5-12-2008

Polymer solar cells

why status and challenges
why,

René Janssen

DPI Annual Meeting, November 25-26, 2008

Global warming in the last millennium

Global energy power demands:

2008
6.5 Billion people use 12 TW

2050
8-10 Billion people use 20 TW

Source: Marty Hoffert, New York University

1
 5-12-2008

Only solar energy can provide 20 TW in 2050

Solar
100000 TW at Earth surface
10000 TW (technical value)
Wind
14 TW
Biomass
5-7 TW
Tide/Ocean
Currents
0.7 TW
Hydroelectric
1.2 TW technically feasible
Geothermal 0.6 TW installed capacity
1.9 TW

Apart from nuclear power, all sources of energy

used nowadays (coal, oil, natural gas, biomass,
water, wind) ultimately originate from the sun.

Source: Arthur Nozik, NREL Courtesy: Christoph Brabec, Konarka

What are the costs for producing solar energy?

2
 5-12-2008

Photovoltaic solar cell efficiencies & costs

40 38
crystalline materials € € € € € €
35
thin film materials € € €

30 organics €

25 24
increasing

20 19 19

15 13
10
10
5
5

0
Status of 2005 3J sc-Si mc-Si CIS a-Si dye polymer

How can we reach 20 TW PV-energy in 2050?

Required power: 20 TW
Sunlight: 1000 W/m2 12 h a day
Efficiency: 10%
Required area: 400,000 km2 = 20×1012/(1000 × 10% × 12/24)
Compare: 500,000 km2 = Spain
How can we reach this area in ~30 years?
Required 36.5 km2 solar cells a day

this is a strip of 365 km by 100 m every day for 30 years

The (only?) solution : plastic solar cells

Roll-to-roll
R ll t ll production
d ti
Lightweight
Low cost

Source: Konarka

3
 5-12-2008

Photosynthesis : or how coal, oil, gas

and biomass are made today

N Mg N
O
N
O O

O
O

Bulk-heterojunction solar cells

light

electron
glass
transfer
transparent electrode
-

absorption - 100 nm
+

metal electrode

donor acceptor nanoscopic mixing of donor and acceptor to

overcome ~10 nm exciton diffusion length

R. H. Friend et al., Nature 1995, 376, 498

A. J. Heeger et al., Science 1995, 270, 1789

4
 5-12-2008

Sub-picosecond hole and electron transfer

Pump 510 nm
O

)n h+ OMe
O
( PCBM
MDMO-PPV O
e-

Pump 670 nm

-0.2
-0.8

-0.1 510 nm
-0.4
670 nm
0.0 0.0
-ΔT//T

Probe 970 nm
0.4 970 nm 0.1 ΔT/T

0.8
0.2
488 nm
1.2 630 nm 0.3

0.5 1.0 1.5 2.0 2.5 -5 0 5 10 15

Energy (eV) Time delay (ps)

Paul van Hal, Appl. Phys. A. 2004 , 79, 41.

Polymer solar cells contain two components:

electron donor and electron acceptor

)n
metal
( interface layer
O MDMO-PPV
semiconducting polymer

transparent
conductive polymer

transparent
conductive oxide

OMe
glass
O

PCBM

Sean Shaheen et al., Appl. Phys. Lett. 2001, 78, 841

Jeroen van Duren, Adv. Funct. Mater. 2002, 12, 665.

5
 5-12-2008

2.5% efficient organic plastic solar cells

VOC = 0.87 V
0
O

)n OMe
O
( -2

J (mA/cm )
2
O FF = 0.60

MDMO-PPV PCBM -4

0.6 JSC = 4.9 mA/cm2

-6
0.5

0.4 0.0 0.2 0.4 0.6 0.8 1.0

V lt
Voltage (V)
0.3
EQE

0.2 (V × I )max
FF = ≤1
0.1
Voc × Isc

0.0 Sean Shaheen et al., Appl. Phys. Lett. 2001, 78, 841
400 500 600 700 800
Martijn Wienk (ECN) and Jeroen van Duren (TU/e)
Wavelength [nm]

Importance of processing and morphology

)n
OMe
O (
O

Toluene η < 1% Chlorobenzene η > 2.5 %

5μm 5μm

200 nm 200 nm

see also: T. Martens et al., Synth. Met. 2003, 138, 243 Joachim Loos and Xiaoniu Yang (TU/e)

6
 5-12-2008

Phase separation at higher temperatures

0 min @ 130°C Δh = 10 nm 10 min @ 130°C Δh = 6 nm

37 min @ 130°C Δh = 170 nm 20 min @ 130°C Δh = 32 nm

More efficient P3HT:PCBM solar cells

C6H13

S n

O
Me
O

Isc = 8.7 mA/cm2

Voc = 580 mV
FF = 0.55
Pavel Schilinsky, Appl. Phys. Lett. 2002, 81, 3885 ηe = 2.8%.
Franz Padinger, Adv. Funct. Mater 2003, 13, 85

7
 5-12-2008

Annealing of P3HT:PCBM solar cells

annealing at 120 °C for 60 min on complete devices

regioregular P3HT Voc = 0.615 V

Mw = 100000 g mol-1 FF = 0.61
Mw/Mn = 2.14 Jsc = 7.2 mA cm-2
ηe = 2.7 %
AM1.5 1000 W/m2
Xiaoniu Yang, TU/e

TEM

after spin coating

from chlorobenzene

P3HT
0.39 nm*
PCBM
0.46 nm

after annealing
at 110 °C for
60 min

P3HT
0.39 nm*
PCBM
0.46 nm
* P3HT whiskers: K.J. Ihn, J. Moulton, P. Smith, J. Polym. Sci. Part B: Polym. Phys. 1993, 31, 735.

8
 5-12-2008

What makes an solar cell efficient?

Absorption efficiency

Or how many photons are absorbed?

Quantum efficiency

Or how many photons are converted into electrons?

Energy efficiency

Or what is the final (chemical) potential of the electrons

generated?

Shockley-Queisser limit: 31% efficiency for a single junction cell

Losses in P3HT:PCBM cells

O
Me
photon energy > 1.9 eV !!!!!
O

FF = 0.65

S n Voc = 0.62 V
PCBM P3HT

6 0.20

14.3% 4.0%
n (x 10 )
-18

0.15
s Out

4
Out
Power In

Power O
Photons In

Electrons

0.10

2
0.05

0 0.00
500 1000 1500 2000 2500 3000 3500 500 1000 1500 2000 2500 3000 3500
Wavenlength (nm) Wavelength (nm)

9
 5-12-2008

Voltage loss: 1.9 → 0.6 eV

P3HT PCBM

33
3.3
driving force for charge separation = 0.9 eV

1.9 eV 4.2
HOMOD-LUMOA = 1.0 eV
5.2 loss at electrodes
2 x ~0.2 = 0.4 V

Voc ≈ 0.62 V

J. Halls et al., Phys Rev B. 1999, 60, 5721

V. Mihailetchi et al., J. Appl. Phys. 2003, 94, 6849
M. C. Scharber et al., Adv. Mater. 2006, 18, 789

Theoretical efficiencies

Eg = α+β
20
Voc ≈ β - 0.4 V
Theroretical Efficiency ((%)

α = 0.2 eV
15 LUMO
α LUMO

10
HOMO β

5 α = 0.9 eV HOMO

0 Donor Acceptor
2.0 1.5 1.0
Band gap energy α + β (eV)
Goal: α+β ≈ 1.4 eV
α small
assuming: FF = 0.7, EQE = 0.9 and
0.2 V loss at each electrode
Martijn Wienk (TU/e)

10
 5-12-2008

PFTBT:PCBM (80 wt. %) solar cells: 4.2%

S
N N LiF / Al
C10 S S
C10 n
Active layer
smu

OMe
PFTBT
PEDOT:PSS
O
ITO
Glass
PCBM

75 % of the absorbed
photons generate current

Voc = 1.0
10V
FF = 0.54
Jsc = 7.7 mA/cm2
ηe = 4.2 %
AM1.5 1000 W/m2

Lenneke Slooff, Appl. Phys. Lett. 2007, 90, 143506

O. Inganäs & M. R. Andersson, Adv. Mater. 2003, 15, 988

PBBTDPP2:PCBM solar cells: 4%

Improved solar cell performance PBBTDPP2 :PCBM (1:2)

0.5

0.4

S N O
0.3 C12H25
EQE

H25C12 S
0.2 O n
N S

0.1

0.0 PBBTDPP2
400 500 600 700 800 900
wavelength (nm)

Ratio 1:2 Jsc (mA/cm2) FF Voc (V) η (%)

PBBTDPP2:[60]PCBM 9.4 0.54 0.63 3.2

PBBTDPP2:[70]PCBM 11.3 0.58 0.61 4.0

M. Wienk, J. Gilot, M. Turbiez, Adv. Mater. 2008, 20, 2556

11
 5-12-2008

PCDTPBT:PCBM solar cells: ~4%

0.5

0.4

0.3
EQE

n
S S
0.2
N N
S
0.1

Eg = 1.40 eV
0.0
400 500 600 700 800 900 1000
Wavelength

Ratio 1:2 Jsc (mA/cm2) FF Voc (V) η (%)

PCPDTBT:[60]PCBM 7.2 0.42 0.69 2.4

PCPDTBT:[70]PCBM 12.7 0.49 0.61 3.8

Munazza Shahid, Jan Gilot (TU/e)

Muhlbacher, Adv. Mater. 2006, 18, 2884. Peet, Nature. Mater. 2007, 6, 497.

Where is the limit?

CT
X
S1 S1 S1
ΔGCT PET
CT
PET
ΔGCRT CRT
Energy [eV]

T1 T1 T1
X
CT
film
Δ
Eg,OPT
film
ETfilm ECT
OPT OPT
EHOMO(D) - ELUMO(A)

0
I IIa IIb

S1 and T1 represent the lowest singlet and triplet states in the D – A combination

12
 5-12-2008

Results from 18 D-A blends in PET

OPT OPT
ΔGCT = ECT − Eg = | EHOMO (D) − ELUMO ( A ) | +0.29 − Eg

Only -0.1 eV
is enough to
make PET

from optical photoinduced absorption, fluorescence, and cell performance

Effective levels scale with ECT and Voc

Implication: There is a loss of 0.5 eV going from ECT to Voc

2 2
[eV]

VOC [V]

1 1
ECT,max
PL

0 0

0 1 2
OPT OPT
| EHOMO(D) - ELUMO(A) | [eV]

OPT OPT
ECT = | EHOMO (D) − ELUMO ( A ) | +0.29 eV

OPT OPT
eVoc = | EHOMO (D) − ELUMO ( A ) | −0.18 eV

13
 5-12-2008

Minimal energy losses?

-ΔGCRT < 0.1 eV prevents CRT

-ΔGCT > 0.1 eV enables PET

film
Eg,OPT
Energy [eV]

film
ECT
film
ET
Δ = 0.3
> 0.6 OPT OPT
> 0.4 |EHOMO(D)-ELUMO(A)|
0.2
eVOC

0 S0

Conclusion: There will be at least 0.6 eV loss from Eg to VOC

Dirk Veldman (TU/e)

How is this in good cells?

Polymer bulk-heterojunctions Eg Voc Eg-Voc η

PCPDTBT : [70]PCBM 1.38 0.62 0.76 5.5
PSiF-DBT : [60]PCBM 1.70 0.90 0.80 5.4
P3HT : [60]PCBM 1.70 0.61 1.09 4.4
PF10TBT : [60]PCBM 1.70 1.00 0.70 4.2
PBBTDPP2 : [70]PCBM 1.43 0.61 0.82 4.0
MDMO-PPV : [70]PCBM 1.70 0.77 0.93 3.0
Small molecule heterojunctions
CuPc : C60 1.55 0.54 1.01 5.0
DCV5T / C60 1.77 0.79 0.79 3.4

Dirk Veldman (TU/e)

14
 5-12-2008

Refinement
Wavelength [nm]

1000

1200
1400
1600
600

800
PF10TBT:[60]PCBM
(0.70 eV, z, {),

10 PCPDTBT:[70]PCBM
etical) Efficiency [%]

0.6 V (0.76 eV, T, V),

PBBTDPP2:[70]PCBM
(0.80 eV, S, U)
5 P3HT:[60]PCBM
(1.09 eV, „, …).
(Theore

EQE = 65%
FF = 0.65 So we may hope
0 for ~11%
2.0 1.5 1.0
Optical band gap energy [eV]

Charge generation in polymer solar cells

Efficiencies of polymer solar cells are increasing to ~5% recently.

Do we really understand the fundamental processes?

15
 5-12-2008

Charge separation only partly understood?

≤ 10 % 80 %

- - -
+ + +

Scientific question:

What are the requirements for full charge separation art very low
electrical fields?

Multi-junction polymer solar cells

wide band low band
gap BHJ metal gap BHJ

e-

e-
h+

h+
n p
recombination junction

30% for a single band gap cell

AM1.5 1000 W/m2 42% for a tandem cell
49% for a triple junction device
68% for an infinite stack

16
 5-12-2008

Aim: create transparent electron and hole

transporting layers to recombine charges

Back cell

Hole transporting layer

e- Recombination layer Electron transporting layer
h+
I,V Front cell
e-

h+

ZnO in
acetone
20 nm.
Light

Solution-processed double junction solar cell

-
1
0.75 V 0.82 V
cm )
2

0
Current Density (mA/c

+ 1.53 V
LiF/Al -1
P3HT:PCBM
PEDOT
-2
ZnO
MDMO-PPV:PCBM front cell
PEDOT:PSS -3 back cell
ITO tandem
Glass -4

0.0 0.5 1.0 1.5 2.0

Voc = 1.53 V Voltage (V)
Isc = 3.0 mA/cm2
FF = 0.40
MPP = 1.8 mW/cm2

Jan Gilot, Appl. Phys. Lett. 2007, 90, 143512

17
 5-12-2008

Solution-processed triple junction solar cell

1
0.75 V
2 x 0.82 V

Current Density (mA/ccm )

2
- 0

-1
2.19 V
+
LiF/Al -2
P3HT:PCBM
PEDOT
ZnO -3 front cell
MDMO-PPV:PCBM back cell
PEDOT -4 triple junction
ZnO
MDMO PPV PCBM
MDMO-PPV:PCBM 00
0.0 05
0.5 10
1.0 15
1.5 20
2.0 25
2.5
PEDOT:PSS
ITO
Voltage (V)
Glass Voc = 2.19 V
Isc = 2.6 mA/cm2
8 solution processed layers on top of each other FF = 0.37
processing time ~10 min. MPP = 2.1 mW/cm2

Jan Gilot, Appl. Phys. Lett. 2007, 90, 143512

Large stacks: six-fold junction -

17 solution processed layers on top of each other LiF/Al +

P3HT:PCBM
1 PEDOT
ZnO
Current Density (mA//cm )

MDMO-PPV:PCBM
2

0 PEDOT
ZnO
MDMO-PPV:PCBM
-1 PEDOT
ZnO
Single junction MDMO-PPV:PCBM
-2 PEDOT
Double junction
Triple junction ZnO
MDMO-PPV:PCBM
Six fold junction
-3 PEDOT
ZnO
0 1 2 3 4 MDMO-PPV:PCBM
MDMO PPV:PCBM
PEDOT:PSS
Voltage (V) ITO
Glass
Voc = 0.75 V Voc = 1.53 V Voc = 2.19 V Voc = 3.46 V
Isc = 3.5 mA/cm2 Isc = 3.0 mA/cm2 Isc = 2.6 mA/cm2 Isc = 1.6 mA/cm2
FF = 0.48 FF = 0.40 FF = 0.37 FF = 0.32
MPP = 1.3 mW/cm2 MPP = 1.8 mW/cm2 MPP = 2.1 mW/cm2 MPP = 1.7 mW/cm2

Single Junction Double Junction Triple Junction Six-fold Junction

18
 5-12-2008

Tandem solar cell using different band gaps

-

S N O
+ S C12H25
LiF/Al
PBBTDPP2:PCBM C12H25 S
O N S n
n PEDOT
ZnO
PF10TBT:PCBM PBBTDPP2
PEDOT:PSS
ITO
Glass
O
Me
O
80
PFTBT:PC60BM [60]PCBM
70 PBBTDPP2:PC60BM
60
Absorption (%)

50
40 PF10TBT
30
S
20 N N
C10H21 S S
10 C10H21 n

0
400 500 600 700 800 900 1000
Wavelength (nm) J. Gilot (TU/e)

Preliminary performance
Preliminary results from solar simulator

-
1

0
A/cm )
2

-1
Current Density (mA

+
LiF/Al
-2
PBBTDPP2:PCBM 78 nm
n PEDOT -3
ZnO
-4
PF10TBT:PCBM 155 nm
PEDOT:PSS -5
ITO
-6
Glass 0.0 0.5 1.0 1.5 2.0
Voltage (V)

TU/e Solar simulator (ECN)

JSC (SR) = 5.3 mA/cm2 JSC (SS) = 5.0 mA/cm2
VOC = 1.57 V VOC = 1.54 V one
FF = 0.47 FF = 0.46 month
η est = 3.9 % η = 3.6 % later

J. Gilot (TU/e) after UV illumination

19
 5-12-2008

To remember and think about

Renewable energy is a must for our future society

Polymer solar cells might be an option to contribute to that goal

because of speed of production at low cost

Efficiencies of about 10% seem within reach but are a challenging goal

New cell designs and concepts may increase efficiencies to ~15%

Acknowledgement

TU/e ECN
Dirk Veldman Sjoerd Veenstra
Jan Gilot Jan Kroon
Stefan Meskers
Martijn Wienk RUG
Arjan Zoombelt Paul Blom
Munazza Shahid Bert de Boer
Shahid Ashraf
Johan Bijleveld TNO
Marc Koetse
Joachim Loos Jorgen Sweelssen
Svetlana van Bavel

20