Nuclear Astrophysics with Rare Isotopes

Nuclear Astrophysics with Rare Isotopes

Nuclear Physics A 827 (2009) 26c–33c Nuclear Astrophysics with Rare Isotopes H. Schatz National Superconducting Cyc...

176KB Sizes 0 Downloads 18 Views

Nuclear Physics A 827 (2009) 26c–33c

Nuclear Astrophysics with Rare Isotopes H. Schatz National Superconducting Cyclotron Laboratory, Dept, of Physics and Astronomy, Joint Institute for Nuclear Astrophysics, Michigan State University, East Lansing, MI 48824, USA

Abstract A summary of the importance of rare isotopes in the understanding of the origin of the elements and accreting neutron stars is presented.

Key words: rp-process, r-process, neutron star PACS: 26.30.-k, 26.30.Hj, 26.50.+x

1. Introduction Rare isotopes play an important role in the nuclear processes occuring naturally in the cosmos [1]. They are a stepping stone for the synthesis of heavy elements in the rapid neutron capture process (r-process), where extremely neutron rich rare isotopes are produced by high fluxes of neutrons. High fluxes of protons, on the other hand, produce extremely proton rich isotopes (or rather isotones) in X-ray bursts and, to some extent, in classical Novae. Finally, extremely neutron rich rare isotopes also exist permanently in the crusts of neutron stars, where their decays are Pauli blocked by degenerate electrons at very high density. The properties of the rare isotopes in these scenarios need to be understood in order to address open questions concerning the origin of elements, nuclear energy generation in stars and stellar explosions, and the nature of neutron stars. The nuclear physics connects observational signatures such as the composition of stellar ejecta, or the characteristics of the nuclear energy generation with the features of astrophysical models and the parameters of the astrophysical system. Understanding of the nuclear physics therefore enables to predict observational signatures, and to interpret observations in terms of model constraints and system parameters. 0375-9474/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.nuclphysa.2009.05.015

H. Schatz / Nuclear Physics A 827 (2009) 26c–33c


2. The r-process The r-process is thought to be responsible for the origin of about 40% of the abundance of the 54 heavy elements beyond ≈Ge. Typically, all the abundance that cannot be produced by the s- or p-process is attributed to the r-process, though there might be contributions from other processes (see below). The site of the r-process is still unknown. Unlike the s-process, so far no observation has linked an observed elemental or isotopic r-only abundance to a specific astrophysical event. The only exception is the detection of 244 Pu [2], an r-process isotope, in deep sea sediments that also show 60 Fe thought to be deposited by a nearby supernova 2.8 Mio years ago [3]. However, the detection is tentative and the correlation in time is very uncertain. A major source of information about the r-process are the abundances found in very metal poor stars with metallicity [Fe/H] (logarithmic ratio of Fe/H to solar Fe/H) typically around −3 (see [4] for a recent review and more references). Unlike the s-process, the r-process is a primary process that generates its own seeds. Therefore, as one goes back in time, one would expect to encounter an early iron poor epoch of galactic chemical evolution where the heavy element composition is, at least locally, dominated by the r-process, with the typical s-process emerging at a later stage when enough iron seed nuclei have been generated. Indeed, a number of such strongly r-process enhanced metal poor stars have been identified. Their composition is believed to provide information of the abundances generated by at most a few r-process events. Three major conclusions have been drawn based on these observations: (1) The excellent agreement of the elemental abundance pattern between Ba and Au with the pattern of the solar abundances attributed to the r-process indicates a robust r-process that in each event produces a very similar composition. (2) The somewhat lower abundances of lighter r-process elements, in particular Sr, Y, and Zr seems to indicate the need of an additional process to contribute to the origin of these elements. (3) The observed abundances of radioactive U and Th relative to the stable isotopes vary significantly from star to star. These variations are much larger than what could be explained by radioactive decay and age variations. Therefore, the abundances of U and Th seem to vary from event to event. Especially the possible existence of an additional, hitherto unknown nucleosynthesis process (or an unknown variant of the known ones) contributing to the solar abundances has generated a lot of excitement. Some have suggested that a second r-process component, a weak r-process in analogy to the different s-process components, is responsible [5] The conclusions depend to some extent on the s-process model used, as they rely on a comparison of the observed r-process abundance pattern in metal poor stars, and the solar abundances attributed to the r-process. The latter is obtained by subtracting the s-process contribution (and a minor p-process contribution) from the solar abundances. Montes et al. [6] use the s-process model of Travaglio et al. [7] and suggest that the so called Light Element Primary Process (LEPP) could explain the missing contribution to the solar abundances. The LEPP had been identified before as a new process needed to explain the origin of some of the Y, Sr, and Zr abundances observed in some metal poor stars by Travaglio et al. [7] and by Qian and Wasserburg [8] (with somewhat different terminology). However, the nature of this new LEPP process is still debated. One possibility is the neutrino-p process in the proton rich neutron driven winds off proton neutron stars in core collapse supernovae [9,10]. Qian and Wasserburg [8] propose the


H. Schatz / Nuclear Physics A 827 (2009) 26c–33c

charged particle process that occurs in the freezeout of a neutrino driven wind and is thought to produce the seeds for an r-process. They propose that in some cases this process is not followed by an r-process resulting in a ”LEPP” event. Farouqi et al. [11] followup on this suggestion and present detailed calculations of the charged particle induced nucleosynthesis in neutrino driven winds. They find that for low entropies indeed the r-process seed generating charged particle processes are not followed by a strong rprocess due to the very low neutron to seed ratio and consequently produce a LEPP like abundance pattern, though some large differences with observations remain unexplained. Alternatively, Pignatari et al. [12] suggest a strong primary s-process occurring at very low metallicities. This would be in line with the calculations of Montes et al. [6] which favor an s-like process over a r-like process for the LEPP. The question of the nature of the LEPP is difficult to address, as observations provide, with a few exceptions, mostly elemental abundance information, and as the r-process peak region around A=130 cannot be observed because the relevant elements don’t have suitable lines for absorption spectroscopy. Consequently it is not clear whether the LEPP exhibits an r-process peak or not. The LEPP abundance patterns that can be extracted from s- and r-process poor stars are therefore rather ”generic”. On the theoretical side, it is clear that the r-process requires extreme neutron densities to generate the A=130 and A=195 abundance peaks, which are created by crossing the N=82 and N=126 neutron shell closures far from stability [13–15]. So far it has been a challenge to identify a scenario where sufficiently large neutron densities are generated. The most frequently discussed potential r-process site is the neutrino driven wind off a proto neutron star formed in a core collapse supernova (see [16] for a recent review and more references). The neutron rich material of the proto neutron star generates neutron fluxes that drive the out flowing matter neutron rich. The relatively low density (high entropy) in the ”bubble” formed between the proto neutron star and the stellar material ejected in the supernova prevents the formation of a lot of seed nuclei via charged particle reactions leading to a high neutron to seed ratio. However, so far, the conditions needed for a full r-process have not been obtained self consistently in such a scenario. For reasonable wind speeds and neutron excesses, entropies are typically too low to produce the heaviest r-process elements. Some have pointed out that a relatively massive protoneutron star of about 2 solar masses would lead to more rapid outflows with reduced seed production and therefore an increased neutron to seed ratio (for example [17]). Alternative scenarios are also explored. Fryer et al. [18] investigate the nucleosynthesis in fall back material in core collapse supernovae. Some of this material can get ejected and the overall trajectory and neutrino exposure seem to enable an r-process. Another possible r-process site is the neutron rich matter ejected in neutron star merger events. If the matter is heated the r-process is similar to the neutrino driven wind scenario though neutron excesses are much larger [19]. If the ejected matter is not heated, a cold decompression of neutron rich matter can also lead to an r-process of a very different nature [20]. A major difficulty of the neutron star merger scenario is the merger rate, which seems to be too low to be consistent with the steady enrichment of r-process elements over time inferred from observations of r-process abundances in metal poor stars [21]. However, given the simplified galactic chemical evolution models used so far and the uncertainties of merger rates and their statistical distribution one cannot exclude this scenario with certainty. Other r-process scenarios include gamma ray burst accretion disks irradiated with an intense neutrino flux [22], outflows from black hole - neutron

H. Schatz / Nuclear Physics A 827 (2009) 26c–33c


star mergers [23], or shocked surface layers of O-Ne-Mg cores [24]. Clearly a lot of theoretical problems need to be solved to arrive at a viable r-process model. However, in the end it is observations and experiments that need to decide which is the correct model. In principle this is possible by comparing the characteristic abundance signature of a specific r-process model with the large body of precision observations of r-process abundance patterns [25]. For example, in some models neutrino irradiation might create characteristic signatures [26]. This requires however accurate knowledge of the underlying nuclear physics that determines the abundance signature for a given set of astrophysical conditions [27]. Most r-process models are characterized by high neutron density and temperature leading to the establishment of at least a partial (n,γ)-(γ,n) equilibrium. In this case, the maximum abundance in an isotopic chain for a given neutron density and temperature is entirely determined by the Saha equation and therefore by neutron separation energies and, to a lesser extent, nuclear partition functions. The r-process proceeds then from isotopic chain to isotopic chain via the β-decay of the this abundant isotope. The β-decay rates tend to be by far slower than the neutron capture and (γ,n) rates and therefore determine the abundance pattern with slow decay rates leading to higher abundances and faster decay rates to lower abundances. Therefore, nuclear masses and β-decay rates tend to be the most important nuclear physics quantities in r-process models. The region around the first major waiting point at the neutron shell closure N = 50, 80 Zn, is best understood, though it is only relevant for the subset of r-process models where there is a neutron capture reaction flow in this region. For example, in neutrino driven wind models with standard parameters the region is bypassed by charged particle reactions (though those models do not successfully produce all r-process nuclei). With the half-life measurement of 78 Ni [28] the half-lives of all major N = 50 waiting points are known experimentally. In addition, this is the mass region where precision Penning Trap mass measurements have finally reached the r-process with mass measurements of the Zn isotopes out to 81 Zn [29,30]. This allows one for the first time to reliably constrain the r-process conditions needed to make 80 Zn a waiting point and to produce the associated A = 80 abundance hump in the observed abundances [29]. Between N = 50 and N = 82 half-life measurements at NSCL have reached the edge of the r-process path [31,32]. In the Zr region, experiments are approaching the 110 Zr region [32], where a predicted possible subshell closure might affect the r-process. The measured unexpectedly large branching for β-delayed neutron emission of 104 Y together with the relatively long β-decay half-lives seem to indicate a lower than expected deformation of the Zr daughters, which might be a hint for a subshell closure at 110 Zr. At the second major waiting point region, the N = 82 shell closure, information is more sparse, though some of the most critical quantities have been measured. In particular, the half-life and more recently the mass of 130 Cd, the major N = 82 waiting point, are known experimentally [33]. However, masses beyond 130 Cd needed to fully constrain this waiting point, and the other N = 82 half-lives beyond 130 Cd and 129 Ag are still unknown. An important question in this context is the possible quenching of the N = 82 shell gap far from stability. It has been shown that the influence of such a shell quenching on nuclear masses removes some of the anomalies in the mass surface caused by shape changes as one approaches the N = 82 shell closure very far from stability (around Zr). This leads to a much better fit of the solar r-process abundances in r-process models [34]. First hints for such a shell quenching effect have already been reported for the 130 Cd


H. Schatz / Nuclear Physics A 827 (2009) 26c–33c

region, based on the systematics of the energy of the first excited 2+ states E(2+ ) of even-even neutron rich Cd isotopes out to 128 Cd [35], and on the measured β decay Q-value of 130 Cd, which can be better reproduced with mass models that include shell quenching [33]. On the other hand, there are alternative theoretical explanations for the anomalous E(2+ ) of 128 Cd that do not require shell quenching [36] and other evidence suggests a strong N = 82 shell gap in the 130 Cd region [37,38]. However, the question is not whether the shell gap at 130 Cd is strong. It is clear that it is strong, and this must be so from an astrophysical point of view to explain the A = 130 abundance peak in the r-process abundance distribution. The question is rather whether a trend of a reduction in shell gap size towards neutron rich nuclei is already noticeable at 130 Cd. Such a small change in shell gap size might manifest itself in some observables but not in others. The different conclusions drawn from different observables therefore do not necessarily contradict each other. However, it is clear that the evidence is ambiguous and that measurements further away from stability are needed to establish or reject the notion of a quenching of the N = 82 shell far from stability where it would affect the r-process. Much less is known about the nuclear physics of the r-process beyond N = 82. Some progress has recently been made in measuring decay properties [39] and the structure of excited states of nuclei towards [40] the r-process path at N = 126. Besides masses and β-decay properties (half-lives and branchings for β delayed neutron emission) other nuclear physics is needed in r-process models depending on the particular model. Recently it has been shown that uncertainties in neutron capture rates near 132 Sn can affect r-process abundance patterns [41] and first measurements using (d,p) transfer reactions have been performed at ORNL [42]. In some fast expanding models, neutron capture on light neutron rich isotopes, such as 17 C and others become important [17,43,25]. Charged particle reaction rates such as α(αn, γ)9 Be or α(t,γ) 7 Li play a critical role in determining the neutron to seed ratio. In r-rprocess models with very high neutron to seed ratios fission processes become part of the r-process [44]. In the extreme, fission cycling can occur [45], where the fission products become again seeds for the r-process that again transforms them into nuclei that undergo fission. Fission cycling has been suggested as a mechanism to stabilize the r-process abundance pattern and to explain the robustness of the pattern observed in metal poor stars. In this case, fission barriers and fission product distributions are needed and first calculations with a complete set of nuclear physics have been performed recently [44]. Finally, depending on the environment, neutrino interactions might influence the r-process and neutrino-nucleus interaction cross sections need to be known to model this effect [26]. Clearly, experiments have only started to scratch at the surface of r-process nuclear physics. Addressing the nuclear physics questions of the r-process will therefore be a major thrust in next generation rare isotope facilities such as RIBF, FAIR, and FRIB. 3. X-ray bursts and the rp-process X-ray bursts are the most frequent thermonuclear explosions in the universe we know of (see [46,47] for recent reviews). They occur on neutron stars that accrete helium and/or hydrogen rich matter from a companion star. The layer of fuel accumulating on the neutron star surface explodes with recurrence times of the order of hours to days, releasing about 1039 -1040 ergs of energy in bursts lasting 10-100 s. There are many open

H. Schatz / Nuclear Physics A 827 (2009) 26c–33c


questions related to X-ray bursts and advances in astrophysics and nuclear physics are needed to address them. Examples include the burst behavior as a function of accretion rate or the occurrence of doubly peaked bursts in some sources [48,49]. The energy released in nuclear reactions during X-ray bursts directly powers the observed burst light curve, and the nuclear reactions determine the time dependence of the energy release and therefore the burst shape, at least for the frequently observed long bursts with time scales of around 100 s [50,51,48,52]. While most of the burned matter remains on the neutron star surface, a small fraction of the freshly synthesized nuclei can be ejected in winds off particularly powerful bursts generating potential observational signatures that could be looked for with advanced X-ray observatories [53]. The burst ashes that remains on the neutron star surface gets incorporated into the curst by the ongoing accretion, where it affects the nature of electron capture reactions that determine the thermal structure of the neutron star and are connected to other observables such as the rare superbursts [54–56] and observations of cooling neutron stars in transients [57,58]. Gupta et al. [56] demonstrated the sensitivity of heat generating electron capture reactions to the initial composition set by X-ray bursts. Recently it has been shown that neutron emission leads to new electron capture pathways in neutron star crusts that reduce the sensitivity of the deeper lying reactions to the initial composition [59]. In the following we limit the discussion of the nuclear physics to the subset of bursters powered by explosive mixed hydrogen and helium burning. In this case, the 3α reaction followed by an αp process produces the seeds for hydrogen burning via the rapid proton capture process. (see [60] for a recent detailed analysis of the reaction sequences). The endpoint of the rp-process depends sensitively on model and parameters, and is typically set by the number of proton capture reactions needed to exhaust the hydrogen [61]. It can therefore vary from system to system, or even from burst to burst. Under extreme conditions with a large hydrogen to seed ratio the rp-process can reach a Sn-Sb-Te cycle that forms a natural endpoint of the rp-process in X-ray bursts [51]. The important nuclear physics for hydrogen burning in X-ray bursts are the β-decay rates, masses, and nuclear reaction rates of the neutron deficient rare isotopes along the path of the αp and rp-processes (see Fig. 1 in [47]). With the recent measurement of the half-life of 96 Cd [62] and the remeasurement of the half-life of 84 Mo [63], both enabled by the new RF Fragment Separator at NSCL, all important half-lives along the rp-process path are now determined experimentally. Major progress has been made with mass measurements, where now Penning traps have reached critical regions of the rp-process path (see [64,65] for the importance of masses in rp-process calculations). A range of measurements in the critical N ≈ Z A = 64 − 72 range, where the most important rp-process waiting points are located [66–69] have together with Coulomb shift calculations that allow the calculation of mirror nuclei masses [70], dramatically improved the accuracy of burst calculations, though some uncertainties owing to unknown masses of very neutron deficient isotopes remain. Masses in the region of the SnSbTe cycle have recently been determined through α-decay measurements [71] . Other new Penning Trap mass measurements have pushed the border of known masses much closer to the rpprocess path, where all unknown masses can now be determined through extrapolations as opposed to global mass models [72]. However, major uncertainties remain due to the largely unknown rp-process reaction rates [73,74]. Owing to the typically low rare isotope beam intensities available today, mostly indirect techniques are employed. Rare isotope beam facilities such as ANL, CRIB,


H. Schatz / Nuclear Physics A 827 (2009) 26c–33c

NSCL, ORNL, and TRIUMF have played an important role in this area (see [47] for a broader review). At the NSCL we use (p,d) transfer reactions far from stability to populate levels in critical nuclei along the path of the rp-process [75,76]. In addition, a number of critical reactions, for example in the CNO region, are within reach of stable beam facilities. An example is the 15 O(α,γ) reaction, which has been shown to be critical in determining overall burst properties due to its role in regulating hydrogen burning via the CNO cycle prior to burst ignition [77,78]. A number of experiments at KVI [79] , ANL [80], and Notre Dame [81] have been performed that populate the critical resonances and constrain the small α-branching in states in 19 Ne that determines the reaction rate, with the Notre Dame group reporting an actual detection at the one sigma level. Clearly new capabilities at stable beam and at next generation rare isotope facilities are needed to transition the field into ”production mode”, where many of the necessary measurements can finally be performed. Until then, there are a number of promising developments at existing facilities that will continue to provide cutting edge results. This includes the TRIUMF/ISAC facility, where a number of new beams of astrophysical importance are being developed and first direct measurements of reaction rates in the rp-process have been performed. The NSCL is developing a low energy reaccelerator facility ReA3 based on the gas stopping of fast beams produced by in-flight fragmentation. This work is supported by the Joint Institute for Nuclear Astrophysics under NSF-PFC grants PHY 02-16723, and NSF PHY 08-22648 as well as by NSF grant PHY 06-06007. References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28]

H. Schatz, Physics Today November 2008, 40. C. Wallner et al., New Astr, Rev. 48 (2004) 145. K. Knie et al., Phys. Rev. Lett. 93 (2004) 171103. C. Sneden, J. J. Cowan, and R. Gallino, Annu. Rev. Astro. Astrophys. 46 (2008) 241. B. Pfeiffer, U. Ott, and K.-L. Kratz, Nucl. Phys. A 688 (2001) 575. F. Montes et al., Ap. J. 671 (2007) 1685. C. Travaglio et al., Ap. J. 601 (2004) 864. Y.-Z. Qian, and G.J. Wasserburg, Phys. Rep. 442 (2007) 237. C. Fr¨ ohlich et al., Phys. Rev. Lett. 96 (2006) 142502. J. Pruet et al., Ap. J. 644 (2006) 1028. K. Farouqi et al., Ap. J. 694 (2009) L49. M. Pignatari et al., Ap. J. 687 (2008) L95. M. Arnould, S. Goriely, and K. Takahashi, Phys. Rep. 450 (2007) 97. Y. Z. Qian, Prog. Part. Nucl. Phys. 50 (2003) 153. J.J. Cowan, F.-K. Thielemann, and J.W. Truran, Phys. Rep. 208 (1991) 267. I.V. Panov and H.-Th. Janka, Astr. Astrophys. 494 (2009) 829. T. Terasawa et al., Ap. J. 562 (2001) 470. C. Fryer et al., Ap. J. 646 (2006) L131. S. Rosswog et al., Astron. Astr. 341 (1999) 499. S. Goriely, et al., Nucl. Phys. A 758 (2005) 587c. D. Argast et al., Astron. Astrophys. 416 (2004) 997. R. Surman and G. C. McLaughlin, Ap. J. 618 (2005) 397. R. Surman, Ap. J. 679 (2008) L117. H. Ning et al., Ap. J. 667 (2007) L159. K. Otsuki et al., AIP Conf. Proc. 847 (2006) 227. Y.-Z. Qian et al., Phys. Rev. C 55 (1997) 1532. K.-L. Kratz et al., Ap. J. 403 (1993) 216. P. Hosmer et al., Phys. Rev. Lett. 94 (2005) 112501.

H. Schatz / Nuclear Physics A 827 (2009) 26c–33c

[29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47] [48] [49] [50] [51] [52] [53] [54] [55] [56] [57] [58] [59] [60] [61] [62] [63] [64] [65] [66] [67] [68] [69] [70] [71] [72] [73] [74] [75] [76] [77] [78] [79] [80] [81]


S. Baruah et al., Phys. Rev. Lett. 101 (2008) 262501. J. Hakala et al., Phys. Rev. Lett. 101 (2008) 052502. F. Montes et al., Phys. Rev. C 73 (2006) 035801. J. Pereira et al., Phys. Rev. C 79 (2009) 035806 I. Dillmann et al., Phys. Rev. Lett. 91 (2003) 2503. B. Chen et al., Phys. Lett. B 355 (1995) 37. T. Kautzsch et al., Eur. Phys. J. A9 (2000) 201. T. R. Rodriguez, Phys. Lett. B 668 (2008) 410. A. Jungclaus et al., Phys. Rev. Lett. 99 (2007) 13250. W.B. Walters et al., Phys. Rev. C 70 (2004) 034314. T. Kurtukian-Nieto et al., AIP Conf Proc. 802 (2005) 73. S. J. Steer et al., Phys. Rev. C 76 (2008) 061302. J. Beun et al., J. Phys. G: Nucl. Part. Phys. 36 (2009) 025201. K. Jones et al., AIP Conf Proc. 1098 (2009) 153. T. Sasaqui et al., Ap. J. 634 (2005) 1173. I. Petermann et al., arXiv:0812.0968v1. P.A. Seeger, W.A. Fowler, and D.D. Clayton, Ap. J. Suppl. 11 (1965) 121. T.E. Strohmayer and L. Bildsten, Compact Stellar X-ray Sources, ed. W. H. G. Lewin and M. van der Klies (Cambridge: Cambridge Univ. Press), arXiv:astro-ph/0301544 (2003). H. Schatz and K.E. Rehm, Nucl. Phys. A 777 (2006) 601. J.L. Fisker, F.-K. Thielemann, M. Wiescher, Ap. J. 608 (2004) L61. A. Watts and I. Maurer, Astron. Astrophys. 467 (2007) L33. O. Koike et al., Astron. Astrophys. 342 (1999) 464. H. Schatz et al., Phys. Rev. Lett. 86 (2001) 3471. S. E. Woosley et al., Ap. J. Suppl. 151 (2004) 75. N.N. Weinberg, L. Bildsten, and H. Schatz, Ap. J. 639 (2006) 1018. A. Cumming et al., Ap. J. 646 (2006) 429. P. Haensel and J.L. Zdunik, Astron. Astrophys. 480 (2008) 459. S. Gupta et al., Ap. J. 662 (2007) 1188. R. E. Rutledge et al., Ap. J. 577 (2002) 346. E.M. Cackett et al., M.N.R.A.S. 372 (2006) 479. S. Gupta et al., Phys. Rev. Lett. 101 (2008) 231101. J.L. Fisker, H. Schatz, and F.-K. Thielemann, Ap. J. Suppl. 174 (2008) 261. H. Schatz et al., Phys. Rep. 294 (1998) 167. D. Bazin et al., Phys. Rev. Lett. 101 (2008) 252501. J. B. Stoker et al., Phys. Rev. C 79 (2009) 015803. H. Schatz, Int. Journ. of Mass Spectr. 251 (2006) 293. A. Parikh et al., arXiv:0811.0727 (2008). J. A. Clark et al., Phys. Rev. Lett. 92 (2004) 192501 D. Rodriguez et al., Phys. Rev. Lett. 93 (2004) 161104 P. Schury et al., Phys. Rev. C 75 (2007) 055801. J. Savory et al., Phys. Rev. Lett. 102 (2009) 132501. B. A. Brown et al., Phys. Rev. C 65 (2002) 045802. C. Mazzocchi et al., Phys. Rev. Lett. 98 (2007) 212501. C. Weber et al., Phys. Rev. C 78 (2008) 054310. A. Parikh et al., Ap. J. Suppl. 178 (2008) 110. K. Smith et al., Proceedings of Science, NIC X (2009) 178. R. R. C. Clement et al., Phys. Rev. Lett. 92 (2004) 172502. D. Galaviz et al., Proceedings of Science, NIC IX (2006) P. 99.100. J.L. Fisker, J. G¨ orres, M. Wiescher, B. Davids, Ap. J. 650 (2006) 332. R.L. Cooper, R. Narayan, Ap. J. 648 (2006) L123. B. Davids et al., Phys. Rev. C 67 (2003) 065808. K.E. Rehm et al., Phys. Rev. C 67 (2003) 065809. W.P. Tan et al., Phys. Rev. Lett. 98 (2007) 242503.