Case study: electromagnetic radiation detection

Transcription

1 Case study: electromagnetic radiation detection many modes of detection rectification requires non-linear element (diode) usually restricted to low frequencies photon detection generation of electron/hole pairs usually restricted to short wavelengths thermal detection convert incoming E&M energy into heat can be fairly independent of frequency/wavelength Dean P. Neikirk 200, last update March 30, 200 Dept. of ECE, Univ. of Texas at Austin

2 Spectral range of common electromagnetic detectors Wavelength m 0 cm cm mm 00 µm 0 µm 3000 Å 300 Å radio microwave sub-millimeter millimeter far infrared IR visible Conventional Bolometer Schottky Diode Antenna-Coupled Microbolometer Photo Diode Liquid Nitrogen Cooled CCD Liquid Helium Cooled Photoconductors 00 MHz GHz 0 GHz 00 GHz THz 0 THz 00 THz 0 5 Hz Frequency 0 6 Hz Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

3 Basic thermal radiation detector: bolometer convert incident electromagnetic radiation to heat, heat capacity C, attached to a heat sink at temp. T s via thermal resistance R. thermal time constant t»c R At dc T bolo = T s + P bolo R temperature change resistance change voltage difference via constant applied bias current to maximize responsivity maximize thermal resistance R th between bolometer and heat sink Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

4 Output voltage and responsivity V = Ib dr dt T = I b dr dt P absorbed Z thermal R = P V incident = I b dr dt dt dp P P absorbed incident = I b Z thermal dr η dt R = I b Z thermal α R bolo η α = R bolo bolo dr dt η = coupling efficiency depends on bias current, temperature coefficient of resistance a, bolometer resistance R bolo, device thermal impedance Z thermal, and coupling efficiency h Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

5 Figure of merit: Noise Equivalent Power figure of merit is really the minimum detectable input power depends on both responsivity and noise for simple resistive bolometer the lowest noise level is Johnson noise noise voltage depends of bandwidth V johnson = 4k B T R BW units for NEP are normally volts/root hertz V NEP = noise R increased responsivity R tends to improve (decrease) NEP Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

6 Detector Size Relative to Wavelength critical in determining "coupling efficiency" much larger than wavelength "classical" absorber detector is its own "antenna" typical figure-of-merit: specific detectivity (D*) typical units cm (root Hz) / watt much smaller than wavelength: "micro-detectors" very poor coupling requires "antenna" structure typical figure-of-merit: Noise Equivalent Power (NEP) D* (effective area) 0.5 / NEP Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

7 Single versus Multi- Mode Antennas single mode: use for point sources one antenna, one detector absorbed power: P = kt effective area l 2 multimode: use for distributed sources n-element antenna "array," n detectors P = nkt effective area n l 2 NEP array = v n NEP single regardless, D* l 2 / NEP single Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

8 Optimum "Resistive" Absorbers: large detectors area >> l 2 "multimode" system impedance matched sheet -absorption must be VERY strong -requires both resistive sheet and perfect mirror 00 % absorption in resistive sheet at design wavelength mirror l/4 resistive sheet R S = 377?/square incident radiation reflectance = 0 Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

9 Optimum "Resistive" Loads: small detectors "detector" area << l 2 - behaves like classical "lumped" circuit element: resistor "absorption" requires an antenna efficiency depends on - antenna gain and beam pattern - detector/antenna impedance match Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

10 Quasi-Optical Detection System detector antenna l substrate substrate lens objective lens Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

11 Trade-offs: Small versus Large Thermal Detectors figure of merit improves as responsivity increases NEP = Noise Voltage / Responsivity responsivity r (Volts/Watt) r = I bias d R d T d T d P depends on: bias current "thermometer" sensitivity: resistance change / temperature rise thermal impedance: temperature rise / power in how do these quantities scale with size? Dean P. Neikirk 200, last update March 30, 200 Dept. of ECE, Univ. of Texas at Austin

12 Optimizing responsivity: alpha (temp. coefficient of resistance: TCR) highest TCR from superconductors at their transition temperature requires cooling system; may present impedance matching (E&M absorption) problem due to low sheet R alpha of several bolometer materials Y-Ba-Cu-O T c : ~ K - Vanadium Oxide (VO x ): ~ K - (room temp.) Semiconducting YBCO: ~ K - (room temp.) Ag: ~ K -, Ni: ~ K -, Au: ~ K - Bismuth (Bi): ~ K - (Room temp.) Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

13 Optimizing alpha: Superconducting Detector Strip temperature coefficient of resistance a: intrinsic semiconductors 50 Resistance dr/dt (AVE) High T C YBCO Strip 00 extrinsic semiconductors and metals: a /T transition edge materials YBCO VO2 Detector Resistance (ž) dr/dt (ž/k) Temperature (K) Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

14 Optimizing responsivity: thermal impedance Limiting mechanisms: heat flow from small bodies radiant: negligible ( W = st 4 W: total radiated power per unit area) conduction: dominant for micron size objects convection: can be significant for > tens of microns but negligible at low pressure Thermal conductivity of various material Silicon (bulk): ~.3 W/cm-K Si 3 N 4 : ~ W/cm-K Bi (thin film): ~0.08 W/cm-K SiO 2 : ~ 0.04 W/cm-K Bi (bulk): ~ W/cm-K Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

15 Thermal properties thermal diffusion equation 2 φ = ρ C κ φ t f is temperature change relative to ambient r is material density C is specific heat (units: energy mass - kelvin - ) k is thermal conductivity (units: power distance - kelvin - ) if fwas a voltage this looks a lot like an electrical problem or if fwas a concentration this looks a lot like a diffusion problem for constant thermal diffusivity D = k/rc φ = t 2 κ ρ C φ Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

16 -d sinusoidal heat flow let s assume the power flowing from the source end of the bar is sinusoidally varying in time power flow l w t y 2 ρ C φ { φ = κ t d 2 = φ z2 x z 2 φ z 2 = ρ C κ φ t T amb + f o φ 2 γ φ o j ω t γ z ( z) = φ e j ω t γ z = o e ρ C κ T amb j ω t γ z ( jω) φ e o γ= j ω ρ C κ φ end = P Z thermal = P t w j ω κ ρ C tanh j ω ρ C κ l Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

17 What does this really mean? consider low frequency : length of bar much less than thermal diffusion length φ end P = t w j ω κ ρ C γ l = L l thermal j ω ρ C = l << κ j ω ρ C P j ω ρ C tanh l l κ t w j ω κ ρ C κ << φ arg ument "low freq" end P l { t w κ power density looks just like the resistance of a bar with conductivity k Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

18 Low frequency thermal response T = l t w P κ the actual input power looks like P in = P ave [ + cos(wt) ] at low frequency the temperature follows (in phase) the input power temperature rises as power increases temperature falls as power decreases power P ave T max temperature time T ambient Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

19 High frequency thermal response consider high frequency : length of bar much longer than thermal diffusion length γ l = L l thermal = j ω ρ C κ l >> φ end = P t w j ω κ ρ C j ω ρ C P tanh l 42 4κ43 4 t w >> j ω κ ρ C L thermal κ φ "high freq" end P { t w power density L thermal κ = P t w 2ωκρ C ( j) Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

20 High frequency thermal response φ "high freq" end P { t w power density the temperature lags the power by 45 ( j) as the frequency increases the magnitude of the temperature change decreases as w L thermal κ P = t w 2ωκρ C power P dc T ave temperature time T ambient Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

21 Superconducting transition edge bolometer Richards, early 60 s few square millimeter black coated glass slide bismuth black used as broad-band absorber suspended by silk threads in a cooled vacuum chamber small superconducting resistor attached threads coated with thin metal to allow electrical access produced very high thermal resistance, moderate thermal capacitance how does this scale with size? need to look at thermal model and various heat transport mechanisms for small objects Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

22 Heat Conduction into Substrate approximate the detectorsubstrate contact area as a hemisphere use thermal spreading resistance R thermal scales as (area) -/2 cut-off frequency scales as /area for device on SiO 2 0µm x 0µm R thermal = 2.5 x 0 4 K/W cut-off f = 8.5 khz µm x µm R thermal = 2.5 x 0 5 K/W cut-off f = 850 khz for large devices clearly need to remove the substrate to increase R! φ = Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin w φ o R ro r e j ω t ( r r ) L T = thermal substrate f cut off substrate o thermal l a L thermal P 2 ro 4243 π κ thermal spreading resis tan ce 2π κ κ ρ C = substrate substrate substrate j area κ ω ρ C a l w 2 π area

23 Thermal losses: convection to air simple spherical model for convective cooling h c : surface coefficient of heat transfer units: W / area Kelvin κ h c = Nu d k : thermal conductivity d: diameter Nu: Nusselt number for free convection from spheres Nu is approximately Gr: Grashof number Pr: Prandtl number Nu = ( Gr Pr) 4 Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

24 Thermal losses: Convection Gr: Grashof number Gr = ρ 2 g β T L 2 µ 3 g is acceleration due to gravity, r is density, m is viscosity, bis the coef. of volume expansion, L is length of object for air near room temperature Gr 3.4 x0 7 L 3 ( L in microns) T ( in Kelvin) Pr: Prandtl number Pr = C µ κ 0.72 ( air near room temp) so as long as L < cm the Nusselt # is Nu = ( Gr Pr) 2 Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

25 Thermal losses: Convection then h c, surface coefficient of heat transfer units: W / area Kelvin h c κ 2 d thermal conduction from sphere is just (area h c ) G c 4π d 2 h c = 8π κ air d or R convect to air 8π κ air area looks just like thermal spreading resistance term! R thermal substrate 2π κ area Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin substrate

26 What about heat sinking through the ends? use a simple bar model assume absorbed power is uniformly dissipated throughout bolometer, sinusoidally varying in time ends of detector are connected to perfect heat sinks thermal diffusion equation is then 2 φ = 2 φ x width thickness 2 = ρ C κ bol bol bol bolometer Length 0 φ t x P input electrical lead: heat sink! = Po 2 Po 2 L t w κ ( j ω + e t ) bol ( j ω + e t ) Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

27 What about heat sinking through the ends? width thickness bolometer Length 0 x electrical lead: heat sink! since is symmetric, solve with x = 0 at center assume absorbed power is uniformly dissipated throughout bolometer, sinusoidally varying in time ends of detector are connected to perfect heat sinks φ ( x) = Po 2 L t w κ bol L 8 2 x 2 2 j ρ e ω cosh cosh [ γ x] [ ] γ L 2 j ω ρ C γ κ if TCR is constant, then really want integral averaged temperature over the whole length of the bolometer + Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin bol κ C bol bol jωt

28 Heat sinking through the ends width thickness bolometer Length 0 x electrical lead: heat sink! integral averaged temperature over the length φ = + L 2 P o jω γ L γ L φ( x) dx = + 24 e tanh t L 24 t w κ bol 2 L 2 3 ( γ L) 2 j ω ρ C γ κ ratio of time varying temp to time varying input power is the thermal impedance L Zthermal( ω) = 2 3 t w κ bol γ L ( γ L) 2 2 γ L tanh Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

29 Heat sinking through the ends bolometer width Length thickness 0 x low frequency: g L << φ 32 γ L<< 2 Po L 24 L t w κ { j ω + e t } high frequency: g L >> bol electrical lead: heat sink! Z thermal φ = P 32 γ L<< j ω ρ C γ κ L 2 t w κ bol φ 32 γ L>> Po 2 L t w κ bol 2 L 2 j ρ bol κ C bol bol ω e jωt Z thermal 32 γ L>> ρ t w L C bol j ω R same as parallel R-C circuit with thermal L 2 t w κ 2 κ = bol Cthermal = ρ t w L C ω bol cut off R thermctherm = 2 bol ρ C bol L Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

30 Small bolometers: heat conduction directly through device leads assume heat sink at ends of bolometer R C thermal = ρ t w L thermal C bol = L 2 t w κ bol = 2 σ R κ bolom electrical width thickness ωcut off bolometer Length 2 κ bol ρ C L bol 2 heat sink - thermal resistance directly proportional to electrical resistance, regardless of material! - for high thermal R want length to cross-sectional area ratio to be large - for high speed want L small! for silicon bar: 0µm x 0µm x 0.5µm R thermal = 0 3 K/W cut-off f =.6 MHz µm x µm x 0.5µm R thermal = 0 3 K/W? cut-off f = 60 MHz Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

31 Actual simulation results for full three-d geometry: Isothermal Profile into Substrate heater element antenna lead 0.0 Y-Direction from the substrate surface (microns) substrate T=.0 T=.02 T=.05 T=. T=.5 T=.3 T=.4 Bismuth Load Quartz Substrate Operating Conditions -Input power = 4 µw -Length = 4 µm -Width = 2 µm X - Direction from the center of the detector (microns) 2 3 Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

32 Fringe Heat Conduction into Leads Detector/heater Antenna Lead Z D T Detector Z D-sub Substrate Z sub heat flow into the antenna from the substrate = + Z net Z D Z sub + Z D sub thermal resistance and speed increase as bolometer shrinks Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

33 Thermal impedance summary convection to air above bolometer conduction to substrate substrate under bolometer air above bolometer conduction out arms of bolometer R thermal air 2π κ thermal arms L 2 area κ total from parallel combination (smallest R dominates) R thermal = thermal R i clearly R thermal gets bigger as device area gets smaller speed gets higher as volume of device shrinks Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin R thermal substrate 2π κ R air thermal convect R area 8π κ substrate = air area area bol

34 Micro bolometers originally this term was coined by Hwang and Schwarz, and referred to bolometers that were smaller than a wavelength T.-L. Hwang, S. E. Schwarz, and D. B. Rutledge, Microbolometers for infrared detection, Applied Physics Letters, vol. 34, pp. 773, 979. required antenna for coupling to increase thermal resistance need to decrease cross sectional area, and increase length, of thermal link can use micromachining to form "free standing" films to remove the substrate bulk or sacrificial layer processes can be used Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

35 Basic classes of antenna coupled microbolometers main classification set by dominant heat transport mechanism R thermal substrate 2π κ substrate area thermal arms L 2 area κ substrate supported : main heat loss to substrate under bolometer air-bridge : main heat loss out ends of bolometer R = bol R < thermal substrate R thermal arms R > thermal substrate R thermal arms < 2π 2 L area κ κ substrate bol 2π 2 L area κ κ substrate < bol Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

36 Air-bridge bolometers need 2π 2 L area κ κ substrate < bol easiest way to get this is to have k sub << k bol use micromachining to form "free standing" films to remove the substrate bulk or sacrificial layer processes can be used thermal resistance is then R thermal arms = L 2 area κ increase by decreasing cross sectional area, increasing length bol Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

37 Optimizing Responsivity R = P V incident = I b dr dt dt dp P P absorbed incident = I b Z thermal dr η dt clearly want max Z thermal, but improvement is not unbounded bias current limitations from I 2 R heating: I max2 R = P max = (T max -T amb )/Z thermal thermal r T max T ambient R bol α bol R bol current density (electromigration/critical currents) r J max α bol l R thermal bol r J max α bol σ bol 2 π σ bol κ sub l w substratedominated detector Instability (problem only if a is positive) r I bias < ( I bias ) 2 dr dt < thermal thermal r < R bol α bol R bol R bol alpha positive Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

38 Tri-layer sacrificial resist process use photoresist as sacrificial layer tri-layer resist, middle layer soluble, top and bottom patterned lift-off patterning of materials electrical lead evaporation Photoresist Air-bridge electrical lead evaporation bolometer evaporation bolometer evaporation electrical lead material bottom sacrificial layer Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

39 Air-bridge Lithographic Technique Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

40 Behavior in IR (~ - ~0 mm) antenna-coupled microbolometer requires sub-micron size for lumped model to apply conductor losses in antenna use "non-resonant" design: simple bow-tie achieve wavelength selectivity using back short (integral micromachined mirror) substrate absorption impacts efficiency micromachining removes substrate from antenna Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

41 Quasi-Optical Detection System detector antenna»l micromachined tuning mirror objective lens Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

42 Impedance Matching: Composite Microbolometers detector lead size and conductivity constraint: R bol 50-00? -cannot use very low or very high conductivity materials load separate function of load and temperature sensor -matched "heater" -high alpha "detector" detector antenna lead substrate Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

43 Composite Microbolometer Heater Signal Line Antenna Detector Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

44 Layout of Simulated Devices Gold Antenna Leads Z Y 2300 Å Evaporated Gold NiCr Heater 2.5 µm x 4 µm smallest region of symmetry X 500 Å Nichrome Heater 800 Å Sputtered SiO µm 2500 Å YBCO Y X YBCO detector signal line substrate Z Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

45 Temperature profiles for various substrates SiO substrate YBCO NiCr free space YBCO substrates: MgO, YSZ, LaAlO 3 with buffer layer: Si, sapphire -most have either high κ or high ε r Temperature above Local Ambient (K) Ksub = 0.04 w/cm/k Ksub = 0.05 w/cm/k Ksub = 0.5 w/cm/k Ksub = 0 w/cm/k SiO 2 SI or sapphire, 80 K Distance Normal to Substrate (microns) 0.5 P in = 4 µw.0 Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

46 YBCO Composite Microbolometer antenna YBCO films -grown on 400 Å MgO buffer -sapphire substrate -courtesy: Alex delozanne, UT-Austin YBCO microbolometer Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

47 Responsivity Measurement y = x bolometers are easily calibrated from dc I-V -slope of resistance versus power curve -plot V/I versus I*V for composite bolometer plot detector voltage versus power in heater e+0 2.0e-6 4.0e-6 6.0e-6 8.0e-6.0e-5 -YBCO / heater : 40 V/Watt Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

48 Noise Voltage T = T = 85.7 T = Vn/(Hz)^/ /root(f) 0-9 Johnson Noise Limit Modulation Frequency (Hz) Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

49 Noise Equivalent Power (NEP) YBCO composite NEP (watts/root Hz) /root(f) Johnson Noise Limit structure - 3 x 0 - W/vHz conventional Bi microbolometer W/vHz at room temperature Modulation Frequency (Hz) Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

50 Micro bolometers originally this term referred to bolometers that were smaller than a wavelength required antenna for coupling but even for large area devices, to increase thermal resistance, need to decrease cross sectional area, and increase length, of thermal link if too large will be slow use micromachining to form "free standing" films to remove the substrate bulk or sacrificial layer processes can be used Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

51 classical micromachined bolometers device area >> l 2 if use CMOS compatible processing can combine with other electronics J.-S. Shie and P. K. Weng, Fabrication of micro-bolometer on silicon substrate by anisotropic etching technique, presented at Transducers '9 99 International Conference on Solid- State Sensors and Actuators, San Francisco, CA, 99. R. A. Wood, High-Performance Infrared Thermal Imaging with Monolithic Silicon Focal Planes Operating at Room Temperature, presented at IEEE International Electron Devices Meeting, Washington, DC, 993. these devices are now widely referred to as microbolometers Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

52 Simple model of large microbolometer assume absorbing region absorbs power uniformly over its entire area bolometer, area A bol, thickness t bol C bol thermal t bol A bol ρ bol C bol heat sink(s) R arm thermal support arm(s), length L arm, width w arm, thickness t bol Larm t w κ arm arm arm Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

53 Equivalent thermal circuit bolometer is REALLY distributed but if assume: power uniformly deposited across bolometer thermal resistance of arm is much larger than that of bolometer bolometer area is approximately isothermal Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

54 Simplified thermal equivalent circuit surface of bolometer is approximately isothermal mainly contributes thermal capacitance arm thermal resistance is large heat sink R arm thermal C t bol thermal t bol ρ arm A bol L w bol C bol arm arm κ arm P in Z thermal = R R arm thermal arm thermal + + j ω C bol thermal arm bol ( j ω R thermal C thermal) arm bol ( ω R C ) 2 thermal thermal conclusions: make arm length-to-cross section ratio LARGE make thickness of bolometer small (area already set by wavelength) Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

55 Bolometer absorber material choices metals skin depth l o = 5 µm Free Carrier Absorption in Silicon impedance matching requires very thin sheets tens of Å few ohms/square semiconductors below-gap absorption weak free carrier absorption absorption constant a»n l 2 would appear very thick layers necessary for efficient absorption heavy doping also required Absorption Length (µm) 00 0 n-type silicon, 0 20 cm -3 0 Wavelength (µm) Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

56 Free carrier absorption in thin silicon films transfer matrix method used for calculations 0.9 air silicon air easily handles multiple layers, complex index of refraction equivalent to microwave network ABCD matrices Absorption t t = 0.7 µm t = 0.36 µm Wavelength (µm) Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

57 Interference effects in moderately absorbing films Normalized Power t = l t = 3l / 4 t = l / 2 reflectance reflectance transmittance transmittance Wavelength (µm) absorbance s i l i c o n incident t = 0.7 µm t = l/2 half wave window t = 3l /4 impedance inverter t = l full wave window Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

58 Impedance Matching using "Backshorts" 0.36 µm thick silicon sheet 0.9 -R s dc 83?/square 0.8 with mirror silver mirror place 2.5 µm behind sheet at λ o = 5 µm: Absorption no mirror l / 2 l / Z air Z Si Z air Wavelength (µm) short open Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

59 0.36 µm Silicon Film with Mirror Normalized Power reflectance Si absorption Ag absorption A g m i r r o r 2.5 µm s i l i c o n incident t = 0.36 µm Wavelength (µm) Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

60 Enhanced Absorption using "Resistive" Coating l / 2 l / Å coating Z air Z Si Z air Absorption Å coating no coating thin Ag layer 0.36 µm Si film, Ag mirror 2.5 µm behind front surface coated with thin Ag film Wavelength (µm) -no coating -30 Å R s dc 6.7? /? -60 Å R s dc 3.3? /? Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

61 Is wavelength selectivity important? visible wavelength sensors: object reflectivity dominates reflectivity/absorption are functions of wavelength produces color in the image most energy emitted by an object at terrestrial temp. (300K) falls in the 3-4 mm range mid to far IR sensing: emissivity of object dominates emissivity is also a function of wavelength atmospheric propagation also influences band choice strong absorption bands H 2 O (2.6, 2.7, 6.3 mm) CO 2 (2.7, 4.3, 5 mm) transmission windows: 3-5 mm (MWIR) 8-4 mm (LWIR) window Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

62 Enhanced absorption in microbolometers Mirror layer thin conductor (absorber) Gap ( odd) 4 λ incident Thin conductor bolometer layer incident Zair Zair Zair Thin mirror layer to match an absorber to free space requires absorber: thin conductor with sheet resistance 377 ohms mirror placed [(odd integer)/4]*l behind absorbing layer essentially a Fabry-Perot cavity Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

63 IR wavelength selective structure and transmission line equivalent model n n- k 2 Incident IR Zair Zn Zn- Zair or Zk Z2 Z Zair Bolometer Layer (Rbolo) Mirror Layer(Rmirror) Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

64 Simulation of absorption of microbolometer Si 3 N 4 SiO 2 Si 3 N 4 Air Gap Si 3 N 4 SiO 2 Si 3 N 4 _COM _COM _COM Unit: Å Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

65 Resolving multi-wavelength ambiguities for a given gap, response occurs at all wavelengths where l = 4*gap/(odd integer) four color detector array gap (air equivalent): 2.5 mm -> l = 0 mm ghost response at 3.3 mm, 2 mm, etc. gap:.5 mm -> l = 6 mm ghost response at 2 mm & shorter gap: 0.83 mm -> l = 3.3 mm all ghost responses below 2 mm gap: 0.5 mm -> l = 2 mm all ghost responses below 2 mm assume external bandpass from just less than 2 mm to just over 0 mm Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

66 Resolving multi-wavelength ambiguities detector signal output S g=2.5mm 2mm (4g/5) 3.3mm (4g/3) 2mm (4g/3) ghost I 0mm I 6mm I 3.3mm I 2mm gap: 2.5mm.5mm 0.83mm 0.5mm I 0mm = S g=2.5mm - m 3.3->0 S g=.83mm - m 2->0 S g=.5mm I 6mm = S g=.5mm - m 2->6 S g=.5mm Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

67 fabrication procedure Cross section view STEP : formation of bottom membrane deposition of bottom membrane layers (Si 3 N 4 /LTO/SI 3 N 4 ) multi-lare stack used to compensate for stress in deposited films nitride in tension oxide in compression backside masking & RIE anisotropic KOH etching Top view Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

68 STEP 2: formation of sacrificial layer and top membrane deposition of sacrificial layer (LPCVD polysilicon deposition) deposition of top membrane layers (Si 3 N 4 /LTO/SI 3 N 4 ) Top view Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

69 STEP 3: back side removal of poly, removal of sacrificial layer remove backside Si 3 N 4 /LTO/SI 3 N 4 layers by RIE polysilicon by KOH patterning & RIE to remove top Si 3 N 4 /LTO/SI 3 N 4 remove sacrificial polysilicon by KOH Top view Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

70 STEP 4: self-aligned microbolometer structure deposition of back side mirror coating layer by evaporator deposition temperature sensitive bolometer layer Top view Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

71 Wavelength selective micromachined IR sensors Top view Top Membrane Bottom Membrane Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

72 Fabrication of proposed structure using IC comparable micromachining techniques bolometer active area bottom membrane Top view Bottom Membrane Top Membrane top membrane Bolometer layer Silicon Nitride Silicon oxide Mirror layer Silicon Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

73 Fabrication process for micromachined microbolometer Deposition of silicon nitride as barrier layer Deposition of sacrificial poly and patterning Deposition of membrane layers (Si 3 N 4 /LTO/SI 3 N 4 ) Patterning of membrane layers using RIE Poly etching and Self-aligned structure using KOH undercutting etching process Deposition of bolometer layer 3 um 26 um 20 um Active area 26 mm x 20 mm, 20 mm leg length, 3 mm leg width Si substrate Si 3 N 4 Poly SiO 2 Cr 20 um Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

74 Self aligned undercutting structure Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

75 Fabrication issues mechanical stability issues tensile nature of Si 3 N 4 + compressive nature of SiO 2 combination of these films weak tensile legs are still quite compliant, fragile Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

76 Thermal impedance approximation T o T o T hot T o T o Active area 26 mm x 20 mm, 20 mm leg length, 3 mm leg width w l = ( + t Rleg Rt t2 t3 t4 Where w = 3 µm, l = 20 µm R R t 2 t 4 R t + l t w κ R leg 4.3 x 0 6 K/W ) Z bolometer R leg 4 Z bolometer. x 0 6 K/W measured thermal impedance under ambient pressure is 6 x 0 4 K/W vs. calculated thermal impedance x 0 6 K/W simple still air estimate gives thermal resistance of about 2x0 5 K/W past work supports two-order reduction due to thermal losses to air at ambient pressure Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

77 Optical setup used for IR testing of microbolometers set-up in A. J. Welch s lab, with thanks to Dan Hammer two device types tested resonant cavity bolometer WITH mirror for enhanced wavelength dependent absorption, first resonance at 3 mm simple bolometer WITHOUT mirror DFG M Chromex Splitter Microbolometer under Mirror Sample xyz stage M det A S CWA A2 A3 M Flip top mount M M Visible HeNe IR HeNe 6 in FL det B A4 Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

78 IR spectrum of ultrafast laser system fairly wide spectral features l o ± Dl bolometer response P absorbed dx dy = dλ η ( λ ) S( λ, x, y) Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

79 3.00E+03 Irradiance on and output voltage of resonant cavity microbolometer 2.50E+03 Mean 2.00E+03.50E+03 Averaged.00E E Measured_with Calculated_with 0.00E Peak wavelength (micron) Peak Wavelength(Micron) Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

80 Responsivity at.5 and 3.39 mm, narrow-band excitation with mirror measured_without measured_with calculated_without calculated_with without mirror without mirror with mirror Wavelength(Micron) Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

81 Measured responsivity of resonant cavity microbolometer Mean_without Mean_with with mirror without mirror with mirror Peak Wavelength (micron) Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

82 Power coupling efficiency simulation_with simulation_without measured_without measured_with Wavelength (micron) Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

83 Noise measurement of the resonant dielectric cavity microbolometer.00e E-04 /f noise dominant Vn/Hz^/2.00E-05.00E-06 NEP (Noise Equivalent Power) at 500 Hz ~ 5 x 0-8 W/(Hz) ½.00E-07 Modulation Frequency (Hz) Detectivity (D*) at 500 Hz 6.9 x 0 4 cm (Hz) /2 /W Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

84 Noise equivalent temperature difference (NETD) smallest temperature change in object space that can be resolved assuming a unit signal to noise ratio NETD T η R thermal C thermal Where T: temperature, η: absorptivity, and C: thermal capacitance temperature resolution for IR microbolometer systems NETD of uncooled camera: claim that C is enough for TV quality public results SBRC (996) : 0 mk Mitubishi Electronics (996) : 500 mk Lockheed Martin (998): less than 00 mk Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin

85 Comparison of current commercial IR detectors Summary of Important Specification of IR dectors Near Infrared Mid Indium Uncooled Microbolometer Long QWIP Antimonide Microbolometer Type InGaAs InSb Microbolometer Microbolometer GaAs QWIP Spectral range um 5.5 um 7 4 um 7 4 um um Array format 320 x x x x x 240 Size 30 x 30 um 30 x 30 um 5 x 5 um 5 x 5 um 30 x 30 um Operating Temperature 29 K Uncooled 77 K Cooled 33 K or 293 K Uncooled 300 K Uncooled 77 K or less Cooled NedT < 20 mk < 00 mk < 00 mk < 30 mk Dean P. Neikirk 200, last update March 30, Dept. of ECE, Univ. of Texas at Austin