COHERENCE SCANNING INTERFEROMETRY

Transcription

1 COHERENCE SCANNING INTERFEROMETRY Part 2. Quantification of errors Suresh K. Ramasamy PhD April 2013 OUTLINE Z based errors XY based errors Part based errors Summary Table Sometimes I lie awake at night and ask, Where have I gone wrong? then a voice says to me, This is going to make more then one night Charles M. Schulz 1

2 SUMMARY TABLE ACKNOWLEDGEMENTS Dr. Jayaraman Raja Dr. Brian Boudreau Sample support: Chris Wichern (Nanofocus) Laura Shaw (University Erlangen Nuremberg) Rose Peet (RPC Photonics) Tatsuya Sugihara (Osaka University) CSI support: Bala Muralikrishnan (NIST) Claudiu Guisca (NPL) Don Cohen (Michigan Metrology) Jimmie Miller (UNCC) Manas Lakshmipathy (Zygo) Peter de Groot (Zygo) id: 2

3 QUESTIONS? "Things should be made as simple as possible, but no simpler." Albert Einstein Z BASED ERRORS Sequential height repeatability Scan length (for systems with no reference signal) Z stage non linearity NA correction Algorithm peak and phase detection, noise reduction / interpolation Focus location Environment changes 3

4 XY BASED ERRORS Camera non linear warping error Sequential repeatability for dimensional measurement Stage reproducibility/ non linearity Stitch errors Objective and turret repositioning errors PART BASED ERRORS Dissimilar materials Shape of part AW of parts ITF 4

5 Z SEQUENTIAL MEASUREMENT REPEATABILITY Reference Flat Standard 10 sequential measurements Average individual data Average of RMS 0.33 nm SCAN LENGTH Step Height Standard 5 scan lengths All magnifications 100 µm scan is taken as basis Step Ht 1.816µm Step Ht µm Step Ht µm Ratio wrt 100µm scan Scan Obj 2.5x Obj 10x Obj. 50x Length Zoom 0.5x Zoom 1.0x Zoom 2.0x Zoom 0.5x Zoom 1.0x Zoom 2.0x Zoom 0.5x

6 Z STAGE NON LINEARITY Reference Flat Standard Titled along X axis to cover entire range (a) Average of 10 msmts. to reduce noise Z lowered by one fringe (b) Average of 10 msmts. RMS of a b Z raised by 2 fringes (c) Average of 10 msmts. RMS of a c Repeat with standard tilted in opposite direction in X axis Repeat the same along Y axis Average of 8 RMS NA CORRECTION Step Height Standard 20 msmts. taken 10x mag is taken as basis Average Step Height in µm Obj. 2.5x Obj. 10x Obj. 50x Run No. Zoom 1x Zoom 2x Zoom 0.5x Zoom 1x Zoom 2x Zoom 0.5x average std. dev NAC

7 ALGORITHM INDUCED 1/2 High2G Normal resolution mode mode ALGORITHM INDUCED 2/2 Deviation Measured Height between Ra in values µm algorithms in nm in nm Z Height in µm 90 Surface Edge Profile Transition of 150nm errors Random on Surface square under wave artifact different FDA Resolution Modes nm 40 nm height 60 nm 100 nm 150 nm 1000 nm Mode 1 Mode 2 Mode 3 Mode Distance 1 Mode in 2mm random squarewave sinusoid random Mode 3 Mode random 1 Mode sinusoid 2 Mode surface 30 nm random surface Mode 100 surface 1nm random Mode surface 2surface Mode surface nm random surface surface Mode 1 Mode 2 Mode 3 7

8 FOCUS VARIATION Step Height Standard first msmt. with bottom surface in focus Second msmt. at +25µm above Critical for cylindrical surface msmts. ENVIRONMENTAL IMPACTS Square wave Standard Combination of XY and Z in one test Height 1 sigma = 0.002µm, Width 1 sigma = 0.048µm Step Height in µm :58:25 18:28:52 18:59:20 19:29:47 20:00:14 20:30:42 21:01:09 21:31:36 22:02:04 22:32:31 23:02:58 23:33:26 0:03:53 0:34:21 1:04:48 1:35:15 2:05:43 2:36:10 3:06:38 3:37:05 4:07:33 4:38:00 5:08:28 5:38:55 6:09:22 Step Width in µm Time in hr:min:sec step ht step width 8

9 CAMERA NON LINEAR WARP XY SEQUENTIAL REPEATABILITY Standard Deviation in µm of 25 measurements Standard [0.25µm (dia.), 0.13µm (center)] Autofocus [0.25µm (dia.), 0.19µm (center)] Std. Dev in µm Dia. CX CY Dia. CX CY Dia. CX CY Dia. CX CY Dia. CX CY Dia. CX CY A B C D E F Without Autofocus With Autofocus 9

10 XY REPRODUCIBILITY Range Value over multiple 3runs Range in µm Runs STITCHING ERRORS ISO Flatness (nm) (Average difference 8 nm) ISO Flatness (nm) Run Delta 10

11 TURRET REPOSITIONING ERROR DISSIMILAR MATERIALS A B A 11