We report on the investigation of the origin of high oxide to nitride polishing selectivity of ceria-based slurry in the presence of picolinic acid. The oxide to nitride removal selectivity of the ceria slurry with pi...We report on the investigation of the origin of high oxide to nitride polishing selectivity of ceria-based slurry in the presence of picolinic acid. The oxide to nitride removal selectivity of the ceria slurry with picolinic acid is as high as 76.6 in the chemical mechanical polishing. By using zeta potential analyzer, particle size analyzer, horizon profilometer, thermogravimetric analysis and Fourier transform infrared spectroscopy, the pre- and the post-polished wafer surfaces as well as the pre- and the post-used ceria-based slurries are compared. Possible mechanism of high oxide to nitride selectivity with using ceria-based slurry with picolinic acid is discussed.展开更多
In this study we have used a combinatorial approach for producing binary and ternary alloy thin film libraries using a lab-scale RF co-sputtering system. Initially we used two elemental sputtering targets, i.e. alumin...In this study we have used a combinatorial approach for producing binary and ternary alloy thin film libraries using a lab-scale RF co-sputtering system. Initially we used two elemental sputtering targets, i.e. aluminum (Al) target and neodymium (Nd) target, to produce a film library of varying composition and successfully identified a suitable composition range (1.95–2.38 at% Nd) in which resistance to hillock formation and resistivity of the film spots were found to be satisfactory in annealed state (350°C, 30 min). In another case, in order to form ternary alloy composition library we have used two sputtering targets, i.e. an Al-0.5 at% Nd alloy target and an elemental Ni target. Though, co-sputtered Al-0.6 at% Nd-0.9 at% Ni alloy films showed satisfactory resistance to hillock formation and low resistivity after annealing, film deposited from a ternary alloy target with the same composition failed to show satisfactory resistance to hillock formation during annealing. In case of Al-0.6 at% Nd-0.9 at% Ni alloy target, 250 nm thick film showed poor resistance to hillock formation than the 500 nm thick film. This clearly showed thickness-dependent hillock performance of Al-0.6 at% Nd-0.9 at% Ni alloy. In this study it was found that, in addition to the process variables, metallurgical microstructure of the alloy sputtering targets had significant effect on the film properties which was not obvious from the results of films deposited using co-sputtering of the individual elemental targets.展开更多
基金supported by the Center for Advanced Materials Processing (CAMP) at Clarkson Universitythe National Integrate Circuit Research Program of China (Grant No. 2009ZX02023-3)+3 种基金the National Basic Research Program of China (GrantNos. 2007CB935400,2010CB934300 and 2006CB302700)the National High Technology Development Program of China (GrantNo. 2008AA031402)the Science and Technology Council of Shanghai,China (Grant Nos. 08DZ2200700,08JC1421700 and09QH1402600)the Chinese Academy of Sciences Visiting Professorship for Senior International Scientists
文摘We report on the investigation of the origin of high oxide to nitride polishing selectivity of ceria-based slurry in the presence of picolinic acid. The oxide to nitride removal selectivity of the ceria slurry with picolinic acid is as high as 76.6 in the chemical mechanical polishing. By using zeta potential analyzer, particle size analyzer, horizon profilometer, thermogravimetric analysis and Fourier transform infrared spectroscopy, the pre- and the post-polished wafer surfaces as well as the pre- and the post-used ceria-based slurries are compared. Possible mechanism of high oxide to nitride selectivity with using ceria-based slurry with picolinic acid is discussed.
文摘In this study we have used a combinatorial approach for producing binary and ternary alloy thin film libraries using a lab-scale RF co-sputtering system. Initially we used two elemental sputtering targets, i.e. aluminum (Al) target and neodymium (Nd) target, to produce a film library of varying composition and successfully identified a suitable composition range (1.95–2.38 at% Nd) in which resistance to hillock formation and resistivity of the film spots were found to be satisfactory in annealed state (350°C, 30 min). In another case, in order to form ternary alloy composition library we have used two sputtering targets, i.e. an Al-0.5 at% Nd alloy target and an elemental Ni target. Though, co-sputtered Al-0.6 at% Nd-0.9 at% Ni alloy films showed satisfactory resistance to hillock formation and low resistivity after annealing, film deposited from a ternary alloy target with the same composition failed to show satisfactory resistance to hillock formation during annealing. In case of Al-0.6 at% Nd-0.9 at% Ni alloy target, 250 nm thick film showed poor resistance to hillock formation than the 500 nm thick film. This clearly showed thickness-dependent hillock performance of Al-0.6 at% Nd-0.9 at% Ni alloy. In this study it was found that, in addition to the process variables, metallurgical microstructure of the alloy sputtering targets had significant effect on the film properties which was not obvious from the results of films deposited using co-sputtering of the individual elemental targets.
