A prerequisite for the commercialization of a drug is preservation of its integrity during its entire lifecycle. However, interactions do exist between the liquid content and the primary packaging resulting in the degradation of the functionalities of the former. The existing glass primary packaging is not adjustable with regard to the active molecule to be stored. A solution to this major problem in the health sector is the application of a barrier coating on the inner surface of bottles. Such coatings aim at fulfilling the targeted requirements (resistance to acid or alkaline solutions as well as formulations with aggressive buffer, reduced cations leaching, inhibition of glass delamination, etc.) and at the same time they must be integrated in the well-established industrial process for the production of the glass bottles.

          Our working hypothesis is that silica SiO2 and silicon oxynitrides SiOxNy are particularly inert materials which, in the amorphous and in the crystalline states present remarkable barrier and mechanical properties. An example is silicate oxide glasses for the conditioning of nuclear wastes. High quality, dense SiO2 coatings can be produced by dip coating, sol gel or colloidal techniques, and alternatively by atomic layer deposition, ALD and by plasma enhanced, PE or plasma impulse, PI or thermal CVD. Wet techniques are multi-steps, involve large amounts of effluents and often meet difficulties to process thin and dense films. PICVD is used by SCHOTT AG to apply SiO2 on tubular pharma vials. This complex process presents low productivity (vial by vial treatment) and does not allow large scale commercialization. Breaking away from the above-mentioned methods, our approach is to use thermal CVD which is simpler since it can be straightforwardly implemented at the back end of the production lines of bottle manufacturing. Provided that the involved chemistry is controlled and the process is appropriately designed, environmentally compatible thermal CVD presents a major industrial advantage as it can uniformly coat the surface of a batch of bottles in one step at a moderate cost. In a recent collaboration, CIRIMAT/SURF, LGC and SGD Pharma demonstrated that this technology can treat very complex substrates such as the confined, internal surface of containers and developed a proprietary CVD process able to uniformly coat the inner surface of pharmaceutical glass bottles by an amorphous alumina film. Nevertheless, an important concentration of leached Al ions from the film was revealed, which is unacceptable in the pharmaceutical context, but this achievement provides the proof of concept of our approach. For this reason, a new collaboration was recently initiated on the engineering and upscaling of a SiO2 CVD process.

          This B2B collaboration does not include the investigation of the process-structure-properties relation in CVD processed SiO2 based materials for barrier applications. This is the objective of HEALTHYGLASS. The feasibility of processing SiO2 and SiOxNY films by CVD has already been demonstrated, but in different perspectives such as the production of hard materials or dielectrics. Our research on SiO2 will focus on the impact of the process chemistry (based on tetraethyl orthosilicate, TEOS and various oxidizing agents, O2, H2O, O3) and of the operating conditions on the process and films characteristics. A multi-scale characterization will evidence the impact of the synthesis routes on the silica network topology, e.g. the presence of siloxane bridges, on the short range order and on the porosity of the films. This part of the work is a priority of the project in order to transfer the findings to the CVD process under implementation and it will be used as a baseline for the processing of SiOxNy films. In SiOxNy, the substitution of O2- by N3- results in density increase, giving rise to improved chemical inertia and to tunable mechanical properties. SiOxNy films from a thermally activated TEOS-based chemistry at temperatures lower than 550°C can be met through the use of hydrazine N2H4 or from O3 in presence of either N2O or NH3, targeting the production of highly reactive oxygen radicals. The deposition processes will be optimized by synergistically combining experimental study and computational fluid dynamics-based modelling. Chemical mechanisms and kinetic laws will be developed by analyzing the chemical composition of the gas at the exit of the CVD reactor.

         The structure of the coatings will be investigated by state of the art spectroscopic and microscopic techniques. The topology of the films at the atomic scale will be investigated by 29Si (and 17O for targeted samples) nuclear magnetic resonance (NMR). The structure of the coating surface will be probed by dynamic nuclear polarization (DNP) NMR. The mismatch between the requested quantity of matter for fine NMR characterization and the throughput of the CVD process will be tackled through careful sample preparation, already successfully demonstrated by the applicants. SEM, AFM, EPMA, spectroscopic ellipsometry, STEM, and EFTEM will be used for the chemical and structural characterizations at the micro and the nano scales. Alternative spectroscopic techniques (XPS, EXAFS, XANES) will be used to clarify questions of the atomic scale structure and mid-range order that may arise in the course of the project. The FTIR spectroscopy will be employed specifically for SiOxNy characterization and for revealing N/O substitutions. The development of residual stress in SiOxNy films which impacts the durability of the material will be tackled (a) by identifying the optimum SiOxNy composition, (b) by varying the O/N ratio along the thickness of the film, and (c) by developing multilayer structures. Stiffness, hardness, Young’s modulus and adherence of the coatings will be investigated by combining nanoindentation at the scale of several hundred nanometers with Atomic Force Microscopy (AFM) and nanoscratch tests. The chemical durability of the coated glass will be tested in conditions defined by the European Pharmacopeia and in accelerated and severe storage conditions (acid or alkaline solutions, with different buffers, complexing agents, etc.) and will be quantified through the determination of the concentrations of extractables by ICP spectrometry, and of the surface alteration at the micro- and atomic-scale thanks to the previously mentioned techniques, completed by 1H and 1H/29Si correlation NMR experiments, SIMS/RBS and APT. Last but not least, the determination of chemical and physical (surface energy, zeta potential, etc.) characteristics of the extreme surface will provide elements for the comprehension of the interactions of the film within its environment, in particular in biologic media.