Investigations were carried out to understand the effect of hydrogen peroxide as an oxidant and benzotriazole ͑BTA͒ as an inhibitor on the chemical mechanical planarization ͑CMP͒ of copper. Cu-CMP was studied using electrochemistry and removal rate measurements in solutions containing the oxidizer and the inhibitor. In the presence of 0.1 M glycine, the copper removal rate was high in the solution containing 5% H 2 O 2 at pH 2 because of a Cu-glycine complexation reaction. The dissolution rate of Cu increased due to the formation of the highly soluble Cu-glycine complex in the presence of hydrogen peroxide. Addition of 0.01 M BTA in the solution containing 0.1 M glycine and 5% H 2 O 2 at pH 2 exhibited a reduction in the Cu removal rate by the formation of a Cu-BTA complex on the surface of copper that inhibits the dissolution. X-ray photoelectron spectroscopy and secondary-ion mass spectroscopy investigations revealed the formation of a Cu-glycine complex, which aided the understanding of the mechanism of Cu-oxidant-inhibitor interaction during polishing.In the era of miniaturization, the need of the semiconductor industry is primarily to increase the packaging density of devices. The performance of integrated circuits depends highly on the type of interconnect materials. In interconnecting technology, the term RC delay, which is the product of the metal resistance ͑R͒ and the capacitance ͑C͒ of the interlevel dielectric, is a major factor for the selection of materials. Copper is gradually replacing aluminum as an interconnect material because it exhibits not only a superior electrical conductivity and thereby less RC delay, but also high resistance to electromigration. 1,2 Hence, chemical mechanical planarization of copper ͑Cu-CMP͒ has emerged as an important component for the manufacturing of integrated circuits. Usually, the Cu-CMP process consists of several steps such as the removal of the surface layer by mechanical action of the pad and abrasive particles, the dissolution of the abraded particles in the CMP solution, and the protection of recessed areas of the surface by using a corrosion inhibitor. The corrosion process is likely to result in pattern defects; therefore, the chemical composition of the CMP slurry is extremely important. 3 The slurry for Cu-CMP contains chemical components to facilitate the oxidation and removal of excess copper as well as passivation of the polished surface. 4 The use of sodium chlorate ͓NaClO 3 ͔, 5 hydrogen peroxide ͓H 2 O 2 ͔, 6-10 and iron nitrate ͓Fe(NO 3 ) 3 ͔ 8 as oxidizers during CMP was attempted in previous studies. Earlier studies 11,12 have shown the effectiveness of glycine as an inhibitor in preventing corrosion of metals. The inhibiting effect of glycine on Cu-CMP was investigated. [13][14][15] proposed that the formation of an oxide film by H 2 O 2 protects the recess region of copper. On the protruded region, the removal of Cu took place both by mechanical abrasion and dissolution with the formation of water-soluble Cu ϩ2 -glycine chelate in the presence...

No abstract

The phase composition and sintering behavior of two commercially available 10mol%Sc2O3–1mol%CeO2–ZrO2 ceramics produced by Daiichi Kigenso Kagaku Kogyo (DKKK) and Praxair have been studied. DKKK powders have been manufactured using a wet coprecipitation chemical route, and Praxair powders have been produced by spray pyrolysis. The morphology of the powders, as studied by scanning electron microscopy, has been very different. DKKK powders were presented as soft (∼100μm) spherical agglomerates containing 60–100nm crystalline particles, whereas the Praxair powders were presented as sintered platelet agglomerates, up to 30μm long and 3–4μm thick, which consisted of smaller 100–200nm crystalline particles. X-ray diffraction analysis has shown that both DKKK and Praxair powders contained a mixture of cubic (c) and rhombohedral (r) phases: 79% cubic +21% rhombohedral for DKKK powders and 88% cubic +12% rhombohedral for Praxair powders. Higher quantities of the Si impurity level have been detected in Praxair powder as compared to DKKK powder by secondary ion mass spectroscopy. The morphological features, along with differences in composition and the impurity level of both powders, resulted in significantly different sintering behaviors. The DKKK powders showed a more active sintering behavior than of Praxair powders, reaching 93–95% of theoretical density when sintered at 1300°C for 2h. Comparatively, the Praxair powders required high sintering temperatures at 1500–1600°C. However, even at such high sintering temperatures, a significant amount of porosity was observed. Both DKKK and Praxair ceramics sintered at 1300°C or above exist in a cubic phase at room temperature. However, if sintered at 1100°C and 1200°C, the DKKK ceramics exist in a rhombohedral phase at room temperature. The DKKK ceramics sintered at 1300°C or above exhibit cubic to rhombohedral and back to cubic phase transitions upon heating at a 300–500°C temperature range, while Praxair ceramics exist in a pure cubic phase upon heating from room temperature to 900°C. However, if heated rather fast, the cubic to rhombohedral phase transformation could be avoided. Thus it is not expected that the observed phase transitions play a significant role in developing transformation stresses in ScCeZrO2 electrolyte upon heating and cooling down from the operation temperatures.

No abstract