Hybrid Wired-Wireless Silicon Carbide-based Optical Sensor for Extreme Sensing Environment Temperature

Transcription

1 Hybrid Wired-Wireless Silicon Carbide-based Optical Sensor for Extreme Sensing Environment Temperature Mumtaz Sheikh (Graduate Research Assistant) Advisor: Prof. Nabeel A. Riza Photonic Information Processing Systems Lab. College of Optics/CREOL, Univ. Central Florida (UCF) Sponsor: US Department of Energy (DOE) Partners: Nuonics, Inc. and Siemens Power Generation, Orlando, FL

2 THE ULTIMATE CHALENGE IN FOSSIL FUEL-BASED POWER GENERATION Dominant Fossil Fuels: Coal and Methane (Natural Gas) Achieve The Following Basic System Attributes: - Clean Energy ( Reduced Green House Gas Emissions) - Improved Energy Conversion Efficiencies More Energy for Less A Solution * Use Extreme Temperature and Pressure Power Plants * J. H. Ausubel, Big Green Energy Machines, The Industrial Physicist, AIP, pp.0-4, Oct./Nov., 004.

3 Recent New Study on Power Plant Operating Temperatures and Pressures,700 ZEPP: Zero Emission Power Plant * J. H. Ausubel, Big Green Energy Machines, The Industrial Physicist, AIP, pp.0-4, Oct./Nov., 004.

4 COAL FOSSIL FUEL TO ELECTRIC POWER PATH CHINA 008 : A NEW COAL POWER PLANT EVERY WEEKS Ref: Call for Undergraduate Research in Energy & Sustainability Summer Internships for 008, Department of Chemical & Biological Engineering, Illinois Institute of Technology, Prof. Donald J. Chmielewski

5 GE s H system Gas Turbine System Test had 3500 gauges and sensors Uses Firing Temperature of 430 C R. Matta, et.al, Power Systems for the st Century, GE Power Systems Publication GER3935B, Oct. 000.

6 Siemens Advanced SGT6-6000G Gas Turbine Produces 58% combined cycle efficiencies with near 50 MW output at 60 Hz. Combustion Section Temperatures ~ 500 ºC

7 THE NEED: PRECISION RELIABLE SENSORS for GAS Temperature and Pressure Measurement to Keep Plants at Operating at Super Critical Conditions Any Options? -- ALL Electronic Sensor Chips? Fail after ~ 500 C (Insulation Breakdown) -- Silica/Glass Fiber Optics (Single Mode) Fail after ~ 900 C (Grating Erasure) What does Power Generation Industry Use Today? Custom Thermo-Couple Probes using Platinum/Rhodium Elements and Magnesia (MgO) or Alumina Insulating Ceramics 900 C BIG RELIABILITY/LIFE TIME PROBLEM Develops Cracks in the Insulation Picks Moisture- Becomes Conductive and Explodes

8 RELIABLE SOLUTION: THICK SINGLE CRYSTAL SILICON CARBIDE Melting Temperature ~ 500 C Resistant to Chemical Attack (Acids, Hot Gases) Mechanically Robust Excellent Atomic Scale Flatness Minimal Optical Wavefront Spoiling High.55 Refractive Index Strong 0% SiC/Air Reflectivity At EYE SAFE Infrared Band (550 nm) Single Crystal SiC Chip -- Handles GPa (0,000 atm) Yield Stress -- Allows Elastic Deformation in the Small Deflection Regime Due to Pressure Natural Interferometer Chip * Refractive Index n Changes With Temperature T & dn/dt (Thermo-Optic Coeff.) Changes Quadratically with T N. A. Riza, et.al, AIP Journal of Applied Physics, Vol.98, 035, 005. PERFECT FRONTEND OPTICAL CHIP FOR ROBUST WIRELESS SENSING

9 PROPOSED HYBRID WIRED-WIRELESS TEMPERATURE SENSOR USING THICK SiC SiC CHIP Tunable Laser Detector LENS M 5 Selection & Alignment Mirrors M WIRED (OPTICAL FIBER) WIRELESS SiC Chips Signal Power Meter and processor Output Signal Detector M M 3 P m = K R FP K Beam Spoiling Alignment Correction Mirror Sensor Selection Mirrors For SiC, can make the -Beam Interference Approximation: Where φ is the Optical path Length (OPL) and φ = Fresnel Power Coefficient= R =R =0.9 for SiC at 550 nm. See N. A. Riza, et.al., IEEE Sensors Journal, June 006. [ R + ( R ) R + ( R ) R R cos ]. φ M 4 4π n( T ) d( T ) HOT Remote Location

10 First SiC Wireless Temperature to 000 C Detected Normalized cos φ (T) Detected Normalized cos φ (T) Data at 547 nm SiC Chip Temperature ( 0 C) Data at 530 nm SiC Chip Temperature ( 0 C) Computer Normalized Cos{φ(t)} Measurements Note: Ambiguous Cos(OPL) Values Versus Temperature

11 Sensor Unwrapped Phase Change Data Δφ (T) in Radians versus SiC Chip Temperature with data taken at 530 nm & 547 nm. Weak quadratic curve fit is achieved. 50 Data at 547 nm Data Curve Fit Sensor Unwrapped Phase Data Δφ(T) (Radian) Δφ (T) = 6.4E-05*T *T SiC Chip Temperature ( 0 C) 50 Data at 530 nm Data Curve Fit Sensor Unwrapped Phase Data Δφ(T) (Radian) Δφ (T) = 6.0E-05*T *T SiC Chip Temperature ( 0 C) φ(t) phase ambiguity is removed by taking into consideration the number of cycles recorded and adding π radians after each cycle to get an unwrapped phase change value across the designed temperature range of room to 000 C.

12 Unwrapped Phase Difference φ (t) Condition for Unambiguous Temperature Measurement φ ( π Radian) Π Radians SiC Chip Temperature ( 0 C) The curve fits to the Δφ and Δφ data are used to calculate φ = Δφ(Τ) Δφ(Τ) and is plotted above ambiguous cycles at 547 nm and 37.5 ambiguous cycles at 530 nm -- Unambiguous temperature measurement via Instantaneous Wrapped Phase Processing using a 3-D Sensor Calibration Table

13 3-D Representation of The Sensor Calibration Chart For The Unambiguous Temperature Measurement of Temperature From Room Conditions To 000 C 4 Δ φ (T) (Radian) SiC Chip temperature ( 0 C) φ (T) (Radian) 3 4 Shown is a helical spring like behavior where for every set of values of Δφ and φ, there is only one value of the temperature corresponding to that data set

14 3-D Calibration Chart Room Temperature to 000 C 4 3 Δ φ (Radian) Chip Temperature ( 0 C) φ (Radian) C Temperature Sensitivity; 30 % Sensor Optical Power Efficiency

15 Experimental Set-Up to Measure Temporal Response of SiC Chip to Conductive Heat Step Function Method: Computer Controlled Digital Step mm diameter Pointed 90 deg-c Heating Element

16 Experimental Set-Up

17 Monitoring Fringe Motion Caused by Conductive Tip Thermal Step Function Captured frames at /30 s intervals Red Cross Tracks a Reference Fringe Position Fringe Motion Occurs each Frame Indicating SiC Response under /30 s

18 Fringe Movement with Time Elapsed after Thermal Step Function Applied to SiC Chip ~.5 Fringe Cycle Shift Over 65 deg-c Temperature Change

19 Temperature Response of the Heated SiC chip to a Thermal Step Function Chip Size; cm x cm X 400 micron Thick Conductive Heat Ramp Response ~ 5 seconds Steady State 90 deg-c Temperature ~ 0 seconds Convective Radiation Cooling ~ 7 seconds

20 Designed and Assembled Laser Beam Smart Transceiver Module for Sensor Motion Mechanics Fiber lens

21 IR Laser Beam Scanning Test Over ± 4.5 degrees Camera Shots at SiC Chip Plane at 60 cm from Fiber Lens

22 Nuonics Fabricated Frontend Probe using a Sintered SiC Tube and Embedded SiC chip SiC Probe

23 SiC Chip Interferograms Before and After Packaging at 633 nm Before After

24 Full Positive and Full Negative Interference Test with Assembled Probe using IR wavelength nm (Room Temperature) Bright State Dark State Shown SiC Chip Retro-reflected IR Camera Images 0: on/off ratio

25 Processed Initial Probe Data over 000 C Temperature Range Point Photo-Detector Data Till ~ 450 ºC IR Camera-based Data from ~ 450 ºC to 000 ºC Requires Active Alignment for Smooth Data Acquisition

26 SiC Probe Optical Response Video With Changing Temperature between deg-c in Oven

27 Two-Wavelength Technique Limitations - Limited design temperature range - Complex signal processing required to determine temperature - Prior calibration required at the design wavelengths Solution A technique that can measure temperature directly and unambiguously

28 Wavelength Tuned Signal Processing ϕ ϕ cos cos R R R R R R R R R FP = ) ( ), ( 4 T t T n π ϕ = k T t T n T t T n π π π ) ( ), ( 4 ) ( ), ( 4 = Δ + = k n n t = Δ + Δ n n n Classic Fabry-Perot Etalon Reflectance Difference between maximas or minimas k is an integer + Δ = Δ N. A. Riza and M. Sheikh, Silicon-carbide-based extreme environment temperature sensor using wavelength-tuned signal processing, Opt. Lett. Vol. 33, No. 0, pp. 9-3, May 008.

29 Wavelength Tuned Signal Processing ( ) C B A n + = ( ) + = Δ ) ( ) ( ) ( T n C T n BC T t k ( ) Δ Δ C n BC n Sellmeier Equation for Silicon Carbide Wavelength separation between maximas or minimas Refractive index, n (T) calculation is explained later

30 Wavelength Tuned Signal Processing Theoretical Experimental Wavelength range: nm SiC chip thickness, t (5 o C): 389 μm No. of cycles, k : 0 Total wavelength change from r.t. to 000 o C: 0.3 nm

31 Refractive index n (T) calculation Refractive index as a function of temperature n ( T ) = Δφ ( ΔT ) t( T ) 4π + n ( Ti ) t( Ti ) t( T ) Δφ is the unwrapped phase difference T i is the initial temperature Raw power at = nm

32 Refractive index n (T) calculation Cosine phase Unwrapped phase Δφ cos( φ) = P m 0.5 ( P P max max + Pmin ) P min P m is the raw optical power

33 Refractive index n (T) calculation Refractive index as a function of temperature, n (T) Varies r.t. to.68 at 000 o C

34 CONCLUSIONS - New Signal Processing Technique Developed and Experimentally Proven - Initial Optical Tests of SiC Chip Temporal Response and Alignment Achieved - Initial Nuonics Probe and Basic Sensor Operations Test to 000 C Achieved FUTURE WORK -Completed First Year of 3-year Research Program -Continue Advances in All-SiC Sensor Design Engineering and Testing Operations

35 Publications -OFS 7, SPIE Proc. Vol. 5855, pp , Bruges, Belgium, May IEEE Sensors Journal, Vol.6, June AIP Journal of Applied Physics, Vol.98, 035, OFS 8, OSA Conf. Proc., Oct IEEE PTL Journal, Dec SPIE Optical Engineering Journal, Jan IEEE Sensors International Conference Proc., Oct Optics Letters Journal, May 008.