Various properties of the components and adhesives were modelled. The compatibility of the components were successfully incorporated into an extended Fox equation to predict the glass transition temperature. The peel strength of the adhesive was modelled in terms of the rheological properties of elastic moduli and loss tangent values at different temperatures. A second model based upon the value of the loss tangent at room temperature was also broadly successful but deviations from predicted behaviour were observed which were attributable to failure of the adhesive joints by a mode not included in the model. The modulus of the adhesive was modelled on the basis of an extended mixture rule in which the extent of compatibility was identified by a parameter n. The value of n varied as a function of adhesive composition and temperature, indicating that the behaviour of the adhesives changed subtly as the compatibility of the phases changed. The value of the parameter could not be directly related to the morphology of the adhesive phases. Fourteen commercially available poly(ethylene-co-vinyl acetate) (EVA) copolymer samples were selected in which there was a systematic change in the melt index, amount of vinyl acetate, and degree of crystallinity. Various hot melt adhesives were made using these copolymers and a standard amount of wax and resin. The materials were examined using differential scanning calorimetry (DSC), oscillatory rheometry (both controlled strain and controlled stress), and transient (creep) rheometry. The adhesives were also investigated using a variety of industrial tests which included peel adhesion and tensile testing at four different rates, open and setting time, shear and peel stress resistance at elevated temperatures, and viscosity determination over a wide range of temperatures. Detailed thermal analysis and characterisation have provided a range of accurate and systematic data on all of the materials and in particular showed that the components of the adhesive did not merely act as a mechanical mixture but had a distinct compatibility. The controlled stress technique was found to more discriminatory than the controlled strain, due to the more precisely controlled heating and cooling of the sample during loading and evaluation. Other key differences between the techniques are attributable to the different thermal histories imposed upon the semi-crystalline adhesive components. Detailed analysis of the complex rheological curves showed several key factors. One of the most important was the modulus crossover temperature Tx which was shown to correlate well with the softening point of the adhesive, its open time, and the heat resistance under shear as determined by the shear adhesion failure temperature (SAFT). It was possible to construct a linear relationship between Tx and SAFT which allowed prediction of this key adhesive parameter. There was no significant relationship established between the softening point of an adhesive and its heat resistance, open time, or critical thermal characteristics, and the use of the softening point as a useful indicator of adhesive performance is contested. The open time was shown to be clearly influenced by the properties of the copolymer. The relationship between open time and melt index is complex and two competing mechanisms are thought responsible. These are the inability to fully wet the substrate for high molecular weights and resistance to complete substrate penetration by capillary effects for adhesives formulated with low molecular weight polymers. Both of these effects cause a reduction in open time. The cloud points of the adhesives were independent of the molecular weight but strongly affected by composition. Degree of crystallinity was also an influence at higher molecular weights. Cloud point correlated slightly with the onset of crystallisation as determined by DSC however differences are extremely small and the method was not deemed robust enough for widespread industrial application. Various properties of the components and adhesives were modelled. The compatibility of the components were successfully incorporated into an extended Fox equation to predict the glass transition temperature. The peel strength of the adhesive was modelled in terms of the rheological properties of elastic moduli and loss tangent values at different temperatures. A second model based upon the value of the loss tangent at room temperature was also broadly successful but deviations from predicted behaviour were observed which were attributable to failure of the adhesive joints by a mode not included in the model. The modulus of the adhesive was modelled on the basis of an extended mixture rule in which the extent of compatibility was identified by a parameter n. The value of n varied as a function of adhesive composition and temperature, indicating that the behaviour of the adhesives changed subtly as the compatibility of the phases changed. The value of the parameter could not be directly related to the morphology of the adhesive phases.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:337096 |
| Date | January 1997 |
| Creators | Doody, Paul David |
| Publisher | Coventry University |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://curve.coventry.ac.uk/open/items/aee7101d-7aef-41a0-a6a1-32d9877f92d1/1 |
Page generated in 0.002 seconds