Evidence of chlorine ion penetration in InP/InAsP quantum well structures during dry etching processes and effects of induced-defects on the electronic and structural behaviour. Issue 9 (August 2015)

• Record Type: Journal Article
• Authors: Landesman, J.P.; Levallois, C.; Jiménez, J.; Pommereau, F.; Léger, Y.; Beck, A.; Delhaye, T.; Torres, A.; Frigeri, C.; Rhallabi, A.
• Is Part Of: Microelectronics and reliability. Volume 55:Issue 9/10(2015)
• Page Start: 1750
• Page End: 1753
• Publication Date: 2015-08
• Subjects: Dry etching -- Photonic materials -- III-V semiconductors -- Quantum wells -- Luminescence -- Defects -- Mechanical stress; Electronic apparatus and appliances -- Reliability -- Periodicals; Miniature electronic equipment -- Periodicals; Appareils électroniques -- Fiabilité -- Périodiques; Équipement électronique miniaturisé -- Périodiques; Electronic apparatus and appliances -- Reliability; Miniature electronic equipment; Periodicals; 621.3815
• DOI: 10.1016/j.microrel.2015.07.029
• Languages: English
• ISSNs: 0026-2714

Abstract

Abstract: In this work, the overall point of interest is the occurrence of artefacts associated with dry etching processes on InP-based structures. By artefacts we mean creation of defects in the remaining material after etching, defects which might be deleterious to both performance of the photonic devices being fabricated, and reliability/lifetime of these devices. A specific sample structure was defined on InP with InAsx P1 − x quantum wells (QWs). These QWs are buried within 1 μm from the surface, for maximum sensitivity to reactive species produced in the etch plasma, and are designed with a gradual As/P composition, such that the luminescence peak produced by each QW is clearly identified. These samples thus possess a "built-in" marker including its own scale. We focused on chemistries with chlorine (SiCl4 /H2 /Ar and Cl2 /N2 ), implemented in an inductively coupled plasma reactor. With such chemistries, etch rates of 0.5 μm/min can be reached. The samples are not really etched, but just exposed shortly to the plasma for the interaction to take place. Actually, we just etch at most a few tens of nanometers. Characterisation was carried out by spectrally-resolved cathodo-luminescence and photo-luminescence. We also measured secondary ion mass spectrometry profiles, which revealed the penetration of chlorine into the samples. High resolution transmission electron microscopy was used, to probe the crystal quality. By comparing doped and un-doped samples, we show that the chlorine observed after exposure consists at least partly in Cl − ions. The other important observation is some mechanical compressive stress, which is also a consequence of the local concentration of Cl impurities after exposure to the plasma. Highlights: Epitaxial structures grown on InP(100), including InAsP QWs with graded composition, buried very close below the surface. Samples were exposed to different king of chlorine‐based plasmas used in dry etching processes Photo‐luminescence, cathode‐luminescence and SIMS were used as the main characterisation methods Modifications of the luminescence lines (shape, spectral shift) are observed even without etching of the material Penetration of Cl inside the structures is evidenced, with associated doping, compensation and mechanical stress effects.