Photochemical CO 2 Reduction: Current Status and Future Prospects. Etsuko Fujita Chemistry Department Brookhaven National Laboratory

Photochemical CO2 Reduction: Current Status and Future Prospects Etsuko Fujita Chemistry Department Brookhaven National Laboratory Utilization of CO...
Author: Clare Pearson
5 downloads 0 Views 1MB Size
Photochemical CO2 Reduction: Current Status and Future Prospects Etsuko Fujita Chemistry Department Brookhaven National Laboratory

Utilization of CO2 in Synthetic Chemistry „

Intermediates or fine chemicals for the chemical industry z z z z z

„ „

−C(O)O− : acids, esters, lactones −O−C(O)O− : carbonates −NC(O)OR− : carbamic esters −NCO : isocyanates −N−C(O)−N : ureas

Use as a solvent Energy rich products z z

OX2-

O RNHC

HCOOH

CO

CH3OH 2e- 2e + 2e2H RNH2 R'RNCORR' 4e R'X R'RNH CH3OH 4H + R R H2 RC ≡CR O CH 4 2 H2 OR'

CO

O

O2 HOOC

CO, CH3OH Hydrocarbons and their derivatives

O O

H2 ONa

H2 ROH H2 NH3 R2 NH

O

O

OH COONa HC

Cn H2n+2 Cn H2n Cn H2n+1 OH

Cn H2n+1 NH2

OR O

HC

The amount of CO2 (115 Mt) used annually by the chemical industry is less than 0.5 % of the total annual CO2 emissions----TOO MUCH CO2 TO CONVERT TO CHEMICALS!

NR2

M. Aresta, 1998

CO2 Hydrogenation CO2 + H2

HCOO−, Me2NCHO, MeOH

■ CO2 to formate, dimethylformamide, etc. In scCO2 with (Me3P)4RuCl2 TON: 4 X 105, TOF: 8 X 103 h-1

■ CO2 to methanol Cu/ZrO2, Cu/ZnO, and Cu/ZrO2/ZnO-based catalysts at 200-250 ºC „

H2 supply: The big problem + H2O

CH4(g) + H2O(g) → 3 H2(g) + CO(g) → 4 H2(g) + CO2(g) High-temperature steam reforming over a nickel catalyst

Renewable H2 is needed for CO2 hydrogenation to be practical

Renewable Energy Resources Solar

1.2 x 105 TW at Earth surface 600 TW practical

Wind

2-4 TW extractable

Tide/Ocean Currents 2 TW gross

Geothermal 12 TW gross over land small fraction recoverable

Energy Gap ~ 14 TW by 2050 ~ 33 TW by 2100

Biomass

Climate Change

5-7 TW gross all cultivatable land not used for food

Hydroelectric 4.6 TW gross 1.6 TW technically feasible 0.9 TW economically feasible 0.6 TW installed capacity

N. Lewis, Caltech

Strategy for CO2 Reduction ■ Reduction of CO2 requires energy ⇒ Solar electricity as energy source (Electrochem) ⇒ Photon as energy source (Photochem) ■ One electron process is unfavorable ⇒ Multi-electron transfer catalysts CO2 + e− → CO2− CO2 + H+ + 2e− → HCO2− CO2 + 2H+ + 2e− → CO + H2O CO2 + 6H+ + 6e− → CH3OH + H2O

Comments Inorg. Chem. 1997, 19, 67 Coord. Chem. Rev. 1999, 185, 373

E° = E° = E° = E° =

− 1.9 V

(vs NHE at pH 7)

− 0.49 V − 0.53 V −

0.38 V

Artificial Photosystem for CO2 Reduction ■ Artificial photosynthesis for CO2 reduction typically requires a photosensitizer, a catalyst and an electron donor ■ Due to the complex nature, a reduction half reaction is normally investigated

hν Energy Storage CO2+ e− CO, HCOO-, CH3OH

■ Products are CO, formate, and H2 ■ Quantum yields reach up to 60 %

D

PS*

D+

PS-



PS

ML+ ML2+

CO2 CO, HCOO-

Homogeneous Photochemical CO2 Reduction Ru(bpy)32+/Co(Me2Phen)32+/TEOA

CO, H2

0.012(CO), 0.065(H2)

Ru(bpy)32+/Ru(bpy)(CO)22+/TEOA

HCOO­

0.14

Ru(bpy)32+/Ni(cyclam)2+/H2A

CO, H2

0.001(CO)

Ru(bpy)32+/Co(cyclam)2+/H2A

CO, H2

Terphenyl/Co(HMD)2+/TEA

CO, (H2, HCOO-)

Terphenyl/Co(cyclam)3+/TEA

CO, H2, HCOO­

Metal porphyrins (or phthalocyanines)/TEA

CO, H2, HCOO­

Re(bpy)(CO)3X/TEOA

CO

0.14

Re(P(OEt)3)(bpy)(CO)3+/TEOA

CO

0.38

CO Re(bpy)(CO)3(CH3CN)+, Re((MeO)2bpy)(CO)3(P(OEt)3)+/TEOA

0.59

0.13(CO)

R

tBu

2− tBu

N

N N

N

N

HN

NH

N

NH HN NH HN

N

N

N

R

R N

N

N

N

N

N

N

N

N

Me2Phen

HMD bpy

cyclam

R

2−

tBu

tBu

Photochemical CO2 Reduction beyond CO ■ CO2 reduction to methane using a system with Ru(bpz)32+, N Ru colloid and TEOA (Φ= 10−4) I. Willner, et al.

N N

bpz

N

■ CO2 reduction to CO (with a trace amount of methanol and methane) using Ti-O or transition-metal grafted zeolites and silicates M. Anpo, et al., H. Frei, et al. ■ p-GaP based photoelectro-reduction of CO2 to methanol in the presence of pyridinium at −0.4 V vs. SCE (pH 5.2) A. Bocarsly, et al. (J. Am. Chem. Soc. 2008, 130, 6342-6344) ● Highly selective reaction ● No overpotential (the standard potential: −0.52 V) ● Mechanism of methanol formation? Interaction of CO2 with pyridinium ion (carbamate formation)? Hydride transfer reactions?

Success and Challenges ■ Light absorption and charge separation were coupled with dark reaction ■ CO2 was photochemically reduced to CO or formate, in some cases, to methanol ■ Quantum yield of CO formation is up to 0.60 Slow reaction rate (~10/h) Low stablity of catalyst/photosensitizer (TN ~200 or less) Mechanism of CO2 reduction? Controlling the selectivity (CO, formate, H2, etc) Beyond CO or formate? Coupling reductive and oxidative half reactions: i.e., to remove a sacrificial electron donor ■ Use of inexpensive catalyst/photosensitizer ■ ■ ■ ■ ■ ■

Outline ■ Terphenyl/Co macrocycle/TEA CO2 binding, Mechanistic investigation of CO formation ■ Re(bpy)(CO)3X/TEA

Understanding reaction mechanisms Improving reaction rates

■ Toward CO2 reduction by photogeneration of renewable hydride donors

Terphenyl/Co Macrocycle/TEA ■ p-Terphenyl as a photosensitizer, triethylamine as an electron donor, and Co macrocycles as catalysts ■ Co macrocycles suppress the hydrogenation of terphenyl: Stable photocatalytic system ■ Efficient formation of CO with Co macrocycles: Φ 313 nm = 0.13 (CO per photon)

TEA/TEOA

TEA+/TEOA+

1TP*

TP-

J. Am. Chem. Soc., 1993, 115, 601



TP

N NH

HN

NH HN

N

NH HN

CoL+

CO2

CoL2+

CO, HCOO-

N

HN

NH

Properties of CoL+ and [CoL-CO2]+

N

CoIL+ + CO2

CoL(CO2)+ K = 1.2 x 104 M-1

CoL(CO2)+ + S

SCoL(CO2)+

S = CH3CN Stretching frequencies in CD3CN: νNH νCN 2+ CoL 3215 1661 I + Co L : 3201 1571 + CoL(CO2) : 3208 1648 SCoL(CO2)+: 3145 1648

Reduced catalyst, CoIL+ (d8, low spin, sq. pl.) Temp. dependent spectra of CoL(CO2)+ in acetonitrile

νCO2 1710 1544

C

H N N

J. Am. Chem. Soc., 1991, 113, 343 Inorg. Chem., 1993, 32, 2657

H3C

O

O

O

Co

C

H N N

H

N N

CH3 H3C

O

Co S

N N

H

CH3

XANES of [CoL(CO2)]+ and [SCoL(CO2)]+

„

„

The [S-CoL(CO2)]+ edge shifts by 1.2 eV relative to that of [CoIIL]2+ This shift reveals significant metal-to-CO2 charge transfer and suggests formation of a Co(III) carboxylate, [S-CoIIIL(CO22−)]+ I

(a)

Normalized Absorptions

„

[Co L] can transfer two electrons to a bound CO2 7705

-

C Co

-

O

N N

N N

CH3 H3C

(b)

[CoII(CO2-)]+ [SCoIII(CO22-)]+

7715

7725

Energy (eV)

O C

H H

Sq. pl.

Oh. CoIII

CoII

CoII

+

O

Sq. pyr. CoI-CO

Co III Co S

N N

H CH3

J. Am. Chem. Soc., 1997, 119, 4549

Laser Flash Photolysis „

„ „ „

Sequential formation of the p-terphenyl radical anion (TP•−), CoIL+, CoL(CO2)+ and [SCoIIIL(CO22−)]+ are observed, indicating that CoIL+ can transfer two electrons to a bound CO2 ET from TP•− to CoL2+: fast, 1.1 x 1010 M-1 s-1 CO2 binding: k = 1.8 x 108 M-1 s-1 and KCO2 = 1 x 104 M-1 Slow thermal reactions including C-OH bond cleavage Time-resolved Spectrum of CoIL+ 3a 1a

0.012

Delta(Absorbance)

0.01

2

²A

10

1

0 400

500 Wavelength / nm

600

2

10

²A

10 ²A

0 0 1.0

Decay of TP•-

J. Am. Chem. Soc., 1995, 117, 6708

1.5 Time / µs

0

450

500

550

600

Wavelength / nm

5

0.5

0.004

0 400

3c

1

0

0.006

0.002

3b

Time-resolved spectrum of the terphenyl radical

700

0.008

0.5

1.0

1.5

Formation of CoIL+

Temperature-dependent transient spectra of CoL-CO2+

Reduction of CO2 to CO by Re(diimine)(CO)3+ Complexes

CO2 reduction by ReL(CO)3X ■ One- and two-electron pathways have been proposed ■ CO2 adducts have not been observed ReL(CO)3Cl

+ e­ - e­

[ReL(CO)3Cl] - e­ + Cl­

­

+ e­ -Cl ­

-Cl -

ReL(CO)3

CO + CO32­

ReL(CO)3(CO2 )

2e-+ CO 2

+ CO2

[ReL(CO)3]­

CO + [AO]­

+ CO2 [ReL(CO)3(CO2 )]­

B.P. Sullivan et al. JCS, Chem. Commun. 1985 F.P.A. Johnson et al. Organometallics 1996 O. Ishitani et al. JACS, 2008

e-+ A

Reactivity of ReL(CO)3 toward CO2 [Re(dmb)(CO)3]2 ■



CO2

2Re(dmb )(CO)3

Re(dmb)(CO)3 reacts with CO2 very slowly! kCO2 < 0.1 M-1 s-1

N NH

CO2 adduct ?

HN N

CoIL+ : kCO2 = 1.8 X 108 M-1 s-1 ■ ■ ■ ■ ■

[Re(dmb)(CO)3]2(13CO2) is the initial product The disappearance of [Re(dmb)(CO)3]2(CO2) is first order in [CO2]: k = 9.7 × 10-4 M-1s-1 Irradiation accelerates the reaction Final products: [Re(dmb)(CO)3]2(13CO3) and Re(dmb)(CO)3(O213COH) Yield of 13CO: 30 - 50 % based on [Re]

J. Am. Chem. Soc., 2003, 125, 11976

Calc. Structure

Observed Photocatalytic Reactions 2TEA

2*[Re(dmb)(CO) 3S]+

2TEA+ 2[Re(dmb)(CO)3S] CO2

hν 2[Re(dmb)(CO)3S]

+

CO + CO32- (or HCO3- )

-2S hν

[Re(dmb)(CO)3]2

2S [Re(dmb)(CO)3]2(CO2) CO2 +2S + hν

Rate-determining step Turnover frequency ~ 10 h−

J. Am. Chem. Soc., 2003, 125, 11976

Investigations in scCO2 ■ [CO2] up to 20 M (cf. ~ 100 mM/atm in conventional solvents) ■ Physical properties of scCO2 (incl. density, viscosity, dielectric constant) are easily tuned with pressure & temp. → selectivity ■ Facile recovery of products from the CO2 solvent – simply release the pressure

With D.C. Grills and M. Doherty

CnF2n+1

N OC OC

F2n+1Cn

N Re

Cl CO

Renewable Hydride Donors toward CO2 and Proton Reduction

CO2 Reduction by Renewable Hydrides The stoichiometric conversion of M−CO2 complexes to M−CO, M−CHO, M−CH2OH and M−CH3 has been previously accomplished by NaBH4 reduction M = Ru(bpy)2(CO)+, Re(Cp)(NO)(CO), etc. Can we replace NaBH4 by a renewable (visible-light­ generated) hydride donor?

H−, −OH−

H−

[M−CO2]+ → [M−CO]+ → M−CH(O)

H+, H−

→ M−CH2(OH)

OH

OH

OH OH O

N N

O

P O

N

NADPH and the Model

OH

OH O

P

O

O

O

N

N

NH2 H

NH2

H

O

„ NADPH acts as the source of two electrons and a proton „ Ru(bpy)2(pbn)2+, a complex with an NADP+ model ligand, catalyzes

the electochemical reduction of acetone to isopropanol: Current efficiency 90 % at -1.14 V vs. Fc/Fc+

„ The first example of electrochemical catalytic reduction of organic

molecules by NADP+/NADPH model complexes N

H

2+

N



2e , 2H

N

N [Ru]

N [Ru]

[Ru] = Ru(bpy)2

CH3CH(OH)CH3

2+

N

+

CH3COCH3

H

H

Koizumi and Tanaka, Angew. Chem. Int. Ed. 2005, 44, 5891

Photochemical and Radiolytic Reactions 2+

N N N

[Ru]

2+

H



N

Et3N/CH3CN

N N

[Ru]

H H

[Ru] = Ru(bpy)2

„ Hydrogenation of the pbn ligand 2.0

1.5

Absorption

takes place upon visible light irradiation (300-600 nm); detected by LCMS; Φ355 = 0.21 „ The x-ray structure of Ru(bpy)2(pbnHH)2+ has been determined „ The results open a new door to photoinduced catalytic hydridetransfer reactions !

After Irradiation

1.0

Before Irradiation 0.5

0.0 300

400

500

600

700

800

Wavelength, nm

Angew. Chem. Int. Ed. 2007, 46, 4169

Mechanism of Generation of [RuII(bpy)2(pbnHH)]2+ 2+

N N N Ru

1

+

N

e−

H

−H +

N



2

N Ru

H N

+

N N

Ru

3 k 8 = 1.2 X 108 M -1 s -1

1 H N



k 7a = 2.2 X 10 8 M -1 s -1

H H

N

1

N

Ru(bpy)2(pbnHH)2+ cannot reduce CO2 or M−CO

„

Theory indicates it is possible to produce stronger hydrides: add the energy of a third photon hν, e−

N

N

„

[RuII(bpy)2(pbnHH)]2+

[Ru]

H

2+

N N Ru

pK a = 11.0

3

H

3 + H+

The disproportionation reaction pools the energy of two photons in RuII(bpy)2(pbnHH)2+

2+

[RuII(bpy)2(pbnHH•-)]+ Inorg. Chem. 2008, 47, 3958

[Ru]

N H

4+ N

Summary „

CO2 is activated by two-electron pathways z Formation of a mononuclear carboxylate species (One M donates 2e− to CO2) z Formation of a binuclear species containing the M-C(O)O-M moiety (Each M center donates 1 e− to CO2)

O M

C

O M

O

C O

„

Slow dark reactions (C−H bond formation/C−O bond cleavage) can be accelerated using scCO2

„

While most photogenerated hydrides (C−H) are weaker than NADPH, theory indicates it is possible to produce stronger hydrides

„

By tuning the electron density of hydride donors, M−CO and M−CHO may be photochemically converted to M−CHO and M−CH2OH, respectively

M

Conclusions „ Solar energy is clearly the most promising source of renewable energy „ Success in these areas is difficult because the feedstock molecules are extremely stable „ Fundamental advances in coupling the solar photophysics of light absorption and charge separation to both oxidative and reductive catalytic reactions are required to achieve solar fuel generation „ Efficient, inexpensive, robust catalysts are needed for fuel generation

Acknowledgment James T. Muckerman (BNL) Dmitry Polyansky (BNL) Diane Cabelli (BNL) David C. Grills (BNL) Mark Doherty (BNL) Patrick Achord (BNL)

Muckerman

Polyansky

Cabelli

Koji Tanaka (IMS, Japan) Takeshi Fukushima (IMS, Japan) B. S. Brunschwig, N. Sutin, C. Creutz Y. Hayashi, T. Ogata, S. Yanagida, L. R. Furenlid, M. W. Renner, D. J. Szalda

Grills

Doherty

Achord

Department of Energy, BES DOE BES Solar Energy Utilization initiative (from 2007) Tanaka

Proposed Mechanism The product yields depend on the rates of the competing reactions → Kinetic and mechanistic studies are needed +

H

H+

MIIIL(H-)

MIL

?

CO2

CO2

MIIIL(CO22-) H+

-OH ­

MIIIL(H2)

MIIIL(-OOCH)

MIIIL(COOH- )

MIIIL + H2

MIIIL + HCOO-

MIIIL + CO + OH-

ML-CO

ML-CHO H- , H+

ML-CH2OH Comments Inorg. Chem. 1997, 19, 67

Suggest Documents