Super-conformal coating and filling of high aspect ratio recessed structures by two-molecule CVD

Complete filling of a deep recessed structure with a second material is a challenge in many areas of nanotechnology fabrication, including MEMS devices, metallization and shallow trench isolation (STI) in integrated circuits. For structures with straight sidewalls, uniform coating methods – including chemical vapor deposition (CVD) or atomic layer deposition (ALD) – typically form a void or region of low density (seam) along the centerline of the feature because the transport of precursor molecules to the bottom becomes rate limiting. These defects undermine device properties and methods are needed to afford complete filling. One approach is to taper the two sidewalls slightly inwards. However, the requirement for tapering is an undesirable constraint both for device design and fabrication. A more general approach is to develop a process for superconformal film growth, in which the growth rate increases with depth. Previously, only specialized processes, including iodine catalyzed Cu growth or high-density plasma CVD, have provided super-conformal coating and complete filling. In this dissertation, I report a newly discovered and general method to achieve super-conformal coating suitable for two-molecule CVD at low temperatures, which makes use of the difference in diffusivities in combination with the competition for surface adsorption sites between the molecules involved. The partial pressures above the opening are set such that the slower diffusing species is in excess, which reduces the growth rate below the peak value. Within the structure, film growth on the sidewalls reduces the partial pressures; the pressure drop is greater for the slower diffusing species, such that the ratio of partial pressures progressively shifts towards a larger growth rate. However, the position at which the peak rate occurs must not be above the bottom of the structure, otherwise the growth rate will fall at greater depths, leading to incomplete filling. We demonstrate and quantify the superconformal coating effect for the CVD of MgO films using Mg(DMADB)2 as the precursor and H2O as the co-reactant at 220 °C. The growth kinetics on planar substrates fit to a first-order adsorption-reaction model. We use a diffusion-reaction formalism to derive a general theory of superconformal growth as a function of AR, species diffusivities, and variation of growth rate vs. partial pressures. The theory predicts, for the MgO system, the possibility of superconformal growth in trenches with AR ≤ 20. We demonstrate the effect experimentally for a macroscopic trench with AR = 18, and for a microscopic trench of AR = 9. We also identify other systems that are promising candidates for superconformal growth. Filling is a dynamic process where the trench progressively narrows with depth; this reduces species transport and may eventually move the partial pressure ratio out of the regime for superconformal coating. We therefore derive two theoretical models that can model and predict the possibility for filling in a V-shaped structure. First, we recast the diffusion-reaction equation for the case of a sidewall with variable taper angle. This affords a definition of effective AR, which is larger than the nominal AR due to the reduced species transport. This model shows that the critical (most difficult) step in the filling process occurs when the sidewalls merge at the bottom to form the V shape trench. For Mg(DMADB)2 / H2O system and a starting AR = 9, this model predicts that complete filling will not be possible, whereas experimentally we do obtain complete filling. We then hypothesize that glancing-angle, long-range transport of species may be responsible for the better than predicted filling. To account for the variable range of species transport, we construct a ballistic transport and reaction model. This incorporates the incident flux from outside the structure, cosine law re-emission from surfaces, and line-of-sight transport between internal surfaces. We cast the transport probability between all positions into a matrix that represents the redistribution of flux after one cycle of collisions. Matrix manipulation then affords a computationally efficient means to determine the steady-state flux distribution and growth rate for a given taper angle. The ballistic transport model predicts a deeper position for the peak of the super-conformal growth rate than the diffusion-reaction model, and successfully explains the observation of complete filling. These models can be used to predict the behavior of any system given a small set of kinetic coefficients to describe the growth rate. This dissertation also reports the growth of a variety of oxide thin films on planar substrates using DMADB-based precursors, synthesized by Professor G. S. Girolami, which contain Ti or rare earth (RE) metal centers. Similar to the results for Mg(DMADB)2, these precursors react with water to yield high quality oxide films. Ti(DMADB)2 affords mixed-phase TiO2 film of high purity at 350 - 450 °C. Y2O3 film containing ~ 4 at. % boron is deposited on Si (100) substrates at 230 - 300 °C. Erbium oxide does not nucleate well on silicon, but deposits well over pre-deposited MgO. RE doping, which is important for the fabrication of devices based on luminescence, is investigated. In-situ Tm doping into MgO and Er doping into Y2O3 are achieved. In addition, we explore the use of the Mg(DMADB)2 precursor for film growth in ALD mode with H2O as the co-reactant. We determine that ALD can be performed in the temperature window 165-210 °C, with a higher growth rate per cycle than has been reported for any other Mg-bearing precursor.