In-situ CMP Endpoint Detection Using Acoustic Emission

Helu, Moneer; Chien, Joshua; Dornfeld, David

doi:10.1016/j.procir.2014.03.025

Cited by 9 publications

(7 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The need for such measurements requires the fitting of the polisher with costly friction plates (which are quite large and complex, and may not always be suitable for certain types of polishers), force sensors, multiplexers and signal amplifiers all of which need periodic calibrations. The scientific literature is filled with alternate methods [16][17][18][19][20][21][22][23][24][25][26][27][28][29][30] of inferring COF in CMP with the most popular of such approaches being centered around measuring platen motor current (PMC). During CMP, the wafer and the platen corotate at desired angular velocity setpoints.…”

Section: Introductionmentioning

confidence: 99%

Correlating Coefficient of Friction and Shear Force to Platen Motor Current in Tungsten and Interlayer Dielectric Chemical Mechanical Planarization at Steady-State Conditions

Headley¹,

Frank²,

Sampurno³

et al. 2020

ECS J. Solid State Sci. Technol.

View full text Add to dashboard Cite

Following up our earlier work that highlighted the relationship between shear force (SF) and platen motor current (PMC), in parallel with the relationship between coefficient of friction (COF) and PMC for various tungsten and interlayer dielectric (ILD) chemical mechanical planarization (CMP) cases at non-steady-state conditions, we explored whether or not PMC could be used as a reliable indicator instead of SF and COF at steady-state conditions. For the 12 cases studied, 72 distinct steps were analyzed. It was determined that PMC somewhat mirrored SF and PMC for long time (i.e. 10 s or longer) intervals after data averaging and applying a trend matching algorithm. SF and PMC trends matched only about 64% of the time (ranges between 45% to 85%) for all 72 steps, while PMC and COF trends matched 62% of the time ranging between 42% and 85%. PMC-SF and PMC-COF correlations were fairly poor at 1-sec time intervals as evidenced by much lower percent match values. Such poor correlations proved that at small time intervals, PMC was not sensitive enough to capture important information regarding myriad fluid dynamics and tribological phenomena and the instantaneous stick-slip occurrences encountered in CMP.

show abstract

Section: Introductionmentioning

confidence: 99%

Correlating Coefficient of Friction and Shear Force to Platen Motor Current in Tungsten and Interlayer Dielectric Chemical Mechanical Planarization at Steady-State Conditions

Headley¹,

Frank²,

Sampurno³

et al. 2020

ECS J. Solid State Sci. Technol.

View full text Add to dashboard Cite

show abstract

“…That the polisher's intuition of when the surface condition plateaus out and when the pad damage sets in plays a critical role in deciding when to stop the polishing process. Consequently, significant process cycle time is consumed by repeated stoppage and surface inspections (visual or through the use of instruments) [8,9]. Quantitative surface roughness metrics do exist, and the most commonly used during polishing in industry is the average roughness parameter, denoted by Ra for one-dimensional profiles or Sa for two-dimensional areas [10].…”

Section: Introductionmentioning

confidence: 99%

A Gaussian Process Model-Guided Surface Polishing Process in Additive Manufacturing

Jin

Iquebal

Bukkapatnam

et al. 2019

Journal of Manufacturing Science and Engineering

View full text Add to dashboard Cite

Polishing of additively manufactured products is a multi-stage process, and a different combination of polishing pad and process parameters is employed at each stage. Pad change decisions and endpoint determination currently rely on practitioners’ experience and subjective visual inspection of surface quality. An automated and objective decision process is more desired for delivering consistency and reducing variability. Toward that objective, a model-guided decision-making scheme is developed in this article for the polishing process of a titanium alloy workpiece. The model used is a series of Gaussian process models, each established for a polishing stage at which surface data are gathered. The series of Gaussian process models appear capable of capturing surface changes and variation over the polishing process, resulting in a decision protocol informed by the correlation characteristics over the sample surface. It is found that low correlations reveal the existence of extreme roughness that may be deemed surface defects. Making judicious use of the change pattern in surface correlation provides insights enabling timely actions. Physical polishing of titanium alloy samples and a simulation of this process are used together to demonstrate the merit of the proposed method.

show abstract

“…On the other hand, there are many alternate methods of inferring frictional effects in CMP that have been reported on in the scientific literature. [24][25][26][27][28][29][30][31][32][33][34][35][36][37][38][39] The most popular of such methods measures the carrier or the platen motor current (CMC or PMC). [40][41][42][43][44][45][46][47][48][49][50][51][52][53][54] During polishings, both wafer and platen rotate at desired velocity setpoints.…”

mentioning

confidence: 99%

Correlating Shear Force and Coefficient of Friction to Platen Motor Current in Copper, Cobalt, and Shallow Trench Isolation Chemical Mechanical Planarization at Steady-State Conditions

Frank

Headley

Sampurno

et al. 2020

ECS J. Solid State Sci. Technol.

View full text Add to dashboard Cite

In a follow up from our earlier work that focused on highlighting the relationship among platen motor current (PMC), shear force (SF), and coefficient of friction (COF) in non-steady-state cases, we examined whether PMC was a reliable indicator that could be used in place of SF and COF data at steady-state conditions. For the 12 cases studied, we examined 60 distinct steady-state steps from their associated Stribeck+ curves. Data averaging, coupled with a trend matching algorithm, showed that for large time spans at steady-state, PMC was a reasonably good indicator of both SF and COF, as 68.5% of the time, PMC and SF trends matched (ranges between 62% to 86%) for all 60 cases. Regarding PMC and COF, we observed trend matching for 68.8% of the time (ranges between 62% and 86%) for all 60 cases. However, correlations between SF and PMC, and also between COF and PMC were found to be very poor (in most cases, non-existent) since at small timescales PMC was not sensitive enough to capture the instantaneous stick-slip occurrences and other important tribological and fluid dynamics phenomena present in CMP.

show abstract

In-situ CMP Endpoint Detection Using Acoustic Emission

Cited by 9 publications

References 13 publications

Correlating Coefficient of Friction and Shear Force to Platen Motor Current in Tungsten and Interlayer Dielectric Chemical Mechanical Planarization at Steady-State Conditions

Correlating Coefficient of Friction and Shear Force to Platen Motor Current in Tungsten and Interlayer Dielectric Chemical Mechanical Planarization at Steady-State Conditions

A Gaussian Process Model-Guided Surface Polishing Process in Additive Manufacturing

Correlating Shear Force and Coefficient of Friction to Platen Motor Current in Copper, Cobalt, and Shallow Trench Isolation Chemical Mechanical Planarization at Steady-State Conditions

Contact Info

Product

Resources

About