Abstract Cubane‐type metal clusters respond uniquely to stimuli like light and electric potential, resulting in behaviors such as crystal‐to‐crystal phase transitions. While structural adaptability is known to be linked to these responses, direct experimental evidence for the associated structural changes has been missing. This study addresses this gap by examining the structural dynamics of the copper(I) iodide cubane (Cu4I4(py)4, py = pyridine) upon photoexcitation using time‐resolved X‐ray liquidography. The results reveal: 1) 100 picoseconds (ps) after excitation, two distinct excited states—the cluster‐centered triplet (3CC) state and the (metal+halide)‐to‐ligand charge transfer triplet (3(M/X)LCT) state—are present; 2) the 3(M/X)LCT state decays with an apparent time constant of 1.21 ns, primarily transitioning to the 3CC state, with a small fraction undergoing decay to the ground state (GS); and 3) the 3CC state eventually returns to the GS. The molecular structures, provided for these states serve as benchmarks for theoretical studies. Importantly, the 3CC structure exhibits significant distortion in the Cu4I4 core and reduced symmetry, findings that are unanticipated by previous models. This comprehensive investigation deepens the understanding of the structural transformations occurring upon photoexcitation, with a potential impact on future applications of these compounds as versatile components in photosensitive metal–organic frameworks.
Metal-semiconductor interfaces are crucial components of optoelectronic and electrical devices, the performance of which hinges on intricate dynamics involving charge transport and mechanical interaction at the interface. Nevertheless, structural changes upon photoexcitation and subsequent carrier transportation at the interface, which crucially impact hot carrier stability and lifetime, remain elusive. To address this long-standing problem, they investigated the electron dynamics and resulting structural changes at the Au/TiO2 interface using ultrafast electron diffraction (UED). The analysis of the UED data reveals that interlayer electron transfer from metal to semiconductor generates a strong coupling between the two layers, offering a new way for ultrafast heat transfer through the interface and leading to a coherent structural vibration that plays a critical role in propagating mechanical stress. These findings provide insights into the relationship between electron transfer and interfacial mechanical and thermal properties.
Molecular ions are ubiquitous and play pivotal roles 1–3 in many reactions, particularly in the context of atmospheric and interstellar chemistry 4–6 . However, their structures and conformational transitions 7,8 , particularly in the gas phase, are less explored than those of neutral molecules owing to experimental difficulties. A case in point is the halonium ions 9–11 , whose highly reactive nature and ring strain make them short-lived intermediates that are readily attacked even by weak nucleophiles and thus challenging to isolate or capture before they undergo further reaction. Here we show that mega-electronvolt ultrafast electron diffraction (MeV-UED) 12–14 , used in conjunction with resonance-enhanced multiphoton ionization, can monitor the formation of 1,3-dibromopropane (DBP) cations and their subsequent structural dynamics forming a halonium ion. We find that the DBP + cation remains for a substantial duration of 3.6 ps in aptly named ‘dark states’ that are structurally indistinguishable from the DBP electronic ground state. The structural data, supported by surface-hopping simulations 15 and ab initio calculations 16 , reveal that the cation subsequently decays to iso -DBP + , an unusual intermediate with a four-membered ring containing a loosely bound 17,18 bromine atom, and eventually loses the bromine atom and forms a bromonium ion with a three-membered-ring structure 19 . We anticipate that the approach used here can also be applied to examine the structural dynamics of other molecular ions and thereby deepen our understanding of ion chemistry.
Understanding protein structure and kinetics under physiological conditions is crucial for elucidating complex biological processes. While time-resolved (TR) techniques have advanced to track molecular actions, their practical application in biological reactions is often confined to reversible photoreactions within limited experimental parameters due to inefficient sample utilization and inflexibility of experimental setups. Here, we introduce serial X-ray liquidography (SXL), a technique that combines time-resolved X-ray liquidography with a fixed target of serially arranged microchambers. SXL breaks through the previously mentioned barriers, enabling microgram-scale TR studies of both irreversible and reversible reactions of even a non-photoactive protein. We demonstrate its versatility in studying a wide range of biological reactions, highlighting its potential as a flexible and multi-dimensional assay framework for kinetic and structural characterization. Leveraging X-ray free-electron lasers and micro-focused X-ray pulses promises further enhancements in both temporal resolution and minimizing sample quantity. SXL offers unprecedented insights into the structural and kinetic landscapes of molecular actions, paving the way for a deeper understanding of complex biological processes.
Energy, structure, and charge are fundamental quantities characterizing a molecule. Whereas the energy flow and structure change in chemical reactions are experimentally characterized, determining the atomic charges of a molecule in solution has been elusive, even for a triatomic molecule such as triiodide ion, I 3 − . Moreover, it remains to be answered how the charge distribution is coupled to the molecular geometry; which I-I bond, if two I-I bonds are unequal, dissociates depending on the electronic state. Here, femtosecond anisotropic x-ray solution scattering allows us to provide the following answers in addition to the overall rich structural dynamics. The analysis unravels that the negative charge of I 3 − is highly localized on the terminal iodine atom forming the longer bond with the central iodine atom, and the shorter I-I bond dissociates in the excited state, whereas the longer one in the ground state. We anticipate that this work may open a new avenue for studying the atomic charge distribution of molecules in solution and taking advantage of orientational information in anisotropic scattering data for solution-phase structural dynamics.
Optical Kerr effect (OKE) spectroscopy is a method that measures the time-dependent change of the birefringence induced by an optical laser pulse using another optical laser pulse and has been used often to study the ultrafast dynamics of molecular liquids. Here we demonstrate an alternative method, femtosecond time-resolved X-ray liquidography (fs-TRXL), where the microscopic structural motions related to the OKE response can be monitored using a different type of probe, i.e., X-ray solution scattering. By applying fs-TRXL to acetonitrile and a dye solution in acetonitrile, we demonstrate that different types of molecular motions around photoaligned molecules can be resolved selectively, even without any theoretical modeling, based on the anisotropy of two-dimensional scattering patterns and extra structural information contained in the q-space scattering data. Specifically, the dynamics of reorientational (libration and orientational diffusion) and translational (interaction-induced motion) motions are captured separately by anisotropic and isotropic scattering signals, respectively. Furthermore, the two different types of reorientational motions are distinguished from each other by their own characteristic scattering patterns and time scales. The measured time-resolved scattering signals are in excellent agreement with the simulated scattering signals based on a molecular dynamics simulation for plausible molecular configurations, providing the detailed structural description of the OKE response in liquid acetonitrile.
Methanol was converted to white-light fluorescent polyaromatics by the catalytic effect of Ca2+ in ion-exchanged zeolite nanocages.
The experimental and calculation results demonstrate that the 3MLppzCT state generated by the spin-forbidden transition rapidly relaxes to 3MLppyCT through internal conversion process with a time constant of ∼450 fs.
Charge transfer (CT) from electron donor (D) to acceptor (A) plays an important role in photoelectric or electrochemical devices and is a useful concept for a molecule with D and A well distinguishable. Here, we report our finding that even in a molecule with D and A not resolvable, CT can be induced by electronic state mixing (ESM) in a symmetric multi-chromophore system (MCS), namely 1,4-di(1-pyrenyl)benzene (Py-Benz-Py). Unlike Py and Py-Benz, Py-Benz-Py exhibits unique photophysical properties attributable to the reduction of the energy gap between two electronic states induced by ESM. The ESM for Py-Benz-Py is due to the extended π-conjugation owing to the further introduction of Py into Py-Benz, and consequently leads to the favorable intramolecular CT, followed by the planarization due to the twisting motion between Py and phenyl moieties. Time-resolved spectroscopic data demonstrate that the twisting process of the Py moiety in acetonitrile occurs with two unequal time constants, suggesting the localized CT state and the asynchronous twisting dynamics of two Py moieties unlike the delocalized CT state in nonpolar and low-polarity solvents leading to the synchronous twisting of two Py moieties. This means that the symmetry-breaking CT in MCSs can induce an asynchronous twisting motion. The results reported here support that a molecule without CT can be turned into another molecule with CT induced by ESM and demonstrate that the excited-state relaxation dynamics can be regulated through the ESM induced by introducing the substituents or changing the environmental factors such as solvent polarities.
Knowledge about factors that govern chemoselectivity is pivotal to the design of reactions that are utilized to produce complex organic substances. In the current study, single-electron transfer (SET)-promoted photoaddition reactions of fullerene C60 with both trimethylsilyl and various alkyl group-containing glycinates and ethyl N-alkyl-N-((trimethylsilyl)methyl)glycinates were explored to evaluate how the nature of N-alkyl substituents of glycinate substrates and reaction conditions govern the chemoselectivity of reaction pathways followed. The results showed that photoreactions of C60 with glycinates, performed in deoxygenated conditions, produced aminomethyl-1,2-dihydrofullerenes efficiently through a pathway involving the addition of α-amino radical intermediates that are generated by sequential SET-solvent-assisted desilylation of glycinate substrates to C60. Under oxygenated conditions, photoreactions of glycinate substrates, except N-benzyl-substituted analogues, did not take place efficiently owing to quenching of 3C60* by oxygen. Interestingly, N-benzyl-substituted glycinates did react under these conditions to form fulleropyrrolidines through a pathway involving 1,3-dipolar cycloaddition of in situ formed azomethine ylides to C60. The ylide intermediates were formed by regioselective H-atom transfer from glycinates by singlet oxygen. Furthermore, methylene blue (MB)-photosensitized reactions of C60 with glycinates under oxygenated conditions took place efficiently to produce fulleropyrrolidines independent of the nature of N-alkyl substituents of glycinates.
Fundamental studies of chemical reactions often consider the molecular dynamics along a reaction coordinate using a calculated or suggested potential energy surface1–5. But fully mapping such dynamics experimentally, by following all nuclear motions in a time-resolved manner—that is, the motions of wavepackets—is challenging and has not yet been realized even for the simple stereotypical bimolecular reaction6–8: A–B + C → A + B–C. Here we track the trajectories of these vibrational wavepackets during photoinduced bond formation of the gold trimer complex [Au(CN)2−]3 in an aqueous monomer solution, using femtosecond X-ray liquidography9–12 with X-ray free-electron lasers13,14. In the complex, which forms when three monomers A, B and C cluster together through non-covalent interactions15,16, the distance between A and B is shorter than that between B and C. Tracking the wavepacket in three-dimensional nuclear coordinates reveals that within the first 60 femtoseconds after photoexcitation, a covalent bond forms between A and B to give A–B + C. The second covalent bond, between B and C, subsequently forms within 360 femtoseconds to give a linear and covalently bonded trimer complex A–B–C. The trimer exhibits harmonic vibrations that we map and unambiguously assign to specific normal modes using only the experimental data. In principle, more intense X-rays could visualize the motion not only of highly scattering atoms such as gold but also of lighter atoms such as carbon and nitrogen, which will open the door to the direct tracking of the atomic motions involved in many chemical reactions. Femtosecond X-ray liquidography is used to track the vibrational wavepacket trajectories of gold atoms in solution, enabling time-resolved observations of the emergence of vibrations and the evolution of the formation of covalent bonds.
We report the generation of gold nanoparticles (AuNPs) from the aqueous solution of chloro(2,2′,2″-terpyridine)gold(III) ion ([Au(tpy)Cl]2+) through X-ray radiolysis and optical excitation at a synchrotron. The original purpose of the experiment was to investigate the photoinduced structural changes of [Au(tpy)Cl]2+ upon 400 nm excitation using time-resolved X-ray liquidography (TRXL). Initially, the TRXL data did not show any signal that would suggest structural changes of the solute molecule, but after an induction time, the TRXL data started to show sharp peaks and valleys. In the early phase, AuNPs with two types of morphology, dendrites, and spheres, were formed by the reducing action of hydrated electrons generated by the X-ray radiolysis of water, thereby allowing the detection of TRXL data due to the laser-induced lattice expansion and relaxation of AuNPs. Along with the lattice expansion, the dendritic and spherical AuNPs were transformed into smaller, raspberry-shaped AuNPs of a relatively uniform size via ablation by the optical femtosecond laser pulse used for the TRXL experiment. Density functional theory calculations confirm that the reduction potential of the metal complex relative to the hydration potential of X-ray-generated electrons determines the facile AuNP formation observed for [Au(tpy)Cl]2+.
Despite numerous experimental and theoretical studies, the proton transfer accompanying the oxidation of 2 '-deoxyadenosine 5 '-monophosphate 2'-deoxyadenosine 5'-monophosphate (5'-dAMP, A) is still under debate. To address this issue, we have investigated the oxidation of A in acidic and neutral solutions by using transient absorption (TA) and time-resolved resonance Raman (TR3) spectroscopic methods in combination with pulse radiolysis. The steady-state Raman signal of A was significantly affected by the solution pH, but not by the concentration of adenosine (2-50 mm). More specifically, the A in acidic and neutral solutions exists in its protonated (AH(+)(N1+H+)) and neutral (A) forms, respectively. On the one hand, the TA spectral changes observed at neutral pH revealed that the radical cation (A(.+)) generated by pulse radiolysis is rapidly converted into A(.)(N6-H) through the loss of an imino proton from N6. In contrast, at acidic pH (<4), AH(.2+)(N1+H+) generated by pulse radiolysis of AH(+)(N1+H+) does not undergo the deprotonation process owing to the pK(a) value of AH(.2+)(N1+H+), which is higher than the solution pH. Furthermore, the results presented in this study have demonstrated that A, AH(+)(N1+H+), and their radical species exist as monomers in the concentration range of 2-50 mm. Compared with the Raman bands of AH(+)(N1+H+), the TR3 bands of AH(.2+)(N1+H+) are significantly down-shifted, indicating a decrease in the bond order of the pyrimidine and imidazole rings due to the resonance structure of AH(.2+)(N1+H+). Meanwhile, A(.)(N6-H) does not show a Raman band corresponding to the pyrimidine+NH2 scissoring vibration due to diprotonation at the N6 position. These results support the final products generated by the oxidation of adenosine in acidic and neutral solutions being AH(.2+)(N1+H+) and A(.)(N6-H), respectively.
Excited-state intramolecular proton transfer (ESIPT) is a fundamental ultrafast photochemical process. Although it has been intensively studied for the development of novel photonic devices such as organic light-emitting diodes, the relation between ESIPT reaction and intramolecular charge transfer (ICT) is still a subject of debate. Furthermore, the effects of the molecular geometry and of the substituent on ESIPT and ICT processes are still unclear. To address these issues, we synthesized a set of four compounds designed to control the electron density distribution of the proton-donating (PD) group and the steric hindrance between the PD and the adjacent phenyl groups: three 2-(1-phenyl-1H-imidazo[4,5-f][1,10]phenanthrolin-2-yl) phenol derivatives, PIPP-Xs (X = H, F, and OMe), and 2-(1-phenyl-1H-imidazo[4,5-f][1,10]phenanthrolin-2-yl) naphthalen-2-ol (PIPN). We then investigated their ESIPT and ICT dynamics as well as the related structural changes using femtosecond transient absorption spectroscopy and theoretical calculations. Although the four compounds commonly exhibit a dual emission originating from the excited enol (E*) and keto (K*) tautomers, their emission properties, such as emission maxima and lifetimes, are systematically modulated by substitution at the para-position of the PD group. The experimental and time-dependent density functional theory calculation results showed that the substitution of an electron-withdrawing group at the para-position of the PD group and the planarity between the PD and proton-accepting (PA) groups play important roles in inducing an efficient ESIPT characterized by increased emission of the K* tautomer. On the other hand, the photoexcitation for PIPP-Xs induces the formation of cis-K*, which is the most stable structure, whereas in PIPN the E* tautomer generated by the photoexcitation is rapidly converted to two species, cis-K* and per-K* with time constants of <0.2 and 0.5 ps, respectively. Furthermore, the per-K* state of PIPN has a charge transfer characteristic, suggesting intramolecular charge migration induced by the formation of per-K* state. This distinctive dynamics of PIPN is due to its pretwisted structure between PD and PA groups. The results provided in this study demonstrate that the molecular geometry plays an important role in the ESIPT and ICT processes.
Despite numerous experimental and theoretical studies, the proton transfer accompanying the oxidation of 2'-deoxyadenosine 5'-monophosphate 2'-deoxyadenosine 5'-monophosphate (5'-dAMP, A) is still under debate. To address this issue, we have investigated the oxidation of A in acidic and neutral solutions by using transient absorption (TA) and time-resolved resonance Raman (TR3 ) spectroscopic methods in combination with pulse radiolysis. The steady-state Raman signal of A was significantly affected by the solution pH, but not by the concentration of adenosine (2-50 mm). More specifically, the A in acidic and neutral solutions exists in its protonated (AH+ (N1+H+ )) and neutral (A) forms, respectively. On the one hand, the TA spectral changes observed at neutral pH revealed that the radical cation (A.+ ) generated by pulse radiolysis is rapidly converted into A. (N6-H) through the loss of an imino proton from N6. In contrast, at acidic pH (<4), AH.2+ (N1+H+ ) generated by pulse radiolysis of AH+ (N1+H+ ) does not undergo the deprotonation process owing to the pKa value of AH.2+ (N1+H+ ), which is higher than the solution pH. Furthermore, the results presented in this study have demonstrated that A, AH+ (N1+H+ ), and their radical species exist as monomers in the concentration range of 2-50 mm. Compared with the Raman bands of AH+ (N1+H+ ), the TR3 bands of AH.2+ (N1+H+ ) are significantly down-shifted, indicating a decrease in the bond order of the pyrimidine and imidazole rings due to the resonance structure of AH.2+ (N1+H+ ). Meanwhile, A. (N6-H) does not show a Raman band corresponding to the pyrimidine+NH2 scissoring vibration due to diprotonation at the N6 position. These results support the final products generated by the oxidation of adenosine in acidic and neutral solutions being AH.2+ (N1+H+ ) and A. (N6-H), respectively.
Although dimer radical ions of aromatic molecules in the liquid-solution phase have been intensely studied, the understanding of charge-localized dimers, in which the extra charge is localized in a single monomer unit instead of being shared between two monomer units, is still elusive. In this study, the formation of a charge-localized dimer radical cation of 2-ethyl-9,10-dimethoxyanthracene (DMA), (DMA)2 .+ is investigated by transient absorption (TA) and time-resolved resonance Raman (TR3 ) spectroscopic methods combined with a pulse radiolysis technique. Visible- and near-IR TA signals in highly concentrated DMA solutions supported the formation of non-covalent (DMA)2 .+ by association of DMA and DMA.+ . TR3 spectra obtained from 30 ns to 300 μs time delays showed that the major bands are quite similar to those of DMA except for small transient bands, even at 30 ns time delay, suggesting that the positive charge of non-covalent (DMA)2 .+ is localized in a single monomer unit. From DFT calculations for (DMA)2 .+ , our TR3 spectra showed the best agreement with the calculated Raman spectrum of charge-localized edge-to-face T-shaped (DMA)2 .+ , termed DT.+ , although the charge-delocalized asymmetric π-stacked face-to-face (DMA)2 .+ , termed DF3.+ , is the most stable structure of (DMA)2 .+ according to the energetics from DFT calculations. The calculated potential energy curves for the association between DMA.+ and DMA showed that DT.+ is likely to be efficiently formed and contribute significantly to the TR3 spectra as a result of the permanent charge-induced Coulombic interactions and a dynamic equilibrium between charge localized and delocalized structures.
Bismuth triiodide, BiI3, is one of the simplest bismuth halides, which have recently attracted considerable attention because of their promising properties. Here, we investigate the structural dynamics of a photoinduced reaction of BiI3 in solution phase using time-resolved X-ray liquidography (TRXL) and density functional theory (DFT) and time-dependent DFT (TDDFT) calculations. The photoreaction was initiated by excitation at 400 nm, which corresponds to the ligand-to-metal charge-transfer transition. The detailed structures and kinetic profiles of all relevant intermediate species from the TRXL data show that the trigonal planar structure of BiI3, which is predicted to be the most stable structure of the lowest excited state by TDDFT calculation, was not observed, and the photoreaction proceeds via two parallel pathways within the time resolution of 100 ps: (i) isomer formation to produce iso-BiI2-I, which relaxes back to the ground-state structure, and (ii) dissociation into BiI2· and I· radicals, which nongeminately recombine to generate ground-state BiI3 and I2.
Conformational isomers of hydroquinone and their 1:1 clusters with water have been spatially separated using a Stark deflector in a supersonic jet. trans-Hydroquinone (HyQ) conformer with zero dipole moment is little influenced by inhomogeneous electric fields, whereas cis conformer with nonzero dipole moment (2.38 D) is significantly deflected from the molecular beam axis into the direction along which the strong field gradient is applied. Resonant two photon ionization carried out by shifting the laser position perpendicular to the molecular beam axis after the Stark deflector then gives an exclusive S1-S0 excitation spectrum of the cis conformer only, making possible immaculate conformer-specific spectroscopy and dynamics. As the spatial separation is apparently proportional to the effective dipole moment strength, conformational assignment could be absolute in the Stark deflector, which contrasts with the hole-burning spectroscopic technique where identification of a conformational isomer is intrinsically not unambiguous. trans- and cis-HyQ-H2O clusters have also been spatially separated according to their distinct effective dipole moment strengths to give absolute spectroscopic identification of each cluster isomer, nailing down the otherwise disputable conformational assignment. This is the first report for the spatial separation of conformational cluster isomers.