Shamban & Microsystems TRIBOLOGY LABS Department of Mechanical Engineering Technion - Israel Institute of Technology The The The The problem problem problem problem of of of of backside backside backside backside wafer wafer wafer wafer contamination contamination contamination contamination is is is is a significant significant significant significant obstacle obstacle obstacle obstacle in in in in moving moving moving moving from from from from 300 300 300 300 mm mm mm mm to to to to 450 450 450 450 mm mm mm mm wafer wafer wafer wafer. The The The The backside backside backside backside particles particles particles particles degrade degrade degrade degrade flatness flatness flatness flatness and and and and create create create create ‘hot ‘hot ‘hot ‘hot spots’ spots’ spots’ spots’ which which which which later later later later become become become become lithography lithography lithography lithography defects defects defects defects on on on on the the the the front front front front side side side side of of of of the the the the wafer, wafer, wafer, wafer, where where where where the the the the actual actual actual actual dies dies dies dies are are are are built built built built. One One One One of of of of the the the the main main main main reasons reasons reasons reasons of of of of this this this this problem problem problem problem is is is is the the the the wear wear wear wear of of of of vacuum vacuum vacuum vacuum nipples nipples nipples nipples due due due due to to to to cyclic cyclic cyclic cyclic loading loading loading loading in in in in chucking chucking chucking chucking incoming incoming incoming incoming wafers wafers wafers wafers. Here Here Here Here we we we we study study study study the the the the possibility possibility possibility possibility of of of of wear wear wear wear reduction reduction reduction reduction by by by by proper proper proper proper material material material material choice choice choice choice and and and and optimization optimization optimization optimization of of of of surface surface surface surface topography topography topography topography of of of of wafer wafer wafer wafer chucks chucks chucks chucks. Fretting motion as a reason of wear Test rig adjustments and sample preparation Reduction of Vacuum Nipple Wear Michael Varenberg, Yuri Kligerman, Haytam Kasem, Grigory Halperin, Alexey Tsipenyuk, Peter Breitman, Tamir Akhavan, Ariella Mansfield & Noam Rozen Examination and testing of original chuck surfaces Theoretical modeling of surface damage To enable cyclic normal and tan- gential loading of contact projec- tions used in wafer chucking, we have modified our test rig built for studies of real contact area for working in a flat-on-flat contact scheme (Fig. 3). The samples tested were cut from several actual wafer chucks (Fig. 4) 250 µm (a) (b) (c) Fig. 1. Typical fretting damage. (a) An excessive amount of fretting wear on a helicopter rear seat attach fitting. (b) Cracks appeared as a result of fretting fatigue in aircraft turbine blade dovetail joint. (c) Fatigue crack emanating from fretted zone. Nominally motionless contacts sub- jected to cyclic load often originating from vibration or temperature chan- ges may be damaged due to fretting motion, which is defined as a rela- tive cyclic motion between two surfa- ces, having a non-uniform distribu- tion of local relative displacement at their contact. Depending on loading conditions, material properties and environment, fretting may cause fret- ting wear or fretting fatigue (Figs. 1 and 2), which can arise in any as- sembly of engineering components. Fig. 2. (a) Schematic of radial fretting de- veloping due to cyclic normal loading alo- ne. (b) Contact area with regions of stick and slip. (c), (d) Fra- gments of radial fret- ting wear scar at dif- ferent magnifications. (a) (b) (c) (d) F D 100 µm 10 µm We acknowledge the support of the Israeli Ministry of Economy (METRO 450 Consortium within the frame of MAGNET program). We thank Michael Gissin (Jordan Valley), Yoram Uziel (Applied Materials) and Ronen Peretz (Intel) for providing chuck samples for tests. X-Y table Microscope stand Microscope Camera (a) (b) (c) Normal loading lever Tangential loading lever Piezo drive Weight holder Self-aligning sample holder Sample Transparent sample holder Transparent sample Fig. 3. (a) Schematic of experimental test rig. (b) Sample holders in assembly. (c) Sample holders and samples in section view . (a) (b) (c) (e) (f) (d) Fig. 4. (a) Self-aligning sample holder. (b) Sample cut from (c) electrostatic chuck (Applied Materials). (d) Vacuum chuck (Jordan Valley) and (e) sample cut from it. (f) New vacuum nipple (JV). Given the presence of relative motion, the forces exerted on the surface act through cer- tain distances so mechanical work is done on the surface. The amount of energy dissipa- ted in contact due to this work actually determines the form and severity of surface damage. This allows us defining a new we- ar criterion based on specific energy and obtained by integating friction stress over sliding distance (Fig. 7). Using this criterion we have studied the effects of ma- terial and surface geometry on chuck wear (Fig. 8). Fig. 8. Maximum work of friction forces (energy dissipated in contact) for two pairs of materials and different geometries. (a) Nipple by Jordan Valley. (b) Support projection by Applied Materials. μ=0.33 μ=1.5 -60 kPa 0 0.002 0.004 0.006 0.008 0.01 0.0 0.2 0.4 0.6 0.8 1.0 Max friction work, mJ/m 2 Pin radius, m PVS Al2O3 10 8 6 4 2 0 Si -90 kPa μ=0.33 μ=1.5 SiO2 PVS 30 20 15 10 5 0 25 Max friction work, mJ/m Max friction work, mJ/m Max friction work, mJ/m Max friction work, mJ/m 2 2 2 2 Fillet radius, m Fillet radius, m Fillet radius, m Fillet radius, m 0.25 0.20 0.15 0.10 0.05 0 (a) (b) Fig. 7. Spatial evaluation of energy dissipated in contact due to the work of friction forces in (a) nipple by Jordan Valley and (b) support projection by Applied Materials. Friction stress Sliding Friction work 6 MPa MPa MPa MPa 0 20 nm nm nm nm 0 60 mJ/m mJ/m mJ/m mJ/m 2 2 2 2 0 3.082 3.083 3.090 3.161 Radial coordinate, mm Radial coordinate, mm Radial coordinate, mm Radial coordinate, mm 40 20 5 10 15 2 4 Friction stress Sliding Friction work 1.0 MPa MPa MPa MPa 0 8 nm nm nm nm 0 4 mJ/m mJ/m mJ/m mJ/m 2 2 2 2 0 0 0.1 0.2 0.3 Radial coordinate, mm Radial coordinate, mm Radial coordinate, mm Radial coordinate, mm 3 2 2 4 6 0.5 1 (a) (b) 20,000 Cycles 200 µm 0 Cycles 40,000 Cycles 60,000 Cycles 0 Cycles 60,000 Cycles 75,000 Cycles 0 Cycles (a) (b) (c) (d) (e) (f) (g) (h) Fig. 6. Surface damage of the samples made from actual chucks and repeatedly loaded in normal direction only in laboratory experiments. (a)-(d) Fragment of the contact area of the ceramic support projection (Applied Materials) loaded with 7 N and seen through the transparent sapphire counter-surface. Dark area visualizes the wear particles accumulated at the interface between the sample and sapphire window. (e), (f) Fragment of the contact area of the Applied Materials support projection before and after the test demonstrating wear particles accumulated on the projection surface. (g), (h) The tip of the polymer support projection (Jordan Valley) before and after the test performed under the load of 5 N demonstrating plastic deformation of the contact area. (a) (b) (c) (d) (e) (f) (g) (h) (i) (j) (k) (m) Fig. 5. Surface topography of (a)-(c) new and (d)-(i) worked nipples of Jordan Valley (from chuck in Fig. 4d). (j) Worked chuck of KLA-Tencor and (k), (m) profiles of its vacuum groves. Surface topog- raphy of new and worked (in real machines) chucks was examined. The measurements show (Fig. 5) wear depths of several micro- meters on ed- ges of working surfaces. The tests perfor- med in the lab show the evo- lution of surfa- ce damage and accumula- tion of wear debris on wa- fer supports (Fig. 6) work- ing as a pollu- tion source.