Transcript Plastic

“Plastic” Electronics and Optoelectronics
Low Cost “Plastic” Solar Cells
Light emitting Field Effect Transistor (LEFET)
CLEO
Baltimore
May 7, 2007
Low Cost “Plastic” Solar Cells:
A Dream or Reality???
We have an energy problem.
Nuclear, hydroelectric, wind etc must all contribute to
the solution.
Solar cells --- Power from the Sun must be --and will be a significant contribution.
Two problems that must be solved with solar:
1. Cost
2. Area
Printing of Plastic Electronics
“inks” ---- with
electronic functionality!
Plastic Substrate
The Dream

Solar Cells
Functional
Ink

“Plastic” Solar Cells
Ultrafast charge separation with quantum
efficiency approaching Unity !
1992
ħ
RO
RO
RO
RO
OR
OR
OR
e-
OR
50 femtoseconds!!
Ultrafast electron transfer is important --Charge transfer is 1000 times faster than any competing
process,
and
Back charge transfer is inhibited
Therefore,
Quantum efficiency for charge separation approaches unity!
Every photon absorbed yields one pair of separated charges!
(Route to photovoltaics and photodetectors)
How do we create a material with
charge-separating junctions everywhere??
ħ
RO
RO
RO
RO
OR
OR
OR
e-
OR
All must be accomplished at nanometer length scale!
OMe
O
S
P3HT
n
PCBM
Bulk Heterojunction Material
Bicontinuous interpenetrating network
Self-assembled nanoscale material with
charge-separating junctions everywhere!
“Bulk” D-A Heterojunction Material
A self-assembled nanomaterial
Electrical Contact
Electrical Contact
Must break the symmetry --- use two different metals with
different work functions.
Electrons will automatically go toward lower work function
contact and holes toward higher work function contact
“Bulk” Heterojunction Material
Before 150oC 150oC anneal
anneal
30 min
150oC anneal
2 hours
OMe
O
S
P3HT
n
PCBM
TEM images of the P3HT/PCBM
interpenetrating network
P3HT/PCBM (1:0.8) prior to annealing
After annealing for 10 minutes at 150oC
Spatial Fourier Transform
P.S.D.(a.u.)
10min annealing
Room temperature


0.01
0.1

1
Spatial Frequency(1/nm)
II. Relevant length scale --- peak corresponds to structure
with “periodicity” of 16-20 nm
I.
III.
Large scale segregation or slow variation in thickness
for distances > 100 nm
Noise at very short length scales
Spatial Fourier Transform
P.S.D.(a.u.)
10min annealing
30min annealing
2hour annealing
0.01
0.1
Spatial Frequency(1/nm)
1
High temperature annealing --- 150o C
1. Stable for long times at High-T
2. 10 minutes is sufficient to lock in the structure
”Period” 160 -200 Å
Distance from any point in the material to a
charge separating interface  40-50 Å
Less than the mean exciton diffusion length --High charge separation efficiency.
Solar Cell Performance
Device structure: ITO/PEDOT/P3HT:PCBM/Al.
Eff = 5.0%
VOC = 0.625 V
FF = 68%
w/o anneal
70oC, 30 min
150oC, 30 min
Series resistance decreased from 113  to 7.9 
New Device Architecture: Optical Spacer
Conventional Device
Glass
ITO
PEDOT Active Layer
Device with Optical Spacer
Al
Glass
ITO
PEDOT Active Layer
Al
Light intensity in
Dead zone
Optical spacer
Device architecture
OR
Ti
OR
OR
+ H2O
OH
Ti
OH
OH
OH
Glas
s
ITO
P3HT
PED :PCB
OT:P M
TiOX
SS
Al
um
inu
m
O
Ti
O
Ti
O
O
Ti
O
O
TiOX
Charge separation layer
Ti
Power Conversion Efficiency
Green light (532 nm)
Efficiency
e = 8.1% → 12.6%
AM1.5 (solar spectrum)
Efficiency
e = 2.3% → 5.0%
Optical spacer: 50% improvement !
Polymer Solar cells: Present status in our labs --- 5% - 6%
What can we expect to achieve?? --- Eff  15%
New Architecture --- optical spacer --- 50% improvement
(already included)
Devices with eff.  5 - 6% fabricated with P3HT
Band gap too large --- missing half the solar spectrum
Opportunity: Potential for 50% improvement using
polymer with smaller band gap
Increase open circuit voltage --Opportunity: Potential for 50% improvement
Tandem Cell --Opportunity: Potential for > 50% improvement
Semiconducting polymers with smaller band gap?
Zhengguo Zhu
(a)
(ZZ50)
S
N
N
(c)
(
S
)n
S
O
OCH 3
(b)
Absorption and photoresponse
out to 900 nm in the IR.
Should be capable of 7% in single cell
and >10% in a Tandem Cell with P3HT
Photocurrent (arb.u.)
n
1E-4
1E-5
1E-6
1E-7
ZZ50/ZZ50:PCBM
1E-8
Photocurrent (arb.u.)
S
1E-4
1E-5
1E-6
1E-7
1E-8
P3HT/P3HT:PCBM
300
400
500
600
700
Energy (eV)
800
900
1000
(a)
S
N
N
(c)
(
S
ZZ50
5.5%
)n
S
O
(b)
S
n
Best performance for a single cell architecture
OCH 3
Tandem Cell
2
Current Density (mA/cm )
(Multilayer architecture equivalent to two solar cells in series)
0
-2
-4
-6
-8
PCPDTBT single cell
P3HT single cell
Tandem cell
-10
-12
0.0
0.2
0.4
0.6
0.8
1.0
1.2
Bias (V)
Open circuit voltage doubled
Efficiency 6.5%
We can do even better ----
Increased performance of
each of the two subcells --Already demonstrated.
Goal for coming months: 10%
Air-stable polymer electronic devices
Thin layer of amorphous TiOx (x<2) improves performance
and
Enhances Lifetime of Polymer Based Solar Cells
Efficiency (%)
Bulk Heterojunction Solar Cells
Single TiOx passivation layer signficantly enhances the lifetime
Factor of 100 !
1
with TiOx
0.1
Conventional
0.01
0
50
100
150
Time (Hours)
Goal: Simple inexpensive barrier materials will be
sufficient for achieving long lifetime.
Lifetime: Light soak of flex devices

Flex OPV with low cost packages pass 1000 hrs @
65°C, 1 sun
1,2
Efficiency [a.u.]
1,0
0,8
0,6
0,4
0,2
0,0
0
200
400
600
800
1000 1200
Time [hours]
Fully flexible devices under 1 sun, 65°C
Flexible devices packaged with commercial, low cost films
(Konarka)
Lifetime: Damp heat stability
Flex OPV with low cost packages pass 500 hrs in damp heat.
Cells are fairly stable, though slow degradation observed
% Efficiency change under 65C/85%RH
40
Structure 1
Structure 4
% Change of Efficiency
20
0
0
200
400
600
800
1000
1200
-20
-40
-60
-80
-100
Time (hours)
Flexible devices packaged with commercial, low cost films
Environmental lifetime (Konarka)
Little (no!) degradation observed after ½ year outdoor testing.
OPV BL2 Module Aging on Rooftop Test Station - under Load
(Encapsulated, with UV Protection)
80%
G4
2
OPV modules loaded
top test
Data on
at 1 roof
Sun (1,000
W/min
) Lowell, MA
80
G6
S5
60%
70
S6
60
50
20%
40
0%
30
20
-20%
Air Temperature (°C)
Change in Power (%) x x
40%
10
-40%
-60%
t0 Efficiencies:
G4: 1.35%
G6: 1.21%
S5: 1.88%
S6: 1.41%
0
-10
Air Temperature
-80%
9/20/06
-20
10/4/06
10/18/06
11/1/06
11/15/06 11/29/06 12/13/06 12/27/06
Date
1/10/07
1/24/07
2/7/07
2/21/07
3/7/07
Proprietary Information of Konarka Technologies Inc.
Reduced efficiency observed in the winter period is not degradation
--- efficiency increases with increasing temperature.
Polymer Solar cells: Present status in our labs --- 5 - 6%
What can we expect to achieve?? --- Eff  15%
New Architecture --- optical spacer --- 50% improvement
(already included)
Devices with eff.  5 - 6% fabricated with P3HT
Band gap too large --- missing half the solar spectrum
Opportunity: Potential for 50% improvement using
polymer with smaller band gap (ongoing ---)
Increase open circuit voltage --Opportunity: Potential for 50% improvement
Tandem Cell --Opportunity: Potential for > 50% improvement
Can we achieve Eff > 10% with the polymer
“bulk heterojunction” nano-material?
Active development program at Konarka Technologies
in USA and in Europe.
Goal: Roll-to-roll manufacturing by printing and
coating technologies
Joint Venture: Konarka - KURZ (Germany)
Low Cost “Plastic” Solar Cells:
A Dream Becoming a Reality
Gate-Induced Insulator-to-Metal Transition --Field induced metallic state in organic FETs
Light Emitting FETs
and --- the relationship between the two
Field Effect Transistor
Drain
Source
Semiconductor Layer
++++
Gate Insulator
Gate Electrode
When voltage is applied between gate and source-drain,
mobile electronic charge carriers (positive or negative) are induced
into the semiconductor.
Can we increase the field induced carrier density to levels sufficient to cross the
insulator-to-metal boundary???
Can we increase the field induced carrier density to
high enough levels sufficient to cross the
insulator-to-metal boundary???
Q = CV = (A/d)V
N = A(/e)(V/d)
nA = (/e)(V/d) electrons (or holes) per unit area
Vmax/d = EBreakdown
These charges are confined to a thin layer (approx 10 nm)
in the semiconductor adjacent to the interface with the
gate dielectric.
Conclusion: n  1020 cm-3 --- approaching metallic densities
Gated four-probe measurement (P3HT)
S
n
P3HT
• Material deposition by spin
casting
source
• Device annealed at 235C
• Carrier mobilities  0.1 cm2/Vs
L = 16 m
drain
W = 1000 m
SiO2
dielectric,
200 nm
gate, Si (n++)
• We measure the voltage drop inside the
transistor channel
• Typically Rcontact < channel resistance
Field-induced M-I Transition
in poly(3-hexyl thiophene) --- PRL (2006)
10
1
10
0
S
n
10
-1
10
-2
10
-3
10
-4
10
-5
0
-2
ln
-1
-1
 ( cm )
P3HT
0.1
0.2
1/T
0.3
1/2
(K
0.4
-1/2
)
0.5
-4
-6
-8
-10
-12
VG = 150 V
Metallic
VG = 50V
Critical line
2
VG = 40 V
Insulator
3
4
5
lnT
Power law indicates critical line
M-I Transition in 2d --- (Zero-T Quantum Phase Transition)
Conclusion:
We can reach metallic densities in organic FETs
(gate–induced through field effect)
n  1020 cm-3
Question:
Can we reach metallic densities (field-induced)
for both electrons and holes in the same device?
Answer: Yes!
Light-emitting Field Effect Transistor (LEFET)
Polymer Light Emitting Field-Effect Transistor
hv
Drain
(Ag)
Conjugated Polymer
+++++++
+++++++
-----------------
Source
(Ca)
Gate Dielectric
Gate
Qualitative description of device operation:
Gate controlled “p-n junction”
Light emission from the overlap recombination zone
Gate voltage controls the electron current,
the hole current, and the brightness!
| I ds|
[ A]
10
8
6
1E-8
4
PMT Current
12
1E-7
| a.u.|
14
2
1E-9
0
30
60
Vgs
90
120
0
150
[ V]
h
150 V
Au
h+ transport
dominates
ground
Ca
Vg < Vsd/2
Ambipolar
transport Vg  Vsd/2
dominates
e- transport
dominates Vg > Vsd/2
Maximum efficiency at crossover point
where electron and hole currents are equal.
| I ds|
[ A]
10
8
6
1E-8
4
2
1E-9
0
30
60
Vgs
90
120
0
150
[ V]
h
Ambipolar
transport Vg  Vsd/2
dominates
PMT Current
12
1E-7
| a.u.|
14
Device physics analyzed in detail by Smith and Ruden
Ids
[A]
1E-6
1E-7
1E-8
0
50
100
150
200
Vgs [V]
experiment
theory
D. L. Smith and P. P. Ruden, Appl Phys Lett 2006, 89, 233519
|Ids|1/2 vs gate voltage,
(where Ids is the total current between the Au and Ca electodes).
I d  Ci
W
Vg  Vth 2
2L
A
µh = 2.1 x10-5 cm2/V/s
20
µe = 1.6 x10-5 cm2/V/s
1/ 2
30
| I ds|
1/ 2
40
10
0
0
25
50
75
Vgs
Vth(holes) = 150V -103V = 47 V
Vth(electrons) = 65 V
100
125
150
[ V]
Large threshold voltages imply disorder
and a high density of traps at the
semiconductor-dielectric interface.
|Ids|1/2 vs gate voltage,
(where Ids is the total current between the Au and Ca electodes).
I d  Ci
W
Vg  Vth 2
2L
A
20
1/ 2
30
| I ds|
1/ 2
40
10
0
0
25
50
75
Vgs
100
125
150
[ V]
The LEFET operates in the ambipolar regime with electrons in the *-band
and holes in the -band only for a narrow range of gate voltages
65 V < VG < 103 V.
(b)
(a)
Calculated curves (Smith and Ruden):
Gate-to-Channel potential and Carrier Densities (electrons and holes) vs x
Gate to channel potential

0 between the hole and electron accumulation regions.
This location defines the center of the light emission zone
(and the center of the charge recombination zone).
D. L. Smith and P. P. Ruden, Appl Phys Lett 2006, 89, 233519
Location of emission zone controlled by gate voltage
VG = 87 V
Good agreement with
Smith and Ruden
VG = 93 V
VG = 99 V
Definitive demonstration of LEFET
Spatial Resolution of the Emission Zone (EZ)
Diagram of a confocal microscrope
(Prof. S. Burrato, UCSB)
Images of the emission zone collected as
EZ is moved across the channel

Cross section plots
of emission intensity
vs. lateral position
Full width at half maximum approx 2 m
(determines recombination rate, Smith & Ruden)
Emission at the extremes
--VG  0 and VG  150 V
Occurs adjacent to the Ca and the Au electrodes:
PLEDs with electron (hole) accumulation layer extending
all the way across the channel as electrode with hole
(electron) injection by tunneling
| I ds|
[ A]
10
8
6
1E-8
4
PMT Current
12
1E-7
| a.u.|
14
2
1E-9
150 V
Au
ground
Ca
Hole
accumulation layer acts as
h+ transport
Vg < Vsd
/2
low resisitance cotnact
across
dominates
the channel
0
30
60
Vgs
90
120
0
150
[ V]
Electron
accumulation layer acts as
e- transport
low resistance contact
across
Vg > V
sd/2
dominates
the channel
Conclusions
At the voltage extremes, the electron (hole) density extends all the way across the 16 m
channel length such that the accumulation layer functions as the electrode for an LED with
opposite carrier injection by tunneling.
In these two limits the carrier densities (electrons in the *- band or holes in the - band)
are sufficiently high that the accumulation layer functions as a low resistance contact,
implying near metallic transport.
These very long distances for electron and hole accumulation, respectively, more than
10,000 repeat units on the semiconducting polymer chain and more than 10,000 interchain
spacings imply the existence of well-defined - and *- bands with a very small density
of deep traps within the band gap.
The LEFET as the route to injection lasers fabricated
from luminescent semiconducting polymers
Light Emitting FET (LEFET) as the
Route to the Polymer Injection Laser
1. The data imply inverted populations:
High density of electrons in *-band
High density of holes in - band
• Reduced ‘electrode absorption losses’ by gate-induced
injection through the source and drain
• Minimized ‘charge induced absorption losses’ because charge
transport is perpendicular to the waveguide direction
2. Minimal loss from the source and drain electrodes
3. Minimal loss from ITO gate electrode
4. Electron & hole densities confined to narrow
region near the gate dielectric
Light emitting FETs
The high gate induced carrier densities imply inverted
population in the LEFET
 High density of electrons in the *–band
and
 High density of holes in the –band.
LEFET Injection Laser: Important Opportunity
Acknowledgement
Professor Kwanghee Lee ---UCSB and GIST (Korea)
Dr. Jin Young Kim
Christoph Brabec and colleagues at Konarka
James Swensen (LEFET)