Speaker
Seolhye Park
Description
See the full Abstract at http://ocs.ciemat.es/EPS2018ABS/pdf/I1.304.pdf
Application of the PI-VM (Plasma Information based Virtual Metrology)
for management of the plasma processes in OLED display manufacturing
S. Park1,2, T. Cho1, J. Lee1, Y. Jang1, J.-J. Hong1, W.-H. Jang1, and G.-H. Kim2
1
Samsung Display Co., Ltd, Chungcheongnam-do, Korea
2
Seoul National University, Seoul, Korea
Plasma processes applied to the display- and semiconducting device manufacturing must be
monitored by virtual metrology (VM) to maintain the process results and increase the
throughput of the processes. Because these processes can be managed according to the VM
results, the prediction accuracy of the VM models predicting such as etch- and deposition
rate, defect particles, etc. are very important. The core algorithms of the existing VM model
are based on statistical methods that analyse the correlation between sensing variables and
performance variables in the widely-known big data pool of the fab. However, in identified
sensing variables obtained from the engineering equipment system (EES), and other sensors,
such as for I-V signal, and optical raw signals, the information about the reacting plasma in
the process reactor is not efficiently included. The inclusion of a ‘good’ parameter, which
efficiently contains information about the state of the process, is important for ensuring the
accuracy of the VM model; therefore, the performance of a statistical VM, without
consideration of the process plasma information, cannot satisfy industrial requirements of
prediction accuracy for the high-definition organic light emitting diode (OLED) display
manufacturing processes. To enhance the performance of the VM model, three types of
reactions in the plasma were parameterized as the core variables of the VM model; therefore,
plasma information (PI) parameters representing the reaction properties in the plasma
volume, sheath, and surface were applied to the VM algorithm and named as the PI-VM [1].
In OLED display manufacturing processes, PI-VM has shown a noticeably enhanced
performance (R2 > 90%) for the dry etching amount prediction compared to the existing
statistical VM (R2 ~ 50%). In OLED manufacturing fab, various problems occurred during
the mass production were predicted and analysed by the application of PI-VM algorithms
modelled according to the characteristics of each process plasmas, and they would be
introduced in the conference.
[1] S. Park, S. Jeong, Y. Jang, S. Ryu, H. –J. Roh, G. –H. Kim, “Enhancement of the Virtual Metrology
Performance for Plasma-Assisted Oxide Etching Processes by Using Plasma Information (PI) Parameters”,
IEEE Trans. Semiconductor Manufacturing, vol. 28, pp. 241-246, August 2015.
