We propose to continue our last experimental run (3-15-61), to interferometrically determine neutron coherent scattering length b_c for silicon to within a few parts per million, using a dual non-dispersive sample. During our 3-15-61 experimental run, we lost one week due to a vacuum pump inadvertently placed on the rooftop of the S18 cabin by the ILL operational staff. The pump could not be shifted during the reactor cycle. We could hence only try to isolate its vibrations from the cabin. The observed phases therefore fluctuated wildly. We could nevertheless select a few of the several hundred recorded interferograms and determine silicon b_c to within 30 parts per million. If in our next run, all sources of vibration and thermal variations are scrupulously kept away from S18, we can achieve the proposed ppm precision.