ICAPS 2020

Permanent URI for this collectionhttp://repository.kln.ac.lk/handle/123456789/21780

Browse

Now showing 1 - 20 of 116

Activation of wood biochar and red brick using natural coconut vinegar
(Faculty of Science, University of Kelaniya, Sri Lanka, 2020) Malka, U.K.M.; De Silva, R.C.L.; De Silva, D.S.M.; Chandrajith, R.
Number of studies have been carried out to determine the efficiency of strong oxidizers in activating natural raw materials used in low cost water purification processes. However, rural communities find it difficult to acquire most of such chemicals. Therefore, this study was aimed to determine the ability of natural coconut vinegar, which is a common domestic acidic solution, in activating abundantly available potential water purifying materials to reduce calcium (Ca2+) ions from water, further reducing the water hardness. In this study mature barks of Glyricidia (Glyricidia sepium), Gadumba (Trema orientalis) and Ipil Ipil (Leucaena leucocephala) were collected and air dried. These were carbonized (400-450 °C) in a closed vessel (2 hours) to produce biochar. Both biochar and brick particles in the range of 2.0-5.6 mm were selected for the analysis. For the activation these samples were soaked in natural coconut vinegar (biochar/brick: vinegar, 1:2 V/V) for 24 hours and completely dried in an oven (120 °C) for 3 hours. Laboratory scale glass columns (2 cm in diameter) were used to calculate Ca2+ adsorption and retaining capacities. Filtrates were analyzed for Ca2+ using flame photometer. Ca2+ adsorption and retaining capacities of each material were calculated per unit bulk volume of the material. Each test was duplicated, and the average was recorded. Untreated red brick and biochar of Glyricidia, Gadumba, Ipil Ipil showed Ca2+ adsorption capacities of 0.44, 0.30, 0.31, 0.27 mg cm3 and retaining capacities of 0.19, 0.01, 0.02, 0.02 mg cm-3 respectively. Activated red brick and biochar of Glyricidia, Gadumba and Ipil Ipil showed Ca2+ adsorption capacities of 0.76, 0.58, 0.68 and 0.63 mg cm-3 and retaining capacities of 0.25, 0.20, 0.23 and 0.15 mg cm-3 respectively. Increase in Ca2+ adsorption and retaining capacities were observed in all the materials tested after activation with vinegar. Further studies are continued to use the vinegar activated natural materials in a low-cost domestic drinking water purification process.
Adulteration detection of Cinnamomum verum with BarHRM technology
(Faculty of Science, University of Kelaniya, Sri Lanka, 2020) Peiris, M.A.L.M.; Silva, F.H.C.; Wijesinghe, W.R.P.
Sri Lanka is the premier exporter of the true cinnamon (C. verum) in the global market. However, Sri Lankan true cinnamon faces a major threat due to severe competition and adulteration from its substitute cassia (e.g. C. aromaticum). It costs one-third of the price of C. verum but it contains coumarin which is a hepatotoxin at substantial amounts (up to 5%) whereas true cinnamon has only trace amounts (about 0.004%). Therefore, it is paramount to detect adulteration of C. verum from its substitute to protect the reputation of true cinnamon. Chemical and morphological methods can detect the adulteration of C. verum but when it comes to admixtures and value-added products, morphological and chemical methods are not accurate. Hence, the objective of the research was to develop a molecular assay to detect adulteration in commercially available cinnamon products. In this study, DNA sequences of C. verum and C. aromaticum were extracted from the National Center for Biotechnology Information (NCBI) using the keyword “Cinnamomum” and selected barcode region “rbcL”. Gene-specific novel markers were manually designed targeting the identified diagnostic SNP sites. Primer properties were analyzed using NetPrimer software and primers with the best qualities were selected. DNA extraction of cinnamon was done using CTAB method with slight modifications. Real-time PCR and melting curve analysis at 65 ⁰C to 95 ⁰C with a ramping rate of 0.05 ⁰C (Qiagen, Germany) was performed. The melting curve analysis and principal component analysis of the data demonstrated a clear distinction between the two species and results confirm that rbcL gene-specific primers can be used to distinguish C. verum from C. aromaticum. Further, this assay has a great potential to quantify adulterants in commercially available cinnamon samples and extremely valuable for an accurate and rapid adulteration detection of cinnamon value-added products in the global and local market.
Aloe vera gel and cinnamon essential oils-incorporated Aloe vera on stem-end rot control of mango (cv. Karthakolomban) using dip treatment
(Faculty of Science, University of Kelaniya, Sri Lanka, 2020) Karunarathna, N.S.N.; Abeywickrama, K.
Stem-end rot (SER) is a major post-harvest disease of mango worldwide. Naturally occurring biologically active compounds from plants are expected to be more suitable and less harmful than synthetic fungicides. Current research was carried out to investigate the applicability of Aloe vera gel and cinnamon bark oil and cinnamon leaf oil-incorporated Aloe vera gel in controlling SER and extending post-harvest shelf life of mango (cv. Karthakolomban). Dip treatments of A. vera gel and cinnamon bark oil (2.0 μL/mL) and cinnamon leaf oil (2.0 μL/mL) incorporated A. vera gel were carried out for 90-day old mango fruits and their pathological, physicochemical, sensory properties and percentage shriveling were evaluated after 10-day storage at 12-14 ℃. After the initial dip treatment trial, a scaling up experiment was conducted using the best treatments where treated and control mangoes were placed in ventilated corrugated fiberboard boxes instead of plastic trays to store mango at 12-14 ℃. Statistical analysis of the results was carried out using MINITAB 18 statistical software. Data with respect to physicochemical properties were analyzed using One-way ANOVA and Tukey’s multiple comparison test. Kruskal Wallis non-parametric test was used to analyze data with respect to pathological, shriveling and sensory properties. Dip treatments of Aloe vera gel in combination with cinnamon leaf and bark oils reduced SER severity of mango to 3.0% in both trials once fruits were subjected to induce ripening. A. vera gel treatment reduced SER severity of mango to 6.0%. A. vera gel, cinnamon bark and leaf oil-incorporated A. vera gel treatments significantly reduced SER severity of mango in comparison to the negative control (distilled water) which showed SER severity of 19.0%. Physicochemical properties namely total soluble solids, titratable acidity, pH, firmness and weight loss of A. vera gel and gel plus oil treated mangoes were similar to the negative control fruits according to the statistical analysis. Percentage shriveling of mango subjected to A. vera gel treatments was reduced to 0- 0.4% compared to the uncoated control fruits which indicated a shriveling of 1.6%. Sensory properties of mango did not demonstrate any drastic alteration between all treatments. Current ecofriendly treatment strategies could be introduced to horticulture industry to reduce post-harvest loss of mango in local trade, during transportation and exportation via air cargo within 10 days.
The antagonistic effect between Abamectin degrading bacteria, Staphylococcus nepalensis and Bacillus thuringiensis
(Faculty of Science, University of Kelaniya, Sri Lanka, 2020) Wijesinghe, J.M.M.; Wimalasekara, S.G.M.R.L.; Perera, T.W.N.K.
Basic ecological concepts govern the composition and the functional relationships among microorganisms of a particular ecosystem of interest. Antagonism is one such functional relationship among bacteria. A bacterium that is antagonistic to another is capable of producing molecules with different structures, cellular targets, spatial range, and mode of action that are harmful to other organisms in the same niche. Even though the antagonism is beneficial to the survival of a bacterium when considering processes such as bioremediation having multiple bacteria capable of performing the same or cooperative objectives, this effect is detrimental. Staphylococcus nepalensis and Bacillus thuringiensis are two Abamectin degrading bacteria. To test whether there is an antagonistic interaction between S. nepalensis and B. thuringiensis, the growth compatibility assay was performed using salt yeast extract (MSYE) agar plates supplemented with Abamectin (25 mgL-1 ). Parallel streak lines were made using previously isolated, Abamectin degrading, S. nepalensis and B. thuringiensis cultures and their identification was confirmed by biochemical tets. Plates were incubated for 72 hours at 37 °C. Growth inhibition zone encircling the S. nepalensis streak was observed after the incubation period. Results from this test were confirmed by measuring the Optical Density (OD) measurements at 600 nm, of MSYE broth cultures of S. nepalensis, B. thuringiensis and a combination of both. Inoculated broths were incubated in a shaker incubator (150 rpm, at 30 °C) for 72 hours, and growth was monitored by measuring OD at 24-hour intervals. OD measurements 72 hours after the inoculation (S. nepalensis - 0.19133, B. thuringiensis - 0.12500, Mixed culture - 0.12000) indicated the fact that the growth of pure cultures is higher, compared to that of the mixed culture of both organisms. Accordingly, the results of OD measurements demonstrated an antagonistic effect between S. nepalensis and B. thuringiensis. As claimed by the results of both tests it is not applicable to use them together to treat Abamectin-associated pollutions. They can still be successfully applied individually as pure cultures. However, both organisms should not be applied to the same site, at the same time, as it would be inimical towards their bioremediation capability.
Anti-diabetic activity of cinnamon (Cinnamomum zeylanicum) loaded nanoparticles
(Faculty of Science, University of Kelaniya, Sri Lanka, 2020) Sathsarani, H.M.W.K.; Jayawardena, B.M.; Dewangani, H.G.N.
Ceylon cinnamon (Cinnamomum zeylanicum) is one of the cinnamon species that shows relatively high anti-diabetic activity. “Sri Vijaya” cinnamon variety (CCSV) is an accession of C. zeylanicum and, it has been identified as a good source of anti-diabetic compounds. The aqueous extract of the quills of CCSV is rich with anti-diabetic compounds. Mainly there are two methods that are used to prepare the aqueous extract. One method is “pressured water extraction” and the other is “decoction”. According to the previously conducted researches, the aqueous extract which is prepared using “pressured water extraction” is more active than the other. Higher stability and the easiness of storage and transportation make powdered drugs and nutraceuticals preferred over liquids. However, most of the powdering techniques such as freeze drying and spray drying decrease the activity of the aqueous extracts. The objective of the present study was to synthesize a powdered nutraceutical from the pressured water extract of the quills of CCSV while conserving the anti-diabetic properties. In this study, cinnamon loaded nano-particles were synthesized using bovine serum albumin (BSA) as the base and citric acid as the cross-linking agent. Since nanoparticles are extremely small in size, their surface area is higher. Because of that their reactivity is also higher compared with the other powdered form of drugs and nutraceuticals. α-amylase inhibitory activity and the α-glucosidase inhibitory activity of nano-particles were determined using dinitrosallicylic acid assay and para nitrophenyl glucopyranoside assay respectively and the corresponding IC50 values were calculated using Graphpad prism software in order to assess the anti-diabetic properties. The inhibitory activity and IC50 values of the aqueous cinnamon extract and the positive control acarbose were determined using the same enzyme assays and they were compared with the values obtained for nano-particles. The obtained data were statistically analysed by one-way analysis of variance (ANOVA) using SPSS software package. The IC50 values of nano-particles, aqueous cinnamon extract and acarbose on α-amylase were 117.60 ± 1.73 µg/mL, 131.27 ± 1.64 µg/mL and 140.37 ± 1.17 µg/mL respectively. The IC50 values of the same compounds on α-glucosidase were 119.25 ± 0.07 µg/mL, 141.25 ± 0.21 µg/mL and 224.45 ± 0.21 µg/mL respectively. IC50 values obtained for nano-particles showed statistically significant difference compared to others. In conclusion, cinnamon loaded nano-particles showed higher inhibitory activity on α-amylase and α-glucosidase than the aqueous extract and acarbose.
Application of artificial neural network in customer analytics: A literature and classification
(Faculty of Science, University of Kelaniya, Sri Lanka, 2020) Amin, R.; Tiroshan, W. A.
Industries are relying on increasing the customer base as a means of growth of the industry, irrespective of churn prediction. It could be argued that churning of customers, have the same or even greater consequence on the company itself. According to a recent study, churn helps the business have a better roadmap about future revenue prediction. Hence, for churn prediction, different researchers have used different methods in distinct sectors which primarily depend on customer participation. It is especially difficult in customer churn to prevent and predict in a world where business models, demands of customers, and services are constantly changing. In such a structure, providers get to know the actual value of sustaining customers in the workplace. Customer churn is a challenging and critical issue for many sectors, with the acquisition cost of customers increasing. Thus, it is a mandatory and absolute necessity for service providers to prevent the churn phenomenon in order to attain the availability of service. Every-year companies are losing up to 30% of customers because of churn and obtaining new customers are 5-10 times costlier than retaining the existing customers. The paper chooses distinct approaches to Artificial Neural Network (ANN). It will create a strategic plan which is practiced on the customer’s analytics according to a particular sequence to classify them into distinctive categories. The four main approaches can be customized for churn prediction. Based on the ability and potential of a customer, customers will be categorized. The classification will be based on different research studies with their unique methodologies and dimensions. The outcome will show the final classification of churn according to ANN. In Custom Analytics (CA), the two proposed dimensions namely, customer retention and customer identification will sort out the identity of customers in four major categorical approaches. They are ANN, Ensemble Approach, Growing Self-Organizing Map (GSOM), and Self-Organizing Map (SOM). Hence, this effective strategy for customer retention would help industries make better informed decisions.
Assessment of the quality of composts in selected commercial compost facilities in Sri Lanka
(Faculty of Science, University of Kelaniya, Sri Lanka, 2020) Udayanthika, S.K.I.; Sewwandi, B.G.N.
Organic food production is one of the fast-growing sectors in Sri Lanka with an understanding of health and environmental impacts caused by the use of chemical fertilizers. Hence, there is a high demand for compost by organic agriculture and it is widely used in home gardening as a substitute for chemical fertilizers. The compost produced by using different raw materials are available in markets at present in Sri Lanka. Even though the quality of the compost has to be in accordance with the Sri Lanka standard specification (SLS 1246:2003) for compost from municipal solid waste and agricultural waste, enough studies have not been carried out on the assessment of compost quality. Therefore, the objectives of this study were to assess the quality of compost collected from the commercial compost producing sites and to investigate the time period that can maintain the SLS standards specifications in compost once it is packed. The compost bags of 1 kg were collected within a day from 4 sites (namely A, B, C and D) in Colombo and Gampaha Districts, which have been packed on the same day and stored in the laboratory till analysis. The composts collected have been made from MSW (sites A, B and C) and garden waste (D). The compost quality parameters were measured in monthly intervals for a period of 6 months with three replicates. Data analysis was done by One-way ANOVA. Data analysis revealed that organic matter content and total carbon content in compost of A were lower than the standard limits during the study period. According to the results, the color of the compost in all composting plants was brownish black and complied with the SLS 1246:2003 standards. The compost did not emit irritable odors in any of the compost samples. Phosphorous content of compost from A and C was lower than the standard limit (0.5%) and Potassium content were lower than the standard minimum limit (1%) in C and D. Nitrogen content in compost from all the sites was lower than the standard minimum limit (1%). Composts from A, B, C and D indicated moderate phytotoxicity according to the germination index % values of 74, 67, 74 and 76, respectively which may be associated with the immaturity of the composts. Sand content was higher than the standard limit (10%) in compost plant D. It was found that most of the composting facilities could maintain the Phosphorous and Potassium within the SLS standard limits. However, Nitrogen content in compost from all the sites were lower than the standard limits and it was decreasing with time during the study period. The results showed that the compost produced in selected sites does not meet the standard specifications given by the Sri Lanka Standards (SLS 1246:2003) and therefore, quality control and continuous monitoring are essential to maintain the quality of compost available in the market. The quality control of compost has to be started from the selection of raw materials up to the detection of maturity of compost in order to produce a good quality product.
Availability and reliability analysis in 5G communication scenarios in IoT applications
(Faculty of Science, University of Kelaniya, Sri Lanka, 2020) Vithanage, D.S.
Due to the increasing usage of wireless communication devices, the demand for fifth-generation (5G) cellular network access is growing rapidly. Facilities that might be seen with 5G technology include, far better levels of connectivity and coverage. 5G cellular networks provide dynamic coverage with respect to time and enduring overlapped cell areas. Due to this reason, 5G network users can be covered by numerous cells and Radio Access Technology (RATs). This could be done, especially by taking full advantage of network capability to facilitate extreme performance that includes supporting hugely inter-tethered devices attributed to IoT applications in 5G. The main challenge in IoT applications is that scalable and efficient connectivity for a massive number of devices sending very short packets, is not done adequately. In such scenarios, IoT devices are expected to select the most appropriate cell based on the channel availability information of each cell. Therefore, efficient cell selection is needed in 5G. Additionally, in a heterogeneous network with overlapping cells, cell selection could be a critical decision for 5G users. The proposed research is aimed at implementing two schemes for cell selection based on the availability and reliability performance in 5G. The study proposes an algorithm by considering two schemes. The first scheme is contingent on the distance. That is the distance to the base-stations must be considered. If the base stations are close to devices, signal strength is high. The second scheme is based upon the channel availability and the distance while priority goes to the channel availability of each cell. These two schemes were simulated by using a simulation program, which was developed in MATLAB. For cell selection, scheme 2 is much fairer than scheme 1 because by using scheme 2, channels availability is balanced through cells. Despite this, the nearest device is allocated to the nearest place by scheme 1 and as a result of that, signal strength is higher in those devices. By considering all the results obtained, it can be concluded that the proposed schemes are efficient cell selection schemes, which can be used to improve the overall system performance.
Beneficial functions of plant materials used in shodhana process of mercury in Ayurveda Rasashastra
(Faculty of Science, University of Kelaniya, Sri Lanka, 2020) Gunaratna, T.A.N.R.; De Silva, W.R.M.; Prajapati, P.K.; De Silva, K.M.N.
Shodhana process in rasashastra is a mandatory process for each material prior to use in herbomineral pharmaceutical preparations. Although different types of mercury (Hg) shodhana”processes are described in rasashastra, in Sri Lanka, Ayurveda herbo-mineral manufacturers use a three-step method with Allium sativum extract, Piper betel extract and the decoction using Terminalia chebula, T. bellerica and Phyllanthus emblica. Although this method is well-known within the Ayurveda community, there are no research evidences available to identify the support and the functions given by the plant materials in the mercury shodhana process. Therefore, this research was carried out to analyse the elemental changes that would occur to commercially available mercury during the shodhana process. Shodhana process was carried out as mentioned in the Rasa Jala Nidhi textbook (volume I) of rasashastra literature under the mercury section (eighth process). As the shodhana process involves three steps, there were four samples to be analysed namely, crude mercury, first step completed Hg, second step completed Hg and final step completed Hg. Samples were microwave digested using HNO3: HCl in 3:1 ratio and diluted prior to the Inductive Coupled Plasma Mass Spectroscopy (ICPMS) analysis. These four Hg samples were then, subjected to ICPMS analysis. Standard 2A was performed to check Ag, Al, As, Ba, Be, Cd, Co, Cr, Cs, Cu, Fe, Ga, K, Li, Mg, Mn, Ni, Pb, Rb, Se, Sr, Tl, U, V, Zn elements and standard 2A Hg was performed to check Hg element. Cu, As and V were measured in He gas mode and rest were measured in no gas mode. The analysis was carried out in triplicate. Origin and R software were used for the comparison. According to the results obtained, Mg, Al, Fe, Co, Zn, Cd, Ba and Pb were present in the crude mercury as noticeable elements, but the element levels were changed with each shodhana step. Most importantly, it clearly shows the reduction of Pb level from 2347.25 ± 0.01 ppb to 173.20 ± 0.02 ppb. Furthermore, trace elements such as Li, Ni, Ga and U were completely removed from mercury after the completion of shodhana process. The reason for the reduction of metal ions can be attributed to metal iron chelation, detoxification procedures with plant bioactive compounds such as organic sulphides, polyphenols and flavonoids. Therefore, these results reflect the benefit of shodhana process and clearly explains the use of plant extracts as a removal agent of unwanted metal ions, which are trapped in Hg.
Calibration of the rolling angle of a Quadrant Photo Detector mounted in the image plane of a dark-field passive LIDAR system
(Faculty of Science, University of Kelaniya, Sri Lanka, 2020) Abeywickrama, S.S.; Perera, H.E.; Jayaweera, H.H.E.
Passive Light Detection and Ranging (LIDAR) has successfully been used for observing insects and their activities. It was reported that such techniques are more efficient compared to traditional approaches. Quadrant Photo Detectors (QPDs) are widely used at the image plane with the use of a modified eye-piece to detect both wing-beats and heading angles of insects. In these systems, knowing the exact orientation of the QPD in the image plane is an imperative task. This study was carried out to propose a method to calibrate the rolling angle of a QPD mounted in the image plane of a Newtonian telescope in a dark-field passive LIDAR system using a Hamamatsu S4349 Silicon QPD. Each segment of the QPD was connected to a data acquisition card through four Trans-impedance amplifiers and programmable gain amplifiers. A white coloured inverted pendulum oscillated across the Field of View (FOV) of the QPD at a known distance was used for calibration. Intensities registered at the individual segments of the QPD were recorded at a rate of 10 kHz while the pendulum swept the FOV. Thirty-six of such measurements were obtained by changing the rolling angles by 10-degree at a time. The four filtered and normalized signals were used to calculate the activation times (full width at 10%) and four unique sinusoidal functions were fitted to the whole range of angles. These coefficients can be used to estimate the rolling angle of the QPD using a test oscillation. It was found that the accuracy of the estimate was ± 6.04 degrees. A ray tracing-based simulation was conducted to simulate this activity and findings from the activity agrees with the theoretical simulation results. It was noted that the highest performance can be obtained when the pendulum oscillates in a plane normal to the optical axis.
Case study of credit risk analysis and creditworthiness prediction at loan approval
(Faculty of Science, University of Kelaniya, Sri Lanka, 2020) Weralupitiya, B. N.; Jayatillake, R. V.
The credit risk is considered as the risk associated with a borrower’s failure to pay the loan or interest amount on time. An increase in Non-Performing Loan (NPL) ratio directly affects the financial performance of the banks as well as the economy of the country as a whole. Therefore, this case study was carried out for a specific bank in Sri Lanka with the objectives of developing a predictive model to assess the creditworthiness of potential loan applicants at the approval and to identify factors associated with time to first default. The data used for this study consisted of bank loan details of 10,626 existing customers in their current loan portfolio and their repayment behavior over 2.5 years. It consisted of 11 continuous and 7 categorical variables including customer’s demographic details, personal financial details and bank-specific ratios. Furthermore, it included 10 transaction variables which all are categorical. The univariate tests such as Mann Whitney Test and Chi-Square Independence Test and graphical analyses identified that apart from variables “Age at Approval” and “CRIB Status at the approval”, all the other variables showed a significant relationship with the variable of interest, “Loan Status”. As only 33% of the respondents were non-performers, the Synthetic Minority Oversampling Technique (SMOTE) was used to handle the class imbalance. Several machine learning techniques such as Logistic Regression, Random Forest, Support Vector Machine, and Artificial Neural Network were applied with and without SMOTE Sampling to achieve the optimal model by comparing the ROCAUC value of each model. The Artificial Neural Network model applied with SMOTE sampling was found to be the best model with a ROC-AUC value of 91.6%. Furthermore, the study data consisted of the borrowers’ default status in every quarter. Therefore, a discrete-time hazard survival model was developed to identify the predictors that affect most of the risk of first default. It was found that the risk of first default to be higher in early quarters and decreases over time with the best fitted discrete survival model. Particularly, “Security type”, “Loan to value ratio”, “Tenure”, “Purpose of the loan” and “Interest rate” were some of the variables found as the most significant variables that associate with the risk of first default.
Catalytically induced pyrolysis of LDPE to liquid fuel
(Faculty of Science, University of Kelaniya, Sri Lanka, 2020) Wickramaarachchi, W.A.A.S.; Premachandra, B.A.J.K.; De Silva, D.S.M.
Plastics are used in a wide range of applications because of their durability, lightweight, easy fabrication, and desired chemical and physical properties. Usually, plastic products are discarded after use to the environment as solid waste. Therefore, the low degradability of plastics and the high demand for plastic products have created a serious environmental issue. Recycling is one of the methods used in plastic waste management. As a recycling method, energy recycling or producing fuel oil from plastic waste has gained a promising interest. In this study, it was expected to convert selected used plastics to fuel oils through a pyrolysis process using a catalyst. A laboratory-scale pyrolysis system was developed and a low-cost conversation process for plastics to fuel oil was investigated in an environmentally friendly manner. Initially, virgin low-density polyethylene (LDPE) was used in this conversion as the control sample. Then waste wrapping materials made of LDPE were subjected to pyrolysis. A two-neck round bottom flask was used as the reactor while the heat was supplied by a LP gas burner. To control overheating and possible heat losses, the reactor was dipped in a soil bath during heating. A constant heating rate and a constant inert gas flow rate to the reactor were maintained throughout the experiment. The gases evolved by the pyrolysis were condensed. The distillate was collected while the uncondensed fraction was trapped first in a non-polar organic solvent and further in a basic aqueous solution to prevent possible hazardous emissions. A locally abundant mineral was tested as a possible catalyst for the pyrolysis to improve the quality of the resulting products. It was observed that the purity of the resulting fuel oil had been improved with the use of the catalyst. The resultant liquid fraction was conveyed for factional distillation and the fractions were characterized with GC-MS and FTIR techniques. According to the GC-MS analysis, the major constituents in the fraction obtained from virgin LDPE through uncatalyzed pyrolysis were decane, undecane and 1-tetradecene. The major constituents obtained through the catalytic pyrolysis of virgin LDPE were cyclopropane, 1- decene, undecane and pentadecane. The pyrolysis of waste LDPE resulted in cyclopentane, decane, undecane and 1-pentadecene as fractional distillates. The mineral tested as the catalyst has given significant improvement in the purity of the oil fractions produced. The combustion characteristics and viscosities of the resultant oils are to be determined and those will be compared with the commercially available fuel oils. The study will be extended for other plastic waste types including mixed waste.
Chemical profile of Terminalia chebula fruit collected from different regions of Sri Lanka and commercial samples from Sri Lanka and India
(Faculty of Science, University of Kelaniya, Sri Lanka, 2020) Suraweera, K.P.C.D.; Wickramarachchi, S.R.; Tennakoon, T.M.S.G.
Terminalia chebula is a valuable medicinal plant used in traditional medicine. The fruit of T. chebula contains a large number of biologically active chemical compounds. Demand for the herbal drugs is increasing every day and maintaining the quality of herbal drugs is very important. Therefore, the objective of this study is qualitative and quantitative comparison of the chemical profiles of T. chebula fruit (without seeds) of commercial samples and authentic samples. Authentic samples of T. chebula were analyzed to see the effect of climatic zone variation on chemical profile and physicochemical parameters. Authentic samples (SLA) were collected from T. chebula plant itself from onsite visit, from Bibila, Buththala, Padhiyathalawa, Gampaha and Colombo and authenticated from the voucher specimen available at Herbarium, Link Natural Products (Pvt) Ltd (LNP). One composite sample was made according to sampling protocol, WHO 1998, from each region for analysis. Commercial samples are a mixture of fruits obtained from several suppliers from different areas. Commercial samples were obtained separately from three different batches of T. chebula commercial stocks from Sri Lanka (SLC) and India (INC) at LNP. Powdered dried fruits were extracted with 70% aqueous methanol and concentrated using rotatory evaporator. Physicochemical parameters were determined according to WHO and European pharmacopoeia methods. Total tannin was determined using Folin-Denis assay. Crude T. chebula fruit extract was separated by thin layer chromatography (TLC) and high performance liquid chromatography (HPLC). Gallic acid content and gallic acid/ellagic acid ratio were calculated using the peak area of HPLC chromatograms. TLC and HPLC showed a large number of compounds in the crude extract of T. chebula fruit. Both showed similar profiles with variations in intensity among the samples. Physicochemical parameters (without water soluble extractive values), total tannin, gallic acid content, gallic acid/ellagic acid ratio are statistically different among SLA samples (P < 0.05). Except physicochemical parameters, other parameters are statistically similar among SLC and INC (P<0.05). Total tannin content (% w/w) of SLA samples was; Padiyathalawa: 33.40 ± 0.17, Buththala: 43.39 ± 0.41, Gampaha: 41.13 ± 0.61, Bibila: 42.31 ± 0.23 and Colombo: 34.12 ± 0.01. Gallic acid content (% w/w) of SLA samples was; Padiyathalawa: 0.49 ± 0.01, Buththala: 0.98 ± 0.01, Gampaha: 1.03 ± 0.02, Bibila: 0.83 ± 0.02 and Colombo: 1.86 ± 0.04. Gallic acid/ellagic acid ratio (% w/w) of SLA samples; Padiyathalawa: 0.15 ± 0.0038, Buththala: 0.18 ± 0.0009, Gampaha: 0.16 ± 0.0003, Bibila: 0.16 ± 0.0041 and Colombo: 0.68 ± 0.0040. Total tannin content, gallic acid content and gallic acid/ellagic acid ratio vary in different batches of commercial samples. Mean of total tannin content (% w/w) of SLC was 49.14 ± 6.09 and INC is 42.79 ± 0.76. Mean of gallic acid content (% w/w) of SLC was 1.13 ± 0.28 and INC is 2.25 ± 0.69. Gallic acid/ellagic acid ratio (% w/w) of SLC was 0.30 ± 0.07 and INC is 0.43 ± 0.05. Chemical composition and quality of T. chebula. dried fruit depend on the geographical location, maturity stage, growth condition and raw material processing condition.
Classifying risk and vulnerability in the supply chain during an epidemic outbreak
(Faculty of Science, University of Kelaniya, Sri Lanka, 2020) Perera, M.A.S.M.; Wijayanayake, A.; Peter, S.
Companies always try to maximize shareholders' value by reducing the cost and maximizing profits in the long terms. However, one of the primary difficulties they face in doing so, is because of disruptions in the supply chain (SC). The supply chain can be disrupted due to natural disasters, manmade catastrophes, strikes, legal disputes, and special cases like epidemic outbreaks. The study explores what causes the supply chain to be disrupted in a company during an epidemic outbreak. It focuses on the Sri Lankan apparel industry as it contributes 6% to Sri Lanka’s GDP and 44% percent to Sri Lanka’s National Export Revenue, which is a significant proportion of the country’s economy. The primary objective of this study is to identify the supply chain risks in order to be prepared, mitigate the effects and ensure business continuity. The study proposes a model to identify the SC risks and vulnerabilities during an epidemic outbreak, and which risks should be prioritized. The model was primarily developed through a systematic review of literature and information collected from experts in the apparel sector was used to validate the findings. Leading apparel manufacturing companies in Sri Lanka were selected through convenience sampling and managers with more than five years’ experience were selected through random sampling. Using the output, the identified risks are then analysed and mapped in a vulnerability matrix considering cost and time factors. The model was tested and validated using 80%-20% rule. 80% of the collected data was used to develop the model and 20% of the collected data was used for testing and validation. Moreover, experts’ opinions were also used to validate the vulnerability matrix. Loss of local key supplier, loss of international key supplier, local port closure, international port closure, transportation link disruption (other than ports), raw materials delays and shortages, human resource shortages, product demand variations, order cancellations and lead time variations are SC risks which are considered for this study. The loss of international key suppliers and order cancellations were classified as high risks, whereas, human resource shortages were classified as the least risk. Though, a generalized vulnerability model is developed in this study considering cost and time factors, it can be customized using different factors and risks depending on the experience and needs of the company. Participants for the survey assumed that customers are international, and suppliers are both local and international. The study can be further developed to identify the SC strategies which should be taken to mitigate the SC disruptions during an epidemic outbreak or during a major global crisis.
CNN based deep learning model for tomato crop disease detection
(Faculty of Science, University of Kelaniya, Sri Lanka, 2020) Gunarathna, M.M.; Rathnayaka, R.M.K.T.
Tomato is one of the most commonly cultivated solanaceous short duration vegetable crops, grown in outdoor and indoor conditions, worldwide. However, unfortunately, many diseases affect these crops which have an impact on quality and the quantity of the produce, agricultural productivity, and causes considerable economic losses to the producers and to the contribution to the growth of the agricultural sector. Therefore, continuous monitoring of the crop is required throughout the growing stage to identify the diseases. The most traditional way of identifying diseases is naked eye observation, which is tedious and time-consuming. Today, advances in computer vision paved by deep learning have led to a situation where disease diagnosis is based on automated recognition. The main objective of this study is to develop an accurate tomato disease classification model which eliminates human error when identifying diseases. Due to a variety of similar disease and pathological problems, even experienced agronomists and plant pathologists often fail to recognize the correct disease. This computer vision system will assist agronomists in detecting a variety of tomato crop diseases. The proposed algorithm consists of four main steps; data collection, data pre-processing, CNN model creation, and evaluation of performance metrics. A leaf is a good indicator of the tomato’s health. Therefore, tomato leaf images belonging to 10 different classes with a resolution of 256x256 were collected from the Internet to build, validate, and test the model. Collected images were normalized and image augmentation techniques were applied to increase the size of the training data set in the preprocessing phase. The CNN model of the study was built from scratch using the Keras library, which runs top of the Tensorflow backend. The model comprises four convolutional blocks followed by batch normalization, max pooling, and dropout layers. Two dense and flatten layers were also included at the end. A time-based learning rate scheduler was used with an initial learning rate of 0.001, momentum of 0.5, an epoch of 15 and a batch size of 27. The final model was able to achieve a training accuracy of 94% and a testing accuracy of 92%. This proposed system would encourage tomato cultivators to detect diseases at an early stage and start treatments without relying on experts. In the future, we hope to build an ensemble approach to classify plant diseases with real-time images towards the development of a decision support system.
Comparative study of novel virgin coconut oil-based mayonnaise with commercial mayonnaise for physico-chemical and sensory parameters
(Faculty of Science, University of Kelaniya, Sri Lanka, 2020) Dilrukshi, D.M.N.; Lakshman, P.L.N.; Manoj, W.G.J.
Mayonnaise is a world-famous dressing that involves higher oil concentrations in the production process. Virgin coconut oil (VCO) as an emerging oil in the food industry is an ideal ingredient to be used in the production of mayonnaise. In this research, VCO was used as a functional ingredient in mayonnaise preparation and a comparative study was conducted between VCO mayonnaise and commercial mayonnaise. Samples of mayonnaise were prepared using trial and error method according to different oil: egg yolk ratios, and the best samples were selected. Chill thaw stable samples out of the best samples were selected by centrifugation, and the best sample was selected through a sensory evaluation using 30 untrained panelists. A five-point hedonic scale was used to access the parameters. The physico-chemical parameters including color, density, acid value, saponification value, iodine value of the best VCO sample and commercial mayonnaise were determined using standard methods and the two samples were compared for their sensory parameters (color, texture, flavor, taste, spreadability, overall acceptability). Results demonstrated significantly a lower acid value, iodine value and higher saponification values for VCO mayonnaise and physical parameters: color and density of the VCO mayonnaise were significantly different (p < 0.05) than the commercial mayonnaise samples. The sensory evaluation resulted a higher mean score of 4.97 for the overall acceptability of VCO mayonnaise while commercial mayonnaise scored 2.75. Studied sensory parameters including color, appearance, taste, flavor and texture of VCO mayonnaise scored higher mean values than commercial mayonnaise though spreadability parameter of commercial mayonnaise scored a higher value of 5.00 and VCO mayonnaise scored a lower value of 4.25. Therefore, the results of the study indicate that, mayonnaise with VCO is considered better, on its quality and organoleptic parameters compared to commercial mayonnaise and some organoleptic modifications are required to achieve better sensory quality.
Comparison of rainwater quality of three areas located in the vicinity of an oil refinery and thermal power plant in Sri Lanka
(Faculty of Science, University of Kelaniya, Sri Lanka, 2020) Hirushan, H.H.; Deeyamulla, M.P.
The chemical composition of rainwater, a form of wet deposition, differs over time due to a broad range of physical, chemical, and biological factors. The purpose of this analysis was to establish and compare the key ionic composition and water quality parameters of bulk deposition samples considering rainfall patterns, rainfall rates and pollutant sources. Three sampling sites were selected for the study in the Gampaha District in Sri Lanka which were separated by 7 km from each other. The first site was in the Makola South (MS) which represented an area in the vicinity of an oil refinery and thermal power plants; the second and third sites were in the University of Kelaniya (UOK) and Orugodawatta (OW) respectively, representing urban environments. Bulk depositions were collected after the container was almost filled avoiding any overflow. The chemical analyses of anions (F- , Cl- , NO3 - , SO4 2- ) in bulk depositions were carried out using the Dionex ICS-900 ion chromatography system and metals (V, Cr, Mn, Fe, Ni, Cu, Zn, As, Cd, Pb) were analyzed using the ICP-MS 7800-Agilent system. The average pH in MS, UOK and OW sites was 6.70, 7.15 and 7.31 respectively, and it was almost neutral due to atmospheric neutralization. The average conductivity values of MS, UOK and OW sites were 40.96 µScm-1 , 35.63 µScm-1 and 38.93 µScm-1 , respectively. The average values of other water quality parameters (salinity, TDS) were higher in the MS site than the other sites. The dominant metals were Na, Cr, Fe, Cu, As, and SO4 2- was the dominant anion in the MS site than the other two sites showing the pollution may be due to the influence of oil refinery and the thermal power plants situated near the MS site. The results indicated that the metal concentrations, anion concentrations and the water quality parameters from the rainwater collected among the MS, UOK and OW sites, the MS site has higher concentrations and higher pollution due to its location being in the vicinity of the oil refinery and thermal power plant. According to the results obtained it can be stated that rainwater analysis can be used as an indirect method to determine ambient air quality.
Comparison of the properties of CZTS semiconductor films grown by sequential and single step electrodeposition techniques
(Faculty of Science, University of Kelaniya, Sri Lanka, 2020) Fernando, W.T.R.S.; Jayathilaka, K.M.D.C.; Wijesundera, R.P.; Siripala, W.
Cu2ZnSnS4 (CZTS) is a promising semiconductor material suitable for application in low-cost and environmentally friendly thin film solar cells due to its superior optoelectronics properties. It is a perfect absorber material due to its high absorption coefficient (>10-4 cm-1 ) and direct optical bandgap (1.4-1.5 eV). Among the CZTS preparation techniques, electrodeposition is an attractive technique because of its simplicity, low cost and easy process controlling capability. In this investigation, a comparative study on CZTS films grown by two different techniques, namely, sequential electrodeposition and single step electrodeposition, has been carried out. Electrodeposition of Cu, Sn and Zn stack layers followed by sulphurisation with H2S is one of CZTS growth techniques. In this study, growth parameters of sequentially electrodeposited CZTS were optimized to obtain best photoactive CZTS thin films. Electrodeposition parameters of Cu, Sn and Zn have been obtained using voltammograms. Cu thin film was electrodeposited on Mo substrate at –0.89 V vs Ag/AgCl in an electrochemical cell containing 0.4 M CuSO4, 3 M lactic acid and NaOH at pH 11. Deposition of Sn thin film on Mo/Cu electrodes was carried out at -1.2 V vs Ag/AgCl in an electrochemical cell containing 0.055 M, 2.25 M NaOH and 8 ml of sorbitol. Zn thin film was electrodeposited on Mo/Cu/Sn at -1.2 V vs Ag/AgCl in an electrochemical cell containing 0.2 M ZnSO4. In order to grow CZTS, Mo/Cu/Sn/Zn thin films were annealed at 550 oC for 60 min in H2S. In the single step electrodeposition, CZTS thin films on Mo substrate were potentiostatically electrodeposited at -1.05 V vs Ag/AgCl for 40 min in a three electrode electrochemical cell containing 0.02 M copper (II) sulfate pentahydrate (CuSO4·5H2O), 0.01 M zinc sulfate heptahydrate (ZnSO4·7H2O), 0.02 M tin sulfate (SnSO4) and 0.02 M sodium thiosulfate (Na2S2O3) at room temperature. 0.2 M tri-sodium citrate (C6H5Na3O7) was used as the complexing agent and tartaric acid (C4H6O6) was used as the pH control solution. The pH of the bath was maintained at 5. The Ag/AgCl and platinum electrodes were used as the reference and the counter electrodes respectively. Then samples prepared were annealed at 550 oC for 30 min in H2S. CZTS films grown by two techniques were characterized using X-ray diffraction, reflectance, dark and light I-V, spectral response and C-V measurements in a PEC containing 0.1 M sodium acetate. Reflectance measurements reveal that the band gap energy of the films is 1.45 eV and I-V and spectral response measurements reveal that CZTS thin films were photoactive and p-type. The results obtained revealed that high quality photoactive CZTS can be prepared using both techniques. However, I-V and spectral response characteristics revealed that photoactive properties of CZTS thin films prepared by single step electrodeposition technique are superior in comparison to sequentially electrodeposited thin films.
Computational investigation of anti-Alzheimer effects of Asiatic acid present in Centella asiatica (Gotukola) and its derivatives
(Faculty of Science, University of Kelaniya, Sri Lanka, 2020) Fernando, R.D.M.N.,; Dahanayake, J.N.
Centella asiatica (Gotukola) is a commonly used medicinal plant that has a wide range of beneficial effects such as antioxidant effects, anti-Alzheimer's disease effects, anti-inflammatory effects, anti-fertility effects, anti-tumor effect and antimicrobial effects. Asiatic acid, pentacyclic triterpenoid is one of the secondary metabolites present in Centella asiatica extract that has all the above pharmacological properties. In this study, the Anti- Alzheimer biological activities of Asiatic acid and its derivatives were mainly focused. Alzheimer’s disease (AD) is a neurodegenerative disease. It results in loss of cognitive activity and memory and creates impairments in signaling among brain cells. Main proteins involved in Alzheimer’s disease are Human amyloid precursor protein (1AAP), Acetylcholine esterase (4PDE), Tau protein (2MZ7), Alzheimer’s beta-A (1IYT) and Alzheimer’s beta-A fibrils (2BEG). In this study, twenty derivatives of Asiatic acid were considered to investigate the anti-Alzheimer activity and one of cholinesterase inhibitor; Donepezil which is commonly used as a clinical drug in Alzheimer was considered as a reference compound. Initially, energy minimized structures of Asiatic acid and its derivatives were obtained using molecular mechanical calculations. Docking studies were carried out for the reference compound, Asiatic acid and suggested derivatives with Alzheimer’s disease related proteins. They were docked using Autodock4.0 to obtain their interactions with target proteins and to determine the amino acid residues in binding pockets. The binding affinities of derivatives with proteins were compared with the binding affinity of parent molecule, Asiatic acid and also with the binding affinity of the reference compound, Donepezil respectively. According to the results, several Asiatic acid derivatives have a higher binding affinity with acetylcholine esterase enzyme and some derivatives showed the high affinity with other proteins. The reasons for their highest binding affinities and further details were obtained by using molecular dynamics (MD) simulations. The parent molecule and several derivatives that have the highest affinity with each protein were then further analyzed using MD simulations. MD simulations were carried out on protein-ligand complexes for 50 ns using CHARMM36 force field. The trajectories obtained from MD simulations were used to calculate the radius of gyration (Rg), root mean square deviation (RMSD), root mean square fluctuation (RMSF), solvent accessible surface areas (SASA), and hydrogen bonding (HB). According to the Rg and RMSD results, the studied protein-ligand complexes were stable throughout simulation time. A significant number of hydrogen bonds were observed between the derivates and protein residues. Further, RMSF and HB results of derivatives were compared with the results of Asiatic acid, in order to investigate the higher binding affinities of the derivatives. The MD analysis results along with docking results indicated that the Asiatic acid derivatives with higher binding affinities according to docking studies have the potential to act as promising anti-Alzheimer agents.
Computational investigation of anti-Alzheimer properties of novel Curcumin derivatives
(Faculty of Science, University of Kelaniya, Sri Lanka, 2020) Amarasinghe, A.K.D.K.K.,; Dahanayake, J.N.
Curcumin, a naturally- occurring principal curcuminoid of turmeric has been used as a remedy in many Asian countries for the past century. Curcumin has shown remarkable results for the effect of various medical conditions such as cancers, liver diseases, heart diseases, osteoarthritis, and also diabetes. In this study, we discuss the effect of Curcumin and its newly synthesized derivatives as potential anti-Alzheimer compounds. Alzheimer’s disease (AD) is a chronic, neurodegenerative disease that can cause dementia that could affect memory, thinking, and behavior. Due to its anti-oxidant, anti-inflammatory, and lipophilic action, Curcumin can improve the cognitive functions in Alzheimer’s disease patients. There are satisfactory proofs for the effect of Curcumin on Alzheimer’s disease such as decreased Beta-amyloid plaques (the main concern regarding AD) and delayed degradation of neurons. The main proteins that are associated with Alzheimer’s disease and highly focused on this article are Amyloid Precursor protein (1AAP), Alzheimer’s Beta – A (1IYT), Alzheimer's Beta A fibrils (2BEG), Acetylcholine esterase (4PQE), and Tau protein (2MZ7). In this computational investigation, energy-optimized structures of selected eight derivatives of Curcumin and parent compound, Curcumin were obtained using DFT calculations. To secure a better understanding of binding interactions of the above-mentioned proteins with our selected derivatives and parent compound as ligands, docking studies were performed. To check the validation of docking results, Donepezil, a clinical drug that is currently used for the AD was used as a reference molecule and docking studies were performed. Among the newly synthesized derivatives, which were suggested as potential anti-Alzheimer agents, two derivatives have shown promising results with higher binding affinities for each protein, according to docking studies. The derivatives that showed the highest binding affinities were selected along with the parent compound, Curcumin for each protein for Molecular Dynamics (MD) simulations. MD simulations were performed on protein-ligand complexes for 50 ns using CHARMM36 force field. The mean radius of gyration (Rg), root mean square deviation (RMSD), root mean square fluctuation (RMSF) and solvent accessible surface areas of the binding pockets were calculated and hydrogen bond analysis (HBA) was also performed. Rg and RMSD results indicated the stability of the protein-ligand complex throughout the simulation time. HBA results showed that ligand has significant number of hydrogen bonds with the ligand. RMSF and HBA results of derivatives were compared with the results of Curcumin, in order to explain the higher binding affinities of the derivatives. The MD analysis results along with docking results reveal that the two derivatives with higher binding affinities according to docking studies have the potential to act as promising anti-Alzheimer agents.