Instrumented Nanoindentation (NI) has been a valuable tool to measure mechanical properties such as hardness and elastic modulus for various materials over the last 20 years. As a further development of hardness testing, it served as an appropriate method to determine mechanical properties, especially on thin films, on a small amount of sample or from special regions on a sample. NI of polymers have only been a marginal phenomenon over the last few years. The basic method in NI evaluation is the O&P method, which assumes elastic material relaxation after deformation. A polymer relaxes viscoelastic after deformation which makes the interpretation of NI data complex. In this work, the emphasis lies on the investigation of polymers by NI. The main goals of this cumulative work are:
(I)Evaluation of the O&P analysis and creep method in combination with material modelling on an isotropic (amorphous) material and compare it to a common macroscopic test.
(II)Expansion of the evaluation from (I) to a more complex semi-crystalline system to evaluate the influence of morphology on the NI results.
(III)Bring results from (I) and (II) into application where the occurrence of polymeric aging are detected by NI.
(IV)Introduce an innovative sample preparation method to investigate polymeric multilayers in cross-section.
After a short introduction to NI of polymers a state-of-the-art section reviews literature ranging from the mechanical behaviour of polymers to NI methods and techniques. In the first publication NI was used to examine the viscoelastic properties of poly(methyl methacrylate) (PMMA) as an amorphous polymer model. An evaluation combining adhesive contact and empiric spring-dashpot models was applied to obtain the instantaneous elastic modulus E0 and the infinitely elastic modulus E∞ from NI creep curves. The value of E0 was compared to moduli obtained with atomic force microscopy-based NI (AFM-NI) and macroscopic compression tests. Furthermore, the elastic modulus was evaluated by the O&P method for the NI and AFM-NI results. Comparison of the elastic modulus E0 from the creep measurements of NI and AFM-NI to compression tests revealed good agreement of the results. However, only the O&P-based AFM-NI results yielded to lower values.
In paper two the material selection was expanded from amorphous to semi-crystalline polymers, as the influence of morphological structures such as spherulites or crystal-lamellae on localized NI depth-force behaviour is discussed controversially in literature. The main objective of paper two was to determine the influence of crystalline zones of semi-crystalline POM (Polyoxymethylene) on NI results. A POM tensile bar was investigated by NI at various positions of the cross-section. Areas at the edge of the sample had a lower modulus than areas in the middle of the cross-section. This agreed with polarized light microscopy results showing a skin layer with less crystallinity close to the sample edge. Semi-crystallinity influenced the NI results obtained for POM. The mean of the modulus distribution over the POM cross-section was in good agreement with macroscopic compression test results.
In paper three a fast and sensitive method to detect polymer aging at an early stage was developed by NI. For this purpose, a commercially available transparent and 50 micrometer thick polyethylene terephthalate (PET) film was aged under different artificial conditions. The evolution of mechanical properties with rising aging time was investigated via NI methods established in paper one and two and macroscopic tensile testing. Physical aging was monitored by the first heating of differential scanning calorimetry (DSC) and chemical aging was studied with gel permeation chromatography (GPC). NI data was compared to tensile test data with good agreement between the results on the nano- and macroscale. A correlation between GPC data and NI creep data indicated a primarily chemical aging of the present PET films. Paper three states that NI is an appropriate method to determine the degradation of PET at an early stage.
In paper four NI on thin, polymeric multilayer film cross-sections were studied. Here NI violated assumptions of semi-infinite and homogeneous samples because NIs were always near free edges and heterophase interfaces. The structural compliance method was used to correct NI results for specimen-scale flexing and edge effects, where the effect of both edges and specimen-scale flexing is to introduce a structural compliance (Cs). The Cs showed a strong position dependence applied to a polymeric multilayer. The effect was larger near the edges and layer interfaces. The Cs correction had little effect on the hardness values; however, the influence on measured elastic modulus was significant. The corrected modulus values tended to be higher than the uncorrected ones in the stiff layers.