1. Organic thin-film field-effect transistors (OTFTs) were fabricated and tested for chemical sensing applications. Pulsed gate operation was found to significantly reduce device baseline drift compared to static operation. 2. Charge transport in the organic semiconductor films occurs via multiple trapping and release of charge carriers. Variable temperature measurements showed thermally activated transport, with the activation energy dependent on gate voltage. 3. Exposure to chemical vapors causes a change in device characteristics due to the interaction of adsorbed analyte molecules with the doped organic semiconductor surface layer. This modifies both the surface doping level and trap energies.

1. 1
CChhaarrggee TTrraannssppoorrtt aanndd CChheemmiiccaall
SSeennssiinngg PPrrooppeerrttiieess ooff OOrrggaanniicc
TThhiinn--ffiillmmss
RRiicchhaarrdd YYaanngg
Material Science & Engineering
University of California, San Diego
06/12/2007

2. 2
PPrroojjeecctt BBaacckkggrroouunndd
AFOSR MURI: Integrated nanosensors for bio/chemical
warfare and explosive agents detection.
Organic thin-film chemical sensors
• Chemiresistors
• ChemFETs
Design objectives
• Sensitivity, Stability, Selectivity
• Integration in sensor platform
Conceptual design (2003)

3. 3
RReesseeaarrcchh AApppprrooaacchh
Charge transport
Device physics
Chemical sensing

4. 4
RReessuullttss BBeeffoorree CCaannddiiddaaccyy
• Analyte identification based on
dispersive charge transport
• Electrode independent chemical
responses in SCLC regime
2 Methanol
0
-2
-4
-6
-8
-10
-12
-14
0 ppm
380 ppm
950 ppm
1900 ppm
9500 ppm
19000 ppm
10-1 100 101 102 103 104 105 106
DG/G (%)
Frequency (Hz)
Appl. Phys. Lett., 88 (2006) 074104 J. Phys. Chem. B, 110 (2006) 361
The above results were based on two-terminal chemiresistors.

5. 5
CChheemmiiccaallllyy SSeennssiittiivvee FFiieelldd--EEffffeecctt
TTrraannssiissttoorrss
gas
Organic semiconductor
thin-film
S D
+ + + + + + + + +
Gate dielectric
Silicon Substrate (n+ )
G
Vg
Id
Vd
Ground
Advantages of ChemFETs as compared to chemiresistors:
• High chemical sensitivity and stability
• High electrical conductivity, therefore, may utilize very thin films

6. 6
DDeevviiccee FFaabbrriiccaattiioonn
25 mm
Photolithography, e-beam
evaporation, lift-off process
Organic thin-film deposited using
molecular beam epitaxy
• Film thickness: 5 - 50 nm
• Growth rate: 0.2 – 1 Å/sec
• Growth temperature: 20 – 200 0C
SiO2
Silicon Substrate (n+ )
Metal Phthalocyanine (MPc)
Metal center: Cu, Co, Fe etc
S
G (Au)
D

7. Low voltage operating ChemFET has been fabricated ( since Feb 2005).
7
DDeevviiccee CChhaarraacctteerriissttiiccss
0 -1 -2 -3 -4 -5
-3.0
-2.5
-2.0
-1.5
-1.0
-0.5
0.0
Source-drain Voltage (V)
( t nerr uC ni ar D m) A
Gate Voltage = -5 V
-4V
-3 V
-2 V
0 V
1, 2 V
30 nm CuPc/ 50 nm SiO2
• Low leakage current
• Ideal FET behavior
• Small threshold voltage
• Low operating voltage

8. 8
CuPc OTFT Characteristics iinn LLiitteerraattuurree
Appl. Phys. Lett. 69, 3066 (1996) J. Appl. Phys. 92, 6028 (2002)
SiO2 thickness = 300 nm
The operation voltages are 10 times too high for ChemFET applications.

9. Back gate process
• Protect gate dielectric with PR
• Dip into HF solution to remove
9
Fabrication IIssssuueess -- GGaattee LLeeaakkaaggee
0 -2 -4 -6 -8 -10
-14
-12
-10
-8
-6
-4
-2
0
Ig (mA)
Vds (V)
Vg
-14 V
+2 V
50 nm SiO2
Silicon Substrate (n+ )
Au
G (Au)
Au
Gate leakage problem persisted in first 3 months
backside SiO2
• E-beam evaporation of Au
Leakage sources and solutions:
• Defective gate oxide: solved by careful growth and inspection
• PR erosion by HF during backside SiO2 etching: solved by
developing BOE etching

10. 10
Fabrication IIssssuuee -- CCoonnttaacctt RReessiissttaannccee
Contact resistance limits current injection
-1 V
0 -2 -4 -6 -8 -10
-1.4
-1.2
-1.0
-0.8
-0.6
-0.4
-0.2
0.0
Ids (mA)
Vds (V)
50 nm CuPc
Vg = -14 V
-10 V
-8 V
-6 V
-4 V
-2 V
0 V
2 V
Source and solution:
• Residual PR forms hole injection blocking layer: solved by
developing cleaning procedure (three cycles of ultrasonication in
trichloroethylene/ acetone/ isopropyl alcohol))
Residual PR

11. -2 V/step
11
DDeevviiccee SSccaalliinngg
-6 7.5 micron channel
-12 10 micron channel
-10
-8
A)
m(Current 0 -2 -4 -6 -8 -10
Drain Source-Drain Voltage (V) -5
-4
-3
-2
-1
0
0 -2 -4 -6 -8 -10
-6
-4
-2
0
-6 15 micron channel
0 -2 -4 -6 -8 -10
-5
-4
-3
-2
-1
0
-6 20 micron channel
0 -2 -4 -6 -8 -10
-5
-4
-3
-2
-1
0
Vgs = -10 V
Vgs = +4 V
nw/L=20,400 +/-1,200 nw/L=30,300
nw/L=14,666 nw/L=15,000
25ML CuPc Thin-films

12. 12
LLiinneeaarr SSccaalliinngg ooff CCuurrrreenntt wwiitthh CChhaannnneell
Saturated region: V Vg = -8 V ds = -10 V
Vg = -6 V
Linear fits with R > 0.9
I W C V
( )2
= m V -
d,sat 2 i g t
5 10 15 20 25
-80
-70
-60
-50
-40
-30
-20
-10
Vg = -4 V
IDS/(nW) (mA*mm)
Channel Length (mm)
L
LLeennggtthh

13. Charge Transport iinn OOrrggaanniicc TTrraannssiissttoorrss
E
kT
m = m q = m exp æç - a
ö¸ eff
0 0 è ø m= effective mobility
eff13
p-type organic semiconductor
Delocalized conduction band
+ Localized
states
EF
+ +
Delocalized valence band
Multiple trapping and release (MTR)
Transport in delocalized band
Trapping and release
G. Horowitz, M. E. Hajlaoui, and R. Hajlaoui,
J. Appl. Phys. 87, 4456 (2000).
m0= free carrier mobility
q= free to total charge ratio
Ea = trap activation energy
Trap energy distribution determines the device characteristics

14. • Transconductance
• Activation energy
g g E
k T
14
VVaarriiaabbllee TTeemmppeerraattuurree SSttuuddyy
8
d
= ¶
¶
m Vds V
g
g I
V =-
0 exp( a )
m m
B
= -
• The charge transport is thermally activated.
• The activation energy depends on the gate voltage.

15. 15
Baseline DDrriifftt RReedduuccttiioonn iinn OOTTFFTTss
-2 0 2 4 6 8 10 12 14 16 18 20 22
1.0
0.8
0.6
0.4
Normalized Id
Vg = -8 V, pulsing
Vg = -4 V, static
Time (hr)
Vg = -8 V, static
0 20 40 60 80 100 120 140
1.05
1.00
0.95
0.90
0.85
0.80
0.75
0.70
Duty Cycle
Normalized Id
Time (minute)
1%
2%
5%
10%
20%
100%
(a)
threshold time
100 1000 10000
30
25
20
15
10
5
0
Drift (%)
Gate Bias Duration (ms)
(b)
• Static gate operation reduce drain
current 40% in 20 h
• Pulsed gating (0.1 Hz, 1% duty cycle)
reduce the drift to less than 1% in 20 h
• There is threshold pulse duration in the
baseline drift

16. 16
Gate
pulse
train
t
PPuullsseedd GGaattiinngg OOppeerraattiioonn
Ev
Vg = 0 V
Ec
Ef
SiO2
Off
State
Ef
Ev
V SiO2 g = -8 V
Ec
tt
On
State
•A pulse train from “off” to
“on” state is applied.
•Break lines represent trap
states located near SiO2
interface and in the bulk.
•“Off State” – at flat band
condition, no charge
accumulation in the channel.
•“On State” – holes
accumulate at the dielectric
interface. There is finite
amount of time (tt) for the
holes get trapped.

17. 17
BBaasseelliinnee DDrriifftt ttoo VVoollaattiillee VVaappoorrss
-2 0 2 4 6 8 10 12 14 16 18 20 22 24
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
(d)
Normalized Drain Current
Time (hr)
(a)
(b)
1900 ppm methanol pulses (c)
(a) 1% 0.1 Hz gate
with methanol
(b) static gate with
methanol
(c) static bias
without
methanol
(d) 1900 ppm
methanol
pulses
20 ML CuPc
• Pulsed gating reduced the baseline drift to 0.09 + 0.016 %/h in exposure to
15 methanol pulses.
• Pulsed gating reduced the error in chemical response by 10%.

18. 18
Baseline DDrriifftt ttoo LLooww VVaappoorr PPrreessssuurree
AAnnaallyytteess
0 4 8 12 16 20 24
1.00
0.95
0.90
0.85
0.80
0.75
Normalized Drain Current
Constant flow gas pulses
Time (h)
(a)
(b)
(c)
(a) 1% 0.1 Hz gate
with 32 ppm
DMMP
(b) 1% 0.1 Hz gate
with 19 ppm
DIMP
(c) Analyte pulse
sequence
• Chemical source of baseline drift has been tested with low vapor pressure
analytes
•There is 10% baseline drift due the tight binding of analytes

19. 19
CChheemmiiccaall DDrriifftt RReedduuccttiioonn
(i)
(ii)
0 4 8 12 16 20 24 28
30
20
10
0.85 (b)
(i)
(ii)
0 4 8 12 16 20 24 28
0.90
0.95
1.00
Normalized Id
Time (h)
0
(iii)
DIMP (ppm)
(a)
(iii)
(a) 20% DIMP duty
cycle
(b) 15% DIMP duty
cycle
(c) 8% DIMP duty
cycle
• Even in the presence of very low volatility analytes, the drift can be reduced to
zero by lowering the duty cycle of the analyte pulse.

20. Physical Structure BBaasseedd SSeennssiinngg MMooddeell
20
T. Someya, et. al. APL, 81, 3079 (2004)
L. Torsi, et. al., Ana. Chem. 77, 308 A (2005)
Assumptions
• Film mobility is determined by traps located
at grain boundary (GB)
• Analytes adsorbed at grain surface change
the GB barrier height EB and therefore
change device mobility and threshold voltage
Grain boundary model
Limitations
• No definite proof of trap state locating at
GBs in organic films by SKPM
• Weak correlation of chemical response with
grain size
• Electronic effect of oxygen doping
ignored
EB: Charge trapping barrier
Polycrystalline pentacene film

21. 21
Scanning KKeellvviinn PPrroobbee MMiiccrroossccooppee
Topography color scale: 20nm Potential color scale: 50mV
220000 nnmm
The potential drop between GBs is less than thermal energy.
Data acquired by Xiaotian Zhou

22. 22
EEvviiddeenncceess ooff OOxxyyggeenn DDooppiinngg
0 10 20 30 40 50
2.2
2.0
1.8
1.6
-3.6
-3.4
-3.2
-3.0
-2.8
• CuPc and F16CuPc sensing films out of vacuum are doped by oxygen
• Oxygen is an acceptor-like dopant as it withdraws electron from
phthalocyanines
• Displacing oxygen reduces p-channel device current, while increases n-channel
device current
-2.6
p-type
p-channel Id (mA)
n-channel Id(mA)
Time (h)
n-type

23. 23
Mo Electronic Moddeell ooff CChheemmiiccaall SSeennssiinngg
• Organic sensing films are doped by chemisorbed oxygen once
outside of vacuum.
adsorption charge transfer Ionization
k1 k2 + - k3 - +
2 2 2 2 MPc + O ؾ® MPc-O ؾ®MPcd -Od ؾ®MPc-O +h
• The surface layer has higher dopant concentration.
SiO2 Air O2
Si CuPc Air
Ef -
x0
Ec
Ev
“delta-doping”
• Chemical analytes adsorption on film surface has 2 effects:
2 2 O /O
s j
– Surface doping level change due to oxygen displacement
– Trapping energy change due to new energy states formed by analyte

24. 24
CChheemmiiccaall SSeennssiinngg MMeecchhaanniissmmss
1.39
1.38
1.37
1.36
1.35
n-channel
Abs (Id) (mA) Time (h)
-2 0 2 4 6 8 1012141618202224262830
1.50
1.45
1.40
1.35
1.30
1.25
1.20
0 60 120 180
1.34
I(mA)
MeOH (1520 ppm)
p-channel
DMMP (68 ppm)
DI

25. 25
TThhee EExxppoonneennttiiaall DDeeccaayyss
-1.015
-1.020
-1.025
-1.030
-1.035
-1.040
0 2000 4000 6000 8000 10000 12000
1.46
1.44
1.42
1.40
1.38
1.36
1.34
Equation: y = A1*exp(-x/t1) + y0
Weighting:
y No weighting
Chi^2/DoF = 2.1604E-6
R^2 = 0.9964
y0 1.33026 ±0.0006
A1 0.11914 ±0.00048
t1 7084.0095 ±73.686
(a). n-channel: DMMP
0 2000 4000 6000 8000 10000 12000
-1.045
(b). p-channel: DMMP
Equation: y = A1*exp(-x/t1) + y0
Weighting:
y No weighting
Chi^2/DoF = 1.3636E-7
R^2 = 0.99732
y0 -1.04583 ±0.0001
A1 0.03262 ±0.00008
t1 5736.22502 ±42.75252
0 2000 4000 6000
1.39
1.38
1.38
1.37
1.37
1.36
1.36
1.35
1.35
Equation: y = A1*exp(-x/t1) + A2*exp(-x/t2) + y0
Weighting:
y No weighting
Chi^2/DoF = 8.5004E-7
R^2 = 0.99005
y0 1.35052 ±0.00026
A1 -0.01143 ±0.00066
t1 57.13782 ±5.90263
A2 0.03778 ±0.0002
t2 2331.16822 ±44.41452
(c). n-channel: MeOH
Id (t) (mA)
-0.99
-1.00
-1.01
-1.02
-1.03
-1.04
Time (second)
raw data
Exponetial fit
0 2000 4000 6000
-1.05
(d). p-channel: MeOH
Equation: y = A1*exp(-x/t1) + A2*exp(-x/t2) + y0
Weighting:
y No weighting
Chi^2/DoF = 1.5898E-7
R^2 = 0.99774
y0 -1.04404 ±0.00007
A1 0.04283 ±0.00025
t1 189.38934 ±2.13073
A2 0.01364 ±0.00019
t2 1622.09746 ±43.98648

26. 26
Concentration DDeeppeeddeennddeenntt
0 2 4 6 8 10 12 14 16 18
1.46
1.44
1.42
1.40
1.38
1.36
50
40
30
20
10
0
Drain Current (mA)
Time (h)
Conc. (ppm)
(a). n-channel: DMMP
-2 0 2 4 6 8 10 12 14 16 18
-1.06
-1.05
-1.05
-1.04
-1.04
-1.03
-1.03
-1.02
0
10
20
30
40
50
Drain Current (mA)
Time (h)
Conc. (ppm)
(b). p-channel: DMMP
n-channel
0 10 20 30 40 50 60
105
90
75
60
45
30
15
0
p-channel
D I (nA)
Concentration (ppm)
1 exp b d I E S
= D = - æ- ö I c çè kT
ø¸

27. 27
BBiinnddiinngg ttoo aa WWeeaakk BBiinnddeerr
p-channel
0 500 1000 1500
75
60
45
30
15
0
n-channel
D I (nA)
Concentration (ppm)
2000
1500
1000
500
0
0
500
1000
1500
2000
1.40
1.38
1.36
-1.46 (b). n-channel
0 2 4 6 8 10 12
-1.44
-1.42
-1.40
-1.38
-1.36
0 2 4 6 8 10 12
1.34
Concentration (ppm)
(a). n-channel
Time (h)
Drain Current (mA)

28. 28
UUllttrraasseennssiittiivvee SSeennssoorr DDeessiiggnn
Merge the 2 interfaces: ultrasensitive ChemFET design
In conventional OTFT sensors (> 10 nm), the chemical sensing and
charge transport interfaces are separated.

29. 29
Chemical RReessppoonnssee CCoommppaarriissoonn
4 ML CoPc
DIMP Air NBMeOH 50 ML CoPc
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
-16
-14
-12
-10
-8
-6
-4
-2
-10
-8
-6
-4
-2
12 TE
DIMP
/1.9
NB
/0.35
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
Appl. Phys. Lett., In Press
8
4
0
/44
EA
/150
MeOH
Conc (ppm)
Time (h)
MeOH
/190
Air
0
EA TE
Response (%)
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
0
EA TE Air DIMP NB

30. 30
S Chemical Seennssiittiivviittyy EEnnhhaanncceemmeenntt
Sensitivity enhancement has been observed on all 5 analytes.
0 exp a E
- = æ ö è ç kT
¸ ø
m m
m 0 is related to carrier density
Ea is the trap energy at
CoPc/SiO2 interface
0 1 2 3 4
16
14
12
10
8
6
4
2
0
Ethyl
acetate
2.2
Sensitivity Enhancement
Dipole Moment (Debye)
Toluene
1.7
MeOH
4.0
Nitrobenzene
15.8
DIMP
3.62
Effective field-effect mobility
In the ultrathin device, the air/CoPc and CoPc/SiO2 interfaces are so
close that analytes affect both carrier density and trap energy.

31. Nitrobenzene
Simulant for TNT
31
Detection ooff NNiittrroobbeennzzeennee VVaappoorrss
1.0 0 1 2 3 4 5 6 7 8 9 10 11 12
0 1 2 3 4 5 6 7 8 9 10 11 12
-2.61
Id (mA)
-2.70
1.0
0.5
0.0
Flow Rate (sccm)
Time (hr)
0.1 0.2
0.6
0.8
-2.79
Time (hr)
Vg = - 8 V
Vds = -4 V
35 ppb 70 ppb
210 ppb
350 ppb
280 ppb
70 ppb nitrobenzene has been detected without a precentrator

32. 2006 – 8 Pack
with a blower in
handheld package
32
2004
PPrroojjeecctt EEvvoolluuttiioonn
2005
2006 – 6 Pack
2004: Three parallel electrodes
2005: Interdigitated electrodes
2006: 6 pack ChemFET for an e-nose.
2006: Handheld package. On-board integration of temperature, humidity
sensors and current amplifier.

33. 33
WWiirreelleessss HHaannddhheelldd PPaacckkaaggee
LLaabbvviieeww iinntteerrffaaccee
VVaappoorr ssaammpplliinngg wwiitthh aa vvaaccuuuumm
IInntteeggrraatteedd PPCCBB
SSeennssoorr eenncclloossuurree
BBlluuee ttooootthh ttrraannssmmiitttteerr

34. 34
SSuummmmaarryy
• Process of low voltage operating and repeatable ChemFETs have
been developed.
• Trap states are found to dominate charge transport in organic
transistors.
• Pulsed gating technique has been developed to reduce drift to less
than 0.1%/h in ChemFETs.
• A ChemFET sensing model has been developed: gas adsorption on
organic semiconductor surface changes both doping concentration
and trap energy.
• Ultrasensitive ChemFETs have been developed to detect explosive
simulant at ppb level.
• The project has evolved from discrete device to integrated circuits
to handheld packages.

35. 35
AAcckknnoowwlleeddggeemmeennttss
Committee Members:
• Prof. Andrew Kummel (Chair)
• Prof. Sungho Jin (Co-chair)
• Prof. Yu-Hwa Lo
• Prof. William Trogler
• Prof. Edward Yu
Collaborators (MSE, Chemistry, Physics)
Jeongwon Park, Xiaotian Zhou, Corneliu Colesniuc, Dr. Karla Miller,
Dr. Amos Sharoni and Dr. Thomas Gredig
Undergrads (ECE, CSE and MAE)
Ti, Tammy, Kate, Casey, Jordan, Sureel, Byron and Vince
Funding from AFOSR MURI

36. 36
R Education/Reesseeaarrcchh BBaacckkggrroouunndd
MM..SS..
SSuurrffaaccee CChheemmiissttrryy
BB..SS..
CChheemmiiccaall EEnnggiinneeeerriinngg
CChheemmiissttrryy
RReesseeaarrcchh OOffffeerr
IInnsstt.. ooff MMaatteerr.. RReess && EEnngg
MM..SS..
AAddvvaanncceedd MMaatteerriiaallss
MMaatteerriiaallss
PPhh..DD..
MMaatteerriiaallss SScciieennccee && EEnnggiinneeeerriinngg

37. R Chemical Reessppoonnssee ooff pp aanndd nn CChhaannnneellss
EEcc
EEff EEDDIIMMPP
EEooxxyyggeenn
EEvv
37
p-channel n-channel
-18 CuPc 50nm, L = 5 micron
-16
-14
-12
-10
-8
-6
-4
-2
0
2
190 ppm DIMP
-20 -15 -10 -5 0 5
Drain Current (mA)
Gate Voltage (V)
air
DIMP
Vds = -6 V
7 F16CuPc 50nm, L = 5 micron
25 20 15 10 5 0 -5
6
5
4
3
2
1
0
Drain Current (mA)
Gate Voltage (V)
air
DIMP
Vds = +6 V
CuPc F16CuPc
EEcc
EEff EEDDIIMMPP
EEooxxyyggeenn
EEvv
Analyte adsorption changes free carrier concentration and trap energy.

38. 0 2 4 6 8 10
38
RRoollee ooff OOxxyyggeenn iinn SSeennssiinngg
2.0
1.9
1.8
1.7
1.6
N(a). n-channel 2
Drain Current (mA) Time (h)
-2 0 2 4 6 8 10 12 14 16 18 20
-2 0 2 4 6 8 10 12 14 16 18 20
1.5
1.4
-1.2
-1.1
-1.0
-0.9
-0.8
-0.7
-0.6
air (b). p-channel
N2
1.3
air
0 2 4 6 8 10
1.7
1.6
1.5
-1.1
-1.0
-0.9
-0.8
-0.7
N2
(b). p-channel
air
Drain Current (mA)
Time (h)
1.4
N2
air
(a). n-channel
NNoott ddiirreecctt ddiissppllaacceemmeenntt..
AA mmiixxttuurree ooff tthhee ccoo--
aaddssoorrppttiioonn aanndd rreemmoottee
aaddssoorrppttiioonn

39. 39
Macroscopic VViieeww ooff CChhaarrggee TTrraannssppoorrtt
•Low voltage region, Ohmic
conduction
= m V
0 J N e
d
• High voltage region, space-charge
limited conduction
2
3
J = em V
9
8 d
J = current density
N0= thermal carrier concentration
e = permittivity of material
V = voltage bias
d = film thickness

40. 40
Scanning KKeellvviinn PPrroobbee MMiiccrroossccooppee
SSKKPPMM::
SSuurrffaaccee ppootteennttiiaall,, eelleeccttrriiccaall ffiieelldd aanndd cchhaarrggee ddiissttrriibbuuttiioonn
An oscillating voltage is applied on the cantilever tip,
Vac sinwt, which creates an oscillating electrostatic
force at the frequency
F dC V V t x
( sin( ) ( ))
= + w -f
2 dz
dc ac
When Vdc = f (x) , the cantilever feels no electrostatic
force, the surface potential f (x) is recorded as the tip
voltage.
First scan: topography
Second scan: potential

41. 41
Microscopic VViieeww ooff CChhaarrggee TTrraannssppoorrtt
E( x) = - d f
(x)
dx
• Ohmic conduction (low voltage):
linear V(x) and uniform E(x) in the
channel. No net charge in the film.
• SCLC (high voltage): parabolic V(x)
and non-uniform E(x) as a consequence
of space charge buildup.

42. • AAtt hhiigghh vvoollttaaggee,, cchheemmiiccaall rreessppoonnssee iiss iinnddeeppeennddeenntt ooff ccoonnttaacctt aanndd hhiissttoorryy
• TThhee iinntteerrffaaccee ttrraappss aarree ffiilllleedd uupp tthhaatt ddoo nnoott aaffffeecctt cchheemmiiccaall sseennssiinngg
42
EElleeccttrrooddee IInnddeeppeennddeenntt CChheemmiiccaall
RReessppoonnssee iinn SSCCLLCC RReeggiioonn
J. Phys. Chem. B, 110 (2006) 361

43. 43
IImmppeeddaannccee SSppeeccttrroossccooppyy
Input Output
( ) v R( ) ( )
Z w = = w - iX w
Resistance: R( ) 1
Reactance: X 1
J. Phys. Chem. B, 110 (2006) 361
i
G( )
w
w
º
( w
) º
w C
( w
)
• Low and high frequency semicircles
co-exists
• The low frequency semicircle
deceases with increasing field
• The 2 semicircles relate to interface
and bulk traps

44. 44
Analyte Identification UUssiinngg IImmppeeddaannccee
Input Output
2 Methanol
0
-2
-4
-6
-8
-10
-12
-14
AC conductivity
0 ppm
380 ppm
950 ppm
1900 ppm
9500 ppm
19000 ppm
10-1 100 101 102 103 104 105 106
DG/G (%)
Frequency (Hz)
Y i
( w ) = ac = G ( w ) + iwC ( w
)
v
ac
G (w) AC conductance.
C (w) capacitance.
• AC conductance change (> 10kHz) is
independent of methanol concentration
above 950 ppm.
• DC conductance changes linearly with
concentration.
Differential AC conductance on 50 nm CoPc thin film w/o analyte

45. 45
AC Conductance vvss.. CCoonncceennttrraattiioonn
4 Isopropanol
2
0
-2
-4
-6
-8
-10
-12
525 ppm
1050 ppm
4200 ppm
5250 ppm
21000 ppm
10-1 100 101 102 103 104 105 106
DG/G (%)
Frequency (Hz)
12
10
8
6
4
2
0
-2
-4
-6
-8
-10
Ethanol
275 ppm
850 ppm
4250 ppm
8500 ppm
17000 ppm
10-1 100 101 102 103 104 105 106
DG/G (%)
Frequency (Hz)
• AC conductance change is concentration independent for ethanol and
isopropanol above critical levels.
• There are distinct binding sites with different analyte absorption energies,
which can be used for analyte identification.
Appl. Phys. Lett., 88 (2006) 074104

46. Low High
46
104
Resonance FFrreeqquueennccyy DDeetteeccttiioonn
Z( w ) 1 ( ) R ( ) i
X
( ) = = w - w
Y
w
Reactance: X 1 - L
( ) ( )
w º w
wC w
Dissipation factor
R
( w
)
DF
=
X
( w
) Impedance Spectroscopy
(2 ppm) (19 ppm)
Dissipation (a.u.) Frequency (kHz)
11.5 11.6 11.7 11.8
700
600
500
400
300
200
100
0
-100
-200
-300
-400
-500
Air
Methanol
Nitrobenzene DIMP
(1900 ppm)
103
103 104 105
Frequency (Hz)
X (w)
-100000
-50000
0
50000
103 104 105
Frequency (Hz)
( X w )
Frequency (Hz)
0
X (w ) ®0
Resonance
Frequency (Hz)
noit api ssi D
Low High
Appl. Phys. Lett., 88 (2006) 074104

47. 47
SSuummmmaarryy –– 22 TTeerrmmiinnaall DDeevviiccee
• Charge transport in organic thin-film is Ohmic at low
field and SCLC at high field.
• Operating Chemiresistors in SCLC region gives contact
independent chemical responses.
• There are co-existence of low frequency and high
frequency transport states in organic thin-film.
• An impedance spectroscopy technique has been
developed to identify chemical analytes based on
dispersive charge transport.