Bachelor of Science

The Travelling Salesman Problem (TSP) is one of the classic problems in combinatorial optimisation. Its applications arise in problems like the delivery services, customer pick-up services, circuit board drilling, and even in X-ray cryxtallography experiments. As most of these applications involve an open tour, we shall therefore discuss the TSP as one involving an open tour.

TSP basically concerns finding the shortest total distance possible. Since the search for the shortest total distance is not easy to obtain especially when the number of cities is large, lower and upper bounds are introduced to at least find a near-optimal or an optimal solution.

In this paper, we used the Prim's algorithm and a particular heuristic algorithm to compute the lower and upper bounds respectively. We coded these algorithms on FORTRAN 77 so that we could investigate how much the upper bound varies fiom the lower bound. By taking the ratio of the upper bound to the lower bound, we could conduct an investigation on the probability distribution of the percentage of the upper bound exceeding the lower bound for different sets of cities.

Our first attempt at computing the upper and lower bounds shows that the former is on average about 30 percent above the latter, while for certain set of cities it maybe as high as 45 percent above. This percentage indicates that we need to improve the lower bound andlor the upper bound. The implementation of the sub-gadient optimisation method on the lower bound shows a relatively good improvement to 20 percent (maximum 30 percent). However, this can be further improved using a sub-case of the 3-opt algorithm on the upper bound to about 8 percent (maximum 16 percent). Though this 8 percent illustrates that the TSP solution has si@cantly improved, fiuther improvements can still be made by using the 3-opt algorithm or some other heuristic algorithms.

Solid state lasers have undergone a renaissance since the development of reliable and cheap laser diodes which can be used as pump sources. This leads to a new class of holosteric lasers which have been demonstrated to be highly efficient, reliable, and are attractive for a wide range of applications. The work described in this academic exercise focuses on the development of a stable gain-switched Nd:YLF laser operating at 1053 nm pumped with a 1.2 W diode array.

A brief history and an overview of diode-pumped solid-state lasers are presented as part of the literature review in the first two chapters. The properties that make LDPSSLs novel and such a fertile field of research are highlighted. Also included are the properties of semiconductor laser diodes and the laser material, Neodynium-doped Yittrium- Lithium- Fluoride.

Chapter three discusses in detail, the general design criterion for the laser setup as well as the rate-equations model which is used as a basis for the simulation program.

In chapter four, the experiments and measurements obtained are presented. The resonator used is a folded, three-mirror cavity resonator in the Fabry-Perot configuration. In this AE, we have demonstrated a diode-pumped Nd:YLF laser operating at 1053 nm in both continuous wave ( CW ) operation and pulsed operation via gain-switching. In the TEM00 operation, the laser produces an maximum output power of 13 mW with a 12.9% transmission output coupler, with a slope efficiency of 1.4%. In the gain-switched operation, generations of single, double and multiple pulses have been observed. These pulses are well simulated by the rate equations model. In the single-pulse generation, pulses of 16 ps duration have been obtained. This is in fairly good agreement with the simulation results which have produced a pulse duration of 12 ps.

This study has been undertaken to determine the general levels of some toxic metals in magazines. The objective of this work was to initiate and stimulate the study of heavy metals due to coloured pigments as well as that of magazines as a whole. The concentration of three toxic metals were determined by atomic absorption method using standard solutions for calibrations. Microwave digestion was employed to release the trace metals into the acid. The results show that three colours in local magazines contain high level of Pb ranging from 1.6 μg/g to 59.0 μg/g. Cr appeared somewhat lower which recorded between 0.4 and 16.1 μg/g. The range of Cd falls between 0 and 7.4 μg/g. The mean total Pb, Cr and Cd concentrations of the various magazines are as high as 12.85, 8.89 and 0.91 mg per copy respectively. These multiplied by their annual sales volume thus indicate that magazine disposal is a source of environmental pollution.

A preliminary study is also undertaken to compare microwave and ultrasound digestion using magazine and paint flakes. Statistical evaluation by regression treatment was applied and high correlation coefficients at 95% confidence level which indicate the good accuracy and precision of the alternative method compared to the standard microwave digestion.

Eccentric muscular actions occur as a result of muscles lengthening during contraction. Previous studies involving such muscular activities have focused on aspects like delayed onset muscle soreness (DOMS), enzyme activity, electrically stimulated muscle characteristics, muscle fibre damage and neuromuscular adaptations (MacIntyre et al., 1995). Although these studies measure physiological components and strength changes, not many researchers have evaluated changes in power after a single intense exercise session.

The present study was designed to measure strength and power following an acute bout of isolated concentric or eccentric leg press exercise. Twenty four subjects (mean age + SD= 23.9 t1.7 years) from the School of Physical Education, Nanyang Technological University were randomly assigned into three groups (n=8). The concentrically exercised (CON) and eccentrically exercised (ECC) groups performed a leg press exercise consisting of 3 sets of 12 repetitions of their maximum isometric strength determined by a modified dynamometer device. Power tests consisted of countermovement jumps (CMJ) and squat jumps (SJ) measured on a force plate, and strength was assessed using isometric leg press dynamometry. Additionally, average peak power and total work on a cycle ergometer for 10 seconds were measured pre-exercise and nine times post exercise. Muscle soreness measurements were made six times pre- and post exercise.

Results revealed greater strength decrements for ECC than CON throughout the study. CMJ heights for both groups show no significant difference (p<.01) and were below baseline even after 48h. SJ heights decreased less than CMJ heights and recovered to baseline by 24h. In peak power and total work, CON recovered to baseline by 2h compared to ECC who recovered by 24h. There was no indication of muscle soreness or DOMS.

Further research needs to be carried out in the area of power changes related to different exercise intensities and muscle groups. Studies should also be designed to examine the relationship of force and velocity in power generation following acute exercise.

A series of 3-(ary1amino)phthalides of the type RC₆H₄NH-CHC₆H₄COO, (R=H; 2-OH; 4-OH; 2-OCH₃; 4-OCH₃; 2-CH₃; 4-CH₃; 2-Cl; 4-Cl; 2-NO₃; 4-NO₂) were synthesized and characterized by 'H, "C NMR and IR spectroscopies. Seven of these phthalides are new compounds.

A convenient method for synthesizing the phthalides is by condensation reaction of 2-formylbenzoic acid with the appropriate substituted aromatic amines in methanol.

IR spectroscopic analysis using KBr disc suggest, in the solid state, that the condensation products exist in the phthalide (closed) form instead of a Schiff base (open form).

'H and ¹³C NMR spectroscopic analysis of the compounds in CDC1, and DMSO-d, confirm that the phthalide form is the predominant form.

In liquid CHCI₃, IR spectroscopy was able to detect both the open and close forms of 3-(0-hydroxyani1ino)phthalide and 3-(0-methoxyani1ino)phthalide thus indicating the presence of both ring and chain tautomers. It has been shown qualitatively that these two compounds exist as a form of ring-chain tautomerism with the ring tautomer being more predominant.

Onchidium tumidium is an osmoconformer and that it is able to regulate its volume at 10% or 90% seawater (SW). However, it is a more effective volun2e regulator in 90% SW than in 10% SW. Results indicate that after an upward transfer from 50% to 90% SW, there were significant increases in the total free amino acids (TFAA) in the whole animal, coelomic plasma (CP), body wall (BW) and internal organs (10). Increased proteolysis was more likely to occur in the I0 than in the BW of 0. tumidium exposed to 90% SW as reflected by a lowered water-soluble protein content in the I0 of slugs exposed to 90% SW compared to those of slugs exposed to 50% SW. It is probable that the higher TFAA in the BW of O.tumidium exposed to 90% SW compared to that of slugs exposed to 50% SW was partially due to NH4+ trapping by α-ketoglutarate in the presence of glutamate dehydrogenase to form glutamate as indicated by a significantly higher amination deamination-1 ratio. Glutamate could then be channeled to form other amino acids via transaminases. When being confronted with a downward transfer from 50% SW to 10% SW, there was possibly a release of free amino acids (FAA) from the BW of 0. tumidium into the CP as indicated by the increase in TFAA in the CP and a decrease in TFAA in the BW. The significantly lower amination deamination-1 ratio in the I0 shows that it is most probably involved in the breakdown of FAA during hypoosmotic stress.

The purpose of the present study was to determine the validities of the 1600-m run and several anthropometric variables for predicting maximal oxygen uptake (VO2max) in Singapore children aged 10-12 years. Fourteen untrained subjects, 12 boys and 2 girls, performed the 1600-m run on a synthetic track and had VO2max determined in the laboratory. A weak correlation of -0.398 (p > 0.10) was found between VO2max and the 1600-m run time. In fact, subscapular thickness alone has the highest correlation coefficient with VO2max in this group (r = -0.863, p < 0.001). Multiple regression analysis (best-subsets) was conducted and the best predictor variables were height, weight, triceps thickness, subscapular thickness and suprailiac thickness resulting in two prediction equations: (1) VO2max (ml.kg-1.min-1) = 79.9 - 16.6 (height, m) - 0.443 (triceps thickness, mm) -0.902 (subscapular thickness, mm)+ -0.358 (suprailiac thickness, mm), with R2 = 0.864, SEE= 2.189 (p < 0.05). and (2) VO2max (ml.ki1.min-1) = 61.8 - 0.212 (weight, kg) -0.425 (triceps thickness, mm) -0.667 (subscapular thickness, mm) + 0.315 (suprailiac thickness, mm) with R2 = 0.857, SEE = 2.248 (p < 0.05). This study thus suggests that three skinfold sites (triceps, subscapular and suprailiac), height and weight are better predictors ofVO2max than the 1600-m run.

The Family Agamidae (Order: Squamata) in Peninsular Malaysia and Singapore is represented by 7 genera and 25 species of lizards. Nine species of lizards from the 3 genera: Acanthosaura, Calotes, and Gonocephalus, were karyotyped in this study. The karyotypes of five lizards, Acanthosaura armata, Gonocephalus bellii, Gonocephalus liogaster, Gonocephalus chamaeleontinus, and Gonocephalus robinsonii were documented here for the first time. No heteromorphic sex chromosome was evident in the species studied. The karyotypes showed chromosomes of two distinct size groups: macrochromosomes and microchromosomes. The number of macrochrolnosomes within each genus was invariant. Of the 3 genera karyotyped, one genus has diploid numbers more than the ancestral karyotype 2n = 36, while the two other genera had less than the 2n = 36 chromosomes.

Karyological data was used to discuss the taxonomic status of species in the Calotes and Gonocephalus species complex. Calotes cristatellus has a karyotype formula 2n = 34(14M + 20m) while other Calotes species have a karyotype of 2n = 34(12M+20m). In light of the karyological data from this and other published studies, there is justification to reassign Calotes cristatellus to the genus Bronchocela. Karyological results were also used to hypothesize possible modes of speciation of Gonocephalus species on either side of the Wallace line.

This study was composed of three parts with the overall aim being to investigate the effects of drop height and rest period on the kinetic measures associated with the performance of 40 drop jumps, a commonly used stretch-shortening cycle (SSC) activity for the development of strength and power The purpose of Part One was to determine an appropriate drop height from between 40 to 70 cm; Part Two investigated the relationship between the kinetic variables and isokinetic strength and power measures and functional estimates of muscular power; and Part Three investigated the effect of a one minute rest between each set of ten drop jumps on the kinetic measures of 40 drop jumps.

Eight male subjects aged 18.7 2 0.5 years were selected from a Singapore Junior College track and field team to participate in this study with all subjects considered to be sprint and jump specialists In Part One, subjects performed one set of 10 drop jumps from each of four heights, specifically, 40 cm (DJ40), 50 cm (DJ50), 60 cm (DJ60) and 70 cm (DJ70). The ground contact time, flight time and peak vertical ground reaction forces were measured from a Kistler mobile force platform (Model 9286) with the rebound height being subsequently calculated. Results of paired t-tests revealed no significant difference in the mean rebound height across the four drop heights, however; based on ground contact times and vertical ground reaction forces the use of a 40 to 50 cm drop height was suggested. A drop height of 50 cm was therefore selected as it was thought this would elicit maximal benefits with respect to the SSC movement, while still being considered safe.

Having established the drop height, one week later subjects were requested to complete 40 consecutive drop jumps and one week later again, subjects completed 40 drop jumps with a one minute rest period between each set of ten jumps. Changes in the kinetic variables of coupling time, ground contact time, rebound height, peak vertical ground reaction force, magnitude of first impact spike and angle of knee flexion were measured across the 40 drop jumps for each condition.

The kinetic variables analyzed from the 40 consecutive drop jumps were correlated with the maximum peak torque and average power measurements obtained during a concentric-eccentric knee extension-flexion task on a Cybex 6000 dynamometer at 180 and 240 degrees per second and functional estimates of muscular power (50 m sprint time, vertical jump height and power). Pearson product-moment correlation analysis revealed a positive relationship between the coupling time of the drop jump and the two performance measures of 50 m sprint and vertical jump power, suggesting that subjects may have been using a greater proportion of slow-twitch muscle fibres when performing these activities. In addition, it was suggested that the isokinetic measurements obtained may not be good indicators of drop jump ability.

The final part of this study investigated the effect of a rest period on the kinetic changes across the 40 drop jumps. Changes in the kinetic variables (as measured in Part Two) were analyzed using a MANOVA statistics for the two different treatments across the 40 jumps. A Student's t-test was subsequently used to determine the differences in the kinetic variables between the two treatments.

No significant changes were observed in the kinetic variables measured across the 40 drop jumps in either condition, with no significant difference in the kinetic variables between the two conditions. Despite this, there was a trend of reduced variability in the kinetic measures associated with the drop jump series with one minute rest intervals, compared to the 40 consecutive drop jumps.

In conclusion, this study recommends the use of a 40 to 50 cm drop height when implementing drop jump training. In addition, despite no significant changes being observed in the kinetic variables of interest, the reduced variability observed when the drop jumps were performed with a one minute rest interval, may suggest that this protocol is beneficial in minimizing "undesirable" jumps when performing a series of drop jumps. Lastly, coupling time was found to be closely related to speed and power performance and thus further studies should investigate the possible relationship between coupling time and muscle fibre composition. If a relationship is established between these variables, this simple kinetic measure may be used in preference to the current invasive technique of muscle biopsy for the prediction of muscle fibre composition.

The plasma focus, beside being used as a fusion device, serves as a simple yet cost-effective device to study the fundamentals of pulse technology and plasma physics as a result of a wide range of plasma phenomena arising from this device. Furthermore, the plasma focus can be a source of fusion neutrons and X-rays depending on the working gas used. As these radiations from the focus have many potential uses, it would be useful to optimize these radiations. Although the mechanism concerning the production of radiation is not clear, the process of radiation could be logically deduced to require a store of energy. This project seeks to find the operational parameters for optimization of kinetic energy and energy input into the plasma.

In a multi-racial society, the identity of the consumer meat sold has a great impact on the different religions and races. In this study, various molecular techniques such as restriction endonucleases digestion, rDNA PCR, RAPD PCR and protein PAGE were used on 11 different meat samples were analyzed for species identification. The species of meat samples include chicken, duck, pig, sheep, ostrich, cow, buffalo, kangaroo, horse, wild boar and deer. Differentiation between the different species were based on DNA and protein banding patterns. From the results obtained, it seemed that RAPD PCR was the best method for species identification and differentiation. Restriction endonucleases digestion could be used to identify the species origin of animals, but the quality of the DNA used must be pure and flee from contaminants. For rRNA analysis, comparison between the species can only be done on the same gel, since only one major band is visible under UV light. Protein analysis of the species is easier to perform than the rest of the methods, but due to the close proximity of the bands, species identification and differentiation is difficult. Other than for species identification the same techniques can be used the characterization, construction of genetic and physical maps, forensic casework, wildlife conservation as well as for investigating the relationship between two geographically distinct species. In addition to these techniques, other methods such as FINS, immunological assay and hybridization probes are also used for identification of the origin of the species.

The Riemann integration is well-known. It is also well-known that if a sequence of Riemann I ntegrable functions converges uniformly to a limit function, then the limit function is also Riemann integrable. However, there are other convergence theorems for the Riemann integral, some of which were not so well-known and some were even discovered recently. We shall look into few such convergence theorems.

The project is divided into two parts, viz.

(i) synthesis and characterization of cobalt complexes with piperazine derivatives and,
(ii) synthesis of 1,Cdiphenyl-1.4-dilphosphorinane.

Stable hydrate isomers of icobalt(I1) complexes were obtained with 1-methylpiperazine acting as a chelating agent. The results indicated that the heterocyclic ligand is effective in stabilizing the +2 oxidation state of cobalt or the formation of cobalt complexes with low coordination number. This is probably associated with the structure of the ligand. Two long-chained cobalt complexes were also synthesized in which the hydrophobic group is made up of piperazine derivatives.

The second part of the project involves the preparation of 1,Cdiphenyl-1,4-phosphorinane. In an attempt to mimic a iiterature method (Hinton and Mann) [15], ethylenebis(triphenylphosphonium) dibromide was treated with sodium borohydride and the results showed that the pheny! group cannot be displaced in contrast with a benzyl group which was removed in the reported method. Hence, the use of organometallic reagents prepared from magnesium and lithium were employed. The results indicated the possibility of using these reagents for the synthesis of the named compound.

The Vehicle Routing Problem (VRP) involves the design of a set of vehicle routes that operates with minimum cost. In many practical examples, the capital costs involved are usually dominant. In my particular variant, the assumption is that most of the capital cost is from purchasing vehicles, so the objective will be to minimize the number of vehicles required.

To solve this problem, we adopted heuristic methods, due to the computational complexity of this problem type. Since optimal solutions are almost impossible to obtain with a reasonable time for large number of cities, near-optimal solutions were sought instead, by using upper and lower bounds.

To compute the upper bound, we found a tour of the pick-up points using a greedy algorithm and then constructed a schedule using a sector partitioning approach. To compute the lower bound, certain constraints were relaxed and a simpler but unrealizable problem was solved. Then the 3-opt algorithm was used to improve both the upper and the lower bounds. These algorithms were tested by a program written in FORTRAN 77 and we analyzed the results by plotting a histogram of the ratios of the upper to lower bounds, for a number of randomly generated instances of the problem. These ratios reflect the relative difference between the two bounds, which in turn indicates the quality of the heuristic solution.

The results for simulations run on similar sets of constraints improved when the 3-opt algorithm was used. However, the histograms did not remain the same for simulations run on different sets of constraints, using the 3-opt algorithm. The lower bound was found to be the likely culprit for the poorer results.

Though the 3-opt algorithms used have shown improvements in the results, we shall show how our approach may be extended to improve the results still further.

Silicon nitride (P-Si3N4) films were synthesised on single crystal silicon (100) substrates using a low frequency, high density, inductively coupled radio frequency plasma source in a mixture of argon, hydrogen and nitrogen plasmas. The influence of DC substrate bias and the processing time on the property of the processed films was investigated. The main mechanisms involved in the process were ion implantation combined with plasma enhanced thermal diffusion. The properties and characteristics of the films analysed using various advanced analytic tools that included infrared spectroscopy (R),X -ray diffraction (XRD), energy dispersive X-ray (EDX) and scanning electron microscopy (SEM). In addition, surface hardness of the processed samples was measured using Vickers microhardness test.

All the results indicate that crystalline silicon nitride films of good quality have been synthesised through the process. Vickers hardness test results testify that the processed silicon's surface hardness has increased.

The resting metabolic rate of 19 obese and 20 nonobese adolescent boys was assessed by indirect calorimetry after an overnight fast. Each subject was measured for a minimum duration of 20 minutes. Body composition was assessed by dual energy x-ray absorptiometry and VO2peak was assessed via a treadmill walking test. The results of the measured RMR were compared with RMRs calculated from five frequently used predictive equations (Harris-Benedict equation (1919), Schofield's equation (1985), the Food Agriculture Organization World Health Organization United Nations University (FAO/WHO/UNU) equation (1) and (2) and the Molnar et al.'s equation (1995)) for adolescents. The relationship between RMR and VO2peak was also examined. The mean (± S.D.) RMR was 1889.11 + 308.13 kcal.day-' in the 19 obese boys and 1528.75 +174.08 kcal.day-' in the 20 nonobese boys. RMR was significantly correlated with lean body mass (r = 0.63, p < 0.01) in the combined group as well as in the obese and nonobese groups (r = 0.66 and r = 0.45 respectively). VO2peak was negatively correlated (r = -0.54, p < 0.01) with RMR in obese and nonobese boys when expressed per kilogram body weight but was positively correlated (r = 0.38, p < 0.05) with RMR when expressed in litres per minute. The results of the present study found that the RMRs of obese boys were higher than those of nonobese boys. RMR and lean body mass was also found to be positively correlated in both obese and nonobese boys. The RMRs predicted by the Harris-Benedict equation and the Molnar et al.'s equation gave good estimations of the measured resting values in the present group of male adolescents and both equations are recommended when actual measurements of RMR are unavailable. Furthermore, results indicated that the relationship between VO2peak and RMR depends on the way VO2peak is expressed.

This project is a study of beginning first-order model theory, up to a point sufficient for an exploration of Vaught's theorem: there is no complete first-order theory with exactly two nonisomorphic countable models. The exploration of this theorem requires, in the first place, the knowledge of the notions of a first-order language, a complete theory, models and isomorphism of models; next, the understanding of theorems such as Completeness Theorem, Compactness Theorem, Lowenheim-Skolem-Tarski Theorem and Omitting Types Theorem. Lastly, atomic models and saturated models are also given a close look at after which these results would enable us to discuss the theorem by Vaught.

Increasing affluence in Singapore during the last decade has led to a higher prevalence of coronary heart disease and obesity among Singaporeans. Recent evidence has suggested an unfavourable blood lipid / lipoprotein profile among Singaporean children and adolescents. Obesity and the physical fitness of Singaporean schoolchildren is of national concern. Very few studies, however, have examined the relationships between blood lipids / lipoproteins, obesity and cardiovascular fitness in Singaporean schoolchildren. Hence, the purpose of this study is to investigate these relationships in obese and non - obese Singaporean schoolchildren aged 13 to 15 years.

78 Chinese boys ( mean age = 13.52 + 0.6 SD ) were recruited for this study. AU subjects underwent 3 different tests : ( a ) body composition analysis - measurements of height and weight, 8 site skinfold thickness, and DEXA scanning to determine fat mass and lean body mass; ( b ) blood biochemistry analysis - drawing of 20ml of venous blood to analyse serum cholesterol, triglyceride, high density and low density lipoproteins, and apoproteins A1 and B ( c ) cardiovascular fitness testing - using the modified Balke protocol to determine 90zpeak of subjects.

Results showed that obese subjects had significantly higher levels of triglyceride and lower levels of VO2peak ( ml.kg-'min.' ) as compared to the non - obese subjects. A significant negative correlation also existed between VO2peak ( ml.kg-'.min-' ), triglyceride and obesity measures like BM1 and skinfold thickness. None of the other blood profile variables, however, had significant associations with obesity and cardiovascular fitness.

This study may be improved by examining the role of habitual physical activity on blood lipid I lipoprotein profile, obesity and cardiovascular fitness in Singaporean male adolescents.

Hydrophilic homopolymer of 2-hydroxyethyl methacrylate (HEMA) and copolymer of N-acryloyl-N'-methyl piperazine (AcrNMP) with HEMA were synthesized by free radical polymerization in dioxane solution at 75oC, using 2,2'-azobisisobutyronitrile (AIBN) as initiator.

Copolymer compositions were obtained using IR and elemental analysis. The reactivity ratios of the copolymerization were obtained by the application of linearization methods of Finemann-Ross (F-R) and Kelen-Tudos (K-T). The results derived from IR analysis clearly showed that the copolymerization yielded an alternating structure with reactivity ratios, r1= 0.263 + 0.01 1, r2 = 0.615 ± 0.097 as obtained by the F-R method and r1 = 0.227 + 0.074, r2 = 0.529 ± 0.145 by the K-T method. Microstructural data calculated statistically by the methods of Igarashi and by the mean sequence equations supported the findings that the copolymer was alternating. The reactivity ratios determined from the elemental analysis also indicated an alternating copolymerization system with r1 = 0.279 ± 0.019, r2 = 0.846 ± 0.185 (F-R) and r1l = 0.261 ± 0.109, r2 = 0.778 ± 0.242 (K-T).

Glass-transition temperatures (Tg) of the copolymers were determined by Werential scanning calorimetry (DSC). However, the results did not demonstrate any meaningful trend. This was attributed to the presence of water in the sample because of the hydrophilic nature of the copolymer. The presence of water in the copolymer will impart a plasticizing effect and this affects the value of T_g.

Crosslinked copolymers were prepared by bulk polymerization of AcrNMP and HEMA, with crosslinker ethylene glycol dimethacrylate (EGDMA) and initiator, AIBN, at 70oC. The crosslinked copolymers were hydrogels. These hydrogels were characterized by the equilibrium swelling measurements at 25 'C. The equilibrium water contents, swelling coefficients and soluble fraction were calculated.

This study has been undertaken to develop two new mediated amperometric biosensors for the determination of H202. The tissue-based biosensor was constructed by the incorporation of the Swede turnip tissue, rich in peroxidase, and ferrocene in a carbon paste matrix. The fabrication of another biosensor modified by the sol-gel immobilisation technique in a "sandwich" configuration was the other emphasis.

The detection principle of the biosensors was based on the enzymatic reduction of H202 by peroxidase and subsequently electron transfer from the electrode to the enzyme, via a redox mediator, ferrocene. The optimisation procedure for the fabrication and analytical performance of the biosensors were studied by amperometry. Possible interference and stability of the biosensors were evaluated The applicability of the biosensors was tested in recoveries of H202 in food and environmental samples.

The tissue composition and mediator loading were the primary parameters for the optimisation fabrication of the turnip tissue-based biosensor. This biosensor was endowed with an extended range of linearity from 0.06 mM to 0.68 mM and a detection limit of 0.015 mM. It exhibited a remarkable measurement reproducibility of 2.44% (n=15) and excellent stability, with no significant activity changes observed within six days of use.

The optimisation fabrication of the sol-gel modified horseradish peroxidase biosensor was focused on H20:Si (vlv), the outer and inner sol-gel film thickness, and the mediator concentration. The linear calibration range of the biosensor under optimised condition was 3 μm to 57 pM with a correlation of 0.996. Its good selectivity warranted a reasonable recovery in analytical applications. Its short response time of 16 seconds rendered it suitable for biosensing applications.