Topic 1 : Effects of Extreme Conditions in Radiation Chemistry
In numerous real situations, radical reactions occur with extreme thermodynamic conditions like temperature over 300°C, with pressure over 100 MPa, or triggered by nuclear reactions which produce elevated Linear Energy Transfer (LET) ionizing particles. These parameters can induce drastic changes in the production and in the reactivity of radical species in water. The most recent data bases show a lack of knowledge. Therefore the effect of the combinations of extreme values of these conditions has never been considered: high temperature water submitted to high LET particles for instance.
In the context of the generation 4 nuclear power plant systems in which temperatures of fluid transfer should increase in order to improve the thermal-cycle yield, one system expected to be developed will involve supercritical water (374°C, 22 MPa). Consequently many fundamental questions on water radiolysis in these conditions appeared since Arrhenius extrapolation could not be applied. How to predict water decomposition in these conditions? That was essentially the aim of Dimitri Saffré PhD (2008-2011) who has performed experiments and MC simulations in a large range of temperature and LET in order to determine radiolytic yields and rate constants through an extreme condition-resistant chemical system: HBr. In situ analysis of Br-, Br2- and Br3- species allowed the determination of HOŸ yields in various conditions of irradiation: X-rays at ESRF (FAME Line, J.L. Hazeman), ns-electron pulses from ALIENOR accelerator in LRAD, ps-electron pulses from ELYSE in LCP/Orsay (M. Mostafavi), pulsed heavy ions in GANIL (E. Balanzat), helion beam in ARRONAX at Nantes (M. Fattahi). During this study a novel high temperature/high pressure cell has been designed and used with high LET particles (figure 1). Only a part of the work was published. Publications of valuable results (figure 2) are still on the way because they are waiting for the comparison with MC simulations performed in collaboration with B. Gervais (CIMAP Caen) and M. Beuve (IPN Lyon). In parallel implementation of water radiation chemistry has been performed in the open source program GEANT4 in the ANR named GEANT4-DNA (2009-2012) by collaborating with S. Incerti (CENBG Bordeaux).
Another high pressure system has been designed recently to reach 400 MPa with a flow and optical windows for pulse radiolysis experiments. This system is devoted to the analysis of protein such as myoglobine under pressure stress in order to probe their tertiary structure and their reactivity towards water radical. (see topic 2)
 Baldacchino, G. et Hickel, B.. In: Hors série « Radiation Chemistry » de l’Actualité Chimique Sciences, E. (Editor) EDP Sciences; 2008.
 Elliot, A.J. et Bartels, D.M. The Reaction Set, In: AECL; 2009
 Saffré, D. et al., (2011) Journal of Physics: Conference Series, 261, 012013
 Incerti, S. et al., (2010) Medical Physics, 37, 4692-4708.
 Nguyen Le, D.T., et al. (2013) NIMS B, 299, 1-7
 Rapport à 6 mois sur l’avancement du projet SIRMIO – compte rendu de la réunion d’avancement du 8 mars 2013, convention Plan Cancer PC201204.
Fig.1: Picture of the optical cell designed for high LET particle irradiations.
Fig.2: Transient signals at 357nm for various high temperatures of Br2- (formation) and Br- (bleaching) in a 10mM NaBr solution irradiated by 130µs pulses of C6+ of 975MeV.
We are currently developing a new approach to analyze the fundamental aspects of the energy Deposition of swift Heavy Ions in Liquid Water (SIRMIO project 2012-2014). It should induce application in real time microdosimetry during hadrontherapy. This project consists in using fluorescence of molecule produced during radiolysis and observed by optical microscopy and fast detection of fluorescence induced by laser excitation. We expect to get images of energy deposition of energetic ions at micrometer scale with high time resolution as well.
After 6 months of work the 4 partners got strong knowledge of the molecules to use, in terms of toxic effect in living cells, of radiolytic mechanisms and also purity, radiolytic yields. During these investigations, new methods in radiation chemistry should be available such as pulse radiolysis method with fluorescence induced by laser, and also a method to obtain the fluorescence lifetime coupled to alpha ionization (developed at IPHC Strasbourg). We expect to use very soon the microscopy in line with Van de Graaf accelerator, microprobe at LEEL in Saclay to reach our objectives.
Topic 2 : Proteins under non specific stress
We are interested in the behaviour of proteins under various stresses. The first stress source is the action of radicals, especially hydroxyl radical. The second one source is the pressure and the third one is the exposure to surface, and more specifically to nanomaterials.
Radical stress on amino acids, peptides and proteins
Within the field of biochemistry, we are particularly interested in the mechanisms implicated in the reactions of Reactive Oxygen Species (ROS) with biological macromolecules and in evaluating protective agents against radical species. Indeed, the interaction of both ionizing radiation and chemicals (alcohol, tobacco, metals, pesticides, asbestos fibers, nanoparticles, drugs…) with living matter can lead, under the sensitizer effect of oxygen, to an uncontrolled production of ROS. When cellular protections (enzymes, antioxidants) against ROS are insufficient, these radical species, which are unstable and highly reactive, will react with their environment and induce oxidative stress. By reacting mainly on macromolecules such as DNA, proteins and lipids, ROS cause subcellular and cellular damage which can be implicated in a number of pathologies.
These works were applied in structural biochemistry. Thus, a new footprinting method for mapping protein surface and protein-ligand interactions has been developed using tritium as a radioactive label.  Rather than a tritium labelling, we studied on an oxidative labelling and detection of residues oxidation by mass spectrometry.
Our results have also demonstrated that radicals, formed during the protein oxidation, can migrate between residues at nanometric distances. In fact, the formation of radical species has not a local character and the mechanism implies that carbon-centered radicals can transfer hydrogen atoms between residues sides chains. These transfers have been identified by coupling radiolysis, isotopic labelling and analyses by mass spectrometry. The similar observation on the natural amino-acids and model peptides show that this transfer phenomenon seems quite general. This could explain the sensitivity of particular protein residues to oxidation.
Fig. 3: Aliphatic aminoa acids, like leucine, can transfer H atoms almost as efficiently as Tyr or Cys.
Pressure stress on proteins
The application of hydrostatic pressure to a protein solution provides a manner to alter the structure of proteins and their solvent interactions. In general, protein-ligand binding is affected by pressures lower than 400 MPa. Furthermore, protein denaturation and unfolding may occur at higher pressures. The effects of pressure on hemeproteins have been the subject of numerous investigations (optical absorption, fluorescence, FTIR, Raman, NMR and SANS). The modifications are observed at the level of the active site of myoglobin and of its secondary structure with an alteration of the electrostatic and hydrogen-bond array. However no study has been reported on the use of radicals as probes of reactivity of proteins under pressure. In order to answer these questions, we gave, for the first time in 2005, the results of protein reactions with radicals generated by pulse radiolysis under pressure up to 100MPa. A new setup - including an optical flow cell and devices which can increase pressure up by controlling the ramp rate - was built and tested. New data were now obtained (2012) on myoglobin with pressure up to 400 MPa.
Adsorption stress on proteins.
We have developed since 2009 a research program to analyse the impact of inorganic nanomaterial on protein. In biological fluids, proteins are indeed the first molecules to come into contact with nanoobjects, which they will absorb on. This adsorption phenomenon controls both the biodistribution of nano-objects and the immune response to the presentation. It is therefore a key process to develop nontoxic nanomaterial and more efficient nanodrugs. But the adsorption induces a stress on the adsorbed proteins: their structure and therefore their function are subject to change and susceptible to induce toxicity. Our study have focused on i) the structural determinants of protein adsorption, ii) the impact of protein adsorption on their structure and dynamics and, iii) the consequence in term of protein function and activity.
To understand the structural determinants of protein adsorption, we have developed proteomic methods in collaboration with Jean Labarre (CEA/DSV). We found that the adsorbed proteins contain a higher number of positively charged amino acids which is consistent electrostatic interactions with silica. The analysis also identified a low aromatic amino acid content (phenylalanine, tryptophan, tyrosine and histidine) in adsorbed proteins. Structural analyses and molecular dynamics simulations of proteins from the two groups indicate that non-adsorbed proteins have higher structural rigidity. The data are consistent with the notion that adsorption is correlated with the flexibility of the protein and with its ability to spread on the surface. Our findings lead us to propose a refined model of protein adsorption.
Fig. 4: Hard proteins, that contain more aromatic residus, are less prone to adsorption because they are less deformable.
To understand the changes induced by adsorption, we have analyzed the behaviour of model proteins such as myoglobin and hemoglobin on various types of nanomaterials. Our results show that adsorption isotherms of pig oxyhemoglobin on silica NP are well fitted by Langmuir model. The adsorption is pH dependent. Although Circular Dichroism spectra of adsorbed hemoglobin reveal a partial loss of helix secondary structure, oxygen affinity of adsorbed hemoglobin increases whereas oxygen binding cooperativity decreases. The effectors binding on adsorbed hemoglobin, decreasing the oxygen affinity of the protein, suggests that hemoglobin molecule keeps its tetrameric form upon adsorption. To study if the silica nanoparticles could act as an effector for haemoglobin, these data are now applied to normal of abnormal human hemoglobin.
Topic 3: Reactivity and dynamics in porous media
As technology progresses to the nano-scale, the question of how chemical reactions are perturbed by the size of the environment needs further consideration. Confinement and interfacial phenomena are expected to dramatically affect reaction rates and mechanisms. We try in this topic to provide a global picture of the dynamic and the reactivity in nanoporous media, with special foci on confined water properties.
Most of these studies are performed using radiolysis to induce a chemical reactivity. Indeed ionising radiations induce reactions throughout the entire sample, whatever its complexity, shape or transparency. Furthermore, in radiolysis, information on the chemical reactivity at the microscopic scale can be obtained from macroscopic analytical strategies.
At the beginning of this topic most of our work was concerned by the pure geometrical confinement effects on radiolysis. Therefore, our studies were conducted only in nanostructured silica of various pore sizes. During the 2009-2013 period, we chose to focus our work on the impact the interface nature on radiolytic phenomena. We carried on a strong effort to develop in the lab the synthesis and the characterization of appropriate materials: i) a liquid atomic layer deposition system was constructed to allow the grafting of nm thick oxide layers  ii) we carried on the large scale preparation of metallic nanosponges with sufficiently large porous volumes to allow the quantification of radiolytic products iii) a surface sol gel method was use to prepare nanocomposite hollow oxide fibers iv) porous materials were grafted with an organosilane layer, and the nature of this monolayer was extensively described using solid state NMR (in collaboration with the LSDRM).
We then used these tailored materials to mesure various radiolytic yields using the gamma source of the lab.
Fig. 5: Ratio of the hydroxyl radical production in confined water with respect to bulk, as a function of the HO capture time after its production
 M. Alam, et al. J. Mat. Chem., 2009 : 19, 4261 2009
 S. Le Caër, et al. J. Phys. Chem. C, 2012, 116 :4748.
 R. Musat, et al. Phys. Chem. Chem. Phys., 2010, 12:12868
 S. Le Caër et al. Phys. Chem. Chem. Phys., 13: 17658. 2011.
 S. Le Caër et al. J. Phys. Chem. C 116, 12916, 2012
 R. Musat et al. Chem. Comm., 46: 2394. 2010.
We could demonstrate that, on oxide surfaces, the nature of the extreme surface controls the interfacial radiolysis, and that the presence of a minimal amount of small band gap oxides on the interface is sufficient to deactivate the H2 production.12 In the vicinity of metal surfaces, the radiolysis events showed a dynamic very different from bulk, with a very important overproduction of radiolytic species followed by their rapid consumption. The organic grafting strategy allowed analysing not only the breaking of OH bonds, but also of CC, CH and SiH in the same layer.13 This study revealed that there was no relationship between the bond stability and its radiosensitivity, and that irradiation led a global desorganisation of the self adsorbed monolayer.In parallel, we carried on our study of confined and surface water properties, both in porous silica and in clays, using vibrational spectroscopy.
For porous silica, the infrared spectra of confined water in the 30–4000 cm−1 spectral range, confirmed our previous ultrafast IR experiment, showing that the water properties were perturbated even in large (50 nm) pores. We detected in the far infrared domain a stiffening and an ordering of the hydrogen bond network.
Fig. 6: Transient spectra measured for the Li+–Montmorillinite (11%humidity: ●; 40% humidity: ▲) and Ca2+– Montmorillinite (11% humidity: ▼; 40% humidity: ◆) at a delay of 800 fs. The spectra are recorded with a pump frequency centered around 2430 cm–1. 
The dynamics of water molecules confined in the interlayer space of montmorillonites (Mt) and in interaction with Li+ and Ca2+ cations was followed using non linear infrared spectroscopy of the O–D bond. The anisotropy experiments evidenced the absence of rotational motions on a 5 ps time scale, proving that the hydrogen bond network in the interlayer space of the clay mineral is locked at this time scale. The overall water behaviour was evidenced to be overwhelmingly dominated by the interaction with the cation.
We are now sufficiently familiar with clays to envision to perform reactivity studies inside the 2D water network they trap. We have started to measure radiolysis products in these systems. We could then evidence that the H2 production is linked to the water amount in the interlayer space. In the future we will use the same 2 photon water photoionization schemes we recently used in porous glasses to follow solvated electron reactions.