302 Charged Particle and Photon Interactions with Matter
12.4.1.2 picosecond pulse-radiolysis systems: a short story and the state of the art
The rst part of the ultrafast pulse-radiolysis story started during the early 1970s and extended to
the mid-1980s. At that time, the electron bunches were emitted by thermionic guns. The rst pulse-
radiolysis experiments were performed using a bunch of picosecond electrons pulses that were sepa-
rated by 350ps as the pump beam (Bronskil etal., 1970). This type of system was also implemented
later during the mid-1980s in Japan (Sumiyoshi and Katayama, 1982; Sumiyoshi etal., 1985). The
Cherenkov light pulses generated in the air by the accelerated electrons were used as the optical
probe. The time resolution and the temporal window of the measured kinetics were limited by the
temporal structure of electron bunches. In 1975, at the Argonne National Laboratory, United States,
the temporal window was extended to several nanoseconds by using single electron pulse generation
(Jonah, 1975; Jonah etal., 1976). An analogue but more advanced conguration named “twin linac
pulse-radiolysis system” (Kobayashi and Tabata, 1985; Tabata etal., 1985) was developed later at
the Tokyo University, Japan; a rst linac was used for the pulse irradiation, while a second generated
the
synchronized Cherenkov optical probe.
The
second phase of the ultrafast pulse-radiolysis history ran from the 1990s to the end of the
2000s and was associated with remarkable progress in shortening the electron bunch duration and
the time resolution of the experiments. Shortening of the electron bunch duration down to 800fs
(Uesaka etal., 1994) was achieved by the implementation of magnetic pulse compressor on the
S-band linac at NERL (Tokyo). Then, a similar device was developed at ISIR (Osaka) to compress
the width of 30 ps electrons bunch down to 125fs (Kozawa etal., 1999), and subpicosecond pulse-
radiolysis measurements were successfully performed (Okamoto etal., 2003). The setup consisted
of a sub-picosecond L-band linac as an irradiation source and a femtosecond Ti:Sa laser, synchro-
nized with the RF eld as an analyzing light (Kozawa etal., 2000). The time jitter between the
optical laser probe and the electron pulse was however several picoseconds and had to be corrected
shot to shot. Nevertheless, this system allowed pulse-radiolysis measurements with 800fs time reso-
lution. Bunch compression is still in use on the current generation of picosecond electrons pulse
accelerators. The performances and the signicant achievements obtained by these systems were
described
by Wojcik etal. (2004).
This
current generation is based on the laser-triggered photocathode RF gun. The electron
bunches are then no more generated by thermionic effect, but are produced by photoemission from
a metal or semi-conductor photocathode hinted by a short UV laser pulse. This laser-triggered
RF gun technology was developed in order to improve the beam quality, notably in terms of more
reduced emittance levels. The laser-driven RF guns immediately became attractive alternatives
for the design of new picosecond pulse-radiolysis systems. First, the lower emittance allows for
a better transport and focusing of the accelerated electron beam, and thus a better pulse-probe
overlap in pulse-radiolysis experiments. Second, the RF-gun design is compact and permits the
construction of smaller facilities at lower costs and easier operation and maintenance conditions.
And last, the electron bunches delivered by the photocathode RF gun can be precisely synchro-
nized with tunable femtosecond optical laser pulses to perform pulse-probe measurements at a
very accurate time resolution that is only limited by the electron pulse duration. Six ultrafast pulse-
radiolysis systems based on a photocathode RF gun and a femtosecond laser source are in operation
all over the world: at Brookhaven National Laboratory (the LEAF facility) (Wishart etal., 2004), at
Osaka University (Kozawa etal., 1999; Saeki etal., 2005; Yang etal., 2006), at Sumitomo Heavy
Industries (SHI, Tokyo) (Aoki etal., 2000), at Waseda University (Tokyo) (Kawaguchi etal., 2005;
Nagai etal., 2007), at the University of Tokyo (NERL) (Muroya etal., 2001a,b, 2005a, 2008), and at
the
University Paris Sud (ELYSE) (Belloni etal., 2004; Marignier etal., 2006).
The
performances and usual operating conditions of the laser-triggered RF guns and of their
associated time-resolved experiments for studying the spurs reactions are described below. They
may be considered as the state-of-the-art pulse-radiolysis systems when this chapter was written
(Muroya etal., 2008). However, new facilities that exploit the laser-plasma acceleration process for