Active Stabilization of Multi-THz Bandwidth Chirp Lasers for Precision Metrology

Transcription

1 Active Stabilization of Multi-THz Bandwidth Chirp Lasers for Precision Metrology Zeb Barber, Christoffer Renner, Steven Crouch MSU Spectrum Lab, Bozeman MT, Randy Reibel, Peter Roos, Nathan Greenfield, Trenton Berg, Brant Kaylor Bridger Photonics Inc. Bozeman MT, 59715

2 Relative Power (db) Relative Power (db) Relative Power (db) Ranging Terminology Two Targets Closely Spaced Range Range Window Targets Resolved Relative Range (mm) Range (m) SNR Single Target Resolution = c/2b -3dB Relative Range (mm) Range Resolution c p c R 2 2B 15 GHz 1 cm Range Precision Range resolution depends only on bandwidth - fair way to compare ranging systems Range precision is dependent upon signal-to-noise ratio (i.e. depends upon Tx powers, receiver size, etc.) Range window depends on system design (pulse rate, backend processing, coherence, etc.) R R max R SNR Range Window c 2 rep

3 Laser Ranging Methods for Length Metrology Coding Incoherent\Direct Coherent Pulsed Sinusoidal Chirped Traditional Pulsed Lidar\Ladar Absolute Distance Meters Chirped AM Ladar (ARL) Fast Detectors, Low Doppler Sensitivity Range-Doppler Lidar MSTAR, Multi- Wavelength Interferometry FMCW Ladar, Swept Source OCT High Doppler Sensitivity, Low Detector Bandwidth Simple, resolution limited by receiver bandwidth Range Ambiguity B R R 1/ rep Range-Doppler Ambiguity max Sinusoidal ranging generally assumes single strong target - Requires cooperative target for accuracy Coherent techniques allow for mixing in optical domain - Reduce receiver bandwidth requirements for high bandwidth waveforms

4 Optical Frequency Optical Frequency Relative Power (db) Optical Frequency f FMCW (Chirped) Ladar FM-CW Chirped Heterodyne Receiver Local Stretched Processor Oscillator df = dt must be within detector f and ADC bandwidth f beat beat = κτ D df dt = 2 R / c τ D Chirped Laser D D B = Chirp Bandwidth Time τ c Splitter Local c Time Tx LO Rx Delayed Signal FFTB Circulator Combiner Delayed Signal B t FM-CW Chirped Heterodyne Receiver Stretched Processor Distant Monostatic Object Tx / Rx df = dt must be within detector and ADC bandwidth df Detector + Digitizer dt = 2 R / c B = Chirp Bandwidth Time FW3dB R c c 2B 2 B Range (km) Fast Fourier Transform F

5 Chirp Nonlinearities f f beat = κτ D Local Oscillator Delayed Signal B Chirp nonlinearities lead to broadening of range return Broadening gets worse with longer delay τ D Chirp Laser τ c t Reference Interferometer Digitization and Post Processing or Sample Triggering [1] E. D. Moore and R. R. McLeod, Optics Express 16, (28). [2] T.-J. Ahn, J. Y. Lee, and D. Y. Kim, Applied Optics 44, (25).

6 Active Linearization Use stable reference delay, τ ref, to generate reference beat note Measure f err on photodetector (proportional to chirp rate) Phase/frequency lock to stable reference oscillator Actuate laser frequency/phase Chirped Laser Output f a o m Fiber Delay Acousto-optic Modulator r e f τ ref f e r r 1 nm Sweeps f r e f Servo Amplifier Digital Phase Detector C. Greiner, B. Boggs, T. Wang, and T. Mossberg, Optics Letters 23, (1998).

7 Frequency [khz] Error From Linearity (MHz) Frequency (MHz) Relative Power (db) Active Linearization Sweep Bandwidth = 4.8 THz Out-of-Loop Residual Sweep Error Unlocked = MHz Locked =.695 MHz Feed Forward =.17 MHz -2 Range Peaks through 5m PM Fiber FW3dB=5mm No Lock Lock Only Feed Forward Time (ms) Relative Range (mm) Improved servo reduced in-loop residuals to 46 khz RMS In Loop Residual Nonlinearity Time [s]

8 PSD db Active Stabilization Benefits Reduced computational load - $1 of analog electronic parts perform calculations in real time Increased coherence Improved SNR Can still use digital compensation Fiber Stablized DFB 5 khz RBW BiasTee Mod Port Slow Lock Frequency [MHz] Stabilized chirped sources Macro ECDL s THz BW, good coherence, slow Micro ECDL s THz BW, ok coherence, med speed DFB lasers 1 GHz, poorer coherence, fast

9 Relative Power (db) Long Range Measurements 1 mm 2 Monostatic Tx / Rx ~9 m Range Bridger Facility 2.5 Retroreflector 25 mm Zoom of Range Peak at 2.2 m - File Number 1 11 micron peak width Range: ~9 m Retro Location Relative Range (mm)

10 Relative Distance (mm) Relative Power (db) Long Range Measurements x 1 Eight targets at 14.2 km standoff Relative Range (m) Precision =.7 mm Time (s)

11 Range Arrival (um) Measured - Actual Mean Position (mm) Relative Power (db) Range [mm] Length Metrology FWHM = 63 microns FW3dB = R = 63 microns Range Profile off Microscope Slide ~7 nm precisions in 2.8 m air (.25 ppm fractional) ~43 nm thickness precision Relative Range (mm) Position Detection - 1 mm steps Measurement Mean Thickness = m STD Across Runs = nm ~2.8 nm precisions Measurement Number Measurement Number

12 Frequency [GHz] Frequency -f(r) [GHz] Chirp Rate - Mean [MHz/ ms] Transmission [Arb] Chirp Calibration.8 HCN Transmission Individual calibrations good to ~2.2 ppm Standard Deviation of 47 calibrations.17 ppm Calibration limited by stated NIST uncertainties Chirp Calibration with HCN = MHz/ms Time [sec] 3 x Measured Chirp Rate Time [ms] Sweep Number

13 MHz MHz THz Chirp vs. Comb Calibrations Calibration performed at NIST with Nate Newbury, Ian Coddington, and Fabrizio Giorgetta Heterodyne signal between comb and chirp nm ~5 THz/sec Linear Fit High order residuals less than 55 khz rms Statistical uncertainty in chirp rate is ~2 ppb Quadratic and Cubic Fit < 1 ppb stability sweep-to-sweep Z. W. Barber et al. Optics Letters, vol. 36, no. 7, pp , Unwrap glitches Time [sec]

14 Y Coordinate (mm) Optical Thickness (mm) Y Coordinate (mm) Optical Thickness (mm) Thickness Measurement Circulator Tx / Rx Splitter 9% 1% Rx 9% 1% Combiner Optical Probe Det. + ADC Sample 2D Linear Stage Sample Wafer Thickness Wedge Removed Mean Optical Thickness = mm Residual Optical Thickness Best Fit Planar Surface Removed mx = my = b = X Coordinate (mm) X Coordinate (mm) -2

15 Y Coordinte (cm) microns Z Coordinate (um) microns Optical Metrology Complex machined optical surfaces 3 Specular reflections and large off normal surface angles require high NA objectives Scaled and Offset Original data Multiple surfaces can be measured simultaneously mm 2 Figures generated by MSU-Spectrum Lab for Wavesource Inc. using Bridger Photonics SLM-M laser system Surface Profile - Sample #3 Scaled and Offset Original Data 2 4 mm mm X Coordinate (cm) 2 4 mm Figures generated by MSU-Spectrum Lab for Wavesource Inc. using Bridger Photonics Y Coordinate (cm) X Coordinate (cm).8

16 Large Volume Metrology Laser Trackers Require cooperative target Crossrange resolution and precision insufficient Proposed FWCW ladar solution Synthetic Aperture Ladar (SAL) to improve crossrange resolution Trilateration to improve crossrange resolution Multistatic SAL Trilateration 3D SAL Image Receiver A c D SA B b Transmitter a C x,y,z 5µm Resolution <1µm Precision

17 Synthetic Aperture Ladar Chirp Laser 9/1 circulator 5/5 target stage SL ADC and Post- Processing Paper Cutouts On Black Background Phase Gradient Autofocus (PGA) used to phase paper images 3M Diamond Reflective Sheeting S. M. Beck et al., Synthetic-aperture imaging laser radar: laboratory demonstration and signal processing, Applied Optics, vol. 44, pp , 25.

18 Conclusions Active linearization of ultra-broadband (> THz) laser sweeps High resolution and precision unambiguous ranging Current accuracy 1-8 of range, 1-9 possible Short and long range metrology applications High resolution Synthetic Aperture Imaging Acknowledgements: Bridger Photonics; Stephen Dunn Wavesource Inc. for optical samples; Esther Baumann, William Swann, Randy Babbitt, Jason Dahl Financial Support: Montana Board of Research and Commercialization Technology NSF GOALI\MCME # NSF SBIR