Surface chemistry studies of the chemical mechanical planarization (CMP) of copper are presented in this paper. Blanket copper samples were polished with an acidic alumina-based slurry which contains an organic acid salt (phthalic acid salt) and an oxidizer false(H2O2false). Surface studies using X-ray photoelectron spectroscopy (XPS) were performed on copper samples after chemical etching or CMP in order to determine the effect that different polishing parameters (i.e., pH and oxidizer concentration) have on the copper surface. XPS studies were also done on samples that were passively soaked in an acidic slurry mixture containing different concentrations of H2O2 to determine how the chemical action alone affects the removal of copper. The etching results revealed that a cuprous oxide false(Cu2Ofalse) forms on the surface of etched metal while polished samples showed CuO and normalCufalse(OH)2. The effect of these copper oxide films on the removal of copper in passive etching and chemical mechanical polishing is discussed. © 2001 The Electrochemical Society. All rights reserved.

We describe the chemical mechanical planarization (CMP) of copper damascene structures using an IC1400 pad and four different types of slurries. Two alumina-based slurries and two silica-based slurries were evaluated. After successful removal of the excess Cu, we examined the topography of the planarized structures using scanning electron microscopy. The effects of the CMP process on spacer erosion, Cu line recess, corrosion of submicrometer Cu lines, liner removal selectivity, and contamination of the patterned structures are presented. It was found that minimizing the etch rate (ϳ10 nm/min) of the slurry is required to achieve reproducible removal rates and unrecessed (etched) damascene structures. No dishing was observed with the utilized pad. We also show that corrosion is prone to occur in low Cu pattern density areas and that the removal of the liner material (tantalum) remains problematic because of its chemical inertness.

We describe chemical mechanical polishing (CMP) of blanket and patterned aluminum films employing a polyurethane pad and a slurry based on alumina particles as the abrasive and hydrogen peroxide as the oxidizer. The experiments were conducted at pressures from 19 to 47 kPa and at linear velocities from 26 to 48 m/min, and yielded Al removal rates from 80 to 250 nm/min. The oxidant concentration has a weak effect on the removal rate of Al. Polishing selectivities of Al to silicon dioxide as high as 130:1 were obtained with the maximum selectivities being observed at regions of low pressures and low velocities. The Preston equation fails to describe the dependence of the removal rate on pressure and velocity, and a power function is proposed instead. X-ray photoelectron spectroscopy was used to examine the surface of Al before and immediately after CMP. These experiments provided information on the thickness of the oxidized Al layer. We found that larger removal rates correlated with a smaller Al-oxide thickness.

Chemical mechanical polishing (CMP) studies of blanket aluminum, patterned aluminum, and SiO 2 thin films using a commercial slurry based on Al 2 O 3 abrasive particles are presented. Both silicon dioxide and aluminum blanket films were polished with two pads of different hardness and structure. The removal rate and the dependence of the removal rate on pressure and linear velocity for both materials varies significantly with pad type. For the softer pad, the Al removal rate depends critically on the surface saturation of the pad with Al 2 O 3 slurry particles. Scanning electron microscopy and X-ray photoelectron spectroscopy were used to study the texture and chemical composition of the soft pad after different polishing conditions. We found saturation of the pad surface with Al 2 O 3 particles but no metallic aluminum on the pad after CMP. Pad reconditioning causes the removal of the abrasive particles from the pad surface. Patterned aluminum samples with a TiN barrier layer were polished in alumina slurry with and without hydrogen peroxide using the soft pad. During CMP of microstructures, both Al and TiN must be removed at similar rates. The removal rate of the TiN film is dramatically enhanced when H 2 O 2 is incorporated into the slurry, whereas polishing of aluminum thin films showed that the oxidizer has no effect on the Al removal rate. Passive soaking of TiN in H 2 O 2 slurry revealed that TiN readily dissolves in the presence of a strong oxidizer, which increases the chemical component of the CMP process.

We describe the development and examine the performance of a slurry containing an organic acid salt, a silica abrasive obtained from the hydrolysis of ethyl silicate ͑TEOS͒, and a passivating agent for the chemical mechanical planarization ͑CMP͒ of Cu damascene structures. The study is performed on full ͑200 mm͒ wafers coated with blanket Cu films to examine removal rates and uniformity and on partial wafers to investigate the CMP of Cu damascene structures. The silica slurry shows moderately high removal rates, good uniformity values, low defectivity, and excellent ability to remove the Ta liner. It is demonstrated that despite the fairly low Cu to SiO 2 selectivity values, a severe erosion of the SiO 2 does not occur. Because of the effective removal of the liner material, long overpolishing times become unnecessary, thus, spacer erosion is avoided. Corrosion induced defects are prevented by the addition of a passivating agent.

To achieve 65 nm technology node requirements, CMP processes must provide improved control of selectivity, topography, wire cross section, and process robustness. Slurries and processes must also be compatible with fragile low k materials by providing low erosion and shear forces. We present data on a unique step 1 bulk Cu removal slurry with high selectivity, removal rates over 8000 Å/min, and extremely low liner removal/erosion in high (90%) density structures. This is achieved through a combination of surface modified abrasives and alternative inhibitors which provide superior performance and reduced electrochemical activity compared to benzotriazole, a commonly used inhibitor. The step 1 slurry was used with a step 2 liner removal slurry that can be chemically tuned to adjust relative selectivities of Cu:Ta:oxide from the nominal ratio of 1:0.9:1.6, allowing its use with a variety of integration schemes. Results of CMP planarization experiments on 200 mm blanket and patterned single damascene test wafers are described, including electrical data which demonstrates low overpolish sensitivity.

This paper describes the potential of a membrane carrier upgrade for legacy CMP tools, such as an IPEC 472. Seventeen different polish parameter settings were applied for both, 150 mm and 200 mm wafers in order to cover a wide range of applications. Resulting from these experiments, a significant performance gain in comparison to stiff carriers becomes evident.