Chemical Mechanical Planarization Polishing Research Articles

Development of a Lab-Scale Tribo-Electrochemical Approach for Chemical Mechanical Planarization

During chemical mechanical planarization (CMP), the electrochemical interactions between slurry additives and the polishing substrate metal films favor the formation of easily removable species that are then detached by mechanical abrasion effects of the CMP pad and nanoparticle abrasives in slurry. Yet, neither the mechanical nor chemical aspects alone dictate the CMP process; rather, it is the intertwined dependencies of tribology and electrochemistry that often control the polish performance. This highlights the importance of the functional synergies, otherwise known as tribo-electrochemistry, which has been largely under-explored in the CMP research community.This work discusses the development of a novel small-scale tribo-electrochemical approach for metal CMP. It covers the tool design aspects, tribo-electrochemical measurements such as open circuit potentiometry (OCP) and polarization resistance, and relevance to CMP applications. The system is developed based on a small-scale CMP polisher coupled to a 3-electrode cell design circuit that is connected to a potentiostat. The design utilizes a common type of CMP metal film as the working electrode, with the accompanying pad and slurry consumables serving as baseline for method development. All other process conditions, including rotating speed, downforce, polish-hold duration, and pad conditioning process follow typical CMP standards.The results show tribo-electrochemical responses to the presence of catalyst (C), oxidizer (O), inhibitor (I), and abrasive (A) in the experimental slurries. The data is of significant value in terms of the anodic & cathodic site functions on the CMP metal surface and in the context of material removal. In addition, notable sensitivity to pad texture, downforce, and temperature effects were observed, and removal rate measurements correlated with polarization resistance for varied levels of the O and I additives. In summary, the tribo-electrochemical aspects of CMP have been probed and an innovative analytical approach has been developed and demonstrated.

Environmentally Friendly Amino Acid-Based Polishing Slurry with Reduced Galvanic Corrosion at Copper/Cobalt Interface for Advanced Cu Interconnect Applications

Chemical mechanical planarization (CMP) is an integral part in semiconductor processing including the back-end-of-line (BEOL) step. A goal of CMP is to uniformly polish the metal interconnects and the liner/barrier materials while minimizing CMP-related defects such as pitting, dissolution, and galvanic corrosion. A variety of azoles are used as corrosion inhibitors such as benzotriazole (BTA) to prevent corrosion defects during polishing. [1,2] These azoles work together with particles in the slurries to polish the overplated metal forming the interconnects. However, azoles are very poorly biodegradable during wastewater treatment and highly toxic to the nitrification process, which requires expensive wastewater treatment challenges.[3,4] Also, it is known that the use of BTA leads to contamination of Cu films by forming undesirable hydrophobic organic residues (e.g., BTA and Cu-BTA complexes),[5] making their removal very difficult during cleaning. To address these challenges aliphatic amino acids were examined as potential corrosion inhibitors to replace problematic azoles in CMP slurries.Three amino acids were considered in a model CMP slurry. These three were chosen for their unique functional groups: methionine, containing a sulfur group, glutamic acid, having an extra carboxylic acid group, and leucine, having a branched hydrocarbon chain. A representative polishing slurry with 35 nm colloidal silica abrasives, 1 wt% hydrogen peroxide as an oxidizer, 0.5wt% arginine as a complexing agent, at pH 8 was used. Three experimental aims were considered (1) corrosion characterization of the three amino acid and BTA electrolytes with linear sweep voltammetry; (2) comparison of the Co and Cu metal removal rate (MRR) in polishing experiments with a proto-type CMP polisher with 8” wafers, and (3) oxidation of the amino acids and BTA in a Fenton assisted electrooxidation reaction. Methionine was most effective in controlling the galvanic corrosion of the Cu-Co couple. The Co and Cu MRR decreased in the presence of three amino acids and showed comparable removal selectivity. The forced degradation of the amino acids under the Fenton-assisted electrooxidation confirmed that they were more readily degraded compared to BTA. Acknowledgments: This work was supported by the Semiconductor Research Corporation grant (Task 3100.001) under the Global Research Collaboration program. References Seo, J., A review on chemical and mechanical phenomena at the wafer interface during chemical mechanical planarization. Journal of Materials Research 2021, 1-23.C. Peethala, H. P. Amanapu, U. Lagudu, and S.V. Babu, J. Electrochem. Soc, 159(6), H582-H588, 2012.D.E. Speed, Chapter 11, Advances in chemical mechanical planarization (CMP) 2nd edition, Babu, S. V., Ed. 2021.G. Li et al. / Chemosphere 241 (2020) 1249932.Seo, J.; Vegi, S. S. R. K. H.; Babu, S.V. ECS Journal of Solid State Science and Technology 2019, 8 (8), P379–P387.

Development of modeling to investigate polyurethane pad hardness in chemical mechanical planarization/polishing (CMP) process

In chemical mechanical polishing (CMP) for IC fabrication, porous pads are critical for achieving extremely consistent throughput in CMP process. The pad hardness, porosity of pad, and pore sizes in the thermoplastic polyurethanes foam fabrication procedure determine the properties of CMP pads. This paper aims to develop a methodology for modeling macro and micro pad hardness. The IC1000 CMP pad has been analyzed with a developed image processing method to determine porosity and pores size. In addition, nano-indentation tests are used to determine physical properties of micro-solid components. For macro hardness analysis, five different sizes of three-dimensional representative volume elements (RVE) have been investigated using the random sequential adsorption technique. Furthermore, simulation of macro pad hardness of RVE models are completed with finite element method. Results of this study not only achieves macro hardness of CMP pad simulation as 3% errors as compared with developed micro-indentation test, but also complete three dimensional porous pad modeling with RVE.

Dynamic Pad Surface Metrology Monitoring by Swing-Arm Chromatic Confocal System

This study aims to develop a dynamic pad monitoring system (DPMS) for measuring the surface topography of polishing pad. Chemical mechanical planarization/polishing (CMP) is a vital process in semiconductor manufacturing. The process is applied to assure the substrate wafer or thin film on wafer that has reached the required planarization after deposition for lithographic processing of the desired structures of devices. Surface properties of polishing pad have a huge influence on the material removal rate (MRR) and quality of wafer surface by CMP process. A DPMS has been developed to analyze the performance level of polishing pad for CMP. A chromatic confocal sensor is attached on a designed fixture arm to acquire pad topography data. By swing-arm motion with continuous data acquisition, the surface topography information of pad can be gathered dynamically. Measuring data are analyzed with a designed FFT filter to remove mechanical vibration and disturbance. Then the pad surface profile and groove depth can be calculated, which the pad’s index PU (pad uniformity) and PELI (pad effective lifetime index) are developed to evaluate the pad’s performance level. Finally, 50 rounds of CMP experiments have been executed to investigate the correlations of MRR and surface roughness of as-CMP wafer with pad performance. Results of this study can be used to monitor the pad dressing process and CMP parameter evaluation for production of IC devices.

Advances in Pad Conditioning for Enhancement of CMP Process Stability over Pad Life

There is an increasing demand for deposition of thick films because of increasing focus on 3D integration of devices. These thick films will require higher removal rates during the chemical mechanical planarization/polishing (CMP) process in order to avoid significantly high polish times for minimizing impact on throughput of the polisher. Aggressive polishing conditions will be required to achieve high removal rates, which in turn will need aggressive pad conditioning to maintain desired surface texture and prevent glazing of the pad. In this paper, the impact of different pad conditioning sweep profiles on CMP pad thickness profile and removal rate WIWNU & WTWNU will be discussed. A methodology is developed that uses a simulator to predict pad thickness profiles for different pad conditioning sweep profiles. Subsequently, these different pad conditioning sweep profiles are evaluated experimentally to study their impact on pad thickness profiles and material removal rates (MRR) on Applied Materials’ 300mm wafer polisher Reflexion® LK. The pad thickness profiles are characterized by confocal microscopy and state-of-the-art full pad surface metrology tool. Pad thickness profiles predicted by the simulator are compared with those measured using confocal microscopy and full pad metrology tool. It is validated that the pad profile trend predicted by the simulator matches with the actual pad thickness profile obtained on the polisher. Advantage of using full pad metrology tool relative to confocal microscopy is highlighted.

Quantification of shear induced agglomeration in chemical mechanical polishing slurries under different chemical environments

The impact of shear on agglomeration in CMP slurries needs to be studied to reduce the defectivity resulting from large agglomerates (>0.5 μm) during Chemical Mechanical Planarization/Polishing (CMP) process. In this work the impact of the change in the pH of CMP slurries on their agglomeration behavior at different quantified shear rates are studied. The threshold value of Camp number (equal to shear rate x time) which represents total shear that is required for causing agglomeration in CMP slurries for different pH values (pH 2, 7 and 10) are measured. The value of the threshold Camp number required to cause agglomeration in a slurry was lowest for acidic pH conditions (Camp number 1E5), highest for basic pH conditions (Camp number 1.5E6) and value of threshold Camp number for neutral pH conditions (Camp number 1.5E5) was between acidic and basic conditions. It was concluded that the value of the threshold Camp number is dependent on the inter-particle forces or surface charges on silica particles in the slurry, and higher inter-particle repulsive forces increases the threshold value of Camp number. Good co-relations between Camp number and the extent of agglomeration in different CMP slurry samples are demonstrated. This method can be used for designing better slurry delivery systems for different slurry chemistries, to reduce the formation of the shear induced agglomerates during slurry handling and in turn reducing the defects during CMP process.

Effect of Surfactant Based Abrasive Free Slurry on CMP Polishing Rate and Planarization of Semi-Polar (11‒22) GaN Surface

An abrasive free slurry has been formulated using ionic and non-ionic surfactants with KMnO4 as an oxidiser. Subsequently, the effect of these surfactants on the material removal rate (MRR) and surface planarity of semi-polar (11‒22) GaN surface have been studied using chemical mechanical planarization (CMP) process. The formulated polishing slurries were characterized for their rheological properties such as shear thickening, thinning and viscosity as a function of shear rate. It was found that the polishing rate and surface planarity depend on the type of surfactant and its concentration. The estimated MRR values of various surfactants are seen to decrease from anionic to cationic to non-ionic in the order SDS>CTAB>TX-100 and the maximum MRR has been found to be 2.58μm/hr for 0.5 wt% SDS surfactant containing slurry, under optimized conditions of other CMP parameters. In compared to the cationic (CTAB) and non-ionic surfactants (TX-100), anionic surfactant (SDS) offered relatively good surface planarity with a remarkable root-mean-square (rms) surface roughness (Rq) of 2 Å over a scan area of 1 × 1 μm.2

Ceria coated hexagonal mesoporous silica core–shell composite particle abrasives for improved chemical–mechanical planarization performance

The structure design of abrasive particles provides an available approach for improving both surface roughness and polishing efficiency in chemical–mechanical planarization/polishing (CMP) applications. In this work, the hexagonal mesoporous silica (H-mSiO2) particles with parallel channels were prepared via a modified tyltrimethylammonium bromide-assisted template method. And the ceria nanoparticles attached to H-mSiO2 was achieved by a solution synthesis technique. The core–shell structure of the as-prepared H-mSiO2–CeO2 composites was characterized in terms of X-ray diffraction, field emission scanning electron microscope, high-resolution transmission electron microscope, nitrogen adsorption/desorption measurement, and STEM–EDX mapping techniques. The oxide–CMP performance of the H-mSiO2–CeO2 composite particles as abrasives was evaluated in terms of surface finish and material removal rate. For comparison, the commercial ceria abrasives and solid silica (sSiO2)–CeO2 composite particles with non-porous sSiO2 cores were also tested under the same CMP conditions. Oxide–CMP results revealed that the H-mSiO2–CeO2 composite abrasives contributed to the finish reduction, efficiency improvement, and scratch elimination with respect to conventional ceria abrasives. By comparing with rigid solid silica (sSiO2)–CeO2 particles, the non-rigid H-mSiO2–CeO2 composites revealed a reduced surface roughness (0.17 nm vs. 0.33 nm, root-mean-square values), a low topographical variation (± 0.4 nm vs. ± 0.8 nm), and an improved removal rate (203 nm/min vs. 144 nm/min). The improved CMP performance might be attributed to the enhanced overall elastic response and reduced particle density, resulting from their hexagonal meso-silica cores with abundant parallel channels. Moreover, the increased Ce3+ concentration also contributed the improvement of polishing efficiency. This work describes an effort to explore the relationship between the meso-silica structure and finishing performance of the ceria-based core–shell abrasives for optimizing oxide–CMP characteristics.

Effect of Pad Type on Passivation Film Formation Relevant to Copper Chemical Mechanical Planarization

Copper (Cu) chemical mechanical planarization (CMP) is an extremely critical step in the manufacturing of ultra-large-scale integrated (ULSI) semiconductor devices. By optimizing the synergistic relationship between a chemical dispersion, referred to as a slurry, and mechanical parameters, global planarization is able to be achieved utilizing CMP. Traditional slurries for Cu CMP promote Cu oxidation, promoting the creation of a passivation film which readily able to be removed by the same mechanical process. However, the efficiency of film removability is strongly correlated to the molecular structure of the passivating agent and also the composition of the polyurethane polishing pad. This work aims to mechanistically understand the relationship between molecular structure of the passivating agent and the pad type on the modulation of the passivation film formed. The film formation kinetics and characteristics were studied through various electrochemical techniques and surface spectroscopy. Results have indicated that the film is directly correlated to the molecular structure of the passivating agent, and the ease of removing the film is strongly contingent on the type of polyurethane pad implemented in the CMP polishing process.

Chemical Mechanical Planarization and Old Italian Violins.

Previous studies have shown that spectral analysis based on force data can elucidate fundamental physical phenomena during chemical mechanical planarization (CMP). While it has not been literally described elsewhere, such analysis was partly motivated by modern violinmakers and physicists studying Old Italian violins, who were trying to discover spectral relations to sound quality. In this paper, we draw parallels between violins and CMP as far as functionality and spectral characteristics are concerned. Inspired by the de facto standard of violin testing via hammer strikes on the base edge of a violin’s bridge, we introduce for the first time, a mobility plot for the polisher by striking the wafer carrier head of a CMP polisher with a hammer. Results show three independent peaks that can indeed be attributed to the polisher’s natural resonance. Extending our study to an actual CMP process, similar to hammered and bowed violin tests, at lower frequencies the hammered and polished mobility peaks are somewhat aligned. At higher frequencies, peak alignment becomes less obvious and the peaks become more isolated and defined in the case of the polished wafer spectrum. Lastly, we introduce another parameter from violin testing known as directivity, Δ, which in our case, we define as the ratio of shear force variance to normal force variance acquired during CMP. Results shows that under identical polishing conditions, Δ increases with the polishing removal rate.

Method for Determining Stability of CMP Slurry

Increasing exacting requirements for defectivity during chemical mechanical planarization/polishing (CMP) process has led to an increased demand of highly stable CMP slurries. Herein, we describe a novel method to obtain values of stability ratio (SR) denoted as ‘W’, a parameter which defines the ability of slurry particles to resist agglomeration. We found that SR is independent of shear rates, is constant for a given slurry and higher the value of SR, lesser is the inclination of the slurry particles to agglomerate even under intense shear stress conditions. We conclude that for determining slurry stability, evaluation of the SR value is a much better metric than traditional methods such as zeta potential, particle size measurements etc. For a 10 wt% silica slurry depending upon conditions such as pH, addition of salts, particle size; values of SR were in the range 1.26–16.53. Based on comparison with CMP defects data, we concluded that (among the slurry samples studied) 10wt% solids, 35 nm basic pH silica slurry with no salt, having SR value 16.53 is most stable and 35 nm acidic pH silica slurry with KCl salt and 135 nm basic pH slurry having SR value 1.26 are relatively most unstable.

Study of Agglomeration Behavior of Chemical Mechanical Polishing Slurry under Controlled Shear Environments

The presence of large oversize particles (>0.5 μm) in the slurry is known to cause polishing defects during chemical mechanical planarization/polishing (CMP) process. One of the reasons for presence of large particles in the slurry is due to usage of pumps during slurry delivery. The shear rate of pumps is known to cause agglomeration problem in CMP slurries, with increasing shear rate leading to increase in agglomeration in the slurries. However, till now all the work on shear-induced agglomeration has been qualitative or at best semi-quantitative in nature. In the present work, a novel technique to study agglomeration behavior as a function of quantified shear rates has been developed. A rheometer was used as a means to subject slurry to different shear rates for various times. The resulting change in oversize particle distribution was measured using a single particle optical sensing (SPOS) system. It was observed that the behavior of silica slurry when subjected to different shear rates falls into three regions namely de-agglomeration, transition region and agglomeration region. This technique provides a deterministic way for slurry stability and can be used as a guide for design of slurry, slurry selection and design of slurry delivery systems.

Study on pad cutting rate and surface roughness in diamond dressing process

Diamond dressing process has an important role in the elimination of scrap materials on the pad surface and regeneration of the pad surface topography before and after Chemical Mechanical Planarization/Polishing (CMP) process for integrated circuit (IC) fabrication. During diamond dressing process, distributions of cutting locus and overlap cutting of diamond grits usually work as a datum surface for pad surface planarization and uniform pad asperities. This paper proposes a novel view to predicting the distribution of pad cutting rate (PCR) and surface roughness based on cutting length and overlap cutting locus. The geometrical model of diamond dressing on the pad has been built and CMP tool parameters are used for the simulation of cutting length and overlap distribution. The distribution of PCR and predict surface roughness is calculated on each pad zone. Simulation results presented a high PCR and roughness is observed in center zones. The verified experiment is in agreement with simulation results where high PCR and roughness is near the center zones, which cause the shape of the pad surface to be concave. The results of this study can be further applied for prediction and optimization of diamond dressing design and improvement of dressing process.

Study on diamond dressing for non-uniformity of pad surface topography in CMP process

Diamond dressing process is critical in conditioning pad surface topography before and after chemical mechanical planarization/polishing (CMP) process for integrated circuit (IC) fabrication. This paper addresses a kinematic model of diamond dressing effect on the profile of pad cutting rate (PCR) with a ring-type diamond dresser through both simulation and experiments. In this kinematic model, the cutting locus distribution, relative velocity of diamond grits on a pad surface, and sliding time have been described as significant factors for non-uniformity of pad surface topography. Moreover, the speed ratio between the diamond dresser and the polishing pad has been investigated with respect to the center distance of the pad and the diamond dresser by the developed method. The model has been verified by experiments of diamond dressing of pad. Experimental results show that the dressing marks and pad cutting rate on the pad surface follow the same trend as simulation results and the final pad surface is obtained as a concave shape. Results of this study can be applied on diamond dressing of pads used in CMP and furthermore can be extended to investigate an optimal diamond dressing process for semiconductor fabrication.

Synergetic effect of organic cores and inorganic shells for core/shell structured composite abrasives for chemical mechanical planarization

Abstract Core/shell structured organic/inorganic composite microspheres has an important potential application in efficient and damage-free chemical mechanical planarization/polishing (CMP) as a kind of novel abrasive due to its uniform non-rigid mechanical property. However, the synergistic effect of material removal between organic cores and inorganic shells of composite abrasives is ambiguous. In this work, oxide-CMP performances of various slurries, containing polystyrene (PS) spheres, solid abrasives (SiO 2 or CeO 2 ), mixed abrasives ((PS + SiO 2 ) or (PS + CeO 2 )), core/shell composites (PS/SiO 2 or PS/CeO 2 ), were investigated by atomic force microscopy. Experiment results indicated that the surfaces polished by composite abrasives exhibited lower surface roughness, fewer scratches as well as lower topographical variations than those by other type of abrasives. The core/shell structure of composite abrasives plays an important role in improving CMP behavior. Moreover, the organic cores are mainly beneficial to decrease surface roughness and mechanical damages, and the inorganic shells are in favor of improving material removal rate.

Tribo-chemical Behavior of Copper in Chemical Mechanical Planarization

Chemical mechanical planarization/polishing (CMP) of copper has emerged as an important component in semiconductor processing. It involves both chemical and mechanical effects, consisting of several steps such as passivation, film dissolution, chemical corrosion, and abrasive abrasion. It is important to understand the individual steps in the removal process. In this paper, investigations were focused on the tribo-chemical behavior of copper in hydrogen peroxide solution at different pH values. The repassivation current and friction coefficient were measured in situ as a function of time. The compositions of unworn and worn surfaces after the tribocorrosion experiments were also analyzed by Raman spectra. The copper exhibited a tendency of repassivation during the potentiostatic tests; furthermore, the repassivation kinetics varied with the surface species and the quantity of passive films formed at different pH values. The results indicate that both relatively high material removal rate and good surface quality may be achieved in the weak acidic solution during CMP.

Material Removal Mechanism of Copper CMP from a Chemical–Mechanical Synergy Perspective

The material removal in chemical–mechanical planarization/polishing (CMP) of copper involves both chemical and mechanical effects. The roles of chemical corrosion, abrasive wear, and their synergistic effects on the material removal mechanism were studied by electrochemical analysis and nano-scratching method using atom force microscopy, respectively. Combining with the results of CMP experiments, dominant factors (chemistry and mechanics) in slurries within the range of pH 3.0–10.0 were assessed. Consequently, a removal mechanism map of copper CMP depending on pH values was constructed. In the alkaline slurry, the wear–corrosion effect predominated in the material removal at pH 8.0 and 9.0; while the copper removal mechanism changed to corrosion–wear action in the acidic slurry from pH 4.0 to 6.0, and good surface quality was also obtained. The results and the strategies provide thorough understanding of the material removal mechanism and further optimization of the CMP process.

Ultra-flat and ultra-smooth Cu surfaces produced by abrasive-free chemical–mechanical planarization/polishing using vacuum ultraviolet light

Abstract The new bonding technologies utilizing intermolecular bonding forces have been developed and attracting attention recently. Cu is known to be a suitable material for the bonding substrate due to its excellent physical properties. And an ultra flatness and an ultra smoothness over a relatively large area are strongly required for the Cu substrate surface. Chemical–mechanical planarization/polishing (CMP) with abrasives is widely adopted for planarizing and smoothing Cu surfaces. But this method has serious problems resulting from abrasives in CMP slurry. Hence, we have developed an abrasive-free polishing (AFP) method that utilizes vacuum ultraviolet (VUV) light in the previous study and an ultra-smooth Cu surface was achieved. However, the problems about a low removal rate and a small finished area remained. To overcome the problem, a new manufacturing process, namely, the process of combining CMP with abrasives and the AFP method was newly developed. First, an ultra-flat surface is achieved using CMP with abrasives. Next, the AFP method is applied for the final polishing step in order to achieve an ultra-smooth surface. As a result, utilizing VUV in situ irradiation and electrolyzed reduced water in the AFP process, the ultra-flat and the ultra-smooth surface produced has a roughness average of

Investigation on the controllable growth of monodisperse silica colloid abrasives for the chemical mechanical polishing application

With the rapid development of chemical mechanical polishing technique as well as its increasing application in IC foundry, the abrasives of slurry are required to have different specifications in terms of size and size’s distribution, which play a vital role in the material removal and defect control. In this study, we monitor in detail the growth process of colloidal silica abrasives changing from the tiny nuclei to large nanoparticles by means of the electron microscopy images. Using the procedure we develop, we are capable of producing monodisperse colloidal silica nanoparticles ranging from 60 to 130 nm in diameter, which are mostly often applied as abrasives in chemical mechanical planarization/polishing (CMP) process of integrated circuit (IC) manufacturing. The physicochemical properties of the silica synthesized by our procedure are also characterized by the X-ray diffraction (XRD) patterns and thermal analysis. The polishing test adopting the colloidal silica as abrasives is performed on silicon wafer to evaluate the CMP properties.

Synchronous, In Situ Measurements in Chemical Mechanical Planarization

Achieving device- and wafer-scale planarity in integrated circuit manufacturing is increasingly challenging as device sizes are reduced to and smaller, and wafer sizes are increased from . Hence a better understanding of chemical mechanical planarization (CMP) processes is needed. We report the synchronous, in situ measurement of wafer forces [coefficient of friction (CoF)] and wafer orientation during CMP polishing of BK7 glass wafers and the ex situ measurement of material removal rate (MRR) over a wide range of applied vertical loads and relative rotational velocities. MRR appears to be Prestonian and was significantly reduced by moving the slurry injection location to the outer edge of the wafer track leading to slurry starvation. Both positive and negative wafer pitch angles were measured, indicating that apparently contradictory modeling reports in the literature may not be in conflict. A phenomenological linkage between wafer orientation and frictional behavior was observed with positive pitch angles (from to ) corresponding to smooth polishing and nose-down angles (from to ) corresponding to stick-slip. Spectral signal analysis shows that the variation of CoF is due to platen rotation effects, while wafer orientation variation is affected by both platen and wafer rotations.