Machine learning–driven in situ process monitoring with vibration frequency spectra for chemical mechanical planarization

Abstract

The objective of this work is to tackle the challenges of monitoring and detecting subtle process changes in chemical mechanical planarization (CMP), an ultraprecision manufacturing process. Monitoring ultraprecision processes is usually of difficulty due to their innate complexity and low signal-to-noise ratio in sensor signals. Especially for subtle signal variations during small process changes, the conventional statistical process control charts could fail to detect such process anomalies from the sensor signals in the time domain. In this paper, frequency spectra representation of the microelectromechanical systems (MEMS) vibration sensor signals during subtle process changes is investigated, and the signal patterns uncovered by frequency spectra are utilized to formulate a machine learning–driven in situ process monitoring approach to detect process anomalies in CMP. The proposed approach overcomes the obstacles of differentiating subtle signal changes by transforming them into the frequency domain with Fourier transform and Hilbert-Huang transform and classifying the resulted frequency spectra with random forest. Based on frequency analysis, it can unveil the differences in the signals obscured in the time domain and suppress the high-frequency noise. Consequently, the presented machine learning–driven in situ process monitoring approach detects process anomalies by differentiating the deviated frequency spectra with machine learning. It is validated on our experimental CMP testbed for anomaly detection, and outperforms three benchmark statistical process control charts. For instance, it detects a slurry shutoff anomaly in CMP about ten times faster than the benchmark methods.

Full text

ORIGINAL ARTICLE
Machine learning–driven in situ process monitoring with vibration
frequency spectra for chemical mechanical planarization
Jia (Peter) Liu
1
&Jingyi Zheng
2
&Prahalada Rao
3
&Zhenyu (James) Kong
4
Received: 8 June 2020 /Accepted: 24 September 2020
#Springer-Verlag London Ltd., part of Springer Nature 2020
Abstract
The objective of this work is to tackle the challenges of monitoring and detecting subtle process changes in chemical mechanical
planarization (CMP), an ultraprecision manufacturing process. Monitoring ultraprecision processes is usually of difficulty due to
their innate complexity and low signal-to-noise ratio in sensor signals. Especially for subtle signal variations during small process
changes, the conventional statistical process control charts could fail to detect such process anomalies from the sensor signals in
the time domain. In this paper, frequency spectra representation of the microelectromechanical systems (MEMS) vibration sensor
signals during subtle process changes is investigated, and the signal patterns uncovered by frequency spectra are utilized to
formulate a machine learning–driven in situ process monitoring approach to detect process anomalies in CMP. The proposed
approach overcomes the obstacles of differentiating subtle signal changes by transforming them into the frequency domain with
Fourier transform and Hilbert-Huang transform and classifying the resulted frequency spectra with random forest. Based on
frequency analysis, it can unveil the differences in the signals obscured in the time domain and suppress the high-frequency noise.
Consequently, the presented machine learning–driven in situ process monitoring approach detects process anomalies by differ-
entiating the deviated frequency spectra with machine learning. It is validated on our experimental CMP testbed for anomaly
detection, and outperforms three benchmark statistical process control charts. For instance, it detects a slurry shutoff anomaly in
CMP about ten times faster than the benchmark methods.
Keywords Chemical mechanical planarization .Ultraprecision manufacturing .Vibration sensor .Frequency spectra .Machine
learning .Anomaly detection
1 Introduction
Chemical mechanical planarization (CMP) is a critical back-
end-of-line process among the sequential processes in semi-
conductor manufacturing. It exerts both mechanical erosion
and chemical corrosion on semiconductor wafers to achieve
nanoscale morphological surface finish [40]. The resulted sur-
face finish from CMP significantly impacts the quality of final
wafers, which will be etched and deposited to manufacture
interconnected layers for semiconductor integrated circuits
[28]. Since CMP is an ultraprecision process, even small pro-
cess drifts could cause severe wafer defects, such as scratches
and burnt surface (in Fig. 1), and lead to wafer scrap and
material waste. Inferior quality of wafers from CMP could
lead to a ~ 35% reduction in throughput and cause as much
as a 100% increase in the cost of ownership [2,38]. It is
therefore desirable to implement in situ process monitoring
in CMP to identify the onset of process anomalies in real time,
enabling corrective actions at an early stage and mitigating
wafer defects and yield losses.
*Jia (Peter) Liu
liu@auburn.edu
Jingyi Zheng
jingyi.zheng@auburn.edu
Prahalada Rao
rao@unl.edu
Zhenyu (James) Kong
zkong@vt.edu
1
Department of Industrial and Systems Engineering, Auburn
University, Auburn, AL 36849, USA
2
Department of Mathematics and Statistics, Auburn University,
Auburn, AL 36849, USA
3
Mechanical and Materials Engineering Department, University of
Nebraska-Lincoln, Lincoln, NE 68588, USA
4
Grado Department of Industrial and Systems Engineering, Virginia
Tech, Blacksburg, VA 24061, USA
https://doi.org/10.1007/s00170-020-06165-1
/ Published online: 19 October 2020
The International Journal of Advanced Manufacturing Technology (2020) 111:1873–1888
Content courtesy of Springer Nature, terms of use apply. Rights reserved.