Abstract
A method to predict plasma etch profile nonuniformity is presented. This was accomplished by using a neural network and a wavelet. The wavelet was used to define a metric of the profile nonuniformity. The method was applied to the etching of tungsten films in a helicon SF 6 plasma. The etch process was characterized by a 2 4-1 fractional factorial experiment. The process parameters that were varied in the design include the radio-frequency source power, the bias power, the substrate temperature, and the SF 6 flow rate. The fluorine concentration [F] measured using optical emission spectroscopy was related to the profile nonuniformity. The model prediction accuracy was optimized as a function of training factors, and the optimized model had a root-mean squared error of 6.43 %. Using the optimized model, we qualitatively estimated etch mechanisms. Decreasing each process parameter generally reduced the profile nonuniformity. For variations either in the source power or temperature, both the profile nonuniformity and [F] were highly correlated. The presented method can be applied to characterize any plasma-processed surfaces.
Original language | English |
---|---|
Pages (from-to) | 817-821 |
Number of pages | 5 |
Journal | Journal of the Korean Physical Society |
Volume | 43 |
Issue number | 5 II |
Publication status | Published - 2003 Nov 1 |
Fingerprint
Keywords
- Neural network
- Plasma etching
- Profile uniformity
- Wavelet
ASJC Scopus subject areas
- Physics and Astronomy(all)
Cite this
Qualitative Interpretation of Etch Profile Nonuniformity Using a Wavelet and a Neural Network. / Lee, Hak Sung; Kim, Byungwhan; Choi, Serk Rim; Hong, Wan Shick; Lee, Kyeong Kyun; Choi, Won Sun; Lim, Myo Taeg.
In: Journal of the Korean Physical Society, Vol. 43, No. 5 II, 01.11.2003, p. 817-821.Research output: Contribution to journal › Article
}
TY - JOUR
T1 - Qualitative Interpretation of Etch Profile Nonuniformity Using a Wavelet and a Neural Network
AU - Lee, Hak Sung
AU - Kim, Byungwhan
AU - Choi, Serk Rim
AU - Hong, Wan Shick
AU - Lee, Kyeong Kyun
AU - Choi, Won Sun
AU - Lim, Myo Taeg
PY - 2003/11/1
Y1 - 2003/11/1
N2 - A method to predict plasma etch profile nonuniformity is presented. This was accomplished by using a neural network and a wavelet. The wavelet was used to define a metric of the profile nonuniformity. The method was applied to the etching of tungsten films in a helicon SF 6 plasma. The etch process was characterized by a 2 4-1 fractional factorial experiment. The process parameters that were varied in the design include the radio-frequency source power, the bias power, the substrate temperature, and the SF 6 flow rate. The fluorine concentration [F] measured using optical emission spectroscopy was related to the profile nonuniformity. The model prediction accuracy was optimized as a function of training factors, and the optimized model had a root-mean squared error of 6.43 %. Using the optimized model, we qualitatively estimated etch mechanisms. Decreasing each process parameter generally reduced the profile nonuniformity. For variations either in the source power or temperature, both the profile nonuniformity and [F] were highly correlated. The presented method can be applied to characterize any plasma-processed surfaces.
AB - A method to predict plasma etch profile nonuniformity is presented. This was accomplished by using a neural network and a wavelet. The wavelet was used to define a metric of the profile nonuniformity. The method was applied to the etching of tungsten films in a helicon SF 6 plasma. The etch process was characterized by a 2 4-1 fractional factorial experiment. The process parameters that were varied in the design include the radio-frequency source power, the bias power, the substrate temperature, and the SF 6 flow rate. The fluorine concentration [F] measured using optical emission spectroscopy was related to the profile nonuniformity. The model prediction accuracy was optimized as a function of training factors, and the optimized model had a root-mean squared error of 6.43 %. Using the optimized model, we qualitatively estimated etch mechanisms. Decreasing each process parameter generally reduced the profile nonuniformity. For variations either in the source power or temperature, both the profile nonuniformity and [F] were highly correlated. The presented method can be applied to characterize any plasma-processed surfaces.
KW - Neural network
KW - Plasma etching
KW - Profile uniformity
KW - Wavelet
UR - http://www.scopus.com/inward/record.url?scp=0344551786&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=0344551786&partnerID=8YFLogxK
M3 - Article
AN - SCOPUS:0344551786
VL - 43
SP - 817
EP - 821
JO - Journal of the Korean Physical Society
JF - Journal of the Korean Physical Society
SN - 0374-4884
IS - 5 II
ER -