Time-Dependent Radiolytic Yield of OH• Radical Studied by

ARTICLE pubs.acs.org/JPCA

Time-Dependent Radiolytic Yield of OH• Radical Studied by Picosecond Pulse Radiolysis Abdel Karim El Omar,† Uli Schmidhammer,† Pierre Jeunesse,† Jean-Philippe Larbre,† Mingzhang Lin,‡ Yusa Muroya,§ Yosuke Katsumura,§ Pascal Pernot,† and Mehran Mostafavi*,† †

Laboratoire de Chimie Physique/ELYSE, UMR 8000 CNRS/Universit
e Paris-Sud 11, Facult
e des Sciences d’Orsay, B^at. 349, 91405 Orsay Cedex, France ‡ Nuclear Science and Engineering Directorate, Japan Atomic Energy Agency, 2-4 Shirakata shirane, Tokaimura, Nakagun, Ibaraki 319-1195, Japan. § Nuclear Professional School, School of Engineering, University of Tokyo, 2-22 Shirakata shirane, Tokaimura, Nakagun, Ibaraki 319-1188, Japan ABSTRACT: Picosecond pulse radiolysis measurements using a pulse-probe method are performed to measure directly the time-dependent radiolytic yield of the OH• radical in pure water. The time-dependent absorbance of OH• radical at 263 nm is deduced from the observed signal by subtracting the contribution of the hydrated electron and that of the irradiated empty fused silica cell which presents also a transient absoption. The time-dependent radiolytic yield of OH• is obtained by assuming the yield of the hydrated electron at 20 ps equal to 4.2  10
7 mol J
1 and by assuming the values of the extinction coefficients of eaq
and OH• at 782 nm (ελ=782 nm = 17025 M
1 cm
1) and at 263 nm (ελ=263 nm = 460 M
1 cm
1), respectively. The value of the yield of OH• radical at 10 ps is found to be (4.80 ( 0.12)  10
7 mol J
1.

’ INTRODUCTION It is well-known that water decomposes under ionizing radiation into eaq
, OH•, H3O+, H•, H2, and H2O2. These radicals and molecular products are formed in the spurs, and the radiolytic yield of induced species after around 200 ns is now well established. Pulse radiolysis is a powerful tool to determine the time-dependent yield of the radicals and is used to follow the transient species directly in various conditions. Among the products of water radiolysis, two of them are of particular interest because of their yields, redox properties, and fast reactivity: the hydrated electron and the OH• radical. As the hydrated electron has a strong absorption band, its time-dependent radiolytic yield was determined first. Bartels et al. measured its decay on a continuous time window up to 200 ns by combining time-correlated absorption spectroscopy and standard time-resolved absorption measurements with ns time-resolution.1 The initial yield of formation of the hydrated electron was obtained by the extrapolation of the fitted decay and it was found to be G = 4.2  10
7 mol J
1. It has been measured directly with a higher time-resolution by Muroya et al. who reported a value for G0 = (4.2 ( 0.2)  10
7 mol J
1.2 In contrast, even if the G-value after 100 ns is well-known, the initial G-value of the OH• radical is still controversial. There is no reliable time-resolved measurement with high time resolution in the UV, where the OH• radical has an absorption band. The only direct measurement of G-values of the OH• radical has been performed by the pioneer work of Jonah and co-workers who used picosecond pulse radiolysis with 200 ps resolution.3 The authors reported the time dependence of G(OH•) by following r 2011 American Chemical Society

the decay of its weak absorption band at 280 nm. According to their observations, a value of 6.1  10
7 mol J
1 was estimated at 200 ps based on an estimated G200ps (es
) = 4.7  10
7 mol J
1 for the hydrated electron. The decay of OH• was obtained by averaging the data obtained in pure water and in 4 M perchloric acid. They report that the OH• radical decays to 0.73 of its initial value from 200 ps to 3 ns. Jay-Gerin and Ferradini reported a G(OH•) at 100 ps of only 4.7  10
7 mol J
1.4 This controversy on the radiolytic yield of OH• at short time was later discussed.5 The difficulties for a direct determination of the time-dependent G-value of OH• result from its very weak absorption coefficient in the middle UV.6 Other evaluations of the G(OH•) were obtained from scavenging studies using formate or hexacyanoferrate ions and halide anions.5,7,8 Recently, the transient absorption signals recorded at 340 nm in 2 M Cl
solution, in neutral and acidic conditions (1 M HCl) corresponding mostly to the formation of ClOH•
and Cl2•
were reported. From the analysis of the decay, the value for G(OH•)100ps was estimated to be around 4.3  10
7 mol J
1.9 This value is in good agreement with the theoretical prediction4 G(OH•) = (4.7 ( 0.25)  10
7 mol J
1 at 100 ps. There are two fundamental problems with the approach using highly concentrated scavenger or highly acidic solution: The direct effect of radiation on the solute and the scavenging of prehydrated Received: August 22, 2011 Revised: September 30, 2011 Published: October 04, 2011 12212

dx.doi.org/10.1021/jp208075v | J. Phys. Chem. A 2011, 115, 12212–12216

The Journal of Physical Chemistry A

ARTICLE

electron should be evaluated.10,11 In the present work, we use a new picosecond pulse radiolysis setup at the electron accelerator ELYSE in Orsay for measuring low absorbance in the middle UV and synchronously in the near-infrared. The pulse-probe measurements provide directly the time-dependent radiolytic yield of the OH• radical in pure water with a time resolution of 10 ps. We discuss the value of the initial yield of the OH• radical and compare it with the data reported in the literature.

’ EXPERIMENTAL SECTION ELYSE is a laser-trigged picosecond electron accelerator based on the radiofrequency photogun technology.12,13 Shortly, a femtosecond light pulse generates photoelectrons on a photocathode which are in turn accelerated by RF fields. The presented transient absorbance measurements were performed at the experimental area EA-1 of ELYSE at a repetition rate of 10 Hz with the electron energy set to 7 MeV and the pulse charge to 4 nC (( 10%). Details on the setup and the typical parameters of measurement and data acquisition can be found elsewhere.14,15 The electron pump and the optical probe pulses are both generated by the same femtosecond Ti:Sapphire laser source (∼2 mJ at 782 nm, pulse duration 100 fs) and are so intrinsically synchronized. Here a special optical configuration is implemented to probe synchronously the absorbance of the OH• radical in the middle UV and of the solvated electron in the near-infrared. After passing the optical delay line, the third harmonic of the initial laser wavelength is generated near to the sample cell. To this aim, a part of the collimated laser wave is frequency doubled in a BBO crystal (type I SHG), the polarization of the remaining laser wave is turned by 90°, and the sum frequency was mixed with the frequency doubled wave in a second BBO crystal (type I SFM). Most of the initial and frequency doubled waves are rejected by a dichroic mirror reflecting the third harmonic and a polarizer. The collinear first and third harmonic probe beams are focused slightly into the sample with a lens. A reference beam is split off before the sample. The further configuration and data acquisition corresponds to the one of the broadband transient absorption setup.15 The signal and the reference beam, here containing both the first and third harmonic, are coupled each in an optical fiber; they are transmitted to a spectrometer (Shamrock SR-303i with grating blazed at 500 nm and 150 L/mm) and dispersed onto a CCD (Newton DU920N-BU, Andor technologies). So, both of the wavelengths in the UV and in the near-infrared are recorded and analyzed synchronously to obtain the change of absorbance relative to a probe pair of signal and reference delayed 10 ms to the electron pulse excitation. A flow cell consisting of two fused silica windows, each 1 mm thick, and with an optical path length of 5 mm was used. The cell was placed 30 mm away from the output window of the ELYSE vacuum tube, behind a 200 μm thin aluminum mirror. The latter was used to direct the optical probe beam collinear with the electron beam through the cell. The diameter of the electron bunch at the position of the cell was in the range of 3
4 mm, and the one of the probe beam is on the scale of 100 μm. The spatial overlap of the electron beam on the probe beam was optimized on the transient absorption signal of water inside the cell before transient absorbance measurements in the empty cell. Under these conditions and at the position of the cell, single shot electro-optic sampling of the electric field copropagating with the relativistic electron bunch revealed a pulse duration around 10 ps

Figure 1. Pulse
probe measurements at 782 nm in an empty fused silica cell and in the same cell containing pure water. The sample thickness of water is 5 mm; the cell consists of two windows of 1 mm each. The experiments are performed under the same conditions with data points every 15 and 5 ps (inset), respectively.

Figure 2. Pulse
probe measurements at 263 nm in an empty fused silica cell and in the same cell containing pure water. The sample thickness of water is 5 mm; the cell consists of two windows of 1 mm each. The experiments are performed under the same conditions with data points every 15 and 5 ps (inset), respectively.

and a rms shot-to-shot jitter