Waste engineering fluids, such as coolants and cutting fluids, are difficult to treat because they have variable physical natures, are particularly toxic and have a very high Chemical Oxygen Demand. The complex and unknown chemical content of the many different products available is also problematic. Current technologies, such as nanofiltration or chemical treatment, are quite effective at reducing the COD of the waste metalworking fluids before disposal. These technologies remove free or emulsified oil and high molecular weight components, but they have their limitations. In addition, the more stringent legislation on waste disposal and effluent discharge induces an economical stress on engineering industries. It can be anticipated that future legislations will introduce eco-toxicology measurements into industrial effluent discharge consents. A modular on-site treatment plant to treat semi-synthetic metalworking fluids was developed in this thesis. The approach was to combine different technologies and to inter-optimise their performances. The technologies used were membrane filtration, bioremediation and chemical treatment. The use of activated carbon was also studied. Membrane filtration included the study of ultrafiltration and nanofiltration. For the bioremediation process, a bio-consortia was developed and tested over 8 months. A final design of the whole process is given. The proposed treatment plant transforms the waste metalworking fluid into two products: very low chemical oxygen demand aqueous phase at 30 mg/l COD and a recovered oil showing a calorific value of 42 kJ/kg which could be a possible commodity. The whole treatment plant is scalled-up for thye treatment of 500 L of waste metalworking fluid per day.

This thesis is concerned with the optimal design of Water Distribution Networks (WDNs). The design involves finding an acceptable trade-off between cost minimisation and the maximisation of numerous system benefits. The primary design problem involves cost-effective specification of a pipe network layout and pipe sizes in order to satisfy expected consumer water demands within required pressure limits. The design of a WDN has many variable parameters such as position and size of the water sources, position and the size of the pipes and position of the treatment plants. However, the layout is constrained by the location of existing facilities such as streets and buildings and other geographic features. The total costs may consist of the cost of network materials such as pipes, construction works and system operation and maintenance. The problem may be extended to consider the design of additional components, such as reservoirs, tanks, pumps and valves. Practical designs must also cater for the uncertainty of demand, the requirement of surplus capacity for future growth, and the hydraulic reliability of the system under different demand and potential failure conditions. The thesis reviews the literature related to water distribution networks, their design and optimisation. It then presents a Genetic Algorithm (GA) formulation to assist in developing the design of a water distribution network. The main aim of this research is to investigate the possibility of combining GAs and GIS in the design optimisation. A decision mechanism is developed which enables the model to reach a meaningful solution and provide a practical design technique for WDNs. The aim is also to provide an experimental analysis of the combined GA and decision mechanism to solve the problem in hand and to assess the robustness of these techniques when applied to different instances. An initial prototype model is presented for the design of a WDN which is used to determine the necessary features of the 'final' model. These features include the world in which the model will be built, the design of the fitness function, chromosome representation, and GA operators. The research mainly concluded that the initial model prototype was useful to determine the necessary features and to produce the final model which enables a variety of necessary factors to be explicitly included in the design of WDNs. This initial model suggested that the final model should include the decision mechanism, which is a matter of policy management and hydraulics, and hydraulic principles which allowed to compare the behaviour of different parameters and to simulate the functioning of the network under different scenarios. Water allocation and distribution policies can be applied according to the importance of the demand area and the ability of the system to deliver sufficient water amounts. These policies link essential hydraulic and institutional relationships as well as water uses and users and allocation decision-making process. It was also found that the representation of the world layout is important. The world is described in GIS in terms of models that define the concepts and procedures needed to translate real-world features into data. The important aspects in the chromosome representation are the node positions, the links. In this case, a chromosome must contain the three-dimensional node coordinates, the connection between nodes, the head required to pump the water. The best model parameters were extracted to be used in real-life situations. The result of tests on an example world demonstrated that the model was successful, and the potential exists for the use of this formulation in more complex and real-world scenarios.

The aim of the present work was to improve the predictive capabilities of current modelling methods used to assess the long-term biosphere impacts of underground repositories for radioactive wastes. A number of issues related to parameter and conceptual uncertainties associated with compartmental biosphere models that simulate transport and accumulation of radionuclides in soils were addressed. The structure of compartmental models used for radiological risk assessments has not evolved noticeably over the past few decades and most of these models rely on simple assumptions. For example, compartmental models used to predict activity concentrations of radionuclides released into soils over very long timescales (typical of the lifetime of underground disposal repositories) assume arbitrary model specifications such as soil layer thickness (the vertical discretisation of the soil column) and length of the time step. Moreover, the majority of the available models assume invariant sorption characteristics of radionuclides with soil depth and hence employ constant solid-liquid distribution coefficient (Kd) values regardless of soil characteristics known to affect radionuclide sorption (e.g. pH, redox potential, moisture content and organic matter). The empirical Kd has a profound effect on long-term predictions of radionuclide behaviour in soil since it determines the degree of radionuclide retardation due to interaction with the soil. It is associated with considerable uncertainty due to differences in experimental conditions and methods used for its measurement and the variation in soil characteristics. In this study, three soil types (arable, grassland and woodland) were incubated under anaerobic conditions and the behaviour of naturally occurring selenium, iodine, rhenium and uranium, expressed as Kd, was investigated. The results indicate that variation in soil characteristics (e.g. moisture content, pH, mineral and organic carbon content) is a significant source of K, variability. Soils relatively higher in organic matter content (e.g. top soils) have higher sorptive capacities for trace elements than mineral subsoils and hence higher Kds. Dynamic, complex behaviour of K, under flooded, anaerobic soil condtions was measured over a 3 week period in soil microcosms. This dynamic behaviour was driven by the shift in soil redox potential which was associated with solubilisation of soil organic and mineral (Fe oxide) phases. Overall, the maximum observed variation in K, over the entire incubation period did not exceed 2 orders of magnitude. Biosphere models were constructed which combined a physically-based water flow model and the compartmental approach and used to simulate the long-term vertical distribution of radionuclides in the soil as well as radionuclide dynamics under different environmental conditions. Investigating radionuclide dynamics on a short timescale could only be achieved using models with a daily time step since short-term variation was obscured by a longer (annual) time step. Simulation results give insights into some of the limitations of available biosphere modelling methods for radiological risk assessment that are often overlooked. For example, soil radionuclide activity concentrations calculated using compartmental models are sensitive to the vertical discretisation (i.e. thickness of soil layers into which the soil column is divided) and time step of the model, hence the structure of the model should not be set arbitrarily. The discretisation procedure proposed in the present study may provide a useful framework to select the appropriate structure of biosphere assessment models. With respect to the effect of uncertainty in K, on model calculations, the results show that equilibrium timescales and radionuclide activity concentrations in the soil at equilibrium increase as the K, increases. For example, the time to reach steady state radionuclide activity concentrations in the vegetated topsoil increased 14-fold and 7-fold, respectively, when K, was increased 28-fold, which is a small variation compared to the uncertainty of Kd commonly reported in the literature (e.g. a few orders of magnitude). The Kd also affects short and long-term radionuclide dynamics in soils; the activity concentration of a radionuclide with low Kd (weakly sorbing) is more responsive to seasonal fluctuations in climatic and hydrological conditions than a radionuclide with a large Kd (strongly sorbing). Radionuclide uptake by plant roots, especially those which access highly contaminated soil layers adjacent to the contaminated aquifer, could be an important mechanism that provides a direct pathway between shallow, contaminated aquifers and the soil surface where elavated contamination poses greater risks.

The application of solid adsorbents for gas separation in pre-combustion carbon capture from gasification processes has gained attention in recent times. This is due to the potential of the technology to reduce the overall energy penalty associated with the capture process. However, this requires the development of solid adsorbent materials with large selectivity, large adsorption capacity, fast adsorption kinetics for CO2 coupled with good mechanical strength and thermal stability. In this work, results on CO2 adsorption performance of three different types of adsorbents; a commercial activated carbon, phenolic resin activated carbons and zeolite templated carbons have been reported at atmospheric and high pressures conditions. The commercial activated carbon was obtained from Norit Carbons UK, the phenolic resin activated carbon was obtained from MAST Carbon Ltd., while the templated carbons were synthesized in the laboratory. A commercial activated carbon was used as bench mark for this study. Surface modification of these carbons was also undertaken and their CO2 uptake measurements at ambient and high pressure conditions were recorded. The commercial and templated carbons were modified by functionalising with amine group, while the phenolic resin carbon was modified by oxidation. The textural properties of the adsorbents was examined using the Micromeritics ASAP, while the CO2 adsorption capacities were conducted using the thermogravimetric analyser (TGA) and the High pressure volumetric analyser (HPVA). Textural properties of synthesized templated adsorbents were seen to depend on the textural characteristics of the parent material. The β-type zeolite produced the carbons with the best textural property. Increase in activation temperature and addition of furfuryl alcohol (FA) enhanced the surface area of most of the templated carbons. The textural property of all the adsorbents under study was seen to differently affect the CO2 uptake capacity at atmospheric (0.1 MPa) and high pressure conditions (up to 4 MPa). Micropore volume and surface area of the commercial activated carbons, phenolic resin activated carbons, and the templated carbons greatly influenced the adsorption trends recorded at ambient conditions. Total pore volumes positively influenced adsorption trend for templated carbons, but not the phenolic resin activated carbons at ambient and high pressure. This also positively influenced the adsorption trend for the commercial activated carbons, but at ambient conditions only. The surface area and the micropore volume have no effect on the adsorption trends for the templated carbons and the commercial activated carbons at high pressure conditions. However, these played a positive role in the adsorption capacities of the phenolic resin activated carbons at the same experimental conditions. Micropore volume and surface area of adsorbents play a major role on the adsorption trends recorded for the modified adsorbents at ambient conditions only. No trend was recorded for adsorption capacities at high pressure conditions. Only the oxidized phenolic resin activated carbon showed a positive adsorption trend with respect to total pore volume at high pressure condition. The amine modified commercial activated carbon showed no positive adsorption trend with respect to the total pore volume at both ambient and high pressure conditions, while the amine modified templated carbon showed no adsorption trend with respect to the textural properties at ambient and high pressure conditions. CO2 uptake measurements for the modified and unmodified templated carbon and phenolic resin carbon, were observed to be higher than those of the commercial activated carbon at ambient and high pressure conditions. Maximum CO2 uptake was recorded at 25 oC. At ambient pressure, the phenolic resin carbon (MC11) showed the highest CO2 uptake of approximately 3.3 mmol g-1, followed by the commercial activated carbon (2.4 mmol g-1), then, the templated carbon (2.4 mmol g-1). At high pressure, the templated carbons (β-AC7-2%) showed the highest CO2 uptake (21.3 mmol g-1), followed by phenolic resin carbon (MC4 - 12.2 mmol g-1), and the commercial activated carbon (6.6 mmol g-1). When samples were modified, the amine modified templated carbon and oxidized phenolic resin carbon showed the highest CO2 uptake of 2.9 mmol g-1 each at ambient pressure, followed by the commercial activated carbon (2.7 mmol g-1). At high pressure conditions, the oxidized phenolic resin carbon showed the highest (10.6 mmol g-1) uptake level, followed by the templated carbon (8.7 mmol g-1), and commercial activated carbon (6.5 mmol g-1).

Climate change is affecting planet Earth. The main cause is anthropogenic emissions of greenhouse gases, the principal one being carbon dioxide, released in the atmosphere as a by-product of the combustion of hydrocarbons for the generation of energy. Carbon capture and storage (CCS) is a technology that would prevent carbon dioxide from being emitted into the atmosphere by safely sequestering it underground. For so doing, CO2 must be captured at large emission points and transported at high pressure to underground reservoirs, where the gas can be injected and stored for thousands of years to come. During surface transportation, leakages from high pressure facilities would pose a risk to the general public, for carbon dioxide is toxic at high concentrations. In this study, atmospheric dispersion of carbon dioxide is studied by the usage of software that solves mathematical equations and algorithms simulating the pollutant dispersion. Dispersion models are used to estimate or predict downwind distances covered by toxic concentrations of the pollutant, emitted from sources such as high-pressure transportation facilities within CCS projects. Two modelling tools from two different classes (Gaussian ALOHA 5.4 and Computational Fluid Dynamics PANACHE 3.4.1) have been evaluated against release field experiments using the statistical model evaluation method proposed by Hanna et al. (1993,2004) and Hanna and Chang (2001), and applied for the consideration of the dense gas CO2, released in large amounts due to leakages. Predictions from the two models have been compared and the limitations of both examined, when dealing with a gas that presents the distinctive physical characteristics of carbon dioxide. The models have been used and compared in simulating representative failure cases within CCS transportation with release parameters taken from the literature. The Computational Fluid Dynamics (CFD) model showed a much higher precision when describing the release of the gas from a HP facility, mainly when dealing with the jet release caused by leakages of any dimensions. When dealing with the transportation of toxic gases, the magnitude of hazards posed by potential failure events within the transportation system is proportional to the extent of the area covered by toxic concentrations of the gas, when modelling representative leakages. Results of this investigation depict a lowering of the Risk involved in the transportation of CO2 by up to an order of magnitude, when modelling the same releases with CFD tools, instead of the more common Gaussian models. The European Union recognizes that deployment of CCS for hydrocarbon power generation, in parallel with the production of renewable energies, is the only way to meet the target for temperature stabilization. For its Impact Assessment on CCS, the EU used results from a risk assessment compiled after the utilization of a Gaussian model. In this thesis, a criticism of this choice is put forward, considering that, when introducing the technology to the general public and regional scale administrators, a Risk Assessment derived using results from Gaussian models can over-estimate the risk in a way not favourable to the purpose.

This thesis introduces three main elements to support decision-making in air quality-related transport planning. The first are novel automatic collection and processing algorithms for traffic flow and geospatial data for input to air pollution models of transport schemes under analysis. The second is a novel strategy to improve the modelling of air quality by the calibration of input background concentrations. The third is a novel 3D air pollution dispersion interface for the 3D visualisation of the air quality predictions in 3D digital city models. Four urban transport schemes were used for the initial development of, and for testing, the applicability and validation of future air quality predictions of the decision-support system based on the above three elements. The automation of the input data collection and processing reduced significantly the time and effort required to set up the air pollution model. The calibration of background concentrations significantly improved the accuracy of, not only the annual mean, but also the hourly, air quality predictions and effectively reduced the model runtime. The 3D air pollution dispersion interface provided an intuitive 3D visualisation of the air quality predictions at and above the ground surface in a single 3D virtual scene. The application of this decision-support system enabled the development of alternative future traffic scenarios so a proposed urban transport scheme might contribute to achieving certain air quality objectives. The validation of the future air quality predictions showed that the methods used for the future projection of air pollution input data slightly increase the error between the modelled and actual annual mean NO2 future concentrations. They also significantly increase the error between the modelled and actual hourly NO2 future concentrations

Due to their high swelling ability, high water retention capacity and low permeability, compacted bentonites have been considered as a key component of cover lining systems for storage of low and medium level toxic wastes and as barrier and backfilling materials for long-term safe storage of high level toxic waste in many countries. This thesis presents an experimental, theoretical and numerical study of thermo-hydromechanical- chemical behaviour of MX80 bentonite in geotechnical applications. The determination of swelling pressures of compacted bentonites is an important aspect of most bentonitebased barrier systems. Swelling pressures of bentonites are usually determined in the laboratory under constant volume conditions using oedometers. Powdered bentonites are usually compacted in stainless steel cylindrical specimen rings and tested immediately after the compaction process is completed. The swelling pressures thus measured are influenced by some post compaction residual stresses. However, bentonites in the form of pellets and bricks are prepared and used in repository conditions that, in turn, are free from any post compaction residual stresses. The influence of post compaction residual stress on swelling pressures of compacted bentonites for a range of dry density that are of interest has not been explored in the past. Such studies are of potential interest for the safe and efficient designs of toxic waste disposal repositories. Many of the waste repositories are commissioned in locations where the ground water either contains significant amount of salts or the repositories are anticipated to receive saline water from sea. Additionally, in some cases, the wastes that are disposed emit very high temperatures (e.g. spent fuel). Under these repository conditions, compacted bentonite barriers are subjected to both thermal and hydraulic loadings at opposite boundaries. A detailed study devoted to appreciate the combined influence of an elevated temperature, bulk fluid type and solute transport due to both hydraulic and thermal gradients are necessary to better the understanding of the mechanical behaviour of compacted bentonites in many practical engineering problems. Constant volume swelling pressure tests were carried out on compacted MX80 bentonite specimens in order to study the influence of post compaction residual stress, electrolyte concentration and temperature (for isothermal conditions) on the swelling pressure. The dry density of the bentonite was varied between 1.1 to 1.9 Mg/m3 to cover a wide range of compaction conditions. Theoretical assessments of swelling pressures were made using the Gouy-Chapman diffuse double layer theory and the Stern theory, as applicable to interacting clay platelet systems. Further, the experimental and the theoretical swelling pressures were compared in order to bring out the applicability of the electrical theories in assessing swelling pressures of bentonites for both compacted saturated and initially saturated slurried conditions. A series of thermal and thermo-hydraulic tests were carried out on bentonite specimens under laboratory scale settings. During the thermal tests, temperatures of 85 and 25 °C were applied at the bottom and top ends of the specimens, respectively. During the thermo-hydraulic tests, in addition to unequal temperature at opposite ends, distilled water was supplied from the top end of the specimens. The temperature and the relative humidity were monitored along predetermined depths of the specimens during both types of testing methods. The swelling pressure variations were monitored at the opposite end of the heat source. Changes in water content, dry density and concentrations of cations and anions along predetermined depths of the specimens were measured after termination of each of the tests. A thermo-hydro-mechanical finite element simulation was undertaken specifically for swelling pressures using the numerical code COMPASS (COde for Modelling PArtially Saturated Soils). Further, the experimental and the simulated results were compared both for thermal and thermo-hydraulic boundary conditions. Compacted bentonite specimens with post compaction residual stresses exhibited lesser swelling pressures as compared to their stress released counterparts. Agreements between the calculated swelling pressures from the Stern theory and the experimental swelling pressure results were found to be reasonable for compaction dry densities of less than 1.45 Mg/m3, whereas at higher dry densities, agreements between the measured swelling pressures and those calculated from the electrical theories were found to be poor. Conversely, compressibility behaviour of initially saturated slurried bentonites was found to be captured well by the electrical theories. On account of vapour flow under thermal gradients, compacted bentonite specimens exhibited swelling pressures at the opposite end of the heat source. The measured swelling pressure for the thermal gradient adopted varied between 0.5 to 1.2 MPa, whereas greater swelling pressures were noted due to an applied thermo-hydraulic gradient. Evaporation, condensation, diffusion and advection processes influenced the distribution of ions in compacted bentonite when subjected to both thermal and thermo-hydraulic gradients. The finite element code, COMPASS, enabled assessing changes in suction and swelling pressure of compacted bentonite satisfactorily under both thermal and thermo-hydraulic hydraulic gradients.

This thesis presents work carried out to further the understanding of the use of waste Acid Mine Drainage (AMD) solids to remove phosphorus from wastewaters. AMD can result in serious pollution and so is often treated by the Coal Authority in the UK the resultant solids being a costly disposal issue. Currently the Water Framework Directive (WFD) is putting increased strain on technology used to remove phosphorus at WasteWater Treatment Works (WWTWs). The work presented in this thesis therefore investigates the use of a problematic waste to provide a novel solution to the issue of phosphorus removal at WWTWs. There has been previous work carried out on the use of both AMD solids and AMD-cement pellets to remove phosphorus from wastewaters. This thesis builds upon this work, firstly the phosphorus removal achieved by unpelletised materials studied in this thesis are compared to those studied by others through the comparison of adsorption isotherms. Mirroring other studies, the principal material studied in this thesis was then pelletised using Portland cement as a binder. Phosphorus removal by these pellets in batch tests was determined and optimised. Phosphorus removal was found to increase with a reduction in pellet size and an increase in test length up to the maximum length studied of three weeks. This increase in performance was attributed to the introduction of a calcium phosphate precipitation removal mechanism as a result of the use of cement as a binder for the pellets. This was highlighted through the correlation of phosphorus removal with a drop in calcium concentrations and pH values. It was concluded that the end product of this precipitation was hydroxyapatite. Continuous column tests were performed on the pellets, it was found that the pellets not directly involved in phosphorus removal were still having their reactivity leached out by passing water and so when these pellets became involved in removal, the columns quickly failed.

The most common technique for carbon dioxide removal from gaseous streams is amine scrubbing, a proven technology in the oil and gas industries. The use of this route in coal fired power plants is not fully understood and the likelihood of solvent degradation is high. Decreased absorption efficiency, undesirable byproducts, the environmental impact of their disposal and increased process costs are the main consequences. In this study, two experimental rigs were designed and commissioned to explore the effects of gas composition and temperature on monoethanolamine degradation. Analytical procedures to detect and quantify its major thermal and oxidative degradation products were also developed. It became apparent early on that solvent degradation, under actual plant conditions, is a slow phenomenon, thus, it was decided to focus on thermal degradation. The present study uniquely enabled the absorption/desorption behaviour of thermally degraded solvents to be evaluated. The major thermal degradation products were quantified. After 14 full absorption/stripping cycles at the presence of 16% oxygen and 15% carbon dioxide, significant concentrations of nitrites and nitrates were detected in the samples. Thermal degradation at 160 oC for 8 weeks reduced monoethanolamine concentration by almost 95%, as evidenced by the chemical analysis, but the remaining solvent retained 22% of its capacity to remove carbon dioxide. Therefore, although not fully quantified, the requirement for monoethanolamine make-up may not be quite as serious as initially believed. There is some evidence to support that the rate of thermal degradation was enhanced as carbon dioxide loading increased and a 20% higher MEA loss was determined in the samples with the rich initial molar loading. A range of degradation products were quantified that correspond to those cited in the literature. 1-(2-hydroxyethyl)-2-imidazolidinone was indicated as the most stable MEA degradation product in the degraded samples at concentrations of up to 17% v/v.

In this work, different aspects on the re-use of wood based fibres have been studied, focusing on ink detachment of flexographic ink from model cellulose surfaces and changes in porous structure of kraft fibres following different treatments. New model systems for evaluation of ink detachment and ink-cellulose interactions were used. Ink detachment was studied using Impinging jet cell equipment, taking into consideration the influence of storage conditions, surface roughness and surface energy of the cellulose substrate. A micro adhesion measurement apparatus (MAMA) was used to directly study ink-cellulose interactions, from which the adhesive properties between ink and cellulose, having various surface energies, could be derived. UV-light, elevated temperatures, longer storage time, decreased surface energy, i.e. making the cellulose surface more hydrophobic, and high surface roughness all negatively affected ink detachment. Attenuated total reflectance - fourier transform infra red (ATR-FTIR) and atomic force microscopy (AFM) was used to evaluate structural and chemical changes of ink and cellulose upon storage at elevated temperature or under UV-light. After storage at elevated temperatures, ATR-FTIR spectra indicated that a hydrolysis or an oxidative reaction took place as a peak at 1710 cm-1 appeared. AFM revealed that storage at elevated temperatures caused the latex particles present in the ink to form a film, most likely due to annealing. Less ink detached from hydrophobic cellulose surfaces. Ink detachment decreased for rougher cellulose substrates due to an increased molecular contact area. Fibre pore structure and water retaining ability influenced fibre/fibre joint strength and different paper strength properties. Investigations took into account the effect of pulp yield, counter-ion types, pH, salt, hornification and strength enhancing additives. Nuclear magnetic resonance relaxation (NMR), inverse size exclusion chromatography (ISEC) and water retention value (WRV) measured the changes that occur in the fibre wall upon varying the conditions. Each different measuring technique contained unique information such that a combination of the techniques was necessary to give as complete a picture as possible over the changes that occurred in the fibre wall upon varying the conditions for the fibre. A correlation between fibre pore radius and sheet strength properties was found, suggesting that fibres with larger pores allow for a larger molecular contact area between fibres to be formed during drying and consolidation of the paper. Fibre/fibre joint strength, fibre flexibility, and the number of efficient fibre/fibre contacts also controlled sheet strength. The effect of different strength enhancing additives on fibre pore structure and paper strength was investigated. Larger pores in the fibres allowed for additives to penetrate into the fibre wall. Additives with low molecular mass (Mw) penetrated into the fibre wall to a larger extent than additives with a high Mw, causing an embrittlement of the fibre. However, low Mw additives gave higher sheet tensile strength despite a leveling out in strength at high additions, indicating that the fibre wall can only adsorb a limited amount of chemical. Polyallylamine hydrochloride (PAH) and polyelectrolyte complexes (PEC) of PAH and polyacrylic acid (PAA) were added separately to the pulp. PEC significantly improved both tensile strength and Z-strength, whereas PAH alone did not increase the strength properties to the same extent unless the sheets were heated to 150°C for 10 minutes. The results suggested that the effect of PEC was dominated by an improvement in fibre/fibre joint strength, whereas the effect of PAH was significantly affected by an improvement of the intra-fibre bond strength