Revisiting the Removal Rate Model for Oxide CMP

Sorooshian, J.; Borucki, L.; Stein, David J.; Timon, Robert P.; Hetherington, Dale L.; Philipossian, Ara

doi:10.1115/1.1866168

Cited by 34 publications

(40 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Based on the initial experimental conditions from which the model was developed, this study attempted to verify the application and use of the model using alternate consumables and an alternate tool set. 5 Removal rate results from this study showed similar systematic scattering effects to in the results shown from the original study. By applying the flash heating model to the results acquired from this study, it was evident that the predictive capability of the flash heating model was more accurate than previously established models.…”

Section: Discussionsupporting

confidence: 87%

See 1 more Smart Citation

Extending the Flash Heating Removal Rate Model to Various Interlayer Dielectric CMP Consumables

Sorooshian

Philipossian

2005

J. Electrochem. Soc.

Self Cite

View full text Add to dashboard Cite

This study extends the application of a previously presented flash heating removal rate model to a series of interlayer dielectric ͑ILD͒ chemical mechanical planarization ͑CMP͒ experiments using Freudenberg XY and perforated groove pads along with Fujimi PL-4217 fumed silica slurry. Polishing tests were conducted at platen temperatures ranging from approximately 24 to 45°C. Application of the model indicates its versatility when using different tool sets, pad types, and slurry chemistries compared to those originally used to develop the model. The flash heating removal rate model was shown to predict experimental data from this study by an average root mean square ͑rms͒ error of 164 Å/min. When compared to the traditional Preston model, which predicted the data by an average rms of 300 Å/min, the flash heating model provides better predictive accuracy and enables one to predict previously assumed scatter associated with removal rate in a systematic and comprehensible manner.Since the development of chemical mechanical planarization ͑CMP͒ in the 1960s, the ability to appropriately model material removal rate has taken several positive steps. Preston's initial observation that material removal during glass polishing was proportional to the pressure and velocity of the process has long been the basis for a majority of the physical models used to predict removal rate in CMP. 1 Models presented by Zhang and Busnaina as well as Tseng and Wang have both attempted to advance the capabilities of determining removal rate data, however each have their own respective shortcomings. 2,3 These downfalls primarily come from the simple fact that experimental removal rate results from CMP often indicate signs of scatter. In this situation, removal rate scatter is defined as any deviation from an average value for a single process condition. Past studies have often neglected the effects of removal rate scatter as an artifact of inconsistencies associated with the CMP tool or the consumables used. Using this assumption, removal rate models have had relative success at achieving the goal of predicting these trends.In an attempt to qualify and compare a series of removal rate models, Stein and Hetherington evaluated series removal rate data acquired from a set of interlayer dielectric ͑ILD͒ polishes on a Speedfam IPEC-472 platform with Rohm and Haas IC-1400 K-groove pad and Cabot SS-12 fumed silica slurry. 4 In their work, the removal rate data taken as the basis for the comparison appeared to be Prestonian, albeit with a discernable amount of scatter ͑ranging anywhere from approximately 1800 to 30 Å/min for a single pressure and velocity condition͒, which was assumed to be an artifact of intricacies involved in the polishing process ͑i.e., slight changes in flow rate, conditioning mechanisms, and pressure͒. When the removal rate data presented by Stein and Hetherington are grouped according to sliding velocity, distinct trends in the removal rate can be seen with increasing pressure ͑see Fig. 1͒. This suggests that what at first appe...

show abstract

Section: Discussionsupporting

confidence: 87%

“…5 The study involved a set of ILD polishes conducted on a Speedfam IPEC-472 platform with a Rohm and Haas IC-1400 K-groove pad and Cabot D7300 fumed silica slurry. Removal rate results from the study showed consistent and significant deviations from Prestonian behavior.…”

mentioning

confidence: 99%

Extending the Flash Heating Removal Rate Model to Various Interlayer Dielectric CMP Consumables

Sorooshian

Philipossian

2005

J. Electrochem. Soc.

Self Cite

View full text Add to dashboard Cite

show abstract

“…In aqueous systems with constant mechanical interactions, it is difficult to determine the rate-limiting processes. Many semi-empirical CMP models are based on this mechanism by including a fractional surface coverage term in the chemical and mechanical removal rate expression [30][31][32][33]. However, there is room for improvement in the existing models if the chemistry of the oxidation, complexation, and inhibition mechanisms could be better defined, to include rate constants and activation energies that can be found in the literature.…”

Section: Introductionmentioning

confidence: 98%

Characterization of copper-hydrogen peroxide film growth kinetics

DeNardis¹,

Rosales-Yeomans²,

Borucki³

et al. 2006

Thin Solid Films

View full text Add to dashboard Cite

“…The decrease, 2000 Å /min, is about 10 times the experimental error. Similar fluctuations can also occur in silicon dioxide polishing data [19]. Figure 6.5 shows a contour plot of the same removal rate data as a function of p and V separately (solid lines) with contours of constant pV superimposed for comparison (dashed lines).…”

Section: A Polishing Examplementioning

confidence: 83%

“…The model contains five parameters, E, A, c p , b, and e, but there is a problem with using E as a fitting parameter. The problem is that a polishing experiment conducted at room temperature in which the chemical rate constant (6.40b) and the reaction temperature T are not directly measured does not provide enough information to reliably estimate E. In fact, for any assumed value of E over a reasonable range (say 0.25-1.1 eV), optimization of the remaining four parameters in the above theory will produce nearly identical root mean square (RMS) fitting errors [19]. The reason for this is that over a narrow range of temperatures, such as in Fig.…”

Section: A Polishing Examplementioning

confidence: 99%

Modeling

Borucki

Philipossian

2007

Microelectronic Applications of Chemical Mechanical Planarization

Self Cite

View full text Add to dashboard Cite

Experimental evidence strongly suggests that material removal in chemicalmechanical polishing (CMP) processes is a result of one or more chemical steps that alter the wafer surface combined with a mechanical step that removes the altered material. Chemical action by itself also removes material by static etching, but generally at a much lower rate than is observed when mechanical action is also present. Similarly, polishing rates observed when a minimally reactive fluid such as water is used instead of slurry are also low. Both chemical and mechanical processes are therefore involved in material removal at commercially practical rates, and the model we describe reflects this dual nature of the process.In this chapter, we derive and apply to data a simple two-step model that involves a chemical step followed by a mechanical removal step. The model is abstract in the sense that most of the specifics of the slurry composition or of the chemical reaction involved are not given in any detail. Although this appears to be a disadvantage, it is necessary for the application of the model to the analysis of removal rates from proprietary slurries whose compositions cannot be directly investigated. When applied as a compact formula, the model can provide a highly accurate description of removal rate variations as a function of polishing pressure and sliding speed. This then makes it possible to extract the relative contributions of chemical and mechanical processes to removal and to confidently interpolate or extrapolate rates based on the calibration data.

show abstract

Revisiting the Removal Rate Model for Oxide CMP

Cited by 34 publications

References 12 publications

Extending the Flash Heating Removal Rate Model to Various Interlayer Dielectric CMP Consumables

Extending the Flash Heating Removal Rate Model to Various Interlayer Dielectric CMP Consumables

Characterization of copper-hydrogen peroxide film growth kinetics

Modeling

Contact Info

Product

Resources

About