Printer-friendly version



The Ultrafast Photochemistry Group was created at ENS in 2000 and is presently led by Pascal Plaza (CNRS senior researcher). Its research activity globally lies in the field of femtochemistry. Our general objective is to understand the primary chemical reactions underlying the conversion of light by organic chromophores in condensed phase. Condensed phase is more specifically understood as the liquid solution or a macromolecular environment. We particularly focus on the early processes, immediately following the absorption of light, and are particularly interested by the active role played by the environment on the reactivity of the chromophores. The processes we study are often ultrafast, taking place in a timescale ranging from femtoseconds (10‑15 s) to picoseconds (10‑12 s). The rationale is that a high reaction rate is an effective way to convert the energy of light into usable chemical energy before it is degraded by unproductive relaxation pathways into light (fluorescence) or heat.


In order to experimentally follow the details of such reactions, an extremely high temporal resolution is required. This is achieved in our group by using the tools of ultrafast spectroscopy. Our main setup is a pump-probe UV-visible transient absorption spectrometer, with a time resolution of the order of 100 fs and broadband probing over the entire visible region. We also use a fluorescence up-conversion setup, capable of recording fluorescence decays at selected wavelengths, with a time resolution of about 200 fs.



Research Topics


Within this framework the group is particularly interested by the processes at the origin of the function of photoactive proteins, that is, proteins that absorb light through a bound chromophore and use its energy to perform functions such as signaling or catalysis. Studying photoactive proteins often also implies studying the isolated chromophore in solution and comparing its photochemical behavior to that of the full chromoprotein. The aim is here to disentangle the respective roles of the chromophore intrinsic photoinduced reactivity and of the protein environment.


In a distinct but still biology-related field the group also pursues the goal of applying the knowledge acquired at the fundamental level to create new chemical objects, displaying interesting photoinduced functionalities. In this direction the group takes special interest in supramolecular constructs able to photochemically control of the movement of metal ions.





Here follow a couple examples of our recent research. More details can be found in our published articles.


1. DNA photorepair by photolyases


It is well-known that UV irradiation produces damages in DNA, mainly cyclobutane pyrimidine dimers (CPD) and (6-4) photoproducts. Photolyases are proteins capable of using visible light to repair those lesions.1 The key photocatalytic element of photolyases is non-covalently bound cofactor, flavin adenine dinucleotide (FAD; Fig. 1a). For DNA photorepair to occur FAD needs to be in its fully reduced FADH form. Excitation of FADH induces then an electron transfer to the lesion, which in turn trigger the repair. If FAD is not fully reduced, an independent photo-induced reaction called photoactivation may reduce the chromophore, from its oxidized FADox form to its semi-reduced FADH form or from FADH to FADH


We studied the photoactivation reaction of different oxidized photolyases, in particular OtCPF1, a (6-4) photolyase (specialized in the repair of (6-4) photoproducts) of a green alga. By selectively exciting FADox at 470 nm we showed that the excited-state decay takes place in about half a picosecond (Fig. 1b), yielding the reduced FAD•– radical anion. The corresponding electron donor was identified as a tryptophan residue, likely lying in the close vicinity of FAD. Subsequent kinetic steps were identified in the picosecond timescale, with essentially no change of shape of the transient absorption spectra. They were tentatively assigned to a series of electron jumps along a chain of three conserved tryptophan residues (Fig. 1b). At each step of the reaction hole hopping is in competition with charge recombination, from the reduced flavin to the oxidized tryptophan.2 New studies are under way to better characterize the nature of the electron transfer intermediates by transient absorption anisotropy experiments.



Fig. 1. a) Structure of FAD; b) Ultrafast photoactivation of OtCPF1; c) Two-photon DNA repair by (6-4) photolyase.


On a different (6-4) photolyase coming from a frog, we on the other hand undertook the study of the DNA photorepair mechanism itself. In collaboration with the groups of K. Brettel (CEA-Saclay) and J. Yamamoto (Osaka University) we showed that two successive photons are required to complete DNA repair (Fig. 1c). The metastable intermediate (X) formed by the first photoreaction is possibly the same oxetane-bridged T(ox)T dimer known to mediate the formation of the lesion under UV irradiation.3




[1]    A. Sancar, Chem. Rev. 103 (2003) 2203.

[2]    J. Brazard et al., J. Am. Chem. Soc. 132 (2010) 4935.

[3]    J. Yamamoto et al., Angew. Chem. Int. Ed. 52 (2013) 7432.



2. The photoswitching of Dronpa


Dronpa1 is a Green Fluorescent Protein (GFP) homologue exhibiting photochromic properties: it can be switched back and forth between a fluorescent state (ON state) and a non-fluorescent state (OFF state) by irradiation at two wavelengths. It was known from X-ray crystallography and NMR that the photochromic reaction is based on cis/trans isomerization of the chromophore central ethylenic bond and protonation/deprotonation of its phenol group2 (Fig. 2a), but the sequence in time of the two steps and their characteristic timescales were much debated.3


We monitored the entire OFFTON photoswitching process of Dronpa in real time, by transient absorption spectroscopy from 100 fs to 1 ns in ENS, and up to milliseconds in collaboration with F. Rappaport (IBPC). We were able to capture both the spectrum and the anisotropy of the main reaction intermediate, a ground-state species formed upon picosecond excited-state decay (Fig. 2b).4 We showed that the blue-shifted absorption of this intermediate is characteristic for a cis-phenol form of the chromophore. In addition, its anisotropy of 0.24 is in agreement with the large reorientation of the phenol group expected upon isomerization (Fig. 2c, inset). We conclude that trans→cis isomerization of the chromophore precedes its deprotonation and occurs on the picosecond timescale, concomitantly to the excited-state decay. We found deprotonation to follow on a much longer timescale of  ̴10 µs (Fig. 2c).4 This second step is probably triggered by the change in local environment of the phenol group caused by trans→cis isomerization.

Fig. 2. a) Chromophore structure in the ON and OFF states of Dronpa. b) Spectrum and anisotropy of the cis-phenol photoswitching intermediate 300 ps after excitation of OFF-state Dronpa. c) Model of the OFF→ON photoswitching of Dronpa. Inset : Alignment of the X-ray-structures of the ON and OFF states showing the reorientation of the chromophore phenol group upon cis/trans isomerization. PDB IDs : 2IOV (ON) and 2POX (OFF).




[1]    Ando et al., Science 306 (2004) 1370.

[2]    Andresen et al., Proc. Nat. Acad. Sci. USA 104 (2007) 13005; Mizuno et al., Proc. Nat. Acad. Sci. USA 105 (2008) 9227.

[3]    Fron et al., J. Am. Chem. Soc. 129 (2007) 4870; Warren et al., Nature Comm. 4 (2013) 1461; Lukacs et al., J. Phys. Chem. B 117 (2013) 11954.

[4]    Yadav et al., J. Phys. Chem. B, in press (DOI: 10.1021/jp507094f)



3. Ionic molecular shuttle


Molecular switching from one state to another in a bistable system, induced by an external stimulus (pH, redox potential, light), is an active area of research aiming at controlling chemical, electrical, and optical processes at the nanoscale level.1 Applications to molecular memories, binary logic computing and molecular machines are concerned.2


In collaboration with the group of B. Valeur (ENS-Cachan) we were interested in the innovative concept of ionic molecular shuttle. It is about implementing molecular bistability via the photocontrolled translocation of a metal ion between the two binding sites of a ditopic receptor. Such system could be used to reversibly write logical information at the molecular level. Our strategy consists in attaching to each binding site a photoactive center which, upon irradiation, can create a positive charge in the vicinity of the bound cation. The photoinduced electrostatic repulsion would then push the cation towards the other site. Ideal devices would bear two different photoactive centers, excitable at two different wavelengths, in order to fully control back and forth operation (Fig. 3a).

Figure 3. a) Principle of a photocontrolled ionic shuttle; b) Calix-DCM2; c) Transient absorption spectra of the Calix-DCM2:K+ 1:1 complex (ML); d) Typical structure of ML calculated by DFT and used to evaluate the interchromophoric distance.


We first studied a model compound containing two identical photoactive centers (DCM), attached to a 1,3-alternate calix[4]biscrown receptor (Calix-DCM2; Fig. 3b).3 Since Calix-DCM2 is symmetrical, the identical initial and final states of the translocation process cannot be distinguished by steady-state spectroscopy. Excited Calix-DCM2 is on the contrary dissymmetrical, which we could use to prove the photoinduced cation exchange by transient absorption spectroscopy. We studied the 1:1 complex with K+ and observed the switching from one excited complexed DCM-crown unit to one uncomplexed DCM-crown unit. The reaction includes two main short components: 0.83 ps and 10 ps, which can be assigned to phototranslocation. A smaller-weighted third component of 101 ps might include a competition between phototranslocation and excitation energy transfer as shown by calculations using Förster’s theory (Fig. 3d).


Other experiments involving a dissymmetrical ionic molecular shuttle are on their way.




[1]    B. L. Feringa in Molecular Switches, Wiley-VCH, Weinheim, (2001).

[2]    Special Issue on Molecular Machines, Acc. Chem. Res. 34 (2001).

[3]    B. Valeur et al., ChemPhysChem 11, (2010) 2416.