Presented at the COMSOL Conference 2008 Hannover

Size: px

Start display at page:

Download "Presented at the COMSOL Conference 2008 Hannover"

Vincent Moody
5 years ago
Views:

1 Presented at the COMSOL Conference 2008 Hannover O. Sardan *,1, D. H. Petersen 1, O. Sigmund 1 and P. Bøggild 1 1 DTU Nanotech, Denmark 2 DTU Mechanical Engineering, Denmark * Corresponding author: DTU Nanotech, Ørsteds Plads, DTU Building 345E, DK-2800, Kgs. Lyngby, Denmark, oezlem.sardan@nanotech.dtu.dk

2 Introduction Electrothermal Microgrippers Nanomanipulation & Assembly Motivation The Microgripper Interface The COMSOL Model Geometry Settings Subdomain Settings Boundary Settings Mesh Settings Solver Settings Results & Discussion 2 DTU Nanotech, Technical University of Denmark

Conventional Three-Beam Design Compact &

3 Conventional Three-Beam Design Compact & flexible design, easy to operate Not strong enough Topology Optimized Design Flexible design, easy to operate Much higher force at the same size! 20 μm 20 μm 3 DTU Nanotech, Technical University of Denmark

4 Same for both three-beam and topology optimized designs Based on resistive heating Open & close COLD CLOSING THE GAP HOT V HOT V COLD 4 DTU Nanotech, Technical University of Denmark

5 Basic nanomanipulation of carbon nanotubes/nanowires On TEM grids for structrual analysis On electrodes for electrical characterization Assembly of nano-devices CNT/NanoBIT-enhanced AFM super-tips Collaboration with V.Eichhorn 20 µm 20 µm 20 µm 5 DTU Nanotech, Technical University of Denmark

It is important to estimate The influence of design variations During image processing after topology optimization During various microfabrication steps The operation temperature and the temperature

6 It is important to estimate The influence of design variations During image processing after topology optimization During various microfabrication steps The operation temperature and the temperature at the end-effectors The effect of the microgripper interface Grey-scale output from the optimization algorithm Improved binary image Both the microgripper & the CNT are damaged 20 µm 6 DTU Nanotech, Technical University of Denmark

temperature at the end-effectors The effect of the microgripper interface 2 mm Si chip PCB Leads

7 It is important to estimate The influence of design variations During image processing after topology optimization During various microfabrication steps The operation temperature and the temperature at the end-effectors The effect of the microgripper interface 2 mm Si chip PCB Leads Contacts Underlying oxide layer 7 mm Actuators μm 7 DTU Nanotech, Technical University of Denmark

8 3D Model Three application modes Conductive Media (dc) Heat Transfer by Conduction (ht) Solid, Stress Strain (smsld) Weak constraints on! 8 DTU Nanotech, Technical University of Denmark

9 2D CAD drawing from a DXF file 3D model by extruding the geometry along the z-axis Contact leads 1 μm-thick oxide underneath Contacts Actuator y x Underlying oxide y z x 9 DTU Nanotech, Technical University of Denmark

10 Silicon dioxide & polysilicon from the MEMS materials library For polysilicon, α = α(t), σ = σ(t) and k = k(t) (Geisberger 2003) For all subdomains Resistive heating Thermal expansion Polysilicon Oxide y z x 10 DTU Nanotech, Technical University of Denmark

11 Electrical domain Thermal domain Fixed temperature underneath the oxide Air convection, with h= h(t) (Geisberger2003) Mechanical Domain Air T 0 =300 K, h=h(t) z T=300 K y x 11 DTU Nanotech, Technical University of Denmark

12 Variable mesh density 1. Meshing the critical edges 2. Meshing the upper boundaries 3. Sweeping through all subdomains (multiple layers) Finer mesh y x y z x 12 DTU Nanotech, Technical University of Denmark

13 Simulation of the actuation behavior Parametric solver with default settings 13 DTU Nanotech, Technical University of Denmark

14 Actuation under an optical microscope at room temperature Simultaneous filtering and image processing Open Close 14 DTU Nanotech, Technical University of Denmark

15 Temperature distribution within the actuator Gap fully closed 1 µm-displacement, V 0 = 5.6 V Ongoing Raman measurements agree with the profile! y x [ o C] 15 DTU Nanotech, Technical University of Denmark

16 Settings: Prescribed displacement at the end-effector boundary Solving only for the mechanical domain Integrating the force over the boundary y=-1 µm y z x 16 DTU Nanotech, Technical University of Denmark

17 Results: Conventional design: k s,y = 2.3 N/m (Agrees well with force measurements ) Optimized design: k s,y = 234 N/m ~10 2 times improvement at the same size! k s,y (TopOpt)= 234 N/m k s,y (3-Beam)= 2.3 N/m 200 μm 17 DTU Nanotech, Technical University of Denmark

18 EU Projects: NANOHAND NANORAC COMSOL A/S (Denmark): Lars Gregersen Thure Ralfs NANORAC 18 DTU Nanotech, Technical University of Denmark

19 Temperature, T Thermal Conductivity, k CTE, α [K] [W/(m K)] [K -1 ] Electrical Conductivity, σ = 1/ρ = β 1+ β 2 T β3 Coefficient Value β β β DTU Nanotech, Technical University of Denmark

20 Temperature Heat Transfer Coefficient, h [K] [W/(m 2 K)] DTU Nanotech, Technical University of Denmark

Presented at the COMSOL Conference 2010 Paris Multiphysics Simulation of REMS hot-film Anemometer Under Typical Martian Atmosphere Conditions

Presented at the COMSOL Conference 2010 Paris Multiphysics Simulation of REMS hot-film Anemometer Under Typical Martian Atmosphere Conditions author: Lukasz Kowalski Universitat Politècnica de Catalunya