Experimental Investigation of Pressure and Velocity on Chemical Mechanical Planarization

2000 ◽  
Author(s):  
Jhy-Cherng Tsai ◽  
Mingyi Tsai
Keyword(s):  
Experimental Investigation ◽  
Taguchi Method ◽  
Material Removal Rate ◽  
Material Removal ◽  
Removal Rate ◽  
Chemical Mechanical Planarization ◽  
Theoretical Models ◽  
Back Pressure ◽  
Wear Model ◽  
Polishing Pressure

Abstract Chemical-mechanical planarization or polishing (CMP) is an emerging process used in surface planarization and polishing for semiconducter wafer with multilevel interconnections. This paper investigates the effects of polishing pressure and velocity on the material removal rate (RR) and the non-uniformity (NU) in the CMP process. Wear models for CMP process from mechanical aspect, including abrasive-based model and flow-based model, are first discussed. Experiments using different polishing pressures, velocities, and back pressures are then designed and conducted based on the Taguchi method. Experimental results showed that RR and NU are consistent with theoretical models in a certain range. Both RR and NU increase as polishing speed increases. As RR and NU also increase with the polishing pressure at low pressure, their values become saturated and then decrease when the pressure exceeds certain value. It further verified that NU can be improved using proper back pressure as predicted by the flow-based wear model.

Experimental Investigation on Effects of Passivants in the Abrasive-Free Polishing of Copper Film

2010 ◽  
Vol 126-128 ◽  
pp. 316-319
Author(s):  
Jhy Cherng Tsai ◽  
Wei Ching Lin
Keyword(s):  
Experimental Data ◽  
Experimental Investigation ◽  
Citric Acid ◽  
Material Removal Rate ◽  
Material Removal ◽  
Removal Rate ◽  
Copper Film ◽  
Copper Base ◽  
Polishing Pressure ◽  
Better Than

Abrasive Free Polishing (AFP) is a polishing technology without abrasives and widely employed in copper-base semiconductor fabrications. This paper investigates the effect of passivants, added to the slurry, on the material removal rate (MRR) and the non-uniformity (NU) via experiments. Two kinds of passivants, Benzotriazole (BTA) and citric acid (CA) are added to the slurry for the experiments. Experimental results showed that the MRR increases when polishing pressure increases while NU decreases at the same time. Both MRR and NU tends to increase when rotational speed increases, though MRR and NU at 40 rpm are lower than that at 30 and 50 rpm in the slurry with CA. Experimental data also showed that AFP using the slurry with CA performs better than that with BTA.

Experimental Investigation on Wafer Polishing of Oxide and Copper Layers

2002 ◽  
Author(s):  
Jhy-Cherng Tsai
Keyword(s):  
Process Parameters ◽  
Chemical Mechanical Polishing ◽  
Material Removal ◽  
Removal Rate ◽  
Back Pressure ◽  
Copper Layer ◽  
Wafer Polishing ◽  
Critical Technology ◽  
And Performance ◽  
Polishing Pressure

Polishing, in particular chemical-mechanical polishing (CMP), is a critical technology for the planarization of wafers. This paper investigates, via experiments, and compares the performance of CMP process with different process parameters for wafers with silicon-dioxide (SiO2) layer and for wafers with copper (Cu) layer. Polishing pressure (P), speed (V), and back pressure (BP) are used as process parameters in this study. Different pads and slurries are also experimented for copper layer as its properties are different from that of conventional oxide layer. Material removal rate (RR) and non-uniformity (NU) are used as indices to measure the performance. Experimental data on oxide layers show RR increases as P and V increase but NU gets worse at the same time. This condition can be improved, for both oxide and copper layers, with suitable BP. Experiments on copper CMP using slurry with abrasives show that RR increases with higher P and V. While NU gets worse with higher P, it can be reduced as V increases using a soft pad. Better NU can be obtained using soft pad though RR is lower in this case. For abrasive-free polishing of copper layer, RR, though relatively lower compared to CMP with regular slurry, is unstable using hard pad despite that NU becomes better at higher P. NU of polished wafer is best at certain pressure but becomes worse at low pressure for hard pad and at high P for soft pad. It is also observed that NU of AFP can be improved with BP and softer pad. Soft pad gives better polishing quality and performance though RR is lower than that using slurry with abrasives.

Sign in / Sign up

Export Citation Format

Share Document