Homogeneous Reduction of CO 2 by Photogenerated Pyridinyl Radicals Francesca Riboni, †,‡ Elena Selli, ‡ M. R. Hoffmann, † and A. J. Colussi* ,† † Linde Center for Global Environmental Science, California Institute of Technology, Pasadena, California 91125, United States ‡ Department of Chemistry, University of Milan, via Golgi 19, 20133 Milano, Italy ABSTRACT: We report that 1-hydropyridinyl radicals (1-PyH • ) photogenerated in solution react with dissolved CO 2 en route to its 2e - reduction into carboxylic acids. The 254 nm excitation of pyridine (Py) in deaerated 2-PrOH/H 2 O mixtures saturated with 1 atm of CO 2 yields a suite of products, among which we identified Na(HCOO) 2 - (m/z - = 113), C 5 H 6 NCOO - (m/z - = 124), and C 5 H 10 O 2 NCOO - (m/z - = 160) species by electrospray ionization mass spectrometry. These products demonstrably contain carboxylate functionalities that split CO 2 neutrals via collisionally induced dissociation. We infer that 1-PyH • [from (1) 3 Py* + 2-PrOH → 1-PyH • + • PrOH] adds to CO 2 , in competition with radical-radical reactions, leading to intermediates that are in turn reduced by • PrOH into the observed species. The formation of carboxylates in this system, which is shown to require CO 2 , Py, 2-PrOH, and actinic radiation, amounts to the homogeneous 2e - reduction of CO 2 by 2-PrOH initiated by Py*. We evaluate a rate constant (2) k 2 (1-PyH • + CO 2 → • Py-1-COOH) ≈ O (10) M -1 s -1 and an activation energy E 2 ≥ 9 kcal mol -1 that are compatible with thermochemical estimates for this reaction. ■ INTRODUCTION The impact of fossil fuel combustion emissions on Earth’s climate would be alleviated by an ideal process that captures atmospheric CO 2 at ∼400 ppm levels and releases it in concentrated form with minimum energy expenditure and capital investment, 1-6 followed by its reductive conversion into fuels and/or chemical feedstocks via an artificial photosynthetic device using inexpensive materials and driven by sunlight. 7-14 The endoergic reduction of CO 2 into organic matter at significant rates, however, is challenging, perhaps as best attested by the fact that evolution took ∼2 billion years to develop photosynthesis. 15 The negative electron affinity of CO 2 in the gas phase 16 translates in solution into a one-electron transfer that converts linear CO 2 into bent CO 2 -• both hindered by unfavorable thermodynamics E°(CO 2 /CO 2 -• )= -2.14 V (versus SCE) and very slow self-exchange rates. 17 The two-electron (hydride) transfer into HCOO - (E° = -0.55 V versus SCE) or the stabilization of CO 2 -• into bound intermediates appear to be more favorable pathways to initiate the reduction of CO 2 toward CH 3 OH and CH 4 . The chemical stabilization of CO 2 -• may have already been achieved in the pyridine (Py)-catalyzed electroreduction of CO 2 . 18,19 The originally proposed mechanism, in which 1- hydropyridinyl radicals (1-PyH • ), 20-22 from the electro- chemical reduction of pyridinium (PyH + ), react homoge- neously with dissolved CO 2 to produce a carbamyl intermediate, however, is held in contention. 19,23-26 It has been challenged by calculations predicting a reduction potential E°(PyH + /PyH • ) ≈-1.4 V (versus SCE) 23 that is significantly more negative than the reported E° = -0.58 V experimental value. 19 On this premise, it was asserted that “under no circumstances can homogeneous pyridinyl radicals in solution be considered active catalysts for CO 2 reduction”. 23 It should be pointed out, however, that such a conclusion implicitly assumes that the discrepancy can only be accounted for by the adsorptive stabilization of 1-PyH • on the cathode. Other explanations, however, are possible. 26 There is evidence that partial charge transfer from an unpolarized electrode to preadsorbed PyH + would also render E°(PyH + /1-PyH • ) less negative than the calculated -1.4 V value. 25 It is therefore possible that the electrochemical reduction of PyH + on Pt at E° = -0.58 V could release 1-PyH • into the bulk electrolyte, as originally proposed. 19 The outstanding issue at this point seems to be whether 1-PyH • in fact reacts with CO 2 homogeneously in water. Here, we address this key question and report experiments in which 1-PyH • is generated photochemically in situ in 2-propanol (2-PrOH)/water mixtures containing dissolved CO 2 . ■ EXPERIMENTAL SECTION All reagents were of chemical grade. Py and 2-PrOH (Sigma- Aldrich) and carbon dioxide (Air Products) were used as received. Py solutions in 10% (v/v) 2-PrOH/Milli-Q water mixtures were adjusted to pH 3 with HCl, where Py (pK a = Special Issue: Mario Molina Festschrift Received: September 25, 2014 Revised: November 10, 2014 Article pubs.acs.org/JPCA © XXXX American Chemical Society A dx.doi.org/10.1021/jp509735z | J. Phys. Chem. A XXXX, XXX, XXX-XXX