Transcript Organic Solar Cells - Materials Science and Engineering

Utilizing Carbon
Nanotubes to Improve
Efficiency of Organic
Solar Cells
ENMA 490 Spring 2006
Motivation

Problem: Lack of power in remote
locations

Possible solution: Organic solar cells are
less expensive and more portable than
conventional solar cells

Main issue: Inadequate efficiency
What We Did
Focus: Increase the efficiency through the
addition of carbon nanotubes
 Research Goal: Model a basic device and
propose an ideal structure for more
efficient power generation
 Experimental Goal: Build selected devices
to test parameters

Project Organization
Research Team
Erik Lowery
Nathan Fierro
Adam Haughton
Richard Elkins
Experimental Team
Erin Flanagan
Scott Wilson
Matt Stair
Michael Kasser
How Organic Solar Cells Work
1. Photon absorption, excitons are
created
2. Excitons diffusion to an
interface
3. Charge separation due to
electric fields at the interface.
4. Separated charges travel to the
electrodes.
High Work Function Electrode
Donor Material
Acceptor Material
Low Work Function
Electrode
E
Critical Design Issues

Exciton creation via photon absorption
 Material

absorption characteristics
Exciton diffusion to junction
 Interfaces
within exciton diffusion length
(nanoscale structure)

Charge separation
 Donor/Acceptor

band alignment
Transport of charge to electrodes
 High
charge mobility
The Active Layer
Composed of an electron donor and
electron acceptor
 3 types of junctions

 Bilayer
 Diffuse
Bilayer
 Bulk heterojunction

Usually the excitons from the electron
donor are responsible for the photocurrent
Electron Acceptor
0.7
MEH-PPV-CN
Electron acceptor




Electrical Properties


CN group
Increased band
alignment
Higher electron affinity
Poor charge mobility
Optical Properties


Peak emission at 558
nm
Peak absorption at 405
nm (~3eV)
1350
MEH-PPV-CN
0.5
1150
0.4
950
0.3
750
550
0.2
350
0.1
150
0
0
1
2
3
-0.1
4
5
6
-50
-250
Energy, eV
Irradiance (W/m^2)

1550
Solar Spectrum
0.6
Absorption (arb. Units)

1750
Electron Donor

Carbon Nanotubes
 Orders
of magnitude
better conductance
than polymers
 Our nanotubes
specifications (Zyvex)




Functionalized
Diameter: 5-15 nm
Length: 0.5-5 microns
MWNT (60% metallic
40% semiconducting)
AFM Amplitude Scan
Electron Donor (cont.)

Carbon Nanotubes
 Optical



Properties
Diameter
SW vs. MW
Chirality (Semiconducting vs metallic)
Modeling
Model Geometry
 Photogeneration of Excitons
 Exciton Transport to Junction
 Electron Hole Separation
 Charge Transport to Electrode

Model Geometry
Incoming Light
X=0
X=L
ITO
CNT
MEH-PPV-CN
Define A to be the area perpendicular to the incoming
light.
Al
Photogeneration of Excitons
Photogeneration of Excitons

Start with Beer-Lambert absorption equation:
S ( x,  )  S Inc ( )e  (  ) x
x
I ( x,  )  
0
S Inc ( )  (  )
e
d
hc
2 x
S Inc ( )  (  )
I ( x)   
e
dd
hc
1 0


Arrive at expression for # Photons absorbed per unit area, per unit
time
Use either blackbody approximation or numerical data for the
solar spectrum (Sinc)
Exciton Transport to Junction

Diffusion Model
du ( x, t )
d 2u ( x, t )
D
 R  u ( x, t )  A  I ( x )
2
dt
dx
Diffusion Term

Decay Term, simple
time-dependent
model
Source Term,
accounts for exciton
generation
Initial and Boundary Conditions
u (0, t )  0
Excitons destroyed at CNT/Electrode Interface
u ( L, t )  0
u ( x , 0)  0
Excitons destroyed at CNT/Polymer Junction
Initially, assume ground state, no excitons
anywhere.
Charge Transport to Electrode

Holes move along CNTs
 Hole

Mobility ~ 3000 cm2/Vs
Electrons move along MEH-PPV-CN
 Electron

Mobility ~ 3.3x10-7 cm2/Vs
Current density is directly related to
mobility; Increased mobility leads to higher
current densities.
Modeling Summary
CNT/MEH-PPV junctions within diffusion
length of exciton generation points
 Thickness Optimization Problem:

 Maximizing
thickness gives more excitons
 Minimizing thickness leads to higher current
Ideal Structure
Nanotubes
ITO
Nanoscale
mixing
MEH-PPV-CN
Al
Nanoscale mixing allows excitons to charge separate before
they recombine
Structure allows for the bulk heterojunction and minimizes
the travel distance to the electrodes
Experimental Design
 Experimental
CNT
design parameters
concentration
Method of mixing
Spin Parameters
Solvents
Device Process Flow
ITO
.4 mm
.7 mm
.2 mm
2.5 mm
Device Process Flow
PEDOT ~100nm
Al contacts ~600 Å
Active Layer
Device Process Flow
LiF ~ 20 Å
Al contacts
Final Product
Nanotube
Experimental Results
Pure Polymer #3 IV Curves
0.01
Device 1
0.008
Device 2
Device 3
0.006
A
Device 4
0.004
0.002
0
-15
-10
-5
0
5
10
15
-0.002
V
Pure polymer devices acted like diodes. Light emission
was observed at higher currents (8 mA)
Experimental Results
0.005
0.004
0.003
0.002
Device 4 dark
0.001
A
Lineary =(Device
dark)
0.0038x - 4
1E-06
0
-1.5
-1
-0.5
0
0.5
1
-0.001
-0.002
-0.003
-0.004
-0.005
V
Pure CNT acted like a resistor, R >350Ω.
1.5
Experimental Design Issues We
Addressed

Nanotube Processing

Method of dispersion
 Type of solvent

Concentration CNT
 amount
of CNT in solvent
 CNT to Polymer

Diffused junction vs. bulk heterojunction
Results Summary
Absorption spectra measured
 AFM to check spatial distribution of
nanotubes
 No successful devices made
 Possible causes:

 CNT
shorting
 Functionalized CNTs might be a problem
Conclusions

Experimental:
 Device
process recipe needs to be refined
 Solve experimental design problems to work on
successful device

Modeling:
 Diffusion
model considerations point towards
improving efficiency by creating nanoscale structure
 Need to consider charge transport in more detail
Acknowledgements

We would like to thank the following
people/organizations:
 Dr.
Gary Rubloff
 Dr. Danilo Romero
 Laboratory for Physical Sciences
 Zyvex