A Diffusion and Reaction Related Model of the Epitaxial Lift-Off Process

Abstract

In the present work a so-called diffusion and reaction related model ͑DR model͒ is derived based on the notion that the overall etch rate in the epitaxial lift-off ͑ELO͒ process is determined both by the diffusion rate of hydrofluoric acid to the etch front and its subsequent reaction rate in the process. In contrast to the model that was previously described in the literature, the DR model yields etch rates which are in quantitative agreement with those obtained experimentally. In order to verify the DR model, the ELO etch rate of AlAs 1−y P y release layers is determined as a function of the phosphorus percentage, the release layer thickness and the temperature. In accordance with the DR model, it is shown that the etch rate is reaction rate related by the dependence on the phosphorus percentage in the release layer, and that the etch rate is diffusion rate related by the dependence on the release layer thickness. From the temperature dependence, an activation energy of 0.31 eV could be deduced for the ELO process under the present conditions. © 2007 The Electrochemical Society. ͓DOI: 10.1149/1.2779968͔ All rights reserved. The "epitaxial lift-off" ͑ELO͒ technique ͑see Fig. 1a͒, in which a III/V device structure is separated from its GaAs substrate by using selective wet etching of a thin Al x Ga 1−x As ͑x Ͼ 0.6͒ release layer and transferred to a foreign carrier, allows the production of singlecrystalline thin films of III/V materials.1 Application of this technique is interesting for the optoelectronics industry, because use of thin-film devices potentially results in a more efficient transfer of generated heat from device to carrier or heat sink and significantly reduces the amount of material needed by reuse of the substrates. This is of particular importance for an intrinsically large area, thus expensive devices like high efficiency III/V solar cells, 2,3 and the integration of III/V based components with, e.g., silicon-based devices. 4,5 Recently, at our institute thin-film GaAs solar cells were made based on the ELO technique, which reached record efficiencies of 24.5%.6 This is close to the highest efficiency of 25.1% reported for regular GaAs cells on a GaAs substrate, 7 which indicates that the ELO process is not detrimental to the quality of the thin-film device.In 1978 Konagai et al. 8 described the separation of devices from a GaAs substrate using the extreme selectivity of hydrofluoric acid ͑HF͒ for Al x Ga 1−x As with a high Al fraction. A wax layer was applied to support the 30 µm thick fragile films during the process. Yablonovitch et al. 9 noted that the tension induced by the wax caused the thinner films, of micrometer thickness, to curl up with a radius of curvature r as they became undercut. This was concluded to be beneficial for removal of the etch products, leading to an increased lateral etch rate V e of the AlAs release layer. By assuming that three moles of hydrogen ͑H 2 ͒ gas are produced for each mole of AlAs etched and that the ability of dissolved H 2 , which has a low solubil...

Arsenic Formation on GaAs during Etching in HF Solutions: Relevance for the Epitaxial Lift-Off Process

“…15 This mechanism is similar to that responsible for the etching of InP by undissociated HCl in acetic acid or concentrated aqueous solutions, 23 in which phosphine (PH 3 ) gas is formed. Niftrik et al 15 used a similar mechanism to explain the etching of AlAs in a HF solution; in that case the formation of AsH 3 was detected during the experiments. The arsine, formed in reaction 1, can give rise to elemental arsenic.…”

HF Species and Dissolved Oxygen on the Epitaxial Lift-Off Process of GaAs Using AlAsP Release Layers

“…This could be the relatively slow or incomplete dissolution of reaction products under these particular conditions resulting in an accumulation of reaction products close to the etch front. 9 According to Eq. 1, the ELO process overall requires 3 moles of HF and 3-6 moles of H 2 O per mole of AlAs.…”

“…Unfortunately, we were unable to trace these data in the literature, so we now use instead data related to the solubility and diffusion of O 2 in water from Ref. 13 9 it is found that O 2 diffusion would limit the ELO etch rate to a maximum of only a few micrometers per hour if it was required in the dominant reaction with AlAs. Because the experimentally obtained etch rates are several orders of magnitude larger, it is clear that O 2 cannot be directly involved in this process.…”

Epitaxial lift-off process for gallium arsenide substrate reuse and flexible electronics

“…The attack of the substrate by the HF and the residues formed during the sacrificial layer etching result in the increase of RMS roughness from 0.3 to 1-4 nms range. Chemical reactions between AlAs and HF have been well studied 10,15,16 6 À n ] (3 À n) þ , on the other hand, are solid and hard to dissolve into the solution. Besides these primary byproducts, solid As 2 O 3 can also be generated on the substrate depending on the oxygen concentration of the etchant 17 .…”