The dynamic vacuum wasca. torr during film formation and the molecular precursors were heated to a temperature defined as (80"-120 "C) so that sublimation/distillation into the heated chamber occurred at a rate of ca. 0.1 g/ h. The temperature gradient within the heated furnace was determined to be less than 50 "C and the furnace temperature, T,, was maintained at 25 "C above the highest temperature at which avisible amount ofthe M(N(SiMe,),),compound passed through the hot wall reactor without decomposition. The volatile products of the deposition reaction were passed through a U-tube filled with glass beads immersed in I-N, and those that were condensed in this manner were analyzed by 'H and "C{'H} NMR and GC/MS. The glass reactor tube was cleaned with 20% HNO, overnight (ca. 12 h), 10% HCI for 2 h followed by extensive washing with deionized water, then acetone, and finally dried in an oven at 175°C for 16 h prior to use. Silicon(100) wafers (from Virginia Semiconductor, Inc.), quartz and glass plates were cut into 1 cm x 2 cm plates, washed with 10% aq. HF for 30 s, followed by immersion in deionized water (10 s) and acetone (10 s), prior to drying. The tube furnace was loaded with the substrates, glass, silicon and quartz alternating randomly throughout its length (ca. 30 cm). The source chamber was loaded (ca. 0.5 to 0.8 g) in a drybox, sealed, and attached to the tube furnace. The system was then evacuated and heated to the predetermined decomposition temperature, Td, for 1 h before the source chamber was opened, evacuated for 10 min, and heated to T,.The films were analyzed by XRD (employing a Scintag XDS-2000 diffractometer, Cu Ka, and SEMiEDAX employing a JEOL LSM 820 microscope with a Kevex detection system (Quantum Detector Kevex Delta 4). The EDAX data were analyzed using Quantex 6.2 software. This technique allows the detection of elements with atomic number 25 to within at best I at-%. The topography of the films was studied by SEM (JEOL 800) with a resolution of less than 1 pm. The quality of adhesion of the film to the substrate was evaluated by the qualitative "Scotch Tape test", wherein two strips of 3M Scotchtape, crossed in an "X" on the surface of the film, are pulled off and examined for particles. For a film to pass the "Scotch Tape test" no particles of the film are visually transferred from the substrate to the tape.