Volume 64B, number 4 PHYSICS LETTERS 11 October 1976 THE TRANSIENT FIELD IN IRON FOR 13 C K. DYBDAL, J.L. EBERHARDT and N. RUD Institute of Physics, University of Aarhus, DK-8000 Aarhus C, Denmark Received 3 September 1976 The spin precession for the fixst j~r= 5/2 + state in laC recoiling with an initial velocity of o/c = 0.044 into magne- tized iron has been measured. A window frame of polycrystalline iron was used to avoid demagnetizing fields. Fringing fields, causing beam-bending effects, are negligible, which reduces the measuring time by a factor of four. The meas- ured precession agrees with the trend observed for other light nuclei at high recoil velocities, but disagrees with the re- sult of a recent experiment on 12C. Recently, large transient magnetic fields deviating from the predictions of the Lindhard-Winther theory [1 ] by more than a factor of three have been found for light nuclei recoiling in polarized iron at high re- coil velocities [2-8]. The anomalies have been explained [6] by capture of polarized Fe 3d electrons into inner s-shell vacancies in the moving ion. The present experiment is the first in a series of ex- periments where the velocity dependence of the tran- sient field for 13C recoiling into polarized iron is studied. The 3.85 MeV level (jTr = ~+) of 13C, which has a mean life of r m = 11.8 -+0.6 ps [9, 10], was populated by the 2H(12C, p)13C reaction. A bombarding energy of 15.5 MeV, corresponding to a strong resonance in the yield [11], leads to a mean recoil velocity of o/c = 0.044. The fact that outcoming protons were not detected causes a spread in the recoil velocity of A(u/c) = +0.004. The 170 keV 3,-rays from the E1 transition between the ~+ state and the 3- state at 3.68 MeV were detected in two 70 cm 3 Ge(Li) detectors at +135 ° with respect to the beam direction and at a distance of 6 cm from the target. For normalization of the 7-ray yields, a 10 cm × 15 cm NaI(T1) detector was placed at 0 °, 40 cm away from the target. The NaI(T1) detector was gain-stabilized. To account for dead-time losses in the two Ge(Li)ADC's (which losses could be different for the two directions of the applied magnetic field), the counting in the NaI(T1) channel was interrupted as long as the ADC's generated a busy-signal. The deuterium targets were produced by evapora- tion of 530/ag/cm 2 titanium onto the iron backing in a deuterium atmosphere. In this way, a deuterium thick- ness of 13/ag/cm 2 was obtained. Despite the fact that F- z r,,. z o N W =E I I I ~ I I I -Ioo o lOO AMPERETURNS I Fig. 1. Hysteresis curve for the bulk magnetization of the Fe polycrystalline frame as obtained from an electromagnetic- induction measurement. the 12C4+ beam current was kept below 150 nA to pre- vent excessive heating of the target, a gradual decrease in yield due to escape of deuterium was observed. As a ferromagnetic backing, a 60 mm X 33 mm poly- crystalline pure t iron window frame was used with 6 mm wide and 2 mm thick legs. A coil of 560 windings was wound around one of the legs. The advantage of this geometry is that no demagnetizing fields will occur in the iron, which makes possible to magnetize the frame with a low applied field. The iron was annealed at 750°C for 80 min and cooled very slowly. A similar geometry ~" Iron with a carbon content of < 0.1%, manufactured by Dom- narfvets Iron Works, Sweden. 414