An attempt has been made to evaluate the kinetics of the thermal and combined thermal/radiative degradation of a peroxide cured polydimethylsiloxane elastomer (PDMS), and a vinylidene fluoride-hexafluoropropylene elastomer (Viton E60-C),by chemical stress relaxation measurements. It was observed that the PDMS elastomer degraded thermally by hydrolytic scission of the main chain and chain reformation by condensation of the silanol chain end groups. However,predominant chain reformation was observed during thermal degradation studies and it was believed that this was a consequence of the reformation of chains cut during the test,and also the reformation of chains cut prior to the test in the post cure operation. These two components of the observed chain reformation response were subsequently rationalized by a model. Thermal degradation of the Viton E60-C elastomer appeared to take place by hydrolytic scission of the amine crosslinkages, and subsequent crosslink reformation by condensation,indicating that the material had not been subjected to a suitable post cure treatment in order to remove water generated by the vulcanization reaction. At temperatures above 200°C predominant crosslinking was observed and this was attributed to the presence of a concurrent reaction which leads to the formation of ring structures. Thermal/radiative degradation studies indicated that temperature and radiation had a synergistic effect on the overall rate of induced chain scission in the PDMS elastomer but the mechanisms responsible for this phenomenon have yet to be established.

Current stress relaxation measurements in compression and the associated equipment are not reliable enough to carry out prolonged tests, especially at high temperatures and in liquid environments. The methods of measurement are outlined with the greatest flexibility in BS 903 Part 34, yet the consideration given to overcome the difficulties encountered is limited. The objectives of this sponsored programme were to examine the best available equipment, to generate a set of accurate data including some at relatively high temperatures; and to formulate a technique to extrapolate these high temperature results to service temperatures.

Cationic surfactants of general structure [RNH2(CH2)3NH3]2+ 2[R'COO]- can be used in rubber processing as multifunctional additives (MFA) which act as processing aids, accelerators and mould release agents. The effects of MFA on gum rubber, carbon black and silica filled compounds were investigated and attempts have been made to explain the mechanism of their action. This was done by studying mixing, vulcanisation behaviour, rheology, and mould release. The principal MFA used was Akzo-Nobel EN444.

An NMR technique has recently been developed to determine the crosslink density in each phase of a polymer blend. This work uses the NMR method to study natural rubber (NR)/acrylonitrile butadiene rubber (NBR) blends, in order that the cure system used to vulcanize such blends can be optimized. A standard injection moulding cure system for NR/Perbunan N1807 (18% acrylonitrile content) gave much more crosslinking in the NBR phase. To reduce this imbalance a less polar analogue of the accelerator was used, resulting in a more even distribution of crosslinks but still with a bias towards the NBR.

Natural rubber-polypropylene (NR/PP)blends have been studied under dif ferent conditions vis. processing parameters, mixer and at different ratios. It has been found that the requirement of a typical thermo­ plastic rubber can be achieved at a ratio of NR to PP of 60:40. It was found that chemical interaction between the two phases, which may take place as a result of mechanochemical reaction, could not be achieved, and the thermoplastic rubber produced was essentially a physical blend. The role of free-radical initiators such as benzoyl and dicumyl peroxide, on the behaviour of NR/PP(60/40) blends was investigated. It was shown that dicumyl peroxide was more effective than benzoyl peroxide in im­ proving mechanical properties of blends. Again no chemical reaction between the two phases could be achieved. Crosslinking reaction in the rubber phase may be responsible for the improvement in tensile properties in the presence of dicumyl peroxide. Ethylene-propylene terpolymer (EPDM)was also used in an attempt to improve the adhesion between the two phases. However, the results shown in this work con­ firm that no such interaction takes place in the presence of EPDM and the overall properties of the blend is lower than in the case of dicumyl peroxide. The performance of different ratios of natural rubber-polypropylene (60/40) composition as a solid phase dispersant (SPD)in polypropylene-polyeth­ ylene (1:1) blends was investigated. The behaviour of the above system was compared with that of EPDM when used as SPD in the same blends; NR/ PP offers better tensile properties while EPDM contributes mainly to impact performance of the PP/PE blends. Photo-oxidative stability of these blends in the presence of NR/PP and EPDM was also investigated. In general, EPDM system offered better overall stability. However, in both cases (i.e., NR/PP and EPDM systems), a synergistic system based on a combination of peroxide decomposer, UV-absorber/screen and free-radical scavenger e.g., HOBP/Cyasorb 1084/Irganox 1076, was shown to be the best photo-antioxidant system. The effect of partial replacement of fresh natural rubber by reclaim rubber (RR)on the properties of NR/PP (60/40) blends was studied. It was found that properties of natural rubber-polypropylene (60/40) blend could be maintained by replacing up to half of the fresh rubber by re­ claim rubber. The behaviour of this composition (i.e.,(NR/RR)/PP at a ratio of 30:30:40)as a solid phase dispersant in PP/PE(1:1) blend was investigated. It was shown that although this system offers a better photo-oxidative stability than a blend which is based entirely on fresh rubber, its impact performance is inferior to blends containing either NR/PP (i.e., no RR) or EPDM as SPD's.

The research evaluated the means in which mild variations in the polymer properties could be modified during the mixing process so that the final mixed compound met the desired rheological parameters. The initial work developed a model to measure the viscosity of the material in the internal mixer during the mixing process. The model assumed the mixer could be treated as two concentric rotary viscometers. The model had a rotor speed dependence which was caused by the assumption not being valid. However, this led to the work continuing in a modified form. It was felt that at a given rotor speed and temperature the torque measured on the rotors would be proportional to the viscosity of the material. This was found to be correct and therefore work continued using this principle. Programs were developed for internal mixers with variable speed rotors and fixed speed rotors. The programs were evaluated using natural rubber with different viscosities and were found to be capable of modifying the polymers to obtain a final mixed compound to a target viscosity. The control program was also used to mix a series of batches at different processing conditions. The rheological properties of these batches were compared to a similar group mixed to a specification based on unit work. The result was the control system gave more rheologically uniform material.

This study is to investigate the fabrication of filled polyurethane elastomer tiles by the reaction injection moulding (RIM) process. The base matrix is composed of a crosslinked polyurethane elastomer formed by the reaction between a diisocyanate (MDI) and a polyol (polytetramethylene glycol) and crosslinked by trimethylol propane (TMP). The two fillers investigated were barium sulphate and Expancel which were used to dissipate and scatter sonic waves in acoustic damping applications.

Thermoplastic rubbers have been shown to have unusual solid state properties which must derive from the structure of the melt prior to solidification. The melt phase has been studied in some detail. The molecular architecture of these block copolymers comprises of hard segments (usually polystyrene) connected by a flexible rubbery chain (polybutadiene or polyisoprene) in a linear or radial structure. Their flow characteristics have been studied and the results correlated with measurements in the solid state. They have been modelled mathematically based on two particular theoretical models. A liquid phase transition was recorded for both models, with appreciable reduction in the apparent activation energy of flow above this temperature. The quality of the domain structure depends on the continuity of the polystyrene phase and has been measured by the stress at yield and by the optical birefringence. A change in response was associated with the liquid-liquid transition.

Fatty diamine cationic surfactants are a recent entrant into the rubber mix formulation and have significant effects on the processing and properties of rubber compounds. Six new fatty diamine cationic surfactants, based on oleic acid, linoleic acid, rubber seed oil, 1,2-ethane diamine and 1,6-hexane diamine, were synthesised; and a simple method of their isolation and purification was developed. The effects of the experimental surfactants as combined activator and accelerator in rubber compounds were investigated. The rubber compounds containing them exhibited high activation energy of vulcanisation; and the surfactants functioned effectively as both activator and accelerator in SBR mixes. One of the new salts (l,2-ethane diamine dioleate, EDDO) and two commercial materials (EN444 and Duomeen TDO) were applied in standard NR and SBR mixes, and their effects on vulcanisation, rheological and vulcanisate properties investigated. Mixes containing the multi functional additives processed better, cured faster and had improved physical/mechanical properties over control mixes containing the same level of fillers. NRJEDDO mixes containing Si69 require a twostage mixing technique for further improvement in vulcanisate properties; similar SBR mixes were only slightly affected. Compounds were prepared in a Farrel (BR model) Banbury and a Francis Shaw KI Intermix internal mixers. Cure measurements were made on a Wallace Precision Cure Analyser and Monsanto (ODR) curemeter. Tensile properties were measured with a Hounsfield electronic tensometer (model SOOL). Filler dispersion was measured by SEM and "dark field" image analysis. Carbon black (N330)-filled NR compounds containing EDDO exhibited higher shear viscosity, higher activation energy of viscous flow and higher extrudate swell than a control compound which contained the same level of RAF black. Silica-filled mixes containing the MF A exhibited higher extrudate swell but lower shear viscosity than the control mixes containing the same level of silica filler. Melt flow measurements were carried out using a Davenport Capillary Rheometer and a Negretti TMS Rheometer. The effects of temperature and die length on shear viscosity, extrudate swell, melt flow behaviour index, n', and extrudate defonnation were investigated. The combined effect of MFA (EN444) level and mixing energy on properties of carbon black (N330)-filled natural rubber compounds was investigated by a factorial design experiment. No dependence of vulcanisation rates and elastic modulus on mixing energy was observed. Vulcanisation rates, mixing time and modulus (M300) showed a decrease with increase in MF A level. Other vulcanisate properties, including tensile strength, tear strength and j hardness, are optimised at MF A level of 2phr and above this level suffered significant deterioration.

Research work has been carried out to critically study parameters affecting peroxide-cured conductive silicone rubbers, using various carbon blacks as the conducting phase.