The Science and Operation of
Gammasphere at ATLAS

Argonne National Laboratory, December 1995

Click here to see document pictures

Contents Page

I.A. Introduction

Gammasphere is a spectrometer of unparalleled sensitivity to nuclear electromagnetic radiation due to its high resolution, granularity and efficiency. This powerful combination of features makes it the ideal device for studying nuclear spectroscopy of rare and exotic processes. Of particular current interest are studies at the very limits of nuclear stability, due to high angular momentum, unusual neutron-to-proton ratio, high mass, and high internal excitation energy. Most of the research during the initial phase of Gammasphere operation has been focused on high angular momentum physics where great progress has been made in establishing a method of correctly assigning quantum numbers to superdeformed states, in finding candidates for hyperdeformation, and in conducting detailed spectroscopy in the second well. This excellent high-spin spectroscopy has been largely based on measurements of gamma rays alone and much progress has been made with Gammasphere as a freestanding device.

To achieve similar progress in the study of nuclei far from the valley of stability and in heavy nuclei Gammasphere needs to be externally triggered by a device which can select the reaction channels of interest. The world's pre-eminent device for channel selection following fusion reactions is the Fragment Mass Analyzer (FMA) at Argonne National Laboratory. Thus, when we combine the power of these two devices it will revolutionize nuclear spectroscopy far from stability and in the heaviest nuclei. We feel that the key physics opportunities at the moment arise from exploitation of the Recoil Decay Tagging (RDT) method, which is described in the technical section of this document. This technique allows prompt g-radiation from a nuclear reaction to be correlated with the radioactive decay of the residues which are produced. In moving Gammasphere, several other classes of experiments involving the high-energy heavy beams and the superb timing of the ATLAS accelerator will become possible. Further, for high-spin experiments, all the beam species which are used in the current LBNL program will still be available.

The broad topic of nuclear spectroscopy near the driplines has been widely discussed in our community due to the interest in future opportunities with radioactive beams. One can find extensive discussions of the interesting science in the proposals for radioactive beam facilities like the Isospin Laboratory (Is91,Is94), Oak Ridge (Or91), Ganil (Sp94), MSU (Ms94), Argonne (Ar95a) and in long-range plans in the USA (Ns95) and Europe (Nu93). However, progress can be made in some of these research areas using intense (> 10 pna) stable beams and a highly channel-selective detector like the proposed combination of Gammasphere and the FMA. This combination is especially applicable to studying nuclei along the proton dripline. These nuclei have been shown to be accessible using stable beams and the FMA, as groundstate proton radioactivities have been isolated all along the dripline from antimony (Z = 51) to bismuth (Z = 83) in recent experiments at ANL and elsewhere. The exciting opportunity is to use these decays as an ultra-sensitive trigger for Gammasphere and study the spectroscopy of nuclei at and beyond the dripline. During the fall of this year we have used this method at ANL with the FMA and the composite AYE-Ball array for preliminary investigations of nuclei at the dripline. While still short in photon detection efficiency, these experiments have demonstrated the enormous potential of the method.

To gain a sense of the importance of highly selective external triggering for reaching the most neutron deficient nuclei, some numerical considerations are useful. In the high spin domain, groundstate superdeformation was studied near the limit of the capabilities of first generation spectrometers, as it was a 1-2% process in reaction channels which had total cross sections of about 200 mb, i.e. at a production cross section level of a few mb. Gammasphere, as a freestanding device, has allowed great progress to be made in finding excited superdeformed bands at the level of a few percent of the population of the groundstate band, i.e. at the level of about 100 microbarns. The direct decay-out of superdeformed bands is at a similar level of cross-section. In contrast, in studies of nuclei far from stability using first generation arrays triggered by electromagnetic separators, detailed spectroscopy was already possible at the 100�s of microbarns and some important measurements were made at the 10�s of microbarns level. With Gammasphere and the FMA the measurements should advance down into the 100�s of nanobarns. With this level of sensitivity in-beam studies of nuclear structure beyond the proton dripline should be straightforward, measurements of Z ~ 102 nuclei become feasible, and the N = Z line can be crossed in many places up to the mass 80 region.

The combination of Gammasphere and the FMA will be unique in the world in 1997 and will widen the scope of Gammasphere physics far beyond that possible with similar detectors in Europe. At present, the 40 detector Ga.Sp array can be coupled to the Legnaro Recoil Separator CAMEL, but the solid angle of the separator is sufficiently small that the device is not competitive. The next generation of European array, at Legnaro, will be freestanding, so it will not be competitive in the areas of research we wish to pursue.

In the following paragraphs we will briefly discuss some of the major physics opportunities we see from installing Gammasphere at the FMA and using the ATLAS accelerator. We have arranged the many diverse topics which may be studied with Gammasphere and the FMA in order of increasing mass. We have deliberately omitted a detailed discussion of high-spin physics although all the necessary beams are available, as this field is covered by the present LBNL program and we wish to emphasize the new opportunities offered at ANL. In the following section, (I.C.) we present other important topics which can exploit the unique accelerator capabilities of ATLAS.

I.B. Research with Gammasphere and the FMA

I.B.1. Light Nuclei

Light nuclei, in the p- and sd-shells, are still the fundamental testing ground for many aspects of our understanding of nuclei, both because the shell model theory is most complete, and because there is an extensive and diverse basis of experimental information. It is an ideal place to evaluate and understand microscopic-macroscopic, SU3, cluster, Hartree-Fock and other models and compare them to the fully microscopic shell model calculations and to data. For these light nuclei, the experimental situation is far from complete, and many rigorous tests of nuclear models remain, especially in the high-spin and high-excitation regime where gamma-ray and particle emission are in competition. One of the key issues is one of truncation; does the shell model space become exhausted in fully aligned configurations, or do intruder configurations from higher shells become important in sustaining collectivity? Band terminations have recently been found in many places in the periodic table, but only here is there a self consistent microscopic theory which allows a full interpretation. One classic case is the high-spin regime in 24Mg, where there have been shell model calculations of high spin states for many years, but no individual state above spin J = 8 has been unambiguously measured. The high spin states in question are above particle emission thresholds, but it is known that some gamma competition remains (Sm86), especially between collective states with similar wavefunctions. The study of this category of states at the limit of stability due to excitation energy and angular momentum is closely related to many dripline issues, and their access should be possible with the combination of Gammasphere and the FMA.

Substantial advances have been recently made in studying states in light nuclei through the measurement of correlated alpha emission from levels (Wu92). Evidence has been found for elongated �chain states� of alpha particles which break up into their constituents. Some of the more compact configurations, like those with 3:1 axis ratio (hyperdeformed), are expected to decay both by alpha breakup and by gamma emission, and it is important to establish whether very elongated configurations can relax back into normal shapes through electromagnetic decay. The radiative branches, although weak, could be selected with great sensitivity by detecting the residue in the FMA and the de- exciting gamma cascade in Gammasphere. In figure 1 we show the results of a trial experiment using the AYE-Ball array in which excited states in 24Mg were populated in the 12C(16O,a)24Mg reaction. In a few hours all the known high spin states (J > 4) in 24Mg were seen and transitions not reported previously were observed.

The measurements of chain states were made using large arrays of Si-strip detectors (DSSD�s) at ANL. Combining technologies to detect a-decays from the compound nucleus (in DSSD�s), the recoiling 24Mg nuclei (in the FMA), and the de-exciting g-rays in Gammasphere should reveal extremely weak radiative branches, even when breakup decay is dominant. The two-body nature of the 12C(16O,a)24Mg and similar reactions allows the excitation energy of the gamma-emitting state to be determined. Accurately measuring both the alpha particle energy and the 24Mg time-of-flight will make the technique ultra-sensitive. These measurements are important, as it is the radiative properties of the states which rigorously determine the deformation involved.

I.B.2. The N = Z Line, the Demise of Isospin Purity and the r-p Process

Between the spherical shell closures for neutrons and protons at N = Z = 28 and N = Z = 50 lies a region of transitional and deformed nuclei which pose some of the most stringent challenges to our understanding of nuclei. In this region the shell model space is large enough that very large collective deformation can develop, which is modulated by the occupancy of individual orbits near the Fermi surface. This leads to rapid shape changes with particle number, and even in a single nucleus, to configuration dependent shapes. The level density is low, so pairing is rather unimportant. The big changes in shape can be measured by changes in the pattern of emitted electromagnetic radiation, allowing a rather direct comparison to be made between experimental observables and the nuclear wavefunction. The nuclei where these effects are most apparent are those with equal numbers of protons and neutrons, where symmetry is greatest and the polarizing influences act in concert. These N = Z nuclei all lie quite far from stability and are difficult to study, as production cross sections are low. At present, only a few states are known in these nuclei, and no detailed spectroscopic measurements have been made. However, extrapolating from the experiments which allowed the first state to be seen in these nuclei (Li87,Li90), the combination of Gammasphere and the FMA should permit great progress. The examination of odd-odd nuclei is of particular interest for extracting residual interactions and correlations between neutrons and protons in these orbits (Wa95b).

Apart from their structural interest, these N = Z nuclei are an ideal laboratory for studying the demise of the isospin purity of states. Along the valley of stability isospin purity is approximately maintained by the growing excess of neutrons. This is not the case along the N = Z line and a rapid decline in the purity of states has been predicted (Bo69,Co95). This demise and its consequences have not yet been explored, again because of difficulties in reaching these nuclei. However, with the proposed device, considerable progress can be anticipated through measuring isospin forbidden electromagnetic decays which can only occur through isospin mixing (Wa69,En91).

Finally, these nuclei are of considerable astrophysical interest, as they lie along the r-p nucleosynthesis path (Wo94). How far up in mass this path goes is still a topic of great interest, as it depends both on the details of the astrophysical site and on the binding energy, shape and beta decay of these nuclei.

I.B.3. The 100Sn Region

A major goal of gamma-ray spectroscopy along the N = Z line is to study 100Sn, the heaviest self-conjugate doubly magic nucleus which can exist. The single particle energies and residual interactions with respect to the 100Sn core can be deduced directly from properties of excited states of 101Sn(1n), 99In(ph), 98Cd(ph-ph), 100In(ph-n) and 102Sn(n-n) which have one or two nucleons outside the 100Sn core. Spectroscopy of 100Sn, although still a distant goal, would reveal properties of the core itself. No excited states in any of the above nuclei are known due to the large spread in fusion-evaporation products in reactions used to produce nuclei in the 100Sn region, leading to production cross-sections anticipated to be in the 10�s of microbarn region. Until now, despite numerous attempts, the closest in-beam approaches towards 100Sn have been the 3- quasiparticle nuclei 97Ag(3ph) and 99Cd(2ph-n).

In-beam gamma-ray experiments in the 100Sn region require both high gamma-ray detection efficiency and ancillary detectors providing excellent reaction channel selection. Relatively small Ge-detector arrays such as OSIRIS at the Hahn-Meitner Institute in Berlin or NORDBALL at the Niels Bohr Institute in Denmark combined with charged particle detectors and neutron detectors already proved to be very powerful tools for in-beam studies in this region and presently define the state-of-the-art for this research. To make further progress into even weaker channels, at the microbarn level, the high efficiency of Gammasphere AND improved channel selection are needed. The most promising way of enhancing channel selection appears to arise from combining contemporary techniques. In this case, efficiently detecting the prompt evaporated charged particles in a device like MICROBALL, the mass of the residue in the FMA, and its stopping characteristics in an ion chamber, should overdefine the events to a degree which unambiguously allows channel identification for each gamma ray. The beta decay of 100In, 102In, 101Sn, 103Sn and some of the very neutron deficient Cd and Ag isotopes is followed by proton emission and, in principle, the Recoil-Decay Tagging method can be used to select gamma rays associated with their excited states. This method has already been used with proton emitters (see II.A.3.) but it has yet to be explored for beta delayed proton emitters. It has the drawbacks that b-delayed proton emission results in broad, featureless proton spectra, and delayed proton branches are usually only a small fraction of all b-decay. The normal RDT method can be applied to the island of alpha radioactivity just above Z = 50 where, for example, the proton-drip line nuclei such as 106Te, 107Te and 108I can be investigated.

I.B.4. Secondary Reaction Studies

The feasibility of utilizing a �double� nuclear reaction to study nuclei which cannot be produced using stable projectile-target combinations merits consideration. In this scheme, a primary beam is used to produce residues which will have sufficient recoil energy to induce further fusion reactions on a secondary target leading to the production of exotic final products. It is envisioned that these secondary reaction products can be identified using the Fragment Mass Analyzer (FMA) in conjunction with an ion chamber to provide Z-resolution.

Currently, we are attempting to measure residues from secondary reactions near 100Sn. The goal of the project is to investigate isotope production rates with secondary reactions in the A = 100 region and maximize the sensitivity of the FMA in conjunction with an ion chamber. If the technique develops as we anticipate, a new avenue of spectroscopic studies on proton rich nuclei which cannot be made with present target/projectile combinations will open up at ATLAS. In the best cases, the production rates will have effective cross sections in the mbarn region, and therefore the use of a device such as Gammasphere will be essential to probe excited states in these exotic secondary residues.

One interesting facet of these double reactions may lie in the domain of high angular momentum. The targets in these experiments are evaporated layers of material, and the time between primary and secondary collisions is ~ 1 fs (1 � 10-15s). The angular momentum involved in forming the first compound nucleus can be as high as 25 , and rather little of this is lost, ~ 5 , (through particle decay or gamma emission) before the second reaction is induced. Thus, in the secondary compound nucleus a new domain of high angular momentum at relatively low excitation energy can be reached. When the focus of interest is high angular momentum studies, reactions can be chosen where only a few residues are produced (as opposed to the high fragmentation far from stability) and effective double reaction cross sections in the many mb range should be achieved.

I.B.5. Isomer Physics

Nuclear isomers are unusual low-lying states with inhibited electromagnetic decay modes, and their study can provide special insights into microscopic nuclear structure and the underlying nuclear Hamiltonian. Isomers can occur from a variety of causes: irregular spin sequences (yrast traps), which reveal the strength of residual interactions between nucleons in specific configurations, or the orthogonality of wavefunctions, for example in K-Isomers, relevant to the symmetries of the Hamiltonian.

The nature of the K quantum number has become an intriguing question in recent years, particularly with regard to its high-spin behavior, where multi-detector arrays have revealed examples that seem to violate established rules of K-selection. These �anomalous� decays are typically only a few percent of the total decay strength of the high-K isomer in question, which in turn are only a few percent of the total population cross-section of the fusion evaporation channel. Theoretical models attempting to explain the mixing of states with widely different values of K, have considered the possibilities of both orientation fluctuations with a fixed, deformed shape, as well as spontaneous deviations from axial symmetry due to quantum-mechanical fluctuations of the shape itself. A systematic investigation of the phenomena is crucial for disentangling the contributions from the different variables that might play a role in the process.

Experimentally, the key to studying isomerism lies in timing. When nuclear states are prepared in a reaction, almost all of them decay by particle or gamma emission on time scales shorter than 10-10 s. What is left is the delayed radiation due to isomers, which can be measured even if it is very weak. The proposed arrangement of Gammasphere at ANL has two aspects which should allow ultra-sensitive isomer experiments. Firstly, the beam structure from the ATLAS accelerator is in bunches of FWHM about 10-10 s which can be spaced to suit the particular experiment. The availability of a chopper at ANL, designed to remove some beam pulses widens the scope of this method beyond what is presently feasible at LBNL. Thus, even very short-lived isomers can be investigated, constrained only by the performance of Gammasphere itself. Our plans to improve the performance of Gammasphere in the low-energy and fast-timing aspects are discussed below, in the technical section.

For longer-lived isomers, in the 10-6 s range, even more sensitive experiments will be possible in which prompt radiation can be measured with Gammasphere at the normal target position, then the isomeric nuclei recoil through the FMA and their delayed radiation detected at the focal plane. These experiments would be essentially background-free, so isomers populated at the sub-microbarn level could be found -- three orders of magnitude more sensitive than contemporary experiments. In this arrangement both the decay of the isomers and the structures built on them can be studied. High- lying, high-K, short-lived (ns) isomers could be studied triggered by their decay to lower-lying, low-K, isomers in the ms range. This configuration would be particularly interesting for investigating the violations of K-selection rules, as it would allow the properties of the bands built on K-isomers to be probed and extremely weak K-violating branches measured.

I.B.6. Nuclear Structure at and beyond the Proton Dripline

The observation of groundstate proton radioactivity in more than 15 nuclei between antimony and bismuth clearly shows that much of the proton dripline can be reached in heavy nuclei in experiments using stable beams (Ho89). In fact, for direct proton radioactivity to compete with beta decay, the last protons need to be unbound by about 1 MeV, well past the literal dripline. Many of the heaviest proton activities have been identified at ANL during the last year using the FMA. These proton activities have already proven to be rich in physics, providing spectroscopic information about single particle states and wavefunctions at the dripline (Pa94,Da95).

With the combination of Gammasphere and the FMA we can build on this unique and rare decay mode and use the proton radioactivities as a trigger for studying excited states in nuclei beyond the dripline. The RDT technique, which provides a tag for the reactions which produced the proton emitters, can be extremely helpful to cleanly isolate these reactions in the analysis, even when cross sections for these exotic nuclei are at the level of microbarns. An example of studying nuclei beyond the dripline is given in figure 2. The proton emitter 147Tm was studied using the AYE-Ball array in a one day experiment. Despite the low efficiency of the array (< 1/7 of Gammasphere) and the short study, excited states in 147Tm could clearly be established, based both on the h11/2 groundstate and on its low spin isomer. The sensitivity of this technique is clear for the future; the backgrounds are sufficiently low that only a few decays from a state need to be detected in order to make a firm identification.

There is currently great theoretical interest in partially bound, and unbound nuclear states at the driplines. While many of the more exotic predictions (skins, collapse of shell structure, changes in spin-orbit splitting) are predicted in the most neutron rich nuclei, there are many issues to explore along the proton dripline. It is interesting to speculate on how these unbound nuclei are formed in the first place. Other than those that are formed directly along the yrast line, excited states can have much larger particle widths (for decays to nuclei nearer stability) than gamma widths (which cool to the groundstate). Are there collective modes which enhance survival against particle emission? Along the yrast line, near the groundstate, particle emission does not compete well with electromagnetic decay, so the shapes and wavefunctions of these nuclei can be investigated by observing the g-cascade observed in Gammasphere as a high-efficiency and high-multiplicity detector, tagged by the decay protons in the focal plane of the FMA. From such measurements the extent to which single particle states and residual interactions change in the vicinity of the dripline can be quantified. Furthermore, the limits of how much angular momentum can be carried can be measured.

It is also interesting to consider that any states which have hindered electromagnetic decay matrix elements, or do not mix strongly with neighboring levels, (shape isomers, yrast traps, pairing vibrations etc.) might well have direct proton or alpha decays to more tightly bound nuclei, perhaps in competition with gamma emission. Thus, equipping the Gammasphere target chamber with segmented large silicon arrays may reveal, in addition to the continuous evaporation spectrum of protons and alphas, a new spectroscopy of direct particle decay between states in one nucleus and related states in another, or a gamma cascade in one nucleus followed by discrete particle emission then more gamma de-excitation. Clearly, this type of phenomena would be closely related to the decay out of superdeformed states. The whole issue of competition between electromagnetic and particle emission, especially in this unique regime, is interesting and very relevant to properties of nuclei in other regions.

Although much of the physics beyond the dripline is of necessity speculative, it is clear that the potential for really new and unusual behavior is high. Developing the experimental techniques for these very difficult measurements opens up an entire vista for more conventional nuclear structure issues in nuclei beyond the present limits of accessibility all the way to (and beyond) the dripline. As one moves further from stability, alpha decay tends to become more common and short-lived, so the effectiveness of the technique improves. We have performed many studies near the dripline using the composite AYE-Ball, the FMA and the RDT technique. As an example, figure 3 shows data from a measurement of an isomer in 156Hf, which lies only 1 proton away from the dripline. The a-decaying isomer was found to be a classical yrast trap. In addition, online analysis clearly isolated excited states in four new isotopes of mercury, 176,177,178,179Hg, and data on dozens of previously unstudied nuclei were obtained.

I.B.7. Hyperdeformation

Recently, there have been several papers (Ga93,Vi95,La95) reporting the existence of gamma-ray sequences with a gamma-ray spacing of ~ 30 keV in Dy and Gd nuclei around A = 150. It had been suggested that these sequences correspond to rotational bands of hyperdeformed nuclei, i.e. nuclei with a 3:1 axis ratio. More recent studies (Sa95) have brought this interpretation into question. Thus, no clear experimental evidence exists for hyperdeformation in heavy nuclei at the moment.

Theoretical calculations predict that the third minimum becomes yrast for spins > 60 . While it is not difficult to bring large amounts of angular momentum into the compound system using heavy-ion reactions, the excitation energy for the residues of interest must also be considered. As a result, the beam energy chosen to populate spins in a nucleus above 70 results in the depletion of this channel relative to the other open channels. For example, in a recent search at Gammasphere to look for nuclei with deformations larger than 2:1, we found it crucial to choose a beam energy such that our nucleus of interest 180Os (4n evaporation) was populated at an intensity four times lower than that of 179Os (5n). This had two major consequences on the data: (1) the spectrum of gamma rays is dominated by the 5n channel, and (ii) a considerable fission background was present. These factors contribute significantly to the background associated with 180Os. Such backgrounds make it difficult to isolate the weakest gamma-ray sequences such as those associated with hyperdeformation.

While the search for hyperdeformed nuclei is one of the main research topics at Gammasphere, it may prove that isolation of these weak channels which are populated at the highest angular momenta are beyond even a device as sensitive as Gammasphere unless enhanced external triggering is used. However, it should be noted that by using the FMA as a channel selector and the BGO shields as a multiplicity filter, Gammasphere would become a much more sensitive device for detecting weak gamma-ray sequences. Indeed, the backgrounds associated with fission and other reaction channels would be eliminated in this setup. With the addition of the microball, one could also probe whether or not the evaporation of charged particles is critical in populating the third minimum.

I.B.8. Deformation and Superdeformation in Light Actinides

One of the early applications of the Nilsson model was the quantification of the regions of deformation away from shell closures, and the fact that deformed shell structure could stabilize a variety of nuclear shapes (Ma63). This was spectacularly demonstrated to be true with the observation of fission isomers, and their interpretation as highly deformed, shell stabilized configurations. After the first fission isomer region was found in the actinide nuclei, extensive calculations revealed that another region of highly deformed isomers should exist in very neutron deficient actinides, well below the N = 126 spherical shell gap, and these nuclei should have considerable oblate deformation in their groundstates. Experimental searches for delayed fission of these states were not successful (Bj70), the researchers concluding that the highly deformed shapes have a sufficiently low inner barrier that they preferentially decay back to the first minima, rather than by delayed fission.

Despite the fact that such states in the second minimum are now called superdeformed and a whole science of calculating and measuring their properties has developed in other mass regions, the light actinide region remains completely unknown. This is largely due to the difficulty in producing nuclides in this region, which can only be formed through fusion of heavy ions, that usually fission during the cooling of the compound nuclei. The residue cross sections have been measured and are usually less than a few millibarns. Considerable efforts (Yo95,Fr95) have been made to study these nuclei, using the FMA, but Gammasphere is needed to provide the necessary gamma-ray efficiency to allow real progress to be made in studying the nuclei of greatest interest. Experiments at Argonne indicate that the RDT method, combined with the segmentation and efficiency of Gammasphere, will make these nuclei accessible and the observation of superdeformation possible.

Recent calculations indicate that these nuclei are particularly interesting in the context of superdeformation because they have well developed second minima lying at very low excitation energy. The observation of such states would form the missing link between the physics of fission isomers and that of high spin: when formed in heavy ion reactions, these states may be expected to have gamma transitions to their bandhead before decaying out to normal states. Despite the large deformation, pairing should be large. The competition between the fission out through the outer barrier and gamma decay through the inner barrier may be found, which could be very relevant for exploring the issue of fission timescales. This competition is very sensitive to angular momentum, and is of intrinsic interest to understanding the reaction process and nuclear viscosity.

I.B.9. Octupole Correlations in Actinide Nuclei

The study of nuclear shapes and the underlying microscopic structures which determine them has been a topic of interest in nuclear spectroscopy for the last decade. One area of interest has been the occurrence of reflection asymmetric shapes. These shapes are relatively unusual, and it is still not clear whether any nucleus has a intrinsic reflection asymmetric groundstate, or just strong octupole (23-pole) correlations. Transitional actinide nuclei present the strongest cases of octupole correlations, which are manifested by bands of states with opposite parity which are connected by enhanced electric dipole decays. The region of nuclei where these features have been observed is quite small and centers around N = 132 and Z = 90. Of particular theoretical interest are the heavier nuclei in this region, particularly the uranium isotopes, which have been predicted to have the strongest octupole correlations (Ch86) of all. However, the nuclei seem particularly sensitive to the influence of even higher multipole terms, including 25-pole (So88) (which enhance the octupole correlations) and 26-pole terms (which diminish them) (Ch92). Thus, an experimental measurement of octupole correlations in these nuclei provides a unique probe into higher order terms in the shape of the nuclear mean field.

Unfortunately, nothing is known about excited states in the key isotopes of interest, 224,225,226U, so the theoretical calculations cannot be tested, nor the correlations measured. This is because there is prohibitive fission competition when trying to produce these nuclei, competition which leaves only 1 in 105 of the fusion products surviving to form a heavy residue. Recent experiments at Dubna (An91,Ye94) have been successful in synthesis to the extent that the groundstate alpha decays, halflives and production cross sections are known. A trial experiment was performed at ANL this fall which indicated that with the seven-fold increase in photon detection efficiency from Gammasphere these experiments should be entirely feasible provided suitable upgrades are made to the FMA focal plane detectors to make them more efficient for very heavy, slow moving residues.

I.B.10. The Structure and Stability Approaching the Heaviest Nuclei

One of the long-term goals in nuclear physics has been the production of superheavy elements with around 114 protons. Recent work at GSI has begun to close in on this goal with the exciting discovery of elements with Z = 110 and 111. All these heavy elements are classically unstable against fission because of the large Coulomb force. The stability of elements with Z up to 111 and the predicted existence of element 114 are due to quantal shell corrections, which create a pocket in the potential energy surface (as occurs for superdeformation and fission isomers), thereby creating a fission barrier which otherwise would not exist. The stability of the superheavy elements is based on the doubly closed spherical shells at Z = 114, N = 184. However, the stability of the heaviest elements with Z between 100 and 111 (which was not anticipated) arises from the fact that the shell correction energy is maximized when the nuclei assume deformed shapes, with not only quadrupole but higher-order deformation. Figure 4 shows the shell correction energy for the heaviest elements calculated in (Mo94).

An experimental verification of the predicted b2 and b4 deformation parameters provides an important check on theory. This test also has implications for the structure and the ease of producing yet heavier elements. Two calculations (Mo94,So94) predict different binding for elements 112 and 113, which imply different fission barriers and a-decay half-lives. Another interesting aspect is the prediction by M�ller that a sub-shell closure occurs at N = 178. The major uncertainty in calculating the shell-correction energy comes from the single-particle energies. If an improved knowledge of single-particle energies confirms this sub-shell, it has implications for producing element 114.

We propose to initiate a program to study the nuclear structure of the heaviest Z elements which can be reached for in-beam spectroscopy. By coupling Gammasphere, the FMA and a Si-strip detector (for recoil decay tagging of a decays), we combine 3 individually powerful devices into an unparalleled detector for studying g rays from the elements with Z � 100. (To study the g rays from such heavy elements, Gammasphere will be augmented by ~ 10 low-energy photon detectors since the lowest lying yrast states with I � 6 have low energy. This is discussed in section II.B.1. In addition, the lowest-energy transitions are also highly converted, leading to the emission of K- and L- x-rays.) This powerful detector combination will make it possible to study ground band rotational transitions and to determine their lifetimes by the recoil distance method. In this manner, we shall be able to deduce b2. In order to obtain information on b4, we need to determine single-particle energies in odd-Z or odd-N isotopes. Of course, these energies are also of direct interest for refining calculations of the shell-correction.

These measurements will provide information not only on the structure of the very heavy elements, but also on the production mechanism of the evaporation residues. The insight gained may be useful for enhancing the production of the superheavy elements. Any gains, even small ones, will be invaluable as the production cross-section for element 114 is extrapolated to be < 1 pb (Ar95b). Currently it is not understood to what extent the production cross-section is limited by the fusibility or fissility of the compound system. We can obtain information on the role of the latter by studying the spin-dependence of the fission barrier Bf( ) through the partial wave content of the evaporation residues (ER). The partial wave distribution of the ER�s may be inferred from the ground band population intensity as a function of spin. Alternatively, the evaporation residue spin distribution may be inferred from the g-multiplicity (or fold) distribution. Using Gammasphere this could be achieved best by removing the BGO shield heavymet collimators. Bf( ) is the most important property governing the survival and stability of the evaporation residues.

It is estimated that the feasibility limit for study occurs near Z = 102, which has been observed to have a production cross-section of 300 nb in the 208Pb(48Ca, 2n) reaction. With a 5 pnA beam and a 0.5 mg/cm2 target, ~ 10-20 counts/d are expected in the full- energy peak of the 8+ � 6+ transition (with an estimated energy of around 210 keV) in gamma-FMA coincidences. The lower transitions become increasingly difficult to observe because of internal conversion. Study of lower Z elements than Z = 102 is more favorable because the production cross-section rises by a factor of ~ 3 for each unit decrease in Z (Ar95b).

I.C. Projects Exploiting the Unique Beams and Beam Properties at ATLAS

I.C.1. Coulomb Excitation of (and with) Heavy Beams

The structure of nuclei at high spin has been primarily investigated using (HI,xn) reactions, which lead to strong population of mainly the yrast states. As a consequence, the region within ~ 1 MeV of the yrast line is largely unexplored. A class of states in this region, which have large wave-function overlap with the groundstate, can be populated by Coulomb excitation but is otherwise inaccessible. Figure 5 schematically illustrates the regions of interest available to Coulomb and nuclear inelastic excitation and the related physics issues. Gammasphere is the ideal device for Coulomb excitation studies. Its high efficiency allows the measurement of very weak decay matrix elements. The large angular coverage allows angular distributions to be measured for all the observed transitions. When Gammasphere is operated with internal particle detectors a full set of correlations of gamma-ray yield with impact parameter and emission angle can be explored. These data will greatly constrain the possible sets of nuclear matrix elements which are involved in excitation and decay. In many cases, unique solutions may be found. Operation of Gammasphere at Argonne for Coulomb excitation studies has several advantages. Most importantly high quality beams of all the heavy elements are available, especially 208Pb, with good timing, intensity, and energy resolution. Thick target experiments, in which some of the nuclei stop before decay have recently been shown to be very sensitive to non-yrast nuclear structures (Wa95a). These latter experiments would also benefit from the wide variety of beams available at ATLAS with energies near the barrier and with excellent time characteristics.

An example of the non-yrast states which are of interest are the double-phonon excitations in 208Pb. The first excited state in the doubly-closed-shell 208Pb is a 3- excitation which has been interpreted as a collective octupole excitation. Within the phonon description, there should exist a quartet of levels with Jp = 0+, 2+, 4+, 6+ which constitute the double-octupole phonon excitations. Since identification of the 2-phonon states in 208Pb constitutes an important test of the phonon description in nuclei there have been many searches for members of the double-octupole phonon multiplet in 208Pb, but there has been no unambiguous identification. Whether this is a consequence of the dissolution of the members among other states, the non-existence of these 2-phonon states, or is due to a lack of detection sensitivity, is not known. The population cross- section for the double-phonon states is maximized in the 208Pb + 208Pb reaction. Taking into account the fact that both projectile and target can be excited gives an approximately 4-fold increase in yield compared to excitation with, say, a Xe projectile. Hence, the availability of Pb beams from ATLAS, coupled with the large sensitivity and granularity of Gammasphere, provide the most powerful combination for studying this problem. The granularity is important because it is important to identify events where the g multiplicity is exactly 3 (e.g. 6+ � 5- � 3- � 0+) as the required signature of double-phonon excitations.

Another long-standing problem in nuclear structure studies is the identification of multiple phonon excitations in deformed nuclei. e.g. double g-phonon or 1 g-phonon � 1 octupole-phonon excitations. Coulomb excitation with very heavy ions is the ideal way for populating multiphonon rotational bands. Recent work at Chalk River (Wa95a) has demonstrated the power of this technique by identifying the rotational members of a purported K = 4+ double g-phonon band in 232Th which was Coulomb excited by 209Bi beams. The availability of Pb or Bi beams at ATLAS permits investigations of this type, where the power of Gammasphere is used to establish the rotational bands of interest. The high-sensitivity of Gammasphere is necessary for the determination of B(El) values from lifetime measurements. For example, line-shape analysis with the Doppler-shift attenuation method requires very clean spectra which can only be produced by multi (2- or 3-fold) gating.

Structure investigations of nuclei near the stability line, which are not normally accessible by (HI, xn) reactions, are feasible with not only Coulomb excitation, but also with transfer plus Coulex. As an example, the population intensities of the ground band members populated in neutron pair-transfer are of interest for studying the decrease of pairing with spin or diabolical pair transfer (and its associated Berry phase).

While 208Pb or 209Bi are expected to be the main beams used for Coulomb excitation studies of the target, it is also expected that the projectile is the nucleus of interest in some cases. Cline et al. (Cl95) have found that this approach is necessary in those cases where the target contamination poses problems. The availability of the complete spectrum of beams from ATLAS is essential in these instances. The narrow beam pulse width is also an advantage for obtaining good mass resolution in time-of-flight measurements of the binary reaction products.

I.C.2. Deep Inelastic Reactions for the Study of Neutron Rich Nuclei

The heavy beams available at ATLAS allow us to systematically explore nuclear structure studies of neutron-rich nuclei produced in deep inelastic scattering reactions. There has been considerable recent interest in nuclei on the neutron-rich side of the valley of stability, both for their basic nuclear structure physics and for their relevance to nuclear astrophysics. This interest has been triggered in part by the promise of reaching very neutron-rich nuclei, possibly out to the region of the astrophysical explosive r- process, through reactions with radioactive beams. The deep-inelastic approach discussed here may become an important precursor to these studies by providing the first step in exploring the neutron-rich side of the valley of stability with good intensity stable beams.

This class of reactions between heavy nuclei has shown large mass and nuclear charge transfers. Collisions between heavy beams with a large neutron excess (238U for example) and neutron-rich target nuclei have produced nuclei on the neutron rich side of the valley of stability at medium angular momentum. These have already been successfully used in first nuclear structure studies. Early decay work has used neutron- rich light to medium-heavy beams on targets of tungsten. Later lead and uranium beams were used on neutron-rich targets to produce nuclei in various mass regions for subsequent beta-gamma groundstate decay studies (Sc91,We93). Recently, high resolution gamma-ray coincidence techniques have been applied, both off-line and in- beam, at the Hahn-Meitner Institute in Berlin (Pa93) and at ATLAS (Br92, Ca94). Neutron-rich Sn isotopes, for example, have been studied up to 126Sn. These data, together with information from neutron-rich fission products led to unique structure studies of the seniority scheme and of subshell occupation. First experiments at Gammasphere (Le94) have focused on neutron rich Yb isotopes. These measurements have been performed using Ca and Sm beams and have shown the promise of the technique.

With Gammasphere and its high multiple-fold efficiency these studies can be extended to weaker channels, and thus further into the unexplored region of neutron-rich nuclei. The intense, high-quality uranium beams, for example, that are available at ATLAS are estimated to allow with Gammasphere studies of neutron-rich isotopes up to about 10 neutrons beyond the heaviest stable isotope. Deep inelastic reactions near the barrier involving uranium beams on the most neutron-rich stable tin target (124Sn) have produced neutron-rich nuclei between osmium and lead up to 18 neutrons beyond the last stable isotope with microbarn cross sections (Ma85).

II. Technical Issues Related to Using Gammasphere at ANL

There are a series of technical questions which are important to consider when evaluating the merits of the move of Gammasphere to ANL. Foremost amongst these are the issues of the uniqueness of the proposed Gammasphere-FMA-RDT combination for selecting particular nuclei, and the characteristics of the beams produced by the ATLAS accelerator. In this section we will discuss these issues and some upgrades to Gammasphere and the FMA which will make the system even more powerful. We also review the current status of the ATLAS accelerator operation and upgrades which are in progress.

II.A. Channel Selection

II.A.1. A Comparison of High-Fold and Recoil Triggering

The selectivity of using a sequence of gamma rays to select a state or nucleus of interest has been widely discussed (Ga88). When the state or states of interest lie in long cascades of gamma decays, and when the state is relatively strong -- more that 10-3 of the reaction cross section, this technique has been demonstrated to be extremely powerful, provided high fold (i.e. four or more gamma rays detected in true coincidence) data can be efficiently acquired. Gammasphere is a device designed to collect such data. However, when there are intense background processes like Coulomb excitation or fission, or when the channel of interest lies below 10-5 of the reaction products, then external triggering, selective only on the process of interest, becomes necessary and restores the power of the device. A quantitative discussion of this competition and the merits of external triggering for very weak reaction channels has been published (En93). Spectra comparing the relative merits of g-g and g-FMA gating are shown in figure 6 for the 160Gd(36S,7n)189Hg reaction, where 189Hg constitutes less than 1% of the reaction cross-section. The spectrum obtained by selecting A = 189 in the FMA and a 403 keV g ray (fig. 6b) is very clean but the statistics are low because the g array consisted of only 10 detectors. In contrast, the spectra obtained by pairwise gating on data from Eurogam I (with 45 detectors) and Gammasphere (with 32 detectors), contain contaminants.

II.A.2. A Comparison of Recoil Separators and Particle Detectors

To date, the only external triggers used with Gammasphere have been light charged particle detectors mounted in the target chamber. Two devices have been used, the CsI 4-� array microball and a large area Si-strip detector. It is important to compare the merits of these devices relative to the FMA when one contemplates moving Gammasphere. Several issues are relevant and the comparison is not always trivial. They include efficiency, resolution, countrate limitations and target purity. For this discussion let us consider the microball.

Microball is a CsI detector with nearly complete angular coverage (Sa95). It can detect and resolve light charged particles. Thus, for a heavy-ion reaction followed by particle evaporation, channel enhancement is achieved through detecting the emitted charged particles. The enhancement is not entirely unambiguous, as channels involving neutron evaporation, or channels with charged particles which avoided detection, can satisfy the channel selection criterion. Further, implicit in using the method is the assumption that the compound nucleus is known, in other words, that the target is mono-isotopic and without contaminants. Despite these shortcomings, it has been shown with first generation arrays that the combination of light particle detectors, neutron detectors, and efficient gamma ray detectors has considerable power for isolating and studying nuclei far from stability (Li82,Ch88,Gr89,Ku92). However, the shortcomings of the method become apparent when the nuclide of interest falls below 1% of the fusion cross section or a few millibarns. It is in this regime that electromagnetic spectrometers such as the FMA become competitive; at cross sections below 0.01% they become essential.

The FMA is a mass-dispersing, energy focusing spectrometer (Da92). It has a mass resolution, M/DM > 300 FWHM. It has an energy acceptance of � 20% of the central energy. The focal plane covers a range of mass/charge (A/q) of � 4% of the central value. The FMA is isochronous. Measuring the time-of-flight of ions through the spectrometer by measuring their arrival at the focal plane against the RF beam structure can usually remove the A/q ambiguities and make the spectrometer a true mass- determining device. We have been refining this utilization of the device for mass determination and background suppression in our double reaction studies (section I.B.4.). This mode of operation will improve most experiments. The range of acceptance is usually sufficient to accept and resolve most residues from near-barrier fusion- evaporation reactions in two atomic charge states, giving an overall efficiency of 5-15%. The quality of primary beam rejection depends on the kinematics of the reaction under study, but is sufficiently good that the focal plane countrate is seldom dominated by scattered beam particles. For operation with Gammasphere it would be used as a zero degree spectrometer with a solid angle of about 2 msr. The mass and energy selection of the FMA can be greatly enhanced by suitable detection systems beyond the focal plane. In light nuclei, the optimum detector arrangement seems to be the use of gas filled ion- chambers (Li90) for which case, using Gammasphere, in-beam spectroscopy is possible near the 1 microbarn level, that is 10-5 of the fusion cross section. In heavier nuclei, especially those far from stability, triggering on subsequent alpha decay using the Recoil Decay Tagging method [RDT] has proven equally, or even more sensitive. Figure 7 depicts a few of the capabilities of the FMA.

During 1995 we have constructed an array of g-ray detectors at the target position of the FMA to fully explore what may be possible with Gammasphere. The AYE-Ball (Argonne, Yale, European Spectrometer) consists of up to 20 Compton suppressed high purity germanium detectors, half of which were of Gammasphere size (> 70% efficiency). The device, and some of the areas of physics addressed are shown in figure 8. The research technique which appeared most novel and promising was using Recoil Decay Tagging (RDT). Using the AYE-Ball composite array, we have been using this technique to investigate many new nuclei in the actinide and rare-earth regions. With insertion of Gammasphere in place of AYE-Ball there is more than a factor seven gain in photopeak efficiency, which scales with the power of the gamma fold under consideration, that is gamma-gamma data will be enhanced by more than a factor fifty. Further improvement of the focal plane detector arrangement should also allow higher countrates and more efficient transmission and identification of the residues of interest, by factors of 2 to 20 depending on the kinematics of the experiment. Thus, more than two orders of magnitude improvement in efficiency may be hoped for when comparing our current setup to the one we plan for 1997. Again, scaling our measurements of the cross-sections which can be usefully exploited now, to those we should reach in 1997 leads us to conclude that coincidence gamma ray spectroscopy should be possible in channels populated with cross sections below 1 microbarn, which is typical for many of the dripline proton emitters which have recently been discovered.

II.A.3. Recoil Decay Tagging (RDT)

Little is known about states in nuclei near, at, or beyond the proton dripline for nuclei with A > 100, nor any excited states in nuclei with Z > 100. This is due to the fact that with stable beams these states can only be populated with very small cross-sections, frequently less than 1 in 106 of the fusion cross section. Even with a powerful channel selection device like the FMA, which isolates reaction products by their mass/charge ratio, there is not sufficient selectivity or background suppression. In the dripline nuclei, the interfering backgrounds come from isobars (nuclei with the same mass, but different atomic number) which have the same mass/charge ratio but lie nearer stability and may be produced 104 times more strongly. In heavy nuclei the backgrounds come from isobars, scattered primary beam, Coulomb excitation and fission fragments.

For nuclei which have alpha or proton decays an ultra-sensitive technique has been developed which can overcome these background problems. The method relies on the separation of the nuclides of interest by an electromagnetic separator (SHIP at GSI, the Recoil Separator at Daresbury, VASSILISSA at Dubna, or the FMA at Argonne), followed by implantation of the ions into large area, highly segmented, double sided strip detector (DSSD) arrays. The arrival and decay of each embedded ion can be followed until that pixel in the DSSD is struck by another reaction product. Thus, the energy and time of the arriving nucleus can be measured, then the energy and time of its decay, and frequently a series of subsequent decays back towards stability. In general, the further from stability one progresses, the higher the decay energy, the shorter the halflife, and the greater the probability of observing chains of the groundstate decays, all of which make the identification more unique. By this method, the new elements with Z > 108 have been identified, all the known proton emitters, dozens of new alpha emitters, and many cases of fine structure in alpha and proton decay have also been observed.

This method can be built upon to allow in-beam particle and gamma-ray spectroscopy on these exotic nuclei, using a method first suggested by Woods (Wo88,Pa95) and called Recoil Decay Tagging (RDT). The key concept is to time-correlate radiation from the production reaction at the target position with flight through the spectrometer and subsequent decay. Although the times involved are all very different , 10-21s for particle emission, 10-15 to 10-9 for gamma emission, 10-6 for flight of the residue and 10-5 to 10+2 for the subsequent decays, these time correlations can be made. It has been shown (Pa95) that cascades of gamma rays can be uniquely associated with nuclei, even right at the dripline. 108,109Te, 109I and 113Xe were identified for the first time by this method. These initial experiments were performed using the gamma array Eurogam I and the Daresbury Recoil Separator.

II.B. Detector Upgrades

II.B.1. Low Energy Gamma-Ray Detectors for Gammasphere

The use of very large, high purity Germanium (HpGe) detectors in Gammasphere has been essential to achieve high photopeak efficiency for photons in the energy range 200 to 4000 keV. Unfortunately, these detectors have shortcomings in the lower energy range of nuclear radiation, below 200 keV, due mainly to their large volume. Their intrinsic energy resolution at low energies is not good, usually worse than 1.5 keV FWHM, and timing resolution also deteriorates badly at low energy. For experiments which need to be sensitive to low energy radiation, for example in the spectroscopy of very heavy nuclei, or for experiments which need excellent timing, like isomer studies, these shortcomings are quite severe. Fortunately, there is a solution which is both straightforward and relatively inexpensive.

The solution involves replacing some of the large detectors in Gammasphere by planar germanium spectrometers (LEPS). These are discs of intrinsic semiconductor, typically 50 mm in diameter and 10 mm thick. Their electrode configuration is that of a true parallel plate capacitor, so the charge-collecting field is very uniform. Typically they have resolutions of 700 eV FWHM at 122 keV, have timing of better than 4 ns for high energy radiation and better than 20 ns for gamma rays as low as 20 keV, and can count at rates in excess of 100 kHz. Because they contain rather little semiconductor, their cost is mainly that of the mechanical housing. The installation of about 10 of these detectors in Gammasphere would have an efficiency of more than 2% for 100 keV radiation, probably contributing more than the other 100 big detectors to most measurements in this energy regime.

We have been planning to develop an array of planar detectors, both for use in Gammasphere, and as a powerful standalone array in its own right. We hope this development can be executed in 1996, so the LEPS array will be available for installation in late 1996, and will be used in many experiments at ANL in 1997.

II.B.2. Channel Plate Detector Time-of-Flight Arm for the FMA

In order to take full advantage of the capabilities of Gammasphere, it is important to upgrade the capabilities of the detectors at the FMA focal plane area. At the present time the focal plane detector is a parallel grid avalanche counter (PGAC), which provides x- and y- position, time, and energy loss information on particles reaching the FMA focal plane. Placed behind this detector at a distance of approximately 40 cm is a double-sided silicon strip detector (DSSD), used for time- and position-correlation measurements of proton and alpha decaying reaction products that are implanted into it, or a highly segmented, high-countrate, ionization chamber. The DSSD also provides energy and time-of-flight information for the reaction products, which is used to discriminate against scattered beam particles. By itself the DSSD is used for decay studies of proton and alpha radioactivities near the proton drip line. It is also used in the recoil decay tagging (RDT) method, where the gamma rays emitted by the parent nuclide at the target position are detected by Gammasphere, and can therefore be correlated with a specific value of A and Z. When the DSSD is replaced by a large ionization chamber, information on the Z of energetic ions having Z < 50 can be obtained.

Some FMA experiments, especially those involving slow-moving heavy-ions, have suffered from large energy loss and scattering introduced by the windows of the PGAC detector and the count rate limitations of this detector. For these reasons, plans are being made to replace this detector by a combination of a thin foil and a large position-sensitive channel plate detector. Electrons emitted by the foil during the passage of a heavy ion will be accelerated and focused onto the channel plate. The excellent timing and high count-rate capability of this detector system should allow us to take full advantage of the FMA-Gammasphere combination to study weakly produced nuclei at the proton drip line, both by their particle decays and by in-beam gamma spectroscopy.

Recently a larger area (40 � 40 mm2) DSSD has been used behind the FMA focal plane, replacing the original 16 � 16 mm2 detector. A new silicon box detector is ready for testing which will be placed in front of the new DSSD. Its purpose is to detect and measure the energy of decay alphas and protons originating in the DSSD which escape through the incident surface, leaving signals which are smaller than the full energy. Using the silicon box, these events can be vetoed, or the detected escape energy can be added back to produce a full-energy signal. The use of the silicon box will significantly enhance the operation of the FMA focal plane implantation system, and allow the unique FMA-Gammasphere combination to do nuclear structure measurements of unprecedented sensitivity at the proton drip line.

We have recently obtained Argonne laboratory discretionary funds in order to develop a position sensitive, channel-plate, time-of-flight arm for use at the focal plane of the FMA and in future radioactive beam experiments. This work will probably be in collaboration with the University of Manchester, U.K. where physicists have extensive experience in this field and have interests in radioactive beam related research and development.

II.C. Accelerator Status and Upgrades

ATLAS has been operating for seven days per week beginning in FY 1994, and delivered over 5200 hours of beam for research that year. With an assumed modest increase in staffing, about two FTE's supported by the DOE Facilities Initiative, we are predicting over 5600 hours for research in FY 1997. With appropriate scheduling and neglecting routine maintenance for a limited time this number could well be as high as 6000 hours during the year that Gammasphere will be operating at ATLAS.

ATLAS provides CW beams with excellent transverse and longitudinal emittances with masses spanning the range from hydrogen to uranium. Currently we use the tandem injector for beams below about mass 60 for species which work well in the negative-ion source. The positive-ion injector with its associated ECR ion source is used for all other beams. In FY 1995, 29 different isotopes were provided for research experiments.

The intrinsically excellent longitudinal emittance of ATLAS makes it possible to provide full intensity beams with both good energy spread (< 0.1%) and good rf time structure (< 200 psec). With the superconducting rebunchers/ debunchers the longitudinal emittance ellipse can be tuned to further optimize either energy spread or time spread as dictated by special experimental requirements. With appropriate detectors, such as avalanche counters at the FMA focal plane or fast liquid scintillators for neutrons, rf timing at the 100 psec (FWHM) level is especially useful for time-of-flight measurements. A beam sweeper is also routinely available at ATLAS to increase the time between rf pulses (normally 82 nsec) when needed.

A major upgrade of ATLAS will be complete in time to benefit the Gammasphere program. Beams from the second ECR ion source, presently under construction, will be available for research by early 1997. This ion source, the design of which is an evolution of the LBNL AECR, will increase both intensities and energies of ATLAS beams, as well as greatly improve scheduling flexibility. With the new source uranium is expected to be available at energies above 6 MeV per nucleon with intensities over 20 pna without the presently required foil stripping in the beamline. With foil stripping beam intensities of at least 1 pna of uranium should be available at energies up to 9 MeV per nucleon. The higher intensities for intermediate masses also make possible useful intensities of very rare isotopes without the use of highly enriched material. Beams in the energy range of 15 to 18 MeV per nucleon would be available for lighter heavy ions, if needed. Beams with intensities of at least 10 pna and with energies of at least 10 MeV per nucleon should be available up to mass 150.

References An91 A. N. Andrevev, Z. Phys. A338, 363 (1991)
Ar95a Concept for an Advanced Exotic Beam Facility Based on ATLAS Ar95b P. Armbruster, Ann. Rev. Nucl. Par. Sci. 35, 135 (1985) and private communication (1995)
Ba92 A. M. Baxter et al., Nucl. Inst. Meth. A317, 101 (1992)
Bj70 S. Bjornholm et al., Nucl. Phys. A156, 561 (1970)
Bo69 A. Bohr and B. Mottelson, Nucl. Structure Volume 1, Benjamin (1969) p. 171
Br92 R. Broda et al., Phys. Rev. Lett. 68, 1671 (1992)
Ca94 M. Carpenter et al., ANL Phys. Div. Annual Report, ANL-95/14 Argonne (1995)
Ch86 R. R. Chasman, Phys. Lett. B175, 254 (1986)
Ch88 T. Chapuran et al., Nucl. Inst. Meth. A272, 767 (1988)
Ch92 R. R. Chasman, Phys. Lett. B208, 187 (1992)
Cl95 D. Cline, private communication (1995)
Co95 G. Colo, P. Van Isaeker and A. Vitturi, GANIL preprint 95-02 and Phys. Rev. C52, R1175 (1995)
Da92 C. N. Davids et al., Nucl. Inst. Meth. B70, 358 (1992)
Da95 C. N. Davids, submitted to Phys. Rev. Letts. (1995)
En91 P. J. Ennis and C. J. Lister, Nucl. Phys. A535, 392 (1991)
En93 P. J. Ennis and C. J. Lister, Nucl. Inst. Meth. A313, 413 (1992)
Fr95 S. J. Freeman et al., Phys. Rev. C50, R1754 (1994); Phys. Rev. C51, R1047 (1995)
Ga88 Gammasphere Proposal LBL-PUB-5202 (1988); I. Y. Lee, Nucl. Phys. A520, 361 (1990)
Ga93 A. Galindo-Uribarri et al., Phys. Rev. Lett. 71, 231 (1993)
Gr89 C. J. Gross et al., Phys. Rev. C39, 1780 (1989)
Ho89 S. Hofman in Particle Emission from Nuclei, Vol. II, Ed. D. N. Poenaru and M. S. Ivascu, CRC Press, Boca Raton, (1989) p. 25
Is91 The Isospin Laboratory, Research Opportunities with Radioactive Beams, LALP Report 91-51
Is94 Overview of Research Opportunities with Radioactive Nuclear Beams (1994)
La95 D. R. Lafosse et al., Phys. Rev. Lett. 74, 5186 (1995)
Le94 I. Y. Lee et al., Proceedings of the Conference on the Physics from Large Gamma Ray Detector Arrays, p. 314 (1994)
Li82 C. J. Lister et al., Phys. Rev. Lett. 49, 308 (1982)
Li87 C. J. Lister et al., Phys. Rev. Lett. 59, 1270 (1987)
Li90 C. J. Lister et al., Phys. Rev. C42, 1191 (1990)
Ma63 E. Marshalek, Rev. Mod. Phys. 35, 108 (1963)
Ma85 W. Mayer et al., Phys. Lett. B152, 162 (1985)
Mo94 Peter M�ller and J. R. Nix, J. Phys. G20, 1681 (1994)
Ms94 The K500 � K1200: A Coupled Cyclotron Facility at the National Superconducting Cyclotron Laboratory, MSUCL-39 (1994)
Ku92 T. Kuroyanagi et al., Nucl. Inst. Meth. A316, 289 (1992)
Ns95 The Long-Range Plan for Nuclear Sciences, NSAC (1995) - to be published
Nu93 European Radioactive Beam Facilities Report of NuPPEC Study Group (1993)
Or91 A Proposal for Physics with Exotic Beams at the Holifield Heavy Ion Research Facility, J. D. Garret and D. K. Gloen, Editors (1991)
Pa93 T. Pawlat et al., Nucl. Phys. A (1993)
Pa94 R. D. Page et al., Phys. Rev. Letts. 73, 3066 (1994)
Pa95 E. S. Paul et al., Phys. Rev. C51, 78 (1995)
Sa95 D. Sarantites et al., Proceedings of the Workshop on First Physics with Gammasphere, Berkeley, CA, December, 1995 - to be published
Sc91 W.-D. Schmidt-Ott et al., Nucl. Phys. A522 , 610 (1991)
Sm86 A. E. Smith, Phys. Letts. B176, 292 (1986)
So88 A. Sobiczewski, Nucl. Phys. A485, 16 (1988)
So94 A. Sobiczewski, Phys. Part. Nucl. 25, 119 (1994)
Sp94 The Spiral Radioactive Ion Beam Facility, GANIL R94-02 (1994)
Vi95 G. Viesti et al., Phys. Rev. C51, 2385 (1995)
Wa69 E. K. Warburton and J. Weneser in Isospin in Nuclear Physics, ed. D. H. Wilkinson, North Holland, p. 175 (1969)
Wa95a D. Ward et al., submitted to Nucl. Phys. A (1995)
Wa95b D. D. Warner et al., submitted for publication and private communication (1995)
We93 C. Wennemann et al., GSI Report GSI-93-1, GSI Darmstadt (1993)
Wo88 P. Woods et al., Fifth Int�l. Conf. on Nuclei Far From Stability, Publ. A.I.P. New York 1988, Ed. I. S. Towner, p. 831
Wo94 S. E. Woosley et al., Astrophy Journal 433, 229 (1994)
Wu92 A. Wuosmaa et al., Phys. Rev. Lett. 68, 1295 (1992)
Ye94 I. Yeremin et al., Nucl. Inst. Meth. A350, 608 (1994)
Yo95 W. Younes et al., Phys. Rev. C52, R1723 (1995)

Figure Captions

Fig. 1. (Top) The detection of 24Mg at the focal plane of the FMA following the 12C(16O,a)24Mg reaction. (Bottom) Gamma rays from the reaction, showing both singles and 24Mg gated spectra. All the known decays from high-spin states were seen in a few hours. Many new transitions, with energies up to 7.9 MeV, were observed.

Fig. 2. An example of the Recoil Decay Tagging (RDT) technique on a proton emitter. In this case 147Tm was studied with the AYE-Ball array. Gamma-ray transitions associated with both the low- and high-spin proton radioactivities were identified. The need for an increase in gamma-ray efficiency, and especially gg coincidence data, is clear and would come from a corresponding study using Gammasphere.

Fig. 3. An example of the RDT technique from an a-emitter near the proton dripline. 156Hf was known to have a a-emitting excited state. Now, through observing the gamma-emitting states, it emerges as a classical yrast trap.

Fig. 4. Shell correction energy for the heaviest nuclei calculated in (Mo94). The shell correction between Z = 100 and 120 creates a fission barrier which is responsible for the stability of the heaviest elements.

Fig. 5. A schematic illustration of the Coulomb excitation process. Heavy beams, especially 208Pb at ~ 6 MeV/u are very powerful for this research.

Fig. 6. A comparison of high fold gating and that obtained for recoil-g-g coincidences. 189Hg was produced in a 7n evaporation channel which constituted < 1% of the reaction products. Figure 6a shows all g rays from an FMA experiment, 6b shows mass gated gg data, while 6c shows Gammasphere ggg triples data and 6d Eurogam triples.

Fig. 7. The implantation facility at the FMA focal plane uses position information from the PPAC detector to establish the mass/charge (M/Q) ratio of reaction products. The decay energy vs. M/Q plot shows the alpha-decays of these recoils after implantation in the double-sided silicon strip detector (DSSD). Each alpha decay can be assigned a parent mass. The projected total M/Q and energy spectra are also shown.

Fig. 8. The AYE-Ball (Argonne-Yale-European Spectrometer). The upper part of the figure shows the 20-detector array, and the lower part shows the scope of experiments. Nuclei which were studied during its 4 month period of operation are marked. Gammasphere will be more than seven times more efficient for single g-ray measurements and more than fifty times more powerful for g-g coincidence studies.

Fig. 9. A mechanical drawing of the Gammasphere-FMA combination. The figure indicates the modifications needed to the FMA and the rails for the Gammasphere support frame.

Fig. 10. Top view of the Gammasphere-FMA combination.

Fig. 11. A schematic outline of the proposed project management plan for Gammasphere at ANL.

Fig. 12. The proposed timelines for the various tasks of site preparation, move, installation and operation of Gammasphere at ATLAS.