MA

X-P

LAN

CK

-INS

TITU

T FÜ

R B

IOA

NO

RG

AN

ISC

HE

CH

EM

IE MAX-PLANCK-INSTITUTFÜR BIOANORGANISCHE CHEMIE

SOLAR PRODUCTION OF HYDROGEN AND OXYGEN INCLUDING STORAGE THEREOF

WITH SILICIDES IN WATER

Prof. Martin Demuth and Dipl. Ing. Peter Ritterskamp

Max-Planck-Institut für Bioanorganische Chemie Stiftstr. 34-36, D-45413 Mülheim an der Ruhr

Phone: +49 208 306-3671/ -3680/ -4 Fax: +49 208 306-3951demuthm@mpi-muelheim.mpg.de http://www.mpibac.mpg.de

PATENT APPLICATION: Generation of Hydrogen and Oxygen from Water and Storage thereof with Silicides 10 2005 040 255.0 (Germany)

and PCT/EP application 2006 / 008333

MA

X-P

LAN

CK

-INS

TITU

T FÜ

R B

IOA

NO

RG

AN

ISC

HE

CH

EM

IETHE CHALLENGE –

WE ARE APPROACHING A SOLUTION

MA

X-P

LAN

CK

-INS

TITU

T FÜ

R B

IOA

NO

RG

AN

ISC

HE

CH

EM

IESCHEMATIC REPRESENTATION OF A SEMICONDUCTOR

(SC, plain vs doped material)

• Electrons are moved from the valence band to the energetically higher

conduction band creating a „„charge hole““ in the valence band enablingthe reduction and oxidation of water to yield hydrogen and oxygen, resp.

• An energetically appropriate bandgap is required for this purpose

• Furthermore, the sc should absorb light in the solar range of 300-800 nm

MA

X-P

LAN

CK

-INS

TITU

T FÜ

R B

IOA

NO

RG

AN

ISC

HE

CH

EM

IECONDUCTION-BAND, VALENCE-BAND AND

BANDGAP ENERGIES OF SOME SEMICONDUCTORS

Potential (NHE), pH=7

MA

X-P

LAN

CK

-INS

TITU

T FÜ

R B

IOA

NO

RG

AN

ISC

HE

CH

EM

IEBANDGAP RANGE OF TITANIUM SILICIDES

(TiSix)

• An ideal semiconductor should have:bandgap > 1.23 eVconduction band < - 0.41 eVabsorption of visible light < 2.2 eV

-1

0

1

2

3

-0.41

+0.82

-0.41

+0.82

H / H2+

O2 / OH -

~3.4 eV360 nm

~1.5 eV800 nm

pH = 7

pH > 7

EV Potential (NHE)

-0.8 sensitizer (perylenes)

MA

X-P

LAN

CK

-INS

TITU

T FÜ

R B

IOA

NO

RG

AN

ISC

HE

CH

EM

IE

FACHBEIRAT UND KURATORIUM · MÄRZ 2004

1 M + H2O → MOn + H2

formation of hydrogen energetically driven by metal oxide formation

forms oxide (layer)

thermally

2

M + O2 → MOnoxidation of

catalyst by oxygen3

M + H2O → M + H2 + 1/2 O2 water splitting

thermally

POSSIBLE (SIDE) REACTIONS IN METAL-DRIVENWATER SPLITTING

photochemically

MA

X-P

LAN

CK

-INS

TITU

T FÜ

R B

IOA

NO

RG

AN

ISC

HE

CH

EM

IE

Ti-

-Ti4+Si4-light H2

2H+

11

H2

H2

44

x

3 3 O2

Current Current investigationsinvestigations concernconcern the the mechanismsmechanisms of and of and 22 3 3

ReactivityReactivity of of TiSiTiSixx

22½ O2 + 2H+18

18H2O

TiSi

+ +

TiO2

SiO2

Si

MA

X-P

LAN

CK

-INS

TITU

T FÜ

R B

IOA

NO

RG

AN

ISC

HE

CH

EM

IEHYDROGEN AND OXYGEN EVOLUTION KINETICS

(including partial hydrogen storage and very efficient oxygen storage)

0

10

20

30

40

50

0 50 100 150 200 250Time (h)

ml

a (hydrogen, reaction under nitrogen) a (oxygen from storage) c (hydrogen, reaction under air)b (hydrogen, reaction under nitrogen) b (oxygen from storage)

24±1 ml

12±1 ml

A (a)

B (a)

60° C 5-8° C

a (H2)

a (O2 ) c (H2)

Phase A: oxidationby water addition (major) and water splitting (minor)

Phase B: exclusive water splitting

b (O2)

b (H2)

2±1 ml

50° C 5-8 50° C

30° C

30° C

MA

X-P

LAN

CK

-INS

TITU

T FÜ

R B

IOA

NO

RG

AN

ISC

HE

CH

EM

IE

0

10

20

30

40

50

60

70

80

0 100 200 300 400 500 600hours

mL

PRESSURE DEPENDENCE OF WATER SPLITTING EFFICIENCY

η = 4.5 – 11.5 % (1.2 bar)(hydrogen evolution shown)

MA

X-P

LAN

CK

-INS

TITU

T FÜ

R B

IOA

NO

RG

AN

ISC

HE

CH

EM

IEHydrogen (sacrificial vs splitting) and oxygen production

UNLOADING OF OXYGEN FROM STORAGE

--10243086

< 614161217

< 3 d,e

5 e12 e15 e43 e

6244042

103

12345

H2 (from water splitting, ml) c

H2 (sacrificial)(ml) b

O2(ml) a

H2 (total in gasphase, ml)

Run

Table. a Gas phase + 3 ml dissolved in water; b H2 (total) - H2 (splitting); c calc: ml O2 x 2; d below detection limit in gas phase; e unloaded by cooling-warming cycle. Exptl. error 5-10%.

MA

X-P

LAN

CK

-INS

TITU

T FÜ

R B

IOA

NO

RG

AN

ISC

HE

CH

EM

IE

-0.41 eV

(3)

4e−

TiO2+ H2

SiO2+ H22H2O

(2)

H2

H2O

-0.43 eVO2

SiO2

(1)

TiO2

O2

(1)

H2O

(1) formation of passivating/catalytic/storage layer (thermal)(2) water splitting (→ O2 + 4H+) (light)(3) reduction 2H+ → 2H● → H2 (thermal)

O2 + 4H+

H2

(4)

(4) reversible storage of H2 and O2 (thermal)

O2(4) (chargedstacks)

MA

X-P

LAN

CK

-INS

TITU

T FÜ

R B

IOA

NO

RG

AN

ISC

HE

CH

EM

IEARE SILICIDE SEMICONDUCTORS IDEAL FOR

WATER SPLITTING? - YES -INCLUDING REVERSIBLE STORAGE OF THE GASES

• Full solar light absorption (350 – 800 nm)

• Conduction band (CB) < H2/H2O (< - 0.41 eV)

• Valence band (VB) > O2/H2O (> + 0.82 eV)

• Cheap and abundant

• Reversible storage of hydrogen and oxygen under different reaction conditions (-> easy separation of the gases)

• Exothermic unloading of oxygen

• High efficiency of water splitting (presently ca 12%)

MA

X-P

LAN

CK

-INS

TITU

T FÜ

R B

IOA

NO

RG

AN

ISC

HE

CH

EM

IERESEARCH TEAM AND COOPERATIONS

Dipl. Ing. Peter RitterskampDr. Andriy Kuklya

Klaus KerpenMarc-André Wüstkamp

Max Planck Institute for Bioinorganic Chemistry

Dr. Claudia Weidenthaler Max Planck Institute for Coal Research

Prof. Dr. Horst KischUniversity of Erlangen

Dr. Jürgen KleinwächterBSR Solar Technologies, Lörrach

MA

X-P

LAN

CK

-INS

TITU

T FÜ

R B

IOA

NO

RG

AN

ISC

HE

CH

EM

IEPROJECT PARTNERS

Hydrogen HeatingsZürich

2HSystems

MA

X-P

LAN

CK

-INS

TITU

T FÜ

R B

IOA

NO

RG

AN

ISC

HE

CH

EM

IE

MA

X-P

LAN

CK

-INS

TITU

T FÜ

R B

IOA

NO

RG

AN

ISC

HE

CH

EM

IENEAR FUTURE GOAL: ONE-FAMILY HOUSE

HEATING TECHNOLOGY BY HPS(Hydrogen Power System)

Demand: 12-15 kW h – 12-15 m3 H2/day

Hydrogen Power System (present):24 kg cat. / 60 m2 / 0.6 m3 H2/day (halogen light)

/ 3 – 12 m3 H2/9 h (solar)

Generation and Storage of H2 and O2

Pump

Combustion(HPS)

MA

X-P

LAN

CK

-INS

TITU

T FÜ

R B

IOA

NO

RG

AN

ISC

HE

CH

EM

IECATALYTIC COMPOSITION – TiSix and/or oxides thereof?

TiSix TiO2: Ti(0) : Six(0) : SiO2

Redox potentials (- 0.41 eV to – 0.43 eV/ + 3.4 – 1.5 eV)active form?

Generation of H2 TiO2 : Ti(0)Generation of O2 Six(0) : SiO2

Light absorber TiSix

MA

X-P

LAN

CK

-INS

TITU

T FÜ

R B

IOA

NO

RG

AN

ISC

HE

CH

EM

IE

Ti-

-Ti4+Si4-light H2

2H+

11

H2

H2

44

x

3 3 O2

Current Current investigationsinvestigations concernconcern the the mechanismsmechanisms of and of and 22 3 3

ReactivityReactivity of of TiSiTiSixx

22½ O2 + 2H+18

18H2O

TiSi

+ +

TiO2

SiO2

Si

MA

X-P

LAN

CK

-INS

TITU

T FÜ

R B

IOA

NO

RG

AN

ISC

HE

CH

EM

IE

-0.41 eV

(3)

4e−

TiO2+ H2

SiO2+ H22H2O

(2)

H2

H2O

-0.43 eVO2

SiO2

(1)

TiO2

O2

(1)

H2O

(1) formation of passivating/catalytic/storage layer (thermal)(2) water splitting (→ O2 + 4H+) (light)(3) reduction 2H+ → 2H● → H2 (thermal)

O2 + 4H+

H2

(4)

(4) reversible storage of H2 and O2 (thermal)

MA

X-P

LAN

CK

-INS

TITU

T FÜ

R B

IOA

NO

RG

AN

ISC

HE

CH

EM

IE

-0.41 eV

(3)

4e-

SiO2+ H2

TiO2+ H22H2O

(2)

H2

(1) formation of passivating/catalytic/storage layer (thermal)(2) water splitting (→ O2 + 4H+) (light)

(4) reversible storage of H2 and O2 (thermal)(3) reduction 2H+ → 2H● → H2 (thermal)

TiSiXH2

(4)

O2 + 4H+

H2O

H2O

TiO2

(1)

-0.43 eV

(1)

SiO2

MA

X-P

LAN

CK

-INS

TITU

T FÜ

R B

IOA

NO

RG

AN

ISC

HE

CH

EM

IE

• Hydrogen can be gained from reforming processes(e.g. catalytically from methane or propane)

Disadvantages: High pressure and temperature are required, besidesmethane and propane being fossil sources

• Biomimetics: Attempts to mimic biosynthesis in green leafs in whichnature produces primarily oxygen and protons (H+), the latter of whichcould potentially be subjeced to enzymatic transformation with ahydrogenase to form hydrogen.

SOURCES OF HYDROGEN

Biosynthesis -Photosystem II

Mn4 cluster

Hydrogenase

+

[NiFe] cluster

MA

X-P

LAN

CK

-INS

TITU

T FÜ

R B

IOA

NO

RG

AN

ISC

HE

CH

EM

IE

• A further and in our opinion in the near future technically realizable

variant of „„farfar--fetchedfetched biomimetics““ concerns the use of novelsemiconductors (scs) as catalysts for „water splitting“ together with sunlight

Semiconductor

sc

Mn4 cluster

+

[NiFe] cluster

MA

X-P

LAN

CK

-INS

TITU

T FÜ

R B

IOA

NO

RG

AN

ISC

HE

CH

EM

IEHYDROGEN AND OXYGEN EVOLUTION KINETICS

0

5

10

15

20

25

30

35

40

45

50

0 50 100 150 200 250

Time (h)

ml

a (hydrogen) a (oxygen evolution from storage) c (hydrogen, reaction under air) b (hydrogen)

24±1 ml

12±1 ml A

B

C

55 Co 5-8 Co

a (H )2

a (O )2

c (H )2

Phase A: InitiationPhase B: Corrosion (major) and splitting (minor)Phase C: Water splitting (exclusive)

MA

X-P

LAN

CK

-INS

TITU

T FÜ

R B

IOA

NO

RG

AN

ISC

HE

CH

EM

IE OUR OLD AND LITERATURE RESULTS

• Experiments with doped titanium dioxide (e.g. TiO2 doped with ruthenium; 300-380-nm irradiation):

→ insufficient solar light absorption and catalysis→ degradation of „catalyst“→ applicable in water of high purity only

CURRENT RESEARCH ACTIVITIES

• Our hitherto most successful result in this area is based on the use of semiconductor-type nanomaterials which had so far not been used for water reduction and oxidation to produce hydrogen and oxygen, resp.:

→ Silicides: TiSix, V2Si, ZrSix, Pt2Si, MnSi2 and Ni2Si→ 300-800-nm irradiation

MA

X-P

LAN

CK

-INS

TITU

T FÜ

R B

IOA

NO

RG

AN

ISC

HE

CH

EM

IE HowHow to to improveimprove 2 vs. 3 ?2 vs. 3 ?

-- reducereduce oxygen oxygen solubilitysolubility in in waterwater layerlayer

-- dopingdoping of of TiSiTiSixx with Pt or with Pt or II II / / IVIV--valentvalent MOMOxx

-- controlcontrol of metal of metal impuritiesimpurities

-- avoidavoid Si and Ti Si and Ti contactscontacts with with waterwater andandoxygen by oxygen by coatingcoating ((NafionNafion membranemembrane) )

MA

X-P

LAN

CK

-INS

TITU

T FÜ

R B

IOA

NO

RG

AN

ISC

HE

CH

EM

IECONCLUSION

Superiority of the silicides:

• Not yet known literature-wise for the title application

• Cheap and abundant

• Thermal and light stability

• Adequate solar light absorption (350-800 nm)

• BONUS: The silicides are able to store reversibly hydrogen at ambient temperature

• Catalysis? 2:1 ratio for H2/O2 generation : yes

• Modifications of the silicide compositions and surface engineering, as well as further technological improvements can likely secure the lead in this field.

MA

X-P

LAN

CK

-INS

TITU

T FÜ

R B

IOA

NO

RG

AN

ISC

HE

CH

EM

IE SUMMARY OF CURRENT AND PLANNED R&D ACTIVITIES

• Further silicide nanostructures (FeSix, B4Si, CoSi2, MnSi2, IrSi2, Ti3C2Si, ZrSi2, TaSi2 CrSi2 etc.)

• Surface engineering: a) Doping of the silicides, b) coating andc) in situ complexation of e.g. conducting HOR-type materials to the semiconductor surface

• Check light and sc concentration, temperature and pH dependence

• Evaluation of the reaction mechanisms, especially the fate of oxygen

• Immobilization of the silicides

MA

X-P

LAN

CK

-INS

TITU

T FÜ

R B

IOA

NO

RG

AN

ISC

HE

CH

EM

IE

M. Grätzel, The artificial leaf,Bio-mimetic Photocatalysis

MA

X-P

LAN

CK

-INS

TITU

T FÜ

R B

IOA

NO

RG

AN

ISC

HE

CH

EM

IE

Schematic of operation of the dye-sensitized electro-chemical photovoltaic cell. The photoanode, made of a mesoporous dye-sensitized semiconductor, receives electrons from the photo-excited dye which is thereby oxidized, and which in turn oxidizes the mediator, a redox species dissolved in the electrolyte. The mediator is regenerated by reduction at the cathode by the electrons circulated through the external circuit. Figure courtesy of P. Bonhôte/EPFL-LPI.

Scanning electron micrograph of the surface of a mesoporous anatase film prepared from a hydrothermallyprocessed TiO2 colloid. The exposed surface planeshave mainly {101} orientation.

„Grätzel„Grätzel--Zelle“Zelle“

M. Grätzel, Nature 414, 338 (2001) ~20 nm particles

Wirkungsgrad ~10%gute Langzeitstabilität preiswertes Material

MA

X-P

LAN

CK

-INS

TITU

T FÜ

R B

IOA

NO

RG

AN

ISC

HE

CH

EM

IE

Figure 1 Principle of operation of photoelectrochemicalcells based on n-type semiconductors. Cell that generates a chemical fuel, hydrogen, through the photo-cleavage of water.

Figure 6 The Z-scheme of photocatalytic water decomposition by a tandem cell.

Photoelektrochemische WasserspaltungPhotoelektrochemische Wasserspaltung

M. Grätzel, Nature 414, 338 (2001)

MA

X-P

LAN

CK

-INS

TITU

T FÜ

R B

IOA

NO

RG

AN

ISC

HE

CH

EM

IEPhotokatalyse mit neuen Photokatalyse mit neuen HalbleiternanomaterialienHalbleiternanomaterialien

-- WasserspaltungWasserspaltung

Titan-Silicid-Katalysatoren :

Vorteile: Nutzung des Bereichs des sichtbaren Lichts (350-800 nm)Einsatz von normalem Wasserhohe H2-Erzeugungsratengute Langzeitstabilität reversible O2- und H2-Speicherung bei RT

Zukunft: Fixierung auf /in polymeren (leitenden) Materialiengetrennte oder kombinierte O2/H2 -Entwicklungvariables Ti:Si-Verhältnis

Alte Experimente mit1) TiO2 dotiert2) TiX (X ≠ O) / TM beschichtet

Nachweis der H2/O2-Entwicklung

Demuth, Ritterskamp, Patentanmeldung 2005

MA

X-P

LAN

CK

-INS

TITU

T FÜ

R B

IOA

NO

RG

AN

ISC

HE

CH

EM

IEREACTION PATHWAYS

MA

X-P

LAN

CK

-INS

TITU

T FÜ

R B

IOA

NO

RG

AN

ISC

HE

CH

EM

IEM

AX-

PLA

NC

K-IN

STIT

UT

FÜ

R

BIO

AN

OR

GA

NIS

CH

E C

HEM

IE

FACHBEIRAT UND KURATORIUM · MÄRZ 2004

# Reaction scheme Description Nature

1 M + H2O → MOn + H2Formation of hydrogen by

metal oxide formationPhotochemical /

Thermal

2 M+O2 → MOnOxidation of catalyst by

oxygen Thermal

3 M + H2O → M+ H2 +1/2 O2 Water splitting Photochemical

3 is not the case: H2 : evolution due to metal oxide formationO2 : no evolution, consumption due oxidationOxygen consumption is not dependent to hydrogen production

1 is not the case:H2 : evolution due to water splittingO2 : evolution due to water splitting (1/2 to H2) and consumption due oxidationAs more hydrogen produced as less measurable consumption.

MA

X-P

LAN

CK

-INS

TITU

T FÜ

R B

IOA

NO

RG

AN

ISC

HE

CH

EM

IEM

AX-

PLA

NC

K-IN

STIT

UT

FÜ

R

BIO

AN

OR

GA

NIS

CH

E C

HEM

IE

FACHBEIRAT UND KURATORIUM · MÄRZ 2004

measured hydrogen production and oxigen consumption

0

10

20

30

40

50

60

70

80

0 50 100 150 200

hours

ml

20C-H2 prod20C-O2 prod50C-H2 prod50C-H2 prod20C-O2 prod50C-O2 prod

hydrogen production and calculated oxygen consumtion at 50 and 20 degry.

0

10

20

30

40

50

60

70

80

0 50 100 150 200

hours

ml

20C-H2 prod20C-O2 cons50C-H2 prod50C-O2 cons

MA

X-P

LAN

CK

-INS

TITU

T FÜ

R B

IOA

NO

RG

AN

ISC

HE

CH

EM

IE

0 50 100 150 200 2500

20

40

60

80

100

120

140

160

180

200

> 30 ml H2 / day / g cat

ml o

f hyd

roge

n (H

2)

time (h)

6-10 ml H2 / day / g cat

HYDROGEN PRODUCTIONWITH TiSi2-4 FROM WATER USING HALOGEN

LIGHT

MA

X-P

LAN

CK

-INS

TITU

T FÜ

R B

IOA

NO

RG

AN

ISC

HE

CH

EM

IE

-0.41 eV

(1)

(3)

4e-

TiO2+ H2

SiO2+ H22H2O

(2)

H2(1)

-0.43 eV

(1) formation of passivating/catalytic/storage layer (thermal)(2) water splitting (→ O2 + 4H+) (light)

(4) reversible storage of H2 and O2 (thermal)(3) reduction 2H+ → 2H● → H2 (thermal)

TiSiXH2

(4)

O2 + 4H+

H2O

H2O