Effect of Laser Shock Wave Cleaning Direction on REFERENCES Particle Removal Behavior at Trenchs 1. J. M. Lee and K. G. Watkins, J. Appl. Phys. 89, 6496 Jin-Su Kim a, Ahmed A. Busnaina b and Jin-Goo Park a (2001) 2. S. H. Lee, et al, Jap. J. Applied Physics, Part 1 44, a Department of Materials Engineering and Bionano 5560 (2005) Technology, Hanyang University, Ansan 426-791, Korea 3. T. G. Kim et al, Microelectronic Engineering 83, b Center for Microcontamination Control, Northeastern 688(2006) University, Boston, MA 02115 4. H. Lim, et al, J. Appl. Phys. 97, 054903 (2005) ABSTRACT Dry cleaning has been studied to control the removal force on pattern wafers as device feature sizes decrease to below 100 nm. Laser shock wave cleaning (LSC) has been introduced to remove particles on patterned wafers . In this method particles are removed by shock wave Fig. 1. Schematic diagram of a laser shock wave system. which generated by a focused laser beam having biconvex lens as shown in Fig. 1. LSC can be applied to BEOL, local cleaning of wafer and extreme ultraviolet (EUV) lithography mask . For the application of LSC to EUV mask cleaning or any BEOL cleaning process, the questions have been raised how the particles behave when patterns exist. For example, EUV lithography mask contains various layers with their topographies as shown Fig. 2. Cross sectional pattern topography of a EUVL in Fig. 2. This topography might create “shadow effect” mask. during cleaning process . In this paper, the effect of pattern topography on the particle removal was investigated on trench patterns using LSC. A laser shock wave was generated by a Q-switched Nd:YAG laser (IMT, Korea) with a maximum pulse energy of 1800 mJ at a wavelength of 1064 nm. To characterize the topographical effect on LSC, samples with trench patterns were aligned for and against the shockwave propagation direction as shown in Figure 3. The pattern width and aspect ratio were 5 µm and 0.25, respectively. PSL microspheres (500 nm red fluorescence, (a) (b) Duke Scientific, USA) were used as the particulate Fig. 3. LSC application in (a) horizontal and (b) vertical contaminant source and were deposited on the STI direction . patterned wafers by dipping them into the IPA solution which contained PSL microspheres. Cleaning performances such as PRE and cleaning area were 80 analyzed by a fluorescence microscope (LV-150, Nikon, 70 Japan). 60 The topographical effect was characterized as a 50 function of laser shock wave propagation direction. Fig. 4 PRE(%) 40 shows the comparison of particle removal efficiency (PRE) in between horizontal and vertical direction of laser 30 shock wave to the trench patterns. PRE was lower in case 20 of vertical direction when compared to horizontal 10 direction. It means that the topography of a trench is 0 critical in removing particles. Fig. 5 (a) and (b) shows the Horizontality Verticality optical microscopic images of the particles remained after cleaning by LSC in vertical and horizontal direction, Fig. 4. Comparison of PRE in between horizontal and respectively. It could be observed that LSC in vertical vertical direction of laser shock wave. direction left more particles than horizontal and also in both cases the remained particles were present at the trench side wall. Based on these results, it could be concluded that LSC in horizontal direction has more PRE than in vertical direction and hence topography is an important consideration during LSC. ACKNOWLEDGEMENTS This work was supported by the Korea Science and Engineering Foundation (KOSEF) grant funded by the (a) (b) Fig. 5. Optical microscope images after LSC in (a) Korea government (MEST) (No. R11-2008-044-00000-0). vertical direction and (b) horizontal direction.