We report the first observations of beam losses due to bound-free pair production at the interaction point of a heavy-ion collider. This process is expected to be a major luminosity limit for the CERN Large Hadron Collider when it operates with 208 Pb 82 ions because the localized energy deposition by the lost ions may quench superconducting magnet coils. Measurements were performed at the BNL Relativistic Heavy Ion Collider (RHIC) during operation with 100 GeV=nucleon 63 Cu 29 ions. At RHIC, the rate, energy and magnetic field are low enough so that magnet quenching is not an issue. The hadronic showers produced when the single-electron ions struck the RHIC beam pipe were observed using an array of photodiodes. The measurement confirms the order of magnitude of the theoretical cross section previously calculated by others. DOI: 10.1103/PhysRevLett.99.144801 PACS numbers: 29.20.Dh, 25.75.ÿq When fully-stripped heavy ions of atomic numbers Z 1 , Z 2 are brought into collision at the interaction point (IP) of a collider, a number of electromagnetic interactions are induced by the intense fields generated by the coherent action of all the Z 1;2 charges in either nucleus (for a review, see Ref.[1]). Some of these ''ultraperipheral'' interactions have much higher cross sections than the hadronic nuclear interactions that are the main object of study. Among them, the bound-free pair production (BFPP), sometimes known as electron capture from pair production, occurs when the virtual photon exchanged by the ions converts into a pair, and the electron is created in an atomic shell of one of the ions:The resulting one-electron atoms have a slightly larger magnetic rigidity than the original bare nucleus.(Magnetic rigidity is defined as p= Qe B for a particle with momentum p and charge Qe that would have bending radius in a magnetic field B). Since the transverse recoil is very small, this ''secondary beam'' will emerge at a very small angle to the main beam. However, it will be bent and focused less by the guiding magnetic elements and may be lost somewhere in the collider ring.It has long been known that this process, together with electromagnetic dissociation of the nuclei, could be a major contribution to the intensity and luminosity decay of ion colliders [2,3]. It was realized more recently [4,5] that, in certain conditions, the BFPP beam will be lost in a well-defined spot, initiating hadronic showers in the vacuum envelope of the beam. The resulting localized heat deposition could induce quenches of superconducting magnets. Detailed calculations have been given elsewhere for 2:76 TeV=nucleon 208 Pb 82 operation of the LHC at CERN [5][6][7][8]. The consequent luminosity limit is expected to occur at a level close to the design performance. It is therefore vital to test our quantitative understanding of the features of the BFPP process in order to ensure safe operation of the LHC, uninterrupted by lengthy quenchrecovery procedures. BFPP has been measured in fixed target experiments [9-11] at lower energy but not,...