english covervol17 no1a - hashemite university · for a reference to an article in a periodical: [...

ISSN 1023-0149

Yarmouk University Publications Deanship of Research and Graduate Studies

ABHATH AL-YARMOUK

Basic Sciences and Engineering

Refereed Research Journal

Vol. 17 No. 1A 1429/2008

ISSN 1023-0149

ABHATH AL-YARMOUK

Basic Sciences and Engineering

Refereed Research Journal

Vol. 17 No. 1A 1429/2008

© Copyright 2008 by Yarmouk University All rights reserved.

No part of this publication may be reproduced without the prior written permission of the Editor-in-Chief.

Opinions expressed in this Journal are those of the authors and do not necessarily reflect the opinions of the Editorial Board or the policy of Yarmouk University

Proof Reading:

Prof. Ibrahim Jibril, Chemistry Dept. Yarmouk Universty

Typesetting and Layout

Manar Malkawi and Fatima Atrooz

Cover Design

Ahmad Al-Tamimi

EDITORIAL BOARD

Editor-in-Chief Professor Abdul Rahman S. Attiyat

Members

Professor Omar Rimawi Professor Mashhoor Al-Refai

Professor Mohmmad Abu-Saleh Dr. Mohammad Sheboul

Professor Hamed Zraiqat Dr. Mohammad Belal Al-Zoubi

Editorial Secretary

Kamilia Adel Al-Haj

In the Name of God Most Gracious Most Merciful

Abhath al-Yarmouk: is a refereed research journal Basic Sciences and Engineering

Abbreviated: A.Y. (Basic Sci. & Eng.)

NOTES TO CONTRIBUTORS Only original unpublished articles will be considered. LANGUAGE: Manuscripts may be written in Arabic or English.

FORMAT: Manuscripts should be submitted in quadruplicate and typed double-spaced, on one side of the paper only, and with 2.5cm margins. The size of the paper should be (27x21.5cm).

• The total number of pages on the manuscript should not exceed 30 pages, including figures, illustrations, references, tables, and appendices.

• Each manuscript should be accompanied by two abstracts, one in Arabic and one in English, of approximately 200 words each, typed on two separate sheets.

• The title of the manuscript, the author’s name, academic rank and affiliation should appear on a separate sheet. The title should also appear on the first page of the manuscript and on each abstract.

• Manuscripts should be sent in print and on a floppy 3.5 computer disk compatible with IBM (Ms Word).

FIGURES AND ILLUSTRATIONS: • Figures and illustrations should be drawn on tracing paper. Their dimensions

should not exceed 19 x 12cm.

FOREIGN NAMES: When foreign names are mentioned in an Arabic manuscript, they should be written in Arabic, followed by the English transcription in parentheses.

DOCUMENTATION: A) Documentation of published references: This should be done within the text

by numbers [5], rerence should be listed at the end of the manuscripts.

For a reference to a book: [ ] Franson N., Clesceri L., Greenberg A. and Eaton A.(eds.), Standard methods

for the examination of water and wastewater, 20th Edition, American Public Health Association, Washington, D.D., USA.(1998).

For a reference to an article in a Periodical: [ ] Smith P. D., 'On tuning the Boyer-Moore-Horspool string searching algorithm',

Software-Practice and Experience, 24(4) (1994) 435-436. For a reference to an article in a book:

[ ] Gordan A.J. & Ford R.A., A Handbook of Practical Data, Techniques, and References, Properties of the Elements. Canada: John Wiley and Sons, Inc, 1972.

B) Documentation of notes and unpublished references: This should be done within the text by writing the word “note” followed by the succession number of the note in brackets, as follows: (Note 1). Then every note is explained in further detail at the end of the manuscript, before the references, under the title Notes, as follows:

Note 1: Al-Nijjar, Tariq H., Levels of Trace Elements in Freshwater fishes of Azraq Oasis-Jordan, Unpublished M.S Thesis, Yarmouk University, 1991.

BOOK REVIEWS: Book reviews of recent academic publications may be considered for inclusion in the Journal.

EDITORIAL CHANGES: The Editor-in-Chief reserves the right to make any editorial changes he deems necessary.

OFFPRINTS: Twenty offprints will be sent free of charge to the principal author of the published manuscript, in addition to one copy of the journal issue in which the manuscript is published.

SUBMISSION OF MANUSCRIPTS AND CORRESPONDENCE: Manuscripts and Correspondence should be sent to:

The Editor-in-Chief Abhath al-Yarmouk (Basic Sciences and Engineering)

Deanship of Research and Graduate Studies Yarmouk University, Irbid, Jordan

SUBSCRIPTION INFORMATION: ABHATH AL-YARMOUK may be obtained through the Exchange Division of the Yarmouk University Library or through the Deanship of Research and Graduate Studies at JD 1.000 per single copy. Annual subscription rates for individuals and institutions: Jordan: JD2. 500; Arab World: JD 8.000 or US$ 12.00; countries outside the Arab World: US$ 18.00 or equivalent.

TABLE OF CONTENTS Articles in English

* Automatic Vehicle Location Software Model and Geographic Information System

Mohammad Qatawneh and Yahya Aref

1

* Harmonic Elimination using Six-Step Symmetry Pulse Width Modulated Waveforms

Ibrahim A. Altawil and Fathi K. Amoura

13

* Profile Measurement using Intensity-Based Saw-Tooth Structured Light Fringe Patterns

Sami Al-Hamdan and Husam Hamad

27

* Effect of Iron Addition on the Magnetization Measurements on the Normal State Y123- Superconducting System

Khitam Khasawinah, Abdel-Rauf Al-Dairy and Yaqoub Hamam

43

* Temporal Growth of Filementation Instability Induced by Large Amplitude Laser Radiation

Mohammad Bawa'aneh, Saud Al-Awfi and Husam El-Nasser

51

* Indoor Radon-222 Concentration Measurements in Some Houses in Dura District During the Winter Season of the Year 2000

Amin Leghrouz, Mohammad Abu-Samreh and K. M. Awawdah

63

* Automatic Quantitative Bladder Measurements using MRI

Abdalmajeid Musa Alyassin 77

* Linear Combination of SZÃSZ-Baskakov Type Operators for (LP)-Approximation

Kareem J. Thamer 93

* Novel Rearrangement in Tertiary Amine N-Oxides. Part II Thermal Rearrangement of Acetylenic Amine Oxides

Abdul-Hussain Khuthier Sharba

105

* Geochemistry of Some Opaque Mineral Oxides from Andesitic Dike, Southern Jordan

Ahmad Al-Malabeh, Tayel El-Hasan, Qaher Al-Qadi and Hasan Al-Fugha

111

* Reducing the Iron Content of Hiswa Clay Deposits, South Jordan by Chemical Leaching

Talal Al-Momani 127

* Biomarkers from Oil Shales in The Upper Cretaceous Amman Formation, NW Jordan

Hatem Darwish and Hakam Mustafa

147

ABHATH AL-YARMOUK: "Basic Sci. & Eng." Vol. 17, No.1A, 2008, pp. 1-11

Automatic Vehicle Location Software Model and Geographic Information System

Mohammad Qatawneh* and Yahya Aref**1

Received on June 8, 2006 Accepted for publication on Dec. 4, 2006

Abstract Due to the huge advancement in technology and especially in the world of wireless

communications, vehicles tracking systems are more and more used these days for deferent purposes, but mainly related to tracking vehicles. The objective of this work is to design a general data model for AVL software and to define the GIS data model to be used in AVL system. Many works are already done in the AVL software design, but its still needs to be validated for the national level. AVL is an integration of modern technologies such as Global positioning system (GPS), Geographic Information System (GIS) and wireless communications. The system is based on receiving the signals from satellites by GPS receiver, which is connected to a Global System for Mobile communication (GSM) modem, and both are fixed on the tracked vehicle. The function of the designed hardware is: first to acquire the position data from the satellite (GPS) and then, to send it by a Short Massage Service (SMS) or General Packet Radio Services (GPRS), to the computer monitor at the control center. GPRS or SMS carries certain information about the geographic position of a certain vehicle to the control center efficiently and according to the needed timings, which could be varied upon request. After the reception of the data, the vehicle position will be displayed at the control center computer using AVL software. In this work, we are interested to point out the importance of the three parts of AVL software and determine the minimum database part to satisfy the local needs in fleet management in Jordan. The paper also shows the importance of building AVL system under GIS environment. There are some locally designed tracking system; provide an efficient tool of control over any fleet of moving vehicles and the capabilities of the well-known international tracking systems at a low price. It offers as well, an easy access to change in functions of the system, and a great possibility of customization to fulfill the local needs. The validation of local softwares to satisfy the proposed model is very important issue.

Keywords: Automatic Vehicle Location (AVL), Geographic Information System (GIS), Global Positioning System (GPS), Global System for Mobile Communication (GSM).

Introduction: Tracking system is a cooperation of three important modern technologies the GIS,

GPS and wireless communications [1]. The first, (GIS) provides the map part of the

© 2008 by Yarmouk University, Irbid, Jordan. * Department of Computer Science, King Abdulla II School for Information Technology University of Jordan,

Amman, Jordan ** Department of Computer Science, Al_Zaytoonah University of Jordan, Amman, Jordan.

Qatawneh and Aref

2

display and the second; (GPS) calculate the position of the moving target on the map, third wireless communication is a media for data transmission between the control base station and vehicle [2]. The cooperation of the three technologies could be seen in Figure-1.

Figure 1 - Schematic diagram of Vehicle tracking system [3]

The main objective of this work is to establish the main relations of different database used in AVL system (the GIS database and the entity database). These relations are very important to design good AVL software to be able to generate the main reports for fleet managements. Also the importance of media is very important considering the cost differences between GSM and GPRS technologies.

We need to use GIS database for the display of the surrounding environment and streets of the geographic site for the application purposes and for the validation and the verification of the tracking software. Then, GIS database implementation for the Roads Network in Jordan was the first and necessary output of AVL system. An application of the software was tracking vehicles at different types of environments, National scale to local scale such as going from main street in Jordan through large cities and get into university Campos.

The Global Positioning System (GPS) is actually a constellation of 24 Earth orbiting satellites (Figure 2). The U.S. military developed and implemented this satellite network as a military navigation system, but opened it up to the private sector [4]. Each of these solar powered satellites circles the globe making two complete rotations every


3

day. The orbits are arranged so that at any time, anywhere on Earth, there are at least four satellites "visible" in the sky.

Figure 2 - GPS Solar powered Satellites Circles the globe

Vehicle tracking and navigational systems made tracking system technology and GPS applications a necessity for increasing number of users. Today's, GPS fitted cars; ambulances, fleets and police vehicles are common sights for such applications. Automatic Vehicle Locating System (AVLS), Vehicle Tracking and Information System (VTIS), Mobile Asset Management System (MAMS), these systems offer an effective tool for improving the operational efficiency and utilization of vehicles [5].

Many research projects show the importance of AVL system in intelligent transportation world wide and show the advantages and benefits of such system, particularly for better customer services [5].

GPS is used in vehicles for both tracking as well as user navigation. Tracking systems enable a base station to keep track of the vehicles without the intervention of the driver while navigation system helps the driver to reach the destination. Tracking systems are development of GPS based navigation; computer and communication. Vehicle tracking systems can provide position and navigation information quickly and accurately at relatively low cost. Tracking systems are two types depending upon the application [6].

In the above contest, we will determine the basic AVL model and the GIS data needed to build the AVL software without going into the complication of GIS analysis and programming. The minimum requirements of the GIS database will help the local

Qatawneh and Aref

4

and small companies to start to use such system in the limited fleet management, such as student transportation.

AVL Components The designed Tracking System hardware consists of the following [7]:

a. A GPS Unit: the position of a vehicle is captured through GPS unit; this unit is installed in the tracked vehicle and connected with transfer unit (GSM modem).

b. Data storage unit: It is required to store the captured data by the GPS unit.

c. Data transfer unit: Captured or Stored data has to be communicated to a central control station for analysis. Network technology used for communication depends upon the mode of tracking the vehicle. The data can either be transmitted instantaneously after capturing as in active mode, or may later be downloaded as in passive mode.

d. Data analysis technologies: Data stored at the control center can be analyzed through a software application. A GIS mapping component is developed with features for tracking the application

The tracking kit in each vehicle consists of a GSM modem and a GPS receiver and controller and antenna. It requires a SIM card like any mobile phone to work on the GSM cellular telephone network. It uses only data and short messages over the network [8]. The tracking operation requires a control center to communicate with the tracking kits. Computer server, wireless modem requires a regular SIM card, communication software to control the communication between the center modem and the car kits.

The system operation in the control center can communicate with each vehicle by sending and receiving short messages (SMS) of GSM network through the wireless modem attached to his mode of operation.

The GPS unit inside the car kit takes care of updating the location data by periodically calculate the data received from the satellite every one second. Therefore, the car kit can be configured to send its location every certain period of time to the control center. This period can vary form one minute to one hour or more.

Software and GIS data Models The AVL system will help public and private entities to organize and control their

fleet (buses, trucks and cars). The control center will know the velocity and the position of each vehicle any time any where. In the universities this will optimize the transportation services for the student's from- to-University.

In this work we will give the main component of AVL software and the GIS data model to be used in such software. There are many software packages that already exist in Jordan and these softwares should contain and follow these models to satisfy the local needs.


5

GIS data Model The main part of this model is the definition of minimum needs of GIS database to

build a cost/benefits GIS model to be used in the AVL software. An example the main local needs of any private university is to determine the best route between university campus and the residential zones of students with all database related to a certain bus and its route. The base relations should be determined clearly between the different databases used in the AVL system as shown in (Figure 3).

Figure 3 - The main relations between AVL database System

There are some steps that should be followed to create GIS data model for AVL services. The following flowchart describe the main blocks to build a GIS for tracking system (Figure 4). The flow chart also shows the procedure of building GIS from National scale to feature details such as main roads in Jordan (Figure 5), to small or sub ways at university compus. Generally, satellite images are used to update the different GIS layers (Figure 6), such as main highways and sub ways.

Qatawneh and Aref

6

Figure 4 – Flowchart of GIS data model for AVL software

The work procedure of designing the GIS model starts by:

1) The Scanning of maps process then,

2) The Registration process followed by,

3) The Digitizing processes and,

4) Finally the Editing process (Figure 4) of the resulted layer maps.

Any GIS project consists of layers viewing the geometry of feature which related with tables contain the attribute about the entire feature [9].


7

Figure 5 – Jordan street layer (vector data) to be loaded in AVL software

Figure 6 – AL- Zaytoonah University 2.5 SPOT Image (raster data)

Qatawneh and Aref

8

Software data Model The AVL software should be composed from three parts:

1. Communication part: it deals with the transmission of data from/to vehicle, via modem installed at the base-station. The data could be sent via SMS to a wireless modem installed at base station or via GPRS at fixed IP address.

2. Data base part: Access, Oracle and SQL-server could be used in the software to organize and manage the database. The design of data base is very important to facilitate the generation of different type of reports. Figure 7 shows the most important relations between the various tables needed to design the AVL database, which gives powerful of the system to generate the various reports needed to analyze and manage the fleets.

3. Map part: this part should be based on the GIS objects or controller software, which gives large advantage in the programming of map based software and deals with raster and vector maps.

The processing engine of the software should be able to connect the communication part, database part and the GIS data together. And also generate the different reports types. Finally, the processing engine of the software should be responsible of the security of different data.

Figure 7 - SQL-Server database relationship for AVL software


9

The cost of existing softwares in Jordan are varies from one thousand dollar to several thousands of dollars. This depends substantially on the technology used to develop such softwares. The web version is generally has higher price than desktop version. The characteristics of a good AVL software and the advantages of one compared to another are:

1- Loading the vector and raster data.

2- Processing the satellite images such as Quick bird and Ikonos.

3- Different geofencing type: fence on the street, fence by street selection, fence by street name, fence by polygons, fence on the district and geofence on point.

4- Give the nearest land mark to the driver.

5- Sent SMS to the driver mobile.

6- Complete report types such as Stop start report, geofence violation report, daily report and analysis of data.

7- GPRS and SMS capabilities.

8- Independent of the hardware.

9- English Arabic interface.

The desk top software gives large advantages for small companies to control their fleet with low costs at high security. Naturally, we have to consider the running cost of wireless services. During the test of one AVL software we have notice a considerable variations of cost for data transmission (SMS and GPRS), between different mobile service provider in Jordan. As example the cost of 1kb is 0.01JD, where you can send more than 10 position location via GPRS, but the cost of the same data is 0.3JD via SMS.

Results and Conclusions The AVL system creates many opportunities for Fleet Management Systems and it

has large impact on transportation services and security. Local AVL systems will help different entities to start to use AVL and to get the benefits of the system at reduced cost. Development of the AVL software is an easy process under GIS softwares. Because the grate advantages of the determination of the main parts of the software reported in this work, but the main difficulties is to find the low cost of the car kit unit and to have a good services mobile provider at low cost.

SQL database design is already implemented and programmed and is ready to be connected with any software. Also the communication part is built independently of the hardware to be used with any tracking carkit unit.

The simple GIS data model proposed is related to the streets, their attributes and the important land marks, but for database design its very important step to be able to mange the huge data and for generation of different reports.

The main feature of the AVL system are :

Qatawneh and Aref

10

1) The car kit or hardware: installed in the tracked Vehicle. It contains a GPS receiver, GSM/GPRS module and Microcontroller based hardware to control the moving vehicle location and sending the position.

2) A base station with a GIS software which contains Digital Maps (Raster or Vector), of the area. It should communicate with the vehicle through SMS/GPRS and have a database that should handle all the tracked Vehicles at the same time.

النموذج الربمجي لنظام تتبع الحافالت وأنظمة املعلومات الجغرافية

محمد القطاونة و يحيى عارف

ملخص

يهدف هذا البحث لوضع األطر األساسية لبناء نظام تتبع الحافالت باستخدام أنظمة المعلومات المالحة العالمي لتحديد الجغرافية ونظام المالحة العالمي واالتصاالت الالسلكية، حيث تم استخدام نظام

الموقع الجغرافي للحافلة ومن ثم إرساله باستخدام وسائل االتصاالت الالسلكية مثل شبكة الخلوي إلى الخادم الرئيسي حيث يتم تحليل البيانات المرسلة باستخدام نظام المعلومات الجغرافية إلظهار الموقع على

ناصر الرئيسة لبناء نموذج لنظام المعلومات الجغرافي في هذا البحث تم تحديد الع. الخارطة الرقمية . المستخدم في أنظمة تتبع الحافالت وتحديد األجزاء الرئيسة إلنشاء هذا النظام

References

[1] Al-Bayari O. and Sadoun B., 'New centralized automatic vehicle location communications software system under GIS environment', International Journal of Communication System, 18 (9) (2005) 833-846.

[2] Strathman J. G., Kimpel T. J., Dueker K. J., Gerhart R. L., Turner K., Taylor P., Callas S. and Griffin D., 'Service Reliability Impacts of Computer-Aided Dispatching and Automatic Vehicle Location Technology: A Tri-Met Case Study', Transportation Quarterly, 54 (3) (2000) 85-102.

[3] Business, (2006), available at:

http://webinfosearch.com/Top/Business/Telecommunications/ Location_and_Tracking/Vehicle/Fleet_Management_Services

[4] Hofmann-Wellenhof B., Lichtenegger H. and Collins J., 4th 'Global Positioning System: Theory and Practice', New York: (1997).


11

[5] Sivaram CMSL and Kulkarni M. N., 'GPS-GIS Integrated system for transportation engineering' , Map India 2001 International Conference, CSDMS, NOIDA, India (2001)

[6] Premal M., and Pavitra A., Project Manager , GPS based fleet management system: Different alternatives, [email protected]

[7] Spring G.; Collura J. and Black K., 'Evaluation of AVL Technologies for Para-transit in Small and Medium Sized Urban Areas', Journal of Public Transportation, 1 (4) (1997) 71-83

[8] Pradeep S. Kingh Kharola MD. BMTC, Bipin Gopalkrishna, 'Geographical Information System and Fleet Management in an Urban Transport Corporation', Indian Journal of Transport Management, 23 (6) (1999) 353-361.

[9] Keith C. K., Third edition 'Getting Started With Geographic Information Systems', Printice –Hall, Inc., USA, (2001).


Harmonic Elimination using Six-Step Symmetry Pulse Width Modulated Waveforms

Ibrahim A. Altawil and Fathi K. Amoura* 1

2

Received on March 29, 2007 Accepted for publication on Nov. 13, 2007

Abstract Selective harmonic elimination technique to synthesize single-phase and three-

phase Pulse Width Modulation (PWM) waveform using six-step symmetry is presented in this paper. The six-step PWM technique inherently generates no triplen, no even harmonic and eliminates non-triplen odd harmonic to the desired order. This means that the dc bus voltage is utilized to the maximum. It also provides a balanced three-phase system at any instant.

Keywords: Harmonics, Pulse Width Modulation (PWM), Voltage Source Inverter (VSI), Selective Harmonics Elimination, Six-Step Symmetry, Newton-Raphson numerical analysis technique.

1. Introduction The performance of voltage source inverters and the quality of their outputs are

seizing a wide area of most progressive research field in the last two decades. A large number of techniques for reducing the total harmonic distortion (THD), harmonics elimination, and reducing the switching loss have been reported [1-7]. Elimination of harmonics, particularly the lower-order ones results in a high-quality output. This in turns leads to a reduction in the current ripples, torque pulsation, and improving the over all system performance [8-12]. Therefore harmonics elimination is the key issue in optimizing the system performance. Two techniques have been used for harmonics elimination. One is based on the symmetry of the quarter of the sine wave that spans from 0o to 90o[2]. The other is based on the folding symmetry of the sine wave that spans from 0o to 60o. The second is called six-step PWM modulation technique [3]. In both techniques, a programmed PWM schemes are used, where the switching angles are pre-calculated, based on approximate model of the switching angles trajectories, and stored in a look-up table in read-only memory (ROM) [4]. The interpolation between the values

© 2008 by Yarmouk University, Irbid, Jordan. * Power Engineering Department, Hajjawi Faculty of Engineering Technology, Yarmouk University, Irbid,

Jordan.

Altawil and Amourah

14

in the look-up tables, to provide quasi-continuous voltage control, is associated with complexity [5]. To avoid memory-based implementation to the PWM firing angles,

several studies have been reported [6][13]. One of the techniques is called area equalization of the PWM. It determines the pulse-width in each sample interval by making the area of the inverter output equal to the area of the sinusoidal reference voltage. This technique has been compared with the programmed PWM technique reported in [2]. The comparison shows a superiority of the programmed PWM technique; because it can produce high-quality waveforms and high voltage gain at a reduced switching frequency. The advance of computer technology and the fast power electronic devices makes the programmed PWM technique more attractive to be used on-line in real time to provide a good dynamic response at comparatively low cost.

The purpose of this paper is to investigate and characterize a PWM technique, for single-phase and three-phase inverters. It is based on six-step PWM technique to eliminate selected harmonics. The six-step PWM technique inherently generates no triplen, no even harmonic and eliminates selected odd harmonics to the desired order. In this technique the dc bus voltage is utilized to the maximum. It provides an output voltage peak up to 1.0pu, with the absence of the triplen harmonics a balanced three-phase system at any instant is guaranteed. This technique is particularly important in single-phase voltage synthesis since the triplen harmonics are inherently absent otherwise they would have to be eliminated individually. Newton-Raphson method is used to solve the non-linear simultaneous transcendental equations (10). In addition an investigation of the range of the harmonics elimination and the switching angle trajectories are established.

2. The principles of six-step symmetry

Figure (1) shows three 120o phase-shifted cosine functions F1 (θ), F2 (θ) and F3 (θ) from –60o to 360o. Since the cosine function is an even function, the waveforms are divided into 60o intervals. The segment of F1 (θ) from 0 to 60o is represented by function f (θ), the segment from -60o to 0 represented by f (-θ),

Where;

ooff 6060)()( ≤≤−−= θθθ (1)


15

-60 0 60 120 180 240 300 360-1.5

-1

-0.5

0

0.5

1

1.5f(-θ) f(θ) f(-θ+120) f(θ-120) f(-θ+240) f(θ-240) f(-θ)

F1(θ)

F2(θ)

F3(θ)

-f(-θ+60) -f(-θ-60) -f(-θ+180) -f(-θ-180) -f(-θ+300) -f(-θ-300)

0

Figure (1) Waveforms of three-phase 120o phase shifted cosine and the relationship between various segments of cosine function

Figure (1) shows how other segments are represented by phase shifting of f (θ), and f (-θ). For example, the segment F2 (θ) from 60o to 120o is the negative of that of f (θ) between 0oand 60o shifted by 60o. Similarly, the segment of F3 (θ) from 60o to 120o is that of f (θ) between –60o and 0o shifted by 120o.

Evaluating the function description of a zero-crossing segment such as F1 (θ) from 60o to 120o, or F2 (θ) from 0o to 60o can be done by using the property of the three phase system where;

F1 (θ)+F2 (θ)+F3 (θ) = 0 (2)

For example, the function description of a zero crossing segment F1 (θ) from 60o to 120o can be computed from those of the other two segments F2 (θ) and F3 (θ) from 60o to 120o, that is -f (-θ+120)+ f (θ-60). Using the above technique the entire function F1(θ) can be described as

Altawil and Amourah

16

≤≤−≤≤+−+−−≤≤−−≤≤+−−≤≤−++−−

≤≤

=

360300)(300240)300()240(

240180)180(180120)180(12060)60()120(600)(

)(

0

1

θθθθθθθθθθθθ

θθ

θ

fff

ff

fff

F

oo

o

(3)

Examining equation (3) shows that one-sixth interval of the waveform is used to describe the entire waveform. Therefore, the function F1 (θ), as defined by equation (3), possesses a six-step symmetry. Placing notches in f (θ) at predetermined angles s'α is employed in this work to eliminate the desired number of harmonics with fundamental amplitude control.

3- Six-step spectrum analysis Figure (2) shows a single-phase inverter configuration, the output voltage Va-b is to be synthesized using the Fourier expansion. The analysis gives the amplitude of the fundamental frequency.

Figure (2) Single phase full-bridge inverter

The function F1 (θ) as defined by equation (3) contains the six-step symmetry. The angles are usually defined for one basic waveform segment from which the PWM waveform is constructed based on symmetry. In this technique the notches are placed on one of the symmetrical folding segment of the waveform, i.e. 0o to 60o period, where, The angles are constrained such that,

S1

S4

S3

S2

Vin

Va-b

D1

D4


17

oM 600 321 ≤≤≤≤≤≤ αααα LL (4)

Figure (3) defines the required variable angles s'α , when the total number of angles M in the 60o segment is even or odd values. The value of M leads to different equation for Fourier coefficients

(a)

(b)

Figure (3) Sketch of the switching angles s'α for (a) M is even (b) M is odd

Assuming the number of lower-order harmonics that can be eliminated is N. The required angles s'α equals M= 1+N. Thus, M equations are required to find the M variables. The coefficients of the Fourier expansion for F1 (θ), in equation (3), are even functions, and contain only cosine terms. Therefore, F1 (θ) can be written in the Fourier expansion form as,

∑∞

=

=1

1 )cos()(n

n nAF θθ (5)

The coefficient An is given by

1

0

)(θf

2α 3α 4α Mα 1α 60 2−Mα 1−Mα

1

0 2α 3α 4α 1−Mα Mα 1α 60

Altawil and Amourah

18

∫=π

θθθπ

2

01 )cos()(1 dnFAn (6)

Using equations (1) and (3) and reducing all the integrals to the same limit (0 to π/3) An becomes as

∫ ++−−=3

0

))]3

cos()(cos()1(1)[((8π

θπθθθπ

dnnnfA nn (7)

From the above equation An is equal to zero when n is even and Mn 3= , where L7,5,3,1=M . Thus, the even and triplen harmonics are inherently eliminated by

this method. For L,4,3,2,1)16( =±= qwhereqn . An becomes as

∫ +=3

0

)6

)(cos(()6

(cos8 π

θπθθππ

dnnfnAn (8)

Substituting for f (θ) reduces An to a single equation as

(9)

oddMMevenMM

NM

nnnnn

AM

kk

kMn

:,23:,13

1

)6

sin()1()6

sin()1()6

cos(81

−−

+=

+−−−= ∑

=

παπππ


19

Therefore, equation (9) can be expressed as shown in equation (10) for M=9, and An is defined as a modulation index, its value varies between 10 ≤≤ nA

0)6

sin()1()6

sin()1()6

cos(8)(

0)6

22sin()1()6

2sin()1()6

2cos(28)(

0)6

sin()1()6

sin()1()6

cos(8)(

1

12

11

1

=

+−−−=

=

+−−−=

=−

+−−−=

∑

∑

∑

=

=

=

M

kk

kMM

M

kk

kM

M

kk

kM

MMMMM

f

f

Af

παπππ

α

παπππ

α

παπππ

α

M

M (10)

3. Newton-Raphson numerical analysis technique Equation (10) is a set of nonlinear equations such that the switching angles ( Mαααα ,,,, 321 LL ) cannot be computed directly. Newton-Raphson numerical technique is used to obtain the solutions. The steps of using numerical technique for solving the set of equations are as follows:

a) Assign a set of initial values 001 to Mαα

TM

][ 003

02

01

0 ααααα LL= (11)

b) Calculate 00 )( ff =α

Where [ ]TM

fffff 003

02

01

0 L=

c) Using Taylor’s series and ignoring the Hessian matrix (second order derivatives) and higher derivatives, the solution of equation (10) to determine the M variables ( s'α ) should be in the following form to fit the Newton-Raphson technique (equation 12), where [J] is the Jacobian matrix

Altawil and Amourah

20

[ ] =− Tff 0

∆

∆∆∆

∂∂

∂∂

∂∂

∂∂

∂∂

∂∂

∂∂

∂∂

∂∂

MM

MMM

M

M

fff

fff

fff

α

ααα

ααα

ααα

ααα

M

M

LL

M

M

LL

LL

3

2

1

0

2

0

1

0

02

2

02

1

02

01

2

01

1

01

(12)

In short form [ ] [ ][ ]α∆=∆ Jf T

d) The left hand side and the Jacobian in the right hand side of equation (12) can be determined from the first previous guess of T

M][ 00

302

01

0 ααααα LL=

e) Inverting the Jacobian the correction needed to converge can be calculated T

M][ 321 αααα ∆∆∆∆ LL

f) Repeat (a) to (e) until [ f∆ ] converges to a very small value in equation (12) equals or below 0.01

4. Simulation results Newton-Raphson method requires a suitable initial guess to guarantee convergence.

For each case (M=9 or 8. etc.,) a good starting guess is selected at modulation index of 0.5 where initial switching trajectories are calculated assuming equal intervals. These results are used as an initial guess for modulation indexes of 0.4 and 0.6 then solutions proceed for higher and lower modulation indexes using as initial guess the results of the previous ones. The angles trajectories in Figure (4 and 5) were obtained by solving equations (10) using Matlab software for M= 9 and 8, respectively.

Figure (4a) shows the variations of trajectory angles versus the modulation index (or the fundamental p.u. voltage magnitude). The selected eliminated harmonics are 5th,

7th, 11th, 13th, 17th, 19th, 23rd and 25th.. Figure (4b) shows the variations of the amplitudes of the first three harmonics (29th, 31st and 35th) normalized to the fundamental peak.


21

Figure (4a) The variation of the switching angles versus the Modulation index for M=9

Figure (4b) The variation of the first three significant harmonics with the normalized amplitude for M=9

Figure (5a) shows the variations of trajectory angles versus the modulation index (or the fundamental p.u. voltage magnitude). The selected eliminated harmonics are 5th, 7th, 11th ,13th, 17th, 19th and 23rd . Figure (5b) shows the variations of the amplitudes of the first three harmonics (25th, 29th and 31st) normalized to the fundamental peak.

Altawil and Amourah

22

Figure (5a) The variation of the switching angles versus the Modulation index for M=8

Figure (5b) The variation of the first three significant harmonics with the fundamental peak

5. Simulation results for six-step PWM Referring to the configuration of single-phase inverter shown in Figure (2). The

process of harmonic elimination for a selective harmonics involves placing notches in the PWM waveforms at calculated angles as explained in the procedure in section 3. To eliminate N harmonics at any modulation index, (N+1) angle is required to define the position of the notches. These angles are used to define one basic waveform pattern segment, which is over a period of 600 as described before, then the entire waveform

signal is constructed based on symmetry. Figure (6a) shows the Va-b voltage of a single–phase inverter, for M=9. Where, 5th, 7th, 11th, 13th, 17th, 19th, 23rd and 25th harmonics are eliminated. Figure (6b) shows the results of the spectrum analysis of the waveform shown in figure (6a), which assures the expected elimination of the selected harmonics


23

mentioned above. Table 1 lists the numerical values of the magnitude of Fourier series coefficients as percentage of the fundamental amplitude.

Figure (6a) Waveform for the output voltage when the selective harmonics are eliminated

Figure (6b) Fourier spectrum for the output voltage, when the selective harmonics are eliminated M = 9

Table 1 Harmonics and their magnitude as percentage of the fundamental for

modulation index equal 1 and the frequency =50Hz % of

fund Freq Order % of



fund Freq Order 0.002 1400 28 0.010 950 19 0.001 500 10 1.006 50 1 0.322 1450 29 0.002 1000 20 0.002 550 11 0.001 100 2

0.003 1500 30 0.001 1050 21 0.002 600 12 0.000 150 3

0.250 1550 31 0.001 1100 22 0.011 650 13 0.001 200 4

0.002 1600 32 0.009 1150 23 0.000 700 14 0.002 250 5

0.000 1650 33 0.001 1200 24 0.001 750 15 0.001 300 6

0.001 1700 34 0.107 1250 25 0.001 800 16 0.009 350 7

0.043 1750 35 0.002 1300 26 0.006 850 17 0.001 400 8

0.001 1800 36 0.002 1350 27 0.001 900 18 0.001 450 9

Altawil and Amourah

24

The same procedure is applied to the three- phase inverter shown in Figure (7). Simulation results using six-step PWM are obtained as shown in Figure (8), for Vab, Vbc, and Vca and the selected harmonics are eliminated for M=9.

Figure (7) Three-phase full-bridge inverter

Figure (8) Waveform for the output voltage of three-phase inverter when the

selective harmonic 5th, 7th 11th 13th 17th 19th 23rd 29th 31st are eliminated

S1

S4

S3

Vin

S2 S6

S5

A B C


25

6. Conclusions Newton-Raphson technique is developed to determine the firing angles of inverter

output voltage using six-step PWM. Spectral analysis shows elimination of triplen, even and odd harmonics up to the desired degree. Angles vary almost linearly which means smooth voltage control. Results in this paper at 50Hz match with results described in [3] at 60 Hz. Therefore, this technique can be applied to both systems. Newten–Raphson method described in this paper shows powerful and fast calculating technique.

الغاء التوافقيات باستخدام التماثل السداسي الخطوات للموجة املعدلة طوليا

ابراهيم الطويل و فتحي عمورة

ملخص

لقد تم في هذه الورقة دراسة طريقة إلغاء توافقيات مختارة لتشكيل موجات معدلة طوليا لطور السداسي للموجة المعدلة طوليا ان التماثل . واحد او ثالثة اطوارباستخدام التماثل السداسي للموجة

يتضمن عدم توليد توافقيات مضاعفة ثالثيا وكذلك التوافقيات الثنائية ويمكن ايضا الغاء التوافقيات ان هذا يعني ان قضيب الجهد الثابت قد تم استحدامه . االحادية الغير مضاعفة ثالثيا ألي درجة مطلوبة

. ثالثي االطوار متزن في اي لحظةوهذه الطريقة تزودنا بنظام. للدرجة القصوى

References [1] Domigan A. Jr, and Embriz-stintandar, ”Harmonics Mitigation Techniques for

Improvement of Power Quality of Adjustable Speed Drives”, Proceeding, IEEE Applied Power Electronics. Conf., (1990) pp. 96-105

[2] Enjeti P. N., Ziogas P. D., and Lindsay J. F., “Programmed PWM Technique to Eliminate Harmonics, Critical Evaluation”, Proceeding, IEEE Industry Application Soc. Conf ( 1988) pp. 418-430.

[3] Ajay Mahoshwari and Khai D.T. Ngo, “Synthesis of Six-Step Pulse Width Modulation Waveforms With Selective Harmonic Elimination”, IEEE Transaction on Power Electronics, 8 (4) (1993).

[4] Sidney R. Bowes, and Paul R. Clark,” Transputer-Based Harmonic Elimination PWM Control of Inverter Drives, IEEE Transaction on Industry Application, 28 (1) (1992) 72-80.

Altawil and Amourah

26

[5] Sidney R. Bowes, and Paul R. Clark,” Simple Microprocessor Implementation of New Regular-Sampled Harmonic Elimination PWM Technique”, IEEE Transaction on Industry Application, 28 (1) (1992) 90-95.

[6] Toufiq J.A, Millitt, B., and Goodmann, M. A., “Novel Algorithm for Generating Near Optimal PWM Waveforms for Ac Traction Drives,” IEE Pro. Pt. B, 133 (2) (1986) 85-94.

[7] Ramshaw R.S, et al., “ PWM Inverter Algorithm for Adjustable Speed Ac Drives Using Non-Constant Voltage Source,” IEEE Trans. On Industry Application, IA (1986) 673-677.

[8] Chiasson J. N. , Tolbert L. M. , McKenzie K. J. and Du Z. “A complete solution to the harmonic elimination problem,” IEEE Trans. Power Electron., 19 (2004) 491.

[9] Kato T., “Sequential homotopy-based computation of multiple solutions for selected harmonic elimination in PWM inverters,” IEEE Trans. Circuits Syst. I, 46(1999) 586.

[10] Xie Y.-X. , Zhou L. and Peng H. “Homotopy algorithm research of the inverter harmonic elimination PWM model,” Proc. Chin. Soc. Elect. Eng., 20( 2000) 23.

[11] Agelidis V. G. , Balouktsis A. and Cosar C. “Multiple sets of solutions for harmonic elimination PWM bipolar waveforms: analysis and experimental verification,” IEEE Trans. Power Electron., 21(2006) 415.

[12] Krein P. T. , Nee B. M. and Wells J. R. “Harmonic elimination switching through modulation,” Proc. IEEE Workshop Comput. Power Electron., (2004) p. 123.

[13] Wells J. R., Geng X. Chapman P. L. , Krein P. T and Nee B. M. “Modulation-Based Harmonic Elimination”, Power Electronics, IEEE Transactions on, 22 (2007) 336-340.


Profile Measurement using Intensity-Based Saw-Tooth Structured Light Fringe Patterns

Sami Al-Hamdan and Husam Hamad* 1

Received on April 2, 2006 Accepted for publication on Nov. 19, 2006

Abstract In this paper, a new technique for creating a 3D map for objects is presented. A structured

light of saw-tooth intensity variation with some inclination angle is projected on the object surface to be measured. The proposed method relies on three main elements; first, a good quality projector to project the saw-tooth fringe pattern on the surface of the object under test. Second, a high quality digital CCD camera to capture the reflected image, and finally, a PC which is interfaced to both the projector and the camera, with the necessary software to generate the necessary fringe pattern and to analyze the captured image to reveal the 3D profile of the object surface. Some test objects are generated using MATLAB and the method is simulated under the assumption of noise-free environment.

Keywords: Fringe analysis, Profilometry, 3D Mapping, Contour measurement, Shape measurement

1. Introduction Three-dimensional surface contouring techniques have numerous applications. For

example, automobile parts whose dimensions need to be checked against their design specifications can be subjected to surface contouring techniques to validate its geometry. Museums can use 3D contouring techniques to establish a permanent digital heritage preservation centre such as the pilot project aimed at the creation of lasting digital 3D models for archival purpose at the “Museo dell’Opera del Duomo” in Florence [1]

Object profile measurement has long been in use for various reasons such as inspection, navigation, or process control. It has gone by several names: 3-D mapping [2], topography [3], profilometry [4], depth mapping [5], 3D-Shape [6], and many others.

In the past, measuring the profile of objects was mainly dependent on contact methods where a probe touching the object surface is used to scan the whole area under

© 2008 by Yarmouk University, Irbid, Jordan. * Department of Computer Engineering, Hajjawi Faculty for Engineering Technology, Yarmouk University,

Irbid, Jordan.

Al-Hamdan and Hamad

28

inspection. Of course, such methods where inherently slow, not very accurate, and only suitable for solid objects with no fear of destruction, i.e. the scratching of the object surface by the probe was not a matter, something that is very annoying when it comes to measuring a highly-polished lens for example.

The need for non-contact non-destructive accurate measuring techniques has led for the development of fringe projection methods. These methods are based on generating a fringe pattern usually of sinusoidal intensity variations and projecting this pattern on the object surface then capturing the reflected image and finally extracting the surface profile. Several techniques for analyzing fringe patterns have been developed, a good review for these methods can be found in [7,8]

The analysis of fringe patterns can be classified into two main categories: intensity-based analysis [3,7,9,10] which were the early techniques, and phase-based analysis which became the preferred choice in recent years [7,11,12].

The most common phase-based fringe pattern analysis are the Phase-Stepping method and the Fourier Transform Method (FTM) [7,8,13,14]

The phase stepping method requires a highly accurate optical setup. In such a system, several fringe patterns are captured such that a certain phase difference between them exists. On the other hand, the Fourier Transform Method uses sophisticated and intensive computational algorithms to analyze the reflected image generated from the projection of a single sinusoidal fringe pattern [15].

Figure 1: Optical setup for profile measurements


29

The method presented here is a return to the old intensity-based techniques but with a new approach. The idea of this method relies on the ability of producing good quality linear fringes with a saw-tooth light intensity function that are projected on the surface of the object under measurement with certain inclination, see figure 1. The reflected image from the surface is then stored on a frame grabber by a CCD camera located at right angles with the surface on which the object is laid on. A computer to produce the height distribution then processes the image.

The theoretical analysis (see next section) shows that this method is very simple from the point of view of computations and programming, thus making real time realization on ‘simple’ computer systems feasible. The optical setup and equipment needed for this method are also quite simple and cheap if compared with other methods. For example, a Data-Show could be used to project a computer generated saw-tooth shaped ‘linear’ fringe pattern on the object under measurement and a CCD camera to collect the reflected image for analysis.

Recently, the use of CCD cameras, LCD projectors, and PC’s has been adopted by several researchers for non-contact profilometry [16…18]

2. Theory Assume that it is possible to produce and project a linearly spaced saw-tooth fringe

pattern that when projected on a flat surface the reflected fringes will have a period P and a light intensity slope A. (figure 2)

Assuming the flat surface is coinciding with the x-y plane, the z-axis coincides with the reflected rays from the surface, and that the light intensity is changing along the x-axis only, the light intensity distribution function I at a point (xi,yj) on the reflected image can be expressed by the following equation:

I(xi,yj ) = A*( xi – KP) + Imin

Al-Hamdan and Hamad

30

Figure 2: Saw-Tooth light intensity variation

K=0,1,2…represents the period number, P is the period of the saw-tooth function, and A is its slope and is given by : A = (Imax – Imin)/P (see figure 2)

Assume now an object has been laid on the flat surface such that the height difference between X1 and X2 along the X-axis is equal to δH (see figure 3). The reflected light intensity at X2 is the same as that of X3 if the object was not present. i.e. if the light ray continued its path until it reached the flat surface at X3. So the light intensity at X2 can be written as follows :

I(X)

X

Imax

0

Imin

Slope =A

Period = P

Y

X


31

I(X2) = I(X1) + A * (X3 – X1) ………………………………..……………(1)

Figure 3: The effect of heights on changing the reflected fringe pattern shape

Using triangulation in figure 3 it can be proved that

δH = β (α δI - δX). ……………………………………………….…………….(2)

Where β=1/tan(θ),α=1/A , δI=I(X2)–I(X1) , δX = X2-X1. For complete derivation of equation 2, see [19].

It should be clarified that δH is calculated between two adjacent points that are laying on the same fringe period or on two adjacent periods. When the light intensities are measured at two points laying on two adjacent periods of the fringe pattern, as shown in figure 4, a correction has to be made on δI to produce the true light intensity difference. This is explained next.

Camera

θ

δH

X1 X2 X3

I(X1)

I(X2)

Saw-Tooth fringe

pattern

Al-Hamdan and Hamad

32

The correct δI for the case shown in figure 4 is given by:

δI = [Imax – I(X1)]+[I(X2) – Imin]

Rearranging,

δI = [Imax– Imin] + [I(X2) – I(X1)]

let D = Imax – Imin, then:

δI = D + [I(X2) – I(X1)]

A correction has to be made for calculating the correct intensity difference δI by adding D to the measured difference whenever the two points are on two consecutive periods of the captured fringe pattern. This occurs when I(X2)<I(X1).

3. Simulation To verify the correctness of the method two object surfaces are simulated using

MATLAB. The first surface to be simulated was a part of a sphere surface. A 512x512 pixel fringe pattern with a fringe period of 32 pixels was used. As it can be noticed in figure 5, the method was very successful in reconstructing the original object shape. Although a shift in the object heights is observed in the reconstructed surface, this is not a problem since we can always allow the captured image to include a reference plane whose reconstructed height is subtracted from all the points on the reconstructed surface.

I(X)

X

Imax

0

Imin

I(X1)

Figure 4

I(X2)

X1 X2


33

To see how accurate is the method, the difference in heights between the original and the reconstructed object was calculated in the middle of the two surfaces that is at line 256 of the fringe pattern. As seen in figure 5 and figure 6 the maximum absolute error was around 0.6 which is around 2% of the maximum height in the middle of the object. It has to be mentioned here that the units of measurements are pixels. In true-life measurements, a magnification factor has to be calculated to convert pixels into true height measurement units.

Simulated spherical object

50 100 150 200 250 300 350 400 450 500

50

100

150

200

250

300

350

400

450

500

The corresponding fringe pattern

Figure 5: Simulation of spherical object

Reconstructed object surface from the fringe pattern

0 100 200 300 400 500 6000

0.1

0.2

0.3

0.4

0.5

0.6

0.7

absolute error with offset removed

Figure 6: Reconstruction of the spherical surface

Al-Hamdan and Hamad

34

The second surface to be simulated is a moderately changing surface. Again, the method presented here was successful in reconstructing the object surface from the corresponding fringe pattern. In this simulation, the fringe pattern period was 64 pixels instead of 32, (see figure 7 and figure 8)

Simulated roof object

50 100 150 200 250 300 350 400 450 500

50

100

150

200

250

300

350

400

450

500

Corresponding fringe pattern

Figure 7: Simulation of a surface that looks like a roof

The MATLAB script for generating and simulating the roof surface is provided in the appendix at the end of this paper.

Reconstructed roof object

Absolute error

Figure 8: Reconstruction of the roof object


35

4. Discussion Equation 2 above was the basis for the computer simulation programs used to

reconstruct the two simulated object surfaces. In the above simulations, we assumed the following:

The variation of the projected fringe pattern was along the x-axis only

The reflectivity of the object surface is homogeneous all over the surface.

Ambient light noise is not present.

Very high quality projector and CCD camera are used.

The method measures the change in height between two adjacent points along the x-axis, which will result in correct height distribution for each line. However, if all lines start from the same reference plane then the overall height distribution for the whole surface will be correct. Otherwise, if for example the object or part of it covers the whole image area then equation 2 will give correct height distribution for points in each line alone but not all points in the image. One solution to this problem is perhaps to project a second linear fringe pattern that has its light intensity distribution increasing linearly from top to bottom and then compute relative height distribution for one line (column), i.e. fixing X at a constant value and changing Y. X could be in the middle of the image for example. This way the relative height distribution of each horizontal line is known relative to its neighbors, and thus a correction for all height distribution can be made for all points in the image to give the correct height distribution of all points relative to each other.

In our theoretical analysis, earlier ideal conditions were assumed. Unfortunately, in real life, noises from different sources add the captured image. Reflection from the object surface is usually non-uniform producing unwanted irradiance variation, which would certainly affects the accuracy of measurements and therefore some measures have to be taken to reduce this effect.

To enhance the process of producing the fringe pattern and the quality of the capture images some of these measures could be used: Painting the object by white color for example, capturing the images on a dark room, and using some image processing algorithms. Preprocessing of the collected image is also necessary to reduce the effect of noise and to enhance the image quality for processing.

It should be pointed out that δH is assumed not large enough to cause cycle slipping of one or more fringe periods. Therefore, this method would be suitable for surfaces with small variations between adjacent points. From a theoretical point of view, this method should be capable of measuring sharp variations on the object surface as long as δH fulfills the condition mentioned above and that these variations are facing the projected light beams as shown in figure 9 below.

Al-Hamdan and Hamad

36

Figure 9: Blind area problem caused by the shadow

The blind area shown in figure 9 is generated because no light beams fall on it. Although the profile of the left hand side of the object in figure 9 has a sharp slope it will not cause problems for the calculations - as long as δH fulfils the criterion of no cycle slipping mentioned earlier.

A solution for the blind area effect mentioned above is perhaps to do the analysis twice, one with the projected fringe pattern facing the left hand side of the object and another one with the fringe pattern facing the right hand side of the object and then combining the two calculations to correct for blind areas errors.

It is worth mentioning also that reducing the angle of projection θ will reduce the effect of shadow. However, reducing θ would reduce the light intensity sensitivity to height variations, i.e. we would like to have bigger δI for small δH something that is achieved by increasing the projection angle θ. Therefore, a compromise has to be made. In practice, viewing angles between 10 and 30 degrees are usually used.

As compared to other methods on the field of fringe projection techniques for non-contact measurements, This method requires less calculations, for example, the very popular Fourier transform method, requires a 2D forward and inverse Fourier transforms for the captured image to be calculated, furthermore, it requires other calculations such


37

as filtering, phase extraction, and phase unwrapping steps. These calculations are more time consuming than those required by this method.

5. Conclusions A new method for non-contact object profile measurements has been proposed.

The idea of the method relies on the ability of producing good quality linear fringes that are projected on the surface of the object under measurement with certain inclination. The reflected image from the surface is then stored on a frame grabber by a CCD camera located at right angles with the surface on which the object is laid on. The image is then processed by a computer to produce the height distribution.

The mathematics for this method is quite easy and the programming of the method is relatively simple. Future work will focus on practical aspects related to the ideas presented in this paper.

The method was tested through simulation. Two object shapes have been generated, and the corresponding fringe pattern was simulated. The fringe pattern was then analyzed and a 3D map for each object was calculated. The method proved to be working. However, in real life a lot of problems may arise that would require additional steps in preprocessing the captured fringe image.

لضوء ذي الشدة املتغرية على قياس طبوغرافية االجسام باستخدام انماط من ا سنان املنشارشكل أ

سامي فتحي الحمدان و حسام احمد حمد

ملخص

تعتمـد هـذه الطريقـة علـى . تعرض هـذه الورقـة طريقـة جديـدة إلنتـاج خارطـة ثالثيـة األبعـاد لألجـسام دته على شكل أسنان المنشار ومـن ثـم إسـقاط هـذه الـصورة المتولـدة إنتاج أنماط من الضوء الذي تتغير ش

ــة ــنمط المــنعكس عــن الجــسم المــراد قيــاس أبعــاده . بزاويــة معين وتحليــل هــذه ، بعــد ذلــك يــتم تــصوير التفتــرض الطريقــة تــوفر ثــالث عناصــر . الــصورة بواســطة الكمبيــوتر الســتخراج األبعــاد الثالثيــة لهــذا الجــسم

وأخيـرا جهـاز حاسـب مـرتبط ، كاميرا رقمية عاليـة الجـودة ، ثانيا، جهاز عرض عالي الجودة ، أوال: رئيسية، المختلفــة مــن كتابــة بــرامج خاصــة للحاســب لتوليــد األنمــاط ال بــد أيــضا. بكــال مــن جهــاز العــرض والكــاميرا

مـن صـحة لعمـل بعـض المحاكـاة للتثبـت MATLABتـم اسـتخدام .ولتحليل الصور المنعكسة عن الجسمافترضت المحاكاة عدم وجود ضوضاء تؤثر علـى جـودة وصـحة الـصور . الطريقة المعروضة في هذه الورقة

.الملتقطة

Al-Hamdan and Hamad

38

References [1] Beraldin J.A., Atzeni C., Guidi G., Pieraccini M., and Lazzari S.,” Establishing a

Digital 3D Imaging Laboratory for Heritage Applications: First Trials” Proceedings of the Italy-Canada 2001 Workshop on 3D Digital Imaging and Modeling Applications, Padova, Italy. April 3–4. (2001).

[2] Pennington T., “Miniaturized 3-D Mapping System Using a Fiber Optic Coupler as a Young’s Double Pinhole Interferometer”, Ph.D. Dissertation, Faculty of the Virginia Polytechnic Institute and State University. (2000).

[3] Rowe S. H. and Welford W. T., “Surface Topography of Non-Optical Surfaces by Projected Interference Fringes” , Nature,216 (1967) 786-787.

[4] Indebetouw G., “Profile Measurement Using Projection of Running fringes”, Applied Optics 17 (18) (1978) 2930-2933

[5] Jain R., Kasturi R. and Schunk B., “Machine Vision”, McGraw Hill Inc., New York, 1995.

[6] Garbat P. and Kujawiñska M., “Combining Fringe Projection Method of 3D Object Monitoring with Virtual Reality Environment: Concept and Exemplary Algorithms”, 1st International Symposium on 3D Data Processing Visualization and Transmission (3DPVT 2002), Padova, Italy, Jun 19-21, 2002.

[7] Reid G., “Automatic Fringe Pattern Analysis: A Review”, Optics and Lasers in Engineering , 7 (1986/87) k37-68,

[8] Robinson D. and Reid G., “Interferogram Analysis”. , IOP Publishing Ltd (1993). [9] Tsuruta T and Itoh Y., “Interferometric Generation of Counter Lines on Opaque

Objects”, Optics Communications, 1(1) (1969) 34-36.

[10] Yatagi T., “Intensity Based Analysis Methods”, In Robinson D. and Reid G., editors of : “Interferogram Analysis, Digital Fringe Pattern Measurement Techniques”, Institute of Physics Publishing, Bristol (1993) Ch4 pp 72-93.

[11] Creath K , “Phase-Measurement Interferometry Techniques”, In E. Wolf, Editor of : “Progress in Optics”, 26 (1988) 350-393.

[12] Creath K, “Temporal Phase Measurement Methods”, In David W. Robinson and Graeme T. Reid, editors of: “Interferogram Analysis, Digital Fringe Pattern Measurement Techniques”, Institute of Physics Publishing, Bristol, (1993) Ch4 pp 94-140.

[13] Takeda M., In, H. and Kobayashi S. “Fourier-transform method of fringe-pattern analysis for computer-based topography and Interferometry”, Journal of the Optical Society of America ,72(1) (1982) 156-160

[14] Takeda M. and Yamamoto H., “Fourier-transform speckle profilometry: three-dimensional shape measurements of diffuse objects with large height steps and/or spatially isolated surfaces”, Applied Optics 37(16) (1998) 3385-3390.


39

[15] Al-Hamdan S., “A Comparison of Two Parallel Computer Architectures in the context of Interferometric Fringe Analysis”, PhD theses, Liverpool John Moores University, (1996).

[16] Tay C., Quan C., Fu Y., Chen L., and Shang H., “Surface profile measurement of low-frequency vibrating objects using temporal analysis of fringe pattern”, Optics & Laser Technology 36(2004) 471-476.

[17] Ailing T. ,Zhuangde J and Yanqun H., “A flexible new three-dimensional measurement technique by projected fringe pattern” Optics & Laser Technology, 38 (2006)585 –589.

[18] Quan L., Yuanhao C. “Fringe contrast-based 3D profilometry using fringe projection”, Optik 116 (2005)123 –128

[19] Al-Hamdan S and Hamad H. “Non-Contact Profile Measurements Using Linear Fringe Patterns” SSCCII-2004 International Symposium of Santa Caterina on Challenges in the Internet and Interdisciplinary Research, Amalfi, Italy Jan 29 – Feb 1, (2004).

Al-Hamdan and Hamad

40

Appendix

%The following script will simulate a roof object and %will produce 4 figures; first the simulated object, %second the simulated fringe pattern, third, the %reconstructed object, and fourth the absolute error %(note the error should be recalclculated by subtracting %the reference plane height)

phi = input('car roof inclination in degrees');

phi=phi*pi/180;

tanphi=tan(phi); % inclination of object surface

H1=zeros(1,256);

for i=100:200

H1(i)=(i-100)*tanphi;

end

H1(201:256)=100*tanphi;

H2=fliplr(H1);

H1= [H1 H2];

H1=[H1;H1;H1;H1;H1;H1;H1;H1];

H1=[H1;H1;H1;H1;H1;H1;H1;H1];

H1=[H1;H1;H1;H1];

H3=zeros(512);

[N,M]=size(H1);

H3(256-(N/2):256+(N/2)-1,:)=H1;

[N,M]=size(H3);

H2=H3(1:16:N,1:16:M);

R1=1:512;

R2=R1(1:16:N);

figure(1);

colormap ([1 1 1]);


41

surf (R2,R2,H2);

axis equal;

P=input ('enter fringe period in pixels : ');

Imax=input ('enter maximum light intensity 0<Imax<255 : ');

Imin=input('enter minimum light intensity 0<Imin<255 : ');

If (Imax<=Imin)

Error ('Imax should be greater than Imin');

end

theta=input ('enter angle of projection in degrees : ');

if ((theta >30) | (theta<10))

error('theta should be between 10 and 30');

end

tanth = tan(theta*pi/180);

DX=1; % the scanning is done pixel by pixel,

DI=Imax-Imin; %

a=(Imax-Imin)/P; % slope of light intensity

FP=ones(size(H3))*Imin; % initialise the fringe pattern with

% minimum light intensity

for i=1:N

for j=2:M

DH = H3(i,j)-H3(i,1); % difference in height between

% adjacent pixels

DX=j;

FP(i,j)= fix(FP(i,1)+a*(DX + tanth*DH));

NF=floor(FP(i,j)/DI);%NF number of fringe cycles

FP(i,j)=FP(i,j) - NF*DI +Imin;

end

end

Al-Hamdan and Hamad

42

figure(2);

C=0:1/255:1;

map=[C' C' C'];

colormap(map);

image(FP);

% reconstruct (HR is the array of the reconstructed surface)

HR=zeros(N,M);

Beta=1/tanth;

alpha=1/a;

for i=1:N

for j=2:M

di=FP(i,j)-FP(i,j-1);%light difference between 2 pixels

if(di<0)

di=di+DI;

end

DH = Beta*(alpha*di-1);

HR(i,j)=HR(i,j-1)+DH;

end

end

figure(3)

colormap ([1 1 1]);

HR2=HR(1:16:N,1:16:M);

Surf (R2,R2,HR2);

axis equal;

err=H3(256,:)-HR(256,:);

figure(4)

plot(err);


Effect of Iron Addition on the Magnetization Measurements on the Normal State Y123-

Superconducting System

Khitam Khasawinah, Abdel-Rauf Al-Dairy and Yaqoub Hamam*1

Received on Sept. 12, 2006 Accepted for publication on Feb. 5, 2007

Abstract Magnetic properties of Y1Ba2Cu3O7-δ superconducting system with Fe added in different

weight percentage have been studied versus the applied magnetic field for different temperatures above the critical temperature, Tc . As expected, the Y1Ba2Cu3O7-δ compound showed nonmagnetic behavior for all temperatures and all fields. The addition of iron caused friction in that system above Tc. A phase transition at about 260K was observed in the susceptibility in the 1% and 4% samples. This could be due to the formation of an anti-ferroelectric or ferroelectric states associated by the oxygen deficiency in some off-center positions of the crystal structure.

Keywords: High temperature superconductors, Magnetization, Doping, Magnetic friction.

Introduction High temperature superconductors are classified as type II superconductors. Doping

processes are used to enhance the critical current density, the critical temperature Tc, and other properties of superconducting materials [1-5]. Y-based superconductors are the most widely studied systems. High Tc superconductors are reported to have fewer electrons per unit volume in the normal state. The grain boundary thickness is greater than or equal to the coherence length so that the barrier penetration is more difficult at weak links and the Cooper pairs can be broken [6].

In the superconducting state, the vortices do not move freely because they are pinned at some pinning centers and a very small Lorentz force is needed to remove a vortex from its pinning center [6]. If the Lorentz force is smaller than the pinning force then the current flows without any dissipation. But if the Lorentz force is large enough, then the vortices move and electrical resistance appears as the transport current exceeds the critical current value, Ic , which can be used as a measure of the pinning force.

It is very helpful to understand superconductivity if we can understand the normal state properties for a superconductor. In addition, preparation conditions like grinding, © 2008 by Yarmouk University, Irbid, Jordan. * Department of Physics, Yarmouk University, Irbid, Jordan.

Khasawinah, Al-Dairy and Hamam

44

compressing, and heat treatment have major effects on their properties. These can help in improving contacts and help in optimizing O2 content which has major effect on the critical temperature and critical current density.

There are several reported articles concerning the properties of the superconducting material in the normal state [7-12]. The effect of iron addition on Y-based systems was previously studied for temperatures below Tc= 92K and an increase of vortex pinning that improves Jc were reported [4, 5, 13,14]. In this work we have prepared Y-based samples and Fe was added in different percentages. The magnetization measurements at various temperatures below Tc are reported in details before [13]. The magnetization measurements versus temperature and various applied fields for different iron content at various temperatures above Tc will be reported and discussed.

Sample Preparation and Experimental Procedure The samples were prepared by the solid-state reaction method using Y2O3, BaCO3,

CuO, and Fe3O4 high purity powder oxides as starting materials. First, we prepared YBa2Cu3O7-δ superconducting system (Y123) from Y2O3, BaCO3, CuO high purity oxides. We divided the material into four samples. Iron in the form of Fe3O4 was added to three samples of Y123 as a percentage ratio of the weight with 1%, 2%, and 4%. Detailed sample preparation was reported before [13].

A small amount of each sample was ground into powder for X-ray diffraction and a diffraction pattern was obtained. Scanning electron microscope (SEM) micrographs were taken for the samples containing iron (1%, 2%, and 4%) and these results were reported before [13,14].

In this work we will focus on the effect of increasing the content of Fe atoms in the Y123 superconductors by studying the magnetic properties of these samples above Tc. The magnetization measurements were carried out using a Vibrating Sample Magnetometer (VSM). The measurements were recorded at different temperatures from 100K to 300K. The magnetic field applied was varied at a rate of 25 Oe/s from 0-8 kOe.

Results and Discussion Room temperature XRD spectra showed that the well defined peaks are almost the

same as those of the orthorhombic structure of the superconductor Y123. The addition of iron caused a slight shift in the position of these peaks. Table 1 shows a comparison between the reported lattice parameters [15] and the calculated ones from the XRD data. SEM data showed presence of grains with their size increasing with increasing iron content which appears very clear for the 2% Fe sample [13,14].

The magnetic properties of the sample above Tc were analyzed and compared with the undoped sample Y123. The measurements of the magnetization, M, versus applied field H for the sample with 0% iron for various temperatures above Tc showed a nonmagnetic behavior at all fields at all temperatures.

Effect of Iron Addition on the Magnetization Measurements on the Normal State Y123- Superconducting System

45

Table1: The lattice parameters for samples with different iron percentages .

Samples studied in this work Lattice parameters

Reported parameters for

0% [15]

0% 1% 2% 4%

a(Ǻ) 3.82

3.8258 3.8263 3.8046 3.8343

b(Ǻ) 3.88

3.8884 3.87735 3.8796 3.8749

c(Ǻ) 11.67

11.6737 11.8044 11.6817 11.4056

Figure (1) shows M versus H for the sample 1% for various temperatures. It showed a diamagnetic behavior at all fields at all temperatures below 260K. It started to show some paramagnetic behavior at low fields for temperatures higher than 260K. We calculated the magnetic susceptibility χ(T) for the 1% sample at low fields (below 1500 Oe) and these values are shown in Figure (2). It is clear that the sample always showed a diamagnetic behavior at temperatures below 260K and high fields.

Figure 1: Magnetization versus applied field for 1% Fe sample at different temperatures.

-0.4

-0.2

0

0.2

0 2000 4000 6000 8000

300280270265260240220200180160140120100

H(Oe)

M(e

mu/

g)


46

Figure 2: The magnetic susceptibility, χ, versus the temperature for 1% Fe sample.

Figure (3) shows M versus H for the 2% Fe sample at different temperatures. The behavior of M versus H is nearly linear diamagnetic behavior for all temperatures .The magnetic susceptibility, χ(T) at each temperature, for this sample was of the order (-5x10-7emu/g Oe). Figure (4) shows M versus H for the sample 4% for various temperatures. It showed, for all temperatures, a dominating paramagnetic behavior at low fields up to about 1500 Oe . After reaching some maximum it showed a diamagnetic behavior for all fields at all temperatures.

Figure 3: Magnetization versus the applied field at different temperatures for 2% Fe sample.

-2

0

2

4

100 150 200 250 300

fe1

T(K)

χ(10

-5)e

mu/

g.O

e

-0.0060

-0.0045

-0.0030

-0.0015

0

0.0015

0 2000 4000 6000 8000

300280260240220200180160140120100

H(Oe)

M(e

mu/g)


47

Figure 4: Magnetization versus applied field for 4% Fe sample at different temperatures.

Figure (5) shows χ (T) at law fields (about 1500 Oe) versus T for the sample with 4% which shows a phase change starting at temperatures near 250K. This appears clearly in Figure (6) which shows maximum magnetization, Mmax, calculated from the magnetization data, versus T.

Figure 5: The magnetic susceptibility, χ, versus the temperature for 4% Fe sample.

-0.002

0

0.002

0 2000 4000 6000 8000

300280260240220200 180160140120100

H(Oe)

M(e

mu/

g)

1

3

5

7

150 250

Fe 4

T(K)

χ(10

-7)e

mu/

g.Oe


48

Figure 6: Maximum magnetization, Mmax, versus the temperature for 4% Fe sample.

The magnetization, M, and the magnetic susceptibility, χ(T) behaviors versus the temperature, T, for the samples with 1% and 4% iron showed a phase transition near 260K, which may be caused by the formation of an anti ferromagnetic states associated with the ordering of the atoms. For the 4% sample we observed a phase transition near 200K but it was weaker than the one near 260K.

For the samples with 2% and 4% iron content, the magnetization was very weak and of the same order of magnitude of the magnetization for the sample with 0% iron. This could be explained, using previously reported SEM micrographs for these samples which suggested that Fe prefers to agglomerate in zones without replacing atoms in the crystal of the superconducting phase ( namely copper atoms ) for high percentages of iron added. In fact, it was much smaller compared to the magnetization of the sample with 1% iron content which was two orders of magnitude higher than others. This appeared clearly when we calculated χ for these samples. These results suggested that the percentage of iron added to samples should not exceed 1% as had been reported before [13].

Conclusions The normal state of Y1Ba2Cu3O7-δ compound for temperatures above Tc is

nonmagnetic. Of course as a superconductor, for temperatures below Tc, this compound is a diamagnetic material. The addition of iron acts to cause friction in that compound. This effect is noticed to be strong with higher iron content and high applied fields and

0

1

2

3

150 250

T(K)

Mm

ax(e

mu/

g)


49

appears very clearly in the range 250K - 260K. The formation of an anti-ferromagnetic transition at T ~ 250K-260K can be associated with ordering of the iron atoms or to the formation of an anti-ferroelectric or ferroelectric states associated by the oxygen deficiency in some off-center positions of the structure.

الحالة العاديةأثر إضافة الحديد على قياسات التمغنط على

Y123لنظام مفرط املوصلية

يعقوب حمام و )الديري(عبدالرؤوف العلي ختام خصاونة،

ملخص

بعــد إضــافة الحديــد إليــه Y123طيــسية لنظــام مفــرط الموصــلية القــد تمــت دراســة الخــصائص المغن المطبقـة ومـع درجـة الحـرارة عنـد درجـات حـرارة أعلـى مـن الدرجـة بنسب مختلفة مع المجاالت المغناطيـسية

غيــر مغناطيــسي عنــد جميــع درجــات YBa2Cu3O7-δو كمــا هــو متوقــع، وجــد أن النظــام . Tcالحرجــة لقــد أدى إضــافة الحديــد إلــى حــدوث احتكــاك عنــد درجــات . مهمــا كــان المجــال المطبــق Tcالحــرارة فــوق

، فــي القابليــة 260Kوث انتقــال للطــور، عنــد درجــة حــرارة قريبــة مــن لــوحظ حــد . Tcحــرارة أعلــى مــن ويمكـن أن يفـسر ذلـك بتكـوين حـاالت مـن . من الحديـد % 4و % 1المغناطيسية للعينات التي تحتوي على

anti-ferroelectric وferroelectric تترافق مع ترتيب في األوكـسجين فـي بعـض المواقـع غيـر المركزيـة .في التركيب

References: [1] Ransley J. H., McBrien P. F., Burnell, G., Tarte E. J., Evetts J. E., Schulz R. R.,

Schneider C. W., Schemhla A., Bielefeldt H., Hilgenkamp H., and Mannhart,` Capacitance measurements on grain boundaries in

Y1– xCaxBa2Cu3O7– `,Phys. Rev. B , 70(2004) 104502.

[2] Zhao Y. E., He, Z. H., Lou Y. Y. and Gawalek W., ` The structure and superconductivity of Y1−xCaxBa2Cu3O7−δ coatings on YSZ single crystal substrate`, Physica C, 415 (2004) 197-202.


50

[3] Lal R., Awan V. B. S., Pandey S. P., Yadav V. S., Varandani,D., Narlikar,A. V., Chhikara A. and Gmelin E., ` Tc degradation in cuprate superconductors from the resistivity of YBa2(Cu1-xMx)4O8 for M=Fe and Ni`, Phys. Rev. B51 (1995) 539-546.

[4] Ciurcha D., Pop A. V., Cozar O., Ilonca, G., Pop, V. and Konopko L. A., Statistical representation of the excess conductivity in Bi-Sr-Ca-Cu-O superconductor` Supercond. Sci. Technol. , 9 (1996) 88-92.

[5] Ilonca G., Pop A. V., Corega C., Geru I. I., Gauter V. G., Konopko L. A., Min Y. and Deltour, R., `Superconducting, magnetic and normal state properties in Bulk (Bi1.6Pb0.4)(Sr1.8Ba0.2)Ca2(Cu1-xCrx)8Oy`, Physica C, 235-240 (1994) 1391-1392.

[6] Buckel and Kleiner, Superconductivity Fundamentals and Applications, Revised and enlarged edition, Wily-VCH Verlag, Berlin, Germany. (2004) 149-302.

[7] Ting Wu and Fossheim K., ` Lattice instabilities in single-phase polycrystalline Y1-xCaxBa2Cu4O8 (x=0, 0.005, 0.1)`, Supercond. Sci. Technol. 6 (1993) 827-836.

[8] Li A., Wang Y. N., Ying X. N., Zhang Q. M. and Chen W. N., ` Effect of zinc substitution on internal friction in YBCO superconductors`, Supercond. Sci. Technol 12 (1999) 645-648.

[9] Ying X. N., Zhang Q. M., Huang Y. N. and Wang, Y. N., ` Normal-state anomalous behaviours studied by the internal friction of YBa2Cu3O7-δ `, Supercond. Sci. Technol. 15 (2002) 1486-1489.

[10] Noguchi S., Inone J., Okuda K., Maeno Y. and Fujita T., `Magnetic susceptibility of Fe-Doped YBa2Cu3O7-δ ` , Jpn. J. Appl. Phys., 27 , (1988), L390- L392.

[11] Bara J. J., Bogacz B. F., Kim C. S., Szytula A. and Tomkowicz, Z., ` Crystal, electrical and magnetic properties of YBa2(Cu1-xFex)3O7-delta `, Supercond. Sci Technol., 4 (1991), 102-106.

[12] Mohamed M. A-K and Jung J, ` Magnetic properties of pure and Fe-doped YBa2Cu3O7-delta superconductors above Tc and at magnetic fields up to 5.5 T`, Supercond. Sci. Technol., 4 (1991), S304-S306.

[13] El Ali A. and Khasawinah Kh. “Critical Current Enhancement by Iron Doping in Y1Ba2Cu3O7-δ Superconducting system”, Accepted for publication in Abhath- Al-Yarmouk.

[14] Khasawinah Kh. And El Ali A., ` Magnetization Measurements on Fe-Doped Y-based Superconductors, Submitted for publication.

[15] Terrel Vandrah A.(Editor).1992. Chemistry of superconductor material preparation, chemistry ,characterization and theory. Noys puplications, USA, pp 469,517,571.


Temporal Growth of Filementation Instability Induced by Large Amplitude Laser Radiation

Mohammad Bawa'aneh * 1, Saud Al-Awfi** and Husam El-Nasser***

Received on March 20, 2006 Accepted for publication on June 28, 2007

Abstract The dispersion relation of the filementation process is derived and solved numerically in the

limit of low order coupling corresponding to no relativistic electron motion in the field of the incident pump wave. It has been found that the increase in electron temperature decreases the peak growth rate value and shifts its position towards short wavelength values. As the ion to electron temperature ratio increases, the instability range shrinks towards long wavelength side. By increasing the incident laser intensity, the range of instability broadens rapidly towards the short wavelength regions and the peak value of the instability growth increases. These results are closely related to the growth rate variation of the ion acoustic waves with changing electron to ion temperature ratio.

Keywords: Plasma; Filementation; Parametric instabilities.

Introduction In laser-plasma interactions, instabilities leading to anomalous heating of plasma

electrons and ions or to scattering of electromagnetic energy out of the plasma are still a subject of intensive study and concern. A coherent light incident into plasma induces density oscillations or fluctuations that couple with the incident wave to produce amplified scattered (induced) wave(s) for suitably matched frequencies and wavenumbers [1,2]. Strongly phase-correlated incident pump waves are essential for the excitation of different collective instabilities in plasma, and therefore, corresponding growth rates can be reduced and effective plasma heating is reached when broad band rather than narrow band laser systems are used [3-6].

Local changes in the dielectric properties of plasma, such as dielectric constant and refractive index, result from natural or induced non-uniformities or perturbations of an incident driving light source [7,8]. Positive changes in the dielectric constant in regions

© 2008 by Yarmouk University, Irbid, Jordan. * Department of Physics, The Hashemite University, Zarqa, Jordan. ** Department of Physics, Taybah University, Medina Munawarah, Saudi Arabia. *** Department of Physics, Al al-Bayt University, Mafraq, Jordan.

Bawa'aneh, Al-Awfi and El-Nasser

52

of high intensity produce a focusing lens, enhance perturbations and then initiate an instability. As pointed out in literature [7,9], filamentation instability is a nonlinear optical effect. In laser-plasma interactions, changes in the plasma dielectric properties are produced by many mechanisms such as plasma density amplification by ponderomotive bunching forces [10-12], thermal effects resulting from plasma heating by collisional absorption [13-16] and relativistic mass variations caused by incident ultra-relativistic intense laser radiation [17-21].

Induced scattering processes (wave-wave interactions in particular) can lead to the growth of both longitudinal and transverse waves in plasma. Self-focusing is one of the most known familiar mechanisms for the growth of transverse waves. When the refractive index for transverse waves makes a positive change, waves are refracted toward the region of increasing refractive index, and hence a local enhancement in the wave energy density tends to cause more waves to refract into that region. This nonlinear optical effect is known as self-focusing, that leads to a filamentation instability in which a beam of radiation breaks up into filaments [22,23].

Self-focusing of intense laser light in a plasma of increasing index of refraction occurs as a result of electron mass increase caused by their large quiver velocities in the light wave, and the reduction of the electron density by the ponderomotive force expulsion of electrons [17,24]. Modifications of the plasma index of refraction by the ponderomotive force and by the relativistic increase of mass were found to be of the same order of magnitude for laser beam channel width of the order of pc ω/ ,

where c is the speed of light and pω the plasma oscillation frequency. Instability threshold of thermal filamentation was found to decrease when accounting for effects of nonlocal electron heat transport. Also, thermal effects should dominate over ponderomotive mechanism for inertial confinement fusion parameter [25].

Because of its role in interacting with other decay waves in laser-produced plasmas, and of its low excitation threshold among other parametric and nonlinear coherent processes in plasmas, filamentation is still of more general interest [26-28]. In altering the spatial and temporal distribution of the driving source intensity, filamentation produces conditions favorable for other instabilities such as stimulated Brillouin and Raman scattering processes by lowering their thresholds and increasing their growth rates [29-33]. Consequence of filamentation, for example, on stimulated Raman scattering (SRS) was considered by Liu and Tripathi in long density scale length plasmas [27]; For electron temperature of 1 keV and incident laser power intensity of a few times 1410 2/ cmW at 1.06 microns, filamentation was found to have no strong effect on SRS, while it enhances SRS growth rate at lower intensities and higher temperatures [26,27]. Barr et. al [33] solved the full SRS equations for linear density profiles and found that feedback linking forward and backward SRS can result in an instability that is absolute for all frequencies.

Among the many theoretical studies of the filamentation instability, Sodha et. al [34] derived an expression for the maximum growth rate of filamentation instability


53

valid for arbitrary intensity of incident laser up to the density at resonance point. Numerical solution of Schrodinger equation coupled with the electron and ion thermal transport equations that describe the filamentation of a laser beam in plasma [35] showed self focusing of the incident beam by one to two orders of magnitude. Drake et. al [36] made a simple formalism for many of the parametric instabilities of light in a homogeneous plasma including the filamentation instability. In their study they showed growth rates as a function of the incident pump power.

In ref. [23] a simple expression for the growth rate of filamentation instability has been derived with some limitation on the validity of that expression. In his work, Kruer used the condition skcppγ , whereγ is the growth rate, k is the wave number and

sc is the speed of light. Also, the simple expression obtained by Kruer describes the

maximum growh rate, where the condition 0/ =dkdγ is used. Another limitation on Kruer's expression is the consideration of cold ions. Since resonances with ion waves are not involved in the filamentation instability, the process is not extremely sensetive to plasma inhomogeneity. Accordingly, in this report, we investigate the full spatial behaviour of ponderomotive filamentation instability induced by a large amplitude laser radiation, where temporal growth rate is discussed. In Sec. 2, the dispersion relation of the filamentation process is derived in the limit of low order coupling for nonrelativistic electron motion in the field of the incident pump wave. Numerical analysis of the dispersion relation and the filamentation temporal growth rate are given in Sec. 3 without the use of any of the above limitations put onγ in ref. [23]. Finally, discussion and conclusions are presented in Sec. 4.

Dispersion Relation To derive the coupled equations for the filamentation of light in plasma as a four

wave process, ions and electrons are treated as conducting fluids that are coupled via the force and density equations of each species and the Maxwell's field equations for the average electromagnetic fields E

r and B

r, namely,

tBE∂∂

−=×∇r

rr,

tE

cVqnB

∂∂

+=×∇ ∑r

rrr20

1α

αααµ

( ) 0=⋅∇+∂∂

ααα Vnt

n rr, ∑=⋅∇

αααε

qnE0

1rr

αααααααα

αα γ nTkBVnqEnqdtVd

nm B ∇−×+=rrrr

r

,

(1)

(2)

(3)


54

where 0µ is the free space permeability, 0ε the free space permittivity, ie,=α ,

and am , aq , an , aT , aγ , and aV are the corresponding mass, charge, particle number density, temperature، ratio of specific heats, and fluid average velocity, respectively.

Vectors Er

and Br

represent the average values of the total electric and magnetic fields in the plasma, respectively. We, also, assume the presence of a large amplitude incident

light wave of the form )cos(),( 0000 trkEtrE ω−⋅=rrrrr

, where 0Er

is the vector

wave amplitude, 0kr

and 0ω are its wavenumber and frequency, respectively.

The derivation steps leading to the coupled equations needed for describing the filamentation process are straightforward, but somehow lengthy. Since the filamentation process involves both high frequency electromagnetic waves and low frequency density fluctuations (ion-acoustic mode), transverse and longitudinal parts of the total current density are usually treated separately. In a plasma of uniform density and temperature with non-relativistic electron motion in the field of the transverse waves, and upon linearizing the electron density about the equilibrium density en0 and the electric field

Er

about the large amplitude incident wave 0Er

, we obtain the following space-time

Fourier transformed equations for the induced (scattered) transverse modes Er~

coupled to the electron density fluctuation en~ via 0E

r,

( ) ( ) ( ) ( )[ ]000000

02

2222 ,~,~2

,~ ωωωωωε

ωωωω +++−−=−− kknkkn

mEe

kEkc eee

pe

rrrrr

rr

( ) ( ) ( ) ( )

+++⋅

+−

−−⋅=−

ωωωω

ωωωω

ωωω

0

000

0

000

0

220222 ,~,~

2,~ kkEEkkEE

mmkeZn

knkcie

ees

rrrrrrrrr

(4)

(5)

where eooepe men εω /2= is the electron plasma oscillation frequency, Z the

single ion charge state، iiBieBes mTkTkZc /)( γγ += the speed of the ion-

acoustic wave, and ω and kr

are the frequency and wave number for the ion-acoustic mode of the plasma.

To use equations (4) and (5) for describing the four wave filamentation process, we assume a first induced electromagnetic mode characterized by the frequency


55

ωωω −= oa and wavenumber kkka

rrr−= 0 , and a second transverse mode with

ωωω += ob and kkkb

rrr+= 0 being the frequency and wave number of this mode,

respectively. Replacement of ),( ωkr

in equation (4) by ),( aak ω−−r

and

),( bbk ωr

results in the following coupled equations;

( ) ( ) ( ) ( )[ ]ωωωωε

ωωωω ,~2,2~

2,~

0000

02

2222 knkknm

EekEkc ee

e

aaaapeaa

rrrr

rr+−−=−−−−

( ) ( ) ( ) ( )[ ]ωωωωε

ωωωω ,~2,2~

2,~

0000

02

2222 knkknm

EekEkc ee

e

bbbbpebb

rrrr

rr+++=−−

( ) ( ) ( ) ( )

⋅+

−−⋅=−

b

bb

a

aa

ie

ees

kEEkEEmm

keZnknkc

ωω

ωω

ωωω

,~,~

2,~ 00

0

220222

rrrrrrr

(6)

(7)

(8)

Necessary information about the four wave filamentation process in a homogeneous plasma of uniform density and temperature are contained in the instability dispersion relation. The system of equations (6-8) becomes a closed one if we ignore higher order modes corresponding to frequencies and wavenumbers other than those involved in the filamentation process, i.e. 02kk ± and 02ωω ± and higher. In this sense, the resulting dispersion relation is that of the linear evolution of the filamentation process.

Upon solving the coupled equations (6-8) with no account to higher order modes, using the dispersion relation of propagating incident light in cold plasma

20

2220 kcpe += ωω and the transverse nature of pure filamentation process described

by the condition 0. 0 =kkrr

, we obtain the following dispersion relation;

( ) [ ] 0)(2

4)( 222222

222222222 =−−−−− kck

kckc ospios ω

νωωωωω ,

(9)

where we introduced the abbreviations 00 / ων eos meE= for the magnitude of the electron quiver velocity in the field of the incident wave, and

peiepi mZm ωω /= for ion plasma frequency.


56

Numerical Results In our numerical analysis, we characterize the filamentation instability by its

growth rate γ normalized to the incident laser frequency 0ω in the limit

0ωγω pp= . Temporal growth of filamentation process in underdense plasma

region is considered. In all calculations, we assume incident light wavelength 0λ of

mµ06.1 , isothermal electrons, one-dimensional adiabatic ions and an equilibrium

electron plasma density en0 of one quarter of the incident light critical density.

Normalized instability growth rate 0/ωγ versus Dekλ for different electron

temperature values is shown in Fig. 1, where Deλ represents the electron Debye length. Curves from high-peaked to low-peaked have the following electron temperature values;

2,1,5.0=eT and keV3 , respectively. Other parameters are 2150 /10 cmWI =

for the incident laser intensity and ei TT 3.0= for the ion temperature. As seen in Fig. 1, the increase in electron temperature results in a decrease in the growth rate peak value and a position shift towards short wavelength values (large k), while the instability range remains constant. This effect is observed only in the long wavelength part (small k) of the instability range, with the rest of the instability range being unaffected by increasing eT .

Figure (1): Normalized growth rate 0/ωγ versus Dekλ for different electron temperature. Curves presented from high-peaked to low-peaked curvey correspond to

keVTe 3,2,1,5.0= , respectively. Other parameters are 25.0/ =crnn , 215

0 /10 cmWI = , mµλ 06.10 = , ei TT 3.0= .


57

Figure 2 shows the normalized instability growth rate 0/ωγ versus Dekλ for

different ion to electron temperature ratios ei TT / . Curves with wider to narrower range

of instability correspond to 1,6.0,3.0,1.0/ =ei TT , respectively. Other parameters

are 2150 /10 cmWI = and keVTe 3= . Contrary to Fig. 1, the instability range

shrinks towards the long wavelength side of the instability range as the ion to electron ratio increases. The magnitude of the peak of instability reduces very slightly, while the peak position remains almost fixed.

Figure (2): Normalized growth rate 0/ωγ versus Dekλ for differention ion to electron temperature. Curves with wider to narrower range of instability correspond to

keVTT ei 1,6.0,3.0,1.0/ = , respectively. Other parameters are 25.0/ =crnn , 215

0 /10 cmWI = , mµλ 06.10 = , keVTe 3= .


58

Variation of the normalized instability growth rate 0/ωγ versus Dekλ with the

incident laser intensity 0I is shown in Fig.3. Curves with narrower to wider range of

instability correspond to 1413120 10,10,10=I and 215 /10 cmW , respectively.

Other parameters are keVTe 3= and ei TT 3.0= . By increasing the intensity of the incident laser beam, the range of instability broadens rapidly towards the short wavelength side. Also, the peak value of the instability increases. For 12

0 10≈I the instability is negligible compared to higher values of incident laser intensity leading to a threshold value of 0I below which the mode can be considered stable.

Discussion and Conclusions Ponderomotive filamentation process has been considered in the limit of low order

coupling corresponding to nonrelativistic electron motion in the field of an incident large amplitude pump wave. In the problem of laser-matter interaction, oscillation of the plasma electrons in the incident laser radiation induces higher field harmonics at frequencies 0ωωω n+=′ and wavenumbers 0knkk

rrr+=′ , where ω and k

r are the

frequency and wavenumber of the low frequency perturbation, respectively. In the weak field limit cos ppν , only low field harmonics are excited [36]. In the presence of low frequency ion density fluctuations, instabilities such as Brillouin and filamentation instabilities derived from the lowest harmonic 1=n are expected to have the strongest growth because a small amount of energy can increase the amplitude of these instabilities.

Numerical solution of the dispersion relation shows that the increase in the electron temperature [for fixed ion to electron temperature ratio] decreases the growth rate peak value, while its position shifts towards short wavelength values [Fig. 1]. This effect is only observed in the long wavelength regime. By varying the ratio of ion to electron temperatures, the instability range shrinks towards the long wavelength end as the temperature ratio increases [Fig. 2]. By increasing the incident laser intensity, the range of instability broadens rapidly towards the short wavelength region and the peak value of the instability growth increases [Fig. 3]. Compared to higher incident laser intensities, values of the order of 212

0 /10 cmWI ≈ have a negligible growth rate and instability

range. This leads to a threshold value of 0I , below which the mode can be considered stable.


59

Figure (3): Normalized growth rate 0/ωγ versus Dekλ for differention incident laser intensity. Curves with narrower to wider range of instability correspond to

151413120 10,10,10,10=I , respectively. Other parameters are 25.0/ =crnn ,

mµλ 06.10 = , keVTe 3= and ei TT 3.0= .

Presence of ion acoustic waves in the filamentation process explains the observed behavior of filamentation growth as the ratio ei TT / is varied; Increasing ei TT / makes the ion acoustic mode increasingly damped, and hence the range of filamentation instability decreases [37]. By decreasing ei TT / , induced modes can persist for a greater number of oscillations and thus are more likely to be observed. The long wavelength dispersion relation of ion acoustic waves kcs≈ω is valid up to ei TT ≈ . However, for

collisionless plasma with ie TT ≈ , more accurate estimates of growth rates and thresholds require kinetic treatment since ion acoustic waves are predominantly damped by ion Landau-damping.


60

This work is an extention to the work done by Kruer [23] where his work has many limitations; two of which are the use of the condition skcppγ in his derivations and

considering the case of cold ions only where iT does not appear as a parameter in his expression. Another, very important, difference between Kruer's work and this work is our study of spacial behaviour of filamentation instability by solving it's spectra with respect to the wavenumber of the low frequency ion wave, whereas ref. [23] derived the maximum growth rate only of the filamentation instability. This generalization allows for a detailed investigation of the instability as a function of the intensity of the incident laser light radiation and the temperature of the plasma species.

Acknowledgment MSB would like to thank the Hashemite university for supporting this work.

فاذ املحفزة بأشعة الليزر ذات السعة العاليةال إستقرارية الننمو

محمد بواعنة، سعود العوفي و حسام الناصر

ملخص

غيـر لقد تم اشتقاق وحل عالقـة التـشتت الختـراق أمـواج الليـزر لمـادة البالزمـا فـي حـدود الـسرعات بزيـادة . ات يقلـل مـن قيمـة تنـامي األمـواج ات، وقد وجدنا أن زيـادة درجـة حـرارة اإللكترونـ النسبية لإللكترون

اســتقرارية االختــراق يقــل، وكــذلك فبزيــادة شــدة النــسبة درجــة حــرارة األيونــات إلــى اإللكترونــات فــإن مــدى هـذه النتـائج مرتبطـة . ل كبيـر اسـتقرارية يزيـد بـشك فـإن مـدى الـال أمواج أشعة الليزر الساقطة علـى البالزمـا

.يونات بتغير نسبة درجة حرارة األيونات إلى اإللكتروناتر ال استقرارية أمواج األبتغي

References [1] Silin V. P., Sov. Phys. JETP 20 (1965)1510 [Zh. Eksp. Teor. Fiz. 47 (1964) 2254].

[2] DuBois D. F. and Goldman M. V., Phys. Rev. 164 (1967) 207. [3] Valeo E., Oberman C., and Perkins F., Phys. Rev. Lett. 28 (1972) 340. [4] Yamanaka C., Yamanaka T., Sasaki T., Yoshida K. and Waki M., Phys. Rev. A 6

(1973) 2335 .

[5] Bodner S. E., Chapline G. F., and DeGroot J., Plasma Phys. 15 (1973) 21 . [6] Ross I. N., Matousek P., Towrie M., Langley A. J. and Collier J. L., Opt. Commun.

144 (1997) 125. [7] Shen Y. R., Prog. Quant. Electr. 4 (1975) 1.


61

[8] Andrew J. Schmitt, Phys. Fluids 31 (1988) 3079.

[9] Marburger J. H., Prog. Quant. Electr. 4 (1975) 35. [10] Kaw P. K., Schmidt G. and Wilcox T., Phys. Fluids 16 (1973) 1522.

[11] Max C., Phys. Fluids 19 (1976) 74. [12] Shkarofsky I. P., Phys. Fluids 23 (1980) 52. [13] Perkins F. W. and Valeo E. J., Phys. Rev. Lett. 32 (1974) 1234. [14] Tripathi V. K. and Pitale L. A., J. Appl. Phys. 48 (1977) 3288.

[15] Lal A. K., Marsh K. A., Clayton C. E., McKinstrie C. J., Li J. S. and T. W. Johnston, Phys. Rev. Lett., 78 (1997) 670.

[16] Silin V. P., Sov. Phys. JETP, 92 (2001) 617 [Zh. Eksp. Teor. Fiz., 119 (2001) 710)].

[17] Guo-Zheng S., Ott E., Lee Y. C. and Guzdar P., Phys. Fluids, 30 (1987) 526. [18] Max C. and Arons J., Phys. Rev. Lett. 33 (1974) 209. [19] Sheng Z. M., Nishihara K., Honda T., Sentoku Y., Mima K. and Bulanov S. V.,

Phys. Rev. E, 64 (2001) 66409. [20] Barr H. C. and Hill L. J., Phys. Plasmas, 10 (2003) 1135.

[21] Honda M., Phys. Rev. E, 69 (2004) 016401.

[22] Melrose D. B., "Instabilities in space and laboratory plasmas", Cambridge university press, Cambridge (1986), pp.100-116 .

[23] Kruer W. L., "The Physics of Laser Plasma Interactions", Addison Wesley Publishing Company (1988), pp. 90-94.

[24] Schmidt G. and Horton W., Comments Plasma Phys. Controlled Fusion 9(1985) 85. [25] Epperlein E. M., Phys. Rev. Lett. 65 (1990) 2145. [26] Barr H. C., Boyd T. J. M. and Coutts G. A., Phys. Rev. Lett. 56 (1986) 2256. [27] Liu C. S., and Tripathi V. K., Phys. Fluids 29 (1986) 4188. [28] Giulietti A. et al., Phys. Rev. E, 59 (1999) 1038. [29] Eliseev V. V., Rozmus W., Tikhonchuk V. T. and Capjack C. E., Phys. Plasmas, 3

(1996) 2215. [30] Huller S., Mounaix Ph. and Pesme D., Physica Scripta, T63 (1996) 151. [31] Huller S., Mounaix Ph., Tikhonchuk V. T. and Pesme D., Phys. Plasmas 4 (1997)

2670. [32] Maximov A. V. et al., Phys. Plasmas, 11 (2004) 2994.


62

[33] Barr H. C., Boyd T. J. M. and Coutts G. A., Phys. Rev. Lett., 60 (1988) 1950.

[34] Sodha M. S., Sharma J. K., Sharma R. P. and Kaushik S. C., J. Appl, Phys., 49 (1978) 599.

[35] Nicolas D. J. and Sajjadi S. G., J. Phys. D: Appl. Phys. 19 (1986) 737. [36] Drake J. F. Kaw P. K., Lee Y. C. and Schmidt G., Phys. Fluids, 17 (1974) 778. [37] Williams E. A. et. al, Phys. Plasmas, 2 (1995) 129.


Indoor Radon-222 Concentration Measurements in Some Houses in Dura District During the

Winter Season of the Year 2000

Amin Leghrouz, Mohammad Abu-Samreh* 1and K. M. Awawdah**

Received on Nov. 25, 2004 Accepted for publication on Sept. 28, 2006

Abstract In this study, we report the indoor radon-222 concentration levels of various dwellings

located in Dura district during the winter season of the year 2000. Measurements were performed using the CR-39 detectors. Over 29 dwellings and 21 classrooms scattered randomly through this district were investigated. The radon concentration levels in the inspected zones were found to vary considerably from 29 to 398 Bq/m3. The regional average was found to be 83 Bq/m3. This corresponds to an approximated annual average effective dose equivalent per population of 2.13 mSv. Most measurements were found to be within the internationally accepted concentration levels.

Keywords: Radon, Level, Dosimeter, Exposure, Isotopes, CR-39 detector, Activity. I. Introduction

Over the past four decades, natural radiation exposure due to radon-222 (222Rn) and its daughters, i.e., Polonium-218 (218Po) and Polonium-214 (214Po), have been recognised as a worldwide problem. They have been considered as a major cause of significant health risk, especially to lung and bones, to general public [1-3]. This is because 222Rn and its natural radioisotopes, such as radon-219 (219Rn) and radon-220 (220Rn), emit alpha particle which is highly effective in damaging lung tissues [4-5]. So, breathing high concentrations of 222Rn gas can cause lung cancer [6]. Besides, radon daughters attach themselves to dust particles present in the air and the attached or unattached radon daughters may be inhaled and become lodged in the nasal passage, trachea and pulmonary tree or pulmonary parenchyma [7].

Radon 222Rn as well as other natural radon isotopes such as 219Rn and 220Rn is naturally produced within that part of earth crust that contains cores of uranium and thorium and their progenies in secular equilibrium [8]. The 222Rn gas emanates from soil, building materials, water supplies, and natural gases of terrestrial origin [9-11]. 222Rn gas enters houses from soil through cracks in concrete floors and walls, floor drains, construction joints, and tinny cracks or pores in hollow-block walls [12]. Thus, © 2008 by Yarmouk University, Irbid, Jordan. * Department of Physics, College of Science , Al-Quds University, Jerusalem, Abu-Deis, West Bank, Palestine ** Department of Physics, College of Science, Hebron University, Hebron, West Bank, Palestine.

Leghrouz, Abu-Samreh and Awawdah

64

radon gas can diffuse through the soil and seep into the buildings and trapped within poorly ventilated buildings. Accordingly, the indoor 222Rn concentration is expected to be high in places having a direct contact with soil and of poor ventilation.

Investigations of natural radiation have received particular attention worldwide and led to extensive surveys in many countries [4]. During the past two decades, a tremendous number of investigations have been conducted and directed to monitor the radon gas levels [13-17]. The recommended indoor action level for the general public is ranged between 150 and 600 Bq/m3 [18-20]. The indoor radon level of 150 Bq/m3, which corresponds to a yearly indoor effective average dose of 1.30 mSv/y has been adopted in USA as a reference point before making any action [20]. The indoor radon concentration average of 39 Bq/m3 and its corresponding "yearly" effective dose population-weighted average value of about 1 mSv have been assigned as the world reference data to the indoor concentrations according to the United Nations Scientific Committee on the Effects of Atomic Radiation (UNSCEAR) 2000 report [4, 7].

The first attempt to measure radon concentration levels in the Southern part of the West Bank, Palestine was performed in Hebron University campus [10]. In this study, we are mainly concerned in investigating the indoor radon concentration levels in houses and public places such as schools in Dura district and its surroundings. Dura, is a city located ten kilometers south-west of old Hebron city and it has more than 60000 inhabitants. The study is motivated by several considerations, among no information or data about radon gas concentration has been reported before for this region. Therefore, this study is expected to provide some data and information about the radon concentration levels concerning the radon exposures to the general public. Besides, some bases for radiation protection countermeasures are recommended too. This includes action level recommendations for the existing houses and for future housing architect design.

II. Methodology This study is directed to smear up a wide area in Dura district (see Figure 1)

searching for places of high indoor 222Rn concentration levels. This includes dwellings, houses and public places such as schools that have been chosen randomly. Measurements were performed during the winter seasons of the year 2000. The passive devices (Solid state nuclear track detectors (SSNTD'S)) of type CR-39 have been used. The passive radon detectors (dosimeters), have been developed and described somewhere else by several workers [1, 17,21-24].

Indoor Radon-222 Concentration Measurements in Some Houses in Dura District During the Winter Season of the Year 2000

65

Figure 1 a) General geological map of the West Bank showing the investigated area

enclosed by a blue circle


66

Figure 1b) Map of Hebron province showing the location of the villages and cities under

investigations.


67

In this study, a typical dosimeter consists of a CR-39 sheet having dimensions of 12.0 mm × 13.0 mm × 1.0 mm inserted in a flat position to the bottom of a plastic cup and held in place by a small piece of blue-tack as shown in Figure 2. The plastic cup has a dimension of 70.0 mm diameter orifice, 50.0 mm diameter base and 65.0 mm deep. The top of the cup was covered with a permeable cling film (Polyethylene foil) to allow only 222Rn gas to pass through the film and to exclude radon daughters from entering the dosimeter [10]. About seventy dosimeters were distributed in over 50 dwellings (single houses, apartments complexes, storage rooms) chosen randomly in Dura district. All dosimeters were fixed at a level of half meter above the ground and left in position for almost two months (during the time interval between 6/1/2000 and 10/3/2000).

Figure 2 Typical CR-39 dosimeter.

The diffused radon gas into the cup decays by the emission of alpha particle, followed by two further alpha particle emissions of the short-lived decay daughters. As alpha particles hit the CR-39 chip, latent holes will be produced on the chip's surface by cracking polymer chains. After the assigned period (∼60 days), the undamaged remaining dosimeters were collected and detectors were taken out. The CR-39 detectors are then chemically etched in 6.25 N-solution of NaOH at temperature of 98±2 0C for fifty minutes [22-24]. During the etching process, the solution has to be stirred

cling film

70.0 mm elastic-band 65.0 mm CR-39 Blue-tack 50.0 mm


68

constantly to enhance the alpha particle tracks on the detector surface. Detectors were then washed with distilled water and dried prior to microscope inspection. An optical microscope, 200 times of magnification was used to count the number of tracks per cm2 occurred in each detector. Typical nuclear track etches picture displayed in Figure 3 as a field microscopic view magnified 200 times.

Figure 3: The field microscopic view magnified 200 times.

The general procedure for counting the number of tracks is summarized by dividing the CR-39 detector surface area into several assigned field of view each has an area of 7.4×10-5 cm-2 [= πr2 = π (4.85×10-3)2, where r is the radius of field of view] as shown in Figure 4. We have examined ten, twenty, thirty, and forty field of view (region of interest) for five detectors chosen arbitrary from the monitored zones. Then, the number of tracks in each area was counted and the average of the measurements is obtained. It was found that the total number of tracks per field for most of the measurements performed on the detectors is roughly equal to the average value for that detector. The deviation of each measurement from the average is found to be within the standard deviation value. Hence, an average of ten field of views can be considered to be a representative to the whole surface of the detector.


69

CR39 detector

Field of view

Figure 4: Schematic representation of field view on the surface of the detector.

In this study, ten field of views were chosen arbitrary to be representative of the whole surface of each detector, and the number of tracks per centimeter-squared was calculated. The obtained average of tracks of the ten fields of views for each detector was used to calculate the radon concentration, CRn, in units of Bq/m3. In calculating CRn, we have adopted the calibration used by the researchers at Yarmouk University which based on Bristol university measurements [9, 13]. Thus, CRn is calculated according to the following equation [9, 13]:

(1) tddtCC

0

00Rn =

where 0C is the value of the radon concentration in the calibration chamber which

is equal to 90 KBq/m3 [9], 0t is the exposure time of detector in radon chamber which is

about 48 hours, d is the tracks density as measured on detector surface in units of

tracks/cm2, d0 (

×= − 240 106.9

7.31cm

tracksd ) is the density of tracks detector measured on

calibrated detectors in radon chamber, and t is the detectors exposure time to indoor radon which is about two months of durations.

Finally, the annual effective dose to population was calculated using the conversion factor value reported in UNCEAR 2000 report [4]. According to UNCEAR 2000 report, the world average 222Ra concentration value of 39 Bq/m3 corresponds to an annual effective dose to the population of about 1 mSv/y [4, 7].

In this study, the well known standard deviation method is used for calculating the standard deviation error for each detector reading according to the following equations:


70

(2) ,1

n

NX

n

ii∑

==

and

( )(3)

2

nXNi

n−

=σ

where X is the average of the number of tracks, iN is the number of tracks in the

ith field of view, n is the total field of views (10), and ∑=

n

iiN

1is the total number of

tracks on each detector. The obtained uncertainty for all detectors was found to be between 10-20 % [9].

It is worth mentioning here that the calculated standard deviation for all detectors was found to be high (~40%). This is because such an error is obtained for a set of detectors that were uncorrelated and the standard deviation error of the regional average in this case has no physical meaning. Thus, this type of error will not be implementing in the present investigation.

III. Results and Discussion The present radon concentration levels data were obtained from 50 dosimeters

collected after 60 days. The rest of the distributed dosimeters were lost. The indoor radon concentrations were calculated using equation (1). The main zones, the number of detectors, N, and the range and the frequency of radon concentration levels of 50 rooms selected from the monitored zones were exhibited in Table 1.


71

Table 1.The range and the frequency of radon level concentrations of 50 houses and classrooms selected randomly from the monitored zones in Dura district.

Range (Bq/m3 )

0-50 51-100 101-150 151-200 Above 200

Zone N

Frequency Frequency Frequency Frequency Frequency

Dura (Houses) 14 6 5 ------ 1 2

Serieh 4 1 2 --- 1 ---

Muraysh 5 2 3 --- --- ---

Dayr al Alasal 3 2 1 --- --- ---

Beayt Mirsim 1 1 --- --- --- ---

Karmah 2 1 1 --- --- ---

Salah Eddin School 17 4 9 3 1 ---

Secondary Male School (Dura)

1 ---- 1 --- --- ---

aL-Majd School 1 1 --- --- --- ---

Bayt al Rush School 1 --- 1 --- --- ---

aL Kum School 1 --- --- 1 --- ---

Total 50 18 23 4 3 2

Statistical methods were employed to analyze the collected data. A summary of minimum (Min), maximum (Max), average (Ave), and annual effective dose equivalent to the population (Dy) of the measured indoor radon concentrations for each monitored zone Dura district is exhibited in Table 2.


72

Table 2. Max, min, Ave radon concentration levels (Bqm-3) and annual effective dose equivalent to the population (Dy) of 50 houses and classrooms selected from the monitored zones.

Radon level concentrations (Bq/m3 ) Dy (mSv/y) Main Zone

N Min Max Ave

Dura (Houses) 14 33 398 129 3.31

Muraysh 5 29 65 45 1.15

Serieh 4 47 174 84 2.15

Dayr al Alasal 3 39 80 55 1.41

Beayt Mirsim 1 55 55 55 1.41

Karmah 2 35 66 51 1.31

Salah Eddin School 17 29 154 72 1.85

Secondary Male School (Dura)

1 95 95 95 2.44

aL-Majd School 1 45 45 45 1.15

Bayt al Rush School 1 64 64 64 1.64

AL Kum School 1 111 111 111 2.85

Total 50 29 398 83 2.13

As it can be seen from Table 2, the minimum indoor radon concentrations of about 29 Bq/m3 has been found in Muraysh zone; while the highest radon concentration level of 398 Bq/m3 has been recorded in a storage room in a house located in central part of Dura city. This is because the floor of the room is basically made of pure soil. This is in agreement with the prediction that the highest output of indoor radon concentrations was originated from soil. It is found that the indoor radon concentrations in the monitored zones are lower than 150 Bq/m3 in general (see Table 2). This value is below the radon reference levels that range from 200-600 Bq/m3 as recommended by ICRP [18], IAEA [2,19] and it is also lower than the USA assigned radon level of 150 Bq/m3 [20]. In less than 5% of the houses radon concentrations were in excess of 200 Bq/m3 where an action is recommended by ICRP [18]. The overall regional average of radon concentrations is about 83 Bq/m3.

In general, values of the radon concentrations that exceed the hazardous set values were found in basements with concrete unpainted walls and having poor ventilation. This might be due to the reduction of air flow rates which increases radon concentrations. Very low radon concentrations of 29 Bq/m3 have been recorded in a large well-ventilated house. This confirms the importance of ventilation in order to avoid radon accumulation to the unaccepted hazardous level. Furthermore, the unpainted constructed houses from local stones were found to have the highest radon concentrations. For instance, indoor radon concentrations of 247 Bq/m3 were recorded in one of the houses constructed in Ngt Nouh.


73

In view of the importance of the protection of the children in schools, about 21 classrooms distributed randomly among six elementary schools in this district were monitored. The Min, the Max, the Ave, and the Dy of radon concentrations in classrooms of the investigated schools are summarized in Table 2. It was found that about 95 % of the classrooms have indoor radon concentrations below 100 Bq/m3.

In order to compare between the indoor radon concentrations results in classrooms located in different floors, about seventeen classrooms were inspected in Salah Eddin Secondary School in central part of Dura city. The obtained results of the indoor radon concentrations for classrooms in the ground and first floors were presented in Table 3.

Table 3. Max, min, Ave, and dose equivalent of the average of indoor levels of two floors in Salah Eddin School.

Radon level concentrations (Bq/m3 ) Main Floor

N Min Max Ave

Dy

(mSv)

Ground Floor 8 29 130 72 1.85

Second Floor 9 32 154 82 2.10

Total 17 29 154 78 2.00

By examining the obtained results of classrooms in Salah Eddin School (see Table 3); a significant difference in the mean radon concentrations between the floors has been noticed. Besides, great variations within concentration values for rooms in the same floor have been also observed. Clearly, the main reason behind such variations is the poor ventilations. For example, indoor radon concentrations of 154 Bq/m3 have been reported in the drawing room in the first floor, which is closed most of the time.

Comparing the school results with that of houses, we found that the average values of radon concentrations in school are smaller than that of houses. This is mainly attributed to several factors, among: Schools have walls and ceilings that are properly painted and almost without any cracks. Besides, schools are well ventilated. Accordingly, remedial efforts should be mainly focused on reducing the emanation of radon from soil by sealing all cracks and joints between the floor and walls and the floor cracks. Moreover, by increasing ventilation rates within homes air infiltration rates will be increased too.

In calculating the annual effective dose equivalent to the population, we have adopted the dose conversion factor in the UNCEAR 2000 report [4]. The annual effective dose equivalent received by people in the investigated zones is found to vary from 1.15 mSv/y to 3.31 mSv/y. The average effective dose equivalent to population for the monitored zones is about 2.13 mSv/y in total. This is almost two times higher than the assigned world average dose equivalent to population of about 1.00 mSv/y [4].


74

IV. Conclusion The obtained results for the indoor radon concentration levels in Dura district

indicate that not all the investigated zones are of high levels. The overall average radon concentration levels in the monitored zones were found to be of the order 83 Bq/m3 in winter season. This corresponds to an overall average dose to population of 2.13 mSv/y, which is higher by a factor of two than the value set by the environmental protection agency [18].

In this study, it was found that ventilation plays an important role in reducing the indoor radon concentration levels. The variation of radon concentration levels observed in winter may be attributed to the changes resulted in the ventilation rates. Thus, improving ventilation of these places will result in reducing radon concentrations. Accordingly, remedial efforts should be focused mainly on reducing the emanation of radon from soil as well as effective consideration to the ventilation of those houses is strongly recommended.

قياس مستويات تراكيز غاز الرادون في هواء عينات من املباني املتعددة في

2000مقاطعة دورا خالل فصل الشتاء لعام

عواودة أبو سمرة ودأمين الغروز، محم

ملخص

مبـاني المتعـددة فـي مـن ال اتفي هذه الدراسة، تم قياس مستويات تراكيز غاز الـرادون فـي هـواء عينـ باسـتخدام مجارعـات راديونيـة معـايرة ولقـد تمـت القياسـات . 2000 لعـام لـشتاء خالل فـصل ا مقاطعة دورا

و منـزل غرفـة 29 مـن رأكثاشتملت الدراسة على . CR-39 سابقا تحوي على كواشف بالستيكية من نوع ووجــد مــن خــالل الدراســة أن .، اختيــرت بــشكل عــشوائي فــي المقاطعــةعــدة مــدارس غرفــة صــفية فــي 21

ووجـد أن معـدل تركيـز الـرادون علـى .متر مكعـب / بيكرل 398 إلى 29تراكيز غاز الرادون تتراوح ما بين يرتبط هذا المعدل بمعـدل جرعـة سـنوية و .متر مكعب /بيكرل 83 مستوى المنطقة الخاضعة للدراسة كان

ة المعنيـة هـي المنطقـ تراكيـز غـاز الـرادون فـي أن وقد دلت النتائج على . ت سيفر ميللي 2.13 للفرد يساوي .ضمن الحدود المقبولة عالميا

References [1] Celebi N. and Alkan H., Evolution of Natural Radiation in the Kesttanbol SPA

Region. Radiat Protect. Dosim., 69(3) (1997) 227.

[2] Baxter M. S., Environmental radioactivity: A perspective on industrial contributions. IAEA Bulletin., 35(2) (1993) 33.


75

[3] UNSCEAR (United Nations Scientific Committee on the Effects of Atomic Radiation)., Sources and Effects of Ionizing Radiation. Report to the General Assembly, with Scientific Annexes, United Nations, New York, NY. United Nations, (1993) 45-56.

[4] UNSCEAR (United Nations Scientific Committee on the Effects of Atomic Radiation). Sources, Effects, and Risks of Ionizing Radiation. Report to the General Assembly. New York, NY, United Nations, (2000) 165.

[5] ICRP (International Commission on Radiological Protection)., Lung cancer risk from indoor exposure due to radon daughters. ICRP Report 50, Oxford: Pergomon Press, 17(1) 1987.

[6] Bunditz R. G., Radon-222 and its daughters, Health Phys., 26 (1974) 145.

[7] Anastasiou T., Tsertos H., Christofides H. and Christodoulides G., Indoor radon concentration measurements in Cyprus using high-sensitivity portable detectors, J. Enviro. Radio., 68 (2003) 159.

[8] Nero A. V. and Nazaroff W. W., Characterizing the Source of Radon Indoors. Radiat. Prot. Dosim., 7(1- 4) (1984) 23.

[9] Kullab M. K., Al-Bataina B. A., Ismail A. M., Abumurad K. M. and Gaith, A., Study of radon-222 concentration levels inside kindergardens in Amman. Radiat. Meas. 28(1-6) (1997) 699-702.

[10] Hassan F. I., Indoor radon concentration measurements at Hebron University campus. An-Najah University Journal for Research., 4(10) (1996) 92.

[11] Camplin G. C., Henshaw D. L., Lock S. and Simmons Z., A national survey of background alpha particle radioactivity. Phys. Educ., 23 (1988) 212.

[12] Kunz E., Sevc J., Placek V. and Horacek J., Lung Cancer in Man in Relation to Different Time Distribution of Radiation Exposure. Health Phys., 36 (1981) 699.

[13] AL-Bataina B. A. and Lsataifeh M. S., Indoors Radon Levels During Autumn in AL-Karak District, Jordan. Mu’tah Lil-Buhuth wad-Dirasat., 15(1) (2000) 123.

[14] Kobal I., Smodis B., Burger J. and Skofljanec M., Atmospheric Radon-222 in Tourist Caves of Slovenia, Yugoslavia. Health Phys., 52(4) (1987) 473.

[15] Popovic D., Djuric G. and Todorovic D., Radionuclides in building materials and radon indoor concentrations. Radiat. Protect. Dosim., 63(3) (1996) 223.

[16] Ahmed J. U., Radon in the human environment: Assessing the picture. IAEA Bulletin., 36 (1994) 32.

[17] Torri A. D, Oppon O. C., Piermattei S., Susanna A. F., Seidel J. L., Tommasino L. and Ardanese L., Measurements of soil and indoor radon in Italy. Nucl. Tracks Radiat. Meas., 15 (1988) 637.


76

[18] ICRP. International Commission on Radiological Protection. Protection against radon-222 at home and work. Publication 65. (England, Elsevier Science, Ltd.) (1994) 56.

[19] Gonzalez A.J., Global levels of radiation exposure: Latest international findings. International Atomic Energy Agency (IAEA) Bulletin., 35(4) (1993) 49.

[20] EPA: USA, The national radon measurement proficiency (RMP) program, Washington, D. C.: Environmental protection Agency; EPA 520/1-88-025; (1988) 32.

[21] Abu-Jarad F. A. and AL Jarallah M. I., Radon in Saudi houses. Radiat. Protect. Dosim., 14 (1986) 243.

[22] Majborn B., Measurements of radon in dwellings with CR-39 track detectors. Nucl. Tracks Radiat. Meas., 12 (1986) 763.

[23] Sohrabi M. and Solaymanian A. R., Indoor radon levels in some regions of Iran. Nucl. Tracks Radiat. Meas., 15 (1988) 613.

[24] Al-Kofahi M. M., Khader B. R., Lehlooh A. D., Kullab M. K., Abumurad K. M., and AL-Batania B. A., Measurements of radon-222 in Jordainan dwellings. Nucl. Tracks Radiat. Meas., 20( 2) (1992) 377.


Automatic Quantitative Bladder Measurements using MRI

Abdalmajeid Musa Alyassin* 1

Received on Sept. 17, 2006 Accepted for publication on Aug. 6, 2007

Abstract Residual Urine Volume [RUV] provides essential information for investigating many

urological patients. The standard clinical method measures RUV using post-void catheterization which involves risks of urinary infection, hemorrhage and trauma [1]. A new quantitatively automated method of the bladder is introduced using MR images. The segmentation part of this method is based on Multispectral segmentation in delineating water-like voxels. The quantitative part estimates volume and surface area of the bladder based on the Divergence Theorem Algorithm. The automated method was tested in vitro and in vivo with MR images. The in vitro accuracy was approximately 3% while the in vivo accuracy was < 5%. The automated technique was compared with the manual tracing and the catheterization techniques, where the correlation coefficients were 0.997 and 0.976 respectively. This non-invasive automated approach is introduced to measure a water-like volume in a timely fashion with a reasonable degree of accuracy using MRI

Keywords: MRI, Quantitative, Segmentation, Residual urine volume, Bladder, Multi-spectral analysis, Divergence theorem algorithm.

Introduction

Medical imaging technology has revolutionized the diagnostic imaging field by introducing new imaging modalities such as X-ray computed tomography (CT), ultrasound, and magnetic resonance imaging (MRI). These imaging modalities provide digitized sectional (tomographic) cuts of an imaged object. Tomographic medical images have afforded physicians tremendous success in diagnosing diseases. Essential diagnostic information about relative location, size, and shape of anatomical structures can be obtained from tomographic images.

A tomographic slice is made up of voxels (volume elements) with ∆x, ∆y, ∆z dimensions that make a small cuboid. Every voxel has two parameters associated with it: center location of the cuboid (x,y,z) and value. The center location corresponds to the center position of the imaged volume element and the value indicates the information representing the constituents of the cuboid. For instance, the voxel values in an MR

© 2008 by Yarmouk University, Irbid, Jordan. * Department of physics, Faculty of Science, Yarmouk University, Irbid, Jordan.

Alyassin

78

image primarily represent proton density and magnetic relaxation characteristic of tissues (T1 “spin-lattice-relaxation” and T2 “spin-spin-relaxation) of an imaged object. Tissues with similar magnetic relaxation characteristic have similar voxel values, and similarly, tissues with different magnetic relaxation characteristic have different values in an image, assuming all tissues have the same proton density concentration.

Researchers have developed methods for viewing objects in 3D from tomographic images. For instance, the Dividing Cubes Algorithm uses the (x,y,z) voxel locations and the corresponding intensity values and generates a sextuple list of (x, y , y, nx, ny, nz) coordinates for viewing purposes, where (nx,ny,nz) are the vector normal components at the (x,y,z) surface location [2]. Other researchers have taken this further by estimating the volume and surface area (Based on the Divergence Theorem Algorithm - DTA) of the 3D objects from the 3D surface representation. The DTA algorithm was only validated in vitro with computer models and CT data of inanimate objects [3]. This paper takes this research even further by providing an in vitro and in vivo evaluation to the developed automated technique using MRI images of the bladder.

The clinical importance of measuring the bladder volume is estimating the RUV, which is defined as the volume of fluid remaining in the urinary bladder immediately after micturition [4]. RUV provides essential information for investigating many urological patients such as those with suspected neuropathic bladder dysfunction, urinary incontinence, urinary tract infection, obstructive uropathic conditions, upper tract dilation, vesicoureteral reflux, and prostatic enlargement [1, 5-7]. RUV estimation serves as index of bladder function, and serial measurements of RUVs may indicate clinical progress. The standard method used in a clinical environment to measure RUV is post-void catheterization [1]. This method involves risks of urinary infection, hemorrhage, and trauma to the urethra. Although, RUV can be estimated through other non-invasive techniques such as the phenolsulfonphthalein (PSP) test, nuclear medicine, radiography, or ultrasound [8-13], they either don’t provide enough accuracy or they use ionizing radiation. Therefore, there is a need for developing a non- invasive technique that uses non-ionizing radiation and most importantly provides an accurate measurement of RUV.

Methods A special multispectral segmentation technique based on region growing was

developed to segment water-like voxels (e.g. urine). Then, a 3D surface representation (pointlist) was generated from the segmented images using the Dividing Cubes Algorithm [2]. The DTA algorithm was then used to estimate the volume and the surface area of the pointlist [3] . The estimated results were then compared to the true volumetric measurements (Figure 1).


79

Figure 1: The flowchart of the main steps used in the automated technique. The T2 median filter was implemented only for the optimization section.

1- Segmentation Technique The segmentation technique is based on multispectral analysis of the dual T2 spin

echo images [14-16]. The first echo is proton density weighted and the second echo is T2 weighted (Figure 2.A and 2.B). The scatter plot generated from the dual echo shows that water like tissues (e.g. urine) have a separated cluster (Figure 2.C).

Multispectral Segmentation

Region Growing

Dual Echo MR Images

T2 Median Filter

Scatter Plot

Radial Histogram Analysis

Annulus ROI Parameters

Water-Like Voxels

Segmented Bladder Voxels

Dividing Cubes Algorithm 3D Bladder Surface Representation

Water-like Cluster

Divergence Theorem Algorithm Quantitative Volume, Surface, & NSI

Analysis

Volumetric Measurements

Seed

Alyassin

80

Figure 2: The dual spin echo MR images. (A) The proton density weighted MR image. (B) The T2 weighted MR image. (C) The scatter plot indicating the urine cluster. (D) Six views of the 3D surface representation of the bladder (the shape of the bladder here is based on the enclosed urine).

After masking out the background clusters (shown in Figure 3.A), a radial histogram is generated from the scatter plot to determine the extents of the urine cluster (Figure 3.C). The x-axis of the radial histogram represents the angle between a ray (profile) emanating from (bottom left) until the extent of the scatter plot (right or top) (Figure 3.B). The increase in the angle represents counter-clockwise rotation of the ray. The y-axis of the radial histogram represents the area under profile (Figure 3.C). Two prominent distributions appear in the radial histogram (Figure 3.C), the first distribution with a maximum of M1 corresponds to the points of the non-urine clusters, and the second distribution with a maximum M2 corresponds to the water-like cluster (urine cluster). The width W of the second distribution is automatically estimated using the gradient descent technique. It approximates the extent (width) of the urine cluster perpendicular to the radial direction at angle M2. An annulus geometrical region of interest (ROI) was used to delineate the urine cluster automatically. The boundaries for


81

the annulus ROI were defined by T1 and T2 thresholds, and the boundary of the background cluster (Figure 3.D). Furthermore, a profile is drawn through M2 to find the center of the urine cluster (Figures 3.D and 3.E). The center of mass for the voxels corresponding to the center of the urine cluster is used as a seed for the region growing segmentation technique (Figures 1 and 3.F).

Figure 3: The main steps in generating the radial histogram and the annulus parameters to the urine cluster and the seeds to region growing segmentation.

Seed

Alyassin

82

The Dividing Cubes Algorithm is used to generate a pointlist of the segmented urine voxels. Then, the volume, the surface area, and the normalized shape index of the urine pointlist were estimated based on the DTA, as follows:

MUNCthe is n if n

zya

MUNCthe is n if n

zxa

MUNCcomponent, normal unit maxmimum the is n if n

yxa e wher

Eq(1) MUNC

a Area Surface

xixi

i

yiyi

i

zizi

i

i iii

)(

)(

)(

1

∆∆=∆

∆∆=∆

∆∆=∆

=∆= ∑∑

one to equals sumtheir and factgors weightingthe are k ,k ,k e wher

Eq(2) anzk anyky anxkVolume

zyx

iiziiz

iiyii

iixiix ∑∑∑ ∆+∆+∆=

Eq(3) volume

Area Surface Index ShapeNormalized

=

199.21

3

The automated quantitative technique was then evaluated in vitro and in vivo using MR imaging modality.

For optimizing the automated method, a special T2 medial technique was developed to reduce the spread of the clusters within the scatter plot while preserving the T2 property between the two echoes. The intensity of the MR signal in the second echo

22

02T

TE eSS−

= , is related to the MR signal in the first echo 2

01

1

TTE

eSS−

= by the

following exponential equation

( )2

12

12

TTETE

eSS

−−

=

Where S0 is the proton density and TE1 and TE2 are the time to echo in MR images and are constants, solving for T2

Eq(4)

SSTETE

T

−

=

2

1

122

ln

The median of T2 values calculated using Eq 1 determines the location of the filtered values for the 1st echo image and the 2nd Echo image. Thus T2 property is conserved in the dual echo images for voxels from corresponding locations as shown in Figure 4 [15].


83

Figure 4: The main steps of the T2 median Filter.

2-Evaluation of the Technique

2.1 In Vitro Evaluation For in vitro evaluation, a scale with precision of 0.1 cc, was used to weigh ten water

balloons before and after injection with distilled water (note that mass equals volume since the density of distilled water 1.00 g/cm3). The difference in weight of the before and after measurements was assumed to be the true volume of water in a balloon. Then these ten water balloons of different volumes ranging from 25 to 1,000 cc were scanned on a GE Signa 1.5-T machine using a dual-echo, multi-slice, T2-weighted spin-echo pulse sequence. TE1, TE2, TR which are conventional MR pulse sequence parameters were 30, 80, 2000 msec respectively, the number of excitations (NEX) was equal to 2, and the slice thickness (ST) was equal to 5 mm. The MR data was segmented and 3D surfaces (pointlists) were generated for the water balloons. The DTA volume measurement based on Eq (2) of the pointlist was compared to the true water volume. The DTA algorithm was also used to estimate surface area based on Eq (1). The normalized shape index (NSI) was also calculated using Eq (3) based on surface area and volume measurements.

2.2 In Vivo Evaluation For in vivo evaluation, the urinary bladder of nine normal subjects was scanned

using a sagittal dual-echo, contiguous multislice, T2 weighted spin-echo pulse sequence

T2 Values

Determines the location of the T2 median

Echo 2

Filtered Echo 1 Filtered Echo 2

Echo 1

Alyassin

84

(Figure 2). TR, TE1, and TE2 ranged from (2,400 to 3,500), (20 to 30), and (80 to 120) msec, respectively. NEX was equal to 2 and the voxel aspect ratio of 1:1:4 were fixed for all studies. To reduce the amount of urine formed during scans the subject fluid intake was restricted for at least 2 hr before scanning.

The MR imaged bladder volume for eight subjects was segmented by manually tracing the boundary of the urinary bladder. Then voxel counting technique was used to estimate the bladder volume. This estimation was then compared with the measurement obtained from the automated method.

In addition, the developed method was used to calculate the volume, surface area, and normalized shape index of each urinary bladder. The excreted urine was measured volumetrically using a graduated cylinder with a precision of +/-5 cc and was compared to the MRI calculated pre-void volume. The assessed pre-void volumes were corrected for the effect of the urine formed during the MR scan using the following linear model:

Eq(5) t R fVV sovoidpre +=−

where V0 is the volume of the urine in the bladder before the start of scan, f is the fraction of urine formed and imaged during the scan, R is the filling rate of the bladder [cc/min], and ts is the pre-void scan time and is equal to 12:48 minutes. The excreted urine volume after the first scan can be estimated using similar linear model as follows:

Eq(6) RUVt RVEUV so −+= where RUV is assumed zero for normal subjects. By substituting Eq 5 for V0 into Eq 6, EUV was estimated by Eq 7.

Eq(7) t R fVEUV so )1( −+=

In addition, the volumetrically measured EUV was compared to the calculated pre- minus post-void urine volume. The assessed pre and post-void volumes were corrected for the effect of the urine formed during the MR study also using a linear model as follows:

Eq(8) tt f RRUVV psvoidpost )( ++=− where tp is the preparation time between the pre-void and post void scan and is estimated to be on the average 3 minutes. Using Equations 5, 6, and 8, the EUV was estimated using both pre and post void MR scans.

Eq(9) )t(t R - EUV V - V psvoid-postvoid-pre +=

Furthermore, the urinary bladder of eight patients with prostatic enlargement was scanned once (pre-void). The scan was acquired on a Signa 1.5-T machine using sagittal dual-echo, contiguous multislice, T2-weighted spin-echo pulse sequence. TR was equal to 3 sec for all patients. The echo acquisition times were fixed, 20 msec for the first echo and 100 msec for the second echo. An attempt was made to restrict fluid intake for


85

at least 2 hr before the scan. However, in four of the eight patients, this was not possible. These four patients underwent other urologic tests that require a large quantity of fluid intake. Immediately after the MR scan, each patient was asked to void and following voiding, each patient was catheterized. The collected voided and catheterized urine volume (CUV) was measured volumetrically using a graduate cylinder (precision +/- 5 cc). The developed technique was used to calculate the pre-void urine volume for each patient. The estimated RUV was compared with the volumetrically measured CUV for each patient.

3- Optimization of the Technique

3.1 Scan Time Reduction To test scan time reduction that leads to urine formation reduction, two normal

subjects were scanned twice as described above. However, the first scan was acquired with NEX = 2 (12.48 min) and the second scan with NEX = 1 (6.24 min). The excreted urine volume was measured with a graduated cylinder and was compared with the calculated pre-void volume (for NEX = 1 and 2) obtained from the automated method for each subject.

3.2 ROI Geometrical Shape The performance of three different geometrical shapes of ROI (slope, annulus,

ellipse) were used in multispectral segmentation then compared in 11 normal subjects in order to evaluate which ROI provided the most accurate measurement of urine volume.

3.3 Image Processing Filters The scatter plot of the middle slice from eleven normal subjects was used to

investigate the reduction in the spread of the urine cluster caused by spatial filtering [15]. Three different spatial filters (Gaussian, median, and T2-Median filters) were applied to the dual-echo MR images. A rough ROI of the urine cluster in the scatter plot for both non-filtered and filtered dual-echo images was first outlined. Then, a standard automatic shrink-wrap technique was applied to the outlined ROI. This technique shrinks the size of the ROI and provides an outline containing the points within the cluster based on a simple background threshold. The percent change in the area and the maximum intensity value of the points within the standard ROI were used to indicate the cluster spread reduction in the scatter plot of the filtered to the non-filtered data.

Results and Discussion

1. In Vitro Validation For the in vitro validation, the accuracy of the volume measurements using MR

data was approximately 3% as shown in Table 1. The accuracy was determined by the

mean error ( %3%100)968.01( =×− ) and the precision was estimated by the standard deviation of the calculated versus true volumes. Although the accuracy was high, but part of the error was due to the vibrations of water within the water balloons inside the MR scanner. Also, the high accuracy of the technique was due the usage of

Alyassin

86

somewhat ideal objects. The water balloons provided good in vitro simulation of the bladder, however they are inanimate objects and not surrounded by real human anatomy which can cause inter-tissue MR relaxations. The accuracy of the NSI for the water balloons indicated the shape was not convoluted, but somewhat spherical (approximately 1). This confirms that the NSI may provide diagnostic information to urologists who are familiar with the shapes and the sizes of normal or abnormal bladder.

Table 1: In Vitro Validation of The Technique Using MRI Data of Different Size Water Balloons

Multispectral Segmentation

True Vol. Cal. Vol. Vol.TrueVol. Cal. NSI

22.8 21.6 0.947 1.192 45.8 43.8 0.956 1.081

68.8 65.8 0.956 1.062

95.8 91.4 0.954 1.065

144.5 137.0 0.948 1.023

180.3 177.0 0.982 1.051

373.6 368.7 0.987 1.041 535.6 526.2 0.982 1.038 702.6 691.8 0.985 1.037 891.0 870.8 0.977 1.039

Mean = 0.968 1.063 %SD = 1.63% 4.84%

Measurements are in [cc] and NSI stands for the normalized shape index

2. In Vivo Validation: For the in vivo evaluation, correlation coefficient between the automated method

and the manual tracing technique in estimating bladder volume for 8 subjects was 0.997 as shown in Table 2. Table 2 indicates that the accuracy of the automated technique in comparison to the manual tracing technique was < 1%. The precision of the manual tracing technique was approximately 5%. This slight increase was due to the inter- and intra-slice user variability in outlining the boundary of the urinary bladder.

Furthermore, for the in vivo validation with normal subjects, the accuracy and precision of the pre-void to the excreted urine volume measurements were < 5% as shown in Table 3. The accuracy was < 1% after the correction for urine formation during the MR scan. Unlike the pre-void data, the accuracy and precision were poor for the pre-post void to the excreted urine volume measurements as shown in Table 4. The correction of urine formation showed better results yet they were still worse than the results in Table 3. This was partly due to the lower quality MR images of the urological patient who had difficulty staying in the MR scanner during the acquisition. These images showed artifacts due to patient motion in the MR images which led to poor estimation of bladder volume. In addition, the additional estimation of parameters in Eq (9) led to increase possible error. Therefore, scan time reduction which reduce the


87

patient discomfort in the MR scanner may lead to improve the accuracy of the automated method.

Table 2: Comparison Between The Developed Technique and Manual Tracing

Manual Tracing

Developed [cc]

Diff [cc] TracingManual

Developed

Correlation Coefficient

Between Developed & Manual Tracing Techniques

551 523 28 0.949 128 135 -7 1.055 632 598 34 0.946 353 334 19 0.946 255 260 -5 1.020 78 81 -3 1.038 353 357 -4 1.011 67 70 -3 1.045

Mean = 7 1.001 SD = 17 4.70%

R2 = 0.997

Table 3: In Vivo Validation of The Technique Using the Pre-void MR Data for Normal Subjects

EUV [cc]

Vpre-img [cc] NSI

EUVV imgpre−

R [cc/min]

Est EUV [cc] EUV

EUV .Est

475 443 1.126 0.933 5.33 457 0.962 380 382 1.410 1.005 -0.33 396 1.042 180 172 1.290 0.956 1.33 186 1.033 240 239 1.099 0.996 0.17 253 1.054 310 279 1.124 0.900 5.17 293 0.945 220 191 1.097 0.868 4.83 205 0.932 353 342 1.159 0.969 1.83 356 1.008 460 455 1.217 0.989 0.83 469 1.020 258 252 1.133 0.977 1.00 266 1.031

Mean = 1.184 0.955 2.241 1.003 SD% = 10.5% 4.6% 2.244 4.5% The bladder filling rate R is estimated as (EUV - Vpre)[cc]/ 6 [min]

Alyassin

88

Table 4: In Vivo Validation of Technique Using Pre And Post-Void Data For The

Normal Subjects

EUV [cc] Vpre [cc] Vpost [cc] NSI EUVV postpre−

[cc]

R [cc/min] Est EUV [cc] EUV

EUV .Est

475 443 48 1.091 0.832 5.35 431 0.907 380 382 10 1.592 0.979 0.52 408 1.074 180 172 8 1.294 0.911 1.09 200 1.111 240 239 14 1.403 0.938 0.97 262 1.092 310 279 26 1.185 0.816 3.77 290 0.935 220 191 12 1.323 0.814 2.73 215 0.977 353 342 45 1.507 0.841 3.73 333 0.943 460 455 11 1.443 0.965 1.09 480 1.043 258 252

Mean = 313 22 1.355 0.887 2.406 1.010

SD = 110 16 0.166 0.069 1.751 0.079

The bladder filling rate R [cc/min] was estimated using Eq (9)

The mean value of the NSI for the nine normal subjects indicated that the shape of the urinary bladder was slightly deviated from the spherical shape. The precision of the NSI measurements revealed 10% variability in the shape of the filled urinary bladder for the tested normal subjects.

Correlation coefficient between the unsupervised method and the catheterization technique in estimating RUV for 8 urological patients was 0.976 as shown in Table 5. The mean and standard error difference between CUV and RUV was possible within (µ ± σ = 14 ± 23 cc) as shown in Table 5.

Table 5: Estimation of the Residual Urine Volume For Dysfunctional Voiders With MR Modality

Patient

EUV [cc]

Vpre-void [cc] NSI RUV

[cc] CUV [cc] Diff [cc]

Correlation Coefficient Between RUV & CUV

`1 520 523 1.115 65 87 -22

2 173 135 1.201 24 30 -6

3 340 598 1.129 320 378 -58

4 240 334 1.124 156 140 16

5 280 260 1.131 0 10 -10

6 70 81 1.369 25 30 -5

7 340 357 1.100 31 62 -31

8 70 70 1.259 14 10 4

Mean = 1.179 -14 SD = 9.35% 23

R2 = 0.976


89

3. Optimization of the Technique The results indicated that the error in the accuracy for both NEX =1 and NEX = 2

were < 5%. Since scan time depends linearly on NEX, this indicates that the scan time can be reduced by half (using NEX = 1) and may still lead to acceptable accuracy and precision. This also opens the door for using even faster MR pulse sequence such as the fast T2 weighted spin echo.

The annulus ROI used in the scatter plot produced the most accurate and precise measurements over the ellipse ROI or the circle ROI. This was expected because the annulus ROI simulated the MR tissue T2 window in the scatter plot.

Furthermore, applying a spatial filter to the original images produced a reduction in the spread of the urine cluster regardless of the filter used. However, the percent mean change in the cluster area within the ROI decreased for filtered data in comparison to non-filtered data. Also, the maximum intensity of the pixels within ROI increased. The reduction in the cluster area and the increase in the maximum intensity were a measure of how well the spread of the urine cluster was reduced. The T2-median filter introduced the greatest reduction in the spread of the urine cluster.

Implementing the suggestions in the optimization steps should lead to a more accurate and reproducible technique. The initial work on scan reduction should reduce the urine formed during MR scan thus improve accuracy. The T2 median filter also showed a promise in reducing the urine cluster and better improves the reproducibility of the technique in comparison to other conventional image processing filters. This technique indicated that the annulus ROI in the scatter plot provided better segmentation results than other ROI shapes. Further investigation is still necessary to validate the above optimization suggestions.

Conclusion The automated method demonstrated the ability to calculate volume of in vivo

water-like fluid such as bladder content with a high degree of accuracy and precision. The in vitro validation showed an accuracy of %3 and the in vivo RUV validation demonstrated an accuracy of < 5%. This technique introduced a new way to extract information based on MR parameters in analysing the multispectral scatter plot. The technique used non-ionizing MR imaging modality with a conventional pulse sequence. Therefore, the data can be acquired readily on any 1.5 Tesla MR scanners. The quantitative analysis was based on the DTA which for the first time was validated with in vivo MR images.

In summary, the automatic technique provided not only quantitative volume measurements but also the surface area and the normalized shape index of the bladder. Further investigation should be conducted to demonstrate the possibility of these biomarkers in the diagnosis of dysfunctional bladder.

Alyassin

90

Acknowledgements

This work would not have been done without the help of Dr. Jack Lancaster and Dr. A. Ghiatas. In addition, the catheterization of the dysfunctional voiders as well as the MR data were done and acquired at the University of Texas Health Science Center in San Antonio, USA.

ستخدام التصوير بالرنني املغناطيسيار مقاييس املثانة اوتوماتيكيا بتقدي

عبدالمجيد الياسين

ملخص

في المثانة بعد التبول معلومات مهمة في تشخيص أمراض ) م-ب-ح(حجم البول المتبقي يوفر اتي عادة مم هي القسطرة األنبوبية وال-ب-والطريقة السريرية المستخدمة لحساب ح. المسالك البولية

ان هذا البحث يقدم حال آخر . تحدث خطرا إما في تكوين تلوث في المسالك البولية او نزيف او صدمةم بطريقة -ب-ومن ثم قياس ح) م-ر(م عن طريق تصوير المريض بالرنين المغناطيسي -ب-لقياس ح

فباستخدام . التجزيئيةم بالعدة المطيافية-هذه الطريقة تعتمد على تجزئة صور ر. اوتوماتيكية جديدة .حجم للمثانى اعتمادا على نظرية األنحراف الحلولي الحسابيةالالطريقة الجديدة تقدرالمساحة الخارجية و

. م لهذه األجسام-وقد تم فحص هذه الطريقة عن قياس أجسام خارج وداخل جسم األنسان بصور رودقتها داخل % 3جسم األنسان تساوي قياس هذه الطريقة ألجسام خارج ويخلص البحث إلى ان دقة

م ومع - التخطيط اليدوي على صور ر وقد تمت مقارنة هذه الطريقة مع%. 5جسم األنسان اقل من على 0.976 و 0.977 فكانت نتيجة عوامل األرتباط المتبادل هي مقياس القسطرة للمسالك البولية

.الترتيب

References

[1] Stoller M. and Millard R., ‘The Accuracy of Catheterized Residual Urine’, Journal of Urology. 141 (1989) 15-16.

[2] Cline H., Lorenson W., Ludke S., Crawford C. and Teeter B., ‘Two algorithms for the dimensional reconstruction of tomograms’, Med Phys. 15 (1988) 320-327.

[3] Alyassin A.M., Lancaster J.L., Downs III J.H. and Fox P.T., ‘Evaluation of New Algorithms for the Measurement of Surface Area and Volume’, Medical Physics, 21 (6), (1994) 741-752.

[4] Bates P., Glen E., Griffiths D., et al, ‘Second Report on the Standardization of Terminology of Lower Urinary Tract Function’, Scand. Journal of Urology Nephral, 11 (1977) 97-199.


91

[5] Heverhagen J.T., Hatlieb T., Boehm D., Klose K.J. and Wagner H.J., ‘Magnetic resonance cystometry: accurate assessment of bladder volume with magnetic resonance imaging’, Urology, 60(2) (2002) 177-181.

[6] Hvarness H., Skjoldbye B. and Jakobsen H., ‘Urinary Bladder Volume Measurements: Comparison of Three Ultrasound Calculation Methods’, Scandinavian Journal of Urology and Nephrology, 36 (2002) 177-181.

[7] Özden E., Turgut A.T., Gögüs C., Kosar U. and Baltaci S., ‘Effect of Premicturitional Bladder Volume on the Accuracy of Postvoid Residual Urine’, J. Ultrasound Med.; 25 (2006) 831-834.

[8] Birch N., Hurst G. and Doyle P., ‘Serial Residual Volumes in Men with Prostatic Hypertrophy’, British Journal of Urology, 62 (1988) 571-575.

[9] Beacock C., Roberts E., Reels R. and Buck A., ‘Ultrasound of Residual Urine: A Quantitative Method’, British Journal of Urology, 57 (1985) 410-413.

[10] Kalis E., Likourinas M., Dermentzolgou F., Samara B. and Goulandris N., ‘Measurement of the volume of Residual Urine using 131 I-Hippuran and the Gamma Camera’, British Journal of Urology, 47 (1975) 567-570.

[11] Mason G. and Mraresh M., ‘Alteration in bladder volume and the ultrasound appearance of the cervix’, British Journal of Obstetrics and Gynaecology, 97 (1990) 457-458.

[12] Haylen B., Parys B. and West. C., ‘Transrectal Ultrasound to Measure Bladder Volumes in Men’, Journal of Urology, 143 (1990) 687-689.

[13] Smith D., ‘Estimation of the Amount of Residual Urine by Means of the Phenolsulfonphthalein Test’, Journal of Urology. 83 (1960) 188-191.

[14] Alyassin A.M., Lancaster J.L. and Ghiatas A., ‘Unsupervised Method of In Vivo Urine Volume Estimation Using MRI’, Proceedings of the International Society for Magnetic Resonance in Medicine, 3 (1996) 1587.

[15] Alyassin A.M. and Cline H.E., ‘Unsupervised Brain Segmentation Using T2 Window’, Proceedings of SPIE Medical Imaging - Image Processing, 4684 (2002) 1759-1769.

[16] Fletcher L.M., Barsotti J.B., and Hornak J.P., ‘A Multispectral Analysis of Brain Tissues', Magn. Reson. Med., 29 (1993) 623-630.


Linear Combination of SZÃSZ-Baskakov Type Operators for (LP)-Approximation

Kareem J. Thamer* 1

Received on Aug. 8, 2001 Accepted for publication on July 7, 2002

Abstract In the present paper, we study some direct results in the (LP)-approximation by linear

combination of Szãsz Baskakov type operators. The estimate of error in the approximation in term of higher order integral modulus of smoothness is obtained using the device of Steklov means.

1991 AMS Classification: 41A36, 41A25.

Keywords: Linear positive operators, Linear combination, Steklov means, ionapproximatLp − .

Introduction Gupta et. al [2] introduced Szãsz Beta operators defined as:

(1.1) ( ) ( ) ( ) ( )∑ ∫

∞

=

∞

=0 0

,,;υ

υυ dttftbxqxfB nnn

, where ),0[ ∞∈x ,

( ) ( ) ( )( )( ) 1,, 1,1

and ! ++

−

++== υ

υ

υ

υ

υ υυ nn

nx

n tnBttbnxexq

, ( )nB ,1+υ

being the Beta function given by

( ) ( )( )1

1++ΓΓ+Γ

υυn

n

. In view of the definition of the Beta function we have

( ) ( ) ( ) ( ) ( )υυυυυ υ

υ +−+ +

++== n

nnn ttn

tptnptb 11

where, ,,1,

.

We may also write (1.1) as:

© 2008 by Yarmouk University, Irbid, Jordan. * Department of Mathematics, Omran University, Maeen, Sana'a- Repubic of Yemen.

Thamer

94

(1.2) ( )( ) ( ) ( )∫

∞

∞∈=0

),0[,,; pnn LfdttftxWxtfB, where

),0[,1 ∞∈≥ xp and ( ) ( ) ( )∑

∞

=+=

0,1,,

υυυ tpxqntxW nnn

.

We note that, howsoever smooth the function may be, the order of approximation

by these operators is, at its best, of the order ( )1−nO . To improve the order of approximation, we apply the technique of linear combination to (1.2) as described by May [5] and Rathore [6] for a sequence of linear positive operators. Our approximation method has the form

(1.3) ( ) ( ) ( )∑

=

=k

jndn xfBkjCxkfB

j0

;,,,, where

( ) ( )∏≠=

=≠−

=k

jii ij

j Ckdd

dkjC

010,0 and 0for , ,

and kddd ,,, 10 K are ( )1+k arbitrary, fixed and distinct positive integers.

In what follows, we have used the notations as below:

Let 3,2,1,],[,0 132231 ==∞<<<<<<< ibaIbbbaaa iii

and ][β denote the integral part of β , also, let ( )],[.. baVB denote the class of

functions of bounded variation on ],[ ba . The semi-norm ( )],[.. baVBf

is defined by the

total variation of f on ],[ ba . Furthermore, C denotes a positive constant not necessarily the same at each occurrence.

The aim of this paper is to proof the following theorems.

THEOREM 1.1. Let 1,),0[ >∞∈ pLf p . If f has 22 +k derivatives on 1I

with ( ) ( )1

12 .. ICAf k ∈+ and

( ) ( )122 ILf p

k ∈+

, then for all n sufficiently large


95

( ) ( )( ) ( )

( )

+≤−

∞++−

),0[221

12,.,

ppp LIL

kkkILn ffnMfkfB

,where

kM is a constant independent of f and n .

THEOREM 1.2. Let ,),0[1 ∞∈ Lf If f has 12 +k derivatives on 1I with ( ) ( )12 .. ICAf k ∈ and

( ) ( )112 .. IVBf k ∈+

, then for all n sufficiently large

( ) ( )( ) ( )

( )( )

( ){ }

),0[12

..

121

121121,.,

∞+++− ++≤−

LIL

k

IVB

kkkILn fffnMfkfB

,where


THEOREM 1.3. Let 1,),0[ ≥∞∈ pLf p . Then for all n sufficiently large

( ) ( ) ( ) ( )( )),0[

11

2/122 ,,,,.,

2 ∞+−−

+ +≤−pp L

kkkILn fnIpnfMfkfB ω

, where


2. (LP)-Approximation

In this section, we shall proof the three theorems stated in section 1. In the sequel, we shall need the following lemmas:

Lemma 2.1. Let ∞<≤ p1 , ),0[ ∞∈ pLf

. Then, for sufficiently small 0>η , the

Steklov mean mf ,η of thm order corresponding to f is defined as:

( ) ( ) ( ) ( )( )∫ ∫− −

−− ∆−+=2/

2/

2/

2/1

1, ,1

η

η

η

ηη η m

mh

mmm duduufufuf KK

where

Thamer

96

1Iu∈ , ∑=

=m

iiuh

1 and ( )ufmh∆ is the

thm order forward difference of the

function f with step length h . Then, we have

(i) mf ,η has derivative up to order m over 1I , ( ) ( )1

1, .. ICAf mm ∈−

η and ( )m

mf ,η

exists a.e. and belongs to ( )1ILp ;

(ii)

( )( )

( ) mrIpfMf rr

rIL

rm

p,,2,1,,,, 1,

2K=≤ − ηωηη

;

(iii) ( )( )11, ,,,

2IpfMff mmILm

pηωη +≤−

;

(iv) ( ) ( )122, ILmILm

ppfMf +≤η

;

(v)

( )( ) ( )12

3, ILm

mIL

mm

ppfMf −

+≤ ηη,

where [ ]( )baCA ,.. stands for the class of absolutely continuous functions on [ ]ba, and sM i ' are certain constants that depend on i but are independent of f and η .

The proof of this lemma easily follows from application of [4,Theorem 18.17].

Lemma 2.2. If for 0Nr ∈ (

0N being the set of nonnegative integers)

( ) ( ) ( ) ( )( )∑ ∫∞

=

∞

++− −−=0 0

,1,,,υ

υυ dtxttpxqrnxT mrrnnmnr

.

Then,

( ) ( ) ( )( ) 1,1

11,1 1,,0,, +>−−++

== rnrn

rxxTxT nrnr and

( ) ( ) ( ) ( )( ) ( )[ ] ( )xTrxrxxmxTxxTrmn mnrmnrmnr ,,,,1,, 1211 −−−+++′=−−− +

( ) ( )xTxmx mnr 1,,22 −++.


97

Hence,

(i)

( ) ( ) ( ) ( )

( )( )2123462223 2222

2,, −−−−++++++++++

=rnrn

rrxrrnxrrnxT nr

;

(ii) ( )xT mnr ,, is a rational function in n and a polynomial in x of degree m ,

(iii) For every [ ) ( ) ( )[ ]( )2/1

,,,,0 +−=∞∈ mmnr nOxTx

.

Lemma 2.3 [3]. For Np∈ and n sufficiently large there holds

( )( ) ( ) ( ) ( ){ }1,,,, 1 oxkpQnxkxuB kpn +=− +−

,

where ( )xkpQ ,, is a certain polynomial in x of degree p and [ )∞∈ ,0x is arbitrary but fixed.

Proof of Theorem 1.1. In view of the assumptions of f , for 2Ix∈ and 1It ∈

( ) ( ) ( ) ( ) ( ) ( ) ( ) ( )∑ ∫+

=

++−+

+−

=12

0

2212

!121

!

k

j

t

x

kkjj

dwwfwtk

xfjxttf

.

Hence, we can write

(2.1)

( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( )∑ ∫+

=

++−+

+−

=12

0

2212

!121

!

k

j

t

x

kkjj

dwwftwtk

xfjxttf φ

( ) ( )( )txtF φ−+ 1, ,

where ( )tφ is the characteristic function of 1I and

( ) ( ) ( ) ( ) ( )∑+

=

−−=

12

0 !,

k

j

jj

xfjxttfxtF

, for all ),0[ ∞∈t and 2Ix∈ .

Thamer

98

Operating on this equality (2.1) by ( )xkBn ,., and breaking the right hand side

into three parts 21 , ΣΣ and 3Σ , say, corresponding to the three terms on the right hand side of (2.1), we have

( )( )

( )( )

( )

+≤Σ ++−

222

2211 IL

kIL

kIL ppp

ffnC, in view of Lemma 2.3 and

[1].

Let fh be the Hardy-Little wood majorant [8] of

( )22 +kf on 1I . Then using slder'oH && inequality and Lemma 2.2, we get

( ) ( ) ( ) ( )

−≡ ∫ ++

t

x

kkn xdwwfwttBJ ,2212

1 φ

( ) ( ) ( )

−≤ ∫ ++ xdwwfxttB

t

x

kkn ,2212φ

( )( ) ( )( )xthxttB f

kn ,22 +−≤ φ

( ) ( )( ){ } ( ) ( )( ){ } pp

fn

qqkn xtthBxtxtB

/1/122 ,, φφ+−≤

( ) ( ) ( )pb

a

p

fnk dtthtxWCn

/1

11

1

,

≤ ∫+−

.

Hence, by Fubini's theorem, Lemma 2.3 and [9, p.244] imply that

( )( ) ( ) ( )

pb

a

p

f

b

an

kIL

dtthdxtxWnCJp

/1

11

1

1

2

2

2,

≤ ∫ ∫+−

( ) ( )

( )1

1ILf

k

pthnC +−≤

( ) ( )( )1

221

IL

kk

pfnC ++−≤

.

Consequently,


99

( )( ) ( )

( )12

2212 IL

kkIL

ppfnC ++−≤Σ

.

For [ ) [ ]11 ,\,0 bat ∞∈ and 2Ix∈ there exists a 0>δ such that δ≥− xt

which gives us

( ) ( )( )( ) ( ) ( ) ( )( )( ) ( ) ( )

−+−≤− ∑

+

=

++++−12

0

222222 ,!

,,1,k

j

jkn

jk

nk

n xxtBj

xfxxttfBxtxtFB δφ

32 JJ += , say.

Applying slder'oH && inequality and Lemma 2.2, that

( ) ( )( )( ) ppn

k xtfBnCJ/11

2 ,+−≤.

Again, applying Fubini's theorem, we get

( )( )

),0[1

22 ∞

+−≤pp L

kIL

fnCJ.

Moreover, using Lemma 2.2 and [1]

( )( )

( )( )

( )

+≤ ++−

222

2213 IL

kIL

kIL ppp

ffnCJ.

Therefore,

( )( ) ( )

( )

+≤Σ +

∞+−

22

22),0[

13 IL

kL

kIL ppp

ffnC.

Combining 21 Σ−Σ , the result follows.

Proof of Theorem 1.2. By the hypothesis, for all 2Ix∈ and for 1It ∈

( ) ( ) ( ) ( ) ( ) ( ) ( ) ( )∑ ∫+

=

++−+

+−

=12

0

1212

!121

!

k

i

t

x

kkii

wdfwtk

xfi

xttf.

Hence,

Thamer

100

( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( )∑ ∫+

=

++−+

+−

=12

0

2212

!121

!

k

i

t

x

kkji

twdfwtk

xfi

xttf φ

( ) ( )( )txtF φ−+ 1, ,

where ( )tφ is the characteristic function of 1I and

( ) ( ) ( ) ( ) ( )∑+

=

−−=

12

0 !,

k

i

ii

xfi

xttfxtF, for almost all 2Ix∈ and for all ),0[ ∞∈t .

Thus

( ) ( )( ) ( ) ( )( ) ( ) ( ) ( )∑ ∫

+

=

+

−

++−=−

12

1

12

!121,,

!,,

k

i

t

x

kn

in

i

n wttBk

xkxtBi

xfxfxkfB φ

( ) ( ) )xkdwwdf k ,,22 +× ( ) ( )( )( )xktxtFBn ,,1, φ−+

321 JJJ ++= , say.

Applying Lemma 2.3 and [1], we obtain

( )( )

( )( )

( )

+≤ ++−

212121

1211 IL

kIL

kIL

ffnCJ.

To estimate 2J , we consider

( ) ( ) ( ) ( )( )21

,1212

IL

t

x

kkn xtwdfwtBK

−≡ ∫ ++ φ

( ) ( ) ( ) dtdxwdfxttxWb

a

t

x

kkn

b

a∫ ∫∫ ++−≤1

1

2

2

1212,.

Following the proof of Proposition 2.2.5 [7, p.50-52] we obtain

( )( ) ( )

( )

≤ ++−

121 ..

1212 IVB

kkIL

fnCJ.

For all [ ) ],[\,0 11 bat ∞∈ and all 2Ix∈ , we can choose a 0>δ such that δ≥− xt

, then


101

( ) ( )( )( ) ( ) ( ) ( ) ( )( )∫∫∞

−≤−0

1,,1,2

2

21dtdxttftxWxtxtFB n

b

aILn φφ

( ) ( ) ( ) ( )( )∑ ∫∫+

=

∞

−−+12

0 0

1,!

1 2

2

k

i

iin

b

a

dtdxtxtxftxWi

φ

54 JJ += , say.

For sufficiently large t , there exist positive constants 0M and C ′ such that

( ) Ct

xtk

k

′>+

−+

+

122

22

for all 20 , IxMt ∈≥ .

By Fubini's theorem

( ) ( ) ( )( )dxduttftxWJ n

b

aM

b

a

M

φ−

+= ∫∫∫∫

∞

1,2

2

2

204

76 JJ += , say.

Now, using Lemma 2.3

( ) ( ) ( ) ( )∫∫ ++− −=2

2

022

0

226 ,

b

a

kn

Mk dxdtxttftxWJ δ

( ) ( )

≤ ∫+−

0

0

1M

k dttfnC, and

( ) ( ) ( )∫∫ +−

′= +

+∞ 2

201

,122

22

7

b

ak

k

nM

dxdttft

xttxWC

J

( ) ( )∫∞

+−≤0

1

M

k dttfnC.

Hence ( )

),0[1

41 ∞

+−≤L

k fnCJ.

Thamer

102

Further, using Lemma 2.2 and [1, p.5] we obtain

( )( )

( )( )

+≤ ++−

2121

1215 IL

kIL

k ffnCJ.

The above estimates of 4J and 5J lead to

( )( ) ( )

( )

+≤ +

∞+−

21121

12),0[

13 IL

kL

kIL

ffnCJ.

The result follows.

Proof of Theorem 1.3. Let ( )tf k 22, +η be the Steklov mean of ( )thk 22 + order

corresponding to ( )tf where 0>η is sufficiently small and ( )tf is defined as zero

outside ),0[ ∞ . Then we have

( ) ( ) ( )( )

( )( )222

22,22,22, ,.,,.,,.,ILkknILknILn

pppfkfBkffBfkfB +++ −+−≤− ηηη

( )222, ILk

pff −+ +η

321 Σ+Σ+Σ= , say.

Letting ( )tφ to be the characteristic function of 3I , we get

( )( )( ) ( )( )( )( ) ( )( )( )( )( )xtfftBxtfftBxtffB knknkn ,1,, 22,22,22, +++ −−+−=− ηηη φφ

54 Σ+Σ= , say.

Clearly, the following inequality holds for 1=p , for 1>p it follows from slder'oH && inequality

( ) ( )( )∫∫ ∫ +−≤Σ3

3

2

2

2

2

., 22,4

b

a

pkn

b

a

b

a

p dtdxtfftxWdx η

Using Fubini's theorem and Lemma 2.3, we get

( ) ( )2222,4 ILkIL pp

ff +−≤Σ η.


103

Proceeding in a similar manner, for all 1≥p

( )( )

),0[22,1

52 ∞+

+− −≤Σpp Lk

kIL

ffnC η.

Consequently, by the property (iii) of Steklov means, we get

( ) ( )( )),0[

11221 ,,,

∞+−

+ +≤ΣpL

kk fnIpfC ηω

.

Since

( )( )

( )( )313

1222,..

1222, IL

kkIVB

kk ff +

+++ = ηη

.

By Theorem 1.1 and 1.2, for all 1≥p there follows

( ) ( )( )

+≤Σ

∞+++

+−

),0[22,2222,

12

3 pp LkIL

kk

k ffnC ηη

( ) ( ) ( )( )

),0[122221 ,,,

∞++−+− +≤

pLkkk fIpfnC ηωη

,

in view of the property (ii) of Steklov means.

Finally, by the property (iii) of Steklov means

( )1223 ,,, IpfC k ηω +≤Σ .

Choosing 2/1−= nη , the result follows.

Acknowledgment: We are extremely thankful to the referees for their valuable comments and

suggestions, which enabled us to improve the presentation of the paper.

Thamer

104

باستخدام الرتكيب الخطي(LP) التقريب في الفضاء (Szasz-Baskakov)باسكاكوف -مؤثرات سازلنوع من

كريم جـــبر ثامـــر

ملخص

باسـتخدام التركيـب الخطـي (LP) في هـذا البحـث قمنـا بدراسـة النتـائج المباشـرة للتقريـب فـي الفـضاء تقدير الخطـأ فـي التقريـب لمقيـاس التكامـل فـي إن. (Szasz-Baskakov) باسكاكوف-لنوع من مؤثرات ساز

Steklov) ســتكلوفة يكــون متــضمنا مــستخدمين بــذلك أدا(Smoothness)نهايــة الرتــب العاليــة بــسالسة means).

References [1] Goldberg S. and Meir A., Minimum moduli of ordinary differential operators, Proc.

London Math. Soc., 23(3) (1971) 1-15.

[2] Gupta Vijay and Srivastava G. S., On simultaneous approximation by Szãsz-Beta operators, Soochow J. Math., 21 (1995) 1-11.

[3] Gupta Vijay and Srivastava G. S., On simultaneous approximation by combinations of Baskakov-Szãsz type operators, Fasciculi Math., 27 (1997) 29-41.

[4] Hewitt E. and Stromberg K., Real and Abstract Analysis, McGraw-Hill, New York (1956).

[5] May C. P., Saturation and inverse theorems for combinations of a class of exponential type operators, Can. J. Math., XXXVIII, 6 (1976) 1224-1250.

[6] Rathore R.K.S., Linear Combinations of Linear Positive Operators and Generating Relations in Special Functions, Ph.D. Thesis, IIT Delhi (India) (1973).

[7] Sinha T.A.K., Restructured Sequences of Linear Positive Operators for Higher Order (LP)-Approximation, Ph.D. Thesis, IIT Kanpur (India) (1981).

[8] Wood B., (LP)-Approximation by linear combination of integral Bernstein type operators, Anal. Numer'r. Theor. Approx., 13(1) (1984) 65-72.

[9] Zygmund A., Trigonometrical Series, Dover Publications, Inc., New York (1955).


N

O

O

CH2 CC CH2 N

R

R

1

N

O

CH2 CC CH2 N

2

Novel Rearrangement in Tertiary Amine N-Oxides. Part II Thermal Rearrangement of Acetylenic Amine Oxides

Abdul-Hussain Khuthier Sharba* 1

Received on Sept. 5, 2002 Accepted for publication on April 9, 2003

Abstract Dialkylprop-2-ynylamine N-oxides (4) undergo thermal rearrangement to prop-1,2-dienyl-N-

hydroxylamines (5) .The rearrangement proceeds by a concerted [2,3]sigmatropic mechanism. The O-allenylhydroxylamines themselves are unstable and undergo a second rearrangement giving acrolein and imine .The second rearrangement proceeds by a concerted mechanism with 1,5-hydrogen migration.

Keywords: Acetylenic amine oxides ,Thermal rearrangement of N-oxides.

Introduction: Tertiary amine oxides are a class of compounds that posses pharmacological values.

A number of acetylenic amines have been found to be pharmacologically active compounds and their chemistry and pharmacology have received considerable attention [1-3].N,N-dialkylaminobut-2-ynylsuccinimides 1 are structurally related to oxotremorine 2 and regarded as specific anticholinergic agents [2].

Moreover,various prop-2-ynylamines that are used as drugs and pesticides exhibit diverse pharmacological effects [4]. They are among the most studied irreversible inhibitors of mitochondrial monoamine oxidase (MAO). Besides carbn oxidation ,N-oxidation was found to be an important route during metabolism of this class of © 2008 by Yarmouk University, Irbid, Jordan. * Department of Chemistry ,College of Science,Al-Mustansirya University, Baghdad, Iraq.

Sharba

106

compounds with the formation of the corresponding acetylenic amine N-oxides as distinct metabolite [5]. However ,acetylenic amine N-oxides are thermally unstable and undergo a rearrangement to O-allenylhydroxylamines thus the metabolic reaction of acetylenic amines is sometimes complicated [6] .A number of acetylenic amines are prepared and converted into the corresponding N-oxides and their rearrangement is described in this work.

Results and Discussion : N , N - dialkylprop – 2 - ynylamine N-oxides (4) undergo a novel thermal

rearrangement when heated in an inert solvent or injected into a gas chromatograph producing prop-1,2-dienyl-N-hydroxylamines (5) (scheme 1) .

The acetylenic N-oxides(4) are indicated by the band near 950-940 cm-1 due to N-O bond stretching while the rearrangement product (5) shows a sharp band near 1950cm-1

due to the asymmetric C=C=C stretching vibration [7].

Mechanism of the Rearrangement : The thermal isomerization of the acetylenic N - oxides 4 to O-

allenylhydroxylamines 5 can be envisaged to proceed by an intramolecular cyclic mechanism in which the reaction is initiated by thermal nucleophilic attack of the electron – rich oxygen on the terminal acetylenic carbon (SNi,scheme 2).

R

R

N+

O-

CH2

C

CR'

R

R

N

4

CH2

R'

CCO

5Scheme 2

R

RN

R

RN CH2CO

Scheme 1

H2C CCC[O]

R

R

N H2C CC

O

R'

R'R'

4 53

R

RN =2CH3,piperidine,2-methylpiperidine,3-methylpiperidine,

4-methylpiperidine,2,6-dimethylpiperidine,morpholine.

R'=H,(CH2)3CH3,C(CH3)3,phenyl

Novel Rearrangement in Tertiary Amine N-Oxides. Part II Thermal Rearrangement of Acetylenic Amine Oxides.

107

The rearrangement is an example of [2,3] sigmatropic migration which has been reported to occur in other acetylenic compounds for example in the sulphinates and sulphenates [9a],the sulfoxides [9b], the ammonium ylids [9c] , and the selenoxides [9d]. Furthermore ,deuterium labeled variant of 4, the N-oxide 6 , gave only the allenic product 7, which contained all of the N- oxide deuterium content of the label (scheme 3), thus clearly pointing to the intramolecularity of the rearrangement.

The allenylhydroxylamines 5 are themselves unstable compounds and undergo a second rearrangement related to the [1,5]hydrogen shifts and analogous to the electrocyclic thermal elimination of ethers of and xanthates [8] (scheme4) .

A notable example in this context is the antidepressive drug pargyline (N-benzyl-N-methylprop-2-ynylamine)which is converted to the corresponding N-oxide by biooxidation in mammal liver microsomal preparations [10].Oxidation of pargyline with m-chloroperbenzoic acid gave pargyline N-oxide in high yield .

When pargyline N-oxide (8) is refluxed in ether , THF, or CCl4,a substantial amount of N-benzylidinemethylamine is produced [11,12] .

A possible mechanism for the formation of N-benzylidinemethylamine can be rationalized through two consecutive rearrangement (scheme 5.).The second rearrangement yields the imine and propenal (acrolein)

N

O

CH2

CC

DN O CC CH2

D

6 7Scheme 3

N

H

O

HC C CH2 C

O

CH CH2

H

N+

Scheme 4

5

Sharba

108

Conclusion : As prop-2-ynylic N-oxides can be expected as metabolites to various prop-2-

ynylamines used as drugs and pesticides a knowledge of the chemical properties of the N-oxides can help explain some of the diverse pharmacological effects seen with these amines 4 . The ease with which the N-oxides generate propenal might be of utmost importance in evaluating their toxicity.

Experimental : General Procedure for the preparation of Acetylenic Amines (3):

Propargyl bromide (0.3 mole)was added to a solution of the secondary amine (0.6 mole) in 100 ml of dry ether with cooling and stirring .The reaction mixture was stirred under reflux for 10 h. The salt (secondary amine hydrobromide) was filtered and washed with ether .The ether was stripped off and the residue was distilled giving pure acetylenic amine (scheme 1.R=H).For acetylenic amines (scheme 1,R = [(CH2)3 ,C(CH3)3 ,ph] the following procedure was followed .The acetylenic compound and the secondary amine(0.3mole) were mixed .The mixture was cooled to 0C° in ice - bath and paraformaldehyde (0.36mole) and CuCl(0.06 g) in peroxide-free dioxane(20ml)were added .To the cooled reaction mixture was added glacial acetic acid (5ml) dropwise with shaking .The temperature was raised slowly to 90° and kept there for 2h.After cooling water (100ml) was added and the reaction mixture was acidified to pH 1 by dropwise addition of 1:1 HCl and extracted with two 50 ml portions of ether .

The aqueous layer was made alkaline with saturated solution of sodium carbonate and extracted with six 30 ml portions of chloroform. The residue obtained after evaporation of the solvent was passed through a column of alumina and eluted with ether .Distillation of the residue in vacou gives pure acetylenic amines.

General procedure for the Preparation of Acetylenic Amine N-Oxides (4) To a solution of the acetylenic amines 3 (0.1 mole) in 50 ml chloroform was added

with stirring a solution of m-chloroperbenzenoic acid (0.1mole) in 50 ml chloroform.

The mixture was stirred for 3h at 20C° and poured into a column of basic alumina (10 times the weight of the oxidation mixture).Traces of unreached amines were

CH2N

O

CH2

C

HC

CH3CH

NO

CCH2

CH3

HHCCHNH3C CH2CHC

O

H

+

Scheme 58

Novel Rearrangement in Tertiary Amine N-Oxides. Part II Thermal Rearrangement of Acetylenic Amine Oxides.

109

removed by washing with chloroform and the N-oxide was eluted with 3:1 chloroform – methanol. The solvent was evaporated in vacuo at 20-25 C° and the solid tertiary N – oxide was washed with dry hexane.

Rearrangement of Acetylenic Amine N- Oxides : A suspension of the acetylenic N- oxide in dry ether or dry CCl4 was stirred at

room temperature for 1h or until most of the solid disappeared .Traces of unreacted N- oxide was filtered off and the solvent evaporated in vacuo giving a pale-yellow oil of the rearrangenment product 5 which is kept at 0C° until used.

إعادة الرتتب الحراري –الجزء الثاني.االمينات الثالثية عادة الرتتب في اكاسيد إ

مينات االستيلينيةألكاسيد األ

عبد الحسين شربة

ملخص

التـي ) 4( امـين - 2 -يضم البحـث دراسـة تـأثير الحـرارة علـى بعـض اكاسـيد ثنـائي الكيـل بروباينيـل وان ) 5(الينيـل هيدروكـسيل امـين -0تعاني من ترتب حراري متحولة الى مركبات من نوع االمينـات االلينيـة ،

الينيــل هيدروكــسيل امــين -0لقــد وجــد ان مركبــات . اعــادة الترتــب تجــري بميكانيكيــة ســيكماتروب توافقيــة غير مستقرة وتعاني من عمليـة اعـادة ترتـب اخـرى معطيـة اكـرولين وااليمـين، والتـي تحـدث عـن طريـق ) 5(

.5،1-هجرة الهيدروجين

References [1] Dalhbom R., Erbing B., Olsson K., George R.and Jenden, D. J.Acta Pharm.

Suecica., 6 (1969) 349.

[2] Karlen B., Lindeke B., Lindgren S., Svensson K.G., Dahlbom R., Jenden D.J. and Giering, J.Med.Chem.,13(1970)651.

[3] Eldeen Z.M., Shubber A, Mosa, N, Muhammed A, Khalyat A. and Ghantous H. Eur. J. Med. Chem.,15(1980)85.

[4] Shirota F.N., DeMaster E.G.and Nagasawa and H.T., J.Med.Chem. ,22(1979)463. [5] (a)Hallstrom G., Lindeke B., Khuthier A. H. and Al-Iraqi M. A.,

Chem.Biol.Interactions 34 (1981) 185 . (b) Weli A. F., Anhnfelt N.O. and Lindeke B., J. Pharm.Pharmacol. 34 (1982) 771.

Sharba

110

[6] Lideke B., Drug Metab.Rev., 13 (1) (1982) 71.

[7] Taylor D.R., Chem.Rev., 67 (1967) 334.

[8] DePuy C.H. and King, R.W., Chem.Rev., 60(1960)431.

[9] (a) Smith G. and Sterling C.H.M. J.Chem.Soc., (C) (1971) 154. (b) Makisumi Y. and Takada S., J.Chem.Comm., (1974) 848. (c) Ollis W.D., Sutherland I.O.and Therbtaranonth Y. and J. Chem. Soc.Chem.Comm., (1973) 657. (d) Reich H. J.and Shah S.K., J. Am.Chem.Soc., 99 (1977) 263.

[10] Lideke B., Hallstrom G., Khuthier A.H.and Al-Iraqi M.A., Proceed.FASEB 38:3 absrt.1884 (1979).

[11] (a) Hallstrom G., Lindeke, Khuthier A.H. and Al-Iraqi, Tetrahedron Lett., 21 (1980) 667. (b) Khuthier A.H. and Al-Iraqi , Iraqi, J. Chem., 15(1) (1990)1.

[12] Khuthier A.H.and Abdel-Baqi, Al-Mustansirya J.Science., 11(1) (2000) 185.


Geochemistry of Some Opaque Mineral Oxides from Andesitic Dike, Southern Jordan

Ahmad Al-Malabeh*, Tayel El-Hasan**, Qaher Al-Qadi*** and Hasan Al-Fugha****

1

Received on Sept. 25, 2006 Accepted for publication on March 29, 2007

Abstract Detailed electron microprobe analysis of the spinel mineral group in andestic rocks from

southern Jordan; shows that the most abundant opaque oxides are titanomagnetite, ilmenite, magnetite, and chromite respectively. Also showed that early formed spinels are highly enriched in the series magnesiochromite – chromite (MgCr2O4 – FeCr2O4) in both compositional types enrichment trend induced by solid–liquid or solid-gas reactions caused an overall increase in Fe and Ti which produces the titanomagnetite, magnetite and ilmenite. The magnetite variety showed clear exsolution features, in which ilmenite lamellae and other ilmenite crystals with core of ilmenite and rims of magnetite. This might be attributed to the equilibrium during the crystallization sequence process. Moreover, the estimated temperatures at which these oxides were formed are variable and believed to be lower than 1300ºC.

Keyword: Spinels, Ex-solution, Andesite, Crystallization, Titanomagnetite, Basement

Introduction and Geologic Setting The northernmost Precambrian basement rocks of the Arabian – Nubian shield

(ANS) are exposed in southwestern Jordan. These rocks were formed during the late Precambrian Pan-African event (1-0.5 Ga ago), [1, 2, 3]. They consist of metamorphic, igneous and volcano-sedimentary sequences [1, 4, 5]. The ANS crust is composed of several extrusive and sedimentary units separated by unconformities and were subsequently deformed, metamorphosed and eventually subjected to intrusive magmatism [6]. While [7] suggested that the crust of the ANS evolved by processes such as orogenic episodes, [8] assumed that it originated through microplate accretion. [9] Summarized the evolution of the ANS, suggesting that its formation began some 950 Ma © 2008 by Yarmouk University, Irbid, Jordan. * Department of Environment and Earth Sciences, The Hashmite University, Zarqa, Jordan. ** Faculty of Sciences, Mu’tah University, Al-Karak-Jordan. *** Department of Environment and Earth Sciences, The Hashemite University, Zarqa, Jordan. **** Department of Environmental and Applied Geology, University of Jordan,Amman, Jordan

.

Al-Malabeh, El-Hasan, Al-Qadi and Al-Fugha

112

ago with rifting, followed by sea-floor spreading and the initiation of subduction which created a series of Juvenile Island arcs between 900 - 650 Ma ago. These island arcs were welded together along ophiolite-bearing suture zones followed by large-scale calc-alkalic magmatism, continental collision and accretion to the East Saharan craton [10, 11]. Moreover, [3] identified four stages of volcanism preserved in the ANS.

The crystalline basement of the Aqaba complex has been described in terms of subduction related processes involving arc and microplate accretion [1]. [12] dated the Aqaba complex between (565 - 620 Ma). [13] studied the oldest plutonic granitoids and dated them to about 620 Ma using the Rb/Sr method. Crustal evolution and its relation to the tectonics of the Arabian plate as well as late Proterozoic igneous activities were the focus of several publications [1, 2, 3, 4, 15, 16].

The crystalline basement of Wadi Turban was intruded by several andesitic dikes (Figure 1). These were the subjects of numerous studies during the last three decades. However, none of them dealt specifically with its oxide opaques content. Therefore, this is the object of the current geochemical investigation.


113

Opaques are usually found as accessories in very low concentrations in all igneous

rocks, particularly in andesites. The main opaque oxides are chromite, ilmenite and titanomagnetite. Chromite is usually one of the first phases to crystallize from mafic–ultramafic magma, thus it is an important petrological and geodynamic indicator [17, 18, 19, 20]. For example, chromites in mantle peridotites are used to identify the tectonic setting of host ophiolites [18]. Chromite has also been used to understand the origin of magmatic sulphide deposits hosted in ultramafic rocks, such as the Jinchuan Ni–Cu–(PGE) sulphide deposit in NW China [21], the Sudbury Ni–Cu–(PGE) sulphide deposit in Canada [22], and the Noril’sk Ni–Cu–(PGE) sulphide deposit in Russia [23]. Moreover, [24] identified two types of chromite (i.e. homogenous and zoned) in the Permian Jinbaoshan Pt–Pd–sulphide bearing ultramafic intrusion in SW China.

The opaque oxide minerals from this particular site were studied for the first time using reflection microscopy and more than fifty electron microprobe analyses. The results indicate that the studied rocks are chemically quite variable.


114

Sampling and Analytical Methods Thirty samples were collected from the dike in Wadi Turban (Figure 2). Thin

polished sections were studied under the reflected light. The chemical composition of 39 opaque mineral oxide grains from randomly selected samples was analyzed using an ARL-GMX electron microprobe. The analytical conditions were 1×10 -8 mA, 25-50 nA specimen current potential, 10 second integration time, and 20 kV acceleration potential: SPI mineral standards were used for the calibration process. An integration time of 10 seconds was used and on–line data reduction was done on a computer using the Bence Albee matrix at the University of Tübingen, Germany. At least three determinations were averaged from each quoted analysis.


115

Results and Discussion

The petrographic investigation indicates that spinels are the most abundant opaque mineral oxides in the andesitic rock. The abundant spinels are titanomagnetite, ilmenite, magnetite and chromite, respectively. The opaque mineral oxides form 1-4 % of the groundmass and occur as inclusions in feldspars and augites. The spinels are translucent red to opaque, and occur as irregular, angular to round anhedral grains, which range in size from 0.02 – 0.05 mm.


116

Cr-spinels are from paragenetic point of view formed in early stages and complex are in composition. They rarely exceed 0.5% of the rock by volume. Crystals of chromite commonly occur as subhedral to euhedral crystals in the groundmass and as inclusions in pyroxene or amphibole.

Magnetite grains occur as subhedral to euhedral crystals in the groundmass and as inclusions in pyroxene. In reflected light, all magnetite crystals are anisotropic and variable in color and reflection. But they are variably paler than the associated ilmenite, which can be easily identified optically by its anisotropic color and pleochroism.

Ilmenite usually occurs as primary crystals in the groundmass or as exsolution lamellae in magnetite, whereas the Fe and Ti-spinels were formed later than chromium spinels. Figure 3 shows grains of magnetite and ilmenite, which clearly demonstrate the exsolution. Figure 4 illustrates very coarse ilmenite lamellae set off by other lamellae in an intersecting texture, and the development of another set along (111) plane. Figure 5 shows ilmenite lamellae that are filled by several generations of fine lamellae. Figure 6 shows subhedral crystals of titanomagnetite containing (111) ilmenite lamellae. These titanomagnetite contains high initial values of Mg and Al. The zoning trend of these ilmenites shows a sharp increase in FeO, while the TiO2 decreases from rim to center.

Similarly in many ultramafic rocks [25, 26] ilmenites occurred as well developed microcrystal (0.2 – 0.4 mm) inclusions in silicates and as abundant, anhedral crystals in the groundmass. Furthermore, chromite occurs as a minor component. It is present as coarse subhedral to euhedral grains interlocked in the silicate matrix.

Electron microprobe analysis results are shown in Table (1). Samples 1, 2, and 3 represent chromite, which has a particularly high chromium content ranging between 31.45 – 35.60% and averaging 34.0 wt% Cr2O3. The magnesium content ranges between 4.97 – 7.11% and averages 5.49 wt% MgO. The specimens show a low titanium content (<1.3 wt% TiO2) but with very high aluminum content that ranges between 11.93 – 13.81 wt% Al2O3. The (MgO/MgO+FeO) ratio ranges from 0.16 – 0.25. The range of the Cr2O3/(Al2O3+ Cr2O3+Fe2O3) ratio was between 0.33 – 0.38. [24] Defined the chemical characteristics for two types of primary chromite. They differ mainly in the zonation of Cr2O3, Al2O3, and Fe2O3

t (total Fe), beside a wide variation in Mg, Ni, and Zn. These chemical features indicate that the type 2a chromite is the type resembling most of our studied chromites. Because they have a relatively high Cr and wide range of Mg, with variable TiO2 and lower Ni and Zn concentrations. [24] attributed the formation of this type as due to serpentinization that causes the rims of chromite crystals to be modified toward magnetite rim.


117


118

Table 1 Representative composition (wt%) of spinel in opaque oxides. Samples 1, 2, 3, and 4 represent chromite, whereas samples 5, 6, 7, 8, and 9 are magnetites.

1 2 3 4 5 6 7 8 9

SiO2 0.10 0.07 0.05 0.09 0.24 0.28 0.22 0.00 0.01

TiO2 0.83 0.87 0.93 1.33 0.09 0.04 3.25 0.13 0.09

Al2O3 13.02 13.81 13.07 11.93 0.15 0.10 0.27 0.20 0.17

Cr2O3 35.60 34.94 33.92 31.54 0.19 0.15 0.05 0.01 0.00

FeOt 40.73 40.29 43.01 47.26 98.87 97.81 97.56 97.32 98.09

MgO 7.16 6.84 6.98 4.97 0.06 0.03 0.04 0.03 0.01

CaO 0.00 0.00 0.00 0.00 0.00 0.00 0.32 0.00 0.00

MnO 0.24 0.23 0.31 1.05 0.04 0.00 0.30 0.11 0.09

Total 97.68 97.05 98.27 98.17 99.64 98.41 102.01 97.80 98.46

Si 0.02 0.01 0.01 0.01 0.04 0.08 0.09 0.00 0.00

Ti 0.14 0.15 0.16 0.23 0.02 0.01 0.12 0.03 0.01

Al 3.40 3.45 3.47 3.23 0.05 0.03 0.02 0.00 0.00

Cr 6.24 6.12 6.05 5.72 0.04 0.03 0.00 0.08 0.10

Fe 6.80 6.93 6.97 8.17 23.63 23.69 23.65 23.82 23.77

Mg 2.36 2.32 2.34 1.67 0.00 0.01 0.00 0.00 0.00

Ca 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01

Mn 0.04 0.05 0.06 0.20 0.01 0.00 0.00 0.05 0.01


119

The geochemistry of the rare magnetites is given in Table (1) (samples 5, 6, 7, 8, and 9). They show a high TiO2, ranging between 0.04 – 3.25% and with average 0.72 wt%; whereas the total iron oxide ranges between 97.32 – 98.87 wt% and averages 97.93 wt% as FeO. However, no variations in the Al2O3, Cr2O3, MgO and MnO contents were observed.

Titanomagnetite is the most abundant oxide and by far the most abundant end-member of the spinel series in the studied andesitic rock. It occurs in various quantities in more than 90 vol% of all the opaques. Table (2) shows that it has high titanium content with an average of 19.05 wt% and low MgO content with an average of 0.50 wt%. The MgO/(MgO+FeO) ratio ranges from 0.01 – 0.04.

Table 2. Representative composition (wt%) of titanomagnetites oxides.

1 2 3 4 5 6 7 8 9 10 11 12 13 14

SiO2 0.02 0.00 0.09 0.18 0.07 0.013 0.15 0.14 0.33 0.20 0.08 0.04 0.00 0.83

TiO2 20.08 10.31 17.99 15.93 20.48 21.32 21.61 21.16 22.65 27.08 28.76 17.16 21.06 20.77

Al2O3 1.64 2.32 0.44 1.40 1.28 1.06 1.04 1.64 1.78 1.37 0.09 0.31 0.09 2.30

Cr2O3 0.11 0.21 0.03 0.18 0.15 0.04 0.00 0.26 0.09 0.24 0.05 0.00 0.09 0.00

FeOt 75.70 86.48 78.74 79.25 74.61 75.09 74.34 73.89 73.35 70.10 67.86 78.51 77.05 72.55

MgO 1.03 0.72 0.40 0.89 0.99 0.03 0.00 0.02 0.03 0.03 0.48 1.66 0.15 0.08

CaO 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.32 0.00 0.00

MnO 0.90 0.55 0.87 0.76 1.10 1.80 1.24 1.89 0.57 1.98 2.16 1.23 0.45 1.08

Total 99.48 100.59 98.56 98.59 98.68 99.35 98.38 99.00 98.80 101.00 99.48 99.23 98.89 97.69

Si 0.00 0.00 0.02 0.05 0.02 0.03 0.04 0.03 0.08 0.04 0.00 0.04 0.00 0.20

Ti 3.66 1.98 3.41 3.05 3.80 3.89 3.98 3.87 4.13 4.69 5.04 3.20 4.19 3.85

Al 0.47 0.70 0.13 0.42 0.37 0.30 0.30 0.47 0.51 0.37 0.02 0.08 0.03 0.66

Cr 0.02 0.04 0.00 0.02 0.03 0.00 0.00 0.05 0.02 0.05 0.00 0.00 0.02 0.00

Fe 15.36 18.51 16.68 16.68 15.17 15.27 15.23 15.10 14.66 13.50 13.20 16.44 15.37 14.48

Mg 0.38 0.27 0.15 0.34 0.37 0.00 0.00 0.00 0.00 0.00 0.16 0.64 0.06 0.63

Ca 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01

Mn 0.19 0.12 0.19 0.15 0.23 0.37 0.26 0.39 0.18 0.39 0.40 0.24 0.10 0.22


120

Table 3. Representative composition (wt%) of ilmenite in opaque oxides.

The results of ilmenite analyses are listed in Table (3). They contain appreciable amounts of geikielite (MgTiO3β). The MgO content varies between 0.16 and 1.46 wt%, and TiO2 content ranges between 23.69 and 46.0% and averages 38.60 wt%. The Ti/(Ti+Fe) ranges from 0.33 to 0.45%. This percentage may indicate a primary ilmenite phase as defined by [27]. The MgO, Cr2O3 and MnO concentrations show only minor variation, whereas the Al content varies widely between 0.0 and 0.74 wt%. Although usually ilmenites can take up about 6% Fe2O3 into solid solutions [28], but it reaches up to 10% in the studied ilmenite. The geikielite (MgTiO3), hematite (Fe2O3) and ilmenite (FeTiO3) contents of the andesitic ilmenite are shown in Figure (7). It shows that all analyzed grains composition is similar and they have ilmenite composition with geikielite affinity.

1 2 3 4 7 8 9 10 11 12 13 14

SiO2 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.10 0.00 0.00 0.13 0.15

TiO2 42.46 40.31 42.53 40.93 45.04 41.06 44.27 46.00 45.51 38.02 38.82 35.47

Al2O3 0.64 0.43 0.28 0.21 0.36 0.39 0.07 0.04 0.00 0.74 0.62 0.89

Cr2O3 0.07 0.12 0.00 0.00 0.06 0.02 0.06 0.00 0.00 0.11 0.12 0.16

FeOt 56.74 57.71 55.86 57.01 54.36 54.86 50.79 49.98 49.70 58.72 57.63 62.88

MgO 0.68 0.83 0.44 0.97 0.77 0.16 0.85 0.81 0.86 1.35 1.46 1.33

CaO 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

MnO 0.67 0.34 0.65 0.38 0.61 0.88 1.91 1.73 1.83 1.35 0.99 0.98

Total 101.26 99.74 99.76 99.50 101.2 97.37 97.95 98.66 97.90 100.29 99.77 101.86

Si 0.00 0.00 0.00 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.03 0.03

Ti 6.71 6.53 6.83 6.62 7.03 6.82 7.17 7.29 7.27 6.18 6.31 5.77

Al 0.16 0.10 0.07 0.05 0.08 0.10 0.02 0.00 0.00 0.02 0.02 0.02

Cr 0.01 0.02 0.00 0.00 0.01 0.00 0.00 0.00 0.00 0.02 0.02 0.02

Fe 9.97 10.90 9.95 10.26 9.43 9.95 8.47 8.79 8.83 10.62 10.41 11.38

Mg 0.21 0.26 0.14 0.31 0.23 0.05 0.00 0.00 0.27 0.43 0.46 0.43

Ca 0.00 0.00 0.00 0.00 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.00

Mn 0.12 0.06 0.12 0.07 0.10 0.16 0.34 0.31 0.33 0.20 0.18 0.18


121


122

Summary and Conclusions

The opaque oxide minerals contained in the andesitic dikes from southern Jordan form only a small constituent of the rock, i.e. < 5 vol%. Electron microprobe analysis was used to determine their chemical composition. The most abundant opaque oxides are titanomagnetite, ilmenite, magnetite and chromite in decreasing order. These spinels are commonly occurring as inclusions in plagioclase and pyroxene but less common in the groundmass. Titanomagnetite and magnetite exhibit well developed crystals whereas chromites and ilmenites show irregular to rounded anhedral crystals.

The low abundance of magnetite crystals may be attributed to the presence of excess titanium during the crystallization process of the magma. Zoning is observed in titanomagnetite with ilmenite lamellae. The early-formed spinels are chromites that are characterized by high Cr2O3, FeO, and MgO contents, such as those reported by [24].


123

The andesitic spinels have high contents of Cr2O3, FeO, and MgO, as well the ilmenites has high TiO2 and FeO with noticeable contents of MgO and MnO.

The crystallization temperature of the opaque oxide minerals can be estimated using the diagram of [26] based on the system hematite – ilmenite - geikielite (Fig. 8). It indicates that the ilmenites have been crystallized at relatively low oxygen fugacity (fO2) and at temperature of less than 1300 °C.

The studied ilmenites are of primary origin, as the Cr2O3 < 1wt%, as shown in Table 3. Whereas, Cr2O3 (> 1wt%) is indicative for secondary origin, [29]. This is supported with the previously mentioned Ti/(Ti+Fe) values, which points towards the primary origin too.

The crystallization sequence of opaque oxide minerals is spinel and ilmenite in pyroxene and amphibole phenocrysts formed pre-eruption, they were followed by isolated ilmenite in the groundmass and post-eruption titanomagnetite and magnetite in the groundmass [30].

Titanomagnetite shows an inter-crystal chemical variation particularly in Cr2O3 and FeO contents as shown in Table 4, where Ti is increasing and Fe is decreasing from center to rim. Such exsolution lamella within the ilmenite crystals is resulted form the equilibration during crystallization.

Table 4. Inter-crystal chemical variation from ilmenite-titanomangnetite (Sample No. 1 and 2), and within titanomagnetite itself (Sample No. 3 and 4).

1 Core 1 Rim 2 Core 2 Rim 3 Core 3 Rim 4 Core 4 Rim

SiO2 0.16 0.12 0.08 0.18 0 0 0 0

TiO2 15.78 20.02 6.3 10.62 23.69 41.68 23..97 43.82

Al2O3 2.38 1.88 2.58 2.19 2.62 0.29 2.74 0.09

Cr2O3 0.18 0.15 0.28 0.17 0.17 0.08 0.08 0.22

FeOt 81.72 74.41 87.99 83.94 72.68 57.11 73.09 54.5

MgO 0.82 0.85 0.44 0.57 1.03 0.95 1.04 0.41

CaO 0 0 0 0 0 0 0 0.32

MnO 0.45 0.73 0.42 0.49 0.32 0.42 0.26 0.89

Total 101.48 98.16 98.09 98.16 100.51 100.53 101.18 100.25

Si 0.04 0.03 0.02 0.05 0 0 0 0

Ti 2.89 3.69 1.29 2.11 4.15 6.66 4.15 6.96

Al 0.68 0.54 0.83 0.68 0.71 0.07 0.05 0.05

Cr 0.03 0.03 0.06 0.04 0.03 0.01 0.01 0.08

Fe 16.66 15.24 19.77 18.28 14.15 10.15 14.09 9.63

Mg 0.3 0.31 0.18 0.22 0.33 0.3 0.35 0.13

Ca 0 0 0 0 0 0 0 0

Mn 0.09 0.15 0.01 0.11 0.06 0.07 0.05 0.16


124

Because the investigated andesitic dike is related to the late tectonic magmatism of the Pan-African-Orogeny and because the spinel lherzolite zone is situated at depths of 30-65 km beneath Jordan [31], therefore, the andesitic magma and their opaque oxides are more likely to be formed by partial melting of spinel lherzolite of the upper mantle.

Acknowledgments

The authors would like to express their sincere thanks to Prof. Dr. S. Kempe of Darmstadt University, and Prof. H. Greiling of Heidelberg University, Germany, for their critical review of the manuscript, which improved its quality.

طع اإلنديسايت، جنوب األردنجيوكيميائية بعض معادن األكاسيد الداكنة في قوا

حسن الفقهاء و قاهر القاضي، طايل الحسن، أحمد مالعبة

ملخص

ودة فـــي قواطـــع صـــخور نتـــائج الدراســـات التفـــصيلية لـــبعض معـــادن مجموعـــة االســـبينل الموجـــ بينـــت معــادن األكاســيد المعتمــة أن باســتخدام جهــاز المــايكروبروب ) وادي التربــان(االنديــسايت فــي جنــوب األردن

.السائدة هي بالترتيب حسب نسبة الشيوع تيتانوماغنتايت وايلمانيت وماغنيتايت و كرومايت

–جموعـــة المغنيـــسيوكرومايت وبينـــت الدراســـة ايـــضا أن االســـبينل المتـــشكل أوال كـــان مـــشبعا مـــن م و أن هذا األشباع في كـال التـركيبين متـأتي مـن تفـاعالت مـن صـلب الـى سـائل أو مـن صـلب الـى . الكرومايت

. التيتانوماغنتايت و االيلمانيت و الماغنيتايتوتكوين زيادة كلية في الحديد و التيتانيوم مما أدىغاز

أدت لوجود قطاعات و بلورات مـن االيلمينايـت فـي مركزهـا أن المحاليل المتداخله في الماغنيتايتبينت أن درجـة الحـرارة . بسبب االتزان خـالل مراحـل التبلـور وذلكايلمينايت يتحول الى ماعنيتايت على األطراف،

. درجة مئوية1300التي تكونت عندها هذه األكاسيد متغيرة و يعتقد أنها كانت أقل من

References

[1] McCourt W.J., The geology, geochemistry and tectonic setting of the granitic rocks of SW-Jordan. NRA, Amman, Unpublished report. (1988), 147-179.

[2] El-Gaby S. and Greiling O.R. (eds), The pan African belt of NE-Africa and adjacent areas, tectonic evolution and economic aspects of a late Proteozoic Orogen. Vieweg and Son, Braunschweig / Wiesbaden. (1988).

[3] Bentor Y.K., The crustal evolution of the Arabo-Nubian massif with special reference to the Sinai Peninsula. Precamb. Res., 28 (1985) 1-74.


125

[4] Bender F., Explanatory note on the geological map of Wadi Araba, Jordan. Geol. Jb. [Reihe B], 10 (1974) 3-62.

[5] Gvirtzman Z. and Garfunkel Z. Vertical movements following intra-continental magmatism: An example from southern. Israel J. Geophys. Res., 102 (B2) (1997) 2645-2658.

[6] Eyal Y. and Eyal M., Mafic dyke swarms in the Arabian-Nubian Shield. Israel J. Earth Sci., 36 (1987) 195-211.

[7] Kroener A., (1984): Late Precambrian plate tectonics and orogeny: A need to redefine the term Pan-African. In: Klerkx, J., and Michot, J. (Eds.): African Geology. Musse Royal De Afrique Central Trevuen (Belgium): 23-28.

[8] Vail J., Pan-African (Late Precambrian) tectonic terrains and the reconstruction of the Arabian-Nubian Shield. Geology., 13 (1985) 839-842.

[9] Stern R.J. and Voegeli D.A., Geochemistry, geochronology, and petrogenesis of a late Precambrian (~ 590 Ma) composite dike from NE-desert of Egypt. Geol. Rdsch. 76 (2) (1987) 325-341.

[10] Schandelmeier H., Richter A. and Harms, U., Deformation of the East Saharan Craton in southeast Libya, south Egypt and north Sudan. Tectonophysics, 140 (1987) 233-246.

[11] Gvirtzman Z. and Garfunkel Z., The transformation of southern Israel from a swell to a basin: Stratigraphic and geodynamic implications for continental tectonics. Earth and Planet. Sci. Letter., 163 (1998) 275-290.

[12] Stoeser D.B. and Camp, V.E., Pan-African microplate accretion of the Arabian shields. Bull. Geol. Soc. Amer., 96 (1985) 817-826.

[13] Brook M. and Ibrahim K., Geochronological and isotope geological investigation of the Aqaba basement complex of southern Jordan. BGS (NERC) Isotope Geol. Unit. Report., (1987) p48.

[14] Bender F., Geologie von Jordanien. Beiträge zur regionalen Geologie der Erde, Bd. 7. Berlin, Stuttgart (Borntraeger). 23 DPP., (1968).

[15] Jarrar G.; Wachendroff H. and Saffarini Gh., A late Proterozoic bimodal volcanic/subvolcanic suite from Wadi Araba, SW Jordan. Precamb. Res. 56 (1992) 51-72.

[16] Jarrar G., Wachendroff H. and Zachmann D., A pan-African alkaline pluton intruding the Saramuj conglomerate, South-west Jordan. Geol. Rdsch., 82 (1993) 121-135.

[17] Irvine T.N., Chromian spinel as a petrogenetic indicator: Part 2. Petrologic applications., Can. J. Earth Sci. 4 (1967) 71– 103.


126

[18] Dick H.J.B. and Bullen T., Chromian spinel as a petrogenetic indicator in abyssal and alpine-type peridotites and spatially associated lavas. Contrib. Mineral. Petrol. 86 (1984) 54– 76.

[19] Liipo J., Vuollo J., Nykanen V., Piirainen T., Pekkarinen L. and Tuokko I., Chromites from the early Proterozoic Outokumpu–Jormua ophiolite belt: a comparison with chromites from Mesozoic ophiolites. Lithos. 36 (1995) 15– 27.

[20] Barnes S.J. and Roeder P.L., The range of spinel compositions in terrestrial mafic and ultramafic rocks. J. Petrol. 42 (2001) 2279–2302.

[21] Barnes S.J. and Tang, Z.L., Chrome spinels from the Jinchuan Ni–Cu sulphide deposit, Gansu Province, People’s Republic of China. Econ. Geol. 94 (1999) 343–356.

[22] Zhou M.-F., Lightfoot P.C., Keays R.R., Moore M.L. and Morrison G.G., Petrogenetic significance of chromian spinels from the Sudbury Igneous Complex, Ontario, Canada. Can. J. Earth Sci. 34 (1997) 1405–1419.

[23] Barnes S.J. and Kunilov V.Y., Spinels and Mg-ilmenites from the Noril’sk and Talnakh intrusions and other mafic rocks of the Siberian flood basalt province. Econ. Geol., 95 (2000) 1701– 1717.

[24] Wang C.Y., Zhou T.M. and Zhao D., Mineral chemistry of chromite from the Permian Jinbaoshan Pt–Pd–sulphide-bearing ultramafic intrusion in SW China with petrogenetic implications. Lithos., 83 (2005) 47– 66.

[25] Haggerty S., Oxidation of opaque mineral oxides in basalts. In: Rumble, D., (Eds): Oxide minerals. Mineral Society of America., 3 (1976) 43-100.

[26] Shee S.R., The oxide minerals of the Wesselton mine Kimberlite, Kimberley, South Africa. In: Kornprobst, J. (Eds): Kimberlites and related rocks. Development in Petrology., 11 (A) (1984) 59-63.

[27] Pe-Piper G., Piper D.J.W. and Dolansky L., Alteration of Ilmenite in the cretaceous sandstones of Nova Scotia, Southeastern Canada. Clays and Clay Minerals., 53 (5) (2005) 490-510.

[28] Deer W.A., Howei R.A. and Zussmann, J., An introduction to the rock forming minerals. Longman, London, (1985) p519.

[29] El-Goresy A., Oxide minerals in lunar rocks. In: Rumble, D., (Eds): Oxide minerals. Mineral Society of America., 3 (1976) 1-43.

[30] Saffarini Gh. and Al-Fugha H., Opaque mineral oxides in Cenozoic Jordanian basalts: A preliminary study. Dirasat., 16 (2) (1989) 33-49.

[31] Al-Malabeh A., Al-Fugha H. and El Hassan, T., Petrology and Geochemistry of a Late Preicambrain Magmatic Rocks from Southern Jordan. N. Jb. Geol. Paläont. Abh., 233(3) (2004) 333-350.


Reducing the Iron Content of Hiswa Clay Deposits, South Jordan by Chemical Leaching

Talal Al-Momani* 1

Received on March 15, 2006 Accepted for publication on Sept. 12, 2006

Abstract Chemical leaching process showed that the extraction of iron content (Fe2O3%) of Hiswa

clay deposits from southeastern Jordan is possible using sodium dithionite, oxalic acid, and sulfuric acid. Iron oxides (mainly hematite) are present as very thin coating and/or cement and do not substitute in the kaolin structure. Measurements of the brightness before and after chemical leaching showed best results, i.e. degree of brightness has increased after chemical leaching. The effect of the different stages of processing and treatments by chemical leaching on the kaolinite structure of Hiswa clay deposits was studied by X-ray diffraction, transmission electron microscopy, and infrared techniques. The results indicate that the Hiswa kaolinite has not been significantly affected.

Keywords: Chemical leaching, Clay deposits, Kaolin, Iron, Jordan..

1. Introduction Kaolin is one of the most important natural industrial substances. The term is

typically used to refer to both the raw clay and the refined commercial product. According to Bates and Jackson (1982)[1], kaolin is a soft, fine, earthy, non-plastic, usually white or nearly white clay (rock) composed essentially of clay minerals of the kaolin group (principally kaolinite). Kaolinite is a common clay mineral of the kaolin group: Al2O3.2SiO2.2H2O, with an ideal composition of 39.5 wt.% Al2O3, 46.54 wt.% SiO2, and 14.0 wt.% H2O.

Kaolin is a naturally occurring white mineral. After refining, it becomes extremely white. It is relatively inert when mixed with other materials and has unique flow characteristics when incorporated or mixed with liquids or placed in fluid systems such as paints and paper industry. However, raw kaolin has few practical uses, and industrial beneficiation is necessary in order to separate non-clay minerals from the kaolin (Harvey and Murray, 1997)[2].

Kaolinite is white when pure and this is probably its most important commercial property, usually referred to as the ‘brightness’ of the kaolin. Its physical and chemical properties affect its end use. Physical properties include brightness, particle size distribution, particle shape and rheology. Color and brightness are important in many © 2008 by Yarmouk University, Irbid, Jordan. * Department of Earth and Environmental Sciences, Faculty of Science, Yarmouk University, Irbid, Jordan.

Al-Momani

128

industrial applications. Natural brightness is important in assessment of kaolin for the paper and filler markets, whereas fired color is important in kaolins for use in ceramics. However, iron oxide staining is a negative feature, and therefore the removal of coloring oxides is essential if a high brightness grade is required. In addition, brightness can be improved by size separation, magnetic separation, and chemical leaching (Bristow, 1995)[3].

The techniques used in mining and refining (beneficiation) of kaolins are divided into wet and dry methods of processing. The principle objectives of kaolin refining are the removal of impurities and the production of the desired particle-size distribution. Wet processing is generally preferred since it is more efficient in concentrating kaolinite and therefore, it is much more commonly used commercially. These processes include: attrition scrubbing, hydrocyclone separation, magnetic separation, chemical leaching, and calcinations.

Chemical leaching is a standard processing technique that is used to improve the brightness and color of kaolin. The crystal structure of kaolinite is generally resistant to attack by most corrosive fluids, so that kaolinite is regarded as an inert powder that does not react with most acids in which it is leached, which is another important commercial property.

The major chemical reaction occurring in the leaching can be achieved using a strong oxidizing agent such as sodium hypochlorite (Murray, 1980)[4], or through a strong reducing agent such as sodium dithionite (Na2S2O4) (Bloodworth et al., 1993)[5]. Kaolins are treated by lowering the pH of the slurry to about 3-4 with a suitable acid, which dissolves the iron and other minor impurities. After washing, dewatering and filtration stage, iron impurities in the crude kaolin remain in the residue. This process enabled kaolin grades to be produced with brightness values ranging from 85 to 88 (Murray, 1980)[4].

In general, different types of clay deposits were investigated by many authors in different areas around the world. Pinheiro et al. (2005)[6] studied the beneficiation of commercial kaolin from Mar de Espanha, Minas Gerais, Brazil. Saikia et al. (2003)[7]studied the characterization, beneficiation and utilization of a kaolinite clay from Assam, India.

Huge kaolin deposits are present in Hiswa clay in southern Jordan. The main goal of the present study is to improve brightness by removing the iron content of Hiswa clay deposits by chemical leaching.

2. Location and Geological Setting of the Study Area The study area lies in Aqaba Region, and it is situated in southern Jordan. The area

is located 20 km to the southeast of Ad-Disa Village at a distance of about 100 km east of Aqaba town (Figures 1).


129

Figure (1): General map showing the study area.

Al-Momani

130

Figure (2): The outcropping Shale Member.

Figure (3): Jointing in the Shale Member.


131

The principal geological sequences cropping out in the study area are of Paleozoic age (Powel et al., 1989)[8]. The Hiswa Sandstone (Graptolite Sandstone Formation) of Khreim Group is subdivided into the Shale Member and the Sandstone-Siltstone Member. The Shale Member (Figure 2) consists of kaolinite of various colors, which range from light gray, dark gray, violet, brown-red, to red. Jointing is very common in this member. Most of the joints are vertical and are filled with ferruginous material (Figure 3). The secondary ferruginous material is responsible for the coloration of the clayey beds.

Three main kaolinte rich beds were recognized as a result of the field investigations:

1- The light-grey kaolinite bed (kaolin A1).

2- The grey-dark-grey kaolinite bed (kaolin A2).

3- The violet-brick-red kaolinite bed (kaolin A3).

The total thickness of kaolin beds ranges between 20 - 24 meters in most of the study area. The thickness of the outcropping Shale Member ranges between 4 and 7 meters. The lower part of the member is clay-rich, while the upper part grades into siltstone. The kaolinite-rich beds are located northwest and southeast of the study area.

3. Materials and Methods

Sampling: Two representative bulk composite samples (Kaolin A2 and Kaolin A3) of different

varieties (about 200 kilograms each) were collected from the Hiswa Shale Member for beneficiation studies. Two varieties of Hiswa clay deposits were identified for processing and treatment depending on their Al2O3 and Fe2O3 contents.

Particle Size Analysis: The particle size of the processed materials is as follows:

a) Four samples (ten grams each) of different fractions (-2.0 mm, -0.063 mm, -0.045 mm, and - 0.020 mm) were used for chemical treatment using sodium dithionite.

b) Four samples (fifty grams each) of -0.063mm fraction of Hiswa kaolin deposits were treated chemically using different excess concentrations (acid : sample) of oxalic acid (no excess, 100% excess, 200% excess, and 300% excess).

c) Four samples (fifty grams each) of -0.063mm fraction were treated chemically using different excess concentrations (acid : sample) of sulfuric acid (no excess, 100% excess, 200% excess, 300% excess, and 400% excess).

Chemical Leaching of Samples: The adopted method in the present work is based on the publications of the Clay

Mineral Society (CMS), United States Bureau of Mines (USBM), and British Geological Survey (BGS); (Hurst, 1979[9]; Raddatz, et al., 1981[10]; Bloodworth et al., 1993[5]).

Al-Momani

132

The chemical leaching of iron was carried out using three different types of agents; a strong reducing agent: sodium dithionite (Na2S2O4), oxalic acid, and sulfuric acid. The temperature during the leaching process was kept in the range of (95 - 100°C). A flow sheet of the chemical leaching of iron for Hiswa kaolin deposits is shown in Figure (4).

Primary crusher (Jaw crusher)

Secondary crusher (Cone crusher)

Screening (dry): 2mm

Attrition Scrubbing- 2 mm

+2 mm

Wet Sieving

Chemical leaching of ironusing (Sodium Dithionite Acid)

Chemical leaching of iron using (Oxalic Acid)

-0.063mm-0.250mm

Chemical leaching of ironusing (Sulfuric Acid)

Figure (4): Flow sheet of chemical leaching of iron from Hiswa kaolin deposits.

Slurry from the leaching system was dewatered on rotary vacuum filter to remove leaching chemicals and to increase solids from approximately 30% to 55-65%. The filtration of the leached slurry was difficult because of the presence of the fine particles of the residue. After drying, the chemical composition and the brightness of the feed and the products were measured.


133

Brightness Brightness is defined as the ratio, expressed as a percentage, of the radiation

reflected by a body to that reflected by a perfectly reflecting MgO standard measured at an effective wavelength of 457 nm (Bloodworth et al., 1993)[5]. Brightness was measured with an Elrepho Photoelectric Reflection Photometer.

Chemical Analysis Chemical analysis of samples before and after leaching was carried out using

Perkin-Elmer 2280 Atomic Absorption Spectrophotometer and CECIL-599 Spectrophotometer. Mineralogical Investigations

Mineralogical investigations using X-ray diffraction techniques (XRD), transmission electron microscopy (TEM), and infrared techniques (IR) were carried out on original and leached samples.

Crystallinity Index: The crystallinity index (CI) was calculated according to Hutchison (1974)[11];

Wilson (1987)[12]; Weaver (1989)[13]; Amigo et al. (1994)[14]; Aparicio and Galan (1999)[15]; Chmielova and Weiss (2002)[16]. Accordingly, the crystallinity index for the representative samples of Hiswa kaolin deposits (before and after chemical leaching) was calculated based on the intensity of the (11 0) and (1 11) reflections. The following equation is used:

Crystallinity Index (CI) = (A + B) / Aτ

Where, A = the height of the (11 0) peak above the sloping background

B = the height of the (1 11) peak above the sloping background

Aτ= the height of the (11 0) peak above the background.

4. Results and Discussion:

Chemical Composition before and after Chemical Leaching The results of the chemical analysis of two representative bulk samples before

treatment are given in Table (1). The chemical results obtained show that SiO2, Al2O3, and Fe2O3 are the major constituents, while MgO, TiO2, Na2O, SO3, and K2O are minor ones. The results obtained indicate also that the bulk sample Kaolin A3 contains higher Fe2O3 percentage (9.15%), compared to Kaolin A2 which contains only 6.28%.

Al-Momani

134

Table (1): Results of the chemical composition of the two representative bulk samples (kaolin A2 and kaolin A3) before chemical leaching (feed).

Iron oxides (mainly hematite) are present as very thin coating and/or cement and do not substitute in the kaolin structure. Fe2O3 controls the color of the clay and reduces its brightness. Therefore, a small content of hematite (Fe2O3) which is red in color is sufficient to give a deep-red color to the clay deposits.

Table (2) gives the results of chemical analysis and brightness after chemical leaching of iron using different agents (sodium dithionite, oxalic acid, and sulfuric acid). The results show that the three leaching processes are suitable for the extraction of iron (Fe2O3%) and improvement of alumina content, brightness and color of Hiswa clay deposits. The results show also that the higher extraction ratio was obtained from samples with high iron content. In contrast, samples with low iron content show relatively lower extraction ratio. As for example, the use of sodium dithionite with samples of low iron (1.82% Fe2O3) resulted in the smallest extraction ratio of about 57%. In addition, higher extraction ratio of iron was obtained at a higher concentration ratio of oxalic acid (3:1) and sulfuric acid (4:1). In all cases however, it is possible to conclude that the chemical leaching by sodium dithionite, oxalic acid, and sulfuric acid can help to extract iron impurities and (effectively) improve the brightness of the Hiswa clay deposits.

Sample No. Oxides (%) Kaolin A2 Kaolin A3

SiO2 52.32

58.73

Al2O3 25.51 19.25 Fe2O3 6.28 9.15

TiO2 1.09 0.83 Na2O 0.08 0.09 K2O 1.42 1.39 MnO 0.01 0.04 MgO 0.15 0.16 CaO 0.08 0.15 P2O5 0.04 0.05 SO3 0.75 0.83 LOI 12.41 9.36

TOTAL 100.18 100.07


135

Table (2): Results of the chemical leaching of Hiswa clay samples.

Fe2O3 (%) Al2O3 (%) Brightness (%) Sample Number

Feed Size

(mm) Before Leaching

After Leaching

% Extraction

Before Leaching

After Leaching

Before Leaching

After Leaching

using sodium dithionite

L-A2 -2.00 6.28 1.15 81.69 25.51 28.26 50.84 73.38

L-266 -0.063 5.37 0.83 84.54 27.62 28.75 56.92 76.70

L-ATT.2 N-M

-0.045

2.90 0.81 72.07 28.53 29.00 61.40 77.85

L-EX.4-OF

-0.020

1.82 0.78 57.14 28.82 29.45 62.56 79.15

using oxalic acid

226-Head -0.063 5.37 --- --- 27.62 --- 56.92 ---

266-No Excess

-0.063

5.37 2.051 61.81 27.62 28.02 56.92 62.67

266-100% Excess

-0.063 5.37 1.550 71.14 27.62 29.16 56.92 68.74

266-200% Excess

-0.063 5.37 1.151 78.57 27.62 29.36 56.92 71.84

266-300% Excess

-0.063 5.37 1.003 81.32 27.62 29.88 56.92 74.96

using sulfuric acid

226-Head -0.063 5.37 --- --- 27.62 --- 56.92 ---

266-No Excess

-0.063

5.37 3.37 37.24 27.62 27.60 56.92 60.95

266-100% Excess

-0.063 5.37 2.77 48.42 27.62 27.85 56.92 62.57

266-200% Excess

-0.063 5.37 2.45 54.38 27.62 28.43 56.92 64.56

266-300% Excess

-0.063 5.37 2.10 60.89 27.62 28.99 56.92 65.92

266-400% Excess

-0.063 5.37 1.47 72.63 27.62 29.19 56.92 70.96

XRD Results before and after Chemical Leaching Results of the random run of XRD of the two representative bulk samples (kaolin

A2 and kaolin A3) before chemical leaching, and results of XRD after processing and treatment are listed in Table (3).

Al-Momani

136

Table (3): Results of the random XRD runs before and after chemical leaching.

Bulk Samples before processing and treatment by chemical leaching.

Serial

No.

Sample

No.

1

KA

2

QZ

3

M/I

4

GYP

5

FEL

6

ALU

7

HE

8

C.I

1 Kaolin-A2 ∗∗∗ ∗∗ ∗ ∗ ∗ 0.65

2 Kaolin-A3 ∗∗∗ ∗∗∗ ∗ ∗ ∗ ∗ ∗ 0.62

Samples after processing and treatment by chemical leaching.

Serial No.

Sample No.

1 KA

2 QZ

3 M/I

4 GYP

5 FEL

6 ALU

7 HE

8 C.I

1 L-A2 ∗∗∗ ∗∗∗ ∗ ∗ 0.82

2 L-266 ∗∗∗ ∗∗∗ ∗ ∗ 0.81

3 L-Att.2 ∗∗∗ ∗∗∗ ∗ ∗ 0.81

4 L-Ex4 ∗∗∗ ∗∗∗ ∗ ∗ 0.87

5 226-N ∗∗∗ ∗∗∗ ∗ ∗ 0.83

6 226-1 ∗∗∗ ∗∗∗ ∗ ∗ 0.81

7 266-2 ∗∗∗ ∗∗∗ ∗ ∗ 0.83

8 266-3 ∗∗∗ ∗∗∗ ∗ ∗ 0.85

Key:

1. KA. : KAOLINITE

2. QZ : QUARTZ

3. M/I : MUSCOVITE / ILLITE 4. GYP. : GYPSUM ∗∗∗ : Major, ∗∗ : Minor, ∗ :Trace

5. FEL. : FELDSPAR

6. ALU. : ALUNITE 7. HE. : HEMATITE

8. C.I: CRYSTALLINITY INDEX

The XRD results obtained for the different samples before leaching show that kaolinite and quartz are usually present as major component in the representative bulk samples. Small amounts of muscovite / illite, alunite and feldspars are also present. Hematite was encountered in kaolin A2 and kaolin A3 bulk samples, while gypsum is found only in kaolin A3 bulk sample.

Examination by XRD before and after the acid treatment indicated that the iron content of the treated products was significantly reduced (Table 3). This suggests that iron oxides in the untreated product might be present as a coating on the surfaces of


137

crystals, along with a small proportion that occurs in the matrix. Furthermore, the treatment leads to increasing the degree of kaolinite crystallinity. The crystallinity index of the studied kaolinite samples after treatment varies between (0.81 and 0.87) and indicates high degree of crystallinity (Table 3).

The results of the oriented XRD samples are given in Figures (5 and 6). The X-ray diffraction results of the clay size fractions (< 2µm) of the two samples indicated that it is composed mainly of kaolinite, while muscovite-illite and quartz are only minor constituents.

Figure (5): Typical oriented XRD patterns for air-dried, saturated with ethylene glycol, and heated at 550°C, of the clay fraction before chemical leaching (kaolin A2 sample). Numbers between brackets indicate (hkl). (K: KAOLINITE; M/I: MUSCOVITE / ILLITE; Q: QUARTZ).

Al-Momani

138

Figure (6): Typical oriented XRD patterns for air-dried, saturated with ethylene glycol, and heated at 550°C, of the clay fraction before chemical leaching (kaolin A3 sample). Numbers between brackets indicate (hkl). (K: KAOLINITE; M/I: MUSCOVITE / ILLITE; Q: QUARTZ).

The results agree with the results of the chemical analysis which indicate that kaolinite and quartz (major components) with minor muscovite/illite are the final products. The results show also that the kaolinite structure after treatment has not been significantly affected.

Results of the Brightness Measurements of the Bulk Samples before and after Chemical Leaching.

The degree of brightness is an important property of kaolinite especially when it is used in the industry. For example, natural brightness is of critical importance in the assessment of kaolinite used for paper and filler applications, whereas fired color in kaolinite is important for ceramic industry.

Results of brightness measurements of the bulk samples of Hiswa clay deposits indicate that Kaolin A2 has higher degree of brightness (50.84%), compared to that of Kaolin A3 which has a brightness of (40.47%). Table (2) gives the results of chemical analysis and brightness after chemical leaching of iron using different agents (sodium dithionite, oxalic acid, and sulfuric acid).


139

Measurements of brightness before and after chemical leaching showed that the degree of brightness has increased after chemical leaching (Table 2). High brightness is required for many industrial applications.

Results of Transmission Electron Microscopy after Chemical Leaching: The result of transmission electron microscopic study after processing and

treatment (Figure 7) shows that the well-defined pseudo-hexagonal plates of kaolinite crystal have not been affected.

Plate (7): Transmission electron micrograph of different particle sizes of well-defined pseudo-hexagonal plates of kaolinite after processing and treatment by chemical leaching (X25, 000).

Results of Infrared Spectroscopy after Chemical Leaching Results of the infrared spectroscopic analysis of the bulk composite samples

(Kaolin A2 and Kaolin A3) before treatment and processing are shown in Figures (8 and 9) respectively. Figure (10) shows the results of the bulk composite samples (Kaolin A2) after chemical leaching.

The hydroxyl-stretching spectrum of kaolinite is made up of four discrete bands. The two high-frequency bands at 3695 and 3670 cm-1 correspond to inner-surface hydroxyls which are perpendicular to ab plane, the medium-frequency band at 3650 cm-1 also corresponds to an inner-surface hydroxyl which is nearly parallel to the sheet, and

Al-Momani

140

finally the low frequency band at 3620 cm-1 for inner OH hydration) (Marel and Beutelspacher, 1976)[17]. For all the analyzed samples, the high-frequency bands at 3695 and 3670 cm-1, the medium- frequency band at 3650 cm-1, and the low frequency band at 3620 cm-1 were detected (Figures 8 and 9).

Figure (8): A. Typical Infrared absorption spectra of the bulk sample before chemical leaching (kaolin A2 sample) from 400 to 4000cm-1 before processing and treatment. B. Enlargement view of the region 400-1600cm-1. C. Enlargement view of the region 3550-3750cm-1 of the same spectra.


141

Figure (9): A. Infrared absorption spectra of the bulk sample (kaolin A3) before chemical leaching from 400 to 4000cm-1 before processing and treatment. B. Enlargement view of the region 3550-3750cm-1 of the same spectra.

Al-Momani

142

According to Marel and Beutelspacher (1976)[17], bands of large intensity are for kaolinite at 3695, 3620, 1100, 1032, 1008, 913, 694, 539, 471, and 431 cm-1. The spectra show wide variations in their intensities; especially at the higher wave numbers (3695, 3620 cm-1 region). A useful spectral region to help to distinguish between well-crystallized kaolinite and low to moderate-crystallized kaolinite is between 3650 and 3695 cm-1 (Russel, 1987[18]; Russel and Fraser, 1994[19]). In general, intense and well-defined bands at 3695 cm-1 and 3620 cm-1 and doublet at 3670-3650 cm-1 indicate a high

Figure (10): A. Infrared absorption spectra of the bulk sample (kaolin A2) from 400 to 4000cm-1 after chemical leaching using sodium dithionite acid. B. Using oxalic acid.


143

degree of crystallinity. Therefore, the crystallinity of the kaolinite of Hiswa clay deposits is considered to be moderate to high crystalline, bands at 3695 cm-1, 3620 cm-1 3670 cm-1 and 3650 cm-1 are intense and well-defined.

5. Conclusions Iron oxides (mainly hematite) are present as very thin coating and/or cement and do

not substitute in the kaolin structure. Results from the different tests of chemical leaching process showed that the extraction of iron content (Fe2O3%) of Hiswa clay deposits is possible using this process. The chemical leaching by sodium dithionite, oxalic acid, and sulfuric acid could help extracting the iron impurities from the Hiswa clay deposits.

The results of the chemical leaching show that the higher extraction percentage was obtained from samples with high iron content. In contrast, samples with low iron content show a lower extraction percentage.

Measurements of the brightness before and after chemical leaching showed that the degree of brightness has been increased after chemical leaching.

The effect of the different stages of processing and treatments by chemical leaching on the kaolinite structure of Hiswa clay deposits was studied by X-ray diffraction, transmission electron microscopy, and infrared techniques. The results indicate that the crystal habit of the Hiswa kaolinite has not been significantly affected by the treatment. The leached clay deposits may find use as a ceramic raw material and filler material for paint, paper, and other materials.

جنوب األردن-حصوة الطينية توضعات تقليل محتوى الحديد في بواسطة الغسل الكيميائي

طالل محمد المومني

ملخص

فــي توضــعات حــصوة الطينيــة علــى شــكل طبقــة رقيقــة أو كمــواد ) الهيماتيــت(توجــد أكاســيد الحديــد بواســطة الغــسل الكيميــائي ويمكــن .الصــقة بــين بلــورات الكــاؤلين وال تــدخل فــي التركيــب األساســي للكــاؤلين

التقليل من الشوائب الموجودة على شكل أكاسيد الحديد في التوضعات الطينية باسـتخدام أحمـاض مختلفـة تفيــد النتــائج بــأن المعالجــة بواســطة .دايثونايــت الــصوديوم، وحــامض األوكزاليــك، وحــامض الكبريتيــك : مثــل

ــة بقيمهــا قبــل الغــسل الكيميــائي تقلــل نــسبة أكاســيد الحديــد وتزيــد در جــة البيــاض ونــسبة األلمنيــوم مقارنوقد أثبتت النتائج المختلفة باستخدام جهـاز حيـود األشـعة الـسينية، والمجهـر اإللكترونـي .المعالجة للكاؤلين

ــأثر باألحمـــاض ــاؤلين ال يتـ ــداخلي للكـ ــوري الـ ــأن التركيـــب البلـ ــراء بـ ــعة تحـــت الحمـ ــات األشـ ــح، وتقنيـ الماسـ .سل الكيميائيالمستخدمة في عملية الغ

Al-Momani

144

References [1] Bates R. L. and Jackson J. A., Glossary of Geology, 2nd. ed.. American Geological

Institute, Virginia, USA. (1982).

[2] Harvey C. C. and Murray H. H., Industrial Clays in the 21st. Century: A Perspective of Exploration, Technology and Utilization. Applied Clay Science, 11 (1997) 285-310.

[3] Bristow C. M., Kaolin. Mineral Industry International, March (1995) 8-13.

[4] Murray H. H., Major Kaolin Processing Developments. International Journal of Mineral Processing, 7 (1980) 263-274.

[5] Bloodworth A. J., Highley D.E. and Mitchell C. J., Kaolin. British Geological Survey, UK. (1993).

[6] Pinheiro P.G., Fabris J.D., Mussel W.N., Murad E., Scorzelli R.B. and Garg V.K., Beneficiation of a commercial kaolin from Mar de Espanha, Minas Gerais, Brazil: Chemistry and mineralogy. Journal of South American Earth Sciences, 20 (2005) 267-271.

[7] Saikia N. J., Bharali D. J., Sengupta P., Bordoloi D., Goswamee R. L., Saikia P. C. and Borthakur P. C., Characterization, beneficiation and utilization of a kaolinite clay from Assam, India. Applied Clay Science, 24 (2003) 93-103.

[8] Powel J. H., Abdelhamid G., Abu Lihie O., Barjous M., Moh’d K., Khalil B., Masri A., Rabba I. and Tarawneh B., Stratigraphy and Sedimentation of the Phanerozoic Rocks in Central and South Jordan, Part A. Ram and Khreim Group. Bulletin 11A. NRA / Geology Directorate, Geological Mapping Division, Amman. (1989).

[9] Hurst V. J., (Editor), Kaolin, Bauxite and Fuller’s Earth. Clay Minerals Society, USA. (1979).

[10] Raddatz A. E., Gomes J. M. and Wong M. M., Laboratory Investigation of Sulfurous Acid Leaching of Kaolin for Preparing Alumina. United States Bureau of Mines, USA. (1981).

[11] Hutchison C. S., Laboratory Handbook of Petrographic Techniques. John Wilew and Sons, New York. (1974).

[12] Wilson M. J., A Handbook of Determinative Methods in Clay Mineralogy. Blackie, Chapman and Hall, New York. (1987).

[13] Weaver C. E., Clays, Muds and Shales. Elsevier, Amsterdam. (1989).

[14] Amigo J. M., Bastido J., Sanz A., Signes M. and Serrano J., Crystallinity of Lower Cretaceous Kaolinites of Tetuel (Spain). Applied Clay Science. 9 (1994) 51-69.

[15] Aparico P. and Galan E., Mineralogical Interference on Kaolinitte Crytallinity Index Measurements. Clays and Clay Minerals. 47 (1) (1999) 12-27.


145

[16] Chmielová M. and Weiss Z., Determination of structural disorder degree using an XRD profile fitting procedure. Application to Czech kaolins. Applied Clay Science, 22 (2002), 65-74.

[17] Marel H. W. and Beutelspacher H., Atlas of Infrared spectroscopy of Clay Minerals and their Admixtures. Elsevier Scientific Publishing Company, Amsterdam. (1976).

[18] Russell J. D., Infrared Methods. In: Wilson, M. J. (editor), A Handbook of Determinative Methods in Clay Mineralogy, 1st. edition. Blackie, USA. (1987).

[19] Russel J. D. and Fraser A. R., Infrared Methods in Clay Mineralogy: Spectroscopic and Chemical Determinative Methods. Edited by M. J. Wilson. Chapman & Hall, London. (1994).

ABHATH AL-YARMOUK: "Basic Sci. & Eng." Vol. 17, No.1A, 2008, pp.147-164

Biomarkers from Oil Shales in The Upper Cretaceous Amman Formation, NW Jordan

Hatem Darwish and Hakam Mustafa * 1

Received on May 28, 2006 Accepted for publication on Feb. 2, 2007

Abstract The presence of biomarkers such as hopanes in oil shales from the Upper Cretaceous

Amman Formation of NW Jordan indicates the presence of immature organic matter.This is also suggested by the predominance of 17 trisnorneohopane over 18 hopane. Steranes, which are traced by GC/MS at m/z = 217, indicate a source input of organic matter derived predominantly from higher marine organisms such as fish. Also, the relatively high proportion of sterane compounds shows that the palaeoenvironment in which the formation was deposited was mainly estuarine to open marine.

Keywords: Biomarkers, Oil Shale, Sterane compounds, Jordan.

Introduction The importance of biological markers in alkane stereochemistry relates to the fact

that the chiral centres of structurally-specific natural products are isomerized during maturation. This isomerization occurs because hydrogen is removed from the centres at elevated temperatures [1]. The triterpane (hopanes) and sterane polycyclics decrease in abundance during the principal stages of hydrocarbon generation, either as a result of degradation or due to dilution by the newly-generated hydrocarbons [2].

For the hopanes, the configuration at the C22 location (either the 22R, or the more stable 22S epimer), which is represented by the ratio of 22S/(22S+22R) C32, is considered to be good evidence of maturity, and increases with increasing maturity [3].

In mature source rocks, 22S epimers predominate over 22R to reach an equilibrium ratio of 60:40. This means that the 22S/(22S+22R) ratio for mature source rocks must be 0.6 [4,5], because the 22S isomer is slightly more stable than the 22R isomer [6].

Tm [17 α (H)- trisnorneohopane] is unstable, while Ts [18 α (H) trisnorhopane] is stable and more resistant to subsequent degradation. With increasing maturity, Tm is converted irreversibly into Ts [7], and the Tm/Ts ratio tends towards zero [8]. This ratio is suitable as a maturity index.

Steranes are complex mixtures of C27, C28 and C29 stereoisomers. They are good indicators of source materials, since C27 steranes are derived predominantly from

© 2008 by Yarmouk University, Irbid, Jordan. * Department of Earth and Environmental Sciences, Yarmouk University, Irbid, Jordan.

Darwish and Mustafa

148

marine organisms, while C29 steranes are derived from higher plants [9], and from higher marine organisms [10]. In general, they are derived from steroids that are present in all organisms more advanced than Cyanobacteria [11].

The aim of the present study is to evaluate the maturity of oil shale deposits located in NW-Jordan, especially those of Kufr Asad and Deir Abu Sa'ed areas due to its high importance as possible source of energy and as possible mother rocks in Jordan and Wadi Sirhan grabens. A second purpose is to determine the depositional environment of this oil shale and the origin of the organic matter in order to help reconstructing the palaeogeography of the area.

Geological Setting After not wide extedned transgressions of the Tethys onto the NW and NE margins

of Jordan during Triassic and Jurassic, predominated in Lower Cretaceous fluviatile environment covering almost the whole country with clastics, mainly sandstones and to lesser amount claystones.

Then at the beginning of Cenomanian transgressed the Tethys again covering large areas of Jordan and extending during Late Cretaceous and Palaeogene gradually southwards forming broad shallow shelf, in which deposited mainly limestone, marl, marly limestone, chert and phosphates.

The ground of the shelf was partly subdivided into long extended deeps and highs originated probably by synsedimentary tectonical movements during Coniacian to Campanian [121,13 and14].

In the deeps deposited in reducing environment during Campanian and Maastrichtian fine grained carbonates rich-in organic matter reaching values of 32%. These oil shales are concentrated in two horizons; one in the upper part of the Amman Formation, which is of Campanian to Early Maastrichtian age, and the other horizon in the Lower Muwaqqar Formation, which is of Maastrichtian-Eocene age [15].

The study area is located in N-Jordan at the eastern margin of the Jordan graben (Fig. 1). The exposed rocks in N-Jordan belong mainly to Upper Cretaceous and Palaeogene, beside the mainly fluviatile and Lacustrine Quaternary deposits and the basalt flows issued in Pliocene to Pleistocene.

The Amman Formation crops out at the surface and in most wadis in North Jordan. Lithologically, it is mainly composed of interbedded marly limestone, marl, limestone and chert layers, bands, and nodules with intercalations of phosphates and locally mainly in the lower part carbonate oil shale.It is overlain by the Al-Hisa phosphorite Formation and underlain by the Umm Ghudran Formation.

This paper deals with the oil shales in the Amman Formation west of Irbid City, between latitudes 35° 38' to 36° 14' N, and longitudes 32° 47' to 32° 57' E, in two locations: Wadi Abu-Zeyad and Kufr Asad (Fig. 1).

They are generally hard, dark-brown to black in colour, and are present as beds and massive bodies within the grey fossiliferous limestones (Fig. 2). They contain abundant


149

fish remains, shells of gastropods and cephalopods, and phosphatic intraclasts [16,17 and 18].

Darwish and Mustafa

150

The organic matter of these oil shales consists mainly of kerogen and to lesser

amount of bitumen [16,17 and18]. The overburden is less than 700m. It is expected that under normal conditions the organic matter is immature. But in the Jordan graben, where since the beginning of its formation in Late Palaeogene to Early Neogene several thousand meters sediments are accumulated and along its margin, where the effect of the movements is highest the oil shale could have been more or less affected and become more mature than elswhere. Therefore the samples have been obtained from exposures not far away from the margin of the Jordan graben.


151

Methods More than 30 samples were collected systematically (one each100 cm) from in

deep incised wadis exposed rocks of the Amman Formation after removal of weathered parts. The samples were in the field enveloped by aluminium paper, then in the lab pulverized < 63m, treated with 20% HCl to remove inorganic carbon, and burned in Leco CS-244 computerized furnace to determine TOC (Fig.3). 7 samples were selected for further analyses( Fig.3), because of the high cost of these analyses; they are 2.2, 2.3 and 2.34 from section 1, 3.5, 3.9 and 3.11 from section 2 and 4.16 from section 3.

The bitumen has been extracted from these samples by Dichloromethane soxhlet appartus, using silicate thimbles for 22 hours.

Free sulfur has been eleminated by using activated copper.

The extracted material is filtered and evaporated by a rotary evaporator at 70C under reduced pressures until a constant weight was reached.

The isolated bitumen was chromatographed on alimina - silica cel, eluting successively with pentane, 1:1 mixture of pentane/ dichloromethane and methanol to collect the saturated fraction, aromatic and the NSO fractions, respectively.

The samples were analyzed using a Hewlett-Packard 5995 GC/MS (30m X 0.25 m.I.d, cross -bonded 100% dimethylpolysiloxane; 90.52mm, film thickness; Split Ratio 30: 1; temperature rate: 120 C-340C).

The GC spectra were identified by comparison with previously-published data, especially those of Philp [11]. Detected compounds consisted of triterpanes with m/z = 191, and steranes with m/z = 217.

Darwish and Mustafa

152

Fig. 3: Schematic flow chart for laboratory analysis. EOM: Extractable Organic Matter; GC: Gas Chromatograph; HC: Hydrocarbon; MS: Mass Spectrometer; NOS:

Heterocompounds; TOC: Total organic carbon.

ROCK SAMPLE POWDERED <63µM

DICHLOROMETHANE SOXHELT EXTRACTION

20% HCL

EVAPORATION AND DRYNESS

FURNACE (LECO-CS244)

TOC EOM

LIQUID CHROMATOGRAPHY

(LC)

SATURATED HC

AROMATICS NSO ASPHALTENE

GC

GC/MS

BIOMARKERS


153

Results:

A number of hopane biomarkers were identified from the samples analyzed (Table 1, Fig. 4).

TABLE 1: LIST OF HOPANES IDENTIFIED IN THE M/Z 191 ION CHROMATOGRAMS IN FIG. 4

Peak No. Compound

1 18α (H)-22,29,30-trisnorhopone (Ts) 2 17α (H)-22,29,30-trisnorneohopone (Tm) 5 17α (H),21β (H)-30-norhopane 10 17α (H), 21β (H)-hopane 11 17β (H), 21α (H)-moretane 14 & 15 22S & 22R- 17α (H), 21β (H)-30,31- bishomohopane

The analysis of these biomarkers and the interpretation of their ratios show, that the oil shales in the study area are deposited in an estuarine or bay to open-marine environment and that the organic matter is mainly derived from zoo- and phytoplanktons beside partly from land-plants or brown algae and partly probably from higher marine animals. They show also that the oil shales did not reach maturity, but may be possible source rocks.

Darwish and Mustafa

154

Fig.4. M/z 191 (Hopane) mass fragmentograms of the saturated hydrocarbon fractions of seven representative samples from the study area(2.2, 2.30, 2.34, 3.5, 3.9, 3.11, 4.16). Peaks identification are listed in Table 1.


155

Darwish and Mustafa

156

Discussion

As shown in Table 2, the relative amounts of 22S/(22S+22R) C32 are less than 0.6. This indicates that source rocks are immature, and the ratio of Tm/Ts is very high, relative to the supposed mature value. Also, the ratio of C29/C30 17α (H)-hopane strongly indicates immaturity in the studied samples; it is around 0.35, while the "mature" level, according to Rulkotter et al., [19], requires a ratio of greater than 1.0.

Table (2): Hopanes biomarkers from 7 analyzed samples. Sample No. 22S/(22S+22R)C32 Tm/Ts C29/C30 C30/C30+Mor

2.2 0.55 10.35 0.31 0.89 2.30 0.51 16.29 0.40 0.87 2.34 0.51 8.35 0.42 0.86 3.5 0.58 7.00 0.42 0.89 3.9 0.57 12.71 0.30 0.90 3.11 0.58 10.74 0.35 0.88 4.16 0.53 8.18 0.29 0.88

The presence of 17a(H), 21 b(H) hopanes indicates that the sediments may be a possible source rock. With increasing thermal stresses, the concentrations of 17a hopane and its isomer moretane will increase [9]. Also with increasing maturity to the level of oil generation, the moretane concentration decreases and the hopane concentration increases, so that the ratio of hopane/(hopane+moretane) reaches nearly unity [20,1].

From the previous discussions of hopane biomarker distribution, it can be seen that the source rock has not reached maturity. This can be clearly observed in Fig. 4, where trisnorneohopane (Tm) greatly exceeds trisnorhopane (Ts). Also, there is the influence of bacterial input which can be traced, according to Waples [11], on the C35 of chromatograms.

The sterane distribution monitored by m/z= 217 (Fig. 5) shows the occurrences of C27-C29 rearranged (diasterane) and regular steranes for all steranes in the examined samples. Peak identifications are listed in Table 3.

Table.3: List of steranes identified in the M/z = 217 ion chromatograms in Fig. 5 Peak No. Compound 1 13 β, 17α -diacholestane (20S) 2 13β, 17α -dicholestane (20R) 10 14α,17α -cholestane (20R) 16 24-methyl - 14α,17α -cholestane(20R) 17 24-ethyl-14α -cholestane(20S) 18 24-ethyl-14β, 17β -cholestane (20R) 19 24-ethyl-14β, 17β -cholestane (20S) 20 24-ethyl-14α, 17β -cholestane (20R)


157

The configuration at the C20 chiral centre, especially for C29 steranes, has an important application for the assessment of the thermal maturity of sedimentary rocks and crude oils [1]. As maturity increases, the 20R epimer is isomerized into the 20S epimer [21, 9]. In mature source rocks, the amount of 20S relative to 20R is approximately 1:1 [21]. For the studied samples, the 20S/20R ratio of C29 steranes ranges from 0.23 to 0.75, with an average of 0.52 (Table 4), which indicates immaturity. Therefore, the rocks in the study area were not exposed to sufficient thermal stress and have not reached equilibrium.

Fig. 5. Sterane mass fragmentograms of the saturated hydrocarbon of seven

selected samples (for peak identification, see Table 3).

Darwish and Mustafa

158


159

Also, the regular 20R epimer can be converted into a mixture of 20R and 20S configurations with increasing temperature, while the rest of the molecules retain their biological configuration [10]. The ratio of 20S/(20S+20R)-C29 steranes is commonly used as a thermal maturity indicator; it rises from zero to 0.5-0.6 in the mature stage [13]. However, it does not exceed 0.43 for the examined samples (Table 4), indicating that the thermal history of these samples was inadequate for petroleum generation.

Darwish and Mustafa

160

TABLE.(4): BIOMARKERS RATIOS OF STERANES FOR 7 SAMPLES.

SAMPLES NO.

C27/C29 C28/C2920S/20S+

20R (C29) 20S/20R

(C29) C27 REAR/

C27 REG. %C27: %C28: %C29 2.2 1.16 1.25 0.28 0.40 0.12

34.0

36.7

29.3

2.30 1.21 1.21 0.34 0.52 0.13

35.4

35.3

29.3

2.34 1.15 1.32 0.19 0.23 0.37

33.1

38.0

28.9

3.5 0.85 1.54 0.40 0.68 0.55

25.1

45.4

29.5

3.9 0.72 1.25 0.36 0.57 0.36

24.3

42.0

33.6

3.11 0.92 1.26 0.39 0.65 0.17

28.9

39.7

31.4

4.16 0.92 1.21 0.38 0.75 0.20

29.4

38.6

32.0

The configuration of C20 in diasteranes (13, 17α , C27 (20S) and (20R)) depends on increasing maturity. The concentration of these rearranged steranes (diasteranes) increases in relation to the concentration of regular C27 steranes with increasing thermal stress or burial depth [22]. The relative abundance of C27-C29 regular steranes can be considered to be an important indication of the source of organic matter in oil and sediments [23], as well as a correlation tool for oil [24].

Regular C27 and C28 steranes are mainly derived from autochthonous aquatic organisms, such as zoo-and phytoplankton. Regular C29 steranes may be derived from continental higher-plant wax or from brown algae and phytoplankton [2,10 and25].

Therefore, the content of regular steranes can be used as a palaeoenvironmental indicator. In the present study, the C27, C28 and C29 contents of seven samples were plotted in a ternary diagram according to Huang and Meinschein [22] and Waples [11], to interpret the origin of the organic source material. The distribution (Fig. 6) shows a close grouping within the centre, indicating, according to this diagram, an estuarine or bay to open-marine depositional setting.


161

Fig.6. Ternary diagram of regular steranes (after Waples, [11]). Distribution of the

studied samples indicating estuarine or bay to marine ecosystem.

These palaevenvironmental indications of the biomarkers are in conform with the field observations and with the analysis of the macerals of this oil shale, which contain cuticles, spores and xylem cells [18].

The discontinuous occurrences of the oil shales, which are separated from each other by mostly primarily reddish coloured marly rocks point to the subdivision of the basin into long extended reducing deeper and oxygenated higher zones. This is also indicated by the allochems (bioclasts and phosphatic clasts) found in the oil shale, which are transported from high energy to low energy environments. The accumulation of organic matter containing pyrite and fish remains beside other nektonic organisms in the oil shales indicates reducing lower and oxidizing higher water horizons of the deeps.

As shown in Table 4, samples from Section 2 have relatively higher abundances of C27 and C28 steranes than C29 steranes, while samples from Sections 3 and 4 have higher C28 steranes than those of C29, and are depleted in C27 steranes. This depletion can be referred to the input of large quantities of higher marine organisms, or may be due to biodegradation which took place during exposure of the rocks, due to the fact that the area is highly jointed and fractured.

Darwish and Mustafa

162

Conclusion

The studied oil shales are deposited in subdivided estuarine or bay to open marine environment.

These oil shales are as indicated by the 20S/20R ratio of C29 steranes strongly immature, and their thermal history as indicated by the ratio 20S/(20S+20R) - C29 steranes is inadequate for petroleum generation. But their richness in organic matter makes them good source of energy. They could be in areas, where they are deeply buried, mother rocks. Therefore, further investigation in such areas like the Jordan graben is needed.

Acknowledgement The authors are deeply grateful to the National Authority-oil shale divsion-

specially Eng. F. Quimari and Mrs. E. Fahmawi for their valuable assistance in the extraction and analyses of the organic matter. This work received financial support from Yarmouk University Research Council (Grant No.2001/8), we wish to express our thank for this support.

،الكريتاسي العلوي-كاشفات عضوية من الصخر الزيتي في تكوين عمان

غرب األردن-شمال

حاتم درويش وحكم مصطفى

ملخص

) العلـوي _ الكريتاسـي ( دل تواجد الكاشفات العضوية كالهوبينات في الصخر الزيتي من تكوين عمـان كـذلك يـشير إلـى عـدم غيـر ناضـجة في شمال غرب األردن علـى أن المـواد العـضوية فـي هـذا الـصخر الزيتـي

ن كمـا تـشير الـستيرينات المـستدل عليهـا مـ . 18 علـى الهـوبين 17النضوج هذا سيادة الترايسنورنيوهوبين ومقيـــاس الطيـــف التثـــاقلي علـــى أن الكائنـــات الحيـــة البحريـــة المتطـــورة / فحوصـــات الكرومـــاتوغراف الغـــازي

وتـشير النـسبة العاليـة . لعضوية في هذا الصخر الزيتي كاألسماك قد أسهمت بشكل كبير في تكوين المواد ا هي بشكل رئيسي مـصبات أنهـر ، لمركبات الستيرين كذلك إلى أن البيئة القديمة التي توضع فيها هذا التكوين

.إلى بحرية مفتوحة


163

References [1] Mackenzie A.S., Applications of biological markers in petroleum geochemistry. In:

Brooks J. and Welte D., (Eds.). Advances in Petroleum Geochemistry. 1, Academic Press, NY, (1984) 115-214.

[2] Tissot B.P. and Welte D.H., Petroleum formation and occurrence, Springer-Verlag, N.Y.(1978) 538.

[3] Lu S. and Kaplan R.I., Diterpanes, triterpanes, steranes and aromatic hydrocarbons in natural bitumens and pyrolysates from different humic coals. Geochim. Cosmochim. Acta, 56 (1992) 2761-2788.

[4] Ensminger A., Van Drosselaer A., Spyckerelle C., Albrecht P. and Ourisson, G., Pentacycle triterpanes of the hopane type as ubiquitous geochemical marker: origin and significance. In: Tissot B. and Bienner, F. (Eds.). Advances in Org. Geochem. Editions Technip., Paris, ( 1973) 245-260.

[5] Selfert, W.K. and Moldowan, J.M., The effect of thermal stress on source-rock quality as measured by hopane stereochemistry. In: Douglas A.G. and Maxwell J.R., (Eds.), Advances in Org. Geoch. Pergamon Press, Oxford, (1980) 229-237.

[6] Mackenzie A.S., Patience R.L. and Maxwell J.R., Molecular changes and the maturation of sedimentary organic matter. In: G. Atkinson and J.J. Zuckerman, (Eds.), Origin and Chemistry of Petroleum. Pergamon Press, Oxford, (1979) 1-32.

[7] Goodwin N.S., Park P.J.D. and Rawlinson A.P., Crude oil biodegradation under simulated and natural conditions. In: Bjoroy, M. et al., (Eds.), Advances in Org. Geochem, John Wileys, UK, (1981)650-658.

[8] Selfert W.K. and Moldowan J.M., Applications of steranes, terpanes and monoaromatics to the maturation, migration and source of crude oils. Geochim. Cosmochim. Acta, 42 (1978) 77-95.

[9] Philp R.P., Fossil Fuel Biomarkers - Application and Spectra. Elsevier, Amsterdam, (1985) 294.

[10] Mackenzie A.S., Brassell S.C., Eglinton G. and Maxwell J.R., Chemical fossils: the geological fate of steroid, Science, 217 (1982) 491-504.

[11] Waples D.W., Geochemistry in petroleum exploration, Reidel Publ. Co., Boston (1985) 232.

[12] Masri M., Geological report of Amman – Zerqa area. Central water Authority, Jordan, (1963)74.

[13] Bender F., Geologie von Jordanien: Beitr. Zur Regionaler Geologie Der Erde, Gebrueder Borntraeger, Berlin- Stuttgart,( 1963)230.

[14] Mustafa H. and Zalmout I, On the Dentitions of Mosasauridae (Marine Reptiles) from the Late Cretaceous (Early Maastrichtian) of the Jordanian Phosphate, Dirassat, 28(1) (2001) 56-62.

Darwish and Mustafa

164

[15] Nazzal J. and Mustafa H., Ammonites from the Upper Cretaceous of North Jordan, Abhath Al-Yarmouk, Yarmouk University, Jordan, 2/2 (1993) 87-120.

[16] Darwish H., Evaluation of organic matter of oil shale Deposits from Amman Formation (Upper Cretaceous) – NW Jordan. Unpublished thesis. Yarmouk University, Irbid. (1993).

[17] Darwish H. and Mustafa H., Gas Chromatography study of Bitumen from Oil Shale of Amman Formation (Upper Cretacous), NW Jordan, Abhath Al-Yarmouk, 6(2)(1997) 93-110.

[18] Darwish H. and Mustafa, H.,Kerogen from Campanian Oil shale of NW Jordan, Abhath AL-Yarmouk, 11(1A) (2002) 95 –104.

[19] Rullkoetter J., Spiro B. and Nissenbaum A., Biological marker characteristics of oils and asphalts from carbonate source rocks in a rapidly subsiding graben, Dead Sea, Israel, Geochim. Cosmochim. Acta, 49 (1985) 1357-1370.

[20] Kinghorn R.R.F., An introduction to the physics and chemistry of petroleum, John Wileys, UK (1983) 420.

[21] Glikson M., Gibson D.L. and Philp. R.P., Organic matter in Australian Cambrian Oil Shales and other Lower Palaeozic Shales. Chem. Geol., 51 (1985) 175-191.

[22] Mackenzie A.S., Patience R.L., Maxwell J.R., Vandenbroucke M. and Durand B., Molecular parameters of maturation in the Toarcian Shales, Paris Basin, France, I. Changes in the configurations of a cyclic isoprenoid alkanes, steranes and triterpanes, Geochim. Cosmochim. Acta, 44 (1980) 1709-1721.

[23] Huang W.Y. and Meinschein W.G., Sterols as ecological indicators, Geochim. Cosmochim. Acta, 43(1979) 739-745.

[24] Lewan M.D., Bjoroy M. and Dolcater D.L., Effects of thermal maturation on steroid hydrocarbons as determined by hydrous pyrolysis of Phosphoria Retort Shale, Geochim. Cosmochim. Acta, 50 (1986) 1977-1987.

[25] Kotarba M.J. and Clayton J.L., A stable carbon isotope and biological marker study of polish bituminous coals and carbonaceous shales, International Journal of coal Geology, 55 (2003) 73 – 94.

محتويات العدد االنجليزيةالبحوث باللغة

* Automatic Vehicle Location Software Model and Geographic Information System

Mohammad Qatawneh and Yahya Aref

1

* Harmonic Elimination using Six-Step Symmetry Pulse Width Modulated Waveforms

Ibrahim A. Altawil and Fathi K. Amoura

13

* Profile Measurement using Intensity-Based Saw-Tooth Structured Light Fringe Patterns

Sami Al-Hamdan and Husam Hamad

27

* Effect of Iron Addition on the Magnetization Measurements on the Normal State Y123- Superconducting System

Khitam Khasawinah, Abdel-Rauf Al-Dairy and Yaqoub Hamam

43

* Temporal Growth of Filementation Instability Induced by Large Amplitude Laser Radiation

Mohammad Bawa'aneh, Saud Al-Awfi and Husam El-Nasser

51

* Indoor Radon-222 Concentration Measurements in Some Houses in Dura District During the Winter Season of the Year 2000

Amin Leghrouz, Mohammad Abu-Samreh and K. M. Awawdah

63

* Automatic Quantitative Bladder Measurements using MRI

Abdalmajeid Musa Alyassin 77

* Linear Combination of SZÃSZ-Baskakov Type Operators for (LP)-Approximation

Kareem J. Thamer 93

* Novel Rearrangement in Tertiary Amine N-Oxides. Part II Thermal Rearrangement of Acetylenic Amine Oxides

Abdul-Hussain Khuthier Sharba

105

* Geochemistry of Some Opaque Mineral Oxides from Andesitic Dike, Southern Jordan

Ahmad Al-Malabeh, Tayel El-Hasan, Qaher Al-Qadi and Hasan Al-Fugha

111

* Reducing the Iron Content of Hiswa Clay Deposits, South Jordan by Chemical Leaching

Talal Al-Momani 127

* Biomarkers from Oil Shales in The Upper Cretaceous Amman Formation, NW Jordan

Hatem Darwish and Hakam Mustafa

147

:التوثيق ]1[باستخدام نظـام التـرقيم يتم ذلك داخل المتن :توثيق المراجع والمصادر المنشورة -أ

.بين قوسينالتـرقيم الـوارد فـي المنـشورة فـي نهايـة البحـث حـسب تعد قائمة بالمـصادر والمراجـع •

.النص :ان المرجع كتابا يكتب هكذاإذا ك

ــع .عــزوز، دريــد [ ] ، مديريــة الكتــب والمطبوعــات، جامعــة حلــب، حلــب، ميكانيــك الموائ1979.

:واذا كان المرجع بحثا في دورية يكتب هكذا: دراسـات محمد، نوري صابر، قياس حـساسية التخريـز أثنـاء كـسر الفـوالذ الكربـوني، [ ]

.24، ص 1985، 9، ع12 مج،العلوم الهندسية والتكنولوجية :واذا كان المرجع مقالة أو فصال في كتاب يكون كالتالي

، طـرق زرع الطفيليـات التـشخيص المخبـري للطفيليـات، . عبـد الحـافظ، سـامي خـضر [ ] .181-175، ص 1986اربد، جامعة اليرموك،

ــشورة -ب ــر المن ــات كلمــة : توثيــق الهــوامش والمــصادر غي ــك فــي المــتن بإثب ــتم ذل ) هــامش(يوتــذكر المعلومــات ). 1هــامش: (متبوعــة بــالرقم المتسلــسل للهــامش داخــل قوســين، هكــذا

:التفصيلية لكل هامش في نهاية البحث تحت عنوان الهوامش وقبل قائمة المراجع فـي هـواء منـازل مدينـة اربـد، األردن، 222لـرادون خضر، بسام راشـد، قيـاس تركيـز ا :1هامش

.11-8، ص 1990رسالة ماجستير غير منشورة، جامعة اليرموك .تقبل للنشر في المجلة مراجعات الكتب الحديثة القيمة :مراجعات الكتب

.يحق لرئيس التحرير إجراء التغييرات التي يراها ضرورية ألغراض الصياغة :التصرفل من ينـشر بحثـه نـسخة واحـدة مـن عـدد المجلـة الـذي ينـشر فيـه البحـث يمنح ك :المستالت

.باإلضافة إلى عشرين مستلة منه ترسل البحوث والمراسالت إلى

رئيس تحرير مجلة "سلسلة العلوم األساسية والهندسية"أبحاث اليرموك

عمادة البحث العلمي والدراسات العليا جامعة اليرموك

اشميةالمملكة األردنية اله-اربد مـن قـسم التبـادل فـي مكتبـة جامعـة اليرمـوك، أو عمـادة أبحـاث اليرمـوك يمكن الحـصول علـى

.البحث العلمي والدراسات العليا لقاء دينار للنسخة الواحدةــا فــي : ياالشــتراك الــسنو دينــاران ونــصف فــي األردن، وثمانيــة دنــانير أو اثنــي عــشر دوالرا أمريكي

. دوالرا أمريكيا أو ما يعادلها في البلدان األخرىالوطن العربي، وثمانية عشر

بسم الله الرحمن الرحيم

أبحاث اليرموك سلسلة العلوم األساسية والهندسية

مجلة علمية محكمة تنشر المجلة البحوث األصلية التي تتوافر فيها

التي لم تقدم للنشر في أي مكان آخرالمنهجية السليمة، و

قواعد وإجراءات النشرــة • ــر :اللغ ــة وال تــستلم البحــوث بغي ــة اإلنجليزي ــة أو باللغ ــة العربي تكتــب البحــوث باللغ

.هاتين اللغتين تقدم البحوث فـي أربـع نـسخ مطبوعـة بفراغـات مزدوجـة وعلـى وجـه :تقديم البحوث •

.سم2.5واحد، وهوامش حجم الواحد منها ب أن ال يزيــد عــدد صــفحات البحــث بمــا فــي ذلــك االشــكال والرســوم والمراجــع يجــ -

.صفحة) 30(والجداول والمالحق عن يقدم الباحث ملخصين لبحثه في صفحتين منفصلتين أحدهما باللغة العربية واآلخر -

. كلمة لكل منهما200باللغة اإلنجليزية في ما ال يزيد على لف ورتبتـه العلميـة والمؤسـسة التـي يعمـل بهـا علـى يكتب عنوان البحث واسم المؤ -

صفحة منفصلة، ثم يكتب عنوان البحث مرة أخرى علـى الـصفحة األولـى مـن البحـث .وعلى صفحة كل ملخص

يقدم البحث بعد الموافقة على نشره مطبوعا ومحفوظا على قـرص كمبيـوتر قيـاس - IBM (Microsoft Word 2000, XP) إنش متوافق مع أنظمة 3.5

:االشكال والرسومات Tracing)تقدم األشكال والرسومات والجداول مرسومة بـالحبر األسـود علـى ورق شـفاف •

Paper) ــومات ــكال والرســ ــذه االشــ ــاد هــ ــاوز أبعــ ــث ال تتجــ ــصلة بحيــ ــفحات منفــ ــي صــ ، وفــ ).سم12×سم19(ية عند ورود أسماء أعالم أجنبية في البحوث المقدمة بالعرب : أسماء األعالم األجنبية •

.فإنها تكتب باللغة العربية تليها األسماء باإلنجليزية بين قوسين

هيئة التحرير

رئيس التحرير األستاذ الدكتور عبدالرحمن عطيات

األعضاء

عمـــــــر الريمـــــــاوياألســـــــتاذ الـــــــدكتور محمــــد أبوصــــالحاألســــــــتاذ الدكتـــــــــور حامـــــــد زريقـــــــات األســـــــتاذ الـــــــدكتور

مشهور الرفاعيتاذ الدكتور ـــــــــــــاألس محــمـــد الشــــــبولـور ـــــــــــــــــالدكتــــ محمد بالل الزعبيـور ـــــــــــــــــــالدكتــ

سكرتري التحرير

الحــاج عادلكــاميليا

2008جميع حقوق الطبع محفوظة لجامعة اليرموك © ال يجوز نشر أي جزء من هذه المجلة أو اقتباسه دون الحصول على

موافقة خطية مسبقة من رئيس التحرير

مجلة يعبر عن آراء أصحابه، وال يعكس آراءما يرد في ال هيئة التحرير أو سياسة جامعة اليرموك

المراجعه اللغوية ستاذ الدكتور ابراهيم جبريلاأل

جامعة اليرموك–قسم الكيمياء

خراجتنضيد وإ منار ملكاوي و فاطمة عطروز

تصميم الغالف

أحمـــــد التميمـــــي

ISSN 1023-0149

أبحـاث اليرمـوك وم األساسية والهندسيةسلسلة العل

مجلة علمية محكمة

م2008/هـ1429 أاألولالعدد عشرالسابع المجلد

ISSN 1023-0149

منشورات جامعة اليرموك عمادة البحث العلمي والدراسات العليا

أبحـاث اليرمـوك سلسلة العلوم األساسية والهندسية

مجلة علمية محكمة

م2008/هـ1429 أاألولالعدد عشرالسابعالمجلد

english covervol17 no1a - hashemite university · for a reference to an article in a periodical: [...

Documents