Blast loading with high pressure intensity propagates within a fraction of second after an explosion. Depending on the amount of energy and wave velocity released, blast loading is highly likely to cause substantial structural damage and leads to a total failure. Taking into account various interests and requirements in the protective structures, the investigation of damage behaviour and structural responses due to this extreme condition is therefore vital. The prime objective of this research is to develop an established framework of numerical modelling of reinforced concrete slabs and protective structures subjected to blast loading. Subsequently investigate the damage mechanisms, dynamic responses and post-failures. The scabbing, spalling, crater and shear plug are of particular interest, with special attention paid to the progressive fracture and its associated velocity. Three major modelling aspects were given the most attention in term of blast loading, material model and fracture modelling. The interaction between non-uniform blast loading and reinforced concrete slabs was modelled using the combined finite-discrete element method. The finite element method was incorporated with a rotating crack approach and discrete element to model the fracture onset and the dynamic post-failures. In the numerical modelling, a mapping method was employed to define blast pressure due to incompatibility of the Jones-Wilkins-Lee method with all compressible material models. The blast loading was determined based on cumulative loads of the incident overpressure, the reflected overpressure and the dynamic wind blast. The calculated blast loading was compared with that obtained from the US Army standard, TMS-1300/UFC-03-340. The Mohr-Coulomb and Von-Mises criteria were applied for the concrete and steel reinforcement respectively. Since the Mohr-Coulomb criterion in concrete can only produce continuum failure, the Rankine with fracture model was introduced to control tensile fracture failure. Meanwhile, a multi-scale simulation was applied to overcome the lack of constitutive material model for the ceramic composite layer. The multi-scale simulation is based on the linear boundary condition, employing a unit cell of ceramic composite with uniaxiallbiaxial loading and a nonlinear anisotropic brittle model. The comparison between numerical and experimental results shows a favourable agreement and gives a reliable prediction on the damage behaviour and structural responses. The fracture and post-failures. however, are still ambiguous and need for further investigation.

The microstructure and properties of alumina and alumina-matrix nanocomposites are strongly affected by the raw powders and the processing route. The significant effect of hydrothermal synthesis condition on the morphology of boehmite particle was observed. UF-SiC has more controllable oxidation behaviour for the preparation of nanometer SiC/mullite/alumina composites than P-SiC. Due to the crystallization and agglomeration of SiC at high temperature, oxidation at low temperature is suggested. The properties of the compacts prepared by pressure filtration were closely related to the state of the starting slurry, the particle size and morphology and the applied pressure. A desired slurry can be obtained by the controlling of pH value, the ionic strength and the addition of suitable dispersant. Due to the non-uniform distribution of the liquid pressure and solid pressure across the compact during pressure filtration, the density of the compacts is not uniform as well. Gradually increasing the pressure to the desired level is suggested in order to obtain a more homogeneous compact density. Cracking upon unloading after pressure filtration and during drying became severe for mono and binary boehmite system due to the high stress caused by its very small particle size. Drying rate and stress, hence cracking can be somehow controlled by the drying condition. All the sintered materials from the pressure filtered compacts of submicron/nanometer alumina and alumina-matrix composite show very homogeneous microstructures with fine grain size, confirming that colloidal pressure filtration is a good way to produce advanced materials

The present study examines the feasibility of fabricating Saffil alumina fibre-reinforced mullite matrix composites via a single-infiltration EFD process, and/or EFD plus pressure filtration. An in-situ EFD cell was developed in order to produce fibrereinforced composites. This cell is able to record the weight gain during EFD, i.e., in real time, thus providing information to establish the exact EFD kinetic model. The critical issues in preparing an aqueous based mullite precursor (suitable for EFD) from nano-size boehmite and fumed silica particles were discussed. It has also been shown that dense, homogeneous green mullite composition and sintered mullite microstructures can be prepared successfully only by controlling the short-range particle-particle interactions of the dissimilar particle species within the suspension. A reaction bonded mullite (RBM) slurry containing alumina and silicon carbide powders was developed in order to obtain zero-shrinkage RBM after sintering. Saffil alumina fibre mats pre-coated with mullite using EFD were infiltrated with RBM slurry using PF. By these means, high green density ( 64.5 % TD) and sintered density (94.6 % TD) were achieved. Room and high temperature 4-point bend behaviour of the monolithic and fibrereinforced CMCs were examined. 30 vol % fibre addition increased both the strength and toughness of the monolithic mullite at room and high temperature. At room temperature, a maximum 4-point bend strength (-500 MPa) was obtained from a fibrereinforced RBM composite having a weak mullite interface between the fibre and the mullite matrix. Both monolithic and fibre-reinforced mullite components produced here were able to keep the similar strength from room temperature up to 1300 °C. A slight decrease was observed at 1400 °C whilst a dramatic strength decrease occurred at 1500 °C, as fibre grain growth occurs at this temperature. Debonding, fibre pull-out, bridging, load transfer and crack deflection mechanisms are found to be responsible for the high strength and toughness. The state-of-damage in Saffil alumina fibre-reinforced RBM CMCs subjected to cyclic fatigue was investigated by means of acoustic emission (AE) monitoring and dynamic forced resonance (FR) techniques. FR measurements showed that as the number of cycles is increased, the Young's modulus decreases, whereas the internal friction increases. The sharp increase in Q" 1 as the number of cycles is increased is evidence for the development of significant microstructural damage, such as matrix cracking and delamination, resulting in the creation of new internal surfaces within the composite.

An effective methodology for discrete fracture in quasi-brittle material is presented within an explicit finite discrete element framework. Simple pragmatic models are envisaged that reflect the data deficiency of the quasi-brittle material and recover the observed physical response within engineering accuracy. Phenomenological strain-softening constitutive models are adopted for the modelling of micromechanical processes in an average sense. An extensional basis for fracture is assumed in both tensile and compressive stress fields, with only the mechanism with which inelastic strain is realised differing between the two stress states. To overcome the mesh dependence introduced by local softening constitutive relationships, the socalled localisation limiters are adopted in the form of the tensile crack band, nonlocal and viscous smeared crack models. Effective localisation lengthscales introduced by these regularisation methods ensure mesh objective failure localisation a priori to discrete crack insertion. A nonlocal map of failure indicators initiates fracture, with discrete cracks inserted into the finite element continuum by the splitting of the discretisation. An isotropic, non-associative Mohr-Coulomb model is derived in principal stress space as a first order approximation to the quasi-brittle response in compression. A model for discrete fracture in tensile and compressive stress fields is proposed, defined by a composite yield surface consisting of the fully anisotropic rotating crack band model coupled with the isotropic, non-associative Mohr-Coulomb model. The novel inclusion of an explicit coupling between the extensional inelastic dilation strain accrued during compressive failure and tensile strength degradation in the dilation directions permits the realisation of discrete fracturing in purely compressive stress fields. The so-called continuum-discrete transition introduces additional degrees of freedom into quasi-brittle systems and permits large deformation to be realised through the process of cataclastic flow. This advancement is considered significant and necessary in the recovery of the observed quasi-brittle response. The effectiveness of the proposed constitutive fracture models is verified by application to a number of physical quasi-brittle fracture systems, including borehole breakout, fracturing around excavations, strip punch tests, dynamic spalling and anchor pullout tests, amongst others.

A delamination damage prognosis system is defined using data available from surface strain field measurements and verified against experimental testing. Optical Grid Deflectometry measurements are interrogated to produce a damage detection and analysis technique. The output of this analysis is used to directly inform damage models in order to assess the effect of detectable damage on structural integrity. Damage · growth is investigated both experimentally and numerically in order to produce correlated models that can be used together with sensor outputs to predict the integrity of the damaged structure.

This thesis describes an investigation of the shear buckling strength of thin metal plates and empty and filled Cold-Formed steel lipped C-shaped cassette sections without intermediate stiffeners in the longitudinal or in the transverse direction of the wide web of the cassettes for various boundary conditions, material properties and geometrical characteristics. Tests were carried out to determine the properties of the metal skin of the cassettes and of the three polystyrene foams that were used for the infilling of the liner trays. Experiments have been carried out on C-shaped cassettes when the wide flanges are stabilised by various lightweight insulation materials fixed in position. A range of alternative numerical analyses complements this experimental research. An extensive literature research has been carried out covering Thin Metal Plates, Cold-Formed Steel Sections, Diaphragms, Composites Structures, C-shaped panels, Sandwich Panels, Polystyrene and Polyurethane materials, and the effect of Intermediate Stiffeners in various cases. The background reading embraced all aspects of the design and behaviour of Cold-Formed steel sections and sheeting. Consideration was given to BS 5950: Part 5,1987 "Code of practice for design of cold formed sections", to the A1S1, 1986 "Design and analysis of cold. formed steel sections", and to Eurocode 3: Part 1.3, 1997 "Coldformed thin gauge members and sheeting". However, a more detailed study was concentrated on the shear buckling of stiffened and unstiffened thin-plates and G shaped panels with and without intermediate stiffeners.

This study involved the use of Laser Doppler Anemometry (LDA) to measure air flows at high temporal and spatial resolution for the calibration of high-precision instrumentation, together with the use of Computational Fluid Dynamics (CFD) modelling tools. The work was performed in conjunction with BSRIA Instrument Solutions (BIS) with the main aim of improving the speed and accuracy of the calibration facilities within the BIS laboratory through the novel application of these technologies. Throughout the exploration of each facility, experimental LDA measurements, theory and virtual CFD models were compared to explore the airflow behaviour. In Part 1 of the investigation, this primarily focussed on a Balometer Calibration Facility (BCF – volume flow device) and an Open Jet Wind Tunnel (OJWT – air velocity device); both controlled using orifice plate pressure drops. The BCF was explored using attached sections of square ducting with optical access for LDA measurements. A traversing method was devised to measure volumetric flow rates, which were compared to the measured orifice plate pressure drops in order to achieve calibration of the facility. Comparing traversing methods showed that a Log-Tchebycheff method provided the most accurate prediction. The BCF calibration method was devised such that is could be applied to other volume flow facilities. These experimental measurements were compared with CFD simulations of ducted airflow in square and circular cross-section geometries for further analysis including the theoretical 1/7th power law velocity profile along with profile method comparisons. A further volume flow device, the Blower Door Calibraion Facility (BDCF) was investigated using a similar process to show the generality of the devised method. The OJWT, a facility typically used to calibrate velocity devices such as anemometers, was calibrated using a simple but novel calibration process involving LDA. To investigate further the OJWT behaviour, CFD models were created to represent OJWT with and without an anemometer and contrasted with equivalent experimental situations. A comparison of these experimental and computational data sets was performed showing representation of experimentally observed phenomena within the CFD model. This included so-called „blockage factors‟, mentioned below. iii In operation, to achieve the calibration of anemometers, the OJWT must make use of Blockage Factors (BFs) in order to correct the calibrated reference velocity to determine the actual velocity experienced by the unit under test. This was explored in Part 2 of the project. Experimental and computational (CFD) investigations were performed to explore the nature of the BFs towards a method of prediction. This would be of great use to industry as an anemometer calibration is not valid on an OJWT if an incorrect BF was applied. Outputs include: Novel methods of calibration were devised for test facilities which successfully achieved UK Accreditation Service (UKAS) certification. LDA was used to show that, in square ducts, the Log-Tchebycheff profiling method performs better than Equal Areas at providing an estimate for average cross sectional velocity when both low- and high- flow rates are considered. To verify CFD models of ducted airflow and of near-field measurements around an anemometer within an OJWT, LDA was shown to be a powerful tool. The basis for a method of BF prediction was proposed. This is based upon the empirical relationship between a statistical analysis of numerous calibrations along with an experimental LDA measurement of the deflection of air around an anemometer (Radius Expansion – RE). This is a first use within the industry and is an original contribution to knowledge. CFD models were also employed in the investigation of BFs, and the empirical relationship was applied to the data extracted from these simulations. Results closely matching the expected values were produced. The empirical relationship (between RE and BF) was demonstrated through estimation of the BF of an unknown Kimo 70 mm anemometer.

No description available.

The aim of the work presented in this thesis is to examine the thermal, dimensional and strength behaviour of glass concrete. The type of concrete used for this study differed significantly from those of other researchers. Firstly, the particle size distribution of glass identically matched the grading of natural aggregate. Secondly the glass used was free of sharp edges, uncontaminated, and produced by a unique implosion technique. The results showed that the process of cement hydration produces a much higher temperature in concrete made with glass cullet. Furthermore, when such concretes were allowed to hydrate under freezing conditions (-10°C or -20°), glass concrete was able to withstand freezing much longer than control concrete. The exposure of glass concrete to both high (60 °C) and low (-20°C) temperatures maintained greater temperature stability. This behaviour is attributed to the thermal properties of glass and the pore structure of glass concrete. These findings suggest that concrete made with recycled glass could have two important applications, namely, cold temperature concreting and in buildings where it is important to maintain greater temperature stability. The benefits of glass aggregate exceeded the enhanced performance of the best available accelerator with natural aggregate. Glass concrete probably has a lower drying shrinkage than any natural aggregate. Because of its low shrinkage, glass concrete is more affected by a change in water to cement ratio. The reduction in shrinkage was related to the glass content of the concrete. A mathematical model is suggested for the prediction of drying shrinkage of concrete made with a mix of glass and natural aggregates. This model produced a very good correlation with the experimentally obtained results. The 28 day compressive strength of standard cured glass concrete was lower than that of control concrete. However, significantly higher rates of strength development are achieved during the first 7 days due to the early acceleration of hydration resulting from the thermal properties of glass. There is also a much greater gain in the strength of glass concrete between the ages of 28 days and 1 year. As a result of these trends, the early age and long term strength of glass concrete is higher than that of control concrete. The post 28 day gain in strength is mainly attributed to better bonding of glass aggregates and the pozzolanic activity of the finest glass particles. When glass concrete is cured at either a high (40°C) or low (-10°C and -20°C) curing temperature, the 28 day compressive strength is higher than control concrete. Glass concrete that had been cured at low temperatures and then subsequently allowed normal curing recovered 100% of its strength, while the recovery for control concrete was just 50%. Thus it can be concluded that glass is better suited for both hot and cold weather concreting.

A significant amount of research has been carried out on the material and structural characteristics of fibre reinforced concrete (FRC). Chapter 1 presents a literature survey of the historical background and development of FRC. New experimental data on the direct tensile, cylinder splitting, flexural, and cube crushing strengths of both plain and fibrous concretes is provided in Chapter 2. An analytical formula for predicting the direct tensile strength of FRC has also been derived. Chapter 3 comprises two tension softening models (i.e. TSM and EMIS models) for simulating the complete pre-cracking and post-cracking responses of cementitious composite. Both models have been validated using various experimental data. Chapter 4 addresses the characterization of the fiber-matrix properties where fibre pull out tests have been conducted for various types of deformed fibres. Chapter 5 covers the nonlinear finite element modelling of the deformed fibre pull out from cementitious matrix. A fibre pullout model has been developed and validated using experimental results. Chapter 6 concentrates on the simulation of the complete loaddeflection response and failure pattern of FRC flat slab at slab-column connection. Parametric studies on the slab thickness and reinforcement ratio have been conducted. Simulations of FRC flat slab and beam structures have also been performed using the EMIS model. Chapter 7 cites the details of a general analytical model for the prediction of ultimate punching shear strength of flat slab at slab-column connection. The analytical model is shown to provide reasonably good improvement when compared to several major design codes. Additionally, a unique analytical expression that provides useful information on the failure mode of flat slab has also been derived. Finally, Chapter 8 summarizes and concludes the outcomes and achievements throughout the research and possible areas for further research are suggested.