T.A. Trainor and the APEX Collaboration

Narrow structures reported by GSI positron experiments EPOS and ORANGE first in positron singles energy spectra and later in electron-positron energy distributions associated with collisions of very heavy ions near the Coulomb barrier have been an outstanding puzzle in nuclear physics for nearly fifteen years. With CW uranium beams of 1-10 pnA intensity from the ATLAS accelerator at Argonne National Laboratory, a beam energy range matching the GSI experiments and a highly efficient apparatus providing a kinematically complete description of produced electrons and positrons APEX¹ was expected to provide a qualitative improvement in the understanding of this phenomenon.

In the initial operating period a program was begun to provide an accurate Monte Carlo (MC) representation of APEX. This representation has been extensively checked with radioactive source and heavy-ion beam studies.

A very stringent test of both APEX functioning and its MC representation has been the detection of Internal Pair Conversion (IPC) electron-positron pairs resulting from Coulomb excitation (in 5.95 MeV/u ²⁰⁶Pb on ²⁰⁶Pb collisions) of the 3^- level in ²⁰⁶Pb followed by subsequent IPC of the E1 transition to the 2⁺ first excited state. Since the IPC occurs in one of the moving collision partners the electron and positron detected energies must be Doppler corrected to narrow the line in the reconstructed electron-positron pair sum-energy spectrum. Because the identity of the source ion cannot be uniquely determined on an event-wise basis the Doppler correction is made with both velocity assumptions (one correct and one incorrect) for each event, resulting in a broader structure beneath a narrow peak, which is indeed quantitatively confirmed in the APEX electron pair sum-energy spectrum.

In a detailed systematic study of the ²⁰⁶Pb data various cuts were made on both the heavy ion kinematics and lepton kinematics (opening angle) which should result in specific changes to both the peak-to-background ratio and the Doppler-corrected peak structure in the pair sum-energy spectrum. The agreement between measurement and simulation was found to be excellent.

A further result of the ²⁰⁶Pb study was the first experimental determination of the branching ratio for IPC of (4±1) × 10^-5, in agreement with the theoretical value of 4.17 × 10^-5. This detailed study of the IPC in ²⁰⁶Pb clearly demonstrates that APEX has adequate sensitivity to observe narrow structures such as those reported by the GSI experiments and to elucidate their kinematic properties.

APEX has now been operational for two years. It has been thoroughly calibrated and has accumulated a very sizable data inventory for the HI collision systems U+Th and U+Ta (7000/µb for U+Th and 24,600/µb for U+Ta). No statistically convincing narrow structures have been observed by APEX in either the positron singles or electron-positron pair coincidence spectra.²

We conclude that the APEX results disallow the presence of any narrow structures in sum-energy spectra at a level which falls at least one order (~40 for a strong HI energy dependence scenario and simple wedge cut), if not two orders (~300 for no HI energy dependence and electron-positron pair kinematic restrictions to a back-to-back decay scenario) of magnitude (99% CL) below the levels reported by the GSI experiments. The conclusion seems inescapable that there is unresolvable disagreement between the APEX and GSI experimental results.