Extended Abstract Book International Conference on Computer ...

Extended Abstract Book International Conference on Computer, Communication,

Chemical, Materials and Electronic Engineering

IC4ME

2-2016

24 ~ 25 March, 2016

Organized by

Faculty of Engineering

University of Rajshahi, Rajshahi 6205, Bangladesh

Extended abstract book of International Conference on Computer, Communication, Chemical, Materials and Electronic Engineering, IC4ME2-2016.

Copyright © 2016, Faculty of Engineering, University of Rajshahi, Rajshahi 6205, Bangladesh

Contact Information:

Dean Faculty of Engineering University of Rajshahi, Rajshahi 6205 Bangladesh

Tel: +880 721 711125 Fax: +880 721 711131 e-mail: [email protected] Alt. e-mail: [email protected]

URL: www.ru.ac.bd/ic4me2-2016

mailto:[email protected]

http://www.ru.ac.bd/icmeie2015/[email protected]

Foreword On behalf of the IC4ME2-2016, I warmly invite you to the International Conference on Computer, Communication, Chemical, Materials and Electronic Engineering, IC4ME2- 2016, at the green premise of the University of Rajshahi during 24~25 March, 2016.

The IC4ME2-2016 is the second of its kind hosted by the Faculty of Engineering of University of Rajshahi. The IC4ME2-2016 continues the successful format from previous years ICMEIE-2015. While having the same overarching goal of presenting cutting-edge results, ideas, techniques, and theoretical advances in the mentioned theme, the IC4ME2-2016 is separately tasked by focusing on emerging topics that complement the areas covered by the main technical program.

The objective of organizing this conference is to bring together leading scientists, researchers and scholars to exchange and share their experiences and research results about all aspects of Electrical, Electronics, Communication, Chemical Engineering and IT, and discusses the practical challenges encountered and the solutions adopted.

The international character of this meeting is illustrated by the participation of researchers from Fiji, Germany, India, Japan, Malaysia, Nepal, and Sri Lanka. Among the huge number of submissions only 56% have been accepted for publication. The conference includes 6 keynote speeches, 2 invited papers and 89 contributed papers distributed in 1 plenary session and 15 oral sessions.

The information presented herein should help open up new avenues for research and provide researchers of the mentioned themes with new ideas to help them improve their production efficiency. We thank the reviewers from various countries who did this thankless job even in their busy schedule.

The editorial team of this book deserves special thanks for their outstanding efforts in reviewing and preparing the abstracts for publication. Sponsorship for this meeting is an important feature of its success. On behalf of the organizing committee of IC4ME2-2016, we thank the University Grants Commission of Bangladesh for their support to promote the meeting.

Finally, we would like to thank the presenters for their willingness to share their latest research and ideas. Without their efforts, this conference would not be possible. We hope you will enjoy your staying at very beautiful campus of the University of Rajshahi.

Abu Bakar Md. Ismail, PhD Chair International Conference on Computer, Communication, Chemical, Materials and Electronic Engineering, IC4ME2-2016 Faculty of Engineering, University of Rajshahi Rajshahi 6205, Bangladesh

Contents

Program Schedule 10

Organizing Committee 13

International Advisory Committee 13

Technical Program Committee 15

Reception Committee 17

Budget, Finance and Sponsor Committee 17

Registration Committee 18

Publication Committee 18

Decoration and Discipline Committee 19

Accommodation Committee 19

Social Event and Tour Committee 20

Food and Refreshment Committee 20

Paper ID Title

Keynote Simulation of Normal Incidence Sound Absorption Coefficients of Perforated Panels with/without Glass Wool by Transmission Line Parameters in a two-port Network

21

Takayoshi Nakai Keynote Settlement of Crystalline Structure of Group II-VI Semiconductor Nanoparticles by Profile

Refinement Technique and Size Determination by Tight Binding Model 21

Surendra K. Gautam Keynote Readerless RFID Transponder: A Concept 22

Mamun Bin Ibne Reaz Keynote Science, Technology and Innovation: For Engineering Our Future 22

Aravind Chinchure Keynote Progress of Low-Temperature Fabrication Technologies of Thin Film Transistors for

Preservation of Global Environment 23

Susumu Horita Keynote Factors that Facilitate and Impede Hitherto Untried R&D in Engineering 24

Sisil Kumarawadu Invited Water-in-oil Microemulsions as Nanoreactors to Prepare Nanoparticles with Tunable

Electrical, Optical, and Antibacterial Properties 25

Md. Abu Bin Hasan Susan, M. Muhibur Rahman and M. Yousuf A Mollah Invited Physics and Technology in Radiation Oncology and Imaging 25

Golam Abu Zakaria 3 Epitaxial Growth of MnAl Films with CrRu Buffer layer 27 M. A. I. Nahid, Mikihiko Oogane, Hiroshi Naganuma, Miho Watanabe and Yasuo Ando

5 Fabrication of Cardiac Biomarker Immunosensor based on Three Different Electrode Surfaces and Contrasting Their Efficiencies

28

Payal Gulati, Prabhash Mishra and Safiul Islam 9 Simulation of the Electrical Activity of Cardiac Tissue by Finite Element Method 29 Tanzina Rahman and Md. Rezaul Islam

10 Baking of Ilmenite on Moistening with Sulfuric Acid followed by Leaching with Dilute Sulfuric Acid Solution

30

Ranjit K. Biswas, Aneek K.Karmakar, Jinnatul Ara and Muhammad A. Gafur

Paper ID Title

11 Thermal Treatment of Ilmenite on Moistening with Concentrated HF followed by Leaching with Dilute Sulfuric Acid Solution

31

Ranjit K. Biswas, Mohammad A. Habib, Aneek K. Karmakar and Mohammad J. Alam 12 Solvent Extraction of V(V) from Nitrate Medium by Tri-n-Octylamine Dissolved in

Kerosene 32

Ranjit K. Biswas, Aneek K. Karmakar and Mottakin 13 Kinetics of Extraction of Ti(IV) from SO4

2- Medium by Cyanex 301 Dissolved in Kerosene 33 Ranjit K. Biswas and Aneek K. Karmakar

14 Temperature Dependence Magnetoresistance of Fe73.5 –Cu1 –Ta3 –Si13.5 –B9 Magnetic Alloy

34

M. Z. Ahsan, F. A. Khan and M. Perven 16 Production and Improvement of Waste Tire Pyrolysis Oil to be Utilized as Biofuel in Diesel

Engine 35

Md. Nurul Islam and Md. Rafsan Nahian 23 A Comparative Study of Electrical and Magnetic Properties of Co Thin Films Deposited on

GaAs (100), Si (100) and Glass Substrates 36

Nafisa Ahmed, A. A. M. Rayhan Kabir, Md. Nahid Akter Shafi and Md. J. Islam 24 Biological Evaluation of Radiotherapy Treatment Plan for Different Field Techniques in 3-

Dimensional Conformal Radiotherapy (3DCRT). 37

Kausar A, Azhari H A and Zakaria G A 25 Design of a Linearly Polarized Multi-Band Transmission Line Feed Microstrip Patch

Antenna for Wireless Communications 38

Sheikh Dobir Hossain, Md. Khalid Hossain and Rebeka Sultana 26 Design and Fabrication of an Unmanned Video Transmitting Tele-Bot using 3G GSM

Network 39

Md. Mamunoor Islam and Mehdi Hasan Chowdhury 29 Effect of Sintering Temperature on Nb+Nd Doped Bismuth Ferrite 40 Sadia Tasnim Mowri, M A Gafur, Quazi Delwar Hossain, Aninda Nafis Ahmed and

Muhammad Shahriar Bashar

30 Silicon Nanocrystals Rich Lanthanum Fluoride Films for Future Electronic Devices 41 Ferdous Rahman, Sk. Rashel Al Ahmed, Md. Golam Saklayen and Abu Bakar Md. Ismail

41 Study on the Displacement Effect at Cylindrical Ionization Chamber with different Radii in High Energy Photon of Flat Beam and True Beams

42

Kumaresh Chandra Paul, Guenther H Hartmann, Enamul Hoque and Golam Abu Zakaria 42 Electrical and Optical Properties of Cu-Nanoparticles- Doped α-Fe2O3 Thin Film Spin-

Coated on Glass Substrate 43

Sanjida Ferdous, Afroza Yasmin, Jinia Sultana and Abu Bakar Md. Ismail 43 Study on Morphological Properties of Cu-NPs Doped α-Fe2O3 Thin Film deposited on Glass

Substrate 44

Jinia Sultana, Afroza Yasmin, Sanjida Ferdous and Abu Bakar Md. Ismail 45 Study on the Content Based Image Retrieval (CBIR) System based on Histogram Technique 45 Rebeka Sultana, Shamim Ahmad and Md. Khalid Hossain

Paper ID Title

47 MRI Segmentation using Fuzzy C-Means Clustering and Bidimensional Empirical Mode Decomposition

46

Gulam Sarwar Chuwdhury, Md. Khaliluzzaman and Md. Rashed-Al-Mahfuz 48 Wear and Morphological Behavior of Electron Beam Dose Irradiated Polyoxymethylene

Copolymer (POM-C) 47

Md. Shahinur Rahman, Heon-Ju Lee, Muhammad Sifatul Alam Chowdhury and Konstantin

Lyakhov

49 Study of Structural and Optical Properties of Pyrolised CuO Films 48 M. Majhar, S. Ahmed, M. Mozibur Rahman and M. K. R. Khan

52 Algorithm for Performance Appraisal using CAW Method 49 M. Z. Ahsan and Md. Mamun-Ur-Rashid Khandker

53 Dynamic Power Management by Reinforcement Learning 50 Safayet Hossain, Muhammad Adnan Ibn-Ismail and Muhidul Islam Khan

56 Friction and Morphological Properties of Ion Implanted Polyoxymethylene Copolymer (POM-C)

51

Md. Shahinur Rahman, Md. Mehedi Hasan, Muhammad Sifatul Alam Chowdhury and

Konstantin Lyakhov

58 Analysis of Annual and Seasonal Precipitation Concentration Index (PCI) of North-Western Region of Bangladesh

52

Ahsan Habib Rasel, Md Monirul Islam and Mumnunul Keramat 60 Assessing Driving Stress using Custom Built Real-Time Sensors 53 Nova Ahmed, Saad Azmeen-Ur-Rahman, Rahat Jahangir Rony and Md. Tanvir Musfique

62 Bitwise Template Fusion of Noisy Images for Enhanced IRIS Recognition System 54 Sohel Ahammed and Biprodip Pal

65 Kinetics of Extraction of Ti(IV) from Sulfate Medium by Cyanex 302 55 Ranjit K. Biswas and Aneek K. Karmakar

66 Autonomous Human Face and Tracking System with Variant Poses, Blur and Illumination 56 Md. Zweel Rana, Monimul Islam and Mohiuddin Ahmad

67 Electrochemical Corrosion Characterization of Artificially Aged Al-6Si-0.5Mg (-1Cu) Alloys in Sodium Chloride Solution

57

Abul Hossain, M. A. Gafur, Fahmida Gulshan and A S W Kurny 69 Effects of Inclusions on the Mechanical Properties of Structural Steel Reinforced Bars 58 Abul Hossain, Fahmida Gulshan and A S W Kurny

72 Utilizing Solar Energy in the Filling Stations of Bangladesh: Technical and Economical Representation

59

Mohammad Jalal Uddin, Muhammad Sifatul Alam Chowdhury, Md. Ridwanul Karim, Md.

Arman Uddin and Md. Bakiuzzaman

74 Conversion of Prawn Shell Waste into Value Added Products for Textile Finishes 60 Md. Mofakkharul Islam, Firoz Ahmed and Md. Ibrahim H. Mondal

75 Textile Performance of Functionalized Cotton Fibre with Silane Coupling Agents 61 Md. Khademul Islam, Md. Abdul Aziz and Md. Ibrahim H. Mondal

76 Synthesis and Characterization of Carboxymethyl Cellulose with Higher Yield from Corn (Zea mays L.) Husk

62

Mst. Sarmina Yeasmin, Md. Abu Sayeed, Md. Abu Raihan Ali and

Md. Ibrahim H. Mondal

Paper ID Title

77 Synthesis and Characterization of Hydrogels from Cellulosic Materials for Green Adsorbent Products

63

Md. Obaidul Haque, Md. Abu Sayeed and Md. Ibrahim H. Mondal 83 Study the Encryption Techniques for Multimedia 64 Md. Martuza Ahamad and Md. Ibrahim Abdullah

84 Influence of Deposition Temperature on the Deposition of SiO2 Films from Reaction of Silicone Oil Vapor and Ozone Gas

65

Arifuzzaman Rajib, Susumu Horita, Atowar Rahman and Abu Bakar Md. Ismail 85 An Improved Representation of Audio Signal in Time-Frequency Plane 66 Kazi Mahmudul Hassan, Ekramul Hamid and Takayoshi Nakai

86 On the Optimization of Number of Message Copies for Multi-Copy Routing Protocols in Scalable Delay-Tolerant Networks

67

Md. Sharif Hossen and Muhammad Sajjadur Rahim 90 Emotional Bangla Speech Signals Classification using K-NN 68 Md. Tohidul Islam, Md. Ekramul Hamid and Somlal Das

92 Content Based Image Searching Using Multidimensional MSF 69 Saiful Islam, Ekramul Hamid and Emdadul Haque

93 Silicon Nanocrystals based Schottky Junction Solar Cell Fabrication and Characterization 70 A.T.M. Saiful Islam, Md. Enamul Karim, Arifuzzaman Rajib and Abu Bakar Md. Ismail

95 Fabrication and Characterization of α-Fe2O3 Homo-Junction Photocathode for Efficient Solar Water Splitting

71

Arifuzzaman Rajib, Atowar Rahman, A.T.M. Saiful Islam and Abu Bakar Md. Ismail 96 Human Recognition based on Adaptive Background Mixture Models and Improved

Histogram of Oriented Gradients 72

Shayhan Chowdhury and Kaushik Deb 97 A Practical Approach to Spectrum Analyzing Unit using RTL-SDR 73 Md. Habibur Rahman and Md. Mamunoor Islam

101 Fabrication of Bismuth Ferrite Multiferroic Perovskite Nanoparticles Using an Aqueous Organic Gel Route

74

Mayeesha Haque, M. S. Parvez, M. S. Islam, M.A. Hakim and M. A. Gafur 102 Purification, Characterization and Anti Proliferative Activity of A Lectin Isolated from

Mulberry Seeds 75

Mst. Jesmin Sultana, Syed Rashel Kabir and M.Taufiq Alam 105 Analysis of GLDAS data for Estimating and Distribution of Evapotranspiration and

Rainfall over Bangladesh 76

Md Ataur Rahman, Md Mainul Islam Mamun and Md Monirul Islam 113 Robustification of Logistic Classifer for Binary Classification in Microarray Gene

Expression Data 77

Md. Shahjaman, Md. Mushfiqur Rahman, Md. Manir Hossain Mollah, Anjuman Ara

Begum, S. M. Shahinul Islam and Md. Nurul Haque Mollah

115 Molecular Evolutionary Analysis of a-Defensin Peptides in Vertebrates 78 Arafat Rahman, M Sahidul Islam, Otun Saha and Titon Chandra Saha

117 Microencapsulated Probiotic Bacteria Protect the Spoilage of Functional Foods 79 Md. Symoom Hossain, Md. Abdul Alim Al-Bari, Mir Imam Ibne Wahed and

Md. Anwar Ul Islam

Paper ID Title

124 Preparation of Highly Cross-Linked Magnetic Polymer Composite Particles and Application in the Separation of Arsenic from Water

80

M. K. Sharafat, H. Ahmad, M. A. Alam and M.M. Rahman 125 Preparation of Hydrophobic Poly((lauryl methacrylate)-Coated Magnetic Nano-Composite

Particles and their Application as Adsorbents for Organic Polutants 81

Rukhsana Shabnam and Hasan Ahmad 126 Statistical Methods for Functional Analysis of Metagenomes 82

Zobaer Akond and Md. Nurul Haque Mollah 127 Simulation of Microalgae and CO2 Flow Dynamics in a Tubular Photobioreactor and

Consequent Effects on Microalgae Growth 83

Saumen Barua, Mohammad Morshed Alam and Ujjwal Kumar Deb 129 MLE based Robust ICIM for QTL Analysis with Backcross Population 84

Md Amanullah, Md. Mamun Monir, Md. Jahangir Alam, Md. Alim Hossen, Mita Khatun

and Md. Nurul Haque Mollah

137 Evaluation of PCA in spatial, frequency and wavelet domains for face recognition 85 Samsi Ara and M. Babul Islam

139 Time-Frequency Coherence Analysis of Multichannel Electroencephalography signalsusing Synchrosqueezing Transform

86

Md. Sujan Ali, Mst. Jannatul Ferdous, Md. Ekramul Hamid and Md. Khademul Islam

Molla

142 Experimental Study on Optical Characterization of Mono Crystalline Silicon Solar Cell 87 Nusrat Chowdhury, Md. Abdur Rafiq Akand and Zahid Hasan Mahmood

143 Canonical Correlation Analysis for SNP based Genome-Wide Association Studies 88 Atul Chandra Singha, Arafat Rahman, Md. Jahangir Alom and Md. Nurul Haque Mollah

148 Frequency Recognition of SSVEP for BCI Implementation using Canonical Correlation Analysis with Adaptive Reference Signals

89

Shalauddin Ahamad Shuza, Md. Rabiul Islam, Md. Kislu Noman and Md. Khademul Islam

Molla

151 Expert Reviewer Detection from Online Experiential Product Reviews 90 Atiquer Rahman Sarkar

152 Satellite Remote Sensing of Chlorophyll-a Concentration Variation over Bay-of-Bengal 91 Md. Monirul Islam and Md. Mainul Islam Mamun

153 FPGA based Pulse Oximeter using VHDL 92 Farhana Binte Sufi, Md. Fahmidur Rahman and Md. Maruful Islam

156 A Comparative Study of Optimization of Solid State Dye Sensitized Solar Cell Technologies for Power Grid Integration and Off-Grid Operation

93

Salman A. Chowdhury, Mobasseer M. Hossain and Md. Abu Sayem 157 Microbial Biosynthesis of Gold Nanoparticles using Aspergillus Foetidus and its

Cytotoxicity against A549 Cell 94

Swarup Roy and Tapan Kumar Das 158 Adsorptive Removal of Cr(VI) From Aqueous Solution onto Charred Vetiver Root 95

Bhoj Raj Poudel and Surendra K Gautam 159 Study on Determining the Average Size and Structure of Nanoparticles 96

Deval P Bhattarai and Surendra K Gautam

10

Program Schedule

Conference Kit Distribution

23 March, 2016 11:00 am – 02:00 pm

05:00 pm – 08:00 pm

Conference Room

Department of CSE

4th

Science Building

University of Rajshahi

24 March, 2016 08:00 am – 08:50 am Senate Bhaban


Inaugural Session 24 March, 2016 09:00 am – 9:45 am Senate Bhaban


Keynote Session 24 March, 2016 10:00 am – 11:30 am

11:45 am – 01:15 pm

Senate Bhaban


Invited Talk 24 March, 2016 03:00 pm – 04:00 pm Senate Bhaban


Technical Session 24 March, 2016 04:30 pm – 06:00 pm 4

th Science Building

25 March, 2016 09:00 am – 12:30 pm 4th

Science Building

Keynote Session: 24 March, 2016 10:00 am - 1:15 pm Room # Senate Bhaban

Chair: Prof. M. Abdus Sobhan, University of Rajshahi, Bangladesh

Title

Simulation of Normal Incidence Sound Absorption Coefficients of Perforated Panels with/without Glass Wool

by Transmission Line Parameters in a Two-Port Network

Takayoshi Nakai

Settlement of Crystalline Structure of Group II-VI Semiconductor Nanoparticles by Profile Refinement

Technique and Size Determination by Tight Binding Model

Surendra K. Gautam

Readerless RFID Transponder: A Concept

Mamun Bin Ibne Reaz

Science, Technology and Innovation: For Engineering Our Future

Aravind Chinchure

Progress of Low-Temperature Fabrication Technologies of Thin Film Transistors for Preservation of Global

Environment

Susumu Horita

Factors that Facilitate and Impede Hitherto Untried R&D in Engineering

Sisil Kumarawadu

11

Invited Talk: 24 March, 2016 03:00 pm - 4:00 pm Room # Senate Bhaban

Chair: Prof. Mamun Bin Ibne Riaz, Kebangsaan Universiti, Malaysia

Title

Water-in-oil Microemulsions as Nanoreactors to Prepare Nanoparticles with Tunable Electrical, Optical, and

Antibacterial Properties

Md. Abu Bin Hasan Susan, M. Muhibur Rahman and M. Yousuf A Mollah

Physics and Technology in Radiation Oncology and Imaging

Golam Abu Zakaria

Technical Session

Session Chair Date & Time Paper ID Venue

1A Prof. Khademul Islam Molla


24 March, 2016

04:30 pm – 6:00 pm

62, 85, 90, 92, 137,

148

Room # 219

4th

Science Building

1B Dr. Surendra Kumar Gautam

Tribhuvan University, Nepal 10, 11, 16, 65, 74, 75

Room # 223

4th

Science Building

1C Dr. Susumu Horita

JAIST, Japan 3, 14, 23, 29, 30, 42

Room # 122

4th

Science Building

1D Prof. Abdur Razzak

RUET

25, 26, 83, 97, 151,

153

Room # 120

4th

Science Building

2A Prof. Md. Rabiul Islam

RUET

25 March, 2016

09:00 am – 10:30 am

45, 47, 52, 53, 66, 96 Room # 120

4th

Science Building

2B Prof. C. M. Mostofa


12, 13, 67, 76, 77,

101

Room # 219

4th

Science Building

2C Dr. Riazul Islam

University of Dhaka 43, 48, 49, 56, 69, 84

Room # 223

4th

Science Building

2D Prof. Abu Reza

University of Rajshahi 113, 126, 129, 143

Room # 122

4th

Science Building

3A

Prof. Golam Abu Zakaria

Anhalt University of Applied

Sciences, Germany

25 March, 2016

11:00 am – 12:30 pm

5, 9, 24, 41, 60, 139 Room # 120

4th

Science Building

3B Prof. Ranjit K Biswas


102, 115, 117, 125,

127, 157

Room # 219

4th

Science Building

3C Prof. Md. Abu Bin Hasan Susan

University of Dhaka

93, 95, 124, 156,

158, 159

Room # 223

4th

Science Building

3D Prof. Mamnunul Keramat


58, 72, 86, 105, 142,

152

Room # 122

4th

Science Building

13

Organizing Committee

Chair : Abu Bakar Md. Ismail, Ph.D

Dean, Faculty of Engineering & Professor

Dept. of Applied Physics & Electronic Engineering

University of Rajshahi, Bangladesh

Co-Chair : Professor Dr. Ranjit Kumar Biswas

Dept. of Applied Chemistry & Chemical Engineering


Secretary : Professor Mamun Ur Rashid Khandker



Treasurer : Professor Dr. Dipankar Das

Dept. of Information & Communication Engineering


Members:

Chairman, Dept. of Applied Physics & Electronic Engineering, University of Rajshahi

Chairman, Dept. of Applied Chemistry & Chemical Engineering, University of Rajshahi

Chairman, Dept. of Computer Science & Engineering, University of Rajshahi

Chairman, Dept. of Information & Communication Engineering, University of Rajshahi

Chairman, Dept. of Materials Science & Electronic Engineering, University of Rajshahi

Professor Dr. M. Mozaffor Hossain,

Dept. of Applied Physics & Electronic Engineering, University of Rajshahi

Professor Dr. M. Abdus Sobhan


Professor Dr. Mumnunul Keramat


Professor Dr. M. Rostom Ali

Dept. of Applied Chemistry & Chemical Engineering, University of Rajshahi

Professor Dr. M. Mamunur Rashid Talukder


Professor Dr. M. Khademul Islam Molla

Dept. of Computer Science & Engineering, University of Rajshahi

Engr. S. M. Rezaul Kabir

Principal, BCMC College of Engineering & Technology, Jessore

International Advisory Committee

Professor Dr. Surendra K. Gautam

Tri-Chandra Campus

Tribhuvan University, Nepal

Professor Dr. Sisil Kumarawadu

Dept. of Electrical Engineering

University of Moratuwa, Sri Lanka

Dr. D.M.G. Preethichandra

Discipline Leader – Electrical Engineering

School of Engineering and Technology

Central Queensland University, Australia

14

Professor Dr. Keikichi Hirose

Dept. of Information and Communication Engineering

The University of Tokyo, Japan

Dr. Keiji Nagai

Associate Professor

Chemical Resource Laboratory

Tokyo Institute of Technology, Japan

Dr. Wu Ping

Associate Professor

Division of Engineering Product Development

Singapore University of Technology and Design, Singapore

Professor Dr. Md. Mamun Bin Ibne Reaz

Dept. of Electrical, Electronic and Systems Engineering

University Kebangsaan Malaysia, Malaysia

Professor Dr. Hartmut Baerwolff

Dept. of Analog & Optoelectronics

Cologne University of Applied Sciences, Germany

Professor Dr. Tomokazu Iyoda

Chemical Resource Laboratory

Tokyo Institute of Technology, Japan

Professor Dr. Norihiko Kamata

Dept. of Functional Materials Science

Graduate School of Science and Engineering

Saitama University, Japan

Professor Yousuke Nakashima

Plasma Research Center

University of Tsukuba, Japan

Dr. Susumu Horita

School of Materials Science

Japan Advanced Institute of Science & Technology, Japan

Professor A B M Shawkat Ali

Department of Computer Science and Information Technology

Dean, School of Science and Technology

The University of Fiji, Fiji

Dr. Golam Zakaria

Dept. of Medical Radiation Physics

University of Cologne, Germany

Anis Haque, Ph.D, P.Eng.

Senior Instructor

Fellow, Institute for Sustainable Energy, Environment and Economy (ISEEE)

Associate Director of Students

Dept. of Electrical and Computer Engineering

University of Calgary, Canada

Dr. Mohammad Abdul Fazal

Senior Lecturer

University of Malaya, Malaysia

15

Technical Program Committee

Co-Chair : Professor Dr. C. M. Mostofa

Dept. of Applied Chemistry & Chemical Engineering


Secretary : Professor Dr. Shamim Ahmad

Dept. of Computer Science & Engineering


Members:

Professor Dr. Md. Ariful Islam Nahid


Professor Dr. M. Shameem Ahsan


Professor Dr. M. Taufiq Alam


Professor Dr. Ekramul Hamid


Professor Dr. A. K. M. Akhter Hossain


Professor Dr. Md. Nurul Haque Mollah

Dept. of Statistics, University of Rajshahi

Professor Dr. Md. Ziaur Rahman Khan

Dept. of Electrical & Electronic Engineering, BUET

Professor Dr. Zahid Hasan Mahmood

Dept. of Electrical & Electronic Engineering, Dhaka University

Professor Dr. M. M. A. Hashem

Dept. of Computer Science & Engineering, KUET

Professor Dr. Mohammad Abdul Goffar Khan

Dept. of Electrical & Electronic Engineering, RUET

Professor Dr. Md. Rafiqul Islam

Dept. of Electrical & Electronic Engineering, KUET

Professor Dr. Shahid Uz Zaman

Dept. of Computer Science & Engineering, RUET

Professor Dr. S. M. Abdur Razzak

Dept. of Electrical & Electronic Engineering, RUET

Professor Dr. Md. Anisur Rahman

Dept. of Computer Science & Engineering, Khulna University

Professor Dr. Mohammad Shahidur Rahman

Dept. of Computer Science & Engineering, Shahjalal University of Science & Technology

Professor Dr. Mohammad Iqbal

Dept. of Industrial & Production Engineering, Shahjalal University of Science & Technology

Professor Dr. Md. Mahbubur Rahman

Dept. of Computer Science & Engineering, MIST

Professor Dr. Mortuza Ali

Dept. of Electrical & Electronic Engineering, Eastern University, Dhaka

16

Professor Rezaul Karim Mazumder

Dept. of Electronics & Telecommunication Engineering, ULAB, Dhaka

Professor Dr. Jugal Krishna Das

Dept. of Computer Science & Engineering, Jahangirnagar University

Professor Dr. Farid Ahmed

Dept. of Physics, Jahangirnagar University

Professor Dr. Md. Shahjahan

Dept. of Applied Physics, Electronics & Communication Engineering, BSMRSTU

Professor Dr. M. Mahbubur Rahman

Dept. of Information & Communication Engineering, Islamic University

Professor Dr. M. Maniruzzaman

Dept. Applied Chemistry & Chemical Engineering, Islamic University

Professor Dr. Momtazul Islam

Dept. Applied Physics, Electronics & Communication Engineering, Islamic University

Dr. Bilkis Ara Begum

Head, Chemistry Division, BAEC, Dhaka

Dr. Shamshad Begum Quraishi

Chief Scientific Officer, Chemistry Division, BAEC, Dhaka

Dr. Samia Tabassum

Senior Scientific Officer, BCSIR, Dhaka

Dr. M. Babul Islam

Associate Professor, Dept. of Applied Physics & Electronic Engg., University of Rajshahi

Dr. Md. Shafiqul Islam

Associate Professor, Dept. of Nuclear Engineering, Dhaka University

Dr. Md. Atowar Rahman

Associate Professor, Dept. of Applied Physics & Electronic Engg., University of Rajshahi

Dr. Md. Nur-Al-Safa Bhuiyan

Associate Professor, Dept. of Information & Communication Engg., University of Rajshahi

Dr. Sabbir Ahmed


Dr. Md. Emdadul Haque


Dr. Asadul Hoque

Associate Professor, Dept. of Materials Science & Engineering, University of Rajshahi

Dr. M. Anwarul Kabir Bhuiya


Dr. G. M. Shafiur Rahman


Mr. Subrata Pramanik

Associate Professor, Dept. of Computer Science & Engineering, University of Rajshahi

Mr. Muhammad Sajjadur Rahim


Dr. N. A. Ruhul Azad

Lecturer & Programme Leader for Engineering Foundation, LBIC, Brunel University, UK

17

Dr. Md. Iqbal Aziz Khan

Assistant Professor, Dept. of Computer Science & Engineering, University of Rajshahi

Mr. Mahboob Qaosar


Mr. A. F. M. Mahbubur Rahman


Mr. M. Rashed Al Mahfuz


Reception Committee

Convener : Professor Rostom Ali

Dept. of Applied Chemistry& Chemical Engineering


Members:

Chairman


Chairman


Chairman


Chairman

Dept. of Information & Communication Engineering, University of Rajshahi

Chairman

Dept. of Materials Science & Engineering, University of Rajshahi

Chairman

Dept. of Electrical & Electronic Engineering, University of Rajshahi

Professor M. Mozaffor Hossain


Professor Dr. M. Abdus Sobhan


Professor Dr. Mumnunul Keramat


Budget, Finance and Sponsor Committee

Convener : Professor Abu Bakar Md. Ismail

Dean, Faculty of Engineering, University of Rajshahi

Members:

Professor Md. Ekramul Hamid


Professor M. Ahsan Habib


Dr. G. M. Shafiur Rahman


18

Dr. Nur Al Safa Bhuyan


Registration Committee

Convener : Professor Mamun Ur Rashid Khandker

Dept. Applied Physics & Electronic Engineering, University of Rajshahi

Members:

Professor Shameem Ahsan


Dr. M. Babul Islam


Muhammad Sajjadur Rahim


Dr. Sabbir Ahmed


Dr. Anwarul Kabir Bhuiya


Dr. M. Iqbal Aziz Khan


Mahboob Qaosar


Publication Committee

Convener : Professor Shamim Ahmad


Members:

Dr. M. Babul Islam


Muhammad Sajjadur Rahim


Dr. Md. Emdadul Haque


Dr. Sabbir Ahmed


Dr. M. Asadul Hoque


Dr. Anik K Karmakar


Foez Ahmed


A.F.M. Mahbubur Rahman


19

Farhana Binte Sufi


Sangeeta Biswas


Decoration and Discipline Committee

Convener : Professor Md. Ekramul Hamid


Members:

Professor Rubaiyat Yasmin


Dr. M. Abdul Matin


Dr. Halida Homyara


Dr. Sinthia Shabnam Mou


Mahboob Qaosar


Sanjoy Kumar Chakravarty


Sangeeta Biswas


Shammi Farhana Islam


Saiful Islam


Mousumi Haque


Accommodation Committee

Convener : Dr. M. Babul Islam


Members:

Professor Shamim Ahmad


Professor Md. Ekramul Hamid


Mirza A. F. M. Rashidul Hasan


Dr. Md. Asadul Haque


Dr. Sabbir Ahmed


20

Md. Morshedul Arefin


Khairul Islam


Sajjadul Kabir


Social Event and Tour Committee

Convener : Professor Shaikh Enayet Ullah


Members:

Professor Syed Mustafizur Rahman


Dr. M. Atowar Rahman




Utpalananda Chowdhury


Sajjadul Kabir


Food and Refreshment Committee

Convener: Professor Md. Rezaul Islam

Dept. Applied Physics & Electronic Engineering, University of Rajshahi

Members:

Professor Dr. M. Mamunur Rashid Talukder


Professor Abu Zafor Muhammad Touhidul Islam


Dr. Aurangzib Md. Abdur Rahman



Dept. of Materials Science & Electronic Engineering, University of Rajshahi

Dr. M. Hasnat Kabir


Md. Rokonuzzaman


Abu Mohammad

Dept. of Materials Science & Electronic Engineering, University of Rajshahi

Dilip Kumar Sarker


21

Simulation of Normal Incidence Sound Absorption Coefficients of

Perforated Panels with/without Glass Wool by Transmission Line

Parameters in a two-port Network

Takayoshi Nakai

Professor

Department of Electrical and Electronic Engineering

Shizuoka University, Japan

ABSTRACT

This paper describes simulation of normal incidence sound absorption coefficients of

perforated panels by transmission line parameters in a two-port network. Maa and

Sakagami have investigated micro perforated panels, MPP. But their theories can treat

only near 1% perforation rates of perforated panels with back cavities. If sound

propagates as a plane wave, sound propagation can be represented as transmission line

parameters in a two-port network. Perforated panels, back cavities, and glass wool

absorption materials are represented as matrix of transmission line parameters,

respectively. Transmission line parameters of a perforated panel with a back cavity are

calculated as multiplication of their matrices. An input impedance can be calculated

from the transmission line parameters. A normal incident absorption coefficient is

calculated from the input impedance. Holes of the perforated panels have losses of

viscous friction and thermal conduction at their walls. Simulations are done in the

condition of 0.25 mm to 5 mm diameters of holes, 0.25 % to 25 % perforation rates,

0.5 mm to 5 mm thickness of the perforated panels with back cavities in which there

are or are not glass wool absorption materials. The results of these simulations are

good agreements with the results of our measurements by transfer function method

except in the condition of more than 1 mm diameter of holes.

Settlement of Crystalline Structure of Group II-VI Semiconductor

Nanoparticles by Profile Refinement Technique and Size

Determination by Tight Binding Model

Dr. Surendra Kumar Gautam

Department of Chemistry

Tribhuvan University, Nepal

ABSTRACT

Group II-VI semiconductor nanoparticles have gained huge interest both in

fundamental research and technical applications due to their unique optical and

electrical properties. They have extensive range of applications in the field of opto-

electronic devices such as light emitting diodes (LEDs), bio-sensors, photo-detectors,

solar cells etc. The properties of such semiconductor nanoparticles are dependent on

22

their crystal structure and size. The settlement of crystalline structure of those

nanoparticles is done by profile refinement of X-ray diffraction (XRD) pattern. The

size quantization effect and blue-shift resulting in the change of band gap with

crystalline size are studied from optical absorption spectra. Particle sizes are calculated

by Tight Binding (TB) model. The verification of particle size and the crystalline

structure of as-synthesized nanocrystals are further carried out from Transmission

Electron Microscopy (TEM) images and Selected Area Electron Diffraction (SAED)

patterns.

Readerless RFID Transponder: A Concept

Mamun Bin Ibne Reaz

Professor

Department of Electrical Electronic and Systems Engineering

Universiti Kebangsaan Malaysia

Bangi, 43600 Selangor

Malaysia

ABSTRACT

RFID is a technology that enables the non-contact, automatic and unique identification

of objects using radio waves. It is projected that the RFID market will be worth more

than US $25 billion in 2018. However, RFID system suffers from limited address

space, local mobility, security and privacy. Above all, it is a monopoly business with

few vendors, which are trying to dominate the market with proprietary standard of

RFID reader. In order to overcome this major issue, we are proposing an RFID tag

communicating with the existing wireless network interface card (WNIC) instead of

the reader. In this novel RFID transponder, IPv6 address will be used to provide a

universal identification number to the objects with seamless address space, global

mobility and data security. The existing EPC (Electronic Product Code) of RFID will

now directly map into IPv6 address by using auto configuration method. This talk will

mainly focus on different key issues related to this readerless RFID system and

suggests the mechanism of reducing significant cost, physical location detection and

usage of global unique address.

Science, Technology and Innovation: For Engineering Our Future

Aravind Chinchure

Chair Professor

Centre for Enterpreneurship & Innovation Symbiosis International University

Pune, India

ABSTRACT

Every third person in this world today subsists with income less than 2 dollars a day.

23

Majority of these people live in Asian countries. The rising socio-economic inequality

and climate change are posing biggest risk to the world. The world is also moving

towards knowledge-based competencies and industries where the emphasis is not on

tangible assets, but on intangible knowledge assets, which is good news for the

developing countries. Today, world’s major knowledge industries are based on

telecommunication, microelectronics, new materials, and information technology. This

offers a great opportunity to researchers from academia and industry from Asian

countries to appropriately engineer the future of the nation and society by developing

innovative solutions to cater to the needs of the people for the sustainable and inclusive

development. To reap the full benefit of the emergence of knowledge-based industry

requires building a new culture in academia and industry that seamlessly connects

science, technology and innovation with relevance. This talk provides several

examples of academic institutions, professors, researchers who have been able to

successfully leverage science, technology and innovations for national and social good.

I will also present new ideas that are emerging and debated in the world on the cost-

conscious frugal science and technology.

Progress of Low-Temperature Fabrication Technologies of Thin

Film Transistors for Preservation of Global Environment

Dr. Susumu Horita

Associate Professor

School of Materials Science

Japan Advanced Institute of Science and Technology, Japan

ABSTRACT

Low-temperature fabrication technology can contribute enhancement of preservation

of global environment due to reduction of not only power or natural source

consumption but also greenhouse effect. So, as an example, we have been studying on

low-temperature fabrication of poly-or microcrystalline silicon (poly-Si) films on

temperature-sensitive and cheap substrates. The appropriate applications of poly Si are

thin-film transistor (TFT) in an active matrix flat panel display, solar cells, and so on

because of its higher reliability and higher mobility. For future application, industry

requires shorter annealing time, lower annealing temperature, more uniform electrical

property such as mobility, threshold voltage, and so on in a whole substrate area. For

meeting these requirements, we have been investing solid-phase crystallization of Si

film on a glass by using pulsed laser annealing(PLA) with crystallization-induction

(CI) layer of yttria-stabilized zirconia [(ZrO2)1-x(Y2O3)x :YSZ]. PLA method can

crystallize Si films effectively at room temperature because of its short pulse duration

time less than 10 ns. The YSZ CI layer can transfer its crystalline information to

crystallized Si on a glass so that we can obtain higher and more uniform crystalline

quality of Si films.

24

As another important technology for TFT, we have low-temperature fabrication of

insulator, e.g, oxide film, in particular, for gate layer. Our group uses atmospheric

pressure CVD (AP-CVD) with silicone oil and ozone gas to obtain Si oxide. As you

know, silicone oil is a safety material for human body, chemically stable, and cheap,

and ozone is an indispensable substance for global environment. By using this

technology, we can produce a SiO2 film at 200 to 300 0C.

In this conference, our previous and current research results on the above low

temperature technologies are presented, including new two-step method in PLA for

much improving film quality and characteristics of Si TFTs fabricated by the above

techniques.

Factors that Facilitate and Impede Hitherto Untried R&D in

Engineering

Dr. Sisil Kumarawadu

Professor

Department of Electrical Engineering

University of Moratuwa, Sri Lanka

ABSTRACT

Completing a research & development (R&D) or postgraduate research project

requires stamina, determination, and willingness to stretch your intellectual and

emotional capabilities. An unmistakable initial momentum is vital to avoid stall out at

some point when working on a research project as it may lead to feel as if there is no

light at the end of the tunnel, or encounter disappointments or unexpected setbacks, or

even embarrassments. Not to be confused with procrastination or self-doubt, which is a

mental road-block, second-guessing at the conception of the R&D project is to guess,

predict or anticipate what you will encounter during the project duration and what you

will finally end up with. This is quite challenging as the research project opportunities

that come along the way of a researcher are rarely repetitions of previous experiences

(hitherto untried). Right second-guessing requires some quality time for an effective

literature review process, peer consultation, reflection, mulling things over and sitting

with your thoughts. The speaker intends to use his close to two decades long

experience in academic and industrial research projects as well as at postgraduate and

undergraduate levels to discuss, with some real world examples, the factors that

facilitate and impede hitherto untried R&D efforts in engineering & technology.

25

Water-in-oil Microemulsions as Nanoreactors to Prepare

Nanoparticles with Tunable Electrical, Optical, and Antibacterial

Properties

Md. Abu Bin Hasan Susan

Professor

Department of Chemistry, University of Dhaka, Dhaka 1000, Bangladesh

ABSTRACT

Nanomaterials with tunable electrical, optical and antibacterial properties have been a

fascinating domain of current research for their promising application in diverse areas.

In this work, we report preparation, characterization and applications of metal and

metallic oxide nanoparticles, core@shell nanoparticles and polymer-nanocomposites

from water-in oil (w/o) microemulsions as nanoreactors. Attempts have been made to

tune electrical, optical and antibacterial properties of the nanoparticles and

nanocomposites by controlling parameters for preparation of w/o microemulsions.

Physics and Technology in Radiation Oncology and Imaging

Golam Abu Zakaria

Professor

Department of Electrical, Mechanical and Industrial Engineering

Anhalt University of Applied Sciences, Koethen, Germany

ABSTRACT

Medical Physics is the application of physics concepts, theories and methods to

medicine or healthcare. On the other hand, biomedical engineering is the application of

engineering principles and design concepts to medicine and biology for healthcare.

Medical physicists and biomedical engineers play a vital and often leading role for any

medical research team. Their activities cover some key areas such as cancer, heart

diseases and mental illnesses. In cancer treatment, they primarily work together on

issues involving imaging and the radiation oncology. Thus the medical

physicist/biomedical engineer plays a mandatory role in every radiation oncology

team.

The capability of controlling the growth of any cancer with radiation dose is always

associated with the unavoidable normal tissue damage. Accordingly, many physical-

technical developments in radiotherapy facilities are aimed to give a maximum

radiation dose to tumour cells and – at the same time - minimize the dose to the

surrounding normal tissue.

26

For that reason, 60-Co irradiation units were developed in the 50ties and medical linear

accelerators in the following decades. Last but not least, neutrons, protons and even

heavier ions have also been applied. At the same time, treatment calculation and

delivery methods have continuously been improved from conventional multi-beam

techniques to tumour shape conformal methods such as radio surgery, Intensity

Modulated Radiotherapy (IMRT), Image Guided Radiotherapy (IGRT), Stereotactic

Body Radiation Therapy (SBRT) tomotherapy and brachytherapy (BT).

The concentration of dose to tumour requires precise information of the shape and the

anatomical geometry of the tumour within the body. The techniques providing such

pieces of Information in a visible form is summarized by the term “Imaging”. X-ray

has played a dominant role almost from the time of its discovery in 1895. Up to now,

the use of x-rays has been extended to tomographic imaging with Computer

Tomography (CT) and other imaging modalities like Ultrasound (US), Magnetic

Resonance Imaging (MRI) or Positron Emission Tomography (PET) which have been

developed over the last decades. By their combined use, the required information level

on the clinical tumour target volume for radiotherapy has been tremendously raised.

The physical and technical development of radiation oncology and imaging are

discussed in this talk covering aspects of biology as well.

27

Epitaxial Growth of MnAl Films with CrRu Buffer Layer

M. A. I. Nahid1, Mikihiko Oogane

2, Miho Watanabe

2, Hiroshi Naganuma

2, and Yasuo Ando

2

1 Dept. of Applied Physics & Electronic Engineering, University of Rajshahi, Rajshahi, Bangladesh

2 Ando Lab, Dept. of Applied Physics, Tohoku University, Sendai, Japan

*[email protected]

KEYWORDS: L10 material, magnetic anisotropy.

The utility of perpendicular MTJ of L10 class

material is promising in future anticipated spin RAM

and STT-RAM devices. In order to fabricate such a

device with higher recording density, it is requisite to

have rare-earth-free magnetic material with the

combination of magnetic properties of high Ku, low MS

and low damping constant (). In such considerations,

-phase MnAl is one of the potential L10 materials[1].

However, the -phase MnAl is metastable in bulk and

easily broken into other phases, which is not

ferromagnetic. The MnAl in thin film form is very

sensitive to the growth temperature, substrate and

buffer layer. We tried various substrate and buffer layer

and found that the -phase MnAl films were epitaxially

grown using CrRu buffer layer on MgO(100) substrate

at growth temperature (TS) ranging from 250C-350C.

The magnetic and structural properties of MnAl alloy

thin films were studied as a function of TS ranging from

150C-500C.

The MnAl films of 40 nm were fabricated using dc

sputtering in a high vacuum of base pressure 2×10-7

Torr on MgO(100) substrate with 50 nm CrRu buffer

layer. The saturation magnetization (MS) and the

effective magnetic anistotropy (Keff) are estimated from

the hysteresis loops. They are plotted as a function of

substrate temperature in fig. 1. At lower TS, the MS and

Keff are relatively small and maximum at 300C and

then decreases. One can estimate, the uniaxial magnetic

anisotropy from the Keff and MS by

equation,22 SeffU MKK . The maximum value

of KU is obtained at 300C of about 2×106 erg/cc which

is comparable to the other reports. The films grown at

300C possess higher magnetic anisotropy constant of

about 2×106 erg/cc. The saturation magnetization in

that sample is obtained about 470 emu/cc. In order to

study the structural properties, the /2 scan x-ray

diffraction of the MnAl films were carried out at the

various TS. In the samples, TS=250C-350C, the (200)

CrRu buffer layer peak is clearly observed.

Figure 1. The saturation magnetization (MS) and effective

magnetic anisotropy constant (Keff) with the variation of TS.

The lattice constant based on the (200) peak is

about 2.77Å. With 45 angle the lattice constant, a of

CrRu is obtained about 3.91 Å, which is very close to

the bulk lattice constant value of MnAl. The lattice

misfit in this case is about 0.7%. At the TS range from

250 to 350C, two peaks of MnAl films are observed at

about 25.3 and 54.4. These peaks correspond to (001)

and (002) of -phase MnAl. Above TS >350C, there is

no peak of MnAl that corresponds to -phase of MnAl.

At TS=300C, the integrated intensity ratio of (001) and

(002) peak is maximum suggesting the higher ordered

sample. The magnetic anisotropy is also maximum in

that case. Therefore it is understood that there is a

strong correlation between structural and magnetic

properties of the MnAl film which can be tuned by the

growth temperature.

[1] Hosoda et al., J. Appl Phys. 111, 07A324 (2012).

Paper ID: 3

mailto:*[email protected]

28

Fabrication of Cardiac Biomarker Immunosensor Based on Three Different

Electrode Surfaces and Contrasting Their Efficiencies

Payal Gulati, Prabhash Mishra and S.S.Islam*

Centre for Nanoscience and Nanotechnology, Jamia Millia Islamia, New Delhi-110025, India

* [email protected]

KEYWORDS: CVD, AMI, PS, GCE, GPMS, ITO, CV, EI, Ag-Ab Immun-ocomplex.

Due to increased threat of Cardio-vascular disease

in this modernized world because of bad eating habits,

there is a requirement for the early detection of

biomarkers related to cardiac abnormalities.

Accordingly, a peculiar label-free electrochemical

immunosensor is developed for the detection of Acute

Myocardial Infarction (AMI) and for few other cardiac

related disorders. Myoglobin being the small size

protein is released in the bloodstream soon after the

commencement of AMI. For the immunosensor

fabrication, three types of electrodes were used porous

silicon, glassy carbon electrode and indium tin oxide

coated glass. All the electrodes were functionalized

with 3-glycidoxypropyltrimethoxy silane (GPMS) for

18 hour at room temperature resulting in highly

reactive epoxy groups on the electrode surface which

promptly reacts with amino group of the Monoclonal

Anti-Myoglobin Human antibody at 4∘C for 24 hour.

To block the non- binding sites on the electrode surface

BSA (Bovine Serum Albumin) treatment was done on

their surfaces for 1hour at room temperature. Finally

sensing on the electrode surface was done after

immobilizing with the Antigen (Myoglobin from

Equine Heart) for 1hour at room temperature in a linear

range from 0.01µg/ml to 1µg/ml.

Cyclic Voltammetry technique was applied to study

the stable Antigen-Antibody immune-complex

formation using three- electrode configuration. All the

electrodes responded to the stepwise changes done on

their surface but out of these, glassy carbon electrode

was found to be highly sensitive as it showed more

current change with respect to small change in the

antigen concentration (as shown in fig1). This

Electrochemical Immunosensing (EI) technique has

certain advantages over other hospital diagnosis

methods like easy to fabricate, low cost, less time

consuming, early detection, and miniaturization which

does not require an expert person to handle this task. EI

technique is suitable for biological prognostics as

electrical signals obtained from the biomolecules

reaction are electrochemical in nature thereby

rendering it a good sensitive and reliable method.

0.0 0.2 0.4 0.6 0.8 1.0

0.060

0.065

0.070

0.075

0.080

0.085

0.090

GCE

I (m

A)

Myoglobin Conc. (microgm/ml)

PS

ITO

Figure 1. Current v/s Myoglobin Conc. plot achieved from

cyclic voltammetry studies

Paper ID: 5

29

Simulation of the Electrical Activity of Cardiac Tissue by Finite Element Method

Tanzina Rahman* and Md. Rezaul Islam

Applied Physics and Electronic Engineering Department, University of Rajshahi, Rajshahi, Bangladesh

*[email protected]

KEYWORDS: Electrical activity of heart, Bidomain &Monodomain model, Finite element method.

The electrical activity is responsible for the periodic

contraction and relaxation cycle of the human heart.

Significant implications of simulating electrical

activities are helpful to understand cardiac

abnormalities like cardiac arrhythmias. Mathematical

modeling of heart provides a better understanding for

the complex biophysical phenomena related to

electrical activity in the heart.

Various electro-physical models have been developed

to simulate electrical properties of cardiac tissue. The

bidomain model represents the cardiac tissue at a

macroscopic scale by relating the transmembrane

potential, the extracellular potential and the ionic

currents.

Monodomain model is used to simulate the electrical

activity of cardiac tissue. It has reduced complexity

than bidomain model. In this model, it is assumed that

the anisotropy of the intracellular and extracellular

spaces is the same, i.e. that the conductivity in the

extracellular space is proportional to the intracellular

conductivity. A Galerkin finite element (FEM) is used

to discretize the monodomain model equation.

Different methods are used to simulate the electro-

physical models. But FEM is used because it provides

better approximate than other methods such as finite

difference method and many more.

The heart tissue is divided into different nodal elements

to simulate the propagation of excitations at a uniform

time step. The surface plot which is shown in the figure

1 represents how excitation is propagated in the cardiac

tissue.

The surface plot represents the propagation of

excitation for 35×35 nodal numbers. In case of 35×35

nodal elements current above the threshold value is

applied to the middle node of the square mesh grid and

less than threshold value to the others node. The

diffusion coefficient dx along the fibers and dy

perpendicular to the fibers but not in the plane. The

value of dx is more than dy. The excitations from the

middle nodes propagated to the nearby horizontal and

vertical nodes. The propagation of excitations along the

horizontal axis is more than the vertical.

It is because value of dx is more than dy. The potential

of those nodes changes consequently.

Figure 1. Propagation of excitation for 35× 35 nodal

elements.

After a uniform time interval, the excitations reached at

the end of the nodes. Then the process started from the

initial condition. Figure-1 consequently shows the

process.

This work has been able to create some insights about

the electrical behavior of human heart, revealing the

nature of the excitation, propagation pattern in the

cardiac tissue. Generating an efficient numerical

solution of these models is a challenging task, and in

fact the physiological accuracy of tissue-scale models

is often limited by the efficiency of the numerical

solution process

Paper ID: 9

30

Baking of Ilmenite on Moistening with Sulfuric Acid Followed by Leaching with

Dilute Sulfuric Acid Solution

Ranjit K. Biswas1*, Aneek K. Karmakar

1, Jinnatul Ara

1 and Muhammad A. Gafur

2

1Department of Applied Chemistry and Chemical Engineering, University of Rajshahi, Rajshahi, Bangladesh

2Pilot Plant and Project Development (PPPD) Division BCSIR Laboratories, Dhaka, Bangladesh

* [email protected]

KEYWORDS: Ilmenite, H2SO4, Baking, Leaching, Chloride feed.

Metallic titanium and its alloys are extensively used in

air-craft industry for their high strength and low

density. These can be prepared by reduction of pure

TiO2. TiO2 also finds application in pigment, plastic,

paper, textile, rubber and ceramic industries. Pure or

pigment grade TiO2 was manufactured earlier from

rutile (naturally occurring TiO2) by the chloride

process, in which, TiCl4 was distilled out on heating

rutile together with coke in presence of Cl2 gas

followed by its decomposition in air to TiO2 and Cl2

gas. By the present time, reserve of rutile on the earth

crust is depleted off, and consequently, now the

titanium industry solely depends on ilmenite. If

ilmenite, on dissolution, forms a solution containing all

Ti and Fe present in it, then the uses of solvent

extraction, selective precipitation, thermal hydrolysis,

filtration and ignition steps may produce pigment grade

TiO2. Otherwise, Fe from ilmenite can be preferentially

removed from ilmenite to get synthetic rutile for the

chloride process. In the first method, the partial

leaching of both Ti and Fe from ilmenite results in low

yield of TiO2 manufacture. The method for complete

dissolution of ilmenite is still unpublished.

As it is very difficult to dissolve ilmenite

quantitatively, hundreds of paper and patents are

available on its dissolution using various ways and

techniques. But none from H2SO4-baking. A new, easy

and attractive technique for ilmenite dissolution is

described in this work. In this work, ilmenite has been

moistened with conc. H2SO4, baked (heated) at a

temperature near the boiling point of H2SO4 and then

leached with very dilute solution of H2SO4. The

optimized baking temperature and time are found to be

300o C and 30 min (for ilmenite to H2SO4 wt. ratio of

1), respectively. The single stage baked mass can be

leached by 0.10 mol/L H2SO4 solution at its boiling

temperature under reflux and at ilmenite to liquid ratio

of 0.01 g/mL for 20 min to extract ~86% Fe and only

1.1% Ti from ilmenite. On the other hand, the 3-stage

baking and 1-stage leaching (Figure 1) remove as much

as 97% Fe and only 2.4% Ti. The residue left after 3-

stage baking and 1-stage leaching is almost white and

identified as rutile (Figure 2), which can be regarded as

a very good quality Chloride feed for manufacture of

pigment grade TO2. Reasonable low temperature and

time required in baking, together with the requirements

of very low concentration of H2SO4 and time in

leaching will attract technologists of this field to adopt

this technique for almost complete removal of Fe from

ilmenite to produce Chloride feed grade TiO2.

Figure 1. Effect of stage-wise baking. Wt. of H2SO4 added in

each stage = 1 g, baking time =30 min, baking temperature

= 200o C (open symbols) and 300o C (closed symbols),

leaching time = 20 min. (,), Ti4+; (,), Fe3+; (,),

Fe2+; (,), Total Fe (Fe3+ + Fe2+).

Figure 2. SEM images of (a) H2SO4 - baked mass and (b)

residue obtained on leaching of H2SO4 - baked mass.

0 1 2 3 40

20

40

60

80

100

Number of baking stage

Wt.

% o

f m

etal

s

dis

solv

ed

Paper ID: 10

31

Thermal Treatment of Ilmenite on Moistening with Concentrated HF followed by

Leaching with Dilute Sulfuric Acid Solution

Ranjit K. Biswas1*, Mohammad A. Habib, Aneek K. Karmakar

1, and Mohammad J. Alam

1

1Department of Applied Chemistry and Chemical Engineering, University of Rajshahi, Rajshahi, Bangladesh

* [email protected]

KEYWORDS: Ilmenite, HF, H2SO4, Baking, Leaching, Chloride feed.

The beach sand of Bangladesh contains considerable

amount of ilmenite, zircon, monazite etc. These are

fractionated at the BAEC‘s Pilot Plant of Beach Sand

Exploitation Centre at Kalatali, Cox‘s Bazar. Ilmenite

is now the principal source of TiO2 - widely used as

white pigment in paint industry. In order to process

ilmenite for the manufacture of TiO2, it is necessary to

decompose it completely or partially by acid. However,

as it is very difficult to dissolve ilmenite (FeTiO3)

quantitatively, enormous studies on ilmenite

dissolution have been patented and published. The

objective of this work is to apply a new technique for

dissolutions of both Ti and Fe, or dissolution of either

Fe or Ti, preferentially, from ilmenite.

In this process, ilmenite has been moistened with conc.

HF, heated at a higher temperature (150-225 C), and

the resultant mass has been leached with dilute sulfuric

acid solution at its boiling temperature under reflux.

The optimized baking temperature and time are

determined to be > 170 C and 20 min (for ilmenite to

HF wt. ratio of 1), respectively. The baked mass can be

leached by 0.50 mol/L H2SO4 solution at solid to liquid

ratio of 0.01 g/mL for 1 h to extract 37% total iron and

7% Ti from ilmenite sample. The stage-wise baking

and single stage leaching improve total iron dissolution

and decrease titanium dissolution (Fig. 1). About 37%

Fe-dissolution from the 1st stage baked mass is

increased to 97% Fe-removal from the 6th

stage baked

mass. On the other hand, 7% Ti-dissolutions from the

1st to 4

th stage baked mass are decreased to ~4.3% Ti-

dissolution from the 6th

stage baked mass.

Comparison of the XRD patterns and SEM images

(Fig. 2) of mother ilmenite with those of the residue

obtained on 96% Fe-removal indicate that ilmenite and

related crystals are completely destroyed to form an

amorphous - almost white product containing most of

titanium in the sample. As the residue contains only

4% originally present iron of the sample, it can be

regarded as a good quality feed material for the

Chloride process of pigment grade TiO2 manuracture.

Figure 1. Effect of stage-wise baking on metals dissolution

from ilmenite. Wt. of ilmenite sample = 1 g, particle size = <

53 µm, amount of concentrated HF added to moisten ilmenite

= 1 g in each stage, baking temperature = 200oC, leaching

agent = 100 mL 0.5 M H2SO4, leaching temperature = boiling

temperature, leaching time = 1 hour.

Figure 2. SEM images of (a) Ilmenite sample and (b) residue

left after leaching of HF - baked mass.

0 1 2 3 4 5 6 70

20

40

60

80

100

Wt.%

of m

eta

ls d

issolv

ed

Number of baking stage

() Total

Fe

() Fe3+

() Fe2+

() Ti4+

() Total Fe

() Fe3+

() Fe2+

() Ti4+

(a)

(b)

Paper ID: 11

32

Solvent Extraction of V(V) from Nitrate Medium by Tri-n-Octylamine Dissolved in

Kerosene

Ranjit K. Biswas*, Aneek K. Karmakar and Mottakin

Department of Applied Chemistry and Chemical Engineering, University of Rajshahi, Rajshahi, Bangladesh

* [email protected]

KEYWORDS: Extraction, stripping, V(V), TOA, loading, NO3--medium.

Technologically important metal, vanadium, is

usually recovered from petroleum fly ashes, tar sand

fly ashes, black shales, hydrodesulphu-rization waste

catalysts, carnotite etc. using roasting, leaching,

extraction-stripping, precipi-tation and reduction steps.

On leaching V dissolves either as V(IV) or V(V);

which are easily inter-transferrable. Extraction

behavior of VO2+

is simple in nature for absence of its

species variation with pH; but that of V(V) is complex

in nature for its extensive species variation with pH.

V(V) exists as VO2+ at pH<1 and starts to hydrolyse

and polymerise with increasing pH to form species,

such as, VO(OH)3, V10O26(OH)24-

, V10O27(OH)5+

,

V10O286-

, V3O93-

etc. within the acidic region. As a

result, extraction of V(V) is little investigated from

mechanistic point of view, though there are reports on

its separation from other metal ions by several

extractants. Consequently, this system is studied from

mechanistic view point.

Tri-n-octylamine (TOA) is found to extract V(V)

rapidly; equilibration time is only 5 min. The extraction

ratio is found to increase with initial [V(V)] in the

aqueous phase and [TOA] in the organic phase. The

variation of extraction ratio with pH is crucial, as

presented in the embodied figure:

Figure 1. Effect of equilibrium and initial pH on extraction.

[V(V)](ini) = 0.50 g/L, Temp. = 300 K, O/A = 1, time = 10

min; [TOA] = 0.10 mol/L.

The variation of extraction ratio with pH(eq) is not

sharp within pH(eq) of 1-2; but it is considerably sharp

within pH(eq) of 2-5. Distribution of different V(V) –

species with pH is collected from literature; and their

extraction equilibrium reactions have been suggested to

guess the expected pH-dependence values at various

pH-values. The calculated pH dependences of -0.5 to -

0.6 in pH(eq) region of 3 to 5 are observed in log D vs.

pH(eq) plots. So the suggested extraction equilibrium

reaction holds good.

The variation of extraction with [NO3-] can be

expressed by a relation: log D = const. + log

1+1.82[NO3-]); where, the value of constant depends of

experimental parameters used. This equation suggests

that the extraction is independent of [NO3-] in its lower

concentration region, but is directly proportional to

[NO3-] in its higher concentration region. With the rise

of temperature, extraction ratio decreases; so the

process is exothermic with H value of -58 kJ/mol.

TOA exhibits a very high loading capacity towards

V(V). 1 mole of TOA can load as much as 2.3 mole

V(V). The extracted V(V)-species can be stripped by

0.75 mol/L NH4OH solution to the extent of 72%; and

the stage-wise stripping is not so effective.

Vanadium(V) can be effectively extracted by TOA

solution in the equilibrium pH region of 1.5-2.5 with

high loading capacity of 2.3 mole V(V)/mole TOA.

1 2 3 4 5 6-2

-1

0

1

2

32 3 4 5 6

log D

pH(ini)

pH(eq)

Paper ID: 12

33

Kinetics of Extraction of Ti(IV) from SO42-

Medium by Cyanex 301 Dissolved in

Kerosene

Ranjit K. Biswas* and Aneek K. Karmakar


* [email protected]

KEYWORDS: Kinetics, extraction, Ti(IV), Cyanex 301, Lewis cell.

In a previous paper, the partitioning of Ti(IV)

between SO42—

H2O/Cyanex 301-kerosene-5% (v/v)

heptan-1-ol has been reported from the equilibrium

view point. This paper deals with the same system

from the kinetics point of view. For kinetic

measurement, a constant interfacial area stirred (Lewis)

cell operated at 3 Hz has been used. The rate per unit

area i.e. flux of Ti(IV) transfer from the aqueous to

organic phase is found to be directly proportional to

[Ti(IV)] in the aqueous phase (if [Ti(IV)] < 1.2 g/L)

and [HA] in the organic phase. It is inversely

proportional to the terms [1+447[H+]) and (1+1.8[SO4

2-

]). The rate constant at 293 K has been determined to

be 10-4.288

. The activation energy, Ea has been measured

to be 37-60 kJ/mol depending on experimental

parameters. The enthalpy change on activation, S± is

always highly negative (-127 to -205 j/mol K).

The empirical flux equation at 293 K can be split to

four different cases, i.e. case 1: at low concentration

region (lcr) of H+ and SO4

2-, case 2: at lcr of H

+ but at

higher concentration region (hcr) of SO42-

, case 3: at

hcr of H+ but lcr of SO4

2- and case 4: at hcr of H

+ and

SO42-

. Considering stability constants of existing

species, partitions of HA, H+ and A

- and acid

dissociation constant of HA, the flux equation in above

cases, take on the same form: F = constant [TiO2+

] [A-].

This equation indicates that the attachment of the first

anionic ligand (A-) to TiO

2+ for forming TiOA

+ is the

slowest step. The addition of 2nd

A- with TiOA

+ and the

diffusion of a reactant species to the reaction site or of

a product species from the reaction site are faster than

this step.

The Ea value of 37-60 kJ/mol indicate that

depending on experimental parameter, the process i.e.

either solely chemically control, or, intermediate

control. High negative S± values suggest that the rate

determining chemical reaction step as cited above

occurs via an SN2 mechanism, i.e. the attach of the 1st

anionic ligand (A-) on hydrated TiO

2+ to form higher

co-ordinated activated complex [TiO(H2O)nA]+ is

slower than the rate of dehydration step to form normal

co-ordinated [TiO(H2O)n-1A]+ and also the rate of

addition of 2nd

A- to form [TiOA2].

Figure 1. Effect of [Ti(IV)] and pH on its rate of forward

extraction by Cyanex 301. [SO42-] = 0.50 mol/L, [Heptanol](o)

= 5% (v/v), Area (Ai) = 0.00375 m2, O/A = 1 (O = 100 mL).

(), pH = 2.50, [Cyanex 301](o) = 0.40 mol/L, Temp. = 318

K; S = 0.95, I = -4.65; (), [Ti(IV)](ini) = 1000 mg/L,

[Cyanex 301](o) = 0.06 mol/L, Temp. = 293 K.

Figure 2. Effect of [Cyanex 301] and [SO4

2-] on its rate of

forward extraction by Cyanex 301. (), pH = 2.50,

[Ti(IV)](ini) = 1000 mg/L, Temp. = 293 K; S = 1.04, I = -

6.585; (), pH = 1.80, [Cyanex 301](o) = 0.40 mol/L,

[Ti(IV)](ini) = 1000 mg/L.

-3.0 -2.5 -2.0 -1.5 -1.0-9.5

-9.0

-8.5

-8.0

-7.5

-7.0

-6.5

-6.0

-5.51.0 1.5 2.0 2.5 3.0 3.5 4.0

log (

F, km

ol/m

2s)

log {[Ti(IV)], mol/L

pH

-1.4 -1.2 -1.0 -0.8 -0.6 -0.4 -0.2 0.0-9.0

-8.5

-8.0

-7.5

-7.0

-6.5-2.5 -2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5

log

(F

, km

ol/m

2s)

log {[Cyanex 301], mol/L}

log {[SO2-

4], mol/L}

Paper ID: 13

34

Temperature Dependence Magnetoresistance of

Fe73.5 –Cu1 –Ta3 –Si13.5 –B9 Magnetic Alloy

M. Z Ahsan1*

, F. A. Khan2 and M. Perven

3

1Military Institute of Science and Technology, Mirpur Cantonment, Mirpur-12, Dhaka-1216

2 Bangladesh University of Engineering and Technology (BUET), Dhaka-1000

*[email protected]

KEYWORD: FINEMET; Magnetic alloy; Normalized resistivity; Magnetoresistance; Microcrystallites.

The Fe-Si-B system exhibits superior soft magnetic properties and thus is very useful in magnetic devices such as EMI filters, current sensors, magnetic sensors, transformer core and magnetic shielding sheets as reported in various literatures. The magnetic alloy with composition of Fe73.5–Cu1–Nb3–Si13.5–B9 was first developed in the form of ribbon by Yoshizawa [1]. It crystallizes in the form of ultrafine grains of -Fe (Si) embedded in the remaining ferromagnetic amorphous matrix. Magnetoresistance (MR) is a magnetotransport property and guiding factor for determining soft magnetic behavior of this class of magnetic alloys. It is phenomenologically known as magnetostriction, which refers to the change of resistance with the applied magnetic field.Its value for Fe73.5–Cu1–Nb3–Si13.5–B9 magnetic alloy found to be nearly zero. This magnetic alloy system gives rise to outstanding magnetic softness and thereby commercially fabricated under the trade name of FINEMET as reported in 1999 [2]. Fe73.5 –Cu1 –Ta3 –Si13.5 –B9 is a similar magnetic alloy, prepared by rapid quenching technique in the form of ribbon, where Niobium (Nb) was substituted by Tantalum (Ta). Many investigations thus far have been performed of this magnetic alloy but significant temperature response on its magnetotransport properties are yet to be reported. As such, this paper intended to reporta comprehensive study on temperature dependent Magnetoresistance of Fe73.5 –Cu1 –Ta3 –Si13.5 –B9magnetic alloy and effect of tantalum (Ta) there on. The room temperature amorphousity of the sample was verified by JDX-8P GEOL X-ray diffractometer.Figur-1 illustrates X-ray diffraction pattern. Two broad peaks found to be within scanning angular range of 5

0 and 65

0, which conforms its bi-

phase nature. The magnetotransportproperties of the sample were studied as a function of temperature using conventional 4-probe technique. The sample exhibits semiconducting behavior at low temperature (35 – 295K) and metallic behavior at high temperature (350 – 700K). The MR found to increase with both the temperature and field in the low temperature range due to the ordering effect of microcrystallites in the

amorphous matrix. A significant dispersion of negative MR obtained at 700 K for the applied magnetic field of 4 KG due to more disorderliness of spins in the amorphous matrix of the sample. Its Zero Field Cool (ZFC) and field cool (FC) measurement demonstrates both the positive and negative slopes of MR below and above a critical transition temperature, 175 K as shown in Figure-2. Up to this critical temperature, the system makes a transition from ferromagnetic to antiferromagnetic and above this temperature from AFM to FM, which is an indicative of reentrant ferromagnetic phase, which leads to be used it in sensors and magnetic devices.

0 50 100 150 200 250 3001

2

3

4

5

6

7

8

MR

Temperature (K)

References: [1] Yoshizawa, S. Oguma, K. Yamauchi, J. Appl. Phys. 64 (1988)

6044. [2] V. Franco, C.F Conde, A Conde, Magnetic properties and

nanocrystallization of Fe63.5Cr10 Si13.5B9Cu1Nb3 alloy, Journal of

Magnetism and Magnetic Materials 203 (1999) 60-62.

Figure-1: XRD Pattern

Figure-6: ZFC and FC Magnetoresistance graph in the

temperature range 40~265K

Paper ID: 14


35

Production and Improvement of Waste Tire Pyrolysis Oil to Be Utilized with Diesel

in CI Engine

Md. Nurul Islam* and Md. Rafsan Nahian

Department of Mechanical Engineering, Rajshahi University of Engineering & Technology, Rajshahi, Bangladesh

* [email protected]

KEYWORDS: Solid tire waste, Pyrolysis, Crude pyrolysis oil, Improvement, Alternative fuel.

The standard of living, quality of life and development

of a nation depend on its per capita energy

consumption. Global energy supply that mainly

depends on fossil fuel is decreasing day by day due to

population growth. By the year 2100, the world

population is expected to be more than 12 billion and it

is estimated that the energy demand will be increased

by five times from present scenario. Besides

reservation of petroleum of the world will be nearly

exhausted by 2040. Due to the fossil fuel crisis, the

development of alternative fuel technologies are drawn

more attraction to deliver the replacement of fossil fuel.

Pyrolysis is one of the promising alternative fuel

technology that produces valuable oil, char and gas

product from organic waste. Approximately 1.5 billion

tires are produced each year in the world which will

eventually enter the waste stream representing a major

potential waste and environmental problem. In

Bangladesh, total waste tires generation of each year is

about 90000 tones. Conversion of these waste tires to

energy through pyrolysis is one of the recent

technology to minimize not only the waste disposal but

also could be utilized as an alternative fuel for internal

combustion engines. Early investigations report that

tire pyrolysis oil extracted from vacuum pyrolysis

method seemed to have properties similar to diesel

fuel. The crude pyrolysis oil begins to distill in 18 min

at 77o

C and ends in 110 min at 184o C. The highest

amount of oil is obtained in 90 min at 174o C. After

distillation, the distilled tire pyrolysis oil (DPTO) is

obtained 53.75% of the total crude tire pyrolysis oil

(TPO). Experimental studies are conducted in Rajshahi

University of Engineering & Technology on a single

cylinder four stroke air cooled diesel engine fueled

with two different blends, 25% distilled tire pyrolysis

oil and 75% diesel oil (DTPO 25), 50% distilled tire

pyrolysis oil and 50% diesel fuel (DTPO 50) & 75%

distilled tire pyrolysis oil and 25% diesel fuel (DTPO

75). The result of the performance of the engine

indicated that the thermal efficiencies of distilled tire

pyrolysis oil-diesel fuel (DTPO-DF) blends are slightly

lower compared to diesel fuel (DF) but convenient. At

0.45 KW brake power and 27%-part load, the thermal

efficiency is 9.5% for DF whereas for blending of

DTPO 25, DTPO 50 & DTPO 75, it is 9.498%, 9.398%

and 9.304% respectively.

Therefore, reduction in thermal efficiency is about

0.0212%, 1.07% & 2.06% for blending of DTPO 25,

DTPO 50 & DTPO 75 compared to DF.

Besides during tire pyrolysis, CO2, CO, SO2, NO2

gases are produced which are responsible for

environmental pollution. The main concern of this

paper is to produce and improve the properties of crude

pyrolysis oil by desulfurizing, distilling and utilize it

with diesel in CI engine to analyze the efficiency for

various compositions.

Paper ID: 16

36

A Comparative Study of Electrical and Magnetic Properties of Co thin Films

Deposited on GaAs (100), Si (100) and Glass Substrates

Nafisa Ahmed1, A. A. M. Rayhan Kabir

1, M.N.A. Shafi2and J. Islam

1*

1Department of Applied Physics and Electronic Engineering, University of Rajshahi, Bangladesh.

2Questsystem Limited, Lalmatia, Dhaka-1207, Bangladesh.

*[email protected]

KEYWORDS: Magnetization, Coercivity, Interfacial resistance, Resistivity, X-ray diffraction.

Thin films of cobalt (Co) have been prepared by e-

beam evaporation technique on GaAs(100), Si(100)

and glass substrates at a pressure of about 3×10-4

Pa.

The GaAs(100) substrate was n-type Si-doped, with an

electron carrier concentration of 1×1024

m−3. On the

other hand, the resistivity of n-type Si(100) substrate

was 0.02 -cm. The thickness of the films ranges from

100 nm to 160 nm. The deposition rate of the films was

about 10 Ås-1

. The as-deposited films have been

annealed in open air for 5 hours at constant temperature

of 673 K. Electrical, magnetic and structural properties

of the films have been studied.

Electrical and magnetic measurements showed that

the interfacial resistance (IR) is present at Co/Si

interface. The IR is also present at Co/GaAs interface.

The coercivity of 100 nm annealed film on Si(100) is

225 Oe. To increase interfacial resistance of Co/Si, 5

nm Co is deposited on Si substrate and annealed in

open air at 673K for 5 hours. Then a film of 100 nm Co

is deposited on the previously annealed sample. The

final sample become Co(100 nm as-deposited)/Co(5

nm annealed)/Si(100)(hereafter referred to as sample-

1). The magnetic properties of ferromagnetic Co film is

influenced by the interfacial resistance (IR). This

property is attractive for spin valve application. The

coercivity of the sample-1 is about 580 Oe. The

enhancement of coercivity is due the thicker interface

in sample-1.

Not only the interfacial resistance but also the

semiconductor substrate influence the magnetic

properties of ferromagnetic film.To study the effect of

annealing on magnetic properties, a Co film of

thickness 5 nm is deposited on GaAs(100) and

annealed in open air at 673K for 5 hours. Then a film

of 100 nm Co is deposited on the previously annealed

sample. The final sample become Co(100 nm as-

deposited)/Co(5 nm annealed)/GaAs(100) (hereafter

referred to as sample -2).The coercivity of the sample-2

becomes 240 Oe. The value of coercivity of the same

thick as-deposited Co film on GaAs (100) is 100 Oe.

The coercivity of 100 nm as-deposited Co film on glass

substrate is 50 Oe and increased to 100 Oe after

annealing.

-1500 -1000 -500 0 500 1000 1500

-0.0010

-0.0005

0.0000

0.0005

0.0010

Mag

neti

zati

on

, M

(em

u)

Magnetic Field, H (Oe)

B Sample-2

C Sample-1

Figure 1. Magnetization process of 100 nm Co thin film on Si

and GaAs substrate.

Electrical resistivity of 100 nm as-deposited Co

film is 4.35×10−6

-m at room temperature. The

resistivity of sample-1 is deceases to 0.42 ×10−6 -m

due to high value of IR present at Co/Si interface. The

change in resistivity of Co film on Si is metallic nature

but the change in resistivity of Co film on GaAs is

semiconducting nature.

To study the crystalline properties of the deposited film,

the XRD measurement is carried out at room

temperature.The XRD study reveals that the Co film grows

in hcp structure and is microcrystalline in nature.

Paper ID: 23

37

Biological Evaluation of Radiotherapy Treatment Plan for Different Field

Techniques in 3-Dimensional Conformal Radiotherapy (3DCRT)

Kausar A1, Azhari H A

1, Chaudhury S

2, Bhuiyan M A

2 and Zakaria G A

1,3

1Department of Medical Physics and Biomedical Engineering, Gono University, Dhaka, Bangladesh

2Oncology Centre, United Hospital Limited, Dhaka, Bangladesh

3Department of Medical Radiation Physics, Gummersbach Hospital, Academic Teaching Hospital of the University

of Cologne, 51643 Gummersbach, Germany

*[email protected]

KEYWORDS: 3DCRT, TCP, NTCP, Treatment plan.

The quality of radiotherapy treatment plan depends

on the physical as well as biological evaluation. The

purpose of the study is to evaluate the 3D radiation

treatment plan biologically in external beam

radiotherapy.

Biologically evaluated treatment plan quantifies the

tumor controlled by radiation dose and the minimum

probability of normal tissue complications as well as

biological factors that influence treatment outcomes.

The biological evaluation is based on the tumor control

probability (TCP) and normal tissue complication

probability (NTCP). The biological evaluation was

performed for Cervical and Prostate cancer treatment

plan for 4-field, 6-field, 7-field and 9-field techniques

using treatment planning system (TPS) VARIAN

ECLIPSE ARIA 11. Microsoft Excel was used for

calculating the biological evaluation of different field

techniques in both cases. The TCP calculation was

based on the Poisson statistics model. TCP model

generally rely on the assumption that tumor control

requires the killing of all tumor clonogens. Poisson

statistics predict that the probability of this occurring

is:

))(exp( DsNPTCP .......................(1)

Where, N=initial number of clonogens and Ps (D) =

cell survival fraction after a dose D.

The normal tissue complication probability (NTCP)

was calculated according to the Lyman-Kutcher-

Burman (LKB) model:

dtt

tNTCP

)2

2exp(

2

1.............(2)

In both cases, the NTCP was calculated for bladder,

rectum and femoral heads. All parameters values both

in Poisson Statistics and LKB models were taken from

the patient data.

The TCP of cervical carcinoma for 4-field, 6-field,

7-field and 9-field are 1, 0.991743, 0.990757 and

0.991458 respectively and the NTCP values for bladder

and rectum for all fields are 0.01 and 1.00 and for right

& left femour is approximately 1.00 correspondingly.

Similarly, in prostate carcinoma, the TCP for four

different field techniques are 0.995872, 0.995848,

0.999777and 0.995907 respectively. However, the

NTCP of the bladder for four different fields are 0.99,

0.01, 0.99 and 0.01 respectively whereas the NTCP for

rectum and right & left femour are 1 for all fields. All

plan variants for both cases shows the satisfactory TCP

result (TCP≥0.5) but the NTCP is very higher than the

tolerance limit (NTCP≤0.05) for rectum, right & left

femour for both cases in all fields excluding of bladder

for cervix in all fields and for prostate of 6-field and 9-

field are within the tolerance limit.

Table 1. Results of biological parameters for four

different field techniques in cervical carcinoma.

Parameters 4field 6 field 7field 9field

TCP 1 .99171 .99075 .99145

NTCP Bladder 0.01 0.01 0.01 0.01

NTCP

Rectum

1.00 1.00 1.00 1.00

NTCP Lt.

femour

1.00 0.99 1.00 1.00

NTCP Rt.

femour

0.99 1.00 0.99 0.99

It was observed that the no treatment plan complies the

biological criteria, therefore, the biological evaluation

of treatment plan need to be practiced by the radiation

oncologist as well as the medical physicist. Moreover,

TPS should integrate the biological evaluation tools

and should be used appropriate parameters for

biologically selecting the best treatment plan.

Paper ID: 24

38

Design of a Linearly Polarized Multi-band Transmission Line Feed Microstrip

Patch Antenna for Wireless Communications

Sheikh Dobir Hossain1*, Md. Khalid Hossain

2 and Rebeka Sultana

3

1Dept. of Physics, Jessore University of Science and Technology, Jessore-7408, Bangladesh

2Institute of Electronics, Atomic Energy Research Establishment, Savar, Dhaka-1349, Bangladesh

3Dept. of Electrical and Electronic Engineering, Shizuoka University, Shizuoka, Japan

* [email protected]

KEYWORDS: Micro-strip Antenna, Smith Chart, Cavity Model, Dual Band Antenna, CST, DGS, Transmission

Line Feed.

A linearly polarized transmission line feeding dual band rectangular micro-strip patch antenna is designed for the applications in wireless communication. The microstrip antenna contains a rectangular patch on the upper layer of the dielectric material with a dielectric constant of 2.4 and there is a ground plane below the dielectric material. Here the introduction of cavity model with transmission line feed has the favor of low profile, high gain and wide bandwidth of the antenna. The antenna has overall size of 46.9 mm by 38.01 mm. It gives bandwidths of about 90 MHz and 115 MHz at the resonance frequencies of 2.45 GHz and 4.1 GHz respectively which are founded to be favorable for wireless communications. The electronic circuit miniaturization increases the importance of wireless communication systems. In commercial and government communication systems, it is required to develop a small size of antenna. This technological trend has focused on the development of micro-strip antennas (MSA) with low cost, minimal weight and low profile antenna which is capable of maintaining high performance over a large spectrum of frequencies. The disadvantageous features of MSA such lower value of efficiency, higher value of Q factor, poor polarization purity and spurious feed radiation have diminished their versatility. However, it is possible to improve the bandwidth (as long as 90 percent) and efficiency (up to about 35 percent) of a microstrip antenna by increasing the thickness of the dielectric materials with cavity model.

The proposed microstrip antenna shown in Figure1 is obtained using copper (annealed) rectangular patch of length 38.01 mm, width 46.9 mm and thickness 0.1 mm on the upper layer of the substrate whose thickness is 2.40 mm. Also there is a ground plane on the lower side of substrate with the thickness of 0.01 mm. Here we have introduced two I slots on the patch with length and width are 14.2 mm and 1.4 mm respectively to obatain the dual band antenna operate at the resonance frequencies of 2.45 GHz and 4.1 GHz respectively.

The studied parameters of the designed antenna are return loss (RL), voltage standing wave ratio (VSWR), smith chart, directivity and gain. The RL indicates how amount of power is lost in the load and does not return as a reflection. Our proposed antenna shows the return losses of -21.25 dB and -27.5 dB at the resonance frequencies of 2.45 GHz and 4.1 GHz, respectively. As it is lower than

the acceptable value of return loss i.e. -10 dB so the designed antenna is perfectly matched and the power loss is minimum. The parameter VSWR determines how well the antenna is matched. Our designed antenna shows the VSWR are of 1.3883 at 2.45 GHz and that of 1.0825 at 4.1 GHz frequency which is below 2 (desired value for good antenna). Hence our designed antenna is perfectly matched with minimum loss. The study of smith chart is very important during the design of a MSA. Using smith chart it is possible to obtain proper impedance matching between antenna and transmission line feeding. The impedance of our designed dual band antenna is 49 ohms which is approximately equal to the desired value of 50 ohms, indicating the minimum power loss.

Figure 1. Geometry of dual band MSA.

The directivities of our designed antenna are 7.2 dBi in positive Z direction with angular width of 75.7 deg and 6.3dBi at an angle 49.0deg from positive Z direction with angular width of 68.5 deg at the resonance frequencies of 2.45 GHz and 4.1 GHz respectively and the gains are 6.698 dB and 6.797 dB at those frequencies, respectively which agree well with the previous results. In this paper, firstly we have looked on the design and simulation of single band antenna and then extend it to dual band antenna. Here we have studied return loss, VSWR, smith chart, directivity, gain, bandwidth and operating frequency. The designed antenna shows good impedance matching of approximately 49 ohm‘s also it provides good gain and efficiency which indicate the designed antenna can be used for various applications like RADAR, Bluetooth, Biomedical instruments etc.

Paper ID: 25

39

Design and Fabrication of an Unmanned Video Transmitting

Tele-bot using 3G GSM Network

Md. Mamunoor Islam* and Mehdi Hasan Chowdhury

Department of EEE, Chittagong University of Engineering & Technology, Chittagong, Bangladesh

*[email protected]

KEYWORDS: Unmanned Tele-bot, Wireless Video Transmission, GSM, 3G Technology, EAT.

As interest in robotics continues to rise day by day,

robots have been increasingly assimilated in practical

life. Appropriate robotic systems can play a vital role

where human has limitations to work. Several systems

can be used to control a robot from distance (such as

Bluetooth, RF, Zigbee etc.) which may have the

limitation of working range. At present the use of GSM

network has increased significantly as it has a massive

coverage area throughout the globe. This paper

demonstrates a technical method of construction of an

unmanned Tele-bot using 3G GSM network which can

be controlled by a cell phone from any part of the

world along with the provision of wireless video

transmission.

Different keys in the dial pad of a phone produce

DTMF (Dual-Tone Multi-Frequency) tones. The Tele-

bot will use the signals generated by this DTMF

technology through a simple telephone call. The 3G

technology of cellular communication has been added

to the Tele-bot for wireless video transmission using

the video call technology. The block diagram of the

operating method is shown in Figure 1.

Figure 1. Block diagram of GSM controlled Tele-bot.

Moreover, Edge Avoiding Technology (EAT) has also

been equipped within the system for avoiding sharp

edges which can damage the robotic system. At the

time of operation, if any sharp edge in the ground

comes in front of the tele-bot, it will stop instantly in

that position. Then the user will be notified so that

proper measures can be taken.

The proposed GSM controlled robotic system has been

implemented practically to observe the performance of

the system. Several keys have been used to control the

robot in different direction and the performance of the

tele-bot is very decent as it requires a small time to

respond accordingly. An observation of wireless video

transmission has also been arranged to study the

performance of the system‘s capability to transmit

video to the users which is illustrated in Figure 2.

Figure 2. Implementation of wireless video

transmission.

A comparative study between the proposed robotic

system and existing systems has been conducted and it

is apparent that existing Bluetooth based system by R.

Pahuja et al. (2014) and Zigbee based system by B.

Bharathi et al. (2013) have limitations of ‗effective

range‘. Moreover, Chowdhury et al. (2013), Pathik et

al. (2014) and Shifat et al. (2014) etc. have introduced

their GSM robotic systems having no video

transmitting facility. The proposed 3G technology

based Tele-bot is capable of both wireless control with

massive effective range and the vital additional feature

of wireless video transmission. The video transmission

efficiency of the Tele-bot can be enhanced significantly

by using 4G and LTE technology which may be

incorporated in future.

As the cellular network is wide-spread across the

globe, the proposed Tele-bot can be monitored from

any part of the world. With the wireless video

transmission and EAT technology added in this robot,

the proposed system should play a significant role in

practical applications where human has restraints to

operate.

Paper ID: 26


40

Effect of Sintering Temperature on Nb+Nd Doped Bismuth Ferrite

Sadia Tasnim Mowri1*, M A Gafur

2, Quazi Delwar Hossain

1, Aninda Nafis Ahmed

2 and Muhammad Shariar

Bashar3

1 Electrical and Electronic Engineering Department, Chittagong University of Engineering and Technology,

Chittagong, Bangladesh. 2Pilot Plant and Process Development Center

3Institute of Fuel Research and Development, Bangladesh Council of Scientific and Industrial Research, Dhaka,

Bangladesh

KEYWORDS: XRD, SEM, Dielectric Constant, Resistivity, Bismuth Ferrite, Niobium Neodymium.

(Bi2O3Fe2O3)0.4(Nb2O5Nd2O3)0.6 was developed by

employing solid state ceramic method to study the

effect of different sintering temperature. Specimens

were sintered at two (850°C and 925°C) different

temperatures. In this paper, X-Ray Diffraction (XRD),

Scanning Electron Microscopy (SEM), Dielectric

properties and D.C resistivity are reported. XRD

analysis implies that three phases were obtained for

sample which was sintered at 850°C and 925°C. One

phase (NdFeO3) is found at both samples.

Bi1.34Fe0.66Nb1.34O6.35 is found for samples sintered at

850°C, whether Bi1.721 δ 0.089Fe 1.056Nb1.134O7 is found

for samples sintered at 925°C (Figure 1).

Figure 1. XRD Patterns of (Bi2O3Fe2O3)0.4(Nb2O5Nd2O3)0.6

SEM image of the samples suggested that grain size

increases with the increasing sintering temperature.

With the increase of frequency, dielectric constant and

loss tangent of the sample decreases (Figure 2).

Figure 2. Effect of frequency on dielectric constant for

(Bi2O3Fe2O3)0.4(Nb2O5Nd2O3)0.6

D.C resistivity of the sample was taken at different

temperature with different voltages. For samples

sintered at 850°C resistivity increases with the

increases of temperature and become maximum in the

temperature range of 150°C-300°C and beyond 300°C

it decreases. On the other hand, a drop of resistivity

was observed within the range of 200°C -250°C for

samples sintered at 925°C. After a drop at 200°C,

resistivity again rises and drops after 250°C and 350°C,

respectively.

NdFeO3 based materials are widely used for the

application of gas sensor, though it is CO and HCs

sensitive and also has high catalytic activities.

Paper ID: 29

41

Silicon Nanocrystals Rich Lanthanum Fluoride Films for Future Electronic Devices

Md. Ferdous Rahman*, Sk. Rashel Al Ahmed, Md. Golam Saklayen and Abu Bakar Md. Ismail

Department of Applied Physics & Electronic Engineering, University of Rajshahi, Rajshahi 6205, Bangladesh

*[email protected]

KEYWORDS: Silicon Nanocrystal, Nonvolatile Memory, Schottky Junction Solar Cell.

Investigation on Silicon nanocrystals (Si-NCs) rich Lanthanum Fluoride (LaF3) film fabricated using a novel one-step chemical method has been reported here. Application of this material has been tested for low-voltage operating non-volatile memory (NVM) and Schottky junction solar cells. Colloidal solution of Si-NCs in hydrofluoric acid (HF) was prepared from meso-porous silicon by ultrasonic vibration (sonication).On a silicon (Si) substrate LaCl3 solution in HCL is allowed to react with the colloidal solution of prepared Si-NCs. LaCl3 reacts with HF of Si-NCs solution and produces LaF3 crystals that deposits on the silicon substrate as a film embedding Si-NCs. This is a novel single step chemical way of depositing LaF3 insulating layer embedding Si-NCs (LaF3:Si-NCs). The XRD and EDX analysis of the deposited film show a polycrystalline and non-stoichiometric nature of LaF3. The presence of Si-NCs was confirmed by SEM and FTIR. The Al/LaF3/Si/Al structure was tested as NVM and a memory window of 525 mV was obtained at a programming and erasing bias of 2V when.The LaF3:Si-NCs films showed strong absorption.I-V characteristics of ITO/LaF3/Si/Al structure showed a dependency on the incident light intensity where current changed under various light illumination. Experimental results show a lot of promise of Si-NCs-rich LaF3 film to be used as an insulating film in non-volatile memory as well as a photoactive material in Schottkey junction solar cell. Recently researchers have been considering nanocrystal-based memory devices as a solution to ultra-large scale integration of electronic nonvolatile memories. One major barrier to such integration of NVMs is the local defect related leakage. Using discrete nanocrystals instead of the conventional continuous floating gate as charge storage nodes, local–defect-related leakage can be reduced efficiently to improve data retention. In this regard, discrete-trap type semiconductor storage materials such as Si nanocrystals (Si-NCs) embedded in a dielectric matrix have been demonstrated as potential candidates for the fabrication of high-speed, high-density, low power-consuming, and nonvolatile memories. Therefore, the poly-silicon–oxide–nitride–oxide–silicon (SONOS)-type structure memories including nanocrystal memories have recently attracted much attention for the application in the next-generation nonvolatile

memories because of their great potential for achieving high program/erase (P/E) speed, low programming voltage and low power performance. In this work our goal has been to create Si-NCs-rich LaF3 film to be used as an insulating film in non-volatile memory as well as a photoactive material in Schottkey junction solar cell.

Figure 1: Schematic diagram of a fabricated device..

After formation the colloidal suspensions of silicon nanocrystals in HF and then it will be allowed to react with Lanthanum chloride (LaCl3) solution in hydrochloric acid (HCl) at room temperature. The chemical reaction between the LaCl3 solution and colloidal solution of Si-NCs in HF can be given as, LaCl3 + HF → LaF3 + HCl The memory measurement has been performed by an impedance analyzer. Capacitance-voltage (C-V) study of the MIS [Al/LaF3/Si/Al]structure reveals that resonant tunneling of electron and charge storage was there when the MIS was biased from accumulation to inversion, which created a memory window. This type of memory window is called hysteresis. The MIS structure showed hysteresis for forward and reverse bias scan, enabling the structure to be used a non volatile memory. The Capacitance-Voltage (C-V) curves observed of the LaF3 layer deposited nonvolatile memory device for various fryquencies and various bias voltage. At a 1MHz frequency hysteresis voltage difference(memory window) (525mV) is so good. So, Finally observed that to study the C-V characteristics of MIS device, it was shown that a memory window of about 525 mV is achievable at a bias voltage of (-2V to +2V), indicating of the stucture used as a non-volatile memory device.

Paper ID: 30

42

Study on the Displacement Effect at Cylindrical Ionization Chamber with Different

Radii in High Energy Photon of Flat Beam and True Beams *Kumaresh Chandra Paul

1,3, Guenther H. Hartmann

2, Enamul Hoque

3, Golam Abu Zakaria

1,4

1Dept. of Medical Physics and Biomedical Engineering, Gono Bishwabidyalay, Dhaka, Bangladesh

2Dept. of Medical Physics in Radiation Oncology, German Cancer Research Center, Im Neuenheimer Feld 280,

69120 Heidelberg, Germany 3Dept. of Physics, Jahangirnagar University, Savar, Dhaka, Bangladesh

4Dept. of Medical Radiation Physics, Gummersbach Hospital, Academic Teaching Hospital of the University of

Cologne, Germany

* [email protected]

KEYWORDS: Cylindrical ionization chamber, displacement effect, true and flat photon beams.

Absorbed dose to water determination in the usual

clinical practice protocols require the placement of the

air-filled ionization chambers in the water phantom,

which introduces several fluence perturbations in high-

energy photon and electron beams. Displacement

perturbation is one of them, which is to be considered

in dosimetry. It is possible to correct the displacement

effect by introducing the chamber-specific quality

correction factor (kQ) or by introducing the concept of

effective point of measurement (EPOM). The EPOM is

the point in the chamber at which the measured dose

would be the same as in the measuring depth in the

absence of the radiation sensitive device or dosimeter

in the water phantom. The aim of this study is to

measure the displacement effect (shift of the EPOM

divided by the radius of the chamber) of cylindrical

ionization chambers by comparing of percentage depth

dose (PDD) in reference to the well established Roos

chamber.

Linear accelerator, specially designed six

cylindrical chambers with the radius of 1 to 6 mm, a

Semiflex ionization chamber, the Roos chamber, 3D

motorized water phantom and electrometers have been

used in the study. PDD was measured in the water

phantom following the International Atomic Energy

Agency (IAEA) TRS-398 protocol. The cylindrical

chambers were placed considering the central axis set

up on the water surface. Mirror imaging system was

followed to set up the cylindrical chambers. 100 cm

source to surface distance (SSD), 10 X 10 cm2 field

size and vertical radiation beams were used for the

study. The beam central axis measurements were

performed from the direction of bottom of the water

phantom to the surface of the water in 6 and 10 MV

true and flat photon beams. OmniPro 7.2 was used for

the dosimetric measurements. Microsoft Excel and

Sigma Plot-10 software were used for the calculation

and drawing the experimental graphs. ―Quick transfer‖

option of Sigma Plot was used to measure the

displacement up to 0.01 mm uncertainty for this

experimental study.

Results: The displacement shifts were found to vary

with beam energy and chamber radius. The

displacement effect obtained a range of 0.25 to 0.57

times r (radius of the chamber) both in true and flat

photon beams, not a constant value (0.6r). The average

experimental values were found lower than that of the

recommendation (0.6r) in the TRS-398 protocol.

The well established Roos chamber was carefully

set as -1.5 mm (minus sign indicates towards the

radiation source) from the front surface of the chamber

and was taken as the reference measurement depth for

the study.

The experimental displacement effect appeared to

be dependent on beam quality, the chamber cavity and

independent on depth after the dose maximum. A

recommended constant value of displacement

correction factor of the protocol does not satisfy with

this experimental study. Monte Carlo simulated study

can be the alternative solution for estimating the

displacement effect.

Paper ID: 41


43

Electrical and Optical Properties of Cu-Nanoparticles- Doped α-Fe2O3 Thin Film

Spin-Coated on Glass Substrate

Sanjida Ferdous*, Afroza Yasmin, Jinia Sultana and Abu Bakar Md. Ismail

Department of Applied Physics and Electronic Engineering, University of Rajshahi,

Rajshahi 6205, Bangladesh

* [email protected]

KEYWORDS: Oxide semiconductor, Photo-catalysis, Copper nanoparticles, Optical absorption.

Iron Oxide (α-Fe2O3) is a most promising earth-

abundant material to be used as catalyst in solar water

splitting that produces Hydrogen. It has a low band gap

of 2 ~ 2.2 eV that supports photo-catalysis. It can

absorb all UV light and a portion of visible light from

360 to 600 nm. But the biggest problems are its low

electrical conductivity and poor absorption in higher

visible solar radiation. Researchers have tried various

approaches to overcome those shortcomings of Fe2O3.

In this background to find a low-cost and simple

solution Copper nanoparticles-doped Iron oxide (Cu-

NPs:Fe2O3) has been investigated in this presentation.

Cu-NPs were mixed with Fe2O3 and the performance of

this photocatalyst was investigated and compared with

Fe2O3 without Cu-NPs. Thin films of CuNPs:Fe2O3 on

glass substrate have been prepared using spin coating

method. Synthesis of iron-oxide was performed from

Iron-chloride (FeCl2) and Urea (NH2-CO-NH2)

solution where ethanol used as solvent. Synthesis of

Cu-NPs was done by centrifuging aqueous solution of

copper sulphate (CuSO4) with sodium borohydried and

hydrazine hydrate used as reducer of Cu2+

ions at

12000 rpm. Synthesised Cu-NPs were immediately

capped by SDS (Sodium Dodecyl Sulphate) to prevent

further growth. The presence of Cu-NPs inside the

prepared Iron-oxide was confirmed by FTIR analysis.

Three and five layers of Cu-NPs:Fe2O3 were deposited

having three different concentrations of Cu-NPs. The

concentration of Cu-NPs was changed by changing the

concentration of CuSO4. Postdeposition annealing

treatments of Cu-NPs:Fe2O3 films were carried out at

500oC. The effects of the chemical composition on the

optical and electrical properties of the thin films were

investigated. The best film characteristics were

achieved using a mass ratio of 11.24% FeCl2 and

0.12% CuSO4. The optimized conditions resulted in a

thin film with average absorbance of 1.12, optical

bandgap of 1.67eV and resistivity of 177.911 Ω-cm.

The absorbance and the resistivity are increasing with

the increased concentration of CuSO4.

300 400 500 600 700 800 900 1000 1100

0.0

0.2

0.4

0.6

0.8

1.0

1.2

L(1) .005M

L(2) .0075M

L(3) .01M

Ab

so

rban

ce (

a.u

)

Wavelength (nm)

Figure 1. Absorbance of Cu-NPs:Fe2O3 for various

concentration of CuSO4.

0.005 0.006 0.007 0.008 0.009 0.010

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

2.0

Resis

tivit

y (

oh

m-m

)

Concentration(M)

Figure 2. Variation of resistivity with various concentration

of CuSO4.

Experimental results show that the electrical and

optical properties of Fe2O3 can be enhanced by doping

Fe2O3 with Cu-NPs that will enable Fe2O3 to be used in

efficient photo-catalysis.

Paper ID: 42

44

Study on Morphological Properties of Cu-NPs Doped α-Fe2O3

Thin Film Deposited on Glass Substrate

Jinia Sultana*, Afroza Yasmin, Sanjida Ferdous and Abu Bakar Md. Ismail

Department of Applied Physics and Electronic Engineering, University of Rajshahi,


* [email protected]

KEYWORDS: Photo-catalysis, Oxide semiconductor, Copper Nanoparticles, Surface morphology.

Iron-oxide (α-Fe2O3) is desired for the fabrication of homo-junction photo-anode, which can perform photo-electrochemical conversion of solar energy into hydrogen fuels by the solar splitting of water. It has a low band gap of 2 ~ 2.2 eV that supports photo-catalysis. But the biggest problemsare its low electrical conductivity and poor absorption in higher visible solar radiation. In this background Copper nanoparticles (Cu-NPs) doped α-Fe2O3 thin films (Cu-NPs:Fe2O3) have been investigated for better catalytic effect. This article reports the structural and morphological studies of Cu-NPs:Fe2O3.The Cu-NPs was prepared in aqueous solution by reducing Cu

2+ ions with Sodium

borohydred and Hydrazine hydred. Synthesis of iron-oxide was performed from Iron-chloride (FeCl2) and Urea (NH2-CO-NH2) solution where ethanol used as solvent. Extraction of the Cu-NPs was done by centrifuging at 12000 rpm of the aqueous solution of copper sulfate (CuSO4) with sodium borohydried and hydrazine hydrate used as reducer of Cu

2+ ions. The

metallic copper produced was immediately capped by SDS (Sodium Dodecyl Sulfate) to prevent further growth. After that, three and five layers of Cu-NPs:Fe2O3films were prepared by simplest spin coating methods, then annealed the films at 500

oC. The

presence of Cu-NPs in the prepared Fe2O3 films was confirmed by FTIR analysis (Figure 2). In the FTIR spectra, the absorption band at around 611.81cm

-1is

attributed to the asymmetric stretch mode of the Fe-O bond. It is also seen that absorption band at 475.18 cm

-1

and at 399.16 cm-1

, of the stretching mode of Fe-O bonds which proves that there is significant amount of Fe2O3 in the sample. The structural (grain size and roughness) properties of the thin films were investigated by AFM (Figure 1). The average surface roughness was found to increase with the thickness of the deposited films and concentration of Cu nanoparticles. The mean grain size was started decreasing with increasing layers for the same concentration and it was increasing with the film thickness and with increasing concentration of copper nano-particles (Figure 3).

Figure 1. AFM image of 0.005M molar concentration of

3-layer Cu-NPs:Fe2O3 film.

Figure 2. FTIR transmittance spectra of Fe2O3 with

embedded Cu-NPs sample.

0.006 0.008 0.0104.4

4.8

5.2

5.6

Gra

in s

ize

(nm

)

Concentration (1/cm3) Figure 3. The variation of grain size with molar

concentration of Cu-NPs for 3Layer Cu-NPs:Fe2O3 film.

Experimental results show that doping Fe2O3 with Cu-

NPs enhances the morphological properties of Fe2O3 that

will enable Fe2O3 to be used in efficient photo-catalysis.

Paper ID: 43

45

Study on the Content Based Image Retrieval (CBIR) System Based on Histogram

Technique

Rebeka Sultana1, Shamim Ahmad

2, and Md. Khalid Hossain

3*

1,2Department of Computer Science & Engineering, University of Rajshahi, Rajshahi 6205, Bangladesh

3Institute of Electronics, Atomic Energy Research Establishment, Savar, Dhaka, Bangladesh

* [email protected]

KEYWORDS: Content Based Image Retrieval, Histogram, Taxicab distance, Correlation, Arithmetic mean,

Geometric mean.

Content Based Image Retrieval (CBIR) has been an

active research area in image processing field where

low-level image features such as color, texture and

shape are extensively used for retrieving similar images

from a large image database. Currently the research

trend of CBIR is to reduce the semantic gap between

low level visual features and high level image

semantics. But this paper focuses only on the primary

feature that is color of the image which extraction

technique is not time consuming at all compare to other

features.

In this paper, histogram is used for extracting the

color feature from images. As histogram is

computationally less expensive, so our intention is to

retrieve similar images from large image database

using only color feature because the results of this first

filtering can be fed to another image retrieval which

may use some other effective features such as shape,

texture, color moment and so on. As a result, small

number of retrieved images found from this first

filtering will provide less time for the second retrieval

system for further processing.

The goal of this research is to study the CBIR

system where histogram technique is used as a way of

extracting color feature from images for both gray and

color of each image which is used then as a main

component for the similarity evaluation between

images. The conventional taxicab distance (Manhattan

distance metric) and correlation technique are

considered to evaluate the similarity between query

image and each image from database. Also three

approaches are used in this paper for comparing the

results with each other. Basically color image has three

color channels: red, green, and blue. For this reason the

first approach is to make color histograms for red,

green, blue, and grayscale separately. Similarity

measurement techniques perform on each histogram

and finally sum up the results of red, green, and blue

histogram where the second approach gives only

separate result for each color histogram. The strategy

of final approach is to make histogram for several

blocks of each image then performs the similarity

measurement techniques block by block between

images and finally arithmetic and geometric mean

operations are performed on all measurement values of

each image. The basic diagram of this histogram based

image retrieval is given below:

Figure 1: Histogram Based Image Retrieval

This experiment provides the best response based on

Manhattan distance with geometric mean from human

perspective point of view in comparison with the

response based on Manhattan distance with arithmetic

mean and correlation technique with arithmetic mean

operation where image blocks are used. And it is hoped

that it can be a good solution for existing image

retrieval system to reduce the computation time. As

image has different types of structure or formulation,

this experiment had some cases where retrieved images

are not so similar compare to query image though the

internal procedure was correct. So, in the future our

expectation is to extend this work for getting more

relevant response by using some other features such as

number of horizontal or vertical edges in an image.

Query

Image

Extract Color

Database

Images

Extract Color

Similarity

Measuremen

t

Retrieved

Images

Paper ID: 45

46

MRI Segmentation using Fuzzy C-Means Clustering and Bidimensional Empirical

Mode Decomposition

Gulam Sarwar Chuwdhury1, Md. Khaliluzzaman

1* and Md. Rashed-Al-Mahfuz

2

1Dept. of CSE, International Islamic University Chittagong (IIUC), Bangladesh

2Dept. of CSE, University of Rajshahi (RU), Bangladesh

* [email protected]

KEYWORDS: Image segmentation, fuzzy C-means, magnetic resonance imaging, wavelet, BEMD, SNR.

Image segmentation is a vital step in medical image processing. Magnetic resonance imaging (MRI) is used for brain tissues extraction in white and gray matter. These tissues extraction help in image segmentation applications such as radiotherapy planning, clinical diagnosis, treatment planning. This paper presents utilization of fuzzy C-means (FCM) clustering by using wavelet and bidimensional empirical mode decomposition (BEMD).The signal to noise ratio (SNR) value is calculated from previous clustering date to examine the best segmentation technique. The experiment with synthetic Brain Web images has demonstrated the efficiency and robustness of the appropriate approach in segmenting noisy medical MRI.

The primitive techniques are based on manual segmentation which is a time consuming process besides being susceptible to human errors. This paper has compared two image decomposition techniques i.e. wavelet and BEMD to determine the best technique which is used before the image clustering technique FCM. For that purpose, first apply FCM on decomposition image then select the best segmented technique from the two segmented approach through the value of SNR. The proposed method is an efficient approach to segment the noisy MRI brain images. Two major stages are involved in proposed methodology i.e. feature extraction and clustering. Feature extraction process is performed by using 2D wavelet decomposition and BEMD. The wavelet decomposition outputs are low pass (approximation component) and high pass (detailed components) at horizontal, vertical, diagonal. To obtain the wavelet features, here Dubechies-1(DAUB1) wavelet is applied to the image. Feature extraction from wavelet decomposition and BEMD are given to Fuzzy C-Means, FCM applied on the feature vector obtained from previous step for clustering. The output image will be segmented into two classes (White Matter, Gray matter) noisy MRI image. Then calculate the SNR for segmented outputs. Finally, provide the output which gives the best SNR for noisy MRI image. The workflow of the proposed method is as shown in Fig. 1.The proposed method is implemented in MATLAB environment, and tested on MRI brain web database.

Figure1. Workflow of the proposed method.

The brain web images are simulated MR images generated by

the Brain Web simulator with different level of noise 0%,

1%, 3%, 5%, 7%, 9% and with different level of INU 0%,

20% and 40%. These images are obtained from Brain Web

Database at the McConnell Brain Imaging Centre of the

Montreal Neurological Institute, McGill University. An

example of images is as shown in Fig. 2(a), 2(b) and 2(c).

(a) (b) (c)

Figure 2. (a), (b) and (c) are T1 simulated brain web images

The evaluation of the segmentation performance in this

paper is measured by

M

x

N

y

M

x

N

y

yxs

yxfSNR

1 1

2

1 1

2

),(

),( (1)

where the input is image and is the

segmented image.

Table1: SNR of the Input Images

Sample

MRI

Input MRI with Smooth &

Adding Gaussian noise

SNR

Wavelet

& FCM

BEMD

& FCM

a

Smooth 1.0175 1.0977

mean= 0 and variance

=0.025 0.7749 0.7789

b

Smooth 0.4249 0.4259

mean=0 and variance

=0.025 0.6588 0.6654

c

Smooth 0.3992 0.3997

mean= 0 and variance

=0.025 0.6452 0.6472

The SNR of the segmented image generated from BEMD and

FCM gives the higher value for noisy MR Image. So, it

should be noted that BEMD decomposition approach is better

than the wavelet algorithm for the use of decomposition

before applying the segmentation algorithm such as FCM. It

is granted that smooth image always provide higher SNR

than noisy images.

Paper ID: 47

47

Wear and Morphological Behavior of Electron Beam Dose Irradiated

Polyoxymethylene Copolymer (POM-C)

Md. Shahinur Rahman1, Heon- Ju Lee

1*, Muhammad Sifatul Alam Chowdhury

2 and Konstantin Lyakhov

1

1Nuclear and Energy Engineering, Jeju National University, Jeju 690-756, South Korea

2Electrial and Electronic Engineering, IIUC, Chittagong, Bangladesh

* [email protected]

KEYWORDS: Electron beam, Optical microscope, Raman spectroscopy, SEM, Wear.

Polyoxymethylene copolymer (POM-C) is the most

prominent engineering thermoplastic consisting of

repeating carbon-oxygen bonds in the form of

oxymethylene groups (OCH2). It is widely used to

make small gear wheels, ball bearings, precision parts,

automotive and consumer electronics. In this study, the

POM-C round blocks were irradiated with 165 KeV

electron beam energy in five doses (100, 200, 300, 500

and 700 kGy) in vacuum condition at room

temperature. The wear and morphological properties of

electron beam dose irradiated POM-C have been

analyzed using optical microscopy, Raman

spectroscopy and scanning electron microscopy

(SEM). The electron beam irradiation transferred the

wear of unirradiated POM-C sample from the abrasive

wear (plough and cracks), adhesive wear

(grooving/striation, micropitting) and scraping to mild

scraping and striation for the 100 kGy dose irradiated

POM-C sample due to cross-linking (macroscopic

networks), chemical free radicals formations and

partial physical modification (smoothness), which can

be concluded from optical microscopic, SEM and

Raman spectroscopic observations. The degree of

improvement for tribological attribute relies on the

electron beam irradiation fluency conditions (energy

and dose rate).

Figure 1. Optical microscopic images of the worn

surfaces at sliding tracks on POM-C surfaces with x20

resolution: (a) unirradiated POM-C; (b) 100 kGy; (c)

200 kGy; (d) 300 kGy; (e) 500 kGy and (f) 700 kGy

electron beam dose irradiated surfaces. Scale bar 50

µm in each case.

Paper ID: 48

48

Study of Structural and Optical Properties of Pyrolised CuO Films

M. Majhar1, S. Ahmed

2, M. Mozibur Rahman

2 and M. K. R. Khan

2*

1Department of Materials Science and Engineering, Rajshahi University, Rajshahi-6205, Bangladesh.

2Department of Physics, Rajshahi University, Rajshahi -6205, Bangladesh.

* [email protected]

KEYWORDS: Spray pyrolysis technique, Cupric oxide, Thin film, and Band gap energy.

Cupric oxide (CuO) thin films were fabricated onto

glass substrates at temperature 350 °C using spray

pyrolysis technique. The source material was the

monohydrate cupric acetate, [(CH3COO)2Cu. H2O] with

different (0.1M, 0.2M and 0.3 M) molar concentrations.

Structural, surface morphology and optical properties of

deposited films were studied by XRD, AFM and UV-VIS

spectrophotometer. The films are monophasic polycry-

stalline in nature. The preferred orientation of the

deposited films was found to be (111) plane and the

crystal structure were identified as monoclinic. The films

were found to be a good absorber of visible EM radiation.

The optical absorbance is ~96% in the range (400-700)

nm after which it decreases with increasing wave length.

The optical band gap of the deposited film was found to

be dependent on film thickness which varies from 1.70-

1.85 eV.

Fig. 1: XRD patterns of CuO films with different molar

concentrations.

0

5

10

15

20

25

30

35

40

400 500 600 700 800 900 1000 1100

t=110 nm

t=170 nm

t=215 nm

Tra

ns

mit

tan

ce, T

(%

)

Wavelength, (nm) Fig. 2: Variation of Transmittance with wavelength.

0

1

2

3

4

5

6

7

8

1 1.2 1.4 1.6 1.8 2 2.2

t=110 nmt=170 nmt=215 nm

hcm

-1 e

V)2

x 1

01

5

Photon energy, h(eV) Fig. 3: Variation of (αhν)2 with photon energy for CuO films.

The thin films were prepared by spray pyrolysis

technique. An aqueous solution of 0.1 M, 0.2M and 0.3 M

[(CH3COO)2Cu.H2O] were used as precursor. The CuO

film was deposited on glass substrate at substrate

temperature ~350ºC.

The XRD patterns of the deposited films are shown in

Fig.1 which confirms that the CuO films are

polycrystalline in nature and could be indexed with

monoclinic structure. The characteristic peaks were

identified comparing with the JCPDS card and these are at

2θ = 35.6020°, 38.7892°, and 53.5223° corresponding to

(hkl) values (111-

), (111), and (020) respectively. All

peak positions are compared with the standard values of

cupric oxide (from JCPDS card). No peak corresponding

to Cu2O phase was found in the XRD pattern that

indicates the formation of pure phase of CuO films.

The optical transmittance as a function of wavelength of

incident radiation for CuO films is shown in Fig.2.The

optical band gap of the deposited film was found to be

dependent on film thickness which varies from 1.70-1.85

eV. A plot of (αhν)2 vs. hν for CuO films of different

thicknesses are shown in Fig.3. The direct band gap of

CuO films have been obtained from the intercepts on the

energy axis after extrapolation of the straight-line of

(αhν)2 vs. hν curves. Depending on the film thickness (110

nm, 170 nm and 215 nm) of the CuO films, the direct

band gap is found to be 1.8 eV, 1.72 eV and 1.72 eV

respectively.

Paper ID: 49

49

Algorithm for Performance Appraisal using CAW Method

M. Z. Ahsan1 and Md. Mamun-ur-Rashid Khandker

2*

1Military Institute of Science and Technology, Mirpur Cantonment, Dhaka, Bangladesh

2Department of Applied Physics and Electronic Engineering, University of Rajshahi, Bangladesh

* [email protected]

KEYWORDS: Performance Score (PS), Overall Performance Index (OPI), Aptitude Index (API), Reflective factors.

Performance appraisal is an HR process, which plays

important role to drive employees for keeping them

aligned with the organizational goal and desired outcome.

Simple Additive Weighing (SAW) method is popularly

used as a tool to calculate individual performance score

(PS). Reflective factors (opportunity and optimum

contribution) are not accounted in this method.

Consequently, the calculated PS found to be less

representative and biased. To address these issues, a new

mathematical model, called Cumulative Average

Weighing Method (CAW) has been developed, where

reflective factors accounted. Three variables (aptitude,

attitude and ability) have been postulated to derive PS

formula.

The formula derived as

)}], where is for

overall average of aptitude, for overall average of

attitude and for overall average of ability score, and

for performance score of immediate previous or last

year, n for number of performance appraisal done in the

current year, and for the performance score at

quiescent year. It is worth mentioning here that the

opportunity factor accounted as

within close limit [01], where is opportunity factor,

determined from the logical weighting (1 or 0) to the

attributes of opportunity, and suffix j is for the number of

assessor. Each variable evaluated by weighted values of

corresponding set of attributes.

This formula realized by software, named Computer

Based Personal Appraisal System (CB-PAS), which has

been developed in Visual Basic at the front and MySql at

the back end in bottom up approach from its tree

algorithm. This software has the options to calculate PS

by both the methods.

A case has been studied in small scale for verifying

representativeness of PS and to supplement the

proposition in this paper. In that three (X, Y and Z)

employees of same position and having almost same

length of service are considered to evaluate PS at the

quiescent year. 35 attributes (10 x personal traits and 25 x

demonstrated traits) have been weighted using 9-point

likert scale against each employee over 05 (2010 to 2014)

years. The analysis, using statistical tools (SD, MAD and

AD) reflects that the PS as calculated by CAW method is

more representative than that of calculated by the SAW

method. The findings, based on SD, are presented in Table

1.

Table-1: Study Result on SD

Employee SD in

CAW

Method

SD in

SAW

Method

Remark

X 0.109 0.139 0 < 0.109 < 0.139

Y 0.115 0.236 0 < 0.115 < 0.236

Z 0.150 0.177 0 < 0.150 < 0.177

Besides, Graphical Model for Score Interpretation

(GMSI) is used as an alternative tool for screening and

selecting the best option using overall performance index

(OPI) and aptitude index (API) and to calculate the rate of

change of aptitude from the slope of the fitted trend line of

API that facilitate the organization to talent management.

The formula used for OPI and API are and

respectively where m is the number of year. Bar

chart of OPI and API is shown in figure-1 to visualize

how to use GMSI for making decision on promotion,

retention etc. However, the postulated variables and

corresponding attributes may not be the same for all

categories, rather depend on the perspectives and

objectives of the organization. As such, flexibility of

choosing variables and attributes remains on hands of the

organization

Figure 1. Bar chart of OPI and API.

Paper ID: 52


50

Dynamic Power Management by Reinforcement Learning

Safayet Hossain, Muhammad Adnan Ibn-Ismail and Muhidul Islam Khan*

School of Engineering and Computer Science

BRAC University, Dhaka, Bangladesh

* [email protected]

KEYWORDS: Dynamic power management, Reinforcement learning, Q learning and SARSA.

Dynamic power management indicates the system

strategies that achieve efficient power consumption by

selectivity turning off when they are idle or busy with

work load. In this paper we present two Reinforcement

learning (RL) based methods on Dynamic power

Management (DPM) for a Network Interface Card

(NIC). We implement Q learning and State Action

Reward State Action (SARSA) in a NIC to find out the

dynamically adaptive behavior according to the

workload.

Research has been devoted to explore different

DPM policies. The existing DPM policies can be

broadly classified into heuristic and stochastic policies.

While the heuristic policies are simple to implement

but provide no guaranties on the power/performance

trade-off. A simple heuristic policy where a device is

put to sleep when it is idle more than a specified

timeout period might be fixed or adopted. For instance,

the device is put to sleep if it is idle or more than break

even time. Break even time is the minimum period for

the device should be turned off so that compensates for

the overhead associated with shutting down and turning

on the device and saves energy compared to the case if

it is remains on.

For Q learning and SARSA learning methods, we

consider a model based on power management system.

We consider a model for the workload of the

considered NIC. We consider a set of states based on

application specific variables and a set of actions.

There is a cost function which helps NIC to learn the

appropriate actions and state transition. We consider

three states and three actions.

For Q and SARSA learning, Q value for each state

and action pair is initially chosen by designer and it

will be updated each time and action by the following

equation 1 t 1 t t( , ) Q(s, t) (s, t) [P minaQ(s ,a) Q(s,a)]tQ s t

Where Q(s,t) is Old value, (s, t) is Learning rate,

1P t is Penalty and is Discount factor.

We consider three actions named Go_Sleep, Go_Idle

and Go_Busy with the energy consumption 0, 0.7 and

0.9 W respectively.

Figure 1. Performance vs latency trade-off for Q earning.

Figure 2. Cumulative average cost over time slots for Q

learning and SARSA.

Figure 1 and 2 show the simulation results. Figure 1

shows that after applying Q learning in a NIC, it can

adaptively learn the performance/latency trade-off.

Figure 2 shows the comparison between Q learning and

SARSA learning for the cumulative average cost over

time slots. We observe that SARSA learning provides

more cumulative cost but converge faster than Q

learning. So, the NIC reaches an optimal stage of

power consumption.

Paper ID: 53


51

Friction and Morphological Properties of Ion Implanted Polyoxymethylene

Copolymer (POM-C)

Md. Shahinur Rahman1*

, Md. Mehedi Hasan2, Muhammad Sifatul Alam Chowdhury

2* and Konstantin Lyakhov

1

1Nuclear and Energy Engineering, Jeju National University, Jeju 690-756, South Korea

2Electrial and Electronic Engineering, IIUC, Chittagong, Bangladesh

* [email protected]

KEYWORDS: POM-C, SEM, Surface roughness, Friction coefficient, Raman spectroscopy.

Polyoxymethylene copolymer (POM-C) round

block was implanted with 120 KeV ions of He to doses

of 5 x 1016

and 1 x 1016

ions cm-2

. It was also implanted

with 120 KeV ions of Ar + He and He + Ne to dose of

1 x 1016

ions cm-2

, respectively. The friction behavior

of both implanted and unimplanted POM-C blocks was

investigated using a pin on disk tribometer against steel

ball. The morphological properties of ion beam

implanted POM-C blocks have been characterized by

scanning electron microscopy (SEM), 3 D nano surface

profiler and Raman spectroscopic methods. The

friction coefficient of He ion implantation at a dose of

5 x 1016

ions cm-2

is lower than that of another ion

doses implanted POM-C blocks and, it is also lower

than the unimplanted one. It also shows the moderate

surface texturing (atomic rearrangement), lower

surface roughness and good chemical structural

behavior compared to both unimplanted and other ion

doses implanted POM-C blocks due to some extent of

carbonization, cross-linking and ions-target atoms

collisions, which is ascertained from SEM, Raman

spectroscopic and 3 D nano surface profiler

observations. The other ion doses implanted POM-C

blocks demonstrate the higher friction coefficient and

surface roughness with polymer surface deformation

(crazing, cracking, pitting and gas evolution, bond

breaking) due to severe chain scission, surface dose

delivered atomic displacements and chemical structural

degradation.

It is concluded that the variation of friction

coefficient behavior of POM-C resulted from its

structural response for ion beam implantation on the

surface. In contrast, further investigation is required to

get the best tribological attributes at optimum ion beam

surface dose delivered.

Figure 1. SEM images on POM-C surfaces with x500

resolution: (a) unimplanted, (b) He 5 x 1016, (c) He 1 x 1016,

(d)Ar + He 1 x 1016 and (e) He + Ne 1 x 1016 ionscm-2

implanted surfaces. Scale bar 10 µm in each case.

Paper ID: 56

52

Analysis of Annual and Seasonal Precipitation Concentration Index of North-

Western Region of Bangladesh *Ahsan Habib Rasel, Md Monirul Islam and Mumnunul Keramat

Dept. of Applied Physics & Electronic Engineering, University of Rajshahi, Rajshahi-6205, Bangladesh*

[email protected]

KEYWORDS: Precipitation Concentration Index, Rainfall, Seasonal rainfall.

The Precipitation Concentration Index (PCI) of

north-western region of Bangladesh is computed. The

PCI is estimated on seasonal and annual

distributions, variations and trends. The seasonal

estimation were based on two seasons dry season

(November to April) and wet season (May to

October). PCI is analyzed at annual and seasonal scale

to identify the pattern of rainfall in the study area for

the period of 2000-2011.

The study regions are Rajshahi and Rangpur divisions

located in the north western part of Bangladesh and

extends from 230

80' N to 260 38' N latitude and

from 880 01' E to 89

070' E longitude.

The daily rainfall data for twelve years (2000-2011)

are collected from Bangladesh Meteorological

Department (BMD) of six stations. The nature of

data collected is depth in millimeter (mm).

In an attempt to define temporal aspects of the rainfall

distribution within a year, the Precipitation

Concentration Index (PCI) proposed by Oliver (1980)

is used. The PCI is estimated on an annual and seasonal

scale. The distribution of PCI is shown by map using

surfer mapping software.

Table 1. Precipitation Concentration Index (PCI)

classification by Oliver.

PCI value Distribution of precipitation

PCI < 10 Uniform precipitation distribution

11 - 15 Moderate precipitation concentration

16 - 20 Irregular distribution

PCI>20 Strongly irregular distribution

The table-2 shows that the rainfall is very much

seasonal. The contribution of wet seasonal rainfall is

more than (90%) and the dry seasonal rainfall is less

than (10%) for all six stations from table-2.

Table 2: Seasonal rainfall, monthly rainfall and PCI

distribution of six stations.

Station Wet

season

Dry

season

Highest

monthly

rain

fall

(%)

Annu

al

PCI (%)

of Rainf

all

PCI (%)

of Rain

fall

PCI

Syedpur 91.2 10.9 8.8 30.7 22.1 18.8

Rangpur 91.4 10.4 8.6 28.7 21.9 18

Dinajpur 93.5 10.4 6.5 29.3 9.9 18.5

Bogra 91.4 10.8 8.6 27.3 18.9 18.3

Rajshahi 92.9 10.4 7.1 26.4 18.8 18.1

Ishwardi 90.5 10.5 9.5 25.0 19.9 17.6

In wet season PCI value is 10.4 to 10.9 that mean

uniform precipitation distribution in wet season. In dry

season PCI value varies from 25.0 to 30.7 and compare

with table-1 it is conclude that strongly irregular

distribution of rainfall in dry period. Annual PCI varies

from 17.6 to 18.8 indicates irregular distribution of

rainfall of the study area.

Figure1 : PCI distribution of the study area.

From the PCI trend analysis two types of observation

are observed. In the northern region the trend is

increasing and in the southern region its decreasing. It

is also observed that the overall precipitation duration

is decreasing. The effect of this trend may have

negative effect on water availability to sustain the

present environmental condition of the study area.

Paper ID: 58

53

Assessing Driving Stress using Custom Built Real-Time Sensors

Nova Ahmed*, Saad Azmeen-Ur-Rahman, Rahat Jahangir Rony and Md. Tanvir Mushfique

ECE, North South University, Dhaka, Bangladesh

* [email protected]

KEYWORDS: Pulse Sensor, Stress detector, Low cost Sensing, Traffic Experiences in Dhaka.

In recent years, stress has become a large problem

that hampers well-being. We focus on the stress

generated at multimodal traffic system of Bangladesh

where vehicles of various speed share the same road

along with unruly traffic of Dhaka, Bangladesh, and

possible ways to detect stress using low-cost

technology.

From the traffic viewpoint, Dhaka has multimodal

traffic (heterogeneous) - vehicles with various speeds.

Various vehicle drivers differ in their level of

education and knowledge on traffic rules (Islam et al.,

2012) as well as roadside safety. Similar disparity in

knowledge among pedestrians make the situation more

complex. Poor road conditions across the city

(Mahmud et al., 2012) and bottlenecks due to

substandard planning exacerbate heavy traffic

situations. The combination of these factors puts the

driver in a unique position and we want to portray the

driver‘s level of stress or well-being (Gulian et al.,

1990). Most research work related to vehicle drivers

have been done predominantly in the context of drivers

in the developed nations. Driving in a developing

country is a highly different experience (Islam et al.,

2012) and our goal is to formally study the driving

experience from the perspective of drivers in

developing nations, starting with Dhaka.

We have developed a custom system (see Figure

1a) that measures pulse-rate in real-time to indicate the

level of discomfort in terms of pulse reading deviation.

The system exploits the basic mechanism of heart

beating - oxy-hemoglobin blood flows out throughout

the body and de-oxy-hemoglobin blood returns to the

heart, resulting in a rhythmic pulsating of the arteries.

The system incorporates an open hardware sensor

(Murphy, 2012) equipped with a 515 nm InGaN on

Sapphire LED that is clipped to the ears of the test

subject. As blood pumps through the blood vessel,

reflectivity changes, which is detected by a photo

sensor. An LCD displays the pulse in real-time and the

system logs the data. Our initial experiments have

indicated correlation among pulse data and road

condition that can directly or indirectly cause stress

among drivers.

Figure 1. Custom real-time pulse sensor and data-logger.

Figure 2. Pulse-rate during various driving conditions.

BPM is observed to rise during key events.

We have shown that certain events during driving

cause an elevation of pulse-rate, providing an

indication of increased stress (see Figure 1b). This

information will be useful to assess the well-being of

drivers, which is very important at personal level as

well as collective level. By making inferences from the

data, we can suggest strategies to improve well-being,

such as opening the window to let some air in or

switching on the radio. We can alleviate the number of

accidents, roadside problems and improve the quality

of life by improving well-being.

Paper ID: 60

54

Bitwise Template Fusion of Noisy Images for Enhanced IRIS Recognition System

Md. Sohel Ahammed* and Biprodip Pal

Rajshahi University of Engineering & Technology (RUET), Rajshahi, Bangladesh

* [email protected]

KEYWORDS: Iris Recognition, Noisy image, Template fusion.

Iris recognition is one of the reliable and accurate

biometric identification systems available today.

However, many results have been published under

favorable conditions. The work presented in this paper

involves developing an iris recognition system

considering noisy data (noisy images). The

performance of many existing system gradually falls

for the noisy images. An image can be noisy in

different ways like at the time of capturing, eyelids,

poor segmentation or any other ways.

Figure 1. Proposed iris recognition system.

In figure 1, the proposed iris recognition system was shown. First of all images were taken from database. Then median filter was used for noise (salt & pepper) removal. A few amount of noise may not be removed. Fusion based approach helped in this case. However, after segmentation and Feature encoding, several templates were found for a single person. Finally modified majority voting system (MVS) was used to generate a final template from several template of the same person by considering common feature. Then Hamming distance was applied for classification.

Figure 2. Proposed voting approach for template fusion.

In figure 2 the proposed voting approach was

shown. Final template was produced by considering the

probability of common feature of the same person.

Several MxN dimensional templates were considered

for voting system. Each pixel value of final template

was calculated by considering the bitwise probability

of having same value in each pixel of MxN templates

of that particular person. According to this modified

biometric template fusion strategy, a final template was

generated considering the uniqueness of those

templates of that particular person.

Table 1. Performance comparison between system without

fusion & proposed system.

Dataset System without fusion

Proposed system

Standard 98.16% 99.3%

Noisy 94.49% 96%

Table 1 shows that proposed system can perform

quite well in both of standard and noisy scenario. For

the determination of the performance of the system,

two databases of digitalized gray scale eye images

were used. CASIA-Iris-Interval V.3 and synthetic

addition of noise with CASIA-Iris-Interval V.3 were

used for the standard and noisy image database

respectively. Each data set of 100 standard eye images

and 100 noisy images (Synthetic added of noise) were

used for experiment; however, trivial hamming

distance was able to carry out an accuracy of 94.49%

on the other hand the proposed approach was able to

accomplish an accuracy of 96%. The performance

increases up-to 99.3% for the standard images (noise

free image).

Table 1 shows that proposed iris recognition system

performs fine in both of standard and non-standard

(noisy) database. The main feature of our proposed

IRIS recognition system was the enhancement of

performance for noisy image. Bitwise template based

fusion approach helped us to achieve this goal.

Paper ID: 62

55

Kinetics of Extraction of Ti(IV) from Sulfate Medium by Cyanex 302

Ranjit K. Biswas* and Aneek K. Karmakar


* [email protected]

KEYWORDS: Kinetics, extraction, Ti(IV), Cyanex 302, Lewis cell.

Titanium (IV), obtainable from leaching of ilmenite

(available in Beach Sand of the south-eastern part of

Bangladesh) can be extracted by bis-(2,4,4-

trimethylpentyl) monothiophosphinic acid (Cyanex

302) dissolved in kerosene containing 5% (v/v) hexn-1-

ol as de-emulsifier (Biswas and Karmakar, 2013). This

system has been investigated from the kinetic view

point in order to obtain an over-all rate equation,

thermodynamic parameters related to kinetics (Ea, S±,

H±, etc.) and to provide a mechanism for the

extraction process.

The kinetics of the considered system has been

investigated on measuring flux (rate/interfacial area)

variations using a constant interfacial area stirred cell

(Lewis cell) operated at 3 Hz. The typical log F vs. log

[Ti(IV)] and log F vs. -log (1+233×10-pH

) plots are

shown in Figure 1. Similarly, log F vs. log [H2A2](o)

and log F vs. -log (1+3.2[SO42-

]) plots are shown in

Figure 2. Based on these results, the empirical flux

equation, at 293 K, derived is: F (kmol/m2s) = 10

-

4.365[Ti(IV)] (1+233[H

+]

-1) [H2A2](o)

0.5 (1+3.2[SO4

2-])

-1.

Activation energy, Ea is measured to be 36-58 kJ/mol

depending on experimental condition (higher value in

lower pH region). The enthalpy and entropy changes

on activation (H± and S

±) are also determined.

The analysis of flux equation cited has been done,

at various regions of [H+] and [SO4

2-], to elucidate the

mechanism of extraction. The rate determining

chemical reaction step, irrespective of reaction

parameters, appears as TiO2+

+A- → TiOA

+; and this

step occurs via an SN2 mechanism. In this mechanism,

the attack of A- to the coordination sphere of

[TiO(H2O)n]2+

in forming the activated higher

coordinated [TiO(H2O)nA]+ is slower than the H2O-

elimination step from [TiO(H2O)nA]+ to form normal

coordinated [TiO(H2O)n-1A]+ and also the step involved

in 2nd

A- addition to [TiO(H2O)n-1A]

+.

Figure 1. Variation of flux with [Ti(IV)] and pH. [SO42-] =

0.1 mol/L, temp. = 293 K. (), pH = 2.0, [H2A2](o) = 0.10

mol/L; (), [Ti(IV)] = 1 g/L, [H2A2](o) = 0.10 mol/L.

Figure 2. Variation of flux with [H2A2](o) and SO4

2-. [Ti(IV)]

= 1 g/L, temp. =293 K. (), pH = 2; (), pH = 1.6, [H2A2](o)

= 0.10 mol/L.

-3.0 -2.8 -2.6 -2.4 -2.2 -2.0 -1.8 -1.6 -1.4 -1.2-8.0

-7.8

-7.6

-7.4

-7.2

-7.0

-6.8

-6.6

-6.4-1.4 -1.2 -1.0 -0.8 -0.6 -0.4 -0.2 0.0

log (

F,

km

ol/m

2s)

log {[Ti(IV)], mol/L

-log (1+233x10-pH

)

-1.6 -1.4 -1.2 -1.0 -0.8 -0.6 -0.4 -0.2-8.6

-8.4

-8.2

-8.0

-7.8

-7.6

-7.4

-7.2

-7.0-1.0 -0.8 -0.6 -0.4 -0.2 0.0

log (

F,

km

ol/m

2s)

log {[H2A

2](o)

, mol/L

-log (1+3.2 [SO2-

4])

Paper ID: 65

56

Autonomous Human Face Detection and Tracking System with Variant Poses, Blur

and Illumination

Md. Zweel Rana*, Monimul Islam and Mohiuddin Ahmad

Department of Electrical and Electronic Engineering (EEE)

Khulna University of Engineering and Technology (KUET), Khulna-9203

* [email protected]

KEYWORDS: Face detection, AdaBoost, pose, illumination, Blur, Arduino.

In recent year face detection and tracking system is

a technology in computer vision that determines the

movement of a face in arbitrary images. Many

computer applications use face detection algorithm to

detect and tracking faces. This paper approaches the

technique for real time variant human face detection

and tracking using a modified version of Viola-Jones

algorithm. However, the detection of a face in real time

environment having uneven illumination, pose and

non-uniform motion blur is quiet difficult. In this

paper, we propose three level approaches to detect

faces and tracking. The low, intermediate, and high

level processing involve the improvement of image

quality, feature extraction, pattern detection and further

improvement of the quality of the image. Then face

tracking for moving camera is performed via servo and

Arduino. This implementation is robust and efficient

to detect and track the face in case of non-uniform

motion, blur, illumination and variant pose. This paper

focuses on automatic face detection and tracking on

video streams for surveillance in public places. This

system can be used for human interaction robot and

security purpose.

The whole process is shown by the following flow

chart in Fig. 1. START

Scan Input Image

Removing BlurFace detect considering pose and illumination

Face Capture

Face detect considering pose and illumination

Face detect considering pose and illumination

Detect Face Position (x,y)

Initiate Servo

Set Servo Position

END

Scan Human

Face

Scan Human

Face

NoYES

NoYES

Figure 1: Flowchart of proposed model.

Real rime results are obtained by detecting several

faces in a single frame simultaneously. All result are

shown sequentially.

Figure 2: Face tracking in indoor or normal environment.

Figure 3: Face detection considering dark and bright illumination.

Figure 4: Face detection considering blurry image.

Figure 5: Face detection considering variant poses.

Table 1: Summary of face detection data.

Condition

No. of face used

Detected

Missed

Precision (%)

Normal 100 100 0 100

Illumination 100 97 03 97

Blur 100 98 02 98

Pose 100 98 02 98

Table I shows the performance of proposed work for

normal environment, light variance, blur and pose.

From these result, it is obvious that our work can

robustly detect human face considering various

parameters.

The performance of this system was enhanced by the

use of open source OpenCV. From the above discussed

result it can be said that this system can be used

robustly for both indoor and outdoor environment

considering various parameters.

Paper ID: 66

57

Electrochemical Corrosion Characterization of Artificially Aged Al-6Si-0.5Mg (-

1Cu) Alloys in Sodium Chloride Solution

Abul Hossain1*, M. A. Gafur

2, Fahmida Gulshan

3, and ASW Kurny

3

1MSTE Plant (KTL), Sylhet Gas Fields Ltd., Golapgonj, Sylhet-3160

2Pilot Plant & Process Development Centre, BCSIR, Dhaka

3Department of Materials and Metallurgical Engineering, BUET, Dhaka

*[email protected]

KEYWORDS: Al-Si alloy, potentiodynamic polarization, electrochemical impedance spectroscopy.

In this study, corrosion behavior of artificially

peakaged Al-6Si-0.5Mg (-1Cu) alloys in 0.1M NaCl

solution has been investigated using potentiodynamic

polarization and electrochemical impedance

spectroscopy (EIS) techniques. The potentiodynamic

polarization curves reveal that 1 wt%Cu content alloy

is less prone to corrosion than the Cu free alloy. The

EIS test results also showed that corrosion resistance or

charge transfer resistance (Rct) increases with the

addition of 1wt%Cu into Al-6Si-0.5Mg alloy. Higher

charge transfer resistance (Rct) was obtained with the

addition of 1wt%Cu and lower Rct value was for Cu

free Al-6Si-0.5Mg alloy. Due to addition of Cu and

thermal modification, the magnitude of open circuit

potential (OCP), corrosion potential (Ecorr) and pitting

corrosion potential (Epit) of Al-6Si-0.5Mg alloy in

NaCl solution were shifted to the more noble direction.

A) Impedance Measurements

B) Table I. Impedance test results

Alloy

Compositions

Rs(Ω) Rct(kΩ) Cp(µF) OCP

(V/SCE)

Al-6Si-0.5Mg 40.37 15.57 1.259 -0.8454

Al-6Si-0.5Mg-

1Cu

44.08 27.13 3.219 -0.6534

Figure 1. Electrical equivalent circuit used for fitting of the

impedance data of Al-6Si 0.5Mg(-1Cu) alloys in 0.1M NaCl

solution.

Figure 2. Nyquist plots for the peakaged Al-6Si-0.5Mg (-

1Cu) alloys in 0.1M NaCl solution. B) Potentiodynamic Polarization Measurements

Table II. Potentiodynamic polarization test results

Figure 3. Potentiodynamic polarization curves of aged Al-

6Si-0.5Mg (-1Cu) alloys in 0.1M NaCl solution. The EIS tests have shown that the addition of 1wt%Cu

into Al-6Si-0.5Mg alloy increase the corrosion

resistance in NaCl. The linear polarization and Tafel

extrapolation plot show that the corrosion current (Icorr)

and corrosion rate (mpy) decrease with the addition of

1wt%Cu into Al-6Si-0.5Mg alloy. The forms of

corrosion in the studied Al-6Si-0.5Mg (-1Cu) alloys

are pitting corrosion as obtained from the

microstructures study with pits observations.

Alloy

Compositions

Icorr

(µA)

Ecorr(mV) Epit

(mV)

Corrosion

rate(mpy)

Al-6Si-0.5Mg 6.300 -764 -480 5.287

Al-6Si-0.5Mg-

1Cu

2.950 -697 -370 2.474

Paper ID: 67

58

Effects of Inclusions on the Mechanical Properties of Structural Steel Reinforced

Bars

Abul Hossain1*, Fahmida Gulshan

2, and ASW Kurny

2

1MSTE Plant (KTL), Sylhet Gas Fields Ltd., Golapgonj, Sylhet-3160

2Department of Materials and Metallurgical Engineering, BUET, Dhaka

* [email protected]

KEYWORDS: Inclusions, reinforced bars, ladle refining, tensile properties.

Inclusions content of pencil ingots and continuously

cast billets (both ladle refined and unrefined) produced

from induction melted liquid steel have been

determined. It has been seen that billets produced from

unrefined or improperly refined melts contain higher

amounts of slag, inclusion and inhomogeneity in the

microstructure while refining in ladle refining furnace

of induction melted assorted scrap gives fairly clean

and refined liquid steel. Metallographic study of the

reinforcing bars produced from properly refined

continuously cast billets show uniform grain size, no

heterogeneity in the microstructure, little or no slag and

little inclusions and better mechanical properties.

Table1. Chemical compositions of the ingots/billets.

Heat

No

Casting

Type

%C %Mn Carbon

Equivalent

PI-1 Pencil

Ingot

0.34 1.10 0.52

PI-2 0.35 1.14 0.54

NB-1 Normal

Standard

Billet

0.32 1.15 0.51

NB-2 0.32 1.07 0.50

QB-1 Quality

Standard

Billet

0.33 1.06 0.51

QB-2 0.31 1.16 0.50

Tensile properties yield strength, ultimate tensile

strength, %elongation and %reduction in area of the

reinforced bars are presented in Fig. 2 and Fig. 3.Yield

strength and ultimate tensile strength of the reinforcing

bars produced from unrefined pencil ingot is lower

than normal and quality standard billets. The

reinforcing bars from quality standard billet shows

highest yield strength due to the lower number and

smaller sizes inclusions content in the steel bars.

All the reinforcing bars from pencil ingot show a

low value of %elongation and %reduction in area (Fig.

3). These may be due to the presence of slags and

larger size in higher number of non-metallic inclusions.

Figure 2. Variation of yield strength and ultimate tensile

strength of the reinforcing bars.

Figure. 3 Percentage of elongation and reduction in area of

the reinforcing bars.

The following conclusions can be drawn from this

study.

Pencil ingots contain different (higher in number

and larger in sized) amounts of inclusions and

give uncertain physical and mechanical

properties.

Normal standard billets were unrefined and

contain slags, inclusions and inhomogeneities in

structure. Final products from these billets give

inferior physical properties.

Quality standard billets produced through proper

refining give clean and refined liquid steels or a little

slags and inclusions. Finished products from these

billets have better mechanical properties.

Paper ID: 69

59

Utilizing Solar Energy in the Filling Stations of Bangladesh: Technical and

Economical Representation

Mohammad Jalal Uddin, Muhammad Sifatul Alam Chowdhury*, Md. Ridwanul Karim, Md. Arman Uddin and

Md. Bakiuzzaman

International Islamic University Chittagong, Chittagong, Bangladesh

KEYWORDS: Solar energy, Renewable energy, Photovoltaic panel, filling stations.

Bangladesh has a large number of fuel stations and

most of the fuel stations maintain large rooftop and

open space. This research is focused on increased use

of solar energy in the fuel stations of Bangladesh in a

convenient way. Some fuel stations of Chittagong city

including ‗M/s Alhas younus & Co. filling station‘ is

considered to carry on the research. This research

performed complete mathematical calculation and

economic representation for setting up photovoltaic

panel in the fuel stations. Mathematical design of the

complete system includes solar power station design,

solar panel design, battery estimation, inverter design,

charge controller design. Economical representation

includes cost calculation, income and detail calculation

of payback period.

Solar power station design:

Roof top area=167.22 m2

Total area=436.63 m2

We can set up 103 solar panel of 250wp on the roof top

area and 268 solar panel on the total area in this filling

station. But we are designing for the filling station that

power consumption. The filling station consumption

power is 50KWh in December. But we are calculating

for 65KWh in a month due to December is the winter

season.

Solar panel design:

Consumption power per day=2.17KWh

Energy to load (KWh/day) = KWp* Radiation *

System efficiency (p)

Or, 2.17= KWp*4.2*0.55

Or, KWp=0.939393KWp or 940Wp

We selected 250wp solar panel from market.

So number of solar panel=3.76≡4

Battery calculation:

Battery sizing watt hour rating:-

DOA* Energy to load (Wh) = Purchase capacity of

battery (Wh)* DOD

Or, 3*2170=Wh*0.7

Or, Wh=9300

Where capacity=Ah capacity*Terminal voltage

Or, 9300 =Ah capacity*12

Or, Ah capacity= 775

We selected 130Ah solar battery from market

So number of battery=5.96≡6

Economical representation:

Figure 1. Generation statistics.

Payback period:

Considering 1 kWh = 12 Taka

Total cost of the system =248642500Taka

Annual income = 23208876Taka

Payback period = cost /Annual income

= 248642500/23208876

=10.72years

Paper ID: 72

60

Conversion of Prawn Shell Waste into Value Added Products for Textile Finishes

Md. Mofakkharul Islam, Firoz Ahmed and Md. Ibrahim H. Mondal*

Polymer and Textile Research Lab, Department of Applied Chemistry and Chemical Engineering, Rajshahi

University, Rajshahi 6205, Bangladesh

*[email protected]

KEYWORDS: Prawn shell waste, Chitosan, Carboxymethyl chitosan.

In this paper, synthesis, characterization and

application of chitosan and its functional derivatives to

enhance the effectiveness of natural biopolymer for

intensified textile and other uses, for the eco-friendly

modified cotton fibres, to avoid chemical modifier has

been reported. The washed fibres were treated with

chitosan and its derivatives, and dyed with reactive

dyes. The treated fibreswere then tested for appearance,

tensile strength, absorbancy, wrinkle recovery and

dyeing properties. Chitosan was made from chitin,

which was obtained from processed prawn shell waste

by a chemical process involving demineralization,

deproteinization, de-colouration, and deacetylation.The

N-octyl chitosan derivative was obtained by reductive

amination of chitosan. Water soluble carboxymethyl

chitosan (CMCh) was prepared by reacting chitosan

with monochloroacetic acid in aqueous alkaline media.

The present study was undertaken to evaluate the

effects of chitosan and its derivatives as modifier on

cotton as well as dyeing behavior. The molecular

weight, degree of deacetylation and ash content of

prepared chitosan were 1,39,958 Da, 85% and 2.33%

respectively. The moisture content, water holding

capacity and total nitrogen content were above 10%,

450% and 6.5% respectively. CMCh had an average

degree of substitution 0.82, determined by titrimetric

analysis. FTIR spectra showed characteristic peaks of

carbonyl group at 1659 cm-1

, amino group at 1600cm-1

,

asymmetric stretching of C-H in the methyl group at

1520 cm-1

and carboxylic group at 1737cm-1

,

respectively. Thus the product of reductive amination

and carboxymethylation of chitosan was investigated

by FTIR spectroscopy and their formation were

confirmed by NMR. The thermal behavior of chitin,

chitosan, N-octyl chitosan, CMCh, and their treated

cotton were investigated by thermogravimetric

analysis. These analyses showed moderate thermal

stability of those treated fibres than untreated fibres.

Modification of cotton by sorption of

N-octyl chitosan and CMCh introduces new functional

groups on the fibre surface with chemical bonding

which assured by FTIR. Surface morphology and

crystallinity of the modified fibres were carried out by

SEM and XRD respectively. As the modified fibres

also showed good dyeability and colour fastness as

well as other properties, the chitosan derivatives as

textile modifier would be helpful to avoid synthetic

petroleum-based chemical modifier.

Paper ID: 74

61

Textile Performance of Functionalized Cotton Fibre with Silane Coupling Agents

Md. Khademul Islam, Md. Abdul Aziz and Md. Ibrahim H. Mondal*

Polymer and Textile Research Lab, Department of Applied Chemistry and Chemical Engineering,

Rajshahi University, Rajshahi 6205, Bangladesh

* [email protected]

KEYWORDS: Cotton fibre, Silane coupling agents, Textile performance.

Modification of cotton fibre was studied by

condensation polymeriza-tion with functionalized

silane coupling agents like vinyltrimethoxy-silane

(VTMS) and 3-Glycidoxy propyl-triethoxysilane

(GTS) in ethanol/water medium. The use of sodium

dodecyl sulfate as a surfactant in ethanol/water mixture

accelerate the dispersion of VTMS and GTS in

ethanol/water mixture which accelerates the hydrolysis,

condensa-tion as well as the fibre surfuce

functionalization. The optimized condition of

modification with VTMS and GTS was 600% and

400%, respectively in ethanol/water mixture (60:40)

containing surfactant by maintaining pH 3.5 at 400 C in

the fibre-liquor ratio of 1:40. The modified cotton fibre

was subjected to evaluate some of the properties like

swelling behavior in different solvents and moisture

absorption. It was observed that swelling behavior and

moisture absorption of modified cotton fibres were

decreased in polar solvents, whereas these increased in

nonpolar solvents.

Fourier transform Infrared spectroscopy was used

to identify the incorporation of silicon containing

species, Energy Disperse X-ray analysis determine the

quantities of atomic silicon which directly reflects it‘s

valence bond with organic moieties, Scanning electron

microscopy and Thermo gravimetric analysis were

used to investigate the surface morphology and thermal

behavior of the modified fibre respectively. The

modification of cotton fibre enhanced the tensile

properties, water repellency, wrinkle recovery and

flexibility due to the Si-O-Si bond in the chain of

trisilanol pendent branch attached through Si-O-C bond

between trisilanol and cellulose substrate.

Paper ID: 75

62

Synthesis and Characterization of Carboxymethyl Cellulose With Higher Yield

From Corn (Zea mays L.) Husk

Mst. Sarmina Yeasmin, Md. Abu Sayeed, Md. Abu Raihan Ali and

Md. Ibrahim H. Mondal*

Polymer and Textile Research Lab, Department of Applied Chemistry and Chemical Engineering

Rajshahi University, Rajshahi-6205, Bangladesh

* [email protected]

KEYWORDS: Corn husk, Carboxymethyl cellulose, Cellulose particle sizes.

Corn husk is a waste part of corn (Zea mays L)

obtained largely from corn harvesting area, rich source

of cellulose. 45.69% pure cellulose was extracted from

raw husk using concentrated caustic soda and acidified

sodium chlorite processes. Carboxymethyl cellulose

(CMC) was synthesized by etherification process from

extracted corn husk cellulose at different particle sizes

(1071, 340, 149, 100 and 74 µm) using 7.5 molL-1

NaOH and 12 molL-1

monochloroacetic acid (MCA) at

55°C for 3.5 h with aqueous ethanolic medium. There

were two reactions occurred simultaneously during

carboxymethyla-tion, the primary reaction which

produced pure CMC, and the side reaction that

produced undesired sodium glycolate and sodium

chloride. CMC was purified several times in a series of

alcohol washes and separations, and sodium glycolate

& sodium chloride contents in the purified CMC were

determined. The yield, degree of substitution (DS),

molecular weight, solubility, water holding capacity,

oil holding capacity and gel content of the prepared

CMC were determined.

The produced cellulose and CMC were identified by

FTIR and the crystallinity was determined by XRD

analysis. Surface morphology of the cellulose and

CMC were analyzed by Scanning Electron Microscopy

(SEM). The yield of CMC was found to be dependent

upon the size of the cellulose. The highest yield of

CMC, 2.40 g/g, obtained at cellulose particle size 74

µm. Solubility, molecular weight and DS increased

with decrease cellulose particle sizes. The purity of the

CMC was 99.98%, sodium chloride and sodium

glycolate contents in the purified CMC were 0.01%

and 0.01% respectively. Synthesized CMC showed

DS, 2.41; water holding capacity, 5.11 g/g; oil holding

capacity, 1.59 g/g; and gel content, 99.96%. The

decreased cellulose particle size allowed higher yields

as well as higher DS of CMC providing plenty of

opportunities for its multipurpose applications, such

foods, pharmaceuticals, cosmetics etc.

Paper ID: 76

63

Synthesis and characterization of hydrogels from cellulosic materials for Green

Absorbent Products

Md. Obaidul Haque, Md. Abu Sayeed and Md. Ibrahim H. Mondal

Polymer and textile Research Lab, Department of Applied chemistry and chemical Engineering,

Rajshahi University, Rajshahi-6205

* [email protected]

KEYWORDS: Cellulose, Hydrogels, Water absorption.

Hydrogels with biodegradable property is highly

appreciable and ultimate expectation of the scientists at

the present time. Copolymerization treatment of

cellulose and cellulose derivatives can play a vital role

in producing green hydrogels and improving the

absorption performance of many absorbent products.

Hydrogels are polymeric substances and due to

hydrophilic functional groups in their structure capable

of holding large amount of water compared to its body

mass. Crosslinking is one of the simplest reactions used

to improve the physical properties of cellulose and

cellulose derivatives. In this work an innovative

cellulose-based environment friendly hydrogel was

synthesized as an alternative to acrylate-based

synthetic hydrogels for personal care products and

other absorption purposes. The cellulose-based

hydrogels were prepared by free radical graft

copolymerization reaction of cotton with acrylic acid

and acrylamide using N,N-methylene-bis-acrylamide

as a crosslinker in the presence of potassium

persulphate (K2S2O8) as an initiator for this system.

The maximum water absorption capacity of the

prepared hydrogels were found 50 g/g in deionized

water. Preparation condition was optimized depending

on monomer concentration, temperature and

crosslinker concentration. pH dependency on water

absorbency, gel content and time for equilibrium water

absorption were also investigated. For structural

characterization FTIR spectroscopy, SEM analysis,

TGA and XRD techniques were performed. The result

reveals that prepared hydrogel could be used

effectively alternate to acrylate based synthetic

hydrogels in personal health care and also as absorbent

materials in the polluted water treatment.

Paper ID: 77

64

Study of Encryption Techniques for Multimedia

Md. Martuza Ahamad* and Md. Ibrahim Abdullah

Dept. of Computer Science & Engineering, Islamic University, Kushtia, Bangladesh * [email protected]

KEYWORDS: Multimedia, Encryption, Decryption, RSA, Blowfish, AES, XOR.

During recent years the telecommunication industry

has made tremendous progress in their development of

systems that offer more bandwidth to the end user.

Multimedia content (e.g., text, audio, video and still

images) is made available for user through

communication network [1]. It is important to secure

the multimedia content when it transfers from sender to

receiver. There are several encryption techniques. Each

algorithm has advantages and limitations. Usually

multimedia content file size is large. Thus encryption

algorithm should be time efficient.

In this work we study time efficiency of most used

encryption algorithms: Blowfish, AES, XOR and RSA

for two types of multimedia content: text and image.

We study the encryption and decryption time of these

algorithms for different types of file size. We have

developed Java simulator to find encrypt and decrypt

time. The computer is HP 4th

Gen. Probook 450 with

Intel® Core™ i5-4200M CPU @ 2.50 GHz, 4 GB

RAM with Ubuntu 14.04 LTS (Precise Pangolin) 64-

bit Operating System.

For our simulation we have used 23 samples of text file

of size 1KB to 1MB and 14 samples of BMP image of

size 10KB to 2200KB. When we develop our

simulation program we use 128bit key size for all of

algorithms. Figure 1 and 2 represent encryption and

decryption time for text. Figure 3 and 4 shows

encryption and decryption time for image.

Figure 1. File size vs Encryption time for text.

Figure 2. File size vs Decryption time for text.

Figure 3. File size vs Encryption time for image.

Figure 4. File size vs Decryption time for image.

In our study, AES is the best performed algorithm than

other most used algorithms and RSA is the poorest

performed algorithm. Blowfish and XOR has average

rate of performance. In AES for a certain range of data

the taken time is same. Blowfish is the second best

performed algorithm and its taken time is linearly

increased with load. In XOR, it perform nice when load

is small but when load is large its performance will be

poor.

Paper ID: 83

65

Influence of Deposition Temperature on the Deposition of SiO2 Films from Reaction

of Silicone Oil Vapor and Ozone Gas

Arifuzzaman Rajib1*

, Susumu Horita2, Atowar Rahman

3 and Abu Bakar Md. Ismail

3

1 Department of Applied Physics, Electronics and Communication Engineering, University of Bangabandhu Sheikh

Mujibur Rahman Science and Technology, Gopalganj, Bangladesh 2 School of Material Science, Japan Advanced Institute of Science and Technology (JAIST), Japan

3Department of Applied Physics and Electronic Engineering, University of Rajshahi


* [email protected]

KEYWORDS: Low-Temperature SiO2 deposition, APCVD, FT-IR spectroscopy.

Low temperature deposition of SiO2 film on a silicon

substrate is desired for high quality gate oxide films to

obtain high performance thin film transistors (TFTs).

This paper reports, the deposition of SiO2films on

silicon substrate by using chemical reaction of silicone

oil vapor and ozone gas at low temperature. An organic

solution as a catalyst at atmospheric pressure has been

used to enhance the deposition rate of SiO2.

The chemically cleaned substrates were loaded into

the chamber (Fig 1), and heated at a deposition

temperature. The silicone oil (SO) was also heated (50

ºC) and vaporized directly by bubbling with N2 gas

through a stainless tube (55 °C). The organic solution

was mixed with silicone oil vapor and N2 gas, and then

flown into the chamber. The ozone was generated by

the ozonizer from O2gasand then introduced into the

chamber together with the silicon oil vapor.

Figure 1. Schematic diagram of APCVD method.

The thickness of the as-deposited films was

measured by the laser ellipsometry. The average

deposition rate was found to be 13.2 and 3.7 nm per

minute with and without organic catalyst, respectively.

The chemical structures of the as-depositedSiO2 films

were studied by Fourier transform infrared (FT-IR)

spectroscopy. FT-IR spectra of the as-deposited films

are very much similar to those of SiO2 films found in

literature.

4000 3000 2000 1000

0.0

0.1

0.2

Absorbed CO2Si-OH(2)

Si-O(1)

Ab

sorb

an

ce (

arb

. u

nit

)

Wavenumber (cm-1)

Flow rate of nitrogen for oil FN2=0.35 slm,

Substrate temp. TS=200°C,

Flow rate of oxygen, FO2=0.5 slm,

Si-O(2)

Si-OH(1)

Absorbed H2O

5 Minutes

10 Minutes

15 Minutes

Deposition time

Figure 2. FT-IR spectra of deposited SiO2 films.

The deposition rate is increased with the increasing

the certain deposition temperature and then it started

decreasing with further increasing the deposition

temperature. Reduction of deposition rate after certain

value of deposition temperature can be interpreted by

the fact that the heat radiation enhances the formation

of SiO2 film in the space and hence reduce the

deposition rate of theSiO2 films.

150 180 210 240 2700

4

8

12

16

Dep

osi

tio

n r

ate

(n

m/m

in)

Deposition temperature (oC)

Deposition time = 10 min,

N2 flow rate for oil = 0.35 slm,

N2flow rate for org. soln.= 0.1 slm,

O2flow rate = 0.5 slm

Org. soln. Temp.

Torg= 34.1 ºC

Torg= 28.1 ºC

Torg=21.9 ºC

Figure 3. Deposition rate vs deposition temperature curve for

different organic solution concentration.

Experimental results show that the deposition

rate of SiO2 on Si substrate can be enhanced and

controlled by controlling the deposition temperature

and catalyst solution temperature.

Paper ID: 84

66

An Improved Representation of Audio Signal in Time-Frequency Plane

Kazi Mahmudul Hassan1*, Md. Ekramul Hamid

2, and Takayoshi Nakai

3

1Dept. of Computer Science & Engineering, Varendra University, Rajshahi, Bangladesh

2 Dept. of Computer Science & Engineering, University of Rajshahi, Rajshahi, Bangladesh

3 Dept. of Electric & Electronic Engineering, Shizuoka University, Hamamatsu-shi, Japan

* [email protected]

KEYWORDS: Time-frequency representation, Hilbert spectrum, 2T-EMD, Filterbank, Fractional Gaussian noise

To analyze non-stationary signal like audio, time-

frequency representation is an important aspect. In case

of representing audio signal in time-frequency-energy

distribution, hilbert spectrum is a popular way which

has several advantages than other methods like STFT,

WT etc. Hilbert-Huang Transform is a prominent

method consists of both Empirical Mode

Decomposition (EMD) and Hilbert Spectral Analysis

(HSA). An enhancement of EMD called Turning

Tangent empirical mode decomposition (2T-EMD) has

recently developed to overcome some limitations like

cubic spline problems, sifting stopping condition etc.

2T-EMD based hilbert spectrum of audio signal

encountered some issues due to the generation of too

many IMFs in the process. In this work, a mutual

implementation of 2TEMD & classical EMD is

proposed which enhances hilbert spectrum

representation of audio signals. This refinement of

hilbert spectrum not only contributes to the future work

of source separation problem but also many other

applications in audio signal processing.

The algorithm of the proposed method is given

below:

St. 0: Choose a threshold point based on filter bank

property of fGn for selecting number of IMFs

using 2T-EMD method.

St. 1: Decompose the signal with 2T-EMD method

to get IMFs set and a residue signal.

St. 2: Decompose residue signal with classical EMD

method and get new IMFs set.

St. 3: Concatenate the 2T-EMD IMFs set with EMD

IMFs set to make the final set of IMFs.

St. 4: Hilbert Transform the IMFs and generates

final Hilbert Spectrum.

A block diagram and outcome of proposed method

applied on a sample audio signal (Speech & Flute

mixture) is shown here.

Figure1: Proposed method applied on a sample signal.

When classical EMD decomposes an audio signal, it

produces less number of IMFs that we required for

successful decomposition. Whereas though 2T-EMD is

an improved version of EMD, due to the new mean

definition and other improvements, it produces many

redundant IMFs. Because of this, the hilbert spectrum

representation of an audio signal becomes quite

meaningless especially in lower frequency region. Here

a method is proposed which is a combined

implementation of 2T-EMD and EMD method that

takes first few IMFs from 2T-EMD and the rest of the

IMFs are from EMD. The threshold point i.e. the

number of IMFs we take from 2T-EMD is selected

based on the satisfaction of Filter Bank property by 2T-

EMD method. Also the reconstruction error rate of

sample audio signal is in 10-16

range which is quite

acceptable in most of the application.

The enhancement of the representation of audio

signal mixture in time-frequency-energy distribution

can contribute significantly in case of source separation

and many other applications.

Paper ID: 85

67

On the Optimization of Number of Message Copies for Multi-Copy Routing

Protocols in Scalable Delay-Tolerant Networks

Md. Sharif Hossen and Muhammad Sajjadur Rahim*

Department of Information and Communication Engineering, University of Rajshahi, Rajshahi-6205, Bangladesh

* [email protected]

KEYWORDS: Delay-tolerant networks, Routing, Simulation, Performance metrics, ONE simulator.

Delay-Tolerant Networks (DTNs), also referred to as

Intermittently Connected Mobile Networks (ICMNs),

are kinds of mobile ad-hoc networks, where there is no

persistent route from source to destination. Unlike

Binary Spray-and-Wait (SNW) routing, where in wait

phase when a node has only a single copy which

cannot be forwarded to any node except for waiting to

direct transmission to destination, in focus phase of

Spray-and-Focus (SNF) routing, this single copy can

be forwarded to a more appropriate relay using single-

copy utility-based routing scheme as follows:

Let every node i maintains a utility value for

every other node j in the network. Then, a node A

forwards to another node B a message destined to node

D, if and only if , where Uth =

threshold value of utility.

Here with varying number of nodes for L copies, we

have evaluated the performance of SNW and SNF

routing techniques for three performance metrics:

delivery probability, average latency, and overhead

ratio. These metrics show that SNF routing exhibits

better performance for only 2% of copies. But in case

of SNW routing, it is 10% of the message copies.

Figure 1. Delivery probability with varying number of nodes

for L copies.

Figure 2. Average latency with varying number of nodes for

L copies.

Figure 3. Overhead ratio with varying number of nodes

for L copies.

Paper ID: 86

68

Emotional Bangla Speech Signals Classification using K-NN

Md. Tohidul Islam*, Md. Ekramul Hamid and Somlal Das

Department of Computer Science and Engineering, University of Rajshahi, Rajshahi, Bangladesh

* [email protected]

KEYWORDS: Emotion Recognition, Bangla Emotional Speech, MFCC, KNN.

Emotion detection from speech is the identification of

emotional or physical state of a human being from his

or her voice. Although emotional state does not alter

the linguistic content, it is an important factor in human

communication because it provides feedback

information in many application such as Ticket

reservation system, call center etc. According to Banse

& Scherer‘s (1996) study on vocal emotional

expression, there are fourteen distinct emotional

categories. For simplicity, we considered only the most

traditional emotions: happy, sad, angry and neutral (no

emotion).

There are a lot of works on emotion detection. Most

of them were done on English language. Besides some

works also done on German, Mandarin, Telugu

language. So emotion detection in Bangla speech is

relatively a new field. The most common challenge is

that there is no emotional Bangla speech database.

We had generated 62 sentences for the corpus. The

sentences were semantically neutral statement (dates

and numbers). We asked 3 persons to pronounce the

sentences according to each emotional mood (happy,

sad, angry and neutral). Thus we have a maximum of

248 sentences for a given student. The recoding system

used was AVS Audio Editor 7.1, with a sampling rate

of 16 kHz and encoded in single channel 16bit PCM.

All recording were transcribed by hand.

MFCCs are the most widely used spectral

representation of speech in many applications,

including speech and speaker recognition. Kim et al.

(2007) argued that statistics relating to MFCCs also

carry emotional information. For each 25ms frame of

speech, 13 MFCC parameters were calculated by

taking the absolute value of the STFT, warping it to a

Mel frequency scale, taking the DCT of the log-Mel-

Spectrum and returning the first 13 components. We

also performed some aggregate functions on these

parameters along all frames. This results a feature

vector of length 56.

Figure 1 shows the block diagram of our emotion

recognition system. We calculated MFCC as the

emotional feature from each input data. Then, the

speech was classified by pattern classification method

(K-NN).

Figure 1. Block diagram of emotion recognition.

Table 1 shows the confusion matrix of our emotion

recognition system from Bangla speech. The rows and

columns represent original and recognized emotion

categories, respectively. For example, first row says

that 96 sentences that belong to angry were recognized

as angry, 00 sentence as happy, 00 sentence as sad and

02 sentence as neutral. So, the recognized accuracy of

anger is 97.96%. We can see that our system do better

in recognizing anger. The mean recognizing rate is

71.43%.

Table 1: Confusion matrix of our System

Angry Happy Sad Neutra

l

Accurac

y

Angry 96 00 00 02 97.96%

Happy 00 59 17 22 60.20%

Sad 00 15 64 19 65.30%

Neutr

al

00 35 02 61 62.24%

In this paper, we built a Bangla emotional database that

cover angry, happy, sad and neutral emotions. We will

continue to get more emotion category into our

database to make it complete so that we can use it to

some real world applications such as Ticket reservation

system, call center.

Testing

Data

Training

Data

Feature

Extraction

(MFCC)

Feature

Extraction

(MFCC)

Classific

ation &

Decision

Making

(K-NN)

Classified

Result

Paper ID: 90

69

Content based Image Searching Using Multidimensional MSF

Saiful Islam1*, Ekramul Hamid

1 and Emdadul Haque

2

1Dept.of Computer Science and Engineering, University of Rajshahi, Rajshahi, Bangladesh

2Dept. of Information and Communication Engineering, University of Rajshahi, Rajshahi, Bangladesh

* [email protected]

KEYWORDS: Markov stationary feature, Markov chain, MMSF, CBIR, Content based image searching.

An effective content-based image feature called

Markov Stationary Feature (MSF) was introduced (J.

Li et al, 2008) for image indexing, searching and

classification in the context of large scale image

databases. It characterizes the spatial co-occurrence of

histogram patterns by Markov chain models.

In the context of MSF characterization, images in

the database are divided into four categories depending

upon the discrimination capability of histogram

analysis: histogram-level distinguishable, intra-bin

distinguishable, extra-bin distinguishable and

histogram undistinguishable images. To form the so-

called MSF, initial and stationary distributions of the

homogeneous Markov chain are combined to encode

the intra-bin and extra-bin relationship of histogram,

respectively. In practice, the MSF method generally

outperforms the corresponding earlier content based

methods (usually suitable for small scale image

database) like Color histogram (G. Pass et al, 1996),

Color Coherence Vector (CCV) (G. Pass et al, 1997),

and Color Auto Correlogram (J. Huang et al, 1999)

features.

Subsequently, some research such as Directed

Markov Stationary Feature (DMSF) (Bingbing Ni et

al, 2009), Multi-direction Markov Stationary Feature

MDMSF (F. Lee et al, 2012), and Markov Stationary

Features and Vector Quantization Histogram

(MSFHQ) (Qiu Chen et al, 2014) based on the same

model have been demonstrated to enhance the

performance of the original MSF.

Though the MSF method and its extended versions

are far better than the earlier methods, they still suffers

difficulty if the underlying image database is

heterogeneous. To overcome the problem, we propose

multidimensional Markov Stationary Feature (MMSF)

model that extends the current MSF by populating

more spatial information by computing

multidimensional co-occurrence matrices of an image

with multiple numbers of histogram bins depending on

the image quantization levels.

Suppose an image I is quantized into nK levels, thus

the set of histogram bins are

. Here, N

indicates the dimension of our proposed NMSF

method. Then, the corresponding co-occurrence

matrices are then calculated as ;

with each element

where d indicates distance between two pixels

and . accumulates the number of co-occurrence

for and . After computing the different co-

occurrence matrices homogeneous Markov chain

model is adopted for every co-occurrence matrix

(essentially converted to a transition matrix) to

characterize the spatial relationship of corresponding

histogram bins (interchangeably, states of in the

Markov chain model). Corresponding initial

distribution and stationary distribution are then

calculated and concatenated (from the co-occurrence

matrix and transition matrix, respectively) to form one

dimensional MSF. The resultant MSFs for every co-

occurrence matrices contribute to form our proposed

NMSF feature space.

The proposed NMSF method can moderately solve

the difficulties prevailing in the original MSF method.

Our experimental results (Fig.1) have justifies the

effectiveness of the proposed system.

Figure 1. Average Precision-recall curves of

Histogram, CAC, original MSF, and proposed MMSF

for randomly selected images from the Corel 1000

database.

Paper ID: 92

70

Silicon Nanocrystals based Schottky Junction Solar Cell Fabrication and

Characterization

A.T.M. Saiful Islam1*, MD. Enamul Karim

2, Arifuzzaman Rajib


2

1Dept. of Applied Physics, Electronics and Communication Engineering, Bangabandhu Sheikh Mujibur Rahman

Science and Technology University, Gopalganj, Bangladesh 2Dept. of Applied Physics and Electronic Engineering, Rajshahi University, Rajshahi, Bangladesh

* [email protected]

KEYWORDS: Schottky junction, silicon nano-crystal, photovoltaic.

A photovoltaic cell can be developed from the

Schottky junction between a semiconductor and a

metal, with or without an insulating layer between

them. This work reports a single-layer Schottky

photovoltaic device that was fabricated by drop-casting

intrinsic silicon nano-crystals (Si-NCs) from colloidal

solution.

In this work a thin layer of indium tin oxide (ITO)

was fabricated on the glass substrate using electron

beam evaporation (E-beam) technique and then a layer

of Si-NCs was developed on that ITO coated glass by

drop casting of Si-NCs and dicholobenzen mixture.

Then a thin layer of Aluminium (Al) was developed

over the Si-NCs layer by using E-beam technique.

Scanning electron microscopy (SEM) was used to

study the surface morphology of fabricated Si-NCs

layer. The photoluminance and the variation of the

absorption on thickness of the Si-NCs layer was

studied using UV flurophotometer. Photo-current

versus voltage (I-V) characteristics was studied with a

Keithley model 2400 source meter and a solar

simulator system.

A layered structure of proposed Schottky junction

solar cell is shown in Figure 1.

Figure 1. A typical schematic of the layered structure

of schottky junction solar cells.

The experimental procedures to fabricate this solar cell

are divided in following steps A. Silicon nano-crystal processing.

B. Cleaning of the glass substrate.

C. Deposition of thin films by E-beam.

D. Preparation of colloidal solution of Si-NCs with

DCB.

E. Final device fabrication and encapsulation.

F. Device characterization

From the experiment it is clear that the absorbance

of the Si-NCs layer increases with increasing of

fineness of the Si-NCs. The I-V characteristics of the

designed Al/Si-NCs/ITO with and without LaF3 were

studies under forward and reverse bias condition. In

this work LaF3 is used because it can lower the work

function of the Al cathode. This help to enhance the

charge collection efficiency by providing proper

energy band matching at the Al/Si interface. The dark

I-V characteristic of the fabricated device shows a

typical rectifying junction behavior with threshold

voltage of 0.3V. The I-V characteristic with LaF3

buffer layer clearly indicates an enhancement (around

15 times) compared to without LaF3 structure.

Figure 2. The I-V characteristics of the fabricated solar cell under 1.5 AM light with LaF3 buffer layer.

Compared to Schottky junction photovoltaic

reported previously, the whole fabrication method (in

this work) is simple, cheap and fast. The performance

parameters such as, Short circuit current density (JSC),

Open circuit voltage (VOC), Fill-factor (FF) and Power

conversion efficiency was low for our fabricated cells.

This may due to fabrication process used to fabricate

the solar cell, processing environment, quality of the

Si-NC etc.

Paper ID: 93

71

Fabrication and Characterization of α-Fe2O3 Homo-Junction

Photocathode for Efficient Solar Water Splitting

Arifuzzaman Rajib1*, Atowar Rahman

2, A. T. M. Saiful Islam


2

1 Department of Applied Physics, Electronics and Communication Engineering, University of Bangabandhu Sheikh

Mujibur Rahman Science and Technology, Gopalganj, Bangladesh 2 Department of Applied Physics and Electronic Engineering, University of Rajshahi,


* [email protected]

KEYWORDS: Photo-catalysis, Metal-Oxide semiconductor, Homo-Junction photocathode.

Production of H2 by water splitting using sun light has

become of great interest among the researchers. Solar

water splitting is considered as an ideal technology to

harvest and store abundant solar energy as clean and

easily transportable hydrogen fuel. Photo-

electrochemical (PEC) cell requires a bias voltage

supplied from an external energy source for efficient

overall water splitting. But the practical PEC water

splitting cell should be a stand-alone device with no

need for an external energy supply. Generally a tandem

cell with another photo-electrode or a photovoltaic

(PV) cell is commonly employed for the stand-alone

devices.

In this paper, we report the fabrication homo-

junction photocathode from a scientifically attractive

material hematite (α-Fe2O3). To make a homo-junction

of Fe2O3, which is naturally, n-type, Zn was doped

Fe2O3 (Zn: Fe2O3) to turn it into p-type. The p-n Fe2O3

homo-junctions were prepared by a simple two-step

spin coating method on ultrasonically cleaned fluorine

doped tin oxide (FTO) (300 nm thick with a sheet

resistance of ~8Ω/sq.) coated glass.

Figure1. Experimental setup for water splitting.

In the first step, 1M Fe2O3 was deposited on FTO

substrate, then annealed for two-hour at 450 ºC in air.

In second step, 0.005M Zn doped Fe2O3 was deposited

on the α-Fe2O3 films, and finally annealed for four-

hour at 450 ºC in air.

The fabricated photocathode was tested in a home-

made experimental set-up with two electrode system

(Fig-1). The thickness of Fe2O3 and Zn: Fe2O3 was also

tuned for the optimum output. The band gap of Fe2O3

and Zn:Fe2O3 were also studied to find the optimum

condition (layer and concentration).

-0.75 0.00 0.75 1.50 2.25

-0.003

0.000

0.003

0.006

0.009

Cu

rren

t (m

A)

Bias voltage (V)

Dark current (without water)

Light current (without water)

Dark current (with water)

Light current (with water)

Figure 2. I-V characteristics for 5 layer of n/p into the water

and without water.

I-V characteristics of photocathode clearly show a

rectifying behavior (Fig-2) with and without water. It is

observed from the I-V characteristic of the junction

that the knee voltage of homo-junction photocathode in

the water was smaller than without water. Reduction of

knee voltage in the water can be interpreted by the fact

that some of the charges came from electrolyte were

accumulated on the junction, and which in turn

decreases the knee voltage.

Experimental results show that homo-junction

photo-cathode of earth-abundant Fe2O3 can be a very

promising electrode for stand-alone PEC cell for solar

water splitting.

Paper ID: 95

72

Human Recognition based on Adaptive Background Mixture Models and Improved

Histogram of Oriented Gradients

Shayhan Ameen Chowdhury and Kaushik Deb*

Department of Computer Science and Engineering, Chittagong University of Engineering and Technology (CUET),

Chittagong 4349, Bangladesh

* [email protected]

KEYWORDS: Histogram of Oriented Gradients, Background subtraction, Gaussian mixture models.

Human recognition in a video surveillance system has

vast application areas including suspicious event

detection, human activity recognition, human

locomotion characterization and fall recognition for

patients etc. In this paper, a framework for human

recognition by generating a background model for

background subtraction process without the prior

knowledge about the backgrounds is proposed.

Initially, every pixel is modeled as a mixture of

Gaussians and use a probabilistic method to update the

model. These Gaussian distributions are then estimated

to find out which are generated from a background.

Each pixel is labeled depending on whether the

Gaussian distribution which exhibits it adequately is

considered part of the background model. Then

morphological operation is used to eliminate

discontinuities in the foreground. After that, the

framework determined which of the ROIs represent

human. Finally, features are extracted from each ROI

for classification. We studied the limitation of

Histogram of Oriented Gradients (HOG) and identified

that it cannot differentiate between some local

structures. To resolve this issue, we proposed a new

feature descriptor improved histogram of oriented

gradients (ImHOG). Various videos containing moving

human are used with variety of conditions to test the

proposed framework and results are presented to prove

the effectiveness.

The escalation of computer vision usages impelled

human recognition as an active research field. HOG

proposed by Dalal et al. (2005) is a very robust feature

descriptor but some local structures cannot differentiate

properly due to mapping gradients of inverse

orientations into same orientation bin. Figure 1 shows

the framework for human recognition. First the input

video is converted into frames. From each frame a

binary image is extracted by subtracting background

process.

Feature

vector

Binary

imageInput frame

Subtracting

background

Labeling and

filtering

Extracting Improved

HOG features

Classifier

(SVM)

Human /

Non-Human

ROI

Figure 1. The proposed framework for human recognition.

Morphological closing operation is applied to

remove holes in the foreground. After that connected

component labeling and filtering is used to find ROIs

and remove non-human regions. The filtering condition

is that the height of the labeled object modeling a

human needs to be two times larger than its width.

Then ROI is processed to generate ImHOG features.

The proposed feature descriptor ImHOG extends HOG

for a cell by including an additional bin which is the

sum of the absolute difference of the gradients of

inverse orientations. ImHOG features then send to

SVM for classification. Processing example of human

recognition is illustrated as Figure 2.

Figure 2. Processing example of human recognition: (a)

input frame, (b) binary image, (c) ROI(s) and (d) recognized

object(s).

We have tested this human recognition framework

on numerous video dataset and acquired an over 94%

human object recognition rate. In this paper, a

framework for human recognition is proposed, which

aimed at recognizing human object from video streams

with high adaptability and precision.

Paper ID: 96

73

A Practical Approach to Spectrum Analyzing Unit using RTL-SDR

Md. Habibur Rahman, Md. Mamunoor Islam*

Department of Electrical & Electronic Engineering, Chittagong University of Engineering & Technology,

Chittagong, Bangladesh

* [email protected]

KEYWORDS: Software defined radio (SDR), RTL-SDR dongle, Spectrum analyzer, Tuner IC.

At present scenario, there has been an immense

advancement in the field of wireless communication in

this modern engineering world. Now-a-days, the

advancement of Software defined radio (SDR) system

has made significant progress which makes it as a

serious substitute of traditional hardware radio

architectures where the mathematical procedures are

obligatory to decode and process radio signals using

analogue circuitry.

Traditionally spectrum analyzing activities are

mainly performed by governmental agencies where

expensive specialized hardware setups are used. With

these radio architectures, a special receiver for almost

each radio communication standard is needed.

Recently, computers have turned out to be powerful

enough to do the required mathematical calculations

using software.

Hence, aim of this paper is to demonstrate a RTL-

SDR based spectrum analyzer which can be used

proficiently as an alternative of existing hardware

spectrum analyzer. This approach will lessen the

complexity of analogue hardware system with the

higher tractability of software based filtering and

demodulation techniques.

Several approaches have been made by the

researchers all over the world employing this device on

their research works such as D. Pfammatter et al.

(2014), K. Tapping et al. (2015), J. Saalmüller et al.

(2015) and T. Zhang et al. (2015).

As RTL-SDR devices are quite cheap

(Approximately 20$) and small sized, this system also

offers cost effectiveness with provision of portability

(T.Rudolph, 2013). An experimental study was

conducted practically with suitable conditions to

examine the feasibility and efficiency of the proposed

system. The block diagram of the methodology is

illustrated below (see Figure 1).

Figure 1. Operational Method.

Various types of signals have been received and

analyzed by this spectrum analyzing system in practical

tryout. In Bangladesh, GSM 900 and GSM 1800 bands

have been used for mobile communication system.

Banglalink, a local mobile operator in Bangladesh uses

895.2-900.2 MHz frequency band for uplink. This

uplink frequency received by RTL-SDR are illustrated

and analyzed below (see Figure 2).

Figure 2. FFT Spectrum of Uplink Frequency.

The experimented data of GSM uplink signal

bandwidth analyzed from the figure is close to real data

which is 200 KHz (B. Furht, M. Ilyas, 2003).

So, RTL-SDR device can be used as an alternative

for spectrum analyzing purposes by which the

complexity in signal analysis can be lessened easily.

Though the performance of this system is not fully

apposite, the system is both user-friendly and cost

effective compared to the traditional hardware system.

So it can be used as a modern spectrum analyzing tool

which has wide range of frequency tuning

conveniences for analyzing the signal more

proficiently.

Paper ID: 97

74

Fabrication of Bismuth Ferrite Multiferroic Perovskite Nanoparticles Using an

Aqueous Organic Gel Route

Mayeesha M. Haque1* , M. S. Parvez

1, M. S. Islam

1, M. A. Hakim

2, M. A. Gafur

3

1Department of Materials Science and Engineering, University of Rajshahi, Rajshahi-6205, Bangladesh

2 Department of Glass and Ceramic Engineering, Bangladesh University of Engineering and Technology (BUET),

Dhaka- 1000 , Bangladesh 3Bangladesh Council of Scientific and Industrial Research (BCSIR), Dhaka- 1000, Bangladesh

* [email protected]

KEYWORDS: Multiferroic, Perovskite, Nanoparticles, Sol-gel. X-ray Diffraction, Scanning Electron Microscopy.

Multiferroic materials exhibit simultaneous

ferroelectric and magnetic orders and gained

considerable interest due to their prospect for advanced

technological applications such as, non-volatile

multiple state memories, magnetically tunable

piezodevices, spintronics, and new generation

photovoltaic devices [1]. The potential of multiferroics

in advanced technological applications is the

inspiration behind the attempt to synthesize bismuth

ferrite (BiFeO3 or BFO) nanoparticles in the present

investigation.

Nano-sized single-phase BiFeO3 particles within the

size range of 30- 50 nm have been synthesized by an

aqueous organic gel route in this experiment. By

controlled heating of a homogeneous solution of

analytical-grade Bi(NO3)3.5H2O and Fe(NO3)3.9H2O

with the presence of HNO3 and C4H6O6 a precursor

powder was formed, which was calcined at varying

temperatures to yield BFO powder. The vibrational,

structural, morphological and thermal properties of the

prepared BFO powders have been studied by FT-IR

(Fourier Transform Infrared) spectroscopy, XRD (X-

ray Diffraction), SEM (Scanning Electron Microscopy)

and TG/DTA (Thermogravimetric and Differential

Thermal Analysis).

The FT-IR study preliminarily indicates the

formation of FeO6 octahedra due to the presence of the

peak around 445 cm-1

whereas presence of BiO6

octahedra can be inferred from the peak present around

550 cm-1

[2]. The XRD spectra show that pure BiFeO3

phase is formed in all the fabricated samples. A single

phase BFO is obtained at a temperature of 500o C and

higher. The calculated lattice parameters a, b and c

correspond to rhombohedrally distorted (R3C) unit cell

of BFO. Lattice constants a and c remained more or

less constant while the value of b increased gradually

with the increase in temperature which caused the

crystal volume to increase as well. The particle size of

the prepared powder has been calculated using

Scherrer‘s formula and remained within the range of

28-39 nm which is below the spin cycloid of BFO (64

nm) [3].

The particle morphology of the prepared BFO

powder samples was studied using SEM. The observed

particle size of the BFO samples increased with the

increase in calcination temperature and remained

within the range of 30 - 50 nm as shown in fig. 1,

which is well in agreement with the particle size

calculated using XRD line broadening.

Figure 1. SEM micrograph of BFO samples calcined at (a)

450o C and (b) 500o C TG/DTA result indicates that the formation of BFO

phase initiates around 300o C or below with some

pyrochlore phase [4]. The result also suggests that

formation of single phase BFO is complete within a

temperature range of 500- 550o C, which is well in

agreement with XRD results.

In this experiment BFO nanoparticles were

synthesized successfully at variable calcination

temperatures via an aqueous organic gel route. The soft

chemical route provided an easy and reproducible

approach to fabricating BFO nanoparticles of desired

morphologies. This experiment can be considered the

cornerstone in developing advanced research schemes

regarding practical application of BFO nanoparticles.

Reference:

1. R. Ramesh, N. A Spaldin, Nature Materials, 6, 21.

(2007)

2. J. Yang, X. Li, J. Zhou, Y. Tang, Y. Zhang, Y. Li,

Journal of Alloys and Compounds. 509, 9271. (2011)

Paper ID: 101

75

Purification, Characterization and Anti Proliferative Activity of A Lectin

Isolatedfrom MulberrySeeds

*Mst. Jesmin Sultana1, Syed Rashel Kabir

2 and M. Taufiq Alam

3

1Dept. of Materials Science and Engineering

2Dept. of Biochemistry and Molecular Biology

3Dept. of Applied Chemistry and Chemical Engineering

*[email protected]

KEYWORDS: Lectin, purification, Heagglutination, affinity chromatography, Antiproliferative.

Lectins are a class of proteins that bind sugar

specifically and agglutinate cells. It is widely

distributed in nature, being found in animals, insects,

plants and microorganisms. Plant lectins have a broad

range of biological implication. Mulberry is the only

food and nutritional source for Bombyx mori L. It is the

host plant of silkworm which produces silk [1].

Mulberry seeds has lot of medicinal values so it is used

as conventional therapeutic plant throughout Southeast

Asia especially in Bangladesh. Literature shows that

galactose and mannose specific lectins has been

purified from mulberry seeds by absar et al. [1, 2]. In

this research a sword bean lectin specific mulberry

seed lectin (MSL) was isolated and purified from the

seeds of morus alba l. (mulberry plant) by extraction

with acetate buffer at pH 4.0, followed by 90%

ammonium sulfate precipitation. The dialyzed

ammonia sulfate precipitate was subjected to affinity

chromatography using an agarose gel-matrix on which

sword bean lectin was immobilized. The newly

prepared affinity matrix was able to purify mulberry

seed lectin in a single step. The purified mulberry seed

lectin is monomer in nature as judged by SDS-PAGE

and its MW was estimated to be 22 kDa. Mulberry seed

lectin powerfully agglutinated with erythrocytes, the

minimum concentration was found to be 0.031 μg/ml

in mice, 0.125 μg/ml in chicken, 8 μg/ml in bovine and

0.25 μg/ml in human blood types A, B and O

erythrocytes. This agglutination was inhibited

specifically by D-glucose and D-mannose,

respectively. Highest lectin activity was observed

between 30-60ºC. MSL lost its activity 78% at 70ºC

and completely at 80 ºC. MSL showed 56% and 70%

inhibition against Ehrlich ascites carcinoma(EAC) cells

in vivo in mice when administered 1.4 mg/kg/day and

2.8 mg/kg/day (i.p) respectively for five days.When

MSL injected into the EAC bearing mice for 10 days

its significantly increased the hemoglobin and RBC

value towards the normal label.

Reference

1. Absar, N., Yeasmin, T., Raza, M. S., Sarkar, S. K., &

Arisaka, F. (2005). Single step purification,

characterization and N-terminal sequences of a mannose

specific lectin from mulberry seeds. Protein Journal, 24,

369–377.

2. Yeasmin, T., Tang, M.A., Razzaque, A., Absar, N.

(2001) Purification and characterization of three

galactose specific lectins from mulberry seeds (Morus

sp.). Eur J Biochem. ; 268:6005-10.

Paper ID: 102

76

Analysis of GLDAS Data for Estimating and Distribution of

Evapotranspiration and Rainfall Over Bangladesh

Md Ataur Rahman, Md Mainul Islam Mamun and Md Monirul Islam*

1Department of Applied physics & Electronic Engineering, University of Rajshahi, Bangladesh

* [email protected]

KEYWORDS: Evapotranspiration, Rainfall, GLDAS, Bangladesh.

Bangladesh is one of the most climate vulnerable

country. Nowadays climate is setting more stress on

agricultural productivity. The climate change may lead

to change in evapotranspiration (ET). The present work

investigates the spatio-temporal variations in ET and

rainfall under climate change in Bangladesh with the

help of Global Land Data Assimilation System

(GLDAS). ET and rainfall are analyzed at annual and

seasonal scale to find out the pattern of ET and rainfall

in the study area during the period 2004-2013.

Fig.1 shows the temporal variation of ET over

Bangladesh for the period 2004-2013. The ET is

exhibiting a good periodic pattern which is increasing

gradually from 2004 to 2013. It is also observed that

the rainfall is decreasing across Bangladesh during the

last decade which is not shown in figure.

Figure 1. Trend in evapotranspiration over Bangladesh

for the period of 2004-2013.

Fig.2 shows the spatial distribution of ET over

Bangladesh for the period 2004-2013. The vertical bar

represents the scaling of total ET. The eastern part of

Bangladesh is showing higher value of ET but the

western is showing lower value of ET. It is also

observed that the eastern part of Bangladesh is showing

higher rainfall rate whenever the western part is

showing lower rainfall rate during the last decade as

shown in Fig.3.

The seasonal variations of ET and rainfall rate over

Bangladesh were observed based on four seasons:

winter, summer, monsoon and post-monsoon seasons

(not shown in figure).

The seasonality of ET exhibits the patterns of

higher values during summer & post-monsoon seasons

whereas lower during winter & monsoon seasons. On

the other hand, it is also observed that the seasonality

of rainfall rate exhibits the patterns of higher values

during summer season whereas lower during winter,

monsoon and post-monsoon seasons. Lower

evaporative energy and lower temperature in winter

may be the reasons for minimum ET and minimum

rainfall rate in winter.

The analysis of annual and seasonal distribution of

ET and rainfall give the rate of ET and rainfall are

lower in the western region than the other regions of

Bangladesh. Thus the western region of Bangladesh

will require more irrigation and vegetation.

Figure. 2. Annual-mean spatial distribution of ET over

Bangladesh for the period of 2004-2013.

Figure. 3. Annual-mean spatial distribution of rainfall rate

over Bangladesh for the period of 2004-2013.

Paper ID: 105

77

Robustification of Logistic Classiferfor Binary Classification in

Microarray Gene Expression Data

Md. Shahjaman1,2*

, Md. Mushfiqur Rahman1, Md. ManirHossain Mollah

1, AnjumanAra Begum

1, S. M. Shahinul

Islam3 and Md. NurulHaque Mollah

1

1Laboratory of Bioinformatics,Department of Statistics, University ofRajshahi,Rajshahi-6205, Bangladesh.

2Department of Statistics, Begum Rokeya University, Rangpur, Bangladesh

3Institute of Biological Science, University of Rajshahi, Rajshahi-6205, Bangladesh

* [email protected]

KEYWORDS: Gene expression, Logistic classifier, Differential expressed (DE) genes, Median absolute deviation

(MAD), robustness.

The aim of classification is to separate/classify the

new objects into one of two or more populations on

the basis of a training data set whose category

membership is known. A number of classifiers exist in

the literature by many authors. Among them logistic

classifier is a standard and popular method for

building prediction models for a binary outcome.

However, most of the classifiers are sensitive to

outliers including logistic classifier when the gene

contamination rate is increased. Also most of the

classifiers suffer with inverse problem of its covariance

matrix in presence of large number of genes (p)

variables with small number of patients/samples (n) in

the training dataset. To overcome the earlier problem,

an attempt is made to detect outliers and modify the

outlying data points in training data set using median

and median absolute deviation (MAD) method, then

we apply the logistic classifier. To overcome the later

problem, we select few top differentially expressed

(DE) genes for binary classification on modified

training dataset using the t-statistic.

We examine the performance of our proposed method

with other classifiers like linear discriminant analysis

(LDA), support vector machine (SVM), naive Baye's

(NB) and classical logistic for simulated and real

datasets in case of both small and large-sample. We

generate the training and test datasets for the data type

D = {D1,D2}, first we consider n1 = 20 and n2 = 20

samples from m = 2 different multivariate normal

populations Np(μ1,V ) and Np(μ2, V ) of different

dimensionsp=5 and 15 respectively.The figure-1 shows

that the common mean difference vs misclassification

error rate in presence of one or two outlying sample in

training dataset across the genome.

Figure 1. Mean difference vs misclassification error rate.

The table 1 shows the average performance results of

five methods (LDA, SVM, NB, Logistic and

Proposed) based on 100 datasets generated for data

type D. It is obvious from the figure 1 and table 1 that

our proposed method show better performance than the

other classifiers in presence of outliers. We also

observed that in absence of outliers our proposed

method keeps equal performance. Table 1.Performance evaluation based on simulated data in

presence of one or two outlier(s) across the genome

MethodsAUCpAUC

LDA 0.650 0.029

SVM 0.816 0.134

NB 0.790 0.125

Logistic 0.700 0.033

Proposed 0.975 0.175

Paper ID: 113

78

Molecular Evolutionary Analysis of α-Defensin Peptides in Vertebrates

Arafat Rahman*, M Sahidul Islam, OtunSaha and Titon Chandra Saha

Noakhali Science and Technology University, Noakhali, Bangladesh

* [email protected]

KEYWORDS: α-Defensin, Antimicrobial Peptides, Vertebrates, Evolutionary Analysis, Selection.

α-Defensin is a group of polypeptides with

antimicrobial activity found in host-defense system and

it is widely distributed in but not limited to mammalian

epithelial cells and phagocytes. These molecules

protect organism from a diverse spectrum of bacteria,

viruses, fungi, and protozoan parasites.

Different studies revealed wide sequence variation

within α-defensin sequences but underlying

evolutionary cause is not well-studied. In this study, α-

defensin gene from 25 vertebrate species has been

comprehensively collected and computationally

analyzed.

Gene and nucleotide database of NCBI was accessed to

extract meta-information of 101 α-defensin genes and

full CDS sequences downloaded from its nucleotide

database by splitting-out intron sequence.

MEGA software used to construct phylogenetic tree

using Neighbor-Joining method. This analysis indicates

that α-defensin gene evolution do not matches with

species evolution and copy-number of this gene differs

among species.

Selection analysis was carried out using DataMonkey

web-server's FEL, SLAC, IFEL, MEME, TOGGLE

and REL program on both propeptide and defensin

super-family codon-aligned sequences to test different

hypothesises. Positively selected sites found on both

propeptide and defensin domain, but the effect of

negative-selection pressure is higher (Table 1 and

Figure 1, respectively).

SWISS-MODEL web-servers were used for homology

modeling of selected α-defensin genes. Structural

variation observed on α-defensin proteins which may

indicates heterogenous structure-function relationship

between species that reflects its interaction with

particular pathogen. This study provides new

perspective on the relationships among α-defensin gene

repertoires which will help to infer its evolution.

Table 1. Evidence of positive selection using SLAC method

on α-defensin domain sequences at 0.1 significance level. Codon dN/dS Normalized

dn/dS

p-Value

7 16.0808 0.961451 0.0290828

9 11.5606 0.691191 0.0592005

11 11.7879 0.70478 0.0905814

13 12.0578 0.720919 0.057693

19 9.92191 0.593217 0.0902638

Figure 1. Negative purifying selection signal measured by

FEL in propeptide part of alpha defensin gene sequences.

Paper ID: 115

79

Microencapsulated Probiotic Bacteria Protect the

Spoilage of Functional foods

Md. SymoomHossain, Md. Abdul Alim Al-Bari*, Mir Imam IbneWahed and Md. Anwar Ul Islam

Department of Pharmacy, University of Rajshahi, Rajshahi-6205, Bangladesh

*[email protected]

KEYWORDS: Probiotics, microencapsulation, biopreservation, functional food.

Probiotic live bacteria have been paid considerable

attention to scientific researchers for treating several

diseases including inflammatory bowel diseases,

colorectal cancer, metabolic syndromes, autoimmune

diseases like rheumatoid arthritis etc., as well as in

diagnostic purposes such as production of biomarkers

e.g., detection of cancers and diabetes. To achieve

beneficial effects for human health, probiotic bacteria

must have good technological properties, survive

gastrointestinal passage and be able to function in the

gut environment. When probiotic bacteria

administered orally, they must be protected from the

stomach acidic condition and can be denatured by bile

acids, antimicrobial compounds and degradative

enzymes before reaching the target sites. These

obstacles limit their arrival to the gut in alive. The

microencapsulation technique protects them from

these unfavorable conditions. In microencapsulation,

the live probiotic bacteria and/or their bioactive

compounds are trapped to protect from the harsh

conditions and to deliver them with improved survival

rate.

Microencapsulation technique of food processing is

an innocuous and ecological approach for food

preservation. To harmonize consumer demands with

the necessary safety standards, traditional means of

controlling microbial spoilage and safety hazards in

foods are being replaced by innovative technologies

including biological antimicrobial systems. This trend

has favored consumption of foods enriched with

physiologically active components including probiotic

bacteria.

Here, we screened pharmaceutical probiotic

products and curds for rejuvenation and identification

of probiotic lactic acid bacteria. We found a

discrepancy between the enumerated viable bacteria

amounts and the claims of the manufacturers.

The preservative activity of microencapsulated

lactic acid bacteria (LAB 1-4)in orange juice (OJ) had

been observed due to the lowering of pH to 3.3 through

acid production and the inhibition of growth of

pathogenic microorganisms which can cause food

spoilage, food poisoning and disease. By contrasting

free form of LAB, the average pH decreased from

3.94 to 2.71 in OJ containing free bacteria after

four weeks of storage. This result suggests that

bacteria in micro encapsulated state had a more

stable survival environment. By doing this, the

shelf life of fermented food can be prolonged.

Figure1. pH change in orange juice micro encapsulated LAB.

The microencapsulated probiotics would be used in

preservation of food and beverage industry in place of

chemical preservatives for the production of functional

foods in the near future.

Paper ID: 117

80

Preparation of Highly Cross-linked Magnetic Polymer Composite Particles and

Application in the Separation of Arsenic from Water

M. K. Sharafat*, H. Ahmad, M. A. Alam and M. M. Rahman

Department of Chemistry, University of Rajshahi, Rajshahi-6205

* [email protected]

KEYWORDS: Styrene, Divinylbenzene, Suspension polymerization, Highly crosslinked polymer.

The contamination of daily used water by toxic

metal ions is a worldwide environmental problem.

Since the degradation of toxic metal ions does not

occur naturally, they have a tendency to accumulate in

organisms and enter the food chains. Among the

different toxic metals, arsenic (As) is a common and

toxic pollutants which is mobilized by natural

weathering reactions, biological activity, geochemical

reactions, volcanic emissions and other anthropogenic

activities. Most often in supplied water As exists in

inorganic forms: arsenite (AsO33−)

and arsenate

(AsO43−

).

Iron oxide magnetic particles have become a

promising research field in separation technology

because of their easy separation by external magnetic

field and can be applied for the removal of toxic metals

from waste water. Highly cross-linked Fe3O4/P(S-

DVB) particles were prepared in this research by

suspension polymerization of styrene (S) and

divinylbenzene (DVB) in presence of nano sized

Fe3O4particles.The mixture was sonicated for 3 min at

10,000 and 5,000 rpm prior to the polymerization

respectively and finally polymerized at 75°C for 24 h.

At first Fe3O4 nanoparticles were prepared by co-

precipitation of Fe2+

and Fe3+

from their alkaline

aqueous solution. To stabilize the magnetic

nanoparticles, the surface of the particles was modified

with oleic acid.TEM images of Fe3O

4 P(S-DVB)

particles and P(S-DVB)/Fe3O

4 composite particles

prepared at sonication speed 10,000 rpm and 5000 rpm

are given below which show that the size of Fe3O4/P(S-

DVB) composite particles prepared at 10000 rpm

sonication and 5000 rpm sonication are larger than

Fe3O4 particles which indicates the possibility of the

formation of P(S-DVB) layer on the surface of Fe3O4

particles.

Figure 1. TEM images of Fe

3O

4 (a), P(S-DVB) (b) particles

and P(S-DVB)/Fe3O

4 (c,d) composite particles prepared at

sonication speed 10,000 rpm (b,c) and 5000 rpm (d).

The morphology and surface structure were also

characterized by FTIR, X-ray photoelectron

spectroscopy, scanning electron microscope, light

microscope, thermo gravimetric analysis and X-ray

diffraction spectroscopy. The theoretical porosity

calculation indicated that the porosity present on the

surface of polymer as well as magnetic polymer

composite.

Table 1. Porosity calculation.

S.

No.

Parameter Polymer Composite

1 Bulk density 0.3945 0.4728

2 Particle

density

3.3143 1.6607

3 Porosity% 88.10 71.53

Finally, the Fe3O4/P(S-DVB) composite particles

were used for the removal of As from As contaminated

water which showed good adsorption capacity

(19.77×10-3

mg/m2) in adsorption compared to the P(S-

DVB) particles(10×10-3

mg/m2) after 2h of mixing.

b

c d

a

500nm

nm

500 nm

50 nm 500 nm

Paper ID: 124

81

Preparation of Hydrophobic Poly ((lauryl methacrylate)-coated Magnetic Nano-

Composite Particles and Their Application as Adsorbents for Organic Pollutants

RukhsanaShabnam* and Hasan Ahmad

Department of Chemistry, University of Rajshahi, Rajshahi-6205

*[email protected]

Keywords: Lauryl methacrylate, Suspension polymerization, Hydrophobic, Nano-composite.

This work introduces a method for the preparation

of magnetic nano-composite particles coated with

highly crosslinked poly (lauryl methacrylate) (PLMA),

a hydrophobic polymer containing long chain alkyl

groups, used for application in waste water treatment.

Magnetic (Fe3O4) nano particles were prepared by

coprecipitation of Fe2+

and Fe3+

from their alkaline

aqueous solutions which was encapsulated with SiO2

using traethylorthosilicate (TEOS). Then precipitation

copolymerization was carried out using LMA and

divinyl benzene (DVB) in presence of Fe3O4/SiO2

particles within stable droplets containing hexadecane–

toluene mixture (4:1 mixture HD-T) as porogen. The

produced magnetic composite particles named

Fe3O4.SiO2/P(LMA-DVB) were characterized by

Fourier Transform IR (FTIR), transmission electron

microscopy (TEM), thermogravimetry (TG) and X-ray

diffractometer (XRD) analyses. Then the prepared

composite particles were used for the removal of

organic pollutants from water. In the FTIR spectrum of

Fe3O4, the strong band at 1059 cm-1

corresponds to

Si―O―Si bonds indicated the bonding of SiO2 to

Fe3O4. The appearance of the signals at 3400 and 1617

cm-1

, the new peaks in the region 2761–2968 cm-1

, the

small signal at 1710 cm-1

corresponded to the stretching

vibration of ester carbonyl group derived from LMA,

the shoulder peak at around 3000 cm-1

confirmed the

formation of trilayered magnetic Fe3O4/SiO2/P(LMA-

DVB) composite particles. The discernible six

characteristic peaks for Fe3O4 that appeared at 30.3°,

35.6°, 43.5°, 54.0°, 57.4° and 63.0° indexed to (220),

(311), (400), (422), (511) and (440), respectively,

which matched well with the standard database of

magnetite. The peak positions remained the same

during different steps of modification, indicating that

the crystalline structure of the magnetite is essentially

maintained. The broad diffraction peak averages at

13.2° represent the amorphous SiO2 and copolymer

layer.

Figure1. XRD spectra of Fe3O4/SiO2/ P(LMA-DVB)

composite particles.

The efficiency of Fe3O4/SiO2/P(LMA-DVB)

composite particles as an adsorbent for organic

pollutants was evaluated on the basis of the adsorption

behavior of some organic pollutants like herbicide

(gramaxone), phenol, salicylic acid and congo red (dye)

on Fe3O4/SiO2/P(LMA-DVB) composite particles. It

was evident from the results that the magnitude of

adsorption is dependent on the nature as well as

molecular size of the pollutants. The adsorption

efficiency reached maximum for hydrophobic large

molecular weight congo red followed by phenol.

Composite particles have minimum loading efficiency

for herbicide. It is most likely that the higher water

solubility of the pollutant is responsible for poor

adsorption efficiency. The removal of different organic

pollutants by magnetic composite particles driven by

hydrophobic interaction was evaluated. They showed

good adsorption capacity particularly for hydrophobic

organic pollutants.

Paper ID: 125

82

Statistical Methods for Functional Analysis of Metagenomes

Zobaer Akond1, 2, 3*

and Md. Nurul Haque Mollah2

1Institute of Environmental Science, University of Rajshahi

2Bioinformatics Lab, Department of Statistics, University of Rajshahi

3Agricultural Statistics and ICT Division, Bangladesh Agricultural Research Institute

* [email protected]

KEYWORDS: random forest, multiple dimensional scaling, linear discriminant analysis.

Metagenomics is the relatively new area of

Bioinformatics which enables to understand the

diversity of microbes, their functions, cooperation and

evolution in particular ecosystem. The voluminous data

generated through DNA sequencing in each

metagenome makes identifying key differences in the

function and taxonomy between communities.

Two statistical techniques are employed for

functional analysis of 212 metagenomes within and

between 10 environments against 27 metabolic

functions (Dinsdale et al., 2013).

Random Forest (RF) is typically used to classify the

data either in supervised or unsupervised manner. A

subset of the data and variables is used to generate the

trees, and thus the approach can predict the

environment to which a metagenome belongs. It

generates a measure of the importance of each variable

calculated by the mean decrease in accuracy (Figure 1).

It finds the 8 most important metabolic functions

among which Photosynthesis has the highest score

70.20; Phages, Prophages has the second highest score

61.31 and Membrane Transport has the eight highest

score 45.29. These 8 most important metabolic

potentials are used for multidimensional scaling

method.

Multidimensional Scaling (MDS) is a visualization

technique that directly scales objects based on either

similarity or dissimilarity matrices (Quinn and Keough,

2002).It is useful for showing which metagenomes

have similar features (Figure 2).

The MDS plots are accomplished with the 10

predefined environments by considering the 8 most

important metabolic variables. In this analysis, the

visualization highlights the separation of the microbes

from human or animal hosts from other samples along

the first dimension and the separation of the aquatic

and mat communities along the second dimension.

Random Forest shows that phage activity is a major

separator of host-associated microbial communities and

free-living. It also finds the 8 most important metabolic

functions. The MDS technique suggests that mat

communities separated from both the animal associated

metagenomes and the aquatic samples by the vitamin

and cofactor metabolism.

Paper ID: 126

83

Simulation of Microalgae and CO2 Flow Dynamics in a Tubular Photobioreactor

and Consequent Effects on Microalgae Growth

Saumen Barua1*

, Mohammad Morshed Alam2 and, Ujjwal Kumar Deb

3

1Department of Mathematics,Sir Ashutosh Govt. College, Chittagong, Bangladesh

2Department of Mathematics,Chittagong College, Chittagong, Bangladesh

3Department of Mathematics, Chittagong University of Engineering & Technology, Chittagong, Bangladesh

*[email protected]

KEYWORDS: Microalgae, Biofuel, Tubular Reactor, CFD, Simulation.

Nowadays energy plays a central role in

development and functioning of the world economy. It

is also true fact that continuous use of liquid fossil fuel

is unsustainable due to depleting sources of fossil fuel.

So, to find an alternatives to fossil fuel have engaged

scientists and entrepreneurs around the world.

Microalgae based biofuel is considered as the most

sustainable alternative to the fossil fuel (Y. Chisti.

2007). In microalgae based biofuel technology, the

main optimal factors for microalgae cultivation are

light, CO2, temperature. CO2 is the main carbon source

for photosynthetic culture of microalgae. The optimal

growth of microalgae depends on CO2 requirement

with different species of algae and geometries of

Photobioreactor (PBR) (Bitoget. al., 2011). In this

study, a two phase flow for CO2 and Microalgae

suspension is considered to understand fluid dynamics

phenomena after injecting CO2 gas inside the tubular

PBR. The growth rate of the microalgae cell is a

function of available light irradiance at a geographical

location while concentration of the cell is related to the

growth rate.

The Tubular PBR is considered in our study have

the radius of 0.025 m while length is 20.94 m. The

surface area and working volume are 3.279m2 and

0.04043 m3 respectively. A fine mesh design is

considered for our simulation with 125691elements

and 1055747 degrees of freedom. For the light energy

from the sun, we considered the location of the

computational domain is placed at Chittagong

University of Engineering & Technology (CUET),

Chittagong with latitude 057222 and longitude

618591 on 21 June (for bright sunny and the

longest day) with average 13.27 day hour. The

ComsolMultiphysics was used to simulate the results.

Figure 1. Cell Concentration for the day seven of the

culture.

Figure 2. Shear rate distribution along the entire

computational domain.

From the simulation after day seven we observed a

very slow growth for the microalgae culture. Figure-1

shows the cell concentration of microalgae culture on

the seventh day from morning (06:00) to the evening

(18:00) increases about 0.013 kg/m3. From this result

we can deduce that the growth related to concentration

of microalgae is increased by day length with respect to

continuous light and CO2.We also study the effect of

CO2 injection inside the PBR on flow phenomena. A

small fluctuation of shear rate around U-loop area is

also found in this study which might reduce the cell

concentration around U-loop areas due to cell fragility

which shown in the Figure-2.

Paper ID: 127

84

MLE Based Robust ICIM for QTL Analysis with Backcross Population

Md. Amanullah1, Md. Mamun Monir

2, Md. Jahangir Alam

1, Md. Alim Hossen

1, Mita Khatun

3 and

Md. Nurul Haque Mollah1

1Laboratory of Bioinformatics, Department of Statistics, University of Rajshahi, Rajshahi, Bangladesh

2Department of Agronomy, College of Agriculture and Biotechnology, Zhejiang University, China

3Department of Physics, Govt. Edward College, National University, Rajshahi, Bangladesh

* [email protected]

KEYWORDS: Quantitative trait loci (QTL), Interval Mapping (IM), Inclusive Composite Interval Mapping (ICIM),

Beta-likelihood.

Quantitative trait loci (QTL) analysis is a widely

used statistical method for detecting the influential

genes of quantitative traits. Inclusive composite

interval mapping (ICIM) is one of the most popular

approaches for QTL analysis to separate the genetic

architecture of a certain trait. In ICIM stepwise

regression is use to select representative markers of

QTLs and earlier to mapping subtracts the influence of

the background variants from phenotypic data. Indeed,

after subtracting influences of the background QTLs,

ICIM approach equivalent to classical interval mapping

(IM) approach. ICIM approach use classical likelihood

function to estimate and predict QTL effects and

position. Therefore, in presence of phenotypic outliers

the approach can give biased estimate of both QTL

effects and positions. In this paper, we proposed robust

ICIM algorithm by maximizing beta-likelihood

function. The proposed method reduces to the

traditional ICIM method when beta tends to 0. The

tuning parameter beta plays an important role on the

performance of the proposed method for QTL

mapping. It controls the trade-off between the

robustness and efficiency of the estimators. The

analysis results of simulated data show that the

proposed method produces more better results in

presence of outliers than classical ICIM for backcross

population; otherwise its keep equal performance.

To investigate the performance of the proposed

method in a comparison with others traditional

approaches, let us first consider three linked QTLs with

15 equally spaced markers in each of chromosomes

where any two successive marker interval size is 5cM.

The true QTL position is located in chromosome 2 with

marker 18, 21 and 25. The true values for the

parameters in the ICIM model are assumed as μ = 0.2,

a = .4 and σ2= 0.2. To test the null hypothesis H0: a= 0

vs. H1: a≠0, we generated 250 trait values. To

investigate the robustness of the proposed method, we

contaminated 20% trait values in this dataset by

outliers. Then we computed LOD scores by IM, ICIM

and the proposed methods for both types of datasets.

Figure 1 shows the LOD scores profile for the

contaminated dataset, where dashed, dotted, solid line

and two dash represents the LOD scores at every 1cM

position in the chromosomes as before for original

(ICIM), IM, ICIM and the proposed method with β =

0.02, respectively. It is seen that the classical IM and

ICIM methods are fail to detect the true QTL position

when the data are contaminated with outliers.

However, in presence of outliers, the highest LOD

score peak occurs in the true QTL position by the

proposed method only (see Figure: 1).

Figure 1. LOD score profile in presence of outliers.

Paper ID: 129

85

Evaluation of PCA in Spatial, Frequency and Wavelet

Domains for Face Recognition

Samsi Ara* and M. Babul Islam

Dept. of Applied Physics and Electronic Engineering, University of Rajshahi, Rajshahi 6205, Bangladesh

* [email protected]

KEYWORDS: Face recognition, PCA, Wavelet transform, ORL face database.

Principle component analysis (PCA) is a linear

technique of dimensionality reduction which transforms

data in the high dimensional space to low dimensional

space. In this paper, the performance of PCA in spatial,

frequency and wavelet domains has been evaluated on

ORL face database. The database contains 400 images of

40 persons, i.e, 10 for each person.

In PCA, the eigenfaces obtained from the eigenvectors

form a basis set of faces which can be used to recognize

face images.

Let I be a 2-D image of size nn and transformed to 1-D

image vector as TnnIII ),(,),2,1(),1,1( .

Now, for M concatenated training images the mean image

[Figure 1(a)] is computed as

M

i

iM

1

1 (1)

The mean-centered image for each image is

ii (2)

The covariance matrix of the mean-centered images is

given by

TAAC (3)

where MA ,,, 21 and the size of C is 22 nn .

Therefore, calculating 2n eigenvectors and eigenvalues

correspond to C needs a tremendous computational effort.

To reduce the computational cost AAT can be used

instead of C as in [1] to obtain M eigenvectors and

eigenvalues. Now, the ith eigenface [Fig. 1(b)] can be

obtained as

MiVVU i

Tii

Tii ,,2,1),( (4)

where iV are the eigenvectors of C. It should be noted that

the )( MP significant eigenvectors correspond to P

highest eigenvalues instead of M eigenvectors can be

used.

Now, a face image can be projected into the eigenface

space as PkU Tkk ,2,1),(

In the recognition process, the Euclidean distance

given in Eq. (5) of each face of training dataset from test

face is used to determine the belonging class of the test

face.

22

kk (5)

where is a vector representing the test image in the

eigenface space.

(a) (b)

Figure 1. (a) Mean face, (b) Eigenface In frequency domain PCA (FPCA), the Fourier

magnitudes of the intensity values of pixels of the images

are used to obtain the eigenface space following the same

process as like in the spatial domain PCA.

Recently, Wavelet Transforms (WTs) have been applied

in different image processing applications. The 2-D DWT

decomposes an image into subbands of lower resolution

which in turn reduces the computational complexity of

estimation of eigenfaces. In this paper, the Daubechies 4

(db4) wavelet [2] has been used to decompose the face

image into 4 subbands, such as, approximate, horizontal,

vertical, and diagonal components. Then the approximate

components are used in the PCA algorithm to obtain the

eigenface space.

From Figure 2 it has been found that the maximum

accuracy is obtained for FPCA whereas the minimum

accuracy is obtained for spatial domain PCA. For wavelet

domain PCA (WPCA) the recognition rate is slightly lower

than that of FPCA but much better than PCA.

Figure 2. Recognition rate for PCA, FPCA & WPCA.

References [1] M. Turkand A. Pentland, ―Eigenfaces for recognition,‖

Journal of Cognitive Neuroscience,vol.3, no. 1,pp.71–

86,1991.

[2] I. Daubechies, ―The wavelet transform, time-frequency

localization and signal analysis,‖ IEEE Trans. Information

Theory, vol. 36, no. 5, pp. 961-1005, 1990.

Paper ID: 137

86

Time-Frequency Coherence Analysis of Multichannel Brain Signals Using

Synchrosqueezing Transform

Md. Sujan Ali1,2

, Mst Janntul Ferdous1,2

, Md. Ekramul Hamid1 and Md. Khademul Islam Molla

1*

1Department of Computer Science and Eng., University of Rajshahi, Rajshahi, Bangladesh

2Department of Computer Science and Eng., Jatiya Kabi Kazi Nazrul Islam University, Mymensingh

* [email protected]

KEYWORDS: Brain computer interface, synchrosqueezing transform, time-frequency coherence.

This paper presents a novel approach to analyze

time-frequency coherence between the channels of

electroencephalography (EEG) signals in brain-

computer interface (BCI) paradigm. The brain signal

EEG is captured by spatially distributed sensors placed

on the scalp. The connectivity of different parts of

brain is an interesting study to the BCI research

community. The neural synchronization of human

brain mostly depends on both time and frequency. The

short-time Fourier transform (STFT) and wavelet

transform are generally used to measure the time-

frequency coherence. The EEG signal is usually

nonstationary in nature and hence it is a great challenge

to implement coherence based analysis. The STFT is

considered to solve such problem, it is not entirely

resolved due to the following reasons: i) within each

short-time period the stationarity of neural data cannot

be assured, ii) the resolution of time frequency

representation is restricted by Heisenberg uncertainty

principle. Although wavelet transform is considered as

data adaptive signal analysis method, it uses basis

function called mother wavelet for signal

decomposition and faces time-frequency resolution

problem.

The synchrosqueezing transform (SST) is one of the

techniques based on the continuous wavelet transform

(CWT) that generates highly localized time-frequency

representations of nonlinear and nonstationary signals.

The SST overcomes the limitations of STFT and

wavelet transform based time-frequency (TF)

representation. It represents the perfect localization of

frequency components. The SST reassigns the energies

of CWT such that the resulting energies of coefficients

are concentrated around the instantaneous frequency

curves of the modulated oscillations. The coherence is

computed using auto and cross spectrum derived from

the TF space of the desired channels.

Figure 1: TF coherence analysis (a) synthetic signal X, (b)

synthetic signal Y, (c) STFT based TF coherence between and

(d) SST based TF coherence between signal X and Y

The TF coherence between two EEG channels is

estimated in this work based on the STFT as well as

SST and compared. The performance of the proposed

SST based time-frequency coherence is evaluated on

both synthetic signals and real data. The experimental

results show that the SST based TF coherence

illustrates enhanced resolution than STFT. Two

synthetic signals

X=[sin(2πf1t),sin(2πf2t)],Y=[sin(2πf1t),sin(2πf2t)] and their

TF coherence are shown in Fig 1, where f1=5Hz and

f2=10Hz. There are different time alignment of two

sinusoids to generate X and Y. It is observed that the

proposed method is more efficient than STFT for

localization of frequency components with higher

resolution in coherence domain. The SST based

method represents a sharper coherence frequency

definition at 5Hz and 10Hz than that of STFT. The

coherence of single frequency is present for wide

ranges of adjacent frequencies in STFT due to the

presence of cross-spectral energy. The proposed SST

based coherence estimation method is applied to real

EEG signals of different motor imagery. The

performance is compared with STFT based measure of

TF coherency.

Paper ID: 139

87

Experimental Study on Optical Characterization of

Mono Crystalline Silicon Solar Cell

Nusrat Chowdhury1*, Md. Abdur Rafiq Akand

2, and Zahid Hasan Mahmood

3

1Institute of Energy, University of Dhaka, Bangladesh

2 Solar Cell Fabrication & Research Division, Institute of Electronics, AERE, Bangladesh Atomic Energy

Commission, Savar, Dhaka, Bangladesh

3 Department of Electrical and Electronics Engineering, University of Dhaka, Bangladesh

* [email protected]

KEYWORDS: Surface Photo Voltage, minority carrier diffusion length, carrier life time.

The experimental works have been done on the

optical characterization of mono crystalline solar cell

by Surface Photo Voltage (SPV) measurement for

fabricated solar cell by observing minority carrier

diffusion length (L) & life time (τ). The core objective

of this research was to reduce the cost of solar cell and

increase the efficiency by analysis the optical

characterization. As the PV cell is one kind of photo

diode so the surface photo voltage is very important

part of characterization of cell fabrication. There is a

direct relation between the minority carrier lifetime and

solar cell efficiency. The method of SPV is a

contactless system that helps to analysis the optical

characterization of semiconductors. It measures the

diffusion length of minority carriers in the region of

essential light absorption inside solar cells and wafers.

The minority carrier diffusion length, L is an important

factor for cell efficiency and spectral response of the

mono crystalline silicon solar cell. It is also necessary

for evaluation of the p-type silicon wafer. In this

experiment the surface photo voltage (SPV) of

fabricated solar cells have been analyzed and observed

minority carrier flow by Light Current Voltage (LIV)

tester as it represented the quality of fabrication

process and efficiency of the solar cell. A simple

computer-controlled, normal incidence measurement

system was designed for SPV measurements of

minority carrier diffusion length and lifetime of Si-

solar cell. Measurement system is based on a mini

monochromator driven with a steeper motor to vary

wavelengths in 400-1200 nm spectral range. Light

induced surface photo voltage is measured as a

function of the wavelength. SPV is measured using a

Standard Research 510 lock in amplifier. A LabVIEW

interface is used for system control and data

acquisition. After calculating the experimental data

obtained from mono crystalline silicon solar cells

measurement, minority carrier diffusion length and life

time were calculated and it was 92 µm and almost

3.135 µs respectively. By using solar simulator‘s (Sun

Simulator K3000 LAB55) platform at 25oC efficiency

was measured and it was 16.05%.

This figure (Figure 1) is used to find the value of

diffusion length that is found by its –x intercept and the

value is 92 𝜇𝑚 and minority carrier lifetime becomes

3.135 .

-100 -50 0 50 100 150 2000

20

40

60

80

100

120

Reciprocal of Penetration depth (1/alpha)

Recip

rocal of

SP

V (

1/V

spv)

Figure 1. 1/alpha vs 1/Vspv represents minority carrier

diffusion length

From sun simulator, we have obtained the

efficiency 16.05% ,open circuit voltage 0.606 V and

short circuit current 61.256 (mA/cm2).

The minority carrier lifetime and efficiency are very

important parameter to explain the quality of solar cell.

We observed that diffusion length sometimes small

valued. It happened due to recombination.

Recombination occurs in many reasons. Doping causes

defects and more recombination.

Paper ID: 142

88

Canonical Correlation Analysis for SNP based Genome-Wide Association Studies

Atul Chandra Singha1*

, Arafat Rahman2, Jahangir Alom

3, Md. Nurul Haque Mollah

3

1Department of Statistics, Noakhali Science and Technology University (NSTU), Bangladesh

2Department of Microbiology, Noakhali Science and Technology University (NSTU), Bangladesh

3Bioinformatics Lab, Department of Statistics, University of Rajshahi, Bangladesh

* [email protected]

KEYWORDS: SNPs, GWAS, CCA, Quantitative traits, Outliers, Minimum β-divergence method.

Genome-wide association studies (GWAS) are

powerful tools for measuring the association between

genotype phenotype pairs in bioinformatics. Most of

the human diseases and traits have a strong genetic

architecture. GWAS is successful in identifying

common genetic variants underlying complex traits or

diseases like cancer, type-II diabetes, cardiovascular

disease, schizophrenia and quantitative traits such as

lipid levels and metabolomics. Now an important

approach to GWAS is to test the association between

multiple single nucleotide polymorphisms (SNPs)

against multiple quantitative phenotypes. Canonical

Correlation Analysis (CCA) is one of the most popular

multivariate statistical techniques to test the association

between multiple SNPs against multiple quantitative

phenotypes. However, it is not robust against

phenotypic contaminations. To overcome this problem,

in this paper an attempt is made to robustify the CCA.

To robustify the CCA, we consider some popular

robust analyzers like Minimum Covariance

Determinant (MCD), Minimum Variance Ellipsoid

(MVE), Orthogonalized Gnanadesikan-Kettering

(OGK) estimators including the Minimum β-

divergence estimator. Using simulated data analysis,

we observed that CCA based on Minimum β-

divergence method (proposed) shows better

performance than classical CCA as well as robust CCA

based on MCD, MVE and OGK estimators in presence

of outliers (see Figure 2). Otherwise proposed method

keeps equal performance to the classical CCA as well

as robust CCA based on MCD, MVE and OGK

estimators (see Figure 1).

Figure 1. Performance comparison in absence of

contamination.

Figure 2. Performance comparison in presence of

contamination.

Paper ID: 143

89

Frequency Recognition of SSVEP for BCI Implementation Using Canonical

Correlation Analysis with Adaptive Reference Signals

Shalauddin Ahamad Shuza1, Md. Rabiul Islam

1, Md. Kislu Noman

1 and Md. Khademul Islam Molla

2

1Department of Computer Science and Eng., Pabna University of Science and Technology, Pabna, Bangladesh

2Department of Computer Science and Engineering, University of Rajshahi, Rajshahi, Bangladesh

2 [email protected]

KEYWORDS: Brain computer interface, Canonical correlation analysis, Visual evoked potential.

Brain computer interface (BCI) is a direct

communication pathway between brain and computer

devices without any muscular movement. It takes

signals generated by the brain activity and converts it

into machine readable command. The frequency

components of brain signal for any task may differ in

different time but it is similar to visual flickering

generated by retina excitement. Based on this

technique, steady-state visual evoked potential

(SSVEP) system is developed. Flickering of LCD

(liquid crystal display) with a constant frequency is

used to generate SSVEP based command in BCI

implementation. Canonical correlation analysis (CCA)

uses artificial sinusoidal signals of flickering frequency

of SSVEP. Such reference has features less than real

SSVEP signals. Therefore, the use of real SSVEP as

reference signals will enhance the frequency

recognition performance. In this paper, the recorded

SSVEP signals of already tested trials are successively

included together with the artificial sinusoids to obtain

the adaptive reference set.

The trial in SSVEP means capturing of

electroencephalography (EEG) for a single guess of

visual stimuli for a specific stimulus frequency. The set

of reference sinusoidal signals T

KYYYY ],,[ 21 including H harmonics of

corresponding stimulus frequencies Kfff ,,, 21 is

used in CCA; where

.2cos,2sin,,2cos,2sin HtfHtftftfY nnnnk

If Xn is the nth

trial of SSVEP signal for kth

stimulus,

CCA(Xn,Y) recognize the frequency as fk. In online

BCI, the stimulus frequency of the trial TC

nX (C is the

total number of EEG channels and T is the length of

signal) is recognized before capturing1nX . It is noted

that the trials121 ,,, nXXX are available during the

recognition of Xn. In the proposed method called data

adaptive CCA (AdCCA), an improved reference signal

is derived using real SSVEP data defined

as Tn

T XXYY 11 ,,,ˆ to recognize current

trial Xn, where

C

c iCi cXX1

1 )( . The block

diagram of the proposed method is shown in Figure 1.

Figure 1. Block diagram of proposed method.

To evaluate the performance of AdCCA, 12

different stimulus frequencies of five subjects are used

for SSVEP frequency recognition. Individual subject

generates 15 trials for each of 12 different stimulus

frequencies. The method does not require training data.

The average recognition accuracies over the five

subjects for different time lengths of capturing SSVEP

signals are illustrated in Table1.

Table 1: Average accuracy of CCA and AdCCA

It is observed that the performance of the proposed

method is better than that of the standard CCA for a

wide range of the length (0.5s to 4s) of SSVEP signals.

The underlying reason of improved performance of

AdCCA is that the real references contain more

frequency components than artificial signals. Such

property of the adaptive reference signals (combination

of artificial and real SSVEP) produces higher

correlation while recognizing using CCA. The

processing time of single trial SSVEP of the proposed

method is short enough to implement online BCI.

Methods .5s 1s 1.5s 2s 2.5s 3s 3.5s 4s

CCA 21 45.4 64.9 73.7 78.2 82.3 85.4 87.3

AdCCA 21 50.1 69.7 79 84.1 86.7 90 92.4

Recognized Trial

Y

Xn Recognition by CCA

DerivenX Add

nX toY

Paper ID: 148

90

Expert Reviewer Detection from Online Experiential Product Reviews

Atiquer Rahman Sarkar

Varendra University, Rajshahi, Bangladesh

[email protected]

KEYWORDS: Expert Reviewer Detection, Product Reviews, Experiential Product.

Consumer reviews have emerged as one of the most

influential factors on a person‘s purchase behavior.

Customers are uncertain about the true quality of the

products buy and this uncertainty is greater for

experiential products like movies, books, songs etc.

because repeated purchase of the same product is quite

uncommon for such products. People spend hours

going through numerous reviews to make an informed

decision and to build confidence on the product they

are planning to buy. Several studies have shown that

people cite product reviews as a top influence in their

purchase behavior. In their study, Zhao et al. (2012)

has found evidence of ―stronger learning from product

reviews than learning from own experience‖ and this is

more so to the cases of experiential products.

McGlohon et al. (2010) mentioned that consumer

recommendations are the most credible form of

advertising among 78% of survey responders. Chen et

al. (2011) mentioned a Wall Street Journal report

(2006) that 71% of online U.S. adults use consumer

reviews for their purchases and 42% of them trust such

a source.

Popular products have thousands of user generated

reviews. It is important to detect expert reviewers from

the reviewer community so that users get to know the

credible expert opinions. The usual approaches of

detecting experts/top reviewers are based on

helpfulness vote. The usual approaches suffer from

some limitations and sometimes can be manipulated.

In this paper we address this problem by developing

a framework for expert detection using a combination

of the traditional techniques. Our goal in this work is to

compare the performance of traditional expert finding

techniques against our proposed technique.

Specifically, we detail an algorithm for generating

composite scores of the reviewers. We then use the

scores to rank the reviewers and compare the

performance of the different algorithms against a large

movie reviews data set from the Amazon.com.

Table 1. Performance comparison between the different

traditional methods and proposed approaches.

Met

hod

of

det

ecti

on

No

. o

f re

vie

ws

mad

e

No

. o

f m

ov

ies

cov

ered

Rating using

only reviews

Rating

considering

helpfulness

Av

g. E

rro

r

Av

g. M

SE

Av

g E

rro

r

Av

g. M

SE

Populari

ty 488028 97804 0.73 0.91 0.75 0.94

Differen

ce 407101 94764 0.68 0.79 0.70 0.82

Ratio 14165 10311 0.61 0.58 0.70 0.77

Propose

d1 119176 54621 0.51 0.39 0.56 0.49

Propose

d2 115454 57516 0.53 0.43 0.52 0.43

Propose

d3 110944 56093 0.55 0.46 0.57 0.53

Propose

d4 110441 57874 0.56 0.49 0.56 0.50

Our major contributions are as follows:

We proposed a diverse model for detecting the

experts/top reviewers.

We compared performance of different models

on a large dataset and found our proposed model

superior over the traditional models.

Paper ID: 151

91

Satellite Remote Sensing of Chlorophyll-a Concentration Variation over

Bay-of-Bengal

Md. Monirul Islam* and Md. Mainul Islam Mamun

Department of Applied Physics & Electronic Engineering, University of Rajshahi, Rajshahi, Bangladesh

* [email protected]

KEYWORDS: Chlorophyll-a, MODIS, Bay-of-Bengal.

Chlorophyll-a (Chl-a) is a green pigment and used

as an indicator of phytoplankton biomass. This study is

done to investigate the concentration of Chl-a

distribution pattern over Bay-of-Bengal. Temporal

variations and spatial distributions of Chl-a are

analyzed during the period of 2005 to 2014 to

determine the phytoplankton biomass over the study

area.

The study region is Bay-of-Bengal and located

between 22.50o

and 15.50o

latitude north and 85o

and

95o

longitude east.

The monthly Chlorophyll-a data is used from

MODIS-Aqua (9km) satellite which was processed by

the NASA Ocean Biology Processing Group (OBPG).

Figure 1 shows the Chl-a variation over the Bay-of-

Bengal for the period of 2005 to2014. The white area

in this image is representing the land surface. The

extremely highest Chl-a concentration is found along

the coastal line especially near Bangladesh. The

increased value of Chl-a concentration designates an

increase of the phytoplankton in the study area. The

highest Chl-a concentration is gradually decreasing

towards the offshore.

Figure 1. Annual-mean spatial distribution of Chlorophyll-a

over the study area for the period of 2005-2014.

Figure 2 shows the Chl-a monthly area averaged

time series over the study area for the period of 2005 to

2014. The Chl-a is minimum in April which is rapidly

increasing to a maximum value in July and then

gradually decreasing towards the minimum value.

Figure 3 shows the yearly Chl-a area averaged time

series over the study area for the period of 2005 to

2014. A strong seasonal dependence of Chl-a

concentration is observed with a slight decreasing

trend. The highest peak of Chl-a is approximately 2.00

mg/m3 found in August 2006.

Figure 2. Monthly area averaged time series of Chlorophyll-a

over the study area for the period of 2005-2014.

Figure 3. Chlorophyll-a area-averaged time series over the

study area for the period of 2005-2014.

This study gives a good idea about the

phytoplankton biomass patterns across the Bay-of-

Bengal on both temporal and spatial basis. Satellite

remote sensing is a useful way to monitor the

phytoplankton biomass variations over large scale

temporal and spatial variations.

Paper ID: 152

92

FPGA based Pulse Oximeter using VHDL

Farhana Binte Sufi*, Md. Maruful Islam and Md. Fahmidur Rahman

Applied Physics and Electronic Engineering, University of Rajshahi, Bangladesh

*[email protected]

KEYWORDS: Brain FPGA, VHDL, Pulse Oximeter, Wearable Sensors, Photoplethymysography (PPG)

This paper describes a Field Programmable Gate

Arrays (FPGA) based Pulse Oximeter using VHDL.

Oximetry refers to the determination of the

percentage of oxygen saturation of the circulating

arterial blood.

Oxygen saturation (SpO2) =

Where [HbO2] is the oxygenated haemoglobin

concentration and [Hb] is the deoxygenated or reduced

haemoglobin concentration.

Photoplethymysography (PPG) is a non-invasive

method for the detection of cardiovascular pulse waves

propagating across the human body. Pulse oximetry is

based on the concept that arterial oxygen saturation can

be made using two wavelengths. The Beer-Lambart

law or the spectrophotometric technique applicable to

haemolyzed blood yields best results for Red (650 nm)

and Infrared (805nm) wavelengths.

Reduced haemoglobin (Hb) has higher optical

extinction in the red region (R) of spectrum than

oxyhaemoglobin (HbO2) and lower optical absorption

in the near infrared region. These differences in the

extinction coefficients can be used for the

determination of the light absorbed by Hb and HbO2 by

the normalised ratio (R):

R=

The ac signal is due to the pulsing of arterial blood

while the dc signal is due to all the non-pulsing

absorbers in the tissue.

An oximeter probe using Red and Infrared LEDs

has been used for this project. The amount of absorbed

light for HbO2 and Hb through tissue is measured with

a photosensor and light to frequency converter that

produces the pulsating signals. The pulsating output of

the probe for over a few seconds is then processed in

the FPGA to measure the difference in the pulse

widths, compared and then processed to calculate the

oxygen saturation in blood. For this project Spartan II

FPGA has been used.

IR and Red

LED-Sensor

probe

FPGA Display

Figure 1. Block diagram of the FPGA based Pulse Oximeter.

FPGAs are highly powerful reconfigurable

devicesand FPGA based technology can offer a direct

interface for real-time high speed data acquisition,

processing, monitoring and controlling abilities in

biomedical signal processing and monitoring systems.

FPGAs give greater flexibility than microcontrollers

for DSP solutions, customizable logic and counters

according to design needs so designers have more

freedom of choice. Unlike processors, FPGAs use

dedicated hardware for processing logic and do not

have an operating system (OS).

FPGAs can be used for wearable sensor systems. A

pulse oximeter based on FPGA has the future scope of

combined monitoring with similar other systems (e.g.

spirometers, etc.), thus providing a cheap, portable,

health monitoring system.

Paper ID: 153

93

A Comparative Study of Optimization of Solid State Dye Sensitized Solar Cell

Technologies for Power Grid Integration and Off-Grid Operation

Salman A. Chowdhury1*, Mobasseer M. Hossain

2, and Md. Abu Sayem

3

1Department of Electrical & Electronic Engineering, TMSS Engineering College, Bogra, Bangladesh

2Department of Computer Science & Engineering, Varendra University, Rajshahi, Bangladesh

2,3Department of Electrical and Electronic Engineering, Rajshahi University of Engineering & Technology,

Bangladesh

* [email protected]

KEYWORDS: SSDSSC, DSSC, Nanotube Photonic Crystals, Photosensitizers.

With the advent of DSSCs as a third generation

photovoltaic device in the 1990s pioneered by

O‘Reagan and Grätzel, with usage of low cost

materials and manufacturing processes, a window of

opportunity for low cost implementation of solar power

has been opened.

A nanostructured, mesoporous metal oxide film

which is photosensitized to the visible spectrum of

light using adsorbed molecular dyes. The dye

molecules absorb visible light, exciting the electrons

and inject those from the excited state into the metal

oxide conduction band. The injected electrons travel

through the nanostructured film to the current collector,

and the dye is regenerated by an electron donor in the

electrolyte solution in liquid-based DSSCs and in case

of SSDSSCs, in the several solid hole transporting

materials (HTMs).

In this paper, the performance analyses of Solid

State Dye Sensitized Solar Cells (SSDSSCs) are

carried out with power grid integration and off-grid

operation in the intertropical region being the main

focus. It has been observed that electrodes of different

compositions such as: TiO2 and ZnO, surface

morphology of the thin films used in SSDSSCs, use of

different types of dyes as photo sensitizers, use of

Nanorods/Nanoparticles as photoanodes - all

correspond to different photoelectric conversion

efficiencies (PCEs) in SSDSSCs. As the SSDSSCs are

free of leakage and corrosion related problems

prevalent in conventional liquid-based Dye Sensitized

Solar Cells (DSSCs), it has the potential to lower the

maintenance cost and improving longevity of a system,

so far however SSDSSCs suffer from low PCEs when

compared with conventional DSSCs, with a maximum

efficiency record of approximately 15 per cent set in

2013 by using solid-state mesoscopic TiO2 solar cells

sensitized with lead iodide perovskite (CH3NH3PbI3)

pigments (D. Liu, et al. 2013), in early 2014 which was

further lifted to 16.7 per cent (N.J. Jeon et al., 2014).

Almost all types of solar cells suffer from a

decreased power output when the incident light is tilted

away from normal since the incident intensity generally

follows a cosine law of the incident angle. The focus of

this paper is to optimize the use of SSDSSC in the

intertropical region. It has been demonstrated

experimentally that the use of nanotube photonic

crystals (NT PC) can partially compensate the cosine

power loss of a DSSC.

Figure 1. The calculated ΔJsc (change of normalized short

circuit current) of different types of DSSCs, as a function of

incident angle.

Figure1 shows the enhancements of normalized

short circuit current after using NT PC of varying

thicknesses, where the numbers preceding NT PC

denotes the lattice constants in nanometers.

With low cost and low maintenance SSDSSCs with

built in Solar angle compensation close to commercial

realization, it is this authors‘ understanding that in the

near future such solar cells will play a crucial role in

helping to boost the economy and supply the power

demands of developing intertropical countries around

the world.

Paper ID: 156

94

Microbial Biosynthesis of Gold Nanoparticles using Aspergillus Foetidus and its

Cytotoxicity against A549 Cell

Swarup Roy and Tapan Kumar Das*

Department of Biochemistry and Biophysics, University of Kalyani, West Bengal-741235, India

* [email protected]

KEYWORDS: Fungi, GNPs, FESEM, TEM, A549 cell line, Cytotoxicity.

We study the extracellular biosynthesis of gold

nanoparticles (GNPs) using the fungal species

Aspergillus foetidus. The biosynthesis of GNPs was

carried out by adding 1 mM final concentration of tetra

chloroauric acid (HAuCl4) solution to the fungal live

cell filtrates. The formation of GNPs were initially

monitored by visual observation of physical changes

and then characterized with the help of UV–visible

spectrophotometer, Fourier Transform Infrared

spectroscopy (FTIR), Dynamic light scattering (DLS),

Zeta potential, X-ray diffraction (XRD), Field emission

scanning electron microscopy (FESEM), Energy-

dispersive X-ray (EDX) analysis and Transmission

electron microscopy (TEM). XRD results revealed

distinctive formation of face centered cubic crystalline

GNPs. From FESEM morphology of the nanoparticles

were found to be roughly spherical and within the size

range of 30-50 nm. The spherical and polydispersed

GNPs in the range of 10-40 nm were observed

following TEM analysis.

We also standardized the parameters pH,

temperature and salt concentration, dilution on the

effect of biosynthesis of GNPs. It was established that

alkaline pH, 1mM gold salt concentration and 75 oC

temperature were the respective optimum parameter for

the biosynthesis of GNPs. Cell cytotoxicity of GNP

was compared with that of normal gold salt solution on

A549 cell. The A549 cell growth in presence of GNPs

was found to be comparatively less toxic than the gold

ion.

References:

1. S Roy, TK Das, GP Maity, U Basu, Mat Sci

Eng B, Vol. 203, 41 (2016).

2. R Bhambure, M Bule, N Shaligram, M

Kamat, R Singhal, Chem Eng Technol. 32,

1036 (2009).

3. VC Verma, SK Singh, R Solanki, S Prakash,

Nano Res Lett. 6, 16 (2011).

4. S Lokina, R Suresh, K Giribabu, A Stephen,

RL Sundaram, V Narayanan, Spectrochimica

Acta A 129, 484 (2014).

Paper ID: 157

95

Adsorptive Removal of Cr(VI) From Aqueous Solution onto Charred Vetiver Root

Bhoj Raj Poudel and Surendra K Gautam*

Dept. of Chemsitry, Tribhuvan University, Nepal

*[email protected]

KEYWORDS: Semiconductor NPs, UV spectra, Transmission Electron Microscopy (TEM).

Adsorption properties of Charred Vetiver Root

(CVR) for heavy metal have been investigated with

chromium as adsorbate from aqueous solutions. The

adsorption of hexavalent chromium [Cr(VI)] onto CVR

was investigated by varying various parameters viz.

contact time, pH, amount of adsorbent and initial

concentration of adsorbate. The maximum adsorption

capacity of Cr(VI) onto CVR was found to be 412.50

mg/gm at pH 1. The adsorption kinetics followed

pseudo-second order and adsorption isotherm followed

Langmuir model. From the repeated adsorption-

desorption test it was seen that the efficiency of CVR

to adsorb Cr(VI) from solution remains almost same.

Thus, CVR was found to be cost effective bio-

adsorbent for efficient removal of Cr(VI) from polluted

water.

Figure 1. Determination of λmax of Cr(VI)

Figure 2. Calibration curve for the determination of Cr(VI)

DPCI complex.

The present research work investigated, by making

simple chemical modification of the vetiver grass root

using charring process, an effective bioadsorbent can

be made for the removal of Cr(VI) metal ion from

aqueous solutions than that of raw vetiver grass root.

The λmax value was found at 530 nm. The maximum

adsorption capacity of Cr(VI) onto CVR was found to

be 412.50 mg/gm at optimum pH 1. The required

equilibrium time for the adsorption of Cr(VI) onto

CVR was found to be 240 minutes. The adsorption

kinetic followed pseudo-second order kinetic model

and adsorption isotherm followed Langmuir isotherm

model.

Paper ID: 158

96

Study on Determining the Average Size and Structure of Nanoparticles

Deval P Bhattarai and Surendra K Gautam*

Dept. of Chemsitry, Tribhuvan University, Nepal

* [email protected]

KEYWORDS: Semiconductor NPs, UV spectra, Transmission Electron Microscopy (TEM).

Semiconductors are the foundation of modern

electronics, including transistors, solar cells, light

emitting diodes (LEDs), quantum dots etc. Researchers

have studied semiconductor nanoparticles intensely and

have developed them for broad applications in solar

energy conversion, optoelectronic devices, molecular

and cellular imaging, ultrasensitive detection, targeted

therapy etc. All applications are size dependent and

their crystalline structures play vital role. Synthesizing

nano-sized semiconductor particles and stabilizing

structure is always challenging. Group II-VI

semiconductor NPs can be synthesized by various ways

as chemical deposition, biomimic, sol-gel,

microemulsion, microwave irradiation, spray pyrolysis

etc. The crystalline structure with phase purity can be

settled using XRD data and SAED pattern. The average

particle sizes are determined with the help of UV

spectra and XRD data which are further verified from

TEM images.

Figure 1: Plot of

Various methods are used for synthesizing Group II-VI

semiconductor NPs. The phase purity crystalline

structures are settled from the XRD and SAED

patterns. The average particle sizes are calculated from

UV-Vis spectrum and TEM image.

))( 2

gVsEh

Paper ID: 159

Extended Abstract Book International Conference on Computer ...

Documents

Transcript of Extended Abstract Book International Conference on Computer ...