Although polymers represent the substrate of choice for flexible devices such as solar cells and OLEDs because they are lightweight, flexible, transparent, inexpensive, and compatible with roll-to-roll processing, they have the drawback to be highly permeable to moisture and oxygen. This poses severe limitations to the performance of the flexible device. This drawback is presently addressed by depositing inorganic (Al2O3, SiO2, Si3N4) thin film barrier layers on the polymer substrate. Despite their impermeable bulk counterpart, the (water vapor) permeation through single barrier layers is driven by several paths, which include nm-sized pores as well as substrate/process induced macro-defects. It is, therefore, of paramount importance to determine and control the density of the macro-defects, as well as to control the inorganic barrier layer microstructure, defined by its (open) nano-porosity. The state-of-the art in barrier layer technology applied to polymer substrates as well as to the direct encapsulation of the (flexible) device is a µm-thick multilayer consisting of inorganic barrier layers decoupled by organic interlayers. This encapsulation solution against water permeation into the device can virtually guarantee a device lifetime of ten years. Although several approaches have been followed in engineering the multilayer, there is still a debate on the effective role of the organic interlayer in affecting the multilayer barrier properties. It is generally considered that the organic interlayer acts as smoothening layer allowing the decoupling between macro-defects either present on the polymer substrate or in the inorganic barrier layer. It is also hypothesized that the organic interlayer infiltrates into the nano-pores present in the barrier layer, therefore affecting the barrier itself at microstructure level. However, this hypothesis has neither been followed by any experimental evidence, nor it has been investigated in the case of organic interlayer deposition methods other than polymerization from its liquid phase. This thesis work aimed to gain insight into the role of the organic interlayer in affecting the multilayer barrier properties. A model system, based on a siloxane chemistry, has been adopted in which the multi-layer is developed by means of two vacuum deposition techniques, i.e. PECVD for the inorganic SiO2-like barrier layer and initiated-chemical vapor deposition (i-CVD) for the poly(V3D3) organic interlayer. This latter allows the polymerization process to develop organic films with full retention of the monomer chemistry. A novel deposition setup has been, therefore, developed to implement both deposition processes in a vacuum chamber, equipped with in situ real time diagnostic tools, such as spectroscopic ellipsometry (SE). As first research step, the i-CVD polymer growth has been studied in situ by means of SE which allowed following all stages of the deposition process from the initial monomer adsorption to the linear film growth and to the thickness losses due to the presence of unreacted monomer units at the end of the deposition process. Moreover, in situ SE measurements allowed characterizing the thickness losses as bulk- related phenomenon and brought new information on the polymerization process which propagate not only at the surface of the growing layer but also in the bulk. Furthermore, a correlation has been made between specific process parameters (i.e. the monomer surface concentration) and the deposition of stable, highly cross-linked polymer layers (i.e. exhibiting no thickness loss upon evacuation). This has allowed to define a process parameter window, i.e. PM/Psat, followed in situ by SE, which controls the deposition of high quality poly(V3D3) layers. The follow-up studies of the i-CVD polymer growth on SiO2-like moisture permeation barrier layers, performed by means i-CVD monomer (V3D3) adsorption/desorption isothermal studies, have highlighted the filling/infiltration of the i-CVD monomer into the open nano-defects/porosity of the SiO2-like layer underneath. This result has, therefore, provided support to the above-mentioned hypothesis on the infiltration of the organic interlayer into the nano-porosity of the SiO2-like barrier layer underneath. Finally, the contribution to the improvement of the barrier performance of the PE-CVD/i-CVD deposited multilayer due the filling of the PE-CVD deposited SiO2-like layer nano-pores has been studied with respect to the smoothening/decoupling effect of the SiO2-like layers macro-defects. Ca test measurements allowed discerning between the water permeation through the macro-defects/pinholes and the permeation through the matrix. The effect of the filling/infiltration of the SiO2-like layer nano-defects has been studied as function of intrinsic porosity of the SiO2-like layer (i.e. as function of the intrinsic WVTR values). It is concluded that the effect of the SiO2-like nano-defect filling by the poly(V3D3) is effective only for SiO2-like layers initially exhibiting an intrinsic WVTR value > 10-3 g m-2 day-1. The above-mentioned results, in combination with an evaluation of the local macro-defects prior and upon deposition of the organic interlayer, and a comparison with parallel studies reported in literature, allow to conclude that the main contribution of the i-CVD layer in improving the multilayer barrier properties is given by the smoothening/decoupling of the macro-defects. The SiO2-like layer microstructure characterization has been carried out on layers deposited on a silicon substrate by means of ellipsometric porosimetry measurements which allowed discerning between the different residual open porosity of the deposited SiO2-like layers. The same characterization in terms of refractive index and residual open porosity, however, should be also carried out in case of polymers as substrates, for example for barrier-on-foil applications. However, polymeric substrates often show optical anisotropy and the proper determination of the barrier layer optical constants can be achieved only by a proper optical characterization of the substrate. In order to perform the microstructure characterization of barrier layers deposited on polymers, the polymer (poly(ethylenenaphtalate), PEN) anisotropy has been characterized by means of the Generalized and Spectroscopic Ellipsometry combined approach. This approach has allowed defining the optical constants of a SiO2-like barrier layer deposited on PEN.
|Qualification||Doctor of Philosophy|
|Award date||12 Jun 2012|
|Place of Publication||Eindhoven|
|Publication status||Published - 2012|