GCEP Distinguished LectureOctober 2008

Third Generation Photovoltaics

Photovoltaics Centre of Excellence supported by the Australian Research Council, the Global Climate and Energy Project and Toyota CRDL

Gavin Conibeer

Deputy DirectorARC Photovoltaics Centre of Excellence

School of Photovoltaics and Renewable energy EngineeringUniversity of New South Wales

School of Photovoltaics & RE Eng.

Stuart Wenham, Martin Green+ Management Committee

LaboratoriesMark Silver

1st Generation:Wafers

2nd Generation:Thin Films

3rd Generation: PVHigh eff and thin film

Silicon PhotonicsSi light emission

Corodination of previously separately funded strands

Laboratory DevelopmentPV and Renewable Energy U/G degrees

ARC Photovoltaics Centre of Excellence

School of Photovoltaics & RE Eng.History

• PV research, UNSW Electrical Eng. 1974 – 1998• Buried contact solar cell

– Martin Green, Stuart Wenham 1986• Crystalline Si on glass spin off company 1995• Separate Centre 1999 – 2005• First UG program - Photovoltaics 2000• PG coursework program 2001• Second UG program – Renewable Energy 2003• New School formally declared 2006

UNSW International Collaborations

•DARPA & U. Delaware (50% efficient solar cell)•Global Climate & Energy Project, Stanford (1)•Global Climate & Energy Project, Stanford (2)•Toyota Central R&D Labs.•Suntech, Wuxi (NYSE)•Nanjing PV Tech, Nanjing (NASDAQ)•JA Solar, Ningjin (NASDAQ)•E-Ton Solar, Taiwan•Asia-Pacific Partnership (Australia, China, India, Japan, Korea, India, USA)

UNSW Photovoltaics R&D Commercialisation

• First Generation Photovoltaics– Buried contact cell (UNSW, BP Solar)– Inkjet printing (UNSW, Suntech, E-Ton)– Semiconductor Fingers (UNSW, Suntech, E-Ton)– Laser doping (UNSW, Suntech)

• Second Generation Photovoltaics– Crystalline Silicon on Glass (UNSW, CSG Solar)

• Characterisation Equipment– Photoluminescence characterisation

(UNSW, BT Imaging)

Undergraduate EducationTwo 4-year Engineering programs (189 students):• Photovoltaics and Solar Energy (2000) (101 students)• Renewable Energy (2003) (88 students)

UAI Distribution 2000-2005

0

10

20

30

40

50

60

70

Below 80 80-84.99 85-89.99 90-94.99 95-100

University Admission Index

# St

uden

ts

3656

3655

3657

3642

Postgraduate Education

• Master of Engineering Science in Photovoltaics and Solar Energy (14 students)– 1.5 year addition to UG;– PV devices; PV systems;

RE technologies;

• Research degrees– PhD (47 students)

– MPhil (10 students)

UNSW Centre for Energy and Environmental Markets

• Interdisciplinary research in energy and environmental markets, policy

• Faculties of Engineering and Commerce & Economics• Environmental sustainability:

– eg. PV load & pricing at Olympic Village• Economic tools & climate change:

– e.g. Market design (Aust. Stock Exchange, CSIRO)• Sustainable technology:

– e.g. Stochastic renewable energy (wind).• Aust. Greenhouse Office

Third Generation PhotovoltaicsOutline

• The importance of Photovoltaics• Three generations of Photovoltaics• The main losses in photovoltaic cells• Third Generation approaches

• Silicon nanostructure tandem cells• Band gap engineering – quantum confinement• Fabrication of materials / devices

• Hot Carrier cells• Contacts – energy filtering • Hot Carrier cooling – energy loss to phonons

• Modification of the solar spectrum• Up- and Down-conversion

• Potentialities and Viabilities• Summary

Transforming the global energy mix: The exemplary path until 2050/ 2100appointed for a term of four years by the federal cabinet (Bundeskabinett)

Meeting the IPCC target of 60% reduction in GHG emission by 2050

Booming Photovoltaics

Global PV market US$6.5 billion in 2006 → $16.4 billion in 2012

Market growth at 35%/yr for last 10 years, 60%+ in 2007Approx 1 million jobs in PV by 2020Approx 1 million jobs in RE by 2010

Driven by rebates/tariffs: Japan, Germany

Now other Euro. Countries and S Australia

USA: Power purchase agreements

Japan: market is stablewith reducing rebates

1988

1991

1994

1997

2000

2003

2006

0

1000

2000

3000

4000

5000

MW

p

USAEuropeJapanRest of WorldTotal

Learning curves

bulk-Si(~10%)

2003

US

$ /kW (~20%)

20000

10000

5000

2000

1000

500

2000.01 0.1 1.0 10.0 100.0

Gas turbines (USA)

1980

1963

1982

19871993

2001

Wind turbines

2002

Cumulative GW installed

1981

Photovoltaics

. more potential for learning

. lower cost at smaller volumes

20000

10000

5000

2000

1000

500

2000.01 0.1 1.0 10.0 100.0

2002

Cumulative GW installed

1981

Photovoltaics

bulk-Si(~10%)

2003

US

$/kW (~20%)

Thin-film PV

2nd Generation

2002

3rd Generation

1st Generation

100

80

60

20

0 100 200 300 400 500

US$0.50/W

US$1.00/W

US$3.50/W

Cost, US$/m2

US$0.10/W US$0.20/W

Present limit

Thermodynamiclimit

40

Effic

ienc

y,%

Photovoltaics: Three Generations

III

III

mc-Si

III-V tandem

c-Si

thin film

a-Si tandem

concentration

*

Efficiency Loss Mechanisms

Two major losses – 50%

Limiting efficiencies 1 sunSingle p-n junction: 31%Multiple threshold: 68.2%

qV

2. Lattice thermalisation

2

2

1. Sub bandgap lossesEnergy

3

Also: 3. Junction loss

4

44. Contact loss

5

5

5. Recombination

1

Third generation options

100%

74%68%

54%49%44%39%31%

0%

58%

circulators

tandem (n )hot carrier

impurity PV & band, up-converterstandem (n = 3)thermal, thermoPV, thermionics

impact ionisationtandem (n = 2)down-converterssingle cell

tandem (n = 6)65%

Eg

Eh

Elintermediate level

Ef

Jl

Jh

JVC

CB

VB

Erela

x

h+

e-

e-e-

h+

E0,e

E2,h

E2,e

One photonMultiple electrons-

E0,h

Intrinsic Intrinsic radiativeradiative and Auger losses includedand Auger losses included

Free choiceor Si cell

Decreasing band gap

Sunlight

42.5%47.5%

50.5%

Number of cells1 2 3

10

20

30

40

00

Si bottomcellFreechoice

29%

33%

45%

AM1.

5GEf

ficie

ncy

Silicon based Tandem CellMartin Green, Gavin Conibeer, Dirk König, Eunchel Cho, Tom Puzzer, Yidan Huang, Shujuan Huang, DengyuanSong, Angus Gentle, Ivan Perez-Wufl, Chris Flynn, Jeana Hao, Sangwook Park, Yong So, Bo Zhang

Silicon based Tandem Cell

Sola

r C

ell 1

Sola

r C

ell 2

Sola

r C

ell 3

Decreasing band gap

Sola

r C

ell 1

Sola

r C

ell 2

Sola

r C

ell 3

Decreasing band gap

Zacharias, 2000

Thin film Si cellEg = 1.1eV

2nm QD, Eg =1.7eV

Si QDs

defect or tunnel

junction

SiO2barriers

Engineer wider band gap Si QDs

Anneal 1100°C – Si precipitation

Dielectric layer

Substrate

Si, Ge or Snrich layer

SubstrateSubstrate

3.02.0 5.54.73.5

100 150 200 250 300 350

Deposition time [sec]

Diameter of Si QDs [nm]

0

0

5000

10000

15000

20000

2500

30000

Integrated PL intensity

(au)1.2

1.3

1.4

1.5

1.6

1.7

1.8

PL

ener

gy [e

V]

Si QD characterisation

XRD Si QDs in oxide dQD 4.5nm

PLoptical energy levels

Range of QD materials

1.0

1.5

2.0

2.5

3.0

3.5

0 1 2 3 4 5 6 7Dot diameter [nm]

PL e

nerg

y [e

V]

Y. Kanemitsu et alH. Takagi et alS. Takeoka et alT. Y. Kim et alT. W. Kim et al Oxide (UNSW) Nitride (UNSW)

Si QDs in oxide/nitride

c-Si

SiC

0.9 eV

1.1 eV

0.5 eV

Si3N4

c-Si

1.9 eV

1.1 eV

2.3 eV

c-Si

SiO2

3.2 eV

1.1 eV

4.7 eV

Alternative matricesGreater σ for Si3N4 but also lower Eact

1.0E-09

1.0E-07

1.0E-05

1.0E-03

1.0E-01

1.0E+01

2.00 2.50 3.00 3.50

1000/T (1/K)

Con

duct

ivity

(S/c

m)

Eσ =0.30eV

Eσ = 0.63eV

Eσ = 0.76eV

Eσ =0.36eV

SiQDs in Si3N4

SiQDs in SiO2

DFT model

-ling

Various material combinationsQuantum Dot / Matrix combinations and current status of investigations

-POSPOSn

--SPGe

SPODSPOEDSPOEDSi

SiCSi3N4SiO2

S = Simulation (ab-initio modelling - DFT)P = Physical (electron microscopy, X-ray difraction)O = Optical (photoluminescence, absorptance)E = Electronic (conductivity, conductivity with Temp.)D = Devices (Diodes, Cells)

Increasing conductivity

Decreasing processing temperature

University of New South Wales, Sydney: Gavin Conibeer, Martin Green, Dirk König, Shujuan Huang, Santosh Shrestha, Chris Flynn, Lara Treiber, Pasquale Aliberti, Andy Hsieh, Rob Patterson, Binesh Puthen Veettil, Martin Kirkengen

Institute Energie Solar, Universitas Polytechnic Madrid: A. Luque, A. Marti, E. Cánovas, A. Martí, P.G. Linares, E. Antolín, D. Fuertes Marrón, C. Tablero

Inst. Research Development Energie Photovoltaic / CNRS, Paris: Jean Francois Guillemoles, Lunmei Huang

University of Sydney: Timothy Schmidt, Raphael Clady, Murad Tayebjee

Hot Carrier solar cellStarted September 2008

Hot Carrier cellExtract hot carriers before they can thermalise:

small Eg

TH

Ef

Hot carrier distribution

• Need to slow carrier cooling

Eδ

Ef(n)

TA

qV=ΔµA

Ese- energy selective contact

h+ energy selective contact

Es

Ef(p)

TA

• Collect carriers over narrow range of energies• Renormalisation of electron (hole) energies

Ross & Nozik, JAP, 53 (1982) 3813Würfel, SOLMAT, 46 (1997) 43 1995Green, 3rd Gen PV (S-Verlag) 2003

Würfel, PIP, 13 (2005) 277Conibeer, TSF, 516(2008) 6948

Takeda et al, SOLMAT, 08

Si QD

Dielectricmatrix

ResonantTransport

Resonant Tunneling Transport

Energy

Energy Selective Contact

Filter

ECEf

Ef

I

V

0

0.01

0.02

0.03

0.04

0 0.5 1 1.5

Gate voltage (V)

Ig(A

) Two different sites on the wafer

NDR at 300K - Repeatable

Hot Carrier cooling

Hot Optical phonon population“phonon bottleneck effect”

Slows further carrier cooling

Decay of Optical phonons to Acoustic is critical

Electrons carry most energy

Cool predominantly via small wave vector optical phonon

emission - timescale of psinelastic – energy relaxation

Optical phonons emitted

Energy

Optical phonon decay

Optical phonon decay

O → LA + LA (Anharmonicity or Klemens mechanism)

Compound – e.g. InN

Element– e.g. Si

Phon

on e

nerg

ies

(den

sity

of s

tate

s) 60

meV

30

0

Allowed phonon energies

E

Nō

Optical phonons(standing waves)

Acoustic phonons(heat in the lattice)

Some evidence for slowed carrier cooling in InN: Chen & Cartwright, APL, 83 (2003) 4984

And for longer phonon lifetimes in GaN, AlSb, InP – all of which have large phonon gaps

0.0

0.5

1.0

1.5

2.0

2.5

3.0B

iN InN

SnO

GaN

AlS

b

InP

SiC

AlN BN

AlP

0%

20%

40%

60%

80%

100%

Eop

tical−

Eac

oust

icE

acou

stic

Eoptical (M

ax -Min)

Eacoustic

Phononic band-gaps for various binary compounds

Gap > Eacoustic

Gap < Eacoustic

Eg = 0.7eV

Phononic gaps in nanostructures

Linear force constant model: l = 4a1 + 4a2– mass ratio = 2; force constant ratio = 5

Nanostructure

Phon

on e

nerg

ies

(den

sity

of s

tate

s)

40

meV

20

0

Phonon propagation in nanostructure

Acoustic phonon reflected from zone edges → standing wave

1D to 3D modelling

•Coherent interference •Periodic QD array – probably fcc – probably core shell QDs•1D modelling to 3D – Lunmei Huang•Long range / short range defects – Andy Hsieh, Binesh PV

Uniform 3D periodic QD array

Colloidal dispersion of nanoparticles

•Core shell nanocrystals – hetero-interface – Guillemoles, IRDEP, Paris

•Colloidal dispersion of Si or other nanocrystals– want uniform spacing and mono-disperse size

Langmuir-Blodgett deposition of monolayers

build up multiple mono-layers

Organosilanes of varying alkyl chain lengths –a)Trimethoxy(propyl)silane, b) Trimethoxy(octyl)silane

Towards a complete cell•Fabrication of slowed cooling absorber

•Energy Selective Contacts

•Transport and Renormalisation of carrier energies

PV cell

Modification of the incident solar spectrum

Multiple exciton generation (MEG)Hanna, Nozik, JAP (2006) 100, 074510Schaller, Klimov, PRL (2004) 92, 186601

h+

e-

e-

e-

h+

E0,e

E2,h

E2,e

One photon

Multiple electrons-

E0,h

QE > 100% to be usefuli.e at least as many photons out as in

or: Re-emit two photons above Eg of celleither: Inject two e-h pairs into cell

Down conversion

PbSe QD array by colloidal dispersion

Bifacial solar cell

Reflector

Phosphors in a transparent medium

QE of a few % is usefulEr doped phosphors – Shalav et al., PL (2005) 86, 013505 Triplet - triplet annihilation - Baluschev et al., APL, 90 (2007), 181103

Up Conversion

Up-converter

Solar cell reflector

Eg

Eh

Elintermediate level

Ef

Jl

Jh

JVC

CB

VB

Erelax

Intermediate band solar cellMarti, Luque et al., PRL (2006) 97, 247701.

Potentialities and ViabilitiesRequirements:

Higher efficiencies Lower costReadily available & benign materials

Tandem cellsAlready provenProblems with reducing costNeed breakthrough in cost structure

Thermal approaches - Hot Carrier cellPotentially very high eff. Long way from proof of concept

Up – Down ConversionCan be applied to existing solar cells – big advantageBut early stages of proof of concept

Summary•Relevance and growth of Photovoltaics•Three PV Generations•Main energy losses•Third Generation approaches•Si nanostructure tandem cells

•Band gap eng.•Range of QD materials•Devices now up to 390mV VOC

•Hot Carrier cells•Energy filter contacts•Phonon bottleneck•Nanostructures - QD based cell

•Up- and Down-conversion•Third generation multi-energy level devices

•tend to involve QD nanostructures •enable tailoring of material properties

Research Staff: Martin Green, Richard Corkish, Gavin Conibeer, Dirk König, Eun-ChelCho, Tom Puzzer, Yidan Huang, Shujuan Huang, Dengyuan Song, Santosh Shrestha, Ivan Perez-Wufl, Angus Gentle, Supriya Pillai

PhD students: Chris Flynn, Jeana Hao, Sangwook Park, Lara Treiber, Yong So, Pasquale Aliberti, Yong So, Andy Hsieh, Bo Zhang, Rob Patterson, Binesh Puthen Veettil, Craig Johnson, Darryl Wang, Dawei Dai

Visiting researchers: Fei Gao, Dong-Ho Kim, Frank Koo, Ke Ma, Veronique Gevaerts, Martin Kirkengen, Martina Schmid

Third Generation Strand (2008)

Thank you for your attention