EUROPEAN ORGANIZATION FOR NUCLEAR RESEARCH
CERN-INTC-2005-017
INTC-P-171
Proposal to the ISOLDE and Neutron Time-of-Flight Experiments Committee
Measurement of Gas and Volatile Elements Production Cross Section in a Molten Lead-
Bismuth Target: addendum
L. Zaninia, M. Fallotc, H. Frånberga,b, F. Gröschela, T. Kirchnerc, U. Kösterb, E.
Manfrina, H. Ravnb, Y. Tallc, W. Wagnera, M. Wohlmuthera
a
Paul Scherrer Institut, CH-5232 Villigen PSI, Switzerland
b
CERN, CH-1211 Geneva, Switzerland
c
SUBATECH, 44307 Nantes-cedex 3, France
Spokesperson: L. Zanini
Contactperson: U. Köster
Abstract
Preliminary results from the IS419 experiment indicate that the ISOLDE facility can provide
valuable data for the MEGAPIE experiment. Production rates for many elements from He to
At, following interaction of the 1.4 GeV proton beam with an ISOLDE target filled with
LBE, were measured. Following the first measurement runs of the IS419 experiment in 2004,
we propose additional measurements during the 2005 run which should yield further essential
information.
Summary of preliminary results from the first measurement run
The first measurements were performed in April/May 20041,2. The spallation target consisted
of a standard ISOLDE cylindrical tantalum container filled with liquid LBE (lead bismuth
eutectic alloy). Protons pulses of 1.4 GeV and variable intensity (up to 1013 protons/pulse with
a rate of one pulse every 16.8 s) impinged on the target.
An additional measurement was planned with the same target at the energy of 1 GeV;
however, following ion source failure, it was decided instead to proceed with a measurement
at the same energy using a target filled with liquid Pb.
1
Yields were measured using three different techniques of common use at ISOLDE. Online
yields of stable isotopes and of some radioactive ones were measured by a Faraday cup
inserted in the beam line. A special data acquisition system3 was developed to trigger the
current measurement by a picoampèremeter with the arrival of the proton beam on target, thus
allowing the measurement of the gas release curves, characteristic of each element.
For short-lived emitting isotopes, beams were directed to a dedicated tape station and yields
were measured with a plastic scintillator detector.
A third measurement method was used for longer lived (T1/2 ≥ 5 min) emitting
radioisotopes; ion beams were implanted on thin Al foils, then after irradiation an offline
detection was performed using a calibrated HPGe detector. Collection measurements were
performed for a number of isotopes.
For a given isotope, the measured yield has two components, one from direct production from
the target and one from the decay of parents. Isotopes were collected in an order chosen so
that the first ones to be measured were the first reaching equilibrium, having parents with
shorter half-lives. In this way most of the measured isotopes were in equilibrium with their
parents, with only a few exceptions.
In order to obtain the absolute production rates from the measured yields, the efficiency of the
ion source had to be measured. For this purpose, known amounts of different gas mixtures
(consisting of Ar/Xe, He/Ne/Ar/Kr/Xe, and 3He/Ar/Xe mixtures) were leaked into the ion
source, thus having the possibility to measure the efficiencies at any time during the
experiment.
For the measurements with the LBE target, the temperatures of the target were of 400 ºC and
600 ºC. The Pb target was at a temperature of 520 ºC. These temperatures are in the range of
the LBE temperature in MEGAPIE during operation, which varies from 300 ºC to 400 ºC
depending on the position inside the target. Temperature differences within these ranges are
not expected to affect the released fraction of the long-lived noble gas and Hg isotopes which
are determining for the overall radiotoxicity assessment. On the other hand, differences are
expected for some isotopes such as I, Cd and Po. Having performed the experiments at higher
temperatures than in MEGAPIE will allow to conclude, in case the release of a specific
isotope is not observed at 600 ºC, that no release should be observed in MEGAPIE for the
same isotope at 300-400 ºC, even for longer irradiation times.
During the first measurement, with the LBE target, it was found that the short term
component of the release curve exhibited discontinuities probably related to splashing effects
in the target which reduced for a few tens of ms the ionization efficiency of the ion source.
While this affects only slightly the absolute release, which is dominated by the long
component, it makes it more difficult to fit the release curve. No such effect was observed
during the second measurement, with the Pb target, where the proton beam intensity was
reduced to 1.5×1012 protons/pulse.
We present in the following the most relevant results obtained from the preliminary data
analysis.
Helium
After hydrogen, helium is the volatile element produced in larger quantity during the
operation of MEGAPIE, and it is therefore of great interest. In Fig. 1 the 4He current
measured by a Faraday cup for 6 s after the arrival of the proton beam on the Pb target is
shown.
The ionization and transmission efficiency from the ion source to the Faraday cup was
measured to be 0.05 % for 3He. Assuming the same ionization and transmission efficiency for
4
He, the production rate for 4He is 0.77 atoms/p, with a systematic uncertainty of about 20 %.
The value obtained with the LBE target is slightly lower, possibly because of a systematic
2
effect from the splashing due to the higher intensity proton beam from the first run. This value
is in good agreement with calculations with MCNPX with the Bertini/Dresner models (see
explanation below), giving 0.84 atoms/p, while the MCNPX run with the INCL4/ABLA
models give a value a factor of 2 lower.
FIGURE 1. Current of 4He (in pA) measured by the Faraday cup.
Mercury
Hg isotopes are extremely important from the safety point of view as during MEGAPIE
operation this element is going to dominate the activity as far as volatile elements are
concerned.
In Figure 2 the measured cumulative production rates for emitting Hg isotopes (measured
with the collection foils technique) are presented and compared with Monte Carlo
calculations. Long-lived Hg isotopes are expected to be completely released at the
temperature of 600 ºC.
The ionization efficiency was not measured for Hg, as it was only measured for noble gases.
In this case only indicative results can be extracted: based on previous results from Ref. [4],
we considered an efficiency of a factor 1.5 higher than the measured Xe efficiency of
3.7(11)%. However, given the importance of the Hg measurement, it would be highly
recommended to have a direct measurement of the ion source efficiency in the Hg mass range.
The measured values are in line with expected cumulative production rates calculated using
the Monte Carlo transport codes FLUKA5 and MCNPX6. The two codes were coupled with
the evolution codes ORIHET37 and SP-FISPACT8, respectively. In the case of MCNPX,
results are shown here with two different model combinations for the intranuclear cascade and
evaporation models. The circles represent results from using the Bertini intranuclear cascade
model with the Dresner evaporation code. The diamonds are obtained using the recently
implemented INCL4/ABLA9 model combination. The trend observed in the data as a function
of the atomic mass is well reproduced by the three calculations. One should note that for
193
Hg, 195Hg and 197Hg, there are isomeric states of 11.1 h, 40 h and 23.8 h half-lives,
respectively. For these three isotopes, equilibrium was not achieved between formation and
decay of the respective isomeric states, a process which is difficult to properly calculate with
3
existing Monte Carlo codes. Overall these results confirm the expected production rates of Hg
isotopes in a thick LBE target.
Figure 2. Production rates for Hg isotopes. Measured points (black squares) are compared with
calculations: open circles: MCNPX (Bertini/Dresner model combination); diamonds: MCNPX
(INCL4/ABLA); stars: FLUKA.
In Figure 3 a comprehensive display of the experimental results for Hg isotopes obtained with
the three techniques with the LBE target is shown. Data refer to measurement with the target
at 400 ºC, with the exception of the collection data which were taken at 600 ºC. However, it is
apparent from the figure that at the two temperatures the released fraction of long-lived
isotopes is very similar.
With the exception for the points for A>204 (where background of stable Pb and Bi is
relevant), there is an excellent agreement between the Faraday cup and collection data. Monte
Carlo calculations will be performed with higher statistical precision to calculate also the tail
of the atomic mass distributions of the production rates.
Figure 3. Measured production rates for Hg isotopes from the Faraday cup (triangles), spectroscopy
from collection data (squares), and detection from the tape station (circles).
4
Xenon and iodine
Results for Xe isotopes, also measured with the LBE target at T=600 ºC, are shown in Fig. 4.
In this case there is a clear disagreement between the values calculated with MCNPX with
Bertini/Dresner, and the results from the other two calculations. The data, with a measured
ionization efficiency of 3.7 % for Xe isotopes seem to favor the other two calculation results,
thus confirming recent experimental findings10.
Similar results are obtained for the iodine isotopes. However, iodine is not completely
released and observed production rates at 600 ºC are a factor 10 lower than the calculated
FLUKA and MCNPX (INCL4/ABLA) values.
FIGURE 4. Same as Fig. 2 but for Xe isotopes.
While production of Hg isotopes from Pb/Bi target is due to direct spallation, the Xe and I
isotopes are the results from a later stage of the spallation process, the fission of highly
excited spallation fragments, or as a two-step process due to neutron induced fission from
high energy spallation neutrons. Thus the evaporation models, the Dresner and ABLA, are
probably most responsible for the differences observed in the calculations.
Tape station data for some Xe isotopes are available and are in rough agreement with the
collection data. It must be noted that for some isotopes such as 125Xe and 127Xe the observed
count rates refer only to transitions from decay from isomeric states.
Polonium and astatine
Among the other isotopes measured, it is of particular interest to discuss the Po and At.
Production rates of 207,208,209,210At of the order of 107 atoms/C (assuming the same ionization
efficiency as for Hg) were detected, with values an order of magnitude lower for 206At. Such
production rates are not of concern for an ADS. Production of At comes from several possible
reactions of Bi, but the most likely, given the high proton energy, is the double charge-
exchange reaction 209Bi(p,- xn)210-xAt. The At decay is responsible for the observed small
quantities of Po isotopes, which contrary to At is expected to be produced in large amounts.
However, as found in Ref. [11], little or no Po should be released at 600 ºC.
5
Other results
Of the other isotopes measured in the first measurement, no release of Br was observed, while
very little amounts of Cd isotopes were detected. For the Kr isotopes, some problems during
the measurement rendered the analysis questionable and such measurement was repeated with
the Pb target. The analysis of these data is in progress.
Motivation for additional measurement
Additional measurements are proposed in 2005 to fulfill the following tasks:
1) Measurement at a lower energy, of 600 MeV, which is very close to the energy of the
MEGAPIE proton beam energy (575 MeV).
2) Measurement with an extended spacing of the proton pulses (e.g. one proton pulse every
60 or 120 s only) to measure the slow release at a reduced target temperature.
3) A fraction of the shifts will be dedicated to additional measurements of the ion source
efficiency. It is apparent that this is the greater source of systematic uncertainty in the data,
and it is therefore recommended to perform additional measurements. This will include not
only the measurement by gas leak, but also by evaporation of known amounts of non noble
gas tracers, in mass regions closer to the masses of interest (for instance gold tracers for Hg).
4) A better characterization of the ion source behavior as a function of the magnetic field
applied will be performed. According to previous experience, there is a rather strong
dependence and we think that this may affect significantly our absolute measurements.
5) The finite response of the plastic scintillator at the tape station to gamma radiation will be
measured using mono-energetic gamma sources.
The experimental setup will be the same as for the previous experiment.
Beam Time request
We request eight shifts for the measurements (4 at 1.4 GeV and 4 at 600 MeV). A good
fraction (about two third of the time) of these measurements can be performed in the
GLM/GHM beamlines fully in parallel with another experiment using all remaining proton
pulses at the HRS. Only the measurements with the monitoring tape station require the use of
the central beamline.
Furthermore, two extra shifts offline for efficiency calibration with gas mixtures and one
extra shift offline for efficiency measurement with non-noble gas tracers are requested. In
total we therefore request 8 on-line and 3 off-line shifts.
References
1. F. Gröschel et al., CERN-INTC-2003-014 (2003).
2. L. Zanini et al, Volatile Elements Production Rates in a 1.4 GeV Proton-Irradiated Molten
Lead-Bismuth Target, Proceedings of the Int. Conference on Nuclear Data for Science and
Technology, Santa Fe, NM, USA, Sept. 26-Oct 1, 2004.
3. E. Manfrin, private communication, 2004.
4. R. Kirchner, Nucl. Instrum. Methods B 126, 125 (1996).
5. A. Fassò et al., in Proceedings of the Monte Carlo 2000 conference, Lisbon, A. Kling, F.
Barao, M. Nakagawa, L. Tavora, P. Vaz eds., Sprinter-Verlag Berlin, p. 159 (2001).
6
6. L. S. Waters et al., MCNPX Users’s Manual Version 2.4.0, LA-CP-02-408 (2002).
7. F. Atchison and H. Schaal, Orihet 3 – Version 1.12, A guide for users, March 2001.
8. C. Petrovich, SP-FISPACT, A computer code for activation and decay calculations for
intermediate energies. A connection of FISPACT and MCNPX, RT/ERG/2001/10, ENEA,
Bologna (2001).
9. A. Boudard et al., Phys. Rev. C 66, 044615 (2002).
10. T. Enqvist et al., Nucl. Phys. A 686, 481 (2001).
11. J. Neuhausen et al., Radiochimica Acta 92, 917 (2004).
7