Authors: Quoc Toan Le, Esen Gül Arslan, Kevin Fundu, Jean Philippe Soulie, Efrain Altamirano-Sanchez
Abstract: The effect of various chemical solutions and mixtures on the etch characteristics, roughness change, and surface composition of NiAl, Al, and Ni films were investigated. Both HCl solution (1.82%) and NH4OH (0.6 and 1.45%) solutions were found to have a detrimental effect on NiAl film in terms of material etching (4-point probe results) and surface roughness change (AFM). Within the concentration range applied, adding H2O2 into the HCl or NH4OH solutions resulted in a significant increase of the etching of the NiAl film. A correlation was observed between the magnitude of etching and increase in surface roughness suggesting that a preferential etching occurred, most likely of grain boundary. Experimental results showed that in the case of 1.82% HCl-H2O2 mixture, NiAl surface can be protected up to 240 s of immersion with the use of a corrosion inhibitor such as triazole (TA).
341
Authors: Teppei Nakano, Quoc Toan Le, Hikaru Kawarazaki, Takayoshi Tanaka, Efrain Altamirano-Sanchez
Abstract: As semiconductor devices continue to scale, it is important to evaluate alternative metals on narrower wiring or via structures. Ruthenium (Ru) is one promising candidate because of its lower resistivity compared to the conventional metals such as copper (Cu), cobalt (Co) and tungsten (W) on narrow space. To prevent leakage problems between metal layers caused by residues on the bottom and sidewalls after the metal patterning process, a cleaning process for Ru metal lines is necessary. Although the industry standard using Ammonia Peroxide Mixture (APM) is effective for removing residues, it was ineffective for Ru semi-damascene stack. Therefore, a new cleaning method involving UV treatment followed by APM, which was tested on metal pitch 18 nm patterned structure, was developed. This method showed promising results and is expected to be used in manufacturing of future semiconductor devices.
335
Authors: Haci Osman Güvenc, Andreas Klipp, Quoc Toan Le
Abstract: Cleaning of semi-damascene structures after direct metal etch (DME) are becoming more challenging as the size of features are getting smaller and number of materials increase in advanced nodes. Typical DME residues in-between Ru lines, generated during/after semi-damascene patterning by DME, mainly consist of Ti-based residues whereas there are TiN layers present as etch stop layer (ESL) below the SiN HM and underneath Ru lines. Wet cleaning of Ti-based residues selective to TiN and Ru in a decent process time is necessary. Undercut of TiN during wet cleaning results in collapsing of the SiN HM and/or Ru lines. We present a novel FOTOPUR® R solution designed to clean Ti-based residues selective to Ru and TiN. Moreover, the new chemistry can further extend the process window without collapsing the Ru lines.
331
Authors: Yuya Akanishi, Quoc Toan Le, Efrain Altamirano-Sanchez
Abstract: Particle removal from BEOL low-k structures is studied using a novel particle removal technique, called Nanolift which removes particles from the substrate by forming a thin polymer film on the surface and removing the polymer film together with the particles. It was confirmed that Nanolift is capable to remove TiFx particles successfully which are generated during the low-k dry etch process for dual damascene structure formation for BEOL interconnect fabrication. Pattern collapse of the fragile low-k structure was confirmed to be prevented by Nanolift in comparison with conventional dual fluid spray cleaning method. FTIR results show that Nanolift leaves no residual polymer remain in low-k films and K-value shift by the Nanolift process was negligible and comparable with the conventional formulated chemistry cleaning process. From these results, Nanolift can be concluded as a suitable cleaning process for advanced BEOL fabrication process.
277
Authors: Quoc Toan Le, E. Kesters, M. Doms, Efrain Altamirano-Sanchez
Abstract: Different types of ALD Ru films, including as-deposited, annealed Ru, without and with a subsequent CMP step, were used for wet etching study. With respect to the as-deposited Ru, the etching rate of the annealed Ru film in metal-free chemical mixtures (pH = 7-9) was found to decrease substantially. X-ray photoelectron spectroscopy characterization indicated that this behavior could be explained by the presence of the formation of RuOx (x = 2,3) caused by the anneal. A short CMP step applied to the annealed Ru wafer removed the surface RuOx, at least partially, resulting in a significant increase of the etching rate. The change in surface roughness was quantified using atomic force microscopy.
302
Authors: Antoine Pacco, Yuya Akanishi, Quoc Toan Le
Abstract: In this work the wet etching of molybdenum thin films was investigated for applications requiring controlled recess without roughening or pattern loading. First, continuous etching of Mo in alkaline and oxidative peroxide solutions was studied. Then, additives like glycine and diethylenetriamine were used and their effect on etch rate and roughness was assessed. Finally, we evaluated if the requirements for a stepwise etching approach for Mo recess using peroxide or ozonated water as the oxidizing step and ammonia as the oxide dissolution agent were met.
295
Authors: Quoc Toan Le, Els Kesters, Yuya Akanishi, Marleen H. van der Veen, Atsushi Mizutani, Frank Holsteyns
Abstract: The etching characteristics of ECD cobalt in different cleaning solutions were characterized using four-point probe, spectroscopic ellipsometry, and X-ray photoelectron spectroscopy. 0.05% HF solution with saturated dissolved oxygen concentration was found to result in a substantial etch of ECD cobalt (~5 nm/min). In contrast, cleaning in the SC1 1:4:100 mixture and the formulated mixture led to a significantly lower etch amount, which could be explained by the formation of a passivation layer at the surface. XPS characterization indicated the formation of a cobalt hydroxide at the surface. The electrical evaluation of the DD structure carried out after cleaning using the formulated chemical mixture and subsequent metallization showed good yield for the 22 nm Kelvin vias, testifying an efficient cleaning of the Co surface at the via bottom.
263
Authors: Harold Philipsen, Sander Teck, Nils Mouwen, Wouter Monnens, Quoc Toan Le
Abstract: The wet-chemical etching of ruthenium in acidic solutions of cerium (IV) has been investigated using electrochemical methods. Etch rates were determined using Rutherford backscattering spectroscopy (RBS) and post-etching surface roughness was investigated using atomic force microscopy (AFM). Low-k material is compatible with the etchant, however, residues were formed.
284
Authors: Hideaki Iino, Yuichi Ogawa, Toru Masaoka, Quoc Toan Le, Els Kesters, Jens Rip, Yusuke Oniki, Yuya Akanishi, Akihisa Iwasaki, Frank Holsteyns
Abstract: The introduction of Co into MOL and BEOL requires a robust wet clean, especially the optimization of the Co rinsing step seems to be critical. The wafer rinsing solutions with a precisely controlled pH and oxidizing additive have been developed to suppress the Co corrosion. In addition, the mechanism of passivation and corrosion of the cobalt surface as well as the passivation stability is discussed.
268
Authors: Yuya Akanishi, Els Kesters, Quoc Toan Le, Frank Holsteyns
Abstract: The impact of dissolved oxygen (O2) on cobalt (Co) corrosion in dilute HF (dHF) solution was studied. It was confirmed that Co etch rate was enhanced as the amount of dissolved O2 in the HF solution increased. The Co etch rate was also found to increase radially outward when performed on a single-wafer spin process in atmospheric air due to the uptake of O2 during the dispense process. The galvanic corrosion of Co was investigated with two types of structures with a Co/Cu interface in different dissolved O2 concentrations, i.e. (1) Co bump structures on Cu and (2) Cu lines with a Co/TaN liner/barrier structure. By controlling both the dissolved and the atmospheric O2 levels, galvanic corrosion prevention at the Co/Cu interface was achieved.
256