Orientador: Vilmar Baldissera / Resumo: Apesar de diversos avanços científicos, mesmo após mais de 200 anos de aplicação da primeira calorimetria com humanos e todas as possibilidades criadas pelas tecnologias desenvolvidas, em especial os aparatos portáteis, o gasto energético (GE) dos exercícios resistidos ainda permanece, de certa forma, elusivo. Desta forma, o presente estudo foi conduzido com o objetivo geral de Quantificar e criar um modelo matemático para a predição do gasto energético de homens e mulheres jovens fisicamente ativos nos exercícios supino reto (SR) e leg press (LP). 11 homens (grupo masculino – GM), com idade (média ± desvio padrão) de 28,7 ± 5,5 anos, altura de 1,77 ± 0,07m, massa corporal de 81,40 ± 8,17kg e percentual de gordura de 16,36 ± 6,79% e 11 mulheres (grupo feminino – GF), com idade de 27,6 ± 4,6 anos, altura de 1,67 ± 0,06m, massa corporal de 60,05 ± 5,76kg e percentual de gordura de 20,20 ± 2,73% se voluntariaram para o estudo. Cada voluntário se apresentou ao Laboratório de Fisiologia do Exercício da UFSCar em oito ocasiões: na primeira visita, os voluntários foram informados sobre os objetivos e procedimentos do estudo, assinaram o termo de consentimento livre e esclarecido, os dados antropométricos mensurados e realizado o ajuste da biomecânica dos exercícios e teste de 1 repetição máxima (1RM) nos exercícios SR e LP. Após 48 a 72 horas, na segunda visita, os voluntários foram submetidos ao reteste de 1RM. Nas seis visitas seguintes, os voluntários foram submetidos ao protocol... (Resumo completo, clicar acesso eletrônico abaixo) / Abstract: Despite many scientific advances, even after over 200 years of the first calorimetry on humans and all the possibilities made especially by portable apparatus, energy expenditure (EE) still remains somehow elusive. In this sense, the present study was conducted to quantify and create a model to predict EE of young physically active men and women on the bench press (BP) and leg press (LP) exercises. 11 men (male group – GM) with age (mean ± standard deviation) 27.8 ± 5.5 years, height 1.77 ± 0.07m, body mass 81.40 ± 8.17kg and body fat percentage of 16.36 ± 6.79% and 11 women (female group – GF) with age 27.6 ± 4.6 years, height 1.67 ± 0.06m, body mass 60.05 ± 5.76kg and body fat percentage of 20.20 ± 2.73% volunteered for the study. Volunteers came to the Exercise Physiology Laboratory of UFSCar on eight visits: in the first visit, volunteers were informed of the aims and proceedings of the study, signed an informed consent and had their anthropometric data analyzed, biomechanical adjustments made and perform 1 repetition maximum (1RM) test on both exercises. After 48 to 72 hours, volunteers came to the second visit to perform 1RM re-test. In the six following visits, volunteers were submitted to the EE protocol at 40, 60 or 80% of their 1RM on the BP and LP exercises. Interval for evaluations of the same exercise were 48 to 72 hours. GM had a significant difference in total energy expenditure between exercises, but such effect was not observed when EE was quantified as calor... (Complete abstract click electronic access below) / Doutor

La fertilidad masculina puede ser medida mediante un espermatograma convencional, sin embargo este examen no incluye la valoración de la integridad del ADN espermático. Esta variable ha sido correlacionada con las tasas de fertilización, viabilidad y desarrollo del embrión, convirtiéndose en una herramienta de importancia clínica tanto para los programas de reproducción animal como los tratamientos de fertilidad asistida. El bioensayo Cometa es capaz de determinar de una manera exacta el valor de la integridad del ADN espermático, lamentablemente este examen no es de uso rutinario por su elevado costo de implementación ya que utiliza microscopia especializada y tinciones fluorescentes para evidenciar la migración del ADN. El objetivo de esta investigación fue evaluar la compatibilidad de las tinciones no fluorescentes Diffquik, Giemsa, de Feulgen y FastBlast en el bioensayo Cometa usando un método visual y automatizado. Se utilizaron 15 eyaculados previamente seleccionados de acuerdo al manual OMS 2010, para luego ser capacitados en búsqueda de homogeneidad adecuada para la experimentación. Cada muestra fue expuesta a una gradiente de Peróxido de hidrogeno (0, 10, 30,60 y 100 mM) por 1 hora a 4°C para luego evaluar el coeficiente de daño mediante el método visual y porcentaje de ADN en la cola mediante el método automatizado. Las pendientes de la regresión lineal en el método visual indican que los valores obtenidos por la tinción control SybrGreen (m=3,69) difieren con Giemsa (m=3,45) y Diffquik (m=2,57). En el método automatizado de igual manera SybrGreen (m=0.83), Giemsa (m=0,79) y Diffquik (m=0,77). Sin embargo SybrGreen es 1,06 veces más efectivo que Giemsa en el visual y 1,05 veces en el automatizado, sugiriendo una compatibilidad con el bioensayo cometa. De igual manera SybrGreen es 1,07 veces más efectivo que Diffquik en el visual y 1,44 veces en el automatizado, concluyendo una compatibilidad solo en el método visual.Male fertility can be measured by a conventional semen analysis, however, this examination does not include the assessment of sperm DNA integrity. This variable has been correlated with fertilization rates, embryo viability and development, becoming a tool of clinical importance for both animal breeding programs and assisted fertility treatments. Comet bioassay is able to determine an exact way the value of sperm DNA integrity, unfortunately this test is not routinely used because of its high cost of implementation because it uses specialized microscopy and fluorescent dyes to demonstrate DNA migration. The objective of this research was to evaluate the compatibility of non-fluorescent dyes Diffquik, Giemsa, Feulgen and Comet FastBlast in the bioassay using a visual and automated method. 15 ejaculates were used previously manually selected according to WHO 2010 and then be trained in finding adequate homogeneity for experimentation. Each sample was exposed to a hydrogen peroxide gradient (0, 10, 30,60 and 100 mM) for 1 hour at 4 ° C and then assess the damage coefficient by visual method and percentage of DNA in the tail by automated method. The slopes of the linear regressions on the visual method indicate that the values obtained by the SybrGreen Control staining (m = 3.69) differ with Giemsa (m = 3.45) and Diffquik (m = 2.57). In the same way automated method SybrGreen (m = 0.83), Giemsa (m = 0.79) and Diffquik (m = 0.77). However SybrGreen is 1.06 times more effective than Giemsa visual and 1.05 times in the automated, suggesting a comet support bioassay. Similarly SybrGreen is 1.07 times more effective than Diffquik visual and 1.44 times in the automated, concluding compatibility only in the visual method. Keywords:

The arithmetic operations such as multiplication and division in binary number system are computationally complex in terms of area, delay and power. Logarithmic Number Systems (LNS) offer a viable alternative combining the simplicity of fixed point number systems and the precision of floating point number systems. However, the computations in LNS result in some loss of accuracy and thus, are limited to mostly signal processing applications; where in certain amount of error is tolerable. In LNS, the cost of computations can be tradeoff with the level of accuracy needed. The Mitchell algorithm proposed incite[mitchell], is a simple approach commonly used for logarithmic multiplication. The method involves a high error margin due to a piecewise straight line approximation of the logarithm curve. Thus, several methods have been proposed in the literature for improving the accuracy of Mitchell's algorithm. In this thesis, we propose a new method for improving the accuracy of Mitchell's logarithmic multiplication using operand decomposition. The operand decomposition process decreases the number of bits with the value of '1' in the multiplicands and reduces the amount of approximation. The proposed method brings down the average error percentage of Mitchell's logarithmic multiplication by around 45%. It can be combined with previous methods to further improve the accuracy. Experimental results are presented to show that both the error range and the average error percentage can be significantly improved by using operand decomposition.

The gluing of unseasoned wood, called green gluing, is a relatively new sawmill process, implying a radically changed order of material flow in the production of value-added wood-based products. It facilitates the enhancement of raw material recovery and value yield by integrating defect elimination and gluing already before kiln drying. The present study evaluates green glued adhesive bonds in flatwise glued beams and finger joints. The main part of this work deals with green gluing using a moisture curing polyurethane adhesive (PUR). Standardised test methods and specially designed, small scale, specimens were used for the determination of the strength, fracture energy and the ductility of both dry- and green glued bonds in tension and in shear. Using the small scale specimens it was possible to capture the complete stress versus deformation curves, including also their unloading part. An optical system for deformation measurement was used for the analysis of bond behaviour. The influence of moisture content during curing and temperature after curing on the adhesive chemical composition and on the mechanical properties was investigated. Furthermore, the moisture transport through the adhesive bond during curing was tested. Finally, microscopy studies were performed for analysis of bond morphology and fracture. The results show that two significant factors influence the shear strength of green glued bonds: wood density and adhesive spread rate. Bonds which fulfil the requirements according to EN 386 could be obtained within a wide range of process parameters. The small specimen tests showed that green glued PUR bonds can reach the same strength and fracture energy, both in shear and in tension, as dry glued bonds with the same adhesive amount. The local material properties of the bonds could be determined, thanks to the failure in the tests taking place within the adhesive bond itself and not in the wood. Following process factors were shown to cause lower bond strength: a) a low adhesive spread rate, b) high pressure and c) short pressing time in combination with low wood density and high moisture content. Moreover, the heat treatment of the cured PUR adhesive during drying influenced the chemical composition of the adhesive, providing for higher strength, stiffness and Tg of the adhesive, caused by an increased amount of highly ordered bidentate urea.

Purpose: To assess the dietary intake and body composition of recreational and competitive female athletes, for the purpose of analyzing the relationships between macronutrient intake and body composition.. The main aim was to determine the relationship between caloric intake, carbohydrate (CHO) intake and protein intake with body fat percentage in active females. Methods: Using an IRB approved protocol, 44 volunteer female recreational and competitive athletes 18 years of age or older were recruited. Interviews were conducted to gather information on within day energy balance by assessing the time and amount of foods/beverages consumed, and the duration and intensity (using a Rating of Perceived Exertion scale) of activity performed on the day of assessment. All analyses were performed using Nutritiming™ (Calorie and Pulse Technologies, Atlanta, GA) to assess energy surpluses, energy deficits, and end of day energy balance. Information on date of birth, race/ethnicity, menstrual status, sleep and wake times, and prior diagnoses of metabolic disease and/or eating disorders were collected at the time of the interview. Height was assessed using a standard stadiometer. Weight and body composition were assessed via Bioelectrical Impedance Analysis (BIA) using InBody 230 (BioSpace Co. USA). The BIA assessment was performed to determine body fat percentage, Basal Metabolic Rate (BMR), Body Mass Index (BMI), segmental body composition, and fat and lean mass in kilograms. Nutrient data were derived using an interviewer-led, 24-hour recall. Results: CHO intake/kg total mass was significantly and inversely correlated with body fat percentage and BMI, (p=0.018 and p=0.001 respectively). Protein intake/kg total mass was also inversely and significantly correlated with body fat percentage (p=0.006). Fat intake was not significantly associated with BMI, body fat percent, or lean mass in kilograms. Total energy intake/kg total mass was inversely associated with BMI (p=0.001), with fat mass (p=0.001), and with body fat percentage, (p=0.001). CHO intake/kg total mass was positively associated with the total number of hours spent in an anabolic (i.e., EB>0) state (p=0.001), and was inversely associated with the total number of hours spent in a catabolic (i.e., EB < 0) state (p=0.001). CHO intake/kg total mass was the only substrate to be significantly correlated with the number of hours spent ± 400 kcal EB over a 24 hour period (p=0.001). Z-scores were created to establish categories of body composition and energy balance values. Utilizing Chi-Square tests, it was determined that more hours spent in an energy surplus (> 400kcal) was associated with higher body fat percent (p=0.042). Conclusions: CHO restriction, whether done intentionally or as a function of an energy restrictive intake, was commonly observed in this subject pool. Of the females surveyed, 79% did not meet their daily energy needs and, on average, consumed 49% of the recommended daily intake of CHO established for active people. The findings that subjects with lower CHO intakes had higher body fat levels, and that CHO was associated with improved maintenance of energy balance, which was also associated with lower body fat percent, suggest that physically active women should not restrict CHO to achieve a desired body composition. It was also observed that end-of-day energy balance was not associated with either energy substrate consumption or body composition.

Banking industry has been playing a significant role in the process of economic development of Taiwan. There exists an inseparable relationship between its operation quality and the steady growth of national economy. Nevertheless, the drastically increasing numbers of bank in the recent years have caused a vicious competition in this industry. Meanwhile, the global economic recession also leads to the continuous rise of the percentage of nonperforming loans, ever-reducing profit earning as well as the financial chaos and crisis in the banking industry. This study adopts a more reliable efficiency evaluation method------ Data Envelopment Analysis(DEA) to judge the operation performance of banks. Besides, the CCR model and BCC model in DEA (Data Envelopment Analysis) are employed to explore whether the source of low efficiency comes from pure technical efficiency or from scale efficiency. Through slack variable analysis, this study suggests some room of improvement for the input and output of the low-efficiency banks. In view that the percentage of nonperforming loans is one of the essential factors for evaluating the operation performance of banks, this study puts it in the output variable of DEA so as to overcome the paradoxical isotonicity being caused. After verification is made by the study, the results show that it is really workable.

Abstract The main purpose of this study is to investigate, through the ¡§Series¡¨ unit in the first volume of high school mathematics, the impact of the updated-version textbooks on the learning effectiveness of students so as to provide references to be used by future revisers of teaching materials. The research subjects were drawn from Kaohsiung City high schools. Three schools were selected (two public and one private). These three high schools, from the 1999 academic year, began to use mathematics textbooks of three different versions, one to each school. A first-year class and a second-year class were drawn from each school, and there were six classes in total. In total there were 237 students receiving the experimentation. The students from the same school, no matter they were first-year or second-year, had been taught by the same teacher in respect of the unit that was included in the study. Among the three schools, School ¥Ò®Õ admitted for the 1998 and 1999 school years new students with scores higher than 580 on the joint entrance examination, School ¤A®Õadmitted for the 1998 and 1999 school years new students with scores from 530 to 580 on the joint entrance examination, and School ¤þ®Õadmitted for the 1998 and 1999 school years new students with scores lower than 530 on the joint entrance examination. The three new versions used in the study were published by Nan-I Book Store, Ta Tung Information Enterprise, Co., Ltd., and Han Lin Publisher respectively. They are high school mathematics textbooks examined and approved by the Ministry of Education (excluding Teacher¡¦s Manual, supplementary materials, journals, reference books, etc). The outdated version was published by National Bureau of Editing and Translation. Mathematics Test written by the researcher was used as the testing tool, and the t-test method was adopted. Moreover, because the variances of the file data collected from a school fluctuated tremendously and the sample space was unequal, a single-factor ANOVA analysis was conducted to indicate whether there existed a significant difference between classes in terms of their learning achievement. The ¡§Questionnaire on Mathematics Opinions¡¨ was administered and the results were represented by ¡¥5¡¦, ¡¥4¡¦, ¡¥3¡¦, ¡¥2¡¦, ¡¥1¡¦, which indicated their scores. An ANOVA analysis was conducted to investigate the differences in students¡¦ learning attitude. After synthesizing the analytical results of this study, the author has come up with the following conclusions: ¤@¡BIt is known from t-test and ANOVA analysis: in School ¥Ò®Õ, there existed a significant difference between the first-year and the second-year students in respect of their learning effectiveness; in School ¤A®Õ, there existed no significant difference between the first-year and the second-year students in respect of their learning effectiveness; in School ¤þ®Õ, there existed no significant difference between the first-year and the second-year students in respect of their learning effectiveness. ¤G¡BAccording to the correct percentage of each question made by students and the error pattern, the test questions were classified into three types: concept, calculation, and comprehensive application (including questions that integrated comprehension, analysis, and application abilities). It was found from the classification of questions that: (1) Textbooks have no impact on both the well-established (School ¥Ò®Õ) and the weak (School ¤þ®Õ) students in respect of the effectiveness of learning concept questions. There is no difference between the outdated and the updated-version textbooks. However, as far as the students with ordinary achievement (School ¤A®Õ) are concerned, the updated-version textbooks have a significant impact on learning effectiveness. (2) Textbooks have an impact on both the well-established (School ¥Ò®Õ) and the weak (School¤þ®Õ) students in respect of the effectiveness of learning calculation questions, and the first-year students have benefited more from the updated-version textbooks than have their second-year counterparts in terms of the effectiveness in learning calculation questions. However, as far as the students with ordinary achievement (School ¤A®Õ) are concerned, the updated-version textbooks have a significant impact on the effectiveness of their learning to reduce cycle fractions, but have no significant impact on their learning other calculation questions. (3) Textbooks have an impact on the effectiveness of learning comprehensive application questions, as far as both the well-established (School¥Ò®Õ ) and the students with ordinary achievement (School¤A®Õ) are concerned. Moreover, the first-year students have benefited less from the updated-version textbooks than have their second-year counterparts in terms of the effectiveness in learning comprehensive application questions. However, as far as the weak students (School¤þ®Õ) are concerned, the updated-version textbooks have a significant impact on the effectiveness of their learning comprehensive application questions. In particular, the updated-version textbooks have an especially significant impact on the effectiveness in learning the comprehensive application questions of regression principle. ¤T¡BFrom investigating the difference between attitude toward and effectiveness in learning mathematics as far as different groups are concerned, it has been found: There existed a significant difference between learning attitude and learning effectiveness as far as the first-year and the second-year students of School ¥Ò®Õ are concerned. There existed no significant difference between learning attitude and learning effectiveness as far as the first-year students of School ¤A®Õare concerned. There existed a significant difference between learning attitude and learning effectiveness as far as the second-year students of School ¤þ®Õ are concerned. There existed no significant difference between learning attitude and learning effectiveness as far as the first-year and the second-year students of School ¤þ®Õ are concerned. ¥|¡BFrom the correlation between the learning attitude (consisting of five sub- categories: learning habit, learning process, learning desire, learning method, and test preparation) and the learning effectiveness of the first-year and second-year students of the three schools, it has been found: ¡]1¡^ As far as the correlations between their learning effectiveness and their learning desire and between their learning effectiveness and their learning method are concerned, there existed a significant difference between the first-year and the second-year students. Because both learning desire and learning method are related to textbooks, the updated-version textbooks had a significant impact on the learning effectiveness of the first-year students of School ¥Ò®Õ. ¡]2¡^ On the average, there existed a significant correlation between the learning effectiveness of the second-year students of School¥Ò®Õ , School ¤A®Õ, and School ¤þ®Õ and each of the five sub-categories of learning attitude (learning habit, learning process, learning desire, learning method, test preparation). Their performance on each sub-category of learning attitude is more or less related to textbooks. Therefore, there existed a significant correlation between the updated-version textbooks and each of the five sub-categories of learning attitude. ¡]1¡^ The correlation between learning effectiveness and learning method presented by the first-year students of School ¥Ò®Õ and School ¤þ®Õ (especially School ¥Ò®Õ) is more significant than that presented by their second-year counterparts. ¡]4¡^ The correlation between learning effectiveness and test preparation presented by the first-year students of School ¥Ò®Õ, School ¤A®Õand School ¤þ®Õ is more significant than that presented by their second-year counterparts.

Aim:  Optimize the usage of transactional emails, going beyond their communicative nature and combining it with marketing purposes. The project has been developed in collaboration with Incomedia, Italian software developer and vendor. Objective: Understand how Incomedia can exploit the benefits of transactional emails in a marketing perspective in order to increase the sales of its software. Limitation: The specificity of the topic, strictly related to Incomedia’s activities, products and consumers. Limits of time and variables tested with the A/B experiment. Theory/Methodology : It helped us to leverage the potential of transactional emails through the improvement of one particular element, the price discount offers. Due to the particularity of the software “medium-price” level, we have choose to do an A/B test experiment of the new transactional email by presenting the discount in two different ways: monetary and percentage terms. Result: The new transactional email, with the price discount, drove us to satisfactory results. The price discount expressed in percentage was better perceived and accepted by consumers; thanks to this, Incomedia during the experiment could highly increase its sales.

The primary prerequisites for powder mixtures/granules intended for tableting is to posses the quality of (i) homogenous composition; (ii) acceptable flowability, (iii) sufficient compressibility; and (iv) anti-adhesiveness. The most important prerequisite for these powder mixture/granulates is undoubtedly the ability to flow, due to its effect on product quality, especially dose and dosage form uniformity. A comprehensive literature study on the flowability of powders revealed that flow is affected by physical properties such as molecular- and interparticle forces, particle size and size distribution, particle shape, particle density, surface structure of the particle, and particle packing geometry. Various flow tests are available to determine powder flow, each measuring a variety of the properties mentioned above, resulting in different flow results and a subsequent variation in the classification of powders. Particle characterization of a wide range of pharmaceutical fillers through SEM and particle size analysis, indicated considerable differences between physical properties of the various fillers, which suggested significant differences in their flow behaviour. Flow tests were conducted determining the critical orifice diameter (COD); percentage compressibility (%C); angle of repose (AoR) and flow rate (FR) of the fillers in the absence and presence of a glidant (0.25% Aerosil® 200). The results confirmed the expected differences in flow obtained from the various tests, with no one of the fillers achieving the same flow behaviour in all the tests. The difference in flow amongst the fillers for a specific test could, to a large extent, been correlated with specific physical properties of the particles within the powder bed. COD results illustrated the influence of particle size and shape and surface structure on the flowability of these materials, with fillers with a smaller average particle size, less spherical shaped particles and uneven / rough surface structures performing poorer than their counterparts. The percentage compressibility (%C) of the materials was affected by the shape and size of the particles and the density of the materials, whilst the packing geometry also affected flow behaviour. Particles with high density and a low internal porosity tended to posses free flowing properties. Powders with a larger difference in the ratio between their respective bulk and tapped densities/volumes presented better flow results. The AoR of the fillers was affected by the cohesiveness and friction between the particles as well as the shape, surface structure and size of the particles. This method was less discriminative in terms of indicating differences in the flow of powders with comparable physical properties. A further drawback of this method was the variation in results between repetitions, which is affected by the way the samples were handled prior to measurement. The flow rate (FR) of the fillers was predominantly affected by the density of the materials and the size, shape, and surface structure of the particles. Powders with a higher density seemed to exhibit a better flow rate, although some of the other factors affected the flow rate more when the densities were very close or identical. The following general rank order for the various fillers (as an average of their performance in all the tests) were established (with no glidant present): Cellactose® 80 > FlowLac® 100 > Prosolv® HD90 * Ludipress® > Emcompress® >Avicel® PH200 > Starlac® » Emcocel® 50M * chitosan » lactose monohydrate. Addition of a glidant failed to change the rank order significantly. During the final stage of the study an attempt was made to modify and/or refine the composite flow index (CFI) proposed by Taylor ef a/. (2000:6) through (i) inclusion of flow rate results in its computation and/or (ii) varying the contribution (percentage) of each test to the CFI (Taylor & co-workers used equal contributions, namely 33 V* %, in their calculation of the CFI). The results indicated that including the results from the flow rate test was not beneficial in terms of providing a more representative CFI (in fact it reduced the accuracy of the index). Next various weight ratios for COD, %C and AoR was used to determine the CFI of each filler, and an optimum ratio was found at 50%:40%:10% (COD:%C:AoR) resulting in the highest CFI for each powder and the widest range for the CFI (largest difference between minimum and maximum values). This ratio was found in the presence and absence of a glidant. At this ratio the CFI discriminated well between the different powders in terms of their flowability. Lastly, the flowability scale for powders as used by the USP (20007:644) for %C and AoR results was adapted and fitted on the CFI results obtained for the various powders. This scale provided an exceptional fit for the powders both in the absence and presence of a glidant) and offered an excellent means for the grouping and classifcation of powders based on their CFI. / Thesis (M.Sc. (Pharmaceutics))--North-West University, Potchefstroom Campus, 2009.

The primary prerequisites for powder mixtures/granules intended for tableting is to posses the quality of (i) homogenous composition; (ii) acceptable flowability, (iii) sufficient compressibility; and (iv) anti-adhesiveness. The most important prerequisite for these powder mixture/granulates is undoubtedly the ability to flow, due to its effect on product quality, especially dose and dosage form uniformity. A comprehensive literature study on the flowability of powders revealed that flow is affected by physical properties such as molecular- and interparticle forces, particle size and size distribution, particle shape, particle density, surface structure of the particle, and particle packing geometry. Various flow tests are available to determine powder flow, each measuring a variety of the properties mentioned above, resulting in different flow results and a subsequent variation in the classification of powders. Particle characterization of a wide range of pharmaceutical fillers through SEM and particle size analysis, indicated considerable differences between physical properties of the various fillers, which suggested significant differences in their flow behaviour. Flow tests were conducted determining the critical orifice diameter (COD); percentage compressibility (%C); angle of repose (AoR) and flow rate (FR) of the fillers in the absence and presence of a glidant (0.25% Aerosil® 200). The results confirmed the expected differences in flow obtained from the various tests, with no one of the fillers achieving the same flow behaviour in all the tests. The difference in flow amongst the fillers for a specific test could, to a large extent, been correlated with specific physical properties of the particles within the powder bed. COD results illustrated the influence of particle size and shape and surface structure on the flowability of these materials, with fillers with a smaller average particle size, less spherical shaped particles and uneven / rough surface structures performing poorer than their counterparts. The percentage compressibility (%C) of the materials was affected by the shape and size of the particles and the density of the materials, whilst the packing geometry also affected flow behaviour. Particles with high density and a low internal porosity tended to posses free flowing properties. Powders with a larger difference in the ratio between their respective bulk and tapped densities/volumes presented better flow results. The AoR of the fillers was affected by the cohesiveness and friction between the particles as well as the shape, surface structure and size of the particles. This method was less discriminative in terms of indicating differences in the flow of powders with comparable physical properties. A further drawback of this method was the variation in results between repetitions, which is affected by the way the samples were handled prior to measurement. The flow rate (FR) of the fillers was predominantly affected by the density of the materials and the size, shape, and surface structure of the particles. Powders with a higher density seemed to exhibit a better flow rate, although some of the other factors affected the flow rate more when the densities were very close or identical. The following general rank order for the various fillers (as an average of their performance in all the tests) were established (with no glidant present): Cellactose® 80 > FlowLac® 100 > Prosolv® HD90 * Ludipress® > Emcompress® >Avicel® PH200 > Starlac® » Emcocel® 50M * chitosan » lactose monohydrate. Addition of a glidant failed to change the rank order significantly. During the final stage of the study an attempt was made to modify and/or refine the composite flow index (CFI) proposed by Taylor ef a/. (2000:6) through (i) inclusion of flow rate results in its computation and/or (ii) varying the contribution (percentage) of each test to the CFI (Taylor & co-workers used equal contributions, namely 33 V* %, in their calculation of the CFI). The results indicated that including the results from the flow rate test was not beneficial in terms of providing a more representative CFI (in fact it reduced the accuracy of the index). Next various weight ratios for COD, %C and AoR was used to determine the CFI of each filler, and an optimum ratio was found at 50%:40%:10% (COD:%C:AoR) resulting in the highest CFI for each powder and the widest range for the CFI (largest difference between minimum and maximum values). This ratio was found in the presence and absence of a glidant. At this ratio the CFI discriminated well between the different powders in terms of their flowability. Lastly, the flowability scale for powders as used by the USP (20007:644) for %C and AoR results was adapted and fitted on the CFI results obtained for the various powders. This scale provided an exceptional fit for the powders both in the absence and presence of a glidant) and offered an excellent means for the grouping and classifcation of powders based on their CFI. / Thesis (M.Sc. (Pharmaceutics))--North-West University, Potchefstroom Campus, 2009.