ELBE gamma-rays shed light on the origin of the chemical elements
Andreas Wagner, Arnd Junghans, Ronald Schwengner
Heavy chemical elements in the cosmos are produced by fusion reactions of two light atomic nuclei or by neutron capture processes. Neutron capture processes in red giant stars produce about half of all nuclei up to lead. The other half is produced during supernova explosions, which occur when heavy stars end their life cycle. During these cataclysmic events generating explosions brighter than an entire galaxy  rapid neutron capture processes and nuclear disintegration by photons lead to the formation of all elements in our solar system which are heavier than iron. Chemical elements with several stable isotopes serve as probes for our understanding of the cosmic nucleosynthesis by using their abundances as fingerprints of the various production processes. Especially the origin of 35 neutron-deficient isotopes around Molybdenum and Samarium remains mysterious, as they are bypassed by stellar neutron capture processes.
Recently, the question of nuclear disintegration induced by the high-temperature photon field present during supernova explosions has received growing attention. During these explosions, the outer layers of the exploding star are heated up to two till three billion Kelvin extending the thermal photon spectrum beyond the neutron or proton separation energy of heavy nuclei. Under these conditions the nucleosynthesis path proceeds through photodisintegration reactions as shown in fig. 1.
Fig. 1: Material driven outwards by the winds from supernova remnant Cassiopeia A (photo: courtesy of Hubble-STScI). The overlay indicates those nuclei studied at the FZD radiation source ELBE in the path of various nucleosynthesis reactions.
Creating a supernova photon spectrum with ELBE bremsstrahlung
At the FZD radiation source ELBE, the thermal photon bath for temperatures of a few billion Kelvin can be replicated using electron bremsstrahlung. Scattered from a metal foil, the electron beam generates an intensive continuum of gamma-rays similar in energy to the field of gamma-rays in a supernova as shown in fig. 2. The ELBE gamma-rays are then used to study nuclear reactions by experiments which aim at verifying or falsifying the input of nuclear theory for key astrophysical cases. In order to study the rare processes involved in gamma-ray induced cosmic nucleosynthesis, the uniquely high intensity of the ELBE radiation is necessary to get a sufficiently strong measurement signal.
Fig. 2: Thermal distribution of photons at a typical supernova temperature of three billion Kelvin (red line) shown together with the bremsstrahlung at ELBE (blue line) in arbitrary units. The overlay of these two spectra together with a predicted nuclear photo-effect cross section shows that the region of interest can be scanned with bremsstrahlung of different end point energy.
One focus of the ELBE experiments is the role of the giant dipole resonance in photon-induced processes. The giant dipole resonance is the most important collective vibration of atomic nuclei. Sufficiently high photon energies lead to a strong absorption of photons in a characteristic energy interval. The cosmic nucleosynthesis is most sensitive to the low energy side of the giant dipole resonance. Thus photon absorption directly above the energy threshold of the nuclear photo-effect but below the peak of the giant dipole resonance is particularly important. However, cross sections close to the threshold are small and complicated to measure. The detectors at the experimental site at ELBE which are used to measure the decay of irradiated samples have high detection efficiency and are shielded against the background from traces of natural radioactivity . A pneumatic system is used to transport the samples quickly from the irradiation positions to the detector. This way short-lived activity can also be detected with a few seconds of decay half life. From the data gathered at ELBE cross sections of the photon induced emission of neutrons, protons and even helium nuclei have been extracted. The latter were detected as an example of heavy nuclei in the energy region of cosmic nucleosynthesis at ELBE for the first time. The nuclides 92Mo and 144Sm serve as crucial test cases as their abundance predictions are off by more than an order of magnitude compared to the observations.
Photon scattering allows for experimental determination of the dipole strength function. In experiments at the remsstrahlung facility at ELBE  we have determined the dipole strength up to the neutron-separation energy for the first time with good precision by including the contributions from resolved resonances as well as from a continuum of unresolved resonances and by applying statistical methods to correct inelastic processes [3-5]. As a result of this novel technique we have achieved a continuous connection of the dipole strength deduced from photon-scattering to data deduced from photodisintegration reactions . Moreover, studying the chain of stable even-mass molybdenum isotopes from 92Mo to 100Mo has shown that the dipole strength depends on properties of the nuclei, such as nuclear deformation . Studying closed-shell N=50 nuclei like 88Sr  and 90Zr, we have found resonance-like structures on top of the tail of the Giant Dipole resonance. This may influence photodisintegration reaction rates determining the production and destruction of specific nuclei in the cosmos.
Dipole strength functions are also important for understanding the inverse process of radiative neutron capture. In this way, the dipole strength distributions in nuclei obtained from experiments will improve both the modeling of cosmic nucleosynthesis reactions and the determination of cross sections of neutron capture reactions which may be used for transmutation processes. In our experiments, atomic nuclei were made visible through their fluorescence light in the ELBE bremsstrahlung: They shine with a continuous spectrum in the ultra-ultra-ultra violet with a small contribution from sharp resonance lines.
The experiments at ELBE are complemented by measurements with radioactive nuclei which are not accessible otherwise at the Gesellschaft für Schwerionenforschung (GSI) in Darmstadt. The dissociation of nuclei in the electromagnetic field of a heavy target nucleus at high beam energies is used to determine the electromagnetic strength function of nuclei with half-lives down to milliseconds. Conducting these measurements, we collabora/*ted with GSI Darmstadt, the Institute of Nuclear Physics at Forschungszentrum Karlsruhe, and the Institute of Nuclear Physics at Technische Universität Darmstadt. The analysis is still in progress. Collaboration also exists with a group from Konan Universityin Kobe, Japan, which uses Laser Compton backscattered photons to investigate photodisintegration reactions.
Fostering scientific exchange on the questions of explosive nuclear synthesis described here, the Institute of Radiation Physics of the FZD organized the International Europhysics Conference "Nuclear Physics in Astrophysics III" in Dresden in March 2007. The conference highlighted the importance of the input of nuclear physics to astrophysical modeling and initiated many fruitful scientific discussions.
Fig. 3: Ratio of radioactive nuclei produced through bremsstrahlung with different electron energy for different photodisintegration processes on 92Mo (92Mo(gamma,n)91Mo (blue) 92Mo(gamma,p)91Nb (green), and 92Mo(gamma,a)88Zr (red), and the reference process 197Au(gamma,n)196Au). The lines show the predictions of two different models for comparison.
Fig. 4: Upper panel: Experimental energy distribution of scattered photons for the Molybdenum nuclides 98Mo (red) and 100Mo (blue). The simulated background dominates the spectrum above the threshold of neutron production (indicated by arrows). Lower panel: Experimental photo-absorption cross section from photon scattering data (full symbols) and photo-production of neutrons (open symbols). The solid line indicates the parameterization of the Giant Dipole Resonance by Lorentzian distributions.
 The physics of core-collapse supernovae, S. Woosley and T. Janka, Nature Physics 1, 147 (2006).
 The photon-scattering facility at the superconducting electron accelerator ELBE, R. Schwengner, R. Beyer, F. Dönau, 1E. Grosse, A. Hartmann, A.R. Junghans, S. Mallion, G. Rusev, K.D. Schilling, W. Schulze and A. Wagner, Nucl. Instr. and Meth. Phys. Res. Section A 555, 211 - 219 (2005).
 Systematics of magnetic dipole strength in the stable even-mass Mo isotopes, G. Rusev, R. Schwengner, F. Dönau, M. Erhard, S. Frauendorf, 1E. Grosse, A.R. Junghans, L. Käubler, K. Kosev, L.K. Kostov, S. Mallion, K.D. Schilling, A. Wagner, 4H. von Garrel, 4U. Kneissl, 4C. Kohstall, 4M. Kreutz, 4H.H. Pitz, 4M. Scheck, 4F.Stedile, 3P. von Brentano, 3J. Jolie, 3A. Linnemann, 3N. Pietralla and 3V. Werner, Physical Review C 73, 44308 (2006).
 Pygmy dipole strength close to the particle-separation energies, G. Rusev, 1E. Grosse, M. Erhard, A.R. Junghans, K. Kosev, K.D. Schilling, R. Schwengner and A. Wagner, European Physical Journal A 27 s01, 171 (2006).
 Dipole-strength distributions up to the particle-separation energies and photodissociation of Mo isotopes, R. Schwengner, N. Benouaret, R. Beyer, F. Dönau, M. Erhard, S. Frauendorf, 1E. Grosse, A.R. Junghans, J. Klug, K. Kosev, C. Nair, N. Nankov, G. Rusev, K.D. Schilling and A. Wagner, Nuclear Physics A 788, 331c (2007).
 Low-energy tail of the giant dipole resonance in 98Mo and 100Mo deduced from photon-scattering experiments, G. Rusev, R. Schwengner, F. Dönau, M. Erhard, 1E. Grosse, A.R. Junghans, K. Kosev, N. Kostov, K.D. Schilling, A. Wagner, 2F. Becvar and 2M. Krticka, submitted to Physical Review C.
 Effect of nuclear deformation on the electric-dipole strength in the particle-emission threshold region, F. Dönau, G. Rusev, R. Schwengner, A.R. Junghans, K.D. Schilling and A. Wagner, Physical Review C, in print.
 Dipole response of 88Sr up to the neutron-separation energy, R. Schwengner, G. Rusev, N. Benouaret, R. Beyer, M. Erhard, 1E. Grosse, A.R. Junghans, J. Klug, K. Kosev, N. Kostov, C. Nair, N. Nankov, K.D. Schilling and A. Wagner, Physical Review C, in print.
1Institute of Nuclear and Particle Physics, Technische Universität Dresden, Germany
2Charles University in Prague, Czech Republic
3University of Cologne, Germany
4Universität Stuttgart, Germany