Yuzhuo Li

Investigation of Copper Removal Mechanisms during CMP

In this paper we continue our (CMP) of copper mechanism study of the chemical mechanical polishing using a novel experimental approach. We investigated the effects of chemical reagents in the polishing slurry on the removal behavior of copper film during its CMP. More specifically, the friction and wear of the copper surface against a polyurethane pad were measured in the presence of slurries containing various chemical constituents and surface-treated abrasive particles. These slurries are typically formulated for Cu CMP in the industry. The surface qualities and wear mechanisms were then investigated using an atomic force microscope. The experimental results indicate that the copper removal mechanism in the presence of an abrasive and DI water alone was dominated by mechanical abrasion, while chemical dissolution of copper is much more pronounced when the slurry was added a complexing agent such as glycine, regardless the involvement of the strong oxidizer hydrogen peroxide. In the presence of a slurry consisting of hydrogen peroxide, glycine, and nanometer-sized silica particles, a series of electrochemical reactions promotes the rapid formation of copper oxide and the redeposition of Cu(OH) 2 precipitation. Based on these tribological investigations we attempt to describe the nature of the material removal, which is one of the key unanswered questions in the CMP community.

Effect of pH and H 2 O 2 on Ta Chemical Mechanical Planarization Electrochemistry and X-Ray Photoelectron Spectroscopy Studies

Tantalum is used as a diffusion barrier and adhesion promoter layer between the dielectric material and copper interconnects. The present study was intended to investigate the effect of oxidizer and solution pH on chemical mechanical planarization of tantalum. High purity Ta disks were used to study the dissolution and oxidation kinetics under static and dynamic conditions using various solutions in acidic and alkaline pH regimes. The electrochemical measurements during dynamic polishing of a Ta disk were carried out using slurries containing silica and alumina particles with hydrogen peroxide at various pH levels. The affected surface layers of the statically etched Ta disk were investigated using X-ray photoelectron spectroscopy (XPS), atomic force microscopy (AFM), and secondary ion mass spectrometry (SIMS). Ta metal was observed to oxidize in aqueous solutions at pH 2, 4, and 12 in absence of HO. The oxidation process follows the parabolic kinetic law and oxidation rate was observed to be higher in an alkaline region than in an acidic region. In the presence of HO, however, Ta dissolved in the alkaline region. The dissolution was found to be greater at pH 12 mainly because of enhanced dissociation of HO in alkaline region. At pH 2, on the contrary, mass gain was observed probably due to an increase in OH content on the top of Ta oxide formed on the surface as confirmed by XPS and SIMS depth profile studies. XPS study revealed that the oxidation of tantalum takes place at a rapid rate and forms soluble oxotantalate and hydroxotantalate in solution at pH 12 in the presence of hydrogen peroxide. AFM study validates both the XPS and the SIMS results, indicating formation of a thin impervious oxide layer on Ta in a solution at pH 2 with 5% HO, and a porous layer was formed on Ta in solution with 5% HO at pH 12. Consequently, the dissolution rate in alkaline region was enhanced which was confirmed by dynamic removal rate measurements, electrochemistry, and XPS studies.

Effect of pH and Ionic Strength on Chemical Mechanical Polishing of Tantalum

Chemical mechanical polishing of tantalum was performed using alumina and silica particles dispersed in deionized water as a function of and ionic strength. The highest polish rate was obtained at values of 3.5 and 8.0 in silica and alumina slurties, respectively. The variation in the polish rate with is due primarily to variations in electrostatic interactions between the particles and the tantalum surface and the effect of changes in average particle size appears to be small. This is confirmed by the measured changes in tantalum polish rate due to changes in ionic strength of the slurry. ©2000 The Electrochemical Society

Electrolyte Composition for Cu Electrochemical Mechanical Planarization

Cu electrochemical mechanical planarization (ECMP) is currently being investigated to replace or supplement Cu chemical mechanical planarization (CMP) due to the introduction of porous low-k dielectric materials, which may not withstand the mechanical force applied during conventional CMP. Electrolytes for Cu ECMP at pH 3 containing 5-phenyl-l-H-tetrazole (PTA), hydroxyethylidenediphosphoric acid, and oxalic acid are investigated using electrochemical methods and polishing of Cu-coated blanket and patterned wafers. The Cu removal rate and the planarization efficiency during Cu ECMP can be approximated using electrochemical measurements of the Cu removal rate, with and without surface abrasion. These results predict a 500 mV potential window within which the Cu removal rate is greater than 600 nm/min and the planarization efficiency is greater than 0.90. However, high planarization efficiencies are only obtained when silica abrasives are included within the ECMP electrolyte. In situ electrochemical impedance spectroscopy results indicate that the interfacial impedance is increased by the presence of silica, suggesting that silica is incorporated into the PTA-based passive film and is thus needed for effective planarization. Electrochemical quartz crystal microbalance experiments indicate that PTA may provide better Cu surface passivation at a high anodic potential than benzotriazole, which is widely used during Cu CMP.

A Liposome-Containing Slurry for Tungsten Chemical Mechanical Polishing

Chemical mechanical planarization (CMP) has become an essential process in the manufacturing of advanced semiconductor devices found in cell phones, computers, and other everyday electronics. The three major applications of CMP involve the planarization of structures and films made of copper, tungsten, and silicon dioxide at wafer level. While CMP of silicon dioxide or interlayer dielectric materials represents the earliest application of CMP and copper CMP has become an enabling and fast-growing segment of CMP, tungsten CMP remains steady in production volume and growth. The yield of a CMP process is heavily dependant upon the performance of its corresponding slurry, a stable colloidal aqueous dispersion of abrasive particles and various chemical additives. It is generally true that, other than the dispersed abrasive particles, all other chemical additives are homogenously dissolved in a single aqueous phase. In this paper, we report the design and performance of a liposome-containing slurry for tungsten CMP. More specifically, in addition to the abrasive particles and other chemical additives commonly found in a metal CMP slurry, this new slurry contains a model catalyst that is encapsulated inside of liposome. The caged catalyst does not interact with the oxidizer outside until the polish. The purpose of encapsulating a catalyst is to increase the pot lifetime of the slurry and reduce the static etch caused by the reaction between the catalyst and the oxidizer if they are exposed to each other directly. In this study, we demonstrated that, by caging the catalyst, the static etch rate is significantly reduced and the pot lifetime is significantly increased. The material removal rate remains high. In this paper, the basic design of such slurry is first introduced. The evidence for the encapsulation of a representative catalyst is presented. The performance of such slurry on the polish of blanket and patterned wafers is discussed.

5-Phenyl-1-H-Tetrazole as a Low-pH Passivating Agent for Copper Chemical Mechanical Planarization

The introduction of porous low-k dielectric materials into semiconductor devices requires the development of low downforce Cu chemical mechanical planarization (CMP). An alternative passivation agent, 5-phenyl-lH-tetrazole (PTA), is proposed here that is effective at lower pH than the traditional CMP passivation agent, benzotriazole (BTA). PTA has previously been reported as a low-pH Cu corrosion inhibitor, but has not been explored for Cu CMP. Cu CMP removal rates and Cu static etch rates are measured for slurries containing 3 wt % H 2 O 2 , 1 wt % glycine, 3 wt % colloidal silica, and PTA concentrations ranging from 0.5 to 3 mM. At pH 3 and PTA concentrations of 0.5-2 mM, PTA provides both effective passivation and Cu removal rates of > 1400 nm/min. Fourier transform-infrared and electrochemical studies are consistent with the formation of an effective passivation layer on Cu in CMP slurries containing PTA concentrations ranging from 0.5 to 2 mM, with the effectiveness of passivation increasing with PTA concentration. The improved low-pH passive film formation for PTA in Cu CMP slurries relative to BTA is most likely due to its much lower pKa (4.3), with a much larger fraction of PTA in the anionic form at low pH.

Experimental Study on Abrasive-Free Polishing for KDP Crystal

Abrasive-free polishing (AFP) has been developed for processing of soft, hygroscopic KH 2 PO 4 (KDP) crystal according to its deliquescence property. A new nonaqueous abrasive-free slurry, composed of water, dodecanol, and Triton X-100, is designed for the AFP process. In this slurry, the function of water, which is enveloped into surfactant micelles and has no direct contact with KDP crystal in the absence of polishing pressure, is to dissolve KDP surface material in the AFP. The experiments show that the material removal rate (MRR) is nonlinearly dependent on polishing pressure and platen speed; MRR as high as 700 nm min -1 at certain polishing condition and a scratch-free polished KDP surface with root-mean-square roughness lower than 2 nm are achieved by the AFP process. Moreover, the repeat utilization times of the slurry are experimentally determined in KDP AFP With a relatively high removal rate and a smooth surface.

Mechanism of Cu removal during CMP in H 2 O 2 -glycine based slurries

Hydroxyl radical generation has been observed during Cu CMP using hydrogen peroxide-glycine based slurries. While the Cu dissolution/polish rates increased with increasing glycine concentration, the copper dissolution rate decreased with increasing peroxide concentration indicating the occurrence of both dissolution and passive film formation during CMP. This is further confirmed by both in situ and ex situ electrochemical experiments.

Cu Electrochemical Mechanical Planarization Surface Quality

Cu electrochemical mechanical planarization (ECMP) has been investigated as an alternative technology to Cu chemical mechanical planarization (CMP). The influence of pH and a complexing agent on the surface quality and Cu removal rate during ECMP is studied on blanket Cu-coated wafers to understand discrepancies in surface quality between low pH ECMP electrolytes and CMP slurries containing 5-phenyl-1-H-tetrazole. The results show that surface roughness increases following Cu ECMP relative to Cu CMP, probably due to the introduction of pad porosity. The surface quality improves as the pH is increased from 3 to 6, suggesting that the formation of some Cu oxide may be desirable, although this decreases the Cu removal rate. The surface quality also improves with the substitution of a less aggressive complexing agent (glycine) for oxalic acid. Measurements by electrochemical impedance spectroscopy suggest that the passive film formed with an applied anodic potential during ECMP is thicker than that which forms during CMP at the open-circuit potential. Impedance measurements show that the trend in surface quality with pH is not always consistent with a previously reported correlation with the degree to which the constant phase element approaches an ideal capacitor.

1H-Benzotriazole Incorporated Pad for Chemical Mechanical Planarization of Copper

Chemical mechanical polishing (CMP) pads are generally designed to play a passive role in terms of chemical reactions that occur on the surface to be polished. For copper CMP, a pad that is capable of initiating and controlling various chemical reactions on the copper surface can bring another dimension of flexibility to the CMP process. This work investigates such a model pad for its potential applications in copper CMP. More specifically, the physical and chemical properties of a polyurethane pad containing 1H-benzotriazole (BTA) are examined. When applied to Cu blanket wafers, the BTA-containing pad yielded a higher removal rate compared to the conventional CMP process with the reference pad. For patterned wafer polishing, the results indicate that the BTA-containing pad is effective in improving the step height reduction efficiency and in minimizing dishing and erosion. The potential applications of a pad with such a design for gaining a fundamental understanding of the copper CMP process are discussed based on these experimental results.