SELECTIVE PORPHYRIN-CATALYZED ELECTROCHEMICAL REDUCTION OF CO2 INTO CO, IN PARTICULAR IN WATER

Abstract

The present invention relates to the use of complexes of water soluble porphyrins of formula (I). The present invention relates to water soluble porphyrins of formula (I), wherein R.sub.1 to R.sub.11 and R.sub.1 to R.sub.8 are as defined in claim 1, their iron complexes, use thereof as catalysts for the selective electrochemical reduction of CO.sub.2 into CO, electrochemical cells comprising them, and methods for reducing electrochemically CO.sub.2 into CO using said complexes or said electrochemical cells, thereby producing CO or syngas.

##STR00001##

Claims

1.-15. (canceled)

16. An iron complex of a porphyrin of formula (I): ##STR00013## wherein R.sub.1 to R.sub.8 and R.sub.1 to R.sub.8 are independently selected from the group consisting of H, OH, F, and C.sub.1-C.sub.6-alkyl, provided that at least 2 of R.sub.1-R.sub.4 are H, at least 2 of R.sub.1-R.sub.4 are H, at least 2 of R.sub.5-R.sub.8 are H, and at least 2 of R.sub.5-R.sub.8 are H, and R.sub.9, R.sub.10 and R.sub.11 are independently selected from a C.sub.1-C.sub.4-alkyl group, or atropisomers thereof, and salts thereof.

17. The iron complex of claim 16, wherein R.sub.9, R.sub.10, and R.sub.11 are a methyl group.

18. The iron complex of claim 16, wherein R.sub.1 to R.sub.8 and R.sub.1 to R.sub.8 are selected from the group consisting of H and C.sub.1-C.sub.6-alkyl.

19. The iron complex of claim 16, wherein the porphyrin of the invention is of formula (Ia): ##STR00014## with R.sub.1 to R.sub.4 and R.sub.9, R.sub.10, and R.sub.11 as defined in claim 16, or atropisomers thereof, and salts thereof.

20. The iron complex of claim 16, wherein the porphyrin of the invention is of formula (Ib): ##STR00015## with R.sub.9, R.sub.10, and R.sub.11 as defined in claim 16, or atropisomers thereof, and salts thereof.

21. The iron complex of claim 20, wherein the porphyrin of the invention is: ##STR00016## or atropisomers thereof, and salts thereof.

22. A two-compartment electrochemical cell comprising at least: a cathodic compartment with a cathode and a cathodic electrolyte solution comprising a cathodic solvent and a cathodic supporting electrolyte, and the substrate CO.sub.2, an anodic compartment with an anode and an anodic electrolyte solution comprising a solvent and an anodic supporting electrolyte, a power supply providing the energy necessary to trigger the electrochemical reactions involving the substrate, and further comprising the iron complex of claim 16.

23. The electrochemical cell of claim 22, wherein CO.sub.2 gas is present only in the cathodic compartment and the CO.sub.2 pressure in the cathodic compartment of the electrochemical cell is of between 1 bar and 30 bars.

24. The electrochemical cell of claim 22, wherein the solvent of the electrolyte solutions is selected from dimethylformamide, acetonitrile, water, or mixtures thereof.

25. The electrochemical cell of claim 24, wherein, when the solvent is water, the cathodic supporting electrolyte is devoid of buffer, and the anodic supporting electrolyte comprises a phosphate buffer.

26. The electrochemical cell of claim 24, wherein, when the solvent comprises dimethylformamide and/or acetonitrile, said solvent further comprises a proton donor with a pKa value in DMF of between 18 and 31.

27. A method of reducing electrochemically CO.sub.2 into CO using an iron complex of a porphyrin of formula (I): ##STR00017## wherein R.sub.1 to R.sub.8 and R.sub.1 to R.sub.8 are independently selected from the group consisting of H, OH, F, and C.sub.1-C.sub.6-alkyl, provided that at least 2 of R.sub.1-R.sub.4 are H, at least 2 of R.sub.1-R.sub.4 are H, at least 2 of R.sub.5-R.sub.8 are H, and at least 2 of R.sub.5-R.sub.8 are H, and R.sub.9, R.sub.10 and R.sub.11 are independently selected from a C.sub.1-C.sub.4-alkyl group, or atropisomers thereof, and salts thereof with iron at the oxidation state of Fe(0), as catalyst or the electrochemical cell of claim 24.

28. The method of claim 27, wherein, in the electrochemical cell, when a mixture of CO and H.sub.2 (syngas) is produced, the pH of the aqueous solution and the potential applied to the cathode are adjusted so as to tune the CO/H.sub.2 molar ratio of the produced gas.

29. The method of claim 27, wherein the iron complex of claim 1 is used as homogenous catalyst or is immobilized on at least one electrode using a binder, optionally containing conductive materials as additives.

30. The electrochemical cell of claim 26, wherein the proton donor is selected from the group consisting of water, trifluoroethanol, phenol, and acetic acid.

Description

DESCRIPTION OF THE DRAWINGS

[0144] FIG. 1 represents the synthetic schema of Fe-o-TMA as used in example 1.

[0145] FIG. 2 represents cyclic voltammograms of the iron porphyrins Fe-o-TMA, Fe-p-TMA, Fe-p-PSULF, FeTPP, FeF5TPP, FeF10TPP and FeF20TPP (see below) in the potential domain of the catalytic CO.sub.2 reduction wave in DMF+0.1 M n-Bu.sub.4NPF.sub.6+0.1 M H.sub.2O+3 M PhOH, at 0.1V/s under 1 atmosphere of CO.sub.2 The current, i, is normalized against the peak current of the reversible one-electron Fe.sup.II/Fe.sup.I reversible wave, i.sub.p.sup.0 obtained at the same scan rate (0.1 V/s). The abscissa axis represents the potential applied to the cathode vs SHE (in V), and the ordinate axis represents the normalized current.

##STR00010##

[0146] FIG. 3. Cyclic voltammetry of Fe-o-TMA, Fe-p-TMA and Fe-p-PSULF (concentration: 1 mM) in DMF+0.1 M n-Bu.sub.4NPF.sub.6+0.1 M H.sub.2O at 0.1 V/s under argon (grey) and under 1 atm. CO.sub.2 in the presence of 3 M PhOH (black). The bottom figures are a blow-up of the upper figures showing the formation of CO on the reverse scan (see example 2). The current axis is normalized toward the peak current i.sub.p.sup.0 of a one-electron reversible wave at the same concentration at same scan rate (0.1 V/s) as can be obtained from the Fe.sup.II/I wave. The abscissa axis represents the potential applied to the cathode vs SHE (in V), and the ordinate axis represents the normalized current.

[0147] FIG. 4. Elimination of the secondary phenomena by raising the scan rate (see Example 2). Cyclic voltammetry of the substituted iron porphyrins (conc.: 1 mM) in DMF+0.1 M n-Bu.sub.4NPF6+0.1 M H.sub.2O under 1 atm. CO.sub.2 in the presence of 3M PhOH at scan rates (V/s), from bottom to top: Fe-p-TMA: 0.1, 1, 5, 10; Fe-p-PSULF: 0.1, 1, 2; Fe-o-TMA: 0.1, 0.5, 1, 2, 6, 10, 20, 30, 48, 96, 115. The current, i, is normalized against the peak current of the reversible one-electron Fe.sup.II/Fe.sup.I reversible wave, i.sub.p.sup.0 obtained at 0.1 V/s. The abscissa axis represents the potential applied to the cathode vs SHE (in V), and the ordinate axis represents the normalized current.

[0148] FIG. 5. a: Catalytic Tafel plots of Fe-o-TMA, Fe-p-TMA, Fe-p-PSULF, FeTPP, FeF5TPP, FeF10TPP and FeF20TPP. b: correlation between TOF.sub.max=k.sub.cat and E.sub.cat.sup.0, recalling the through-structure substituents effect and showing the coulombic interaction effects of positively and negatively charged substituents, including the huge gain in reactivity and overpotential brought about by positively charged substituents when located in ortho position of the phenyl ring.

[0149] FIG. 6. depicts the benchmarking of all catalysts based on catalytic Tafel plots derived from cyclic voltammetry experiments in DMF or acetonitrile (see Costentin et al Proc. Natl. Acad. Sci. U.S.A. 2014, 111, 14990-14994 for details and references). The abscissa axis represents the overpotential (in Volts), and the ordinate axis represents log(TOF).

[0150] See example 3 for experimental details.

[0151] a: example 3; b: in the presence of 3M PhOH, c: see structure above.

[0152] py=pyridine, tpy=2,2:6,2-terpyridine, bpy=2,2-bipyridine, Mebimpy=2,6-bis(1-methyl benzimidazol-2-yl)pyridine

##STR00011##

[0153] The vertical arrows indicate the overpotential values at which the electrolysis were carried out (see text and FIG. 8).

[0154] FIG. 7. Preparative-scale electrochemical CO.sub.2-to-CO conversion catalyzed by Fe-p-TMA (at an overpotential of 450 mV, vertical arrow in FIG. 7) and Fe-o-TMA (at an overpotential of 220 mV, vertical arrow in FIG. 7) in DMF+0.1 M n-Bu.sub.4NPF.sub.6+0.1 M H.sub.2O under 1 atm. CO.sub.2 in the presence of 3M PhOH (catalyst conc.: 0.5 mM). CO (full circle) and H.sub.2 (open circle) faradaic yields (top) and CO-vs-H.sub.2 selectivity (bottom), defined as: % CO/(% CO+% H.sub.2).

[0155] FIG. 8. Long run (84 hours) preparative-scale electrochemical CO.sub.2-to-CO conversion catalyzed by Fe-o-TMA (at an overpotential of 220 mV, vertical arrow in FIG. 7) in DMF+0.1 M n-Bu.sub.4NPF.sub.6+0.1 M H.sub.2O under 1 atm. CO.sub.2 in the presence of 3 M PhOH (catalyst concentration: 0.5 mM). a. Charge evolution and b. Faradaic yields (CO: full symbols, H.sub.2: open symbols) as a function of time during 24 h on a carbon electrode (circles), then 18 h on a mercury electrode (squares), then a second 18 h period again on a mercury electrode (diamonds) and finally another 24 h period again on a carbon electrode (triangles).

[0156] a. The abscissa axis represents the time (in hours), and the ordinate axis represents the charged passed (in Coulomb). b. The abscissa axis represents the time (in hours), and the ordinate axis represents the faradaic yield (in %).

[0157] FIG. 9. Cyclic voltammetry of Fe-p-TMA, and Fe-p-TMA (concentration: 0.5 mM) in H.sub.2O+0.1 M KCl at 0.1 V/s under argon (grey) and under 1 atm. CO.sub.2+0.5M KHCO.sub.3 (black) at pH 7.1. The bottom figures are a blow-up of the upper figures. The abscissa axis represents the potential applied to the cathode vs SHE (in V), and the ordinate axis represents the current (in A).

[0158] FIG. 10. Long run (60 hours) preparative-scale electrochemical CO.sub.2-to-CO conversion catalyzed by Fe-o-TMA (at an applied potential of 0.86V, 0.91V and 0.96V vs. NHE) in H.sub.2O+0.1 KCl+0.5M KHCO.sub.3 under 1 atm. CO.sub.2 (catalyst concentration: 0.5 mM). Left ordinate axis represents the charge evolution (in Coulomb) and right ordinate axis represents the selectivity for CO (black squares) and H.sub.2 (open circles) (in %) as a function of time during 60 h on a carbon electrode.

[0159] FIG. 11. Preparative-scale electrochemical CO2-to-CO conversion catalyzed by Fe-p-TMA (at an applied potential of 0.96V vs. NHE, grey line) and Fe-o-TMA (at an applied potential of 086V, 0.91V and 0.96V vs. NHE, black line) in H.sub.2O+0.1 KCl+0.5M KHCO.sub.3 under 1 atm. CO.sub.2 (catalyst concentration: 0.5 mM). CO (red dotsfull circle) and H.sub.2 (blue open dotscircle) faradaic yields (top) and CO-vs-H.sub.2 selectivity (bottom), defined as: % CO/(% CO+% H.sub.2). The abscissa axis represents the time (in hours), and the ordinate axis represents the current (in mA).

EXAMPLES

[0160] The following examples, while relating to particular embodiments of the invention, are intended for purposes of illustration only and are not intended to limit the scope of the invention.

[0161] As used herein, TOF.sub.0 represents the TurnOver Frequency at zero overpotential. The value of TOF.sub.0 is obtained from extrapolation of the TOF vs. overpotential curve at zero overpotential. The TOF vs. overpotential curve is obtained from the experimental measurement of the current density (I) as function of potential (E) using cyclic voltammetry. For example, in the case of a simple mechanism (i.e. if the chemical steps in the catalytic loop are equivalent to a single step characterized by an apparent catalytic constant) the following relationship can be used:

[00001]$\mathrm{TOF}=\frac{I}{F.Math.\sqrt{\frac{D}{k_{\mathrm{cat}}}}.Math.C_{\mathrm{cat}}^0}$

with D being the diffusion coefficient of the catalyst, C.sub.cat.sup.0 at being its concentration in solution and k.sub.cat the catalytic rate constant. The value of TOF.sub.0 is preferably obtained from extrapolation of the TOF vs. overpotential curve at zero overpotential. Said TOF vs. overpotential curve is for instance obtained such as described in Costentin et al ChemElectroChem, 2014, 1, 1226-1236, or calculated as detailed in Costentin et al, Science 2012 338, 90.

Example 1. Synthesis and Characterization of Fe-o-TMA

[0162] Chemicals.

[0163] Dimethylformamide (Acros, >99.8%, extra dry over molecular sieves), the supporting electrolyte NBu.sub.4PF.sub.6 (Fluka, purriss.) were used as received. All starting materials were obtained from Sigma-Aldrich, Fluka, Acros and Alfa-Aesar, and were used without further purification. CHCl.sub.3 and CH.sub.2Cl.sub.2 were distilled from calcium hydride and stored under an argon atmosphere.

[0164] Materials.

[0165] .sup.1H NMR spectra were recorded on a Bruker Avance III 400 MHz spectrometer and were referenced to the resonances of the solvent used. The mass spectra were recorded on a MALDI 4800 TOF/TOF. The elemental analysis were performed by the Microanalysis Service of the Institut de Chimie des Substances Naturelles (ICSN-CNRS), Avenue de la Terrasse, 91198 Gif-sur-Yvette cedex, France.

[0166] Porphyrins.

[0167] Iron(III) 5,10,15,20-tetrakis(4-N,N,N-trimethylanilinium)-porphyrin pentachloride (Fe-p-TMA) was prepared as described elsewhere in Costentin et al. (Proc. Natl. Acad. Sci. U.S.A., 2015, 112, 6882).

[0168] Chloro iron(III) 5,10,15,20-tetrakis(4-sulfonatophenyl)porphyrin (tetrabutylammonium salt form) (Fe-p-PSULF) was prepared in situ by the deprotonation of commercial Chloro iron(III) 5,10,15,20-tetrakis(4-sulfonatophenyl)porphyrin (acid form) (LivChem Logistic) with 4 eq. of TBAOH.

##STR00012##

Synthesis of Chloro iron(III) 5,10,15,20-tetra-(o-N,N,N-trimethylanilium)-porphyrin tetra trifluoromethanesulfonate (Fe-o-TMA)

5,10,15,20-tetra(o-nitrophenyl)-21H,23H-porphyrin. (1)

[0169] Compound 1 was synthesized following a previously reported procedure by Collman et al. (J. Am. Chem. Soc. 1975, 97, 1427-1439).

[0170] Pyrrole (4.62 mL; 66.17 mmol, 1 eq) was added dropwise to a solution of 2-nitrobenzaldehyde (10 g, 66.2 mmol, 1 eq.) in boiling acetic acid (200 mL). After 20 min of stirring at reflux, the solution was cooled in an ice bath and chloroform (25 mL) was added. The mixture was filtered on a glass frit and the purple powder was thoroughly washed with chloroform and methanol until filtrates were colorless. The product was recovered and dried in the oven (1.8 g, 13.6%)

,-5,10,15,20-tetra(o-aminophenyl)-21H,23H-porphyrin. (2)

[0171] Compound 2 was synthesized following a previously reported procedure by Collman et al. (J. Am. Chem. Soc. 1975, 97, 1427-1439).

[0172] Porphyrin 1 (1.8 g; 2.265 mmol, 1 eq.) was dissolved in concentrated HCl solution (60 mL). Excess SnCl.sub.2.2H.sub.2O (7.67 g; 34 mmol, 15 eq.) was added, and the mixture quickly heated at 70 C. for 25 min. After careful neutralization with ammonia, chloroform was added and the red mixture was stirred vigorously for 1 h. The organic phase was kept, and the aqueous phase was extracted with chloroform 3 times. The organic phases were combined, filtered, reduced under vacuum, washed with dilute ammonia and water, dried on Na.sub.2SO.sub.4 and evaporated, affording a statistical mixture of the 4 atropisomers of the desired product with impurities (1.86 g).

[0173] Pure ,,,-5,10,15,20-(o-aminophenyl)-21H,23H-porphyrin can be obtained after purification by column chromatography on silica gel (gradient elution from CH.sub.2Cl.sub.2/Et.sub.2O=100/0 to 95/5). The others atropisomers can be recovered after elution of CH.sub.2Cl.sub.2/MeOH 90/10. Refluxing the 3 others atropisomers in toluene for 4 h affords a new statistical mixture of 5,10,15,20-(o-aminophenyl)-21H, 23H-porphyrin.

[0174] .sup.1H NMR (400 MHz, CDCl.sub.3) 8.91 (s, 8H, -H), 7.88 (dd, J=7.4, 1.4 Hz, 4H, ArH), 7.60 (td, J=8.0, 1.5 Hz, 4H, ArH), 7.17 (td, J=7.4, 1.1 Hz, 4H, ArH), 7.11 (dd, J=8.1, 0.8 Hz, 4H, ArH), 3.51 (s, 8H, ArNH.sub.2), 2.67 (s, 2H, Pyr-NH).

[0175] HR-MALDI-TOF ([M+H+].sup.+) calculated for C.sub.44H.sub.35N.sub.8: 675.2980, found: 675.3359 IR (cm.sup.1): 3464w, 3359w, 3317w, 2920w, 2855w, 1613m, 1491m, 1472m, 1449m, 1348w, 1297m, 1261w, 1213w, 1185w, 1158w, 965s, 800s, 741vs, 648s

Chloro iron(III) ,,,-5,10,15,20-tetra(o-aminophenyl)-porphyrin. (3)

[0176] A solution of 2 (165 mg; 0.245 mmol; 1 eq.) and 2,6-lutidine (0.57 mL; 4.89 mmol; 20 eq.) in freshly distilled THF (50 mL) is degassed with argon for 10 min. FeBr.sub.2 (500 mg; 1.69 mmol; 6.9 eq.) is added and the reaction mixture is stirred for 4 hours under argon. After evaporation in vacuo, the residue is solubilized in CH.sub.2Cl.sub.2 and washed with water (until aqueous phase gets colorless) and with saturated NaCl solution. The organic phase is then dried overs MgSO.sub.4, filtered, evaporated and purified by column chromatography on silica gel (gradient elution from CH.sub.2Cl.sub.2/MeOH=100/0 to 90/10) affording the desired product as a purple powder (185 mg) in 98% yield.

[0177] HR-MALDI-TOF ([MCl.sup.].sup.+) calculated for C.sub.44H.sub.32FeN.sub.8: 728.2094, found: 728.1423 UV-vis (DMF, .sub.max/nm, log()/L.Math.mol.sup.1.Math.cm.sup.1): 414 (4.84), 576 (3.92), 626 (3.55) IR (cm.sup.1): 3464w, 3359w, 3204vw, 2916vw, 1612m, 1576m, 1495m, 1450m, 1334w, 1299m, 1267vw, 1201w, 1157w, 1067w, 997vs, 860w, 802m, 748s, 721m

,,,-5,10,15,20-tetra(o-N,N-dimethylaminophenyl)-21H,23H-porphyrin (5)

[0178] To a solution of porphyrin 2 (170 mg; 0.25 mmol; 1 eq.) in MeCN/MeOH (2/4 mL) is added formaldehyde (37% in H.sub.2O; 750 L; 10.1 mmol; 40 eq.) and sodium cyanoborohydride (79.2 mg; 1.26 mmol; 5 eq.). Glacial acetic acid (4.5 mL) is then poured dropwise and the reaction mixture is stirred for 4 hours at room temperature. The reaction mixture is diluted with CH.sub.2Cl.sub.2 (20 mL) and cautiously washed with saturated Na.sub.2CO.sub.3 solution (230 mL) and water (130 mL). The organic phase is dried over Na.sub.2SO.sub.4, evaporated in vacuo and the residue is purified by column chromatography on silica gel (gradient elution from Petroleum ether/Ethyl acetate=100/0 to 90/10) affording the desired product as a purple powder (112 mg) in 57% yield.

[0179] .sup.1H NMR (400 MHz, CDCl.sub.3) 8.75 (s, 8H, -H), 7.99 (dd, J=7.4, 1.6 Hz, 4H, ArH), 7.76-7.64 (m, 4H, ArH), 7.40 (d, J=7.5 Hz, 4H, ArH), 7.30 (td, J=7.4, 1.0 Hz, 4H, ArH), 2.23 (s, 24H, NCH.sub.3), 2.30 (s, 2H Pyr-NH).

[0180] .sup.13C NMR (101 MHz, CDCl.sub.3) 154.25 (C.sub.qArNMe.sub.2), 137.66 (CH.sub.Ar), 134.45 (C.sub.q), 131.04* (CH.sub.), 129.20 (CH.sub.Ar), 120.06 (CH.sub.Ar), 118.62 (C.sub.q), 117.91 (CH.sub.Ar), 43.55 (N(CH.sub.3).sub.2) *. *Broad signal for C.sub. and C.sub. could not be detected.

[0181] HR-MALDI-TOF ([M+H+].sup.+) calculated for C.sub.52H.sub.51N.sub.8: 787.4232, found: 787.4155.

[0182] UV-vis (DMF, .sub.max/nm, log()/L.Math.mol.sup.1.Math.cm.sup.1): 429 (5.20), 527 (4.21), 563 (3.88), 603 (3.78), 660 (3.61)

[0183] IR (cm.sup.1): 3314vw, 2924w, 2855w, 2787w, 1714vw, 1591w, 1558w, 1493m, 1472m, 1449m, 1429m, 1345m, 1316w, 1217m, 1158m, 1098w, 1050m, 967s, 945s, 801vs, 758s, 724vs

Chloro iron(III) ,,,-5,10,15,20-tetra(o-N,N-dimethylaminophenyl)-porphyrin. (4)

[0184] Path A (see FIG. 1). To a solution of porphyrin 3 (185 mg; 0.24 mmol; 1 eq.) in MeCN/CH.sub.2Cl.sub.2 (5/5 mL) is added formaldehyde (37% in H.sub.2O; 721 L; 9.69 mmol; 40 eq.) and sodium cyanoborohydride (76.1 mg; 1.21 mmol; 5 eq.). Glacial acetic acid (5 mL) is then poured dropwise and the reaction mixture is stirred for 4 hours at room temperature. The reaction mixture is evaporated, the residue is solubilized in CH.sub.2Cl.sub.2 (30 mL) and cautiously washed with saturated Na.sub.2CO.sub.3 solution (230 mL), saturated NaCl solution (215 mL) and water (120 mL). The organic phase is dried over MgSO.sub.4, filtered, evaporated in vacuo and the residue is purified by column chromatography on silica gel (gradient elution from CH.sub.2Cl.sub.2/MeOH=100/0 to 95/5) affording the desired product as a purple powder (158 mg) in 75% yield.

[0185] Path B. A solution of 5 (30 mg; 0.038 mmol; 1 eq.) and 2,6-lutidine (89 L; 0.76 mmol; 20 eq.) in freshly distilled THF (8 mL) is degassed with argon for 10 min. FeBr.sub.2 (158 mg; 0.53 mmol; 14 eq.) is added and the reaction mixture is stirred for 20 hours at 40 C. Some remaining starting material could be observed by TLC, so more FeBr.sub.2 (78 mg; 0.267 mmol; 7 eq.) are added and the reaction mixture is stirred for 5 h at 45 C. After evaporation in vacuo, the residue is solubilized in CH.sub.2Cl.sub.2 and washed with water (until aqueous phase gets colorless) and with saturated NaCl solution. The organic phase is then dried over MgSO.sub.4, filtered, evaporated and purified by column chromatography on silica gel (CH.sub.2Cl.sub.2/Ethyl Acetate 90/10 followed by CH.sub.2Cl.sub.2/MeOH 9/1) affording the desired product as a purple powder in 61% yield.

[0186] HR-MALDI-TOF ([MCl].sup.+) calculated for C.sub.52H.sub.48FeN.sub.8: 840.3346 found: 840.2511.

[0187] UV-vis (DMF, .sub.max/nm, log(s)/L.Math.mol.sup.1.Math.cm.sup.1): 426 (4.91), 583 (3.86), 631 (3.59), 703 (3.22)

[0188] IR (cm.sup.1): 2929w, 2843w, 2782w, 1591w, 1494m, 1429m, 1330m, 1201w, 1156w, 1107w, 1065w, 1053w, 997vs, 946m, 801m, 757m

Chloro Iron(III) 5,10,15,20-tetra(o-N,N,N-Trimethylanilinium)porphyrin tetra(trifluoromethanesulfonate) (Fe-o-TMA

[0189] To a solution of porphyrin 4 (40 mg; 0.046 mmol; 1 eq.) in dry DMF (3 mL) under argon is added methyl trifluoromethanesulfonate (517 L; 4.56 mmol; 100 eq.). The solution is stirred for 24 h at 100 C. under argon. Most of the DMF is evaporated under vacuum and the residual black waxy solid is solubilized in water (10 mL) and washed with CH.sub.2Cl.sub.2 (420 mL). The aqueous phase is evaporated and the brown solid is solubilized in water (5 mL) and purified by dialysis (membrane MWCO=1000 g/mol) against deionised water (590 min). After removal of water, the residue is dried in vacuo for 24 h affording a dark brown powder (65 mg) in 92% yield.

[0190] ESI-MS: [M+Cl.sup.].sup.4+ calculated for C.sub.56H.sub.60C.sub.1FeN.sub.8: 233.8489; found: 233.8489.

[0191] UV-vis (DMF, .sub.max/nm, log(s)/L.Math.mol.sup.1.Math.cm.sup.1): 430 (4.90); 561 (3.81); 624 (3.43); 664 (3.28).

[0192] IR (cm.sup.1): 3176w, 2827vw, 1471m, 1260vs (TfO), 1225vs (TfO), 1158vs (TfO), 1028vs (TfO), 948w, 834w, 759w, 666w, 634vs (TfO).

Example 2. Cyclic Voltammetry and Electrolysis

[0193] Methods and Instrumentation.

[0194] Cyclic Voltammetry.

[0195] The working electrode was a 3 mm-diameter glassy carbon (Tokai) disk carefully polished using decreasing size of diamond paste (from 15 to 1 m), ultrasonically rinsed in absolute ethanol and dried before use or a mercury drop deposited on a 1 mm diameter gold disk. The counter-electrode was a platinum wire and the reference electrode was an aqueous SCE electrode. All experiments were carried out under argon or carbon dioxide atmosphere at 25 C., the double-wall jacketed cell being thermostated by circulation of water. Cyclic voltamograms were obtained by use of a Metrohm AUTOLAB instrument. Ohmic drop was compensated using the positive feedback compensation implemented in the instrument.

[0196] Preparative Scale Electrolysis. Electrolyzes were performed using a Princeton Applied Research (PARSTAT 2273) potentiostat. The experiments were carried out in a two-compartment cell with a glassy carbon crucible (the volume of the solution was 4 mL and active surface area was 12.7 cm.sup.2) or a mercury pool (active surface area was 5.1 cm.sup.2). The reference electrode was an aqueous SCE electrode and the counter electrode was a platinum wire in a bridge separated from the cathodic compartment by a glass frit, containing 0.2M NEt.sub.4CH.sub.3CO.sub.2+0.1 M NBu.sub.4PF.sub.6 DMF solution. The electrolysis solution was purged with CO.sub.2 during 20 min prior to electrolysis. Particular care was exerted to minimize the ohmic drop between working and reference electrodes. This was performed as follows: the reference electrode was directly immersed in the solution (without separated bridge) and put progressively closer to the working electrode until sustained oscillations appeared. It was then moved slightly away until the remaining oscillations were compatible with the catalytic current. The appearance of oscillations in this cell configuration does not require positive feedback compensation as it does with microelectrodes. The potentiostat+positive feedback compensation device system is equivalent to a self-inductance. Oscillations thus appear as soon as the resistance that is not compensated by the potentiostat comes close to zero as the reference electrode comes closer and closer to the working electrode surface.

[0197] Gas Detection. Gas chromatography analyses of gas evolved in the headspace during the electrolysis were performed with an Agilent Technologies 7820A GC system equipped with a thermal conductivity detector. CO and H.sub.2 production was quantitatively detected using a CP-CarboPlot P7 capillary column (27.46 m in length and 25 m internal diameter). Temperature was held at 150 C. for the detector and 34 C. for the oven. The carrier gas was argon flowing at 9.5 mL/min at constant pressure of 0.5 bars. Injection was performed via a 250-L gas-tight (Hamilton) syringe previously degassed with CO.sub.2. Conditions allowed detection of H.sub.2, O.sub.2, N.sub.2, CO, and CO.sub.2. Calibration curves for H.sub.2 and CO were determined separately by injecting known quantities of pure gas.

[0198] Results.

[0199] FIG. 2 shows the Fe.sup.I/0 cyclic voltammetric responses obtained under 1 atm CO.sub.2 in the presence of 3 M phenol, with Fe-o-TMA, Fe-p-TMA, Fe-p-PSULF, FeTPP, FeF5TPP, FeF10TPP and FeF20TPP.

[0200] It immediately appears, before any treatment of the raw data, that the Fe-o-TMA is a far much better catalyst than any other molecules in the series both in terms of current and potential.

[0201] As seen in FIG. 3, the formation of CO is clearly attested by the observation of a cathodic shift of the Fe.sup.I/Fe.sup.II reoxidation wave on the reverse scan in the cyclic voltammetric catalytic responses of Fe-p-TMA, Fe-o-TMA and Fe-p-PSULF. It matches what can be observed in the cyclic voltammetry of the Fe.sup.II/Fe.sup.I couple in the presence of CO.

[0202] It is noticed (FIGS. 2 and 3) that the current-potential responses of Fe-p-TMA, Fe-o-TMA and Fe-p-PSULF show peaks instead of the plateaus expected for fast catalytic processes. It is assumed that this is due to the interference of secondary phenomena such as substrate or co-substrate consumption, inhibition by product and possibly other phenomena that all increase with the charge passed. One way of fighting the interference of such phenomena is to raise the scan rate, and thereby, decrease the charge passed, so as to get back to a S-shaped current potential responses and derive the rate constant from the ensuing plateau current.

[0203] This is the treatment that has been applied to the raw cyclic voltammetric data as shown in FIG. 4. It is worth noting, en passant, that the scan rates required to reach an S-shaped CV responses are, as expected, the larger the stronger catalysis (in increasing order: Fe-p-PSULF, Fe-p-TMA, Fe-o-TMA).

[0204] FIG. 7 shows the faradaic yields and the CO/H.sub.2 selectivity factor for Fe-o-TMA (at an overpotential of 220 mV) and Fe-p-TMA (at an overpotential of 450 mV) over 7 hours electrolysis time. In both cases the CO-vs.-H.sub.2 selectivity is excellent, even slightly better in the first case than in the second. During this seven-hour experiments both catalysts appear stable. Fe-o-TMA appears stable even during much longer electrolysis times, up to at least 84 hours as reported in FIG. 8 at an overpotential of 220 mV at which the TOF is 10.sup.6 s.sup.1.

Example 3: Benchmarking of Fe-o-TMA with Prior Art Complexes in Organic Medium

[0205] Benchmarking with other catalysts in terms of overpotential and turnover frequency in water does not seem possible at the moment.

[0206] A comparison with the characteristics of other catalysts obtained in an aprotic solvent such as DMF or acetonitrile was thus made.

[0207] Combination of the foot-of-the wave analysis with increasing scan rates, which both minimize the effect of side-phenomena, allowed the determination of the turnover frequency as a function of the overpotential, leading to the catalytic Tafel plot for the Fe-o-TMA catalyst shown as the upper curve in FIG. 5. The turnover frequency (TOF), takes into account that the molecules that participate to catalysis are only those contained in the thin reaction-diffusion layer adjacent to the electrode surface in pure kinetic conditions. The overpotential, , is the difference between the standard potential of the reaction to be catalyzed and the electrode potential. Correlations between TOF and provide catalytic Tafel plots that are able to benchmark the intrinsic properties of the catalyst independently of parameters such as cell configuration and size. Good catalysts stand in the upper left corner and bad catalysts in right bottom corner. These plots allow one to trade between the rapidity of the catalytic reaction and the energy required to run it. The other Tafel plots shown in FIG. 6 are simply the repeat of what has been established in details in Costentin et al Proc. Natl. Acad. Sci. U.S.A. 2014, 111, 14990-14994.

[0208] The values of k.sub.cat required to establish the catalytic Tafel plots and the k.sub.cat vs. Eat for these three porphyrins may then be derived from the plateau current values, using the following equations. For the catalytic plateau current: i.sub.pl=FSC.sub.cat.sup.0{square root over (D.sub.cat)}{square root over (2k.sub.cat)}, and for the one-electron diffusion current at 0.1 V/s:

[00002]$i_p^0=\mathrm{FS}0.446C_{\mathrm{cat}}^0.Math.\sqrt{D_{\mathrm{cat}}}.Math.\sqrt{\frac{\mathrm{Fv}\left(=0.1.Math..Math.V.Math.\text{/}.Math.s\right)}{\mathrm{RT}}}.$

Using the ratio

[00003]$\frac{i_{\mathrm{pl}}}{i_p^0}=2.24.Math.\sqrt{\frac{2.Math.k_{\mathrm{cat}}}{0.1}.Math.\frac{\mathrm{RT}}{F}}$

avoids determining S (electrode surface area), C.sub.cat.sup.0 and D.sub.cat (concentration and diffusion coefficient of the catalyst, respectively). Thus:

[00004]$k_{\mathrm{cat}}={\left(\frac{i_{\mathrm{pl},v}}{i_{p,0.1}^0}\right)}^2.Math.\frac{0.1}{{2.24}^2}.Math.\frac{F}{2.Math.\mathrm{RT}}$

[0209] The values of k.sub.cat thus obtained were then used to locate the catalytic Tafel plots of porphyrins (FIG. 5a) and to introduce the corresponding data points in the kcat vs. E.sub.cat.sup.0 correlation diagram (FIG. 5b).

[0210] A through-space effect clearly appears for Fe-p-TMA in so far that its representative point stands above the through-structure correlation line. That this effect results from the positive charges borne by the substituents is confirmed by the observation that the introduction of negative charges in similar positions produces the reverse effect (FIG. 6b). Although clearly present, these effect are necessarily small since the charges are rather distant from the reacting center, viz., the initial adduct between CO.sub.2 and the iron(0) complex.

[0211] These observations encouraged us to introduce positive charged substituents closer to the reacting center in spite of the expected synthetic difficulties related to steric congestion of this type of substituted iron-porphyrin (see the Experimental Section below). The resultFe-o-TMAproved to be the best catalyst of the whole iron porphyrin series with considerable gains in terms of overpotential and of turnover frequency as can be seen in FIGS. 5 and 6. The maximal turnover frequency is as high as 10.sup.6 s.sup.1 and the turnover frequency at zero overpotential larger than 300 s.sup.1.

[0212] The reason for this leap forward is most likely the stabilization of the initial iron(0) adduct by the interactions of the negative charge borne by the oxygens of CO.sub.2 in this adduct with the nearby positive charges borne by the trimethylanilinium substituents on the porphyrin phenyls. The presence of phenol in large concentration then helps the proton-assisted reductive cleavage of one of the carbon-oxygen bond of CO.sub.2. This is a particularly striking example of the power of close-distance through-space interactions in boosting of catalysis. Another type of close-distance through-space interactions as already been implemented to boost CO.sub.2 reduction catalysis, albeit with a lesser efficiency, namely the introduction of ortho phenol groups in the porphyrin phenyls in which stabilization of the initial Fe.sup.0 CO.sub.2 adduct is achieved by means of H-bonding. Comparison between the coulombic-interaction and H-bond-interaction series is shown in FIG. 6, as well as with the best examples that can be found in the literature. It again appears that Fe-o-TMA is the champion in all categories.

Example 4: Catalysis in Water

[0213] Cyclic Voltammetry

[0214] The same experimental setup was used as in example 2.

[0215] The cyclic voltammetric response of Fe-p-TMA and Fe-o-TMA in water is reported on FIG. 9. Under Argon, Fe-o-TMA features two reversible waves attributed to the Fe.sup.II/Fe.sup.I and Fe.sup.I/Fe.sup.0 couples while for Fe-p-TMA, only the Fe.sup.II/Fe.sup.I couples appears as a reversible wave. This already indicates a poor reactivity of the Fe.sup.0 state of Fe-o-TMA toward the proton reduction reaction.

[0216] Under CO.sub.2, the two catalysts present a strong catalytic wave. The highest being the one of Fe-o-TMA. Furthermore, the onset of the catalytic wave of Fe-o-TMA is approximately 125 mV more positive than the one of Fe-p-TMA.

[0217] For very efficient catalysts, the substrate diffusion can become the limiting factor of the catalysis. In this specific case of total catalysis, the peak current of the catalytic wave is independent of the kinetic constant and the catalyst concentration and can be calculated with the following equation:

[00005]$i_p=0.52{\mathrm{FSC}}_{\mathrm{CO}.Math..Math.2}^0.Math.\sqrt{\frac{D_{\mathrm{CO}.Math..Math.2}.Math.\mathrm{Fv}}{\mathrm{RT}}}$$i_p=1.27.Math..Math.\mathrm{mA}$$S=0.07.Math..Math.{\mathrm{cm}}^2\left(d=3.Math..Math.\mathrm{mm}\right)$$C_{\mathrm{CO}.Math..Math.2}^0={30.10}^{-6}.Math..Math.\mathrm{mol}.Math.{\mathrm{cm}}^{-3}$$D_{\mathrm{CO}.Math..Math.2}={1.10}^{-5}.Math..Math.{\mathrm{cm}}^2.Math.s^{-1}$$F=96500.Math..Math.C.Math.{\mathrm{mol}}^{-1}$$V=0.1.Math..Math.V.Math.\text{/}.Math.s.Math..Math.R=8.314.Math..Math.J.Math.{\mathrm{mol}}^{-1}.Math.K^{-1}$$T=293.Math..Math.K$

[0218] The experimental peak current for Fe-o-TMA is measured at 1.03 mA, which indicates that a total catalysis regime is almost reached.

[0219] Theses cyclic voltammetric experiments clearly indicate that Fe-o-TMA is a better catalyst than Fe-p-TMA both in term of kinetic rate and overpotential.

[0220] Preparative-Scale Electrolysis in Water.

[0221] Electrolyses were performed using a Princeton Applied Research PARSTAT 4000 potentiostat interfaced with VersaStudio software. The experiment was carried out in a two-compartment cell. The working electrode was a glassy carbon crucible (active surface area was 15 cm.sup.2). The reference electrode was an aqueous SCE electrode and the counter electrode a platinum grid in a bridge separated from the cathodic compartment by a glass frit. Ohmic drop was minimized by immersing directly the reference electrode into the solution as close as possible to the working electrode. The electrolysis cell was purged with CO.sub.2 for 15 min before electrolysis then sealed for quantitative experiments or under a continuous flux for the long-time-scale electrolysis to avoid the CO.sub.2 consumption.

[0222] The cathodic electrolyte solution consisted in aqueous 0.5 mM Fe-o-TMA in 0.1M KCl and 0.5M KHCO.sub.3 (pH 7.2) saturated with CO.sub.2, while the anodic electrolyte solution consisted in an aqueous solution of phosphate buffer (0.2M) at pH 7.1.

[0223] Gas Detection.

[0224] Gas chromatography analyses of gas evolved in the headspace during the electrolysis were performed with the same techniques than detailed in Example 3.

[0225] Results

TABLE-US-00001 Potential (V Selectivity Catalyst vs.NHE) Time Current Charge TON (CO/H.sub.2) (CO/H.sub.2) Fe-p-TMA 0.96 V 7 h 45 ~360 A 11.7 37.8/1.6 96/4 Fe-o-TMA 1.86 V 24 h ~800 A 64.8 111/0 100/0 Fe-o-TMA 0.91 V 9 h 10 3 -> 2.6 mA 92.8 159/0 100/0 Fe-o-TMA 0.91 V 8 h 40 3.4 mA 110 190/1 99.5/0.5 Fe-o-TMA 0.96 V 2 h 35 + 14 -> 2.7 mA 86.3 + 660/6.25 99/1 2 h 55 + 71.6 + 6 h 13 + 121.1 + 2 h 50 + 62.4 + 3 h 15 44.29

[0226] A series of preparative scale electrolyses were performed at three different applied potentials (0.86V, 0.91V and 0.96V vs. NHE) over a total of 60 h (FIG. 10). During the whole experiment, the system remained very selective towards CO.sub.2 reduction with a faradaic efficiency above 99% for CO. The almost linear evolution of the charge with time indicates a quite good stability of the catalyst.

[0227] Comparison with Fe-p-TMA

[0228] A control electrolysis of 7 h 45 at 0.96V vs. NHE has been performed with Fe-p-TMA, with the same setup used for Fe-o-TMA, The measured current density was more than an order of magnitude lower for Fe-p-TMA compared to Fe-o-TMA (FIG. 11). Moreover, the selectivity was slightly lower for Fe-p-TMA (96% selectivity for CO, versus >99% for Fe-o-TMA).

SUMMARY

[0229] Without wishing to be bound by theory, it may be concluded that through-space substituent effects on the catalysis of the electrochemical CO.sub.2-to-CO conversion by iron(0)-tetraphenyl porphyrins has been first investigated and evidenced by the introduction of four positively charged trimethylanilinium groups in para-position of the TPP phenyls (Fe-p-TMA). The assignment of this catalysis boosting effect to the coulombic interaction of these positive charges with the negative charges borne by the initial Fe.sup.0CO.sub.2 adduct has been further confirmed by the negative catalytic effect observed when the four positive charges in Fe-p-TMA are replaced by four negative charges borne by sulfonate groups installed in the para-position of the TPP phenyls (Fe-p-PSULF). Optimization of the catalysis by means of coulombic stabilization of the initial Fe.sup.0 CO.sub.2 adduct was reached when four positively charged trimethylanilinium groups are introduced in the ortho position of the TPP phenyls (Fe-o-TMA). The exceptional efficiency of the resulting catalyst is unprecedented with maximal turnover frequency as high as 106 s.sup.1 and is reached at a low overpotential of 220 mV. The selectivity for CO production is close to 100% while the catalyst appears extremely stable upon long term electrolysis, with no significant alteration for more than three and a half days.