Professional Documents
Culture Documents
Engineering
Prof. David L. Carroll
ISSN: 1947-3931
www.scirp.org/journal/eng/
9 771947 393005
08
Editor-in-Chief
Prof. David L. Carroll
Editorial Board
Dr. Rehan Ahmed
H e r i o t-Wa t t U n i v e r s i t y, U K
J o r d a n U n i v e r s i t y o f S c i e n c e a n d Te c h n o l o g y, J o r d a n
Te c h n i c a l U n i v e r s i t y o f L o d z , Po l a n d
P u t ra U n i v e r s i t y, M a l a y s i a
E c o l e C e n t ra l e d e N a n t e s , F ra n c e
Dr. P. Chandramohan
A n n a U n i v e r s i t y, I n d i a
T h e U n i v e r s i t y o f To k y o , J a p a n
S o u t h e a s t U n i v e r s i t y, C h i n a
C o v e n t r y U n i v e r s i t y, U K
Ka n a g a w a U n i v e r s i t y, J a p a n
I N H A U n i v e r s i t y, Ko r e a ( S o u t h )
S H U -T E U n i v e r s i t y, Ta i w a n ( C h i n a )
Prof. Yong-Gang Lv
C h o n g q i n g U n i v e r s i t y, C h i n a
I E E E P h o t o n i c s S o c i e t y, U K
U . S . D e p a r t m e n t o f E n e r g y / N E T L- P i t t s b u r g h , U S A
M o s c o w Po w e r E n g i n e e r i n g U n i v e r s i t y, R u s s i a
Ts i n g h u a U n i v e r s i t y, C h i n a
H a r b i n I n s t i t u t e o f Te c h n o l o g y, C h i n a
T i a n j i n U n i v e r s i t y, C h i n a
Wu h a n U n i v e r s i t y, C h i n a
Prof. Zhao Xu
Te c h n i c a l U n i v e r s i t y o f D e n m a r k , D e n m a r k
Tr e n d M i c r o , U S A
Wa s e d a U n i v e r s i t y, J a p a n
C h a n g s h a U n i v e r s i t y o f S c i e n c e a n d Te c h n o l o g y, C h i n a
C e r e s I n c ., T h o u s a n d O a k s , U S A
C h o n g q i n g U n i v e r s i t y, C h i n a
Editorial Assistant
Yanna Li
eng@scirp.org
TABLE OF CONTENTS
Volume 2
Number 8
August 2010
Electrochemical Generation of Zn-Chitosan Composite Coating on Mild Steel and its Corrosion Studies
K. Vathsala, T. V. Venkatesha, B. M. Praveen, K. O. Nayana580
Lie Group Analysis for the Effects of Variable Fluid Viscosity and Thermal
Radiation on Free Convective Heat and Mass Transfer with Variable Stream Condition
P. Loganathan, P. P. Arasu625
Statistical Modeling of Pin Gauge Dimensions of Root of Gas Turbine Blade in Creep Feed Grinding
Process
A. R. Fazeli635
Wind Turbine Tower Optimization under Various Requirements by Using Genetic Algorithm
S. Yldrm, . zkol641
ENG
Engineering
Journal Information
SUBSCRIPTIONS
Engineering (Online at Scientific Research Publishing, www.SciRP.org) is published monthly by Scientific Research Publishing, Inc.,
USA.
Subscription rates:
Print: $50 per issue.
To subscribe, please contact Journals Subscriptions Department, E-mail: sub@scirp.org
SERVICES
Advertisements
Advertisement Sales Department, E-mail: service@scirp.org
Reprints (minimum quantity 100 copies)
Reprints Co-ordinator, Scientific Research Publishing, Inc., USA.
E-mail: sub@scirp.org
COPYRIGHT
Copyright2010 Scientific Research Publishing, Inc.
All Rights Reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by
any means, electronic, mechanical, photocopying, recording, scanning or otherwise, except as described below, without the
permission in writing of the Publisher.
Copying of articles is not permitted except for personal and internal use, to the extent permitted by national copyright law, or under
the terms of a license issued by the national Reproduction Rights Organization.
Requests for permission for other kinds of copying, such as copying for general distribution, for advertising or promotional purposes,
for creating new collective works or for resale, and other enquiries should be addressed to the Publisher.
Statements and opinions expressed in the articles and communications are those of the individual contributors and not the statements
and opinion of Scientific Research Publishing, Inc. We assumes no responsibility or liability for any damage or injury to persons or
property arising out of the use of any materials, instructions, methods or ideas contained herein. We expressly disclaim any implied
warranties of merchantability or fitness for a particular purpose. If expert assistance is required, the services of a competent
professional person should be sought.
PRODUCTION INFORMATION
For manuscripts that have been accepted for publication, please contact:
E-mail: eng@scirp.org
Abstract
Quality of cement is evaluated via group of tests. The most important, and close to understanding, is the
compressive strength test. Recently, Egyptian standards adopted the European standards EN-196 and
EN-197 for specifying and evaluating quality of cements. This was motivated by the large European investments in the local production of cement. The current study represents a comparative investigation, experimental and numerical, of the effect of different parameters on evaluation of compressive strength. Main parameters are shape of specimens and type of sand used for producing tested mortars. Three sets of specimens
were made for ten types of cements. First set were 70.6 mm cubes molded according to old standards using
single sized sand. Second group were prisms molded from standard sand (CEN sand) according to the recent
standards. Third group were prisms molded from local sand sieved and regenerated to simulate same grading
of CEN sand. All specimens were cured according to relevant standards and tested at different ages (2,3,7,10
and 28 days). Results show that CEM-I Type of cement does not fulfill, in all of its grades, the strength requirements of Ordinary Portland cement OPC specified in old standards. Also, the use of simulated CEN
sand from local source gives strengths lower than those obtained using standard certified CEN sand. A limited number of tests were made on concrete specimens from two most common CEM-I types to investigate
effect on concrete strength and results were also reported. Numerical investigation of the effect of specimen
shape and type of sand on evaluation of compressive strength of mortar specimens, presented in the current
study, applies one of the artificial intelligence techniques to simulate and predict the strength behavior at
different ages. The Artificial Neural Network (ANN) technique is introduced in the current study to simulate
the strength behavior using the available experimental data and predict the strength value at any age in the
range of the experiments or in the future. The results of the numerical study showed that the ANN method
with less effort was very efficiently capable of simulating the effect of specimen shape and type of sand on
the strength behavior of tested mortar with different cement types.
Keywords: Cement Type, Sand Type, Mortar Specimen, Strength, Modeling, Artificial Neural Network
1. Introduction
For decades, engineers used to apply cement based on
certain classification [1-3]. This classification refers to
its composition and consequently relevant properties.
Among these properties, strength was the main target of
using a specific type of cement. Ordinary Portland cement (OPC), sulphate resisting cement (SRC) and white
cement share almost same values for compressive
strength at different ages. One type: namely rapid hardCopyright 2010 SciRes.
560
H. HODHOD ET AL.
models to simulate flow behavior in open channel infested by submerged aquatic weeds. Mohamed [14] proposed an artificial neural network for the selection of
optimal lateral load-resisting system for multi-story steel
frames. Mohamed, in her master thesis, proposed the
neural network to reduce the computing time consumed
in the design iterations. Abdeen [15] utilized ANN technique for the development of various models to simulate
the impacts of different submerged weeds' densities, different flow discharges, and different distributaries operation scheduling on the water surface profile in an experimental main open channel that supplies water to different distributaries.
2. Problem Description
To study the effect of specimens shape and types of sand
used for producing tested mortars on evaluation of compressive as well as flexural tensile strengths, experimental and numerical techniques will be presented in this
study. The experimental program and its results will be
described in detail in the following sections. After that,
numerical approach will be discussed to show the efficiency of numerical techniques. The numerical models
presented in this study utilized Artificial Neural Network
technique (ANN) using the data of experiments and then
can predict the strength value in the range of the experiment or in the future.
3. Experimental Program
The experimental program focuses on evaluating compressive strength of mortar made from new cement types.
Ten types of cements with different grades and rate of
hardening were procured from local market in Egypt.
Compressive strength was evaluated for each type using
the cubic specimens (70.6 mm side length) and using the
testing of part of prism (40*40*160 mm). The last method was employed twice. First with local sand following
the same grade specified in ES4756 (and EN 196), and
second with certified CEN sand according to same standards. Specimens were tested at ages of 2, 3, 7, 10 and
28 days.
Concrete mixes with same proportions were cast from
different types of cements. Slump and compressive
strength were measured for each mix to investigate the
effect of type of cement on concrete properties. Compressive strength was measured at 3, 7 and 28 days.
4.1. Water
Tap water was used for mixing and curing of all speciENG
H. HODHOD
mens.
4.2. Cement
Ten types of cement were used. All were supplied in
bags carrying the symbols of both ES4756 and EN-197.
They were all produced locally in Egypt by different
Cement Companies. The ten types covered CEM I (ordinary Portland cement) with different grades and rates of
hardening. The types also included white cement, sulphate resisting cement (SRC) and CEM II type cements.
Table 1 shows the investigated types of cement.
ET AL.
561
4.3. Sand
Two types of sand were used for the current study: CEN
sand that was imported from France in sealed transparent
bags (Figure 1), and local siliceous sand. Local siliceous
sand was used in its natural grading (Figure 2) for casting concrete. This sand was sieved to get the single size
sand required for testing mortar cubes according to old
ES (still effective as part of local code of practice ECP
203/2001 app.3. The local sand was also used to regenerate the CEN sand by collecting different sizes in the percentages specified in EN-196. The grading of this regenerated sand, and limits of CEN sand, are shown in Figure 3.
Percentage Passing
4.4. Gravel
Local siliceous gravel was used for casting concrete specimens. Gravel has a maximum nominal size of 20 mm.
100
80
60
40
20
0
0.01
0.1
10
Type of Cement
Manufacturer
CEM I-52.5N
SINAI
El-MASRIYA
HELWAN
CEM I-42.5N
(SRC)(**)
CEM I-42.5N (SRC)
CEM I-42.5N
El-MASRIYA
CEM I-42.5R
HELWAN
CEM I-42.5N
HELWAN
(White)
7
CEM I-32.5R
ELKAUMIYA (NCC)
8
CEM I-32.5R (SRC)
ASSIUT (CEMEX)
9
CEM II-B-S 32.5N
HELWAN
10
CEM II-B-L -32.5N
ELKAUMIYA (NCC)
(*) Sand having a grading similar to CEN sand.
(**) This cement will be denoted in figures as SRC-1.
6
ENG
H. HODHOD
562
ET AL.
Percentage Passing
100
80
60
40
20
0
0.01
0.1
10
similar to CEN was used. Constituents were mixed mechanically using 5 liter mixer. Steel molds were used for
casting. For all sets, specimens were compacted using
vibrator and left covered with impervious sheet for 24
hours. Then specimens were demolded and immersed in
water till day of testing. Concrete cubes (with 150 mm
side length) were cast to evaluate concrete strength. Mix
proportions were water: cement: sand: gravel = 0.6:1:
1.5:3.0. Constituents were mixed mechanically using 140
liter tilting type mixer. Dry materials were mixed first for
about one minute. Then, water was added gradually and
mixing continued till uniform mix was obtained. Concrete was cast in steel molds and compacted using a vibrating table. Specimens were covered with plastic
sheets for 24 hours. Then molds were removed and specimens were wet cured till age of testing.
5. Test Results
The test results are explained in the following sections.
H. HODHOD
ET AL.
563
Figure 5. Compressive strength of mortar specimens produced under different conditions, for different types of cement.
ENG
H. HODHOD ET AL.
564
Figure 6. Strength factor for mortar prisms produced from CEN sand (left: rapid setting cements, right: normal setting cements).
Figure 7. Strength factor for mortar prisms produced from regenerated sand (left: rapid setting cements, right: normal setting cements).
Figure 8. Strength factor for mortar cubes (left: rapid setting cements, right: normal setting cements).
H. HODHOD
ET AL.
565
Figure 9. Flexural strength of mortar prisms produced different sand types, for different types of cement.
ENG
H. HODHOD ET AL.
566
Figure 10. Flexural strength vs. compressive strength (left: cen sand, right: regenerated sand).
30
25
CEM I 52.5N
CEM I 42.5N-SRC
20
CEM I 42.5N
CEM I 42.5R
CEM I 42.5N white
15
CEM I 32.5R
CEM I 32.5R (SRC)
10
CEM II-B-S32.5N
CEM II B-L32.5N
5
0
0
10
20
30
Age - days
Neural networks are models of biological neural structures. Briefly, the starting point for most networks is a
model neuron as shown in Figure 15. This neuron is connected to multiple inputs and produces a single output.
Each input is modified by a weighting value (w). The
neuron will combine these weighted inputs with reference
to a threshold value and an activation function, will determine its output. This behavior follows closely the real
neurons work of the humans brain. In the network structure, the input layer is considered a distributor of the signals from the external world while hidden layers are considered to be feature detectors of such signals. On the
other hand, the output layer is considered as a collector of
the features detected and the producer of the response.
30
25
20
CEM I 52.5N
CEM I 42.5N-SRC
15
CEM I 42.5N
CEM I 42.5R
10
5
0
0
10
20
ET AL.
Concrete Compressive Strength (MPa)
H. HODHOD
567
30
25
20
CEM I 32.5R
CEM I 32.5R-SRC
15
CEM II-B-S32.5N
CEM II B-L32.5N
10
5
0
0
30
10
20
30
Age - days
Age - days
30
25
20
CEM I 42.5R
15
CEM I 32.5R
CEM I 32.5R-SRC
10
5
0
0
10
20
30
Figure 13. Measured concrete compressive strength (left: high grades of cement, right: low grades of cement).
30
25
CEM I 52.5N
20
CEM I 42.5N-SRC
CEM I 42.5N
15
10
CEM II B-L32.5N
5
0
0
Age - days
10
20
30
Age - days
Figure 14. Measured concrete compressive strength (left: rapid setting cements, right: normal setting cements).
f x
Figure 15. Typical picture of a model neuron that exists in
every neural network.
(1)
Y j Fth U j t j
(2)
and
1
1 e x
(3)
ENG
H. HODHOD ET AL.
568
and,
ej
(4)
Yj 1 Yj d j Yj
(5)
7. Simulation Models
To fully investigate numerically the effect of specimen
shape and type of sand on the strength behavior of tested
mortar with different cement types, seven models are
Type of Cement
CEM I-52.5N
CEM I-42.5N
(SRC) (**)
CEM I-42.5N
(SRC)
CEM I-42.5N
CEM I-42.5R
CEM I-42.5N
(White)
CEM I-32.5R
CEM I-32.5R
(SRC)
CEM II-B-S-32.5N
CEM II-B-L-32.5N
2
3
4
5
6
7
8
9
10
I or II
32.5 or 42.5
or 52.5
N or R
A or B
S or L
SRC or
White
Manufacturer
52.5
19
100
42.5
19
100
60
67
42.5
19
100
1
1
42.5
42.5
19
23
100
100
0
0
42.5
19
100
50
32.5
23
100
60
67
60
60
0
0
32.5
23
100
67
2
2
32.5
32.5
19
19
60
60
24
12
50
50
0
0
Table 3. Key input variables and output for all ANN models.
Input Variables
Model
I
or II
32.5 or
42.5 or
52.5
N or
R
A or
B
S or
L
SRC or
White
Manufacturer
Days
Initial Time
Final Time
Compn. Str.
Compn. Str.
Compn. Str.
Output
Flex. Ten.
Str.
Flex. Ten.
Str.
ENG
H. HODHOD
ET AL.
569
The parameters of the various network models developed in the current study for the different simulation
models are presented in Table 5. These parameters can
be described with their tasks as follows:
Learning Rate (LR): determines the magnitude of the
correction term applied to adjust each neurons weights
during training process = 1 in the current study.
Momentum (M): determines the life time of a correction term as the training process takes place = 0.9 in
the current study.
Training Tolerance (TRT): defines the percentage
error allowed in comparing the neural network output to
the target value to be scored as Right during the training process = 0.001 in the current study.
Testing Tolerance (TST): it is similar to Training
Tolerance, but it is applied to the neural network outputs
and the target values only for the test data = 0.003 in the
current study.
Input Noise (IN): provides a slight random variation
to each input value for every training epoch = 0 in the
current study.
Ten.: Tensile
Several neural network architectures are designed and
tested for all simulation models investigated in this study
to finally determine the best network models to simulate,
very accurately, the effect of specimen shape and type of
sand on the strength behavior of tested mortar with different cement types based on minimizing the Root Mean
Square Error (RMS-Error). Figure 16 shows a schematic
diagram for a generic neural network. The training procedure for the developed ANN models, in the current
study, uses the experimental data presented in the previous sections of the current study. After the ANN models
are settled for all cases, prediction procedure takes place
to predict the compression as well as tensile strengths at
different age-days rather than those days measured in the
experiment (internal and after 28 days).
Table 4 shows the final neural network models for the
seven simulation models and their associate number of
neurons. The input and output layers represent the key
input and output variables described previously for each
simulation model.
Input # 1
Output # 1
Input # 2
Output # 2
Hidden layer
3 neurons
Hidden layer
3 neurons
No. of Layers
Initial Sitting
Time
Final Sitting
Time
Cube Str.
Prism Str.
(CEN sand)
Prism Str.
(Regenerated)
Prism Flex. Str.
(CEN)
Prism Flex. Str.
(Regenerated)
Input Layer
Output Layer
ENG
H. HODHOD ET AL.
570
Training
Epochs
MPRE
RMS-Error
Initial
Sitting
Time
Final
Sitting
Time
Cube
Str.
Compn.
Str. (CEN)
Flex. Str.
(CEN)
1146
4985
672361
301098
179853
315475
505672
0.067
0.0005
0.034
0.0005
1.175
0.0008
0.174
0.0004
0.281
0.0003
1.512
0.0016
0.321
0.0002
9. Conclusions
Based on the experimental investigation conducted in the
course of the current research, the following can be concluded:
1) There is an inverse proportion between setting time
and cement grade, and a direct proportion between grade
and water requirement for standard consistency.
2) Applying old cement standards, for testing and
evaluating mortar compressive strength of cement mortar, results in rejection of new cement types. Using of
C E MI52.5N
25.0
Compressive Strength- Mpa
Simulation
Parameter
20.0
15.0
10.0
Experiment
ANN Training
5.0
ANN Prediction
0.0
0
10
20
30
40
50
Age - days
ENG
H. HODHOD
ET AL.
571
No. 2
No. 3
No. 4
No. 5
No. 6
No. 7
No. 8
No. 9
No. 10
70.0
85.0
75.0
100.0
90.0
80.0
110.0
80.0
120.0
100.0
ANN
69.9
84.9
75.0
100.0
89.9
79.9
109.9
80.0
119.9
100.0
Exp.
140.0
140.0
180.0
210.0
200.0
170.0
225.0
185.0
240.0
200.0
ANN
140.0
140.0
180.0
210.0
200.0
170.0
225.0
185.0
240.0
200.0
C E MI52.5N
25
20
15
10
Experiment
ANN Training
C E MI52.5N
30.0
Compressive Strength- Mpa
30
Compressive Strength- Mpa
No. 1
Exp.
ANN Prediction
25.0
20.0
15.0
10.0
Experiment
ANN Training
5.0
ANN Prediction
0.0
0
10
20
30
40
50
10
Age - days
20
30
40
50
Age - days
Figure 18. Prism compressive strength (left: cen sand, right: regenerated sand).
C E MI52.5N
12
8
Experiment
ANN Training
C E MI52.5N
16
Flexural Strength - MP
Flexural Strength - MP
16
ANN Prediction
12
8
Experiment
ANN Training
ANN Prediction
0
0
10
20
30
40
Age - days
50
10
20
30
40
50
Age - days
Figure 19. Prism flexural strength (left: cen sand, right: regenerated sand).
H. HODHOD ET AL.
572
201-206.
10. Acknowledgements
The Authors would like to express their gratitude towards Prof. Dr. Farouk El-Hakim of 15th May institute
for Civil and Arch. Engineering, and undergraduate students (4th yearcivil) for the help they provided during
the experimental part of this research.
11. References
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
K. A. Kheireldin, Neural Network Application for Modeling Hydraulic Characteristics of Severe Contraction,
Proceedings of the 3rd Internetional Conference, Hydroinformatics, Copenhagen, 24-26 August 1998, pp. 41-48.
ENG
Abstract
In the present work, the expansion coefficient due to hydrogen incorporation was measured for the axial direction of a Zircaloy-4 cooling channel, similar to that installed in the Atucha I PHWR, Argentina, trying to
simulate the nuclear power reactor operating conditions. As a first step, the solubility curve of hydrogen in
Zircloy-4 was determined by two techniques: differential scanning calorimetry and differential dilatometry.
The comparison with classical literature curves showed a good agreement with them, although the calorimetric technique proved to be more accurate for these determinations. Dilatometry was able to detect the end of
hydride dissolution from concentrations around 60 wppm-H up to 650 wppm-H, where the eutectoid reaction:
takes place (at 550oC). We assume that this ability is a good indicator of the aptitude of the
technique to measure dimensional changes in the given hydrogen concentration range. Then, the expansion
of Zircaloy-4 homogeneously hydrided samples was measured at 300oC, the typical operating temperature of
a nuclear power reactor, obtaining a relative expansion of 2.21 10-4% per wppm-H. Considering the relative expansion observed for Zircaloy-4 at room temperature due to hydriding, starting from a hydrogen free
sample, the total relative expansion rate is calculated to be 5.21 10-4% per wppm-H.
Keywords: Thermal Analysis, Dimensional Change, Hydrides, Zircaloy-4
1. Introduction
Most of the core structural components of the nuclear
power reactors are made of Zicaloy-4, a reference zirconium alloy in many structural nuclear applications. During reactor operation, the initial dimensions of the Zrbase components could increase due to three different
degradation processes: hydrogen pick up, irradiation growth and creep.
The hydrogen incorporated into the matrix is a fraction
of the total amount of hydrogen produced during the
corrosion reaction between the zirconium and the coolant,
according to the reaction:
Zr + 2H2O ZrO2 + 4H
The crystalline defects produced by the fast neutron
irradiation induce changes in the initial dimensions of the
components depending on the fabrication texture. On the
other hand, the creep contribution to these processes depends on the magnitude of the external stress applied to
Copyright 2010 SciRes.
the component.
The pick up of hydrogen atoms by the metal induces
an expansion of its initial length. This expansion continues after crossing the solubility limit at the reactor operating temperature, since the hydrogen in excess to that
limit precipitates as ZrH1,5+x after some supersaturation in
solid solution. Due to the higher specific volume of the
zirconium hydride with respect to the zirconium matrix,
the onset of precipitation induces an additional dimensional change. This change in length depends on both,
the orientation at which the hydrides precipitated in the
matrix and the crystalline texture of the component.
The material under study in the present work is Zircaloy-4 taken from cooling channels similar to those installed in the Atucha I PHWR. These tubes have a fully
recrystallized microstructure, which induces hydride
precipitation at the grain boundaries. In addition, these
components show a strong texture in a quasi-radial direction: the c axis of the -Zr hexagonal cell is oriented in a
cone surrounding the radial direction of the tube [1]. The
ENG
P. VIZCANO
574
2. Experimental Procedure
2.1. Material and Sample Preparation
The Zircaloy-4 samples were taken from an off cut of a
cooling channel similar to those installed in the Atucha I
reactor. Rectangular samples of dimensions 10 5 1.7
mm were cut from the tube, with the length of the sample
parallel to the axial direction of the tube, as it is shown in
Figure 1.
The tubes are cold-shaped and welded (by the tungsten
inert gas method) from fully recrystallized Zircaloy-4
sheets. The typical Kearns texture factors were measured
in a previous work for the [0002] pole (c axis of the
hexagonal cell). The range of values was: Faxial = 0.050.07, Ftangential = 0.22-0.26, Fradial = 0.67-0.73. Thus, about
6% of the c poles are aligned in the axial direction, 24%
in the tangential and 70% in the radial direction [1]. The
microstructure was fully recrystallized with a grain size
of 15-20 m. It can be observed in Figure 2.
10 mm
5 mm
Figure 1. Orientation and dimensions of the dilatometric
samples.
ET AL.
2.2. Hydriding
The hydrogen was incorporated by the cathodic charge
technique. The process was carried out in an electrolytic
cell at 80 2C. A diluted aqueous solution of sulfuric
acid was used as electrolyte, circulating a current density
of 5 mA/cm2 through the sample from periods of 18 to
96 h. As a result, hydride layers of different thickness
(from a few microns up to 50 microns) formed in the
samples.
The hydrogen was diffused into the bulk during the
dilatometric experiments. After the experiments, the
samples were polished to eliminate the oxide and any
remaining hydride layer on the surfaces. Finally, the hydrogen content was measured using a LECO RH-404
hydrogen meter. The error of these determinations is of
2%.
The hydrogen range of the samples hydrided in this
way varied from 50 to 650 wppm-H.
20 m
P. VIZCANO
0.20
Extrapolation lines
0.05
DSC curve
Derived curve
-4.0
0.10
-3.0
0.15
-2.0
d q/dt (mJ/sec )
msT
-1.0
dq/dt (mJ/sec)
0.25
0.00
pT
-5.0
25 C
100
200
300
400
o
500
-0.05
Temperature ( C)
3. Results
3.1. Diffusion in the Bulk and TTSSd
Determinations
A typical diffusive dilatometric run is shown in Figure 4.
The differential apparatus needs a hydrogen-free reference sample (in fact it contains 20 wppm-H). Since the
reference is identical to the sample before hydrogen
charging, the expansion of the reference compensates
and cancels the thermal expansion of the -Zr phase in
the hydrided sample. Thus, the expansion measured with
the differential dilatometer only depends on the hydrogen
concentration of the hydrided sample. During the heating
stage, the hydride layer at the sample surface dissolves
and the hydrogen atoms diffuse into the bulk, raising the
concentration in solid solution. This process increases
continuously the dimensions of the sample as it is obser-
ved in Figure 4. Given enough time at the plateau temperature (550C), the hydride layer ends its dissolution
and hydrogen distributes homogeneously into the sample.
Depending on the thickness of the layer, it will dissolve
during the run or during the time at the plateau temperature. Yet, it is possible that a fraction remains undissolved. This will occur if at the plateau temperature the
solubility limit is reached without a complete dissolution
of the hydride layer. During the cooling stage, the sample
reduces its length but the differential expansion does not
return to zero: at room temperature, a difference in length between the sample and the reference still subsists
since the hydrogen diffused into the bulk is now precipitated as hydrides.
From the description given above, it is inferred that at
the first dilatometric run the hydrogen distribution is
controlled by the diffusion process. During this transient,
TTSSd cannot be determined. Thus, after an additional
mechanical polishing to eliminate any possible remaining hydride layer at the surface, TTSSd was measured in
the second run.
Figure 5 shows a dilatometric curve obtained after the
diffusive cycle, in the second run. During heating, the
sample increases its length again but when the dissolution finishes, the slope of the curve changes; at this point
TTSSd is determined. In Figure 5 this change in the
slope or knee is observed, for a sample containing 255
wppm-H, at 403C. This point is identified as the knee
temperature, keT. Another possible criterion, which is
not used in the present work, is to determine TTSSd at
the dilatometric derived curve: the change in the slope at
the knee produces a discontinuity, a step in the derived
curve, as it is shown in Figure 5 too. It is not an ideal
step; the discontinuity extends in a temperature range
of 40C to 50C. At the middle height of the step, the
step temperature, sT, can be determined. The step criterion proves to be more accurate than the knee criterion.
3.0
600
Expansion
14
Temperature
300
6
200
4
Lengh increase due to the
hydrogen incorporation
100
0
5000
10000
15000
Time (sec)
20000
2
0
25000
0.0020
403 C
keT
0.0015
2.0
L(m)
10
400
Expansion curve
Derived curve
2.5
12
L (m)
Temperature(o(
C))
500
575
dL/dt (m/sec)
cT
0.0
ET AL.
1.5
0.0010
401 C
1.0
sT
0.0005
0.5
o
0.0
50 C
200
300
400
o
500
0.0000
Temperature ( C)
P. VIZCANO
576
for other zirconium alloys [3], but no difference was observed between them for Zircaloy-4. Thus, TTSSd was
measured at the knee point (keT). In any case, it was observed that within an uncertainty interval of 2-4C, both
temperatures are virtually identical, Figure 5.
The dilatometric TTSSd data are plotted in Figure 6
as TTSSd vs the hydrogen concentration, [H]. The solubility equation obtained from these data is:
CkeT = 2.86 105 exp (-4730/keT)
(1)
(2)
(3)
Temperature(o(C
) )
450
400
350
300
250
200
150
100
Dissolution data
0
100
200
300
[H] (wppm)
400
500
600
Temperature ((o
C))
500
400
300
200
100
pT data
cT data
0
100
200
300
400
500
600
[H] (wppm)
Figure 7. Dissolution calorimetric data measured at pT and
cT and fitting curves.
ET AL.
4. Discussion
4.1. Terminal Solid Solubility
The uncertainty of TTSSd determinations with the dilatometric technique can be estimated from Figure 5.
Where the expansion curve changes its slope (end of
dissolution), its derivative shows a step. This step extends over a temperature interval of about 40 to 50C, an
interval larger than the 25C between pT and cT in the
calorimetric curve (Figure 3). The knee temperature
virtually agrees with the temperature at the middle height
of the step in Figure 5. However, there is a higher intrinsic uncertainty in the dilatometric measurements with
respect to the calorimetric ones which affects the accurateness of TTSSd determinations. This uncertainty increases in the low hydrogen range, where the signal of
hydride dissolution is weak. It becomes evident in Figure 6, from 60 to 130 wppm-H, where the scatter of the
data is large. For these data, TTSSd error varies from
18C to 15C at the upper extreme of the interval. For
higher concentrations (in our case, concentrations higher
than 187 wppm-H) the error decreases to 10C, becoming constant for concentrations higher than 250 wppm-H,
where an error of 8C can be assumed.
Concerning DSC determinations, as it can be inferred
from the criteria commented in 2.4, there are some discrepancies regarding the exact point where TTSSd
should be located in the DSC curve [3-7]. As a brief
summary we can say that: Z. L. Pan, measuring Youngs
modulus as functions of temperature and hold time during quasistatic thermal cycles to Zr-2.5Nb hydrided samples, concluded that the most reliable point to associate
TTSS is msT [5]. D. Khatamian found the best correlation for pT contrasting TTSSd determinations at pT, msT
and cT for unalloyed zirconium and Zr-20wt%Nb hydrided/deuterided samples with neutron diffraction
measurements [6]. Recently, the authors of the present
work determined TTSSd for Zr-2.5Nb with pressure tube
microstructure by DSC using DD as a contrasting technique. In this work, the difficulty of determining the best
point to measure TTSSd on the dissolution curves has
been discussed thoroughly. Yet, since the accurateness of
the DSC data is higher than the DD, it was not possible
to obtain conclusions about the best point for TTSSd
determination on the DSC curve from this comparison
[8].
In any case, it is evident that the selection of one of the
three criteria based on the measurements made with a
contrasting technique does not provide physical meaning
to the choice, turning it into the true dissolution point.
In fact, the certainty of this choice will be strongly dependant on the capability of the contrasting technique to
detect the disappearance of very small hydrides at the
final stage of the dissolution process. In the present cir-
ENG
P. VIZCANO
577
500
DSC (cT)
400
Temperature ( C)
ET AL.
DSC (pT)
DD (knee)
300
200
Kearns
Zuzek
100
0
50
100
150
200
250
300
[H] (wppm)
350
400
L
1
) room 3.2 10-4 3.00 10-6 [H ]
L0
wppm
(2)
where (L/Lo)room is the relative increase after the hydrogen diffusion into the bulk at room temperature.
As the cooling channels operate in the two-phase field,
in order to determine the total expansion in service, the
contribution of both, the fraction of H atoms in solid solution and that of the zirconium hydrides at the reactor
operating temperature (300C) should be added to the
growth due to the hydrides already present at room temperature. The measurements made at 300C are listed in
Table 2.
The relation found between the expansions and the
hydrogen concentration is linear too. Both, the data and
Table 1. Interception, slope, standard errors (SE) and lower and upper confidence limit (LCL and UCL). The standard deviation (SD) and R-value (R) are also given (97% of confidence).
Intercept
Slope
Statistics
Value
SE
LCL
UCL
Value
SE
LCL
UCL
SD
5.6
2.0
0.6
10.7
0.054
0.004
0.044
0.064
0.9
ENG
578
P. VIZCANO
Table 2. Relative expansion at 300C.
lo (m)
l/lo
0.85
9781
8.69 E-5
5.0
10332
4.81E-4
398
5.7
10035
5.64E-4
358
5.65
10035
5.63E-4
446
9.5
9784
9.75E-4
650
12.9
9968
1.29E-3
[H] (wppm)
l (m)
128
227
ET AL.
the regression line are shown in Figure 9 and the statistics parameters in Table 3. The linear equation is:
L
1
)300 C -1.54 10-4 2.2110-6 [H ]
(3)
L0
wppm
where (L/Lo)300C is the relative length increase of the
hydrided sample at 300C. Then, combining (2) and (3),
the total length increase is calculated as follows:
L
L
L
)TOTAL
) room
)300 C
(4)
L0
L0
L0
Table 3. Interception, slope, standard errors (SE) and lower and upper confidence limit (LCL and UCL). The standard deviation (SD) and R-value (R) are also given (97% of confidence).
Intercept ( 10-4)
Slope ( 10-6)
Statistics
Value
SE
LCL
UCL
Value
SE
LCL
UCL
SD
1.54
1.3
5.5
2.5
2.21
0.33
1.22
3.2
0.96
1.3 10-4
0.0030
0.0025
TOTAL
L/Lo
L/Lo
0.0020
room
L/Lo
0.0015
0.0010
0.0005
300 C
L/Lo
0.0000
100
200
300
400
[H] (wppm)
500
600
Figure 9. Relative expansion at 300C, at room temperature [1] and the sum of both effects.
5. Conclusions
The present work was focused on two main objectives:
hydrogen solubility measurements and the determination
Copyright 2010 SciRes.
The hydrogen solid solubility curve for Zircaloy-4 was determined by two techniques, differential
scanning calorimetry and differential dilatometry. The
comparison with classical literature curves showed a
good agreement with them. The solubility curves obtained with calorimetry, measuring TTSSd at the peak and
completion temperatures are:
CpT = 1.85 105 exp (4362/pT)
CcT = 1.78 105 exp (4546/cT)
and the one obtained by dilatometry, measuring TTSSd
at the knee temperature is:
CkeT = 2.86 105 exp (4730/keT)
Although the coincidence between them is good, the
calorimetric technique is more accurate for these measurements.
6. References
[1]
P. VIZCANO
[2]
[3]
[4]
[5]
[6]
ET AL.
579
[8]
P. Vizcano, A. D. Banchik and J. P. Abriata, Calorimetric Determination of the -Hydride Dissolution Enthalpy
in Zircaloy-4, Metallurgical and Materials Transactions
A, Vol. 35A, No. 8, 2004, pp. 2343-2349.
[9]
J. Kearns, Terminal Solubility and Partitioning of Hydrogen in the Alpha Phase of Zirconium, Journal of Nuclear Materials, Vol. 22, No. 3, 1967, pp. 292-303.
[10] E. Zuzek, J. P. Abriata and A. San Martn, H-Zr (Hydrogen-Zirconium), Bulletin of Alloy Phase Diagrams,
Vol. 11, No. 4, 1990, pp. 385-395.
[11] P. Vizcano, A. D. Banchik and J. P. Abriata, Solubility
of Hydrogen in Zircaloy-4: Irradiation Induced Increase
and Thermal Recovery, Journal of Nuclear Materials,
Vol. 304, No. 2-3, 2002, pp. 96-106.
ENG
Abstract
A Zinc-Chitosan composite coating was generated on mild steel from zinc sulphate-sodium chloride electrolyte by electrodeposition. The electrolyte constituents were optimized for good composite coating. The corrosion resistance behavior of the composite was examined by weight loss, polarization and impedance
methods using 3.5 wt% NaCl neutral solution as medium. Separate polarization profiles were recorded for
composite coating and compared with zinc coated sample. SEM images of coatings were recorded for the
pure and composite coating.
Keywords: Composite Coating, Chitosan, SEM, Impedance, Electrodeposition
1. Introduction
Zinc electroplating is an industrial process and is widely
used to coat on steel for enhancing its service life. As
zinc is electrochemically more active than steel and
hence it sacrificially protect the steel from corrosion.
However zinc itself undergoes corrosion leading to the
formation of zinc compounds called white rust on its
surface. This tendency of formation of white rust reduces
the life of the coating from the expected period. Therefore to enhance the life span of the zinc coating and to
avoid the white rust formation the alternative methods
like surface modification is adopted. The earlier modification methods are associated with chromate based formulations and the procedure is very simple to generate
passive chromate films on corroding zinc coatings. The
use of chromate passivation is prohibited because of
pollution hazards. An alternate to this chromation is to
generate surface films or surface barriers with specific
organic molecules or with certain addition agents [1-6].
Also the service life of zinc coating is enhanced by including the inert materials in its coating. The inclusion is
done by codeposition of these materials with zinc and
thus generating composite coating. These zinc composite
coatings exhibit better corrosion resistance property.
Nowadays the nanosized materials are codeposited to get
better zinc composite with better corrosion resistance
[7-10].
A survey of literature reveals that the conducting
Copyright 2010 SciRes.
2. Experimental
2.1. Plating Process
Zinc and Zn-chitosan coatings were electrically deposited from sulphate-chloride bath. The constituents of the
bath were 250 g/L ZnSO47H2O, 40 g/L NaCl, 30 g/L
H3BO3 and 0 g/L chitosan (88% deacetylated). In all the
ENG
K. VATHSALA
ET AL.
581
zinc coating
composite coating
8
7
6
5
4
3
2
0
50
100
150
200
250
300
350
400
Time (h)
ENG
K. VATHSALA
582
ET AL.
zinc coating
com posite coating
16
14
12
-Z''(ohm)
10
8
6
4
2
0
-2
5
10
15
20
25
30
35
40
45
Z '(o h m )
Figure 4. SEM images for the two samples. (a) Zinc coating,
(b) composite coating.
4. Discussion.
The experimental results of the present investigations
inferred that the chitosan can be included in the deposit
easily. It acquires a positive charge by protonation in
Current density(cm-2)
500
Figure 5. SEM images for two samples after anodic polarization for (a) zinc coated; (b) composite coated sample.
zinc coating
composite coating
400
300
1m
200
X 4000 1m
X 4000
100
-1.03
-1.02
-1.01
-1.00
-0.99
-0.98
Potential (V)
chit NH 2 H 3O chit NH 3 H 2 O
K. VATHSALA
ET AL.
583
5. Conclusions
Zn-chitosan composite was generated by electrodeposition from sulphate bath. The precipitated chitosan was
codeposited along with zinc. The performance of composite coating was established from the results of weight
loss, polarization, impedance and salt spray test. In all
these studies Zn-chitosan composite exhibits better anti
corrosion performance. The SEM images of surface provide an evidence for the presence of chitosan in coating
and crystalline nature. The composite showed uniform
and lower corrosion rates than that of zinc coating.
6. Acknowledgements
The authors are grateful to University Grant Commission,
New Delhi, Govt. of India [Major Research Project F.32220/2006(SR)] for providing financial assistance.
7. References
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
ENG
584
K. VATHSALA
ET AL.
ENG
Department of Electronics Engineering, National Taiwan University of Science and Technology, Taipei, Taiwan, China
2
Department of Optics and Photonics, National Central University, Jhungli, Taiwan, China
3
Department of Electro-Optical Engineering, National United University, Miaoli, Taiwan, China
4
Department of Mechanical Engineering, National Taiwan University, Taipei, Taiwan, China
E-mail: skliaw@mail.ntust.edu.tw
Received February 2, 2010; revised March 23, 2010; accepted March 27, 2010
Abstract
Several high-performance and tunable erbium-doped fiber lasers are reviewed. They are constructed by using
fiber Bragg gratings (FBGs) or short-wavelength-pass filters (SWPFs) as wavelength tunable components
inside the laser cavity. Broadband wavelength tuning range including C- and/or S-band was achieved, and
tunable laser output with high slope efficiency, high side-mode suppression ratio was obtained. These fiber
lasers can find vast applications in lightwave transmission, optical test instrument, fiber-optic gyros, spectroscopy, material processing, biophotonic imaging, and fiber sensor technologies.
Keywords: Fiber Bragg Grating (FBG), Short-Wavelength-Pass Filter (SWPF), Tunable Fiber Laser, Optical
Communication
1. Introduction
In recent years, fiber lasers have found a variety of
applications in the testing of fiber components, fiber
sensing and wavelength division multipling (WDM)
systems, in which they are used to act as a backup
source with ITU-T grids [1]. Also, fiber lasers are useful for spectroscopy, sensing protection, and fiber-optic
gyro [2]. Partially because of their features, such as
low wavelength sensitivity to temperature, low-intensity noise, and all-fiber construction, their advantages
over non-fiber-based laser sources are potentially lowintensity noise, high output power, and compatibility
with fiber components. Previous works have proposed
design and/or characteristics valuation of fiber lasers, including multiple-ring cavity fiber laser [3], two separate
erbium-doped fiber lasers [4], distributed feedback fiber
lasers [5], and Brillouin erbium-doped fiber laser pumped using fiber Bragg grating (FBG) [6]. These fiber lasers, however, have fixed wavelengths that are not suitable for wavelength routing, reconfigurable switching
and/or network protection. On the contrary, tunable erbium-doped fiber lasers could well fit such requirements.
Nowadays, a variety of tunable fiber lasers have been
demonstrated such as tunable single-frequency fiber lasers [7], coherent combining tunable lasers [8], tunable
Copyright 2010 SciRes.
ENG
586
Wavelength
1553
1552
1551
1550
1549
-1.0
-0.5
0.0
0.5
1.0
Turns of Screw
(a)
Second grating
First grating
-50
1552.68nm
1540.5nm
-60
Power (dBm)
dn
( )
(1)
n2
d
T
2n {1 ( )[ P12 ( p11 p12 )]} [
]T
2
n
1554
-70
-80
1540.5nm
-85.3dBm
1535.6nm
-85.23dBm
-90
1530
1547.7nm
-85.21dBm
1545.47nm
-85.73dBm
1540
1560.66nm
-83.93dBm
1557.6nm
1552.68nm -84.06dBm
-86.12dBm
1550
1560
Wavelength (nm)
(b)
Figure 1. (a) Wavelength tunable FBG versus turns of micrometer screw; (b) superposed transmission spectra of two
tunable FBGs (Total tuning range may cover the C band).
ENG
587
(a)
(b)
588
60
50
40
30
20
10
0
0
20
40
60
80
100
120
140
160
(a)
(b)
ENG
589
30
25
20
15
10
5
Figure 5. Configuration of proposed BFM-based liner-cavity tunable fiber laser in backward pump scheme with use of
residual pump power [11] (PM: power meter).
EDF length = 1m
EDF length = 3m
EDF length = 4m
0
0
25
50
75
100
125
150
(a)
40
Experimental Curve
@ 1544.3 nm
35
30
25
20
15
10
5
0
0
10
(a)
(b)
40
EDF length = 1m
EDF length = 3m
EDF length = 4m
35
30
25
20
15
10
5
0
0
20
40
60
80
100
120
140
(b)
6(b). The transfer efficiency versus pump power for different lengths of EDF is shown in Figure 7(a). We find
that the transfer efficiency is increased as the EDF length
increases in the beginning. Then it reaches a constant
value of 21.5% as the pump power is larger than 70 mW.
Copyright 2010 SciRes.
590
(a)
(b)
ENG
(a)
591
5. Discussion
5.1. Advantages of FBG-Based Tunable Fiber
Lasers
(b)
592
6. Conclusions
Two kinds of tunable fiber-filter-based EDF fiber lasers
have been reviewed. Both of them have broadband wavelength tuning range including C- and/or S-band. Using
FBG in strain mechanism, we have proposed and demonstrated a tunable FBG-based fiber laser that employs
one OC, two homemade TFBGs. The configuration consists of a linear cavity to achieve a wavelength tuning
range of 31.5 nm with 0.05 nm linewidth and over 60 dB
SNR. The power variation over the entire tuning range is
less than 0.1 dB with power equalization by using
low-cost VOAs. Another way is to employ a BFM and
tunable FBG at either cavity end of fiber cavity. The
BFM acts as a broadband rear-end reflector both for
lasing signal and pump source. For wavelength tunable
demonstration, power variation over the whole C-band is
less than 1.0 dB without the usage of power equalization. The time to reach stable laser operation is less than
11 ms after switching between the two FBGs, and the
continuous tuning resolution is less than 0.2 nm in the
whole range. For the SWPF-based tunable fiber laser
using temperature tuning mechanism, two tunable
SPWFs based erbium-doped fiber lasers were reviewed.
The side-polishing and fused-tapering techniques were
used to achieve thermo-optic tunable short-wavelengthpass function based on material dispersion discrepancy
and variations of waveguide structures. The tuning efficiency is 50 nm/C, cut-off efficiency is 1.2 dB/nm, and
rejection efficiency is 55 dB, individually. The widely
tunable SWPFs were applied to achieve broadband and
high-tuning-efficiency S- and/or C-band EDF ring lasers,
which can be tuned close to the short-wavelength edge of
gain bandwidth, and the tuning range is 26 nm with the
signal-to-ASE-ratio of around 40 dB, and the FWHM
linewidth is about 0.5 nm. All of them have graceful
features of simple structure, compactness, ease of connection to fiber components, high-efficiency, and continuous tunability. They are promising for vast applications in lightwave transmission, optical test instrument,
fiber-optic gyros, spectroscopy, material processing, fiber sensing, WDM backup light sources, as well as in biophotonics.
ENG
7. Acknowledgements
The authors were partially supported by the National
Science Council (NSC) (Project Nos. NSC 982221E011-017, NSC 97-2923E-011-001-MY3, NSC 982218
E008-004, NSC 98-2221-E-239-001-MY2). We thank
Jang W. Y., Wang C. J., Hung K. L., Jhong G. S., Chi S.,
Tseng S. M., Huang C. M., Lai Y. for discussion, T.
Wang and Z. G. Shieh for kind help.
8. References
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
593
Y. C. Zhao, S. Winnall and S. Fleming, Tunable FiberRing Laser Based on Broad-Band Fiber Bragg Grating
and Bending Effects, Microwave and Optical Technology Letters, Vol. 46, No. 6, 2005, pp. 562-563.
ENG
Abstract
The present paper investigates fire resistance of a simply-supported composite concrete-trapezoidal steel
sheet slab. The objective is to find out if a steel sheet, as a moisture diffusion barrier, may substantially effect
the hydro-thermal situation in the concrete part of the cross-section. The numerical integration of the equations of a coupled hygro-thermal boundary-value problem (Tenchev, R.T., Li, L.Y. and Purkiss, J.A. (2001)
Num. Heat Transfer Part A, 39(7), 685-710), with and without considering the barrier, shows that the barrier
does not really effect the magnitude and the development of temperatures over the cross-section, while there
is a significant effect on the pattern of moisture transport and the magnitude of vapour pressure. Particularly
high magnitudes of vapour pressure (about 4.5 MPa) were shown in cases where the steel sheet was considered in analyses, which indicates a possible micro damage of concrete in the web of the section, although
spalling probably cannot take place due to the steel sheet cover. As the typical composite slab investigated
here is not sufficiently fire resistant without any additional reinforcement bars placed in the web, further investigations are directed to finding an optimal position and area of these bars. Following a simplified procedure given in EC2 (Eurocode 2, Design of Concrete Structures, Part 1.2 (2004) Structural fire design, European Committee for Standardization) and assuming that the present composite slab is subject to the uniform
traction q = 1.52 kN/m, yields that placing one bar with the area 1.153 cm2 4 cm away from the edge suffices
for the 60 min fire resistance of the slab.
Keywords: Fire Analysis, Composite Structures, Heat and Mass Transfer
1. Introduction
Composite concrete-trapezoidal steel plate slabs are widely used structural elements in buildings and bridges.
During the placement of concrete the trapezoidal steel
plate replaces panelling, while upon hardening of concrete the two materials work as a composite slab, the
steel plate representing the reinforcement. For the better
vertical load redistribution, concrete is additionally reinforced with a steel mesh at the upper part of the crosssection (here called the flange). The reinforcement in the
web of the concrete part of the section needs rarely to be
applied for non-accidental actions. By contrast, when the
composite slab is exposed to fire, the steel plate is directly exposed to high temperatures resulting in a substantial decrease of its bearing capacity. The reinforcement in the web and its position within the concrete slab
then become essential. In fact, both the position and the
area of the additional reinforcement turn out to be essenCopyright 2010 SciRes.
ENG
T. HOZJAN ET AL.
J L E L
(1)
(2)
JV E L
Air conservation:
G A
t
(3)
J A
Energy conservation:
C Tt
k T Cv T E E L D
D
t
(4)
595
(5)
J G G V v G G G DVA V
G
(6)
(7)
JL L vL
KKG
KK L
PG
(8)
PL
(9)
T. HOZJAN
596
ET AL.
cem
3
V , 0.0089 kg/m , initial porosity of concrete por 0.15,
0
16
Figure 1. The composite plate and 2D finite element mesh over the concrete cross-section.
Table 1. Boundary conditions of composite slab.
case
T
A1
PG
V
A2
edge 1
edge 2
qT qT (TISO 834 )
qT qT (T 20 C)
PG
n
V
symmetry
o
PG
PG 0.1 MPa
n
0
qV qV ( V , )
qT qT (TISO 834 )
qT qT (T 20 C)
PG
PG 0.1 MPa
PG 0.1 MPa
qV qV ( V , )
qV qV ( V , )
n
T
n
PG
ENG
T. HOZJAN ET AL.
597
t = 30 min
t = 60 min
0.14
0.14
0.14
0.12
0.12
0.12
0.1
0.1
0.1
0.08
0.08
0.08
0.06
0.06
0.06
0.04
0.04
0.04
0.02
0.02
0.02
0
0 0.02 0.04 0.06 0.08 0.1 0.12
0
0 0.02 0.04 0.06 0.08 0.1 0.12
1000
900
800
700
600
500
400
300
200
ENG
T. HOZJAN
598
yre [cm]
zre [cm]
2.5
2.5
2.5
2.5
2.5
ET AL.
t = 30 min
t = 60 min
0.14
0.14
0.14
0.12
0.12
0.12
0.1
0.1
0.1
0.08
0.08
0.08
0.06
0.06
0.06
0.04
0.04
0.04
0.02
0.02
0.02
0
0 0.02 0.04 0.06 0.08 0.1 0.12
0
0 0.02 0.04 0.06 0.08 0.1 0.12
0
0 0.02 0.04 0.06 0.08 0.1 0.12
t = 30 min
t = 60 min
0.14
0.14
0.14
0.12
0.12
0.12
0.1
0.1
0.1
0.08
0.08
0.08
0.06
0.06
0.06
0.04
0.04
0.04
0.02
0.02
0.02
0
0 0.02 0.04 0.06 0.08 0.1 0.12
0
0 0.02 0.04 0.06 0.08 0.1 0.12
4.5
4
3.5
3
2.5
2
1.5
1
0.5
0
0 0.02 0.04 0.06 0.08 0.1 0.12
4.5
4
3.5
3
2.5
2
1.5
1
0.5
Figure 4. Distribution of pore pressures over the cross-section at t = 15, 30, and 60 min. (a) case A1; (b) case A2.
ENG
T. HOZJAN ET AL.
599
t = 60 min
0.14
0.14
0.14
30
0.12
0.12
0.12
25
0.1
0.1
0.1
20
0.08
0.08
0.08
0.06
0.06
0.06
0.04
0.04
0.04
0.02
0.02
0.02
0
0 0.02 0.04 0.06 0.08 0.1 0.12
0
0 0.02 0.04 0.06 0.08 0.1 0.12
15
10
5
0
0 0.02 0.04 0.06 0.08 0.1 0.12
t = 30 min
t = 60 min
0.14
0.14
0.14
30
0.12
0.12
0.12
25
0.1
0.1
0.1
20
0.08
0.08
0.08
0.06
0.06
0.06
0.04
0.04
0.04
0.02
0.02
0.02
0
0 0.02 0.04 0.06 0.08 0.1 0.12
0
0 0.02 0.04 0.06 0.08 0.1 0.12
0
0 0.02 0.04 0.06 0.08 0.1 0.12
15
10
5
0
Figure 5. Distribution of free water content over the cross-section at t = 15, 30, and 60 min. (a) case A1; (b) case A2.
Figure 6. (a) Variation of temperature with time for different vertical positions of reinforcement bars; (b) Variation of temperature with vertical position zre at t = 30, 60, and 90 min.
As the steel sheet is directly exposed to high temperatures, it loses its bearing capacity shortly after fire begins.
The presence of additional reinforcement bars, if any, is
essential in such cases. That is why fire resistance of
composite slab in terms of the resistance time can be
rather well estimated on the basis of actual temperatures
in the additional reinforcement bars placed in the web. It
is well known that creep strains of mild steel start increasing at about 400C [11]. At roughly 500C, the
Copyright 2010 SciRes.
600
T. HOZJAN
5. References
[1]
[2]
[3]
[4]
[5]
R. T. Tenchev, L. Y. Li and J. A. Purkiss, Finite Element Analysis of Coupled Heat and Moisture Transfer in
Concrete Subjected to Fire, Numerical Heat Transfer
Part A, Vol. 39, No. 7, 2001, pp. 685-710.
[6]
Z. P. Baant and M. F. Kaplan, Concrete at High Temperatures: Material Properties and Mathematical Models, Longman, Harlow, 1996.
[7]
[8]
[9]
4. Discussion
The simply-supported composite concrete-trapezoidal
steel sheet slab is a widely used structural element in
engineering. Often its fire resistance applied from below
is of prime importance. The steel sheet acts as a moisture
diffusion barrier. In order to find out if such a barrier
may substantially affect the hydro-thermal situation in
the concrete part of the cross-section, we compared the
numerical results of the coupled hygro-thermal boundary-value problem obtained with and without considering
the barrier. The results showed that the barrier does not
really affect the magnitude and the development of temperatures over the cross-section, while the effect on the
pattern of moisture transport and the magnitude of vapour pressure is significant. Particularly high magnitudes
of vapour pressure (about 4.5 MPa) were experienced in
analyses when considering the steel sheet, which indicates a possible micro damage of concrete in the web of
the section, although explosive spalling probably cannot
Copyright 2010 SciRes.
ET AL.
ENG
T. HOZJAN ET AL.
[10] G. A. Khoury, C. E. Majorana, F. Pesavento and B. A.
Schrefler, Modelling of Heated Concrete, Magazine of
Concrete Research, Vol. 54, No. 2, 2002, pp. 77-101.
601
[11] G. Williams-Leir, Creep of Structural Steel in Fire: Analytical Expressions, Fire and Materials, Vol. 7, No. 2,
1983, pp. 73-78.
ENG
Abstract
The initial oxidation behavior of Ni-6.5Co-17.8Cr-3.7Al-0.5Y alloy is investigated at 800C-1000C. X-ray
diffraction results show that the dominant Cr2O3 phase and secondary -Al2O3 and NiO phases are observed
on the surface of samples at all initial stages (oxidized for 16 hours). YAlO3 and -Al2O3 can only be detected at low temperature (800C) while the spinel NiCr2O4 is only observed at 900C and 1000C. Though
the growth rates of -Al2O3 and Cr2O3 are comparable at 900C, the former becomes much lower than the
latter when the temperature changes to 1000C. Scanning electron microscopy (SEM) images show that the
-Al2O3 grows from some irregular ditches in the chromia scale at 900C. However, cracking and spalling
are more serious at 1000C without -Al2O3-grown-ditches, which is in accordance with the growth rates of
these oxides at different temperatures. The cracking can be explained by the results of Raman determination
which indicate that the stress on the surface of specimen oxidized at 1000C is higher than that at 900C.
Owing to this condition, a preoxidation treatment on the NiCoCrAlY alloy for 16 hours is prepared at 900C,
and then thermal cycling oxidation test is conducted at 1000C for 200 hours. The result indicates that the
initial preoxidation treatment at 900C improves the oxidation resistance of alloy at 1000C.
Keywords: NiCoCrAlY, Oxidation Kinetics, Initial Oxidation, Al2O3, Cr2O3
1. Introduction
NiCrCoAlY alloys are often used as bond coatings of
thermal barrier coatings (TBCs) to protect the substrate
from oxidation at high temperature and to provide the
necessary adhesion of the ceramic to the substrate [1-3].
Some authors [4-7] indicated that the protection offered
by MCrAlY (M=Ni, Co or a combination) alloys against
high temperature oxidation relies on the ability of the
alloy to develop and maintain a continuous, dense and
slow growing -Al2O3 scale. The formation of a continuous of alumina (Al2O3) layer during the oxidation of
the substrate at high temperatures could result in a dramatic slowing down of the oxidation process, because
Al2O3 formation has a slower rate of oxidation compared
to other oxidations. Generally, the protectiveness of the
alloy surface at long-term stage is frequently determined
by the initial stage of oxidation [8-10].
*The project was supported by the State Key Development Program for
Basic Research of China (Grant No. 2007CB707700).
Besides alumina, chromia (Cr2O3) also plays an important role during the high temperature oxidation
[11-13]. However, the effect of the interaction among the
oxides on the oxidation resistance of alloy has not been
discussed in detail. In particular, there has been a lack of
attention to the initial stages of oxidation on NiCoCrAlY
alloy to date.
Raman spectroscopy has been used as a non-destructive technique for determining the stresses in oxide
scales for decades due to the bands in the Raman spectra
of specimens shift with pressure [11-13]. The salient
features of the Raman technique are that it does not require a special environment, and it provides a high resolution.
The aim of this study is to improve the service life of
the alloys. The evolution of the oxide scale on the surface of NiCoCrAlY alloy at initial stage oxidized at
800C-1000C were investigated. After reporting the
experimental results, a method to improve the oxidation
resistance was proposed by utilizing the interaction chaENG
C. ZHU
2. Experimental
The original powder was commercially available, and its
component was 6.5%Co, 17.8%Cr, 3.7%Al, 0.5%Y,
balance Ni (wt.%) with an average particle size of 16.34
m.
The powder was heated and compacted into the form
of a cylindrical rod of green density equal to 89 5% of
its theoretical density at inner temperature of 800C for 2
hours with a pressure of 300 MPa. Subsequently, the rod
was annealed in vacuum at 1000C for 1 hour in order to
homogenize and recrystallize the alloy.
Disc shaped specimens (diameter 15 mm and thickness 1 mm) were cut from the rod using spark-machining.
The specimen surface was ground and polished. After
each preparation step the specimens were thoroughly
cleaned ultrasonically with alcohol.
Isothermal oxidation was performed in static air at 1
atm pressure in a resistance furnace which has a maximum operating temperature of 1300C. All the specimens were put into the furnace at the same time after the
test temperature was reached. Then, oxidized specimens
were removed out from the furnace after a chosen time
and air-cooled to room temperature. The initial oxidation
tests were performed at 800C, 900C and 1000C for 2,
4, 8 and 16 hours.
A thermal cycling oxidation test was conducted at
1000C for 200 hours. In order to keep consistent of
oxidation condition in the test, the specimens were preoxidized at 900C and 1000C for 16 hours. Then the
oxidation behavior of the specimens was evaluated by
measuring the weight gains of the samples for 184 hours.
The 12-hour cycle consisted of 11 hours holding at
1000C, followed by cooling in air for 1 hour. The precision of the balance was 0.1 mg.
Raman spectroscopy was used here in order to determine the stresses in chromia scales formed at 900C and
1000C on NiCoCrAlY alloy. Chromia has the same
structure as corundum, and therefore it should have
seven Raman active bands (A1g + 5Eg) [11-13]. The most
intense mode is the 549 cm-1 A1g vibration [11] and this
one was used for monitoring the stress.
The spectroscopy was measured at room temperature
using the Renishaw Ramanscope 1000 (Renishaw,
Gloucestershire, UK) in conjunction with an Olympus
BH-2 microscope. During the measurements, the laser
(HeNe, 632.8 nm) was focused at a position on the surface of the sample and the laser spot size was set about
35 m. The Raman spectroscopy acquired was analyzed
by the commercial Renishaw WiRe software to obtain
the peak shift fitted by GaussianLorentzian function.
The surface morphologies and polished cross sections
of the specimens were observed using a scanning elecCopyright 2010 SciRes.
ET AL.
603
C. ZHU ET AL.
604
(a)
(b)
(c)
(c)
(d)
Figure 3. SEM images of the specimen surface (oxidized for
16 h at 800C) (a) surface image; (b) higher magnification;
(c) and (d) EDX analysis in selected region.
ENG
C. ZHU
(e)
(f)
605
(g)
ET AL.
Figure 5. The cross-sectional microstructure and the elemental maps of NiCoCrAlY alloy (oxidized for 16 h) at (a)
800C; (b) 900C and (c) 1000C.
ENG
606
C. ZHU ET AL.
Figure 6. Raman spectrum for the chromia formed on NiCoCrAlY alloys oxidized for 16 h at 900C and 1000C.
4. Conclusions
Initial oxidation tests of Ni-6.5Co-17.8Cr-3.7Al-0.5Y
alloy specimens was performed at 800C, 900C and
1000C for 16 hours. Cr2O3 was the predominant phase
at all three temperatures and the dense Cr2O3 scale play-
ENG
C. ZHU
5. Acknowledgements
The project was supported by the State Key Development Program for Basic Research of China (Grant No.
2007CB707700).
6. References
[1]
N. P. Padture, M. Gell and E. H. Jordan, Thermal Barrier Coatings for Gas-Turbine Engine Applications, Science, Vol. 296, No. 5566, 2002, pp. 280-284.
[2]
[3]
[4]
[5]
[6]
ET AL.
607
ENG
Abstract
Highly nonlinear birefringent Photonic Crystal Fibre (PCF) that exhibits low losses and small effective mode
area across a wide wavelength range has been presented. The effects of angular orientation on bending losses
of the proposed PCFs have been thoroughly investigated by employing a full vectorial finite element method
(FEM). It has been demonstrated that it is possible to design a bending-insensitive nonlinear PCF with a birefringence in the order of 10-2 and a nonlinear coefficient of 49 W-1km-1 at the wavelength of 1.55 m. Also,
significant improvements on key propagation characteristics of the proposed PCFs have been demonstrated
by carefully altering the desired air hole diameters and the hole-to-hole spacing. It is demonstrated that two
zero dispersion wavelengths can be achieved by the proposed design.
Keywords: Nonlinear Coefficient, Effective Mode Area, Confinement Loss and Birefringence
1. Introduction
Photonic Crystal Fibers consisting of a central defect
region in a regular lattice of air holes have attracted significant research attention. These fibers provide extra
degrees of freedom in manipulating optical properties
[1-2]. PCFs can be divided into two categories according
to the mechanism used to guide the light: photonic-bandgap (PBG) guidance and effective index guidance. The
PBG fibers use a perfectly periodic structure exhibiting a
PBG effect of the crystal lattice at the operating wavelength to guide light in a low-index core region. In PBG
fibres, the core can be created from the lower refractive
index material, which could be solid glass or a large air
hole (in the case of an air-glass PCF) [3-5]. On the other
hand, the effective index-guiding PCFs rely on total internal reflection (TIR) to confine light in the region of
missing air hole forming a central core. The presence of
air holes decreases the effective index of the cladding,
making light guidance possible by TIR. This guiding
method is more analogous to the operation of a conventional step-index fibre [1,3,6].
PCFs have remarkable properties, strongly depending
on the design details such as low sensitivity to bend losses even for high mode areas, where, low or high mode
areas leading to very strong or weak optical nonlinearities. PCF technology, now allows the fabrication of fiCopyright 2010 SciRes.
H. ADEMGIL
lished results by Yue et al. [11] and Sun et al. [12] have
demonstrated that it is possible to design PCFs with relatively large birefringence in the order of 10-3-10-2. To our
knowledge proposed PCFs with elliptical air holes [1112] exhibit the highest birefringence to date. However,
fabrication becomes challenging by the use of several
rings of elliptical air holes in cladding region. Moreover,
controlling the elliptical air holes during the fabrication
process might be difficult [10,15].
The design of PCF structures with small mode areas
that lead to high nonlinear coefficient , is an ongoing
challenge. By varying the size of the air holes in the
cladding region and the hole to hole spacing, desired
effective mode areas can be obtained [7,8]. Small core
diameter that leads to low effective mode area can be
reduced by having a relatively small hole to hole spacing.
Previously published results such as Poli et al. [16] and
Saitoh et al. [17] have demonstrated theoretically, that it
is possible to design PCFs with nonlinear coefficients of
about 30 and 44 W-1km-1, respectively, at 1.55 m telecommunication wavelength. However, these structures
are purely theoretical and the hole to hole spacing, , is
around 0.9 m. From the point of view of fabrication,
small hole to hole spacing might be problematic to
manufacture.
In recent years, highly birefringent PCFs with nonlinear properties have received growing attention in telecommunication and supercontiniuum applications [7,1819]. Previously published results by Lee et al. [7] and
Yamamoto et al. [18], have experimentally demonstrated
that it is possible to design highly nonlinear PCFs with a
relatively large birefringence in the order of 10-3 at 1.55
m telecommunication wavelengths. Lee et al. [7] has
demonstrated a birefringent PCF having nonlinear coefficient , of 31 W-1km-1 for the use of optical
code-division multiple access (OCDMA) applications.
Similarly, Yamamoto et al. [18] has demonstrated highly
birefringent PCF with Ge-doped core having nonlinear
coefficient, , of 19 W-1km-1.
Moreover, recently published papers such as Kudlinski
et al. [20] and Cumberland et al. [21] have shown that
PCFs with two zero dispersion wavelengths (ZDW)
demonstrate stronger power spectral densities than single
ZDW PCFs. Therefore PCFs with two ZDW can be beneficial in supercontinuum applications. Kudlinski et al.
[20] have demonstrated that it is possible to design two
ZDW PCF with a nonlinear coefficient of 31 W-1km-1.
For many applications it is essential to design PCFs
that exhibit simultaneous high birefringence, low losses,
and high nonlinear coefficient across a wide wavelength
window. Additionally, bending losses can be a critical
issue in the sensing and communication applications [22].
Bending is one of the important issues regarding the
practical development of PCFs. When an optical fibre is
bent, the field profile deforms outwards in the direction
Copyright 2010 SciRes.
ET AL.
609
ENG
H. ADEMGIL ET AL.
610
PCF structure is not an issue. Theoretical and experimenttal investigations by Suzuki et al. [10] have shown
that it is possible to fabricate even complex PCF structures by adapting conventional stack and draw methods.
However, stack and draw methods [15] are limited to
closely-packed geometries such as triangular or honeycomb lattices and cannot easily generate different geometries. Alternatively, drilling methods allow great flexibility for both the hole size and spacing, but the structures are generally limited to a small number of holes.
Another alternative fabrication method is to use the solgel method which allows for independent adjustment of
the hole size and spacing. The sol-gel method [15] provides additional design flexibility that will be necessary for such PCF structures. Additionally, recently published results by Vu et al. [22] demonstrated experimenttally, that the fabrication of bending insensitive PCFs is
possible and that these fibers are robust against high
amounts of bending.
In this work, we have employed full vectorial finite
element method (FEM) to investigate key modal properties of the proposed index guided PCF. The modal solution approach based on FEM is more flexible and reliable
than other techniques. It can represent any arbitrary
cross-section more accurately and has been widely used
to find the modal solutions of a wide range of optical
waveguides [6,26]. The FEM formulation for modal
analysis based on anisotropic perfectly matched layers
(PML) is capable of handling as many modes as required
and analyse leaky modes. By using PMLs boundary condition, propagation characteristics and optical properties
of leaky modes in PCFs, it can be precisely evaluated
[27]. The modal analysis has been applied on the crosssection in the x-y plane of the PCF as the wave is propagating in the z direction.
In this study, birefringence, confinement loss, effective mode area, nonlinear coefficient properties of the
proposed PCFs are reported thoroughly. Also, significant
improvements of propagation characteristics of the PCFs
are demonstrated. Following this introduction, a brief
theoretical analysis is provided in Section 2. The simulation results are reported in Section 3 and, finally, conclusions are drawn in the last section.
2. Theory
(1)
dxdy
4
E dxdy
(4)
ENG
H. ADEMGIL
3. Simulation Results
In Figure 1, the proposed PCF design with an indexguiding core surrounded by a triangular array of air holes
is presented. The diameter of air holes and hole-to-hole
spacing is denoted by d and , respectively. The refractive index of pure silica is set equal to 1.45. In order to
reduce the confinement losses, five rings of air holes are
considered.
Previously published results by Ortigosa et al. [14]
have shown that, by varying the hole diameters along the
two orthogonal axes high birefringence can be achieved.
Therefore, in our design in order to achieve ultrahigh
birefringence, the air hole diameter sizes, d1, along the
x-direction are increased. To enhance the birefringence
further, the first column of the air hole group is shifted
outwards by /2. As a result, the PCF core becomes
more asymmetrical which results in a significant increase
in the birefringence.
In order to keep the birefringence at the optimum level
and reduce confinement losses, we next investigate the
size of the air holes in the cladding region. It is known
that confinement losses [2,26] can be reduced by increasing the size of the air holes in the inner cladding
area, d. However, according to our simulations this has a
negative effect on the birefringence. Hence, there is a
tradeoff between ultrahigh birefringence and low confinement losses. Alternatively, by increasing the number
Copyright 2010 SciRes.
ET AL.
611
ENG
612
H. ADEMGIL ET AL.
Figure 3. Variation effective mode area, Aeff of the fundamental mode as a function of the operating wavelength.
fundamental and first order modes reduce with increasing hole-to-hole spacing.
As expected, losses of higher order modes are much
higher than the fundamental mode. It is worth noting that,
confinement loss for y-polarized mode of both fundamental and first order mode are higher than x-polarized
mode. In this study we mainly concentrate on the behavior of the fundamental modes. Therefore, further analysis of propagation properties focuses on the fundamental
modes.
Figure 3 shows the variation of effective mode area as
a function of wavelength. It can be noted that Aeff is increasing with the increasing hole to hole spacing. The
effective mode area steadily increases when the wavelength increases. It is worth noting that the effective area
is much smaller than that of the conventional fibres at
1.55 m wavelength. This would contribute to increase
the nonlinearities produced by refractive index power dependence [1].
Having the freedom to control the optical properties of
the PCF by hole to hole spacing and placement whilst
maintaining strong confinement of the mode, allows for
the realization of high nonlinear effects. With ultra-high
nonlinearities, we can generate supercontinuum with relatively low pumping power. This is a very important
advantage. The devices can be made smaller, cheaper
and become more portable [28].
Variation of the nonlinear coefficient as a function of
wavelength is presented in Figure 4. As presented in
Equation (5) the nonlinear coefficient is inversely proportional to the effective area. Small effective mode area
leads to high nonlinear coefficient that would be useful
in the context of supercontiniuum generation and soliton
pulse transmission [7,9]. In this regard, the nonlinear
coefficient steadily increases when the wavelength and
hole to hole spacing, decreases. Our design shows that
the nonlinear coefficient, , for = 1.7 m and = 2 m
at 1.55 m operating wavelength is 26 W-1km-1 and 20
W-1km-1, respectively.
ENG
H. ADEMGIL
ET AL.
613
614
H. ADEMGIL ET AL.
of the order of 10-2 is achieved. The birefringence is sensitive to the varying wavelength, and it increases as the
wavelength increases. It can be noted that the birefringence for = 1 m and = 1.6 m at 1.55 m operating
wavelength is 2.65 10-2 and 1.01 10-2, respectively.
Finally, the chromatic dispersion profile can be easily
controlled by varying the hole diameter and the hole to
hole spacing [28]. Controllability of chromatic dispersion
in PCFs is a very important problem for practical applications to optical communication systems, dispersion
compensation, and nonlinear optics [12]. At short wavelength, the modal field remains confined to the silica
region, but at longer wavelengths the effective cladding
index decreases. Thus, as we change the size of air hole d
or the separation between them , ZDW can be altered
to any value. This unusual dispersion characteristic of
PCFs allows them to be used in non-linear fiber optics.
One can see that it is possible to shift the zero dispersion
wavelength from visible to near-infrared (IR) regions by
appropriately changing the geometrical parameters (d
and ).
At short wavelength, the modal field remains confined
to the silica region, but at longer wavelengths the effective cladding index decreases. Thus, as we change the
size of air hole d or the separation between them ,
ZDW can be altered to any value. This unusual dispersion characteristic of PCFs allows them to be used in
non-linear fiber optics. One can see that it is possible to
shift the zero dispersion wavelength from visible to
near-infrared (IR) regions by appropriately changing the
geometrical parameters (d and ). As may be seen from
Figure 9 when hole to hole spacing, = 1 m and =
1.2 m the proposed PCF has a single ZDW, 0.8 m and
0.84 m respectively. On the other hand, when = 1 m
and = 1.2 m two ZDW is achieved. The first ZDW
for both cases is around 0.8 m. However, according to
simulation results the second ZDW for = 1 m and
4. Conclusions
In summary, we have presented a highly nonlinear birefringent PCF. Simultaneous, birefringence, and nonlinear
(coefficient) properties of the proposed PCF have been
reported in this paper that to the best of our knowledge,
are much higher than any other results published so far in
literature. Moreover, two ZDW that is beneficial for supercontiniuum applications has been achieved. Also, it is
shown that a low effective area has a positive effect on
the bending losses and the proposed structure is bending
insensitive. The proposed PCF structure configuration is
straightforward when compared to many fabricated PCF
structures in literature. Therefore, fabrication of the proposed PCFs is believed to be possible and is not beyond
the realms of todays existing PCF technology. These
reported results can be widely used for the supercontinuum generation, polarization control in fiber-optic
sensors and telecommunication applications.
ENG
H. ADEMGIL
5. References
ET AL.
615
[1]
[2]
G, Renversez, B. Kuhlmey and R. McPhedran, Dispersion Management with Microstructured Optical Fibers:
Ultraflattened Chromatic Dispersion with Low Losses,
Optics Letter, Vol. 28, No. 12, 2003, pp. 989-991.
[3]
[15] R. T. Bise and D. J. Trevor, Sol-Gel Derived MicroStructured Fiber: Fabrication and Characterization, Optical Society of America, Optical Fiber Communications
Conference (OFC), Washington, DC, Vol. 3, March
2005.
[4]
[5]
[6]
[7]
[8]
[9]
[16] F. Poli, A. Cucinotta, S. Selleri and A. H. Bouk, Tailoring of Flattened Dispersion in Highly Nonlinear Photonic
Crystal Fibers, IEEE Photonics Technology Letter, Vol.
16, No. 4, 2004, pp. 1065-1067.
[17] K. Saitoh and M. Koshiba, Highly Nonlinear Dispersion-Flattened Photonic Crystal Fibers for Supercontinuum Generation in a Telecommunication Window,
Optics Express, Vol. 12, No. 10, 2004, pp. 2027-2032.
[18] T. Yamamoto, H. Kubota, S. Kawanishi, M. Tanaka and
S. Yamaguchi, Supercontinuum Generation at 1.55 M in
a Dispersion-Flattened Polarization-Maintaining Photonic
Crystal Fiber, Optics Express, Vol. 11, No. 13, 2003, pp.
1537-1540.
[19] M. Lehtonen, G. Genty, M. Kaivola and H. Ludvigsen,
Supercontinuum Generation in a Highly Birefringent
Microstructured Fiber, Applied Physics Letter, Vol. 82,
No. 14, 2003, pp. 2197-2199.
[20] A. Kudlinski, B. A. Cumberland, J. C. Travers, G.
Bouwmans, Y. Quiquempois and A. Mussot, CW Superconttinuum Generation in Photonic Crystal Fibres
with Two Zero-Dispersion Wavelength, AIP Conference
Proceedings, Sao Pedro, August 2008, pp. 15-18.
[21] B. A. Cumberland, J. C. Travers, S. V. Popov and J. R.
Taylor, 29 W High Power CW Supercontinuum
Source, Optics Express, Vol. 16, No. 8, 2008, pp. 59545962.
[22] N. H. Vu, I. K. Hwang and Y. H. Lee, Bending Loss
Analyses of Photonic Crystal Fibers Based on the FiniteDifference Time-Domain Method, Optics Letter, Vol. 33,
No. 2, 2008, pp. 119-121.
[23] T. Martynkien, J. Olszewski, M. Szpulak, G. Golojuch,
W. Urbanczyk, T. Nasilowski, F. Berghmans and H.
Thienpont, Experimental Investigations of Bending Loss
Oscillations in Large Mode Area Photonic Crystal Fibers, Optics Express, Vol. 15, No. 21, 2007, pp. 1354713556.
[24] K. Saitoh and M. Koshiba Single-Polarization SingleMode Photonic Crystal Fibers, IEEE Photonics Technology Letter, Vol. 15, No. 10, 2003, pp. 1384-1386.
[12] Y. S. Sun, Y.-F. Chau, H.-H. Yeh, L.-F. Shen, T.-J. Yang
and D. P. Tsai, High Birefringence Photonic Crystal Fiber with Complex Unit Cell of Asymmetry Elliptical Air
Holes Cladding, Applied Optics, Vol. 46, No. 22, 2007,
pp. 5276-5281.
ENG
616
H. ADEMGIL ET AL.
ENG
Abstract
In recent years, the narrow bandgap antimonide based compound semiconductors (ABCS) are widely regarded as the first candidate materials for fabrication of the third generation infrared photon detectors and
integrated circuits with ultra-high speed and ultra-low power consumption. Due to their unique bandgap
structure and physical properties, it makes a vast space to develop various novel devices, and becomes a hot
research area in many developed countries such as USA, Japan, Germany and Israel etc. Research progress
in the preparation and application of ABCS materials, existing problems and some latest results are briefly
introduced.
Keywords: Antimonide Based Compound Semiconductors (ABCS), IR Laser, IR Detector, Integrated Circuit,
Functional Device
1. Introduction
Antimonide based compound semiconductors (ABCS)
mainly refer to the antimonide based binary, ternary and
quaternary compound semiconductor materials, consisting of the III-group elements (Ga, In, Al, etc.) and Sb, As
and other V-group elements, such as GaSb, InSb, AlGaSb, InAsSb, AlGaAsSb, InGaAsSb and so on. Their
crystal lattices are around 6.1 and they together with
the InAs-based materials have been routinely called the
6.1 III-V family materials. Antimonide based semiconductors with narrow bandgap as the basic feature, in
the condition of lattice matched or nearly matched with
strain with GaSb, InAs, InP and other commonly used
substrates, their bandgap can be adjusted in a wide range
coveraging from near-infrared wavelength 0.78 m (AlSb)
to far-infrared spectral regions 12 m (InAsSb). The
heterojunctions formed between them can have type-I,
type-II staggered and type-II misaligned band lineups.
The unique band structure and excellent physical properties of ABCS based materials provide great freedom and
flexibility for band engineering and structural design of
materials and create a broad space for development of
high-performance microelectronics, opto-electronic devices and integrated circuits. Applications could include
active-array space-based radar, satellite communications,
ENG
C. LIU
618
ET AL.
InSb
GaSb
AlSb
InAs
GaAs
InP
GaN
Energy gap/(eV)
0.18
0.70
1.63
0.36
1.42
1.35
3.39
Electron mobility/
(cm2/V.s)
8104
5 000
200
3104
8 500
5 400
900
4.0
4.0
1.0
1.0
2.7
226
194
80
0.067
0.082 (L)
0.45 (H)
0.077
0.12 (L)
0.55 (H)
0.2
0.6
0.014
0.018 (L)
0.4 (H)
0.042
0.4
0.12
0.98
0.024
0.025 (L)
0.37 (H)
Thermal conductivity
(W/cm.K)
0.15
0.4
0.7
0.27
0.5
0.7
1.3
17.9
15.7
12.04
15.1
12.8
12.5
Effective mass
(m0)
Electron
Hole
C. LIU
ET AL.
619
Figure 1. Energy gap & spectral wavelength versus lattice constant, showing the evolution of HEMTs and HBTs transistors
for high-frequency and low-power operation [1].
620
C. LIU ET AL.
insulating properties and difficult to measure their electrical properties. The origin of O impurity is very complex, and both the purity of the metal organic sources and
the epitaxial environment and process conditions are
closely related. The development of new organic aluminum source such as TMAA, TTBAl, EDMAA etc. is
precisely in order to inhibit the serious C contamination
problem [4-5]. Thus, growing AlSb and their multielement materials using MOCVD is the most challenging
work in all the III-V epitaxial materials technologies.
The epitaxial growth of antimonide materials using
MBE was following earlier pioneering work of the IBM
group of L.L Chang and L. Esak, first on InAs/GaSb and
InAs/AlSb films [3]. Different from MOCVD process,
MBE uses ultra-high vacuum epitaxial environment with
single-element materials for molecular beam sources and
is easy to implement epitaxy of atomic layer and in situ
real-time monitoring, avoiding the C-pollution problem
which exits in Al-containing materials growing by
MOCVD and greatly reducing the concentration of O
doping. In fact most of the prototype devices having
complex fine structures and low-dimensional structures
(quantum wells, quantum wires and quantum dots) were
first achieved using materials grown by MBE. It is
noteworthy that, no matter MOCVD or MBE method,
the use of substrates whose surface orientation have a
small angle offset (i.e., low-density atomic step on the
surface of the substrate) seem to be more accessible
high-quality epitaxial layers. Experiments confirmed that
the use of GaSb (100) substrates miscut 2 towards (110)
or 6 towards (11) B may get higher crystal quality of
InGaAsSb and AlGaAsSb epitaxial layers [5]. To overcome the difficulty that antimonides have no semi-insulating substrate materials, the use of GaAs, Si and other
heterogeneous substrate materials for epitaxy of ABCS
films have also attracted great attention. H. Toyota, etc.
[10] reported that they grown high-quality GaSb/AlGaSb
multi-quantum well (MQW) structures with a 5nm AlSb
initiation layer and a relatively thick GaSb buffer layer
(0.5-2.0 m) grown on Si (001) substrates by molecular
beam epitaxy. The photoluminescence (PL) emission
around 1.55 m wavelength was observed for GaSb/
AlGaSb MQW structure at room temperature. Low dislocation density, high-quality GaSb epitaxial films on
GaAs (001) substrates stripe-patterned with SiO2 is also
prepared by MOCVD with low temperature epitaxial
lateral overgrowth (ELO) method [11].
Apart from common binary, ternary and quaternary
antimonides being composed of Al, Ga, In, As and Sb, in
order to extend the applications of antimonide-based
materials in the far-infrared band ( 5 m), easy to adjust
the material lattice constant to match the substrates lattice constant of GaSb, InAs et al. and develop new functional materials, recently some ternary, quaternary antimonides containing N ( 2%), P or Bi ( 2%) and fiveelements antimonides such as AlGaInAsSb, GaInNAsSb
Copyright 2010 SciRes.
etc have also aroused peoples concern and research interest [12-14]. T. Ashlet, etc. [12] found that the addition
of a small percentage of nitrogen ( 2%) to GaSb, InSb,
and GaInSb materials would significantly change their
energy band structures (bandgap become smaller) which
is very conducive to develop multi-band infrared detectors.
C. LIU
ET AL.
621
Si:1 ML InAs
622
C. LIU ET AL.
C. LIU
4. Conclusions
As the first candidate materials for fabrication of the
third generation large-scale focal plane arrays infrared
(photon) detectors, integrated circuits with ultra-high
speed and ultra-low power consumption and new high
efficiency thermophotovoltaic devices, the research and
development of antimonide based compound semiconductor materials and device applications are in the ascendant, attracting increasingly widespread concern and
research interests of researchers and institutions in the
world. Compared to currently more mature GaAs-based
and InP-based materials growth and device manufacturing process, the growth technology of antimonide based
micro-structure materials such as heterojunctions, superlattice quantum wells and self-aligned quantum dots continues to face considerable great difficulties and technical
challenges and the manufacturing process of various antimonides devices are far from mature. Therefore, there
Copyright 2010 SciRes.
ET AL.
623
5. Acknowledgements
This work was supported by the National Natural Science Foundation of China (Grant No. 60876004).
6. References
[1]
B. R. Bennett, R. Magno, J. B. Boos, et al., Antimonide-Based Compound Semiconductors for Electronic Devices: A Review, Solid-State Electronics, Vol. 49, No. 12,
2005, pp. 1875-1895.
[2]
[3]
[4]
[5]
[6]
A. Rogalshi, Material Considerations for Third Generation Infrared Photon Detectors, Infrared Physics &
Technology, Vol. 50, No. 2-3, 2007, pp. 240-252.
[7]
[8]
M. G. Mauk and V. M. Andreev, Gasb-Related Materials for TPV Cells, Semiconductor Science and Technology, Vol. 18, No. 5, 2003, pp. S191-S201.
[9]
M. Rosker and J. Shah, DARPAs Program on Antimonide Based Compound Semiconductors (ABCS),
IEEE GaAs Digest, 2003, p. 293.
ENG
624
C. LIU ET AL.
[10] H. Toyota, T. Sasaki and Y. Jinbo, Growth and Characterization of Gasb/Algasb Multi-Quantum Well Structures on Si (0 0 1) Substrates, Journal of Crystal Growth,
Vol. 310, No. 1, 2008, pp. 78-82.
[11] K. Zaima, R. Hasimoto, M. Ezaki, et al., Dislocation
Reduction of Gasb on Gaas by Metalorganic Chemical
Vapor Deposition with Epitaxial Lateral Overgrowth,
Journal of Crystal Growth, Vol. 310, No. 23, 2008, pp.
4843-4845.
[12] T. Ashley, L. BuckleG. W. Smith, et al., Dilute Antimonide Nitrides for Very Long Wavelength Infrared
Applications, Proceedings of SPIE, Orlando, 17 May
2006, pp. 62060L.
[13] P. H. Jefferson, L. Buckle , B. R. Bennett, et al., Growth
of Dilute Nitride Alloys of Gainsb Lattice-Matched to
Gasb, Journal of Crystal Growth, Vol. 304, No. 2, 2007,
pp. 338-341.
[14] D. Jackrel, A. Ptak, B. Seth, et al., Gainnassb Solar
Cells Grown by Molecular Beam Epitaxy, IEEE 4th
World Conference on Photovoltaic Energy Conversion.
Waikoloa, HI, 2006, pp. 783-786.
[15] J. B. Hacker, J. Bergman, G. Nagy, et al., An UltraLow Power Inas/Alsb HEMT W-Band Low-Noise Amplifier, IEEE MTT-S International Microwave Symposium, June 2005, pp. 1029-1032.
[16] M. K. Kwang, H. S. Jung, E. K. Dong, et al., The
Growth of a Low Defect Inas HEMT Structure on Si by
Using an Algasb Buffer Layer Containing Insb Quantum
Dots for Dislocation Termination, Nanotechnology, Vol.
20, No. 22, 2009, p. 225201.
[17] A. Rogalski. New Material Systems for Third Generation Infrared Photo Detectors, Opto-Electron Review,
Vol. 16, No. 4, 2008, pp. 458-482.
[18] M. Walther, R. Rehm, J. Schmitz, et al., AntimonyBased Superlattices for High-Performance Infrared Imagers, Proceedings of SPIE, Orlandoof, 17 March 2008, pp.
69400A-69400A-8.
[19] E. Corbin, M. J. Shaw, M. R. Kitchin, et al., Systematic
Study of Type II Ga1Xinxsb/Inas Superlattices for Infra-Red Detection in the 10-12 M Wavelength Range,
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
ENG
Abstract
Natural convective boundary layer flow and heat and mass transfer of a fluid with variable viscosity and
thermal radiation over a vertical stretching surface in the presence of suction/injection is investigated by Lie
group analysis. Fluid viscosity is assumed to vary as a linear function of temperature. The symmetry groups
admitted by the corresponding boundary value problem are obtained by using a special form of Lie group
transformations viz. scaling group of transformations. An exact solution is obtained for translation symmetry
and numerical solutions for scaling symmetry. The effects of fluid viscosity and thermal radiation on the dimensionless velocity, temperature and concentration profiles are shown graphically. Comparisons with previously published works are performed and excellent agreement between the results is obtained. The conclusion is drawn that the flow field and temperature profiles are significantly influenced by these parameters.
Keywords: Scaling Group of Transformations, Free Convective Flow, Temperature-Dependent Fluid
Viscosity, Suction/Blowing, Thermal Radiation
1. Introduction
The study of natural convection flow for an incompressible viscous fluid past a heated surface has attracted the
interest of many researchers in view of its important applications to many engineering problems such as cooling
of nuclear reactors, the boundary layer control in aerodynamics, crystal growth, food processing and cooling
towers. In this paper, symmetry methods are applied to a
natural convection boundary layer problem. The main
advantage of such methods is that they can successfully
be applied to non-linear differential equations. The symmetries of differential equations are those continuous
groups of transformations under which the differential
equations remain invariant, that is, a symmetry group
maps any solution to another solution. The symmetry
solutions are quite popular because they result in the reduction of the number of independent variables of the
problem.
A class of flow problems with obvious relevance to
polymer extrusion is the flow induced by the stretching
motion of a flat elastic sheet. In a melt-spinning process,
the extrudate from the die is generally drawn and simulCopyright 2010 SciRes.
626
P. LOGANATHAN
ET
AL.
2. Mathematical Analysis
We consider a free convective, laminar boundary layer
flow and heat and mass transfer of viscous incompressible fluid over a vertical stretching sheet emerging out of
a slit at origin (x = 0, y = 0) and moving with non-uniform velocity U(x) in the presence of thermal radiation
(Figure 1).
The governing equations of such type of flow are, in
the usual notations,
u v
0
(1)
x y
Figure 1. Physical model of boundary layer flow over a vertical stretching surface.
ENG
P. LOGANATHAN
u
u
v
x
y
1 T u 2 u
[ g (T T ) g * (C C )]
T y y x 2
T
T
T
1 qr
v
x
y c p y 2 c p y
(3)
C
C
2C
v
D 2
x
y
y
(4)
(2)
u U ( x), v V ( x), C Cw , T Tw at y 0
u 0, C C , T T as y
(5)
when the viscous dissipation term in the energy equation
is neglected (as the fluid velocity is very low). Here u
and v are the components of velocity respectively in
the x and y directions, is the coefficient of fluid
viscosity, is the fluid density (assumed constant), T
is the temperature, is the thermal conductivity of the
fluid, D is diffusional coefficient, is the volumetric
x
y c p y 2 3 c p k * y 2
(6)
T T
,v
,
and
(7)
x
y
C w C
Tw T
where u is the stream function. The stream wise
velocity and the suction/injection velocity are taken
as
U ( x) c x , V ( x) V0 x
m
m1
2
(8)
ET AL.
627
[1 (T T )]
y xy x y 2
2
3
v
v* [a (1 )] 3 g ( * )
2
b
y y
y
(9)
16 1T3 2
)
c p 3 c p k * y 2
y x x y
(10)
D 2
y x x y
y
(11)
where b(Tw T ), v*
2
xm ,
V0 x ,
y
x
1 at y 0;
0, 0, 0 as y
y
(12)
We now introduce the simplified form of Lie-group
transformations namely, the scaling group of transformations (Mukhopadhyay et al. [19]),
: x* x e 1 , y* y e 2 , * e 3 ,
u* u e 4 , v* v e 5 , * e 6 , * e 7
(13)
P. LOGANATHAN
628
v*e (3 2 3 6 ) *
*2
y y
(14)
3 *
3 *
v* [a ]e (3 2 3 ) *3 v*e (3 2 3 6 ) * *3
y
y
g
( e
* 7
e
*
)
*
1
1
These relations give 6 7 0, 2 1 3 .
4
3
1
The boundary conditions yield 4 m1 1 ,
2
m 1
1
1
5
1 1 (as m ) .
2
4
2
In view of these, the boundary conditions become
x ,
V0 x
*
x
y
1
( )
* 4
, y* y e
1
4
, * e
3
1
4
u u u
*
1
2
, v v v
*
1
4
1
4
, *
*
3F 0
4V0
, 1 at
3
0 and F 0, 0, 0 as
F 1, F
31
,
4
(19)
(20)
(21)
(22)
4
3F 0
Sc
, 0
*
(18)
(17)
1
4
x* x x 1 , y* y y
( * )
2 F 2 3FF 4 ( )
u * u e , v* v e
, * ,, *
Expanding by Taylors method in powers of and
keeping terms up to the order we get
2
4 D 3F 0
*
0, * 0, * 0 as y*
0 and
y*
The set of transformations reduces to
4(a )v* F 4 v* F 4 g
, 1 at y
*
16 T
4
c p 3 c k
1 2 2 2 3 3 2 3 6 3 2 3
6 7 ; 1 2 3 6
2 2 6 and 1 2 3 7 2 2 7
3
4
, x F ( ), * ( ), * ( )
*
3
1
*
p
(16)
2 *
De
y*2
The system will remain invariant under the group of
transformations , we would have the following relations
among the parameters, namely
x* x e
y x
1
4
* * * *
*
*
*
x y*
y x
(1 2 3 7 )
1
*2
* *
(15)
(2 2 7 )
2 F 2 3FF 4 v* F
16 1T3 (2 2 6 ) 2 *
e
c p 3 c k *
y*2
p
dx
dy
d
du
dv
d d
3
x 1
0
0
y 1 1 u 1 v 1
4
4
2
4
With the help of these relations, the (14), (15) and (16)
become
e (1 2 3 6 ) *
*
*
x y *
y x
*
AL.
* 2 * * 2 *
*
e (1 2 2 23 ) *
*
*
x y*2
y x y
*
ET
(25)
(26)
*c p
is the Prandtl number,
k*
v*
N
is the Radiation parameter, Sc is the
3
D
4 1T
Schmidt number. The boundary conditions take the following forms.
where
Pr
v* c p
(24)
f 1, f S , 1,
at * 0 and f 0, 0, 0 as *
(27)
ENG
P. LOGANATHAN
1
g 4 2
4
where S V0 ( 1 ) v , S 0 corresponds to suction
3
b
and S 0 corresponds to injection.
3. Numerical Solution
The set of non-linear ordinary differential Equations (24)
to (26) with boundary conditions (27) have been solved
by using the R. K. Gill method, (Gill [24]) along with
Shooting Technique with , Pr, Sc, a and N as prescribed parameters. The numerical solution was done
using Matlab computational software. A step size of
= 0.001 was selected to be satisfactory for a convergence criterion of 10 7 in nearly all cases. The value of
was found to each iteration loop by assignment
statement = + . The maximum value of ,
to each group of parameters , Pr, Sc, a and N, determined when the values of unknown boundary conditions
at = 0 not change to successful loop with error less
ET AL.
629
Prandtl number Pr , Schmidt number Sc and radiationparameter N . For illustrations of the results, numericalvalues are plotted in the Figures 2-9. In all cases we
take a 1.0 .
In the absence of diffusion equations, in order to ascertain the accuracy of our numerical results, the present
study is compared with the available exact solution in the
literature. The temperature profiles for Prandtl number
Pr are compared with the available exact solution of
1.2
1
0.8
0.6
( )
Pr = 0.3
Pr = 0.5
0.4
0.2
Pr = 1.0
0
0.5
1.5
Figure 2. Influence of Prandtl number over the temperature profiles. Sc 0.0, N 0.1, a 1.0, S 0.5, 0.5 .
1.2
1.0
0.8
0.6
( )
0.4
N = 0.5
0.2
0.0
N = 1.0
N = 3.0
0.0
-0.2
0.5
1.0
1.5
Figure 3. Effects of thermal radiation over the temperature profiles. Pr 0.3, a 1.0, S 0.5, 0.5, Sc 0.62 .
ENG
P. LOGANATHAN
630
ET
AL.
1.6
1.4
= 1.0
1.2
= 0.5
1
f '() 0.8
= 0.1
0.6
0.4
0.2
0
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
Figure 4. Effects of fluid viscosity over the velocity profiles. Sc 0.62, N 0.1, a 1.0, S 0.5, Pr 0.71 .
1.2
0.8
0.6
( )
0.4
= 0.1
= 0.5
= 1.0
0.2
0.5
-0.2
1.5
Figure 5. Effects of fluid viscosity over the temperature profiles. Sc 0.62, N 0.1, a 1.0, S 0.5, Pr 0.71 .
ENG
P. LOGANATHAN
ET AL.
631
1.6
1.4
1.2
S = 3.0
S = 5.0
f '()
0.8
S = 8.0
0.6
0.4
0.2
0
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
Figure 6. Effect of suction over the velocity profiles. S 1.0, N 0.1, a 1.0, 0.5, Pr 0.71 .
1.2
0.8
0.6
( )
S = 3.0
0.4
S = 5.0
0.2
S = 8.0
0
0
0.5
1.5
-0.2
Figure 7. Influence of suction over the temperature profiles. Sc 0.62, N 0.1, a 1.0, 0.5, Pr 0.71 .
ENG
P. LOGANATHAN
632
ET
AL.
1.2
0.8
0.6
( )
S = 3.0
0.4
S = 5.0
0.2
S = 8.0
0
0
0.5
-0.2
1.5
Figure 8. Effects of suction over the concentration profiles. Sc 0.62, N 0.1, a 1.0, 0.5, Pr 0.71 .
1.2
0.8
Sc = 0.22
0.6
( )
0.4
Sc = 0.62
0.2
Sc = 0.78
0
0
-0.2
0.5
1.5
Figure 9. Effects of Schmidt number over the concentration profiles. S 1.0, N 0.1, a 1.0, 0.5, Pr 0.71 .
ENG
P. LOGANATHAN
qr
increases
y
as decreases which in turn increases the rate of radiative heat transferred to the fluid and hence the fluid temperature increases. In view of this explanation, the effect
of radiation becomes more significant as N 0 ( N
0) and can be neglected when N . Also, it is seen
from Figure 3 that the larger the N , the thinner the
thermal boundary layer thickness for both values of Pr .
In addition, radiation demonstrates a more pronounced
influence on the temperature distribution of ( Pr 0.3 )
than that of ( Pr 1.0 ). It is noticed from the figure that
the temperature decreases with the increasing value of
the radiation parameter N . The effect of radiation parameter N is to reduce the temperature significantly in
the flow region. The increase in radiation parameter
means the release of heat energy from the flow region
and so the fluid temperature decreases as the thermal
boundary layer thickness becomes thinner.
Figure 4 exhibits the velocity profiles for several values of with Pr = 0.71 in presence of suction (S = 0.5)
when N = 0.1. In the case of uniform suction, the velocity of the fluid is found to increase with the increase of
the temperature-dependent fluid viscosity parameter
at a particular value of except very near the wall as
well as far away of the wall (at = 5). This means that
the velocity decreases (with the increasing value of )
at a slower rate with the increase of the parameter at
very near the wall as well as far away of the wall. This
can be explained physically as the parameter increases, the fluid viscosity decreases the increment of the
boundary layer thickness.
In Figure 5, variations of temperature field ( ) with
for several values of (with Pr = 0.71 and N = 0.1)
in presence of suction (S = 0.5) are shown. It is very
clear from the figure that the temperature decreases with
the increasing of whereas the concentration of the
fluid is not significant with the increasing of . The
increase of temperature-dependent fluid viscosity parameter ( ) makes decrease of thermal boundary layer
thickness, which results in decrease of temperature profile ( ) . Decrease in ( ) means a decrease in the
velocity of the fluid particles. So in this case the fluid
particles undergo two opposite forces: one increases the
fluid velocity due to decrease in the fluid viscosity (with
increasing ) and other decreases the fluid velocity due
to decrease in temperature ( ) (since ( ) decreases with increasing ). Near the surface, as the
temperature ( ) is high so the first force dominates and
far away from the surface ( ) is low and so the second
Copyright 2010 SciRes.
ET AL.
633
force dominates here. Now we concentrate in the velocity and temperature distribution for the variation of suction parameter S in the absence and presence of temperature-dependent fluid viscosity parameter .
Figure 6 presents the effects of suction on fluid velocity when the fluid viscosity is uniform, i.e., = 0.5.
With the increasing value of the suction (S > 0) ( =
0.5, Pr = 0.71 and N = 0.1), the velocity is found to decrease (Figure 6), i.e., suction causes to decrease the
velocity of the fluid in the boundary layer region. The
physical explanation for such a behavior is as follows. In
case of suction, the heated fluid is pushed towards the
wall where the buoyancy forces can act to retard the fluid
due to high influence of the viscosity. This effect acts to
decrease the wall shear stress. Figures 7 and 8 exhibit
that the temperature ( ) and concentration ( ) in
boundary layer also decrease with the increasing suction
parameter S (S > 0) ( = 0.5, Pr = 0.71 and N = 0.1).
The thermal and solutal boundary layer thickness decrease with the suction parameter S which causes an increase in the rate of heat and mass transfer. The explanation for such behavior is that the fluid is brought closer to
the surface and reduces the thermal boundary layer
thickness in case of suction. As such then the presence of
wall suction decreases velocity boundary layer thicknesses but decreases the thermal and solutal boundary
layers thickness, i.e., thins out the thermal and solutal
boundary layers.
Figure 9 illustrates the influence of Schmidt number
Sc on the concentration. As Schmidt number Sc increases,
the mass transfer rates increases. Hence, the concentration decreases with increasing Sc. It is evident from this
figure that the concentration ( ) takes its limiting
value C, for higher values of the dimensionless distance
. From this figure, we observe that when the concentration difference C is maintained constant, the dimensionless concentration profile decreases, in the since that
the values of the Schmidt number increases. The variation in the thermal boundary layer is very small corresponding to a moderate change in Schmidt number.
There are very small changes in velocity and temperature
distributions when moderate changes in Schmidt number.
5. Conclusions
By using Lie group analysis, first find the symmetries of
the partial differential equations and then reduce the
equations to ordinary differential equations by using
scaling and translational symmetries. Exact solutions for
translation symmetry and numerical solution for scaling
symmetry are obtained. From the numerical results, it is
predict that the effect of increasing temperaturedependent fluid viscosity parameter on a viscous incompressible fluid is to increase the flow velocity which in
ENG
P. LOGANATHAN
634
6. References
[1]
[2]
[3]
[4]
[5]
[6]
E. M. A. Elbashbeshy and M. A. A. Bazid, Heat Transfer in a Porous Medium over a Stretching Surface with
Internal Heat Generation and Suction or Injection, Applied Mathematics and Computation, Vol. 158, No. 3,
2004, pp. 799-807.
[7]
[8]
[9]
ET
AL.
ENG
Abstract
Creep feed grinding is a recently invented process of material handling. It combines high quality of the piece
surface, productivity, and the possibility of automatic control. The main objectives of this research is to study
the influences of major process parameters and their interactions of creep feed grinding process such as
wheel speed, workpiece speed, grinding depth, and dresser speed on the pin gauge dimensions of root of gas
turbine blade by design of experiments (DOE). Experimental results are analyzed by analysis of variance
(ANOVA) and empirical models of pin gauge dimensions of root are developed. The study found that higher
wheel speed along with slower workpiece speed, lower grinding depth and higher dresser speed, cause to
obtain best conditions for pin gauge dimensions of root.
Keywords: Creep Feed Grinding, Pin gauge dimension, Analysis of Variance, Regression,
Interactive Effect
1. Introduction
Grinding has traditionally been associated with small
rates of material removal and fine finishing operations.
Using an approach known as creep-feed grinding (Figure 1), a large-scale metal removal similar to milling can
be achieved. Using this approach, higher material removal rates can be performed by selection of a higher
depth of cut and lower workpiece speed. The correct
selection of the cutting conditions and the wheel specifications can provide a greater material removal rate and a
finer surface quality. One of the most important applications of creep-feed grinding is the production of the aerospace parts used in jet engines such as turbine vanes,
and blades where parts should have high strength to the
fatigue loads and creep strains. These parts are made
from nickel-based super-alloys such as Inconel, Udimet,
Rene, Waspaloy, and Hastelloy. They provide a higher
strength to weight ratio, and maintain high resistance to
corrosion, mechanical thermal fatigue, and mechanical
and thermal shocks [1].
Vafaeesefat modeled and predicted the grinding forces
of the creep feed grinding of supper-alloy materials using
neural network. This model was then used to select the
working conditions (such as depth of cut, the wheel
speed, and workpiece speeds) to prevent the surface
burning and to maximize the material removal rate. The
Copyright 2010 SciRes.
A. R. FAZELI
636
temperature in dry grinding that failed to take into account the cooling effect of the fluid.
Lavine et al. [7] presented a conical grain model, with
grain slope set to one. Lavine and Jen [8] derived a separate thermal model among the fluid, wheel and workpiece to predict the occurrence of boiling.
Wange et al. [9] depicted that the grinding energy
when the fluid begins to cause boiling is defined as the
critical grinding energy for the workpiece burning. The
results showed that the workpiece burning can be predicted or evaluated to avoid the working conditions of
burning occurrence.
Shafto et al. [10] proposed that workpiece burning
could be explained by the phenomenon of fluid film
boiling.
Ohishi and Furukawa [11] derived the relationship
between the grinding heat flux and grinding zone temperature at burning using the fraction of the grinding
energy entering into the workpiece at 10%.
Wange et al. [12] modeled the grinding force of the
creep feed grinding using the improved back propagation
neural (BPN) network in view of avoiding the workpiece
burning. The results showed that the grinding energy can
be accurately predicted by the application of the grinding
force model and that a larger size of wheel is available to
have a better working efficiency.
Pin gauge dimension is one of the important geometrical dimensions in root of gas turbine blade that plays an
important role in correct assembly of blade on disk of
turbine. If this important dimension is not properly controlled and goes out of its tolerance range (within 0.062),
the blade will not fit on the disk of turbine.
Figure 2 shows the gas turbine blade. Figure 3 illustrates the pin gauge dimension that is one of the important geometrical dimensions in root of gas turbine blade.
In this research, the influences of major process parameters and their interactions of creep feed grinding
process such as wheel speed, workpiece speed, grinding
depth and dresser speed on the root geometrical dimensions of gas turbine blade is studied using design of experiments (DOE).
It is desirable to know the effects of the major parameters and interactive influences among the process
parameters on root geometrical dimensions and relationship between root geometrical dimensions and process
parameters to obtain the best conditions of parameters
for optimum production.
For modeling and determining the influences of main
parameters and interaction effects among parameters of
the process on root geometrical dimensions, design of
experiments method (DOE) has been employed. DOE is
a statistical method which is used to find the significance
of interactive effects among variables and relations
among process parameters using variance analysis. Finally, using this model and the suitable pin gauge dimension, input parameters has been achieved for optimum
production.
2. Description of Material
We have chosen Inconel 738 LC supper-alloy as the experimental sample. This supper-alloy provides higher
strength to weight ratio, and maintains high resistance to
corrosion, mechanical thermal fatigue, and mechanical
and thermal shocks. The chemical composition of this
supper-alloy is presented in Table 1.
3. Experimental Modeling
3.1. The Output Parameters
Output parameter, pin gauge dimension measured in
terms of mm with inside micrometer 0-25, 0.01 mm precision.
Input parameters were selected from the various parameters of creep feed grinding process such as the properties of the work piece material, tools, dresser rotational
speed, rigidity of machine tools and type of coolant. The
selected parameters are:
ENG
A. R. FAZELI
Table 1. The chemical compositions of Inconel 738 LC supper-alloy.
Min
Max
Element
Min
Element
Percentage
C
Max
0.09
0.13
Percentage
Nb
0.6
1.1
Cr
15.7
16.3
Ta
1.5
Co
2.4
2.8
Al
3.2
3.7
Fe
0.3
Ti
3.2
3.7
Si
0.05
(AI + Ti)
6.5
7.2
Mn
0.05
0.007
0.009
0.003
Zr
0.03
0.06
Mo
1.5
Ni
Bal.
637
Table 2. The parameter levels.
Parameters
Low
Level
High
Level
17
25
100
180
0.05
0.15
0.6
0.9
0.3
0.6
0.04
0.08
A. R. FAZELI
638
P1 (mm)
P2 (mm)
P3 (mm)
V (m/s)
F (mm/min)
E (m/rev)
0.75
0.45
0.06
140
0.1
21
0.75
0.45
0.06
140
0.1
21
0.75
0.45
0.06
140
0.1
21
ENG
A. R. FAZELI
Parameters
Coefficient
P-Value
Constant
0.3
0.001
P1
0.683
0.174
P2
0.067
0. 758
P3
0.460
0.001
0.04
0.74
0.017
0.012
0.012
P1 P2
0.222
0.19
P1 P3
1.166
0.322
P1 f
0.001
0.032
P1 E
1.366
0.04
P1 V
0.028
0.01
R-Sq = %97.89
sion.
Figure 10 summarizes the dresser speed on the pin
gauge dimension at grinding depth (first section). The
results show that an increase of dresser speed combined
with the increase of grinding depth (first section), produces small absolute value of difference between the
measured dimensions and nominal dimension of pin
gauge dimension.
Reasonably, with higher wheel speed, grinding depth
(first section) and slower workpiece speed, machining
forces apply equally on the other side of root. Therefore
R-Sq(adj) = %89.43
639
Figure 8. Effect of the wheel speed on the pin gauge dimension at grinding depth (first section).
(2)
R3 0.13 0.453(P1) 0.033(P2) 0.0005(f )
0.255(E) 0.006(V) 0.683(P1 E) 0.272(P1 P2)
0.0008(P1 f ) 0.11(P2 V) - 0.018(P1 V)
(3)
5. Discussion
Figure 8 summarizes the wheel speed on the pin gauge
dimension at grinding depth (first section). The results
show that an increase of wheel speed combined with the
increase of grinding depth (first section), produces small
absolute value of pin gauge dimension.
Figure 9 shows the effect of workpiece speed on the
pin gauge dimension at grinding depth (first section).
The results show that a decrease of workpiece speed
combined with the increase of grinding depth (first section) produces small absolute value of pin gauge dimenCopyright 2010 SciRes.
ENG
A. R. FAZELI
640
6. Conclusions
In this study, the creep feed grinding process has been
optimized by selection of significant input parameters
including the wheel speed, dresser speed, grinding depth
and slower workpiece speed. Finally, by means of
ANOVA, the main effects of the input parameters and
their interactions on the pin gauge dimension were determined. Based on the statistical analysis of the experimental data the following conclusions can be obtained.
1) According to the variance analysis and the effect of
interactions between the input parameters, it can be concluded that with higher wheel speed, dresser speed,
grinding depth (first section) and lower grinding depth
(second and third section) and slower workpiece speed,
the pin gauge dimension decreases and as a result the pin
gauge dimension reaches a suitable level.
2) In the creep feed grinding process center points
have insignificant effects on the pin gauge dimension. It
means that the process can be modeled with two levels
for each input parameters.
3) Finally, with the large number of effective parameters in the creep feed grinding process, consideration of
the creep feed grinding process through the design of
experiments is shown to be the efficient method for
achieving the acceptable results.
7. References
[1]
[2]
S.-B. Wang and H.-S. Kou, Selections of Working Conditions for Creep Feed Grinding. Part (I)Thermal Partition ratios, International Journal of Advanced Manu-
[3]
[4]
[5]
[6]
[7]
[8]
[9]
S.-B. Wang and H.-S. Kou, Selections of Working Conditions for Creep Feed Grinding. Part (II): Workpiece
Temperature and Critical Grinding Energy for Burning,
International Journal of Advanced Manufacturing Technology, Vol. 28, No. 1-2, 2006, pp. 38-44.
ENG
Abstract
The purpose of this studyy is to optim
mize the mass of 1.5 MW wind turbinee steel tower performing Genetic
Algorithm method
m
(GA). In accordancce with ASCE
E 7-98, AISC
C-89 and IEC
C61400-1 , the impact of lloads on
tower is calculated withinn the highest safety condittions against buckling
b
strength of each sections of toower by
means of GA
A codes. Thee stifness alon
ng tower is eensured entirrely while thee mass of tow
wer is mitigaated and
optimized.
Keywords: Mass
M Optimizzation, Genetiic Algorithm, Wind Turbiine Tower
1. Introducction
(K r )
1
..Fy
2.Cc 2
Fa
(1)
L
L
3( K ) ( K )3
5
r
r
3
8Cc
8Cc 3
Copyright 2010
2
SciRes.
L
r
of steel ttower, K
) is thee slenderness ratio
r
= 2 forr the cantileveer type of structure, and L aand r are
the lenggth of the toweer and radius of
o section, resppectively.
The maaterial coefficient Cc is calculated by:
where ( K
Cc
2. 2 .E
Fy
(2)
ENG
S. YILDIRIM ET AL.
642
fa fb
1
Fa Fb
(4)
3. Fatigue Load
The DEL (Damage Equivalent Load) method facilitates
to determine the steel tower preliminary dimensions in
any circumstances which fatigue load histogram data
does not exist. The SN curve for the DEL method can be
expressed in the following [2]:
loglog s n loglog 80 MPa
2 106 n
m
(5)
N z
Mf z
rmax S z .m N0
(6)
where:
Mf ( z ) is the moment produced by the fatigue DEL
thrust along steel tower. S z is the section modulus
that varies along the height of tower. rmax is the
maximum allowable stress range at N 0 cycles (typi-
h
max max Mx, yB
(7)
max(Mx,yT) = Maximum moment range at tower top
x or y direction .
max(Mx,yB) = Maximum moment range at tower
base x or y direction.
Safety Factor of DEL is 1.0.
Consequence failure factor and material factor: sd.m
= 1.15 1.1 = 1.265
Number of cycles: 5.29 108 for 1.5 MW turbine
This represents a 20 year lifetime.
1 0.01. r if r 212
t
t
0
(9)
0.70
r
212
if
t
r
0.1
0.01
.
Fy
if B . cr
0.175. B . cr
2
The maximum applied stress a combined with normal
stress and shear stress is calculated by a :
fa fb
3 fv 2
(12)
S. YILDIRIM ET AL.
1 fb 0.15 Fb
(13)
Fa Fb
where:
fa
is the applied compression stress
Fa is the allowable stress
fb
is the applied bending stress
Fb is the allowable bending stress.
5. Earthquake Load
This section is based on ASCE 7-98 Earthquake Load
Specification. Even though earthquake load seems to be
not much significant effect on design of steel tower because of the fact that wind turbine towers are placed in
low seismic areas, earthquake load should be taken into
consideration so as to be more precise in designing of
steel tubular tower (see Figure 1).
The maximum considered earthquake spectral response acceleration for short periods ( SMS ) and at 1
adjusted for site class effects, should be
second
determined by [3]:
S MS Fa.Ss; S M 1 Fv.S1
(14)
where :
Mapped maximum considered earthquake spectral response acceleration at a period of 1s as determined
in accordance with Section 9.4.1 (ASCE 7-89).
Ss Mapped maximum considered earthquake spectral response acceleration at short periods as determined
in accordance with Section 9.4.1 (ASCE 7-89).
Fa and Fv are defined in Tables 9.4.1.2.4a and b
respectively in accordance with Section 9.4.1 (ASCE
7-89). According to the ASCE 7-89 9.4.1.2.5 design
spectral response acceleration at short periods, S DS and
at 1 s period S D1 shall be determined from Equations
9.4.1.2.5-1 and 9.4.1.2.5-2 respectively:
S DS
2
2
S MS ; S D1 S M 1
3
3
(15)
643
T
(17)
S a T S DS 0.4 0.6
T T0
T
0 , if
S
Otherwise
DS
TS S DI / S DS ; T0 0.2.Ts
(18)
T is the structural period.
Spectral acceleration response can be found as in Figure 2.
W w z dz WHead Mass
(19)
ENG
S. YILDIRIM ET AL.
644
I
(20)
Cs T Sa t .
R
I is the importance factor. R is the reduction factor
equal to 1.
Base Shear is V C s T W (Figure 3)
(21)
dary layer is 213 m and 1 is 11.5 for exposure D category in accordance with ASCE 7-98.
The towers are assumed to be located in flat unobstructed area for direct wind exposure Category D where
wind flows over the open water and flat terrain.
Importance factor is 1.0 for low occupancy concerning
the wind turbine erection and installation.
1 1.7 I g 2 Q 2 g 2 R 2
z
Q
R
G f 0.925
1 1.7 g v I z
(22)
where:
The topographic factor K zt is 1.0 for the flat area.
K d is 0.95 for a round cylinder tower in accordance
with Table 6-6 in ASCE 7-98.
The terrain exposure coefficient is determined as per
Table 6-5 of ASCE 7-98 or by the following formula [3]:
2
15 ft 1
) ifz 15 ft
2.01(
zg
Kz z
(23)
2
z 1
otherwise
2.01( z )
g
(24)
where
The intensity factor of turbulence :
I z 0.15(33 ft / z )1/ 6
(25)
0.577
2 ln(3600n1 )
(27)
7.47 N1
Rn
(1 10.3N1 )5/3
(28)
(29)
(Figure 5)
(30)
Vz z Fz x dx
(Figure 6)
(31)
Mz z Fz z .( x z ) dx
(Figure 7)
(32)
ENG
S. YILDIRIM ET AL.
645
the height of tower. E and I are elastic modulus and moment of inertia of the tower. Wt and Wtow are the
weight of head mass and tower mass respectively.The
natural frequency in Hz is calculated by: ft wt / 2 .
Genetic Algorithms is one of the methods used in optimization problems. Particularly, it is based on natural
selection. Genetic Algorithms is dependent on that the
best generation has to live in nature. Although many genetic algorithms have been said with different structures,
all of them comprises of three basic operations. Genetic
Algorithm uses reproduction, crossover and mutation
operators to define fitness and to create new solutions.
Reproduction is simply a process to make decision which
strings should remain and how many copies of them
should be produced in the pool. The decision is made by
comparing the fitness of each string. The fitness indicates
survival potential and reproduction efficiency of the
string in the next generations. For an Optimization problem, the fitness function is the objective function of optimization problem as shown in Figure 8.
9. Optimization Problem
ENG
S. YILDIRIM ET AL.
646
INPUT DATA
1) General Specifications
2) Material Characteristics
3) Structural Parameters
4) Load conditions
5) GA Parameters
Load Calculation
6) ASCE 7-98/IEC/Eurocode
Loads
7) Safety Factors
OPTIMIZATION PROGRAM
1) Natural frequency
2) Extreme Loads
3) Fatigue damage
4) Fitness function
OUTPUT PARAMETERS
1) Wall thicknesses of each segment
ENG
S. YILDIRIM ET AL.
of turbine .
Design flow is shown as follows. This GA structure
minimizes tower mass subject to general dimensions,
design loads and some design restrictions. Load calculation depends on wind turbine design requirements of the
standard IEC61400-1 and ASCE 7-98. All extreme loads
of tower sections are calculated by the load combination
[5] as shown in Figure 11.
11. Conclusions
[1]
[2]
[3]
[4]
[5]
12. References
ENG
Abstract
This paper describes a device capable of producing net impulse, through two synchronized masses, which
move along a figure-eight-shaped orbit. In addition to the detailed description of the mechanical components
of this device, particular attention is paid to the theoretical treatment of the innovative principle on which the
device is based. In more details, the mechanical system consists of two independent but simultaneous rotations, the former being related to the formation of the figure-eight-shaped path and the latter to an additional
spinning. Based on the parametric equations of motion of the lumped masses, and considering semi-static
tensile deformation of the connecting rods carrying them, it was found that the resultant impulse towards the
direction of the spin vector includes a non-vanishing term that is linearly proportional to the time. In addition,
reduced but encouraging experimental results are reported. These findings sustain the capability of the proposed mechanism to achieve propulsion.
Keywords: Inertial Propulsion, Centrifugal Force, Net Impulse, Rotating Figure-Eight, Mechanism
1. Introduction
The matter of the inertial propulsion utilizing eccentric
masses aiming at producing limited motion of the object
to which they are attached, is an old topic [1-4]. The
general impression is that these masses lead to periodic
oscillations in which the synchronization of participating
masses plays a significant role [5], while chaotic phenomena may also appear [6,7].
In the particular case of possible space propulsion,
fifty years ago Norman Dean (a civil service employee
residing in Washington DC) proposed the use of two
contra-rotating eccentric masses in order to convert rotary motion to unidirectional motion [4]. He claimed that
in this way one could achieve thrust thus producing motion of the object to which this system was attached.
Since then, Deans mechanism was internationally named as Dean drive or Dean space drive (the interested
reader may consult, for example, http://en.wikipedia.
org/wiki/Dean_drive). However, despite the extremely
high number of internet references as well as the many
articles sited in popular mechanics or science fiction magazines, a very small number of scientific papers exist in
the open literature. A careful search reveals an old paper
ENG
C. G. PROVATIDIS
649
650
C. G. PROVATIDIS
locity shaft motor where 1 is the speed reducetion of the transmission between the motor M1 and the
right half of the shaft (2). Thus power transmission is
performed through P1-S1 towards the mass a. Similarly,
the rest half of the power produced by the motor M1 is
transmitted through P1-S2 towards the other mass b. A
characteristic of this mechanism is that the second planet
gear, P2, is fixed thus causing rolling of the spin gears S1
and S2 on P2. Obviously, the rotation of the planet gear
P1 enforces the spin gear S1 to rotate about its local axis
(initially coinciding with the global z-axis) and also enforces the casing (11) to rotate around x-axis.
2.2. Operation
In brief, the motor M1 rotates at an angular velocity
motor and drives the planet gear (P1) at an angular ve-
Figure 3. Abstractive sketch of the planetary system. The mechanical power flows through the planet gear (P1) to the spin
gears (S1) and (S2). The lumped masses a and b are attached to the free ends of the rods (4) and (5), respectively, which
are firmly connected with the aforementioned spin gears (S1) and (S2).
ENG
C. G. PROVATIDIS
cos t
R 01 sin t
initial time, t = 0. In more details, during the time interval t, not only the two rods rotate around the axes of the
spin gears but also the casing in such a way that
casing . In the general case dealt in this paper, the
entire system rotates about the z-axis at an angular velocity z , using a second motor M2 shown in Figure 1 as
well as in Figure 2.
cos t
0
1
R12 0 cos t
0 sin t
cos z t
R 23 sin z t
(2)
sin t
cos t
0
sin z t
cos z t
0
(3)
(4)
x t
y t R
z t
x
y
t 0
(5)
r sin t R cos t
2
x t
y t R
z t
x
y
t 0
r sin t R cos t
(6)
sin t
R a R 23 R 12 R 01
651
(1)
(7)
(8)
(9)
ENG
C. G. PROVATIDIS
652
z
2
R 4
rcost 2 r
z
(10)
(21)
+z
Fxa Fxb mr
2
2
(20)
2
2
xa t z r cos t cos z t
(11)
Fya Fyb mr
(22)
za t
2r cos 2t Rcost
(12)
(13)
(14)
zb t rsin2t Rsint
(15)
(16)
(17)
(25)
nrzb zb zb / r
where
Based on the abovementioned kinematics, the components of the inertial force exerted on the mass a can be
calculated by:
while those on the mass b:
nrxb xb xb / r
nryb yb yb / r
(18)
5. Inertial Forces
(24)
nrza za za / r
and
zb t 2 2r cos 2t +Rcost
zk
nrya ya ya / r
yk
nrxa xa xa / r
2 z r sin t cos z t
xk
(23)
2
2
yb t z r cos t sin z t
z 2 r cos t cos z t
(19)
xa R sin t sin z t
, xb R sin t sin z t
(26)
, zb R cos t
ENG
C. G. PROVATIDIS
Fz Fza Fzb 2m
ra
(27)
EA
(28)
Similarly, for the rod that corresponds to the mass particle mass b it holds:
Frb r
rb
(29)
EA
r
2
z
Frb mr z cos t
2
1 sin t 2
2
cos t (31)
3
cos t
3
(32)
EA
z cos t
2
m
EA
z r
2
1 sin t
(33)
t 2 cos 2 t 2 z r 2 cos 3 t
(34)
Similarly, for the rod that corresponds to the mass particle mass b it holds:
rb
EA
(35)
rb r
EA
z r
2
cos
r
4
cos
t 2
(37)
(38)
4m
2
(40)
(41)
(42)
EA
cos 2 t
2
z
(43)
7. Impulse
The impulse caused by the vertical resultant force
is given by:
Fz t
(44)
(45)
where
I z , rigid 2 m r sin 2
(46)
and
(47)
with
m
2
t 2 cos t
2
representing the vertical resultant force due to the rigidbody motion of the rods, and
c0
(36)
t 2 cos t 2 z r cos t
2
2 cos t
2 z cos
cos t 1 sin t
2
with
2 r 2 cos 2 t 2
cos
cos
EA
I z t Fz d
2 z cos t
(39)
(30)
1 sin t 2
Fz 4m cos 2t r
Fz, tension
ra rb cos 2t
653
EA
2
z
m 2 9
2
2 2
z 3 r
EA 8
2
m 2
c4
z 2 r 2
c2
(48)
4 EA
m
2
c6
24 EA
z2 r 2
ENG
C. G. PROVATIDIS
654
lead to zero values every 180, 90 and 60 degrees, respectively, thus not practically contributing to the propulsion.
Concerning particularly the second harmonic ( sin 2 ),
not only the elastic part of the impulse vanishes every
180 degrees, but also that caused by the rigid-body motion (cf. Equation (46)). However, in addition to the three
harmonic terms, (47) includes also the term c t ,
which increases linearly with the time t. The analytical
expression of the term c in (48) depicts that for a given
angular frequency and given elasticity properties of
the two rods, the value of the net impulse is fully controlled by the angular frequency .
0
9. Discussion
It is remarkable that the findings of this paper are in consistency with previous experimental results (8% weightreduction) related to ideally rigid gyroscopes [16]. Since
the proposed device is essentially a flexible (elastic) gyroscope of which the masses operate in a hemisphere, it
is believed that similar accurate experiments with those of [16] should be performed for the current device.
Copyright 2010 SciRes.
ENG
C. G. PROVATIDIS
10. Conclusions
This work contributes to the field of inertial propulsion,
proposing a new concept for the production of net impulse through rotating masses. In the beginning, it was
shown that when a lumped mass moves along a circumference, it repeats its position every 360 degrees, thus its
initial linear momentum is repeated and no net impulse is
finally produced for a whole revolution; this case corresponds to the notorious Dean drive. Then, it was theoretically shown that when two lumped masses move along
a specific figure-eight-shaped path at a phase difference
of 180 degrees, in such a way that the latter path lays on
the surface of a hemisphere that additionally spins about
its axis of symmetry, the involved inertial forces lead to a
non-vanishing net impulse. Intuitively, this claim is true
because when the ratio of the spinning angular velocity
over the first one (formation of the figure-eight-shaped
path) is not an integer number, every mass does not repeat its initial position; in other words, when the mass
completes the figure-eight-shaped path (every 360 degrees), the linear momentum has a different value that
what it had at the initial position. In terms of combined
structural mechanics and kinematics, every mass is connected to the center of the hemisphere through an elastic
rod that is imposed to highly variable tensile deformation.
This finding implies a variable radius of the hemisphere
Copyright 2010 SciRes.
655
11. References
[1]
[2]
[3]
[4]
[5]
I. I. Blekhman, Synchronization in Science and Technology, ASME Press, New York, 1988.
[6]
[7]
[8]
[9]
I. I. Blekhman, Vibrational Mechanics: Nonlinear Dynamic Effects, General Approach, Applications, World
Scientific, Singapore, 2000.
[10] M. G. Millis and N. E. Thomas, Responding to Mechanical Antigravity, NASA/TM-2006-214390, AIAA2006-4913, December 2006. http://gltrs.grc.nasa.gov/reports/2006/TM-2006-214390.pdf
[11] M. G. Millis, Assessing Potential Propulsion Breakthroughs, In: E. Belbruno, Ed., Annals of the New York
Academy of Sciences, Vol. 1065, New York, December
2005, pp. 441-461.
[12] C. G. Provatidis, Some Issues on Inertia Propulsion
Mechanisms Using Two Contra-Rotating Masses, Theory of Mechanisms and Machines, Vol. 8, No. 1, April
ENG
C. G. PROVATIDIS
656
2010, pp. 34-41.
[13] C. G. Provatidis and V. T. Tsiriggakis, A New Kinematics Theory in Physics and Presentation of a Device for
Gravity Studies, Proceedings 9th International Scientific-Practical Conference on Research, Development and
Applications of High Technologies in Industry, A. P.
Kudinov, Ed., Vol. 1, St. Petersburg, April 2010, pp. 386393.
[14] C. G. Provatidis and V. T. Tsiriggakis, A New Concept
and Design Aspects of an Antigravity Propulsion
Mechanism Based on Inertial Forces, Proceedings 46th
AIAA/ASME/SAE/ASEE Joint Propulsion Conference &
Exhibit, Nashville, July 2010.
[15] C. G. Provatidis, A Novel Mechanism to Produce Fig-
ure-Eight-Shaped Closed Curves in the Three-Dimensional Space, In: D. Tsahalis, Ed., Proceedings of 3rd
International Conference on Experiments/Process/System
Modeling/Simulation & Optimization, Athens, July 2009.
[16] R. Wayte, The Phenomenon of Weight-Reduction of a
Spinning Wheel, Meccanica, Vol. 42, No. 4, August
2007, pp. 359-364.
[17] M. Tajmar, Homopolar Artificial Gravity Generator Based on Frame-Dragging, Acta Astronautica, Vol. 66, No.
9-10, May-June 2010, pp. 1297-1301.
[18] V. O. Kononenko Vibrating Systems with a Limited
Power Supply, Iliffe Books Ltd, London, 1969.
[19] A. H. Nayfeh and D. T. Mook Nonlinear Oscillations,
John Wiley & Sons, Inc., New York, 1979.
ENG
C. G. PROVATIDIS
APPENDIX A
Remarks concerning the figure-eight-shaped curve
For the purpose of completeness, details are provided
here for the figure-eight-shaped curve.
First, in the absence of the spinning motion, i.e., when
0 , a careful inspection of (5) and (6) reveals that:
1) At the initial time instance ( t 0 ), in fact the rods
obtain their horizontal position parallel to x-axis,
i.e., a on the left and b on the right side.
2) At the time instance given by t 2 , both
masses obtain their vertical position.
3) At the time instance given by t , the masses
mutually interchange their (horizontal) position.
4) At the time instance given by t 3 2 , the
masses are again found at their vertical position.
5) At the time instance given by t 2 , the masses
obtain their initial (horizontal) position, and so on.
6) Therefore, the distance between the two masses
varies from the minimum 2 R (vertical position) to
z
657
xa ya z a xb yb zb rsphere
, with rsphere r 2 R 2
Second, in the case of a non-vanishing spinning angular velocity ( 0 ), the lumped masses do not generally
follow the same path, as clearly shown in Figures
5(b)-5(f), particularly when the ratio z of the angular velocities is not an integer number.
z
1a
1a
1a
ENG
Ezenwa Alfred Ogbonnaya1, Hyginus Ubabuike Ugwu2, Charles Agbeju Nimibofa Johnson3
Department of Marine Engineering, Rivers State University of Science and Technology, Port Harcourt, Nigeria
Department of Mechanical Engineering, Michael Okpara University of Agriculture (MOUA), Umuahia, Nigeria
3
Department of Marine Engineering, Niger Delta University, Bayelsa, Nigeria
E-mail: {ezenwaogbonnaya, canjohnson2000}@yahoo.com, ubadyke2001@yahoo.com.au
Received March 25, 2010; revised June 24, 2010; accepted June 28, 2010
Abstract
Surge and stall are the two main types of instabilities that often occur on the compressor system of gas turbines. The effect of this instability often leads to excessive vibration due to the back pressure imposed to the
system by this phenomenon. In this work, fouling was observed as the major cause of the compressor instability. A step to analyze how this phenomenon can be controlled with the continuous examination of the vibration amplitude using a computer approach led to the execution of this work. The forces resulting to vibration in the system is usually external to it. This external force is aerodynamic and the effect was modeled
using force damped vibration analysis. A gas turbine plant on industrial duty for electricity generation was
used to actualize this research. The data for amplitude of vibration varied between 15 and 15 mm/s while
the given mass flow rate and pressure ratio were determined as falling between 6.1 to 6.8 kg/s and 9.3 to 9.6
respectively. A computer program named VICOMS written in C++ programming language was developed.
The results show that the machine should not be run beyond 14.0 mm vibration amplitude in order to avoid
surge, stall and other flow-induced catastrophic breakdown.
Keywords: Computerized Solution, Instabilities, Vibration, Gas Turbine Compressors, Operational Limits
1. Introduction
The economics of power generation with gas turbines
(GT) is now quite attractive due to its low capital cost, its
high reliability and flexibility in operation. Another outstanding feature is its capability of quick starting and
using wide variety of fuel from natural gas to residual oil
or powdered coal. Due to better material being made
available and with the use of adequate blade cooling, the
inlet gas temperature of the turbine blades can now exceed 1200C as a result of which the overall efficiency of
GT plant can be 35%. This is almost the same as that of a
conventional steam power plant. Based on these developments, occurrence of instabilities in the compressor
system would no doubt result to performance deterioration of the overall efficiency of the GT. There are two
basic types of instabilities that could be encountered in
the GT compressor system namely the rotating stall and
surge. Both types of instabilities have damaging consequences to the compressor.
According to Iwakiri, et al. [1], rotating stall causes
the compressor to operate with extremely low frequencies resulting in excessive high internal temperature that
has an adverse effect on blade life. Surge causes severe
problems such as excessive pressure built-up at the inlet
and cyclic loading on the compressor. These instabilities
might lead to the inability of the compressor blade to
produce the required loading and the engine might sustain catastrophic damage as a result of the excessive vibration [2].
ENG
E. A. OGBONNAYA ET AL.
659
Surge line
Fouled
Design point
Pa
P1
Heavy stage
loading
Light stage
.
m To
Po
mx cx kx 0
(1)
E. A. OGBONNAYA
660
ET
AL.
c
k
2
2 ,
n
m
m
K
C-damper
.
..
F cos t
x 2 x n 2 x o
m
..
x 2 x n2 x 0 is;
Direction of
shaftrotation
X 1 C1e
2 - t
C2 e
2 - 2 t
X 2 C1 C 2t e t
.
Cx
Kx
(7)
X3 e
t
A cos
2 t B sin
2 t
D
F0cost
F0
cos t
M
2 D 2 x
F0
cost
M
and x 2
D 2 D 2
since f (D2) cost = f ( 2) cost,
..
mx
Figure 3. Free body diagram of the Compressor Model.
..
F0cost + k (dst + x) = m x
..
2 D - 2 2
X 2 2 n 2
2
4 D n
F0
cos t
M
F0 2 Dsin t - n2 2 cos
2
M
4 2 2 n2 2
2 2
2
2 2
F0 4 n cos t -
2
M
4 2 2 n2 2
F0cost + kdst + kx = m x
(2)
(3)
where tan
2 2
n
(4)
..
F0cost + kx = m x
..
and
m x + kx = F0cost
.
For damping force analysis C x
where:
Copyright 2010 SciRes.
(5)
(6)
2 2
2
2
F0 4 n
M 4 2 2 n2 2 2
which,
Frequency, F =
cost
ENG
E. A. OGBONNAYA ET AL.
amplitude.
661
Start
F0
M 4 2 2 n2 2
(8)
4
2
2
n
(9)
where;
F0 = aerodynamic force (N)
M = mass of the shaft (kg)
X = vibration displacement amplitude (mm/s)
= force frequency (Hz)
n = natural frequency (rad/s)
= product of coefficient of damping (Nsm-1).
This is the equation used to model the flowchart and
consequently design the program to simulate a solution
to the vibration effects of instabilities in GT compressors.
Figure 4 shows the flowchart for VICOMS written in
C++ language for obtaining the aerodynamic force resulting to vibration in the GT system. It has one loop as
shown and can go round several iterations until the first
speed is equal to or less than the last speed to make the
program stop. This is when the surge would have been
uncontrollable as to cause damage to the plant. Hence,
VICOMS would predict when the GT should undergo
maintenance check.
The flowchart in Figure 4 led to the evaluation of the
computer program code written in C++ programming
language. The program helped in the calculation of aerodynamic force as stated in Equation (9).
Declare / Define
variables used
1 1, n 314
Input first, speed last speed
and step value
n = 2N / 60
Input active load
Input X1 [ i ] and X2 [ i ]
F01 i X1 i m 4 2 2 n2 2
2
F i X i m 4 2 2 n2 2
2
02
Yes
Stop
ENG
E. A. OGBONNAYA
662
Speed
rp
(kg/s)
(RPM)
Active
Load
Fre-
Vibration
quency
Amplitude
(Hz)
Brg.2
15.0
6.49
9.30
3005
50
51.2
4.8
6.7
9.40
3053
50
51.3
4.8
15.0
6.2
9.30
3051
50
51.2
4.6
15.0
6.8
9.50
3080
50
51.3
4.9
10.0
6.4
9.60
3063
50
51.0
4.9
15.0
6.3
9.50
3063
50
50.4
5.1
15.0
6.1
9.40
3074
45
51.1
5.1
15.0
6.3
9.50
3076
40
51.3
5.3
7.0
6.5
9.40
3077
40
56.7
5.0
15.0
6.66
9.40
3081
36
51.5
5.1
6.40
A work has been carried out on the computerized solution of instabilities in GT compressor. The test engine is
AFAM III, GT 17, TYPE 13D.
The compressor system suffered surge and stall, which
resulted to instabilities in the test engine due to fouling.
It was shown that fouling leads to the stuffing of the
compressor stages. This also results in the reduction of
the compressor surge margin and dramatic instability of
the operation of the whole GT compressor system, culminating to vibration.
The mass flow rate, pressure ratio, shaft speed and vibration amplitude in the system were collected from the
two end bearings of the compressor system in the GT
plant. A model was consequently developed to analyze
the data collected in order to determine the corresponding aerodynamic force causing vibration in the system.
The mathematical model was used to run a program code
named VICOMS written in C++ programming language.
The results and the graph showed that the GT should not
be run beyond 14 mm.
20
15
Amplitude of Vibration (mm)
AL.
4. Conclusions
(mm/sec)
Brg1
ET
10
5
0
-5
5. Acknowledgements
-10
-20
Time in Hours
3070
3060
6. References
3050
[1]
[2]
R. Kurz, Surge Control Design System Design, Proceedings of ASME Turbo Expo, Orlando, Florida, 8-12
June 2009. www.teurbexpo.org
J.-L. Lu, W. Chu and K. Peng, Numerical and Experimental Research of Stall Inception on Subsonic Axial
Flow Compressor Rotor, Journal of Aerospace Engineering, Vol. 48, No. 2, 2010, pp. 3-4.
T. Okada, A. Kawajiri, O. Yataka and O. Eisuke, Stall
Inception Process and Prospects for Active Hub-Flap
Control in Three Stage Axial Flow Compressor, Journal
of Thermal Science, Co-Published with Springer-Varlag
GmbH, 2008, pp. 4-8.
3040
3030
3020
3010
[3]
3000
6
6.2
6.4
6.6
6.8
[4]
ENG
E. A. OGBONNAYA ET AL.
[5]
[6]
[7]
[8]
[9]
663
ENG
E. A. OGBONNAYA
664
ET
AL.
Nomenclature:
ENG
Abstract
The results of comparative DC simulation tests confirm that a proposed modification to the feedback circuit
of a Flipped Voltage Follower (FVF), to produce a type of Folded Flipped Voltage Follower (FFVF), is
capable of maximising the linear DC operating range for given values of supply rail voltage and operating
current.
Keywords: Analog, Voltage Follower
1. Introduction
The name Flipped Voltage Follower (FVF) was coined
by Carvajal et al. [1] to describe a class of pre-existing,
and new, low power/low voltage analogue circuits.
A prototype FVF is a two transistor source-follower in
which the input mosfet is forced to operate at a sensibly
constant DC drain current, set by ancillary circuitry, despite variation in input voltage or load current. This is
achieved by the action of shunt negative feedback.
The overall result is a source-follower with decreased
output impedance and increased linearity in its voltage
transfer characteristic. The so-called Super Source-Follower [2] can be regarded as a member of the FVF
family: in fact, it has been called a Folded Flipped Voltage
Follower (FFVF) [3].
In Figure 1, M1 and M2 are inter-connected to form an
N-channel FVF the operating current for which in supplied by MX, the output mosfet of a simple 1:1 current
mirror formed from MW and MX. The mirror input current,
IX, is set by choice of RB.
A capacitor, CS [1], may be required to produce a
specified phase margin in the loop-gain frequency response.
M1 passes an effectively constant current so the incremental voltage gain of the FVF is close to unity providing it operates in its linear region. Unfortunately, as
has been noted in [1], the valid linear range decreases with
threshold voltage. This is most easily seen by applying
Equations (1) and (2), which follow to the case in which
the characteristics of M1 and M2 are identical.
From [4],
ID
N
2
VGS VTN
2
(1)
2I D
N
(2)
and, VDS(min)
+VDD
1:1
MX
MW
VD
IX
RB
M1
VG
VS
M2
CS
-VSS
ENG
666
2I X
2VTN
N
(4)
2. Proposed Circuit
Figure 2 shows the proposed FFVF circuit. It differs from
that of Figure 1 (and that of [3]), by the way in which the
feedback connection is made from the drain of M1 to the
gate of M2. Instead of the direct link of Figure 1 an additional mosfet, M3, is included and forced to operate at a
sensibly constant current, IZ, provided by the high output
resistance Widlar-type current mirror formed from MY, MZ
and RZ. MX performs the same function as Figure 1 but, in
this case its output current is IY so in normal operation the
current in M1 is (IY-IZ).
M3 performs two functions. The first is to provide a
feedback current, which is converted to a feedback voltage at the gate of M2. The second function of M3 is to
keep the drain source voltage of M1 constant, preferably
at the minimum level for saturation, with variation in VG.
Using Equations (1) and (2), that can be achieved if,
2I Z
VTP
P
2 IY I Z
N
(6)
+VDD
1:1
MW
MX
VD
VG
M1
IY
RB
VS
M3
M2
IZ
MY
MZ
CS
RZ
ENG
P
N
From a DC standpoint the choice of the ratio IZ/IY is not
critical provided Equation (6) is satisfied. However, from
a small signal viewpoint the choice of IZ affects the
loop-gain characteristics via its effect on the dynamic
parameters of M3.
667
1.5V
1.0V
0.5V
0V
VVDD,
VZZ
V
VGG
-0.5V
3. Results
VVZZ
-1.0V
VSS
V
-1.5V
-1.5V
VVDD
VGG
V
0V
V
VSS
-1.0V
-0.5V
-0.0V
0.5V
1.0V
-0.0V
0.5V
1.0V
1.5V
4. Conclusions
The superior DC performance of the proposed FFVF,
compared with that of the FVF is clearly evident.
5. References
[1]
[2]
[3]
[4]
-0.5V
-1.0V
-0.5V
0.5 V where the voltage traces for VD (< VG) and VS are
parallel to that of VG, in accordance with the theory presented. (Above VG = 1 V the onset of triode behavior in
MX causes non linearity)
1.0V
0.5V
-1.0V
VG
VG
Figure 4. Voltage traces for Figure 2 (See text for circuit
details).
1.5V
-1.5V
-1.5V
VVDD
1.5V
VG
VG
Figure 3. Voltage traces for Figure 1 (See text for circuit details).
ENG
Abstract
Manual process of creating bonding diagram is known to be time consuming and error prone. In comparison,
offline programming (OLP) provides a much more viable option to reduce the wire bonding creation time
and error. OLP is available in two versions, i.e., vendor specific OLP and direct integration offline programming (Di-OLP). Both versions utilize the bonding diagram and computer aided design data to speed up
bonding program creation. However, the newly proposed Di-OLP is more flexible as it can be used to create
bonding program for multiple machine platforms in microelectronics industry. Some special features of
Di-OLP method are presented. The application of generic OLP however, is applicable to machines that recognize ASCII text file. The user needs to know the data format accepted by machine and convert the data
accordingly to suit its application for different machine platforms. Di-OLP is also a practical method to replace the time consuming manual method in production line.
Keywords: Wire Bonding, Offline Programming, Computer Aided Design, Direct Integration Offline
Programming, Bondlist
1. Introduction
Semiconductor industry is moving in the trend of increased integration and miniaturization. This has resulted in increasing number of bond pads on a chip. These
pads will later be wire bonded to a leadframe via a process called wire bonding [1,2]. Wire bonding process is
basically a process where interconnection between chip
and leadframe is established via thin gold wires. Wire
bond machines utilize precise control of bonding force,
ultrasonic vibration, bonding temperature and bonding
time to establish the connection between gold wire to
bond pad or leadframe. The trend of increased integration has resulted in new challenges for wire bond process;
mainly because more wires are bonded on a chip and pad
pitch has become smaller [3]. A single semiconductor
product can contain as much as 600 wires and pitch distance can be as low as 50 micron or smaller [4]. One of
the main challenges from this trend is the traditional
method to manually prepare the bonding program has
become very time consuming and error prone.
In order to carry out automatic wire bonding, a wire
bond machine requires a set of pre-program instruction.
These instructions will be saved as a wire bond program
(WBP). The WBP is also called wire bond recipe. The
WBP mainly consists of three sets of bonding instruction.
Copyright 2010 SciRes.
They are material handling, bonding parameter and bonding path instruction. The material handling information
such as magazine dimension, leadframe-indexing pitch
can be keyed into machine relatively fast as in most production floor these dimensions are standardized. Bonding
parameters on the other hand do require slightly more
effort if optimization is required. However when proper
characterization is carried out, such as grouping of the
bonding parameter by different types of bond pad material, bonding capillary, etc, it allows user to re-use the
bond parameter when coming across the same bonding
condition. This will allow wire bond parameter to be
keyed into WBP with relative ease.
The standardization and characterization option, however do not apply to the bonding path component of
bonding program. Bonding path component is required
to guide bond head to the correct position during the
bonding process. Bonding path component can be represented by a set of bonding coordinates with each consists
of two points that are connected to form a representing
connectivity between bond pad and leadfinger as shown
in Figure 1. According to the conventional manual
method, every new product would require the user to
manually input all bonding path coordinates into the
bonding program one by one.
The conventional manual method of inputting bonding
coordinates into bonding program is suitable for product
ENG
Y. L. FOO ET AL.
Wire,
W1 = xb1, yb1, xw1, yw1
W2 = xb2, yb2, xw2, yw2
W3 = xb3, yb3, xw3, yw3
= .
= .
= .
= .
= .
= .
Wn = xbn, ybn, xwn, ywn
Figure 1. A bonding diagram and a set of bonding coordinates representing bonding path component. Cross symbol shows the coordinate system origin at the center of the
chip and leadframe.
with relatively low pin count (less than 48 wires). However as wire count increases and bond pitch becomes
smaller the manual bonding program preparation method
becomes very tedious and time consuming. Moreover
errors are likely to occur with this method. Errors are
likely to occur due to the fact that operator only has access to locally enlarge image of bond pad and leadframe
during the manual bonding program preparation process.
This narrow vision makes it very difficult for operator to
recognize the exact bonding bond pad position where
bonding wire is to be placed. The risk of misplacing the
wire on the adjacent bond pad increases, as bond pad
pitch becomes smaller.
These limitations of manual bonding program creation
process have driven user and equipment vendor to explore other option, to speed up bonding program creation process and reduce human errors. The solution is
known as offline programming (OLP) method. This method extracts the bonding coordinates from the computer
aided design (CAD) drawing and utilizes it to simultaneously program all wires [5]. It greatly reduces the time
required to create the bonding program [6-8].
669
670
Y. L. FOO ET AL.
sions of OLP softwares from different wire bond machine vendor. This will increase the implementation cost
of vendor specific OLP. The other disadvantage of vendor specific OLP is, it only works on specific version of
CAD application. OLP from wire bond vendor A might
requires AutoCAD 2005 to work with, while OLP from
wire bond vendor B might requires AutoCAD 2006 to
create WBP. This means if a production line consists of
wire bond machine from different vendors, one might
need to license two or more versions of CAD software.
This would translate into additional licensing fees. The
cost of implementation and the inflexibility of vendor
specific OLP are key factors that hinder wide spread of
vendor specific OLP.
Another alternative to the vendor specific OLP is to
use the bondlist created by bonding diagram creation
software. Bondlist information can be converted into
machine recognizable format and carries out OLP. The
method on how bondlist information can be utilized for
OLP is presented in the following section. The term direct integration offline programming (Di-OLP) is used to
differentiate this method from vendor specific OLP.
Center of leadframe
In bondlist all the bonding location on the leadfinger is represented by a set of x and y coordinate.
Figure 4. Bonding location on leadframe is referred to coordinate system origin at the center of the leadframe drawing.
ENG
Y. L. FOO ET AL.
(1)
671
o
Poo
Pno
Pno
(2)
Y. L. FOO ET AL.
672
y
P (x, y)
6. References
P (x, y)
[1]
5. Conclusions
Manual process of creating bonding diagram is found to
be time consuming and error prone. OLP provides a
much more viable option to reduce the wire bonding
creation time and error. OLP is available in two versions,
vendor specific OLP and Di-OLP as described. Both
versions utilize the bonding diagram CAD data to speed
up bonding program creation. However, the proposed
Di-OLP is more flexible as it can be used to create
bonding program for multiple machine platforms. The
application of generic OLP however, is applicable to
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
S. Kalpakjian, Manufacturing Engineering and Technology, 3rd Edition, Surface Technology, Kansas, 1995.
S. DiBartolomeo, Advance Packaging, Penn Well,
Nashua, 2000.
R. R. Tummala, V. Sudaram, F. Liu, G. White, S. Bhattacharya, R. M. Pulugurtha, M. Swaminathan, J. Laskar,
N. M. Jokerst and S. Y. Chow, High Density Packaging
in 2010 and Beyond, IEEE International Symposium on
Electronic Materials and Packaging, Taiwan, 2002, pp.
30-36.
L Nguyen, I. Singh, C. Murray, J. Jackson, J. DeRosa and
D. Ho, 70 m Fine Pitch Wire Bonding, IEEE International Electronics Manufacturing Technology Symposium, Adelaide, 1998, pp. 394-400.
T. C. Chang, R. A. Wysk and H. P. Wang, ComputerAided Manufacturing, 3rd Edition, Prentice Hall, New
York, 1998.
S. K. Prasad, Advanced Wire Bond Interconnection,
Springer, Berlin, 2004.
C. J. Oh, Y. J. Lee, Y. J. Han and C. S. Ahn, A New
System for Reducing the Bonding Process Cycle Time
and Increasing the Accuracy of Bonding Diagram, IEEE
International Conference on System, Man and Cybernetics, Vol. 5, 2004, pp. 4301-4305.
Y. L. Foo, A. H. You and C. W. Chin, Direct Integration
Offline Programming Method in Wire Bonding Process,
11th International Conference on Electronic Materials
and Packaging, Taiwan, 2009, pp. 1-5.
G. G. Harman, Wire Bonding in Microelectronics Materials, Processes, Reliability and Yield, 2nd Edition,
McGraw-Hill, New York, 1997.
ENG
ENGINEERING
A Journal Published by Scientific Research Publishing, USA
www.scirp.org/journal/eng
9 771947 393005
Editor-in-Chief
Prof. David L. Carroll
08
Editorial Board
Dr. Rehan Ahmed
Prof. MohD A. M. Al-Nimr
Prof. Jan Awrejcewicz
Prof. Shahnor Basri
Prof. Alain. Bernard
Dr. P. Chandramohan
Dr. Hongyang Chen
Prof. Ming Chen
Dr. Rahat Iqbal
Prof. Shouetsu Itou
Prof. Jae Moung Kim
Prof. Chui-Chi Lee
Prof. Yong-Gang Lv
Prof. John Marsh
Dr. Mehrdad Massoudi
Dr. Koduri Venkata Surya Ramam
Prof. Sergey M. Smolskiy
Prof. Hongbin Sun
Prof. Baolin Wang
Prof. Chengshan Wang
Prof. Gaofeng Wang
Prof. Zhao Xu
Dr. Wei Yan
Dr. V.P. Yashnikov
Prof. Tsutomu Yoshihara
Prof. Xiangjun Zeng
Dr. Hongyu Zhang
Prof. Luowei Zhou
H e r i o t-Wa t t U n i v e r s i t y, U K
J o r d a n U n i v e r s i t y o f S c i e n c e a n d Te c h n o l o g y, J o r d a n
Te c h n i c a l U n i v e r s i t y o f L o d z , Po l a n d
P u t ra U n i v e r s i t y, M a l a y s i a
E c o l e C e n t ra l e d e N a n t e s , F ra n c e
A n n a U n i v e r s i t y, I n d i a
T h e U n i v e r s i t y o f To k y o , J a p a n
S o u t h e a s t U n i v e r s i t y, C h i n a
C o v e n t r y U n i v e r s i t y, U K
Ka n a g a w a U n i v e r s i t y, J a p a n
I N H A U n i v e r s i t y, Ko r e a ( S o u t h )
S H U -T E U n i v e r s i t y, Ta i w a n ( C h i n a )
C h o n g q i n g U n i v e r s i t y, C h i n a
I E E E P h o t o n i c s S o c i e t y, U K
U . S . D e p a r t m e n t o f E n e r g y / N E T L- P i t t s b u r g h , U S A
Universidad of Concepcion, Chile
M o s c o w Po w e r E n g i n e e r i n g U n i v e r s i t y, R u s s i a
Ts i n g h u a U n i v e r s i t y, C h i n a
H a r b i n I n s t i t u t e o f Te c h n o l o g y, C h i n a
T i a n j i n U n i v e r s i t y, C h i n a
Wu h a n U n i v e r s i t y, C h i n a
Te c h n i c a l U n i v e r s i t y o f D e n m a r k , D e n m a r k
Tr e n d M i c r o , U S A
Russian Academy of Sciences, Russia
Wa s e d a U n i v e r s i t y, J a p a n
C h a n g s h a U n i v e r s i t y o f S c i e n c e a n d Te c h n o l o g y, C h i n a
C e r e s I n c ., T h o u s a n d O a k s , U S A
C h o n g q i n g U n i v e r s i t y, C h i n a
ENGINEERING is an international journal dedicated to the latest advancement of engineering. The goal of
this journal is to provide a platform for engineers and academicians all over the world to promote, share,
and discuss various new issues and developments in different areas of engineering. All manuscripts must
be prepared in English, and are subject to a rigorous and fair peer-review process. Accepted papers will
immediately appear online followed by printed hard copy. The journal publishes original papers including
but not limited to the following fields:
Aerospace Engineering
Agricultural Engineering
Chemical Engineering
Civil Engineering
Electrical Engineering
Environmental Engineering
Industrial Engineering
Materials Engineering
Mechanical Engineering
Mining Engineering
Nanotechnology
Nuclear Engineering
Power Engineering
Test Engineering
Transportation Engineering
We are also interested in: 1) Short Reports2-5 page papers where an author can either present an idea
with theoretical background but has not yet completed the research needed for a complete paper or
preliminary data; 2) Book ReviewsComments and critiques.
eng@scirp.org
TABLE OF CONTENTS
August 2010
Volume 2 Number 8
Lie Group Analysis for the Effects of Variable Fluid Viscosity and Thermal Radiation on
Free Convective Heat and Mass Transfer with Variable Stream Condition
P. Loganathan, P. P. Arasu............................................................................................................ 625
Statistical Modeling of Pin Gauge Dimensions of Root of Gas Turbine Blade in Creep
Feed Grinding Process
A. R. Fazeli.................................................................................................................................635
Copyright
2010 SciRes.