Visible-light-driven methane formation from CO2 with a molecular iron catalyst

Article abstract of DOI:10.1038/nature23016

Converting CO₂ into fuel or chemical feedstock compounds could in principle reduce fossil fuel consumption and climate-changing CO₂ emissions. One strategy aims for electrochemical conversions powered by electricity from renewable sources, but photochemical approaches driven by sunlight are also conceivable. A considerable challenge in both approaches is the development of efficient and selective catalysts, ideally based on cheap and Earth-abundant elements rather than expensive precious metals. Of the molecular photo- and electrocatalysts reported, only a few catalysts are stable and selective for CO₂ reduction; moreover, these catalysts produce primarily CO or HCOOH, and catalysts capable of generating even low to moderate yields of highly reduced hydrocarbons remain rare. Here we show that an iron tetraphenylporphyrin complex functionalized with trimethylammonio groups, which is the most efficient and selective molecular electro- catalyst for converting CO₂ to CO known, can also catalyse the eight-electron reduction of CO₂ to methane upon visible light irradiation at ambient temperature and pressure. We find that the catalytic system, operated in an acetonitrile solution containing a photosensitizer and sacrificial electron donor, operates stably over several days. CO is the main product of the direct CO₂ photoreduction reaction, but a two-pot procedure that first reduces CO₂ and then reduces CO generates methane with a selectivity of up to 82 per cent and a quantum yield (light-to-product efficiency) of 0.18 per cent. However, we anticipate that the operating principles of our system may aid the development of other molecular catalysts for the production of solar fuels from CO₂ under mild conditions.

Full text of DOI:10.1038/nature23016

LEttER
doi:10.1038/nature23016
Visible-light-driven methane formation from CO2
with a molecular iron catalyst
heng Rao1, Luciana C. Schmidt1,2, Julien Bonin1 & Marc Robert1
Converting CO2 into fuel or chemical feedstock compounds could generate the active Fe0 state. Using electron donors with high reducing
in principle reduce fossil fuel consumption and climate-changing ability should thus be favourable, and adding 0.2 mM of Ir(ppy)3
CO2 emissions1,2. One strategy aims for electrochemical conversions (4, where ppy is phenylpyridine; Table 1) as photosensitizer
powered by electricity from renewable sources3–5, but photochemical (E0(Ir(ppy)3+/Ir(ppy)3* ≈ −1.73 V versus the saturated calomel
approaches driven by sunlight are also conceivable6. A considerable electrode (SCE) and E0(Ir(ppy)3/Ir(ppy)3− ≈−2.19V versus SCE)23 to
challenge in both approaches is the development of efficient and the solution indeed enhanced the photocatalytic CO2 reduction so that
selective catalysts, ideally based on cheap and Earth-abundant 47 h of irradiation gave a turnover number in CO relative to 1 of 198
elements rather than expensive precious metals7. Of the molecular (entry 2 in Table 1 and Fig. 1b). Adding 0.1M trifluoroethanol (entry
photo- and electrocatalysts reported, only a few catalysts are stable 3 in Table 1) slightly increased the turnover number further to 240,
and selective for CO2 reduction; moreover, these catalysts produce probably owing to trifluoroethanol facilitating the C–O bond cleavage
primarily CO or HCOOH, and catalysts capable of generating even step. With the photosensitizer, products included not only CO but also
low to moderate yields of highly reduced hydrocarbons remain 10% hydrogen and 12% methane that correspond to turnover numbers
rare8–17. Here we show that an iron tetraphenylporphyrin complex of 24 and 31 (entry 2 in Table 1 and Fig. 1b). No other gaseous product
functionalized with trimethylammonio groups, which is the most was formed, and analysis of the liquid phase failed to detect methanol
efficient and selective molecular electro- catalyst for converting CO2 or formaldehyde by 1H NMR or formate (HCOO−) by ion chromatog-
to CO known18–20, can also catalyse the eight-electron reduction raphy. The presence of 0.1M of trifluoroethanol increased the selec-
of CO2 to methane upon visible light irradiation at ambient tivities (and turnover numbers) for H2 and CH4 to 19% (73) and 18%
temperature and pressure. We find that the catalytic system, (66), respectively (entry 3 in Table 1 and Fig. 1b). Blank experiments
operated in an acetonitrile solution containing a photosensitizer and (entries 1 and 5–8 in Table 1) confirmed that no methane is formed in
sacrificial electron donor, operates stably over several days. CO is the the absence of sensitizer, CO2, catalyst, electron donor or light.
main product of the direct CO2 photoreduction reaction, but a two-
In isotope labelling experiments conducted under a 12CO2 or a 13CO2
pot procedure that first reduces CO2 and then reduces CO generates atmosphere, gas chromatography/mass spectrometry analysis (Fig. 2)
methane with a selectivity of up to 82 per cent and a quantum yield identified as reaction product 12CH4 (m/z=16) or 13CH4 (m/z=17),
(light-to-product efficiency) of 0.18 per cent. However, we anticipate respectively, confirming that methane originates from CO2 reduction.
that the operating principles of our system may aid the development Increasing the irradiation time increased the amount of CO2 reduction
of other molecular catalysts for the production of solar fuels from products generated (Fig. 1c). The longest irradiation time of 102h
CO2 under mild conditions.
produced CO, CH4 and H2 with turnover numbers (and selectivities)
Iron tetraphenylporphyrins electrochemically reduced to the Fe0 of 367 (78%), 79 (17%) and 26 (5%), respectively (Table 1, entry 4 and
species have been shown to be the most efficient molecular catalysts Fig. 1c). These values correspond to a methane production rate of
for the CO2-to-CO conversion18,19. The nucleophilic Fe centre binds 763μmol per hour per gram of catalyst (μmol h−1 g−1), which exceeds
to CO2 and the Fe–CO2 adduct is further protonated and reduced to the rate of many other catalysts12–15,17,24 that generate methane from
afford CO upon cleavage of one C–O bond. Substitution of the four CO2. The linear evolution of both CO and CH4 over more than 80h
para-phenyl hydrogens by trimethylammonio groups18,20 (Fe-p-TMA 1, and the stable absorption spectrum of the system under irradiation
Table 1) led to CO formation with selectivity close to 95% in water at (Extended Data Fig. 1), with no evidence for degradation of the sensi-
pH 7 as well as in aprotic solvent such as N,N-dimethylformamide tizer 4 or catalyst 1, illustrate the stability of the catalytic system.
(DMF), at low overpotentials and with excellent stability (1 day). The
Evolution of the different products (Fig. 1c) shows that methane
fact that the standard redox potential of the FeI/Fe0 is not very negative production starts only after a large amount of CO has built up, suggest-
(E0 =−1.50V versus SCE in DMF)20 combined with the high intrinsic ing that CO is an intermediate in the methane formation process. We
activity towards CO2 reduction makes these catalysts good candidates have also previously shown by ultraviolet–visible spectroscopy22 that
for photochemical reduction of the gas.
irradiation of a CO2 saturated solution of 1 without a sensitizer (in that
Catalyst 1 was firstly used as a photocatalyst without a photosen- case, only CO is obtained) led to the formation of detectable amounts
sitizer under visible light irradiation (wavelength λ>420nm) with of FeIICO species. We may thus hypothesize that this iron–carbonyl
triethylamine (TEA, 50mM) as sacrificial electron donor. Illumination adduct is an intermediate for further reduction towards methane in
of a 1atm CO2-saturated solution of acetonitrile containing 2μM of the presence of a strong reducing agent. To explore the influence of
1 at room temperature for 47h selectively produced CO, with a turnover CO on methane formation more directly, experiments were then con-
number in CO relative to catalyst concentration of 33 (entry 1 in Table 1 ducted in a 1atm CO-saturated acetonitrile solution under visible light
and Fig. 1a). No side products were observed, and the linear production irradiation (λ>420nm), with 4 as sensitizer and TEA as sacrificial
of CO with time indicates good stability of the catalytic system.
electron donor (Fig. 1d). In a 47-h irradiation experiment, this slightly
A factor that can potentially limit the catalytic rate of this system21,22 lowered H2 production and increased CH4 production by almost a
is the three-electron reduction of the initial FeIII porphyrin species to factor of three compared to the experiment using a CO2-saturated
1Université Paris Diderot, Sorbonne Paris Cité, Laboratoire d’Electrochimie Moléculaire, UMR 7591 CNRS, 15 rue Jean-Antoine de Baïf, F-75205 Paris Cedex 13, France. 2INFIQC-CONICET,
Departamento de Química Orgánica, Facultad de Ciencias Químicas, Universidad Nacional de Córdoba, 5000 Córdoba, Argentina.
0 0 M o n t h 2 0 1 7
| V o L 0 0 0 | n A t U R E | 1
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.

reSeArCH letter
Table 1 | Catalytic performance and structures of catalysts and sensitizer
Turnover numbers
CH4
Entry
[1] (μM)
Gas
[4] (mM)
[TEA] mM
λ (nm)
Time (h)
CO
H2
1
2
2
2
2
2
-
2
2
2
CO2
CO2
CO2
CO2
Ar
CO2
CO2
CO2
CO2
-
50
50
50
50
50
50
-
>420
>420
>420
>420
>420
>420
>420
Dark
47
47
47
102
47
47
23
23
47
33
198
240
367
-
3
5
-
-
31
66
79
-
-
-
-
26
-
2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
24
73
26
43
1
-
-
15
3*
4
5
6
7
8
50
50
9†
>420
139
10
11
12*
13
14
2
2
2
-
CO
CO
CO
CO
CO
0.2
0.2
0.2
0.2
0.2
50
50
50
50
50
>420
>420
>420
>420
Dark
47
102
102
47
-
-
-
-
-
89
140
159
-
18
28
34
-
2
23
-
-
Summary of the reaction conditions used when evaluating the catalytic performance of catalysts 1 and 2. Above the table are shown the molecular structures of the iron-based catalysts Fe-p-TMA
1, Fe-o-OH 2 and FeTPP 3; and of the visible-light photosensitizer Ir(ppy)3 4.
*In the presence of 0.1 M trifluoroethanol.
†Catalyst 2.
solution: 83% of product was CH4 and 17% of product was H2 (entry did not give any reduction product (entries 13 and 14 in Table 1),
10 in Table 1), with the CH4 formation rate of 1,865μmol h−1 g−1
.
while a longer irradiation time of 102h enhanced the selectivity for
Blank experiments in the absence of 1 or in the absence of light methane further to 87% (entry 11 in Table 1 and Fig. 1d). Addition of a
0
5
10 15 20 25 30 35 40 45 50
0
5
10 15 20 25 30 35 40 45 50
35
30
25
20
15
10
5
a
250
200
150
100
50
b
H2
CO
CH4
0
0
350
d
c
150
125
100
75
50
25
0
300
250
200
150
100
50
0
0
10 20 30 40 50 60 70 80 90 100
0
10 20 30 40 50 60 70 80 90 100
Irradiation time (h)
Figure 1 | Photochemical reduction of CO2 under visible light
over an extended irradiation time in the presence of 0.2 mM of 4. d, Under
a CO atmosphere and with 0.2 mM of sensitizer 4 present, H2 and CH4 are
produced (filled symbols); adding 0.1 M of trifluoroethanol increases their
production rate (open symbols). Data points are the results of at least two
individual experiments and typical uncertainty on turnover numbers is
about 5%, corresponding to the size of the data points. Source data for this
figure is available in the online version of the paper.
irradiation. Shown is the formation of gaseous products (in terms of
turnover numbers) as a function of irradiation time, using an acetonitrile
solution saturated with 1 atm CO2 (a, b, c) or 1 atm CO (d) and containing
2 μM of catalyst 1 and 50 mM of TEA. a, With no sensitizer, only CO is
produced. b, When 0.2 mM of sensitizer 4 is present, H2, CO and CH4 are
produced (filled symbols); adding 0.1 M of trifluoroethanol increases their
production rate (open symbols). c, H2, CO and CH4 product evolution
2
| n A t U R E | V o L 0 0 0 | 0 0 M o n t h 2 0 1 7
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.

letter reSeArCH
5
4
Figure 2 | Methane detection. Typical gas
Under 12CO2
16
Under 13CO2
17
chromatogram observed during long-term
irradiation of a solution containing 2μM of catalyst 1,
50 mM of TEA and 0.2 mM of sensitizer 4, under
12CO2 or 13CO2 atmosphere. The inset shows the
mass spectra of methane generated under a 12CO2
or 13CO2 atmosphere.
100
90
80
70
60
50
40
30
20
10
0
100
90
80
70
60
50
40
30
20
10
0
16
15
3
15
2
14
O2–N2
12 13 14 15 16 17 18
12 13 14 15 16 17 18
1
m/z
m/z
CO
0
H2
CH4
–1
1.5
2.0
2.5
3.0
3.5
4.0
4.5
Retention time (min)
weak acid in moderate concentration (trifluoroethanol 0.1M) slightly potential value of the excited iridium complex (E0(4+/4*)≈−1.73V
increased the methane formation rate (from a turnover number of versus SCE), which is more negative than all three redox couples related
140 to 159) with some loss of selectivity (from 87% to 82%; entry 12 to the Fe porphyrin (FeIII/FeII, FeII/FeI and FeI/Fe0). After electron
in Table 1). The successful methane evolution under these conditions transfer, the oxidized 4+ is reduced by the sacrificial electron donor
over 102h with an average rate of 1,467μmol h−1 g−1 illustrate the TEA upon irradiation, thereby closing the catalytic cycle and generat-
robustness, activity and selectivity of the catalytic system.
ing the protonated triethylamine TEAH+ that could then act as proton
When replacing catalyst 1 by Fe-o-OH (2, Table 1), which carries –OH donor, as seen before27. Figure 3 sketches a plausible mechanism
groups at all ortho, ortho′ positions of the four phenyl rings25 instead based on these considerations, which involves a postulated formyl
of trimethylammonio groups at the para positions, methane was intermediate29,30 that may be stabilized by through-space interactions
also evolved although in slightly smaller amounts (turnover num- between the positive charges of the trimethylammonio groups and the
ber 26 after 47 h irradiation and 14% selectivity, Table 1, entry 9). partial negative charge on the CHO species bound to the metal. With
The standard redox potential E0(FeI/Fe0)=−1.575V versus SCE26 in complete reduction of the FeIICO adduct necessitating six electrons, the
DMF for catalyst 2 is only 75mV more negative than for 1, and as in quantum yield for CH4 formation is Φ=0.18% (see Methods).
the latter case, the substituents on the phenyls may help stabilizing
reaction intermediates (through internal H bonds involving the –OH
hυ
N
N
groups). In contrast, the non-substituted tetraphenyl Fe porphyrin
(FeTPP 3, Table 1) only gives CO and H2 (with turnover numbers/
selectivities of 84/79% and 22/21%, respectively) under the same
irradiation conditions27, probably owing to its much more negative
standard redox potentials (for example, E0(FeI/Fe0)=−1.67V versus
SCE in DMF)28 and the absence of phenyl ring substituents for stabilizing
intermediate species involved in hydrocarbon production. The ability
to produce methane is thus not restricted to catalyst 1, but is probably a
more general property of Fe porphyrins that have a sufficiently positive
standard redox potential and are functionalized with substituents that
can stabilize intermediates involved in the catalytic cycle.
N
Fe
N
*
Ir(ppy)3
Ir(ppy)3
Cl
N
N
hυ
TEA
N
N
(×2)
+
Ir(ppy)3
hυ
FeI
2e–, H+
FeIICO
*
Ir(ppy)3
Ir(ppy)3
CO
+ H2O
hυ
TEA
+
1e–, 2H+
–
Ir(ppy)3
Another key parameter for CO2 reduction beyond the two-elec-
tron production of CO is the driving force for charge transfer from
the excited state of the sensitizer. When replacing 4 by the less-reduc-
FeICHO
FeII
FeICO2
Fe0
ing ruthenium complex Ru(bpy)32+ (where bpy is 2,2′-bipyridine;
E0(Ru(bpy)32+/Ru(bpy)3+)≈−1.33V versus SCE and E0(Ru(bpy)3
/
3+
4e–, 5H+
CO2
CH4
+ H2O
Ru(bpy)32+*)=−0.81V versus SCE)23, only CO and H2 and no CH4
were obtained, possibly because the Ru excited state or its reduced
form are not able to trigger the carbonyl reduction from the FeIICO
adduct. Emission quenching experiments between the excited state of
the sensitizer 4* and 1 on one hand and 4* and TEA on the other hand
revealed very weak quenching with TEA, while it is very efficient and
Figure 3 | Sketch of the proposed mechanism for CO2 reduction to CH4
by catalyst 1. Initially, the starting FeIII porphyrin (top left) is reduced
with three electrons to the catalytically active Fe0 species. The Fe0 species
reduces CO2, with the resultant FeI regenerated through electron transfer
from the excited photosensitizer (right-hand side cycle). The CO produced
binds to FeII and is further reduced with a total of six electrons (transferred
from the excited sensitizer) and six protons to generate methane, via a
postulated FeI-formyl (FeICHO) intermediate (left-hand side cycle). For
more detailed discussion, see text. hυ, light irradiation.
diffusion-controlled with 1 (kq ≈1.7×1010
M
−1 s−1, Extended Data
Fig. 2), suggesting that direct electron transfer occurs from the excited
sensitizer 4* to the Fe porphyrin. This is in line with the standard redox
0 0 M o n t h 2 0 1 7
| V o L 0 0 0 | n A t U R E | 3
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.

reSeArCH letter
18. Azcarate, I., Costentin, C., Robert, M. & Savéant, J.-M. A study of through-space
charge interaction substituent effects in molecular catalysis leading to the
design of the most efficient catalyst of CO2-to-CO electrochemical conversion.
J. Am. Chem. Soc. 138, 16639–16644 (2016).
19. Bonin, J., Maurin, A. & Robert, M. Molecular catalysis of the electrochemical
and photochemical reduction of CO2 with Fe and Co metal based complexes.
Recent advances. Coord. Chem. Rev. 334, 184–198 (2017).
20. Costentin, C., Robert, M., Savéant, J.-M. & Tatin, A. Efficient and selective
molecular catalyst for the CO2-to-CO electrochemical conversion in water.
Proc. Natl Acad. Sci. USA 112, 6882–6886 (2015).
We anticipate that further spectroscopic investigation, in conjunc-
tion with quantum calculations, will help decipher in greater detail the
reduction mechanism at play. This insight should aid the development
of more efficient catalytic systems that make use of Earth-abundant
Fe-based molecular complexes to reduce CO2 into CO and then into
CH4 under mild conditions and driven by visible light.
Online Content Methods, along with any additional Extended Data display items and
Source Data, are available in the online version of the paper; references unique to
these sections appear only in the online paper.
21. Bonin, J., Robert, M. & Routier, M. Selective and efficient photocatalytic CO2
reduction to CO using visible light and an iron-based homogeneous catalyst.
J. Am. Chem. Soc. 136, 16768–16771 (2014).
22. Rao, H., Bonin, J. & Robert, M. Non-sensitized selective photochemical
reduction of CO2 to CO under visible light with an iron molecular catalyst.
Chem. Commun. 53, 2830–2833 (2017).
received 6 January; accepted 6 June 2017.
Published online 17 July 2017.
23. Prier, C. K., Rankic, D. A. & MacMillan, D. W. C. Visible light photoredox catalysis
with transition metal complexes: applications in organic synthesis. Chem. Rev.
113, 5322–5363 (2013).
24. Wang, H., Chen, Y., Hou, X., Ma, C. & Tan, T. Nitrogen-doped graphenes as
efficient electrocatalysts for the selective reduction of carbon dioxide to
formate in aqueous solution. Green Chem. 18, 3250–3256 (2016).
25. Costentin, C., Drouet, S., Robert, M. & Savéant, J.-M. A local proton source
enhances CO2 electroreduction to CO by a molecular Fe catalyst. Science 338,
90–94 (2012).
26. Costentin, C., Passard, G., Robert, M. & Savéant, J.-M. Pendant acid-base
groups in molecular catalysts: H-bond promoters or proton relays?
Mechanisms of the conversion of CO2 to CO by electrogenerated iron(0)
porphyrins bearing prepositioned phenol functionalities. J. Am. Chem. Soc.
136, 11821–11829 (2014).
27. Bonin, J., Chaussemier, M., Robert, M. & Routier, M. Homogeneous
photocatalytic reduction of CO2 to CO using iron(0) porphyrin catalysts:
mechanism and intrinsic limitations. ChemCatChem 6, 3200–3207
(2014).
1. Jhong, H.-R. M., Ma, S. & Kenis, P. J. A. Electrochemical conversion of CO2 to
useful chemicals: current status, remaining challenges, and future
opportunities. Curr. Opin. Chem. Eng. 2, 191–199 (2013).
2. Aresta, M., Dibenedetto, A. & Angelini, A. Catalysis for the valorization of
exhaust carbon: from CO2 to chemicals, materials, and fuels. Technological use
of CO2. Chem. Rev. 114, 1709–1742 (2014).
3. Qiao, J., Liu, Y., Hong, F. & Zhang, J. A review of catalysts for the
electroreduction of carbon dioxide to produce low-carbon fuels. Chem. Soc.
Rev. 43, 631–675 (2014).
4. Parajuli, R. et al. Integration of anodic and cathodic catalysts of Earth-
abundant materials for efficient, scalable CO2 reduction. Top. Catal. 58,
57–66 (2015).
5. Tatin, A. et al. Efficient electrolyzer for CO2 splitting in neutral water using
Earth-abundant materials. Proc. Natl Acad. Sci. USA 113, 5526–5529
(2016).
6. Sahara, G. & Ishitani, O. Efficient photocatalysts for CO2 reduction. Inorg. Chem.
54, 5096–5104 (2015).
28. Bhugun, I., Lexa, D. & Savéant, J.-M. Homogeneous catalysis of electrochemical
hydrogen evolution by iron(0) porphyrins. J. Am. Chem. Soc. 118, 3982–3983
(1996).
29. Appel, A. M. et al. Frontiers, opportunities, and challenges in biochemical
and chemical catalysis of CO2 fixation. Chem. Rev. 113, 6621–6658
(2013).
7. Takeda, H., Cometto, C., Ishitani, O. & Robert, M. Electrons, photons, protons
and Earth-abundant metal complexes for molecular catalysis of CO2 reduction.
ACS Catal. 7, 70–88 (2017).
8. Shen, J. et al. Electrocatalytic reduction of carbon dioxide to carbon monoxide
and methane at an immobilized cobalt protoporphyrin. Nat. Commun. 6, 8177
(2015).
30. Davies, S. G., Hibberd, J. & Simpson, S. J. Disproportionation of the iron
carbonyl hydride (n5-C5H5)Fe(CO)H(Ph2PCH2CH2PPh2) to the iron methyl
(n5-C5H5)Fe(Ph2PCH2CH2PPh2)Me. J. Chem. Soc. Chem. Commun.1404–1405
(1982).
9. Weng, Z. et al. Electrochemical CO2 reduction to hydrocarbons on a
heterogeneous molecular Cu catalyst in aqueous solution. J. Am. Chem. Soc.
138, 8076–8079 (2016).
10. Manthiram, K., Beberwyck, B. J. & Alivisatos, A. P. Enhanced electrochemical
methanation of carbon dioxide with a dispersible nanoscale copper catalyst.
J. Am. Chem. Soc. 136, 13319–13325 (2014).
11. Xie, M. S. et al. Amino acid modified copper electrodes for the enhanced
selective electroreduction of carbon dioxide towards hydrocarbons. Energy
Environ. Sci. 9, 1687–1695 (2016).
12. Wu, T. et al. A carbon-based photocatalyst efficiently converts CO2 to CH4 and
C2H2 under visible light. Green Chem. 16, 2142–2146 (2014).
13. AlOtaibi, B., Fan, S., Wang, D., Ye, J. & Mi, Z. Wafer-level artificial photosynthesis
for CO2 reduction into CH4 and CO using GaN nanowires. ACS Catal. 5,
5342–5348 (2015).
14. Liu, X., Inagaki, S. & Gong, J. Heterogeneous molecular systems for
photocatalytic CO2 reduction with water oxidation. Angew. Chem. Int. Ed. 55,
14924–14950 (2016).
15. Wang, Y. et al. Facile one-step synthesis of hybrid graphitic carbon nitride and
carbon composites as high-performance catalysts for CO2 photocatalytic
conversion. ACS Appl. Mater. Interf. 8, 17212–17219 (2016).
16. Zhu, S. et al. Photocatalytic reduction of CO2 with H2O to CH4 over ultrathin
SnNb2O6 2D nanosheets under visible light irradiation. Green Chem. 18,
1355–1363 (2016).
Acknowledgements This work was partially supported by the CNRS, Défi
Transition énergétique “Emergence CO2” (the PERIODIC project). H.R. thanks
the China Scholarship Council for a PhD fellowship (CSC student number
201507040033). We thank D. Clainquart (Université Paris Diderot) for
assistance in gas chromatography/mass spectrometry analysis and I. Azcarate
for porphyrin synthesis.
Author Contributions M.R. conceived the research, J.B. and M.R. directed the
project and co-wrote the paper. J.B. conceived the experimental setup. J.B.,
L.C.S. and H.R. conducted experiments. H.R., J.B. and M.R. analysed results.
All the authors contributed to the scientific interpretation and reviewed the
manuscript.
Author Information Reprints and permissions information is available at
www.nature.com/reprints. The authors declare no competing financial
interests. Readers are welcome to comment on the online version of the
paper. Publisher’s note: Springer Nature remains neutral with regard
to jurisdictional claims in published maps and institutional affiliations.
Correspondence and requests for materials should be addressed to
J.B. (julien.bonin@univ-paris-diderot.fr) and M.R. (robert@univ-paris-diderot.fr).
17. Yu, L. et al. Enhanced activity and stability of carbon-decorated cuprous oxide
mesoporous nanorods for CO2 reduction in artificial photosynthesis. ACS Catal.
6, 6444–6454 (2016).
4
| n A t U R E | V o L 0 0 0 | 0 0 M o n t h 2 0 1 7
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.

letter reSeArCH
MeThOds
Ultra gas chromatograph equipped with a CP 7514 column (Agilent Technologies)
Synthesis of catalysts 1 and 2. The synthesis of chloro iron(iii) 5,10,15, and coupled to a DSQ II mass spectrometer in positive ionization mode, using a
20-tetra(40-N,N,N-trimethylanilinium)porphyrin (Fe-p-TMA, 1)20 and chloro TriPlus headspace autosampler.
iron(iii) 5,10,15,20-tetrakis(2′,6′-dihydroxyphenyl) porphyrin (Fe-o-OH, 2)25 have Turnover number calculation. Turnover number is practically defined as the
been described elsewhere.
number of catalytic cycles per catalyst amount. The number of moles of H2, CO and
Photochemical measurements. Irradiations of acetonitrile (99.9% extra-dry, CH4 was determined by converting peak integrations from gas chromatography
Acros Organics) solutions containing triethylamine (99% pure, Acros Organics) as measurements into moles in the sample headspace by using individual calibration
sacrificial electron donor, and fac-(tris-(2-phenylpyridine))iridium(iii) (Ir(ppy)3, 4; curves and taking into account the irradiated sample volume (3.5ml).
99%, Aldrich) as sensitizer were realized in a closed 1cm×1cm quartz suprasil Quantum yield calculation. The number of incident photons was measured using
cuvette (Hellma 117.100F-QS) equipped with home-designed headspace glassware. the classical iron ferrioxalate (K3Fe(C2O4)3) chemical actinometer, following the
Solutions were saturated with argon (>99.998%, Air Liquide), 12CO2 (>99.7%, Air procedure reported previously32 and using known parameters for calculations33.
Liquide), 13CO2 (99 atom% 13C, Aldrich) or 12CO (> 99.997%, Air Liquide) for Using three independent measurements, we determined that the number of
20 min before irradiation. A Newport LCS-100 solar simulator, equipped with an incident photons to the sample was (2.18 0.17)×1019 photons per hour. The
AM1.5 G standard filter allowing 1 Sun irradiance, was used as the light source CO-to-CH4 reduction being a six-electron process, the overall quantum yield Φ
combined with a Schott GG420 longpass filter and 2-cm-long glass (OS) cell filled of the process was determined using the following equation:
with deionized water to prevent catalyst absorbance and to cut off infrared and
Number of CH4 molecules formed × 6
low ultraviolet.
ΦCH (%) =
× 100
4
Number of incident photons
Spectrophotometric measurements. Ultraviolet–visible absorption data were
collected with an Analytik Jena Specord 600 ultraviolet–visible spectrophotometer.
Emission quenching measurements were conducted with a Cary Eclipse fluores-
cence spectrophotometer (Agilent Technologies), with the excitation wavelength
set at 420nm and the emission spectrum measured between 430nm and 700nm.
Emission intensities used for the Stern–Volmer analysis were taken at 517nm, that
is, the emission maximum of 4. The lifetime of the emissive excited state of 4 was
taken as 1.9μs, as reported before31.
Taking 159 as the highest turnover number for CH4 (Table 1, entry 12), we obtain
a quantum yield Φ of about 0.18% after 102h of irradiation.
Data availability. The data that support the findings of this study are available
from the corresponding authors upon reasonable request. Original experimental
data that support the findings of this study are available from the corresponding
authors upon reasonable request. Source data for Fig. 1 is available in the online
version of the paper.
Reduction products analysis. Gaseous products analysis was performed with an
Agilent Technology 7820A gas chromatography system equipped with a capil-
lary column (CarboPLOT P7, length 25m, inner diameter 25mm) and a thermal
conductivity detector. Calibration curves for H2, CO and CH4 were established
separately. Control experiments, with no catalyst, no CO2 or no light were
conducted under the same conditions otherwise as the full system. Ionic chroma-
tography measurements were performed with a Thermo Scientific Dionex ICS-
1100 system. Mass spectra were obtained by a ThermoFisher Scientific TRACE
31. Dedeian, K., Djurovich, P. I., Garces, F. O., Carlson, G. & Watts, R. J. A new
synthetic route to the preparation of a series of strong photoreducing agents:
fac-tris-ortho-metalated complexes of iridium(III) with substituted
2-phenylpyridines. Inorg. Chem. 30, 1685–1687 (1991).
32. Alsabeh, P. G. et al. Iron-catalyzed photoreduction of carbon dioxide to
synthesis gas. Catal. Sci. Technol. 6, 3623–3630 (2016).
33. Montalti, M., Credi, A., Prodi, L & Gandolfi, M. T. Handbook of Photochemistry 3rd
edn (CRC Press, 2006).
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.

reSeArCH letter
Extended Data Figure 1 | Evolution of the absorption spectrum
with time. The absorption spectrum of a CO2-saturated acetonitrile
solution containing 2 μM of 1, 0.2 mM of 4, 0.05 M of TEA upon visible
(>420 nm) light irradiation remains stable over the course of experiments,
highlighting the stability of the system. The inset shows the absorption
spectrum of 2 μM of catalyst 1 in acetonitrile (no sensitizer 4), revealing
that in the photocatalytic mix, >90% of photons above 420 nm are
absorbed by 4.
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.

letter reSeArCH
Extended Data Figure 2 | Sensitizer 4 emission quenching after
excitation at 420 nm. a, Upon increasing concentration of TEA in a
0.1 mM acetonitrile solution of 4, no emission quenching is observed,
as confirmed by the Stern–Volmer analysis (inset). b, Upon increasing
concentration of 1 in a 0.2 mM acetonitrile solution of 4, emission
quenching is observed corresponding to a diffusion-controlled quenching
rate of (1.7 0.1) ×1010
M
−1 s−1 as determined by Stern–Volmer analysis
(inset). a.u., arbitrary units. I0/I is the emission intensity without quencher
divided by the emission intensity with a known concentration of quencher.
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.