Proton-Coupled Energy Transfer in Molecular Triads

Abstract

A photochemical mechanism was experimentally discovered and denoted proton-coupled energy transfer (PCEnT). A series of anthracene-phenol-pyridine triads formed the local excited anthracene state after light excitation at ca. 400 nm, which led to fluorescence around 550 nm from the phenol-pyridine unit. Direct excitation of phenol-pyridine would have required light around 330 nm, but the coupled proton transfer within the phenol-pyridine unit lowered its excited state energy so that it could accept excitation energy from anthracene. Singlet-singlet energy transfer thus occurred despite the lack of spectral overlap between the anthracene fluorescence and the phenol-pyridine absorption. Moreover, theoretical calculations indicated negligible charge transfer between the anthracene and phenol-pyridine units. PCEnT was suggested as an elementary reaction of possible relevance to biological systems and future photonic devices.

One sentence summary:

An elementary reaction was discovered, with potential importance for light-activated chemistry, photobiology and photonic devices.

Introduction

Coupling of proton transfer to electronic transitions exerts a great influence on the energetics and kinetics of these processes. Photoacids are well-studied examples, where light excitation leads to a rapid proton release, and in some cases, even produces long-lived pH changes and pH-gradients. (1, 2) Another example is excited state intramolecular proton transfer (ESIPT) that generates an excited tautomer state of the light-absorbing molecule; systems where this state is fluorescent have applications for optoelectronic materials. (3) Often, ESIPT instead allows for ultrafast dissipation of excess electronic energy as heat, and is therefore an important photo-stabilizing mechanism in plastics (4) as well as in biomolecules such as DNA (5).

More recently, the charge transfer mechanism of proton-coupled electron transfer (PCET) has generated great interest. (6) PCET is a key reaction behind biological energy conversion in photosynthesis, respiration and nitrogen fixation (7), as well as in synthetic chemical systems for photoredox catalysis and renewable energy conversion and storage. (8–11) It is also involved in DNA synthesis and repair, photoreceptor function and many other radical enzyme reactions. (12–15) Of particular interest is the concerted PCET mechanism, where the electron and proton are transferred without the formation of high energy intermediates, thereby offering a low-barrier reaction pathway.

In all the above reactions, the coupled transfer of the proton changes the energy landscape of the overall process, which obviously affects its dynamics and rate. There are also important effects of the much greater mass of the proton, whose wavefunction is more localized than that of the electron(s) involved, which may impose a greater sense of directionality on the process. Understanding and analysing such effects, theoretically and experimentally, is a fundamentally interesting challenge and of practical importance.

In 2019, our groups used a series of anthracene-phenol-pyridine (An-PhOH-Py) triads (Fig. 1), (16) to demonstrate the first example of concerted PCET occurring in the Marcus inverted region. (17) This behaviour, where the rate decreases with increasing driving force, is important to slow down recombination after photochemical charge separation, but was previously considered unlikely for concerted PCET reactions because of the nearly barrierless tunnelling to higher vibrational product states. (18) Nevertheless, photoexcitation of the An subunit of the triads results in the formation of a local excited state (LES) on the anthracene (Eq. 1). (16, 19) In tenths of picoseconds, the LES forms the charge separated state (CSS) by electron transfer from the PhOH group to the ^1*An concerted with proton transfer to the pyridine (Eq. 2). The charges subsequently recombine to the ground state reactants (Eq. 3), once again in a concerted electron-proton transfer reaction but with a large driving force (>2.0 eV in CH₃CN). This process consequently places the reaction in the inverted region, as was shown by the rate dependence on driving force and solvent polarity.

Fig. 1.

Structures of the studied anthracene-phenol-pyridine ([An-PhOH-py]) triads. The enol form is the stable ground state form, and the electronically excited keto form is the local electron-proton transfer, LEPT, state.

$$\text{An}-\text{PhOH}-\text{py}\stackrel{h\nu }{\to }{}^{1*}\text{An}-\text{PhOH}-\text{py}\text{(LES)}$$ (1)

$${}^{1*}\text{An}-\text{PhOH}-\text{py}\to {\text{An}}^{•-}-{\text{PhO}}^{•}-{\text{pyH}}^{+}(\text{CSS})$$ (2)

$${\text{An}}^{•-}-{\text{PhO}}^{•}-{\text{pyH}}^{+}\to \text{An}-\text{PhOH}-\text{py}$$ (3)

The CSS intermediate was only observed in triads 1–3, however, and in triads 4–8, a similar quenching of the LES state did not result in an observable CSS. (16) This result is puzzling, as the charge recombination driving force is larger for 4–8, meaning the reaction would be more deeply in the inverted region and thus even slower than for 1–3. Theoretical investigations have suggested the involvement of another state that would lie lower than the CSS in 4–8. (20, 21) This is the so-called local electron-proton transfer state (LEPT), where the PhOH-py fragment is in a local, electronically excited tautomeric (keto) state *[PhO-pyH] in which the phenolic proton has transferred to the pyridine. The LEPT state is the same excited tautomeric state as that formed by ESIPT after excitation of the separate phenol-pyridine compound, (22) and similar to other ESIPT compounds. (3) This state was proposed to be populated after excitation in 4–8 (20, 21). We herein note that it can possibly be populated from the CSS (Eq. 4), or be formed directly from the LES (Eq. 5) by a process here denoted proton-coupled energy transfer, PCEnT (vide infra).

$${\text{An}}^{•-}-{\text{PhO}}^{•}-{\text{pyH}}^{+}\to \text{An}{-}^{*}[\text{PhO}-\text{pyH}](\text{LEPT})$$ (4)

$${}^{1*}\text{An}-\text{PhOH}-\text{py}\stackrel{\text{PCEnT}}{\to }\text{An}{-}^{*}[\text{PhO}-\text{pyH}](\text{LEPT})$$ (5)

Here we present experimental and theoretical evidence for direct formation of the fluorescent LEPT state from the LES state in a 77 K butyronitrile glass, in which formation of the CSS is inhibited. This is the first experimental example of a process we here denote PCEnT, in which excitation energy was transferred from ^1*An to ^1*[PhO-pyH], coupled to proton transfer within the PhOH-py unit (Eq. 5). This finding was surprising, as it is a singlet-singlet transition without any detectable spectral overlap between the donor fluorescence and acceptor absorption. Moreover, PCEnT occurred with negligible charge transfer between the An and PhOH-py units, which is in sharp contrast to PCET reactions. This result significantly expands the scope of concerted processes where proton transfer controls reactivity. Many excited state charge and energy transfer reactions in proteins and DNA occur between hydrogen-bonded pigments, making PCEnT possibly important in natural systems. Yet, such a reaction has, to the best of our knowledge, not previously been reported.

Results and Discussion

The lack of observed CSS formation in triads 4–8, together with the computationally predicted excited state keto tautomer (LEPT) (20), laid the basis for further investigation. The triad absorption spectra agree well with a sum of contributions from the separate components: a vibrationally structured anthracene band around 400 nm and a PhOH-py band around 330 nm (Fig. 2B and Fig. S1). The LEPT state should fluoresce around 500 nm in aprotic, rigid media, as has been shown for the excited keto-tautomer, *[PhO-pyH] of the separate phenol-pyridine compound. (22) Therefore, we examined the triads (1, 2, 4–7) by fluorescence spectroscopy in rigid butyronitrile glass at T = 77 K. We also studied these triads computationally and found that the minimum-energy LEPT geometry exhibits an approximately 90° twist between the phenol and pyridine fragments, whereas the planar LEPT geometry is a first-order saddle point. This twisting, which is known to quench the fluorescence (22, 23), is expected to be hindered in the glass at 77 K. Thus, the computational studies were performed for both the planar and twisted LEPT geometries.

Fig. 2.

Steady state absorption (room temperature) and fluorescence (77 K) spectra in butyronitrile. A) schematic Jablonski diagram of the possible transitions at 77 K for the triads: radiative decay (downward solid line) and nonradiative decay. The CSS energy depends strongly on the chemical environment (16, 19) and is higher than the LES at 77 K. B) Absorption (solid line) and fluorescence spectra normalized to the first An emission peak (dashed line, λ_exc: 400 nm) of 1 (black) and the reference compounds of similar concentrations (4–7 μM): 2,4-di-tertbutyl-6-(pyridin-2-yl)phenol (PhOH-py, blue) and 9-cyano-10-methylanthracene (CN-MeAn, green). C) Fluorescence spectra of triads 1, 2, 4–7 with excitation at 400 nm (1–2), 365 nm (6) and 375 nm (5 and 7). Note that 7 was measured with twice the absorption at the excitation wavelength. D) Absorption spectrum (black line), fluorescence excitation spectrum (λ_em=455 and 550 nm, solid line) and fluorescence spectrum (λ_exc=400 nm, dashed line) of 2.

The 77 K fluorescence spectrum of 1, after selective excitation of the An unit at 400 nm, showed a structured ^1*An fluorescence at 420–580 nm, accompanied by a broad band with a maximum around 550 nm (Fig. 2B). The latter is expected for the LEPT state (22), and was similarly observed for all measured triads with an electron donating group on the pyridine: 1–2 and 5–6 (Fig. 2B and C). The LEPT emission shifted to lower energy with less electron-donating groups (2 > 1 > 6), consistent with the computational results in Tables S8 and S12. As 1 and 5 share the same functional group (R₂=4-Me) on the pyridine, the LEPT energy should be similar. Still, the peak for 5 was blueshifted by 11 nm, which indicated a small but significant coupling to the An unit, which differs in its functional group (1: CN vs. 5: H) between these triads. The relative intensity of the ^1*An and LEPT fluorescence was independent of triad concentration (Fig. S2), and control experiments with 9-CN-10-Me-An only showed the structured ^1*An fluorescence (Fig. 2B). This excluded any excimer or other anthracene complex as responsible for the broad fluorescence around 550 nm.

Fluorescence excitation spectroscopy provided unambiguous evidence for LEPT state formation from excited ^1*An (Fig. 2D). The shape of the excitation spectrum of the 550 nm LEPT emission peak (purple line) matched that of the absorption spectrum (black line), with contributions from both the PhOH-py (330 nm) and the An (300–430 nm) absorption bands. The excitation spectrum of the 450 nm ^1*An emission (green line) showed instead a weaker relative contribution from the 330 nm PhOH-py band. The 550 nm LEPT state could thus be formed either by exciting the PhOH-py unit, which would cause formation of the LEPT state by the well-established ESIPT mechanism, or directly via excitation of the An unit followed by energy transfer from ^1*An.

The excitation energy transfer from the anthracene donor to the phenol-pyridine unit to yield the excited keto-tautomeric state (LEPT) is a type of mechanism that has never been described before, and we denote it proton-coupled energy transfer, PCEnT (Fig. 2A). Energy transfer directly to the *[PhOH-py] state would be strongly uphill, as is seen from its low-wavelength absorption spectrum around 330 nm, but proton transfer lowers the phenol-pyridine excited state energy to make the processes energetically feasible. The quantum yield for the LEPT formation was deemed to be similar irrespective of which fragment was excited, based on the good agreement between the shapes of the absorption spectrum (Fig. 2D) and the 550 nm excitation spectrum, where the respective PhOH-py and An amplitudes was fairly equal in the two spectra. This fact shows that PCEnT from ^1*An was about as efficient as ESIPT from *[PhOH-py].

To follow the formation and decay of the LEPT state in real time, ultrafast transient fluorescence spectroscopy measurements of 6 were performed with excitation at 370 nm, in the mid-to-red part of the An absorption band (Fig. 3). The observed features matched those seen in the steady state experiments, and global fits were performed within the peak regions of the LES (380–490 nm) and the LEPT (520–655 nm) fluorescence (Fig. 3B). The rise time of the LEPT peak was fitted to ~44 ps and a similarly short decay component could be found for the LES (46 ps), which reinforced the notion that the LEPT state could be populated directly from the LES as depicted in the overall reaction scheme in Fig. 2A.

Fig. 3.

Contour map (A) of the transient fluorescence emission intensity counts (dark blue=highest intensity) of 6 in 77 K butyronitrile (0.2 mM) between 390–640 nm and a 2 ns time window. Spectra at given times after excitation with 370 nm light (B). The ^1*An fluorescence reaches a maximum at ca. 30 ps (IRF_FWHM = 65 ps) and decreases simultaneously with the the rise of LEPT fluorescence that maximizes after ca. 140 ps. C) Time-resolved fluorescence traces (fit:lines, data:circles) within the LES (445 nm, blue line) and LEPT (570 nm, green line) bands, and the instrumental response function (IRF, dashed line).

We can exclude LEPT formation via a sequential reaction LES-CSS-LEPT at 77 K, because the CSS would be substantially destabilized in a solvent glass (Fig. 2A), due to its greater dipole moment compared to both the LES and LEPT (Fig. 4B). The solvent glass at 77 K prevented the large degree of twisting of the An group to reach the CSS optimized structure (Fig. 4B and Table S1), which further destabilized the CSS energy. The CSS lies less than 0.2 eV below the LES at room temperature, (16) and small-molecule charge separation is typically destabilized by more than 0.5 eV by freezing out solvent repolarization in polar solvents. (25) Our gas phase calculations in Table S6 indeed placed the CSS higher than the LES. Moreover, the LEPT fluorescence was formed faster in 6 (~44 ps) than in 1 (~69 ps), in spite of ca. 0.5 eV greater room-temperature driving force for CSS formation in the latter, (16) which is clear experimental proof against a reaction involving the CSS.

Fig. 4.

Potential energy surfaces and optimized structures for the electronic states discussed. A) The schematic potential energy surfaces for the ESIPT and PCEnT reactions, color coded marking the electronically excited fragment (An:green or PhOH-py:purple). Note that ESIPT is barrierless on the excited state potential energy surface (purple), whereas PCEnT follows the reorganization of the heavy nuclei including the solvent. In fluid solution the LEPT state will twist along the PhOH-py bond, where 90° leads to a conical intersection, and nonradiative decay. This twist is impeded in the 77 K butyronitrile glass, although a partial twist could occur to lower the energy of the LEPT state relative to the LES (Table S11). B) The optimized structures and the natural bond orbital (NBO (29)) charges on the anthracene (blue), phenol (red), and pyridine (purple) subunits of 1 for the GS, LES, CSS and planar (not twisted, first-order saddle point) LEPT state in the gas phase. The charges reflect the electron density, where a more negative number equals a more electron dense fragment (the methylene spacer between anthracene and phenol is not included in any of the fragments, and therefore those charges are excluded in this analysis).

The photophysics of triads 1–8 have been previously studied by transient absorption (TA) at ambient temperature, (16) and 6 was also examined by time-resolved fluorescence spectroscopy at various temperatures. (24) In the TA study at room temperature, the initial anthracene local excited states (LESs) in 1–3 decayed to a long-lived state (up to τ= 755 ps) with spectral features in agreement with the charge separated state (CSS) being formed by PCET, including obvious features of the anthracene anion. The same mechanism was assigned to the decay of the LES of 4–8 based on the close correlation of LES decay rates with the free energies of PCET for each triad. These triads did not show a long lived transient in the TA spectra at ambient temperatures. However, computations later placed the LEPT state below the CSS for 4 and 6, (20) indicating that rapid CSS-to-LEPT conversion and/or a contribution of PCEnT from the LES would be reasonable.

The previously reported spectrally-resolved ultrafast fluorescence spectra of 6 at ambient temperatures showed an H/D kinetic isotope effect (KIE) = 4, and did not show any evidence of emission from an LEPT state. At lower temperatures, in a 2-methyl-tetrahydrofuran (2-Me-THF) glass or a polymethylmethacrylate (PMMA) film, the sub-ns integrated fluorescence followed a single exponential with τ ≅ 150 ps, with an unexplained long-timescale (» ns) component. The data in 2-Me-THF were obtained in the presence of 3% methanol, which was observed in the current study to quench the LEPT emission. Therefore the two studies are not directly comparable. In retrospect, these data could be consistent with the LES either decaying by PCET or by the new PCEnT pathway reported here in butyronitrile glasses; additional studies are needed.

The observation of PCEnT to directly form the excited tautomeric LEPT state was surprising, and needs mechanistic clarification. As both the ^1*An and the LEPT state are fluorescent, PCEnT is clearly a singlet-singlet process. However, there was no detectable spectral overlap between the donor fluorescence and acceptor ground-state absorption (Fig. 2B), which is a requirement for Förster (dipole-dipole) energy transfer. The absence of low-energy absorption of the keto-form [PhO⁻-pyH⁺] is because the ground state, at the LEPT geometry, lies ~0.5 eV higher than the enolic ground state (Table S4) and was thus not populated to any significant extent. Instead, as the reaction occurred at quite a small donor-acceptor distance (~6 Å between the central rings of An and PhOH), the coupling was presumably dominated by exchange and/or penetration terms. (26) Any stronger coupling mechanism was probably prevented by the methylene spacer between the subunits. We propose therefore that PCEnT could be described by a nonadiabatic surface crossing (Fig. 4A), in analogy to Marcus-type theoretical descriptions of PCET. (6, 7) Thermal fluctuations of heavy nuclei lead to a configuration where electronic energy transfer and proton tunneling occur with energy conservation (crossing point of the reactant and product free energy curves). It is different from PCET, however, in that there is no charge transfer between donor and acceptor. The product state is a fluorescent excited state, when phenol-pyridine twisting is hindered, and can thus readily be detected. Moreover, this mechanism differs from an ESIPT reaction, where proton transfer is driven by intramolecular charge redistribution, and the electronic coupling is usually strong, which leads to ultra-fast reactions. The PCEnT reaction described here was instead driven by a remote, weakly coupled electronically excited state.

The rearrangement from enol to keto form of the phenol-pyridine acceptor unit is much greater than just the transfer of a proton, as it involves significant charge redistribution and changes in bond lengths and angles. These combined effects change the excitation energy of the phenol-pyridine acceptor unit and allow it to come into resonance with that of the anthracene. This behavior is the fundamental basis for coupling of proton transfer to energy transfer and distinguishes PCEnT from typical energy transfer reactions. The details of this coupling and the role of coupling with different proton vibrational states remain to be further explored.

Fig. 5 demonstrates the PCEnT process observed in a molecular dynamics trajectory for triad 6 propagated on the S₁ state, which changed from LES to LEPT character as the proton transferred from the phenol to the pyridine. Representative configurations along this trajectory, along with an analysis of the fragment charges and dominant natural transition orbitals, show that the character of the S₁ state changed from LES to a mixture of LES and LEPT and then to LEPT. The nonadiabatic coupling between the S₁ and S₂ adiabatic electronic states (27, 28) exhibited a peak at the avoided crossing, providing evidence that this process is nonadiabatic. Throughout this process, the charge on the anthracene fragment remained virtually zero, indicating that the LES to LEPT transition did not involve charge transfer between anthracene and the phenol-pyridine group, consistent with the PCEnT mechanism. Moreover, the net charge transfer from the phenol to the pyridine was also nearly zero after accounting for the charge of the transferring proton. In the framework of the vibronically nonadiabatic PCET theory (6), the transferring proton would be treated quantum mechanically and would tunnel during the LES to LEPT transition.

Fig. 5.

Avoided crossing region between the S₁ and S₂ adiabatic electronic states for triad 6, corresponding to a crossing between the LES and LEPT diabatic electronic states, and analysis of the S₁ excited state for relevant configurations. The energy curves were obtained from an excited state adiabatic molecular dynamics trajectory and are plotted along a unitless general reaction coordinate rather than time; this trajectory is far from equilibrium and is not directly comparable to experiment due to the reaction conditions (section S6.4). Configurations A, B, and C were obtained from this trajectory. The dominant natural transition orbitals (NTOs), corresponding to the transitions from occupied to virtual orbitals that characterize the S₁ excited state, are shown for these three configurations. For A, the NTOs are localized on the anthracene (LES), whereas for C, the NTOs are localized on the phenol-pyridine fragment (LEPT state), where the proton has mostly transferred. For B, an approximately equal mixture of LES and LEPT is observed, as indicated by the two NTOs. Above the orbital drawings of each configuration, the numbers are the charges (in a.u., the elementary charge of an electron) of the anthracene, phenol, and pyridine fragments for the S₁ state, in that order (excluding the charge of the CH₂ bridging group and the transferring hydrogen, which has a charge of ~0.5 a.u.).

In summary, we have experimentally proven the involvement of the excited *[PhOH-Py] state (LEPT) in the photochemical reactions of these triad molecules. This mechanism competes with PCET from the LES and could explain why the CSS was not observed in 4–8. The LEPT state was formed by a novel proton-coupled energy transfer (PCEnT) mechanism. This is a singlet-singlet exchange energy transfer, coupled to proton transfer in the acceptor unit, that can be described as a nonadiabatic transition to the excited product state. Its free energy barrier in turn is determined by the reorganization energy associated with the solvent and other heavy nuclei, similar to Marcus ET theory or PCET theories. (6, 17) In contrast to PCET, there is negligible charge transfer between donor and acceptor.

This mechanism can have several intriguing implications for both natural and synthetic systems. PCEnT may allow for the photosensitization of otherwise dark states, such as seen here by the LEPT formation from the low energy excitation of the LES, rather than the direct excitation of the PhOH-py unit at higher energy (ca 330 nm). Moreover, as energy transfer occurs over long distances but proton transfer is short-range, we suggest that this mechanism may be of importance to direct energy migration in e.g. light-harvesting systems and photonic materials, where the PCEnT acceptor would form a low-energy site. PCEnT could also form the basis of a molecular switch for energy transfer, where proton transfer could be controlled independently by hydrogen-bonding or (de)protonation, thus effectively turning energy transfer on or off. We speculate that these processes could already be operational in natural systems with hydrogen-bonded light-absorbers, such as porphyrins in photosynthetic assemblies, DNA and flavin proteins, but have yet to be identified.

Data and materials availability:

Excel file with the absorption, emission, emission excitation spectra, transient absorption and streak camera data, as well as the optimized geometry for triads 1 and 6 and their NBO charges, is available as supplementary material (data S1). Experimental methods and procedures, additional data and analysis are included in the Supplementary Materials.