Computer simulations of a maximum length sequence modulated photoacoustic spectrometer

Transcription

1 Computer simulations of a maximum length sequence modulated photoacoustic spectrometer Arash Soleimani Karimabad Ralph T. Muehleisen Civil, Architectural, and Environmental Engineering asoleima@iit.edu muehleisen@iit.edu

2 Outline PAS Basics MLS Idea System Block Diagram Laser Model Acoustic Propagation Model Noise and Error Sources Conclusions

3 Basics of Photoacoustic Spectroscopy (PAS) Laser pulses generate sound pulses Gas absorbs light and heats up to make sound Sound pulses measured by microphone Lock-in amplifier rejects background noise Analysis used to get optical absorption coefficient

4 Idea The utility and benefits of using maximum length sequence (MLS) modulated lasers was discussed at the last ASA meeting Remove optical chopper and lock-in amplifier Measure multiple wavelengths at the same time Gas/Liquid/Solid measurements without a resonant cell Result : Multi-wavelength PAS that is cheaper, lower power, more rugged and more portable than current systems Goal: Develop a computer simulation of a new type of PAS system using MLS modulation of multiple low power laser diodes

5 System Block Diagram Multiple Low Power Laser Diodes MLS Modulation Absorbing Sample Pressure Waves (Photoacoustic Signals) Electret Microphone Electric Signals Absorption Spectrum

6 Diode Laser Rate Equations Reference: S. A. Javro and S. M. Kang, Transforming Tucker s linearized laser rate equations to a form that has a single solution regime, J. Lightwave Technol., vol. 13, pp , Sept

7 Simulink Laser Model Carrier Density Integrator 1.0 Constant 3 N_0 Gain 9 simin 1/(q*V_act) From Workspace 1.0 Constant 1 2 Out 2 N_e/tau _n Gain 10 Gain 1 s Integrator Scope 2 g_0 Gain 1 Current Input by MLS Signals 1/tau _n Gain 2 Product Gain 7 Gamma Gain 6 Gamma 1.0 Constant 2 Gain 5 Photon Density Integrator Beta 1 s Scope 1 epsilon Gain 3 (V_act*Eta*h*c)/(Gamma *tau_p*lambda ) Scope 14 Optical Laser Power Output Integrator 1 Scope Gain 4 1/tau _p 1 Out 1 Gain 8

8 Simulink Laser Model simin In1 Out1 1 Current Input by MLS Signals Optical Laser Power Output Laser Model System Scope

9 Simulink Laser Model

10 Laser Power [ W] Input Current [ma] 2aPA6: Arash Soleimani -Karimabad, MLS Photoacoustics Typical Laser Diode Simulation Overshoot and Ringing Laser Power output for 10 ma MLS input Laser Power MLS Signal t [ms]

11 Laser Power [ W] Laser Power [ W] 2aPA6: Arash Soleimani -Karimabad, MLS Photoacoustics MLS Laser Simulation Results Zoom of Pulse Rising Edge Zoom of Pulse Falling Edge Delay and ringing are very short compared to the acoustic time scale and are filtered by the acoustic wave temporal response so they have no affect the sound generation The modulated laser pulse acts essentially like a unit step function t [ns] t [ns]

12 Simulating Acoustic Propagation 1 p I L( r, t) p c t c t where I ( t) P( t) A and is the optical absorption coefficient L beam Since the system is linear, the pressure at a point in space p(t) can be written as a convolution of a source input convolved with the temporal response I L p( t) hl( t) ha( t) t where h ( t) is the response of an infinitely narrow laser L and h ( t) is the response of a perfect laser impulse ( t) a p

13 Acoustic Response Components L L Laser Beam Mic to measure p(t) d r 0 sin 0 For a Gaussian Beam of width a h () t a K e a r0 ct asin 0 2 For a narrow beam of length 2L and an absorption coefficient of L Ke 2 ( ) cosh L d 2 hl t c t 2 c 2 c t d c

14 Sorry Because of time lost in modeling the lasers, we have not yet implemented the propagation model into Simulink. We also have not quantified the sources of noise and error and have not added them to the model either

15 Identified Sources of Noise and Error Input Current Source Current noise, pulse ringing, and slew rate effects can be limited with proper laser drive circuitry Laser Spontaneous emission and absorption, shot noise can be a problem on cheap lasers Background Acoustic Noise This is the limiting noise source for most situations. A good estimate of the background noise level and spectrum is essential for estimating instrument sensitivity. Microphone/Amplifer Thermal noise, 1/f noise, shot noise are small compared to acoustic background noise Sampling and Digitization These are not insignificant, but can be small compared to acoustic noise

16 Conclusions MLS modulation of the input current of a typical laser diode can generate an optical output that is a near perfect MLS function The received acoustic signal can be estimated by consideration of acoustic wave generation as linear processes and the convolution theorem There are a multitude of noise sources in our system but the measurements will be limited mainly by ambient acoustic noise rather than instrumentation noise We will continue our work and hopefully present a complete model with analysis in Paris 08