Gas Plasma Etching of Chromium Films

Abstract

The crystallization behaviour of Fe 74 Dy 6 B 20 and Fe 70 Dy 10 B 20 amorphous alloys was carefully investigated by differential scanning calorimetry, Mössbauer spectrometry and x-ray diffraction up to 800 • C. Calorimetric studies were performed in limited temperature ranges that were progressively extended. For Fe 74 Dy 6 B 20 , after partial crystallization into the tetragonal Fe 3 B compound, the remaining amorphous part segregates into two amorphous 'phases', respectively enriched and impoverished in dysprosium. Tetragonal Fe 3 B further transforms into orthorhombic Fe 3 B. Metastable Dy 3 Fe 62 B 14 compound then forms from the Dy-impoverished amorphous fraction, and subsequent crystallization of the Dy 1+ε Fe 4 B 4 phase occurs in the Dy-enriched fraction. Finally, Dy 3 Fe 62 B 14 decomposes into bcc iron, Dy 1+ε Fe 4 B 4 and iron borides. The nature of the first crystallization product suggests the existence of local environments of t-Fe 3 B type for this Dy concentration. The crystallization process of Fe 70 Dy 10 B 20 strongly differs from that of Fe 74 Dy 6 B 20 . Segregation phenomena occur in the amorphous state prior to any crystallization. If the nature of the first crystallization product is assumed to be correlated with short-range order in the amorphous state, our results suggest that the local environments differ from those of Fe 74 Dy 6 B 20 , as they probably involve dysprosium atoms. This behaviour would agree with a previous Mössbauer study performed on the as-quenched amorphous alloys, providing evidence for a structural modification of the iron environments in the rare earth concentration range 8-9 at.%.

Cr and CrOx etching using SF6 and O2 plasma

“…The plasma etching of Cr is currently almost exclusively reserved for chlorine-oxygen-based chemistry by forming chromyl chloride (CrO 2 Cl 2 ) as the final etch product. [29][30][31][32][33][34][35][36][37][38][39][40][41] The consensus is that both chlorine (e.g., Cl 2 plasma) and oxygen radicals (e.g., O 2 plasma) are needed to create the volatile chromyl chloride product. It was demonstrated by Abe et al 29 roughly 45 years ago and a few years later explained by Nakata et al 30 As both O and Cl radicals are needed to remove a Cr atom, possibly allowing any intermediate CrO x Cl y reaction path, the etch rate (ER) shows an optimum for a specific ratio between the supplied Cl 2 and O 2 gas.…”

“…[1][2][3] When aiming at higher aspect ratio features or thicker Cr layers, also RIE lag (the smaller gaps etch slower) will play a role. 1,38,39 On top of these Cr issues, the high O 2 concentration erodes the photoresist (both vertically and laterally) and the selectivity (typically less than two) 1,20,35,36 and resist retraction 1,12,34,37 will restrict the CD performance. Additionally, there is a tradeoff between the requested straightness of the etch profile and mask selectivity.…”