HAWAII NATURAL ENERGY INSTITUTEwww.hnei.hawaii.eduDOE Hydrogen Program Review 2007, Arlington

Photoelectrochemical Hydrogen Production: UNLV-SHGR Program Subtask

This presentation does not contain any proprietary or confidential information

Robert PerretUniversity of Nevada Las Vegas

Research Foundation

Eric L. MillerUniversity of Hawaii at Manoa

15 May 2007 #PDP37

2

• Project start date: 1 Oct. 2004• Project end date: 31 Dec. 2007• Percent complete: 70%

• Barriers for photoelectrochemical(PEC) H2 production technologies:

–Y: Materials Efficiency–Z: Materials Durability–AB: Bulk Materials Synthesis–AC: Device Configuration Designs

• Total project funding: $2.81M–DOE share: $1,685,861–Contractor share: $1,123,361

• FY06 Funding : $ 958k• Funding for FY07: $0

Timeline

Budget

Barriers

• University of Hawaii at Manoa / Eric L. Miller• University of Nevada, Las Vegas / Clemens Heske• University of California, Santa Barbara /

Eric McFarland• National Renewable Energy Laboratory /

Mowafak Al-Jassim & John Turner• MVSystems Incorporated / Arun Madan• Intematix Corporation / Xiaodong Xiang• Altair Nanotechnologies Incorporated /

Vesco Manev

Collaborators / PIs

OVERVIEW

3

The primary objective is to assist DOE in the development of hydrogen-production technology utilizing solar energy to photoelectrochemically split water. The primary focus is on low-cost thin film materials (such as metal oxides) and novel multi-junction thin film devices (such as the UH-Hybrid Photoelectrode- HPE)

OBJECTIVES

Identify and develop new PEC film materials compatible with high-efficiency, low-cost H2 production devices: Target: 1.6 mA/cm2 – 6.5 mA/cm2 AM 1.5 photocurrentDemonstrate functional multi-junction device incorporating best-available PEC film materials: Target: 2 - 8 % STH efficiency under AM 1.5 illuminationDevelop avenues, integrating new theoretical, synthesis and analytical techniques, for optimizing future PEC materials and devicesExplore avenues toward manufacture scaled devices and systems

catalystFoil Substrate

Solid-state-junctionConductive Interface (e.g., ITO)

Photoactive PEC Film

light

HPE Deviceelectrolyte

DOE PEC Program Targets

Specific UNLV-SHRG PEC Project Goals

4 General Approach PhilosophyThe main approach of our collaborative network of world-leaders in materials R&D focuses on integrating state-of-the-art theoretical, synthesis and analytical techniques to identify and develop the most promising materials classes to meet the PEC challenges in efficiency, stability and cost.

THEORY: Materials & Interface Modeling–Theoretical Calculations of Semiconductor Band Structures

SYNTHESIS: Materials Discovery / Development–Physical and Chemical Vapor Deposition –Combinatorial & Manufacture Scale Synthesis Techniques

ANALYSIS: Materials & Device Characterization–Physical/Solid-State Electronic/Optoelectronic Properties–Solid-Solid & Solid-Liquid Interface Characteristics –Photoelectrochemical Behavior Analysis

THEORYTHEORY

SYNTHESIS

SYNTHESISANAL

YSIS

ANAL

YSIS PECPEC

R&D Feedback Loop:

Application to Focus Materials:

tu

●W-compounds ●Zn-compounds ●Fe-compounds ●Si-compounds ●Cu-chalcopyrites

5 SUMMARY

Collaborative Research Team Established: combining materials theory, synthesis and characterizations, to expedite the materials discovery and development process for improved PEC devices.Critical Experience, Protocols, and Infrastructure Developed:

–Versatile synthesis tools established for fabricating PEC materials & devices–Comprehensive characterization protocols established for PEC materials & devices– Rapid-throughput synthesis & screening techniques developed to facilitate discovery– Manufacture scale process demonstrated for HPE device fabrication

Focus Materials Classes Established: including WO3-, ZnO- Fe2O3-, silicon- , and copper chalcopyrite-based thin films.Key Targets Met in Focus Materials Experiments:

–Photocurrents in excess of 3 mA/cm2 in tungsten-based compound films–Photocurrents in excess of 6.5 mA/cm2 in Si- and chalcopyrite-based films (with additional bias

constraints to be corrected in band edge alignment modifications)–STH Device efficiencies in excess of 3% in WO3-based multijunction structures under 1 sun

FUTUREContinued Development of Research Team Capabilities and CollaborationContinued Optimization of Performance and Durability in Focus MaterialsContinued Discovery of New Materials and Possible Down-Selection of Old MaterialsSelection of Best Materials for Incorporation in High-Efficiency PEC Devices

PROGRESS

6 Approach: Theory

Background of Density-Functional Theory (DFT):

K-S equation)()()()(8

22

2

rrrVrm

hj

KSjjeffj ψεψψ

π=+∇−

)()()()( rVrVrVrV XCNHeff ++=

:)(rVXC Exchange-correlation potentials: LDA, or GGA

Applications:Total energies, Electronic structures such as band structure anddensity of states, Optical properties such as optical absorptioncoefficient and transition properties

To provide theoretical guidance and understanding on materials studied by the PEC teamTo search for new materials and new concepts

Theoretical modeling of PEC materials:

7 Approach: Theory

Collaboration on WO3with the PEC team:

Results on ZnO:N-incorporation leads to bandgap reduction in ZnOGa-N cluster doping may form intermediate band in ZnOGroup-IB elements (Cu, Ag, Au) lead to VBM up shift in ZnOp-type and gap reduced ZnO is possible by Cu and Ag incorporationGa-N cluster doping may form intermediate band in ZnO

Specific Applications in PEC Materials

Determined conduction and valence band positions for cation and anion doped WO3 –Nitrogen incorporation leads to bandgap reduction in WO3–Calculated optical absorption coefficient for WO3:N–Calculated a minimal shift for the conduction band minimum with molybdenum doping.

New Materials:Theoretical results from WO3 and ZnO to be included in modeling of new PEC

material systems

8 Approach: SynthesisCollectively, our team commands a broad portfolio of thin film synthesis techniques to facilitate the discovery and development of PEC materials and devices for high-performance, low-cost hydrogen production systems, including. Team synthesis capabilities include:

Physical Vapor Deposition SystemsChemical Vapor Deposition SystemsSpray Pyrolysis Fabrication SystemsSol-Gel Fabrication Systems

Reactive sputtering system for compound material films (including oxides, sulfides, etc.)

Co-Evaporation system for copper chalcopyrite films (CIGS, CGS, etc.)

Plasma-enhanced CVD system for low temperature synthesis of metal oxide films

9 Approach: SynthesisThe team also offers a range of advanced techniques to facilitate rapid discovery of new materials classes and the establishment of large-scale device fabrication. Specific combinatorial synthesis and manufacture-scale equipment includes:

Combinatorial Synthesis Systems–Automated Pyrolysis systems–Automated Physical-Vapor-Deposition systems

Manufacture-Scale Film Technology–Large-scale vacuum-system cluster tools–Cassette systems for large sheet depositions

Automated combinatorial pyrolysis system

Automated combinatorial physical-vapor-deposition system

Cluster tool for vacuum-deposition of manufacture scale thin-film devices

10 Approach: Characterizations

Morphology: Scanning Electron Microscopy, Atomic Force Microscopy, Spectroscopic Ellipsometry

Microstructure: X-Ray Diffractometry,EBSD, Transmission Electron Microscopy

Chemistry: Secondary Ion Mass Spectrometry,X-ray Photoelectron Spectroscopy, SynchrotronX-ray Emission and Absorption Spectroscopy

AFM of WO3 Films

←Pure WO3

N2-doped WO3→

Materials Properties

11 Approach: CharacterizationsElectronic Structure

Electronic Surface Bandgap, Band Edges, Band Alignment, Fermi Energy, Work Function, Electrical Properties

UV Photoelectron Spectroscopy,Inverse Photoemission

Impedance Spectroscopy, UV-Vis Spectroscopy

Conductive Atomic Force Microscopy, Scanning Kelvin Probe Microscopy, Scanning Tunneling Spectroscopy

Scanning ProbeMicroscope

IPES

XPS, UPS,Auger

Glovebox

-14 -12 -10 -8 -6 -4 -2 0 2 4 6 8

WO3

0.42-2.85

-2.94

0.38-2.86

0.40

0.37

3.0 mTorr

1.0 mTorr

0.5 mTorr

0.25 mTorr

0 mTorr

IPES

Nor

mal

ized

Inte

nsity

Energy rel. EF (eV)

UPS He II

EF

p(N2)

-2.91

-2.91

0.40

12 Approach: CharacterizationsOptical and Photoelectrochemical Properties

Optical-Photoelectrochemical Combinatorial Screening Systems

Solar Cell I-V Curve Testing System, Solar Cell Spectral Response Measurement SystemPhotoluminescence and Cathodoluminescence

Diffuse Reflectance Spectrometry

13

Our team utilizes its collective expertise and capabilities in theoretical materials modeling, in materials synthesis, and in materials screening and characterization to identify and develop an expansive set of promising PEC thin-film materials classes. The five classes of “focus materials” currently under investigation include:

Tungsten-Based Compound Films (U. Hawaii, Intematix)–Modified Tungsten Oxide Compounds with Anion/Cation Substitutions

Zinc-Based Compound Films (NREL)–Modified Zinc Oxide Compounds with Anion/Cation Substitutions

Iron-Based Compound Films (UCSB, Altair Nano)–Novel Iron-Based Compound Materials, including Fe2O3 Nanorods

Silicon-Based Compound Films (MVSystems)–Amorphous Silicon Carbide Films with p- and n- type Doping

Copper Chalcopyrite Compound Films (U. Hawaii)–Copper-Indium-Gallium-Selenium-Sulfur Compounds

Specific Approach: Focus Materials

14 Progress: Tungsten-Based Compound Films

Hybrid Photoelectrode with pure WO3 films

0

2

4

6

8

10

0.0 0.5 1.0 1.5 2.00

2

4

6

8

10

12

Integrated HPE performance level

a-Si/a-Si tandem, filtered by WO3 film

a-Si/a-Si tandemstainless steel substrate

High performance WO3 film

Cur

rent

(mA

/cm

2 )

Voltage (V)

STH

efficiency (%)

0

2

4

6

8

10

0.0 0.5 1.0 1.5 2.00

2

4

6

8

10

12

Integrated HPE performance level

a-Si/a-Si tandem, filtered by WO3 film

a-Si/a-Si tandemstainless steel substrate

High performance WO3 film

Cur

rent

(mA

/cm

2 )

Voltage (V)

STH

efficiency (%)

Glass substrate

Glass substrate

Non-Conductive Epoxy

WO3

RTV SiliconeITOa-Si

(nip-nip) TCO

TCO

Glass substrate

Glass substrate

Non-Conductive Epoxy

WO3

RTV SiliconeITOa-Si

(nip-nip) TCO

TCO

00.0

0.5

1.0

1.5

2.0

2.5

3.0

20 40 60 80 100C

urre

nt (m

A/c

m2 )

Load Line Analysis Device Configuration time (min)

Performance

Tungsten oxide is a model material to study PEC hydrogen generation…

Sufficient absorption to generatemodest photocurrents

Good electron transport properties

High Stability in Electrolytes

Thin film process scalable

Demonstrated in prototype multi-junction devices.

promiseNon ideal band edge alignment –requires supplemental bias

Bandgap requires reduction to increase photocurrents

The photostability over extended time periods and for new tungsten-alloy compositions requires validation

challenges

15 Progress: Tungsten-Based Compound Films

1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.00

50

100

150

200

250

0 mTorr N2 2 mTorr N2 3 mTorr N2 4 mTorr N2 6 mTOrr N2

hν (eV)

(αhν

)1/2 (c

m-1ev

)1/2

UV edge plotted as indirect bandgapCalculated current density based on UV absorption edge for a 1 μm thick filmShaded region represents absorbed portion of AM1.5G

Pure WO3 – 2.59 mA/cm2

4 mTorr N2 – 6.30 mA/cm2

With sufficient nitrogen in the sputtering ambient, bandgap reductions are realized, however this increase in absorption does not correlate with an increase in measured photocurrents . . .

0.0 0.4 0.8 1.2 1.6 2.0 2.40.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5 0 mTorr N2 2 mTorr N2 3 mTorr N2 4 mtorr N2 6 mTorr N2

Cur

rent

Den

sity

(mA

/cm

2 )

Potential (V vs. SCE)

0.33 M H3PO4AM 1.5G illumination

PEC testing of WO3:N

16

2 0 2 1 2 2 2 3 2 4 2 5 2 6 2 7 2 8 2 9 3 0

6 m T o r r N 2 4 m T o r r N 2 3 m T o r r N 2 2 m T o r r N 2 0 m T o r r N 2

D e g re e s (2 - th e ta )

Cou

nts

(nor

mal

ized

)

2 0 2 1 2 2 2 3 2 4 2 5 2 6 2 7 2 8 2 9 3 0

6 m T o r r N 2 4 m T o r r N 2 3 m T o r r N 2 2 m T o r r N 2 0 m T o r r N 2

D e g re e s (2 - th e ta )

Cou

nts

(nor

mal

ized

)Influence of Nitrogen on Structure

X-ray diffraction pattern shows an initial decrease in crystallinity for monoclinic phase followed by transformation to a stabilized cubic phase for nitrogen partial pressures greater than 3mTorr. The TEM diffraction pattern for nitrogen doping shows smearing, indicating a highly defective lattice.

Progress: Tungsten-Based Compound Films

17Tungsten-Based FilmsPath Forward:

Elimination of lattice defects for the nitrogen doped WO3–by adjusting sputtering parameters–by post sputtering annealing–by exploring quaternary systems

Pursue bandgap reduction using different anion/cation species–oxy-sulfides (in addition to the oxy-nitrides)–by metallic cations

Based on the need to minimize requisite bias voltage, we will investigate valence band modifications, as well as the effect of composition on conduction band level and its alignment with the hydrogen reduction potential.

SUMMARY

FUTURE WORK

Tungsten oxide has been demonstrated in a multi-junction hybrid phototelectrode configuration.Nitrogen is observed to reduce the bandgap to ~2 eV.Incorporation of nitrogen results in a highly defective lattice. This degrades the transport properties resulting in poor PEC performance.Theory indicates that other dopants can favorably alter band edges.

18 Progress: Zinc-Based Compound FilmsZnO is inexpensive, nontoxic, and easy to synthesize. It also has adirect bandgap and high electron mobility thereby making it a good candidate for PEC splitting of water.

Conduction band edge only slightly mismatched to drive the hydrogen reaction.

Impurity doping can shift the valence band edge, providing for bandgap reduction.

Can also fabricate as P-type ZnO

promiseBandgap is too large to effectively utilize the visible light

Requires stability improvement against photocorrosion

challenges

• Impurity doping using anion or cation species

Bandgap reduction Explore p-type ZnO• Impurity doping using group 1B

metals

strategies for PEC improvement

19 Progress: Zinc-Based Compound FilmsStructural evolution for RF sputtering of ZnO:N

High quality pure ZnO thin films are synthesizedN-incorporation into ZnO thin films was achievedBandgap reduction in ZnO thin films was achieved by N-incorporationPhoto-response in ZnO:N thin films was demonstrated

400 500 600 700 800 900 1000 1100 12000.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0 ZnO ZnO:N at 80 W ZnO:N at 100 W ZnO:N at 120 W ZnO:N at 150 W ZnO:N at 200 W Zn3N2

Wavelength (nm)

Abs

orba

nce

Optical absorption spectra from ZnO:N samples

-0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2

0

10

20

(c) As-grown ZnO Annealed ZnO ZnO:N

Light power: 75 mW/cm2

Phot

ocur

rent

(μA

cm-2

)

Potential (V vs. Ag/AgCl)

200 400 600 800 10000

50

100

Tra

nsm

ittan

ce (%

T)

Wavelength (nm)

UV/IR filter Red filter

Photocurrent-voltage curves of the samples under light illumination with the UV/IR filter

Solution: Na2SO4

pH: 6.8

30 35 40 45 50 55

* *

*

ZnO(101)

Zn3N2

(440)Zn3N2

(400)

Zn3N2

ZnO:N at 200 W

ZnO:N at 150 WZnO:N at 120 W

ZnO:N at 100 W

ZnO:N at 80 W

ZnO

Inte

nsity

(a.u

.)

2θ (degree)

Substate

ZnO(002)

*

30 35 40 45 50 55

* *

*

ZnO(101)

Zn3N2

(440)Zn3N2

(400)

Zn3N2

ZnO:N at 200 W

ZnO:N at 150 WZnO:N at 120 W

ZnO:N at 100 W

ZnO:N at 80 W

ZnO

Inte

nsity

(a.u

.)

2θ (degree)

Substate

ZnO(002)

*

20

Bandgap reduction in ZnO thin films was achieved by Cu-incorporationP-type ZnO thin films with reduced bandgaps were achieved by Cu incorporationPhoto-response in ZnO:Cu thin films was demonstrated

400 500 600 700 800 900 10000.00.10.20.30.40.50.60.70.80.91.01.11.2

ZnO ZnO:Cu(4) ZnO:Cu(6) ZnO:Cu(10) ZnO:Cu(12)

Wavelength (nm)

Abs

orba

nce

(b)

Optical absorption spectra from ZnO:Cu samples Photocurrent-voltage curves of p-type ZnO:Cu samples

-0.6 -0.4 -0.2 0.0 0.2 0.4 0.6

-300

-250

-200

-150

-100

-50

0

50

Photo currents Undoped ZnO ZnO:Cu(4) ZnO:Cu(6) ZnO:Cu(10) ZnO:Cu(12)

Phot

ocur

rent

(μA

cm-2)

Potential (V vs. Ag/AgCl)

Solution: Na2SO4

pH: 6.8

Cu increase Grain size and surface roughness decreased

Progress: Zinc-Based Compound Films

21Zinc-Based Compound FilmsPath Forward:

SUMMARY

FUTURE WORK

Preliminary work shows that nitrogen can be effectively incorporated into ZnO films to reduce bandgapNeed to improve crystallinity for ZnO:N to realize good PEC properties.Theoretical calculations show that doping with group 1B transition metals (Cu, Ag) results in bandgap reduction.Cu was incorporated into ZnO films with mixed results – bandgapreduction is observed and Cu is found at Zn sites for select films, however the measured photocurrents are low.

The process for RF sputtering of ZnO films requires optimization.–Study the film growth as a function of substrate temperature.–Incorporation of impurity dopants via co-sputtering will be employed.

Band gap reduction–Impurity band generation by new impurities–Acceptor-donor co-doping

Stability improvement

22 Progress: Iron-Based Compound Films

strategies for PEC improvement

• Growth of crystalline oxide• Direct growth along the preferred

electron conduction paths• High Surface Area Materials

Photon To Electron Conversion Increase

Shift of Bandpositions

•Quantum size effects•Transition metal doping

Improve Kinetics of Water Oxidation

• Identify and deposit optimized surface oxidation electro-catalyst

Bandgap ~ 2 eV (40% solar light absorption).Abundant and inexpensiveHigh Stability in Electrolytes (pH>3)

Carrier TransportValence Band EdgeWater Oxidation Kinetics

Low optical absorption

challenges

Iron Oxide, as a commonly-found material with bandgap well-suited for the direct solar water splitting of water, is considered the “Holy Grail” of PEC materials- but its performance has been severely limited by opto-electronic properties…

promise

23 Progress: Iron-Based Compound FilmsSynthesis of nanorod electrodes

Synthetic control of nanorod structure

1) 35 ± 3 nm 2) 21 ± 2 nm 3) 20 ± 3 nm 4) 12 ± 3 nm

SubstrateFeCl3+Salt Quartzsubstrate

Ti/Pt

β-FeOOH

α-Fe2O3

nanorods

Thermaltreatment

RuO2

Electrocatalystdeposition

24

360 400 440 480 520 560 600 640 680

0

2

4

6

8

10

12

14

16

18

20S116C

RuO2 No Catalyst Cobalt

0.1M NaOH 400mV appliedbias vs Ag/AgCl N2 Purged

IPC

E (%

)

Wavelength (nm)360 400 440 480 520 560 600 640 680

0

2

4

6

8

10

12

14

16

18

20

IPC

E (%

)

Wavelength (nm)

S116C RuO2

No Catalyst Cobalt

0.1M NaOH 100mV appliedbias vs Ag/AgCl N2 Purged

Applied BiasApplied Bias

100 mV100 mV 400 mV400 mV

No CatalystNo Catalyst 0.3%0.3% 1.0%1.0%

RuORuO22 1.6%1.6% 0.9%0.9%

CobaltCobalt 0.2%0.2% 1.0%1.0%* Efficiency from IPCE data

1) 2) 3)

1) Photoelectrochemical performance of 21 and 15 nm nanorods under applied bias showingand efficiency of 1.3% and 1% respectively. (2,3) IPCE of 35nm Fe2O3 nanorods with an applied bias of 100mV and 400mV, graph 2 and 3 respectively, showing the dramatic improvement of the nanorods with the electrodeposition of RuO2 at low applied bias while at high applied bias the oxygen evolution catalyst is not improving the performance of the nanorods. 4) Table of results showing the Efficiency of the nanorods from graph 2 and 3.

4)

20 30 40 50 60

S

SS

S

Cou

nts

(a.u

.)

2 θ (deg.)

Hematite Nanorods (0.5M NaNO3) Hematite Nanorods (1M NaNO3)

H-(012)H-(104)

H-(110)

H-(113)

H-(024)H-(300)

S S-SubstrateH-Hematite

50 nm

Nanorod characterization & performanceProgress: Iron-Based Compound Films

25 Path Forward: Iron-Based Compound Films

SUMMARYEarly stages of research and development on iron oxide nanorods synthesis have shown:

Control over the size and morphology of the nanostructuresImproved photoelectrochemical performance as compared with spray

pyrolysis deposited filmsPhotoelectrochemical efficiencies of ~1 %

This synthesis method is applicable to wide variety of substrates and can be adapted to any size of substrates.

Control gross structure of photocatalysts– Nanorod synthesis conditions (pH, T, electrolyte composition)– Explore use of framework templates (deposit in cubic phase nanopores)

Control opto-electronic properties– Nanorod size reduction to increase VB Confinement at d<6nm– Nanorod growth to be along (110) plane– Nanorod doping by in situ growth or high temperature diffusion

Control kinetics– Selective deposition of surface electrocatalysts (start with Pt, Ni, Au, Ru)

FUTURE WORK

26 Progress: Silicon-Based Compound FilmsAmorphous silicon carbide is an electrochemically stable and photoactive material with tunable bandgap, which would enable the fabrication of “all-silicon” multi-junction water-splitting devices.

Tunable bandgap of 2.0-2.3 eVand good optoelectronic qualityLarge knowledgebase from a-Si technologya-SiC shows good corrosion resistanceEnables “all-silicon multi-junction device” to be fabricated in a “cluster tool” machine

Non-ideal band edge alignment –requires supplemental bias

Kinetic limitations apparent for bare a-SiC electrodesLong term corrosion and photo-corrosion behavior is not known

promise challenges

1.7

1.9

2.1

2.3

2.5

0.0 0.1 0.2 0.3 0.4 0.5 0.6CH4/(CH4+SiH4) in source gases

Tauc

Eg

(eV

) Amorphous SiPanel

Bandgaptuning by carbon content(Gas flow ratio)

27 Progress: Silicon-Based Compound FilmsAll-Silicon Hybrid PEC Device Designs

n i p n i p

a-S

iCEg

2-2.

3eV

a-Si

Eg=1

.75e

V

nc-S

iEg

~1.1

eV

Sta

inle

ss S

teel

Su

bstra

te

O2C

atal

yst

photon 1photon 2

photon 3

qV1

qV2

qV3

elec

troly

te

elec

troly

te

photons

photons

(HER)

(OER)

h+

e-

* US patent #6,258,408B1: MVSystemslarge-scale cluster tool design reel-to-reel cassette*

light

H2 evolutionPhotoactivep-type a-SiC layer

ohmic interface

a-Si and/or nc-Si solar cell

Stainless steel substrate O2 evolution catalyst

O2 evolution

electrolyte

Cluster-tool fabrication equipment

Device schematic Band diagram using A-Sic PEC film

28

0

20

40

60

80

100

300 500 700 900 1100Wavelength (nm)

Opt

ical

Tra

nsm

issi

on (%

)

a-SiC Eg=2.0eV, 0.20 um

a-SiC Eg=2.0eV, 0.31 um

-15

-12

-9

-6

-3

0

-1.4 -1.2 -1.0 -0.8 -0.6Potential (V vs SCE)

Cur

rent

(mA

/cm

2 )

Progress: Silicon-Based Compound Films

Photocurrent of >9mA/cm2

with a-SiC photocathode

Route to >10% STH Efficiency

photo-electrode

photo-electrode bandgap

(eV)

current available (m A/cm 2)

photovoltaic layer

configuration

current available

(filtered by top layer) (m A/cm 2)

possible STH (% )

achieved STH to date(% )

W O 3 2.6-2.8 4.3-3.3 a-S i/a-S i 8.0 5.3 3.1SiC 2 14.4 a-S i/nc-S i 6.0 7.4SiC 2.35 8.1 a-S i/nc-S i 7.8 ~10SiC ~2 ~14 nc-Si ~14 ~17

1-sunAM1.5 G

0.33M H3PO4

a-SiC EG=2.0eV, 0.31 um

Optical properties tailored through bandgap and thickness

29Silicon-Based Compound FilmsPath Forward:

SUMMARY

FUTURE WORK

One-sun photocurrent of ~9 mA/cm2 demonstrated for a-SiC electrodea-SiC films appear stable in cathodic regime (short-term tests)Flatband potential and photocurrent onset indicate non-ideal valence band maximum position and kinetic limitationsa-SiC is compatible with automated fabrication of multijunction devices in cluster tool deposition machine

Optimize optoelectronic quality of a-SiC at EG=2.0eV-2.3eV– systematic variation of PECVD process parameters

Comprehensive PEC characterization a-SiC photoelectrodes– band positions, electrode kinetics– long term stability

Fabricate and characterize complete monolithic a-SiC/a-Si multijunctionPEC device

30 Progress: Copper Chalcopyrite Films

Direct bandgap and good carrier transport propertiesHigh PEC photocurrents demonstrated for p-type Cu(In,Ga)Se2 electrodesBandgap and band edges “tunable” by composition Synergy with PV CIGS multijunctiondevice research and development

Valence band edge of the Cu(In,Ga)Se2films too highKinetic limitations apparent for bare electrodesLong term corrosion and photo-corrosion behavior is not knownHigh-temperature fabrication steps

Bandgap Bowing Rangesof Cu(In,Ga)Se2 and CuGa(Se,S)2S.-H. Wei and A. Zunger, JAP 78, 3846 (1995);H. Matsushita et. al, JJAP 29, 484 (1990)

promise challenges

Cu(In,Ga)Se2 is proven as an efficient absorber for thin-film solar cells and its optoelectronic properties are equally well suited for photoelectrolysis.

1.0

1.5

2.0

2.5

0.0 0.2 0.4 0.6 0.8 1.0

Composition Parameter X

Band

gap

(eV)

1.0

1.5

2.0

2.5CuGa(Se1-XSX)2

CuIn1-XGaXSe2

CuGaSe2

CuInSe2

CuGaS2

CuGaSe2

31 Progress: Copper Chalcopyrite Films

Electrodes are stable during short-term PEC testingValence band edge too positive in current samples – high bias requiredPhotocurrent onset at least 0.3V more cathodic than flatband potentialResistance of back contact (high/SnO2:F vs low/Mo) has strong impact on photocurrent curve

before before

after after

2μm

CuIn0.2Ga0.8Se2(3-stage process)

CuGaSe2(3-stage)

Stability: morphology before/after PEC testing

0

200

400

600

800

1000

1200

-2.0 -1.5 -1.0 -0.5 0.0Potential (V vs SCE)

J2 (mA

2 /cm

4 )

CIGS/Mo

CIGS/SnO2

3-stage CGS/SnO2

2-stage CGS/SnO2

Photocurrent onset potential

CIGS/Mo:Illuminated open-circuit

=0.07 V vs SCE=0.31 V vs NHE

Stability, and Band Edge Positions

32

25 30 35 40 45 50 55

2 Theta (degrees)

log(

Inte

nsity

)Progress: Copper Chalcopyrite Films

-40

-30

-20

-10

0

-1.0 -0.8 -0.6 -0.4 -0.2 0.0Potential (V vs SCE)

Cur

rent

(mA

/cm

2 )

-40

-30

-20

-10

0

-0.4 -0.2 0.0 0.2 0.4 0.6 0.8Bias Voltage (V)

Cur

rent

(mA

/cm

2 )

28.3 mA/cm2 28.0 mA/cm2

VOC=663mVFF=0.613η=11.3%

in 0.5M H2SO4

CuIn1-xGaxSe2

Mo back contact

Soda-lime glass

light

CdS

CuIn1-xGaxSe2

ITO

Soda-lime glass

i-ZnO

light

Mo back contact

PEC electrode Solar cell1-Sun Photocurrent: similar in PEC and PV Devices (CuIn0.4Ga0.6Se2)

CuGaSe23-stage process

SnO2

(112)

(220) (204)

(312)(116)

CuGaSe22-stage process

-40

-30

-20

-10

0

-2.0 -1.5 -1.0 -0.5 0.0Potential (V vs SCE)

Cur

rent

(mA

/cm

2 )CuGaSe2 deposition conditions, film texture, and photocurrent curves

Preliminary data shows possible effect of texture [(112)/(220)+(204)] on saturation photocurrent and dark current.

in 0.5M H2SO4

AM1.5G

XRD: CGS samples from 2 different recipes Photocurrent 1-sun AM1.5G

3-stage

2-stage

33Copper Chalcopyrite Films

SUMMARY

FUTURE WORK

Path Forward:

1-sun photocurrents of 21-28 mA/cm2 demonstrated for wide-bandgap CIGS and CGS electrodesWide-bandgap CIGS and CGS Films appear stable in cathodic regimeFlatband potential measurement indicates valence band maximum too highLate photocurrent onset indicates kinetic limitations at electrode surface

Assemble comprehensive body of PEC data on existing CIGS electrodesStudy materials with wider bandgap and lower valence band

– CuGaSe2 with Cu-poor surfaces– Alloy with sulfur for wider bandgap CuGa(Se,S)2

Explore metal nanoparticle deposition on electrode surface for kinetic improvements