Optical modeling and simulation of thin-film Cu(In,Ga)Se2 solar cells

University of LjubljanaFaculty of Electrical Engineering

Optical Optical modelingmodeling andand simulationsimulation ofofthinthin--film film Cu(In,Ga)SeCu(In,Ga)Se22 solar cellssolar cells

J. Krc1, A. Campa1, G. Cernivec1, J. Malmström2,M. Edoff2, F. Smole1 and M. Topic1

1University of Ljubljana, Faculty of Electrical Engineering,Tržaška 25 Si-1000 Ljubljana, Slovenia

2Uppsala University, Ǻngström Solar Center,P.O. Box 534, SE-75120 Uppsala, Sweden

2


Outline

• Introduction

• Optical modeling of thin-film solar cells

• Optical simulator SunShine

• ResultsOptical simulation of thin Cu(In,Ga)Se2 solar cells

• Conclusions

3


Introduction

Electrical energy

Solar energy Solar cell (PV modules)

> 30 % growth in module production per year

4


Introductionconventionalwafer-based solar cells thin-film (TF) solar cells

• lower material consumption• low temperature processes• possibility of being flexible

cell thickness:a few 100 um

cell thickness:a few um or less

5


Introduction

Mo

glasssubstrate

CIGSabsorber

ZnOCdS

ZnO:Al

incidentlight

n-typep-type

2 um

-

+

V

J

40 nm50 nm

400 nm

300 nm

Cu(In,Ga)Se2 (CIGS) thin-film solar cell:

Transparent ConductiveOxide (TCO) front contact

pn-junction

metal contact andsubstrate

6


Introduction

Mo

glasssubstrate

CIGSabsorber

ZnOCdS

ZnO:Al

incidentlight

back contact

n-typep-type

2 um

-

+

V

J

40 nm50 nm

400 nm

300 nm

Optical features:• multilayer structure

multiple R and T!

TF solar cell as an optical system:

7


Introduction

Mo

glasssubstrate

CIGSabsorber

ZnOCdS

ZnO:Al

incidentlight

back contact

n-typep-type

2 um

-

+

V

J

40 nm50 nm

400 nm

300 nm

• multilayer structure• layers in nm range

interferences!


Optical features:

8


Introduction

• multilayer structure• layers in nm range• textured interfaces

Mo

glasssubstrate

CIGSabsorber

ZnOCdS

ZnO:Al

incidentlight

back contact

n-typep-type

2 um

-

+

V

J

40 nm50 nm

400 nm

300 nm

light scattering!


Optical features:

9


Optical modeling of TF solar cells

10


Mo

glasssubstrate

CIGSabsorber

ZnOCdS

ZnO:Al

back contact

n-typep-type

-

+

V

J

nano-textured surface of CIGS layer

Optical modeling of light scattering

11


• Specular (non-scattered) light• Scattered (diffused) light

Light scattering at nano-textured interfaces:


12


Descriptive scattering parameters:

( )R scattADF ϕ

( )T scattADF ϕ

( ) ( )T dif T T dif totI ADF Iϕ ϕ= ⋅ϕ

T dif totI

1. Haze (scattering level)

2. Angular distribution functions (directions)


13


Optical simulator SunShine

14



• 1-D semi-coherent optical model [J. Krc et al., Progress in Photovoltaics 11 (2003) 15.]

scatteredlight

specularlight

• specular light -electromagnetic waves (coherent)

• scattered light -light rays (incoherent)

15




• Complex refractive indexes and thicknesses of layers

N(λ) = n(λ) – jk(λ)

Main input parameters (structure description):

d1

N5

N1

N2

N3

N4

N6

N0

N0

d2

d3

d4

d5

d6

16






• Haze and ADF of textured interfacesand root-mean-square rougness, σrms, of interfaces

Calibrated equations of scalar scattering theory(for details refer to our NUSOD paper)

σrms, n, λ HR, HT for internal interfaces

17






• H and ADF of textured substratesand measured root-mean square rougness of interfaces

Main ourput results:

• Optical reflectance from the structure

• Absorptances in individual layers

• Charge-carrier generation-rate profile

18


Simulations

Mo

glasssubstrate

CIGSabsorber

ZnOCdS

ZnO:Al

back contact

n-typep-type

360 nm

40 nm50 nm

390 nm

300 nm

(lower material consumption, shorther deposition times)

• To analyse and optimise optical and electrical propertiesNumerical modeling&simulation an important tool!

• Thinning down CIGS absorber below 1 um

Extremely thinCIGS absorber

Simulated structure:

19


SimulationsIncluded in simulations:

• Experimentally obtained complex refractive indexes of layers[ O. Lundberg, PhD. Thesis, Uppsala University, 2003]

n k

Wavelength, λ (nm)400 600 800 1000 1200

Ref

ract

ive

inde

x, n

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

CIGS

CdS

Wavelength, λ (nm)400 600 800 1000 1200

Ext

inct

ion

coef

ficie

nt, k

1e-6

1e-5

1e-4

1e-3

1e-2

1e-1

1e+0

1e+1

CIGS

CdS

20


SimulationsIncluded in simulations:

• Experimentally obtained complex refractive indexes of layers[ O. Lundberg, PhD. Thesis, Uppsala University, 2003]

• Measured H and ADF [ J. Krc et al., Proc. of E-PVSEC, Barcelona, 2005, p. 1831.]

Example of ADFT measurement:

21


Simulation results

22


Simulation results

Wavelength, λ (nm)400 600 800 1000 1200

A, Q

E, R

tot

0.0

0.2

0.4

0.6

0.8

1.0symbols - measurementlines - simulations

QE

Rtot

ACIGS ACIGS QE

if ideal extraction ofcharge carriers fromCIGS

Total reflectance and quantum efficiency

23


Simulation results

Wavelength, λ (nm)400 600 800 1000 1200

A, Q

E, R

tot

0.0

0.2

0.4

0.6

0.8

1.0symbols - measurementlines - simulations

AZnO:Al

AMo

determination of opticallosses in the structure

Absorptances in non-active layers

24


Simulation results

for further electricalanalysis of the structure

Position, x (nm)

0 100 200 300 400

Gen

erat

ion

Rat

e P

rofil

e, G

L (cm

-3s-1

)

1e+17

1e+18

1e+19

1e+20

1e+21

1e+22

1e+23C

dS

CIG

S

λ = 450 nm (1.52 mW/cm2)

λ = 900 nm (0.68 mW/cm2)

AM 1.5 (100 mW/cm2)

Charge-carrier generation-rate profile

25


Conclusions

• good agreement in sim. and meas. total reflectance of thin CIGS solar cell

• calibration of the simulator with realistic optical parameters (refractiveindexes, scattering and others) is important

• starting point for optical optimisation and electrical analysis of the structure

26


Further workcombined optical + electrical analysis of the structure

SunShine&Aspin simulators

External characteristics and parameters of the solar cell:

Voltage, V (V)0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7

Cur

rent

den

sity

, J (m

A/c

m2 )

-24

-20

-16

-12

-8

-4

0

JSC = 21.0 mA/cm2 21.1 mA/cm2

VOC = 592 mV 595 mV

FF = 0.68 0.65Eff. = 8.1 % 8.2 %

measurement(symbols)

simulation(line)

Wavelength, λ (nm)400 600 800 1000 1200

QE

0.0

0.2

0.4

0.6

0.8

1.0symbols - measurementline - opt. + el. simulation

QE

see extended NUSOD paper submitted to OQE

27


SimulationIncluded in simulations:

• Reduced reflectance of CIGS/Mo interface by 30 %

(formation of MoSe2 interfacial layer) [ J. Krc et al., Proc. of E-PVSEC, Barcelona, 2005, p. 1831.]

Wavelength, λ (nm)400 600 800 1000 1200

Sim

ulat

ed R

efle

ctan

ce, R

0.0

0.2

0.4

0.6

0.8

1.0

as calculated for idealCIGS/Mo interface

calibrated for realisticCIGS/Mo interface

28


Simulation

Haze of internal interfaces

1 2 2...34 cos cos( )

1rms inc outn n

TH eπσ ϕ ϕ

λ⋅ −

−= −

214 cos( )1

rms incn

RH eπσ ϕ

λ⋅

−= −

cR(σrms, λ)

cT(σrms, λ)

cR, cTcalibrationfunctions

J. Krc et al. J. Appl. Phys. 92/2 (2002) 749-755.J. Krc et al. Thin Solid Films 426/1-2 (2003) 296-304

M. Zeman et. al. JAP 88 (2000)basic theory

calibration introduced

Modified equations of scalar scattering theory:

C.K. Carniglia, Optical Engineering 18/2 (1979)P. Beckmann, A. Spizzichino, Pergamon Press (1963)

initial values:cR = 1cT = 0.5

29


Simulation

Wavelength, λ, (nm)400 500 600 700 800

Haz

e, H

T

0.0

0.2

0.4

0.6

0.8

1.0

n/Ag(σr=26 nm)

SnO2/air(σr=40 nm) i/n (σr=26 nm)

p/i(σr=40 nm)

SnO2/p(σr=40 nm)

Wavelength, λ, (nm)400 500 600 700 800

Haz

e, H

R

0.0

0.2

0.4

0.6

0.8

1.0

n/Ag(σr=26 nm)

SnO2/p(σr=40 nm)

i/n (σr=26 nm)

air/Ag(σr=37 nm)

p/i(σr=40 nm)

HR HT

Application of the calibrated theory to internal interfaces

30


Optical modeling and simulation of thin-film Cu(In,Ga)Se2 solar cells

Documents

Transcript of Optical modeling and simulation of thin-film Cu(In,Ga)Se2 solar cells