium if the residence time of the HCl over the Al in the boat is not sufficiently large, as commonly observed in the Ga source. By increasing the H.sub.2 carrier flow rate with a constant HCl flow rate, the residence time of the boat is decreased and one can observe whether Al incorporation in the solid is affected.

[0038] FIG. 3 shows the Al solid content (left) and growth rate (right) of this series of samples. Initially, as the H.sub.2(Al) flow rate increases from 75 to 300 sccm, x.sub.Al increases slightly from 0.58 to 0.61, while the growth rate increases from 5 to 7 μm/h. While the specific cause of these trends is unclear, these results are opposite of what one would expect if the HCl conversion was incomplete, implying that this parameter is not limiting the growth. Above 300 sccm of H.sub.2(Al), the growth rate and x.sub.Al decrease drastically, implying that the residence time of HCl in the boat is too short, and the source conversion reaction is no longer reaching completion. The reduced generation of AlCl.sub.3 leads to decreased x.sub.Al in the solid, and combined with the increased concentration of unreacted HCl in the reactor, suppresses the growth rate. It was noted that the large HCl(Al)/HCl(Ga) ratio of 27 (assuming complete conversion of all HCl to AlCl.sub.3 and GaCl) needed to achieve x.sub.Al=0.5-0.6 suggests that the species reaching the substrate surface is AlCl.sub.3, as opposed to the much more reactive AlCl as indicated by FIG. 1. FIG. 1 plots the equilibrium constant, K.sub.eq, for the growth of GaAs and AlAs from different group III precursors and a) As.sub.4 or b) AsH.sub.3, calculated from thermochemical data. These calculations neglect to include equilibrium between other gas phase precursors.

[0039] Next, it was investigated whether the AlCl.sub.3 generated in the Al source was decomposing into AlCl and HCl before reaching the substrate. Changing the source temperature, T.sub.S, is a useful method to alter the chemistry within the reactor without changing T.sub.D or reactant flows. Previously, T.sub.S was varied to affect decomposition of AsH.sub.3 in the reactor independently of other growth parameters. FIG. 4 displays the results of a similar experiment, in which T.sub.S varied under constant reactant flows with constant T.sub.D to determine whether this would alter the distribution of AlCl.sub.x species in the reactor. x.sub.Al is relatively constant as T.sub.S varies between 650 and 800° C. Growth rate is also relatively constant until showing a decrease at TS=800° C. It is possible that AsH.sub.3 decomposition increased at this temperature and led to decreased growth rate. The insensitivity of Al solid content and growth rate to T.sub.S, combined with the large HCl(Al)/HCl(Ga) required to obtain compositions above x.sub.Al>0.50, strongly suggest that the Al growth species in the reactor is AlCl.sub.3, and that it is not substantially decomposing to AlCl.

[0040] Experiments were performed to understand the growth parameter space for growth of Al.sub.xGa.sub.1-xAs from AlCl.sub.3 by HYPE. FIG. 5 shows the effect of deposition temperature with constant HCl(Al) flow. x.sub.Al increases strongly with T.sub.D and the growth varies weakly, passing through a maximum at 650° C. The trend of increasing x.sub.Al with T.sub.D agrees with the equilibrium curves in FIG. 1, which predict that the driving force for AlAs growth from AlCl.sub.3 increases with T.sub.D while the driving force for GaAs growth from GaCl simultaneously decreases. The growth rate is relatively insensitive in this temperature range because of these opposite trends in K.sub.eq for each binary. This result further suggests that AlCl.sub.3 is the dominant Al-precursor in the reactor, because growth from AlCl and GaCl is expected to exhibit a monotonic growth rate decrease based on FIG. 1.

[0041] The effects of the both the flow rate and the nature of the group V precursor on Al.sub.xGa.sub.1-xAs growth were investigated. FIG. 6 shows the effect of AsH.sub.3 flow rate on x.sub.Al and growth rate. Without being limited by theory, x.sub.Al increases weakly as the AsH.sub.3 flow rate is increased from 45 to 100 sccm. The growth rate doubles within this range, implying that increased AsH.sub.3 flow rate is increasing both Ga and Al incorporation in the solid because of the relative insensitivity of x.sub.Al to this parameter. The nature of the group V species has a much stronger effect on Al.sub.xGa.sub.1-xAs growth. It has been shown that GaAs growth rate could be enhanced by limiting decomposition of the AsH.sub.3 precursor into As.sub.2/As.sub.4. In that case AsH.sub.3 decomposition is limited by increasing the flow rate of H.sub.2 carrier input with the AsH.sub.3, which increases the velocity of the AsH.sub.3 through the reactor and decreases the amount of time it spends in the 800° C. source zone.

[0042] FIG. 7 shows x.sub.Al and growth rate for a series of Al.sub.xGa.sub.1-xAs samples grown with varying AsH.sub.3 carrier flow rate. Note that H.sub.2 flow rate was compensated in another reactor port so that the total H.sub.2 flow rate and reactant dilution level in the reactor were constant. x.sub.Al increases strongly with AsH.sub.3 carrier flow rate, and growth rate increases as well. These results imply that the presence of uncracked AsH.sub.3 is key to the incorporation of Al. This can be understood by considering that K.sub.eq for growth of AlAs from AlCl.sub.3 and As.sub.4 is extremely low, as seen in FIG. 1, while K.sub.eq for AlAs growth from AlCl.sub.3 and AsH.sub.3 is nearly five orders of magnitude larger. We further note that K.sub.eq for AlAs growth from AlCl.sub.3 and AsH.sub.3 is still below unity at 650° C., however, indicating that the equilibrium calculations do not tell the entire story. It is likely that the presence of unreacted AsH.sub.3 modifies the kinetics at the substrate surface. The AsH.sub.3 can provide reactive H-radicals that help drive the kinetic reduction of the otherwise highly stable AlCl.sub.3 molecule, helping to consume surface AlCl.sub.3, explaining the trends observed in FIG. 7.

[0043] The external Al generator allows for the controlled deposition of Al-containing compounds by HYPE. All of these various growth trends were combined to achieve Al.sub.xGa.sub.1-xAs in the compositional space between x.sub.Al=0-1. FIG. 8 shows (004) x-ray diffraction curves for samples with Al content varying from 0.11 to 0.93. The Al-generator also allows for deposition of Al-phosphide compounds by HYPE, which up to this point have never been demonstrated by this growth technique.

[0044] FIG. 9 shows transmission measurements of (a) Al.sub.xIn.sub.1-xP and (b) Al.sub.xGa.sub.yIn.sub.1-x-yP epilayers with compositions closely lattice-matched to GaAs. These wide band gap materials are useful in many III-V devices. For example, they can be readily integrated into devices such as solar cells to provide transparent passivation for front and rear surfaces or as the active layers in LED devices that emit at green wavelengths.

[0045] FIG. 10 shows x.sub.Al for Al.sub.xIn.sub.1-xP epilayers grown near the lattice-matched composition as a function of AlCl.sub.3/InCl ratio. Relative to Al.sub.xGa.sub.1-xAs growth, small AlCl.sub.3/InCl ratios are preferred to achieve a 50/50 solid composition, implying that the growth of Al.sub.xIn.sub.1-xP is controllable through use of the AlCl.sub.3 precursor. The growth of these phosphide materials for the first time as disclosed herein allows for new and more efficient devices to be grown by HVPE.

EXPERIMENTAL

[0046] Materials were grown in an atmospheric pressure, dual-chamber HVPE reactor shown in FIG. 2. GaCl and InCl were generated in situ from HCl and elemental Ga and In in the upper temperature zones 1 and 2 at a temperature of 800° C., except where stated. Substrates were (100)-oriented GaAs with an offcut of 6° towards the (111)A plane. AlCl.sub.3 was generated externally of the reactor in a separate quartz boat enclosed in clamshell furnace. The Al furnace temperature was 400° C. in order to promote generation of AlCl.sub.3 over AlCl. Al precursor generation was controlled by the flow rates of HCl and H.sub.2 carrier to the boat as indicated in FIG. 2. The process lines that deliver the Al precursor to the reactor were heated to 200° C. using insulated heat tapes to prevent solidification of the AlCl.sub.3 and subsequent clogging of the lines. The Al-line is plumbed into the reactor through an alumina tube that extends through the majority of the 800° C. upper source zones. The alumina tube is inert to reaction with AlCl.sub.3 or decomposition byproducts, and has an inner diameter of 4 mm to promote a high velocity through the higher temperature source zone.

[0047] GaAs/Al.sub.xGa.sub.1-xAs/GaAs structures were grown and analyzed for Al solid content and Al.sub.xGa.sub.1-xAs growth rate. The deposition temperature in zones 3 and 4 (T.sub.D) was 650° C. except where noted. AsH.sub.3 was the group V precursor. Al.sub.xGa.sub.1-xAs lattice constant was measured using high resolution x-ray diffraction of the (004) plane and used to compute x.sub.Al via Vegard's law. Epilayer thickness and growth rate were determined by fitting of the sample reflectance using a transfer matrix method and data for n and k were calculated. Al.sub.xIn.sub.1-xP and Al.sub.xGayIn.sub.1-x-yP epilayers were also grown at a temperature of 650° C. from AlCl.sub.3, InCl, GaCl and PH.sub.3. Composition of the quaternary was determined through measurement of the lattice constant by x-ray diffraction and band gap determination from spectroscopic transmission measurements. Transmission samples were fabricated by bonding the epilayer to a glass handle with transparent epoxy and selectively etching away the absorbing substrate using an ammonium hydroxide/hydrogen peroxide based etchant.

[0048] Using methods and devices disclosed herein, a method for the controlled deposition of Al-containing III-V materials by HVPE through use of an external AlCl.sub.3 generator was demonstrated. The generation of AlCl.sub.3 vs. AlCl was selected for through the use of a 400° C. source temperature, enabling reliable control of the solid Al-composition. It was shown that the AlCl.sub.3 molecule was insensitive to decomposition at typical source and deposition temperatures in the reactor as used herein. The effects of growth conditions such as deposition temperature, V/III ratio, and group V species on Al.sub.xGa.sub.1-xAs solid composition and Al.sub.xGa.sub.1-xAs growth rate were determined. It was discovered that conditions selecting for AsH.sub.3 over As.sub.2/As.sub.4 strongly promoted incorporation of Al in the film. Control over Al.sub.xGa.sub.1-xAs composition in the entire range from x.sub.Al=0-1 as well as the growth of near-lattice-matched Al.sub.xIn.sub.1-xP and Al.sub.xGa.sub.yIn.sub.1-x-yP was achieved for the first time by HYPE. These results allow for the growth of new high-performance optoelectronic devices by HYPE.

[0049] The foregoing disclosure has been set forth merely to illustrate the invention and is not intended to be limiting.