Device-level vacuum-packaged infrared sensors on flexible
substrates
Aamer Mahmood
Advisor:Prof. Donald P. Butler
Microsensors LaboratoryDepartment of Electrical Engineering
University of Texas at ArlingtonArlington, TX 76019
2
Outline
– Introduction– MEMS– Infrared radiation and detection– Bolometers– Flexible substrates
– Bolometers on flexible substrates
– Device-level vacuum-packaged microbolometers
– Fabry Perot cavity based tunable infrared microspectrometer
3
Microelectromechanical Systems (MEMS)
• Micro-Electro-Mechanical Systems (MEMS) is the integration of mechanical elements, sensors, actuators, and electronics on a common substrate through microfabrication technology.
4
Infrared radiationPlanck’s law
-0.2
0
0.2
0.4
0.6
0.8
1
1.2
0 5 10 15 20 25 30 35
BlackbodyGraybody
No
rmal
ized
exi
tan
ce
Wavelength(m)
5
Infrared Detectors• Photon Detectors
– Incident radiation generates photo carriers• Photovoltaic detectors• Photoconductive detectors• Photoemissive detectors
• Thermal Detectors– Incident radiation causes change in temperature that
causes a change in a detector property e.g.• Bolometers (change in temperature causes the
detector resistance to change)• Pyroelectric detectors (change in temperature causes
the detector capacitance to change)• Thermocouples (use Seebeck effect)
6
Bolometers and IR detection
eff
thth G
C
2)(1 theff
signalac
G
PT
7
Bolometers and IR detection
• Temperature induces a change in the detector resistance
η = absorptivity, = angular frequency of incident radiation, τ = detector thermal time constant, Psignal = the magnitude of the incident flux fluctuation
signaleff
b PG
IRV
2/122 )1(
signaleffb
b PG
V
RR
RI
2/1222 )1()(
Rb
R amp
V
RbI amp
8
Bolometer Figures of Merit
dT
dR
RTCR
1
2/122 )1(
eff
b
signalV G
IR
P
V
2/1222 )1()(
effb
b
signalI G
V
RR
R
P
I
Temperature coefficient of resistance
Responsivity
Normalized change in resistance w.r.t. temperature
Output/Input
9
Bolometer Figures of Merit
Detectivity
nVV V
AfD
nII I
AfD
Signal-to-noise ratio normalized to the detector area and frequency bandwidth
10
Bolometer materials
Material TCR (300K) Salient features
YBCO -3 to -3.5 %K-1 Room temp sputtering, no heat treatment
VOx -2 %K-1 Low noise
A-Si -2.7 %K-1 High doping with impurities, Crystalization by high temp annealing
P-Si -1 to -2 %K-1 High temperature annealing
P-Si_Ge alloy ~ -2 %K-1 High temperature deposition
11
Flexible substrates
• Polyester and Polyimide used as flexible substrates
• Polyimide is thermally stable
• Polyimide has a Tg of ~400°C
• Polyimide is chemically resistant to most clean room etchants
Property Units Polyimide Polyester
Thickness Range mils 0.3 - 5 0.25 - 14
Dielectric Constant 1 Mhz 3.4 3.2
Volume Resistivity W-cm 10 18 10 18
Tensile Strength (at 25oC) psi 40 000 27 000
Tear Initiation Strength gms 800 1200
Operational Temperature min/max oC .-200 to +300 .-60 to +105
Coefficient of Thermal
Expansion (at 20 oC)1/oC 10x10-6 20x10-6
Change in Linear
Dimension(150 oC, 30min)% >0.15 >1.5
Acid Resistance _ Good Good
Alkali Resistance _ Poor Poor
Grease/Oil Resistance _ Good Good
Organic Solvent Resistance
_ Good Good
Water Absorbtion % (24 hrs.) 3 >0.8
http://www-ee.uta.edu/zbutler/Smart_skin_for_web.ppt
12
Flexible systems
• Advantages of flexible substrate micro sensors– Low cost– Lightweight– Conformable to non planar surfaces– Software based printed IC processes– High degree of redundancy
13
Flexible systems
• Flexible electronics for personal communication (flexible electronic paper)
• Smart clothing (Wireless communications with smart sensors and actuators in the ambient)
• BioMEMS (flexible electrodes for neural prostheses, vision prosthesis)
• Conductive polymers (compound eye, piezoresistive strain sensors)
• Flexible energy sources (photovoltaic cells, organic solar cells)
14
Sensors on flexible substrates (Smart Skin)
• Sensor Arrays on flexible substrates– Infrared sensors– Pressure/Tactile Sensors– Flow sensors– Humidity sensors– Velocity sensors
15
Evolution of “smart skin” in the micro sensors lab
• First generation (2001-2002)– Used solid Kapton sheets pasted on to wafers
• Second generation (2003-2004)– Spin on Kapton used (no micromachining, not
separated from carrier wafer)
• Third generation (2004)– Spin on Kapton used (micromachined devices, peeled
off carrier wafer)
• Fourth generation (2005)– Vacuum packaging at the device level
16
Microbolometers on flexible substrates
17
Microbolometer FabricationTrench Geometry
Si
PI5878
Si3N4SrTiO3
PI2610Si3N4
SrTiO3Ti
AuYBCO
Al
18
Microbolometers on a flexible substrate
19
Temperature Coefficient of Resistance (TCR)
dT
dR
RTCR
1
0
5
10
15
20
25
30
-4.5
-4
-3.5
-3
-2.5
-2
240 250 260 270 280 290 300 310 320
Res
ista
nc
e (M
W)
%T
CR
(K-1)
Temperature (K)
20
Effects of Joule Heating
-3
0
3
-2 0 2
Vo
ltag
e (V
)
Current ()
)(2 TRIVIP bI
Iradtheff PGGG -
)()()( 02 TTTRITPTG bIeff -
)exp()( 0 kTERTR a
21
Responsivity/Detectivity
2/122 )1(
eff
v G
IRVR
nvv V
AfRD
100
1000
104
105
106
100
1000
104
105
106
1 10 100 1000
970 na730 na540 na
390 na214 na136 na
970 na730 na540 na
390 na214 na136 na
Re
spo
nsi
vity
(V
/W)
De
tectivity (c
m H
z1
/2/W)
Frequency (Hz)
22
Effects of substrate heating
23
Area scans of bolometersDevice 1b4 (Trench Geometry)
24
Area scans of bolometersDevice DD15 (Mesa Geometry)
25
Conclusion
• Microbolometers on flexible substrates have been fabricated
• Mean measured thermal conductance = 5.61x10-7 W/K • Max room temperature responsivity RV = 7.4x103 V/W• Max room temperature detectivity D*= 6.6x105 cmHz1/2/W• Measured room temperature TCR = -2.63%/K• Measured room temperature resistance = 3.76MΩ
26
Device-level vacuum-packaging
27
Device-level vacuum packaging
Si
PI5878
Si3N4SrTiO3
PI2610TiAu
Al
Si3N4
OTMSYBCO
28
Device-level vacuum packaging
Optical Window
Detector
Al Mirror
Bond Pad
29
Device-level vacuum packaging
Design Considerations
– Optical window transmission characteristics– Optical window structural analyses– Cavity vacuum– Thermal analyses
30
Optical Transmission Characteristics
Transmission characteristics of thin Aluminum Oxide film M. Aguilar-Frutis, M. Garcia, C. Falcony, G. Plesch and S. Jimenez-Sandoval, “A study of the dielectric characteristics of aluminum oxide thin films deposited by spray pyrolysis from Al(acac)3,” Thin Solid Films, vol 389, Issues 1-2, pp 200-206, 15 June 2001.
31
Complex relative permittivity of Al2O3
0
1
2
3
4
5
6
-1
0
1
2
3
4
5
0 5 10 15 20 25 30 35 40
'"
Wavelength(m)
-1
0
1
2
3
4
0 5 10 15 20 25 30 35 40
tan
Wavelength(m)
32
Optical Transmission Characteristics of polyimide
PI 5878G
0
20
40
60
80
100
120
0.9 1.9 2.9 3.9 4.9 5.9 6.9 7.9 8.9 9.9 10.9 11.9 12.9
Wavelength(um)
Tra
ns
mis
sio
n (
%)
TransmissionVerification
NoiseVerification
5 per. Mov. Avg.(TransmissionVerification)
33
Structural analysis of vacuum element
• Mechanical Strength – Ceramic Al2O3 has a tensile strength of 260
MPa– ZnSe has an apparent elastic limit of 55.1
MPa
34
Structural integrity of vacuum element
35
Al2O3 stress analysis
100
101
102
103
104
0.1 1 10 100
Al2O
3 stress vs. radius of curvature
Mises stress (MPa)Tensile strength (MPa)Compressive strength (MPa)
Str
es
s (M
Pa
)
r (cm)
36
Permeability through Al2O3
• Permeability is the flow rate through a specimen once steady state has been achieved
• He Permeability through Al2O3 at room temperature is
~100-1000 atoms/s/cm/atm•
n=number of moles
R=universal gas constant=8.314J/(mole.K) 10-6
10-5
10-4
10-3
10-2
10-1
1 10 100 1000 10000
Permeability=100atoms/s/cm/atm
Permeability=1000atoms/s/cm/atm
Pre
ssu
re (
mT
)
Time (Days)
V
nRTP
37
Thermal analysis (analytic)
Incomplete micromachining
Heat source ((300+ΔT)K)
Heat sink (300K)
Au
Ti arm
Al2O3
Au
Tipatch
Si3N4
Al2O3
Si3N4
Al2O3
Si3N4
PI2610
Top air
Au
Tipatch
Si3N4
Lower air
Si3N4
Lower air
Complete micromachining
Ruptured cavity
GA GB GC GD GE GF GG
38
Thermal analysis (numeric)
Gth ≈ 5x10-6 W/K (Vacuum)
≈10-4 W/K (air)
39
Microbolometer fabrication
Trench Geometry(Not to scale)
40
Fabrication(Silicon wafer)
41
Fabrication(PI 5878G)
42
Fabrication(Nitride)
43
Fabrication(Al)
44
Fabrication(Sacrificial Polyimide PI 2610)
45
Fabrication(Support Nitride)
46
Fabrication(Ti arms)
47
Fabrication(Au contacts)
48
Fabrication(YBCO detector pixel)
49
Fabrication(Photodefinable PI2737 sacrificial mesa)
50
Fabrication(Al2O3)
51
Section of vacuum cavity before micromachining
Al2O3
Sacrificial PI2737 mesa
Sacrificial PI2610
Al mirrorNitride
Nitride
52
Fabrication(Partially micromachined)
53
Fabrication(Fully micromachined)
54
Fabrication(Sealed vacuum cavity)
55
Fabrication(Superstrate PI 5878G)
56
Single microbolometer
57
Fabrication of encapsulated devices
Partially micromachined
device
Fully micromachined
device
SEM graph of an unsealed
micromachined device
58
Fabrication of encapsulated devices
Sealed device SEM graph of sealed device SEM graph of
cross section of vacuum cavity
Vacuum cavity
59
VI curve
-40
-30
-20
-10
0
10
20
30
40
-0.6 -0.4 -0.2 0 0.2 0.4 0.6
Vol
tage
(V
)
Current (A)
Measured Gth=3.73x10-6 W/K
60
Temperature Coefficient of Resistance (TCR)
40
60
80
100
120
140
160
180
-5
-4
-3
-2
-1
0
280 285 290 295 300 305 310 315
Res
ista
nce
(M
W)
TC
R (%
K-1)
Temperature (K)
R(300K)=53.4 MΩ
TCR(300K)=-3.7%/K
61
Current Responsivity (RI)
10-1
100
101
102
1 10 100 1000
10.09V7.20V5.48V3.66V
Res
po
ns
ivit
y (A
/W)
Frequency (Hz)
RI=61.3 μA/W
@ 5Hz
Current Responsivity (RI)
=Output current/Input power
62
Detectivity (D*)
102
103
104
105
106
1 10 100 1000
10.09V7.20V5.48V3.66V
Det
ec
tiv
ity
(cm
Hz1/
2 /W)
Frequency (Hz)
D* = 1.76x105 cm-Hz1/2/W
Detectivity (D*)
= Area and frequency normalized signal to noise ratio
63
Conclusion
• Device level vacuum encapsulated microbolometers on flexible substrates have been fabricated
• Theoretical thermal conductance in vacuum is 5x10-6 W/K • Measured thermal conductance is 3.73x10-6 W/K (Intact
Vacuum cavity)
• Measured room temperature TCR is -3.7%/K, resistance is 53.4MΩ
• Measured RI is 61.3 μA/W, D*=1.76x105cm-Hz1/2/W
64
• This work is supported by the National Science Foundation
The End
