Homogeneous solar fuels photocatalytic systems often require several additives in solution with the catalyst to operate, such as a photosensitizer (PS), Brønsted acid/base, and a sacrificial electron donor (SED). Tertiary amines, in particular triethylamine (TEA) and triethanolamine (TEOA), are ubiquitously deployed in photocatalysis applications as SEDs and are capable of reductively quenching the PS’s excited state. Upon oxidation, TEA and TEOA form TEA•+ and TEOA•+ radical cations, respectively, which decay by proton transfer to generate redox non-innocent transient radicals, TEA and TEOA, respectively, with redox potentials that allow them to participate in an additional electron transfer step, thus resulting in net one-photon/two-electron donation. However, the properties of the TEA and TEOA radicals are not well understood, including their reducing powers and kinetics of electron transfer to catalysts. Herein, we have used both pulse radiolysis and laser flash photolysis to generate TEA and TEOA radicals in CH3CN, and combined with UV/Vis transient absorption and time-resolved mid-infrared spectroscopies, we have probed the kinetics of reduction of the well-established CO2 reduction photocatalyst, fac-ReCl(bpy)(CO)3 (bpy = 2,2′-bipyridine), by these radicals [kTEA• = (4.4 ± 0.3) × 109 M−1 s−1 and kTEOA• = (9.3 ± 0.6) × 107 M−1 s−1]. The ∼50× smaller rate constant for TEOA indicates, that in contrast to a previous assumption, TEA is a more potent reductant than TEOA (by ∼0.2 V, as estimated using the Marcus cross relation). This knowledge will aid in the design of photocatalytic systems involving SEDs. We also show that TEA can be a useful radiolytic solvent radical scavenger for pulse radiolysis experiments in CH3CN, effectively converting unwanted oxidizing radicals into useful reducing equivalents in the form of TEA radicals.

Inspired by nature, artificial photosynthesis (AP) is an encouraging route to convert CO2, water, and sunlight into fuels and fine chemicals. On the reductive side of AP, evolution of CO is the simplest outcome from CO2 reduction by transition metal-based catalysts, requiring either two electrons (2e) or 2e + 2H+, producing CO32− or H2O, respectively, as byproducts alongside CO. Since the first reports by Lehn and co-workers1,2 that fac-ReX(bpy)(CO)3 (X = Cl, Br; bpy = 2,2′-bipyridine) can act as an efficient, self-sensitized, homogeneous photocatalyst for the reduction of CO2 to CO, this and many other families of transition metal complexes have been heavily investigated for photocatalytic CO2 reduction, often in the presence of an added photosensitizer for light harvesting. The Lehn-type (M = Re and Mn)3–12 and numerous other catalysts bearing carbonyl (CO) and/or bpy ligands have proved to be ideal for mechanistic studies of CO2 reduction because of the substitutional versatility at the bipyridine ligand and their easily detected spectroscopic signatures in the visible and mid-infrared regions. However, these favorable characteristics are counterbalanced against the large number of reactive and excited state species in solution during catalysis since the reactions occur in a complex mixture predominantly containing the catalyst, a sacrificial electron donor (SED), and often a photosensitizer and/or a Brønsted acid/base.13,14 Consequently, mechanistic understanding of proton and electron transfer steps is often limited to thermodynamic parameters such as redox potentials and acid dissociation constants unless specialized techniques such as time-resolved spectroscopy are employed.

While SEDs are unlikely to be used in a practical AP process, they are, in some cases, recyclable15 and also useful for studying photochemical reductive half-reactions, such as CO2 or proton reduction, since they are a convenient source of electrons with which to reduce a catalyst. The tertiary amines, triethylamine (TEA) and triethanolamine (TEOA), are two of the most widely and recently used SEDs in photocatalysis applications in the year 2023 preceding this contribution.16–30 While there is a trend in photocatalytic CO2 reduction to replace them with more potent reductants, such as 1,3-dimethyl-2-phenyl-2,3-dihydro-1H-benzo[d]imidazole (BIH),20,21,26,27,31,32 the rapidly growing field of metallaphotoredox catalysis, which uses light to drive the catalytic transformation of organic molecules into value-added chemicals with transition metal catalysts, relies heavily on the use of TEA and related aliphatic amines as SEDs.33–35 In a common scenario, the photogenerated metal-to-ligand charge transfer (MLCT) excited state of a photosensitizer or photocatalyst, e.g., [fac-ReCl(bpy)(CO)3]* (Scheme 1, reaction 1), is reductively quenched by an amine, generating an aminium radical cation, amine•+ (Scheme 1, reaction 2). Since a large excess of amine is typically used (>1 M), rapid proton transfer (PT) occurs from the αC–H bond of amine•+ to another amine molecule,36–39 generating a neutral amine radical and a trialkylammonium cation (Scheme 1, reaction 3). Notably, the assignment of PT is challenged by more contemporary observations of the amine•+ engaging in hydrogen atom transfer with substrates containing αC–H bonds to furnish chemoselective photoredox catalysis.40 

SCHEME 1.

(1) Photoexcitation of fac-ReCl(bpy)(CO)3 (1); (2) Reductive quenching of the MLCT excited state of 1 by TEA;51,56 (3) Proton transfer between TEA•+ and TEA;36–39 (4) One-electron reduction of 1 by TEA. Note that analogous reactions also occur with TEOA.

SCHEME 1.

(1) Photoexcitation of fac-ReCl(bpy)(CO)3 (1); (2) Reductive quenching of the MLCT excited state of 1 by TEA;51,56 (3) Proton transfer between TEA•+ and TEA;36–39 (4) One-electron reduction of 1 by TEA. Note that analogous reactions also occur with TEOA.

Close modal

For TEA, the rate constant for the PT reaction in acetonitrile (CH3CN) has been estimated to range from 1 × 108–2 × 109 M−1 s−1.41 Accordingly, in the presence of 1 M TEA, PT will occur on a timescale of 0.5–10 ns. The rapidity of this PT is responsible for the sacrificial nature of tertiary amines as electron donors since it prevents charge recombination by back electron transfer. The amine radicals that are formed are reactive species, capable of adding to double bonds and aromatics and engaging in radical-radical coupling reactions.42,43 However, they are also strong reductants that can often reduce the ground state of a photosensitizer or catalyst (Scheme 1, reaction 4), resulting in overall one-photon/two-electron SED reactivity, which improves the quantum efficiency of the photocatalytic process. While the amine radical coupling reactivity has had enormous importance for pharmaceutical and medicinal applications,44–46 in general, the redox properties of amine are still not well understood as a SED, making it difficult to investigate by conventional methods.

In this work, as part of an investigation to develop the mechanistic framework for the reductive activation of homogeneous photocatalysts in CH3CN, we have employed pulse radiolysis (PR) and laser flash photolysis (LFP) techniques to generate TEA and TEOA radicals in CH3CN and probed the one-electron reduction of the CO2 reduction catalyst, fac-ReCl(bpy)(CO)3 (1), by these radicals (Scheme 1, reaction 4). In LFP, the amine radicals were generated following pulsed 355 nm laser excitation of 1, according to reactions 1–3 shown in Scheme 1. PR is a complementary technique that uses a high-energy electron pulse from an accelerator to ionize the solvent.47,48 This produces solvated electrons, which very rapidly reduce 1, together with various other solvent-derived radicals, at least one of which we found can oxidize TEA or TEOA, thus initiating the PT reaction 3 and the subsequent secondary reduction of 1 by TEA or TEOA (reaction 4). In these experiments, we seek to expand on the previous determinations that TEA and TEOA are secondary reductants towards metal-based complexes in aqueous solution and DMF,49–53 and metal-organic-frameworks in CH3CN.54 It was found that TEA is kinetically much faster than TEOA at reducing 1 in CH3CN; their rate constants for electron transfer differ by a factor of ∼50×. Importantly, the order of the reducing power of these two radicals is the opposite of what has previously been discussed in the literature,52,53,55 which has important implications in photocatalysis applications when choosing a SED. Our discovery that tertiary amines such as TEA can be oxidized by radiolytically generated solvent radicals in CH3CN, generating TEA radicals as useful secondary reductants, also opens up the possibility of using TEA as a radical scavenging agent, enabling cleaner mechanistic pulse radiolysis investigations in CH3CN.

As a general remark, the chemicals used in this work were purchased from Sigma-Aldrich and most often further purified using common bench scale protocols (vida infra). Transient absorption spectroscopy was performed either by using the pulse radiolysis or flash photolysis technique previously established in our lab. Equipment used to perform spectroelectrochemistry and cyclic voltammetry was purchased from commercial sources, and the electrochemistry techniques were adopted from the literature.

Acetonitrile (CH3CN, Sigma-Aldrich, HPLC Plus, ≥99.9%) on a scale of 250 mL was purified by freeze–pump–thaw degassing, followed by rapid stirring over a 0.05% by weight Na/Hg amalgam (1 mL) for two hours in a nitrogen-filled glovebox. The resulting CH3CN with a white precipitate suspended in it was separated from the excess amalgam by decantation. CH3CN was then separated from the white precipitate by vacuum transfer at 10−6 Torr. The purity of the CH3CN was tested by measuring the lifetime of the solvated electron generated by PR, with lifetimes on the order of ∼1 µs indicating excellent purity (Fig. S1). Triethylamine (TEA, Sigma-Aldrich, Puriss, ≥99.5%) was purified on a 25 mL scale by freeze–pump–thaw degassing and then storing over ∼0.5 mL of sodium potassium alloy (NaK, Sigma-Aldrich, Potassium 78 wt. %, Sodium 22 wt. %) for several days, followed by vacuum transfer at 10−6 Torr to separate from the NaK. Triethanolamine (TEOA, Sigma-Aldrich, ≥99.0%) was degassed by argon bubbling and stored over activated 3 Å sieves for at least one week before use. 1,4-diazabicyclo[2.2.2]octane (DABCO, Sigma-Aldrich, ≥99%) was purified on a 500 mg scale by sublimation at 0.1–1 Torr and room temperature. N,N-dimethyl-p-toluidine (DMPT, Sigma-Aldrich, 99%) was fractionally distilled on a 100 mL scale and degassed by freeze–pump–thaw. fac-ReCl(bpy)(CO)3 was synthesized and purified according to literature preparation.57 Tetrabutylammonium hexafluorophosphate (TBAPF6, Sigma-Aldrich, ≥99.0% for electrochemical analysis) was used without further purification. All materials were stored in a nitrogen-filled glove box for sample preparation.

PR experiments were carried out at the 9 MeV BNL Laser Electron Accelerator Facility (LEAF).58 The electron pulses are generated by a Ti:sapphire oscillator-based ultrafast laser system (Spectra Physics) that delivers ∼3 ps 266 nm pulses to a Mg photocathode. For UV/Vis/near-IR detection of transient radicals, the optical probe path consisted of a pulsed 75 W xenon arc lamp, a 0.5 cm pathlength quartz optical cuvette with an airtight Teflon valve, a selectable 10 nm bandpass interference filter, and either a germanium photodiode (GEP-600L, 2–3 ns response time) or FND-100Q silicon photodiode (∼1.5 ns response time, EG&G) detector. For kinetic measurements at 510 nm, a GG 495 nm long-pass filter was placed between the xenon arc lamp and sample cuvette to prevent photoexcitation of fac-ReCl(bpy)(CO)3. For time-resolved infrared (TRIR) detection,59 the electron pulses were produced using a Nd:YAG laser that delivers ∼7 ns 266 nm pulses (Spectra Physics, Quanta Ray GCR 170-10) to the Mg photocathode. A continuous wave mode-hop-free (CW-MHF) external-cavity quantum cascade laser (EC-QCL, DRS Daylight Solutions, 21054-MHF) tuned to specific wavelengths was used as the mid-IR probe light source. The IR beam was split into probe and reference beams before the sample, and the probe beam was focused into a home-built, airtight IR solution flow cell (1.93 mm pathlength for the analyzing IR beam), which was equipped with 0.35 mm thick CaF2 windows. The signals from the two HgCdTe IR detectors (Kolmar Technologies, Inc., KMPV9-0.5-J2, DC-20 MHz) were simultaneously digitized on an oscilloscope (Teledyne LeCroy, HRO 66Zi, 12-bit, 600 MHz). The response time of the IR detection system (10%–90% rise) was measured to be ∼40 ns. Solutions for TRIR were prepared inside a N2-glovebox, housed in gas-tight syringes, and delivered to the IR cell with remote-controlled syringe pumps (typically by mixing up to three streams from three syringes to control reagent concentrations).

For LFP, we used the third harmonic (355 nm) of a Nd:YAG pulsed laser (Spectra Physics, Quanta-Ray LAB-170–10, 8–10 ns fwhm) to photoexcite the sample. TRIR measurements were performed using two CW EC-QCLs (DRS Daylight Solutions, 21054-MHF or MIRcat-2400) as mid-IR monitoring light sources. The mid-IR light from the EC-QCL was tuned to a specific wavelength and split into two beams: the probe and the reference beams. The probe beam was passed through a liquid IR flow cell (Harrick Scientific Products, Inc., DLC-S25, 1 mm pathlength), and both beams were directed to a matched pair of fast rise time IR detectors (Kolmar Technologies, Inc., KMPV9-0.5-J2, DC-20 MHz). The dual-beam optical geometry and TRIR instrumentation were previously reported in detail for experiments initiated by pulse radiolysis.59 Solutions for TRIR were prepared inside a N2-glovebox, housed in gas-tight syringes, and delivered to the IR cell with remote-controlled syringe pumps (typically by mixing up to three streams from three syringes to control reagent concentrations).

The SEC cell is a quartz cuvette equipped with a gold honeycomb working electrode with a 1.7 mm electrode path length and gold wire as the counter electrode (Pine Research, AKSTCKIT3). A solid-state Ag/AgCl leakless (eDAQ, ET072) reference electrode was used in the SEC experiment and referenced to the ferrocenium/ferrocene (Fc+/Fc) couple by cyclic voltammetry using a 3 mm diameter glassy carbon disk working electrode and platinum wire before and after the SEC experiment. UV–Vis spectra of samples prepared at 2.5 mM fac-ReCl(bpy)(CO)3 in CH3CN containing 50 mM TBAPF6 were measured on an Agilent 8453 spectrophotometer. Samples were sparged for at least 5 mins with a steady stream of argon saturated with CH3CN vapor from a pre-bubbler. A constant potential of −1.78 V vs Fc+/0 was applied by using a CH Instruments 620 E potentiostat, and spectra were taken every 5–10 s until the solution came to equilibrium. Cyclic voltammetry was performed in a 4-neck round bottom flask using the aforementioned potentiostat and electrode in CH3CN solutions containing 100 mM TBAPF6 as the supporting electrolyte.

PR of CH3CN results in the instantaneous generation of solvated electrons, esolv, as the primary reducing radicals. esolv exists as an equilibrium mixture of two forms: a cavity electron that absorbs strongly in the near-IR region at λmax = 1450 nm and a molecular dimer anion, CH3CN2, that absorbs more weakly at λmax = 550 nm.60 Both forms of esolv are extremely potent reductants, and we have previously shown that they are efficiently scavenged by 1 in CH3CN to generate 1•− [reaction. (5)] with a near-diffusion controlled rate constant of k5 = 6.0 × 1010 M−1 s−1 by monitoring esolv decay kinetics at 1460 nm as a function of [1].61 Similar measurements in this work have reproduced this rate constant (Fig. S2),
1+ esolv-1.
(5)

The transient absorption spectrum measured between 440 and 600 nm, 2 µs after PR of a N2-saturated CH3CN solution of 1, peaks at λmax = 510 nm, with a high tail at shorter wavelengths [Fig. 1(c)]. This spectrum is assigned to the intact, one-electron reduced product, 1•−, by comparison with a previously reported spectrum of one-electron reduced fac-ReCl(4,4′-(CH3)2-bpy)(CO)3, and the fact that this complex is known to live much longer than 10 ms.62 Indeed, a kinetic trace monitored at 510 nm following PR indicates a lifetime much longer than the microsecond regime (Fig. S3). This assignment was supported by electrochemical measurements. Isolation of the first (most positive) redox event in the cyclic voltammogram of 1 demonstrates reversibility, which has previously been accepted as the bpy ligand-based reduction and oxidation [Fig. 1(a)].63 At potentials near E1/2 (∼−1.74 V vs Fc+/0) on the time scale of the CV experiment at a scan rate of 100 mV s−1, the subsequent chemical steps initiated by the loss of the chloride ligand are not observed. Further support was found by UV–Vis spectroelectrochemistry (SEC) of a solution containing 1 at an applied potential of −1.78 V vs Fc+/0. Spectra collected within 5 s of applying potential at the electrode resulted in a similar spectrum to that seen via pulse radiolysis [Fig. 1(c)]. Notably, when the potential was held during SEC at the first reversible reduction, additional spectral features were observed and assigned to the solvento complex, fac-Re(CH3CN)(bpy)(CO)3 (465 nm), and dimer, [Re(bpy)(CO)3]2 (590, 780, 940 nm), the formation of which is initiated by chloride ligand loss [Fig. 1(b)].62 

FIG. 1.

(a) Cyclic voltammogram of 1 mM 1 in CH3CN with 1 mM Fc and 100 mM TBAPF6 at 100 mV s−1; red dot denotes applied potential in SEC experiment. (b) UV–Vis SEC of 2.5 mM 1 in CH3CN with 50 mM TBAPF6 at −1.78 V vs Fc+/0 for ∼60 s of electrolysis; blue trace after ∼5 s. (c) Normalized spectra of 1•− observed by pulse radiolysis 2 µs after electron pulse (black dots) and by UV–Vis SEC after ∼5 s of electrolysis at −1.78 V vs Fc+/0 (blue line). All solutions are N2 saturated.

FIG. 1.

(a) Cyclic voltammogram of 1 mM 1 in CH3CN with 1 mM Fc and 100 mM TBAPF6 at 100 mV s−1; red dot denotes applied potential in SEC experiment. (b) UV–Vis SEC of 2.5 mM 1 in CH3CN with 50 mM TBAPF6 at −1.78 V vs Fc+/0 for ∼60 s of electrolysis; blue trace after ∼5 s. (c) Normalized spectra of 1•− observed by pulse radiolysis 2 µs after electron pulse (black dots) and by UV–Vis SEC after ∼5 s of electrolysis at −1.78 V vs Fc+/0 (blue line). All solutions are N2 saturated.

Close modal

PR was performed on a 1 M solution of N,N-dimethyl-p-toluidine (DMPT) in O2-saturated CH3CN (Fig. 2). Since PR of CH3CN generates oxidizing radicals in addition to esolv,61 depending on the oxidation potential of a solute and its concentration, it may be oxidized by these radiolytic oxidants. O2 saturation of CH3CN solutions has previously been used to scavenge esolv, thus preventing reduced solutes from forming and permitting clean spectra of one-electron oxidized solutes to be obtained following PR,64–67 although it should be noted that this has only been demonstrated for a handful of specific solutes, none of which were tertiary amines. The attendant superoxide radical that is generated, O2•−, is a very mild reductant relative to esolv.68 The transient absorption spectrum in Fig. 2 exhibits a broad band with λmax = 470 nm. This spectrum is identical to the reported spectrum of DMPT•+ in CH3CN measured by stopped-flow mixing,69 confirming that DMPT•+ is generated in the PR experiment (Scheme 2, Fig. 2).

FIG. 2.

Transient absorption spectrum of DMPT•+ obtained 400 ns following pulse radiolysis of 1 M DMPT in O2-saturated CH3CN (black dots); Lorentzian fit (red line).

FIG. 2.

Transient absorption spectrum of DMPT•+ obtained 400 ns following pulse radiolysis of 1 M DMPT in O2-saturated CH3CN (black dots); Lorentzian fit (red line).

Close modal
SCHEME 2.

Major constituents generated by PR of an O2-saturated CH3CN solution containing 1 M DMPT.

SCHEME 2.

Major constituents generated by PR of an O2-saturated CH3CN solution containing 1 M DMPT.

Close modal
Pulse radiolysis of a 1 M solution of TEA in CH3CN results in a transient absorption spectrum identical to that obtained in the absence of TEA (λmax = 510 nm) but with approximately twice the amplitude [Fig. 3(a)], indicating that the yield of 1•− is doubled. As can be seen in Fig. 3(b), the formation kinetics of 1•− change from monophasic in the absence of TEA (green trace) to biphasic in the presence of TEA (orange trace). This double exponential growth of 1•− is observed at concentrations of TEA as low as 10 mM. As [TEA] is increased, the amplitude of the fast-growth component remains constant, at the same level as in the absence of TEA, while the final plateau amplitude that is reached after the slower-growth component increases [Fig. 3(c)], and is attained on the same time scale for all TEA concentrations. The fast component is due to the scavenging of esolv by 1 [reaction (5)], and as discussed later, we assign the slower component of 1•− growth to the reduction of 1 by radiolytically generated TEArad [reaction (6)],
1+ TEArad1+TEA+.
(6)
Since the 550 nm absorption band of esolv overlaps with the spectrum of 1•−, its rapid formation and subsequent decay as it is scavenged by 1 complicate the double-exponential fitting of the 510 nm growth kinetic traces of 1•−. We therefore turned to the complementary technique of pulse radiolysis coupled with time-resolved infrared spectroscopy (PR-TRIR)59 so that we could monitor 1•− formation via one of its known ν(CO) mid-IR absorption bands,62 free from any interfering absorptions. Maintaining the concentration of TEA at 1 M and varying the concentration of 1 results in the series of 1•− PR-TRIR kinetic growth traces (measured at 1865 cm−1) shown in Fig. 3(d). In this case, with 3 mM 1, since the esolv scavenging lifetime (5.5 ns) is shorter than the response time of the TRIR detection system (∼40 ns), the fast growth component is not resolvable. However, the rate of growth of the slower component is observed to increase as a function of [1], always reaching the same final amplitude since the concentration of reducing equivalents (esolv + TEArad) is the same in each measurement. The second-order rate constant for the reduction of 1 by TEArad was determined as k6 = (4.1 ± 0.4) × 109 M−1 s−1 from the slope of a plot of the observed rate constant for the slower component growth (obtained from double exponential fits of the data; see supplementary material) vs [1] [Fig. 3(e), Fig. S4]. The related bicyclic tertiary amine 1,4-diazabicyclo[2.2.2]octane (DABCO) was also deployed as a control in place of TEA (Scheme 3). Strikingly, a kinetic trace measured at 510 nm after PR of 1 in CH3CN in the presence of 1 M DABCO is substantially different from the observed kinetics in the presence of 1 M TEA [Fig. 3(f)]. Most notably, the enhancement in total yield of 1•− and secondary growth in the kinetic trace that are observed with TEA are now absent, with only the fast scavenging of esolv by 1 [reaction (5)] occurring in the presence of DABCO.
FIG. 3.

Pulse radiolysis: (a) Transient absorption spectra obtained 500 ns after the electron pulse for 3 mM 1 in N2-saturated CH3CN in the absence (red) and presence (black) of 1 M TEA. (b) Kinetic traces measured at 510 nm on neat CH3CN (blue), 1 M TEA in CH3CN (red), 3 mM 1 in CH3CN (green), and 3 mM 1 in CH3CN with 1 M TEA (orange). (c) Kinetic traces measured at 510 nm for 3 mM 1 in CH3CN (red) and for solutions containing TEA at the following mM concentrations: 1000 (blue), 500 (green), 250 (burgundy), 95 (violet), and 47 (teal), 24 (green), 10 (orange). (d) PR-TRIR kinetic traces for variable [1] in CH3CN with 1 M TEA, observing 1•− at λmax = 1865 cm−1; [1] (mM) = 0.4 (blue), 0.6 (green), 0.8 (red), 1.4 (violet), and 1.8 (black). (e) Plots of kobs for the secondary growth component of 1•− vs [1] for the measurements with 1 M TEA (black dots) with linear regression in red (k6 = (4.1 ± 0.4) × 109 M−1 s−1), and with 1 M TEOA (black squares) with linear regression in blue (k7 = (1.2 ± 0.1) × 108 M−1 s−1). (f) Kinetic traces measured at 510 nm on 3 mM 1 in CH3CN containing 1 M TEA (blue) and 1 M DABCO (red).

FIG. 3.

Pulse radiolysis: (a) Transient absorption spectra obtained 500 ns after the electron pulse for 3 mM 1 in N2-saturated CH3CN in the absence (red) and presence (black) of 1 M TEA. (b) Kinetic traces measured at 510 nm on neat CH3CN (blue), 1 M TEA in CH3CN (red), 3 mM 1 in CH3CN (green), and 3 mM 1 in CH3CN with 1 M TEA (orange). (c) Kinetic traces measured at 510 nm for 3 mM 1 in CH3CN (red) and for solutions containing TEA at the following mM concentrations: 1000 (blue), 500 (green), 250 (burgundy), 95 (violet), and 47 (teal), 24 (green), 10 (orange). (d) PR-TRIR kinetic traces for variable [1] in CH3CN with 1 M TEA, observing 1•− at λmax = 1865 cm−1; [1] (mM) = 0.4 (blue), 0.6 (green), 0.8 (red), 1.4 (violet), and 1.8 (black). (e) Plots of kobs for the secondary growth component of 1•− vs [1] for the measurements with 1 M TEA (black dots) with linear regression in red (k6 = (4.1 ± 0.4) × 109 M−1 s−1), and with 1 M TEOA (black squares) with linear regression in blue (k7 = (1.2 ± 0.1) × 108 M−1 s−1). (f) Kinetic traces measured at 510 nm on 3 mM 1 in CH3CN containing 1 M TEA (blue) and 1 M DABCO (red).

Close modal
SCHEME 3.

Major constituents generated by PR of a N2-saturated CH3CN solution containing 1 M DABCO and 3 mM 1.

SCHEME 3.

Major constituents generated by PR of a N2-saturated CH3CN solution containing 1 M DABCO and 3 mM 1.

Close modal
A PR-TRIR experiment was also performed on 1 in CH3CN in the presence of 1.5 M TEOA as a function of [1]. It is known that PR of CH3CN solutions of weak Brønsted acids results in rapid scavenging of esolv by the acid.70 Therefore, reaction (5) does not occur in the presence of 1.5 M TEOA. However, an exponential growth of 1•− is still observed (Fig. S5), which we assign to the reduction of 1 by radiolytically generated TEOArad radicals [reaction (7)]. A plot of the observed rate constant for the growth of 1•− as a function of [1] reveals a second-order rate constant for the reduction of 1 by TEOArad, k7 = (1.2 ± 0.1) × 108 M−1 s−1 [Fig. 3(e)],
1+TEOArad1+TEOA+.
(7)
LFP-TRIR experiments were performed on 1 in CH3CN in the presence of either 2 M TEA or 2 M TEOA. In these experiments, 1•− was generated by reactions (1)–(4) (Scheme 1). Second-order rate constants for reaction (2) have previously been reported (k2(TEA) = 2.6 × 107 M−1 s−1 and k2(TEOA) = 8.0 × 107 M−1 s−1).51,56 Therefore, at the 2 M concentration of amine used, the fast growth of 1•− from the reductive quenching of 1* by TEA or TEOA occurs within the response time of the LFP-TRIR apparatus (∼25 ns). However, the subsequent reduction of 1 by TEAphot or TEOAphot radicals [reactions (8) and (9), respectively] is observed as a secondary exponential growth of 1•−, which occurs more slowly with TEOAphot than with TEAphot [Fig. 4(a) and Fig. S6]. The total yield of 1•− is governed by the amount of 1* generated upon photoexcitation, and therefore the kinetic traces reflect an increasing yield of 1•− as [1] increases. The second order rate constants for the reduction of 1 by the photochemically generated TEAphot and TEOAphot radicals [reactions (8) and (9), respectively] were determined from plots of the observed rates of the secondary growth of 1•− vs [1], as k8 = (4.7 ± 0.5) × 109 M−1 s−1 and k9 = (6.5 ± 0.6) × 107 M−1 s−1, respectively [Fig. 4(b)],
1+ TEAphot1-+TEA+,
(8)
1+ TEOAphot1-+TEOA+.
(9)
FIG. 4.

Laser flash photolysis: (a) LFP-TRIR kinetic traces for variable [1] in N2-saturated CH3CN with 2 M TEA, observing 1•− at 1865 cm−1 following 355 nm pulsed laser excitation: [1] = 0.6 mM (blue), 0.9 mM (green), 1.5 mM (red), 1.8 mM (violet), 2.4 mM (teal), and 2.8 mM (purple). (b) Plots of kobs for the secondary growth component of 1•− vs [1] for the measurements with 2 M TEA (black dots, from data in panel A) with linear regression in red [k8 = (4.7 ± 0.5) × 109 M−1 s−1] and with 2 M TEOA (black squares, from data shown in Fig. S6), with linear regression in blue [k9 = (6.5 ± 0.6) × 107 M−1 s−1], respectively.

FIG. 4.

Laser flash photolysis: (a) LFP-TRIR kinetic traces for variable [1] in N2-saturated CH3CN with 2 M TEA, observing 1•− at 1865 cm−1 following 355 nm pulsed laser excitation: [1] = 0.6 mM (blue), 0.9 mM (green), 1.5 mM (red), 1.8 mM (violet), 2.4 mM (teal), and 2.8 mM (purple). (b) Plots of kobs for the secondary growth component of 1•− vs [1] for the measurements with 2 M TEA (black dots, from data in panel A) with linear regression in red [k8 = (4.7 ± 0.5) × 109 M−1 s−1] and with 2 M TEOA (black squares, from data shown in Fig. S6), with linear regression in blue [k9 = (6.5 ± 0.6) × 107 M−1 s−1], respectively.

Close modal

The radiolysis of CH3CN results in the ionization and excitation of solvent molecules, eventually culminating in a mixture of micromolar level concentrations of solvated electrons (esolv), several neutral solvent-derived radicals (R), and protonated solvent molecules (CH3CNH+), via the sequence of rapid reactions shown in Scheme 4.61, esolv is a potent reductant, capable of rapidly reducing most dissolved solutes. While the PR and SEC techniques differ in their temporal resolution (i.e., sub-second vs seconds) and mode of electron transfer (i.e., homogenous vs interfacial), the speciation of 1•− formed upon one-electron reduction of 1 in CH3CN, either by esolv in a PR measurement [reaction (5)] or at an electrode in an SEC experiment, is identical on time scales less than 5 seconds, as determined by UV–Vis spectroscopy [Fig. 1(c)]. The high tail in the spectrum at wavelengths shorter than 510 nm is consistent with contributions from near-energy-equivalent ligand-to-metal (pπ* to dπ) and/or inter-ligand (pπ* to pπ*) transitions.71 Therefore, we can cogently assign the spectra generated independently by PR and SEC to 1•−, and we can conveniently use the absorbance maximum at 510 nm to monitor the kinetics of electron transfer reactions to generate 1•−.

SCHEME 4.

The sequence of reactions involved in the radiolysis of CH3CN by ionizing radiation. Solvent-derived radicals are shown in blue font.

SCHEME 4.

The sequence of reactions involved in the radiolysis of CH3CN by ionizing radiation. Solvent-derived radicals are shown in blue font.

Close modal

There have been reports of some solutes being simultaneously oxidized and reduced during the radiolysis of CH3CN solutions,64–67 with O2 addition being used as a way to selectively generate only solute•+ via scavenging of esolv by O2 to make O2•−. We have previously discussed that while the primary solvent radical cation, CH3CN•+, would be an extremely potent oxidant [Eo(CH3CN•+/0(sol)) ≈ +4.6 V vs SCE in CH3CN], it likely has only a fleeting existence, probably comparable to the time scale of its thermal equilibration, due to a very facile and rapid proton transfer reaction with a neutral CH3CN molecule that generates CH2CN and CH3CNH+ (Scheme 4).61 Therefore, one or more of the other solvent-derived radicals is/are more likely to be responsible for solute oxidation. However, the identity of the oxidant(s) and their oxidizing power(s) are not yet known, and this is the subject of ongoing research in our laboratory.

The one-electron oxidation of TEA by PR would offer a unique way to rapidly generate TEA•+ radical cations, which would then deprotonate according to the established pathway shown in reaction (3), Scheme 1, thus permitting a kinetic investigation of reaction (4), Scheme 1, i.e., the one-electron reduction of 1 by TEA radicals. However, we first need to establish that the PR of a CH3CN solution of TEA will result in the generation of TEA•+. Since TEA is a saturated aliphatic amine, TEA•+ has no useful optical absorption band to probe other than in the UV region, where all the other products of radiolysis will absorb too. We therefore used DMPT as a surrogate, since it is also a tertiary amine and has a very similar oxidation potential to TEA in CH3CN; Eo(TEA•+/0) = +0.96 V vs SCE72 and Eo(DMPT•+/0) = +1.1 V vs SCE.73,74 Importantly, DMPT’s aromatic ring results in DMPT•+ having a distinct absorbance in the visible region (λmax = 470 nm), as reported by Oyama and co-workers as a transient species generated using a chemical oxidant with stopped-flow mixing.69 The transient absorption spectrum we obtained 400 ns after PR of a 1 M O2-saturated solution of DMPT in CH3CN (Fig. 2) is identical to Oyama’s spectrum of DMPT•+ in CH3CN, confirming that we are oxidizing DMPT in our PR experiment. Therefore, we can expect that PR of a 1 M solution of TEA in CH3CN will also produce TEA•+ radical cations. We note that TEA has minimal reactivity toward esolv [Fig. 3(b) and Fig. S7], and therefore, a PR measurement on 1 in the presence of TEA results in esolv being exclusively scavenged by 1 [reaction (5), k5 = 6.0 × 1010 M−1 s−1; τ5 = 5.5 ns with 3 mM 1]. In addition to the indirect oxidation of a solute in a PR experiment by radiolytically generated solvent radicals, if it is present at a molar level concentration (e.g., 1 M TEA), a fraction may also be directly ionized by the electron pulse (Scheme 5). For 1 M TEA in CH3CN, a simple estimation based on the relative concentrations of electrons present in the TEA and CH3CN molecules predicts that ∼15% of ionizations will occur at TEA molecules (see supplementary material). In any case, regardless of the mode of oxidation of TEA, both will result in the same TEA•+ species, and it is inconsequential for our experiments.

SCHEME 5.

Pulse radiolysis of a solution can result in (a) indirect oxidation of the solute by radiolytically generated, solvent-derived oxidizing radicals ([Ox]), and (b) direct oxidation (ionization) of the solute by the electron pulse. The direct pathway typically only becomes relevant at high solute concentrations (molar levels).

SCHEME 5.

Pulse radiolysis of a solution can result in (a) indirect oxidation of the solute by radiolytically generated, solvent-derived oxidizing radicals ([Ox]), and (b) direct oxidation (ionization) of the solute by the electron pulse. The direct pathway typically only becomes relevant at high solute concentrations (molar levels).

Close modal

Having established that TEA will be oxidized by PR of a CH3CN solution, it is reasonable to assume that the well-established deprotonation reaction, reaction (3), will follow to generate the strongly reducing α-amino radical, TEA. We note that this reaction is sometimes also referred to in the literature as an H-atom transfer (HAT) from TEA to TEA•+,75 as opposed to a PT from TEA•+ to TEA, as we have shown in reaction (3). While the exact nature of this reaction is an interesting conundrum, it is not important for our experiments since both pathways arrive at identical products (TEA + TEAH+). Goez used photo-CIDNP to estimate a range of 1 × 108–2 × 109 M−1 s−1 for the PT rate constant [reaction (3)], and therefore, with 1 M TEA, this reaction occurs rapidly on a timescale of 0.5–10 ns.41 The reducing power of the TEA radical in CH3CN was previously estimated by Wayner and Griller using modulated photolysis to generate TEA and phase-sensitive voltammetry to measure its potential.76 This resulted in a value of E1/2(TEA+/•) = −1.12 V vs SCE, which would imply that TEA is incapable of reducing 1 since E1/2(10/•−) = −1.34 V vs SCE in CH3CN. Flash photolysis experiments by Kutal and Ferraudi on the almost identical ReBr(bpy)(CO)3 complex in N,N-dimethylformamide (DMF) in the presence of TEA also implied that photochemically generated TEA does not reduce that particular Re complex in DMF.52 Nevertheless, we felt it was important to test whether or not TEA can reduce 1 in CH3CN since it is often referred to as a strongly reducing radical in the literature, and if it is shown to reduce 1, this would have implications for the design of experiments in solar fuels and photoredox catalysis.

The doubling in the amplitude of the 510 nm band of 1•− that we observed following PR of a 3 mM CH3CN solution of 1 in the presence of 1 M TEA, relative to in the absence of TEA [Fig. 3(a)], strongly implies that 1 is being further reduced by a species derived from TEA, since esolv is the only solvent-derived reductant (Scheme 6), and it will be quantitatively scavenged by 1, even in the presence of TEA. As discussed earlier, the most obvious candidate for this additional reductant is TEA, which immediately puts into question the previously estimated E1/2(TEA+/•) = −1.12 V vs SCE value.76 Probing the formation of 1•− at 510 nm revealed two-step biphasic kinetics [Fig. 3(b)], with the fast step being due to the reduction of 1 by esolv [reaction (5)] and the slower step assigned to the reduction of 1 by radiolytically generated TEArad radicals [reaction (6)]. The dependence of the final amplitude of the 510 nm kinetic trace, i.e., the final concentration of 1•−, on the concentration of TEA used [Fig. 3(c)] is consistent with more efficient oxidation of TEA (both indirect and direct, see Scheme 5) as its concentration is increased and, therefore, more TEArad being produced by reaction (3). As explained in the Results section, we turned to PR-TRIR so that we could more easily probe the kinetics of 1•− formation via one of its well-known ν(CO) IR bands62 without any interference from esolv, one form of which absorbs in the 500 nm region.60 This allowed the clean double-exponential kinetic fitting of the 1865 cm−1 PR-TRIR kinetic traces shown in Fig. 3(d) and Fig. S4 as a function of [1], resulting in a straight line fit of the observed rate of the secondary growth component [Fig. 3(e)], and a second-order rate constant for the reduction of 1 by TEArad of k6 = (4.1 ± 0.4) × 109 M−1 s−1.

SCHEME 6.

Mechanism of the reduction of 1 by esolv and radiolytically generated TEArad or TEOArad radicals produced by PR of a CH3CN solution of 1 in the presence of TEA or TEOA. [Ox] represents one or more of the radiolytically generated, solvent-derived oxidizing radicals. Note that in the presence of TEOA, 1 does not react with esolv. Parentheses designate the chemical formula and associated rate constants pertaining to TEOA.

SCHEME 6.

Mechanism of the reduction of 1 by esolv and radiolytically generated TEArad or TEOArad radicals produced by PR of a CH3CN solution of 1 in the presence of TEA or TEOA. [Ox] represents one or more of the radiolytically generated, solvent-derived oxidizing radicals. Note that in the presence of TEOA, 1 does not react with esolv. Parentheses designate the chemical formula and associated rate constants pertaining to TEOA.

Close modal

Our data provide strong evidence that TEA is capable of very efficiently reducing 1 in CH3CN, with a rate constant that is not far below the diffusion-controlled limit of ∼1010 M−1 s−1. However, as a control, we repeated the PR measurements with the related bicyclic tertiary diamine, 1,4-diazabicyclo[2.2.2]octane (DABCO), which is shown in Scheme 3. DABCO has an oxidation potential that is slightly more negative than that of TEA,77 and as such, it will be even easier to oxidize in a PR experiment than TEA. However, unlike the straight-chained aliphatic amine radical cation, TEA•+, DABCO•+ is rather stable and does not engage in a PT reaction with another DABCO molecule (Scheme 3).78 As such, α-amino DABCO radicals, which would be analogous to TEA radicals, are not generated. Nelsen identified reversibility in the redox process associated with the DABCO•+/0 couple in CH3CN, suggesting that DABCO•+ is stable on the time scale of the cyclic voltammetry experiment.77 McKinney and Geske independently demonstrated that the enhanced stability of DABCO•+ is rooted in spin state delocalization, primarily between both nitrogen atoms.79 The lack of a secondary growth in the 510 nm kinetic trace of 1•− formation following PR of a 3 mM solution of 1 in CH3CN in the presence of 1 M DABCO [Fig. 3(f)] clearly shows that the secondary reduction of 1 requires the presence of an amine that is capable of generating reductive α-amino radicals, such as TEA. This further supports our assignment of TEA as the secondary reductant.

The related tertiary amine, TEOA, has also been extensively used as a SED in photochemical CO2 reduction over the course of several decades, perhaps more so than TEA. It is recognized that, like TEA, it can also act as a one-photon/two-electron donor when used to reductively quench the photoinduced excited state of a photocatalyst or sensitizer via the formation of a strongly reducing, carbon-based TEOA radical.51–53,80 However, while the oxidative chemistry of TEOA is very similar to that of TEA, there is a subtle difference in that the PT (or HAT) reaction that occurs between TEOA•+ and another molecule of TEOA to generate TEOA can lead to two possible radicals: one α-amino and the other α-hydroxy (Scheme 7). The α-amino-TEOA radical is analogous to the α-amino-TEA radical and will transform into an iminium cation when it transfers an electron to a substrate. However, the α-hydroxy-TEOA radical will transform into an aldehyde upon substrate reduction, accompanied by the release of a proton. It has been shown that α-aminoalkyl radicals are highly stabilized and become more stable with increasing C- or N-alkylation,81 and it has been suggested that α-amino-TEOA will be the dominant form of the TEOA radical.75 However, to the best of our knowledge, no attempt has been made to quantify this, and the actual distribution of the two forms is unknown.

SCHEME 7.

Degradation pathways for the TEOA•+ radical cation via its reaction with TEOA.80 

SCHEME 7.

Degradation pathways for the TEOA•+ radical cation via its reaction with TEOA.80 

Close modal

The redox potentials for the one-electron oxidation of TEA and TEOA have been reported in CH3CN as E(TEA•+/0) = +0.96 V72 and E(TEOA•+/0) = +0.59 V vs SCE.82 It has thus been assumed in the literature55 that a similar trend will also apply for the potentials associated with the one-electron oxidation of the TEA and TEOA radicals and, therefore, that TEOA should be the superior electron donor. To test this hypothesis, we repeated our PR-TRIR measurements on 1 in CH3CN but in the presence of TEOA. Since Brønsted acids such as TEOA rapidly scavenge esolv in CH3CN,70 reaction (5) is shut down, and the 1865 cm−1 PR-TRIR kinetic traces associated with the growth of 1•−, therefore, do not contain an initial fast component due to reaction (5). Instead, they exhibit exponential growth on a much longer timescale than that observed for TEArad [compare Fig. 3(d) and Fig. S5], with a rate constant for the reduction of 1 by radiolytically generated TEOArad of k7 = (1.2 ± 0.1) × 108 M−1 s−1, which is a factor of 34× lower than the rate constant for reduction by TEArad [k6 = (4.1 ± 0.4) × 109 M−1 s−1]. This is a significant decrease and suggests that, in contrast to previous assumptions, TEOA is actually a weaker reductant than TEA. An overview of the reduction of 1 by esolv and radiolytically generated TEArad or TEOArad radicals is shown in Scheme 6.

Since TEOA reacts with esolv in the PR experiments and we cannot rule out the possibility that the product of this reaction could also be a reductive species that may contribute to the observed rate of formation of 1•−, we performed complementary LFP experiments on 1 in CH3CN in the presence of TEA or TEOA, where 1•− is produced according to reactions (1–4), Scheme 1. Again, in contrast to Kutal and Ferraudi’s results in DMF,52,53 we did observe a secondary growth of 1•− due to the reduction of 1 by photochemically generated TEAphot radicals [Fig. 4(a)], with a rate constant, k8 = (4.7 ± 0.5) × 109 M−1 s−1 [Fig. 4(b)]. This is very similar to the rate constant we measured by PR-TRIR [k6 = (4.1 ± 0.4) × 109 M−1 s−1]. Exponential fitting of the analogous kinetic traces in the presence of TEOA (see supplementary material) resulted in a second-order rate constant for the reduction of 1 by photochemically generated TEOAphot radicals [k9 = (6.5 ± 0.6) × 107 M−1 s−1] that is slightly lower (by a factor of 1.8×) than the equivalent rate constant we measured by PR-TRIR with radiolytically generated TEOArad radicals [k7 = (1.2 ± 0.1) × 108 M−1 s−1]. However, as we pointed out earlier, there is a possibility that an additional reductant is generated in the PR experiments with TEOA, which could contribute to the observed rate of 1•− formation and thus the slight discrepancy between k7 and k9. Nevertheless, the LFP experiments confirmed that TEOAphot radicals are indeed considerably less efficient than TEAphot radicals at reducing 1 in CH3CN. Combining our data from PR and LFP gives mean values for the second-order rate constants for the reduction of 1 by TEA [kTEA• = (4.4 ± 0.3) × 109 M−1 s−1] and TEOA [kTEOA• = (9.3 ± 0.6) × 107 M−1 s−1] and, therefore, kTEA•/kTEOA• = 47 ± 0.3. An overview of the reduction of 1 by photochemically generated TEAphot or TEOAphot radicals is shown in Scheme 8.

SCHEME 8.

Mechanism of the reduction 1* by TEA and TEOA and the reduction of 1 by the TEAphot and TEOAphot radicals that are generated by fragmentation of TEA•+ and TEOA•+, probed by LFP of a solution of 1 in the presence of TEA or TEOA. Parentheses designate the chemical formula and associated rate constants pertaining to TEOA.

SCHEME 8.

Mechanism of the reduction 1* by TEA and TEOA and the reduction of 1 by the TEAphot and TEOAphot radicals that are generated by fragmentation of TEA•+ and TEOA•+, probed by LFP of a solution of 1 in the presence of TEA or TEOA. Parentheses designate the chemical formula and associated rate constants pertaining to TEOA.

Close modal
While it has long been known that TEA and TEOA can both act as one-photon/two-electron SEDs in photocatalysis reactions through reductive quenching of an excited state followed by a second electron transfer from the carbon-based TEA or TEOA radicals that are produced by degradation of their radical cations, very little is known about the redox properties of the TEA and TEOA radicals. Knowledge of the reducing powers of these radicals would be tremendously useful when designing a photocatalytic reaction process so that the appropriate SED could be chosen, the radical of which would be capable of efficiently reducing the catalyst, thus ensuring two-electron donation from the SED and maximizing the overall quantum efficiency. Our results show that the experimental estimate of E1/2(TEA+/•) = −1.12 V vs SCE in CH3CN76 is too positive and that, in reality, it must be more negative than −1.34 V vs SCE (the reduction potential of 1). Furthermore, our discovery that TEOA transfers an electron to 1 with a rate constant that is ∼50× lower than from TEA implies that TEOA is a weaker reductant than TEA, but that E1/2(TEOA+/•) is still more negative than −1.34 V vs SCE in CH3CN. This assumes that the majority of the TEOA radicals exist in the α-amino-TEOA form (Scheme 7). If this assumption is true, we can make a simple estimate of the difference between the reduction potentials of TEA+ and TEOA+ by using the Marcus cross relation [Eq. (10)],83 
kxy=(kxxkyyKxyfxy)1/2,
(10)
where kxy is the rate constant for the cross reaction involving the reduction of 1 by amine [reaction (4)], kxx is the self-exchange rate constant for the reduction of 1 by 1•−, kyy is the self-exchange rate constant for the reduction of amine+ by amine, Kxy is the equilibrium constant for the cross reaction involving the reduction of 1 by amine [reaction (4)], lnfxy = (lnKxy)2/(4 ln{(kxxkyy)/νn2}), and νn is the nuclear frequency factor. Again, if TEOA exists mainly in the α-amino-TEOA form, it is a reasonable assumption that kyy will be similar for the two amines, and for cross reactions with relatively small ΔEo values, fxy is ∼1. Based on these assumptions, from the ratio of the measured rate constants for the reduction of 1 by TEA and TEOA (kTEA•/kTEOA• = 47 ± 0.3), Eq. (10) can be used to estimate the ratio of the equilibrium constants for reaction (4) for the two amines, resulting in KTEA•/KTEOA• = 2209 ± 28. Since the electrochemical potential for a redox reaction is ΔE = 2.303(RT/F)log K, where R is the molar gas constant, T is the temperature, F is the Faraday constant, and K is the equilibrium constant, the difference between the reduction potentials of TEA+/• and TEOA+/• can be estimated as ΔE = E1/2(TEA+/•) − E1/2(TEOA+/•) ≈ −0.2 V. We note that this is only a very simple estimation, and in future work, we will attempt to experimentally determine the reduction potentials of TEA+ and TEOA+ through redox equilibria measurements with a family of Re complexes with different reduction potentials. The smaller rate constant for the reduction of 1 by TEOA compared to by TEA might be due to the electronegative oxygen atoms in TEOA inductively withdrawing electron density from the central N-atom in the α-amino form of TEOA. However, inductive effects rapidly dissipate with distance, so we cannot rule out the possibility that a sufficient population of α-hydroxy TEOA is present to affect the observed rate constant since α-hydroxy-TEOA will have a significantly different reactivity toward 1 and experience stronger inductive withdrawing effects.

The rate of reduction of a catalyst by an SED, or a radical derived from it, is only one issue that needs to be considered when choosing an SED for photocatalysis. There are other factors involved, such as solubility and the propensity of the SED to become involved in the catalytic reaction itself. For example, in photocatalytic CO2 reduction with catalysts such as 1, it is now known that a deprotonated TEOA actually binds to the Re-center, and then a CO2 molecule inserts into the Re–O bond to form a Re-carbonate species to initiate catalysis.84 This cannot occur with TEA as the SED and may be one reason why TEOA has been used more in photochemical CO2 reduction reactions than TEA.

PR is an incredibly powerful technique for the rapid production of various ions and radical species and their subsequent interrogation by time-resolved spectroscopy, and in many ways, it can simplify mechanistic investigations. For example, a reduced catalyst can easily be prepared via its reduction by esolv without the need for a photosensitizer. Unfortunately, several undesirable solvent-derived radicals are also generated during radiolysis (see Scheme 4 for CH3CN radiolysis products). In CH3CN PR, we typically add formate anion (HCO2) to the solution since the relatively weak C–H bond in HCO2 causes it to engage in H-atom transfers with about 60% of the radiolytic solvent radicals, neutralizing them and forming CO2•− in the process.61 Fortunately, CO2•− is a strong reductant, so the yield of one-electron reduced solute is boosted beyond what is achieved from esolv scavenging alone due to the secondary reduction of the solute by CO2•−.

While HCO2 has proved to be a valuable radical scavenger for reductive PR studies in CH3CN, there are still some issues. For example, as mentioned earlier, it does not scavenge all the radiolytic radicals. In addition, it can often react with the solute under investigation prior to radiolysis. The radiolytic oxidative chemistry of aliphatic tertiary amines such as TEA investigated in this work suggests that this class of compounds holds great promise as an alternative radiolytic radical scavenger/suppressor in CH3CN PR, while also providing a boost in the yield of one-electron reduced solute via the formation of strongly reducing α-amino amine radicals. TEA functions differently than HCO261 in that it does not engage in H-atom transfer reactions with the solvent radicals. Instead, it intercepts one or more of the oxidizing solvent radicals (or is directly ionized by the electron pulse, Scheme 5) and is thus oxidized to TEA•+. The established oxidative degradation chemistry of TEA•+ then occurs, resulting in the formation of the strongly reducing TEArad radical as a useful secondary reductant [Eq. (3)]. Therefore, TEA is essentially being used to convert some of the oxidizing radicals into reductants. This radical scavenging process reduces the concentration of solvent radicals, and if at a high concentration of TEA (e.g., ≥1 M), some of the solvent radical cations, CH3CN•+, also happen to be intercepted, it will prevent the formation of some of the other radicals in the first place (Scheme 4). TEA is also a strong base in CH3CN [pKa(TEAH+) = 18.46],85 so it will efficiently scavenge the radiolytic protons (CH3CNH+). Tertiary amines that are more sterically hindered than TEA could also be used to minimize interactions with the solute under investigation. We plan to investigate a variety of tertiary amines for their performance as radiolytic radical scavengers for PR in CH3CN and other organic solvents.

The aliphatic tertiary amines, TEA and TEOA, are ubiquitous in solar fuels and photoredox catalysis schemes, where they are commonly employed as SEDs for the initiation and study of reductive processes. It is well established that TEA and TEOA are one-photon/two-electron donors, since they can transfer a first electron to the photoinduced excited state of a photocatalyst or sensitizer, after which their radical cation decomposes into a carbon-based radical, TEA or TEOA, which is strongly reducing and can transfer a second electron to a catalyst. However, little is known about the redox properties of the TEA and TEOA radicals, with one measurement of E1/2(TEA+/•) = −1.12 V vs SCE in CH3CN and a suggestion that TEOA will be a stronger reductant than TEA. In this work, we have used a combination of pulse radiolysis and laser flash photolysis on CH3CN solutions of a common CO2 reduction photocatalyst, fac-[ReCl(bpy)(CO)3] (1), in the presence of TEA or TEOA, to generate TEA and TEOA radicals and probe their electron transfer reactivity toward 1. In contrast to a previously reported LFP measurement in DMF, we find that both TEA and TEOA can transfer an electron to 1 in CH3CN, with rate constants that differ by a factor of ∼50×, i.e., kTEA• = (4.4 ± 0.3) × 109 M−1 s−1 and kTEOA• = (9.3 ± 0.6) × 107 M−1 s−1. Therefore, TEA radicals are more efficient than TEOA at reducing 1 in CH3CN, and since they can reduce 1, we can revise the value of E1/2(TEA+/•) down to <−1.34 V vs SCE (the reduction potential of 1). A simple estimation based on the Marcus cross relation suggests that E1/2(TEOA+/•) will be ∼0.2 V more positive than E1/2(TEA+/•). This work has afforded mechanistic insight into tertiary amines that are commonly used in photocatalyst applications. While it unequivocally indicates that TEA and TEOA participate in catalyst activation reduction steps, it also implies their likely participation as reductants toward catalyst intermediates within the photocatalytic cycle. Knowledge of the redox properties of such radicals is essential when designing a photocatalysis system.

This work has also highlighted the fact that TEA has utility as a promising scavenger of radiolytically generated, solvent-derived oxidizing radicals in pulse radiolysis measurements of CH3CN solutions. Oxidation of TEA by one or more of the radiolytic radicals results in TEA•+, which rapidly decomposes into a strongly reducing TEA radical by proton transfer to another TEA molecule. Therefore, TEA will reduce the concentration of deleterious radicals by converting some of the oxidizing radicals into additional reducing equivalents in the form of TEA, thus boosting the yield of a one-electron reduced solute beyond what is achieved by esolv scavenging alone. We are currently investigating TEA and related tertiary amines for their effectiveness at radiolytic radical scavenging.

The supplementary material contains additional transient absorption and TRIR kinetic traces obtained from pulse radiolysis and laser flash photolysis experiments and their kinetic fits, details of the direct ionization calculations, and a description of the kinetic models employed in the exponential fits of the data.

This work and use of the Laser Electron Accelerator Facility of the Accelerator Center for Energy Research at BNL were supported by the U.S. Department of Energy (DOE), Office of Science, Office of Basic Energy Sciences, Division of Chemical Sciences, Geosciences and Biosciences under contract No. DE-SC0012704. M.A.V. was supported by the U.S. Department of Energy, Office of Science, Office of Workforce Development for Teachers and Scientists (WDTS) under the Science Undergraduate Laboratory Internships (SULI) program. We acknowledge Dr. Andrew Cook and Dr. James Wishart for their insightful discussions. This was produced by Brookhaven Science Associates, LLC under Contract No. DE-SC0012704 with the US Department of Energy. The US Government retains a non-exclusive, paid-up, irrevocable, world-wide license to publish or reproduce the published form of this paper, or allow others to do so, for US Government purposes.

The authors have no conflicts to disclose.

Cody R. Carr: Investigation (equal); Supervision (equal); Visualization (equal); Writing – original draft (equal); Writing – review & editing (equal). Michael A. Vrionides: Investigation (supporting); Visualization (supporting); Writing – review & editing (supporting). David C. Grills: Conceptualization (lead); Investigation (lead); Project administration (lead); Supervision (lead); Visualization (equal); Writing – original draft (equal); Writing – review & editing (equal).

The data that support the findings of this study are openly available in Zenodo at https://doi.org/10.5281/zenodo.10070473

1.
J.
Hawecker
,
J.-M.
Lehn
, and
R.
Ziessel
, “
Electrocatalytic reduction of carbon dioxide mediated by Re(bipy)(CO)3Cl (bipy = 2,2′-bipyridine)
,”
J. Chem. Soc., Chem. Commun.
(
6
),
328
330
(
1984
).
2.
J.
Hawecker
,
J.-M.
Lehn
, and
R.
Ziessel
, “
Efficient photochemical reduction of CO2 to CO by visible light irradiation of systems containing Re(bipy)(CO)3X or Ru(bipy)32+–Co2+ combinations as homogeneous catalysts
,”
J. Chem. Soc., Chem. Commun.
(
9
),
536
538
(
1983
).
3.
S. A.
Chabolla
,
C. W.
Machan
,
J.
Yin
,
E. A.
Dellamary
,
S.
Sahu
,
N. C.
Gianneschi
,
M. K.
Gilson
,
F. A.
Tezcan
, and
C. P.
Kubiak
, “
Bio-inspired CO2 reduction by a rhenium tricarbonyl bipyridine-based catalyst appended to amino acids and peptidic platforms: Incorporating proton relays and hydrogen-bonding functional groups
,”
Faraday Discuss.
198
,
279
300
(
2017
).
4.
S. Y.
Sung
,
D.
Kumar
,
M.
Gil-Sepulcre
, and
M.
Nippe
, “
Electrocatalytic CO2 reduction by imidazolium-functionalized molecular catalysts
,”
J. Am. Chem. Soc.
139
(
40
),
13993
13996
(
2017
).
5.
M. L.
Clark
,
P. L.
Cheung
,
M.
Lessio
,
E. A.
Carter
, and
C. P.
Kubiak
, “
Kinetic and mechanistic effects of bipyridine (bpy) substituent, labile ligand, and Brønsted acid on electrocatalytic CO2 reduction by Re(bpy) complexes
,”
ACS Catal.
8
(
3
),
2021
2029
(
2018
).
6.
D. C.
Grills
,
M. Z.
Ertem
,
M.
McKinnon
,
K. T.
Ngo
, and
J.
Rochford
, “
Mechanistic aspects of CO2 reduction catalysis with manganese-based molecular catalysts
,”
Coord. Chem. Rev.
374
,
173
217
(
2018
).
7.
A. N.
Hellman
,
R.
Haiges
, and
S. C.
Marinescu
, “
Rhenium bipyridine catalysts with hydrogen bonding pendant amines for CO2 reduction
,”
Dalton Trans.
48
(
38
),
14251
14255
(
2019
).
8.
M. R.
Madsen
,
J. B.
Jakobsen
,
M. H.
Rønne
,
H.
Liang
,
H. C. D.
Hammershøj
,
P.
Nørby
,
S. U.
Pedersen
,
T.
Skrydstrup
, and
K.
Daasbjerg
, “
Evaluation of the electrocatalytic reduction of carbon dioxide using rhenium and ruthenium bipyridine catalysts bearing pendant amines in the secondary coordination sphere
,”
Organometallics
39
(
9
),
1480
1490
(
2020
).
9.
J.
Mukherjee
and
I.
Siewert
, “
Manganese and rhenium tricarbonyl complexes equipped with proton relays in the electrochemical CO2 reduction reaction
,”
Eur. J. Inorg. Chem.
2020
(
46
),
4319
4333
(
2020
).
10.
E.
Fujita
,
D. C.
Grills
,
G. F.
Manbeck
, and
D. E.
Polyansky
, “
Understanding the role of inter- and intramolecular promoters in electro- and photochemical CO2 reduction using Mn, Re, and Ru catalysts
,”
Acc. Chem. Res.
55
(
5
),
616
628
(
2022
).
11.
L.
Rotundo
,
S.
Ahmad
,
C.
Cappuccino
,
D. E.
Polyansky
,
M. Z.
Ertem
, and
G. F.
Manbeck
, “
A dicationic fac-Re(bpy)(CO)3Cl for CO2 electroreduction at a reduced overpotential
,”
Inorg. Chem.
62
(
20
),
7877
7889
(
2023
).
12.
Y.
Yamazaki
,
M.
Miyaji
, and
O.
Ishitani
, “
Utilization of low-concentration CO2 with molecular catalysts assisted by CO2-capturing ability of catalysts, additives, or reaction media
,”
J. Am. Chem. Soc.
144
(
15
),
6640
6660
(
2022
).
13.
H.
Takeda
,
K.
Koike
,
T.
Morimoto
,
H.
Inumaru
, and
O.
Ishitani
, “
Photochemistry and photocatalysis of rhenium(I) diimine complexes
,” in
Advances in Inorganic Chemistry
,
Inorganic Photochemistry Vol. 63
, edited by
R.
van Eldik
and
G.
Stochel
(
Academic Press
,
2011
), pp.
137
186
.
14.
Z.
Guo
and
H.
Yang
, “
Homogeneous catalytic CO2 photo/electro-conversion
,” in
CO2 Conversion and Utilization
, edited by Z. Zhang (
Wiley-VCH
,
2023
), pp.
177
188
.
15.
G. A.
Lowe
, “
Enabling artificial photosynthesis systems with molecular recycling: A review of photo- and electrochemical methods for regenerating organic sacrificial electron donors
,”
Beilstein J. Org. Chem.
19
,
1198
1215
(
2023
).
16.
Y.
Kuramochi
,
Y.
Suzuki
,
S.
Asai
,
T.
Suzuki
,
H.
Iwama
,
M. S.
Asano
, and
A.
Satake
, “
Significance of the connecting position between Zn(II) porphyrin and Re(I) bipyridine tricarbonyl complex units in dyads for room-temperature phosphorescence and photocatalytic CO2 reduction: Unexpected enhancement by triethanolamine in catalytic activity
,”
Chem. Sci.
14
(
33
),
8743
8765
(
2023
).
17.
S.
Navalón
,
A.
Dhakshinamoorthy
,
M.
Álvaro
,
B.
Ferrer
, and
H.
García
, “
Metal–organic frameworks as photocatalysts for solar-driven overall water splitting
,”
Chem. Rev.
123
(
1
),
445
490
(
2023
).
18.
B.
Domingo-Tafalla
,
T.
Chatterjee
,
F.
Franco
,
J.
Perez Hernandez
,
E.
Martinez-Ferrero
,
P.
Ballester
, and
E.
Palomares
, “
Electro- and photoinduced interfacial charge transfers in nanocrystalline mesoporous TiO2 and TiO2/iron porphyrin sensitized films under CO2 reduction catalysis
,”
ACS Appl. Mater. Interfaces
15
(
11
),
14304
14315
(
2023
).
19.
Q.-S.
Huang
,
C.
Chu
,
Q.
Li
,
Q.
Liu
,
X.
Liu
,
J.
Sun
,
B.-J.
Ni
, and
S.
Mao
, “
Three-phase interface construction on hydrophobic carbonaceous catalysts for highly active and selective photocatalytic CO2 conversion
,”
ACS Catal.
13
(
17
),
11232
11243
(
2023
).
20.
P.-Y.
Ho
,
S.-C.
Cheng
,
F.
Yu
,
Y.-Y.
Yeung
,
W.-X.
Ni
,
C.-C.
Ko
,
C.-F.
Leung
,
T.-C.
Lau
, and
M.
Robert
, “
Light-Driven reduction of CO2 to CO in water with a cobalt molecular catalyst and an organic sensitizer
,”
ACS Catal.
13
(
9
),
5979
5985
(
2023
).
21.
J.
Canivet
and
F. M.
Wisser
, “
Metal–organic framework catalysts for solar fuels: Light-driven conversion of carbon dioxide into formic acid
,”
ACS Appl. Energy Mater.
6
(
18
),
9027
9043
(
2023
).
22.
B.
Kim
,
D.
Kwon
,
J.-O.
Baeg
,
M.
Austeria P
,
G. H.
Gu
,
J.-H.
Lee
,
J.
Jeong
,
W.
Kim
, and
W.
Choi
, “
Dual-atom-Site Sn-Cu/C3N4 photocatalyst selectively produces formaldehyde from CO2 reduction
,”
Adv. Funct. Mater.
33
(
19
),
2212453
(
2023
).
23.
L.
Wang
,
H.
Zhang
,
Z.
Zhang
,
J.
Zhang
,
Y.
He
,
Q.
Li
,
J.
Bao
,
M.
Fang
, and
Y.
Wu
, “
Highly efficient and selective visible-light photocatalytic CO2 reduction to CO using a 2D Co(II)-Imidazole MOF as cocatalyst and Ru(bpy)3Cl2 as photosensitizer
,”
Chem.: Asian J.
18
(
15
),
e202300297
(
2023
).
24.
K.
Kaya
,
D.
Ditz
,
A.
Jaworski
,
J.
Chen
,
S.
Monti
,
G.
Barcaro
,
S.
Budnyk
,
A.
Slabon
, and
R.
Palkovits
, “
Enhanced solar CO2 photoreduction to formic acid by platinum immobilization on bipyridine covalent triazine framework with defects
,”
Adv. Sustainable Syst.
7
(
8
),
2300071
(
2023
).
25.
L.
Wang
,
L.
Wang
,
S.
Yuan
,
L.
Song
,
H.
Ren
,
Y.
Xu
,
M.
He
,
Y.
Zhang
,
H.
Wang
,
Y.
Huang
,
T.
Wei
,
J.
Zhang
,
Y.
Himeda
, and
Z.
Fan
, “
Covalently-bonded single-site Ru-N2 knitted into covalent triazine frameworks for boosting photocatalytic CO2 reduction
,”
Appl. Catal., B
322
,
122097
(
2023
).
26.
Y.
Kinoshita
,
N.
Deromachi
,
T.
Kajiwara
,
T.-a.
Koizumi
,
S.
Kitagawa
,
H.
Tamiaki
, and
K.
Tanaka
, “
Photoinduced catalytic organic-hydride transfer to CO2 mediated with ruthenium complexes as NAD+/NADH redox couple models
,”
ChemSusChem
16
(
6
),
e202300032
(
2023
).
27.
Z.-L.
Xie
,
N.
Gupta
,
J.
Niklas
,
O. G.
Poluektov
,
V. M.
Lynch
,
K. D.
Glusac
, and
K. L.
Mulfort
, “
Photochemical charge accumulation in a heteroleptic copper(I)-anthraquinone molecular dyad via proton-coupled electron transfer
,”
Chem. Sci.
14
(
37
),
10219
10235
(
2023
).
28.
H. D.
Huerta-Zerón
,
N.
Rockstroh
,
M.
Lang
,
A.-E.
Surkus
,
V.
Brüser
,
S.
Lochbrunner
,
H.
Junge
, and
M.
Beller
, “
Photocatalytic CO2 reduction with a TiO2-supported copper photosensitizer and an iron-based CO2 reduction catalyst
,”
Catal. Sci. Technol.
13
(
13
),
3940
3945
(
2023
).
29.
C.
Su
,
Z.
Chen
,
Q.
Feng
,
F.
Wei
,
A.
Mo
,
H.-H.
Huang
,
H.
Hu
,
H.
Zou
,
F.
Liang
, and
D.
Liu
, “
Electronic effects promoted the catalytic activities of binuclear Co(II) complexes for visible-light-driven CO2 reduction in a water-containing system
,”
Dalton Trans.
52
(
14
),
4548
4553
(
2023
).
30.
M.
Takahashi
,
T.
Asatani
,
T.
Morimoto
,
Y.
Kamakura
,
K.
Fujii
,
M.
Yashima
,
N.
Hosokawa
,
Y.
Tamaki
, and
O.
Ishitani
, “
Supramolecular multi-electron redox photosensitisers comprising a ring-shaped Re(I) tetranuclear complex and a polyoxometalate
,”
Chem. Sci.
14
(
3
),
691
704
(
2023
).
31.
P.
De La Torre
,
J. S.
Derrick
,
A.
Snider
,
P. T.
Smith
,
M.
Loipersberger
,
M.
Head-Gordon
, and
C. J.
Chang
, “
Exchange coupling determines metal-dependent efficiency for iron- and cobalt-catalyzed photochemical CO2 reduction
,”
ACS Catal.
12
(
14
),
8484
8493
(
2022
).
32.
Q.
Lei
,
H.
Yuan
,
J.
Du
,
M.
Ming
,
S.
Yang
,
Y.
Chen
,
J.
Lei
, and
Z.
Han
, “
Photocatalytic CO2 reduction with aminoanthraquinone organic dyes
,”
Nat. Commun.
14
(
1
),
1087
(
2023
).
33.
J. W.
Beatty
and
C. R. J.
Stephenson
, “
Amine functionalization via oxidative photoredox catalysis: Methodology development and complex molecule synthesis
,”
Acc. Chem. Res.
48
(
5
),
1474
1484
(
2015
).
34.
J. C.
Bawden
,
P. S.
Francis
,
S.
DiLuzio
,
D. J.
Hayne
,
E. H.
Doeven
,
J.
Truong
,
R.
Alexander
,
L. C.
Henderson
,
D. E.
Gómez
,
M.
Massi
,
B. I.
Armstrong
,
F. A.
Draper
,
S.
Bernhard
, and
T. U.
Connell
, “
Reinterpreting the fate of iridium(III) Photocatalysts—Screening a combinatorial library to explore light-driven side-reactions
,”
J. Am. Chem. Soc.
144
(
25
),
11189
11202
(
2022
).
35.
S. L.
Goldschmid
,
N. E.
Soon Tay
,
C. L.
Joe
,
B. C.
Lainhart
,
T. C.
Sherwood
,
E. M.
Simmons
,
M.
Sezen-Edmonds
, and
T.
Rovis
, “
Overcoming photochemical limitations in metallaphotoredox catalysis: Red-light-driven C–N cross-coupling
,”
J. Am. Chem. Soc.
144
(
49
),
22409
22415
(
2022
).
36.
P. J.
DeLaive
,
T. K.
Foreman
,
C.
Giannotti
, and
D. G.
Whitten
, “
Photoinduced electron transfer reactions of transition-metal complexes with amines. Mechanistic studies of alternate pathways to back electron transfer
,”
J. Am. Chem. Soc.
102
(
17
),
5627
5631
(
1980
).
37.
D. G.
Whitten
, “
Photoinduced electron transfer reactions of metal complexes in solution
,”
Acc. Chem. Res.
13
(
3
),
83
90
(
1980
).
38.
L. C.
Portis
,
V.
Bhat
, and
C. K.
Mann
, “
Electrochemical dealkylation of aliphatic tertiary and secondary amines
,”
J. Org. Chem.
35
(
7
),
2175
2178
(
1970
).
39.
S. D.
Ross
, “
The mechanism of anodic dealkylation of aliphatic amines in acetonitrile
,”
Tetrahedron Lett.
14
(
15
),
1237
1240
(
1973
).
40.
P.
Bellotti
,
H.-M.
Huang
,
T.
Faber
, and
F.
Glorius
, “
Photocatalytic late-stage C–H functionalization
,”
Chem. Rev.
123
(
8
),
4237
4352
(
2023
).
41.
M.
Goez
and
I.
Sartorius
, “
Photo-CIDNP investigation of the deprotonation of aminium cations
,”
J. Am. Chem. Soc.
115
(
24
),
11123
11133
(
1993
).
42.
B. G.
Stevenson
,
E. H.
Spielvogel
,
E. A.
Loiaconi
,
V. M.
Wambua
,
R. V.
Nakhamiyayev
, and
J. R.
Swierk
, “
Mechanistic investigations of an α-aminoarylation photoredox reaction
,”
J. Am. Chem. Soc.
143
(
23
),
8878
8885
(
2021
).
43.
D.
Gunasekera
,
J. P.
Mahajan
,
Y.
Wanzi
,
S.
Rodrigo
,
W.
Liu
,
T.
Tan
, and
L.
Luo
, “
Controlling one- or two-electron oxidation for selective amine functionalization by alternating current frequency
,”
J. Am. Chem. Soc.
144
(
22
),
9874
9882
(
2022
).
44.
A.
McNally
,
C. K.
Prier
, and
D. W. C.
MacMillan
, “
Discovery of an α-amino C–H arylation reaction using the strategy of accelerated serendipity
,”
Science
334
(
6059
),
1114
1117
(
2011
).
45.
M. H.
Shaw
,
V. W.
Shurtleff
,
J. A.
Terrett
,
J. D.
Cuthbertson
, and
D. W. C.
MacMillan
, “
Native functionality in triple catalytic cross-coupling: sp3 C–H bonds as latent nucleophiles
,”
Science
352
(
6291
),
1304
1308
(
2016
).
46.
C.
Le
,
Y.
Liang
,
R. W.
Evans
,
X.
Li
, and
D. W. C.
MacMillan
, “
Selective sp3 C–H alkylation via polarity-match-based cross-coupling
,”
Nature
547
(
7661
),
79
83
(
2017
).
47.
Recent Trends in Radiation Chemistry
, edited by J. F. Wishart and B. S. M. Rao (
World Scientific
,
2010
).
48.
Y.
Tabata
,
Pulse Radiolysis
(
CRC Press
,
1991
).
49.
M.
Kirch
,
J.-M.
Lehn
, and
J.-P.
Sauvage
, “
Hydrogen generation by visible light irradiation of aqueous solutions of metal complexes. An approach to the photochemical conversion and storage of solar energy
,”
Helv. Chim. Acta
62
(
4
),
1345
1384
(
1979
).
50.
P. J.
DeLaive
,
B. P.
Sullivan
,
T. J.
Meyer
, and
D. G.
Whitten
, “
Applications of light-induced electron-transfer reactions. Coupling of hydrogen generation with photoreduction of ruthenium(II) complexes by triethylamine
,”
J. Am. Chem. Soc.
101
(
14
),
4007
4008
(
1979
).
51.
K.
Kalyanasundaram
, “
Luminescence and redox reactions of the metal-to-ligand charge-transfer excited state of tricarbonylchloro-(polypyridyl)rhenium(I) complexes
,”
J. Chem. Soc., Faraday Trans. 2
82
(
12
),
2401
2415
(
1986
).
52.
C.
Kutal
,
A. J.
Corbin
, and
G.
Ferraudi
, “
Further studies of the photoinduced reduction of carbon dioxide mediated by tricarbonylbromo(2,2′-bipyridine)rhenium(I)
,”
Organometallics
6
(
3
),
553
557
(
1987
).
53.
C.
Kutal
,
M. A.
Weber
,
G.
Ferraudi
, and
D.
Geiger
, “
A mechanistic investigation of the photoinduced reduction of carbon dioxide mediated by tricarbonylbromo(2,2′-bipyridine)rhenium(I)
,”
Organometallics
4
(
12
),
2161
2166
(
1985
).
54.
Y.
Benseghir
,
A.
Solé-Daura
,
D. R.
Cairnie
,
A. L.
Robinson
,
M.
Duguet
,
P.
Mialane
,
P.
Gairola
,
M.
Gomez-Mingot
,
M.
Fontecave
,
D.
Iovan
,
B.
Bonnett
,
A. J.
Morris
,
A.
Dolbecq
, and
C.
Mellot-Draznieks
, “
Unveiling the mechanism of the photocatalytic reduction of CO2 to formate promoted by porphyrinic Zr-based metal–organic frameworks
,”
J. Mater. Chem. A
10
(
35
),
18103
18115
(
2022
).
55.
E.
Fujita
and
B. S.
Brunschwig
, “
Homogeneous redox catalysis in CO2 fixation
,” in
Electron Transfer in Chemistry Vol. 4
, edited by V. Balzani (
Wiley-VCH
,
2001
), pp.
88
126
.
56.
M. D.
Doherty
,
D. C.
Grills
, and
E.
Fujita
, “
Synthesis of fluorinated ReCl(4,4′-R2-2,2′-bipyridine)(CO)3 complexes and their photophysical characterization in CH3CN and supercritical CO2
,”
Inorg. Chem.
48
(
5
),
1796
1798
(
2009
).
57.
J. M.
Smieja
and
C. P.
Kubiak
, “
Re(bipy-tBu)(CO)3Cl−improved catalytic activity for reduction of carbon dioxide: IR-spectroelectrochemical and mechanistic studies
,”
Inorg. Chem.
49
(
20
),
9283
9289
(
2010
).
58.
J. F.
Wishart
,
A. R.
Cook
, and
J. R.
Miller
, “
The LEAF picosecond pulse radiolysis facility at Brookhaven National Laboratory
,”
Rev. Sci. Instrum.
75
(
11
),
4359
4366
(
2004
).
59.
D. C.
Grills
,
J. A.
Farrington
,
B. H.
Layne
,
J. M.
Preses
,
H. J.
Bernstein
, and
J. F.
Wishart
, “
Development of nanosecond time-resolved infrared detection at the LEAF pulse radiolysis facility
,”
Rev. Sci. Instrum.
86
(
4
),
044102
(
2015
).
60.
D. C.
Grills
and
S. V.
Lymar
, “
Solvated electron in acetonitrile: Radiation yield, absorption spectrum, and equilibrium between cavity- and solvent-localized states
,”
J. Phys. Chem. B
126
(
1
),
262
269
(
2022
).
61.
D. C.
Grills
and
S. V.
Lymar
, “
Radiolytic formation of the carbon dioxide radical anion in acetonitrile revealed by transient IR spectroscopy
,”
Phys. Chem. Chem. Phys.
20
(
15
),
10011
10017
(
2018
).
62.
Y.
Hayashi
,
S.
Kita
,
B. S.
Brunschwig
, and
E.
Fujita
, “
Involvement of a binuclear species with the Re−C(O)O−Re moiety in CO2 reduction catalyzed by tricarbonyl rhenium(I) complexes with diimine ligands: Strikingly slow formation of the Re–Re and Re–C(O)O–Re species from Re(dmb)(CO)3S (dmb = 4,4′-Dimethyl-2,2′-bipyridine, S = solvent)
,”
J. Am. Chem. Soc.
125
(
39
),
11976
11987
(
2003
).
63.
B. P.
Sullivan
,
C. M.
Bolinger
,
D.
Conrad
,
W. J.
Vining
, and
T. J.
Meyer
, “
One- and two-electron pathways in the electrocatalytic reduction of CO2 by fac-Re(bpy)(CO)3Cl (bpy = 2,2′-bipyridine)
,”
J. Chem. Soc., Chem. Commun.
(
20
),
1414
1416
(
1985
).
64.
K.
Bobrowski
and
P.
Das
, “
Transient phenomena in the pulse radiolysis of retinyl polyenes. 4. Environmental effects on absorption maximum of retinal radical anion
,”
J. Phys. Chem.
89
(
26
),
5733
5738
(
1985
).
65.
K.
Bobrowski
,
G.
Dzierzkowska
,
J.
Grodkowski
,
Z.
Stuglik
,
Z.
Zagórski
, and
W.
McLaughlin
, “
A pulse radiolysis study of the leucocyanide of malachite green dye in organic solvents
,”
J. Phys. Chem.
89
(
20
),
4358
4366
(
1985
).
66.
S.
Nad
and
H.
Pal
, “
Photoinduced electron transfer from aliphatic amines to coumarin dyes
,”
J. Chem. Phys.
116
(
4
),
1658
1670
(
2002
).
67.
J. R.
De la Fuente
,
G.
Kciuk
,
E.
Sobarzo-Sanchez
, and
K.
Bobrowski
, “
Transient phenomena in the pulse radiolysis of oxoisoaporphine derivatives in acetonitrile
,”
J. Phys. Chem. A
112
(
41
),
10168
10177
(
2008
).
68.
D. A.
Armstrong
,
R. E.
Huie
,
W. H.
Koppenol
,
S. V.
Lymar
,
G.
Merényi
,
P.
Neta
,
B.
Ruscic
,
D. M.
Stanbury
,
S.
Steenken
, and
P.
Wardman
, “
Standard electrode potentials involving radicals in aqueous solution: Inorganic radicals (IUPAC Technical Report)
,”
Pure Appl. Chem.
87
(
11-12
),
1139
1150
(
2015
).
69.
M.
Goto
,
H.
Park
,
K.
Otsuka
, and
M.
Oyama
, “
Kinetics of the decay reactions of the N,N-Dimethyl-p-Toluidine cation radical in acetonitrile. Acid−Base interaction to promote the CH2–CH2 bonding
,”
J. Phys. Chem. A
106
(
35
),
8103
8108
(
2002
).
70.
W. A.
Mulac
,
A.
Bromberg
, and
D.
Meisel
, “
Protonation of the reduced species in irradiated acetonitrile
,”
Radiat. Phys. Chem.
26
(
2
),
205
209
(
1985
).
71.
L.
Yang
,
A.-M.
Ren
,
J.-K.
Feng
,
X.-J.
Liu
,
Y.-G.
Ma
,
M.
Zhang
,
X.-D.
Liu
,
J.-C.
Shen
, and
H.-X.
Zhang
, “
Theoretical studies of ground and excited electronic states in a series of halide rhenium(I) bipyridine complexes
,”
J. Phys. Chem. A
108
(
32
),
6797
6808
(
2004
).
72.
J. W.
Arbogast
,
C. S.
Foote
, and
M.
Kao
, “
Electron transfer to triplet fullerene C60
,”
J. Am. Chem. Soc.
114
(
6
),
2277
2279
(
1992
).
73.
N. V.
Rees
,
O. V.
Klymenko
,
R. G.
Compton
, and
M.
Oyama
, “
The electro-oxidation of N,N-dimethyl-p-toluidine in acetonitrile: A microdisk voltammetry study
,”
J. Electroanal. Chem.
531
(
1
),
33
42
(
2002
).
74.
The reported potential versus Ag+/0 was converted to SCE by adding 0.3 V, according to
V. V.
Pavlishchuk
and
A. W.
Addison
, “
Conversion constants for redox potentials measured versus different reference electrodes in acetonitrile solutions at 25 °C
,”
Inorg. Chim. Acta
298
,
97
102
(
2000
).
75.
A.
Rodenberg
,
M.
Orazietti
,
B.
Probst
,
C.
Bachmann
,
R.
Alberto
,
K. K.
Baldridge
, and
P.
Hamm
, “
Mechanism of photocatalytic hydrogen generation by a polypyridyl-based cobalt catalyst in aqueous solution
,”
Inorg. Chem.
54
(
2
),
646
657
(
2015
).
76.
D. D. M.
Wayner
,
D. J.
McPhee
, and
D.
Griller
, “
Oxidation and reduction potentials of transient free radicals
,”
J. Am. Chem. Soc.
110
(
1
),
132
137
(
1988
).
77.
S.
Nelsen
and
P.
Hintz
, “
Electrochemical oxidation of tertiary amines. Effect of structure upon reversibility
,”
J. Am. Chem. Soc.
94
(
20
),
7114
7117
(
1972
).
78.
E.
Hasegawa
,
Y.
Tamura
,
K.
Suzuki
,
A.
Yoneoka
, and
T.
Suzuki
, “
Photoreactions of 4-(tribromomethyl)-4-methyl-2,5-cyclohexadienone and its derivatives with amines: Radical cyclization and ring expansion reactions promoted through photoinduced electron transfer processes
,”
J. Org. Chem.
64
(
24
),
8780
8785
(
1999
).
79.
T. M.
McKinney
and
D. H.
Geske
, “
The triethylenediamine cation radical
,”
J. Am. Chem. Soc.
87
(
13
),
3013
3014
(
1965
).
80.
Y.
Pellegrin
and
F.
Odobel
, “
Sacrificial electron donor reagents for solar fuel production
,”
C. R. Chimie
20
(
3
),
283
295
(
2017
).
81.
T. J.
Burkey
,
A. L.
Castelhano
,
D.
Griller
, and
F. P.
Lossing
, “
Heats of formation and ionization potentials of some α-aminoalkyl radicals
,”
J. Am. Chem. Soc.
105
(
14
),
4701
4703
(
1983
).
82.
H.
Sun
and
M. Z.
Hoffman
, “
Reductive quenching of the excited states of ruthenium(II) complexes containing 2,2′-bipyridine, 2,2′-bipyrazine, and 2,2′-bipyrimidine ligands
,”
J. Phys. Chem.
98
(
45
),
11719
11726
(
1994
).
83.
N.
Sutin
, “
Theory of electron transfer reactions: Insights and hindsights
,”
Prog. Inorg. Chem.
30
,
441
498
(
2009
).
84.
T.
Morimoto
and
O.
Ishitani
, “
Modulation of the photophysical, photochemical, and electrochemical properties of Re(I) diimine complexes by interligand interactions
,”
Acc. Chem. Res.
50
(
11
),
2673
2683
(
2017
).
85.
J.
Coetzee
and
G.
Padmanabhan
, “
Properties of bases in acetonitrile as solvent. IV. Proton acceptor power and homoconjugation of mono- and diamines
,”
J. Am. Chem. Soc.
87
(
22
),
5005
5010
(
1965
).

Supplementary Material