Clean GaAs Research Articles

Comparative studies of the surface reactivity of triethylgallium on semiconductor and dielectric surfaces

The adsorption of triethylgallium (TEG) on GaAs in the presence of submonolayer quantities of Al or In has been investigated. It is found that these additives modify the reactivity patterns observed for TEG on the clean GaAs surface; the presence of Al suppresses the amount of diethylgallium (DEG) desorbing whereas In has the opposite effect. This arises since the kinetic parameters governing the desorption of DEG, which occurs in competition with its surface decomposition, are changed when Al or In are deposited on the surface. Comparative studies on the adsorption of TEG on Si3N4 have also been performed and the results are used to gain an insight into phenomena observed during the growth of III-V semiconductors by chemical beam epitaxy.

Atomic and electronic structures of GaAs(110) and their alkali-adsorption-induced changes.

We performed density-functional-theory total-energy and force calculations of clean and Na-covered GaAs(110) for two coverages: \ensuremath{\Theta}=1/2 (one Na atom per pair of Ga and As) and \ensuremath{\Theta}=1/4. The atomic positions of the adsorbate and of the top three substrate layers are fully relaxed. We discuss the nature of the adsorbate-substrate interaction, the adsorbate-adsorbate interaction, how the adsorbate modifies the clean-surface geometry, and how the atomic and electronic structures are correlated. It is found that the adsorption is partially ionic. The highest occupied adsorbate-induced ``band'' reflects the properties of a Mott-insulator state, where the lowest unoccupied state is a one-dimensional Bloch state.

Dynamical strain at semiconductor interfaces: Structure and surface-atom vibrations of GaAs(110) and GaAs(110)–p(1×1)–Sb

The dynamical force fields of the clean GaAs(110) surface, an isolated Sb chain, and the GaAs(110)–p(1×1)–Sb(1‐ML) interface have been computed utilizing tight‐binding total energy models that have been used successfully to describe the atomic and electronic structure of the clean and adsorbed cleavage faces of the tetrahedrally coordinated compound semiconductors. Since the main consequences of the different chemical bonding in these two cases are manifested in changes in the force field associated with the dynamics of the top‐layer atoms, we explore these consequences using a restricted dynamical model in which only the top‐layer atoms are allowed to vibrate. The resulting vibrational energies are in remarkably good agreement with experimental measurements, and hence afford a vehicle to obtain quantitative relationships between the nature of the surface chemical bonds and the vibrational energies of the surface atomic species.

Reflection high-energy electron diffraction observation of GaAs surface-prepared ultrasonic running de-ionized water treatment

GaAs (001) surface ultrasonically cleaned under running de-ionized water (U-RDIW) is investigated by reflection high-energy electron diffraction (RHEED). The RHEED observations of U-RDIW-treated surfaces show a spotty (1×1) pattern at room temperature and a (2×1) streaky surface reconstruction pattern at 360 °C. The experimental results indicate that chemically clean and damage-free GaAs surfaces can be produced by U-RDIW treatment. We discuss surface structures before/after heating using a hydrogen- terminated model.

Near-band gap transitions in the surface photovoltage spectra for GaAs, GaP and Si surfaces

The spectral surface photovoltage dependences of the clean GaAs(100) surface as well as the real GaP(100) and Si(111) surfaces have been investigated in the vicinity of the energy band-gaps. The spectral regions have been determined, where optical band-to-band direct (GaAs, GaP) and indirect (GaP, Si) electron transitions are dominant. From the graphical analysis the photothresholds of these processes have been estimated, consistent with the literature data.

GaAs surface oxidation and deoxidation using electron cyclotron resonance oxygen and hydrogen plasmas

We have used x-ray photoelectron spectroscopy to investigate electron cyclotron resonance oxygen plasma oxidation and hydrogen plasma deoxidation of GaAs surfaces. Experimental evidence shows that ECR oxidation at room temperature forms a stoichiometric oxide layer that is primarily composed of As2O5 and Ga2O3. A hydrogen plasma is shown to clean GaAs surfaces and recover surface order, even at room temperature. Sample heating (200–300 °C) expedites the removal of Ga2O3 and improves surface order. In addition, no significant differences in oxidation and deoxidation were observed for sample biasing, either positively or negatively.

Photoemission study of the formation of the CaF2/GaAs(100) interface.

The formation of the interface between ${\mathrm{CaF}}_{2}$ and GaAs(100) has been studied by high-resolution photoemission spectroscopy. ${\mathrm{CaF}}_{2}$ thin films were deposited in ultrahigh vacuum by evaporation on clean GaAs(100), and subsequently annealed in steps up to 590\ifmmode^\circ\else\textdegree\fi{}C. By 550 \ifmmode^\circ\else\textdegree\fi{}C, a monolayer of Ca was found to react with As at the interface with an associated loss of approximately one Ga and one F by evaporation per formula unit of ${\mathrm{CaF}}_{2}$. The thickness of the interfacial layer terminates at one monolayer for the range of annealing conditions studied. The interfacial valence-band offset was found to be 8.5 eV and the Fermi level approaches the GaAs valence-band edge with increasing ${\mathrm{CaF}}_{2}$ overlayer thickness.

Surface science studies of MOVPE processes: a review of progress

The increasing use of lower total pressures in GaAs MOVPE growth necessarily results in an increased dependency of the properties of the epilayer upon surface chemistry. This paper describes some of the advances in our understanding of the possible surface chemical pathways involved in MOVPE growth that have resulted from surface science studies. It is demonstrated that although such experiments are often carried out under idealised static conditions, they reveal details of the underlying fundamental chemical pathways likely to be operating under real growth conditions. Data are presented which describe the interactions of three species employed in the growth of GaAs-H 2, AsH 3 and (C 2H 5) 3Ga on GaAs (100) substrates. It is demonstrated that at H 2 pressures of 0.5 mbar and temperatures as low as 700 K, clean GaAs (100) surfaces may be generated. This treatment depletes As from the topmost layers of the substrate, most likely via AsH 3 formation, however As levels may be restored via the heterogeneous decomposition of AsH 3. The extent of As deposition under these conditions is consistent with a low AsH 3 effective sticking coefficient. TEG is shown to adsorb reactively on an As-stabilised GaAs (100) surface at 300 K yielding a species which has the stoichiometry 1Ga:4C±0.5, with the implication that a C 2 unit is liberated via reactive adsorption at 300 K. The resulting surface species is shown to generate both ethene and ethene in the ratio 2:1 in a sequential decomposition pathway at higher temperatures which leaves little residual carbon. This ratio is compared with that expected for a completely intramolecular decomposition pathway. Coverage estimates made from AES data suggest that TEG saturates at a fractional coverage of 0.3±0.1 monolayers with respect to substrate surface atom density. It is demonstrated that these estimates are in reasonable agreement with the saturation coverages predicted for certain species on the basis of Van der Waals radii. It is further demonstrated that the surface science approach can facilitate development of novel methods capable of operating under real growth conditions. This is illustrated via results for the dynamic adsorption of (C 2H 5) 3Ga on GaAs (100) at 300 K obtained using the technique of optical second harmonic generation. Analysis of these data using a simple model generates a value for the saturation coverage of TEG at 300 K which is in good agreement with that estimated from the AES data and also suggest that the system adopts non-Langmuirian adsorption kinetics together with dissociative molecule-surface interactions. By analogy with metal/adsorbate systems, the sticking coefficient for TEG on GaAs (100) at 300 K would appear to increase with increasing exposure. It is suggested that although the fit of the experimental SHG data to a Langmuir isotherm is quite poor, the availability of As surface sites plays a major role in the adsorption kinetics. The surface science/SHG data are compared to results obtained elsewhere and the role of this fundamental surface chemistry is discussed in the light of commonly accepted MOVPE technology.

Clean and Damage-Free GaAs Surface Prepared by Using the Ultrasonic Running Deionized Water Treatment.

Dissolution of arsenic and gallium oxides on GaAs surface during running deionized water (RDIW) and ultrasonic-running deionized water (U-RDIW) treatments is investigated by X-ray photoelectron spectroscopy (XPS). The U-RDIW treated (001) GaAs surfaces are also investigated by reflection high-energy electron diffraction (RHEED). The deionized water used in these treatments is thoroughly deoxygenated to less than 0.3 ppm by bubbling nitrogen gas. RDIW treatment removes arsenic oxide more rapidly than gallium oxide. U-RDIW treatment accelerates the removal of both oxides from GaAs surface and produces oxide-free GaAs surface within 1 hr. After U-RDIW treatment, RHEED observation of the acid-etched surfaces shows a spotty (1×1) pattern at room temperature and a (2×1) streaky reconstruction pattern at 360 C. Ammonium-etched surfaces, on the other hand, show a (2×4) streaky reconstruction pattern at 310 C. The experimental results indicate that chemically-clean and damage-free GaAs surfaces can be produced by U-RDIW treatment. They also show that As/Ga ratios between the acid- and ammonium-etched GaAs surfaces after the U-RDIW treatment are different.

Dynamical low-energy electron-diffraction analysis of bismuth and antimony epitaxy on GaAs(110).

The atomic geometry of Bi adsorbed on GaAs(110) is determined using low-energy electron diffraction (LEED) and compared with calculated atomic geometries of GaAs(111)-p(1\ifmmode\times\else\texttimes\fi{}1)-Sb and clean GaAs(110). The analysis of the one-monolayer, epitaxical films is facilitated by comparing LEED intensity data measured for each system under identical experimental conditions and analyzed using a common multiple-scattering model. The overlapping chain geometry, recently proposed as a possible alternative to the previously determined geometry for the GaAs(110)-p(1\ifmmode\times\else\texttimes\fi{}1)-Sb system, was tested for both the Sb and Bi systems. Comprehensive multiple-scattering calculations indicate, however, that the previously determined geometry provides the superior fit to the LEED intensity measurements. Several improvements to the LEED analysis methodology, including simultaneous multidimensional optimization, are described.

Products of thermal decomposition of triethylgallium and trimethylgallium adsorbed on Ga-stabilized GaAs(100)

We report mass spectrometric studies of the products of thermal decomposition of triethylgallium (TEGa), and trimethylgallium (TMGa) adsorbed on Ga-stabilized GaAs(100) in ultrahigh vacuum. Adsorbed layers were prepared by dosing clean GaAs, near room temperature, to either saturated coverage or coverages below saturation. Subsequent heating leads to desorption of products, detected by a differentially pumped quadrupole mass spectrometer. Total carbon coverage was monitored by X-ray photoelectron spectroscopy (XPS). The substrate temperature was measured by infrared laser interferometric thermometry. At saturated coverage, TEGa decomposes upon heating (1–2°C/s) to yield one third Ga-alkyl product (Ga(C 2H 5) 2) between 0 and 300°C, and two thirds hydrocarbon products (mostly C 2H 4 and some C 2H 5) between 250 and 390°C. About 25% of the Ga deposited from TEGa remains on the surface after all products desorb. Below 40% of saturated coverage, only the higher temperature hydrocarbon products desorb, and all adsorbed Ga remains on the surface. TMGa behaves similarly; starting at saturated coverage, Ga-alkyl product (either Ga(CH 3) 2 or a mixture of Ga(CH 3) 2 and TMGa) desorbs at low temperature (50–380°C) and hydrocarbon product (CH 3) desorbs at higher temperature (250–425°C). However, for TMGa the yield of the Ga-alkyl is about twice the CH 3 yield. No ethane, or butane was detected in TEGa decomposition, nor was any ethane found for TMGa decomposition, indicating that association and disproportionation reactions are unimportant. Also no As-alkyl products were detected. The slower rate of desorption of CH 3 for TMGa decomposition, compared to C 2H 4 and C 2H 5 desorption from TEGa decomposition, qualitatively explains higher carbon levels in GaAs films grown with TMGa versus TEGa.

Low-energy cathodoluminescence experiment with polarized electrons and a negative-electron-affinity GaAs target.

We report new measurements of circularly polarized recombination radiation from highly doped p-type GaAs emitted under impact of longitudinally polarized electrons with an initial kinetic energy varied between 2 and 10 eV. The work function of the target crystal is lowered by cesiation and oxidation of the surface. The results are discussed and compared with our earlier measurements on a clean GaAs crystal surface and with photoluminescence data.

Electronic structure and dynamics of thin Ge/GaAs(110) heterostructures

Using angle-resolved picosecond laser photoemission we have investigated both occupied and transiently excited empty states at the surface of Ge grown epitaxially on GaAs(110). We observe a normally unoccupied, Ge layer derived state whose separation from the valence-band maximum of the system is 700±50 meV at six monolayers Ge coverage. The evolution of the electronic structure is followed as a function of coverage and correlated with low-energy electron diffraction. The time dependence of the transiently occupied Ge signal is compared with that of the clean GaAs(110) surface and shows that electrons are <me;40p>prevented from diffusing into the GaAs bulk by the conduction-band offset of 330±40 meV.

Surface photovoltage spectroscopy investigations of the electronic surface states on clean and oxygen-exposed polar GaAs(100) and GaAs(111) surfaces

Surface photovoltage spectroscopy (SPS) combined with Auger electron spectroscopy (AES) control have been used in the investigation of the electronic surface states on atomically clean, polar n-type GaAs(100) and GaAs(111) surfaces prepared by means of the ion bombardment annealing (IBA) procedure and subsequently exposed to oxygen. For both of the clean surfaces the shape of the surface photovoltage spectrum corresponds to the onset of the band-to-band semiconductor transitions. After oxygen adsorption the amplitude of the surface photovoltage signal for photon energy hν > E g increases significantly, especially for the GaAs(111) surface. Moreover, for photon energy hν < E g an additional wide shoulder in the surface photovoltage appears for both of the investigated GaAs surfaces. It becomes more prominent with increasing oxygen exposures and extends down to photon energies of 1.26 and 1.22 eV, respectively and was attributed to oxygen-induced electronic surface states localized below the middle of the band gap at the surface.

Methyl exchange reaction of trimethylindium on GaAs(100) and the preferential etching of gallium

We have observed a heterogeneous methyl exchange reaction between trimethylindium (TMIn) and Ga atoms on GaAs(100) which could be important in controlling stoichiometry in III–V compound semiconductor films containing In and Ga, prepared from group-III-alkyl sources. Adsorbed layers were prepared by dosing clean GaAs (either Ga- or As-stabilized) with TMIn near room temperature, to saturated coverage (1.7 × 10 14 In atoms/cm 2). Total carbon and indium coverages were monitored by X-ray photoelectron spectroscopy (XPS). Samples were heated to ~ 500° C at a rate of 1.6° /s and desorption products were detected by a differentially pumped quadrupole mass spectrometer. TMIn decomposes to form a Ga-alkyl product (either Ga(CH 3) 2, Ga(CH 3) 3, or a mixture) which desorbs between 70 and 350 ° C, and CH 3 which is detected between 220 and 450 ° C. The Ga-alkyl yield was higher for Ga-stabilized surfaces, while the yield of CH 3 was greater for As-rich surfaces. On both Ga- and As-rich surfaces, the yields of Ga(CH 3) x and CH 3 are, within experimental error, the same as those measured for trimethylgallium decomposition. The preferential “etching” of Ga by TMIn is a likely explanation for the observed reduction in the Ga incorporation probability in InGaAs films prepared from metal-organic molecular beam epitaxy. It could also explain the In-rich surfaces left after CH 4/H 2 plasma etching of InGaAsP.

Investigation of In contacts on atomically clean GaAs(110) surfaces

We have examined In–GaAs(110) contacts fabricated in situ both by depositing In at room temperature followed with annealing and by depositing In on heated surfaces [n- and p-type GaAs(110) surfaces (250 and 300 °C, respectively) using synchrotron radiation photoemission. It is found that In deposition on GaAs(110) surfaces with temperature below room temperature does not induce measurable reaction, whereas reaction takes place with InAs as product when the room-temperature-formed interface is annealed or when In is deposited on the heated substrate. However, there is no evidence of InxGa1−xAs graded heterojunction formation at the interface in either case. Outdiffusion of the excess As from the substrate is suggested to provide a source for InAs formation instead of the previously proposed In–GaAs replacement reaction. Electrical measurements were also performed on in situ In/n-GaAs(110) contacts before and after annealing. Reproducible high-barrier heights were recorded (1.02±0.02 eV) with good ideality factors (1.05±0.01) for the unannealed contacts, which is significantly different from the previously published values (0.76 and 0.78 eV). No dramatic change has been recorded in terms of barrier height upon annealing up to 375 °C although the ideality factor increases to around 1.2±0.1. The implication of this study to ohmic contact formation mechanisms will be discussed.

Near-band gap transitions in the surface photovoltage spectra for GaAs, GaP and Si surfaces

Abstract The spectral surface photovoltage dependences of the clean GaAs(100) surface as well as the real GaP(100) and Si(111) surfaces have been investigated in the vicinity of the energy band-gaps. The spectral regions have been determined, where optical band-to-band direct (GaAs, GaP) and indirect (GaP, Si) electron transitions are dominant. From the graphical analysis the photothresholds of these processes have been estimated, consistent with the literature data.

Novel electron-beam lithography for i n s i t u patterning of GaAs using an oxidized surface thin layer as a resist

The first demonstration of in situ electron-beam (EB) lithography is reported, where a photo-oxidized surface thin layer of GaAs is used for a resist. The in situ EB lithography sequence consists of five processes, i.e., preparation of a clean GaAs surface, photo-oxidation for a resist film formation, direct patterning of the oxide resist by EB-induced Cl2 etching, Cl2 gas etching of GaAs surface for pattern transfer, and thermal treatment in an arsenic ambient for resist removal and surface cleaning. The GaAs wafer is never exposed to air throughout all of the above processes to avoid an unintentional surface contamination. The minimum electron dose required for patterning of the GaAs oxide resist is about 5×1016 cm−2. An overgrown layer on the patterned GaAs surface shows a good surface morphology, which strongly indicates that this technology makes it possible to repeat crystal growth and surface patterning.

De-excitation spectroscopy of L2,3 core excitons in CaF2 thin films

Spectator and participator Auger decay processes have been observed in CaF2 under resonant excitation of the L2,3 core exciton transitions with synchrotron radiation. The CaF2 thin films were deposited on clean GaAs (100) substrates. At the interface the valence band offset is 7.7 eV, with the Fermi level a few tenths of an electron volt above the top of the GaAs valence band.

In situ characterization of organometallic growth of ZnSe using grazing incidence X-ray scattering

We have used in situ grazing incidence x-ray scattering to probe the near atmospheric pressure (100 Torr) OMVPE growth of ZnSe on GaAs. For the first time, we have directly observed the establishment of a clean GaAs surface in a hydrogen ambient with a characteristic (2×4) reconstruction and followed the development of the ZnSe epitaxial film during the initial stages of growth. Our results indicate the presence of a very stable and well-ordered p(2×1) reconstruction during growth, despite the presence of the ambient carrier gas and organic reaction by-products. In addition, by observing the changes in X-ray reflectivity due to the introduction of source compounds, we have also been able to investigate kinetic effects during alternating source growth.